Task 3.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Cognizant - Artificial Intelligence Virtual Experience Program"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Task 3 - Model Building and Interpretation\n",
    "Building a predictive model and interpreting the results back to the business.\n",
    "\n",
    "---------------------------------------------------------------\n",
    "The client has provided 3 datasets, it is now your job to combine, transform and model these datasets in a suitable way to answer the problem statement that the business has requested. \n",
    "\n",
    "Most importantly, once the modeling process is complete, we need you to communicate your work and analysis in the form of a single PowerPoint slide, so that we can present the results back to the business. The key here is to use business-friendly language and to explain your results in a way that the business will understand. For example, ensure that when you’re summarizing the performance of the results you don’t use technical metrics, but rather convert it into numbers that they’ll understand. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing Libraries\n",
    "\n",
    "import pandas as pd\n",
    "from datetime import datetime\n",
    "import sklearn\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>transaction_id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>product_id</th>\n",
       "      <th>category</th>\n",
       "      <th>customer_type</th>\n",
       "      <th>unit_price</th>\n",
       "      <th>quantity</th>\n",
       "      <th>total</th>\n",
       "      <th>payment_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>a1c82654-c52c-45b3-8ce8-4c2a1efe63ed</td>\n",
       "      <td>2022-03-02 09:51:38</td>\n",
       "      <td>3bc6c1ea-0198-46de-9ffd-514ae3338713</td>\n",
       "      <td>fruit</td>\n",
       "      <td>gold</td>\n",
       "      <td>3.99</td>\n",
       "      <td>2</td>\n",
       "      <td>7.98</td>\n",
       "      <td>e-wallet</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>931ad550-09e8-4da6-beaa-8c9d17be9c60</td>\n",
       "      <td>2022-03-06 10:33:59</td>\n",
       "      <td>ad81b46c-bf38-41cf-9b54-5fe7f5eba93e</td>\n",
       "      <td>fruit</td>\n",
       "      <td>standard</td>\n",
       "      <td>3.99</td>\n",
       "      <td>1</td>\n",
       "      <td>3.99</td>\n",
       "      <td>e-wallet</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>ae133534-6f61-4cd6-b6b8-d1c1d8d90aea</td>\n",
       "      <td>2022-03-04 17:20:21</td>\n",
       "      <td>7c55cbd4-f306-4c04-a030-628cbe7867c1</td>\n",
       "      <td>fruit</td>\n",
       "      <td>premium</td>\n",
       "      <td>0.19</td>\n",
       "      <td>2</td>\n",
       "      <td>0.38</td>\n",
       "      <td>e-wallet</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>157cebd9-aaf0-475d-8a11-7c8e0f5b76e4</td>\n",
       "      <td>2022-03-02 17:23:58</td>\n",
       "      <td>80da8348-1707-403f-8be7-9e6deeccc883</td>\n",
       "      <td>fruit</td>\n",
       "      <td>gold</td>\n",
       "      <td>0.19</td>\n",
       "      <td>4</td>\n",
       "      <td>0.76</td>\n",
       "      <td>e-wallet</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>a81a6cd3-5e0c-44a2-826c-aea43e46c514</td>\n",
       "      <td>2022-03-05 14:32:43</td>\n",
       "      <td>7f5e86e6-f06f-45f6-bf44-27b095c9ad1d</td>\n",
       "      <td>fruit</td>\n",
       "      <td>basic</td>\n",
       "      <td>4.49</td>\n",
       "      <td>2</td>\n",
       "      <td>8.98</td>\n",
       "      <td>debit card</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0                        transaction_id            timestamp  \\\n",
       "0           0  a1c82654-c52c-45b3-8ce8-4c2a1efe63ed  2022-03-02 09:51:38   \n",
       "1           1  931ad550-09e8-4da6-beaa-8c9d17be9c60  2022-03-06 10:33:59   \n",
       "2           2  ae133534-6f61-4cd6-b6b8-d1c1d8d90aea  2022-03-04 17:20:21   \n",
       "3           3  157cebd9-aaf0-475d-8a11-7c8e0f5b76e4  2022-03-02 17:23:58   \n",
       "4           4  a81a6cd3-5e0c-44a2-826c-aea43e46c514  2022-03-05 14:32:43   \n",
       "\n",
       "                             product_id category customer_type  unit_price  \\\n",
       "0  3bc6c1ea-0198-46de-9ffd-514ae3338713    fruit          gold        3.99   \n",
       "1  ad81b46c-bf38-41cf-9b54-5fe7f5eba93e    fruit      standard        3.99   \n",
       "2  7c55cbd4-f306-4c04-a030-628cbe7867c1    fruit       premium        0.19   \n",
       "3  80da8348-1707-403f-8be7-9e6deeccc883    fruit          gold        0.19   \n",
       "4  7f5e86e6-f06f-45f6-bf44-27b095c9ad1d    fruit         basic        4.49   \n",
       "\n",
       "   quantity  total payment_type  \n",
       "0         2   7.98     e-wallet  \n",
       "1         1   3.99     e-wallet  \n",
       "2         2   0.38     e-wallet  \n",
       "3         4   0.76     e-wallet  \n",
       "4         2   8.98   debit card  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Importing Datasets\n",
    "\n",
    "sales = pd.read_csv('sales.csv')\n",
    "sales.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>product_id</th>\n",
       "      <th>estimated_stock_pct</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>4220e505-c247-478d-9831-6b9f87a4488a</td>\n",
       "      <td>2022-03-07 12:13:02</td>\n",
       "      <td>f658605e-75f3-4fed-a655-c0903f344427</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>f2612b26-fc82-49ea-8940-0751fdd4d9ef</td>\n",
       "      <td>2022-03-07 16:39:46</td>\n",
       "      <td>de06083a-f5c0-451d-b2f4-9ab88b52609d</td>\n",
       "      <td>0.48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>989a287f-67e6-4478-aa49-c3a35dac0e2e</td>\n",
       "      <td>2022-03-01 18:17:43</td>\n",
