Telematics_Hierarchial_Clustering.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "accelerator": "TPU",
    "colab": {
      "name": "Telematics Hierarchial Clustering.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyMbQlOeAXnoX2DHHh+S7ror",
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/Sankarlalr23/Automotive_Telematics_Analytics/blob/main/Telematics_Hierarchial_Clustering.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nGY3BZXa7KCy"
      },
      "source": [
        "The Workbook is crashing while Running. Better not to use as it need a lot of RAM Space.\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2-WQPEXvqctD"
      },
      "source": [
        "import warnings                         #'do not disturb' mode\n",
        "warnings.filterwarnings('ignore')\n",
        "\n",
        "import numpy as np                      # vectors and matrices\n",
        "import pandas as pd                     # tables and data manipulations\n",
        "import matplotlib.pyplot as plt         # plots\n",
        "import seaborn as sns                   # more plots\n",
        "\n",
        "%matplotlib inline"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qCQqhPUcqug-",
        "outputId": "066dd86a-ac43-40ff-9cea-147536e31844"
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive/')\n",
        "\n",
        "train_path='/content/drive/My Drive/Xen.AI/'\n",
        "data = pd.read_csv('/content/drive/My Drive/Xen.AI/telematics_data.csv')   # data importing"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mounted at /content/drive/\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 391
        },
        "id": "NSG8rDmSquok",
        "outputId": "d0c38b31-7ef1-4abd-a33e-e05c0db402de"
      },
      "source": [
        "data.head()"
      ],
      "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>year</th>\n",
              "      <th>month</th>\n",
              "      <th>day</th>\n",
              "      <th>vin</th>\n",
              "      <th>wmi</th>\n",
              "      <th>engine_cd</th>\n",
              "      <th>device_type</th>\n",
              "      <th>msg_ts</th>\n",
              "      <th>vin_utc_key</th>\n",
              "      <th>packet_id</th>\n",
              "      <th>reason_code</th>\n",
              "      <th>script_version</th>\n",
              "      <th>deviceid</th>\n",
              "      <th>latitude</th>\n",
              "      <th>longitude</th>\n",
              "      <th>driveridcode1</th>\n",
              "      <th>obdvin</th>\n",
              "      <th>vehiclebattvolts</th>\n",
              "      <th>fuel_total</th>\n",
              "      <th>engine_rpm</th>\n",
              "      <th>coolant_tmp</th>\n",
              "      <th>fuel_lvl_pct</th>\n",
              "      <th>driving_tme_sec</th>\n",
              "      <th>cruise_tme_sec</th>\n",
              "      <th>idle_tme_sec</th>\n",
              "      <th>idle_fuel_total</th>\n",
              "      <th>pto_tme_total</th>\n",
              "      <th>pto_fuel_total</th>\n",
              "      <th>gear</th>\n",
              "      <th>fuel_tmp</th>\n",
              "      <th>oil_tmpl</th>\n",
              "      <th>throt_pos</th>\n",
              "      <th>mpg</th>\n",
              "      <th>acel_pos</th>\n",
              "      <th>engine_pct_load</th>\n",
              "      <th>torque</th>\n",
              "      <th>oil_lvl</th>\n",
              "      <th>oil_psi</th>\n",
              "      <th>coolant_psi</th>\n",
              "      <th>intak_air_tmp</th>\n",
              "      <th>manifold_tmp</th>\n",
              "      <th>vehiclespeed</th>\n",
              "      <th>heading</th>\n",
              "      <th>odometerinmeters</th>\n",
              "      <th>odometer</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>2020</td>\n",
              "      <td>4</td>\n",
              "      <td>18</td>\n",
              "      <td>3HSDZTZR9KN352697</td>\n",
              "      <td>3HS</td>\n",
              "      <td>TZ</td>\n",
              "      <td>Xirgo</td>\n",
              "      <td>2020-04-18 06:50:40</td>\n",
              "      <td>3HSDZTZR9KN3526972020-04-18 06:50:40</td>\n",
              "      <td>36</td>\n",
              "      <td>9</td>\n",
              "      <td>204.0</td>\n",
              "      <td>173101817</td>\n",
              "      <td>38.799149</td>\n",
              "      <td>-90.603680</td>\n",
              "      <td>0.0</td>\n",
              "      <td>3HSDZTZR9KN352697</td>\n",
