code file .ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "4fc8b738",
   "metadata": {},
   "source": [
    "### Objectives of this notebook\n",
    "<ul>\n",
    "    <li>Implementing both k means and hierarchical Agglomerative clustering and performing cluster analysis </li>\n",
    "    <li>Using matplotlib and seaborn and drawing plots for Exploratory Data Analysis</li>\n",
    "    <li>Using evaluation metrics for unsupervised learning and evaluating our model </li>\n",
    "    \n",
    "</ul>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4e4a8e2d",
   "metadata": {},
   "source": [
    "## Dataset(Real world):\n",
    "We will use cluster analysis to generate a big picture model of the weather at a local station using a minute-graunlarity data. In this dataset, we have in the order of millions records. How do we create clusters out of them to segment out similar weather conditions?\n",
    "\n",
    "This data comes from a weather station located in San Diego, California. The weather station is equipped with sensors that capture weather-related measurements such as air temperature, air pressure, and relative humidity. Data was collected for a period of three years, from September 2011 to September 2014, to ensure that sufficient data for different seasons and weather conditions is captured.\n",
    "\n",
    "The minute weather dataset contains raw sensor measurements captured at one-minute intervals.\n",
    "Each row, or sample, consists of the following variables:\n",
    "\n",
    "* **rowID:** \tunique number for each row\t(*Unit: NA*)\n",
    "* **hpwren_timestamp:**\ttimestamp of measure\t(*Unit: year-month-day hour:minute:second*)\n",
    "* **air_pressure:** air pressure measured at the timestamp\t(*Unit: hectopascals*)\n",
    "* **air_temp:**\tair temperature measure at the timestamp\t(*Unit: degrees Fahrenheit*)\n",
    "* **avg_wind_direction:**\twind direction averaged over the minute before the timestamp\t(*Unit: degrees, with 0 means coming from the North, and increasing clockwise*)\n",
    "* **avg_wind_speed:**\twind speed averaged over the minute before the timestamp\t(*Unit: meters per second*)\n",
    "* **max_wind_direction:**\thighest wind direction in the minute before the timestamp\t(*Unit: degrees, with 0 being North and increasing clockwise*)\n",
    "* **max_wind_speed:**\thighest wind speed in the minute before the timestamp\t(*Unit: meters per second*)\n",
    "* **min_wind_direction:**\tsmallest wind direction in the minute before the timestamp\t(*Unit: degrees, with 0 being North and inceasing clockwise*)\n",
    "* **min_wind_speed:**\tsmallest wind speed in the minute before the timestamp\t(*Unit: meters per second*)\n",
    "* **rain_accumulation:**\tamount of accumulated rain measured at the timestamp\t(*Unit: millimeters*)\n",
    "* **rain_duration:**\tlength of time rain has fallen as measured at the timestamp\t(*Unit: seconds*)\n",
    "* **relative_humidity:**\trelative humidity measured at the timestamp\t(*Unit: percent*)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "c16466c5",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Import necessary libraries\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn.cluster import KMeans,AgglomerativeClustering\n",
    "from sklearn.metrics import silhouette_score,davies_bouldin_score\n",
    "from sklearn.preprocessing import StandardScaler,normalize,RobustScaler,MinMaxScaler\n",
    "import scipy.cluster.hierarchy as shc\n",
    "import collections\n",
    "from sklearn.decomposition import PCA, KernelPCA,TruncatedSVD\n",
    "%matplotlib inline "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d0728f91",
   "metadata": {},
   "source": [
    "### Read the dataset and display some rows"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "4177190d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>rowID</th>\n",
       "      <th>hpwren_timestamp</th>\n",
       "      <th>air_pressure</th>\n",
       "      <th>air_temp</th>\n",
       "      <th>avg_wind_direction</th>\n",
       "      <th>avg_wind_speed</th>\n",
       "      <th>max_wind_direction</th>\n",
       "      <th>max_wind_speed</th>\n",
       "      <th>min_wind_direction</th>\n",
       "      <th>min_wind_speed</th>\n",
       "      <th>rain_accumulation</th>\n",
       "      <th>rain_duration</th>\n",
       "      <th>relative_humidity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>2011-09-10 00:00:49</td>\n",
       "      <td>912.3</td>\n",
       "      <td>64.76</td>\n",
       "      <td>97.0</td>\n",
       "      <td>1.2</td>\n",
       "      <td>106.0</td>\n",
       "      <td>1.6</td>\n",
       "      <td>85.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>60.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>2011-09-10 00:01:49</td>\n",
       "      <td>912.3</td>\n",
       "      <td>63.86</td>\n",
       "      <td>161.0</td>\n",
       "      <td>0.8</td>\n",
       "      <td>215.0</td>\n",
       "      <td>1.5</td>\n",
       "      <td>43.0</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>39.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>2011-09-10 00:02:49</td>\n",
       "      <td>912.3</td>\n",
       "      <td>64.22</td>\n",
       "      <td>77.0</td>\n",
       "      <td>0.7</td>\n",
       "      <td>143.0</td>\n",
       "      <td>1.2</td>\n",
       "      <td>324.0</td>\n",
       "      <td>0.3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>43.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>2011-09-10 00:03:49</td>\n",
       "      <td>912.3</td>\n",
       "      <td>64.40</td>\n",
       "      <td>89.0</td>\n",
       "      <td>1.2</td>\n",
       "      <td>112.0</td>\n",
       "      <td>1.6</td>\n",
       "      <td>12.0</td>\n",
       "      <td>0.7</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>49.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>2011-09-10 00:04:49</td>\n",
       "      <td>912.3</td>\n",
       "      <td>64.40</td>\n",
       "      <td>185.0</td>\n",
       "      <td>0.4</td>\n",
       "      <td>260.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>0.1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>58.8</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",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1587252</th>\n",
       "      <td>1587252</td>\n",
       "      <td>2014-09-10 23:55:29</td>\n",
       "      <td>915.9</td>\n",
       "      <td>74.66</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>11.0</td>\n",
       "      <td>2.1</td>\n",
       "      <td>357.0</td>\n",
       "      <td>1.6</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>46.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1587253</th>\n",
       "      <td>1587253</td>\n",
       "      <td>2014-09-10 23:56:29</td>\n",
       "      <td>915.9</td>\n",
       "      <td>74.66</td>\n",
       "      <td>9.0</td>\n",
       "      <td>1.6</td>\n",
       "      <td>19.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>353.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>46.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1587254</th>\n",
       "      <td>1587254</td>\n",
       "      <td>2014-09-10 23:57:29</td>\n",
       "      <td>915.9</td>\n",
       "      <td>74.66</td>\n",
       "      <td>15.0</td>\n",
       "      <td>1.7</td>\n",
       "      <td>42.0</td>\n",
       "      <td>2.3</td>\n",
       "      <td>353.0</td>\n",
       "      <td>1.2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>46.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1587255</th>\n",
       "      <td>1587255</td>\n",
       "      <td>2014-09-10 23:58:29</td>\n",
       "      <td>915.9</td>\n",
       "      <td>74.66</td>\n",
       "      <td>8.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>24.0</td>\n",
       "      <td>2.2</td>\n",
       "      <td>355.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>46.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1587256</th>\n",
       "      <td>1587256</td>\n",
       "      <td>2014-09-10 23:59:29</td>\n",
       "      <td>915.9</td>\n",
       "      <td>74.66</td>\n",
       "      <td>12.0</td>\n",
       "      <td>1.9</td>\n",
       "      <td>21.0</td>\n",
       "      <td>2.2</td>\n",
       "      <td>357.0</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>46.3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1587257 rows × 13 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           rowID     hpwren_timestamp  air_pressure  air_temp  \\\n",
       "0              0  2011-09-10 00:00:49         912.3     64.76   \n",
       "1              1  2011-09-10 00:01:49         912.3     63.86   \n",
       "2              2  2011-09-10 00:02:49         912.3     64.22   \n",
       "3              3  2011-09-10 00:03:49         912.3     64.40   \n",
       "4              4  2011-09-10 00:04:49         912.3     64.40   \n",
       "...          ...                  ...           ...       ...   \n",
       "1587252  1587252  2014-09-10 23:55:29         915.9     74.66   \n",
       "1587253  1587253  2014-09-10 23:56:29         915.9     74.66   \n",
       "1587254  1587254  2014-09-10 23:57:29         915.9     74.66   \n",
       "1587255  1587255  2014-09-10 23:58:29         915.9     74.66   \n",
       "1587256  1587256  2014-09-10 23:59:29         915.9     74.66   \n",
       "\n",
       "         avg_wind_direction  avg_wind_speed  max_wind_direction  \\\n",
       "0                      97.0             1.2               106.0   \n",
       "1                     161.0             0.8               215.0   \n",
       "2                      77.0             0.7               143.0   \n",
       "3                      89.0             1.2               112.0   \n",
       "4                     185.0             0.4               260.0   \n",
       "...                     ...             ...                 ...   \n",
       "1587252                 1.0             1.8                11.0   \n",
       "1587253                 9.0             1.6                19.0   \n",
       "1587254                15.0             1.7                42.0   \n",
       "1587255                 8.0             1.8                24.0   \n",
       "1587256                12.0             1.9                21.0   \n",
       "\n",
       "         max_wind_speed  min_wind_direction  min_wind_speed  \\\n",
       "0                   1.6                85.0             1.0   \n",
       "1                   1.5                43.0             0.2   \n",
       "2                   1.2               324.0             0.3   \n",
       "3                   1.6                12.0             0.7   \n",
       "4                   1.0               100.0             0.1   \n",
       "...                 ...                 ...             ...   \n",
       "1587252             2.1               357.0             1.6   \n",
       "1587253             1.8               353.0             1.4   \n",
       "1587254             2.3               353.0             1.2   \n",
       "1587255             2.2               355.0             1.4   \n",
       "1587256             2.2               357.0             1.5   \n",
       "\n",
       "         rain_accumulation  rain_duration  relative_humidity  \n",
       "0                      NaN            NaN               60.5  \n",
       "1                      0.0            0.0               39.9  \n",
       "2                      0.0            0.0               43.0  \n",
       "3                      0.0            0.0               49.5  \n",
       "4                      0.0            0.0               58.8  \n",
       "...                    ...            ...                ...  \n",
       "1587252                0.0            0.0               46.3  \n",
       "1587253                0.0            0.0               46.2  \n",
       "1587254                0.0            0.0               46.3  \n",
       "1587255                0.0            0.0               46.2  \n",
       "1587256                0.0            0.0               46.3  \n",
       "\n",
       "[1587257 rows x 13 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df =pd.read_csv('minute_weather.csv')\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "beec4afe",
   "metadata": {},
   "source": [
    "### Data Preprocessing:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "6a1abd27",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>rowID</th>\n",
       "      <th>hpwren_timestamp</th>\n",
       "      <th>air_pressure</th>\n",
       "      <th>air_temp</th>\n",
       "      <th>avg_wind_direction</th>\n",
       "      <th>avg_wind_speed</th>\n",
       "      <th>max_wind_direction</th>\n",
       "      <th>max_wind_speed</th>\n",
       "      <th>min_wind_direction</th>\n",
       "      <th>min_wind_speed</th>\n",
       "      <th>relative_humidity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>2011-09-10 00:00:49</td>\n",
       "      <td>912.3</td>\n",
       "      <td>64.76</td>\n",
       "      <td>97.0</td>\n",
       "      <td>1.2</td>\n",
       "      <td>106.0</td>\n",
       "      <td>1.6</td>\n",
       "      <td>85.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>60.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>2011-09-10 00:01:49</td>\n",
       "      <td>912.3</td>\n",
       "      <td>63.86</td>\n",
       "      <td>161.0</td>\n",
       "      <td>0.8</td>\n",
       "      <td>215.0</td>\n",
       "      <td>1.5</td>\n",
       "      <td>43.0</td>\n",
       "      <td>0.2</td>\n",
       "      <td>39.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>2011-09-10 00:02:49</td>\n",
       "      <td>912.3</td>\n",
       "      <td>64.22</td>\n",
       "      <td>77.0</td>\n",
       "      <td>0.7</td>\n",
       "      <td>143.0</td>\n",
       "      <td>1.2</td>\n",
       "      <td>324.0</td>\n",
       "      <td>0.3</td>\n",
       "      <td>43.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>2011-09-10 00:03:49</td>\n",
       "      <td>912.3</td>\n",
       "      <td>64.40</td>\n",
       "      <td>89.0</td>\n",
       "      <td>1.2</td>\n",
       "      <td>112.0</td>\n",
       "      <td>1.6</td>\n",
       "      <td>12.0</td>\n",
       "      <td>0.7</td>\n",
       "      <td>49.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>2011-09-10 00:04:49</td>\n",
       "      <td>912.3</td>\n",
       "      <td>64.40</td>\n",
       "      <td>185.0</td>\n",
       "      <td>0.4</td>\n",
       "      <td>260.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>0.1</td>\n",
       "      <td>58.8</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",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1587252</th>\n",
       "      <td>1587252</td>\n",
       "      <td>2014-09-10 23:55:29</td>\n",
       "      <td>915.9</td>\n",
       "      <td>74.66</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>11.0</td>\n",
       "      <td>2.1</td>\n",
       "      <td>357.0</td>\n",
       "      <td>1.6</td>\n",
       "      <td>46.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1587253</th>\n",
       "      <td>1587253</td>\n",
       "      <td>2014-09-10 23:56:29</td>\n",
       "      <td>915.9</td>\n",
       "      <td>74.66</td>\n",
       "      <td>9.0</td>\n",
       "      <td>1.6</td>\n",
       "      <td>19.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>353.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>46.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1587254</th>\n",
       "      <td>1587254</td>\n",
       "      <td>2014-09-10 23:57:29</td>\n",
       "      <td>915.9</td>\n",
       "      <td>74.66</td>\n",
       "      <td>15.0</td>\n",
       "      <td>1.7</td>\n",
       "      <td>42.0</td>\n",
       "      <td>2.3</td>\n",
       "      <td>353.0</td>\n",
       "      <td>1.2</td>\n",
       "      <td>46.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1587255</th>\n",
       "      <td>1587255</td>\n",
       "      <td>2014-09-10 23:58:29</td>\n",
       "      <td>915.9</td>\n",
       "      <td>74.66</td>\n",
       "      <td>8.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>24.0</td>\n",
       "      <td>2.2</td>\n",
       "      <td>355.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>46.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1587256</th>\n",
       "      <td>1587256</td>\n",
       "      <td>2014-09-10 23:59:29</td>\n",
       "      <td>915.9</td>\n",
       "      <td>74.66</td>\n",
