Koneoppimisen malleja.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.metrics import (accuracy_score,confusion_matrix)\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn import tree\n",
    "from sklearn.cluster import KMeans\n",
    "import pydotplus\n",
    "from IPython.display import Image\n",
    "%matplotlib inline\n",
    "\n",
    "#Vaikuttaa kaavioiden ulkoasuun:\n",
    "sns.set()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Viini</th>\n",
       "      <th>X1</th>\n",
       "      <th>X2</th>\n",
       "      <th>X3</th>\n",
       "      <th>X4</th>\n",
       "      <th>X5</th>\n",
       "      <th>X6</th>\n",
       "      <th>X7</th>\n",
       "      <th>X8</th>\n",
       "      <th>X9</th>\n",
       "      <th>X10</th>\n",
       "      <th>X11</th>\n",
       "      <th>X12</th>\n",
       "      <th>X13</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>14.23</td>\n",
       "      <td>1.71</td>\n",
       "      <td>2.43</td>\n",
       "      <td>15.6</td>\n",
       "      <td>127</td>\n",
       "      <td>2.80</td>\n",
       "      <td>3.06</td>\n",
       "      <td>0.28</td>\n",
       "      <td>2.29</td>\n",
       "      <td>5.64</td>\n",
       "      <td>1.04</td>\n",
       "      <td>3.92</td>\n",
       "      <td>1065</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>13.20</td>\n",
       "      <td>1.78</td>\n",
       "      <td>2.14</td>\n",
       "      <td>11.2</td>\n",
       "      <td>100</td>\n",
       "      <td>2.65</td>\n",
       "      <td>2.76</td>\n",
       "      <td>0.26</td>\n",
       "      <td>1.28</td>\n",
       "      <td>4.38</td>\n",
       "      <td>1.05</td>\n",
       "      <td>3.40</td>\n",
       "      <td>1050</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>13.16</td>\n",
       "      <td>2.36</td>\n",
       "      <td>2.67</td>\n",
       "      <td>18.6</td>\n",
       "      <td>101</td>\n",
       "      <td>2.80</td>\n",
       "      <td>3.24</td>\n",
       "      <td>0.30</td>\n",
       "      <td>2.81</td>\n",
       "      <td>5.68</td>\n",
       "      <td>1.03</td>\n",
       "      <td>3.17</td>\n",
       "      <td>1185</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>14.37</td>\n",
       "      <td>1.95</td>\n",
       "      <td>2.50</td>\n",
       "      <td>16.8</td>\n",
       "      <td>113</td>\n",
       "      <td>3.85</td>\n",
       "      <td>3.49</td>\n",
       "      <td>0.24</td>\n",
       "      <td>2.18</td>\n",
       "      <td>7.80</td>\n",
       "      <td>0.86</td>\n",
       "      <td>3.45</td>\n",
       "      <td>1480</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>13.24</td>\n",
       "      <td>2.59</td>\n",
       "      <td>2.87</td>\n",
       "      <td>21.0</td>\n",
       "      <td>118</td>\n",
       "      <td>2.80</td>\n",
       "      <td>2.69</td>\n",
       "      <td>0.39</td>\n",
       "      <td>1.82</td>\n",
       "      <td>4.32</td>\n",
       "      <td>1.04</td>\n",
       "      <td>2.93</td>\n",
       "      <td>735</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Viini     X1    X2    X3    X4   X5    X6    X7    X8    X9   X10   X11  \\\n",
       "0      1  14.23  1.71  2.43  15.6  127  2.80  3.06  0.28  2.29  5.64  1.04   \n",
       "1      1  13.20  1.78  2.14  11.2  100  2.65  2.76  0.26  1.28  4.38  1.05   \n",
       "2      1  13.16  2.36  2.67  18.6  101  2.80  3.24  0.30  2.81  5.68  1.03   \n",
       "3      1  14.37  1.95  2.50  16.8  113  3.85  3.49  0.24  2.18  7.80  0.86   \n",
       "4      1  13.24  2.59  2.87  21.0  118  2.80  2.69  0.39  1.82  4.32  1.04   \n",
       "\n",
       "    X12   X13  \n",
       "0  3.92  1065  \n",
       "1  3.40  1050  \n",
       "2  3.17  1185  \n",
       "3  3.45  1480  \n",
       "4  2.93   735  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df=pd.read_excel('http://taanila.fi/wine.xlsx')\n",
    "df.rename(columns={'Y':'Viini'}, inplace=True)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Tehdään fature-matriisi,jota varten pudotetaan Y-muuttuja\n",
    "X = df.drop('Viini', axis=1)\n",
    "\n",
    "#Tehdään target eli viini\n",
    "y = df['Viini']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "#jaetaan data testidataan ja opetusdataan\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Sovitetaan opetusdata ja testidata K-nearest neighbors-malliin (KNN)\n",
    "malli = KNeighborsClassifier(n_neighbors=1)\n",
    "malli.fit(X_train, y_train)\n",
    "\n",
    "#Mallin mukaisten ennusteiden laskeminen opetusdatalle ja testidatalle:\n",
    "y_train_malli = malli.predict(X_train)\n",
    "y_test_malli = malli.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Oikeaan osuneiden ennusteiden osuus opetusdatassa = 1.0\n",
      "Confusion-matriisi opetusdatalle = \n",
      " [[40  0  0]\n",
      " [ 0 57  0]\n",
      " [ 0  0 36]]\n",
      "\n",
      "Oikeaan osuneiden ennusteiden osuus testidatassa = 0.7555555555555555\n",
      "Confusion-matriisi testidatalle = \n",
      " [[15  2  2]\n",
      " [ 1 12  1]\n",
      " [ 1  4  7]]\n"
     ]
    }
   ],
   "source": [
    "#Oikeaan osuneiden ennusteiden osuus datoissa\n",
    "as_train = accuracy_score(y_train, y_train_malli)\n",
    "as_test = accuracy_score(y_test, y_test_malli)\n",
    "\n",
    "#Confusion-matriisit datoille\n",
    "cm_train = confusion_matrix(y_train, y_train_malli)\n",
    "cm_test = confusion_matrix(y_test, y_test_malli)\n",
    "\n",
    "print('Oikeaan osuneiden ennusteiden osuus opetusdatassa = {}'.format(as_train))\n",
    "print('Confusion-matriisi opetusdatalle = \\n {}'.format(cm_train) + \"\\n\")\n",
    "print('Oikeaan osuneiden ennusteiden osuus testidatassa = {}'.format(as_test))\n",
    "print('Confusion-matriisi testidatalle = \\n {}'.format(cm_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Kun naapureiden lukumääräksi määritellään 1, osuu on KNN-malli täydellisesti opetusdatan osalta. Testidatassa se kuitenkin osuu \n",
    "vain 75%"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Uusi dataframe\n",
    "df2=pd.read_excel('http://taanila.fi/wine.xlsx')\n",
    "df2.rename(columns={'Y':'Viini'}, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Tehdään fature-matriisi,jota varten pudotetaan Y-muuttuja\n",
    "X = df2.drop('Viini', axis=1)\n",
    "\n",
    "#Tehdään target eli viini\n",
    "y = df2['Viini']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "#jaetaan data uudestaan testidataan ja opetusdataan\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Sovitetaan opetusdata ja testidata Gaussian Naive Bayes-maliin\n",
    "malli = GaussianNB()\n",
    "malli.fit(X_train, y_train)\n",
    "\n",
    "#Mallin mukaisten ennusteiden laskeminen opetusdatalle ja testidatalle:\n",
    "y_train_malli = malli.predict(X_train)\n",
    "y_test_malli = malli.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Oikeaan osuneiden ennusteiden osuus opetusdatassa = 0.9924812030075187\n",
