I_A_Predição_de_vendas.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.7.9"
    },
    "papermill": {
      "default_parameters": {},
      "duration": 183.02005,
      "end_time": "2021-01-28T19:27:27.012857",
      "environment_variables": {},
      "exception": null,
      "input_path": "__notebook__.ipynb",
      "output_path": "__notebook__.ipynb",
      "parameters": {},
      "start_time": "2021-01-28T19:24:23.992807",
      "version": "2.2.2"
    },
    "colab": {
      "name": "I.A_Predição_de_vendas.ipynb",
      "provenance": []
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:24:30.603508Z",
          "iopub.status.busy": "2021-01-28T19:24:30.602489Z",
          "iopub.status.idle": "2021-01-28T19:24:30.607102Z",
          "shell.execute_reply": "2021-01-28T19:24:30.607633Z"
        },
        "papermill": {
          "duration": 0.043807,
          "end_time": "2021-01-28T19:24:30.607963",
          "exception": false,
          "start_time": "2021-01-28T19:24:30.564156",
          "status": "completed"
        },
        "tags": [],
        "id": "ae33qfuxkagU"
      },
      "source": [
        "#Library for data manipulation and analysis\n",
        "import pandas as pd"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:24:30.756892Z",
          "iopub.status.busy": "2021-01-28T19:24:30.756206Z",
          "iopub.status.idle": "2021-01-28T19:24:30.758453Z",
          "shell.execute_reply": "2021-01-28T19:24:30.758964Z"
        },
        "papermill": {
          "duration": 0.039766,
          "end_time": "2021-01-28T19:24:30.759154",
          "exception": false,
          "start_time": "2021-01-28T19:24:30.719388",
          "status": "completed"
        },
        "tags": [],
        "id": "nICvd2GckagX"
      },
      "source": [
        "import os"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:24:30.836648Z",
          "iopub.status.busy": "2021-01-28T19:24:30.833821Z",
          "iopub.status.idle": "2021-01-28T19:24:33.192932Z",
          "shell.execute_reply": "2021-01-28T19:24:33.193512Z"
        },
        "papermill": {
          "duration": 2.401957,
          "end_time": "2021-01-28T19:24:33.193862",
          "exception": false,
          "start_time": "2021-01-28T19:24:30.791905",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZqbTzBNjkagY",
        "outputId": "c32c2632-a920-4838-a8f7-4ec832d205ec"
      },
      "source": [
        "import numpy as np\n",
        "from numpy.random import seed\n",
        "from numpy.random import randn\n",
        "\n",
        "\n",
        "\n",
        "from sklearn import linear_model\n",
        "from statsmodels.stats.outliers_influence import variance_inflation_factor\n",
        "from statsmodels.graphics.gofplots import qqplot\n",
        "import statsmodels.api as sm\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "%matplotlib inline\n",
        "plt.style.use('seaborn-whitegrid')\n",
        "\n",
        "import seaborn as sns\n",
        "sns.set(style=\"whitegrid\")\n",
        "\n",
        "import math\n",
        "%config IPCompleter.greedy=True\n",
        "\n",
        "import csv\n",
        "import time\n",
        "\n",
        "import scipy.stats as stats"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/statsmodels/tools/_testing.py:19: FutureWarning: pandas.util.testing is deprecated. Use the functions in the public API at pandas.testing instead.\n",
            "  import pandas.util.testing as tm\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:24:33.265723Z",
          "iopub.status.busy": "2021-01-28T19:24:33.265042Z",
          "iopub.status.idle": "2021-01-28T19:24:33.270322Z",
          "shell.execute_reply": "2021-01-28T19:24:33.269636Z"
        },
        "papermill": {
          "duration": 0.042515,
          "end_time": "2021-01-28T19:24:33.270466",
          "exception": false,
          "start_time": "2021-01-28T19:24:33.227951",
          "status": "completed"
        },
        "tags": [],
        "id": "_dqfJHugkagY"
      },
      "source": [
        "import warnings\n",
        "warnings.filterwarnings('always')\n",
        "warnings.filterwarnings('ignore')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:24:33.341163Z",
          "iopub.status.busy": "2021-01-28T19:24:33.340502Z",
          "iopub.status.idle": "2021-01-28T19:24:33.343086Z",
          "shell.execute_reply": "2021-01-28T19:24:33.343595Z"
        },
        "papermill": {
          "duration": 0.040281,
          "end_time": "2021-01-28T19:24:33.343788",
          "exception": false,
          "start_time": "2021-01-28T19:24:33.303507",
          "status": "completed"
        },
        "tags": [],
        "id": "lCfrmo0vkagZ"
      },
      "source": [
        "##1.\tRead the ‘csv’ file provided "
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:24:33.418500Z",
          "iopub.status.busy": "2021-01-28T19:24:33.417628Z",
          "iopub.status.idle": "2021-01-28T19:24:34.356148Z",
          "shell.execute_reply": "2021-01-28T19:24:34.355212Z"
        },
        "papermill": {
          "duration": 0.979161,
          "end_time": "2021-01-28T19:24:34.356338",
          "exception": false,
          "start_time": "2021-01-28T19:24:33.377177",
          "status": "completed"
        },
        "tags": [],
        "id": "VE6FdAWkkagZ"
      },
      "source": [
        "dataset = pd.read_csv(\"/content/drive/MyDrive/Colab Notebooks/archive/train_data.csv\", names=['date','product_identifier','department_identifier','category_of_product','outlet','state','sales'],sep=',', header=0)\n",
        "prod_price = pd.read_csv(\"/content/drive/MyDrive/Colab Notebooks/archive/product_prices.csv\",sep=',', header=0,names=['outlet','product_identifier','week_id','sell_price'])\n",
        "date_week=pd.read_csv(\"/content/drive/MyDrive/Colab Notebooks/archive/date_to_week_id_map.csv\",sep=',',header=0, names=['date','week_id'])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pJOoVCqFlA2t",
        "outputId": "50803d0d-8321-46d5-cba8-f8a4fd9d8d77"
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:24:34.428327Z",
          "iopub.status.busy": "2021-01-28T19:24:34.427352Z",
          "iopub.status.idle": "2021-01-28T19:24:34.431461Z",
          "shell.execute_reply": "2021-01-28T19:24:34.432149Z"
        },
        "papermill": {
          "duration": 0.041564,
          "end_time": "2021-01-28T19:24:34.432333",
          "exception": false,
          "start_time": "2021-01-28T19:24:34.390769",
          "status": "completed"
        },
        "tags": [],
        "id": "HNVUNy18kaga"
      },
      "source": [
        "##a.\tGet the dimensions of the data"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:24:34.503825Z",
          "iopub.status.busy": "2021-01-28T19:24:34.502744Z",
          "iopub.status.idle": "2021-01-28T19:24:34.509472Z",
          "shell.execute_reply": "2021-01-28T19:24:34.510162Z"
        },
        "papermill": {
          "duration": 0.044534,
          "end_time": "2021-01-28T19:24:34.510372",
          "exception": false,
          "start_time": "2021-01-28T19:24:34.465838",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0j8TLvAKkaga",
        "outputId": "c307e796-83f3-452b-bfc8-cceae8bd52aa"
      },
      "source": [
        "dataset.shape, prod_price.shape, date_week.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((395000, 7), (59000, 4), (821, 2))"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:24:34.583857Z",
          "iopub.status.busy": "2021-01-28T19:24:34.583176Z",
          "iopub.status.idle": "2021-01-28T19:24:34.585416Z",
          "shell.execute_reply": "2021-01-28T19:24:34.585914Z"
        },
        "papermill": {
          "duration": 0.041348,
          "end_time": "2021-01-28T19:24:34.586115",
          "exception": false,
          "start_time": "2021-01-28T19:24:34.544767",
          "status": "completed"
        },
        "tags": [],
        "id": "umujGMIGkagb"
      },
      "source": [
        "##merge csv files"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:24:34.658075Z",
          "iopub.status.busy": "2021-01-28T19:24:34.657409Z",
          "iopub.status.idle": "2021-01-28T19:24:35.007918Z",
          "shell.execute_reply": "2021-01-28T19:24:35.007347Z"
        },
        "papermill": {
          "duration": 0.387553,
          "end_time": "2021-01-28T19:24:35.008081",
          "exception": false,
          "start_time": "2021-01-28T19:24:34.620528",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 419
        },
        "id": "C5iun1Vkkagb",
        "outputId": "b5b32d86-9616-4c19-fae0-41480516b927"
      },
      "source": [
        "df=pd.merge(prod_price,date_week, on=['week_id'], how='inner')\n",
        "dataset=pd.merge(dataset,df, on=['date','product_identifier','outlet'], how='inner')\n",
        "dataset"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>date</th>\n",
              "      <th>product_identifier</th>\n",
              "      <th>department_identifier</th>\n",
              "      <th>category_of_product</th>\n",
              "      <th>outlet</th>\n",
              "      <th>state</th>\n",
              "      <th>sales</th>\n",
              "      <th>week_id</th>\n",
              "      <th>sell_price</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>2012-01-01</td>\n",
              "      <td>74</td>\n",
              "      <td>11</td>\n",
              "      <td>others</td>\n",
              "      <td>111</td>\n",
              "      <td>Maharashtra</td>\n",
              "      <td>0</td>\n",
              "      <td>49</td>\n",
              "      <td>2.94</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2012-01-01</td>\n",
              "      <td>337</td>\n",
              "      <td>11</td>\n",
              "      <td>others</td>\n",
              "      <td>111</td>\n",
              "      <td>Maharashtra</td>\n",
              "      <td>1</td>\n",
              "      <td>49</td>\n",
              "      <td>7.44</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2012-01-01</td>\n",
              "      <td>423</td>\n",
              "      <td>12</td>\n",
              "      <td>others</td>\n",
              "      <td>111</td>\n",
              "      <td>Maharashtra</td>\n",
              "      <td>0</td>\n",
              "      <td>49</td>\n",
              "      <td>0.97</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>2012-01-01</td>\n",
              "      <td>432</td>\n",
              "      <td>12</td>\n",
              "      <td>others</td>\n",
              "      <td>111</td>\n",
              "      <td>Maharashtra</td>\n",
              "      <td>0</td>\n",
              "      <td>49</td>\n",
              "      <td>4.97</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2012-01-01</td>\n",
              "      <td>581</td>\n",
              "      <td>21</td>\n",
              "      <td>fast_moving_consumer_goods</td>\n",
              "      <td>111</td>\n",
              "      <td>Maharashtra</td>\n",
              "      <td>0</td>\n",
              "      <td>49</td>\n",
              "      <td>4.88</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394995</th>\n",
              "      <td>2014-02-28</td>\n",
              "      <td>2932</td>\n",
              "      <td>33</td>\n",
              "      <td>drinks_and_food</td>\n",
              "      <td>333</td>\n",
              "      <td>Kerala</td>\n",
              "      <td>2</td>\n",
              "      <td>161</td>\n",
              "      <td>2.78</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394996</th>\n",
              "      <td>2014-02-28</td>\n",
              "      <td>2935</td>\n",
              "      <td>33</td>\n",
              "      <td>drinks_and_food</td>\n",
              "      <td>333</td>\n",
              "      <td>Kerala</td>\n",
              "      <td>8</td>\n",
              "      <td>161</td>\n",
              "      <td>0.20</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394997</th>\n",
              "      <td>2014-02-28</td>\n",
              "      <td>3004</td>\n",
              "      <td>33</td>\n",
              "      <td>drinks_and_food</td>\n",
              "      <td>333</td>\n",
              "      <td>Kerala</td>\n",
              "      <td>0</td>\n",
              "      <td>161</td>\n",
              "      <td>2.50</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394998</th>\n",
              "      <td>2014-02-28</td>\n",
              "      <td>3008</td>\n",
              "      <td>33</td>\n",
              "      <td>drinks_and_food</td>\n",
              "      <td>333</td>\n",
              "      <td>Kerala</td>\n",
              "      <td>0</td>\n",
              "      <td>161</td>\n",
              "      <td>1.98</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394999</th>\n",
              "      <td>2014-02-28</td>\n",
              "      <td>3021</td>\n",
              "      <td>33</td>\n",
              "      <td>drinks_and_food</td>\n",
              "      <td>333</td>\n",
              "      <td>Kerala</td>\n",
              "      <td>0</td>\n",
              "      <td>161</td>\n",
              "      <td>2.08</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>395000 rows × 9 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "              date  product_identifier  ...  week_id sell_price\n",
              "0       2012-01-01                  74  ...       49       2.94\n",
              "1       2012-01-01                 337  ...       49       7.44\n",
              "2       2012-01-01                 423  ...       49       0.97\n",
              "3       2012-01-01                 432  ...       49       4.97\n",
              "4       2012-01-01                 581  ...       49       4.88\n",
              "...            ...                 ...  ...      ...        ...\n",
              "394995  2014-02-28                2932  ...      161       2.78\n",
              "394996  2014-02-28                2935  ...      161       0.20\n",
              "394997  2014-02-28                3004  ...      161       2.50\n",
