Irises.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/sfczekalski/anaconda3/lib/python3.7/site-packages/statsmodels/tools/_testing.py:19: FutureWarning: pandas.util.testing is deprecated. Use the functions in the public API at pandas.testing instead.\n",
      "  import pandas.util.testing as tm\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from scipy.stats import pearsonr\n",
    "\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import accuracy_score, confusion_matrix\n",
    "from sklearn import tree\n",
    "from IPython.display import SVG\n",
    "from graphviz import Source\n",
    "from IPython.display import display"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# First look at the data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After importing necessary packages, I will read the data from the file and take a look at the first rows"
   ]
  },
  {
   "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>Sepal.Length</th>\n",
       "      <th>Sepal.Width</th>\n",
       "      <th>Petal.Length</th>\n",
       "      <th>Petal.Width</th>\n",
       "      <th>Species</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5.1</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4.7</td>\n",
       "      <td>3.2</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.2</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4.6</td>\n",
       "      <td>3.1</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.2</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>5.0</td>\n",
       "      <td>3.6</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Sepal.Length  Sepal.Width  Petal.Length Petal.Width Species\n",
       "1           5.1          3.5           1.4         0.2  setosa\n",
       "2           4.9          3.0           1.4         0.2  setosa\n",
       "3           4.7          3.2           1.3         0.2  setosa\n",
       "4           4.6          3.1           1.5         0.2  setosa\n",
       "5           5.0          3.6           1.4         0.2  setosa"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv('data/Graduate - IRISES dataset (2019-06).csv', sep='|')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that the file has been read correctly. Let's invetigate further: check our dataframe's shape, if it contains any missing values or duplicate rows, its data types and some summary statistics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(150, 5)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Sepal.Length    False\n",
       "Sepal.Width      True\n",
       "Petal.Length    False\n",
       "Petal.Width     False\n",
       "Species         False\n",
       "dtype: bool"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isna().any()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>Sepal.Length</th>\n",
       "      <th>Sepal.Width</th>\n",
       "      <th>Petal.Length</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>150.000000</td>\n",
       "      <td>149.000000</td>\n",
       "      <td>150.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>5.779333</td>\n",
       "      <td>3.061745</td>\n",
       "      <td>3.758000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.197721</td>\n",
       "      <td>0.433963</td>\n",
       "      <td>1.765298</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-4.800000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>5.100000</td>\n",
       "      <td>2.800000</td>\n",
       "      <td>1.600000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>5.800000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>4.350000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>6.400000</td>\n",
       "      <td>3.300000</td>\n",
       "      <td>5.100000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>7.900000</td>\n",
       "      <td>4.400000</td>\n",
       "      <td>6.900000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Sepal.Length  Sepal.Width  Petal.Length\n",
       "count    150.000000   149.000000    150.000000\n",
       "mean       5.779333     3.061745      3.758000\n",
       "std        1.197721     0.433963      1.765298\n",
       "min       -4.800000     2.000000      1.000000\n",
       "25%        5.100000     2.800000      1.600000\n",
       "50%        5.800000     3.000000      4.350000\n",
       "75%        6.400000     3.300000      5.100000\n",
       "max        7.900000     4.400000      6.900000"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that petal width's summary is not printed! One needs to always check the data types of the variables - it might be the reason of missing summary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Sepal.Length    float64\n",
       "Sepal.Width     float64\n",
       "Petal.Length    float64\n",
       "Petal.Width      object\n",
       "Species          object\n",
       "dtype: object"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data cleansing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Indeed, the column was interpreted as object, instead of floating point values. We need to fix it! After trying to use standard \"astype(float)\" command I noticed it is impossible to do it that easily - someone has written a comma instead of a dot."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>Sepal.Length</th>\n",
       "      <th>Sepal.Width</th>\n",
       "      <th>Petal.Length</th>\n",
       "      <th>Petal.Width</th>\n",
       "      <th>Species</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>133</th>\n",
       "      <td>6.4</td>\n",
       "      <td>2.8</td>\n",
       "      <td>5.6</td>\n",
       "      <td>2,2</td>\n",
       "      <td>virginica</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Sepal.Length  Sepal.Width  Petal.Length Petal.Width    Species\n",
       "133           6.4          2.8           5.6         2,2  virginica"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "comma_indicies = df['Petal.Width'].str.contains(',')\n",
    "df[comma_indicies]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that there is one such value. In this case, I will replace it manually - of course if there was more of such corrupted data I would try a suitable regex. After this fix our \"astype(float)\" command serves us as expected and we can take a look at the summary of all the columns."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.loc[comma_indicies, 'Petal.Width'] = 2.2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['Petal.Width'] = df['Petal.Width'].astype(float)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Sepal.Length    float64\n",
       "Sepal.Width     float64\n",
       "Petal.Length    float64\n",
       "Petal.Width     float64\n",
       "Species          object\n",
       "dtype: object"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>Sepal.Length</th>\n",
       "      <th>Sepal.Width</th>\n",
       "      <th>Petal.Length</th>\n",
       "      <th>Petal.Width</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>150.000000</td>\n",
       "      <td>149.000000</td>\n",
       "      <td>150.000000</td>\n",
       "      <td>150.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>5.779333</td>\n",
       "      <td>3.061745</td>\n",
       "      <td>3.758000</td>\n",
       "      <td>1.199333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.197721</td>\n",
       "      <td>0.433963</td>\n",
       "      <td>1.765298</td>\n",
       "      <td>0.762238</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-4.800000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.100000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>5.100000</td>\n",
       "      <td>2.800000</td>\n",
       "      <td>1.600000</td>\n",
       "      <td>0.300000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>5.800000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>4.350000</td>\n",
       "      <td>1.300000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>6.400000</td>\n",
       "      <td>3.300000</td>\n",
       "      <td>5.100000</td>\n",
       "      <td>1.800000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>7.900000</td>\n",
       "      <td>4.400000</td>\n",
       "      <td>6.900000</td>\n",
       "      <td>2.500000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Sepal.Length  Sepal.Width  Petal.Length  Petal.Width\n",
       "count    150.000000   149.000000    150.000000   150.000000\n",
       "mean       5.779333     3.061745      3.758000     1.199333\n",
       "std        1.197721     0.433963      1.765298     0.762238\n",
       "min       -4.800000     2.000000      1.000000     0.100000\n",
       "25%        5.100000     2.800000      1.600000     0.300000\n",
       "50%        5.800000     3.000000      4.350000     1.300000\n",
       "75%        6.400000     3.300000      5.100000     1.800000\n",
       "max        7.900000     4.400000      6.900000     2.500000"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "What is also really important to check is presence of duplicate rows. Imagine that there are two copies of one of the rows: the first in the training set and the other in validation set. The performance of our classifier would be overestimated!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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>Sepal.Length</th>\n",
       "      <th>Sepal.Width</th>\n",
       "      <th>Petal.Length</th>\n",
       "      <th>Petal.Width</th>\n",
       "      <th>Species</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>102</th>\n",
       "      <td>5.8</td>\n",
       "      <td>2.7</td>\n",
       "      <td>5.1</td>\n",
       "      <td>1.9</td>\n",
       "      <td>virginica</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>143</th>\n",
       "      <td>5.8</td>\n",
       "      <td>2.7</td>\n",
       "      <td>5.1</td>\n",
       "      <td>1.9</td>\n",
       "      <td>virginica</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Sepal.Length  Sepal.Width  Petal.Length  Petal.Width    Species\n",
       "102           5.8          2.7           5.1          1.9  virginica\n",
       "143           5.8          2.7           5.1          1.9  virginica"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[df.duplicated(keep=False)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There is one such raw in our dataset! Let's remove it"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.drop_duplicates()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>Sepal.Length</th>\n",
       "      <th>Sepal.Width</th>\n",
       "      <th>Petal.Length</th>\n",
       "      <th>Petal.Width</th>\n",
       "      <th>Species</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [Sepal.Length, Sepal.Width, Petal.Length, Petal.Width, Species]\n",
       "Index: []"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[df.duplicated(keep=False)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(149, 5)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's check missing values now"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It turns out, that one measurement of sepal width is missing. What also looks suspicious is that sepal length got some negative values. Let's take a closer look at the row with missing information first, and then investigate what might be the reason of the negative values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "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>Sepal.Length</th>\n",
       "      <th>Sepal.Width</th>\n",
       "      <th>Petal.Length</th>\n",
       "      <th>Petal.Width</th>\n",