       "      <td>12.0</td>\n",
       "      <td>1.9</td>\n",
       "      <td>21.0</td>\n",
       "      <td>2.2</td>\n",
       "      <td>357.0</td>\n",
       "      <td>1.5</td>\n",
       "      <td>46.3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1587257 rows × 11 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           rowID     hpwren_timestamp  air_pressure  air_temp  \\\n",
       "0              0  2011-09-10 00:00:49         912.3     64.76   \n",
       "1              1  2011-09-10 00:01:49         912.3     63.86   \n",
       "2              2  2011-09-10 00:02:49         912.3     64.22   \n",
       "3              3  2011-09-10 00:03:49         912.3     64.40   \n",
       "4              4  2011-09-10 00:04:49         912.3     64.40   \n",
       "...          ...                  ...           ...       ...   \n",
       "1587252  1587252  2014-09-10 23:55:29         915.9     74.66   \n",
       "1587253  1587253  2014-09-10 23:56:29         915.9     74.66   \n",
       "1587254  1587254  2014-09-10 23:57:29         915.9     74.66   \n",
       "1587255  1587255  2014-09-10 23:58:29         915.9     74.66   \n",
       "1587256  1587256  2014-09-10 23:59:29         915.9     74.66   \n",
       "\n",
       "         avg_wind_direction  avg_wind_speed  max_wind_direction  \\\n",
       "0                      97.0             1.2               106.0   \n",
       "1                     161.0             0.8               215.0   \n",
       "2                      77.0             0.7               143.0   \n",
       "3                      89.0             1.2               112.0   \n",
       "4                     185.0             0.4               260.0   \n",
       "...                     ...             ...                 ...   \n",
       "1587252                 1.0             1.8                11.0   \n",
       "1587253                 9.0             1.6                19.0   \n",
       "1587254                15.0             1.7                42.0   \n",
       "1587255                 8.0             1.8                24.0   \n",
       "1587256                12.0             1.9                21.0   \n",
       "\n",
       "         max_wind_speed  min_wind_direction  min_wind_speed  relative_humidity  \n",
       "0                   1.6                85.0             1.0               60.5  \n",
       "1                   1.5                43.0             0.2               39.9  \n",
       "2                   1.2               324.0             0.3               43.0  \n",
       "3                   1.6                12.0             0.7               49.5  \n",
       "4                   1.0               100.0             0.1               58.8  \n",
       "...                 ...                 ...             ...                ...  \n",
       "1587252             2.1               357.0             1.6               46.3  \n",
       "1587253             1.8               353.0             1.4               46.2  \n",
       "1587254             2.3               353.0             1.2               46.3  \n",
       "1587255             2.2               355.0             1.4               46.2  \n",
       "1587256             2.2               357.0             1.5               46.3  \n",
       "\n",
       "[1587257 rows x 11 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#We drop rain_duration and rain_accumulation columns because they are mostly zero\n",
    "df.drop(columns = ['rain_duration','rain_accumulation'],inplace = True)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "27abf4c8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 1587257 entries, 0 to 1587256\n",
      "Data columns (total 11 columns):\n",
      " #   Column              Non-Null Count    Dtype  \n",
      "---  ------              --------------    -----  \n",
      " 0   rowID               1587257 non-null  int64  \n",
      " 1   hpwren_timestamp    1587257 non-null  object \n",
      " 2   air_pressure        1587257 non-null  float64\n",
      " 3   air_temp            1587257 non-null  float64\n",
      " 4   avg_wind_direction  1586824 non-null  float64\n",
      " 5   avg_wind_speed      1586824 non-null  float64\n",
      " 6   max_wind_direction  1586824 non-null  float64\n",
      " 7   max_wind_speed      1586824 non-null  float64\n",
      " 8   min_wind_direction  1586824 non-null  float64\n",
      " 9   min_wind_speed      1586824 non-null  float64\n",
      " 10  relative_humidity   1587257 non-null  float64\n",
      "dtypes: float64(9), int64(1), object(1)\n",
      "memory usage: 133.2+ MB\n"
     ]
    }
   ],
   "source": [
    "df.info() #Lets inspect our dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "e9c301f3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>rowID</th>\n",
       "      <th>air_pressure</th>\n",
       "      <th>air_temp</th>\n",
       "      <th>avg_wind_direction</th>\n",
       "      <th>avg_wind_speed</th>\n",
       "      <th>max_wind_direction</th>\n",
       "      <th>max_wind_speed</th>\n",
       "      <th>min_wind_direction</th>\n",
       "      <th>min_wind_speed</th>\n",
       "      <th>relative_humidity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1.587257e+06</td>\n",
       "      <td>1.587257e+06</td>\n",
       "      <td>1.587257e+06</td>\n",
       "      <td>1.586824e+06</td>\n",
       "      <td>1.586824e+06</td>\n",
       "      <td>1.586824e+06</td>\n",
       "      <td>1.586824e+06</td>\n",
       "      <td>1.586824e+06</td>\n",
       "      <td>1.586824e+06</td>\n",
       "      <td>1.587257e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>7.936280e+05</td>\n",
       "      <td>9.168301e+02</td>\n",
       "      <td>6.185144e+01</td>\n",
       "      <td>1.619654e+02</td>\n",
       "      <td>2.774272e+00</td>\n",
       "      <td>1.634030e+02</td>\n",
       "      <td>3.399813e+00</td>\n",
       "      <td>1.668264e+02</td>\n",
       "      <td>2.133130e+00</td>\n",
       "      <td>4.760837e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>4.582018e+05</td>\n",
       "      <td>3.051593e+00</td>\n",
       "      <td>1.183362e+01</td>\n",
       "      <td>9.520812e+01</td>\n",
       "      <td>2.060758e+00</td>\n",
       "      <td>9.236723e+01</td>\n",
       "      <td>2.423167e+00</td>\n",
       "      <td>9.746275e+01</td>\n",
       "      <td>1.745345e+00</td>\n",
       "      <td>2.621454e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>9.050000e+02</td>\n",
       "      <td>3.164000e+01</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>1.000000e-01</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>7.000000e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>3.968140e+05</td>\n",
       "      <td>9.148000e+02</td>\n",
       "      <td>5.270000e+01</td>\n",
       "      <td>6.200000e+01</td>\n",
       "      <td>1.300000e+00</td>\n",
       "      <td>6.800000e+01</td>\n",
       "      <td>1.600000e+00</td>\n",
       "      <td>7.700000e+01</td>\n",
       "      <td>8.000000e-01</td>\n",
       "      <td>2.470000e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>7.936280e+05</td>\n",
       "      <td>9.167000e+02</td>\n",
       "      <td>6.224000e+01</td>\n",
       "      <td>1.820000e+02</td>\n",
       "      <td>2.200000e+00</td>\n",
       "      <td>1.870000e+02</td>\n",
       "      <td>2.700000e+00</td>\n",
       "      <td>1.800000e+02</td>\n",
       "      <td>1.600000e+00</td>\n",
       "      <td>4.470000e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1.190442e+06</td>\n",
       "      <td>9.187000e+02</td>\n",
       "      <td>7.088000e+01</td>\n",
       "      <td>2.170000e+02</td>\n",
       "      <td>3.800000e+00</td>\n",
       "      <td>2.230000e+02</td>\n",
       "      <td>4.600000e+00</td>\n",
       "      <td>2.120000e+02</td>\n",
       "      <td>3.000000e+00</td>\n",
       "      <td>6.800000e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.587256e+06</td>\n",
       "      <td>9.295000e+02</td>\n",
       "      <td>9.950000e+01</td>\n",
       "      <td>3.590000e+02</td>\n",
       "      <td>3.230000e+01</td>\n",
       "      <td>3.590000e+02</td>\n",
       "      <td>3.600000e+01</td>\n",
       "      <td>3.590000e+02</td>\n",
       "      <td>3.200000e+01</td>\n",
       "      <td>9.300000e+01</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              rowID  air_pressure      air_temp  avg_wind_direction  \\\n",
       "count  1.587257e+06  1.587257e+06  1.587257e+06        1.586824e+06   \n",
       "mean   7.936280e+05  9.168301e+02  6.185144e+01        1.619654e+02   \n",
       "std    4.582018e+05  3.051593e+00  1.183362e+01        9.520812e+01   \n",
       "min    0.000000e+00  9.050000e+02  3.164000e+01        0.000000e+00   \n",
       "25%    3.968140e+05  9.148000e+02  5.270000e+01        6.200000e+01   \n",
       "50%    7.936280e+05  9.167000e+02  6.224000e+01        1.820000e+02   \n",
       "75%    1.190442e+06  9.187000e+02  7.088000e+01        2.170000e+02   \n",
       "max    1.587256e+06  9.295000e+02  9.950000e+01        3.590000e+02   \n",
       "\n",
       "       avg_wind_speed  max_wind_direction  max_wind_speed  min_wind_direction  \\\n",
       "count    1.586824e+06        1.586824e+06    1.586824e+06        1.586824e+06   \n",
       "mean     2.774272e+00        1.634030e+02    3.399813e+00        1.668264e+02   \n",
       "std      2.060758e+00        9.236723e+01    2.423167e+00        9.746275e+01   \n",
       "min      0.000000e+00        0.000000e+00    1.000000e-01        0.000000e+00   \n",
       "25%      1.300000e+00        6.800000e+01    1.600000e+00        7.700000e+01   \n",
       "50%      2.200000e+00        1.870000e+02    2.700000e+00        1.800000e+02   \n",
       "75%      3.800000e+00        2.230000e+02    4.600000e+00        2.120000e+02   \n",
       "max      3.230000e+01        3.590000e+02    3.600000e+01        3.590000e+02   \n",
       "\n",
       "       min_wind_speed  relative_humidity  \n",
       "count    1.586824e+06       1.587257e+06  \n",
       "mean     2.133130e+00       4.760837e+01  \n",
       "std      1.745345e+00       2.621454e+01  \n",
       "min      0.000000e+00       7.000000e-01  \n",
       "25%      8.000000e-01       2.470000e+01  \n",
       "50%      1.600000e+00       4.470000e+01  \n",
       "75%      3.000000e+00       6.800000e+01  \n",
       "max      3.200000e+01       9.300000e+01  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()#Lets find the distribution of the dataframe.This method only describes about series of dtype int,float"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9b5932e2",
   "metadata": {},
   "source": [
    "### Heatmap to identify highly correlated variables:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "22c6cb3a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,8))\n",
    "sns.heatmap(df.corr(),\n",
    "            annot=True,\n",
    "            linewidths=.5,\n",
    "            center=0,\n",
    "            cbar=False,\n",
    "            cmap=\"YlGnBu\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "474e2ba8",
   "metadata": {},
   "source": [
    "We can see that min_wind_speed,max_wind_speed, avg_wind_speed are very highly correlated varibles (having degree of correlation as 0.97 and 0.93) hence we can use only avg_wind_speed which has very high correlation with max and min avg wind speeds for features of our interest. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6a5297f8",
   "metadata": {},
   "source": [
    "### Find duplicates present in dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "e3b19173",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0          False\n",
       "1          False\n",
       "2          False\n",
       "3          False\n",
       "4          False\n",
       "           ...  \n",
       "1587252    False\n",
       "1587253    False\n",
       "1587254    False\n",
       "1587255    False\n",
       "1587256    False\n",
       "Length: 1587257, dtype: bool"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.duplicated()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "808f6e6c",
   "metadata": {},
   "source": [
    "We can see that there are no duplicates present in our dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d57a283b",
   "metadata": {},
   "source": [
    "### Detecting outliers by IQR method:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "feaccef2",
   "metadata": {},
   "outputs": [],
   "source": [
    "p0 = df.relative_humidity.min()\n",
    "p100 = df.relative_humidity.max()\n",
    "q1 = df.relative_humidity.quantile(0.25)\n",
    "q2 = df.relative_humidity.quantile(0.5)\n",
    "q3 = df.relative_humidity.quantile(0.75)\n",
    "iqr = q3-q1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "9721d345",
   "metadata": {},
   "outputs": [],
   "source": [
    "lc = q1 - 1.5*iqr\n",
    "uc = q3 + 1.5*iqr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "0ab53334",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "p0 =  0.7 , p100 =  93.0 , lc =  -40.249999999999986 , uc =  132.95\n"
     ]
    }
   ],
   "source": [
    "print( \"p0 = \" , p0 ,\", p100 = \" , p100 ,\", lc = \" , lc ,\", uc = \" , uc)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cea6db4e",
   "metadata": {},
   "source": [
    "If lc < p0 → There are NO Outliers on the lower side\n",
    "\n",
    "If uc > p100 → There are NO Outliers on the higher side\n",
    "\n",
    "We can observe that both the conditions are satisfied by our dataset. So, there are no outliers present in our dataset.\n",
    "\n",
    "We can also get the pictorial representation and detect any outliers present using Box Plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "9426bff1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.relative_humidity.plot(kind ='box')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "56b496af",
   "metadata": {},
   "source": [
    "### Missing values treatment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "a6a07a24",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "rowID                   0\n",
       "hpwren_timestamp        0\n",
       "air_pressure            0\n",
       "air_temp                0\n",
       "avg_wind_direction    433\n",
       "avg_wind_speed        433\n",
       "max_wind_direction    433\n",
       "max_wind_speed        433\n",
       "min_wind_direction    433\n",
       "min_wind_speed        433\n",
       "relative_humidity       0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "43b6ec9b",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "rowID                 0.00000\n",
      "hpwren_timestamp      0.00000\n",
      "air_pressure          0.00000\n",
      "air_temp              0.00000\n",
      "avg_wind_direction    0.02728\n",
      "avg_wind_speed        0.02728\n",
      "max_wind_direction    0.02728\n",
      "max_wind_speed        0.02728\n",
      "min_wind_direction    0.02728\n",
      "min_wind_speed        0.02728\n",
      "relative_humidity     0.00000\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "print((df.isnull().sum()/df.shape[0])*100.0)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9b0693e6",
   "metadata": {},
   "source": [
    "We can observe that avg_wind_direction,avg_wind_speed,max_wind_direction,max_wind_speed have null values(about 0.027%).As we have very less percentage of null values we can simply drop them."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "72ac640d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>rowID</th>\n",
       "      <th>hpwren_timestamp</th>\n",
       "      <th>air_pressure</th>\n",
       "      <th>air_temp</th>\n",
       "      <th>avg_wind_direction</th>\n",
       "      <th>avg_wind_speed</th>\n",
       "      <th>max_wind_direction</th>\n",
       "      <th>max_wind_speed</th>\n",
       "      <th>min_wind_direction</th>\n",
       "      <th>min_wind_speed</th>\n",
       "      <th>relative_humidity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>2011-09-10 00:00:49</td>\n",
       "      <td>912.3</td>\n",
       "      <td>64.76</td>\n",
       "      <td>97.0</td>\n",
       "      <td>1.2</td>\n",
       "      <td>106.0</td>\n",
       "      <td>1.6</td>\n",
       "      <td>85.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>60.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>2011-09-10 00:01:49</td>\n",
       "      <td>912.3</td>\n",
       "      <td>63.86</td>\n",
       "      <td>161.0</td>\n",
       "      <td>0.8</td>\n",