       "      <td>ce8f3a04-d1a4-43b1-a7c2-fa1b8e7674c8</td>\n",
       "      <td>0.58</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>af8e5683-d247-46ac-9909-1a77bdebefb2</td>\n",
       "      <td>2022-03-02 14:29:09</td>\n",
       "      <td>c21e3ba9-92a3-4745-92c2-6faef73223f7</td>\n",
       "      <td>0.79</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>08a32247-3f44-4002-85fb-c198434dd4bb</td>\n",
       "      <td>2022-03-02 13:46:18</td>\n",
       "      <td>7f478817-aa5b-44e9-9059-8045228c9eb0</td>\n",
       "      <td>0.22</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0                                    id            timestamp  \\\n",
       "0           0  4220e505-c247-478d-9831-6b9f87a4488a  2022-03-07 12:13:02   \n",
       "1           1  f2612b26-fc82-49ea-8940-0751fdd4d9ef  2022-03-07 16:39:46   \n",
       "2           2  989a287f-67e6-4478-aa49-c3a35dac0e2e  2022-03-01 18:17:43   \n",
       "3           3  af8e5683-d247-46ac-9909-1a77bdebefb2  2022-03-02 14:29:09   \n",
       "4           4  08a32247-3f44-4002-85fb-c198434dd4bb  2022-03-02 13:46:18   \n",
       "\n",
       "                             product_id  estimated_stock_pct  \n",
       "0  f658605e-75f3-4fed-a655-c0903f344427                 0.75  \n",
       "1  de06083a-f5c0-451d-b2f4-9ab88b52609d                 0.48  \n",
       "2  ce8f3a04-d1a4-43b1-a7c2-fa1b8e7674c8                 0.58  \n",
       "3  c21e3ba9-92a3-4745-92c2-6faef73223f7                 0.79  \n",
       "4  7f478817-aa5b-44e9-9059-8045228c9eb0                 0.22  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "stock = pd.read_csv('sensor_stock_levels.csv')\n",
    "stock.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>temperature</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>d1ca1ef8-0eac-42fc-af80-97106efc7b13</td>\n",
       "      <td>2022-03-07 15:55:20</td>\n",
       "      <td>2.96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>4b8a66c4-0f3a-4f16-826f-8cf9397e9d18</td>\n",
       "      <td>2022-03-01 09:18:22</td>\n",
       "      <td>1.88</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>3d47a0c7-1e72-4512-812f-b6b5d8428cf3</td>\n",
       "      <td>2022-03-04 15:12:26</td>\n",
       "      <td>1.78</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>9500357b-ce15-424a-837a-7677b386f471</td>\n",
       "      <td>2022-03-02 12:30:42</td>\n",
       "      <td>2.18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>c4b61fec-99c2-4c6d-8e5d-4edd8c9632fa</td>\n",
       "      <td>2022-03-05 09:09:33</td>\n",
       "      <td>1.38</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0                                    id            timestamp  \\\n",
       "0           0  d1ca1ef8-0eac-42fc-af80-97106efc7b13  2022-03-07 15:55:20   \n",
       "1           1  4b8a66c4-0f3a-4f16-826f-8cf9397e9d18  2022-03-01 09:18:22   \n",
       "2           2  3d47a0c7-1e72-4512-812f-b6b5d8428cf3  2022-03-04 15:12:26   \n",
       "3           3  9500357b-ce15-424a-837a-7677b386f471  2022-03-02 12:30:42   \n",
       "4           4  c4b61fec-99c2-4c6d-8e5d-4edd8c9632fa  2022-03-05 09:09:33   \n",
       "\n",
       "   temperature  \n",
       "0         2.96  \n",
       "1         1.88  \n",
       "2         1.78  \n",
       "3         2.18  \n",
       "4         1.38  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "storage = pd.read_csv('sensor_storage_temperature.csv')\n",
    "storage.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 7829 entries, 0 to 7828\n",
      "Data columns (total 10 columns):\n",
      " #   Column          Non-Null Count  Dtype  \n",
      "---  ------          --------------  -----  \n",
      " 0   Unnamed: 0      7829 non-null   int64  \n",
      " 1   transaction_id  7829 non-null   object \n",
      " 2   timestamp       7829 non-null   object \n",
      " 3   product_id      7829 non-null   object \n",
      " 4   category        7829 non-null   object \n",
      " 5   customer_type   7829 non-null   object \n",
      " 6   unit_price      7829 non-null   float64\n",
      " 7   quantity        7829 non-null   int64  \n",
      " 8   total           7829 non-null   float64\n",
      " 9   payment_type    7829 non-null   object \n",
      "dtypes: float64(2), int64(2), object(6)\n",
      "memory usage: 611.8+ KB\n"
     ]
    }
   ],
   "source": [
    "# Check dataset info\n",
    "\n",
    "sales.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 15000 entries, 0 to 14999\n",
      "Data columns (total 5 columns):\n",
      " #   Column               Non-Null Count  Dtype  \n",
      "---  ------               --------------  -----  \n",
      " 0   Unnamed: 0           15000 non-null  int64  \n",
      " 1   id                   15000 non-null  object \n",
      " 2   timestamp            15000 non-null  object \n",
      " 3   product_id           15000 non-null  object \n",
      " 4   estimated_stock_pct  15000 non-null  float64\n",
      "dtypes: float64(1), int64(1), object(3)\n",
      "memory usage: 586.1+ KB\n"
     ]
    }
   ],
   "source": [
    "stock.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 23890 entries, 0 to 23889\n",
      "Data columns (total 4 columns):\n",
      " #   Column       Non-Null Count  Dtype  \n",
      "---  ------       --------------  -----  \n",
      " 0   Unnamed: 0   23890 non-null  int64  \n",
      " 1   id           23890 non-null  object \n",
      " 2   timestamp    23890 non-null  object \n",
      " 3   temperature  23890 non-null  float64\n",
      "dtypes: float64(1), int64(1), object(2)\n",
      "memory usage: 746.7+ KB\n"
     ]
    }
   ],
   "source": [
    "storage.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Date data format conversion\n",
    "\n",
