JM2_Regression_Project_v0.6.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Phase 0: Imports"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Loading required python packages."
   ]
  },
  {
   "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",
    "import seaborn as sns\n",
    "from statsmodels.graphics.correlation import plot_corr\n",
    "%matplotlib inline\n",
    "\n",
    "#models\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.linear_model import Ridge\n",
    "from sklearn.linear_model import Lasso\n",
    "from sklearn.neighbors import KNeighborsRegressor\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.tree import DecisionTreeRegressor\n",
    "from sklearn.svm import SVR\n",
    "from sklearn.ensemble import GradientBoostingRegressor \n",
    "\n",
    "#metrics, procssing and model selection\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn import metrics"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Phase 1: Data Collection"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As kaggle already provided us with data in the form of csv files, all we have to do is load them into dataframes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "#loading csv's\n",
    "train=pd.read_csv('df - train_set.csv')\n",
    "test=pd.read_csv('df - test_set.csv')\n",
    "test_id=test['Index']\n",
    "test=test.drop('Index',axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Phase 2: Data Cleaning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Kaggle tries to provide us with clean data but it's always a good idea to check for ourselves."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 64376 entries, 0 to 64375\n",
      "Data columns (total 13 columns):\n",
      " #   Column            Non-Null Count  Dtype  \n",
      "---  ------            --------------  -----  \n",
      " 0   Province          64376 non-null  object \n",
      " 1   Container         64376 non-null  object \n",
      " 2   Size_Grade        64376 non-null  object \n",
      " 3   Weight_Kg         64376 non-null  float64\n",
      " 4   Commodities       64376 non-null  object \n",
      " 5   Date              64376 non-null  object \n",
      " 6   Low_Price         64376 non-null  float64\n",
      " 7   High_Price        64376 non-null  float64\n",
      " 8   Sales_Total       64376 non-null  float64\n",
      " 9   Total_Qty_Sold    64376 non-null  int64  \n",
      " 10  Total_Kg_Sold     64376 non-null  float64\n",
      " 11  Stock_On_Hand     64376 non-null  int64  \n",
      " 12  avg_price_per_kg  64376 non-null  float64\n",
      "dtypes: float64(6), int64(2), object(5)\n",
      "memory usage: 6.4+ MB\n"
     ]
    }
   ],
   "source": [
    "#Checking for null values\n",
    "train.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Province            0\n",
       "Container           0\n",
       "Size_Grade          0\n",
       "Weight_Kg           0\n",
       "Commodities         0\n",
       "Date                0\n",
       "Low_Price           0\n",
       "High_Price          0\n",
       "Sales_Total         0\n",
       "Total_Qty_Sold      0\n",
       "Total_Kg_Sold       0\n",
       "Stock_On_Hand       0\n",
       "avg_price_per_kg    0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "None of the columns contain any null values.\n",
    "\n",
    "Moreover, we note that the DataFrame consists of 13 columns.(12 predictors and 1 response)\n",
    "\n",
    "The predictor variables are broken down into the following types:\n",
    "\n",
    "    -object: Province, Container, Size_Grade, Commodities and Date\n",
    "\n",
    "    -float: Weight_Kg, Low_Price, High_Price and Sales_Total\n",
    "\n",
    "    -integer: Total_Qty_Sold, Total_Kg_Sold and Stock_On_Hand\n",
    "    \n",
    "with the response variable(avg_price_per_kg) being a float.\n",
    "    \n",
    "As we can see in the next cell, the table contains information about 'Commodoties that we are not interested in, we filter only for 'APPLE GOLDEN DELICIOUS'."
   ]
  },
  {
   "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>Province</th>\n",
       "      <th>Container</th>\n",
       "      <th>Size_Grade</th>\n",
       "      <th>Weight_Kg</th>\n",
       "      <th>Commodities</th>\n",
       "      <th>Date</th>\n",
       "      <th>Low_Price</th>\n",
       "      <th>High_Price</th>\n",
       "      <th>Sales_Total</th>\n",
       "      <th>Total_Qty_Sold</th>\n",
       "      <th>Total_Kg_Sold</th>\n",
       "      <th>Stock_On_Hand</th>\n",
       "      <th>avg_price_per_kg</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>CAPE</td>\n",
       "      <td>EC120</td>\n",
       "      <td>1L</td>\n",
       "      <td>12.0</td>\n",
       "      <td>APPLE GRANNY SMITH</td>\n",
       "      <td>2020-03-10</td>\n",
       "      <td>108.0</td>\n",
       "      <td>112.0</td>\n",
       "      <td>3236.0</td>\n",
       "      <td>29</td>\n",
       "      <td>348.0</td>\n",
       "      <td>0</td>\n",
       "      <td>9.30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CAPE</td>\n",
       "      <td>M4183</td>\n",
       "      <td>1L</td>\n",
       "      <td>18.3</td>\n",
       "      <td>APPLE GOLDEN DELICIOUS</td>\n",
       "      <td>2020-09-09</td>\n",
       "      <td>150.0</td>\n",
       "      <td>170.0</td>\n",
       "      <td>51710.0</td>\n",
       "      <td>332</td>\n",
       "      <td>6075.6</td>\n",
       "      <td>822</td>\n",
       "      <td>8.51</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>GAUTENG</td>\n",
       "      <td>AT200</td>\n",
       "      <td>1L</td>\n",
       "      <td>20.0</td>\n",
       "      <td>AVOCADO PINKERTON</td>\n",
       "      <td>2020-05-05</td>\n",
       "      <td>70.0</td>\n",
       "      <td>80.0</td>\n",
       "      <td>4860.0</td>\n",
       "      <td>66</td>\n",
       "      <td>1320.0</td>\n",
       "      <td>50</td>\n",
       "      <td>3.68</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>TRANSVAAL</td>\n",
       "      <td>BJ090</td>\n",
       "      <td>1L</td>\n",
       "      <td>9.0</td>\n",
       "      <td>TOMATOES-LONG LIFE</td>\n",
       "      <td>2020-01-20</td>\n",
       "      <td>60.0</td>\n",
       "      <td>60.0</td>\n",
       "      <td>600.0</td>\n",
       "      <td>10</td>\n",
       "      <td>90.0</td>\n",
       "      <td>0</td>\n",
       "      <td>6.67</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>WESTERN FREESTATE</td>\n",
       "      <td>PP100</td>\n",
       "      <td>1R</td>\n",
       "      <td>10.0</td>\n",
       "      <td>POTATO SIFRA (WASHED)</td>\n",
       "      <td>2020-07-14</td>\n",
       "      <td>40.0</td>\n",
       "      <td>45.0</td>\n",
       "      <td>41530.0</td>\n",
       "      <td>927</td>\n",
       "      <td>9270.0</td>\n",
       "      <td>393</td>\n",
       "      <td>4.48</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Province Container Size_Grade  Weight_Kg             Commodities  \\\n",
       "0               CAPE     EC120         1L       12.0      APPLE GRANNY SMITH   \n",
       "1               CAPE     M4183         1L       18.3  APPLE GOLDEN DELICIOUS   \n",
       "2            GAUTENG     AT200         1L       20.0       AVOCADO PINKERTON   \n",
       "3          TRANSVAAL     BJ090         1L        9.0      TOMATOES-LONG LIFE   \n",
       "4  WESTERN FREESTATE     PP100         1R       10.0   POTATO SIFRA (WASHED)   \n",
       "\n",
       "         Date  Low_Price  High_Price  Sales_Total  Total_Qty_Sold  \\\n",
       "0  2020-03-10      108.0       112.0       3236.0              29   \n",
       "1  2020-09-09      150.0       170.0      51710.0             332   \n",
       "2  2020-05-05       70.0        80.0       4860.0              66   \n",
       "3  2020-01-20       60.0        60.0        600.0              10   \n",
       "4  2020-07-14       40.0        45.0      41530.0             927   \n",
       "\n",
       "   Total_Kg_Sold  Stock_On_Hand  avg_price_per_kg  \n",
       "0          348.0              0              9.30  \n",
       "1         6075.6            822              8.51  \n",
       "2         1320.0             50              3.68  \n",
       "3           90.0              0              6.67  \n",
       "4         9270.0            393              4.48  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>Province</th>\n",
       "      <th>Container</th>\n",
       "      <th>Size_Grade</th>\n",
       "      <th>Weight_Kg</th>\n",
       "      <th>Commodities</th>\n",
       "      <th>Date</th>\n",
       "      <th>Low_Price</th>\n",
       "      <th>High_Price</th>\n",
       "      <th>Sales_Total</th>\n",
       "      <th>Total_Qty_Sold</th>\n",
       "      <th>Total_Kg_Sold</th>\n",
       "      <th>Stock_On_Hand</th>\n",
       "      <th>avg_price_per_kg</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CAPE</td>\n",
       "      <td>M4183</td>\n",
       "      <td>1L</td>\n",
       "      <td>18.3</td>\n",
       "      <td>APPLE GOLDEN DELICIOUS</td>\n",
       "      <td>2020-09-09</td>\n",
       "      <td>150.0</td>\n",
       "      <td>170.0</td>\n",
       "      <td>51710.0</td>\n",
       "      <td>332</td>\n",
       "      <td>6075.6</td>\n",
       "      <td>822</td>\n",
       "      <td>8.51</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>CAPE</td>\n",
       "      <td>JG110</td>\n",
       "      <td>2M</td>\n",
       "      <td>11.0</td>\n",
       "      <td>APPLE GOLDEN DELICIOUS</td>\n",
       "      <td>2020-04-14</td>\n",
       "      <td>50.0</td>\n",
       "      <td>50.0</td>\n",
       "      <td>16000.0</td>\n",
       "      <td>320</td>\n",
       "      <td>3520.0</td>\n",
       "      <td>0</td>\n",
       "      <td>4.55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>W.CAPE-BERGRIVER ETC</td>\n",
       "      <td>JE090</td>\n",
       "      <td>2S</td>\n",
       "      <td>9.0</td>\n",
       "      <td>APPLE GOLDEN DELICIOUS</td>\n",
       "      <td>2020-04-16</td>\n",
       "      <td>55.0</td>\n",
       "      <td>55.0</td>\n",
       "      <td>990.0</td>\n",
       "      <td>18</td>\n",
       "      <td>162.0</td>\n",
       "      <td>1506</td>\n",
       "      <td>6.11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>CAPE</td>\n",
       "      <td>M4183</td>\n",
       "      <td>1S</td>\n",
       "      <td>18.3</td>\n",
       "      <td>APPLE GOLDEN DELICIOUS</td>\n",
       "      <td>2020-05-04</td>\n",
       "      <td>80.0</td>\n",
       "      <td>120.0</td>\n",
       "      <td>32020.0</td>\n",
       "      <td>388</td>\n",
       "      <td>7100.4</td>\n",
       "      <td>443</td>\n",
       "      <td>4.51</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>69</th>\n",
       "      <td>EASTERN CAPE</td>\n",
       "      <td>IA400</td>\n",
       "      <td>1S</td>\n",
       "      <td>400.0</td>\n",
       "      <td>APPLE GOLDEN DELICIOUS</td>\n",
       "      <td>2020-09-28</td>\n",
       "      <td>1800.0</td>\n",
