Computer_Hardware_Senan Mehdiyev.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Importing libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "import seaborn as sns\n",
    "import sklearn.model_selection as sms\n",
    "import sklearn.preprocessing as sp\n",
    "import sklearn.linear_model as slm\n",
    "import sklearn.metrics as sm\n",
    "import sklearn.ensemble as ens\n",
    "from warnings import filterwarnings \n",
    "filterwarnings('ignore') # to ignore warnings\n",
    "import sklearn.linear_model as slm\n",
    "import sklearn.svm as svm"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Loading data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Seller Name</th>\n",
       "      <th>Model Name</th>\n",
       "      <th>Machine Cycle Time in nanoseconds</th>\n",
       "      <th>Minimum Main Memory in Kilobytes</th>\n",
       "      <th>Maximum Main Memory in kilobytes</th>\n",
       "      <th>Cache Memory in Kilobytes</th>\n",
       "      <th>Minimum Channels in Units</th>\n",
       "      <th>Maximum Channels in Units</th>\n",
       "      <th>Published Relative Performance</th>\n",
       "      <th>Estimated Relative Performance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>amdahl</td>\n",
       "      <td>470v/7</td>\n",
       "      <td>29</td>\n",
       "      <td>8000</td>\n",
       "      <td>32000</td>\n",
       "      <td>32</td>\n",
       "      <td>8</td>\n",
       "      <td>32</td>\n",
       "      <td>269</td>\n",
       "      <td>253</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>amdahl</td>\n",
       "      <td>470v/7a</td>\n",
       "      <td>29</td>\n",
       "      <td>8000</td>\n",
       "      <td>32000</td>\n",
       "      <td>32</td>\n",
       "      <td>8</td>\n",
       "      <td>32</td>\n",
       "      <td>220</td>\n",
       "      <td>253</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>amdahl</td>\n",
       "      <td>470v/7b</td>\n",
       "      <td>29</td>\n",
       "      <td>8000</td>\n",
       "      <td>32000</td>\n",
       "      <td>32</td>\n",
       "      <td>8</td>\n",
       "      <td>32</td>\n",
       "      <td>172</td>\n",
       "      <td>253</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>amdahl</td>\n",
       "      <td>470v/7c</td>\n",
       "      <td>29</td>\n",
       "      <td>8000</td>\n",
       "      <td>16000</td>\n",
       "      <td>32</td>\n",
       "      <td>8</td>\n",
       "      <td>16</td>\n",
       "      <td>132</td>\n",
       "      <td>132</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>amdahl</td>\n",
       "      <td>470v/b</td>\n",
       "      <td>26</td>\n",
       "      <td>8000</td>\n",
       "      <td>32000</td>\n",
       "      <td>64</td>\n",
       "      <td>8</td>\n",
       "      <td>32</td>\n",
       "      <td>318</td>\n",
       "      <td>290</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Seller Name Model Name  Machine Cycle Time in nanoseconds  \\\n",
       "0      amdahl     470v/7                                 29   \n",
       "1      amdahl    470v/7a                                 29   \n",
       "2      amdahl    470v/7b                                 29   \n",
       "3      amdahl    470v/7c                                 29   \n",
       "4      amdahl     470v/b                                 26   \n",
       "\n",
       "   Minimum Main Memory in Kilobytes  Maximum Main Memory in kilobytes  \\\n",
       "0                              8000                             32000   \n",
       "1                              8000                             32000   \n",
       "2                              8000                             32000   \n",
       "3                              8000                             16000   \n",
       "4                              8000                             32000   \n",
       "\n",
       "   Cache Memory in Kilobytes  Minimum Channels in Units  \\\n",
       "0                         32                          8   \n",
       "1                         32                          8   \n",
       "2                         32                          8   \n",
       "3                         32                          8   \n",
       "4                         64                          8   \n",
       "\n",
       "   Maximum Channels in Units  Published Relative Performance  \\\n",
       "0                         32                             269   \n",
       "1                         32                             220   \n",
       "2                         32                             172   \n",
       "3                         16                             132   \n",
       "4                         32                             318   \n",
       "\n",
       "   Estimated Relative Performance  \n",
       "0                             253  \n",
       "1                             253  \n",
       "2                             253  \n",
       "3                             132  \n",
       "4                             290  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# retrieving data\n",
    "#df=pd.read_csv(\"D:\\\\Data\\\\Gamboo\\Computer_Hardware\\Computer Hardware\\Computer Hardware.csv\")\n",
    "df=pd.read_csv(\"Computer Hardware.csv\")\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### EDA, Preprocessing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To check the general form of columns which may need some **cleaning operations like handling null(missing values), converting datatype, and others**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 208 entries, 0 to 207\n",
      "Data columns (total 10 columns):\n",
      " #   Column                             Non-Null Count  Dtype \n",