              "      <td>14.2</td>\n",
              "      <td>106165.0</td>\n",
              "      <td>1144.0</td>\n",
              "      <td>79.0</td>\n",
              "      <td>41.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>2167380.0</td>\n",
              "      <td>1503.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>76.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.0</td>\n",
              "      <td>11.0</td>\n",
              "      <td>12.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>340.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>4.0</td>\n",
              "      <td>10.0</td>\n",
              "      <td>1005.0</td>\n",
              "      <td>960.0</td>\n",
              "      <td>380663000.0</td>\n",
              "      <td>380663.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2020</td>\n",
              "      <td>5</td>\n",
              "      <td>20</td>\n",
              "      <td>3HSDZTZR9LN055320</td>\n",
              "      <td>3HS</td>\n",
              "      <td>TZ</td>\n",
              "      <td>Xirgo</td>\n",
              "      <td>2020-05-20 03:01:09</td>\n",
              "      <td>3HSDZTZR9LN0553202020-05-20 03:01:09</td>\n",
              "      <td>36</td>\n",
              "      <td>9</td>\n",
              "      <td>204.0</td>\n",
              "      <td>172301945</td>\n",
              "      <td>33.618963</td>\n",
              "      <td>-84.333847</td>\n",
              "      <td>0.0</td>\n",
              "      <td>3HSDZTZR9LN055320</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>584.0</td>\n",
              "      <td>61.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>56.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.0</td>\n",
              "      <td>26.0</td>\n",
              "      <td>11.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>228.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>24.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>11.0</td>\n",
              "      <td>700.0</td>\n",
              "      <td>88781990.0</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2020</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>3HSDZTZR0KN354211</td>\n",
              "      <td>3HS</td>\n",
              "      <td>TZ</td>\n",
              "      <td>Xirgo</td>\n",
              "      <td>2020-04-01 20:02:39</td>\n",
              "      <td>3HSDZTZR0KN3542112020-04-01 20:02:39</td>\n",
              "      <td>36</td>\n",
              "      <td>9</td>\n",
              "      <td>204.0</td>\n",
              "      <td>173511306</td>\n",
              "      <td>41.139005</td>\n",
              "      <td>-101.173123</td>\n",
              "      <td>0.0</td>\n",
              "      <td>3HSDZTZR0KN354211</td>\n",
              "      <td>NaN</td>\n",
              "      <td>113134.0</td>\n",
              "      <td>1145.0</td>\n",
              "      <td>81.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1788120.0</td>\n",
              "      <td>1172.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>41.0</td>\n",
              "      <td>42.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>320.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>25.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>2709.0</td>\n",
              "      <td>416436440.0</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>2020</td>\n",
              "      <td>5</td>\n",
              "      <td>5</td>\n",
              "      <td>3HSDWTZR8LN845362</td>\n",
              "      <td>3HS</td>\n",
              "      <td>TZ</td>\n",
              "      <td>Xirgo</td>\n",
              "      <td>2020-05-05 06:17:36</td>\n",
              "      <td>3HSDWTZR8LN8453622020-05-05 06:17:36</td>\n",
              "      <td>36</td>\n",
              "      <td>9</td>\n",
              "      <td>204.0</td>\n",
              "      <td>173404340</td>\n",
              "      <td>38.856184</td>\n",
              "      <td>-98.139251</td>\n",
              "      <td>0.0</td>\n",
              "      <td>3HSDWTZR8LN845362</td>\n",
              "      <td>14.0</td>\n",
              "      <td>149410.0</td>\n",
              "      <td>1368.0</td>\n",
              "      <td>81.0</td>\n",
              "      <td>54.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>786600.0</td>\n",
              "      <td>497.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>86.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.0</td>\n",
              "      <td>19.0</td>\n",