      "Confusion-matriisi opetusdatalle = \n",
      " [[40  0  0]\n",
      " [ 0 56  1]\n",
      " [ 0  0 36]]\n",
      "\n",
      "Oikeaan osuneiden ennusteiden osuus testidatassa = 0.9333333333333333\n",
      "Confusion-matriisi testidatalle = \n",
      " [[16  3  0]\n",
      " [ 0 14  0]\n",
      " [ 0  0 12]]\n"
     ]
    }
   ],
   "source": [
    "#Oikeaan osuneiden ennusteiden osuus datoissa\n",
    "as_train = accuracy_score(y_train, y_train_malli)\n",
    "as_test = accuracy_score(y_test, y_test_malli)\n",
    "\n",
    "#Confusion-matriisit datoille\n",
    "cm_train = confusion_matrix(y_train, y_train_malli)\n",
    "cm_test = confusion_matrix(y_test, y_test_malli)\n",
    "\n",
    "print('Oikeaan osuneiden ennusteiden osuus opetusdatassa = {}'.format(as_train))\n",
    "print('Confusion-matriisi opetusdatalle = \\n {}'.format(cm_train) + \"\\n\")\n",
    "print('Oikeaan osuneiden ennusteiden osuus testidatassa = {}'.format(as_test))\n",
    "print('Confusion-matriisi testidatalle = \\n {}'.format(cm_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Gaussian Naive Bayes-malli onnistuu ennustamaan opetusdatan lähes täydellisesti ja testidatan melko hyvin: 93% "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Uusi dataframe\n",
    "df3=pd.read_excel('http://taanila.fi/wine.xlsx')\n",
    "df3.rename(columns={'Y':'Viini'}, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Tehdään fature-matriisi,jota varten pudotetaan Y-muuttuja\n",
    "X = df3.drop('Viini', axis=1)\n",
    "\n",
    "#Tehdään target eli viini\n",
    "y = df3['Viini']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "#jaetaan data uudestaan testidataan ja opetusdataan\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Oem\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:433: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\Oem\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:460: FutureWarning: Default multi_class will be changed to 'auto' in 0.22. Specify the multi_class option to silence this warning.\n",
      "  \"this warning.\", FutureWarning)\n"
     ]
    }
   ],
   "source": [
    "#Sovitetaan opetusdata ja testidata logistisen regression malliin\n",
    "malli = LogisticRegression()\n",
    "malli.fit(X_train, y_train)\n",
    "\n",
    "#Mallin mukaisten ennusteiden laskeminen opetusdatalle ja testidatalle:\n",
    "y_train_malli = malli.predict(X_train)\n",
    "y_test_malli = malli.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Oikeaan osuneiden ennusteiden osuus opetusdatassa = 0.9849624060150376\n",
      "Confusion-matriisi opetusdatalle = \n",
      " [[39  1  0]\n",
      " [ 0 56  1]\n",
      " [ 0  0 36]]\n",
      "\n",
      "Oikeaan osuneiden ennusteiden osuus testidatassa = 0.9111111111111111\n",
      "Confusion-matriisi testidatalle = \n",
      " [[18  1  0]\n",
      " [ 1 12  1]\n",
      " [ 0  1 11]]\n"
     ]
    }
   ],
   "source": [
    "#Oikeaan osuneiden ennusteiden osuus datoissa\n",
    "as_train = accuracy_score(y_train, y_train_malli)\n",
    "as_test = accuracy_score(y_test, y_test_malli)\n",
    "\n",
    "#Confusion-matriisit datoille\n",
    "cm_train = confusion_matrix(y_train, y_train_malli)\n",
    "cm_test = confusion_matrix(y_test, y_test_malli)\n",
    "\n",
    "print('Oikeaan osuneiden ennusteiden osuus opetusdatassa = {}'.format(as_train))\n",
    "print('Confusion-matriisi opetusdatalle = \\n {}'.format(cm_train) + \"\\n\")\n",
    "print('Oikeaan osuneiden ennusteiden osuus testidatassa = {}'.format(as_test))\n",
    "print('Confusion-matriisi testidatalle = \\n {}'.format(cm_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Logistinen regressio onnistuu ennustamaan opetusdatan lähes täydellisesti ja testidatan melko hyvin: 91% "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[[0.9953236969091906, 0.003901048256518427, 0.0007752548342909642],\n",
       " [0.9973359661712653, 0.0020359819078957465, 0.0006280519208389809],\n",
       " [0.998412647276359, 0.0002656066409693722, 0.0013217460826716399],\n",
       " [0.9959925360896443, 1.6826052546960613e-07, 0.004007295649830306],\n",
       " [0.080899932964534, 0.9114567256185591, 0.007643341416906983]]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X1 = df3.drop('Viini', axis=1).head()\n",
    "malli.predict_proba(X1).tolist()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ensimmäinen rivi on 99,532 % todennäköisyydellä viini1,  0,390 % todennäköisyydellä viini2 ja 0,0775 % todennäköisyydellä viini3."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Uusi dataframe\n",
    "df4=pd.read_excel('http://taanila.fi/wine.xlsx')\n",
    "df4.rename(columns={'Y':'Viini'}, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Tehdään fature-matriisi,jota varten pudotetaan Y-muuttuja\n",
    "X = df4.drop('Viini', axis=1)\n",
    "\n",
    "#Tehdään target eli viini\n",
    "y = df4['Viini']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "#jaetaan data uudestaan testidataan ja opetusdataan\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Sovitetaan opetusdata ja testidata päätöspuu-malliin\n",
    "malli = tree.DecisionTreeClassifier(max_depth=3)\n",
    "malli.fit(X_train, y_train)\n",
    "\n",
    "#Mallin mukaisten ennusteiden laskeminen opetusdatalle ja testidatalle:\n",
    "y_train_malli = malli.predict(X_train)\n",
    "y_test_malli = malli.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Oikeaan osuneiden ennusteiden osuus opetusdatassa = 1.0\n",
      "Confusion-matriisi opetusdatalle = \n",
      " [[40  0  0]\n",
      " [ 0 57  0]\n",
      " [ 0  0 36]]\n",
      "\n",
      "Oikeaan osuneiden ennusteiden osuus testidatassa = 0.9111111111111111\n",
      "Confusion-matriisi testidatalle = \n",
      " [[17  2  0]\n",
      " [ 1 12  1]\n",
      " [ 0  0 12]]\n"
     ]
    }
   ],
   "source": [
    "#Oikeaan osuneiden ennusteiden osuus datoissa\n",
    "as_train = accuracy_score(y_train, y_train_malli)\n",
    "as_test = accuracy_score(y_test, y_test_malli)\n",
    "\n",
    "#Confusion-matriisit datoille\n",
    "cm_train = confusion_matrix(y_train, y_train_malli)\n",
    "cm_test = confusion_matrix(y_test, y_test_malli)\n",
    "\n",
    "print('Oikeaan osuneiden ennusteiden osuus opetusdatassa = {}'.format(as_train))\n",
    "print('Confusion-matriisi opetusdatalle = \\n {}'.format(cm_train) + \"\\n\")\n",
    "print('Oikeaan osuneiden ennusteiden osuus testidatassa = {}'.format(as_test))\n",
    "print('Confusion-matriisi testidatalle = \\n {}'.format(cm_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Max depth 3 antaa antaa jo parhaimman tuloksen mihin malli pystyy. Päätöspuu ennustaa opetusdatan täydellisesti ja testi datan melko hyvin: 91%"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Päätösåuun graafinen esittäminen\n",
    "dot_data = tree.export_graphviz(malli, out_file=None, \n",