       "      <td>1800.0</td>\n",
       "      <td>1800.0</td>\n",
       "      <td>1</td>\n",
       "      <td>400.0</td>\n",
       "      <td>2</td>\n",
       "      <td>4.50</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Province Container Size_Grade  Weight_Kg  \\\n",
       "1                   CAPE     M4183         1L       18.3   \n",
       "7                   CAPE     JG110         2M       11.0   \n",
       "24  W.CAPE-BERGRIVER ETC     JE090         2S        9.0   \n",
       "40                  CAPE     M4183         1S       18.3   \n",
       "69          EASTERN CAPE     IA400         1S      400.0   \n",
       "\n",
       "               Commodities        Date  Low_Price  High_Price  Sales_Total  \\\n",
       "1   APPLE GOLDEN DELICIOUS  2020-09-09      150.0       170.0      51710.0   \n",
       "7   APPLE GOLDEN DELICIOUS  2020-04-14       50.0        50.0      16000.0   \n",
       "24  APPLE GOLDEN DELICIOUS  2020-04-16       55.0        55.0        990.0   \n",
       "40  APPLE GOLDEN DELICIOUS  2020-05-04       80.0       120.0      32020.0   \n",
       "69  APPLE GOLDEN DELICIOUS  2020-09-28     1800.0      1800.0       1800.0   \n",
       "\n",
       "    Total_Qty_Sold  Total_Kg_Sold  Stock_On_Hand  avg_price_per_kg  \n",
       "1              332         6075.6            822              8.51  \n",
       "7              320         3520.0              0              4.55  \n",
       "24              18          162.0           1506              6.11  \n",
       "40             388         7100.4            443              4.51  \n",
       "69               1          400.0              2              4.50  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Filter for Golden Delicious Apples\n",
    "train=train[train['Commodities']=='APPLE GOLDEN DELICIOUS']\n",
    "test=test[test['Commodities']=='APPLE GOLDEN DELICIOUS']\n",
    "train.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We may as well remove the \"Commodities\" column as it has no more useful information."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 1952 entries, 1 to 64310\n",
      "Data columns (total 12 columns):\n",
      " #   Column            Non-Null Count  Dtype  \n",
      "---  ------            --------------  -----  \n",
      " 0   Province          1952 non-null   object \n",
      " 1   Container         1952 non-null   object \n",
      " 2   Size_Grade        1952 non-null   object \n",
      " 3   Weight_Kg         1952 non-null   float64\n",
      " 4   Date              1952 non-null   object \n",
      " 5   Low_Price         1952 non-null   float64\n",
      " 6   High_Price        1952 non-null   float64\n",
      " 7   Sales_Total       1952 non-null   float64\n",
      " 8   Total_Qty_Sold    1952 non-null   int64  \n",
      " 9   Total_Kg_Sold     1952 non-null   float64\n",
      " 10  Stock_On_Hand     1952 non-null   int64  \n",
      " 11  avg_price_per_kg  1952 non-null   float64\n",
      "dtypes: float64(6), int64(2), object(4)\n",
      "memory usage: 198.2+ KB\n"
     ]
    }
   ],
   "source": [
    "#Remove Commodities\n",
    "train=train.drop([\"Commodities\"],axis=1)\n",
    "test=test.drop([\"Commodities\"],axis=1)\n",
    "train.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Phase 3: Exploratory Data Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3.1: Univariate Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 3.1.1: Numeric data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>Weight_Kg</th>\n",
       "      <th>Low_Price</th>\n",
       "      <th>High_Price</th>\n",
       "      <th>Sales_Total</th>\n",
       "      <th>Total_Qty_Sold</th>\n",
       "      <th>Total_Kg_Sold</th>\n",
       "      <th>Stock_On_Hand</th>\n",
       "      <th>avg_price_per_kg</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1952.000000</td>\n",
       "      <td>1952.000000</td>\n",
       "      <td>1952.000000</td>\n",
       "      <td>1952.000000</td>\n",
       "      <td>1952.000000</td>\n",
       "      <td>1952.000000</td>\n",
       "      <td>1952.000000</td>\n",
       "      <td>1952.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>40.460912</td>\n",
       "      <td>174.307377</td>\n",
       "      <td>215.648053</td>\n",
       "      <td>20053.533811</td>\n",
       "      <td>174.510758</td>\n",
       "      <td>2960.176332</td>\n",
       "      <td>408.393955</td>\n",
       "      <td>6.778893</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>99.655169</td>\n",
       "      <td>373.553578</td>\n",
       "      <td>433.546159</td>\n",
       "      <td>39005.069445</td>\n",
       "      <td>308.810797</td>\n",
       "      <td>6097.416527</td>\n",
       "      <td>724.450582</td>\n",
       "      <td>2.248744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>3.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.250000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>9.000000</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>1325.000000</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>219.600000</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>5.460000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>12.000000</td>\n",
       "      <td>80.000000</td>\n",
       "      <td>108.000000</td>\n",
       "      <td>5495.000000</td>\n",
       "      <td>64.000000</td>\n",
       "      <td>853.500000</td>\n",
       "      <td>126.500000</td>\n",
       "      <td>6.670000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>18.300000</td>\n",
       "      <td>127.250000</td>\n",
       "      <td>160.000000</td>\n",
       "      <td>21082.500000</td>\n",
       "      <td>200.000000</td>\n",
       "      <td>3093.525000</td>\n",
       "      <td>468.000000</td>\n",
       "      <td>8.280000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>400.000000</td>\n",
       "      <td>2300.000000</td>\n",
       "      <td>3300.000000</td>\n",
       "      <td>369464.000000</td>\n",
       "      <td>4237.000000</td>\n",
       "      <td>74000.000000</td>\n",
       "      <td>6400.000000</td>\n",
       "      <td>21.240000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Weight_Kg    Low_Price   High_Price    Sales_Total  Total_Qty_Sold  \\\n",
       "count  1952.000000  1952.000000  1952.000000    1952.000000     1952.000000   \n",
       "mean     40.460912   174.307377   215.648053   20053.533811      174.510758   \n",
       "std      99.655169   373.553578   433.546159   39005.069445      308.810797   \n",
       "min       3.000000     2.000000     5.000000       5.000000        1.000000   \n",
       "25%       9.000000    50.000000    60.000000    1325.000000       12.000000   \n",
       "50%      12.000000    80.000000   108.000000    5495.000000       64.000000   \n",
       "75%      18.300000   127.250000   160.000000   21082.500000      200.000000   \n",
       "max     400.000000  2300.000000  3300.000000  369464.000000     4237.000000   \n",
       "\n",
       "       Total_Kg_Sold  Stock_On_Hand  avg_price_per_kg  \n",
       "count    1952.000000    1952.000000       1952.000000  \n",
       "mean     2960.176332     408.393955          6.778893  \n",
       "std      6097.416527     724.450582          2.248744  \n",
       "min         3.000000       0.000000          0.250000  \n",
       "25%       219.600000       9.000000          5.460000  \n",
       "50%       853.500000     126.500000          6.670000  \n",
       "75%      3093.525000     468.000000          8.280000  \n",
       "max     74000.000000    6400.000000         21.240000  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Weight_Kg           3.328254\n",
       "Low_Price           4.222672\n",
       "High_Price          3.828819\n",
       "Sales_Total         4.250317\n",
       "Total_Qty_Sold      4.445169\n",
       "Total_Kg_Sold       5.529586\n",
       "Stock_On_Hand       3.391815\n",
       "avg_price_per_kg    0.082452\n",
       "dtype: float64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.skew()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that all of our numeric predictors are highly skewed. While our response variable is fairly symmetrical."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Weight_Kg            9.109698\n",
       "Low_Price           17.073369\n",
       "High_Price          13.630768\n",
       "Sales_Total         24.200524\n",
       "Total_Qty_Sold      32.338123\n",
       "Total_Kg_Sold       45.153801\n",
       "Stock_On_Hand       14.671224\n",
       "avg_price_per_kg     1.042921\n",
       "dtype: float64"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.kurtosis()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As expected the kurtosis indicates strong outliers in the predictor variables. This is highlighted by the fact that the mean of our predictor values lean much closer toward their minimum values while the maximum values are many standard deviations away."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 8 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "features=[]\n",
    "for x in train.columns:\n",
    "    if np.issubdtype(train[x].dtype, np.number)==True:\n",
    "        features.append(x)\n",
    "train[features].plot(kind='density', subplots=True, sharex=False, figsize=(10, 10));"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "These variables appear to be normally distributed with either some skewness or some completely seperate bumps. This is caused by the outliers.\n",
    "\n",
    "We remove the outliers by converting values to their z-score and removing those z-scores with an absolute value greater than 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",
       "      <th>Province</th>\n",
       "      <th>Container</th>\n",
       "      <th>Size_Grade</th>\n",
       "      <th>Weight_Kg</th>\n",
       "      <th>Date</th>\n",
       "      <th>Low_Price</th>\n",
       "      <th>High_Price</th>\n",
       "      <th>Sales_Total</th>\n",
       "      <th>Total_Qty_Sold</th>\n",
       "      <th>Total_Kg_Sold</th>\n",
       "      <th>Stock_On_Hand</th>\n",
       "      <th>avg_price_per_kg</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CAPE</td>\n",
       "      <td>M4183</td>\n",
       "      <td>1L</td>\n",
       "      <td>18.3</td>\n",
       "      <td>2020-09-09</td>\n",
       "      <td>150.0</td>\n",
       "      <td>170.0</td>\n",
       "      <td>51710.0</td>\n",
       "      <td>332</td>\n",
       "      <td>6075.6</td>\n",
       "      <td>822</td>\n",
       "      <td>8.51</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>CAPE</td>\n",
       "      <td>JG110</td>\n",
       "      <td>2M</td>\n",
       "      <td>11.0</td>\n",
       "      <td>2020-04-14</td>\n",
       "      <td>50.0</td>\n",
       "      <td>50.0</td>\n",
       "      <td>16000.0</td>\n",
       "      <td>320</td>\n",
       "      <td>3520.0</td>\n",
       "      <td>0</td>\n",
       "      <td>4.55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>W.CAPE-BERGRIVER ETC</td>\n",