              "394998  2014-02-28                3008  ...      161       1.98\n",
              "394999  2014-02-28                3021  ...      161       2.08\n",
              "\n",
              "[395000 rows x 9 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:24:35.082231Z",
          "iopub.status.busy": "2021-01-28T19:24:35.081589Z",
          "iopub.status.idle": "2021-01-28T19:24:35.084051Z",
          "shell.execute_reply": "2021-01-28T19:24:35.084483Z"
        },
        "papermill": {
          "duration": 0.04187,
          "end_time": "2021-01-28T19:24:35.084667",
          "exception": false,
          "start_time": "2021-01-28T19:24:35.042797",
          "status": "completed"
        },
        "tags": [],
        "id": "eak_d07dkagb"
      },
      "source": [
        "##b.\tGet the summary"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:24:35.162334Z",
          "iopub.status.busy": "2021-01-28T19:24:35.161709Z",
          "iopub.status.idle": "2021-01-28T19:24:35.265628Z",
          "shell.execute_reply": "2021-01-28T19:24:35.266192Z"
        },
        "papermill": {
          "duration": 0.146934,
          "end_time": "2021-01-28T19:24:35.266365",
          "exception": false,
          "start_time": "2021-01-28T19:24:35.119431",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 297
        },
        "id": "eQrC7NhRkagc",
        "outputId": "70534c68-40df-402f-8192-306f1046bca5"
      },
      "source": [
        "dataset.describe()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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_identifier</th>\n",
              "      <th>department_identifier</th>\n",
              "      <th>outlet</th>\n",
              "      <th>sales</th>\n",
              "      <th>week_id</th>\n",
              "      <th>sell_price</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>395000.000000</td>\n",
              "      <td>395000.000000</td>\n",
              "      <td>395000.000000</td>\n",
              "      <td>395000.000000</td>\n",
              "      <td>395000.000000</td>\n",
              "      <td>395000.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>1509.960000</td>\n",
              "      <td>24.460000</td>\n",
              "      <td>211.200000</td>\n",
              "      <td>1.228919</td>\n",
              "      <td>105.070886</td>\n",
              "      <td>4.987644</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>809.799518</td>\n",
              "      <td>6.337863</td>\n",
              "      <td>91.161291</td>\n",
              "      <td>3.595266</td>\n",
              "      <td>32.578749</td>\n",
              "      <td>3.874444</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>74.000000</td>\n",
              "      <td>11.000000</td>\n",
              "      <td>111.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>49.000000</td>\n",
              "      <td>0.050000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>926.000000</td>\n",
              "      <td>21.000000</td>\n",
              "      <td>113.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>77.000000</td>\n",
              "      <td>2.680000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>1325.000000</td>\n",
              "      <td>22.000000</td>\n",
              "      <td>221.500000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>105.000000</td>\n",
              "      <td>3.980000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>1753.000000</td>\n",
              "      <td>31.000000</td>\n",
              "      <td>331.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>133.000000</td>\n",
              "      <td>6.480000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>3021.000000</td>\n",
              "      <td>33.000000</td>\n",
              "      <td>333.000000</td>\n",
              "      <td>293.000000</td>\n",
              "      <td>161.000000</td>\n",
              "      <td>44.360000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       product_identifier  department_identifier  ...        week_id     sell_price\n",
              "count       395000.000000          395000.000000  ...  395000.000000  395000.000000\n",
              "mean          1509.960000              24.460000  ...     105.070886       4.987644\n",
              "std            809.799518               6.337863  ...      32.578749       3.874444\n",
              "min             74.000000              11.000000  ...      49.000000       0.050000\n",
              "25%            926.000000              21.000000  ...      77.000000       2.680000\n",
              "50%           1325.000000              22.000000  ...     105.000000       3.980000\n",
              "75%           1753.000000              31.000000  ...     133.000000       6.480000\n",
              "max           3021.000000              33.000000  ...     161.000000      44.360000\n",
              "\n",
              "[8 rows x 6 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:24:35.342517Z",
          "iopub.status.busy": "2021-01-28T19:24:35.341911Z",
          "iopub.status.idle": "2021-01-28T19:24:35.345365Z",
          "shell.execute_reply": "2021-01-28T19:24:35.345840Z"
        },
        "papermill": {
          "duration": 0.043963,
          "end_time": "2021-01-28T19:24:35.346019",
          "exception": false,
          "start_time": "2021-01-28T19:24:35.302056",
          "status": "completed"
        },
        "tags": [],
        "id": "JFWzf4sIkagc"
      },
      "source": [
        "##c.\tPrint the first 5 rows and last 5 rows."
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:24:35.432063Z",
          "iopub.status.busy": "2021-01-28T19:24:35.426681Z",
          "iopub.status.idle": "2021-01-28T19:24:35.437887Z",
          "shell.execute_reply": "2021-01-28T19:24:35.437110Z"
        },
        "papermill": {
          "duration": 0.056004,
          "end_time": "2021-01-28T19:24:35.438028",
          "exception": false,
          "start_time": "2021-01-28T19:24:35.382024",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "id": "0qSvb5P2kagc",
        "outputId": "84f9a4f1-fe53-40b3-db57-0cff48ce4bfb"
      },
      "source": [
        "dataset.head(5)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>date</th>\n",
              "      <th>product_identifier</th>\n",
              "      <th>department_identifier</th>\n",
              "      <th>category_of_product</th>\n",
              "      <th>outlet</th>\n",
              "      <th>state</th>\n",
              "      <th>sales</th>\n",
              "      <th>week_id</th>\n",
              "      <th>sell_price</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>2012-01-01</td>\n",
              "      <td>74</td>\n",
              "      <td>11</td>\n",
              "      <td>others</td>\n",
              "      <td>111</td>\n",
              "      <td>Maharashtra</td>\n",
              "      <td>0</td>\n",
              "      <td>49</td>\n",
              "      <td>2.94</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2012-01-01</td>\n",
              "      <td>337</td>\n",
              "      <td>11</td>\n",
              "      <td>others</td>\n",
              "      <td>111</td>\n",
              "      <td>Maharashtra</td>\n",
              "      <td>1</td>\n",
              "      <td>49</td>\n",
              "      <td>7.44</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2012-01-01</td>\n",
              "      <td>423</td>\n",
              "      <td>12</td>\n",
              "      <td>others</td>\n",
              "      <td>111</td>\n",
              "      <td>Maharashtra</td>\n",
              "      <td>0</td>\n",
              "      <td>49</td>\n",
              "      <td>0.97</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>2012-01-01</td>\n",
              "      <td>432</td>\n",
              "      <td>12</td>\n",
              "      <td>others</td>\n",
              "      <td>111</td>\n",
              "      <td>Maharashtra</td>\n",
              "      <td>0</td>\n",
              "      <td>49</td>\n",
              "      <td>4.97</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2012-01-01</td>\n",
              "      <td>581</td>\n",
              "      <td>21</td>\n",
              "      <td>fast_moving_consumer_goods</td>\n",
              "      <td>111</td>\n",
              "      <td>Maharashtra</td>\n",
              "      <td>0</td>\n",
              "      <td>49</td>\n",
              "      <td>4.88</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "         date  product_identifier  ...  week_id sell_price\n",
              "0  2012-01-01                  74  ...       49       2.94\n",
              "1  2012-01-01                 337  ...       49       7.44\n",
              "2  2012-01-01                 423  ...       49       0.97\n",
              "3  2012-01-01                 432  ...       49       4.97\n",
              "4  2012-01-01                 581  ...       49       4.88\n",
              "\n",
              "[5 rows x 9 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:24:35.518363Z",
          "iopub.status.busy": "2021-01-28T19:24:35.517560Z",
          "iopub.status.idle": "2021-01-28T19:24:35.527259Z",
          "shell.execute_reply": "2021-01-28T19:24:35.527796Z"
        },
        "papermill": {
          "duration": 0.053672,
          "end_time": "2021-01-28T19:24:35.527977",
          "exception": false,
          "start_time": "2021-01-28T19:24:35.474305",
          "status": "completed"
        },
        "tags": [],
        "id": "oNCwcbfokagc",
        "outputId": "a3574252-97f8-41e9-efd7-be15d28c5fd2"
      },
      "source": [
        "##c.\tPrint the first 5 rows and last 5 rows.\n",
        "dataset.tail(5)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>date</th>\n",
              "      <th>product_identifier</th>\n",
              "      <th>department_identifier</th>\n",
              "      <th>category_of_product</th>\n",
              "      <th>outlet</th>\n",
              "      <th>state</th>\n",
              "      <th>sales</th>\n",
              "      <th>week_id</th>\n",
              "      <th>sell_price</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>394995</th>\n",
              "      <td>2014-02-28</td>\n",
              "      <td>2932</td>\n",
              "      <td>33</td>\n",
              "      <td>drinks_and_food</td>\n",
              "      <td>333</td>\n",
              "      <td>Kerala</td>\n",
              "      <td>2</td>\n",
              "      <td>161</td>\n",
              "      <td>2.78</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394996</th>\n",
              "      <td>2014-02-28</td>\n",
              "      <td>2935</td>\n",
              "      <td>33</td>\n",
              "      <td>drinks_and_food</td>\n",
              "      <td>333</td>\n",
              "      <td>Kerala</td>\n",
              "      <td>8</td>\n",
              "      <td>161</td>\n",
              "      <td>0.20</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394997</th>\n",
              "      <td>2014-02-28</td>\n",
              "      <td>3004</td>\n",
              "      <td>33</td>\n",
              "      <td>drinks_and_food</td>\n",
              "      <td>333</td>\n",
              "      <td>Kerala</td>\n",
              "      <td>0</td>\n",
              "      <td>161</td>\n",
              "      <td>2.50</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394998</th>\n",
              "      <td>2014-02-28</td>\n",
              "      <td>3008</td>\n",
              "      <td>33</td>\n",
              "      <td>drinks_and_food</td>\n",
              "      <td>333</td>\n",
              "      <td>Kerala</td>\n",
              "      <td>0</td>\n",
              "      <td>161</td>\n",
              "      <td>1.98</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394999</th>\n",
              "      <td>2014-02-28</td>\n",
              "      <td>3021</td>\n",
              "      <td>33</td>\n",
              "      <td>drinks_and_food</td>\n",
              "      <td>333</td>\n",
              "      <td>Kerala</td>\n",
              "      <td>0</td>\n",
              "      <td>161</td>\n",
              "      <td>2.08</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "              date  product_identifier  department_identifier  \\\n",
              "394995  2014-02-28                2932                     33   \n",
              "394996  2014-02-28                2935                     33   \n",
              "394997  2014-02-28                3004                     33   \n",
              "394998  2014-02-28                3008                     33   \n",
              "394999  2014-02-28                3021                     33   \n",
              "\n",
              "       category_of_product  outlet   state  sales  week_id  sell_price  \n",
              "394995     drinks_and_food     333  Kerala      2      161        2.78  \n",
              "394996     drinks_and_food     333  Kerala      8      161        0.20  \n",
              "394997     drinks_and_food     333  Kerala      0      161        2.50  \n",
              "394998     drinks_and_food     333  Kerala      0      161        1.98  \n",
              "394999     drinks_and_food     333  Kerala      0      161        2.08  "
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:24:35.607913Z",
          "iopub.status.busy": "2021-01-28T19:24:35.606278Z",
          "iopub.status.idle": "2021-01-28T19:24:35.610602Z",
          "shell.execute_reply": "2021-01-28T19:24:35.611242Z"
        },
        "papermill": {
          "duration": 0.046537,
          "end_time": "2021-01-28T19:24:35.611436",
          "exception": false,
          "start_time": "2021-01-28T19:24:35.564899",
          "status": "completed"
        },
        "tags": [],
        "id": "7F2spyBdkagd"
      },
      "source": [
        "##5.\tVerify if there are any null values in each of the columns. (don’t impute now)."