       "      <th>Species</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>82</th>\n",
       "      <td>5.5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3.7</td>\n",
       "      <td>1.0</td>\n",
       "      <td>versicolor</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Sepal.Length  Sepal.Width  Petal.Length  Petal.Width     Species\n",
       "82           5.5          NaN           3.7          1.0  versicolor"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[df['Sepal.Width'].isna(), :]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Sepal.Length</th>\n",
       "      <th>Sepal.Width</th>\n",
       "      <th>Petal.Length</th>\n",
       "      <th>Petal.Width</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>50.000000</td>\n",
       "      <td>49.000000</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>50.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>5.936000</td>\n",
       "      <td>2.777551</td>\n",
       "      <td>4.260000</td>\n",
       "      <td>1.326000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.516171</td>\n",
       "      <td>0.312427</td>\n",
       "      <td>0.469911</td>\n",
       "      <td>0.197753</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>4.900000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>5.600000</td>\n",
       "      <td>2.600000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>1.200000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>5.900000</td>\n",
       "      <td>2.800000</td>\n",
       "      <td>4.350000</td>\n",
       "      <td>1.300000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>6.300000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>4.600000</td>\n",
       "      <td>1.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>7.000000</td>\n",
       "      <td>3.400000</td>\n",
       "      <td>5.100000</td>\n",
       "      <td>1.800000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Sepal.Length  Sepal.Width  Petal.Length  Petal.Width\n",
       "count     50.000000    49.000000     50.000000    50.000000\n",
       "mean       5.936000     2.777551      4.260000     1.326000\n",
       "std        0.516171     0.312427      0.469911     0.197753\n",
       "min        4.900000     2.000000      3.000000     1.000000\n",
       "25%        5.600000     2.600000      4.000000     1.200000\n",
       "50%        5.900000     2.800000      4.350000     1.300000\n",
       "75%        6.300000     3.000000      4.600000     1.500000\n",
       "max        7.000000     3.400000      5.100000     1.800000"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[df['Species'] == 'versicolor'].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that sepal width is really important when one would try to differentiate between irises from versicolor class and other classes. The row with the missing value describe object from that class, so it is really important to impute this value properly. \n",
    "\n",
    "I will use median of this variable from that class. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "median_versicolor_sepal_width = np.median(df.loc[(df['Species'] == 'versicolor') & \n",
    "                                                 ~(df['Sepal.Width'].isna()), 'Sepal.Width'])\n",
    "\n",
    "df.loc[df['Sepal.Width'].isna(), 'Sepal.Width'] = median_versicolor_sepal_width"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "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>Sepal.Length</th>\n",
       "      <th>Sepal.Width</th>\n",
       "      <th>Petal.Length</th>\n",
       "      <th>Petal.Width</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>149.000000</td>\n",
       "      <td>149.000000</td>\n",
       "      <td>149.000000</td>\n",
       "      <td>149.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>5.779195</td>\n",
       "      <td>3.062416</td>\n",
       "      <td>3.748993</td>\n",
       "      <td>1.194631</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.201760</td>\n",
       "      <td>0.433477</td>\n",
       "      <td>1.767791</td>\n",
       "      <td>0.762622</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-4.800000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.100000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>5.100000</td>\n",
       "      <td>2.800000</td>\n",
       "      <td>1.600000</td>\n",
       "      <td>0.300000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>5.800000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>4.300000</td>\n",
       "      <td>1.300000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>6.400000</td>\n",
       "      <td>3.300000</td>\n",
       "      <td>5.100000</td>\n",
       "      <td>1.800000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>7.900000</td>\n",
       "      <td>4.400000</td>\n",
       "      <td>6.900000</td>\n",
       "      <td>2.500000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Sepal.Length  Sepal.Width  Petal.Length  Petal.Width\n",
       "count    149.000000   149.000000    149.000000   149.000000\n",
       "mean       5.779195     3.062416      3.748993     1.194631\n",
       "std        1.201760     0.433477      1.767791     0.762622\n",
       "min       -4.800000     2.000000      1.000000     0.100000\n",
       "25%        5.100000     2.800000      1.600000     0.300000\n",
       "50%        5.800000     3.000000      4.300000     1.300000\n",
       "75%        6.400000     3.300000      5.100000     1.800000\n",
       "max        7.900000     4.400000      6.900000     2.500000"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After imputing the missing value, I will dive in into sepal lengths. Why are some values negative?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Sepal.Length</th>\n",
       "      <th>Sepal.Width</th>\n",
       "      <th>Petal.Length</th>\n",
       "      <th>Petal.Width</th>\n",
       "      <th>Species</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>-4.8</td>\n",
       "      <td>3.4</td>\n",
       "      <td>1.9</td>\n",
       "      <td>0.2</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Sepal.Length  Sepal.Width  Petal.Length  Petal.Width Species\n",
       "25          -4.8          3.4           1.9          0.2  setosa"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[df['Sepal.Length'] <= 0.0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "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>Sepal.Length</th>\n",
       "      <th>Sepal.Width</th>\n",
       "      <th>Petal.Length</th>\n",
       "      <th>Petal.Width</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>50.000000</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>50.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>4.814000</td>\n",
       "      <td>3.428000</td>\n",
       "      <td>1.462000</td>\n",
       "      <td>0.246000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.431142</td>\n",
       "      <td>0.379064</td>\n",
       "      <td>0.173664</td>\n",
       "      <td>0.105386</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-4.800000</td>\n",
       "      <td>2.300000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.100000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>4.800000</td>\n",
       "      <td>3.200000</td>\n",
       "      <td>1.400000</td>\n",
       "      <td>0.200000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>5.000000</td>\n",
       "      <td>3.400000</td>\n",
       "      <td>1.500000</td>\n",
       "      <td>0.200000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>5.200000</td>\n",
       "      <td>3.675000</td>\n",
       "      <td>1.575000</td>\n",
       "      <td>0.300000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>5.800000</td>\n",
       "      <td>4.400000</td>\n",
       "      <td>1.900000</td>\n",
       "      <td>0.600000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Sepal.Length  Sepal.Width  Petal.Length  Petal.Width\n",
       "count     50.000000    50.000000     50.000000    50.000000\n",
       "mean       4.814000     3.428000      1.462000     0.246000\n",
       "std        1.431142     0.379064      0.173664     0.105386\n",
       "min       -4.800000     2.300000      1.000000     0.100000\n",
       "25%        4.800000     3.200000      1.400000     0.200000\n",
       "50%        5.000000     3.400000      1.500000     0.200000\n",
       "75%        5.200000     3.675000      1.575000     0.300000\n",
       "max        5.800000     4.400000      1.900000     0.600000"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[df['Species'] == 'setosa'].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There is only one such value. After looking at summary of irises from setosa class we notice that the absoulte value of the sepal length is very close to the mean for the whole class. Someone has probably just put a minus sign before the number by mistake! Let's fix that"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.loc[df['Sepal.Length'] <= 0.0, 'Sepal.Length'] = np.abs(df.loc[df['Sepal.Length'] <= 0.0, 'Sepal.Length'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "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>Sepal.Length</th>\n",
       "      <th>Sepal.Width</th>\n",
       "      <th>Petal.Length</th>\n",
       "      <th>Petal.Width</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>50.00000</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>50.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>5.00600</td>\n",
       "      <td>3.428000</td>\n",
       "      <td>1.462000</td>\n",
       "      <td>0.246000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.35249</td>\n",
       "      <td>0.379064</td>\n",
       "      <td>0.173664</td>\n",
       "      <td>0.105386</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>4.30000</td>\n",
       "      <td>2.300000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.100000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>4.80000</td>\n",
       "      <td>3.200000</td>\n",
       "      <td>1.400000</td>\n",
       "      <td>0.200000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>5.00000</td>\n",
       "      <td>3.400000</td>\n",
       "      <td>1.500000</td>\n",
       "      <td>0.200000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>5.20000</td>\n",
       "      <td>3.675000</td>\n",
       "      <td>1.575000</td>\n",
       "      <td>0.300000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>5.80000</td>\n",
       "      <td>4.400000</td>\n",
       "      <td>1.900000</td>\n",
       "      <td>0.600000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Sepal.Length  Sepal.Width  Petal.Length  Petal.Width\n",
       "count      50.00000    50.000000     50.000000    50.000000\n",
       "mean        5.00600     3.428000      1.462000     0.246000\n",
       "std         0.35249     0.379064      0.173664     0.105386\n",
       "min         4.30000     2.300000      1.000000     0.100000\n",
       "25%         4.80000     3.200000      1.400000     0.200000\n",
       "50%         5.00000     3.400000      1.500000     0.200000\n",
       "75%         5.20000     3.675000      1.575000     0.300000\n",
