02_knapsack_2D_stochastic.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Genetic Algorithms With Pareto Fronts\n",
    "\n",
    "<img src=\"https://i.pinimg.com/originals/1a/5c/a8/1a5ca8373a83e8e84f8f268ee3ecf1de.png\" width=700>\n",
    "\n",
    "\n",
    "This notebook may be used in Google Colab. To access press this button:    \n",
    "<a href=\"https://bit.ly/genetic-algorithm-colab\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"First: Open In Colab\"/></a>\n",
    "\n",
    "Here you will learn about implementing Pareto Optimisation with a Genetic Algorithm.  \n",
    "\n",
    "This notebook follows the previous one in which we learned about the fundamentals of Pareto Fronts, where we examined the Knapsack problem in an exhaustive search space ([Github](https://bit.ly/pareto-front), [Google Colab](https://bit.ly/pareto-front-colab)). Here we will learn to apply in a search space that is computationally large.  \n",
    "\n",
    "Most real life situations involve search spaces that are intractable, meaning that we might never have enough computational power to access all of the feasible solutions (baring from introduction of quantum computing). One approach is to search in a stochastic (random) manner. Not completely random, though, we would like to both exploit what we know works as well as explore new regions in the Decision and Objective spaces.  \n",
    "\n",
    "To this end we will define and use the Genetic Algorithm as our stochastic search mechanism of choice, where Pareto Fronts are used as its *Selection Function* as detailed below. \n",
    "\n",
    "**Main Take Aways**  \n",
    "* In large combinatorial spaces we are not likely to have access to all the feasible solutions, and need to resort to subsets as proxies.\n",
    "* The difference between the ***true Pareto front*** of all feasible solutions and ***Pareto front approximations*** of subset solutions.         \n",
    "* ***Niching*** generates a better ***diversity*** of solutions resulting in a better Pareto approximation.  \n",
    "* The basics of Genetic Algorithms with an emphasis on Pareto fronts as the selection function.   \n",
    "* You will see how the Genetic Algorithm learns by tracking the evolution of the Pareto front approximations.\n",
    "\n",
    "\n",
    "**The Knapsack Problem**   \n",
    "As in the previous notebook, in this challenge you will fill virtual knapsacks with packages that have monetary value and weight.   \n",
    "The objective is to minimise the final weight of the knapsack and maximise the total monetary value of its contents.    \n",
    "(You will also be able to change the objectives from min-max to the other options: max-max and min-min, max-min.)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Recommended Notebook Usage**\n",
    "\n",
    "We first recommend that you be familiar with the fundamentals of Pareto fronts as introduced in the previous notebook ([Github](https://bit.ly/pareto-front), [Google Colab](https://bit.ly/pareto-front-colab)). You will notice that some conventions used there are applied here, too.\n",
    "\n",
    "Do each of the following steps at a time and answer the summary questions to ensure that the main lessons have been understood. \n",
    "\n",
    "<ul>\n",
    "      \n",
    "<details>\n",
    "        <summary>Setup, Decision & Objective Spaces</summary> \n",
    "            The data generation of packages and knapsacks is the same as before with the exception that we will explore subsets of all feasible knapsacks, due to the fact that the Decision Space may be intractable. The Objective Space and objectives are the same. \n",
    "</details>\n",
    "    \n",
    "<details>\n",
    "        <summary>Pareto Front Approximations</summary> \n",
    "            Here you will learn how niching which improves diversity results in better Pareto front approximations.\n",
    "</details>\n",
    "    \n",
    "<details>\n",
    "        <summary>Genetic Algorithms Components</summary> \n",
    "            You will get an understanding of the different parts that make the Genetic Algorithm with an emphasis of the Pareto optimisation mechanism implemented in the selection function.\n",
    "</details>\n",
    "    \n",
    "<details>\n",
    "        <summary>Knapsack Optimisation</summary> \n",
    "            Here you will see the Genetic Algorithm learn better solutions for the Knapsack Problem while comparing results of Pareto Optimisation vs. Single Objective Optimisation.  \n",
    "</details>  \n",
    "    \n",
    "</ul>\n",
    "    \n",
    "There is no need to understand every line of Python, especially in the first run! Reading the cell headers is sufficient to understand what a cell does.  It is more important to understand the text in the Markdown cells and the visuals created.  \n",
    "\n",
    "We recommend doing the first run until the [Summary](#Summary) section as is. \n",
    "Afterwhich we recommend that you explore different types of results by modfiying the variables flagged with ***CHANGE ME!***. We list in the [Summary](#Summary) section suggested actions to take, but feel free to be creative!\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src=\"https://tinyurl.com/vrdj29w9\" width=\"700\">  \n",
    "\n",
    "**One final note for PyCon USA participants**  \n",
    "\n",
    "Don't worry if you don't manage to to go through everything in depth during the allocated time. You will be able to access to this notebook later."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:46.102435Z",
     "start_time": "2020-11-11T07:22:45.823552Z"
    }
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:46.145289Z",
     "start_time": "2020-11-11T07:22:46.120648Z"
    }
   },
   "outputs": [],
   "source": [
    "SMALL_SIZE = 12\n",
    "MEDIUM_SIZE = 16\n",
    "BIGGER_SIZE = 20\n",
    "\n",
    "plt.rc('font', size=SMALL_SIZE)          # controls default text sizes\n",
    "plt.rc('axes', titlesize=SMALL_SIZE)     # fontsize of the axes title\n",
    "plt.rc('axes', labelsize=MEDIUM_SIZE)    # fontsize of the x and y labels\n",
    "plt.rc('xtick', labelsize=SMALL_SIZE)    # fontsize of the tick labels\n",
    "plt.rc('ytick', labelsize=SMALL_SIZE)    # fontsize of the tick labels\n",
    "plt.rc('legend', fontsize=SMALL_SIZE)    # legend fontsize\n",
    "plt.rc('axes', titlesize=BIGGER_SIZE)  # fontsize of the figure title\n",
    "\n",
    "plt.rcParams['figure.figsize'] = 12, 8"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Packages \n",
    "\n",
    "Let's create some data!\n",
    "\n",
    "<img src=\"https://uk.packsize.com/wp-content/uploads/sites/11/2015/05/wp-boxes.jpg\" width=300>\n",
    "\n",
    "Same as in the previous tutorial, here we create packages by defining for each its weight and monetary value according to a distribution.   \n",
    "\n",
    "Parameters of interest:  \n",
    "* `N_PACKAGES` - the total number packages to choose from.  \n",
    "* `value_distribution_mode` - indicates the relationship between the package weight and monetary value.   \n",
    "The weights are randomly distributed where: \n",
    "    * `value_distribution_mode=random` - monetary value distribution also random\n",
    "    *  `value_distribution_mode=squared` - monetary value is squared that of the weights (up to some noise quantified by `noise_factor`).  \n",
    "    \n",
    "This will results in \n",
    "* `packages` a dictionary containing information of the `N_PACKAGES` packages.\n",
    " It has the form `{package_number: package}`, where each `package` has the form `{'weight': weight, 'value': value}`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:47.032106Z",
     "start_time": "2020-11-11T07:22:47.027141Z"
    }
   },
   "outputs": [],
   "source": [
    "seed = 3141592653 # CHANGE ME! to get a different distribution\n",
    "\n",
    "N_PACKAGES = 20   # CHANGE ME! the total number of packages to choose from\n",
    "\n",
    "WEIGHT_MEAN = 0.1 # CHANGE ME! the mean of the weight distribution\n",
    "WEIGHT_SD = 10    # CHANGE ME! the standard deviation of the weight distribution\n",
    "\n",
    "VALUE_MEAN = 0.1  # CHANGE ME! when value_distribution_mode='random': the mean of the monetary value distribution \n",
    "VALUE_SD = 100    # CHANGE ME! when value_distribution_mode='random': the standard deviation of the monetary value distribution \n",
    "\n",
    "value_distribution_mode = 'random' # CHANGE ME! options: 'random', 'squared'\n",
    "noise_factor = 1./2  # CHANGE ME! when value_distribution_mode='squared': the standard deviation of the noise introduced\n",
    "# ---------------\n",
    "\n",
    "np.random.seed(seed)\n",
    "\n",
    "weights = np.abs(np.random.normal(WEIGHT_MEAN, WEIGHT_SD, N_PACKAGES))\n",
    "\n",
    "if 'squared' == value_distribution_mode:\n",
    "    values =  weights ** 2 \n",
    "    values += values * np.random.normal(0, noise_factor, len(weights))\n",
    "    values = np.abs(values)\n",
    "elif 'random' == value_distribution_mode:\n",
    "    values = np.abs(np.random.normal(VALUE_MEAN, VALUE_SD, N_PACKAGES))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:47.355120Z",
     "start_time": "2020-11-11T07:22:47.349265Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{0: {'weight': 16.045064716419265, 'value': 122.44920604368212},\n",
       " 1: {'weight': 10.257933643337433, 'value': 34.08475741925324},\n",
       " 2: {'weight': 7.427142915985984, 'value': 7.3477194585976235},\n",
       " 3: {'weight': 16.1061446548582, 'value': 75.21497126311662}}"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# packages dictionary\n",
    "packages = {idx: {'weight': weights[idx], 'value': values[idx]} for idx in range(N_PACKAGES)}\n",
    "\n",
    "{ipackage: package for ipackage, package in packages.items() if ipackage < 4}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:47.893777Z",
     "start_time": "2020-11-11T07:22:47.664119Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Packages')"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# visualising the package weight-value distribution\n",
    "\n",
    "plt.scatter([packages[idx_package]['weight'] for idx_package in packages],\n",
    "            [packages[idx_package]['value'] for idx_package in packages],\n",
    "            color = 'purple', alpha=0.04\n",
    "           )\n",
    "\n",
    "for idx_package, package in packages.items():\n",
    "    plt.annotate(f'{idx_package:02.0f}', xy=(package['weight'], package['value']), fontsize=MEDIUM_SIZE)\n",
    "    \n",
    "plt.xlabel('weight [kg]')\n",
    "plt.ylabel('value [£]')\n",
    "plt.title('Packages')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#  Decision and Objective Spaces\n",
    "\n",
    "\n",
    "## Decision Space Size\n",
    "There are two parameters of interest that decide how many knapsacks are possible: \n",
    "* `N_PACKAGES` which was determined in [Packages](#Packages).\n",
    "* `N_PACKAGES_PER_KNAPSACK` introduced here determines the number of packages that fit into a knapsack.  \n",
    "\n",
    "\n",
    "We can easily calculate the number of combinations defined as an unordered arrangement of $k$ objects from a set of $n$, which is called  *$n$ choose $k$*,    \n",
    "which is calculated as:  \n",
    "$$\n",
    "_nC_k=\\frac{n!}{(n-k)!k!}\\equiv {n \\choose k}\n",
    "$$\n",
    "\n",
    "Here we use     \n",
    "`k = N_PACKAGES_PER_KNAPSACK`  \n",
    "`n = N_PACKAGES`,  \n",
    "with assumptions for simplicity:  \n",
    "* A package can appear only once in a a knapsack.  \n",
    "* The order in which a package is assigned to a knapsack does not matter. \n",
    "\n",
    "**Note**  \n",
    "You do not need to understand the equation above.  \n",
    "Below you are provided with a calculator that executes this equation using the `math.factorial` function.    \n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:48.278826Z",
     "start_time": "2020-11-11T07:22:48.276081Z"
    }
   },
   "outputs": [],
   "source": [
    "from math import factorial"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:48.528455Z",
     "start_time": "2020-11-11T07:22:48.523354Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'There are 184,756 combinations of 20 choose 10.'"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def pretty_int(x):\n",
    "    if x > 1.e6:\n",
    "        return f'{int(x):0.1e}'\n",
    "    else:\n",
    "        return f'{int(x):,}'\n",
    "\n",
    "N_PACKAGES_PER_KNAPSACK = 10\n",
    "# -----------\n",
    "    \n",
    "n = N_PACKAGES\n",
    "k = N_PACKAGES_PER_KNAPSACK\n",
    "\n",
    "n_combinations = factorial(n)//factorial(k)//factorial(n-k)\n",
    "\n",
    "f'There are {pretty_int(n_combinations)} combinations of {n} choose {k}.'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Knapsacks\n",
    "\n",
    "Here we will create the `n_knapsacks` knapsacks using the `create_knapsacks` function which does two functions: \n",
    "* Decision Space: assigns packages to the knapsack  \n",
    "* Objective Space: calculates that knapsack weight and the sum of the monetary value of its contents using the `knapsack_objective` function.\n",
    "\n",
    "\n",
    "<img src=\"https://www.chatelaine.com/wp-content/uploads/2019/07/5-most-popular-backpacks-feature.jpg\" width=500>\n",
    "\n",
    "\n",
    "Each knapsack contains one feasible combination of `N_PACKAGES_PER_KNAPSACK` packages (of the total `N_PACKAGES` packages). \n",
    "\n",
    "* The `knapsack_objective` function determines the knapsack objective value (`weight`, `value`) by summing those of all its `packages`.  \n",
    "* The `create_knapsacks` function creates `n_knapsacks`. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:48.981455Z",
     "start_time": "2020-11-11T07:22:48.979324Z"
    }
   },
   "outputs": [],
   "source": [
    "# the properties of the knapsack that we want to optimise\n",
    "\n",
    "objectives = ['weight', 'value']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:49.207971Z",
     "start_time": "2020-11-11T07:22:49.205146Z"
    }
   },
   "outputs": [],
   "source": [
    "# mapping the Decision Space to the Objective Space: calculates for each knapsack its `objective` value\n",
    "\n",
    "def knapsack_objective(knapsack, objective, packages):\n",
    "    knapsack[objective] = np.sum([packages[idx_package][objective] for idx_package in knapsack['packages']])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:49.459509Z",
     "start_time": "2020-11-11T07:22:49.455218Z"
    }
   },
   "outputs": [],
   "source": [
    "# creates a dictionary of `n_knapsacks` knapsacks\n",
    "\n",
    "def create_knapsacks(n_knapsacks=1000, seed=0, n_packages=N_PACKAGES, n_packages_per_knapsack=N_PACKAGES_PER_KNAPSACK):\n",
    "    knapsacks = {}\n",
    "    for idx_knapsack in range(n_knapsacks):\n",
    "        if seed:\n",
    "            np.random.seed(idx_knapsack + seed)\n",
    "        this_idxs_packages = np.random.choice(range(n_packages), n_packages_per_knapsack, replace=False)\n",
    "\n",
    "        knapsacks[idx_knapsack] = {}\n",
    "        knapsacks[idx_knapsack]['packages']= this_idxs_packages\n",
    "\n",
    "        for objective in objectives:\n",
    "            knapsack_objective(knapsacks[idx_knapsack], objective, packages)\n",
    "            \n",
    "    return knapsacks"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Pareto Front Approximation\n",
    "\n",
    "In a combinatorially large space we are not likely to brute force our way to obtain the ***true Pareto front*** of the full feasible solution space.  At best we can calculate ***Pareto front approximations*** of subsets.\n",
    "\n",
    "\n",
    "## Niching  \n",
    "Here we generate `n_instances` subsets that we will call *instances*, each containing `n_knapsacks`, afterwhich we will calculate Pareto front approximations for each. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:50.241110Z",
     "start_time": "2020-11-11T07:22:50.029508Z"
    }
   },
   "outputs": [],
   "source": [
    "# creating `n_instances` dictionaries of `n_knapsacks` knapsacks\n",
    "\n",
    "n_instances = 4\n",
    "n_knapsacks = 1000   # CHANGE ME! The number of knapsacks (consider n_combinations in you choice)\n",
    "\n",
    "knapsacks_instances = {}\n",
    "for idx in range(n_instances):\n",
    "    knapsacks_instances[idx] = create_knapsacks(n_knapsacks, seed=idx * 100000)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:50.694020Z",
     "start_time": "2020-11-11T07:22:50.261167Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x288 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "nplots = n_instances\n",
    "ncols = 4\n",
    "nrows = nplots // ncols + np.sum(nplots % ncols != 0)\n",
    "\n",
    "plt.figure(figsize=(4 * ncols, 4 * nrows))\n",
    "\n",
    "for idx_instance in knapsacks_instances:\n",
    "    plt.subplot(nrows, ncols, idx_instance + 1)\n",
    "    \n",
    "    plt.scatter([knapsacks['weight'] for _, knapsacks in knapsacks_instances[idx_instance].items()],\n",
    "                [knapsacks['value'] for _, knapsacks in knapsacks_instances[idx_instance].items()],\n",
    "                s=10, alpha=0.7, color='purple')\n",
    "\n",
    "    plt.xlabel('weight [kg]')\n",
    "    plt.ylabel('value [£]')\n",
    "    plt.title(f'Knapsacks Instance {idx_instance}', fontsize=MEDIUM_SIZE)\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the above we see that even though the distributions in Objective Space (weight, value) trace similar trends, each has its own distinct features that will impact the Pareto Front.  \n",
    "\n",
    "Can you identify by eye the Pareto Front for each instance?    \n",
    "(Beforehand you will need define the Optimisation Objectives! E.g, minimising weight and maximising value.)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Optimisation Objective\n",
    "\n",
    "In the knapsack problem we want to optimise for two objectives:    \n",
    "* ↓ minimise the knapsack weight \n",
    "* ↑ maximise the moentary values of the packages in the knapsack\n",
    " \n",
    " \n",
    "**Note**    \n",
    "The default setting here is ↓↑, but feel free to explore the other combinations: ↓↓, ↑↓ or ↑↑. (See the ***CHANGE ME!*** for `objective_mode`).    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:51.292537Z",
     "start_time": "2020-11-11T07:22:51.289922Z"
    }
   },
   "outputs": [],
   "source": [
    "from operator import lt as less_than, gt as greater_than\n",
    "from operator import (truediv as div, mul)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:51.583001Z",
     "start_time": "2020-11-11T07:22:51.580352Z"
    }
   },
   "outputs": [],
   "source": [
    "# Objective Space declaration\n",
    "\n",
    "# CHANGE ME!\n",
    "objective_mode, heuristic, soh_unit = {'weight': 'min', 'value': 'max'}, 'value/weight', '£/kg'\n",
    "#objective_mode, heuristic, soh_unit = {'weight': 'min', 'value': 'min'}, '1/value/weight', '1/£/kg'\n",
    "#objective_mode, heuristic, soh_unit = {'weight': 'max', 'value': 'max'}, 'value*weight',  '£*kg'\n",
    "#objective_mode, heuristic, soh_unit = {'weight': 'max', 'value': 'min'}, 'weight/value', 'kg/£'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:51.833173Z",
     "start_time": "2020-11-11T07:22:51.830379Z"
    }
   },
   "outputs": [],
   "source": [
    "# These objects are used to calculate the relationships between the knapsacks.\n",
    "\n",
    "# for Single Objective Optimisation\n",
    "direction_to_multiplier = {'min': div, 'max': mul}\n",
    "\n",
    "# for Pareto Optimal selection\n",
    "mode_to_operator = {'min': less_than, 'max': greater_than}\n",
    "objective_operator = {key: mode_to_operator[objective_mode[key]] for key in objective_mode.keys()}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Pareto Front Calculation\n",
    "\n",
    " \n",
    "\n",
    "We will use the `knapsacks_to_pareto_front` function where:  \n",
    "* input: dictionary of knapsacks\n",
    "* output: list of indexes of knapsacks that are Pareto optimal\n",
    "\n",
    "This is the same algorithm used in the previous notebook for which we detailed the pseudocode. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:52.398472Z",
     "start_time": "2020-11-11T07:22:52.392339Z"
    }
   },
   "outputs": [],
   "source": [
    "def knapsacks_to_pareto_front(knapsacks):\n",
    "    objective_values = {}\n",
    "\n",
    "    for objective in objectives:\n",
    "        objective_values[objective] = [knapsacks[idx][objective] for idx in knapsacks]\n",
    "        \n",
    "    idxs_pareto = []\n",
    "    \n",
    "    idx_objects = list(knapsacks.keys())\n",
    "\n",
    "    for idx in range(len(objective_values[objective])):\n",
    "        is_pareto = True\n",
    "\n",
    "        this_weight = objective_values['weight'][idx]\n",
    "        this_value = objective_values['value'][idx]\n",
    "\n",
    "        other_weights = np.array(objective_values['weight'][:idx] + objective_values['weight'][idx + 1:])\n",
    "        other_values = np.array(objective_values['value'][:idx] + objective_values['value'][idx + 1:])\n",
    "\n",
    "        for jdx in range(len(other_weights)):\n",
    "            other_dominates = objective_operator['weight'](other_weights[jdx], this_weight) & objective_operator['value'](other_values[jdx], this_value)   \n",
    "\n",
    "            if other_dominates:\n",
    "                is_pareto = False\n",
    "                break\n",
    "\n",
    "        if is_pareto:\n",
    "            idxs_pareto.append(idx_objects[idx])\n",
    "            \n",
    "    return idxs_pareto"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:53.196504Z",
     "start_time": "2020-11-11T07:22:52.675180Z"
    }
   },
   "outputs": [],
   "source": [
    "# calculating the pareto front for each instance\n",
    "\n",
    "pareto_instances = {}\n",
    "for idx_instance in knapsacks_instances:\n",
    "    pareto_instances[idx_instance] = knapsacks_to_pareto_front(knapsacks_instances[idx_instance])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:53.628637Z",
     "start_time": "2020-11-11T07:22:53.197944Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABHQAAAEUCAYAAABKwGFLAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAAEAAElEQVR4nOz9e7gk51Ueir9fXbqqu6sv+zZ7ZjQzkq2RRoqMjUGAbTixCCRwgmV84Tk/PZh7gF9yCISQAyQBMjMH5yQcfhASIAEHiENwIhJsY2QSE0gs5YIT7GBsS5ZGGluai2Zm7917775Uddf9+/3x9VddVV3Vl727d++953ufZx5pd3dVV1XXt2qtd631LkIphYCAgICAgICAgICAgICAgIDA0YG06AMQEBAQEBAQEBAQEBAQEBAQEJgOgtAREBAQEBAQEBAQEBAQEBAQOGIQhI6AgICAgICAgICAgICAgIDAEYMgdAQEBAQEBAQEBAQEBAQEBASOGAShIyAgICAgICAgICAgICAgIHDEIAgdAQEBAQEBAQEBAQEBAQEBgSMGQejMAYSQ7yKEUELI+dTrX0EI2SGEfJoQsrqo4ztoEEKeJoT8tznt+xVCyG/Nad8/TAh51zz2fdAghDxCCPmPhBCTELJNCPkXhJDlRR+XwGwgbE4SwuYsFoSQ1xFCfpUQ8r8IIS4hhC76mARmC2FzkhA2Z7EghHwfIeTfE0JeJYRYhJBnCSE/SggpLPrYBGYDYXOSEDZnsSCEfBsh5L8TQrYIIU7/mv06IeTcIo5HEDoHBELIWwD8EYCXAPwFSmljwYckMB4/DOA4GJ3TAJ4GUATwLQB+AMDXA/goIUTYgGMKYXOOJH4Yx8DmAPhyAH8ZwHUAn1rwsQgcEITNOZL4YRwPm/P3ANwB8DcAvA3AbwP4aQAfWORBCcwXwuYcSfwwjofNWQHwnwB8L4C/BOD/6f/3vxNCKgd9MMpBf+HdCELIWwF8FMBnAPxlSml7wYckcHfhRwGoAB6nlDYBgBByC8AzAN4B4EMLOzKBuUDYHIEF419RSv8lABBC3gvgzQs+HoE5Q9gcgQXjyyilW7G/P04IIQAuE0JeSyn94qIOTGA+EDZHYJGglP7j1EvPEEKuAfgYGLHzwYM8HpGdnzMIIX8RwH8A8EkA35A2OP3ywfcSQn6IEPIyIaRDCHmGEPJI6nN/qV9OepsQ0u2Xk/4tQoic+twrhJDf6pefXiWE2ISQPyWEfG3qc19BCPnDfvtNjxDyRULIP429v9YvmX+x/303CCH/mhByT8Y5voEQ8uHYvq4QQv7OmOvyU/1S/G/r/32SEPIvCSG3+qVrtwkhHyWEnJjwUvP9Pta/pm8nhPwSIaTR//dbhJB66rN/gxDyfP+YdwkhnyKEvJNfRwD3AnhPf3+UEPL+/nvnCSH/qv978Wv3zwghS6n9v58QcpMQ8kZCyH/tX8eXCCF/NeO4X9Pf553++X+REPKPU595KyHkP/XvEYsQ8geEkNdNcFneDuD3OZkDAJTS/wKWPf/mCbYXOEIQNif3ugibk/zs3GwOpTSc5hoKHG0Im5N7XYTNSX52njZnK+PlT/b/O/R7ChxtCJuTe12EzUl+dm42Jwfb/f/6e9x+zxAVOvPFNwH4GbB2l3dSSns5n/s2AFfASkULAH4WwEcIIQ9RSvlN8Vqw0q5fBGADeBTAJQBrAP52an+PgZW8/wQAB8CPA/gPhJA3UEqvEEIMAH8A4E8AfBeADoD7ALwlto/l/vf8HQBbAE4D+FtgpWQPUUptACCEfGX//K4C+JsAbgJ4AMDrs06UsBafXwbw7WAVI3/Qf+tfgS3yHwVwA8A6gK8DUMq5ZuPwj8GY+28FcAHA/wsgAPCd/eN4D4CfA/B/A/ivYO1Ir++fNwC8E8C/B2P+L/Vf4w7D6f4x/jCAXbDf5u/2P5/ORFcB/GsAv9D/ru8G8M8IIVcopR/vH8trwH6LLljZ8EsAzoExvOh/5psAfATA74PdLwD7Xf8rIeT1lNIbWReBEFIE8BoAv5bx9nMA/lzWdgJHFsLmpCBszsHaHIG7DsLmpCBszqGxOW8FEAJ4ccrtBA43hM1JQdicxdkcwsg/pX89fh7A58Hug4MFpVT8m/E/sIVM+/+uAtBGfJaC3WRq7LVv6b/+lpxtCNjN8xNgN70Ue+8VAC6As7HXKgB2wMrgAWawKIDXT3FOMoCz/e3eGXv9v4AtwNKIbZ8G8N8A6GAlaFsAviL1GRPAD+3hWr8C4Ldifz/WP8Z/mfrcL4EZURL7+0+n2feIzykAvqb/vW+Mvf7+/mtfG3tNA2Nw3xd77Tf75396xHdcBfCfUq9VATQA/MKI7U73j+GvZrz3WwC+sMi1Iv7N5p+wOUPbCpszeO1AbU7Gft4LgC5qbYh/8/knbM7QtsLmDF5bqM3pb/d6AD0A/3xRa0T8m+0/YXOGthU2Z/DawmxO/7P8vvwkgFOLWB+i5Wq++H0A94MxsaPwh5RSL/b35/r/jZSyCSGn+mV618CMigfmKNcBpEvn/geNsYqU0k7/WDjD+RKAJoBfJUyl+2zWQRFC/hoh5DOEEBOsfOx6/60L/fdLAL4awAcopd0x51gBYyy/DMBXU0o/mXr/kwB+tF+q9yWEEDJmf+Pw+6m/Pwe24Ndj3/elhJBfJIR8ff9cJgIhpEAI+buEkBcIIT2w3+K/9t++kPp4l/bZYgCglDpg2aK4CvpfAvBRSumtnO97AOw++gAhROH/wFjnTwD485Meu8Cxh7A5AwibA2FzBOYOYXMGEDYHi7c5hJBTYFn3LwD4kUm3EzgyEDZnAGFzsHCb83VgVVh/Bey++UOSakM7CAhCZ774mwB+HcBFQsiPj/jcTupvp/9fHYhK6X4PTLn/vQD+AoCvAPD345+LYSPjOzbQ7yOmlLYAfC2AWwD+KYDrhPWNvpt/mBDyg/33/ghMjfwrAbwp9X1LYPfQzRHnxnEOzED9B0ppVvnr/6d/jj8G4LMAXiWE/D2y9ylMI68pGHP71wB8FZgx3CGEfIgQct8E+/4HYKWCvwVW+vmVGCi2p3+L3YztndTnVjD6GvKHyq+DGbj4v7f1t89DE4w1Xsp4bxnD10ngaEPYnAGEzUkey0HZHIG7C8LmDCBsTvJYDtzmEEJWAPwhWLXFN/SDboHjBWFzBhA2J3ksB25zKKWfoZR+glL6GwC+AUzKYkjPZ94QGjrzBQXw/WA32D8khDiU0l/Yw37uByvl+3ZK6W/xFwkhj+d8fj3ntVejA6P0zwC8u89GPgrGdP9bwnpBnwXwBFgZ2t+Kfd9rUvvcBetPnkRw7jmw/s5/RQjpxffbP55NsHHaP0AIuQDWj3kZrITwn02w/6lAWZ3cr4Ix6UtgTO7PgY26/Koxmz8B4Dcppe/lLxDWO7tXNDD6GnKRrb8D9hBIw83bkFLaJUyE7JGMt/8c2KQrgeMDYXMGEDYnH3OzOQJ3HYTNGUDYnHzM3eYQQqpgQeQKgP+NUvrqmE0EjiaEzRlA2Jx8HLifQyn9IiFkB8D5abfdL0SFzpxB2bSP7wTw7wD8I5Khwj0BeMlaVDpICFEBvCfn82+Kl/oRQipgbOcnMo7Pp5T+DwA/BXY/PBz7Ti/18e9ObdsF69/8NsLEd0eCUvpvwIS0fogQ8o9GfO4KpfTvghm1vSqNTwxK6S6l9LcB/NvU9zlggl5pjL02U+I/Anhbv0w4C1fAek4foZR+KuPfZ8fs//cAfBMhpMZfIIR8DZhQ2u/t47gFDiGEzUl8XticbMzb5gjcRRA2J/F5YXOyMVeb02/t+H2wIRB/iVJ6dR/HKnDIIWxO4vPC5mTjwP0cwqaorYC1ex4oRIXOAYBSGhBCvhVMZf2f9tnkfzHFLp4HcA3A3yeEBGA3/N8c8fkNAP+REHIJAyX2MoCfBgBCyNvA2O3fBfBy/70fAlNk54bpYwB+nBDyd8FUwv8CmKBYGv8XWJXHJwghPwdW3vZaAF9KKf3B9Icppf+2fw7/hhAiU0p/qE80/BGADwB4oX9+3wxWdvgfx12cvYAQ8j4MzncTwINg6vDx7/s8gP+tf73uAGhQSl8BuzbfSQj5HJig1ruQVLGfFhcB/GUAf0wI+X/6+7wHwDdSSr+NUkoJIT8Aps5fADOODbDMwFsAXKeU/vyI/f8smHr77xFC/gGAGpgy/f8E8OF9HLfAIYWwOQMIm5OJudqcfnD1l/t/PtR/jf+Wr1BKP7WPYxc4hBA2ZwBhczIxbz/ng2CtJ38DQJkQ8qbYe1+g2WPNBY4whM0ZQNicTMzbz/lvYDHUC2DC0K8Hm1h2E8A/38dx7w10AUrMx/0fBkrs51OvF8DGrwUAvrX/GgXw3tTn7uu//l2x174UjLHtgt0s/zeA7+1/7r7Y514B6z/8XjCG0AHwaQB/IfaZC2Dlby+D3YRb/eP6qthnimDleFtgi/OjYJkPCuBS6njfCOApML2WHtjN/eOx958G8N9S27yjf2y/DCao9atgpYMmgDaYsNa3TnCtX0G2EvvX5/wm9/X//s7+cW32j+NlAP8IQDW2zUNgglzd/rbv77++CuBJMJZ7F8xYfkXGb/Z+ADczjvlpAE+nXrsfwL8BMyZ2/7f7+dRn3tz/HXb7n3mlfxxvnuA6fQlYX7nV3/79AFYWvVbEv9n8g7A5wubQw2NzYvdT1r/3L3q9iH/7/wdhc4TNoYfK5uTZm8Txin9H9x+EzRE2hx4qm/NzYKLQ7f71/TxYAv3EItYHHzMmcExAmF7Kf6OUftuij0VAQOD4Q9gcAQGBg4SwOQICAgcJYXMEDjuEho6AgICAgICAgICAgICAgIDAEYMgdAQEBAQEBAQEBAQEBAQEBASOGETLlYCAgICAgICAgICAgICAgMARg6jQERAQEBAQEBCYAQghf50Q8ilCiEMIeX/qva8jhLxACOkSQj5OCLk39p5GCPkNQkibEHKHEPIjk24rICAgICAgcPdCEDoCAgLHFiK4EhAQOGDcAvBeAL8Rf5EQsgrgQwB+CsAygE+BTUThuATgAQD3AvhaAD9GCPnGCbcVEBAQEBAQuEtxV7dcra6u0vvuu2/RhyEgcCzxv/7X/2pQStcWeQyEkHcBCAF8A4AipfS7+q+vgo0v/F6wsZA/DeB/o5S+qf/+PwDwNQDeDuAkgI+DjU382LhtR0HYHAGB+eEw2BwOQsh7AZyJ2ZzvB7Mhb+n/XQYbpfpGSukLhJBb/ff/Y//9nwbwAKX0iXHbjjoOYXMEBOaHw2RzDguEzREQmA9G2RvloA/mMOG+++7Dpz71qUUfhoDAsQQh5Nqij4FS+qH+sTwK4EzsrXcBeI5S+u/6718C0CCEPNQPkL4TLIDaBbBLCPnnAL4LwMcm2DYXwuYICMwPh8HmjMAjAD7D/6CUWoSQLwB4hBCyAeBU/P3+/79j3LYAhmxOnwD6fgA4d+6csDkCAnPCIbc5C4HwcwQE5oNR9ka0XAkICNyNGAqQwKpuHiGELCE7uHpk3LZzPmYBAYGjCwNAK/VaC0Cl/x5S7/P3xm07BErp+yilj1JKH11bE8UDAgICAgICxxmC0BGYCpefvownn30y870nn30Sl5++fMBHJCCwJxxYcEUI+f6+js+ntra29nXQxxHCpgjcJTABVFOvVQF0+u8h9T5/b9y2M4dYkwICAnc7hB0UOEq4q1uuBKbDpacv4fIzlyERxgM+8bonoveefPZJvOdD70FIQ1BQXHrs0oKOUkBgIkwaXNmp98ZtOwRK6fsAvA8AHn300btXtCwDwqYI3EV4DqyVE0Ckg3M/WPvmLiHkNoA3APjD/kfe0N9m5LazPkixJgUEBO52CDsocNQgKnQEJsLlpy/j8jOMjQ5piPd86D0Rcx03bgBw+ZnLgrkWOOx4DixgAjAcXAG4HX8fw8FV5rZzPuZjBWFTBI4jCCEKIUQHIAOQCSE6IUQB8GEAryOEvLv//t8D8NmY7tZvAvhJQsgSIeQhAN8H4P3998ZtOxOINSkgIHC3Q9hBgaMIQegITIQLqxciphoYGLl3PPmOhHEDAIlIuLB6YRGHKSCQwFEOro47hE0ROKb4SQA9AH8bwLf1//8nKaVbAN4N4O8D2AXwVQCeiG13EUyL6xqAZwD8LKX0YwAwwbYzgViTAgICdzuEHRQ4ijhwQocQ8tf7ehIOIeT9qfe+jhDyAiGkSwj5OCHk3th7GiHkNwghbULIHULIj0y6rcD+8cTrnsAH3vWBISP3kSsfGTJuH3jXBxLliQICC8SRDa6OO4RNETiOoJReopSS1L9L/ff+iFL6EKW0SCl9jFL6Smw7h1L6PZTSKqV0nVL686n95m47K4g1ORv4tg+7ZcO3/UUfioCAwJQQdlDgKGIRFTq3ALwXwG/EXySErAL4EICfArAM4FMAfjv2kUsAHgBwL4CvBfBjhJBvnHBbgRkgy8jFIYybwGHDUQ6u7gYImyIgcLgg1uT+4Ns+mteasDYtNK81BakjMHeIRPnsIeygwFHDgRM6lNIPUUp/F8B26q13gelX/DtKqQ1G4Lyh3+4AMEHAn6aU7lJKnwfwzwF814TbCswIT7zuCTz+4OOZ7z3+4OPCuAkICEwFYVMEBA4XxJrcO3zHB5EICuUCiETgO4LQEZg7RKJ8DhB2UOAo4TBp6DwC4DP8D0qpBdby8AghZAnAqfj7/f9/ZNy26S8RI4T3hyeffRJPvfhU5ntPvfhU7og/AQEBgSwImyIgcLhw1NbkLFqcZtUmpWgKaEjhWi5oSKFoYpiswHwhEuXzwVGzgwJ3Nw4ToWMAaKVeawGo9N9D6n3+3rhtE6CUvo9S+iil9NG1tbV9H/TdhLS6exppNXgBAQGBURA2RUDgcOGorclZtDjNsk1K0RXU762jfKKM+r11ABB6OgKLwoEkyo8jjpodFBA4TISOCaCaeq0KoNN/D6n3+XvjthWYAbKMm0QkfPOFb85UgxdGTkBAYBSETREQOFw4imtyFi1Os26TUnQFek0HAKGnI7BIHEiiHDhe3Q9H0Q4KCBwmQuc5AG/gfxBCygDuByv52wVwO/5+//+fG7ftnI/5rsGVxpVMdffffeJ3M9XgrzSuLOIwBQQEjggOm00Rk2kE7nYctjU5CWbR4jSvNimhpyOwYBxYovw4dT8cRTsoILCIseUKIUQHIAOQCSE6IUQB8GEAryOEvLv//t8D8FlK6Qv9TX8TwE8SQpb6PZzfB+D9/ffGbSuwT1x87CIuvvUigGF197Qa/MW3XsTFxy4u7FgFBAQOPw6TTRGTaQQEpluTP/XVP4Uff+OPT7RW5kmWplucFH0PhM4M9pG5X6GnI7BYiET5HnCYfJN5QCSvjicW8XT5SQDxu//bAFymlF4ihLwbwC8B+C0A/xNAXEL8IoB/BuAagB6An6GUfgwAKKVbY7YVmAEuPXYJBAQXVi8Mqbvzv680rhw54yYgILAYHBabEs+ku5YL3/FnFtQJCBwlTLImn7/zPH7w7A/C2rRAQzqSBOFkKZHI2M/uFYqu7Hufs9hH1j7r99aZPdFmv/9Fwbf9Y3dORxn9pLiCWKIcgA+W7P7Zfnz0+8hPlH8KwDpYovy7+++N2/ZY47D4JrPGQdhjgcWAUEoXfQwLw6OPPko/9alPLfowBASOJQgh/4tS+uiij+Mw4bDbnMtPX850YADWV57lwOxlm8MG4eQcDwibM4x52By7ZcPatCICtHyiHGnG7OezAocfwlYmcRhsDiHkEpKJcmCQKP96sGT3vWDJ7u+ilL7S304DS5R/CwaJ8p+P7Td321E47H7OJDgOfk0WhD0+2hhlb+5eKywgICAgEOHS05dw+ZnLUSlx3JGJiwRSUFx67NKetzmMOK6ZdAGBeWCaViLRdnS8IKoZDx8opZfAxopnvfdHADJHjVNKHQDf0/831bbHGcfFr8mCsMfHF+KXFBCYAKLEWOAoY1y26Vc/9at4+trTAAaTGwDmyKQnPlx+5jIISPT/02wzy4zWpGty0s/No+VCQOA4YhoCVJClxwsiIBQ4Spi20uby05cX6tfMG8IeH1+IX1JAYAxEibHAUcak2aY44uM4n3rxqaGJDxdWL0T/z9+bdJtZYNI1KdaugMB8MA0BKsjS4wMREAocFeyl0ubC6oWF+TUHBWGPjyfELyogMAaixFjgqGKabBMAEBBQ0OjzH7nykcT+0hMfACT2Mek2+8Wka1KsXQEBAYHZQgSEAocde6204X7KQfs1ogtAYL848LHlAgJHDaLEWOCogmebOLhj844n3zFE5khEwl//yr+e+HwcWQ5MeoTnJNvMApOuyb2sXTHSU0BAQEBA4OhiWt8nXmlz0H4NryS2Ni00rzWF7yGwJwhCR0BgDHiJcflEWbRsCBwpZDkmPNuUdmg+8K4P4J/87/8Ejz/4eOa+Hn/w8UwH5onXPTH1NvvFpGty2rUrHCsBAYH9QBDCAgKLx7S+T9Zo8oPya+KVxEQi8J2jYzuEvTs8EISOgMAEUHQFek0XZI7AkcM02SbeJ56Fp158Ck8+++TQ63vZZhaYdE1Os3aPsmMlIHAccZQCBkEICwgcHkzj+6TtzEH6NUe1C0DYu8MFQegICAgIHHNMkm3KE0jmiIsDcuxlmzgOW7C2H8fqsJ2LgMBRx1ELGAQhLCBwuDCJ75O2Mx/40w/sy6+ZFovqAtivzyLs3eGCIHQEBAQEjjnGZZt+6D/8UGZf+dsfeHtmH/qTzz6ZSeZIRMI3X/jm3G3isJs2Np7dQOtG69AEa5M4VllO0FELPAUEDhp7CR4OS8Aw6bEf1Uz7pBCktcBRwySVNnE78+HrH8Z3fPQ7pvKFZoGD7gKYhc9y3O3dUYMgdAQEBASOMSapovmlP/mlIQfmV978K/j1L/91/ObbfnPIkbnSuIIrjSvD27zpV/A77/idzN71K40r0d++7WP76jZ6Oz3YOzYCL5g4WJt3UDHKscpzgg5L4CkgcBix1+BhkoBh3vZgmmM/znp7grQWOGqYtIL4g1/4YGRnrravZmrs/PY3/TZ+5c2/MuTXPH/n+SNJcs7CZznO9u4oQhA6AkceImskIJCNSatoKCgISPT+r7z5V/DEI0+ASATvvv/dCYLm4lsv4uJj/X9vvRht82tf+2t412veBd/xh3rX+TYcfDynqqvwbA++7U88gWqRQUWeEyQyVQIC+dhr8DAuYDgIezDtsR9XvT1BWgscJUxTQfwdH/0O/IH1ByifKOPvf9PfT/g1XGNH0RS889w7E6TOT331T+EHz/7gcILnCMQks/JZjqu9O4oQv8AB4fLTl3Fh9UKmMvqTzz6JK40riYBHYDJwh45IBDSkgiUWEIghq4omLoAcd3goKB679zF87+u/F99Q/obEg57brbSduvTYJVCf4kx4Bm8/+faEY5C3DcCcCbkgQ1/SoTgKVs6vTC1a7FouI4YOcL3nOUE88ORE1SyPybf9uexXQOCgsJ/gQdHZfc+DpPg6OAh7cNjJ2oOyD4f9OggIxDGN7xPSEFfbV6HXdADMryEgiZiNP+O//eS3o7RSwtX2Vfz4G38c1qaVsD8AjkRMMk+fRWAxEL/gAeDS05dw+ZnLEasbJ3XihoWC4tJjlxZ0lIcT3FkBBUAwZHgWHeAJCBxmXHzsIihoZH/i4zn5f7n9ufjWi5H9yQoS8sZ0Xv76y7lBRd42e3UmFh1UZB13/Ny5QzgrCMJa4Dhgv8FD3jqYpT3Is2HzCHxmRcIcpH0QAaDAUcK0vk866ZSVYOfk8nu+7D0A2PpL25/DEpNMYmP4+QgcD4hfcs64/PRlXH7mMoBBvyaATJb48jOXQUDumkqdcQaHOyuBF6B9o43qmSrkgpxwWuIOnddjrRu+JkgdAQGOrGwTR14VzbQP+r04BnvdZlRQcRDZ6vhxzzugOizOoYDAfrGf4CFvHUxDMoyyDePW8SwDn1najIO2DyIAFDhK2IvvMw3y7M+iK9lEIujuhPiF54wLqxcgESlR2seV0Z968amhksALqxcWdaj7wrSB1CQGhzsrkiKByASSKkW92xGh0zeodsuG1/PgtB3YTVsYMAGBGEY5LHlVNPvBPImVvKBiEU5MPKDq7nRhbpowThgz+95FVyQJCBwGjFoHk5AM42xDmhixWzYUR5mL/ZolCSPsg4DAaMzb90nbn8NQyZZnY0T79vGGEEWeM9LioAAjdT5y5SO5/Z1HDXsRJpxEYI87K6EfggYUoRdmOi3coKpFNXN/R0GgTEDguCB3EtSYdbjfdZpnU+a5/rmN6u500b7RhtNyZirOKqZIHD8QQh4mhPxnQkiLEHKVEPLO2HtfRwh5gRDSJYR8nBByb+w9jRDyG4SQNiHkDiHkRxZzBgePSddB3lof52+kK33NDXNuQsuzJGGEfRAQEEgjy8YseqCEwPwhCJ0DQBapE8dBkTnzCmz2Mv1gEqeGOyu1szWcffNZ1M7Vcp2WvP0JIyYwCiK4ykbaVkxjO7Lswbh1aDdtbDy7gdaN1p7XaZ4T03ixgZ0v7KDxYmP2wVnfRmlVDdUzVZRWSjOfACOmSBwfEEIUAB8B8FEAywC+H8BvEUIeJISsAvgQgJ/qv/cpAL8d2/wSgAcA3AvgawH8GCHkGw/u6BeLcetglI0Z52/EiRFj3chNDs3qPGZJwgj7IHDUcfnpy3jy2Scz33vy2Sdx+enLB3xEe8dhiDmybIyYUnf8IZ4ABwSumfORKx8Zeu/xBx8/EDJnXu0Ie8k4TVqWOGnPdt7+hAaFQB5iwdWvAPiLAN4K4ClCyBsB7IAFV98L4CkAPw0WXL2pv/klDIKrkwA+Tgj5PKX0Ywd5DvNA2lYY6wbMDXNi25FJrIxYh77tY/vqNno7Pai6Cn1Z39M6zbIB5oaJzu0OtLIGx3JgrBswdGNf1yfre40TBvyeL1ofBMbhIQCnAfwjSikF8J8JIf8dwLcDuAHgOUrpvwMAQsglAA1CyEOU0hcAfCeA76KU7gLYJYT8cwDfBeDI25xZYJSNmcTfiKZpaT7spr3ntXxQYqSifULgKINP/n2h8cLQ0Bg++ffC6oUjNzTmsMQcQ61g+6wMFPbm8EP8KgcErpmThadefApPPvvkXEmdeRqZPGfJbtqw2zb0qg69Pjz9ZdYCe1n7UzSFaet0HEiKtOdASxizYwkRXGVgSE+ibU9lO6YVCuSfU3UVnu1BsZU9r9MsG0AoAQUFoWRP+5z0exfdNy9wZEEAvA5AFcBn+IuUUosQ8gUAjxBCNgCcir/f//93ZO6QkO8Hq/7BuXPn5nPUhwyTVOHsJzk0CQ5Kx2va7xH+i8BhAp/8S8CezcBAX/SPb/wxfvmTv4yQhon3j8rQmHnrWu11LR+UXRO2ZnEQV/sAkJ5mlUZ6+lUW0otk2kUzbyOTdpbspo0bn7gBIhPQgOLsm89mkjoHBrr3TYVi/F2Fuz64StsKvarD3DCnrsDLEwoERVTuq+iMvJELMvQlHYqjYOX8yszWl17TYZw2EHohtKo287HicYwSaxYOjkAfVwBsAvhRQsg/AmudeiuAjwMwAGylPt8CUOm/x/9OvzcESun7ALwPAB599NF9PP32h4O892dJqo4jf/LO66Cy89N8j/BfBA4T4pN/acoxD2mIX/yTX4z+jr9/VIbG7NcO7Wca3yTHtpe1P6m9EbZmsRBXes7IInMkIuHxBx9PTLkaRerstwUCOPgMst22QWSC8moZVsNilToLIHR8x2f98Kt7d7AOSwmlwMxxrIIrXsKcRQrzEuZJsltZtoL3YO/HdvDtsh7487JNiq5g9cHVqSfwzWos+qwcHEEKHQ9QSj1CyDsA/CKAHwfTyfm3ABwAJhiRHEcVQKf/Hv/bTr13KLGIe3/WVb95x5N3Xgc1dWqa7xH+i8BhQnry7yQ4akNj9kycTDmN76DW8qT2RtiaxUJc6TnjSuNK7jSrNNkT0hBXGleG9rHfFoi4QzTPDHUcelUHDSishgUasCz/IjALB0uMBj2eOE7BFS9hjvehc8TtzKR96FkVNrN4MOc98CfZ/35KjSf9/CiHai8B6iwcHJH1Ol6glH4WjDgGABBC/hjAvwSrI/3O2OtlAPeDtX7uEkJuA3gDgD/sf+QNAJ47qOOeFsf13ufnJckSeu0e7JYd6XIdVOJs3PckbOU05I8gjgXmDO6bjOpaiOOokTn7wTibuahYZFK7JnR6FotDdcUIIQ8D+GUAXw6WHf9RSumH++99Xf+9cwD+J5h+xbX+exqAfwbgWwB0Afy/lNKfP/gzGMbFxy6CgkbBVtwwpQ3bxbdezMygp0dqygUZXs8DgIke0ItwiPS6jrNvPjtSQ4cf3zwX8CwcLKGPcXxx1IOry09fxuc2P4cPPv9BAMlKPwD4nc//Dj78wocjx2mRfei+zaZdpW1X3AYAyFxneyJT+vsFBUAw0dodKd48ZYCad77TQmS9jhcIIa8H8CLYlNH/E6x98/1gpPDPEkLeDeD3Afw9AJ/ta3YBwG8C+ElCyKcArAP4PgDfPY9jnEW13yyCj8N47yuaArtpo3mjCUVVIKlSYsrUfsnvSX2iUS2eaVs5if9yGMkzgdngsMVW05A6BzE05rBg0ml8+4lF5pkYOwr6Y8cZh+ZqHeeJM5ceuwQCkukg8b9HOUh8kdgtG17PQ+AGABBpQowMKhboEOl1RuTwkcezCNL2gllUFxxEKbfAweMoBFd5iAsLpsUFv/WD3zrUnw4spg+dr39zw4RaVAEgoWfDbQAnPtSiOlwdk65SbNlQHCXXaeC2JXADtG+2UT1bhazKe5rQxfc3DTkTt22+40MpKjDWjb3ZEIpoZLxckEWF4NHHt4P5MiqA/wrgL1JKHQBbfXvzSwB+Cyy4ijsMF8GCq2sAegB+Zh4+zqyq/WaSTJljRnqawGaIdHZ9EBAoujJY47OoYNynT+TbPsxNE4EboLRSiny+ScaaH0byTGD/OKyx1ajJv3EcxNCYPIyzEft9P41JbOZ+YpGDiLmmrYjm5yrsz/5xmK7WsZ44MyqbNYmhUnQFiqMwPZhyYfDaOMZ0we1CI1sY9rmARXmewAxwqIOrPKSFBdOkTh6Zc9Cly3z9u5aL7mYXKw+uRAGQb/uw2zYCL0BpuQSn4wAUmXpX6SpFr+dlEj/8O3lQI6kSiEwgKdJEgVeWQxW3YcB0RLokS+jt9CBJEszQnNoZ820f5oYZHUftbE3YuiMOSumPAvjRnPf+CMwXynrPAfA9/X9zQdyupHX90i3ik1T77TcRMq/q2GmntjRebEQ+VO1MjbWQB4Bne/BtP0n87icAnEJ8NL2fiMT2GIkNYCoCeGKdDOF3HTUcythq1OTfOCYZGjMPjNWz2ef7eZhn8th3fAReAEmREPrhQkmTLG1YIW2xPxz2KzbziTNHGXshZxbdLjTKQdkP2XRUyvOE83O4cZiDq1FICwtyUicPBGQhfeh8/RfrRVhbFjp3OnBaDsprZWw9t4XSWgndrS4AQFJYRUCWPYjbMd/24bSd7LaoVFBTXiuDBszGgALltfLYY047VHEblvV+5j76tq3X6YFQguJSEWEw3oEammbY/+7SMsu2p39iYV8EZom0XeHBFA++0nqAB1HtN48AZ5pkkt2yGSndceHZHgInQGm1NDSVbxYB3iQ+Ud5+4rYCAGRNnkq7cBJf8aj4XQJjsdBpnuMm/6axCFJnnI3Y7/tzPfacNnZQoH2jHU0frp2pHcjxZB5j6vqAQEhb7BOH6YodyMSZozZCOI69kjPxipj433HsRW9iou/WFHg9D07HgaRIQ5ksY93Y0/f5jh9l4ENvsUxzHoTzIzAvZPWgZ1XlcLz9wtvn7ghlkQs8QAkRorhcZJPvCmWohgqyQ1AwClA0BVpVg3GCmfI8+8YDO1/zYTftKOjhLUnxsl0e1GhVDdUzVbRutqBoSlTtwgOwibQq9kGk2y2bZcOCcGq9M2PdgO/kt3kJ+yIwa2TZlZCGQ20RR12odNo1HbohQABVU6EW1ah9Ml7Fl9XqNCrA422jaZ9rLKmSJyyfOie/58OBA7tpz6w6QLRFHEkcqmme05I5HHlDY+aFsXo2qffjfkjWejyoipO4X5BuY9frOqpnqlHcNCIHOHdkXR8hbbE/HJord1ATZw5qhPC8sJcbPitQiDsQe9WbiO9/bGBEk5/fdyBCgfbNGNN8Nsk0H4bMtXB+BOaJaYQF59WHHieCzQ1zaE3HiQ2u/7X7yi6qp6ugAUXoh5ALMowTA32ZSSpYONnh2z6a15sghLVT1c7UIidBVtl+uYZEfB0C2ePTs7AfIt3QDeg1faJt4/aiu9PF9tXtSGcoq81L2BeBeWCcXTkKZM645/80a1qv6aieq6J9vQ2pIEEtq4m1OGmrU1bbKJEI2jfaqJ6pQi7ICbuZh7xAcdIqxv1g0S38AtPjsE3zTE/+5ZCIhB/4ih/AL3/ylxPv81byvKEx88I4GxF/P8//WUTFSdwv4G3skiGh1+lB0RXIBRlEIiCELHT9Lrp75DjiUF3Boz5x5jAgs7c6J1DghifSe5hSb4J/38gyY8dnuj+ryWBq34EIAapnq1EvaJxpHjd6+KAMiHB+BOaNSYUF51GyHF9nPDOVlZ1WdKb/RQiB32W2xtw2cepLTrG2hQnXYlbmKfRDtF9tRxkoAFg5vzJU9Zdeh5OQIWlbsVd7Mem2cXvBdTlGtXkJ+yIwL4yyK4dl6kzeszzr+Q8MV/5NvC51BSe/5CTq5+psXaZ8Cd8ergqME9Tx/aQJlzAImc+lTu5zjQqE8qoYZ2UbRBB2NHGYYqusyb9XGleioTFvOfuWxOTfh1Yfmmiq3jwwllztv2+37OyquQm6I2Z+zDG/QFIk+I6PxksNEMpiu/q5+siuiFnGSJMQ68KGzA6H6koe5YkzhwF5REZeoMDLfgGwbBEhUdZ8UvZ2XGCUF3TsNxBRNAWymmSa0w5W+pgOukVBOD8C88aTzz6J37vye7nvp6dfzZLUia99vu5GlSf7DhNB9roe9LoOq2HBODn55KdE5qntAAQo1otovtKE4zsor5QjLYl0kBPPpPH/jrJBk+hhzHpdKzqbhGW3bdTO1GC3Rgdkwr4IzAujBEsXOXWGY5phC3bLht209/XcV3QFOnQ0m6yS2dxgxQtqUY3IZQBRVeCoar844RL6IfO5vMl9rvh+Rr2fJe4+C1shgrCjh8MWW42a/Pst578F7je6+EL7CyOn6E2DeSdyI2mJtgNJTUpLHHRbdNyPKK+V4Ts+JEmKdPxAEFX+pjHL4xUt4QePw3Z1j+TEmcOC3N7q2MO9vFaGuWEOTYsBgPKJMmtdmkLTZmyfaU7QMSoQmcT4pvcLDI8/BpK6E4toUchzfg5DS5jA0QbvRR+lnQMMkzqz6kOPr31ZlSMSwnf9RCabo7xWRne7C72uQ6/qmeTLpN8nqUxAOQxCVM9WEbhBVHXIx6OnW78AjGw9jWOUrZiXo8InWhGJwO9Npi8mgiuBWWOcxsWips7EMc2wBWDyiuBRz+X4d1pbFgIvgGZoUIsqtKoWJc8mIU/i/kvtzHQ+16SI2wa7aWP76jZruZiinV7g2ODQxVZZFTf82fpNS98EWqOZfsS0OFBigQy+c1RyeZJj3kt84Ns+7JYd+UDmhglj3UDBKEym4zfDGGke8ZaIm0bjUF2Rozpx5rBgFLkSf7jzQCreZ81fz2Nuc79zgixxZrvACKJjGm2LyGGJlTwCGHKwxl2fg4RgrgX2i2/5t9+CDz3/oUwyJz2+PD7SPN2Hvp8HZLryxW7ZuPOZO1B0BbtXd3H2zWeh1/XE/V5aLsF3maaNrA7rTIw6niwS13f8SD9rnHZEovV0uwu7bUOv6pnl0KNsxbyI4aypD9PaYwGB/SCLzJGIhMcffDwx5WrRpE7e+swatgBgovajcc9l/p3dnS66jS5AgMZLDVROVVA7W5ua8I3bpay2sVkNqfBtH9tXt9Hb6UHVVejLutDbustwVGKruZAAY5IzsyAIImmJvm/BZS3ykssj95VqLTfWjSH9vFHbuaYLq2Fh9YFVhAinmh41yxhp1vGWiJvGQ1yNQ4JZGBbf9gGJCfIZ60ygPq66zpEu+93vgttvljh+7ns16JGztd2NKpHSwdBhaVEQYqYC+8Glpy/hg89/MPGaRCS886F34nTlNH7pT34p8R4Fxbsffje+5MSXDJE5+31AxitfOrc7sBoWTjx8Ak7HYYRJXU/c70A22TrJ8WRO0UrZtVE2LW4j2jfbKLklbD23NSRIyveVq1MxJ2J4FvsVGSyB/SAtWBoXQE6TPXut9pvFPZrXUpRnPyZ57o9tH+9/p7lpYuk1SyiUC+jt9qKJV5Puh1+DeCY9MX48LrKcIZg8Lfh5q7oKz/ag2IrQ2xI4lJjHs3UU+TsrgiAha9Ff7+P8nTxEmqayBPOWidAPJ5pUx7crLhXR3eqi1+yhUC5MpQE4yxhp1vGWiJvGQ1yNQ4BZGBa7aePGJ25EU58UbSDUlbfPdIZ9XsJdI8uYMyZw7cWgKzrrG+VlxXw8MZDMgMVf49vNGmPLrQ9JpZDA0cPlpy/j8jOXE68REHzgXR8AgMwWrItvvZjZiz7NA3KUACkf12usG9h5aQedjQ5UTYVeZYRq+n7PyzZlHQ9/PW+KRBrcpnFtsKz3zE0TVcIE1XtyL1eQNM8J2o+jMk0F0rS2SWSwBPaLLMFSXoGTnn61l6kzMw2i4s9zboe8AKXlpDD7pATSJM9lRVfY5LyejzAIUTAKw4mjMfvJy6RHx8sDOkWaWjA577zkgswE6B0FK+dXhF0QOJSYR9I1b5+zJAji3xGXtRjl7+Tuq28/eu0eAi8AkQkCLxh7fHy7ECGM05NV9eSdy6zsw0z3JeKmsRBXZEJcfvpypoAXwMqU96PCPgvDYm6Z8GwPlZMVOB0HZsOEqqlj9xnPsKedrFlVDY2bgpVuMxhn0HOPq9+iMEoMMe9cZ4VJHNbDUikkcPRwYfUCJCIN6VtwEdP0uM93PfyuXGHBSR+Qefd0elxv9UwV61+6Dr2uw1gzoNf7hM6E9/uQ9gUdrNX4FK3udhfmpjlSfJSv+3RmKx6QBV4wsSBp1sSreRAu+3GARAZLYBYYJVjK/96rv7NXEplvm6VRAwCNFxtwTRedWx1UTldQMFhm2rd9NF5sIPRDSIqE1QdXR1b9pVs68yqcJx1nPEqLJyuTDsQCsz0KJqfBk11220atWovaYIX/IXAYMUsSYNQ+Z00QxL+DE7N5rZST6IOaGybaN9uwNizQgDKtrTHfnx6jHrhBwgc6yutexE3jIa7IBLj09KUoYwUke8bjZcgUdE+q7Ps1LL7N2gx62z3YOzZKayUYq8bYKSnR9jmZ8VkQH6Oy7oqmZJ77KIMeBZEuY6xXzq8MAscJxBCzXpspoTOhwzqPh5bA8Uc6Sw6wlqr0eOF0dj0LkwYwefc0fz0+rnflgezs7yT3e/p44t/L9XF4u1SVVOH3/OR0mwmFCOPfM4kg6ayqCuZNuEzzHDnKjp3A/DGKrNmPZs5eSGSuQ8Fbk4x1I1Gtp+gKOrc7kBUZVsOCVtVQMFi7g92y0bndgVbW4FgOjHUDht5vR88RCo6T1Xk6Fnn2LL6usvSvfJvZJz6MwjhtZO57Wvs0LgHGrxefzjVJteM0EPZE4KghiyCY5j4eVbWc9/o0+qB6XcfSa5cgqRJCL4wEl8edk6IrUYVQsV6Ea7swN03oVf3A1/2s7YKIm0ZDXJkxiLc4pIUA0z3ll5+5DAIydeZq36X2jo9ivYhzX32O9XfftwTjpBHpV4wzOFlO1n6Cj8T+c7LugRfAtxkhM825+47PWOddG57tYfvqNtZftx4t9KEANUNPY55le6IsUGDeyCJ14piEzOHICmDSD/u8ezr++rhxvZMi/cAemqLVtlFySpAUKSJ1swKw+PZ5rROTHisnxov1YqI1Yupzm7NtmPQ5IlqzBGaFaR32ie/RmP/hdByAAoXVfuVt2076Jq4PQllbglxgdkgtqoMEDmWC8IQOIqJxQsF70bGYRAOMJ6OcjgO9pifElNPXiR/HNGs5azJe2pcbun77JJaFPRGYFOM6HZ6/8zz+zlf8nZm2Mo9CfI3Eic9x93HeuhvVGj5tTMVbJYlEpqogBtgxdDe76LzaAQ0plu5fgrVpRVXOsxCHntTeCbtwcBBXdwzSLQ6c1MlqcZCIhAurF/b0PekAY5oFxgMFSZFQOVmJBJGzgpa8RZblZO0l+MjafzrrHngB7J0kIZPOZuWdP9+HZ3tQdXVQ2piTOZt2ZPp+IcoCBQ4CnFBOV+YAwNvOv23qLPo4Z4NXweVlkue9lqJpLwWFtSns9ODbPhMcTBHQwPRChGnEp8xw58jaslA5VZlJ6wMf2z5rTEJUidYsgVlgrw77RNV6MfIzDEKEbojudpdpwfQzza7lwut50Gs6iitFhH4Ix3Tg9lxAQpRQMk4bCL1w0LoV8y3iQsF8Uh/fzut56O0yHYtivYgwGE3mjlpXXOPH6Tjwu8x/sbasyFdLHxcnnHhQl3dt48RTZ6uDbqOL8oly4vcY0jCLXb+ZTJ8R9kRgAkza6dBr9fBjr/uxTHIyC/shDuLb2i0biq4MaXBlbhe757s7g6lWo/bBbYrTdhCGIbSqBl8bJKS47h/3saZJ0MTbSo11RmivPLiC5o0m5IKM0nIJ3Z0u7JaNwAsgKVLCHqavHz/HcaQ731foD9tGu2XDNV0Ul/aXCBOYHOLqjkFWNjyk4Z5aHCZFXikwR5aew6SBVd7DN+1k8X1miYsO7TNjSpUkS+i1e7Bb9vAUCDtGyGjK0EIfWS2gM1G/+PUZ5ZBkOY/zLtsTZYEC8wYnlLPw0asfxQf+9AN4z5e9Z+L9xZ0NSc1/2GdNjttvdndUyTPfd9zpKp0oQZZlNG800W104dv+kJh6nHiaNvs05ORpClYeXGFC68W996GnWx8Oyk5MUpEpIDAt5hnIx/0Pr+eBFAl820ftLNOAUXQlei9wA8iajPKJMohCIElSYj+rD65GE6WsLQutGy3UztQSQsG1M7VEZp0TLbIqs7XSdcf7GjnrKl6Z03ylCUVXIkI3r8LQ6TjwLOYj6Uv5Y8a53TZvmXAsB3JBRuVUJRFAZfmH0XfPgIhfpD0RrV5HA9N0Ovzssz8LGlD88Gt/GHpNH18ts59ugnhbd78dcpL7OH7P+7YPAoIwCEFAxu7Dd310t7pQdRV2k8VHzetsSiihBMZpI9L6GuUj8HvfbibbSvk1CxGiuFQEgMHxFBhxPer6ZWmPZh4DBdo32iAyO2etoiX8OHPDhNWw0N3qwjhtCD/jACCu8ASYZYvDOGSVAtstG4qjRAsii+yYtMJn2oevucHKjc0Nc0hQkH9P3BHRa3qk50MJHQSH/XJEEETiXjz7lD6GSaoF1l+3PpMHuXAIBI4a0g5QGiEN8R0f/Q7IBRlPvO6J6e7xVJ/2NM7SfoiTLG2MSMgv7XQ5PsIwhKIqLPsThLli6nvJ3mVp9wCMaJcKEnav7kajhCfNImZdy7hdn9U1HXeNR1VkCghMg4NoIVQcBWoxOdyB68AAiN4DABCgWC9mJ6schRG0/cpgAFg5vxKtXd5WqRZVeF0PpsTGiZdXy9AqGmRNjqb2jTreUdN0SislLN23BHPbBFFJ4polstleGLVZWA0LkIEVbSX3O411A6EfolasoXm9id5uLxKFTlxLPf/v/diZRdkT0dJxdDBtp8Nr9NdA0ZWJ/I792KFE23hBRu3seM0qIHnPy6qMa398DZIkQS2pOPOVZ6AUlQS5wf0ItahCLshwOy7CMERgBTA3WYyllTVQMBH0caRU/N63Ni2ETghaZm2l3CbEW7D49zttZ9g+pmUxMKHOKAGqZ6qgIUXzehPdbZZc49dFLapYfWAVvWZvKKm/V4h4bTTEFZkQo1ocHn/w8ZmQOcCgzI2XAqPFWp+4IKBe18cutnEVLpM+fEcJCsaPNyr5vd1hxsj1UagWUDlZgWu5UTVN+0Y7CoTiztRQBU3fwHS3u/AdNgYwjbhDsp+M+agJGJNsnzaawtAIzBNZZI5EJLzt/Nvw0asfHXKYAjfAN5S/IaFZxdun4uAP4CG70m9D8G0/k3yNtt8ncZLWdujuDCZZpY+hdrYG3/FZqW8QjhRT30v2Lq0NxLV70qPOAzeISq0nOef4fr2eFwmj5m07i4Bl0opMAYFpsZdAfmrNndSa6TV76O30QChBcaUIWZMBYGwrEW9jilcG86mYPJvcfrUN646F4nIR2pIGraKxfVMKv+fDgQNzwxwSMU5fkzx/xrVcSAUJpeVSoooonc0urhQhKRKctoNuswu1rMK38+0WT6IRiaByirXc88pn3/YPRPR9EfYk3fYybvKhwOLwxOuewO98/nfwwec/GL2W1ekAAG+//+34nj//PRO3Be6HUMwlYfstUOnXsqqH26+24bZdQAZkRY7sCt8mre1HJAK/56N9vc0qAAMKCgrHckAoidrIR8F3mJYoT5jrKzpUXYVW1RItW/Hz9DVmZ9JV2OlrAGRrjya+v1/NRCmbxicX5GRbqjYYoV4oFzIF4qeFIHDHQ1yNCTGqxeGpF5/Ck88+ORNSh1et8FLg8loZgRsMxnojW9smq+1p1JSXSRdClqBgAv1gy+k4cC0X9XN1yJoM3/YTPeeSIoHIBJIqgUgkYfSGrkGfYeZE0Kj2hP0s8jhh1dvtMcN2LjuwGpX5T0/hEIZGYF640rgyRObw6sDf/J+/ie/+2HcjxIDU+fztz+Pr7/36Ic2qUUFHJIzeDzR4lilPwBPYP3ESD8i6O92I/LWbrOUzfgxxPZ+xTtyEhFTiuPJaFFKjzvk5TixyGNtvXrZsv9d06DtHZC9Ftktgv5jGl9jLszq9ZtqvtqNMNpFIlPkd10qk6LFW7VRlMCez6+fq8LoeC7i6PkIvhHHCgFbVmO6FH2L76jZcy0VpuTSVZhDXzpILcsKf8x1W/RfPZlfvqcK3fex+YRcI2Uh2SZVw71vujY43fn5ZAVnzWhOO6aD5chP1++rQKlru6OKjqoETJf5iz4v05EOBw4HLT19OkDmj8LtXfxc/82c/g594009M/HzaD6GYVa2WpSmTZbuitVMtQCISAj+A3bIjUiWe8A79EKWVEvS6Dq2iodvoRhXGWlVD/Rz7njyymB+b7/iMELrJ2p1oQHHqjacyq4Iyr11GKJe+BqMIsvj1AYBCpQDP9qIqJ77NrKv2jqqdOkiIqzEBJmlxiPeE7gd5D+e4NkQ6kMlqXZhFKbRe0yNBQc78xhFpQoDAvG1C1mS0rrVgnDaiCpzyWpn1h95hgn2aoUEqSImqm6xKF7vVz9aXCiPFCBOaPZ2+Zo+eLTSYtT2hJCrjLq4WAYqhUcgjDXm5AKftAARzaUsREIjj4mMXQUEjccF4q+f/8dD/Aaft4K9+4q8ipCF+5IEfwQ+e/UGmNTFCswrIJjH4GuQCf6PGZkYaPB0nIbg3CnnEiblponqmitJKCZ3bHYAgqvaLH0O0RvtBUZbjMSkhlYf4ek2PEo4Ek0dUEmad86hsWeKzM2hpGZWBFNkugYPEXh3yoTUTy2Sng59RgV1eqzZfZ9yPoJTCbtrQ6zqsLQsr51dg9kxsX92GvWtDr+gIKkHm8Wc93+OaiLylAUhO4Etns23YLGvfdgAA1qYFc8Nkeh05ldfR9/WFT60NC07HYdNt+j4LgCFf0Xf8KCl1lDS1uG2LPy9EoHc4kW65GgU+XGZRVaRZdgrIbkNSNAVqWYVW0eCaLmSZEbbNa022Pvt+EdfH4aLFxroRae1w3yGrcjpxXCltv9KJEgrlAkI/hFJUMquC0jFLZhV2BkZd+3QLvNN2okrHuI817e83LjZapFbXUYG4ImOQ1+Lw+IOPJ3o/Z03qjGNLE4xyygDl6UmMg920owksXHhw9cHVfKa2/72qoaK0WkJ5vQyv6zGmWB+QTYEToLfTAwXF7iu7A4HR/v4aLzYQeqx1QtbYmL7dL+4CBGMny2QZzIgdHxO0cMKqt8t6PCvrFbhdN1G6m9uyEC+jVln59DhDI4IogVng0mOXQECGxn8qmoJ3nnsnfNvHCxsv4Cff9JNwLRdalbUO6FV9ZJVK2u7s6QFKx39k5HfqTDDU2rTQ3e6OXFtjx2Zy7YoJCKk4eAAGAjS/2EShVkBxqYiTX3Iys6pwkkrCXIwiyWaU5co6JpHtEjho7Nch5/4IFywelcketY9RZHZ5rYzN5zejaVq8NctYN+CaLvSKzlqwbH/IFnHh5XilLoAhTcTyWjlRVQQM+2s6dBgnDHQbXeiGzoInd7I1y6srKaWRqDM/3kSb0vZgOg+ATILssEPRFRgnDPi9ycRsBRYD7qd86we/FXSEk0BAZqZHumekqnpBkUt6xm2S3bIRuEFiwpVe05nGlRdG1Tjx16fxHdJEiu/4kGRpaKT5JDHLuDhlUmKF7zuaotW290TETRIbxSsd5zUl9KhDXJExGNXikCZ7QhriSuPKzI8hb4HER+umF+q0i8pu2rjxiRtRCd/ZN5+NSJ28LBQo4PU8uKYLv+ezsaFNB5qhRQw1Ny7GCeYQeV0PBaPAypr7ZdTmLROqocLcNFE5VUFxqRgt3tAPM6dkxQ1OlsEcRcbEr2t8AobbdROtHrwXPcsIZlVSjQu8RBAlMCtcfOzi0Gv8nvz2+rdH/eftm21Uz1ahVTSUT5SnctinIRSi7M/q/u7tdFXNyvmVaC2nj2HUeuI93tNmnuOi9L1WD1vPbqF2rob29TaME8ZgpCcvfe47Z9Osad9mI4yJRKLKo/h2dtOGuWkCQFSROWs7IbJdAgeNWRCUiq4M6fjN6tgUXQFq7P/TrVmKpqBgMPFlSmlkl4BBMOJaLrqbXaw8uBJNmQKQ0ERUbCVzPWcR26e/7DQUjYk5q2UVxpoxka6IorP2MgCorFdAQaO/4zYx3jKadQxHBbMivgUE0v5Hea0cDWoAsklPbpP0mp7opuDrk2tcxfX+AESyExNX9sf9GUojYnjoeCaMWfJiykmIlTgBnm6T30vbo++MHoEe/20OekroUYK4GmMwqsUhPf3q4lsvZgZa80BWm9Wk01ayYLdtEJmgvFqG1bAipnWcdgx/v3K6wsS8Ckqy9FVTIKkSvC0PoReyXlIvjFhl3/ZBCQUBgaL2hbT8EDRgJdAFIymolWVw8gzmJEFL3Bjz0t1CuYDGS6xqqGAUcq9tlhM2CpMcj2jJEtgPhu5nUo0yRnt5AE66zay0WrKqavKOIU/3J54pB4adsFHHw19XdRWtWy2WBZMkRt64frR9Wj8LmIw04tsGboD2zTYAJMYh200brzzzCjq3O+hudbH651ax8sDK1ILtkyCuRSRsjcBBYJ5O+KyenVmtWXzSHfdn+Gs8KUUkgmK9CGvLSkyZ4p9RSyqUkpIggiY5jtNffnqoJTXLJ0u/lj4HAAndC62qRcHYIkndWf1mIrg73ODJ71HVOQBAQWfW6bAXpP0PXhXH4xYg38/PqyCJEyC82gfI95nSEhR2y0bregtEIgjDEHpNj9rbrS0rMy7JI26ipFHLBlrDz/5Jk86R7XF8Np10y4ReZ+TUKImMXMRGoNOARpOQ4xAJ8fEQV2MC5LU4AAOjc6VxZd9kzl4Cn3ib1X6UxPWqDhpQtG+34fd8IMwWAbNbNrrb/ckLXR+yJkdZZq2qwQ7taCIFL/ONl0nzEuZ4aXHlVIWN3azWmDAYH22eQaJkLWq9pqN+b50ZqRjyDFvWdVb0QauH02E9+umKn/1iHEMuWrIEZgVFn7wUfRZO9ai11rzWzJyylbkOteRUG9/24Wv5TkW6Sq7xYgO9nR6cjoP1R9ahFtWEsz9ujfGsvKRIUWbc7bogEoFe0aPqmsBjpdUAC458mxE+o6bRADGHcYVtKxfkhN2224yY1ioavK7Hxpj6YeSAAfsnYNLXYNRzQxDMAkcB8352pisQ7ZYNu8laLJy2A6kgoVgvRlOm+JpKVxuO0snI80vy/Ja8Nq+sbSNNwlg1Tly4dRHrW/g7dw/SnQ6jMK9Oh0mQJln0qo7m9SbMWyYooZBUaaBVk+Hr5FWQ8P/yZI7vMHuQJoCykkWu6WLr81tYeu0SQjeMJh3bO0wiw+k4OP3G0wnbkkdw+jab7MvPp3KqkkgWTdOWlU7s220bTsuBcdqYnhzuj0CX1D5xliXcLKqKx0JckQkxiqyZBZM87cNt1je3Xtdx6o2nsPn8Jox1A1aDCenF+0EBoHW9ha3Pb0UEUu3e2pBgs92y0Wv20L7ZRvN6E/Vz9ZFjPtM6PVmOS/T3iPPmozt5uxQnhMZV+PDviJwvx0dxhZE5k4wXngajMkmCgRaYJTjhkSY645ilU511b/NS2vSULSCbMI4fs9fz4LQd2E0797ji32lumOjc7kBWZFh3LOxWdqEZWlKAfYI2TL2mY+NzG1ALKvSajtWHV5kNIYwwck0X3a0uAFZdAwpsfX4LYRBi4zMbuPdr7oVxMrstJGG/+loclmOhdaPFAr6qDkmW4HQceJYHAoIwCNG63kJvu5fphE2Lsa1qGVl9EXAJHGbs9dmZN8Gy8WIDoR9CUiSsPrg6rKsBIHAD2Lt9wXmwkcFcFNS3fTRvNNHb6UWt26P0ssbqgWVUZHNB9ua1JpbvX4ZaUkfq6uS1YfDrF//7ICD8nbsHFx+7iM9tfm6iSVfvfvjdB9bpkEZWYspwmPQDH8vNydxRg1Ky7mff8RM2Y+O5DWgVDWpRTcQe0bCVjgNQQFIkNkWq40LWZSgFBU7Lgd22oymgedNL0/AdNr1PNVSmbbrLBsnoGAzamaR9kR+nJEvo7faYTTppwOt6QxIZE113TQGlFJ7l5Q+JmCAhfrcnn+7Os54xZnEj7eXhtpeS+TwHxnf8qDKGC+ZxFfaonaF/jEuvXULgBpB1mY3gS/eVtgB72wZRCZpfbIL6FMXlYmZAkEXeZDkuceOZVb4Yv35xsb+0c5S+znbLhuIoUdl0abkERVNY+1j/eNLjhfl+ZmE4EgEUdxodP9GGISCwH8SJzjzh4L041ZPYPd52kJ6yBWRPjYhsETB2IkNcywuE/U0ogazJ0JeZg1JeKSeyZeMqgHzbh9lgny+tlCJiqbxahm/76NzuQCtrIBKBXJBhnDDQvNlkgoh+AN/ysfn8ZqQ/NnQ99OQYZmvTihw87pTd99b7Eho6ANC+2YZqqIzgyekxnxSjSr0TlTv9TKAIuAQOO/aS4MojUeyWHa1zxxpMcOH2qXaWtQNY2xY802NDJGoDvUFOCDWvNWHdsWA1rGjiTfy7E37PGA0Ju2XDNV1WNYwQdptNsrKbNna/sIvO7Q6MkwbKq2VguFshs5rRbtnRpL5Zte5PA5Fxv3vw5LNP4sMvfDjxmkQkfMPpb8Af3PqDRPXOh1/4MJ589smFCSOnE1NpSQcgf+JVr9lD+9U21LIa2YkIFEzOYpfJWQRewGQoDDahl/spvDJHUiQETgBzw4QkS4wsOW1EFYDd7S6LE/rfPcnzmbeM9m71YG1YzH+53mRt5rGkdVbVbjqx7vU8mLfY1EHrjgXjFJuuvPya5b3HxGNIb9HdMBp33xnPGLO6kabRoJimZH7csQLI1IOQCzIzSKmHu6RKCN0QoRcicAIEKwHs0B46BkooqEcBGVBLaiSCPCrrFBdRjgiXtj1MwOhKpiMSV14HgDBgxxn/3nRAx6tv0loYUTmy5sNu2tHvAjq7rHU0zrTPTvPj822fBZAZWTPBQs8WhJD7APxTAG8G4AD4HQA/TCn1CSFfCuDXATwM4HkAf4VS+mf97QiAfwjge/u7+jUAf5vyH/KQYGxFyh6d6kntnqIz3Yi0yCgwLOSeLuMNnCB3rHdai6Z6tgoaUhSMAoIgQKFUgFbRhnq6R1UAcUHkwAlg3mJtVdSnKJQK0bV0LReyIkeEq7lhwm252PjsBuSiDFVXIRfkkQQUrxr0NR+tG60B2dUP7PS6niif9m0frestmJsmFFVBrVrbV/CTl+kaauPN+I0E9o/jbnMWgXHZ2yzk2UaeyJIVGYSSSEeDVyvbLRvWlgW5IKO53WTfX1QGxKjDBkSUlkpQSyokWUpkrTP9np6PxgsNKLoCSZYSGhK8ethqWOhudWGcNqBVNGy9uoXOnQ6Ky0VQUEiyhNbN1kgyOe0/cn+qtFwamQiLH8us/I+9/GYCRxNZw2X+xTf+C3zT0jfh9+78Hr7349879+EyeRg71SmLDI3FBPHXtl/ahiRJrFrugeSgg+2r2yAgaN1ooXquGlUAWg0r8mMqpysInIB1P5ytsVZJmWDl/Aq8nofqPdVon8XlIjzLGySmMtrT00kvLoGhFJnIe2mpBLfnIpSSxJJe1zNb59Papb3dHqrLVYCweEstqdj+wnZUeTRpjBS1tOZVN+XErvw3mSYxeZxjqON1NgvArMpGc53sVJDDtWj2VF6ccazxfQGIKlOybnZFZ8ZAr+kwGyYkKiUFkPuf12tMF6e704WiK9FI8jRJZW6abMxfSkQ53cPKhfviBEzcEUmPatcqGm5/+jbMO2akVM+zVoo+EC6TCzICNxh57unfZS+/d15VVHycKVGYSHTlVAXd7S5zzFKOlWCh54J/CmATwCkAdQB/COD/JIT8CoCPAPiF/mf+vwA+Qgh5gFLqAvh+AO8A8Aawx+UfAngZwK8c7OGPxiRE8aisbN7Db5p1kCUyCgyP6o3rPPgO06ThwsZp8O+nIYXnePAsD2EYwnMZOdK1ugh8RswoFQVaRUvYAMVREg6E3bLRutGK1uPaI2ugEkXtbA3V01V0t7tMgFBV0Gv1UD/L1l7gBiiuFHHiDSfg9/xo1HEeARUnoLmt8h2mAzaqKk/WZFROVdi6P7f/dZ/OQgIZ+gH9Ftrj6vwsEMfa5swLkwRe09yjWbaR69LE13l8upTX82BtWnAtN5rqKSkSaECx8sBK5D9IigTH6mvxrRWTbd+pahy7ZaN1swVFZxOtSmulRKaaBzurD6yi1+xFgve9nR46tzoAgPJqGaWVUkQKj7oOcdvNA8G4DZ8mqJqHHRI4frj42EV8/JWP45lrz0TDZb7l/Legea2Jt598O77n/Pfg11/6dVDQuQ6XGZccn6S1Gxj2XXybVeYGToBCjU3xtds29Lqe8PVDP2Q+xZkqS/7oMty2C0mR0LrRgrVpMZJny4oqcXiFtaRIABg5ZLdtqEUVaw+voXmzCafjJJJTACJ9LSKRaAKVXJBRv5fZtN2ruzA3WYykL+sJYmnptUuQVTmznay700XzehNOy0HQC+C2XQR2AK3GhuK4XReSJKG8Wh7pG04qpQFkd1bE297iCf1JtX+OYwx1fM5kQZhl2WjWwy3eq9i53WEleqoUvT/Nd+Yda9qJn4SgUDUV7RttyJqcyLrz86ifqyNwA2gPaqCgmRmqwItNeuHjQTOILW5Q4u1PTospvAODKTH8szZslNZKsDaZDlA8axUXLsuryhn3u9CQorvTHZBFY65V48VG9Ltx7Qtu5Pg4U7WsQlKlkY7VqCD6OLPOc8ZrAPwSpdQGcIcQ8jEAjwB4DMw+/kI/A/5PCCH/F4C/AOBjAL4TwM9RSm8CACHk5wB8Hw5ZcDUJUZyn2TBScHMKu8fvTb/H9qlX9Wg95lUL+TYjR7IIY/7ZXrOHnS/swNqwYN5hWSWtrKG0UorI2TufuQMjMHD707dx6o2nokx6+vj5Pvl6LC4XsXJ+JQrkeKXg8vnlKFMWlWJ7IYrVIkrnSwDFYNR6bE2mW0I3ntuAa7oglEBf0VFeG4yUz2rHUItq5CCNKkveD/LuFWFPZo5jbXPmgXmRCVmkslpUceKRE+g1e2yd1/WED2IRCzSgaF5rAjKw8toVOB0nCuB44stYNxKtWpGWWWqii1bRWNVeVYdnM58k4U/1bVWIEIVyAYquwL3lAhSon6vD7bqsFeOkMVGrdtz2xSux4yOIfZuRzTbsxLVxLZdpiWD6ts+0Pte+ZQqEz3NkcOnpS3jm2jMgIPiBr/iBqJ2qfm8d//qz/xq/cfU3QEHx1nvfikuPXZrLMUzSETDpPR2vduPaWlpVg6zK6LV60Coa9GpMPLnv69ttGzSkUSWvsWbADFlim9sKXq3Lk9Y8+eP1PFhbFto32iivldF+tQ0iEQRugKAQQJIluLaL5o0m/B6rEuxudmGcNEBkAkmVok6JtAhxnFgyN8xoKES8nYzHPe0bbWg1DW7bRe3eGryuh5XKCqwtC/YOa+O0ehYKRmEoNhz1W4yq1svy19JDgabR/jmubeS5Z0II+Z497vN3KaU7e9z2yGHeZaP8Ru51eonJS6MqaUYda9ZYvWmOny+I+IQX40SGCFZfNDm+4Cbdx1Cgx41nv/2JZ8tLayU4LQfVM9WhawawHtSEcYxp5UxSkTTqGvIWEt73ys8rvR+7ZcO8ZUI1VHhbrMLK0JlIoqwyrQ/FHowz9R0/d5zopNoXR5V1XpDN+QUATxBCngawBOB/B/BTYAHWZ1PtDJ/tv84DsM/E3vtM/7UhEEK+Hyy7jnPnzu3xMPeOUURxHjnYvNaEa7qwGhZWH1gdcuAntXuc0Oxud9H4fAO1+2qgAcXJN5wcEs+L73OSkbqu6YJ6FKXVEiRFwtJrlpj4suWBBqxyp1AqoHauhu52F5svbKJ2Ty3TgQBY5ktf0qE4SmIijblpglCCbqML87aJ6tlqRL7wfaTbU+OktW/7qJ2pJVpCiUSglRnZLUlSwkEcIoAPUGvibsuYH1ebc9yQV128X78rj1Tm5AmvrInbRkop9CUdlaACpajA6TigAY0COP55HTqazSactoPdl3dBQCBpEgpGAeW1MvMJTC+ajiepEgpyYWi0edrWAoDX89DbYePRq/dUsXqBJYr4MdgtO5c4GWe7eWvI7U/fRvVsFbIqD0SYN5nvVTlVmcoOZU3vGdeScVw1MxZhcxbZ5nn56cu4/MxlAGws+S9/8pfxlrNvwROvewK/c/V38N0f++6o3eqZa8/g8tOXp6rQmZTYy/R5tNGaeuO+N35Py6qM1YdX4bQdnHj4ROQ/JHz9koLamdogsRRLWJfXylGlHg1pIpmm13WoRRVhEILIBKqhorRWQuAFkQ6OuWHCbtrQahqcjoP6uToooXC7LmhA4Zqsjbq8Voais/Z3IhEQQhLEEg3YZE1CyCD26NsMc9NE9UwVhXIBjU4DXtdDwShEg2/aN9soLhXhdt382BDDVYq8SjnvumfZwHTb2yS+y0H6UovAqLP5NUSddxODAvgzAHcNoQPM1wnmN7LdstnN3xfmmnZ0bTrjnhAKTbHNowxjIrujypkLlpfx8gd32lBOso9R18LcNFFySqAhZWy058PasqIgTNFjuh0p4zhpVc5IpMiqaIRpxmhmgOkJERBQQgfXJ6PVxbcHmjl5mfJJtC+OMOu8CJvzX8DIljYAGcC/BPC7AH4S0aDoCC0Alf7/G6n3WwAMQghJOzuU0vcBeB8APProowvTu5i0xJXfT8WlIrpbXfSaPZYZTj38MomilHPFBUYd04G1baG0UkK31Y1GfQ+NDdeT93ueo8ZFQiEDXstDoV5gI8pPK9FUGbtlw2256NzpwLd9VE5WEmskvfazyqg5qdTd7YLIBPDZdJtR14Bfw/h0L4BV7oAAWkXDxuc30N3sMq2f+7SBk5JFAK8bQmtifjiWNmfRJPKsoWhKpHklqdKQlt0sxXyzhk3EgzcAKJ8oY+WBlah1PJN06NvRMGQixlpVQ7FcBCQW2IZeCGvLAiUUnVsdlNZKKFQKiSq9uPYFr/QDZTakuFIECFBaKrFEFSFoXm8CQMLXIYRl2uN+SdxuJSpn+r5NGIQgOwTUp3AdF6bE/KeVB1fQ2+1NPckm7qM4bQcgGCt4P7KC9Gj7PIuwOQtr87ywegESkRIaOe/50Hvw5LNP4qkXnxrS1rmwemHik5qG2MvyeeLx1SRTNRPf7QwnhzniLZbx7+DvZfo8NWZ7okrATYtN2jU9QAL8Xr/NPGC2o2Cw75UUiRGsRQWFSgGFUgFumxFUlVMVZhspGFnUT0Ib68aQneP+Gk9yZx2nccJg1T9BGO2bf07HQDx6bFxHgd0v7iIMwkgzbNqW2r34RPMuwFg0xp3Nu8CMyKT7emk/ByPEArOh6AoM3diznsEkGfdpeknHjY7jYqVux40ClbShzJpUNSmpo1d13PmzO2hdb6Hb6MJpO6jew6p0+Oi+uG5HvFXLd3zImhxN5uIOGW8DyeqxTRvi9IPBt310d7rwTA+U0sQIQa4nFPohClUWFOeNTc8SHcs6//R1Omas84HZHEKIBJb5fh+At4AFTL8B4GcA3AZQTW1SBdDp/7+Zer8KwDysNmeaEtd4lrq4UkTBKMA4wbS7RhG+8bXPAwgAbPKUIqPX6GFb2WbZ7K/SEXgBzE0znxROrcP0scoFGZX1CkxiYvmBZZRPlBNOiKIrqN1bQ2+3B6XIhMdHrZE4Qcr/SySCYr2I5sus2karalFgNSqrzCfWWdtW1FLBA7PmtSZ8ywcFhVpRhzRx0gQwP7bj5nwcIhw7m3NYSORxyCQu8qr9HB9e14vGc8c1HcaJ+U56LHnDJhK6M46f9FkoI3kbLzYSAQ4nodrX2/A6HuxtVj24fH4ZK+dXYLdtVEkVkiJBKSooLhchyVIiQcS1L/hPzPUD1aKKs191lpHtRgGEJMkSPk4YFJAkKRqTvHJ+ZUAMAVGAl9aiCL0wEmOXNAlhGLKW1KKKglHI9E1GIe6jcMmAUbpu8UrqWYr5HyIcaGyFBbZ58vaq93zoPQlS5yNXPpL4HNfWmWa61TTE3qhEaVpTbxKCMOse5GspixQaNW00foyKrsCmNto32wiDEOYtEye//CRkVUZ5rcwE08mgEyFetdK81kQYhIlpWJyo4UnotPh5fC3zY/RtPzf+iMducUoys4omx1/kbV9qSWWj3Fs2/A1/qoq7vfpEx9mXGnVWAYBblNJrk+yo76gEYD/zXiHEAlNIBzJ7clQmyLhnlcCNMox5wYy5acLtuPC6HjyblRPrFR1hgU2c4q1PPOO0l7JZfmzGSQNUYs5H6A/2b+hG4jijVq1+/2f1TBV2yHo9b3/6dtTLfuqNpyJhVu7cbL+0jeaNJpsuc28t0sCJG7Xm9SY6tzuwd20WnMUCPkVn/fRZWhqT6uOMwzFinQ/a5iwDOAfm6DgAHELIvwDwXgA/AuBvpbLfrwfwy/3/fw7M3vxJ/+839F9bKKYRMR6lGcUzSr1mD27HxXZrG7ImjyyT9x2fPZxjY7hXzq/AOG2gt9vD2TefRelECb1mj2XDmqxd0u8lK3WyBITjQQcnOyunKnBNF4VSgfWnN+2IpLVbTDiQEIKCXohsESdyE9cmFlSaG2ZESPE2qRAhyutltK63WB/5Vjciq9LXPl6CTSmF03HQ2+0lRraHfojiUhH6EiujjjtFaQJ42sBJ6EpMDWFzFoSEll5fF4KCDlW4Aiww6O30oJU19HZ6ME4MRDDtFlvnoR/C63oJH2Cq43EGeoW9di/pS/TJGathobvZxdJrl5gIel0f0jiMB2zGuoHQY0KoWy9sobhSTLRQ26GN0A+jrDvLTSISQeXaF57lRVUtvAJa0ZjOnnFiINosqRJ8x0fjpQZCJ4RjOkysVFNgb9vYkXbQ3eoChBHiBaOQ0CpLDJeoauhud5lmzh7b/Dmygr2hSuOMtiwgWyfyiPs8i4itfgELbPPMInXi2AuZA0xP7OXFLtPuJ6vKfmQb+5gW96H7mADVs9VoHRTKBUiyFCWJ48cRP7dcsiV2fvy7ORnOk2qTxiD8tazYLYq5crRD45ALMrSyhq7bhdWwmD0rFdDr9PZsw+925N61lNLs8SL5nw8BTLVNBoRYYAyz6hOOZ9zTzG2ElFBffGzmNMcaeAF2X9mNekQDj2WuSqsl+D3miOh1PeoLndTIpYMu3/Fh79pMU8hjjlzndieaFhXP9nFDx/s/ufPSutWC53iorFfgdByYDROqNmDpzS0TzRtNNp69F8A1B8cYL1FUiypOPHwCm89vQi2rmSLR0TnEWtEm0ceZ+Dc+BqzzQdscSmmDEPIygL9GCPn/gWXLvxPMoXkazIn6oT6h/H39zf5z/7+/CeBHCCH/Huzx+bcA/OJej2UWGGUv9uL4oAXY2zZUQ4W5aaJyqjJycgF3auJjuEGA1QdXYbfYWlWLKuvhLirQDC1T9DjtWHBB4jQZtfrgKsxNE1pVS4wUNjdMdG53oipBrcoECuM2L64x0XixwYLBngdVUyMyGoi1SfWDGUopC7oyiuXTbQWhH4L6FF7PQ/N6E8vnl6PxnnwCDq/4iV/3OAE8zZrmo1EVTYmmWRx1mzBvCJuzOEQESr+V3NwyIclSosI1DkIJKCgIJRFZwqtZOnc62HpuC+WTZUiqlElWjxLl5c9mu2mjt90DJTRzP4EbDE1b8W1G0sY1Drk948LprulCX9Kxcv8KXMuNMuQAhrLuvu2j1WoBhLVZdG53QCkTT04LGad1OHirqSRJLIG304XTcaBoCsIgRKFYgBmYUEsqtLLGWjkcP1OLgu+r1+xBUiTUztZmZk+y/JWsFpZRBNJR9XkWFFstvM3zidc9gSeffXKoMgcAHn/w8anJHGB2xN40+xnlZyWEe/tVurwLIe335w2e4N/BxZMlmSXZuXAyl64YdRy8Utpps6RZvG07Ln4eJbj7ybCs2CTzGmSQP/x1Xqm0fXUbxXoRruVCKSrRZFBgkLhyLSbuTijB7hd3mU1T2ZTAPclh3OU4bFfrF3CXiwXmTUaZajx5RlXPWGOVUj1PiBhPkPXlx1paLqF+Xx3WtgWv5yHwAsgFGeX1Mno7PbhdF6WVEkIwlj7PAKbLgKNSxr7y+/oj69it7EKrMGX57ZeYg2TdYaP3eAYvzhzrVR3WpoXudheUUnhdD72tHno7PehVHfX76kmDpitQVAVBL4Bns4w7N8RpLRJJkVg1QhZZhuEefF4pAAyCyyOccTrKeBeY3flxsGDqPwP4m5RSlxDyDrD2zX8I1ub5jn5FIAD8KoDXAvhc/+9f67+2MIyyF3t1fHj7D6FsDGd3p5s7SUXRk9pV8elz6ZZRgDkd6ZHBnLSNCxXKBTmbBNUHPd3x6QehHzLB4TILgiil0JfZWPC09oaiK+jc7kAra7B3bbiaCwISEVLRNaRAd6sbEd5ZFTrptoLACQAC1E7VQEGx+fwmE0TUFKw9tBYFfEPE2ISBSjpA3b66HdlJY804aroSdxOOjc3ZD6JEk8/ae5SiArWcWnd96DUdxmlW7RKvslMc1l4tF2SEQYj62TrUojq0/ShR3rh/4bs+CtUCikvFRLWP77BJc5qhYfvFbXTudNhAhrOsA660UkpoHMZtVKSBqEoJsocHasBAt4eTsgAQ2AFUQ4W9Y0MqSNAr7DzjCaXoWsaSRgCi79IqGlbOr8B3BtM9JVmCYzrwLA/Vs9WItM59LsS87r1WAE6SnEwnHURANxscljZPrpmThadefApPPvvknkmdWdwneftJP2fNTROBF7BqklglX7pChlf7tm+2IzsR9/u5/9Pd7GLlwRWECBPjuAGm1VU7W4vWb1y6Io9UMTdNVkm41YV5x0TnTgcrD6xg9cHVhMh7QuD4pUE1TVYV89C1Sq3VuF/FKxntXRu9Ro9VCVY0IMSAsHJ81M/VmbZYP7EXeEH0etxOCkyOfV8tQogK4BSl9PoMjmfhLPIikdVqMG0ZYN6o4XFGj2d1ebsBf4inHaE8wiK+wLWKFo3VowFF69UWI3PaLohEsP3iNozTBmpna1GQFydxOMFRKDODaUrmoH+9n0kDgKAXQFlRYLdtyAUZssZGBma2NPUFTgHAalionq6isl5BeaWM1o0WG+lHMsiWewdK6uuPrAMYFmPkztioDFY6+5RXtjhti8XdiFnanL4O12M5730awJfnvEcB/Fj/36FAfA1mTWyY1vGJZ1EkVYKxYkQTm/L2E9euGuX0Z5Xgp9cVz2pxEeIsRyO9H14BZ+/aTJjvtMFIH9eHUmC2IvCCqKLHd1mbmAMHFBRL9y3BaTusGicmpm63bJROlFAoF1gQllGhEw/eeOWNuWGCgoKGFFpVi9a/Xtf3tdaH9D7qOkAAe5edH9fqmDfulhYvYXNmj/jaLa8OJrxkEcaKrkQBQHxCJ7d5RCJQNRVEIYnkUFSB42SL8nZ3umjebIKAJaP0Gkv6bH1+C1JBgqRKUaWKtWmxyS2nWfAWrwzU63pks9JtRLyikPs65bUymtebaLzUAKEkykYDjJTt7bAWTaKw94x1AxQUgRfA2rKiqp+4/l4iiPQCOB2HkWAx8eLo+1fL2P7CNht33LetWbaIk1iSwfR4zA0z0raZtmJ8kuTkrKotjgtmaHMW3ub55LNP5rZbAQOhZAB7InU4Zv08yiKCAzfA9kvbUFQFakVNVPLFK/eJ1B8VLrN1zFum4u8X60VYWxZ6uz1IqsTWuRugtMImAEeVchlVynmkitNxcOtTtxhZHoao31sfktGIJ8N6zeQE5Un8xCG/K2VfVV1F5WQFd567A1mVowojTlg5pgOn7WD5NcuQCzJcy4VaVqGW1SFSXGBy5F4xQsi/BvMh3jNmH18G4I/BCJg947CwyItE+qEHApY92jLHLrBI+DjF+E7KcvISZp5d5+SH3bYRuAEK5QLMWyZCPxyopKeCq7hqum/7ePWTr6JzpwMEQMEooLjKBFY9ixFD/JzTBsF3fDgtB51XO6CERoJ83FjXztTgu37UPgUCELAy7PrZOmRNHiLB7BYbd+5ZrNLGajCxUrWoorhcjAgkfi34efF2Ef56/Di7211sPLfByiJVaWSAltXysh/dnOOIg7Y5hwmXn76MC6sXMp2ZJ599ElcaV8aO8owTCtNObMjbX9TWFGuPymw3SombpsXO8zK0/Li4gxO3fYo+ECrk+xlVeu/bPprXmwi9EBQU9dfWoVd1tG4yrr+71UV5rQxrywIANkq0pgMU8LoeJFmCccJA/RxzVBJi6v3zk+Q+8UvzBf/igqbL9y8jcAMsv2Z5qLVhPxh6VoBdc97GWj5RHvqd5unsHrXRwRx3s805DIjWdGzCS6aIpj2oMElP6OSBBW9DSlf41u+tw+/5aL3agm7okApMlJe3HMRtAg0p24fMKngCN8DGcxtwTRehw+zK+iOsHSxdYZi2T3mDDnybCYJyIih0w8jehF7IgkfbYwFOSUVvhwVcakWN/JR4Rp5/h92yWZuWxdperS0r8rPi19oGa6+wd2zYuzZuffoWTr/x9JBuEdcNMm+ZoITCs72E3k5WFVTe7zdpy++sqi0OMw7a5iy6zTOLzJGIhMcffDwx5Wq/pM4snkdDUg9xoqLjsBbLns+mv/ku1h5Zg6RIQy1HUfWhF2aPAe+vrdALUVwuorRagt2yETgBdl/ehWsygoOTNLzSB0Cy8nndiEhuEKa9VTAKqJ2twbVd+C0fbtdFRakM+SyJ6kFFitZ0ea088XWK/LxYVbWkSiiozGern2M+GA1pdI0c00Hj+QYCL0Bvu4d7Hr0HkAbj1OdJ6B73BNSoM3orgL8Rf6FPulwF8DildNZifAtnkReNTMb1ejN6oNotO1NcCkhOZOGMb8EYHjUMjLipY+O4+eQIRVfQvtlmujeEQi2qaF1rIfQH4n8AEgJYPBMU+AFUXQWlFKETonWthdJqCTRIluilq5FkVUb1TDUhyCcXZFhbVlQqCADt220mdGpoWL5/mQkw97N36f54c4OVHnLhYr2mRwaE95TmORtxdfr4cTptB/a2DSxjMGK4L+Q1advbYZjUcIiM3EHbnEOBS09fwuVnLkMiLNCIOzNxZ4iC4tJjl0bui7chpCc2ABmaERP85vFMTt59ygXwXNNFd6uLypkKuptdVM9UIy0X3xktup7n8NOQorvThW+PdzTslo3WKy3Ypo2gF2BXZT3ZPEiiIYVqqKhqVZZp7wsBLr12KTouToKk9a7iuhXcLod+CElJCv7FdUGITFgrhixBX2LEUbrCIH0dJ12Hme0JMWcxreM1D/LlmBDSd6XNOYwYFcyPayeNb5cmh80NE1uf30IYhOg1erj3a+6FXtcTmnp88qWiKYxg6a9Fp8Oq9XgLp6qrEdk8rpokryXizufu4M6fsX/l9TLKq2UW0LXtKIFWWi7hxCMnACAai67X9ISfAtpv/+hn9H3bh9WwEHphQoR96NgoYG1ZsJts6haAhG5R3A4Z6wYTca8X4XZd+Pbwc2BUZXj8txXVNxEWYXMW1uZ5pXFliMzhAshpsiekIa40ruzpBCd9HuUOjsh6Rsaes2EQMm0+Z1Ax3Gv2IKsytIoG806SRE4TzZn3fWxaFR/uYm6ZoBLFUnlpIGXRr9jhfks8rgncANamNRji4IdQi2qUmKqeqUa2IyuhZugGFE1JJPOzbHHW8AjeTsaTY77rR3GV3bLZBL7+1Cyuy9ptdBF4AYq1Iigodl7ZQXm1DGvTyhTEnxWOQwJqHEadTRVAM/UaAXAfAG3WB7JoFvkwIKuMLfRCqIYKAjJy+lTECCNE5VRlIi2XUcKpXFivsFxgC1aToZaZYCglNCJaeGap9UoLpEBAXRplhbSyhsAO0HqVEUBqWYVW11AoFeC7w9VIQ1MQbH9Qfqczh0aSJdz69C20X22zFoptG+feci5y4swNc6h1iZcOr7x2BXc+dwchZQ4bbyuLX/ehyoKMqqmoCsLx0N3pDo0YHlWNkC7FjqvkL8K4HDIjd6A25zDg8tOXcfmZywCGM1RpZ+fyM5dBQMZX6ozob86bGjWW1IllgoYqUlo2Orc7LFBoWNBX9Gg6C5FItK7SoutpxyrL4Y+qBrVsR4MHE77tw24zAU/P8uB2XXRbXRhrLCCRFRlEJgg9Rg7HR6XLqgwikajypvFigwmsUwLVUKEU+5Ml+uuUn69W1uBYToLIjeuCJCbXxKoG8s5jmnWYdb0UXcltd5sH+TJp1v2Q466zORyzqAo8KEx1r/Uz0b7tQy7I7N6XCarrVVgNC747GNltUzsSGjZOsJZSa8sCDSh828f6I+uw23ZkD+Ii5qOCnniWPn7MdstG+0Ybnu1B0RWUVksoLhdhm8x/qaxX4PZcVE9XI90M3vLJ7XC8CinwYpn7gowTD59grWsZpC4/vub1Jggh6Da6KK2UmE3XBqPM08+KQrnAkmqqnBBvjldeTlIZnuX/3KXkzoHbnEW2eV587CIoaJS0ik+zSk+/uvjWi/iJN/1EbvXrKExiI0Y9Z7OekXpNT0x9k4jECJeAQikpQMi+l7eLRq3cTv4UUT4NmBCCyskKujtdtG62YN2xcPtPbyMMQ/S22SRinkDubncZOVRlpAi3b4EbMG0d04HvMHsFgmid8opp3iqZl+TzHZ9NBC0VMvVr4teN26J4O1ngBdj+wjazVU1WeMBjUN/xWYVk3y5pZQ09rQcKNgVUr7NqQT4ZNUsQfxY4JgmokRh1NlcAfDsh5D9ltC7Nq1XpUIgFLtLRSTsJkirB22IkSqGaXXHDt5skAzIu0xUfV9zb6cFpOTBOG6ifrQNAQtwvclKaNsxNc7BPmxEVtXtrkHQJkAC9qsO8Y6LxQgNrD6+hdqY2VBWTPvc0wWM3bfTaPXhdD8VaEVqNCZ7app2YThU/p/T0ikKpgG6jC93QcfvTtyN2mbPxoxj6RAtX0wahLFCDDBSXi5HR49UIpeVSYixg/JgOC4lyyIzcImzOQnFh9QIkIg2VHXMBwXRm68LqhbH7zCKGx02NGvWb57U7xEEoI3BAMURm8GOIi67bLRvWVSsKPNItWIMdD6oGs9Z248UGWq+0YG6aUMsq2q+20d3uwm250Fd0qAVWIajXddTO1VA+UR7dKur4kaiy7/QdL4lg9+ouSidKAEXU0imrMrMBOdc+Mbkm/Ru0bCjOYIKL7zDnjF8ffp55k3ni/z/Jeh3l7O41uDomWfe7zuYAs60KPAjk3WvpezcSFSYYZMcB7H5hF1aDETVKYfCcB5LaXL7jo3q2GlXs6Us6jJNGlKSK246sSZzp5/rQMbeYjUc42F4tqtANHVvPb4HILJvNNTe6m13U7q2hda2F3m4PxaVi1AbuWi6K9SJKayVQMtDzG9W6ZrdsmLdMqIaK0lIJalmFvqRH5E+6kjKdaBsisKaoDI+2OUT+Dz+eA7Zhd53NufTYJRCQzLiK/32lcQU/8aaf2PO9MYmNGBn/5DwjuV/iayyOqGm1qDWJy1F0t7uZ1Wvp7wcQTQNu32xHgyYK5QLksgxSIPCaHoKNAK/+yatYf3gdxrqBjec2QCSCxksNKBqrwPZ6HrqNLprXm1B11pppt2zo9YGN4t/FuxqME8bQcAiv58F3fPR2euhudZlGWJoIjl03HlMpmhK1k9ktG/aODbJEEkkufhwbz27A2rKgGRqMdQOnHz2N9q02KicraN9qs6RZZUAsz2MdHpME1EiMOqP3AvgQgL9MCHkRgINBV/77CCFcvyatbbNnHAaxwMPk6Ci6EjGdAHIZ3/jnuWORx3CPu6kVnbVsFOtFaIYGc9NMfG96Wg3/zvJ6GXJBZhnt/mv82HdKOzA3TIQBC/BCNxyqirFbNvyNwajPrIoWbqjqr6lj+4Vt9LZ7KK2VYKwaMDeZAjwXMeTbcKPFp1cYJ9lniUoidjmqIgBGMvSR09ivBuJlkLIuw27a2PjsBswNk03dUGXYbeY8Ld23BL/nRw8nu8XGmBaXimN1jubtbBwyI3fgNmfRSGeoAObwp0d7pjNb45AmRxLtOVV9ZIthGuNIv/gEmpNvOInaOTbJKZ3J9R0fcNj+zC0T1GctDPqSPrbysLvThd2yo8CDB1+hF4IU2LFJBQnGSQPFtSLclgtKKaQCEyK859F7IClS9B28Fz6ezeZtS5IiwbEcNm5TZZN3ejs9WBsWI5BbNggI2q+2sfya5SHtrLzMvbVpoSt1ISkSG5EeH1VKgfZNVsFkt20WoPWMyNbEJ/Okp/Twtle7ZaN1vRWJMfJWMPOOCbNhQjd06Ev6kLM7zoEeqY2RRcIdLdx1NmceVYHTIK23NTEpOaaqjVfzcVFhfVmPCOGzbz6b1JroDAYUALGqXE1JVOzxY+MVeHnfnTd1ZkjEXVOgqKzSplAqYP2N61g9v5qYMmrv2pBkKSJJWjdbMDdMLJWXYN5i/pjdstHd7DKdwZCicqaC7avbUctC+jpFAW0/6JRUCWpFxdpDa+zzMTudrqQctcYVra8F4jMtEN7eMcqfibeJLTqJtCBy6a6zOQBG2hDu16TbJae9N9IxQ7oVcNSgmXEJivT7AKvm7Ta6oCHF8v3Lmbp1nFThbeNEYpUwkIHGCw1oZQ03Pn0D9dfUYTdtqLqK4lKRtUy5fb0t04VW1tC+3YZx0kB5lbWfl9ZKsHdtaHUNXtdD62YLgRtEgxICL4B5x8Tu1V0QhYCA4PSXnR7o7fRFjCVZwuoDq+g1ewkh9ejcY3FCvP2c/1eraPA6HijoUJLLbtno7fTgdxlppBZVLJ9fZhVBfb/NMZ0EsTwPHJME1EjknhGl9HcJIY8BeA9Y9Qv3Wml/O3XcPo4aFu3opBGfkDDpzTfuATXJTc0f0p3bnahPUq/pUSY7/YBXNFZqSCTCMj6x0XiGbgCU9WjLqpzI2HMHo3m9CfOOCeuOheJyEeWTZaw/ss4yTRlG+eTrTqJ+lhk7Y40ZH3PTjLJydsuGDj3hYHExQqIwR0UiEnzbR+dOB7IqR32foxj6+PnGjZte1eF2+k5SWUXBKEDWZHR3ulB0BV7Xg6zJkdNobgzGCmax4ZP+lnbTjpzUvfadHiYjdzfaHCCb1IljWjInjazfmBMik/zmk1R31M/Vh4Iz/l40YY4AzZebTC+iH7QEXgDIQNkuJyZyxY/dWDew+dwmm5RnujA3TdTPseo5SZVAXYpeqwe1pLLWUKiw7ljQahoUXYk0cuLtZ4EboHGlwQjkqh6RSnpNZ/susnJqXors22ysMpGY+LpEJFBKo1bLUaQHt3HdnS5CN4S2pMHQjWT5MwGqZ6vwuh62rmwh6AW4Fd7C0r1LbIynF0aTeZyOA1CgsDqo9rGbTPR96/NbWHrtEkI3ZImAJnDl966AqATUo7jw9gsJAiqe6c8ilw9bNn3WuBttzjyqAidFPMBp32gndLamva+yKg8VTYGqq6ylyVYiW8UrV/gxcNF4GtIhcnUvlc58cAInW/O0xuy2jcrpCgoGm5YXTfrst4dxHwoAwoC1z0NiI8yVggLHcyKbZJwyYN4xIWlSJIScblngVYyhHyIM2O8qFSTYTRvVc9Xh4C1GLHGbk/c7clIucAK4XReFUmEsmROvTgAw1wBuEiyiQvlutDmTYhYJxnjMEHohrIaF1QdWESK74ixL7zL3+PRh/6Zzq8Mm4hWkqHomPlo88IKopShwAkgFCb2dHsxNE9Sj0FfYNLrKPRU8sPYAGi81UKwXWdVKtR/HUAIKyvwdOmjBKq+WYZw0mN3QVWg1LTEowW7ZaN5swtw0mVbophn5OXERY4DZGx4r5Z07l4nIElC3WzZCL0xMCY7eVxWo6yqcphORvnbTjqYir5xf2ZP0xLQJ72OQgBqJkWdGKf0vYKPEAQCEEAWAC+CvUEr/tP/amwD893ke5EFhkY5OGnt1pOMPqKx2HyD/pk6L4XmWB9dy4Vpubm8jD9a0mgbfZoHdUFBWVHDikROwNi3QkIJiEFi5pov29TYkVYpIEBBgW2PZJl5JEO/P5kaZw27ZbLymLKHxUgPUZ4aPi4Tx9gjeAlE7w0omXdtFd6OLEGyk3vJrlyfStEkHyNxBdE0XnuVBMzQUKgVUK1XYTdZ/7/Zc1M7WohLrUWx41m+Zdjbspo0bn7gRZdLOvvnsvkidw2Lk7jabw8FJ43RlDgA8/uDj+xrlCWSQklP85qNKmXP70VM91wBAKYVSVKCWVNhXbRQqBdAuRbFWHDmRy3eYwHqhXGCVLNfbkGQJhXIhEjinlCIMQug1HaUTJRinDRSXitHEJ0VXEiXXkipBKSqRwwcZWNFWBkKD/Yxa7UwNSpEJp7dutkBAsPvKLiRdYkRQVY8Ilfh14MfNvzP0Q5SWSiyD1a8IzBKAbt9sw9pgrWjta21QnyKwAxSMAihhlUqSMpjSwwPBwA0QhmE0HpQHYmaDVSNWT1fRvtWG2TBhnBwIt5sbJrqbXVhbFiqnKiNLrRedTZ8X7jabM6+qwEkQrb++aHhcZ2va+yod/OlVHX7PZ1VojsKCBORPowMB7I4N1VOhGVpEaE6UQOtXFTstRrD0mj0U60UAyRau6LxjZEZ3qzukcZOuWObJH94yhpBNxaKUwtqyYN7q+1yOD1VV4blephByXO+rs9FB9XQV64+so9fsoXpPdej68OMK3GBQRRWD3bRhbpmwmzaK9SLMTRO9HdYK1tvuJZJp6WvOf/vSMqtsppRG+h6LsimLqlC+22zOpNhvgpETmL3dHpy2g5XzKwjcAM0bTRSXikOkzX4SFr7jQ5IklNfLICBsUqYkRYkWfg52y4a1w9qNtKoGWZchSRKMdQO3PnkLve0e616QCIorRdxj3IMgCCDL8iAe61dAK5oCraqh/WobIEDnVgeV0xUALK6xW3bkF2gVDaBAb6eH9qttyAUZFBR2c1iPi9sbTsTEbYLdtLH53Ca6O12mSWb7WNVXh/zJ1QdXM3+3eAV3ea0cxTz7TSQf92TTXjDV2VNKfULI14L1gHJsgokSH3ks0tFJY6+OdLxFgWfA4u0+ud+XUb4sqdIgQ+36aN5osiAm3kcee0hz0eAhxltToFW0iA3nbCyRCIpLRbRvteFaLnqtgWK8oimR1se4/mx+zr1OD4EbDMRggVzm127ZTACsHqJ9o43N26wCYOWBldxJYvw6RefWN9bmhslKuQGsv349at1oXm8yh2qjA0mW0LzeRP1cPRKvHsWGx88ry9mw2zaITKBVmKNmbplzU4dfJI67zeHgpHEWnnrxKTz57JNztTfjkJWZyhLaS9uFQrkAu22j+XKTVdLdMiOxv+q5KjzLg6IpuXaOkw5um02r4RVwXJSdExfWlgW1rGL35V1Uz1RROVWJbBlfY77DxPl4yxT1abQPpTA47nhGDWA2RCkqkS2pnqkmWrUADGXr4wSPsW4gDEK0brZAQbFy/8qQXfJt5lCFNGSTMxQZjuWgfm8dalkdBKx9ksl3fLRvtqFVNbRutNC63oJSZD3tsiYzsdP+SHbqUbRvtUE9CmN10DrCyeXavbWh1trodz9cLZkHgrvB5sy7KjAP/H7K0tmael/6QPMv/necFGlebw5NXuL3feiHaFxpgBACp+NkEppZiFcc7r68i9JqCY7poLxShlpUE7Yy2iZFZqSn1UTnoCkJwXq+Juvn6nj1T19F65UW3K6L0Atx9s1nUTldgVbR2ARQLVsImWf3FbV/7fuZeEVTEjoaXMQ0EqLXmQ9jOEZURXTjEzfg2R562z2c++pzIBJroeeDIXzbR7M5fuiG1/Ng3mZ6bLtf2N1XQmo/OCwVyneDzZkU+0kwcgJTVmRYG1ZEashq9uT3SRLgQLIiLd6eGdc4LS4XIanS0LNSLshAMEg8L68tw++xdvHVh1eh1TQ4LYe1U9tepBm4/eI22q+2o8SV7/QTMNtd7H6Rafr1tnrMRyix4Q2GztYuAcHNT95Eb7eH6nqV+V3VAvS6DqthDQ3IiBO4vO0t3sZqbpnwLR/F5WKk1ZNuQ8373fLInkkKC0bGrHuIkY+7GPvUZ0QpfSb19xcBfPfMjmjBWJSjk8Z+HGm9zrLGfCRn/GbPHdnnDI8V1ioatl/aBiEEW89vwThtoLRcQuVUJSI9IgKpXwKYNb4v66Hp235EbPCpVHbbRvOVJiOSbKYqz8uhK6cqA+HCjBJJ7tjx8cqqzhwr3kOfdX0lRWLBl2lDMzQWJI6YJBYnvXhpdeiHUfWQojP2nF8XY91gQob1ImRNjsROeSvZ2FaXEc4GD+5aN1vsYdG0F5rpmieOu81Jt3OmkW7/PGhkifrF1wAw3GIQJ5adNguWiitFGCcNllnusaCKEMbIxCdyxbNDPPg68cgJFFeKKFQLQIiEKDtvWVKLbAofCCIHKH7MgRdg94u7bFqFKmH14VW4nb6WVRBGQZXdYmQOkQhc08Xmc5sonygPgp6U8CiAqHyYE8lpR4OAIHADBH4Au20PVQLw81w5v4LmtSakAtO4UMsqZJUFaaWVErrbbCIGkQg79uUirA0LRGZB8fob11E7U4ucUuOkgQtvv8Aqc1aNqDqH/0Zez0PzWpNNUyypQxWD6aD5bsFxtznA/KsCsxB/pqUnJu0VnDyNKvz6RIVrulG7heu4g2Ctf9/vvLTD1vRJA3pVj+79XB+p/3q01n1WbaeWVAR2gF6zB0lhvku6fTTRpq3KCTInPlWvUClAq2mJaTncj5FkCWpRZaS0yXwlvc6O21g3ooDM3GBtFcYJI5Ed1+7TEq2xPCAK/ZBdC8tDcbnI7Ft/omjjpQYTsW/aLAkn95NwN9to3WjBOMmIeUmW2NAOXYmEYvOGbviOzyoFdm02rrhhsdbxGKFzkIHXYalQvhtszkGAUAJZY2usUC2gtFoaXk99TJIAj6rr3ACNFxoIAnZ/1++to36untA4BZC4b+2WHemRbn1+C77nY+PzGwjcAHpNj4hdVWcdBv4dH07bYTo4sWnCIOw+JYQgDEN4tgfP8uC7Pryul6hA5qTw9kvbMDdNlFZLzHbobEBEPIFmt20EbgB7l/k8TsdBea0cXS+7zap5NEND904X1o6FpfuW0LzejM55HCmTviajPhOfBjqq6oYLM/eavSHt1DxkFS3EJ4EdB5LnaB/9nDDK0Xnb+bcdSFC1l8xBQgS4n/GNE0IjS9RiY4V5C5EkS9HYT0lifdryCTkiPQBEgQkf28cz15xIik9ziRMr6fMDEAUNnTsdyLIcHZtW1aBVNEZeEKC72R3qvVd0xk4rmhJll3gglHd9Vx9chV7ToZZUmJsmrC1r5CSxOCPstB2AIFE9JKlJsVNjnbV9dG532LhkTY6MFX945LWZxI8z6z29ruPkG05i9+VdVk2lSBNXcR13lvooIYvMkYiExx98PNHmuR9SZz+/d9pm6HU9WgMAEgRmmgjg2R29orMJBzW2jvnEvIke5jwQcpheVf3MIMvOv9M4wUQCwyBkPeX9YCnedkAkVpFjNSzIBRlhO4RW06KJfV7Pi9Yubzno7nThd30oJQXF5cE65ms2cuD6gu1ArGS5H/R5PSYU6FouJJn1v28+u4lCqRDpjcWJ8dALUdAKKK4VIUkSqqdZvzlvPeVrvFAqsNaLDRNSQYJclOG7PjSqDWUYjZNJIif+G+k1HYEboFjPz7zx80kEzTO2G8ImHTwWVRWYVe3HX58W6QEDdsuOklPFpSK6W110NjqRdoPdtAeVKKcZuUxD1qLNq1CyfKR4y9TuF3fhWi7TzlAlpsF3sozSSom1YWW0j8b9Hb/HhIG5QDOfqhe1Y8am5XCCGxSsVbVtw+t5qJ2pJTRweICz89IOtl7YAghQXi3jzFedyZ/QZTNyqnW9hV6rx/Q3DGbXecUzoSQivOWCDL/nY+eLOyzgpDRqsc8jt4eqqflvT4Hdq4PJY7zCmR+baKUQ2AuyCMxRAyD4ujQ3zUQCPB63cL/f7bnYfnkbxUqRaW9ZLHnE11fWM4w/172uB6kgwe/62PnCDkCAtYfW2PCUfutmb7sXVfqUVktD04R9m1Xlej0P3c0uay9fK6FyupJIxPiOD6fjQKtocDvMBykuF7F8fhlhGCbavfWqDmvTgmf3WzZLSsL+8DZW46SBIAhA0O+YuG1BkqVMfyCd+ObxY3xQA59GqOh9UXWXae9w/yaPgIvvP3ADmLdNlNZKKKgFjEO6Gmv76jZrbb/ZTk47PsK2JvfICSFfBPBOSulnJtkRIUQCcBXA45TS52Z0fAvBKEfno1c/emDtD9NmDuI3LDAcbI1UkCdAea2MkIbwLC+asrCLXUiSFGWke002WSo+9s5u2dEi5P2bPJhJCw6mg7540BV6IYpLRciqzAygykZ3+raPnZdZBokQwqbXpHrvI2OqK1h/3fpEwQEPOs1NE71mjzlUeSqAiLH5/RGFUkGKhAu5UXfaTuL6cueRH4fTdli/7D71AwD2IOFtL5NWcR1mZ+lutDlXGleGyBxeAZgme0Ia4krjSt6uIuRV1Ozl906XtQKpiVkjxvjGR47LmpxoMwAwZAuy7FO8BYCARJNcAMDcNKEUFChFBSffcBK2acNYzRbs83oeOlsdBHbAiKNN5gioZZURxlUtmvYgSUzg0LfY9/NJM/HgJm53EoRXn9DhVXhez0NgB+jc6sA2bSAAwjBE4AfoXuti29uGoirQl3Ws3M9au+550z0olJloKs+W86CpvFaOJgbyFhFKKXo7PRCQqAVsUii6Aq2sRRWEk9wDs9bRWaRNuhttDnA4qgL3+7vz1ic+YKC4UoTXYz4Cd9KN00aUYS6UC1HFCQ0pW69LrC1x5fzKSB8prv0T+iEc0wEhBJWTFZRWSzBOGInJMZnVAH17cfvTtyPtu1NvPBVN1SOUwDhtRFU0aYJ79YFV6FUdO6/soLxahmZoUTbe3DBZpc3LOwAAVVNhNSzc/ORNLN23FE3xS1933/Whr+gRscVtLp+kJSlSdC4r51ew+tAq7LaNpXuX2L4cf2iq1iSJSL2enDzG7WVeKy//rlEZ/6OCu9XmHAR4ojZdwT/qfuFJIV7dn45bjHUDXs/D7hd22aSmkD1r4z53ni2LV7h6the1gREQpjnVtLH60Go0hbdyssK0APuVd4muBsdngxN6HpsSd6KEYrWYWH+KzlrDfYetJeOkgeJyEV7XQ+OlBqsOXDMgn5EjEmjl/EpErsR9uvi0Yd9hAsy3P30b7Vtt9LZ7WH5gOTN+SRAnjS52r+2iWC8i8AIWD0GPphESiaC71YVclEEDCrWkJgilrJiGk/h8amlxqcjixDFdKPEqSd/2o5a59LTjo2hTOEYd+X0AxtNeA5D+Nto+jmfhOAyOziQYxQbnBlv9AAPImMJAmQ4Fr9BRyyrCIETtXA2hy66FtWOhcroSORx80fJ9x8WHeVlt4GSX36YR70cNvRCSzKYsBG4Ac9OEZ3pspF9ZjXrvA5cJl9rUHsrsx52XrCogRVMivR9JllA/UwcF6+2P99GmrzMPLrkDpFU11M7Wos+aG2x8epSFq7ER5nzUu9fzEHgBnLYDz/RQqORXBI3Dnqq45hyc7RP34S6zORcfuwgKisvPXB5q50y3f15868WxU/VGVdQM6dzEtBuyyoWBbJvC7+X45zIdGi05DW5UbzrXpIprLPDWBV4Oa26Y6O50sXNtBxJlD3PzlokTrzvBWkzPVpkoZz27DFg3mGAyFKC8XoZe1+F1vahV0W7a6G334HtsvGbQC+DKLvQlHcYpY6jVi1+vPP0c3jpWWilh9cFVtG63YO/aUcAJAhRrrDIm8AIoWt8Z6/mJscnAgPzmx+q7rKWCk+kEBGt/bg2u5aJxtYH6mfpEmhSJjGbGdIqse2DWOjoLtkn34S6zOQdRFTgJ9vu78zbF1QdW0dnogEoUhBCUVpI6NQAjtXtNVnGillS0rrVY+2afzIkIhRwfKb4GWtdbjNgJwmh6nt2KrWsgtzqYa9/xViPf9bH64GqUnOHrPItcAgFWL6wmdILMDSZK3LnVwdL9SyAKwe4Lu5A1GW7HhVJgVdPGmjFEThXKhWiKDrctnNgCMFTxbG4wzRtjzYBqqPBML3qGTKKNkUZ68lheK288ecjfy0sSHhHch7vM5hwk0vffJPdjooLO9hOJWV6F41lM38bpONDqGk48ciJRFZNnyxR90D2wgQ3IKhMnlgty1A4GsDXAq4U54ZDoatBY14Fvs9jC7bqwqzaWzy8n/Ci9ruOeL78HtTM1tG62oGoqXv30q6yiUJLgtBxQKUbc1HWsv2490gPlcVV8WrGiK7BhszHpbTZRs3mjicrJYd2xhB9ne7B3bIRuiF6rl5DqUHUV1rbF9L0KCqxtC4Zs4MQjJ3JboOIkfuCy5FzoD3TYRiUJ4r8xT4wFbsDiSX/vWm6HCeOO/n9wjYMJQcd/5PAiz9F52/m34aNXP3pgjs44jGODRwVbQPYUBj42l2eg+Bjv2tkafNvHrU/fgmd5sG5ZUFQlEvflwVrt7KAfHkC0WLIcnHQgyY+dOzbAgHDh43hDn2k86Es6TjzEFjwnT3Y2dwAJqKxXEqN3005C4ATobbPgiYYUS/cvRSKBjuVEBqJQKbBpGbyVTFOi9q541YFruZkPC9/x0bjagKqrrGWjrKB9q43yWpmVO5ouSqsleI6H5fPL+3JIpq3imndwNgPcVTYHAC49dgkEBBdWLwzZE/73lcaVicgcc5NNaIpEyoHM3zut3VBcYTpPaUc5z6Zk3fNZ1TV59igekPCJUivnVwbZrP5YYd4i4bQdbH9hmwURjg9VV7H+unUQlcC1XRCZIAwYyZtuG+KBX3m1jILBqoWcloPWtVY0Qnfp/qUoS2aUDNarHoSglDk+xpoR2aRRhBcQI7qdAdFdqBRwau0Uuo0ua51osnPsNXvwbC+axJU1VjX++2bpbRTrRXRudZg48istGPcY6NzoTCQ0Gs9oJgQfUw7xPIVDD4FNuqtszjyqAveC/f7u8XZMp+mgtFZKjMOOE8jcrkiKBK/LREx5Npwndkb5SHwNmJsm1l+/zqp0ez5CGkLSpZGCx/GEEij7Lqthwe8NpkjpNR3NJmsjiGsBJWxLTF+MV9DwIRBmg7WVaHUNp7/sNLOLTUaMB3cCeBbzNdLXPe27Da3tDH/HOG2wDHlBgrnZT2D1K3j2Kmw8qro80erecZiPtpqqoDp6LZt3lc05CGTdA9PcF1HSRPOH2gYVjbVdFww25KF2ppZb/ZFnyzjRkhZAjyfC7ZYNc9PMbG/mNqhxtcF2SAGrwdquAQxN2eTVveYWq2SWZAlO04niu/jkYkVXoDhs5HouwU6B5itNNuxlx2aE1FIpcf35uXF9Grkgw95hk7dkVYbVsGCsG6xicElnfhsN4ew4rPW1II2MaeIkfq/ZY8Mf6vog7kz5v1lVkvHr6Tv+aPt3xDDq6C/vcZ+39rjdwnFYHJ1xGMcGjwq28j7D2V+eGY47M3bLRm+3B1Cgu92FpLGRe2kNnHhpLJGGM2X8wdt4sQHzlsmcqpjAcnRsvMSwH6DufmEX1KNQayrOPXQOxkkjMaq8u9NF6IZw2g6Ky0XWi6n5Q46A23WhGiqo2e9p7ZNXek1nJZeuj8AJotaxzec3mViozogk3v/eerUF3dBRqBQSThY3NqCA1/Zgt202faeowjhpwAxMuLYLz2Q98F7Hi67ZRL/7DJyWeQdn+8RdZ3M4RpE1k5DG8b7iOImaW1Hj+AntBrfrQpO1iW1KGnkOTXrbeOBkt2z4rg+n6URkCHcw4o4Fnyy184UdRpK6PtyOi+5uF9SjKOhMT6Z9vQ0KCr83yFRxnYioWq4go36OBXdEZoKg7Vtt1s7Zd4iIQiDJErMzbaZfYW1aCc0ufi68rTJOZnNnUFZllFfLsDusioZrdBCJoLhcxNr6GuyWjdbNVlTJFy8TH/qNU78Zv4auyQIzx3RQXCuieqqaKTSa+9vp41vzpiWOp8GCbdJdZ3NmXRW4V+z3d4+TLFz/QtGVodbOeKDBxdKH9ClivkJc1yf9fXpVR+Cw4RHWloWqWoV1y4rsV5xIyqo6UYsqjFMGlKKCoBewkcrXmpmVlHFCPE5+2y0WUHK77ts+m9JHAHvXRnG5CEIJiEJQWitB0RSU18vRREB+3eLTwfKufdRm3m+lL6+VUT9XZy2whMC6Y0HRWdtnPECcFqOqy4FBUkJSJABTaEMeTtx1Nmc/mMTvzboHgGRlV1r+YNR3GetG5J9wjTy+Ziil0bpNEy6+6w8NFsg6D+4vpGOo9nNtWFtWpFuTSUhorE26uFJEb7sHx3KgV/RIG6Z5vcm6LPrJOUVXUFoqwTM9dF7tQCpK0SThhJ83gkDm5IfX9RD6IWRdhl7XEXhBpAfWvN6EeYvZKH1Zx/oj6zDWDZRPlqPzJYREVdQAsPLACuyWjZ0v7kAtqYnWqSzwY+RTguMaYln+b5xYS99H8/RpFoXcs6GU7tXoHFkcFkdnHCZhg/NaGbI+H5XN9m/wzPG1qgK37aL1SguBzwQHT37JyUwhQWPdGHwfTX6f77BxfaqhgoCw7FLfIJgbJpyOA6fj4MTDJ6AUGau89JolBH4AvapDKSaNT6/dg1yQsXJ+BXaTZazimf24I1AoFdDb7qG304NruWjfaiN0QlTPVAHKWOXm9SY6dzrweh60iga/68OzPSiOAr/H+t+9noedqztYfWgVVoM5NKBgJE3PQ/NGE90dJtzcudNBaa0EIhE0rzdBKQX1KIK1IHLOxj1k+G+U57RMS/QcVkN2N9qcWWEUiQoMV9QompLQbiiuFCO9hD1ly3XWjsg1EUZlWKJx5i0bO1d32KhtVR44MBn2ylg3sPrwKtrX2yiggPqZOipnKzBWWXBUaVbQudWBa7lwLTfS2ok7GQCw/ggLOHQwe9N+tQ3zlgnzjonqPVWm4aXJOPXGUyxzf9WHZ3mwd+2I1E0Ha+n1GA/C+Pdz8jo+ZUbRlcxe+dxrnPrNjNPMQe02uigusdat3VeyhUaHfoeUzdhr+8usMuOLskl3q82ZVVXgfrHf313Rk/oX8QlSwHBVm3HaiIYhpO9bro/XvtlGlSQFlKNj7JPLns0qfvW6Ds3QAGnY5iYSSv0hCjypJmsyHMr09Lh2Fa9KjE9s4dfHbtkseLptMpuyYWLl/ApqZ2psMg0pM1sUUNTO1BCGIVbKK3A6DhQ9e0DEJELn3K5zAVFzw0xMwWrfbiPwApRXy5EdATBUgT3J75geOR8//6zkIf/tRmpDHkLcrTZnL5iUrMt6fgGsakWSpaGBBnmEC/+uXrMH8zZrMdy9uouzbz6b+5wEkLAxfs/Hqr6a+A6+b8d00Hy5iaX7llCoFBLyEI0XG9i6sgUaME08pahEeoFxGGsGSmslhEGI0loJtdOssodP6tJqGty2i9q9tWgClnHKQLfZZZpiKotlPHvQ1p4mpXirJdfyWjm/wtpRLQ8IgW6ji+5GFwQEWkWDtWkxqYyChO52F4EXYFtjBO/K/SvobnejqsZuo4ul1wx0vRRNgdN04HaYtg0nm9LXj6/5zKrv2O+i1bQocRd/Dhz3CVfA+Jaruw4H7ejsxSEe9fDj+0wbwbxMmN20sfHcBjMgqhIFCXHoNR21e2ughGVqavfUEhNREiJY213YbTsKUriTYm6YjE3VBlo5vseOXzOYQbDbNnZf2gUAdG53cM+X38PaKPyA9b6X1YSjY6wbMCUTYch62cMwjAx4iDDRvsCvk6zJ6O50UVxlozdX7l9JjHYHAFBAKbA2K0VXoNgKyqtlbL+8HWXs26+20X61jdJyCQWjgPJamWXuCmwymL7CnB6v64GAsJ7cYiEKSoOAtX1lMf2Z90lOwDVqKschrcIRmAMm0atJfF5PtjjmaehMCt7bTCQSaS1kPZBBB9lWAFg9v4peqwca0Og88jL3J7/kJIwTRlTRwkU8eVuAtcUy5aquMke/bUfkcQEFRgbHstTGuoHQD2GsG2i/2oaxbsDasJjo8dJA0BwBYHdsVsVUHWS2uK3ptZPToeJBWJy8Dv0wOta09kT8Wpl3zMSY8fhazvzN+r33hUoB937NvUzotKrnVueM0zuK6xeNy2ZmOUnC5hwN7Lcq8LBgVKVPuqot9MKo+iWd+Y6qfUgVhVIhMbI7akHtt1+DsKlTgRfAsRw2GCFlc+PVLZ7tgVKKzp0OJIVlyLdubkXiyOW1MtsopwuHV6I4HQeqrqJYKyL0QihFBaffeDoSjbe2LFZhGEoRWcwDl7ifOBWBm2q7AhBlyetn68ze9CsMQZFZgc1/i1G2YVSlYNpGZl3nQ9xGLrBHTHqf5t0DWdPa8vYR/672rXY0NdNqWDC3TCBEZgVIno2J++F200b7VhudWx14pge1rKKm1aJpWvaujVf+yyswX2VTK6v3VBNtXWmNnPv+/H0JQXG9rkeVipIioXW9hW6TJXokVYJWZRU9xpoRDUAI3ADWpoXWjRZqZ2owN01GfDlM9F0uynCaDtSSCqfjML2fE2yqVuV0BUsPLEFRWTwHt3+tmz04HUZKu6YLc8NklU5dn41ALyqQZRmSKkX6qIqmoHqmChpSNK83o6Ez6QmDWRqp8d/f63mDdvT+tL4soXU+4YqAYPeVXdRfU4dmaEehsm8sjvbRzwkH5ejsp1R01MMvT88ivW/f9rF9dRvtm224bRfVs9WEMYp/1+qDq1CKCkjItBs8cyBcFzkuseyW32NlirwtyrzFpjAUygXUz9WjbH5gByitlNC51UHjSgNO04FSUGBbNhpXG6icqGDtobWhyqF4AMmdLKWgoLfdQ6PdiFqvFI1Vz2w8twFCCBM0q2koLZfQdBhjHte/UItq1J/N+7h5ZY7bc7HxmQ1odQ0SkaDVNbiWizBkQSEP8FbPryK0Q1CZov6aelQ9FIbs/Gv3MUMduPm9nkO/d84DKxFYdlhgqUM/aiXIAvvEpBUyHPxBlxXYjNsmyyn3HTYqmLcxxu/nvMCfC9MVKoVIgyGeKRqqKupXtARuAEmWsPncJra9bRTKBVROVSIxdkL6gtBVnWnVbDHNjEI1KUCu19j7gRdA1VV4lof2zTYq91TgtJ1BG1RZQfflLgghCLyAEeTn6rCbNpo3mkAIVsXXd7ai4+2vWXOTOYPGKSPS0Bg1GvmL/+mLUIoKqEdx/1+6P1rj/LOc0OGYpG0l/tvlPR+y9ItGVQImiPy+k8SJNmFzBNKYZ5IhnqmN/82r2nq7rDXBWDdy1Uji1T5cQDkeBPL9c/vF7Q2QXYkSr26RiITda7uonK6guFwc0iz0XX+0fgWYveQJJbtpR/ZM0QeTPZfPL8N3BsMZ7JAl19JDI3g7hW/7jKBKtVckziOjHYpXONXO1qLrEgW3KRI7LhQ/zjbspVJwFKEncLQxKVmXdw/EtbPSAw2yvsvreXA6Dpt8JEtRtatS6PvrsQpoLigOikTlbJT04W1AXoDGCw2mPdNk61erabCrrH2LEIJbn76FwA6gVlQQQlBeLyfGoQ8RGvVkwobbLrtpo3WtFVWe1M/VEQYhuo0uAjtA81YTxmkD1ZNVyJoMe7c/fWvbgiRJkBWZTRGWJXg9D+ZtVpFn79ggDxEgBOSSjNXTq6ifreP2p2+jt9OLJvaV1kq48cc30NvpsYm+YYhCkSW8aUBBA9YS75kea1klVdihzZL/Fospi/XUJNERyez4722sG9Gk5DBI2p240Lpv+yAgMLdYR4i1YQ21nx1VHO2jP+LIK9+b9MGUd6PnBv9ZDjkI/K4Pc8uE3bZZqwHN7jesn60zIcDURJSs7FZvtwcAcDoOGxNK6NBCNTQjEgWjoFi7sIadqzvotXvobfUQnA3QudNB7VwtYdzsls3IoL6B9W0/0qlYeXAlEmTmlUFWw0Lz5SZUXWVtFK4P644FSikqZyqJc+EizL7D+sX1mo5mqwkiEyzft4zADSJR5faNNtSSCupTaDUNalkFwCbh1O6rJUZ/FpeLQ9my6NxjveB5v33WA4tP5eg1WRsZoSTqMT9KJcgC+8e4Cpn0Z6cl/HjrQuiFkFQpoXsFgFWt3WhHGefamdpg25Sd4hlfYJAtnrhdkLdadtg0qmKtCEVTouoXraKx/8dwFRJ/aHOYG2ZUzXL2zWfRvNlE8LmAaep0vWhbx2LtEpIqwW278CymfeV0HHS3unB7LohEsIGNKKvGv4uColAuoHW9BeO0EbUspNcnJ9db11roNXs4eeYkrG0LrdstGKvGyLU86rfO+r35GNas9g602NQIXuU4shIw9pzhUzmEzRHIwl5tzsS+0KhBEefqsBoWqz7reGhebw61RHBEvsyGyYidnR4oKLSKBvOOOfHxR5nh/vG3tlpwOy6sLSvKHic0C/st51n6FcAgcadXdIQem8hCYuU8/Fx9m2XCCRnYF3PLhGu6LNBBOJhm4wWMSL9/ZZjwwcA+GmsDvUTuuyhaagJP7FrwCmxOogOsskFSpcyEYeL677HaZpwNFDiamIasy0sCGbqR2WKZCwroVR2r51cj/0DRFTRebKDb6IKGNEpGRWvmXD1ROcsrdInUH4ktExinDHimB6XI2prKq6xN0ut6LAkFCq2sQS7KWLuwlmhfdE03GqYACaifHbY/vt1PqmkSTn7pSbSuteD1vGgd6TWdJdWNApbvX8bm5zfRvtWGZmigPkXgB3A7LvRlJm2hF3TY2zZKJ0pwWg6UsgLFVKAUBv6NXtehllSWWC8qMIoGVi+swm7bkBWZJaZCOogZl3Ws3L8C3/Uh63KkVxq6IWu9ajOt03iLaJZNyKvasZt2RNwBgzgIGAitc0Fo3upPQxr5L0cdR/8MjjCyRKimcXrybvR4Fike/Gc65KBMOM9Q4HdZlqh5vQlgMBoyvi8+ESVBKvT/5gzx9ovbUTtV5XQFYRiiuMzInM5GB9tf3EZ5tQzN0KJ9axUNrZstnHjkBJo3mpDAMlf2BssiAYOg0rxlRvsPnADWloXSiRK6m0y3RlIlFPQCq1rZZXo5qq4CEutdX3lwBa7lwjhhQFKSqupRv7g+GNOpV3XQgMJqWCjWijj1xlOwOzY6r3agllR4XTalpn62zo4z5/qkfwcACTKJs/l84k+6ZSJ+nPH9BC7TFzLWDTa1A9mTjQSOL6bJbO4lC2q3bJi3TDaudosJDMYnSXFtibjTHgUkPBvs+ENaDnmZ9Sgg6vmJFiLu4Nktm/Wab/fg+E4UOPCKwF6z3wbVP860QKndtrH9AhP2LNQKOPklJ+G0HPRaPYRBGJXj+o6P0AnhdT1ABzyHbWs3mVC8tWmht9uDBInZojsWpAIT9tOrOty2C0mVICkSulusykfRB5lASZEACmxf3WakrEwQuiGa15tQCgpqp1gVwF7Wcla5caK1lAx/3tww0d3swtqyUDk1GEk6qqqHk99pglpAgGNamzMtATRy/4TZBb2uT0Qq8O+XdRm7L++ifl89mnYZb8/Oer6nj93reUw/otFlI3oJy8ivnF9JVFRy29a81kTrdguNKw2U18qsPaumRZNbuo0uPNNDeb2M3nay1dNu2ti+ug3f9WHeMlG/rw65IMN3GDHVebUDfYVV0HVud6CVWYWx3Ulq0NgtG83rTWx8hlU1l9ZKOPOVZ6LJoulJiOlrkG4L9W0f7Zsxsv9sbehax7cX1TYCceyVrOPPM05KjhsQ4Dt+okJfKSowTg4Syb7js9HfBQkUrAKY2wM+2jxx3P34jCdIosqbkwZq99RYm9QG08QK7AB6RYdiKFh9gGlzOqYTtZVbdyy0b7ZBCUX1HtYBEU+q2U0bNz5xA2EYwnzVhFbRYJw2IpmL7avboCFF9VQV+jIjt7o7Xdi7NqxNC0v3LuHkG06yyVGrJRhrTDJDX2EkUId2YDdtUFDUz9bhWi5aN1twTRdOy4Fx2ojIXkmREDohXNNFrVrD6gOrQ5WMdpN9Lxdb16oaSitMviJL/zFtE7J0s9JTVfl1yRJa59pjXDMxXh1+lHH0z+AII32jTuv0pLcH8gmhPIeci24FXgC35cJYN1h1TV/Ar7udLKWPqkyQUwpY09Hb7UFfYtUpBaMAvaazyVO2j/b1NnMwtnpYfXgVxklmcMw7ZnSsp7/sNLZL2yAFdrzxc/Asj2W9SwVABjzPQ+lECdVTVSgam3BRq9bQvN5E46UGAjdA6DBVdtdiwRWlNCrB5C1bHLxlo1AqRNVEep1l8Pn4P17u6Pf8qFfc7yVF/OK/Ufzv+O8Qf5+LHto7rARy3MSI+H6408SZ6bzJRgLHF9NkNrOI5Lxy+zgooSAgoCTZs8ArxXiprttxsfHcBlRdjSpAONlcXi0nyJtR0yncjovbf3obxmkDkixFY7jjWTcuNGicYH+bG2bURy0pUvQQj68Xp+0w21QuABLQvt5mWbAWG39cMAqonamBSAShx9opq/dUQQmF1JZQXimj2+jCNV1oNY1pWZQUJg56ogzXcmHdtiCrMlq3WigtlUBA0L7ehtNiQqiqpkakLrfzRGJZulNvPIXKmQpq99SGNHQmXctZE3YAJFtL0yXMfYd25cEV9HZ7iWkdefdX3L5NW20lcPdg2sqLtL/CtSby7q14u4SkSMn9U6C71YXVsIAAUB5WRmpExb9bq2ooGAXWFmX7Y7PEcVsjyUxHwneYbdQNHcV6EaXVUtTuFK+otJs2vvifv4jd67vovNrB6S87DVVTsXT/Epy2g/p9daZv5bEpf17XSyS7tq9uw7xjwm7aUMoKHNPB8mtYVbFmaNh6fguhF2LnlR14podCuQBCCZSCkjg3APC67H2lyK7r5gub0Cs6rIaF1QdWo+o9INvnjJP9vuMn2sviRHLaTnHdsrRGxiwh9AWPP3zbx53P3UmQkvf9+ftGkjqjbJTv+JAkCeX1MquMk5CbaMkS8OUjyVs3W1A1NUq08JilfKKMzs0OiitFbL+4HSWSyf+fvT+PkWXN8sOw3xd7ZEQulZm13XvrvvvWfq/f6+npmaHEIemRQNILBEOgRcEmZEukZIq2BQuwYHiDaZKiYFgkAVqCRMugBIsey7IlWpRkmiYtUqRNUlxGw5npnnnT3W+/ey2ZlVvsq/84+X0ZERmZlVW3ql7d13kGg363qjIyMjK+E+c757coNABqPWzBn/owmgYaO40lWnswJdfO1n4LACAbMjoPOmIQ33unJwbVsiqLIVn7fhvjx2MoNlFTG11q5nAU9cF3DhBMCG2Txim8wFvUDLoCzdYQu7GovZjEEIeEzmv0GmKNFa+7QJODwR24aB22CO089FbqP1aH2UmQiLpmVT0C1NPROcvD2rVWDs9f17h0NmOM2QB6AF7keR5f9PfbWB/VG/CyyIri61ep/a9bAEbHwP5H+wKGm6VEqQBQOkZdc6dKHQgmATUm/BTRlD5D7FADxtq1xLHylGySw2lYamRxPRm9pcM+tBF5EbSetni454B75sIduKS9s9dA91EX7ilxIIsOF8bEgHPsCG2bHDle/upLzF7OcPIrJ+i83cHOWzu49717pYXunDiYvZhh9HiEzlFHTJMUQwEyIJyGcM9cMWHLkmyJSrY2KmgF/m/+v+7ArbUUXPreCw8fWZWJ4lIRJP2mFCvbnHNxXBaaXHViWkmlmofRNtA8bBJkt7VYk1XEmazJCGchJl9OIJsyJCah84hE2b2hVxI05udQRz+KnAhxEIOp5KKQhMmSDXcSJBh9MQKTGbxTD0c/f7TEo15FQ5Vksh2OgxixF0MyJSgmFU5ZmEHSJUiyRBO2NBMblTRMoVkaWYNGVOSZfROdNzpQLdLhibwITCVhvtb9FjSbNk6DT2jqPns2g7ljovdur4Soi/0YSAH7oS3cuPj3dZm1nAQJnFOnREmVDXmJWrrUnOETRWSiEX+Z+2sV2up1jG3Oud64LPKiuF5jP0bsx0vIkNqY95qLmyowYOfNHXTe6IhhVZ1GVPW9syRDnhJdQNbKz1gAtMbiZR083lyavSTXPUVVcPBTB9RQySgnTZ5NShP+JEzgDByige81EQ6pzlA0BXJDht7U4Q096JYOSZUw/HSI5v2maAzxz8ok0vkyDRNWzxL1X+zHkBWZPlOSIw1TonD1qOFttI3SZ1MbKl17L4bSUKgZtWPCO/Pgj31q9mw4hOS1GaeXLQ3QLulE9CrNmFfRrbzp2Oac64skTEpNySzNluqH0t+vYDbwUHSlRCWsc6zkx1kl4Mvd4U4+PkEwJFcqfi/7Qx+yKsPcMRE5ROGOAxJPlhRJ0C3BgNANIelSqSlttGioO3oygiRJsPvLullcZ4s3o0efjRDOSCvw8LuHZBqRY+l1vCGjWRpkXUbOclh9C2c/XIi64z7Rm7Ikw+jLERRNgdk2wRhbQjMmQYI0ShF7MWInxsmvn6Dz5rzxNd9vrRowchQizyG8plm5T6qpnar7zVVC9KvukbvcCN74rBhj/00AfxzAd0GPzX8IwK8wxv4dAH89z/N//2ZO8ScnXgVuuqpps4riU0WRVHmmAJag9ElIjjJZmgmdilLRFcQEs5vb5WlNmoYL+pKuwB/4yHPiLHbf7IrziH0Sycpz4qvLugxd1oUmDOeGN+9TIp2dzWDYBpr3mpB1WVjg8b91Th1MHk8weTaB1bdg37eRhRmQAdOTKaSGBNVSF/QDLBa6YtC0PYnKv0ujVIiIcVtkzdIwPZ7CG3o4VA9Lm6BSUQmIplkRrSC62n4MllPjiaMd1qIsLnmvvA7JqBrbnHO5uMzGn/+tc+LUUqlqNbQedgRKjUcVccbpDbyZkyMXOSKchoCEsj5LDVqI616FTiicXSRZWrLh5lMpq29h+mKK8bMxOg86NFGvCCAmQQJIlK9kXaa8duxA62gY/MYA7nMX3tjD3kd76D7qQm2qsPYs+ntNFkKH4ydjDD8ZIvIiNLoNtI/ayNMcO2/tCM0snn8kSYK1a4nrlkYpZE2GPyJdjuTjBI1uA9autbIRddkoCjFOn00FJbV1RDbMPPfX5Y6LcsoS2nANnfSubZY2jW3Oubm4bH7i92ISJOQSuQatU6RL1KGKZU1GGqXIkYsB1TpNquJkvdjo4OfjnDiIZhG5pDzqiCEMfz1fz52HHUweT5BnOdoPCfXX3G8i8qIlxI/dt8EYQzAjakOe5whnIc4/O0fzsInu210opoL+O33kSU5oAYmJNShrMqw9C7EXk5vNvCbi03PnzCHEcdNAc78Jra0h8RO4Z66YVPNrwR0Fk4j+PpiQNgWncRRryIuGkPx61An28/xfdCKKvAjOqbM0qb+O/HIVuvFNxzbnXH8oulJqSjZ2G0v1A49N7ivFWKYSXoTuq72/GEj+oQtETkTDoocdxF4sXHrzLIdma5BVGXsf7gGgWofTHpMggXtKFCznxBFDOKNrwDvzBM1ynTGOYiiCdYAMIsclYY25hb5w6+OyFu6AZC40i9CLnGUw/GyI2I2ReimmIJpldR8a+zEhDINY7Lc0iyQykjCpbSjx74lT01VDhbFjXPhMqdv3XBYtuuk9chdiozNijP0eAP8RgP8cwP8CwJ8s/PpLAL8fwDbpXENcdhoLlAVLeQFTFeaqUnzqjlFX4PP/TcIEelMnVXPZF8KnXM/CPXEx/GIIb+Bh5+EOuu91YbTnjjSKhDROoZgK9j7aQxRE0AwNxs7CvjeYBvBOPUiqhHOcQ2/raB40F8ifMYkgz57N0NhrwOyQGGrkRqJBFDKauhkdA5IsYeetHaRxSk2WnMSQ/bGPYBgg7sdwXjhik8PRCuI6mIqARvKpWxIm1DWf2yLzJPjiv3yBPM1x9ptn+Pbv+Tb63+ovNlZzcWZJo4m/d+qh9x4VTklUoIHMQrIg/el7S3SHV71XXpdkVIxtzrneWNfQq1KpVtKgnowFxTCYBFRIVB6O3FlKbxIVqXXUEhxqvvYG04HQZ6k2EfjGrP9uH/7YR//dPiABdn+ZA8+1raYvpnBeOLAPaSpVnbJxfjmTGVEX3uxAszS07rXAFIZoGkG1VCgvFNroNInmqTWoGHLPXGrWZIEQFmybROuUZAlaWxP5llMNuLMMbwp3jjoIxyFmpzNYexa6b3dx/vm50KWw9+3aRtRF393S91woyLS2JiipSyiCFbnjVXIKb3pvqlNy12Kbc+5W8Hsx0ROBKFuF1inmoSKquKhvMfxsCEVVMP5yDADQbX1lMV/XvOT3ezAJwEBipoqpIJyFSxoMXKCTSQz2PVsMepxTB/6YdLp4w4LnPaNj4Nv/rW9j8mJCGhNnLsJJiOnTKdUuQx+NbgNZnCGaRPBUD+6xC6tvlfQjeu/0lvQTkyCB1bMw/moMdV+lHKcppMnjknBqkebN87L4PJ16Cvcmg6WiYP/4yRh2uKhP+eu5eUTkRZg+naL1gHRCLpINuGx+ucpm7iZjm3NuJhRDKTUl12nobHpfFZ/vK993HW2LU9OzHMF5QEib+bMyzwj1ohgK7EOimPNh9vjJmHR4FEJQBwjgD31IGmkF8rUUjkIkHullSZoEq28BQO19zmsK58SBd+JB0iRCE7UMDH40AGQAKeicdELoBNMArQctNHoNYRzDtYF4foucCKquYvx0jmDWpKVrDFB+dM9cAESHzRISegfKhi7OiUP0r9YcQWgoUA2VTHGylJz/UI/mWSuUf0ngxF1sBNfFpmf0RwH8u3me/0HGmIJy0vkNAP/CtZ/ZNjYu5EuCpU6M9sP2UjdSiE819aUFsG7DX0WZFIVP+QLDBBg/G9PCjDP4k3mS0ZWS883uB7tUDMlUFAUjohilcQrvxIPRMSDrxNV0z1wgJwoHQC4JWZYhizPRJFIMBXEQAwxgKCy2OVIp9ojuFfsxlFRB+w2C88mqDHPXhGIomDyfiE0kT6rDT4ZgComTdt/p0nU0lnmoiq4gcOjzhLMQwSjA53/9c7Gh5Iged+giz3LsfXsPOcvhj3yiNBScLXjyLlLekuB6ksbrkowqceM5hzHmVH5kAvg/5Hn+L85//7sA/BkADwH8fQB/IM/zx/Pf6QD+LQD/JAAPwJ/M8/xPv+o53USsW991VKq6+wXAkh0tn/pUH459Y3mSFUwCmB2iASzpsxjlXMThtpIiCWqAc+qIxjI/Lte2Gn4xhGqraOw0EHkRgmlQmu4WkTy8waroinA5CM5p+pyGKbTWwqEvmkbQmhr0ti6uBd8oMonB7JrQWhrs3eXmaxIkOPtNgiSPPhvh8HuHhA7QFKR+itiPoVu6gFgnnWQl37vOXSwYByVBVRFzt7EszeC8cLD30R68U0/QHq5rA1N7j+RYK376GsS2zrmFuCxadB1ap9igXCfQLRrGHv19OA3Rvr+wHV+HPAPK93sSJnAHLrIog9E0YHSNJdh+6Zx9oliBAZPHE8iGTBbFpiro13wDx9f38195DvfEhXfqkUX6vTZ23t4RVun9D/sIJySaOnk2Edpi61ADrXstKIYiLJd5g949c2Ef2migsXYzu+q7Eo23Qu4uXc8LaFV8o2y0DXJKnW8aq7XKdTRjrrKZu+HY5pwbimpTcuXfXeK+KuYGjsRVdKVUzxgdQ2hbBZMABhamJ1z+QW2qaOw2yChmMEM4CuEPfZg9E3pbFwMY59QRAuahG4q6Ko1TeEMPcRBj8mxCeSFOkCbkwMskJs5paT3O68HZyxme/dIzNHYaUC2VEDhDD0xi8M48mH0Tw8+H0Js6VFMVOqGRGxEF9ags82DAgGZrCJ0QRodoXJIiidqgKDVRRAsVG9DAgg7uj304L6nu47bosipDaShIThMYOwZe/upLtI5akFV5aVBdzDv+1BcOo0XU+aaxyT1yF1gQm77rBwD+5/P/ziu/G4F4n9u4xhBcwXnzoA5VUeQkrhUsjRJ4Zx6M1vICAFZzwasbQQ5drnKhOXSYbzq46FYSJuUGkAQhjhe5kRDny5IMSURTOEkj15fm/aYozpIgwejLEbIkgz/2sfPODqZPpzj74RnCcYjGbkN0fvnnVXQFiqWQzXiSInIj7H97H+17bTgDB4lLKBzGSPiUK9snMTW9JFVC7MYYP11MlLjeUHHR2n2bUESjQGyYxk/HMFoGueR4CbSmhjyhxlLzsFlCUPFNapGWxS2FV3HJN00c/O+Qv5auVzeec/I8F+OWOX/9GMCfn/+7D+AvAPiDAP4igH8FwH8A4LfOX/LHALwL4A0ABwD+BmPsN/M8/yuvel7XHesaeoqx7FwH1N8vS3a0BTHlqt5KdZJV1GfJMuKzI4fYCIkJ9lwglDsf8Ea1/8KnTVScgeU09eabH1mRkQUZTj4+AXLSy+DTXfowc5jywIUkS9j7YA+KqZBQ4ZRyThIkOP61Y4w+HwEZ8OC3PiDo8xw95517ohDh1Cp/7COaRRiMB+g87JSKlBId7OUUpz8+RTyJyWEioJyAFHj5/ZcAgCzLcPCdgyUx0Dp3MSVQBOIoT3MhFg1AuI3FAVFvja5BTamKc8SV76WanBL7C3HWVeKnmxzzDmywtnXODcdV0aJ1aJ3qs6xYpBefq8WmTBzERPtpKEt6XnU1jzi/yoZk7/09cr4yFGF7WxVZ5v89/JRoArEXwxt4MHdNIY6eeAmmL6Zo7jcRuqGYXIfjEKqpovlGE8hoXen2QgPLPXURK3O0sL5odPO1WdL2qJy7vWcThSrJ0HzQxPT5FGlAQzQ+wKp+Z3XrszrsW/W91tGq6milikFOqYlfLzZ7Xc2Yy27mbjhuPOf8pAyurhqb3lfF3OCPfUweT4T+5cF3D9B7t4fxkzEiJ8L4qzHShCjWrQct2Pu22GOFsxBGiwZpo69G8M98IANmzgyyIUPW5cX9byhCioHl9EA12tRE5iYMnMZo79l48fwFVFVFNI2W6jLxOUJy63r5g5fwzjykYUquVj1C8SV+AkmVoBoqGGOigcMkBmvXqm0SFUO3daQB7bkAcjDmdWJVaqLuGcARe8GMaii9qWN2MkMwo/0QH+5JCu3RuMbQUj6Za5k5LxxBQd95e2flfvpV7pG7woLY9B2nAPorfvcIwNm1nM02ANRwBec2c9WH6eCTAbIkQ5aSLbgkSyXBUk41iMMY/pmP/e/ug8lMLIAilWn6bAoApSludSMItqwazqcyjXYD0SSC1tbI8SVMCSoYZfRARw5Ls5CodEzkgKIqmI1nSOMUsi5j76M9QCIr8ka3gemLKU5/eErnw4DWfWpEBWPaTE1fTJEFGZjC0DxsQjZkBNMAsUvioowRLNpoG5geT2H1LZhdE83DJp7/8nNyoZhG0G2dLPwmAWRDRo4ckirBPSX00OzFDM3DpthAFr8H+8DGt3/Pt/HJX/mENqxhhpNfPyF00rOpEFTtPOyQ6ruhlAqw6vGUUKl1oCl+75skjrpmXJ3YW/U1d2RjBdx+zvm9AE4B/K35v/8JAB/nec4bPH8MwIAx9n6e5z8CQaH/QJ7nIwAjxti/DeAPALhzDZ2LpgtVhEwS1osDdh52oJgKOb0VXA02eYDxB6Jz4sB56WDydILxl2PY92w0ug3R6CyK1cVBjNANkee5sMy0dizkyAWlB4BwZho/GUPRF9bCPL8xicHokoC71bPENJufV+In8IYetIYGc8eEP/AR+7GYxvNCpP1ggXzEBPDPfciKjNEX1GxudBsLEUFOB3s5hfPcQftRG86pA9VSSduiaZA9+MCFJEk4cU5gtAz0v7V8y/NmfRInohFWbBaNn43RN/oit8iaDLC5VtFc0PW6mjlFtFDnIT0LeAOa67dxGPZryE2/8Zzzk765ugpatPpc2mTzVZfTrL5F1ObGYu3XOb1VjR4wQUn7rv2gLZwunRMH51+cI3ZjMIUtNXb5eaqGitnpjEwP2jqyKEPsxUBGdRDfsIlh1Lx+UC0V7QdttO61xMQ9CRJhMiEcc/wEzvF84zKnLHENnfPPz5HGKdIwFZoczokD79RD6IZo3m+i93YPTGFLTdhV67P686pBRnVoUKRV1dFKi9/TJrXKNyhuPOd8UwZXN1mfbtLkK+aG6fMp4jBGo9cgTRgvgnNGeoRxFOPkByeQdRmqqWL2ghzrOMKFo/G5XISkSJg9nyGNUmi2hoOfOhACxUmYQLM1pGkKs2eK3LL/4T6G+lA87422gd67PUiyBLUxd9BcMVAJRgE+/6ufI3Ii+BMf1qEFc9fE3rf2aMClSEiTFHpHJwTgkwn8IclsFA1b+HUrXh/VVGH1Lcg6mWNYPUs0vgXqriA14Z17tXpZwTgAMmDyZILp0ymQA5PWBPauLZq+aZSSaP2crlVX13LjGiYzuCfuyubPq94jd4UFsek7/lUA/yvG2F8GMJv/LJ8XGP9jAH/5Jk7uJy2KiBteBMRBDCVQlm7WYBKUoHi77++W4GTAYkrc3G/CP/cRTILSAgCIr9joNQBgaYpbtxGs3tR8ER/+zCHsezZkQ4aikf7E9PkUiq5g9nIG+54tBIohAe0HbQEVZjIrcVydEwfeuYfxV2MoJsHtWM4gqRLab7Sh6ArOPz3H7PkMsUNCzN03uzBaBtIwBVIgDmLSdAhTMDBYBxa50RxRMdh/vy8myXpTx/TZFP65DwZGx2CApJPtsW7pSzaBxei80cFbv/MtOC8J0jd9MUWe5rD2LGF9au1ZAi4++mIkBMWqrkIXbb43TRx1zbh1VqB3bGMF3H7O+f0AfjHPcz4l+xDA9/kv8zx3GWOfA/iQMXYC4LD4+/l//55rPqdriatMoLiFbBGFxxs4SZAgMa72AAtmNBmO3VgUD829Zq3IehqSiOnw0yGsXQtZkMEfkRsEFynmr8mQQbVUpFEq0DRJQG5ZakMlfvnc7jcJk1Jz1t634Z67UCwFjW4DqqGi0WvAaBsleke1QGI5OcpABjRTE4UC/RI4/N4hnMFcH2devKRpSvpBBjVuVV3F+WfnkHQJyj+gcylSqDglLnIjIAfSIKXr7yeiWWTv2xg/Hos1Wyfoeh1ruRYtZCwa0AA9Qy6a4BXjrhRC87jxnPNN2VxdNS5Lm1mng7BJ8NeHTkgN5EO6/NxGuIgwC/JANCgBCL2eLMlK2nfFPDB5NsHoyxEkJpFdb5LD7JolbR9Zk6E2VORJDtVU4Q09NLqka6UYCunrSRLVX7ukvaU0FOhtHe2Hbai6ukAoFU0ucsAZOrD6802TTk48TKZaKY1SPP2lp5g+mUKzNZg9k/ITWzTBZ8czIAdUS71UvbFUX2A9CpijNkt6hYVN4SrdtiKF6w7WKNcRt13nvJaDq7vw3Rdzl2rRs9sbekAOaA3SpCoxJRiQpilkyOSEa2fIkZM7Fmh/1nu7hzRMkfjEDOi8OR+a6YrQ4PSGHhp7Dci6LM6lji1gwIDZNUX9VocaBIDACdC810QSJGAviKGw88YOginRuMNZiEa3gTSgfZDeov0Pd7DyBh46b3ag23q5+Vq4PsgBq0f7n8iJEISB+D2XmvDOvVq9rCJFNPoggn/uExIyWeh8iRrnaH2Nw7XMLmr+XOe98XWyIDZ91/81gF8C8GMA/28QNPB/CeCnALRxTRuZn+TpVXUzBYBUvENlSXCPRxGKpxhleF0SLER+ARK36r/XX7Kn5PDluinuuo1gFXqfIUOj1xBioP644FrgRgjHIYIswPirMYkGa/MNB+aLIc/hnDhQTRUAIOsydh7twB/7CJ0QektH635LFATdd7vEMR04kBUZSUznIqsycdonlGRZTvDI9sM2OkeLh0DJRtMgu09uIy4zGXmeE7Ln+RT2vo12q13a3C5BjUH0ByYx5AnZssdOTIVUl/R6uA6QO3Cpgz0NhatQMfiGrk5Jf9PEcenC+W5trIBbyjkAwBh7A8A/AuC/X/ixjeXp2ARAc/47/u/q7+qO/4cA/CEAePjw4TWc8eXjMhMoSZbgvHCQJQutg8sW8EvHnqM7vHMP51+e0yZFp3UXjANyoZjTmYpaGX2tD0mW0H3UBVNYyblFIIc4RHccQJIkhJMQ3be7cE7JLSt2YgFPVjSl1JxNgkRoSXgDDwwM/Q/6i6Krht7Bc6rRM8gCOErBFCo0Ez/B8NMCVfZBB+MnY0weT6BbpMXDj51nObIsg6zL6L3bg2IqAm3Dvw9FJ0qcc+pAt3Ux5bL2LASzAPa+jda9lph0CfHAG5puV6m91TyzzkK0Lu5KITSPW8s583gtN1evEpehzXCNijRKazVV+N+sO1YSkki4e+IinIVQGyqMhwYUU0Gn0xEUZ+4a0zpqAaBNjN7SydBAluCeufBHPiR1YRfMc6JqqIi8CHmci+l4VdvHOXWo7vESJHEC1VDhDlyiNYCh9aBVomFDIoFRc8dE7MSlz8NztHviAjKIPm7NKWZMEVbr4ZR0/dIwhRu5It9xlxy9pYuaLZgERJOv6PatWp91P19XtxS/fwBLgwOgLIZaRFcWRdfvWI1yHXHbOefGBlc3Wefche++NCw5aqP/br+koZMEczSNnuLgowNBO1YaCtIkhXfmoXXUwvjJGADd75MJlZB6R4eiUvOXI2CYRI1ZxVRg7pjCBaqIfOPXpnh+nA5edcEC5rqcKtm4x24MSEDzoAmza2J2PIN76kJWZaETqOgKZFkmrdMkRTgJEc5Cyj2AsD3na7SoY8YdQXNGTe6qw/L42RjWrlXK7fwc+R7Y7JpC/qJIL63mmFX6XdXv7KZqo8s8124yNnrXPM+/Yoz9DIB/GcB/HUAK4BdAhcQfyfP8xXWczE/y9KqYsAAqKHis6jza98gaky8UcaxCc8g+pInvKpX3i27Cuo1gHfS+uFAUg+yQY582UuEkBJOpeZLntIlRJJp0tY/aoihLo7T0+RN/vrBT4mEWF7HZNWEdWMjyDHpDhyRJmDybCIEta9dCOA3RvNeEP/LRetBaggAX6RTumQvnhYPZ8QzWoQW9paP3dg9m10TrfovoJmugxhzlJBvEz8ySDHEQo/tWVxQswXhOB5uLPWdRVrquVd2kOkTNZSDnl0kwd2xjdWs5Zx7/NIC/nef5l4WfOQBalb9rgaZoTuHfQeV3S5Hn+Z8F8GcB4Od+7ueqPPm7E3OthciJEHkR2ma7ZItb3bjz5ssm91cRUdi+1wbTGA5+5gDIaVpedXXjD3UmMai6Ss0cVRbrsBiKoRAFauiDqQzhOITW1oRFsKfTFC0LM4RxCK2llTSrnGOyILb6FnKWlxrbvEASn4Ov0fnrzT5ROCEBika6HEWqLHfYyZIMZscU0OzET6C3dTGVUm0VwVmA9CjF4JMBAKKMhtOQ9HYkCF46p1ByZA6fdDV2Gzj7+EzQLa5jilncMBttmgDGXgyzu4B/v0ohc1cKIeDWcw7wmm6uXjU2ai5zl8i4QAfnm53CMOlC2mcOuGcuglmANE5FQ0PUKnOKc5Zmgo4uBN9b5GyXIUPzsCm0s9xTF5OnJESqWRq0pgbGGFiHiXOpar/YezaCcUANIkhEIx0GQBfLRhaGAnvXhnfqLblk8Tw8OyOqumZqhOJuKKL24ci8YExuOHmeI5pFMHZIdNg/9xE5EUInxO63yKji+PvHUAxFCLhzygfPX9WNUHHdVr+HdUhgoH5wwGkoADW6ASxv4O9YjXIdcZs556YHVzdZ59yV776YuxRj0cTkCGajRTXR3u/ao8FUlMBoGkSlNAgBPDueke5pksMduMiRY++DPcTeorkp6h/GViJLVqKWJkAwDEpIWtuwS3/f+1YPiZ+gea8J79TD8DNyKPbOPASzAI1eA733emSlLjOotgo1pTwpqzLyLEc4DWH1LRIdnhEDpFjH2eGi7snSTFw7XkOBkcOVrMtEE88XjV6A9oB8f8hrLv4MKMZF6K3LIDpfJW7rfdaew6Z/mOf5M5QTwU3HT9T0qi5h8YdknTiuYtSLmQLLzSFrz1r5kK3r8l4UwSTA+PEYskqOVHxzASw6pcGEHFg4CsU5c+AcO1A0BdMnU+g7uuhuz17MoDU1RFOa/BebGZFLlIlid5p/dsVQcPIbJ2AglA3X1bD37YV4Ykg81aKAKz8G/6zBhBACaZ5iNpxB1VSMPh+JidveB3ul61Sy5p1/3iRICD3QMkqOW0W9js4bHXLymqODJFUSn3MT3aTid1Y7jaxMKy+TYO7SxorHLeacfwbAv1r52cegnAIAYIxZAN4G5aARY+wlgO+CINOY//fHt3CuNxK8IAEDZi9nkDUZ4yfjlfbi1SnRJsERheaOif4Hi6kRn7oUrTQ5Us/as9ZOVoo01SROEJ2T9s7sKXHSOWXz8HuHAENpAs0/szf0MH0+Ret+C2bPBEAbFH5ufFLsnDgIZyHpZTAmirJgHKB11IIbuGBgS1RZfgxuycstQPvv9tF7u4e9D/Zow3mUonWvhdnxDGlIzZvBbw4wPZmic9RBo9/A/b37paKJT/+5+Lx/TiiCq/LEq9e2qsOl6AokSRJOfMCrFzJ3oRDicVs553XeXN1G8Bqm0S3TwYFF0R9MArFJWoXecU4cyJoM99hFc7+JOIihNwvDsnndlcUZ8pTqL+/UEzSAYjMjCROBeuGi4713eqL2KWpL1EXRDCEJEpz96KzWyCIYBzj5+IS0BStNbMWgxurxbxzDG3vIc6Jw1NErFV2Be+hCNmRIioTu2114Qw+yQi5bQRrg+S8/h97W4Q5c9N7uYfZyhhfff4H2/XZJi8fet5dqRLFBmwT1tKxKPVKslWI/pmvOcrHhK9I1kyDBZEIOpFzX8S7WKNcRt1jn3Ojg6ibj6/zuN9HuEflqbukdzAIkfiKc7Oz9hdi3pFBDl+eicBLCOXUgyzKcMwfjJ2OYHapD6uofscea06NkVSat0I4BA1RbJHECDZrQ3au6l3LKaZ6SoLtkkG5O834Tdm5DNmWBiBl9NULnzQ7yjNDI9sGCtuqcOhh8OiBJDEUqDd25zqI/9smVeL6ui/ucxl4DsiGLRlhx31rcw1TpZXXXfp0mz09K3OVP/I3VtKiLJdTIBhDDVYWwopO6dzgLxUIC6hPTVbipSZDAPXaFTTqnIPBj8aKo914PkROByQyt+y0S89u3aKo03xT5Ax/T4yl2tB2iPWmLAkYJFITjENEsIhvcBxUb3Jw2NIMfDwhWfOYi8om/qZhU+EyeTcDAyJ6v34Bma0sQROeE6Bbjz8bI4xzBLAAYWYgrplKyBS1a8/IONyTAHbjovkkOERwuWDupK/DIS99DQTxxlW7Suu/DOXEwfTaF3tKR57koHovHvhClc4c2VrcVjLHfBuA+5kjAQvzHAP4UY+z3AvhLAP4IgB/Mm8cA8IsA/jBj7JcB7AP45wH8s7dz1tcfxXzDYfhZkq20F9/4uIWcU0UUOicLEU9r14J75kLv6IgmEXrv9aCaqtjArDp2EbqvWjRB4tpVeU5Q3yxeOC4FkwBqpCIYB6IJvPvBLhgYiQ7uGHBPXZpssxyarcFoGWj0GpRfvAgSkxDOQrIfbaiLPDsvtuqossXGi2ZpcAcuxk/HMHdMtI/aAm3DC77ETxDOQtIGMlXhaAVWbqLx6X/iE7WE0y2ugyfO0VrmjokoiDB+Rteaa/p8QygPX1e8tpur24jigIs7MxWbB1zwu6jNsEr/pdFroPsOaez5Yx+RG9VrTh2R652iLQQ8q9pzSUhOWapBuYkjWPgmizeji0O4WoOCNoTTFDey4Mc4+fgE4y/HUA2VkMqFJlESJDj9zVMMfzgEUxhSl4SOg0kANluu4WRdRqPfEJstTmWKwxgNqyGGSqPPRjj59RNkOU3gOWIpz3J4Qw/ewIO1Z9VPwGuGkXWfefyErJJZzmD0DNh7Ng3G5iLJxWaXc+KAgcEduNh7f++VnkHbEPFaD674d7+KXnMTsen+iK8Bb+hRHd7RSWvmYUcIFBf3d8E4wOmPTqFqlCvcgYvZ0xnMngn/3MfB9wi9bLQN0TDm5zP4ZIDZyxkiJ8L02RSyISMNU0yfT9E8IHOYyIlIc2euuzd+PIa9b5c1CqOUTGsMMnE4+8EZfOaTlk+vA71Fbr96i6jikiwta+QxIIuzlc51/Jy5sUUSJaWhV5oQajINU4HMA+r3TsXcACyoneLaFzR5gnGw1EAvymR805rCPDb6NIyx/9MFf5LneX5tHeabnF7dFhT5Kors1YfVK0MMC4OflW4FVcTJBkW6YiiwDiziWtpaqSnBJAazYxKF6cShTZAbYfx0jHgaI57FkE0ZvbdJkT0YBUiD+ftHFavbuQWvsD1ni8/CrdZbhy2EsxDnn59DkiWMfjyCe+xi/6f2RaPJn/oYfj5EjhzRNBIW5LxxppoqObZ4CXFdTzwAoAm/tnClSoIE42djaG0NjW4D/sinTVSSI3ZjPP/l5+g86giOpxAA5N9BBT1T/G540rtIN6kaPMGff3oOf+KjedAUmzqe+FRTfe1EBG8x5/x+AH8hz/PSxijP87N5M+ffBPDvgTS7fl/hT/4oSLPrMQAfwJ94XeiddSEm1UkmtKA0W1sLn79QvyJYOPFJSpmaybUtYj9GlmbI8jm0t6EinIbwRz7llhp4bV3TGwB6b/fgnrlCkwsgx6U0SuGcOMhi0q/qv9tHFEZwBy41nHOG7jtdtB8SvHf01QiSJkHRFECideud06YmiRNkcUbTMNtA7FFDO0toI7RKiLjYeOGigXmWI5yFGD8eI4kTGLZB63++IcqRI5pEOP/8HHma03v5NToXxnohZF4EJWECe6+eelv33TknpEF0/hW5+DQPmiKnFt0Qvylxy3XOa725uulYiQjUF1a0q7QZxDEKOU2SJEiKBFmTBSKkiPgFIGhWdbbZPO8I3T9DEWuAUwiyOEMwDdB52EHsxzSZN+za4ZzRNgTCmqOHRY46Iec7SCBtwEIkYUK0h4YMzSC6lTfyYPWsJaQSbxippkoiqWEidBS584x36kE1VXS/1UXi0+u4XhDXF4vdmJCVXZOazeHq/MO/qypqJ5jSueiWTi6ikkRSAHPkT3W4lUYpYi9GFmXlgRq/DjfoeHSbcVs555syuLppceQlVNmKwXodGp4PbRphA3mWY/TViNAxtkEom8J7BJOAXCqfT4E52NUbesiQwTvx8OP/9Mfov9NHOAnFfgWgtcHrB1mRobU1SExCcB5g8OkAZz86w8FHB9CbOpKIbMgllQZQiqmI9eacUq5hMkMcxTj99VPkEtHNm/ea6L7VpUZLRUy45OA3rwWKTVlBZR8HGD8bi2F8MAvw+G8/hn3PhvPCQfOwCc3SEIexcCU2ugahJxlICxDLoujPf/k5Rl+OoDZU9N7tieF8dWA2/GQoNCC5rmtRn3bdnuh1zi2bnu3vRKk9AADoghom4/n/X2fc2PTqNqDI15F0XgViyJsUWn+RhIAaTnIFcZKneSnxVCMYB2KCxXVi+OQHKBRQc865YpKDw+zpDHmWQ9Jp4eu6Dm/oYefRDloPWjD7Jm1Yolh0kztvdARnMo1SUfgIbn3Baj0JEsReDM3QAIU2cZIiAQwY/niI2fEM48dj6JYOs2di8mxCiWo+Ocoz2sDqLZ1cukwF7fttTJ5PhM1w4id4+asvkaUZnBcODn72AJqlEQUjiMFkRkidU7LGmzybEBqgUExxqkIVIVTkjVp7FiXFCj1k3XedxRn0Dk3yvBEVaOaOCX/kAwyvq4jgreScPM//B2t+99cAvL/idyGAf27+/3c2Nn04FfON1beI+91aLW65SY6rOvFVqZnTp1NkGa0nc4dygCRLwsK8Ol2pajUUJ05FXZniJCaYBMhzggqbOya8Mw/+2BfFh27piPwI7YdtojI8OYbz0oE/8tF+0Eb7jTY6DzsIplSAuWcukiSBoimUz+IMmq1Bb+pLOkCrrrFz6sDsmQgnIaZPp/j0L32KzhsdMMbwrX/8W+Jz3//Z+7D2LNj3bDCFITwP4Z17CGcheu/0So2Zusk1LxzPPz3H2Y/OSFusb+HRP/JoqalTV8jyRvfjv/MY4SSELMuQDRmyLpcE5tfFa1Yc3UrO+aZsrm466u5pxVhY0Va1GepeX2x0cjpndfNRi6CpUByKv+daNbzx+uJXXxDiWCJEiXviQrM0ZGlGa3mF/gc/5/HjMSKHGsydhx1otoaYxZA0CZ2jTqmpruik+cPA4J7PbXgzRta+WDRaeTNm9NkI/sRHY7eBxE+w9+HeIgdHCdr321BMBXpTx+O/8xjxkxg5I4fOzqMOZicz5AlNv09/eEqfvUJfr/uuljTX5s42s5MZFFWB/khfO9wKp2EJCVWsXW56U3/LcVt7q2/E4GoT5sKVj113X12APuOi3vyZbbQMnH18hmAawD0h4fHIo4E2cggKlqITCjCchoRClGW4Jy6JFYcxdFtH82ETsipTU4g/r3PAO/XgDlxkUQbZpN/HHr3Gm3o4+/QM9753D4pJDsP+yBd7lsRPaBjVNoQbsKIpiLUYjZ0GolmENErp8+zTurN2qR5UtDKltLhvkTVZfA/BOMDTv/sUWZrh5NdPYO6YMLoGZEVGOKbGUhIn6L7ZLVNYJzTkUk1VCEcXGy/OiYMXv/ICWZIJJ2LFVEQtwgdm/sgvUTmdUweRG9G/k4xcyfr198/rnls2OtM8zx/V/Zwx9gsA/o8A/rvXeE7Aaz69uq6kswm8tA5KhpxoP+E0JJX0+e9qET8MaB21yJnJj0paFsUHNl+kvPHDhfPqhPKCSSBshd1TF+7IRTSNkMYp0jBF970u7D0bjb0GjJaB8ZMxWdPpiiiE+BTL3reFULBz4ggRYs3SYHQMxCHZljMwjL8ao/teF4qpiIXbvEedYN3WRYOJsTlkG1kJCtk+oqJv8mQCJjF0jjpio+ecOkQd229RgylK0Hu7BwAYfjYEA8PoqxElJIOcLrI4g9bX4A08zF7MYO1bQqTMwMLNp4gyADYQeizeIzpZleZxDq2pQbVUGE0DWZoJnYuvW0juKvE15JxvXBSbn0mYLDUBqsHvM+fYEZoxq3LQpjmu6MRXfb3RMaA2VDT3myIX1KFLSpoZepkOYe/bVPAUmk/F13LLSu5gY9+bN5Vy4OWvvhT5TNFpquyf+7B2LeR5jvYjcrLgx3v+4rngrGcJWahLjPjw7plbgkavu8ZGy8Dzl8/hD32Mn4+BHGjsNmhKP3AER10xqFhBRo43sUvTrCiLhH3nqqk156qncYrzL8/J6r2tI0uzcoGI9YWsP/LBwGhyJgEMbG2jr/Qdv2bF0S3mnG/E5urrCm5FW23M1EU1f62iO5cc/OY0K6FVMV5QD3nNwH8//GyI2Isxez5D4AaQNRl5lMPateCfL4RCi8M5gBoiyMm6OJpFwmY89mPS2uMbxBrkkd7SsffRHs5+dIbWXguSLMHatSDrcmmy7Z/7UBoKkpeJ0Ac6+fgEsibDPycXUvueDXvPxvmX50AKMJWhfb8N1VQRzSJEswjBKCCLdUspaTteNOE22kZpMKDoClr3WvS6GnH7Yr6QNAkqVBhtYwkNeJOb+tuO28o535TB1arm6EVxGR2cKpJuFfqszg2U76ekYwmKoiD1iFKuNTUoioLeez0kYYLZS9LJ02wNnaMOIi/C0W8/Qp7nmJ3MEM9iTJ9O0bzXFGs6CUgPp3W/hfajNpzjhQNgOA6h2Ap2OjtU0+Q5kjARdEswQJaJisXRhUbXgBIo0Fs6xl+MkSHD+SfnUDRa47wO4O9z9uwMraMWrUdDETmRMQb3zBV5GQxgMhPSGmlM+yW9rUOHDtVUYdiLYR1nJVi7FrFFZAn+yIeiKbD6lvgukpB0SpECo+cjIdCc+EkJqRNMAoEaiv0YaZjCH/pwz1xyHNaVlffP655bXulM8zz/m4yx/z2AfwPA77iOE/omTK+umnQuG9VOMYAyIkQuC1eugjDnGSURljNMnkzg6I44Du8+B9MATGY0pR64SKJEbDyqwaHDsR/DG3qYPZ9BkiWYOybMHRONnQa5VrXmx2UM9r5Nf3s8Q6PXEJuSYFoRPgzIctM796h4AjVQDn7qAOf2ObrvdtF7q0dww76FydMJvHMPZteE2TUhazKiaYTBbFASfAUWlChZpw2brMulznue5pi+mMIf+Og8pI6xvW8LVFPrQQuTZxNxrOI9IGmSEEBMggSDE3IJy7OcuK6gJlwcxoimEZoHTWRYzUktfo/99/piM1nVzSn+d3GjexFV5q5O1W8i53xTg0PX+QSkrglQ95pNHmirdLqKscqJr0jnyaIMrYetkohdkSNfPJ9gGsAdkl0md0Xgm4u65lPxtbIhU1PrbWpqBZMA1q6FOIiFqw1ADShZl9HYaZSF9RjQebOD6dMp3KGL88/OyaI0ThcOeBusVX4s+54Nf0h89fGnY8xezKBoCuy+vfQ6Pp1yBkSbMpoLuij/nMVrwfNuMCE0JVMY/DMfaUKTN450XLWhLTbTvYFHegBnZOneeti60MXmsvfSXY/rzjnflM3V1xUXoZcv+/yq1mt8LY2fjJHFGa1NXYF35sG+Zy8/V3UFoRsiT3JY+xaZO0hE2eaCpDw3CYcXkMOL1tIw+HgA+z41mbl+2arn8vgp6Ww195rIoozW+xwhLPkSQoRCH4zlRJtlKiN0wHzQFLuxoD5FToSXL1+SI9bUF5vMjGWQJRnt+20xEFBNlfJBW18rRF1EULeOWgIVoJqq2KCBLX28JQrtkl5HzfcV+0R5TfTNcstdrm2Ksa1z6uMqzIXL6uDUIenq7j9/WkaCiP2DKkO1VFj7FvSOLqhK7qmL4WdDTL6awNw1MfpqhDd+xxuiHlEbKoIRrd2dN3eQ+Al23tyB0TEW6yomrdDGLqFp+LkdfO8AaUiCxf7QBzJg8nwC3dYh67IYVEu6tORMzK+Rc+wQIvjAFqYKAISBS5Zl8Mc+7VtyknXwzjxoTQ16WxfPeVmTkac5Jk8nUEwFD37LA4ROCL2jC1aFe+ai+053qdHN9YGyMEOIEJpNWohJQAgho2UgQAD7wMb+d/apyR6X6bNFnVK+X7QPbcRejM7Dzlp31qIeTxIQOul1iuvIaF8A+N41HIfHaz+9KiYd5BAL47ofIMUHYDgLBZQsnIZEXTgoi1euhTDPha38kU9Uqkr32WhTQ8MduMjTXHSNqxFMAjgvHDCV4fQ3T6E1NTT3moiTGFbPQutBCztv7YhJTxqnGH0xEoLIyCGaE8XCgAtnSRpxy2Mnhv22jTRJMflqAvfERZ7nyLMczik1pJwTsiNu7DQgaRLaD6g4ke5JmB3PyJoTNSgAQ1m6dkbHwNHPH2H8bIzmQZPs/oYeTj4+EQ+K/Q/30Xunh+FnQ7peWQ5Zk0kHSJMgyRK0FhUrzgsHqq3CH/noPOrQ1Gzswzl24J648Ac+2o/atRoidQ8n2yg310oPoEtACl+Tqfp155xvZPBN/yro+qrXXKoZvYa0ypuNqybiqqkiTEPSzppH0Racu6vwB+zs+Qx6S0cwDbD/4T5Ngri7wXDZ3YA3ncaPx3CPXaH71Tf6RDV9MYV3Rk4xqqWi/25/ZQMqCRKhx+G8dCDrMsy+SZNkm5rfzqlDguzzKR5vTFWb44pOVFd9R4e1b6H/rT5kU8beu3uwD+ylNWjv26I54wwdNPebSzafRfQSp33kaQ7nhPjq+x/R9Up8avIVRWFXfueMKKDNwyacEwdm39yYanWle+luxzbnbBC3sWFe9x7rnl91vwMgnvl8LTgnZO19+hunaOw3EE5CtL/TRpZkYqrMaztZIwcpo2XAaBkkKtwgREkaEiKZrzUAwuElSzJRH9j3bFh7Fk2uO/XoN+585Rw7CM4DojCoshiq6U2dBNXnyGPFWIjQ3/vePYEqyrMciZIIhA7TGKJZRFSFMx/WoUU6ZuchYj/G5MkErfst6G0dzYMm/HOadAP1Olo8t0uqJGzgOS1jVS4oXs8STasGnQSUkeAcic6Ha6tec9G9cUdjm3NqYhPmQjE2HlJt2Czif+ecOIiDWDQx+GvsfRtJnKDzqAPFVOCqLoIxoffCWQhj18DOGztiMK4YZADDtUy1BgkQFzUM+Wfgrn85cjQPm5i+mGLwyYB0b97uUuNUV+GcUS0iKRIavQa0loYsWtBUq2LvnYcdhLOQGj+PJ6XGdZ7liJ0Yk68mYDJDGpEb1t639xD7MRq9BpIgEQ1WvaXj8HuHcM5oH5YjJ5QQAN3WkWVUX1WbusW9qNqgvVE+LzDdMxfTp1PsPNpBMA3Qe7eHLMkQO/R+dS7GAOXN0Rcj4XZaR/OsOwdeg/JBIf8O7noj+JXOjDGmgKzBn13L2eCbM73iX/pNPEDqHoB8wXhD6ixKmrT24Vm8MYsQZk7TqfIQFVPB0c8fCXrDOupGznKy/FZVaA0NURah0Wqg/aCNvQ/3REeaSQxaQ4PckMEyRot03ukOpoGAE6ZhimBGi9U/92Hv24jcCNPnU6iGir2P9uCPfNh7trDDlGxJUKesngVVV5FECRKfEmM4CWk6/nhconHxzVvdtVMMBUbTwHQyhTf0qJA4dpCGKU3JQOKqRtsQm0wORVR0ZYF2mgTIWU7OXCeu+Ix5QpbOWkMDU1hpUse/tyJF6ypT74sebnd9qn4TOeebGoqhiAYjh9letKnetKip0+laVSTVTUF4g8TqWiUnl6KlpbFjCGqWc+oAjAqUeBYLFEnRWaLFaBosGhXzh7N/7lOe0WXELjVeFF0hOLImQVEVQjMyLDWgipuANErJYrNrIDgPMH02xeFPH6LRbWDw4wGCUYCTXzuB1qTiKY2pAd17h+yAs3gh0idrMmSZ3GRkSUb7oF2iuxbXoHPqCC0ivalDa2kk4l74O/5aXlQBgGZTDjJaBk23OgbCabi0ti9Cb2bIiMpZaeJftHm/yjT1LsY252wWt7FhvnAgseb5JZoNsgR/5sM5cYTOF6cfGW1DOHNCBtVYeY7gPEDokK7LyQ9OsPNoB1qTNlzuwIW5Y8I9JZe+nbd3YO/b5KrZ0GhQc0L5izu8cMth1VKRp6QbsSo/c2rX7NkM4TSEtWdBb+tieo8cGD8Zk7bGmYvmYRPtIxoGFemoS5Sv+bEH0YA2m06Ie9+7RzVUkmHnzR3MXs7QPGpCUYmmwNEBSUC6RNVhUBKQJTljbElMtZbudoF+0apQDAVKSHpCkiyV8uuq++6u1zbF2Oac64vLDBaK9comSHZOvWwftRf1wtzNLYszSC0JB985EPQk79zD+KsxNXOCuSD6fIgVuzR4M/smQieE1bOEsUkSJuK5LqvEHHj2xTNMn03hD3103+pSQ7hlIJyE1CxWZSEE33nQEfphdag2Xs813mnAH/uLGsBY6P71v90n7Zoc8Cc+aQHumGIAX2yw5lmO/rt9dI5IfxA5NWWcFzTEn2GG7ttdQe/nuZy7n04eT2gfmRN4QFKoQWx0DTR2G1QzBglROpvGEg0UILSPP/IRuiGMngHVUGuRgUsxb3YJQfc57fZ1aARvdFaMsb9e82MNwHsAegD+h9d5Ut+UuIkHyLoHIH/w620deZYvuT+sKoaqhTeAEg+RJ0GOVFl3bgDIKjjJYN+z0brXQhImaD1olS2Q59NzDq/LMU+4eS5gfd6A3KZyECdda2iIprTArH1LWAJyIWHumsPRLoqqQGkoCGYBJaT3+wDIyrP1oCW0OPgkPQmIo8lFwIpaEUmQ4PjXjzF9MkU+/z+rb8E5pkJN1dWlSRT/vjmEmF9rAwaah03MTmbQLE1Y/oVhiNghUcLmbrPE4+d2qDyp84fNZSGBq0TeivDxuzBV3+ac6wmjY2D/o/3LURA2mIC9yn1SajTNkTgcTaQYysLScrCgLITTEGz+fznLxXGEuwFr1dIAjLYBs2uSzacXieYrchBqDhLSJBW0sepnL1p3G20D/rkP/5QmR2mcovWgJV6jN3UMfjyAPCL3v8ZOA7FHDSSWM7HOnTMHkiRh99u71Fwq2CTzRr176sKTPGqktA1kYYYgD4SVKT9H/h1wh60kTKA3dbgnLoafD9HoNdB52KFGOc9zYbK0eVyF3qxOwoVWADYbVlymQK7GbdMjtjnn1eI2NszV9wgmAZRwQctZl5d4zTF5PEESJwKZU6QfhbMQ7ksX/tQXtYXZMxG5EZjEcP7ZOVjOoFoq2nqbqOd7NtEvE5pcmx1TiAAPPh0gciI8+/vPoNk0ee+/1weTGfSODlkmurXe1MWwB5PyGk9CsvqNvRjOmUN6G7/tSNRTHAXYfqMN59QRk/ciHZVvlnhwDT9FJyRP7M4p+w0VWUS5whuSwULnAa3tujzLp+JAWSTV2rOIis4gUOqKrizRNavfZxU5sPI+KDSPsiQr5dd1NOG7UNsUY5tzbj6uMlhYJ3wMlBEzRQphEiYLNzeL5CD4nilLM+hNHW/8jjcQzKhJEM5CnH9xDlVXRd0jqRKsvoVGrwFv6AnEPwCxp0vCBI3dhnDG5bQkTqkOZyFp02iK0E7kuaLuWe6cOJg+n1ItociQVRknv3EiXmvv2QjG1PxOY6qXZFUuXWPeYC3lZoPEigFq6PDhubVvkX4YFytGJv7eaBsEUpAkRH6EcBpCb1MDJxgFRGfbtUSTZfJsItDJQm8nSMgJMc8x/nwMSSYUZe+d3sWDqCoFFysMhe5gbHpWEpbB9TMAfwHA/z3P8//vdZ7UNyVu4gGy7gHIp9b8d9UifV3BVS28FUMRU+ALz6nSdFB0Bcaegda9lligdZuFItUr8iLquvrUlDI7JhgYZENG+6gtust8isaLGAHN1RWcfnwKSSKtmkavIeC+XLCUJ1+9pSPIAjHNjv1YdNetXUugh8ZPxrDDxftNn0yRRAmQQdhuth62MH0yJerFfPPFE661Sx32uuaJ0TIwezGDpEkYPxmjeUiQyWBG1Ioi7Yw7YPTf7dMkT5PJntlQRAd/k6kWv+7V5t34MdkLBk6Avff37spUfZtzrikuC1He9Jh1gsSbxqpGExfrS09T2kR5DEgBxVJIxM9UoLUWUGReMNRZDfPfc40p59TB+Y/PqSjxU3Te6qD/Qb+Uo6oi81zrh+tndN7oYPjpEGqDnGYUfaGvNXs5g6zIaO41MTsmd5idt3bQPmqThTCfjoULyoPRo9xV1O8YPxnDH/pI4gTthzRpz5Ej9UkwkMOLkUMUi/z8ee5KggRMZvAGHvxzH5ImUZ6b56DqdH3d91wt1Irc+ouKnFUOZRdNuS4qqG8otjnnFeImN8x1qGT+7K5a0K56fvF1PoypvoicSGhBsJxBbapkATylTY+iKUR7aBkIZyEmT0gfDwzIUxIdbbfa4hnvHDuQVRnxLEb7QRv2PjVKpi+mePb3nhH61tLQfa+L7jtduKcuglGALM4w/HwIxhgih+gbzcOmEPrk67+x16BjdLSFbte8seGPfQRDQv7yjdGqxhcf/HHUJtfT2M13kYSEnGw/asMbetj/YF/kmGKerV77IsqZX2tFV0p1Yd26v8o9U8wNAEStd5FI9h1FDG5zzi0Er4GK+nxrnz8FNB9H25Rofav0dnQFkiIhdEOwnIkGjGjWYv6sNomS6bx04J17yNOc1ntDQfsB1QviWctAZgh+JD4LQO61oRGi0Wug0W2QAx0gLNGTkJ7zJcfcCZmkIAOYwkTzI3ZjSDKZq0RRhGAcIM/yku4ir6OCaYA0TNHoNgTV3WgZJRSRP/YJsdjSCSH0Rkecn2LQtQsnYUmsmOeTYByQyPw0AFKg81aHar6uIfRGV6GTi/ssPvhr7DVEoz0JkwvrkFqAwzi4U43gVbHRmeV5/o/e8Hl8I+MmHiAXTaDWPRw3eXgWxbeCCfE+zY65BGWtFuqRG8E79dB7r1dWPedJDMvFOKd6cSte3dbhDqjISf2UkuGc3lWEDCuGIizU+WYymJBbhGqrxHl/2EZ/vy+SWrGxUkyweZ6Xk9PAAQNRwYowXsUghyyEQBzG4jgH3zmgQidKYO8WEEjFKf/8PIGFLtDgxwMopiIacPx8+GYMWCStot2yZmmiMNUsDd65V9Ls2QQOWNzg8+t29qMzpHEKf+Dj7d/99lok1m3ENufc7eDCxhe5Ya2L6muKiJBgEoB5DMGI8ofdsIVuTt1GbV2eVYyFxtTwkyEkJgEmwBgTDXEBly6gTriQZ+dhR0y+FUNB/50+mMaQR7lALR79/BGcUwfhUQjGGLwJCbHLulyyU08CsuS1ehb8sY/W/VaJipCECbI4g9E1wEAFZRKROKKkSoicSEykps+maOwSp76IUAinIQBg52iH+OuGBKNpwBt4MHdMuj6bQI/59Vvx3Nj0WVLnUHbRlKtYUFedRG5qI7bNOa8WN7VhXoVK5muprmmxCuXBEbOKRoje3ts9sfYB4OyHZ6QdNV+f/rkvttvWvoXYi9E8bELWZbFZMtoGpi9IAFgxya2F/1xSJcQeNYeMrlGiIE2eToiqLTHMns4Qh7EY5vijuUOWQTVF+0GbaqSWDt3WBXqXX5csyqC1NTT3m/BGHibPJkj8RDRS+GYp9mNyr3IjQg90jdJQ0DlxCKk0DpF4Cc5+fEY2x/MajH+/1WsPlPMB1/cqDqOiMFrWObsKeiIsiycbHUPk142oWnejkQNgm3NuMy5DCeXPPH/mI41SZCmJoxdpfbU0ZWPZpEQ0ksYB0jjF5OlENG2Gnw9JL8fSBJWa36PBlBozxz84hjtwgZQo56IOqXkfvodo9BrwzikP8L2BvW9j/GSMs984I5pnS4Vukybh7OUMwSiA2TNpgDX00DpsiZoEWAAGjLaB8eOxoLo3ogbOPj5D60ELAOmKcTOOYBLA2rNEk7n3Tq+Uuxu9BvyRD62liXon8iLIhoyW3SK00YSMN5IgQftemzSAsIxOLtaGnAUROREa3QY1qy6BtqnmiTvYCK6Nu3tm35C47gfIRROoizY2F4k1J2FCzZzzAO45JRGrZyEKFg9jYFmM0+yYcM9c0t7JaaOzjtssGkIAiSfbGp4Pn8M+tIVAl9E1ROFS1bWobiYB0u6pUjKKibb0uefc1TQm0eU0XKjHe2ceZFNG7MRQH5J9Z+gShLHRbyDPcux9uCcaS0W0TBXpVDxPPsWSVIkmaIweLPzfaZSWBF4VTRHXl9st88KLd4yTgK7LVeGAiq4gcOg6mG0Sea3aGm9jG9W4TnpFdX0r4XxDltLvVUvdyHL9ovc32gbaD9uYPp1CyiWE0xDhJBS6O0voR0DQQlnORP6zDi1IErk+8PPniCV7z4Zz5qDRbcDsmCXrYsVQkOhU2EXhohFSPXdJlRCfxQjdEKqtotFtQNZkMZFWdEXQYd1TF5IilSZpex/swXlJXHXFJOcs57SMNKqKCK6LVc+Vi4qci6Zoa99TX+MkcocLqp/0uIkN8ypUslhLa9A61TDaVOhnSQatpS2QsGOqZzRbE/emrMiw92xIikTCmwCsviUap86JAzajGkhvEX0qDmLkWKzr3js9uAPSjsjiDI1+A+375Ipp7dIUPY1S2sTIMkZfjeCf++i+2xWIII504VP3ItVKrK8W5VBv5OH4HxzD7Jtwjkn3gtsMc/MMJjFB78Bkfh461WLjJ2Oc/vAU4y/G0JoazI4JzdZw/2fvi++2lMcqQ7JiY7o4jJqdzBCOQ7QetISOWNH967I1S7WZfNcaNdu4e3HZmkXUGzlIq+bcR/edrnCAWifaXTUpCSYBvKGH2ItFY0Fv6kjjVNh71+1vRG7RiAYlyVJJ+27JDKWwNqp7A+fMQezFaOw14J65kGQJ51+ew+pb2P1gFy9/7SViP4axQxRvpaEsmS/w3FqkukuKBCd14E98SLK0aEjN91fBmAbnZsdc5Gadfp8hg2ZrsHep2cQ1dtIwBdMYkphkMOw9G5PHEzHUXqfJxeua/nt9AiWo5dqJszMug7Z5XfLLyjNkjP0zlzlQnue/+Oqns41NYt3NddGNx3+3qlPN4WpxEEO3ddrQHM8QTkIhOloSEZ4X6oqhoHnYpGJjXuz4Y7+W21y0tnzxKy/IvcqWkEs50ois94wdQ0D0Bp8MyHlLkYRoaZ3NbrFQ4+cAFChp80SpmiohhxjBFht7DUi6hNZRS6jIc67o8a8dwz/30X6jjTzLcfDdg8XmbA5dFiKuXaP0GZ1TRwg7F6dYWUwuYo29BpBDfM7JZIJwElJjKWrAO/VIgDBMlja0fDNVR+visYkGhWIo2HufNoBJnEBVlsVPbyu2Oef1ieuiV9ROzXRFUK+UhnJhM6d6vHUN7f67faHJEzl07tyWt/qZeN7grgvjJ2NIkiS0pryBB/fUxeTJBMEsgGqowplCUYkmwXK2dF51LgrFc+WbTO83PCi6grMfnuHwe4cAo+lXMCGaBs+7VQczo2Og+24XsRdDbaiC19952KGCrb0sIniVYuXCZ03hetZN0S46NkdrVbXcrjO2Oefm4to0kObNR65xJ+gNaxAj6yavVeHzYmPE3rfRPiItKj4Z5rp8nOadBMmykUKYUL4KKF/x9zY6Bt76R99C980uQjdE+14bRscQOU9v6oAEtJwW1IYKSIB1aKH3Vg+Tx+Qqw0WG1YYqkEhGp0z7kDWipo+fjWH2TciGDKYwyLIMxhjCaUhDNkUiKjpjYFMGSZOEnoZiKEQbk2VEfgR736bJOWNL13NVk7f4N1xQ3b5nQzEV6DZZGw8/GV4Zccev9abiyXcxtjnn64lNa5YkSDD4ZIDIjQSKRLVJtDz24lLTd5Pg+47ZS2pqtt8gGrVsyNh5tANZlSlnRcumCOLfAStp/a16n+LaKO4NONU7mkSUVxiDsqtA1mUEU8pr1j4NqswdE7EbkwFE2yCKVZyWtLOMtiEomKET0jGfkavd7vu7yBlp+3A3U45CKr6+mj/s0EaWkBNX5BJKBxngDlwht1Ecatc1c+r2tUbHKA0DXhe0zVVi3af5c5c4Tg5gm3Rek7hIS6coWMrRNnpLXwiAYYUYZ0GEmRdKddxmAadXJehNHV7gwZ/40Js69j7YE5A8LrrHHV5CNxQLum5Cwws1rkPhvHBouts1oegKsiRb0MLAMPiEaE+SJOHBP/SgzFsFCfsNPhlAa2to3WsR7JEtipYkJLHCLM0QTAMojQVHndPWps+mAGgjVuwqc8HAoo6NYiiYPJ+QTs7cASfP81p19jr00SbJrS6MjoHuO4sN4NeY5P7cJf52m3O+xihuuKtxka1wCW23ojl7lYfuhQ4488IqjVPhNCFBgtoikb269zVgiEk0b05HboTpiymyOBMInmBMcGXDNqB8S4HaVMEYE/afVSpXUeusrtHN8wHPO8EsIN57QXTU2rWIfmVQocqdK5Iwgdkx0b7fhndOsOhoFpHY+mGzhAbciP50RfeiixCjm7zeNuzaSdw1xp+7xN9uc86GcV2OV3zN8vXB3WR4rEKMrNtwVRuR1Y1ecWDDRUSnz6eYvZxh79t7wlE09mOEM2r6cCrBykbyt/ri37yBJMmSQO103uggizPo39Wh6ApN8hm5X82ez+CPfYwfj9Ho06AHoMbv0qYosPH87zxHEiUYPxtD1VRYhxbMrgm9pVPdMT8HWZeRhos8EMwCuMcuZF2GbuqQNAmGbazcwF40WKzT6qu6p14GcXcbLmq3FH/uEn+7zTnXFJs+j4JJgMnjCSF45zTH/e/so/1G+0pabklI1Me9b+/h7Idn0CyNUCe71BCJ3RhpRKiU8eMx7H271Kjd/3Bf7ElWvfc6w5sq1Vs2ZIwfk1Cwf+aj0aPBsqzJCM4D+Oc+Jk8n2PvOHmZPZ7B2LbhnrthbcSMWfj2dUwe7394Vv4+8iGqhjgGjZSAOCD05fTlden3xsxTdlmVNhtE0BL08CRZD7VWftXZf+xOG5lv3qd68tbPYxq3GRZ3qqmApMOdDr4DXAgvBy+LUpVqMAwsxT45UUU0Vu9/ZRRqm2PuALM0HnwxEMwYAgnNqDPFp97rpEG8kZXEG1VaFO0QWk/VuGqfET59R11mzNGqWFDq3eZbDPXaBBqnKc2FDrpQuYm7FxzdTXGhU8Fi7hK7J2aJI5FE3IZRkCeF5iJCFZHl+Sg0kSZJEIVb7fdYkqGJy8869Je569W/5BvBrVnHf5pxbiuuannOngU3cj1ahcWrFBa/w0L0IUi0ayfMms9pQETkRdEknrnnHqH1f/nP+Ot68TtwE7rlLzZGmBomRGKI38IRIIBccrVK56j4zp18EowBpkmL6ZCqum9kl7RuOIFQMoqUZHUMIj/JJe7EgTAI61+ZhE/7IFwXpZUT+XoVedx3F0w0XYNuccwNxXZRMfpyqm0w1XqV5uLaeCElcOPZieGcenv+Xz9H/oA+9qSOaRSKfXOYeVXRy2+I1jqRKIk8U6yRJnaODVBnN++SKKesypidTNL4i63CeQwV93lSw++Eupi+nUE0VaZrC2DEEik80qgp0c4A2c4pGjWjN1HDwMwfovdeDvbe8gd302VF8P44cSDrJlRF3l7mnbtsh75KxzTnXFJf9nmtr5ZpjJHECSESLsg9t6G19Zf184XvO13uWZGgdtdB52Fno4Bh9OKcOtKZWynHVfFSUqKg933ntIKnUJC7SsooN7wwZCbO/0yUh4jnVy9whq3TFUDB9QW5Xk68msPYson2jgXC2sFIvHps7YE0eTzD6coQ0SbHzcAftN9rQWzryPIfEJIy+HKHzqFOSyShR7Y2yJMjpx6dwTh36Dg7skpNYXR5Y1bz5JiNyqrHy0+V5/vg2T2QbtxerbvIlLYvCzb8OXluELK960BZ5oUVhwyJSpdqMScMUzpkDMBD/+mGr5G6zcjo015iIz2KkcYoszxCMAjR6pH+jNTV6nySDf+5DNmSiL+wvkoykSXBPXSiagu6jLuIoxt77e2XqBwNaR8QhjZwFZJJDxINpAO/UQ+tBq5TE6ja1sR/j/NNzRH6E5kETkirB7JloHhCN7DIipvwa5FlOE/qnU7QetIRWyNIE8ZroM68a25xzO3Ft0/OaByuwWnSu1GQcLpqMV7EUrfv7i+7j4u/9c5/gwrMQVp9ETLmoKnIIu90SNWm+6eJQZkVXAJmKPvfMBUDIGb2lL2iWbPm8qq4XxfNLQqK72j0bRods0mVdxuzlTIizFi3H+aaTw5r9mU+NpAolM0uzBR2sJv+vpardkfxwE7HNOTcT13XPXOY4r9L4W/VaPh1OAtrgZWkGd0hN3NiNoRpqLSXpovey94leoJoqYi9eUCXmwYdhHOXsD31Mn0wRnAeInRhHv+VIaPNxDT9ZJQF2WZdhtAzIiozzz88x/mJM+mHzoVCxSQYAsk7NHOfUgW7pSGJqRPff7dfWcbXT8RX5o+7vr4q4uwxl5i4jebY553riOr7nVfdn56iD8dMxFJWE/K/azClFTvdwsUHKGyJVl85iM3T8eFwSNwawrBU21/nJ0ky49XK0Ll8nHN2S5ZlA4hg7BrIsI1pTmGF2MsPk8QRZnOHcO0eWZmj0GsiyDFbPqjU1UAxyRA6cAPmXOdSGitANETohZFNGFmWkLZbmkBW5nLd0RbhgFZtEzgm5f8V+jOA8gGIqQpqC76/CSYgc+RJiaNWg/ychfjI+5TaWonqTX5QcL2qg1D1og3EgqFs8AZQmbTmWbI+LzZjIjaBZGnrv9BB7MVoPWmsnMkVbwP57pJUx/HwI98RF4ifI0xzN+0Q5sHYtJEGC8y/P0TAagqPP4ZH7H+5TQtcoSURuBMVc3ujIKim6e2ceNaTGgfidO3Bh7ZaTILC84TXa5NIQuRH0tk5Np4ZGU3+Z+O5FEVNgmT9a9/0K4bIHLWiWVnLOqPvbn5Qu9k96XNf0fNW6v8iFj7sjcE0uXkStPN+KlfhK/a8L7uPimrj/W+4j8RM8+6VnyPMcs2czIKO/mT4jxxrn1EGe5mg/aBPNouAGoxiU09oP2hg/GUNraMizHN23u4K6uW5KxBvc3A6VWxQLuuv8b1RdBTKQW5VM4qzV4pI3hCePJ0hiynPGd4ylc11V5CRBguNfPxaUy4PvHCzl/m1+2MZl4rruGX6cOmrnbQRfk97Qo/ygK0R5Suh3cRALZ61NopjLJEXC5PFEoHTq6BR8wxe6Ie7/3H1oLQ3Tp1N6X1+Be+ouOVb13unh5OMT+Oc+dEuHYiqk1zcfClXzdjAOyNFr6JP+RZajdb++1hIoR1nC7GwGMLIzX6XJtYpSe5WN96baOdf1fNvG3Y7r+J5X3Z/739lH+yE1QK9yv1bfQzVVSLZUW4OvG7APPhlg+NkQwShA52EHqq1C0RQ61oyOZYB0brSWhnBCKJvh50PoTV042wHUBHJOHYQjoome/Yi0+SSdpC+a7zaRsYzcN+/REJpbhbOUIc/r6zq+z4ucCP7Ah9pQ4Rw79Nk0Bc9/+TnRvrwETGHoPOwgiRLREDd2jNrvTtZkNPebCCaBYD/wuikJE5x/do7uu90lxBC/5vza/iTFxp+WMfZfA/A/AvAtAEvVd57nb13jeW3jGmMTWOKrwuqr7llVwWC1oZZcTpxjB0//3lORdI5+/khQG7gdXzAOqHOcLKbLqz7f4JMBxo/HiF1Sad99fxeKQe5bWZghduKS1Xqe55AUCbIso/tml8QOC5zLDBnMHRMAVk6EkiBBEtGUv7HXQKPbwOyYJukcwpiEy84udRteo22g0W0gbaaCL8o3YkWkQDE5X2i9WIBDDj8ZXlg4rmuWfR2buW3OuZm4tun5ikLkIhc+7o5QFNlbd+9V9Wf4hsKf1hdH6+7R4kSMMYb+e320HrYQu7HQ9WIygz/ycfxrxwCAyVcT7P/0/mLCXVgPvFiz+pZo+nY6m/H0nRcOVFtFfBaTQ5ZBqJzeOz1hWeqcOsJly75nC5fBokOVYtDUbxgPyVlrWL4uFxU5zomDk++fUNN37FMOrkzn6wYAqz7jHac7rI1tzrm+uM7JaJXaeR3H3fQ+NTpUT8RODNVSMf5yjEiLkAYplKYiNhubvF8pl7UNIQJa1ZMRwqxOBG/gwepbSIMUcldG52EHO492oBgKXOYiT3OikKcp2g/aIodIkoSoT7RLd+CSQ1elwRyMA5z84ATewMPkOblq7X17b7Xte6F57Jw4sPdtuGcujJZRP72vPGsu46636rqt+/45morXSa8DonCbc64W11HHrLs/i4i5Vz3PKsWyWoNXByx8D5XFGTRTgz/0yUq80UKWZRh8OgDLSb9T0ckx13lJjQ2jZYAxRvumubMdckDra2ASQ+iGpKmX5kh8onMDhD5sHjSRhRmYzGAf2Nh5YwfNgyYNm1s6nWdh7QbjAC9+9QVilyhlekeH3tahxzo6b3QgqRKZ48gMjd0GZENG4ASIvRgSk1Y2xIsmN43dhvgukpBcmMNJiDzLEY5DQW8vIpruKjpvVVxXzbTRKxlj/xiAvwjgrwF4H8BfAdAA8NsBPAbwt658Btt4pbjoRtgUIvuqyZG/d9HOHAxQDZVU0lPSyFFMBYmf4NP/7FPMXswQ2RGa95slq2zFWIiRctvyqhhi6TOG1O31hz6cYwferxECoPctguhNn9FEK0kSmH0TaZQinIaYvZgBAE5+/QQH3z0Q71EV86u7vsE4wFd/8yu4A4Jf82JJUshWcPDpQAiPyppc0tBZBQtc9fMipS2chsKda5PGm2Is4N1mx0TkRWv1dJau7deUILc55+biOhEXdZu2tWg+ox5ivCqqjWYAojhKYqInKbqysQsWP4eig5JqEm0CgHCPGj8dI5yRCDskgMkLt6rieijq1fCCsJg7uGZYsSFr79tIAtLWSsMUoRsSpQNlamriE/WBFzN8k7VK/FCzNCiagjAOy9fwIrHoiJpbTGaUQ186Apl4mWfKJu91l2Obc24/bnrYtO59L3Of2vs2uu924Y98tB600DpqYfTFCGbLXLhOXXBO1c/B12ydngxv+GbI4J156LzRweHPHEIyJHQedAR9IhgHNBE/peYKn1YbbQNm14Ssyjj75AyqrmLwwwHcUxfth+2FDmKQkDWwIaN9vw2zay7p/RWD1xPOS4f0NRqqsHGuRWVWhn1Xcdfb9PsvfqcASiYddzW2OefqcR11zHXcn+tCOOBVmrec3l1tjjhnDpwTB1qD8kIaznWuUiDLMvTe7gldTXPHhD/2kcUZmodN7DzaQTALYHQX1KzIjYSIO2c82Hs2XLhEf5rXNb13ekhCchE0WgYgAUbTWEIbFz8TcmD42RCxG1PDXWXI0xy6rUOSJMiKjMGPB/AGHiRJQjAKEPohWMIQBzEavQa673YFZar6vVTdCIEFBTbPcnIdC2O4Zy45aOJ6nhXBOBDMkcvUlVeN66yZNn3V/wbAnwHwLwGIAfzhPM9/hTH2HoD/D4C/fKV338YrxSY3wiqdi7rXXSY5LrnVcJvuub1dEiTUyMlSBNMAzYOmKDqCaQCtqUGzNYROCG2mLVllF3ne68QQgbJejDf2kPop0jgl/Zo3WrAPbcQ+cTiDSUDuEmkGvaWj/60+uWi19VITpdo9r0YwJfcLRVcgKzJ1ped0iGASIE9ok5clGdwzt9TtX7XhXfnzAlUlCUjf5zKNNyGC6kUX6ulU42uEL29zzjXGXUJNrMs1FzWaeVMjciMwlyF2Yww/G2L/o/2NcxbXyKkTd09CEhE++9EZ3GOXms5OhJ03dzB+PC7ZFVcFDOuQdIwxuEMXuk0ugc4LB1mSQVIkqJYK79QTlshVy1J+/GLuWKVXVpxoaa0ymnHdGk6CBIqmQG/riIOYXtsxEM1WN37XHu/1pjtsc84txqaF7FWHTWtRZCvqomJ+qA5W+u/1hQB5lmTQGpqgSm8yJKnLZYquiM0DsEAHJEGCyIsIiZvniL0YZtespZg6pw5kQ16yFea/24l3gJQat+OnYzB5gXQq6oXoto7mQXMt/RUAkAPe0EM4JerG/nf3xYaw9nPPz7dYH14mN2ysnVP4Tvn7vga5Z5tzXiGu4zvmx1j1bL1q7VTMb7EfC1OF2CfXpyLKPgkSPP27TxFMCU1//+fuAxLQvNfE3kd7onGqmFQTcTeoLM2QRRm8oQetSWLHqxY86gAAau5JREFURXH1usF08pDYEwwMOXLBBig6AzcPmzCOqKGRhAkSP8HJxydI4xTRJMLOWztIwgRZRGgepaFA0iQc/bYj0txJyKkq8iPsfrAL79RDFEYIBgHMHRNmy4SsyWCMCVfPKp1y1bCQa+k0ug2cf3UOdMjVuG/0V+aKTb/DYBzg6d99CiZTc4ozR24yrrNm2vRV7wP4IwAy0ONOAYA8zz9hjP0xUFL6D690Btu4cmxyI9Td4Kted1FyLG6IqgLHzomDNFq4JuR5DlmTyYqzqUO1VJx/dg5/5EOWZciSjOaDJqJZhDd/4c2lRbPJQ7y4SLtvdzH8dIhwGiKZJdS0yTNEk4im4LMQzYMm8jTH7HiGPM2RpznZjsqSKKI2XUiKpsB56cAdukBGYqhcC8iAAUmVkGe5cJdYN1W6KNHwqdjws6HQ2LnM9KlY+LUetGqh0Stfe030nCvENudcU9ShSq57ErXqfddtkJY2WSs2eHVW4rIqI8pIO0LRldqJV92xeY5qHbUgq/KSfg9vIrlnLnbe3sHoyxEaHWpQp3EKYJkyWVcQhrMQaZgiciMSgH7pIHuUlSx7FVOBLMtUrM2FVTnkOwloWlZdb6vW46qJ1joKQvGamD0TlmZBMUiXw3nh4OBnD2obv+tywteYL64jtjnnFmPTQvYqk/gLNQHnNIhwGkJSJSCnIVcap5g+nUJraUgjQhXbBwvqYlWsOHIjnP34DEbLgHPiLOlPrfscwKKOGj8ZA4DQvEjCBLIiI/Ii9N7vYeetnbUaO3WIR/67YBxg9nJGqCBVKVG8inohwTgAJCzQgjWbQQCYPJtAb+lQGyqMjoHOQ7q2dU6n/LsYfDJA7MZCPL4o7F78ztZRdS+sk17P3LPNOXckavdKr4CeqDYYOWWJW4gX814wDYiWtNPAuXyOyIugWzrtKTRFvH+RMhlMaKjMTCZYDHwfVUSZFGsu/r9FF2NeuxSdgbNkkR8A4Pk/eI6zj89I01NiaD8iM5vpiykdNwMOPzoUn5kxRnkpSEh/LAXadhvTl+SglRs51JYKq2cJk4zhZ0Mh/HyRlET7qC0csBRVwezljGjr+/aS5tplvkP+PVh9C+7ALTFHbiquM29t+soMQJLnec4YOwPwEMAvzX/3AsDbVz6DbVw5NrkRVj0MN72B6po4zin9b3O/iQwZLQKJodEj1wS9pUPRFYRTsrkbfDrA+OkYzomDncYOvKkHs29CMRVYXYtEsSrNlIse4tXut2Io2PvOHvrf7uP0B6fQezoanQasXQvNwybOfvMM5o4JpjBEDk2+lEMFWltDGhAFyz1z0Xunt7SAawuNuUOXYimYPZsJ5Xa+uSuKm8qqvMQd5yLOzokjOvV1gn+CTztPrkW49mU34YquIMg3tyve5Hu4wdjmnGuK6saJr9ebRFGIZsF8g9R60Cq5GVx0nkUXrDrEnFhfuoI8z0vraIlSVEAPco0cSZEENbTaCOLHD2chFE1B7MXI8xxJkCyheuoaHdydIZpGmL6cQpIkoc9jdgkmLSkS7D0bo89GCGch8pTcGoJJIAosq28J0cPqVH7Vxqf6uddREJJwYZOexRkAYOfRDvKc4NqapQk6RfU74zmyTgvgNRZQ3uacW4zLFLKXfd5tPPUsWOGmUYo4iBFMAkyeTCBp5C719u9+e3ngZJCzy/jpGN6ZhzzOMXkyEa59F30OkZPmtGx/5ENWZVh9i5pMsoS9j/YExYvnnCRIapvj63IC1yTkNUeV4qUYCpRAweiLEZjMMPjhAEbXgGaS0LusyyK3cmqZ0TII0TfXNlx3vTl9TFIlkUurDZ9qLXQZU47Sd/L65Z5tzrkjUXf/bOrgu64eKCHyjIWFeDHvGS0DeZojzVKYOyb0pg7NmrtS7q8wNZi7WxbRvEAZZZL4CQ5++mCJRrlUU+kLM5qc5dBaWok27p16lB+TmP7OjUm0+LAJSSPjFsVUhHFM0ZbdaBs4/+Ic3rkHK7QgqRKsPQv99/oIZyExD+bX9TK0ymgageWEMmJ5mcJR1FxTDAWRG1EjG1kJjbn0nWm0b02jFJIsLTFHbiKuM29t+sofA3g0/+9fBvA/YYz9FwASAP9TAF9d+Qy2ceXYeHJRsyHa5HXFxcObEZqlITgPkMTUZW4eNoU1buRGpBczF+7k0GSzS82bLKRNw+zZjBTOLRWzZzPsvLUjpuWbPsR5ASHJEhULOhVf9qFNzlJ7DUgKoWQkXUL33S5N1qYBnGOH+JdOjMZ+AwwMwShAHCzTN1Z1d4NxAH/kI/aIdpX4NMEvop24iKqiK7WIpsiJ4A5c9N/tw5t6mL6cwupZYuMLoNS04p+b2xJuGq/KLb9K8+gaYptzrimWCov5NPkmJ5lifSrzBooqLTUHVtGrii5YXEdLMcp2n0bHEFOmuolXdf1Gswijr0awD23kaY4syahAqUCfi8e/9717ApqchEnJHnNVUWe0DbIYbRlIw1TkoDzOadrTNIT9eBIm0Ns6Tcv9BM7Agaqr0Boaxl9SA1yzNKJaPOzQ9zTfzJUoVassgy+gIHC0ZhyQ24RiKUBOumeSLCFLMirUCvdHnbBrNb6mfHEdsc05txg3uQG/qFmUhElpM5QEibD9HX05gt7UYe1akDSpNKWt3v8AbQIUkxq/SbSgHa1cl4Vm9+jLEcBA9Icsh97UCTEEEinVbNpY1TXH7X1bbE54buDHLzp+cmTRunPik2m9qWP6fArnzMHut3bhnDpU481F3wEaBjKVQZVVQdfgP4/ciBBGQYJEX+Th0Anhj3xCRU8XGl8iPxdqIb7pusr98Brmnm3OuUNR94xcm0fWoD8UQxESE1WkTDXvKYaCo58/QjANsP/hPtyBC8WgfUMVRVwyd5mfm3PiIJgGaN9vI5jRXsbcMTF5OsHoqxGSYL3EQrHxCywPaiRVgtbQEPsxdr+9i+7bXSR+gue//ByKqRDjISf0cbUuMDoGum/R/kv9NiEQG72GGOInAbmF1un1VKNY0yStBDmIBcH3NMW/kWQJzsBB7MfIIpK+aB42gRwYfDIQ1HfuMMrzZufNDsJpiL0P9l4JnXMZqt515a1Nj/B/BfDB/L//KEjA69n83ymAf+qVz2QbV4qr3Aib3milxTPfOKVxClmV0X2ni9gjgU8+talq6gAAckqKnQcdIANZaXaoWEqTlNTOleXN3oWfe55o/amPJE6gKkRt8s99NB800TpslZTZOcLIG3gYfTWCrMiANO/ITsLFpsYo06OKyYHbBCqBgsGPBwim1LluHbWoGx4ktTDNqu4GR0iYOya8Mw+zkxlmz2fUKJICYeMHoLQhkzUZ7pkrGkSbfvevKbd8m3OuKVYVEDc5yeTrM0syaqDE5ebAOnoVd8HSGhpOf+MU58k56VXcs8XDl3+uVRMvHmLy7pELw/jZGPd/+j7sA3ttIwhYNI349JhzvesQQNXGN0crpnGKcBzSRGluL7zzaAfe0MPk2QTusYvTH57C7ttCayxLMkRuBGvPAnJg8mSCPCF74SJVjOfZlQXlBcWoYiwjCTlKsH3UrrUHfs01ci6Kbc655bipZ9FFzaKltWEoaB21ICkSFEvB9OkUkiYtTWmr97/RMtDYbSBLyY1FUUmLQtFoup9GNOQpIn/5MRrdBk2D4xSdow7RLOY6fABEU6bYHM/SDHEQIw1TnAxOEDkRsjBDjhw7bxK6LgkTBMMAaZzC6BrY/3C/ZDhR14xGDngDj5y15nkqDVMo6gqKp6aIxpNz4gCAaDBxKpugXukKXYcogWZqpRqLfzZeC/ljn8TdXw+61HXENufc4bgoj6x7HhYNDqr1et06NDqkWRNMaO2uG1BVh8MnPziBYip4/vefo3W/henTKUZfjqCZJIK8yf6q2PgtxdzoQbVU6B0dhz99CKNt4OTpCRSDHLbUlgpJklZqtXINIUmhv1EMyglcT2udO2hpv1rI27ImY//D/WXEor5wFYu8CLIio/vuYr+ahAnpplo6QjcUDqP8u2wdthC1IoE2ukp8XeYQG71Dnud/pvDf/4Ax9h0A/w2QGvtfy/P8N2/o/LZRiOsQNr2UBaSfEF+6qUPWZDGhFguzYCVeTVBiAtZfwAG5qKBqqfDPfQFrK06CN1UY54k2mFA3OhjRhEkzyAnK0z3Iqiw6zby4yhNSR3cGDnbf3xXnH0znKu01mhCxHwvrYEmRAEbTs4OPDnDyGycw+ybMHVNMqzhMkzeBFKMeIZEhg31vDoXMqIALpgGUhiLOoVp0FmlXF+mGFD/D68Yt3+ac6406lN5NPmCKhVD7wXJzYJ2OF9eE8EfUrDU7JmSd9LhqdcLWFF0cheIOXEyeEnLm5NdPqBHNxcLrGkGVXMun+d7Qw/gpbdj4eiw2fdMopY2QoUDWZOx+axdP/s4ThCOC8QbnAWSdKJhG26Cp1ksTnbc75L7nkg0nU6gBlMUZFFWB2lBJ76vS/F51Hfn519E4i1FEOm00SSrQyi6LFLzrsc05r0dsWgddlOOKtEFgAdNv32tj/4N9JFGyVIdUn6X2vo1Hv/AIwTQAMuD4+8eEuAsS7DzaQZZkS8hffozpiyncoUt6YN5yDgrGVLME0wCKRggc54UDMKrNzJ4J3dIR5AFSPxVUiNglWoQ39JDGKYb64r3rzCy4jlYSU83WftBGlpLYafuNNjoPOyKHJCE54qmWitiJcfIxNZVYzmDfIw2LEvIpTETt55w5hGqaD/mK15PXQjwvf4OaxGtjm3PufqzLI+tq66sOPy5zTDAgiRMopoLWvRZOf3iK6fEU3Xe7mD2foXW/JVDCxWHaOimLEuVsHOD0h6di2G20DSRRIgZXnHopqzIktWzWUqchJN6jRtdvVbO5ul+9CNXJkVFZkqFttjF+MkbsLeihwSSopWpd5z7p6xp8Xekd8jx/BuDfueZz2caauK6O36Y3WjAO8PJXXyLLMsxeztB+2BaCfXUUiGrULQ7eAeaLiv+dKBbmSu+XURhXDAVW38L55+eQdRmzpzPsfbRHm42+VYIkcy2a5kETeocmYc4JaWvETgzN1GqPb+/byOIM5s7C9tsf+nBPXDCZwd63IWmSQDHVNYGKRVEVIZEECUafj5BnuYAf8utaFVHkG9A6pfx1cMrXkFteim3Oef3iqoVQsVmbZRn8c7Lm5Npcl3kvjkIZPx6LnwXTAMPPh9j/cL/kUMVzBVCG5HYedgQVbPTlCFmUwR/7sPYtdN7oUB45dRAMA0g66eTImrxYw292aPo88qEaKryBh51H5BKhGioUbT7B9hN0HnXQuteiJroui0k/Y0xQxYpIJ55ruLhrHULwomfFZZt7iqHAG3jQW/qlkIKvW2xzzt2L66iD6miDmz4jV6EdjY6BwScDuAMXjV4DwTTA9OWUDCEqyF9eOz3/B8/hD32kcYoszmAf2ALZYnQMoW8VTANkeYZGv4H+R/2FS1SQIHRC5FFOg7E4g6TSsG30dITYjWHtWWLIhAmWNPuCKTWNJFWC1bMEEki1iE5VrfE4NY3JDOE0hNmnplKOXOhw1eV1ramhrbcFWqmIVHjda5PrjJvMOYyx3wdCAT0EcAzgD+R5/rcYY78L5LT1EMDfn//88fw1OoB/C8A/CcAD8CfzPP/TN3F+dzHqmh/Vn60bJl2lQXDZY9p9G3lMDeI8zWE0DRhNA+yI0dC6Yyw1cOvyZ11ePP3hKYJxgNij5svMI0fgIJtTOncMKKEi1nT1nKvnOn4yFvsio2eIBi5Q1hgV179mv7pJw5cP65jE0DxslhrFBgzY92xRUxZBCXUUuavE1zVE3+hdGGO/CuD/DOD/luf5yc2e0jbq4ro6ftUbrSrWy4Nzqlv7LaQxIVu4Xk2WZEKwa+X7zIsWZ+DA7i/befJJVfHnzqlzocJ4VaQ5jVIMfjyAbpMWhbFrwOgayOKMtCwKquntB22MvxpDMRWkPjnW8AJGMRQYOwYkWVpCvvDkwCdv9q6N1mELw8+HkFUZ5o6J6RMSP+Uw42ITKEszYZNc+vwFtAKHe2dJVoL6VTdLPNFzuogkS/CnRAWrhUuuOM5dj23O+WbHRYV8XfP3qpNbvuFyz1yMvhgBErnJyKpMFK65RoVwkjCUJUgup4I1+rRZ02INsiYjdmNMnk0gSRLSJEX33S6yNMP02VQg8yRFEjlUa5J7TuRGkBQJ7aM22g/aSKJENG8iNyJa1dGCVpWEyUoaFAChnwFcz7NiVSE7fjxG5EaIZhHZpKdX17tY9T5fV2xzzt2Pa7m316ADVx1rSe+rrnk8pxa5AxfIgd67PaTRHK1XcXVKokQ4xgDkFtPoNxaT9/l5BtMA7pmLNEyRuISgMVoGdFtH7+2FVbjQ0MnJwtfet4n6ZVEO4nqG3qmH3ns9REEkqJbTZ1PSC5IlNHaJKlpnDgEAYBC1SuTQtQudECxnYnNUJxp/Ua6/rrV/l/LJJnFbOYcx9l8F8CcA/HdAosuH85/3AfwFAH8QwF8E8K8A+A8A/Nb5S/8YgHcBvAHgAMDfYIz9Zp7nf+WmzvWuRF3zA8AS5YmvO7EW1zQrXxVdWHdM+8DGt/7xb8EZODBs0gnNkgxaS1sSQ16XP3ntYXZMRGGE0x+dIhgF8M48aG0NTGNoH7ZFQ5nLWZToToay8lyTMEGWZKIBLEmSuCbc4KJqnPEqjRGev6xdq1QfKcZqR9BVFLmrxCrjiJuMTd/lJYA/BeBPMsb+c1AC+k/yPA/Wv2wb1xXX1fErLrKq/XhJFHSuuu4OXEgSTWz8kV+y3V1XTHGED5MZZk9nAm1TxwEVQp+F98zTfElhfEmrwiAet2IsbP2SIClRM0qJy1TQ/1YfkkrTLJ5MIjdCOA0ROzEkTapFvvBrpjd1TJ5NoEBBc7+J2YsZRp+P4J/76L7TFXSIYhPoQgcPnQo+JjGa1FcEBavfn2KQbohz4sB54SBnOSRVujBxvGbFzq3lnO3k6vZik41RMXhjZ93xLmr4JCHRtu7/lvsYfTmCoivUaE0WjgelIidKliC5ikH5aZJPkDiUM1RbFQ0grauRFtccOeef+9AaGvyJj/2P9mH/nE2TbYkhmAYL63KjLHBsdJY3Q2s3mmGyRHG4yrOi+L0AqJ3iJSFpqDGZIY1T+CNfiLZeJb4unvma2NY5dzyuow6qO8ZFNISL7lOuGcgHSEpHEXbeJYRtgaqQpRn9W5vXPlVHHF1BOAuhuoTis3YtKKYidHbq1gpH86mGiuZBE2mYwrhvCHq7e+ZSHZfP81ZDg97RobU1dN/prqVn8mvHa5UiDR/A2nxV/dlN1CJ3MJ9sEreVc/5lAH88z/O/N//3cwBgjP0hAB/nef7n5//+YwAGjLH38zz/EYDfD6p7RgBGjLF/G8AfAPDNb+jUND+ARa3ArbYVQ1np5Fka3FaceTelF9bVTNXX2Ac27IP1wucAxACf10x8/fJmxvT5FOMvx9B3dEhMgjcg2mYapnj4Wx8iCZMl165NzpeHpEiCSt856gA5MPxsCP/cF9fV6Bii5rsKgq96nQGAMVbSMqu7hq86LKhzhF5lHHFTsdHZ5nn+jzHG9kACXf89AP8+gBlj7D8C8H/J8/xv3OA5bgPXC03lN/M6Sz6jYwjVdQ4/CyYBNUM2aFJwhE8VbVNcNDwhFlE0/D05Z7xoZ158LedhKroCSZYg6zJCL8T+d/Ypsc0XVbXwkzUqRhhjpaKp86iDHLT4uJVodXoHAM6xIxpB1h4VWFmakWvWnKd5ERRz1XcbTIJlQcFVhZWx4Ilu0mB73Yqd28o528nVzcUqvYar3IOrECODTwaiqdk8bJYcC4pCepIqQdVpuo0cRE9QJJEXig4tRtuA2aOGj9bUxIbPOXHIvSpK0XrUgr1LBZlz4iBLFxoQSZAgOCc3vTiI4Z652P9woWGhWurKHFotAC+kgKygttZRyYrHLTbBAKwVcRd5JQemT4lukWc5Gv3G0gTwMvF18cxXxbbOuRux7r6/jjqoegygvoEpzueC+7To1KQ3dbQetiBJpLUnBjA1ue/t3/k2Xn7/JTWY5052VVr2ve/dw+nHp/DOPaRJiobdWNnMARbU8jiISf+na4j3z5AJ+gGnPww+HYDlTAiVbtJgL9Yq3CkQANRIRTAOlnS7Vur2xDS8qkMDrXMIW9l4u8Z8clvDr9vIOYwxGcDPAfh/MsY+A2AA+E8A/M8AfAjg+4XzcRljnwP4kDF2AqqHvl843PcB/J4V7/OHAPwhAHj48OGrnvaNxcYomRXNY/4zfn+tc/IsvqdzSvIOWkPD7OVMaF2tRMPhanV7cb9U/Ldz7OD0h6dATuY0nTc7AokitLFMFbPxDIEbIA+JwcFNcIwdQ+SXaoO6rmlV5/rVedhBOCOTCFmXxbViEsPk8QTGjgHVUkvNossiZYp5IJyFSEOitK5yMS7WiVcdFqwyxrjtumbjd8nz/BTAvwbgX2OMfQDgnwYlod/PGHuW5/kbN3OK2+DxKhCwuiR20Q3MVdd5cArEJslwFdqm+J51KBrePR0/HoPNykmh+FpZkwUFQW/qwj5v9MVIbDLqIG/VYpBPtDg1qojaqQqJ8YQslNnnf6voCtpvtEsd98sWBIqhQAmV5Wn7umt8CRTQXds8bRK3lHO2k6sbiLoH+1XuQd58KGo/FBEjWZxBtVUwsBLipvre3JKzN+4JsdNiM6K6ScmRI8syGG2DqAumIhq9s+czSG2J1v7+8uYwQADFUpDGqdCwKDWuCxs3YAV3fMNC7qLmTR1kfPDJQPDYeROqJLSIeh0MMKD1oCUQjjxXXzW+Lp75utjWOV9vbHLfv0odVHeMdYMtYPV9yp/xwThA5ERQG0TjzNMcslGhWFVyXzAJoJgKjv7hI0HZqMsJRsfAvZ+9tzHtVDHKznWxH9MwrW1AMcv0CGNiwB/5sPdsSIq0cU1QrVXCWQjkgNYvoxaKdsTV50Aak0B8HJCwclGvZxWKm9PJspjqnfbDdvl6zNEHSZCQhtlrghy8hZyzD0AFoYn/KwBiAP8pgD8MwAZwVvn7CYDm/Hf839Xf1X2OPwvgzwLAz/3cz+WveM43Epf5blc1j/nPrF1LaHDWOXlW3zONUkyfTQEZmDybIOklUBvqUpOh9NpC3pi+mGLw2QCdB53SvqyIDOHrpMq8CMYBPv3PPhVSE41doneKBtS8URNMaRAlKRKSKEH3rS5UUxUNj3UNatEYWlfjMcDes0u1hqzKUBvkoLX7we5SLrr0XqqQryVFQuIntS7GVxFcXhUlsEGY1O4hbyOu9E55nv+QMfbHAXwM4F8F8OA6T2pLgbjeWLX4rjLt2rSY4ggf59RZog9UE2JVzycJVnPc6843mATQW7pAAzlnDpChFvJWnYA7Jw7cgQvvzIN9z0b7qL3UtOJIgNilaTsA4aBV1+Da9KFRl4jXbXDqoIwbo4Du4ObpMnETOee2Jlc/iVH7YF9zD65D4PgjH9Ekwt5He8iwQKJx5E18FiNnObSWJgqTOiE9jEmrQm/pgiLBo7pJcc9cEkWWJYyfj6E3dYSTEEbHWKKd1m1EjCZNs4y2IZq+RWcK/vNV9NPLNL9WNW9WQcaLPPY6IdNVeY1z3DnC8VVziGJcnwjhTcRN1jnbGqc+vo7BwyZ5qQ51wjdpoy9GNHFWZZg9E637reXGC69twgR5lpdo3dx6uEgRqDav19FOq8Gd65wTB6MvRghGVFMdfPdggSCcBHBOHSRBQpPxnkF5cQXVeynyhStnEpD+V9HVyjl2EDohvIGH1oPWQsiZbx4D2mQxxhCcB5g+mwpUcgnFfb5AcbunLvyhD9VSMfpyRIilbkM0qgXaICC9sdcROXhDOcef/++/kef5SwBgjP1pUEPnbwJoVf6+BWAGwCn8O6j87rWMy363dfud4s94g6DOybP6nlwsffCjAcJJCPeli8OfOSw1GZbeX1844h3/yjHs+7aQsOAD6cmTCbIkg3vmYuetHcFq4GsumJBTVeqnRCdnDOEsFAYLfBDUftAmIfe2DqNlIByH0Ns6Gt1GSch83XW8aJ9R93s+GNJsrdaR6yI0X913VtLu4Ro9RlnLbFWdeJW1XgIbqPLa++Em49LvxBj7naAO8j8B6uD+EoD/3XWd0JYCcf2xLoldx7RrVSgGwfTTKMX48bjUSKomxGJXmRc1AEqNnuIUq3jOVTQQd425KGknIelP9N/twx/7pYl98e+DSQDnhQPVVgEGyPpCsLT6t/y4dRupWtjxvGvfOmpBVuWVImvrmnKbdLKv0ry7K3GDOedWJlevCxT5OuMiKtAmDdBgEmD2cgZZkeGcOjCPTTR6DfFAVgxFIG+K1MzqeyOnZseT/+IJkjAhMeQP+isn8d65B2/gAQwCydJ52IHeIuepVZSp4roHIEQDeW6rWn0HE5rsmzskQlhE8dj79qUasJdpoEmKhNC9WMi0Lkdcp9DfdYsQXmfcZJ2zrXFWx9cxeKjmJWBRb6zSGOTrTVIlKKYCa99CnuZo3W+Vmib8eM6JIzYX1q5Vqk+C6QIhVES7vJL5haEsKF+6guGPh1BMRSB9siRDMKQGij/2kUXZhVRv/pn4/zIwjL4cofOoA1mTaai2a+Hk4xOMn4yh6iryNBfXofgc4CiiNCJ0g9pQETmRyAO8Biw2w6fxFP7ERwYymVAbqkAYADTA45vYogjqpa/d1zT8uqmck+f5iDH2DFRBix/P//djENqYn4MF4G0QOnnEGHsJ4LsA/ur8T747f82dicsgOK77u93kmVV8zzRKCSn8Tg/HPzgGUxban3XNVJ6bBp8NYN8nM5bi0No793D28RnsPRvuwCXdLp3WeRqSe53e0sltz1IBALIp481feBPGjrGU43bf20VwHkDWZdj3bOx+a7eWVr3qOl60zyj+fim/ViinAOVhb+gh9mLkWY7hZ0P03uldrPVV2WPuf7R/aYbKZeKu7K82elfG2Ecgfuc/BeoYfwXgXwdxPD+95nPaUiCuOS6CDN/UDbhJN5wvvCLsGaDNED/H8RPa7K0SHqvT++FuLOsWKr8uGS527cpZDgYSAlw1TS4ibuo2lCUI5HwzJ/i3ytzxYp6oLktXuS2o+m3FLeWcW5lcvQ5Q5OuOVQ+4untw3b3NcuJaW/sWGruNUiOVH8+AgeMnx4i9GGpDxcF3DpaKhtnxjGx2OyYin0TQl6ZH83N2Th3svLkDSZFw/P1j5CzH+WfnMLrGQgunhuJUzbPVpgd3cnBO6PYpogPVJun7FFE8lykQLtNA67/XpwZORI596xo44juq5JdNhf5uS/PiOuIW65xtjbMivq7CmOel4kTYPXNJE+dea3kQxmuHOEOekgOobMi1yDuuSdXoNaipGyVIfMobeUZ0dI5UlhQJAFaihS7j+Ge06FyGPx7Cn/qkI+HGkHUZZseEe+Yi9mIxtV63Dot6ZaEbUg02b3JrtiborkbbQPtBG+6xS9brSYr2g3bJOhlYoIg4wmDyeII0puHWzts7ACAaRM6JA2/oIZyG0EwNiUeUWUkuT/OrNVcdlZV/lnX3123eg7eYc/5dAP8iY+yvgAZX/xKA/xeA/xjAn2KM/V4AfwnAHwHwg3m+AYBfBPCHGWO/DBqA/fMA/tlrPK9XisvS476O/FJ8T72p49kvPYM38mAf2mgeNhFOQ0KqrWimKoaCzoMOZk9nS0NrtUHDpdAJyT3Tj6BZ5MDJm5qKrkBv6ui/30c4C7H3wZ4QUHZOnNJQKQkS7H+4j8AJsPf+4u/WfabqdazWeKtEnatU1yJyudhQn72cIRyHaL/RBoAl3VWety9a0zc95L4L+6tN3/0HoAn0nwfwi3me/+2bOJktBeJmou7GvQ2ecN0mY9XCq/tbXuR4px7sA3ut8FhV72eThbpuQRfP02gbZNE7twKs28wE46AE61tHn+CaILMXM0RuBKYwNJOmoDFclq4C3L3N0TXEjeecb/rk6uuOTR9wq+5to23AvmcjizPoLX2pmcPXqHPs4OT7J+Let/dIFwZYrAuzbZKDXExF1d4HeyvXu71nI/ET+Oc+ZFVG60ELk+cTgSTpvNGpbbquLRDmxQo/R+eUbIQ7DzuI/RiNXmOJd31RYVS91ps00IJxAOfMQTAmp61NkTGr8suruAPdQRrojeecrUDpxfF1FcZcJy90QiRugtiL4Q998Uwv3p/F9ca1/PgaWNqoAAL5N/p8hMZuA5IqlTT3eJOY10hVGmISJDj+9WNMn0whaZLQBVt3nYyOgYPvHkA2ZeRxDkmREAcxIp/Wr9k10XrQolz2ZIzZyxkhjmrWYRIu9Mokldyyxk/HiKbk9hdNIrQetJD4RE/rvtslyqoirRROV4wFlYzJDExmcE9cSIoESSb3Un59nFMHO2/tkHPgyEej31hqEm3i3noZbbJbugdvZW8FQvv1AXwCGkL9hwD+t3meB/Nmzr8J4N8D0Tx/X+F1fxRE83wMGoD9ibuECLxK3ft15BexxvUE9qGN2I2RpRmmL6bIkxxIAWPHWHn+xaG1olGuGD8dI01SJD5pj1p9C617LSiGQs3PC4ZDVckJPlRq9BrQXR2KuX7wv1HdsG695YBz6ojvj+e+qpPx3rf3cPbDM6FPI0mUHzj9HoBgOyThZrSsy3yGVZ/r60bj1MWmZ/LfBvAX8zwPb/JkcAsUiNe50HmVWNoc3EIDoLrJAFa7SVT/VmzE5pOkyIvWCo8B9Z3gup+vuy787+tEVVcdIxgHePFrLxA7C2cJvoHjUZoegWz0FIOEU/UOTaKKU7e6RhgkcunaBAK5bkL1msRt5Zxv5OTqdYp1zYhV665IWRx8MkDkR2j0G4i9WOg6cPpmGqbwhz6a95vQbA2H3z0sTZ7q1rvRNnD+1TkAYPJ8AkVV0NxvCt0cALX5c2XeKazP4jlxd65iA+oqzRF+vdat9WAc4OnffYo4oI3qw9/+cGMx1FXN+VdxB/q60Bhr4jZyzlag9A5GMZ+MvxxDMRUYTQPqvrrSJnzVeqtD6nFxdSYxIJu7zDzsACgIIHN0z5MxUaLGAfpGXzSJpk+mSKIECLG0nlbVOJz+lcYpgkmAPM8RjubUqqNOeThVoCnV6ftxvbIkTpBlGSKHzCmiWQRr1xKuLmBYWy9VgzsFRk6ExKemNvKFpbJikK27e0r6hZqt1ddAKyb/pet09wZft1Ln5HkeA/gX5v9f/d1fA/D+iteFAP65+f/fubiDQ4GVwRvGqqHC6ll4+asvkaYpsoh07JSGsvb8udbM4JMBJl9NMH0xhaRLaOw10H+/jyzJRGM5yIK1wyFgWXKibqh02cH/kqPdmkHQ+MkY/jnRPVsPW+Kcqk7GRttA950uFF3B5PkEiZsgnISw79mCXuaf++K814lMX/T9bJKzbls0/TKx0Vnkef7/uOkTmceNUyC2hQ7FqybCja3/ConkQjeJStLhdKii1eZK4bFVquuXWHz8M9WJMlchzsWCZ/jZELEbIxjTrVdNzHWNrcnTCfI8h9W3hLXoyuZWkOCrv/kVsiyDJEl49AuP1k7o102oXpe4xZzzjZxcvW6xcnO04udJmCCNUgQjcqmL3Aj+uS8E/bhGBT8Gkxl6nR452Znr0W3BJMDk2QQSI/45U0nwtKqbs66BuspBIZgEyJGDSQy993rkMjPfnFSFmted41U2IsE0AJMZmgdNslU/ddA8aG6U++uaL6/qDrSq0Py64pZyzlag9A4GX1+NXgOdRx2ETghjx6AByhqb8Lo6SDHqxb6NtkEuWMjBckY5YjxGOAsFFYLrdumWjtANYe/bAsUiaRIQAHEYi5zjnDiColAVUubnUjShmD6fQoIkchBvThfdNYNJgGAcLOn7dR52xOfyBz6yNAMDA2QgxzLd87LrWtEVsankNDR+jKIGUfvBesHjdbXtXWsA3GKd842MOzgUWAohQj6vyafPSHhY0iToJhk0qA11SXi49lhzpBzTGIy2AbWhIk1TxF6McBwSesdfFnGvi6rkRN1Q6aJnfPVzLu21Vqw33oRRGyrUHRWSLC0zEfIc1i65hFq7FoafDZF4CbIkg9pUxflytoM/oSY5byRd5l641D7x7jWFRdyNs5jHlgJxe/EqifCqXds6bZlVYseXPb+VneCazZoSKisn/kVR5iRIEE6JL161FS3CAgHAaNKkS7XKiblK3eJRtBatwrn55+fHGD8dwx24sLoLF68qpDAYB6KABKtHD2xjOb6pk6u7EtcFTV1qAsxRfHEQQ9Hpga/taGTp2V5oUnCBYSd3aoWMASDxE0yeT2DYBrTmvAmkK1ANFXEQw7CNWiG+YgOVT9UlRRLT6TROISmS0JdQdEVslEZfjtDoN6DZWq3uxk3Qk7h4fDgL0dhtYOfRzko6RF1UN2kXnVNdDr/L063biG2Nc/3xqjkmCch63Dl1YLQN6E19I+HNdUOkOrHvKoVUMRS4Zy4GPyIHzcnTCQ5/+hAsZ6Lpw8NoG+i80UHsUjOndb+FF7/yAu6pixw5NEvD/of7JRfAau2R6HRe/shHEido7jWpiVxw4JJVmcToj2dQTXVJ34/TxBI/gfPCEQjDOkHTja//HCmg9TXMXs4ABoH2KaIhiwigdbGudnwdGgDbuFzclaHAWqfOcx/hLMT+h/vUINVkaLYGxtilaEKKTs6eeUTPXdVW0bnXgd7Wodt6aY3wGgmo18db1Xiue8ZXDR3qPm/dHoznLV4n8YG5c+IgnIVwj11YBxY6rY44Dh98xX5MNKogEdcmGAeIPBKS33t/D8EkQJZkpLv1OcAUJnLapb67SzRp7lpTuBh350wWsaVA3FJcJhEWF++mN3+xM1206OQQXufEWZ4CVaZLm5wfh+bxRkxJdb1CdfDHPiRZEhuvKhRXkiVkcQatSdbFwTCAd+4BObDz5g6JIs9FDiVZQhqnBKvuGlAaSikxr9u8cFHAYqITn6c68dMUIAUiPwLS+b8LwakUTGbI0xyH3zu81oRzV/mi27jbcSElZ4P7qi6HcJtaa9dCMA0QnoeQIMFsmnScGs64mNpU3isYB3j5qy+RpRn8gY83fscbMNoGgnEAY8cAmzIxJVoF8efCfcWpuqIrGH0+QpaS9Xn7QbuUN8FAVpzhQuiU5xR/6iOYBCWr4jqU3zrnv7prXBWPL+apq6zvTTZH1et2l6dbtxjbGucaYlVuuOxwavDJALOXM6RRCgaGvQ/3ajdXm9IJiu51xQYLgJIjn6IrCGchYjcmKtUcnWP2TEiSJBzogDL9FDlw8vEJTn7zBFlIzSEGBn/kQ1Ilui55sITSBagx5FouVEmFN/REU6mxS8gY3dZx/P1juAMXWZShsdsQ9saipjIUmoSbChRNuVRTuC6Km8Y4iInGNa/XuMvVZeuZdbXjXWkAbOObE6tqnaJTp3vsYtafodFtCLrlZZ+7PA8U8wjPEUUjmDojlup7bOIyyZs+px+fgkkM4ydj9I2+eL9N0Di8ucwH2HxN73+4D6fvwOybsHftUvNJCZUSapCH2lARezH0po7zz88haRKCYYCc5WjsNyAx0gBb9Xkuo+G67ju4q03hu3Mmi9hSIO5YVJPVJpa6/DWRE8EduOi/2xd2k0bbEFA+SS27PK1rDq3S0hh8MkDkRIiDGL23e0v243zxBeMAZz86g6zI8Me+6B4DtKBjPxZTpziIIUkSjK4hjs0FmQEg9mNhaWz2zCUNHGAzDQkAGHwyQJZkRAfRlaXi1N63sf/T+4jdGKq1gBry4FQKq08IniRKri3h/KRP1Ldx9Vh3/29yX4kcMhdG773XQ4ZMUAKYRJsM+8BGlhD6ZhWVZ1URz9dOa78l1k5xSpTnOelqPB6vdVSoTtX5Z1d1FVm8cIDJsxz+1BdUjsnjCabyFJIilXR1JFVayif8M3ABdgDwzrxa57/qNY79WEzXO51O7d9cZX1fdnN0l6dbtxjbGucVY1VuuDTUPiQIv27pyK1cNISLaxxAfeNoha5UUWiU6zwIjZ44Lbl1dt/sYvJ0IoZckReh0Wug9aC1ev3PayfDNuCFHtIghfW2hUa/gWASIJyGmEwmgsJUpFFFboTESahZLTNkeQYmM2i2BkmWEDgBFJOESGcvZ+i+20X7qL2EOuAbwXU00VXf2yqK2vCzIaEI8xyyJiP2Y7JSz4KNKCTb2MbXFeu0YiI3gtk2YR3QGi2aO1zlXlYMpTTs4VGnQbrWue6C+qx4rGhGaCDnhSP2V3VonDrED2/m+Oc+VEOFYi0oUWaXmjlLEhE1OmS8Ac6dhvkAjFPXtZZGVu2KJIZlq+qhizRcL/pe7mpT+M6d0ZYCcfeiuvA3sdQVG5qGisiNMDuhzvQScsYhC+HIjaDb+trmUJ2KeTAJMHk8QTgNBf2CNzyKNCSjQ7SGLMzgDT2kPiGDeHLiRUWWZGRt7EUIJyFiJybqhCQJQWbeEc/ijKZwaVbf3a6BK1YLGt7B1y0ds5MZWvdasPrWktDqwXcOVl5vTqXgloY8mV5HwtlO1H9y41WRWes275cpOLgwuj/yodlEieKv5XQlXjytojuuOv+6tQPUT4n4BKlOQLlIpeANa1mVodoqYicWx+SNIkmVEHsxcpbD7CxyCJOZ+HcdPbRYHPEm1irnvyLq0HnhCKHVkuvLLa/vuzzduq3Y1jivHqtyw2UbhIquQFIkhG4IljPoLb003eao3yzJlhpHRTpB8XleFBrl9YUYYCmScOtM4xSKqeDoHz7Ci199gdHjEZyBA/elizzPcfCdg5XnrFkazJ4Jxhj0HR17396DYipCO4xTG0pGDBKDaqpwTh1EXoRoSiiiPM0FCsfu2xh/MRYW6lXhZC7syjd4vFm9yTpe2zyuuACClXV9qiYT29jGXYpVzd1gEkBRFdJ3OeosOXVed3C2wiaDk3Uad8V1qhgKcpaDgQlxdVtfDPU5kq6ICnROHHqTeXO8SGHnTIY6F2Bv6ME5dWDv2bUo63vfu0dIH11BnlNey9KMqOtNAy8/eymYCu0H7fK12WDA/rrXI6/32W/jVqJu4V94888t6YJhAJYxhLMQ3Te7S1OZcBZi59EO8jxfCd1NwoUAahzESyrmSUwOUKqhio1NEiSChpQECQ6+e0BuEz0DaZrC3rOht/TSouZUiyzNIKsy9j7cW8AAdaWkoQOQ68MqXY7iZ+QJaPxkDABL1BE+3VfUNWJ+a643p1I4p861b5K2E/WfzLgOZNa6zfulCo4wgmZraPQbpWZt5EaQVZke3JXp7abnv27t1BZpK3jiVVcXA8tNHn5NbMMWTZ9iDilq/cR+jNiPl9B6PF8xiSF0QuTJauc/fv7+1C81jkoaGyuoqjcZ34TCaRtfb/B7u2iaUEW0bHScAoWBHzeYBkjjFI1uA+EsBHLA3KlvHC0hAQvnxRvOpZ8nGfKUBlnemQdr14J75kK3dchMRvNhE0xhRMNa0Sjh52y0DRJYbemCosnzVTEvckp37MfIkgzWvoXOww6yOENjryF09/jfGV0D4ThElmUlBA7PqeEsxPGvHMO+b0OSJLJtRw0lbUOKWvH6iIl8q6yDdt15aUsj38Z1Rl2tw5u7ex/uwR/7aN1v3di9VqSOspzBvmdfqGu1qj6rrlPFUNA8bCJyaC2mUQrnhJA6nPkwfTYlymZbR6PbgHfuEeKubYj6wtgxoIRKrV5QnuXwhh7Jb7AWEj8Rw7JiFKUqRAOqgCRqPWhRTRVnS1pbPwl7mW/eJ9rGtQdvTFShdKuCQ3IliSZfRstAFmU4/eEpbXb4Yq6ZytS+/3yxBtNAiPPxYsBoG+gcdTB+Ooai0uRK0RU4pw6YzKA3dYy+GAEy0DnqoPd2j5KXrkDWymLEm0yPixtFAGKztuqaJCGp0udqjtiLIevyMkyRb/we6UtJ+DJuYsixRA951aheE+C1t0PfxgXBp7BplJbEKa8KD75MMVH9myIUn28u6u7JogYVgLU6FsXPKV5Xs3ZWneMqnnj1+HyTWJcfio2dOq0fLsi+ZIs+p4a6AxcsZ2g9bMHaW6Z78uBWp9Xm82XzWPWabaLZs80P27ipuArSa6V+wnwtFpHA02dTAICkSABoCrxJ42jd85L/vP2gjWBKP5NUCeycwepZmDybwB8TLUG11At1HIzOws1vFXK6us4bvQYkRVpoGs4dvPjfeUMPoy9GSP0Ukibh8d9+jLd/99uEcOYbPVuD2TfJ2n3eDOIbSi4M33nYuZBGcVHtJRrY15RLiuYcr7sL6DbuXmza3F0XV31+lqijyEVD46L3XMcsKNKdjLYB59SBrMvIkxzezIPRoaHU2Y/OYPUtpElKg2l9oZPDnUa5CHzd5+Jr3zl10GItNLrra86l61z4b1mTwSRWP+B6BXTw61LX3N0z28adiU3Es0p/P3/w2/s2pi+nCGchsoQSTBFds2nHVDEUtB+0Mf5qDMVUCPr8dk/8bv87+2g/pCkRL7Y4leL8s3OMn47RPGxi8ngCxVDWOlhs+tmKlsirNjbBJMDw0yEGPxwAMiW1zhudJaRTdbpfPMZdsNITxdVWT+cbHyWth/mmptr4vK7YCKlRafoWaYir7kkAtToWdZ+TO9Xx4qMqSFxXPFym6csLok2vgfhceiJQSNXcmEWZaEpLsiRet+4cqo3ioh3ppt/FJiLX2/ywjeuKi4royyC9NtLsChfW5QA9r+29ZSvfi2Lt83I+nDJa5BSVRinyNIfaUHHvZ++J2mUToeFNkNPVesXoGGKyXp3KpzHZH7vHLkInJBqXQYglo2OI94udGP6AdMC80EPvnV6JOh66IRSzXmNjXe5clQuvI5byvaFcuHHcxjZeJS7bQHiV52cddXRVzbZJXq07b0VTMPxkiGAaQJZkxF6M6bMpps+ncE4cdB910X7QhtExYO1aJYTdqiZ48VzsPRuJn1wJQcOPc5HWVjGnbNqkeZ3qmrt5Vtu4U3HZZkGxM9056sA9d5EnOYymIQqadRPw2mOaCvrf6tfC6fiEDVg0UhRDweH3DvH47zxGa78Ff+Qj9mOYfRPIceVFuUkTqijY6Jw6aD1oAQzQ24TAqXaq6woXgZKYQ7/rrnspIV0STripw9CmkOltfDNCbGq65U3N1/U9X3Rf192TAGp1LFa9LgkThJMQs+ezlYLE1bWwadP3qutEMZZRkTwnKLoCvalj/GSMxE/E9G/VRq5Og+Iq8OOLPtc2P2zjuuK6i+hN7s3imuCi5dX7G1hunm76nkUxd77JSaIEh7uH5Bh1yenvRvVTDrinLjzJE6ifVVP5JEiQZzm673Ux/mIMAJBkqaQrxifpBz97AM3SxKAOKAvDK9pCEPYiJONtRDCh5rjZMcVn/SZTL7ZxN+JSTee5tISkSoidWGjJbNpELlJH1zVQ6vJqXX1TrX8mzyaQJAmyJKP1oEV6OIaC3ls9OEMHaktdMqS5zOCr80bnSgiaqzwrLuPAep11zU0jfbaZbBsXxmWL/+JmpPduDz30hI5MHc1po4Q1fy2TGIkjBwkSfb2KudEx0H+3T3zsYwfmjonmfrOkI1F87abUJi5suiqKgo0zdYbIj6BZWu3Gq/b1NdBvWS1ft1dJhpdxGNoUMr2Nb0aUNjXq8qbm1s/ngk3LqnvyIqhz9XO2HrTgDb0lnRng8gXDdayTKioSINRRGqdwz1wopgKzbWL327uQFGlZ4O+Cc7js9HCjY27zwzauKa67ObjJvblqTVy1uVR9TwBCpHx2NoM7cMVUutO54oBpTf2UBAnGT8bwzj1kUYbWw5b4+dL03SDk8vCzIZqHTbTutwjNs2uX9C4UYzFJL1IbFF0paYbxjeVt0BQ2oYE6Jw68Uw/umYvmYXMtSnsb2/haIgemz6bIsgzOcwcHP3MgtGQ2berUuV9d1JwANrA4n7/e7JmInAiRF8Hat4CUclpTaeLwpw6XEHb8tcV/F49Zh+K77HosHsc79zZqhK17vlzF1Xmj87wFpM82k23jwrgKdLC4Gem80SkJWdV1hDc9h2ASCDtLLszFk8DSVBpEF7H3baimCklbFjHmiJ4lS9ILzolP2qquMUAZodR+o72ggW2YrOqg30bLKCXGV0mGmxTLq47/k+5Q802Pq2z0b+OcVp3HqvPdRJ9nSYcnSGpFzusm7VX3qU3O6TL6M0vvOdcP48gpWZdhdk1IinTlhs1lp+UXNtfu4L2zjdczrrs5uO7evAh9d9XmUl2OcU4cjB+P4Y996A0drcPWlezWq1GXW7iuRmOngRw5JElaQgkVa5eq4Ohlc24ddfyiPPeqOWJTKp1qqsLe2N63l0RZt7GNrz0Y0DpqkYhwDmi2VuteeZnYpDmxUX6bC6abHROqqaL1oEXUzSApOQmve29Oh78qq6B67OpxvHMP06dTtB60LmyErXvvOnTzddQ1t4Fg3lZc29goLgsdXNdsuPLEy1CElXCWZBh+Sva9mq0Ju+I6Ma/Spq2wKINxgOFnQ6RRimgalSxJq3/LzzsJSbD0Ivu7y6BlloqgYrLJc7JRfTIuN5xqEtLGKKNNppUr/uaym8BtvH7xun3Hdee7yWeo/s0mjlyr3KcAavJyShRvNPO4rP5MtfAqur7IqozOUYdeF5LIZzAJgEkZan3RNbgK7fKiY75u98427mbcRHOwtlmzQS3yKhuP6nsabTKIsPs2ps+mmJ3M0Og2Vh5z0zVa9xkUvcaSHRC0jsiJStPsyzRarppzv04qXYa5vXENYvN1ET3dxjc3FF2BrMoAiOq4yr3yMlGkGmbIVjYnVjY3CgNvniP2PtwTzRvFUFY2Ry+inF4nxcretzF+NkZjt3GhkUcwDqgJ1TZqqa51+f466prbQDBvM9c2rj0uunFfpVOp6OTycv7ZObwzD+E0hNE2EM5C3PvevXoxrzoYYJBg+Bk1hBhjSOMU/siHpEq04OeWg7EfEy9UVwTqiNvwAautfi9b3PD34RuyIhrJG3hwBy767/ZFw6mKlgEuhk0Wz+2yCIZtkbON1yFepTAvTraL/y6uhVXuU8E4wFf/v6/gDlwgBw6+e4D97+yXjnkZ/Zm6wosj86qfbfDJAJPHEyRxgs5Rp/S+667TlWiX2zywjVuK22gObtQQuMZnoWKQ+Lpqq7AOLDT6DXSO6tfVputv1WdQjIWuRhLMPxendaQZnBcODn/mEMGYNje85rnJtc51QrIsQ+zGROuqoYlsGq9CpRPntM1z27gDUbxP2w/ar0wJrKMa1jUnVq0PoQXqRGL/oejKSjfipc+zgnJ6nRSrUqMoZ/DOPCiGsiRRwSMYB3j6d5+CyQx5muPo549uNN/fxnFL73HtR9zGT3xcdOMuLfR8cytsxaDpd+zGxEU/mcE78yDrsnDQ2sgecF70qIaKOIhhdA00+g04///27j1KkvOs7/j36e7Z6bnszOxdkrVrWZaEQQoSBznYPjaSHUM4cMAygmBLMbfj+CDHQOIQLiEyiuxgiE+4BF9iBXO3DDhIjnUSDHZsWfgGyAbFyNi6YK8kJO3u7O5cemZ6Znr6zR9VNVvTU91dXV3VXV3z+5wzR9qu6u737Zl6+q3nvZ2ucf5r59lY2uDgFQdZfmbZ23bceY2F8DSooBes3yF4pXKJ2tM1mo3mjilcwWik0nSJ1TOrrC2sbW/LHnwW26MDQrvWxFq4OsEIBpE867dh3un52wmVNrtP1ZfqNLeaTB6aZHNtk43VnddgL+vPbK5tbm/7uWMh46gRBusNVs+tsvzsMuV9ZRaeXGD2xGzXm6Sk0y4VD6RI4vaathvd00vjvFH3EsXVQ1VKpZK362WbZA60v/52jZrrUIdKtUKVKgsL3pp89cU6k0cnvXaXgY0Zy88ss3Z+jUa9weErD7NR3+hpQdaeODj/D+e3E9/lsfKu0dtJpuLHmSYGbCflwzfLinOShjRHeaXR9o6aatjp+mj3/T9xYCLy/iNOHVo7ndvt3NmLdomi1t0JI0fn+O208clx1pfXt3fwiyp7JgmXjO+pFLUkE+3+cFu3l8OxPfKl22rjwePV2SoTBycoj5W9IDOzb9cOWl3L5w9vrB6sUqlXtrfeXHpiibHJMWqnapSqJcx5AW1jdWPHzghJMsutdcJ5I3zWltZobDYolUpsbW5d6GELDRWevmR6xwieqPpoQVLZy/ptmPfTW1+dqVIql7ZvVPZNXmj4xNlSs3VU3vrSeuT6XLs4WH562VsoeV+F8enxWHWNFS/8RHuj3shs63qRYUraa9pr8jh8fnhKZttkTt1LPASjgcPTqqPet+MIlFBcC2LR2MQYbsuxcnqFrY0t5k7MsXhykeVnl1lfXGfGuq9DkYjB5NFJyuPl7U0uwomqxFPxY45IDDaamDk+401d1WYPkoI0RnmlPVIszlTDuM/vdv/R9jVars00Rqh0ShRF7U64QxPOPXaOymSFEiWec/1zEpUhrxS5JBORi/RFBKxON1Ft54aHLuiZS2dYfGrRS9D0cNMRFRSWHl5ibXHNW/zrwAT7L94PTWhuNSmPlVMbBtm6Vka9WmfpH5eona7htpz3PhFl7DbiRlOkZC/rt2HeT299da7KZTdctmMNnV5vUoJReWZGc6tJczPGYqkGBy87CIDbckxdNBWr4RZnGkJ43vzs8VnFFCmEXteFinyNULtl9Wz3nVXC50PnJEQ4ZoC/KYJ/I9VuJG6n12tdk2/qyBSNeoN90/u2O9WaDe+mrVL12lGlSsnbTbSHpHicEQqVcW/K2frSOs2NJqWx0o71DbMaKbM9GnqshJWNUqW0nUzSZg/SrzT+dtP+++/3niCYDVFfqjN1ZCqVhcTTGqGSJFHUqDdYmV9h9sQszWaTmYtnqEwU61ovVm0kF9rdxEQGrA43UXECXHXOW119R7Y25vStHWVeb3hDoE/MsV5bZ+rIFIevOLx9LK0v+qi1MqpzVQ487wClsRKbtU3qS/UdjbS476spUrKXJWnAtN6ExG0YRJ0TxKId5/baSPPXuLCy0ah7MaljXfxRNNWZKo16g0PPP5So3ruO++WePDi5HaeypoVJJWtp9YIH7ZbVs6veiI8uI1p6STZHjahJ8jrbz4kY/Tf/lXlvOtZMlf2X7Gd81pum0Kg3OP/4edbOrXmdS8dntz+3jvEi5udaqV5Y1wdaFnHPaKRMeLSTmbd2RrOxc8FZtZ2kH2n87Wbx99/P33XrbsV5vka6la1Rb1A7XcMwpg5PsVnfpDw+uFHHg2rb5PO3IyOtbe9Vm9XDu+0us3pulUa9wdSRKe/1IxoP1dlqT421qJEywRbn1bnq9q5ZQCo3SK11ag3a5X1ltja2WDmzQrlazma4s0jB9dLo6DQCsNNz5h+Zp7np9S4fvupwouu97euvNxifG/fW1TpVY3Xei31tY4G/1WmpUqLZaG73OG3v5BBzO9FdU8AGPA1BC5PKIKTVCx60W2qna8zYzHbis93r9TTatk2yKJiy2WnqZqfybu8QuuXFCTNjs75JZbJyYaer9caOeILFuzbri3U2ahtMHJjoug17pVqJXOMri1HGraOdpo5OeUmqPkdai4Sl8beb9DWyShYUZW2p7emWm9791dSRKSqTlR33eIN4/0G0bUbvtyO516n3qt0uVO0aQdPHpjn72Fkq4xUWnlhget1rCEQuFthtJ5lQ4Ou2qwx0H+kTXgsnah2gyC3J2wTtXhqHItJe3AZOkgZLfbFO7ekaY9NjbJ7xdqbrtABxL420oEdsY3GD9ZV1yvvK3vpdK+0XKQ3WArOSbfc4R+3kEE7qxLn56lTuLBqQRWk8Sr6lmaisVL1ESGOtkXqverg9sG9yn7f5w/wqU0enYq81uOt1/brXz9VZPbvKzPEZqvtbOq8i4kmcdlXtVI2V+RVWz6x6U7eSLnia8iiAqKluva4lIhJHGn+7vb5GlsmCrDt1gi3Rgb7WJO36PqHRxtB50eSOr5NgEfxgEfZBtW3UYpLUdeq96vnCNe9iL5VLzD96oWc8EA40HadvRYzIaberTDij26g3OHTFoY693EHiZ/LQhXpC+23Eo4J2L41DTU0QidbTmjUJGyzOHIbhzMU6P24jLbwrxfKzy+BgY2Wj45SOqMRL7XQNK3tDi1fmV3bs5NDLzVdUubNqQGphUhmEtEeB9DJNs9t10/q9Pn10mvpCnflH59msbVLaV2L/xft3JWHjtFeCslZnq5z64inGp8epn61z7OuP7Ti3XX06XZtB3Dp85WHWFta2p1MlmfreTtKbP8UVKbIsO0KyGDEXCEY6Lz+zzNbGFpMHJzl69dFU1ukJv0d485mNlQ3KY10WTe7wWkkXwQ8Wtgcyj0GKbpKJtHqvgi/kteW17R2nmlvNyG3Dw3PFW7UbkRO1q0xjvcHW5hb1c960hfXaOpdcd8mOYLNjnrufhQ3XM0mgDQdQHNuJodbeME1NEInWy3WXpMFSna16N1WNpre7Xoq9veFdJSYPTW4vSNht1F5r4qU6U/V2sJlfwW05qjM741brzVdPPYIZNSCzbDyKhKU9CiTO68UZ5RL1vT59bJrmZpOxE2MsPLHA2vk19k3v3Do43F7ZrG9y9rGzHLvmWGSZGhveulxBsrex0dh1Tmt9ul2bO3bT8bc1TrONEr75M2dMXzLddapr3LLvFWZ2P/AiIPiF/6Nz7uv8Y7cAbwcOAx8FftQ5d84/dhB4H/