Alam_Mini_Prog_A2.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "false"
   },
   "source": [
    "# **Implement a Selection Sort Algorithm**\n",
    "MSDS 432 Module 2\n",
    "\n",
    "Nameyeh Alam"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "Collapsed": "false",
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import time\n",
    "import pandas as pd \n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import random\n",
    "import string\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\", category=Warning)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "false"
   },
   "source": [
    "### Define Selection Sort"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "Collapsed": "false"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2, 3, 5, 6, 10]\n"
     ]
    }
   ],
   "source": [
    "# base code for Selection Sort\n",
    "# as described in Chapter 2 of Grokking Algorithms (Bhargava 2016)\n",
    "\n",
    "# Finds the smallest value in an array\n",
    "def findSmallest(arr):\n",
    "  # Stores the smallest value\n",
    "  smallest = arr[0]\n",
    "  # Stores the index of the smallest value\n",
    "  smallest_index = 0\n",
    "  for i in range(1, len(arr)):\n",
    "    if arr[i] < smallest:\n",
    "      smallest_index = i\n",
    "      smallest = arr[i]      \n",
    "  return smallest_index\n",
    "\n",
    "# Sort array\n",
    "def selectionSort(arr):\n",
    "  newArr = []\n",
    "  for i in range(len(arr)):\n",
    "      # Finds the smallest element in the array and adds it to the new array\n",
    "      smallest = findSmallest(arr)\n",
    "      newArr.append(arr.pop(smallest))\n",
    "  return newArr\n",
    "\n",
    "print(selectionSort([5, 3, 6, 2, 10]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "false"
   },
   "source": [
    "# 1. Generate random data  (4 Data Types)\n",
    "- common random number seed (5) is used to generate each of the arrays<br>\n",
    "- five data arrays of length 5,000, 10,000, 15,000, 20,000, and 25,000 of uniformly distributed random numbers (you may use either integers or floating point)\n",
    "    - #### Extra Credit #1 (6 points) :  \n",
    "        - Do the same thing for each of the following four data types:  (1) uniformly distributed random integer, (2) uniformly distributed random float, (3) randomly generated text string of length 5, and (4) randomly generated text string of length 15.  \n",
    "        - The assignment calls for us to profile the Selection Sort algorithm on a single data type over a dimension of size, but this extra credit will profile the algorithm across both size AND data type"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "false",
    "cell_style": "center"
   },
   "source": [
    "### 1.1 Generate random data - INTEGERS\n",
    "-  Ensure that a common random number seed is used to generate each of the arrays"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "Collapsed": "false",
    "cell_style": "center"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "int"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.random.seed(5)\n",
    "array1 = np.random.randint(low=1, high=1000000, size=5000).tolist()\n",
    "np.random.seed(5)\n",
    "array2 = np.random.randint(low=1, high=1000000, size=10000).tolist()\n",
    "np.random.seed(5)\n",
    "array3 = np.random.randint(low=1, high=1000000, size=15000).tolist()\n",
    "np.random.seed(5)\n",
    "array4 = np.random.randint(low=1, high=1000000, size=20000).tolist()\n",
    "np.random.seed(5)\n",
    "array5 = np.random.randint(low=1, high=1000000, size=25000).tolist()\n",
    "\n",
    "# verify data type\n",
    "type(array1[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "false",
    "cell_style": "center"
   },
   "source": [
    "### 1.2 Generate random data - FLOATS\n",
    "-  Ensure that a common random number seed is used to generate each of the arrays"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "Collapsed": "false",
    "cell_style": "center"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "float"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.random.seed(5)\n",
    "fl_array1 = np.random.random(size=5000).tolist()\n",
    "np.random.seed(5)\n",
    "fl_array2 = np.random.random(size=10000).tolist()\n",
    "np.random.seed(5)\n",
    "fl_array3 = np.random.random(size=15000).tolist()\n",
    "np.random.seed(5)\n",
    "fl_array4 = np.random.random(size=20000).tolist()\n",
    "np.random.seed(5)\n",
    "fl_array5 = np.random.random(size=25000).tolist()\n",
    "\n",
    "#verify data type\n",
    "type(fl_array1[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "false"
   },
   "source": [
    "<p style=\"page-break-after:always;\"></p>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "false"
   },
   "source": [
    "### 1.3 Generate random data - STRINGS - length 5\n",
    "-  Ensure that a common random number seed is used to generate each of the arrays"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "Collapsed": "false",
    "cell_style": "center"
   },
   "outputs": [],
   "source": [
    "np.random.seed(5)\n",
    "str_5_array1 = [''.join(random.choices(string.ascii_letters, k = 5)) for _ in range(5000)]\n",
    "np.random.seed(5)\n",
    "str_5_array2 = [''.join(random.choices(string.ascii_letters, k = 5)) for _ in range(10000)]\n",
    "np.random.seed(5)\n",
    "str_5_array3 = [''.join(random.choices(string.ascii_letters, k = 5)) for _ in range(15000)]\n",
    "np.random.seed(5)\n",
    "str_5_array4 = [''.join(random.choices(string.ascii_letters, k = 5)) for _ in range(20000)]\n",
    "np.random.seed(5)\n",
    "str_5_array5 = [''.join(random.choices(string.ascii_letters, k = 5)) for _ in range(25000)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "false"
   },
   "source": [
    "### 1.4 Generate random data - STRINGS - length 15\n",
    "-  Ensure that a common random number seed is used to generate each of the arrays"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "Collapsed": "false"
   },
   "outputs": [],
   "source": [
    "np.random.seed(5)\n",
    "str_15_array1 = [''.join(random.choices(string.ascii_letters, k = 15)) for _ in range(5000)]\n",
    "np.random.seed(5)\n",
    "str_15_array2 = [''.join(random.choices(string.ascii_letters, k = 15)) for _ in range(10000)]\n",
    "np.random.seed(5)\n",
    "str_15_array3 = [''.join(random.choices(string.ascii_letters, k = 15)) for _ in range(15000)]\n",
    "np.random.seed(5)\n",
    "str_15_array4 = [''.join(random.choices(string.ascii_letters, k = 15)) for _ in range(20000)]\n",
    "np.random.seed(5)\n",
    "str_15_array5 = [''.join(random.choices(string.ascii_letters, k = 15)) for _ in range(25000)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "false"
   },
   "source": [
    "# 2. Execute Selection Sort\n",
    "- Execute the base algorithm (Selection Sort) for each of the arrays, noting the execution time with each execution. \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "false",
    "cell_style": "center"
   },
   "source": [
    "### 2.1 Execute Selection Sort - INTEGERS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "Collapsed": "false",
    "cell_style": "center"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sort time =  1.2628004850000005 seconds\n",
      "Sort time =  5.584781576000001 seconds\n",
      "Sort time =  12.108416 seconds\n",
      "Sort time =  20.188089211999998 seconds\n",
      "Sort time =  40.36099103 seconds\n"
     ]
    }
   ],
   "source": [
    "start = time.perf_counter()\n",
    "array1 = selectionSort((array1))\n",
    "end = time.perf_counter()\n",
    "print(\"Sort time =  %s seconds\" % (end - start))\n",
    "int_sort_time1 = (end - start)*1000 # convert to milliseconds \n",
    "\n",
    "start = time.perf_counter()\n",
    "array2 = selectionSort((array2))\n",
    "end = time.perf_counter()\n",
    "print(\"Sort time =  %s seconds\" % (end - start))\n",
    "int_sort_time2 = (end - start)*1000 # convert to milliseconds \n",
    "\n",
    "start = time.perf_counter()\n",
    "array3 = selectionSort((array3))\n",
    "end = time.perf_counter()\n",
    "print(\"Sort time =  %s seconds\" % (end - start))\n",
    "int_sort_time3 = (end - start)*1000 # convert to milliseconds \n",
    "\n",
    "start = time.perf_counter()\n",
    "array4 = selectionSort((array4))\n",
    "end = time.perf_counter()\n",
    "print(\"Sort time =  %s seconds\" % (end - start))\n",
    "int_sort_time4 = (end - start)*1000 # convert to milliseconds \n",
    "\n",
    "start = time.perf_counter()\n",
    "array5 = selectionSort((array5))\n",
    "end = time.perf_counter()\n",
    "print(\"Sort time =  %s seconds\" % (end - start))\n",
    "int_sort_time5 = (end - start)*1000 # convert to milliseconds "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "false",
    "cell_style": "center"
   },
   "source": [
    "### 2.2 Execute Selection Sort - FLOATS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "Collapsed": "false",
    "cell_style": "center"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sort time =  1.5647980609999905 seconds\n",
      "Sort time =  6.201333333000008 seconds\n",
      "Sort time =  14.360392242000003 seconds\n",
      "Sort time =  25.924489211999997 seconds\n",
      "Sort time =  40.530579394 seconds\n"
     ]
    }
   ],
   "source": [
    "start = time.perf_counter()\n",
    "fl_array1 = selectionSort((fl_array1))\n",
