26-linear regression ml model.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "f8f10f82-ce6b-43a9-b6b4-e1c5850893ea",
   "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>country</th>\n",
       "      <th>order_value_EUR</th>\n",
       "      <th>cost</th>\n",
       "      <th>date</th>\n",
       "      <th>category</th>\n",
       "      <th>customer_name</th>\n",
       "      <th>sales_manager</th>\n",
       "      <th>sales_rep</th>\n",
       "      <th>device_type</th>\n",
       "      <th>order_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Sweden</td>\n",
       "      <td>98320.37</td>\n",
       "      <td>77722.25</td>\n",
       "      <td>8/23/2020</td>\n",
       "      <td>Games</td>\n",
       "      <td>Konopelski LLC</td>\n",
       "      <td>Maxie Marrow</td>\n",
       "      <td>Tarrah Castelletti</td>\n",
       "      <td>Tablet</td>\n",
       "      <td>70-0511466</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>France</td>\n",
       "      <td>46296.26</td>\n",
       "      <td>40319.41</td>\n",
       "      <td>5/15/2020</td>\n",
       "      <td>Games</td>\n",
       "      <td>Wisoky Inc</td>\n",
       "      <td>Othello Bowes</td>\n",
       "      <td>Amelina Piscopiello</td>\n",
       "      <td>Tablet</td>\n",
       "      <td>77-3489084</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Portugal</td>\n",
       "      <td>140337.34</td>\n",
       "      <td>115708.14</td>\n",
       "      <td>2020-04-09 00:00:00</td>\n",
       "      <td>Appliances</td>\n",
       "      <td>Hegmann Group</td>\n",
       "      <td>Celine Tumasian</td>\n",
       "      <td>Corene Shirer</td>\n",
       "      <td>PC</td>\n",
       "      <td>65-8218141</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>France</td>\n",
       "      <td>203604.46</td>\n",
       "      <td>175344.16</td>\n",
       "      <td>6/26/2019</td>\n",
       "      <td>Electronics</td>\n",
       "      <td>Kirlin and Sons</td>\n",
       "      <td>Othello Bowes</td>\n",
       "      <td>Crysta Halls</td>\n",
       "      <td>Mobile</td>\n",
       "      <td>29-5478106</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>UK</td>\n",
       "      <td>63979.04</td>\n",
       "      <td>56032.84</td>\n",
       "      <td>10/22/2019</td>\n",
       "      <td>Games</td>\n",
       "      <td>Schoen-Keeling</td>\n",
       "      <td>Jessamine Apark</td>\n",
       "      <td>Genevra Charrisson</td>\n",
       "      <td>PC</td>\n",
       "      <td>27-3437546</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    country  order_value_EUR       cost                 date     category  \\\n",
       "0    Sweden         98320.37   77722.25            8/23/2020        Games   \n",
       "1    France         46296.26   40319.41            5/15/2020        Games   \n",
       "2  Portugal        140337.34  115708.14  2020-04-09 00:00:00   Appliances   \n",
       "3    France        203604.46  175344.16            6/26/2019  Electronics   \n",
       "4        UK         63979.04   56032.84           10/22/2019        Games   \n",
       "\n",
       "     customer_name    sales_manager            sales_rep device_type  \\\n",
       "0   Konopelski LLC     Maxie Marrow   Tarrah Castelletti      Tablet   \n",
       "1       Wisoky Inc    Othello Bowes  Amelina Piscopiello      Tablet   \n",
       "2    Hegmann Group  Celine Tumasian        Corene Shirer          PC   \n",
       "3  Kirlin and Sons    Othello Bowes         Crysta Halls      Mobile   \n",
       "4   Schoen-Keeling  Jessamine Apark   Genevra Charrisson          PC   \n",
       "\n",
       "     order_id  \n",
       "0  70-0511466  \n",
       "1  77-3489084  \n",
       "2  65-8218141  \n",
       "3  29-5478106  \n",
       "4  27-3437546  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "# Replace 'your_file.xlsx' with the path to your Excel file\n",
    "sales_data = 'sales data.xlsx'\n",
    "\n",
    "# Read the Excel file into a pandas DataFrame\n",
    "final_df = pd.read_excel(sales_data)\n",
    "\n",
    "# Now you can work with the DataFrame 'df'\n",
    "final_df.head()  # Display the first few rows of the DataFrame\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "894b6246-c79f-46f4-a656-d944bc19843f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a pivot table with median sales, median cost, and purchase frequency by customers\n",
    "pivot_table = final_df.pivot_table(\n",
    "                              index='customer_name', \n",
    "                              values=['order_value_EUR', 'cost'], \n",
    "                              aggfunc={'order_value_EUR': 'median', 'cost': 'median', 'customer_name': 'count'})\n",
    "\n",
    "# Rename the 'Customer' column to 'Purchase Frequency'\n",
    "pivot_table.columns=['Median Cost', 'Purchase Frequency', 'Median Cost']\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "dcb4e909-a4e2-4fb1-a463-e3008f6a4b64",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                    Median Cost  Purchase Frequency   Median Cost\n",
      "customer_name                                                    \n",
      "Hegmann Group         115708.14                   1  1.403373e+05\n",
      "Kirlin and Sons       249788.24                   2  5.009896e+07\n",
      "Konopelski LLC         77722.25                   1  9.832037e+04\n",
      "Labadie and Sons      110924.13                   1  1.270754e+05\n",
      "Romaguera-Dietrich     55529.13                   3  6.968143e+04\n",
      "Schoen-Keeling         56032.84                   1  6.397904e+04\n",
      "Wisoky Inc             40319.41                   1  4.629626e+04\n"
     ]
    }
   ],
   "source": [
    "# Print the pivot table\n",
    "print(pivot_table)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "c0009dbe-05f0-4fdf-aea4-9759af6c9f1f",
   "metadata": {},
   "outputs": [],
   "source": [
    "sorted_pivot_table = pivot_table.sort_values(by = 'Purchase Frequency', ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "a0b160f1-f622-420a-ac43-102e95fbdfa6",
   "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>Median Cost</th>\n",
       "      <th>Purchase Frequency</th>\n",
       "      <th>Median Cost</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>customer_name</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Romaguera-Dietrich</th>\n",
       "      <td>55529.13</td>\n",
       "      <td>3</td>\n",
       "      <td>6.968143e+04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Kirlin and Sons</th>\n",
       "      <td>249788.24</td>\n",
       "      <td>2</td>\n",
       "      <td>5.009896e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Hegmann Group</th>\n",
       "      <td>115708.14</td>\n",
       "      <td>1</td>\n",
       "      <td>1.403373e+05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Konopelski LLC</th>\n",
       "      <td>77722.25</td>\n",
       "      <td>1</td>\n",
       "      <td>9.832037e+04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Labadie and Sons</th>\n",
       "      <td>110924.13</td>\n",
       "      <td>1</td>\n",
       "      <td>1.270754e+05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Schoen-Keeling</th>\n",
       "      <td>56032.84</td>\n",
       "      <td>1</td>\n",
       "      <td>6.397904e+04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Wisoky Inc</th>\n",
       "      <td>40319.41</td>\n",
       "      <td>1</td>\n",
       "      <td>4.629626e+04</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    Median Cost  Purchase Frequency   Median Cost\n",
       "customer_name                                                    \n",
       "Romaguera-Dietrich     55529.13                   3  6.968143e+04\n",
       "Kirlin and Sons       249788.24                   2  5.009896e+07\n",
       "Hegmann Group         115708.14                   1  1.403373e+05\n",
       "Konopelski LLC         77722.25                   1  9.832037e+04\n",
       "Labadie and Sons      110924.13                   1  1.270754e+05\n",
       "Schoen-Keeling         56032.84                   1  6.397904e+04\n",
       "Wisoky Inc             40319.41                   1  4.629626e+04"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sorted_pivot_table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "90850a50-0988-4d0b-8fb0-3e5def30a604",
   "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>Median Cost</th>\n",
       "      <th>Purchase Frequency</th>\n",
       "      <th>Median Cost</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>customer_name</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Romaguera-Dietrich</th>\n",
       "      <td>55529.13</td>\n",
       "      <td>3</td>\n",
       "      <td>6.968143e+04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Kirlin and Sons</th>\n",
       "      <td>249788.24</td>\n",
       "      <td>2</td>\n",
       "      <td>5.009896e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Hegmann Group</th>\n",
       "      <td>115708.14</td>\n",
       "      <td>1</td>\n",
       "      <td>1.403373e+05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Konopelski LLC</th>\n",