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:24:35.690048Z",
          "iopub.status.busy": "2021-01-28T19:24:35.689064Z",
          "iopub.status.idle": "2021-01-28T19:24:35.820168Z",
          "shell.execute_reply": "2021-01-28T19:24:35.820617Z"
        },
        "papermill": {
          "duration": 0.171871,
          "end_time": "2021-01-28T19:24:35.820843",
          "exception": false,
          "start_time": "2021-01-28T19:24:35.648972",
          "status": "completed"
        },
        "tags": [],
        "id": "9DrbKF1bkagd",
        "outputId": "c9ecdcae-33b5-4b2c-d5e6-5dcdda1fd112"
      },
      "source": [
        "dataset.isnull().sum()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "date                     0\n",
              "product_identifier       0\n",
              "department_identifier    0\n",
              "category_of_product      0\n",
              "outlet                   0\n",
              "state                    0\n",
              "sales                    0\n",
              "week_id                  0\n",
              "sell_price               0\n",
              "dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:24:35.903707Z",
          "iopub.status.busy": "2021-01-28T19:24:35.902639Z",
          "iopub.status.idle": "2021-01-28T19:24:35.906714Z",
          "shell.execute_reply": "2021-01-28T19:24:35.907165Z"
        },
        "papermill": {
          "duration": 0.048786,
          "end_time": "2021-01-28T19:24:35.907481",
          "exception": false,
          "start_time": "2021-01-28T19:24:35.858695",
          "status": "completed"
        },
        "tags": [],
        "id": "1nCzTBi9kagd",
        "outputId": "12ce8eb5-3593-4c3e-b497-1008085adaa2"
      },
      "source": [
        "##d.\tExplore the data types of each column.\n",
        "dataset.dtypes"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "date                      object\n",
              "product_identifier         int64\n",
              "department_identifier      int64\n",
              "category_of_product       object\n",
              "outlet                     int64\n",
              "state                     object\n",
              "sales                      int64\n",
              "week_id                    int64\n",
              "sell_price               float64\n",
              "dtype: object"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 19
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:24:35.991346Z",
          "iopub.status.busy": "2021-01-28T19:24:35.990544Z",
          "iopub.status.idle": "2021-01-28T19:24:36.079599Z",
          "shell.execute_reply": "2021-01-28T19:24:36.080146Z"
        },
        "papermill": {
          "duration": 0.133641,
          "end_time": "2021-01-28T19:24:36.080330",
          "exception": false,
          "start_time": "2021-01-28T19:24:35.946689",
          "status": "completed"
        },
        "tags": [],
        "id": "JfQHL0pfkage"
      },
      "source": [
        "##converting as category\n",
        "col = ['category_of_product', 'state']\n",
        "dataset[col] = dataset[col].astype('category')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:24:36.164373Z",
          "iopub.status.busy": "2021-01-28T19:24:36.163378Z",
          "iopub.status.idle": "2021-01-28T19:24:36.167881Z",
          "shell.execute_reply": "2021-01-28T19:24:36.168366Z"
        },
        "papermill": {
          "duration": 0.049899,
          "end_time": "2021-01-28T19:24:36.168537",
          "exception": false,
          "start_time": "2021-01-28T19:24:36.118638",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-qnJW2RZkage",
        "outputId": "8106dde8-78cc-4454-a95e-70cc30e9168b"
      },
      "source": [
        "##e.\tUnderstand if all the columns are in appropriate datatypes or not.\n",
        "dataset.dtypes"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "date                       object\n",
              "product_identifier          int64\n",
              "department_identifier       int64\n",
              "category_of_product      category\n",
              "outlet                      int64\n",
              "state                    category\n",
              "sales                       int64\n",
              "week_id                     int64\n",
              "sell_price                float64\n",
              "dtype: object"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:24:36.252472Z",
          "iopub.status.busy": "2021-01-28T19:24:36.251587Z",
          "iopub.status.idle": "2021-01-28T19:24:36.254752Z",
          "shell.execute_reply": "2021-01-28T19:24:36.254102Z"
        },
        "papermill": {
          "duration": 0.04787,
          "end_time": "2021-01-28T19:24:36.254898",
          "exception": false,
          "start_time": "2021-01-28T19:24:36.207028",
          "status": "completed"
        },
        "tags": [],
        "id": "8HAvs4Nxkage"
      },
      "source": [
        "##check for insights using scatterplot\n",
        "def scatter(dataset, column):\n",
        "    plt.figure()\n",
        "    plt.scatter(dataset[column] , dataset['sales'])\n",
        "    plt.ylabel('sales')\n",
        "    plt.xlabel(column)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:24:36.355529Z",
          "iopub.status.busy": "2021-01-28T19:24:36.350009Z",
          "iopub.status.idle": "2021-01-28T19:24:42.935794Z",
          "shell.execute_reply": "2021-01-28T19:24:42.935096Z"
        },
        "papermill": {
          "duration": 6.642485,
          "end_time": "2021-01-28T19:24:42.935957",
          "exception": false,
          "start_time": "2021-01-28T19:24:36.293472",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "3UO4DoDfkagf",
        "outputId": "827077f3-3d0b-49b5-c079-e4ca9592a0cf"
      },
      "source": [
        "scatter(dataset, 'product_identifier')\n",
        "scatter(dataset, 'department_identifier')\n",
        "scatter(dataset, 'category_of_product')\n",
        "scatter(dataset, 'outlet')\n",
        "scatter(dataset, 'state')\n",
        "scatter(dataset, 'week_id')\n",
        "scatter(dataset, 'sell_price')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYwAAAEMCAYAAADXiYGSAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3df1AU9/0/8OfdhTsFRUADOZRIQiO9yX0TzV1Hmxkdg7Fgg2intRKipnHML4uTxGhDiAHjjxiQpJkkGGKT6QxTIi1JKpWIoHGY/Ghj9YzWKyo2VSThCvLDRPlxwu1+/+DDBYQ7FtzbvYPn4y/Z9/143du9e+6+9727GlEURRAREQ1Bq3YBREQUGBgYREQkCQODiIgkYWAQEZEkDAwiIpLkJrUL8BVBENDW1oagoCBoNBq1yyEiCgiiKKKrqwshISHQavvvU4zawGhra0NNTY3aZRARBaQZM2Zg4sSJ/ZYpFhhr167FN998A61Wi+DgYLz44oswmUw4f/48MjIycPnyZYSFhSEnJwexsbEA4LVtKEFBQQB6PrRer/fRpwosdrsdZrNZ7TL8CvukP/bHQGOtT65du4aamhr3b2hfigVGTk6OO60OHTqEzMxM/PWvf0V2djbS0tKwZMkSlJaWIisrC4WFhQDgtW0ovcNQer0eBoPBNx8qALEvBmKf9Mf+GGgs9slgQ/mKHfTuu2tz9epVaDQaNDc3o7q6GsnJyQCA5ORkVFdXo6WlxWsbEREpT9FjGC+88AK++OILiKKId999Fw6HA1FRUdDpdAAAnU6HyMhIOBwOiKLosS0iIkLJsomICAoHxvbt2wEAe/fuRW5uLp566imfv6fdbvf5ewQSm82mdgl+h33SH/tjIPZJD1VmSS1duhRZWVm45ZZb0NDQAJfLBZ1OB5fLhcbGRhiNRoii6LFtOMxm85gcfxyMzWaDxWJRuwy/wj7pj/0x0FjrE6fT6XFDW5FjGG1tbXA4HO6/Dx8+jEmTJmHy5MkwmUwoKysDAJSVlcFkMiEiIsJrG1GgqrLVYfW2SqQ8W4rV2ypRZatTuyQiyRTZw+jo6MBTTz2Fjo4OaLVaTJo0CQUFBdBoNNi8eTMyMjKwa9cuhIaGIicnx/08b21EgabKVoe3Sk7C2eUCAFxq7cBbJScBAPMtMWqWRiSJIoExZcoU/OUvfxm0LS4uDiUlJcNuIwo0heWn3WHRy9nlQmH5aQYGBQReS4pIIU2tHcNaTuRvGBhECpkSPn5Yy4n8DQODSCGrFplgCNL1W2YI0mHVIpNKFRENz6i9+CCRv+k9TlFYfhpNrR2YEj4eqxaZePyCAgYDg0hB8y0xDAgKWBySIiIiSRgYREQkCQODiIgkYWAQEZEkDAwiIpKEgUFERJIwMIiISBIGBhERScLAICIiSRgYREQkCQODiIgkYWAQEZEkDAwiIpKEgUFERJIwMIiISBIGBhERScLAICIiSRgYREQkCQODiIgkYWAQEZEkNynxJq2trfjd736HixcvQq/XY/r06diyZQsiIiIQHx+PGTNmQKvtya7c3FzEx8cDAA4fPozc3Fy4XC7ceeed2LFjB8aPH69EyUREdB1F9jA0Gg3WrFmDiooK7Nu3DzExMcjLy3O3FxcXo7S0FKWlpe6waGtrw4svvoiCggIcPHgQISEheO+995Qol4iIBqFIYISFhWH27Nnuv2fOnIn6+nqvz/n0009hNpsRGxsLAEhNTUV5ebkvyyQiIi8UGZLqSxAE7NmzBwkJCe5lK1euhMvlwrx587Bu3Tro9Xo4HA5ER0e7HxMdHQ2Hw6F0uURE9H8UD4ytW7ciODgYK1asAABUVVXBaDTi6tWr2LhxI/Lz8/HMM8/I9n52u1221xoNbDab2iX4HfZJf+yPgdgnPRQNjJycHNTW1qKgoMB9kNtoNAIAJkyYgGXLluGPf/yje/mRI0fcz62vr3c/djjMZjMMBoMM1Qc+m80Gi8Widhl+hX3SH/tjoLHWJ06n0+OGtmKB8dprr8Fut2P37t3Q6/UAgO+++w4GgwHjxo1Dd3c3KioqYDKZAABz587F1q1bceHCBcTGxqK4uBiLFi1SqlwiooBSZatDYflpNLV2YEr4eKxaZMJ8S4ys76FIYJw7dw7vvPMOYmNjkZqaCgCYNm0a1qxZg6ysLGg0GnR3d2PWrFl46qmnAPTscWzZsgWPP/44BEGAyWTCCy+8oES5RD6lxBebxpYqWx3eKjkJZ5cLAHCptQNvlZwEAFnXLUUC44477sDZs2cHbdu3b5/H591///24//77fVUWkeKU+mLT2FJYftq9TvVydrlQWH5a1vWKZ3oTKcjbF5topJpaO4a1fKQYGEQKUuqLTWPLlPDBr4DhaflIMTCIFKTUF5vGllWLTDAE6fotMwTpsGqRSdb3YWAQKUipLzaNLfMtMUhfdjduDh8PDYCbw8cjfdndgTlLioh69H6BOUuK5DbfEuPz9YiBQaQwJb7YRL7AISkiIpKEgUFERJIwMIiISBIGBhERScLAICIiSRgYREQkCQODiIgkYWAQEZEkDAwiIpKEZ3oTEY0Co+aOe0RE5DtK3ZiLQ1JERAFOqRtzMTCIiAIc77hHRESS8I57REQkiVI35uJBbyKiAKfUjbkYGEREo4ASN+bikBQREUnCwCAiIkkUCYzW1lY8+uijSExMxOLFi5Geno6WlhYAwIkTJ5CSkoLExESsXr0azc3N7ud5ayMiImUpEhgajQZr1qxBRUUF9u3bh5iYGOTl5UEQBGzcuBFZWVmoqKiA1WpFXl4eAHhtIyIi5SkSGGFhYZg9e7b775kzZ6K+vh52ux0GgwFWqxUAkJqaigMHDgCA1zYiIlKe4rOkBEHAnj17kJCQAIfDgejoaHdbREQEBEHA5cuXvbaFhYVJfj+73S5r/YHOZrOpXYLfYZ/0x/4YiH3SQ/HA2Lp1K4KDg7FixQocPHjQ5+9nNpthMBh8/j6BwGazwWKxqF2GX2Gf9Mf+GGis9YnT6fS4oa1oYOTk5KC2thYFBQXQarUwGo2or693t7e0tECr1SIsLMxrGxERKU+xabWvvfYa7HY78vPzodfrAfRs/Xd2duLYsWMAgOLiYiQlJQ3ZRkREylNkD+PcuXN45513EBsbi9TUVADAtGnTkJ+fj9zcXGRnZ8PpdGLq1KnYuXMnAECr1XpsIyIi5SkSGHfccQfOnj07aNs999yDffv2DbuNiIiUxTO9iYhIEgYGERFJwsAgIiJJGBhERCQJA4OIiCRhYBARkSQMDCIikoSBQUREkjAwiIhIEsWvVks01lXZ6lBYfhpNrR2YEj4eqxaZMN8So3ZZRENiYBApqMpWh7dKTsLZ5QIAXGrtwFslJwGAoUE3zNcbIxySIlJQYflpd1j0cna5UFh+WqWKaLR4+4MTePX947jU2gERP2yMVNnqZHsPBgaRgppaO4a1nEiKKlsd9v+jdsByuTdGGBhECpoSPn5Yy4mk2L33lMc2OTdGGBhEClq1yARDkK7fMkOQDqsWmVSqiEaDK+1dHtvk3BjhQW8iBfUegOQsKVKKnBsjDAwihc23xDAgSFYTg4MG3cswBGn9Y5bUl19+iX/+85+yFUJERCPz2NL/B51W02+ZTqtB+rKZsr6P5MBYsWIFbDYbAGD37t1Yv349nn32WRQUFMhaEBERDc98SwyeTp2Fm8PHQwPg5vDxeDp1lux7spKHpM6dO4eZM3vSqqSkBIWFhQgJCcGDDz6IJ554QtaiiIhoeJQY6pQcGIIgQKPR4OLFixBFET/60Y8AAN99953PiiMiIv8hOTAsFgu2bNmCS5cuYeHChQCAixcvIjw83GfFERGRNG9/cAIHjlyEIIjQajVImn0rnvyVSscwduzYgdDQUMTHxyM9PR0A8N///herVq2StSAiIhqetz84gf3/qIUgiAAAQRCx/x+1ePuDE7K+j+Q9jPDwcKxfv77fsvnz58taDBERDd+BIxc9LpdzL0PyHsa1a9fw+9//HgsWLIDFYgEAfP755/jTn/4kWzFERDR8vXsWUpePlOTAePnll1FTU4O8vDxoND3zfe+44w7s2bNH0vNzcnKQkJCA+Ph41NTUuJcnJCQgKSkJS5YswZIlS/DZZ5+5206cOIGUlBQkJiZi9erVaG5ullouEdGYodEMvlyr9dAwQpKHpA4dOoTKykoEBwdDq+3JmaioKDQ0NEh6/oIFC7Bq1So89NBDA9reeOMNzJgxo98yQRCwceNG7NixA1arFbt27UJeXh527NghtWQiolGvylYHeNiRSJp9q6zvJXkPIygoCC5X/+v4t7S0ICwsTNLzrVYrjEaj5MLsdjsMBgOsVisAIDU1FQcOHJD8fCKisaCw/PSgeWEI0so+S0ryHkZSUhKee+45PP/88wCAxsZGvPzyy3jggQduuIgNGzZAFEVYLBasX78eoaGhcDgciI6Odj8mIiICgiDg8uXLkkMK6Ake+kHv2fr0A/ZJf+yPgfy5Ty55uHy5s0uQvW7JgfHMM88gLy8PKSkp6OjoQGJiIpYtW4bf/va3N1RAUVERjEYjrl27hu3bt2PLli3Iy8u7odfsy2w2w2AwyPZ6gcxms7knLFAP9kl/7I+B/L1Pbi5vHjQ0bg4fP6K6nU6nxw1tyUNSer0emZmZ+Oqrr/D3v/8dx48fR2ZmJvR6/bAL6qt3mEqv1yMtLQ3Hjx93L6+vr3c/rqWlBVqtdlh7F0REo52S91jxuodRV+f5XrBtbW3uf8fEjOz6Je3t7XC5XJg4cSJEUcT+/fthMvV8SLPZjM7OThw7dgxWqxXFxcVISkoa0fsQEY1WSt5jxWtgLFy4EBqNBqLoeS6vRqPB6dND3zN227ZtqKysRFNTEx555BGEhYWhoKAA69atg8vlgiAIiIuLQ3Z2NgBAq9UiNzcX2dnZcDqdmDp1Knbu3DnMj0dENPopdY8Vr4Fx5swZ2d5o06ZN2LRp04Dle/fu9fice+65B/v27ZOthuGqstXxzmhERP+Hd9zzoMpWh7dKTsLZ1TOV+FJrB94qOQkADA0iGpMkB0Z3dzfef/99HD16FK2trf2GqYqKinxSnJoKy0+7w6KXs8uFwvLTDAwiGpOGdbXaP//5z7Barfj3v/+Nn/3sZ2hubsacOXN8WZ9qmjzMbfa0nIhotJMcGJWVlfjDH/6Ahx9+GDqdDg8//DDy8/Nx5MgRX9anminh44e1nIhITVW2OqzeVomUZ0uxeltlzyVDZCY5MDo7O93nTIwbNw4dHR2Ii4tDdXW17EX5AyXnNhMR3YjeY66XWjsg4odjrnKHhuRjGHFxcTh16hTuuusumM1mvPnmm5gwYQKioqJkLchfKDm3mYjoRih1zFVyYGRmZuKmm3oenpGRgc2bN6O9vR1bt26VrRh/o9TcZiKiG6HUMVfJgdHe3o6pU6cCAIKDgxEZGQmdTofp06fLWhDRaMZze8gXpoSPH/R6UnIfc5V8DOOll16CTtczpp+TkwOXywWNRoMXX3xR1oKIRiulxplp7FHqmKvkPYyGhgZER0eju7sbn3/+OQ4fPoygoCDMnTtX1oKIRiue20O+Mt8Sg9Pnm3HgyEUIggitVoMF1mmyr1eS9zAmTJiApqYmHD16FHFxcQgJCQHQc0IfEQ2N5/aQr1TZ6vDJsW/c9/AWBBGfHPtGvVlSK1aswK9+9St0dXUhMzMTAHD8+HHcfvvtshZENFopNc5MY4/fzZJ67LHHsHDhQuh0Otx6a899YqOiorBt2zbZiiEazX7y40js/0ftoMuJboSnu+55Wj5Sw7r44G233eb1byLy7OiZRo/Ln1S4FhpdNBpgsLtQaDTyvg+vVusFp0CSnHgMg3zF0y2LvNzKaEQkH/QeazgFkuTG65NRoGNgeODtIBLRSPD6ZBToOCTlAYcPSG68Phn5yk06DbpdA8efbtLJexCDgeEBp0CSL/D6ZOQLLmHwgxWelo8Uh6Q84PABEQUKpQ56cw/DAw4fEFEgUHIiDgPDCw4fEJG/G2oiTpWtTrbfMQaGFzwPg4j83VATceS8PAiPYXjA8zCIKBAMNRFHzpmdigRGTk4OEhISEB8fj5qaGvfy8+fPY/ny5UhMTMTy5ctx4cIFSW1K4HkY5CtVtjqs3laJlGdLsXpbJTdC6IYMNRFHzpmdigTGggULUFRU5L5jX6/s7GykpaWhoqICaWlpyMrKktSmBJ6HQb7APVeS23xLDCIm6j22yzmzU5HAsFqtMBqN/ZY1NzejuroaycnJAIDk5GRUV1ejpaXFa5sSqmx1gIfzXSYEBylSA41O3HMlX2i5ck2R91HtGIbD4UBUVJT7tq86nQ6RkZFwOBxe23ytylaH14u/kn3+MhHAPVdSnpwbI6N+lpTdbh/W43eVfOv17Mgr7V2w2Ww3WpZqArl2X1GyT0KDdfiu3TXocn/5v/GXOvxJIPfJpdYO2epXLTCMRiMaGhrgcrmg0+ngcrnQ2NgIo9EIURQ9tg2X2WyGwWCQ/PiO97/x2m4I0sJisQy7Dn9gs9kCtnZfUbpP1qDnGEbfYSlDkA5rlt4Nix9M2eY6MlBA9ImX362JwUHDqt/pdHrc0FZtSGry5MkwmUwoKysDAJSVlcFkMiEiIsJrm9qudQtql0ABbL4lBunL7sbN4eOhAXBz+HikL7ub5/dQQFBkD2Pbtm2orKxEU1MTHnnkEYSFheHjjz/G5s2bkZGRgV27diE0NBQ5OTnu53hrUxOPbdCN4hUESElX27tkey1FAmPTpk3YtGnTgOVxcXEoKSkZ9Dne2oiISJqAOw+DiIjU8ZMfR8r2WqN+ltRwebqZeq+beT8MGiFem4zUcPRMI56U6bUYGNcZ6hiFnGlNY0fvGd69s6N6z/AGwNAgnxrsRnAjxSGp62i13m9pePRMo0KV0GjCM7xJLRoZ79LKwLiOMMQtDXlGLo2Ep/XmUmsHryNFPiXnzE4GxnWGOkahD2KX0fB5m6nCiw9SoOCv33UGu5d3X84unrhHw7dqkQk6D8OdHJqiQMHAGAT3Ikhu8y0xCB7neY6JnAcmiXyFs6T6uH4mC5GcvJ1xO9RkCyJ/wE3pPgabyUIkF2/HMYaabEHkDxgYfXAGFPmStzuf8YRQCgQMjD54Nz3ypUP/rB10uVarkfU2mkR98TwMogBTZavDyf80D9omCCLP9iaf4XkYPiLnZYCJ+tq995TaJRDdMAZGH3JeBpioryucIUWjAAOjD44jkxqSZt+qdglEkjAw+uA4MqnBdNtktUsgkoSB0Qev50NqeL34K657FBAYGH3wej6kBpcgct2jgMDA6IMn7pFaeC0pCgQMjD44S4rUwplSFAgYGH2M41VqSSW8lhQFAv5C9lHX2KZ2CTRG8VpSFAgYGER+4Cc/jlS7BKIhMTCI/MDRM41ql0ABSskp2X5xA6WEhATo9XoYDAYAwIYNGzB37lycOHECWVlZcDqdmDp1Knbu3InJk9U/yanKVseT/EhWnKFHI6XklGy/2cN44403UFpaitLSUsydOxeCIGDjxo3IyspCRUUFrFYr8vLyfFqD1HFkzpknuXGGHo2UkhsbfhMY17Pb7TAYDLBarQCA1NRUHDhwwKfvGT05WNLjuDVIcuMMPRopJTc2/GYt3bBhAxYvXozNmzfj+++/h8PhQHR0tLs9IiICgiDg8uXLPqvhX18Pfr+C63FrkOTGGXo0UkpeNNUvjmEUFRXBaDTi2rVr2L59O7Zs2YKFCxfK8tp2u13yY6XeaGT6ZA1sNtsIK1JXoNbtS/7SJ6zDf/lzn0yU8Bi56veLwDAajQAAvV6PtLQ0PPnkk1i1ahXq6+vdj2lpaYFWq0VYWNiwXttsNrsPpg/p/W8kPay2WYTFYhlWHf7AZrMFZN2+pFifSFi3/OH/huvIQAHRJ0OsX8Op3+l0etzQVn1Iqr29HVeuXAEAiKKI/fv3w2QywWw2o7OzE8eOHQMAFBcXIykpSc1S3XjdHyLyF29/cEKx91J9D6O5uRnr1q2Dy+WCIAiIi4tDdnY2tFotcnNzkZ2d3W9arT/gdX+IyF+Uf1mr2HupHhgxMTHYu3fvoG333HMP9u3bp3BFQ+N1f4jIX0g99ioH1YekAhGv+0NEYxEDYwR4728i8hdKjpAzMIiIApiSI+QMjBHYvfeU2iUQESmOgTECV9q71C6BiEhxDAwiIpKEgUFERJIwMIiISBIGBhERScLAICIiSRgYREQkCQODiIgkYWAQEZEkDAwiIpKEgUFERJIwMIiISBIGBhERScLAICIiSRgYREQkCQODiIgkYWAQEZEkDAwiIpKEgUFERJIwMIiISBIGBhERSeL3gXH+/HksX74ciYmJWL58OS5cuKB2SUREY5LfB0Z2djbS0tJQUVGBtLQ0ZGVlqV0SEdGY5NeB0dzcjOrqaiQnJwMAkpOTUV1djZaWFpUrIyIae25SuwBvHA4HoqKioNPpAAA6nQ6RkZFwOByIiIiQ9Bp2u90ntdlsNp+8rq8Fat2+5C99wjr8V6D3iVz1+3VgyMFsNsNgMEh78PvfSH5di8UyworUY7PZArJuX1KsTySsW/7wf8N1ZCC/7xOZ1y2n0+lxQ9uvh6SMRiMaGhrgcrkAAC6XC42NjTAajSpXRkQ09vh1YEyePBkmkwllZWUAgLKyMphMJsnDUcO179Ulsj6OqNdQ6wzXKRopJdctvx+S2rx5MzIyMrBr1y6EhoYiJyfHp+/HLy75Ctct8hWl1i2/D4y4uDiUlJSoXQYR0Zjn10NSRETkPxgYREQkCQODiIgk8ftjGCMliiIA4Nq1aypX4l+cTqfaJfgd9kl/7I+BxlKf9P5m9v6G9qURB1s6Cly5cgU1NTVql0FEFJBmzJiBiRMn9ls2agNDEAS0tbUhKCgIGo1G7XKIiAKCKIro6upCSEgItNr+Ry1GbWAQEZG8eNCbiIgkYWAQEZEkDAwiIpKEgUFERJIwMIiISBIGBhERScLAICIiSQI2MHJycpCQkID4+Ph+Z3SfP38ey5cvR2JiIpYvX44LFy5Iagt0nvojISEBSUlJWLJkCZYsWYLPPvvM3XbixAmkpKQgMTERq1evRnNzsxql+0RrayseffRRJCYmYvHixUhPT0dLSwsA7597rPZJfHw8Fi9e7F5Pzp49637e4cOHkZSUhIULF+Lpp59GR0eHWh9BdmvXrkVKSgqWLl2KtLQ0nD59GsDY/R0Zkhigjh49KtbX14v33XefePbsWffylStXinv37hVFURT37t0rrly5UlJboPPUH9f/3cvlcon333+/ePToUVEURTE/P1/MyMhQrF5fa21tFb/88kv336+88or4/PPPe/3cY7VPRFEUZ8yYIV69enXAc65evSree++94vnz50VRFMXMzEzxzTffVKReJXz//ffufx88eFBcunSpKIpj93dkKAG7h2G1Wgfc27u5uRnV1dVITk4GACQnJ6O6uhotLS1e20aDwfrDG7vdDoPBAKvVCgBITU3FgQMHfFWe4sLCwjB79mz33zNnzkR9fb3Xzz1W+8SbTz/9FGazGbGxsQB6+qS8vNyXZSqq77WSrl69Co1GM6Z/R4Yyqq5W63A4EBUVBZ1OBwDQ6XSIjIyEw+GAKIoe23x1j3B/sWHDBoiiCIvFgvXr1yM0NBQOhwPR0dHux0REREAQBFy+fBlhYWEqVis/QRCwZ88eJCQkeP3cY7VPeq1cuRIulwvz5s3DunXroNfrB/RJdHQ0HA6HGiX7zAsvvIAvvvgCoiji3Xff5e+IFwG7h0HSFBUV4W9/+xs+/PBDiKKILVu2qF2S4rZu3Yrg4GCsWLFC7VL8xvV9UlVVhY8++ghFRUX4z3/+g/z8fJUrVM727dtRVVWFZ555Brm5uWqX49dGVWAYjUY0NDTA5XIBAFwuFxobG2E0Gr22jWa9n0+v1yMtLQ3Hjx93L+87HNHS0gKtVjvqtqRzcnJQW1uL119/HVqt1uvnHqt9AvywnkyYMAHLli3zuJ7U19eP2u/M0qVLceTIEdxyyy38HfFgVAXG5MmTYTKZUFZWBgAoKyuDyWRCRESE17bRqr29HVeuXAHQc8ni/fv3w2QyAQDMZjM6Oztx7NgxAEBxcTGSkpJUq9UXXnvtNdjtduTn50Ov1wPw/rnHap9899136OzsBAB0d3ejoqLCvZ7MnTsXp06dcs8EKi4uxqJFi1SpXW5tbW39htcOHz6MSZMm8XfEi4C9vPm2bdtQWVmJpqYmhIeHIywsDB9//DG+/vprZGRk4Pvvv0doaChycnJw++23A4DXtkA3WH8UFBRg3bp1cLlcEAQBcXFx2LRpEyIjIwEAx48fR3Z2NpxOJ6ZOnYqdO3diypQpKn8SeZw7dw7JycmIjY3FuHHjAADTpk1Dfn6+1889FvtkzZo1yMrKgkajQXd3N2bNmoXMzEyEhIQAAA4dOoSdO3dCEASYTCa88sorCA4OVvOjyKKpqQlr165FR0cHtFotJk2ahOeeew533nnnmP0dGUrABgYRESlrVA1JERGR7zAwiIhIEgYGERFJwsAgIiJJGBhERCQJA4NIRUeOHMG8efPULoNIEgYGkYLi4+NRW1s7oue++eab2LBhg8wVEUnHwCAiIkkYGEQj8PXXX2PlypWwWq144IEH8MknnwDoueJrSUmJ+3EfffQRHnzwQQDAQw89BABYsmQJZs2ahf379w943YaGBqxbtw5z5sxBQkICCgsLAfRcZvydd95BeXk5Zs2ahZSUFF9/RKIBRtXlzYmU0NXVhSeeeAK//OUv8d5778Fms2Ht2rX48MMPvT6vqKgI8fHxKC0txfTp0wH0HMPoJQgCnnzySSQkJODVV19FQ0MDfvOb3+C2227DvHnz8Pjjj6O2thZ5eXk+/XxEnnAPg2iYTp48ifb2djz22GPQ6/X46U9/ivvuuw8ff/zxDb3uqVOn0NLSgvT0dOj1esTExODXv/71oHsiRGrgHgbRMDU2NuKWW25xXxoc6LmxUENDww297rfffovGxkb3Hf+Anstn9/2bSE0MDKJhioyMxP/+9z8IguAODYfDgdjYWDQ2NqKjo8P92KamJsmvazQaMbMpnGgAAAEdSURBVG3aNFRWVg7artFobqxwohvEISmiYbrrrrswbtw4vPvuu+jq6sKRI0dw+PBh/PznP4fJZMLBgwfR0dGB2tpafPDBB/2eO2XKFNTV1Xl83ZCQEOzevRudnZ1wuVyoqanBv/71LwA993v59ttvIQiCzz8j0WAYGETDpNfrUVBQgE8//RRz5szBSy+9hNzcXMTFxeHhhx9GUFAQ7r33Xjz33HNYvHhxv+emp6cjIyMDVqt1wLEJnU6HgoICnDlzBgsWLMCcOXOwadMmXL16FQDcN3OaPXs2fvGLXyjzYYn64P0wiIhIEu5hEBGRJAwMIiKShIFBRESSMDCIiEgSBgYREUnCwCAiIkkYGEREJAkDg4iIJGFgEBGRJP8fgvs/VuyFBG8AAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:24:43.038218Z",
          "iopub.status.busy": "2021-01-28T19:24:43.037061Z",
          "iopub.status.idle": "2021-01-28T19:24:43.419887Z",
          "shell.execute_reply": "2021-01-28T19:24:43.419146Z"
        },
        "papermill": {
          "duration": 0.437893,
          "end_time": "2021-01-28T19:24:43.420049",
          "exception": false,
          "start_time": "2021-01-28T19:24:42.982156",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 833
        },
        "id": "OgdBNIHikagf",
        "outputId": "981ff294-39d6-4de6-c1c3-666eb40d48bb"
      },
      "source": [
        "fig = plt.figure(figsize=(18, 14))\n",
        "corr = dataset.corr()\n",
        "c = plt.pcolor(corr)\n",
        "plt.yticks(np.arange(0.5, len(corr.index), 1), corr.index)\n",
        "plt.xticks(np.arange(0.5, len(corr.columns), 1), corr.columns)\n",
        "fig.colorbar(c)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.colorbar.Colorbar at 0x7fd21eca4f98>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 18
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1296x1008 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:24:43.559880Z",
          "iopub.status.busy": "2021-01-28T19:24:43.559108Z",
          "iopub.status.idle": "2021-01-28T19:25:38.296749Z",
          "shell.execute_reply": "2021-01-28T19:25:38.297245Z"
        },
        "papermill": {
          "duration": 54.829754,
          "end_time": "2021-01-28T19:25:38.297436",
          "exception": false,
          "start_time": "2021-01-28T19:24:43.467682",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "Upk10Sv7kagf",
        "outputId": "112dee23-f119-404c-907c-d70dc163c3dc"
      },
      "source": [
        "#vars() function in Python,vars() method takes only one parameter and that too is optional,\n",
        "#It takes an object as a parameter which may be can a module, a class, an instance, or any object having __dict__ attribute.\n",
        "sns.pairplot(dataset, vars=['product_identifier','department_identifier','category_of_product','outlet','state','sales'])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<seaborn.axisgrid.PairGrid at 0x7fd21ec5a908>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 19
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x1080 with 42 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:25:38.406526Z",
          "iopub.status.busy": "2021-01-28T19:25:38.405881Z",
          "iopub.status.idle": "2021-01-28T19:25:38.677511Z",
          "shell.execute_reply": "2021-01-28T19:25:38.676960Z"
        },
        "papermill": {
          "duration": 0.328437,
          "end_time": "2021-01-28T19:25:38.677670",
          "exception": false,
          "start_time": "2021-01-28T19:25:38.349233",
          "status": "completed"
        },
        "tags": [],
        "id": "ICUu4wr9kagg",
        "outputId": "91de6b26-53cf-41c7-c199-831678a68940"
      },
      "source": [
        "for name, group in dataset.groupby([\"outlet\", \"department_identifier\"]):\n",
        "    plt.title(name)\n",
        "    plt.scatter(range(len(group)), group[\"sales\"])\n",
        "    plt.show()\n",
        "    break"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:25:38.789708Z",
          "iopub.status.busy": "2021-01-28T19:25:38.788603Z",
          "iopub.status.idle": "2021-01-28T19:25:38.949401Z",
          "shell.execute_reply": "2021-01-28T19:25:38.948795Z"
        },
        "papermill": {
          "duration": 0.218607,
          "end_time": "2021-01-28T19:25:38.949537",
          "exception": false,
          "start_time": "2021-01-28T19:25:38.730930",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 419
        },
        "id": "E1ZvJ21ekagg",
        "outputId": "a56bb164-5939-4606-adea-af099e9513f2"
      },
      "source": [
        "##use label encoder-LabelEncoder encode labels with a value between 0 and n_classes-1 where n is the number of distinct labels\n",
        "from sklearn.preprocessing import OneHotEncoder\n",
        "from sklearn.preprocessing import LabelEncoder\n",
        "dataset['state_encoded'] = LabelEncoder().fit_transform(dataset['state'])\n",
        "dataset[['state', 'state_encoded']]"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>state</th>\n",
              "      <th>state_encoded</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Maharashtra</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Maharashtra</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Maharashtra</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Maharashtra</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Maharashtra</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394995</th>\n",
              "      <td>Kerala</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394996</th>\n",
              "      <td>Kerala</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394997</th>\n",
              "      <td>Kerala</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394998</th>\n",
              "      <td>Kerala</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394999</th>\n",
              "      <td>Kerala</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>395000 rows × 2 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "              state  state_encoded\n",
              "0       Maharashtra              1\n",
              "1       Maharashtra              1\n",
              "2       Maharashtra              1\n",
              "3       Maharashtra              1\n",
              "4       Maharashtra              1\n",
              "...             ...            ...\n",
              "394995       Kerala              0\n",
              "394996       Kerala              0\n",
              "394997       Kerala              0\n",
              "394998       Kerala              0\n",
              "394999       Kerala              0\n",
              "\n",
              "[395000 rows x 2 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:25:39.061316Z",
          "iopub.status.busy": "2021-01-28T19:25:39.060460Z",
          "iopub.status.idle": "2021-01-28T19:25:39.064442Z",
          "shell.execute_reply": "2021-01-28T19:25:39.063822Z"
        },
        "papermill": {
          "duration": 0.063062,
          "end_time": "2021-01-28T19:25:39.064581",
          "exception": false,
          "start_time": "2021-01-28T19:25:39.001519",
          "status": "completed"
        },
        "tags": [],
        "id": "Z0NVEjvGkagh"
      },
      "source": [
        "#### Maharastra=1; Kerala = 0; Telangana=2"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:25:39.176067Z",
          "iopub.status.busy": "2021-01-28T19:25:39.175065Z",
          "iopub.status.idle": "2021-01-28T19:25:39.340949Z",
          "shell.execute_reply": "2021-01-28T19:25:39.340282Z"
        },
        "papermill": {
          "duration": 0.223957,
          "end_time": "2021-01-28T19:25:39.341095",
          "exception": false,
          "start_time": "2021-01-28T19:25:39.117138",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 419
        },
        "id": "9shNpvx3kagh",
        "outputId": "ea0da54d-a44e-4088-8a91-e00da2fa64e1"
      },
      "source": [
        "dataset['cat_prod_encoded'] = LabelEncoder().fit_transform(dataset['category_of_product'])\n",
        "dataset[['category_of_product', 'cat_prod_encoded']]"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>category_of_product</th>\n",
              "      <th>cat_prod_encoded</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>others</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>others</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>others</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>others</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>fast_moving_consumer_goods</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394995</th>\n",
              "      <td>drinks_and_food</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394996</th>\n",
              "      <td>drinks_and_food</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394997</th>\n",
              "      <td>drinks_and_food</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394998</th>\n",
              "      <td>drinks_and_food</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394999</th>\n",
              "      <td>drinks_and_food</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>395000 rows × 2 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "               category_of_product  cat_prod_encoded\n",
              "0                           others                 2\n",
              "1                           others                 2\n",
              "2                           others                 2\n",
              "3                           others                 2\n",
              "4       fast_moving_consumer_goods                 1\n",
              "...                            ...               ...\n",
              "394995             drinks_and_food                 0\n",
              "394996             drinks_and_food                 0\n",
              "394997             drinks_and_food                 0\n",
              "394998             drinks_and_food                 0\n",
              "394999             drinks_and_food                 0\n",
              "\n",
              "[395000 rows x 2 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 21
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:25:39.452130Z",
          "iopub.status.busy": "2021-01-28T19:25:39.451098Z",
          "iopub.status.idle": "2021-01-28T19:25:39.454195Z",
          "shell.execute_reply": "2021-01-28T19:25:39.454728Z"
        },
        "papermill": {
          "duration": 0.060715,
          "end_time": "2021-01-28T19:25:39.454915",
          "exception": false,
          "start_time": "2021-01-28T19:25:39.394200",
          "status": "completed"
        },
        "tags": [],
        "id": "9LOhaEjKkagh"
      },
      "source": [
        "#### drinks_and_food=0; fast_moving_consumer_goods=1; others=2"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:25:39.591928Z",
          "iopub.status.busy": "2021-01-28T19:25:39.591230Z",
          "iopub.status.idle": "2021-01-28T19:25:39.717447Z",
          "shell.execute_reply": "2021-01-28T19:25:39.716742Z"
        },
        "papermill": {
          "duration": 0.209181,
          "end_time": "2021-01-28T19:25:39.717587",
          "exception": false,
          "start_time": "2021-01-28T19:25:39.508406",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 419
        },
        "id": "v35Twevnkagh",
        "outputId": "920819b9-15e0-4863-8f1b-0d6e4279a8ee"
      },
      "source": [
        "dataset['Month'] = pd.to_datetime(dataset['date']).dt.month\n",
        "dataset = dataset.drop(columns=[\"date\",\"week_id\",\"state\",\"category_of_product\"])\n",
        "dataset"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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_identifier</th>\n",
              "      <th>department_identifier</th>\n",
              "      <th>outlet</th>\n",
              "      <th>sales</th>\n",
              "      <th>sell_price</th>\n",
              "      <th>state_encoded</th>\n",
              "      <th>cat_prod_encoded</th>\n",
              "      <th>Month</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>74</td>\n",
              "      <td>11</td>\n",
              "      <td>111</td>\n",
              "      <td>0</td>\n",
              "      <td>2.94</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>337</td>\n",
              "      <td>11</td>\n",
              "      <td>111</td>\n",
              "      <td>1</td>\n",
              "      <td>7.44</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>423</td>\n",
              "      <td>12</td>\n",
              "      <td>111</td>\n",
              "      <td>0</td>\n",
              "      <td>0.97</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>432</td>\n",
              "      <td>12</td>\n",
              "      <td>111</td>\n",
              "      <td>0</td>\n",
              "      <td>4.97</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>581</td>\n",
              "      <td>21</td>\n",
              "      <td>111</td>\n",
              "      <td>0</td>\n",
              "      <td>4.88</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394995</th>\n",
              "      <td>2932</td>\n",
              "      <td>33</td>\n",
              "      <td>333</td>\n",
              "      <td>2</td>\n",
              "      <td>2.78</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394996</th>\n",
              "      <td>2935</td>\n",
              "      <td>33</td>\n",
              "      <td>333</td>\n",
              "      <td>8</td>\n",
              "      <td>0.20</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394997</th>\n",
              "      <td>3004</td>\n",
              "      <td>33</td>\n",
              "      <td>333</td>\n",
              "      <td>0</td>\n",
              "      <td>2.50</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394998</th>\n",
              "      <td>3008</td>\n",
              "      <td>33</td>\n",
              "      <td>333</td>\n",
              "      <td>0</td>\n",
              "      <td>1.98</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394999</th>\n",
              "      <td>3021</td>\n",
              "      <td>33</td>\n",
              "      <td>333</td>\n",
              "      <td>0</td>\n",
              "      <td>2.08</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>395000 rows × 8 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "        product_identifier  department_identifier  ...  cat_prod_encoded  Month\n",
              "0                       74                     11  ...                 2      1\n",
              "1                      337                     11  ...                 2      1\n",
              "2                      423                     12  ...                 2      1\n",
              "3                      432                     12  ...                 2      1\n",
              "4                      581                     21  ...                 1      1\n",
              "...                    ...                    ...  ...               ...    ...\n",
              "394995                2932                     33  ...                 0      2\n",
              "394996                2935                     33  ...                 0      2\n",
              "394997                3004                     33  ...                 0      2\n",
              "394998                3008                     33  ...                 0      2\n",
              "394999                3021                     33  ...                 0      2\n",
              "\n",
              "[395000 rows x 8 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 22
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:25:39.831742Z",
          "iopub.status.busy": "2021-01-28T19:25:39.830903Z",
          "iopub.status.idle": "2021-01-28T19:25:39.834415Z",
          "shell.execute_reply": "2021-01-28T19:25:39.833903Z"
        },
        "papermill": {
          "duration": 0.06328,
          "end_time": "2021-01-28T19:25:39.834560",
          "exception": false,
          "start_time": "2021-01-28T19:25:39.771280",
          "status": "completed"
        },
        "tags": [],
        "id": "XCHK13Qakagi"
      },
      "source": [
        "##using different machine learning algorithm"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:25:39.948170Z",
          "iopub.status.busy": "2021-01-28T19:25:39.947422Z",
          "iopub.status.idle": "2021-01-28T19:25:39.956111Z",
          "shell.execute_reply": "2021-01-28T19:25:39.956598Z"
        },
        "papermill": {
          "duration": 0.067769,
          "end_time": "2021-01-28T19:25:39.956811",
          "exception": false,
          "start_time": "2021-01-28T19:25:39.889042",
          "status": "completed"
        },
        "tags": [],
        "id": "5gn2ywdLkagi"
      },
      "source": [
        "def knn():\n",
        "    knn = KNeighborsRegressor(n_neighbors=10)\n",
        "    return knn\n",
        "\n",
        "def extraTreesRegressor():\n",
        "    clf = ExtraTreesRegressor(n_estimators=100,max_features='auto', verbose=1, n_jobs=1)\n",
        "    return clf\n",
        "\n",
        "def randomForestRegressor():\n",
        "    clf = RandomForestRegressor(n_estimators=100,max_features='log2', verbose=1)\n",
        "    return clf\n",
        "\n",
        "def svm():\n",
        "    clf = SVR(kernel='rbf', gamma='auto')\n",
        "    return clf\n",
        "\n",
        "def predict_(m, test_x):\n",
        "    return pd.Series(m.predict(test_x))\n",
        "\n",
        "def model_():\n",
        "#     return knn()\n",
        "    return extraTreesRegressor()\n",
        "#     return svm()\n",
        "#     return nn()\n",
        "#     return randomForestRegressor()    \n",
        "\n",
        "def train_(train_x, train_y):\n",
        "    m = model_()\n",
        "    m.fit(train_x, train_y)\n",
        "    return m\n",
        "\n",
        "def train_and_predict(train_x, train_y, test_x):\n",
        "    m = train_(train_x, train_y)\n",
        "    return predict_(m, test_x), m\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:25:40.072304Z",