    "end = time.perf_counter()\n",
    "print(\"Sort time =  %s seconds\" % (end - start))\n",
    "fl_sort_time1 = (end - start)*1000 # convert to milliseconds \n",
    "\n",
    "start = time.perf_counter()\n",
    "fl_array2 = selectionSort((fl_array2))\n",
    "end = time.perf_counter()\n",
    "print(\"Sort time =  %s seconds\" % (end - start))\n",
    "fl_sort_time2 = (end - start)*1000 # convert to milliseconds \n",
    "\n",
    "start = time.perf_counter()\n",
    "fl_array3 = selectionSort((fl_array3))\n",
    "end = time.perf_counter()\n",
    "print(\"Sort time =  %s seconds\" % (end - start))\n",
    "fl_sort_time3 = (end - start)*1000 # convert to milliseconds \n",
    "\n",
    "start = time.perf_counter()\n",
    "fl_array4 = selectionSort((fl_array4))\n",
    "end = time.perf_counter()\n",
    "print(\"Sort time =  %s seconds\" % (end - start))\n",
    "fl_sort_time4 = (end - start)*1000 # convert to milliseconds \n",
    "\n",
    "start = time.perf_counter()\n",
    "fl_array5 = selectionSort((fl_array5))\n",
    "end = time.perf_counter()\n",
    "print(\"Sort time =  %s seconds\" % (end - start))\n",
    "fl_sort_time5 = (end - start)*1000 # convert to milliseconds "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "false"
   },
   "source": [
    "### 2.3 Execute Selection Sort - STRINGS - length 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "Collapsed": "false",
    "cell_style": "center"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sort time =  1.7095689699999923 seconds\n",
      "Sort time =  7.7064416970000025 seconds\n",
      "Sort time =  17.290823273 seconds\n",
      "Sort time =  28.18400097 seconds\n",
      "Sort time =  44.81899684899997 seconds\n"
     ]
    }
   ],
   "source": [
    "start = time.perf_counter()\n",
    "str_5_array1 = selectionSort((str_5_array1))\n",
    "end = time.perf_counter()\n",
    "print(\"Sort time =  %s seconds\" % (end - start))\n",
    "str_5_sort_time1 = (end - start)*1000 # convert to milliseconds \n",
    "\n",
    "start = time.perf_counter()\n",
    "str_5_array2 = selectionSort((str_5_array2))\n",
    "end = time.perf_counter()\n",
    "print(\"Sort time =  %s seconds\" % (end - start))\n",
    "str_5_sort_time2 = (end - start)*1000 # convert to milliseconds \n",
    "\n",
    "start = time.perf_counter()\n",
    "str_5_array3 = selectionSort((str_5_array3))\n",
    "end = time.perf_counter()\n",
    "print(\"Sort time =  %s seconds\" % (end - start))\n",
    "str_5_sort_time3 = (end - start)*1000 # convert to milliseconds \n",
    "\n",
    "start = time.perf_counter()\n",
    "str_5_array4 = selectionSort((str_5_array4))\n",
    "end = time.perf_counter()\n",
    "print(\"Sort time =  %s seconds\" % (end - start))\n",
    "str_5_sort_time4 = (end - start)*1000 # convert to milliseconds \n",
    "\n",
    "start = time.perf_counter()\n",
    "str_5_array5 = selectionSort((str_5_array5))\n",
    "end = time.perf_counter()\n",
    "print(\"Sort time =  %s seconds\" % (end - start))\n",
    "str_5_sort_time5 = (end - start)*1000 # convert to milliseconds "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "false",
    "cell_style": "center"
   },
   "source": [
    "### 2.4 Execute Selection Sort - STRINGS - length 15"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "Collapsed": "false",
    "cell_style": "center"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sort time =  2.1625934549999783 seconds\n",
      "Sort time =  8.205712484000003 seconds\n",
      "Sort time =  16.332644362999986 seconds\n",
      "Sort time =  29.447379878999982 seconds\n",
      "Sort time =  48.91004509099997 seconds\n"
     ]
    }
   ],
   "source": [
    "start = time.perf_counter()\n",
    "str_15_array1 = selectionSort((str_15_array1))\n",
    "end = time.perf_counter()\n",
    "print(\"Sort time =  %s seconds\" % (end - start))\n",
    "str_15_sort_time1 = (end - start)*1000 # convert to milliseconds \n",
    "\n",
    "start = time.perf_counter()\n",
    "str_15_array2 = selectionSort((str_15_array2))\n",
    "end = time.perf_counter()\n",
    "print(\"Sort time =  %s seconds\" % (end - start))\n",
    "str_15_sort_time2 = (end - start)*1000 # convert to milliseconds \n",
    "\n",
    "start = time.perf_counter()\n",
    "str_15_array3 = selectionSort((str_15_array3))\n",
    "end = time.perf_counter()\n",
    "print(\"Sort time =  %s seconds\" % (end - start))\n",
    "str_15_sort_time3 = (end - start)*1000 # convert to milliseconds \n",
    "\n",
    "start = time.perf_counter()\n",
    "str_15_array4 = selectionSort((str_15_array4))\n",
    "end = time.perf_counter()\n",
    "print(\"Sort time =  %s seconds\" % (end - start))\n",
    "str_15_sort_time4 = (end - start)*1000 # convert to milliseconds \n",
    "\n",
    "start = time.perf_counter()\n",
    "str_15_array5 = selectionSort((str_15_array5))\n",
    "end = time.perf_counter()\n",
    "print(\"Sort time =  %s seconds\" % (end - start))\n",
    "str_15_sort_time5 = (end - start)*1000 # convert to milliseconds "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "false"
   },
   "source": [
    "# 3. Organize Data \n",
    "- Organize the results of the study into a table showing the size of data array and the time taken to sort the array. \n",
    "- Discuss the differences in timing and how they relate to data type and length of array.  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "false"
   },
   "source": [
    "### 3.1 Create Dataframe - INTEGERS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "Collapsed": "false",
    "cell_style": "center"
   },
   "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>array_length</th>\n",
       "      <th>int_sort_time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5000.0</td>\n",
       "      <td>1262.800485</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10000.0</td>\n",
       "      <td>5584.781576</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>15000.0</td>\n",
       "      <td>12108.416000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>20000.0</td>\n",
       "      <td>20188.089212</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>25000.0</td>\n",
       "      <td>40360.991030</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   array_length  int_sort_time\n",
       "0        5000.0    1262.800485\n",
       "1       10000.0    5584.781576\n",
       "2       15000.0   12108.416000\n",
       "3       20000.0   20188.089212\n",
       "4       25000.0   40360.991030"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_int = pd.DataFrame(np.array([[len(array1),int_sort_time1],\n",
    "                            [len(array2),int_sort_time2], \n",
    "                            [len(array3),int_sort_time3],\n",
    "                            [len(array4),int_sort_time4],\n",
    "                            [len(array5),int_sort_time5]]), \n",
    "                            columns=['array_length','int_sort_time'])\n",
    "df_int"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "false"
   },
   "source": [
    "### 3.2 Create Dataframe - FLOATS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "Collapsed": "false"
   },
   "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>array_length</th>\n",
       "      <th>fl_sort_time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5000.0</td>\n",
       "      <td>1564.798061</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10000.0</td>\n",
       "      <td>6201.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>15000.0</td>\n",
       "      <td>14360.392242</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>20000.0</td>\n",
       "      <td>25924.489212</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>25000.0</td>\n",
       "      <td>40530.579394</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   array_length  fl_sort_time\n",
       "0        5000.0   1564.798061\n",
       "1       10000.0   6201.333333\n",
       "2       15000.0  14360.392242\n",
       "3       20000.0  25924.489212\n",
       "4       25000.0  40530.579394"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_fl = pd.DataFrame(np.array([[len(fl_array1),fl_sort_time1],\n",
    "                                [len(fl_array2),fl_sort_time2], \n",
    "                                [len(fl_array3),fl_sort_time3],\n",
    "                                [len(fl_array4),fl_sort_time4],\n",
    "                                [len(fl_array5),fl_sort_time5]]), \n",
    "                            columns=['array_length','fl_sort_time'])\n",
    "df_fl"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "false"
   },
   "source": [
    "### 3.3 Create Dataframe - STRINGS - length 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "Collapsed": "false"
   },
   "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>array_length</th>\n",
       "      <th>str5_sort_time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5000.0</td>\n",
       "      <td>1709.568970</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10000.0</td>\n",
       "      <td>7706.441697</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>15000.0</td>\n",
       "      <td>17290.823273</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>20000.0</td>\n",
       "      <td>28184.000970</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>25000.0</td>\n",