              "      <td>19.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>424.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>13.0</td>\n",
              "      <td>30.0</td>\n",
              "      <td>1095.0</td>\n",
              "      <td>838.0</td>\n",
              "      <td>426460520.0</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2020</td>\n",
              "      <td>4</td>\n",
              "      <td>7</td>\n",
              "      <td>3HSDZTZRXLN085409</td>\n",
              "      <td>3HS</td>\n",
              "      <td>TZ</td>\n",
              "      <td>N2</td>\n",
              "      <td>2020-04-07 20:14:07</td>\n",
              "      <td>3HSDZTZRXLN0854092020-04-07 20:14:07</td>\n",
              "      <td>36</td>\n",
              "      <td>9</td>\n",
              "      <td>204.0</td>\n",
              "      <td>190815-0009</td>\n",
              "      <td>33.351086</td>\n",
              "      <td>-87.017387</td>\n",
              "      <td>0.0</td>\n",
              "      <td>3HSDZTZRXLN085409</td>\n",
              "      <td>14.1</td>\n",
              "      <td>33886.0</td>\n",
              "      <td>1167.0</td>\n",
              "      <td>82.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>4755420.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>504.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>88.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>65.0</td>\n",
              "      <td>67.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>320.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1090.0</td>\n",
              "      <td>458.0</td>\n",
              "      <td>107975440.0</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   year  month  day  ... heading odometerinmeters  odometer\n",
              "0  2020      4   18  ...   960.0      380663000.0  380663.0\n",
              "1  2020      5   20  ...   700.0       88781990.0       NaN\n",
              "2  2020      4    1  ...  2709.0      416436440.0       NaN\n",
              "3  2020      5    5  ...   838.0      426460520.0       NaN\n",
              "4  2020      4    7  ...   458.0      107975440.0       NaN\n",
              "\n",
              "[5 rows x 45 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 3
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XSOw8yAhqu2s"
      },
      "source": [
        "# Data frame with important variables only\n",
        "df = pd.DataFrame(data, columns = ['vin','msg_ts','device_type','fuel_total','engine_rpm','coolant_tmp','oil_tmpl',\n",
        "                       'throt_pos','torque','oil_psi','manifold_tmp','vehiclespeed','odometerinmeters'])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JpHA8sRtqu_Z"
      },
      "source": [
        "df['msg_ts']=pd.to_datetime(df['msg_ts'])  # Converting msg_ts to date time"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "t18pi4XXq0MB"
      },
      "source": [
        "df = df.reset_index()\n",
        "df = df.drop('index', axis =1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "13kIhdrXq0Vi",
        "outputId": "298c5007-7bab-4b72-e4d5-6e628300d447"
      },
      "source": [
        "df.isnull().sum()/len(df) * 100 "
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "vin                  0.000000\n",
              "msg_ts               0.000000\n",
              "device_type          0.000000\n",
              "fuel_total          11.810623\n",
              "engine_rpm           0.000000\n",
              "coolant_tmp         11.096828\n",
              "oil_tmpl            11.075469\n",
              "throt_pos           11.072168\n",
              "torque               7.697760\n",
              "oil_psi             11.084984\n",
              "manifold_tmp        10.909447\n",
              "vehiclespeed         5.309576\n",
              "odometerinmeters     5.280256\n",
              "dtype: float64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MQL95XARq0cw"
      },
      "source": [
        "df = df.dropna(axis = 0, how = 'any')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jUbyWqiuq0iZ"
      },
      "source": [
        "df = df[df[\"device_type\"] == 'Xirgo']\n",
        "df = df[df[\"engine_rpm\"] >= 1000]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "M69U7ZxFq0n_"
      },
      "source": [
        "df = df.drop('device_type', axis =1)\n",
        "\n",
        "df = df.reset_index()\n",
        "df = df.drop('index', axis =1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4mN0qRXMq0tU",