       "      <td>215.0</td>\n",
       "      <td>1.5</td>\n",
       "      <td>43.0</td>\n",
       "      <td>0.2</td>\n",
       "      <td>39.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>2011-09-10 00:02:49</td>\n",
       "      <td>912.3</td>\n",
       "      <td>64.22</td>\n",
       "      <td>77.0</td>\n",
       "      <td>0.7</td>\n",
       "      <td>143.0</td>\n",
       "      <td>1.2</td>\n",
       "      <td>324.0</td>\n",
       "      <td>0.3</td>\n",
       "      <td>43.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>2011-09-10 00:03:49</td>\n",
       "      <td>912.3</td>\n",
       "      <td>64.40</td>\n",
       "      <td>89.0</td>\n",
       "      <td>1.2</td>\n",
       "      <td>112.0</td>\n",
       "      <td>1.6</td>\n",
       "      <td>12.0</td>\n",
       "      <td>0.7</td>\n",
       "      <td>49.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>2011-09-10 00:04:49</td>\n",
       "      <td>912.3</td>\n",
       "      <td>64.40</td>\n",
       "      <td>185.0</td>\n",
       "      <td>0.4</td>\n",
       "      <td>260.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>0.1</td>\n",
       "      <td>58.8</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",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1587252</th>\n",
       "      <td>1587252</td>\n",
       "      <td>2014-09-10 23:55:29</td>\n",
       "      <td>915.9</td>\n",
       "      <td>74.66</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>11.0</td>\n",
       "      <td>2.1</td>\n",
       "      <td>357.0</td>\n",
       "      <td>1.6</td>\n",
       "      <td>46.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1587253</th>\n",
       "      <td>1587253</td>\n",
       "      <td>2014-09-10 23:56:29</td>\n",
       "      <td>915.9</td>\n",
       "      <td>74.66</td>\n",
       "      <td>9.0</td>\n",
       "      <td>1.6</td>\n",
       "      <td>19.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>353.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>46.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1587254</th>\n",
       "      <td>1587254</td>\n",
       "      <td>2014-09-10 23:57:29</td>\n",
       "      <td>915.9</td>\n",
       "      <td>74.66</td>\n",
       "      <td>15.0</td>\n",
       "      <td>1.7</td>\n",
       "      <td>42.0</td>\n",
       "      <td>2.3</td>\n",
       "      <td>353.0</td>\n",
       "      <td>1.2</td>\n",
       "      <td>46.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1587255</th>\n",
       "      <td>1587255</td>\n",
       "      <td>2014-09-10 23:58:29</td>\n",
       "      <td>915.9</td>\n",
       "      <td>74.66</td>\n",
       "      <td>8.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>24.0</td>\n",
       "      <td>2.2</td>\n",
       "      <td>355.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>46.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1587256</th>\n",
       "      <td>1587256</td>\n",
       "      <td>2014-09-10 23:59:29</td>\n",
       "      <td>915.9</td>\n",
       "      <td>74.66</td>\n",
       "      <td>12.0</td>\n",
       "      <td>1.9</td>\n",
       "      <td>21.0</td>\n",
       "      <td>2.2</td>\n",
       "      <td>357.0</td>\n",
       "      <td>1.5</td>\n",
       "      <td>46.3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1586824 rows × 11 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           rowID     hpwren_timestamp  air_pressure  air_temp  \\\n",
       "0              0  2011-09-10 00:00:49         912.3     64.76   \n",
       "1              1  2011-09-10 00:01:49         912.3     63.86   \n",
       "2              2  2011-09-10 00:02:49         912.3     64.22   \n",
       "3              3  2011-09-10 00:03:49         912.3     64.40   \n",
       "4              4  2011-09-10 00:04:49         912.3     64.40   \n",
       "...          ...                  ...           ...       ...   \n",
       "1587252  1587252  2014-09-10 23:55:29         915.9     74.66   \n",
       "1587253  1587253  2014-09-10 23:56:29         915.9     74.66   \n",
       "1587254  1587254  2014-09-10 23:57:29         915.9     74.66   \n",
       "1587255  1587255  2014-09-10 23:58:29         915.9     74.66   \n",
       "1587256  1587256  2014-09-10 23:59:29         915.9     74.66   \n",
       "\n",
       "         avg_wind_direction  avg_wind_speed  max_wind_direction  \\\n",
       "0                      97.0             1.2               106.0   \n",
       "1                     161.0             0.8               215.0   \n",
       "2                      77.0             0.7               143.0   \n",
       "3                      89.0             1.2               112.0   \n",
       "4                     185.0             0.4               260.0   \n",
       "...                     ...             ...                 ...   \n",
       "1587252                 1.0             1.8                11.0   \n",
       "1587253                 9.0             1.6                19.0   \n",
       "1587254                15.0             1.7                42.0   \n",
       "1587255                 8.0             1.8                24.0   \n",
       "1587256                12.0             1.9                21.0   \n",
       "\n",
       "         max_wind_speed  min_wind_direction  min_wind_speed  relative_humidity  \n",
       "0                   1.6                85.0             1.0               60.5  \n",
       "1                   1.5                43.0             0.2               39.9  \n",
       "2                   1.2               324.0             0.3               43.0  \n",
       "3                   1.6                12.0             0.7               49.5  \n",
       "4                   1.0               100.0             0.1               58.8  \n",
       "...                 ...                 ...             ...                ...  \n",
       "1587252             2.1               357.0             1.6               46.3  \n",
       "1587253             1.8               353.0             1.4               46.2  \n",
       "1587254             2.3               353.0             1.2               46.3  \n",
       "1587255             2.2               355.0             1.4               46.2  \n",
       "1587256             2.2               357.0             1.5               46.3  \n",
       "\n",
       "[1586824 rows x 11 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.dropna(inplace =True)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "9999ca09",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 1586824 entries, 0 to 1587256\n",
      "Data columns (total 11 columns):\n",
      " #   Column              Non-Null Count    Dtype  \n",
      "---  ------              --------------    -----  \n",
      " 0   rowID               1586824 non-null  int64  \n",
      " 1   hpwren_timestamp    1586824 non-null  object \n",
      " 2   air_pressure        1586824 non-null  float64\n",
      " 3   air_temp            1586824 non-null  float64\n",
      " 4   avg_wind_direction  1586824 non-null  float64\n",
      " 5   avg_wind_speed      1586824 non-null  float64\n",
      " 6   max_wind_direction  1586824 non-null  float64\n",
      " 7   max_wind_speed      1586824 non-null  float64\n",
      " 8   min_wind_direction  1586824 non-null  float64\n",
      " 9   min_wind_speed      1586824 non-null  float64\n",
      " 10  relative_humidity   1586824 non-null  float64\n",
      "dtypes: float64(9), int64(1), object(1)\n",
      "memory usage: 145.3+ MB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "93757ed9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1586824"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3f1ab5bf",
   "metadata": {},
   "source": [
    "### Down sampling:\n",
    "The dataset is very large hence we can downsample and only take some datapoints among all present.\n",
    "To keep memory manageable and to run HAC faster let us take only some part of the dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "c26fc9d4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>rowID</th>\n",
       "      <th>hpwren_timestamp</th>\n",
       "      <th>air_pressure</th>\n",
       "      <th>air_temp</th>\n",
       "      <th>avg_wind_direction</th>\n",
       "      <th>avg_wind_speed</th>\n",
       "      <th>max_wind_direction</th>\n",
       "      <th>max_wind_speed</th>\n",
       "      <th>min_wind_direction</th>\n",
       "      <th>min_wind_speed</th>\n",
       "      <th>relative_humidity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>321</th>\n",
       "      <td>321</td>\n",
       "      <td>2011-09-10 05:21:49</td>\n",
       "      <td>911.4</td>\n",
       "      <td>64.94</td>\n",
       "      <td>82.0</td>\n",
       "      <td>0.7</td>\n",
       "      <td>95.0</td>\n",
       "      <td>0.9</td>\n",
       "      <td>63.0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>44.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>467</th>\n",
       "      <td>467</td>\n",
       "      <td>2011-09-10 07:47:49</td>\n",
       "      <td>912.2</td>\n",
       "      <td>53.96</td>\n",
       "      <td>178.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>190.0</td>\n",
       "      <td>1.2</td>\n",
       "      <td>163.0</td>\n",
       "      <td>0.6</td>\n",
       "      <td>90.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>540</th>\n",
       "      <td>540</td>\n",
       "      <td>2011-09-10 09:00:49</td>\n",
       "      <td>912.5</td>\n",
       "      <td>55.22</td>\n",
       "      <td>61.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>71.0</td>\n",
       "      <td>2.3</td>\n",
       "      <td>44.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>86.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>818</th>\n",
       "      <td>818</td>\n",
       "      <td>2011-09-10 13:38:49</td>\n",
       "      <td>913.2</td>\n",
       "      <td>64.22</td>\n",
       "      <td>237.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>254.0</td>\n",
       "      <td>4.2</td>\n",
       "      <td>221.0</td>\n",
       "      <td>1.2</td>\n",
       "      <td>62.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1129</th>\n",
       "      <td>1129</td>\n",
       "      <td>2011-09-10 18:49:49</td>\n",
       "      <td>913.4</td>\n",
       "      <td>57.74</td>\n",
       "      <td>207.0</td>\n",
       "      <td>2.7</td>\n",
       "      <td>221.0</td>\n",
       "      <td>3.1</td>\n",
       "      <td>196.0</td>\n",
       "      <td>2.2</td>\n",
       "      <td>80.5</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",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1585629</th>\n",
       "      <td>1585629</td>\n",
       "      <td>2014-09-09 20:52:29</td>\n",
       "      <td>916.1</td>\n",
       "      <td>68.18</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>2.3</td>\n",
       "      <td>357.0</td>\n",
       "      <td>1.5</td>\n",
       "      <td>45.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1585671</th>\n",
       "      <td>1585671</td>\n",
       "      <td>2014-09-09 21:34:29</td>\n",
       "      <td>916.3</td>\n",
       "      <td>68.36</td>\n",
       "      <td>348.0</td>\n",
       "      <td>0.2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.3</td>\n",
       "      <td>338.0</td>\n",
       "      <td>0.1</td>\n",
       "      <td>35.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1585785</th>\n",
       "      <td>1585785</td>\n",
       "      <td>2014-09-09 23:28:29</td>\n",
       "      <td>916.4</td>\n",
       "      <td>68.00</td>\n",
       "      <td>55.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>61.0</td>\n",
       "      <td>1.2</td>\n",
       "      <td>49.0</td>\n",
       "      <td>0.9</td>\n",
       "      <td>37.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1586086</th>\n",
       "      <td>1586086</td>\n",
       "      <td>2014-09-10 04:29:29</td>\n",
       "      <td>915.4</td>\n",
       "      <td>67.46</td>\n",
       "      <td>34.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>47.0</td>\n",
       "      <td>1.7</td>\n",
       "      <td>18.0</td>\n",
       "      <td>1.1</td>\n",
       "      <td>63.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1586350</th>\n",
       "      <td>1586350</td>\n",
       "      <td>2014-09-10 08:53:29</td>\n",
       "      <td>916.4</td>\n",
       "      <td>74.48</td>\n",
       "      <td>355.0</td>\n",
       "      <td>1.2</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>348.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>33.3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>6981 rows × 11 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           rowID     hpwren_timestamp  air_pressure  air_temp  \\\n",
       "321          321  2011-09-10 05:21:49         911.4     64.94   \n",
       "467          467  2011-09-10 07:47:49         912.2     53.96   \n",
       "540          540  2011-09-10 09:00:49         912.5     55.22   \n",
       "818          818  2011-09-10 13:38:49         913.2     64.22   \n",
       "1129        1129  2011-09-10 18:49:49         913.4     57.74   \n",
       "...          ...                  ...           ...       ...   \n",
       "1585629  1585629  2014-09-09 20:52:29         916.1     68.18   \n",
       "1585671  1585671  2014-09-09 21:34:29         916.3     68.36   \n",
       "1585785  1585785  2014-09-09 23:28:29         916.4     68.00   \n",
       "1586086  1586086  2014-09-10 04:29:29         915.4     67.46   \n",
       "1586350  1586350  2014-09-10 08:53:29         916.4     74.48   \n",
       "\n",
       "         avg_wind_direction  avg_wind_speed  max_wind_direction  \\\n",
       "321                    82.0             0.7                95.0   \n",
       "467                   178.0             1.0               190.0   \n",
       "540                    61.0             2.0                71.0   \n",
       "818                   237.0             3.0               254.0   \n",
       "1129                  207.0             2.7               221.0   \n",
       "...                     ...             ...                 ...   \n",
       "1585629                 4.0             2.0                14.0   \n",
       "1585671               348.0             0.2                 1.0   \n",
       "1585785                55.0             1.0                61.0   \n",
       "1586086                34.0             1.4                47.0   \n",
       "1586350               355.0             1.2                 3.0   \n",
       "\n",
       "         max_wind_speed  min_wind_direction  min_wind_speed  relative_humidity  \n",
       "321                 0.9                63.0             0.5               44.3  \n",
       "467                 1.2               163.0             0.6               90.1  \n",
       "540                 2.3                44.0             1.4               86.3  \n",
       "818                 4.2               221.0             1.2               62.7  \n",
       "1129                3.1               196.0             2.2               80.5  \n",
       "...                 ...                 ...             ...                ...  \n",
       "1585629             2.3               357.0             1.5               45.0  \n",
       "1585671             0.3               338.0             0.1               35.0  \n",
       "1585785             1.2                49.0             0.9               37.7  \n",
       "1586086             1.7                18.0             1.1               63.3  \n",
       "1586350             1.4               348.0             1.0               33.3  \n",
       "\n",
       "[6981 rows x 11 columns]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rand_id = np.random.choice(len(df),7000)\n",
    "sample_df = df.loc[df['rowID'].isin(rand_id)]\n",
    "sample_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "d4752613",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\lalit\\AppData\\Local\\Temp\\ipykernel_1948\\897907050.py:1: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  sample_df.drop(columns = ['rowID','hpwren_timestamp'], inplace = True)\n"
     ]
    },
    {
     "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>air_pressure</th>\n",
       "      <th>air_temp</th>\n",
       "      <th>avg_wind_direction</th>\n",
       "      <th>avg_wind_speed</th>\n",
       "      <th>max_wind_direction</th>\n",