       "      <td>44818.996849</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   array_length  str5_sort_time\n",
       "0        5000.0     1709.568970\n",
       "1       10000.0     7706.441697\n",
       "2       15000.0    17290.823273\n",
       "3       20000.0    28184.000970\n",
       "4       25000.0    44818.996849"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_str5 = pd.DataFrame(np.array([[len(str_5_array1),str_5_sort_time1],\n",
    "                                [len(str_5_array2),str_5_sort_time2], \n",
    "                                [len(str_5_array3),str_5_sort_time3],\n",
    "                                [len(str_5_array4),str_5_sort_time4],\n",
    "                                [len(str_5_array5),str_5_sort_time5]]), \n",
    "                            columns=['array_length','str5_sort_time'])\n",
    "df_str5"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "false"
   },
   "source": [
    "### 3.4 Create Dataframe - STRINGS - length 15"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "Collapsed": "false"
   },
   "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>array_length</th>\n",
       "      <th>str15_sort_time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5000.0</td>\n",
       "      <td>2162.593455</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10000.0</td>\n",
       "      <td>8205.712484</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>15000.0</td>\n",
       "      <td>16332.644363</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>20000.0</td>\n",
       "      <td>29447.379879</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>25000.0</td>\n",
       "      <td>48910.045091</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   array_length  str15_sort_time\n",
       "0        5000.0      2162.593455\n",
       "1       10000.0      8205.712484\n",
       "2       15000.0     16332.644363\n",
       "3       20000.0     29447.379879\n",
       "4       25000.0     48910.045091"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_str15 = pd.DataFrame(np.array([[len(str_15_array1), str_15_sort_time1],\n",
    "                                [len(str_15_array2),str_15_sort_time2], \n",
    "                                [len(str_15_array3),str_15_sort_time3],\n",
    "                                [len(str_15_array4),str_15_sort_time4],\n",
    "                                [len(str_15_array5),str_15_sort_time5]]), \n",
    "                            columns=['array_length','str15_sort_time'])\n",
    "df_str15"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "false"
   },
   "source": [
    "### 3.5 Merge Dataframes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "Collapsed": "false"
   },
   "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>index</th>\n",
       "      <th>array_length</th>\n",
       "      <th>fl_sort_time</th>\n",
       "      <th>int_sort_time</th>\n",
       "      <th>str15_sort_time</th>\n",
       "      <th>str5_sort_time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>5000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1262.800485</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>10000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5584.781576</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>15000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>12108.416000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>20000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>20188.089212</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>25000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>40360.991030</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0</td>\n",
       "      <td>5000.0</td>\n",
       "      <td>1564.798061</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1</td>\n",
       "      <td>10000.0</td>\n",
       "      <td>6201.333333</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2</td>\n",
       "      <td>15000.0</td>\n",
       "      <td>14360.392242</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>3</td>\n",
       "      <td>20000.0</td>\n",
       "      <td>25924.489212</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>4</td>\n",
       "      <td>25000.0</td>\n",
       "      <td>40530.579394</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0</td>\n",
       "      <td>5000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1709.568970</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>1</td>\n",
       "      <td>10000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>7706.441697</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>2</td>\n",
       "      <td>15000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>17290.823273</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>3</td>\n",
       "      <td>20000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>28184.000970</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>4</td>\n",
       "      <td>25000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>44818.996849</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0</td>\n",
       "      <td>5000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2162.593455</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>1</td>\n",
       "      <td>10000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8205.712484</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>2</td>\n",
       "      <td>15000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>16332.644363</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>3</td>\n",
       "      <td>20000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>29447.379879</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>4</td>\n",
       "      <td>25000.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>48910.045091</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    index  array_length  fl_sort_time  int_sort_time  str15_sort_time  \\\n",
       "0       0        5000.0           NaN    1262.800485              NaN   \n",
       "1       1       10000.0           NaN    5584.781576              NaN   \n",
       "2       2       15000.0           NaN   12108.416000              NaN   \n",
       "3       3       20000.0           NaN   20188.089212              NaN   \n",
       "4       4       25000.0           NaN   40360.991030              NaN   \n",
       "5       0        5000.0   1564.798061            NaN              NaN   \n",
       "6       1       10000.0   6201.333333            NaN              NaN   \n",
       "7       2       15000.0  14360.392242            NaN              NaN   \n",
       "8       3       20000.0  25924.489212            NaN              NaN   \n",
       "9       4       25000.0  40530.579394            NaN              NaN   \n",
       "10      0        5000.0           NaN            NaN              NaN   \n",
       "11      1       10000.0           NaN            NaN              NaN   \n",
       "12      2       15000.0           NaN            NaN              NaN   \n",
       "13      3       20000.0           NaN            NaN              NaN   \n",
       "14      4       25000.0           NaN            NaN              NaN   \n",
       "15      0        5000.0           NaN            NaN      2162.593455   \n",
       "16      1       10000.0           NaN            NaN      8205.712484   \n",
       "17      2       15000.0           NaN            NaN     16332.644363   \n",
       "18      3       20000.0           NaN            NaN     29447.379879   \n",
       "19      4       25000.0           NaN            NaN     48910.045091   \n",
       "\n",
       "    str5_sort_time  \n",
       "0              NaN  \n",
       "1              NaN  \n",
       "2              NaN  \n",
       "3              NaN  \n",
       "4              NaN  \n",
       "5              NaN  \n",
       "6              NaN  \n",
       "7              NaN  \n",
       "8              NaN  \n",
       "9              NaN  \n",
       "10     1709.568970  \n",
       "11     7706.441697  \n",
       "12    17290.823273  \n",
       "13    28184.000970  \n",
       "14    44818.996849  \n",
       "15             NaN  \n",
       "16             NaN  \n",
       "17             NaN  \n",
       "18             NaN  \n",
       "19             NaN  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df=pd.concat([df_int, df_fl, df_str5, df_str15], axis=0, join='outer')\n",
    "df=df.reset_index()\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "false"
   },
   "source": [
    "## 4. Generate Plots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "Collapsed": "false",