    "                         feature_names=X.columns,\n",
    "                         class_names=['Viini1', 'Viini2', 'Viini3'],\n",
    "                         filled=True, rounded=True,  \n",
    "                         special_characters=True)  \n",
    "graph = pydotplus.graph_from_dot_data(dot_data)  \n",
    "Image(graph.create_png())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Uusi dataframe\n",
    "df5=pd.read_excel('http://taanila.fi/wine.xlsx')\n",
    "df5.rename(columns={'Y':'Viini'}, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Tehdään fature-matriisi,jota varten pudotetaan Y-muuttuja\n",
    "X = df5.drop('Viini', axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1.25166667e+01, 2.49420290e+00, 2.28855072e+00, 2.08231884e+01,\n",
       "        9.23478261e+01, 2.07072464e+00, 1.75840580e+00, 3.90144928e-01,\n",
       "        1.45188406e+00, 4.08695651e+00, 9.41159420e-01, 2.49072464e+00,\n",
       "        4.58231884e+02],\n",
       "       [1.38044681e+01, 1.88340426e+00, 2.42617021e+00, 1.70234043e+01,\n",
       "        1.05510638e+02, 2.86723404e+00, 3.01425532e+00, 2.85319149e-01,\n",
       "        1.91042553e+00, 5.70255319e+00, 1.07829787e+00, 3.11404255e+00,\n",
       "        1.19514894e+03],\n",
       "       [1.29298387e+01, 2.50403226e+00, 2.40806452e+00, 1.98903226e+01,\n",
       "        1.03596774e+02, 2.11112903e+00, 1.58403226e+00, 3.88387097e-01,\n",
       "        1.50338710e+00, 5.65032258e+00, 8.83967742e-01, 2.36548387e+00,\n",
       "        7.28338710e+02]])"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Sovitetaan opetusdata ja testidata K-means clustering-malliin\n",
    "malli = KMeans(n_clusters=3)\n",
    "malli.fit(X)\n",
    "\n",
    "#Ryhmien keskukset (sepal_length, sepal_width, petal_length, petal_width):\n",
    "malli.cluster_centers_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>col_0</th>\n",
       "      <th>lkm</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>K</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>69</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>62</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "col_0  lkm\n",
       "K         \n",
       "0       69\n",
       "1       47\n",
       "2       62"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Ryhmiin kuulumiset:\n",
    "X['K'] = malli.predict(X)\n",
    "pd.crosstab(X['K'], 'lkm')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "</style>  \n",
       "<table id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4\" > \n",
       "<thead>    <tr> \n",
       "        <th class=\"blank level0\" ></th> \n",
       "        <th class=\"col_heading level0 col0\" colspan=8>X1</th> \n",
       "        <th class=\"col_heading level0 col8\" colspan=8>X10</th> \n",
       "        <th class=\"col_heading level0 col16\" colspan=8>X11</th> \n",
       "        <th class=\"col_heading level0 col24\" colspan=8>X12</th> \n",
       "        <th class=\"col_heading level0 col32\" colspan=8>X13</th> \n",
       "        <th class=\"col_heading level0 col40\" colspan=8>X2</th> \n",
       "        <th class=\"col_heading level0 col48\" colspan=8>X3</th> \n",
       "        <th class=\"col_heading level0 col56\" colspan=8>X4</th> \n",
       "        <th class=\"col_heading level0 col64\" colspan=8>X5</th> \n",
       "        <th class=\"col_heading level0 col72\" colspan=8>X6</th> \n",
       "        <th class=\"col_heading level0 col80\" colspan=8>X7</th> \n",
       "        <th class=\"col_heading level0 col88\" colspan=8>X8</th> \n",
       "        <th class=\"col_heading level0 col96\" colspan=8>X9</th> \n",
       "    </tr>    <tr> \n",
       "        <th class=\"blank level1\" ></th> \n",
       "        <th class=\"col_heading level1 col0\" >count</th> \n",
       "        <th class=\"col_heading level1 col1\" >mean</th> \n",
       "        <th class=\"col_heading level1 col2\" >std</th> \n",
       "        <th class=\"col_heading level1 col3\" >min</th> \n",
       "        <th class=\"col_heading level1 col4\" >25%</th> \n",
       "        <th class=\"col_heading level1 col5\" >50%</th> \n",
       "        <th class=\"col_heading level1 col6\" >75%</th> \n",
       "        <th class=\"col_heading level1 col7\" >max</th> \n",
       "        <th class=\"col_heading level1 col8\" >count</th> \n",
       "        <th class=\"col_heading level1 col9\" >mean</th> \n",
       "        <th class=\"col_heading level1 col10\" >std</th> \n",
       "        <th class=\"col_heading level1 col11\" >min</th> \n",
       "        <th class=\"col_heading level1 col12\" >25%</th> \n",
       "        <th class=\"col_heading level1 col13\" >50%</th> \n",
       "        <th class=\"col_heading level1 col14\" >75%</th> \n",
       "        <th class=\"col_heading level1 col15\" >max</th> \n",
       "        <th class=\"col_heading level1 col16\" >count</th> \n",
       "        <th class=\"col_heading level1 col17\" >mean</th> \n",
       "        <th class=\"col_heading level1 col18\" >std</th> \n",
       "        <th class=\"col_heading level1 col19\" >min</th> \n",
       "        <th class=\"col_heading level1 col20\" >25%</th> \n",
       "        <th class=\"col_heading level1 col21\" >50%</th> \n",
       "        <th class=\"col_heading level1 col22\" >75%</th> \n",
       "        <th class=\"col_heading level1 col23\" >max</th> \n",
       "        <th class=\"col_heading level1 col24\" >count</th> \n",
       "        <th class=\"col_heading level1 col25\" >mean</th> \n",
       "        <th class=\"col_heading level1 col26\" >std</th> \n",
       "        <th class=\"col_heading level1 col27\" >min</th> \n",
       "        <th class=\"col_heading level1 col28\" >25%</th> \n",
       "        <th class=\"col_heading level1 col29\" >50%</th> \n",
       "        <th class=\"col_heading level1 col30\" >75%</th> \n",
       "        <th class=\"col_heading level1 col31\" >max</th> \n",
       "        <th class=\"col_heading level1 col32\" >count</th> \n",
       "        <th class=\"col_heading level1 col33\" >mean</th> \n",
       "        <th class=\"col_heading level1 col34\" >std</th> \n",
       "        <th class=\"col_heading level1 col35\" >min</th> \n",
       "        <th class=\"col_heading level1 col36\" >25%</th> \n",
       "        <th class=\"col_heading level1 col37\" >50%</th> \n",
       "        <th class=\"col_heading level1 col38\" >75%</th> \n",
       "        <th class=\"col_heading level1 col39\" >max</th> \n",
       "        <th class=\"col_heading level1 col40\" >count</th> \n",
       "        <th class=\"col_heading level1 col41\" >mean</th> \n",
       "        <th class=\"col_heading level1 col42\" >std</th> \n",
       "        <th class=\"col_heading level1 col43\" >min</th> \n",
       "        <th class=\"col_heading level1 col44\" >25%</th> \n",