       "      <td>JE090</td>\n",
       "      <td>2S</td>\n",
       "      <td>9.0</td>\n",
       "      <td>2020-04-16</td>\n",
       "      <td>55.0</td>\n",
       "      <td>55.0</td>\n",
       "      <td>990.0</td>\n",
       "      <td>18</td>\n",
       "      <td>162.0</td>\n",
       "      <td>1506</td>\n",
       "      <td>6.11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>CAPE</td>\n",
       "      <td>M4183</td>\n",
       "      <td>1S</td>\n",
       "      <td>18.3</td>\n",
       "      <td>2020-05-04</td>\n",
       "      <td>80.0</td>\n",
       "      <td>120.0</td>\n",
       "      <td>32020.0</td>\n",
       "      <td>388</td>\n",
       "      <td>7100.4</td>\n",
       "      <td>443</td>\n",
       "      <td>4.51</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>80</th>\n",
       "      <td>CAPE</td>\n",
       "      <td>JE090</td>\n",
       "      <td>2S</td>\n",
       "      <td>9.0</td>\n",
       "      <td>2020-03-13</td>\n",
       "      <td>55.0</td>\n",
       "      <td>55.0</td>\n",
       "      <td>4840.0</td>\n",
       "      <td>88</td>\n",
       "      <td>792.0</td>\n",
       "      <td>181</td>\n",
       "      <td>6.11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Province Container Size_Grade  Weight_Kg        Date  \\\n",
       "1                   CAPE     M4183         1L       18.3  2020-09-09   \n",
       "7                   CAPE     JG110         2M       11.0  2020-04-14   \n",
       "24  W.CAPE-BERGRIVER ETC     JE090         2S        9.0  2020-04-16   \n",
       "40                  CAPE     M4183         1S       18.3  2020-05-04   \n",
       "80                  CAPE     JE090         2S        9.0  2020-03-13   \n",
       "\n",
       "    Low_Price  High_Price  Sales_Total  Total_Qty_Sold  Total_Kg_Sold  \\\n",
       "1       150.0       170.0      51710.0             332         6075.6   \n",
       "7        50.0        50.0      16000.0             320         3520.0   \n",
       "24       55.0        55.0        990.0              18          162.0   \n",
       "40       80.0       120.0      32020.0             388         7100.4   \n",
       "80       55.0        55.0       4840.0              88          792.0   \n",
       "\n",
       "    Stock_On_Hand  avg_price_per_kg  \n",
       "1             822              8.51  \n",
       "7               0              4.55  \n",
       "24           1506              6.11  \n",
       "40            443              4.51  \n",
       "80            181              6.11  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Remove outliers using z-score\n",
    "for i in features:\n",
    "    z_mean=train[i].mean()\n",
    "    z_std=train[i].std()\n",
    "    train['zscore']=abs((train[i]-z_mean)/z_std)\n",
    "    train=train[train['zscore']<3]\n",
    "    train=train.drop('zscore',axis=1)\n",
    "train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>Weight_Kg</th>\n",
       "      <th>Low_Price</th>\n",
       "      <th>High_Price</th>\n",
       "      <th>Sales_Total</th>\n",
       "      <th>Total_Qty_Sold</th>\n",
       "      <th>Total_Kg_Sold</th>\n",
       "      <th>Stock_On_Hand</th>\n",
       "      <th>avg_price_per_kg</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1644.000000</td>\n",
       "      <td>1644.000000</td>\n",
       "      <td>1644.000000</td>\n",
       "      <td>1644.000000</td>\n",
       "      <td>1644.000000</td>\n",
       "      <td>1644.000000</td>\n",
       "      <td>1644.000000</td>\n",
       "      <td>1644.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>12.742640</td>\n",
       "      <td>84.369221</td>\n",
       "      <td>102.372263</td>\n",
       "      <td>10779.321776</td>\n",
       "      <td>118.315693</td>\n",
       "      <td>1494.513990</td>\n",
       "      <td>319.843674</td>\n",
       "      <td>6.918425</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>4.048532</td>\n",
       "      <td>45.477576</td>\n",
       "      <td>52.496762</td>\n",
       "      <td>14792.191679</td>\n",
       "      <td>148.502259</td>\n",
       "      <td>1906.601324</td>\n",
       "      <td>452.675560</td>\n",
       "      <td>2.110512</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>3.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.550000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>9.000000</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>1053.750000</td>\n",
       "      <td>13.000000</td>\n",
       "      <td>164.700000</td>\n",
       "      <td>11.750000</td>\n",
       "      <td>5.560000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>12.000000</td>\n",
       "      <td>70.000000</td>\n",
       "      <td>96.000000</td>\n",
       "      <td>4480.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>732.000000</td>\n",
       "      <td>134.000000</td>\n",
       "      <td>6.745000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>18.300000</td>\n",
       "      <td>118.500000</td>\n",
       "      <td>140.000000</td>\n",
       "      <td>14294.000000</td>\n",
       "      <td>169.250000</td>\n",
       "      <td>1984.050000</td>\n",
       "      <td>430.000000</td>\n",
       "      <td>8.392500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>18.300000</td>\n",
       "      <td>220.000000</td>\n",
       "      <td>260.000000</td>\n",
       "      <td>96260.000000</td>\n",
       "      <td>778.000000</td>\n",
       "      <td>8655.900000</td>\n",
       "      <td>2361.000000</td>\n",
       "      <td>13.070000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Weight_Kg    Low_Price   High_Price   Sales_Total  Total_Qty_Sold  \\\n",
       "count  1644.000000  1644.000000  1644.000000   1644.000000     1644.000000   \n",
       "mean     12.742640    84.369221   102.372263  10779.321776      118.315693   \n",
       "std       4.048532    45.477576    52.496762  14792.191679      148.502259   \n",
       "min       3.000000     2.000000     5.000000      5.000000        1.000000   \n",
       "25%       9.000000    50.000000    60.000000   1053.750000       13.000000   \n",
       "50%      12.000000    70.000000    96.000000   4480.000000       60.000000   \n",
       "75%      18.300000   118.500000   140.000000  14294.000000      169.250000   \n",
       "max      18.300000   220.000000   260.000000  96260.000000      778.000000   \n",
       "\n",
       "       Total_Kg_Sold  Stock_On_Hand  avg_price_per_kg  \n",
       "count    1644.000000    1644.000000       1644.000000  \n",
       "mean     1494.513990     319.843674          6.918425  \n",
       "std      1906.601324     452.675560          2.110512  \n",
       "min         3.000000       0.000000          0.550000  \n",
       "25%       164.700000      11.750000          5.560000  \n",
       "50%       732.000000     134.000000          6.745000  \n",
       "75%      1984.050000     430.000000          8.392500  \n",
       "max      8655.900000    2361.000000         13.070000  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Weight_Kg           0.283771\n",
       "Low_Price           0.697714\n",
       "High_Price          0.446422\n",
       "Sales_Total         2.067655\n",
       "Total_Qty_Sold      1.841766\n",
       "Total_Kg_Sold       1.729352\n",
       "Stock_On_Hand       2.179969\n",
       "avg_price_per_kg   -0.118437\n",
       "dtype: float64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.skew()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Weight_Kg          -1.050116\n",
       "Low_Price          -0.246510\n",
       "High_Price         -0.805569\n",
       "Sales_Total         4.419355\n",
       "Total_Qty_Sold      3.279708\n",
       "Total_Kg_Sold       2.380347\n",
       "Stock_On_Hand       4.958846\n",
       "avg_price_per_kg   -0.109211\n",
       "dtype: float64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.kurtosis()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Most of the kurtosis values are less than 3 which indicates we have removed the outliers. \n",
    "\n",
    "Some of our predictor values still display high skewness but this is to be expected.\n",
    "\n",
    "For example, when looking at a column like Sales_Total, many places can have sales up to a certain point but after a certain amount only the biggest place can make that much in sales. This would skew our data toward the more 'acheivable' Sales_total values."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Convert date to datetime to avoid mismanagement when handling categorical data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Date (object) --> Date (datetime)\n",
    "train['Date'] = pd.to_datetime(train['Date'])\n",
    "test['Date'] = pd.to_datetime(test['Date'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We also require our numeric data to be standardized (for lasso and ridge regression) before the addition of dummy variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Weight_Kg',\n",
       " 'Low_Price',\n",
       " 'High_Price',\n",
       " 'Sales_Total',\n",
       " 'Total_Qty_Sold',\n",
       " 'Total_Kg_Sold',\n",
       " 'Stock_On_Hand']"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#we dont want to scale the response variable\n",
    "features.remove('avg_price_per_kg')\n",
    "features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 1644 entries, 1 to 64310\n",
      "Data columns (total 12 columns):\n",
      " #   Column            Non-Null Count  Dtype         \n",
      "---  ------            --------------  -----         \n",
      " 0   Province          1644 non-null   object        \n",
      " 1   Container         1644 non-null   object        \n",
      " 2   Size_Grade        1644 non-null   object        \n",
      " 3   Weight_Kg         1644 non-null   float64       \n",
      " 4   Date              1644 non-null   datetime64[ns]\n",
      " 5   Low_Price         1644 non-null   float64       \n",
      " 6   High_Price        1644 non-null   float64       \n",
      " 7   Sales_Total       1644 non-null   float64       \n",
      " 8   Total_Qty_Sold    1644 non-null   float64       \n",
      " 9   Total_Kg_Sold     1644 non-null   float64       \n",
      " 10  Stock_On_Hand     1644 non-null   float64       \n",
      " 11  avg_price_per_kg  1644 non-null   float64       \n",
      "dtypes: datetime64[ns](1), float64(8), object(3)\n",
      "memory usage: 167.0+ KB\n"
     ]
    }
   ],
   "source": [
    "#scaling and adding scaled data back to train dataframe\n",
    "scalar=StandardScaler()\n",
    "train[features] = scalar.fit_transform(train[features])\n",
    "train.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 3.1.3: Categorical Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x648 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Plot countplots of categorical data\n",
    "fig, ax =plt.subplots(1,3, figsize=(20, 9))\n",
    "sns.countplot(x=train['Size_Grade'], ax=ax[0])\n",
    "sns.countplot(x=train['Container'], ax=ax[1])\n",