    "cell_style": "center"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAasAAAErCAYAAABgoay1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOydd1hUx9fHv3d36b1IEURpioItECNWlKIRe49ogpro7000Go2xJaJRLLEXNBFbYomKsUSTqBQJKtYoIDYEGyodRDrs7nn/INywwMJSBZzP8/A87Ny5Z87MvXvP3jln5nBERGAwGAwGoxEjeNsKMBgMBoNRFcxYMRgMBqPRw4wVg8FgMBo9zFgxGAwGo9HDjBWDwWAwGj3MWDEYDAaj0dMsjNW3334LOzu7t61GnRMbGwuO43D16tV6b2vixIkYOHBgvbfDqDvEYjE4jsPhw4fftioAmu/3kFE3BAUFgeM4JCYm1uj8JmGsvL29wXFcub+SL+mCBQtw6dKlWrcTGhqKoUOHwsLCAhzHYfXq1TWWFRYWBnd3d7Ro0QJqampo06YNxowZg/j4+FrrWRv27dsHkUhUrtzPzw+//vprg+gQFRWFESNGwNTUFKqqqjA3N8eQIUMQGRlZa9ny+leWkgd9RX+zZ8+utR51jYuLCz799FOZMpFIhISEBAwfPrxe2+7Vq5fcsSr5e/HiRZ19DxsjAwcOhFAoxJ9//vm2VZFLr1698L///e9tqwEAaNOmDVasWFGnMqv+VjcSevfujaNHj8qU6erqAgA0NTWhqalZ6zays7Ph4OCAiRMnYsaMGTWWEx0dDQ8PD0ybNg3r16+HlpYWnj59ijNnziArK6vWetYHOjo6DdJOUlIS+vXrB3d3d/z+++8wMjLCixcvcO7cOWRkZNRKdmFhYbXP+fHHHzFs2DCZMg0NjVrp0ZCYmJjUexu///67zNiamppi27ZtGDVqFF9mZGQEgUBQJ9/DxsbTp08RFhaGOXPmYOfOnRg0aFCl9aVSKYgIQqFQpryoqAhKSkr1qWrzhpoAn3zyCbm6uso9vnjxYmrXrp1M2bp166hly5akpqZGAwcOpH379hEASkhIUKhNMzMzWrVqVY30Xbt2LRkbG1dZLyEhgSZNmkSGhoakqalJPXv2pIsXL/LHHz16RADoypUrCp9DRBQTE0MjR44kPT09UlNTo06dOtGff/5JgYGBBEDmb+rUqURE5OXlRQMGDOBlSKVSWr16NbVp04aUlJTIysqKtmzZUm6Mli1bRjNmzCBdXV0yMjKiefPmkUQikdvngIAA4jiOcnNzKx2bFy9e0JgxY0hHR4dUVVXJxcWFbt26xR8v6cuff/5JPXr0IBUVFdq6davc/pWlqKiIANCvv/4qV4eDBw+SsrIy3bx5ky/btWsXqaqqUmRkJF924MAB6tSpE6moqFCbNm1o7ty5lJOTwx+XSqW0efNmsrOzI2VlZTIyMqKxY8fKjGPZe630Pe/l5VWuXxcvXqywD4qOW1BQEPXs2ZNUVVXJ3t6eAgMD5Y5DWQDQ/v37y5WX/R6WfD506BBZWVmRmpoajRgxgrKysujo0aNka2tLWlpaNGbMGHrz5o2MrKrGtCzdunWj//u//ytXbmtrS4sXLyYioqioKHJzcyMdHR1SV1cnOzs7OnjwYJX9XbRoEY0dO5bi4+NJWVmZXr58WWG/Dx06RG3btiWhUEj379/nv1MbN24kCwsL4jiO8vLy6K+//qI+ffqQnp4e6ejoUN++fenGjRu8vAkTJtCHH35YTo9evXrJvZ+JiHr27EnTp0+Xe/zNmzc0Y8YMMjU1JTU1NXrvvffo5MmT/PGS501AQAANGjSI1NTUyMrKqtwYxcbGkpubG6moqJCFhQXt2LFDpu2ePXuWu1/j4+Nrfe81S2N15MgREolEtGXLFoqJiaHdu3eTiYlJnRirkgEvayBKc+DAARKJRHT27Fm5dXJycqht27Y0ZswYunnzJj169IiWLVtGKioq9PDhQyIqb6wUOefly5fUokULcnd3p8uXL1NsbCydOHGCzp49SwUFBbRp0yYSCoWUkJBACQkJlJmZSUTljdWmTZtITU2N/P39KSYmhvz8/EhZWZn27dsnM0a6urr0ww8/UExMDB08eJCEQiH98ssvcvt96dIlAkB79+6Va9SkUik5OjpS165d6fLlyxQZGUmjRo0ifX19SktLk7kO7du3p9OnT9Pjx4/p6dOncvtXFkWMFRGRt7c32djYUFZWFt27d4/U1dVp+/bt/HF/f3/S19en/fv3U1xcHIWGhpK9vT15e3vzdRYtWkSamprk5+dHDx8+pJs3b5Kvr6/MOFZmrF6/fk3Ozs40YcIEvl+FhYXl+lCdcevSpQudO3eOHj58SBMmTCA9PT25Y1WW6hgrDQ0NGjJkCEVFRdGFCxdIX1+fPDw8yNPTkyIjIyk0NJQMDQ1p0aJF1RrTsmzbto309PSooKCAL7ty5QoB4L8b7du3p4kTJ9K9e/coLi6O/vjjD/rjjz8q7WtRURGZmJjw9dzd3Wn58uXl+q2urk4uLi507do1evDgAWVlZZGXlxdpaWnRqFGjKCIigqKiokgikdCxY8coICCAHj58SNHR0eTt7U0GBgaUnp5ORERhYWEkEAjo2bNnfBsPHjwgAHT16lW5ulZmrKRSKfXu3Zv69etHly5dori4ONqxYwcpKSlRaGgoEf33vLGysqKAgAB69OgRff311yQSiSguLo6IiCQSCTk4OJCzszNdv36dbt++TR4eHqStrc23nZaWRubm5jR//nz+fpVIJLW+95qMsRIKhaShocH/WVlZ8cfLfkm6detW7saeO3dunRir8PBwateuHf3zzz9yzxWLxeTt7U0cx5GBgQENHDiQ1qxZQ/Hx8Xwdf39/srCwILFYLHNu7969ae7cuURU3lgpcs6CBQvI1NRU7q/QvXv3klAoLFde1liZmJjQwoULZerMmDGDbG1t+c9mZmY0YsQImTqurq40ceLEigfmXxYtWkQikYi0tbWpX79+tHTpUnrw4AF//OzZswRApiw3N5eMjIz4h3zJjX/o0CGF+leWkge9ioqKzH2loaFBx44d4+tlZ2eTnZ0djR8/njp27EgjR46UkWNmZkb+/v4yZcHBwcRxHL1584YyMzNJRUWFNm7cKFeXqowVEVHfvn3L/aoua6yqM26nTp3i68THx/O/eBWhOsZKSUmJN5RERNOmTSOhUEipqal82eeff04ffPCBzHhUNqYVkZaWRsrKyjLX7v/+7/+oe/fu/Gd1dfUK9a6MY8eOkYmJCf+dO3jwILVu3Vrmh9bixYtJIBDQixcvZM718vIiPT29St8IiYqfF1paWnT48GG+rEOHDuTj48N//vrrr6lTp06VyqnMWAUGBpKqqmq58Zs0aRKNGjWKiP573mzevJk/XlhYSGpqarRr1y4iIvrzzz+J4zh6/PgxXyclJYVUVVVl2m7dunU5o17be69JBFgAwAcffICIiAj+Lzg4WG7d+/fvo3v37jJlzs7OdaKHs7MzHjx4gPfee09uHaFQiL179+Lly5fYsmUL7OzssGPHDrRv3x4XL14EANy4cQMvX76Ejo4O73PT1NTElStX8OjRowrlKnLOP//8g169ekFdXb3GfczIyEBiYiL69OkjU963b1/ExcWhoKCAL+vSpYtMHTMzMyQlJVUq39fXF0lJSdizZw+6deuGo0ePomPHjjhy5AgA4O7duzA2Nka7du34c9TU1PD+++/j7t27MrK6detWoz6WsGbNGpn7KiIiAgMGDOCPa2ho4PDhwzh69CgyMjKwe/du/lhCQgJevnyJL7/8UuZ6DBkyBESE2NhYREdHo6CgAB4eHrXSUxGqM26lr5uZmRkAVHndakKrVq2gr6/PfzYxMYGZmRkMDAxkypKTkwEoNqYVoa+vD09PT/zyyy8Aiv2XR44cwSeffMLXmTdvHiZPnox+/fph2bJliIiIqFL/n376CV5eXrz/acSIEXj9+jXOnz8vU69ly5b8OJbG3t6+3HcxLi4OEydOhI2NDbS1taGjo4Ps7Gw8e/aMrzNt2jTs2bMHUqkURUVF+Pnnn/HZZ59Vqa88bty4gYKCApiamsqM6+HDh8s9b0rfG0pKSmjRogV/b9y7dw/GxsawtLTk6xgaGsLGxkZhXWp67zWZAAs1NbVqDQjHcfWojWKYmppiwoQJmDBhAlavXo3OnTvj+++/R2BgIKRSKRwcHHDs2LFy58lz8Ct6Tm37Tv9uxF9WDlWwQb+ysrLMZ47jIJVKq2xDX18fo0aNwqhRo7By5Uq4u7vj22+/xbhx4ypsu6T9suW1DYYwNjau8r66ePEiOI7D69evkZKSwgf2lPRz27Zt5Qw7UPyg/ueffwBUfk0EAkG5sS0qKqpWP0pQdNxKX7eSY4pct+pSNqCA47gKy0raVmRM5fHJJ59gzJgxSE1NxcWLF5GTk8PfTwCwdOlSTJo0CWfPnkVISAh8fX2xaNEiLF26tEJ5jx8/RlBQEIKDg7Fp0ya+XCKRYOfOnTJLPeTdhxWVDxo0CC1btsT27dthbm4OZWVlODs7ywSxfPLJJ1i4cCHOnj2L3NxcZGdnY+LEiXL7XhVSqRQGBga4cuVKuWNlv8NVfadr+3yp6b3XZN6sqkP79u3LXZSGWKtUGSoqKrC0tOR/QTo5OSEuLg66urqwsbGR+TM1Na1QhiLnODo64tKlS8jNza1QhrKyMh+tJA99fX2YmJjg77//likPCwuDjY0NVFRUajIEchEIBGjbti0/Nvb29khMTMTDhw/5Onl5ebh58ybs7e0rlaVI/6pDVFQU5s2bh127dsHV1RXjx4/nHyotW7aEqakpYmJiyl2PknFycHCAsrIyzp07J7cNIyMjvHr1iv9MROV+9SsrK0MikVSqa23GrbGgyJjKY9CgQdDR0cGvv/6K/fv3Y8iQIdDT05OpY21tjS+++AK//fYblixZgh07dsiV5+/vj44dOyIyMlLmzfvw4cM4ffp0jdYLJSUlISYmBosWLYKHhwc6dOgAJSUlpKamytTT1dXF2LFj4e/vD39/f4wZM4b/kVQTnJyckJqaCrFYXG5MLSwsFJbToUMHJCYm4unTp3xZampquTdeRe7X6tJk3qyqw9y5czFx4kS8//778PDwwKVLl3Dw4EEAlf8qyM7O5ge9qKgICQkJiIiIgJaWFqytrQEAV65cweTJk3Ho0CG5U4Hbt2/HnTt3MGLECFhbW6OwsBAnT57E+fPn8d133wEAJk2ahM2bN8PT0xMrVqyAra0tkpKSEBwcjI4dO2LIkCHl5CpyzowZM+Dv748RI0bAx8cHpqamiI6OhrKyMgYMGABLS0sQEc6cOYPu3btDTU2twnDjhQsXYsGCBbC2tkafPn0QFBSEnTt3YufOndW7GGU4efIkjh07hnHjxqFt27bgOA4XLlzAzz//jLFjxwIAPDw84OjoiI8++gjbtm2DlpYWli1bBrFYjOnTp1cqX9H+lZCZmVnuoaOqqgpdXV3k5uZi/PjxGDVqFLy9vTFkyBB07twZ8+fPx8aNG8FxHHx9ffG///0P2traGDZsGEQiEe7du4fz589jx44d0NbWxldffYXvvvsOKioqcHNzQ05ODs6ePYsFCxYAANzc3ODv74+hQ4eiVatW2L59O168eCHzo8XS0hLh4eGIi4uDjo5OhQ+u2oxbY0GRMZWHkpISPvroI+zYsQNxcXEyMxCZmZn49ttvMXLkSFhaWiI9PR3nzp1Dhw4dKpRVVFSEvXv3YtasWXBwcJA5Zm9vj7lz52LPnj1YtGhRtfpnaGgIfX197Ny5E61bt0Zqaiq++eYbqKmplas7ffp09OnTB1KpFGFhYQrJT0tLK/dDR0dHBx4eHnBxccGwYcOwZs0adO7cGenp6bh8+TI0NTUxZcoUheQPGDAA9vb2mDhxIjZt2gSRSISFCxdCWVlZ5tlqaWmJS5cuIT4+HmpqajJTwTWmSq9WI6Amoetr164lU1NTUlVVpYEDB9L27dsJAL1+/VqunIpCuwHItK1INODNmzdp0qRJZG1tTWpqaqSrq0uOjo7k5+cn45hNSUmhadOmkampKSkpKZGZmRmNHDmSIiIiiKji0PWqziEiun//Pg0dOpS0tbVJTU2NOnfuTH/99Rd/fMaMGdSiRQuFQ9dFIhFZWVnJOF6JFAsMKEtsbCxNmzaN7OzsSENDg7S0tMjBwYFWrlxJeXl5fL2SEGxtbW0+BLt0UEvJdagoYKai/pWlJDihor9hw4YREdGnn35K1tbWMk7p0NBQEgqFdObMGb7st99+ow8++IBUVVVJS0uLunTpIuNclkqltH79erK1tSUlJSUyMjKi8ePH88czMzNpwoQJfPj/999/X24cHz16RD179iQNDQ2FQterO26QEzRREfLqygtdL42Pjw9ZW1vLlC1fvpxat24tU1bVmMrj5s2bBIBatGhBRUVFfHlOTg6NHz+eWrduzS8fGD9+fLmgiBKOHj1KACg2NrbC47NnzyZLS0uSSqUV9pOo/HeqhODgYHJwcCAVFRWys7Oj48ePVxiQQETk4OBAHTp0qLLfRBWHjAMgT09PfgzmzZtHrVu3JiUlJTI2NqaBAwfShQsXiKji5w1R+WCJ2NhYcnV1JRUVFWrVqhXt2LGDunbtSrNnz+brXLt2jbp06UKqqqrlQtdreu9x/1Zu9ixZsgQ//fRTvTiRGQwGo64pKipCq1atsHDhQsyaNettqyOXzMxMmJub44cffsD//d//1Vs7zXIaMD8/H1u2bMHAgQOhrq6OkJAQbNiwoVFuo8NgMBilkUqlSEpKwvbt21FQUKDwFF1DcfLkSaioqMDOzg5JSUlYunQpRCIRRo8eXa/tNktjxXEcgoODsXbtWmRlZcHS0hJLlizB3Llz37ZqDAaDUSmPHz+Gra0tTE1NsW/fPmhpab1tlWTIycnBwoUL8ezZM2hqasLJyQmXL19GixYt6rXdd2YakMFgMBhNl2YZus5gMBiM5gUzVox6JT4+Hq6urtDQ0GgUC7UZilNRWpLqwnEcDhw4UEcaMd5lmLFi1CsrV65EcnIyIiIikJCQUOfyvb294ebmVq1zVqxYgTZt2tS5Lu8aT58+rTC3VX0nYGTX792kWQZYMBoPjx49Qrdu3WBra1srOYWFheW2gWlONOX+nTp1SmaPRkWSXzIY1YW9WTHqjZKozD179oDjOHh7ewMo3qx0/Pjx0NXVhZqaGlxcXHDz5k3+vNDQUHAchz/++AO9evWCqqqqwjtnlLxplewQULILQkpKCoDiTMLfffcdnj17xr8JlOwNJxaLsXTpUlhaWkJVVRX29vb46aefZOQ/efIEHh4eUFVVhYWFBfz8/MpNlykih+M4bNmyBRMmTICOjg68vLwAFL+JWllZQUVFBS1atMCAAQOQl5cnt7+HDh3CBx98AB0dHRgaGsLT0xMxMTH88ZK3n6NHj2LIkCFQV1eHlZUV9u/fLyPn2bNnGDhwINTU1GBhYYGtW7cqNN7Af9tzlfwZGhrKrVvVtScifPbZZ7C2toaamhqsrKywaNEifvPkyq7fqVOn0LVrV6irq0NXVxfdunXD7du3Fe4Ho5Gj0NJoBqMGJCQkyORhev36NUmlUurWrRt17tyZLl68SFFRUTR27FjS1dWllJQUIiK6cOECAaB27drRqVOn6PHjxzLpVUpTdqeHTz75hLS1tWn8+PF0584dunz5MllYWNDHH39MRMUpM+bPn0/m5uZ8rp2srCz+3I4dO9K5c+fo8ePHdPjwYdLR0eHTI0ilUurcuTN169aNrl27Rrdv36YPP/yQtLW1ZXbKqEoOUfGqfX19fdqyZQvFxsbSw4cP6bfffiMtLS36/fff6dmzZ3T79m3auHFjpYkq9+zZQ6dPn6bY2Fi6desWDRkyhGxsbPi8Tk+ePCEAZGlpSUeOHKFHjx7R/PnzSSgUUkxMDN+vrl27kpOTE129epVu375Nbm5upKWlVWmyvxLZle3mglK7Eyhy7SUSCS1evJiuXr1KT548oVOnTpGJiQktWbKk0uuXkJBASkpKtGbNGnr8+DHdu3ePDh48SFFRUXJ1YzQtmLFi1Ctl8zAFBQURALp79y5flp+fTyYmJrRs2TIi+s9YVZbEsYSKjJWhoSHl5+fzZatWrSITExP+c0Xb+zx+/Jg4jqP79+/LlC9btow6d+5MRETnz58nAPTo0SP+eFpaGqmpqfF9VEQOUfFDfMqUKTJ1NmzYQLa2tlRYWFhlv+WRlpZGAOjSpUtE9J9BWb9+PV+nqKiINDQ06McffySi/7ZgKklSSESUnJxMqqqqChkrNTU1mXxgZY1yibFS5NpXxIYNG8jGxob/XNH1u3XrFgGgJ0+eVDI6jKYMm1xmNCh3796FgYGBzAaiKioq+OCDD+osV1X79u1ldudWJMfWzZs3QURwcnKSKReLxXwuo3v37pXL3aOvry+TP0oROSWU7d/YsWOxZcsWtG7dGh4eHnB1dcXw4cMrXRQaERHB52ZKTU3ld5t/9uwZevbsydcrnUNIJBLB2NhYJkeRoaEh2rZty9dp0aKFTL8qY+/evXB0dOQ/GxkZVVhP0Wvv7++PXbt24enTp8jJyYFYLK4yhUSnTp0wYMAAODg4wN3dHS4uLhg5cmSlKUUYTQtmrBgNTn3nqqooHw9Vsfa95GEYHh5eLlleab2qCr9XVA5Qvn9mZmZ48OABLly4gJCQECxfvhzz58/HtWvXKnzo5ubmwsPDA7169cKePXtgYmICoHhX8NK5kYDKcxRVNPbVwczMTOFcc1Vd+4CAAHzxxRdYvXo1+vbtC21tbQQEBGDx4sWVyhUKhfjrr79w48YNBAUF4bfffsOCBQsQEBCAwYMHV79TjEYHC7BgNCj29vZITU3FvXv3+LKCggJcv369wXIuVZRrp+TN4Pnz5+Xy/ZSkh+nQoQNSUlJkcvdkZGTIBDQoIqcyVFRUMHDgQPzwww+4c+cOcnNzcfLkyQrr3r9/HykpKfD19UW/fv3Qvn17ZGRkVDuXl729PVJSUmQyxqampsr0qy5Q5NqHhYWha9eumDNnDhwdHWFrayuTOwmQnyuJ4zh069YNixYtQlhYGPr27Yu9e/fWaR8Ybw9mrBgNSv/+/dGtWzdMmDABly9fRnR0ND7++GPk5+fX647NpbG0tERiYiKuXLmC1NRU5ObmwsbGBlOmTMFnn32G/fv3IzY2FpGRkdizZw/WrFkDoDjvVOfOnfHxxx/jxo0biIyMxKRJkyASifg3A0XkyGP37t3w9/dHZGQknj17hoMHDyIrK0tuzqXWrVtDRUUFW7duRVxcHIKDgzFr1qxqvyW5urqic+fOmDhxIq5fv46IiAh4eXnVeQi6Ite+Xbt2uHPnDk6dOoW4uDhs3rwZx48fl5FT0fULDw/H8uXLce3aNTx//hzBwcGIioqSO3aMpgczVowGheM4nDx5EnZ2dvD09MT777+PxMREBAYGVhryXJcMHz4cY8aMgaenJ1q0aIEffvgBALBz50589dVX8PX1RYcOHeDq6oqff/4ZVlZWvO4nTpyAhoYGevfujcGDB+PDDz9Eu3btoKqqysuvSo489PT0sHfvXri4uKB9+/bYsGEDdu7cCVdX1wrrGxoa4sCBAwgMDIS9vT2+/vprrFu3DgJB9b7WJddER0cHffr0weDBgzFo0CC5yUVriiLXfvr06Zg0aRImT56Mrl274tq1a+XSzld0/XR0dHDlyhUMGzYMtra2mDJlCry8vPhkp4ymD9vIlsGoBVlZWTA3N8eKFSswc+bMt60Og9FsYQEWDEY1+P333yESidC+fXskJydj2bJl4DgOY8eOfduqMRjNGmasGIxqkJubi++//x5Pnz6FhoYGHB0dcenSJRgbG79t1RiMZg2bBmQwGAxGo4cFWDAYDAaj0cOMFYPBYDAaPcxYMRgMBqPR884HWLx69apG5xkaGiI1NbWOtak9TK/qwfSqHkyv6tFc9WrZsmUdaqMY7M2KwWAwGI0eZqwYDAaD0ehhxorBYDAYjZ533mdVFiJCfn4+pFJppRuCJiUl8am2GxPNSS8igkAggKqqaq1SWDAYjKYPM1ZlyM/Ph5KSUpU7TotEonLJ9BoDzU0vsViM/Px8qKmp1YNWDAajqcCMVRmkUmmdp0Zg1ByRSNQo3xQZjKbI308ysT8iBam5D2CoLsKkLi3Q11LnbaulEOypXAY23dT4YNeEwag9fz/JhN+Vlyig4lCFlFwx/K68BIAmYbAa1FhJpVIsWLAA+vr6