ADS_intro_1.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "disabled-withdrawal",
   "metadata": {
    "tags": []
   },
   "source": [
    "##### Algorithms and Data Structures (Winter - Spring 2022)\n",
    "\n",
    "* [Table of Contents](ADS_TOC.ipynb)\n",
    "* <a href=\"https://colab.research.google.com/github/4dsolutions/elite_school/blob/master/ADS_intro_1.ipynb\"><img align=\"left\" src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open in Colab\" title=\"Open and Execute in Google Colaboratory\"></a>\n",
    "* [![nbviewer](https://raw.githubusercontent.com/jupyter/design/master/logos/Badges/nbviewer_badge.svg)](https://nbviewer.org/github/4dsolutions/elite_school/blob/master/ADS_intro_1.ipynb)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1a88a468-f08f-4e27-a685-1c0b7f66a67d",
   "metadata": {},
   "source": [
    "# Introduction\n",
    "\n",
    "A course in <i>Algorithms and Data Structures</i> does well to begin with some overview of scope.  What might we cover?  The short answer is:  everything.  \n",
    "\n",
    "Data structures include the periodic table, recipe books, tabular data, entire libraries, not just arrays, queues, stacks and heaps, trees and networks (graphs).  \n",
    "\n",
    "We may analyze some in terms of the others.  A pandas DataFrame is made from the pandas Series, in turn built atop the numpy array, which imitates native Python array concepts. \n",
    "\n",
    "Yes, the Python ecosystem is featured herein, with encouragment to use one's knowledge of Python as a stepping stone to other languages e.g. Julia, Javascript, Java and C/C++ (not a complete list -- I mention J in class a lot, and even Hoon, as uber-exotic yet super cool).\n",
    "\n",
    "Algorithms accompany all these data structures (graphs, linked lists, stacks and heaps), lending them their dynamism and their ultimate utility. How do we store and retrieve, update and delete?  How do we [sort](ADS_sandbox_7.ipynb), when the elements have greater than and lesser than relationships? What mathematical operations might we perform?  \n",
    "\n",
    "The idea of \"math objects\" obtains from the idea of core concepts having the attributes of both nouns and verbs, substance and action, a status quo (current state) and \"change ability\" (transformability).  Those hoping to stamp out OOP will of course be unhappy with its further incursion, but not on some either/or basis.\n",
    "\n",
    "If our structures represent geometric objects, we might want to rescale them, spin them, distort them, measure them, alter their properties.  Think of [Blender](https://blender.org), another prominent player within the Python ecosystem.\n",
    "\n",
    "Perhaps we're looking for routes on a map, involving minimum effort i.e. the [least weighty path](ADS_sandbox_6.ipynb) through a directed network, or the lightest spanning tree.\n",
    "\n",
    "In this implementation of Algorithms and Data Structures (ADS), we do not shy away from the data structures associated with \"array based computing\" meaning, in this case, we consider the namespaces of both numpy and pandas.  \n",
    "\n",
    "The idea here is to equip a 21st century high schooler with the electronic desktop publishing skills needed to produce professional quality tables and plots. The ability to produce Jupyter Notebooks in shared repositories is useful in many walks of life.\n",
    "\n",
    "On the other hand, at the middle and high school level, *Algorithms and Data Structures* is [likely to draw heavily from CP](https://www.geeksforgeeks.org/5-best-books-for-competitive-programming/) (\"competitive programming\").  We go there too.  The world of CP has evolved into a succinct summary of many computer science topics.  Whether a student is committed to a CP lifestyle or not, much value derives from working out, now and then, as if one were."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "central-obligation",
   "metadata": {
    "tags": []
   },
   "source": [
    "# Algorithms\n",
    "\n",
    "The contemporary term \"algorithm\" derives from the proper name of a Persian polymath on faculty with the House of Wisdom in Baghdad over a thousand years ago. Iraq had not yet been invented as a country back then.\n",
    " \n",
    "His name in Persian was: محمد بن موسی خوارزمی, which gets Latinized as Muḥammad ibn Mūsā al-Khwārizmī (c. 780 – c. 850), or Al Khwarizmi for short, which sounds a little bit like \"Al Gorithm\" (but not a lot).  \"Algorithm\" sounds a lot more like \"Al Gore\" (a joke, why?  See note [1] below).\n",
    "\n",
    "Note to teachers:  I recommend using this statue as a kind of \"home base\" for your version of *Algorithms and Data Structures*, presuming you have forked or are cloning this repo.\n",
    "\n",