       "      <td>77722.25</td>\n",
       "      <td>1</td>\n",
       "      <td>9.832037e+04</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    Median Cost  Purchase Frequency   Median Cost\n",
       "customer_name                                                    \n",
       "Romaguera-Dietrich     55529.13                   3  6.968143e+04\n",
       "Kirlin and Sons       249788.24                   2  5.009896e+07\n",
       "Hegmann Group         115708.14                   1  1.403373e+05\n",
       "Konopelski LLC         77722.25                   1  9.832037e+04"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# sort by range\n",
    "top_4 = sorted_pivot_table[0:4]\n",
    "top_4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "a5e3c43b-57cf-43cd-a068-baeeb5b56948",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Romaguera-Dietrich', 'Kirlin and Sons', 'Hegmann Group',\n",
       "       'Konopelski LLC'],\n",
       "      dtype='object', name='customer_name')"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "top_4.index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "1210677d-9649-41b2-a401-eeddd361b3b2",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# List of special customer names\n",
    "loyal_customer_names = ['Romaguera-Dietrich', 'Kirlin and Sons', 'Hegmann Group', 'Konopelski LLC']\n",
    "\n",
    "# Function to assign customer type\n",
    "def customer_type(name):\n",
    "    if name in loyal_customer_names:\n",
    "        return 'Loyal customer'\n",
    "    else:\n",
    "        return 'Regular customer'\n",
    "\n",
    "# Assigning new variable 'customer_type'\n",
    "final_df['customer_type'] = final_df['customer_name'].apply(customer_type)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "06ceac12-3834-4f9a-9dda-d55bd7eb76d5",
   "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>country</th>\n",
       "      <th>order_value_EUR</th>\n",
       "      <th>cost</th>\n",
       "      <th>date</th>\n",
       "      <th>category</th>\n",
       "      <th>customer_name</th>\n",
       "      <th>sales_manager</th>\n",
       "      <th>sales_rep</th>\n",
       "      <th>device_type</th>\n",
       "      <th>order_id</th>\n",
       "      <th>customer_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Sweden</td>\n",
       "      <td>98320.37</td>\n",
       "      <td>77722.25</td>\n",
       "      <td>8/23/2020</td>\n",
       "      <td>Games</td>\n",
       "      <td>Konopelski LLC</td>\n",
       "      <td>Maxie Marrow</td>\n",
       "      <td>Tarrah Castelletti</td>\n",
       "      <td>Tablet</td>\n",
       "      <td>70-0511466</td>\n",
       "      <td>Loyal customer</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>France</td>\n",
       "      <td>46296.26</td>\n",
       "      <td>40319.41</td>\n",
       "      <td>5/15/2020</td>\n",
       "      <td>Games</td>\n",
       "      <td>Wisoky Inc</td>\n",
       "      <td>Othello Bowes</td>\n",
       "      <td>Amelina Piscopiello</td>\n",
       "      <td>Tablet</td>\n",
       "      <td>77-3489084</td>\n",
       "      <td>Regular customer</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Portugal</td>\n",
       "      <td>140337.34</td>\n",
       "      <td>115708.14</td>\n",
       "      <td>2020-04-09 00:00:00</td>\n",
       "      <td>Appliances</td>\n",
       "      <td>Hegmann Group</td>\n",
       "      <td>Celine Tumasian</td>\n",
       "      <td>Corene Shirer</td>\n",
       "      <td>PC</td>\n",
       "      <td>65-8218141</td>\n",
       "      <td>Loyal customer</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>France</td>\n",
       "      <td>203604.46</td>\n",
       "      <td>175344.16</td>\n",
       "      <td>6/26/2019</td>\n",
       "      <td>Electronics</td>\n",
       "      <td>Kirlin and Sons</td>\n",
       "      <td>Othello Bowes</td>\n",
       "      <td>Crysta Halls</td>\n",
       "      <td>Mobile</td>\n",
       "      <td>29-5478106</td>\n",
       "      <td>Loyal customer</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>UK</td>\n",
       "      <td>63979.04</td>\n",
       "      <td>56032.84</td>\n",
       "      <td>10/22/2019</td>\n",
       "      <td>Games</td>\n",
       "      <td>Schoen-Keeling</td>\n",
       "      <td>Jessamine Apark</td>\n",
       "      <td>Genevra Charrisson</td>\n",
       "      <td>PC</td>\n",
       "      <td>27-3437546</td>\n",
       "      <td>Regular customer</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    country  order_value_EUR       cost                 date     category  \\\n",
       "0    Sweden         98320.37   77722.25            8/23/2020        Games   \n",
       "1    France         46296.26   40319.41            5/15/2020        Games   \n",
       "2  Portugal        140337.34  115708.14  2020-04-09 00:00:00   Appliances   \n",
       "3    France        203604.46  175344.16            6/26/2019  Electronics   \n",
       "4        UK         63979.04   56032.84           10/22/2019        Games   \n",
       "\n",
       "     customer_name    sales_manager            sales_rep device_type  \\\n",
       "0   Konopelski LLC     Maxie Marrow   Tarrah Castelletti      Tablet   \n",
       "1       Wisoky Inc    Othello Bowes  Amelina Piscopiello      Tablet   \n",
       "2    Hegmann Group  Celine Tumasian        Corene Shirer          PC   \n",
       "3  Kirlin and Sons    Othello Bowes         Crysta Halls      Mobile   \n",
       "4   Schoen-Keeling  Jessamine Apark   Genevra Charrisson          PC   \n",
       "\n",
       "     order_id     customer_type  \n",
       "0  70-0511466    Loyal customer  \n",
       "1  77-3489084  Regular customer  \n",
       "2  65-8218141    Loyal customer  \n",
       "3  29-5478106    Loyal customer  \n",
       "4  27-3437546  Regular customer  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "final_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "9046f751-d876-4f33-8827-2c3d1531e4c6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "customer_type\n",
       "Loyal customer      7\n",
       "Regular customer    3\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "final_df['customer_type'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "7cc280f6-6ff0-41b2-861d-cda632b0ebca",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Give me a python code to extract day, month and a year form date variable in a dataframe"
   ]
  },
  {
   "cell_type": "raw",
   "id": "67887405-221f-4f7d-87c1-8c0151a0ff33",
   "metadata": {},
   "source": [
    "# Convert 'date' column to datetime type\n",
    "final_df['date'] = pd.to_datetime(final_df['date'])\n",
    "\n",
    "# Extract day, month, and year\n",
    "final_df['day'] = final_df['date'].dt.day\n",
    "final_df['month'] = final_df['date'].dt.month\n",
    "final_df['year'] = final_df['date'].dt.year\n",
    "\n",
    "final_df.head()\n",
    "final_df['year'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "5eec806a-ab8c-47c5-8f4c-7c2f6108a7fb",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Give me a python code to assign 1 for \"loyal customer\" and 0 for \"regular customer\" in a variable CustomerType in a dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "3a9452f8-c394-4428-bb87-7a9b21bc0e31",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Gennie\\AppData\\Local\\Temp\\ipykernel_20836\\57327186.py:1: FutureWarning: Downcasting behavior in `replace` is deprecated and will be removed in a future version. To retain the old behavior, explicitly call `result.infer_objects(copy=False)`. To opt-in to the future behavior, set `pd.set_option('future.no_silent_downcasting', True)`\n",
      "  final_df['Customer_Type_encoded'] = final_df['customer_type'].replace({'Loyal customer': 1, 'Regular customer': 0})\n"
     ]
    }
   ],
   "source": [
    "final_df['Customer_Type_encoded'] = final_df['customer_type'].replace({'Loyal customer': 1, 'Regular customer': 0})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "376e1e88-f203-47f0-97c3-0f2cf798d954",
   "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>country</th>\n",
       "      <th>order_value_EUR</th>\n",
       "      <th>cost</th>\n",
       "      <th>date</th>\n",
       "      <th>category</th>\n",
       "      <th>customer_name</th>\n",
       "      <th>sales_manager</th>\n",
       "      <th>sales_rep</th>\n",
       "      <th>device_type</th>\n",
       "      <th>order_id</th>\n",
       "      <th>customer_type</th>\n",
       "      <th>Customer_Type_encoded</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Sweden</td>\n",
       "      <td>98320.37</td>\n",
       "      <td>77722.25</td>\n",
       "      <td>8/23/2020</td>\n",
       "      <td>Games</td>\n",
       "      <td>Konopelski LLC</td>\n",
       "      <td>Maxie Marrow</td>\n",
       "      <td>Tarrah Castelletti</td>\n",
       "      <td>Tablet</td>\n",
       "      <td>70-0511466</td>\n",
       "      <td>Loyal customer</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>France</td>\n",
       "      <td>46296.26</td>\n",
       "      <td>40319.41</td>\n",