       "      <th>max_wind_speed</th>\n",
       "      <th>min_wind_direction</th>\n",
       "      <th>min_wind_speed</th>\n",
       "      <th>relative_humidity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>321</th>\n",
       "      <td>911.4</td>\n",
       "      <td>64.94</td>\n",
       "      <td>82.0</td>\n",
       "      <td>0.7</td>\n",
       "      <td>95.0</td>\n",
       "      <td>0.9</td>\n",
       "      <td>63.0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>44.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>467</th>\n",
       "      <td>912.2</td>\n",
       "      <td>53.96</td>\n",
       "      <td>178.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>190.0</td>\n",
       "      <td>1.2</td>\n",
       "      <td>163.0</td>\n",
       "      <td>0.6</td>\n",
       "      <td>90.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>540</th>\n",
       "      <td>912.5</td>\n",
       "      <td>55.22</td>\n",
       "      <td>61.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>71.0</td>\n",
       "      <td>2.3</td>\n",
       "      <td>44.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>86.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>818</th>\n",
       "      <td>913.2</td>\n",
       "      <td>64.22</td>\n",
       "      <td>237.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>254.0</td>\n",
       "      <td>4.2</td>\n",
       "      <td>221.0</td>\n",
       "      <td>1.2</td>\n",
       "      <td>62.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1129</th>\n",
       "      <td>913.4</td>\n",
       "      <td>57.74</td>\n",
       "      <td>207.0</td>\n",
       "      <td>2.7</td>\n",
       "      <td>221.0</td>\n",
       "      <td>3.1</td>\n",
       "      <td>196.0</td>\n",
       "      <td>2.2</td>\n",
       "      <td>80.5</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>1585629</th>\n",
       "      <td>916.1</td>\n",
       "      <td>68.18</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>2.3</td>\n",
       "      <td>357.0</td>\n",
       "      <td>1.5</td>\n",
       "      <td>45.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1585671</th>\n",
       "      <td>916.3</td>\n",
       "      <td>68.36</td>\n",
       "      <td>348.0</td>\n",
       "      <td>0.2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.3</td>\n",
       "      <td>338.0</td>\n",
       "      <td>0.1</td>\n",
       "      <td>35.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1585785</th>\n",
       "      <td>916.4</td>\n",
       "      <td>68.00</td>\n",
       "      <td>55.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>61.0</td>\n",
       "      <td>1.2</td>\n",
       "      <td>49.0</td>\n",
       "      <td>0.9</td>\n",
       "      <td>37.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1586086</th>\n",
       "      <td>915.4</td>\n",
       "      <td>67.46</td>\n",
       "      <td>34.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>47.0</td>\n",
       "      <td>1.7</td>\n",
       "      <td>18.0</td>\n",
       "      <td>1.1</td>\n",
       "      <td>63.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1586350</th>\n",
       "      <td>916.4</td>\n",
       "      <td>74.48</td>\n",
       "      <td>355.0</td>\n",
       "      <td>1.2</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>348.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>33.3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>6981 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         air_pressure  air_temp  avg_wind_direction  avg_wind_speed  \\\n",
       "321             911.4     64.94                82.0             0.7   \n",
       "467             912.2     53.96               178.0             1.0   \n",
       "540             912.5     55.22                61.0             2.0   \n",
       "818             913.2     64.22               237.0             3.0   \n",
       "1129            913.4     57.74               207.0             2.7   \n",
       "...               ...       ...                 ...             ...   \n",
       "1585629         916.1     68.18                 4.0             2.0   \n",
       "1585671         916.3     68.36               348.0             0.2   \n",
       "1585785         916.4     68.00                55.0             1.0   \n",
       "1586086         915.4     67.46                34.0             1.4   \n",
       "1586350         916.4     74.48               355.0             1.2   \n",
       "\n",
       "         max_wind_direction  max_wind_speed  min_wind_direction  \\\n",
       "321                    95.0             0.9                63.0   \n",
       "467                   190.0             1.2               163.0   \n",
       "540                    71.0             2.3                44.0   \n",
       "818                   254.0             4.2               221.0   \n",
       "1129                  221.0             3.1               196.0   \n",
       "...                     ...             ...                 ...   \n",
       "1585629                14.0             2.3               357.0   \n",
       "1585671                 1.0             0.3               338.0   \n",
       "1585785                61.0             1.2                49.0   \n",
       "1586086                47.0             1.7                18.0   \n",
       "1586350                 3.0             1.4               348.0   \n",
       "\n",
       "         min_wind_speed  relative_humidity  \n",
       "321                 0.5               44.3  \n",
       "467                 0.6               90.1  \n",
       "540                 1.4               86.3  \n",
       "818                 1.2               62.7  \n",
       "1129                2.2               80.5  \n",
       "...                 ...                ...  \n",
       "1585629             1.5               45.0  \n",
       "1585671             0.1               35.0  \n",
       "1585785             0.9               37.7  \n",
       "1586086             1.1               63.3  \n",
       "1586350             1.0               33.3  \n",
       "\n",
       "[6981 rows x 9 columns]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sample_df.drop(columns = ['rowID','hpwren_timestamp'], inplace = True)\n",
    "sample_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "19057ff5",
   "metadata": {},
   "outputs": [],
   "source": [
    "features = ['air_pressure', 'air_temp', 'max_wind_direction','avg_wind_direction','min_wind_direction', 'avg_wind_speed', 'relative_humidity']\n",
    "#Taking only features of interest for clustering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "26ce256a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>air_pressure</th>\n",
       "      <th>air_temp</th>\n",
       "      <th>max_wind_direction</th>\n",
       "      <th>avg_wind_direction</th>\n",
       "      <th>min_wind_direction</th>\n",
       "      <th>avg_wind_speed</th>\n",
       "      <th>relative_humidity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>321</th>\n",
       "      <td>911.4</td>\n",
       "      <td>64.94</td>\n",
       "      <td>95.0</td>\n",
       "      <td>82.0</td>\n",
       "      <td>63.0</td>\n",
       "      <td>0.7</td>\n",
       "      <td>44.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>467</th>\n",
       "      <td>912.2</td>\n",
       "      <td>53.96</td>\n",
       "      <td>190.0</td>\n",
       "      <td>178.0</td>\n",
       "      <td>163.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>90.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>540</th>\n",
       "      <td>912.5</td>\n",
       "      <td>55.22</td>\n",
       "      <td>71.0</td>\n",
       "      <td>61.0</td>\n",
       "      <td>44.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>86.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>818</th>\n",
       "      <td>913.2</td>\n",
       "      <td>64.22</td>\n",
       "      <td>254.0</td>\n",
       "      <td>237.0</td>\n",
       "      <td>221.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>62.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1129</th>\n",
       "      <td>913.4</td>\n",
       "      <td>57.74</td>\n",
       "      <td>221.0</td>\n",
       "      <td>207.0</td>\n",
       "      <td>196.0</td>\n",
       "      <td>2.7</td>\n",
       "      <td>80.5</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1585629</th>\n",
       "      <td>916.1</td>\n",
       "      <td>68.18</td>\n",
       "      <td>14.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>357.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>45.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1585671</th>\n",
       "      <td>916.3</td>\n",
       "      <td>68.36</td>\n",
       "      <td>1.0</td>\n",
       "      <td>348.0</td>\n",
       "      <td>338.0</td>\n",
       "      <td>0.2</td>\n",
       "      <td>35.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1585785</th>\n",
       "      <td>916.4</td>\n",
       "      <td>68.00</td>\n",
       "      <td>61.0</td>\n",
       "      <td>55.0</td>\n",
       "      <td>49.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>37.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1586086</th>\n",
       "      <td>915.4</td>\n",
       "      <td>67.46</td>\n",
       "      <td>47.0</td>\n",
       "      <td>34.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>63.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1586350</th>\n",
       "      <td>916.4</td>\n",
       "      <td>74.48</td>\n",
       "      <td>3.0</td>\n",
       "      <td>355.0</td>\n",
       "      <td>348.0</td>\n",
       "      <td>1.2</td>\n",
       "      <td>33.3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>6981 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         air_pressure  air_temp  max_wind_direction  avg_wind_direction  \\\n",
       "321             911.4     64.94                95.0                82.0   \n",
       "467             912.2     53.96               190.0               178.0   \n",
       "540             912.5     55.22                71.0                61.0   \n",
       "818             913.2     64.22               254.0               237.0   \n",
       "1129            913.4     57.74               221.0               207.0   \n",
       "...               ...       ...                 ...                 ...   \n",
       "1585629         916.1     68.18                14.0                 4.0   \n",
       "1585671         916.3     68.36                 1.0               348.0   \n",
       "1585785         916.4     68.00                61.0                55.0   \n",
       "1586086         915.4     67.46                47.0                34.0   \n",
       "1586350         916.4     74.48                 3.0               355.0   \n",
       "\n",
       "         min_wind_direction  avg_wind_speed  relative_humidity  \n",
       "321                    63.0             0.7               44.3  \n",
       "467                   163.0             1.0               90.1  \n",
       "540                    44.0             2.0               86.3  \n",
       "818                   221.0             3.0               62.7  \n",
       "1129                  196.0             2.7               80.5  \n",
       "...                     ...             ...                ...  \n",
       "1585629               357.0             2.0               45.0  \n",
       "1585671               338.0             0.2               35.0  \n",
       "1585785                49.0             1.0               37.7  \n",
       "1586086                18.0             1.4               63.3  \n",
       "1586350               348.0             1.2               33.3  \n",
       "\n",
       "[6981 rows x 7 columns]"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sample_df = sample_df[features]\n",
    "sample_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "00755358",
   "metadata": {},
   "source": [
    "### Feature Scaling:\n",
    "\n",
    "It is the final step in data preprocessing. In feature scaling we put all our variables in same range so that one variable doesnot dominate the other.We have 4 feature scaling methods.\n",
    "1. Standard scaler\n",
    "2. Min Max Scaler\n",
    "3. Robust Scaler\n",
    "4. Normalizer\n",
    "\n",
    "We will explore first three methods in this notebook and find out which one performs better for our dataset.  "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "44f9483a",
   "metadata": {},
   "source": [
    "### Standard Scaler:\n",
    "\n",
    "It resizes the distribution to have mean 0 and unit variance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "6c7287db",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-1.7600764 ,  0.26312084, -0.74617646, ..., -1.07640569,\n",
       "        -1.00330497, -0.13636408],\n",
       "       [-1.49908826, -0.64631985,  0.28729986, ..., -0.0414829 ,\n",
       "        -0.85800936,  1.60158126],\n",
       "       [-1.40121771, -0.54195781, -1.00726522, ..., -1.27304102,\n",
       "        -0.37369064,  1.45738492],\n",
       "       ...,\n",
       "       [-0.12890051,  0.51657152, -1.1160522 , ..., -1.22129488,\n",
       "        -0.85800936, -0.38681036],\n",
       "       [-0.45513569,  0.47184493, -1.26835397, ..., -1.54212095,\n",
       "        -0.66428187,  0.58461761],\n",
       "       [-0.12890051,  1.05329062, -1.74701669, ...,  1.87312427,\n",
       "        -0.76114561, -0.55377454]])"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = StandardScaler().fit_transform(sample_df)\n",
    "X"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9b284900",
   "metadata": {},
   "source": [
    "### Robust Scaler:\n",
    "\n",
    "Scale features using statistics that are robust to outliers. This Scaler removes the median and scales the data according to the quantile range (defaults to IQR: Interquartile Range). The IQR is the range between the 1st quartile (25th quantile) and the 3rd quartile (75th quantile)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "c1d34c3a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-1.35897436,  0.14423077, -0.6       , ..., -0.9       ,\n",
       "        -0.57692308, -0.01612903],\n",
       "       [-1.15384615, -0.44230769,  0.01290323, ..., -0.13076923,\n",
       "        -0.46153846,  1.03917051],\n",
       "       [-1.07692308, -0.375     , -0.75483871, ..., -1.04615385,\n",
       "        -0.07692308,  0.9516129 ],\n",
       "       ...,\n",
       "       [-0.07692308,  0.30769231, -0.81935484, ..., -1.00769231,\n",
       "        -0.46153846, -0.16820276],\n",
       "       [-0.33333333,  0.27884615, -0.90967742, ..., -1.24615385,\n",
       "        -0.30769231,  0.42165899],\n",
       "       [-0.07692308,  0.65384615, -1.19354839, ...,  1.29230769,\n",
       "        -0.38461538, -0.26958525]])"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scaler = RobustScaler()\n",
    "rob_x = scaler.fit_transform(sample_df)\n",
    "rob_x"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "57e9f03e",
   "metadata": {},
   "source": [
    "### Minmax Scaler:\n",
    "\n",
    "It rescales data from the original range so that all values are within the new range of 0 and 1.It does not change the distribution of original dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "bda0e2b1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.27876106, 0.50140056, 0.26462396, ..., 0.17548747, 0.03870968,\n",
       "        0.46238938],\n",
       "       [0.31415929, 0.33053221, 0.52924791, ..., 0.454039  , 0.05806452,\n",
       "        0.96902655],\n",
       "       [0.32743363, 0.35014006, 0.19777159, ..., 0.12256267, 0.12258065,\n",
       "        0.92699115],\n",
       "       ...,\n",
       "       [0.5       , 0.54901961, 0.16991643, ..., 0.13649025, 0.05806452,\n",
       "        0.38938053],\n",
       "       [0.45575221, 0.54061625, 0.13091922, ..., 0.05013928, 0.08387097,\n",
       "        0.67256637],\n",
       "       [0.5       , 0.64985994, 0.00835655, ..., 0.96935933, 0.07096774,\n",
       "        0.34070796]])"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scaler = MinMaxScaler()\n",
    "min_x = scaler.fit_transform(sample_df)\n",
    "min_x"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6d407a63",
   "metadata": {},
   "source": [
    "### Principal Component Analysis:\n",
    "\n",