DtwDzwc865uwdbemkn64Rlp9jWT8dyY71Bs9GkXCmzNr/mbTzzWLlt3OpVVAd+P9Mse233tI4MjLpfzcLejG4SWzjLGeeCSLLIaCfhJE2pUqK51ey6bXh9wdsJIrztb7v1e4J55a0LNTfqDepLdeoLdTC2G0lAZNY3agesOIE2ascNaD+6J5XV9DXCRwqq1wZOrzd3laq3qGcW109UvKxUKz0nxatzVY6/+HjkGjqtN19JtzLNogGZ9o22SF50u26ivtcDpbESpYo3Oidq6+CgvbJZ32SsOrbdoRQ1AggHjTVvt5ftf9fjdTZ1muIVjltpJ30b6w02ahvgvM8i1u5/Hcq+h73JOfeb4QfM7GrgvcB3AV8A7gLeDbzGP+VdwAZwDLgO+N9m9pBz7uFBFVraG1bCst+O5cp4hVKlRH2hzub6JlPHpry2TkodRFEd+EFsgnjroraWt6d2Zcv5WU4p2/G+mb+DjKzwUN6lJ5eYuXSG8r5y24s3ySKjcVSq3iJ61dlq18DVrjHRLvC1S/QcuuIQq2dXKVVK24/VF+vbyaJuWd84gban3cA6lLcXGuEjRTaIBk6WNwi99o63s737nz9VIZwg6ncrU/V4i/Qm7iiX4Hsdx47FfMdnxpk9PrtjmlX4tQ5dcWh7rcHyvt27t4S/9/dfsp/yhLfxwtbGFgsnF3pqB8Rp56Wa9HWwemaVlfkVcHDRtRdp6lR6bgXuc849AGBmtwN/b2b7gSZwM3CNc64GfMrMPgy8DvjZYRVYdkrSHum3U7ffpG3QMVadrbL41CLV2SrlsfR2neoUTxMloHps9wyrnaSoKG0FF22pUsLKRmmshJWs7cWb9cKWcQJXp4RH1PPbJnqqFSYOTrBR26Cx2vCCAt5izKVyibWlNRpzje354mFx53u3TT51qUM/gUKLj0rRFa1HNml9Ot149fP5FO3zFelF0puhTtdNp1Eurc+N3OVzrsqxa461LVfr65XHy2yNbyVqB3RrQ0S1Ufq6gTQ48PwDzF42y+bqJrMnZhV/knm7mf0S8BXg551z9wNXA58JTnDOPW5mG8BVeAmdhnPukdBrPATcMLgiS9rS6NRNY6RuECemj033nfiImunQ66jBbjEqyWjuQccpRUVpa3t4fqOJ23I0N5vbOx90On+YC9AlSXhEXXiN9QYTcxNMHZraXmuiOluldqpG7ekazhylsdKuhE0v873bfV5de/P6CBR5+B2J7AXDntpYX6yzsbLBxFz3rYRFZLfWazjTXWVijnLpNAq5bcKodQrAjNeWSdIOiNOG6JaA6nV6RtB7P1Yd0y5VyfwM8CW86VOvAe4zs+uAaWCx5dxFYD+wBSy1ObaLmb0BeAPAiRMn0iq3pKzd9M5e75nSGoHSz/1M0HleO1VjbGKsY8dVxymvMReWzzu17qSt8EUbtUZMp/OHORw/lWleEWtNVKr+IoWNpneTtLX7JilY7Gt8ahyH6zjfO/x5tS6CnFV2Ny+/I5FBGFZSJcmuCGmWM9jNavX0KitnVth/8X4lb0V6EHUND2qEa6fv6SSdMkHbJbyuVqVaSTzSqJc2RBrTM9Rm6Y9z7i9D//xdM3st8J1ADZhpOX0GWMYbodPuWNR73IW3Bg/XX399vC0gZeDSmo407JG6QXzeWNlg9fQqh6461LbjqlsMibuw/LA76brJX4kkV0ZhmFkW2gWA6myV+kJ9e3Hm1sZUsNjX+so65ozyeNkLAuPtkzrAQLPDRfkdpUG7P+RX3/O8h7heVC83MVHlDF6jnznuwRboa+ejd7OK+/nmvREjkoXIa7hDMiXt66Td93SSBEeQ4LWSUTtV67vTqNtzd3wW/me2enaVxnqD8f3j1OltS/Ney6qY1ZW/zQkPA9cGD5rZ5cA48AheQqdiZlc65x71T7nWf46MqCTTkZLK8joMyj0xN8HKmRVqp2qUxkpMHZmKPL/bCMZuC8uPwvqj+SqNSMrqC/XI3V46CQeh1uG9caZDHb7qsDcv1B8OGN4SPSoAxM0OS2a0+0POpPHlOcz1onrpRW8tZ+vi6/3McW/SZN/07t2s4m7rOQqNGJEstNswIer7f1DXSae2ScfnDTAWRsWWSrXCyvwKlWqFZ/7mma4bbKT5/ns9ZpnZHPAtwCfxOq5+APhW4CeBMeCzZvYyvHbOncA9zrll/7n3AHea2evx2jmvAl4y4CpIyuJMR8qiQw3666jaUYdQG2fi4ARbG1vsq+7bkbCO/VrVGAvLj8D6o/kqDeoxl/TUF+o8+dknsbLhthzHX3y8a1InTmOgW7CoVL1dueqLdS/IdAkAcbcd7VZu9UilSrs/DFEaX57DXC+ql170XUOgof+6xxhiHLzH6tlVzj52lupsdVfMG4VGjEgW2l1DUd//g7hO+klUDDIWbm+mUS6xPL/MyvwKYxNjbC5vMj41HmuDjTTeXzFr2xjwNuAFeOvifBm4KVjs2Mx+DHg/cAj4GPAjoee+Efgt4DRwFrhNnVbpSNJmz6KdHxXn4saaTuXpp6MqTj3D5R6fGWd9ab2va77bwvKjsP5oadgFaONNzrlp/ydI5gQ95q/D6xVfxesxD4R7zG8F3uM/R/ao+lIdKxtTh6ewslFfqnd9TjgIBQ2OpOIGgCA7PHFwguqBamR2uJP6Qp1Tf3eKxScXWTi5QKOevMx71NvNbN7MPm1mN/qPXY23owPg7f6AF1+u8n+idn+IjDdm9gYze9DMHjxz5kwW5S+cNHdRmDo6NZRe2kq10nGXu/B54XIGiZV+Gw6d3j/8+QaNn6iYF+f3EIxEVNyJx8zuN7O6mdX8n6+Ejt1iZifNbMXMPuR3VAXHDprZvf6xk34Hl2Qo9jU8gMZ+P22TQcbCyniFzbVN5h/1Noeon6t7C5aaY2N1Y3uDjV4/p7hxZhRuvAbJOXfGOfdC59x+59ycc+5FzrmPho7f7Zw74Zybcs69Kugk94+dc87d5B87oU7ydAQJk5XTK7Hb7EmeE1drnIsTa7qVp1NHVaf41Us9g3IPpM005PZkHPkrUXvqMR9hwxhBUp2p4rYcK/MruC1vZ4du+mkMdNs6r1O9g+xwsN15L+95+uHT1E7XGN8/zvRF0+qR6k3muz9oscDeJVknot3r5OFa6HVLzKwXAQ1/vlNHptrueNN1pI+mNySlaZ4FknRdm552lukzUTGoWFip+ptHbDaZOzHH4slFNtc22X/xfm/61Xhle3oneLvxxVnHK26cSeu7QyQrSUaRDXLkWayOnC7lab0Owet87ha/ktRzUNd8XtqT7eS1ZG83s18CvgL8vHPufrze788EJzjnHjezoMe8SXSP+Q2tL6yt9QZvWI3+6lyV4y8+3tMaOkkDQ7s6dgsA4UYdsD0kMVhzBzrPOa0v1lk9t8rm2ib1c3UqE95oH4lnELs/SDJ5+/JMmpROEv8GUffwe1Sq7Xe86VQWTW9IlTqtRlgv12zSmJB10iiubq8bbB5hJWP6kmmmj03v6vnu5TPoNc7k7btDJCzRTnUDHHkWJ9bEKU+Sjqqk9dQ1n8+ETqY95uotH7xhNvqrc/ESOVGja3qRKOPe0qCpzlUTzTkt7yuz/9h+6ot1Zi+d3fNBrU/a/UF26ScpXV+ss1HbYOLARNttNeOWIaseqKSNIU1vSCyzTivJvx1rWJ1bpXa6xvTR3TvRtQpfp902fMiqIy3uOoPdbt56aTMpzkiRJEnODnrkWbc2QdI65GW0TRHl7pNSj3nxJP0yHtQ0rTQaPknq2NqggZ2rzUP3xVGrs1X2X7yfZqPJ5JFJb7caiUW7P0hcSZPSwXbBK/MrrJ5ZZfqS6UQ3I3md2qTGVyKZT/PUSOR8297K+9wqS08uMXPpDI21RuR1HdUOirPhQ1YdaXFft+sNYQ9tJsUZKZoknSh5G4WSVXnyVs9RMQqfmHrMR1zSocKDuoFJZUedJNnqUINmc22T8ZnxHdsHQ/c5p5Wqt026GjqJaPcHiSVxUnq9wdjEGIevPMzawtr2Fr69aNQb1E7X2NrcYvLgZO6mNqnx1ZtBdFppJHK+Be2F2ukaM5fOMHko+rpu1w4Kb/iwMr/ijdRpSehkNaolrdfttc2kOCMiYdrdd6dcfQLqMS+WfqYx5W0BsFiv02MdgwZNfbHO5tom60vr1Jv1HcmrWHNO1dBJxDl3Bnhhh+N3A5G7Ovg7QdyUTckkb7rdfLRrWASxpUmTfVP7qM52n/7Z+roLJxfY2thi6SlvgEZ5rLdd8CT31Gm1B1WqFaaPTtNYa7Rte7RrB8XZ8CGrUS1pvq7aLiKSRF5HLQ9T3mqvHvOC6Pdiy9sCYFm+d2W9wtjEWGTyKk6DR1lqkey1uxY7xbpK1dvxJVjrIunaXJOHJgG8UXwx1tqQfFKnlYR1a3u0awfF3fAhzYRJv+sMioikZZCd/qNyj5WrkqnHvDj6vdjytgBYpu/d51bpylKLDE+nWBesoWMlo3aq1vsovlBsKO8rK5kz+tRpJTt0igmd2kFxN3xIQ97bGaNywyUi6RhUp3/eY19YPksluZLkyzKNi63fJEuevuQ7laWf5FVwM1kql1hbXqO+WGe6qoWRRQalU6wbRGI7T3FOOlOnlfQqD6NhhrlTabf4Nko3XCKSjkF1+g8z9vUqn6WS3Ej6ZTnsXQny9CUfd5vPJOWrjFfYXNtk+ZllzBmlSonqbO9TO0RGSZ6SGJ1iXdaJ7TzFOREppmFtGx4nvo3SDZfIKMpTeyssqm2UdlmHFfuSyG/JJBf6+bIc5Aib1nPz9CWfZVmCNTqam00mDkzQ3GqqQSOFlsckRrtYl3ViO09xTkSKaVgddHHi2yjdcImMmn7aW4NOBGXRNsy8DZfiZ6TIJx3luWem47k5+pLPuizV2Sr1hTrNrebQ6yqStVFLYmQ5ZSJPcU5E8iP1nuohTP2KE9+GPRpcpMiStreSJFf6jVlZtQ2zin1pJ6AU+aSjPPfMBBd/o7773OpsNTdf8ll8huHAB2wvjqjpVlJ0SmJcoJsZEWmV9EYhb1Mr4sa3PKwzJFJESdtbvSZX0khujFrbMO0EVL5rK7mQx56Z8MW/uba5/Xj43Dx9yae9fWhr3ccmxnBNR3V2MLteiAyLkhg75SnOicjwJblR6PeGKqtkkOKbyPAkbW/1mlxJI7kxam3DtBNQ+a6t7FndLszwxQ8wPjPuffGPwEXcr/pinY2VDSbmJmhuNsEYmeknImlQI19EJFqSG4V+bqiyXNcsb6OGRPaaJO2tXpMraSU3RqltmHYCajRqLXtSpwuz9eLfK1ONGvUGtVM1lp9a5vxXzzN1bIrqTHVkhhiKiIhIdpLcKPRzQ9UpGdRPQiarRJGSRCLZ6yW5Mmqja9KSZgJqb3xiUjh79eIPGk6VyQpby1uUrMTciTkw9tTnICIiItF6vVHop03VLhnU9zSuDBY5zeMuiSLSPWYpEduZPhEZWaM0tC4tlfEKjXoD13RMHZpifGYcDK2dIyIiIoklbVO1Swb1m5DJYpHTUdslUUSUiI1Dn4bICKlUKxy64hBnHztLZbxCeV9Z06xERERkaKKSQf0mZLIYiT1qO+GIiBKxcejTEIkpL8P9qnNVjl1zLBdlEZFiyku8E9lrinLtpZGQSXsk9l6dri8yypSI7U6fiEgMeRvutxenm4lI+qJuHvMW70RGSR4WAu63DKntvFLdGVfykEhR+0mku7xcr6BEbBylYRdAZBSEh/tZyWisN4ZdJBGRvgQ3jyunV1g4uUCj7sU1xTuRZNpdU7Gfn8K1108Z+i3/oF9X4jOzK82sbmZ/EHrsFjM7aWYrZvYhMzsYOnbQzO71j500s1uGU3IZtDxer5VqZc/saJyEEjoiMWi4n4gUTfjmcWtji9rpGo16Q/FOJKF+EzJpXHv9lCGrZK6SxLnwLuCvg3+Y2dXAe4HXAceAVeDdLedv+MduBd7jP0cKTtfr6FErTYYmT8P5utFwPxEZlEHExka9QaPeYHNtk0a9wdJTS8zYDI21BnPPnVO8E0kgq4WAe4kJ/ZQhq2SuksTDZWavARaAzwBX+A/fCtznnHvAP+d24O/NbD/QBG4GrnHO1YBPmdmH8ZI/Pzvg4suAjer1Okr3lWnbW7WV3BjFNRo071pEsjaI2Bh+D4BytczM8RkmD05u7yChoc0ivctiIeBeY0I/Zciq80qdYsNjZjPAncArgNeHDl2Nl+ABwDn3uJltAFfhJXQazrlHQuc/BNzQ5j3eALwB4MSJE6mWXwZvFK/XUbyvTJOmXMlQaDifiMhug4iN4fcYmxijOlOlPFYeud44kTxKe62HJDGhnzJktVaF1sAYmrcC73POPdXy+DSw2PLYIrDfP7bU5tguzrm7nHPXO+euP3LkSApFlmEbtet1r99XjsZvSQpnVIfziYhkaRCxsfU9qrNVqrPVkeqNE9kr1F6SpMzsOuCVwDdFHK4BMy2PzQDLeCN02h0TyZ29Hif3Vm0lN0ZxOJ+ISNYGERvbvYfisEj+qL0kfbgRuAx4wszAG3lTNrNvAD4CXBucaGaXA+PAI3gJnYqZXemce9Q/5Vrg4YGVXKQHez1O7q3aSq5oTRoRkd0GERsVf0VGh65XSegu4A9D//4pvATPbcBR4LNm9jLgC3jr7NzjnFsGMLN7gDvN7PXAdcCrgJcMrOQiPdrLcVJr6IjInmdmV5pZ3cz+IPTYLWZ20sxWzOxDZnYwdOygmd3rHztpZrcMp+QiIiIiuznnVp1zzwY/eNOs6s65M865h4EfA94PnMZbH+eNoae/EZjwj30AuM1/jojkTC4TOrq5EpEBexfw18E/zOxq4L14W3QeA1aBd7ecv+EfuxV4j/8cERERkdxxzt3hnPuXoX/f7Zw74Zybcs69yjl3LnTsnHPuJv/YCefc3cMptYh0k8uEDrq5EpEBMbPXAAvA/w09fCtwn3PuAedcDbgd+F4z229mU8DNwO3OuZpz7lPAh/Hik4iIiIiIyEDkLqGjmysRGRQzm8GbN/7mlkNXAw8F/3DOPY6XNL7K/2k45x4Jnf+Q/xwREREREZGByFVCZxA3V2b2BjN70MwePHPmTJrFF5HR81bgfc65p1oenwYWWx5bxJtjPg0stTm2i2KOiIiIiIhkIVcJHQZwc+Wcu8s5d71z7vojR46kUGQRGUVmdh3wSuBXIw7XgJmWx2aA5S7HdlHMERERERGRLORmb6/QzdU3RRzudAPV7HCso89//vPzZnay58LudhiYT+F18kL1ybdRqc9zh12ALm7E277zCTMDLzlcNrNvAD4CXBucaGaXA+PAI3gxp2JmVzrnHvVPuRbouvtDCjFnVH73cRSlLkWpB4x+XfIecwYuxXZOYNT/RsKKUhfVY3gUc1p0iTmj+DvupGj1AdUpz9rGm9wkdBjCzZVzLpXucjN70Dl3fRqvlQeqT74VrT5DdBfwh6F//xReDLoNOAp81sxeBnwBbyroPc65ZQAzuwe408xeD1wHvAp4Sbc37DfmFOl3X5S6FKUeUKy6iCetdk6gSH8jRamL6iF50inmFO13XLT6gOo0qvKU0Bn4zZWI7F3OuVW8HfMAMLMaUHfOnQHOmNmPAe8HDgEfA34k9PQ3Ar8FnAbOArc557omkUVERERERNKSm4SObq5EZJicc3e0/Ptu4O42554Dbsq+VCIiIiIiItFyk9BpNWI3V3cN8b2zoPrkW9HqI/EV6XdflLoUpR5QrLpINor0N1KUuqgeMiqK9jsuWn1AdRpJ5pwbdhlERERERERERKQHedu2XEREREREREREulBCR0RERERERERkxCih0yMzu9/M6mZW83++Ejp2i5mdNLMVM/uQmR0cZlm7CdUh+Nkys9/wj11mZq7l+O3DLnOYmb3JzB40s3Uz+52WY//MzL5sZqtm9gkze27o2LiZ/ZaZLZnZs2b25oEXPkK7+pjZi8zso2Z2zszOmNkHzezi0PE7zGyz5Xd1+VAqIakpQqwZ5RhTpPii2CJJKAYNV1FikOLP3lOE2BE2ynEkUJR4EqbYcoESOsm8yTk37f98HYCZXQ28F3gdcAxvx653D7GMXYXqMA1cBKwBH2w5bS503lsHX8qOngbehrfD2TYzOwzcA9wOHAQeBP4odModwJXAc4GXAz9tZt8xgPJ2E1kf4ADegl6X4ZV5GfjtlnP+KPz7dM79Q9aFlYEY6Vgz4jGmSPFFsUWSUgwanqLEIMWfvWmkY0fYiMeRQFHiSZhiiy+3u1yNoFuB+5xzDwD42dm/N7P9zrnl4RYtlpvxtn3/i2EXJC7n3D0AZnY9cGno0PcCDzvnPugfvwOYN7MXOOe+DPwQ8MPOufPAeTP7H8APAx8ZYPF3aVcf59yfhs8zs3cCnxxs6SRHRjXWjFSMKVJ8UWyRlCkGDUBRYpDij4SMauwIG6k4EihKPAlTbLlAI3SSebuZzZvZp83sRv+xq4GHghOcc48DG8BVgy9eIj8E/J7bve3ZSTN7ysx+28/ijoLW38UK8DhwtZkdAC4OH/f//+qBlrA/3wo83PLYd/tDCx82s9uGUSjJRJFiTVFiTJHji2KLtFIMyp+ixiDFn2IpUuwIK0ocCRQ1noQVPrYoodO7nwEuB56DN5zrPjN7PjANLLacuwjsH2zxeufPlbwB+N3Qw/PAC/GGqn0zXj3eP/jSJdLpdzEd+nfrsdwzs28E3gL8+9DDfwx8PXAE+FfAW8zstUMonqSrMLGmYDGmkPFFsUUiKAblU+FikOJP4RQmdoQVLI4EChdPwvZKbFFCp0fOub90zi0759adc78LfBr4TqAGzLScPoM3by/vXgd8yjn31eAB51zNOfegc67hnDsFvAn4djMbhYu40++iFvp367FcM7MrgD8FftI5tz3U0zn3Jefc0865LefcZ4BfB75vWOWUdBQs1hQpxhQuvii2SBTFoNwqVAxS/CmegsWOsCLFkUCh4knYXootSuj0zwGGN5Tr2uBBf7XsceCRIZWrFz/IzmxzlGBo4Sj8zbT+LqaA5+PNET0PPBM+7v9/61C8XPF7BT4GvNU59/tdTg/+JqVYRjnWFCnGFCq+KLZIDxSD8qEwMUjxZ88Y5dgRVqQ4EihMPAnba7FlVP7YcsHM5szsn5tZ1cwqZnYr3ry8j+ANr/tuM3uZfzHcCdyT90W+zOwleEMiP9jy+LeY2deZWcnMDgH/DbjfOdc6LG9o/N9BFSgD5eD3AtwLXGNmN/vH3wL8P39xL4DfA/6jmR0wsxfgDbf7nSFUYYd29TGz5wAfB97pnPvvEc97lV8XM7N/CvwE8L8GW3pJU5FizajGmCLFF8UW6ZVi0PAVJQYp/uwtRYodYaMaRwJFiSdhii0hzjn9xPzBm2v313hDzRaAzwHfFjp+C/AEsIL3h3Fw2GWOUaf3Ar8f8fhrga/6dXkG74K+aNjlbSnjHXhZ1fDPHf6xVwJfxtta8H7gstDzxvG2uFsCTgFvHnZdOtUH+AX//2vhn9DzPgCc9R//MvATw66Lfvr+WyhMrBnVGFOk+KLYop8EfzOKQcMvdyFikOLP3vopUuxoqddIxpFQOQsRT+LUaS/GFvMrJiIiIiIiIiIiI0JTrkRERERERERERowSOiIiIiIiIiIiI0YJHRERERERERGREaOEjoiIiIiIiIjIiFFCR0RERERERERkxCihIyIiIiIiIiIyYpTQkdSZ2R1m5hI+934z+1SM824yszfHfM0bzcyFfi4LHfuamf1BkrJGvM9LW97nijReV0Q6U8xRzBEZJMUcxRyRQVLMUczpRAkdycJvAi/O+D1uAmIFnZB/jVeuZ1Ivjech//XfltHri0g0xRwRGSTFHBEZJMUcaasy7AJI8TjnngKeGnY5InzJOfe5rF7cObcMfM7MXpDVe4jIboo5ijkig6SYo5gjMkiKOYo5nWiEzh5kZt/sD1t7aeixH/cfe1vosSv9x74r9NjzzOz9ZnbGzNbN7G/N7NUtr79rWKCZHTGzD5jZkpmdN7PfNrPv8V//xogyvtLMvmBmq2b2d+H3MLPfAX4IeE5oCN7X+v9kwMzKZnaXX85Xhh5/rZl92czqZvZFv+z3m9n9abyvSJEp5rSnmCOSPsWc9hRzRNKnmNOeYk72NEJnb/obYAF4BRDMqXwFsOb/l9BjDeABADM7DvwlcBr4t8AZ4AeAPzGzm5xzH+7wnvcA/wT4OeAx4GbgN9qc+3zg14G3A/PAvwM+aGYvcM49BrwVOAK8EPge/znrMerdkZlNAB/AG9p3o3PuC/7j3wa8H/gw3lDEI8CvAVXgkX7fV2QPUMyJoJgjkhnFnAiKOSKZUcyJoJgzGEro7EHOuaaZPQC8HLjTzErADcB7gJ8ws2nnXM0//nl/uBvAHYABNzjnzvqP/ZkfjO7Euyh3MbNvB14K/IBz7o9Dz/swcCLiKYeBb3XOPeo//wt4czP/BfCLzrnHzewMsJHWMD8zOwDcB1wMvMQ593jo8H8CvgS82jnn/PP/DngQBR2RrhRzIsuomCOSEcWcyDIq5ohkRDEnsoyKOQOiKVd718eBF5tZFbgOmAP+C1429mX+OS8HPhF6zncA/wdYNLNK8AP8GXCtmc20ea8XAVvAvS2P/8825z8aBBwA59xpvMx1VIBKwyV42fRJWgKOmZWB64E/CQKOX6bPA1/NqDwiRaSYc4Fijkj2FHMuUMwRyZ5izgWKOQOkETp71yeAceAlwDcBDznnTpm3rd3LzewJ4ChecAocBX7Q/4lyCFiKePxi4LxzbrPl8VNtXudcxGPreMPwsvCNeGX/Wedca5kOA2N4Qa9Vu/KLyG6KORco5ohkTzHnAsUckewp5lygmDNASujsXV/Em0P5CrygEwSXj+MNv3sS2AA+HXrOWeAvgF9u85pPt3n8GeCAmY21BJ5jyYqeuo/gbYv3y2ZWd879eujYPLCJF3BbHQOeGED5RIpAMecCxRyR7CnmXKCYI5I9xZwLFHMGSAmdPco55/xVxL8N+Hrg3f6hj+MtmLUE/JVzbjX0tI/gLWr1sHNurYe3+xxQBl4N/HHo8e9PVnrAyypP9PH8HZxz7zCzLeDXzKzknPtV//EtM3sQuNnM7gjN8/xm4Hko6IjEopizk2KOSLYUc3ZSzBHJlmLOToo5g6OEzt72CeBdeHMw/8J/7G+AZfxFvVrOfwvwV8ADZvZO4GvAAeAa4HLn3I9GvYlz7s/N7NPAXWZ2GG8l9u8DrvVPaSYo+5eAg2Z2G94CWnXn3BcTvE64nL/iB55f9QPPf/UP/QLw58C9ZnYX3lDBO4BnE5ZdZK9SzNlZTsUckWwp5uwsp2KOSLYUc3aWUzFnALQo8t4WLMr1oHNuCbysKfDJluP4x57AW8TqIeAXgY/ird5+Azvng0Z5NV4W+pfxMslV4Hb/2GKCsv8m8Id+Of4KbxX1vvlDAn8ceIeZ/bT/2EeBW/Gy7fcCP4O33d+zJCu7yF6lmNNCMUckU4o5LRRzRDKlmNNCMSd7FlpcWmSg/Ez0jwAHnXPrGb7PjXgB9JXAJ51zjRRe81K8bPh/ds69NfR4BW9hs/cBVzrnHuv3vUQkHYo5IjJIijkiMkiKOXuTplzJQJjZDwOzwMPAPrxt+m4D3pFlwGnxMb8sz3POfS3uk8xsAvgV//nzwOXATwOreNns4LyXcmF4pYgMkWKOiAySYo6IDJJijgSU0JFBWQH+DfB8vC39vgr8B+AdA3jvzwMvDP273Yrx7WwBFwHvxNuCbwUvuHy/c+6Z0Hl/2/I+T/ZcUhFJi2KOiAySYo6IDJJijgCaciUiIiIiIiIiMnK0KLKIiIiIiIiIyIhRQkdEREREREREZMQooSMiIiIiIiIiMmKU0BERERERERERGTFK6IiIiIiIiIiIjBgldERERERERERERsz/B/CSi4fBri4dAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1152x288 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# visualising the knapsack set instances with an emphasis on the Pareto fronts\n",
    "\n",
    "nplots = n_instances\n",
    "ncols = 4\n",
    "nrows = nplots // ncols + np.sum(nplots % ncols != 0)\n",
    "\n",
    "plt.figure(figsize=(4 * ncols, 4 * nrows))\n",
    "\n",
    "for idx_instance in knapsacks_instances:\n",
    "    plt.subplot(nrows, ncols, idx_instance + 1)\n",
    "    \n",
    "    plt.scatter([knapsacks['weight'] for _, knapsacks in knapsacks_instances[idx_instance].items()],\n",
    "                [knapsacks['value'] for _, knapsacks in knapsacks_instances[idx_instance].items()],\n",
    "                s=10, alpha=0.1, color='purple')\n",
    "    \n",
    "    plt.scatter([knapsacks_instances[idx_instance][idx_knapsack]['weight'] for idx_knapsack in pareto_instances[idx_instance]], \n",
    "                [knapsacks_instances[idx_instance][idx_knapsack]['value'] for idx_knapsack in pareto_instances[idx_instance]], \n",
    "            marker='x', s=100, linewidth=4, color='green', label='Pareto Front')\n",
    "\n",
    "    plt.xlabel('weight [kg]')\n",
    "    plt.ylabel('value [£]')\n",
    "    plt.title(f'Knapsacks Instance {idx_instance}', fontsize=MEDIUM_SIZE)\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Did you manage to correctly identify the Pareto front for each instance? \n",
    "\n",
    "We clearly see that each instance is a subset of the full space resulting in wildly different Pareto front approximations. \n",
    "\n",
    "Next we will see that we can combine the four instances to one to obtain a mutual Pareto front approximation which is better than each individual."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:53.634239Z",
     "start_time": "2020-11-11T07:22:53.631104Z"
    }
   },
   "outputs": [],
   "source": [
    "# knapsacks_pareto_combined is a set of indexes of that knapsacks that are pareto optimal of the combined set of all instances \n",
    "\n",
    "knapsacks_pareto = {}\n",
    "instance_pareto_details = {}\n",
    "for idx_instance in knapsacks_instances:\n",
    "    for idx_knapsack in pareto_instances[idx_instance]:\n",
    "        idx_pareto = f'{idx_instance}-{idx_knapsack}'\n",
    "        knapsacks_pareto[idx_pareto] = knapsacks_instances[idx_instance][idx_knapsack]\n",
    "        instance_pareto_details[idx_pareto] = idx_instance\n",
    "\n",
    "\n",
    "knapsacks_pareto_combined = knapsacks_to_pareto_front(knapsacks_pareto)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:53.967160Z",
     "start_time": "2020-11-11T07:22:53.773557Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x11e2ea850>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# visualising each instance pareto front (light Xs) and the combined pareto front (thick Xs)\n",
    "\n",
    "plt.scatter([knapsacks['weight'] for _, knapsacks in knapsacks_pareto.items()],\n",
    "            [knapsacks['value'] for _, knapsacks in knapsacks_pareto.items()],\n",
    "            marker='x', s=100, alpha=0.7, color='lightgreen', label='knapsack in a Parto Front of an instance')\n",
    "\n",
    "plt.scatter([knapsacks_pareto[idx_knapsack]['weight'] for idx_knapsack in knapsacks_pareto_combined], \n",
    "            [knapsacks_pareto[idx_knapsack]['value'] for idx_knapsack in knapsacks_pareto_combined], \n",
    "        marker='x', s=100, linewidth=4, color='green', label=f'Pareto Front when combining all {n_instances} instances')\n",
    "\n",
    "for idx_knapsack in knapsacks_pareto_combined:\n",
    "    plt.annotate(instance_pareto_details[idx_knapsack], [knapsacks_pareto[idx_knapsack]['weight'] + 0.5, knapsacks_pareto[idx_knapsack]['value']], fontsize=20, color='green')\n",
    "\n",
    "plt.xlabel('weight [kg]')\n",
    "plt.ylabel('value [£]')\n",
    "plt.title(f'Pareto Knapsacks')\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This visualisation clearly shows that combining multiple instances/subsets to one results in a Pareto front approximation that is better than those of each individual.\n",
    "\n",
    "\n",
    "The take away from this is it that is recommended to combine solutions from different initial seeds.  This approach is called ***niching***. Niching is a manner in which you can explore more effectively the Decision Space without runing into local minima/maxima in the Objective Space. Local minima (when minimising) and maxima (when maximising) is when an algorithms is stuck at a result that is optimal locally, but not globally, e.g: \n",
    "\n",
    "<img src=\"https://upload.wikimedia.org/wikipedia/commons/6/68/Extrema_example_original.svg\" width=300>\n",
    "\n",
    "Another way of thinking about niching is that it is a mechanism to obtain a more diverse set of solutions which results in a better Pareto front approximation.\n",
    "\n",
    "\n",
    "As an over-simplistic analogy consider our ancestors leaving the [Cradle of Humankind](https://en.wikipedia.org/wiki/Cradle_of_Humankind) about 50,000 years ago. In effect they seeded different populations in Europe, Far East and Australia while others remained in Africa. Each developed its own culture to make the world a richer place in terms of diversity, as opposed to everyone being the same. This analogy is imperfect in many ways, but hopefully gives a picture to have in mind. \n",
    "\n",
    "<img src=\"https://images.ctfassets.net/cnu0m8re1exe/3O1R69w3U6bLrBS6ig5J9E/a875a35bd1ae07693908b12ca2e5e73a/bae1HR-1024x532.jpg?w=650\" width=500>\n",
    "\n",
    "\n",
    "This analogy is our segue way to the next topic of having an initial population \"learn\" to create more fit ones ( \"learning\" doesn't mean that the algorithm will understand, but rather improve its results based on random actions). In the next section we will learn about Genetic Algorithms.   "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Genetic Algorithm Components\n",
    "\n",
    "\n",
    "## Genetic Algorithm\n",
    "\n",
    "**What?**   \n",
    "Inspired by Darwin's theory of evolution, in 1960 John Holland introduced [Evolutionary Computation](https://en.wikipedia.org/wiki/Evolutionary_computation) as a family of algorithms for optimisation. The main idea is to have a population-based trial and error problem solvers with a stochastic optimisation character.  \n",
    "\n",
    "[Genetic Algorithms](https://en.wikipedia.org/wiki/Genetic_algorithm) is one technique which is commonly used to generate high-quality solutions to optimisation and search problems. It relies on biologically inspired operators such as mutation, crossover and selection, which we will explore in detail.  \n",
    "\n",
    "Perhaps it is best to have in mind a quote from the 17th century English writer and gardener John Evelyn:  \n",
    "\n",
    "<img src=\"https://www.azquotes.com/picture-quotes/quote-explore-everything-keep-the-best-john-evelyn-121-91-15.jpg\" width=500>\n",
    "\n",
    "\n",
    "**Why?**  \n",
    "We will use it for the knapsack problem because:  \n",
    "* It works well for intractable combinatorial spaces  \n",
    "* Diversity, Niching - it finds multiples solutoins and we start from different starting points. \n",
    "* It is simple to implement\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "There are many ways to construct a Genetic Algorithm, e.g, in the following illustration.\n",
    "\n",
    "<img src=\"http://aqibsaeed.github.io/img/ga.png\" width=500>  \n",
    "\n",
    "(image credit: [Aaqib Saeed](http://aqibsaeed.github.io/2017-08-11-genetic-algorithm-for-optimizing-rnn/)).\n",
    "\n",
    "We will now explore the various components of the Genetic Algorithm with an emphasis on the role of Pareto Optimisation in the selection function.  \n",
    "\n",
    "Troughout we will refer to a knapsack as an ***individual*** wherase a set of them is a ***population***.  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##  Initial Population\n",
    "`create_knapsacks` is the same function we used in [Niching](#Niching).  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:54.976680Z",
     "start_time": "2020-11-11T07:22:54.973334Z"
    }
   },
   "outputs": [],
   "source": [
    "N_INITIAL_POPULATION = 20\n",
    "seed = 3141592653\n",
    "\n",
    "population = create_knapsacks(N_INITIAL_POPULATION, seed=seed)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Cross Over Algorithm\n",
    "\n",
    "One of two transformation functions used in the Genetic Algorithm.  \n",
    "\n",
    "The following image uses DNA snippets to illustrate the how the crossover tranformation may be generate two children solutions from two parents. \n",
    "\n",
    "![crossover](https://user-images.githubusercontent.com/6064016/97842646-dc0c2400-1cdf-11eb-8b59-f0c70274d606.png)\n",
    "\n",
    " \n",
    "\n",
    "In this analogy:  \n",
    "* A DNA strip corresponds to a knapsack\n",
    "* The nucleotides (`A`, `C`, `G`, `T`) correspond to the packages (except for the fact that we are using unique packages).\n",
    "\n",
    "Here we present a few functions \n",
    "* `cross_over_2_indvs` - generates two children by crossing over (mating) two parents \n",
    "* `crossover_knapsacks` - generates `len(knapsacks)` children for input `knapsacks`  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Parents To Children"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:55.862073Z",
     "start_time": "2020-11-11T07:22:55.857314Z"
    }
   },
   "outputs": [],
   "source": [
    "# Cross over of a pair of parents to a pair of children\n",
    "\n",
    "def cross_over_2_indvs(ind1, ind2):\n",
    "    # here we assume:\n",
    "    # both individuals are non empty lists with the same length\n",
    "    # a crossover cuts through the middle of the lists\n",
    "    # Caveat! the [n_//2:] operation below means that we do not want the packages to be sorted because that is likely to cause drop in diversity.\n",
    "    \n",
    "    assert len(ind1) == len(ind2)\n",
    "    \n",
    "    mutuals = list(set(ind1) & set(ind2))\n",
    "    \n",
    "    non_overlap1 = [ind for ind in ind1 if ind not in mutuals]\n",
    "    non_overlap2 = [ind for ind in ind2 if ind not in mutuals]\n",
    "    \n",
    "    ind1_ = non_overlap1.copy()\n",
    "    ind2_ = non_overlap2.copy()\n",
    "    \n",
    "    n_ = len(ind1_)\n",
    "    \n",
    "    ind1_[n_//2:] = non_overlap2[n_//2:]\n",
    "    ind2_[n_//2:] = non_overlap1[n_//2:]\n",
    "    \n",
    "    ind1_ += mutuals\n",
    "    ind2_ += mutuals\n",
    "    \n",
    "    return ind1_, ind2_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:56.145812Z",
     "start_time": "2020-11-11T07:22:56.141088Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "parent 1: A, B, C, D, E\n",
      "parent 2: 1, 2, 3, 4, C\n",
      "\n",
      "crossed over to\n",
      "child 1: A, B, 3, 4, C\n",
      "child 2: 1, 2, D, E, C\n"
     ]
    }
   ],
   "source": [
    "# dummy example - for this example we are not using packages in knapsacks but rather characters in a list\n",
    "\n",
    "parent1 =  ['A', 'B', 'C', 'D', 'E']\n",
    "parent2 = [1, 2, 3, 4, 'C']\n",
    "\n",
    "print(f'parent 1: {\", \".join(parent1)}')\n",
    "print(f'parent 2: {\", \".join(map(str,parent2))}')\n",
    "\n",
    "child1, child2 = cross_over_2_indvs(parent1, parent2)\n",
    "print(f'\\ncrossed over to\\nchild 1: {\", \".join(map(str, child1))}\\nchild 2: {\", \".join(map(str, child2))}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**A deeper look**\n",
    "\n",
    "In this dummy example the `C` character appears in both parents and is passed on to both children (rather than one child having two `C`s and the other none). This corresponds to the `mutuals` term in the `cross_over_2_indvs` function. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:57.122303Z",
     "start_time": "2020-11-11T07:22:57.117430Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "parent 1: 14, 12, 4, 1, 5, 9, 19, 7, 15, 13\n",
      "parent 2: 0, 17, 12, 19, 1, 15, 7, 9, 11, 4\n",
      "\n",
      "crossed over to\n",
      "child 1: 14, 17, 11, 1, 4, 7, 9, 12, 15, 19\n",
      "child 2: 0, 5, 13, 1, 4, 7, 9, 12, 15, 19\n"
     ]
    }
   ],
   "source": [
    "# knapsack working example\n",
    "\n",
    "idx_indv1 = 0  # change me! index of knapsack\n",
    "idx_indv2 = 1  # change me! index of knapsack\n",
    "# -------\n",
    "\n",
    "parent1 = population[idx_indv1]['packages']\n",
    "parent2 = population[idx_indv2]['packages']\n",
    "\n",
    "print(f'parent 1: {\", \".join(map(str,parent1))}')\n",
    "print(f'parent 2: {\", \".join(map(str,parent2))}')\n",
    "\n",
    "child1, child2 = cross_over_2_indvs(parent1, parent2)\n",
    "print(f'\\ncrossed over to\\nchild 1: {\", \".join(map(str, child1))}\\nchild 2: {\", \".join(map(str, child2))}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###  Population\n",
    "`crossover_knapsacks` is a wrapper for `cross_over_2_indvs`.  \n",
    " \n",
    "\n",
    "It separates the input knapsacks to pairs of parents and then runs each pair through `cross_over_2_indvs`. This results in `len(knapsacks)` children. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:57.685523Z",
     "start_time": "2020-11-11T07:22:57.680302Z"
    }
   },
   "outputs": [],
   "source": [
    "def crossover_knapsacks(population, packages, idx_start=None):\n",
    "    n_population = len(population)\n",
    "    \n",
    "    idxs_parents = list(population.keys())\n",
    "    \n",
    "    idxs_parent2 = idxs_parents[1::2]\n",
    "    idxs_parent1 = idxs_parents[::2][:len(idxs_parent2)]\n",
    "    \n",
    "    children = {}\n",
    "    \n",
    "    if idx_start is None:\n",
    "        idx_child_counter = np.max(idxs_parents) # TODO: this does not generalise well since indviduals will be dropped\n",
    "    else:\n",
    "        idx_child_counter = idx_start\n",
    "        \n",
    "    for idx_parent1, idx_parent2 in zip(idxs_parent1, idxs_parent2):\n",
    "        idx_child_counter += 1\n",
    "        children[idx_child_counter] = {}\n",
    "        children[idx_child_counter + 1] = {}\n",
    "        \n",
    "        children[idx_child_counter]['packages'], children[idx_child_counter + 1]['packages'] = cross_over_2_indvs(population[idx_parent1]['packages'], population[idx_parent2]['packages'])\n",
    "        \n",
    "        idx_child_counter += 1\n",
    "    \n",
    "    return children\n",
    "       \n",
    "children = crossover_knapsacks(population, packages)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:58.428830Z",
     "start_time": "2020-11-11T07:22:57.936703Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Examining the Decision Space (packages) of one pair of parents their resulting cross over children.\n",
    "\n",
    "idx_parent1 = 4\n",
    "assert idx_parent1 % 2 == 0\n",
    "knapsacks_plot = {'parent 1': population[idx_parent1], 'parent 2': population[idx_parent1 + 1],\n",
    "                  'child 1': children[min(list(children.keys())) + idx_parent1],\n",
    "                  'child 2': children[min(list(children.keys())) + idx_parent1 + 1]\n",
    "                 }\n",
    "\n",
    "colors = {'parent 1': 'purple', 'parent 2': 'green', 'both': 'orange'}\n",
    "\n",
    "# parents_packages = list(set(list(population[0]['packages']) + list(population[1]['packages'])))\n",
    "# min_weight = np.min([packages[package]['weight'] for package in parents_packages])\n",
    "# max_weight = np.max([packages[package]['weight'] for package in parents_packages])\n",
    "# min_value = np.min([packages[package]['value'] for package in parents_packages])\n",
    "# max_value = np.max([packages[package]['value'] for package in parents_packages])\n",
    "\n",
    "min_weight = np.min([packages[package]['weight'] for package in packages])\n",
    "max_weight = np.max([packages[package]['weight'] for package in packages])\n",
    "min_value = np.min([packages[package]['value'] for package in packages])\n",
    "max_value = np.max([packages[package]['value'] for package in packages])\n",
    "\n",
    "ncols = 2\n",
    "nrows = 2\n",
    "\n",
    "plt.figure(figsize = (5 * ncols, 5 * nrows))\n",
    "\n",
    "for ipanel, knapsack_idx in enumerate(knapsacks_plot):\n",
    "    plt.subplot(nrows, ncols, ipanel + 1)\n",
    "    \n",
    "    idx_packages_in_knapsack = knapsacks_plot[knapsack_idx]['packages']\n",
    "    \n",
    "    plt.scatter([packages[idx_package]['weight'] for idx_package in idx_packages_in_knapsack],\n",
    "                [packages[idx_package]['value'] for idx_package in idx_packages_in_knapsack],\n",
    "                color = 'purple', alpha=0.04\n",
    "               )\n",
    "    \n",
    "    #counts_ = {'both': 0, 'parent 1': 0, 'parent 2': 0}\n",
    "    for idx_package in idx_packages_in_knapsack: \n",
    "        in_parent1 = idx_package in knapsacks_plot['parent 1']['packages']\n",
    "        in_parent2 = idx_package in knapsacks_plot['parent 2']['packages']\n",
    "        \n",
    "        if in_parent1 & in_parent2:\n",
    "            color = colors['both']\n",
    "            #counts_['both'] += 1\n",
    "        elif in_parent1:\n",
    "            color = colors['parent 1']\n",
    "            #counts_['parent 1'] += 1\n",
    "        elif in_parent2:\n",
    "            color = colors['parent 2']\n",
    "            #counts_['parent 2'] += 1\n",
    "        else:\n",
    "            print('problem!')\n",
    "            break\n",
    "        \n",
    "        plt.annotate(f'{idx_package:02.0f}', xy=(packages[idx_package]['weight'], packages[idx_package]['value']), alpha=1, color=color)\n",
    "        \n",
    "    #print(knapsack_idx, counts_)\n",
    "    \n",
    "    plt.title(knapsack_idx)\n",
    "    \n",
    "    plt.xlim(min_weight * 0.95, max_weight * 1.05)\n",
    "    plt.ylim(min_value * 0.95, max_value * 1.05)\n",
    "    plt.xlabel('weight [kg]')\n",
    "    plt.ylabel('value [£]')\n",
    "    \n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The color coding is as follows: \n",
    "* Purple packages are those that originate from only parent 1\n",
    "* Green packages are those that originate from only parent 2\n",
    "* Orange packages are those that originate from both parent 1 and 2\n",
    "\n",
    "We see that both children's packages are crossed over from both parents."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Mutation\n",
    "\n",
    "The following uses DNA snippets to illustrate how the mutation tranformation may be applied to one parent solutions to generate one child.\n",
    "\n",
    "![mutation](https://user-images.githubusercontent.com/6064016/97843727-b54eed00-1ce1-11eb-943f-c725b7fc49ed.png)\n",
    "\n",
    "\n",
    "We will use the function `mutate_knapsack` that takes in an individual knapsack and replaces `n_mutate` packages to created a mutated child knapsack."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:58.830029Z",
     "start_time": "2020-11-11T07:22:58.826010Z"
    }
   },
   "outputs": [],
   "source": [
    "def mutate_knapsack(indv_package_idxs, all_package_idxs, n_mutate=1, seed=None):\n",
    "    remaining_package_idxs = list(set(all_package_idxs) - set(indv_package_idxs))\n",
    "    \n",
    "    np.random.seed(seed)\n",
    "    loc_mutate = np.random.choice(range(len(indv_package_idxs)), n_mutate, replace=False)\n",
    "    new_package_idxs = np.random.choice(remaining_package_idxs, n_mutate, replace=False)\n",
    "    \n",
    "    indv_package_idxs_ = indv_package_idxs.copy()\n",
    "\n",
    "    for i, loc in enumerate(loc_mutate):\n",
    "        indv_package_idxs_[loc] = new_package_idxs[i]\n",
    "    \n",
    "    return indv_package_idxs_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:59.132350Z",
     "start_time": "2020-11-11T07:22:59.127960Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "individual original: 14, 12, 4, 1, 5, 9, 19, 7, 15, 13\n",
      "individual mutated : 14, 11, 4, 1, 2, 9, 19, 7, 15, 13\n",
      "8 packages remain and 2 were changed\n"
     ]
    }
   ],
   "source": [
    "all_packages = list(packages.keys())\n",
    "n_packages_mutate = 2    # CHANGE ME!\n",
    "seed = 2\n",
    "\n",
    "idx_indv = 0\n",
    "individual = population[idx_indv]['packages']\n",
    "\n",
    "print(f'individual original: {\", \".join(map(str,individual))}')\n",
    "individual_mutated = mutate_knapsack(individual, all_packages, n_mutate=n_packages_mutate, seed=seed)\n",
    "print(f'individual mutated : {\", \".join(map(str,individual_mutated))}')\n",
    "print(f'{len(set(individual) & set(individual_mutated))} packages remain and {len(set(individual) - set(individual_mutated))} were changed')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Fitness Evaluation\n",
    "\n",
    "Fitness means the values of the objectives (weight, value). Here we use `knapsack_objective`, the same algorithm as in the previous notebook.\n",
    "\n",
    "Here we will apply the algorithm to three of the children generated above (two from crossover, one from mutation) and examine how competitive they are against the original population."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:22:59.732175Z",
     "start_time": "2020-11-11T07:22:59.727120Z"
    },
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'child 1': {'packages': [14, 17, 11, 1, 4, 7, 9, 12, 15, 19]},\n",
       " 'child 2': {'packages': [0, 5, 13, 1, 4, 7, 9, 12, 15, 19]},\n",