       "      <td>5/15/2020</td>\n",
       "      <td>Games</td>\n",
       "      <td>Wisoky Inc</td>\n",
       "      <td>Othello Bowes</td>\n",
       "      <td>Amelina Piscopiello</td>\n",
       "      <td>Tablet</td>\n",
       "      <td>77-3489084</td>\n",
       "      <td>Regular customer</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Portugal</td>\n",
       "      <td>140337.34</td>\n",
       "      <td>115708.14</td>\n",
       "      <td>2020-04-09 00:00:00</td>\n",
       "      <td>Appliances</td>\n",
       "      <td>Hegmann Group</td>\n",
       "      <td>Celine Tumasian</td>\n",
       "      <td>Corene Shirer</td>\n",
       "      <td>PC</td>\n",
       "      <td>65-8218141</td>\n",
       "      <td>Loyal customer</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>France</td>\n",
       "      <td>203604.46</td>\n",
       "      <td>175344.16</td>\n",
       "      <td>6/26/2019</td>\n",
       "      <td>Electronics</td>\n",
       "      <td>Kirlin and Sons</td>\n",
       "      <td>Othello Bowes</td>\n",
       "      <td>Crysta Halls</td>\n",
       "      <td>Mobile</td>\n",
       "      <td>29-5478106</td>\n",
       "      <td>Loyal customer</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>UK</td>\n",
       "      <td>63979.04</td>\n",
       "      <td>56032.84</td>\n",
       "      <td>10/22/2019</td>\n",
       "      <td>Games</td>\n",
       "      <td>Schoen-Keeling</td>\n",
       "      <td>Jessamine Apark</td>\n",
       "      <td>Genevra Charrisson</td>\n",
       "      <td>PC</td>\n",
       "      <td>27-3437546</td>\n",
       "      <td>Regular customer</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    country  order_value_EUR       cost                 date     category  \\\n",
       "0    Sweden         98320.37   77722.25            8/23/2020        Games   \n",
       "1    France         46296.26   40319.41            5/15/2020        Games   \n",
       "2  Portugal        140337.34  115708.14  2020-04-09 00:00:00   Appliances   \n",
       "3    France        203604.46  175344.16            6/26/2019  Electronics   \n",
       "4        UK         63979.04   56032.84           10/22/2019        Games   \n",
       "\n",
       "     customer_name    sales_manager            sales_rep device_type  \\\n",
       "0   Konopelski LLC     Maxie Marrow   Tarrah Castelletti      Tablet   \n",
       "1       Wisoky Inc    Othello Bowes  Amelina Piscopiello      Tablet   \n",
       "2    Hegmann Group  Celine Tumasian        Corene Shirer          PC   \n",
       "3  Kirlin and Sons    Othello Bowes         Crysta Halls      Mobile   \n",
       "4   Schoen-Keeling  Jessamine Apark   Genevra Charrisson          PC   \n",
       "\n",
       "     order_id     customer_type  Customer_Type_encoded  \n",
       "0  70-0511466    Loyal customer                      1  \n",
       "1  77-3489084  Regular customer                      0  \n",
       "2  65-8218141    Loyal customer                      1  \n",
       "3  29-5478106    Loyal customer                      1  \n",
       "4  27-3437546  Regular customer                      0  "
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "final_df.head(5)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "8fea9e2c-459b-4cd5-9171-f9d6a94e92d2",
   "metadata": {},
   "outputs": [],
   "source": [
    "# give me a python code to create a new dummy var for the variables country, category, device_type in a dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "316a7a6c-40cd-4198-b0a9-fc045ec94739",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create dummy variables for multiple categorical variables\n",
    "prep_df = pd.get_dummies(final_df, columns=['country', 'category', 'device_type'], dtype=int)\n",
    "\n",
    "# Concatenate the dummy variables with the original DataFrame\n",
    "df = pd.concat([final_df, prep_df], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "2d725cac-09e0-4fd7-9ee0-c2fca217cd45",
   "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>country</th>\n",
       "      <th>order_value_EUR</th>\n",
       "      <th>cost</th>\n",
       "      <th>date</th>\n",
       "      <th>category</th>\n",
       "      <th>customer_name</th>\n",
       "      <th>sales_manager</th>\n",
       "      <th>sales_rep</th>\n",
       "      <th>device_type</th>\n",
       "      <th>order_id</th>\n",
       "      <th>...</th>\n",
       "      <th>country_Sweden</th>\n",
       "      <th>country_UK</th>\n",
       "      <th>category_Appliances</th>\n",
       "      <th>category_Clothing</th>\n",
       "      <th>category_Electronics</th>\n",
       "      <th>category_Games</th>\n",
       "      <th>device_type_MAC</th>\n",
       "      <th>device_type_Mobile</th>\n",
       "      <th>device_type_PC</th>\n",
       "      <th>device_type_Tablet</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Sweden</td>\n",
       "      <td>98320.37</td>\n",
       "      <td>77722.25</td>\n",
       "      <td>8/23/2020</td>\n",
       "      <td>Games</td>\n",
       "      <td>Konopelski LLC</td>\n",
       "      <td>Maxie Marrow</td>\n",
       "      <td>Tarrah Castelletti</td>\n",
       "      <td>Tablet</td>\n",
       "      <td>70-0511466</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>France</td>\n",
       "      <td>46296.26</td>\n",
       "      <td>40319.41</td>\n",
       "      <td>5/15/2020</td>\n",
       "      <td>Games</td>\n",
       "      <td>Wisoky Inc</td>\n",
       "      <td>Othello Bowes</td>\n",
       "      <td>Amelina Piscopiello</td>\n",
       "      <td>Tablet</td>\n",
       "      <td>77-3489084</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Portugal</td>\n",
       "      <td>140337.34</td>\n",
       "      <td>115708.14</td>\n",
       "      <td>2020-04-09 00:00:00</td>\n",
       "      <td>Appliances</td>\n",
       "      <td>Hegmann Group</td>\n",
       "      <td>Celine Tumasian</td>\n",
       "      <td>Corene Shirer</td>\n",
       "      <td>PC</td>\n",
       "      <td>65-8218141</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>France</td>\n",
       "      <td>203604.46</td>\n",
       "      <td>175344.16</td>\n",
       "      <td>6/26/2019</td>\n",
       "      <td>Electronics</td>\n",
       "      <td>Kirlin and Sons</td>\n",
       "      <td>Othello Bowes</td>\n",
       "      <td>Crysta Halls</td>\n",
       "      <td>Mobile</td>\n",
       "      <td>29-5478106</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>UK</td>\n",
       "      <td>63979.04</td>\n",
       "      <td>56032.84</td>\n",
       "      <td>10/22/2019</td>\n",
       "      <td>Games</td>\n",
       "      <td>Schoen-Keeling</td>\n",
       "      <td>Jessamine Apark</td>\n",
       "      <td>Genevra Charrisson</td>\n",
       "      <td>PC</td>\n",
       "      <td>27-3437546</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 35 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    country  order_value_EUR       cost                 date     category  \\\n",
       "0    Sweden         98320.37   77722.25            8/23/2020        Games   \n",
       "1    France         46296.26   40319.41            5/15/2020        Games   \n",
       "2  Portugal        140337.34  115708.14  2020-04-09 00:00:00   Appliances   \n",
       "3    France        203604.46  175344.16            6/26/2019  Electronics   \n",
       "4        UK         63979.04   56032.84           10/22/2019        Games   \n",
       "\n",
       "     customer_name    sales_manager            sales_rep device_type  \\\n",
       "0   Konopelski LLC     Maxie Marrow   Tarrah Castelletti      Tablet   \n",
       "1       Wisoky Inc    Othello Bowes  Amelina Piscopiello      Tablet   \n",
       "2    Hegmann Group  Celine Tumasian        Corene Shirer          PC   \n",
       "3  Kirlin and Sons    Othello Bowes         Crysta Halls      Mobile   \n",
       "4   Schoen-Keeling  Jessamine Apark   Genevra Charrisson          PC   \n",
       "\n",
       "     order_id  ... country_Sweden  country_UK  category_Appliances  \\\n",
       "0  70-0511466  ...              1           0                    0   \n",
       "1  77-3489084  ...              0           0                    0   \n",
       "2  65-8218141  ...              0           0                    1   \n",
       "3  29-5478106  ...              0           0                    0   \n",
       "4  27-3437546  ...              0           1                    0   \n",
       "\n",
       "   category_Clothing category_Electronics category_Games device_type_MAC  \\\n",
       "0                  0                    0              1               0   \n",
       "1                  0                    0              1               0   \n",
       "2                  0                    0              0               0   \n",
       "3                  0                    1              0               0   \n",
       "4                  0                    0              1               0   \n",
       "\n",
       "  device_type_Mobile device_type_PC device_type_Tablet  \n",
       "0                  0              0                  1  \n",