    "sns.countplot(x=\"Province\", data=train, ax=ax[2])\n",
    "ax[2].tick_params(labelrotation=45)\n",
    "fig.tight_layout(pad=4.0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Most of the Golden Delicious Apples are found in the cape, have a size grade of 1S and are in containers 'M4183', 'JE090 and 'EC120'."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Before continuing the analysis we should note that numerical data is required for regression analysis. As such we will now transform our categorical data into numerical values using dummy variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_dummies = pd.get_dummies(train,drop_first=True)\n",
    "test_dummies = pd.get_dummies(test,drop_first=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "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>Weight_Kg</th>\n",
       "      <th>Date</th>\n",
       "      <th>Low_Price</th>\n",
       "      <th>High_Price</th>\n",
       "      <th>Sales_Total</th>\n",
       "      <th>Total_Qty_Sold</th>\n",
       "      <th>Total_Kg_Sold</th>\n",
       "      <th>Stock_On_Hand</th>\n",
       "      <th>avg_price_per_kg</th>\n",
       "      <th>Province_EASTERN CAPE</th>\n",
       "      <th>...</th>\n",
       "      <th>Container_M9125</th>\n",
       "      <th>Size_Grade_1M</th>\n",
       "      <th>Size_Grade_1S</th>\n",
       "      <th>Size_Grade_1U</th>\n",
       "      <th>Size_Grade_1X</th>\n",
       "      <th>Size_Grade_2L</th>\n",
       "      <th>Size_Grade_2M</th>\n",
       "      <th>Size_Grade_2S</th>\n",
       "      <th>Size_Grade_2U</th>\n",
       "      <th>Size_Grade_2X</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.373103</td>\n",
       "      <td>2020-09-09</td>\n",
       "      <td>1.443585</td>\n",
       "      <td>1.288619</td>\n",
       "      <td>2.767888</td>\n",
       "      <td>1.439367</td>\n",
       "      <td>2.403481</td>\n",
       "      <td>1.109645</td>\n",
       "      <td>8.51</td>\n",
       "      <td>0</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>-0.430568</td>\n",
       "      <td>2020-04-14</td>\n",
       "      <td>-0.755970</td>\n",
       "      <td>-0.997932</td>\n",
       "      <td>0.353042</td>\n",
       "      <td>1.358536</td>\n",
       "      <td>1.062677</td>\n",
       "      <td>-0.706778</td>\n",
       "      <td>4.55</td>\n",
       "      <td>0</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>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>-0.924725</td>\n",
       "      <td>2020-04-16</td>\n",
       "      <td>-0.645992</td>\n",
       "      <td>-0.902659</td>\n",
       "      <td>-0.661991</td>\n",
       "      <td>-0.675722</td>\n",
       "      <td>-0.699108</td>\n",
       "      <td>2.621121</td>\n",
       "      <td>6.11</td>\n",
       "      <td>0</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>1.373103</td>\n",
       "      <td>2020-05-04</td>\n",
       "      <td>-0.096103</td>\n",
       "      <td>0.335889</td>\n",
       "      <td>1.436375</td>\n",
       "      <td>1.816581</td>\n",
       "      <td>2.941145</td>\n",
       "      <td>0.272146</td>\n",
       "      <td>4.51</td>\n",
       "      <td>0</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>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>80</th>\n",
       "      <td>-0.924725</td>\n",
       "      <td>2020-03-13</td>\n",
       "      <td>-0.645992</td>\n",
       "      <td>-0.902659</td>\n",
       "      <td>-0.401640</td>\n",
       "      <td>-0.204205</td>\n",
       "      <td>-0.368576</td>\n",
       "      <td>-0.306811</td>\n",
       "      <td>6.11</td>\n",
       "      <td>0</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 33 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    Weight_Kg       Date  Low_Price  High_Price  Sales_Total  Total_Qty_Sold  \\\n",
       "1    1.373103 2020-09-09   1.443585    1.288619     2.767888        1.439367   \n",
       "7   -0.430568 2020-04-14  -0.755970   -0.997932     0.353042        1.358536   \n",
       "24  -0.924725 2020-04-16  -0.645992   -0.902659    -0.661991       -0.675722   \n",
       "40   1.373103 2020-05-04  -0.096103    0.335889     1.436375        1.816581   \n",
       "80  -0.924725 2020-03-13  -0.645992   -0.902659    -0.401640       -0.204205   \n",
       "\n",
       "    Total_Kg_Sold  Stock_On_Hand  avg_price_per_kg  Province_EASTERN CAPE  \\\n",
       "1        2.403481       1.109645              8.51                      0   \n",
       "7        1.062677      -0.706778              4.55                      0   \n",
       "24      -0.699108       2.621121              6.11                      0   \n",
       "40       2.941145       0.272146              4.51                      0   \n",
       "80      -0.368576      -0.306811              6.11                      0   \n",
       "\n",
       "    ...  Container_M9125  Size_Grade_1M  Size_Grade_1S  Size_Grade_1U  \\\n",
       "1   ...                0              0              0              0   \n",
       "7   ...                0              0              0              0   \n",
       "24  ...                0              0              0              0   \n",
       "40  ...                0              0              1              0   \n",
       "80  ...                0              0              0              0   \n",
       "\n",
       "    Size_Grade_1X  Size_Grade_2L  Size_Grade_2M  Size_Grade_2S  Size_Grade_2U  \\\n",
       "1               0              0              0              0              0   \n",
       "7               0              0              1              0              0   \n",
       "24              0              0              0              1              0   \n",
       "40              0              0              0              0              0   \n",
       "80              0              0              0              1              0   \n",
       "\n",
       "    Size_Grade_2X  \n",
       "1               0  \n",
       "7               0  \n",
       "24              0  \n",
       "40              0  \n",
       "80              0  \n",
       "\n",
       "[5 rows x 33 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_dummies.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3.2: Multivariate Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We start by evaluating the relationship between the date and our response variable"
   ]
  },
  {
   "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>avg_price_per_kg</th>\n",
       "      <th>Date</th>\n",
       "      <th>Day</th>\n",
       "      <th>Month</th>\n",
       "      <th>Month_year</th>\n",
       "      <th>Quarter</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>8.51</td>\n",
       "      <td>2020-09-09</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "      <td>2020-09-09</td>\n",
       "      <td>2020Q3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>4.55</td>\n",
       "      <td>2020-04-14</td>\n",
       "      <td>14</td>\n",
       "      <td>4</td>\n",
       "      <td>2020-04-14</td>\n",
       "      <td>2020Q2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>6.11</td>\n",
       "      <td>2020-04-16</td>\n",
       "      <td>16</td>\n",
       "      <td>4</td>\n",
       "      <td>2020-04-16</td>\n",
       "      <td>2020Q2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>4.51</td>\n",
       "      <td>2020-05-04</td>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>2020-05-04</td>\n",
       "      <td>2020Q2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>80</th>\n",
       "      <td>6.11</td>\n",
       "      <td>2020-03-13</td>\n",
       "      <td>13</td>\n",
       "      <td>3</td>\n",
       "      <td>2020-03-13</td>\n",
       "      <td>2020Q1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    avg_price_per_kg       Date  Day  Month Month_year Quarter\n",
       "1               8.51 2020-09-09    9      9 2020-09-09  2020Q3\n",
       "7               4.55 2020-04-14   14      4 2020-04-14  2020Q2\n",
       "24              6.11 2020-04-16   16      4 2020-04-16  2020Q2\n",
       "40              4.51 2020-05-04    4      5 2020-05-04  2020Q2\n",
       "80              6.11 2020-03-13   13      3 2020-03-13  2020Q1"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#seperating date,month and year\n",
    "training_day=pd.DataFrame([])\n",
    "training_day['avg_price_per_kg']=train['avg_price_per_kg']\n",
    "training_day['Date']=train['Date']\n",
    "training_day['Day']=[x.day for x in train['Date']]\n",
    "training_day['Month']=[x.month for x in train['Date']]\n",
    "training_day['Month_year']=[pd.to_datetime(x, format='%m%Y') for x in train['Date']]\n",
    "training_day['Quarter'] = pd.PeriodIndex(train.Date, freq='Q')\n",
    "training_day.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='Date', ylabel='avg_price_per_kg'>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Plotting the mean avg_price_per_kg against the Date\n",
    "avg_date=training_day.groupby('Date')['avg_price_per_kg'].mean()\n",
    "sns.scatterplot(x=avg_date.index,y=avg_date)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we can see from this plot, there isn't much of a relatioship between these two variables, in the following cells we attempt to see if there is a relationship between the components of the date (Day,Month,Year) and our response variable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='Day', ylabel='avg_price_per_kg'>"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "avg_day=training_day.groupby('Day')['avg_price_per_kg'].mean()\n",
