2.Customer Segmentation Report.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Customer Segmentation Report"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The main bulk of our analysis will come in this part of the project. Here, we use unsupervised learning techniques to describe the relationship between the demographics of the company's existing customers and the general population of Germany. By the end of this part, we should be able to describe parts of the general population that are more likely to be part of the mail-order company's main customer base, and which parts of the general population are less so."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import Libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T05:09:21.314931Z",
     "start_time": "2019-08-12T05:09:14.381390Z"
    }
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "from joblib import dump, load\n",
    "\n",
    "import missingno as msno\n",
    "\n",
    "from time import time\n",
    "from collections import OrderedDict, Counter\n",
    "\n",
    "from tqdm.auto import tqdm, trange\n",
    "tqdm.pandas()\n",
    "\n",
    "from utils.transformers import *\n",
    "from utils.functions import *\n",
    "\n",
    "import pyarrow\n",
    "\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.cluster import KMeans\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "sns.set(style=\"darkgrid\")\n",
    "\n",
    "# magic word for producing visualizations in notebook\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load the Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T05:11:34.812896Z",
     "start_time": "2019-08-12T05:11:30.671833Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\radue\\Anaconda3\\envs\\udacity_ml_capstone\\lib\\site-packages\\pyarrow\\pandas_compat.py:708: FutureWarning: .labels was deprecated in version 0.24.0. Use .codes instead.\n",
      "  labels = getattr(columns, 'labels', None) or [\n",
      "C:\\Users\\radue\\Anaconda3\\envs\\udacity_ml_capstone\\lib\\site-packages\\pyarrow\\pandas_compat.py:735: FutureWarning: the 'labels' keyword is deprecated, use 'codes' instead\n",
      "  return pd.MultiIndex(levels=new_levels, labels=labels, names=columns.names)\n",
      "C:\\Users\\radue\\Anaconda3\\envs\\udacity_ml_capstone\\lib\\site-packages\\pyarrow\\pandas_compat.py:752: FutureWarning: .labels was deprecated in version 0.24.0. Use .codes instead.\n",
      "  labels, = index.labels\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ANREDE_KZ_1</th>\n",
       "      <th>ANREDE_KZ_2</th>\n",
       "      <th>CAMEO_DEUG_2015_1</th>\n",
       "      <th>CAMEO_DEUG_2015_2</th>\n",
       "      <th>CAMEO_DEUG_2015_3</th>\n",
       "      <th>CAMEO_DEUG_2015_4</th>\n",
       "      <th>CAMEO_DEUG_2015_5</th>\n",
       "      <th>CAMEO_DEUG_2015_6</th>\n",
       "      <th>CAMEO_DEUG_2015_7</th>\n",
       "      <th>CAMEO_DEUG_2015_8</th>\n",
       "      <th>...</th>\n",
       "      <th>MIN_GEBAEUDEJAHR</th>\n",
       "      <th>ANZ_KINDER</th>\n",
       "      <th>ANZ_STATISTISCHE_HAUSHALTE</th>\n",
       "      <th>EINGEZOGENAM_HH_JAHR</th>\n",
       "      <th>EXTSEL992</th>\n",
       "      <th>VERDICHTUNGSRAUM</th>\n",
       "      <th>VK_DHT4A</th>\n",
       "      <th>VK_DISTANZ</th>\n",
       "      <th>VK_ZG11</th>\n",
       "      <th>LNR</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.359845</td>\n",
       "      <td>-0.292527</td>\n",
       "      <td>-0.302057</td>\n",
       "      <td>-0.098898</td>\n",
       "      <td>0.023455</td>\n",
       "      <td>-0.395656</td>\n",
       "      <td>-0.000407</td>\n",
       "      <td>0.137664</td>\n",
       "      <td>0.018645</td>\n",
       "      <td>910215</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.359845</td>\n",
       "      <td>-0.292527</td>\n",
       "      <td>0.357993</td>\n",
       "      <td>0.048946</td>\n",
       "      <td>0.023455</td>\n",
       "      <td>-0.519516</td>\n",
       "      <td>0.731728</td>\n",
       "      <td>1.102566</td>\n",
       "      <td>1.527603</td>\n",
       "      <td>910220</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.359845</td>\n",
       "      <td>-0.292527</td>\n",
       "      <td>-0.008701</td>\n",
       "      <td>-0.542432</td>\n",
       "      <td>-2.641994</td>\n",
       "      <td>-0.395656</td>\n",
       "      <td>1.097796</td>\n",
       "      <td>0.459298</td>\n",
       "      <td>0.018645</td>\n",
       "      <td>910225</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1.213535</td>\n",
       "      <td>-0.292527</td>\n",
       "      <td>-0.375395</td>\n",
       "      <td>-0.838120</td>\n",
       "      <td>-0.376363</td>\n",
       "      <td>-0.519516</td>\n",
       "      <td>0.365661</td>\n",
       "      <td>0.780932</td>\n",
       "      <td>1.904842</td>\n",
       "      <td>910226</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.359845</td>\n",
       "      <td>-0.292527</td>\n",
       "      <td>-0.302057</td>\n",
       "      <td>0.048946</td>\n",
       "      <td>0.023455</td>\n",
       "      <td>-0.395656</td>\n",
       "      <td>-1.098609</td>\n",
       "      <td>-0.827238</td>\n",
       "      <td>-0.735833</td>\n",
       "      <td>910241</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 686 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   ANREDE_KZ_1  ANREDE_KZ_2  CAMEO_DEUG_2015_1  CAMEO_DEUG_2015_2  \\\n",
       "0            1            0                  0                  0   \n",
       "1            0            1                  0                  0   \n",
       "2            0            1                  0                  0   \n",
       "3            0            1                  0                  1   \n",
       "4            1            0                  0                  0   \n",
       "\n",
       "   CAMEO_DEUG_2015_3  CAMEO_DEUG_2015_4  CAMEO_DEUG_2015_5  CAMEO_DEUG_2015_6  \\\n",
       "0                  0                  0                  0                  0   \n",
       "1                  0                  0                  0                  0   \n",
       "2                  0                  1                  0                  0   \n",
       "3                  0                  0                  0                  0   \n",
       "4                  0                  0                  0                  1   \n",
       "\n",
       "   CAMEO_DEUG_2015_7  CAMEO_DEUG_2015_8  ...  MIN_GEBAEUDEJAHR  ANZ_KINDER  \\\n",
       "0                  0                  1  ...         -0.359845   -0.292527   \n",
       "1                  0                  1  ...         -0.359845   -0.292527   \n",
       "2                  0                  0  ...         -0.359845   -0.292527   \n",
       "3                  0                  0  ...          1.213535   -0.292527   \n",
       "4                  0                  0  ...         -0.359845   -0.292527   \n",
       "\n",
       "   ANZ_STATISTISCHE_HAUSHALTE  EINGEZOGENAM_HH_JAHR  EXTSEL992  \\\n",
       "0                   -0.302057             -0.098898   0.023455   \n",
       "1                    0.357993              0.048946   0.023455   \n",
       "2                   -0.008701             -0.542432  -2.641994   \n",
       "3                   -0.375395             -0.838120  -0.376363   \n",
       "4                   -0.302057              0.048946   0.023455   \n",
       "\n",
       "   VERDICHTUNGSRAUM  VK_DHT4A  VK_DISTANZ   VK_ZG11     LNR  \n",
       "0         -0.395656 -0.000407    0.137664  0.018645  910215  \n",
       "1         -0.519516  0.731728    1.102566  1.527603  910220  \n",
       "2         -0.395656  1.097796    0.459298  0.018645  910225  \n",
       "3         -0.519516  0.365661    0.780932  1.904842  910226  \n",
       "4         -0.395656 -1.098609   -0.827238 -0.735833  910241  \n",
       "\n",
       "[5 rows x 686 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "azdias = pd.read_parquet('temp/azdias_processed.gzip')\n",
    "azdias.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T05:12:06.422103Z",
     "start_time": "2019-08-12T05:12:04.738510Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 891221 entries, 0 to 891220\n",
      "Columns: 685 entries, ANREDE_KZ_1 to VK_ZG11\n",
      "dtypes: float32(15), uint8(670)\n",
      "memory usage: 627.3 MB\n"
     ]
    }
   ],
   "source": [
    "azdias = azdias.drop('LNR', axis=1)\n",
    "azdias.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Dimensionality Reduction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Investigate the variance accounted for by each principal component."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Utility Functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T05:25:06.390586Z",
     "start_time": "2019-08-12T05:25:06.381611Z"
    }
   },
   "outputs": [],
   "source": [
    "def pca_results(full_dataset, pca, show_plot=True):\n",
    "    '''\n",
    "    Create a DataFrame of the PCA results\n",
    "    Includes dimension feature weights and explained variance\n",
    "    Visualizes the PCA results\n",
    "    '''\n",
    "\n",
    "    # Dimension indexing\n",
    "    dimensions = ['Dimension {}'.format(i) for i in range(1,len(pca.components_)+1)]\n",
    "\n",
    "    # PCA components\n",
    "    components = pd.DataFrame(np.round(pca.components_, 4), columns = full_dataset.keys())\n",
    "    components.index = dimensions\n",
    "\n",
    "    # PCA explained variance\n",
    "    ratios = pca.explained_variance_ratio_.reshape(len(pca.components_), 1)\n",
    "    variance_ratios = pd.DataFrame(np.round(ratios, 4), columns = ['Explained Variance Ratio'])\n",
    "    variance_ratios.index = dimensions\n",
    "    \n",
    "    # PCA explained cumulative variance\n",
    "    cumsum = pca.explained_variance_ratio_.cumsum().reshape(len(pca.components_), 1)\n",
    "    variance_cumsum = pd.DataFrame(np.round(cumsum, 4), columns = ['Explained Cumulative Variance'])\n",
    "    variance_cumsum.index = dimensions\n",
    "\n",
    "    if show_plot == True:\n",
    "        # Create a bar plot visualization\n",
    "        fig, ax = plt.subplots(figsize = (14,8))\n",
    "        ax.plot(np.arange(len(variance_cumsum)), variance_cumsum)\n",
    "        ax.set_ylabel(\"Explained Cumulative Variance\")\n",
    "        ax.set_xlabel(\"Number of Principal Components\")\n",
    "\n",
    "    # Return a concatenated DataFrame\n",
    "    return pd.concat([variance_cumsum, variance_ratios, components], axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T05:28:22.870756Z",
     "start_time": "2019-08-12T05:28:22.864772Z"
    }
   },
   "outputs": [],
   "source": [
    "# Map weights for the first principal component to corresponding feature names\n",
    "# and then print the linked values, sorted by weight.\n",
    "def print_pcs(df, pca, comp, k=5):\n",
    "    components = pd.DataFrame(np.round(pca.components_, 4), columns = df.columns)\n",
    "    pc = components.iloc[comp-1].sort_values(ascending=False)\n",
    "    print(f'Weights for PC{comp}')\n",
    "    print(f'Top {k} weights')\n",
    "    print(pc.head(k))\n",
    "    print('\\n')\n",
    "    print(f'Bottom {k} weights')\n",
    "    print(pc.tail(k))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T10:01:22.819167Z",
     "start_time": "2019-08-12T10:01:22.812182Z"
    }
   },
   "outputs": [],
   "source": [
    "def comp_weights(pca, dimension_max, dimension_min , num_features, df):\n",
    "    components = pd.DataFrame( np.round(pca.components_, 4), columns = df.keys())\n",
    "    components.index = [i for i in range(1, len(pca.components_)+1)]\n",
    "\n",
    "    weights_pos = components.iloc[dimension_max-1]\n",
    "    weights_neg = components.iloc[dimension_min-1]\n",
    "    fig, ax = plt.subplots(figsize = (15, 4), ncols=2, nrows = 1)\n",
    "    weights_pos.sort_values(ascending=False)[:num_features].plot(kind='bar', ax = ax[0]);\n",
    "    weights_neg.sort_values(ascending=False)[-num_features:].plot(kind='bar', ax = ax[1], color = 'red');\n",
    "    \n",
    "    return fig"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### PCA"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Apply PCA to the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T05:23:38.416921Z",
     "start_time": "2019-08-12T05:22:32.668683Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "PCA(copy=True, iterated_power='auto', n_components=None, random_state=777,\n",
       "    svd_solver='auto', tol=0.0, whiten=False)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pca = PCA(random_state=777)\n",
    "pca.fit(azdias)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T05:25:09.226389Z",
     "start_time": "2019-08-12T05:25:08.670397Z"
    }
   },
   "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>Explained Cumulative Variance</th>\n",
       "      <th>Explained Variance Ratio</th>\n",
       "      <th>ANREDE_KZ_1</th>\n",
       "      <th>ANREDE_KZ_2</th>\n",
       "      <th>CAMEO_DEUG_2015_1</th>\n",
       "      <th>CAMEO_DEUG_2015_2</th>\n",
       "      <th>CAMEO_DEUG_2015_3</th>\n",
       "      <th>CAMEO_DEUG_2015_4</th>\n",
       "      <th>CAMEO_DEUG_2015_5</th>\n",
       "      <th>CAMEO_DEUG_2015_6</th>\n",
       "      <th>...</th>\n",
       "      <th>KBA13_ANZAHL_PKW</th>\n",
       "      <th>MIN_GEBAEUDEJAHR</th>\n",
       "      <th>ANZ_KINDER</th>\n",
       "      <th>ANZ_STATISTISCHE_HAUSHALTE</th>\n",
       "      <th>EINGEZOGENAM_HH_JAHR</th>\n",
       "      <th>EXTSEL992</th>\n",
       "      <th>VERDICHTUNGSRAUM</th>\n",
       "      <th>VK_DHT4A</th>\n",
       "      <th>VK_DISTANZ</th>\n",
       "      <th>VK_ZG11</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Dimension 1</th>\n",
       "      <td>0.1488</td>\n",
       "      <td>0.1488</td>\n",
       "      <td>0.0003</td>\n",
       "      <td>-0.0003</td>\n",
       "      <td>0.0018</td>\n",
       "      <td>0.0052</td>\n",
       "      <td>0.0041</td>\n",
       "      <td>0.0049</td>\n",
       "      <td>0.0013</td>\n",
       "      <td>0.0004</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0114</td>\n",
       "      <td>0.0196</td>\n",
       "      <td>0.0330</td>\n",
       "      <td>-0.0177</td>\n",
       "      <td>0.0082</td>\n",
       "      <td>0.0080</td>\n",
       "      <td>-0.0029</td>\n",
       "      <td>-0.0830</td>\n",
       "      <td>-0.0716</td>\n",
       "      <td>-0.0498</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 2</th>\n",
       "      <td>0.2154</td>\n",
       "      <td>0.0666</td>\n",
       "      <td>0.0069</td>\n",
       "      <td>-0.0069</td>\n",
       "      <td>-0.0048</td>\n",
       "      <td>-0.0112</td>\n",
       "      <td>-0.0088</td>\n",
       "      <td>-0.0103</td>\n",
       "      <td>-0.0019</td>\n",
       "      <td>-0.0033</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0350</td>\n",
       "      <td>-0.0079</td>\n",
       "      <td>0.0143</td>\n",
       "      <td>0.0383</td>\n",
       "      <td>0.0594</td>\n",
       "      <td>-0.0125</td>\n",
       "      <td>0.0179</td>\n",
       "      <td>-0.0009</td>\n",
       "      <td>0.0282</td>\n",
       "      <td>0.0493</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 3</th>\n",
       "      <td>0.2699</td>\n",
       "      <td>0.0544</td>\n",
       "      <td>-0.0206</td>\n",
       "      <td>0.0206</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0085</td>\n",
       "      <td>-0.0085</td>\n",
       "      <td>-0.0132</td>\n",
       "      <td>0.0003</td>\n",
       "      <td>0.0018</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0368</td>\n",
       "      <td>-0.0288</td>\n",
       "      <td>-0.0215</td>\n",
       "      <td>0.0650</td>\n",
       "      <td>-0.0029</td>\n",
       "      <td>0.0077</td>\n",
       "      <td>0.0407</td>\n",
       "      <td>0.0182</td>\n",
       "      <td>0.0040</td>\n",
       "      <td>-0.0025</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 4</th>\n",
       "      <td>0.3164</td>\n",
       "      <td>0.0465</td>\n",
       "      <td>0.0172</td>\n",
       "      <td>-0.0172</td>\n",
       "      <td>0.0081</td>\n",
       "      <td>0.0079</td>\n",
       "      <td>0.0014</td>\n",
       "      <td>0.0015</td>\n",
       "      <td>0.0059</td>\n",
       "      <td>-0.0025</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0399</td>\n",
       "      <td>-0.0150</td>\n",
       "      <td>0.0058</td>\n",
       "      <td>-0.0079</td>\n",
       "      <td>0.0051</td>\n",
       "      <td>-0.0088</td>\n",
       "      <td>0.0101</td>\n",
       "      <td>0.0013</td>\n",
       "      <td>0.0061</td>\n",
       "      <td>0.0092</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 5</th>\n",
       "      <td>0.3595</td>\n",
       "      <td>0.0431</td>\n",
       "      <td>0.0890</td>\n",
       "      <td>-0.0890</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0003</td>\n",
       "      <td>0.0006</td>\n",
       "      <td>-0.0003</td>\n",
       "      <td>0.0004</td>\n",
       "      <td>0.0040</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0105</td>\n",
       "      <td>-0.0005</td>\n",
       "      <td>-0.0104</td>\n",
       "      <td>0.0230</td>\n",
       "      <td>-0.0092</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0154</td>\n",