    "sales['timestamp'] = pd.to_datetime(sales['timestamp'], format='%Y-%m-%d %H:%M:%S')\n",
    "stock['timestamp'] = pd.to_datetime(stock['timestamp'], format='%Y-%m-%d %H:%M:%S')\n",
    "storage['timestamp'] = pd.to_datetime(storage['timestamp'], format='%Y-%m-%d %H:%M:%S')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Function to covert timestamps to hourly\n",
    "\n",
    "def hourly(df_name, col_name):\n",
    "    \n",
    "    timelist = []\n",
    "    data = df_name[col_name].tolist()\n",
    "    \n",
    "    for each in data:\n",
    "        each = each.strftime('%Y-%m-%d %H:00:00')\n",
    "        timelist.append(each)\n",
    "        \n",
    "    df_name['timestamp'] = timelist"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>transaction_id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>product_id</th>\n",
       "      <th>category</th>\n",
       "      <th>customer_type</th>\n",
       "      <th>unit_price</th>\n",
       "      <th>quantity</th>\n",
       "      <th>total</th>\n",
       "      <th>payment_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>a1c82654-c52c-45b3-8ce8-4c2a1efe63ed</td>\n",
       "      <td>2022-03-02 09:00:00</td>\n",
       "      <td>3bc6c1ea-0198-46de-9ffd-514ae3338713</td>\n",
       "      <td>fruit</td>\n",
       "      <td>gold</td>\n",
       "      <td>3.99</td>\n",
       "      <td>2</td>\n",
       "      <td>7.98</td>\n",
       "      <td>e-wallet</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>931ad550-09e8-4da6-beaa-8c9d17be9c60</td>\n",
       "      <td>2022-03-06 10:00:00</td>\n",
       "      <td>ad81b46c-bf38-41cf-9b54-5fe7f5eba93e</td>\n",
       "      <td>fruit</td>\n",
       "      <td>standard</td>\n",
       "      <td>3.99</td>\n",
       "      <td>1</td>\n",
       "      <td>3.99</td>\n",
       "      <td>e-wallet</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>ae133534-6f61-4cd6-b6b8-d1c1d8d90aea</td>\n",
       "      <td>2022-03-04 17:00:00</td>\n",
       "      <td>7c55cbd4-f306-4c04-a030-628cbe7867c1</td>\n",
       "      <td>fruit</td>\n",
       "      <td>premium</td>\n",
       "      <td>0.19</td>\n",
       "      <td>2</td>\n",
       "      <td>0.38</td>\n",
       "      <td>e-wallet</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0                        transaction_id            timestamp  \\\n",
       "0           0  a1c82654-c52c-45b3-8ce8-4c2a1efe63ed  2022-03-02 09:00:00   \n",
       "1           1  931ad550-09e8-4da6-beaa-8c9d17be9c60  2022-03-06 10:00:00   \n",
       "2           2  ae133534-6f61-4cd6-b6b8-d1c1d8d90aea  2022-03-04 17:00:00   \n",
       "\n",
       "                             product_id category customer_type  unit_price  \\\n",
       "0  3bc6c1ea-0198-46de-9ffd-514ae3338713    fruit          gold        3.99   \n",
       "1  ad81b46c-bf38-41cf-9b54-5fe7f5eba93e    fruit      standard        3.99   \n",
       "2  7c55cbd4-f306-4c04-a030-628cbe7867c1    fruit       premium        0.19   \n",
       "\n",
       "   quantity  total payment_type  \n",
       "0         2   7.98     e-wallet  \n",
       "1         1   3.99     e-wallet  \n",
       "2         2   0.38     e-wallet  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hourly(sales, 'timestamp')\n",
    "sales.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>product_id</th>\n",
       "      <th>estimated_stock_pct</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>4220e505-c247-478d-9831-6b9f87a4488a</td>\n",
       "      <td>2022-03-07 12:00:00</td>\n",
       "      <td>f658605e-75f3-4fed-a655-c0903f344427</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>f2612b26-fc82-49ea-8940-0751fdd4d9ef</td>\n",
       "      <td>2022-03-07 16:00:00</td>\n",
       "      <td>de06083a-f5c0-451d-b2f4-9ab88b52609d</td>\n",
       "      <td>0.48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>989a287f-67e6-4478-aa49-c3a35dac0e2e</td>\n",
       "      <td>2022-03-01 18:00:00</td>\n",
       "      <td>ce8f3a04-d1a4-43b1-a7c2-fa1b8e7674c8</td>\n",
       "      <td>0.58</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0                                    id            timestamp  \\\n",
       "0           0  4220e505-c247-478d-9831-6b9f87a4488a  2022-03-07 12:00:00   \n",
       "1           1  f2612b26-fc82-49ea-8940-0751fdd4d9ef  2022-03-07 16:00:00   \n",
       "2           2  989a287f-67e6-4478-aa49-c3a35dac0e2e  2022-03-01 18:00:00   \n",
       "\n",
       "                             product_id  estimated_stock_pct  \n",
       "0  f658605e-75f3-4fed-a655-c0903f344427                 0.75  \n",
       "1  de06083a-f5c0-451d-b2f4-9ab88b52609d                 0.48  \n",
       "2  ce8f3a04-d1a4-43b1-a7c2-fa1b8e7674c8                 0.58  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hourly(stock, 'timestamp')\n",
    "stock.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>temperature</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>d1ca1ef8-0eac-42fc-af80-97106efc7b13</td>\n",
       "      <td>2022-03-07 15:00:00</td>\n",
       "      <td>2.96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>4b8a66c4-0f3a-4f16-826f-8cf9397e9d18</td>\n",
       "      <td>2022-03-01 09:00:00</td>\n",
       "      <td>1.88</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>3d47a0c7-1e72-4512-812f-b6b5d8428cf3</td>\n",
       "      <td>2022-03-04 15:00:00</td>\n",
       "      <td>1.78</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0                                    id            timestamp  \\\n",
       "0           0  d1ca1ef8-0eac-42fc-af80-97106efc7b13  2022-03-07 15:00:00   \n",
       "1           1  4b8a66c4-0f3a-4f16-826f-8cf9397e9d18  2022-03-01 09:00:00   \n",
       "2           2  3d47a0c7-1e72-4512-812f-b6b5d8428cf3  2022-03-04 15:00:00   \n",
       "\n",
       "   temperature  \n",