       " 'individual_mutated': {'packages': array([14, 11,  4,  1,  2,  9, 19,  7, 15, 13])}}"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# defining the new generation\n",
    "\n",
    "population_born = {'child 1': {'packages':child1}, 'child 2': {'packages':child2}, \n",
    "                   'individual_mutated': {'packages': individual_mutated}\n",
    "                  }\n",
    "population_born"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:23:00.034996Z",
     "start_time": "2020-11-11T07:23:00.029543Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'child 1': {'packages': [14, 17, 11, 1, 4, 7, 9, 12, 15, 19],\n",
       "  'weight': 92.8669222089192,\n",
       "  'value': 770.2954735883773},\n",
       " 'child 2': {'packages': [0, 5, 13, 1, 4, 7, 9, 12, 15, 19],\n",
       "  'weight': 101.83251049309172,\n",
       "  'value': 773.4176186474931},\n",
       " 'individual_mutated': {'packages': array([14, 11,  4,  1,  2,  9, 19,  7, 15, 13]),\n",
       "  'weight': 103.12995451169887,\n",
       "  'value': 558.442956775247}}"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# evaluating the fitness of the new generation\n",
    "\n",
    "for _, indv in population_born.items():\n",
    "    for objective in objectives:\n",
    "        knapsack_objective(indv, objective, packages)\n",
    "population_born    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:23:00.567955Z",
     "start_time": "2020-11-11T07:23:00.379313Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'value [£]')"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# visualising the population and the new population\n",
    "\n",
    "plt.scatter([knapsacks['weight'] for _, knapsacks in population.items()],\n",
    "                [knapsacks['value'] for _, knapsacks in population.items()],\n",
    "                s=100, alpha=0.4, color='purple', label='original population')\n",
    "\n",
    "plt.scatter([knapsacks['weight'] for _, knapsacks in population_born.items()],\n",
    "                [knapsacks['value'] for _, knapsacks in population_born.items()],\n",
    "                marker='s',s=200, alpha=0.4, color='orange', label='cross overs and mutations')\n",
    " \n",
    "\n",
    "plt.legend()\n",
    "plt.xlabel('weight [kg]')\n",
    "plt.ylabel('value [£]')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The visual above shows the distribution of the parent population (circles) and some selected children (squares).   \n",
    "\n",
    "Results might differ by configuration, but generally speaking:  \n",
    "There is not reason to be surprised if the children solutions/knapsacks are not Pareto optimal. That is the stochastic nature of this technique! At times we will get children that beat their parents, but not always (as an exterme case - we shouldn't always expect the children of Albert Einstein or Michael Jordan to perform better than their parents).\n",
    "\n",
    "<img src=\"https://i.ytimg.com/vi/-YZNnUDrpbE/maxresdefault.jpg\" width=500>\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Selection\n",
    "\n",
    "This is where the Pareto front technique gets used!  \n",
    "\n",
    "Imagine that we have a population competing to continue to the next generation. We need to choose a subset of the \"most fit\".   \n",
    "\n",
    "We could just take the Pareto optimal solutions.   \n",
    "That might not be adequate in exterme cases where we have a small set (e.g, of one), because this will result in a lack of diversity.  \n",
    "\n",
    "For this reason we need to choose in advance the number individuals to progress to the next iteration. We'll call this `n_selection`.  \n",
    "\n",
    " \n",
    "There are various algorithms that use the concept of Pareto optimality to choose the most diverse `n_selection` solutions.  \n",
    "\n",
    "Here I describe a fairly simplistic one and will provide material on more advanced techniques.  \n",
    "\n",
    "Let's look at is pseudocode, first.  \n",
    "\n",
    "**Pseudocode**\n",
    "\n",
    "* input\n",
    "    * `population` - list or dictionary of solutions from which we want to select from\n",
    "    * `n_selection` - determines the number of individuals to select for the next generation\n",
    "    * `exact_n_selection` - boolean determining if we need exactly `n_selection` or to choose the all solutions from the final pareto front\n",
    "* output\n",
    "    * `pareto_fronts` - a list where each item is a list of indexes of the individuals that comprise of the pareto front. \n",
    "\n",
    " \n",
    "```python\n",
    " \n",
    "population_remaining = population.copy()\n",
    "individuals_selected = []\n",
    "pareto_fronts = []\n",
    "while len(individuals_selected) < n_selection:\n",
    "    these_selected = find_pareto_optimal(population_remaining)\n",
    "    \n",
    "    # --- fro exact_n_selection=True we update these_selected ---\n",
    "    if exact_n_selection:\n",
    "         n_iterim_selected = len(these_selected) + len(individuals_selected)\n",
    "        if n_iterim_selected > n_selection:\n",
    "            n_subset = n_selection - len(individuals_selected)\n",
    "            these_selected = choose_random(these_selected, n_subset)\n",
    "    # ------\n",
    "             \n",
    "    pareto_fronts.append(these_selected)\n",
    "    \n",
    "    individuals_selected.append(these_selected)  # updating the selected\n",
    "    population_remaining -= these_selected # updating the remaining population\n",
    "    \n",
    "```\n",
    "\n",
    "Example:  \n",
    "E.g, for `n_selection=7` could return `pareto_fronts=[[17, 49, 83], [5, 58, 25, 49]]`. This means that individuals  `[17, 49, 83]` comprise of the first Pareto front, but because it has less than 7 items, we also select an additional 4 from the next Pareto front (after excluding the first) resulting in `[5, 58, 25, 49]`. \n",
    "\n",
    "\n",
    "For the purpose of this demo we'll call this algorithm a Naive Pareto Selector.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:23:01.388416Z",
     "start_time": "2020-11-11T07:23:01.380723Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[[1, 15, 19], [4], [6, 13, 14], [5, 18], [0, 2]]"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# naive pareto selector applied on the knapsacks\n",
    "\n",
    "N_SELECTION = 10           # CHANGE ME!  the number of individuals to be selected (examine in visual below)\n",
    "exact_n_selection = False  # CHANGE ME!  determines the number of individuals selected for the final pareto front (visual below)\n",
    "\n",
    "idxs_knapsacks_selected = []\n",
    "\n",
    "population_remaining = population.copy()\n",
    "\n",
    "individuals_selected = []\n",
    "pareto_fronts = []\n",
    "while len(individuals_selected) < N_SELECTION:\n",
    "    these_selected = knapsacks_to_pareto_front(population_remaining)\n",
    "    \n",
    "    n_these_selected_and_selected = len(these_selected) + len(individuals_selected)\n",
    "    if exact_n_selection:\n",
    "        if n_these_selected_and_selected > N_SELECTION:\n",
    "            n_these_subset = N_SELECTION - len(individuals_selected)\n",
    "            these_selected = np.random.choice(these_selected, n_these_subset, replace=False).tolist()\n",
    "\n",
    "    pareto_fronts.append(these_selected)\n",
    "    \n",
    "    individuals_selected = [individual for pareto_front in pareto_fronts for individual in pareto_front]\n",
    "    \n",
    "    for individual in these_selected:\n",
    "        del population_remaining[individual]\n",
    "        \n",
    "pareto_fronts    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:23:01.922358Z",
     "start_time": "2020-11-11T07:23:01.696633Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# visualising the outcome of the naive pareto selector\n",
    "\n",
    "max_size = 1000\n",
    "for ipareto, pareto_front in enumerate(pareto_fronts):\n",
    "    size = max_size * (1. - ipareto/6)\n",
    "    plt.scatter([population[idx_indv]['weight'] for idx_indv in  pareto_front],\n",
    "                [population[idx_indv]['value'] for idx_indv in  pareto_front],\n",
    "                marker='*',s=size, alpha=0.4, label=f'selected Pareto {ipareto} ({len(pareto_front)})',)\n",
    "\n",
    "plt.scatter([knapsacks['weight'] for _, knapsacks in population.items() if _ not in individuals_selected],\n",
    "                [knapsacks['value'] for _, knapsacks in population.items() if _ not in individuals_selected],\n",
    "                 s=100, alpha=0.4, color='purple', label='not selected')\n",
    "    \n",
    "plt.legend()\n",
    "plt.xlabel('weight [kg]')\n",
    "plt.ylabel('value [£]')\n",
    "pass"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this visual we can see the how the Naive Pareto Selector works.  \n",
    "Given the optimisation objectives it first selects all the Pareto optimal solutions, marked as blue large stars.  \n",
    "If this number is less than `n_selection` it goes for the next Pareto front marked as slighly smaller orange stars. This continues until `n_selection` is reached.   \n",
    "\n",
    "For those interested in more advance selection function algorithms, we suggest looking into state of the art algorithms: \n",
    "* `SPEA2` - Strength Pareto Evolutionary Algorithm \n",
    "* `NSGA-II` -  Non-dominated Sorting Genetic Algorithm\n",
    "* `IBEA` - Indicator-Based Evolutionary Algorithm  \n",
    "\n",
    "They account for further details that are beyond the scope of this notebook:  \n",
    "* Elitism - a fraction of the selected individuals progress to the next generation, bypassing the mutation and crossover processes\n",
    "* Density estimation - optimised to reduce the density of selected solution in the Objective Space \n",
    "* Preserving edge results - preserves edge solutions in the Objective space\n",
    "* Pareto Front heuristics for comparisons\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This concludes the components of the Genetic Algorithm.  \n",
    "Let's put them together to see learning in action!  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Knapsack Optimisation\n",
    "\n",
    "Now we'll put all the components together and examine the evolution of the Pareto front estimates of the Knapsack problem!\n",
    "\n",
    "<img src=\"https://upload.wikimedia.org/wikipedia/commons/f/fd/Knapsack.svg\" width=250>\n",
    "\n",
    "The next two cells comprise of scripts that generate the initial population and apply selection.  \n",
    "\n",
    "We will compare results of two different selectors: \n",
    "* `naive_pareto_selector` - as practiced in [Selection](#Selection) . Remember that this uses the advantages of Pareto optimisation: gives a broader scope for exploring the landscape.  \n",
    "* `single_objective_selector` - Single Objective Optimisation discusses in the previous notebook. This is the more common approach of combining multiple paremters into one single heuristic. \n",
    "\n",
    "Whereas the Single Objective Optimisation is likely to be faster and demand less cognitive effort from the user, it is more likely to result in suboptimal solutions copared to the Pareto optimal selector. \n",
    "\n",
    "Still skeptic? You should be.  \n",
    "Let's find out by applying to the knapsack problem!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Just beforehand, some more maintenance.  \n",
    "\n",
    "We need to determine the probablity of each parent to transform to children. Some will crossover, some will mutate:  \n",
    "* `PROB_CROSS_OVER` - probability of parent to create a child by crossover\n",
    "* `PROB_MUTATION` - probability of parent to create a child by mutation  \n",
    "\n",
    "* `PARENTS_COMPETE` is a boolean that determiness if parents compete with the children for the next generation. This is where the analogy to humans breaks down ... unless you are a fan of [The Highlander](https://en.wikipedia.org/wiki/Highlander_(film)). Geek paraphrasing:  \n",
    "\n",
    "> \"There can be ~only~ [$\\ge$] one!\").  \n",
    "\n",
    "\n",
    "<img src=\"https://i.ytimg.com/vi/Oc9fUbjJbAU/maxresdefault.jpg\" width=700>\n",
    "\n",
    "\n",
    "If  `PARENTS_COMPETE=False` the parents do not compete, but they can still create a children as an exact replicates with probability   \n",
    "`replicate_prob= 1 - PROB_CROSS_OVER - PROB_MUTATION`  \n",
    "\n",
    "Friendly reminder:  In the first stage there is no need to understand all the lines of Python below. It's enough to have a general idea what each function does."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:23:03.599962Z",
     "start_time": "2020-11-11T07:23:03.597023Z"
    }
   },
   "outputs": [],
   "source": [
    "# -- selection --\n",
    "PARENTS_COMPETE = True   # CHANGE ME! True: parents compete with children to be selected for the next generation. False: only children compete\n",
    "\n",
    "# -- mating --\n",
    "PROB_CROSS_OVER = 0.6    # CHANGE ME! probability that any parent may craete a child by mating with another parent\n",
    "\n",
    "# -- mutation --\n",
    "PROB_MUTATION = 0.4      # CHANGE ME! probability that any parent may create a child by mutating itself\n",
    "N_PACKAGES_MUTATE = 2    # CHANGE ME! number of packages used for the process of mutation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:23:03.917162Z",
     "start_time": "2020-11-11T07:23:03.910549Z"
    }
   },
   "outputs": [],
   "source": [
    "# selection function algorithms\n",
    "\n",
    "# same algorithm as above\n",
    "def naive_pareto_selector(population, n_select=20, exact_n_selection = False):\n",
    "    population_remaining = population.copy()\n",
    "    \n",
    "    individuals_selected = []\n",
    "    pareto_fronts = []\n",
    "    \n",
    "    while len(individuals_selected) < n_select:\n",
    "        these_selected = knapsacks_to_pareto_front(population_remaining)\n",
    "\n",
    "        n_these_selected_and_selected = len(these_selected) + len(individuals_selected)\n",
    "        if exact_n_selection:\n",
    "            if n_these_selected_and_selected > n_select:\n",
    "                n_these_subset = n_select - len(individuals_selected)\n",
    "                these_selected = np.random.choice(these_selected, n_these_subset, replace=False).tolist()\n",
    "\n",
    "        pareto_fronts.append(these_selected)\n",
    "\n",
    "        individuals_selected = [individual for pareto_front in pareto_fronts for individual in pareto_front]\n",
    "\n",
    "        for individual in these_selected:\n",
    "            del population_remaining[individual]\n",
    "            \n",
    "    return pareto_fronts,  individuals_selected\n",
    "\n",
    "\n",
    "# calculating the single objective heuristic for each knapsack (soh) \n",
    "# same function as in the previous notebook\n",
    "def knapsack_to_soh(knapsack):\n",
    "    # generalising \n",
    "    # (weight,value) : returns single-objective-heuristic\n",
    "    # (↓, ↑) : returns knapsack['value'] / knapsack['weight'] \n",
    "    # (↑, ↑) : returns knapsack['value'] * knapsack['weight'] \n",
    "    # etc.\n",
    "    \n",
    "    value_heuristic    = direction_to_multiplier[objective_mode['value']](1, knapsack['value'])\n",
    "    weight_heuristic   = direction_to_multiplier[objective_mode['weight']](1, knapsack['weight'])\n",
    "    return  value_heuristic * weight_heuristic\n",
    "\n",
    "# selection function for Single Objective Optimisation: chooses the top `n_select` individuals\n",
    "def single_objective_selector(population, n_select=20, objective='value_weight_ratio'):    \n",
    "    population_sorted = {k: v for k, v in sorted(population.items(), key=lambda indv: indv[1][objective], reverse=True)} \n",
    "    \n",
    "    return {k: population_sorted[k] for k in list(population_sorted.keys())[:n_select]} "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:23:04.245645Z",
     "start_time": "2020-11-11T07:23:04.235793Z"
    }
   },
   "outputs": [],
   "source": [
    "# algorithms for creation of children\n",
    "\n",
    "# creates the children\n",
    "def create_children(idx_indv_start, idxs_parents, population_parents, n_packages_mutate=None):\n",
    "    idxs_chidren = {}\n",
    "    \n",
    "    if n_packages_mutate is None:\n",
    "        n_packages_mutate = N_PACKAGES_MUTATE\n",
    "    \n",
    "    # children replicated\n",
    "    idxs_chidren['replicated'] = list(np.arange(idx_indv_start, idx_indv_start + len(idxs_parents['replicate'])))\n",
    "    population_children = {idxs_chidren['replicated'][counter]: population_parents[idx_indv] for counter, idx_indv in enumerate(idxs_parents['replicate'])}\n",
    "\n",
    "    if idxs_chidren['replicated']:\n",
    "        idx_indv_max = np.max(idxs_chidren['replicated'])\n",
    "    else:\n",
    "        idx_indv_max = idx_indv_start\n",
    "        \n",
    "    # children mutated\n",
    "    idxs_chidren['mutated'] = np.arange(idx_indv_max + 1, idx_indv_max + 1 + len(idxs_parents['mutate']))\n",
    "\n",
    "    for (idx_parent, idx_child) in zip(idxs_parents['mutate'], idxs_chidren['mutated']):\n",
    "        population_children[idx_child] = {}\n",
    "\n",
    "        seed_ = seed + idx_child\n",
    "\n",
    "        parent = population_parents[idx_parent]['packages']\n",
    "        population_children[idx_child]['packages'] = mutate_knapsack(parent, all_packages, n_mutate=n_packages_mutate, seed=seed_)\n",
    "\n",
    "    idx_indv_max = np.max(idxs_chidren['mutated']) \n",
    "    \n",
    "    # children cross-overs\n",
    "    population_parents_mate = {idx_parent: population_parents[idx_parent] for idx_parent in population_parents if idx_parent in idxs_parents['mate']}\n",
    "    population_children_crossovers = crossover_knapsacks(population_parents_mate, packages, idx_start=idx_indv_max)\n",
    "\n",
    "    idxs_chidren['crossed_over'] = [] # not sure if required. delete?\n",
    "    for idx_child in population_children_crossovers:\n",
    "        assert not idx_child in population_children\n",
    "        population_children[idx_child] = {}\n",
    "        population_children[idx_child] = population_children_crossovers[idx_child]\n",
    "        \n",
    "        idxs_chidren['crossed_over'].append(idx_child)\n",
    "        \n",
    "    return population_children, idxs_chidren\n",
    "    \n",
    "\n",
    "# determines the parents indices for replication, mutation, mating\n",
    "def population_tranformations_idxs(idxs_parents, cx_prob, mutate_prob, parents_compete=True, seed=None):\n",
    "    np.random.seed(seed)\n",
    "    \n",
    "    if parents_compete:\n",
    "        # no need for replicates\n",
    "        replicate_prob = 0\n",
    "        mutate_prob = 1. - cx_prob  # we are concerned with only mutating a mating\n",
    "    else:\n",
    "        replicate_prob = 1. - cx_prob - mutate_prob\n",
    "    \n",
    "    transformation_idxs_parents = {}\n",
    "    \n",
    "    transformation_idxs_parents['replicate'] = set(np.random.choice(list(idxs_parents),  int(len(idxs_parents) * replicate_prob), replace=False))\n",
    "    transformation_idxs_parents['mutate'] = set(np.random.choice(list(idxs_parents),  int(len(idxs_parents) * PROB_MUTATION), replace=False))\n",
    "    #print(type(idxs_parents), type(transformation_idxs_parents['replicate']), type(transformation_idxs_parents['mutate']))\n",
    "    transformation_idxs_parents['mate'] = idxs_parents - transformation_idxs_parents['replicate'] - transformation_idxs_parents['mutate']\n",
    "\n",
    "    # randomising (keeping list in order to keep random order. perhaps best done in the cross-over function ...)\n",
    "    transformation_idxs_parents['mate'] = list(np.random.choice(list(transformation_idxs_parents['mate']), len(transformation_idxs_parents['mate']), replace=False))\n",
    "    \n",
    "    return transformation_idxs_parents"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:23:04.568515Z",
     "start_time": "2020-11-11T07:23:04.551246Z"
    }
   },
   "outputs": [],
   "source": [
    "# the main part of the algorithm\n",
    "\n",
    "# logging the individuals\n",
    "def _update_log(gen_num, this_gen, all_gens):\n",
    "    for idx_indv in this_gen:\n",
    "        if 'generations' not in all_gens[idx_indv]:\n",
    "            all_gens[idx_indv]['generations'] = [] \n",
    "        all_gens[idx_indv]['generations'].append(gen_num)\n",
    "        \n",
    "\n",
    "# visualising pareto fronts\n",
    "def plot_pareto_front(population, objective_to_sort='weight', label=None, marker='-o', color=None, linewidth=None, alpha=0.7, markersize=None):\n",
    "\n",
    "    pareto_front_idxs = knapsacks_to_pareto_front(population)\n",
    "\n",
    "\n",
    "    population_to_sort = {idx_indv: population[idx_indv] for idx_indv in population if idx_indv in pareto_front_idxs}\n",
    "    population_sorted = {k: v for k, v in sorted(population_to_sort.items(), key=lambda indv: indv[1][objective_to_sort], reverse=False)}\n",
    "    plt.plot([indv['weight'] for _, indv in population_sorted.items()],\n",
    "             [indv['value'] for _, indv in population_sorted.items()], marker,\n",
    "                     alpha=alpha, color=color, label=label, linewidth=linewidth, markersize=markersize)\n",
    "    \n",
    "\n",
    "# the genetic algorithm\n",
    "def knapsack_genetic_algorithm(n_generations, n_initial_population=None, unique_indvs=False, n_packages_mutate=None,  \n",
    "                               cx_prob=None, mutate_prob=None, seed=None, n_select=None, exact_n_selection=False, \n",
    "                               parents_compete=None, selector='pareto_naive', visualise=True, display_pareto_front=True):\n",
    "    if not cx_prob:\n",
    "        cx_prob = PROB_CROSS_OVER\n",
    "    if not mutate_prob:\n",
    "        mutate_prob = PROB_MUTATION\n",
    "    if not n_packages_mutate:\n",
    "        n_packages_mutate = N_PACKAGES_MUTATE\n",
    "    if not n_initial_population: \n",
    "        n_initial_population = N_INITIAL_POPULATION\n",
    "    if parents_compete is None:\n",
    "        parents_compete = PARENTS_COMPETE\n",
    "    if n_select is None:\n",
    "        n_select = n_initial_population\n",
    "        \n",
    "    # initial population\n",
    "    population_parents = create_knapsacks(n_initial_population, seed=seed)\n",
    "    \n",
    "    if unique_indvs:\n",
    "        population_parents = {frozenset(population_parents[key]['packages']): population_parents[key] for key in population_parents}\n",
    "    \n",
    "    # evaluating fitness of the initial population\n",
    "    if 'single_objective_optimisation' == selector:\n",
    "        for idx_indv, indv in population_parents.items():\n",
    "            indv['value_weight_ratio'] = knapsack_to_soh(indv)\n",
    "    \n",
    "    # creating log of all individuals\n",
    "    population_all_generations = {idx_indv: {'packages': population_parents[idx_indv]['packages']} for idx_indv in population_parents}\n",
    "    \n",
    "    for gen_num in range(n_generations):\n",
    "        # keeping track of highest valued individual (useful when unique_indvs=False)\n",
    "        idx_indv_max = len(population_all_generations) - 1\n",
    "        \n",
    "        # --- population tranformations ---\n",
    "        # dictionary containing the the transofrmations (replicate, mutate, mate) and the idxs of parents that will go through each transformation\n",
    "        idxs_parents_tranformations = population_tranformations_idxs(set(population_parents.keys()), cx_prob, mutate_prob, parents_compete=parents_compete, seed=seed)\n",
    "        \n",
    "        # generating the children popultation basedon the tranformation of the parents \n",
    "        population_children, idxs_chidren = create_children(idx_indv_max, idxs_parents_tranformations, population_parents, n_packages_mutate=n_packages_mutate)\n",
    "        if unique_indvs:\n",
    "            population_children = {frozenset(population_children[key]['packages']): population_children[key] for key in population_children}\n",
    "        \n",
    "        for idx_indv, indv in population_children.items():\n",
    "            if idx_indv not in population_all_generations:\n",
    "                population_all_generations[idx_indv] = {} \n",
    "                population_all_generations[idx_indv]['packages'] = population_children[idx_indv]['packages']\n",
    "        \n",
    "        # --- evaluating the children fitness ---\n",
    "        for idx_indv, indv in population_children.items():\n",
    "            for objective in objectives:\n",
    "                if objective not in indv:\n",
    "                    knapsack_objective(indv, objective, packages)\n",
    "                    \n",
    "            if 'single_objective_optimisation' == selector:\n",
    "                indv['value_weight_ratio'] = knapsack_to_soh(indv)\n",
    "        \n",
    "                    \n",
    "        # --- selection ---\n",
    "        # population for selection (children only or parents+children)\n",
    "        if parents_compete:\n",
    "            # pooling parents and children together to compete for selection\n",
    "            population_competing = {**population_parents, **population_children} # was called population_parents_and_children\n",
    "            _update_log(gen_num, population_competing, population_all_generations)\n",
    "        else:\n",
    "            # only children compete for selection\n",
    "            population_competing = population_children.copy()\n",
    "            _update_log(gen_num, {**population_parents, **population_children}, population_all_generations)\n",
    "\n",
    "        # the selection function: Naive Pareto Selector or Single Objective Optimisation \n",
    "        if 'single_objective_optimisation' == selector:\n",
    "            individuals_selected = single_objective_selector(population_competing, n_select=n_select)\n",
    "            pareto_fronts = None\n",
    "        else:\n",
    "            pareto_fronts, individuals_selected = naive_pareto_selector(population_competing, \n",
    "                              n_select=len(population_parents), exact_n_selection=exact_n_selection)\n",
    "\n",
    "        \n",
    "        # --- plotting ---\n",
    "        if visualise:\n",
    "            alpha = 0.4\n",
    "\n",
    "            pf_marker = '-'\n",
    "            color = 'gray'\n",
    "            linewidth = 2\n",
    "            if gen_num in [0, n_generations - 1]: \n",
    "                if 0 == gen_num:\n",
    "                    marker = 's'\n",
    "                    size = 40\n",
    "                    color = 'purple'\n",
    "                    linewidth = 5\n",
    "                else:\n",
    "                    marker = 'x'\n",
    "                    size = 100\n",
    "                    color = 'green'\n",
    "                    linewidth = 5\n",
    "\n",
    "                label = f'competitors gen {gen_num + 1}'\n",
    "                plt.scatter([knapsacks['weight'] for _, knapsacks in population_competing.items()],\n",
    "                                [knapsacks['value'] for _, knapsacks in population_competing.items()],\n",
    "                                 s=size, alpha=0.7, color=color, label=label, marker=marker, linewidth=4)\n",
    "                label= f'pareto front gen {gen_num + 1}'\n",
    "            else:\n",
    "                label = None\n",
    "                \n",
    "            if display_pareto_front:\n",
    "                plot_pareto_front(population_competing, objective_to_sort='weight', label=label, marker=pf_marker, color=color, linewidth=linewidth, alpha=alpha)\n",
    "                \n",
    "\n",
    "        \n",
    "        \n",
    "        # --- next generation ---\n",
    "        population_parents = {idx_indv: population_competing[idx_indv] for idx_indv in population_competing.keys() if idx_indv in individuals_selected}\n",
    "        \n",
    "    plt.legend()\n",
    "    plt.xlabel('weight [kg]')\n",
    "    plt.ylabel('value [£]')   \n",
    "    \n",
    "    return population_all_generations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:23:05.442233Z",
     "start_time": "2020-11-11T07:23:05.150388Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# running the knapsack problem with a genetic algorithm\n",
    "\n",
    "n_generations = 20           # CHANGE ME! Number of iterations\n",
    "n_initial_population = 20    # CHANGE ME!\n",
    "parents_compete = True       # CHANGE ME! True: parents and children compete for selection, False: only children compete\n",
    "unique_indvs = False         # CHANGE ME! True: no knapsack duplication, False: may have knapsack duplication\n",
    "n_select = 20                # CHANGE ME! number of competing individuals selected for the next generation\n",
    "exact_n_selection = True    # CHANGE ME! True: selector limited to n_select, False: selector might go above n_select (relevant for selector='pareto_naive')\n",
    "\n",
    "display_pareto_front = True  # Visualises the Pareto front \n",
    "seed = 1                     # CHANGE ME! For different distributions and runs\n",
    "\n",
    "selector = 'pareto_naive'    # Leave this selection function method. Below we compare with 'single_objective_optimisation'\n",
    "population_all_pareto = knapsack_genetic_algorithm(n_generations, seed=seed, selector=selector, \n",
    "                                                   unique_indvs=unique_indvs, \n",
    "                                                   n_initial_population=n_initial_population,\n",
    "                                                   n_select=n_select, exact_n_selection=exact_n_selection,\n",
    "                                                   parents_compete=parents_compete,\n",
    "                                                   display_pareto_front=display_pareto_front\n",
    "                                                  )\n",
    "\n",
    "plt.title('Pareto Optimisation')\n",
    "ax = plt.gca()\n",
    "x_lim = ax.get_xlim()\n",
    "y_lim = ax.get_ylim()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The above visual shows:  \n",
    "* purple squares - all competing individuals in the first generation\n",
    "* purple solid line - Pareto front of the first generation  \n",
    "\n",
    "If  `n_generations>1` you will also see  \n",
    "* greeen Xs - all competing individuals in the last generation\n",
    "* green solid line - Pareto front of the last generation  \n",
    "* gray thin lines (for `n_generations>2`) - Pareto fronts of all intermediate generations  \n",
    "\n",
    "Tracking the Pareto front lines you should get a clear picture of how the genetic algorithm found better solutions.  \n",
    "\n",
    "By increasing `n_generations` you will see improvement, until a certain point in which there is no reall cost benefit.  This does not mean that we have found the true Pareto front of this problem. More likely it got stuck in one (or more) local minima/maxima. \n",
    "\n",
    "Solving for this is out of the scope of this demo and requires further exploration. We discus this in [Summary](#Summary).\n",
    "\n",
    "In the meanwhile let's compare results with the same algorithm that uses a different selection function: Single Objective Optimisation.  \n",
    "\n",
    "How do you think that they will compare?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:23:08.669025Z",
     "start_time": "2020-11-11T07:23:08.417005Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Single Objective Optimisation: value/weight')"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "selector = 'single_objective_optimisation'\n",
    "population_all_soh = knapsack_genetic_algorithm(n_generations, seed=seed, selector=selector, \n",
    "                                                   unique_indvs=unique_indvs, \n",
    "                                                   n_initial_population=n_initial_population,\n",
    "                                                   n_select=n_select, exact_n_selection=exact_n_selection,\n",
    "                                                   parents_compete=parents_compete,\n",
    "                                                   display_pareto_front=display_pareto_front\n",
    "                                                  )\n",
    "\n",
    "plt.xlim(*x_lim)\n",
    "plt.ylim(*y_lim)\n",
    "plt.title(f'Single Objective Optimisation: {heuristic}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This visual is the same as before but now using the Single Objective Optimisation approach for the selection function. \n",
    "\n",
    "Examine the plot to see if you can pick up on any difference.  \n",
    "In the next few cells we will do direct comparisons between them."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:23:09.214021Z",
     "start_time": "2020-11-11T07:23:09.188239Z"
    }
   },
   "outputs": [],
   "source": [
    "# Evaluating objective values for populations (weight, value) as well as single objective heuristic: value/weight \n",
