diff --git a/credit_card_customer_data.xlsx b/credit_card_customer_data.xlsx
new file mode 100644
index 0000000..5b51f8c
Binary files /dev/null and b/credit_card_customer_data.xlsx differ
diff --git a/credit_card_customer_segmentation.ipynb b/credit_card_customer_segmentation.ipynb
new file mode 100644
index 0000000..524f5a0
--- /dev/null
+++ b/credit_card_customer_segmentation.ipynb
@@ -0,0 +1,2227 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<hr>\n",
+    "<h3>ALLLife Bank Credit Card Customer Segmentation</h3>\n",
+    "<hr>\n",
+    "\n",
+    "<b>Background</b>: AllLife Bank wants to focus on its credit card customer base in the next financial year. They have been advised by their marketing research team, that the penetration in the market can be improved. Based on this input, the Marketing team proposes to run personalised campaigns to target new customers as well as upsell to existing customers. Another insight from the market research was that the customers perceive the support services of the back poorly. Based on this, the Operations team wants to upgrade the service delivery model, to ensure that customers queries are resolved faster. Head of Marketing and Head of Delivery both decide to reach out to the Data Science team for help.\n",
+    "\n",
+    "<b>Data Description</b>: Data is of various customers of a bank with their credit limit, the total number of credit cards the customer has, and different channels through which customer has contacted the bank for any queries, different channels include visiting the bank, online and through a call centre. \n",
+    "\n",
+    "<b>Key Questions:</b> \n",
+    "1. How many different segments of customers are there?\n",
+    "2. How are these segments different from each other?\n",
+    "3. What are your recommendations to the bank on how to better market to and service these customers?"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Importing the Libraries"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import warnings\n",
+    "warnings.filterwarnings('ignore')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "\n",
+    "from sklearn.preprocessing import StandardScaler\n",
+    "\n",
+    "import seaborn as sns \n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "from scipy.spatial.distance import cdist\n",
+    "from sklearn.cluster import KMeans\n",
+    "import numpy as np \n",
+    "\n",
+    "from scipy.spatial.distance import pdist\n",
+    "from scipy.cluster.hierarchy import dendrogram, linkage,cophenet"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#Reading the dataset \n",
+    "\n",
+    "df=pd.read_excel('credit_card_customer_data.xlsx')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "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>Sl_No</th>\n",
+       "      <th>Customer Key</th>\n",
+       "      <th>Avg_Credit_Limit</th>\n",
+       "      <th>Total_Credit_Cards</th>\n",
+       "      <th>Total_visits_bank</th>\n",
+       "      <th>Total_visits_online</th>\n",
+       "      <th>Total_calls_made</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>87073</td>\n",
+       "      <td>100000</td>\n",
+       "      <td>2</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>1</td>\n",
+       "      <td>2</td>\n",
+       "      <td>38414</td>\n",
+       "      <td>50000</td>\n",
+       "      <td>3</td>\n",
+       "      <td>0</td>\n",
+       "      <td>10</td>\n",
+       "      <td>9</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   Sl_No  Customer Key  Avg_Credit_Limit  Total_Credit_Cards  \\\n",
+       "0      1         87073            100000                   2   \n",
+       "1      2         38414             50000                   3   \n",
+       "\n",
+       "   Total_visits_bank  Total_visits_online  Total_calls_made  \n",
+       "0                  1                    1                 0  \n",
+       "1                  0                   10                 9  "
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "#Viewing top 2 rows \n",
+    "\n",
+    "df.head(2)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "The cols : Sl_No and CustomerKey are IDs which can be eliminated as they are unique and will not have any relevant role in forming the clusters so we remove them"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "cols_to_consider=['Avg_Credit_Limit','Total_Credit_Cards','Total_visits_bank','Total_visits_online','Total_calls_made']"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "subset=df[cols_to_consider]  #Selecting only the above columns "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### EDA "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Checking for Missing Values "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Avg_Credit_Limit       0\n",
+       "Total_Credit_Cards     0\n",
+       "Total_visits_bank      0\n",
+       "Total_visits_online    0\n",
+       "Total_calls_made       0\n",
+       "dtype: int64"
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "subset.isna().sum() "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "No missing values were found "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Checking for the statistically summary "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>Avg_Credit_Limit</th>\n",
+       "      <th>Total_Credit_Cards</th>\n",
+       "      <th>Total_visits_bank</th>\n",
+       "      <th>Total_visits_online</th>\n",
+       "      <th>Total_calls_made</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <td>count</td>\n",
+       "      <td>660.000000</td>\n",
+       "      <td>660.000000</td>\n",
+       "      <td>660.000000</td>\n",
+       "      <td>660.000000</td>\n",
+       "      <td>660.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>mean</td>\n",
+       "      <td>34574.242424</td>\n",
+       "      <td>4.706061</td>\n",
+       "      <td>2.403030</td>\n",
+       "      <td>2.606061</td>\n",
+       "      <td>3.583333</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>std</td>\n",
+       "      <td>37625.487804</td>\n",
+       "      <td>2.167835</td>\n",
+       "      <td>1.631813</td>\n",
+       "      <td>2.935724</td>\n",
+       "      <td>2.865317</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>min</td>\n",
+       "      <td>3000.000000</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>25%</td>\n",
+       "      <td>10000.000000</td>\n",
+       "      <td>3.000000</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>1.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>50%</td>\n",
+       "      <td>18000.000000</td>\n",
+       "      <td>5.000000</td>\n",
+       "      <td>2.000000</td>\n",
+       "      <td>2.000000</td>\n",
+       "      <td>3.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>75%</td>\n",
+       "      <td>48000.000000</td>\n",
+       "      <td>6.000000</td>\n",
+       "      <td>4.000000</td>\n",
+       "      <td>4.000000</td>\n",
+       "      <td>5.000000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>max</td>\n",
+       "      <td>200000.000000</td>\n",
+       "      <td>10.000000</td>\n",
+       "      <td>5.000000</td>\n",
+       "      <td>15.000000</td>\n",
+       "      <td>10.000000</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "       Avg_Credit_Limit  Total_Credit_Cards  Total_visits_bank  \\\n",
+       "count        660.000000          660.000000         660.000000   \n",
+       "mean       34574.242424            4.706061           2.403030   \n",
+       "std        37625.487804            2.167835           1.631813   \n",
+       "min         3000.000000            1.000000           0.000000   \n",
+       "25%        10000.000000            3.000000           1.000000   \n",
+       "50%        18000.000000            5.000000           2.000000   \n",
+       "75%        48000.000000            6.000000           4.000000   \n",
+       "max       200000.000000           10.000000           5.000000   \n",
+       "\n",
+       "       Total_visits_online  Total_calls_made  \n",
+       "count           660.000000        660.000000  \n",
+       "mean              2.606061          3.583333  \n",
+       "std               2.935724          2.865317  \n",
+       "min               0.000000          0.000000  \n",
+       "25%               1.000000          1.000000  \n",
+       "50%               2.000000          3.000000  \n",
+       "75%               4.000000          5.000000  \n",
+       "max              15.000000         10.000000  "
+      ]
+     },
+     "execution_count": 8,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "subset.describe()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "The min and max value of 'Avg_Credit_Limit' is very larger as compared to the other columns \n",
+    "To bring the data to the same scale let's standardize the data.\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Standard Scaler ( Z Score )"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "scaler=StandardScaler()\n",
+    "subset_scaled=scaler.fit_transform(subset)   "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "subset_scaled_df=pd.DataFrame(subset_scaled,columns=subset.columns)   #Creating a dataframe of the above results"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Data Visualizations "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### HeatMap :  To check for correlated features "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 2 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "sns.heatmap(subset_scaled_df.corr(),annot=True)\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "There is no significant correlation observed among the different features "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### PairPlots"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<seaborn.axisgrid.PairGrid at 0x1a1a00b0d0>"
+      ]
+     },
+     "execution_count": 12,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 900x900 with 30 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "sns.pairplot(subset_scaled_df,diag_kind=\"kde\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Looking at the diagonal of the pairplot we can assume the data to be a mixture of gaussians , looing at the peaks of the gaussians we can say that the optimal number might come between 2-3 , but to be sure , let's make an elbow  plot\n",