        "outputId": "e47d51ea-8c99-4d33-f8a3-6348fbcc7222"
      },
      "source": [
        "df.isnull().sum()/len(df) * 100 "
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "vin                 0.0\n",
              "msg_ts              0.0\n",
              "fuel_total          0.0\n",
              "engine_rpm          0.0\n",
              "coolant_tmp         0.0\n",
              "oil_tmpl            0.0\n",
              "throt_pos           0.0\n",
              "torque              0.0\n",
              "oil_psi             0.0\n",
              "manifold_tmp        0.0\n",
              "vehiclespeed        0.0\n",
              "odometerinmeters    0.0\n",
              "dtype: float64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 578
        },
        "id": "_pz37lEkq0xi",
        "outputId": "7dc726c1-42f3-4691-989f-73ee0f739ae5"
      },
      "source": [
        "corrmat = df.corr()\n",
        "f, ax = plt.subplots(figsize=(12, 9))\n",
        "sns.heatmap(corrmat,vmax=1, square=True, annot = True)\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x648 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lD93SnCUq09D",
        "outputId": "7af90fdf-1b23-4f95-87fe-04eabae1f4b9"
      },
      "source": [
        "dups = df.pivot_table(index = ['vin'], aggfunc ='size')\n",
        "dupsdf=pd.DataFrame(dups)\n",
        "dupsdf=dupsdf.apply(lambda x: x.sort_values(ascending=False))\n",
        "print(dupsdf.index[:100])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Index(['3HSDZTZR5LN134290', '1HSERTZRXLH809470', '3HSDZTZRXLN395656',\n",
            "       '3HSDZTZR3LN154022', '3HSDZTZR0LN880114', '3HSDZTZR3KN315175',\n",
            "       '3HSDZTZR1KN174476', '3HSDZTZR8KN398425', '3HSDZTZR3LN833708',\n",
            "       '3HSDZTZR4LN056083', '3HSDZTZR7LN630689', '3HSDZTZR6LN065318',\n",
            "       '3HSDZTZR6LN056148', '3HSDZTZR8LN139435', '3HSDWTZR2LN845423',\n",
            "       '3HSDZTZR5KN140444', '3HSDZTZR4LN822992', '3HSDZTZR4KN126504',\n",
            "       '3HSDWTZR5LN844783', '3HSDZTZR2LN112232', '3HSDZTZR8KN537162',\n",
            "       '3HSDZTZR9KN136686', '3HSDZTZR6KN540223', '3HSDZTZR4KN227364',\n",
            "       '3HSDZTZR1LN081863', '3HSDZTZR0KN279915', '3HSDZTZR2LN817113',\n",
            "       '3HSDZTZR0LN437333', '3HSDWTZR9LN810071', '3HSDWTZR6LN844811',\n",
            "       '3HCDZTZRXLL056502', '3HSDZTZR0LN384388', '3HSDZTZR3LN054647',\n",
            "       '3HSDWTZR3LN845348', '3HSDZTZR6LN056070', '3HSDZTZR9LN055320',\n",
            "       '3HSDWTZR7LN064378', '3HSDZTZR1LN125327', '3HSDWTZR9LN844849',\n",
            "       '3HSDZTZR6LN112511', '3HSDWTZR2LN845308', '3HSDZTZRXLN443043',\n",
            "       '3HSDZTZR1LN290908', '3HSDWTZR8LN845362', '3HSDWTZR1LN845090',\n",
            "       '3HSDZTZR2LN075067', '3HSDZTZR0LN630677', '3HSDWTZRXLN844861',\n",
            "       '3HSDZTZR0LN062656', '3HSDWTZRXKN315831', '3HSDWTZR0LN845338',\n",
            "       '3HSDZTZR0LN291791', '3HSDWTZR8KN308697', '3HSDWTZR2LN844773',\n",
            "       '3HSDZTZR4KN547798', '3HSDWTZRXLN845024', '3HSDWTZR5LN845352',\n",
            "       '3HSDWTZRXKN310631', '3HSDWTZR9LN845466', '3HCDZTZR6LL833600',\n",
            "       '3HSDWTZR5LN845092', '3HSDZTZR7KN314305', '3HCDZTZR6LL833595',\n",
            "       '3HSDWTZR5LN845108', '3HSDWTZR3KN308686', '3HSDWTZR7LN844803',\n",
            "       '3HSDZTZR0KN354211', '3HSDWTZR8LN845216', '3HSDWTZR8LN845149',\n",
            "       '3HSDZTZR2LN630681', '3HSDWTZR9LN845094', '3HSDWTZR7LN810070',\n",
            "       '3HSDWTZR9LN845337', '3HCDZTZR1LL158318', '3HSDWTZR0LN844822',\n",
            "       '3HSDWTZR9KN310622', '3HSDWTZR1LN845199', '3HSDZTZR9KN352697',\n",
            "       '3HSDZTZRXLN085409', '3HCDZTZR2LL070362', '3HCDYTZN8LL296583',\n",
            "       '3HCDZTZR1LL880100', '3HCDZTZR1ML681274', '3HSDZTZR5LN164115',\n",
            "       '3HSDWTZR3LN211277', '3HSDWTZR1LN887792', '3HSDZTZR9LN085496',\n",
            "       '3HCDZTZR6ML684705', '3HSDZTZR1LN871454', '3HSDZTZR2LN230944',\n",
            "       '3HSDWTZR2LN854882', '3HCDZTZR5LL880102', '3HCDZTZR3LL880101',\n",
            "       '3HSDZTZR3MN096740', '3HCDZTZR9LL880099', '3HSDZTZR3LN164114',\n",
            "       '3HCDZTZR7LL880103', '3HSDWTZR6LN854884', '3HCDZTZR0ML681251',\n",
            "       '3HSDZTZR3LN852131'],\n",
            "      dtype='object', name='vin')\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MmSM6dA_q1DI"
      },
      "source": [
        "top10_vins = list(dupsdf.index[:5])\n",