       "0         2.96  \n",
       "1         1.88  \n",
       "2         1.78  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hourly(storage, 'timestamp')\n",
    "storage.head(3)"
   ]
  },
  {
   "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>timestamp</th>\n",
       "      <th>product_id</th>\n",
       "      <th>quantity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2022-03-01 09:00:00</td>\n",
       "      <td>00e120bb-89d6-4df5-bc48-a051148e3d03</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2022-03-01 09:00:00</td>\n",
       "      <td>01f3cdd9-8e9e-4dff-9b5c-69698a0388d0</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2022-03-01 09:00:00</td>\n",
       "      <td>03a2557a-aa12-4add-a6d4-77dc36342067</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             timestamp                            product_id  quantity\n",
       "0  2022-03-01 09:00:00  00e120bb-89d6-4df5-bc48-a051148e3d03         3\n",
       "1  2022-03-01 09:00:00  01f3cdd9-8e9e-4dff-9b5c-69698a0388d0         3\n",
       "2  2022-03-01 09:00:00  03a2557a-aa12-4add-a6d4-77dc36342067         3"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Save aggregate values for timestamps\n",
    "\n",
    "sales_agg = sales.groupby(['timestamp', 'product_id']).agg({'quantity': 'sum'}).reset_index()\n",
    "sales_agg.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>timestamp</th>\n",
       "      <th>product_id</th>\n",
       "      <th>estimated_stock_pct</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2022-03-01 09:00:00</td>\n",
       "      <td>00e120bb-89d6-4df5-bc48-a051148e3d03</td>\n",
       "      <td>0.89</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2022-03-01 09:00:00</td>\n",
       "      <td>01f3cdd9-8e9e-4dff-9b5c-69698a0388d0</td>\n",
       "      <td>0.14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2022-03-01 09:00:00</td>\n",
       "      <td>01ff0803-ae73-4234-971d-5713c97b7f4b</td>\n",
       "      <td>0.67</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             timestamp                            product_id  \\\n",
       "0  2022-03-01 09:00:00  00e120bb-89d6-4df5-bc48-a051148e3d03   \n",
       "1  2022-03-01 09:00:00  01f3cdd9-8e9e-4dff-9b5c-69698a0388d0   \n",
       "2  2022-03-01 09:00:00  01ff0803-ae73-4234-971d-5713c97b7f4b   \n",
       "\n",
       "   estimated_stock_pct  \n",
       "0                 0.89  \n",
       "1                 0.14  \n",
       "2                 0.67  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "stock_agg = stock.groupby(['timestamp', 'product_id']).agg({'estimated_stock_pct': 'mean'}).reset_index()\n",
    "stock_agg.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>timestamp</th>\n",
       "      <th>temperature</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2022-03-01 09:00:00</td>\n",
       "      <td>-0.028850</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2022-03-01 10:00:00</td>\n",
       "      <td>1.284314</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2022-03-01 11:00:00</td>\n",
       "      <td>-0.560000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             timestamp  temperature\n",
       "0  2022-03-01 09:00:00    -0.028850\n",
       "1  2022-03-01 10:00:00     1.284314\n",
       "2  2022-03-01 11:00:00    -0.560000"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp_agg = storage.groupby(['timestamp']).agg({'temperature': 'mean'}).reset_index()\n",
    "temp_agg.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "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>timestamp</th>\n",
       "      <th>product_id</th>\n",
       "      <th>estimated_stock_pct</th>\n",
       "      <th>quantity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2022-03-01 09:00:00</td>\n",
       "      <td>00e120bb-89d6-4df5-bc48-a051148e3d03</td>\n",
       "      <td>0.89</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2022-03-01 09:00:00</td>\n",
       "      <td>01f3cdd9-8e9e-4dff-9b5c-69698a0388d0</td>\n",
       "      <td>0.14</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2022-03-01 09:00:00</td>\n",
       "      <td>01ff0803-ae73-4234-971d-5713c97b7f4b</td>\n",
       "      <td>0.67</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2022-03-01 09:00:00</td>\n",
       "      <td>0363eb21-8c74-47e1-a216-c37e565e5ceb</td>\n",
       "      <td>0.82</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2022-03-01 09:00:00</td>\n",
       "      <td>03f0b20e-3b5b-444f-bc39-cdfa2523d4bc</td>\n",
       "      <td>0.05</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             timestamp                            product_id  \\\n",
       "0  2022-03-01 09:00:00  00e120bb-89d6-4df5-bc48-a051148e3d03   \n",
       "1  2022-03-01 09:00:00  01f3cdd9-8e9e-4dff-9b5c-69698a0388d0   \n",
       "2  2022-03-01 09:00:00  01ff0803-ae73-4234-971d-5713c97b7f4b   \n",
       "3  2022-03-01 09:00:00  0363eb21-8c74-47e1-a216-c37e565e5ceb   \n",
       "4  2022-03-01 09:00:00  03f0b20e-3b5b-444f-bc39-cdfa2523d4bc   \n",
       "\n",
       "   estimated_stock_pct  quantity  \n",
       "0                 0.89       3.0  \n",
       "1                 0.14       3.0  \n",
       "2                 0.67       NaN  \n",
       "3                 0.82       NaN  \n",
       "4                 0.05       NaN  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Merge datasets together\n",
    "\n",
    "merged = stock_agg.merge(sales_agg, on=['timestamp', 'product_id'], how='left')\n",
    "merged.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>timestamp</th>\n",
       "      <th>product_id</th>\n",
       "      <th>estimated_stock_pct</th>\n",
       "      <th>quantity</th>\n",
       "      <th>temperature</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2022-03-01 09:00:00</td>\n",