       "max         5.80000     4.400000      1.900000     0.600000"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[df['Species'] == 'setosa'].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Univariate analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Looks much better now! Summary statistics are useful for noticing bugs like the ones I've noticed, but sometimes we may miss something. This is where visaulizations come to play - we can plot distributions of our variables and understand them better."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df['Species'].hist();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The classes are evenly distributed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.boxplot(column=['Sepal.Length', 'Sepal.Width', 'Petal.Length', 'Petal.Width']);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Boxplots allow us to take a look at distributions of continuous variables. We can see that sepal width variable has some values below the first quartile and above the third quartile, but not very much. There is no outliers in our data.\n",
    "\n",
    "Histograms and distribution plots will allow us to take a closer look."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot(df['Sepal.Length']);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot(df['Sepal.Width']);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAEGCAYAAABrQF4qAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nO3deXQc9Znv//fTrX3fZW2W5UWWZCxsvAEOZjV7IMkPhnXCTLghzE0mN5Nk5pCZ+SVzyV0yySRzZ87kEhjgNySBECDAGDAxhMUsxou8W5Jly7as3drV2qXu/v7+UJsIIaO23a3qrn5e5+hYXV1V/ciSPqr+1reeEmMMSiml7MthdQFKKaWCS4NeKaVsToNeKaVsToNeKaVsToNeKaVsLsrqAqbLysoyCxYssLoMpZQKK7t37+4yxmTP9FzIBf2CBQuoqqqyugyllAorInLyTM/p0I1SStmcBr1SStmcBr1SStmcBr1SStmcBr1SStmcBr1SStmcBr1SStmcBr1SStmcBr1SStlcyF0Zq87NMzsarS7hE+5eN9/qEpRSPnpEr5RSNqdBr5RSNqdBr5RSNqdBr5RSNqdBr5RSNqdBr5RSNqdBr5RSNqdBr5RSNqdBr5RSNqdBr5RSNqdBr5RSNqdBr5RSNqdBr5RSNqdBr5RSNqdBr5RSNqf96JWygN4/QM0lPaJXSimb06BXSimb06BXSimb06BXSimb06BXSimb8yvoReR6EakTkXoReWiG578tIjUickBE3hKR4inPeURkn+9jUyCLV0opNbtZp1eKiBP4ObARaAZ2icgmY0zNlNX2AquNMcMi8hfAj4E7fM+NGGNWBLhupZRSfvLniH4tUG+MOW6MGQeeBW6duoIx5h1jzLDv4XagMLBlKqWUOlf+BH0B0DTlcbNv2ZncD7w+5XGciFSJyHYR+cJMG4jIA751qjo7O/0oSSmllL/8uTJWZlhmZlxR5F5gNXD5lMXzjTGtIrIQeFtEDhpjjn1iZ8Y8BjwGsHr16hn3rZRS6tz4c0TfDBRNeVwItE5fSUSuAf4OuMUYM3Z6uTGm1ffvceBdYOV51KuUUuos+RP0u4AlIlIiIjHAncAnZs+IyErgUSZDvmPK8nQRifV9ngWsB6aexFVKKRVksw7dGGPcIvINYAvgBJ40xlSLyMNAlTFmE/ATIAl4XkQAGo0xtwDlwKMi4mXyj8qPps3WUUpNMeHxMjDqRgRinA4SY7XvoDp/fv0UGWM2A5unLfv+lM+vOcN224Dl51OgUpHgRNcQu0/2cqi1n3G39+PlCzITWVWcxvKCNGKi9PpGdW70cEEpC014vLx6oJVdDb3ERjmoLEilODMBgP4RN/uaevndnha2HunizjVF5KfFW1yxCkca9EpZpHtwjKd3NNLuGmXDkmyuKsv51FH7lUuzOdoxyIt7mnlk6zFuWp7HxQszLapYhSt9L6iUBVyjEzzx4Qn6Rya475IFXH/BvBmHZkSE0txk/vKqJSzOTmLT/lY+rO+yoGIVzjTolZpjA6MTPLWtgeExD3++fgFL5yXPuk1ibBT3XlzMsvwUXjvYxu6TPXNQqbILDXql5pDXa/jGM3s55RrlrrXzKUxP8Htbp0O4Y3URi3OSeHFPC0dODQSxUmUnGvRKzaFfbT/J1iOd3FyZ79eR/HRRTgf3rismNyWO56uaGBidCEKVym406JWaIw1dQ/zo9cNcsTSbdSUZ57yfmCgHd6wpYtzj5fndzXiNdg1Rn02DXqk54PUa/vqF/UQ5hf/9peX4Liw8Z7kpcdy0PJ/6jkE9OatmpUGv1Bz4bVUTuxp6+f7NFeSlBmYu/JoF6VTkpfBmzSl6h8YDsk9lTxr0SgXZ8Libn715hFXF6dy2KnC3ahARbq7MQwReO9gWsP0q+9GgVyrInnj/BJ0DY/ztjWXnPWQzXVpCDFctzaGmzUVdu87CUTPToFcqiLoGx3j0veNcW5HLquJzPwH7WdYvySIrKZZXD7Ti9nhn30BFHA16pYLo396uZ2TCw99cXxa014hyOLi5Mo/uoXF2NeiFVOrTNOiVCpKuwTF+s7ORL60sYHFOUlBfa0lOEiVZibxb1/mJ7pdKgQa9UkHzy20NjLm9fO3yRUF/LRFhY3kuA2Nuth/vDvrrqfCiQa9UEAyNuXnqo5NsrMgN+tH8aQuyEinNTWLrkU5GJzxz8poqPGjQKxUEv93VRP/IBA/OwdH8VBvL5zEy4WHbMb2ISv2RBr1SATbh8fLEBydYsyCdVcXpc/raBenxlM1LZtuxbh2rVx/ToFcqwN6qPUVL3whfvWyhJa9/2ZJshsc97GnsteT1VejRoFcqwJ7e0Uh+ahxXl+da8voLMhMoSo/ng/ouPF5teKY06JUKqIauId4/2sWda+fjdAT2Klh/iQgbSrPpGRqnurXfkhpUaNGgVyqAfrOzcfIGIWuKLK2jPC+FrKQY3j/ahdE2xhFPg16pABlze3iuqolrK3LJTYmztBaHCOsXZ9HSN0Jjz7CltSjradArFSC/P9RO7/AE96wrtroUAFYUpREX7eAjvYAq4mnQKxUgz1c1U5QRz6WLMq0uBYDYKCer5qdzqKUfl95yMKJp0CsVAG39I3x4rIsvrSzEYdFJ2JmsW5iJ16DNziKcBr1SAfDS3haMgS9dVGB1KZ+QlRRLaW4SO0/06FTLCOZX0IvI9SJSJyL1IvLQDM9/W0RqROSAiLwlIsVTnrtPRI76Pu4LZPFKhQJjDC/uaWHNgnSKMxOtLudTLl6YycCom5o2l9WlKIvMGvQi4gR+DtwAVAB3iUjFtNX2AquNMZXAC8CPfdtmAD8A1gFrgR+IyNxeE65UkB1s6ae+Y5AvXRS42wQGUmluMmnx0VTp8E3E8ueIfi1Qb4w5bowZB54Fbp26gjHmHWPM6Tlc24HTP/HXAW8aY3qMMb3Am8D1gSldqdDwu93NxEQ5uHF5ntWlzMghwqridOo7BvUm4hHKn6AvAJqmPG72LTuT+4HXz2ZbEXlARKpEpKqzs9OPkpQKDRMeL68caGNjeS6p8dFWl3NGp5urVZ3U/jeRyJ+gn2kKwYxndUTkXmA18JOz2dYY85gxZrUxZnV2drYfJSkVGrYd66ZnaJxbVuRbXcpnSkuIYUluErtP6knZSORP0DcDU6/nLgRap68kItcAfwfcYowZO5ttlQpXr+5vJTk2istLQ/8AZXVxBq5RN0c7BqwuRc0xf4J+F7BEREpEJAa4E9g0dQURWQk8ymTId0x5agtwrYik+07CXutbplTYG3d72VLdzsZlucRFO60uZ1bleSkkxkZR1aDDN5Fm1qA3xriBbzAZ0LXAc8aYahF5WERu8a32EyAJeF5E9onIJt+2PcAPmfxjsQt42LdMqbD3/tFOXKNuPl8Z2sM2pzkdwsqiNOraBxgec1tdjppDUf6sZIzZDGyetuz7Uz6/5jO2fRJ48lwLVCoQntnRGPB9PlfVRHy0k6be4aDsPxhWzk/jg/ouDrT0c/HC0GjVoIJPr4xV6hxMeLzUtrlYlp9ClCN8fo3yUuOZlxLHXr37VEQJn59QpULI0VMDjLm9LC9MtbqUs7aiKI2m3hG6BsdmX1nZgl9DN8pePF5DbZuLXQ09NPYMU5SRwJKcJFbNTychVn8k/HGo1UV8tJOFWUlWl3LWLixKY0t1O/ua+rjGotsdqrmlR/QRZmTcwyNb63lmZyMdA2NcUJCKa2SC1w+184v3jtE7rFdOzsbt9XK43UVFXopltws8H6nx0SzKTmJfU5/efSpC6OFbBBl3e3nqowZOucb4k9VFVBam4pDJoGroGuKX2xt47L3j/Pn6BeQkW3uHpFB2vHOI0Qkvy/JTrC7lnK2Yn8YLu5tp7BkOyUZsKrD0iD5CeLyGX+84SVPPMHesLmJFUdrHIQ+wICuRr162EI/X8MQHJxgZ91hYbWirbu0nJsrBopzwG7Y5bVl+CtFOYW9jn9WlqDmgQR8hPjrWRX3HIF9cWcAFBTOfQMxLjee+SxcwNOZm86G2Oa4wPHiNoabVRdm8ZKKd4fvrExvlZFl+Kgdb+nF7vFaXo4IsfH9Sld8GRid463AHS3OTWb0g4zPXLUiL57Il2ew+2Ut9x+AcVRg+GrqHGBr3sCw//GbbTLeiKI2RCQ+H27Ulgt1p0EeAN6pP4fYYbvKzje5VZTlkJcXw0t5mxt16tDdVdauLKIdQmhu+wzanLcpOIjk2in1NOnxjdxr0NtfUM8zuxl7WL84iKznWr22inQ6+uLKQ3uEJPjreHeQKw4cxk9NSF+ckERsV+r1tZuN0CBf6WiJon3p706C3uT/UniIpNoorl55dd8WSrESW5CTxQX0XEzqGC0C7a5S+4QnK88J3ts10K4rS8BjDqwe0qaydadDb2CnXKEc7Brl0USax59Bd8YqlOQyNufUWdD61bZNj2WXzki2uJHDyUuPISY7llf168t3ONOhtbNuxLqIcwppZTsCeSUlWIsWZCbx3tAu3V4/qD7e7KEyPJzkudO8kdbZEhMrCVHad7KGtf8TqclSQaNDb1NCYm72Nfaycn0biebQ1uKI0h/6RCfZF+Hxr1+gEzb0jthq2Oa2yIA1j4LUDelRvVxr0NrWroQe313Dpoqzz2k9pbhJ5qXFsO9Yd0ZfL1/mGbcrn2S/os5JjWZafwqsa9LalQW9DHq9h+/FuluQkkZtyfq0MRIS1JRm0u0Zp7o3ct/a17S7SEqLJTfFv5lK4ubkyn31NfTT1DFtdigoCDXobOtY5iGvUfc5j89NdWJhGtFOoOhmZJ2XH3V7qOwYpn5eCSPg1MfPHzZWT11joUb09adDb0L6mPuKiHQGbHRIX7aSyII39zf2MuSOvB86xzkHcXmPL8fnTijISWFGUptMsbUqD3mbG3B6qW/tZXpBGVAB7saxekM6428vB5v6A7TNc1La5iI1ysCArwepSgurmyjyqW10c79TWF3ajQW8ztW0uJjyGFUVpAd3v/IwEcpJj2RVhc+q9xnC4fYDS3OSwumXgubi5Mh8RHb6xI3v/5EagvY19pCVEU5wZ2KNPEWF1cTpNvSN0uEYDuu9Q1tI7wuCYm/I8+1wkdSbzUuNYU5zBK/t1+MZuNOhtZGB0gvqOQVYUfrLXfKBUFqUhwIGWyBm+qW134RAozbV/0AN8/sI8jnYMUqcdLW1Fg95GDrb0Y5i8J2gwpMRFU5KVyIHm/oiZU3+4bYDizEQSYiLjZmzXX5CHQ9CTsjajQW8jNW0uspNjz3vu/GepLEyja3CMtn77D9/0Do3T7hql3Ea9bWaTnRzLJYsyeWV/a8T8MY8EGvQ2MTzupqFriIogTwFclp+CQ+BABMy+qW13AVBm42mVM/l8ZT4N3cMcanFZXYoKEA16m6hrH8BrCHrQJ8ZGsTgniYMtfbY/4jvcNkB2UixZSfa8GvZMrls2jyiH8NpBnX1jFxr0NlHT5iIlLoqC9Pigv1ZlQRq9wxO2bokwOuHheNdgRMy2mS49MYb1i7N47aAO39iFX0EvIteLSJ2I1IvIQzM8v0FE9oiIW0Rum/acR0T2+T42Bapw9UejEx6OnBqgPC8lKLNtpqvIT8HpEA7aePbNkVOT75DKbNjEzB83VebR1DNi6+9xJJk16EXECfwcuAGoAO4SkYppqzUCfwY8M8MuRowxK3wft5xnvWoG7x/tYsJjqMifm1CKi3ayKDuRmjaXbY/4DrcPkBDjZH6Ar0cIF9dVzCPaKdq62Cb8OaJfC9QbY44bY8aBZ4Fbp65gjGkwxhwA9O4UFnijup24aAclWYlz9prleSn0DI1zamBszl5zrni8hrr2AcrmJc/JO6RQlJoQzecWZ/HqgTbb/jGPJP4EfQHQNOVxs2+Zv+JEpEpEtovIF86qOjUrr9fwTl3HnF+if7rBV22b/WZmnOwZYmTCE7HDNqfdVJlPS98I+yNghpXd+ZMMMx3SnM2f+PnGmNXA3cD/EZFFn3oBkQd8fwyqOjs7z2LXqrrVRdfgOEvn+MrNlLhoitLjqWm1X9AfbhvA6RCW5CRZXYqlNlbk+oZv9OKpcOdP0DcDRVMeFwJ+f+eNMa2+f48D7wIrZ1jnMWPMamPM6uzsbH93rYB36joAWGLBJfoVeSm09I3QPzIx568dLMYYattcLMpOPKcbqttJanw0G5Zk85oO34Q9f4J+F7BEREpEJAa4E/Br9oyIpItIrO/zLGA9UHOuxapPe7eugwsLU0k6j/vCnis7Dt90Do7RPTQe8cM2p91UmUdr/yh7myL7nsHhbtagN8a4gW8AW4Ba4DljTLWIPCwitwCIyBoRaQZuBx4VkWrf5uVAlYjsB94BfmSM0aAPkL7hcfY19XH50hxLXj87OZbMxBhbBf1h371hA3XTlnB3TUUuMU6Hzr4Jc34dBhpjNgObpy37/pTPdzE5pDN9u23A8vOsUZ3Be0e78Bq4Ymn2xwE1l0SEivwUttV3MzrhIc4GQx21bS7yU+NIS4ixupSQkBIXzYbSbDYfbOPvbizH4YjMWUjhTq+MDWPv1nWQnhDNhYXB6Vbpj4q8FDzGUHcq/NvaDo65aewZjrjeNrO5uTKPtv5R9jb1Wl2KOkca9GHK6zVsretkQ2k2TguPsooyEkiMjbLF7Jsj7QMYoFzH5z/h6vIcYqIceuepMKZBH6Zq2lx0D41zeam1s5QcIpTPS+bIqQHc3vC+Xq62fbJfUH5a8No8h6PkuGiu8A3feL06+yYcadCHqQ/quwD43JIsiyuZHL4Zc3s50TlkdSnnbMLj5eipQcrmpSARejXsZ7mpMo9TrjF2N+rwTTjSoA9TH9Z3sTQ3mZxk648+F+UkEe0UasJ49s3xziHGPV7KIrBbpT+uLs8lNkpn34QrDfowNDrhYeeJHtYvtv5oHiDa6WBJTjK1bS68YXphTU1bPzFRDhZlR/bVsGeSFBvFlUtz2HywDY8O34QdDfowtOdkL2NuL+sXZ1pdyscq8lNwjbpp7Qu/HvVeY6hpG2BpbjLRTv2VOJObKvPoGBijqqHH6lLUWdKf6jD0QX0XToewbmHoBH1ZbjIOISyHbxq7hxkac89Zm+dwdVVZDnHRDr3zVBjSoA9DHx7rZmVRmiVtD84kITaK4sxESy7cOl81bS6cDpnzxnDhJjE2iqvKcth8sF2Hb8KMBn2Y6R+e4GBzX8iMz09VPi+ZdtcoPUPjVpfiN2MM1a39LMpOtMWVvcF20/J8ugbH2HlCh2/CiQZ9mPnoeDdeExrTKqcLxyZn7a5ReocnWJaXanUpYeHKsmzio528dlBbF4cTDfow82F9FwkxTkvbHpxJZlIsOcmxYRX01a0uBHRapZ8SYqK4qjyH3x9qx+0J7wvkIokGfZj5sL6LdSUZxESF5reuPC+Fhu4h+obDY/imptXF/MwEkuOirS4lbNy8PI+uwXEdvgkjoZkWakatfSMc7xoKyfH50yryUvAaeLcu9O8U1j04RrtrlGX5OmxzNq5YmkNCjJNXdfZN2NCgDyMfhlDbgzMpSI8nOTaKN2tOWV3KrE5PBV2m3SrPSnyMk6vLc3X4Joxo0IeRD+u7yEqKCelpgA4RyvKS2XqkkzG3x+pyPlN1q4u81DjSE7X3/Nm6aXkePUPjbD+uwzfhQIM+TBhj+KC+m/WLs0K+6VZ5XgqDY+6QDoGB0Qmaeob1IqlzdMXSbBJjdPZNuNCgDxNHTg3SNTjG+kWhO2xz2qLsJOKjnbxZ0251KWdU0+bCgE6rPEdx0U6uqcjl9UPtjLt1+CbUadCHidNtideH8Pj8adFOBxtKs/hDTQcmRJuc1bS6yEiMITcl1upSwtYtF+bTNzzBB/Whf+I90mnQh4kP67soyUqkIC3e6lL8srFiHu2uUQ61hN6c+pFxD8c7h1iWp73nz8dlS7JJS4jm5b06fBPqNOjDwITHy47j3SHVrXI2V5Xl4BB4szb0Zt/UtPXjMYYLCnTY5nzERDm4aXkeb9acYmjMbXU56jNo0IeB/U19DI17+FwIz5+fLiMxhtXFGSE5zfJAcz8ZiTEUpofHu6NQduuKAkYmPPwhBP+gqz/SoA8DH9R3IQIXh1BbYn9srMilts1FU8+w1aV8bHDMzbHOQSoLUnXYJgBWF6eTnxrHy3tbrC5FfQYN+jDwYX0XywtSSUsIr/ne118wD4DNIXQF5aGWfrwGKkOwV1A4cjiEW1YU8N7RLroHx6wuR52BBn2IGxpzs7cxNNsSz6YoI4HKwtSQCvoDzf3kJMfqbJsAunVFPh6vYfOh0J1OG+k06EPczhM9uL0mrMbnp7pxeR77m/tDYvimf2SCk91DVBbqsE0glc1LpjQ3if/U4ZuQpUEf4j6o7yI2ysGq4nSrSzknN16QB8Drh6w/qj/Y3IcBKgt02CaQRIRbVxRQdbI3JP6gq0/ToA9xH9Z3sWZBRtje/Wh+ZgIXFKTw2kHr39bvaeyjMD2erGQdtgm0Wy7MB+CVAzqnPhT5FfQicr2I1IlIvYg8NMPzG0Rkj4i4ReS2ac/dJyJHfR/3BarwSNAxMMrh9gEuDaP58zO5cXke+5v6aO617mivtW+EdtcoF80Pz3dGoa4oI4HVxen8p148FZJmDXoRcQI/B24AKoC7RKRi2mqNwJ8Bz0zbNgP4AbAOWAv8QET0N81PHx3rBgiL/jaf5ablk8M3Vp6U3dPYi9MhVBbqRVLBcuuKfOpODXC4PfSuho50/hzRrwXqjTHHjTHjwLPArVNXMMY0GGMOANO7G10HvGmM6THG9AJvAtcHoO6I8N6RLtITosP+Cs7izEQuLErjJYuO9sbdXvY19VGel0JCTJQlNUSCG5fn4XSItkQIQf4EfQHQNOVxs2+ZP/zaVkQeEJEqEanq7NQGSTDZlvj9o52sX5yF0xH+M0S+tLKA2jaXJfeTfbeug+FxDxfN15OwwZSZFMvlpdm8vLcFjzc0m9lFKn+CfqaU8fe76Ne2xpjHjDGrjTGrs7Oz/dy1vR05NUjHwBgbltjj/+PzF+YT5RBesmAK3gu7m0mKjWJJTujesMUubltVSLtrlPeP6gFbKPEn6JuBoimPCwF/35udz7YR7b0jk78ooXzbwLORkRjDFUtzeGlvy5zefq69f5S3Dnewcn6aLd4Zhbqry3NIS4jm+d3NVpeipvAn6HcBS0SkRERigDuBTX7ufwtwrYik+07CXutbpmbx3tFOFuckkR8mbYn98f9cVEDnwBgf+k4yz4VndjbiNYZ1JeE9cylcxEY5+cKKAt6sPkXf8LjV5SifWYPeGOMGvsFkQNcCzxljqkXkYRG5BUBE1ohIM3A78KiIVPu27QF+yOQfi13Aw75l6jOMTnjYeaKHy2xyNH/aVeU5pMRF8eKeuTnaG3d7+c3ORq4ozSZD7ws7Z25fXci4x8um/frmPVT4NQXBGLMZ2Dxt2fenfL6LyWGZmbZ9EnjyPGqMOLsaehhze9lQao/x+dNio5zcsiKf56ua6R0aD/pNubdUt9M5MMafXlJMe7823Jory/JTqchL4fmqZr58yQKry1HolbEh6f2jXcQ4HawrybC6lIC79+Jixtxenqtqmn3l8/Sr7Scpyojn8tKcoL+W+qTbVxdysKVf59SHCJ1UHILeO9LJ6gXptpzzXTYvhbUlGfx6x0n+y2ULg3aCtLbNxc4TPXzvhjI9CeuHZ3Y0BnR/bo/BKcIPX6nhpsr8s9r27nXzA1qL0iP6kNPhmmx7cJlNplXO5MuXFNPUM8K7dR1Be41H3j1GYoyTO9YUzb6yCrjE2CjK8pLZ19Snc+pDgAZ9iHn/aBeA7U7ETnXdsnnkJMfyy49OBmX/J7qGePVAK/deXBx2N2uxk1XF6QyNe6jT4RvLadCHmPePdpKVFENFXorVpQRNtNPB3evms/VIJ8c7BwO+/0ferSfK6eD+y0oCvm/lvyU5ySTHRVF1stfqUiKeBn0I8XoNH9R38bnFWThsPq58z7pi4qId/Ns79QHdb0vfCC/uaeGuNUXkJMcFdN/q7DgdwsqiNI6cGmBgdMLqciKaBn0IqW130TU4buvx+dOyk2P504uLeXlvC8cCeFT/yLuTfzgeuHxRwPapzt1Fxel4Dext7LO6lIimQR9CImF8fqqvXb6I2Cgn//rW0YDsr659gN/sbOKutfMpsNEVxeEsJzmO4swEdjX04DV6UtYqGvQh5P2jnZTNSyYnJTKGHLKSYrnv0gVs2t/K0VMD57UvYwz//ZVqkmKj+PbG0gBVqAJhXUkG3UPjHO8csrqUiKVBHyJGxj3sOtFru6thZ/PAhoUkRDv50euHMedxxLelup1tx7r5zrWlQb/iVp2dZfmpJMQ42XFi7nocqU/SoA8R2451Me7x2qYtsb8yEmP41jWlvHW445x7owyOufnhq7WUzUvm7rV6sU2oiXY6WDU/ndo2F64RPSlrBQ36EPGH2g6SYqNYa8O2B7P5yudKWFGUxj9sqqZz4Ox60hhj+N6LB2nrH+GHX7iAKKf+SIeiNSUZeA1UndSehlbQ34oQYIzh7cOn2FCaRUxU5H1LnA7hJ7dVMjTm4fv/eeishnB+vf0kr+xv5bvXLWXNgsj7IxkuspJiWZyTxM4TPXqlrAUiL1VCUHWri1OuMa4qy7W6FMssyU3mrzaW8vqhdn68pc6vsK9q6OGHr9Zy5dJsHtyg0ylD3cUlmbhG3dRYcDvJSGe/rllh6K3aDkTgiqWRNT4/3YOXL6Spd5hH3j1GfLSTb1695IzrvlHdzjef3Ut+Whw/+5MVtr/AzA7K8pJJT4hm27Eulof5De/DjQZ9CHj78ClWFqWRlRRrdSmWEhH+x60XMDrh4WdvHuFk9zDfva6UvNQ/zokfGnPz1EcN/NOWOpYXpvHEfat1lk2YcIhwycJMNh9qp6VvRK91mEMa9BbrcI2yv7mfv75uqdWlhASHQ/jJbReSkxzHkx+c4LWDrdxwQR4ZiTGMTHh4ZV8rA2NuNlbk8i93rrBlK2c7W1WcwR9qO/joWBe3rdLOonNFf0ss9o6vVe/V5XpzjNOcDuGhG8q4Z918fvbmET461s3A6AQTXsMNF8zjvksXsLIoDREdrgk38TFOVjPsURoAABEKSURBVM5Po+pkL9dfkEdSrEbQXND/ZYttqT5FQVo8S3OTrS4l5BRlJPDPd6z4+LExRsPdBi5dlMWOEz1sP97NNeWROwFhLumsGwsNjE7wwdEubrhgngaYH/T/yB6yk2Mpn5fM9uPdjLu9VpcTETToLfT24Q7GPV5uWD7P6lKUmlOXLclmeNzD7kbtVT8XNOgt9PtD7eQkx7KyKN3qUpSaU8WZCRSlx/PB0U69gGoOaNBbZGTcw7t1nVy3bJ7OAVcRR0TYUJpN7/AE1a39Vpdjexr0Ftl6pIORCQ83XKDDNioyleelkJUUw9YjnefVuVTNToPeIq8faic9IToim5gpBZMXUF1emkNb/yh17ed3PwL12TToLTA64eGt2g6urZin3RZVRFtRlEZ6QjRv13XoUX0QacpY4K3aDgbH3NyyIt/qUpSylNMhXLE0h+beEY52BO7eweqT/Ap6EbleROpEpF5EHprh+VgR+a3v+R0issC3fIGIjIjIPt/HLwJbfnh6aW8LuSmxXLww0+pSlLLcyvlppMVH8/ZhPaoPllmDXkScwM+BG4AK4C4RqZi22v1ArzFmMfDPwD9Oee6YMWaF7+PBANUdtnqHxtl6pINbVxTg1Nk2ShHlcHD50mwae4Y5ckqP6oPBnyP6tUC9Mea4MWYceBa4ddo6twJP+T5/Abha9DLGGb12sI0Jj+FWHbZR6mOritPJSIzhjZp2vDqvPuD8CfoCoGnK42bfshnXMca4gX7g9LhEiYjsFZGtInLZTC8gIg+ISJWIVHV2dp7VFxBuXt7bQmluEhV5KVaXolTIiHI4uKY8l7b+UV492GZ1ObbjT9DPdGQ+/U/umdZpA+YbY1YC3waeEZFPJZwx5jFjzGpjzOrsbPvefKOpZ5iqk718YWWB9m1RaprKwlTmpcTx0zfqmPBoD5xA8ifom4GpjaMLgdYzrSMiUUAq0GOMGTPGdAMYY3YDx4DS8y06XD2/uxkRuHXF9DdESimHCNcuy+Vk9zDP7my0uhxb8SfodwFLRKRERGKAO4FN09bZBNzn+/w24G1jjBGRbN/JXERkIbAEOB6Y0sOL2+Plt7saubw0W++so9QZLM1N5uKFGfzszSP0D09YXY5tzBr0vjH3bwBbgFrgOWNMtYg8LCK3+FZ7AsgUkXomh2hOT8HcABwQkf1MnqR90BjTE+gvIhy8fbiDU64x7lo73+pSlApZIsL/e3MF/SMT/MtbR60uxzb8uvGIMWYzsHnasu9P+XwUuH2G7X4H/O48a7SFZ3Y2kpsSy9VleicppT7LsvxU7lgzn19+1MDd6+azOCfJ6pLCnl4ZOweaeobZeqSTO1YXacsDpfzwnWtLiY928sNXa/QiqgDQ1JkDz1U1IcAdOmyjlF+ykmL51sZSth7pZPPBdqvLCXsa9EE2OuHhmR2NXLk0R0/CKnUW7rukmAsKUviHV6rpH9ETs+dDgz7IXtjdTPfQOF/dsNDqUpQKK1FOB//7i5V0D47x498ftrqcsKZBH0Qer+Hf3z/OhUVprNO+80qdteWFqfzZpSU8vaORnScicsJeQGjQB9Eb1e2c7B7maxsW6pWwSp2j71xbyvyMBL7z/D4Gx9xWlxOWNOiDxBjDL947TnFmAtct09sFKnWuEmOj+OmfXEhz7wj/49Uaq8sJSxr0QfL+0S72N/Xx1csWajtipc7TmgUZfG3DIp7d1cQfak5ZXU7Y0aAPAq/X8OMthylMj+f21YVWl6OULfzVxiWU56Xw3Rf209I3YnU5YUWDPgheO9jGoRYX395YSmyU0+pylLKF2Cgn//eei3B7DF9/eg/jbu1w6S8N+gCb8Hj56Rt1LM1N1i6VSgVYSVYiP76tkn1NffyvzbVWlxM2NOgD7NmdjTR0D/M31y/VsXmlguDG5Xl8ZX0J/7Gtged2Nc2+gdKgD6SOgVF+vKWOixdmcJU2L1MqaL53YxmXLcnib186yLZjXVaXE/I06APo4VdqGJvw8j+/uFznzSsVRNFOBz+/5yJKshJ58Fe7qe8YsLqkkKZBHyDv1HXw6oE2vn7lYhZla1tVpYItJS6aJ/9sDTFRTu59fCdNPcNWlxSyNOgDoH94gr9/6RCLshN58ArtaaPUXCnKSOBX969lZMLD3Y9vp61fp13ORIP+PBlj+OsX9nPKNco/3X6hTqdUao6V56Xwy6+spXdogrv/fQfNvXpkP50G/Xl68sMG3qg5xUM3lLFyfrrV5SgVkS4sSuOpr6yhe3CM2x75iKOndMx+Kr9uJahmtquhhx+9Xss15bnc/7kSq8sJKc/saLS6BBVhVhVn8NuvXcKXn9zJ7Y9+xKP3rmLdwkyrywoJekR/jg63u7j/P3ZRlJ7AP91eqbNslAoB5XkpvPDgJWQkxHDP4zv49faTVpcUEjToz0FTzzBffmIn8TFOfnn/WtISYqwuSSnlU5yZyEtfX89lS7L4+5cP8d3n9zMU4e2NNejPUn3HIHc+tp3RCQ+//Mo6CtMTrC5JKTVNanw0j9+3hm9etZjf7Wnmxn99nz2NvVaXZRkN+rNQ1dDDbb/Yxpjbw9P/5WKWzku2uiSl1Bk4HcK3r13Kbx+4BLfHcNsj2/iHTdW4RiPv/rMa9H4wxvDUtgbufnwH6QkxvPgX61lemGp1WUopP6wtyeD1b13GvRcX89RHDVz90608u7ORCU/kdL/UoJ9Fh2uU+5+q4gebqrl0USYvPHgJ8zN1uEapcJISF83Dt17Af359PYXp8Tz04kE2/mwrL+5pjoh2xzq98gxGxj38+/vH+cXWY7i9hv9+yzK+fEmxzq5RKoxVFqbx4l9cyh9qO/jpG3V8+7n9/OPvD/PlSxZw+6pCclLirC4xKDTop+kaHOOZHY38avtJOgfGuH7ZPB66oYwFWYlWl6aUCgARYWNFLleX5bD1SCdPfHCCn2yp46dv1LGhNJvPV+ZzVVkO6Yn2mU2nQQ+4Rid453AHrx1o490jnYy7vVyxNJv/esVi1pZkWF2eUioIHA7hyrIcrizL4VjnIC/uaealPS18p24/Toewan46a0rSWVWczkXz08N6GrVfQS8i1wP/AjiBx40xP5r2fCzwS2AV0A3cYYxp8D33PeB+wAN80xizJWDVn4Nxt5fGniGqW10cauln54keDrb04zWQmxLLPevmc8+6YhbnaAdKpSLFouwk/vq6Mr6zcSkHW/p5s+YU7x3t5Bdbj+PxGgCW5CSxvDCVRdlJlGQlUpKVyILMROJjQr+/1axBLyJO4OfARqAZ2CUim4wxNVNWux/oNcYsFpE7gX8E7hCRCuBOYBmQD/xBREqNMZ5AfyFjbg9VDb0MjLoZHHMzMDrB4Kib/pEJ2vpHaekbobVvhM7BMczk940Yp4MVRWl848rFXFaazar56Tj0rlBKRSyHQ7iwKI0Li9L47nVLGR53s7+pn90ne6g62cu2+m5e3NPyiW0yE2PISoolK9n3b1IsyXFRJMVGkXj6I8ZJlNNBlENwOmTKvw4cDjAGvMYQH+1kSW7gp237c0S/Fqg3xhwHEJFngVuBqUF/K/APvs9fAP5NJs9a3go8a4wZA06ISL1vfx8Fpvw/Ghx1c8/jOz61PD7aSV5aHAVp8Vy5NIe8tDiK0hOoyE9hUXYSMVE68UgpNbOEmCguWZTJJYv+2DNnaMxNQ/cQJ7qGON45RLtrlK6BMToHx9jT2EvXwDgjE+d2LLuiKI2Xv74+UOV/zJ+gLwCm3pixGVh3pnWMMW4R6Qcyfcu3T9v2U3fMFpEHgAd8DwdFpM6v6v10OJA7m10WEGn3Nou0rznSvl6Yw6/5nrl4Ef/M+ff5JCDfOOfNi8/0hD9BP9NYhvFzHX+2xRjzGPCYH7WEPBGpMsastrqOuRRpX3Okfb2gX3O482fcohkomvK4EGg90zoiEgWkAj1+bquUUiqI/An6XcASESkRkRgmT65umrbOJuA+3+e3AW8bY4xv+Z0iEisiJcASYGdgSldKKeWPWYdufGPu3wC2MDm98kljTLWIPAxUGWM2AU8Av/KdbO1h8o8BvvWeY/LErRv4ejBm3IQYWwxBnaVI+5oj7esF/ZrDmhjzqSFzpZRSNqJzC5VSyuY06JVSyuY06ANERK4XkToRqReRh6yuJ9hEpEhE3hGRWhGpFpH/ZnVNc0VEnCKyV0RetbqWuSAiaSLygogc9n2/L7G6pmASkb/y/UwfEpHfiEjYt7TUoA+AKW0ibgAqgLt87R/szA18xxhTDlwMfD0CvubT/htQa3URc+hfgN8bY8qAC7Hx1y4iBcA3gdXGmAuYnIByp7VVnT8N+sD4uE2EMWYcON0mwraMMW3GmD2+zweY/OX/1FXPdiMihcBNwONW1zIXRCQF2MDkzDqMMePGmD5rqwq6KCDed01QAja49keDPjBmahNh+9A7TUQWACuBTzcbsp//A/wNYP/bEk1aCHQC/59vuOpxEbHtzRmMMS3APwGNQBvQb4x5w9qqzp8GfWD41erBjkQkCfgd8C1jjMvqeoJJRG4GOowxu62uZQ5FARcBjxhjVgJDgG3PQYlIOpPvxkuY7LibKCL3WlvV+dOgD4yIbPUgItFMhvzTxpgXra5nDqwHbhGRBiaH564SkV9bW1LQNQPNxpjT79ZeYDL47eoa4IQxptMYMwG8CFxqcU3nTYM+MPxpE2ErvjbUTwC1xpifWV3PXDDGfM8YU2iMWcDk9/htY0zYH+19FmNMO9AkIkt9i67mky3K7aYRuFhEEnw/41djg5PPeivBADhTmwiLywq29cCfAgdFZJ9v2d8aYzZbWJMKjr8EnvYdxBwH/tzieoLGGLNDRF4A9jA5s2wvNmiFoC0QlFLK5nToRimlbE6DXimlbE6DXimlbE6DXimlbE6DXimlbE6DXoUdEfGIyD5fd8HnRSRhlvX/1s/9NohI1gzL3xWRoN0kWkSuEJFLpzz+DxG5LVivpyKPBr0KRyPGmBW+7oLjwIOzrO9X0FvoCmxw9aUKXRr0Kty9DywGEJF7RWSn72j/UV/f+B8x2Ylwn4g87VvvZRHZ7es5/sC5vKhv3z8RkV0ickBEvuZbfoXvHcDp/u1P+66wRERu9C37QET+VURe9TWEexD4K1+Nl/leYoOIbBOR43p0r86XBr0KW742sjcweXVuOXAHsN4YswLwAPcYYx7ij+8A7vFt+hVjzCpgNfBNEck8h5e/n8nOhmuANcBXRaTE99xK4FtM3ptgIbDed/OKR4EbjDGfA7IBjDENwC+Af/bV+L5vH3nA54CbgR+dQ31KfUxbIKhwFD+l7cL7TPbceQBYBezyHUDHAx1n2P6bIvJF3+dFwBKg+yxruBaonHK0nerbzziw0xjTDOCrcwEwCBw3xpzwrf8bX81n8rIxxgvUiEjuWdam1Cdo0KtwNOI7av+Yb3jkKWPM9z5rQxG5gskOhZcYY4ZF5F3gXG4VJ8BfGmO2zLD/sSmLPEz+ns3UyvqzTN3H2W6r1Cfo0I2yi7eA20QkB0BEMkSk2PfchK+lMkweeff6Qr6MydsgnostwF+c3q+IlM5yQ47DwELfmDxMDjOdNgAkn2MdSs1Kg17ZgjGmBvh74A0ROQC8yeQ4N0x2HzzgOxn7eyDKt84Pge0z7U9ENotI/pRFr4lIs+/jeSZvJVgD7BGRQ0yOv5/xHbIxZgT4r8DvReQD4BTQ73v6FeCL007GKhUw2r1SqTkiIknGmEHfMNPPgaPGmH+2ui5lf3pEr9Tc+arv5Gw1k0NIj1pcj4oQekSvlFI2p0f0Sillcxr0Sillcxr0Sillcxr0Sillcxr0Sillc/8/A3pCPJRUkEkAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot(df['Petal.Length']);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAEGCAYAAAB1iW6ZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nO3deXzU9b3v8dcnK9kg+0JIIEAgbMoSWdwFtKgVqrUWtS6tFXtaq6c97a3tOdfjtdeeLqc97T21te7WqtRqq5yKIqIgOwRlT0ICBAgQsofs6+f+kcHGGMgAk/xm+Twfjzyc38w3Mx9G8uab73wXUVWMMcb4viCnCzDGGOMZFujGGOMnLNCNMcZPWKAbY4yfsEA3xhg/EeLUCycmJuqoUaOcenljjPFJ27Ztq1TVpL4ecyzQR40aRV5enlMvb4wxPklEDp3uMRtyMcYYP+FWoIvIAhEpFJFiEXmoj8f/S0S2u772iUit50s1xhhzJv0OuYhIMPA4cDVQCmwVkWWquvdUG1X9To/23wamDUCtxhhjzsCdHvpMoFhVD6hqG7AUWHSG9rcCr3iiOGOMMe5zJ9DTgSM9rktd932GiIwEsoD3T/P4EhHJE5G8ioqKs63VGGPMGbgT6NLHfafb0Wsx8Jqqdvb1oKo+qaq5qpqblNTnrBtjjDHnyJ1ALwUyelyPAI6dpu1ibLjFGGMc4U6gbwWyRSRLRMLoDu1lvRuJyHggDtjo2RKNMca4o99AV9UO4H5gBZAPvKqqe0TkURFZ2KPprcBStQ3WjTHGEW6tFFXV5cDyXvc93Ov6Ec+VZfrz8ubDTpfwidtmZTpdgjEGWylqjDF+wwLdGGP8hAW6Mcb4CQt0Y4zxExboxhjjJyzQjTHGT1igG2OMn7BAN8YYP2GBbowxfsIC3Rhj/IQFujHG+AkLdGOM8RMW6MYY4ycs0I0xxk9YoBtjjJ+wQDfGGD9hgW6MMX7CAt0YY/yEBboxxvgJC3RjjPETFujGGOMn3Ap0EVkgIoUiUiwiD52mzS0isldE9ojIy54t0xhjTH9C+msgIsHA48DVQCmwVUSWqereHm2ygR8Cl6hqjYgkD1TBxhhj+uZOD30mUKyqB1S1DVgKLOrV5l7gcVWtAVDVcs+WaYwxpj/uBHo6cKTHdanrvp7GAeNEZL2IbBKRBX09kYgsEZE8EcmrqKg4t4qNMcb0yZ1Alz7u017XIUA2cCVwK/C0iMR+5ptUn1TVXFXNTUpKOttajTHGnIE7gV4KZPS4HgEc66PNm6rarqoHgUK6A94YY8wgcSfQtwLZIpIlImHAYmBZrzZvAFcBiEgi3UMwBzxZqDHGmDPrN9BVtQO4H1gB5AOvquoeEXlURBa6mq0AqkRkL/AB8H1VrRqooo0xxnxWv9MWAVR1ObC8130P97itwHddX8YYYxxgK0WNMcZPWKAbY4yfsEA3xhg/4dYYujG+4uXNh50u4RO3zcp0ugQTYKyHbowxfsIC3Rhj/IQFujHG+AkLdGOM8RMW6MYY4ycs0I0xxk9YoBtjjJ+wQDfGGD9hgW6MMX7CAt0YY/yEBboxxvgJC3RjjPETFujGGOMnLNCNMcZPWKAbY4yfsEA3xhg/YYFujDF+wgLdGGP8hFuBLiILRKRQRIpF5KE+Hr9bRCpEZLvr6+ueL9UYY8yZ9HumqIgEA48DVwOlwFYRWaaqe3s1/bOq3j8ANRpjjHGDOz30mUCxqh5Q1TZgKbBoYMsyxhhzttwJ9HTgSI/rUtd9vX1RRHaKyGsiktHXE4nIEhHJE5G8ioqKcyjXGGPM6bgT6NLHfdrr+n+AUap6AfAe8EJfT6SqT6pqrqrmJiUlnV2lxhhjzsidQC8Feva4RwDHejZQ1SpVbXVdPgXM8Ex5xhhj3OVOoG8FskUkS0TCgMXAsp4NRCStx+VCIN9zJRpjjHFHv7NcVLVDRO4HVgDBwLOqukdEHgXyVHUZ8ICILAQ6gGrg7gGs2RhjTB/6DXQAVV0OLO9138M9bv8Q+KFnSzPGGHM23Ap0Y8zpdXYpFQ2ttHd00aVKRGgwiTHhTpdlApAFujHnoKOrix1HatlxpI7DNU20dXR96vHQYOGvH5Vy7eQ0Fk0dTvLQIQ5VagKJBboxZ6FLlS0Hq1mzr4K65nYSo8OZlhFLZnwkkWEhBAnUt3ZwvLaZky0dPLY8n/94O59rJ6fxz/OzyU6JcfqPYPyYBboxbqpvaee1baUUlTcwKiGSG6elk50cjUgfSzUy47htVibF5Q28tq2UP206xPLdx7lxajo/vG4CSTYkYwaABboxbjhS3cQfNx2itb2TRVOHM3NUfN9B3svY5GgeujaH+y4fzR8+PMCz6w7yXv4J/vX6CdySm+HWcxjjLts+15h+HKpq5Nn1BwkPCeJbV41lVlbCWQdxXFQYD12bw9v/fBk5aUP5weu7uOeFPKob2waoahOILNCNOYOSykaeW19CdHgI9142mpTz/HBzTFI0S++dzSM3TGRdUSXX/uZDNh2o8lC1JtBZoBtzGtWNbby46RBDI0K59/LRDIsI9cjzBgUJd1+Sxd++dTFRYSHc/vRmXthQgmrvLZKMOTsW6Mb0obWjkxc3lQBw15yRDB3imTDvadLwYbx5/yVcOS6Jf1+2hx/9bRftnV39f6Mxp2GBbkwvqspr20opP9nKrTMzSYgeuBkpMUNCeerOXL511Rhe2XKEe17Io7G1Y8Bez/g3C3RjeskrqWHPsZMsmJzK2OToAX+9oCDh+5/L4edfvID1xZXc+tQmKhta+/9GY3qxQDemh5qmNt7afZzRSVFcMjZxUF/7losyePKOGew7Uc/Nv9/A4aqmQX194/ss0I1x6VLl9Y9KAfjitBEEOTBHfN6EFF76+mxqm9u56fcb2H20btBrML7LAt0Yl7ySGg5UNHL95DTiosIcq2PGyDhe+8YcwoKFxU9uYtuhasdqMb7FAt0YoKmtgxV7yshKjCJ3VJzT5TA2OYbXv3kxSTHh3PnMFjbbXHXjBlv674c6u5StJdWsK66ktb0TBYYPi2D+xBQy4yOdLs8rvZdfTkt7J5+/IM1rluOnDYvgz0tmc+tTm7j7ua08c1cuFw/yuL7xLdZD9zOHqxr5zaoilu04RsyQECalD2PS8GEcP9nCE2v289LmQ7S0dzpdplcpO9nCloNVzMyKJ21YhNPlfEry0CEsXTKHzPhIvvr8Vj7cV+F0ScaLWaD7kWO1zTy3oYTOri7umD2SJZeN5gtT07lxWjrfu2Yc8yekkH/8JM9vKKG1w0L9lOU7jxMeEszVE1KcLqVPSTHhvLJkNmOSovn6C3l8UFDudEnGS1mg+4nqxjae31DCkNBg7r1sNBPShn5q6CA8JJi5OcksviiT0pom/rjx0GcOZQhExeUNFFc0MDcnmchw7x2BjI8K4+V7ZzE+NYb7XtxmPXXTJwt0P9DW0cXzGw7S2aV89eJRxEaefobG5PRhfGlGBiWVjbyx/eggVul9VJWVe8sYFhHKzKx4p8vpV2xkGC/eM5MxydEseTHPNvUyn2GB7gdWF5ZT2dDGbbMy3Trq7MKMWK7KSWb7kVr2HAvcec4FZfUcqWlm7vhkQoN940chNjKMP90zk4y4SL72/Fab0mg+xa2/xSKyQEQKRaRYRB46Q7ubRURFJNdzJZozOXGyhbVFlUzPjGVMkvvL1K8an8zwYUN44+OjNATg3iFdqryXf4L4qDCmj3R+muLZSIgO56WvzyI5Jpy7n93KztJap0syXqLfQBeRYOBx4FpgInCriEzso10M8ACw2dNFmr6pKm9uP0pYSBALJqed1fcGBwk352bQ0tHFsh3HBqhC77X32EmO17UwLyeZ4CDvmKZ4NpKHDuHle2czLDKUO57ZQv7xk06XZLyAOz30mUCxqh5Q1TZgKbCoj3Y/Bn4OtHiwPnMGO4/WUVLVxLWTU4k+hw/0UocO4arxyew+WsfBysYBqNA7qSqr95WTEBXGhRmxTpdzzobHRvDKvbOJCA3mzme3cKTa9n4JdO4EejpwpMd1qeu+T4jINCBDVf9+picSkSUikicieRUV9in9+ehS5YOCcpJjws9ryOCy7ESGDglhxZ6ygDlgoai8gWO1LVwxLsmR/Vo8KSM+khfvmUlbRxd3PLPZdmkMcO4Eel9/4z/5yReRIOC/gH/p74lU9UlVzVXV3KSkJPerNJ9RcLye8vpWrhx/fqEUGhzE/AkpHK5uCphf21cXVjAsIpSpmb7bO+8pOyWGZ+++iLKTLXz1ua0B+ZmI6ebO7+mlQEaP6xFAz0HXGGAysNo17zkVWCYiC1U1z1OFmn84NWQQFxnKlPTzD6VpmXGsLa5kxZ4TjE8d6pNjyu4qqWykpKqR66ekERI0sDNbXt58eECfv7dbcjP406ZDfOG367nz4pGf+vPdNitzUGsxznDnb/RWIFtEskQkDFgMLDv1oKrWqWqiqo5S1VHAJsDCfABt2F9FaU0zl49L8kj4BgcJn5uYQkVDKzuO+PeMiQ+LKogMC+aiUd4/7/xs5aQO5aZpIyiuaOC1baV0BcgQmvmHfgNdVTuA+4EVQD7wqqruEZFHRWThQBdoPuuJNfuJGRLCjEzPTbebkDaU1KFD+LCowm+DoLy+hYKyemaPTiAsxDfmnZ+t6SPjWDAplZ2ldazYU+Z0OWaQuTU1QlWXA8t73ffwadpeef5lmdM5VNXI2qJK5k9IIcSDi2FEhMvHJfJqXimFZfVMSBvqsef2FhuKqwgJEmaPTnC6lAF1WXYiNU1trC2qJDkmnBkj/e+3EdM3/+ym+LGlW48QHCTMGIDFMFPSY4mNDGWNH+4T0tDawUeHa5iaEXtOUzx9iYjw+QuGMzYpmjc+PsaByganSzKDxALdh7R1dPGXvCPMzUlmWESox58/OEi4dGwih6ubKPGzeelbDlbR0aWDfk6oU4KDhFtnZhIXFcZLmw5zqMq//n+avlmg+5BV+Se692yZOXAzFnJHxhMZFszaIv/ppXd0drHpQDXjUqJJcWOvG38RERbMXXNGAvC157dS19zucEVmoFmg+5CXtxwmPTaCy8cN3Bz+sJAgZo6Kp6CsnpqmtgF7ncG082gdDa0dAdM77ykhOpzbZ2VyqKqJ+1/+iI5O2zLZn1mg+4jSmibWFlVyS27GgM8Tv8i1lezWg76/k5+qsnF/FUkx4Yw9i83L/MnopGgeu3Eya4sq+dXKfU6XYwaQBbqPOLWB1k3T0/tpef7iIsPISY1h66Ean+/RHalp5mhtM3NGJ3jNWaFO+PJFmSy+KIPfrd7P+wUnnC7HDBALdB+xbPsxZoyMI2OQDnmeNTqBxtYO9hzz7e0ANuyvJDwkiGl+ssz/fDyycBIT04bynT/voLTGNvLyRxboPqCwrJ6CsnoWTR0+aK85Njma+KgwNh303VNxTra0s/toHTNGxhEeEux0OY4bEhrM726fTleX8q2XP7YjCP2QBboPWLbjKMFBwnVTzm7P8/MRJMKsrHgOVTVxvK550F7Xk7YcrEYV5vj5QqKzMSoxil986QJ2HKnlJ8vznS7HeJgFupfrPsTiGJeMTSQxOnxQX3tGZhwhQcJmH/xwtKOriy0HqxmXEkPCIL9v3m7B5DTuuTSL5zeU8NbO406XYzzIAt3LfXS4ltKaZhZdOHjDLadEhodwwYhYth+upaW9c9Bf/3zsdk1VnDPGeud9eejaHKZnxvKD13faeLofsUD3cv+z4xhhIUFcMynFkdefPTqets4uPvaxXRg37q8iISqMscmBOVWxP6HBQfxm8TQAvveXHXR1+eeGbIHGAt2LqSor9pRxxbgkYoZ4fqm/O0bERZIeG8HmA1U+c6JRaU0TR2qamTMmwedPJBpIGfGR/O/PT2DTgWqe31DidDnGAyzQvdjO0jqO17WwYFKqo3XMHh1PeX0rB31kP5CN+6sICwliuge3F/ZXt+RmMC8nmZ+9U0BxuW3i5ess0L3YO3vKCAkS5k1IdrSOC0bEEhEazOYD3v/haH1LOztL65ieGceQUJuq2B8R4T++OIXIsGC+++p22n18IVmgs0D3UqrKO7vLmDMmgdjIMEdrCQ0OYnpmLHuPnaS+xbs3eNpysJpOVS62qYpuS44ZwmM3TmFnaR2/+2C/0+WY82CB7qWKyhs4WNnINQ4Pt5wyMyuBTlW2HapxupTTau3oZPPBasanxJAYY1MVz8Z1U9JYNHU4//1+EbtK65wux5wjC3Qv9c7uMkTgcxOdmd3SW1JMOKOTothaUu21R9S9tfM4Da0dXGxTFc/Jowsnkxgdzvdf22FDLz7Kv49u8WHv7C5jemYcyV60f/esrARe2XKYohP1jE/1riPqVJXn1peQFB1uUxX78PLmw261mz8hmT9tPswDr3zMZdkDs03zbbMGbj//QGc9dC90rLaZvcdPco2X9M5PmZAWQ3R4iFeuHN12qIZdR+uYMyawd1U8XxPShpKTGsOq/HJq/WQ//EBige6F3i8oB3B8dktvIUFB5I6Ko7Cs3ut+2J/bUMLQISE2VfE8iQg3XDgcRfm7bQvgc9wKdBFZICKFIlIsIg/18fg3RGSXiGwXkXUiMtHzpQaO9wvKyYyPZIwXHshw0SjX4Rcl3tNLP1bbzDu7y1g8M5OwEOujnK+4yDDm5qSw9/hJ8o/79vbJgabfv/0iEgw8DlwLTARu7SOwX1bVKao6Ffg58CuPVxogmts6WV9cydycZK8cOoiLDGN8agx5JTV0esly8Rc3HUJVuWP2SKdL8RuXjk0kOSac/9l5zLbZ9SHudGdmAsWqekBV24ClwKKeDVS15z/jUYB3/KT7oI0HKmnt6PK64ZaeZmbFU9/awV4v6L01t3XyypbDXD0xZdAO/wgEwUHCoqnp1Da1fzIEaLyfO4GeDhzpcV3quu9TRORbIrKf7h76A309kYgsEZE8EcmrqPCfU+U9aVV+OZFhwcx0nevpjcalxBAbGcoWLzj84s3tR6ltauerl2Q5XYrfyUqMYkZmHOuKKyg72eJ0OcYN7gR6X7/3f6YHrqqPq+oY4AfAv/X1RKr6pKrmqmpuUtLAnVzvq1SV9wvKuSw70atP2AkSYeaoePZXNFJR3+pYHV1dylNrDzAxbSizvPgfQF+2YHIq4SHBvLXzmM9szhbI3An0UiCjx/UI4NgZ2i8FvnA+RQWqgrJ6jte1MC/Hu6Yr9mXGyDiCg4SNByodq2Fl/gn2VzRy3xWjvfLzBn8QFR7CvAnJ7K9opPBEvdPlmH64E+hbgWwRyRKRMGAxsKxnAxHJ7nF5PVDkuRIDx6mxyitzvP+3l5ghoVw4IpZth2ocmcKoqvx+9X4y4iO4fhCP5gtEs7ISSIwO4+1dZV7zQbjpW7+BrqodwP3ACiAfeFVV94jIoyKy0NXsfhHZIyLbge8Cdw1YxX5sVf4JLhgxjOQY71kdeiaXjE2gvVN5eYt7qxA9afPBarYfqWXJZaMJCbapigMpOEhYMCmNioZWr5quaj7LraX/qrocWN7rvod73H7Qw3UFnKqGVj4+UsuD87L7b+wl0oZFMDYpmhc2lPD1S0cP6hzwJ9bsJyEqjC/lZvTf2Jy3CWkxZCVG8V7+CaZmxNrWxF7KujZeYs2+ClTxifHzni4Zm8CJk60s3zV4qwp3ltayurCCr12aZcEySESE66ak0dzWyepCm8borSzQvcSqgnKSYsKZNNy7Nr3qT3ZKDGOTo3lizf5BO5fyl+/uIzYylDvn2EKiwZQeG8G0zFjW76+iutG7tn4w3SzQvUB7ZxcfFlYwd3wyQUG+NVsjSIRvXjmGgrJ63t1bNuCvt+1QNWv2VXDf5WMcO2c1kF09MZUggRV7Bv7/tTl7FuheYGtJNfWtHcz14tWhZ7LwwuGMSojkN6uKB3yu8i/f3UdidBh3XWy9cycMiwjlsuwkdh2t43B1k9PlmF4s0L3ABwXlhAUHcenYRKdLOSchwUHcPzeb/OMnWbn3xIC9zob9lWzYX8U/XTmWyDDbyt8pl2UnEhUeMii/kZmzY4HuBVYVlDNrdDxR4b4bUl+YOpyRCZH8ZlXRgPTSO7uUH/89n+HDhnC7HZDgqPCQYK4cl8SBikaKyxucLsf0YIHusJLKRg5UNDIvxzeHW04JCQ7igbnZ7Dl2kje3n2kh8bl5Zcth8o+f5EfXT7CZLV5gZlY8wyJCWbm3zLYE8CIW6A47tTp0ro9NV+zLjdPSuWDEMH6yPJ+G1g6PPW9tUxu/fLeQWVnxtirUS4QGBzE3J5kjNc0UlNmWAN7CAt1h7xeUMzY5mswE39/6NShI+PcbJlFe38rjHxR77Hl/tXIfdc3tPLJwku3Z4kWmZ8aREBXGyr0nvPbg8EBjge6g+pZ2Nh+s8vnhlp5mjIzjpunpPLP2ICWVjef9fOuLK/njxkPcOWcUE9J8a46+vwsOEuZPSKHsZAu7SuucLsdgge6odUWVtHcqc/0o0AEeWpBDWEgQ//KXHXR0nvtpN7VNbfzLqzsYnRTFDxbkeLBC4ylTRgwjdegQ3ss/YRt3eQELdAe9X1DO0CEhzBjpXwcbJw8dwmM3TmbboRp+/d65bbypqvzob7uoamzl/y2eRkSYfRDqjYKku5de1djGR4drnC4n4FmgO6SrS/mgsJwrxif75W6Bi6amc0vuCB5fXcy6orPfM/33a/azfFcZ3716PJPThw1AhcZTJqTFMCIugvcLys/rNzJz/vwvSXzEjtJaKhvamOsDe5+fq0cWTmJMUjTffuUjdh91f4z1pc2H+Pk7hSyaOpz7Lh89gBUaTxARrpmYSl1zO1sPWS/dSRboDlmVX05wkHDVeP8aP+8pMiyEZ+7KJTIshFuf2tTvr+Sqyp+3Hubf3tjNVeOT+M8vXehze9sEqjFJUYxMiGRNYTnt1kt3jAW6Q97LP8GMkXHERoY5XcqAGpkQxZ/vm018VBh3PL2ZP24s6fMHvrapjftf+ZgfvL6LOaMT+N3tMwj1w6EofyWusfSTLR12CIaDfHetuQ8rrWmioKyeH10XGDM3RsRF8up9c3hw6cc8/OYenltfws0zRpARH0mwCGv2lbNy7wnqWzr4/ufG840rxhBsPXOfMzoxilEJUawprOCiUfH2D7IDLNAdcGp16PwJvr861F0pQ4fwyr2zeb+gnJ+/U8gvVhR+8tjQISFcPi6Jb1wxxj4A9WHdvfRknl53kC0Hq7nERzeb82UW6A5YufcEoxOjGJ0U7XQpg0pEmDchhXkTUmho7eBoTTPN7Z1MGj7UenN+YnRSNFmJUazZ191LH8xjCY2NoQ+6htYONh+oZp6P7n3uKdHhIYxPjWFqRqyFuZ+Z7/oHe8vBKqdLCTj2kzTI1u6roK2zi3kBNNxiAktWYhRjkqJYU1RJW4fNeBlMbgW6iCwQkUIRKRaRh/p4/LsisldEdorIKhGx42RO4738coZFhJLrZ6tDjelp/oQUGls72Gy99EHVb6CLSDDwOHAtMBG4VUQm9mr2MZCrqhcArwE/93Sh/qDTtTr0qvFJfrk61JhTRiZEMTY5mg/3VVgvfRC5kyozgWJVPaCqbcBSYFHPBqr6gaqeOmBwEzDCs2X6h48P11Dd2GbDLSYgzM9JprGtk00HrJc+WNwJ9HTgSI/rUtd9p3MP8HZfD4jIEhHJE5G8iooK96v0E+/llxMSJFwx3n+X+xtzSmZCFNnJ0XxYVEFre6fT5QQEdwK9rxUefe6TKSJfAXKBX/T1uKo+qaq5qpqblBR4obYq/wQzs+IZOiTU6VKMGRTzJ6TQ1NbJRuulDwp3Ar0UyOhxPQL4zKGRIjIf+Fdgoaq2eqY8/3G4qomi8oaAWkxkTEZ8JONTYlhbVEmL9dIHnDuBvhXIFpEsEQkDFgPLejYQkWnAH+gO83LPl+n73ss/AQTW6lBjAOZNSKa53Xrpg6HfQFfVDuB+YAWQD7yqqntE5FERWehq9gsgGviLiGwXkWWnebqA9V7+CbL95OxQY87GiLhIclJjWGe99AHn1tw5VV2uquNUdYyqPua672FVXea6PV9VU1R1qutr4ZmfMbDUNrWx5WA18yda79wEpnk5KTS3d7Jh/9kfdmLcZ5OhB8HKvSfo6FKunZzqdCnGOCI9LoIJqTGsK66krrnd6XL8lgX6IHhndxnpsRFMsZ0ETQCbNyGFlvYunlt/0OlS/JYF+gCrb2lnbVElCyanImJ7fJvANTw2golpQ3lm3UHqmqyXPhAs0AfY+wXltHV22XCLMXTPeKlv6eCZdQecLsUvWaAPsHd2l5EcE870TNuMy5i0YRFcOzmVZ9eXUNvU5nQ5fscCfQA1t3WyurCCz01KtcOOjXF5cH42Da0dPL3WxtI9zQJ9AK0uLKe5vdOGW4zpISd1KNdPSeO59QepabReuidZoA+gZTuOkRgdxsyseKdLMcarPDg/m6b2Tp5aa2PpnmSBPkDqW9pZVVDO5y8YbnufG9PLuJQYrp+SxgsbSqi2XrrHWNIMkBV7TtDW0cUNFw53uhRjvNKD87p76U9+aL10T7FAHyDLdhxjRFwE0zNjnS7FGK+UnRLDwguH88eNJVQ12AatnmCBPgCqGlpZX1zJDRcOt8VExpzBA/OyabFeusdYoA+A5buO09mlLLThFmPOaExSNIumpvPCxhIq6q2Xfr4s0AfAG9uPkZ0cTU5qjNOlGOP1vj13LG0dXfxhzX6nS/F5FugeVlzewLZDNdw8Y4QNtxjjhtFJ0XxhWjp/2nyI8voWp8vxaRboHvbatlKCg4Qbp5/pHG1jTE8PzM2mvVN5YrWNpZ8PC3QP6ujs4q8flXLluCSSY4Y4XY4xPmNUYhQ3uXrpR6qbnC7HZ1mge9CHRRWU17fypdyM/hsbYz7lu9eMQ4D/fLfQ6VJ8lgW6B/0lr5T4qDDm5iQ7XYoxPidtWAT3XjaaN7cfY/uRWqfL8UkW6B5S1dDKe/kn+MLUdMJC7G015lx848oxJEaH8dhbe1FVp8vxOZY8HrJ06xHaO5VbZ9pwizHnKjo8hO9cPY6tJTW8vbvM6XJ8Tog7jURkAfAbIBh4WlV/2uvxy4FfAxcAi1X1NU8X6s06Ort4adMhLh6TQHZK4CtiwrwAAA0mSURBVM09f3nzYadLMH7ky7kZvLjxEI+9lc9V45OJCAt2uiSf0W8PXUSCgceBa4GJwK0iMrFXs8PA3cDLni7QF7yXX86xuhbunDPK6VKM8XkhwUE8snASR2ubecIWG50Vd4ZcZgLFqnpAVduApcCing1UtURVdwJdA1Cj1/vjxhLSYyOYP8E+DDXGE2aPTuCGC4fzxJr9No3xLLgT6OnAkR7Xpa77DFB0op4N+6u4fXam7XtujAf96LocgkR49O97nS7FZ7iTQH2tXz+nj59FZImI5IlIXkVFxbk8hdd5dv1BwkKC+LLNPTfGo9KGRfDg/GxW7j3Bij32Aak73An0UqBnWo0Ajp3Li6nqk6qaq6q5SUlJ5/IUXqWsroXXtx3lltwRJESHO12OMX7nnkuzyEmN4d/f3EN9S7vT5Xg9dwJ9K5AtIlkiEgYsBpYNbFm+4em1B+hU5b7LxzhdijF+KTQ4iP+4aQon6lv45bv7nC7H6/Ub6KraAdwPrADygVdVdY+IPCoiCwFE5CIRKQW+BPxBRPYMZNHeoKaxjZc2H2bRhcPJiI90uhxj/Na0zDjumD2SFzaWkFdS7XQ5Xs2tT/FUdbmqjlPVMar6mOu+h1V1mev2VlUdoapRqpqgqpMGsmhv8NyGEprbO/mnK613bsxA+18LckiPjeB7f9lBU1uH0+V4LZuWcQ5qm9p4fv1BrpmYEpALiYwZbNHhIfzi5gspqWriZ28XOF2O17JAPwePf1BMQ2sH/3LNeKdLMSZgzBmTwFcvGcULGw+xrqjS6XK8kgX6WTpS3cQLGw7xxekjGG9HzBkzqH6wIIexydF859XtdgZpHyzQz9KvVu5DpHvvZmPM4BoSGsxvb5vGyeZ2vvvqdrq6bEfGnizQz8Ku0jre2H6Ur12aRdqwCKfLMSYg5aQO5ZGFk1hbVMnvba+XT7FAd1NHZxc/+tsuEqPDbWaLMQ5bfFEGN1w4nF++W8jqwnKny/EaFuhuen5DCbuO1vHIDZMYOiTU6XKMCWgiws++OIXxqUP59isfs7+iwemSvIIFuhuOVDfxy3f3MTcnmeumpDpdjjEGiAwL4ak7ZxAWHMS9L+RR12RbA1ig96OrS/nhX3chAo8umoRIX3uVGWOcMCIukt9/ZQalNc3c88JWmts6nS7JURbo/fjd6mLWFVfyb9dPZEScLfE3xtvMzIrn14unsu1wDd98aRvtnQF5LANggX5GWw5W86uV+7jhwuF2VqgxXuy6KWn83y9M5oPCCr7z5+0BG+punSkaiE6cbOGBVz4mMz6Sn9w42YZajPFyt88aSX1LBz99u4D2zi7++9bphIUEVp81sP60bjrZ0s5dz26hvqWd390+gxib1WKMT/jGFWN4+PMTWbHnBPe9mBdwG3lZoPfS1tHFN17cRnF5A7//ygwmDh/qdEnGmLPwtUuz+MmNU1izr4IvPbGRsroWp0saNBboPbS0d/LNl7axYX8VP7/5Ai4f5/unKhkTiG6blcnTd+VSUtnIosfX8fHhGqdLGhQW6C4nW9q589ktrCoo58eLJnHT9BFOl2SMOQ9zc1J4/ZsXExocxJee2MgTa/b7/d4vFuhASWUjtzyxkY8O1fDrL0/ljjmjnC7JGOMBOalDeeuBy7hmUgo/fbuArzyzmZLKRqfLGjABH+hv7TzO5/97HWUnW3j27otYNDXd6ZKMMR40LCKUx2+bzn/cNIVdpXVc8+sP+X+rimhp979FSAE7bbGsroUfv7WXt3YeZ1pmLL+9bTrpsbaDojH+SES4dWYmc3OSefR/9vKrlft4efNhHpyfzc0zRhAa7B9924AL9JMt7by48RC/+6CY9i7ln+dn880rxwbcfFVjAlHK0CE8fvt0vrK/ip+vKOCHf93Fb98v5q6LR/LlizIZFuHbU5QDJtAPVzWxdOthXtx0iPqWDuZPSObhz08iM8GW8xsTaOaMSeCv/3Qx7xeU8+SHB/jJ8gL+a2UR10xK4cZp6VwyNtEne+1+HegHKxtZXVjO27vL2HKwGhFYMCmVb101lsnpw5wuzxjjIBFh3oQU5k1IYffROl7Zcpi/7zzOm9uPERMewuXjkrhiXBIXZcUzKiHSJ1aLuxXoIrIA+A0QDDytqj/t9Xg48EdgBlAFfFlVSzxb6ul1dSllJ1vYX9FAYVk924/U8vHhWo7WNgMwNjma739uPDdNT7eThowxnzE5fRiP3TiFh2+YyJrCCt4vKOf9gnLe2nUcgMToMHJSh5KdEk12cgzjUqIZnRRNXGSoVwV9v4EuIsHA48DVQCmwVUSWqereHs3uAWpUdayILAZ+Bnx5IApeW1TBWzuPU1HfSnl9KxX1rVQ2tNLRY37p8GFDmJoZy5LLR3Pl+CRGJkQNRCnGGD8THhLMNZNSuWZSKl1dyv6KBraUVPPRoVqKyutZuuUIzT1mx4QGC0nR4STFdH/FRoYRGRZMRGgwQ0KDiQgLJlgERelS6FJFFa4YlzQgowTu9NBnAsWqegBARJYCi4Cegb4IeMR1+zXgtyIiqurxWfwHKxtZVVD+yZuYkxpDUkw4abERjE2KZmxyNEkx4Z5+WWNMgAkKErJTYshOieH2WSOB7tGAo7XNFJXXc7CyiQpXp7KioZWjtS3sPXaS5vZOmto6ae04/Y6PwyJCHQv0dOBIj+tSYNbp2qhqh4jUAQlAZc9GIrIEWOK6bBCRwnMpOoAk0us9NJ9h79GZed37c7vTBXzWoL9Hd/wM7jj3bx95ugfcCfS+Boh697zdaYOqPgk86cZrGkBE8lQ11+k6vJm9R2dm70///Ok9cmdeTinQ83SHEcCx07URkRBgGFDtiQKNMca4x51A3wpki0iWiIQBi4FlvdosA+5y3b4ZeH8gxs+NMcacXr9DLq4x8fuBFXRPW3xWVfeIyKNAnqouA54BXhSRYrp75osHsugAYsNT/bP36Mzs/emf37xHYh1pY4zxD763ttUYY0yfLNCNMcZPWKB7IRFZICKFIlIsIg85XY83EpFnRaRcRHY7XYs3EpEMEflARPJFZI+IPOh0Td5GRIaIyBYR2eF6j/6P0zWdLxtD9zKurRb20WOrBeDWXlstBDwRuRxoAP6oqpOdrsfbiEgakKaqH4lIDLAN+IL9PfoH6d6EJUpVG0QkFFgHPKiqmxwu7ZxZD937fLLVgqq2Aae2WjA9qOqH2FqH01LV46r6ket2PZBP94pu46LdGlyXoa4vn+7hWqB7n762WrAfRHPORGQUMA3Y7Gwl3kdEgkVkO1AOrFRVn36PLNC9j1vbKBjjDhGJBl4H/llVTzpdj7dR1U5VnUr3CviZIuLTw3cW6N7Hna0WjOmXa1z4deAlVf2r0/V4M1WtBVYDCxwu5bxYoHsfd7ZaMOaMXB/4PQPkq+qvnK7HG4lIkojEum5HAPOBAmerOj8W6F5GVTuAU1st5AOvquoeZ6vyPiLyCrARGC8ipSJyj9M1eZlL6N6hda6IbHd9Xed0UV4mDfhARHbS3ZFaqap/d7im82LTFo0xxk9YD90YY/yEBboxxvgJC3RjjPETFujGGOMnLNCNMcZPWKAbnyAina6pd7tF5C8iEtlP+x+5+bwlIpLY675FIvJGj+sfuk7jOnV9g4gsc91efmouc6/neEREvue6fbeIDD/TaxrjCRboxlc0q+pU186KbcA3+mnvVqCfxgZgTo/rOcBJEUl2XV8MrAdQ1etcqwzP5G5geD9tjDlvFujGF60FxgKIyFdce1pvF5E/uDZb+ikQ4brvJVe7N0Rkm2vf6yVnenJVrQDqRGSs6650upfQX+y6vpju0P9Ub1tE/tW1j/17wHjXfTcDucBLrnoiXM/xbRH5SER2iUiOZ94WE+gs0I1PEZEQ4Fpgl4hMAL4MXOLaYKkTuF1VH+IfPfrbXd/6NVWdQXe4PiAiCf281AbgYhEZDxQBm1zXIcAFdK8s7FnXDLq3aZgG3ARcBKCqrwF5rrqmqmqz61sqVXU68Hvge+f6fhjTU4jTBRjjpgjXNqfQ3UN/BlgCzAC2dm9dQgTd26D25QERudF1OwPIBqrO8Hrr6e6JB9O9xcAW4GG6A7tQVVt6tb8M+JuqNgGcGmM/g1ObZW2j+x8AY86bBbrxFc2uXvgnXBtQvaCqPzzTN4rIlXRvvDRHVZtEZDUwpJ/X2wB8m+5Af0pV60VkCHAlrvHzPpzNPhqtrv92Yj+HxkNsyMX4slXAzac+rBSReBEZ6Xqs3bV9LMAwoMYV5jnAbDeeey/dH2ReBnzsum873R/Gbuij/YfAjSIS4Try7YYej9UDMWfx5zLmnFigG5/lOh/z34B3XTvmraR7Bz2AJ4Gdrg9F3wFCXG1+TPd4+Ge4piAOdz230n3CT6WqtruabARG00egu457+zPdof863cNCpzwPPNHrQ1FjPM52WzTGGD9hPXRjjPETFujGGOMnLNCNMcZPWKAbY4yfsEA3xhg/YYFujDF+wgLdGGP8xP8H9HDA0GcIb/IAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot(df['Petal.Width']);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The distributions look ok, there is not much skew in them. Petal length and petal width seem to be bimodal distributions - we can hope that there might be differences in values of this variables depending on the class. Pairplot will help us discover such dependencies."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Multivariate analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 804.75x720 with 20 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.pairplot(df, hue=\"Species\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "At the diagonal we can see the same variable distributions as I plotted before, but this tame different classes are marked with different colors. By looking at this plots we can see that the distribution of petal length and width for class setosa differ from other classes. We learned that versicolor and verginica are more similar to each other than to setosa only by looking at our data!\n",
    "\n",
    "The scatter plots help us discover some correlations between the variables, and for example petal length and petal width seem to be positively correlated. It is quite logical - the longer the petal the wider it is. There are also some other dependencies, like the one between sepal length and petal width, or sepal length and petal length."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correlation between peral width and length: 0.963, p-val: 2.081610827010091e-85\n"
     ]
    }
   ],
   "source": [
    "corr1, p1 = pearsonr(df['Petal.Width'], df['Petal.Length'])\n",