+    "\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Elbow Plot"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "1 2.006922226250361\n",
+      "2 1.4571553548514269\n",
+      "3 1.1466276549150365\n",
+      "4 1.0463825294774463\n",
+      "5 0.9908683849620168\n",
+      "6 0.9429753673519395\n",
+      "7 0.9229356170857781\n",
+      "8 0.8893571274402587\n",
+      "9 0.8664426937686484\n"
+     ]
+    },
+    {
+     "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": [
+    "\n",
+    "clusters=range(1,10)\n",
+    "meanDistortions=[]\n",
+    "\n",
+    "for k in clusters:\n",
+    "    model=KMeans(n_clusters=k)\n",
+    "    model.fit(subset_scaled_df)\n",
+    "    prediction=model.predict(subset_scaled_df)\n",
+    "    distortion=sum(np.min(cdist(subset_scaled_df, model.cluster_centers_, 'euclidean'), axis=1)) / subset_scaled_df.shape[0]\n",
+    "                           \n",
+    "    meanDistortions.append(distortion)\n",
+    "\n",
+    "    print(k,distortion)\n",
+    "plt.plot(clusters, meanDistortions, 'bx-')\n",
+    "plt.xlabel('k')\n",
+    "plt.ylabel('Average distortion')\n",
+    "plt.title('Selecting k with the Elbow Method')\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Looking at the elbow method we can take K=3 , so let's apply Kmeans with k=3 \n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "KMeans(algorithm='auto', copy_x=True, init='k-means++', max_iter=300,\n",
+       "       n_clusters=3, n_init=15, n_jobs=None, precompute_distances='auto',\n",
+       "       random_state=2345, tol=0.0001, verbose=0)"
+      ]
+     },
+     "execution_count": 14,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "\n",
+    "kmeans = KMeans(n_clusters=3, n_init = 15, random_state=2345)\n",
+    "kmeans.fit(subset_scaled_df)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "array([[-0.59579625, -1.05962278, -0.9015185 ,  0.32299678,  1.14810882],\n",
+       "       [-0.02106178,  0.37368962,  0.6663945 , -0.55367163, -0.55300488],\n",
+       "       [ 2.83176409,  1.86222621, -1.10576269,  2.82731942, -0.87432983]])"
+      ]
+     },
+     "execution_count": 15,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "centroids = kmeans.cluster_centers_\n",
+    "centroids"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "centroid_df = pd.DataFrame(centroids, columns = subset_scaled_df.columns )"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>Avg_Credit_Limit</th>\n",
+       "      <th>Total_Credit_Cards</th>\n",
+       "      <th>Total_visits_bank</th>\n",
+       "      <th>Total_visits_online</th>\n",
+       "      <th>Total_calls_made</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <td>0</td>\n",
+       "      <td>-0.595796</td>\n",
+       "      <td>-1.059623</td>\n",
+       "      <td>-0.901518</td>\n",
+       "      <td>0.322997</td>\n",
+       "      <td>1.148109</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>1</td>\n",
+       "      <td>-0.021062</td>\n",
+       "      <td>0.373690</td>\n",
+       "      <td>0.666395</td>\n",
+       "      <td>-0.553672</td>\n",
+       "      <td>-0.553005</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>2</td>\n",
+       "      <td>2.831764</td>\n",
+       "      <td>1.862226</td>\n",
+       "      <td>-1.105763</td>\n",
+       "      <td>2.827319</td>\n",
+       "      <td>-0.874330</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   Avg_Credit_Limit  Total_Credit_Cards  Total_visits_bank  \\\n",
+       "0         -0.595796           -1.059623          -0.901518   \n",
+       "1         -0.021062            0.373690           0.666395   \n",
+       "2          2.831764            1.862226          -1.105763   \n",
+       "\n",
+       "   Total_visits_online  Total_calls_made  \n",
+       "0             0.322997          1.148109  \n",
+       "1            -0.553672         -0.553005  \n",
+       "2             2.827319         -0.874330  "
+      ]
+     },
+     "execution_count": 17,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "centroid_df"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "The above are the centroids for the different clusters "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Adding Label to the dataset"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "dataset=subset_scaled_df[:]  #creating a copy of the data "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 19,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "dataset['KmeansLabel']=kmeans.labels_"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>Avg_Credit_Limit</th>\n",
+       "      <th>Total_Credit_Cards</th>\n",
+       "      <th>Total_visits_bank</th>\n",
+       "      <th>Total_visits_online</th>\n",
+       "      <th>Total_calls_made</th>\n",
+       "      <th>KmeansLabel</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <td>0</td>\n",
+       "      <td>1.740187</td>\n",
+       "      <td>-1.249225</td>\n",
+       "      <td>-0.860451</td>\n",
+       "      <td>-0.547490</td>\n",
+       "      <td>-1.251537</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>1</td>\n",
+       "      <td>0.410293</td>\n",
+       "      <td>-0.787585</td>\n",
+       "      <td>-1.473731</td>\n",
+       "      <td>2.520519</td>\n",
+       "      <td>1.891859</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>2</td>\n",
+       "      <td>0.410293</td>\n",
+       "      <td>1.058973</td>\n",
+       "      <td>-0.860451</td>\n",
+       "      <td>0.134290</td>\n",
+       "      <td>0.145528</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>3</td>\n",
+       "      <td>-0.121665</td>\n",
+       "      <td>0.135694</td>\n",
+       "      <td>-0.860451</td>\n",
+       "      <td>-0.547490</td>\n",
+       "      <td>0.145528</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>4</td>\n",
+       "      <td>1.740187</td>\n",
+       "      <td>0.597334</td>\n",
+       "      <td>-1.473731</td>\n",
+       "      <td>3.202298</td>\n",
+       "      <td>-0.203739</td>\n",
+       "      <td>2</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>5</td>\n",
+       "      <td>-0.387644</td>\n",
+       "      <td>-0.787585</td>\n",
+       "      <td>-1.473731</td>\n",
+       "      <td>-0.547490</td>\n",
+       "      <td>1.542593</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>6</td>\n",
+       "      <td>1.740187</td>\n",
+       "      <td>0.135694</td>\n",
+       "      <td>-1.473731</td>\n",
+       "      <td>2.861408</td>\n",
+       "      <td>-0.553005</td>\n",
+       "      <td>2</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>7</td>\n",
+       "      <td>-0.520633</td>\n",
+       "      <td>-0.787585</td>\n",
+       "      <td>-1.473731</td>\n",
+       "      <td>-0.547490</td>\n",
+       "      <td>-0.902271</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>8</td>\n",
+       "      <td>-0.786612</td>\n",
+       "      <td>-1.249225</td>\n",
+       "      <td>-1.473731</td>\n",
+       "      <td>-0.206600</td>\n",
+       "      <td>-0.553005</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>9</td>\n",
+       "      <td>-0.839808</td>\n",
+       "      <td>-0.325946</td>\n",
+       "      <td>-1.473731</td>\n",
+       "      <td>-0.547490</td>\n",
+       "      <td>1.193326</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   Avg_Credit_Limit  Total_Credit_Cards  Total_visits_bank  \\\n",
+       "0          1.740187           -1.249225          -0.860451   \n",
+       "1          0.410293           -0.787585          -1.473731   \n",
+       "2          0.410293            1.058973          -0.860451   \n",
+       "3         -0.121665            0.135694          -0.860451   \n",
+       "4          1.740187            0.597334          -1.473731   \n",
+       "5         -0.387644           -0.787585          -1.473731   \n",
+       "6          1.740187            0.135694          -1.473731   \n",
+       "7         -0.520633           -0.787585          -1.473731   \n",
+       "8         -0.786612           -1.249225          -1.473731   \n",
+       "9         -0.839808           -0.325946          -1.473731   \n",
+       "\n",
+       "   Total_visits_online  Total_calls_made  KmeansLabel  \n",
+       "0            -0.547490         -1.251537            1  \n",
+       "1             2.520519          1.891859            0  \n",
+       "2             0.134290          0.145528            1  \n",
+       "3            -0.547490          0.145528            1  \n",
+       "4             3.202298         -0.203739            2  \n",
+       "5            -0.547490          1.542593            0  \n",
+       "6             2.861408         -0.553005            2  \n",
+       "7            -0.547490         -0.902271            0  \n",
+       "8            -0.206600         -0.553005            0  \n",
+       "9            -0.547490          1.193326            0  "
+      ]
+     },