       "      <td>0.0108</td>\n",
       "      <td>-0.0181</td>\n",
       "      <td>-0.0467</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 6</th>\n",
       "      <td>0.3818</td>\n",
       "      <td>0.0222</td>\n",
       "      <td>0.0370</td>\n",
       "      <td>-0.0370</td>\n",
       "      <td>-0.0001</td>\n",
       "      <td>-0.0046</td>\n",
       "      <td>-0.0047</td>\n",
       "      <td>-0.0129</td>\n",
       "      <td>-0.0041</td>\n",
       "      <td>-0.0050</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0516</td>\n",
       "      <td>0.0217</td>\n",
       "      <td>-0.0367</td>\n",
       "      <td>-0.0148</td>\n",
       "      <td>-0.0196</td>\n",
       "      <td>0.0149</td>\n",
       "      <td>-0.0091</td>\n",
       "      <td>-0.0193</td>\n",
       "      <td>-0.0388</td>\n",
       "      <td>-0.0684</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 7</th>\n",
       "      <td>0.4023</td>\n",
       "      <td>0.0206</td>\n",
       "      <td>-0.0246</td>\n",
       "      <td>0.0246</td>\n",
       "      <td>0.0043</td>\n",
       "      <td>0.0059</td>\n",
       "      <td>0.0031</td>\n",
       "      <td>0.0020</td>\n",
       "      <td>0.0019</td>\n",
       "      <td>0.0061</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0535</td>\n",
       "      <td>-0.0020</td>\n",
       "      <td>0.0269</td>\n",
       "      <td>0.0279</td>\n",
       "      <td>0.0107</td>\n",
       "      <td>0.0012</td>\n",
       "      <td>0.0364</td>\n",
       "      <td>-0.0161</td>\n",
       "      <td>-0.0173</td>\n",
       "      <td>-0.0050</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 8</th>\n",
       "      <td>0.4199</td>\n",
       "      <td>0.0175</td>\n",
       "      <td>-0.0235</td>\n",
       "      <td>0.0235</td>\n",
       "      <td>-0.0055</td>\n",
       "      <td>-0.0085</td>\n",
       "      <td>0.0003</td>\n",
       "      <td>0.0008</td>\n",
       "      <td>0.0023</td>\n",
       "      <td>0.0053</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0681</td>\n",
       "      <td>0.0257</td>\n",
       "      <td>-0.0090</td>\n",
       "      <td>0.0530</td>\n",
       "      <td>0.0843</td>\n",
       "      <td>-0.0088</td>\n",
       "      <td>-0.0052</td>\n",
       "      <td>0.0648</td>\n",
       "      <td>0.0749</td>\n",
       "      <td>0.0837</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 9</th>\n",
       "      <td>0.4365</td>\n",
       "      <td>0.0167</td>\n",
       "      <td>-0.0279</td>\n",
       "      <td>0.0279</td>\n",
       "      <td>0.0043</td>\n",
       "      <td>0.0061</td>\n",
       "      <td>0.0028</td>\n",
       "      <td>0.0019</td>\n",
       "      <td>-0.0002</td>\n",
       "      <td>0.0048</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0025</td>\n",
       "      <td>-0.0144</td>\n",
       "      <td>-0.0077</td>\n",
       "      <td>-0.0248</td>\n",
       "      <td>0.0240</td>\n",
       "      <td>0.0078</td>\n",
       "      <td>0.0020</td>\n",
       "      <td>0.0071</td>\n",
       "      <td>0.0080</td>\n",
       "      <td>0.0116</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 10</th>\n",
       "      <td>0.4515</td>\n",
       "      <td>0.0150</td>\n",
       "      <td>0.0083</td>\n",
       "      <td>-0.0083</td>\n",
       "      <td>-0.0048</td>\n",
       "      <td>-0.0038</td>\n",
       "      <td>-0.0033</td>\n",
       "      <td>0.0051</td>\n",
       "      <td>-0.0009</td>\n",
       "      <td>-0.0076</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0820</td>\n",
       "      <td>-0.0164</td>\n",
       "      <td>-0.0050</td>\n",
       "      <td>0.0258</td>\n",
       "      <td>-0.0212</td>\n",
       "      <td>-0.0035</td>\n",
       "      <td>-0.0084</td>\n",
       "      <td>0.0131</td>\n",
       "      <td>0.0189</td>\n",
       "      <td>0.0174</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 11</th>\n",
       "      <td>0.4650</td>\n",
       "      <td>0.0135</td>\n",
       "      <td>-0.0091</td>\n",
       "      <td>0.0091</td>\n",
       "      <td>0.0002</td>\n",
       "      <td>-0.0016</td>\n",
       "      <td>-0.0010</td>\n",
       "      <td>-0.0028</td>\n",
       "      <td>-0.0002</td>\n",
       "      <td>-0.0047</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0273</td>\n",
       "      <td>0.0232</td>\n",
       "      <td>0.0464</td>\n",
       "      <td>-0.0176</td>\n",
       "      <td>-0.0197</td>\n",
       "      <td>0.0707</td>\n",
       "      <td>-0.0047</td>\n",
       "      <td>-0.1182</td>\n",
       "      <td>-0.1343</td>\n",
       "      <td>-0.1209</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 12</th>\n",
       "      <td>0.4781</td>\n",
       "      <td>0.0131</td>\n",
       "      <td>-0.0011</td>\n",
       "      <td>0.0011</td>\n",
       "      <td>0.0037</td>\n",
       "      <td>-0.0028</td>\n",
       "      <td>-0.0036</td>\n",
       "      <td>-0.0084</td>\n",
       "      <td>0.0020</td>\n",
       "      <td>0.0008</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0372</td>\n",
       "      <td>-0.0538</td>\n",
       "      <td>-0.0291</td>\n",
       "      <td>-0.0303</td>\n",
       "      <td>-0.0132</td>\n",
       "      <td>-0.0068</td>\n",
       "      <td>0.0179</td>\n",
       "      <td>0.0194</td>\n",
       "      <td>0.0091</td>\n",
       "      <td>-0.0066</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 13</th>\n",
       "      <td>0.4898</td>\n",
       "      <td>0.0118</td>\n",
       "      <td>0.0038</td>\n",
       "      <td>-0.0038</td>\n",
       "      <td>-0.0004</td>\n",
       "      <td>-0.0094</td>\n",
       "      <td>-0.0062</td>\n",
       "      <td>-0.0063</td>\n",
       "      <td>-0.0016</td>\n",
       "      <td>0.0091</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0108</td>\n",
       "      <td>-0.0149</td>\n",
       "      <td>0.0067</td>\n",
       "      <td>0.0620</td>\n",
       "      <td>0.0538</td>\n",
       "      <td>0.0063</td>\n",
       "      <td>0.0047</td>\n",
       "      <td>-0.0075</td>\n",
       "      <td>-0.0046</td>\n",
       "      <td>0.0073</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 14</th>\n",
       "      <td>0.5009</td>\n",
       "      <td>0.0110</td>\n",
       "      <td>0.0006</td>\n",
       "      <td>-0.0006</td>\n",
       "      <td>-0.0004</td>\n",
       "      <td>0.0008</td>\n",
       "      <td>0.0016</td>\n",
       "      <td>0.0008</td>\n",
       "      <td>0.0013</td>\n",
       "      <td>-0.0010</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0165</td>\n",
       "      <td>0.0054</td>\n",
       "      <td>-0.0196</td>\n",
       "      <td>-0.0154</td>\n",
       "      <td>0.0669</td>\n",
       "      <td>-0.0273</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>0.0206</td>\n",
       "      <td>0.0355</td>\n",
       "      <td>0.0448</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 15</th>\n",
       "      <td>0.5116</td>\n",
       "      <td>0.0107</td>\n",
       "      <td>-0.0019</td>\n",
       "      <td>0.0019</td>\n",
       "      <td>0.0005</td>\n",
       "      <td>-0.0049</td>\n",
       "      <td>-0.0025</td>\n",
       "      <td>-0.0016</td>\n",
       "      <td>0.0064</td>\n",
       "      <td>-0.0008</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0533</td>\n",
       "      <td>0.0109</td>\n",
       "      <td>-0.0205</td>\n",
       "      <td>0.0457</td>\n",
       "      <td>0.1119</td>\n",
       "      <td>0.0212</td>\n",
       "      <td>0.0454</td>\n",
       "      <td>0.0014</td>\n",
       "      <td>0.0031</td>\n",
       "      <td>0.0075</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 16</th>\n",
       "      <td>0.5214</td>\n",
       "      <td>0.0098</td>\n",
       "      <td>-0.0027</td>\n",
       "      <td>0.0027</td>\n",
       "      <td>0.0019</td>\n",
       "      <td>-0.0005</td>\n",
       "      <td>0.0007</td>\n",
       "      <td>0.0032</td>\n",
       "      <td>0.0039</td>\n",
       "      <td>-0.0020</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.2428</td>\n",
       "      <td>0.0244</td>\n",
       "      <td>0.0043</td>\n",
       "      <td>-0.0200</td>\n",
       "      <td>0.0613</td>\n",
       "      <td>-0.0110</td>\n",
       "      <td>-0.0140</td>\n",
       "      <td>0.0119</td>\n",
       "      <td>0.0269</td>\n",
       "      <td>0.0368</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 17</th>\n",
       "      <td>0.5310</td>\n",
       "      <td>0.0096</td>\n",
       "      <td>-0.0109</td>\n",
       "      <td>0.0109</td>\n",
       "      <td>0.0026</td>\n",
       "      <td>-0.0013</td>\n",
       "      <td>-0.0041</td>\n",
       "      <td>-0.0004</td>\n",
       "      <td>0.0011</td>\n",
       "      <td>0.0010</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0122</td>\n",
       "      <td>-0.1091</td>\n",
       "      <td>0.0437</td>\n",
       "      <td>0.0115</td>\n",
       "      <td>-0.2802</td>\n",
       "      <td>-0.0133</td>\n",
       "      <td>0.0450</td>\n",
       "      <td>0.0043</td>\n",
       "      <td>-0.0214</td>\n",
       "      <td>-0.0393</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 18</th>\n",
       "      <td>0.5398</td>\n",
       "      <td>0.0088</td>\n",
       "      <td>-0.0025</td>\n",
       "      <td>0.0025</td>\n",
       "      <td>-0.0038</td>\n",
       "      <td>-0.0074</td>\n",
       "      <td>-0.0049</td>\n",
       "      <td>-0.0028</td>\n",
       "      <td>-0.0038</td>\n",
       "      <td>0.0084</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0280</td>\n",
       "      <td>-0.1746</td>\n",
       "      <td>-0.0765</td>\n",
       "      <td>-0.0060</td>\n",
       "      <td>-0.0367</td>\n",
       "      <td>0.0279</td>\n",
       "      <td>0.0011</td>\n",
       "      <td>0.0243</td>\n",
       "      <td>0.0072</td>\n",
       "      <td>-0.0128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 19</th>\n",
       "      <td>0.5481</td>\n",
       "      <td>0.0083</td>\n",
       "      <td>0.0038</td>\n",
       "      <td>-0.0038</td>\n",
       "      <td>0.0046</td>\n",
       "      <td>0.0050</td>\n",
       "      <td>0.0028</td>\n",
       "      <td>-0.0038</td>\n",
       "      <td>0.0018</td>\n",
       "      <td>-0.0046</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0616</td>\n",
       "      <td>0.0428</td>\n",
       "      <td>-0.0289</td>\n",
       "      <td>0.0042</td>\n",
       "      <td>0.0812</td>\n",
       "      <td>0.0060</td>\n",
       "      <td>-0.0149</td>\n",
       "      <td>-0.0065</td>\n",
       "      <td>-0.0039</td>\n",
       "      <td>-0.0052</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 20</th>\n",
       "      <td>0.5563</td>\n",
       "      <td>0.0081</td>\n",
       "      <td>-0.0108</td>\n",
       "      <td>0.0108</td>\n",
       "      <td>0.0023</td>\n",
       "      <td>0.0043</td>\n",
       "      <td>0.0035</td>\n",
       "      <td>-0.0026</td>\n",
       "      <td>0.0026</td>\n",
       "      <td>0.0010</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0337</td>\n",
       "      <td>-0.0641</td>\n",
       "      <td>-0.0539</td>\n",
       "      <td>0.1230</td>\n",
       "      <td>-0.0546</td>\n",
       "      <td>0.0150</td>\n",
       "      <td>-0.1048</td>\n",
       "      <td>0.0206</td>\n",
       "      <td>-0.0052</td>\n",
       "      <td>-0.0294</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 21</th>\n",
       "      <td>0.5642</td>\n",
       "      <td>0.0079</td>\n",
       "      <td>-0.0059</td>\n",
       "      <td>0.0059</td>\n",
       "      <td>-0.0009</td>\n",
       "      <td>-0.0033</td>\n",
       "      <td>-0.0027</td>\n",
       "      <td>0.0005</td>\n",
       "      <td>-0.0024</td>\n",
       "      <td>0.0049</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0865</td>\n",
       "      <td>-0.0894</td>\n",
       "      <td>-0.0020</td>\n",
       "      <td>0.0026</td>\n",
       "      <td>-0.0872</td>\n",
       "      <td>0.0022</td>\n",
       "      <td>-0.0117</td>\n",
       "      <td>0.0109</td>\n",
       "      <td>0.0072</td>\n",
       "      <td>0.0013</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 22</th>\n",
       "      <td>0.5718</td>\n",
       "      <td>0.0076</td>\n",
       "      <td>-0.0027</td>\n",
       "      <td>0.0027</td>\n",
       "      <td>-0.0054</td>\n",
       "      <td>-0.0061</td>\n",
       "      <td>-0.0002</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0036</td>\n",
       "      <td>-0.0027</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.1209</td>\n",
       "      <td>0.0632</td>\n",
       "      <td>0.0526</td>\n",
       "      <td>0.1087</td>\n",
       "      <td>-0.0866</td>\n",
       "      <td>-0.0086</td>\n",
       "      <td>-0.0495</td>\n",
       "      <td>0.0119</td>\n",
       "      <td>0.0079</td>\n",
       "      <td>0.0064</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 23</th>\n",
       "      <td>0.5792</td>\n",
       "      <td>0.0074</td>\n",
       "      <td>0.0105</td>\n",
       "      <td>-0.0105</td>\n",
       "      <td>0.0015</td>\n",
       "      <td>0.0040</td>\n",
       "      <td>0.0043</td>\n",
       "      <td>-0.0026</td>\n",
       "      <td>0.0006</td>\n",
       "      <td>-0.0049</td>\n",
       "      <td>...</td>\n",
       "      <td>0.1171</td>\n",
       "      <td>0.0825</td>\n",
       "      <td>0.0085</td>\n",
       "      <td>0.0564</td>\n",
       "      <td>-0.0216</td>\n",
       "      <td>0.0029</td>\n",
       "      <td>-0.0125</td>\n",
       "      <td>0.0010</td>\n",
       "      <td>0.0009</td>\n",
       "      <td>-0.0065</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 24</th>\n",
       "      <td>0.5863</td>\n",
       "      <td>0.0072</td>\n",
       "      <td>-0.0055</td>\n",
       "      <td>0.0055</td>\n",
       "      <td>0.0018</td>\n",
       "      <td>-0.0049</td>\n",
       "      <td>-0.0022</td>\n",
       "      <td>-0.0013</td>\n",
       "      <td>0.0013</td>\n",
       "      <td>0.0070</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0105</td>\n",
       "      <td>-0.0205</td>\n",
       "      <td>0.0133</td>\n",
       "      <td>-0.0138</td>\n",
       "      <td>0.0141</td>\n",
       "      <td>-0.0062</td>\n",
       "      <td>-0.0515</td>\n",
       "      <td>0.0054</td>\n",
       "      <td>0.0072</td>\n",
       "      <td>0.0177</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 25</th>\n",
       "      <td>0.5931</td>\n",
       "      <td>0.0068</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0027</td>\n",
       "      <td>0.0013</td>\n",
       "      <td>-0.0005</td>\n",
       "      <td>-0.0009</td>\n",
       "      <td>0.0007</td>\n",
       "      <td>-0.0023</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0898</td>\n",
       "      <td>0.0143</td>\n",
       "      <td>0.0512</td>\n",
       "      <td>0.0432</td>\n",
       "      <td>-0.0308</td>\n",
       "      <td>0.0063</td>\n",
       "      <td>0.0003</td>\n",
       "      <td>-0.0455</td>\n",
       "      <td>-0.0504</td>\n",
       "      <td>-0.0499</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 26</th>\n",
       "      <td>0.5996</td>\n",
       "      <td>0.0065</td>\n",
       "      <td>-0.0056</td>\n",
       "      <td>0.0056</td>\n",
       "      <td>0.0006</td>\n",
       "      <td>0.0014</td>\n",
       "      <td>0.0014</td>\n",
       "      <td>-0.0050</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>-0.0028</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0536</td>\n",
       "      <td>0.0724</td>\n",
       "      <td>0.0224</td>\n",
       "      <td>0.0783</td>\n",
       "      <td>0.0285</td>\n",
       "      <td>-0.0068</td>\n",
       "      <td>-0.0167</td>\n",
       "      <td>-0.0108</td>\n",
       "      <td>-0.0109</td>\n",
       "      <td>-0.0041</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 27</th>\n",
       "      <td>0.6061</td>\n",
       "      <td>0.0064</td>\n",
       "      <td>0.0066</td>\n",
       "      <td>-0.0066</td>\n",
       "      <td>0.0045</td>\n",
       "      <td>0.0053</td>\n",
       "      <td>-0.0048</td>\n",
       "      <td>0.0021</td>\n",
       "      <td>0.0002</td>\n",
       "      <td>-0.0064</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0123</td>\n",
       "      <td>-0.0533</td>\n",
       "      <td>0.0040</td>\n",
       "      <td>0.0009</td>\n",
       "      <td>-0.0091</td>\n",
       "      <td>0.0158</td>\n",
       "      <td>-0.0429</td>\n",
       "      <td>-0.0316</td>\n",
       "      <td>-0.0343</td>\n",
       "      <td>-0.0396</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 28</th>\n",
       "      <td>0.6124</td>\n",
       "      <td>0.0063</td>\n",
       "      <td>-0.0039</td>\n",
       "      <td>0.0039</td>\n",
       "      <td>-0.0006</td>\n",
       "      <td>-0.0006</td>\n",
       "      <td>0.0036</td>\n",
       "      <td>0.0003</td>\n",
       "      <td>0.0015</td>\n",
       "      <td>0.0018</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0195</td>\n",
       "      <td>0.0413</td>\n",
       "      <td>0.1935</td>\n",