    "print(f'Correlation between peral width and length: {round(corr1, 3)}, p-val: {p1}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correlation between peral width and length: 0.821, p-val: 1.5156011375418409e-37\n"
     ]
    }
   ],
   "source": [
    "corr2, p2 = pearsonr(df['Petal.Width'], df['Sepal.Length'])\n",
    "print(f'Correlation between peral width and length: {round(corr2, 3)}, p-val: {p2}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correlation between peral width and length: 0.874, p-val: 7.292419251744654e-48\n"
     ]
    }
   ],
   "source": [
    "corr3, p3 = pearsonr(df['Petal.Length'], df['Sepal.Length'])\n",
    "print(f'Correlation between peral width and length: {round(corr3, 3)}, p-val: {p3}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are strong linear relationships between the variables. Their corresponding p-values are very small, indicating that the correlation truly exist.\n",
    "\n",
    "It is very useful piece of information, because it helps us choose our classification model. If there are dependencies between variables one shouln't use linear models because the presence of such dependencies violate its assumption that the variables are independent from each other. In this situation we should model the dependencies directly, or use different kind of model, that is able to handle that, for example a decision tree."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Classification"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The data has been cleaned, and we got some understandying of it. I will work on classification now. First, I will use KNN and decision tree for my baselines, and depending on their performance I may also check how some more sophisticated models perform compared to the baseline. \n",
    "\n",
    "KNN is a simple classifier that stores all the examples from training set, and decides on testing examples label based on nearest neighbours search. It uses euclidean distance, so it is important to scale our data, so all the dimensions contribute to the distance equally."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "KNN accuracy: 1.0\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_train_test = df.copy()\n",
    "y, X = df_train_test.pop('Species'), df_train_test\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)\n",
    "\n",
    "knn = Pipeline(steps=[\n",
    "    ('scaler', StandardScaler()),\n",
    "    ('knn_classifier', KNeighborsClassifier(n_neighbors=5))\n",
    "])\n",
    "\n",
    "knn.fit(X_train, y_train)\n",
    "knn_pred = knn.predict(X_test)\n",
    "print(f'KNN accuracy: {accuracy_score(y_test, knn_pred)}')\n",
    "\n",
    "knn_cm = confusion_matrix(y_test, knn_pred)\n",
    "plt.figure(figsize=(6, 4))\n",
    "sns.heatmap(knn_cm, cmap='Blues', annot=True, square=True, cbar=False, \n",
    "            xticklabels=knn['knn_classifier'].classes_, \n",
    "            yticklabels=knn['knn_classifier'].classes_)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To my suprise KNN works really well for this problem! It is always worth to check how simple methods perform, because if they solve the task there is no need to use complex methods that are often more difficult to implement and interprete."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I will also try a decision tree, because it has this nice property that one can always check the rules it has used to model the data.\n",
    "\n",
    "Decision trees are models that partition the data space in a way that Gini index the regions is minimized- that is they try to find regions in which data-points from the same class lie."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Decision tree accuracy: 1.0\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPgAAAD4CAYAAADB0SsLAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAWDklEQVR4nO3de7hVdZ3H8fdHDwgIalzOwQuDTkBOXqYSZ3QsBS0zUZM06ckujgqGNWm3mazx2qPWjPXMZENcdEzTmSYzRwTCTAGJdAa8IFhqXogoOIyXMBECzvnOH2sdPOK5LDxrnbX38vN6Hp6z1tpr7/3dP9Znr+v+LUUEZlZNu5RdgJkVxwE3qzAH3KzCHHCzCnPAzSqsoeg36P/Oz/gwfRdeXPqdskuwCujXgDqa7jW4WYU54GYV5oCbVZgDblZhDrhZhTngZhXmgJtVmANuVmEOuFmFOeBmFeaAm1WYA25WYQ64WYU54GYV5oCbVZgDblZhDrhZhTngZhXmgJtVmANuVmEOuFmFOeBmFeaAm1WYA25WYQ64WYU54GYV5oCbVZgDblZhDrhZhTngZhXmgJtVWKUDPv3SM/nNPVez7NavbJ92yJh9WXjjF1j6w6/wo385j0G79yuxwtqyZPF9nDLh/Zx0wvu4ftbMssupSfXWRpUO+PfvfIAPfvrfXjPtu5d8lH/89h0cfsZVzF6wnM998riSqqstLS0tXHXlFUybfh23z57L/HlzePqpp8ouq6bUYxtVOuBLHnqaFza88pppo0c28vMHk/+Uex94nFOPe0cZpdWclSseZcSIkew3YgR9+vblhBMnsHDBPWWXVVPqsY0qHfCO/PLptZw07hAAPvS+d7Ff01tKrqg2rG9uZvjew7ePNzY10dzcXGJFtace2yhTwCUNk3SNpHmS7m37V3RxRTjvsls474yjWXLL3zNwwG5s2dpSdkk1IYjXTZNUQiW1qx7bqCHjfLcA/wVMAD4FfBL4v85mljQFmALQsN84GoYe1MMy8/PkqmZOPj/ZLx/1Z4184D21U1uZmpqGs27tuu3j65ubaWxsLLGi2lOPbZR1E31IRFwPbI2IRRFxNnBEZzNHxMyIGBsRY2sp3ADD3jIQSL55vzz5/cz60c9Lrqg2HHTwIaxevYo1a37L1i1bmD9vLseMP7bssmpKPbZR1jX41vTvWkkTgN8D+xVTUn5uvPos3nPYaIbuNZCn5n+Nr02fx8D+u3HepKMBuOPeR7jpjgdKrrI2NDQ0cNFXL2HqlHNpbW3h1ImnMWrU6LLLqin12EaKeP1+xetmkk4CFgMjgGuBPYDLI2J2d8/t/87PdP8Gb2IvLv1O2SVYBfRroMODAZnW4BExJx3cAIzPqygzK1bWo+j/JGkPSX0k3SPpOUkfK7o4M+uZrAfZjo+Il4CTgDXAGOBLhVVlZrnIGvA+6d8Tgf+MiBcKqsfMcpT1KPqdkh4HNgHnSxoGbC6uLDPLQ6Y1eER8GTgSGBsRW4GNwAeLLMzMei7TGlxSH+DjwNHppXmLgOkF1mVmOci6if5dkv3waen4x9Np5xZRlJnlI2vAD4+Iv2w3fq+k5UUUZGb5yXoUvUXSW9tGJP054J9hmdW4rGvwLwELJD0DCBgJnF1YVWaWi6wB/zkwGngbScAfL6wiM8tN1k30+yPiTxHxaEQsj4g/AfcXWZiZ9VyXa3BJw4F9gf6S3gnbf7GyBzCg4NrMrIe620R/P3AWyW+/v9Vu+kvAVzp6gpnVji4DHhE3AjdKOi0ibuulmswsJ1n3wZdIul7STwAkvV3SOQXWZWY5yBrwG4C7gH3S8SeBCwupyMxykzXgQyPih0ArQERswxe6mNW8rAHfKGkIJB1DSzqCpPsmM6thWS90+TwwG3irpCXAMOD0wqoys1xkXYO/FfgA8Dck++K/JvuXg5mVJGvAL077ZHsL8F5gJsnPRc2shmX+NVn6dwIwPSLuAPoWU5KZ5SVrwH8naQZwBjBP0m478VwzK0nWkJ5Bsu99QkT8ARiMu002q3lZ72zyCvDjduNrgbVFFWVm+fBmtlmFOeBmFeaAm1WYA25WYZnuD94Tm7fh+4N34airF5RdQs1bcpHvWN2dzu4P7jW4WYU54GYV5oCbVZgDblZhDrhZhTngZhXmgJtVmANuVmEOuFmFOeBmFeaAm1WYA25WYQ64WYU54GYV5oCbVZgDblZhDrhZhTngZhXmgJtVmANuVmEOuFmFOeBmFeaAm1WYA25WYQ64WYU54GYV5oCbVZgDblZhDrhZhTngZhXmgJtVmANuVmENZRfQm5Ysvo9vfP1KWltamXjahzln8pSySyrdJScfyHtGD+GFjVuYNGMpAFPHHcAxY4bSGsGLG7dy6exf8dzLW0qutDbU2zL0plmDt7S0cNWVVzBt+nXcPnsu8+fN4emnniq7rNLduXwtf/cfy18z7aZfrOYjM5fy0VnLWPzr55h89P7lFFdj6nEZetMEfOWKRxkxYiT7jRhBn759OeHECSxccE/ZZZXu4dUb2LBp22umbdzSsn24f99dIXq7qtpUj8tQlwGXtKukm3urmCKtb25m+N7Dt483NjXR3NxcYkW17fzxBzD3s0dywsFNfHfRs2WXUxPqcRnqMuAR0QIMk9R3Z15U0hRJyyQtu37WzB4VmJfoYDUkqYRK6sO0Bc8y4dv3M39lM5MO37fscmpCPS5DWQ6yrQKWSJoNbGybGBHf6uwJETETmAmweVttbOA1NQ1n3dp128fXNzfT2NhYYkX14Scrm/nXjxzKjEWryi6ldPW4DGXZB/89MCedd1C7f3XloIMPYfXqVaxZ81u2btnC/HlzOWb8sWWXVZNGDO6/ffiYMUNZ9fwrJVZTO+pxGep2DR4RlwNIGpSMxsuFV1WAhoYGLvrqJUydci6trS2cOvE0Ro0aXXZZpbty4tsZO3Iv9hrQh3kXHMmMRas4atRgRg4ZQASs3bCZq+Y9UXaZNaEelyFFdL0FLelg4PvA4HTSc8AnIuKxLG9QK5voteqoqxeUXULNW3LR+LJLqHn9GujwYECWTfSZwOcjYmREjAS+AMzKszgzK0aWgO8eEdtXMxGxENi9sIrMLDdZjqI/I+liks10gI8BPjFqVgeyrMHPBoYBPwZuT4f/tsiizCwfWY6ivwh8thdqMbOcdRpwSXfSxVXIEXFKIRWZWW66WoNf02tVmFkhOg14RCxqG06vRR+Tjj4REVuLLszMeq7bfXBJ44AbSa5JFzBC0icj4r5iSzOznspymuybwPER8QSApDHAfwKHFVmYmfVcltNkfdrCDRARTwJ9iivJzPKSZQ2+TNL1vHqhy5nAg8WVZGZ5yRLwqcCnSc6FC7gPmFZkUWaWjywBbwD+ta2DB0m7ArsVWpWZ5SLLPvg9QP924/2BnxVTjpnlKUvA+7Xv5CEdHlBcSWaWlywB3yjpXW0jkg4DNhVXkpnlJcs++IXArZJ+n47vDUwqriQzy0uWX5MtlXQg8DaSo+iP+1JVs/rQ1a/Jjo2IeyV9aIeHRksiIn5ccG1m1kNdrcGPAe4FTu7gsSDpAMLMalhXvya7NP3r3lvM6lS3R9ElXSBpDyWuk/SQpON7ozgz65lMfbJFxEvA8UAjSX9sXy+0KjPLRZaAt3WofiJwQ0QsbzfNzGpYloA/KOmnJAG/K72FUWuxZZlZHro8D67k3qiXkHSV/ExEvCJpCO422awudBnwiAhJ/x0Rh7Wb9jzwfOGVmVmPZdlEf0DS4YVXYma5y3It+njgU5JWARtJDrBFRBxaZGFm1nNZAv6Bwqsws0J0e39wAEnvBkZHxA2ShgEDIyLTDQh9f3DrKd9DvXsPXjz+jd0fXNKlwD8AF6WT+gA351eamRUly0G2icApJPvfRMTvgUFFFmVm+cgS8C2RbMcHgKTdiy3JzPKSJeA/lDQD2EvSZJIOF2cVW5aZ5SHLUfRWYDHwEskNCC+JiLsLrcrMcpEl4IOAc4AXgB8AjxZakZnlpttN9Ii4PCIOIrm7yT7AIknuF92sDmTZB2+zHlhHch16YzHlmFmespwHnyppIckdToYCk32Zqll9yLIPPhK4MCIeKboYM8tXln7Rv9wbhZhZ/nZmH9zM6owDblZhDrhZhTngZhXmgJtVmANuVmEOuFmFOeBmFeaAm1WYA25WYQ64WYU54GYV5oCbVZgDblZhDrhZhTngZhXmgJtVmANuVmEOuFmFOeBmFZalV9XKWLL4Pr7x9StpbWll4mkf5pzJU8ouqaa4fV7vkpMP5D2jh/DCxi1MmrEUgKnjDuCYMUNpjeDFjVu5dPaveO7lLSVX2rE3zRq8paWFq668gmnTr+P22XOZP28OTz/1VNll1Qy3T8fuXL6Wv/uP5a+ZdtMvVvORmUv56KxlLP71c0w+ev9yisvgTRPwlSseZcSIkew3YgR9+vblhBMnsHDBPWWXVTPcPh17ePUGNmza9pppG7e0bB/u33fX9MbatSnzJrqkRqBf23hErC6kooKsb25m+N7Dt483NjWx4lHfR7GN22fnnD/+ACYcMpyX/7SN875fu/cEyXLrolMk/Rp4FlgErAJ+UnBduYsOvmYllVBJbXL77JxpC55lwrfvZ/7KZiYdvm/Z5XQqyyb614AjgCcj4gDgOGBJV0+QNEXSMknLrp81M4cye66paTjr1q7bPr6+uZnGRt9DsY3b5435ycpmjj1wWNlldCpLwLdGxPPALpJ2iYgFwDu6ekJEzIyIsRExtlaOxB508CGsXr2KNWt+y9YtW5g/by7HjD+27LJqhtsnuxGD+28fPmbMUFY9/0qJ1XQtyz74HyQNBO4DbpG0HtjWzXNqTkNDAxd99RKmTjmX1tYWTp14GqNGjS67rJrh9unYlRPfztiRe7HXgD7Mu+BIZixaxVGjBjNyyAAiYO2GzVw174myy+yUIro+BChpd2ATydr+TGBP4JZ0rd6tzdtq+Rij1YOjrl5Qdgk178GLx3d4wCTLGrwRWBsRm4EbJfUHmoBMATez8mTZB78VaG033pJOM7MalyXgDRGx/Tq8dLhvcSWZWV6yBPz/JJ3SNiLpg8BzxZVkZnnJsg/+KZKj598BBPwW+EShVZlZLroNeEQ8DRyRnipTRPyx+LLMLA+dBlzSxyLiZkmf32E6ABHxrYJrM7Me6moNvnv6d1BvFGJm+es04BExI/17ee+VY2Z56nYfXNIwYDKwf/v5I+Ls4soyszxkOYp+B7AY+BnJRS5mVieyBHxARPxD4ZWYWe6yXOgyR9KJhVdiZrnLEvALSEK+SdJLkv4o6aWiCzOznstyoYtPk5nVqa4udDkwIh6X9K6OHo+Ih4ory8zy0NUa/PPAFOCbHTwWgPvzMatxXV3oMiX9O773yjGzPGW50OVDHUzeAKyIiPX5l2RmeclyHvwc4EigrWOsccADwBhJV0TE9wuqzcx6KEvAW4G/iIhmAElNwHeBvybpadUBN6tRWc6D798W7tR6YExEvABsLaYsM8tDljX4YklzeLWjxdOA+9LulP9QWGVm1mNZAv5p4EPAu0m6bLoJuC2SDtV9hN2shnUZcEm7AndFxHuB23qnJDPLS5f74BHRArwiac9eqsfMcpRlE30zsELS3cDGtokR8dnCqjKzXGQJ+Nz0n5nVmSy/JruxNwoxs/x19WuyH0bEGZJWwOvvEBoRhxZamZn1WFdr8AvSvzcA/0tyRxMzqyOdHkWPiLXp4CBgBnAzcBKwOSJ+0wu1mVkPKbleJcOM0qHAJJIr2dak58brjqQpETGz7Dpqmduoa/XUPlmuRW+zHlgHPA80FlNOr5hSdgF1wG3Utbppn24DLmmqpIXAPcBQYLIPsJnVhyznwUcCF0bEI0UXY2b5ynIe/Mu9UUgvqot9p5K5jbpWN+2T+SCbmdWfnTnIZmZ1xgE3q7DKB1zSWZL2KbuOWifpCkk7fW2DpHFpjz91Q9I+kn70Bp53naS3dzPPpyR94o1Xl6/K74Onp/i+GBHLyq6lbJJE8n/emuNrjiNp35Myzt8QEdvyev881XJtb1RdrsEl7S5prqTlklZKmiTpMEmLJD0o6S5Je0s6HRgL3CLpEUn9JR0n6WFJKyT9u6Td0tf8uqRfSnpU0jXptJMl/U86/8/SHmVLJ+kbks5vN36ZpC9I+pKkpelnuDx9bH9Jv5I0DXgIGCHpe2m7rZD0uXS+76XthaTDJf0ibd//lTRIUj9JN6TPeVjS67rrkjRY0n+n7/9AevVjW30zJf2UpMuvXtNFW61Mx8+SdKukO4GfStpF0jRJj0maI2leu3ZZKGlsOvyypCvTNnqgbdlIX/+L6fCodLlZLukhSW+VNFDSPen4CkkfLLQBIqLu/pFcLjur3fiewC+AYen4JODf0+GFwNh0uB/Jj2bGpOM3ARcCg4EneHWLZq/071vaTTsX+GbZnz2t5Z3AonbjvwQ+QXL6RiRf3HOAo4H9Sbq+PiKd9zDg7nbPbfus3wNOB/oCzwCHp9P3IDmd+gXghnTagcDqtD3HAXPS6dcCl6bDxwKPpMOXAQ8C/WukrY4GVqbjZwFrgMHp+OnAvLQNhwMvAqd3sCwFcHI6/E/AP7b7rF9Mh/8HmNhu2RuQtuUe6bShwFNty1gR/7Jc6FKLVgDXSPoGyYL8InAwcHeyFcquwNoOnvc24NmIeDIdv5GkU8nvkPRcc52kuelrAuwH/JekvUkW/GeL+Tg7JyIeltSYHlsYRvL5DwWOBx5OZxsIjCYJ4m8i4oF0+jPAn0u6lqQjj5/u8PJvA9ZGxNL0vV4CkPRukgATyU0pfwOM2eG57yb58iUi7pU0RK929zU7Ijb1/NPvnE7aavUOs90dSTfgkHyGWyPZjVknaQEd28Kry8mDwPvaPyhpELBvRNye1rE5nd4HuErS0SRfvPsCTSSXgeeuLgMeEU9KOgw4EbgauBt4LCKO7Oap6uT1tkn6K+A44CPAZ0jWQNcC34qI2em+5mX5fIJc/IhkbTMc+AHJmvrqiJjRfiZJ+/ParrZelPSXwPtJvtzOAM5u/xQ6+P0/nbRdhnnaXmtjB4/1lh3bakfta8vyOQG2RroaBlp4fZY6e50zSb5oDouIrZJWkazdC1Gv++D7AK9ExM3ANSR3WRkm6cj08T6SDkpn/yPJT14BHgf2lzQqHf84sEjSQGDPiJhHssn+jvTxPYHfpcOfLPIzvQE/IPkyOp1kAb4LODv9LEjaV9LrfhQkaSiwS0TcBlwM7Hh76MeBfSQdns4/SFIDyV1szkynjQH+jGS3pr3284wDnmvbAijZjm3VlZ8Dp6X74k0kuyA7Lf3caySdCiBpN0kDSJap9Wm4x5NcCl6YulyDA4cA/yypleTuKlOBbcC3003CBuBfgMdI9i2nS9pEco+1vwVuTRfapcB0kn3wOyT1I/nm/Vz6Ppel8/6O5H5sB/TKp8sgIh5LNwN/F8lv99dK+gvg/nQ35WXgYyRrl/b2BW6Q1PblftEOr7tF0iTgWkn9gU3Ae4FpJO24gqStz4qIP6Xv1eay9LUfBV6hRr4Ud2yrdKumM7eRbMmtBJ4k2Y/e8Abf+uPADElXkCynHwZuAe6UtAx4hOQLtTCVP01mtrMkDYyIlyUNIenN6KiIKGQfuWj1ugY3K9IcSXuRHFj9Wr2GG7wGN6u0ujzIZmbZOOBmFeaAm1WYA25WYQ64WYX9P3n9ePNObSN9AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "decision_tree = DecisionTreeClassifier(random_state=42)\n",
    "\n",
    "decision_tree.fit(X_train, y_train)\n",
    "tree_pred = decision_tree.predict(X_test)\n",
    "print(f'Decision tree accuracy: {accuracy_score(y_test, tree_pred)}')\n",
    "\n",
    "tree_cm = confusion_matrix(y_test, tree_pred)\n",
    "plt.figure(figsize=(6, 4))\n",