    "sns.scatterplot(x=avg_day.index,y=avg_day)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-There is no clear trend amongst this data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='Month', ylabel='avg_price_per_kg'>"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYMAAAEGCAYAAACHGfl5AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAAAWbUlEQVR4nO3df5BdZZ3n8fenSZxAQgYNDVEUYxwKRixF7HVBxWVEapD11+xYDJbMODNbwzL+iIq7ju5uOTq1NTWuFitoKYOuiooow6DiiAyWOitawlQHI4LoKjFCEJIms5IQaQn0d/+4pzOdpjvpm9zbp2/n/aq61ff8uOd+T3K7P/c855znSVUhSTq4DbVdgCSpfYaBJMkwkCQZBpIkDANJErCk7QL2x5FHHllr1qxpuwxJGijr16+/v6qGZ1o2kGGwZs0aRkdH2y5DkgZKkp/PtsxmIkmSYSBJmocwSPLWJLcnuS3JlUmWTVt+epIHkmxoHu/qd02SpD319ZxBkmOAdcAzquqhJFcB5wKfnLbqjVX1sn7WIkma3Xw0Ey0BDk2yBDgM+MU8vKckqQt9DYOqugd4P3AXcC/wQFXdMMOqpyb5fpKvJjlxpm0lOT/JaJLRsbGxPlYtSQvPxESxcexBvnvn/Wwce5CJid52MtrXMEjyeOCVwNOAJwHLk5w3bbVbgKdW1bOBDwJfnGlbVXVZVY1U1cjw8IyXyUrSojQxUVx/+32cfcmNvOajN3P2JTdy/e339TQQ+t1M9BLgZ1U1VlW7gGuA509doaq2V9WDzfPrgKVJjuxzXZI0MDZt28mFV21gfNcEAOO7Jrjwqg1s2razZ+/R7zC4CzglyWFJApwB3DF1hSSrm2UkeV5T07Y+1yVJA2PL9vHdQTBpfNcEW3eM9+w9+no1UVXdnORqOk1BjwDfAy5LckGz/FLg1cCfJ3kEeAg4txxxR5J2O3rlMpYtHdojEJYtHeKow5ft5VXdySD+3R0ZGSm7o5B0sJg8ZzDZVLRs6RAXnXMSZ524mqGhzHk7SdZX1chMywaybyJJOpgMDYWzTlzNCetOY+uOcY46fBlrVi3vKgj2xTCQpAEwNBTWDq9g7fCK/my/L1uVJA0Uw0CSZBhIkgwDSRKGgSQJw0CShGEgScIwkCRhGEiSMAwkSRgGkiQMA0kShoEkCcNAkoRhIEnCMJAkYRhIkjAMJEkYBpIkDANJEoaBJAnDQJKEYSBJwjCQJGEYSJIwDCRJGAaSJAwDSRKGgSQJw0CShGEgScIwkCQxD2GQ5K1Jbk9yW5IrkyybtjxJLkny0yS3Jjm53zVJkvbU1zBIcgywDhipqmcChwDnTlvtpcBxzeN84CP9rEmS9Fjz0Uy0BDg0yRLgMOAX05a/EvhUddwEHJHkifNQlySp0dcwqKp7gPcDdwH3Ag9U1Q3TVjsGuHvK9OZm3h6SnJ9kNMno2NhYv0qWpINSv5uJHk/nm//TgCcBy5OcN321GV5aj5lRdVlVjVTVyPDwcO+LlaSDWL+biV4C/KyqxqpqF3AN8Pxp62wGnjJl+sk8tilJktRH/Q6Du4BTkhyWJMAZwB3T1rkW+KPmqqJT6DQl3dvnuiRJUyzp58ar6uYkVwO3AI8A3wMuS3JBs/xS4DrgbOCnwK+AP+lnTZKkx0rVY5rnF7yRkZEaHR1tuwxJGihJ1lfVyEzLvANZkmQYSJIMA0kShoEkCcNAkoRhIEnCMJAkYRhIkjAMJEkYBpIkDANJEoaBJAnDQJJEF11YJ9nBY0cgewAYBd5WVRt7WZgkaf50M57BRXRGIPssnaEqzwVWAz8GPg6c3uviJEnzo5tmorOq6m+rakdVba+qy4Czq+rzwOP7VJ8kzcnERLFx7EG+e+f9bBx7kImJwRurpU3dHBlMJDkHuLqZfvWUZQv+X31ioti0bSdbto9z9MplrFm1nKGhtF2WpB6YmCiuv/0+LrxqA+O7Jli2dIiLzjmJs05c7e/5HHVzZPBa4A+BrcCW5vl5SQ4F3tiH2npm8oNy9iU38pqP3szZl9zI9bff5zcHaZHYtG3n7iAAGN81wYVXbWDTtp0tVzY4ugmDX1bVy6vqyKoarqqXA49W1UNV9e1+FdgLflCkxW3L9vHdv9+TxndNsHXHeEsVDZ5uwuDLSVZOTiT5beDLvS+p9/ygSIvb0SuXsWzpnn/Oli0d4qjDl7VU0eDpJgz+mk4grEjyXDrnDs7rT1m95QdFWtzWrFrOReectPv3fPKcwZpVy1uubHDM+QRyVX0lyVLgBuBw4FVV9ZO+VdZDkx+U6SeX/KBIi8PQUDjrxNWcsO40tu4Y56jDvUikW6na+0nUJB9kz6uFXgxsBDYBVNW6fhU3m5GRkRodHe3qNZNXE/lBkXSwSrK+qkZmWjaXI4Ppf3XXH3hJ829oKKwdXsHa4RVtlyJJC84+w6CqLp/LhpL8fVX9/oGXJEmab73sqG5tD7clSZpHvQwD7+CSpAFlF9aSpJ6GgZfmSNKAmlMYJDkkyWf2sdpf9KAeSVIL5hQGVfUoMJzkcXtZ54aeVSVJmlfddGG9CfhOkmuB3T28VdVFvS5qMbDLbEmDpJsw+EXzGKLTHYVmYd/qkgZNN30TvQcgyfKqsu/nvZity+wT1p3mHdCSFqQ5X02U5NQkPwTuaKafneTD+3jN8Uk2THlsT/KWaeucnuSBKeu8a392ZCGxy2xJg6abZqIPAL8LXAtQVd9P8qK9vaCqfgycBJ0rkoB7gC/MsOqNVfWyLmpZ0Ca7zJ4aCHaZLWkh6+o+g6q6e9qsR7t4+RnAnVX1827ecxDZt7qkQdPNkcHdSZ4PVHOJ6TqaJqM5Ohe4cpZlpyb5Pp0T1P+5qm6fvkKS84HzAY499tgu3nb+2be6pEGzz/EMdq+YHAlcDLyEzhHFPwJvrqptc3jt4+j8oT+xqrZMW7YSmKiqB5OcDVxcVcftbXv7M56BJB3sDnQ8AwCq6n7gtftZw0uBW6YHQbPd7VOeX5fkw0mObN5PkjQPurmaaG2SLycZS7I1yZeSzLXb6tcwSxNRktVJ0jx/XlPTPo82JEm9080J5M8CVwFPBJ4E/B2znwPYLclhwJnANVPmXZDkgmby1cBtzTmDS4Bza65tV5KknujmBHKq6tNTpj+T5I37elFV/QpYNW3epVOefwj4UBd1SJJ6rJsw+GaSdwCfozOQzR8AX0nyBICq+pc+1CdJmgfdhMEfND//07T5f0onHBz2UpIGVDdXEz1tb8uTnFlVXzvwkiRJ862XI529t4fbkiTNI4e9lCT1NAy8HFSSBlQvw0CSNKC6uZpoXzb1cFuSFgGHfx0ccw6D5k7itwHHVtWfJTkOOL6q/gGgqv5Dn2qUNIAc/nWwdNNM9Ang18CpzfRm4H/0vCJJi8Jsw79u2uaouQtRN2Hw9Kr6n8AugKp6CK8gkjQLh38dLN2EwcNJDqW5aijJ0+kcKUjSY0wO/zqVw78uXN2EwV8C1wNPSXIF8HXg7X2pStLAc/jXwTLnkc4AkqwCTqHTPHRTWwPQONKZNBgmryZy+NeFoScjnSX5PeAbVfWVZvqIJK+qqi/2pkxJi83QUFg7vIK1wyvaLkX70FUzUVU9MDlRVb+k03QkSRpw3YTBTOv28qY1SVJLugmD0SQXJXl6Mx7y/wLW96swSdL86SYM3gQ8DHyezvjH48Ab+lGUJGl+dTO4zU7gHX2sRZLUkn2GQZIPVNVbknyZGbqprqpX9KUySdK8mcuRwaebn+/vZyGSpPbsMwyqan2SQ4A/q6rz5qEmSdI8m9MJ5Kp6FBhO8rg+1yNJakE39wlsAr6T5Fpgdx+0VXVRr4uSJM2vbsLgF81jCDi8P+VIktrQzaWl7wFIsrIzWTv6VpUkaV7N+aazJCNJfgDcCvwgyfeTPLd/pUntmJgoNo49yHfvvJ+NYw8yMTH3nn2lQdVNM9HHgddX1Y0ASV5IZyjMZ/WjMKkNjturg1U33VHsmAwCgKr6NmBTkRYVx+3VwaqbMPjnJH+b5PQk/y7Jh4F/SnJykpP7VaA0nxy3VwerbpqJTmp+Th/D4Pl0uql4cS8Kkto0OW7v1EBw3F4dDLq5muh39rY8yeuq6vIDL0lqz+S4vdPPGThurxa7Xg5O82bAMNBAGxoKZ524mhPWnea4vTqodHPOYF8e89uS5PgkG6Y8tid5y7R1kuSSJD9NcqvnH9S2yXF7T1l7JGuHVxgEOij08shgpu6tf0xzrqHp7O4e4AvTVnspcFzz+LfAR5qfkqR50ssw2NfXpzOAO6vq59PmvxL4VFUVcFOSI5I8saru7WFtkujcR7Fp2062bB/n6JU2gelf9TIMvrOP5ecCV84w/xjg7inTm5t5e4RBkvOB8wGOPfbY/a9SOkh5Q532Zs5hkOTCGWY/AKyvqg1V9ca9vPZxwCuAd860eIZ5MzU5XQZcBjAyMmL/AFKXZruh7oR1p7F2eEXL1alt3ZxAHgEuoPOt/Rg639JPBz6a5O37eO1LgVuqassMyzYDT5ky/WQ6vaNK6iFvqNPedBMGq4CTq+ptVfU2OuEwDLwI+ON9vPY1zNxEBHAt8EfNVUWnAA94vkDqvckb6qbyhjpN6iYMjgUenjK9C3hqVT0E/Hq2FyU5DDgTuGbKvAuSXNBMXgdsBH4KfBR4fRc1SZqjyRvqJgPBG+o0VTcnkD9L52qfLzXTLweuTLIc+OFsL6qqX9E5qpg679Ipzwt4Qxd1SNoP3lCnvUnnb/EcV+6MX/BCOid9v11Vo/0qbG9GRkZqdLSVt5akgZVkfVWNzLSsm6uJLgY+X1UX96wySdKC0M05g1uA/950G/G+JDOmi9rh6FySDkQ3vZZeDlye5AnA7wPvTXJsVR3Xt+o0J95MJOlA7U9Hdb8FnACsAX7U02q0XxydS9KBmnMYJHlvkp8AfwXcBjy3ql7et8o0Z95MJOlAdXNp6c/ojGq2FvgN4FlJqKpv9aUyzZmjc0k6UN00Ez0KfAO4HngP8I/Au/tQk7rkzUSSDlQ3RwbrgH8D3FRVv5PkBDqhoJZ5M5GkA9VNGIxX1XgSkvxGVf0oyfF9q0xdmRydy94nJe2PbsJgc5IjgC8CX0vy/7B3UUlaFLq5z+D3mqfvTvJN4DfpnD+QJA24/RrprKr+T68LkSS1Z39uOpMkLTKGgSTJMJAkGQaSJAwDSRKGgSQJw0CShGEgSWI/bzqT1BsTE8WmbTvZsn2co1fawaDaYxhILXG4Ui0kNhNJLXG4Ui0khoHUEocr1UJiGEgtmRyudCqHK1VbDAOpJQ5XqoXEE8hSSxyuVAuJYSC1yOFKtVDYTCRJMgwkSYaBJAnDQJKEYSBJYh7CIMkRSa5O8qMkdyQ5ddry05M8kGRD83hXv2uSJO1pPi4tvRi4vqpeneRxwGEzrHNjVb1sHmqRJM2gr2GQZCXwIuCPAarqYeDhfr6nJKl7/W4mWguMAZ9I8r0kH0sy0732pyb5fpKvJjmxzzVJkqbpdxgsAU4GPlJVzwF2Au+Yts4twFOr6tnAB4EvzrShJOcnGU0yOjY21seSJeng0+8w2Axsrqqbm+mr6YTDblW1vaoebJ5fByxNcuT0DVXVZVU1UlUjw8PDfS5bkg4ufQ2DqroPuDvJ8c2sM4AfTl0nyeokaZ4/r6lpWz/rkiTtaT6uJnoTcEVzJdFG4E+SXABQVZcCrwb+PMkjwEPAuVVV81CXJKmRQfy7OzIyUqOjo22XIUkDJcn6qhqZaZl3IEuSDANJkmEgScIwkCRhGEiSMAwkSRgGkiQMA0kShoEkifnpjkI6IBMTxaZtO9myfZyjVy5jzarlDA2l7bKkRcUw0II2MVFcf/t9XHjVBsZ3TbBs6RAXnXMSZ5242kCQeshmIi1om7bt3B0EAOO7Jrjwqg1s2raz5cqkxcUw0IK2Zfv47iCYNL5rgq07xluqSFqcDAMtaEevXMaypXt+TJctHeKow5e1VJG0OBkGWtDWrFrOReectDsQJs8ZrFk101DakvaXJ5C1oA0NhbNOXM0J605j645xjjrcq4mkfjAMtOANDYW1wytYO7yi7VKkRctmIkmSYSBJMgwkSXjOQPvB7iGkxccwUFfsHkJanGwmUlfsHkJanAwDdcXuIaTFyTBQV+weQlqcDAN1xe4hpMXJE8jqit1DSIuTYaCu2T2EtPjYTCRJMgwkSYaBJAnDQJKEYSBJAlJVbdfQtSRjwM/brmOOjgTub7uIPlnM+waLe//ct8F1IPv31KoanmnBQIbBIEkyWlUjbdfRD4t532Bx75/7Nrj6tX82E0mSDANJkmEwHy5ru4A+Wsz7Bot7/9y3wdWX/fOcgSTJIwNJkmEgScIw6JskT0nyzSR3JLk9yZvbrqnXkhyS5HtJ/qHtWnopyRFJrk7yo+b/79S2a+qVJG9tPo+3JbkyyUCPSpTk40m2JrltyrwnJPlakp80Px/fZo37a5Z9e1/zubw1yReSHNGr9zMM+ucR4G1V9dvAKcAbkjyj5Zp67c3AHW0X0QcXA9dX1QnAs1kk+5jkGGAdMFJVzwQOAc5tt6oD9kngrGnz3gF8vaqOA77eTA+iT/LYffsa8Myqehbwf4F39urNDIM+qap7q+qW5vkOOn9Qjmm3qt5J8mTg3wMfa7uWXkqyEngR8L8Bqurhqvplq0X11hLg0CRLgMOAX7RczwGpqm8B/zJt9iuBy5vnlwOvms+aemWmfauqG6rqkWbyJuDJvXo/w2AeJFkDPAe4ueVSeukDwNuBiZbr6LW1wBjwiaYJ7GNJFsWYnlV1D/B+4C7gXuCBqrqh3ar64uiquhc6X8qAo1qup1/+FPhqrzZmGPRZkhXA3wNvqartbdfTC0leBmytqvVt19IHS4CTgY9U1XOAnQxuM8MemrbzVwJPA54ELE9yXrtVaX8k+W90mqKv6NU2DYM+SrKUThBcUVXXtF1PD70AeEWSTcDngBcn+Uy7JfXMZmBzVU0exV1NJxwWg5cAP6uqsaraBVwDPL/lmvphS5InAjQ/t7ZcT08leR3wMuC11cMbxQyDPkkSOu3Od1TVRW3X00tV9c6qenJVraFzAvIbVbUovmFW1X3A3UmOb2adAfywxZJ66S7glCSHNZ/PM1gkJ8enuRZ4XfP8dcCXWqylp5KcBfwF8Iqq+lUvt20Y9M8LgD+k8615Q/M4u+2iNCdvAq5IcitwEvDX7ZbTG83RztXALcAP6Pz+D3TXDUmuBL4LHJ9kc5L/CPwNcGaSnwBnNtMDZ5Z9+xBwOPC15m/KpT17P7ujkCR5ZCBJMgwkSYaBJAnDQJKEYSBJwjCQZpSkknx6yvSSJGP720Nr0xPq66dMn77YenvVYDMMpJntBJ6Z5NBm+kzgngPY3hHA6/e1ktQWw0Ca3Vfp9MwK8BrgyskFTZ/5X2z6lb8pybOa+e9u+qH/pyQbk6xrXvI3wNObG4Xe18xbMWXchCuau4KlVhgG0uw+B5zbDADzLPbsdfY9wPeafuX/K/CpKctOAH4XeB7wl00fVe8A7qyqk6rqvzTrPQd4C/AMOr2lvqCP+yLtlWEgzaKqbgXW0DkquG7a4hcCn27W+wawKslvNsu+UlW/rqr76XSSdvQsb/HPVbW5qiaADc17Sa1Y0nYB0gJ3LZ0xAE4HVk2ZP1OTzmTfLr+eMu9RZv89m+t6Ut95ZCDt3ceBv6qqH0yb/y3gtdC5Mgi4fx/jVeyg08GYtCD5TUTai6raTGdM5OneTWc0tFuBX/GvXSbPtp1tSb7TDG7+VeArva5VOhD2WipJsplIkmQYSJIwDCRJGAaSJAwDSRKGgSQJw0CSBPx/S3XIv3As+eoAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "avg_month=training_day.groupby('Month')['avg_price_per_kg'].mean()\n",
    "sns.scatterplot(x=avg_month.index,y=avg_month)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The middle months seem to have lower means while the end and start months have higher means"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='Quarter', ylabel='avg_price_per_kg'>"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYkAAAEGCAYAAACQO2mwAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAAAei0lEQVR4nO3dfZRdVZnn8e+vkmCRSiIYKuGdShwaJC6IeI2AyouIk7AaEWXFsHxptJ0YhxeF6R5Zqx207bFnBCetCBojg28tbyIItBCgbXUQFamEAokYCKGAEEiKoHmTkoR65o9zKpy6uedW3co9dW9Vfp+1zrr37r3Puc+GVD11zj5nb0UEZmZmlbQ0OgAzM2teThJmZpbLScLMzHI5SZiZWS4nCTMzyzW+0QHU03777RcdHR2NDsPMbFRZvnz5CxHRXqluTCWJjo4OOjs7Gx2GmdmoIumpvDpfbjIzs1yFJwlJF0laKekRSddJai2rP1nSJkld6XZppm6upFWSVku6pOhYzcxsoEKThKSDgAuBUkS8ERgHLKjQ9N6ImJ1uX0j3HQdcBcwDjgLOkXRUkfGamdlAI3G5aTywt6TxwERg3RD3mwOsjog1EfEycD1wZkExmplZBYUmiYh4Fvgy8DTwHLApIu6u0PR4SQ9JulPSrLTsIOCZTJu1aZmZmY2Qoi837Uvy1/8M4ECgTdKHypqtAA6LiGOArwE/7t+9wiF3mY1Q0kJJnZI6e3p6hhVnX1+wpmcrv37iBdb0bKWvz5MemplB8Zeb3gU8GRE9EbEduBk4IdsgIjZHxNb0/R3ABEn7kZw5HJJpejAVLlVFxNKIKEVEqb294m2+VfX1BctWPs/pV9zLOd+6n9OvuJdlK593ojAzo/gk8TRwnKSJkgScCjyabSBp/7QOSXPSmDYCDwCHS5ohaS+SAe/b6h1g98ZtXHxjF73b+wDo3d7HxTd20b1xW72/ysxs1Cn0YbqIuF/STSSXlHYADwJLJS1K65cAZwOflLQDeAlYEMkiFzsknQ/cRXJX1DURsbLeMa7f3LszQfTr3d7Hhi29zGyfVO+vMzMbVQp/4joiPgd8rqx4Sab+SuDKnH3vAO4oLjqYPqWV1gktAxJF64QWpk1urbKXmdmeYY9/4rpjahuL58+mdULyn6J1QguL58+mY2pbgyMzM2u8MTV303C0tIi5s/bnyAvfwYYtvUyb3ErH1DZaWirdXGVmtmfZ45MEJIliZvskj0GYmZXZ4y83mZlZPicJMzPL5SRhZma5nCTMzCyXk4SZmeVykjAzs1xOEmZmlstJwszMcjlJmJlZLicJMzPL5SRhZma5nCTMzCyXk4SZmeUqPElIukjSSkmPSLpOUmtZ/QclPZxuv5J0TKauW9LvJHVJ6iw6VjMzG6jQJCHpIOBCoBQRbyRZhnRBWbMngZMi4mjgn4ClZfWnRMTsiCgVGauZme1qJNaTGA/sLWk7MBFYl62MiF9lPv4GOHgEYjIzsyEo9EwiIp4Fvgw8DTwHbIqIu6vs8rfAndlDAHdLWi5pYaUdJC2U1Cmps6enp16hm5kZxV9u2hc4E5gBHAi0SfpQTttTSJLEZzLFb4uIY4F5wHmSTizfLyKWRkQpIkrt7e1174OZ2Z6s6IHrdwFPRkRPRGwHbgZOKG8k6WjgauDMiNjYXx4R69LXDcAtwJyC4zUzs4yik8TTwHGSJkoScCrwaLaBpENJkseHI+KxTHmbpMn974F3A48UHK+ZmWUUOnAdEfdLuglYAewAHgSWSlqU1i8BLgWmAl9P8gg70juZpgO3pGXjgWsjYlmR8ZqZ2UCKiEbHUDelUik6O/04hZlZLSQtz3vMwE9cm5lZLicJMzPL5SRhZma5nCTMzCyXk4SZmeVykjAzs1xOEmZmlstJwszMcjlJmJlZLicJMzPL5SRhZma5nCTMzCyXk4SZmeVykjAzs1xOEmZmlstJwszMchWeJCRdJGmlpEckXSeptaxekq6QtFrSw5KOzdTNlbQqrbuk6FjNzGygQpOEpIOAC4FSRLwRGAcsKGs2Dzg83RYC30j3HQdcldYfBZwj6agi4zUzs4FG4nLTeGBvSeOBicC6svozge9F4jfAPpIOAOYAqyNiTUS8DFyftjUzsxFSaJKIiGeBLwNPA88BmyLi7rJmBwHPZD6vTcvyygeQtFBSp6TOnp6eeoZvZrbHK/py074kf/3PAA4E2iR9qLxZhV2jSvnAgoilEVGKiFJ7e/vuhmxmZhlFX256F/BkRPRExHbgZuCEsjZrgUMynw8muSSVV25mZiOk6CTxNHCcpImSBJwKPFrW5jbgI+ldTseRXJJ6DngAOFzSDEl7kQx431ZwvGZmljG+yINHxP2SbgJWADuAB4Glkhal9UuAO4DTgdXAn4GPpnU7JJ0P3EVyV9Q1EbGyyHjNzGwgRexymX/UKpVK0dnZ2egwzMxGFUnLI6JUqc5PXJuZWS4nCTMzy+UkYWZmuZwkzMwsl5OEmZnlcpIwM7NcThJmZpZryA/TSdrCrnMnbQI6gf8WEWvqGZiZmTVeLU9cLyaZO+laksn3FgD7A6uAa4CT6x2cmZk1Vi2Xm+ZGxDcjYktEbI6IpcDpEXEDsG9B8ZmZWQPVkiT6JM2X1JJu8zN1Y2duDzMz26mWJPFB4MPABmB9+v5DkvYGzi8gNjMza7BaxiT+FBFnZAskzYiIl4Bf1jcsMzNrBrWcSdwuaUr/B0lvAG6vf0hmZtYsakkS/0ySKCZJejNwE1C+FKmZmY0hQ77cFBE/kTQBuBuYDLw3Ih4vLDIzM2u4QZOEpK8x8O6lKcAa4AJJRMSFVfY9ArghUzQTuDQivpJp8/ckg+L98bwBaI+IFyV1A1uAV4AdeYtimJlZMYZyJlG+1NvyoR48IlYBswEkjQOeBW4pa3M5cHna5gzgooh4MdPklIh4YajfaWZm9TNokoiI7w7lQJJ+FBHvr9LkVOCJiHiqSptzgOuG8n1mZgZ9fUH3xm2s39zL9CmtdExto6VFdTt+LbfADmbmIPULqJIAJE0E5jLwmYsA7pYUwDfTp7zL91sILAQ49NBDa43ZzGzU6usLlq18notv7KJ3ex+tE1pYPH82c2ftX7dEUc9ZYHOfupa0F/Ae4IdV9j8DuK/sUtPbIuJYYB5wnqQTd/nSiKURUYqIUnt7+zBDNzMbfbo3btuZIAB6t/dx8Y1ddG/cVrfvGKmpwucBKyJifZU2u5xpRMS69HUDyVjGnMIiNDMbZdZv7t2ZIPr1bu9jw5beun1HPZNEtXObqmMNkl4LnATcmilrkzS5/z3wbuCR+oRqZjb6TZ/SSuuEgb/GWye0MG1ya92+Y0hJQtI4Sf86SLPP5Ow7ETgNuDlTtkjSokyzs4C7IyJ7jjQd+KWkh4DfAj+JiGVDidfMbE/QMbWNxfNn70wU/WMSHVPb6vYdihjaBK6S7gLOiIiX6/btdVYqlaKzs/yOXTOzsav/7qYNW3qZNnl4dzdJWp73HFotdzd1A/dJug3Y+Rd/RCyuKRozM6ublhYxs30SM9snFXL8WpLEunRrIZmWw8zMxrha5m76R0gGkcvGDszMbIwa8t1Nko6X9Hvg0fTzMZK+XlhkZmbWcLXcAvsV4D8DGwEi4iFgl4fbzMxs7KjpOYmIeKas6JU6xmJmZk2mloHrZySdAEQ6zcaFpJeezMxsbKrlTGIRcB5wEMmU37PTz2ZmNkbVcnfTC7y6OJCZme0Barm7aaak2yX1SNog6VZJg00PbmZmo1gtl5uuBW4EDgAOJJn22wsEmZmNYbUkCUXE9yNiR7r9K1XWkDAzs9GvlrubfibpEuB6kuTwAeAnkl4HULZYkJmZjQG1JIkPpK+fKCv/GEnS8PiEmdkYU8vdTTOq1Us6LSLu2f2QzMysWdRzZbov1fFYZmbWBApdvlTSEZK6MttmSZ8ua3OypE2ZNpdm6uZKWiVpdToeYmZmI6iWMYnB7HKnU0SsInkyG0njSJ7UvqXCvvdGxF9nC9L2V5EsfboWeEDSbRHx+zrGbGZmVdTzTGIwpwJPRMRTQ2w/B1gdEWvSJVOvB84sLDozM9tFPZNE9yD1C8h/+O54SQ9JulPSrLTsICA76+zatGwASQsldUrq7OnpqTVmMzOropZpOSZK+h+SvpV+PlzSzktEEfG+KvvuBbyH5CntciuAwyLiGOBrwI