       "      <td>-0.0245</td>\n",
       "      <td>0.0337</td>\n",
       "      <td>-0.0337</td>\n",
       "      <td>0.0163</td>\n",
       "      <td>-0.1345</td>\n",
       "      <td>-0.1201</td>\n",
       "      <td>-0.0834</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 29</th>\n",
       "      <td>0.6185</td>\n",
       "      <td>0.0061</td>\n",
       "      <td>0.0015</td>\n",
       "      <td>-0.0015</td>\n",
       "      <td>0.0007</td>\n",
       "      <td>-0.0020</td>\n",
       "      <td>-0.0031</td>\n",
       "      <td>0.0016</td>\n",
       "      <td>-0.0009</td>\n",
       "      <td>-0.0006</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0921</td>\n",
       "      <td>-0.0386</td>\n",
       "      <td>0.0617</td>\n",
       "      <td>0.0431</td>\n",
       "      <td>0.0270</td>\n",
       "      <td>-0.0081</td>\n",
       "      <td>0.0232</td>\n",
       "      <td>-0.0662</td>\n",
       "      <td>-0.0610</td>\n",
       "      <td>-0.0463</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 30</th>\n",
       "      <td>0.6244</td>\n",
       "      <td>0.0059</td>\n",
       "      <td>-0.0028</td>\n",
       "      <td>0.0028</td>\n",
       "      <td>-0.0001</td>\n",
       "      <td>0.0011</td>\n",
       "      <td>-0.0016</td>\n",
       "      <td>-0.0011</td>\n",
       "      <td>-0.0015</td>\n",
       "      <td>-0.0022</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0872</td>\n",
       "      <td>0.0141</td>\n",
       "      <td>0.0087</td>\n",
       "      <td>0.1455</td>\n",
       "      <td>0.0142</td>\n",
       "      <td>-0.0068</td>\n",
       "      <td>-0.0181</td>\n",
       "      <td>-0.0049</td>\n",
       "      <td>-0.0124</td>\n",
       "      <td>-0.0130</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 656</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0029</td>\n",
       "      <td>-0.0007</td>\n",
       "      <td>-0.0147</td>\n",
       "      <td>-0.0003</td>\n",
       "      <td>0.0043</td>\n",
       "      <td>-0.0003</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 657</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0006</td>\n",
       "      <td>-0.0002</td>\n",
       "      <td>0.0027</td>\n",
       "      <td>-0.0006</td>\n",
       "      <td>-0.0012</td>\n",
       "      <td>-0.0025</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 658</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0007</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>-0.0062</td>\n",
       "      <td>-0.0005</td>\n",
       "      <td>0.0076</td>\n",
       "      <td>-0.0008</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 659</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0016</td>\n",
       "      <td>0.0004</td>\n",
       "      <td>-0.0067</td>\n",
       "      <td>-0.0013</td>\n",
       "      <td>0.0024</td>\n",
       "      <td>-0.0010</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 660</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0011</td>\n",
       "      <td>-0.0005</td>\n",
       "      <td>-0.0068</td>\n",
       "      <td>-0.0011</td>\n",
       "      <td>-0.0043</td>\n",
       "      <td>-0.0019</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 661</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0009</td>\n",
       "      <td>0.0009</td>\n",
       "      <td>-0.0029</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0019</td>\n",
       "      <td>0.0004</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 662</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0002</td>\n",
       "      <td>0.0005</td>\n",
       "      <td>0.0022</td>\n",
       "      <td>0.0011</td>\n",
       "      <td>-0.0011</td>\n",
       "      <td>0.0013</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 663</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0005</td>\n",
       "      <td>-0.0019</td>\n",
       "      <td>0.0030</td>\n",
       "      <td>0.0002</td>\n",
       "      <td>0.0016</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 664</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0007</td>\n",
       "      <td>0.0002</td>\n",
       "      <td>-0.0037</td>\n",
       "      <td>0.0007</td>\n",
       "      <td>-0.0007</td>\n",
       "      <td>0.0012</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 665</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0025</td>\n",
       "      <td>-0.0011</td>\n",
       "      <td>0.0017</td>\n",
       "      <td>0.0005</td>\n",
       "      <td>-0.0009</td>\n",
       "      <td>-0.0018</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 666</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0021</td>\n",
       "      <td>-0.0011</td>\n",
       "      <td>0.0035</td>\n",
       "      <td>0.0004</td>\n",
       "      <td>-0.0051</td>\n",
       "      <td>0.0020</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 667</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0003</td>\n",
       "      <td>0.0002</td>\n",
       "      <td>-0.0016</td>\n",
       "      <td>-0.0005</td>\n",
       "      <td>0.0015</td>\n",
       "      <td>0.0020</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 668</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0002</td>\n",
       "      <td>-0.0002</td>\n",
       "      <td>-0.0022</td>\n",
       "      <td>-0.0001</td>\n",
       "      <td>0.0029</td>\n",
       "      <td>-0.0019</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 669</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0011</td>\n",
       "      <td>0.0012</td>\n",
       "      <td>0.0047</td>\n",
       "      <td>0.0006</td>\n",
       "      <td>-0.0076</td>\n",
       "      <td>0.0039</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 670</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0005</td>\n",
       "      <td>0.0014</td>\n",
       "      <td>0.0038</td>\n",
       "      <td>-0.0001</td>\n",
       "      <td>-0.0057</td>\n",
       "      <td>0.0055</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 671</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>0.0002</td>\n",
       "      <td>-0.0065</td>\n",
       "      <td>0.0005</td>\n",
       "      <td>0.0010</td>\n",
       "      <td>-0.0003</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 672</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0005</td>\n",
       "      <td>-0.0003</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0004</td>\n",
       "      <td>0.0008</td>\n",
       "      <td>-0.0022</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 673</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0005</td>\n",
       "      <td>0.0013</td>\n",
       "      <td>-0.0059</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>0.0018</td>\n",
       "      <td>-0.0016</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 674</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0007</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>-0.0031</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0001</td>\n",
       "      <td>-0.0013</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 675</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0011</td>\n",
       "      <td>-0.0003</td>\n",
       "      <td>0.0019</td>\n",
       "      <td>0.0003</td>\n",
       "      <td>-0.0051</td>\n",
       "      <td>0.0022</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 676</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0007</td>\n",
       "      <td>-0.0005</td>\n",
       "      <td>0.0010</td>\n",
       "      <td>0.0002</td>\n",
       "      <td>0.0016</td>\n",
       "      <td>0.0002</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 677</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0004</td>\n",
       "      <td>-0.0034</td>\n",
       "      <td>-0.0002</td>\n",
       "      <td>0.0022</td>\n",
       "      <td>0.0004</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 678</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0008</td>\n",
       "      <td>0.0003</td>\n",
       "      <td>0.0014</td>\n",
       "      <td>-0.0002</td>\n",
       "      <td>-0.0020</td>\n",
       "      <td>-0.0013</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 679</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0008</td>\n",
       "      <td>0.0002</td>\n",
       "      <td>0.0041</td>\n",
       "      <td>0.0002</td>\n",
       "      <td>0.0006</td>\n",
       "      <td>0.0015</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 680</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0006</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>0.0012</td>\n",
       "      <td>0.0011</td>\n",
       "      <td>-0.0009</td>\n",
       "      <td>0.0028</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 681</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0004</td>\n",
       "      <td>0.0002</td>\n",
       "      <td>0.0025</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0011</td>\n",
       "      <td>0.0009</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 682</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0012</td>\n",
       "      <td>0.0015</td>\n",
       "      <td>-0.0030</td>\n",
       "      <td>0.0002</td>\n",
       "      <td>0.0030</td>\n",
       "      <td>-0.0044</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 683</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0021</td>\n",
       "      <td>-0.0002</td>\n",
       "      <td>-0.0018</td>\n",
       "      <td>0.0015</td>\n",
       "      <td>0.0025</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 684</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0007</td>\n",
       "      <td>-0.0004</td>\n",
       "      <td>0.0034</td>\n",
       "      <td>-0.0003</td>\n",
       "      <td>0.0069</td>\n",
       "      <td>-0.0033</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dimension 685</th>\n",
       "      <td>1.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.7071</td>\n",
       "      <td>0.7071</td>\n",
       "      <td>0.0013</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>-0.0001</td>\n",
       "      <td>-0.0001</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>0.0002</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>-0.0000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>685 rows × 687 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               Explained Cumulative Variance  Explained Variance Ratio  \\\n",
       "Dimension 1                           0.1488                    0.1488   \n",
       "Dimension 2                           0.2154                    0.0666   \n",
       "Dimension 3                           0.2699                    0.0544   \n",
       "Dimension 4                           0.3164                    0.0465   \n",
       "Dimension 5                           0.3595                    0.0431   \n",
       "Dimension 6                           0.3818                    0.0222   \n",
       "Dimension 7                           0.4023                    0.0206   \n",
       "Dimension 8                           0.4199                    0.0175   \n",
       "Dimension 9                           0.4365                    0.0167   \n",
       "Dimension 10                          0.4515                    0.0150   \n",
       "Dimension 11                          0.4650                    0.0135   \n",
       "Dimension 12                          0.4781                    0.0131   \n",
       "Dimension 13                          0.4898                    0.0118   \n",
       "Dimension 14                          0.5009                    0.0110   \n",
       "Dimension 15                          0.5116                    0.0107   \n",
       "Dimension 16                          0.5214                    0.0098   \n",
       "Dimension 17                          0.5310                    0.0096   \n",
       "Dimension 18                          0.5398                    0.0088   \n",
       "Dimension 19                          0.5481                    0.0083   \n",
       "Dimension 20                          0.5563                    0.0081   \n",
       "Dimension 21                          0.5642                    0.0079   \n",
       "Dimension 22                          0.5718                    0.0076   \n",
       "Dimension 23                          0.5792                    0.0074   \n",
       "Dimension 24                          0.5863                    0.0072   \n",
       "Dimension 25                          0.5931                    0.0068   \n",
       "Dimension 26                          0.5996                    0.0065   \n",
       "Dimension 27                          0.6061                    0.0064   \n",
       "Dimension 28                          0.6124                    0.0063   \n",
       "Dimension 29                          0.6185                    0.0061   \n",
       "Dimension 30                          0.6244                    0.0059   \n",
       "...                                      ...                       ...   \n",
       "Dimension 656                         1.0000                    0.0000   \n",
       "Dimension 657                         1.0000                    0.0000   \n",
       "Dimension 658                         1.0000                    0.0000   \n",
       "Dimension 659                         1.0000                    0.0000   \n",
       "Dimension 660                         1.0000                    0.0000   \n",
       "Dimension 661                         1.0000                    0.0000   \n",
       "Dimension 662                         1.0000                    0.0000   \n",
       "Dimension 663                         1.0000                    0.0000   \n",
       "Dimension 664                         1.0000                    0.0000   \n",
       "Dimension 665                         1.0000                    0.0000   \n",
       "Dimension 666                         1.0000                    0.0000   \n",
       "Dimension 667                         1.0000                    0.0000   \n",
       "Dimension 668                         1.0000                    0.0000   \n",
       "Dimension 669                         1.0000                    0.0000   \n",
       "Dimension 670                         1.0000                    0.0000   \n",
       "Dimension 671                         1.0000                    0.0000   \n",
       "Dimension 672                         1.0000                    0.0000   \n",
       "Dimension 673                         1.0000                    0.0000   \n",
       "Dimension 674                         1.0000                    0.0000   \n",
       "Dimension 675                         1.0000                    0.0000   \n",
       "Dimension 676                         1.0000                    0.0000   \n",
       "Dimension 677                         1.0000                    0.0000   \n",
       "Dimension 678                         1.0000                    0.0000   \n",
       "Dimension 679                         1.0000                    0.0000   \n",
       "Dimension 680                         1.0000                    0.0000   \n",
       "Dimension 681                         1.0000                    0.0000   \n",
       "Dimension 682                         1.0000                    0.0000   \n",
       "Dimension 683                         1.0000                    0.0000   \n",
       "Dimension 684                         1.0000                    0.0000   \n",
       "Dimension 685                         1.0000                    0.0000   \n",
       "\n",
       "               ANREDE_KZ_1  ANREDE_KZ_2  CAMEO_DEUG_2015_1  CAMEO_DEUG_2015_2  \\\n",
       "Dimension 1         0.0003      -0.0003             0.0018             0.0052   \n",
       "Dimension 2         0.0069      -0.0069            -0.0048            -0.0112   \n",
       "Dimension 3        -0.0206       0.0206             0.0000            -0.0085   \n",
       "Dimension 4         0.0172      -0.0172             0.0081             0.0079   \n",
       "Dimension 5         0.0890      -0.0890            -0.0000            -0.0003   \n",
       "Dimension 6         0.0370      -0.0370            -0.0001            -0.0046   \n",
       "Dimension 7        -0.0246       0.0246             0.0043             0.0059   \n",
       "Dimension 8        -0.0235       0.0235            -0.0055            -0.0085   \n",
       "Dimension 9        -0.0279       0.0279             0.0043             0.0061   \n",
       "Dimension 10        0.0083      -0.0083            -0.0048            -0.0038   \n",
       "Dimension 11       -0.0091       0.0091             0.0002            -0.0016   \n",
       "Dimension 12       -0.0011       0.0011             0.0037            -0.0028   \n",
       "Dimension 13        0.0038      -0.0038            -0.0004            -0.0094   \n",
       "Dimension 14        0.0006      -0.0006            -0.0004             0.0008   \n",
       "Dimension 15       -0.0019       0.0019             0.0005            -0.0049   \n",
       "Dimension 16       -0.0027       0.0027             0.0019            -0.0005   \n",
       "Dimension 17       -0.0109       0.0109             0.0026            -0.0013   \n",