        "df_top10 = df[df['vin'].isin(top10_vins)]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZhbqwUrmJoQs"
      },
      "source": [
        "# Apply Hierarchical Clustering to Time Series"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qzHE5L7iq1J0",
        "outputId": "3331d2cd-2d7a-493a-d151-deb2eee26dc1"
      },
      "source": [
        "df_top10.columns"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['vin', 'msg_ts', 'fuel_total', 'engine_rpm', 'coolant_tmp', 'oil_tmpl',\n",
              "       'throt_pos', 'torque', 'oil_psi', 'manifold_tmp', 'vehiclespeed',\n",
              "       'odometerinmeters'],\n",
              "      dtype='object')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 690
        },
        "id": "DF84tSgPqvHz",
        "outputId": "1cb211bd-9187-43f7-cda5-9d703f6bfc2c"
      },
      "source": [
        "pip install sktime"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting sktime\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/ed/0b/ee4c2a9f2ef22eea4e202c4740142f3dfb8a3e5f9f1b36731b39b58ca432/sktime-0.6.0-cp37-cp37m-manylinux2014_x86_64.whl (5.7MB)\n",
            "\u001b[K     |████████████████████████████████| 5.7MB 4.9MB/s \n",
            "\u001b[?25hRequirement already satisfied: pandas>=1.1.0 in /usr/local/lib/python3.7/dist-packages (from sktime) (1.1.5)\n",
            "Collecting statsmodels>=0.12.1\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/da/69/8eef30a6237c54f3c0b524140e2975f4b1eea3489b45eb3339574fc8acee/statsmodels-0.12.2-cp37-cp37m-manylinux1_x86_64.whl (9.5MB)\n",
            "\u001b[K     |████████████████████████████████| 9.5MB 30.6MB/s \n",
            "\u001b[?25hRequirement already satisfied: numba>=0.50 in /usr/local/lib/python3.7/dist-packages (from sktime) (0.51.2)\n",
            "Requirement already satisfied: numpy>=1.19.0 in /usr/local/lib/python3.7/dist-packages (from sktime) (1.19.5)\n",
            "Collecting scikit-learn>=0.24.0\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/a8/eb/a48f25c967526b66d5f1fa7a984594f0bf0a5afafa94a8c4dbc317744620/scikit_learn-0.24.2-cp37-cp37m-manylinux2010_x86_64.whl (22.3MB)\n",
            "\u001b[K     |████████████████████████████████| 22.3MB 1.4MB/s \n",
            "\u001b[?25hRequirement already satisfied: wheel in /usr/local/lib/python3.7/dist-packages (from sktime) (0.36.2)\n",
            "Requirement already satisfied: pytz>=2017.2 in /usr/local/lib/python3.7/dist-packages (from pandas>=1.1.0->sktime) (2018.9)\n",
            "Requirement already satisfied: python-dateutil>=2.7.3 in /usr/local/lib/python3.7/dist-packages (from pandas>=1.1.0->sktime) (2.8.1)\n",
            "Requirement already satisfied: scipy>=1.1 in /usr/local/lib/python3.7/dist-packages (from statsmodels>=0.12.1->sktime) (1.4.1)\n",
            "Requirement already satisfied: patsy>=0.5 in /usr/local/lib/python3.7/dist-packages (from statsmodels>=0.12.1->sktime) (0.5.1)\n",
            "Requirement already satisfied: setuptools in /usr/local/lib/python3.7/dist-packages (from numba>=0.50->sktime) (56.1.0)\n",
            "Requirement already satisfied: llvmlite<0.35,>=0.34.0.dev0 in /usr/local/lib/python3.7/dist-packages (from numba>=0.50->sktime) (0.34.0)\n",
            "Collecting threadpoolctl>=2.0.0\n",
            "  Downloading https://files.pythonhosted.org/packages/f7/12/ec3f2e203afa394a149911729357aa48affc59c20e2c1c8297a60f33f133/threadpoolctl-2.1.0-py3-none-any.whl\n",
            "Requirement already satisfied: joblib>=0.11 in /usr/local/lib/python3.7/dist-packages (from scikit-learn>=0.24.0->sktime) (1.0.1)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.7/dist-packages (from python-dateutil>=2.7.3->pandas>=1.1.0->sktime) (1.15.0)\n",
            "Installing collected packages: statsmodels, threadpoolctl, scikit-learn, sktime\n",
            "  Found existing installation: statsmodels 0.10.2\n",
            "    Uninstalling statsmodels-0.10.2:\n",
            "      Successfully uninstalled statsmodels-0.10.2\n",
            "  Found existing installation: scikit-learn 0.22.2.post1\n",
            "    Uninstalling scikit-learn-0.22.2.post1:\n",
            "      Successfully uninstalled scikit-learn-0.22.2.post1\n",
            "Successfully installed scikit-learn-0.24.2 sktime-0.6.0 statsmodels-0.12.2 threadpoolctl-2.1.0\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.colab-display-data+json": {
              "pip_warning": {
                "packages": [
                  "statsmodels"
                ]
              }
            }
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ut5EVIt3Jtzt"