       "        <th class=\"col_heading level1 col45\" >50%</th> \n",
       "        <th class=\"col_heading level1 col46\" >75%</th> \n",
       "        <th class=\"col_heading level1 col47\" >max</th> \n",
       "        <th class=\"col_heading level1 col48\" >count</th> \n",
       "        <th class=\"col_heading level1 col49\" >mean</th> \n",
       "        <th class=\"col_heading level1 col50\" >std</th> \n",
       "        <th class=\"col_heading level1 col51\" >min</th> \n",
       "        <th class=\"col_heading level1 col52\" >25%</th> \n",
       "        <th class=\"col_heading level1 col53\" >50%</th> \n",
       "        <th class=\"col_heading level1 col54\" >75%</th> \n",
       "        <th class=\"col_heading level1 col55\" >max</th> \n",
       "        <th class=\"col_heading level1 col56\" >count</th> \n",
       "        <th class=\"col_heading level1 col57\" >mean</th> \n",
       "        <th class=\"col_heading level1 col58\" >std</th> \n",
       "        <th class=\"col_heading level1 col59\" >min</th> \n",
       "        <th class=\"col_heading level1 col60\" >25%</th> \n",
       "        <th class=\"col_heading level1 col61\" >50%</th> \n",
       "        <th class=\"col_heading level1 col62\" >75%</th> \n",
       "        <th class=\"col_heading level1 col63\" >max</th> \n",
       "        <th class=\"col_heading level1 col64\" >count</th> \n",
       "        <th class=\"col_heading level1 col65\" >mean</th> \n",
       "        <th class=\"col_heading level1 col66\" >std</th> \n",
       "        <th class=\"col_heading level1 col67\" >min</th> \n",
       "        <th class=\"col_heading level1 col68\" >25%</th> \n",
       "        <th class=\"col_heading level1 col69\" >50%</th> \n",
       "        <th class=\"col_heading level1 col70\" >75%</th> \n",
       "        <th class=\"col_heading level1 col71\" >max</th> \n",
       "        <th class=\"col_heading level1 col72\" >count</th> \n",
       "        <th class=\"col_heading level1 col73\" >mean</th> \n",
       "        <th class=\"col_heading level1 col74\" >std</th> \n",
       "        <th class=\"col_heading level1 col75\" >min</th> \n",
       "        <th class=\"col_heading level1 col76\" >25%</th> \n",
       "        <th class=\"col_heading level1 col77\" >50%</th> \n",
       "        <th class=\"col_heading level1 col78\" >75%</th> \n",
       "        <th class=\"col_heading level1 col79\" >max</th> \n",
       "        <th class=\"col_heading level1 col80\" >count</th> \n",
       "        <th class=\"col_heading level1 col81\" >mean</th> \n",
       "        <th class=\"col_heading level1 col82\" >std</th> \n",
       "        <th class=\"col_heading level1 col83\" >min</th> \n",
       "        <th class=\"col_heading level1 col84\" >25%</th> \n",
       "        <th class=\"col_heading level1 col85\" >50%</th> \n",
       "        <th class=\"col_heading level1 col86\" >75%</th> \n",
       "        <th class=\"col_heading level1 col87\" >max</th> \n",
       "        <th class=\"col_heading level1 col88\" >count</th> \n",
       "        <th class=\"col_heading level1 col89\" >mean</th> \n",
       "        <th class=\"col_heading level1 col90\" >std</th> \n",
       "        <th class=\"col_heading level1 col91\" >min</th> \n",
       "        <th class=\"col_heading level1 col92\" >25%</th> \n",
       "        <th class=\"col_heading level1 col93\" >50%</th> \n",
       "        <th class=\"col_heading level1 col94\" >75%</th> \n",
       "        <th class=\"col_heading level1 col95\" >max</th> \n",
       "        <th class=\"col_heading level1 col96\" >count</th> \n",
       "        <th class=\"col_heading level1 col97\" >mean</th> \n",
       "        <th class=\"col_heading level1 col98\" >std</th> \n",
       "        <th class=\"col_heading level1 col99\" >min</th> \n",
       "        <th class=\"col_heading level1 col100\" >25%</th> \n",
       "        <th class=\"col_heading level1 col101\" >50%</th> \n",
       "        <th class=\"col_heading level1 col102\" >75%</th> \n",
       "        <th class=\"col_heading level1 col103\" >max</th> \n",
       "    </tr>    <tr> \n",
       "        <th class=\"index_name level0\" >K</th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "        <th class=\"blank\" ></th> \n",
       "    </tr></thead> \n",
       "<tbody>    <tr> \n",
       "        <th id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4level0_row0\" class=\"row_heading level0 row0\" >0</th> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col0\" class=\"data row0 col0\" >69.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col1\" class=\"data row0 col1\" >12.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col2\" class=\"data row0 col2\" >0.7</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col3\" class=\"data row0 col3\" >11.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col4\" class=\"data row0 col4\" >12.1</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col5\" class=\"data row0 col5\" >12.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col6\" class=\"data row0 col6\" >12.8</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col7\" class=\"data row0 col7\" >14.1</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col8\" class=\"data row0 col8\" >69.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col9\" class=\"data row0 col9\" >4.1</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col10\" class=\"data row0 col10\" >2.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col11\" class=\"data row0 col11\" >1.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col12\" class=\"data row0 col12\" >2.6</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col13\" class=\"data row0 col13\" >3.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col14\" class=\"data row0 col14\" >5.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col15\" class=\"data row0 col15\" >10.8</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col16\" class=\"data row0 col16\" >69.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col17\" class=\"data row0 col17\" >0.9</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col18\" class=\"data row0 col18\" >0.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col19\" class=\"data row0 col19\" >0.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col20\" class=\"data row0 col20\" >0.8</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col21\" class=\"data row0 col21\" >0.9</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col22\" class=\"data row0 col22\" >1.1</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col23\" class=\"data row0 col23\" >1.7</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col24\" class=\"data row0 col24\" >69.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col25\" class=\"data row0 col25\" >2.