      "---  ------                             --------------  ----- \n",
      " 0   Seller Name                        208 non-null    object\n",
      " 1   Model Name                         208 non-null    object\n",
      " 2   Machine Cycle Time in nanoseconds  208 non-null    int64 \n",
      " 3   Minimum Main Memory in Kilobytes   208 non-null    int64 \n",
      " 4   Maximum Main Memory in kilobytes   208 non-null    int64 \n",
      " 5   Cache Memory in Kilobytes          208 non-null    int64 \n",
      " 6   Minimum Channels in Units          208 non-null    int64 \n",
      " 7   Maximum Channels in Units          208 non-null    int64 \n",
      " 8   Published Relative Performance     208 non-null    int64 \n",
      " 9   Estimated Relative Performance     208 non-null    int64 \n",
      "dtypes: int64(8), object(2)\n",
      "memory usage: 16.4+ KB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.columns = ['Seller_Name', 'Model_Name', 'MYCT', 'MNM', 'MXM', 'CACH','CHMIN','CHMAX', 'PRP', 'ERP']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>MYCT</th>\n",
       "      <th>MNM</th>\n",
       "      <th>MXM</th>\n",
       "      <th>CACH</th>\n",
       "      <th>CHMIN</th>\n",
       "      <th>CHMAX</th>\n",
       "      <th>PRP</th>\n",
       "      <th>ERP</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>208.000000</td>\n",
       "      <td>208.000000</td>\n",
       "      <td>208.000000</td>\n",
       "      <td>208.000000</td>\n",
       "      <td>208.000000</td>\n",
       "      <td>208.000000</td>\n",
       "      <td>208.000000</td>\n",
       "      <td>208.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>204.201923</td>\n",
       "      <td>2880.538462</td>\n",
       "      <td>11824.019231</td>\n",
       "      <td>24.096154</td>\n",
       "      <td>4.644231</td>\n",
       "      <td>17.740385</td>\n",
       "      <td>105.177885</td>\n",
       "      <td>98.850962</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>260.833016</td>\n",
       "      <td>3883.839300</td>\n",
       "      <td>11747.916663</td>\n",
       "      <td>37.417999</td>\n",
       "      <td>6.787198</td>\n",
       "      <td>24.913375</td>\n",
       "      <td>161.090223</td>\n",
       "      <td>154.974961</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>17.000000</td>\n",
       "      <td>64.000000</td>\n",
       "      <td>64.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>15.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>50.000000</td>\n",
       "      <td>768.000000</td>\n",
       "      <td>4000.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>27.000000</td>\n",
       "      <td>28.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>110.000000</td>\n",
       "      <td>2000.000000</td>\n",
       "      <td>8000.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>49.500000</td>\n",
       "      <td>45.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>225.000000</td>\n",
       "      <td>4000.000000</td>\n",
       "      <td>16000.000000</td>\n",
       "      <td>32.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>24.000000</td>\n",
       "      <td>111.500000</td>\n",
       "      <td>99.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1500.000000</td>\n",
       "      <td>32000.000000</td>\n",
       "      <td>64000.000000</td>\n",
       "      <td>256.000000</td>\n",
       "      <td>52.000000</td>\n",
       "      <td>176.000000</td>\n",
       "      <td>1150.000000</td>\n",
       "      <td>1238.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              MYCT           MNM           MXM        CACH       CHMIN  \\\n",
       "count   208.000000    208.000000    208.000000  208.000000  208.000000   \n",
       "mean    204.201923   2880.538462  11824.019231   24.096154    4.644231   \n",
       "std     260.833016   3883.839300  11747.916663   37.417999    6.787198   \n",
       "min      17.000000     64.000000     64.000000    0.000000    0.000000   \n",
       "25%      50.000000    768.000000   4000.000000    0.000000    1.000000   \n",
       "50%     110.000000   2000.000000   8000.000000    8.000000    2.000000   \n",
       "75%     225.000000   4000.000000  16000.000000   32.000000    6.000000   \n",
       "max    1500.000000  32000.000000  64000.000000  256.000000   52.000000   \n",
       "\n",
       "            CHMAX          PRP          ERP  \n",
       "count  208.000000   208.000000   208.000000  \n",
       "mean    17.740385   105.177885    98.850962  \n",
       "std     24.913375   161.090223   154.974961  \n",
       "min      0.000000     6.000000    15.000000  \n",
       "25%      5.000000    27.000000    28.000000  \n",
       "50%      8.000000    49.500000    45.000000  \n",
       "75%     24.000000   111.500000    99.500000  \n",
       "max    176.000000  1150.000000  1238.000000  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABDoAAAF4CAYAAAC1hZl5AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAAAsD0lEQVR4nO3de5hlZX0n+u+vacAWtLnYIgUYTEv0mEk02vESNVHwYDQmeibqxPSJ6JiHM4yHJBN7JibmZJijydEZTGJi0hkSjcR0vOTiwKACppEEHS9pFBsUiZQRgRJpGmi5CTT9nj/2Krtoqm/03rV3rfp8nqeevdbaa7/1q7eqd6/13e96V7XWAgAAANAHy8ZdAAAAAMCwCDoAAACA3hB0AAAAAL0h6AAAAAB6Q9ABAAAA9IagAwAAAOiNkQYdVXVEVf1NVX21qq6uqudU1VFV9Ymq+lr3eGS3b1XVH1TVtVW1uaqePsraAAAAgP4Z9YiOdyW5sLX25CRPTXJ1kjcn2dhaOynJxm49SV6S5KTu6/Qk60dcGwAAANAz1VobTcNVK5NckeT725xvUlXXJHlBa+1bVXVskktba0+qqv/eLX9g1/1GUiAAAADQO6Mc0fGEJFuS/HlVfbGq/qyqDktyzJzw4qYkx3TLxyW5fs7rb+i2AQAAAOyT5SNu++lJzmytfa6q3pWdl6kkSVprrar2a0hJVZ2ewaUtOeyww57x5Cc/eVj1shs33HBDkuT4448fcyUAAAAsBpdffvktrbVV4/jeoww6bkhyQ2vtc93632QQdHy7qo6dc+nKzd3zNyY5Yc7rj++2PUhr7Zwk5yTJmjVr2qZNm0ZVP51169YlSc4+++wxVwIAAMBiUFXXjet7j+zSldbaTUmur6ondZtOSfKVJOcnOa3bdlqS87rl85O8trv7yrOTbDM/BwAAALA/