       "1                  0              0                  1  \n",
       "2                  0              1                  0  \n",
       "3                  1              0                  0  \n",
       "4                  0              1                  0  \n",
       "\n",
       "[5 rows x 35 columns]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "28e2ffab-81be-4095-a807-adf282f0e914",
   "metadata": {},
   "outputs": [],
   "source": [
    "y=prep_df['cost']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "7ae187f9-95f7-42e7-bc54-03bca2c8a746",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     77722.25\n",
       "1     40319.41\n",
       "2    115708.14\n",
       "3    175344.16\n",
       "4     56032.84\n",
       "5    110924.13\n",
       "6     55529.13\n",
       "7     55529.13\n",
       "8     55529.13\n",
       "9    324232.32\n",
       "Name: cost, dtype: float64"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "4d3ca571-6bc8-482a-b104-6e0701c996c5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['order_value_EUR', 'cost', 'date', 'customer_name', 'sales_manager',\n",
       "       'sales_rep', 'order_id', 'customer_type', 'Customer_Type_encoded',\n",
       "       'country_France', 'country_Italy', 'country_Korea', 'country_Portugal',\n",
       "       'country_Sweden', 'country_UK', 'category_Appliances',\n",
       "       'category_Clothing', 'category_Electronics', 'category_Games',\n",
       "       'device_type_MAC', 'device_type_Mobile', 'device_type_PC',\n",
       "       'device_type_Tablet'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prep_df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "9470f778-f6cd-4598-b602-d4174b1b52a1",
   "metadata": {},
   "outputs": [],
   "source": [
    "# give me a python code to drop 5 categorical variables from a dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "d5305281-632c-4e16-aa3b-20ea5ec26ade",
   "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>order_value_EUR</th>\n",
       "      <th>cost</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>98320.37</td>\n",
       "      <td>77722.25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>46296.26</td>\n",
       "      <td>40319.41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>140337.34</td>\n",
       "      <td>115708.14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>203604.46</td>\n",
       "      <td>175344.16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>63979.04</td>\n",
       "      <td>56032.84</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>127075.41</td>\n",
       "      <td>110924.13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>69681.43</td>\n",
       "      <td>55529.13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>69681.43</td>\n",
       "      <td>55529.13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>69681.43</td>\n",
       "      <td>55529.13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>99994324.23</td>\n",
       "      <td>324232.32</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   order_value_EUR       cost\n",
       "0         98320.37   77722.25\n",
       "1         46296.26   40319.41\n",
       "2        140337.34  115708.14\n",
       "3        203604.46  175344.16\n",
       "4         63979.04   56032.84\n",
       "5        127075.41  110924.13\n",
       "6         69681.43   55529.13\n",
       "7         69681.43   55529.13\n",
       "8         69681.43   55529.13\n",
       "9      99994324.23  324232.32"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# List of categorical variables to drop\n",
    "categorical_vars_to_drop = [  'customer_type', 'Customer_Type_encoded',\n",
    "       'country_France', 'country_Italy', 'country_Korea', 'country_Portugal',\n",
    "       'country_Sweden', 'country_UK', 'category_Appliances',\n",
    "       'category_Clothing', 'category_Electronics', 'category_Games',\n",
    "       'device_type_MAC', 'device_type_Mobile', 'device_type_PC',\n",
    "       'device_type_Tablet', 'date', 'customer_name', 'sales_manager', 'sales_rep', 'order_id']\n",
    "\n",
    "# Drop the categorical variables from the DataFrame\n",
    "x = prep_df.drop(columns=categorical_vars_to_drop, axis=1)\n",
    "x\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "a77b76dd-8fca-410e-bf39-9334915de12c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['order_value_EUR', 'cost'], dtype='object')"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "c2c6456b-0851-4d09-a038-2fa66f5e1ac4",
   "metadata": {},
   "outputs": [],
   "source": [
    "# give me a python code to scale a dataframe with standardscaler."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "0262101f-b5b4-450d-b973-c76863c41283",
   "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>order_value_EUR</th>\n",
       "      <th>cost</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>98320.37</td>\n",
       "      <td>77722.25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>46296.26</td>\n",
       "      <td>40319.41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>140337.34</td>\n",
       "      <td>115708.14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>203604.46</td>\n",
       "      <td>175344.16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>63979.04</td>\n",
       "      <td>56032.84</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>127075.41</td>\n",
       "      <td>110924.13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>69681.43</td>\n",
       "      <td>55529.13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>69681.43</td>\n",
       "      <td>55529.13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>69681.43</td>\n",
       "      <td>55529.13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>99994324.23</td>\n",
       "      <td>324232.32</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   order_value_EUR       cost\n",
       "0         98320.37   77722.25\n",
       "1         46296.26   40319.41\n",
       "2        140337.34  115708.14\n",
       "3        203604.46  175344.16\n",
       "4         63979.04   56032.84\n",
       "5        127075.41  110924.13\n",
       "6         69681.43   55529.13\n",
       "7         69681.43   55529.13\n",
       "8         69681.43   55529.13\n",
       "9      99994324.23  324232.32"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Initialize the StandardScaler\n",
    "scaler = StandardScaler()\n",
    "\n",
    "# Fit and transform the DataFrame\n",
    "scaled_data = scaler.fit_transform(x)\n",
    "x\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "5f66e56a-dcdc-4c84-8dd6-983835260c07",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-0.33334695, -0.35176779],\n",
       "       [-0.3350829 , -0.80601252],\n",
       "       [-0.33194493,  0.10955789],\n",
       "       [-0.32983382,  0.83381702],\n",
       "       [-0.33449286, -0.61517828],\n",
       "       [-0.33238745,  0.05145772],\n",
       "       [-0.33430258, -0.62129566],\n",
       "       [-0.33430258, -0.62129566],\n",
       "       [-0.33430258, -0.62129566],\n",
       "       [ 2.99999666,  2.64201294]])"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scaled_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "1c0c1cae-9f51-4564-99e0-dd1bd79248cb",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "850c96d6-8612-4932-8306-402c775439ea",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# Assuming you have your data stored in variables x and y\n",
    "\n",
    "# Split the data into training and testing sets\n",
    "train_x, test_x, train_y, test_y = train_test_split(scaled_data, y, test_size=0.2, random_state=42)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "634321b1-28b9-4c21-a3cf-b03381153792",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-0.33238745,  0.05145772],\n",
       "       [-0.33334695, -0.35176779],\n",
       "       [-0.33430258, -0.62129566],\n",
       "       [-0.33194493,  0.10955789],\n",
       "       [ 2.99999666,  2.64201294],\n",
       "       [-0.33449286, -0.61517828],\n",
       "       [-0.32983382,  0.83381702],\n",
       "       [-0.33430258, -0.62129566]])"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "59ff9254-a15e-4c94-bd36-2d4e769c2495",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-0.33430258, -0.62129566],\n",
       "       [-0.3350829 , -0.80601252]])"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "0baf7dc1-61f3-4d04-816e-1520c1ceaf9c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5    110924.13\n",
       "0     77722.25\n",
       "7     55529.13\n",
       "2    115708.14\n",
       "9    324232.32\n",
       "4     56032.84\n",
       "3    175344.16\n",
       "6     55529.13\n",
       "Name: cost, dtype: float64"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "1b289606-8ae7-448c-8fad-58628004c56a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "8    55529.13\n",
       "1    40319.41\n",