       "Dimension 18       -0.0025       0.0025            -0.0038            -0.0074   \n",
       "Dimension 19        0.0038      -0.0038             0.0046             0.0050   \n",
       "Dimension 20       -0.0108       0.0108             0.0023             0.0043   \n",
       "Dimension 21       -0.0059       0.0059            -0.0009            -0.0033   \n",
       "Dimension 22       -0.0027       0.0027            -0.0054            -0.0061   \n",
       "Dimension 23        0.0105      -0.0105             0.0015             0.0040   \n",
       "Dimension 24       -0.0055       0.0055             0.0018            -0.0049   \n",
       "Dimension 25       -0.0000       0.0000             0.0027             0.0013   \n",
       "Dimension 26       -0.0056       0.0056             0.0006             0.0014   \n",
       "Dimension 27        0.0066      -0.0066             0.0045             0.0053   \n",
       "Dimension 28       -0.0039       0.0039            -0.0006            -0.0006   \n",
       "Dimension 29        0.0015      -0.0015             0.0007            -0.0020   \n",
       "Dimension 30       -0.0028       0.0028            -0.0001             0.0011   \n",
       "...                    ...          ...                ...                ...   \n",
       "Dimension 656       0.0000      -0.0000             0.0029            -0.0007   \n",
       "Dimension 657       0.0000      -0.0000             0.0006            -0.0002   \n",
       "Dimension 658       0.0000       0.0000             0.0007             0.0001   \n",
       "Dimension 659       0.0000      -0.0000             0.0016             0.0004   \n",
       "Dimension 660       0.0000      -0.0000            -0.0011            -0.0005   \n",
       "Dimension 661       0.0000      -0.0000             0.0009             0.0009   \n",
       "Dimension 662       0.0000      -0.0000             0.0002             0.0005   \n",
       "Dimension 663       0.0000      -0.0000            -0.0005            -0.0019   \n",
       "Dimension 664       0.0000      -0.0000            -0.0007             0.0002   \n",
       "Dimension 665       0.0000      -0.0000            -0.0025            -0.0011   \n",
       "Dimension 666       0.0000      -0.0000            -0.0021            -0.0011   \n",
       "Dimension 667       0.0000      -0.0000             0.0003             0.0002   \n",
       "Dimension 668       0.0000       0.0000            -0.0002            -0.0002   \n",
       "Dimension 669       0.0000      -0.0000             0.0011             0.0012   \n",
       "Dimension 670       0.0000      -0.0000             0.0005             0.0014   \n",
       "Dimension 671       0.0000      -0.0000             0.0001             0.0002   \n",
       "Dimension 672       0.0000       0.0000            -0.0005            -0.0003   \n",
       "Dimension 673       0.0000      -0.0000             0.0005             0.0013   \n",
       "Dimension 674       0.0000       0.0000            -0.0007             0.0001   \n",
       "Dimension 675       0.0000       0.0000            -0.0011            -0.0003   \n",
       "Dimension 676       0.0000       0.0000             0.0007            -0.0005   \n",
       "Dimension 677       0.0000      -0.0000             0.0000             0.0004   \n",
       "Dimension 678       0.0000      -0.0000             0.0008             0.0003   \n",
       "Dimension 679       0.0000       0.0000            -0.0008             0.0002   \n",
       "Dimension 680       0.0000       0.0000             0.0006             0.0001   \n",
       "Dimension 681       0.0000      -0.0000             0.0004             0.0002   \n",
       "Dimension 682       0.0000      -0.0000             0.0012             0.0015   \n",
       "Dimension 683       0.0000      -0.0000            -0.0021            -0.0002   \n",
       "Dimension 684       0.0000       0.0000             0.0007            -0.0004   \n",
       "Dimension 685       0.7071       0.7071             0.0013             0.0001   \n",
       "\n",
       "               CAMEO_DEUG_2015_3  CAMEO_DEUG_2015_4  CAMEO_DEUG_2015_5  \\\n",
       "Dimension 1               0.0041             0.0049             0.0013   \n",
       "Dimension 2              -0.0088            -0.0103            -0.0019   \n",
       "Dimension 3              -0.0085            -0.0132             0.0003   \n",
       "Dimension 4               0.0014             0.0015             0.0059   \n",
       "Dimension 5               0.0006            -0.0003             0.0004   \n",
       "Dimension 6              -0.0047            -0.0129            -0.0041   \n",
       "Dimension 7               0.0031             0.0020             0.0019   \n",
       "Dimension 8               0.0003             0.0008             0.0023   \n",
       "Dimension 9               0.0028             0.0019            -0.0002   \n",
       "Dimension 10             -0.0033             0.0051            -0.0009   \n",
       "Dimension 11             -0.0010            -0.0028            -0.0002   \n",
       "Dimension 12             -0.0036            -0.0084             0.0020   \n",
       "Dimension 13             -0.0062            -0.0063            -0.0016   \n",
       "Dimension 14              0.0016             0.0008             0.0013   \n",
       "Dimension 15             -0.0025            -0.0016             0.0064   \n",
       "Dimension 16              0.0007             0.0032             0.0039   \n",
       "Dimension 17             -0.0041            -0.0004             0.0011   \n",
       "Dimension 18             -0.0049            -0.0028            -0.0038   \n",
       "Dimension 19              0.0028            -0.0038             0.0018   \n",
       "Dimension 20              0.0035            -0.0026             0.0026   \n",
       "Dimension 21             -0.0027             0.0005            -0.0024   \n",
       "Dimension 22             -0.0002            -0.0000             0.0036   \n",
       "Dimension 23              0.0043            -0.0026             0.0006   \n",
       "Dimension 24             -0.0022            -0.0013             0.0013   \n",
       "Dimension 25             -0.0005            -0.0009             0.0007   \n",
       "Dimension 26              0.0014            -0.0050             0.0001   \n",
       "Dimension 27             -0.0048             0.0021             0.0002   \n",
       "Dimension 28              0.0036             0.0003             0.0015   \n",
       "Dimension 29             -0.0031             0.0016            -0.0009   \n",
       "Dimension 30             -0.0016            -0.0011            -0.0015   \n",
       "...                          ...                ...                ...   \n",
       "Dimension 656            -0.0147            -0.0003             0.0043   \n",
       "Dimension 657             0.0027            -0.0006            -0.0012   \n",
       "Dimension 658            -0.0062            -0.0005             0.0076   \n",
       "Dimension 659            -0.0067            -0.0013             0.0024   \n",
       "Dimension 660            -0.0068            -0.0011            -0.0043   \n",
       "Dimension 661            -0.0029             0.0000             0.0019   \n",
       "Dimension 662             0.0022             0.0011            -0.0011   \n",
       "Dimension 663             0.0030             0.0002             0.0016   \n",
       "Dimension 664            -0.0037             0.0007            -0.0007   \n",
       "Dimension 665             0.0017             0.0005            -0.0009   \n",
       "Dimension 666             0.0035             0.0004            -0.0051   \n",
       "Dimension 667            -0.0016            -0.0005             0.0015   \n",
       "Dimension 668            -0.0022            -0.0001             0.0029   \n",
       "Dimension 669             0.0047             0.0006            -0.0076   \n",
       "Dimension 670             0.0038            -0.0001            -0.0057   \n",
       "Dimension 671            -0.0065             0.0005             0.0010   \n",
       "Dimension 672            -0.0000            -0.0004             0.0008   \n",
       "Dimension 673            -0.0059             0.0001             0.0018   \n",
       "Dimension 674            -0.0031            -0.0000            -0.0001   \n",
       "Dimension 675             0.0019             0.0003            -0.0051   \n",
       "Dimension 676             0.0010             0.0002             0.0016   \n",
       "Dimension 677            -0.0034            -0.0002             0.0022   \n",
       "Dimension 678             0.0014            -0.0002            -0.0020   \n",
       "Dimension 679             0.0041             0.0002             0.0006   \n",
       "Dimension 680             0.0012             0.0011            -0.0009   \n",
       "Dimension 681             0.0025            -0.0000            -0.0011   \n",
       "Dimension 682            -0.0030             0.0002             0.0030   \n",
       "Dimension 683            -0.0018             0.0015             0.0025   \n",
       "Dimension 684             0.0034            -0.0003             0.0069   \n",
       "Dimension 685            -0.0001            -0.0001             0.0001   \n",
       "\n",
       "               CAMEO_DEUG_2015_6  ...  KBA13_ANZAHL_PKW  MIN_GEBAEUDEJAHR  \\\n",
       "Dimension 1               0.0004  ...            0.0114            0.0196   \n",
       "Dimension 2              -0.0033  ...           -0.0350           -0.0079   \n",
       "Dimension 3               0.0018  ...           -0.0368           -0.0288   \n",
       "Dimension 4              -0.0025  ...            0.0399           -0.0150   \n",
       "Dimension 5               0.0040  ...           -0.0105           -0.0005   \n",
       "Dimension 6              -0.0050  ...           -0.0516            0.0217   \n",
       "Dimension 7               0.0061  ...           -0.0535           -0.0020   \n",
       "Dimension 8               0.0053  ...            0.0681            0.0257   \n",
       "Dimension 9               0.0048  ...            0.0025           -0.0144   \n",
       "Dimension 10             -0.0076  ...           -0.0820           -0.0164   \n",
       "Dimension 11             -0.0047  ...           -0.0273            0.0232   \n",
       "Dimension 12              0.0008  ...           -0.0372           -0.0538   \n",
       "Dimension 13              0.0091  ...            0.0108           -0.0149   \n",
       "Dimension 14             -0.0010  ...           -0.0165            0.0054   \n",
       "Dimension 15             -0.0008  ...           -0.0533            0.0109   \n",
       "Dimension 16             -0.0020  ...           -0.2428            0.0244   \n",
       "Dimension 17              0.0010  ...           -0.0122           -0.1091   \n",
       "Dimension 18              0.0084  ...            0.0280           -0.1746   \n",
       "Dimension 19             -0.0046  ...           -0.0616            0.0428   \n",
       "Dimension 20              0.0010  ...           -0.0337           -0.0641   \n",
       "Dimension 21              0.0049  ...           -0.0865           -0.0894   \n",
       "Dimension 22             -0.0027  ...           -0.1209            0.0632   \n",
       "Dimension 23             -0.0049  ...            0.1171            0.0825   \n",
       "Dimension 24              0.0070  ...           -0.0105           -0.0205   \n",
       "Dimension 25             -0.0023  ...            0.0898            0.0143   \n",
       "Dimension 26             -0.0028  ...            0.0536            0.0724   \n",
       "Dimension 27             -0.0064  ...            0.0123           -0.0533   \n",
       "Dimension 28              0.0018  ...            0.0195            0.0413   \n",
       "Dimension 29             -0.0006  ...           -0.0921           -0.0386   \n",
       "Dimension 30             -0.0022  ...            0.0872            0.0141   \n",
       "...                          ...  ...               ...               ...   \n",
       "Dimension 656            -0.0003  ...           -0.0000           -0.0000   \n",
       "Dimension 657            -0.0025  ...            0.0000           -0.0000   \n",
       "Dimension 658            -0.0008  ...            0.0000            0.0000   \n",
       "Dimension 659            -0.0010  ...           -0.0000            0.0000   \n",
       "Dimension 660            -0.0019  ...            0.0000           -0.0000   \n",
       "Dimension 661             0.0004  ...           -0.0000           -0.0000   \n",
       "Dimension 662             0.0013  ...            0.0000           -0.0000   \n",
       "Dimension 663             0.0001  ...            0.0000            0.0000   \n",
       "Dimension 664             0.0012  ...            0.0000           -0.0000   \n",
       "Dimension 665            -0.0018  ...            0.0000            0.0000   \n",
       "Dimension 666             0.0020  ...           -0.0000            0.0000   \n",
       "Dimension 667             0.0020  ...           -0.0000            0.0000   \n",
       "Dimension 668            -0.0019  ...            0.0000            0.0000   \n",
       "Dimension 669             0.0039  ...            0.0000            0.0000   \n",
       "Dimension 670             0.0055  ...            0.0000            0.0000   \n",
       "Dimension 671            -0.0003  ...           -0.0000            0.0000   \n",
       "Dimension 672            -0.0022  ...           -0.0000            0.0000   \n",
       "Dimension 673            -0.0016  ...            0.0000           -0.0000   \n",
       "Dimension 674            -0.0013  ...           -0.0000            0.0000   \n",
       "Dimension 675             0.0022  ...           -0.0000           -0.0000   \n",
       "Dimension 676             0.0002  ...            0.0000            0.0000   \n",
       "Dimension 677             0.0004  ...           -0.0000            0.0000   \n",
       "Dimension 678            -0.0013  ...           -0.0000           -0.0000   \n",
       "Dimension 679             0.0015  ...           -0.0000            0.0000   \n",
       "Dimension 680             0.0028  ...           -0.0000           -0.0000   \n",
       "Dimension 681             0.0009  ...           -0.0000            0.0000   \n",
       "Dimension 682            -0.0044  ...            0.0000            0.0000   \n",
       "Dimension 683             0.0001  ...           -0.0000            0.0000   \n",
       "Dimension 684            -0.0033  ...            0.0000           -0.0000   \n",
       "Dimension 685             0.0002  ...            0.0000            0.0000   \n",
       "\n",
       "               ANZ_KINDER  ANZ_STATISTISCHE_HAUSHALTE  EINGEZOGENAM_HH_JAHR  \\\n",
       "Dimension 1        0.0330                     -0.0177                0.0082   \n",
       "Dimension 2        0.0143                      0.0383                0.0594   \n",
       "Dimension 3       -0.0215                      0.0650               -0.0029   \n",
       "Dimension 4        0.0058                     -0.0079                0.0051   \n",
       "Dimension 5       -0.0104                      0.0230               -0.0092   \n",
       "Dimension 6       -0.0367                     -0.0148               -0.0196   \n",
       "Dimension 7        0.0269                      0.0279                0.0107   \n",
       "Dimension 8       -0.0090                      0.0530                0.0843   \n",
       "Dimension 9       -0.0077                     -0.0248                0.0240   \n",
       "Dimension 10      -0.0050                      0.0258               -0.0212   \n",
       "Dimension 11       0.0464                     -0.0176               -0.0197   \n",
       "Dimension 12      -0.0291                     -0.0303               -0.0132   \n",
       "Dimension 13       0.0067                      0.0620                0.0538   \n",
       "Dimension 14      -0.0196                     -0.0154                0.0669   \n",
       "Dimension 15      -0.0205                      0.0457                0.1119   \n",
       "Dimension 16       0.0043                     -0.0200                0.0613   \n",
       "Dimension 17       0.0437                      0.0115               -0.2802   \n",
       "Dimension 18      -0.0765                     -0.0060               -0.0367   \n",
       "Dimension 19      -0.0289                      0.0042                0.0812   \n",
       "Dimension 20      -0.0539                      0.1230               -0.0546   \n",
       "Dimension 21      -0.0020                      0.0026               -0.0872   \n",
       "Dimension 22       0.0526                      0.1087               -0.0866   \n",
       "Dimension 23       0.0085                      0.0564               -0.0216   \n",
       "Dimension 24       0.0133                     -0.0138                0.0141   \n",
       "Dimension 25       0.0512                      0.0432               -0.0308   \n",
       "Dimension 26       0.0224                      0.0783                0.0285   \n",
       "Dimension 27       0.0040                      0.0009               -0.0091   \n",