RjmiI0nOTLKhqg5J8vUkr88gXPlwVb0hyXVJXt3t+7EkL01ybZK7u30BAAAA9tlIg47W2hVJ1szz1Cnz7NuSvHGU9QAAAAD9Nsq7rgAAAAAsKEEHAAAA0BuCDgAAAKA3BB0AAABAbwg6AAAAgN4QdAAAAAC9IegAAAAAekPQAQAAAPSGoAMAAADoDUEHAAAA0BuCDgAAAKA3BB0AAABAbwg6AAAAgN4QdAAAAAC9IegAAAAAekPQAQAAAPSGoAMAAADoDUEHAAAA0BuCDgAAAKA3BB0AAABAbwg6AAAAgN4QdAAAAAC9IegAAAAAekPQAQAAAPSGoAMAAADoDUEHAAAA0BuCDgAAAKA3BB0AAABAbwg6AAAAgN4QdAAAAAC9IegAAAAAekPQAQAAAPSGoAMAAADoDUEHAAAA0BuCDgAAAKA3BB0AAABAbwg6AAAAgN4QdAAAAAC9IegAAAAAekPQAQAAAPSGoAMAAADoDUEHAAAA0BuCDgAAAKA3BB0AAABAbwg6AAAAgN4QdAAAAAC9MdKgo6q+UVVXVtUVVbWp23ZUVX2iqr7WPR7Zba+q+oOquraqNlfV00dZGwAAANA/CzGi44Wttae11tZ0629OsrG1dlKSjd16krwkyUnd1+lJ1i9AbQAAAECPjOPSlZcnObdbPjfJK+Zs/4s28NkkR1TVsWOoDwAAAFikRh10tCQXV9XlVXV6t+2Y1tq3uuWbkhzTLR+X5Po5r72h2/YgVXV6VW2qqk1btmwZVd0AAADAIrR8xO0/r7V2Y1U9Nsknquqrc59srbWqavvTYGvtnCTnJMmaNWv267UAAABAv410REdr7cbu8eYkH0nyzCTfnr0kpXu8udv9xiQnzHn58d02AAAAgH0ysqCjqg6rqkfNLic5NclVSc5Pclq322lJzuuWz0/y2u7uK89Osm3OJS4AAAAAezXKS1eOSfKRqpr9Pn/VWruwqv4pyYer6g1Jrkvy6m7/jyV5aZJrk9yd5PUjrA0AAADooWpt8U5zsWbNmrZp06ZxlzF069evz/T09LjL+J7ZWlavXj3mSh5s9erVOeOMM8ZdBgAAALuoqstba2vG8b1HPRkpD8P09HSuvfrqfN/Ko8ZdSpLkkAcGYdj9M98ecyU7Xbft1nGXAAAAwAQSdEyo71t5VH7z+S8edxkT622XXTTuEgAAAJhAI73rCgAAAMBCEnQAAAAAvSHoAAAAAHpD0AEAAAD0hqADAAAA6A1BBwAAANAbgg4AAACgNwQdAAAAQG8IOgAAAIDeEHQAAAAAvSHoAAAAAHpD0AEAAAD0hqADAAAA6A1BBwAAANAbgg4AAACgNwQdAAAAQG8IOgAAAIDeEHQAAAAAvSHoAAAAAHpD0AEAAAD0hqADAAAA6A1BBwAAANAbgg4AAACgNwQdAAAAQG8IOgAAAIDeEHQAAAAAvSHoAAAAAHpD0AEAAAD0hqADAAAA6A1BBwAAANAbgg4AAACgNwQdAAAAQG8IOgAAAIDeEHQAAAAAvSHoAAAAAHpD0AEAAAD0hqADAAAA6A1BBwAAANAbgg4AAACgNwQdAAAAQG8IOgAAAIDeEHQAAAAAvSHoAAAAAHpj5EFHVR1UVV+sqgu69SdU1eeq6tqq+lBVHdJtP7Rbv7Z7/sRR1wYAAAD0y0KM6PjlJFfPWX9Hkt9rrT0xyW1J3tBtf0OS27rtv9ftBwAAALDPRhp0VNXxSX4qyZ9165Xk5CR/0+1ybpJXdMsv79bTPX9Ktz8AAADAPhn1iI7fT/Kfkuzo1o9OcntrbXu3fkOS47rl45JcnyTd89u6/R+kqk6vqk1VtWnLli0jLB0AAABYbEYWdFTVy5Lc3Fq7fJjtttbOaa2taa2tWbVq1TCbBgAAABa55SNs+7lJfqaqXprkEUkeneRdSY6oquXdqI3jk9zY7X9jkhOS3FBVy5OsTLJ1hPUBAAAAPTOyER2ttV9vrR3fWjsxyc8luaS1tjbJJ5O8stvttCTndcvnd+vpnr+ktdZGVR8AAADQPwtx15Vd/VqSX62qazOYg+M93fb3JDm62/6rSd48htoAAACARWyUl658T2vt0iSXdstfT/LMefb5bpJXLUQ9AAAAQD+NY0QHAAAAwEgIOgAAAIDeEHQAAAAAvSHoAAAAAHpD0AEAAAD0hqADAAAA6A1BBwAAANAbgg4AAACgNwQdAAAAQG8IOgAAAIDeEHQAAAAAvbF83AUAANBf69evz/T09AG3MzMzkySZmpo64LaSZPXq1TnjjDOG0hYAk0XQAQDAxLvnnnvGXQIAi4SgAwCAkRnWqIl169YlSc4+++yhtAdAf5mjAwAAAOgNQQcAAADQG4IOAAAAoDcEHQAAAEBvCDoAAACA3hB0AAAAAL0h6AAAAAB6Q9ABAAAA9IagAwAAAOgNQQcAAADQG4IOAAAAoDcEHQAAAEBvCDoAAACA3hB0AAAAAL0h6AAAAAB6Q9ABAAAA9IagAwAAAOgNQQcAAADQG4IOAAAAoDcEHQAAAEBvCDoAAACA3hB0AAAAAL0h6AAAAAB6Q9ABAAAA9MbycRfAQ83MzOTu27flbZddNO5SJtZ1t9+aR+aBcZcBAADAhDGiAwAAAOgNIzom0NTUVO7PQfnN57943KVMrLdddlEOnjpm3GUAAAAwYYzoAAAAAHpjryM6quoVSZ6Y5MrWmkkjAAAAgIm1xxEdVfXHSf5DkqOTvLWq/p8FqQoAAADgYdjbiI4fT/LU1toDVfXIJJcleevoywIAAADYf3ubo+O+1toDSdJauztJ7WvDVfWIqvp8VX2pqr5cVf+l2/6EqvpcVV1bVR+qqkO67Yd269d2z5/4MH8mAAAAYInaW9Dx5Kra3H1dOWf9yqravJfX3pvk5NbaU5M8LclPVtWzk7wjye+11p6Y5LYkb+j2f0OS27rtv9ftBwAAALDP9nbpyv/2cBturbUkd3arB3dfLcnJSX6+235ukrOSrE/y8m45Sf4myburqrp2AAAAAPZqj0FHa+26+bZX1bIkr0ky7/Nz9jsoyeUZ3LXlj5JMJ7m9tba92+WGJMd1y8club77vturalsGk6Desk8/CQAAALDk7e2uK4+uql+vqndX1ak1cGaSryd59d4ab6090Fp7WpLjkzwzyZMPtOCqOr2qNlXVpi1bthxocwAAAECP7G2OjvcneVKSK5P8YpJPJnllkle01l6+r9+ktXZ799rnJDmiqmZHkhyf5MZu+cYkJyRJ9/zKJFvnaeuc1tqa1tqaVatW7WsJAAAAwBKwtzk6vr+19kNJUlV/luRbSR7fWvvu3hquqlVJ7m+t3V5VK5L87xlMMDoblnwwyWlJzutecn63/pnu+UvMzwEAAADsj70FHffPLrTWHqiqG/Yl5Ogcm+Tcbp6OZUk+3Fq7oKq+kuSDVfW2JF9M8p5u//ckeX9VXZvk1iQ/tz8/CAAAAMDego6nVtV3klS3vmLOemutPXp3L2ytbU7yI/Ns/3oG83Xsuv27SV61r4UDAAAA7Gpvd105aKEKAQAAADhQeww6quoRSf5dBreH3ZzkvXNuDQsAAAAwUfZ215Vzk6zJ4K4rL03yzpFXBAAAAPAw7W2OjqfMuevKe5J8fvQlAQAAADw8exvRMfeuKy5ZAQAAACbavt51JRncaWWf77oCAAAAsNDcdQUAAADojb1dugIAAACwaAg6AAAAgN4QdAAAAAC9IegAAAAAekPQAQAAAPSGoAMAAADoDUEHAAAA0BuCDgAAAKA3BB0AAABAbwg6AAAAgN4QdAAAAAC9IegAAAAAekPQAQAAAPSGoAMAAADoDUEHAAAA0BuCDgAAAKA3BB0AAABAbwg6AAAAgN4QdAAAAAC9IegAAAAAekPQAQAAAPSGoAMAAADoDUEHAAAA0BuCDgAAAKA3BB0AAABAbwg6AAAAgN4QdAAAAAC9IegAAAAAekPQAQAAAPSGoAMAAADojeXjLoD5Xbft1rztsovGXUaS5KY770iSPO7wR425kp2u23Zrnjh1zLjLAAAAYMIIOibQ6tWrx13Cg9w3fWeS5OAJChaeOHXMxPUTAAAA4yfomEBnnHHGuEt4kHXr1iVJzj777DFXAgAAAHtmjg4AAACgNwQdAAAAQG8IOgAAAIDeEHQAAAAAvWEyUgAAHmL9+vWZnp4edxnfM1vL7CTpk2L16tUTN5E8wFIn6AAA4CGmp6fz1as356gjx13JQNsxeLz5ps3jLWSOW28bdwUAzGdkQUdVnZDkL5Ick6QlOae19q6qOirJh5KcmOQbSV7dWrutqirJu5K8NMndSV7XWvvCqOoDAGDPjjoyedkprnTenQs27hh3CQDMY5T/c21P8qbW2lOSPDvJG6vqKUnenGRja+2kJBu79SR5SZKTuq/Tk6wfYW0AAABAD40s6GitfWt2REZr7Y4kVyc5LsnLk5zb7XZukld0yy9P8hdt4LNJjqiqY0dVHwAAANA/CzIWsapOTPIjST6X5JjW2re6p27K4NKWZBCCXD/nZTd023Zt6/Sq2lRVm7Zs2TK6ogEAAIBFZ+RBR1UdnuRvk/xKa+07c59rrbUM5u/YZ621c1pra1pra1atWjXESgEAAIDFbqRBR1UdnEHIsaG19nfd5m/PXpLSPd7cbb8xyQlzXn58tw0AAABgn4ws6OjuovKeJFe31n53zlPnJzmtWz4tyXlztr+2Bp6dZNucS1wAAAAA9mpkt5dN8twkv5Dkyqq6otv2G0nenuTDVfWGJNcleXX33McyuLXstRncXvb1I6wNAAAA6KGRBR2ttU8lqd08fco8+7ckbxxVPQAAAED/LchdVwAAAAAWgqADAAAA6A1BBwAAANAbgg4AAACgNwQdAAAAQG8IOgAAAIDeEHQAAAAAvSHoAAAAAHpD0AEAAAD0hqADAAAA6A1BBwAAANAbgg4AAACgNwQdAAAAQG8IOgAAAIDeEHQAAAAAvSHoAAAAAHpD0AEAAAD0hqADAAAA6A1BBwAAANAbgg4AAACgNwQdAAAAQG8IOgAAAIDeEHQAAAAAvSHoAAAAAHpD0AEAAEDvbd26NW9605ty6623jrsURkzQAQAAQO9t2LAhV111VTZs2DDuUhgxQQcAAAC9tnXr1lx88cVpreWiiy4yqqPnlo+7AID9tX79+kxPTx9wOzMzM0mSqampA25r9erVOeOMMw64HQAAhm/Dhg3ZsWNHkmTHjh3ZsGFDzjzzzDFXxagY0QEsWffcc0/uueeecZcBAMCIXXLJJdm+fXuSZPv27dm4ceOYK2KUjOgAFp1hjZxYt25dkuTss88eSnsAAEymk08+ORdeeGG2b9+e5cuX55RTThl3SYyQoKPHhjW8f7aN2ZPCA2WIPwAAsJDWrl2biy++OEmybNmyrF27dswVMUouXWGvVqxYkRUrVoy7DAAAgIfl6KOPzqmnnpqqyotf/OIcddRR4y6JETKio8eMmgAAABhYu3ZtrrvuOqM5lgBBB7AghnUp1TAN+7KsYXF5FwDA8B199NF55zvfOe4yWACCDmBBTE9P52tfuSqPX3nouEv5nkMeuC9Jcu+NXxtzJTt9c9u94y6BHti6dWt+53d+J295y1sMzQUAlhxBB7BgHr/y0Lz5ed837jIm2ts/dd24S6AHNmzYkKuuuiobNmzImWeeOe5yAAAWlMlIAaBHtm7dmosvvjittVx00UW59dZbx10SAMCCEnQAQI9s2LAhO3bsSJLs2LEjGzZsGHNFAAALS9ABAD1yySWXZPv27UmS7du3Z+PGjWOuCABgYZmjA1gQMzMzuWvbveag2Itvbrs3h9XMuMtgETv55JNz4YUXZvv27Vm+fHlOOeWUcZcEALCgjOgAgB5Zu3Ztli0b/Pe+bNmyrF27dswVAQAsLCM6gAUxNTWVe9td7rqyF2//1HU5dGpq3GWwiB199NE59dRT89GPfjQvfvGL3V4WAFhyBB0A0DNr167NddddZzQHALAkCToAoGeOPvrovPOd7xx3GQAAY2GODgAAAKA3RhZ0VNV7q+rmqrpqzrajquoTVfW17vHIbntV1R9U1bVVtbmqnj6qugAAAID+GuWlK+9L8u4kfzFn25uTbGytvb2q3tyt/1qSlyQ5qft6VpL13SPQI9+csNvL3nzXfUmSxx52yJgr2emb2+7NSceNuwoAAFi8RhZ0tNb+sapO3GXzy5O8oFs+N8mlGQQdL0/yF621luSzVXVEVR3bWvvWqOoDFtbq1avHXcJD3Dc9nSQ59LjJqe2k4yazrwAAYLFY6MlIj5kTXtyU5Jhu+bgk18/Z74Zum6ADeuKMM84YdwkPsW7duiTJ2WefPeZKAACAYRnbZKTd6I22v6+rqtOralNVbdqyZcsIKgMAAAAWq4UOOr5dVccmSfd4c7f9xiQnzNnv+G7bQ7TWzmmtrWmtrVm1atVIiwUAAAAWl4UOOs5Pclq3fFqS8+Zsf21395VnJ9lmfg4AAABgf41sjo6q+kAGE48+pqpuSPKfk7w9yYer6g1Jrkvy6m73jyV5aZJrk9yd5PWjqgtY/NavX5/pbiLRAzHbxuxcHQdi9erVEzkPCQAALDWjvOvKa3bz1Cnz7NuSvHFUtQDMZ8WKFeMuAQAAGLKFvusKwAEzcgIAANidsd11BQAAAGDYBB0AAABAbwg6AAAAgN4QdAAAAAC9IegAAAAAesNdVwCAJWf9+vWZnp4+4HZmZmaSJFNTUwfc1urVq91VCgCGQNABABNgWCfeiZPvhXTPPfeMuwQAYBeCDgDoGSffezes8GbdunVJkrPPPnso7QEAB07QAQATYJijJpx8MwwzMzPZti25YOOOcZcysbbelmzfMTPuMhgTl8AtHH3N/hJ0AAAAS8KwLxMcxgi62TaG0dbMzMxQfr6+nsQb8bh0CDoAAHiIqampLF92S152ipv07c4FG3fksY878E+GWTjT09P5ytWb86ijhtDYQckhhx94M9vb4PGQw+864LYeyF25/tu3HFAbd9x6wGUMncsN2V+CDgAAYMl41FHJs15S4y5jYn3u421obQ1zBM0wzNYyG3hMir6OoBknQQcAHIBJO4hLJvNAzkEcwNIzPT2dL391c1YcPe5KBu7rMpyvb9k83kLmuGfruCvoJ0EHAByA6enpXHP15jxu5eR8OrjsgcGR3LaZK8dcycBN24b36SAAi8uKo5MnvcwlcLtzzQUmfB4FQQcAHKDHray87gUHj7uMifW+S+8fSjtGz+wbo2eWrmFPtJm4OwWwOAk6AIBFYXb0zKojxl3JTtV9EHfrtyZjGPSW28ddAX3h7hTAYiboAAAWjVVHJK862eHL7vz1JdvHXQJjNMxRE+5OASxmLpYCAAAAekPQAQAAAPSGsZ8AADBGJtrdN8OY1HRmZiZ3bEs+93F3g9qdO25NZh6YGXcZcEAEHQAAMEbT09O5+urNOeLIcVey045uot1v3TQZE+3eftu4KwAWE0EHAACM2RFHJi88tcZdxsT65MXDGYExNTWVBw66Jc96ib7enc99vGXqmAO/rTCMkzk6AAAAgN4QdAAAAAC9IegAAAAAesMcHSxpw5rlfGZmJvfcc88QKhquFStWZGrqwK+xHMYs5wDA/GZmZrJt2/Dmoeij229L2g53AgH2jaCDJe2yyy7LLbfcMu4yRuauu+4ays83MzMj6AAAABYFQQdL2sqVK4cyEuPee+/Njtn7sE2QZcuW5dBDDz3gdlauXDmEagCA+UxNTWXbtsn64OXOOwaPhz9qvHXMNYxRqiysmZmZ3P2d5JoLJu84eVLcvTWZud9opWETdLCkrV+/ftwlwEQb5uVdyfAOUl1OBfTJ6tWrx13CQ0zfNXjvP/Zxk1HbsY+bzH4CJpOgA4CRm8Q5bIZlZmYmd9ze8r5L7x93KRPrpttb7opPqxajW29LLtg4GZ/EfqcbYfDoCRphcOttyWMfd+DtTGJwu27duiTJ2WefPeZKWMympqby3YNvyZNe5h4Yu3PNBTsytcpopWETdACwW8M6+HbADIvPpH16fkc3wuCxEzLCIBmEHJPWTwAIOgB6aViXnAzLbC2zgcekGMYlMFNTU7lmwq6tv/XOwZ0bjjq8xlxJp4Zz2dLMzEy+sy3560u2D6Gofrr59uS7bTijZyZtlIHAFIB9JegA6KHp6en881c257iVk3Giu/yBwYn3XTdeOeZKdrpx23Bu4ziJn+be0gVLK6cmo7aVU8Prp/u3D07mJ8X2BwaPyw8abx2z7pcBAYCgA6CvjltZOfP5B37Xnb76w8vuHUo7k/apd9LfT76f//znT9RIpWTnaKVJCrwmqRYW1jBH8w1zJN6kTSB9x63J5z4+nLB7GO7u5p955ITMP3PHrUmOGV5792ydnLuu3Ltt8HjoBN1Q8J6tSVaNu4r+EXQAAIvCME+UJu3yrmTyTgZZ2lasWDHuEkZiEoPA6TsH70UnHDMhtR0zvH4aVjszMzNDmdj8ge8O2mh14H/fK1asGM7d5FZN5t/lYifoAOihmZmZ3LmtDW3UQh/duK3l8HInEA5MX08GWZwEZXs3iX3U11F4yfD6exJvd9/XcHrYfT0ugg4AYMnp48HppBrWQfOwJzXu60kK9JF/q3s3zIBiGKNnhtHGgRB0APTQ1NRU7mpbzdGxB3942b05bBhDTofEtfWwZ0bPMGkmMcTznr10TU9PZ/NXN6cec4ANHdJ9HaBxz4Ij6ADoqRsn6NKVW+4a/Hf3mMMm4y4wyaB/fuC4cVcxGk4ImSROumDPvGczDOO+VGRXNeYJXwUdAD00aZNa3dR9WnXYcZNT1w8cN1n95GQQYHHxvs3EuT9pt4y7iM6Yb3cu6ADooUk7+OrzRGsAAOM2rFuwD22Oju3m6ABgQk3i9ceJa5ABAOaa1DvcjIugA4CRc/0xAMDkG+YHSR/4wAeG1tb+EnQAsFtGTQAAsNgsG3cBc1XVT1bVNVV1bVW9edz1AAAAAIvLxAQdVXVQkj9K8pIkT0nymqp6ynirAgAAABaTiQk6kjwzybWtta+31u5L8sEkLx9zTQAAAMAiMklBx3FJrp+zfkO3DQAAAGCfLLrJSKvq9CSnd6t3VtU146xnCXlMklvGXcQSor8Xjr5eOPp6YenvhaOvF46+Xlj6e+Ho64WjrxfOk8b1jScp6LgxyQlz1o/vtj1Ia+2cJOcsVFEMVNWm1tqacdexVOjvhaOvF46+Xlj6e+Ho64WjrxeW/l44+nrh6OuFU1WbxvW9J+nSlX9KclJVPaGqDknyc0nOH3NNAAAAwCIyMSM6Wmvbq+r/TnJRkoOSvLe19uUxlwUAAAAsIhMTdCRJa+1jST427jqYl8uFFpb+Xjj6euHo64WlvxeOvl44+nph6e+Fo68Xjr5eOGPr62qtjet7AwAAAAzVJM3RAQAAAHBABB3ss6q6tKr2OENxVX2jqh4zz/bXVdW7R1ddf1XV+6rqld3yXn8HDFTViVV11QG2cWhV/X1VXVFV/2ZYtS0Fu+v/qvqVqnrknPWPVdURC1rcBBjG3+eo7O59nIevql5QVReMu46+qKo7x13DUuI9Yfiq6qyqWjfuOpYa/T6/qvqlqrq6qjYs0Pd7XVVNzVn/s6p6yrC/z0TN0QEwYX4kSVprTxtzHX3yK0n+MsndSdJae+lYq1mEqmp5a2377tYZD78HABapf5/kRa21G/a245D+r3tdkquSzCRJa+0XD7C9eRnRsQRU1f+oqsur6stVdXq37c6q+m/dtr+vqmd2owW+XlU/0+2zoqo+2CV8H0myYk6b66tqU/f6/7LLtzyzqr5QVVdW1ZMX7iedPHvo+9/rtm2sqlXd9qdV1WeranNVfaSqjtxL26/p+viqqnrHQvw8i9DyqtrQ/Q3/TVU9sqp+q6r+qeu3c6qqku+l2V/p+v+DVfXYDE7If7Qb0bF6vD/KorRr//9Skqkkn6yqTyZL/pPC+f4+v9cfVbWmqi7tls+qqvdX1aeTvH+e9ROr6pLu73djVT2+e933RoR163d2j8uq6o+r6qtV9YluZM0r59T2kPfxqvqJ7t/CFVX1xap61MJ003hV1Wu7fv1SVb1/l+deV1XnV9UlSTbu8txPdv37hST/eiFrXux27fOqekJVfab7e3zbLvv+Wrf9S1X19nHVPEnm6b+frqrPdf9u/76qjun2O6uqzq2qy6rquqr611X1X7v+vLCqDp7T7H/qtn++qp7YvX5VVf1t93/qP1XVc7vt51XVa7vl/6sW6FPiSVdVb6mqf66qTyV5UrdtddfXl3e/h9n322NqcCz4pe7rx8Za/CK2m37/0e7fyBU1OB+ayBGWC6Gq/iTJ9yf5eFW9qQbnLptrcE7yw90+8x2D7PW9o+Y55u6ONdYk2dD1/4qaM2K9dnN+U4Pzp9/u/j18dvZ9bI9aa756/pXkqO5xRQbp2dFJWpKXdNs/kuTiJAcneWqSK7rtv5rBbX6T5IeTbE+yZpc2D0pyaZIf7ta/keTMbvnfJ/mzbvl1Sd497r6YoL5f223/rdl+SbI5yU90y/9vkt/vlt+X5JXd8qXdm8NUkm8mWZXByKxLkrxi3D/vJH0lObHr6+d26+9Nsm72d9Jte3+Sn+6WZ5Ic2i0f0T2+IMkF4/5ZFuPXHvr/G0keM2e/B60vla996Z/u3/ql3fJZSS5PsmI36/8zyWnd8r9N8j+65e+9f3Trd3aPr8zgLmfLkjwuyW1z3md29z7+P+fUe3iS5ePuxwX4Pf1gkn+e8zs5apfnX5fkhnm2PyLJ9UlOSlJJPuy95OH3eZLzk7y2W3/jnL/jlyT5X0keOd/vZyl+7ab/jszOGxD8YpJ3dstnJflUdh7/3Z0HHxu+olv+RpK3dMuvnf1bTvJXSZ7XLT8+ydXd8jFJrk3y/K4Wv5fkGUmuTPLIJI/u+mddBgHpSd0+z0pySbf8oSS/0i0flGTluH+Gxfi1h36/Kslzun3enuSqcdc65n76RpLHJPnDJP+523Zydp4TnpWHHoPsy3vH7o65L013Tjl3PXs4v8ngmGn29f81yW/u7ecyomNp+KWq+lKSzyY5IYMDr/uSXNg9f2WSf2it3d8tn9ht//EMPtFOa21zBifis17dfUr1xQz+U517XdXfdY+Xz2lrqZqv73dk8B9YMujf51XVygxOrv+h235uBv2/Oz+awQnQljYYPrZhL/svVde31j7dLf9lkucleWH3ydaVGbyJ/2D3/OYM0uX/M4NQjwM3X/+z0/72z/mttXt2s/6cDE46ksHBxN7ael6Sv26t7Wit3ZTkk7s8P9/7+KeT/G4NRuYc0ZbGZRonZ9BPtyRJa+3Wefb5xDzbn5zkX1prX2uDo7K/HHGdfTJfnz83yQe65+eOqnlRkj9vrd09Z9+lbr7+Oz7JRd3/e/8xO//fS5KPzzn+OygPPjY8cc5+H5jz+Jxu+UVJ3l1VV2QQRj26qg5vrX07gw9yPpnkTX4vSQahz0daa3e31r6TQX89IsmPJfnrrg//e5Jju/1PTrI+SVprD7TWti18yb0wX78nyaNaa5/plv9q/pcuSc9L9x7bWrskydFV9ejuuV2PQfblvWN3x9y7s6fzm/uSzM51tU/nmObo6LmqekEG/xE9p7V2dw2GQT8iyf3dwVcyOPG+N0laazuqao9/F1X1hAzS0B9trd1WVe/r2