    "<a data-flickr-embed=\"true\" href=\"https://www.flickr.com/photos/kirbyurner/51843248720/in/dateposted-public/\" title=\"Khwarizmi_Amirkabir_University_of_Technology\"><img src=\"https://live.staticflickr.com/65535/51843248720_4251c8dc12.jpg\" width=\"255\" height=\"500\" alt=\"Khwarizmi_Amirkabir_University_of_Technology\"></a><script async src=\"//embedr.flickr.com/assets/client-code.js\" charset=\"utf-8\"></script>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "third-heading",
   "metadata": {
    "tags": []
   },
   "source": [
    "But *what is* an \"algorithm\" anyway?\n",
    "\n",
    "An algorithm is a step-by-step recipe, not necessarily numerical. It's very closely related to the mathematical idea of \"a function\" i.e. a procedure for taking in a set of inputs, doing \"the same thing\" every time (i.e. doing the algorithm) and thereby generating the corresponding outputs.\n",
    "\n",
    "When you follow a recipe for baking a cake, you start with certain ingredients, measuring cups, other containers (data structures), an oven for adding heat over time (called \"baking\"), and you end up, one hopes, with a cake.\n",
    "\n",
    "In computing, we aim for deterministic outcomes, meaning the exact same ingredients will produce the exact same cake every time, \"as if a robot were doing it\".  Humans have been knowned to aim for this same kind of perfection.\n",
    "\n",
    "Allen Downey extracts one connotation of \"algorithm\" which is:  \"followed mindlessly by a machine\".  \n",
    "\n",
    "This connotation of \"mindlessness\" follows from \"deterministic\" in the sense of their being no room for alternative interpretations, leading to different outcomes.  \n",
    "\n",
    "At the level of computer languages, such precise specifications are finally meetable.  We are able to recreate our results with great or perfect  accuracy, when beginning from the same inputs.\n",
    "\n",
    "[Quoting from <i>Think Python</i>](https://greenteapress.com/thinkpython2/html/thinkpython2008.html#sec87), by Allen Downey:\n",
    "\n",
    "<blockquote>\n",
    "    One of the characteristics of algorithms is that they do not require any intelligence to carry out. They are mechanical processes where each step follows from the last according to a simple set of rules.<br />\n",
    "    \n",
    "    \n",
    "Executing algorithms is boring, but designing them is interesting, intellectually challenging, and a central part of computer science.<br />\n",
    "\n",
    "Some of the things that people do naturally, without difficulty or conscious thought, are the hardest to express algorithmically. Understanding natural language is a good example. We all do it, but so far no one has been able to explain how we do it, at least not in the form of an algorithm.\n",
    "   </blockquote>\n",
    "\n",
    "Clearly cooking is more creative and may involve making snap judgements. We might use the cooking metaphor more in connection with what, in machine learning, we call \"hyperparameters\".  These algorithms come with \"settings\" or \"knobs\" and need to be fine tuned. \n",
    "\n",
    "You might think of algorithms as \"what we fall back on\" and \"do by rote\".  You might also think of them as evolving.  Even if the primitive algorithms do not change, how they're combined may well be.\n",
    "\n",
    "An athlete has a playbook of maneuvers known \"by heart\" but may also always be learning new ones.  Algorithms and Data Structures offers similar opportunities:  to both learn by rote, and to grow.  \n",
    "\n",
    "What we mean by \"mindlessly\" need not mean \"recklessly\" or \"carelessly\" (although it could, depending on circumstances).  Think about how you would, and already do, use all these words in real life.  How you think and talk about algorithms and data structures is probably evolving too."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "accomplished-holly",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Algorithms in Science Fiction\n",
    "\n",
    "What is science fiction in one era, may become science fact in another.  Take the idea of a chess-playing automaton.  An automaton is a machine that appears to have agency, to act on its own, in response to stimuli, much as we humans do.  \n",
    "\n",
    "If a machine is so life-like we cannot tell it from a human being, at least from its conversation, we might say it \"passes the Turing Test\". \n",
    "\n",
    "Suppose we discover we've been fooled.  What we considered our mental peer is suddenly reduced to a machine in our eyes.  This change in perspective may be an experienced as \"uncanny\", as in \"weird\" and/or \"unsettling\".  \n",
    "\n",
    "Much science fiction (e.g. the movie *Artificial Intelligence* directed by Steven Speilberg) explores this \"uncanny valley\" dividing robot from not robot.  What if we cannot really tell, if X is a robot or a human?  The Swedish science fiction series *Real Humans* likewise explores uncanny valley quite a bit.\n",