    "\n",
    "for idx_indv, indv in population_all_pareto.items():\n",
    "    for objective in objectives:\n",
    "        if objective not in indv:\n",
    "            knapsack_objective(indv, objective, packages)\n",
    "                    \n",
    "    indv['value_weight_ratio'] = knapsack_to_soh(indv)\n",
    "    \n",
    "for idx_indv, indv in population_all_soh.items():\n",
    "    for objective in objectives:\n",
    "        if objective not in indv:\n",
    "            knapsack_objective(indv, objective, packages)\n",
    "                    \n",
    "    indv['value_weight_ratio'] = knapsack_to_soh(indv)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:23:09.874246Z",
     "start_time": "2020-11-11T07:23:09.512370Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The pareto run generated 420 individuals and the SOH generated 420\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# visualising all individuals of both methods during for all generations\n",
    "\n",
    "\n",
    "def jitter(x, loc=1., scale=0.005):\n",
    "    return x * np.random.normal(loc, scale)\n",
    "\n",
    "alpha = 0.2\n",
    "size = 10\n",
    "jitter_factor = 0.0  # CHANGE ME! e.g, 0.001. Set to a small factor to jitter the results. This is useful when wanting to visualise duplicates\n",
    "\n",
    "\n",
    "print(f'The pareto run generated {len(population_all_pareto):,} individuals and the SOH generated {len(population_all_soh):,}')\n",
    "\n",
    "indvs_pareto_weights = np.array([knapsacks['weight'] for idx_knapsack, knapsacks in population_all_pareto.items()])\n",
    "indvs_pareto_values  = np.array([knapsacks['value']  for idx_knapsack, knapsacks in population_all_pareto.items()])\n",
    "plt.scatter(jitter(indvs_pareto_weights, scale = jitter_factor), jitter(indvs_pareto_values, scale = jitter_factor), \n",
    "            s=size, alpha=alpha, color='green', label='Pareto Optimisation individuals', marker=None, linewidth=4)\n",
    "\n",
    "\n",
    "indvs_soh_weights = np.array([knapsacks['weight'] for idx_knapsack, knapsacks in population_all_soh.items()])\n",
    "indvs_soh_values  = np.array([knapsacks['value']  for idx_knapsack, knapsacks in population_all_soh.items()])\n",
    "plt.scatter(jitter(indvs_soh_weights, scale = jitter_factor), jitter(indvs_soh_values, scale = jitter_factor),\n",
    "            s=size, alpha=alpha, color='red', label='Single Objective Optimisation individuals', marker='s', linewidth=4)\n",
    "\n",
    "\n",
    "plot_pareto_front(population_all_pareto, color='green', label='Pareto Optimisation Pareto Front')\n",
    "plot_pareto_front(population_all_soh, color='red', label='Single Objective Optimisation Pareto Front', marker='-s')\n",
    "\n",
    "indv_ = single_objective_selector(population_all_pareto, n_select=1)\n",
    "key = list(indv_.keys())[0]\n",
    "plt.plot(indv_[key]['weight'], indv_[key]['value'], '*', markersize=20, color='green', alpha=1, label=f'Pareto Optimisation max {heuristic}')\n",
    "\n",
    "indv_ = single_objective_selector(population_all_soh, n_select=1)\n",
    "key = list(indv_.keys())[0]\n",
    "plt.plot(indv_[key]['weight'], indv_[key]['value'], '*', markersize=20, color='red', alpha=0.6, label=f'Single Objective Optimisation max {heuristic}')\n",
    "\n",
    "\n",
    "plt.xlabel('weight [kg]')\n",
    "plt.ylabel('value [£]')\n",
    "plt.title(f'Individuals of all {n_generations} generations')\n",
    "\n",
    "plt.legend(fontsize=10)\n",
    "# plt.xlim(*x_lim)\n",
    "# plt.ylim(*y_lim)\n",
    "pass"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note above that in addition to the Pareto fronts of both runs, we also highlight with a star the optimal single objective heuristic of each run.\n",
    "\n",
    "**Diversity**     \n",
    "Now we address the issue of diversity in the Decision Space.  \n",
    "\n",
    "In the next two cells we will count knapsack duplicates. \n",
    "Note that the `sorted` function ensures that knapsacks like `[1,3,2]` are aggregated with `[1,2,3]` and the other permutations (`[2,3,1]` ...).  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:23:10.381136Z",
     "start_time": "2020-11-11T07:23:10.212113Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Duplication of Individuals in Single Objective (showing top 4)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(2, 5, 7, 9, 10, 12, 15, 16, 17, 18)     38\n",
       "(5, 7, 9, 10, 12, 15, 16, 17, 18, 19)    26\n",
       "(2, 5, 7, 10, 12, 15, 16, 17, 18, 19)    12\n",
       "(1, 5, 7, 9, 10, 12, 15, 16, 17, 18)      8\n",
       "dtype: int64"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# examining by eye the top knapsack duplications for all generations (Single Objective Optimisation)\n",
    "\n",
    "import pandas as pd\n",
    "\n",
    "top = 4\n",
    "\n",
    "print(f'Duplication of Individuals in Single Objective (showing top {top})')\n",
    "pd.Series([tuple(sorted(knapsack['packages'])) for idx_knapsack, knapsack in population_all_soh.items()]).value_counts().head(top)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:23:10.567384Z",
     "start_time": "2020-11-11T07:23:10.558891Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Duplication of Individuals in Pareto Optimisation (showing top 4)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(7, 9, 10, 12, 14, 15, 16, 17, 18, 19)    43\n",
       "(2, 7, 9, 10, 12, 14, 15, 16, 17, 18)     39\n",
       "(0, 7, 9, 10, 12, 14, 15, 16, 17, 18)     18\n",
       "(0, 2, 4, 7, 10, 12, 15, 16, 17, 18)       3\n",
       "dtype: int64"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# examining by eye the top knapsack duplications for all generations (Naive Pareto Selector)\n",
    "\n",
    "\n",
    "print(f'Duplication of Individuals in Pareto Optimisation (showing top {top})')\n",
    "pd.Series([tuple(sorted(knapsack['packages'])) for idx_knapsack, knapsack in population_all_pareto.items()]).value_counts().head(top)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " If `unique_indvs = False` you should see that in both selection function options we end up with duplications, i.e, lack of diversity.  \n",
    "\n",
    "This may be addressed in a few manners. \n",
    "* set `unique_indvs = True` this will ensure that only unique solutions are competing for the next generations. This is done by ensuring that the keys of knapsack dictionaries are the packages sets (using `frozenset`). \n",
    "\n",
    "\n",
    "If you conclude that you still have diversity problems, consider this as a list of options to explore rather than instructions to follow: \n",
    "\n",
    "* use a larger initial population: `n_initial_population`. (This may be considered niching - starting from more solutions).      \n",
    "* select more individuals per round by vayring `n_select`    \n",
    "* ensure no replications by setting `PROB_CROSS_OVER + PROB_MUTATION = 1`  \n",
    "* increase the number of packages mutating `N_PACKAGES_MUTATE` \n",
    "* change the problem by increasing the number of total packages `N_PACKAGES` ([Packages section](#Packages)) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Examining a single generation**\n",
    "\n",
    "Whereas above we examined all the individuals in all generation generated by the two runs,  \n",
    "in the following we examine for a selected generation.  \n",
    "\n",
    "* `n_generation_examine` generation to examin (for final do `-1` or `'final'`)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:23:12.082928Z",
     "start_time": "2020-11-11T07:23:11.843835Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "In this generation (no. 20)\n",
      "the Pareto Optimisation run had 40 individuals and the SOH Optimisation generated 40\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtwAAAH9CAYAAAApsatnAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAACGuklEQVR4nOzdd3iUVfr/8fdJgUASkBYIIE2REhISCIpAKFJVQBdlFcuKrr1+wbK2VdzFXRVUFtQfi66Lrg1FKSq6iIKAgBJYQKqAhiKhCqRAIOX8/nhmhkwyCQlkMimf13XNlcxT72fmIdxz5j7nGGstIiIiIiLiH0GBDkBEREREpCpTwi0iIiIi4kdKuEVERERE/EgJt4iIiIiIHynhFhERERHxIyXcIiIiIiJ+pIRbRM6aMaaVMcYaY6YXWD7dtbzVWRy7r+sY40qxzzjXPn3P9LxnyhiTYoxJKe/z+mKMqWOMmeyKKcf1msQHOq6qyvX6Lgp0HCJS8SjhFqniXEmABtyvnl4A7gN+BP4OPAPsDWhElVhF+jBVUsaYq40xU4wxS4wxaa6/B++cZp9IY8yzxpjNxpgsY8xhY8x/jTH9yytukaomJNABiEiV9hjwHPDrWRzjB6ADcLBMIqpehgI/WWuHBTqQaqIDcCzQQRTwJNAZyAB2A+2L29gYUw9YCnQENgBTgQjgCmCBMeZWa+2//BqxSBWkhFtE/MZamwqknuUxjgGbyyaiaqcpsDjQQVQX1tqKeJ+OwUm0twF9gIWn2X4cTrL9CXCNtTYHwBjzOJAMTDHG/Ndau9tvEYtUQSopEamG8tdcu37/wBhz0PX1cbIxZmgR+0UaY14yxux2bbvZGDOWIv6WFKzhNsZ0dz2fVUxsm4wxJ4wx9V3Pi6zhNsZ0NcZ8aYxJd31dvsAYc/HprrmI9YsKlt4YY2oYY+41xswzxuxwxfWb6zyXFnUNPo5dwxhzvzFmtevr+WOu8oQ5xpgBpThOtDHmVde+J40xB4wxnxhjuhZxLQbo4y4rKml9sTFmsDHmO2NMput6Zxtj2hdXk2+MucgYM9MYs9cV2y5jzD+NMU19bLvIdZwQY8zjxpitrtd2lzHmeWNMjSLicsewy3WOfcaY94wx7Xxs6461jTHmPmPMOmPMcfdrUJr31n0PAi2BlvleT6/7qajX2BhT1xjzd2PMFuNdolHovc9/vxtj4o0xnxtjjrjumW+NMT18vTZFsdYutNZutdaWtKzsd66fT7mTbddx9gMvAbWAW0oTg4iohVukumuJU7LxM/AfoD5wDTDHGDPAWutpDTPG1AS+BroBa4F3gXOAP+O0nJ2WtXaFMWYLcJkxpoG19lD+9caYC3G+8v7YWvtbccdyJR4LgBo4rXHbgHhgEfBNSeIpgfrAP4BlwFfAASAaGAbMM8bcZq19owTHmQ6MAtYDbwPHcVqfewFDXNdRLGNMa5yv+pviXN/7wLnASOByY8xV1trP8p1vEfA0sMP1HCClBOe5FngPyAI+xPmGogewHOd997XPLcA04AQwF9gFtAVuBYYZY7pba3f62PU9IAn4AkgDLgMeAaKAmwucYwjO+xwKfIrzfjcHRriuv5+1drWPc/zDdY7PgXlArmt5ad7bFJz69/9zPZ+U7/hrfL0m+eI+B/gOp9V4pWvfhsDvgfnGmLustf/0sWsizmuxHHgDaAFcBXxtjIm31m4p7rxnoYnr588+1rmX9Qf+4qfzi1RN1lo99NCjCj8A6/xT91rWyr0ceLrAusGu5fMKLH/ctfxjICjf8tbAb6510wvsM921vFW+ZY+5lt3rI9ZXXeuG5VvW17VsXL5lBqfMxAJXFDjGA/mura+Pa55e8Lyu9Yt8vE41geY+tq2Lkzz/BtQqsC4FSCmwbR7O1/HBPo7VoITv439d8T9RYHkPIAc4BET4eO8XleJeiQQO4yTOnQusey7f65r//bwAOImTADcrsE9/nAR3lq/XGlgF1M+3PNx1nFygSb7l9VxxHQQ6FjhWJ5z65NVF3Hu/Aq19XOtZv7dF/FtbVGDZP13L/wmYfMvbAkddr3X+19N9v1tgdIFj3eFa/lpJ39MC+7uP/U4x2+xxbdPRx7r/c63beybn10OP6vxQSYlI9bYDGJ9/gbX2v8BO4MIC296Mkzg+Yq3Ny7f9L8DkUpzzP67j3JR/oauM4FpgP06LZ3F6AO2AxdbaOQXWvQJsL0U8RbLWnrA+alWttUeBN3ESwW6nOwzOB4QTONdd8FiHCu1RgDGmOTAI5315ocD+y3Bau+vjtPaejStwvrV411pbsDV7PHDExz534bQ6P2Ct9eoca639GqfFe5gxJtLHvn+y+b7JsNZm4nxzEoTTwuv2B1dcT1trNxY4x3rgdSDBGNPRxzlecN2jXsrovS2W656+AecDwWPWWk9Zh7V2K86/mxo411fQd9ba6QWWvYnz4argv82y9Lnr5zPGmGD3QmNMI5x6cHBeGxEpBZWUiFRva6y1uT6W7wI8tdCuZOl8YJe11lcyuwinfOG0rLW7jTFfAwONMR3zJVDDcJLGl22+2tEidHH9/NbH8XONMUuB80oSz+kYY2KAh4HeOCUHYQU2aVbc/tbaNGPMpzjXt8YY8zGwBPjeOh1CSyLB9XOJtTbbx/pvcBK7BJySlTPlPs/SgiustRnGmDU4raT5ue+TPsYYXwlqFBCM0xK+qsC6ZB/b73L9zJ/Uuc/R2fgej/0C188OwMYC637wsT1w9u9tCbQDauMkz75KpL7BGUUkwce6Qq+NtTbbGLMP/ya8T+F8y3U1zv36Nc43D1fgfFvQAh8fHEWkeEq4Raq3I0Usz8G7I2Rd1899RWxf2rGdpwMDcVq5/+Ra5m7xfqsE+5d1PD4ZY7rjJEUhOPXrc3FqjfNw6sWvwClNOJ1rcK7zOpxaYIAsY8xM4CFrbVHX4ea+3qJGfHEvP6cEsZTkPEXF42t5A9fPh09z7IiCC6y1R3xs5/6wFZxvmfsct5X2HBRxL5The1ucs3nfjhSxTw7er02Zstamuj44/RlnWMm7cUp5ZuDUvG/F+RZKREpBCbeIlMRR18/GRaxvUsTyoszCSW5uMM5wYw2AS4G1PkoZyioed6tcUX/3zvGx7EmcURn6WWsX5V9hjHkMJyk7LWvtcZzh1sYZY87FaVEdjdMq3QqnU19x3Ndb1OscXWC7M5Xm+lnU6+prufucda21aT7WlwX3OTpba9eVct+iRucok/f2NMrrfStTrg+A97oeHsaYS1y/riz3oEQqOdVwi8hpWWvTcXWKM8b4KtXoW8rjHccZAaMpMACn5TeEkrVuA7hHoyg0Ooqr7rSXj30Ou36e62OfOpwqS8jvfOC3gglZUecuCWvtLmvtuzhf228DehljGpxmt/+5fvYyxvj6wNDP9dPXKB2l4TlPwRXGmAiclt+CVrh+nu5Dw9nwxznO5L3NpXSty1twJsLp7BqtpKCyet/Ki7vW/L2ARiFSCSnhFpGS+jfO34znjTGevx2u4eruP4PjTXf9/IPrkYPTYa4kluEkM72NMQVbIu/FR/2260PDZqBn/s51rgTdPb5wQSlAfWNMXP6Fxpg/4iTMp2WMaWSMifWxKhynBCIHZ5SPIrk6932F0xr+fwWOfxHOB5bDON8cnI05OK2t1xtjOhdY9yS+vwV4BcgGXjbGFPrQ4hrv+mwT5X/jlFg87Ro6suA5gowxfUt5zBRK/94eAhoZY3zdK4VYa0/i3NORwF8LnOc8nH832TgdiSsE12tZqDTHGHMjzr/TZcDs8o5LpLJTSYmIlNSLwJU4YwGvNsb8FycB+z3ObIbDS3Mwa+13xphtOONIhwKfWmdyjZLsa12J0VfAx8aY/ONw9we+xBnfuqAJwL+A74wxH+GMNd3Pdf61OFNg5zcJJ/laaoz5ECcZTcRpAZ6J07HsdJoB/zPG/Aisw+kUWAenPrYJMNn1YeB07sQZz3mCMWYQTqc69zjcecDNJTxOkVwdPO/BSQCXua7ZPQ53Z5xOqn3I12nOWrvZNQ73m8AGY8yXwE84r2kLnFbpA5xmSvHTxHXIGHM1zgeKFa6OfBtwykXOxelU2YDCnR6LM4nSv7fucei/NMYsxhl5Zq219tNizvMozmtwr6s2eiGnxuGOxBkes9AoKmXFGHMlzr9bOFXacnG+CXsOWmsfyrdLbWCfMeYrnNF+8oCeOK/xJmBk/lGKRKRklHCLSIlYa0+4ZsYbh9MJ8AGcVsLxOIlQqRJul7c41fJX0nISdzzfuVpOn8Wp/wb4Hqe8ZTA+Em5r7ZvGGAOMxemkeRinVfdxnPHFC27/pTFmGE7r7jU4JQU/4CTpbShZwp2CM4JLX9d+DXHGeN6Ck4x9UIJjYK392RiT6IrlMtfx0nA+XDxrrS2Tulpr7bvGmN9wOs1dg5NULsZJuCa6NksrsM87xpi1wIM41zgIyMQZ03kmToe7s43ra1dr9EM4728SzjcDe3A6PxZ6/05zvDN5b8fjfMgchpOEBuPct0Um3Nba34wz++ljOMM2jsWZ+OgHYIK1dn5p4j4D8RQYghPn+tq4ft+B85q6ncC5J3vhdGwGp6PkE8CkUoysIyL5mHzDgoqIiPjkKr35GahhrY0+3fYiInKKarhFRMTDGHOOMaZ2gWUGpyW4BWdfJy4iUu2ohVtERDyMMUNwSkDm45TDRADdcUoTdgGJJa21FxERhxJuERHxcI06Mx6nRrkRTl+f3cBnwN9KMEmPiIgUoIRbRERERMSPVMMtIiIiIuJHVX5YwIYNG9pWrVoFOgwRERERqeJWrVp10FrbqODyKp9wt2rViuTk5ECHISIiIiJVnDFmh6/lKikREREREfEjJdwiIiIiIn6khFtERERExI+UcIuIiIiI+JESbhERERERP1LCLSIiIiLiR0q4RURERET8SAm3iIiIiIgfKeGWcpVxMoPff/R7Mk5mBDoUERERkXKhhFvK1dc/f81HGz/im1++CXQoIiIiIuVCCbeUq1mbZzk/N80KcCQiIiIi5UMJt5Qbay2f/fQZAJ/+9CnW2gBHJCIiIuJ/Sril3Gw8sJGsnCwAjuccZ9PBTQGOSERERMT/lHBLuZm3dR45eTkA5Nk85m2dF+CIRERERPxPCbeUmw83fMiJ3BMAZOVk8eGGDwMckYiIiIj/hQQ6AKk6rppxFZ9s/qTE26/csxLzjCly/Yj2I/j4mo/LIjQRERGRgFELt5SZ5wY8R3yTeMJDw8/qOOGh4SQ0SeC5Ac+VUWQiIiIigaOEW8pM2wZtSb4tmWf6PkOtkFoEmdLdXkEmiFohtfhLv7+QfHsybRu09VOkIiIiIuVHCbeUqeCgYB7s8SBr71xLXOO4Erd21w6tTefGnVl751rGXjy21Mm6iIiISEVV7lmNMaaDMeYbY8xRY8w2Y8zvXMtbGWOsMSYj3+PP+faraYx50xiTZozZa4wZW96xS8m5W7sf6/UYYSFhxW4bFhLG470eV6u2iIiIVEnl2mnSGBMCzAGmAgOBPsCnxpgE4KRrs3OstTk+dh8HtAVaAk2AhcaYjdbaL/0euJyR4KBgOkV1okZwDc/4277UCK5BbONYtWqLiIhIlVTeGU57oCnwsrU211r7DfAdcGMJ9r0J+Ku19rC1dhPwOjDab5FKmZi1eRbpJ9KL3Sb9RLqmehcREZEqqyI0KRqgU77nO4wxu40x/zbGNAQwxtQDooG1+bZbC8SUX5hSWu6p3C2npnAPMkHUDK7p1ZptsZrqXURERKqs8k64twD7gYeNMaHGmEE4ZSW1gYNAN5ySka5AJPCua78I18+j+Y511LVNIcaY240xycaY5AMHDpT9VUiJbDywkeM5xz3Pa4c4HSM/HfUpnRt39upQqaneRUREpKoq14TbWpsNXAlcDuwFHgQ+BHZbazOstcnW2hxr7T7gXmCQMSYSyHAdok6+w9UBfNYqWGunWWsTrbWJjRo18tPVyOnM2zqP3LxcAEKDQvnrJX8l+fZkBp43kJW3rfQaPjA3L1dTvYuIiEiVVO4lJdbaddbaPtbaBtbawUAb4Adfm7p+BllrDwOpQOd86zsDG/wbrZyNDzd8SHZuNnVq1uHz6z73Gu6v4PCB2XnZmupdREREqqRyn9rdGBMH/IST7N+NU5s93RhzEXAE2ArUAyYDi6y17jKSt4EnjTHJQGPgNuDm8o1eSqNxRGO6NetGUoskBp430Oc27uEDJ62YxKKUReUboIiIiEg5CESnyRtxWqv3A/2BgdbaEzgt3V/ilImsB04Ao/Lt9zSwHdgBfAtM0JCAFduTvZ+kUXgjftfhd8Vu527t/vS6T8spMhEREZHyU+4t3Nbah4GHfSx/H3i/mP1OALe4HlIJzNk8h6jaUVzU7KJAhyIiIiISMBVhWECpgn4+/DPrD6xn6AVDCQ4KDnQ4IiIiIgGjhFv8Ys7mOYSFhDHovEGBDkVEREQkoJRwS5k7fPwwi3cuZmCbgYTXCD/9DiIiIiJVmBJuKXPu8beHXTAs0KGIiIiIBJwSbilTJ3NPMm/bPC5sdiHRkdGBDkdEREQk4JRwS5n6NuVb0k6kMbzd8ECHIiIiIlIhKOGWMmOtZc6WObQ+pzWxUbGBDkdERESkQlDCLWVm3b517Di6gyvaXYExJtDhiIiIiFQISrilzMzZMoe6NeuS1DIp0KGIiIiIVBhKuKVM/Jr2Kyv3rOTytpdTI7hGoMMRERERqTCUcEuZmLtlLiFBIVza9tJAhyIiIiJSoSjhlrOWcTKDr3/5mr4t+3JO2DmBDkdERESkQlHCLWdt/vb5nMg9oaEARURERHxQwi1nJTcvl09/+pS4qDha12sd6HBEREREKhwl3HJWlu9ezsFjB7mi/RWBDkVERESkQlLCLWdl9ubZREdE061pt0CHIiIiIlIhKeGWM7bl4Ba2HNrC8HbDNdGNiIiISBGUcMsZm7tlLuGh4QxoMyDQoYiIiIhUWEq45YwcPHaQpbuWMui8QYSFhAU6HBEREZEKSwm3nJHPf/ocay1DLxga6FBEREREKjQl3FJqWTlZfLn9S3qc24Oo8KhAhyMiIiJSoSnhllL75pdvyDiZwRXtNBSgiIiIyOko4ZZSsdYyd8tc2tZvS/uG7QMdjoiIiEiFp4RbSmVV6ip+Tf+VK9pdoaEARUREREpACbeUypzNc6hfqz49W/QMdCgiIiIilYISbimxHUd2sGbfGoa2HUpIUEigwxERERGpFJRwS4nN3TKXGsE1GHL+kECHIiIiIlJpKOGWEjmadZSFKQu5pNUlRNaMDHQ4IiIiIpWGEm4pkS+3fUl2XjbD2w0PdCgiIiIilYoSbjmtnLwcPt/6OV2ju3Ju3XMDHY6IiEill5uXG+gQpByp55uc1pIdSzicdVit2yIiImdpf+Z+UtNTycnLISQohOjIaM3aXA0o4ZZiWWuZvXk259Y5l4QmCYEOR0REpNLan7mfXUd3AZBn88jJy/E8V9JdtamkRIq18cBGfj7yM8PbDddENyIiImchNT0VgN3pu9lycAu703d7LZeqSwm3FGvOljlE1oikX6t+gQ5FRESk0srNyyUnL4c8m0d6VjrGGNKz0j0t3arprtqUcEuR9mbsZcXuFVx6/qXUDKkZ6HBEREQqreCgYEKCQggyQUSGRWKtJTIskiATREhQCMFBwYEOUfxINdxSpM9++owgE8RlbS8LdCgiIiKVXnRkNLuO7qJ5ZHPyIvIIMkGe5VK1KeEWn45lH2P+9vn0atGLBrUbBDocERGRSs/dMVKjlFQ/SrjFpwU/L+B4znGuaHdFoEMRERGpMqLCo4gKjyI3L1dlJNWIarilkDybx9wtc+nQsANtG7QNdDgiIiJVjpLt6kUJtxTyw68/sC9zH1e2vzLQoYiIiIhUekq4pZA5m+cQVTuKi5pdFOhQRERERCo9JdziZftv21l/YD1DLxiqr7tEREREyoASbvEyd8tcwkLCGHTeoECHIiIiIlIlKOEWj9+O/8binYsZ2GYg4TXCAx2OiIiISJWghFs8vtj6Bbl5uQy7YFigQxERERGpMpRwCwAnc08yb9s8Lmx2oWa8EhERESlDSrgFgG9TviXtRBrD2w0PdCgiIiIiVYoSbsFay5wtc2h9Tmtio2IDHY6IiIhIlaKEW1i7by07ju7ginZXYIwJdDgiIiIiVYoSbmHulrnUrVmXpJZJgQ5FREREpMpRwl3N/Zr2Kyv3rOTytpdTI7hGoMMRERERqXJCAh2ABNbcLXMJCQrh0raXBjoUERGRiiMtDXbuhBMnvJfXrAktWkCdOoGJSyolJdzVWMbJDL7+5Wv6tuzLOWHnBDocERGRimPnTrjvPkhNLbyuWTP4+uvyj0kqLSXc1dh/t/2XE7knNBSgiIhIQSdOOMl2rVreyyMj4ddfAxOTVFpKuKupnLwcPtv6GXFRcbSu1zrQ4YiIiATeiROwd6/zWLIEjhxxHjk5UK8eREQEOkKppJRwV1PLdy3n4LGD3JV4V6BDERERKT+ZmU7L9d69sGeP9++//XZqu6NH4dgxCAtz6raDgwMXs1R6SrirqTlb5hAdEU23pt0CHYqIiEjZsdZJllNTvR/upDo93Xv7+vWhSRNISIDo6FOP1FS4/fbCJSUiZ0AJdzW05eAWthzawh1d79BENyIiUvlYCwcPFk6q3Y+srFPbGgONGkHTptCzp/OzSRMnqW7SxGnB9sU9OklkZOF1+r9TSkkJdzU0Z8scwkPDGdBmQKBDERER8S0nB/btO9UyvXevd2t1Ts6pbUNCoHFjJ5mOjXUSaXdi3bixs760WrRwRiMp2EHSGDj//LO7Nql2yj3hNsZ0AF4FugIHgIettbNc6/q71rUAvgdGW2t3uNbVBP4fcDVwDHjBWvtSecdf2R08dpDvdn3HFe2uICykiE/1IiIi5SEr61QnxYKt1Pv3Oy3ZbmFhTqt0ixZw0UXeSXXDhhBUxnP51amjof+kzJRrwm2MCQHmAFOBgUAf4FNjTALwG/AJcCvwKfBXYAbQ3bX7OKAt0BJoAiw0xmy01n5ZntdQ2X3202dYaxl6wdBAhyIiItVBRkbhOmp3gp2/kyI45RvR0dC+PfTr511TXbeuSjmk0irvFu72QFPgZWutBb4xxnwH3AjsAjZYaz8CMMaMAw4aY9pbazcDN+G0eB8GDhtjXgdGA0q4SygrJ4svt31Jj3N7EBUeFehwRESkKrDWGTrPVy313r2+Oyk2bQpdungn1NHREB4ekEsQ8beKUMNtgE5AHWCte6G1NtMYsx2IMcbsA6Lzr3f9fmU5xlnpffPLN2RmZ3JFuysCHYqIiFQmeXlFd1Lcu7dwJ8WoKCeB7tXLO6Fu0sQZYk+kminvhHsLsB942BjzMtAPp6xkIRCBU9Od31Eg0rXO/bzgukKMMbcDtwO0aNGirGKv1Ky1zN0yl7b129K+YftAhyMiIhVNdrZTN+0rqd63z7uTYmiokzw3aQKdO58a9SM62km2z6STokgVVq7/Iqy12caYK4EpwJ+AZOBD4ASQgdPKnV8dIN21zv08q8A6X+eZBkwDSExMtL62qW5Wpa7i1/RfeejihzQUoIhIdeXupFhw1I/UVDhwwLuTYq1aTgLdqhVcfPGpFuqmTaFBA9VTi5RCuX8Etdauw2nVBsAYswx4C7A4ddru5eHAeTh13YeNMalAZ+Ar1yadgQ3lFXdlN2fzHOrXqk/PFj0DHYqIiPhTenrhDorun4cPe29bp46TSHfseCqZdrdU16mjpFqkjARiWMA44CcgCLgbpzZ7Ok6L9QRjzFXA58BTwDpXh0mAt4EnjTHJQGPgNuDm8o2+ctpxZAdr9q3hD3F/ICRIX/OJiFRq1jqJc1HTk2dmem/foIGTQCcmFq6nVidFkXIRiOzrRpyh/0KBJcBAa+0J4IAr2X4FeAdnHO5r8+33NM443DuA48DzGhKwZOZumUuN4BoMOX9IoEMREZGSyM313UnRXQbingURnPGnGzd2Eug+fQon1TVqBO46RAQITEnJw8DDRaxbgDN0oK91J4BbXA8poaNZR1mYspD+rfsTWdNnH1MREQmE7GynM6KvTor79/vupNi0qdNJMf/05I0aqZOiSAWnf6FV3JfbviQ7L5vh7YYHOhQRkeonK+tUEl2wo+LBg96dFGvXdpLoNm2gRw/vpFqdFEUqNSXcVVh2bjafb/2crtFdObfuuYEOR0Sk6rHWmUnR16gfqanOhDD51a3rJNCdOnmP+tGkiTopilRhSrirsKU7l3I467Bat0VEzoa1zhTkRU1PXrCTYsOGTjJ94YWF66lr1w7MNYhIQCnhrqKstczePJtz65xLQpOEQIcjIlKx5eY641AXNZPiyZOntg0OdiZ3adoU2rXzTqobN1YnRREpRAl3FbXhwAZ+PvIz93S7RxPdiIiAkzQX7KTobq3ev99Jut1q1DhVP92li3dS3aiRk3SLiJSQEu4qau6WuUTWiKRfq36BDkVEpPwcO+a7ljo1FQ4d8u6kGB7uJNDnnQdJSd7Tk9evr3pqESkzSriroL0Ze1mxewUjO46kZkjNQIcjIlJ2rIW0tKKnJz961Hv7c85xEui4OO9a6qZNISJCSbWIlAsl3FXQZz99RpAJ4rK2lwU6FBGR0rPWaY0uanryY8dObWvMqU6K3bt7T0/epAnUqhW46xARcVHCXcUcyz7G/O3z6dWiFw1qNwh0OCIivuXknOqkmH968tRUp866YCfFxo2dJLpjR+966qgodVIUkQpPCXcV89X2rziec5wr2l0R6FBEpLo7edJ3PfXevU5SnZd3atsaNZyW6WbNIDHRO6lu2FCdFEWkUlPCXYXk2Tw+/elTOjTsQNsGbQMdjohUB5mZxXdSzC883Emq27Z1Oinmn0mxXj3VU4tIlaWEuwr5fvf37Mvcxy0JtwQ6FBGpKtydFIuanjwtzXv7evWcJDo+3nsWxehoiIwMyCWIiASaEu4qZO6WuUTVjuKiZhcFOhQRqUzcnRSLmp78+PFT2xrjjEMdHQ09ehSenjwsLHDXISJSQSnhriK2/7ad9QfWc0v8LQQHqdZRRArIyXEmd/E1PfnevZCdfWrbkJBTnRRjYgp3UgwNDdx1iIhUQkq4q4g5W+YQFhLGoPMGBToUEQmUEyeKrqc+cMC7k2LNmk4C3bw5dOtWuJNiUFDgrkNE5Czk5uVWuMZHJdxVwG/Hf2PJziVcev6lhNcID3Q4IuJPmZm+R/3Yswd++81728hIp8yjXTvo29c7qT7nHHVSFJEqZX/mflLTU8nJyyEkKIToyGiiwqMCHRaghLtKmLd1Hrl5uQy7YFigQxGRs2WtM1uir1bq1FRIT/fevn59J4FOSPDuoBgd7cykKCJSDezP3M+uo7sAZ9S2nLwcz/OKkHQr4a7kTuae5IttX3BhswuJjowOdDgiUhJ5eXDwYNHTk2dlndrWGKduOjoaevXynp5cnRRFRABITU8FYHf6btKz0okMi6R5ZHNS01OVcMvZW5SyiLQTaQxvNzzQoYhIfjk5zuQuBZPqPXuc5Tk5p7YNCTnVMh0b6z09eVSUs15ERHzKzcslJy+HPJtHelY6xhjSs9LJi3BauitCTbf+ildi1lrmbJ5D63NaExsVG+hwRKqfrKxTiXT+xHrPHqeTorWntg0LcxLoli2he3fveuoGDdRJUUTkDAUHBRMSFEJOXg6RYZGeFu4gE0RIUEjAk21Qwl2prd23lp1pO/m/i/4Po85PIv6RkeG7lnrvXt+dFJs2hQ4d4JJLvJPqunXVSVFExE+iI6PZdXQXzSObkxeRR5AJ8iyvCJRwV2JzNs+hbs26JLVMCnQoIpWXtXD4cNHD6WVkeG9fv76TVHfp4iTS+TsqhmuUIBGRQHDXaWuUEilTv6b9SnJqMtd1uo4awTUCHY5IxebupOhrevK9e707KQYFOXXTTZpA797esyg2aeKMXy0iIhVOVHgUUeFRFaJmuyAl3JXU3C1zCQkK4dK2lwY6FJGKITvbmUnR16gfBTsphoaeapXu3Nl7evJGjdRJUUSkEqtoyTYo4a6U0k+k8/UvX9O3ZV/OCTsn0OGIlJ+sLN8TvuzdW7iTYq1aTiLdqhVcfHHhToqqpxYRkXKihLsSmr99PidyT2goQKma0tOLnvTlyBHvbevUcRLojh29E+roaGedkmoREakAlHBXMjl5OXy29TPiouJoXa91oMMRKT13J8WipifPzPTevmFDp9yjWzfvhLpJE3VSFBGRSkEJdyWzfNdyDh47yF2JdwU6FJGi5eZ6d1IsmFifOHFq26AgaNzYSaL79DnVQbFpU2d5DXUKFhGRyk0JdyUzZ8scoiOi6da0W6BDkeru5EnfMym6Oynm5p7atkYNJ3lu2hTi472H0lMnRRERqeL0v1wlsuXgFrYc2sIdXe/QRDdSPo4fL9w50T2s3qFD3p0Ua9d2Eug2baBnT+/SD3VSFBGRakwJdyUy7YdphIeGM6DNgECHIlWFtd6dFAtOT370qPf2des6SXRsbOFOipGRSqpFRER8UMJdCZhnvJOYGRtmYJ+2RWwtUoC1zhTkRdVT5++kaIzTGt20KVx0UeFOirVrB+46REREKikl3BVcwWQ7/3Il3eKRm+uMQ11UUn3y5Kltg4OdmRSbNoX27b2nJ1cnRRERkTKnhFuksnB3UvSVVO/fX7iTortlumtX7+nJGzVykm4REREpF0q4K7CNuzeedn3H5h3LKRopF8eOFa6jdndULNhJMTzcSajPPx+SkrynJ69XT/XUIiIiFYQS7grsdMm0ku1KyFpISyt6evKCnRTPOcdJpOPiCndSjIhQUi0iIlIJKOEWKWvWOq3RRU1Pfvz4qW2NcWZSjI6G7t0Ld1KsVStw1yEiIiJlQgl3BWeftj47TqrDZIDl5BTfSTE7+9S2ISFOZ8QmTSAm5tSEL9HRzvLQ0MBdh4iIiPidEu5KwJ1cq2a7nJ086T17YsFOinl5p7atWdNJoJs3h8RE7+nJGzZ0pi8XERGRakkJdyWiZNsPMjN9z6To7qSYX3i4k0BfcAH07u09Pbk6KYqIiEgRlHBL1ebupOhr1I89e5xZFvOrV89JoOPjC9dTR0YG5BJERESkclPCLZWftXDwYNHTk2dlndrWGGcc6uho6NmzcFIdFha46xAREZEqSQm3VA45OU7dtK966n37fHdSjI52OinmT6obN3bWi4iIiJQTZR5ScZw4UXwnxfyTvoSFOS3SLVrAhRd6J9XqpCgifpCbl0twkGZpFZHSU8It5Sszs3DnRPfjt9+8t42MdBLodu2gXz/v6cnPOUedFEWkXOzP3E9qeio5eTmEBIUQHRlNVHhUoMMSkUpECbc/nDgBb74Jt9ziDBdXnVgLR44UPT15wU6K9es7SXWXLt611O6ZFEVEAmh/5n52Hd0FQJ7NIycvx/NcSbeIlJQSbn/YvBk+/hh69IDOnQMdTdnLy/PupFiwo2LBTopRUU4C3atX4U6K1e0DiYhUKqnpqQDsTt9NelY6kWGRNI9sTmp6qhJuESkxJdz+kJzslEckJ1fehDsnx+mM6Cup3rfPWe8WEnKqVTouzjupjopSJ0URqZRy83LJycshz+aRnpWOMYb0rHTyIpyWbtV0i0hJKRMqa9bCt99C69aweLFTVlJRa42zsorupHjgQOFOik2bQsuWcPHF3tOTN2igTooiUuUEBwUTEhRCTl4OkWGRnhbuIBNESFCIkm0RKTEl3GXNPUNhs2bw669OQhsdHbh40tOLTqoPH/betk4dJ9aOHb07KDZt6qyrqB8cRET8JDoyml1Hd9E8sjl5EXkEmSDPchGRklLCXdZ+/NGpcTbGaSH+8Uf/JtzWOolzUdOTZ2R4b9+ggRNPYqJ3Uh0d7UxdLiIiHu46bY1SIiJnQwl3WTnvPPj558LLv/gC2rSB7dvP/NjuToq+Rv1ITXVGRXELCjrVSbF378KTvqiToohIqUSFRxEVHqWabRE5Y0q4S+uVV2DePCcJzs9Xsp1/3ZAhhZcHBcFll8G99zozJebvpJg/sS7YSTE09FSrdOfO3kl1o0bqpCgi4gdKtkXkTCkzK63f/95JhFetclqMa9Uq2X4tW0JuLhw7dqoEpHFj2LbN6Vh58KB3J8VatZwEulUrp5Ni/qS6QQPVU4uIiIhUEkq4SysqCsaPhwUL4LXXnElemjQ5/X7ffOOUfpw44STLrVo5U5CHhECnTt6jfkRHq5OiiIiISBWhhPtMBAXBoEEQHw+TJjmt3adzzjlw8qRTAnLvvU6SXbu2nwMVERERkUBTwn023K3dX3wB8+cXv229evDAA07NtlquRURERKoNzVZytoKCoHnz029Xo4aznZJtERERkWql3BNuY0wrY8w8Y8xhY8xeY8wrxpgQ1zprjMk0xmS4Hm/k288YY543xhxyPZ43poJkr8nJp+88aa2znYiUr7Q0WL/eKf3K/1i/3lknIiLiZ4EoKXkN2A9EA+cAXwF3A5Nd6ztba7f52O924EqgM2Bd+/0CTPVvuKfhnsr9qqucmmxrnaH8Tp50WrWbNHFatTMzK/5U7yJV0c6d3mPVu5044azr1Kn8YxIRkWolECUlrYEPrbVZ1tq9wJdATAn2uwl40Vq721r7K/AiMNp/YZZQaqozpF+tWnD8OKSkQGwsvPyy83PHDmd57dpw4ICTjItI+XEn25s2eT/yrxMREfGjQCTck4BrjTG1jTHNgEtxkm63xa5Sk0+MMa3yLY8B1uZ7vpaSJer+9eOPp1q109LgoYecjpQdOsBf/woPPugs37v31FTvIiIiIlJtBCLhXoyTKKcBu4FkYLZrXR+gFdAe2AN85q7vBiKAo/mOcxSI8FXHbYy53RiTbIxJPnDggD+u4ZQlS5yxuGNjYdo0Z7hAd0ju4QOnTXO+tj5yxNleRKSay83LDXQIIiLlplxruI0xQTit2dOAHjhJ9JvA88Aj1trFrk1PGmMewEnKOwA/AhlAnXyHqwNkWJt/ekaHtXaa6xwkJiYWWl+mGjSAZ56BgQOLrs3OP1nOhg1+DUdEpCLbn7mf1PRUcvJyCAkKIToymqjwqECHJSLiV+XdabI+0AJ4xVp7AjhhjPk3MB54xMf2FnBnsRtwOkz+4Hre2bUssMaMKdl27tbuQYP8G4+IeKtZ06nV7tDB9zopN/sz97Pr6C4A8mweOXk5nudKukWkKivXkhJr7UGckUXuMsaEGGPOwekMuc4YE2OMiTfGBBtjInA6Rf4KuHo38TYw1hjTzBjTFHgQmF6e8YtIJdSihe/EumZNZ52Um9T0VAB2p+9my8Et7E7f7bVcRKSqCsSwgCNwOk7+CcgFvgHG4NR1/z+gOZAJLAOGWmuzXfv9E2iDU14C8IZrmYhI0erU0dB/FUBuXi45eTnk2TzSs9IxxpCelU5ehNPSnZuXS3BQcKDDFBHxi3JPuK21a4C+PlbtA9oVs5/FKTvxVXoiIlKhVfeEMjgomJCgEHLycogMiyQ9K53IsEiCTBAhQSHV+rURkaovEC3cIiLVhjoJnhIdGc2uo7toHtmcvIg8gkyQZ7mISFWmhFtExE/USdCb+5r1AUREqhsl3CIifpK/k6C7hKJ5ZHNS01OrbZIZFR5FVHhUtS+xEZHqJRAT34iIVHlFdhK0pzoJVmdKtkWkOlHCLSLiB+5OgkEmiMiwSKy16iQoIlJNqaRERMRP1ElQRERACbeIiN+ok6CIiIASbhERv1InQRERUQ23iEg5ULItIlJ9KeEWEREREfEjJdwiIiIiIn6khFtERERExI+UcIuIiIiI+JESbhERERERP1LCLSIiIiLiR0q4RURERET8qMiJb4wxt5zhMWdba387w31FREQqFE1aJCJnq7iZJt8ALGBKcTwLrAGUcIuISKW2P3M/qemp5OTlEBIUQnRkNFHhUYEOS0QqodNN7T4CJ4Eu6bG2nlU0IiIiFcD+zP3sOroLgDybR05ejue5km4RKa3iEu5cYI+1dkdJDmSMCXLtY8siMBERkUBJTU8FYHf6btKz0okMi6R5ZHNS01OVcItIqRWZcFtrQ0tzIGttHlCqfURERCqa3LxccvJyyLN5pGelY4whPSudvAinpVs13SJSWhqlREREJJ/goGBCgkIIMkFEhkVirSUyLJIgE0RIUIiSbREptdPVcJ+WMSYUiLbW7iyDeERERAIuOjKaXUd30TyyOXkReQSZIM9yEZHSKrKF2xjznjHm3RIcowvwS9mFJCIiElhR4VGcW/dcT0t3SFAI59Y9V/XbInJGimvh7gM8kH+Bq2PkNmCYtXaDPwMTEREJpKjwKKLCo1SzLSJnrbga7jrAkQLLDNAKqOmneERERCoUJdsicraKS7i3ADcaY3xNfKOh/0RERERESqC4kpLxwCfAZcaYn4ATnJp1cpoxJt31ex0/xiciIiIiUqkV2cJtrZ0N9MVJuo/hjLEdgtO6HeJ67l4mIiIiIiI+FJssW2sXA4vdz40xIcBJ4I/W2tWuZd2B7/wZpIiIiIhIZVWqiW+stTlAP5z6brf9wNtlGZSIiIiISFVR6nIQa+23BZ7/DNxcZhGJiIiIiFQhmtpdRERERMSPiptp8mdjTOeSHsgYE+TaJ6ZsQhMRERERqfyKa+FuBdQoxbE0KY6IiIiISAGnq+Fe4XvemyJpQhwRERERkXyKS7ifOcNj7jnD/UREREREqpwiE25r7Zkm3CIiIiIi4qJZIkVEpPJKS4OdO+HECe/lNWtCixZQp05g4hIRyUfDAoqISOXlK9kGZ9nOneUfj4iID2rhFhGRysudbG/a5L28QwffibiISACohVtERERExI+UcIuIiIiI+FGpE25jTIQxpqUxJtQfAYmIiIiIVCUlruE2xgwF/gJ0xpng5kJgtTHmDeAba+17/glRRESkCDVrOrXaHTr4XiciUgGUqIXbGHMlMAc4CPypwH6/ADeVeWQiIiKn06KF78TaPSygiEgFUNIW7qeBf1trbzXGhAAv5Fu3Hri7zCMTERE5nTp1oFOnQEchIlKsktZwdwBmuH63BdYdBhqUWUQiIiIiIlVISRPuNKBhEetaAQfKJBoRERERkSqmpCUlXwGPGWO+ANJdy6wxpiZwL/CFP4ITkWpEU3SLiEgVVdIW7ieAJsAW4A2cspJHgTVAc2CcH2ITkepEU3SLiEgVVaIWbmttijGmC/AMMBjIBXoDXwJPWWv3+C9EEakWNEW3iIhUUSUeh9tauxv4ox9jERERERGpcjS1u4iIiIiIH5WohdsY8+ZpNrHWWrV+i4iIiIgUUNKSkksoPP52fSASOOJ6iIicOU3RLSIiVVRJO0228rXcGNMbmApcX4YxiUh11KJF8cMCioiIVFIl7jTpi7V2sTHmZWAK0KtsQhKRaklTdIuISBVVFp0mfwYSyuA4IiIiIiJVzlkl3MaYEGA0sLsU+7Qyxswzxhw2xuw1xrziOg7GmHhjzCpjzDHXz/h8+xljzPPGmEOux/PGGHM28YuIiIiI+FtJRyn5xsfiGsAFQAPgzlKc8zVgPxANnIMzbfzdxpipwBxgkmubO4A5xpi21tqTwO3AlUBnnA6cXwG/4NSQi4iIiIhUSCVt4Q4CTIFHOvAJ0N9a+3opztka+NBam2Wt3YszW2UM0BfnA8Aka+0Ja+1k13kuce13E/CitXa3tfZX4EWc1nURERERkQqrpKOU9C3Dc04CrjXGLALqAZcCf8ZJutdZa/MPP7jOtdydlK/Nt26ta5mIiIiISIUViJkmF+Mkymk4td/JwGwgAjhaYNujOGN942P9USDCVx23MeZ2Y0yyMSb5wIEDZRu9iIiIiEgpFNnCbYz5Q2kOZK19+3TbGGOCcFqrpwE9cJLoN4HngVSgToFd6uCUrgBkFFhfB8go0CLujmWa6xwkJiYWWi8iIiIiUl6KKymZXorjWOC0CTfO7JQtgFestSeAE8aYfwPjgbHAg8YYky+JjgNedf2+AafD5A+u551dy0REREREKqziEu7WZX0ya+1BY8wvwF3GmIk4Ldw34dRqLwJygftdI5bc5trNPULK28BYY8w8nAT/QZwJd0REREREKqwiE25r7Q4/nXMETsfJP+Ek2N8AY6y1J40xVwJvAM8Bm4ArXUMCAvwTaAP86Hr+hmuZiIiIiEiFZXyUQFcpiYmJNjk5OdBhiIiIiEgVZ4xZZa1NLLi8RMMCug4wCLgLaAeEFVxvrW1zVhGKiIiIiFRBJRoW0BhzGfAFUBtoD2wGdgLnAnnAt/4KUERERESkMivpONx/xhkt5DLX8yddk+HEAME4ybiIiIiIiBRQ0oS7PfApTmu2xVWKYq39CRiHk5CLiIiIiEgBJU2484Ac1/jYB3DG0nbbA5xX1oGJiIiIiFQFJU24twCtXL8nA/9njIk2xjTCGQ87pexDExERERGp/Eo6Ssm7QAfX708DC4Ddrue5wHVlHJeIiIiISJVQooTbWvtqvt9XGWNigSE4o5YssNZu9FN8IiIiIiKVWonH4c7PWrsbZ6ZHEREREREpRknH4f6fMeb/jDGN/R2QiIiIiEhVUtJOk6nABGCXMeYLY8y1xphCs02KiIiIiIi3EiXc1trLgGbAI0Aj4D1gnzHmTWNMPz/GJyIiIiJSqZW0hRtr7X5r7SRrbSLODJOvApcAC4wxO/wVoIiIiIhIZVbihDs/a+0m4C/AEzgT3zQvy6BERERERKqKUifcxphLjDH/BvYBb+OMx31fWQcmIiIiIlIVlGhYQGNMJ+AGnAlumuPMLPkP4D/W2q1+i05EREREpJIr6Tjc64CjwEfA29bapf4LSURERESk6ihpwv174FNr7Ql/BiMiIiIiUtWUdGr3mf4ORERERESkKjqjUUpERERERKRklHCLiIiIiPiREm4RERERET9Swi0iIiIi4kdKuEVERERE/EgJt4iIiIiIHynhFhERERHxIyXcIiIiIiJ+pIRbRERERMSPlHCLiIiIiPiREm4RERERET9Swi0iIiIi4kdKuEVERERE/EgJt4iIiIiIHynhFhERERHxIyXcIiIiIiJ+pIRbRERERMSPlHCLiIiIiPiREm4RERERET9Swi0iIiIi4kdKuEVERERE/EgJt4iIiIiIHynhFhERERHxIyXcIiIiIiJ+pIRbRERERMSPlHCLiIiIiPiREm4RERERET9Swi0iIiIi4kdKuEVERERE/EgJt4iIiIiIHynhFhERERHxIyXcIiIiIiJ+pIRbRERERMSPlHCLiIiIiPiREm4RERERET9Swi0iIiIi4kdKuEVERERE/EgJt4iIiIiIHynhFhERERHxo5DyPJkxJqPAolrAa9ba+4wxrYBfgMx865+31v7VtW9N4P8BVwPHgBestS/5P2oRERHxkpYGO3fCiRPey2vWhBYtoE6dwMQlUkGVa8JtrY1w/26MiQD2Ah8V2Owca22Oj93HAW2BlkATYKExZqO19ks/hSsiElhKaqSi8nVfgrNs507o1Kn8YxKpwMo14S7gKmA/sKSE298EjLbWHgYOG2NeB0YDSrhFpGpSUiMVlfu+3LTJe3mHDr7vWZFqLpAJ903A29ZaW2D5DmOMBb4CHrbWHjTG1AOigbX5tlsLXFkukYqIBIKSGhGRKiEgnSaNMS2BPsBb+RYfBLrhlIx0BSKBd13r3KUoR/Ntf9S1ja/j326MSTbGJB84cKAsQxcRERERKZVAjVJyI7DUWvuLe4G1NsNam2ytzbHW7gPuBQYZYyIBd2fL/AWLdYB0Xwe31k6z1iZaaxMbNWrkp0sQERERETm9QJWU/AF47jTbuEtNgqy1h40xqUBnnFITXL9v8FN8IiIiUpSaNZ2ypg4dfK8TES/lnnAbY3oAzSgwOokx5iLgCLAVqAdMBhZZa91lJG8DTxpjkoHGwG3AzeUUtohI+VNSIxVVixbFj6AjIl4C0cJ9E/CJtbZgOUgb4G9AFJCG05I9Kt/6p3HG4d4BHMcZo1sjlIhI1aWkRiqqOnU0So5IKZjCg4RULYmJiTY5OTnQYYiIiIgvGm9eqhBjzCprbWLB5ZraXURERALndOPNi1QBgRyHW0RERKo7jTcv1YBauEVERERE/EgJt4iIiIiIHynhFhERERHxI9Vwi4iISOBovHmpBtTCLSIiIoHTooXvxFrjzUsVohZuERERCRxNoiPVgFq4RURERET8SAm3iIiIiIgfKeEWEREREfEjJdwiIiIiIn6khFtERERExI+UcIuIiIiI+JESbhERERERP1LCLSIiIiLiR0q4RURERET8SAm3iIiIiIgfKeEWEREREfEjJdwiIiIiIn6khFtERERExI+UcIuIiIiI+JESbhERERERP1LCLSIiIiLiR0q4RURERET8SAm3iIiIiIgfKeEWEREREfEjJdwiIiIiIn6khFtERERExI+UcIuIiIiI+JESbhERERERP1LCLSIiIiLiRyGBDiAQsrOz2b17N1lZWYEORUQqsbCwMJo3b05oaGigQxERkQqsWibcu3fvJjIyklatWmGMCXQ4IlIJWWs5dOgQu3fvpnXr1oEOR0REKrBqWVKSlZVFgwYNlGyLyBkzxtCgQQN9UyYiIqdVLRNuQMm2iJw1/R0REZGSqLYJd6AFBwcTHx9Pp06dGDlyJMeOHTvrYy5atIhly5aVer/Zs2cTFxdHhw4diI2NZfbs2aU+19SpU3n77bdLdd4ePXqUNlQAJk2a5PV6XXbZZRw5cuSMjpXfuHHjaNasmed9mTt37lkfE+Bvf/tbqfdp1aoVsbGxxMfHEx8ff0bva0HTp09nz549Z30cERERKR0l3CXwbcq3jPxoJInTEhn50Ui+Tfn2rI9Zq1Yt1qxZw/r166lRowZTp04t0X45OTlFrjuThHvt2rU89NBDzJkzh02bNjF37lweeugh1q1bV+x+Bc9155138oc//KFU5z7TJLJgwj1v3jzOOeecMzpWQWPGjGHNmjV89NFH3HLLLeTl5ZVov9zc3CLXnUnCDbBw4ULWrFnDmjVrCn04Ke58RVHCLSIiEhhKuE/j25RveXD+g+zP3E/jiMbsz9zPg/MfLJOk2y0pKYlt27bx6aefctFFF5GQkMCAAQPYt28f4LS83njjjfTs2ZMbb7yRAwcOcNVVV9GtWze6devGd999R0pKClOnTuXll18mPj6eJUuWkJKSwiWXXEJcXBz9+/dn586dhc49ceJEHn/8cU+nr9atW/PYY48xYcIEAPr27csDDzzgafX94YcffJ5r3LhxTJw40bPPmDFjSExMpEOHDqxcuZIRI0bQtm1bnnzySc+5IyIiAEhNTaV3796ecyxZsgSAu+66i8TERGJiYnj66acBmDx5Mnv27KFfv37069cPcFqDDx48CMBLL71Ep06d6NSpE5MmTQIgJSWFDh06cNtttxETE8OgQYM4fvx4se9Jhw4dCAkJ4eDBg1x55ZV07dqVmJgYpk2b5hX/gw8+SOfOnVm+fDnvvPMOF154IfHx8dxxxx3k5uby6KOPcvz4ceLj47n++uuLjLEkCp6vNNc6c+ZMkpOTuf7664mPjz/t9YuIiEjZqZajlOT3+qrX+fnwz0Wun7V5FpknM6kZUtOz7ETOCe747A5+1/53PvdpU68Nt3W9rUTnz8nJ4YsvvmDIkCH06tWLFStWYIzhjTfe4IUXXuDFF18EYOPGjSxdupRatWpx3XXXMWbMGHr16sXOnTsZPHgwmzZt4s477yQiIoKHHnoIgGHDhnHTTTdx00038eabb3L//fcXKhfZsGGDZ3u3xMREXn31Vc/zY8eOsWbNGhYvXswtt9zC+vXrC53r66+/9jpGjRo1SE5O5h//+AdXXHEFq1aton79+px33nmMGTOGBg0aeLZ97733GDx4ME888QS5ubme1utnn32W+vXrk5ubS//+/Vm3bh33338/L730EgsXLqRhw4Ze51y1ahX//ve/+f7777HWctFFF9GnTx/q1avH1q1bef/993n99df5/e9/z8cff8wNN9xQ5Pvy/fffExQURKNGjXjzzTepX78+x48fp1u3blx11VU0aNCAzMxMLrroIl588UU2bdrE888/z3fffUdoaCh333037777Ls899xyvvPIKa9asKTbGhISEQjH069eP4OBgatasyffff+91vjO51ldeeYWJEyeSmJhY5HWLiIhI2av2CffpHM06Su3Q2l7LagTX4GjW0bM6rrvVE5wW7j/+8Y9s2bKFa665htTUVE6ePOk11Njw4cOpVasWAAsWLGDjxo2edWlpaWRkZBQ6x/Lly/nkk08AuPHGG3nkkUfOKNZRo0YB0Lt3b9LS0kpULz18+HAAYmNjiYmJITo6GoA2bdqwa9cur4S7W7du3HLLLWRnZ3PllVd6XpcPP/yQadOmkZOTQ2pqKhs3biQuLq7Icy5dupTf/e53hIeHAzBixAiWLFnC8OHDad26tee4Xbt2JSUlxecxXn75Zd555x0iIyOZMWMGxhgmT57MrFmzANi1axdbt26lQYMGBAcHc9VVVwHOB45Vq1bRrVs3wHl/o6KiShyjr4S74IeK/Ocri2sVERGR8lHtE+7TtURvO7yN/Zn7qVOzjmdZ2ok0osKj+PuAv5/xed013Pndd999jB07luHDh7No0SLGjRvnWedOrADy8vJYsWIFYWFhZ3x+t44dO7Jq1So6d+7sWbZq1SpiYmI8zwuOxFCSkRlq1nS+EQgKCvL87n5esA69d+/eLF68mM8//5zRo0czduxYkpKSmDhxIitXrqRevXqMHj36rIZfyx9DcHBwkSUVY8aM8WrxX7RoEQsWLGD58uXUrl2bvn37euIICwsjODgYcMZkvummm/j738/8njid/OcrTkmvVURERMqHarhP495u95J5MpO0E2nk2TzSTqSReTKTe7vdW+bnOnr0KM2aNQPgrbfeKnK7QYMGMWXKFM9zd+IeGRlJenq6Z3mPHj344IMPAHj33XdJSkoqdKyHHnqIv//9755W0JSUFP72t7/x4IMPeraZMWMG4LSq1q1bl7p16xY619nYsWMHjRs35rbbbuPWW29l9erVpKWlER4eTt26ddm3bx9ffPGFZ/uizp2UlMTs2bM5duwYmZmZzJo1y+c1l8bRo0epV68etWvXZvPmzaxYscLndv3792fmzJns378fgN9++40dO3YAEBoaSnZ2dpnGeCbHKcv3TEREREqu2rdwn06fVn14cdCLvLLyFX45/Aut67Xm3r730qdVnzI/17hx4xg5ciT16tXjkksu4ZdffvG53eTJk7nnnnuIi4sjJyeH3r17M3XqVIYNG8bVV1/NnDlzmDJlClOmTOHmm29mwoQJNGrUiH//+9+FjhUfH8/zzz/PsGHDyM7OJjQ0lBdeeMFTkgBOy2pCQgLZ2dm8+eabAIXOdTYWLVrEhAkTCA0NJSIigrfffpvWrVuTkJBA+/btOffcc+nZs6dn+9tvv50hQ4bQtGlTFi5c6FnepUsXRo8ezYUXXgjArbfeSkJCwlmVVAwZMoSpU6fSoUMH2rVrR/fu3X1u17FjR8aPH8+gQYPIy8sjNDSUV199lZYtW3L77bcTFxdHly5dePfdd33GWFpncq2jR4/mzjvvpFatWixfvtxToiTVW25eLsFBp//mREREzpyx1gY6Br9KTEy0ycnJXss2bdpEhw4dAhRR5dK3b191tBMpRmX9e7I/cz+p6ank5OUQEhRCdGQ0UeGF+x2IiEjJGWNWWWsLJU1q4RYRqWb2Z+5n19FdAOTZPHLycjzPlXSLiJQ9JdxSrEWLFgU6BBEpY6npqQDsTt9NelY6kWGRNI9sTmp6qhJuERE/UKdJEZFqJDcvl5y8HPJsHulZ6RhjSM9K97R05+aVfhZTEREpnhJuEZFqJDgomJCgEIJMEJFhkVhriQyLJMgEERIUog6UIiJ+oJISEZFqJjoyml1Hd9E8sjl5EXkEmSDP8qpII7GISKAp4RYRqWbcddpVfZQSjcQiIhWFSkoC5NlnnyUmJoa4uDji4+P5/vvvAWc85fzTtpdGSkoKnTp1KtU+u3fv5oorrqBt27acd955PPDAA5w8eRKA6dOnc++9vif46dGjxxnFOHv2bK/re+qpp1iwYMEZHSs/ay3jx4+nbdu2XHDBBfTr148NGzacdr/p06ezZ88ez/PSvv7Jycncf//9pY73yJEjvPbaa57ne/bs4eqrry71cXzp27cv7dq1o3PnzvTs2ZMtW7ac9TFTUlJ47733Sr1PrVq1iI+P9zzc99bZ+Nvf/nbWxxAn6e7cpDPxTeLp3KRzlUtE3SOxuOvV3SOx7M/cH+jQRKQaUsJ9OiNGQJcuhR8jRpzxIZcvX85nn33G6tWrWbduHQsWLODcc88F4I033qBjx45lFX2xrLWMGDGCK6+8kq1bt/LTTz+RkZHBE088cdp9ly1bdkbnLJhw/+Uvf2HAgAFndKz8Xn31VZYtW8batWv56aefeOyxxxg+fPhpp4MvmHCX9vVPTExk8uTJpY63YMLdtGlTZs6cWerjFOXdd99l7dq13HTTTTz88MMl2icnJ6fIdWeScAOcd955rFmzxvOoUaNGic5XHCXcZauqllrkH4lly8Et7E7f7bVcRKQ8KeE+nZQUaNas8OMsZi9MTU2lYcOG1KxZE4CGDRvStGlTwGmddE/UExERwRNPPEHnzp3p3r07+/btA2D79u10796d2NhYnnzySSIiIgqdIzc3l4cffphu3boRFxfHP//5z0LbfPPNN4SFhXHzzTcDEBwczMsvv8ybb77JsWPHANi1axd9+/albdu2PPPMM559859zwoQJnvM8/fTTnuVvv/02cXFxdO7cmRtvvJFly5Yxd+5cHn74YeLj49m+fTujR49m5syZfPnll4wcOdKz76JFixg6dCgA8+fP5+KLL6ZLly6MHDmSjIyMQtfy/PPP88orr1C7dm0ABg0aRI8ePXj33Xc98Y4ZM4aYmBj69+/PgQMHmDlzJsnJyVx//fXEx8dz/PjxQq//ww8/TExMDAMGDOCHH36gb9++tGnThrlz5xaK89tvv/W05CYkJJCenk5GRgb9+/enS5cuxMbGMmfOHAAeffRRtm/fTnx8PA8//LDXtxNZWVncfPPNxMbGkpCQ4JlNc/r06YwYMYIhQ4bQtm1bHnnkkUKvQ0G9e/dm27ZtpKSkkJSURJcuXejSpYvnA9OiRYtISkpi+PDhdOzYscj75tFHH2XJkiXEx8fz8ssvFxnj6RQ8X2mv9dFHH+X48ePEx8dz/fXXl+icUv1oJBYRqWhUw/366/Dzz0Wv37MH0tIKL8/IgMce871PmzZw221FHnLQoEH85S9/4YILLmDAgAFcc8019OlTeKr4zMxMunfvzrPPPssjjzzC66+/zpNPPskDDzzAAw88wKhRo5g6darPc/zrX/+ibt26rFy5khMnTtCzZ08GDRpE69atPdts2LCBrl27eu1Xp04dWrRowbZt2wD44YcfWL9+PbVr16Zbt25cfvnlXrNOzp8/n61bt/LDDz9grWX48OEsXryYBg0aMH78eJYtW0bDhg357bffqF+/PsOHD2fo0KGFyicGDBjA7bffTmZmJuHh4cyYMYNrr72WgwcPMn78eBYsWEB4eDjPP/88L730Ek899ZRn37S0NDIzM2nTpo3XMRMTEz1lJZmZmSQmJvLyyy/zl7/8hWeeeYZXXnmFV155pciZNDMzM7nkkkuYMGECv/vd73jyySf56quv2LhxIzfddBPDhw/32n7ixIm8+uqr9OzZk4yMDMLCwgCYNWsWderU4eDBg3Tv3p3hw4fz3HPPsX79etasWQPgNSX7q6++ijGGH3/8kc2bNzNo0CB++uknANasWcP//vc/atasSbt27bjvvvs834748umnnxIbG0tUVBRfffUVYWFhbN26lVGjRnk+WKxevZr169fTunVrpk2b5vO+ee6555g4cSKfffYZAC+++KLPGN3X7Ob+UAHQs2dPRo4c6XW+oo5T1LU+99xzvPLKK57XTcQX90gsOXk5RIZFesYa10gsIhIoSrgDICIiglWrVrFkyRIWLlzINddcw3PPPcfo0aO9tqtRo4an9bRr16589dVXgFOSMnv2bACuu+46HnrooULnmD9/PuvWrfOUKRw9epStW7d6JdwlMXDgQBo0aADAiBEjWLp0aaGEe/78+SQkJACQkZHB1q1bWbt2LSNHjqRhw4YA1K9fv9jzhISEMGTIED799FOuvvpqPv/8c1544QW+/fZbNm7cSM+ePQE4efIkF198camuASAoKIhrrrkGgBtuuIERJSgJqlGjBkOGDAEgNjaWmjVrEhoaSmxsrFeC7NazZ0/Gjh3L9ddfz4gRI2jevDnZ2dk8/vjjLF68mKCgIH799VfPNxVFWbp0Kffddx8A7du3p2XLlp4ktH///tStWxeAjh07smPHDp8J9/XXX0+tWrVo1aoVU6ZMITs7m3vvvZc1a9YQHBzsOR7AhRde6Lkvirpv8peCFBdjXFyc13bukhK3RYsWeZ2vLK5VxJfqNhKLiFRsSriLaYkG4L//dUpICvr1V/j738/4tMHBwfTt25e+ffsSGxvLW2+9VSjhDg0NxRjj2b40Na/WWqZMmcLgwYOL3KZjx46F6obT0tLYuXMn559/PqtXr/ac363gc2stjz32GHfccYfX8ilTppQ4Vrdrr72WV155hfr165OYmEhkpDNG8MCBA3n//feL3K9OnTqEh4fz888/e7Vyr1q1yuc3B76uw5f8r39QUJCnBCgoKMjne/Hoo49y+eWXM2/ePHr27Ml///tfVqxYwYEDB1i1ahWhoaG0atXqtHXlxXHHAMXfE++++67XB6Nx48bRuHFj1q5dS15enldLdHh4uOf3ou6bspxxNP/5ilPSaxXxpbqMxCIilYNquANgy5YtbN261fN8zZo1tGzZssT7d+/enY8//hiADz74wOc2gwcP5v/9v/9HdnY2AD/99BOZmZle2/Tv359jx47x9ttvA07d94MPPsjo0aM9tdBfffUVv/32G8ePH2f27Nmelub853nzzTc9ddW//vor+/fv55JLLuGjjz7i0KFDAPz2228AREZGkp6e7jPmPn36sHr1al5//XWuvfZaz7V+9913nhKXzMxMr9ZZt4cffpj777+f48ePA7BgwQKWLl3KddddB0BeXp7nw8V7771Hr169ThtPaW3fvp3Y2Fj+9Kc/0a1bNzZv3szRo0eJiooiNDSUhQsXsmPHjtOeNykpyVN7/tNPP7Fz507atWt3VrEdPXqU6OhogoKC+M9//kNuru8a1qLum4LxllWMZ3Kc0NBQT3wixanqI7GISOWhFu7TadXKdwfJVq3O+JAZGRncd999HDlyhJCQEM4//3ymTZtW4v0nTZrEDTfcwLPPPsuQIUM8X7vnd+utt5KSkkKXLl2w1tKoUSNPGYqbMYZZs2Zx991389e//pW8vDwuu+wyr1EgLrzwQq666ip2797NDTfc4Gk1dbf8Dho0iE2bNnnKPCIiInjnnXeIiYnhiSeeoE+fPgQHB5OQkMD06dO59tprue2225g8eXKh1vXg4GCGDh3K9OnTeeuttwBo1KgR06dPZ9SoUZw4cQKA8ePHc8EFF3jte99993H48GFiY2MJDg6mSZMmzJkzh1q1agFOq+oPP/zA+PHjiYqKYsaMGQCMHj2aO++8k1q1arF8+fISvwe+TJo0iYULFxIUFERMTAyXXnop6enpDBs2jNjYWBITE2nfvj0ADRo0oGfPnnTq1IlLL72Ue+65x3Ocu+++m7vuuovY2FhCQkKYPn26V2vvmbj77ru56qqrePvttxkyZEiRrcxF3TdxcXEEBwfTuXNnRo8eXWYxnslxbr/9duLi4ujSpYsnWRcpjmq2RSTQjLW2/E5mTMHhJWoBr1lr73Ot7w+8CrQAvgdGW2t3uNbVBP4fcDVwDHjBWvvS6c6ZmJho3Z3D3DZt2kSHDh3O8moC59ixY9SqVQtjDB988AHvv/++Z/SL8nDo0CG6dOniaa2tDCIiInyObiJytir73xMRESk7xphV1tpCIzGUa0mJtTbC/QCaAMeBj1wBNgQ+Af4M1AeSgRn5dh8HtAVaAv2AR4wxQ8or9oyTGfz+o9+TcTLwSduqVauIj48nLi6O1157jRdffLHczr1nzx4uvvhinx01RURERKSwQJaUXAXsB5a4no8ANlhr3Qn4OOCgMaa9tXYzcBNOi/dh4LAx5nVgNPBleQT79c9f89HGj7gh7gaGtxt++h38KCkpibVr1wbk3E2bNvVZQ13RqXVbREREAiWQnSZvAt62p2paYgBPFmmtzQS2AzHGmHpAdP71rt9jyilWZm2e5fzcNKu8TikiIiIiVUBAEm5jTEugD/BWvsURwNECmx4FIl3rKLDevc7X8W83xiQbY5IPHDhw1vFaa/nsJ2fCj09/+pTyrHsXERERkcotUC3cNwJLrbW/5FuWAdQpsF0dIN21jgLr3esKsdZOs9YmWmsTGzVqdNbBbjywkawcZ+zk4znH2XRw01kfU0RERESqh0Al3H/Au3UbYAPQ2f3EGBMOnIdT130YSM2/3vX7Bj/HCcC8rfPIyXMm3cizeczbOq88TisiIiIiVUC5J9zGmB5AM1yjk+QzC+hkjLnKGBMGPAWsc3WYBHgbeNIYU88Y0x64DZheHjF/uOFDTuQ6Y0Bn5WTx4YYPz/qYwcHBxMfH06lTJ0aOHMmxY8fO+piLFi1i2bJlpd7PPc5yhw4diI2NLTRed0nONXXqVM8EOiXVo0eP0oYKOONd53+9LrvsMo4cOXJGx6ro+vbtS8FhLUsjOzubLl26nFUMt956Kxs3bix2m9GjRxcaVx0gJSWF995776zOLyIiUtkFYpSSm4BPrLVe5SDW2gPGmKuAV4B3cMbhvjbfJk/jjMO9A2c4weettWUyQslVM67ik82fFLm+RnANr+dr963FPFP01OAj2o/g42s+LvactWrVYs2aNQBcf/31TJ06lbFjx5421pycHEJCfL9tixYtIiIiolSJ7Nq1a3nooYf46quvaN26Nb/88gsDBw6kTZs2xMXFFblfwXPdeeedJT6n25l8OIBTE/+4Z8OcN0/fOBRl6dKlhWYHLa033njjjPd1J9zuGT9FRESqo3Jv4bbW3mGtvbGIdQuste2ttbWstX2ttSn51p2w1t5ira1jrW1ckklvSuq5Ac8R3ySe8FDfs++dzD1Z7HO38NBwEpok8NyA50p1/qSkJLZt28ann37KRRddREJCAgMGDGDfvn0AjBs3jhtvvJGePXty4403cuDAAa666iq6detGt27d+O6770hJSWHq1Km8/PLLxMfHs2TJElJSUrjkkkuIi4ujf//+7Ny5s9C5J06cyOOPP07r1q0BaN26NY899hgTJkwAnBbWBx54wNMa/8MPP/g817hx45g4caJnnzFjxpCYmEiHDh1YuXIlI0aMoG3btjz55JOec0dEOH1hU1NT6d27t+ccS5Y4I0XeddddJCYmEhMTw9NPPw3A5MmT2bNnD/369aNfv34AtGrVioMHDwLw0ksv0alTJzp16sSkSZMAJ+nr0KEDt912GzExMQwaNMgzBXx+o0eP5q677qJ79+60adOGRYsWccstt9ChQwdGjx7t2c5XXEePHqVdu3Zs2bIFgFGjRvH66697Hf/LL79k5MiRnueLFi1i6NChRR6zIPfrBTBz5kxPTL7uh/znvPTSS5kwYQKTJ08GYMyYMVxyySUAfPPNN1x//fUAzJ8/n4svvpguXbowcuRIz1CK+VvZ//Wvf3HBBRdw4YUXctttt3Hvvfd6zrV48WJ69OhBmzZtPK3djz76KEuWLCE+Pp6XX37Z53WJiIhUedbaKv3o2rWrLWjjxo2FluXk5tiJ3020tcbXskHPBFnGUeJH0DNBttb4WvbFZS/a3LzcQsf2JTw83FprbXZ2th0+fLh97bXX7G+//Wbz8vKstda+/vrrduzYsdZaa59++mnbpUsXe+zYMWuttaNGjbJLliyx1lq7Y8cO2759e892EyZM8Jxj6NChdvr06dZaa//1r3/ZK664olAcCQkJds2aNV7L1qxZYxMSEqy11vbp08feeuut1lprv/32WxsTE+PzXPmf9+nTxz7yyCPWWmsnTZpko6Oj7Z49e2xWVpZt1qyZPXjwoNdrMHHiRDt+/HjnfcjJsWlpadZaaw8dOuRZ1qdPH7t27VprrbUtW7a0Bw4c8Jzb/Tw5Odl26tTJZmRk2PT0dNuxY0e7evVq+8svv9jg4GD7v//9z1pr7ciRI+1//vOfQq/FTTfdZK+55hqbl5dnZ8+ebSMjI+26detsbm6u7dKli2f/ouKaP3++7d69u33//fft4MGDCx0/OzvbnnvuuTYjI8Naa+2dd97piaOoY/bp08euXLnS6/Wy1tqPPvrI3nTTTdbaou8Ha63t1q2bzczMtMuXL7dXX321tdbaXr162W7dutmTJ0/acePG2alTp9oDBw7YpKQkT2zPPfecfeaZZ7xi+PXXX23Lli3toUOH7MmTJ22vXr3sPffc43ntrr76apubm2s3bNhgzzvvPGuttQsXLrSXX355odeiKvH190RERKonINn6yEcDOfFNhRIcFMyDPR5keLvh/H7m79l6aCuZ2Zmn3a92aG3aNWjHjKtn0LZB2xKf7/jx48THxwNOC/cf//hHtmzZwjXXXENqaionT570tDoDDB8+nFq1agGwYMECr5ratLQ0nxO7LF++nE8+cUplbrzxRh555JESx5ffqFGjAOjduzdpaWklqpcePtyZHCg2NpaYmBiio6MBaNOmDbt27aJBgwaebbt168Ytt9xCdnY2V155ped1+fDDD5k2bRo5OTmkpqaycePGYstcli5dyu9+9zvCw51vKkaMGMGSJUsYPnw4rVu39hy3a9eupKSk+DzGsGHDMMYQGxtL48aNiY2NBSAmJoaUlBTi4+OLjGvgwIF89NFH3HPPPT4nJgoJCWHIkCF8+umnXH311Xz++ee88MILZ3St+RV1Pxw9epT69etTu3ZtunbtyqpVq0hLS6NmzZp06dKF5ORklixZwuTJk1mxYgUbN270lJ+cPHmSiy++2Os8P/zwA3369KF+/foAjBw50msSpCuvvJKgoCA6duzo+XZGREREAjvTZIXUtkFbkm9L5rmlzzF+yXjPcIC+hIWE8Xivx3ks6TGCTOmqc/LXcLvdd999jB07luHDh7No0SLGjRvnWedOIgHy8vJYsWIFYWFhpTqnLx07dmTVqlV07nxqAJhVq1YRE3NqTiFjvOvVCz73pWbNmgAEBQV5fnc/z8nJ8dq2d+/eLF68mM8//5zRo0czduxYkpKSmDhxIitXrqRevXqMHj2arKyi34uSxgNOh1VfJSUlifuXX34pMq68vDw2bdpE7dq1OXz4MM2bNy90/GuvvZZXXnmF+vXrk5iYSGRkZLHHzC//655/fVH3w4wZMxg8eDAAoaGhtG7dmunTp9OjRw/i4uJYuHAh27Zto0OHDmzfvp2BAwfy/vvvn/a1LEr+18tqrHoRERGPQM40WWEFBwXTKapToc6SBdUIrkFs49hSJ9tFOXr0KM2aNQPgrbcKjpp4yqBBg5gyZYrnuTtxj4yMJD39VF/UHj168MEHHwDw7rvvkpSUVOhYDz30EH//+989Lb4pKSn87W9/48EHH/RsM2PGDMBpQa5bty5169YtdK6zsWPHDho3bsxtt93GrbfeyurVq0lLSyM8PJy6deuyb98+vvjiC8/2RZ07KSmJ2bNnc+zYMTIzM5k1a5bPaz4bxcX18ssv06FDB9577z1uvvlmsrOzC+3fp08fVq9ezeuvv86111572mPm17hxYzZt2kReXh6zZp2a8bSo+8Fdv+3m/hDTu3dvkpKSmDp1KgkJCRhj6N69O9999x3btm0DIDMz06v1GpxvIr799lsOHz5MTk4OH39cfMdgKPq9EhERqU6UcBdh1uZZpJ8oPlFIP5FeplO9jxs3jpEjR9K1a1caNmxY5HaTJ08mOTmZuLg4OnbsyNSpUwGnHGLWrFmejoxTpkzh3//+N3FxcfznP//hH//4R6FjxcfH8/zzzzNs2DDat2/PsGHDeOGFFzzlFwBhYWEkJCRw55138q9//cvnuc7GokWL6Ny5MwkJCcyYMYMHHnjA87x9+/Zcd911XiNt3H777QwZMsTTadKtS5cujB49mgsvvJCLLrqIW2+9lYSEhLOKraCi4tqyZQtvvPEGL774IklJSfTu3Zvx48cX2j84OJihQ4fyxRdfeDpMFnet+T333HMMHTqUHj16eEp0wPf9kJuby7Zt22jfvr1nu6SkJFJTU7n44otp3LgxYWFhng8kjRo1Yvr06YwaNYq4uDguvvhiNm/e7HX+Zs2a8fjjj3PhhRfSs2dPWrVqRd26dYt9veLi4ggODqZz587qNCkiItWWqepf/SYmJtqC4xhv2rSJDh06FLmPtZZGExpx6Pghz7IgE0TN4JqcyD1Bns3zLG9QqwEHHj5QojKLyqhv375MnDiRxMTEQIcipbB06VLeeecdz4exspKRkUFERAQ5OTn87ne/45ZbbuF3v/tdmZ6jsjnd3xMREak+jDGrrLWFkia1cPuw8cBGjuecqvGtHVqbxPqxJOfeSmL9WK/hAzXVu1REvXr1KvNkG5xvYdzDN7Zu3Zorr7yyzM8hIiJS1ajTpA/zts4jNy/X06r9135/5f9q9SPooYdZNuFNJh1fyJ8X/pkTuSfIzctl3tZ5dGzUMdBh+8WiRYsCHYJUIO6x1kVERKTk1MLtw4cbPiQ7L5vOjTuz9s61jL14LEGrVsNvvxG8+n882ONB1t65lrjGcWTnZZfJVO8iIiIiUjUp4fahSUQTJgycQPLtyc7Y2tbCt99C69aweDFY6xk+8IUBL9A4vHGgQxYRERGRCkolJT58et2n3gtSU+HQIWjWDH79Ffbuhehoz2Q5D/Z40PeBRERERKTaUwt3Sfz4I+TlgTFOa/ePPwY6IhERERGpJJRwl8SSJeCe6bF2bef5WXr22WeJiYkhLi6O+Ph4vv/+ewBuvfVWr2m6SyMlJYVOnTqVap/du3dzxRVX0LZtW8477zweeOABTp48CcD06dO59957fe7Xo0ePM4px9uzZXtf31FNPsWDBgjM6Vn7WWsaPH0/btm254IIL6NevHxs2bDjtftOnT2fPnj2e56V9/ZOTk7n//vtLHe+RI0d47bXXPM/37NnD1VdfXerjVBYRERFntf+KFSu47bbbzuoYJblnW7VqxcGDBwstX7RoEcuWLTur84uISPWlkhKAV16BefOcVmxfsrOd+m2Ac86BZctgyBDf2wYFwWWXQRGJKsDy5cv57LPPWL16NTVr1uTgwYOeJPeNN944iwspHWstI0aM4K677mLOnDnk5uZy++2388QTTzBhwoRi9z3T5GP27NkMHTqUjh2dUV3+8pe/nNFxCnr11VdZtmwZa9eupXbt2syfP5/hw4ezYcOGQlOe5zd9+nQ6depE06ZNgdK//omJiWc0Rrk74b777rsBaNq0KTNnziz1caqLL774giFF/ZsrobNJmBctWkRERMQZf9AUEZHqTS3cAL//PcTHO+UiUVHQsqX34/zzITjY2TY42HlecJuoKGf/hATneMVITU2lYcOG1KxZE4CGDRt6Er6+ffvinqgnIiKCJ554gs6dO9O9e3f27dsHwPbt2+nevTuxsbE8+eSTPlsPc3Nzefjhh+nWrRtxcXH885//LLTNN998Q1hYGDfffLPr0oJ5+eWXefPNNzl27BgAu3btom/fvrRt25ZnnnnGs2/+c06YMMFznqefftqz/O233yYuLo7OnTtz4403smzZMubOncvDDz9MfHw827dvZ/To0cycOZMvv/ySkSNHevZdtGiRZybG+fPnc/HFF9OlSxdGjhxJRkZGoWt5/vnneeWVV6hduzbgTHfeo0cP3n33XU+8Y8aMISYmhv79+3PgwAFmzpxJcnIy119/PfHx8Rw/frzQ6//www8TExPDgAED+OGHH+jbty9t2rRh7ty5heL89ttviY+PJz4+noSEBNLT08nIyKB///506dKF2NhY5syZA8Cjjz7K9u3biY+P5+GHH/b6diIrK4ubb76Z2NhYEhISWLhwIeB8OBgxYgRDhgyhbdu2PPLII4VeB3BaaR977DHi4+NJTExk9erVDB48mPPOO88zNndRca1cuZK4uDiysrLIzMwkJiaG9evXex3/0Ucf5dVXX/U8HzduHBMnTizymPnlf70A7r33XqZPnw7AqlWr6NOnD127dmXw4MGkpqZ6tvv6668ZMGAAl19+OevWrQMgISHB84Htqaee4vXXXweKvh/d92xeXh5333037du3Z+DAgVx22WVeH3amTJniuYbNmzeTkpLC1KlTefnll8tkZlUREal+lHCDkyyPHw8PPghpaU4nyZLOwGmts31amrP/+PHO8YoxaNAgdu3axQUXXMDdd9/Nt99+63O7zMxMunfvztq1a+ndu7cnoXjggQd44IEH+PHHH2nevLnPff/1r39Rt25dVq5cycqVK3n99df55ZdfvLbZsGEDXbt29VpWp04dWrRowbZt2wD44Ycf+Pjjj1m3bh0fffQRBWftnD9/Plu3buWHH35gzZo1rFq1isWLF7NhwwbGjx/PN998w9q1a/nHP/5Bjx49GD58OBMmTGDNmjWcd955nuMMGDCA77//nszMTABmzJjBtddey8GDBxk/fjwLFixg9erVJCYm8tJLL3nFkJaWRmZmJm3atPFanpiY6CkryczM9Dzv06cPzzzzDFdffTWJiYm8++67rFmzhlq1ahV6/S+55BI2bNhAZGQkTz75JF999RWzZs3iqaeeKvSaT5w4kVdffZU1a9awZMkSatWqRVhYGLNmzWL16tUsXLiQBx98EGstzz33HOeddx5r1qwp9G3Cq6++ijGGH3/8kffff5+bbrqJrKwsANasWcOMGTP48ccfmTFjBrt27fLx7kOLFi1Ys2YNSUlJng81K1as8CSgRcXVrVs3hg8fzpNPPskjjzzCDTfcUKhM6ZprruHDD08Nhfnhhx9yzTXXFHnMksjOzua+++5j5syZrFq1iltuuYUnnngCgIMHDxIaGkrdunVJSkpiyZIlHD16lJCQEL777jsAlixZQu/evYu8H/P75JNPSElJYePGjfznP/9h+fLlXusbNmzI6tWrueuuu5g4cSKtWrXizjvvZMyYMZ7XVEREpDSUcLsFBcGgQTBtGnTqBDt2wPHjxe9z/DikpEBsrLPfoEFOx8rTiIiIYNWqVUybNo1GjRpxzTXXeFr58qtRo4anNbBr166kpKQATkmKuzX4uuuu83mO+fPn8/bbbxMfH89FF13EoUOH2Lp162ljK2jgwIE0aNCAWrVqMWLECJYuXVroPPPnzychIYEuXbqwefNmtm7dyjfffMPIkSNp2LAhAPXr1y/2PCEhIQwZMoRPP/2UnJwcPv/8c6644gpWrFjBxo0b6dmzJ/Hx8bz11lvs2LGj1NcRFBTENddcA8ANN9xQ6Dp8qVGjhqeMITY2lj59+hAaGkpsbKznvcivZ8+ejB07lsmTJ3PkyBFCQkKw1vL4448TFxfHgAED+PXXXz3fVBRl6dKl3HDDDQC0b9+eli1b8tNPPwHQv39/6tatS1hYGB07dizytRg+fLgn7osuuojIyEgaNWpEzZo1OXLkSLFxPfXUU3z11VckJyf7bEVPSEhg//797Nmzh7Vr11KvXj3OPffcM7pWty1btrB+/XoGDhxIfHw848ePZ/fu3YBzjw0aNAiApKQkFi9ezHfffcfll19ORkYGx44d45dffqFdu3ZF3o8FX9+RI0cSFBREkyZN6Nevn9f6ESNGAN7/5kRERM6GargLcrd2f/EF/OMfp2q3fdm7Fx54wKnZLkGinV9wcDB9+/alb9++xMbG8tZbbzF69GivbUJDQzGu4wYHB5OTk1Pi41trmTJlCoMHDy5ym44dOxaqG05LS2Pnzp2cf/75rF692nN+t4LPrbU89thj3HHHHV7Lp0yZUuJY3a699lpeeeUV6tevT2JiIpGRkVhrGThwIO+//36R+9WpU4fw8HB+/vlnr1Zud4mCLwWvw5f8r39QUJCnBCgoKMjne/Hoo49y+eWXM2/ePHr27Ml///tfVqxYwYEDB1i1ahWhoaG0atXK01p9JtwxQPH3RP5Y8+/jjv3dd98tMq5Dhw6RkZFBdnY2WVlZhLs7DOczcuRIZs6cyd69ez0fZIo7pltISAh5+fpKuNdba4mJiSnU2gxO/fbYsWMB6NatG8nJybRp04aBAwdy8OBBXn/9dc83NUXdj6Xhfr1K+29ORESkKGrh9iUoCJo3h9DQ4rerUcPZrpTJ9pYtW7xa3dasWUPLli1LvH/37t35+OOPAfjggw98bjN48GD+3//7f2RnZwPw008/eco13Pr378+xY8d4++23Aafu+8EHH2T06NGeWuivvvqK3377jePHjzN79mx69uxZ6Dxvvvmmp676119/Zf/+/VxyySV89NFHHDp0CIDffvsNgMjISNLT033G3KdPH1avXs3rr7/Otdde67nW7777zlPikpmZ6Wntze/hhx/m/vvv57jrW4kFCxawdOlSzzcAeXl5ng8X7733Hr169TptPKW1fft2YmNj+dOf/kS3bt3YvHkzR48eJSoqitDQUBYuXOhpkS7uvElJSZ7a859++omdO3fSrl27MonRrai4AO644w7++te/cv311/OnP/3J5/7XXHMNH3zwATNnzvR821LcMd1atmzJxo0bOXHiBEeOHOHrr78GoF27dhw4cMCTcGdnZ7Nhwwastaxbt474+HjA+dbh3HPP5aOPPuLiiy8mKSmJiRMn0rt3b6Do+zG/nj178vHHH5OXl8e+fftYtGjRaV+vsrxPRESk+lELd1GSk0+fSFvrbNe5c6kOnZGRwX333ecpOzj//POZNm1aifefNGkSN9xwA88++yxDhgyhbt26hba59dZbSUlJoUuXLlhradSoEbNnz/baxhjDrFmzuPvuu/nrX/9KXl4el112GX/7298821x44YVcddVV7N69mxtuuMEzIoe75XfQoEFs2rSJiy++GHDKZd555x1iYmJ44okn6NOnD8HBwSQkJDB9+nSuvfZabrvtNiZPnlyodT04OJihQ4cyffp03nrrLQAaNWrE9OnTGTVqFCdOnABg/PjxXHDBBV773nfffRw+fJjY2FiCg4Np0qQJc+bM8dRlh4eH88MPPzB+/HiioqKYMWMGAKNHj+bOO++kVq1aPltXS2PSpEksXLiQoKAgYmJiuPTSS0lPT2fYsGHExsaSmJhI+/btAWjQoAE9e/akU6dOXHrppdxzzz2e49x9993cddddxMbGEhISwvTp071aqcvC9ddf7zOut99+m9DQUK677jpyc3Pp0aMH33zzDZdcconX/jExMaSnp9OsWTOio6OLPWZ+5557Lr///e/p1KkTrVu3JiEhAXAS6ZkzZ3L//fdz9OhRcnJy+L//+z+OHz9OQkKC1zcSSUlJfP3119SqVYukpCR2797tqasu6n6Mytev4qqrruLrr7+mY8eOnHvuuXTp0sXnv6H8hg0bxtVXX82cOXOYMmWK6rhFRKRUTEk7NVVWiYmJtmBHv02bNtGhQ4eid7IWbrrJ+Vm7tvNz7144edJp1W7SxEnGMzOdUUumTy91K/fZOHbsGLVq1cIYwwcffMD777/vc0QIfzl06BBdunQ5o1rqQImIiPA5uolUbOPHj+f888/3fONRVjIyMoiIiODQoUNceOGFfPfddzRp0uSMjnXavyciIlJtGGNWWWsLjResFm5fUlPh4EGnXOT4cSfZ7tYNbrgB3nnHadVu3NhJxnfv9kz1Xl5WrVrFvffei7WWc845hzfffLPczr1nzx769u3LQw89VG7nlOrrySef9Mtxhw4dypEjRzh58iR//vOfzzjZFhERKQkl3L78+OOpVu2gIHjoIRg40GnF/utfYcECeO01OHLk1FTv5ZhwJyUlsXbt2nI7X35Nmzb1WUNd0al1W/IrSd22iIhIWVGnSV+WLHGSaV/D/RUcPvDIkTKZ6l1EREREqia1cPvSoAE888ypVm1f3MMHLlgArslVREREREQKUsLty5gxJdvO3drtmpRDRERERKQglZSIiIiIiPiREu4ACQ4OJj4+nk6dOjFy5EiOHTt21sdctGgRy5YtK/V+s2fPJi4ujg4dOhAbG1tovO6SnGvq1KmeCXRKqkePHqUNFXDGu87/el122WUcOXLkjI4lIiIi4m9KuEshNy+3zI5Vq1Yt1qxZw/r166lRowZTp04t0X7FTTV9Jgn32rVreeihh5gzZw6bNm1i7ty5PPTQQ6xbt67Y/Qqe68477+QPf/hDqc59Jh8OoHDCPW/ePM4555wzOpaIiIiIvynhLoH9mftZu3cta/auYe3etezP3H/6nUohKSmJbdu28emnn3LRRReRkJDAgAED2LdvHwDjxo3jxhtvpGfPntx4440cOHCAq666im7dutGtWze+++47UlJSmDp1Ki+//DLx8fEsWbKElJQULrnkEuLi4ujfvz87d+4sdO6JEyfy+OOP07p1awBat27NY489xoQJEwDo27cvDzzwgKc1/ocffvB5rnHjxjFx4kTPPmPGjCExMZEOHTqwcuVKRowYQdu2bb3GVY6IiAAgNTWV3r17e86xxDXqy1133UViYiIxMTE8/fTTAEyePJk9e/bQr18/+vXrB0CrVq04ePAgAC+99BKdOnWiU6dOTJo0CYCUlBQ6dOjAbbfdRkxMDIMGDfJMAS8iIiLib0q4T2N/5n52Hd1FTl4OeTaPnLwcdh3dVWZJd05ODl988QWxsbH06tWLFStW8L///Y9rr72WF154wbPdxo0bWbBgAe+//z4PPPAAY8aMYeXKlXz88cfceuuttGrVijvvvJMxY8awZs0akpKSuO+++7jppptYt24d119/Pffff3+h82/YsIGuXbt6LUtMTGRDvpFXjh07xpo1a3jttde45ZZbfJ6roBo1apCcnMydd97JFVdcwauvvsr69euZPn06hw4d8tr2vffeY/DgwaxZs4a1a9cSHx8PwLPPPktycjLr1q3j22+/Zd26ddx///00bdqUhQsXsnDhQq/jrFq1in//+998//33rFixgtdff53//e9/AGzdupV77rmHDRs2cM455/Dxxx+X7o0SEREROUMapeQ0UtNTAdidvpv0rHQiwyJpHtmc1PRUosKjzvi4x48f9ySWSUlJ/PGPf2TLli1cc801pKamcvLkSU+rM8Dw4cOpVasWAAsWLGDjxo2edWlpaT4ndlm+fDmffPIJADfeeCOPPPLIGcU6atQoAHr37k1aWlqJ6qWHDx8OQGxsLDExMUS7JgZq06YNu3btokGDBp5tu3Xrxi233EJ2djZXXnml53X58MMPmTZtGjk5OaSmprJx40bi4uKKPOfSpUv53e9+R3h4OAAjRoxgyZIlDB8+nNatW3uO27VrV1JSUkr5KoiIiIicGSXcxcjNy/W0bKdnpWOMIT0rnbwIp6U7Ny+X4KDgMzq2u4Y7v/vuu4+xY8cyfPhwFi1axLhx4zzr3EkkQF5eHitWrCAsLOyMzp1fx44dWbVqFZ07d/YsW7VqFTExMZ7npsBY5AWf+1KzZk0AgoKCPL+7nxesQ+/duzeLFy/m888/Z/To0YwdO5akpCQmTpzIypUrqVevHqNHjyYrK+uMrjF/POB0WFVJiYiIiJQXlZQUIzgomJCgEIJMEJFhkVhriQyLJMgEERIUcsbJdlGOHj1Ks2bNAHjrrbeK3G7QoEFMmTLF89yduEdGRpKenu5Z3qNHDz744AMA3n33XZ+lHw899BB///vfPS2+KSkp/O1vf+PBBx/0bDNjxgzAaUGuW7cudevWLXSus7Fjxw4aN27Mbbfdxq233srq1atJS0sjPDycunXrsm/fPr744gvP9kWdOykpidmzZ3Ps2DEyMzOZNWuWz2sWERERKU9q4T6N6Mhodh3dRfPI5uRF5BFkgjzLy9q4ceMYOXIk9erV45JLLuGXX37xud3kyZO55557iIuLIycnh969ezN16lSGDRvG1VdfzZw5c5gyZQpTpkzh5ptvZsKECTRq1Ih///vfhY4VHx/P888/z7Bhw8jOziY0NJQXXnjBU34BEBYWRkJCAtnZ2bz55psAhc51NhYtWsSECRMIDQ0lIiKCt99+m9atW5OQkED79u0599xz6dmzp2f722+/nSFDhnhqud26dOnC6NGjufDCCwG49dZbSUhIUPmIiIiIBJSx1gY6Br9KTEy0ycnJXss2bdpEhw4dSnyM/Zn7SU1PJScvh5CgEKIjo8+qfrsy6du3LxMnTiQxMTHQoYhUSKX9eyIiIlWXMWaVtbZQ0qQW7hKICo8iKjzqrGq2RURERKR6UsJdCtUx2V60aFGgQxARKVpaGuzcCSdOeC+vWRNatIA6dQITl4hIPuo0KSIilZevZBucZT4m+xIRCQS1cIuISOXlTrY3bfJe3qGD70RcRCQA1MItIiIiIuJHSrhFRERERPxICXcA9OvXj//+979eyyZNmsRdd93lc/u+fftScGjD0sjOzqZLly6MGTOGSZMmeZYPHjyYW2+91fP8wQcf5KWXXiryOE899RQLFiwo9lzjxo1j4sSJhZYfOXKE1157rfTBi4iIiFRySriLk5YG69fDqlXej/XrnXVnaNSoUZ4ZIN0++OADRo0adbYR+7R06VJ69uxJz549WbZsGeBMD3/w4EE2bNjg2W7ZsmX06NGjyOP85S9/YcCAAWcUgxJuEfGLmjWdnx06eD/yrxMRCTAl3MXxU+/3q6++ms8//5yTJ08CznTqe/bs4f333ycxMZGYmBiefvppn/tGRER4fp85cyajR48G4MCBA1x11VV069aNbt268d1333m2+/LLL7n00kvp0aMHy5cvB2DDhg106tSJyMhIDh8+zIkTJ9i0aRNdunRh1apV9OnTh65duzJ48GBSU1MBGD16NDNnzgRg3rx5tG/fnq5du3L//fczdOhQz/k2btxI3759adOmDZMnTwbg0UcfZfv27cTHx/Pwww+f8WsnIuKlRQvfibV7WEARkQpAo5QUx0+93+vXr8+FF17IF198wRVXXMEHH3zA73//ex5//HHq169Pbm4u/fv3Z926dcTFxZXomA888ABjxoyhV69e7Ny5k8GDB7PJFffChQt5+umnqV27NiEhIezcuZNly5Zx8cUX8+uvv7J8+XLq1q1LbGwsxhjuu+8+5syZQ6NGjZgxYwZPPPGEZ0p3gKysLO644w4WL15M69atC7XMb968mYULF5Kenk67du246667eO6551i/fj1r1qw549dNRKSQOnWgU6dARyEiUiwl3AHiLitxJ9z/+te/+PDDD5k2bRo5OTmkpqaycePGEifcCxYsYOPGjZ7naWlpZGRkcPToUerXr0/t2rUB6NGjB8uWLWPZsmWMHTuWX3/9lWXLllG3bl169uzJli1bWL9+PQMHDgQgNzeX6Ohor3Nt3ryZNm3a0Lp1a8+1TJs2zbP+8ssvp2bNmtSsWZOoqCj27dt3Vq+ViIiISGWmhDtArrjiCsaMGcPq1as5duwY9evXZ+LEiaxcuZJ69eoxevRosrKyCu1njPH8nn99Xl4eK1asICwszGv7GTNmMHjwYM9zdx33jz/+SKdOnTj33HN58cUXqVOnDjfffDPWWmJiYjylJ2eiZr6vd4ODg8nJyTnjY4mIiIhUdqrhDpCIiAj69evHLbfcwqhRo0hLSyM8PJy6deuyb98+vvjiC5/7NW7cmE2bNpGXl8esWbM8ywcNGsSUKVM8z92lG+76bbcePXrw2WefUb9+fYKDg6lfvz5Hjhxh+fLl9OjRg3bt2nHgwAFPwp2dne3VsRKgXbt2/Pzzz6SkpABOUn86kZGRpKenl+i1EREREalKlHAXx8+930eNGsXatWsZNWoUnTt3JiEhgfbt23PdddfRs2dPn/s899xzDB06lB49eniVekyePJnk5GTi4uLo2LEjU6dOJTc3l23bttG+fXvPdrGxsRw8eJDu3bt7Latbty4NGzakRo0azJw5kz/96U907tyZ+Ph4z8gmbrVq1eK1115jyJAhdO3alcjISOrWrVvstTZo0ICePXvSqVMndZoUERGRasVYawMdg18lJibagmNYb9q0iQ7uxLk4aWm+Rypx936vU6cMIy17S5cu5Z133mHq1KllfuyMjAwiIiKw1nLPPffQtm1bxowZU+bnEanoSvz3REREqjxjzCprbWLB5arhLk4l7/3eq1cvevXq5Zdjv/7667z11lucPHmShIQE7rjjDr+cR0RERKSyU8ItZ2TMmDFq0RYREREpAdVwi4iIiIj4UbVNuKt67bqI+J/+joiISElUy4Q7LCyMQ4cO6T9LETlj1loOHTpUaOx7ERGRgqplDXfz5s3ZvXs3Bw4cCHQoIlKJhYWF0bx580CHISIiFVy1TLhDQ0M905KLiIiIiPhTtSwpEREREREpL0q4RURERET8SAm3iIiIiIgfVfmp3Y0xB4AdgY4jABoCBwMdhFRIujekOLo/pCi6N6Q4uj8cLa21jQourPIJd3VljEm21iYGOg6peHRvSHF0f0hRdG9IcXR/FE8lJSIiIiIifqSEW0RERETEj5RwV13TAh2AVFi6N6Q4uj+kKLo3pDi6P4qhGm4RERERET9SC7eIiIiIiB8p4RYRERER8SMl3FWEMaatMSbLGPNOvmXXGWN2GGMyjTGzjTH1AxmjlD9jzCLXfZHhemzJt073RzVnjLnWGLPJdQ9sN8YkuZb3N8ZsNsYcM8YsNMa0DHSsUn7y/b1wP3KNMVPyrdf9UY0ZY1oZY+YZYw4bY/YaY14xxoS41sUbY1a57o1Vxpj4AIdbYSjhrjpeBVa6nxhjYoB/AjcCjYFjwGuBCU0C7F5rbYTr0Q50fwgYYwYCzwM3A5FAb+BnY0xD4BPgz0B9IBmYEag4pfzl+3sRATQBjgMfAej+EJz/K/YD0UA80Ae42xhTA5gDvAPUA94C5riWV3tKuKsAY8y1wBHg63yLrwc+tdYuttZm4PxxHGGMiQxAiFLx6P6QZ4C/WGtXWGvzrLW/Wmt/BUYAG6y1H1lrs4BxQGdjTPtABisBcxVOcrXE9Vz3h7QGPrTWZllr9wJfAjFAXyAEmGStPWGtnQwY4JKARVqBKOGu5IwxdYC/AGMLrIoB1rqfWGu3AyeBC8ovOqkg/m6MOWiM+c4Y09e1TPdHNWaMCQYSgUbGmG3GmN2ur4VrUfjeyAS2u5ZL9XMT8LY9NaSZ7g+ZBFxrjKltjGkGXMqppHud9R7+bh26NwAl3FXBX4F/WWt3F1geARwtsOwozlfHUn38CWgDNMMZI/VTY8x56P6o7hoDocDVQBLO18IJwJPo3hAXV212H5zSADfdH7IYJ4lOA3bjlBXNRvdGsZRwV2KuzggDgJd9rM4A6hRYVgdI93NYUoFYa7+31qa7vt57C/gOuAzdH9XdcdfPKdbaVGvtQeAldG+ItxuBpdbaX/It0/1RjRljgnBasz8BwoGGOPXaz6N7o1hKuCu3vkArYKcxZi/wEHCVMWY1sAHo7N7QGNMGqAn8VP5hSgVicWrqdH9UY9bawzgtU/m/+nX/XvDeCAfOcy2X6uUPeLdug+6P6q4+0AJ4xdWQcwj4N86H9Q1AnDHG5Ns+Dt0bgBLuym4azh+6eNdjKvA5MBh4FxhmjEly/UH8C/CJtVafNKsJY8w5xpjBxpgwY0yIMeZ6nJEovkT3hzj/Sd5njIkyxtQDxgCfAbOATsaYq4wxYcBTOHWZmwMYq5QzY0wPnFK0jwqs0v1Rjbm+DfsFuMv1/8o5OHX+64BFQC5wvzGmpjHmXtdu3wQi1opGCXclZq09Zq3d637gfJ2TZa09YK3dANyJk1jtx6mhujuA4Ur5CwXGAweAg8B9wJXW2p90fwhO/4+VON9qbAL+BzxrrT2AMzLFs8Bh4CLg2kAFKQFzEz4+hOv+EJyRaobg/N+yDcgGxlhrTwJX4nwzcgS4Bef/nJOBCbNiMd6dSUVEREREpCyphVtERERExI+UcIuIiIiI+JESbhERERERP1LCLSIiIiLiR0q4RURERET8SAm3iIiIiIgfKeEWEanEjDHjjDFnNL6rMWaRMWZpCba70hgztoTH7GuMsfkerfKtSzHGvHMmsfo4T68C5zm/LI4rIuIPSrhFRCq3N4CL/XyOK4ESJdz53IMTV2qZR+P4/+3dWajVVRTH8e/PITWi0hyaQUMwqCwsKLEcsAgCScwkgqjefDCKwAaoTM2wSxNZgRT1IoZlgkE4hCOCiSMORCqJRlZa5q3M6bp62PvQv9M5kvee/70qvw9chPX/772XPq27XXv/t+T5p5c0v5lZw3Tp6ATMzKz1IuJ74PuOzqOGHRGxtqzJ8xcQ10oaVNYaZmaN4h1uM7MSSRqSWx6GFWKTcmx6ITYwx+4rxPpLmiPpgKRjkjZLGls1/39aSiT1kTRXUrOkQ5I+kjQmzz+iRo6jJW2UdETStuIakj4mfeb7qkL7xp62/8uApM6SZuc8RxfiD0n6RtJRSVtz7iskrWjEumZm7c073GZm5doE/AaMAir90qOAv/KfFGIngVUAkq4BvgZ+Bp4CDgATgPmS7o+IhadZ83PgRuA5YBcwDninzrvXAW8DrwIHgaeBTyUNiohdwDSgD3AbMCaPOfY//t6nJakHMJfUFjIiIjbm+N3AHGAhqY2lD/AW0B34tq3rmpl1BBfcZmYliohTklYBI4GpkjoBw4H3gSckXRQRf+TnG3KrBMAUQMDwiPglxxbnQnwqqSD9D0n3AMOACRExrzBuIXBtjSG9gbsiYmcev5HUd/0gMCMidks6ABxvVIuIpJ7AF8AVwNCI2F14/DKwAxgbEZHf3wasxwW3mZ2j3FJiZla+ZcAdkroDNwOXAq+RdorvzO+MBJYXxtwLfAkcltSl8gMsBgZLurjOWrcDLcCCqvhndd7fWSm2ASLiZ9Kueq3ivBGuJO30X0hVsS2pM3ArML9SbOecNgDflZSPmVnpvMNtZla+5UA3YChwC7AlIn7KV/KNlLQX6EsqzCv6Ao/kn1ouA5prxK8ADkXEiar4T3Xm+bVG7BiphaMMN5FyfzYiqnPqDXQlFfzV6uVvZnbWc8FtZla+raT+6FGkgrtSWC8jtW7sA44DawpjfgFWAzPrzPlDnfh+oKekrlVFd7/Wpd5wi0hX+s2UdDQi3i48OwicIP2yUa0fsLcd8jMzazgX3GZmJYuIyDds3A1cD7yXHy0jHVZsBtZFxJHCsEWkA4XbI+KvM1huLdAZGAvMK8THty57IO1492jD+H+JiCZJLcBbkjpFxJs53iJpPTBO0pRCD/cQoD8uuM3sHOWC28ysfSwH3iX1V6/OsU3A7+QDlVXvvwisA1ZJmgXsAXoCNwADIuLxWotExBJJa4DZknqTbil5ABicXznVitx3AL0kTSQdXjwaEVtbMU8xzzdy0f1mLrpfz49eApYACyTNJrWZTAF+bGXuZmYdzocmzczaR+VA5PqIaIa0owusrHpOfraXdIBwCzADWEq62WQ4/+71rmUsaYd8JmmXuzvwQn52uBW5fwB8kvNYR7phpM1yO8kkoEnS5BxbCjxM+p+ABcAzpKsKf6R1uZuZdTgVDoKbmdl5Ku+SPwb0iog236N9mnVGkH55GA2sjIiTDZjzatJO/SsRMa0Q70I6VPohMDDfG25mdtZxS4mZ2XlG0qPAJcB24ALSFYMTgaYyi+0qX+Vc+kfEnv87KH8Q5408/iAwAJgMHCHttFfeG8Y/rTlmZmc1F9xmZuefP4EnSV+R7Ea6w/p5oKkd1t5A+iplRb3bVOppAS4HZpGuD/yTVFiPj4j9hfc2V62z74wzNTNrJ24pMTMzMzMrkQ9NmpmZmZmVyAW3mZmZmVmJXHCbmZmZmZXIBbeZmZmZWYlccJuZmZmZlcgFt5mZmZlZif4GV1Q+HYiYhRMAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# visualising a selected generation\n",
    "\n",
    "n_generation_examine = -1   # CHANGE ME!\n",
    "\n",
    "jitter_factor = 0           # CHANGE ME! e.g, 0.001\n",
    "\n",
    "# ----\n",
    "if ('final' == n_generation_examine) or (-1 == n_generation_examine):\n",
    "    n_generation_examine = n_generations - 1\n",
    "    \n",
    "assert n_generation_examine < n_generations\n",
    "    \n",
    "# indexes of this generation\n",
    "idx_indvs_pareto = [idx_indv for idx_indv, indv in population_all_pareto.items() if n_generation_examine in population_all_pareto[idx_indv]['generations']]\n",
    "idx_indvs_soh = [idx_indv for idx_indv, indv in population_all_soh.items() if n_generation_examine in population_all_soh[idx_indv]['generations']]\n",
    "\n",
    "# population of this generation\n",
    "population_gen_pareto = {idx_knapsack: knapsack for idx_knapsack, knapsack in population_all_pareto.items() if idx_knapsack in idx_indvs_pareto}\n",
    "popuation_gen_soh     = {idx_knapsack: knapsack for idx_knapsack, knapsack in population_all_soh.items() if idx_knapsack in idx_indvs_soh}\n",
    "\n",
    "print(f'In this generation (no. {n_generation_examine + 1})\\nthe Pareto Optimisation run had {len(idx_indvs_pareto):,} individuals and the SOH Optimisation generated {len(idx_indvs_soh):,}')\n",
    "\n",
    "indvs_pareto_weights = np.array([knapsacks['weight'] for idx_knapsack, knapsacks in population_all_pareto.items() if idx_knapsack in idx_indvs_pareto])\n",
    "indvs_pareto_values  = np.array([knapsacks['value']  for idx_knapsack, knapsacks in population_all_pareto.items() if idx_knapsack in idx_indvs_pareto])\n",
    "plt.scatter(jitter(indvs_pareto_weights, scale = jitter_factor), jitter(indvs_pareto_values, scale = jitter_factor), \n",
    "            s=size, alpha=alpha, color='green', label='Pareto Optimisation', marker=None, linewidth=4)\n",
    "\n",
    "\n",
    "\n",
    "indvs_soh_weights = np.array([knapsacks['weight'] for idx_knapsack, knapsacks in population_all_soh.items() if idx_knapsack in idx_indvs_soh])\n",
    "indvs_soh_values  = np.array([knapsacks['value']  for idx_knapsack, knapsacks in population_all_soh.items() if idx_knapsack in idx_indvs_soh])\n",
    "plt.scatter(jitter(indvs_soh_weights, scale = jitter_factor), jitter(indvs_soh_values, scale = jitter_factor),\n",
    "            s=size, alpha=alpha, color='red', label='Value/Weight', marker='s', linewidth=4)\n",
    "\n",
    "\n",
    "plot_pareto_front(population_gen_pareto, color='green', label='Pareto Optimisation Pareto Front')\n",
    "plot_pareto_front(popuation_gen_soh, color='red', label='Single Objective Optimisation Pareto Front', marker='-s')\n",
    "\n",
    "indv_ = single_objective_selector(population_gen_pareto, n_select=1)\n",
    "key = list(indv_.keys())[0]\n",
    "plt.plot(indv_[key]['weight'], indv_[key]['value'], '*', markersize=20, color='green', alpha=1, label=f'Pareto Optimisation max {heuristic}')\n",
    "\n",
    "indv_ = single_objective_selector(popuation_gen_soh, n_select=1)\n",
    "key = list(indv_.keys())[0]\n",
    "plt.plot(indv_[key]['weight'], indv_[key]['value'], '*', markersize=20, color='red', alpha=0.6, label=f'Single Objective Optimisation max {heuristic}')\n",
    "\n",
    "\n",
    "plt.legend(fontsize=10)  #(bbox_to_anchor=(1., 1))\n",
    "plt.xlabel('weight [kg]')\n",
    "plt.ylabel('value [£]')\n",
    "plt.title(f'Individuals of generation {n_generation_examine} ')\n",
    "\n",
    "# plt.xlim(*x_lim)\n",
    "# plt.ylim(*y_lim)\n",
    "pass"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:23:12.327625Z",
     "start_time": "2020-11-11T07:23:12.321464Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Duplication of Individuals in Single Objective Optimisation in generation 19 (showing top 4)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(2, 5, 7, 9, 10, 12, 15, 16, 17, 18)     32\n",
       "(1, 2, 5, 7, 8, 9, 10, 12, 15, 17)        1\n",
       "(5, 7, 9, 10, 11, 12, 15, 16, 17, 19)     1\n",
       "(2, 4, 5, 10, 11, 12, 15, 16, 17, 18)     1\n",
       "dtype: int64"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# examining by eye the top knapsack duplications for one generation (Single Objective Optimisation)\n",
    "\n",
    "top = 4\n",
    "\n",
    "print(f'Duplication of Individuals in Single Objective Optimisation in generation {n_generation_examine} (showing top {top})')\n",
    "pd.Series([tuple(sorted(knapsack['packages'])) for idx_knapsack, knapsack in popuation_gen_soh.items()]).value_counts().head(top)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:23:12.777271Z",
     "start_time": "2020-11-11T07:23:12.770900Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Duplication of Individuals in Pareto Optimisation in generation 19 (showing top 4)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(7, 9, 10, 12, 14, 15, 16, 17, 18, 19)    18\n",
       "(2, 7, 9, 10, 12, 14, 15, 16, 17, 18)     14\n",
       "(7, 8, 9, 10, 12, 14, 15, 16, 17, 19)      1\n",
       "(2, 7, 9, 12, 13, 14, 15, 16, 18, 19)      1\n",
       "dtype: int64"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# examining by eye the top knapsack duplications for on generation (Naive Pareto Selector)\n",
    "\n",
    "print(f'Duplication of Individuals in Pareto Optimisation in generation {n_generation_examine} (showing top {top})')\n",
    "pd.Series([tuple(sorted(knapsack['packages'])) for idx_knapsack, knapsack in population_gen_pareto.items()]).value_counts().head(top)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For completeness, there are a few more additional considerations for solving combinatorial problems with Genetic Algorithms:\n",
    "\n",
    "* Stop criteria: you need to determine the number of iterations  \n",
    "* Statistical significance: do you trust results of just one seed run or should you combine multiple? In [Pareto Front Approximation](#Pareto-Front-Approximation) we discussed niching where we saw that the resulting Pareto front of multiple seed runs is more likely to be better than each individual. This is worth contemplating, as always, within the boundaries of cost effectiveness.  \n",
    "\n",
    "Another takeaway is the importance of visualising. Here we put a lot of focus on visualising the Objective Space and a little on the Decision Space. Depending on your project's setup you can decide how much focus to put on each to improve decision to be made.  \n",
    "\n",
    "Also, our Objective space was 2D (`weight`, `value`). Further consideration of visualisation is required for more dimensions. E.g, 2D slices in the Objective Space using colors-grading/shapes/sizes for additional numerical/categorical dimensions.  \n",
    "\n",
    "\n",
    "**Actions To Take**\n",
    "- [ ] In [Knapsack Optimisation](#Knapsack-Optimisation) change to `n_generations=1` and increase to `2`, `3` etc. to see the step-by-step improvement of the Pareto front. \n",
    "- [ ] In [Knapsack Optimisation](#Knapsack-Optimisation) change by large steps `n_generations=10`, `20`, `30` etc. to see at what stage improvement of the Pareto front slows. \n",
    "- [ ] In [Knapsack Optimisation](#Knapsack-Optimisation) change  `parents_compete`. How does this impact the results? \n",
    "- [ ] In [Knapsack Optimisation](#Knapsack-Optimisation) change  `unique_indvs`. A `True` value ensures that all knapsacks are unique (`False` allows for duplicates). How does this impact the results?\n",
    "- [ ] In [Knapsack Optimisation](#Knapsack-Optimisation) change the value of `seed`. Has this changed any conclusions that you derived so far?  \n",
    "- [ ] In [Knapsack Optimisation](#Knapsack-Optimisation) change the value of `n_initial_population` to `100`. Is this useful?\n",
    "- [ ] In [Optimisation Objective](#Optimisation-Objective) change the optimisation objective to minimising both weight and monetary value.   \n",
    "This is done, e.g, by setting `objective_mode={'weight': 'min', 'value': 'min'}`.   \n",
    "Try to guess in advance which knapsacks are considered Pareto Optimal!   \n",
    "Try all four combinations.  \n",
    "- [ ] In [Packages](#Packages) change `value_distribution_mode=\"squared\"` to examine rules of the $value\\sim weight^2$ distribution.\n",
    "- [ ] Create your own distribution! Create your own custom `value_distribution_mode` in [Packages](#Packages).   \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Summary  \n",
    "\n",
    "Congratulations! I hope that you enjoyed learning about applying Pareto Optimisation with Genetic Algorithms.  \n",
    "\n",
    "We took a brief look at comparing results when using the commonly practiced Single Objective Optimisation.  \n",
    "Results here are more likely to be suggestive than conclusive. This is mostly due to the stochastic nature of this technique.  \n",
    "For robust conclusions one would have to run multiple seeds, but that is outside the scope of this notebook.  \n",
    "\n",
    "That said, the results suggest that Single Objective Optimisation plows its way along a narrow direction, e.g maximising the `value/weight` as seen in the default setting whereas Pareto Optimisation is agnostic of this and explores all directions.   \n",
    "\n",
    "\n",
    "Note that if you want to apply Pareto fronts with Genetic Algorithms, we suggest that you do not use the scripts above, but rather use as a reference for pedagogical purposes. In [Resources](#Resources) we suggest useful packages. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Decision Making\n",
    "\n",
    "In the previous notebook where we applied Pareto Fronts to exhaustive search spaces we discussed what decisions may be made with the final results ([Github](https://bit.ly/pareto-front), [Google Colab](https://bit.ly/pareto-front-colab)).  \n",
    "\n",
    "Everything said there applies to our current method of the stochastic search.   \n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Resources\n",
    "\n",
    "The following packages are useful for prototyping and contain state of the art Pareto optimal selection functions.  \n",
    "\n",
    "* `DEAP` - evolutionary computation framework for rapid prototyping and testing of ideas.  ([site](https://deap.readthedocs.io/en/master/) )\n",
    "\n",
    "* `pymoo` a framework that offers state of the art single- and multi-objective optimization algorithms and many more features related to multi-objective optimization such as visualization and decision making ([site](http://pymoo.org/), [github](https://github.com/msu-coinlab/pymoo/tree/6bad624cc70d12856b47c17cf0f9908a1b3653c5), [arxiv](https://arxiv.org/pdf/2002.04504.pdf) ). "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Appendix: Visualising Decision Space"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-11-11T07:23:16.146205Z",
     "start_time": "2020-11-11T07:23:15.556515Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# visulalising the decision space\n",
    "\n",
    "width = 0.4\n",
    "alpha = 0.6\n",
    "\n",
    "package_idxs, package_counts = np.unique([package_idx for idx_indv, indv in population_all_pareto.items() for package_idx in indv['packages']], return_counts=True)\n",
    "package_counts_pareto = {package_idx: package_count for package_idx, package_count in zip(package_idxs, package_counts)}\n",
    "\n",
    "package_idxs, package_counts = np.unique([package_idx for idx_indv, indv in population_all_soh.items() for package_idx in indv['packages']], return_counts=True)\n",
    "package_counts_soh = {package_idx: package_count for package_idx, package_count in zip(package_idxs, package_counts)}\n",
    "\n",
    "\n",
    "package_idxs = np.array(list(package_counts_pareto.keys()))\n",
    "\n",
    "plt.figure(figsize=(16,5))\n",
    "\n",
    "plt.subplot(1, 2, 1)\n",
    "\n",
    "plt.bar(package_idxs - width/2, package_counts_pareto.values(), color='green', width=width, alpha=alpha, label='Pareto Optimisation', hatch='/')\n",
    "plt.bar(package_idxs + width/2, package_counts_soh.values() , color='red', width=width, alpha=alpha, label=f'SOH {heuristic}')\n",
    "plt.xticks(np.arange(len(package_idxs)), package_idxs)\n",
    "\n",
    "plt.xlabel('package index')\n",
    "plt.ylabel('counts')\n",
    "plt.legend()\n",
    "\n",
    "# ----\n",
    "\n",
    "plt.subplot(1, 2, 2)\n",
    "plt.scatter([packages[idx_package]['weight'] for idx_package in packages],\n",
    "            [packages[idx_package]['value'] for idx_package in packages],\n",
    "            color = 'purple', alpha=0.04\n",
    "           )\n",
    "\n",
    "\n",
    "for idx_package, package in packages.items():\n",
    "    fontsize_ = MEDIUM_SIZE\n",
    "    plt.annotate(f'{idx_package:02.0f}', xy=(package['weight'], package['value']), fontsize=fontsize_)\n",
    "    \n",
    "plt.xlabel('weight [kg]')\n",
    "plt.ylabel('value [£]')\n",
    "plt.title('Packages')\n",
    "\n",
    "\n",
    "pass"
   ]
  }
 ],
 "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": "3.8.6"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "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()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}