       "      <td>00e120bb-89d6-4df5-bc48-a051148e3d03</td>\n",
       "      <td>0.89</td>\n",
       "      <td>3.0</td>\n",
       "      <td>-0.02885</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2022-03-01 09:00:00</td>\n",
       "      <td>01f3cdd9-8e9e-4dff-9b5c-69698a0388d0</td>\n",
       "      <td>0.14</td>\n",
       "      <td>3.0</td>\n",
       "      <td>-0.02885</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2022-03-01 09:00:00</td>\n",
       "      <td>01ff0803-ae73-4234-971d-5713c97b7f4b</td>\n",
       "      <td>0.67</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.02885</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2022-03-01 09:00:00</td>\n",
       "      <td>0363eb21-8c74-47e1-a216-c37e565e5ceb</td>\n",
       "      <td>0.82</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.02885</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2022-03-01 09:00:00</td>\n",
       "      <td>03f0b20e-3b5b-444f-bc39-cdfa2523d4bc</td>\n",
       "      <td>0.05</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.02885</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             timestamp                            product_id  \\\n",
       "0  2022-03-01 09:00:00  00e120bb-89d6-4df5-bc48-a051148e3d03   \n",
       "1  2022-03-01 09:00:00  01f3cdd9-8e9e-4dff-9b5c-69698a0388d0   \n",
       "2  2022-03-01 09:00:00  01ff0803-ae73-4234-971d-5713c97b7f4b   \n",
       "3  2022-03-01 09:00:00  0363eb21-8c74-47e1-a216-c37e565e5ceb   \n",
       "4  2022-03-01 09:00:00  03f0b20e-3b5b-444f-bc39-cdfa2523d4bc   \n",
       "\n",
       "   estimated_stock_pct  quantity  temperature  \n",
       "0                 0.89       3.0     -0.02885  \n",
       "1                 0.14       3.0     -0.02885  \n",
       "2                 0.67       NaN     -0.02885  \n",
       "3                 0.82       NaN     -0.02885  \n",
       "4                 0.05       NaN     -0.02885  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "merged = merged.merge(temp_agg, on='timestamp', how='left')\n",
    "merged.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 10845 entries, 0 to 10844\n",
      "Data columns (total 5 columns):\n",
      " #   Column               Non-Null Count  Dtype  \n",
      "---  ------               --------------  -----  \n",
      " 0   timestamp            10845 non-null  object \n",
      " 1   product_id           10845 non-null  object \n",
      " 2   estimated_stock_pct  10845 non-null  float64\n",
      " 3   quantity             3067 non-null   float64\n",
      " 4   temperature          10845 non-null  float64\n",
      "dtypes: float64(3), object(2)\n",
      "memory usage: 508.4+ KB\n"
     ]
    }
   ],
   "source": [
    "merged.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "timestamp                 0\n",
       "product_id                0\n",
       "estimated_stock_pct       0\n",
       "quantity               7778\n",
       "temperature               0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Finding columns with null values\n",
    "\n",
    "merged.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "timestamp              0\n",
       "product_id             0\n",
       "estimated_stock_pct    0\n",
       "quantity               0\n",
       "temperature            0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Filling in missing values\n",
    "\n",
    "merged['quantity'] = merged['quantity'].fillna(0)\n",
    "merged.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Adding category and unit price columns\n",
    "\n",
    "prod_catg = sales[['product_id', 'category']]\n",
    "prod_price = sales[['product_id', 'unit_price']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Category - 7529\n",
      "Price - 7529\n"
     ]
    }
   ],
   "source": [
    "# Cheking for duplicates\n",
    "\n",
    "print('Category -', prod_catg.duplicated().sum())\n",
    "print('Price -', prod_price.duplicated().sum())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Deleting duplicates\n",
    "\n",
    "prod_catg = prod_catg.drop_duplicates()\n",
    "prod_price = prod_price.drop_duplicates()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Category - 0\n",
      "Price - 0\n"
     ]
    }
   ],
   "source": [
    "# Cheking for duplicates\n",
    "\n",
    "print('Category -', prod_catg.duplicated().sum())\n",
    "print('Price -', prod_price.duplicated().sum())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "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>timestamp</th>\n",
       "      <th>product_id</th>\n",
       "      <th>estimated_stock_pct</th>\n",
       "      <th>quantity</th>\n",
       "      <th>temperature</th>\n",
       "      <th>category</th>\n",
       "      <th>unit_price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2022-03-01 09:00:00</td>\n",
       "      <td>00e120bb-89d6-4df5-bc48-a051148e3d03</td>\n",
       "      <td>0.89</td>\n",
       "      <td>3.0</td>\n",
       "      <td>-0.02885</td>\n",
       "      <td>kitchen</td>\n",
       "      <td>11.19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2022-03-01 09:00:00</td>\n",
       "      <td>01f3cdd9-8e9e-4dff-9b5c-69698a0388d0</td>\n",
       "      <td>0.14</td>\n",
       "      <td>3.0</td>\n",
       "      <td>-0.02885</td>\n",
       "      <td>vegetables</td>\n",
       "      <td>1.49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2022-03-01 09:00:00</td>\n",
       "      <td>01ff0803-ae73-4234-971d-5713c97b7f4b</td>\n",
       "      <td>0.67</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-0.02885</td>\n",
       "      <td>baby products</td>\n",
       "      <td>14.19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2022-03-01 09:00:00</td>\n",