+     "execution_count": 20,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "dataset.head(10)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Visualizing the clusters"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Since the number of dimensions is 5 in the dataset , it is not possible to create 5-d Visualization so we can take any 2 random features and make a scatter plot observing the different  clusters "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXIAAAD4CAYAAADxeG0DAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nO3deZwcdZ3/8denuvqanvvIfd+BBAKEcARQDjHICquACou34o9dd0UXr1VX0d9v19VVF3fVhVVEVhRQEVBQRDlFxIQjIQeBkINck5nMPdN31ff3R3WOyfTcPd1Tk8/z8ZgHk66u7/dT3dXv7qku6iPGGJRSSvmXVeoClFJKjY4GuVJK+ZwGuVJK+ZwGuVJK+ZwGuVJK+Zxdiknr6+vNnDlzSjG1Ukr51nPPPXfQGNNw7O0lCfI5c+awbt26UkytlFK+JSK78t2uh1aUUsrnNMiVUsrnNMiVUsrnNMiVUsrnNMiVUsrnSnLWilJDYZwWTOLnkN0OwVOQ6GWIVTbMMRox8Z9B5jlw42CAQA2EzgCTBec1CC5BolcgVuUQxjuQq+l1JLQKopciEsG4XZjEvZDZDIF5ICHIvgz2PCR6JRKoG9ljkNmI6fkZZF8FKwzB05Gyq5BAnzPQRs2YLKT+gEk9DlID9mzIvARS5j0+wcUFnzN/HSlIPohJ/QUC03PbO+XIcqcJ03M3pP4AbjME6iB2HVb0r7zlmZcw8fuADBK5BEJnIiJjW7Oz39vP3H1I6EyIvAWR0JHl6ecxiV8BBoleioROL+j8UoqrH65cudLo6YdqICazGdN6LZgMkAKiYFUjdfcOORRNeh2m7UNgkoA7wD0jIFGk7meIPWuA8V7AtL3fewMgDVIGVj1U/ye0vi83T+KYtcIgNlJ7JxI8YUh1H+J23wLd/4W3/YdYXq21dyDB5cMabyDGpDGt74PsZjBxQPDe9XJzEoKKT2PF/qZgc+atw+3CtFwJ7oFcHSGQAFLzAyS0EpN+EdP6HiDZd+XQBRA6Cbr/G0gDLkjUC9XKfxmzMDepP2PaPgJkgUxuv5iG1N2DWOW4nV+D+J1H1RyBsndgVX5u2HOJyHPGmJXH3q6HVtS4ZDo+DaabIyGWALcZ0/2toa1vDKb9k7kwGCjEAZJgOjCdNw08Xseh8dK5G+PgNELbR8F00DfE8eo3Pd72DINx9ucJcQDXG699eOMNKvEAZDfltg+OhHhuTpLQ9VWM21bYeY9hem4FZ+9RdaTBJDDtN+K6LqbjRvKGOED6Uej+Tm557jk3CUg8BJnnx6Zec6imBJDJ3RgHZzem54eY7DaI/zi33OR+EhC/G5PZUrA6NMjVuGPcDsi+lmdJFpKPDG0QZy+4LcOY1YX0nwZYfACcA3kWpMHdzaBvFtnXvO0aqtSTDPjydHZh3NahjzcIk/yVF3oDkSCkninYnHklH+LwG+XR3FbIvNjPc9DrjvkGxSR/X4Di8nB2gtuVZ4F3eIjUE3if1I+VgdTjBStDg1yNQwN8dSPhoQ0hIQb/JH6s4AjHG+qf7MP4SkrCQxh3gHqHS6JDvF+kcHPm1d/z64JVTu+/FPLJ95gFhr59wyVh+t0vJAKEvPnz1jTEfXkINMjVuCNWDEJn0Tf4whC9amhjBCZBcClD38VDEH3rADXVQvAk+r4oIxBc4a3fLxtCZ3nbNVThC+k/tAIQWolYFUMfbxBS9q6hhV14dcHmzKvsGuDYNwsL7CVYwUUQXM7Ab3D5Q1MGeG5HQwLTwZ5H3/0sCtFrILKmvzUhcknB6tAgV+OSVP0bBGaCxIAoEIHQKqT8I0Mfo/o/wJpC/pAVvN0/6n05ZS9GKj47yHjfhMDU3jWFz4XqWyF4Qu62aG5c8ZZLDAIzve0ZBrEqkJpve2P0CqcQBGYgVV8f1niDCr0BolfjfTlbxpHHLLcNEkNqbkEK+CkyHym7GsLne/MS9ea2piA1N3vLq78Fgenkja7yG5Hqbx9Zj9x2VH4eseeOXc3V/wnWpNzjVAaEIfImpOxK7+yiqq95t+UeRwhD5f9DAlMLV4OetaLGK2NcSK8FZw8Elw77rA9vDAfSf8ZkXwU3AeKAVQf2coSMdyzeng/BFUM6q8Gr6RnvS87gciS4KHe7gcwG7zRBex6GMJJ9GQIzIHQ6IiP7zGTcbkg9iXF2eacB2osgdMaIxxt0PmcvpJ4FqwpjL0Eyf/HCKXweMlaHJ/LVkd0G6fUQmJI7ffDIm9mh58CkXgDnFe/NNXYdVu5spkOPGWQgfK7319RY12sc7zsWpwlCKxB7fu/lbmeuJuM9llbViObp76wVDXKllPIJPf1QKaUmKA1ypZTyOQ1ypZTyOQ1ypZTyOQ1ypZTyOQ1ypZTyOQ1ypZTyOQ1ypZTyOQ1ypZTyOQ1ypZTyOQ1ypZTyOQ1ypZTyOQ1ypZTyuWG0LMlPRGYCdwBT8Fpl3GqMuXm04ypl0usxPbdAdgeETkViHxmwOXKvdd0OTM8PvdZwVjUSey+E3wRkMPGfQOI+IADRq7wu7ZKvIcH4Ztw2TM9tkPi1199UyiByERK7DglMLm1tqScx3bdAdhsYA/YcpPxDEH7ToJcLNs5+TNctXg9OkwCrAiKXIbH397n8q8lsPDJP8CSk/P8M+drjxu3C9NwOqYdBKpCyd0PkkjFr0jyWRn0ZWxGZCkw1xjwvIhXAc8BfG2M297eOXsZWDcYkH8W034DXfNjgtcaK5DrdLxh4Xbcb0/JWcJo50v8xCmXvg8xfILOJIw18oxBejVXz3THakrFh3A7MwbeC2ww4Ry2xQKqQ+vuRwJSS1Ob23A5d36Rvk+QIxD6AVXFDv+saZy/m4OVguujdIckGazJS/yvEKvfum3oa03Y9vfeRMFL7UyS4dMAajRvHtPw1OPs53OBaohB9F1blwA1GSmnMLmNrjNlvjHk+93sXsAWYPtpx1fHLGIPp/CJeEBx6MTtg4piufx98/fjd4LTQu4lvAuL/A5nN9A6YBKSfxmQ2FKr8ojDx//UaEvcKcQAXTBem+79LURbGjfcT4ni39fxgwKbRpvu/vL8u+rS5y4LbgonfdeS+/e4jXx28zsQvveYgh0LcuxHid2KcxkHXH28KeoxcROYApwDP5ll2nYisE5F1zc3NhZxWTTSmDdy2fAsg/dzg66efJH+QCJCnU7zJQtpnfyGmniJvt3kAspB+upjVHDX1KyADNc8Oep2U+pN6hv6bXCdzXXZybxjO3vx3y7w4eJ3pp8i7j0gIMusHX3+cKViQi0g58AvgBmNM57HLjTG3GmNWGmNWNjQ0FGpaNRHJAE2KrZrB1w9Mof9dO0//TgmB5bN90hrksEmptseqA5Ppf7lxB64tUD/A4OK1dYNcB/pgP3erHqxKsKaSv1GzC9ZANYxPBQlyEQnihfidxph7CzGmOn6JhCF6GXBso98oxD40+Ppl76FvYAfAmtnPp0UbIheNrNgSkdj76dtt/pAoEvtwMcs5TOyZEFxG/pC0IDAN7P57r3p197ddYST2Hu9+EoCyq/LcNwqxDw5eZ9k19H0jsLwQD5466PrjzaiDXLyveH8AbDHGfHP0JSkFUvnPEL4ACIGUA2GIvReJXjX4usEToepfQSpyn+4jYJ+A1N2O1Pwg1/G8DIh6He5r7yhqY+FCkNAKqPwyEOPIy9gCIlDxcSRyfulqq/lOLgyPftMMeM9B7W0DnhUikTdD+d/jBbR1ZF0ph6p/9Z7bQ/et+DRELqbXPlJ2tXf2yWA1BhdC1TdAqo7aRxYjtT86bs9aOQd4CniJIwe3/skY81B/6+hZK2qojHMQ3EYIzDl8tsKQ1zWZ3DHbSu+T4uHbXa/bvQQgMN+XL9xDjElD9lWMm0asINjzx82bknH2YZwmAMSq6/UcDLquGwdnO8ZkEAmBvdD7b977tnrHywOzEatyeDWaTG5fKB/yqa2l1N9ZK6MO8pHQIFdKqeEbs9MPlVJKlZYGuVJK+ZwGuVJK+ZwGuVJK+ZwGuVJK+ZwGuVJK+ZwGuVJK+ZwGuVJK+ZwGuVJK+ZwGuVJK+ZwGuVJK+ZwGuVJK+ZwGuVJK+ZwGufI9Y9K4Xd/APXA6buNy3NYPYrLbCzS2i9t9C27T2biNy3BbrsFkNh1Z7rTgtt+Ie+Bk3AMrcDs+i3E7RjSXG/8VbvNFuI0n4jZfgkk+Nqra3cSDx4z36KDrGGNwe27HbTont71XYdIvDLxOei3uwbd78zSdhXvgQtzGJbiNi3Ebl+Yes/y92I3bidvxOdwDp3iPYfsnvEsX+4gxLm7PbbhNq3OP2Tsw6eK2i9PL2Crfc9v+NtfD8lAjXfGuL13/WyQwupZnbsdNkLiXXr0+JYrU/RICMzAH1+Sa+GZzC4MQmIXU/9rrYjPUeeK/gM6b6N1HMoLUfBsJv3H4dcfvh84v9B2v+j+QyAX9r9f1Tej5Eb17m0aQuruQYN/OPiazAdNyLfl7pB5FypC6+xB7zpF1jYtpuRyy24FD7eFssBqQht95naJ8wO38GsTvpPdjFkXq7kaCSwo6l17GVk1IJvv6MSEOYMCkMPE7Rze22w6Jn9OnYbNJY7pvgeQjuSbR2aMWZrxGGOmnhj6PMdD9DfqGYRLT9Y2RFd/99X7G+3r/dbhx6Lmdvg2qU5ju7+Rfp+vmPPPku2MK03Nr79vSfwJnD0dCHCALpgOSDw8+5jhg3G6I/y/DeczGgga58rfsq17z5D7SA3drHwrnda/re98FkNmIyb4CpqfvYpOCzCvDmCgNbmv+RdmdwxgnN73Jgtucf6Hzev8ruvu8rkl9R4R+Do2Q3TrEqrzHrPe6r4BJ55kujslsGeK4Jebs66cPrAvZ4m2DBrnyN3tOP13bg2AvHt3YgRn5gwYLggsRe16u9+cxJAz23GFMFALpp0VZYPowxslNLzZYtf2MN7X/Fa0pYLL5l9nz+7l93hCrssBedEwtc/t5Ey5D7AVDHLfEAlP72f8EirgNGuTK18SeD6FTgWOOp0rocMf1EY9t1ULkUvp2ag8jsf8DkTfnGvce/TIKgFUNwziuLSJQ/lHg2F6bEaTi4yMpHWJ/n3c8yvsfT6zyfjrTR5Dyj+Zf53Cj5EFIGIl9pPdt4fO8rvW9mjRbYEUhcsngY44DYlVA9Ary7iPlf1u0OjTIle9J9XchejkQAgSCJyG1dyIDffoc6thV/xfKrsl98hYILERq/wcJLkEkgtT9HELnAAHvJ/xGpPYeJO8hmQHmKXs3VNwIUuPNY02Gyq94XeVHUnfZ1VDxabDqcuNNgsqbsKKXDrxexecg9oFcV3qBwFyk5rtIaEX++4dOR2q+DYHZ3v0lCtT0vlNgIVLzfa9z/dHrSgCpvQvC5+OFuQWhs5HanyFWnr90ximp/ALE3pd7UxcIzENqvocETypeDXrWipoovH3Z8Q4tFHlsYxxAEBn9ZyOvc/zw3ggKPZ63vdlhrWdMBrAREYzJYoyFiDuk58MYFzDDOtNnvBnL/e+Q/s5aGbsZlSoyEWGsdunBxi5kABUyxEc6nre9w/yr4qh5RGxEYKh/9BfiDbDUxnL/G4z/Hz2llDrOaZArpZTPaZArpZTPaZArpZTPaZArpZTPaZArpZTPaZArpZTPaZArpZTPaZArpZTPaZArpZTPaZArpZTPaZArpZTP+faiWclslrtf2kB1JEJDLIYxhqaeHra2NDO7ppaaSITKcIRJZTFSTpZXWlp49eBBUq7DeTNnsmTSFCaXlwOwq72NA909rJgyheZ4nNc72pgcq6AqEqGubODLaWYch31dXYhA1nE40NNNdSRK2LY50NVNRThMZSTM1PIKgoEjF1ba1d7G717bRk00SkUwxJKGSTz9+k7akgmuO2Ulm1oOsqejg4sXLCSRydCRSjK9opJ4JkNLPI6Id6EhYwwiELFtJsXK+ePrO2jq7sYAK6fNYHZ1TZ+a93R2sLujnUmxckIBm0w2y5aWZmKhIHOr6wCDIBhgUixGNOhdDKklHqc9maCpp5vqSBmhQIBNTQeYFIsxtaICAyQzGVqTSU6aPIXykNc0YGPTAVLZLPVlMRpiMcqChb0olN8Zt9vrEBSYguRttOB/R7Zxat6LeBmnCXCRwJTCz+20AEmwpuUubDVQfbnmGu5BDMa7VHFgcsFrKjRfXsb2int+wguN+0ddx8yKSuLZDC2JY/vteQQ4fdoMbl5z6eHQP9qdG17k355+ing2gzvI41geDPK5c8/n8iVLWfk/3yWeyddVJD8BwoEArgHHuLjGMNRnrSoc5vH3fJCqaJTmnm4uv+tOGnu6hzxvKBDgyqXLeLX1IGv37R3yvACnTJ7KpoNNpB3n8G22WLzn5BV89pw3ELCO7z8IjUljOr8EiQeAgNdmrfwTWLFrS11awRiTwnR8AZIP4W2jDRWfwip7p7c8uxPTfgNkt+Fdy3s6Uv3NvI2ehz23cwDT/gnIrMdrWFGLVH8NCa06qr5Dz8GvvPlxADf3Y7z1AvORmpvHRdei/i5j67sg/8Jjj3DnS6PsxTgMAsytruGRd7+/17v5I69t44aHHySR7ac1Vh5R2yYgQvcwQrwQaiIRnrvu71j1/e9xMB4f9voCwwrwwYQtiw+ddjr/eNY5BRzVf9yOz+dC/OjmxVGk+utI5OJSlVVQbvsnIflbejfHjiI1N0PobEzzG3L9So/aw6QcaXgMsapGPK8xBnNwTa5HqXPUkihS/yBiz/Dqy/sc5CFVSMMTJW940V+Q++4j0V0bXyrqfAZo7Olm3f69vW7/zrpnhxXiAIlstughDtCWTPLsnt0jCnEobIgDpFyX2198ftC/YiYy48YhcT99AySB6f5eKUoqOON2QfI39A5xOLyNqT+ASdJnDzPZ3CfkUcg8B+4Beoc4QBaTuMubxiT6eQ7yMBlI/XZ0NY2hggS5iNwmIk0isnHwe4+OU4IXvwCN3b0PRzR2dRW9jtH4854BuqeXQCKbJeMc+yI7jpgOvD0rD7exqKWMGbe1nw7zgNMIzv5+mlsnMc7ePLcPg9NI/sc3A9nca8Ed4DnIUxPOgdHVNIYK9Yn8dmBNgcYaUJld/O9ns67L8km9v/A4deq0Ie8Chwz3/oX0jhOL1z9wKKZVVBAuwXM5bliTQMJ5FggETyl6OWMiMJX8EWN5DbODJ+cPeilDQqN8DILLvU/2fUQhdEaujIZ+noM8JAJF7ME5XAUJcmPMk0BrIcYazL+/qbjdtUNWgDULFjHnmLM/PnHW6sNncwxVWTDEGdNmFLK8ITl35iymVlRw3qw5I1o/IILdz7f9IxGyLL543gUFG8+PRAJQ8Rl6d1+3QKLIAJ3u/UQkBOU3AtGjbj20jX8PwVMhuILej0HYa+QcHt3+IfZsiKw5Zu6g94Vn9K9z9R16DqL5hui9nr0YQmeNqqaxVLAvO0VkDvBrY8yyfpZfB1wHMGvWrNN27do14rl+tmkjn3/0ETLGHfEYIRGuX3kGB5MJfr55IxnXpSIUIpHJkHZdAiJMjpVz3Wmn8zfLT857hsW21ha+9ec/8fTuXSQzGbKuy7EVHTrj5MK58/n4WauZV1PLxx9+iPu3bhlSnVHbZlIsRirrMLemhrZEgtc7Osjmtt0YgyVCVThM2nFpTx053ifAlSecyL9ddOSPpf/35OP8+KX1pJwsFke+pz96nYAIlmURsgIsmzSJG85czastB/neur9woKc77+GtY2O+IRbjk2edwy+2bGbtvj04xhC0LE6aNJkbzz6XM2bMHNL2T3Qm9QSm+7vg7IPgCqTiY+Pi7IhCMslHMT3f8w53BE9DKv4Bsed5y0wa0/NDSPwcjAPRy5DYhxErNvp5jYOJ3wXxH4NJQORipPx6xOr9oazXcxCY5d3X2eHVY1VB2TuR2IcQifQzU/GM+VkrgwX50UZ7+qFSSh2PJsxZK0oppXrTIFdKKZ8r1OmHPwWeARaLyB4R+WAhxlVKKTW4gpz/ZYy5uhDjKKWUGj49tKKUUj6nQa6UUj6nQa6UUj6nQa6UUj6nQa6UUj6nQa6UUj6nQa6UUj6nQa6UUj6nQa6UUj7n2yv7t8TjrN27h0d2bGN3Rwet8Tgpx6EsGCRkB8g6BktgaUMD3Zk0W5oPMqe6imUNU9hysAkDBIC9XV1kjUsmk2FPrvt8NBDg7Jmzed8pp7Fq+gzsPJewPdjTzScfeZi2VIJLFyzmzJmzWNYwCYDnG/exq70dYwwzq6pZXFfPun17ue3Fdezu6GRKeTkzKquojkQ4d9ZswnaQJ3ZuJxIMMqOiEjtgkcw6JNNpdnd1sqC2jjXzF/B8434qQmEW19fz3P59PLpjO/u7u1hQU8v8mjrCdoBUNktDrJyV06azsekAiUyGFVOm8p21z9LY3cUHV5zKtrZWXmk5yLSKStqTCbozGabEyqmKRKgMh1naMAljDK+2tjC3uoayYJANBxrpSCYJBiwssdja0kxnwmvhlXCy2JZFY3cXVeEIq6ZNZ1dnB+3JBDWRKG+YM5fzZs8t3s4xDMakIf0s4EDoDEQGuza1UuOP75ovA3z72T/xX3/5M9ki1F4dCnPH269i2VEdgv7x4Yf4ZZ7ric+rqSWVzdLU00PG9a7ybUGfa5SPVMS2cV1D2h24RdqhOSMBG4MhNcyWagIIQjRok8xmcY0Zdd/OumgZj77n/VSES39N50NM6k+Y9o8edYsDlV/Hik6Mxsdq4pkwl7F9ctdObnluXVFCHKA9neK99/3icH/JXe1teUMcYHtbK3u7Og+HOBQuxAGS2eygIX70nEknO+wQB68VrouhJ5PBKUCIA7Qk4rz3/nsLMFJhGLcT0349mO6jfhLQcSPG2V/q8pQaFt8F+R3rXyCRLW4n+ozj8KfdXsPWj/32waLOPZGsbxxHAZl8pE/zdo+LGW0Hd6WKzHdB3nFUK7NiMRi601637650qujzTxTFP4g3ANNN7yZ3h2TAdBW7GqVGxXdBfsmCRYQDgaLOmXXdwz0mrzv19KLOPZHURsbRF4nh1eTd/SWChN9Q9HKUGg3fBfnVy05idnUNVgG7ug/EFuFjZ5xNfVkZAO9cdhKxYDDvfS0gVOQ3GT/51pq3lLqEw8ReANEroddZKmUQeiMETytVWUqNiC/PWklmM/xyy2ZuX/8CO9tayQxjG4IiZIf4Bd6qadP5x7PP4fRpM/osu+5X9/HojtdwgVgwyFsXLeXDp60kmc3yoxefZ0NTI65rWNowifk1tdy3ZRPbO9oPr28BYdtmXnUttmWxo70VRKgMhQkGLFxj6EqlSDkOk2IxVs+cw/7uLqrCYeZW1/D7Ha/x8sFmMo5D0ApQHysjGAhgi8X0ykpWTpvOa62t9GTSBCyLp3buIO26zKqqIuu4tCTiBCxvnqzjErYDVEUiVEeirJo+A2NgW2sLc2tqiARsnt27m/ZkEhdDxnFoTyRyX6QKCLiui+v9i1gwhGNcsq6LLRbLJ0/hS2+8gCX1DUN+norBGAPpP2ISvwDjINHLIHwhIr77fKOOE/2dteLLIFdKqePRhDn9UCmlVG8a5Eop5XMa5Eop5XMa5Eop5XMa5Eop5XMa5Eop5XMa5Eop5XMa5Eop5XMa5Eop5XMa5Eop5XMa5Eop5XMa5Eop5XMa5Eop5XN2qQsYCWMMa/ft5ZdbNvPwa6+SzGYJB2ymlpfREKugqaeb7kyatOPSmUwQCARYUl/P9IoqBMi6Djvb22mOx8k6DjMqK1lYV8ef9+yhJ5MmFgwSDNgsrKnlH848u1fj5Xya4z08+MpWutNppsRi3LHhRdqTSS6YM48vnPdGbnl+Lfds3kg4YPPZ1eexcvoMvvnMH3l27x46UylmV1XztiUn0JqI88SunbQlE4QCAVZNn8Ga+Qs5efIUNjQd4NEd29nV0UbQsnBcg2sMIdsmkUmzpbmZ1mSCtONQHy3jHScsJxoKkshmWDF5Kq+1tdKRSnL2zFmcOmUaGw408tTru4iFQvzVwsVUhsP8ZturPLlrB6+0HKS+rIy3LTmBxu5uXmltoSwYZMWUqVyyYBFlwSCu63LnxvU8+MpW4pkMVZEILfE4lggrpkzlnFmzSWazrD/QSHsywYKaOi5ZuIh5NbXF2UmGyGRfh9TvwDgQudC7TrlSPuO7y9g6rsv1Dz7AYzu24xSpediHV5zGZ897Y95lj+/cwd8+9ADGDL9b/VDYIkyKldOSSJByssNeX/BarAVEcI0hYgepiUZojcdJOy7BgPdHWcS26U6ncQbYH2wRKsIR7nzbVbz3/p/THI8Pae5DwlaAG846m4+ctmrY2zEW3J47oeureO2qDRCA8o9glX+0xJUpld+EuYztA1tf5smdO4oW4gD/8+JzNPd097k9lc3y97/5FcnsyLrVD0XWGPZ1d40oxOFIkDq5ZhqJbIZ9XV0kHQcX780n5Th0pFIDhvihWtqSCa6972eDhvjRcx+Sch1ufvYZdrS3jWhbCsk4jbkQTwEZIOv93n0rJvNKaYtTaph8F+Q/37KRtHGLPu9tLz7f57Zn9+5BitRybjxpTSRGvK7jujy87dUCVjNCqT/g/c1wrAwm+dtiV6PUqPguyEsl30v++Ivwwhgf730DFTEuClRqyHwX5FedsJxQCXoqfuCUPoelWDW9by/PiU6AumjZiNe3RFgzf1HhChqp8IX0PfgDYCPRS4pdjVKj4rsgv2zxEi6cP59AET81/e3KVdSX9Q2vsG3znUveStS2idr2mFRkizCrsopwIDCi9QO5j7+2ZWGLELVtZlV549mWRcS2iQQC1ESih+/bn6BlURuNctcV72RKrHzQuY8dLWRZfPLsc5ldXT2ibSkkCUyGys8D4dxPyPtv+Uf1zBXlOwU5a0VE1gA3AwHg+8aYrw50/9E2XzbGsP5AI/e/vJlfv7qVRCZLxLaZXl7OlIpKGnu66EymyLourYkEtmVx0uQpTK2oAAxZF3a2tdIU7yHtOMypqWFJXT1P7NpBTyZDRTCEHQiwqLaOj515Novq6gespzUR5zfbXqUrlWJmZRU/XP8crfEEb16wkBvPXM2PXnqRn2xYT8QO8rlz38DJk6fwn2v/zDO7Xz98+uGVJy6jububp17fRUsiTti2WTVtBm9esJBlkz74BLgAAA9+SURBVCazqekAj+7czu6ODoJWgKzrfd0bCtgkshk2Nx2gOR73Tj8sK+Pak1ZgWxaJbIbTpk7n1dYW2pMJVs+czUmTp7C5uYknd+2kPBTiLQsXUR4K8/vt23hq1y5ebmmivizGlUtPZF93F6+0tBALhlg+eTJvnr+AiO2dfviLLZu4/5WXSWYy1ESiNPV0Y4nFqVOncc6s2SSymdzph0nm19RyyYJFzKyqGvHzPhaMsx+SvwOyEL4IsWeXuiSl+tXfWSujDnIRCQCvAG8C