    "\n",
    "Alan Turing helped establish contemporary computer science, and like many before and since, speculated as to the possibility of \"mechanized\" human intelligence. Speculators along these lines have created the field of Artificial Intelligence (AI). \n",
    "\n",
    "#### The Turk\n",
    "\n",
    "\n",
    "The first apparent chess-playing automaton was named [<i>The Turk</i>](https://interestingengineering.com/the-turk-fake-automaton-chess-player), and was actually a \"magic act\" in the sense of involving deception.  \n",
    "\n",
    "![The Turk](https://upload.wikimedia.org/wikipedia/commons/thumb/5/52/The_Turk.jpg/640px-The_Turk.jpg)\n",
    "<br/><i>[The Turk](https://commons.wikimedia.org/wiki/File:The_Turk.jpg) - Mechanical Turk or Automaton Chess Player (German: Schachtürke), exposed in Heinz Nixdorf MuseumsForum, Paderborn, Germany.</i>\n",
    "\n",
    "Spectators believed they were looking at an automaton, whereas in fact a midget inside, tucked in among the gearworks, actually played the game in real time, in very cramped circumstances.  Such a midget once beat Napoleon, whom we imagine was taken in by the hoax.  \n",
    "\n",
    "Today, however, game playing automatons are science fact, from [IBM's Deep Blue](https://www.ibm.com/ibm/history/ibm100/us/en/icons/deepblue/) and [Watson](https://www.ibm.com/watson) to [Google's AlphaGo](https://www.deepmind.com/research/highlighted-research/alphago).  Nor need these be expensive esoteric computers.  Chess playing software is by now ubiquitous.\n",
    "\n",
    "An automaton (robot) by Hanson Robotics, dubbed Sophia, stars in [short amusing skits](https://youtu.be/Ml9v3wHLuWI) and emulates facial expressions convincingly enough to appear human.  Sophia demonstrates what we imagine real Artificial Intelligence might be like i.e. a seemingly human intelligence powered by mindlessly executing algorithms.  \n",
    "\n",
    "Short of that so far unattainable goal, algorithms have gotten a lot better at speech recognition, and indeed pattern recognition of all sorts.  These dramatic improvements have come as a consequence of breakthroughs in the area of Machine Learning (ML).\n",
    "\n",
    "A few of the very many science fiction movies featuring AI:\n",
    "\n",
    "* [A.I. Artificial Intelligence](https://www.imdb.com/title/tt0212720/?ref_=fn_al_tt_4):  the Spielberg version of a production started by Stanley Kubrick.\n",
    "* [Her](https://www.imdb.com/title/tt1798709/?ref_=fn_al_tt_1): a man falls in love with an operating system, which then evolves beyond his control\n",
    "* [C.O.D.E Guardian](https://www.youtube.com/watch?v=PQ-UP2zieSI):  a World War 2 movie homage featuring life-like giant military robots (CG)\n",
    "* [Logan's Run](https://www.imdb.com/title/tt0074812/?ref_=fn_tt_tt_1): a futuristic city in which no one lives beyond age thirty, is policed by an AI computer\n",
    "* [Ex Machina](https://www.imdb.com/title/tt0470752/?ref_=fn_al_tt_1): a billionaire inventor genius creates life-like robots who plot to escape their prison"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "888e0b48-ddf6-435b-91b9-56602e19ba39",
   "metadata": {
    "tags": []
   },
   "source": [
    "### The Origins of Arithmetic Algorithms\n",
    "\n",
    "The idea of \"numerical recipes\" is at the basis of  arithmetic and is what we learn in school as addition, subtraction, multiplication and division. When school teachers talk about elementary school math, they talk about teaching algorithms, such as how to perform long division, or how to find a square root.\n",
    "\n",
    "Speaking of square roots and how to find them, another word attributed to Al Khwarizmi is \"surd\".  The surd, also called the radical sign, is the $\\sqrt{}$ symbol."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "6fe7c7d2-d4e7-489f-9213-5a685a6e5220",
   "metadata": {},
   "outputs": [],
   "source": [
    "rt2 = lambda x: pow(x, 1/2)\n",
    "rt3 = lambda x: pow(x, 1/3)\n",
    "rt5 = lambda x: pow(x, 1/5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "dece80c6-1225-4719-be56-2111b4b242ac",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.4142135623730951"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rt2(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "71b9c436-bb82-412f-90a0-6e763e5b337d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.618033988749895"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "phi = (1 + rt2(5))/2\n",