WLBgAZKTk7Fp0yZkZ2fD0tISM2fOhEgkQlFREbZt24bHjx9DS0sLs2fPhpGREQDgxIkTCAkJgUAgwOTJk9GlSxcAQEREBPbu3QupVApXV1cMHz68IbvGYDAYjZr91+NRQLKP/AISYP/1+CZhrBo0GvDPP/+EmZkZ//nAgQPw9PTEli1boKGhgZCQEABASEgINDQ0sHXrVnh6euLgwYMAgBcvXiA8PBwbNmzA4sWLsXv3bkilUkilUuzevRuLFi3Cxo0bcfnyZbx48aIhu1anDB06tMo6/v7+yMvLqzcdzp49W2Va8yNHjiAxMZH//PXXX9d5KnQGg1E3pBZV7DOWV97YaDBjlZaWhlu3bvFZT4kId+/eRffu3QEALi4uuHHjBgDg5s2bcHFxAQB0794d0dHRICLcuHEDPXr0gJKSEoyMjGBiYoLY2FjExsbCxMQExsbGEIlE6NGjBy+rvpFeDYVk/lRIPhsGyfypkF4NrbXM33//vco6u3btqjdjJRaLFTJWAQEBSEpK4j+vW7cObdu2rRedGAxG7VAX51dYbljwuoE1qRkNZqz27duHiRMn8v6HrKwsqKur8xFi+vr6SE9PBwCkp6fDwMAAACAUCqGuro6srCyZ8tLnlC03MDDgZdUn0quhoP1+QHoKAALSU0D7/WptsGxtbQEA4eHhGD16ND777DP06dMHM2bMABFh9+7dSEpKwpgxYzB69OgKZUgkEsyePRv9+/eHq6srdu7cCQCIjo7G4MGD4ebmhqlTp/whw00AACAASURBVOL16+IbdfTo0Vi1ahVGjRoFPz8/BAYGYsWKFXB3d8fTp0/LyT9z5gwiIyMxY8YMuLu7Iy8vD6NHj0ZkZCTfB19fXwwcOBDjxo3DrVu3MHr0aDg7O+P8+fO8jsuXL8egQYPg5uaG/fv312rcGAxGeYgIv0alIEdJDSiTEUpFUgiv5EtvSbPq0SA+q3/++Qc6OjqwsrLC3bt3q6xfUYotjuMqLK+sfkUEBQUhKCgIALB69WoYGhrKHE9KSuKjAQsmD5Kro0TegcIC0O4NkOzeIPdclb1/yj1WQkmod3R0NMLCwmBiYoLBgwfj1q1bmD59Ovz9/XH8+HEZI1363Lt37yIpKQlhYWEAgMzMTIhEIsyePRsrV65Ejx49sGbNGmzatAkrVqwAx3HIzs7GqVOnAADPnj2Du7s7hgwZUqF+w4cPx88//wwfHx/eb8hxHIRCIUQiEXJzc9GrVy/4+PjA29sbq1evRkBAAGJiYjBz5kwMGjQIhw4dgo6ODs6fP4+CggIMGTIE/fv3R+vWrWXHS0Wl3HWqK0QiUb3Jrg1Mr+rB9KoYIsKOy09x+E4aBCSF+6uruGXQHqkqujAseA2v50EYPHYQ1Brh2JWlQYzVw4cPcfPmTdy+fRuFhYXIy8vDvn37kJubC4lEAqFQiPT0dOjr6wMofjNKS0uDgYEBJBIJcnNzoampyZeXUPqc0uVpaWnQ09OrUBc3Nze4ubnxn8tu5lhQUFDv65TEYrFCdSQSCbp06QIjIyNIpVJ06NABT58+haOjI4gIEomknCyRSASxWAwzMzM8e/YMCxYsgKurK/r27Yv09HRkZmaiW7duEIvFGDVqFKZPnw6xWAwiwuDBg3l5Uqm0QvmlKatD6c/Kysro06cPxGIx2rVrxy/stbW1RXx8PMRiMS5cuID79+/j9OnTAIrftmNjY2X8mkDxNamvzUCb60aj9QXTq3q8Tb2ICLv+ScaZhxkQQorZ9w6hZ3YckHoReJ0O6BmCGzEJOfaOyKmmjm9jI9sGMVYTJkzAhAkTAAB3797F6dOn8eWXX2LDhg24evUqevbsidDQUDg5OQEAHB0dERoairZt2+Lq1auwt7cHx3FwcnLCli1bMHjwYGRkZCAhIQE2NjYgIiQkJCA5ORn6+voIDw/Hl19+WWu9hf7yfUcikQgFcz/5dwqwDPotIFyzu9btA4CysvJ/+giFChk6ANDV1UVgYCBCQ0Oxb98+nD59GkuXLq30HHV19dqoKoNIJOLfbgUCAVRUVPj/S/dhxYoVvH+SwWDUDVIi/Hg9CediX0PEEb6+sx/d0u5DMHc5uHYdG61xr4y3ujegl5cXzpw5g5kzZyI7Oxv9+/cHAPTv3x/Z2dmYOXMmzpw5Ay8vLwBAq1at4OzsjDlz5sDX1xdTp06FQCCAUCjElClT4Ovri6+++grOzs5o1apVvevPjZgEKKvIFiqrFJfXM5qamsjOzpZ7PD09HVKpFJ6enpg3bx7u3LkDbW1t6Ojo4Nq1awCA3377jQ9wqUh+Tk5OpTpoaGhUqkNV9O3bF7/88guKiooAAHFxccjNza2xPAaDAUikhK1XE3Au9jWUBcCCh7+iW+pdcEM/Ateu49tWr8Y0+KJge3t72NvbAwCMjY2xatWqcnWUlZUxZ86cCs8fOXIkRo4cWa78vffew3vvvVe3ylaBoLsLpADoxH4gPRXQL36tFnR3qfe2vby8MHHiRBgZGeHYsWPljickJGDOnDmQSqUAgIULFwIANm3ahAULFiA/Px8WFhbYsKFi39qwYcMwb9487N69Gzt37kSbNm3K1Rk7diwWLFgAVVVVhSIYyzJhwgTEx8dj4MCBICLo6+tjz5491ZbDYDCKEUsJG8Nf4dKzLKgIOSxKOouOryKADl3ADao4GKupwJG8qIV3hLLJF3NzcxWaDivxDTU2mqNeil6TmtBYp0OYXtWD6QUUSaRYe+kVrr3IhppIgG+F99D+3B5ARx+CJZvAaevWmV4s+SKDwWAwqk2BWIpVYS9x7UU2NJQFWGaZi/bn9wKcAILP5soYqqaKQtOAqampePbsGXJycqChoYHWrVs3yjDRd5XBgwfzm72WhPhv2bIF7du3rxP5ixYtKrfI+tNPP8W4cePqRD6Dwag5+WIpfENfICopF1oqQix7XwttNi8HiMANa9p+qtLINVZisRhBQUEIDAxEcnIyTExMoKqqivz8fCQmJsLIyAju7u5wc3Nju5S/Zc6cOcP/Xx/TgCtXrqxTeQwGo27ILZJg+YUXuJeSB11VIb7vZwbz3SuArEygfWdwg8a8bRXrDLlWZt68eXBwcMC0adNga2sLgeC/GUOpVIrY2FhcvHgR33zzjVwnPYPBYDDqh+wCCZZeiMejtHwYqImw3M0CpiFHQTHRgI4eBJ/OASdoGvv+KYJcY7V06VLo6FS8E69AIEDbtm3Rtm1bvHnzpt6UYzAYDEZ53uSLsSQkHk8yCmCkoYTlrq1g/PwupH8GFPupPp0LTrvijRGaKnIDLOQZKgAoLCzkp5q0tbXrXisGg8FgVEhGnhiLg57jSUYBTLWUsNLdAsaSLEh3byj2Uw0dD86u09tWs85RKBrwl19+QWxsLADg1q1bmDx5Mry9vXHz5s16VY7BYDAY/5GaW4RFgc/xPLMQ5trKWOneGoaqAkj91zdLP1VpFDJWly5d4neEOHbsGGbOnIlvvvkGv/76a70q966ye/du9O3bF46Ojli8eHG9tBEfH48TJ05UWic6OhrBwcH85/Pnz2Pbtm31og+DwaicpOxCLAp8jldZhWijqwJfdwvoq4lAZw4DzdRPVRqFwvgKCgqgoqKCrKwsJCUl8Vv0NMZFeA3N308ysT8iBam5YhiqizCpS4taZ938+eefceDAAVy5cgVRUVF1pOl/iMVi3liNGDFCbr27d+8iKiqKz0Hm4eEBDw+POteHwWBUTkJWIb4Neo7UXDFs9FWxtH8raKkIQfdug/442mz9VKVRyFi1bNkSFy9eRGJiIjp1Kp4LffPmjcwmq+8ifz/JhN+1RBRIijcBSckVw+9acebcmhqs+fPn4/nz55g8eXKV65hOnz6NjRs3QiAQQFtbG8ePH0d+fj7mzZuHqKgoCIVC+Pj4oGfPnjhy5AiCg4NRUFCA3Nxc5OXlITY2Fu7u7hgzZgymTZsmI7uwsBDr1q1Dfn4+rl+/jhkzZiA/Px9RUVHw9fXF7NmzoaqqitjYWLx8+RIbNmxAQEAA/vnnH3Tt2hWbNm0qHqO//8a6detQVFQECwsLbNy4ERoaGjUaGwbjXSQ+swDfBccjI08MO0M1LOlnDg1lIeh1GqS7SvxUHzVLP1VpFDJWU6dOxb59+yASifC///0PABAZGckbrubKsIMPqn1OgYSwITwBG8IT5NY55WUn99iaNWsQGhqKgIAABAYGVtrWpk2bcPDgQZiamiIzMxMAsHfvXgBAcHAwYmNj8dFHH+HixYsAivOKBQUFQU9PD+Hh4fjxxx/xyy+/VChbWVkZX3/9NW+cgOI09qXJzMxEQEAAzp8/D29vb5w8eRLr1q3DoEGDEB0djZYtW2Lz5s04cuQItLW1sXnzZuzcuRNfffVVpf1iMBjFPM3Ix5LgeGQWSOBgrI5v+5pDTUkAkkhk/VSezdNPVRqFjJWNjQ1WrFghU9a7d2/07t27XpRiKIaTkxO++uorDBkyBB9++CEA4Nq1a/D29gZQfN3Mzc3x+PFjAECfPn3k5vmqCe7u7uA4DnZ2djA0NOR3zGjbti1evHiBhIQExMTEYNiwYeA4DoWFhXB0dKyz9hmM5kxsWj6WhjxHVqEUXUw1sKiPGVRExWEG74qfqjRyjVV0dLRCAhwcHOpMmcZGZW9AIpEI3gEPkJJbfreIFuoi7BphU5+qASh+C7t16xaCg4Ph4eGB8+fPy82mDNRtvirgv1xbpfNVlXwWi8UQCATo06cPtm/f3mg32GUwGiMPUvKw7EI8coukeN9ME/N7t4SS8F9DdS/iXz8VB8HUOc3aT1UaucZqx44dMp/T09PBcRy0tLSQlZUFIoKBgcE7HR02qUsLGZ8VAKgIOUzq0qJB2n/69CmfGiUwMBCvXr2Cs7MzTpw4gV69eiEuLg4vX76EtbU17ty5I3OuIvmqqsqZVRUl0YxPnjyBra0t8vLy8OrVK1hbW9dYJoPR3IlOysXy0Hjkiwk9LLQwp0dLKAmLE5nS63RId60v9lMN+Qhc+85vWduGQ66x8vPz4/8/fvw4srOzMW7cOKioqKCgoABHjhyBlpZWgyjZWCkJoqjraEBFWbFiBZ48eQIiQq9evWBvbw87Ozt8/fXXcHV1hVAoxMaNG2Xeekpo3749hEIh3NzcMHbs2HIBFgDQo0cP+Pn5wd3dHTNmzKi2fgYGBti4cSO++OILFBYWAgC++eYbZqwYDDncTsjByr9foFBC6NtGG7OcTSEU/GuopJJiQ5WVCdh1Ajd47FvWtmFRKJ/V1KlT8dNPP8lsWCsWizF9+nTs3l11+vbCwkL4+PhALBZDIpGge/fuGDt2LPz8/HDv3j1+euqLL75AmzZtQETYu3cvbt++DRUVFXz++eewsrICAISGhuL48eMAihMxlqREf/z4Mfz8/FBYWIiuXbti8uTJfFr1ymD5rBoGls+qejC9qkdz0OvGi2ysvvgSYinBzVoHn3cz4Q0VAEhPHSr2VWnrQrBkMzidmk//NcV8VgoFWJSEKNvZ/efDiYuLq/AXe0UoKSnBx8cHqqqqEIvFWLJkCbp06QIAmDRpUrnU6rdv30ZiYiK2bNmCR48eYdeuXVi5ciWys7Nx7NgxrF69GgCwYMECODk5QVNTE/7+/pg+fTpsbW2xatUqREREoGvXrgrpx2AwGG+T8OdvsO7SK0gI+NBWF9PeN4ag1I/tYj/VkWI/1adza2WomioKGatx48Zh5cqVcHR0hIGBAdLS0nDr1i1MnTpVoUY4joOqqioAQCKRQCKRVPrWc/PmTfTp0wccx6Ft27bIyclBRkYG7t69i06dOkFTUxMA0KlTJ0RERMDe3h55eXlo27YtgOKotxs3bjQbY7V582aZNCBAcQ6rWbNm1Yn80NBQPjy9BAsLC4XemhkMRu34+0kmNl1JgJSA4e314d21hczzUdZPNf6d8lOVRiFj1adPH1hZWeHq1avIyMiAmZkZRo0aBXNzc4UbkkqlmD9/PhITEzFgwADY2tri/Pnz+PXXX3Hs2DE4ODjAy8sLSkpKSE9Pl0nuaGBggPT0dKSnp8PAwIAv19fXr7C8pH5zYdasWXVmmCrCxcWFn05lMBgNR1Dca2y7mggCMNbBABM6GcoaqtJ+qnYdwQ1+dxOeKpw10dzcHKNHj65xQwKBAGvXrkVOTg7WrVuH58+fY8KECdDV1YVYLMZPP/2EU6dOYfTo0RWGX8t7EyvJjKsoQUFBCAoKAgCsXr26XMbjpKQkhZNJNtakk81NLxUVlXrLTC0SiRpl1mumV/VoinqdiErA1qvFO95Mc26NT7q1Klcn+/Au5Dy8A4GuPvTnr4RQz6BcnbrWq7Gi0NMjOzsbv//+O549e4b8/HyZY8uWLatWgxoaGujQoQMiIiIwdOhQAMU+rX79+uH06dMAit+MSjv/0tLSoKenB319fdy7d48vT09PR4cOHfipydL19fX1K2zfzc0Nbm5u/OeyTsbCwkIQUZUP1uYYyFCf1FQvsViMoqKienOeNwfHfEPC9Koe8vQ6dT8de24lAwCmvGcETyu1cvXofiSkR/cCHAdM+QoZEgLqqI/NNsBi8+bNEIvFcHZ2rtF+gG/evIFQKISGhgYKCwtx584dDBs2DBkZGdDT0wMR4caNG/zO7k5OTjh79ix69uyJR48eQV1dHXp6eujSpQt+/fVXfu1PZGQkJkyYAE1NTaipqSEmJga2trYICwvDwIEDq60nUBxMkp+fj4KCgkr9aiUh/I2N5qQXEUEgEPD+TgajORAQnYoDkcWGYvr7xhjUtnywBGVmMD9VGRQyVjExMdi1axeUlJRq1EhGRgb8/PwglUpBRHB2doajoyOWLVvGZxpu3bo1v9ana9euuHXrFr788ksoKyvj888/B1C8SHXUqFFYuHAhAGD06NF8sMWnn36K7du3o7CwEF26dKlxcAXHcVBTU6uyXlP7Jfe2aax6MRgNBRHhUFQqjkangQMwo7sJ3Kx1y9eTSiD1Xwe8ef3O+6lKo9A6q++++w5ffPEFTExMGkKnBqXsOitFaawPX6ZX9WB6VQ+mV/Uo0YuIsO92Ck7eT4eAA2Y5m8JFzuYB0t8PgU4fBrR0IPDZUi9h6s12GtDBwQErV66Ei4sLdHVlfwn079+/XhRjMBiM5oCUCLtuJuGPmNcQcsDXvVqih4V2hXXpfiTozLu9nkoeChmrBw8ewMDAoNz+cgAzVgwGgyEPiZSw/VoiAuMyIRJwmN+7JbqZV7xNnYyfavB4cB26NLC2jRuFjJWPj09968FgMBjNComU4BsYg8C4TCgLOSzqa46uphUnHuXXU5X4qYYwP1VZFF74kp2djX/++Qfp6enQ19eHo6MjH9zAYDAYjP8QSwkbLr/C5edZUBVx+M6lFRyM5e9vSWeOAg+iiv1Un859J/JTVReBIpViYmIwc+ZMBAYG4tmzZwgKCsLMmTMRExNT3/oxGAxGk6JIIsWaiy9x+XkWNJSFWNq/CkN1P7J4g9oSP5VuxWtE33UUerPat28fPv30U/Ts2ZMvCw8Px969e7Fq1ap6U47BYDCaEgViKVaFvcTthBxoKguwaaQDWgjlry+U9VONY36qSlDozSohIQHOzs4yZd27d0diYmK9KMVgMBhNjbwiKZaHvsDthBzoqAixws0C7Y3l5/wjqQTS3RtK+anGN6C2TQ+FjJWJiQnCw8Nlyq5cuQJjY+N6UYrBYDCaEjmFEiy7EI87SbnQUxNhhbsFLPUq33mF/ggA7kcyP5WCKDQN6O3tjdWrV+Ovv/6CoaEhUlJSkJCQgAULFtS3fgwGg9GoySooNlSP0vJhqC7CclcLtNSufFs6uh8JOv3rv36qOcxPpQAKGat27dph69atuHXrFjIyMuDo6Ij33nuPRQMyGIx3msx8MXxC4vEkowDGmkpY7toKxppVGKpyfqrmkXevvlHIWKWnp0NZWRl9+vThy7Kzs/kwdgaDwXjXSM8TY0nwc8RnFqKlljKWu7WCoXrl+6cyP1XNUchntXbt2nLJDNPT07Fu3bp6UYrBYDAaMyk5RVgc+AzxmYWw0FHGSneLKg0VANCfpf1Uc5ifqhoo9Gb16tUrWFhYyJRZWFjg5cuX9aIUg8FgNFaSsgvxbVA8knOKYKmngu/7t4K2atWPUnoQBfr93/VUU+eA062bRIrvCgq9WWlra5cLU09MTISWlvywTAaDwWhuvHxTiIWBz5GcUwRbA1WscLVQzFC9KfFTScENGgPOnvmpqotCb1b9+vXD+vXrMX78eBgbGyMxMRFHjhxhm9gyGIx3hueZBVgS9BwZ+RJ0aKGG7/qZQ12p6mm84n3/NgCZGUBbB3BDPmoAbZsfChmr4cOHQyQSYf/+/UhLS4OhoSH69euHwYMH17d+DAaD8dZ5nJ4Pn5B4vCmQoJOxOha7mENVpNDElKyf6rO54ITMT1UTFDJWAoEAQ4cOxdChQ2vUSGFhIXx8fCAWiyGRSNC9e3eMHTsWycnJ2LRpE7Kzs2FpaYmZM2dCJBKhqKgI27Ztw+PHj6GlpYXZs2fDyMgIAHDixAmEhIRAIBBg8uTJ6NKleHuSiIgI7N27F1KpFK6urhg+fHiNdGUwGIzSPErLw9KQeGQXSuHYUgPze5tBRUFDVRh9i/mp6gjFRhxAVFQUduzYgdWrVwMA4uLiEB0drdC5SkpK8PHxwdq1a/HDDz8gIiICMTExOHDgADw9PbFlyxZoaGggJCQEABASEgINDQ1s3boVnp6eOHjwIADgxYsXCA8Px4YNG7B48WLs3r0bUqkUUqkUu3fvxqJFi7Bx40ZcvnwZL168qO5YMBgMhgz3k3OxJLjYUH1gromFfRQ3VPQmA5kbfIr9VB8yP1VtUWjU//rrL/j7+8PU1BT3798HACgrK+Pw4cMKNcJxHFRVi7cekUgkkEgk4DgOd+/eRffu3QEALi4uuHHjBgDg5s2bcHFxAVC8B2F0dDSICDdu3ECPHj2gpKQEIyMjmJiYIDY2FrGxsTAxMYGxsTFEIhF69OjBy2IwGIyacCcpB0svxCO3SIperbXwTW8zKAkVNFT/+qmkGWlAW3twQ5mfqrYoNA34559/4rvvvoORkRFOnToFADAzM8OrV68UbkgqlWL+/PlITEzEgAEDYGxsDHV1dQj/nb/V19fn13Klp6fDwKD4dVkoFEJdXR1ZWVlIT0+Hra0tL7P0OSX1S/5/9OiRwroxGAxGaW69ysaqsJcolBD6WWpjZndTCAWcwufTn8eA+5HgtHXBffY181PVAQoZq7y8PBgaGsqUicViiEQK526EQCDA2rVrkZOTg3Xr1lW6RouIypVxHFdheWX1KyIoKAhBQUEAgNWrV5frl6KIRKIan1ufML2qB9OrerwLel18nIaVYS9RJCEMdTDGvP42EMh5nlREYfQtZPy775/+3O8hsmlXJ3rVJY31OlaGQtamffv2OHnyJEaOHMmX/fXXX7C3t692gxoaGujQoQMePXqE3NxcSCQSCIVCma2bDAwMkJaWBgMDA0gkEuTm5kJTU5MvL6H0OaXL09LSoKenV2H7bm5ucHNz4z+npqZWuw8AYGhoWONz6xOmV/VgelWP5q7X5WdvsP7yK0gI8GynhymddJFe6tlSFfQmA9J13wFSKbhBYyHq5NQsx6tly5Z1qI1iKDQBO2XKFFy/fh1ffPEF8vPzMWvWLFy9ehWffPKJQo28efMGOTk5AIojA+/cuQMzMzPY29vj6tWrAIDQ0FA4OTkBABwdHREaGgoAuHr1Kuzt7cFxHJycnBAeHo6ioiIkJycjISEBNjY2sLa2RkJCApKTkyEWixEeHs7LYjAYDEUIfZKJdf8aqpEd9PGZo5HcGZqKIKkU0t0bi9dT2XZgfqo6RqE3Kz09PaxatQpxcXFISUmBgYEBbGxsIBAo5mzMyMiAn58fpFIpiAjOzs5wdHSEubk5Nm3ahMOHD8PS0pJfZNy/f39s27YNM2fOhKamJmbPng0AaNWqFZydnTFnzhwIBAJMnTqV12HKlCnw9fWFVCpFv3790KpVq5qMB4PBeAc5H/sa268lggCM72iA8R0Nq2WoAID+OgbciwA0tSH4bB7zU9UxHMlzBFVCdHQ0BAIBOnToUB86NSjVCRIpTXOfDqlrmF7Vg+lVPWqj1x8PM7DzZhIAYFKXFhhtX/21UPQwGtL13wIkhWDWUnAO79Var/qk2U4D+vj44MGDBwCAkydPYvPmzdi8eTOOHz9er8oxGAxGfXLiXhpvqKY6GtXMUL15Dan/uv/2/fvXUDHqFoWMVXx8PNq2bQsACA4Oho+PD3x9fREYGFivyjEYDEZ9cfROKvbdTgEA/O99Ywy1q35uvmI/1QYgM/1fP9WEulaT8S8K+axKZgpLdl43NzcHAD5ogsFgMJoKRISDkakIuJsGAQfM+MAErta6NZPF/FQNhsJp7ffs2YOMjAy8//77AFiKEAaD0fQgIuy9lYxTDzIg4ICverREnzbaNZMVEw06dQgAIJj6FTg9tu9ffaLQNOAXX3wBdXV1tG7dGmPHjgVQHJgwaNCgelWOwWAw6gopEX66kYRTDzIgEgDf9DaruaEq7af6cDQ4B8c61pZRFoXerLS0tDBhguxc7HvvMScig8FoGkikhO3XExEUlwklAYcFfczgZKZZI1n8eqrX6YBNB3DDvOpYW0ZFyDVWx48f53esOHLkiFwB48aNq3utGAwGo46QSAmbriQg7OkbKAs5LO5rji6mGjWWV+ynuv2vn4rt+9dQyDVWZbcvYjAYjKZGkYSw/vIrXInPgqpIgCX9zGFvpF5jeeX8VPpNa3+9poxcY/XZZ5/x/3/++ecNogyDwWDUFYUSKX64+BI3XuZAQ0kAn/6t0M5QrcbyKCuT+aneInKNVVJSkkICjI2N60wZBoPBqAsKxFKs/PsFIhJzoaUixLL+rWCtr1pjefx6KuanemvINVZffvmlQgIq82cxGAxGQ5NbJIFv6AtEJ+dBR1WI7/u3Qhu9mhsqAKCzvwF3bwOaWsxP9ZaQa6yYEWIwGE2NnEIJll14gYepedBXE2G5ayuY66jUSibF3AWdPAgAEEyZw/xUbwnFsycyGAxGI+ZNgQRLQ+IRl56PFuoiLHezgKmWcq1kFvup1v7rpxoFriPzU70t5BqrJUuWKLRF/rJly+pUIQaDwVCUv59kYn9EClJzH0DAARICTDSVsNzVAkaaSrWSLeunag9u2MQ60ppRE+Qaq5LcUgwGg9EY+ftJJvyuJaJAUrx3qYQADsAQO71aGyqgrJ+K7fv3tpFrrFxcXBpQDQaDwage+yNSeENVAgE4eS8dg9tVfwd1GTkxd0GnSvxUbD1VY0CusQoLC0OfPn0AACEhIXIFKPIGlpqaCj8/P7x+/Rocx8HNzQ2DBg3C0aNHERwcDG3t4v25PvroI34bpxMnTiAkJAQCgQCTJ09Gly5dAAARERHYu3cvpFIpXF1dMXz4cABAcnIyNm3ahOzsbFhaWmLmzJkQiZhLjsFojhARUnLFFR5LlVOusOyS9VRSKbiBo8B1dKqVPEbdIPdpfvnyZd5YXbx4Ua4ARYyVUCjEpEmTYGVlhby8PCxYsACdOnUCAHh6emLo0KEy9V+8eIHw8HBs2LABGRkZWL58OTZv3gwA2L17N7799lsYGBhg4cKFcHJygrm5OQ4cOABPT0/07NkTO3fuREhICDw8PKoeAQaD0aTIPQ1zsAAAIABJREFULZJg69VEuccN1Wv+I5WkUkj3bARep/3rp2LrqRoLcq/qwoUL+f99fHxq1Yienh709PQAAGpqajAzM0N6errc+jdu3ECPHj2gpKQEIyMjmJiYIDY2FgBgYmLCL0Tu0aMHbty4ATMzM9y9exezZs0CUDyFGRAQwIwVg9HMeJqRjzUXX+JVVhGUBACBg1j631SgipDDpC4taiyfzh0Hom/9t56Kzc40Gqp1JXJzc5Gfny9Tpq9fvbnh5ORkPHnyBDY2Nnjw4AHOnTuHsLAwWFlZ4eOPP4ampibS09Nha2sr00aJcTMw+C9njIGBAR49eoSsrCyoq6tD+K8DtHR9BoPRPAh5nIkd1xNRKCG00VXB/N5meJSW9280oBiG6iJM6tICfS11aiSfHt0DnTwAoMRPVXOjx6h7FDJWUVFR2LlzJ1JSUsodq87i4fz8fKxfvx7e3t5QV1eHh4cHRo8ezcv55Zdf8Pnnn/OZictSUbki4fWlCQoKQlBQEABg9erVMDSsmeNUJBLV+Nz6hOlVPZhe1eNt6FUglmBj6GOcvlu8BdzgDsaY088KKiIhOlkBo963hkgkglhcc1+V9M1rpO1eD0ilUB/hBa1+A+tEd3Yd6w6FjNWPP/6IUaNGoWfPnlBWrtkiO7FYjPXr16N379744IMPAAC6uv+lknZ1dcWaNWsAFL8xld7pPT09nX+DK7sbvJ6eHrS0tJCbmwuJRAKhUChTvyxubm5wc3PjP6emptaoP4aGhjU+tz5helUPplf1aGi9ErIKsebiSzzJKICykMP0943hZq2LrNcZyKojvUgqhXTrciAtBbC2Q77HKBTUUR+b63Vs2bJlHWqjGAplCi4qKkK/fv2gqqoKgUAg86cIRIQff/wRZmZmGDx4MF+ekZHB/3/9+nW0atUKAODk5ITw8HAUFRUhOTkZCQkJsLGxgbW1NRISEpCcnAyxWIzw8HA4OTmB4zjY29vj6tWrAIDQ0FA4ObEIHgajKXMlPgtz/nqKJxkFMNVSwg8DWsPNWrfqE6sJnTsBRP8DaGhBMG0e81M1UhS6Kp6enjh16hSGDx9e7Wk3AHj48CHCwsJgYWGBefPmASgOU798+TKePn0KjuPQokULTJs2DQDQqlUrODs7Y86cORAIBJg6dSpvGKdMmQJfX19IpVL069ePN3BeXl7YtGkTDh8+DEtLS7aomcFoooilhP0RKTh5v9jv7NxKCzO7m0BDue4X5Rb7qfYDAARTZjM/VSOGI3kOolIkJCTA19cXWVlZ0NLSkjm2bdu2elOuIXj16lWNzmuur/f1BdOreryreqXlFmHtpVe4n5IHIQd4v2eEIe30qvyRXBO9KOsNpN/PAl6ngRswEoLR3rXQvO70agia4jSgQm9WGzZsgJ2dHZydnWvss2IwGIzKiEjIwYb/b+/e46Kus8ePvz7DHQdwGFBSu3gj84KgYF6RFM2yi5palla6bbVkfn/1rS1391vuo2x1y8Us3SzL0m27ud6qTUtRNK94wRteMM1LigPMiNxhZt6/PwZGUcABGUA8z7+Y91w+Z2Y+cHhfPue96Qw5xTaMfp68PKAVd4TWflff6lS4nqp9J7QRUvevsXMpWZlMJmbOnOnyHJUQQrjKrhTf7M/mi71ZKCAyzJ8X+7UiyNd9c0fqR5mnut649A1FR0ezf/9+Z9UJIYSoCxeKrCRuPsuus/lowLhuIYzpasRDV/O5cVepo2moZWXzVBNlnup64VKyKi0t5e9//zt33HEHQUEVL7ibPHmyWwITQjRth7MK+fvG38gqsBLg48H/9mtF1E3N3HpMlXsB+4dldf/uHonWPcatxxN1x6VkdfPNNztX3QkhxLVQSvH9EQsLd5mw2uH2ED9e7t+K0GbXvq1Htce127EvnA2WrLJ5qgluPZ6oWy4lqzFjxrg7DiHEDaCg1Mb7WzPYdNJxSe/9nQw8EdkCLw/3DfuVUz8ug307HPNUv5d5qutNlSsmfv31V5dewNXHCSFubL9aivjfH06w6WQufp46/jigFU/1bFk/ieryeSqjzFNdb6r81+Ljjz/G39+fAQMG0Llz5wrliywWC2lpaWzYsIGioiLZ2l4IUa3Li9D+cUBrWgfWz2UwKu+SeaqhMk91vaoyWb3xxhvs3LmTn376iQ8++ACdToefnx+FhYUopejWrRt33323c7NEIYS4XInNzkc7zvHj0RwABrUL4tmYlvh41s9lMI7rqcrmqdrdjjZS5qmuV9UO2vbs2ZOePXtitVrJyMggPz+fZs2acdNNNzm34xBCiMqczS3h7xt/49hlRWjrk/ppuWOeyl+P7uk/yjzVdcylb87T05M2bdq4OxYhRBOx9VQuc7acJb/UTpjei1cGtKZdsG+9xqCOHkQtXQSU1f2TearrmvybIYSoM5cXoe19s54pvW9ySxHa6jjmqd6+ZJ6qV70eX9Q9SVZCiDqRXVDKOz+fIa2sCO0TUS14oNPVi9DWNZmnapokWQkhrtmejHxmbTpDTpGNYD9P/ti/FXe0cE8R2qtRP62QeaomqEbfYlZWFmazmfDwcHfFI4S4jtiVYskBRxFau4LuZUVom7uxCG11HPNUnwEyT9XUuHRGZWVl8e677zovAF68eDFbt24lNTWVZ5991p3xCSEaqQvFNmZvPsPOM44itA93M/Jw1xC3FqGtjsq7gP2j8nmqETJP1cS4dLHDhx9+SFRUFJ999hmeZV3qiIgI9u7d69bghBCN0+GsQl7873F2nsknwMeD1+5qw6MRoQ2XqMrnqczl81SPN0gcwn1c6lkdPXqUV199tcJ+Vv7+/hQUFLh0kKysLObOncv58+fRNI34+Hjuvfde8vLySExMJDMzk9DQUF544QX0ej1KKRYuXMju3bvx8fEhISGBdu3aAbB+/XqWLl0KwKhRo4iLiwPg2LFjzJ07l5KSEqKiopg4cWK9T+wK0dQppfj+sIVPdp0rK0Lry8v9W7u9CG1V7FvXo5YtxmTOdDR4+cj+VE2USz2roKAgMjIyKrSdPn2akJAQlw7i4eHBhAkTSExMZPr06axevZrTp0+zfPlyunXrxpw5c+jWrRvLly8HYPfu3WRkZDBnzhyefvppFixYAEBeXh5Llizhrbfe4q233mLJkiXk5eUB8NFHH/HMM88wZ84cMjIySE1NdflDEEJcXUGpjddXHebDHY5Edf/tBqbH39qwiWrxXChPVADKhkpPa5B4hHu5lKzuv/9+Zs6cybp167Db7fz8888kJiby4IMPunQQg8Hg7Bn5+fnRunVrzGYzKSkpDBw4EICBAweSkpICwI4dO4iNjUXTNMLDw8nPz8disZCamkpERAR6vR69Xk9ERASpqalYLBYKCwsJDw9H0zRiY2OdryWEuHYnzhfz0qoTrD2Sha+njj/2b8VT0fVThLYq6j+fQUlxxUar1VmwVjQtLvWVBw0ahF6vZ+3atRiNRpKTk3n44Yfp1avmE5gmk4njx4/ToUMHcnJyMBgMgCOhXbhwAQCz2Vyh12Y0GjGbzZjNZoxGo7M9ODi40vbyxwshrt26YznMKytC287oz//2bUmbQJ8Gi0cV5KNWL4Pz2ZU/wJxVvwGJeuHywG6vXr1qlZwuVVRUxKxZs3jyySfx96/6Ggyl1BVtVc0/aZpW6eOrsmbNGtasWQPAjBkzXB7KvJynp2etn+tOElfNSFxVK7baeTf5GCv2O6YAht3RgqlDbsdTc/33rS6pkmIKfvgP+UsWofIuVPk4XWiLBv/syjWG77EyjTWu6ricrA4ePMjx48cpKiqq0D5q1CiXnm+1Wpk1axYDBgzgzjvvBBxzYRaLBYPBgMViITAwEHD0jLKyLv53lJ2djcFgIDg4mLS0i+PRZrOZzp07YzQayc7OrvD4S7c0uVR8fDzx8fHO25cepyZCQkJq/Vx3krhqRuKqXEZuCTPLitB66TSejmnJkPZBeGqq3uNSdhtqyzrUyn9f7DWFd4Xbu8LqZRWHAr19UA881mi+04b+HqtyrXG1atWqDqNxjUvJ6pNPPmHLli106tQJb++Le9C4utpOKcUHH3xA69atue+++5zt0dHRJCcnM2LECJKTk4mJiXG2r1q1in79+pGeno6/vz8Gg4HIyEi++OIL56KKPXv28Oijj6LX6/Hz8+PIkSN07NiRDRs2MGzYMJc/BCHERdtO5/Lu5oYtQgtlIyx7tmFfuhjOnnI0tmmLbtTj0LUHmqZhb9HKMUdlyQJDCNrICeh6x9V7rML9XEpWGzduZNasWVX2Vq7m8OHDbNiwgVtuuYWXX34ZgHHjxjFixAgSExNJSkoiJCSEF198EYCoqCh27drFlClT8Pb2JiEhAQC9Xs9DDz3E1KlTARg9ejR6vR6Ap556innz5lFSUkJkZCRRUVG1ilWIG5XNrvjXnkyWpjnme+9so2dKn5vQ13MRWgB1ZD/2pYvgl0OOhpCWaA8+htYrFu2SS2h0veOgd1yj7cGIuuNSsgoJCcHLq/bLUzt16sTXX39d6X2vvfbaFW2apvHUU09V+vhBgwYxaNCgK9rbt2/PrFmzah2jEDcyc6GVtzf+RlpmIToNnogK5cFOwfVfhPb0cUdPat8OR0NAENrwh9EG3o3m2TBL5EXj4FKyevbZZ5k/fz79+vUjKCiown2dO3d2S2BCiPqxt6wI7fkiGwY/T17u34ou9VyEVmVmoFb+G7UtGZQCHz9HyaShD6L5NkxBXNG4uJSsjh07xu7duzl48GCFOSuAf/7zn24JTAjhXnal+M+BbP5dVoQ2oqU//9uvFc396q/6g7pwHvX916jkVWCzgocnWtw9aMPHogUEXf0FxA3DpbPyiy++4JVXXiEiIsLd8Qgh6kFusY3EsiK0AGO7GnmkW/0VoVVFBagfl6N+XAHFhaBpaL3vQnvwUbSQlvUSg7i+uJSsfHx8ZLhPiCYiPbuQv2/8DVO+lQBvHS/0bUXP1vp6ObYqLUVtWIX6/mvIzXE0RsSgGzkerU3beolBXJ9cSlYPP/wwn376KaNHj3ZeC1Xu0uK2QojGSynFf4+cdxah7Wj05ZUB9VOEVtltqG0bUCs+h2yTo7F9J3SjnkAL7+L244vrn0vJqnxe6qeffrrivq+++qpuIxJC1LnCUjtzt51l44lcAIbfbmBiVAu31/ZTSsG+HY5l6L+dcDS2ugXdyAnQvZfsjCBc5lKyev/9990dhxDCTU6eL2bmxt84faEEX08dk+8MY8BtgVd/4jVSRw9iX/oZlFdBDw5Be+AxtD5xaLr6v3ZLXN9cSlahobI1tBDXo/XHc5i3LYNim+KWIG9eGdCaNkHuLUKrfjuJfdki2LPd0aAPQLt3rGOVn5d39U8WogpVJqv58+fzzDPPAPDee+9V2V2fPHmyeyITQtRaic3Ogh0mVh89D0Bc20D+0CsMX0/3zTGr7EzHtVJb1oGyg7cP2pAH0YaORPNv5rbjihtDlcmqRYsWzp/DwsLqJRghxLU7l+coQvuLuWIRWnfND6m8C6j/foNa91+wloKHB9qAe9HuexgtyOCWY4obT5XJauTIkfz888/079+fMWPG1GdMQoha2n46l9lbzpJfYqdlWRHa9m4qQquKi1BrVqJWL4XCAgC0mAFoIx5Da1H/VblF01btnNVHH31E//796ysWIUQtXV6EtlcbPf/T+yb0PnW/kEFZraiNP6K++xIuOIYZ6RKFbtTjaLe0r/PjCQFXSVY12dRQCNEwzIVW3vn5Nw6YHEVoJ0SGMvKOui9Cq+x2ijb+hH3xPyHTsSEjbcMdSaqTVLcR7lVtsrLb7ezfv7/aF+jatWudBiSEcN2+c/m88/MlRWj7taJLy7ot/KqUgrRU7Es/I+fkMUdjWGt0IyZAjz5yrZSoF9Umq9LSUj744IMqe1iapsk1WEI0ALtSLD1g5vO9mdgVdGvpz0tuKEKrjh/B/p/P4PA+AHTGUNTwh9H6DkbzkGulRP2p9sz29fWVZCREI5NbbGP25jPsKCtCO6aLkXERdVuEVmWcxr7sX7Brs6PBvxnaPaMJGfMk2bm5dXYcIVxVL3sBzJs3j127dhEUFOTcIPHrr79m7dq1zlqD48aNo0ePHgAsW7aMpKQkdDodEydOJDIyEoDU1FQWLlyI3W5n8ODBjBgxAgCTycTs2bPJy8ujbdu2PP/883h61t82B0LUl0uL0OrLitBG12ERWmXJRn37BWrTGrDbwcsbbfD9aMMeQmumR/PxAUlWogHUywKLuLg4hg0bxty5cyu0Dx8+nAceeKBC2+nTp9m8eTP/+Mc/sFgsvPHGG7z77rsAfPzxx/zlL3/BaDQydepUoqOjadOmDf/6178YPnw4/fr148MPPyQpKYmhQ4fWSexCNAZKKValn2fBThNWu6Kj0Zc/9m9NC33dFKFV+XmoH5agkr6D0hLQ6dBi70a77xE0g7FOjiHEtag2WS1atKhODtK5c2dMJpNLj01JSaFv3754eXnRokULwsLCOHr0KOC4OLllS8deN3379iUlJYXWrVtz4MAB/ud//gdwJMZvvvlGkpVoMgpL7czblsGGExcAuDe8OZN6tMDL49qrUajiYlTSd6hVS6DAMaxIz77oRoxHC2tzza8vRF1p0LGy1atXs2HDBtq1a8fjjz+OXq/HbDbTsWNH52OCg4Mxmx3XjhiNF//DMxqNpKenk5ubi7+/Px5lk72XPl6I693JnGJmbigvQqvx3J03EVsHRWiVzYbatAb17Rdwvuz3pVOEY8uOth2rf7IQDaDBktXQoUMZPXo04NhmZNGiRSQkJFQ59FhZe22WzK5Zs4Y1a9YAMGPGDEJCQmr8GgCenp61fq47SVw109ji+vGQiQ82n8CUe4hAX0/yi61YFdwW7M/04Z24LfjalqUrpSjeso68zz/EfuYkAJ7twtFP+APeLmzZ0dg+r3ISV8001riq02DJqnnz5s6fBw8ezMyZMwFHjyk7O9t5n9lsJjg4GKBCe3Z2NgaDgYCAAAoKCrDZbHh4eFR4fGXi4+OJj4933s7KyqpV/CEhIbV+rjtJXDXTmOJKPp7D3LIK6QA5RVYAOoX48tfBbfC1F5CVVVDr11cH9zj2lfo13dEQGoY2Yjz26P7k6nRwye9XVRrT53UpiatmrjWuVq3qv5xWjZJVTk4ORUVFFdrK55BqymKxYDA4ilxu376dm2++GYDo6GjmzJnDfffdh8Vi4ezZs3To0AGlFGfPnsVkMhEcHMzmzZuZMmUKmqbRpUsXtm7dSr9+