/7d6vQttIlraURUYqIUnt7+1C7Y2ZmQ1DLmcS3gb8Ax6ef1wL/c4j7zgNWRMT68oqI2BwRW9P3dwATJO2XHv+QTNODSdbYNjOzEVJLknh9RFwGbAeIiJeo/Nd+JeeQc6lJ0v6SlL6fk8a0EXgAOFzSjPRMZAFwWw3xmpnZbqrl7qaXJe1NeslH0utJziyqkjSR5A6lT2TKFgFExBLgbOCTknYALwELIiKAHZLOB+4CxgHXRMTKGuI1M7PdpOT38RAaSqcBnwWOAu4G3gacGxE/Lyy6GpVKpejs7Gx0GGZmo4qk5RFRqlRXy7Qc90haARxHcpnpU+lCRGZmNkbVcnfTWcCOiPhJRPwbyeWg9xYWmZmZNVwtA9efi4hN/R8i4k/A5+oekZmZNY1akkSltvWc1sPMzJpMLUmiU9JiSa9P17v+F2B5UYGZmVnj1ZIkLgBeBm4geXK6FziviKDMzKw51HJ30zbA03Wbme1BBk0Skr4SEZ+WdDuV5056TyGRmZlZww3lTOL76euXiwzEzMyaz6BJIiKWpwsA/ZeI+NAIxGRmZk1iSAPXEfEK0J5OtGdmZnuIWp5z6Abuk3QbsK2/MCIW1zsoMzNrDrUkiXXp1gJMLiYcMzNrJrXcAvuPAJKmJB9jS2FRWVPr6wu6N25j/eZepk9ppWNqGy0tQ11axMxGkyEnCUklktXpJqefNwEfiwg/db0H6esLlq18notv7KJ3ex+tE1pYPH82c2ft70RhNgbV8sT1NcB/jYiOiOggedr629V2kHSEpK7MtlnSp8vafFDSw+n2K0nHZOq6Jf0u3dcLRTSB7o3bdiYIgN7tfVx8YxfdG7cNsqeZjUa1jElsiYh7+z9ExC8lVb3kFBGrgNkA6W20zwK3lDV7EjgpIv4oaR6wFHhrpv4Ur1vRPNZv7t2ZIPr1bu9jw5ZeZrZPalBUZlaUWpLEbyV9k2St6gA+APxc0rEAEbFikP1PBZ6IiKeyhRHxq8zH3wAH1xCTjbDpU1ppndAyIFG0Tmhh2uTWBkZlZkWpJUnMTl/L15A4gSRpvHOQ/ReQJJhq/ha4M/M5gLslBfDNiFhavoOkhcBCgEMPPXSQw9vu6pjaxuL5s3cZk+iY2tbo0MysAENe43rQA0l/ExHfzanbi+T22VkRsT6nzSnA14G3R8TGtOzAiFgnaRpwD3BBRPy/vBi8xvXI6L+7acOWXqZN9t1NZqNdtTWuaxm4HsynqtTNA1ZUSRBHA1cDZ/YnCICIWJe+biAZy5hTv3BtuFpaxMz2SRw3cz9mtk9ygjAbw+qZJKr9pjiHnEtNkg4FbgY+HBGPZcrbJPXfbtsGvBt4pH7hmpnZYOq5/GjF61aSJgKnAZ/IlC0CiIglwKXAVODrkgB2pKc904Fb0rLxwLURsayO8ZqZ2SDqmSQqnklExJ9JkkC2bEnm/ceBj1fYbw1wTHm5mZmNnHpebrqvjscyM7MmUMu0HBdXKN4ELI+Irog4v35hmZlZM6jlTKIELAIOSreFwMnAtyT99/qHZmZmjVbLmMRU4NiI2Aog6XPATcCJwHLgsvqHZ2ZmjVTLmcShwMuZz9uBwyLiJeAvdY3KzMyaQi1nEtcCv5F0a/r5DOC69BmG39c9MjMza7haFh36J0l3AG8nud11UUT0z4HxwSKCMzOzxqrl7qavAjdExFcLjMfMzJpILWMSK4DPSlot6fJ0pTozMxvDhpwkIuK7EXE6ySR7jwFfkvR4YZGZmVnDDeeJ6/8EHAl0AH+oazRmZtZUhpwkJPWfOXyBZDbWN0fEGYVFZmZmDVfLLbBPkqxCNxN4DXC0JKotAmRmZqNbLUniFeA/SNag7gKOA37N4MuWmpnZKFXLmMSFwFuApyLiFOBNQE8hUZmZWVOoJUn0RkQvgKTXRMQfgCOKCcvMzJpBLUliraR9gB8D96TTc6yrtoOkIyR1ZbbNkj5d1kaSrkifv3hY0rGZurmSVqV1l9QQq5mZ1UEt03Kclb79vKSfAa8Fqi4nGhGrgNkAksYBzwK3lDWbBxyebm8FvgG8NW1/FcnSp2uBByTdFhGeJ8rMbIQMa/nSiPjFMHY7FXgiIp4qKz8T+F5EBMkEgvtIOoDkOYzV6TKmSLo+beskYWY2Quq5fOlgFgDXVSg/CHgm83ktry5sVKl8AEkLJXVK6uzp8Ti6mVk9jUiSkLQX8B7gh5WqK5RFlfKBBRFLI6IUEaX29vbdC9TMzAYY1uWmYZgHrIiI9RXq1gKHZD4fTDIgvldOuZmZjZCRutx0DpUvNQHcBnwkvcvpOGBTRDwHPAAcLmlGeiayIG1rZmYjpPAzCUkTSe5Q+kSmbBFARCwB7gBOB1YDfwY+mtbtkHQ+cBcwDrgmIlYWHa+Zmb2q8CQREX8GppaVLcm8D+C8nH3vIEkiZmbWACN5d5OZmY0yThJmZpbLScLMzHI5SZiZWS4nCTMzyzVSD9OZmQ1JX1/QvXEb6zf3Mn1KKx1T22hpqTQBg40EJwkzaxp9fcGylc9z8Y1d9G7vo3VCC4vnz2burP2dKBrEl5vMrGl0b9y2M0EA9G7v4+Ibu+jeuK3Bke25nCTMrGms39y7M0H0693ex4YtvQ2KyJwkzKxpTJ/SSuuEgb+WWie0MG1ya4MiMicJM2saHVPbWDx/9s5E0T8m0TG1rcGR7bk8cG1mTaOlRcydtT9HXvgONmzpZdpk393UaE4SZtZUWlrEzPZJzGyf1OhQDF9uMjOzKpwkzMwsl5OEmZnlGomV6fYBrgbeCATwsYj4dab+74EPZuJ5A9AeES9K6ga2AK8AOyKiVHS8Zmb2qpEYuP4qsCwizk7Xqp6YrYyIy4HLASSdAVwUES9mmpwSES+MQJxmZlam0CQhaQpwInAuQES8DLxcZZdzgOuKjMnMzIau6DGJmUAP8G1JD0q6WlLFp2IkTQTmAj/KFAdwt6Tlkhbm7LdQUqekzp6ennrHb2a2Rys6SYwHjgW+ERFvArYBl+S0PQO4r+xS09si4lhgHnCepBPLd4qIpRFRiohSe3t7ncM3M9uzFZ0k1gJrI+L+9PNNJEmjkgWUXWqKiHXp6wbgFmBOQXGamVkFhSaJiHgeeEbSEWnRqcDvy9tJei1wEnBrpqxN0uT+98C7gUeKjNfMzAYaibubLgB+kN7ZtAb4qKRFABGxJG1zFnB3RGQnjZ8O3CKpP85rI2LZCMRrZmYpRUSjY6ibUqkUnZ2djQ7DzGxUkbQ87zk0P3FtZma5nCTMzCyXk4SZmeVykjAzs1xOEmZmlstJwszMcjlJmJlZLicJMzPL5SRhZma5nCTMzCyXk4SZmeVykjAzs1xOEmZmlstJwszMcjlJmJlZLicJMzPLVXiSkLSPpJsk/UHSo5KOL6s/WdImSV3pdmmmbq6kVZJWS7qk6FjNzGygkVi+9KvAsog4O13CdGKFNvdGxF9nCySNA64CTgPWAg9Iui0idlkj28zMilHomYSkKcCJwP8FiIiXI+JPQ9x9DrA6ItZExMvA9cCZhQRqZmYVFX25aSbQA3xb0oOSrpbUVqHd8ZIeknSnpFlp2UHAM5k2a9OyASQtlNQpqbOnp6fuHTDbXX19wZqerfz6iRdY07OVvr6xs668jX1FJ4nxwLHANyLiTcA2oHxsYQVwWEQcA3wN+HFargrH2+WnKyKWRkQpIkrt7e11C9ysHvr6gmUrn+f0K+7lnG/dz+lX3Muylc87UdioUXSSWAusjYj70883kSSNnSJic0RsTd/fAUyQtF+67yGZpgcD6wqO16yuujdu4+Ibu+jd3gdA7/Y+Lr6xi+6N2xocmdnQFJokIuJ54BlJR6RFpwIDBp4l7S9J6fs5aUwbgQeAwyXNSAe8FwC3FRmvWb2t39y7M0H0693ex4YtvQ2KyKw2I3F30wXAD9Jf9GuAj0paBBARS4CzgU9K2gG8BCyIiAB2SDofuAsYB1wTEStHIF6zupk+pZXWCS0DEkXrhBamTW5tYFRmQ6fk9/HYUCqVorOzs9FhmO3UPybRf8mpdUILi+fPZu6s/WlpqTTsZjbyJC2PiFKlupE4kzDbY7W0iLmz9ufIC9/Bhi29TJvcSsfUNicIGzWcJMwK1tIiZrZPYmb7pEaHYlYzz91kZma5nCTMzCyXk4SZmeVykjAzs1xOEmZmlmtMPSchqQd4qtFxDGI/4IVGB1EnY6UvY6Uf4L40q2bvy2ERUXHyuzGVJEYDSZ15D62MNmOlL2OlH+C+NKvR3BdfbjIzs1xOEmZmlstJYuQtbXQAdTRW+jJW+gHuS7MatX3xmISZmeXymYSZmeVykjAzs1xOEoOQdIikn0l6VNJKSZ9Ky18n6R5Jj6ev+6blU9P2WyVdWXasD0h6OD3OZVW+84uSnpG0taz8REkrJO2QdPYI9OU0Scsl/S59fWfmWG9Oy1dLuqJ/dcEK31mx3e70pcn6sSgt75L0S0lHjeK+nCupJ+1Ll6SPj9J+/EumD49J+tNQ+9Gk/TlM0k+V/O74uaSDa+3PbokIb1U24ADg2PT9ZOAx4CjgMuCStPwS4Evp+zbg7cAi4MrMcaYCTwPt6efvAqfmfOdx6fduLSvvAI4GvgecPQJ9eRNwYPr+jcCzmWP9FjgeEHAnMC/nOyu2252+NFk/pmTavAdYNor7cm723+xo7UdZmwtIVrUctf0Bfgj8Tfr+ncD3h/P/aLjbiH3RWNmAW4HTgFXAAZl/UKvK2g34gQPeAvx75vOHga8P8l1bc8q/wzCSxHD7kpaLZO3x16Rt/pCpOwf4ZoV9Bm1Xj740Qz8y5XeO1r6U/5sdrf0oa/cr4LTR3B9gJXBw5tib6/H/aKibLzfVQFIHyV8M9wPTI+I5gPR12iC7rwaOlNQhaTzwXuCQ4qKtbhh9eT/wYET8BTgIWJupW5uWlRtqu2Frhn5IOk/SEyR/ZV44mvsCvD+9rHGTpGH9+2ySfiDpMGAG8B/D6UfmOB00tj8PpccEOAuYLGnqsDozDE4SQyRpEvAj4NMRsbnW/SPij8AngRuAe4FuYEc9YxyqWvsiaRbwJeAT/UUVmlW6l3qo7YalWfoREVdFxOuBzwCfHSyOnNiaoS+3Ax0RcTTw7ySXRGvSJP3otwC4KSJeGSyOKvE1Q3/+DjhJ0oPAScCzjODvDieJIZA0geQfyg8i4ua0eL2kA9L6A4ANgx0nIm6PiLdGxPEkp62PSxqXGWT7QlF96FdrX9JBsluAj0TEE2nxWiA7eHYwsK5CXyq2G8P9uJ7kDHFU9iUiNqZ//QJ8C3jzaOxHxgLgulr60Iz9iYh1EfG+iHgT8A9p2abh9qtmI3ltazRuJBn+e8BXysovZ+AA1mVl9edSdn0XmJa+7gt0AX81yHfXdUyi1r4A+5Ce6lY41gMkA+z9g2yn53xn1XbD6Usz9QM4PNPmDKBzFPflgEybs4DfjMZ+pHVHkJytq9afk2brD8kMsi3p+y8CXxhOn4a7jdgXjdaN5E6lAB4m+cXeBZxOcrfST4HH09fXZfbpBl4EtpL8hXBUWn4d8Pt0W1DlOy9L9+tLXz+flr8l/byNZGBsZZF9Ibl0si3TtotXE10JeAR4Argy74cxr93u9KXJ+vFVkoHFLuBnwKxR/P/kf6V9eSjty5GjsR9p3eeB/z1SP/cF/385O/2+x4CrgdcU8bsub/O0HGZmlstjEmZmlstJwszMcjlJmJlZLicJMzPL5SRhZma5nCTMBiHpYEm3pjN/rpF0paTX1OnY50o6sB7HMiuCk4RZFel0zTcDP46Iw4HDgb1JnmXZ3WOPI3nosqYkkc79ZTYinCTMqnsn0BsR3waIZB6gi4CPSDpfmTVDJP2bpJPT99+Q1JmuRfCPmTbdki6V9EuSmT5LwA/S6Rn2TtcU+EW6JsFdmSkgfi7pnyX9AvjUCPXdDP9FYlbdLGB5tiAiNkvqpvrPzz9ExIvp2cJPJR0dEQ+ndb0R8XYAJQv7/F1EdKZzBX0NODMieiR9gGQaho+l++0TESfVr2tmg3OSMKtODH3Wzqz5khaS/IwdQLJgTX+SuCFnnyNIFqy5J12UbBzwXKY+bz+zwjhJmFW3klfn8gdA0hRgOsmcU3+VqWpN62eQTO/8loj4o6Tv9NeltuV8l0jmsDo+pz5vP7PCeEzCrLqfAhMlfQR2Djb/H5IJ2J4EZktqSRfomZPuM4XkF/omSdOBeVWOv4VkeUxIpo9vl3R8+l0T0vUJzBrGScKsikhmwDwLOFvS4yRnD30R8UXgPpJE8Tvgy8CKdJ+HgAdJzkKuSdvl+Q6wRFIXyeWls4EvSXqIZCbRE+reKbMaeBZYsxpIOoFkyvf3RcTywdqbjXZOEmZmlsuXm8zMLJeThJmZ5XKSMDOzXE4SZmaWy0nCzMxyOUmYmVmu/w9+wm/XnqOFYwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "avg_quarter=training_day.groupby('Quarter')['avg_price_per_kg'].mean()\n",
    "sns.scatterplot(x=avg_quarter.index.to_timestamp(),y=avg_quarter)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we saw by the trend in the months plot, the response variable appears to fall as we go toward the middle of the year and then rise as the year ends"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Dropping the non-numeric 'Date' column and adding a Month column\n",
    "train_dummies['Month']=training_day['Month']\n",
    "train_dummies=train_dummies.drop('Date',axis=1)\n",
    "testing_day=pd.DataFrame([])\n",
    "testing_day['Month']=[x.month for x in test['Date']]\n",
    "test_dummies['Month']=testing_day['Month']\n",
    "test_dummies=test_dummies.drop('Date',axis=1)"
   ]
  },
  {
   "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>Weight_Kg</th>\n",
       "      <th>Low_Price</th>\n",
       "      <th>High_Price</th>\n",
       "      <th>Sales_Total</th>\n",
       "      <th>Total_Qty_Sold</th>\n",
       "      <th>Total_Kg_Sold</th>\n",
       "      <th>Stock_On_Hand</th>\n",
       "      <th>avg_price_per_kg</th>\n",
       "      <th>Province_EASTERN CAPE</th>\n",
       "      <th>Province_NATAL</th>\n",
       "      <th>...</th>\n",
       "      <th>Size_Grade_1M</th>\n",
       "      <th>Size_Grade_1S</th>\n",
       "      <th>Size_Grade_1U</th>\n",
       "      <th>Size_Grade_1X</th>\n",
       "      <th>Size_Grade_2L</th>\n",
       "      <th>Size_Grade_2M</th>\n",
       "      <th>Size_Grade_2S</th>\n",
       "      <th>Size_Grade_2U</th>\n",
       "      <th>Size_Grade_2X</th>\n",
       "      <th>Month</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.373103</td>\n",
       "      <td>1.443585</td>\n",
       "      <td>1.288619</td>\n",
       "      <td>2.767888</td>\n",
       "      <td>1.439367</td>\n",
       "      <td>2.403481</td>\n",
       "      <td>1.109645</td>\n",
       "      <td>8.51</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>-0.430568</td>\n",
       "      <td>-0.755970</td>\n",
       "      <td>-0.997932</td>\n",
       "      <td>0.353042</td>\n",
       "      <td>1.358536</td>\n",
       "      <td>1.062677</td>\n",
       "      <td>-0.706778</td>\n",
       "      <td>4.55</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>-0.924725</td>\n",
       "      <td>-0.645992</td>\n",
       "      <td>-0.902659</td>\n",
       "      <td>-0.661991</td>\n",
       "      <td>-0.675722</td>\n",
       "      <td>-0.699108</td>\n",
       "      <td>2.621121</td>\n",
       "      <td>6.11</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>1.373103</td>\n",
       "      <td>-0.096103</td>\n",
       "      <td>0.335889</td>\n",
       "      <td>1.436375</td>\n",
       "      <td>1.816581</td>\n",
       "      <td>2.941145</td>\n",
       "      <td>0.272146</td>\n",
       "      <td>4.51</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>80</th>\n",
       "      <td>-0.924725</td>\n",
       "      <td>-0.645992</td>\n",
       "      <td>-0.902659</td>\n",
       "      <td>-0.401640</td>\n",
       "      <td>-0.204205</td>\n",
       "      <td>-0.368576</td>\n",
       "      <td>-0.306811</td>\n",
       "      <td>6.11</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 33 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    Weight_Kg  Low_Price  High_Price  Sales_Total  Total_Qty_Sold  \\\n",
       "1    1.373103   1.443585    1.288619     2.767888        1.439367   \n",
       "7   -0.430568  -0.755970   -0.997932     0.353042        1.358536   \n",
       "24  -0.924725  -0.645992   -0.902659    -0.661991       -0.675722   \n",
       "40   1.373103  -0.096103    0.335889     1.436375        1.816581   \n",
       "80  -0.924725  -0.645992   -0.902659    -0.401640       -0.204205   \n",
       "\n",
       "    Total_Kg_Sold  Stock_On_Hand  avg_price_per_kg  Province_EASTERN CAPE  \\\n",
       "1        2.403481       1.109645              8.51                      0   \n",
       "7        1.062677      -0.706778              4.55                      0   \n",
       "24      -0.699108       2.621121              6.11                      0   \n",
       "40       2.941145       0.272146              4.51                      0   \n",
       "80      -0.368576      -0.306811              6.11                      0   \n",
       "\n",
       "    Province_NATAL  ...  Size_Grade_1M  Size_Grade_1S  Size_Grade_1U  \\\n",
       "1                0  ...              0              0              0   \n",
       "7                0  ...              0              0              0   \n",