px1b/f4QJbw39ge+n5X7vE8Orv2bUvyxxmkyNdX1VnZ+Tv5qQzeTH86yVuq6ocWrMr+mq//2Wm+/tmenZeV7vp+cdde1ufzvfaqalmSQ/axtoe8j7fW3l5VH03y0iSfrqoXt9a+uo/t9dm+/B44cN4/Hr4/TPK7rbXzu2OTs+Y8N/f4b9djw7nHcG2e5WVJnt1a++483/OHkmzN4BNa5rcsye3NPGAsDrv+X7fH946qekR2f8z9cMz9Hvt0jmlER/+tTHJbd6L95CTP3o/X/mOSn0+SqvpXGVy+kgyGft2VZFt3fdRLhlhvn+yu75dlMGw8GfTvp7qk/raqen63/ReS/EN27/NJfqKqHlNVByV5zV72X6oeX1Wznzz9fAbD7JLklqo6PN3voTsBPKG19skkv5bB7+7whS62h+br/zuSLIm5HfbBfP3zjQyG2ibJz+5HW/8ryc91y2uTXNYtz23vZzIYZpoMRmf8bA3m6jgmg8u09qiqVrfWrmytvSPJP2UwaqHvLknyqqo6Okmq6qh9fN1Xk5xYO+f2ec0oiuup+fr803nw3/esTyR5fXV3ctqP30+fzdd/K5Pc2D1/2sNs99/MeZz9JPziJGfO7lBVT+sen5nBseGPJFnXfUC21P1jkld08xE8KoMPVe5O8i9V9aok6eYveGq3/8YkZ3TbD+pG/rL/5uv3JLmjqp7VLf/c/C9dki5L9x7bhaK3dCNhHo7ZUONBx9yd3R0LDvX8Zsl+2r6EXJjk31XV1UmuyeASin21Psmfd6+9OoNhQmmtfamqvpjBgdz1GRyA8FC76/u7kjyzqn4zyc3ZefBwWpI/6Q7Yvp7k9btruLX2rap6cwbDQivJR1tr543mx1jUrknyxqp6b5KvZPA3fWQG12belMHJWjIYcveX3YFEJfmD1trtNZinlIdvvv6/L8mFVTXTWnvhWKsbv/n65/NJ3lNVb83gmtV9dWYG79f/McmW7Hz/+NMk53WX0F2YnZ/I/G2SU7rve32SLyTZ29DoX6mqF2bwac2Xk3x8P+pblFprX66q307yD1X1QJIvVtXfZfAJ1W/tun9VfSzJL7bWZmowAfVHq+ruDA4eBXz7YL4+T/LLSf6qqn4tyXlz9r2wO7neVFX3ZTDvzG+MoeyJsZv+OyuDyyNuyyAIeTjBw5FVtTmDT3Fng7tfSvJH3fblSf6xqn45g/ed13f/Dt6U5L1VdfKcT2OXnNbaF6rqQ0m+lMGx3+zxx9ok67tjwoOTfLDb55eTnFNVb8jg0+szsjNgYh/tod/fkORPq2pHBifSLg0aOCuDf6+bMwjiHm4wmu44+k/z0GPuZDB/2J9U1T3ZeSnc0M9vagm/58BYVNWdrTWjBYCx6q6lv7P75PfzGUw0etO46wKAUZr9/69bfnOSY1trvzzmshgyIzoAYGm6oKqOyGDejrcKOQBYIn6qqn49g3Ph6zK4gxY9Y0QHAAAA0BsmIwUAAAB6Q9ABAAAA9IagAwAAAOgNQQcAAADQG4IOAOAhquotVfXlqtpcVVdU1bP2sO/7quqV3fKlVbVmCN//fVV1Y1Ud2q0/pqq+caDtAgD9J+gAAB6kqp6T5GVJnt5a++EkL0py/Yi+10F7ePqBJP92FN8XAOgvQQcAsKtjk9zSWrs3SVprt7TWZqrqGVX1D1V1eVVdVFXH7qmRqjq1qj5TVV+oqr+uqsO77d+oqndU1ReSvGoPTfx+kv9QVct3affwqtrYtXtlVb28235iVX21Gw3yz1W1oapeVFWfrqqvVdUzu/0Oq6r3VtXnq+qLs68HAPpB0AEA7OriJCd0YcEfV9VPVNXBSf4wyStba89I8t4kv727BqrqMUl+M8mLWmtPT7Ipya/O2WVra+3prbUP7qGObyb5VJJf2GX7d5P8H127L0zyzqqq7rknJnlnkid3Xz+f5HlJ1iX5jW6ftyS5pLX2zO71/62qDttDHQDAIrJ877sAAEtJa+3OqnpGkudnEAR8KMnbkvyrJJ/oMoWDknxrD808O8lTkny62/+QJJ+Z8/yH9rGc/y/JeUk+OmdbJfmdqvrxJDuSHJfkmO65f2mtXZkkVfXlJBtba62qrkxyYrfPqUl+pqrWdeuPSPL4JFfvY00AwAQTdAAAD9FaeyDJpUku7UKCNyb5cmvtOfvYRCX5RGvtNbt5/q59rONrVXVFklfP2bw2yaokz2it3d9NUvqI7rl75+y3Y876juw87qkkP9tau2ZfagAAFheXrgAAD1JVT6qqk+ZseloGox1WdROVpqoOrqof3EMzn03y3Kp6Yrf/YVX1Aw+zpN/O4NKTWSuT3NyFHC9M8n372d5FSc6cvdylqn7kYdYFAEwgQQcAsKvDk5xbVV+pqs0ZXILyW0lemeQdVfWlJFck+bHdNdBa25LkdUk+0LXxmQzmzNhvrbUvJ/nCnE0bkqzpRpq8NslX97PJtyY5OMnm7vKWtz6cugCAyVSttXHXAAAAADAURnQAAAAAvWEyUgBgbKrqj5I8d5fN72qt/fk46gEAFj+XrgAAAAC94dIVAAAAoDcEHQAAAEBvCDoAAACA3hB0AAAAAL0h6AAAAAB64/8H2m/ImyYJrEAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1296x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(18,6))\n",
    "sns.boxplot(x=df['Seller_Name'], y=df['PRP'])\n",
    "plt.xlim([-1, 10])\n",
    "plt.ylim([0, 600])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Splitting data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df.drop(columns=['Seller_Name', 'Model_Name', 'ERP', 'PRP'])\n",
    "y = df[['PRP']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(166, 6) (42, 6) (166, 1) (42, 1)\n"
     ]
    }
   ],
   "source": [
    "train_X, test_X, train_y, test_y = sms.train_test_split(X, y, test_size=0.2, random_state=25)\n",
    "print(train_X.shape, test_X.shape, train_y.shape, test_y.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Data Normalization(Scaling)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**RobustScaler** is one of the preferred scalers mainly to **avoid outliers impact** if there is a chance of being"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "scaler_rb=sp.RobustScaler()\n",
    "scaled_tra=scaler_rb.fit_transform(train_X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>MYCT</th>\n",
       "      <th>MNM</th>\n",
       "      <th>MXM</th>\n",
       "      <th>CACH</th>\n",
       "      <th>CHMIN</th>\n",
       "      <th>CHMAX</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.419306</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.75</td>\n",
       "      <td>2.8</td>\n",
       "      <td>1.263158</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.161388</td>\n",
       "      <td>-0.410667</td>\n",
       "      <td>-0.291667</td>\n",
       "      <td>-0.25</td>\n",
       "      <td>-0.2</td>\n",
       "      <td>0.842105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.346908</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-0.333333</td>\n",
       "      <td>-0.25</td>\n",
       "      <td>0.2</td>\n",
       "      <td>-0.105263</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       MYCT       MNM       MXM  CACH  CHMIN     CHMAX\n",
       "0 -0.419306  0.666667  0.000000  0.75    2.8  1.263158\n",
       "1  1.161388 -0.410667 -0.291667 -0.25   -0.2  0.842105\n",
       "2 -0.346908  0.000000 -0.333333 -0.25    0.2 -0.105263"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "names = list(train_X.columns)\n",
    "train_X=pd.DataFrame(scaled_tra, columns=names)\n",
    "train_X.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>MYCT</th>\n",
       "      <th>MNM</th>\n",
       "      <th>MXM</th>\n",
       "      <th>CACH</th>\n",
       "      <th>CHMIN</th>\n",
       "      <th>CHMAX</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.419306</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>1.75</td>\n",
       "      <td>0.4</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-0.015083</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-0.333333</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.558069</td>\n",
       "      <td>-0.333333</td>\n",
       "      <td>-0.333333</td>\n",
       "      <td>-0.25</td>\n",
       "      <td>-0.2</td>\n",
       "      <td>-0.210526</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       MYCT       MNM       MXM  CACH  CHMIN     CHMAX\n",