    "sns.heatmap(tree_cm, cmap='Blues', annot=True, square=True, cbar=False, \n",
    "            xticklabels=decision_tree.classes_, \n",
    "            yticklabels=decision_tree.classes_)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The also perform great on our dataset. Let's plot the tree and see what features it has used!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/svg+xml": [
       "<svg height=\"790pt\" viewBox=\"0.00 0.00 680.00 790.00\" width=\"680pt\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
       "<g class=\"graph\" id=\"graph0\" transform=\"scale(1 1) rotate(0) translate(4 786)\">\n",
       "<title>Tree</title>\n",
       "<polygon fill=\"#ffffff\" points=\"-4,4 -4,-786 676,-786 676,4 -4,4\" stroke=\"transparent\"/>\n",
       "<!-- 0 -->\n",
       "<g class=\"node\" id=\"node1\">\n",
       "<title>0</title>\n",
       "<polygon fill=\"#fcfffd\" points=\"250.5,-782 116.5,-782 116.5,-699 250.5,-699 250.5,-782\" stroke=\"#000000\"/>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"183.5\" y=\"-766.8\">Petal.Length &lt;= 2.45</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"183.5\" y=\"-751.8\">gini = 0.665</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"183.5\" y=\"-736.8\">samples = 104</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"183.5\" y=\"-721.8\">value = [31, 37, 36]</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"183.5\" y=\"-706.8\">class = versicolor</text>\n",
       "</g>\n",
       "<!-- 1 -->\n",
       "<g class=\"node\" id=\"node2\">\n",
       "<title>1</title>\n",
       "<polygon fill=\"#e58139\" points=\"169,-655.5 56,-655.5 56,-587.5 169,-587.5 169,-655.5\" stroke=\"#000000\"/>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"112.5\" y=\"-640.3\">gini = 0.0</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"112.5\" y=\"-625.3\">samples = 31</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"112.5\" y=\"-610.3\">value = [31, 0, 0]</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"112.5\" y=\"-595.3\">class = setosa</text>\n",
       "</g>\n",
       "<!-- 0&#45;&gt;1 -->\n",
       "<g class=\"edge\" id=\"edge1\">\n",
       "<title>0-&gt;1</title>\n",
       "<path d=\"M158.6676,-698.8796C152.0413,-687.7735 144.8684,-675.7513 138.1967,-664.5691\" fill=\"none\" stroke=\"#000000\"/>\n",
       "<polygon fill=\"#000000\" points=\"141.1031,-662.6093 132.9736,-655.8149 135.0917,-666.1959 141.1031,-662.6093\" stroke=\"#000000\"/>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"126.8502\" y=\"-676.3534\">True</text>\n",
       "</g>\n",
       "<!-- 2 -->\n",
       "<g class=\"node\" id=\"node3\">\n",
       "<title>2</title>\n",
       "<polygon fill=\"#fafefc\" points=\"321.5,-663 187.5,-663 187.5,-580 321.5,-580 321.5,-663\" stroke=\"#000000\"/>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"254.5\" y=\"-647.8\">Petal.Length &lt;= 4.75</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"254.5\" y=\"-632.8\">gini = 0.5</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"254.5\" y=\"-617.8\">samples = 73</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"254.5\" y=\"-602.8\">value = [0, 37, 36]</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"254.5\" y=\"-587.8\">class = versicolor</text>\n",
       "</g>\n",
       "<!-- 0&#45;&gt;2 -->\n",
       "<g class=\"edge\" id=\"edge2\">\n",
       "<title>0-&gt;2</title>\n",
       "<path d=\"M208.3324,-698.8796C213.4909,-690.2335 218.9808,-681.0322 224.3076,-672.1042\" fill=\"none\" stroke=\"#000000\"/>\n",
       "<polygon fill=\"#000000\" points=\"227.4436,-673.679 229.5617,-663.2981 221.4322,-670.0924 227.4436,-673.679\" stroke=\"#000000\"/>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"235.6851\" y=\"-683.8366\">False</text>\n",
       "</g>\n",
       "<!-- 3 -->\n",
       "<g class=\"node\" id=\"node4\">\n",
       "<title>3</title>\n",
       "<polygon fill=\"#3fe685\" points=\"244,-544 121,-544 121,-461 244,-461 244,-544\" stroke=\"#000000\"/>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"182.5\" y=\"-528.8\">Petal.Width &lt;= 1.6</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"182.5\" y=\"-513.8\">gini = 0.059</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"182.5\" y=\"-498.8\">samples = 33</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"182.5\" y=\"-483.8\">value = [0, 32, 1]</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"182.5\" y=\"-468.8\">class = versicolor</text>\n",
       "</g>\n",
       "<!-- 2&#45;&gt;3 -->\n",
       "<g class=\"edge\" id=\"edge3\">\n",
       "<title>2-&gt;3</title>\n",
       "<path d=\"M229.3179,-579.8796C224.0867,-571.2335 218.5195,-562.0322 213.1177,-553.1042\" fill=\"none\" stroke=\"#000000\"/>\n",
       "<polygon fill=\"#000000\" points=\"215.9608,-551.0421 207.7896,-544.2981 209.9717,-554.6658 215.9608,-551.0421\" stroke=\"#000000\"/>\n",
       "</g>\n",
       "<!-- 6 -->\n",
       "<g class=\"node\" id=\"node7\">\n",
       "<title>6</title>\n",
       "<polygon fill=\"#9355e9\" points=\"392.5,-544 262.5,-544 262.5,-461 392.5,-461 392.5,-544\" stroke=\"#000000\"/>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"327.5\" y=\"-528.8\">Petal.Width &lt;= 1.75</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"327.5\" y=\"-513.8\">gini = 0.219</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"327.5\" y=\"-498.8\">samples = 40</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"327.5\" y=\"-483.8\">value = [0, 5, 35]</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"327.5\" y=\"-468.8\">class = virginica</text>\n",
       "</g>\n",
       "<!-- 2&#45;&gt;6 -->\n",
       "<g class=\"edge\" id=\"edge6\">\n",
       "<title>2-&gt;6</title>\n",
       "<path d=\"M280.0319,-579.8796C285.391,-571.1434 291.0979,-561.8404 296.6282,-552.8253\" fill=\"none\" stroke=\"#000000\"/>\n",
       "<polygon fill=\"#000000\" points=\"299.6135,-554.6522 301.8592,-544.2981 293.6467,-550.9919 299.6135,-554.6522\" stroke=\"#000000\"/>\n",
       "</g>\n",
       "<!-- 4 -->\n",
       "<g class=\"node\" id=\"node5\">\n",
       "<title>4</title>\n",
       "<polygon fill=\"#39e581\" points=\"113,-417.5 0,-417.5 0,-349.5 113,-349.5 113,-417.5\" stroke=\"#000000\"/>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"56.5\" y=\"-402.3\">gini = 0.0</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"56.5\" y=\"-387.3\">samples = 32</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"56.5\" y=\"-372.3\">value = [0, 32, 0]</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"56.5\" y=\"-357.3\">class = versicolor</text>\n",
       "</g>\n",
       "<!-- 3&#45;&gt;4 -->\n",
       "<g class=\"edge\" id=\"edge4\">\n",
       "<title>3-&gt;4</title>\n",
       "<path d=\"M138.4313,-460.8796C126.0898,-449.2237 112.6798,-436.5587 100.351,-424.9148\" fill=\"none\" stroke=\"#000000\"/>\n",
       "<polygon fill=\"#000000\" points=\"102.5068,-422.1366 92.8335,-417.8149 97.7004,-427.2258 102.5068,-422.1366\" stroke=\"#000000\"/>\n",
       "</g>\n",
       "<!-- 5 -->\n",
       "<g class=\"node\" id=\"node6\">\n",
       "<title>5</title>\n",
       "<polygon fill=\"#8139e5\" points=\"238,-417.5 131,-417.5 131,-349.5 238,-349.5 238,-417.5\" stroke=\"#000000\"/>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"184.5\" y=\"-402.3\">gini = 0.0</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"184.5\" y=\"-387.3\">samples = 1</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"184.5\" y=\"-372.3\">value = [0, 0, 1]</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"184.5\" y=\"-357.3\">class = virginica</text>\n",
       "</g>\n",
       "<!-- 3&#45;&gt;5 -->\n",
       "<g class=\"edge\" id=\"edge5\">\n",
       "<title>3-&gt;5</title>\n",
       "<path d=\"M183.1995,-460.8796C183.3788,-450.2134 183.5722,-438.7021 183.7538,-427.9015\" fill=\"none\" stroke=\"#000000\"/>\n",
       "<polygon fill=\"#000000\" points=\"187.2546,-427.8724 183.9233,-417.8149 180.2556,-427.7547 187.2546,-427.8724\" stroke=\"#000000\"/>\n",
       "</g>\n",
       "<!-- 7 -->\n",
       "<g class=\"node\" id=\"node8\">\n",
       "<title>7</title>\n",
       "<polygon fill=\"#ffffff\" points=\"393.5,-425 259.5,-425 259.5,-342 393.5,-342 393.5,-425\" stroke=\"#000000\"/>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"326.5\" y=\"-409.8\">Petal.Length &lt;= 4.95</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"326.5\" y=\"-394.8\">gini = 0.5</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"326.5\" y=\"-379.8\">samples = 8</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"326.5\" y=\"-364.8\">value = [0, 4, 4]</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"326.5\" y=\"-349.8\">class = versicolor</text>\n",
       "</g>\n",
       "<!-- 6&#45;&gt;7 -->\n",
       "<g class=\"edge\" id=\"edge7\">\n",
       "<title>6-&gt;7</title>\n",
       "<path d=\"M327.1502,-460.8796C327.0814,-452.6838 327.0083,-443.9891 326.937,-435.5013\" fill=\"none\" stroke=\"#000000\"/>\n",
       "<polygon fill=\"#000000\" points=\"330.4352,-435.2683 326.8512,-425.2981 323.4355,-435.3272 330.4352,-435.2683\" stroke=\"#000000\"/>\n",
       "</g>\n",
       "<!-- 14 -->\n",
       "<g class=\"node\" id=\"node15\">\n",
       "<title>14</title>\n",
       "<polygon fill=\"#853fe6\" points=\"545.5,-425 411.5,-425 411.5,-342 545.5,-342 545.5,-425\" stroke=\"#000000\"/>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"478.5\" y=\"-409.8\">Petal.Length &lt;= 4.85</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"478.5\" y=\"-394.8\">gini = 0.061</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"478.5\" y=\"-379.8\">samples = 32</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"478.5\" y=\"-364.8\">value = [0, 1, 31]</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"478.5\" y=\"-349.8\">class = virginica</text>\n",
       "</g>\n",
       "<!-- 6&#45;&gt;14 -->\n",
       "<g class=\"edge\" id=\"edge14\">\n",
       "<title>6-&gt;14</title>\n",
       "<path d=\"M380.3125,-460.8796C392.3818,-451.368 405.3039,-441.1843 417.6788,-431.432\" fill=\"none\" stroke=\"#000000\"/>\n",
       "<polygon fill=\"#000000\" points=\"420.0296,-434.0356 425.7173,-425.0969 415.6968,-428.5377 420.0296,-434.0356\" stroke=\"#000000\"/>\n",
       "</g>\n",
       "<!-- 8 -->\n",
       "<g class=\"node\" id=\"node9\">\n",
       "<title>8</title>\n",
       "<polygon fill=\"#39e581\" points=\"246,-298.5 133,-298.5 133,-230.5 246,-230.5 246,-298.5\" stroke=\"#000000\"/>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"189.5\" y=\"-283.3\">gini = 0.0</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"189.5\" y=\"-268.3\">samples = 2</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"189.5\" y=\"-253.3\">value = [0, 2, 0]</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"189.5\" y=\"-238.3\">class = versicolor</text>\n",
       "</g>\n",
       "<!-- 7&#45;&gt;8 -->\n",
       "<g class=\"edge\" id=\"edge8\">\n",
       "<title>7-&gt;8</title>\n",
       "<path d=\"M278.584,-341.8796C265.0385,-330.1138 250.3092,-317.3197 236.8002,-305.5855\" fill=\"none\" stroke=\"#000000\"/>\n",
       "<polygon fill=\"#000000\" points=\"238.8503,-302.7303 229.0054,-298.8149 234.2599,-308.015 238.8503,-302.7303\" stroke=\"#000000\"/>\n",
       "</g>\n",
       "<!-- 9 -->\n",
       "<g class=\"node\" id=\"node10\">\n",
       "<title>9</title>\n",
       "<polygon fill=\"#c09cf2\" points=\"394.5,-306 264.5,-306 264.5,-223 394.5,-223 394.5,-306\" stroke=\"#000000\"/>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"329.5\" y=\"-290.8\">Petal.Width &lt;= 1.55</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"329.5\" y=\"-275.8\">gini = 0.444</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"329.5\" y=\"-260.8\">samples = 6</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"329.5\" y=\"-245.8\">value = [0, 2, 4]</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"329.5\" y=\"-230.8\">class = virginica</text>\n",