9gBrgauNMZv7W2e0Qa6UUsejsTz9cBWwzRiz3RiTBu4CLi/AuEoppYagEEE+Hdh91L/35G7rRUSuE5F1IrKuubm5ANMqpZSCwgR5vm/I+hyvMcbcaoxZaYxZ2dDQUIBplVJKQWGCfA8w86h/zwD2FWBcpZRSQ1CIIF8LLBSRuSISAt4FPFCAcZVSSg3BqK9+aIzJishHgYfxTj+8zRizadSVKaWUGpKCXMbWGPMQ8FAhxlJKKTU8vvs/O5VSSvWmQa6UUj6nQa6UUj6nQa6UUj6nQa6UUj6nQa6UUj5XkNMPi6Gno4cHb/09zz2ynsisatYuFF4OJiGQu0KAm7sqgCVgDCBHLh5gDGRcrzVNQI60qHGN91Z29HW4XXNkvaxheXeE+7/0D4cXN/f0cMf6F3hs13Z60hmmlJdz2tTpbG5u4tk9u0k6WSwgZNs4rkvACjCtooLz58ylK5Vid1cnNeEIz+x5nfZk0psyN7YFXHfq6Vw4fz4/3vAiLYkEb56/kL9auJgHX93KvS9voiWeoCIc4ozpM+lKpXhk+zY6kkkCIlw4bwFfPO98frrpJe7Y8MLh8edWV3PbW9/OY7t2cMf6F9jT2YFrDPNqavnuWy5jQV0df9mzhy8/+Sh7uzqZVVnNovo6Njc3055M0pNO0ZPJHO7pWREKccGceVy3chVL6xvY39XJNff+jF0d7VjACQ2TqAiH2NPRQVsyRTyTBqAqEuGyRUv4wCkrx93lbJXys4Jcj3y4hnsZ246DnVx/2qfoPNhFT8Bl1xdPBTsXyMYcCeKjfx9Mf+sd+zsQ6sjw8j9/lp3tbbzt7jvpSqUYy66hh99/8LrbYwyOMWTc0c1qQd66b1h1Fv/xl2eGPV7EtvnSeRfwmUd/N6z1orbNT654JydPnjLsOZU6no3lZWzH3N1fu4/2Ax2kEmma3jnvSIhD7+AeTjPI/tY79ncR0lVBfnT/Y/zLH5+gc4xDHLwAP/T2msxmSTrOqEMc8oc4MKIQB6+2f3rskWGvl8hm+dyjw19PKZWfLw6tPPPAOjLpLADJBZUl6d5769q1dEyP5O3yeDxzR/gX3csHm0lls4RtX+yCSo1rvvhEHquOHf5d0mP9eTi/ikCQWDBUkrknItuysC1f7H5KjXu+eCW9/WOXEomFAah+Yv/hY9dFkZvrfz/2Aa5dfvKImyCPZ9ERfioWoCFPU+rB2CJcvngpAQ1ypQrCF6+k89+1mrdefzHBSJAZ69qJvtzhBWwxfoA12Toaaqu4/vQzuHj+AoJHBZA1xMM8gtfRvmwIoVkbiRILBikPhQgHAqyZv4CQFeg1l5C/o8e86pq8t1eHw3k/AU+OxXj03R+gKhwe0nYcYomwqK6eX1/9HmZUVA5rvTNnzOKLb7hgWPMppfrni7NWDmk70M4r616jblotyZogP3jsaZ5p3Et3NoO9P04mIrjlQWqyAUSEblxCoQC1Vog58SA7WluJR4RwfRnZRIZwwsU1QltdACdoUdmeRbrSuI5DpjrMrFgF//0P76W2qqJXHXs7O9nc3ER3OkVdWYxTp05ja0szz+/dx4sH9lEdLeOkhkns7OggHAiwfPJkTpo8lY5Uktc7OlhcX8+zu3fz2K4dzKqs4mAizgv793HG9Jl85YKLcI1hfeN+OlIpTp06jcpwmM5Uiuf376M1EacqHGHZpMl0p1NsbDrA77e/RmUkwo1nnUNNNMq+rk5ebGzk7o0v4RiHz5/7RpY0TKItkeD5fXt5bNd2utJp/mb5yayafqQnyBM7d/BC435Wz5zF5PJytrYcpDuVpiOVpLGri+3tbQQDFqtnzubkyVM4oWESkntz2dh0gK/+8Qlqo2V8+fwLeenAAVricYzAozu2I8AFc+dy0uSpzKupHdX+o9Txqr+zVnwV5EopdTzz9emHSiml+qdBrpRSPqdBrpRSPqdBrpRSPqdBrpRSPqdBrpRSPqdBrpRSPqdBrpRSPqdBrpRSPqdBrpRSPqdBrpRSPqdBrpRSPqdBrpRSPufLPlsHdjXzwy/cxRN3P00244zpXLNPnMF/PvMvRMujYzrPaKWTae75+v389oeP4WQdzn/Xav7mc1cQq4oNvrJSytd8dxnb1sY2PrTsE3S1dhe4qv6FoiEe7LmzaPMNlzGGT154E1uefZV0Ig1AMGwzbf4U/vuFr2MHffl+rZQ6xoS5jO0vvvUgPR09RZ0znUjzmx/8oahzDsfmZ15h69pth0McIJPK0vT6QZ6+b20JK1NKFYPvgnzDk5txneL/FfH4PX8q+pxDtXXtNpxs36bUie4km5/ZWoKKlFLF5Lsgn75gSknmnX/ynJLMOxSTZzdgh/oePgmXhZg2f3IJKlJKFZPvgvyqGy8jXBYq+rwf/Ndrij7nUJ1x6amUVUSwrN5tl+2gzQXXnFuiqpRSxeK7IJ9/8hy+cPcnqJ489M7toxGwLf79sS8RCASKMt9I2EGbbz31FRavWoAdsgmGbeYsm8U3Hr+JipryUpenlBpjvjtr5RDXdWnd30Z7cydNe1qYs3Q6qUSG/dsb2b5hF3bEpqqmglQiQ7QiwikXLOe5R9bTuKsJk3WZuXQG0ViEzrZuZi+ZRv2MenZs2ElZVQw7GKCrtYe6aTUsWDG3QFtdHJ0tXThZh5rJ1aUuRSlVYP2dteLbIFdKqePNhDn9UCmlVG8a5Eop5XOjCnIRuUpENomIKyJ9Pu4rpZQae6P9RL4ReDvwZAFqUUopNQKjugiHMWYLgIgMdlellFJjpGjHyEXkOhFZJyLrmpubizWtUkpNeIN+IheR3wP5/r/4zxlj7h/qRMaYW4FbwTv9cMgVKqWUGtCgQW6MuagYhSillBoZPf1QKaV8brSnH75NRPYAZwEPisjDhSlLKaXUUI32rJVfAr8sUC1KKaVGQA+tKKWUz/m6mePWddsIloVIdCSom1oDtkvT9hbqZtRg2QFwIFweBiDVncKO2pisS7wjQdXkCrIpF9dxMQGLjv1tLF65gEQiRWdzJx0tnVTWlTNlljZmmOjSjoMxhrDt65eDOo75cs+9/rRPsu2FnUWb7wv3fJzzrjy7aPOp4mjq6eazf/gdT+7aBRhOnzaDr150MbOq9BLAyl98d2jlX669uaghDvCVd3yLdDo9+B2Vb2Rdlyt/9lOe3LUTx7g4xvCXfXu44p6f0KPPtfIZ3wX5Yz/5Y0nm/cybvlKSedXYeHzndtoSCZyjrsfvGkMik+XBV7VhtfIX3wV5qezctLvUJagC2tneTtpx+twez2bY3tZagoqUGjkN8iFaft7SUpegCmhxfT3BPH1YY8EgJ07SL7iVv/guyN/xqctLMu9N9366JPOqsbF65mzmVFUTso68BGzLojZaxsXzFpSwMqWGz3dB/uGvXst5V55RvAkFvr/xG8WbTxWFJcJPr3gn71x2ElXhMOWhEJcvXsq977hGT0NUvqPNl5VSyie0+bJSSk1QGuRKKeVzGuRKKeVzGuRKKeVzGuRKKeVzGuRKKeVzGuRKKeVzGuRKKeVzGuRKKeVzGuRKKeVzGuRKKeVzGuRKKeVzGuRKKeVzvr1e546Nu/j+Z35Cd0c34UiIhpm1tDR20tbYRiaVpaK6jPmnzWXPy/uxbZtT3rSMDY9vIp1yOOVNywkGAmRSGdqbOmhtbMNxDJW1FSR6khjXZeXFK1h02jxmLZ2BiJR6c5VSql++vIztFQ3vp7Olu4AV9W/m4ml8+YHPMGPh1KLMp5RS/Zkwl7H95IVfKlqIA+zeuo9PXXgTTp7+jkopNR74LshffGxT0efs7uhhwxObiz6vUkoNhe+CvFTamzpLXYJSSuXluyAPRoJFnzObdjhx9eKiz6uUUkPhuyC/6d4bizqfHQpw2fUXM2lmfVHnVUqpofJdkJ++5lS+/KtPY4fG+MxJgVlLp/NPd97AR77x3rGdSymlRsGX55GfdelKfpP8aanLUEqpccF3n8iVUkr1pkGulFI+p0GulFI+p0GulFI+p0GulFI+p0GulFI+V5KrH4pIM7Arz6J64GCRyykF3c6JRbdzYhnP2znbGNNw7I0lCfL+iMi6fJdonGh0OycW3c6JxY/bqYdWlFLK5zTIlVLK58ZbkN9a6gKKRLdzYtHtnFh8t53j6hi5Ukqp4Rtvn8iVUkoNkwa5Ukr53LgLchG5SkQ2iYgrIr46BWgoRGSNiGwVkW0i8plS1zMWROQ2EWkSkY2lrmUsichMEXlMRLbk9tmPlbqmsSAiERH5i4isz23nTaWuaayISEBEXhCRX5e6luEYd0EObATeDjxZ6kIKTUQCwHeAS4ATgKtF5ITSVjUmbgfWlLqIIsgC/2iMWQqcCfzdBH0+U8AFxpiTgRXAGhE5s8Q1jZWPAVtKXcRwjbsgN8ZsMcZsLXUdY2QVsM0Ys90YkwbuAi4vcU0FZ4x5EmgtdR1jzRiz3xjzfO73LrwAmF7aqgrPeLpz/wzmfibcWRIiMgO4FPh+qWsZrnEX5BPcdGD3Uf/ewwR84R+PRGQOcArwbGkrGRu5Qw4vAk3AI8aYibid/wF8CnBLXchwlSTIReT3IrIxz8+E+3R6DMlz24T7ZHO8EZFy4BfADcaYzlLXMxaMMY4xZgUwA1glIstKXVMhichfAU3GmOdKXctIlKRnpzHmolLMOw7sAWYe9e8ZwL4S1aIKQESCeCF+pzHm3lLXM9aMMe0i8jjedyAT6cvs1cBlIvIWIAJUisiPjTHXlriuIdFDK8W1FlgoInNFJAS8C3igxDWpERIRAX4AbDHGfLPU9YwVEWkQkerc71HgIuDl0lZVWMaYzxpjZhhj5uC9Lh/1S4jDOAxyEXmbiOwBzgIeFJGHS11ToRhjssBHgYfxvhi7xxizqbRVFZ6I/BR4BlgsIntE5IOlrmmMrAbeDVwgIi/mft5S6qLGwFTgMRHZgPdh5BFjjK9Oz5vo9H/RV0opnxt3n8iVUkoNjwa5Ukr5nAa5Ukr5nAa5Ukr5nAa5Ukr5nAa5Ukr5nAa5Ukr53P8HRFlS27XZ798AAAAASUVORK5CYII=\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "\n",
+    "\n",
+    "plt.scatter(dataset['Avg_Credit_Limit'], dataset['Total_Credit_Cards'], c=kmeans.labels_,)  \n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "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.scatter(dataset['Avg_Credit_Limit'], dataset['Total_visits_bank'], c=kmeans.labels_,)  \n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXwAAAD4CAYAAADvsV2wAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nO3dd3gc1bn48e87s13VkiX3bmPcbTA2PfSYEgiBJJQkhORe0kgguSncH7nJTW7avem5ISQECJDk0gIECARCDQSIQYAxNrZxw1hyU7Paauuc3x+zllV2pZW0kmzt+3kePdqdnTnnzM7uO6MzR+cVYwxKKaVGP2ukG6CUUmp4aMBXSqk8oQFfKaXyhAZ8pZTKExrwlVIqT3hGugGZjB071kyfPn2km6GUUoeVV199tc4YU5HutUM24E+fPp2qqqqRboZSSh1WRGRHpte0S0cppfKEBnyllMoTGvCVUipPaMBXSqk8oQFfKaXyxCE7SieX7vz+/Txy89O01DXjOA7xSByxLcZNG8vEWeM5+qwlnHzxcRSNLeAb5/8PW1/fTsWUsfzXQ18jUBDguXtfYtvaHUTbY9Rs3k08lqCorIi97+wjEUtw1JmL+Pwv/wXbtkd6V5VSKiM5VGfLXL58uRnssMxkMsmFZR+nvSXS57qWx8JJOD2We3wenKSDk+z5WmdiCX/ccSMVk8oH3F6llBosEXnVGLM83Wujukvnvz70k6yCPZA22AMkYok+gz2AcQxfPOk/+tU+pZQaTqM64K9+9LVhrW/vO7XDWp9SSvXHqA74HJq9VUopNSJGdcA/+qzFw1pfxRTtv1dKHbpGdcD/xp/+DX/Ql9W6tif9W2F7bcSSPrcXgR8/+61+tU8ppYbTqB6W6fP5eLD5Dm69/k6e+P3fadsfxkkmScaTiGVROr6YiTMnsPysJZx6yQkUlxdy3arv8u7GasrHl/GtB79KUWkBz9z9Alte204kHGHXlr0kE0mChQEadjeSSCRZfNJ8vnjLZwgG/SO9y0opldGoHpaplFL5Jm+HZSqllDpIA75SSuUJDfhKKZUnNOArpVSe0ICvlFJ5YtABX0SmiMgzIrJBRNaLyDVp1hER+YWIbBGRtSJy1GDrVUop1T+5GIefAP7NGPOaiBQBr4rIE8aYtzqtczYwJ/WzErgx9VsppdQwGfQVvjFmtzHmtdTjFmADMKnbahcAdxjXP4FSEZkw2LqVUkplL6d9+CIyHVgGrO720iRgZ6fn1fQ8KSillBpCOQv4IlII3Adca4xp7v5ymk16/IuviFwlIlUiUlVbq1MNK6VULuUk4IuIFzfY/9EYc3+aVaqBKZ2eTwZ2dV/JGHOTMWa5MWZ5RUVFLpqmlFIqZdA3bUVEgFuADcaYn2RY7SHgahG5C/dmbZMxZvdg685G3a4Gbv/m3fzj/tVE2qJgDMmkg3H6N4eQP+Rj7jGz2fduHfverevIguX1ezn6rMV84ruXMWPh1KHYhZza+PJmbvuPu9i2dgcTZ4/nY//5YY46fdFIN0spNQwGPXmaiJwIPA+8CRzIBfj/gKkAxphfp04KvwRWAWHgSmNMrzOj5WLytOb6Fj654Is01TYz1JPEBQoD/O9L32P6gil9rzxC1r2wkeve+19Ew7GOZf6Qj6/d8QVO+oAOmlJqNOht8rRBX+EbY/5B+j76zusY4HODrau/Hrrxcdqa2oY82ANE26Lc9o27+M/7vjLkdQ3UTV+5o0uwB4iGY/z6S7dx4oUrcM/LSqnRalT/p+3aZ9cTjyaGpS5jDBv+uXlY6hqobW/sSLu8rqaBaHss7WtKqdFjVAf8SXMmZJWtKlcqD/EUh2PGl6ZdHgj58QW8w9wapdRwG9UB/8JrzsUXyC7F4WD5Q34+8h8XD0tdA3XZ9R8gEOqalcsf8vGBa8/Fskb1R0EpxSgP+FOPnMS3HvgKYyeXYeXgSt/j96QtJ1Qc5LM/+zgrzz160HUMpVVXnsZHvvlBgkUB/CE//qCP8z+7io9849A+USmlciMvUhwaY6iraUBESCaTCNC4r5ltb2ynoLQQsaC5tgVv0IMgFJSEsDweTNKhqLyQkvIiAgUBxk4qo701QnNDC/v37idYFCRYGKR8whhsj52Ttg6HeCxO4579lFQU49c8vEqNKr2N0smLgK+UUvlCc9oqpZTSgK+UUvlCA75SSuUJDfhKKZUnNOArpVSe0ICvlFJ5QgO+UkrlCQ34SimVJzTgK6VUntCAr5RSeWLQCVAOde9urKFmy25q3t7F3nfrCDeFadi7n0DIz5HHzaFy8lgKikP4Aj4mz51IW1OYqr+tYeemXXg9HpacOp/GPU3UVtez/c0djJteyXHvW45xHOpqGgi3Rlh57lFMmjVhpHdVKaV6NWrn0mnY08jX3/cDtq55pyP/7FA6cuUcfv7Cd3SaYaXUiMrLuXS++f7/Ycvr24cl2ANsXL2Zn3/mt8NSl1JKDUROAr6I3Coi+0RkXYbXTxGRJhFZk/r5Ri7qzWTX1j1sf/NdjDO8f708ccffh7U+pZTqj1z14d8G/BK4o5d1njfGnJej+nrVVNeC7bWhfThqOygRH578uUopNRA5ucI3xjwHNOSirFyYuXgqzjBf3QNMmDlu2OtUSqlsDWcf/nEi8oaI/FVEFgxlRf6gn0//+GP4gsOTzxYAga/dfvXw1aeUUv00XMMyXwOmGWNaReQc4M/AnO4richVwFUAU6dOHVSF5/7rmUyZO4k7f/AAG156m3BLe9o+fY/XJlQcJFQSYsbCqex8exe7Nu/GSbrrWh4LDF1u/vqCXnwBH9FwFBBmLpnGtb++itlLZwyqzUopNZRyNixTRKYDfzHGLMxi3XeA5caYukzraIpDpZTqvxEfliki40VEUo9XpOqtH466lVJKuXLSpSMidwKnAGNFpBr4JuAFMMb8GrgY+IyIJHDHzlxiDtX/+FJKqVEqJwHfGHNpH6//EnfYplJKqREyav/TVimlVFca8JVSKk9owFdKqTyhAV8ppfKEBnyllMoTGvCVUipPaMBXSqk8oQFfKaXyxKjLabvl9e088tsnaK5vZfy0sfzz0deo3rTbnfxMwOP1kIi589bbPpui0hBO0hAJx3CSjjunfep/gC1bKCwtoKSihHBLmPaWCJZt4fV7iLRGScST+IJelp22iM/94hOMnVg2gnuulFK9G1U5bR+9+Ul+de3viEfiwz4fvmVb3LT2x0ybN3lY61VKqc5GfPK04dDWHOaGa35HNBwbkeQnTtLhu5f+dNjrVUqpbI2agL/+hU14vPaItmHH+uoRrV8ppXozagJ+oMDf0fc+Uixr1LydSqlRaNREqAUnzMVf4B/RNhx3ftpuM6WUOiSMmoBv2zY/eOzrlFYUEyoKDm8+W2Dc9Aqu++MXhrVOpZTqj1E1LHPm4mncVXMTrz+9jramMPNPOILn7nmJ155Yy66teygsLWDBSXN55a9vYBzD3GNmUjllLImEw/59TYSb2mncu5/9dU04CUPF1HKOOn0RgZCfaCTOnu178Xg9jJ9Wwc63d9Nc30L5pFJO+sBxLDtt0UjvvlJK9WpUDctUSql8lxfDMpVSSvVOA75SSuUJDfhKKZUnchLwReRWEdknIusyvC4i8gsR2SIia0XkqFzUq5RSKnu5usK/DVjVy+tnA3NSP1cBN+aoXqWUUlnKScA3xjwHNPSyygXAHcb1T6BURCbkom6llFLZGa4+/EnAzk7Pq1PLuhCRq0SkSkSqamtrh6lpSimVH4Yr4EuaZT3+AcAYc5MxZrkxZnlFRcUwNEsppfLHcAX8amBKp+eTgV3DVLdSSimGL+A/BHwsNVrnWKDJGLN7mOpWSilFjubSEZE7gVOAsSJSDXwT8AIYY34NPAqcA2wBwsCVuag3G/F4nHBTmGBhEMtjkYgl8Pq8JJ0k4eYwoeICHCdJpDVCqDiEbdvEo3EsjwUOxKIxfEEfyVgS22MjtpCIJrA9NrbXxjgGj3dUTUmklBqlchKpjDGX9vG6AT6Xi7qytefdWq6YdbWby3aIjZtewef/95OsPPfoIa9LKaUGatT+p+1Hp392WII9wN53avmvD/+EN/6+fljqU0qpgRiVAf+x3z097HVGwzHu+M97hr1epZTK1qgM+M/e/eKI1Fv9tt6HVkodukZlwH/vlaeMSL0zFk0dkXqVUiobozLgn/rhExEr3f96DR1/yMfHv/3hYa1TKaX6Y1QGfIB763+HL+gd8npEhNnLZvD9v36dI1fMGfL6lFJqoEbtAPKSkgIeafu/kW6GUkodMkbtFb5SSqmuNOArpVSe0ICvlFJ5QgO+UkrlCQ34SimVJzTgK6VUntCAr5RSeUIDvlJK5QkN+EoplSc04CulVJ7QgK+UUnli1M6lA7C/tomffuo3vF21lWg4RkFJkOPffwz+oJ+aLXto3N3Ivpp64u1xCksLWHLKQk677ATefH4jb720CSfpMH7GOGrfraV+937CzWGmHDmJy66/iLdf2cKaZ9YhlsXxFxzD6ZedNNK7q5RSvRI33ewgCxFZBfwcsIGbjTE/6Pb6x4EfAjWpRb80xtzcW5nLly83VVVVA27T/T//Czd+8fYBb99fJRXF/GH7DQRCgWGrUymluhORV40xy9O9NuguHRGxgRuAs4H5wKUiMj/NqncbY5amfnoN9oOVTCaHNdgDNNU2818f/Mmw1qmUUv2Riz78FcAWY8w2Y0wMuAu4IAflDtgDv3h0ROp99Ym1I1KvUkplIxcBfxKws9Pz6tSy7i4SkbUi8icRmZKuIBG5SkSqRKSqtrZ2wA2KtEYGvO1g5KJ7TCmlhkouAn66XILdI9/DwHRjzGLgSSBtf4sx5iZjzHJjzPKKiooBN+iiL5434G0HY/ay6SNSr1JKZSMXAb8a6HzFPhnY1XkFY0y9MSaaevpb4Ogc1JtRsDDIqitPHcoqevB4bb75py8Pa51KKdUfuRiW+QowR0Rm4I7CuQS4rPMKIjLBGLM79fR8YEMO6u3Vv93yWY6/cCW/vPq31FU34hgHj9dm0uzxFJUV0bBnP021zUTCUZykg9fnYeq8SZx00bGse2EjO9ZXk0w6FJQEibREaW5sIZlIMmZcKWddcSrVm3ex6eUtiAhHn7GYT/7gcgqKQ0O9W0opNWCDDvjGmISIXA08jjss81ZjzHoR+TZQZYx5CPiCiJwPJIAG4OODrTcbx513NMedN6R/TCil1GEjJ+Pwh8Jgx+ErpVQ+GtJx+EoppQ4PGvCVUipPaMBXSqk8oQFfKaXyhAZ8pZTKExrwlVIqT2jAV0qpPKEBXyml8oQGfKWUyhMa8JVSKk+M6py2q35zE29Hmg8uSDgUvlaP1RSj+ZQJ4O10vjNgtSfAMThBD1Z7grLHaohX+Gk6flzXdTuVV/zSPua/HuZrt3yOJacs6PLy8+++w9effpLdrS0UeL3MKSunvr2dsmCQ8mCItXv30BaPEfR6OXPmbK5cchR/3fI2v39zDXVtbTidyrKg23PBa1sEPB6ml45hfGEhb+zZQ2s8Rsjj5fSZM7l84VJ+8MJzvFxTjQicNXM2Vx11DDdWrebZHdtpTyQA8Nse5o2toDwUYmtjA6WBAJUFBbyxdw+tMbe8hZWVNEejrNu3l2gyCYAtQsDjoTxUwCULF/PxJcvwe7p+pD7/14f565bNOMbgESGRmsrDI8J1J5xM3BjueON1attaSRiDBVQUFHLFkmV8fOkyAh5vP4+6UiqTUTuXzsKf/ZSwJEG6TddvjDtbv5D+NVLLO78v3dfrto1vZxtTfrKO/37iPzjq9MUA/G3rZj79yENZt1cAK1VPMhd5humZlGAo+SybRePGcffFl3Tsx5m/v5WtjY29bpepnR6xWFBZyZ8+eCm2pX+IKpWtvJtL5/UdO9MHe3CXWZL5tQPLDzzuLdin1otNKSAyMciPPvGrjsXXP/1Ev9pscAN9LoL9gfKGU8xJsqGulud2vAPAjv2NfQZ7yNzOhHHY0lDPs+9sz10jlcpzozLg3/D8C8NeZ9uScupqGjqeN7S3D3sbRlo4HueVXdUA3PvWukGX1xaP83KqPKXU4I3KgL9s4sRhr9NbG8HnP9jf7MnDboiAx8P4wiIAFlaMy0l54woKB12OUso1KqPS50472X2QrnvEmPTL08lmXWMgaSisquO8z5zVsfjCI+dl2drRwyMW7ztiLgCr5hyB3Vd3WB9sEd6fh++jUkNlVAZ8gJvOONd9cCBop34kmiT4ViMkTY/Xuv/YjVGIJHtdR6IOk366jmPPWcanf3RFR/3fO+0sjp00pUe7grYHj2Vhi3S8+ZYIBV4vn1x2NJWhgrRZ4XtjibjlpQKsJULI6+Wc2XM6lgF4LYtVs+fglZ6H/UAZIY8XWwSPWB0B28Jta6Z2ecRiUlExv7/wYkoDwY7lD374cjy9BP0Sv5+JRUVpyx0fKuD2919MWVDTRiqVK6N2lM4Bd65+lZd37KDE6ycQM1AX5rV9eyjw+5k/cTwN9c34/F6mzB7P/NIKkuEYL++qZn5FJdNLSqmtbuCNPbuoibZROHMsjYkojZEI3s37mR4s4sS5s1lyygKKy4rS1l8XDvPKrmqOLK9gYlER2/c3Uh4KUeoPsG1/I47jICLMKB2D3+PBMYZtjQ3Uh9uoawuTBIJeD5MKi9jd1kJbNIZtWYwJhSj2+RGgsrCwo7ykk8QS62B5jsPfd2wn6PVy7OSpALTGYuxsbmJ3czPRZIKpJWOYUFRIgdfnti8YYkwwyNbGBhKOg51qXzgeZ3dLMzubm7HFYmJxEcYYgl4vM0rHIBmC+z/e3c7GunoumreA13fv4p81O7l80RKmlY7BGMP2/Y3UtrVR19ZGRUEhYwtCvZanlMqst1E6oz7gK6VUPhnyYZkiskpENonIFhG5Ls3rfhG5O/X6ahGZnot6lVJKZW/QAV9EbOAG4GxgPnCpiMzvttongUZjzGzgp8B/D7ZepZRS/ZOLK/wVwBZjzDZjTAy4C7ig2zoXALenHv8JOF20g1YppYZVLgL+JGBnp+fVqWVp1zHGJIAmoLx7QSJylYhUiUhVbW1tDpqmlFLqgFwE/HRX6t3vBGezDsaYm4wxy40xyysqKnLQNKWUUgfkIuBXA50HnE8GdmVaR0Q8QAnQgFJKqWGTi4D/CjBHRGaIiA+4BOg+TeRDwIH/SroYeNocquNBlVJqlBr0fPjGmISIXA08DtjArcaY9SLybaDKGPMQcAvwexHZgntlf8lg61VKKdU/OUmAYox5FHi027JvdHocAT6Yi7qUUkoNzKidS0cppVRXozrFIcDe1lZqmpt4ZVc1AY+X0kCQXS1NJJNgWYbNDfXYlsWs0jGcMmMWU0tKebdpP2v37KE9GeeyRUtZv28v1c1NTC0ppTkaobYtzJ62VipDBSyorGR66RiC3vSp+OrCYVZXv8uYYJBif6BjuJKDO3RpbChERaiALY0NFHp9TCgqYmtDAz7bZlJhITe++grxZJIzZ8xCLIv2ZJy9La0cObaCEr+fP2/ayIKKSo6fOrWjzqTjsLWxgbpwK43tEVZOnkp7PMYbe3ZTUVCIiLCvtYXyUIhCnx9LhPGFRZSHDk5U1tTezr0b1jOtpIQzZ80BYHtjI+v37WFcYSEVBYVMKSrmxZqdtEQjTCsZgzEQTcbZ3dqG4zhYlsWU4mImFBVRmZrmuC4cpjbcxozSUtrjCfa0tjC1pJTq5maefWcbJ0+bzryKyqH4KAyKcVohuRPsiYhVMtLNUWpARu1cOvsj7Xz+r3/hhZ3v5rBV6fksiy8edwKfOnpFxzLHcfjIA/fyz5reE3gIYFsWPtsmnkxicGe1jCUSJPvRBo9YPHzpR9nd2sKXHn+U/dFIv/bBY1mcOXMWPzrzbK5+9CGeSWWuOtDGsaEQteFwl2XZfnIsYMWkyQS9Xl7c+S4eyyaSiAPue9ee7LqnJX4/L155FUGfr1/7MBSMcTAtP4TwH0C8YOIQPB8p/hbugDOlDi15OXnapffdzSs11V0Sfw+lgG3zo7PO5pw57nzw1z35GPe8tX6Yand5RPDYNpFUcvL+8lkWM8eUsbG+Lsctc/XnJDGpqJjnr/zXIWlHfzhtv4OWnwGdM5gFIPQRrOKvjlSzlMoo73LaVjc3sWbP7mEL9gCRZJIbq17ueP7Axg3DWLsrYcyAgz1AzHGGLNhD//Ls1rQ0D1k7+qXtFroGe4AItP8fh+rFklKZjMqAX9vWhte2h7/ecFvH44QznKeb0SmZ7E+n1hBxmtIvNxEgPqxNUWqwRmXAP6J87LAHXAFWTJzc8bwsGMy88hDyDXJOukNlRjsB7BE4affgXZh+uT0N9/8MlTp8jMqAX+Dz8cWVx+MbxkTiBT4fXzruhI7n3z/9vcNW9wHzyscysaRkwAc1YNt8udM+5JIH92Z0tq5YctSQtKO/pPj/AUEOflUECCDF38i8kVKHqFF70xbgqe1b+elLL7Kxbl9W/fley2J2WRnbGhuJproTCr0+Ek6SWNIBDIaufdFB28O5R8zl8yuOY0pJ1+F6z7/7Dl9/+km3P9oYPLaNx7I6bl56LYspJSVMLS5lU30dQa+HqcWlbG6oxwKqW5ppix/sNvCJkExtG/R4SDoOkWSyI9n3D888m+ZolFter+LON9+gMRLBAGMCAeLJJC2xWEfaQMcYLMBn2wQ8XuZXVPL5FceycvIUXt9dw6cfeZjacBuWCKdMm8Ep06bz09Uvsj8SwbYsJhQWUuTzs7m+joQxeC0Lu9voIknt45EVlXxhxXEEPR5ufPVlapqbmTWmjLZYjD1trQQ9HjY31BN3HPy2zTUrj+PTy1f2+3gPFRPfjGn7FcTXgWcOUvgZxLtopJulVFp5OUpHKaXyUd6N0lFKKdWTBnyllMoTGvCVUipPaMBXSqk8oQFfKaXyhAZ8pZTKExrwlVIqT2jAV0qpPKEBXyml8sSgAr6IlInIEyKyOfV7TIb1kiKyJvXz0GDqVEopNTCDvcK/DnjKGDMHeCr1PJ12Y8zS1M/5g6xTKaXUAAw2R9sFwCmpx7cDzwJfG2SZOfPmph389MG/sT3ehowJUE+MiJPEFovSQIA55eWcNHU6r+/exeqaapoi7YBgW8LUklLOO2IuWxsb2NPaSnVTE/XhMHHjYIvgsz3MKSvjS8efyElTp/eoe0PtPv701nrebqgj6TgIwpSSEmaPKedv2zbTFIly/JQplAQCrN23l+ZIhNpwG4Jw/JSpXHfCyTy+dTO/W/MaTdEIU4tLOX3mLIIeD09u38rb9XUkHcP4wkJOnzmLD81fRG24jR/84+9srKvDMQ6Ti4pZPH48L1VXs7ulGcdAkc/HtJISygsKWDp+PNsa9rNm724mFBZRWVDAG3v3MK6gkFOnz+Cd/fsZEwxy8fwFFPn83LdhPatrdhJJJDiivJwPHLmABZXjAPjb1s3cWPUykUSCD8xbwBWLl/LUO9v484a3eGvfPloTMRwDxkki4k60Nrm4mHPmzKW2rQ3bEs6fO4+FqfKyZYyB2MuYyONgBZHABYj3iAF9XozTgGm7D+JVYFLpHK1S8B7jPk5udidPC74fsYoHVEeX+uJvYyIPghNBAmeBb0XH5HZd1jMGYi9iIk+ChMB7BMTfBLxI8H2Id37fdRkHYs9hIs+CVYIEP4B4pg2s3ck9mPb7IbkH8R8H/jMQSZ/TOWMZiZ2Y9vtS+wF4FiGhixHP5N437FLGdkz4TxB/C8QD3qVI6CLEHt/3tk4YIn/BxN8EzywkeGHGXMUmWYsJ3wfx18DEwDvPrcczO+u2Qm6PwUANavI0EdlvjCnt9LzRGNOjW0dEEsAaIAH8wBjz5wzlXQVcBTB16tSjd+zYMeC23fLA03x3axV4LLBSXyJjYJDzxafz0UVL+NapZ3Q8/03Vy/xs9YsdM24OB69lER+CHABey0JEsEWIJZMkO31e/LaHq1esZGNtLY9sebvHdo7jZJ2X10LweWw+ffQxfGHl8VltY4zBNH0Noo+nEpJYgBeKvoxV8LEsa06VFV+Pqb8ciECvc6sGQXxI+T2IZ0a/6ujMabsDWn6Em0TFAQmA/71IyX93CfrGGMz+ayH6LF0zb0nqxweFn8Yq/GzmfTMJTOOnIP5q6kTmcX9Kvo8VPLdf7TbRlzD7Pw0mCcTcE5A9Eyn/P0QC2ZUReQqz/4tAlINzz7r7IqW/QAKn9lmGE34Qmr+eKuMAC/AjY25C/JlnWzXJWkz9ReA0A2EgkDqmd/UI4ib2OqbxytTnq/Pnwg9F12MVXNL3DpPbY9CXQc2WKSJPAulOmdcDt2cZ8CcaY3aJyEzgaeB0Y8zW3uodzGyZ0VichT/5CcnQ8CWZfvlfPsPYUIialmbOuOPWYQ32IynXJxq/7eGxy69gWmlpn+seDD7dUxD6kYpnEHts1vU6tedAckuWawt4j8Eq/0PW5XdmkrWY2tPoGqwACSKlv3avmg+sG30Ws/+aNPvYmR8Z+yjimZK+vvaHMU1fp0eqRgkilf9EJLtkPcYkMbUnglPf7ZUAFH4eq7DvHMTGxDD7jgXTmn4FKUIqX+o1uYxxWjH7jsc9OadhVSAVzyOSvsfa2f8ViDyCe/3ZUTF4l2KV392prcY9Tk5Nhpb4kcrnECvtrcuubW7/C6b5+jTHMejurxXqs4xsDWq2TGPMGcaYhWl+HgT2isiEVCUTgH0ZytiV+r0Nt9tn2QD3JStPvrQWYw9v7qY/vrkGgGff2Z72z/LRysnx9NoGw9PvbMtu3chj6QOh2BB7Lvs6k/WQfDfr9cFAvApjBpjiMPa828Yexba7+9Rl0V/7CPYp0WczvmTaH6ZnXl4AG2L9uKhKbM7QlghEHs6ujPgbfaxgDnbzZBJb7XbhZCyiDRK9XE9Gn6JrsD9Q71qM6XQSTlanObl1Ih6IvtB7Ww+U3v5w5s9qfPimgR/sTduHgCtSj68AHuy+goiMERF/6vFY4ATgrUHW26ug39evhNk5qdPjfgB9to2VRwE/1yc3SwRftqkNxU/6j3CqqyNb4qF/KdZJ1TvQr4+P9MkkrdQ+dZJxHzuvk2a7HmX01pYsic/tFu13Hd3r6+29Nj2lD6AAABvJSURBVG49fbWjN8bpY51MJ4sD3WQHnnrp83ORbZrLjO+PoV/HYJAGG/B/AJwpIpuBM1PPEZHlInJzap15QJWIvAE8g9uHP6QB/5QVC/BFHHCGJ+wL8JFFSwE4c+asnF/1HsqsXGfBNfDeWXOyWlWC7yf9l8UB/ylZVylWCXiXkv3XwZO6UTnAnLv+U0h/n8CX2qdObQteRJ8BwRgInJHxZQl9CNJ223jBd3Qfje3EngH2eHqerIJI8NLsyvAuAinI/LoUgmdB72X4VpL5WAnYk3q/GRq8EOgegD3gf0+XriSxx4NnNpkzPRvwn9R7Ww+U1esxSNv7MiQGFfCNMfXGmNONMXNSvxtSy6uMMf+SevyiMWaRMWZJ6vctuWh4byzL4oYzz8VqT0LCcb8QxvT/Ii5L3zntTEI+94NSGgjys/eeg3+QV/o+K/tgYolQ4s/2Cit7XsuiwOul0OdjQmFhlzZZIgQ8Hn626hy+cvyJPbatDIXwZOhDTVdPyOvFb3v40VmrGBvKrj9TvPOh6BrA736ZJAQEkdL/RazCrMroKKv0x2BPBNKNNhEO3BBECsAzAyn5Vr/K71KaVYiU/i/uDeBQKhD4oeiaHiNuxLcECj+NG/SDHAz+fiAEBKDkh4hVlrlC34kQvDS1TdDdBylExvwG6a1rpHu7RZAxN4BVlgraAfcncBZ0O1FlLsNCxvwGpISuV9oeoDjVpt4/NyI+ZMyvcfe/8/fEC1a528beti+6FrwLUu97wN0XeypS8t2e65b+HKxKun4ubNzP2a+yvv+B7wQIXkbPY3BTv47BYI3qFIdt7RFufuBptjY2UDKljE3RJhrb2wl4PEwoKuLoCZM4ZuIk1u7dw9/ffYdtDfXEk0kK/QGWTZjAZQuX8sbe3expbaWmuYmNdbU0RaOEvB6K/QGWT5zEZ5evpCxNgGqORnh6+za2NTbis23a43GmlJSwbPwE7tuwnvpwmPOPnEc8meTt+noiiTjbGxsBOG3mLC48cj7bGuu56dUq6tvDLKoczzETJxHyelldU03VrmocY5hdVs4xEydz8rTpxJNJfv3qy7yxdw/RRJL5FRWcMXMOD2xYx9p9ezAGJhYWsaCyktJAkPmVlexrbeO5d99hRukY5owp42/btzKxsIjz5szljX17KPYHOG3GDHy2hxd27mBjXS0t0SgzSsdw+sxZFPvdkRl7W1v5VdVq2uIxrli8jEXjxrOloZ7n393BKzXVNEcjJByHuJMkaHuIG8ORYyu4ZMEi1tfuwxaL02bMpCSQ3UiPzkxyL0Sfd/9s9p/a72DfUY5JQuwFTHwTOG0gSbDKwbMQECS5Gezp4FuZk64s47RC9BkwUfCfhNiZh6Sa5G6I/gMkiPEuRmJVbv+v/5SMwwl7lJF4F2L/BKvIfZ+yHFXToxwTc99vpxZ8y/s9PNEtIwrRZzGJLWAE8c5KtSn77g3jhFNlbAXxIp454D85qyGixhiIr4HEJrCngu/YjCcaYxIQfR6T2Awmgnhmpj5nvfylkqnexE6IvZQ6Bqdkf8LoB81pq5RSeUJz2iqllNKAr5RS+UIDvlJK5QkN+EoplSc04CulVJ7QgK+UUnlCA75SSuUJDfhKKZUnNOArpVSe0ICvlFJ5QgO+UkrlieGbpm2YxRIJPvHQA7xY3Z/EFt10nmeoj8myVs2cw6/Oy5yf/bXdu/j2s0+xtjZtjhgClsUnjzqGP21Yz942NxtQoc9HWyyWdpLP8kCQLx53ApcuXJzVRF6N7e187x/P8uCmjSRSGaosoCQQoDUWo9jnJ+Tzsae1hQKfj2Kfn92tLfg9HpKOQ3vCTRgxb2wFy8ZP4J631pFwHDyWxccWLyXhONz71rqO9Q4o9fv58nEn8ti2LbxcU03I4+XShYuZWlLKL1a/yJ7Uvnotm0nFRextbcPBcPqMWXzj5FOpKOj/BFWZmNgaTMt33RyoVikUfBIJfRySOzHN33EntZIABC9Gir6IZD3Heyo/bct3IPaaW4aMAWdXqryLkKIvgYlgWr4Pkcdxp9Y9Eyn+f73PdJlt/U4Y0/JjiDwAJg7+E5Hir4M1HtP2W2i7HUwzeJcgxdcj3j6mIO6rvsT21Hv2Ih1TPdtzkZKvI74VPdc3EUzLT6H9T+5kcb7j3fbZkzFtt0L4VnCawLsAKbrenSW0E6fl59B2E25KSBsCl2CVfrP3NhqDab8XWm8Ap87NXVv0NcR/wuD2PVmLafkeRJ5ycxEEznXLzUGe46E2aidPO+v3v2NLY0MOW9S3i46czw/POrvH8g21+7j43jt7BMPB8lsW1x53Ap86uucXrLOE47Dqj7fxTmNjr9laB8MSyToPgCeV77a3ttgiVBYU8tTHriTg6V+C7HRMfAOm/sN0TYsXhOBFEHkITAsH58/2g28lVtnNPQtKV3ZyF6buXDfTUlp+8B4Nzp5UZq0DnwMP2OORsY/1a5bIHvUbg2m4HOJrgVhqqQVSCv5TIfIoXTJeSQgpf2DAOXlNsg5Tt8o9gfQQQMruQHxLuyx1Gj4BsVc4mNbRAimCwLnQ/kDX9hFEyu/tSEbvtPwC2n6ZpqqLsEq/n7GdTtvvoPVn3TJNBZCym9OelLJhTBRTe5Y7U2jHcfS6U2aXP9Tn1M7DIe8mT9vWWD/swR7gvo3p87r84uWXch7sAaKOww2vrCbeR/7cv7+znb2trUMW7KF/qQ4TfQR7gKQxNEUj/HXz5sE1LMW0/ooeOWRph/Y73SvOLn9HRSH2sjt1bzZlt90KJtbLGlGIvwLJ3XRNrZeAZEMq5d4gJNZBYj0Hgz2A4ybLjnQPpoCJYtp+M+DqTPj/Ukm904lgWn/Wdf3426lUip3ff8cto/3unu0jimn71cGnbTdlqOp+nAz5lI1JQOsv06QVjGBafpKh7VmIPAamia7HMe6mQ4y9NPByh8moDPirazIlHR4Zb9XWDlnZCcehvj3c6zqb6uuG5IQz1MLxOG/Vpe8C67fEW2TOgNP9RICb9jDLgE98HW5XQ1/SBcmwGxAHI76Z9FmZIhmWJyG+fhD1df5LIo1Et/1JbKZropID0rzvADjd2pepLgNkSIbuNGU+CfeW77YPJr7RPZH2fMGdW/8QNyoD/rLxE0a6CV3MKSsfsrItEcYEek+iMHNMWUfO3cNJyOvN3Xtn95aqLk13ikm4Kf2y4ZlL37fDBDdDVPfFITehxmBk7Jrxk/4kZ4HniIHX551H+sxgKd3fN88MMqV0TE+6ta+3HLQZkt1YxZkTnfeW/rAPbrKXNBnZxJf952UEjcqAf+TYCiYXDf8NlFUz0+di/fzK4wgMQcD1WxafXLYcfx9lnz5jJqWBYK6zz3bRn7JtkT7Xt0QIerycd8SRg2lWByn6HD3zmAYgcH6aRNQ+8C5CvHOzK7vgE30ks/a6eVrtMrpe6VpuDtfAWVnVk7n4palg0zkISyoD2HvpeaLxIQWfGnB1Erq8l/0NIIVf6Lq+d376k4T43D78Hu3zI4WfPfg0dEX6qvxnYVnpQ5iIFwr+JU0e2QBSeG2GtmcheA5YIbqGTg9YY8F/8sDLHSaDCvgi8kERWS8ijohkzMQrIqtEZJOIbBGR6wZTZ7YevfwKFlVmThmXlY5cuH3nw10xcXLGUTpLxo3npvPez7SS0ozbW8ClCxZT3Ck3rd/OnNc25PXyhZXHc83K43pvGOC1be770KWcPnNWz3I8XiT1uzzonhQCHg9jg+5VjNey8HT6Uk0tLuG9M2d1BGxLhPOPOJLz5x6JnWa0UNDj4UvHHs+SceOR1D59aMEivnr8SV3y8ApQGSrAFsES4YQp03jgw5cR8g7+hi2AeBcjY24Ee6ZbmxRCwZVIyfeQsjvBe1SqFT4Ino+MydBvnK5szzRkzO2p5NuCm/u2IvWqD4LvQ8puQcruTiUwt90f38lI+Z/6NRoobf0iSNntEDgbN6ha4F2OlN+NlP4IQpel8v26V85SdkvHDdEB1WePR8r+DzxdR9JgTUZKf474V/bcZszNEHwf7lW9gPdopOwupOQHbkCXAne5PcvNa9tpFJFV/DUIXMrBcCXgXwUlP++9nQWfhYLPp/LnCtiToeRHiL9nDuas913cG8r4jk+1xwP+09x9GWhS+2E0qFE6IjIP92+13wBfNsb0GFYj7rvwNnAmUA28AlxqjEl/hzMlVykOHcch5jhEYzECto2xbfcwWVaPPzLDcbcfNuT1guMQSSTwWja2JTi4ydEt3B2OJxJYxtBuDKX9yMOacBzEGJLGkHQcPCIkUwnBD4glElipQHvgZmg4Hifg8eA4DpZluVfJA8ireqC8WDLZUWfCcTrK6/w46ThYqcexRAKPZXW5oookEl3abYzBMQYDHfvo6/R65/I6vx8Hkr0fGOljjMHOcOWWC8YkALvH+2dMErAGla+2cxmZyjPG/eQNxYgO9/vs9Ag+7vJkzhNmG5PEGEHEyarswbbPcSJY1gDyHpvEEOz70B3HwehtlM6g3gFjzIZUBb2ttgLYYozZllr3LuACoNeAnyuWZRGwrLRdKt0PU+erayyLwgxdJRbg8bl/0vb32uzA1XLnj3v3a9jOQfJAMCxM1ccgA+GB8jq/H52v4Ds/7hx0fWnei+7vqYh0ucrvfr2TLoh7ui2zRPr8n4fByvTFz8UVWucyMpU3lAHC/S72rNddnvtuRRE7dbiy26fBtm8gwd4tfyj2/dAK9NkYjhZPAnZ2el6dWtaDiFwlIlUiUlU7hCNblFIqH/V52hORJ4HxaV663hjzYBZ1pLtcS9uPZIy5CbgJ3C6dLMpWSimVpT4DvjHmjEHWUQ1M6fR8MrBrkGUqpZTqp+Ho0nkFmCMiM8T9//FLgIeGoV6llFKdDHZY5oUiUg0cBzwiIo+nlk8UkUcBjDsk4mrgcWADcI8xZhD/5qeUUmogBjtK5wHggTTLdwHndHr+KPDoYOpSSik1OIffuCKllFIDogFfKaXyhAZ8pZTKExrwlVIqTxx+c+YOwMa6Wu5Z/yY7m5owGLY2NBB3kpQFQiSMQ3koxBkzZnHMpMms2b2LR7e8TdxJgoG6cDs+22JGcSke22JXa2sqgYfBY1nMG1vBOXPmcsLU/k25uqe1hVdqatgbbqUyVMCycRN4fucOXtu9i3A8zuzyci444kju2+AOaHrfEXP5VdUrNLS3cdLU6ayYNIXKUAFVu2uo2lWDz7b5wLwFLExNGGeMYc2e3by5bw9v1dYiwLyKShzHgBj8todwIo5HLAIeL7PLyzhq/EQcY1hdU832/Q04ScPmxnriySTjCotYMn48iyvH88quGjyWO8FZ0Otlc309G+trmVpSSms0ylPbtzJzTBmnTJtO1e5dVO2qIeE4TCwsZGN9HXXhdkIeDxOKiplfUUHSGAyGYl+AeRUVzK+o7Nd7aZwmiP0T8IH/+P6lJjQOxKsgWQu+pYjd9Z/AjdMGsRepaWlmbeNMppROHPCkfCb+FiS2gWdO1jNxZl12YgvEN4JnKngW9Ws+IOO0plIV2qn3r/fptnPJJPdC7FWwSjGeOUi8yp1IzXdsn1nAjIm7SUdMG/hW9kgVaZwGTHQ1OLuBAiCK2JNT+9h1igaT2AHxN8GeAN6jsn7/Dh7TWYh3Xhb7W+cmw5Eit83SdWIVE9/o5mHwzHRnGc2xUZvi8ICP3H8PL1bv7HvFQZpUVMxTH70y7ZwznRlj+J8Xn+d3r79K3HH6moSz35aNm8CvzzufK/58H5sb6klmeXx9ts34wkIi8QSNkXbiGTIJAQRtD7btTuw2p6ycTfV1WAjhRDZJQPpux9Jx47n1gouyminTCd8Nzd9JzX3ufkllzG8Q3zF9bmuSNZiGj4LTmPrf77ibf7b4PxERnPYnME1fJpJwSBoHSwzXvXImu+Mrue2CiyjyZ3diMU4rpvFfIL7BzYFqkuBb5rZTBjY3TEfZJobZ/3mIvui+B8Zxg0XZbYhV0uf2Tvuj0HRd6v0zgEnNePmeQbWr73YbN/NU+LZUu+O4iU6C7nuENzWr56L028fXYRo+QUfiGZOAwmuxCj/p7lfb7dDyw9Trnb8DNuBHxtyA+E9wJ39r+ipE/gZ4QAxYE9w0jXYFmRgnjGm8yj1JHDim3sXImJsQK818+YDTeqObX7cjyPuRst8h3nkY045p/BTE1oDY7nH0LkDG/Bax+pfXubfJ00Z1wL/5tSq+94+/56hFfTt79hxuOCdzInOAp7Zt5ZrHHslJcMxkWkkpO5v2D2lKw6Hks20unreA75x2Zq/rmcQWTN0H6JFJSgqQyhf7vFJ16t4PiY10Tc4RREq+Db7jMbWn9yi7PWHz3sc+ysqpy/hxmvzFaetp+ndof5iumZv8ELoEq/j6rMrIWHbLL6Dt5m7t9IL/NKwx/9vrtia5283P2iPzVACp/DtijRlU23qtO/IMpunaNCkIO5FSpPKFnlfBJoHZdwKYxm4bBJGy20A8mPrLSZ9h7EDZQaTiBUz7/dDyY7qmWbTBdwxW2R0ZN3eavgHt99P1mPog+EGskp7J1U3sZUzDv9IjnaM1Fql4HtP8XWi/l67HwgfB92OVfCfzfqTbtXzLaXvArWteHdb6ntzWd+q037+5ZkiDPcCOwzjYgzt18/0Z8gN3ZsL3kTG1YLT3E71JVKdS3XV/p9ox4T9A5K9kSoJw+sTN/OXtjVnl8TXGpAn2ANFUwBik9rvpGdjiEH0a02ueXSDyCGn3USR1xTt0TPiPvQd7ABIQe7nn4thq0qc9jGDa78aEuwfOdA0QiD4J4T/SM6duEmKvYpzuJ5RO2h9M04ZYKodwmurCd5H2BGTaIf5aarvubY5B+4Pk8qJ8VAf8yDDncc2m+6Q12scHUQFu0O/zg25agXQJ3B1wes/ziwm7fzqn47Sl8pb2/Px4LYcCT7wjn0HfTNpy3Jdy8FnImEzcSXWT9LKp00raE6ZJps/bmkumbeDrmTAZ52R0WsC00GfGogNJ3jOedKyMx8f9XGY4dpmOqZOpTZJqR6bPQm4vDkd1wD99Rs8MT0PpiCzyr557xNxeM1nlQsiTmyxRI0WAlZMm93njTAJnpDI5dWMc8B/feyWeWaTNMYvfzRzlP5l0eVvjjs3f90xh6bjxeLM4jiJWp2xanVmprEmD5H8Pab/GniP67PsV/ymQ9h6CgP+kwbetN4GzSf/+d2Li4OuZPQvfigwnsyASOBsJvDf956Jr4eA7EfxnkHbsijUWrPQ350XEbUOPYyrgOzb9NoGzgTRdjCYB3qNT26Ur75hBJeTpblQH/G++5zQKcpQiry+2CL9YdV6f6126cDGzysrwW0MT9AMeDzee+75+Jy0X3HSGvixPRgdSIQZsO6f5em0Rivx+vn3K6X2v7DspFTQPfLkFCELhZxA73YzeB4nYSOl/4wadA+0Pgj0RKfi4O0IieCEOQRwDjoG2hIcHd8xlZ3hin/cXutRV8m03pWJHupwASBEyyP57ACn6CkgpB4Onz72HUfK9vjf2LoHAKg4GotT7F7oslax76EjoQ+CZmSbnLLhhKQBFX01741msEii6zl3nQAiTEHgXu/vjPxO8y8icnigAoY8jnqlI0dVgVXDwPfC6/ful/9NroJXi/3RH2nTU4U8d02+k3yB4Hnjnc/CzmtrH4q8jVqG7nRRx8Dj6QQqR4p73AwZjVN+0BQjHYvz3i8/zyOZNtEajqSGVXVnApOJiTpk2g3/s3MHOpqbUUMGDbBEshKQ5uP2BXLDvmT6D/zj5VMYVFmbVpmgiwSObN/Hwpo3UhtsoC4ZYOK6SF3bsYGtjA44xjA0VMG9sBZvq6wCoLCjgrdp9xJJJxoYKOHfOEYwNhfjbti28s38/tmVxxoxZ/PuJJ1PkD7CrpZk/rl3DE9u2sretFUuEYn+AAq/34H4Zg4gQ8npZMWkyly9aSnsizh/WruGNvXtoiUZpjERIJJMU+f2smDSZBZWVrNu3D69lcdG8BRw5toJ731rHa7t3MbGoiB3797Oudi/jCgo5bspUqnbVsL2xkXgyice2aYtGSOCmQAx5fZQFQ+7JSQxF/gAnTZ3GJQuXMDbU1xXagV1w3P7qyKPuFzV4EeI7KqttAUziHbd/NVmD+E9y88+mgpAxBmIvEm29n3f2N/J4zXzEfzyXLlxCRUH/Rk6YZB2m/W53pI53IRL6UI9hhANlnGa33zq+BjyzkdAliJ3d0FF3H5/HtD8MYiGBC1PDBYc265hbdwwij2Aiz4BVDp4pEF8LUoKEPoh4F/a+fXw9JnwPmCb3qt5/ZkdmK2OSEH0S0/4AJHbj3qvxgHcWEvpwl1FcxmnFtP/ZvTfgmYYEL0E8k/tuv9Pgfnbib7kjakIf7vWYGhOHyOOY6JPuDenQh7oMvXTLuwfi68A7Hwl+GLH77jXoLm9H6SilVL7J21E6SimlDtKAr5RSeUIDvlJK5QkN+EoplSc04CulVJ7QgK+UUnnikB2WKSK1wI4ML48F6oaxOSNF93N0yYf9zId9hEN7P6cZY9JO9XnIBvzeiEhVpnGmo4nu5+iSD/uZD/sIh+9+apeOUkrlCQ34SimVJw7XgH/TSDdgmOh+ji75sJ/5sI9wmO7nYdmHr5RSqv8O1yt8pZRS/aQBXyml8sRhG/BF5IMisl5EHBE57IZH9UZEVonIJhHZIiLXjXR7hoqI3Coi+0Rk3Ui3ZaiIyBQReUZENqQ+r9eMdJuGgogERORlEXkjtZ/fGuk2DSURsUXkdRH5y0i3pT8O24APrAM+ADw30g3JJRGxgRuAs4H5wKUiMr/3rQ5btwGrRroRQywB/JsxZh5wLPC5UXo8o8BpxpglwFJglYikz/c3OlwDbBjpRvTXYRvwjTEbjDGbRrodQ2AFsMUYs80YEwPuAi4Y4TYNCWPMc0DDSLdjKBljdhtjXks9bsENEpNGtlW5Z1ytqafe1M+oHBEiIpOBc4GbR7ot/XXYBvxRbBKws9PzakZhgMhHIjIdWAasHtmWDI1UN8caYB/whDFmVO4n8DPgq9AjW+oh75AO+CLypIisS/MzKq94U9IlEx2VV0r5REQKgfuAa40xzSPdnqFgjEkaY5YCk4EVItJ7UtrDkIicB+wzxrw60m0ZCM9IN6A3xpgzRroNI6AamNLp+WRg1wi1ReWAiHhxg/0fjTH3j3R7hpoxZr+IPIt7f2a03ZA/AThfRM4BAkCxiPzBGPOREW5XVg7pK/w89QowR0RmiIgPuAR4aITbpAZIRAS4BdhgjPnJSLdnqIhIhYiUph4HgTOAjSPbqtwzxvy7MWayMWY67nfz6cMl2MNhHPBF5EIRqQaOAx4RkcdHuk25YIxJAFcDj+Pe4LvHGLN+ZFs1NETkTuAlYK6IVIvIJ0e6TUPgBOCjwGkisib1c85IN2oITACeEZG1uBctTxhjDqshi/lAp1ZQSqk8cdhe4SullOofDfhKKZUnNOArpVSe0ICvlFJ5QgO+UkrlCQ34SimVJzTgK6VUnvj/mfsi7vQ92bEAAAAASUVORK5CYII=\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plt.scatter(dataset['Avg_Credit_Limit'], dataset['Total_calls_made'], c=kmeans.labels_,)  \n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### The clusters we are visualizing seems to do a good job but the preferred way will be to reduce the dimensions to 3 or less (if possible )  and then try to plot the clusters\n",
+    "HINT: Try PCA before clustering "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Analyse the Clusters "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Let us make a visualization to observe the different clusters by making boxplots , \n",
+    "for the clusters we expect to observe statistical properties which differentiates clusters with each other "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1440x1080 with 8 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "dataset.boxplot(by = 'KmeansLabel',  layout=(2,4), figsize=(20, 15))\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Looking the box plot we can observe differentiated clusters "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Hierarchical Clustering "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Now that we have tried Kmeans , let's try hierarchical clustering with different dendograms for the same dataset and choosing the best using the cophenetic coefficient by using different types of linkages "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 25,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x936 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "single 0.7391220243806552\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x936 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "complete 0.8599730607972423\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x936 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "average 0.8977080867389372\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x936 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "ward 0.7415156284827493\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x936 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "median 0.8893799537016724\n"
+     ]
+    }
+   ],
+   "source": [
+    "linkage_methods=['single','complete','average','ward','median']\n",
+    "results_cophenetic_coef=[]\n",
+    "for i in linkage_methods :\n",
+    "    plt.figure(figsize=(15, 13))\n",
+    "    plt.xlabel('sample index')\n",
+    "    plt.ylabel('Distance')\n",
+    "    Z = linkage(subset_scaled_df, i)\n",
+    "    cc,cophn_dist=cophenet(Z,pdist(subset_scaled_df))\n",
+    "    dendrogram(Z,leaf_rotation=90.0,p=5,leaf_font_size=10,truncate_mode='level')\n",
+    "    plt.tight_layout()\n",
+    "    plt.title(\"Linkage Type: \"+ i +\" having cophenetic coefficient : \"+str(round(cc,3)) )\n",
+    "    plt.show()\n",
+    "    results_cophenetic_coef.append((i,cc))\n",
+    "    print (i,cc)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 26,
+   "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>LinkageMethod</th>\n",
+       "      <th>CopheneticCoefficient</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <td>0</td>\n",
+       "      <td>single</td>\n",
+       "      <td>0.739122</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>1</td>\n",
+       "      <td>complete</td>\n",
+       "      <td>0.859973</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>2</td>\n",
+       "      <td>average</td>\n",
+       "      <td>0.897708</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>3</td>\n",
+       "      <td>ward</td>\n",
+       "      <td>0.741516</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>4</td>\n",
+       "      <td>median</td>\n",
+       "      <td>0.889380</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "  LinkageMethod  CopheneticCoefficient\n",
+       "0        single               0.739122\n",
+       "1      complete               0.859973\n",
+       "2       average               0.897708\n",
+       "3          ward               0.741516\n",
+       "4        median               0.889380"
+      ]
+     },
+     "execution_count": 26,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "results_cophenetic_coef_df=pd.DataFrame(results_cophenetic_coef,columns=['LinkageMethod','CopheneticCoefficient'])\n",
+    "results_cophenetic_coef_df"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Looking at the best cophenetic coefficient we get is for \"Average\" linkage.\n",
+    "\n",
+    "But looking at dendogram 'ward' and 'complete' show good difference between clusters.\n",
+    "\n",
+    "So choosing 'complete' because it has high cophenetic coefficirnt and good cluster segregation.\n",
+    "\n",
+    "Lets make a dendogram for the last 25 formed clusters using complete linkage to have a better view since the above dendograms are very populated "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 27,
+   "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": [
+    "#use truncate_mode='lastp' to select last p formed clusters\n",
+    "plt.figure(figsize=(10,8))\n",
+    "Z = linkage(subset_scaled_df, 'complete', metric='euclidean')\n",
+    "\n",
+    "dendrogram(\n",
+    "    Z,\n",