       "      <td>0363eb21-8c74-47e1-a216-c37e565e5ceb</td>\n",
       "      <td>0.82</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-0.02885</td>\n",
       "      <td>beverages</td>\n",
       "      <td>20.19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2022-03-01 09:00:00</td>\n",
       "      <td>03f0b20e-3b5b-444f-bc39-cdfa2523d4bc</td>\n",
       "      <td>0.05</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-0.02885</td>\n",
       "      <td>pets</td>\n",
       "      <td>8.19</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             timestamp                            product_id  \\\n",
       "0  2022-03-01 09:00:00  00e120bb-89d6-4df5-bc48-a051148e3d03   \n",
       "1  2022-03-01 09:00:00  01f3cdd9-8e9e-4dff-9b5c-69698a0388d0   \n",
       "2  2022-03-01 09:00:00  01ff0803-ae73-4234-971d-5713c97b7f4b   \n",
       "3  2022-03-01 09:00:00  0363eb21-8c74-47e1-a216-c37e565e5ceb   \n",
       "4  2022-03-01 09:00:00  03f0b20e-3b5b-444f-bc39-cdfa2523d4bc   \n",
       "\n",
       "   estimated_stock_pct  quantity  temperature       category  unit_price  \n",
       "0                 0.89       3.0     -0.02885        kitchen       11.19  \n",
       "1                 0.14       3.0     -0.02885     vegetables        1.49  \n",
       "2                 0.67       0.0     -0.02885  baby products       14.19  \n",
       "3                 0.82       0.0     -0.02885      beverages       20.19  \n",
       "4                 0.05       0.0     -0.02885           pets        8.19  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Merging into main dataset\n",
    "\n",
    "merged = merged.merge(prod_catg, on=\"product_id\", how=\"left\")\n",
    "merged = merged.merge(prod_price, on=\"product_id\", how=\"left\")\n",
    "merged.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 10845 entries, 0 to 10844\n",
      "Data columns (total 7 columns):\n",
      " #   Column               Non-Null Count  Dtype  \n",
      "---  ------               --------------  -----  \n",
      " 0   timestamp            10845 non-null  object \n",
      " 1   product_id           10845 non-null  object \n",
      " 2   estimated_stock_pct  10845 non-null  float64\n",
      " 3   quantity             10845 non-null  float64\n",
      " 4   temperature          10845 non-null  float64\n",
      " 5   category             10845 non-null  object \n",
      " 6   unit_price           10845 non-null  float64\n",
      "dtypes: float64(4), object(3)\n",
      "memory usage: 677.8+ KB\n"
     ]
    }
   ],
   "source": [
    "merged.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Converting timestamp format\n",
    "\n",
    "merged['timestamp'] = pd.to_datetime(merged.timestamp, format='%Y-%m-%d %H:%M:%S')"
   ]
  },
  {
   "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>product_id</th>\n",
       "      <th>estimated_stock_pct</th>\n",
       "      <th>quantity</th>\n",
       "      <th>temperature</th>\n",
       "      <th>category</th>\n",
       "      <th>unit_price</th>\n",
       "      <th>day_of_month</th>\n",
       "      <th>day_of_week</th>\n",
       "      <th>hour</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>00e120bb-89d6-4df5-bc48-a051148e3d03</td>\n",
       "      <td>0.89</td>\n",
       "      <td>3.0</td>\n",
       "      <td>-0.02885</td>\n",
       "      <td>kitchen</td>\n",
       "      <td>11.19</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>01f3cdd9-8e9e-4dff-9b5c-69698a0388d0</td>\n",
       "      <td>0.14</td>\n",
       "      <td>3.0</td>\n",
       "      <td>-0.02885</td>\n",
       "      <td>vegetables</td>\n",
       "      <td>1.49</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>01ff0803-ae73-4234-971d-5713c97b7f4b</td>\n",
       "      <td>0.67</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-0.02885</td>\n",
       "      <td>baby products</td>\n",
       "      <td>14.19</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0363eb21-8c74-47e1-a216-c37e565e5ceb</td>\n",
       "      <td>0.82</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-0.02885</td>\n",
       "      <td>beverages</td>\n",
       "      <td>20.19</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>03f0b20e-3b5b-444f-bc39-cdfa2523d4bc</td>\n",
       "      <td>0.05</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-0.02885</td>\n",
       "      <td>pets</td>\n",
       "      <td>8.19</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             product_id  estimated_stock_pct  quantity  \\\n",
       "0  00e120bb-89d6-4df5-bc48-a051148e3d03                 0.89       3.0   \n",
       "1  01f3cdd9-8e9e-4dff-9b5c-69698a0388d0                 0.14       3.0   \n",
       "2  01ff0803-ae73-4234-971d-5713c97b7f4b                 0.67       0.0   \n",
       "3  0363eb21-8c74-47e1-a216-c37e565e5ceb                 0.82       0.0   \n",
       "4  03f0b20e-3b5b-444f-bc39-cdfa2523d4bc                 0.05       0.0   \n",
       "\n",
       "   temperature       category  unit_price  day_of_month  day_of_week  hour  \n",
       "0     -0.02885        kitchen       11.19             1            1     9  \n",
       "1     -0.02885     vegetables        1.49             1            1     9  \n",
       "2     -0.02885  baby products       14.19             1            1     9  \n",
       "3     -0.02885      beverages       20.19             1            1     9  \n",
       "4     -0.02885           pets        8.19             1            1     9  "
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Timestamp extraction\n",
    "\n",
    "merged['day_of_month'] = merged['timestamp'].dt.day\n",
    "merged['day_of_week'] = merged['timestamp'].dt.dayofweek\n",
    "merged['hour'] = merged['timestamp'].dt.hour\n",