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col26\" class=\"data row0 col26\" >0.7</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col27\" class=\"data row0 col27\" >1.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col28\" class=\"data row0 col28\" >2.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col29\" class=\"data row0 col29\" >2.6</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col30\" class=\"data row0 col30\" >3.1</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col31\" class=\"data row0 col31\" >3.7</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col32\" class=\"data row0 col32\" >69.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col33\" class=\"data row0 col33\" >458.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col34\" class=\"data row0 col34\" >79.9</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col35\" class=\"data row0 col35\" >278.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col36\" class=\"data row0 col36\" >406.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col37\" class=\"data row0 col37\" >470.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col38\" class=\"data row0 col38\" >520.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col39\" class=\"data row0 col39\" >590.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col40\" class=\"data row0 col40\" >69.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col41\" class=\"data row0 col41\" >2.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col42\" class=\"data row0 col42\" >1.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col43\" class=\"data row0 col43\" >0.7</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col44\" class=\"data row0 col44\" >1.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col45\" class=\"data row0 col45\" >2.1</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col46\" class=\"data row0 col46\" >3.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col47\" class=\"data row0 col47\" >5.8</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col48\" class=\"data row0 col48\" >69.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col49\" class=\"data row0 col49\" >2.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col50\" class=\"data row0 col50\" >0.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col51\" class=\"data row0 col51\" >1.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col52\" class=\"data row0 col52\" >2.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col53\" class=\"data row0 col53\" >2.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col54\" class=\"data row0 col54\" >2.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col55\" class=\"data row0 col55\" >3.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col56\" class=\"data row0 col56\" >69.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col57\" class=\"data row0 col57\" >20.8</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col58\" class=\"data row0 col58\" >3.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col59\" class=\"data row0 col59\" >10.6</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col60\" class=\"data row0 col60\" >19.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col61\" class=\"data row0 col61\" >21.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col62\" class=\"data row0 col62\" >22.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col63\" class=\"data row0 col63\" >28.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col64\" class=\"data row0 col64\" >69.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col65\" class=\"data row0 col65\" >92.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col66\" class=\"data row0 col66\" >10.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col67\" class=\"data row0 col67\" >78.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col68\" class=\"data row0 col68\" >86.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col69\" class=\"data row0 col69\" >88.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col70\" class=\"data row0 col70\" >98.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col71\" class=\"data row0 col71\" >123.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col72\" class=\"data row0 col72\" >69.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col73\" class=\"data row0 col73\" >2.1</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col74\" class=\"data row0 col74\" >0.6</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col75\" class=\"data row0 col75\" >1.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col76\" class=\"data row0 col76\" >1.6</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col77\" class=\"data row0 col77\" >2.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col78\" class=\"data row0 col78\" >2.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col79\" class=\"data row0 col79\" >3.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col80\" class=\"data row0 col80\" >69.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col81\" class=\"data row0 col81\" >1.8</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col82\" class=\"data row0 col82\" >0.9</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col83\" class=\"data row0 col83\" >0.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col84\" class=\"data row0 col84\" >1.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col85\" class=\"data row0 col85\" >1.8</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col86\" class=\"data row0 col86\" >2.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col87\" class=\"data row0 col87\" >5.1</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col88\" class=\"data row0 col88\" >69.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col89\" class=\"data row0 col89\" >0.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col90\" class=\"data row0 col90\" >0.1</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col91\" class=\"data row0 col91\" >0.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col92\" class=\"data row0 col92\" >0.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col93\" class=\"data row0 col93\" >0.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col94\" class=\"data row0 col94\" >0.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col95\" class=\"data row0 col95\" >0.7</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col96\" class=\"data row0 col96\" >69.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col97\" class=\"data row0 col97\" >1.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col98\" class=\"data row0 col98\" >0.6</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col99\" class=\"data row0 col99\" >0.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col100\" class=\"data row0 col100\" >1.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col101\" class=\"data row0 col101\" >1.