+    "    truncate_mode='lastp',  # show only the last p merged clusters\n",
+    "    p=25 # show only the last p merged clusters\n",
+    ")\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Let's take a maximum distance around 3.2 to form the different clusters as we would like to have 3 clusters \n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 28,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "max_d=5\n",
+    "from scipy.cluster.hierarchy import fcluster\n",
+    "clusters = fcluster(Z, max_d, criterion='distance')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 29,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{1, 2, 3}"
+      ]
+     },
+     "execution_count": 29,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "set(clusters)  # So there are 3 clusters which are formed "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Assign the clusters label to the  data set"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 30,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "dataset2=subset_scaled_df[:] #Create a duplicate of the dataset"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 31,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "dataset2['HierarchicalClusteringLabel']=clusters"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 32,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>Avg_Credit_Limit</th>\n",
+       "      <th>Total_Credit_Cards</th>\n",
+       "      <th>Total_visits_bank</th>\n",
+       "      <th>Total_visits_online</th>\n",
+       "      <th>Total_calls_made</th>\n",
+       "      <th>HierarchicalClusteringLabel</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <td>0</td>\n",
+       "      <td>1.740187</td>\n",
+       "      <td>-1.249225</td>\n",
+       "      <td>-0.860451</td>\n",
+       "      <td>-0.547490</td>\n",
+       "      <td>-1.251537</td>\n",
+       "      <td>3</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>1</td>\n",
+       "      <td>0.410293</td>\n",
+       "      <td>-0.787585</td>\n",
+       "      <td>-1.473731</td>\n",
+       "      <td>2.520519</td>\n",
+       "      <td>1.891859</td>\n",
+       "      <td>2</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>2</td>\n",
+       "      <td>0.410293</td>\n",
+       "      <td>1.058973</td>\n",
+       "      <td>-0.860451</td>\n",
+       "      <td>0.134290</td>\n",
+       "      <td>0.145528</td>\n",
+       "      <td>3</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   Avg_Credit_Limit  Total_Credit_Cards  Total_visits_bank  \\\n",
+       "0          1.740187           -1.249225          -0.860451   \n",
+       "1          0.410293           -0.787585          -1.473731   \n",
+       "2          0.410293            1.058973          -0.860451   \n",
+       "\n",
+       "   Total_visits_online  Total_calls_made  HierarchicalClusteringLabel  \n",
+       "0            -0.547490         -1.251537                            3  \n",
+       "1             2.520519          1.891859                            2  \n",
+       "2             0.134290          0.145528                            3  "
+      ]
+     },
+     "execution_count": 32,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "dataset2.head(3)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Analyse the clusters "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 33,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1440x1080 with 8 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "dataset2.boxplot(by = 'HierarchicalClusteringLabel',  layout=(2,4), figsize=(20, 15))\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Here also we observe differentiated clusters."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Silhouette Score"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 34,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "0.5157182558881063"
+      ]
+     },
+     "execution_count": 34,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "from sklearn.metrics import silhouette_score\n",
+    "silhouette_score(dataset.drop('KmeansLabel',axis=1),dataset['KmeansLabel'])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 35,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "0.5026435522438492"
+      ]
+     },
+     "execution_count": 35,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "from sklearn.metrics import silhouette_score\n",
+    "silhouette_score(dataset2.drop('HierarchicalClusteringLabel',axis=1),dataset2['HierarchicalClusteringLabel'])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Silhouette Score is better when closer 1 and worse when closer to -1\n",
+    "\n",
+    "Here Kmeans score is slightly better tha Hierarchical"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Comparing Kmeans and Hierarchical Results"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 36,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>Avg_Credit_Limit</th>\n",
+       "      <th>Total_Credit_Cards</th>\n",
+       "      <th>Total_visits_bank</th>\n",
+       "      <th>Total_visits_online</th>\n",
+       "      <th>Total_calls_made</th>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>KmeansLabel</th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <td>0</td>\n",
+       "      <td>-0.595796</td>\n",
+       "      <td>-1.059623</td>\n",
+       "      <td>-0.901518</td>\n",
+       "      <td>0.322997</td>\n",
+       "      <td>1.148109</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>1</td>\n",
+       "      <td>-0.021062</td>\n",
+       "      <td>0.373690</td>\n",
+       "      <td>0.666395</td>\n",
+       "      <td>-0.553672</td>\n",
+       "      <td>-0.553005</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>2</td>\n",
+       "      <td>2.831764</td>\n",
+       "      <td>1.862226</td>\n",
+       "      <td>-1.105763</td>\n",
+       "      <td>2.827319</td>\n",
+       "      <td>-0.874330</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "             Avg_Credit_Limit  Total_Credit_Cards  Total_visits_bank  \\\n",
+       "KmeansLabel                                                            \n",
+       "0                   -0.595796           -1.059623          -0.901518   \n",
+       "1                   -0.021062            0.373690           0.666395   \n",
+       "2                    2.831764            1.862226          -1.105763   \n",
+       "\n",
+       "             Total_visits_online  Total_calls_made  \n",
+       "KmeansLabel                                         \n",
+       "0                       0.322997          1.148109  \n",
+       "1                      -0.553672         -0.553005  \n",
+       "2                       2.827319         -0.874330  "
+      ]
+     },
+     "execution_count": 36,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "Kmeans_results=dataset.groupby('KmeansLabel').mean()\n",
+    "Kmeans_results"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 37,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>Avg_Credit_Limit</th>\n",
+       "      <th>Total_Credit_Cards</th>\n",
+       "      <th>Total_visits_bank</th>\n",
+       "      <th>Total_visits_online</th>\n",
+       "      <th>Total_calls_made</th>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>HierarchicalClusteringLabel</th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <td>1</td>\n",
+       "      <td>2.831764</td>\n",
+       "      <td>1.862226</td>\n",
+       "      <td>-1.105763</td>\n",
+       "      <td>2.827319</td>\n",
+       "      <td>-0.874330</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>2</td>\n",
+       "      <td>-0.594183</td>\n",
+       "      <td>-1.086676</td>\n",
+       "      <td>-0.886364</td>\n",
+       "      <td>0.350347</td>\n",
+       "      <td>1.196606</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>3</td>\n",
+       "      <td>-0.037852</td>\n",
+       "      <td>0.348490</td>\n",
+       "      <td>0.614820</td>\n",
+       "      <td>-0.544055</td>\n",
+       "      <td>-0.531891</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                             Avg_Credit_Limit  Total_Credit_Cards  \\\n",
+       "HierarchicalClusteringLabel                                         \n",
+       "1                                    2.831764            1.862226   \n",
+       "2                                   -0.594183           -1.086676   \n",
+       "3                                   -0.037852            0.348490   \n",
+       "\n",
+       "                             Total_visits_bank  Total_visits_online  \\\n",
+       "HierarchicalClusteringLabel                                           \n",
+       "1                                    -1.105763             2.827319   \n",
+       "2                                    -0.886364             0.350347   \n",
+       "3                                     0.614820            -0.544055   \n",
+       "\n",
+       "                             Total_calls_made  \n",
+       "HierarchicalClusteringLabel                    \n",
+       "1                                   -0.874330  \n",
+       "2                                    1.196606  \n",
+       "3                                   -0.531891  "
+      ]
+     },
+     "execution_count": 37,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "Hierarchical_results=dataset2.groupby('HierarchicalClusteringLabel').mean()\n",
+    "Hierarchical_results"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Carefully observing the above results we can say that : \n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "\n",
+    "Cluster 0 of Kmeans appears similar to Cluster 2 of Hierarchical \n",
+    "\n",
+    "\n",
+    "Cluster 1 of Kmeans appears similar to Cluster 3 of Hierarchical \n",
+    "\n",
+    "\n",
+    "Cluster 2 of Kmeans appears similar to Cluster 1 of Hierarchical \n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Let's rename "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "\n",
+    "Cluster 0 of Kmeans  and Cluster 2 of Hierarchical as G1\n",
+    "\n",
+    "Cluster 1 of Kmeans  and Cluster 3 of Hierarchical as G2\n",
+    "\n",
+    "Cluster 2 of Kmeans  and Cluster 1 of Hierarchical as G3\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 38,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>Avg_Credit_Limit</th>\n",
+       "      <th>Total_Credit_Cards</th>\n",