       "Dimension 28       0.1935                     -0.0245                0.0337   \n",
       "Dimension 29       0.0617                      0.0431                0.0270   \n",
       "Dimension 30       0.0087                      0.1455                0.0142   \n",
       "...                   ...                         ...                   ...   \n",
       "Dimension 656      0.0000                      0.0000               -0.0000   \n",
       "Dimension 657     -0.0000                      0.0000                0.0000   \n",
       "Dimension 658     -0.0000                     -0.0000                0.0000   \n",
       "Dimension 659      0.0000                      0.0000               -0.0000   \n",
       "Dimension 660      0.0000                      0.0000               -0.0000   \n",
       "Dimension 661      0.0000                      0.0000                0.0000   \n",
       "Dimension 662      0.0000                      0.0000               -0.0000   \n",
       "Dimension 663     -0.0000                     -0.0000               -0.0000   \n",
       "Dimension 664     -0.0000                     -0.0000                0.0000   \n",
       "Dimension 665     -0.0000                      0.0000                0.0000   \n",
       "Dimension 666      0.0000                      0.0000               -0.0000   \n",
       "Dimension 667     -0.0000                     -0.0000               -0.0000   \n",
       "Dimension 668      0.0000                      0.0000               -0.0000   \n",
       "Dimension 669      0.0000                     -0.0000               -0.0000   \n",
       "Dimension 670     -0.0000                     -0.0000                0.0000   \n",
       "Dimension 671     -0.0000                     -0.0000               -0.0000   \n",
       "Dimension 672     -0.0000                     -0.0000               -0.0000   \n",
       "Dimension 673     -0.0000                     -0.0000               -0.0000   \n",
       "Dimension 674     -0.0000                     -0.0000               -0.0000   \n",
       "Dimension 675      0.0000                      0.0000                0.0000   \n",
       "Dimension 676     -0.0000                      0.0000               -0.0000   \n",
       "Dimension 677      0.0000                     -0.0000               -0.0000   \n",
       "Dimension 678     -0.0000                     -0.0000                0.0000   \n",
       "Dimension 679      0.0000                     -0.0000                0.0000   \n",
       "Dimension 680      0.0000                     -0.0000               -0.0000   \n",
       "Dimension 681      0.0000                      0.0000               -0.0000   \n",
       "Dimension 682      0.0000                     -0.0000                0.0000   \n",
       "Dimension 683     -0.0000                     -0.0000                0.0000   \n",
       "Dimension 684      0.0000                      0.0000                0.0000   \n",
       "Dimension 685     -0.0000                      0.0000               -0.0000   \n",
       "\n",
       "               EXTSEL992  VERDICHTUNGSRAUM  VK_DHT4A  VK_DISTANZ  VK_ZG11  \n",
       "Dimension 1       0.0080           -0.0029   -0.0830     -0.0716  -0.0498  \n",
       "Dimension 2      -0.0125            0.0179   -0.0009      0.0282   0.0493  \n",
       "Dimension 3       0.0077            0.0407    0.0182      0.0040  -0.0025  \n",
       "Dimension 4      -0.0088            0.0101    0.0013      0.0061   0.0092  \n",
       "Dimension 5       0.0000            0.0154    0.0108     -0.0181  -0.0467  \n",
       "Dimension 6       0.0149           -0.0091   -0.0193     -0.0388  -0.0684  \n",
       "Dimension 7       0.0012            0.0364   -0.0161     -0.0173  -0.0050  \n",
       "Dimension 8      -0.0088           -0.0052    0.0648      0.0749   0.0837  \n",
       "Dimension 9       0.0078            0.0020    0.0071      0.0080   0.0116  \n",
       "Dimension 10     -0.0035           -0.0084    0.0131      0.0189   0.0174  \n",
       "Dimension 11      0.0707           -0.0047   -0.1182     -0.1343  -0.1209  \n",
       "Dimension 12     -0.0068            0.0179    0.0194      0.0091  -0.0066  \n",
       "Dimension 13      0.0063            0.0047   -0.0075     -0.0046   0.0073  \n",
       "Dimension 14     -0.0273            0.0001    0.0206      0.0355   0.0448  \n",
       "Dimension 15      0.0212            0.0454    0.0014      0.0031   0.0075  \n",
       "Dimension 16     -0.0110           -0.0140    0.0119      0.0269   0.0368  \n",
       "Dimension 17     -0.0133            0.0450    0.0043     -0.0214  -0.0393  \n",
       "Dimension 18      0.0279            0.0011    0.0243      0.0072  -0.0128  \n",
       "Dimension 19      0.0060           -0.0149   -0.0065     -0.0039  -0.0052  \n",
       "Dimension 20      0.0150           -0.1048    0.0206     -0.0052  -0.0294  \n",
       "Dimension 21      0.0022           -0.0117    0.0109      0.0072   0.0013  \n",
       "Dimension 22     -0.0086           -0.0495    0.0119      0.0079   0.0064  \n",
       "Dimension 23      0.0029           -0.0125    0.0010      0.0009  -0.0065  \n",
       "Dimension 24     -0.0062           -0.0515    0.0054      0.0072   0.0177  \n",
       "Dimension 25      0.0063            0.0003   -0.0455     -0.0504  -0.0499  \n",
       "Dimension 26     -0.0068           -0.0167   -0.0108     -0.0109  -0.0041  \n",
       "Dimension 27      0.0158           -0.0429   -0.0316     -0.0343  -0.0396  \n",
       "Dimension 28     -0.0337            0.0163   -0.1345     -0.1201  -0.0834  \n",
       "Dimension 29     -0.0081            0.0232   -0.0662     -0.0610  -0.0463  \n",
       "Dimension 30     -0.0068           -0.0181   -0.0049     -0.0124  -0.0130  \n",
       "...                  ...               ...       ...         ...      ...  \n",
       "Dimension 656    -0.0000           -0.0000   -0.0000      0.0000  -0.0000  \n",
       "Dimension 657     0.0000           -0.0000   -0.0000      0.0000  -0.0000  \n",
       "Dimension 658    -0.0000           -0.0000   -0.0000      0.0000   0.0000  \n",
       "Dimension 659     0.0000           -0.0000   -0.0000     -0.0000   0.0000  \n",
       "Dimension 660     0.0000            0.0000    0.0000     -0.0000   0.0000  \n",
       "Dimension 661    -0.0000            0.0000   -0.0000      0.0000  -0.0000  \n",
       "Dimension 662     0.0000           -0.0000    0.0000     -0.0000   0.0000  \n",
       "Dimension 663     0.0000            0.0000    0.0000     -0.0000   0.0000  \n",
       "Dimension 664    -0.0000            0.0000    0.0000     -0.0000  -0.0000  \n",
       "Dimension 665     0.0000            0.0000   -0.0000      0.0000  -0.0000  \n",
       "Dimension 666    -0.0000           -0.0000    0.0000      0.0000  -0.0000  \n",
       "Dimension 667     0.0000            0.0000    0.0000      0.0000  -0.0000  \n",
       "Dimension 668    -0.0000            0.0000    0.0000     -0.0000   0.0000  \n",
       "Dimension 669    -0.0000           -0.0000   -0.0000     -0.0000   0.0000  \n",
       "Dimension 670     0.0000            0.0000    0.0000     -0.0000   0.0000  \n",
       "Dimension 671    -0.0000           -0.0000   -0.0000     -0.0000   0.0000  \n",
       "Dimension 672     0.0000            0.0000    0.0000      0.0000  -0.0000  \n",
       "Dimension 673    -0.0000           -0.0000    0.0000     -0.0000   0.0000  \n",
       "Dimension 674    -0.0000           -0.0000    0.0000     -0.0000   0.0000  \n",
       "Dimension 675    -0.0000            0.0000    0.0000     -0.0000   0.0000  \n",
       "Dimension 676    -0.0000            0.0000   -0.0000      0.0000  -0.0000  \n",
       "Dimension 677     0.0000           -0.0000    0.0000     -0.0000  -0.0000  \n",
       "Dimension 678     0.0000           -0.0000   -0.0000     -0.0000   0.0000  \n",
       "Dimension 679     0.0000           -0.0000    0.0000      0.0000   0.0000  \n",
       "Dimension 680     0.0000           -0.0000    0.0000     -0.0000   0.0000  \n",
       "Dimension 681     0.0000           -0.0000   -0.0000      0.0000   0.0000  \n",
       "Dimension 682    -0.0000           -0.0000   -0.0000     -0.0000   0.0000  \n",
       "Dimension 683     0.0000            0.0000   -0.0000      0.0000  -0.0000  \n",
       "Dimension 684    -0.0000            0.0000   -0.0000      0.0000  -0.0000  \n",
       "Dimension 685     0.0000           -0.0000    0.0000     -0.0000   0.0000  \n",
       "\n",
       "[685 rows x 687 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "results = pca_results(azdias, pca)\n",
    "results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pick a number of components"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T05:26:37.263163Z",
     "start_time": "2019-08-12T05:26:37.259176Z"
    }
   },
   "outputs": [],
   "source": [
    "n_components = 150"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T05:26:37.896957Z",
     "start_time": "2019-08-12T05:26:37.884994Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9082"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "explained_variance = results.iloc[n_components, 0]\n",
    "explained_variance"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Re-apply PCA to the data while selecting for number of components to retain"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T05:29:36.952650Z",
     "start_time": "2019-08-12T05:28:27.526352Z"
    }
   },
   "outputs": [],
   "source": [
    "pca = PCA(n_components=n_components, random_state=777)\n",
    "azdias_pca = pca.fit_transform(azdias)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### PCA Results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T05:30:13.347879Z",
     "start_time": "2019-08-12T05:30:13.331950Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Weights for PC1\n",
      "Top 5 weights\n",
      "D19_GESAMT_ONLINE_QUOTE_12     0.4903\n",
      "D19_VERSAND_ONLINE_QUOTE_12    0.4737\n",
      "ONLINE_AFFINITAET              0.1393\n",
      "D19_BANKEN_ONLINE_QUOTE_12     0.0913\n",
      "D19_GESAMT_ANZ_12              0.0885\n",
      "Name: 0, dtype: float64\n",
      "\n",
      "\n",
      "Bottom 5 weights\n",
      "W_KEIT_KIND_HH             -0.0958\n",
      "D19_VERSAND_ONLINE_DATUM   -0.2964\n",
      "D19_VERSAND_DATUM          -0.2990\n",
      "D19_GESAMT_ONLINE_DATUM    -0.3124\n",
      "D19_GESAMT_DATUM           -0.3141\n",
      "Name: 0, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "print_pcs(azdias, pca, comp=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T05:30:20.985989Z",
     "start_time": "2019-08-12T05:30:20.970033Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Weights for PC2\n",
      "Top 5 weights\n",
      "ALTER_HH         0.3310\n",
      "SEMIO_REL        0.2465\n",
      "SEMIO_PFLICHT    0.2112\n",
      "FINANZ_SPARER    0.2055\n",
      "ORTSGR_KLS9      0.1711\n",
      "Name: 1, dtype: float64\n",
      "\n",
      "\n",
      "Bottom 5 weights\n",
      "MOBI_REGIO          -0.1399\n",
      "FINANZ_MINIMALIST   -0.1573\n",
      "FINANZ_VORSORGER    -0.1607\n",
      "SEMIO_LUST          -0.1662\n",
      "SEMIO_ERL           -0.1789\n",
      "Name: 1, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "print_pcs(azdias, pca, comp=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T05:30:26.857627Z",
     "start_time": "2019-08-12T05:30:26.842668Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Weights for PC3\n",
      "Top 5 weights\n",
      "ORTSGR_KLS9         0.3036\n",
      "EWDICHTE            0.2092\n",
      "SEMIO_ERL           0.1452\n",
      "FINANZ_HAUSBAUER    0.1156\n",
      "SEMIO_LUST          0.1153\n",
      "Name: 2, dtype: float64\n",
      "\n",
      "\n",
      "Bottom 5 weights\n",
      "FINANZ_ANLEGER   -0.1413\n",
      "SEMIO_REL        -0.1725\n",
      "INNENSTADT       -0.2327\n",
      "BALLRAUM         -0.2348\n",
      "ALTER_HH         -0.2721\n",
      "Name: 2, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "print_pcs(azdias, pca, comp=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T05:30:34.859377Z",
     "start_time": "2019-08-12T05:30:34.852396Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.14879457, 0.21543767, 0.2698714 , 0.31638538, 0.35952412,\n",
       "       0.38175618, 0.4023303 , 0.41986534, 0.43653928, 0.45154193,\n",
       "       0.46499263, 0.47805131, 0.48984636, 0.50086212, 0.51157054,\n",
       "       0.52141132, 0.53100713, 0.5398324 , 0.54814447, 0.55626691,\n",
       "       0.56420539, 0.57179247, 0.57917487, 0.58632662, 0.59314383,\n",
       "       0.59963251, 0.60606662, 0.61235082, 0.61846485, 0.62437557,\n",
       "       0.63012401, 0.63579734, 0.64137774, 0.64671545, 0.65187342,\n",
       "       0.6569394 , 0.66195205, 0.66670932, 0.67143234, 0.67589032,\n",
       "       0.68026605, 0.68456008, 0.68880638, 0.69302904, 0.69707645,\n",
       "       0.70107254, 0.7050183 , 0.70888582, 0.71259461, 0.71629607,\n",
       "       0.71987792, 0.72340219, 0.72689514, 0.73034185, 0.73372489,\n",
       "       0.73703295, 0.74028574, 0.74353046, 0.74667187, 0.74979614,\n",
       "       0.75289535, 0.75593423, 0.75892637, 0.7618655 , 0.7647813 ,\n",
       "       0.76764514, 0.77049595, 0.77329392, 0.7760069 , 0.77866851,\n",
       "       0.78130755, 0.78390968, 0.78644534, 0.78895741, 0.79141351,\n",
       "       0.79384975, 0.79627608, 0.79867961, 0.80107031, 0.80340481,\n",
       "       0.80571806, 0.80801078, 0.81028741, 0.81253022, 0.81475161,\n",
       "       0.81689093, 0.81902478, 0.82113462, 0.82321821, 0.82526287,\n",
       "       0.82726805, 0.82923987, 0.83116998, 0.83305303, 0.83490408,\n",
       "       0.83674635, 0.83855738, 0.8403337 , 0.84210193, 0.84384432,\n",
       "       0.84556693, 0.84726228, 0.84895302, 0.85061801, 0.85226645,\n",
       "       0.8538735 , 0.85546126, 0.85704454, 0.85860721, 0.86015325,\n",
       "       0.86166359, 0.86316259, 0.86465168, 0.86608835, 0.86752072,\n",
       "       0.86893474, 0.87033158, 0.8716922 , 0.87303834, 0.8743645 ,\n",
       "       0.87567152, 0.87696245, 0.87824014, 0.8794945 , 0.88073095,\n",
       "       0.88194398, 0.88314442, 0.88431958, 0.88548554, 0.88663991,\n",
       "       0.88778364, 0.88891123, 0.89003367, 0.89112444, 0.8922056 ,\n",
       "       0.89325508, 0.89428491, 0.89531191, 0.89632534, 0.8973284 ,\n",
       "       0.89832372, 0.89931544, 0.9002792 , 0.90123939, 0.90218335,\n",
       "       0.90311624, 0.90404429, 0.90495466, 0.90585423, 0.90673402])"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pca.explained_variance_ratio_.cumsum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T05:30:56.372921Z",
     "start_time": "2019-08-12T05:30:56.361953Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['temp/pca.joblib']"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dump(pca, 'temp/pca.joblib')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Clustering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T07:04:13.722481Z",
     "start_time": "2019-08-12T05:35:39.207202Z"
    },
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running KMeans with 2 cluster(s)\n",
      "Finished KMeans with 2 cluster(s) in 0.0m:53.3224892616272s\n",
      "Running KMeans with 3 cluster(s)\n",
      "Finished KMeans with 3 cluster(s) in 2.0m:1.0543417930603027s\n",
      "Running KMeans with 4 cluster(s)\n",
      "Finished KMeans with 4 cluster(s) in 2.0m:23.61949372291565s\n",
      "Running KMeans with 5 cluster(s)\n",
      "Finished KMeans with 5 cluster(s) in 4.0m:56.41005301475525s\n",
      "Running KMeans with 6 cluster(s)\n",
      "Finished KMeans with 6 cluster(s) in 5.0m:15.610645055770874s\n",
      "Running KMeans with 7 cluster(s)\n",
      "Finished KMeans with 7 cluster(s) in 6.0m:6.284068584442139s\n",
      "Running KMeans with 8 cluster(s)\n",
      "Finished KMeans with 8 cluster(s) in 7.0m:51.431763648986816s\n",
      "Running KMeans with 9 cluster(s)\n",
      "Finished KMeans with 9 cluster(s) in 7.0m:45.77589559555054s\n",
      "Running KMeans with 10 cluster(s)\n",
      "Finished KMeans with 10 cluster(s) in 7.0m:57.756840467453s\n",
      "Running KMeans with 11 cluster(s)\n",
      "Finished KMeans with 11 cluster(s) in 9.0m:9.520852327346802s\n",
      "Running KMeans with 12 cluster(s)\n",
      "Finished KMeans with 12 cluster(s) in 7.0m:56.797954082489014s\n",
      "Running KMeans with 13 cluster(s)\n",
      "Finished KMeans with 13 cluster(s) in 7.0m:26.21763849258423s\n",
      "Running KMeans with 14 cluster(s)\n",
      "Finished KMeans with 14 cluster(s) in 10.0m:17.30099606513977s\n",
      "Running KMeans with 15 cluster(s)\n",
      "Finished KMeans with 15 cluster(s) in 8.0m:33.37239933013916s\n"
     ]
    }
   ],
   "source": [
    "scores = []\n",
    "\n",
    "clusters = list(range(2,16,1))\n",
    "# Over a number of different cluster counts...\n",
    "for k in clusters:\n",
    "    # run k-means clustering on the data and...\n",
    "    print(f'Running KMeans with {k} cluster(s)')\n",
    "    start = time() # Get start time\n",
    "    kmeans = KMeans(n_clusters = k, random_state=777)\n",
    "    model = kmeans.fit(azdias_pca)\n",
    "    # compute the average within-cluster distances.\n",
    "    score = model.score(azdias_pca)\n",
    "    scores.append(np.abs(score))\n",
    "    \n",