    "phi"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "413d175e-9488-4974-a7d6-998f06e83ffa",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "123.00000000000003"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rt5(123**5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "9fc69853-a8a8-434c-8cfa-259d7b064087",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2.0"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rt3(8)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2200099b-15dc-442e-8bdc-44e17d03caa3",
   "metadata": {
    "tags": []
   },
   "source": [
    "This idea of \"numerical recipes\" spread to the Italian peninsula and to the rest of Europe, through the Republic of Pisa, where the famous Leonardo Fibonacci (c. 1170 - c. 1250) wrote *Liber Abaci*, based on Al Khwarizmi's work, which he had been exposed to in North Africa.  \n",
    "\n",
    "Here's how storyteller Buckminster Fuller tells the tale in his [Synergetics 2](https://archive.org/details/synergetics2expl00full/page/n35/mode/2up?q=carthage) (1979):\n",
    "\n",
    "<blockquote>\n",
    "When the archaeologists' artifact-proven history of mathematics opens 4,000 years ago in Babylon and Mesopotamia, it is already a very sophisticated science. Mathematics may well have had its beginnings much earlier in India or Indochina, as it is an art and science that has traveled consistently westward. Over 3,000 years ago the Greeks made further magnificent contributions to geometry, algebra, and calculation. Then about 2,000 years ago the Roman Empire all but obliterated mathematics. A little more than 1,000 years ago Arabs and Hindus traveling through North Africa began to restore some of the ancient mathematics to the westward-evolving culture. When al-Khwarizmi's original a.d. 800 treatise on algebra was republished in Latin in Carthage in 1200, it required a further 200 years for his elucidation of the function of zero — the cipher — to be diffused into the university systems of Europe.</blockquote>   \n",
    "\n",
    "If you learned to use Arabic numerals, [Fibonacci of Pisa](https://mathshistory.st-andrews.ac.uk/Biographies/Fibonacci/) told his readers, you could get free of relatively clunky Roman numbers and the abacus.  You could learn to compute on paper or on a chalkboard.\n",
    "\n",
    "You might be thinking \"if algorithm means step-by-step number crunching then isn't any computer program an algorithm?\"  \n",
    "\n",
    "That's pretty close to how people use the word nowadays.  People talk about the \"recommendation algorithm\" used by online merchants to suggest other related products you might want to borrow or buy.  Many algorithms are \"semi-numerical\" meaning they involve more than just number crunching.\n",
    "\n",
    "Think of the globe as a data structure, storing information about the world, and of a projection from that globe, to a flat surface, as an algorithm.  \n",
    "\n",
    "We need data types and algorithms that mirror our practices around [spatial and temporal information](ADS_project_2.ipynb), in order to create data visualizations such as Google Earth.\n",
    "\n",
    "<a data-flickr-embed=\"true\" href=\"https://www.flickr.com/photos/kirbyurner/51978151490/in/dateposted-public/\" title=\"Spaceship Moon\"><img src=\"https://live.staticflickr.com/65535/51978151490_ee196a9d2c.jpg\" width=\"500\" height=\"246\" alt=\"Spaceship Moon\"></a><script async src=\"//embedr.flickr.com/assets/client-code.js\" charset=\"utf-8\"></script>\n",
    "\n",
    "*screenshot of MSG Sphere website, April 2022*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "animated-crime",
   "metadata": {},
   "outputs": [],
   "source": [
    "from IPython.display import YouTubeVideo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "terminal-album",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/jpeg": "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\n",
      "text/html": [
       "\n",
       "        <iframe\n",
       "            width=\"400\"\n",
       "            height=\"300\"\n",
       "            src=\"https://www.youtube.com/embed/oRkNaF0QvnI\"\n",
       "            frameborder=\"0\"\n",
       "            allowfullscreen\n",
       "            \n",
       "        ></iframe>\n",
       "        "
      ],
      "text/plain": [