       "Name: cost, dtype: float64"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "65398fd9-0b25-4681-a0c5-8dc208d5be12",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.metrics import mean_absolute_error"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "030ea846-9511-4472-934b-0c9ab1673e8e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Give me a python code build Linear Regression ML modelon train data, \n",
    "# make prediction on test data, measure accuracy with MAPE, a plot to visualize y_pred against test_y\n",
    "# and kde plot to visualize the disturbution of a residual"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "bf37185d-7f54-4e4f-a04b-4376c1fc461b",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.metrics import mean_absolute_percentage_error"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "3f3832e5-2874-4cec-8678-004ef21e5fee",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Train linear regression model\n",
    "model = LinearRegression()\n",
    "\n",
    "# Train (fit) the model z\n",
    "model.fit(train_x, train_y)\n",
    "\n",
    "# Make predictions of the test data\n",
    "y_pred = model.predict(test_x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "28c85e28-42ee-4700-9118-b41f248ebe03",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "c044a923-6b41-4923-b158-08440d35629a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define a function to calculate Mean Absolute Percentage Error (MAPE)\n",
    "def calculate_mape(test_y, y_pred):\n",
    "    return np.mean(np.abs((test_y - y_pred) / test_y)) * 100\n",
    "# Calculating the Mean Absolute Percentage Error (MAPE) using the custom function\n",
    "mape = calculate_mape(test_y, y_pred)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "f54edae4-99e6-42fa-b112-163840677960",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.metrics import mean_absolute_percentage_error"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "211175c2-8024-4b31-a928-0684a994f177",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(y_pred, test_y, color=\"blue\")\n",
    "plt.xlabel('Prediction')\n",
    "plt.ylabel('Actual data')\n",
    "plt.title('Linear Regression prediction vs actual data')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "aaf86767-ba96-4f4f-b6a1-d296962b975c",
   "metadata": {},
   "outputs": [],
   "source": [
    "# pip install seaborn\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "c1e5f713-7968-4453-bfc2-b6f7e4faeeb9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAioAAAHHCAYAAACRAnNyAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8fJSN1AAAACXBIWXMAAA9hAAAPYQGoP6dpAABiVElEQVR4nO3dd3RUZeI+8Gcmk5nUmfTeC6GFEHoo0jsKdl2VoLKWhXWVdVdwd62r6E8RrGANoijSe+/SISEh9ATSO+l1ksy8vz9Y8jUkQBKS3CnP55w5x7m5M/PkZpx5uPe975UJIQSIiIiIDJBc6gBEREREt8KiQkRERAaLRYWIiIgMFosKERERGSwWFSIiIjJYLCpERERksFhUiIiIyGCxqBAREZHBYlEhIiIig8WiQnSX3nrrLchksk55rREjRmDEiBEN9/fv3w+ZTIbVq1d3yuvPmDEDAQEBnfJabVVRUYGZM2fCw8MDMpkML7/8stSRmpDJZHjrrbfuuF5nvLduvIf279/foa9D1FYsKkR/sHTpUshksoablZUVvLy8MH78eHz22WcoLy9vl9fJzs7GW2+9hfj4+HZ5vvZkyNla4v3338fSpUvx4osv4qeffsJTTz11y3UDAgIa/b1tbW0xYMAALFu2rBMTE9HtKKQOQGSI3nnnHQQGBqKurg65ubnYv38/Xn75ZXzyySfYuHEjevXq1bDuv//9b8ydO7dVz5+dnY23334bAQEB6N27d4sft3Pnzla9TlvcLtu3334LvV7f4Rnuxt69ezFo0CC8+eabLVq/d+/e+Pvf/w4AyMnJwXfffYfo6GhotVr8+c9/7pCM1dXVUCj48UvUEvw/hagZEydORL9+/Rruz5s3D3v37sWUKVNw33334cKFC7C2tgYAKBSKDv/Sqaqqgo2NDZRKZYe+zp1YWlpK+votkZ+fj+7du7d4fW9vbzz55JMN92fMmIGgoCAsXLiww4qKlZVVhzwvkSnioR+iFho1ahT+85//IC0tDT///HPD8ubGEezatQtDhw6Fg4MD7OzsEBYWhtdffx3A9TEB/fv3BwA8/fTTDYcdli5dCuD6OJSePXsiNjYW99xzD2xsbBoee/MYlRt0Oh1ef/11eHh4wNbWFvfddx8yMjIarRMQEIAZM2Y0eewfn/NO2Zobo1JZWYm///3v8PX1hUqlQlhYGD7++GPcfGF2mUyG2bNnY/369ejZsydUKhV69OiB7du3N7/Bb5Kfn49nn30W7u7usLKyQkREBH788ceGn98Ya5GSkoItW7Y0ZE9NTW3R89/g6uqKrl274sqVK42W6/V6LFq0CD169ICVlRXc3d3x/PPPo7i4uNF6p06dwvjx4+Hi4gJra2sEBgbimWeeabItbh6jcujQIfTv3x9WVlYIDg7G119/3SRbampqo7/H7Z4zLS0Nf/nLXxAWFgZra2s4Ozvj4YcfbtH2SEpKwoMPPggPDw9YWVnBx8cHjz32GEpLS+/4WKL2xj0qRK3w1FNP4fXXX8fOnTtv+a/tc+fOYcqUKejVqxfeeecdqFQqJCcn4/DhwwCAbt264Z133sEbb7yB5557DsOGDQMADB48uOE5CgsLMXHiRDz22GN48skn4e7ufttc7733HmQyGV577TXk5+dj0aJFGDNmDOLj4xv2/LRES7L9kRAC9913H/bt24dnn30WvXv3xo4dO/CPf/wDWVlZWLhwYaP1Dx06hLVr1+Ivf/kL7O3t8dlnn+HBBx9Eeno6nJ2db5mruroaI0aMQHJyMmbPno3AwECsWrUKM2bMQElJCf72t7+hW7du+Omnn/DKK6/Ax8en4XCOq6tri39/AKivr0dmZiYcHR0bLX/++eexdOlSPP3003jppZeQkpKCL774AqdPn8bhw4dhaWmJ/Px8jBs3Dq6urpg7dy4cHByQmpqKtWvX3vY1ExMTGx731ltvob6+Hm+++eYd/+63c/LkSRw5cgSPPfYYfHx8kJqaisWLF2PEiBE4f/48bGxsmn1cbW0txo8fD61Wi7/+9a/w8PBAVlYWNm/ejJKSEmg0mjZnImoTQUQNYmJiBABx8uTJW66j0WhEZGRkw/0333xT/PF/pYULFwoAoqCg4JbPcfLkSQFAxMTENPnZ8OHDBQCxZMmSZn82fPjwhvv79u0TAIS3t7coKytrWL5y5UoBQHz66acNy/z9/UV0dPQdn/N22aKjo4W/v3/D/fXr1wsA4r///W+j9R566CEhk8lEcnJywzIAQqlUNlqWkJAgAIjPP/+8yWv90aJFiwQA8fPPPzcsq62tFVFRUcLOzq7R7+7v7y8mT5582+f747rjxo0TBQUFoqCgQCQmJoqnnnpKABCzZs1qWO/3338XAMTy5csbPX779u2Nlq9bt+6O7x8hrm+LN998s+H+tGnThJWVlUhLS2tYdv78eWFhYdHovZWSknLLv83Nz1lVVdVknaNHjwoAYtmyZQ3LbryH9u3bJ4QQ4vTp0wKAWLVq1W1/B6LOwkM/RK1kZ2d327N/HBwcAAAbNmxo88BTlUqFp59+usXrT58+Hfb29g33H3roIXh6emLr1q1tev2W2rp1KywsLPDSSy81Wv73v/8dQghs27at0fIxY8YgODi44X6vXr2gVqtx9erVO76Oh4cHHn/88YZllpaWeOmll1BRUYEDBw60+XfYuXMnXF1d4erqivDwcPz00094+umn8dFHHzWss2rVKmg0GowdOxbXrl1ruPXt2xd2dnbYt28fgP/722/evBl1dXUten2dTocdO3Zg2rRp8PPza1jerVs3jB8/vs2/1x/3pNXV1aGwsBAhISFwcHBAXFzcLR93Y4/Jjh07UFVV1ebXJ2ovJlNUDh48iHvvvRdeXl6QyWRYv359qx5fU1ODGTNmIDw8HAqFAtOmTWt2vf3796NPnz5QqVQICQlp9lgxmbaKiopGpeBmjz76KIYMGYKZM2fC3d0djz32GFauXNmq0uLt7d2qgbOhoaGN7stkMoSEhLR6fEZrpaWlwcvLq8n26NatW8PP/+iPX8Q3ODo6Nhnn0dzrhIaGQi5v/JF1q9dpjYEDB2LXrl3Yvn07Pv74Yzg4OKC4uLjR9k9KSkJpaSnc3NwaSs2NW0VFBfLz8wEAw4cPx4MPPoi3334bLi4umDp1KmJiYqDVam/5+gUFBaiurm7yNwSAsLCwNv9e1dXVeOONNxrGDrm4uMDV1RUlJSW3HWsSGBiIOXPm4LvvvoOLiwvGjx+PL7/8kuNTSDImM0alsrISEREReOaZZ/DAAw+0+vE6nQ7W1tZ46aWXsGbNmmbXSUlJweTJk/HCCy9g+fLl2LNnD2bOnAlPT8+7+pcPGY/MzEyUlpYiJCTklutYW1vj4MGD2LdvH7Zs2YLt27fjt99+w6hRo7Bz505YWFjc8XVaM66kpW41cZhOp2tRpvZwq9cRNw287UwuLi4YM2YMAGD8+PHo2rUrpkyZgk8//RRz5swBcH0grZubG5YvX97sc9wYB3Nj8r1jx45h06ZN2LFjB5555hksWLAAx44dg52d3V1lvd3f8GZ//etfERMTg5dffhlRUVHQaDSQyWR47LHH7liaFyxYgBkzZmDDhg3YuXMnXnrpJcyfPx/Hjh2Dj4/PXf0ORK1lMntUJk6ciP/+97+4//77m/25VqvFq6++Cm9vb9ja2mLgwIGNZmK0tbXF4sWL8ec//xkeHh7NPseSJUsQGBiIBQsWoFu3bpg9ezYeeuihJgMGyXT99NNPAHDHYiqXyzF69Gh88sknOH/+PN577z3s3bu34RBBe882mpSU1Oi+EALJycmNztBxdHRESUlJk8fevDeiNdn8/f2RnZ3d5FDYxYsXG37eHvz9/ZGUlNTkC7a9XwcAJk+ejOHDh+P9999HZWUlACA4OBiFhYUYMmQIxowZ0+QWERHR6DkGDRqE9957D6dOncLy5ctx7tw5rFixotnXc3V1hbW1dZO/IQBcunSp0f0bA3xv/js2t0dp9erViI6OxoIFC/DQQw9h7NixGDp0aLPvgeaEh4fj3//+Nw4ePIjff/8dWVlZWLJkSYseS9SeTKao3Mns2bNx9OhRrFixAmfOnMHDDz+MCRMmNPvhcCtHjx5t+JfXDePHj8fRo0fbOy4ZoL179+Ldd99FYGAgnnjiiVuuV1RU1GTZjYnTbhwCsLW1BdD0C6etli1b1qgsrF69Gjk5OZg4cWLDsuDgYBw7dgy1tbUNyzZv3tzkNObWZJs0aRJ0Oh2++OKLRssXLlwImUzW6PXvxqRJk5Cbm4vffvutYVl9fT0+//xz2NnZYfjw4e3yOje89tprKCwsxLfffgsAeOSRR6DT6fDuu+82Wbe+vr5hWxUXFzfZO3Tz3/5mFhYWGD9+PNavX4/09PSG5RcuXMCOHTsaratWq+Hi4oKDBw82Wv7VV181+7w3Z/n888+b3fvyR2VlZaivr2+0LDw8HHK5/LaHsIg6iskc+rmd9PR0xMTEID09HV5eXgCAV199Fdu3b0dMTAzef//9Fj1Pbm5uk9MF3d3dUVZWhurq6g7ZXU/S2LZtGy5evIj6+nrk5eVh79692LVrF/z9/bFx48bbTtj1zjvv4ODBg5g8eTL8/f2Rn5+Pr776Cj4+Phg6dCiA66XBwcEBS5Ysgb29fcNevsDAwDbldXJywtChQ/H0008jLy8PixYtQkhISKNTqGfOnInVq1djwoQJeOSRR3DlyhX8/PPPjQa3tjbbvffei5EjR+Jf//oXUlNTERERgZ07d2LDhg14+eWXmzx3Wz333HP4+uuvMWPGDMTGxiIgIACrV6/G4cOHsWjRotuOGWqLiRMnomfPnvjkk08wa9YsDB8+HM8//zzmz5+P+Ph4jBs3DpaWlkhKSsKqVavw6aef4qGHHsKPP/6Ir776Cvfffz+Cg4NRXl6Ob7/9Fmq1GpMmTbrl67399tvYvn07hg0bhr/85S8NJaxHjx44c+ZMo3VnzpyJDz74ADNnzkS/fv1w8OBBXL58uclzTpkyBT/99BM0Gg26d++Oo0ePYvfu3bc9DRy4Xshnz56Nhx9+GF26dEF9fT1++uknWFhY4MEHH2zbBiW6G5Kec9RBAIh169Y13N+8ebMAIGxtbRvdFAqFeOSRR5o8Pjo6WkydOrXJ8tDQUPH+++83WrZlyxYBoNlTAcn43Dg9+cZNqVQKDw8PMXbsWPHpp582Og32hptPT96zZ4+YOnWq8PLyEkqlUnh5eYnHH39cXL58udHjNmzYILp37y4UCkWjU06HDx8uevTo0Wy+W52e/Ouvv4p58+YJNzc3YW1tLSZPntzoVNcbFixYILy9vYVKpRJDhgwRp06davKct8t28+nJQghRXl4uXnnlFeHl5SUsLS1FaGio+Oijj4Rer2+0Hm465feGW502fbO8vDzx9NNPCxcXF6FUKkV4eHizp+m29vTkW627dOnSJqcCf/PNN6Jv377C2tpa2Nvbi/DwcPHPf/5TZGdnCyGEiIuLE48//rjw8/MTKpVKuLm5iSlTpohTp041em7cdCqxEEIcOHBA9O3bVyiVShEUFCSWLFnS5L0lxPXTjp999lmh0WiEvb29eOSRR0R+fn6T5ywuLm7YXnZ2dmL8+PHi4sWLTbb3zacnX716VTzzzDMiODhYWFlZCScnJzFy5Eixe/fuFm1TovYmE0LCUWwdRCaTYd26dQ1n7vz222944okncO7cuSaD+ezs7JqMSbkxidTNZw7dc8896NOnDxYtWtSw7MZgNY6IJyIian9mcegnMjISOp0O+fn5DTNttkVUVFSTeSl27dqFqKiou41IREREzTCZolJRUYHk5OSG+ykpKYiPj4eTkxO6dOmCJ554AtOnT8eCBQsQGRmJgoIC7NmzB7169cLkyZMBAOfPn0dtbS2KiopQXl7ecJn7G4PhXnjhBXzxxRf45z//iWeeeQZ79+7FypUrsWXLls7+dYmIiMyCyRz62b9/P0aOHNlkeXR0NJYuXYq6ujr897//xbJly5CVlQUXFxcMGjQIb7/9NsLDwwFcv2hbc6f5/XET7d+/H6+88grOnz8PHx8f/Oc//2n2Qm9ERER090ymqBAREZHpMZt5VIiIiMj4sKgQERGRwTLqwbR6vR7Z2dmwt7dv9ynJiYiIqGMIIVBeXg4vL68mFxu9mVEXlezsbPj6+kodg4iIiNogIyPjjhe6NOqicmPa7IyMDKjVaonTEBERUUuUlZXB19e3RZe/MOqicuNwj1qtZlEhIiIyMi0ZtsHBtERERGSwWFSIiIjIYLGoEBERkcFiUSEiIiKDxaJCREREBotFhYiIiAwWiwoREREZLBYVIiIiMlgsKkRERGSwWFSIiIjIYLGoEBERkcFiUSEiIiKDxaJCREREBotFhYiIiAyWQuoAREQAUFVbj6sFlcgsrkJRZR1Kq+ugFwIAYKdSwNFWCQ+1FYJdbeFkq2zR5eGJyPixqBCRJAortNh3qQDHrxbiRGoR0gqrGn4mA2CrUkD+vy5SVatDvV40/NzJVom+/o7oH+CIkWFuCHGzY3EhMlEyIYS482qGqaysDBqNBqWlpVCr1VLHIaI7qNTWY/OZbKyJzcKptCIIAQS42CDUzR6BLrbwdrCGu9oKdioF5PL/Kx5CCFTX6VBYUYvs0mqkF1bhcl45kvIroK3Xw9/ZBlN7e+Phvj7wdbKR8DckopZozfc3iwoRdbjM4ip8e/AqVsVmorpWh3AfDQYFOqOPvyM01pZtft7aej3OZpfiZEoRjqcUoaZOh3u6uOLFEcEYGOjEvSxEBopFhYgMQk5pNRbuuow1cVmwsbTA2O7uGNnVDS52qnZ/rZo6HY5eLcT2s7lIL6pCHz8HzJ3YDQMCndr9tYjo7rCoEJGkaup0+GpfMr45eBVKhRz3RXhjdDc3WFladPhrCyFwOr0Ea+IycfVaJcZ1d8d/pnTnISEiA8KiQkSSOZR0Da+vS0ROaTUmhXvivggv2Cg7f9y+XggcuVKI306mo6KmHnPGdcEzQwKhsOCsDERSY1Ehok5XU6fDe1su4KdjaejuqcbMoYHwdLCWOhZq6nRYeSoDO87lItxbg88ej4S/s63UsYjMGosKEXWqS7nlmPVLHNILq/DEID+M7eZucANZk/LK8eX+ZJTX1OO9+3vi/kgfqSMRma3WfH9zHygR3ZUtZ3Iw9ctDqK3X47/TemJcdw+DKykAEOpuj/fvD0dff0e88lsC3tp4DnU6vdSxiOgOOOEbEbWJXi+wYNclfLnvCgYHO+O5e4KgUnT8YNm7YaNU4MXhwQh1s8OPR9NwIacMS57sC0dbpdTRiOgWuEeFiFqttl6PV1bG46t9V/D4AD/MHhli8CXlBplMhrHdPfDvSd1wIacMDyw+goyiqjs/kIgkwaJCRK1Soa3HjJgT2JqYg5dGh+K+CC+DPNRzJ1091Xjrvh6ortXh/q8O41x2qdSRiKgZLCpE1GJlNXV46vvjiM8owbyJ3TAoyFnqSHfFU2ONt+7rAbW1JR7/9hjOZJZIHYmIbsKiQkQtUlpVhye/PY7LeeV4fVI3dPM0jTPtNNaW+NekbnBXW+FP3x5HbFqx1JGI6A9YVIjojiq19YiOOYGUa5X416TuCHa1kzpSu7JRKjBvQjf4OFoj+ocTOJvFw0BEhoJFhYhuq6ZOh5k/nsKl3HK8NrErAl1Mc7I0a6UF/jE+DO5qFZ76/jiS8yukjkREYFEhotvQ6QVe+vU0YtOK8Y/xYSa3J+VmNkoFXpvQFXYqBZ787jhyS2ukjkRk9lhUiOiW3ttyAbsv5OGl0aEmMyblTuytLDF3YjfU6/WYEXMCFdp6qSMRmTXJi0pWVhaefPJJODs7w9raGuHh4Th16pTUsYjMXszhFPxwOAXRgwPQ199R6jidyslWiX+O74r0oirMWh6Hes5gSyQZSYtKcXExhgwZAktLS2zbtg3nz5/HggUL4OhoXh+KRIbm4OUCvLv5PCaHe2Jcdw+p40jC18kGr4zpgkPJ1/D+1otSxyEyW5JOof/hhx/C19cXMTExDcsCAwMlTEREaYWVmP1rHHr5OOBPA/ykjiOpnt4aTB/kjx8Op6CHlxoP9uWFDIk6m6R7VDZu3Ih+/frh4YcfhpubGyIjI/Htt9/ecn2tVouysrJGNyJqP5Xaesz88RRslQrMHhkCudz4Zpxtb2O7u2NEmCvmrU1EQkaJ1HGIzI6kReXq1atYvHgxQkNDsWPHDrz44ot46aWX8OOPPza7/vz586HRaBpuvr6+nZyYyHQJIfCvdYnIKK7CnLFdYKviNUuB69cGemZIIPycbfDi8liUVtVJHYnIrMiEEEKqF1cqlejXrx+OHDnSsOyll17CyZMncfTo0Sbra7VaaLXahvtlZWXw9fVFaWkp1GrzOCOBqKOsOJGOuWsTMXtkCIaEuEgdx+AUlGvx+rpERAU545vpfY3y+kZEhqKsrAwajaZF39+S7lHx9PRE9+7dGy3r1q0b0tPTm11fpVJBrVY3uhHR3buQU4Y3N57D6K5uLCm34GqvwgvDg7HrQh6+P5QidRwisyFpURkyZAguXbrUaNnly5fh7+8vUSIi81NTp8NLv56Gu9oK06MCpI5j0Pr6O2JSuCc+2HaRV1sm6iSSFpVXXnkFx44dw/vvv4/k5GT88ssv+OabbzBr1iwpYxGZlQ+3X0RqYSVmjwyBUiH51EoG77H+vvBxtMbfVsSjpk4ndRwikyfpp1L//v2xbt06/Prrr+jZsyfeffddLFq0CE888YSUsYjMxsHLBYg5nIrHB/jB18lG6jhGwdJCjr+MCEFaYSU+2Mb5VYg6mqSDae9WawbjEFFjpdV1GLfwANztrfDaxK6Qc3Boq2w/m4Mfj6bhl5kDMZjjeohaxWgG0xKRdN7bch5l1fV47p4glpQ2GNfDA9091fjH6jOo5PWAiDoMiwqRGdp/KR8rT2XiiUF+cLZTSR3HKMllMjx3TxCuVWjx4XYeAiLqKCwqRGamQluPeWsTEe6twagwN6njGDV3tRUe6++LZUfTcPxqodRxiEwSiwqRmflk52UUVdbiz8MCOWlZOxjXwwNd3O0wb10itPU8C4iovbGoEJmRxMxSLD2Sggf7+MDV3krqOCZBLpPh2aFBSCuswjcHrkodh8jksKgQmYl6nR5z156Br5MNJoZ7SB3HpPg52WByuCc+35uMlGuVUschMiksKkRm4pcT6TifXYZnhwRCIef/+u3tgT7ecLCxxJsbzsKIZ30gMjj8tCIyA4UVWny84xJGhLki1N1e6jgmSaWwwFOD/HEw6Rp2nc+TOg6RyWBRITIDH++8BJ0QeKy/n9RRTFpff0f09tXg7U3nOb0+UTthUSEycYmZpVhxIgMP9/WF2tpS6jgmTSaTYfqgAOSV1eBrDqwlahcsKkQmTAiBtzedg6+TDcZ0c5c6jlnwdLDGpHBPLN6fjNzSGqnjEBk9FhUiE7b9bC5OpRXjiYF+sJBzzpTOMrW3F1SWFvhoB2esJbpbLCpEJkpbr8P7Wy8g0tcBvXwcpI5jVmyUCjzU1wdr4rKQmFkqdRwio8aiQmSilh1JQ1ZJNf40kANopTAyzA2+jtZ4d/N5nq5MdBdYVIhMUGlVHT7bm4RRXd3h42gjdRyzZCGX4U8D/XEitQh7L+ZLHYfIaLGoEJmgxQeuoLZejwf7eEsdxaxF+GjQw0uND7dfhE7PvSpEbcGiQmRickqrEXM4BZPCPeFgo5Q6jlmTyWR4rL8fLudVYN3pLKnjEBklFhUiE/Pp7iQoFXJM6eUpdRQCEOJmhwGBTvhk5yVOAkfUBiwqRCYk5VolVp3KxLTe3rBRKqSOQ//zaD9f5JbV4Jfj6VJHITI6LCpEJmTR7stwsLHk5G4GxsvBGsNCXfHl/mRU13KvClFrsKgQmYikvHJsjM/G1N5eUCr4v7aheSDSGyVVdVh2NFXqKERGhZ9mRCZi4e7LcLFXYWSYm9RRqBluaiuM6OKKxQeuoEJbL3UcIqPBokJkAi7klGFrYi7u7+0NhQX/tzZU90d6o1Jbjx+PpEodhcho8BONyAR8vjcJ7moVhnVxkToK3Yaz3fU9Xt8cvMq9KkQtxKJCZOQu55VjW2Iu7o3wgkLO/6UN3X0RXqjU1uPnY2lSRyEyCvxUIzJyX+xNgrOdEsNDXaWOQi3gbKfCPV1c8e3BqzwDiKgFWFSIjNiVggpsPpNzfW8Kx6YYjfsivFBcVYtfTnBeFaI74ScbkRFbvD8ZGmtLjOjCM32MibvaCkNDXbDkwBVo67lXheh2WFSIjFRWSTXWnc7G5HDOm2KM7ovwxrVyLdbG8RpARLfDTzciI/Xd71dhbWmBUV25N8UYeTtYo3+AE5YcuMIrKxPdBosKkREqqqzFryfSMa6HO6yVFlLHoTa6r7cX0gqrsP1srtRRiAwWiwqREVp6JBVCAON7eEgdhe5CsKsdwr01+HJfMoTgXhWi5rCoEBmZqtrrM5uODHOD2spS6jh0l+6L8ML5nDIcSr4mdRQig8SiQmRkVsdmorymDpPCuTfFFPTwUiPA2QbfHrwqdRQig8SiQmREdHqBbw9excBAZ7jaW0kdh9qBTCbD5F5eOJh0DRdyyqSOQ2RwWFSIjMjOc7nIKK7G5F6eUkehdjQoyAnOtkp89zv3qhDdjEWFyEgIIbDk4BV091Qj2NVO6jjUjhRyOSb09MCG+GzkltZIHYfIoLCoEBmJuPQSJGSUYlI496aYolFd3aBUyPHj0VSpoxAZFBYVIiPxw6EUeGisEOnnIHUU6gA2SgVGdHHFL8fTebFCoj9gUSEyAlkl1dh+NhcTenhALpNJHYc6yPgeHiivqcO605xWn+gGFhUiI7DsaCqsLOUY3sVV6ijUgdzUVujr74jvD13lBHBE/8OiQmTgqmrr8cvxdIwIc4OVJafLN3UTe3riSkElfk/iBHBEAIsKkcFbG5eFSm09xvdwlzoKdYKuHvYIdLHF94dSpI5CZBBYVIgMmBACPx5JRV9/R07wZiZkMhnG93DHwcsFSL1WKXUcIslJWlTeeustyGSyRreuXbtKGYnIoBy9Woik/AqM687p8s1JVJAL7KwU+OlYmtRRiCQn+R6VHj16ICcnp+F26NAhqSMRGYxlR1Lh42iNHl5qqaNQJ1Iq5BgZ5oaVJzNQVVsvdRwiSUleVBQKBTw8PBpuLi4uUkciMghZJdXYeT4P47q7Q8ZTks3OmG7uqKytx/rT2VJHIZKU5EUlKSkJXl5eCAoKwhNPPIH09PRbrqvValFWVtboRmSqlh9Lg5WlBYaF8pRkc+Rqr0Jff0csPZLCU5XJrElaVAYOHIilS5di+/btWLx4MVJSUjBs2DCUl5c3u/78+fOh0Wgabr6+vp2cmKhzaOt1WHEyA/eEuvKUZDM2trsHLudV4GRqsdRRiCQjEwZU1UtKSuDv749PPvkEzz77bJOfa7VaaLXahvtlZWXw9fVFaWkp1GoewyfTsSE+C39bEY+PH4qAt6O11HFIInoh8OqqBPQPcMRnj/eROg5RuykrK4NGo2nR97fkh37+yMHBAV26dEFycnKzP1epVFCr1Y1uRKbo52Np6OGlZkkxc3KZDKO7umNrYi6uVWjv/AAiE2RQRaWiogJXrlyBpyevDkvm61JuOU6mFmN0V07wRsDwLq6Qy2RYeSpD6ihEkpC0qLz66qs4cOAAUlNTceTIEdx///2wsLDA448/LmUsIkktP54GB2tL9A9wlDoKGQA7KwUGBTlh+bF06PQGc6SeqNNIWlQyMzPx+OOPIywsDI888gicnZ1x7NgxuLryLAcyT1W19VgTm4kRYW5QWBjUDk+S0Nju7sgqqcbBywVSRyHqdAopX3zFihVSvjyRwdmckIOqWh1GdWVZp/8T7GqHAGcb/HoiHSO7ukkdh6hT8Z9sRAZk+Yk0RPhqeF0fakQmk2FkVzfsuZCPvLIaqeMQdSoWFSIDcT67DAkZpRgVxkG01NTQEBcoLGRYxUG1ZGZYVIgMxK8n0uFoY4lIfwepo5ABslEqEBXkjF9OcFAtmRcWFSIDUFVbj3WnszC8ixsUcv5vSc0b3c0N2SU1+D2Jg2rJfPATkcgAbE3MRYW2HiPDOIiWbi3Y1Q7+zjZYcZKHf8h8sKgQGYAVJ9IR7q2Bm5qDaOnWZDIZRnRxw+7zeZyplswGiwqRxJLzK3AqrZh7U6hFhoa4QCYD1sZlSh2FqFOwqBBJbNWpDNirFOgX4CR1FDICdlYK9A9wwoqTGTCga8oSdRgWFSIJ1dbrsTo2E0NCXWDJmWiphUaGueFqQSVi04qljkLU4fjJSCShvRfzUFhZi5FhnG2UWq67lxruahUH1ZJZYFEhktBvJzMQ7GoLPycbqaOQEZHLZLgn1BWbz2SjQlsvdRyiDsWiQiSR/LIaHLhcgOFduDeFWm94F1do6/TYeiZH6ihEHYpFhUgia09nQSGXY3Cws9RRyAg526kQ7q3Bb5xSn0wciwqRBIQQ+O1kBvoHOsJWJelFzMmIDQ9zRWxaMa4WVEgdhajDsKgQSSAuvQQp1yp52IfuSj9/J9iqLLAqlnOqkOliUSGSwOrYDLjaKdHDSy11FDJiSoUcg4NdsCY2E/U6vdRxiDoEiwpRJ6up02FTQg6GhbpCLpNJHYeM3PAursgv1+JQ8jWpoxB1CBYVok6249z1CxAOC+WU+XT3glxs4eNojTU8/EMmikWFqJOtjs1EVw97eGh4AUK6ezKZDMNCXbHzfB5Kq+ukjkPU7lhUiDpRbmkNDidf494UaldDQ1xQp9NjC+dUIRPEokLUidaezoRCLsegIF6AkNqPk60S4d4arI7lnCpkelhUiDqJEAKrYzPRL8ARNkrOnULta1ioa8Np70SmhEWFqJOcySzF1YJK3MPDPtQB+gc4wUZpgXVxHFRLpoVFhaiTrDudBUcbS/T01kgdhUyQUiHHwEAnrInLgl4vpI5D1G5YVIg6QZ1Ojw3xWRgS4gILOedOoY4xLNQVWSXVOJVWLHUUonbDokLUCQ5cKkBxVR2GhrhIHYVMWJiHPVztVVh3mod/yHSwqBB1gjVxmQhwtoG/s63UUciEyWUyDA1xweaEHNTU6aSOQ9QuWFSIOlhpVR12X8jD0BAOoqWONyzEBeXaeuy+kCd1FKJ2waJC1MG2ns2BTi8wOMRZ6ihkBjwdrBHqZod1cVlSRyFqFywqRB1sXVwWwr01cLRRSh2FzMTQEBccuFyAospaqaMQ3TUWFaIOlFlchROpRRjCQbTUiQYFOUMA2HImW+ooRHeNRYWoA22Iz4ZKIUf/AE6ZT51HbW2JCB8N1vDwD5kAFhWiDiKEwJq4TPQPcIKVpYXUccjMDA1xQXxGCVI5pT4ZORYVog5yLrsMVwsqediHJNHH3xHWlhZYH8+9KmTcWFSIOsi601nQWFsinFPmkwRUCgsMCHTCurgsCMEp9cl4sagQdQCdXmBjQjaigpw5ZT5JZkiIC9KKqpCQWSp1FKI2Y1Eh6gDHrhaioFzLwz4kqR6eajjaWGIDD/+QEWNRIeoA609nwUNthWBXTplP0pHLZRgU5IyNCdmo1+mljkPUJiwqRO2spk6HbWdzMTjEGTIZD/uQtIaEuKCwohZHrhRKHYWoTVhUiNrZ3ov5qNDWY2gwD/uQ9IJcbOHlYMWzf8hosagQtbMN8VkIdrWFp4O11FGIIJPJMDjYBdvP5qK6lldUJuPDokLUjkqr67D3Yj4Gc28KGZAhwS6oqtVh78V8qaMQtRqLClE72nE2F/U6gUFBvFIyGQ4PjRVCXG159g8ZJRYVona0ISEL3b3UcLLllZLJsEQFu2DfpXyUVtdJHYWoVQymqHzwwQeQyWR4+eWXpY5C1Cb5ZTU4eqWQh33IIEUFO6NeJ7DjbK7UUYhaxSCKysmTJ/H111+jV69eUkcharPNZ3Igl8kwIJBXSibD42ijRA8vNc/+IaMjeVGpqKjAE088gW+//RaOjo5SxyFqsw0JWejt6wA7lULqKETNGhzsgmNXC5FfViN1FKIWk7yozJo1C5MnT8aYMWPuuK5Wq0VZWVmjG5EhSC+sQkJGKaKCOYiWDFf/QCdYyGXYfCZH6ihELSZpUVmxYgXi4uIwf/78Fq0/f/58aDSahpuvr28HJyRqmU1nsqFSyNHHj3sFyXDZqRSI8HHAhgQe/iHjIVlRycjIwN/+9jcsX74cVlZWLXrMvHnzUFpa2nDLyMjo4JRELbMhPgt9/R1hZWkhdRSi24oKdkZCRinSC6ukjkLUIpIVldjYWOTn56NPnz5QKBRQKBQ4cOAAPvvsMygUCuh0TWdQVKlUUKvVjW5EUruUW47LeRU824eMQh8/R6gUcmw6ky11FKIWkWzU3+jRo5GYmNho2dNPP42uXbvitddeg4UF/2VKxmFjQhZsVRaI8NFIHYXojqwsLdDX3xEb4rMwa2SI1HGI7kiyomJvb4+ePXs2WmZrawtnZ+cmy4kMlRACG+OzMSDACQoLycemE7VIVLAzFuy8jEu55QjzsJc6DtFt8ZOV6C4kZJYio7iah33IqET4OMBWZYFNCTz8Q4bPoCZ82L9/v9QRiFplU0I2HKwt0d2T46XIeFhayDEgwAkbE7Lx93FdIJPJpI5EdEvco0LURjq9wKaEbAwMcoZczg96Mi5RwS5IL6pCYlap1FGIbotFhaiNTqYWIb9ci8Gc5I2MUHdPNTTWljz8QwaPRYWojTYlZMPVToUQNzupoxC1moVchoGBTtiUkAO9Xkgdh+iWWFSI2qBOp8fWxBwMCnKCnMf3yUhFBTsjt6wGp9KKpY5CdEssKkRtcDj5Goqr6hDFs33IiHVxt4eznZKHf8igtamoXL16tb1zEBmVTQk58NRYIcDZRuooRG0ml8kwKNAZWxJzUK/TSx2HqFltKiohISEYOXIkfv75Z9TU8HLhZF609TrsOJeLqCBnntZJRi8q2BlFlbU4drVI6ihEzWpTUYmLi0OvXr0wZ84ceHh44Pnnn8eJEyfaOxuRQTpwqQAV2npE8WwfMgFBLrZwV6t4+IcMVpuKSu/evfHpp58iOzsbP/zwA3JycjB06FD07NkTn3zyCQoKCto7J5HB2JSQDT8nG/g48rAPGT+ZTIaoIGdsPZuD2noe/iHDc1eDaRUKBR544AGsWrUKH374IZKTk/Hqq6/C19cX06dPR05OTnvlJDIIVbX12HUhD4OCuDeFTMegIGeU19TjUDL/kUmG566KyqlTp/CXv/wFnp6e+OSTT/Dqq6/iypUr2LVrF7KzszF16tT2yklkEPZezEdNnR5RLCpkQq7vIbTGxnge/iHD06Zr/XzyySeIiYnBpUuXMGnSJCxbtgyTJk2CXH699wQGBmLp0qUICAhoz6xEktuckI1gV1t4aKykjkLUbmQyGQYFOWNbYg5q6nSwsrSQOhJRgzbtUVm8eDH+9Kc/IS0tDevXr8eUKVMaSsoNbm5u+P7779slJJEhKK+pw96LBTzsQyYpKsgZlbU67L+UL3UUokbatEdl165d8PPza1JOhBDIyMiAn58flEoloqOj2yUkkSHYfSEPtTo9iwqZJC8HawQ422BTQg4m9PSUOg5RgzbtUQkODsa1a9eaLC8qKkJgYOBdhyIyRBvjsxHmbg8XO5XUUYg6xKAgZ+y5kIdKbb3UUYgatKmoCNH8BawqKipgZcVj92R6Sqvq8HvSNQwKcpI6ClGHGRTkjJp6PfZc5OEfMhytOvQzZ84cANcHXr3xxhuwsfm/eSR0Oh2OHz+O3r17t2tAIkOw41wudHqBAYE87EOmy11thWBXW2xOyMZ9EV5SxyEC0Mqicvr0aQDX96gkJiZCqVQ2/EypVCIiIgKvvvpq+yYkMgAbE7LR3UsNJ1vlnVcmMmJRQS747VQ6ymrqoLaylDoOUeuKyr59+wAATz/9ND799FOo1eoOCUVkSAortDh6pRDRgwOkjkLU4QYFOeHn42nYdS4PD/b1kToOUdvGqMTExLCkkNnYdjYXAgIDAzk+hUyfs50KYR722HSGk7+RYWjxHpUHHngAS5cuhVqtxgMPPHDbddeuXXvXwYgMxaaEbPT00kBtzd3gZB6igpzx87E0lFTVwsGGhztJWi3eo6LRaBouaa/RaG57IzIVeWU1OJFShEG8UjKZkYGBTtALge1nc6WOQtTyPSoxMTHN/jeRKduamAMLuQz9A3jYh8yHg40S3TzV2HQmG48N8JM6Dpm5No1Rqa6uRlVVVcP9tLQ0LFq0CDt37my3YESGYFNCNnr5aGCnatMkzkRGKyrIGUevFOJahVbqKGTm2lRUpk6dimXLlgEASkpKMGDAACxYsABTp07F4sWL2zUgkVSyS6oRl17CKfPJLPX/3+DxbTz8QxJrU1GJi4vDsGHDAACrV6+Gh4cH0tLSsGzZMnz22WftGpBIKlvO5MDSQoa+/o5SRyHqdGorS4R7a7Apnmf/kLTaVFSqqqpgb28PANi5cyceeOAByOVyDBo0CGlpae0akEgqGxOy0dvXATZKHvYh8xQV7IyTqUXIK6uROgqZsTYVlZCQEKxfvx4ZGRnYsWMHxo0bBwDIz8/n/CpkEtIKK5GYVYooHvYhM9bP3wkKCxm2nMmROgqZsTYVlTfeeAOvvvoqAgICMHDgQERFRQG4vnclMjKyXQMSSWHzmRyoFHJE+vGwD5kvW5UCvbwdOPkbSapNReWhhx5Ceno6Tp06he3btzcsHz16NBYuXNhu4YiksikhG338HGFlaSF1FCJJRQU743R6CTKLq+68MlEHaFNRAQAPDw9ERkZCLv+/pxgwYAC6du3aLsGIpJKcX4GLueU87EMEoI+fI5QWch7+Icm0aZRgZWUlPvjgA+zZswf5+fnQ6/WNfn716tV2CUckhc1nsmFtaYEIXwepoxBJzlppgd5+DtiUkI3nhwdLHYfMUJuKysyZM3HgwAE89dRT8PT0bJhan8jYCSGwMSEbff0doVS0eYcjkUkZHOSMRXuSkHqtEgEutlLHITPTpqKybds2bNmyBUOGDGnvPESSuphbjqsFlXiwDy9vT3RDbz8HWFnKsflMNmaPCpU6DpmZNv2T0dHREU5OvPYJmZ7NZ7Jhp1Kglzcvrkl0g0phgb7+jtiUwHEq1PnaVFTeffddvPHGG42u90Nk7IQQ2Bifjf4BTlBY8LAP0R8NCnLGpbxyJOWVSx2FzEybDv0sWLAAV65cgbu7OwICAmBpadno53Fxce0SjqgzJWaVIqO4Gk9FBUgdhcjgRPg4wFZpgU1ncjBnrL3UcciMtKmoTJs2rZ1jEElvU0I2NNaW6O7J2ZWJbmZpIUe/ACdsjM/CK2NCeRIFdZo2FZU333yzvXMQSUqvF9h8JgcDAp1gIecHMFFzooKcceByAc5ll6Enx3FRJ2nzgfiSkhJ89913mDdvHoqKigBcP+STlZXVbuGIOktcejFySmswOJiTvBHdSg9vNdRWCmzm5G/UidpUVM6cOYMuXbrgww8/xMcff4ySkhIAwNq1azFv3rz2zEfUKTYlZMPZVoku7jz2TnQrCrkc/QOcsDEhC0IIqeOQmWhTUZkzZw5mzJiBpKQkWFlZNSyfNGkSDh482G7hiDpDvU6PzWdyMDDIGXIedye6rcHBzsguqcHpjBKpo5CZaFNROXnyJJ5//vkmy729vZGbm3vXoYg60/GUIhRW1vKwD1ELdPVQw9HGEpsSeEVl6hxtKioqlQplZWVNll++fBmurq53HYqoM21KyIa7WoUgTg1OdEdyuQwDg5yx+UwOdHoe/qGO16aict999+Gdd95BXV0dAEAmkyE9PR2vvfYaHnzwwRY/z+LFi9GrVy+o1Wqo1WpERUVh27ZtbYlE1Ca19XpsTcxBVJAzT7ckaqHBQc4oKNfieEqh1FHIDLSpqCxYsAAVFRVwdXVFdXU1hg8fjpCQENjb2+O9995r8fP4+Pjggw8+QGxsLE6dOoVRo0Zh6tSpOHfuXFtiEbXa70kFKKupx6AgHvYhaqkQNzu42as4pT51Cpm4i6Hbhw8fRkJCAioqKtCnTx+MGTPmrgM5OTnho48+wrPPPnvHdcvKyqDRaFBaWgq1mpN0Uev9bcVpxKYV4/892It7VIha4dcT6ThwuQCn/j0GlrzkBLVSa76/Wz3hm16vx9KlS7F27VqkpqZCJpMhMDAQHh4eEEK0+cNep9Nh1apVqKysRFRUVLPraLVaaLXahvvNjZMhaqnqWh12nsvDlF6eLClErTQ42BkbE7JxKOkaRnZ1kzoOmbBW1WAhBO677z7MnDkTWVlZCA8PR48ePZCWloYZM2bg/vvvb3WAxMRE2NnZQaVS4YUXXsC6devQvXv3ZtedP38+NBpNw83X17fVr0d0w96L+aiu02FwsIvUUYiMjp+TDbwdrHn2D3W4VhWVpUuX4uDBg9izZw9Onz6NX3/9FStWrEBCQgJ2796NvXv3YtmyZa0KEBYWhvj4eBw/fhwvvvgioqOjcf78+WbXnTdvHkpLSxtuGRkZrXotoj/amJCFYFdbeGis7rwyETUik8kQFeyM7edyUVOnkzoOmbBWFZVff/0Vr7/+OkaOHNnkZ6NGjcLcuXOxfPnyVgVQKpUICQlB3759MX/+fERERODTTz9tdl2VStVwhtCNG1FblNXUYd/FAg6iJboLg4OcUVWrw76L+VJHIRPWqqJy5swZTJgw4ZY/nzhxIhISEu4qkF6vbzQOhagj7DyXh1qdHlEsKkRt5ulgjSAXW2zg4R/qQK0aTFtUVAR3d/db/tzd3R3FxcUtfr558+Zh4sSJ8PPzQ3l5OX755Rfs378fO3bsaE0solbbGJ+Fbp72cLZTSR2FyKhFBTtj1alMlNfUwd7KUuo4ZIJatUdFp9NBobh1t7GwsEB9fX2Lny8/Px/Tp09HWFgYRo8ejZMnT2LHjh0YO3Zsa2IRtcq1Ci0OJxciKoiDaInuVlSQM2p1euw4lyd1FDJRrdqjIoTAjBkzoFI1/6/Q1h6y+f7771u1PlF72JqYA8iAgUFOUkchMnrOdip087DHxvgsPNTXR+o4ZIJaVVSio6PvuM706dPbHIaoM2yIz0a4twZq7qYmahdRwc748UgarlVo4cLDqdTOWlVUYmJiOioHUafILK5CbFox/jIiWOooRCZjYJAzfjyahq2JOZgeFSB1HDIxnPeYzMqmhBwoLeTo58/DPkTtRW1liXBvDTbE8+wfan8sKmRW1sdnoY+/A6yVFlJHITIpg4OdEZtWjMziKqmjkIlhUSGzcSm3HJdyyzGEU+YTtbt+/k5QKeTYyDlVqJ2xqJDZ2JiQBVuVBSJ8HaSOQmRyrJUW6OPniA2nWVSofbGokFkQQmD96WwMCHDmJemJOsiQEBdcyivHxVxe2Z7aDz+xySzEpRcjq6QaQ0M4ZT5RR4nw0cDeSsFBtdSuWFTILGyIz4azrRJdPXghS6KOorCQY0CAEzbEZ0GvF1LHIRPBokImr06nx+YzORgU5Ay5XCZ1HCKTNiTEBdklNYhNb/l134huh0WFTN6hpGsoqqzFkBCe7UPU0cI87OFip8T601lSRyETwaJCJm/96Sz4OFojwNlG6ihEJk8uk2FwsAs2n8lBbb1e6jhkAlhUyKRVauux43wuhgS7QCbjYR+izjA0xAWl1XU4cLlA6ihkAlhUyKTtPJ+Lmjo9hvBsH6JO4+tkA39nG6w7nSl1FDIBLCpk0tafzkY3T3u42ltJHYXIrAwJdsGeC/koq6mTOgoZORYVMlkF5VocSrqGwZwyn6jTDQ52Rm29HtvP5kodhYwciwqZrE0J2ZDJgEGBPOxD1Nmc7VTo7qXGujie/UN3h0WFTNbauExE+jnAzkohdRQiszQ0xAXHrhYip7Ra6ihkxFhUyCQl55fjbHYZhoa4Sh2FyGwNCHSCpYWcU+rTXWFRIZO0Ni4LdioFIv0cpI5CZLZslAr09XfE2rhMCMEp9altWFTI5Oj1AutOZ2Hg//41R0TSGRrqgst5FTifwysqU9vwU5xMzonUIuSU1mBYKA/7EEmtl48GGmtLDqqlNmNRIZOzNi4T7moVurjbSR2FyOwp5HJEBTljfXwW6nWcUp9aj0WFTEp1rQ5bzuRgaAinzCcyFMNCXXCtohaHkq9JHYWMEIsKmZSd53NRWavjYR8iAxLoYgsfR2usieWU+tR6LCpkUtbEZqKrhz3c1Zwyn8hQyGQyDAt1xc7zeZxSn1qNRYVMRl5ZDQ4lX8PQUE6ZT2Rohoa4oE6nx9YzOVJHISPDokImY/3pLFjIZYgK4pT5RIbGyVaJnt4arInj4R9qHRYVMglCCKyOzUQ/fyfYKDllPpEhuifUFSdTi5FWWCl1FDIiLCpkEhKzSpGUX4HhXTiIlshQ9QtwhI3SAms4pwq1AosKmYTVsZlwslUi3FsjdRQiugWVwgKDgpyxOjYDej2n1KeWYVEho6et12H96SwMDXGBXM65U4gM2fAursguqcGxq4VSRyEjwaJCRm/PhXyU1dTjHh72ITJ4oW528NRYYRXnVKEWYlEho7fqVAZC3e3g7WAtdRQiugOZTIZ7Ql2xLTEH5ZxThVqARYWMWn5ZDQ5evoZhIdybQmQshoW6QFuvxxbOqUItwKJCRm1N3PW5U4aEcO4UImPhbKdCLx8NVp7KkDoKGQEWFTJaQgj8diodAwI5dwqRsRnexQ1x6SVIzq+QOgoZOBYVMlqn0oqReq0KI8J42IfI2PQLcIS9SoFVsdyrQrfHokJGa+XJDLirVejmqZY6ChG1kqWFHENCXLA6NhN1Or3UcciAsaiQUarQ1mPzmRzcE+oKuYxzpxAZoxFhriisqMWBSwVSRyEDxqJCRmlzQjZq6nScMp/IiPk72yLIxRYrTqZLHYUMGIsKGaVfT6YjwtcBznYqqaMQ0V0YEeaKfRcLkFdWI3UUMlAsKmR0LuaWISGjFKPC3KSOQkR3aUiICxQWMqzmTLV0CywqZHRWnMiAxtoSkf4OUkchortko1RgYKATVpxI54UKqVmSFpX58+ejf//+sLe3h5ubG6ZNm4ZLly5JGYkMXE2dDmvjMnFPqAsUcvZsIlMwsqsbMoqrcZQXKqRmSPpJf+DAAcyaNQvHjh3Drl27UFdXh3HjxqGyslLKWGTAdpzLRVlNPUZ25WEfIlMR5m4Pbwdr/HqCg2qpKUmn89y+fXuj+0uXLoWbmxtiY2Nxzz33SJSKDNkvx9PR3VMNTw0vQEhkKmQyGUaEuWLlqQwUVmg5SJ4aMah956WlpQAAJyenZn+u1WpRVlbW6EbmIzm/AsdTijCKe1OITM49/5tqYE0cB9VSYwZTVPR6PV5++WUMGTIEPXv2bHad+fPnQ6PRNNx8fX07OSVJ6dcT6VBbKTAgsPkiS0TGS21liQGBTlh+PB1CcFAt/R+DKSqzZs3C2bNnsWLFiluuM2/ePJSWljbcMjJ4jQhzUVOnw+rYTNzTxRWWFgbztiWidjS6qzvSCqtw9AoH1dL/MYhLzs6ePRubN2/GwYMH4ePjc8v1VCoVVCoeuzRH287moLS6jod9iExYVw97+DhaY/nxdAwOcZE6DhkISf9pKoTA7NmzsW7dOuzduxeBgYFSxiEDtvxYOnp6cRAtkSmTyWQYGeaGHedyUVCulToOGQhJi8qsWbPw888/45dffoG9vT1yc3ORm5uL6upqKWORgbmYW4ZTacUY3c1d6ihE1MFuXGh05Ske2qfrJC0qixcvRmlpKUaMGAFPT8+G22+//SZlLDIwy4+lw9HGEv0CHKWOQkQdzM5KgahgZyw/lgYdZ6olSDxGhSO76U4qtPVYE5eJCT09OBMtkZkY290dBy4XYP+lfO5JJcM564eoOetPZ6GmTscLEBKZkWBXOwS72mLZ0TSpo5ABYFEhgyWEwE9H09DX35EzVRKZmTHd3HHwcgHSCnlJFXPHokIG60RKES7llWMMd/0SmZ2oYGfYqhRYfpzX/zF3LCpksH48mgpvB2uEe2ukjkJEnUylsMCIMFesOJGO6lqd1HFIQiwqZJBySqux42wexnV3h0wmkzoOEUlgbDd3lNfUY318ltRRSEIsKmSQlh9Lh1Ihx7BQV6mjEJFE3NRW6OvviKVHUnmWqBljUSGDU1Onwy8n0nFPF1dYKy2kjkNEEhrb3R2XcstxIqVI6igkERYVMjhbzuSgqLIWY7tzEC2RuQv31sDbwRoxR1KljkISYVEhgyKEwPeHUtDb9/qHExGZN5lMhvE93LHzXC4yi6ukjkMSYFEhg3IipQjnc8owoYen1FGIyEAMC71+GJgTwJknFhUyKD8cToGPozV6+fCUZCK6zsrSAqPC3PDr8XRUauuljkOdjEWFDEZ6YRV2nsvD+B4ePCWZiBoZ18MDlbXXr/1F5oVFhQzG0iOpsFUpMCzUReooRGRgXOxUGBDohB8OpUDPqyqbFRYVMgil1XVYcTIdo7u5QaXgKclE1NSknp5ILazC7gt5UkehTsSiQgZhxYl01On0GN/DQ+ooRGSgQt3tEeZhj29/vyp1FOpELCokudp6PX44nILBwS5wtFFKHYeIDNjknp44mVqM0+nFUkehTsKiQpLbfCYbeWVaTA7nKclEdHt9/R3hobHiXhUzwqJCkhJC4JuDV9Hb1wG+TjZSxyEiAyeXyzCppwe2n81FeiEngDMHLCokqQOXC3Axt5x7U4ioxe7p4go7lYJ7VcwEiwpJavH+Kwhxs0MPL7XUUYjISKgUFhjfwwMrT2WgoFwrdRzqYCwqJJm49GIcTynCvb28OMEbEbXKuO4ekMmApUdSpI5CHYxFhSSzZP8VeDlYo1+Ao9RRiMjI2FkpMKqrO5YdSUN5TZ3UcagDsaiQJJLyyrHzfB6mhHtCzr0pRNQGk3p6oLpOh+XH06WOQh2IRYUk8eW+ZDjbKjldPhG1mbOdCsNCXfHtwauoqdNJHYc6CIsKdbq0wkpsTMjGvRFeUFjwLUhEbTe1txeKq2qx4gT3qpgqfktQp1u8/wrUVpYYGeYmdRQiMnLuaisMDnbBkgNXUVuvlzoOdQAWFepUWSXVWB2bicm9PKFU8O1HRHdvWm9v5JXVYE1cptRRqAPwm4I61eL9ybC2tMCYbu5SRyEiE+HtaI2BQU74cl8y6nTcq2JqWFSo02SVVGPFiQxM7uUJK0sLqeMQkQl5INIHmcXVWBPLvSqmhkWFOs1X+5Jho7w+oyQRUXvydbLBoCAnfL43mWNVTAyLCnWKzOIq/HYyA5N7eXFvChF1iAcifZBdUs2xKiaGRYU6xZf/25syrjvHphBRx7i+V8UZn+9Jgrae86qYChYV6nCp1yqx8mQm7o3g3hQi6lgP9vFBblkNVpzIkDoKtRMWFepwC3ddhsbGEuO6c2wKEXUsb0drDA11wWd7k1BVWy91HGoHLCrUoS7klGFjQjbuj/TmvClE1CkejPRBaVUdlh5JlToKtQN+c1CHWrDzEtzUKowIc5U6ChGZCTe1FUZ1dcOS/VdQWs0rKxs7FhXqMKdSi7D7Qj4e6usLhZxvNSLqPNMivaGt12Px/itSR6G7xG8P6hBCCLy39QICXWwxONhZ6jhEZGYcbZSYHO6JmMMpyC6pljoO3QUWFeoQO87l4XR6CR4f4Ae5TCZ1HCIyQ1P+N2/TJ7suSR2F7gKLCrW7Op0eH2y/gAgfDcK9NVLHISIzZa20wAOR3lgTm4ULOWVSx6E2YlGhdvfriXSkXavC4wP8pI5CRGZuVDc3eGis8N6WCxBCSB2H2oBFhdpVaVUdFuy8jBFhrvB3tpU6DhGZOYVcjj8N8MOh5GvYdylf6jjUBiwq1K4+3ZOE2no9HunnK3UUIiIAQF9/R/T0UuPdzRdQp+MFC42NpEXl4MGDuPfee+Hl5QWZTIb169dLGYfu0pWCCvx4NBX39faCg41S6jhERAAAmUyGJwf5I62wEj8fS5M6DrWSpEWlsrISERER+PLLL6WMQe1ACIF3Np2Hk60Sk3p6Sh2HiKgRf2dbjAxzwye7LqOwQit1HGoFhZQvPnHiREycOFHKCNROdl/Ix4HLBZgztgunyicig/RIf18cTynC/9txCR8+2EvqONRC/Eahu1ZTp8NbG88hwkeDfv6OUschImqW2soSD/fzwcqTGYjPKJE6DrWQURUVrVaLsrKyRjeS3pIDV5BXVoPoqADIOLkbERmwMV3d4e9sg/+sPwudnqcrGwOjKirz58+HRqNpuPn68swSqV0tqMBX+65gci9PeDpYSx2HiOi25HIZnh4SiMSsUvxynANrjYFRFZV58+ahtLS04ZaRkSF1JLMmhMC/1p2Fo60l7o/0ljoOEVGLdHG3x+iubvhw+yXkldVIHYfuwKiKikqlglqtbnQj6ayPz8LRq4WYMTgQKoWF1HGIiFrssQF+sJDL8M6m81JHoTuQtKhUVFQgPj4e8fHxAICUlBTEx8cjPT1dyljUAkWVtXh38wVEBTmjt6+D1HGIiFrFTqXAU4P8sSUxB3sv5kkdh25D0qJy6tQpREZGIjIyEgAwZ84cREZG4o033pAyFrXA2xvPobZej+lR/lJHISJqk8HBzujtq8G8tYkoq6mTOg7dgqRFZcSIERBCNLktXbpUylh0B7vP52FDQjamR/lzBloiMloymQzPDg1CWXU95m+9KHUcugWjGqNC0iutrsPr6xIR6euAoSEuUschIrorLnYqPD7AD7+eSMfh5GtSx6FmsKhQq7y98RwqtPV4dmgg50whIpMwupsbenip8Y9VCSjnISCDw6JCLbb9bA7Wns5CdFQAnO1UUschImoXcpkMz98ThOKqOry96ZzUcegmLCrUIvnlNZi7NhH9AxwxLJSHfIjItLjaW2F6lD9Wx2Zhx7lcqePQH7Co0B3p9QL/WHUGQgAzhwbxkA8RmaThXVzRz98Rc9ecQT4ngjMYLCp0RzFHUnHgcgGevycIamtLqeMQEXUImUyGPw8LAgC8sjIeel4LyCCwqNBtnc0qxQfbLmBiTw9E+vHKyERk2tTWlnhxRAiOJBfim9+vSh2HwKJCt1FeU4fZv8bB28Eajw/wkzoOEVGnCPfW4N4IL3y04xJi04qkjmP2WFSoWUIIvLbmDPJKtXhpdCgsLfhWISLz8XA/H4S42WHW8tMoqqyVOo5Z47cPNevHI6nYmpiL54cHwVNjLXUcIqJOpZDL8dKoUFTX6fC3Faeh43gVybCoUBMnU4vw3y3Xx6UMDHSWOg4RkSScbJWYNTIEh5KuYdHuy1LHMVssKtRITmk1XvgpFqHudvjTQI5LISLzFu6twaP9ffH53mRsS8yROo5ZYlGhBjV1Ojy3LBYyGfC30V2gkPPtQUR0X4QXBgU5Yc7KBFzMLZM6jtnhNxEBuD6p26urEnAptxxzxoZBw/lSiIgAXJ9f5fl7guGuVuGZpSdRUK6VOpJZYVEhAMAnuy5j85