      },
      "source": [
        "from sktime.distances.elastic_cython import dtw_distance\n",
        "\n",
        "# df_col = pd.DataFrame(data = df_top10, columns = ['fuel_total', 'engine_rpm', 'coolant_tmp','oil_tmpl','throt_pos', \n",
        "#                                                        'torque', 'oil_psi', 'manifold_tmp', 'vehiclespeed','odometerinmeters'])\n",
        "df_col = pd.DataFrame(data = df_top10, columns = ['engine_rpm','coolant_tmp'])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IFKmwlYtJt32"
      },
      "source": [
        "series_list = df_col.values"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ij3nnPVKJt8M"
      },
      "source": [
        "series_list = series_list.reshape(series_list.shape[0], series_list.shape[1],1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dpFQhzJWLZXh"
      },
      "source": [
        "for i in range(len(series_list)):\n",
        "    length = len(series_list[i])\n",
        "    # series_list[i] = np.array(series_list[i]).reshape((length,2))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5KSeOjv7LHUE",
        "outputId": "0b344fb9-3aa6-4e01-e3f6-38db8e91e8df"
      },
      "source": [
        "series_list"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[[1240.],\n",
              "        [  81.]],\n",
              "\n",
              "       [[1319.],\n",
              "        [  83.]],\n",
              "\n",
              "       [[1226.],\n",
              "        [  87.]],\n",
              "\n",
              "       ...,\n",
              "\n",
              "       [[1260.],\n",
              "        [  81.]],\n",
              "\n",
              "       [[1141.],\n",
              "        [  81.]],\n",
              "\n",
              "       [[1264.],\n",
              "        [  80.]]])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 21
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "y6ymlD8pJuDU"
      },
      "source": [
        "# Initialize distance matrix\n",
        "n_series = len(series_list)\n",
        "distance_matrix = np.zeros(shape=(n_series, n_series))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "h8vqtfLki_bv",
        "outputId": "60bc0d8c-cace-4e6b-bf99-7845efe76956"
      },
      "source": [
        "print(series_list[0])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[1240.]\n",
            " [  81.]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "background_save": true
        },
        "id": "AcVJ68jgJuGx"
      },
      "source": [
        "# Build distance matrix\n",
        "for i in range(n_series):\n",
        "    for j in range(n_series):\n",
        "        x = series_list[i]\n",
        "        y = series_list[j]\n",
        "        if i != j:\n",
        "            dist = dtw_distance(x, y)\n",
        "            distance_matrix[i, j] = dist"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "background_save": true
        },
        "id": "EIEAAoRgdRNy"
      },
      "source": [
        "#Binary data\n",
        "np.save('distance_matrix.npy', distance_matrix)\n",
        "\n",
        "#Human readable data\n",
        "np.savetxt('distance_matrix.txt', distance_matrix)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "background_save": true
        },
        "id": "oX4W-1xadyVu"
      },
      "source": [
        "# distance_matrix = np.load('distance_matrix.npy')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "background_save": true
        },
        "id": "hvLzguvHqvT7"
      },
      "source": [
        "from scipy.cluster.hierarchy import single, complete, average, ward, dendrogram\n",
        "\n",
        "def hierarchical_clustering(dist_mat, method='complete'):\n",
        "    if method == 'complete':\n",
        "        Z = complete(distance_matrix)\n",
        "    if method == 'single':\n",
        "        Z = single(distance_matrix)\n",
        "    if method == 'average':\n",
        "        Z = average(distance_matrix)\n",
        "    if method == 'ward':\n",
        "        Z = ward(distance_matrix)\n",
        "    \n",
        "    fig = plt.figure(figsize=(16, 8))\n",
        "    dn = dendrogram(Z)\n",
        "    plt.title(f\"Dendrogram for {method}-linkage with correlation distance\")\n",
        "    plt.show()\n",
        "    \n",
        "    return Z"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cMr3McdecOQs"
      },
      "source": [
        "linkage_matrix = hierarchical_clustering(distance_matrix)"
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}