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col102\" class=\"data row0 col102\" >1.8</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row0_col103\" class=\"data row0 col103\" >3.6</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4level0_row1\" class=\"row_heading level0 row1\" >1</th> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col0\" class=\"data row1 col0\" >47.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col1\" class=\"data row1 col1\" >13.8</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col2\" class=\"data row1 col2\" >0.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col3\" class=\"data row1 col3\" >12.8</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col4\" class=\"data row1 col4\" >13.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col5\" class=\"data row1 col5\" >13.8</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col6\" class=\"data row1 col6\" >14.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col7\" class=\"data row1 col7\" >14.8</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col8\" class=\"data row1 col8\" >47.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col9\" class=\"data row1 col9\" >5.7</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col10\" class=\"data row1 col10\" >1.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col11\" class=\"data row1 col11\" >3.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col12\" class=\"data row1 col12\" >4.8</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col13\" class=\"data row1 col13\" >5.6</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col14\" class=\"data row1 col14\" >6.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col15\" class=\"data row1 col15\" >8.9</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col16\" class=\"data row1 col16\" >47.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col17\" class=\"data row1 col17\" >1.1</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col18\" class=\"data row1 col18\" >0.1</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col19\" class=\"data row1 col19\" >0.9</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col20\" class=\"data row1 col20\" >1.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col21\" class=\"data row1 col21\" >1.1</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col22\" class=\"data row1 col22\" >1.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col23\" class=\"data row1 col23\" >1.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col24\" class=\"data row1 col24\" >47.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col25\" class=\"data row1 col25\" >3.1</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col26\" class=\"data row1 col26\" >0.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col27\" class=\"data row1 col27\" >2.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col28\" class=\"data row1 col28\" >2.9</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col29\" class=\"data row1 col29\" >3.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col30\" class=\"data row1 col30\" >3.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col31\" class=\"data row1 col31\" >4.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col32\" class=\"data row1 col32\" >47.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col33\" class=\"data row1 col33\" >1195.1</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col34\" class=\"data row1 col34\" >171.6</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col35\" class=\"data row1 col35\" >970.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col36\" class=\"data row1 col36\" >1055.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col37\" class=\"data row1 col37\" >1150.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col38\" class=\"data row1 col38\" >1285.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col39\" class=\"data row1 col39\" >1680.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col40\" class=\"data row1 col40\" >47.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col41\" class=\"data row1 col41\" >1.9</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col42\" class=\"data row1 col42\" >0.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col43\" class=\"data row1 col43\" >1.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col44\" class=\"data row1 col44\" >1.7</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col45\" class=\"data row1 col45\" >1.7</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col46\" class=\"data row1 col46\" >1.9</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col47\" class=\"data row1 col47\" >4.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col48\" class=\"data row1 col48\" >47.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col49\" class=\"data row1 col49\" >2.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col50\" class=\"data row1 col50\" >0.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col51\" class=\"data row1 col51\" >2.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col52\" class=\"data row1 col52\" >2.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col53\" class=\"data row1 col53\" >2.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col54\" class=\"data row1 col54\" >2.6</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col55\" class=\"data row1 col55\" >2.7</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col56\" class=\"data row1 col56\" >47.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col57\" class=\"data row1 col57\" >17.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col58\" class=\"data row1 col58\" >3.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col59\" class=\"data row1 col59\" >11.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col60\" class=\"data row1 col60\" >16.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col61\" class=\"data row1 col61\" >16.8</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col62\" class=\"data row1 col62\" >18.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col63\" class=\"data row1 col63\" >30.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col64\" class=\"data row1 col64\" >47.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col65\" class=\"data row1 col65\" >105.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col66\" class=\"data row1 col66\" >11.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col67\" class=\"data row1 col67\" >89.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col68\" class=\"data row1 col68\" >97.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col69\" class=\"data row1 col69\" >102.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col70\" class=\"data row1 col70\" >112.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col71\" class=\"data row1 col71\" >139.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col72\" class=\"data row1 col72\" >47.