kc/GVkMAJdbKWOQ0RkUKwsLfD3cWGortVh5o8nUV2rkzqS2WBRIaw8mYEv9iXjTwP8OHiWiOgWXOxUeHVcGC7mluOVlfE8E6iTsKiYuT0X8jBvbSJGd3XDlF6eUschIjJoQa52mD0qBDvP5eKtjecgBMtKR2NRMWOxaUWYtTwOffwd8MyQQF5skIioBfr5O+GZoYH46VgaPt+bLHUck6eQOgBJ42xWKZ6OOYlAV1vMHhkKuZwlhYiopUZ3dUdpVR0+2XUZaisFZgwJlDqSyWJRMUOX88rx5PfH4Wp//XirUsEda0RErXV/pDeqanV4a9N5qCwteE20DsKiYmaS88vxp2+PwcHaEnMndIONkm8BIqK2kMlkeGKgH+p0ery+NhEKuQwP9/OVOpbJ4beUGbmYW4Y/fXscdioLzJ3YDXZW/PMTEd0NmUyG6MEB0OkF/rn6DOr1gntW2hm/qczE2axSPPn9cThYW2LepG5QW3FCNyKi9iCXyfDM0EBYyGWYtzYR2jodx6y0IxYVM3Ak+Rr+vOwUPB2s8dqErrBT8c9ORNSe5DIZZgwOgKWFHG9tOo+iqjq8MiaUZ1O2A35jmbgtZ3Lw8m+n0d1TjZfHdIGVpYXUkYiITNKNMStqKwU+25OEaxVavHNfDygseMLC3WBRMVFCCCw5cBUfbr+IISHOeOGeYP7PQkTUwWQyGe7r7Q17a0t89/tVZBVX44s/RcKeh9vbjN9cJkhbr8Nra87gw+0X8UCkN2aNCGFJISLqRCPD3PDahK44mVqEh5ccRUZRldSRjBa/vUxMXlkNHv36GNadzsKLw4PxcD9fHiMlIpJALx8HvHVvDxRX1eLeLw7hSPI1qSMZJRYVE3LsaiEmf3YIGUVVeGNKD9zTxVXqSEREZs3XyQbvTu0JP0cbPPn9cSw5cAV6XsywVVhUTIBOL/Dp7iT86dtjcLNX4b/TeiLEzU7qWEREBMDeyhL/nNAVU3p54YNtF/H00pMoqqyVOpbRkAkjvvRjWVkZNBoNSktLoVarpY4jicziKvx9ZQJOpBTh/j7eeDDSh9ftISIyUPEZxVi8/wpUlhb4+OEIDDfTPd+t+f7mHhUjJYTAylMZGL/wIK4WVODfk7vh4b6+LClERAast68j5j/QC54aK0T/cAL/WX8Wldp6qWMZNO5RMUIZRVWYtzYRh5Kv4Z5QF0QPDuA1e4iIjIgQArvO5+GXE+lwtlNi/gO9zGrvSmu+v1lUjIi2Xofvfk/B53uTYK9S4JmhQejt6yB1LCIiaqP8shp8dygFiVmlmNLLE/+e3B0eGiupY3U4FhUTI4TAngv5+O/W80gvrMKEnp54qI8PrJWcZZaIyNgJIfB70jX8ciIddTo9Zo8KwTNDAk16JnEWFROSkFGC97dewPGUIvT0UmN6VAB8nWykjkVERO2sUluP1XGZ2HU+D272Krw6LgzTIr1hYYJjD1lUTMC57FIs3HUZuy/kw8fRGo8P8EOkrwMnbyMiMnE5pdVYcSIDJ1KLEOxqi1fGdsGknp4mdbIEi4qREkLgVFoxvtyXjP2XCuChtsIDfbwxJNjFpN6gRER0Z8n5FVgdm4GEzFIEudrixeHBmNrbG0qF8Z+wy6JiZGrr9dh2Ngc/HEpBQmYpfBytcV+EF6KCnaGQG/8bkoiI2i4prxwbE7JxKq0YrvYqREf54/EBfnC2U0kdrc1YVIxE6rVKrDiZgVWnMlBYWYtwbw0m9PBAbz8HyHmIh4iI/iCzuArbz+bi96Rr0AuBiT098PhAPwwKdDa6ve4sKgasqLIW28/mYk1cJmLTimGrtMCwUFeM6urGQbJERHRH5TV1OHj5GvZezEN2aQ28HKzwUB8f3BvhhVB3e6njtQiLioHJL6vBrgt52HE2F4eTCyEgEO6twbBQV/QPcDKJ441ERNS5hBC4lFeOg5ev4XhKIapqdQhzt8fEcA+M7e6O7p5qgz0Bg0VFYrX1esRnlOD3pALsu5iPs9llkMuAbp5qDAx0woBAZ2isLaWOSUREJqK2Xo8zmSU4llKI0+klqKrVwVNjhRFhrhjexRVRQS7Q2BjO9w6LSierqq3HmcxSnEotwomUIpxILUJNnR72VgqEe2vQ29cBkb6OsLPiNPdERNSx6nV6nM8pw+mMEpzJLEF2SQ1kAHp4qTEwyBn9A5zQx98BbvbSzYBrdEXlyy+/xEcffYTc3FxERETg888/x4ABA+74OCmKSoW2Hpdyy3E+pwznskoRn1GCpLwK6ISAjdICXdzt0c1TjZ5eagQ42xrdACciIjItBeVanM0uxbnsMlzOLUdBhRYA4KmxQm9fB/T01qCntwbdPOzhaq/qlMNFRlVUfvvtN0yfPh1LlizBwIEDsWjRIqxatQqXLl2Cm5vbbR/bkUWlTqfHqdRipFyrRMq1CiTnVyApvwKZxdUAAAu5DH5O1vB3skWImx1C3Ozg62jDYkJERAatsEKLpPzr32vXv+MqUV2nAwA42lgixM0Ooe72CHKxRaCLLSL9HOFkq2zXDEZVVAYOHIj+/fvjiy++AADo9Xr4+vrir3/9K+bOnXvbx3ZkUfnm4BW8v/Ui5DLA1V4FLwdreDtYw8fRBn5ONvB2sOYgWCIiMnp6IZBfpkVGURXSi6uQVVyN7NJq5JbWQFuvx4AAJ6x8IapdX7M139+SDpqora1FbGws5s2b17BMLpdjzJgxOHr0qITJgKpaHRxsLPH5Y5FQWLCQEBGRaZLLZPDQWMFDY4X+gU4Ny/VC4JuDV3Htf4eKpCJpUbl27Rp0Oh3c3d0bLXd3d8fFixebrK/VaqHV/t8GKy0tBXC9mbW3msoKVFWUY83xpHZ/biIiImOQmlMMlaVFu3/P3ni+lhzUMarTUObPn4+33367yXJfX98Oe81PO+yZiYiIjIPmtY553vLycmg0mtuuI2lRcXFxgYWFBfLy8hotz8vLg4eHR5P1582bhzlz5jTc1+v1KCoqgrOzc7OjlMvKyuDr64uMjAyDmmels3E7XMftcB23w3XcDtdxO1zH7XBdZ20HIQTKy8vh5eV1x3UlLSpKpRJ9+/bFnj17MG3aNADXy8eePXswe/bsJuurVCqoVI0vwuTg4HDH11Gr1Wb9xruB2+E6bofruB2u43a4jtvhOm6H6zpjO9xpT8oNkh/6mTNnDqKjo9GvXz8MGDAAixYtQmVlJZ5++mmpoxEREZHEJC8qjz76KAoKCvDGG28gNzcXvXv3xvbt25sMsCUiIiLzI3lRAYDZs2c3e6jnbqlUKrz55ptNDheZG26H67gdruN2uI7b4Tpuh+u4Ha4zxO0g+YRvRERERLfCmcyIiIjIYLGoEBERkcFiUSEiIiKDxaJCREREBsvoi8qXX36JgIAAWFlZYeDAgThx4sRt11+1ahW6du0KKysrhIeHY+vWrZ2UtGO1ZjssXboUMpms0c3KyqoT03aMgwcP4t5774WXlxdkMhnWr19/x8fs378fffr0gUqlQkhICJYuXdrhOTtaa7fD/v37m7wfZDIZcnNzOydwB5g/fz769+8Pe3t7uLm5Ydq0abh06dIdH2dqnw9t2Q6m+PmwePFi9OrVq2ESs6ioKGzbtu22jzG19wLQ+u1gKO8Foy4qv/32G+bMmYM333wTcXFxiIiIwPjx45Gfn9/s+keOHMHjjz+OZ599FqdPn8a0adMwbdo0nD17tpOTt6/Wbgfg+qyDOTk5Dbe0tLROTNwxKisrERERgS+//LJF66ekpGDy5MkYOXIk4uPj8fLLL2PmzJnYsWNHByftWK3dDjdcunSp0XvCzc2tgxJ2vAMHDmDWrFk4duwYdu3ahbq6OowbNw6VlZW3fIwpfj60ZTsApvf54OPjgw8++ACxsbE4deoURo0ahalTp+LcuXPNrm+K7wWg9dsBMJD3gjBiAwYMELNmzWq4r9PphJeXl5g/f36z6z/yyCNi8uTJjZYNHDhQPP/88x2as6O1djvExMQIjUbTSemkAUCsW7futuv885//FD169Gi07NFHHxXjx4/vwGSdqyXbYd++fQKAKC4u7pRMUsjPzxcAxIEDB265jql+PvxRS7aDOXw+CCGEo6Oj+O6775r9mTm8F2643XYwlPeC0e5Rqa2tRWxsLMaMGdOwTC6XY8yYMTh69Gizjzl69Gij9QFg/Pjxt1zfGLRlOwBARUUF/P394evre8dGbapM8f1wN3r37g1PT0+MHTsWhw8fljpOuyotLQUAODk53XIdc3g/tGQ7AKb9+aDT6bBixQpUVlYiKiqq2XXM4b3Qku0AGMZ7wWiLyrVr16DT6ZpMte/u7n7LY+u5ubmtWt8YtGU7hIWF4YcffsCGDRvw888/Q6/XY/DgwcjMzOyMyAbjVu+HsrIyVFdXS5Sq83l6emLJkiVYs2YN1qxZA19fX4wYMQJxcXFSR2sXer0eL7/8MoYMGYKePXvecj1T/Hz4o5ZuB1P9fEhMTISdnR1UKhVeeOEFrFu3Dt27d292XVN+L7RmOxjKe8EgptCnzhUVFdWoQQ8ePBjdunXD119/jXfffVfCZCSFsLAwhIWFNdwfPHgwrly5goULF+Knn36SMFn7mDVrFs6ePYtDhw5JHUVSLd0Opvr5EBYWhvj4eJSWlmL16tWIjo7GgQMHbvklbapasx0M5b1gtEXFxcUFFhYWyMvLa7Q8Ly8PHh4ezT7Gw8OjVesbg7Zsh5tZWloiMjISycnJHRHRYN3q/aBWq2FtbS1RKsMwYMAAk/hinz17NjZv3oyDBw/Cx8fntuua4ufDDa3ZDjczlc8HpVKJkJAQAEDfvn1x8uRJfPrpp/j666+brGvK74XWbIebSfVeMNpDP0qlEn379sWePXsalun1euzZs+eWx9uioqIarQ8Au3btuu3xOUPXlu1wM51Oh8TERHh6enZUTINkiu+H9hIfH2/U7wchBGbPno1169Zh7969CAwMvONjTPH90JbtcDNT/XzQ6/XQarXN/swU3wu3crvtcDPJ3gtSj+a9GytWrBAqlUosXbpUnD9/Xjz33HPCwcFB5ObmCiGEeOqpp8TcuXMb1j98+LBQKBTi448/FhcuXBBvvvmmsLS0FImJiVL9Cu2itdvh7bffFjt27BBXrlwRsbGx4rHHHhNWVlbi3LlzUv0K7aK8vFycPn1anD59WgAQn3zyiTh9+rRIS0sTQggxd+5c8dRTTzWsf/XqVWFjYyP+8Y9/iAsXLogvv/xSWFhYiO3bt0v1K7SL1m6HhQsXivXr14ukpCSRmJgo/va3vwm5XC52794t1a9w11588UWh0WjE/v37RU5OTsOtqqqqYR1z+Hxoy3Ywxc+HuXPnigMHDoiUlBRx5swZMXfuXCGTycTOnTuFEObxXhCi9dvBUN4LRl1UhBDi888/F35+fkKpVIoBAwaIY8eONfxs+PDhIjo6utH6K1euFF26dBFKpVL06NFDbNmypZMTd4zWbIeXX365YV13d3cxadIkERcXJ0Hq9nXjNNubbzd+9+joaDF8+PAmj+ndu7dQKpUiKChIxMTEdHru9tba7fDhhx+K4OBgYWVlJZycnMSIESPE3r17pQnfTpr7/QE0+vuaw+dDW7aDKX4+PPPMM8Lf318olUrh6uoqRo8e3fDlLIR5vBeEaP12MJT3gkwIITpv/w0RERFRyxntGBUiIiIyfSwqREREZLBYVIiIiMhgsagQERGRwWJRISIiIoPFokJEREQGi0WFiIiIDBaLChFJJjU1FTKZDPHx8bdcZ//+/ZDJZCgpKWnX15bJZFi/fn27PieRVA4ePIh7770XXl5enfLebsnrrV27FuPGjYOzs/Md/z+/HRYVIrqtGTNmQCaTQSaTwdLSEoGBgfjnP/+Jmpqau35uX19f5OTkoGfPnu2QlMh8VVZWIiIiAl9++aXBvF5lZSWGDh2KDz/88K5ey2ivnkxEnWfChAmIiYlBXV0dYmNjER0dDZlMdtcfQBYWFiZxRVoiqU2cOBETJ0685c+1Wi3+9a9/4ddff0VJSQl69uyJDz/8ECNGjOiQ1wOAp556CsD1Pad3g3tUiOiOVCoVPDw84Ovri2nTpmHMmDHYtWsXgOtXX50/fz4CAwNhbW2NiIgIrF69uuGxxcXFeOKJJ+Dq6gpra2uEhoYiJiYGQPOHfrZu3YouXbrA2toaI0eObPIh99Zbb6F3796Nli1atAgBAQEN90+ePImxY8fCxcUFGo0Gw4cPR1xc3C1/v9raWsyePRuenp6wsrKCv78/5s+f37aNRWSAZs+ejaNHj2LFihU4c+YMHn74YUyYMAFJSUlSR7sj7lEholY5e/Ysjhw5An9/fwDA/Pnz8fPPP2PJkiUIDQ3FwYMH8eSTT8LV1RXDhw/Hf/7zH5w/fx7btm2Di4sLkpOTUV1d3exzZ2Rk4IEHHsCsWbPw3HPP4dSpU/j73//e6ozl5eWIjo7G559/DiEEFixYgEmTJiEpKQn29vZN1v/ss8+wceNGrFy5En5+fsjIyEBGRkarX5fIEKWnpyMmJgbp6enw8vICALz66qvYvn07YmJi8P7770uc8PZYVIjojjZv3gw7OzvU19dDq9VCLpfjiy++gFarxfvvv4/du3cjKioKABAUFIRDhw7h66+/xvDhw5Geno7IyEj069cPABrt+bjZ4sWLERwcjAULFgAAwsLCkJiY2OpDTKNGjWp0/5tvvoGDgwMOHDiAKVOmNFk/PT0doaGhGDp0KGQyWUMJIzIFiYmJ0Ol06NKlS6PlWq0Wzs7OAICLFy+iW7dut32e1157DR988EGH5bwVFhUiuqORI0di8eLFqKysxMKFC6FQKPDggw/i3LlzqKqqwtixYxutX1tbi8jISADAiy++iAcffBBxcXEYN24cpk2bhsGDBzf7OhcuXMDAgQMbLbtRgFojLy8P//73v7F//37k5+dDp9OhqqoK6enpza4/Y8YMjB07FmFhYZgwYQKmTJmCcePGtfp1iQxRRUUFLCwsEBsbCwsLi0Y/s7OzA3D9HxgXLly47fPcKDWdjUWFiO7I1tYWISEhAIAffvgBERER+P777xvO1tmyZQu8vb0bPUalUgG4PuguLS0NW7duxa5duzB69GjMmjULH3/8cZuyyOVyCCEaLaurq2t0Pzo6GoWFhfj000/h7+8PlUqFqKgo1NbWNvucffr0QUpKCrZt24bdu3fjkUcewZgxYxqNtSEyVpGRkdDpdMjPz8ewYcOaXUepVKJr166dnKxlWFSIqFXkcjlef/11zJkzB5cvX4ZKpUJ6ejqGDx9+y8e4uroiOjoa0dHRGDZsGP7xj380W1S6deuGjRs3Nlp27NixJs+Vm5sLIQRkMhkANJmf4fDhw/jqq68wadIkANfHvly7du22v5darcajjz6KRx99FA899BAmTJiAoqIiODk53fZxRIagoqICycnJDfdTUlIQHx8PJycndOnSBU888QSmT5+OBQsWIDIyEgUFBdizZw969eqFyZMnt+vr+fn5AQCKioqQnp6O7OxsAMClS5cAAB4eHq07208QEd1GdHS0mDp1aqNldXV1wtvbW3z00UfiX//6l3B2dhZLly4VycnJIjY2Vnz22Wdi6dKlQggh/vOf/4j169eLpKQkcfbsWTFlyhQxYMAAIYQQKSkpAoA4ffq0EEKItLQ0oVQqxauvviouXrwoli9fLjw8PAQAUVxcLIQQ4vz580Imk4kPPvhAJCcniy+++EI4OjoKf3//hnyRkZFi7Nix4vz58+LYsWNi2LBhwtraWixcuLBhHQBi3bp1QgghFixYIH755Rdx4cIFcenSJfHss88KDw8PodPpOmKTErW7ffv2CQBNbtHR0UIIIWpra8Ubb7whAgIChKWlpfD09BT333+/OHPmTIe8nhBCxMTENLvOm2++2arXYlEhottqrqgIIcT8+fOFq6urqKioEIsWLRJhYWHC0tJSuLq6ivHjx4sDBw4IIYR49913Rbdu3YS1tbVwcnISU6dOFVevXhVCNC0qQgixadMmERISIlQqlRg2bJj44YcfGhUVIYRYvHix8PX1Fba2tmL69Onivffea1RU4uLiRL9+/YSVlZUIDQ0Vq1atEv7+/rcsKt98843o3bu3sLW1FWq1WowePVrExcW11yYkorsgE+Kmg71EREREBoITvhEREZHBYlEhIiIig8WiQkRERAaLRYWIiIgMFosKERERGSwWFSIiIjJYLCpERERksFhUiIiIyGCxqBAREZHBYlEhIiIig8WiQkRERAaLRYWIiIgM1v8HiYNh70JuOUEAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot KDE of residuals\n",
    "residentuals=test_y - y_pred\n",
    "sns.kdeplot(residentuals, fill=True)\n",
    "plt.xlabel('Residuals')\n",
    "plt.ylabel('Density')\n",
    "plt.title('Distribution of Residuals')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "82992d82-ad69-47b6-a3c8-84a290b1a1fe",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean Absolute Percentage Error (MAPE): 0.00%\n"
     ]
    }
   ],
   "source": [
    "# print MAPE\n",
    "print(f\"Mean Absolute Percentage Error (MAPE): {mape:.2f}%\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5e32df44-d0a1-43f1-8e28-c525df4048e2",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8a77c82b-e6cc-4595-bc6e-43295730432d",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0699cc79-20b7-467a-8218-6bec2f87c030",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.12.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}