       "<IPython.lib.display.YouTubeVideo at 0x7fa49844d760>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "YouTubeVideo(\"oRkNaF0QvnI\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "laughing-finder",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/jpeg": "/9j/4AAQSkZJRgABAQAAAQABAAD/2wCEAAUDBAkJCAkICAkICAkICAkJCQgJCQgICAgICAgICAgICAgIChAOCAgPCQgIDiEOEB0RHx8fCA4iJCIeJBAeExIBBQUFCAcIDwkJDxINEA0SEhISEhUVEhISFRUSEhISEhUVFRUSEhUSEhISFRUSEhYVFRIVFRUVFRUVFRUVEhUVFf/AABEIAWgB4AMBIgACEQEDEQH/xAAdAAABBAMBAQAAAAAAAAAAAAAABAUGBwIDCAEJ/8QAYhAAAQMCAwQDCQcMDQgKAwEAAgEDBAASBREiBhMhMgcxQhQjQVFSYWJygghxgZKTovAVJDNDU1SRobLC0dQJFxg2c3R1lLGzwdLTNDVjg7S14eIWJURVZIWjxPHyJoTDpP/EABwBAQABBQEBAAAAAAAAAAAAAAACAQMFBgcECP/EAD8RAAIBAwEDCQcCBQMDBQAAAAABAgMEESEFEjEGExRBUWFxgZEiMlKhscHRFUIzYnLh8BYjU4Ky8QclNDVD/9oADAMBAAIRAxEAPwDjKiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCiiigCip+nRTiH3WF8o9/gV490WTh63YXvC48S/D3nKq4BAaKnK9GM77rE+O9/g1gXRvM+6xPjvf4VUBCaKmqdHEz7rE+O9/hVknRnM+6xPjvf4NVwCEUVOV6Mpv3WJ8d7/ArNjounGVouRPjvf4NUBA6KsT9qLEPu8D5SR+r0J0SYh93hfKSP1eq4BXdFWUnQ1iXVvoHH/SSP1etLnRJiA/b4HD/AEkj9XqgK7oqw06JMQ+6wvlH/wDArEuief8AdoXyj/8Ag0BX1FT1eiud91hfKPf4NYr0WzvusL5R7/BoCCUVOv2sJ33WH8d//BpQnRLiH3WF8o//AIFAV7RVh/tSYh1b+B8pI/V6JHRLOAbifgfKSM1//wA9AV5RU6TownfdYnx3/wDBpUz0RYiX22EPrOPj/wDwoCu6KsZvogxAitF2EvpXyLfxx8/xUua6C8VL7bA+Vkfq1AVXRVpP9CGJBzScNT0d9JVfmxqSr0PYh93gL/rJH50dKAreirHToexH7rB+Ukfq9Zp0NYl93gcf9LIX+iPVcArWirNk9C2Jt8zsL48rLPyc1jZEqebOtC9D+Ifd4Hyr/wDgVQFc0VYhdEOIfdYXykj9XrD9qXEPusL5R7/AoCvqKsL9qXEPusL5R/8AwKP2psQ+7wvlJH6vQFe0VYKdE2IfdYXx3/8AArL9qXEPu8L5SR+r0BXlFWF+1JiH3WF8eR+r1sZ6IcRIhADiGZkIA2JyCNwzW0AbFGMyJSVERE8K0BXNFWxtB0DYzCPdyCiCSWiRXS90LqjccfenGQHXQ6l3aknBclXKkA9DeJ23byEg+UTjyJ+HcUBW9FT5zouli5u1l4dd6LkkkH37Yyr+Clg9Dk4hQwlYY4JeQ7JL/wBtlQk4tatFaUVZSdDWI/d4Hykj9XrBeh3EPu8D5SR+r0IlcUVYv7UOIfdoPykj9Xrz9qLEPu8D5V//AAKAryhEqxR6IMQ+7QflJH6vWbHQ9PIrSk4e36RuSrfjDGX8NUlJRWWVjFyeEVvlRVwB7n7FCTMZWGKi+Heyf1avf3PmK/fOG/KSv1WsW9u2MdHVh6mRWx7t8KcinqKuL9z3iv3zhnysr9Vo/c94r984Z8rK/Vap+vbP/wCaHqS/Rr3/AI5FPULVtTegTFWxuV/DyFObJ2StqeNfrbqpnd6I8QH7bC+A5H6vXvtryjcx3qMlNdqPFXtatB4qRcX3ld0VYidEOIfd4Hykj9XreHQziJfb8P8AlZH6tV8sFaUVY0jofxIOZyF628etX1V3FYB0SYgX2+B8o/8Aq9AV5RVir0P4h93gfKSP1esw6HsQLlfgfKvJ/Sx/RQFpnJK0ezp5R5i9Yuuk5GVeMxnXNQNuGPlC2Rj6twp5P08NBwX/ALg98kfzeFQlcUk8OS9UTVGo9Un6Go3K1pW5MOfL7Q98k7+itw4c79we+SP9FU6VR+JeqJcxU+F+jNAj6tb0bt1fQv71b2MMd094e+TO342WVOTOFG3qJtwi/gz/ACcvxr+iqdKpfEvVFOj1fhfoxBFgkWo9I+T9OWnEAER0W/S3V1el1r1/BSsIjnabc9Ww/ncOZfw+9Xow3O0Dgj2e9l80RSnSaPxL1Q6PU+F+jEBN3fT+9zZ/RKVg2IDcekvJ7X/LlW87mx0NPD6W7IjL1iy/opsejPl9reEf4M/0ZU6VS+JeqHR6nwv0YoSRcXN2riLteVWhfp6vpVsajOCNu6eL2T+L1eH8VbGIbvMTbnyZc34KdJo/EvVDmKvwv0NVtvNSV1aXvMu/c3vkz/RSI4rv3J75M/0VTpNH4l6odHqfC/RmoErQ8XZGtz8d/wC5PfJH+itsDDneYmnPkz/RTpNH4l6odHq/C/Rmlhu0fS/Jra2Vvo0pKEYj9ieL2D/RppKcF8uZp75M7f6KdJo/EvVDmKnwv0E70wi5f+als1i0Wx7QgN3lXHrIdXaQiy9mlGF4C444Iui4IkYiRWkIiJFqLwaUG5fBy08FH1EW6cLUX2sxEdXlZaRXqyyqvSqPxL1Q6PV+F+jGSHhhF6PlfTL6cacFjabbrR9mnDcuFdpcG3lEWytu9bq+HNcq1yN41b3srh7QiRn8YUtHxeenSqXxL1Q6PU+F+jMohC2NoCI+k5/zL4fzq8dkOHzuWjby8o29oRt/o9KkZuPl9oLVzbwD7XkiPEfxJ79Jlhulzi9b5Itnb6tuX6adJo/EvVDo9X4X6MVG40Oq4SIuUR1atPMX/H8daCcIuUbfp838NAxiHlac9oC/OTL4a8Vp37m98mX6P6Kp0qj8S9UOj1fhfozxU8rV7V1ZZ+T9PjVJMA2LddEXZbrMBgiEe+d9kHcukQjiqEJL6Vqp15KiLThiuzOGG8UWJMkx5LVo3zm78PknbdaM2OGcUk8JGipmOSKngr0mj8S9UOYqfC/QhBn6Re1XjrZDbeNtwiQ6uYS1CQ+C3+iluMbPy4zu6kNWl2SDKQBD5QOtKQmPw8LkzRK3uE64xut059biNo2FqEREC9tNK5J1oReKqdJo/EvVBUKnwv0GdEu7NvpVi9aP09XT/wAa3ux3e028Pqtl+SKZDWkozgiRE24PsmPtXZVJXNJ6KS9UHRqLXdfoJyX8n/6/NrYVYF9PzRrEF9r8mrxZPSrBSrag+VShiLcJF5Pa5e15NAJQT6dqnJjDD72eq0tQ2iV3MWrTy+Pr7NeErbbdg6i5iLyi7NvhtQabMS2j3RC1c4RHaIgPKIlaNxeSKCVUbwTjFyeEPGJv7pu43CNwdLYFcZX9kefT/wDNaFnE43bayYk0ROc5Wj2gITXwDb1Z/wBOUHxrFifEQIhG07bhEh+2md7o56yQCtRfEKJTYeOENrG8vtK3vY2j61wpnaurrSoc5nge2nQSXtCyFL3bhW7kSAi1ly69QneKopEgkOSJl8OeVKsLxQozzhlcZSLd40Nws3jp3okSKu9Ubc0VUz8VR3EpZb7fsjaRiQ6SEBE7BAtJceIkXHjzVrGTuxcAnWyIHW7i16StuMdXNkVv0Wqa40Lrw9GTmTj4ui5uT1NBcQ3WndcXk9pLR4ekteYZtS33MRGVzmoSPmtIbRtMR8d3X4KhKstNCLoyyLyt2LV3qiI5LcvpVqYkXEQs2g3eRctzpCQkNpHlndqLNF4aattORKKjFYSRZEHFBdZYMREG3d4NtoCbRjqG8gyuFUHwpnqRfCtIpONiJDYTboiRCYiXfRLl72HaHzrb5s/BHHcZFsSaBgjbHskWkgIdenLyeGadVbW4zRDqakbsxu0jqaK4dJGWSCSjcmpE5vDlU4zlwITowb3mTZp/lu/NrNxdV3lVGWcZdG64W3bbR0uDf2risFcrU0p7PV10ug4oTrjYlbcXNpIRG3mMSJVu97hw8XGrilHtPHK3miR4dirrHIWntAWoC9nyvOnGppgmIDJb3giQ2lYWkrb7FO0D7aoAkSp1omSrwVFWFYZCFwSfeImojRCLjo2kZmQ3DHjCXA5RDx8QpqXgmRLsNxLevO2jummcMxMWGGyKxgPqfKIrSLibpFqJ1eJLxXJERE1/a/Ju2v05Y3J/Evuusyezdt3Fm8Z3ofC/s+omqUVFcD2m0iMj0QI/JLskY5qpAXhLLgvvpUoQvCnHPteCuTbT2TXsKm5VXhJcH5nRLDaVG8hvU34rrRlTDimANkROi63HARuK8HzH2BZbMvKXqyTx0+5V7Vdk7WrbPq85T4fuj1NFNo7OpXlLclx6n1p/5xIKcGN/3lE+QxL9VrNuHGHT9Uo3yGJfqtZ7YYNu+/tD3tfsg+QXlD6C/i97qYsJjG+5ugtIrScuIhBpoAG43X3S4NMAOpTXgiCvvV2vZ20aN7RVWm+PFdafYzlt7ZVLSq6c1r29TXaiXYVhzTpbocQjHpIriaxAQAGxuN10zjILTQpxUlVESkMvC22n3WD+yR3nGi3ZFYRNGTREInxtW0lTq8HVTfiMxvcuRYpFuiG51+0gOcbeoSIC1NRRLiDK8cxQi45C2t2weIcVxD+UJn+1O17zyCGVEtLTqH0ub1a0iJXeT630zpUsgSG74w/nfTz1pTyf/rQoWH7mvZtjEYkwXHXQVmXwsstJDYazXWC6kIfB40qZ9IGyDWHtNOtm67e4oLfZp0ZplYCeSvX5qZegLEgbnnHDD/qUAuOw+5D3yyLgaZfR2Yb+ROyS6s0QURLERMkRVtHpdjbzDTP7i62f4S3a/NcWtI2ns+3uba4qKK5yDl7WudMP6aG0WF5XoV6MHJ7k1HTq19n6lc7G7JniAukDosi0ojcoqVxFmuSZKnVkP4Ur3bLZJ3DxaM3RdF1SHSKjaQoip1queaXfgqweiGJusNRwuG+cdcL3hXdj8Frefw1n0tw78NNzr3Jg5+Pdr+AXCX4Kxq5OUf0nnsPnNzfzl+PDPZoZB7bq/qPNZ/29/dxheHHHbqQno/2PaxBp11x11uxyxLLNWgTXO8F8adXnrDbDZVmHJiMNOvn3Sdp32aAIwBFTIU4rcXXn1VO+iaNZhoF4XTdNfjqA/NBKhW3Mze46A+Bl6K0PwGBr891U9mo19n21DZlKUoe3UcFnXPtPL+Wgo3tetf1YqT3IKenVosfXUfJ3RnGbZM0fkLYBlx3PgRV+51VyLXR2JNqTDgDxI2zQfWIFRKpdNgsT+9l+WY/v1DlPsPddPodJ8HvbuX2Yz8yfJ/aykp9JqLisbzS7c4H7ZPYFmXCakG++BOiWYhurEtMw03Aq9SeOoftZhaRJjsYCI0astI8syzbA1ztRE8NXRsLAdj4eyw+NjgIdw5iVtzhknEFVOpU6qqbpM/ztJ99r+qbqO39mUbXZlKpGG7N7qk9c5cW3nzK7Gv6te/qQlLegt5x7PeWMeQxQIxuugyHO6YgPvmSJq9FM8181WkPRbG++ZX/o/wCHTD0OYVvJRyiHTGDIf4VzNPxBf8dKsefjgNTY0NeaSDpeqoIhB8BZOJ7Nejk1se2la8/dpPfklHOe3C9WWdvbTr9J5q3bW5FuWPDL9EUxtngncMsmEIlCwTAyyzIF4askRM0NDT4KlGyOwTMqE1JN98FduuQN1bpMw4Zgq+Cnnplwu+M3LHmjHaX8G6oj+I7PgUqgWFbYTY7QsMuoIBdaO6bXrJSXiqKvWq1j69Cz2ZtGcbmDnTlHMEtcZa71ww0eyjVudoWMHQklOLxJvuT7nx0ZOv2rY33zK/8AR/w6P2rY33zK/wDR/wAOnjo2xZ+XDV6QV574wutENIoOXBERPCtR3pK2omQ5otR3UBtYwHaoAesnHkUriRV5QHhWyXFDZNG0jdypexLGMcdeGmTBUau0atw7aNT2lnr008iM9IWzTWHkwLbjp75DJb7NO7UMsrBTyqiyrTlj2OyJhNlJcQ92hIOQANt+V3KiZ8qU3Hy/BXN9o1qFS5bt1uQeMLs0Wet9ZvdjSq07dRrPM9cvzeOpdRa/7V8bLPfyurP7T/hVWmHYa7IfSOymZqRCPgQRTrU18CInHP8A+K6JDl9n+yqq6HUHu2Vdz7vT47d4l+XzK3na+w7bn7elTXNqo3vNdaST/wDBp+zNq3Co16km5uCjjPe2h2wvoxjiP1w+64fasybD8aKS+/mlb5XRlEIe9OSGi8dwGnwoQ/0ZU9bcR5xxxTDzQDQ83EzQTMMi4ARcEK63ry9/wLWkjHMZhFm+T4j/AKYBcbXzXoip8Va9d/T2dYPmp20pRwvbSyvXPHtLFnO9vPbhXSln3W8P0xgY9osM7lkuxiO/dEI325Z5ghpwzXLgVN1LscxIpT5yDQUJ225EzyuEADTmqrkuWdIq5ldThz0+a93ee74Z0+Rv1rGXNR5z3sLe8ca/MGxIiFB4kWkRTrIi4IlWnE6MGVbDePvi5YN9m6svt1ZXNquWfjqK9F+Fd0Yg2RDm3GTfF66aWh+Pq9hatjHsdbiORWy/7U/uvVS0sj+UVofbrduS+yaE7eVzdrMW1GOeHHGfNvBqfKDaNWNeNC3bTim3j1x5JZKh292aSA6Aipm06Fwmdud4laYLaiJwuBerw1G0q7OlTCu6MPNweeMu+H1RRUcT4ikvvilUolYXlLs5WV5uw0hP2o93U16/Uy2wb3pVrmesovD+z9Cz8L6NYzrDTqvyEV1sDJE3OWZgi5J3vq41CdscAKBJJrUYKl7Rl2w8+XC8S4cPMvDOrw2c/wAkj/xZn+rSmzbzAhnRFAbd6GtkvTy5c/JJOH4PFW47S5NUKtlm3juTSUljOumq8zVtn7drU7rFaTlBtp56u/y+hXHR9so1iAPK466G7MESyzjmirxvBfFTH0z7OBAZUGnDPes5qTtum1wU7ApwqwOhIVQJiElqi6AkK8CQkQsxIfAqUydP8Zt1xlp19uKJsEO/dvVoC3o27yxFJAVbRUkRcrs14IteOy2dQjs+hX3f9znIZl1/xEvpoeu6vajvatLee5uy06vcz9TnJNPMQ/TyuFeG6Q/3ac9pdlpkG03mL2HV71KYIZEJ8btJNSmswuXwISovmSmV5dVvkDb7XaroZpoobdL6fNra279NVJgWmrHsVJhxpoLRJ8XCvK4hGwmht05alvGqlYx3nhD/AIjtX3JGcibpkikXE2+5cJtCJMlIBu1NROCyynjRBJE51qA7X4++TxSSbECdAdIkRW7oRG0iNSW7Lxr2a9xXacnRufcucjgTQjbyiJkVgl5N5EvV2vNUeCSUkguu9IbrhEbiK4vJJEu41bep7oQUPHtHCFjthWkLh3ANtttw6R5hJMuF1JsSjNCJFcQX+iF1vaESDlLwcU6vDTXizIja6GsSEtREVol2yK3mJPB61aIm9l3XuboQPSZ6NPatHipCmlfw1LdWMknN5wbmXt0TZgZCLRNlaVuq0kK27tCvVTiTROCRkI3PERDdbqIi1F6VIXI4xtO9bduMrjAr7QG3k4aSIbs1yz0pXv1WIJLlw3EAELYlyt3aRMR9Xii/DVcFE8cRJIcEbiPVaJCPLbf5XVlw8CeilOGyLeq8DtIewRCJH2dN2QkXL15UwOOi4+QD9iK7UPlc3tZ06YeBNa2nLbeYrbj9Lm8Q/g4cKq1oRhLLJHElutP6SbuO60StMm/S1Ium65PHSp7aGY64UYystIdZFp8oSsyQfovVUcaRrfC7HcK50BG10bCDyi8Rkto+Lw/BvxQd+3vSctfHSI284kQ//HtLVndRf3mSbHMbEStNsTIC59N5W+kPNp6l4rSfC++3G6Qi09baRFdb2bryBSst60TyuHGofHIu974byErbRIhIh1W6uz4skzp2SYLjjRAJC0Wkhu1AQ6Cu46xuIVXqXV8FFHd4DfzxLlwvFnxZFohiS2I9zcZ11gDtFwd8VvhuW7NSXNVsTPqSnfZvFyJx/wCtMO04fiRaYgDy4fIK0i8lepU8KEtVpsztBuyJh3VZdYQ6hK490To+kKmCZdeRL4uE+2SmXC6Rai+pmKXFp1WwJdpae0o2rV/Ka04nhrU2pZXAzhYo4TgNBDgETxgxuhiAJO7wwEQ1KiXKduV3DPLPqraO0rsZ8WJTTANDy7hrdXAeoHw8BtEPFMutC4edij4i4042+0VjrRC42faAxLSQ3IqXJ1ovj96p1sX3ZOF98xYIY7py7nHWo29c7ojysTASsVXb4m+bUsltSWSJkji5eG9saN5SdKrHKfy70+pk7W6qW1RTpvDX+YfcOEd4TETAhITG4SHqIfHWyo+zjDTEl0JDUjD+6Dfkj3S27ZbYRgDQZmZukY270VVCV3PIckzkC/T9HmJOrKuNbc2JV2dV3XrB+7Lt7n3o6bsnatO9p5Wkl7y+/gzwwEhyLUJaSHyh8VRPbtNxGbBgWWBflHvRYaFq9ppqKUUHyFNYi8UokTPJVtVUVRRUllJ8ThNvsk07ymnwiXZIfSQuKVLYG2Z7Or54wekl3dq70U2xsuN7Rx+5e6/t4MqcSLdlbzWlb8Uqle3i/wDWeIFy/wDWEz/andXw1E8diORidaPmAStLyht0mPoqP4ONSbbHEh+qeJNmI/5ymD8WU9aX9ldro1Y1IKcHlSWU+45bUpypycZLDWjQ0iY8w+1WwZPZLs/TmrQ8g9n6e11FSe7+79Peq6WixuinaKS421IkPvSChyUbAnSvMWGhac3V5ajDN13K5VyRck4IiJ0xtTG3+HyAHjfGOzzlZcP40SuVOjBkghHd2pBF7O4ZHV8X+iuoejudvsNjkq5qAbovHm13v8aCi/DWlbMrqW0Lu1lwlr8sM2jaNFxsrevHjHT55X3EuNfWOBkHUTcUWf8AWGKNZ/GJVpRhK934OIlxV+KTRevYrZfORaj/AE2TrY7McftrhGXqNB1fHcFfZrb0LT74rrBdbLtw+o6lyfPRyr8b6P6n0L9nNbuO/j9CxKzl0DpX7ucznu4f9xKdno/c0BkHNO5jBf5iQM3Px51R0SUTuIA+XW5MA/jyBK34M8quHpLnozhr/lOjuR/1ukvmXL8FUvgv+VR/4yx/WhWF5V11Gvb20OEMP5pL5GW5OUW6NavLjJNfLLOhpT1jROF1ACkuXXkg58KhCdKUT7hK/Az/AItTLGV+tnv4Fz8ha5yGslyp23cWDpqi0t5POVnhg8HJ/ZVG9U3Vz7LWMPHHJ0Vs/iQy4zcgBIRcQshPK7SRBxtVU8FUz0m/52k++3/Ut1aPRkv/AFXG94/61yoHtBh3dW0RM9Yk60R/wQMNmf4URU98kq1yh37zZ1D4qkoesl+WXNiuNte1vhhGfopInnRxhfc2HtIQ5G6m+Px3O5Egl5xCwfZpk2k2VnP4kk1p1hBbJrdCROIQi1aVpWgqcTv+A6lu0GIpFiOyMkXdNqqD4FLLIA+ErU+Gq0TpQl5/YY/H+E/vV7dpVtn2lGlbXEmlBJx3c/t4N479TyWFK9ualSvRSbllPOOvite4tTE4gvMuMHxFwCBfeJMvw1zrNiq06bR8zRkC+sC2r8HCrw2C2hWfGJwxEHAcIDEc8upFFRzVV4iSfgWq+6XsM3U1HhTTJC5f4Rq0D/CG7X4FrFcrqMLyzhd0dVHr/lf4ZkeTdWdtdStqmjl9V+US/oZ/zeX8Zd/obqI9M3+cR/ijX9c/Ut6Gl/6vL+Mu/wBDdRHpl/ziP8Ua/rn6jtV/+wU/CBXZ/wD9zPxmQuvD5fgr2vD5S9WubQ95eJvkuDOlh5fZ/srneFiLsWTvmSsMDL3iG7UKp4QXwp+hK6HbXT7P9lVB0Xx2XZshp9AcE2j0GgqKlvg6kLwp18K6tyloyr1baEJbkm3iXY8Jo5zsGtGjSrznHfSUcrtWWiQYP0nskgjKZNkvLDvjfv5cCH3slqZYdOjzWb2SB5o9K8M084GJpmi8U4L46j8no5w8luFHW/MDpW/Pup8wTC40BggZybbRSMyIs9WSIpmReiI+LqrK7NhtGDcbx05wx7y4/ZY7TH38rGSUrZTjLPB8PuyoekjBQhzVFlMmngR0Q8AZqqGCebNM/h81RlakvSNjQTJpG1xabBGgLy8lJSMfMpFl8CL4