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col73\" class=\"data row1 col73\" >2.9</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col74\" class=\"data row1 col74\" >0.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col75\" class=\"data row1 col75\" >2.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col76\" class=\"data row1 col76\" >2.6</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col77\" class=\"data row1 col77\" >2.9</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col78\" class=\"data row1 col78\" >3.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col79\" class=\"data row1 col79\" >3.9</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col80\" class=\"data row1 col80\" >47.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col81\" class=\"data row1 col81\" >3.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col82\" class=\"data row1 col82\" >0.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col83\" class=\"data row1 col83\" >2.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col84\" class=\"data row1 col84\" >2.7</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col85\" class=\"data row1 col85\" >3.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col86\" class=\"data row1 col86\" >3.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col87\" class=\"data row1 col87\" >3.9</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col88\" class=\"data row1 col88\" >47.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col89\" class=\"data row1 col89\" >0.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col90\" class=\"data row1 col90\" >0.1</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col91\" class=\"data row1 col91\" >0.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col92\" class=\"data row1 col92\" >0.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col93\" class=\"data row1 col93\" >0.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col94\" class=\"data row1 col94\" >0.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col95\" class=\"data row1 col95\" >0.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col96\" class=\"data row1 col96\" >47.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col97\" class=\"data row1 col97\" >1.9</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col98\" class=\"data row1 col98\" >0.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col99\" class=\"data row1 col99\" >1.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col100\" class=\"data row1 col100\" >1.6</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col101\" class=\"data row1 col101\" >1.9</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col102\" class=\"data row1 col102\" >2.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row1_col103\" class=\"data row1 col103\" >3.0</td> \n",
       "    </tr>    <tr> \n",
       "        <th id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4level0_row2\" class=\"row_heading level0 row2\" >2</th> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col0\" class=\"data row2 col0\" >62.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col1\" class=\"data row2 col1\" >12.9</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col2\" class=\"data row2 col2\" >0.7</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col3\" class=\"data row2 col3\" >11.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col4\" class=\"data row2 col4\" >12.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col5\" class=\"data row2 col5\" >13.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col6\" class=\"data row2 col6\" >13.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col7\" class=\"data row2 col7\" >14.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col8\" class=\"data row2 col8\" >62.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col9\" class=\"data row2 col9\" >5.7</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col10\" class=\"data row2 col10\" >2.7</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col11\" class=\"data row2 col11\" >1.7</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col12\" class=\"data row2 col12\" >3.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col13\" class=\"data row2 col13\" >4.8</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col14\" class=\"data row2 col14\" >7.6</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col15\" class=\"data row2 col15\" >13.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col16\" class=\"data row2 col16\" >62.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col17\" class=\"data row2 col17\" >0.9</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col18\" class=\"data row2 col18\" >0.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col19\" class=\"data row2 col19\" >0.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col20\" class=\"data row2 col20\" >0.7</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col21\" class=\"data row2 col21\" >0.9</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col22\" class=\"data row2 col22\" >1.1</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col23\" class=\"data row2 col23\" >1.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col24\" class=\"data row2 col24\" >62.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col25\" class=\"data row2 col25\" >2.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col26\" class=\"data row2 col26\" >0.8</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col27\" class=\"data row2 col27\" >1.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col28\" class=\"data row2 col28\" >1.7</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col29\" class=\"data row2 col29\" >2.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col30\" class=\"data row2 col30\" >3.1</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col31\" class=\"data row2 col31\" >3.8</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col32\" class=\"data row2 col32\" >62.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col33\" class=\"data row2 col33\" >728.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col34\" class=\"data row2 col34\" >94.8</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col35\" class=\"data row2 col35\" >600.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col36\" class=\"data row2 col36\" >652.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col37\" class=\"data row2 col37\" >702.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col38\" class=\"data row2 col38\" >791.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col39\" class=\"data row2 col39\" >937.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col40\" class=\"data row2 col40\" >62.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col41\" class=\"data row2 col41\" >2.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col42\" class=\"data row2 col42\" >1.