       "0 -0.419306  2.000000  0.666667  1.75    0.4  0.000000\n",
       "1 -0.015083  0.000000 -0.333333  0.00    0.2  0.000000\n",
       "2  0.558069 -0.333333 -0.333333 -0.25   -0.2 -0.210526"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scaled_test=scaler_rb.transform(test_X)\n",
    "test_X=pd.DataFrame(data=scaled_test, columns=names)\n",
    "test_X.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>MAE</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RMSE</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>R2_score</th>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: []\n",
       "Index: [MAE, RMSE, R2_score]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# to merge all scores and metrics results altogether\n",
    "performance=pd.DataFrame(data=None, index=['MAE','RMSE','R2_score'])\n",
    "performance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# creating function to evaluate model at one run\n",
    "def evaluate(model,model_name):\n",
    "    performance[model_name]=[sm.mean_absolute_error(test_y, model.predict(test_X)),\n",
    "    sm.mean_squared_error(test_y, model.predict(test_X))**0.5,\n",
    "    sm.r2_score(test_y, model.predict(test_X)) ]\n",
    "pd.options.display.float_format = '{:.3f}'.format"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Modelling"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Multiple linear regression model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>MLinear</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>MAE</th>\n",
       "      <td>32.231</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RMSE</th>\n",
       "      <td>46.875</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>R2_score</th>\n",
       "      <td>0.916</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          MLinear\n",
       "MAE        32.231\n",
       "RMSE       46.875\n",
       "R2_score    0.916"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lm=slm.LinearRegression()\n",
    "lm.fit(train_X, train_y)\n",
    "evaluate(lm, 'MLinear')\n",
    "performance[['MLinear']]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Lasso regression model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1.00000000e+01, 1.20679264e+01, 1.45634848e+01, 1.75751062e+01,\n",
       "       2.12095089e+01, 2.55954792e+01, 3.08884360e+01, 3.72759372e+01,\n",
       "       4.49843267e+01, 5.42867544e+01, 6.55128557e+01, 7.90604321e+01,\n",
       "       9.54095476e+01, 1.15139540e+02, 1.38949549e+02, 1.67683294e+02,\n",
       "       2.02358965e+02, 2.44205309e+02, 2.94705170e+02, 3.55648031e+02,\n",
       "       4.29193426e+02, 5.17947468e+02, 6.25055193e+02, 7.54312006e+02,\n",
       "       9.10298178e+02, 1.09854114e+03, 1.32571137e+03, 1.59985872e+03,\n",
       "       1.93069773e+03, 2.32995181e+03, 2.81176870e+03, 3.39322177e+03,\n",
       "       4.09491506e+03, 4.94171336e+03, 5.96362332e+03, 7.19685673e+03,\n",
       "       8.68511374e+03, 1.04811313e+04, 1.26485522e+04, 1.52641797e+04,\n",
       "       1.84206997e+04, 2.22299648e+04, 2.68269580e+04, 3.23745754e+04,\n",
       "       3.90693994e+04, 4.71486636e+04, 5.68986603e+04, 6.86648845e+04,\n",
       "       8.28642773e+04, 1.00000000e+05])"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Return numbers spaced evenly on a log scale, it means that creates 50 numbers between 1 and 5\n",
    "np.logspace(start=1, stop=5, num=50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "10.0"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lasso=slm.LassoCV(alphas=np.logspace(start=1, stop=5, num=50),cv=5)\n",
    "out_lasso = lasso.fit(train_X, train_y)\n",
    "out_lasso.alpha_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Lasso</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>MAE</th>\n",
       "      <td>32.910</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RMSE</th>\n",
       "      <td>54.377</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>R2_score</th>\n",
       "      <td>0.886</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Lasso\n",
       "MAE      32.910\n",
       "RMSE     54.377\n",
       "R2_score  0.886"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "evaluate(out_lasso, 'Lasso')\n",
    "performance[['Lasso']]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Ridge regression model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "95.40954763499938"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ridge=slm.RidgeCV(alphas=np.logspace(start=1, stop=5, num=50),cv=5)\n",
    "out_ridge=ridge.fit(train_X, train_y)\n",
    "ridge.alpha_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Ridge</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>MAE</th>\n",
       "      <td>32.270</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RMSE</th>\n",
       "      <td>61.360</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>R2_score</th>\n",
       "      <td>0.855</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Ridge\n",
       "MAE      32.270\n",
       "RMSE     61.360\n",
       "R2_score  0.855"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "evaluate(out_ridge, 'Ridge')\n",
    "performance[['Ridge']]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### ElasticNet regression model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "elastic_net=slm.ElasticNetCV(l1_ratio=0.45,alphas=np.logspace(start=1, stop=5, num=50),cv=5)\n",
    "out_elastic_net=elastic_net.fit(train_X, train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10.0\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([-4.2463035 , 17.171761  , 13.40732559, 12.62111071, 10.90057509,\n",
       "       10.06853192])"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(out_elastic_net.alpha_)\n",