    "    end = time() # Get end time\n",
    "    m, s = divmod(end - start, 60)\n",
    "    print(f'Finished KMeans with {k} cluster(s) in {m}m:{s}s')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Investigate the change in within-cluster distance across number of clusters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T07:05:09.918856Z",
     "start_time": "2019-08-12T07:05:09.211780Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(clusters, scores, linestyle='--', marker='o', color='b');\n",
    "plt.xlabel('Number of Clusters');\n",
    "plt.ylabel('Score');\n",
    "plt.title('Score vs. Number of Clusters')\n",
    "plt.savefig('temp/kmeans.png')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Re-fit the k-means model with the selected number of clusters and obtain cluster predictions for the general population demographics data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T07:44:41.377829Z",
     "start_time": "2019-08-12T07:36:21.519330Z"
    }
   },
   "outputs": [],
   "source": [
    "kmeans = KMeans(n_clusters = 8, random_state=777)\n",
    "kmeans = kmeans.fit(azdias_pca)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T07:48:31.908884Z",
     "start_time": "2019-08-12T07:48:31.904897Z"
    }
   },
   "outputs": [],
   "source": [
    "population_labels = kmeans.labels_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T07:48:33.084465Z",
     "start_time": "2019-08-12T07:48:33.073506Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['temp/kmeans_model.joblib']"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dump(kmeans, 'temp/kmeans_model.joblib')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Customers data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T08:17:21.615930Z",
     "start_time": "2019-08-12T08:17:11.529016Z"
    }
   },
   "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>LNR</th>\n",
       "      <th>AGER_TYP</th>\n",
       "      <th>AKT_DAT_KL</th>\n",
       "      <th>ALTER_HH</th>\n",
       "      <th>ALTER_KIND1</th>\n",
       "      <th>ALTER_KIND2</th>\n",
       "      <th>ALTER_KIND3</th>\n",
       "      <th>ALTER_KIND4</th>\n",
       "      <th>ALTERSKATEGORIE_FEIN</th>\n",
       "      <th>ANZ_HAUSHALTE_AKTIV</th>\n",
       "      <th>...</th>\n",
       "      <th>VK_ZG11</th>\n",
       "      <th>W_KEIT_KIND_HH</th>\n",
       "      <th>WOHNDAUER_2008</th>\n",
       "      <th>WOHNLAGE</th>\n",
       "      <th>ZABEOTYP</th>\n",
       "      <th>PRODUCT_GROUP</th>\n",
       "      <th>CUSTOMER_GROUP</th>\n",
       "      <th>ONLINE_PURCHASE</th>\n",
       "      <th>ANREDE_KZ</th>\n",
       "      <th>ALTERSKATEGORIE_GROB</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>9626</td>\n",
       "      <td>2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>10.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>2.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>3</td>\n",
       "      <td>COSMETIC_AND_FOOD</td>\n",
       "      <td>MULTI_BUYER</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>9628</td>\n",
       "      <td>-1</td>\n",
       "      <td>9.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>FOOD</td>\n",
       "      <td>SINGLE_BUYER</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>143872</td>\n",
       "      <td>-1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>11.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3</td>\n",
       "      <td>COSMETIC_AND_FOOD</td>\n",
       "      <td>MULTI_BUYER</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>143873</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>9.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>1</td>\n",
       "      <td>COSMETIC</td>\n",
       "      <td>MULTI_BUYER</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>143874</td>\n",
       "      <td>-1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>14.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "      <td>FOOD</td>\n",
       "      <td>MULTI_BUYER</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 369 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      LNR  AGER_TYP  AKT_DAT_KL  ALTER_HH  ALTER_KIND1  ALTER_KIND2  \\\n",
       "0    9626         2         1.0      10.0          NaN          NaN   \n",
       "1    9628        -1         9.0      11.0          NaN          NaN   \n",
       "2  143872        -1         1.0       6.0          NaN          NaN   \n",
       "3  143873         1         1.0       8.0          NaN          NaN   \n",
       "4  143874        -1         1.0      20.0          NaN          NaN   \n",
       "\n",
       "   ALTER_KIND3  ALTER_KIND4  ALTERSKATEGORIE_FEIN  ANZ_HAUSHALTE_AKTIV  ...  \\\n",
       "0          NaN          NaN                  10.0                  1.0  ...   \n",
       "1          NaN          NaN                   NaN                  NaN  ...   \n",
       "2          NaN          NaN                   0.0                  1.0  ...   \n",
       "3          NaN          NaN                   8.0                  0.0  ...   \n",
       "4          NaN          NaN                  14.0                  7.0  ...   \n",
       "\n",
       "   VK_ZG11  W_KEIT_KIND_HH  WOHNDAUER_2008  WOHNLAGE  ZABEOTYP  \\\n",
       "0      2.0             6.0             9.0       7.0         3   \n",
       "1      3.0             0.0             9.0       NaN         3   \n",
       "2     11.0             6.0             9.0       2.0         3   \n",
       "3      2.0             NaN             9.0       7.0         1   \n",
       "4      4.0             2.0             9.0       3.0         1   \n",
       "\n",
       "       PRODUCT_GROUP  CUSTOMER_GROUP ONLINE_PURCHASE ANREDE_KZ  \\\n",
       "0  COSMETIC_AND_FOOD     MULTI_BUYER               0         1   \n",
       "1               FOOD    SINGLE_BUYER               0         1   \n",
       "2  COSMETIC_AND_FOOD     MULTI_BUYER               0         2   \n",
       "3           COSMETIC     MULTI_BUYER               0         1   \n",
       "4               FOOD     MULTI_BUYER               0         1   \n",
       "\n",
       "  ALTERSKATEGORIE_GROB  \n",
       "0                    4  \n",
       "1                    4  \n",
       "2                    4  \n",
       "3                    4  \n",
       "4                    3  \n",
       "\n",
       "[5 rows x 369 columns]"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "customers = pd.read_csv('data/Udacity_CUSTOMERS_052018.csv', sep=';', low_memory=False)\n",
    "customers.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Load the preprocessing pipeline and apply on customers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T08:17:22.321056Z",
     "start_time": "2019-08-12T08:17:22.306066Z"
    }
   },
   "outputs": [],
   "source": [
    "clean_pipeline = load('temp/preprocessing_pipeline.joblib') "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T08:17:23.253553Z",
     "start_time": "2019-08-12T08:17:22.966291Z"
    }
   },
   "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>LNR</th>\n",
       "      <th>AGER_TYP</th>\n",
       "      <th>AKT_DAT_KL</th>\n",
       "      <th>ALTER_HH</th>\n",
       "      <th>ALTER_KIND1</th>\n",
       "      <th>ALTER_KIND2</th>\n",
       "      <th>ALTER_KIND3</th>\n",
       "      <th>ALTER_KIND4</th>\n",
       "      <th>ALTERSKATEGORIE_FEIN</th>\n",
       "      <th>ANZ_HAUSHALTE_AKTIV</th>\n",
       "      <th>...</th>\n",
       "      <th>VHN</th>\n",
       "      <th>VK_DHT4A</th>\n",
       "      <th>VK_DISTANZ</th>\n",
       "      <th>VK_ZG11</th>\n",
       "      <th>W_KEIT_KIND_HH</th>\n",
       "      <th>WOHNDAUER_2008</th>\n",
       "      <th>WOHNLAGE</th>\n",
       "      <th>ZABEOTYP</th>\n",
       "      <th>ANREDE_KZ</th>\n",
       "      <th>ALTERSKATEGORIE_GROB</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>9626</td>\n",
       "      <td>2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>10.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>9628</td>\n",
       "      <td>-1</td>\n",
       "      <td>9.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>143872</td>\n",
       "      <td>-1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>143873</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>2.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>9.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>143874</td>\n",
       "      <td>-1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>14.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 366 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      LNR  AGER_TYP  AKT_DAT_KL  ALTER_HH  ALTER_KIND1  ALTER_KIND2  \\\n",
       "0    9626         2         1.0      10.0          NaN          NaN   \n",
       "1    9628        -1         9.0      11.0          NaN          NaN   \n",
       "2  143872        -1         1.0       6.0          NaN          NaN   \n",
       "3  143873         1         1.0       8.0          NaN          NaN   \n",
       "4  143874        -1         1.0      20.0          NaN          NaN   \n",
       "\n",
       "   ALTER_KIND3  ALTER_KIND4  ALTERSKATEGORIE_FEIN  ANZ_HAUSHALTE_AKTIV  ...  \\\n",
       "0          NaN          NaN                  10.0                  1.0  ...   \n",
       "1          NaN          NaN                   NaN                  NaN  ...   \n",
       "2          NaN          NaN                   0.0                  1.0  ...   \n",
       "3          NaN          NaN                   8.0                  0.0  ...   \n",
       "4          NaN          NaN                  14.0                  7.0  ...   \n",
       "\n",
       "   VHN  VK_DHT4A  VK_DISTANZ  VK_ZG11  W_KEIT_KIND_HH  WOHNDAUER_2008  \\\n",
       "0  3.0       5.0         3.0      2.0             6.0             9.0   \n",
       "1  NaN       6.0         6.0      3.0             0.0             9.0   \n",
       "2  4.0      10.0        13.0     11.0             6.0             9.0   \n",
       "3  2.0       6.0         4.0      2.0             NaN             9.0   \n",
       "4  4.0       3.0         5.0      4.0             2.0             9.0   \n",
       "\n",
       "   WOHNLAGE ZABEOTYP ANREDE_KZ ALTERSKATEGORIE_GROB  \n",
       "0       7.0        3         1                    4  \n",
       "1       NaN        3         1                    4  \n",
       "2       2.0        3         2                    4  \n",
       "3       7.0        1         1                    4  \n",
       "4       3.0        1         1                    3  \n",
       "\n",
       "[5 rows x 366 columns]"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "extra_columns = ['ONLINE_PURCHASE', 'PRODUCT_GROUP', 'CUSTOMER_GROUP']\n",
    "customers_extra = customers[extra_columns + ['LNR']]\n",
    "customers = customers.drop(extra_columns, axis=1)\n",
    "customers.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T08:19:39.182481Z",
     "start_time": "2019-08-12T08:17:23.825989Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\radue\\Anaconda3\\envs\\udacity_ml_capstone\\lib\\site-packages\\pyarrow\\pandas_compat.py:708: FutureWarning: .labels was deprecated in version 0.24.0. Use .codes instead.\n",
      "  labels = getattr(columns, 'labels', None) or [\n",
      "C:\\Users\\radue\\Anaconda3\\envs\\udacity_ml_capstone\\lib\\site-packages\\pyarrow\\pandas_compat.py:735: FutureWarning: the 'labels' keyword is deprecated, use 'codes' instead\n",
      "  return pd.MultiIndex(levels=new_levels, labels=labels, names=columns.names)\n",
      "C:\\Users\\radue\\Anaconda3\\envs\\udacity_ml_capstone\\lib\\site-packages\\pyarrow\\pandas_compat.py:752: FutureWarning: .labels was deprecated in version 0.24.0. Use .codes instead.\n",
      "  labels, = index.labels\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ANREDE_KZ_1</th>\n",
       "      <th>ANREDE_KZ_2</th>\n",
       "      <th>CAMEO_DEUG_2015_1</th>\n",
       "      <th>CAMEO_DEUG_2015_2</th>\n",
       "      <th>CAMEO_DEUG_2015_3</th>\n",
       "      <th>CAMEO_DEUG_2015_4</th>\n",
       "      <th>CAMEO_DEUG_2015_5</th>\n",
       "      <th>CAMEO_DEUG_2015_6</th>\n",
       "      <th>CAMEO_DEUG_2015_7</th>\n",
       "      <th>CAMEO_DEUG_2015_8</th>\n",
       "      <th>...</th>\n",
       "      <th>MIN_GEBAEUDEJAHR</th>\n",
       "      <th>ANZ_KINDER</th>\n",
       "      <th>ANZ_STATISTISCHE_HAUSHALTE</th>\n",
       "      <th>EINGEZOGENAM_HH_JAHR</th>\n",
       "      <th>EXTSEL992</th>\n",
       "      <th>VERDICHTUNGSRAUM</th>\n",
       "      <th>VK_DHT4A</th>\n",
       "      <th>VK_DISTANZ</th>\n",
       "      <th>VK_ZG11</th>\n",
       "      <th>LNR</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.359845</td>\n",
       "      <td>-0.292527</td>\n",
       "      <td>-0.448734</td>\n",
       "      <td>-1.429498</td>\n",
       "      <td>0.823089</td>\n",
       "      <td>0.471363</td>\n",
       "      <td>-0.366474</td>\n",
       "      <td>-1.470506</td>\n",
       "      <td>-1.490312</td>\n",
       "      <td>9626.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.359845</td>\n",
       "      <td>-0.292527</td>\n",
       "      <td>-0.302057</td>\n",
       "      <td>0.492479</td>\n",
       "      <td>-0.642908</td>\n",
       "      <td>-0.395656</td>\n",
       "      <td>-0.000407</td>\n",
       "      <td>-0.505604</td>\n",
       "      <td>-1.113073</td>\n",
       "      <td>9628.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.359845</td>\n",
       "      <td>-0.292527</td>\n",
       "      <td>-0.448734</td>\n",
       "      <td>-1.133809</td>\n",
       "      <td>-1.042725</td>\n",
       "      <td>-0.519516</td>\n",
       "      <td>1.463863</td>\n",
       "      <td>1.745834</td>\n",
       "      <td>1.904842</td>\n",
       "      <td>143872.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.359845</td>\n",
       "      <td>-0.292527</td>\n",
       "      <td>-0.448734</td>\n",
       "      <td>-0.985965</td>\n",
       "      <td>-3.175084</td>\n",
       "      <td>-0.519516</td>\n",
       "      <td>-0.000407</td>\n",
       "      <td>-1.148872</td>\n",
       "      <td>-1.490312</td>\n",
       "      <td>143873.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.359845</td>\n",
       "      <td>-0.292527</td>\n",
       "      <td>-0.008701</td>\n",
       "      <td>-0.985965</td>\n",
       "      <td>0.023455</td>\n",
       "      <td>-0.395656</td>\n",
       "      <td>-1.098609</td>\n",
       "      <td>-0.827238</td>\n",
       "      <td>-0.735833</td>\n",
       "      <td>143874.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 686 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   ANREDE_KZ_1  ANREDE_KZ_2  CAMEO_DEUG_2015_1  CAMEO_DEUG_2015_2  \\\n",
       "0          1.0          0.0                1.0                0.0   \n",
       "1          1.0          0.0                0.0                1.0   \n",
       "2          0.0          1.0                0.0                0.0   \n",
       "3          1.0          0.0                0.0                0.0   \n",
       "4          1.0          0.0                0.0                0.0   \n",
       "\n",
       "   CAMEO_DEUG_2015_3  CAMEO_DEUG_2015_4  CAMEO_DEUG_2015_5  CAMEO_DEUG_2015_6  \\\n",
       "0                0.0                0.0                0.0                0.0   \n",
       "1                0.0                0.0                0.0                0.0   \n",
       "2                0.0                0.0                1.0                0.0   \n",
       "3                0.0                1.0                0.0                0.0   \n",
       "4                0.0                0.0                0.0                0.0   \n",
       "\n",
       "   CAMEO_DEUG_2015_7  CAMEO_DEUG_2015_8  ...  MIN_GEBAEUDEJAHR  ANZ_KINDER  \\\n",
       "0                0.0                0.0  ...         -0.359845   -0.292527   \n",
       "1                0.0                0.0  ...         -0.359845   -0.292527   \n",
       "2                0.0                0.0  ...         -0.359845   -0.292527   \n",
       "3                0.0                0.0  ...         -0.359845   -0.292527   \n",
       "4                1.0                0.0  ...         -0.359845   -0.292527   \n",
       "\n",
       "   ANZ_STATISTISCHE_HAUSHALTE  EINGEZOGENAM_HH_JAHR  EXTSEL992  \\\n",
       "0                   -0.448734             -1.429498   0.823089   \n",
       "1                   -0.302057              0.492479  -0.642908   \n",
       "2                   -0.448734             -1.133809  -1.042725   \n",
       "3                   -0.448734             -0.985965  -3.175084   \n",
       "4                   -0.008701             -0.985965   0.023455   \n",
       "\n",
       "   VERDICHTUNGSRAUM  VK_DHT4A  VK_DISTANZ   VK_ZG11       LNR  \n",
       "0          0.471363 -0.366474   -1.470506 -1.490312    9626.0  \n",
       "1         -0.395656 -0.000407   -0.505604 -1.113073    9628.0  \n",
       "2         -0.519516  1.463863    1.745834  1.904842  143872.0  \n",
       "3         -0.519516 -0.000407   -1.148872 -1.490312  143873.0  \n",
       "4         -0.395656 -1.098609   -0.827238 -0.735833  143874.0  \n",
       "\n",
       "[5 rows x 686 columns]"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "customers = clean_pipeline.transform(customers)\n",