       "24               0  ...              0              0              0   \n",
       "40               0  ...              0              1              0   \n",
       "80               0  ...              0              0              0   \n",
       "\n",
       "    Size_Grade_1X  Size_Grade_2L  Size_Grade_2M  Size_Grade_2S  Size_Grade_2U  \\\n",
       "1               0              0              0              0              0   \n",
       "7               0              0              1              0              0   \n",
       "24              0              0              0              1              0   \n",
       "40              0              0              0              0              0   \n",
       "80              0              0              0              1              0   \n",
       "\n",
       "    Size_Grade_2X  Month  \n",
       "1               0      9  \n",
       "7               0      4  \n",
       "24              0      4  \n",
       "40              0      5  \n",
       "80              0      3  \n",
       "\n",
       "[5 rows x 33 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_dummies.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 1644 entries, 1 to 64310\n",
      "Data columns (total 33 columns):\n",
      " #   Column                         Non-Null Count  Dtype  \n",
      "---  ------                         --------------  -----  \n",
      " 0   Weight_Kg                      1644 non-null   float64\n",
      " 1   Low_Price                      1644 non-null   float64\n",
      " 2   High_Price                     1644 non-null   float64\n",
      " 3   Sales_Total                    1644 non-null   float64\n",
      " 4   Total_Qty_Sold                 1644 non-null   float64\n",
      " 5   Total_Kg_Sold                  1644 non-null   float64\n",
      " 6   Stock_On_Hand                  1644 non-null   float64\n",
      " 7   avg_price_per_kg               1644 non-null   float64\n",
      " 8   Province_EASTERN CAPE          1644 non-null   uint8  \n",
      " 9   Province_NATAL                 1644 non-null   uint8  \n",
      " 10  Province_ORANGE FREE STATE     1644 non-null   uint8  \n",
      " 11  Province_TRANSVAAL             1644 non-null   uint8  \n",
      " 12  Province_W.CAPE-BERGRIVER ETC  1644 non-null   uint8  \n",
      " 13  Province_WEST COAST            1644 non-null   uint8  \n",
      " 14  Container_DT063                1644 non-null   uint8  \n",
      " 15  Container_EC120                1644 non-null   uint8  \n",
      " 16  Container_EF120                1644 non-null   uint8  \n",
      " 17  Container_EG140                1644 non-null   uint8  \n",
      " 18  Container_JE090                1644 non-null   uint8  \n",
      " 19  Container_JG110                1644 non-null   uint8  \n",
      " 20  Container_M4183                1644 non-null   uint8  \n",
      " 21  Container_M6125                1644 non-null   uint8  \n",
      " 22  Container_M9125                1644 non-null   uint8  \n",
      " 23  Size_Grade_1M                  1644 non-null   uint8  \n",
      " 24  Size_Grade_1S                  1644 non-null   uint8  \n",
      " 25  Size_Grade_1U                  1644 non-null   uint8  \n",
      " 26  Size_Grade_1X                  1644 non-null   uint8  \n",
      " 27  Size_Grade_2L                  1644 non-null   uint8  \n",
      " 28  Size_Grade_2M                  1644 non-null   uint8  \n",
      " 29  Size_Grade_2S                  1644 non-null   uint8  \n",
      " 30  Size_Grade_2U                  1644 non-null   uint8  \n",
      " 31  Size_Grade_2X                  1644 non-null   uint8  \n",
      " 32  Month                          1644 non-null   int64  \n",
      "dtypes: float64(8), int64(1), uint8(24)\n",
      "memory usage: 231.5 KB\n"
     ]
    }
   ],
   "source": [
    "train_dummies.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "All of our variables are now numeric and we can begin building regression models"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Phase 4: Model Building"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.1: Feature Selection"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.2: Modelling Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.2.1: Random Forest Model"
   ]
  },
  {
   "attachments": {
    "image.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The Random Forest Model utilise multiple randomly created decision trees by combining their output to generate a prediction\n",
    "![image.png](attachment:image.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The  combination of learning models will increase the accuracy of the result. The random forest algorithm allows easy  measurement of the relative importance of the features used on the prediction. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Utilising the Random Forest Model:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1. Preprocessing:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Setting up the basics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "X=train_dummies.drop('avg_price_per_kg',axis=1)\n",
    "y=train_dummies['avg_price_per_kg']\n",
    "\n",
    "# standardize features\n",
    "scaler = StandardScaler()\n",
    "X_scaled = scaler.fit_transform(X)\n",
    "\n",
    "# Train test split\n",
    "x_train, x_test, y_train,y_test = train_test_split(X_scaled,y,test_size=0.2,random_state=6)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.Training "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Fitting the model to our data and applying hyperparamers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(max_depth=5)"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Our forest consists of 100 trees with a max depth of 5\n",
    "RF = RandomForestRegressor(n_estimators=100, max_depth=5)\n",
    "RF.fit(x_train,y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3.Testing "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To understand the performance of this model, we will get the mean square error and R^2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE: 0.7261098238901945\n",
      "R^2 0.8860726913621104\n"
     ]
    }
   ],
   "source": [
    "# Get predictions\n",
    "y_pred = RF.predict(x_test)\n",
    "\n",
    "# Compute RMSE\n",
    "print(\"RMSE:\",np.sqrt(metrics.mean_squared_error(y_test,y_pred)))\n",
    "print(\"R^2\",metrics.r2_score(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Then we will evaluate it against the expected output"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Create figure and axes\n",
    "f, ax = plt.subplots(figsize=(5,5))\n",
    "\n",
    "# Plot on axes\n",
    "ax.set_title('Actual vs Predicted')\n",
    "ax.set_xlabel('Actual')\n",
    "ax.set_ylabel('Predicted')\n",
    "ax.scatter(y_test,y_pred)\n",
    "ax.plot(y_test,y_test,'r')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4. Adjusting Hyperparameters "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The standard hyperpameter values will not give the best model performance. For that reason, adjustments to the model hyperparameters must take place to get the best performing model."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Creating different hyperparameters of the \"n_estimators\" to compare the results:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 2 trees in forest\n",
    "forest_1 = RandomForestRegressor(n_estimators=2, max_depth=5, random_state=23)\n",
    "\n",
    "# 20 trees in forest\n",
    "forest_2 = RandomForestRegressor(n_estimators=20, max_depth=5, random_state=23)\n",
    "\n",
    "# 100 trees in forest\n",
    "forest_3 = RandomForestRegressor(n_estimators=100, max_depth=5, random_state=23)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Training the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(max_depth=5, n_estimators=2, random_state=23)"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "forest_1.fit(x_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(max_depth=5, n_estimators=20, random_state=23)"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "forest_2.fit(x_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(max_depth=5, random_state=23)"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "forest_3.fit(x_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Evaluating the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Create figure and axes\n",
    "f, ax = plt.subplots(figsize=(15,5), nrows=1, ncols=3,sharey=True)\n",
    "\n",
    "# Create list of titles and predictions to use in for loop\n",
    "pred = [forest_1.predict(x_test), forest_2.predict(x_test), forest_3.predict(x_test)]\n",
    "title = ['trees = 2', 'trees = 20', 'trees = 100']\n",
    "\n",
    "# Loop through all axes to plot each model's results \n",
    "for i in range(3):\n",
    "    rmse = round(np.sqrt(metrics.mean_squared_error(pred[i],y_test)))\n",
    "    ax[i].set_title(title[i]+\"  (RMSE: \"+str(rmse)+ \")\")\n",
    "    ax[i].set_xlabel('Actual')\n",
    "    ax[i].set_ylabel('Predicted')\n",
    "    ax[i].plot(y_test,y_test,'r')\n",
    "    ax[i].scatter(y_test,pred[i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.3: Simple modelling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "X=train_dummies.drop('avg_price_per_kg',axis=1)\n",
    "y=train_dummies['avg_price_per_kg']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split train dataset into train and test sets\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, \n",
    "                                                    y, \n",
    "                                                    test_size=0.20,\n",
    "                                                    random_state=1,\n",
    "                                                    shuffle=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearRegression()"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Linear Regression\n",
    "linreg = LinearRegression()\n",
    "linreg.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Ridge()"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Ridge Regression\n",
    "ridge = Ridge()\n",
    "ridge.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Lasso(alpha=0.01)"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Lasso Reg\n",
    "lasso = Lasso(alpha=0.01)\n",
    "lasso.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVR()"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Support Vector model\n",
    "svr = SVR()\n",
    "svr.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GradientBoostingRegressor()"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Gradient Boosting model\n",
    "params = {}\n",
    "gbr=GradientBoostingRegressor()\n",
    "gbr.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsRegressor(n_neighbors=3)"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#KNN model\n",
    "knn= KNeighborsRegressor(n_neighbors = 3)\n",
    "knn.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeRegressor()"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Decision Tree\n",