    "Linear dimensionality reduction using Singular Value Decomposition of the data to project it to a lower dimensional space. The input data is centered but not scaled for each feature before applying the SVD."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "92b0ace2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(6981, 7)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[-0.91878122,  0.37826533, -0.56481021, ..., -0.21901997,\n",
       "         0.51749503,  0.08985377],\n",
       "       [ 1.32271683,  1.12808753, -1.11151023, ..., -0.17585723,\n",
       "        -0.11373328,  0.04068561],\n",
       "       [-0.65363543,  2.02634422, -0.84460876, ..., -0.11203041,\n",
       "        -0.31381547,  0.2226488 ],\n",
       "       ...,\n",
       "       [-1.93584057,  0.01256594, -0.64706425, ..., -0.25358669,\n",
       "        -0.00416141,  0.14523688],\n",
       "       [-1.82433476,  0.85237304, -0.76663309, ..., -0.31320699,\n",
       "        -0.60535107,  0.24380147],\n",
       "       [ 0.69119461, -1.98038722, -0.44181603, ...,  2.1436536 ,\n",
       "         0.08678517,  1.87128995]])"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pca = PCA()\n",
    "#Transform the data\n",
    "df_pca = pca.fit_transform(X)\n",
    "print(np.shape(df_pca))\n",
    "df_pca"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3299bdae",
   "metadata": {},
   "source": [
    "### Kernel Principal Component Analysis:\n",
    "\n",
    "Non-linear dimensionality reduction through the use of kernels like ‘linear’, ‘poly’, ‘rbf’, ‘sigmoid’, ‘cosine'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "c96078f7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-0.91878122,  0.37826533, -0.56481021],\n",
       "       [ 1.32271683,  1.12808753, -1.11151023],\n",
       "       [-0.65363543,  2.02634422, -0.84460876],\n",
       "       ...,\n",
       "       [-1.93584057,  0.01256594, -0.64706425],\n",
       "       [-1.82433476,  0.85237304, -0.76663309],\n",
       "       [ 0.69119461, -1.98038722, -0.44181603]])"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "kernel_pca = KernelPCA(n_components=3,kernel='linear')\n",
    "kernel_pca_df = kernel_pca.fit_transform(X)\n",
    "kernel_pca_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "30792795",
   "metadata": {},
   "source": [
    "### Truncated SVD:\n",
    "\n",
    "This transformer performs linear dimensionality reduction by means of truncated singular value decomposition (SVD). Contrary to PCA, this estimator does not center the data before computing the singular value decomposition."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "a699ff3f",
   "metadata": {},
   "outputs": [],
   "source": [
    "trun_svd = TruncatedSVD(n_components = 2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6248921e",
   "metadata": {},
   "source": [
    "Find number of clusters suitable for our dataset using Silhouette Coefficient\n",
    "\n",
    "Silhouette Coefficient or silhouette score is a metric used to calculate the goodness of a clustering technique. Its value ranges from -1 to 1.\n",
    "\n",
    "Silhouette Score = (b-a)/max(a,b)\n",
    "\n",
    "where\n",
    "\n",
    "a= average intra-cluster distance i.e the average distance between each point within a cluster.\n",
    "\n",
    "b= average inter-cluster distance i.e the average distance between all clusters.\n",
    "\n",
    "1: Means clusters are well apart from each other and clearly distinguished.\n",
    "\n",
    "0: Means clusters are indifferent, or we can say that the distance between clusters is not significant.\n",
    "\n",
    "-1: Means clusters are assigned in the wrong way.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "beffb3b0",
   "metadata": {},
   "source": [
    "### Assumptions of K-Means:\n",
    "\n",
    "Before we go on just implementing K-Means algorithm I would suggest to look into assumptions made by the model.\n",
    "1. k-means assumes the variance of the distribution of each attribute (variable) is spherical;\n",
    "2. all variables have the same variance;\n",
    "3. the prior probability for all k clusters is the same, i.e., each cluster has roughly equal number of observations;\n",
    "\n",
    "If anyone of the assumptions fail k-means will fail.\n",
    " "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "c90c2611",
   "metadata": {},
   "outputs": [],
   "source": [
    "k_means = KMeans(n_clusters=10,init ='k-means++',verbose =0,max_iter = 380)\n",
    "#Run the  kmeans clustering algorithm method = kmeans++ chooses centroids in a way which speeds up convergence\n",
    "model = k_means.fit(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "29d06bab",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Generate cluster predictions and store in y_hat\n",
    "y_hat = k_means.predict(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "5adceaab",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.23428444853216873"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "labels = k_means.labels_\n",
    "silhouette_score(X,labels,metric= 'euclidean')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "98060ed2",
   "metadata": {},
   "source": [
    "Two values are of importance here — distortion and inertia. Distortion is the average of the euclidean squared distance from the centroid of the respective clusters. Inertia is the sum of squared distances of samples to their closest cluster centre."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b2867c39",
   "metadata": {},
   "source": [
    "### Elbow plot:\n",
    "In the Elbow method, we are actually varying the number of clusters ( K ) from 1 – 10. For each value of K, we are calculating WCSS ( Within-Cluster Sum of Square ). WCSS is the sum of squared distance between each point and the centroid in a cluster. When we plot the WCSS with the K value, the plot looks like an Elbow. As the number of clusters increases, the WCSS value will start to decrease. WCSS value is largest when K = 1. When we analyze the graph we can see that the graph will rapidly change at a point and thus creating an elbow shape. From this point, the graph starts to move almost parallel to the X-axis. The K value corresponding to this point is the optimal K value or an optimal number of clusters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "a4283053",
   "metadata": {},
   "outputs": [],
   "source": [
    "#for each value of k, we can initialise k_means and use inertia to identify the sum of squared distances of samples to the nearest cluster centre\n",
    "sum_of_squared_distances = []\n",
    "K = range(1,12)\n",
    "for k in range(1,12):\n",
    "    k_means = KMeans(n_clusters=k,init = 'k-means++')\n",
    "    model = k_means.fit(X)\n",
    "    sum_of_squared_distances.append(k_means.inertia_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "2d2c0bb2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(K, sum_of_squared_distances, 'bx-')\n",
    "plt.xlabel('k')\n",
    "plt.ylabel('sum_of_squared_distances')\n",
    "plt.title('elbow method for optimal k')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "113c3b9a",
   "metadata": {},
   "source": [
    "We can see that optimal value of k where elbow shape is formed is k =4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "f000c9c5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "scores of pca  with StandardScaler\n",
      "Silhoutte score of kmean is  0.27098086224020884\n",
      "davies bouldin score of kmean is  1.4225561362395358\n",
      "scores of pca with MinMaxScaler\n",
      "Silhoutte score of kmean is  0.34849957721572256\n",
      "davies bouldin score of kmean is  1.063210646684248\n",
      "scores of pca with RobustScaler\n",
      "Silhoutte score of kmean is  0.2477199751016768\n",
      "davies bouldin score of kmean is  1.4766813923910407\n",
      "scores of kernel pca with StandardScaler\n",
      "Silhoutte score of kmean is  0.3407303240716901\n",
      "davies bouldin score of kmean is  1.0535858967910139\n",
      "scores of kernel pca with MinMaxScaler\n",
      "Silhoutte score of kmean is  0.45441309372140615\n",
      "davies bouldin score of kmean is  0.8430974116344568\n",
      "scores of kernel pca with RobustScaler\n",
      "Silhoutte score of kmean is  0.3413835161115402\n",
      "davies bouldin score of kmean is  0.9994492044522848\n",
      "scores of truncated svd with StandardScaler\n",
      "Silhoutte score of kmean is  0.41328993920044776\n",
      "davies bouldin score of kmean is  0.8660153955535671\n",
      "scores of truncated svd with MinMaxScaler\n",
      "Silhoutte score of kmean is  0.42915749013493376\n",
      "davies bouldin score of kmean is  0.8204809710349248\n",
      "scores of truncated svd with RobustScaler\n",
      "Silhoutte score of kmean is  0.39382221534612183\n",
      "davies bouldin score of kmean is  0.8912134896505916\n"
     ]
    }
   ],
   "source": [
    "kmean = KMeans(n_clusters = 4)\n",
    "df_pca = pca.fit_transform(X)\n",
    "model = kmean.fit(df_pca)\n",
    "y_hc = kmean.predict(df_pca)\n",
    "labels = kmean.labels_\n",
    "print(\"scores of pca  with StandardScaler\")\n",
    "print('Silhoutte score of kmean is ' , silhouette_score(df_pca, labels))\n",
    "print('davies bouldin score of kmean is ' , davies_bouldin_score(df_pca, labels))\n",
    "df_pca = pca.fit_transform(min_x)\n",
    "model = kmean.fit(df_pca)\n",
    "y_h = kmean.fit_predict(df_pca)\n",
    "labels_kmean = kmean.labels_\n",
    "print(\"scores of pca with MinMaxScaler\")\n",
    "print('Silhoutte score of kmean is ' , silhouette_score(df_pca, labels_kmean))\n",
    "print('davies bouldin score of kmean is ' , davies_bouldin_score(df_pca, labels_kmean))\n",
    "df_pca = pca.fit_transform(rob_x)\n",
    "model = kmean.fit(df_pca)\n",
    "y_min = kmean.fit_predict(df_pca)\n",
    "labels_kmean = kmean.labels_\n",
    "print(\"scores of pca with RobustScaler\")\n",
    "print('Silhoutte score of kmean is ' , silhouette_score(df_pca, labels_kmean))\n",
    "print('davies bouldin score of kmean is ' , davies_bouldin_score(df_pca, labels_kmean))\n",
    "df_pca = kernel_pca.fit_transform(X)\n",
    "model = kmean.fit(df_pca)\n",
    "y_min = kmean.fit_predict(df_pca)\n",
    "labels_kmean = kmean.labels_\n",
    "print(\"scores of kernel pca with StandardScaler\")\n",
    "print('Silhoutte score of kmean is ' , silhouette_score(df_pca, labels_kmean))\n",
    "print('davies bouldin score of kmean is ' , davies_bouldin_score(df_pca, labels_kmean))\n",
    "df_pcade = kernel_pca.fit_transform(min_x)\n",
    "model_mine = kmean.fit(df_pcade)\n",
    "y_hce = kmean.fit_predict(df_pcade)\n",
    "labels_kmean_min = kmean.labels_\n",
    "print(\"scores of kernel pca with MinMaxScaler\")\n",
    "print('Silhoutte score of kmean is ' , silhouette_score(df_pcade, labels_kmean_min))\n",
    "print('davies bouldin score of kmean is ' , davies_bouldin_score(df_pcade, labels_kmean_min))\n",
    "df_pca = kernel_pca.fit_transform(rob_x)\n",
    "model = kmean.fit(df_pca)\n",
    "y_min = kmean.fit_predict(df_pca)\n",
    "labels_kmean = kmean.labels_\n",
    "print(\"scores of kernel pca with RobustScaler\")\n",
    "print('Silhoutte score of kmean is ' , silhouette_score(df_pca, labels_kmean))\n",
    "print('davies bouldin score of kmean is ' , davies_bouldin_score(df_pca, labels_kmean))\n",
    "df_pca = trun_svd.fit_transform(X)\n",
    "model = kmean.fit(df_pca)\n",
    "y_min = kmean.fit_predict(df_pca)\n",
    "labels_kmean = kmean.labels_\n",
    "print(\"scores of truncated svd with StandardScaler\")\n",
    "print('Silhoutte score of kmean is ' , silhouette_score(df_pca, labels_kmean))\n",
    "print('davies bouldin score of kmean is ' , davies_bouldin_score(df_pca, labels_kmean))\n",
    "df_pcad = trun_svd.fit_transform(min_x)\n",
    "model_min = kmean.fit(df_pcad)\n",
    "y_hc = kmean.fit_predict(df_pcad)\n",
    "labels_kmean_min = kmean.labels_\n",
    "print(\"scores of truncated svd with MinMaxScaler\")\n",
    "print('Silhoutte score of kmean is ' , silhouette_score(df_pcad, labels_kmean_min))\n",
    "print('davies bouldin score of kmean is ' , davies_bouldin_score(df_pcad, labels_kmean_min))\n",
    "df_pca = trun_svd.fit_transform(rob_x)\n",
    "model = kmean.fit(df_pca)\n",
    "y_min = kmean.fit_predict(df_pca)\n",
    "labels_kmean = kmean.labels_\n",
    "print(\"scores of truncated svd with RobustScaler\")\n",
    "print('Silhoutte score of kmean is ' , silhouette_score(df_pca, labels_kmean))\n",
    "print('davies bouldin score of kmean is ' , davies_bouldin_score(df_pca, labels_kmean))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "acb2cd07",
   "metadata": {},
   "source": [
    "### Visualizing the clusters formed:\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "8a33a4d7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "min_x = np.array(min_x)\n",
    "plt.scatter(min_x[y_hce==0, 0], min_x[y_hce==0, 1], s=20, c='red', label ='Cluster 1')\n",
    "plt.scatter(min_x[y_hce==1, 0], min_x[y_hce==1, 1], s=20, c='yellow', label ='Cluster 2')\n",
    "plt.scatter(min_x[y_hce==2, 0], min_x[y_hce==2, 1], s=20, c='green', label ='Cluster 3')\n",
    "plt.scatter(min_x[y_hce==3, 0], min_x[y_hce==3, 1], s=20, c='cyan', label ='Cluster 4')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5fb27867",
   "metadata": {},
   "source": [
    "### Create plots to understand clusters formed:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "21355316",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "scores of pca with MinMaxScaler\n",
      "Silhoutte score of kmean is  0.34848166459415214\n",
      "davies bouldin score of kmean is  1.0635860305184126\n"
     ]
    }
   ],
   "source": [
    "df_pca = pca.fit_transform(min_x)\n",
    "model = kmean.fit(df_pca)\n",
    "y_min = kmean.fit_predict(df_pca)\n",
    "labels_kmean = kmean.labels_\n",
    "print(\"scores of pca with MinMaxScaler\")\n",
    "print('Silhoutte score of kmean is ' , silhouette_score(df_pca, labels_kmean))\n",
    "print('davies bouldin score of kmean is ' , davies_bouldin_score(df_pca, labels_kmean))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "a22ed426",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Function that creates a DataFrame with a column for Cluster Number\n",
    "\n",
    "def pd_centers(featuresUsed, centers):\n",
    "    colNames = list(featuresUsed)\n",
    "    colNames.append('prediction')\n",
    "\n",
    "    # Zip with a column called 'prediction' (index)\n",
    "    Z = [np.append(A, index) for index, A in enumerate(centers)]\n",
    "\n",
    "    # Convert to pandas data frame for plotting\n",
    "    P = pd.DataFrame(Z, columns=colNames)\n",
    "    P['prediction'] = P['prediction'].astype(int)\n",
    "    return P"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "65584ae5",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Function that creates Parallel Plots\n",
    "from itertools import cycle, islice\n",
    "from pandas.plotting import parallel_coordinates\n",
    "\n",
    "def parallel_plot(data):\n",
    "    my_colors = list(islice(cycle(['b', 'r', 'g', 'y', 'k']), None, len(data)))\n",
    "    plt.figure(figsize=(15,8)).gca().axes.set_ylim([-3,+3])\n",
    "    parallel_coordinates(data, 'prediction', color = my_colors, marker='o')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "4f9fb644",
   "metadata": {},
   "outputs": [],
   "source": [
    "centers = model.cluster_centers_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "674da1b1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>air_pressure</th>\n",
       "      <th>air_temp</th>\n",
       "      <th>max_wind_direction</th>\n",
       "      <th>avg_wind_direction</th>\n",
       "      <th>min_wind_direction</th>\n",
       "      <th>avg_wind_speed</th>\n",
       "      <th>relative_humidity</th>\n",
       "      <th>prediction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.203315</td>\n",
       "      <td>0.126014</td>\n",
       "      <td>0.616840</td>\n",
       "      <td>-0.042715</td>\n",
       "      <td>0.058747</td>\n",
       "      <td>-0.280452</td>\n",
       "      <td>0.002522</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.173454</td>\n",
       "      <td>0.272241</td>\n",
       "      <td>-0.045300</td>\n",
       "      <td>-0.016278</td>\n",
       "      <td>-0.004278</td>\n",
       "      <td>0.022081</td>\n",
       "      <td>-0.000795</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.552758</td>\n",
       "      <td>-0.078820</td>\n",
       "      <td>-0.065967</td>\n",
       "      <td>0.014289</td>\n",
       "      <td>-0.022036</td>\n",
       "      <td>0.023722</td>\n",