    "azdias = pd.read_parquet('temp/azdias_processed.gzip')\n",
    "customers = pd.DataFrame(customers, columns=azdias.columns)\n",
    "customers.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Reduce the memory used by the dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T08:23:03.593475Z",
     "start_time": "2019-08-12T08:19:39.926187Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Memory usage of dataframe:  5.8488311767578125  MB\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4adfaed2754a45e7b53fbf1cd986cc9e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=4), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "___MEMORY USAGE AFTER COMPLETION:___\n",
      "Memory usage is: 3.8383216857910156 MB\n",
      "This is 65.63% of the initial size\n",
      "Memory usage of dataframe:  1003.0615386962891  MB\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c08c492a37a24373b0497576dc5bbb54",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=686), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "___MEMORY USAGE AFTER COMPLETION:___\n",
      "Memory usage is: 134.15589141845703 MB\n",
      "This is 13.37% of the initial size\n"
     ]
    }
   ],
   "source": [
    "customers_extra, _ = reduce_mem_usage(customers_extra)\n",
    "customers, _ = reduce_mem_usage(customers)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Put back the extra columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T08:23:04.801739Z",
     "start_time": "2019-08-12T08:23:04.217708Z"
    }
   },
   "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>ANREDE_KZ_1</th>\n",
       "      <th>ANREDE_KZ_2</th>\n",
       "      <th>CAMEO_DEUG_2015_1</th>\n",
       "      <th>CAMEO_DEUG_2015_2</th>\n",
       "      <th>CAMEO_DEUG_2015_3</th>\n",
       "      <th>CAMEO_DEUG_2015_4</th>\n",
       "      <th>CAMEO_DEUG_2015_5</th>\n",
       "      <th>CAMEO_DEUG_2015_6</th>\n",
       "      <th>CAMEO_DEUG_2015_7</th>\n",
       "      <th>CAMEO_DEUG_2015_8</th>\n",
       "      <th>...</th>\n",
       "      <th>EINGEZOGENAM_HH_JAHR</th>\n",
       "      <th>EXTSEL992</th>\n",
       "      <th>VERDICHTUNGSRAUM</th>\n",
       "      <th>VK_DHT4A</th>\n",
       "      <th>VK_DISTANZ</th>\n",
       "      <th>VK_ZG11</th>\n",
       "      <th>LNR</th>\n",
       "      <th>ONLINE_PURCHASE</th>\n",
       "      <th>PRODUCT_GROUP</th>\n",
       "      <th>CUSTOMER_GROUP</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.429498</td>\n",
       "      <td>0.823089</td>\n",
       "      <td>0.471363</td>\n",
       "      <td>-0.366474</td>\n",
       "      <td>-1.470505</td>\n",
       "      <td>-1.490312</td>\n",
       "      <td>9626</td>\n",
       "      <td>0</td>\n",
       "      <td>COSMETIC_AND_FOOD</td>\n",
       "      <td>MULTI_BUYER</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.492479</td>\n",
       "      <td>-0.642908</td>\n",
       "      <td>-0.395656</td>\n",
       "      <td>-0.000407</td>\n",
       "      <td>-0.505604</td>\n",
       "      <td>-1.113073</td>\n",
       "      <td>9628</td>\n",
       "      <td>0</td>\n",
       "      <td>FOOD</td>\n",
       "      <td>SINGLE_BUYER</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.133809</td>\n",
       "      <td>-1.042725</td>\n",
       "      <td>-0.519516</td>\n",
       "      <td>1.463863</td>\n",
       "      <td>1.745834</td>\n",
       "      <td>1.904842</td>\n",
       "      <td>143872</td>\n",
       "      <td>0</td>\n",
       "      <td>COSMETIC_AND_FOOD</td>\n",
       "      <td>MULTI_BUYER</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.985965</td>\n",
       "      <td>-3.175084</td>\n",
       "      <td>-0.519516</td>\n",
       "      <td>-0.000407</td>\n",
       "      <td>-1.148872</td>\n",
       "      <td>-1.490312</td>\n",
       "      <td>143873</td>\n",
       "      <td>0</td>\n",
       "      <td>COSMETIC</td>\n",
       "      <td>MULTI_BUYER</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.985965</td>\n",
       "      <td>0.023455</td>\n",
       "      <td>-0.395656</td>\n",
       "      <td>-1.098609</td>\n",
       "      <td>-0.827238</td>\n",
       "      <td>-0.735833</td>\n",
       "      <td>143874</td>\n",
       "      <td>0</td>\n",
       "      <td>FOOD</td>\n",
       "      <td>MULTI_BUYER</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 689 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   ANREDE_KZ_1  ANREDE_KZ_2  CAMEO_DEUG_2015_1  CAMEO_DEUG_2015_2  \\\n",
       "0            1            0                  1                  0   \n",
       "1            1            0                  0                  1   \n",
       "2            0            1                  0                  0   \n",
       "3            1            0                  0                  0   \n",
       "4            1            0                  0                  0   \n",
       "\n",
       "   CAMEO_DEUG_2015_3  CAMEO_DEUG_2015_4  CAMEO_DEUG_2015_5  CAMEO_DEUG_2015_6  \\\n",
       "0                  0                  0                  0                  0   \n",
       "1                  0                  0                  0                  0   \n",
       "2                  0                  0                  1                  0   \n",
       "3                  0                  1                  0                  0   \n",
       "4                  0                  0                  0                  0   \n",
       "\n",
       "   CAMEO_DEUG_2015_7  CAMEO_DEUG_2015_8  ...  EINGEZOGENAM_HH_JAHR  EXTSEL992  \\\n",
       "0                  0                  0  ...             -1.429498   0.823089   \n",
       "1                  0                  0  ...              0.492479  -0.642908   \n",
       "2                  0                  0  ...             -1.133809  -1.042725   \n",
       "3                  0                  0  ...             -0.985965  -3.175084   \n",
       "4                  1                  0  ...             -0.985965   0.023455   \n",
       "\n",
       "   VERDICHTUNGSRAUM  VK_DHT4A  VK_DISTANZ   VK_ZG11     LNR  ONLINE_PURCHASE  \\\n",
       "0          0.471363 -0.366474   -1.470505 -1.490312    9626                0   \n",
       "1         -0.395656 -0.000407   -0.505604 -1.113073    9628                0   \n",
       "2         -0.519516  1.463863    1.745834  1.904842  143872                0   \n",
       "3         -0.519516 -0.000407   -1.148872 -1.490312  143873                0   \n",
       "4         -0.395656 -1.098609   -0.827238 -0.735833  143874                0   \n",
       "\n",
       "       PRODUCT_GROUP  CUSTOMER_GROUP  \n",
       "0  COSMETIC_AND_FOOD     MULTI_BUYER  \n",
       "1               FOOD    SINGLE_BUYER  \n",
       "2  COSMETIC_AND_FOOD     MULTI_BUYER  \n",
       "3           COSMETIC     MULTI_BUYER  \n",
       "4               FOOD     MULTI_BUYER  \n",
       "\n",
       "[5 rows x 689 columns]"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "customers_exta = pd.merge(customers, customers_extra, how='inner', on='LNR')\n",
    "customers_exta.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T08:23:10.976045Z",
     "start_time": "2019-08-12T08:23:05.417682Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\radue\\Anaconda3\\envs\\udacity_ml_capstone\\lib\\site-packages\\pyarrow\\pandas_compat.py:114: FutureWarning: A future version of pandas will default to `skipna=True`. To silence this warning, pass `skipna=True|False` explicitly.\n",
      "  result = infer_dtype(pandas_collection)\n"
     ]
    }
   ],
   "source": [
    "customers_exta.to_parquet('temp/customers_processed.gzip', compression='gzip')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Apply PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T08:23:13.565818Z",
     "start_time": "2019-08-12T08:23:11.655197Z"
    }
   },
   "outputs": [],
   "source": [
    "customers = customers.drop('LNR', axis=1)\n",
    "customers_pca = pca.transform(customers)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Obtain clustering predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T08:23:14.429956Z",
     "start_time": "2019-08-12T08:23:14.211703Z"
    }
   },
   "outputs": [],
   "source": [
    "customers_labels = kmeans.predict(customers_pca)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T08:16:49.591925Z",
     "start_time": "2019-08-12T08:15:20.785Z"
    }
   },
   "source": [
    "Construct dictionary that maps cluster labels to frequency in customer population"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T08:23:44.864229Z",
     "start_time": "2019-08-12T08:23:44.829349Z"
    }
   },
   "outputs": [],
   "source": [
    "counts_customer = Counter(customers_labels)\n",
    "n_customers = customers_pca.shape[0]\n",
    "\n",
    "customer_freqs = {label: 100*(freq / n_customers) for label, freq in counts_customer.items()}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Construct dictionary that maps cluster labels to frequency in general population"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T08:24:00.296914Z",
     "start_time": "2019-08-12T08:24:00.151336Z"
    }
   },
   "outputs": [],
   "source": [
    "counts_population = Counter(population_labels)\n",
    "n_population = azdias_pca.shape[0]\n",
    "\n",
    "population_freqs = {label: 100*(freq / n_population) for label, freq in counts_population.items()}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Compare the proportion of data in each cluster for the customer data to the proportion of data in each cluster for the general population"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T08:31:17.357450Z",
     "start_time": "2019-08-12T08:31:17.349482Z"
    }
   },
   "outputs": [],
   "source": [
    "customer_clusters = pd.DataFrame.from_dict(customer_freqs, orient='index', columns=['% of data'])\n",
    "customer_clusters['Cluster'] = customer_clusters.index\n",
    "customer_clusters['DataSet'] = 'Customers Data'\n",
    "population_clusters = pd.DataFrame.from_dict(population_freqs, orient='index', columns=['% of data'])\n",
    "population_clusters['Cluster'] = population_clusters.index\n",
    "population_clusters['DataSet'] = 'General Population'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T08:34:28.341680Z",
     "start_time": "2019-08-12T08:34:28.333702Z"
    }
   },
   "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>% of data</th>\n",
       "      <th>Cluster</th>\n",
       "      <th>DataSet</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>22.314403</td>\n",
       "      <td>0</td>\n",
       "      <td>Customers Data</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>9.328888</td>\n",
       "      <td>1</td>\n",
       "      <td>Customers Data</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>23.881827</td>\n",
       "      <td>2</td>\n",
       "      <td>Customers Data</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>14.220045</td>\n",
       "      <td>3</td>\n",
       "      <td>Customers Data</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>3.389999</td>\n",
       "      <td>5</td>\n",
       "      <td>Customers Data</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.918853</td>\n",
       "      <td>4</td>\n",
       "      <td>Customers Data</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>24.361342</td>\n",
       "      <td>7</td>\n",
       "      <td>Customers Data</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1.584643</td>\n",
       "      <td>6</td>\n",
       "      <td>Customers Data</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>8.802194</td>\n",
       "      <td>7</td>\n",
       "      <td>General Population</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>13.934591</td>\n",
       "      <td>6</td>\n",
       "      <td>General Population</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10.089304</td>\n",
       "      <td>3</td>\n",
       "      <td>General Population</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>15.639443</td>\n",
       "      <td>0</td>\n",
       "      <td>General Population</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>4.955112</td>\n",
       "      <td>5</td>\n",
       "      <td>General Population</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>20.925113</td>\n",
       "      <td>2</td>\n",
       "      <td>General Population</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>13.462542</td>\n",
       "      <td>1</td>\n",
       "      <td>General Population</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>12.191701</td>\n",
       "      <td>4</td>\n",
       "      <td>General Population</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   % of data  Cluster             DataSet\n",
       "0  22.314403        0      Customers Data\n",
       "1   9.328888        1      Customers Data\n",
       "2  23.881827        2      Customers Data\n",
       "3  14.220045        3      Customers Data\n",
       "5   3.389999        5      Customers Data\n",
       "4   0.918853        4      Customers Data\n",
       "7  24.361342        7      Customers Data\n",
       "6   1.584643        6      Customers Data\n",
       "7   8.802194        7  General Population\n",
       "6  13.934591        6  General Population\n",
       "3  10.089304        3  General Population\n",
       "0  15.639443        0  General Population\n",
       "5   4.955112        5  General Population\n",
       "2  20.925113        2  General Population\n",
       "1  13.462542        1  General Population\n",
       "4  12.191701        4  General Population"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_clusters = pd.concat([customer_clusters, population_clusters])\n",
    "all_clusters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T08:39:35.331572Z",
     "start_time": "2019-08-12T08:39:34.939587Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdoAAAFgCAYAAAActbi8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAAgAElEQVR4nO3deZydZXn/8U+WSaKQsBk3RCNVLkTBaARcQBZxR4WfLbTBatSgVkFrxYgW2USwKaVa94ogIosKhSooIgZaFASNorJ4uRTaIhEhIAQLWZz8/rifgUOcTM5M5j5PZubzfr14Zc45z3PONSdDvnPf537ua9LatWuRJEl1TG67AEmSxjODVpKkigxaSZIqMmglSapoLAXtVGBO86ckSWPCWAqtJwA3L19+H/39rpSW1J7Zs2dOarsGjR1jaUQrSdKYY9BKklSRQStJUkVVP6ONiGOAg5qbF2fmoog4HdgD+ENz/3GZeUHNOiRJaku1oI2I/YCXAM8C1gKXRMSBwHOAF2bmslqvLUnSpqLmiHYZ8J7MXAUQETcBT2z+Oy0itgUuoIxo+yvWIUlSayb1oqlARDwV+B6wJ/AR4O3APcBFwDmZ+bkunmYOcHOtGiVpGLy8R12rfh1tRDwduBh4b2YmcGDHYx8HXg90E7QAeB2tpLbNnj2z7RI0hlRddRwRLwC+AxyZmWdExM4R8dqOQyYBq2vWIElSm2ouhtoOuBA4ODOXNHdPAj4aEUuA+4C3AGfUqkGSpLbVnDo+ApgBnBIRA/d9BjiJ8nltH3B+Zp5TsQZJklrVk8VQo2QO7nUsaRPgXscaDneGkiSpIoNWkqSKxlKbPEli5qwZzJjeN6JzH1i5mhX3PjDKFUlDM2gljSkzpvcxf9FZIzr37MWHsAKDVr3l1LEkSRUZtJIkVWTQSpJUkUErSVJFLoYax1ydKUntM2jHMVdnSlL7nDqWJKkig1aSpIoMWkmSKjJoJUmqyKCVJKkig1aSpIoMWkmSKjJoJUmqyKCVJKkig1aSpIoMWkmSKjJoJUmqyKCVJKkig1aSpIoMWkmSKjJoJUmqaEw3fp85awYzpvcN+7wHVq5mxb02NZck1Temg3bG9D7mLzpr2OedvfgQVmDQSpLqc+pYkqSKDFpJkioyaCVJqsiglSSpIoNWkqSKDFpJkioyaCVJqsiglSSpIoNWkqSKDFpJkioyaCVJqsiglSSpIoNWkqSKDFpJkioyaCVJqmhM96NVPf1rVjN79sxhn7dm1UruvmdVhYokaWwyaDWoyVP7WLp44bDPm7foVMCglaQBTh1LklSRQStJUkUGrSRJFRm0kiRVZNBKklSRQStJUkVVL++JiGOAg5qbF2fmoojYDzgFeATw5cw8qmYNkiS1qdqItgnUlwDPAuYC8yLir4DTgNcATwN2jYiX16pBkqS21Zw6Xga8JzNXZeZq4CZgB+CXmXlzZq4BvgT8RcUaJElqVbWp48y8YeDriHgqZQr545QAHrAMeMJwnnebbTYflfpGsr2guuN7q02ZP5/qtepbMEbE04GLgfcCayij2gGTgP7hPN/y5ffR378W2Lj/Ye64Y8WIzx0r2voHZSK8t2rPxv5cj8bPp2Gt4ai66jgiXgB8BzgyM88AbgUe13HIY4HbatYgSVKbqo1oI2I74ELg4Mxc0tx9TXkongLcDMynLI6SJGlcqjl1fAQwAzglIgbu+wywADi/eewbwHkVa5AkqVU1F0O9C3jXeh5+Zq3XlSRpU+LOUJIkVWTQSpJUkUErSVJFBq0kSRUZtJIkVWTQSpJUkUErSVJFBq0kSRUZtJIkVWTQSpJUkUErSVJFBq0kSRUZtJIkVWTQSpJUkUErSVJFBq0kSRUZtJIkVWTQSpJUkUErSVJFBq0kSRUZtJIkVWTQSpJUkUErSVJFBq0kSRUZtJIkVWTQSpJUkUErSVJFBq0kSRUZtJIkVWTQSpJUkUErSVJFBq0kSRUZtJIkVWTQSpJUkUErSVJFBq0kSRUZtJIkVWTQSpJUkUErSVJFBq0kSRUZtJIkVWTQSpJUkUErSVJFBq0kSRUZtJIkVWTQSpJU0dS2C2hD/5rVzJ49c0Tnrlm1krvvWTXKFUmSxqsJGbSTp/axdPHCEZ07b9GpgEErSeqOU8eSJFVk0EqSVFH1qeOImAVcBeyfmbdExOnAHsAfmkOOy8wLatchSVIbqgZtROwOfA7YoePu5wAvzMxlNV9bkqRNQe2p40OBdwC3AUTEI4EnAqdFxE8j4riIcPpakjRuVQ25zFyYmVd23PVYYAnwJuC5wJ7Am2vWIElSm3p6eU9m/hdw4MDtiPg48HrK9HJXttlm8wqVDc9Ir8GdKHx/tCnz51O91tOgjYidgR0y8/zmrknA6uE8x/Ll99HfvxZo73+YO+5Y0crrDpfvj8ajjf25Ho2fT8NawzGiqeOI2GyErzcJ+GhEbBURfcBbAFccS5LGrQ2OaCPiNcCHgM0oQTkF2BoY9q90mfnTiDgJ+B7QB5yfmecM93kkSRorupk6Phk4Cngb8A+Uz1jvHc6LZOacjq8/BXxqOOdLkjRWdTN1/IfM/DLwfeAB4G+A/atWJUnSONFN0D4QEdOBXwFzM7MfWFu3LEmSxodupo6/BlwMvAG4OiL2BO6sWpUkSePEBke0mXki8KbM/A1wAPCfwGtrFyZJ0njQzarjb2bmywEy80fAjyLi+5SdnaRhmTlrBjOm9w37vAdWrmbFvQ9UqEiS6lpv0EbEeZRmAH8WET/teKgPWFm7MI1PM6b3MX/RWcM+7+zFh7ACg1bS2DPUiPYIYA5le8TDO+5fA9xYsSZJksaN9QZtZt4C3BIR0aw0ftBG7AwlaQLZaotpTJ02fdjnrVm1krvvWVWhIqn3ull1/KqIOB7YnI3cGUrSxDJ12nSWLl447PPmLToVMGg1PnRzHe3JwInA/wBvBy4BPlOzKEmSxgt3hpIkqSJ3hpIkqSJ3hpIkqaINBm1mnhgRX8rM3zQt814I2NpulLk6U5LGp6E2rHjhOrfnNF/+GNgR+F29siYeV2dK0vg01Ij2k82fjwSeBNxA2axiZ8qGFXPrliZJ0ti33sVQmblzZu4M/BB4YWY+MzPnUfY4/nWvCpQkaSzrZtVxZOZVAzeaxgJPqVeSJEnjRzerju+PiAXAmZSdoRYCv69ZlCSpNyJib+BC4BbKv/FrgRMy87z1HL8F8LHMXDDEcz6SsrHRXMpugt8DDs/MQRvSdPOcY1k3I9o3Ae+kbFZxP7AAeGPFmiRJvXVZZs7NzGcCLwKOiYh56zl2K+AZG3i+vwV+nZm7NMf+Efi7IY7v5jnHrG4u77kJeHZEbN3cvqt6VZKkVmTm8og4GXhDs2/CXwKPAq4BXgd8BNgpIv4FOIbS4W07YDZwZGZ+BdgCmBwRkzOzv9kv/zEAEbEHZWvfGZQdB9/R+ZyZ+c4efrs90c2IFigBa8hK0oRwI/BUYC/g+c3X2wPzgCOBG5tA3B+4JDN3B14JHN2c/2ngAGBZRJwJzMvM6yJiGvCPwMsycy6lt/nr13nOcafroJUkTSj3U6aA3wJ8jBK0D2uRmplnAr+OiEXAcQOPN