    "decision_tree = DecisionTreeRegressor()\n",
    "decision_tree.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor()"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Random Forest\n",
    "random_forest = RandomForestRegressor(n_estimators=100)\n",
    "random_forest.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Compare models\n",
    "\n",
    "# Make training set predictions for each model\n",
    "train_lin = linreg.predict(X_train)\n",
    "train_ridge = ridge.predict(X_train)\n",
    "train_lasso = lasso.predict(X_train)\n",
    "train_Knn = knn.predict(X_train)\n",
    "train_svr = svr.predict(X_train)\n",
    "train_gbr = gbr.predict(X_train)\n",
    "train_DcT = decision_tree.predict(X_train)\n",
    "train_RnF = random_forest.predict(X_train)\n",
    "\n",
    "# Make test set predictions for each model\n",
    "test_lin = linreg.predict(X_test)\n",
    "test_ridge = ridge.predict(X_test)\n",
    "test_lasso = lasso.predict(X_test)\n",
    "test_Knn = knn.predict(X_test)\n",
    "test_svr = svr.predict(X_test)\n",
    "test_gbr = gbr.predict(X_test)\n",
    "test_DcT = decision_tree.predict(X_test)\n",
    "test_RnF = random_forest.predict(X_test)\n",
    "\n",
    "# Dictionary of results\n",
    "results_dict = {'Training MSE':{\n",
    "                        \"Least Squares\":metrics.mean_squared_error(y_train, train_lin,squared=False),\n",
    "                        \"Ridge\": metrics.mean_squared_error(y_train, train_ridge,squared=False),\n",
    "                        \"LASSO\": metrics.mean_squared_error(y_train, train_lasso,squared=False),\n",
    "                        \"SVR\": metrics.mean_squared_error(y_train, train_svr,squared=False),\n",
    "                        \"GBR\": metrics.mean_squared_error(y_train, train_gbr,squared=False),\n",
    "                        \"KNN\": metrics.mean_squared_error(y_train, train_Knn,squared=False),\n",
    "                        \"Decision Tree\": metrics.mean_squared_error(y_train, train_DcT,squared=False),\n",
    "                        \"Random Forest\": metrics.mean_squared_error(y_train, train_RnF,squared=False)\n",
    "                },\n",
    "                'Test MSE':{\n",
    "                        \"Least Squares\": metrics.mean_squared_error(y_test, test_lin,squared=False),\n",
    "                        \"Ridge\": metrics.mean_squared_error(y_test, test_ridge,squared=False),\n",
    "                        \"LASSO\": metrics.mean_squared_error(y_test, test_lasso,squared=False),\n",
    "                        \"SVR\": metrics.mean_squared_error(y_test, test_svr,squared=False),\n",
    "                        \"GBR\": metrics.mean_squared_error(y_test, test_gbr,squared=False),\n",
    "                        \"KNN\": metrics.mean_squared_error(y_test, test_Knn,squared=False),\n",
    "                        \"Decision Tree\": metrics.mean_squared_error(y_test, test_DcT,squared=False),\n",
    "                        \"Random Forest\": metrics.mean_squared_error(y_test, test_RnF,squared=False)\n",
    "                }\n",
    "                }\n",
    "results=pd.DataFrame(results_dict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "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>Training MSE</th>\n",
       "      <th>Test MSE</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Least Squares</th>\n",
       "      <td>6.460531e-01</td>\n",
       "      <td>0.716214</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ridge</th>\n",
       "      <td>6.462297e-01</td>\n",
       "      <td>0.717136</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LASSO</th>\n",
       "      <td>6.677167e-01</td>\n",
       "      <td>0.739192</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SVR</th>\n",
       "      <td>5.329361e-01</td>\n",
       "      <td>0.621075</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GBR</th>\n",
       "      <td>3.926035e-01</td>\n",
       "      <td>0.578749</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KNN</th>\n",
       "      <td>6.120626e-01</td>\n",
       "      <td>0.970947</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Decision Tree</th>\n",
       "      <td>1.096238e-15</td>\n",
       "      <td>0.700207</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Random Forest</th>\n",
       "      <td>2.061689e-01</td>\n",
       "      <td>0.536327</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               Training MSE  Test MSE\n",
       "Least Squares  6.460531e-01  0.716214\n",
       "Ridge          6.462297e-01  0.717136\n",
       "LASSO          6.677167e-01  0.739192\n",
       "SVR            5.329361e-01  0.621075\n",
       "GBR            3.926035e-01  0.578749\n",
       "KNN            6.120626e-01  0.970947\n",
       "Decision Tree  1.096238e-15  0.700207\n",
       "Random Forest  2.061689e-01  0.536327"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 1644 entries, 1 to 64310\n",
      "Data columns (total 32 columns):\n",
      " #   Column                         Non-Null Count  Dtype  \n",
      "---  ------                         --------------  -----  \n",
      " 0   Weight_Kg                      1644 non-null   float64\n",
      " 1   Low_Price                      1644 non-null   float64\n",
      " 2   High_Price                     1644 non-null   float64\n",
      " 3   Sales_Total                    1644 non-null   float64\n",
      " 4   Total_Qty_Sold                 1644 non-null   float64\n",
      " 5   Total_Kg_Sold                  1644 non-null   float64\n",
      " 6   Stock_On_Hand                  1644 non-null   float64\n",
      " 7   Province_EASTERN CAPE          1644 non-null   uint8  \n",
      " 8   Province_NATAL                 1644 non-null   uint8  \n",
      " 9   Province_ORANGE FREE STATE     1644 non-null   uint8  \n",
      " 10  Province_TRANSVAAL             1644 non-null   uint8  \n",
      " 11  Province_W.CAPE-BERGRIVER ETC  1644 non-null   uint8  \n",
      " 12  Province_WEST COAST            1644 non-null   uint8  \n",
      " 13  Container_DT063                1644 non-null   uint8  \n",
      " 14  Container_EC120                1644 non-null   uint8  \n",
      " 15  Container_EF120                1644 non-null   uint8  \n",
      " 16  Container_EG140                1644 non-null   uint8  \n",
      " 17  Container_JE090                1644 non-null   uint8  \n",
      " 18  Container_JG110                1644 non-null   uint8  \n",
      " 19  Container_M4183                1644 non-null   uint8  \n",
      " 20  Container_M6125                1644 non-null   uint8  \n",
      " 21  Container_M9125                1644 non-null   uint8  \n",
      " 22  Size_Grade_1M                  1644 non-null   uint8  \n",
      " 23  Size_Grade_1S                  1644 non-null   uint8  \n",
      " 24  Size_Grade_1U                  1644 non-null   uint8  \n",
      " 25  Size_Grade_1X                  1644 non-null   uint8  \n",
      " 26  Size_Grade_2L                  1644 non-null   uint8  \n",
      " 27  Size_Grade_2M                  1644 non-null   uint8  \n",
      " 28  Size_Grade_2S                  1644 non-null   uint8  \n",
      " 29  Size_Grade_2U                  1644 non-null   uint8  \n",
      " 30  Size_Grade_2X                  1644 non-null   uint8  \n",
      " 31  Month                          1644 non-null   int64  \n",
      "dtypes: float64(7), int64(1), uint8(24)\n",
      "memory usage: 218.7 KB\n"
     ]
    }
   ],
   "source": [
    "train_dummies.drop('avg_price_per_kg',axis=1).info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 685 entries, 0 to 684\n",
      "Data columns (total 33 columns):\n",
      " #   Column                         Non-Null Count  Dtype  \n",
      "---  ------                         --------------  -----  \n",
      " 0   Weight_Kg                      685 non-null    float64\n",
      " 1   Low_Price                      685 non-null    float64\n",
      " 2   High_Price                     685 non-null    float64\n",
      " 3   Sales_Total                    685 non-null    float64\n",
      " 4   Total_Qty_Sold                 685 non-null    int64  \n",
      " 5   Total_Kg_Sold                  685 non-null    float64\n",
      " 6   Stock_On_Hand                  685 non-null    int64  \n",
      " 7   Province_EASTERN CAPE          685 non-null    uint8  \n",
      " 8   Province_NATAL                 685 non-null    uint8  \n",
      " 9   Province_ORANGE FREE STATE     685 non-null    uint8  \n",
      " 10  Province_TRANSVAAL             685 non-null    uint8  \n",
      " 11  Province_W.CAPE-BERGRIVER ETC  685 non-null    uint8  \n",
      " 12  Province_WEST COAST            685 non-null    uint8  \n",
      " 13  Container_DT063                685 non-null    uint8  \n",
      " 14  Container_EC120                685 non-null    uint8  \n",
      " 15  Container_EF120                685 non-null    uint8  \n",
      " 16  Container_EG140                685 non-null    uint8  \n",
      " 17  Container_IA400                685 non-null    uint8  \n",
      " 18  Container_JE090                685 non-null    uint8  \n",
      " 19  Container_JG110                685 non-null    uint8  \n",
      " 20  Container_M4183                685 non-null    uint8  \n",
      " 21  Container_M6125                685 non-null    uint8  \n",
      " 22  Container_M9125                685 non-null    uint8  \n",
      " 23  Size_Grade_1M                  685 non-null    uint8  \n",
      " 24  Size_Grade_1S                  685 non-null    uint8  \n",
      " 25  Size_Grade_1U                  685 non-null    uint8  \n",
      " 26  Size_Grade_1X                  685 non-null    uint8  \n",
      " 27  Size_Grade_2L                  685 non-null    uint8  \n",
      " 28  Size_Grade_2M                  685 non-null    uint8  \n",
      " 29  Size_Grade_2S                  685 non-null    uint8  \n",
      " 30  Size_Grade_2U                  685 non-null    uint8  \n",
      " 31  Size_Grade_2X                  685 non-null    uint8  \n",
      " 32  Month                          685 non-null    int64  \n",
      "dtypes: float64(5), int64(3), uint8(25)\n",
      "memory usage: 64.9 KB\n"
     ]
    }
   ],
   "source": [
    "test_dummies.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The test set contains an extra container 'IA400', we drop this before adding the test_dummies to the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_dummies=test_dummies.drop('Container_IA400',axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "# create submission dataframe\n",
    "# Create Dataframe of Order_No and Time from Pickup to Arrival\n",
    "submission = pd.DataFrame(\n",
    "    {'Index': test_id,\n",
    "     'avg_price_per_kg': random_forest.predict(test_dummies)\n",
    "    })"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# save DataFrame to csv file for submission\n",
    "submission.to_csv(\"AnswerDcT.csv\", index=False)"
   ]
  }
 ],
 "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}