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col43\" class=\"data row2 col43\" >0.9</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col44\" class=\"data row2 col44\" >1.6</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col45\" class=\"data row2 col45\" >2.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col46\" class=\"data row2 col46\" >3.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col47\" class=\"data row2 col47\" >5.7</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col48\" class=\"data row2 col48\" >62.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col49\" class=\"data row2 col49\" >2.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col50\" class=\"data row2 col50\" >0.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col51\" class=\"data row2 col51\" >1.7</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col52\" class=\"data row2 col52\" >2.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col53\" class=\"data row2 col53\" >2.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col54\" class=\"data row2 col54\" >2.6</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col55\" class=\"data row2 col55\" >3.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col56\" class=\"data row2 col56\" >62.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col57\" class=\"data row2 col57\" >19.9</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col58\" class=\"data row2 col58\" >2.7</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col59\" class=\"data row2 col59\" >13.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col60\" class=\"data row2 col60\" >18.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col61\" class=\"data row2 col61\" >20.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col62\" class=\"data row2 col62\" >21.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col63\" class=\"data row2 col63\" >27.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col64\" class=\"data row2 col64\" >62.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col65\" class=\"data row2 col65\" >103.6</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col66\" class=\"data row2 col66\" >16.6</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col67\" class=\"data row2 col67\" >70.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col68\" class=\"data row2 col68\" >93.2</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col69\" class=\"data row2 col69\" >101.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col70\" class=\"data row2 col70\" >110.8</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col71\" class=\"data row2 col71\" >162.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col72\" class=\"data row2 col72\" >62.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col73\" class=\"data row2 col73\" >2.1</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col74\" class=\"data row2 col74\" >0.6</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col75\" class=\"data row2 col75\" >1.1</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col76\" class=\"data row2 col76\" >1.7</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col77\" class=\"data row2 col77\" >2.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col78\" class=\"data row2 col78\" >2.6</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col79\" class=\"data row2 col79\" >3.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col80\" class=\"data row2 col80\" >62.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col81\" class=\"data row2 col81\" >1.6</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col82\" class=\"data row2 col82\" >1.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col83\" class=\"data row2 col83\" >0.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col84\" class=\"data row2 col84\" >0.7</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col85\" class=\"data row2 col85\" >1.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col86\" class=\"data row2 col86\" >2.6</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col87\" class=\"data row2 col87\" >3.8</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col88\" class=\"data row2 col88\" >62.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col89\" class=\"data row2 col89\" >0.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col90\" class=\"data row2 col90\" >0.1</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col91\" class=\"data row2 col91\" >0.1</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col92\" class=\"data row2 col92\" >0.3</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col93\" class=\"data row2 col93\" >0.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col94\" class=\"data row2 col94\" >0.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col95\" class=\"data row2 col95\" >0.6</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col96\" class=\"data row2 col96\" >62.0</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col97\" class=\"data row2 col97\" >1.5</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col98\" class=\"data row2 col98\" >0.6</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col99\" class=\"data row2 col99\" >0.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col100\" class=\"data row2 col100\" >1.1</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col101\" class=\"data row2 col101\" >1.4</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col102\" class=\"data row2 col102\" >1.9</td> \n",
       "        <td id=\"T_50a3da00_81af_11e9_bec3_9c5c8ebf7da4row2_col103\" class=\"data row2 col103\" >3.3</td> \n",
       "    </tr></tbody> \n",
       "</table> "
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x2a8aeaf0a20>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Tunnuslukuja ryhmittäin:\n",
    "X.groupby('K').describe().style.format('{:.1f}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Viini</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>K</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>50</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>46</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>13</td>\n",
       "      <td>20</td>\n",
       "      <td>29</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Viini   1   2   3\n",
       "K                \n",
       "0       0  50  19\n",
       "1      46   1   0\n",
       "2      13  20  29"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Katsotaan miten muodostetut ryhmät suhteutuava lajikkeisiin:\n",
    "pd.crosstab(X['K'], df['Viini'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ensimmäiseen ryhmään meni suuri määrä kaikkia viinejä, mutta ryhmä vaikuttaisi olevan viini 3. Toinen ryhmä on selkeästi viini 1 ja kolmas ryhmä melko selvästi viini 2"
   ]
  }
 ],
 "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": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}