    "out_elastic_net.coef_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Elastic_Net</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>MAE</th>\n",
       "      <td>54.753</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RMSE</th>\n",
       "      <td>108.898</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>R2_score</th>\n",
       "      <td>0.544</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Elastic_Net\n",
       "MAE            54.753\n",
       "RMSE          108.898\n",
       "R2_score        0.544"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "evaluate(out_elastic_net, 'Elastic_Net')\n",
    "performance[['Elastic_Net']]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### SVR regression model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We will do **Hyperparameter tuning** to find **otimal best** parameters in some models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'C': 5, 'kernel': 'linear'}"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "parameters = {'kernel':('linear', 'rbf', 'poly', 'sigmoid'), 'C':[1, 5, 10, 50, 100, 500, 1000] }\n",
    "svr = svm.SVR()\n",
    "clf = sms.GridSearchCV(svr, parameters)\n",
    "clf.fit(train_X, train_y)\n",
    "clf.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>SVR</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>MAE</th>\n",
       "      <td>33.552</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RMSE</th>\n",
       "      <td>86.282</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>R2_score</th>\n",
       "      <td>0.714</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            SVR\n",
       "MAE      33.552\n",
       "RMSE     86.282\n",
       "R2_score  0.714"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "evaluate(clf,'SVR')\n",
    "performance[['SVR']]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Ensemble methods"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### RandomForest regression model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'n_estimators': 180}"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "parameters = { 'n_estimators':[i for i in range(150,200,10)] }\n",
    "rfr = ens.RandomForestRegressor(random_state=25)\n",
    "clf3 = sms.GridSearchCV(rfr, parameters)\n",
    "clf3.fit(train_X, train_y)\n",
    "\n",
    "clf3.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>R_forest</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>MAE</th>\n",
       "      <td>16.559</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RMSE</th>\n",
       "      <td>26.583</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>R2_score</th>\n",
       "      <td>0.973</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          R_forest\n",
       "MAE         16.559\n",
       "RMSE        26.583\n",
       "R2_score     0.973"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "evaluate(clf3, 'R_forest')\n",
    "performance[['R_forest']]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### ExtraTree regression model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'n_estimators': 160}"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "parameters = { 'n_estimators':[i for i in range(80,200,10)] }\n",
    "etr = ens.ExtraTreesRegressor(random_state=25)\n",
    "clf4 = sms.GridSearchCV(etr, parameters)\n",
    "clf4.fit(train_X, train_y)\n",
    "\n",
    "clf4.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ExTree</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>MAE</th>\n",
       "      <td>18.076</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RMSE</th>\n",
       "      <td>28.738</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>R2_score</th>\n",
       "      <td>0.968</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          ExTree\n",
       "MAE       18.076\n",
       "RMSE      28.738\n",
       "R2_score   0.968"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "evaluate(clf4, 'ExTree')\n",
    "performance[['ExTree']]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### GradientBoosting regression model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'learning_rate': 0.01, 'n_estimators': 250}"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "parameters = { 'learning_rate':[0.1,0.01,0.001], 'n_estimators':[50,100,150,200,250] }\n",
    "gbr = ens.GradientBoostingRegressor(random_state=25)\n",
    "clf5 = sms.GridSearchCV(gbr, parameters)\n",
    "clf5.fit(train_X, train_y)\n",
    "\n",
    "clf5.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>GradientBR</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>MAE</th>\n",
       "      <td>21.079</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RMSE</th>\n",
       "      <td>31.492</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>R2_score</th>\n",
       "      <td>0.962</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          GradientBR\n",
       "MAE           21.079\n",
       "RMSE          31.492\n",
       "R2_score       0.962"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "evaluate(clf5, 'GradientBR')\n",
    "performance[['GradientBR']]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### AdaBoost regression model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'learning_rate': 0.1, 'n_estimators': 200}"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "parameters = { 'learning_rate':[0.1, 1, 2], 'n_estimators':[50,100,150,200]}\n",
    "abr = ens.AdaBoostRegressor(random_state=25)\n",
    "clf6 = sms.GridSearchCV(abr, parameters)\n",
    "clf6.fit(train_X, train_y)\n",
    "\n",
    "clf6.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>AdaBR</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>MAE</th>\n",