       "      <td>0.001771</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.317895</td>\n",
       "      <td>-0.287152</td>\n",
       "      <td>-0.008340</td>\n",
       "      <td>0.015651</td>\n",
       "      <td>0.013104</td>\n",
       "      <td>0.007530</td>\n",
       "      <td>-0.001097</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   air_pressure  air_temp  max_wind_direction  avg_wind_direction  \\\n",
       "0     -0.203315  0.126014            0.616840           -0.042715   \n",
       "1      0.173454  0.272241           -0.045300           -0.016278   \n",
       "2     -0.552758 -0.078820           -0.065967            0.014289   \n",
       "3      0.317895 -0.287152           -0.008340            0.015651   \n",
       "\n",
       "   min_wind_direction  avg_wind_speed  relative_humidity  prediction  \n",
       "0            0.058747       -0.280452           0.002522           0  \n",
       "1           -0.004278        0.022081          -0.000795           1  \n",
       "2           -0.022036        0.023722           0.001771           2  \n",
       "3            0.013104        0.007530          -0.001097           3  "
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "P = pd_centers(features, centers)\n",
    "P"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3f3d9ee2",
   "metadata": {},
   "source": [
    "### Dry days"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "51b31179",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "parallel_plot(P[P['relative_humidity'] < 0])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "92a5f19c",
   "metadata": {},
   "source": [
    "### Warm days"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "ae579b9f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA50AAAHXCAYAAAA/cD5pAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8qNh9FAAAACXBIWXMAAAsTAAALEwEAmpwYAAAsVElEQVR4nO3de7hkZ10n+u+vk5DOjQAhICGkOwoEQlQ85KiARlFUdLyh8Siinghz+ngcL3AOj6NmxsvMiYJ4BnTwhopR6QEZEVFwuClNuIVAMISkEwKGBAIcwCiXEBKSzjt/rLXZ1bv3pap7v7Uv/fk8Tz17VdXa611V9au11ne9a62q1loAAACghx0bPQMAAABsX0InAAAA3QidAAAAdCN0AgAA0I3QCQAAQDdCJwAAAN3MJXRW1c6quqKq3lNV11bVr86jXQAAADZWzeN3OquqkpzUWrutqo5L8pYkP9tau7x74wAAAGyYY+fRSBuS7W3j3ePGW/+0CwAAwIaa2zmdVXVMVV2V5BNJXt9ae8e82gYAAGBjzKWnM0laaweSPLqq7pPkFVV1XmvtmslxqmpPkj3j3cc8/OEPn9fssYXdcMMNUStMS70wLbXCLNQL01IrzOKGG27459ba6Rs9H0dqLud0HtJo1S8lub219purjNM2Yt7YeqoqaoVpqRempVaYhXphWmqFWVTVla218zd6Po7UvK5ee/rYw5mqOiHJtyS5fh5tAwAAsHHmdXjtg5L8aVUdkyHovqy19qo5tQ0AAMAGmdfVa69O8lXzaAsAAIDNY24XEgIAAJinu+66K7fcckvuuOOOjZ6VVe3cuTNnnnlmjjvuuI2elS6ETgAAYFu65ZZbcsopp2T37t2pqo2enWW11nLrrbfmlltuydlnn73Rs9PF3H6nEwAAYJ7uuOOOnHbaaZs2cCbDFY1PO+20Td8beySETgAAYNvazIFzwVaYxyMhdAIAAHT0tKc9LQ94wANy3nnnbfSsbAihEwAAoKOLLroor3nNazZ6NjaM0AkAAJBk795k9+5kx47h79696zPdCy64IPe73/3WZ2JbkKvXAgAAR729e5M9e5Lbbx/u33zzcD9JnvrUjZuv7UDoBAAAtr1nPCO56qqVn7/88uTOOw9+7Pbbk6c/PfnDP1z+fx796OT5z1+f+dvOHF4LAAAc9ZYGzrUeZ3p6OgEAgG1vrR7J3buHQ2qX2rUr2bevwwwdRfR0AgAAR71LLklOPPHgx048cXj8SD3lKU/JYx/72Lzvfe/LmWeemT/+4z8+8oluIXo6AQCAo97CxYIuvjj50IeSs84aAud6XEToJS95yZFPZAsTOgEAADIETFeqXX8OrwUAAKAboRMAAIBuhE4AAGDbaq1t9CysaSvM45EQOgEAgG1p586dufXWWzd1qGut5dZbb83OnTs3ela6cSEhAABgWzrzzDNzyy235JOf/ORGz8qqdu7cmTPPPHOjZ6MboRMAANiWjjvuuJx99tkbPRtHPYfXAgAA0I3QCQAAQDdCJwAAAN0InQAAAHQjdAIAANCN0AkAAEA3QicAAADdCJ0AAAB0I3QCAADQjdAJAABAN0InAAAA3QidAAAAdCN0AgAA0I3QCQAAQDdCJwAAAN0InQAAAHQjdAIAANCN0AkAAEA3QicAAADdCJ0AAAB0I3QCAADQjdAJAABAN0InAAAA3QidAAAAdCN0AgAA0I3QCQAAQDdCJwAAAN0InQAAAHQjdAIAANCN0AkAAEA3QicAAADdCJ0AAAB0I3QCAADQjdAJAABAN0InAAAA3QidAAAAdCN0AgAA0I3QCQAAQDdCJwAAAN0InQAAAHQjdAIAANCN0AkAAEA3QicAAADdCJ0AAAB0M5fQWVUPqao3VtX+qrq2qn52Hu0CAACwsY6dUzt3J/l/WmvvrqpTklxZVa9vre2fU/sAAABsgLn0dLbWPtZae/c4/Nkk1yV58DzaBgAAYOPMq6fzi6pqd5KvSvKOZZ7bk2TPwv19+/bNbb7Y2tQKs1AvTEutMAv1wrTUCkebaq3Nr7Gqk5O8KcklrbW/WmPcNs95Y+uqqqgVpqVemJZaYRbqhWmpFWZRVVe21s7f6Pk4UnO7em1VHZfk5Un2rhU4AQAA2B7mdfXaSvLHSa5rrf2XebQJAADAxptXT+fjk/xokm+qqqvG23fMqW0AAAA2yFwuJNRae0uSmkdbAAAAbB5zO6cTAACAo4/QCQAAQDdCJwAAAN0InQAAAHQjdAIAANCN0AkAAEA3QicAAADdCJ0AAAB0I3QCAADQjdAJAABAN0InAAAA3QidAAAAdCN0AgAA0I3QCQAAQDdCJwAAAN0InQAAAHQjdAIAANCN0AkAAEA3QicAAADdCJ0AAAB0I3QCAADQjdAJAABAN0InAAAA3QidAAAAdCN0AgAA0I3QCQAAQDdCJwAAAN0InQAAAHQjdAIAANCN0AkAAEA3QicAAADdCJ0AAAB0I3QCAADQjdAJAABAN0InAAAA3QidAAAAdCN0AgAA0I3QCQAAQDdCJwAAAN0InQAAAHQjdAIAANCN0AkAAEA3QicAAADdCJ0AAAB0I3QCAADQjdAJAABAN0InAAAA3QidAAAAdCN0AgAA0I3QCQAAQDdCJwAAAN0InQAAAHQjdAIAANCN0AkAAEA3QicAAADdCJ0AAAB0I3QCAADQjdAJAABAN0InAAAA3QidAAAAdCN0AgAA0I3QCQAAQDdCJwAAAN3MLXRW1Yuq6hNVdc282gQAAGBjzbOn89IkT5pjewAAAGywuYXO1tplSf5lXu0BAACw8ZzTCQAAQDfHbvQMTKqqPUn2LNzft2/fxs0MW4paYRbqhWmpFWahXpiWWuFoU621+TVWtTvJq1pr500xbpvnvLF1VVXUCtNSL0xLrTAL9cK01AqzqKorW2vnb/R8HCmH1wIAANDNPH8y5SVJ3p7knKq6paqePq+2AQAA2BhzO6eztfaUebUFAADA5uDwWgAAALoROgEAAOhG6AQAAKAboRMAAIBuhE4AAAC6EToBAADoRugEAACgG6ETAACAboROAAAAuhE6AQAA6EboBAAAoBuhEwAAgG6ETgAAALoROgEAAOhG6AQAAKAboRMAAIBuhE4AAAC6EToBAADoRugEAACgG6ETAACAboROAAAAuhE6AQAA6EboBAAAoBuhEwAAgG6ETgAAALoROgEAAOhG6AQAAKAboRMAAIBuhE4AAAC6EToBAADoRugEAACgG6ETAACAboROAAAAuhE6AQAA6EboBAAAoBuhEwAAgG6ETgAAALoROgEAAOhG6AQAAKAboRMAAIBuhE4AAAC6EToBAADoRugEAACgG6ETAACAboROAAAAuhE62bL27k127x6Gd+8e7gMAAJvLsRs9A3A49u5N9uxJbr99uH/zzcP9JHnqUzduvgAAgIMJnczd3Xcnn/3s4u0znzn4/jTP3XBDcuDAwdO9/fbkJ38yqUrOPTc555zkhBM25jUCAACDaq1t9Dwsq6raZp23o01ryec/v3o4nCU83nHHdO0ee2xy73snp5xy6O3lL58cs5IcWis7diRf+qVDAJ28PfKRyYknrsc7w1ZUVbFsYRpqhVmoF6alVphFVV3ZWjt/o+fjSOnp3KaW9iYeac/iPfdM1+5JJx0aEB/ykOWD4ymnrBwqTzklOf74oddyObt3D4fULnXWWcmrX53s3z/crr12+Pt3fze8J8kwzd27F0Poox61GEZPPvlw3m0AAGAlejo3iYXexGnD4VrPf/7z07V77LHTBcBpnjvppOSYY/q+TwsOPqdz6Ok88cTkhS9c/pzOu+5KPvCBg4Po/v3J+96XfOELi+OdddbBQXQhjJ566nxeF/3Zw8y01AqzUC9MS60wi+3S0yl0HoG7705uu+3wew8nn7vttkPPUVzJcr2JswbEaXoTN7u9e5OLL05uvrmya1fLJZfMfhGhu+9O/umfFkPoQii9/vrkzjsXxzvzzIMP0X3Uo4Ywet/7ru9roj8re6alVpiFemFaaoVZCJ2dVVU73CCxkuV6E4+kZ/FwehMPNxwu3E4+eX69iVtFj4X3gQPJBz94aM/oddcd/Lk/6EGH9oyee25y2mnrOjusIyt7pqVWmIV6YVpqhVkInZ1VVVs4ZPL5z0+e9KQjPy/xs5+dvjfxxBOnC4fThMet3Ju4Fcxz4X3PPcO5pJNBdOH2uc8tjveABxwaRB/1qOT00+cym6zCyp5pqRVmoV6YllphFkJnZwuhc1rHHLM+5yXqTdx6NsPC+557kg9/+NDDdPfvH3Z2LLj//Q+9gNG55yYPfKAdE/OyGeqFrUGtMAv1wrTUCrMQOjtbGjpf+MLVw+POnTbaj1abeeHdWvKRjxx6mO611yaf/vTiePe976E9o+eem5xxhrpeb5u5Xthc1AqzUC9MS60wC6Gzs8nQuWtXctNNGzs/bF5bceHdWvKxjx3aM3rttcm//uvieKeeeughuueeO1zYSBg9PFuxXtgYaoVZqBempVaYhdDZ2eQ5nSv9DAYk22vh3VryiU8ceoju/v3JJz+5ON4ppwxXz13aO3rWWcmOHRs3/1vBdqoX+lIrzEK9MC21wiyEzs56XL2W7eloWXh/8pOHXrzo2muTj398cZyTThrC6NKe0d27hdEFR0u9cOTUCrNQL0xLrTALobOzrfA7nWwOR/vC+9Zbh59yWdo7+tGPLo5zwgnJIx5x6AWMvvRLj76LZh3t9cL01AqzUC9MS60wC6GzM6GTaVl4L+9TnxrC6NKfd/nwhxfHOf74xTA6eXvoQ4ffl92O1AvTUivMQr0wLbXCLITOzoROpmXhPZvPfObgntGF3tGbb14c57jjknPOOfQw3Yc+NLnXvTZu3teDemFaaoVZqBempVaYhdDZmdDJtCy818dttyXXX3/oBYw++MHhAkfJ0Pv5sIcdegGjhz986DXdCtQL01IrzEK9MC21wiyEzs6ETqZl4d3X7bcvhtHJntEbb0zuuWcY55hjhl7QpT/vcs45w2/obibqhWmpFWahXljL3r3JxRcnN99ccbFMpiV0ztpQ1ZOS/FaSY5L8UWvt2WuML3QyFSv6jfH5zyc33HDoBYw+8IHkwIFhnB07hosVLb2A0SMekZx44sbMt3phWmqFWagXVrN3b7Jnz7AjN6n4WUCmJXTO0kjVMUluSPItSW5J8s4kT2mt7V/lf4ROpmJFv7nceWfy/vcfegGjG25I7r57GKcqOfvsQ3tGH/GI5OST+86femFaaoVp6L2aXmvDETJ33z3snNwMf+fV1o03Lq4DF0JnMpy2cvbZwxFDR3o79tj1mc5GTH+1aR+tP/m2uGw5P629qzZ6fo7UvK5P+dVJPtBauzFJquqlSb4nyYqhE9iajj8+Oe+84TbprruGMLr0MN3XvnZ4bsGuXYf2jD7ykcm97z3f18HRa2FFnwy/cStEsJKDe6+GC7Lt2ZN84QvJhRdur9C0HvO2cBTMZrMQeI70786dKz9/ww3Lt3333cljHrP4/qx2+8IX1h5n8r1e67aVbNXAfLi3N70pef7zhx3528W8ejovTPKk1tq/He//aJKvaa391JLx9iTZM959zBvf+Mbu88bW94QnPCFqZes6cKDykY/szM03n5SbbjopN910Ym6++aR86EMn5q67Fndvnn76Hdm9+/bs2vW5g/6efPLdq0z9UOqFBUOvS+XAgcXbP/zD6fnd331o7rzzmCz0RtzrXgfy9Kd/MI973K25556ktUpri39Xeiypg55bv8eGxw/nsWR4zQc/1/ex4e/Kjx08/pE9tvh+Lf/YMP7sj02+vsnHvvCFHWOdJJO9V5vJjh0txxyzeFu4v/g3hzy33PMrTyfLTvfgcVceZxhe/vnF8fo9v2PHcPRNbz/0Q1+bj3984SIHi7XywAfekZe+9PL+M7CMe+4Z6n1YDi4Or3b/wIG1xzn8+8N3bLKN9bg/uZxf+thq9yfnaa37Bw707I7dHj2dmyp0Lvkfh9cyFYfAbU8LhyNN9ozu3z/83MvnP7843hlnHPrTLueem9zvfgdPzyFwi5Ye3rbc7XCf6zXdHs8tBKSVbc4QMQ8LG+I7dhw8POtj6zGNzTzd3/iNyXft4Hr5zd9cn96zaf6u9Ny8AhVrc07n9jeEz/W5Pe5xi78cIHTO0kjVY5P8Smvt28b7v5AkrbVfX+V/hE6mInQeXQ4cGA5hW3oBo+uuSz73ucXxHvjAxRB6223JS16ycJjKsLI/4YTkec9LnvzkzROC5vXcZvm6LN1oXno73OeO5H+f+czJOTw4RLz4xZs7AK1HW1VCyix27578jePFetm1K7nppo2ZJzYvOz+Z1sHLFqFz+kaqjs1wIaFvTvKRDBcS+uHW2rWr/I/QyVSETpJhD+OHP3zoBYz2708++9nJMTe+9+q44zZf2Jr3dDdrD4wQwSz0XnE4bLewloOXLdsjdM7lQkKttbur6qeSvDbJMUletFrgBJjVjh1DMNi1K/mO71h8vLUh+Ky0fn/BC+Yftti8Lrnk4AvDJMPP+1xyycbNE5vXQrAceq+G5Y/eK+BILV22bAdz+53OWenpZFr2GLIWvVfMwiFwHA7rIqalVpiF3+nsTOhkWhberMUhcBwOyxZmoV6YllphFtsldDrQC9j2nvrUIWDu2jXc37VL4AQAmBc9nWx59hgyC/XCtNQKs1AvTEutMAs9nQAAALAGoRMAAIBuhE4AAAC6EToBAADoRugEAACgG6ETAACAboROAAAAuhE6AQAA6EboBAAAoBuhEwAAgG6ETgAAALoROgEAAOhG6AQAAKAboRMAAIBuhE4AAAC6EToBAADoRugEAACgG6ETAACAboROAAAAuhE6AQAA6EboBAAAoBuhEwAAgG6ETgAAALoROgEAAOhG6AQAAKAboRMAAIBuhE4AAAC6EToBAADoRugEAACgG6ETAACAboROAAAAuhE6AQAA6EboBAAAoBuhEwAAgG6ETgAAALoROgEAAOhG6AQAAKAboRMAAIBuhE4AAAC6EToBAADoRugEAACgG6ETAACAboROAAAAuhE6AQAA6EboBAAAoBuhEwAAgG6ETgAAALoROgEAAOhG6AQAAKAboRMAAIBuhE4AAAC6EToBAADoRugEAACgG6ETAACAboROAAAAuhE6AQAA6EboBAAAoBuhEwAAgG6ETgAAALoROgEAAOime+isqh+oqmur6p6qOr93ewAAAGwe8+jpvCbJ9yW5bA5tAQAAsIkc27uB1tp1SVJVvZsCAABgk3FOJwAAAN2sS09nVb0hyZcs89TFrbVXzjCdPUn2LNzft2/fkc8cRwW1wizUC9NSK8xCvTAttcLRplpr82moal+SZ7XW3jXl+G1e88bWVlVRK0xLvTAttcIs1AvTUivMoqqubK1t+YuxOrwWAACAbubxkylPrqpbkjw2yaur6rW92wQAAGBzmNvhtbNyeC3TcpgKs1AvTEutMAv1wrTUCrNweC0AAACsQegEAACgG6ETAACAboROAAAAuhE6AQAA6EboBAAAoBuhEwAAgG6ETgAAALoROgEAAOhG6AQAAKAboRMAAIBuhE4AAAC6EToBAADoRugEAACgG6ETAACAboROAAAAuhE6AQAA6EboBAAAoBuhEwAAgG6ETgAAALoROgEAAOhG6AQAAKAboRMAAIBuhE4AAAC6EToBAADoRugEAACgG6ETAACAboROAAAAuhE6AQAA6EboBAAAoBuhEwAAgG6ETgAAALoROgEAAOhG6AQAAKAboRMAAIBuhE4AAAC6EToBAADoRugEAACgG6ETAACAboROAAAAuhE6AQAA6EboBAAAoBuhEwAAgG6ETgAAALoROgEAAOhG6AQAAKAboRMAAIBuhE4AAAC6EToBAADoRugEAACgG6ETAACAboROAAAAuhE6AQAA6EboBAAAoBuhEwAAgG6ETgAAALoROgEAAOhG6AQAAKAboRMAAIBuhE4AAAC6EToBAADoRugEAACgG6ETAACAboROAAAAuukeOqvquVV1fVVdXVWvqKr79G4TAACAzWEePZ2vT3Jea+0rktyQ5Bfm0CYAAACbQPfQ2Vp7XWvt7vHu5UnO7N0mAAAAm8Oxc27vaUn+YqUnq2pPkj0L9/ft2zeHWWI7UCvMQr0wLbXCLNQL01IrHG2qtXbkE6l6Q5IvWeapi1trrxzHuTjJ+Um+r03RaFVNMxqkqqJWmJZ6YVpqhVmoF6alVphFVV3ZWjt/o+fjSK1LT2dr7YmrPV9VFyX5ziTfLEkCAAAcPbofXltVT0ryc0m+obV2e+/2AAAA2DzmcfXaFyQ5Jcnrq+qqqvr9ObQJAADAJtC9p7O19tDebQAAALA5zaOnEwAAgKOU0AkAAEA3QicAAADdCJ0AAAB0I3QCAADQjdAJAABAN0InAAAA3QidAAAAdCN0AgAA0I3QCQAAQDdCJwAAAN0InQAAAHQjdAIAANCN0AkAAEA3QicAAADdCJ0AAAB0I3QCAADQjdAJAABAN0InAAAA3QidAAAAdCN0AgAA0I3QCQAAQDdCJwAAAN0InQAAAHQjdAIAANCN0AkAAEA3QicAAADdCJ0AAAB0I3QCAADQjdAJAABAN0InAAAA3QidAAAAdCN0AgAA0I3QCQAAQDdCJwAAAN0InQAAAHQjdAIAANCN0AkAAEA3QicAAADdCJ0AAAB0I3QCAADQjdAJAABAN0InAAAA3QidAAAAdCN0AgAA0I3QCQAAQDdCJwAAAN0InQAAAHQjdAIAANCN0AkAAEA3QicAAADdCJ0AAAB0I3QCAADQjdAJAABAN0InAAAA3QidAAAAdCN0AgAA0I3QCQAAQDdCJwAAAN0InQAAAHQjdAIAANCN0AkAAEA3QicAAADddA+dVfWfq+rqqrqqql5XVWf0bhMAAIDNYR49nc9trX1Fa+3RSV6V5Jfm0CYAAACbQPfQ2Vr7zMTdk5K03m0CAACwORw7j0aq6pIkP5bk00meMI82AQAA2HjV2pF3PFbVG5J8yTJPXdxae+XEeL+QZGdr7ZdXmM6eJHvGu+clueaIZ46jwf2T/PNGzwRbhnphWmqFWagXpqVWmMU5rbVTNnomjtS6hM6pG6s6K8nftdbOm2Lcd7XWzp/DbLHFqRVmoV6YllphFuqFaakVZrFd6mUeV6992MTd70lyfe82AQAA2BzmcU7ns6vqnCT3JLk5yU/MoU0AAAA2ge6hs7X2/Yf5ry9c1xlhO1MrzEK9MC21wizUC9NSK8xiW9TLXM/pBAAA4OjS/ZxOAAAAjl5CJwAAAN0cduisqr+rqvus47xwFJu1nqrqFzvODptcVZ1fVb894//8SlU9a8pxd1fVNYfb1irT/cUl99+2HtNleVX1E1X1YzP+z76qmurS9FV1UVW94HDbWmGau6vqhyfur1v9bQdV9d1V9fOdpj3T97GqvrGqXjXD+DdV1f0Pp61VpnlRVZ0xcf+Pqurc9Zg2h6f3cqe3qrq0qi7c6PnYLKrqtjWev09V/eTE/TOq6i/Xsf0vrmfW20rb3pPbS1X1n6rqiePwM6rqxMNt77AvJNRa+45lZrIynCd6z+FMs6qOba3dfbjzNGNbx7TWDsyjLdZ2GPX0i0l+rfuMsSm11t6V5F0b2dZhLq8OqtvW2uOOcPZYRWvt9ze6rcOok91JfjjJfxunO7da3wpaa3+T5G86TXtu38fl2jrMZcpFSa5J8tFxuv/2yOeOIzHP5Q7r4wjzy32S/GSS302S1tpHk2yJ0L7ctvcy4/zSxN1nJHlxktsPp72pejqr6q+r6sqquraq9oyP3VRV9x/3yr6vqv4sw4LvIStM47aqet44jb+vqtPHx/dV1fOr6l1JfraqHlNVbxrbe21VPWgc72eqan9VXV1VLx0f+4aqumq8/WNVnbJ0z2NVvaCqLpqY5+dU1buT/EBVfWtVvb2q3l1V/72qTj6cN5HZHGk9VdWzk5wwfu57x8d+pKquGB/7g6o6Znz8tqp67tjWG6rqq8eau7Gqvnsc56KqeuX4+Pur6pfn9mZsMePnc/24J/SGqtpbVU+sqreO791Xj7e3j9/Jt9Xwk0mpqmdW1YvG4S+vqmtW2mNWVe8d9x5WVd1a417jqvqzqvqWye/5uEfuRROf689MTOficT7fkuScNV7bY6rqPVX1niT/buLxpW39eVW9NcmfV9XpVfXyqnrneHv8ON7JVfUn4+u4uqq+f4W6vW38W2OdXjP+zw9OtL2vqv5yfN/3VlUdzme3kZZ+52voCXjuxPOTPYb/cVwGvKWqXlIr9E5X1QOq6spx+CurqlXVWeP9f6qqE+vgvbX7alj+XzHWxNePj59QVS+tquuq6hVJTljjtfz4+P9XJHn8xONL25pmvfbQGpZL76lhPfRlSZ6d5OvHOnnmkvq73/heXl1Vl1fVV0y0vex3YCup6ZYvk7VyaVX9dg3LmRtrld6ZqvqdWlzmv6IWl0VPq6pLxuGF7+OK37uqetL42LuTfN8ar+e0qnrdWPd/lKQmnpts681V9TdJ9lfVMeOy4J3j5/x/TvzPvx+XD++pqmePr/f8JHvHejmhJnrMquop4/jXVNVzJtuuqkvG6VxeVQ+c4WPaMmqbLHfGmri0FtcPz5yY9m+Nn/01VfXV4+MnjcuDK2pYD3/PxHQOqa0avGB8/W9I8oAjfe+3gjp0e/M/Trw3v7rM+CfXkF/ePX4O3zM+9ewkXzZ+Ds+tg4+UuryqHjUxjX01HL2y7Ge0ijOq6jU1LAd/Y2J6t00MX1hVl47Dl1bV743t3zguZ1401tulE/9zUy0efbHs9tI4rQtrWK+ckeSNVfXGGpadz58Y7/+oquet+ipaa2vektxv/HtChg/mtCQ3Jbl/hr2y9yT52jWm0ZI8dRz+pSQvGIf3Jfndcfi4JG9Lcvp4/weTvGgc/miS48fh+4x//zbJ48fhkzP03H5jkldNtPuCJBeNwzcl+blx+P5JLkty0nj/3yf5pWneD7cju61TPd02MfzIsRaOG+//bpIfm6i7bx+HX5HkdWOdfWWSq8bHL0rysXE+Fubp/I1+nzbjbfx87k7y5Rl2Wl2Z5EUZNqa+J8lfJ7l3kmPH8Z+Y5OXj8I7xO/fkDD03j1+lnd9P8m+SnJfknUn+cHz8/UlOmvyeJ/mVcblx/FhDt46f8WOSvDfJieM8fSDJs1Zp8+okF4zDz01yzTi8tK0rk5ww3v9vSb5uHD4ryXXj8HOSPH9i2vddWreT95N8f5LXJzkmyQOTfCjJg8a2P53kzPH9e/tCe1vplkO/8w9M8oGJ5/9Hkq9L8r8muSrJziSnjJ/3ap/ZteNn+1NjnTw1ya4kb5/4vJ41Du9L8v+Nw9+R5A3j8P+dxfXMV2So72W//+Nn8qEkpye5V5K3ZnFdtrStadZr70jy5HF451irX6y3Zervvyb55XH4m7K4DPuVLPMd2OjP/TDqZHfWXr5cNPGeX5rkv4/jnjtZU8tM+4eSPHccviLJ5ePwnyT5tnF44fv4jVnmezd+Rh9O8rBxnl42+Vkt0+ZvZ9yuyLA8a0nuv0xbn0ty9nh/T5L/MA4fn2FZeXaSbx8/4xOXfKf2Tdbrwv0MG4cLtXpskn9I8r3jOC3Jd43Dv7HQ3na7Zfssdx6T5PUT9+8zMe2FdeMFWVxn/VqSH1kYN8kNGdabK9XW92Vx/XNGkk8luXCjP7851MfujNubSb41w8+SVIbv/KuyuD2w8F09Nsm9x+H7Z9imqHE61yyZ7sJn8cwkvzoOPyjJ+1b7jFaYz4uS3Jjk1LFGb07ykMl5G4cvTHLpOHxpkpdmcdn5mRy8XH30ON5N42tZcXtpnNaFk+OPwycn+acsbnu/LcmXr/aeT3t47c9U1ZPH4YdkWOBOurm1dvka07gnyV+Mwy9O8lcTzy08fk6GjczX17BT8ZgMYSAZNgj3VtVfZ1jxJMMK/7/U0GvwV621W2rtToCFtr42w0rqreP/3CvDioX+1qOeJn1zhi/MO8fP8oQknxif+0KS14zD701yZ2vtrqp6b4YFw4LXt9ZuTZKq+qsMKyKHtC3vg6219yZJVV2b5O9ba23iPT01yZ9W1cMybNwclySttXtqOOrg6iR/0Fp76yptvDnDSvTmJL+XZE9VPTjJv7bWPrfM9/zVrbU7k9xZVZ/IsHHx9Ule0Vq7fZzXFQ/Jq+Gchvu01i4bH/rzDBt5y/mb1trnx+EnJjl3Yn7uXcMRE0/MsJGb8bX/6yqvNRnq7SVtOOT/41X1pgwbQp9JckVr7ZZxPq/K8B6/ZY3pbTZLv/NnJ7mxqr42wwbeIzIsz382yStba3ckuaOq/naN6b4tQ2/jBRlW4k/KsJJ98wrjL6x3rszi9/+CDOEgrbWrq+rqVdr7miT7WmufTJKq+oskD19h3FXXa1V1SpIHt9ZeMbZ9xzjNVZrP12XYQZHW2j/U0JN27/G55b4Dt6w2sU1qreXLUn/dhkPi9tfqPXZvTvKMGs533J/kvjX0OD82yXI9w8t9724b5+/94+MvzrAhv5ILMvaGttZeXVUrLQeuaK19cBz+1iRfUYu9tqdmWEc+McmfLCzPWmv/skq7ybD8mKzVveP8/HWG9eLCEWFXJvmWNaa1VW2X5c6NSb60qv5rkldn2Hm+4CXjNC6rqnuP67JvTfLdtdhbuzPDTtGVauuCLK5/PlpV/7Dqq99ebm6tXV5Vv5nh/fnH8fGTM7w3l02MW0l+raouyJBpHpxhObual2X4vH45yf+WZOFcz5U+o+tWmM7ft9Y+nSRVtT/DTo4Pr9H2304sOz++ZLm6O8OOlgVTby8taK3dNtbKd1bVdRnC53tX+581Q2dVfWOGhd1jW2u3V9W+DG/OpM+tNZ3l5neZ/68k17bWHrvM+P8mwxfju5JcXFVf3lp7dlW9OsPeo7dW1bdl2Fs0edjwSvNaGYLGUw5j3jlMneqpkvxpa+0XlnnurjbugsmwkLgz+WIAmqz/tuT/lt5n0Z0Tw/dM3L8nwzLlPyd5Y2vtyVW1O8Pe2AUPy7DhdkZWd1mGQ1zPSnJxht7RC7PySn1yng7kCM5Xn8Jkfe7I0Ct/x+QIU+z8msU8X9u6W+U7/9IMK+HrM6zs2mG8b5dlWFnuSvLKDEestAwbZstZeC/n8T6uul4bQ+d62tJ1MmGt5ctq469YQK21j4wb5E/KUDf3y1B/t7XWPrvGdOe5TKkkP91ae+3kCOP2zXqZXC9u5VpZ0XZa7rTW/rWqvjLJtyX5iQzz/7SFp5eOnqGGvr+19r7JJ2p4ocvV1prn9W1jk8vpX2+t/cEq4z41w5EDjxk7L27KoduvBxmXO7fWcCrED2b4/BbaO+QzWsVKy6PJz3/pvEwuO5cuV9frO/9HGa5VcX2Go0ZWNc05nadm6F24vaoekaGH8HDsyOKJtT+c5ffUvy/J6VX12CSpquOq6lFVtSNDV/IbM3y5T01yclV9WWvtva2152Q4xOERGXpGzq2q48cVzDevMD+XJ3l8VT10bOukqlppjzXrZ73q6a6qOm4c/vskF1bVA5Ivnve0a8bpfcv4fyck+d4Mez85PKcm+cg4fNHCg1V1aoY9uxckOa1WOf+qtfbhDId8PKy1dmOG5cWzcvBex7VcluR7azh35pQMO6xWau9TST5VVV83PvTUKdt4XZKfXrhTVY8eB1+fg88Lve84OFm3k96c5AdrOOfm9Azv0RVTzsNmt9J3/hUZDvt5SoYNwWT43n1XVe0ce4y/c41pvznJjyR5/9jb9S8ZdkLO0hN8WYZ1UqrqvAyHuq3kHUm+YexhPC7JD0wx/WXXa2PQuaWqvnd8/PgaznH+bIZD/Jbz5oy1OW5U/3Nr7TNTzAODyzNcCOOyDO/ls7LyjqzlXJ9kdw3n3iZD7a5msra+Pcl9Vx89SfLaJP/XwnKiqh5eVSdlWKb8+Fgjqar7jeOvVC9XZKjV+9dwjYOnJHnTFO1vF9tmuVPDOXc7WmsvT/IfkvwvE08vnP//dUk+PfaGvTbJT48hM1X1VeO4K9XWZVlc/zwoyRNmeB3bxWuTPG38/FNVD17YppxwapJPjIHzCRl2OiSrL7OT4aiXn0tyamttoUd7pc9oVh+vqkeOOenJa469smm3lw56ra21d2Q4iuCHM/a6r2aa0PmaJMeOXafPzrDQPhyfS/LVNZxc+01J/tPSEVprX8gQTJ9Tw8U8rkryuAyHI7147CL+xyS/PW4kPqOGk6evTnJXkv8xbqy+LMPx+y/LYlf50rY+mWGD+CXj/789Q2ilr/Wqpxcmubqq9rbW9mdYEL9u/Cxfn+HY+VlckeTlGQ79fHkbrhjJ4fmNJL9eVf+Yg/emPS/J77TWbkjy9CTPXmahPukdGc5zSIaV/IMzw0q9tfbuDAv792Q4d+eda/zLjyf5nRoOpZt21/fPJDm/hgsP7M/iXsz/N8Phe9eMy7KFlfgX63bJdF6Rofbek+Hcq59rrf3/U87DZrfsd3485Pi6JLtaa1eMj70zw5VJr87wmb03w7l1y2qt3ZThs1rYGfGWJJ+a4nDmSb+XYSfmdRnWS1eu0t7HMpyv9fYMG6orHQo1+T8rrdeS5EczHAJ4dYZD9r4kw2s/UMNFXp65ZHK/kuQx4/jPTvK/T/cSGb05w/nmH0jy7gy9nVOHzvGIhj1JXl3DhYQ+sca//GqSC2o4nO37MpxjuZY/ynD477vH7aU/GOf5NRm+G+8al1ELh+VdmuT3a7yQ0MS8fizJzyd5Y4blypWttVdO9UK3h22z3Mmw7ts3fu4vTjJ5VNcd47r29zOsV5PhaKPjMqxrrh3vJyvUVob1z/vH5/4sR+GpZq2112W4RsPbx6zxlzk0SO7NsL5/b5Ify7ATKm04Neut4/r+uTnUX2Y43eZlE4+t9BnN6uczHCr/tiyejjizGbaXXpjkNVX1xonHXpbkrdPUfy0eYdFXVd3WWnN1WDadGs4zPL+19lMbPS9wtKuqk8dzRU7MsFG3Z1whAnSxFZc7NRwy/Cw7ydlINVxd/Xmttb9fa9xtdxw/AFvaC2u42MvODOdqb+oNP2BbsNyBGYynMF6R5D3TBM6kQ09nVb0jw6WYJ/3oWlc0guWop+2tqn48w5UDJ721tfbvlht/ndr8nUz8vuLot1pra54Ez8bYiM/Msmdrqqovz3D16Ul3tta+pmObc1+O0Z/lDvNUw8XCnrPk4Q+21o7kXM1NZW6H1wIAAHD0meZCQgAAAHBYhE4AAAC6EToBAADoRugEAACgG6ETAACAbv4nIW/APDhXploAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "parallel_plot(P[P['air_temp'] > 0.2])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "56a728ff",