21Wd6LsKHgncFpEHAVEc/8VEXEd8NLmvnFtvUEbEQc0fw5/uyJJ0li2M2VdzqWUz1fPoEwdT+o8KCKOAN4N/BI4fuDxiDgR2CwzL8nMdwP7AfMpC6OWNp8HzwV2A07oyXfUoqFGtB9q/ry6F4VIktoXEY8BDqdsUnRtZn6OMip9HiUz1vDQ+p69KKuFL6BszzuQKY8HjoyIgWB+OmU6OoEdIuJpzWOfBw5e5znHnaG+sXsj4hfAthHx03UfbFaTSZLGvv2aqdx+ygj2BMpo9qKI+Bnls9RrgCcCVwKrIuJ04OPApyNiFfAToD8iHkEZ5X4CuDki7qcE7Fsz8/6IeCNwNjCdshjqC5SAXhURp2fmuLuqZaigfRnwLMpvHIf3phxJUi9l5hXAlut5eK/13L9bx9c7DPL4/cAh63m9b1OypdMf13nOcWW9QZuZK4D/jIhXArdRVpv1Adc0j0mSpA3oZk58C+AK4HbKB9lPiIj9O7dllCYiWxtK6kY3QftPwCGZeTlAROwLnEJpLiBNWLY2lNSNbq6jnTkQsgCZuYTSOk+SJG1AN0G7NiKeNHCjaQD/x2oVSZI0jnQzdXw88P2IuIzS1eGlwNurViVJ0jjRTVOBCyPiJmBfymKok5pGA5KkDqtWr/nttL6pj6nwvLdP65v62KGOiYhZwEmUS3LWAHcD72l6iA9LRBwK3JeZ54yk3hoi4hbg/ygLHKYBv6E0MVi6gfOOo3QnurJ2jevT1U4cmZmUC44lSesxrW/qY+YvOmvUn/fsxYcMGd4RMRn4BnA5MDcz10TEPsA3I2KnzFw+zJd8AeVqk03NK5p9lGkuPf1WROyYmXcOcc5elPelNeN2yytJmkD2oezadExm9gNk5uXNLkxTmubux2bm3gAR8QVKkP4bcA4wMFo+jjJqfDWwb0QsA66jbFz0RMpI+QOZeUlEHNvctwOlRd6HKb1sd6fsEvWXmbk2Io4EDqLMiH4LeB/wJOASytaO9wPvAf6VkkkPAG/MzF8O9Q1n5sURcS1lD+V/iYjDgL+mNDZYBfxVU8tzgFMj4kBg66bOR1I26Xh3Zv57l+/xiNm9R5LGvmcB1w2E7IDM/EZm/m6I8w4EbsnMecCbgT0z8zLga8DRmfktyjaLS5ptd/+c0olnYIS9M7A3pcPP6cA/UBq4PxvYJSJeRtnsaNemxm15aMeoAF6XmS+mbNn4T5n5HEp/224vH70e2LGZNj8A2DsznwFcBByWmV8EfggszMyfUXY5XJiZzwYW0qOGBgatJI19/ZSR4HBdBRwQERdSwvBDgxyzL2VES2b+F2XP492bx76dmWuA/waWZeaNze3fAFtRuvbsDiwFfkQZXT69Ofd3A9PAwMXAJyLi88A9lL2Qu7EWuD8z76WMbP8yIk4CXgVsPsjxrwOeEREfpIyiBztm1HUdtBGxX0RcHxG/jogFFWuSJA3PD4Fnd3TLAUq7uuaz2rU8vMVdH0AzPbsjcBawJ3Bt83lvp3VvT+Khjx07d15ZM0hdU4CPdrTF250ydQtlypimjvMoo+BrKaPbz6z/W32YXYAbI2I7Sqe5LYFvUhoVTBrk+CspeyovbeoY7JhRN1Q/2r517no3D/028v6aRUmShuVK4HfAMRExBSAiXgq8kdKe7k5g+4iYERFbU0KV5nPN4zLzq5TLNh8NzOLhbeuWUKaViYjtKQulum2fugT464jYPCKmAhdSpp8fJiK+DOyamZ8FPkgJ3SFFxKso09FfoYzGf5WZ/wz8gDIlPqU5dA0wtfm+dwCOpoTxazqOqWqoEe1/RMQrOm7fC7yWUtz9g58iSeq1zFxLWcD0Z8D1TWvT91FW6d6emTdQpmdvAL5KCWaALwLRtMK7EnhvZv4euAz4QET8OfBOysKon1GCcmFmLuuyrq8D51Omm6+nLKw6Y5BDTwT+PiJ+BCwG/mY9T/mNiLiuaen3LuClTZObS4HJEXEjZYr658CTm3MuoYyQd6RMgd8A3ATMBB4ZEZt1871sjKFWHb8COKG5nurvKG/231J6CP5F7cIkaaxZtXrN7Ru6FGekzzutb+iLRJpLXP56iMfftp6HXjnIsecC53bctf8gxxzb8fUtwJyO23t3fH0Cf7roaN3jf0IZla5XZs4Z4rH7gBev57GTgZObm1dR8mxATzZfGqpN3u+BwyJiF8qy6+8DJ2Smo1lJGsSGNpXYiOet8bTqkaE+o53STB0/GngJ8AtgSUS8tlfFSZI01g31Ge15lCXSC4BPZuaZlMB9XkRc0oPaJEka84aaj3hGZh7YrD6+DqD50PmIiNixJ9VJkjTGDRW0V0bEN4EZlG26HpSZP69alSRJ48RQi6HeFBE7Aysz8xc9rEmSpHFjyKVszd6QkqQu9K9e9dvJfdNG/fKe/tWrbp/cN63KimbV55pxSRolk/umPWbp4oWj/rzzFp26wfBudl56H2U/37WUXY/OoPQQXzvqRW24ngWUTf4XrHP/FcATgPsoWyBOBj6UmV8ZxdeeA1wx1LW3EbEb8NrMfF9EvBp4TmYePVo1dDJoJWl8+BTwGOB5mfn7pqPNBZRN+j/ZamV/amFmXgHQfET5g4j4Vmbe08MadqK8X2Tm1ygdi6qoGrTNX/RVwP6ZeUtE7AecAjwC+HJmHlXz9SVpIoiIJ1BGsts2mw2RmfdGxDtouuU0re0+C2xH6fbz/sy8rOkruy3wVEqf2FMz88PNnsn/SGmDNwX4Qmb+c9PbdnFz3/XAByhbG24JPL45ruuRYWb+LCLuA54SETdR2uQ9s6nx5Mz8YjM63p+yr8PjgK9Tuu/sxeB9dq/oeG+eQWn1t3lz/kmUXa+OBzaPiL+ndBvaOzMXRMRzgY9RFgLfCbw1M3/VjMSvpewTPRs4PDO/2c33WK1NXkTsDnyXsokzEfEI4DTKXslPA3aNiJfXen1JmkB2A27MzLs778zMn2fm+c3NjwGnNb1nXw18NiJmNo/tQtknYXfgyIjYEji0eY5nN8//mojYszl+B2DfzHwDpcH6OZn5XEp/2r+NiEd1W3jT/AAggWOB5U1P2X2BY5vdCQH2oGz/+3RKv9oDu3yJhZRdDXcF9gH+sfll5Gjga5k50E2IiJhGCeHDMvOZlD2Sz+l4rmmZ+TxKk52ue9nWHNEeCrwDOLO5vRvwy8y8GSAivkR507r6jUATW/+a1cyePXPDBw5izaqV3H3Pqg0fKI1tD34O2zQDOIoy6nygCZn9KE3Sj28O66M0IQC4PDNXAb+LiLuALZrj50bEvs0xm1OC9EYgB6Z5M/PkiNgnIo6gNH2fBmxoo/5Tm1HsVOAu4KDMvK95rTc3z3tnRPw7ZUR9L/DvmXl78/2dSwni87p4X94DvCwi3t/UP1QP2h2AuzPzB00NX42If42ILZrHBzZruh7YuovXBioGbWYuBIiIgbseD3R2fFhG+UB8WLbZpid9eoc00n/wa5kI9Uye2sdIF5nMW3Qqs2dPH+WKNs6m9ne2Kar1Ho3T9/6HwE4RMSsz7236u543sCioOWYKZRR6F0BEPI7SWu8AHt40fqB37RRgUWb+W3P8oygLmJ5LRwe3iPgnYHtKs/YLKQG9oT6vD35Gu46het+uWee4Naynz+46vgLcTZluPpcyAl+fwWZ5B94LeOh9Wvd1h9TLxVCT6fiNi1Jk/3CfZPny++jvL0/T1v8wd9yxYtSfc2O+l/XVsym+P5tiTSNV4+9svNkUf65H473f1MI6M/8nIs4EzoiINzaLoaZSPtf8Y3PYEkq3mhMiYidKW7w5QzztEuDQiPg6pWvbd4HBOgC9GHhbZl4VEa+kfN470j6vA71v39kE+wHA/6NMbb+sGVmupITl0XT02QUeSfn89NuD1LdjZt4WEW+Hspc/D++5OyCBbSJi18z8QUQcBPx3Zt7VMWgctl4G7a2UD7EHPBa4rYevL0lV9a9edXs3l+KM5Hkn903b0GFvp7SAu7wJkpmU0ezAWpjDgX9tetVOAl6XmSuGCJDPUBZI/ZiSFadn5hXNYqhOJwFnRsT9wP9SRtdPZmSOBz7V9L6dAnw4M3/UfE57B/AN4FHAlzLzWwARMdBn9xYe6rPb6VjguxHxAPCT5rgnUxY2HRsRH6H0ryUzV0bEwcAnmj61dwEHj/B7eVAvg/YaSoPhpwA3A/Mpi6MkaVyotalEFyFLZvZT+q6evJ7Hb2MDfWWb23M6br5zkOOvoHxuOnD7HB6+YKjTFwY5f+8/PezBx+6lrJ4ezM/XvSa3OWd9fXbnNI+fQrnaZTBP6fj6C83xV1MWha37Ont3fH0LQ88GPEy1VcfryswHKJ2Azqd8mP5zuvsgW5KkMav6iLbzt6PM/A7l+ihJkrqSmV9gkNHxWNGzEa0kSRORQStJUkXudTyKZs6awYzpg13GJUmaqAzaUTRjeh/zF501onPPXnzIKFcjSdoUOHUsSVJFBq0kSRUZtJIkVWTQSpJUkYuhJE0YI223aKtFbQyDVtKEMdJ2i/MWnQoYtBoZp44lSarIoJUkqSKDVpKkigxaSZIqMmglSarIoJUkqSKDVpKkigxaSZIqMmglSarIoJUkqSKDVpKkigxaSZIqMmglSarIoJUkqSKDVpKkigxaSZIqMmglSarIoJUkqSKDVpKkigxaSZIqMmglSarIoJUkqSKDVpKkigxaSZIqMmglSarIoJUkqSKDVpKkigxaSZIqMmglSarIoJUkqSKDVpKkigxaSZIqMmglSarIoJUkqSKDVpKkigxaSZIqMmglSarIoJUkqSKDVpKkigxaSZIqMmglSapoahsvGhGXA48GVjd3vTUzr2mjFkmSaup50EbEJGAH4EmZuabXry9JUi+1MXUczZ+XRsRPIuKwFmqQJKkn2pg63gr4DnA40AdcERGZmd/u5uRtttm8Zm1dmT17ZtslPIz1bNimVlONevrXrGby1L6en1vLRPg708TQ86DNzKuBqwduR8TngVcAXQXt8uX30d+/FmjvB/+OO1YMer/1FOurBzbNmkZqY76XWvUsXbxwROfOW3TqmHmPNoWfIUNXw9HzqeOI2CMiXtRx1yQeWhQlSdK40sbU8ZbA8RHxfMrU8RuAt7VQhyRJ1fV8RJuZFwEXAz8GlgKnNdPJkiSNO61cR5uZHwQ+2MZrS5LUS+4MJUlSRQatJEkVGbSSJFVk0EqSVJFBK0lSRQatJEkVGbSSJFVk0EqSVJFBK0lSRQatJEkVGbSSJFVk0EqSVJFBK0lSRa1075G0fjNnzWDG9L62y5A0SgxaaRMzY3of8xedNezzzl58SIVqJG0sp44lSarIoJUkqSKDVpKkigxaSZIqMmglSarIoJUkqSKDVpKkigxaSZIqMmglSarIoJUkqSKDVpKkigxaSZIqMmglSarIoJUkqSKDVpKkigxaSZIqMmglSarIoJUkqSKDVpKkiqa2XYCkTdvMWTOYMb2v7TKkMcuglTSkGdP7mL/orBGde/biQ0a5GmnscepYkqSKDFpJkioyaCVJqsiglSSpIoNWkqSKDFpJkiry8h5NaF4jKqk2g1YTmteISqrNqWNJkioyaCVJqsiglSSpIoNWkqSKDFpJkioyaCVJqqiVy3siYj5wFNAHfDQzP9lGHZIk1dbzEW1EbAt8GNgDmAu8JSJ26nUdkiT1Qhsj2v2AJZl5F0BEnAf8OXD8Bs6bAjB58qSH3fmorTYbURHTZm0zovMGq6HTSOuBkdc0luoB/87A96gbm3g9c4BbgTUjLkQTxqS1a9f29AUj4v3AZpl5VHN7IbBbZr5lA6fuAVxZuz5J6tKTgVvaLkKbvjZGtJOBznSfBPR3cd4PgD2BZcAfK9QlScNxa9sFaGxoI2hvpQTmgMcCt3Vx3krgu1UqkiSpkjaC9jLg2IiYDfwBeC2woWljSZLGpJ6vOs7M3wB/D1wOXAecnZnX9roOSZJ6oeeLoSRJmkjcGUqSpIoMWkmSKjJoJUmqyKCVJKkig1aSpIpa6d7Tpk2xc1BEzAKuAvbPzFtaruUY4KDm5sWZuajNegAi4njKfthrgc9n5iktlwRARJwMPCozF2wCtVwOPBpY3dz11sy8psV6XgUcA2wGXJqZ72qxloXAYR13PRk4MzMPW88p0qiaUJf3NJ2DvgvMo+w0dRXwV5l5Y4s17Q58DtgR2KHNoI2I/YDjgH0ooXYJ8InMvKDFmvaidHvam/LL0Y3AyzIz26qpqetFwLmUX0YWtFzLJMqOa0/KzNY3uY+I7Sn7ku8O3A4sAU7MzG+2WhgQEU8HLgSel5l3tl2PJoaJNnX8YOegzPwDMNA5qE2HAu+gu20oa1sGvCczV2XmauAm4IltFpSZ/wHs0wTIoymzMH9os6aI2JoS/ie2WUeHaP68NCJ+EhFtj9QOBL6cmbc2P0cHA62NrtfxaeADhqx6aaJNHT+eEiYDlgG7tVQLAJm5ECAiNnRodZl5w8DXEfFUyhTyC9qrqMjM1RFxHHAE8FXgNy2X9FnK7mbbtVzHgK2A7wCHU0b9V0REZua3W6rnKcCqiPga5Re1i4APtlTLg5oZm0dk5lfbrkUTy0Qb0Y60c9CE0kyvfRt4b2b+su16ADLzGGA2JdwObauO5vO+/83M77RVw7oy8+rMfH1m3tOM1D4PvKLFkqZSZo/eDDyPMoX8hhbrGfBWYJP4fF8Ty0QL2luBx3Xc7rZz0IQRES+gjI6OzMwzNoF6doyIuQCZ+X/AvwG7tFjSwcBLIuI64Hjg1RHxzy3WQ0Ts0XxmPGASDy2KasNvgcsy847MvB+4gJZnjiJiGrAX8LU269DENNGmju0cNISI2I6yUOTgzFzSdj2N7YHjImIPymzEa4DT2iomM1888HVELAD2zsx3t1VPY0vg+Ih4PmXq+A3A21qs5yLgjIjYElgBvJzyc9WmXYBfNGszpJ6aUCNaOwdt0BHADOCUiLiu+a/Nf7DJzG8AFwM/BpYCV2XmuW3WtKnJzIt4+Ht0WmZe3WI91wCLKSv8bwT+Gzi9rXoa22OjdrVkQl3eI0lSr02oEa0kSb1m0EqSVJFBK0lSRQatJEkVGbSSJFU00a6j1RgWEVOAdwHzKT+704CvA0dTtkW8PjNPHuFzXwrMdw9cSaPNEa3Gkk9TtvR7UWbOBXalbKh/6ig894s3fIgkDZ/X0WpMiIg5wA3A4zLz3o77H0tpfPAqmhFtRKwFZg+MTgduAw9QNk54KmWP66WU/W8/DywArqfsEdwPfIKyIX4fcG5mntjUcCWlq9EcYK/M7GxSIUl/whGtxop5wA2dIQuQmb/NzPO7fI4DgZkdo2GA7TPzjc3X+2Tm/wJnUnZXmkfZo3e/iDioOeYJwIcycwdDVlI3/IxWY0U/G/+L4XeBEyPiCkp3oo9m5q86D4iIzSibz28dER9q7t4cmAtcC6wBWtveUNLY44hWY8U1wNMiYmbnnRGxbURcDDxineMnNY9PG7gjM2+m9Eo9CZgFXBYRr1rnvCnNuc/PzKJrAwIAAADNSURBVLnN6Pe5PNTkfWXThF6SumLQakzIzNuAs4DTImIWQPPnp4DlwP0dh98BPKf5ev7AnRHxN5TPaC/NzPcB3wKe3Tz8R6CvmZr+PvB3zTlbAt+jdA2SpGEzaDWWvJ3SDeaqph/sNc3thesc907gkxHxI+BpwMBnqV+kjFhvjIilwBbAvzSPfRX4j4h4BiWcnxsRP2te45zMPKvetyVpPHPVsSRJFTmilSSpIoNWkqSKDFpJkioyaCVJqsiglSSpIoNWkqSKDFpJkir6/6F3SyVDKdMJAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 494.225x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.catplot(x='Cluster', y='% of data', hue='DataSet', data=all_clusters, kind='bar')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T09:31:25.003544Z",
     "start_time": "2019-08-12T09:31:24.711983Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Proportion of general population in cluster 7: 8.802193844175575%\n",
      "Proportion of customer population in cluster 7: 24.3613424331601%\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# What kinds of people are part of a cluster that is overrepresented in the\n",
    "# customer data compared to the general population?\n",
    "\n",
    "over_cluster = 7\n",
    "\n",
    "print(f'Proportion of general population in cluster {over_cluster}: {population_freqs[over_cluster]}%')\n",
    "print(f'Proportion of customer population in cluster {over_cluster}: {customer_freqs[over_cluster]}%\\n')\n",
    "\n",
    "pc1_avg = (azdias_pca[population_labels == over_cluster][0].sum() + customers_pca[customers_labels == over_cluster][0].sum()) / \\\n",
    "          (n_population + n_customers)\n",
    "\n",
    "pc2_avg = (azdias_pca[population_labels == over_cluster][1].sum() + customers_pca[customers_labels == over_cluster][1].sum()) / \\\n",
    "          (n_population + n_customers)\n",
    "\n",
    "pc3_avg = (azdias_pca[population_labels == over_cluster][2].sum() + customers_pca[customers_labels == over_cluster][2].sum()) / \\\n",
    "          (n_population + n_customers)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T10:01:45.211303Z",
     "start_time": "2019-08-12T10:01:44.540808Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# This line gets the index of the most important PCA feature for over-represented cluster\n",
    "pca_feat_max = pd.Series(kmeans.cluster_centers_[over_cluster]).sort_values(ascending=False).index[0]\n",
    "pca_feat_min = pd.Series(kmeans.cluster_centers_[over_cluster]).sort_values(ascending=True).index[0]\n",
    "\n",
    "# This line plots the PCA components weights against one another\n",
    "fig = comp_weights(pca, pca_feat_max, pca_feat_min, 5, customers)\n",
    "plt.suptitle(f'Group {over_cluster}: Popular with company');\n",
    "plt.savefig('temp/over_cluster.png', dpi=fig.dpi, bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T10:58:06.429362Z",
     "start_time": "2019-08-12T10:58:06.408418Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Popular with company\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>SEMIO_VERT</th>\n",
       "      <th>SEMIO_SOZ</th>\n",
       "      <th>SEMIO_KULT</th>\n",
       "      <th>SEMIO_FAM</th>\n",
       "      <th>ANREDE_KZ_1</th>\n",
       "      <th>ANREDE_KZ_2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.093707</td>\n",
       "      <td>2.018105</td>\n",
       "      <td>2.987535</td>\n",
       "      <td>5.889674</td>\n",
       "      <td>0.419833</td>\n",
       "      <td>0.580167</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   SEMIO_VERT  SEMIO_SOZ  SEMIO_KULT  SEMIO_FAM  ANREDE_KZ_1  ANREDE_KZ_2\n",
       "0    1.093707   2.018105    2.987535   5.889674     0.419833     0.580167"
      ]
     },
     "execution_count": 168,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pca_inverse = pca.inverse_transform(kmeans.cluster_centers_[over_cluster,:])\n",
    "df = pd.DataFrame(pca_inverse[np.newaxis], columns=customers.columns)\n",
    "\n",
    "scaler = clean_pipeline.named_steps['transform_pipeline'].transformers_[2][1].named_steps['std_scaler']\n",
    "df[numeric_features] = scaler.inverse_transform(df[numeric_features])\n",
    "\n",
    "popular_features = ['SEMIO_VERT', 'SEMIO_SOZ', 'SEMIO_KULT', 'SEMIO_FAM', 'ANREDE_KZ_1']\n",
    "print('Popular with company')\n",
    "df[popular_features]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T09:56:50.740569Z",
     "start_time": "2019-08-12T09:56:50.405430Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Proportion of general population in cluster 4: 12.191701048337057%\n",
      "Proportion of customer population in cluster 4: 0.9188529209191659%\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# What kinds of people are part of a cluster that is underrepresented in the\n",
    "# customer data compared to the general population?\n",
    "\n",
    "under_cluster = 4\n",
    "\n",
    "print(f'Proportion of general population in cluster {under_cluster}: {population_freqs[under_cluster]}%')\n",
    "print(f'Proportion of customer population in cluster {under_cluster}: {customer_freqs[under_cluster]}%\\n')\n",
    "\n",
    "pc1_avg = (azdias_pca[population_labels == under_cluster][0].sum() + customers_pca[customers_labels == under_cluster][0].sum()) / \\\n",
    "          (n_population + n_customers)\n",
    "\n",
    "pc2_avg = (azdias_pca[population_labels == under_cluster][1].sum() + customers_pca[customers_labels == under_cluster][1].sum()) / \\\n",
    "          (n_population + n_customers)\n",
    "\n",
    "pc3_avg = (azdias_pca[population_labels == under_cluster][2].sum() + customers_pca[customers_labels == under_cluster][2].sum()) / \\\n",
    "          (n_population + n_customers)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T10:02:01.406442Z",
     "start_time": "2019-08-12T10:02:00.744242Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# This line gets the index of the most important PCA feature for over-represented cluster\n",
    "pca_feat_max = pd.Series(kmeans.cluster_centers_[under_cluster]).sort_values(ascending=False).index[0]\n",
    "pca_feat_min = pd.Series(kmeans.cluster_centers_[under_cluster]).sort_values(ascending=True).index[0]\n",
    "\n",
    "# This line plots the PCA components weights against one another\n",
    "fig = comp_weights(pca, pca_feat_max, pca_feat_min, 5, customers)\n",
    "plt.suptitle(f'Group {over_cluster}: Unopular with company');\n",
    "plt.savefig('temp/under_cluster.png', dpi=fig.dpi, bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-08-12T10:05:42.098884Z",
     "start_time": "2019-08-12T10:05:42.063980Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Unpopular with company\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>D19_GESAMT_ONLINE_QUOTE_12</th>\n",
       "      <th>D19_VERSAND_ONLINE_QUOTE_12</th>\n",
       "      <th>ONLINE_AFFINITAET</th>\n",
       "      <th>D19_BANKEN_ONLINE_QUOTE_12</th>\n",
       "      <th>D19_GESAMT_ANZ_12</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.088116</td>\n",
       "      <td>0.013217</td>\n",
       "      <td>2.217593</td>\n",
       "      <td>0.012309</td>\n",
       "      <td>1.037286</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   D19_GESAMT_ONLINE_QUOTE_12  D19_VERSAND_ONLINE_QUOTE_12  ONLINE_AFFINITAET  \\\n",
       "0                    0.088116                     0.013217           2.217593   \n",
       "\n",
       "   D19_BANKEN_ONLINE_QUOTE_12  D19_GESAMT_ANZ_12  \n",
       "0                    0.012309           1.037286  "
      ]
     },
     "execution_count": 167,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pca_inverse = pca.inverse_transform(kmeans.cluster_centers_[under_cluster,:])\n",
    "df = pd.DataFrame(pca_inverse[np.newaxis], columns=customers.columns)\n",
    "\n",
    "scaler = clean_pipeline.named_steps['transform_pipeline'].transformers_[2][1].named_steps['std_scaler']\n",
    "df[numeric_features] = scaler.inverse_transform(df[numeric_features])\n",
    "\n",
    "unpopular_features = ['D19_GESAMT_ONLINE_QUOTE_12', \n",
    "                      'D19_VERSAND_ONLINE_QUOTE_12', \n",
    "                      'ONLINE_AFFINITAET', \n",
    "                      'D19_BANKEN_ONLINE_QUOTE_12', \n",
    "                      'D19_GESAMT_ANZ_12']\n",
    "print('Unpopular with company')\n",
    "df[unpopular_features]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Results - Compare Customer Data to Demographics Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Popular with company - Cluster 7**\n",
    "\n",
    "* SEMIO_VERT (1) # affinity indicating in what way the person is dreamily (highest affinity)\n",
    "* SEMIO_SOZ (2) # affinity indicating in what way the person is social minded (very high affinity)\n",
    "* SEMIO_KULT (3) # affinity indicating in what way the person is cultural minded (high affinity)\n",
    "* SEMIO_FAM (6) # affinity indicating in what way the person is familiar minded (very low affinity)\n",
    "* ANREDE_KZ_1 (0) # gender: female\n",
    "\n",
    "**Unpopular with company - Cluster 4**\n",
    "\n",
    "* D19_GESAMT_ONLINE_QUOTE_12 (0) # amount of online transactions within all transactions in the complete file: no Online-transactions within the last 12 months\n",
    "* D19_VERSAND_ONLINE_QUOTE_12 (0) # amount of online transactions within all transactions in the segment mail-order: no Online-transactions within the last 12 months\n",
    "* ONLINE_AFFINITAET (2) # online affinity: (average affinity)\n",
    "* D19_BANKEN_ONLINE_QUOTE_12 (0) # amount of online transactions within all transactions in the segment bank: no Online-transactions within the last 12 months\n",
    "* D19_GESAMT_ANZ_12 (1) # transaction activity TOTAL POOL in the last 12 months: very low activity"
   ]
  },
  {
   "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": "3.7.3"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": true,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {
    "height": "calc(100% - 180px)",
    "left": "10px",
    "top": "150px",
    "width": "393.188px"
   },
   "toc_section_display": true,
   "toc_window_display": true
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}