    "merged.drop(columns=['timestamp'], inplace=True)\n",
    "merged.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>product_id</th>\n",
       "      <th>estimated_stock_pct</th>\n",
       "      <th>quantity</th>\n",
       "      <th>temperature</th>\n",
       "      <th>unit_price</th>\n",
       "      <th>day_of_month</th>\n",
       "      <th>day_of_week</th>\n",
       "      <th>hour</th>\n",
       "      <th>category_baby products</th>\n",
       "      <th>category_baked goods</th>\n",
       "      <th>...</th>\n",
       "      <th>category_meat</th>\n",
       "      <th>category_medicine</th>\n",
       "      <th>category_packaged foods</th>\n",
       "      <th>category_personal care</th>\n",
       "      <th>category_pets</th>\n",
       "      <th>category_refrigerated items</th>\n",
       "      <th>category_seafood</th>\n",
       "      <th>category_snacks</th>\n",
       "      <th>category_spices and herbs</th>\n",
       "      <th>category_vegetables</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>00e120bb-89d6-4df5-bc48-a051148e3d03</td>\n",
       "      <td>0.89</td>\n",
       "      <td>3.0</td>\n",
       "      <td>-0.02885</td>\n",
       "      <td>11.19</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>9</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>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>01f3cdd9-8e9e-4dff-9b5c-69698a0388d0</td>\n",
       "      <td>0.14</td>\n",
       "      <td>3.0</td>\n",
       "      <td>-0.02885</td>\n",
       "      <td>1.49</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>9</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>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>01ff0803-ae73-4234-971d-5713c97b7f4b</td>\n",
       "      <td>0.67</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-0.02885</td>\n",
       "      <td>14.19</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>1</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>3</th>\n",
       "      <td>0363eb21-8c74-47e1-a216-c37e565e5ceb</td>\n",
       "      <td>0.82</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-0.02885</td>\n",
       "      <td>20.19</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>9</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>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>03f0b20e-3b5b-444f-bc39-cdfa2523d4bc</td>\n",
       "      <td>0.05</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-0.02885</td>\n",
       "      <td>8.19</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>9</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>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 30 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                             product_id  estimated_stock_pct  quantity  \\\n",
       "0  00e120bb-89d6-4df5-bc48-a051148e3d03                 0.89       3.0   \n",
       "1  01f3cdd9-8e9e-4dff-9b5c-69698a0388d0                 0.14       3.0   \n",
       "2  01ff0803-ae73-4234-971d-5713c97b7f4b                 0.67       0.0   \n",
       "3  0363eb21-8c74-47e1-a216-c37e565e5ceb                 0.82       0.0   \n",
       "4  03f0b20e-3b5b-444f-bc39-cdfa2523d4bc                 0.05       0.0   \n",
       "\n",
       "   temperature  unit_price  day_of_month  day_of_week  hour  \\\n",
       "0     -0.02885       11.19             1            1     9   \n",
       "1     -0.02885        1.49             1            1     9   \n",
       "2     -0.02885       14.19             1            1     9   \n",
       "3     -0.02885       20.19             1            1     9   \n",
       "4     -0.02885        8.19             1            1     9   \n",
       "\n",
       "   category_baby products  category_baked goods  ...  category_meat  \\\n",
       "0                       0                     0  ...              0   \n",
       "1                       0                     0  ...              0   \n",
       "2                       1                     0  ...              0   \n",
       "3                       0                     0  ...              0   \n",
       "4                       0                     0  ...              0   \n",
       "\n",
       "   category_medicine  category_packaged foods  category_personal care  \\\n",
       "0                  0                        0                       0   \n",
       "1                  0                        0                       0   \n",
       "2                  0                        0                       0   \n",
       "3                  0                        0                       0   \n",
       "4                  0                        0                       0   \n",
       "\n",
       "   category_pets  category_refrigerated items  category_seafood  \\\n",
       "0              0                            0                 0   \n",
       "1              0                            0                 0   \n",
       "2              0                            0                 0   \n",
       "3              0                            0                 0   \n",
       "4              1                            0                 0   \n",
       "\n",
       "   category_snacks  category_spices and herbs  category_vegetables  \n",
       "0                0                          0                    0  \n",
       "1                0                          0                    1  \n",
       "2                0                          0                    0  \n",
       "3                0                          0                    0  \n",
       "4                0                          0                    0  \n",
       "\n",
       "[5 rows x 30 columns]"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Convert categories to numeric variables\n",