       "      <td>26.686</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RMSE</th>\n",
       "      <td>44.649</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>R2_score</th>\n",
       "      <td>0.923</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          AdaBR\n",
       "MAE      26.686\n",
       "RMSE     44.649\n",
       "R2_score  0.923"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "evaluate(clf6, 'AdaBR')\n",
    "performance[['AdaBR']]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Voting regression model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>MLinear</th>\n",
       "      <th>Lasso</th>\n",
       "      <th>Ridge</th>\n",
       "      <th>Elastic_Net</th>\n",
       "      <th>SVR</th>\n",
       "      <th>R_forest</th>\n",
       "      <th>ExTree</th>\n",
       "      <th>GradientBR</th>\n",
       "      <th>AdaBR</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>MAE</th>\n",
       "      <td>32.231</td>\n",
       "      <td>32.910</td>\n",
       "      <td>32.270</td>\n",
       "      <td>54.753</td>\n",
       "      <td>33.552</td>\n",
       "      <td>16.559</td>\n",
       "      <td>18.076</td>\n",
       "      <td>21.079</td>\n",
       "      <td>26.686</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RMSE</th>\n",
       "      <td>46.875</td>\n",
       "      <td>54.377</td>\n",
       "      <td>61.360</td>\n",
       "      <td>108.898</td>\n",
       "      <td>86.282</td>\n",
       "      <td>26.583</td>\n",
       "      <td>28.738</td>\n",
       "      <td>31.492</td>\n",
       "      <td>44.649</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>R2_score</th>\n",
       "      <td>0.916</td>\n",
       "      <td>0.886</td>\n",
       "      <td>0.855</td>\n",
       "      <td>0.544</td>\n",
       "      <td>0.714</td>\n",
       "      <td>0.973</td>\n",
       "      <td>0.968</td>\n",
       "      <td>0.962</td>\n",
       "      <td>0.923</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          MLinear  Lasso  Ridge  Elastic_Net    SVR  R_forest  ExTree  \\\n",
       "MAE        32.231 32.910 32.270       54.753 33.552    16.559  18.076   \n",
       "RMSE       46.875 54.377 61.360      108.898 86.282    26.583  28.738   \n",
       "R2_score    0.916  0.886  0.855        0.544  0.714     0.973   0.968   \n",
       "\n",
       "          GradientBR  AdaBR  \n",
       "MAE           21.079 26.686  \n",
       "RMSE          31.492 44.649  \n",
       "R2_score       0.962  0.923  "
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "performance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>MLinear</th>\n",
       "      <th>Lasso</th>\n",
       "      <th>Ridge</th>\n",
       "      <th>Elastic_Net</th>\n",
       "      <th>SVR</th>\n",
       "      <th>R_forest</th>\n",
       "      <th>ExTree</th>\n",
       "      <th>GradientBR</th>\n",
       "      <th>AdaBR</th>\n",
       "      <th>VotingR</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>MAE</th>\n",
       "      <td>32.231</td>\n",
       "      <td>32.910</td>\n",
       "      <td>32.270</td>\n",
       "      <td>54.753</td>\n",
       "      <td>33.552</td>\n",
       "      <td>16.559</td>\n",
       "      <td>18.076</td>\n",
       "      <td>21.079</td>\n",
       "      <td>26.686</td>\n",
       "      <td>16.933</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RMSE</th>\n",
       "      <td>46.875</td>\n",
       "      <td>54.377</td>\n",
       "      <td>61.360</td>\n",
       "      <td>108.898</td>\n",
       "      <td>86.282</td>\n",
       "      <td>26.583</td>\n",
       "      <td>28.738</td>\n",
       "      <td>31.492</td>\n",
       "      <td>44.649</td>\n",
       "      <td>26.012</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>R2_score</th>\n",
       "      <td>0.916</td>\n",
       "      <td>0.886</td>\n",
       "      <td>0.855</td>\n",
       "      <td>0.544</td>\n",
       "      <td>0.714</td>\n",
       "      <td>0.973</td>\n",
       "      <td>0.968</td>\n",
       "      <td>0.962</td>\n",
       "      <td>0.923</td>\n",
       "      <td>0.974</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          MLinear  Lasso  Ridge  Elastic_Net    SVR  R_forest  ExTree  \\\n",
       "MAE        32.231 32.910 32.270       54.753 33.552    16.559  18.076   \n",
       "RMSE       46.875 54.377 61.360      108.898 86.282    26.583  28.738   \n",
       "R2_score    0.916  0.886  0.855        0.544  0.714     0.973   0.968   \n",
       "\n",
       "          GradientBR  AdaBR  VotingR  \n",
       "MAE           21.079 26.686   16.933  \n",
       "RMSE          31.492 44.649   26.012  \n",
       "R2_score       0.962  0.923    0.974  "
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# estimators for Voting model are prefitted models those were evaluated before\n",
    "vr = ens.VotingRegressor([('GradientB', clf5), ('ExtraTree', clf4), ('RandomF', clf3)])\n",
    "vr.fit(train_X, train_y)\n",
    "\n",
    "evaluate(vr, 'VotingR')\n",
    "performance"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I merged evaluation metrics and scores of all models we trained into **\"performance\"** in order to make the comparison. So, \n",
    "that can be seen some models performed better like **GradientBoost, ExtraTree, RandomForest, Voting regression** models.\n",
    "Particularly, doing of **hyperparameter tuning** boosted performance on closer optimal values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "According to the comparison of models performances, the **Voting regression** is the optimal(best for now) model to pick and to utilize as the major"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Thanks**"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}