       "</g>\n",
       "<!-- 7&#45;&gt;9 -->\n",
       "<g class=\"edge\" id=\"edge9\">\n",
       "<title>7-&gt;9</title>\n",
       "<path d=\"M327.5493,-341.8796C327.7559,-333.6838 327.9751,-324.9891 328.189,-316.5013\" fill=\"none\" stroke=\"#000000\"/>\n",
       "<polygon fill=\"#000000\" points=\"331.693,-316.3831 328.4463,-306.2981 324.6953,-316.2067 331.693,-316.3831\" stroke=\"#000000\"/>\n",
       "</g>\n",
       "<!-- 10 -->\n",
       "<g class=\"node\" id=\"node11\">\n",
       "<title>10</title>\n",
       "<polygon fill=\"#8139e5\" points=\"246,-179.5 139,-179.5 139,-111.5 246,-111.5 246,-179.5\" stroke=\"#000000\"/>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"192.5\" y=\"-164.3\">gini = 0.0</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"192.5\" y=\"-149.3\">samples = 3</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"192.5\" y=\"-134.3\">value = [0, 0, 3]</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"192.5\" y=\"-119.3\">class = virginica</text>\n",
       "</g>\n",
       "<!-- 9&#45;&gt;10 -->\n",
       "<g class=\"edge\" id=\"edge10\">\n",
       "<title>9-&gt;10</title>\n",
       "<path d=\"M281.584,-222.8796C268.0385,-211.1138 253.3092,-198.3197 239.8002,-186.5855\" fill=\"none\" stroke=\"#000000\"/>\n",
       "<polygon fill=\"#000000\" points=\"241.8503,-183.7303 232.0054,-179.8149 237.2599,-189.015 241.8503,-183.7303\" stroke=\"#000000\"/>\n",
       "</g>\n",
       "<!-- 11 -->\n",
       "<g class=\"node\" id=\"node12\">\n",
       "<title>11</title>\n",
       "<polygon fill=\"#9cf2c0\" points=\"398.5,-187 264.5,-187 264.5,-104 398.5,-104 398.5,-187\" stroke=\"#000000\"/>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"331.5\" y=\"-171.8\">Petal.Length &lt;= 5.45</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"331.5\" y=\"-156.8\">gini = 0.444</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"331.5\" y=\"-141.8\">samples = 3</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"331.5\" y=\"-126.8\">value = [0, 2, 1]</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"331.5\" y=\"-111.8\">class = versicolor</text>\n",
       "</g>\n",
       "<!-- 9&#45;&gt;11 -->\n",
       "<g class=\"edge\" id=\"edge11\">\n",
       "<title>9-&gt;11</title>\n",
       "<path d=\"M330.1995,-222.8796C330.3372,-214.6838 330.4834,-205.9891 330.626,-197.5013\" fill=\"none\" stroke=\"#000000\"/>\n",
       "<polygon fill=\"#000000\" points=\"334.1289,-197.3555 330.7975,-187.2981 327.1299,-197.2378 334.1289,-197.3555\" stroke=\"#000000\"/>\n",
       "</g>\n",
       "<!-- 12 -->\n",
       "<g class=\"node\" id=\"node13\">\n",
       "<title>12</title>\n",
       "<polygon fill=\"#39e581\" points=\"324,-68 211,-68 211,0 324,0 324,-68\" stroke=\"#000000\"/>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"267.5\" y=\"-52.8\">gini = 0.0</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"267.5\" y=\"-37.8\">samples = 2</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"267.5\" y=\"-22.8\">value = [0, 2, 0]</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"267.5\" y=\"-7.8\">class = versicolor</text>\n",
       "</g>\n",
       "<!-- 11&#45;&gt;12 -->\n",
       "<g class=\"edge\" id=\"edge12\">\n",
       "<title>11-&gt;12</title>\n",
       "<path d=\"M307.6688,-103.9815C302.6044,-95.1585 297.2475,-85.8258 292.1532,-76.9506\" fill=\"none\" stroke=\"#000000\"/>\n",
       "<polygon fill=\"#000000\" points=\"295.1807,-75.1942 287.167,-68.2637 289.1097,-78.6789 295.1807,-75.1942\" stroke=\"#000000\"/>\n",
       "</g>\n",
       "<!-- 13 -->\n",
       "<g class=\"node\" id=\"node14\">\n",
       "<title>13</title>\n",
       "<polygon fill=\"#8139e5\" points=\"449,-68 342,-68 342,0 449,0 449,-68\" stroke=\"#000000\"/>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"395.5\" y=\"-52.8\">gini = 0.0</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"395.5\" y=\"-37.8\">samples = 1</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"395.5\" y=\"-22.8\">value = [0, 0, 1]</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"395.5\" y=\"-7.8\">class = virginica</text>\n",
       "</g>\n",
       "<!-- 11&#45;&gt;13 -->\n",
       "<g class=\"edge\" id=\"edge13\">\n",
       "<title>11-&gt;13</title>\n",
       "<path d=\"M355.3312,-103.9815C360.3956,-95.1585 365.7525,-85.8258 370.8468,-76.9506\" fill=\"none\" stroke=\"#000000\"/>\n",
       "<polygon fill=\"#000000\" points=\"373.8903,-78.6789 375.833,-68.2637 367.8193,-75.1942 373.8903,-78.6789\" stroke=\"#000000\"/>\n",
       "</g>\n",
       "<!-- 15 -->\n",
       "<g class=\"node\" id=\"node16\">\n",
       "<title>15</title>\n",
       "<polygon fill=\"#c09cf2\" points=\"540.5,-306 414.5,-306 414.5,-223 540.5,-223 540.5,-306\" stroke=\"#000000\"/>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"477.5\" y=\"-290.8\">Sepal.Width &lt;= 3.1</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"477.5\" y=\"-275.8\">gini = 0.444</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"477.5\" y=\"-260.8\">samples = 3</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"477.5\" y=\"-245.8\">value = [0, 1, 2]</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"477.5\" y=\"-230.8\">class = virginica</text>\n",
       "</g>\n",
       "<!-- 14&#45;&gt;15 -->\n",
       "<g class=\"edge\" id=\"edge15\">\n",
       "<title>14-&gt;15</title>\n",
       "<path d=\"M478.1502,-341.8796C478.0814,-333.6838 478.0083,-324.9891 477.937,-316.5013\" fill=\"none\" stroke=\"#000000\"/>\n",
       "<polygon fill=\"#000000\" points=\"481.4352,-316.2683 477.8512,-306.2981 474.4355,-316.3272 481.4352,-316.2683\" stroke=\"#000000\"/>\n",
       "</g>\n",
       "<!-- 18 -->\n",
       "<g class=\"node\" id=\"node19\">\n",
       "<title>18</title>\n",
       "<polygon fill=\"#8139e5\" points=\"672,-298.5 559,-298.5 559,-230.5 672,-230.5 672,-298.5\" stroke=\"#000000\"/>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"615.5\" y=\"-283.3\">gini = 0.0</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"615.5\" y=\"-268.3\">samples = 29</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"615.5\" y=\"-253.3\">value = [0, 0, 29]</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"615.5\" y=\"-238.3\">class = virginica</text>\n",
       "</g>\n",
       "<!-- 14&#45;&gt;18 -->\n",
       "<g class=\"edge\" id=\"edge18\">\n",
       "<title>14-&gt;18</title>\n",
       "<path d=\"M526.416,-341.8796C539.9615,-330.1138 554.6908,-317.3197 568.1998,-305.5855\" fill=\"none\" stroke=\"#000000\"/>\n",
       "<polygon fill=\"#000000\" points=\"570.7401,-308.015 575.9946,-298.8149 566.1497,-302.7303 570.7401,-308.015\" stroke=\"#000000\"/>\n",
       "</g>\n",
       "<!-- 16 -->\n",
       "<g class=\"node\" id=\"node17\">\n",
       "<title>16</title>\n",
       "<polygon fill=\"#8139e5\" points=\"530,-179.5 423,-179.5 423,-111.5 530,-111.5 530,-179.5\" stroke=\"#000000\"/>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"476.5\" y=\"-164.3\">gini = 0.0</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"476.5\" y=\"-149.3\">samples = 2</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"476.5\" y=\"-134.3\">value = [0, 0, 2]</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"476.5\" y=\"-119.3\">class = virginica</text>\n",
       "</g>\n",
       "<!-- 15&#45;&gt;16 -->\n",
       "<g class=\"edge\" id=\"edge16\">\n",
       "<title>15-&gt;16</title>\n",
       "<path d=\"M477.1502,-222.8796C477.0606,-212.2134 476.9639,-200.7021 476.8731,-189.9015\" fill=\"none\" stroke=\"#000000\"/>\n",
       "<polygon fill=\"#000000\" points=\"480.3724,-189.7851 476.7884,-179.8149 473.3726,-189.844 480.3724,-189.7851\" stroke=\"#000000\"/>\n",
       "</g>\n",
       "<!-- 17 -->\n",
       "<g class=\"node\" id=\"node18\">\n",
       "<title>17</title>\n",
       "<polygon fill=\"#39e581\" points=\"661,-179.5 548,-179.5 548,-111.5 661,-111.5 661,-179.5\" stroke=\"#000000\"/>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"604.5\" y=\"-164.3\">gini = 0.0</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"604.5\" y=\"-149.3\">samples = 1</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"604.5\" y=\"-134.3\">value = [0, 1, 0]</text>\n",
       "<text fill=\"#000000\" font-family=\"Times,serif\" font-size=\"14.00\" text-anchor=\"middle\" x=\"604.5\" y=\"-119.3\">class = versicolor</text>\n",
       "</g>\n",
       "<!-- 15&#45;&gt;17 -->\n",
       "<g class=\"edge\" id=\"edge17\">\n",
       "<title>15-&gt;17</title>\n",
       "<path d=\"M521.9184,-222.8796C534.3579,-211.2237 547.8743,-198.5587 560.301,-186.9148\" fill=\"none\" stroke=\"#000000\"/>\n",
       "<polygon fill=\"#000000\" points=\"562.9741,-189.2065 567.8782,-179.8149 558.1879,-184.0985 562.9741,-189.2065\" stroke=\"#000000\"/>\n",
       "</g>\n",
       "</g>\n",
       "</svg>"
      ],
      "text/plain": [
       "<IPython.core.display.SVG object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "graph = Source(tree.export_graphviz(decision_tree, \n",
    "                                    out_file=None, \n",
    "                                    feature_names=df.columns[0:4], \n",
    "                                    class_names=list(decision_tree.classes_), \n",
    "                                    filled = True))\n",
    "\n",
    "display(SVG(graph.pipe(format='svg')))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Another nice property of decision trees is that they provide a natural way to measure feature importance, by checking which feature contributed to the splitting criterion the most. We can see that petal length is really important. \n",
    "\n",
    "One can notice that in the plotted tree it was the most common feature used to partition the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>importance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Petal.Length</th>\n",
       "      <td>0.892694</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Petal.Width</th>\n",
       "      <td>0.088020</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sepal.Width</th>\n",
       "      <td>0.019286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sepal.Length</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              importance\n",
       "Petal.Length    0.892694\n",
       "Petal.Width     0.088020\n",
       "Sepal.Width     0.019286\n",
       "Sepal.Length    0.000000"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(decision_tree.feature_importances_, \n",
    "             columns=['importance'], index=X_train.columns).sort_values('importance', ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Summary"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this notebook, I analyzed the dataset I was given. After perfoming some sanity checks I identyfied some problems with the data - some cleaning was necessary to perform. Looking at distribution plots of individual features help me understand them better. Pairplots turned out to be useful to discover dependencies between them and decide on type of model to use. I've used KNN and decision tree classifiers, that are both simple and powerful. In addition, the decision trees offer useful interpretation techniques, like measure of feature importance and nice visual representation."
   ]
  },
  {
   "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"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}