   "metadata": {},
   "source": [
    "### Cool days"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "772d200a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "parallel_plot(P[(P['relative_humidity'] > 0) & (P['air_temp'] < 0.2)])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b91997e3",
   "metadata": {},
   "source": [
    "   All the cluster centroids which fall under a category sunny or cool or dry indicates that the cluster points of particular centroid exhibit sunny or cool or dry nature ."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "95efd2be",
   "metadata": {},
   "source": [
    "### Drawbacks of K-Means clustering:\n",
    "We have to decide number of clusters in before hand and it always tries to make clusters of same size.\n",
    "Lets implement another clustering algorithm(Hierarchical Clustering)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2e7f8f2e",
   "metadata": {},
   "source": [
    "## Hierarchical Agglomerative clustering(HAC):\n",
    "\n",
    "   In this algorithm we assign each point to individual cluster and at each iteration we merge closest pair of clusters and repeat this process until a single cluster is left. \n",
    "   How do we decide which points are similar?Take distance between centroids of clusters and the points having least distance can be merged.It is a distance-based algorithm. We store all the distances between any two points in a proximity matrix. \n",
    "    \n",
    "### Choosing number of clusters:\n",
    "   To get number of clusters in HAC we use dendrogram. A dendrogram is a tree like structure which records sequence of merges/splits.We set a threshold and draw a line. We choose it in such a way that it cuts the tallest vertical line. Total number of clusters will be number of vertical lines intersected by threshold.\n",
    "\n",
    "Lets go ahead and implement what we have learnt. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "35df3fed",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlgAAAGuCAYAAABMRXhFAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8qNh9FAAAACXBIWXMAAAsTAAALEwEAmpwYAABECklEQVR4nO3deZxcV33n/e+vNy2WhC25sWx5EZvVj8HEMGIxeeYJCcnEGCcmYTEksY2GiUnCJCFhwpYJZGZIYJ4QCMkAwYwtzO4FEjtCeB5gWMIDGNpgLGO6jTGWJSNZ7W7ZWnqvPvPHPafq9O1b+62uqu7P+/XqV1fX3c69davut885da455wQAAID89LS7AAAAACsNAQsAACBnBCwAAICcEbAAAAByRsACAADIGQELAAAgZwQsAF3HzF5jZt9odzkAoBwCFoBcmNmDZjZlZsfN7DEz+6aZ/Z6Z8TkDYNXhgw9Ann7NObdR0nmS3i3pzZKuW84CmFlfJ68PwOpAwAKQO+fc48652yRdIelqM3uGma0xs/eY2UNm9oiZ/aOZrZMkM3uhmR00szea2REzO2Rmu8L6zGyLmd1mZsfM7DuSnhJvz8ycmb3ezH4s6cf+ud81s/vNbMIve1Y0/78zs1Eze9zMPmhmXzOz/+CnvcbM/n8ze5+ZjUv6SzN7ipn9bzMbN7NHzeyTZnZqtL4HzezPzOxuMztpZteZ2Rlm9gVfo/clMzutdUccQKchYAFoGefcdyQdlPRvldRonS/pIklPlbRN0tuj2bdKeoJ//rWSPhCFkg9ImpZ0pqR/73/SXirpeZIuMLNfkvQuSa/0y+yX9BlJMrPTJd0i6a2StkgalfSC1LqeJ+kBSWdI+itJ5td3lqT/S9I5kv4ytczLJP2K38dfk/QFSW+TNKjks/aPyhwmACsQAQtAq/1M0mZJ10j6E+fchHPuuKS/lvSqaL45Sf/VOTfnnNsr6YSkHWbWqyS8vN05d9I5d4+kGzK28y6/7ilJvy3peufc95xzM0rC1MVmtl3SpZJ+6Jz7nHNuXtLfSzqcLrNz7h+cc/POuSnn3P3OuS8652acc2OS3ivpF1LL/INz7hHn3MOS/lXSHc657zvnpiX9k6RnNXLwAHQn+hYAaLVtSj5r1ku608zC8yapN5pv3AeeYFLSBiU1QH2SDkTT9mdsJ55+lqTvhT+ccyd8c982P+1ANM2Z2cEK65KZnSHp/Upq4jYq+ef0aGqZR6LHUxl/b8goM4AVihosAC1jZs9REmr+WUnIeLpz7lT/8wTnXC2hY0zSvJJmueDcjPlc9PhnSjrah3KcoqQ58GFJhySdHU2z+O+MdUlJbZuTdKFzbpOk31ESEAEgEwELQO7MbJOZXaak39MnnHM/kPQRSe8zsyf6ebaZ2a9WW5dzriDpc0o6m683swskXV1lsU9L2mVmF5nZGiUB6Q7n3IOSPi/pQjN7qf+G4OuV9P+qZKOSJsvHzWybpD+rVm4AqxsBC0Ce/sXMjitpYvtzJX2VwrcB3yzpfknfNrNjkr4kaUeN6/2PSprYDkv6qKTdlWZ2zn1J0l9I+qySGqunyPf3cs49KukVkv5fSeOSLpA0LGmmwir/i6RnS3pcSUD7XI3lBrBKmXPpmnAAWD38QKgHJf22c+4r7S4PgJWBGiwAq46Z/aqZneqbD9+mpD/Vt9tcLAArCAELwGp0saSfSHpUyZhVL/XDOwBALmgiBAAAyBk1WAAAADnriIFGTz/9dLd9+/Z2FwMAAKCqO++881Hn3GCleaoGLDO7XtJlko44554RPf+HSsaPKUj6vHPuTf75tyq5j1hB0h855/5XtW1s375dw8PD1WYDAABoOzPLupvEIrXUYH1U0v+Q9LFoxb8o6XJJP+ecm4kGDrxAyVgzT1dyO4ovmdn5fqBAAACAVaFqHyzn3NclTaSe/n1J7/Y3UZVz7oh//nJJn/E3RP2pkkEFn5tjeQEAADpeo53cz5f0b83sDjP7mr/fmJTccyy+SepB/9wSZnaNmQ2b2fDY2FiDxQAAAOg8jQasPkmbJT1fyT25bvI3TK2Zc+5a59xO59zOwcGK/cQAAAC6SqMB66Ckz7nEdyQtSDpdyZ3q4zven+2fAwAAWDUaDVj/LOkXJcnMzpc0oGRE5NskvcrM1pjZkyQ9TdJ3cignAABA16hlmIZPS3qhpNPN7KCkd0i6XtL1ZnaPpFlJV7tkSPgfmtlNku6VNC/p9XyDEAAArDYdcaucnTt3OsbBAgAA3cDM7nTO7aw0D7fKAQAAyBkBCwAAIGcELAAAgJwRsAAAAHJGwAIAAMgZAQsAACBnBCwAAICcVR1oFJV96o6HdOtd3A0IQGtdftE2/dbzzm13MQDUiBqsJt1618O699CxdhcDwAp276Fj/CMHdBlqsHJwwZmbdOPrLm53MQCsUFd8+FvtLgKAOlGDBQAAkDMCFgAAQM4IWAAAADkjYAEAAOSMgAUAAJAzAhYAAEDOCFgAAAA5I2ABAADkjIAFAACQMwIWAABAzghYAAAAOSNgAQAA5IyABQAAkDMCFgAAQM4IWAAAADkjYAEAAOSMgAUAAJAzAhYAAEDOCFgAAAA5I2ABAADkjIAFAACQMwIWAABAzghYAAAAOSNgAQAA5IyABQAAkDMCFgAAQM4IWAAAADkjYAEAAOSsasAys+vN7IiZ3ZMx7Y1m5szsdP+3mdnfm9n9Zna3mT27FYUGAADoZLXUYH1U0iXpJ83sHEn/TtJD0dMvlvQ0/3ONpA81X0QAAIDuUjVgOee+LmkiY9L7JL1Jkoueu1zSx1zi25JONbMzcykpAABAl2ioD5aZXS7pYefcD1KTtkk6EP190D+XtY5rzGzYzIbHxsYaKQYAAEBHqjtgmdl6SW+T9PZmNuycu9Y5t9M5t3NwcLCZVQEAAHSUvgaWeYqkJ0n6gZlJ0tmSvmdmz5X0sKRzonnP9s8BAACsGnXXYDnn9jnnnuic2+6c266kGfDZzrnDkm6TdJX/NuHzJT3unDuUb5EBAAA6Wy3DNHxa0rck7TCzg2b22gqz75X0gKT7JX1E0h/kUkoAAIAuUrWJ0Dn36irTt0ePnaTXN18sAACA7sVI7gAAADkjYAEAAOSMgAUAAJAzAhYAAEDOCFgAAAA5I2ABAADkjIAFAACQMwIWAABAzghYAAAAOSNgAQAA5IyABQAAkDMCFgAAQM4IWAAAADkjYAEAAOSMgAUAAJAzAhYAAEDOCFgAAAA5I2ABAADkjIAFAACQMwIWAABAzghYAAAAOSNgAQAA5IyABQAAkDMCFgAAQM4IWAAAADkjYAEAAOSMgAUAAJAzAhYAAEDOCFgAAAA5I2ABAADkjIAFAACQMwIWAABAzghYAAAAOSNgAQAA5IyABQAAkDMCFgAAQM4IWAAAADkjYAEAAOSsasAys+vN7IiZ3RM99zdmNmJmd5vZP5nZqdG0t5rZ/WY2ama/2qJyAwAAdKxaarA+KumS1HNflPQM59wzJd0n6a2SZGYXSHqVpKf7ZT5oZr25lRYAAKALVA1YzrmvS5pIPff/Oefm/Z/flnS2f3y5pM8452accz+VdL+k5+ZYXgAAgI6XRx+sfy/pC/7xNkkHomkH/XNLmNk1ZjZsZsNjY2M5FAMAAKAzNBWwzOzPJc1L+mS9yzrnrnXO7XTO7RwcHGymGAAAAB2lr9EFzew1ki6T9CLnnPNPPyzpnGi2s/1zAAAAq0ZDNVhmdomkN0n6defcZDTpNkmvMrM1ZvYkSU+T9J3miwkAANA9qtZgmdmnJb1Q0ulmdlDSO5R8a3CNpC+amSR92zn3e865H5rZTZLuVdJ0+HrnXKFVhQcAAOhEVQOWc+7VGU9fV2H+v5L0V80UCgAAoJsxkjsAAEDOCFgAAAA5I2ABAADkjIAFAACQMwIWAABAzghYAAAAOSNgAQAA5IyABQAAkDMCFgAAQM4avtkzgO7wqTse0q13cc/1bnbvoWOSpCs+/K02lwTNuPyibfqt553b7mJgmVCDBaxwt971cPECje50wZmbdMGZm9pdDDTh3kPH+EdnlaEGC1gFLjhzk2583cXtLgawalH7uPpQgwUAAJAzAhYAAEDOCFgAAAA5I2ABAADkjIAFAACQMwIWAABAzghYAAAAOSNgAQAA5IyABQAAkDMCFgAAQM4IWAAAADkjYAEAAOSMgAUAAJAzAhYAAEDOCFgAAAA5I2ABAADkjIAFAACQMwIWAABAzghYAAAAOSNgAQAA5IyABQAAkDMCFgAAQM4IWAAAADkjYAEAAOSMgAUAAJAzAhYAAEDOqgYsM7vezI6Y2T3Rc5vN7Itm9mP/+zT/vJnZ35vZ/WZ2t5k9u5WFBwAA6ES11GB9VNIlqefeIunLzrmnSfqy/1uSXizpaf7nGkkfyqeYAAAA3aNqwHLOfV3SROrpyyXd4B/fIOml0fMfc4lvSzrVzM7MqawAAABdodE+WGc45w75x4clneEfb5N0IJrvoH9uCTO7xsyGzWx4bGyswWIAAAB0nqY7uTvnnCTXwHLXOud2Oud2Dg4ONlsMAACAjtFowHokNP3530f88w9LOiea72z/HAAAwKrRaMC6TdLV/vHVkm6Nnr/Kf5vw+ZIej5oSAQAAVoW+ajOY2aclvVDS6WZ2UNI7JL1b0k1m9lpJ+yW90s++V9Klku6XNClpVwvKDAAA0NGqBizn3KvLTHpRxrxO0uubLRQAAEA3YyR3AACAnBGwAAAAckbAAgAAyBkBCwAAIGcELAAAgJwRsAAAAHJGwAIAAMgZAQsAACBnBCwAAICcEbAAAAByRsACAADIGQELAAAgZwQsAACAnBGwAAAAckbAAgAAyBkBCwAAIGcELAAAgJwRsAAAAHJGwAIAAMgZAQsAACBnBCwAAICcEbAAAAByRsACAADIWV+7CwAA6ExHb7xJx/bsaXcxVoSZ039RkrT/yg+1uSTdb9Nll+m0K17Z7mJURcACAGQ6tmePpkdGtHZoqN1F6Xrvf/Qr7S7CijA9MiJJBCwAQHdbOzSk8z7+sXYXA5Ak7b/yqnYXoWb0wQIAAMgZAQsAACBnBCwAAICcEbAAAAByRsACAADIGQELAAAgZwQsAACAnBGwAAAAckbAAgAAyBkBCwAAIGcELAAAgJwRsAAAAHLWVMAysz8xsx+a2T1m9mkzW2tmTzKzO8zsfjO70cwG8iosAABAN2g4YJnZNkl/JGmnc+4ZknolvUrSf5f0PufcUyUdlfTaPAoKAADQLfpyWH6dmc1JWi/pkKRfkvRbfvoNkv5S0ocqrmV0VHrhCxc/98pXSn/wB9LkpHTppUuXec1rkp9HH5Ve/vKl03//96UrrpAOHJCuvHLp9De+Ufq1X0u2/brXLZ3+n/+z9Mu/LN11l/SGNyyd/td/Lb3gBTr/J/v0qn/+R+nTmxZP/7u/ky66SPrSl6R3vnPp8h/+sLRjh/Qv/yL97d8unf7xj0vnnCPdeKP0oYzDd8st0umnSx/9aPKTtnevtH699MEPSjfdtHT6V7+a/H7Pe6Q9exZPW7dO+sIXksf/7b9JX/7y4ulbtkif/Wzy+K1vlb71rcXTzz5b+sQnksdveENyDGPnny9de23y+JprpPvuWzz9oouS4ydJv/M70sGDi6dffLH0rnclj1/2Mml8fPH0F71I+ou/SB6/+MXS1NTi6ZddJv2n/5Q8Tp93Utece/rmN6W3vW3p9NS59/afHUueD+co517ymHNv6fTUuXfGj0YWl7XOc28Jzr3kMefe0uk1fu6tGTuiU79/19J9aPe5l6HhGizn3MOS3iPpISXB6nFJd0p6zDk372c7KGlb1vJmdo2ZDZvZ8NzcXKPFAAAA6DjmnGtsQbPTJH1W0hWSHpN0s6RbJP2lbx6UmZ0j6Qu+CbGsnTt3uuHh4YbK0W5XfDj5L+bG113c5pIA2ThH0aj9V14lSTrv4x9rc0mARKeck2Z2p3NuZ6V5munk/suSfuqcG3POzUn6nKSfl3SqmYWmx7MlPdzENgAAALpOMwHrIUnPN7P1ZmaSXiTpXklfkRQaaK+WdGtzRQQAAOguzfTBukNJk+D3JO3z67pW0psl/amZ3S9pi6TrcignAABA12jqW4TOuXdIekfq6QckPbeZ9QIAAHQzRnIHAADIGQELAAAgZwQsAACAnBGwAAAAckbAAgAAyBkBCwAAIGcELAAAgJwRsAAAAHJGwAIAAMgZAQsAACBnBCwAAICcEbAAAAByRsACAADIGQELAAAgZwQsAACAnBGwAAAAckbAAgAAyBkBCwAAIGcELAAAgJwRsAAAAHJGwAIAAMgZAQsAACBnBCwAAICcEbAAAABy1tfuAjRseLe075Z2l0I6fHnye/c721sOSbrw5dLOXe0uBQAAq173Bqx9t0iH90lbL2xrMW4899a2br/o8L7kNwELAIC2696AJSXhatfn212KzrD7Je0uAQAA8OiDBQAAkLPursECAABd4eiNN+nYnj1NrWN6ZESStP/Kq5paz6bLLtNpV7yyqXVUQw0WAABouWN79hQDUqPWDg1p7dBQU+uYHhlpOujVghosAACwLNYODem8j3+srWVotvarVtRgAQAA5IyABQAAkDOaCIEKbr7vZu19YG+7i9GU0YlfkCTtuv3aNpekeZc++VK94vxXtLsYAFAVAQuoYO8DezU6Maodm3e0uygNe9azvtbuIuRidGJUkghYALoCAQuoYsfmHdp9ye52F2PV23U7dykA0D3ogwUAAJAzAhYAAEDOmgpYZnaqmd1iZiNm9iMzu9jMNpvZF83sx/73aXkVFgAAoBs0W4P1fkm3O+eGJP2cpB9JeoukLzvnnibpy/5vAACAVaPhgGVmT5D0/0i6TpKcc7POucckXS7pBj/bDZJe2lwRAQAAukszNVhPkjQmabeZfd/M/qeZnSLpDOfcIT/PYUlnNFtIAACAbtJMwOqT9GxJH3LOPUvSSaWaA51zTpLLWtjMrjGzYTMbHhsba6IYAAAAnaWZgHVQ0kHn3B3+71uUBK5HzOxMSfK/j2Qt7Jy71jm30zm3c3BwsIliAAAAdJaGA5Zz7rCkA2YWhrh+kaR7Jd0m6Wr/3NWSbm2qhAAAAF2m2ZHc/1DSJ81sQNIDknYpCW03mdlrJe2X9MomtwEAANBVmgpYzrm7JO3MmPSiZtYLAADQzRjJHQAAIGfc7BkA2uzojTfp2J497S7GEtMjI5Kk/Vde1eaSLLXpsst02hX0QEHnogYLANrs2J49xTDTSdYODWnt0FC7i7HE9MhIRwZSIEYNFgB0gLVDQzrv4x9rdzG6QifWqAFp1GABAADkjIAFAACQMwIWAABAzghYAAAAOSNgAQAA5IyABQAAkDMCFgAAQM4IWAAAADkjYAEAAOSMgAUAAJAzAhYAAEDOCFgAAAA5I2ABAADkjIAFAACQMwIWAABAzghYAAAAOSNgAQAA5IyABQAAkDMCFgAAQM4IWAAAADkjYAEAAOSMgAUAAJAzAhYAAEDOCFgAAAA5I2ABAADkjIAFAACQMwIWAABAzghYAAAAOSNgAQAA5IyABQAAkDMCFgAAQM762l2AFW94t7TvltZv5/Ddye/dL2n9ti58ubRzV+u3gxXj5vtu1t4H9ja1jpGJEUnSrtubP/cuffKlesX5r2h6PQBQDjVYrbbvFunwvtZvZ+szk59WO7xveQIjVpS9D+zV6MRoU+sY2jykoc1DTZdldGK06bAHANVQg7Uctl4o7fp8u0uRj+WoIcOKtGPzDu2+ZHe7i5FLDRgAVEMNFgAAQM6aDlhm1mtm3zezPf7vJ5nZHWZ2v5ndaGYDzRcTAACge+RRg/XHkn4U/f3fJb3POfdUSUclvTaHbQAAAHSNpgKWmZ0t6SWS/qf/2yT9kqTQC/oGSS9tZhsAAADdptkarL+T9CZJC/7vLZIec87N+78PStqWtaCZXWNmw2Y2PDY21mQxAAAAOkfDAcvMLpN0xDl3ZyPLO+eudc7tdM7tHBwcbLQYAAAAHaeZYRp+XtKvm9mlktZK2iTp/ZJONbM+X4t1tqSHmy8mAABA92i4Bss591bn3NnOue2SXiXpfzvnflvSVyS93M92taRbmy4lAABAF2nFOFhvlvSnZna/kj5Z17VgGwAAAB0rl5HcnXNflfRV//gBSc/NY70AAADdiJHcAQAAckbAAgAAyBkBCwAAIGcELAAAgJzl0skd6GQ333ez9j6wt6FlRyZGJEm7bt9V97KXPvlSveL8VzS0XQBAd6MGCyve3gf2anRitKFlhzYPaWjzUN3LjU6MNhzqAADdjxosrAo7Nu/Q7kt2L9v2GqnxArrN0Rtv0rE9e5Z9u9MjSc3y/iuvWvZtb7rsMp12xSuXfbvoPtRgAQAacmzPnmLYWU5rh4a0dqj+muVmTY+MtCVQojtRgwUAaNjaoSGd9/GPtbsYy6IdNWboXtRgAQAA5IyABQAAkDOaCFeD4d3SvlvyWdfhu5Pfu1+Sz/oufLm0kw7hAICVhRqs1WDfLdLhffmsa+szk588HN6XX/ADAKCDUIO1Wmy9UNr1+XaXYrG8asEAAOgwBCwAALBiVBufrdZx1Jod84wmQgAAsGJUG5+tlnHU8hjzjBosAACwojQ7PlseY54RsAAAXWm5b9XTjlv0cGue7kUTIQCgKy33rXqW+xY93Jqnu1GDBQDoWiv5Vj3demuecjWLlWoAV2JNHTVYAAAgN+VqFsvVAK7UmjpqsAAAQK7qqVns1pq6aghYQAU333ez9j6wt+7lRiaS/9523d7YbYAuffKlesX5r2hoWQBA+xGwgAr2PrBXoxOj2rF5R13LDW1uvCPs6MSoJBGwgC6V17cb8/zW4krs49TpCFhAFTs279DuS3Yv2/YarfUC0BlCH6Rmv3GY1zcWQ1AjYC0vAhYAADnrpG83rtQ+Tp2OgJWH4d3Svluypx2+O/ld7sbGF75c2kmNBQAAK0lnBqxKgSWoFlxirQ4x+26RDu+Ttl64dNrWZ5Zf7vC+5DcBCwCAFaUzA1alwBJUCi6x5QoxWy+Udn2+vmVqCYcAAKDrdGbAkhoLLFkIMQAAYJl1bsACAABL1DsMRKPDPTC0Q3O4VQ4AAF2k3ptcN3KT6pV6+5rlRA0WgBWp3Cj8lUbZZwR9dItWDwPB0A7NowYLwIoURuFPG9o8lDnS/ujEaEO3RQKALNRgAVix6hmFnxH0AeRpZQSsRgf6ZJBPAC1WS4fkejoh0/EY6A4ro4kwjJuVZeszs8fMOryv+mCmANCkWjok19oJmY7HQPdYGTVYUv3jZjE+FpZRuQ7XWSp1wk5rd6fsWver1n1q9/60Sl4dkul4DHSPhgOWmZ0j6WOSzpDkJF3rnHu/mW2WdKOk7ZIelPRK59zR5osKdK/Q4XrH5h1V583qgJ0ldOBuZyCpdb9q2adO2B8Aq1O6KT+r2b7e5vlmarDmJb3ROfc9M9so6U4z+6Kk10j6snPu3Wb2FklvkfTmJrYDrAj1dLiuRad0ys5rvzplfwCsPqEpPzTVp5vsQ+BaloDlnDsk6ZB/fNzMfiRpm6TLJb3Qz3aDpK+KgAWgQzA+FoAslZryG2mez6WTu5ltl/QsSXdIOsOHL0k6rKQJMWuZa8xs2MyGx8bG8igGAFTF+FgAlkPTndzNbIOkz0p6g3PumJkVpznnnJm5rOWcc9dKulaSdu7cmTkPALQC42MBaLWmApaZ9SsJV590zn3OP/2ImZ3pnDtkZmdKOtJsIdGFKo1NFlQaoyy2Sscrq/YNvdX+zTwA6GQNNxFaUlV1naQfOefeG026TdLV/vHVkm5tvHjoWpXGJgvKjVEWW8XjlZVrygrKNWnFaN4CgPZopgbr5yVdKWmfmd3ln3ubpHdLusnMXitpvySGHF6t6h2bLMsqH6+s2W/o0bwFAO3RzLcIvyHJykx+UaPrBYDVKM9b6uRxO528ysOtfbBarZyR3IEmZfV5KtfPiX5NyFt6HJ4std5OR6pvvJ5WlSevsqB1ygXpcuGZwFw7AtZKUq5jebnO5Ku083g5WaOSl/vavsSI48hfLbfUqbVmqVytUj0XyGZv8cOtfTpfuSCdFZ4JzPUhYK0koWP51gsXP1/uZtcSASullj5P9GtCO1WrWapUq8QFEllqDdIE5vqs/IBVb61O0K21O7V2LF/lnceBbtZozRIXyPxUqkms1DeNJrbVY+UHrHpqdQJqd9ojKwxXCsLdGoIBdL1KNYnlahGpQVxdVn7AkuofLoDanfbICsPlgjAhGECb1VuTSA3i6rI6AlanSNfQ0Pl8KZo4AQArAAFrOaVraOh8DnSUWofqYJiO5VXvUAIB/Z3QTgSs5VathoaaGaBtahmqg2E6ll89QwkEK6W/U1a4zAqWqzVM1np8pOU/Rp0TsOLms3TT2UpuMqul2bDZ/a+n8/hKPtYrVKWbQle6ITS1MNmqDdXBMB3tsVr7O2WFy3SwXClhshG1HB+pPceocwJW3HwWN51VajILwSEOC90WEKo1G+bRZFhr53GaJ7tSVq1LUO5m0NTCAN2jWrjstjCZrnVqtkaulvDdjmPUOQFLym4+q9RkVgwOPix0a0Co1GyYV5NhLZ3Hm9kWHfjbqt6bQrezFqZcjVul2jaJGrdO18lNNe1SS5CQVs/xCNK1Tiu1Rq6zAlYj4uCwXP2XCBNL0YG/iI7SlZWrcStX2yZR49YNOrmppl2qBQlpdR2PWKVap26rkSun+wNWOxAmstGBXxIdpWvRTTVuqF2nNtW000pr3kPtCFiNIkwsvzq+CBHXIqVrj5aj5oiO0ugUcTNVuolqJTRNNTKEw0rYbyyPcH6F8+nojTfVfO70tLJgXWV4d3Kx3v2S5OJ9+O7S38O1/6eNFgo1h1JSaxj3vUt9SzLUIklJ7VGoQRqdGC37jbtOcvN9N2tkYkQjEyPadfsu3Xzfze0uErpUfHFYOzRUbKaaHhkpey+9bhLvXyze19hK2e9OdPTGm7T/yqs0PTKi6ZER7b/yKh298aZ2F6sp6Wbees4darCCRr7FiOVXxxchsmqRuqXmaO8De2Uy7di8Y9U3J9Yr1F6Gmsub77t51R+7rGaqldQ0Vc8QDs3ud7pGYyXVBjYrHUbq7V+WdWxbfVxrqaEK51e95w4BK1bvtxiBFgoBsVtCYd4abeYNtZdDm4eKNZadGrDKNd9xsW5evU2HtR7zrM78q7WjepY47NYbSJoNaI2ItxlqN/PaHgErbythbC60RDv7haW/2VhuSIRO+mZj/GWB+EsCtdTodUs4jT/cl/OishrUM/p7vcc8XWO2kmoD262ZgNbsNvPe3soNWO0KOitlbK68De8uvRbDu1fl8WgmMOS5bSl7SIRObIrs5mbeWtVysa7UUV3Kr3lqpXWIr7XpkIBUWbc2i7a7hrizA1Z8Ua43ILUz6LRjbK5GLdctivbdIslKj5c5YKX75ey6fVdbamvaGRj4ZmP3yqrpCvKs8Sq3HWrVVrdubRZtdw1xZwescFHeemFjAambgk67LGfn/vhWPcss7pcjLa2tCd/ak7LDV7uHfVjtyh3/Wo99q5YPluMcSNfGpP87z+s/81Z0iF+uGrhuL9NyqrdWqt5m0aM33tTQ0Aa1qKdmKi53KFP8fmmlzg5YUikkdUJAWs5mx+XcVjs795erQWvBvsY1OOnammrf2mtX814c/OJvwzUbGJZbswE16/jXc+xbsXzQrqbVdnQIblS6JmFubKxY3oXjxxcNnbBcwWa5agU7VatrpeIvGOTZcTysr5GaqVCm0KE91opA2BkBa/JR6fAjyeNG++eE5sRTBvMtW2w5mx1XS1+urBq0Nu1rtY7R7WjeC8EvPA4X8WYDw3LLI6Cmj3+9xz7v5RtdTyXxh3wtNVKNdAiuZ+DEPL82ny5rYXy8JcGmnjKX66PV6j5ZzfR1y/Oej4121q/1PM36QkFeGi17uQ7trQiEHRKwjkp6QvK4lv456dqd8PfMsaW3rcm7JqieZsd0x+56y7X1wuS5fbdIcqXBT6XF81ZaVzP92JZLugatidqzlTgGUvqeffHzzQSG5daOgDp6NAlwO07LPoadJus/7EY/6MsFqXq+lt7KWrJWjc3VDTV7zfR1K3fPx7hWUGptzWCj52m94X65O6b3btmiwvi4pkdGlpQtlKd3y5aa19cZAUuqr39OXLuTMYp32Xml+mpH0uHk+CFp45n1lTPu2N1IudLzZc1baV37bpEKs1LvgLT/G8n+7LulfNCqNZBWu+F1nkEuK1CH8Jh6fu+jX2/bGEhZHekl+mi10/HZ4zXPW60fXtb81cJ8uYtEpQtNXl8ZrxSk6tlGmDer/0qrL3iN1qAt11f9m6nhayZglls2q1ZQajxk5n2e1hvuw3aqlT+vMFYYH1+0/Xgdx/bs0cLx43XVynVOwKpXPX2zKtU6VQoUSzrZu/oCVti2JB0/LJ08srgJs9basFpqeCqtq3cg2Y/z/u/k70ohs+7gl3HD63LLHD8knRxLHtfTFJwVqMPrk37+zCdWberLupCWU+7bh5KWXFzTHeml9jfbVfv2ZKd23k/3PWvl+sMxqXf0/FoGNA0Xid4tWzT53e8WLwKtHNwwluf4PuX6r7Tya/DLWRvVyFf6K5WvkRqPrDKly1JJ3k2eeZync2NjKoyPF2+Zs3ZoSJsuu6wY1ivVYtU6b541v3k2a3ZOwKp08a3U1NasaoEiDnKhDI04eSS7CbNe1caTKnesspobyx3LRoNfUG6Zk2NqeKiGcoG6gS9BZF1IK82b9e3D8Di+uEr5NNuFi/+WdVs0uK65PoXVvj3ZzrG5Kkn3PavH2NSYxqfGKwazcudAvQOU1jJ/+MAO/x3HF4NWDG7YSukyL8fX4JerNiqrI34cisP+ZAWGrPLVW+OR1cm6UqfsWtdZSw1bufni/Wv02PcPDmruwQeX9HGKH9fSMT1+3Mqa3zx1TsCqdPGt1NSWJQ4ZpwxWr3WqFijC+gqz2dtYzr5N1caTqrVZslrTaqtUawpepgFJy10Ys2pOyn37sJ6LcaVvA2bVpJyYO5E5GGi9669U/qzpQdZxyQo65UaElyrXgNUyLlm5vmdh+XLLDq4b1P5j+6sGs1rOgbf861v0yMlHKjYD1tq02MoOv638Snw11Tobt7NszX5pIA7FYbmswJDeViM3OC4XOpoJDrXWAJabL6h33+Lau/7B7H8Ss94P6fCaNW+rvp04PTqqhcnJXNYVdE7AkipffOvtoxVCxsmx+pv1stY3c0xas2npNhodo6sZ8bHIqrGqdhxbOexFs8GzUoCsoyaz0SamSjUn8TprqV2K5x+bGiv7bUCTafO6zRp+ZLhYcyWpbC1WVpgq923DasvVKj0SfJAVAsemxrT/2H6NTY3VdL9Aqf4as1beDDs+ll878DWZrGIzYDNlqeXClRVQ0s+18ivxtarUqb5a2cotm77g1hvUmm06qicUl6uhicX76WZnNfqc5xabwerdXq1qrQGsNF+5fUufh9OjpdrgevsrSeXDa1ZZGxHOp4f/7E1LAuDCsWNl55NUbOoNTZW16KyAVa90TVUshIy4WS/r4t+MdFjJCgCh6TPUfmU1zTVTa1Nv7V49yh2vSp3g8wie5QJiuX2Ny7nhGdLGrTU1McWBY6B3QLOFWW1Zt6VszUm8zvGpcQ2uG6zYnJeef8u6LRqfGtfIxMiigBO2NzE1UZx3Q/8GnZg7UVzX+FTyn/Su23eVDWuVanzS5anU+b9cH7VqI8EHu27fpf3H9lcNoNVq1kJTX7kwW08N4l1H7tKCFpYEy3KhORzLkYmRstuJy5c1T3zxKfeffC0X5ayAkrVcvRedrHBXT3+frHJW6lTfyLLlLrhhei21Y/X0+WlW/C20dN+rozfepCPveY8Wjh9X//btKoyPa2FysqGmv1Yqd95mvYbp8zAOKZIWHYd0oAkhJmwzfk2mR0e1dkf93/yt9g9LOJ9OfPWrxQAYlyNrvqAwPq6ejRu1cPy4jrznPTWVpzMC1vyUVFi7tB9WNemaqlrnjy/+sfhC/dnfLY2rFWrACrOV+y5lBYCTY0ntl/UmPzPHspvuCrPS/LT0zX9Injt8t9S3Rpo5LhXmFgev+DidMpgdSB7x+3fGhaVy19JcmrU/6eMVmhmPHyp9M1HK7rcmZTexNiprX5fUWG6VJG1et1njU+MamxrLvEDvfWCv5hbmNFOY0ZqFNZopzEhSxWAQLrz7Ht2nkYkRjU2NVWzOiy/U41PjFQNOPG8sLBdqSJxc1ebDrG+4xduopJ4+apWkg2SlsmbVrI1PjVdtKq1UKzcyMaI1vcnrOu/mi/uWHkA2HZpr3U5oipSyz5n44hM+wAvj48ULTnhcLnyE2oC1O3YsunDvv/IqzY2NNdy3J1xQ5sZKn5lxWRvt7xOWzWrOCtvs3bKlbCiqpSksvc+11tzV2o+nWfGFOjwOoS5st2fjRvUPDpZtOmtEvc2wlTrgp8/bdDnnjxzR/MREMSyVa+oLYWRhclLq7dXC8eQbvWF94fhkBZZ0UIvPn3R54n1Pn9NxDWilYxLKliXdbBr2qxadEbAWCkmIOPqgNLAhee6b/0M6djB5/rTti4NBCBhxuAgX+WqhqFITWXyhvu/27E7pM8ekL769tI14gNONZ0qnnJ6ULT2kgyssXk/Yh1Cu3oEkYG08s9QkqU3J/rvC4tHOC7NS75rk73JNoNOPlx73Dvj1aXGZpSTEzc+UypHV3Hjhy5Pnw37Gx/Dko6VjV66mKquJNWyjXI1iOD4h6EqlYxy//mHfi8HrkeIqwgV6y9oti5reYv09/ZopzBR/S9VrTsJycQ1TreKAU6n2a25hrvi7v6d/UQ1JOoBlyfqGW1qlvkM7Nu/QpU++VCMTIw11uA81cLUMk5FV2zgyMaLJ+ez+ELU0va7pXaMTcyeK03qsRwtuYcnycY1lpeMah/H3Dr+3Yrni8yz+cC6Mjy9qNpl78EFJpQtOuHBJyUUjvsiEi1EcftIXmnSNQHBszx65uTm56WmN707e1wvHj8vNzmrdM59ZHD8phL24tqfaBa1SH5tYvO+VQlEj/ZjSF/i8+/GUu0iXq2WJX/O4xqNSmE5vq9w3D9PnSChPvU3EcQf8rGNeri+WJM1PTCQBw2xpbVRfnzQ/r/lDh2QDA8kChdK1b2FysviPw8LkpNY/+9mSpMnvfjdz8NRwfi1570Tnehyq0sNUxDWgWcckNNXG5sbGNH/okCQtmVbuuXI6I2BJpQASB6YQLtIhYuOZ0sRPpNmTi59fs6kUJKQksElJcKm1+SyraTEozEmyZBtzU6UgFQexUGN1yhPL19qE8DXxk8rlmj0p9fZL8/7YxEElq5xhKIgl5fZlmD0hPfStZPshSM7PLA6NWU1wcfA89vDiYFiuP1g6GIX9iVWqUQzH50e3Ja9f2I+NZ5amFWZLQUtKHj/xtGJACiFlfLpUE1LLN/TimpOwrrf861s0MjFSbEoM6w7GJsc0Pp284cvV2swtzGnfo8l+bj1la7EzewiAYd0n507KyUlSUvuyMF9c78jEiKbnpxfNH/cxq9ZsFcvq/5U1PRyTRr7RODIxUjx2YR/C8/HfcfAM2w3bjvcr1MzF5YpruMIxmvXnfDiOcXnGpsaKYWluYa64X1PzUxo9Oqodp+1Ycg5JSage6BkolnFsKvlgPzl3MhnQ9AEtqXELF4KFEyeknh5JyYd6uPjEH9bzExNLglAsDj9ZH/Jx7cnE9dcXL8S9W7bI+vvlpqfVPzhYvGi6+XnNjY0VL0JSckHKqsGIazzi6fOHDi0JM7WEpHhgzHQYzHo8tW+frL9fNjAgNzu7KICkg2W5ZsV0LUilsJPupB2vL+zbwrFjmrr7btnAwJL9nD9yRAuTk7K+viU1Hlk1MSE4udlZuZmZJUEi9AVys7OSmdz0dLHWJwSH9CCZcVnTf8fHLz5O4bwJx3lhampRTVdhfLx47sX/AMTnkJScW25+fumBLRSW9HUKxyFdo7Zw/Lj6Nm9edL7Hxym8h+JQNT0ysuh8CPOkz7OFyUnJrBgUY3P790vOLVq+uMzCQnFaLTonYMWOH1ocLgqz0sPDybRN20qhyhVKtRlzU1L4MAwX8nStUZCudUpvO+4zFYcWV0ia+VxB6l+XBJOwrfRteo7+tDR/WrzNcrf3CWGxkDFfOqhISZPgzAl/zKYXr39+xv9hSZmOPiide3FpehxMyw0jEQe6rGZOaWkNYGj2DMfSFZLl3ULyeq3fLMlJm85Knh8blQozi1+XULMXHsdCzVxhthTGe88oBqQN/Rs0U5jR1PyUJOn7R76v0aOji0JNuICG2qiTcyfVYz3FTYRmoK8d+JpOzJ1Q73yvCq6wZN3j0+OLajjufOTOJYEsrvWKw8qBEwe04Ba0QRsya8UKrrCoObLHehaV5YZ7btD4dKnGLW62qtYxP93/K72u0MyaDpS1CLVIX9r/pUXBRFLxWN3wwxs0uG6wGGhCOXds3qGRiRGdmDtRDD37j+3Xe4ffqy3rtiwqV7xsOEZZ4n5tcc1lWL7gCjo+e7xYe7bgFrShf8OS4zU2NaZ9j+5bFOLSA5oevfEmTd19d3KRCf/BFwqli21v6XNhemQkCVZzc+pZv74YhMIFa+ruu+UKheLFY+H4cfVs3JgsGzUjSqWL7Nzhw8ULcRzuwoVfZlKhoLkDBzKP1dqhoeRbVSeTz5pQ4xFPD2VfOH58UXPM3NjYoqBYmJgofjsrDi1hH8NFvVLwcdPTctPTxeMX17zEoWN89+7iBTuEsvi4lKsFiTszh+a8sL/hd3z8i+WamZGbmSnuQzEITExIhYJs/fpF+zF34MCiGp1i7aWvFerZuFFuZqYYCNJ9htTbqx6/zrhpbW5sLFm3mTQ/v+g4hNcvlu5zFF7v2YMHF52n4bebnS2dy70Z17S0Qplrr2dr1sjNzBQDuq1Zk9nsVqwt8+LyhdfUzc4u+kdmYXKyeIwKjz1WbJ488rd/W6pxrFS+VIAq1mhV2afM/XR1pLFW2XlWrxu+xn+QrX1C0m9o/zdKM8QBYM2m5AI88RM/0SS5UvAJ0n/3rUvCh1RafvNTS4N/vuAPk1qcEArCNtO1YmG9fWsXB5kwrbd/cZNclr61pcfz08nfISCs8c2C8zOSkiYN9fT78NhTem7zU3xt2Qlp7cZSk2B6v9PlC9NOGUy2EUZ5L9aMRQHrkX3S/GyyT6EpMRyfMGhpVnOdVKq9qnQM08c4/O5bmzQbL8yljnOP1DdQOm5hPdF+XXbONj3U1ysnyWRLajCCXkvCSdx0lLahf4O2rNui/cf2F/vypJeP/17Xt67Up8vPHy8Xl2dNaOKVFq23knCxTweIEBw29G/QQO+AJqYn1Gd9Wtu3VlvWbSmGp8n5SfX39C86Jqf0n6LZwuyiIDhTmNHOM3ZqbGpMB44fKAaNoc1DNXVyf84nnqPpQum9saZ3TVL7E32RIITgnWfs1OjRUU3OTWp93/pF5ZxdmC2+Nhv6NxTL1mM9xYAUjkn6uGcJr5nJZGZacAvq9f8Axa9lvI5Q9lDmUDs1/MjwkvWft+k87T+2Xxv6N+gfPneqTrnnwaWFMEs+wHt7kw/s8HeY7C88PRs3Fmsz4ufDb5mpZ8OG4sXH1qyRm5+X9fWV5ovDXfEg9JYukmFaVIaejRtLtRd+3euf8xxJpTCVvriF8vZu2aK5Bx8shr/4Ah3WHWoMejZsKM4fyhK2E4TaivgCG1/4s7Y3cN55i5axtWvlpqcls2S/orKmQ0YIhb1btiT7FY5/b2+x5qJ4/FOvV/x6huNXrAWJXtNFIaK3V/3nnCMpqjWJXteeTZu0dscOTX73u4vKGfY3lLd/+3b1Dw4m80XHMoTv3s2bkxAzMJCEbalYvqzXe9Fr5ddXrt9R8VhOTCzpN1WTGtaf2d/JLPmnoVDIPNeLXyKI9kHSktcvLkOmULuVkZMuGB250zm3s9LudV7AkpLwUAxQWnohzQo3tQjLpZe33qSf18RPMi7aofqwBcepGDyi4FSrSkGq7uV9SJVJ1pPUzs1NZQfW4mvgA+vMiaTsITSu2VjqmJ+uTUsL+x+ve+CU7DBWo11bn6jhdWurzyhVvBinpQNVuXmk5GId5q+2XPrbgo2WI/18HLoqhYKwXDx/OqyF9dUasC68IftboOF4x2WN9z+EznKvSTxvvI44MNV7PLPEoTuEYpOp13q1tm/touMS67VeLbgFOTl94COmwUfrr/VLVlThA38Z1rHkIhRqLELzSHRxk1Tsd1MManGTS3x9yQiTkkrb6u1V76mnFpt/0gEovW9xCAjBMr3fiy7QUVBSX5/kXHF7C1NTyfMZtTeLyp6x/jgIVypvev/DMejZsGHRvi46/hmvYwhYkpaEzWD9c55TDGbhcWa4qFbmYqGWlj0uT62dvpdNuvx5vKdSujdgtUrVi3UDQWdVCAGs873g3G06XksV9ioRgkHBFcrW5tWix3p0zsZztOc3Ko9PI5UPWN2uUo1o2t99eF5nLc1gqFWFC/qieaS6+sQUF60nbHSjWo4fmlJLwOqpNLEZZnaJmY2a2f1m9pZWbacuVWtCCFfZuueNSrhazMlp3s03Fa4kacEt6ODxgzmVqjvVcwzPONrCgqwGtYQD5xoOESs6XEmEqw7RkoBlZr2SPiDpxZIukPRqM7ugFdsCsDyqNZGipJfrG7DqtepbhM+VdL9z7gFJMrPPSLpc0r0t2h6AZXDxpy5udxEAoCu0KmBtkxR///egpOfFM5jZNZKu8X+esP9yrPEho4Gie9pdAEBU1wMr3nnVZmjbOFjOuWslXduu7QMAALRKqzq5PyzpnOjvs/1zAAAAK16rAtZ3JT3NzJ5kZgOSXiXpthZtCwAAoKO0pInQOTdvZv9R0v+S1CvpeufcD1uxLQAAgE7TEQONAgAArCQtG2gUAABgtSJgAQAA5IyABQAAkLO2jYNlZudJ+rCkJ0naJGmDL8+MpLWSCpLmJU34v03StKTTJK2TdFzSg5LO9fM6SUckPdGvY0HSZkmzfr0n/e8FSaf4x/2Spvy6+/20Xv/7USUDpD7BT98m6YSko0oGGBuQ9Jgvyzq/jPPbkaQ1frmCpGOStvpyFfyys36ZOV+WNX75x3z5ev20GUmn+m33+G3N+2kDKt1A0fnjNOfXvU7SmH/+AUnP8Mc53Kxv2m9zzu/7hKT1frlQjvV+3rV+v04oGW5joz/uk34bZ/uyrfHbNv+74Nf/kJJhOzb4sofjVfDzDfhjMO+XVfSazPryDPhtTPnH836+eV/W0/z0GV/uPr+O+ahMJ/20cJx6fBn6/Dbm/fEZiMowreQ8ON0v1+PLFc4x55d/3D/fH71G00pey8P+mJ2UtMWv/5hfbkrJeSr/Gmzw2z3uy9rv55vzyzzBLxPujj7ny2H++VP9c6Fc076s6/z6w3Fa8M+N+7KF/R+I1jeg0nkSzPrfkyqdz+H9G8rzqN+ncH7Jrz8cu3DcFvxzzq+3xx//HiXn2jq/jhn/XPi8Oumnhff1gF8urKfgp0vJ6zLgpzk/X7//He5iftLv+5SS92l43xZUeh9adEzC++oJKr1vJ6N9HPPH5KCkHX49U0rOn3Q5H1VyTmyIXif5bfX64x9em/X+mJ3w2w3Hcspv7xG/nvB6Kjp2P1NyDof3jvnyhHM+HJuj0esUzi35dfb458Jy4bNu3v9s8usN740N/vesX996//i4knN5sy/7lKRBld6Xc36evqjsT1HpM0IqnSthP48qea2f4o/rgkrvwc3+9ZnzyxyW9CMln4nnRuUdUPIe2aTS50x4TX7qyxjOnfA+nVVy3dns533MH6tT/PbCZ2iPSq95r5Jzw/l1TPj5t0TbeaJK171wHI4o+fx9ql9HOAdm/b5+X8nn7Dl+3VN+v8Ln0rySc+cUv+y038aUfz22+DKH98ImJa9xOG4blXxenOaXcypdv8I1WH7ex/18k37atJLPpl6/7vCe2arSey2sr98f+/DZEY7dev/4cb++dSp9Rm1U6Zq6zu/no/44hs/kgkqfF+FmlOG6UvDL9Pn9DsdhQMm56Pw2wv4elfQbzrk7VUHbOrmb2aRKHyYAAADdwjnnKrYCtrOJsG21ZwAAAE0oVJuhnSHnWZL+TNJOSWeq1Hy3oFJTTGiaCM1JM1pcdT7plw1VrWGHw+NQxTippKpvrZLqvlB9/ZiSasDTVGq2C02EoYlPKlWjh+raWGh6KETzhOUWot+hOjsOtRYtH5pKTIuFqu4pJVWkvdE8ofoxvUxoUgjVquH4he2FqvbQjBbKJ5Wal+Iq1NDEklX20IQxkCpXvM6wX8eVHPewrlA12+PLE6qzXfQTmozi/ZVfLjSRpI+Bi36nX4v4sUXliKVfh9DUGM6/0Ey2VqXzNSzXEy0Tngs/Myod8/h1DE1mkypVP29QUpV/QkmVdZgetteXWrdF6w7bDU3QplLTTnzMw2solY5nev/DeXDSl2tASVV52NZctHxo/h5QqWkkrCs+V+PzPgjv/X5li8splT4vwrrCvq7R4vMlrHdWyXvoCX5aeN3j9+gxJc0Yjyk5V8/wj/uVvBahySbsUzgveqPnwnEI0+ImudD0HIvLHzePp/c7/owI74t4ery/oSmvX4uPd63i+RdUOm9Ds1yWrM+i+FjF51SlspT7TJNKr33Yx/RnYTj+4fM4lDlWy7FIn69h3eFcid+H4TUJx+WkFjfRhrLG+xReJ2WUL3wuS6XzKF4+Xa5a9iueFpoF4/WHJrzQ7BiunfFn25RKTcdxF434MyeUP/57TEnTdPi8CU3aBZWaWdOvddZrL5Veh7CN8JkVjk9BS49n3EUilDXrenlCpdctnh6OUXh/hi4rn5T05jLlLO1IG5sIH1DS/woAAKCbLDjnyv3DIam9ASt0FgMAAOgmVQNWO/tgvUvSIZWqGoPQXNJI8stappUJst51x/sWN1+UW49LLZc1vdHjFC9b6zrS23MZP+n5sh6XmxY34ynj72rrqEc9y5XbZrV1pKdVW6bc9PTrVG696cdZ85ZbX1YTanr58C2d0Gy9IOl70XPp8zu9TwWV37escsXPL2RMi9c7X2b5Wl6j+L0olb6FFqaHb/NmlbnSc/H6q5Wh3LmeXnf6syDrmNerltdEWvxZXct7odx5lLX9ap+B4RhV+8yqdO5Xej9We89Uer7cfOXKXG09tR6rrM/Pap8DtR6/es7xSvNnvRfDZ0e6/NXWVW295ZaL39uV1Pv+ebzaDO2swTqqUn+Icm2u9XBa2kYNAEAewjWm1ue71Urbn1apWoPVzia6dOe4ZmV1EG0UJxiwVLX/xnjPNIfPnc5W7rVZaa/ZStufVjlcbYZ2NhHeo/qq+tLTyy1bbr5qP7Wso5x6qjdbpVz1d7l5Gqmyb2Y/alm21vLlsa1q89fzetfTdFDp2NbTpFGrRo5Fo8e+PdXhlZVrGmmkaaLRbde6ntV4Yau32a2TlHu/Lud7Lm+1fubU+7nX6Sq9V8tlhqrDNLSzifBxJV+L7OQPFZobUQ61DQCw+oRc8JhzbnOlGdvZRDijZGyZcrVoWWNVzCgZqyNd7qwxQOLxR7Kk5y/Xth6G5E9vq9p4KumxX6TFy9VzcS63naxxZlxqWljeqfxYW+X2pZZ9jMuXta9Z0ysdl/R2q42lU88YO1nlr7RsPeP2VFp/pb/j9YRzNh53JZ43/VxWWSuVL+v4lytfVtmqSY8xU+lcz3rN02XLmr/a85WOdzm1njuVzsF0WWp538TKvSfS6yo3T9b6Km2v3DLVtllu2+Xeu9XKVW7erPkq7Xu67FmvVaXPjmrLlCtb1r6E5bL2qdL1o9Jz1caGqnZNqfQ5kH4N0p8ztVyvyq0za75yn/fVzodKn9e1vAcrXXOyyph+HC97UsmtiSpqWw1WJWb215JerOQedwUl92C6wzn3u6n5/ljJQKVO0mf80xdJerdzbsHMXuac+2wN23uBpG9nLeO3Ma7kPk0/p+SehC9QMtjaQ0ruKRXuZzWm5B5XT5T080oulrf5eR9RMijaS5UM3jYh6Ra/ztOVBMenq3QvvHAyzEq6Xcn9BG9Wch+qdyj5BsPnJf22pG8que/S5ZI+ouT+Z1udc5f4Y/kcJffK+kdJP/T78kdKBnv9saQPSPp9JYn8JWb2dufcf/X7f4GkDykZ0PXHSgZfPN/v74OSznLObfPz/bFK9447U9I3JO2R9DJJvyPpNZL+jaSXKxnctSDpfl/er0l6sj9em51z7zazT/v1/ZOk1yu5z9ijSgaxLCi5b9dpKt1X8jwl58KkkgEiPyLpl3x5Dku6TNIPJH1c0qskXeinbVTyhYu7ldwb6zFJByRdoGTgyR5Jb1VyDt7tj8sLJD3Xl+edSgbh/KikN0n6e7+tfyNpv6Q7/Gv3Akm/IOlv/P5M+eV+Vckgl2+TtF3S+yT9oT8eH5Z0naRv+XX8ppJzblzJufgn/lj8UMlgec9X6T5h90ra6+f/pKTf88foBkmf8q9luEfYUSXflovv2XmrknNnXtK7nHPfUhX+PPhNSS9Rcl73+e1/ym/76ZKe6Z/bIOnbSt4rP1Nynv+WknvE9Sm579rjfvtnK3mN71Hy7eON/tj8ByXvpX+UdLWSc+M3lbyPPumLda5z7ldSr9m5Ss7RNyh5Tz7fOfdnZnaNpOuccwUz+7Rz7tV+uZNKzuPfUPL+mffbfKrfp9cqeV1PV/LeOkvSs51zf25mvUrO73sk/bpKAy9KpW8ozkvareQz5ReVvNe2S3qhf+5rkn5FyTmxVsm5f7bfzwf99uYkDfv5z1MyzuBWlT5vfuK3eZpfNtwzMkz/DSXn7YBf/zv943slfcGv/6W+DGGg2/BP8gH/Wp2l5B/SfpUGsFzjyxnuMxsGu/2KkvP4TD/PhX5fDij5bAr34lyQ9HHn3B+Y2XX+7+3+2J+u0mDKYcDi/b4cA347R/08p6h0sezV4gE2w/0vH5b0NJXu49kblXlOyXv5WiXn86ckfV3Snyo5Jx5T6Vx8lpLPu34l790t/rjJl3VSpfuMFpRcH071x+5hX5anK7kOPK7S55z8/Hf5sm3z+xTu53pApUFYw30pH/f7vsEfi28oeQ8+ruQz71RJn/XH6GfOuT1m9jdKzsffVHKO3Ozn/6B/jWaVvMffp+Tz6ywln7Hn+/U/IunFzrkdZna7pE/4cr1bybX6u0o+q35Fyet/VMk596zouIz49Vyk5P3wgJLX/XF/nGf8a3S5kvs3Tim5Bof7Vs5J+lf/mn1SyfUgDGR9mpLPoLOVXIPmfBnCfv+cks/kH/jX4ilKXu8fSJJzbq9q0JEBCwAAoJu1s5M7AADAikTAAgAAyBkBCwAAIGcELAAAgJz9Hy4pktPnA5bEAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 720x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize =(10,7))\n",
    "plt.title('Dendrogram')\n",
    "dend = shc.dendrogram(shc.linkage(X, method='ward',metric ='euclidean'))\n",
    "plt.axhline(y=80, color='r', linestyle='--')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "151e635e",
   "metadata": {},
   "source": [
    "### Choosing threshold for dendogram:\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "fb62176f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "scores of pca with StandardScaler\n",
      "Silhoutte score of Hierarchical is  0.24032542358029713\n",
      "davies bouldin score of Hierarchical is  1.4997315733510441\n",
      "scores of pca with MinMaxScaler\n",
      "Silhoutte score of Hierarchical is  0.2968755883540477\n",
      "davies bouldin score of Hierarchical is  1.2729805371912448\n",
      "scores of pca with RobustScaler\n",
      "Silhoutte score of Hierarchical is  0.22962335175833598\n",
      "davies bouldin score of Hierarchical is  1.4951211286341513\n",
      "scores of kernel pca with StandardScaler\n",
      "Silhoutte score of Hierarchical is  0.3166498681331434\n",
      "davies bouldin score of Hierarchical is  1.0365208564056163\n",
      "scores of kernel pca with MinMaxScaler\n",
      "Silhoutte score of Hierarchical is  0.3922168907539009\n",
      "davies bouldin score of Hierarchical is  0.8976287068012444\n",
      "scores of kernel pca with RobustScaler\n",
      "Silhoutte score of Hierarchical is  0.30525520352164903\n",
      "davies bouldin score of Hierarchical is  0.9573764758543659\n",
      "scores of truncated svd with StandardScaler\n",
      "Silhoutte score of Hierarchical is  0.3258956067060118\n",
      "davies bouldin score of kmean is  0.9937258574917316\n",
      "scores of truncated svd with MinMaxScaler\n",
      "Silhoutte score of Hierarchical is  0.3452930948438733\n",
      "davies bouldin score of Hierarchical is  0.9400835893988322\n",
      "scores of truncated svd with RobustScaler\n",
      "Silhoutte score of Hierarchical is  0.34281836763243106\n",
      "davies bouldin score of Hierarchical is  0.9759004846285431\n"
     ]
    }
   ],
   "source": [
    "cluster = AgglomerativeClustering(n_clusters=4, affinity='euclidean', linkage='ward')  \n",
    "df_pca = pca.fit_transform(X)\n",
    "y = cluster.fit_predict(df_pca)\n",
    "labels_hac = cluster.labels_\n",
    "print(\"scores of pca with StandardScaler\")\n",
    "print('Silhoutte score of Hierarchical is ' , silhouette_score(X, labels_hac))\n",
    "print('davies bouldin score of Hierarchical is ' , davies_bouldin_score(X, labels_hac))\n",
    "df_pca = pca.fit_transform(min_x)\n",
    "y = cluster.fit_predict(df_pca)\n",
    "labels_hac = cluster.labels_\n",
    "print(\"scores of pca with MinMaxScaler\")\n",
    "print('Silhoutte score of Hierarchical is ' , silhouette_score(df_pca, labels_hac))\n",
    "print('davies bouldin score of Hierarchical is ' , davies_bouldin_score(df_pca, labels_hac))\n",
    "df_pca = pca.fit_transform(rob_x)\n",
    "y = cluster.fit_predict(df_pca)\n",
    "labels_hac = cluster.labels_\n",
    "print(\"scores of pca with RobustScaler\")\n",
    "print('Silhoutte score of Hierarchical is ' , silhouette_score(df_pca, labels_hac))\n",
    "print('davies bouldin score of Hierarchical is ' , davies_bouldin_score(df_pca, labels_hac))\n",
    "df_pca = kernel_pca.fit_transform(X)\n",
    "y = cluster.fit_predict(df_pca)\n",
    "labels_hac = cluster.labels_\n",
    "print(\"scores of kernel pca with StandardScaler\")\n",
    "print('Silhoutte score of Hierarchical is ' , silhouette_score(df_pca, labels_hac))\n",
    "print('davies bouldin score of Hierarchical is ' , davies_bouldin_score(df_pca, labels_hac))\n",
    "df_pca = kernel_pca.fit_transform(min_x)\n",
    "y_hc = cluster.fit_predict(df_pca)\n",
    "labels_hac = cluster.labels_\n",
    "print(\"scores of kernel pca with MinMaxScaler\")\n",
    "print('Silhoutte score of Hierarchical is ' , silhouette_score(df_pca, labels_hac))\n",
    "print('davies bouldin score of Hierarchical is ' , davies_bouldin_score(df_pca, labels_hac))\n",
    "df_pca = kernel_pca.fit_transform(rob_x)\n",
    "y = cluster.fit_predict(df_pca)\n",
    "labels_hac = cluster.labels_\n",
    "print(\"scores of kernel pca with RobustScaler\")\n",
    "print('Silhoutte score of Hierarchical is ' , silhouette_score(df_pca, labels_hac))\n",
    "print('davies bouldin score of Hierarchical is ' , davies_bouldin_score(df_pca, labels_hac))\n",
    "df_pca = trun_svd.fit_transform(X)\n",
    "y_h = cluster.fit_predict(df_pca)\n",
    "labels_hac = cluster.labels_\n",
    "print(\"scores of truncated svd with StandardScaler\")\n",
    "print('Silhoutte score of Hierarchical is ' , silhouette_score(df_pca, labels_hac))\n",
    "print('davies bouldin score of kmean is ' , davies_bouldin_score(df_pca, labels_hac))\n",
    "df_pcad = trun_svd.fit_transform(min_x)\n",
    "y_h = cluster.fit_predict(df_pcad)\n",
    "labels_hac = cluster.labels_\n",
    "print(\"scores of truncated svd with MinMaxScaler\")\n",
    "print('Silhoutte score of Hierarchical is ' , silhouette_score(df_pcad, labels_hac))\n",
    "print('davies bouldin score of Hierarchical is ' , davies_bouldin_score(df_pcad, labels_hac))\n",
    "df_pca = trun_svd.fit_transform(rob_x)\n",
    "y = cluster.fit_predict(df_pca)\n",
    "labels_hac = cluster.labels_\n",
    "print(\"scores of truncated svd with RobustScaler\")\n",
    "print('Silhoutte score of Hierarchical is ' , silhouette_score(df_pca, labels_hac))\n",
    "print('davies bouldin score of Hierarchical is ' , davies_bouldin_score(df_pca, labels_hac))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "25031cee",
   "metadata": {},
   "source": [
    "The results for both K-Means and HAC shows that kernel pca is better as it enables dealing with more complex data patterns, which would not be visible under linear transformations alone.\n",
    "\n",
    "Order of performance:\n",
    "**Kernel Pca>Truncted SVD> PCA**\n",
    "\n",
    "The results also show that **MinMax Scaler** works the best among other feature scaling techniques\n",
    "\n",
    "\n",
    "The Silhouette score used to study the separation distance between the resulting clusters :kmean with kernel_pca and MinMaxScaler shows best score 0.456.( higher values indicating better clustering)\n",
    "\n",
    "the Davies-Bouldin score is defined as the average similarity measure of each cluster with its most similar cluster, where similarity is the ratio of within-cluster distances to between-cluster distances. Thus, clusters which are farther apart and less dispersed will result in a better score and kmean with kernel_pca and MinmaxScaler shows best score 0.83 than HAC .( lower values indicating better clustering)\n",
    "\n",
    "As we are selecting data points randomly for the dataframe. The numbers may slightly vary but in overall K-means with kernel PCA and Minmax scaling performed the best. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "d6f5a2c0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "min_x = np.array(min_x)\n",
    "plt.scatter(min_x[y_hc==0, 0], min_x[y_hc==0, 1], s=20, c='red', label ='Cluster 1')\n",
    "plt.scatter(min_x[y_hc==1, 0], min_x[y_hc==1, 1], s=20, c='yellow', label ='Cluster 2')\n",
    "plt.scatter(min_x[y_hc==2, 0], min_x[y_hc==2, 1], s=20, c='green', label ='Cluster 3')\n",
    "plt.scatter(min_x[y_hc==3, 0], min_x[y_hc==3, 1], s=20, c='cyan', label ='Cluster 4')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "825c746b",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}