          "iopub.status.busy": "2021-01-28T19:25:40.071282Z",
          "iopub.status.idle": "2021-01-28T19:25:40.076212Z",
          "shell.execute_reply": "2021-01-28T19:25:40.076867Z"
        },
        "papermill": {
          "duration": 0.065108,
          "end_time": "2021-01-28T19:25:40.077069",
          "exception": false,
          "start_time": "2021-01-28T19:25:40.011961",
          "status": "completed"
        },
        "tags": [],
        "id": "a3v33dfdkagj"
      },
      "source": [
        "def calculate_error(test_y, predicted, weights):\n",
        "    return mean_absolute_error(test_y, predicted, sample_weight=weights)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:25:40.191778Z",
          "iopub.status.busy": "2021-01-28T19:25:40.191050Z",
          "iopub.status.idle": "2021-01-28T19:25:40.195393Z",
          "shell.execute_reply": "2021-01-28T19:25:40.194915Z"
        },
        "papermill": {
          "duration": 0.062106,
          "end_time": "2021-01-28T19:25:40.195529",
          "exception": false,
          "start_time": "2021-01-28T19:25:40.133423",
          "status": "completed"
        },
        "tags": [],
        "id": "VHJzpo1Ekagj"
      },
      "source": [
        "from sklearn.model_selection import KFold"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:25:40.316044Z",
          "iopub.status.busy": "2021-01-28T19:25:40.315336Z",
          "iopub.status.idle": "2021-01-28T19:25:40.656751Z",
          "shell.execute_reply": "2021-01-28T19:25:40.656014Z"
        },
        "papermill": {
          "duration": 0.405593,
          "end_time": "2021-01-28T19:25:40.656909",
          "exception": false,
          "start_time": "2021-01-28T19:25:40.251316",
          "status": "completed"
        },
        "tags": [],
        "id": "FGPGBhsBkagj"
      },
      "source": [
        "##K-Folds cross-validator\n",
        "#Provides train/test indices to split data in train/test sets. Split dataset into k consecutive folds (without shuffling by default).\n",
        "kf = KFold(n_splits=5)\n",
        "splited = []\n",
        "# dataset2 = dataset.copy()\n",
        "for name, group in dataset.groupby([\"outlet\", \"department_identifier\"]):\n",
        "    group = group.reset_index(drop=True)\n",
        "    trains_x = []\n",
        "    trains_y = []\n",
        "    tests_x = []\n",
        "    tests_y = []\n",
        "    if group.shape[0] <= 5:\n",
        "        f = np.array(range(5))\n",
        "        np.random.shuffle(f)\n",
        "        group['fold'] = f[:group.shape[0]]\n",
        "        continue\n",
        "    fold = 0\n",
        "    for train_index, test_index in kf.split(group):\n",
        "        group.loc[test_index, 'fold'] = fold\n",
        "        fold += 1\n",
        "    splited.append(group)\n",
        "\n",
        "splited = pd.concat(splited).reset_index(drop=True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:25:40.790404Z",
          "iopub.status.busy": "2021-01-28T19:25:40.789410Z",
          "iopub.status.idle": "2021-01-28T19:25:40.794892Z",
          "shell.execute_reply": "2021-01-28T19:25:40.794234Z"
        },
        "papermill": {
          "duration": 0.082098,
          "end_time": "2021-01-28T19:25:40.795064",
          "exception": false,
          "start_time": "2021-01-28T19:25:40.712966",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 419
        },
        "id": "7EsOw_vAkagj",
        "outputId": "194d629e-9e5e-4a93-b3d2-41a918c92ed4"
      },
      "source": [
        "splited"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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_identifier</th>\n",
              "      <th>department_identifier</th>\n",
              "      <th>outlet</th>\n",
              "      <th>sales</th>\n",
              "      <th>sell_price</th>\n",
              "      <th>state_encoded</th>\n",
              "      <th>cat_prod_encoded</th>\n",
              "      <th>Month</th>\n",
              "      <th>fold</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>74</td>\n",
              "      <td>11</td>\n",
              "      <td>111</td>\n",
              "      <td>0</td>\n",
              "      <td>2.94</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>337</td>\n",
              "      <td>11</td>\n",
              "      <td>111</td>\n",
              "      <td>1</td>\n",
              "      <td>7.44</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>74</td>\n",
              "      <td>11</td>\n",
              "      <td>111</td>\n",
              "      <td>0</td>\n",
              "      <td>2.94</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>337</td>\n",
              "      <td>11</td>\n",
              "      <td>111</td>\n",
              "      <td>5</td>\n",
              "      <td>7.44</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>74</td>\n",
              "      <td>11</td>\n",
              "      <td>111</td>\n",
              "      <td>1</td>\n",
              "      <td>2.94</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394995</th>\n",
              "      <td>2932</td>\n",
              "      <td>33</td>\n",
              "      <td>333</td>\n",
              "      <td>2</td>\n",
              "      <td>2.78</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>4.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394996</th>\n",
              "      <td>2935</td>\n",
              "      <td>33</td>\n",
              "      <td>333</td>\n",
              "      <td>8</td>\n",
              "      <td>0.20</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>4.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394997</th>\n",
              "      <td>3004</td>\n",
              "      <td>33</td>\n",
              "      <td>333</td>\n",
              "      <td>0</td>\n",
              "      <td>2.50</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>4.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394998</th>\n",
              "      <td>3008</td>\n",
              "      <td>33</td>\n",
              "      <td>333</td>\n",
              "      <td>0</td>\n",
              "      <td>1.98</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>4.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>394999</th>\n",
              "      <td>3021</td>\n",
              "      <td>33</td>\n",
              "      <td>333</td>\n",
              "      <td>0</td>\n",
              "      <td>2.08</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>4.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>395000 rows × 9 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "        product_identifier  department_identifier  ...  Month  fold\n",
              "0                       74                     11  ...      1   0.0\n",
              "1                      337                     11  ...      1   0.0\n",
              "2                       74                     11  ...      1   0.0\n",
              "3                      337                     11  ...      1   0.0\n",
              "4                       74                     11  ...      1   0.0\n",
              "...                    ...                    ...  ...    ...   ...\n",
              "394995                2932                     33  ...      2   4.0\n",
              "394996                2935                     33  ...      2   4.0\n",
              "394997                3004                     33  ...      2   4.0\n",
              "394998                3008                     33  ...      2   4.0\n",
              "394999                3021                     33  ...      2   4.0\n",
              "\n",
              "[395000 rows x 9 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 28
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:25:40.912647Z",
          "iopub.status.busy": "2021-01-28T19:25:40.912020Z",
          "iopub.status.idle": "2021-01-28T19:25:41.050462Z",
          "shell.execute_reply": "2021-01-28T19:25:41.049816Z"
        },
        "papermill": {
          "duration": 0.199689,
          "end_time": "2021-01-28T19:25:41.050610",
          "exception": false,
          "start_time": "2021-01-28T19:25:40.850921",
          "status": "completed"
        },
        "tags": [],
        "id": "yitqJgCAkagk"
      },
      "source": [
        "from sklearn.ensemble import ExtraTreesRegressor, RandomForestRegressor"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:25:41.169071Z",
          "iopub.status.busy": "2021-01-28T19:25:41.168288Z",
          "iopub.status.idle": "2021-01-28T19:25:41.171286Z",
          "shell.execute_reply": "2021-01-28T19:25:41.171741Z"
        },
        "papermill": {
          "duration": 0.064345,
          "end_time": "2021-01-28T19:25:41.171915",
          "exception": false,
          "start_time": "2021-01-28T19:25:41.107570",
          "status": "completed"
        },
        "tags": [],
        "id": "74MixHo7kagl"
      },
      "source": [
        "from sklearn.metrics import mean_absolute_error"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:25:41.294419Z",
          "iopub.status.busy": "2021-01-28T19:25:41.293731Z",
          "iopub.status.idle": "2021-01-28T19:27:21.772176Z",
          "shell.execute_reply": "2021-01-28T19:27:21.771499Z"
        },
        "papermill": {
          "duration": 100.545163,
          "end_time": "2021-01-28T19:27:21.772328",
          "exception": false,
          "start_time": "2021-01-28T19:25:41.227165",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zkGc60DYkagl",
        "outputId": "9af4a9cc-10a9-4bab-9bf3-d73634842b88"
      },
      "source": [
        "best_model = None\n",
        "error_cv = 0\n",
        "best_error = np.iinfo(np.int32).max\n",
        "for fold in range(5):\n",
        "    dataset_train = splited.loc[splited['fold'] != fold]\n",
        "    dataset_test = splited.loc[splited['fold'] == fold]\n",
        "    train_y = dataset_train['sales']\n",
        "    train_x = dataset_train.drop(columns=['sales', 'fold'])\n",
        "    test_y = dataset_test['sales']\n",
        "    test_x = dataset_test.drop(columns=['sales', 'fold'])\n",
        "    print(dataset_train.shape, dataset_test.shape)\n",
        "    predicted, model = train_and_predict(train_x, train_y, test_x)\n",
        "    weights = test_x['outlet'].replace(True, 5).replace(False, 1)\n",
        "    error = calculate_error(test_y, predicted, weights)\n",
        "    error_cv += error\n",
        "    print(fold, error)\n",
        "    if error < best_error:\n",
        "        print('Find best model')\n",
        "        best_error = error\n",
        "        best_model = model\n",
        "error_cv /= 5"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(316000, 9) (79000, 9)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
            "[Parallel(n_jobs=1)]: Done 100 out of 100 | elapsed:   19.8s finished\n",
            "[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
            "[Parallel(n_jobs=1)]: Done 100 out of 100 | elapsed:    0.5s finished\n",
            "[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "0 1.2387110128948498\n",
            "Find best model\n",
            "(316000, 9) (79000, 9)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=1)]: Done 100 out of 100 | elapsed:   21.0s finished\n",
            "[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
            "[Parallel(n_jobs=1)]: Done 100 out of 100 | elapsed:    0.5s finished\n",
            "[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "1 1.1442954243400743\n",