+       "      <th>Total_visits_bank</th>\n",
+       "      <th>Total_visits_online</th>\n",
+       "      <th>Total_calls_made</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <td>G1</td>\n",
+       "      <td>-0.595796</td>\n",
+       "      <td>-1.059623</td>\n",
+       "      <td>-0.901518</td>\n",
+       "      <td>0.322997</td>\n",
+       "      <td>1.148109</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>G2</td>\n",
+       "      <td>-0.021062</td>\n",
+       "      <td>0.373690</td>\n",
+       "      <td>0.666395</td>\n",
+       "      <td>-0.553672</td>\n",
+       "      <td>-0.553005</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>G3</td>\n",
+       "      <td>2.831764</td>\n",
+       "      <td>1.862226</td>\n",
+       "      <td>-1.105763</td>\n",
+       "      <td>2.827319</td>\n",
+       "      <td>-0.874330</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "    Avg_Credit_Limit  Total_Credit_Cards  Total_visits_bank  \\\n",
+       "G1         -0.595796           -1.059623          -0.901518   \n",
+       "G2         -0.021062            0.373690           0.666395   \n",
+       "G3          2.831764            1.862226          -1.105763   \n",
+       "\n",
+       "    Total_visits_online  Total_calls_made  \n",
+       "G1             0.322997          1.148109  \n",
+       "G2            -0.553672         -0.553005  \n",
+       "G3             2.827319         -0.874330  "
+      ]
+     },
+     "execution_count": 38,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "Kmeans_results.index=['G1','G2','G3']\n",
+    "Kmeans_results"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 39,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>Avg_Credit_Limit</th>\n",
+       "      <th>Total_Credit_Cards</th>\n",
+       "      <th>Total_visits_bank</th>\n",
+       "      <th>Total_visits_online</th>\n",
+       "      <th>Total_calls_made</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <td>G1</td>\n",
+       "      <td>-0.594183</td>\n",
+       "      <td>-1.086676</td>\n",
+       "      <td>-0.886364</td>\n",
+       "      <td>0.350347</td>\n",
+       "      <td>1.196606</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>G2</td>\n",
+       "      <td>-0.037852</td>\n",
+       "      <td>0.348490</td>\n",
+       "      <td>0.614820</td>\n",
+       "      <td>-0.544055</td>\n",
+       "      <td>-0.531891</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>G3</td>\n",
+       "      <td>2.831764</td>\n",
+       "      <td>1.862226</td>\n",
+       "      <td>-1.105763</td>\n",
+       "      <td>2.827319</td>\n",
+       "      <td>-0.874330</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "    Avg_Credit_Limit  Total_Credit_Cards  Total_visits_bank  \\\n",
+       "G1         -0.594183           -1.086676          -0.886364   \n",
+       "G2         -0.037852            0.348490           0.614820   \n",
+       "G3          2.831764            1.862226          -1.105763   \n",
+       "\n",
+       "    Total_visits_online  Total_calls_made  \n",
+       "G1             0.350347          1.196606  \n",
+       "G2            -0.544055         -0.531891  \n",
+       "G3             2.827319         -0.874330  "
+      ]
+     },
+     "execution_count": 39,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "Hierarchical_results.index=['G3','G1','G2']\n",
+    "Hierarchical_results.sort_index(inplace=True)\n",
+    "Hierarchical_results"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 40,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.axes._subplots.AxesSubplot at 0x1a1bc0ba90>"
+      ]
+     },
+     "execution_count": 40,
+     "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": [
+    "Kmeans_results.plot.bar()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 41,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.axes._subplots.AxesSubplot at 0x1a1a13c590>"
+      ]
+     },
+     "execution_count": 41,
+     "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": [
+    "Hierarchical_results.plot.bar()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### By both the methods of Clustering we get comparable clusters"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Cluster Profiles and Marketing Recommendation"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Since both the clustering alogrithms are giving similar clusters so we can assign labels from any one of the algorithm to the original (non scaled) data  to analyse clusters profiles\n",
+    "( here we are assigning labels of Kmeans , same could be done using hierarchical labels) "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 42,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "subset['KmeansLabel']=dataset['KmeansLabel']\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Understanding each feature characterstics within different clusters "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 43,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Avg_Credit_Limit\n",
+      "             count      mean      min       max\n",
+      "KmeansLabel                                    \n",
+      "0            224.0   12174.0   3000.0   50000.0\n",
+      "1            386.0   33782.0   5000.0  100000.0\n",
+      "2             50.0  141040.0  84000.0  200000.0\n",
+      "\n",
+      "\n",
+      "\n",
+      "Total_Credit_Cards\n",
+      "             count  mean  min   max\n",
+      "KmeansLabel                        \n",
+      "0            224.0   2.0  1.0   4.0\n",
+      "1            386.0   6.0  2.0   7.0\n",
+      "2             50.0   9.0  5.0  10.0\n",
+      "\n",
+      "\n",
+      "\n",
+      "Total_visits_bank\n",
+      "             count  mean  min  max\n",
+      "KmeansLabel                       \n",
+      "0            224.0   1.0  0.0  2.0\n",
+      "1            386.0   3.0  1.0  5.0\n",
+      "2             50.0   1.0  0.0  1.0\n",
+      "\n",
+      "\n",
+      "\n",
+      "Total_visits_online\n",
+      "             count  mean  min   max\n",
+      "KmeansLabel                        \n",
+      "0            224.0   4.0  1.0  10.0\n",
+      "1            386.0   1.0  0.0   3.0\n",
+      "2             50.0  11.0  6.0  15.0\n",
+      "\n",
+      "\n",
+      "\n",
+      "Total_calls_made\n",
+      "             count  mean  min   max\n",
+      "KmeansLabel                        \n",
+      "0            224.0   7.0  1.0  10.0\n",
+      "1            386.0   2.0  0.0   4.0\n",
+      "2             50.0   1.0  0.0   3.0\n",
+      "\n",
+      "\n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "for each in cols_to_consider:\n",
+    "    print (each)\n",
+    "    print ( subset.groupby('KmeansLabel').describe().round()[each][['count','mean','min','max']])\n",
+    "    \n",
+    "    print (\"\\n\\n\")\n",
+    "    \n",
+    "    "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Analysis of clusters and questions answered :\n",
+    "    "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### 1. How many different segments of customers are there? \n",
+    "\n",
+    "Answer : Total numbers of segments are 3\n",
+    "    \n",
+    "    \n",
+    "  "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### 2. How are these segments different from each other? (Cluster profiles )\n",
+    "  \n",
+    "\n",
+    "Answer: \n",
+    "    \n",
+    "\n",
+    "**Label 0 can be considered low valued customers**\n",
+    "   \n",
+    "    This group comprises of about 34% of the customers ( 224/660 )\n",
+    "    \n",
+    "    These customers have a mean \"Avg_Credit_Limit \" around 12200 and have 2 credit card on an average and the maximum number of credit card as 4.\n",
+    "    \n",
+    "    They are the ones who makes the most number of customer care calls to the bank as the average calls made is 7 \n",
+    "\n",
+    "\n",
+    "\n",
+    "**Label 1 can be considered medium valued customers** \n",
+    "    \n",
+    "    This group forms the majority of the customers having about 58% customers in total  ( 386/660 )\n",
+    "    \n",
+    "    These customers have  \"Avg_Credit_Limit \" ranging from 5000.0 to 100000.0 \n",
+    "    \n",
+    "    These are the ones which make the maximum number of visits to the bank as the average visits to bank is 3.\n",
+    "    \n",
+    "    They are the ones who are least active online as the maximum visit onine is just 3\n",
+    "\n",
+    "\n",
+    "\n",
+    "**Label 2 can be considered  high value customers** \n",
+    "    \n",
+    "    These are the least in number i.e. only 50 customers comprising 7.5% of total customers (50/660) .\n",
+    "    \n",
+    "    These customers have a minimum \"Avg_Credit_Limit \" of 84000 and have atleast 5 Credit cards .\n",
+    "    \n",
+    "    These are the ones which make the minimum number of visits to the bank as the maximum visit to bank is 1 amongst all 50     customers.\n",
+    "    \n",
+    "    They are mostly using online services as the average visit online is 11. \n",
+    "\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### 3. What are your recommendations to the bank on how to better market to and service these customers? (Business Recommendations )"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "1. Customers in the medium group ( having Label 1 ) are not engaged much in online activities , one of the exercise can be to engage them online. If they join online , promotions and offers can be communicated to them with much ease.\n",
+    "\n",
+    "\n",
+    "\n",
+    "2. Customers in low group ( label 0 ) can further be binned to check if there are any extreme groups having high average credit limit.These customers can be given more offers and new credit cards so that we can have them in medium group (label 1 )  over a period of time. Similarly we can perform this for medium customers (label 1)  and try to have them in high group (label 2) over a period of time .\n",
+    "\n",
+    "\n",
+    "\n",
+    "3. Customers in low group ( label 0 ) make the most number of customer care calls, these customers can be told about different offers to try and move them to  medium group over a period of time .\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.7.4"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}