/Vi/fj3R0dG1ikmIhrQ4NdOZqC6VXWC9pmrp6sQvjiSVttvRENgc7f5H0PoPRZNVs+I64dKZmpqayj//+U/Onz9/xX2u7BQ8e/Zs0tLSyM3N5dlnn2Xs2LEcOHCAX3/9FU3TCA0N5emnnwbg5ptvpk+fPrz44ovodDp+97vfodM5flEnTZrE9OnTsdvt3HXXXc4E99hjjzF79my+/PJL2rZty6BBg1z+AIRoDA5nFZJZYK30vqwq2q9Gmc6gln+OStnoaPDzR7t7FFr8A2g+7iluK4S7aMqF9enPP/88999/P3FxcXh7N63N086cOVOr5zXV7r27SFyVO32hmH+lZrHlVNXXLoX6e7JgZAeXX1PlWFDffYna+CPYbODphTZoONo9o9H017Y4o6E/r6pIXDXTZIcB8/LyGDJkiNQAE6KOmAutfLk3i59+OY9dgbeHRmSYP6kZBZRcMhTo46ExIdK1nbpVQT5q9TLUmhVQUgyaDq3fYLT7H0Uzym7f4vrmUrIaNGgQ69atk+E1Ia5RfomNZWlmVh4yU2xT6DQY2iGIR7qFYPT3Ivl4DotTM8kqsBLi78mEyFAGtg2q9jVVaQlq3feo/y6B/LIeWuSd6EZMQGt9Sz28KyHcz6VklZ6ezg8//MCKFSsqrOID+Otf/+qWwIRoSkptdlaln+er/dnkFtsA6H2zngndQ2kT5ON83MC2QQxsG+TSMI2y21Bb1qFW/hvMZY/t2NlxrVSHO9z2XoRoCC73rKRXJUTN2ZViw68X+HxPFqb8UgA6h/rxRFQLOoX61eo1lVKwZxv2pYvh7ClHY+tb0T30BHTtKcP1oklyKVnFxcW5OQwhmhalFLvP5rMoNZPjlmIAbgny5vHIFkS3blbrhKKO7HcsQ//lkKPB2MKxjXyvWDSdbNkhmi6XL7JYt24dGzZscF50Gxsby1133eXO2IS4LqVnF7JodyZ7zzku4DX6e/JoRAh3tQ2q9TYe6vRxR09q3w5HQ0AQ2vCxaLHD0LzqppitEI2ZS8lq6dKlJCcnc//99zvH0leuXInFYmHUqFHujlGI68LZ3BL+tSeTn084Fjk089YxuouR4eEGfFy8qNe+dT1q2WLOWbLAEAKDH0A79QtqWzIoBT5+aEMfRBs6As23bncEFqIxcylZrV27lmnTphEaenH5a/fu3Xn99dclWYkb3vlCK1/tz2J1+nlsCrx0GvfdbuChLkYCfFwfmrNvXY9aPNex7BzAnAnffIwC8PBEi7sH7d4xaIHNq3sZIZokl5JVcXGxc5PEcgEBAZSUlLglKCGuBwWlNlYcNLP8oJkiq2MZ+uB2QYyLCCG0Wc2H5tSyxRcT1aW8fdBNew8tNKwOohbi+uRSsoqMjGTOnDk89thjhISEkJmZyRdffEH37t3dHZ8QjU6pTfHj0fN8tT+LnCLHMvSY1nomRIZya3Ofqzz7SkopOHbY0ZOqTEmJJCpxw3MpWU2aNIlPPvmEl19+GavViqenJ3369GHixInujk+IRsOuFJtO5PKvPZlk5DmWod8e4ssTkS3o0rLm80fKWorasQm19tuLldArE3x9beUghDu4lKz8/f2ZPHkyCQkJ5ObmEhAQ4CwuK8SNYG9GPp/tzuSo2bHrQOtAbyZEhtK7jb7Gy9DVBQsqeTUq+QfIsTgamwVA+05wcI9jW/ly3j5oIyfU1dsQ4rpVZbIymUy0aNECgHPnzlW479JtQmq7RYgQ14Nj5iIWpWay+6xjy3eDn2MZ+uB2NV+Grk4cRa391lEF3VpWSb31rWiD7kO7Mw7Nx8e5GpCy1YDayAnoesfV8bsS4vpTZbJ66aWXWLRoEQBTpkyp8gVc2SJEiOvNubwSPt+TRfKvFwDw99IxqnMwD3QKdnkZOoCyWlG7t6KSvoWjBx2Nmuao3TfoPugUUaFnpusdB73jGm21biEaSpXJqjxRgSQkceO4UGTl6wPZ/HDkPFa7wlOncW94c8Z0MRLo6/pGhSr3AmrjatT6Hxy9JAC/Zmj949HuGi4LJoSoIZd++z755BMmTZp0Rfunn37Kk08+WdcxCVHviqx2Vh4ysyzNTEGpHQ2Iuy2QR7uH0FLv+h5u6vRx1NrvHBfxls89hbVGG3Q/Wp+70HxrVw9QiBudS8kqOTm50mS1YcMGl5LVvHnz2LVrF0FBQcyaNQtw7JGVmJhIZmYmoaGhvPDCC+j1epRSLFy4kN27d+Pj40NCQgLt2rUDYP369SxduhSAUaNGOWsWHjt2jLlz51JSUkJUVBQTJ06UYp7CJVa7YlW6hS/3ZmEpW4be46ZmTIgMpV2wa7vpKrsNUrdjT/oODu+7eEfXnugG3w+dI9FkQZIQ16TaZJWUlASAzWZz/lzOZDIREBDg0kHi4uIYNmwYc+fOdbYtX76cbt26MWLECJYvX87y5csZP348u3fvJiMjgzlz5pCens6CBQt46623yMvLY8mSJcyYMQOAV199lejoaPR6PR999BHPPPMMHTt25G9/+xupqalERUXV6IMQNxalFFtP5fH59yc4db4QgA7BvjwRFUpEWDPXXiM/D/XzT6h130O2ydHo44fWd5Bj0URYa3eFL8QNp9pktXHjRgCsVqvz53JBQUE899xzLh2kc+fOmEymCm0pKSlMmzYNgIEDBzJt2jTGjx/Pjh07iI2NRdM0wsPDyc/Px2KxcODAASIiItDr9QBERESQmppKly5dKCwsJDw8HIDY2FhSUlIkWYkqHThXwKe7TRzJdqxqDdN7MSEylH63BLjUI1dnTzlW9W1Zd7HiRGiYY+v4vvFo/q4lOyGE66pNVq+//joAX375JY888kidHjgnJweDwQCAwWDgwgXHqiuz2UxIyMWLII1GI2azGbPZjNFodLYHBwdX2l7+eCEud+J8MYtTTaT85liGHuTrwVO9b6PvTZ54XmUZurLbYf9O7Gu/hbTUi3fc0d0x1Netp2zRIYQbuTRnNXbsWOx2e6X31fXFwUqpK9qq+m9X07RKH1+dNWvWsGbNGgBmzJhRITHWhKenZ62f604S15UyLhTx8daT/HDQhAL8vDx4tGdrHolqTaC/D9bya54qYS/Ipyjpewr+uwTb2dOORm8f/OLuwX/4aDxvaeeWmOV7rBmJq2Yaa1zVcSlZjRs3rsr7arusPSgoCIvFgsFgwGKxOAvlGo3GCteXZGdnYzAYCA4OJi0tzdluNpvp3LkzRqOR7OzsCo8PDg6u8rjx8fHEx8c7b9f2WpbGeh2MxHVRbrGNJQey+f6whVK7wkODYR2bM7ZbCM19PSm4YMHfu/K41LkzqHXfozatgSLHnBbBoY6hvv5DKGkWQAmAm96TfI81I3HVzLXG1apVqzqMxjUuJav333+/wm2LxcLy5cuJjo6u9YGjo6NJTk5mxIgRJCcnExMT42xftWoV/fr1Iz09HX9/fwwGA5GRkXzxxRfk5eUBsGfPHh599FH0ej1+fn4cOXKEjh07smHDBoYNG1bruMT1r9hq57vDFv6Tlk1+iWNEYMCtATzWPZSbAqpehq6UgrRUx1Df/p2O/aMAwruiG3wfdL8TzUOG+oRoCC4lq0v3sSq/PXnyZKZOncqgQYOu+vzZs2eTlpZGbm4uzz77LGPHjmXEiBEkJiaSlJRESEgIL774IgBRUVHs2rWLKVOm4O0bNBoMAAAcxUlEQVTtTUJCAgB6vZ6HHnqIqVOnAjB69GjnYounnnqKefPmUVJSQmRkpCyuuEHZ7Ip1x3P4954ssgsdQ3sRYf48EdmCDsaql6Gr4iLUliRU0vdw9pSj0dML7c6BjlV9bhrqE0K4TlM1nfQpk5WVxcsvv8zChQvrOqZ6debMmVo9r6l2793FnXEppdj+Wx6LUzM5leO4ELetwYcnoloQGeZf5ZynyjqH79YkCn5aCQWORRc0D0aLuxct9m60gCC3xOuKG/F7vBYSV8002WHA9957r8IvfHFxMQcPHmTAgAFuC0wIVxzMLGDR7kzSMh3zSi2aeTG+ewgDbgtEV0mSUkrBkf3Y13wLe7ZToMoWDrXvhDb4frSoPmierpdVEkLUD5d+K8PCKtYx8/HxYciQIURERLglKCGu5lROMYtTM9l22jGHGejjwdiuRoZ1bI6Xx5UrVFVJMWpbsmPvqN9OOBo9PPHtP4SSfkPR2nasz/CFEDXkUrIaM2aMu+MQwiXZBaV8sTeLtcdysCvw8dB48I5gRnYOxt/rysUPypyJWv9f1IYfIT/X0RjYHG3gMLSB9xDUvmOjHKYRQlTkciHbfv36cfvttzvbDh8+zJYtW6SQragXeSU2lh7I5tvDFkpsCl3ZMvSHu4UQ7FfxNFZKwS8HUWu+Re3eAuXXCN7awTHUF90fzcurAd6FEKK2XEpWmzZt4vHHH6/Q1q5dO95++21JVsKtSmx2fjhynm/2Z5Fbtgy97y0BjO8eSuvAisvQVWkpKmUjKuk7OHHU0ajTocUMQBt0n2NeSgocC3FdcilZaZp2RQULu91e4+oRQrjKZlck/3qBf+/JJLPAsQy9Sws/nohqwe0hFbfZUOfNqORVjm3ic3McjfoAtFjHUJ8WfH1dqS+EuJJLyapTp058+eWXjB8/Hp1Oh91u55tvvqFTp07ujk/cYJRS7DyTz6LUTE6cdxSJvTXIh8ejQunZqlmFnpE6fsRRUHbHJrCVlUxqc5tjqK9XLJq3T0O8BSGEG7iUrCZOnMiMGTN45plnnOvzDQYDr7zyirvjEzeQI1mFfJaayf5zBQCE+HvyWPdQBt4WiEdZoVlltaJ2bnIM9R077HiipoOo3ugGPwDhXWSoT4gmyKVkZTQamTlzJkePHiU7Oxuj0UiHDh3qvIituDGduVDC4j2ZbD7pWK2n99YxpquRe8MNeJctQ1e5OReH+s6XVdX3b4Y2YKjjIt6Qlg0VvhCiHrh89aPdbsdms6GUIjw8nKIix15Avr6u7aYqxOUshVa+2pfF6qPnsSvw9tC4/3YDo7oY0Xs7lqGrk8ccQ33bN4C11PHEm252DPX1jkPzkfNPiBuBS8nq5MmTzJw5Ey8vL7Kzs+nbty9paWkkJyfzwgsvuDtG0cQUlNpYlmZmxUEzxWXL0OPbBzEuIoQQfy+UzYbauRn72pWQXlZpX9MgIsaxd9Qd3WWoT4gbjEvJ6qOPPuLhhx8mNjaWiRMnAo7df+fPn+/W4MT1L/l4DotTM8kqOITR35OuLfzYdbaAC8U2AO5so2d8ZCi3BPmg8nOxr1qJWvdfMGc6XsDXD61fvGNrjhb1X49MCNE4uJSsTp8+fUUdQF9fX0pKStwSlGgako/nMHdbBsU2xyUOWQVW1v/qmJfqFOLHk1Gh3NHCH/XbCewrv0NtXQfl51SLVo6K5/0Gofn6N9RbEEI0Ei5vEXLs2DHat2/vbDt69OgVNQOFKGdXik92mZyJ6lKBPh78Lb412r6d2BZ/C4f2XryzcxS6+PuhSw80WcAjhCjjUrJ6+OGHmTFjBkOGDMFqtbJs2TJ++uknnnnmGXfHJ64jxVY7qRn5bD+dR8pveeQU2Sp9XG6xFfV/CajMDEeDjy9an0GOob6bbq7HiIUQ1wuXklXPnj2ZOnUqSUlJdO7cmczMTF566SXatZNN6W5054us7Pgtj22n80g9m0/JJT0pHQo7Vy6ECCk6D5kZYGzhGOrrH4/mr6/PsIUQ1xmXl663a9fuiuRktVrxvMa9f5577jl8fX3R6XR4eHgwY8YM8vLySExMJDMzk9DQUF544QX0ej1KKRYuXMju3bvx8fEhISHBGdP69etZunQpAKNGjSIuLu6a4hJVO51TzLbTeWw/ncfhrEIuHejraPSlV2s9MW30HJ/3Lh/cfA/FHhdr+PnYSnjs15/QPfcniIhB08k28UKIq3Mp07zxxhtMnjwZg8HgbDtx4gTvv/8+b7/99jUH8frrrxMYGOi8vXz5crp168aIESNYvnw5y5cvZ/z48ezevZuMjAzmzJlDeno6CxYs4K233iIvL48lS5YwY8YMAF599VWio6Od296La2OzKw5lFbK9LEGdyb24sMZTp9E9zJ9ebfREh3hgPJGGSt2J+tcubjFnohXk83m7e8jyaU5I8XkeO/YDsaY9aJGvNeA7EkJcb1xKVm3btuXll19m0qRJ9OnThxUrVrBixQrGjRvnlqBSUlKYNm0aAAMHDmTatGmMHz+eHTt2EBsbi6ZphIeHk5+fj8Vi4cCBA0RERDiTU0REBKmpqfTv398t8d0Iiqx2dp91zD/t+C3PudQcIMBbR3RrPTGt9URhwffQDtTKnZCehr28Rh+AphFrSiXWlFrxxYND6+ldCCGaCpeS1fjx4+nZsyfvv/8+n3/+OQaDgb/97W91thpw+vTpAAwZMoT4+HhycnKcvTiDwcCFCxcAMJvNhIRcrKBtNBoxm82YzWaMRqOzPTg4GLPZXCex3UjMhWXzT6dy2ZNRQKn94gBfmN6LO9vo6dXCm05Zh9HtX4NasRPMmReHATWdYxuOrj3QuvbEfvY0/GselBRfPIi3D9rICfX6voQQ1z+XJ5xMJhOFhYW0bNmS4uLiOrvG6o033iA4OJicnBzefPNNWrWq+sLPyrYkqaqSQVXta9asYc2aNQDMmDGjQvKrCU9Pz1o/151qEpdSiuPmAjb+Yubn42bSMnKd92lAl7AA+rcLpq9/IWHpOyjdtIWSg3vAanUmKC2wOT49euPdozc+3e9EFxhU4RiFgYHkff4B9iwTupAW6B97Fr+Bd9fRu712TeF7rE8SV81IXHXHpWT1j3/8g5MnT/LnP/+Z9u3bs2rVKl5//XVGjhzJAw88cE0BBAcHAxAUFERMTAxHjx4lKCgIi8WCwWDAYrE457OMRmOFLcizs7MxGAwEBweTlpbmbDebzXTu3LnS48XHxxMfH++8Xdstzcurzzc2V4vLZlekZRY4558y8kqd93l7aHQPa0avlt5E5x2n+aEk1I+O3lN++YM0DdrdjtatJ1rXnnBLe0p1OkqB/JJSuPzYXXqivfURLcviygfyG9Hndr1+jw1F4qqZphpXdZ0Kd3EpWQUGBvL3v/8db2/Hqq5hw4YRERHB+++/f03JqqioCKUUfn5+FBUVsXfvXkaPHk10dDTJycmMGDGC5ORkYmJiAIiOjmbVqlX069eP9PR0/P39MRgMREZG8sUXX5CXlwfAnj17ePTRR2sdV1NTUGpzzD+dymPHmTzySi5upBnk40F062b08i8iImMvPrt2QvqBCr0nAoLQuvSArj3QukSh6QMrPY4QQriLS8nqqaeeuqKtVatWvPnmm9d08JycHN555x0AbDYb/fv3JzIykvbt25OYmEhSUhIhISG8+OKLAERFRbFr1y6mTJmCt7c3CQkJAOj1eh566CGmTp0KwOjRo2/4lYDZBaXO3tPecwVYL5l/ah3oTUyYL3eWnqHjse3oVuyCbNPFJ5f3nrqW9Z5ubS/VJIQQDUpT1exN/8knnzBp0iTn7aSkJAYNGuS8/c477/DSSy+5N0I3O3PmTK2e19i690opTpwvZp/ZzvojJo6ai5z3aUCnUD96BVqJMR+m1aEtzt6TU0AQWpco6NrTLb2nxvZ5lZO4akbiqpmmGlejGwZMTk6ukKwWL15cIVnt27fPfZGJq7LaFQdMF+efTPkV55+iWvrSi2x6/raTwKRt0nsSQly3qk1W1XS6RAPJL7Gx64zj+qedZ/LIL704/9Tc14N+Yb50Nx+iW/omfJL3Vuw96QPRuvZw9J46R6EFyNyTEOL6UG2ykg3uGofM/PL5p1z2mwqwXsxP3BzoRS/vPGKy9tMhZR26y3tPbcMdvadu0dJ7EkJct6pNVjabjf379ztv2+32K26LuqeU4rilmG2nc9l+Oo9jlosX1eo06NJcR0zpWWJObOWmn7dV6D1pgc2hc6T0noQQTUq1ySooKIh//vOfztt6vb7C7Uvr+YlrU2pT7DcVsL0sQWUVXExAvh4aUc1KiLmQTs8DSQSYTl58YoXeU09CevYmW6p3CCGamGqT1dy5c+srjhtSXrGNnWcc22vsOpNP4SXjewZvjRidhV4Zu+l6YB3epRdX96EPvGTlXo8KvScZ5hNCNEXXtr+HqLFzeSXO1XsHTAVcupHurT42ehWfIvroz7Q/vQ9d+WW5l/WeHHNPsrWGEOLGIcnKzexK8Yu5iO2nHT2oE+crzj9188onxnyImENJtMzPvPjECr2nKLSAoEpeXQghbgySrNygxGZnX0YB28q2dzcXXpx/8tMpetgziTmVQo8T29FbCx13SO9JCCGqJMmqhpKP57A4NZOsgkOE+HsyITKUgW2DuFBsY2fZ9u67z+ZTdMn8k1FXSq+848Qc20SX7CN4qbK9ofQBaF0GSu9JCCGuQpJVDSQfz2HutgyKyyaaMgusvLvlLF/vy+JMXimXlN+jLbnEmPYRczKFdnm/oYGj93RbR+d+T9zWQXpPQgjhAklWNbA4NdOZqMrZFJzOLcUDRfeSDGJO7SDGtI/Q4vOOB+gD0HoNhG7SexJCiNqSZFUDWQWlOMrCXkYpPv35dZrZihy9p1s7oHUbJr0nIYSoI5KsaiCk5AKZ3lf2jEKLz6PveSd061F23ZP0noQQoi7JFaQ18NjR/+JjK6nQ5mMr4bFjq9D9/n/R9b5LEpUQQriBJKsaiLWe5g+HlxBaZEFTitAiC384vIRY6+mGDk0IIZq0JjUMmJqaysKFC7Hb7QwePJgRI0bU6etrIycQu3gusVtTLzZ6+6BNeK5OjyOEEKKiJtOzstvtfPzxx/zpT38iMTGRTZs2cfp03fZ4dL3jHIkpONSxkCI4FG3Cc+h6x9XpcYQQQlTUZHpWR48eJSwsjJYtWwLQt29fUlJSaNOmTZ0eR9c7DnrHNdrtqoUQoilqMj0rs9mM0Wh03jYajZhlqwwhhGgSmkzPSil1RVtlOx2vWbOGNWvWADBjxgxCQkJqdTxPT89aP9edJK6akbhqRuKqGYmr7jSZZGU0GsnOznbezs7OxmAwXPG4+Ph44uPjnbdrO5TXWIcBJa6akbhqRuKqmaYaV6tWreowGtc0mWHA9u3bc/bsWUwmE1arlc2bNxMdHd3QYQkhhKgDTaZn5eHhwaRJk5g+fTp2u5277rqLm2++uaHDEkIIUQc0VdlkjxBCCNGINJlhwPr26quvNnQIlZK4akbiqhmJq2YkrrojyUoIIUSjJ8lKCCFEo+cxbdq0aQ0dxPWqXbt2DR1CpSSumpG4akbiqhmJq27IAgshhBCNngwDCiGEaPSazHVWdeG5557D19cXnU6Hh4cHM2bMIC8vj8TERDIzMwkNDeWFF15Ar9ejlGLhwoXs3r0bHx8fEhISnN3q9evXs3TpUgBGjRpFXFxcrWM6c+YMiYmJztsmk4mxY8eSn5/P2rVrCQwMBGDcuHH06NEDgGXLlpGUlIROp2PixIlERkYC176Fyrx589i1axdBQUHMmjULoE4/n2PHjjF37lxKSkqIiopi4sSJlZbMciWuxYsXs3PnTjw9PWnZsiUJCQk0a9YMk8nECy+84LwCv2PHjjz99NPVHr+q91ibuL7++us6+95MJhOzZ88mLy+Ptm3b8vzzz+PpefVf6criSkxM5MyZMwAUFBTg7+/P22+/Xa+fV1ZWFnPnzuX8+fNomkZ8fDz33ntvg59jVcXV0OdYVXE1hnPMLZRwSkhIUDk5ORXaFi9erJYtW6aUUmrZsmVq8eLFSimldu7cqaZPn67sdrs6fPiwmjp1qlJKqdzcXPXcc8+p3NzcCj/XBZvNpp566illMpnUV199pVasWHHFY06dOqVeeuklVVJSos6dO6cmT56sbDabstlsavLkySojI0OVlpaql156SZ06dapGxz9w4ID65Zdf1Isvvuhsq8vP59VXX1WHDx9WdrtdTZ8+Xe3atavWcaWmpiqr1eqMsTyuc+fOVXjcpao6flXvsTZx1eX3NmvWLPXzzz8rpZSaP3++Wr16da3jutRnn32mvvnmG6VU/X5eZrNZ/fLLL0oppQoKCtSUKVPUqVOnGvwcqyquhj7HqoqrMZxj7iDDgFeRkpLCwIEDARg4cCApKSkA7Nixg9jYWDRNIzw8nPz8fCwWC6mpqURERKDX69Hr9URERJCamlrdIVy2b98+wsLCCA0NrTbevn374uXlRYsWLQgLC+Po0aMVtlDx9PR0bqFSE507d77iv726+nwsFguFhYWEh4ejaRqxsbEux1dZXN27d8fDwwOA8PDwq1bgr+74Vb3H2sRVlZp+b0opDhw4QO/evQGIi4urk7iUUmzZsoV+/fpV+xru+LwMBoOzZ+Tn50fr1q0xm80Nfo5VFVdDn2NVxVWV+jzH3EGGAS8zffp0AIYMGUJ8fDw5OTnOgrgGg4ELFy4Aji1JLq1aXL4lyeVblQQHB9fZViWbNm2q8Edk9erVbNiwgXbt2vH444+j1+sxm8107Nix0uNfvoVKenr6NcdUV5+PO7d4SUpKom/fvs7bJpOJP/7xj/j5+fHII49wxx13VHv8qt5jbdXF95abm4u/v7/zj2VdnWcHDx4kKCiIm266ydnWEJ+XyWTi+PHjdOjQoVGdY5fGdamGPscujevQoUON+hyrLUlWl3jjjTcIDg4mJyeHN998s9rKwsrFLUmqa68Jq9XKzp07efTRRwEYOnQoo0ePBuCrr75i0aJFJCQkVBpXTeOtCzX9fKqK+1otXboUDw8PBgwYADj+EMybN4+AgACOHTvG22+/zaxZs9x2/Ms19u/t8n+IGuLzKioqYtasWTz55JP4+/tX+bj6Pseqiquhz7HL42rs51htyTDgJYKDgwEICgoiJiaGo0ePEhQUhMViARzd+PJJS6PRWKHEfvmWJMHBwRW2KjGbzZVuVVJTu3fvpm3btjRv3hyA5s2bo9Pp0Ol0DB48mF9++cUZ1+XHDw4OdnkLlZqqq8+nsvjKv4/aWr9+PTt37mTKlCnOXz4vLy8CAgIAx3UmLVu25OzZs9Uev6r3WBt19b0FBARQUFCAzWar8PhrYbPZ2L59e4UeQn1/XlarlVmzZjFgwADuvPPOal+vPs+xyuKChj/HKourMZ9j10KSVZmioiIKCwudP+/du5dbbrmF6OhokpOTAUhOTiYmJgaA6OhoNmzYgFKKI0eO4O/vj8FgIDIykj179pCXl0deXh579uxxrri5Fpf/x1t+YgNs377dWWE+OjqazZs3U1paislk4uzZs3To0MFtW6jU1edjMBjw8/PjyJEjKKXYsGHDNcWXmprKihUreOWVV/Dx8XG2X7hwAbvdDsC5c+c4e/YsLVu2rPb4Vb3H2qir703TNLp06cLWrVsBxx/Na/0+9+3bR6tWrSoMCdXn56WU4oMPPqB169bcd999zvaGPseqiquhz7Gq4mrM59i1kIuCy5w7d4533nkHcPyH2b9/f0aNGkVubi6JiYlkZWUREhLCiy++6Fw2+/HHH7Nnzx68vb1JSEigffv2gGP8etmyZYBj2exdd911TbEVFxfzhz/8gffff985/PDee+/x66+/omkaoaGhPP30086e0tKlS1m3bh06nY4nn3ySqKgoAHbt2sVnn33m3EJl1KhRNYpj9uzZpKWlkZubS1BQEGPHjiUmJqbOPp9ffvmFefPmUVJSQmRkJJMmTXJpOKKyuJYtW4bVanUuJChfPrx161a+/vprPDw80Ol0jBkzxvkLWNXxqzoHahPXgQMH6ux7O3fu3BXLir28vGoV16BBg5g7dy4dO3Zk6NChzsfW5+d16NAhXnvtNW655Rbn9z5u3Dg6duzYoOdYVXEtXLiwQc+xquLatGlTg59j7iDJSgghRKMnw4BCCCEaPUlWQgghGj1JVkIIIRo9SVZCCCEaPUlWQgghGj1JVkI0Qs899xx79+6t9+OWV/UvvxBUiMZCkpUQN7CGSopC1JQkKyFqqLJeh/REhHAvKWQrBLB8+XLWrl1LTk4ORqORcePG0atXL8BRZmbt2rW0b9+e5ORk7r77bsLCwq5oi4uLY/78+Zw4cQJN0+jevTu/+93vaNasGStXruTIkSO89NJLzmN+8sknzkoC1bHb7axcuZK1a9eSn59P165defrpp9Hr9ZhMJiZPnkxCQgJfffUVJSUlDB8+3FmBoKSkhA8//JCdO3fSvHlz4uLi+OGHH/jggw947733yMrKYubMmeh0OkaPHk2fPn0A2LhxY6WvJ0RDkZ6VEEDLli3561//yqeffsqYMWN47733KtRYS09Pp2XLlixYsMD5h7uytpEjRzJ//nwSExPJzs7mm2++AWDAgAHs2bOH/Px8wNET27x5M7GxsVeN7YcffiAlJYVp06Yxf/589Ho9CxYsqPCYQ4cO8e677/J///d/LFmyhNOnTwPwzTffkJmZyXvvvcdf/vIXNm7c6HzO888/T0hICK+88gqLFy/mwQcfvOrrCdFQJFkJAfTp04fg4GB0Oh19+/Z1bkxXzmAwcM899+Dh4YG3t3elbWFhYURERODl5UVgYCDDhw8nLS3N+dg77riDLVu2AI4iqAEBAc7N86qzZs0aHnnkEYxGI15eXowZM4Zt27ZVGHocM2YM3t7e3Hbbbdx6662cOHECgC1btjBy5Ej0ej1Go5F77rnHpc+jqtcToqHIMKAQOCpdf/fdd2RmZgKOyvu5ubnO+y/d5K+qtpycHBYuXMjBgwcpKirCbrdXKEY6cOBAfvzxR+Lj49m4caNLvSqAzMxM3nnnnQoFV3U6HTk5Oc7b5VvHAPj4+FBUVAQ4KnBfvkGhK6p6PSEaiiQrccPLzMxk/vz5vPbaa4SHh6PT6Xj55ZdrvFnev//9bwDeeecdAgIC2L59O5988onz/piYGBYsWMDJkyfZuXMn48ePd+l1jUYjf/jDH+jUqdMV95lMpmqf27x5c7Kzs2nTpg1AhX2LhLieyDCguOEVFxejaZpzw7t169Zx6tSpGr9OYWEhvr6+NGvWDLPZzLffflvhfm9vb+68807mzJlDhw4dKu2tVWbIkCF8+eWXzl7fhQsXSElJcem5ffr0Yfny5eTl5WE2m1m1alWF+5s3b37VhCdEYyDJStzw2rRpw3333cef//xnfv/733Py5Eluv/32Gr/OmDFjOH78OE888QR/+9vfnKsJLxUXF8fJkyddHgIEuPfee4mOjubNN9/k8ccf589//jPp6ekuPXf06NEEBwczefJk3njjDXr37l1hP6IRI0bwn//8hyeffJKVK1e6HJMQ9U32sxKiHmVlZfH//t//48MPP3RupFmffvzxRzZt2sRf//rXej+2ENdCelZC1BO73c53331H37596y1RWSwWDh06hN1u58yZM3z77beV9viEaOxkgYUQ9aCoqIjf//73hIaG8qc//anejmu1Wvnoo48wmUz4+/vTr18/7r777no7vhB1RYYBhRBCNHoyDCiEEKLRk2QlhBCi0ZNkJYQQotGTZCWEEKLRk2QlhBCi0ZNkJYQQotH7/94K4km/1J1cAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<function matplotlib.pyplot.close(fig=None)>"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plt.style.use('ggplot')\n",
    "# sns.set_context('poster')\n",
    "\n",
    "plt.plot( 'array_length', 'int_sort_time', data=df, marker='o',linewidth=2,linestyle='solid')\n",
    "plt.plot( 'array_length', 'fl_sort_time', data=df, marker='o',linewidth=2,linestyle='solid')\n",
    "plt.xlabel('array length')\n",
    "plt.ylabel('Execution Time (milliseconds)')\n",
    "plt.title('Fig 1: Selection Sort Execution Time vs Array Length\\n for Integers and Floats')\n",
    "plt.legend()\n",
    "plt.show()\n",
    "plt.close()\n",
    "\n",
    "plt.plot( 'array_length', 'str5_sort_time', data=df, marker='o',linewidth=2,linestyle='solid')\n",
    "plt.plot( 'array_length', 'str15_sort_time', data=df, marker='o',linewidth=2,linestyle='solid')\n",
    "plt.xlabel('array length')\n",
    "plt.ylabel('Execution Time (milliseconds)')\n",
    "plt.title('Fig 2: Selection Sort Execution Time vs Array Length\\n for Strings w/ Lengths 5 & 15')\n",
    "plt.legend()\n",
    "plt.show()\n",
    "plt.close()\n",
    "\n",
    "plt.plot( 'array_length', 'int_sort_time', data=df, marker='o', linewidth=2,linestyle='solid')\n",
    "plt.plot( 'array_length', 'fl_sort_time', data=df, marker='o', linewidth=2,linestyle='solid')\n",
    "plt.plot( 'array_length', 'str5_sort_time', data=df, marker='o', linewidth=2,linestyle='solid')\n",
    "plt.plot( 'array_length', 'str15_sort_time', data=df, marker='o', linewidth=2,linestyle='solid')\n",
    "plt.xlabel('array length')\n",
    "plt.ylabel('Execution Time (milliseconds)')\n",
    "plt.title('Fig 3: Selection Sort Execution Time vs Array Length\\n for Numeric and String Data')\n",
    "plt.legend()\n",
    "\n",
    "plt.show()\n",
    "plt.close"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "false",
    "cell_style": "center"
   },
   "source": [
    "# Executive Summary"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "Collapsed": "false",
    "cell_style": "center"
   },
   "source": [
    "As noted in Week 1, algorithm speed is measured in terms of how quickly the run time increases as the size of the input increases, written in Big O notation, O(n), where n is the number of operations an algorithm will make, in the worse case scenario.\n",
    "\n",
    "\n",
    "Above, we implemented the selection sort algorithm, which sorts a set of elements. In order to sort a list, you have to check each item on the list; so this checking operation takes O(n) time, which needs to be performed n times, resulting in algorithm speed of O(n × n) time or O(n^2) time. Data types can also impact algorithm speed, since different data types have different size ranges. For example, an unsigned 8-bit integer type has a range of 0 through 255, whereas an unsigned 16-bit integer types has a range of 0 through 65,535 (McConnell, 2004). We profiled the selection sort algorithm for multiple data types over a dimension of size. For a single data type, we expected sort execution time to increase as the size of the array increased, based on our knowledge that the speed of the algorithm is O(n^2). We can see this trend clearly in Figure 3; as array length increases across data types, sort time also increases. \n",
    "\n",
    "\n",
    "In Figure 1, we compared the sort time for integers vs floats. We can see that the algorithm sorts integers slightly faster than floats, which intuitively makes sense since integers are simple and floats are more complex. At array length of 20,000, the difference between sort times is very clear, however, at array length of 25,000, the sort times are much closer together. Similarly, in Figure 2, we compared the sort time for strings of length 5 and strings of length 15. Here, we also expected the small string type (length 5) to be sorted faster than the larger string type (length 15), however, the times are very close together, with a visible difference at array length of 25,000, where the longer string has a slightly larger sort time. \n",
    "\n",
    "\n",
    "Based on Big O speed of the sort algorithm, our results did not perfectly align with our expectations. This may be because of overhead, or other processes running on the machine while the algorithm was sorting. Also, Python is not the best programming language to test this on, since performance differences are more apparent in low-level programming languages like C. \n",
    "\n",
    "\n",
    "Sorting algorithms are very useful to data engineers, since certain tasks, such as binary search from last week, only work if your data is in sorted order. When using an algorithm like this one, data engineers should also be mindful of the data types they're using. If the Big O run time is O(n^2), as is the case with selection sort, choosing a data type that processes faster will make an overall performance difference in the application. \n"
   ]
  }
 ],
 "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": "3.7.6"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {
    "height": "390.972px",
    "width": "403.993px"
   },
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": true
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "position": {
    "height": "422.4px",
    "left": "1016.93px",
    "right": "20px",
    "top": "242px",
    "width": "350.925px"
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}