            "Find best model\n",
            "(316000, 9) (79000, 9)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=1)]: Done 100 out of 100 | elapsed:   20.0s finished\n",
            "[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
            "[Parallel(n_jobs=1)]: Done 100 out of 100 | elapsed:    0.5s finished\n",
            "[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "2 1.180879889459888\n",
            "(316000, 9) (79000, 9)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=1)]: Done 100 out of 100 | elapsed:   20.9s finished\n",
            "[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
            "[Parallel(n_jobs=1)]: Done 100 out of 100 | elapsed:    0.5s finished\n",
            "[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "3 1.1098180587476256\n",
            "Find best model\n",
            "(316000, 9) (79000, 9)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=1)]: Done 100 out of 100 | elapsed:   20.3s finished\n",
            "[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "4 1.1869525643415655\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=1)]: Done 100 out of 100 | elapsed:    0.5s finished\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:27:21.904331Z",
          "iopub.status.busy": "2021-01-28T19:27:21.903687Z",
          "iopub.status.idle": "2021-01-28T19:27:21.907371Z",
          "shell.execute_reply": "2021-01-28T19:27:21.906788Z"
        },
        "papermill": {
          "duration": 0.07263,
          "end_time": "2021-01-28T19:27:21.907525",
          "exception": false,
          "start_time": "2021-01-28T19:27:21.834895",
          "status": "completed"
        },
        "tags": [],
        "id": "mz1r8Ld7kagl"
      },
      "source": [
        "if error < best_error:\n",
        "        print('Find best model')\n",
        "        best_error = error\n",
        "        best_model = model"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:27:22.039904Z",
          "iopub.status.busy": "2021-01-28T19:27:22.039206Z",
          "iopub.status.idle": "2021-01-28T19:27:22.043404Z",
          "shell.execute_reply": "2021-01-28T19:27:22.042680Z"
        },
        "papermill": {
          "duration": 0.072922,
          "end_time": "2021-01-28T19:27:22.043555",
          "exception": false,
          "start_time": "2021-01-28T19:27:21.970633",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GDuX7i81kagm",
        "outputId": "2f953f7b-0e4a-45cd-d060-9db827b931ae"
      },
      "source": [
        "print(best_error)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "1.1098180587476256\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:27:22.176792Z",
          "iopub.status.busy": "2021-01-28T19:27:22.175798Z",
          "iopub.status.idle": "2021-01-28T19:27:22.182365Z",
          "shell.execute_reply": "2021-01-28T19:27:22.181729Z"
        },
        "papermill": {
          "duration": 0.074988,
          "end_time": "2021-01-28T19:27:22.182504",
          "exception": false,
          "start_time": "2021-01-28T19:27:22.107516",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wmxmNfA-kagm",
        "outputId": "fcb0ce48-d392-41c6-9e99-64f61ca01cf4"
      },
      "source": [
        "##Decision tree\n",
        "print(best_model)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "ExtraTreesRegressor(bootstrap=False, ccp_alpha=0.0, criterion='mse',\n",
            "                    max_depth=None, max_features='auto', max_leaf_nodes=None,\n",
            "                    max_samples=None, min_impurity_decrease=0.0,\n",
            "                    min_impurity_split=None, min_samples_leaf=1,\n",
            "                    min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
            "                    n_estimators=100, n_jobs=1, oob_score=False,\n",
            "                    random_state=None, verbose=1, warm_start=False)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:27:22.314097Z",
          "iopub.status.busy": "2021-01-28T19:27:22.313167Z",
          "iopub.status.idle": "2021-01-28T19:27:22.316807Z",
          "shell.execute_reply": "2021-01-28T19:27:22.316275Z"
        },
        "papermill": {
          "duration": 0.071309,
          "end_time": "2021-01-28T19:27:22.316948",
          "exception": false,
          "start_time": "2021-01-28T19:27:22.245639",
          "status": "completed"
        },
        "tags": [],
        "id": "FsasWMFxkagm"
      },
      "source": [
        "##Using test data"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:27:22.449413Z",
          "iopub.status.busy": "2021-01-28T19:27:22.448673Z",
          "iopub.status.idle": "2021-01-28T19:27:22.532332Z",
          "shell.execute_reply": "2021-01-28T19:27:22.531584Z"
        },
        "papermill": {
          "duration": 0.15262,
          "end_time": "2021-01-28T19:27:22.532483",
          "exception": false,
          "start_time": "2021-01-28T19:27:22.379863",
          "status": "completed"
        },
        "tags": [],
        "id": "MZy_OBsikagm"
      },
      "source": [
        "dataset_test = pd.read_csv(\"/content/drive/MyDrive/Colab Notebooks/archive/test_data.csv\", names=['id','date','product_identifier','department_identifier','category_of_product','outlet','state'],sep=',', header=0)\n",
        "prod_price = pd.read_csv(\"/content/drive/MyDrive/Colab Notebooks/archive/product_prices.csv\",sep=',', header=0,names=['outlet','product_identifier','week_id','sell_price'])\n",
        "date_week=pd.read_csv(\"/content/drive/MyDrive/Colab Notebooks/archive/date_to_week_id_map.csv\",sep=',',header=0, names=['date','week_id'])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:27:22.667303Z",
          "iopub.status.busy": "2021-01-28T19:27:22.666211Z",
          "iopub.status.idle": "2021-01-28T19:27:22.798066Z",
          "shell.execute_reply": "2021-01-28T19:27:22.798525Z"
        },
        "papermill": {
          "duration": 0.203184,
          "end_time": "2021-01-28T19:27:22.798729",
          "exception": false,
          "start_time": "2021-01-28T19:27:22.595545",
          "status": "completed"
        },
        "tags": [],
        "id": "OCb8FFBHkagn",
        "outputId": "233a51c3-e39b-4a81-e950-0b108306d1cd"
      },
      "source": [
        "df=pd.merge(prod_price,date_week, on=['week_id'], how='inner')\n",
        "dataset_test=pd.merge(dataset_test,df, on=['date','product_identifier','outlet'], how='inner')\n",
        "dataset_test"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>id</th>\n",
              "      <th>date</th>\n",
              "      <th>product_identifier</th>\n",
              "      <th>department_identifier</th>\n",
              "      <th>category_of_product</th>\n",
              "      <th>outlet</th>\n",
              "      <th>state</th>\n",
              "      <th>week_id</th>\n",
              "      <th>sell_price</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>2014-03-01</td>\n",
              "      <td>74</td>\n",
              "      <td>11</td>\n",
              "      <td>others</td>\n",
              "      <td>111</td>\n",
              "      <td>Maharashtra</td>\n",
              "      <td>162</td>\n",
              "      <td>3.43</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>2014-03-01</td>\n",
              "      <td>337</td>\n",
              "      <td>11</td>\n",
              "      <td>others</td>\n",
              "      <td>111</td>\n",
              "      <td>Maharashtra</td>\n",
              "      <td>162</td>\n",
              "      <td>7.78</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>2014-03-01</td>\n",
              "      <td>423</td>\n",
              "      <td>12</td>\n",
              "      <td>others</td>\n",
              "      <td>111</td>\n",
              "      <td>Maharashtra</td>\n",
              "      <td>162</td>\n",
              "      <td>0.97</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4</td>\n",
              "      <td>2014-03-01</td>\n",
              "      <td>432</td>\n",
              "      <td>12</td>\n",
              "      <td>others</td>\n",
              "      <td>111</td>\n",
              "      <td>Maharashtra</td>\n",
              "      <td>162</td>\n",
              "      <td>4.97</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5</td>\n",
              "      <td>2014-03-01</td>\n",
              "      <td>581</td>\n",
              "      <td>21</td>\n",
              "      <td>fast_moving_consumer_goods</td>\n",
              "      <td>111</td>\n",
              "      <td>Maharashtra</td>\n",
              "      <td>162</td>\n",
              "      <td>4.88</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15495</th>\n",
              "      <td>15496</td>\n",
              "      <td>2014-03-31</td>\n",
              "      <td>2932</td>\n",
              "      <td>33</td>\n",
              "      <td>drinks_and_food</td>\n",
              "      <td>333</td>\n",
              "      <td>Kerala</td>\n",
              "      <td>166</td>\n",
              "      <td>2.78</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15496</th>\n",
              "      <td>15497</td>\n",
              "      <td>2014-03-31</td>\n",
              "      <td>2935</td>\n",
              "      <td>33</td>\n",
              "      <td>drinks_and_food</td>\n",
              "      <td>333</td>\n",
              "      <td>Kerala</td>\n",
              "      <td>166</td>\n",
              "      <td>0.20</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15497</th>\n",
              "      <td>15498</td>\n",
              "      <td>2014-03-31</td>\n",
              "      <td>3004</td>\n",
              "      <td>33</td>\n",
              "      <td>drinks_and_food</td>\n",
              "      <td>333</td>\n",
              "      <td>Kerala</td>\n",
              "      <td>166</td>\n",
              "      <td>2.68</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15498</th>\n",
              "      <td>15499</td>\n",
              "      <td>2014-03-31</td>\n",
              "      <td>3008</td>\n",
              "      <td>33</td>\n",
              "      <td>drinks_and_food</td>\n",
              "      <td>333</td>\n",
              "      <td>Kerala</td>\n",
              "      <td>166</td>\n",
              "      <td>1.98</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15499</th>\n",
              "      <td>15500</td>\n",
              "      <td>2014-03-31</td>\n",
              "      <td>3021</td>\n",
              "      <td>33</td>\n",
              "      <td>drinks_and_food</td>\n",
              "      <td>333</td>\n",
              "      <td>Kerala</td>\n",
              "      <td>166</td>\n",
              "      <td>2.08</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>15500 rows × 9 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "          id        date  product_identifier  department_identifier  \\\n",
              "0          1  2014-03-01                  74                     11   \n",
              "1          2  2014-03-01                 337                     11   \n",
              "2          3  2014-03-01                 423                     12   \n",
              "3          4  2014-03-01                 432                     12   \n",
              "4          5  2014-03-01                 581                     21   \n",
              "...      ...         ...                 ...                    ...   \n",
              "15495  15496  2014-03-31                2932                     33   \n",