    "\n",
    "merged = pd.get_dummies(merged, columns=['category'])\n",
    "merged.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 10845 entries, 0 to 10844\n",
      "Data columns (total 30 columns):\n",
      " #   Column                          Non-Null Count  Dtype  \n",
      "---  ------                          --------------  -----  \n",
      " 0   product_id                      10845 non-null  object \n",
      " 1   estimated_stock_pct             10845 non-null  float64\n",
      " 2   quantity                        10845 non-null  float64\n",
      " 3   temperature                     10845 non-null  float64\n",
      " 4   unit_price                      10845 non-null  float64\n",
      " 5   day_of_month                    10845 non-null  int64  \n",
      " 6   day_of_week                     10845 non-null  int64  \n",
      " 7   hour                            10845 non-null  int64  \n",
      " 8   category_baby products          10845 non-null  uint8  \n",
      " 9   category_baked goods            10845 non-null  uint8  \n",
      " 10  category_baking                 10845 non-null  uint8  \n",
      " 11  category_beverages              10845 non-null  uint8  \n",
      " 12  category_canned foods           10845 non-null  uint8  \n",
      " 13  category_cheese                 10845 non-null  uint8  \n",
      " 14  category_cleaning products      10845 non-null  uint8  \n",
      " 15  category_condiments and sauces  10845 non-null  uint8  \n",
      " 16  category_dairy                  10845 non-null  uint8  \n",
      " 17  category_frozen                 10845 non-null  uint8  \n",
      " 18  category_fruit                  10845 non-null  uint8  \n",
      " 19  category_kitchen                10845 non-null  uint8  \n",
      " 20  category_meat                   10845 non-null  uint8  \n",
      " 21  category_medicine               10845 non-null  uint8  \n",
      " 22  category_packaged foods         10845 non-null  uint8  \n",
      " 23  category_personal care          10845 non-null  uint8  \n",
      " 24  category_pets                   10845 non-null  uint8  \n",
      " 25  category_refrigerated items     10845 non-null  uint8  \n",
      " 26  category_seafood                10845 non-null  uint8  \n",
      " 27  category_snacks                 10845 non-null  uint8  \n",
      " 28  category_spices and herbs       10845 non-null  uint8  \n",
      " 29  category_vegetables             10845 non-null  uint8  \n",
      "dtypes: float64(4), int64(3), object(1), uint8(22)\n",
      "memory usage: 995.5+ KB\n"
     ]
    }
   ],
   "source": [
    "merged.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Removing product id, no longer needed\n",
    "\n",
    "merged.drop(columns='product_id', inplace=True)\n",
    "merged.head(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(10845, 28)\n",
      "(10845,)\n"
     ]
    }
   ],
   "source": [
    "# Creating target variable y and independant variable x\n",
    "\n",
    "X = merged.drop(columns=['estimated_stock_pct'])\n",
    "y = merged['estimated_stock_pct']\n",
    "print(X.shape)\n",
    "print(y.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Setting folds K\n",
    "K = 10\n",
    "\n",
    "# Setting split % to 75% training and 25% teasting\n",
    "split = 0.75"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing functions\n",
    "\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import mean_absolute_error\n",
    "from sklearn.preprocessing import StandardScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fold 1: MAE = 0.236\n",
      "Fold 2: MAE = 0.237\n",
      "Fold 3: MAE = 0.236\n",
      "Fold 4: MAE = 0.236\n",
      "Fold 5: MAE = 0.237\n",
      "Fold 6: MAE = 0.237\n",
      "Fold 7: MAE = 0.236\n",
      "Fold 8: MAE = 0.237\n",
      "Fold 9: MAE = 0.236\n",
      "Fold 10: MAE = 0.235\n",
      "Average MAE: 0.24\n"
     ]
    }
   ],
   "source": [
    "accuracy = []\n",
    "\n",
    "for fold in range(0, K):\n",
    "\n",
    "  # Instantiate algorithm\n",
    "  model = RandomForestRegressor()\n",
    "  scaler = StandardScaler()\n",
    "\n",
    "  # Create training and test samples\n",
    "  X_train, X_test, y_train, y_test = train_test_split(X, y, train_size=split, random_state=42)\n",
    "\n",
    "  # Scale X data, we scale the data because it helps the algorithm to converge\n",
    "  # and helps the algorithm to not be greedy with large values\n",
    "  scaler.fit(X_train)\n",
    "  X_train = scaler.transform(X_train)\n",
    "  X_test = scaler.transform(X_test)\n",
    "\n",
    "  # Train model\n",
    "  trained_model = model.fit(X_train, y_train)\n",
    "\n",
    "  # Generate predictions on test sample\n",
    "  y_pred = trained_model.predict(X_test)\n",
    "\n",
    "  # Compute accuracy, using mean absolute error\n",
    "  mae = mean_absolute_error(y_true=y_test, y_pred=y_pred)\n",
    "  accuracy.append(mae)\n",
    "  print(f\"Fold {fold + 1}: MAE = {mae:.3f}\")\n",
    "\n",
    "print(f\"Average MAE: {(sum(accuracy) / len(accuracy)):.2f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x1440 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotting importance of features\n",
    "\n",
    "features = [i.split(\"__\")[0] for i in X.columns]\n",
    "importances = model.feature_importances_\n",
    "indices = np.argsort(importances)\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(10, 20))\n",
    "plt.title('Feature Importances')\n",
    "plt.barh(range(len(indices)), importances[indices], color='b', align='center')\n",
    "plt.yticks(range(len(indices)), [features[i] for i in indices])\n",
    "plt.xlabel('Relative Importance')\n",
    "plt.show()"
   ]
  }
 ],
 "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": 4
}