              "15496  15497  2014-03-31                2935                     33   \n",
              "15497  15498  2014-03-31                3004                     33   \n",
              "15498  15499  2014-03-31                3008                     33   \n",
              "15499  15500  2014-03-31                3021                     33   \n",
              "\n",
              "              category_of_product  outlet        state  week_id  sell_price  \n",
              "0                          others     111  Maharashtra      162        3.43  \n",
              "1                          others     111  Maharashtra      162        7.78  \n",
              "2                          others     111  Maharashtra      162        0.97  \n",
              "3                          others     111  Maharashtra      162        4.97  \n",
              "4      fast_moving_consumer_goods     111  Maharashtra      162        4.88  \n",
              "...                           ...     ...          ...      ...         ...  \n",
              "15495             drinks_and_food     333       Kerala      166        2.78  \n",
              "15496             drinks_and_food     333       Kerala      166        0.20  \n",
              "15497             drinks_and_food     333       Kerala      166        2.68  \n",
              "15498             drinks_and_food     333       Kerala      166        1.98  \n",
              "15499             drinks_and_food     333       Kerala      166        2.08  \n",
              "\n",
              "[15500 rows x 9 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 46
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:27:22.932677Z",
          "iopub.status.busy": "2021-01-28T19:27:22.931761Z",
          "iopub.status.idle": "2021-01-28T19:27:22.936964Z",
          "shell.execute_reply": "2021-01-28T19:27:22.936301Z"
        },
        "papermill": {
          "duration": 0.075898,
          "end_time": "2021-01-28T19:27:22.937128",
          "exception": false,
          "start_time": "2021-01-28T19:27:22.861230",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "AIXtEqEnkagn",
        "outputId": "46a47765-bd66-4f32-b797-6fce3a7cf85f"
      },
      "source": [
        "##checking datatypes for test data\n",
        "dataset_test.dtypes"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "id                        int64\n",
              "date                     object\n",
              "product_identifier        int64\n",
              "department_identifier     int64\n",
              "category_of_product      object\n",
              "outlet                    int64\n",
              "state                    object\n",
              "dtype: object"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 38
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:27:23.071115Z",
          "iopub.status.busy": "2021-01-28T19:27:23.070043Z",
          "iopub.status.idle": "2021-01-28T19:27:23.082690Z",
          "shell.execute_reply": "2021-01-28T19:27:23.082056Z"
        },
        "papermill": {
          "duration": 0.082252,
          "end_time": "2021-01-28T19:27:23.082965",
          "exception": false,
          "start_time": "2021-01-28T19:27:23.000713",
          "status": "completed"
        },
        "tags": [],
        "id": "zJVQzEn9kagn"
      },
      "source": [
        "col = ['category_of_product', 'state']\n",
        "dataset_test[col] = dataset_test[col].astype('category')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:27:23.218490Z",
          "iopub.status.busy": "2021-01-28T19:27:23.217592Z",
          "iopub.status.idle": "2021-01-28T19:27:23.223141Z",
          "shell.execute_reply": "2021-01-28T19:27:23.222555Z"
        },
        "papermill": {
          "duration": 0.07649,
          "end_time": "2021-01-28T19:27:23.223300",
          "exception": false,
          "start_time": "2021-01-28T19:27:23.146810",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bxAcCnKokago",
        "outputId": "0f94f5dc-4045-4852-d840-39ceb3e60a50"
      },
      "source": [
        "dataset_test.dtypes"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "id                          int64\n",
              "date                       object\n",
              "product_identifier          int64\n",
              "department_identifier       int64\n",
              "category_of_product      category\n",
              "outlet                      int64\n",
              "state                    category\n",
              "dtype: object"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 40
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:27:23.365919Z",
          "iopub.status.busy": "2021-01-28T19:27:23.364919Z",
          "iopub.status.idle": "2021-01-28T19:27:23.382624Z",
          "shell.execute_reply": "2021-01-28T19:27:23.382038Z"
        },
        "papermill": {
          "duration": 0.094722,
          "end_time": "2021-01-28T19:27:23.382798",
          "exception": false,
          "start_time": "2021-01-28T19:27:23.288076",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 419
        },
        "id": "mSQK6D5Fkago",
        "outputId": "5ff33305-eeba-4a9e-e637-1eee7b55699f"
      },
      "source": [
        "##label encoder to convert categorical data, or text data, into numbers, which our predictive models can better understand.\n",
        "dataset_test['state_encoded'] = LabelEncoder().fit_transform(dataset_test['state'])\n",
        "dataset_test[['state', 'state_encoded']]"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>state</th>\n",
              "      <th>state_encoded</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Maharashtra</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Maharashtra</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Maharashtra</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Maharashtra</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Maharashtra</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15495</th>\n",
              "      <td>Kerala</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15496</th>\n",
              "      <td>Kerala</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15497</th>\n",
              "      <td>Kerala</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15498</th>\n",
              "      <td>Kerala</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15499</th>\n",
              "      <td>Kerala</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>15500 rows × 2 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "             state  state_encoded\n",
              "0      Maharashtra              1\n",
              "1      Maharashtra              1\n",
              "2      Maharashtra              1\n",
              "3      Maharashtra              1\n",
              "4      Maharashtra              1\n",
              "...            ...            ...\n",
              "15495       Kerala              0\n",
              "15496       Kerala              0\n",
              "15497       Kerala              0\n",
              "15498       Kerala              0\n",
              "15499       Kerala              0\n",
              "\n",
              "[15500 rows x 2 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 41
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:27:23.518870Z",
          "iopub.status.busy": "2021-01-28T19:27:23.518012Z",
          "iopub.status.idle": "2021-01-28T19:27:23.522054Z",
          "shell.execute_reply": "2021-01-28T19:27:23.522739Z"
        },
        "papermill": {
          "duration": 0.074648,
          "end_time": "2021-01-28T19:27:23.522950",
          "exception": false,
          "start_time": "2021-01-28T19:27:23.448302",
          "status": "completed"
        },
        "tags": [],
        "id": "ntmWbxg5kago"
      },
      "source": [
        "#### Maharastra=1; Kerala = 0; Telangana=2"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:27:23.683679Z",
          "iopub.status.busy": "2021-01-28T19:27:23.682434Z",
          "iopub.status.idle": "2021-01-28T19:27:23.697590Z",
          "shell.execute_reply": "2021-01-28T19:27:23.698244Z"
        },
        "papermill": {
          "duration": 0.100195,
          "end_time": "2021-01-28T19:27:23.698429",
          "exception": false,
          "start_time": "2021-01-28T19:27:23.598234",
          "status": "completed"
        },
        "tags": [],
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 419
        },
        "id": "87m42yt8kago",
        "outputId": "511a2be1-3ea0-48e9-fd34-321de6ce59b8"
      },
      "source": [
        "dataset_test['cat_prod_encoded'] = LabelEncoder().fit_transform(dataset_test['category_of_product'])\n",
        "dataset_test[['category_of_product', 'cat_prod_encoded']]"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>category_of_product</th>\n",
              "      <th>cat_prod_encoded</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>others</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>others</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>others</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>others</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>fast_moving_consumer_goods</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15495</th>\n",
              "      <td>drinks_and_food</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15496</th>\n",
              "      <td>drinks_and_food</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15497</th>\n",
              "      <td>drinks_and_food</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15498</th>\n",
              "      <td>drinks_and_food</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15499</th>\n",
              "      <td>drinks_and_food</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>15500 rows × 2 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "              category_of_product  cat_prod_encoded\n",
              "0                          others                 2\n",
              "1                          others                 2\n",
              "2                          others                 2\n",
              "3                          others                 2\n",
              "4      fast_moving_consumer_goods                 1\n",
              "...                           ...               ...\n",
              "15495             drinks_and_food                 0\n",
              "15496             drinks_and_food                 0\n",
              "15497             drinks_and_food                 0\n",
              "15498             drinks_and_food                 0\n",
              "15499             drinks_and_food                 0\n",
              "\n",
              "[15500 rows x 2 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 42
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "execution": {
          "iopub.execute_input": "2021-01-28T19:27:24.763710Z",
          "iopub.status.busy": "2021-01-28T19:27:24.762992Z",
          "iopub.status.idle": "2021-01-28T19:27:24.827162Z",
          "shell.execute_reply": "2021-01-28T19:27:24.826472Z"
        },
        "papermill": {
          "duration": 0.135739,
          "end_time": "2021-01-28T19:27:24.827322",
          "exception": false,
          "start_time": "2021-01-28T19:27:24.691583",
          "status": "completed"
        },
        "tags": [],
        "id": "AaUyTNt_kagq"
      },
      "source": [
        "##index=false--avoid pandas creating an index in saved csv file\n",
        "dataset_test.to_csv('final_output.csv', index=False)"
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}