aY8Lhk8+0yHW8YgPo5rkRe8g8fgrle1IUql9ONuvZcsRxw8u7J0PZ0507OMq3FRy88fPvwW30Q4XuYO/LnlHf590Ohsfe6y9utO3mykybJbdZdYAGQyC9TvEyLMz4CqdkPiVMFJuNH4ZC1Ha/ADYf2IlVd+2hL+4x/8A1P71dHvZWFlZ0rS5k0sLhnVxw29O/U0W0V5d3M7iglnPXjTOcLXu0LUaBVaQXbVJQtNOYFK3IkTNEzTr665+2iw4osp+P9yMkHzgWppfhBUq3ej3aop4vI6gAbJDwDPIgMdK8VVc7hP8VRjppwtBcZmCme8Tcn6w5m2vwpenwJXj5SwpX2z4XNHVQeU+vDeH81k9ewp1LO9lb1dHPTHVnivl9Sw9nv8AI4/8Xa/q0qI9Gm0N7j8B0luaN0mFXwtbxbg98VXNPMvo1LtnlTuOP1f5Oz/VBVDHLNiaT7RZG1IcIS84mXAvRVOCp4UJa9G2tqSsejVVw4SXamln8ljZVgrzn4Pjo4vseX/4L3wzCm2HpDrfDukgMx8G8FFEiH39K++ir4aqb3TKaB/ixeb7aHaq2tnMVblRm5DfUacR7QmPAwX0kVMqqv3RUfe2N3CF0Zy0iIRS5HBIRzLhqVBH2uHHKsntCVOVtCVLGHOm1jhrNP8AueGxU1Xkp8VCaee6LXyOfMNxh+NpjncBOg6bB64T5gVwb2Ife3svGaKviyqSA7g077K39QpZdtgTkYQ4XpReLsDMuCbtTFE4qngqKjEtHVpt5tPLbzDWLajcIjaNxCN7l1o6uYhBFW1OvqJfEi1sBiB7x7Z6TDJsntwbT1xMSoz4SIskfKadBeXzEgrp6qgXSMQ7i+8QcC4dIiZ2kQkQ+FRzIQ4+C2rdjEWBiMxgSxBx4N13c3YWBd9Ab2RFpVOW6lxJY8rCoo8QXrWittsUB/E3XyabaCQVwsRmhCOG7EWjJhi/TeQEVueXX4MkRnQvUY5kQRuSZOXBy8xEWrSI3e/y/BqSlk/EhJsRFsWrR1EJfZC7V3H0aV4+jbbm7aYERDSQ6rz7RCXHSXmReylNrrIlcJNWkQ3DcRXCFokQl5X9Omq4zqX2saCOBiHe7C3lpaSEbe1dy+0NSIbSJore9tMi3aPKRFeB3e1w9nzVG8Fwl/ekIgWnUP4bRLj6yLUmNBjCJFaNwDpLVrHTqH2Sqjki5SpvGonOK1cI3eiXo+twz89Ip0Td8zhEJabius9UbvEJdX9FPeEwzfIdOl24SO3V2SK0ezlp+Alr2XCEZYBqeEWt5zFaJF5Qiqe91pnxXqqm+T5vOo0Fhm6Ft0R0naLdo85XXFb7+nJKWbkge3Vt1t28IbcrrRIg6/B1e/71O0TD5jrljQCAhc7pG2xq7mIi5c+pOPaHx0yTHhalkwBau0XNaZcS8Xv8V7VVUskXDdMBYacIRIiaKOQkWm4SC7e6uryss08FeNMuOucxE5eQlcNpW+UIZ6Rz9/lrezMERcPd33AQ26tflXcc7V0p4KTJLc3e9G4SvERt7IlfcXq6QT2qY7Cg7LHIe8PFaQly+SJFzdXMvw8KcJmHkTFoEIuRBuEmiKwh5rXLeBEhWrn76VowttyWwQulcX2v8nSXXbzJ7NN2zuMlGImHhcNshIbRG64buYi6uHiqGrLiaXmOLTW8cbESFq8mx1cwkRCbpEYppJbAREXqTzrVjdHSDvJYgREIYXijRF2CMIE37H5Q22kq+FT81V/h5sPkQGJBdc4IuWlcQ9q0V5ktHh1ZDU+2IdbjE40doCWG4k2HNaTjkCUIjd5yMURPGXh66Rks68S3WTcPZFODwHJchthrSRcx9lpsdRul6Ijxy8K5InFUqa7XuCwUaNH0Nwt42IlquLTvTPwEZGToKvhQfF1OGxuFdwxifMbZJgLjnas1D3PH+OQGSe+nHd51GdoS78I+Q0I/GI3fyXRq4Y4bJ7z7BD3I44MaURF3NcLsUXftrT8d65oxTmucRM0zVEREp22Z2hIXe5HWGGBAnd441c0DZhz3hfuWguHLvaAmZpwVSpvBbm3Giu5d62Q8wOta7wt4iVglwHiqiCZ1G0eNonRIW3bxtdFy8gdbIwfArxUXBFSBpxCFRVUy60VUXwbR2fSvaEqVRaP1T6mu9Hrsrypa1VUhxXzXYy5KEpk2VxMTZYaO4HSj78QIiPvG9Npory43KICdq5ra60ua3Lk+Vw7aNhOyryoz4rr7V1NeJ1iyvIXVFVIcH8n1ryIz0hYNv4xuAPfWQMh8ZhaV4W9ovCnnzTw0w7aYhbLxeNaQ73F5jm9FwQuEZZ6Hw3aq8A23IiKGSmSrnwRLEqs9vMNJqWTuohlETtxatZFc8JeleV3vGnireeRW2M5tKj74fdfdGpcqdmYxcwXdL7P7MZAKsnyrBpfm1qJa6OaSWSuOYfE71EKS+3zE6/umifPSJOg0iruY6oKZCSqvBVVeOSbYPSDYJC0sloLuw/YhF6oHlcv6KuX3OeNfVHDZMl5qMjyYibbrrTAMrJc7khGsh8R0k+V6IpCgoticM883/pe20g7PYS/i01reNsq2AsNo0jr7rrggDTV+SZ8SJfMBL4KwFbk7Z1a0qzT3p6tptfRmXp7auYU1STW6uCaT+qOcJfSKJl3zfvElwiRvieXh03ktqVtwvpBFC73v495W3A6IXdajdYScM+HtUvH3aGCeHBsR/DC5fevTP3q6khIw62DzQtG26AuNmIiom2aCYGK5cqoqLVn/AErYb29iWe3eefXJd/X7vG7lY7N2OPoczltC9IcKOiSpLomIixecgjMmjNLEFS46TTP+hM8m7aLHncNdFqbDkxHbbxFwgG623iBiqieXDUCqiL580roTpY2xj7PYPKxp+OTzcTcITTO7F0+6JLMUbSPJOBPIvHwCtVX0U+6fwzaDGImDM4ZNZdl7+x18o5thuIz0orkElXiLCpw8aUlyVsJPLUm/6n+RDlDdwWE1jwX4IZ+2WRab5nFOUpC22+lcXL9PNWodrmN2R2FoG624NQ3W+9XW/czf3MPij+ive5m/uYfFT9FKnJWwn7yk/GTf3EeUF3D3Wo+CX4OQ2elJAG0DlgidgZFiJcvgQSy49dYB0mAJ73OSJrzOi+O8UeA5X3Iq9ScFVOpKuDpj6e9ntnnCiyC7tnDzQYYNOutXJcPdLpqLcfsraqqWRIuWS51Sj3u2Gris2ZEhu0qWJgJKOfBSFMOVBJU8CKuXjWpf6YstFiWnD2paeGpT9dudfd9rj7MfwWPPeedwcsXivS8Ti3Bc0Iy97pVd8Rsu8LGlRFUxuRetF051XibfsfcnPjh+mrG6HfdT4Vjk+NhRQZsCXLXdtfYZEW8QI7N8CiacBLJbcuHgq+322gEjMWgEUIiIkAUEU4qREvBBROKqtRnyUsJvMoyl4yb+5Wnyhu4LEWo+S/ByEz0ni1pb7paz67HxDPT4ciTPxVlL6TRMe+d0uClxa370TxrrJbffqWdJPutsAw91yNhsVzGnWitJ1rdRYBKK5ELcoxM3creYQUV4ZKqVEsL92tEJwRl7OGyz2jYnNSXB9VpyG0JfGSq/6Wsd3dxLHZvPHpkp+v3Wd7TPburP0HaRtdKhvNwyj4lGee3ZNx7zZcd357ppW2s0vvNLUVEXNUypLP2veWb3G9EmnOvbaGM6Wcm8xEmm7DW5FIXRVEXLnz8NdA9FfSVg20cfunCnwdJq3exnBRuZEUuRHWF4oi5FkY3CtpZKuXDDpo2/jbNYYWLyIrkkBkNNEDG6F25+4ULM8kyS38dHyWsXHdaljs3pY9Mhbeuk95Yz24X4I9H6LpJAJG+0BqAqQWqdhW6gv4IWS8M6p6XtaDTjjTrTjTjRk242RBeBgVhgQ55XISEnXlp66cv3beEf90Yn8pF/v0qwb3ZOzrrojKw7EoomWp/dRZAB6RoDqHb6qEvmq1/pHZ3wP1f5Ln+pL74l6L8DSvScf3SZ/OD/AL1Iv+nTV127cuErucLrvKuz0r4fHXUuxmP4disRvEMMeYlxnuR5se12gcAhQ2nUz4gaIqeFKqjpu90Ph2zWKJhcrDJMo1itSd6x3OIWuk6KDa4qLcm7L8NXZ8l7GfvKT8ZSf3LcdvXUfdwvCMfwQSL0vPN6RelW+d4T/LVaTYl0oq//AJQslwfJN8VD4meX4qtLD+nPD3dkpO1o4dISNFkCwUX637oJSlR4twlnZbdIFfeFarb92jgn/cs/40L+9V2fJ61nHdlvtdjnLHpktR2zXjLeSjnt3Y5+g1N7dsF2C+OH6aUw9s83Whjsv70ytasMEMjPQIiqKmWeeWeaddLP3aOC/wDcs/40L+9SrC/djYO++ywGDzxJ51toVIomSK4aAi8Cz7VeZckdnJ5UX6v8nofKS8aw5L0X4NO1W3qxpDsM5EiTu+9u2vnut4Q62tZa7epeHXmngWmL9sBj7k58o1+mrb6eOn3Dtl8Raw+Xh0mU5IhtzBdY7nQEE35DFhb1UW9FjkufppUn6B+kqFtRAfxCJDciNx5hRCB9GVMjBll5TTdZpbk+KfAtTqclbCfvKT8ZN/cjDlBdw91qPhFfgoGL0lNtFm1v2yVNRA6IZ+iVipdW57pMR0cnVkmmfUb96eiuREqV0R0y7axdnsJdxeRFKS0y6wBNNI0hrv3RaFUU+HBSzquehr3R2GbR4s3hEXDJUdx1p53ev9zKCCwF6paCqua9VVXJaxUd1KWOzeePTJR7eum957ue3dWfoQBOlBUEURyUgonAUkKiIPiTV1Vo/wCnLRat25x7RGGr8dWn06+6Aw3ZfEWcNlYdJlOPwm5iOMdziAg6/IYQC3iotyFHJfaStHQp7ovCdo8U+pTEF+E+Ud19opPc6g6TFimy3ulVd5ulcP3mipPktYz0kpPxk39ysNv3cPdaj5L8FexOkFAHJtZLaEudoP2Jny55ASJny8aTY/tl3UyQGrpkSWiTrt9uoS7a5immuvkjB5AfFH9Fc17be60wXD8Rm4d9TJcruKU7GJ9ruVGnXGDJpwgQyzsvEkRfDln4auUuTdnTlFxUvYaa9p4ynlaZwW6m27mec49pNP2VnVYeuCv2cLdmCL7Not3bp91wwZjg6IjaV5Kl5kGqwLiVRPJF4U74ZsYwQ3OvvPD2jashRR8r66lgZu+LLditXj0KdLkHaPDJmJsQ3ojGHum24D26cMrGAkGYC1mnKScPGlUttH7tbDwctw/ApEtlLu+SpTMI/Na00zIS1ePWqVn8mIHCPimH4U2+/CYEzFhzembstWXWhArgfjq8jcsObgYimfHKuUsSdJ8nQ3gjYZEIiVtw322iQ+IrkyTq/p606PfdgYLOkBHxXDnMH3p2DJ3rU6IF3BFkGjTRtAvlIJInhyTNUhnu8ib+qmH7oRtXCgMXAFCC05Uq0hVOC55Z5+HhUk1jBcpPU58wyCQkF/K1c4Q3XEQWc9vWQiZAuXpU2q1bJIiuMXRuIiJbiu5tXtUulOE2LTpaCs0m24QCQkVuq0CXscU9GluwmGFOfEjId20Y+jeN2rT5/wA6qS0R66cVJqKJpsns67iDxOlpbFq0rezaWnl5i8Hs1J5WxAsQnDICtEt5qG0iERLmHs6SJfwVZvRvhTTTAtNNiI6jL0tV2ou1Uvn4QEkbD5btQ23CQ+SQ9qsfKTbMtHdhocv4Pss+6IvgTkWM804Rbsd6bQ26HjZHU6BdS28UQkXwUx4N32faQ84kPeyJoTEdIkJPJqHTlxyz4+PKuu4Ow0FsR7wNwhuxuIu9jbaIiOekU8VQed0bOOt9w9zNi60dzUxsRu7m3okQjaqFegkWnJUVRTjmtSjKS4lHGEloytYGIi6/9RYl2/xO5ojttAYwhvStdyRTu3XEskREFETxrGse2MF+W600Q78SsLmLfkFvII8xWkOrqRMlWr8k9DwxCF+B398bS375d+bIbtIW5d6ITJFFMlTrTqpmf6Kn5kjv0YWm7fsvdLwkwdoiRNg0ALIFwAaRUVRTNrNc7sklvST0G7CSy2jmqdhz7BF5IHa4QlcIkOghEx5iUrvNnSiNixNjytkJabSEdOm6zVxtXSvHhq8FWr02bHN4Qy0wyThNvlaTrghYIgN42AOQtGiiKIg5cPGtVR9UhJkd62JOjq8rVyiWnmLTV6E3JHkq01GWjFeHYu42Lpi3bZp7JAJXaREe14lrfg0YnWyJ0WybEbnCuEBuIit0553JqpnYAnGyHlu1W2lbd6Xk8fCtLsCUd4IulyjaIjq/+vweep4RZy8m3DcTYF4QFsicA7R8km7SuIi+LxqxMEhsS5cQSJ4BExK0RIAtHVrdy06gFERF4qdVNIQm3iK0iIit06h3Y3EJFbwtUS66s/YxwiiC11E8BAGo7hdMdDQ28rq6Cu60sLLjxqcIJyRCc2osuvaB7S21/rT9a0haEvSsIjz8KPJUIxpe/u+iW7+SEWvzKmZRSfcI++DqucAgIREbhGwDHO7K4RTNA8FQ9zDZJERExIuIri707zFqLwUMcIGnLSEh5gISH1hK4fnDSdrB23ZZQwc3TbTrtrrhARBhwkb5ulwFDNmOJO9XFLupARKdCwyT2Y0j5J3+5Sh3AcQbF9+PDmm7KGJEYsjSCXd9zxX5TokLa2qp7llFXgqOyE42rQGjBVu7uxUhFi24WGhKOZRmmu52mWRF0DUARkmWLgTNU4XN5opy/DJgvsg6PaTl8ku0Be8WaVFNrtmnx3DUXCpegO+SmG5Dwv32nujaCGCAbRkaZqrqqhgikqt8HfYvD5bEQd/Dltb2XLuNyNIAQEI8AmlcvBEACI3kQl4Ko5eCtO5YbI6Tb89Be3T18Y9a8uKNm5N7S5itzUn7NT5Pq9eA+007WYX3TGcbH7IOtr+EHs+0Nw+1TtXlcptbmVvVjVi9YNNeR0K4oRrU5U5cJLDKWttH0i1FSYiqRbfQNxLK3SL47wfaLWPwHcvvElRtFrvljdQuqEKseE0n/byZx+6t5W9WVN8U8f54nVfuSP8AM0vLwYs7/sGH1QX7Irtvvp8HAGjWyEHdssRttKVIGyMBeETbj3l701OvwXr7l2Y3G2fxCS+QtNR8QkvumXKDTWHwDM/VQAJfgr5+dIW0rmNY1NxN4rCxCa46m8XSw0Z2sgSiirY0yjYZpmuTXhr0lg92j2KlQsLwrFXk7xjQyyj8CQh7hkdznfn5VwknjQq709w1tx9VNlmorpZycGc7hPyijIN8I8k6gRld0n8VWqk90rtVsnO2Nh4RhOKsPycE7j7iaFuQJvg013E8BG6yiamnSdXimaspx8C1/wC4R22+pu04QXStjY00UQruADKDN2Ea+USmhsonjlUB1f7tr94uM/8Al/8AveBXGfuKP39YN/5h/ujEK7M921+8XGf/AC//AHvArjP3FH7+sG/8w/3RiFAfTOqn91T0hns9s1JmRysmSTGFCLh3uRIEyV5EJFRSbZbecTNFTMBReurYrln9kgAvqDhZfa0xfIvXKFKs/EJ0Bx30b7HTtosYaw2IW8kzDM3ZD5GQgI5uyJMlziWScVVeKqqoiZqSV2VgXuLcBbZEZs7FpT9qXutHFis3dpW2VYcIEXxERVTv7HY8A7UyhPJCPBJItZ9o0m4e4Qj6W7A194Fq5/dqjtTv8I/6MfVy3cze6/qUswQuvi7nf9yrldlvcrvSy8NAZ7H+5RjYNj2G4xhmJSHGoMi92LOBs3DAgcBSakxxBEJL87VHj40pq/ZCekJ6HAh4DFdJosT3j80gLIihM97ajl4d068RKuWXCNl1EqLzn0W9Ku0T+O4Qw/jmNOtPYvh7brR4hLIDA5rImBirmRASEqKK8FRVRanv7IwyY7TwXFz3Z4IwIL4Lgmz7099LwX2koCEe5f6Ez2qmv715yJh8EWykvtiKvOOPX7qPHvS1DVAMlJULK1OGpKvfpN9xpBDDnn8BlzynMNEYRphx3mphAiluRJphtWXj6kJVVM8kVERVJHH9jefbXBMVaS3ehiomadrdOxGRaVfRUmnsvVWuqlWgPk10PbcyMAxqJiscnPrd0RktCVvdMMyRJMYxzRCuDil3UQgvWKKncHu8Xhc2LJ0FzA50EwVOohNTUV+ESzrgTbGQ07iM51jLcuzZLjVvVujfMm8vNYqV3B7rJow6NMPB37IAYILnj3gxxE/x50Bzf7kPo8w/aLHn8PxQXTYbwx+SItOqye9bkRGh1Ci8LXj4e94qm/uvPc/4bs3h0XFcKfl2OzghuxZJg8mbseQ+DrLgiJDl3MaKK3Z3oqKlq50R0ebcYngcsp2DyViSTZJgnUajyM2DNszCyQ2Y8TabXPLPT56VdIfSXjWObv6sYg/NFjPdNkjTTQEXAjFhgAC/LhdlnlwoC9v2OjaF9vHJ2GXksWVhpySZ7IyYr8cG3RTqFVakOiqp16c+VKZP2Qn994fyRE/rZdXh7hvocPCYhY/NJopWKxmxiNtGDwsYc7u5CEToZiTrpC0WQqqIjQ8cyVBo/wDZCf33h/JET+tl0BcPuXtiWMf6NncHkuvR2peIP3usWb0dxMjSRt3iKPEmRTinUq0w9JnuRsIwzBcTxNrEcUddgYfKltg53Juzcjsm6Inayi2Ko5LkqVZPuA/3mtfyhN/LCrF90H+9TaD+RMQ/2R2gPmL0bYE3iWM4ZhrpG21iGIxIjjgW7wG5EhtoybuRUvQT4ZovvV2thnuMsGYfafDE8WUmXW3REu5MlVs0MUXvHVmNcfdAP769nv5dwz/bWa+sFAcC/sjv75sP/kJj/eGJVY37H7tLh8PZ2c1NnQojhY084LcmSxHMg7hw8bxB00VQzFUz6uC+Kq5/ZHf3zYf/ACEx/vDEqqHo56Hcfx6M5MweB3Yw0+sY3O6oEe18QbdULJT4Euh0FzRMtXmWgOyfdq7V4bJ2PmsRcQw+S6UiEotMS477pIMtolUQbcVVyFM+qucvcFfvzjfxKd/UVDdueg3aXBoTmI4nhvcsRogBx7uzDnsidMW203ceSZrmZInBKmXuCv35xv4lO/qKAff2Rr99ML+QI3+8MTqhuj7ad/CMUg4rH+ywZLb6DdbvREu+skWS5A40pgq5LwNavn9ka/fTC/kCN/vDE6gu0HR/vNhMK2jjtpdHnzYOIEI6iadkEsJ818kHb21Xj/lDadmgO8ulvpJYw3ZSTtDGcExdgg5h59aOvzgAYJW9ob3QNU8QF4q+WTrhEqkSqREtykXElUuKqqrxVV8dT3aXpQmTNmsM2bdz3GFy5D4uXKqutuIncjRZ/cd9NRPBk62iZWcX7pl2B+o2zmypOhZMxUMUxCTnziL6YV3IyufEbI1iqK9ROuUB0P8Ase8Pf7MY0xdZv8Qdavyut3uHsBdlwuyzzyqa4D7krZNhhG5EaXPcy1SH5kho1XzBENttE8SZL8NRX9jd/wAw4p/K/wD7KPT37ubpJn4PhcSDhZOMSMXN8XJTV29Yixha3otGPFp0zkNIhpxRBPLJVRUA4t6dtnIOFbR4nhmGOrIhw5O7aNTFwhLdNm9HIx4ETLxOsrnx70ufGru2j2cl4hs1sxvbRkhggCJlzdxjKnDAH0g3Pc6+ZM8uum33KXucvq8A4xixk1hYSCFqK2tr082DteRw/wDs8RDEm1VNS2miW5IddGdOMABnsNBaw1HwpLRAREGwaORugEEVLRUhBtMuq9OFSi8PIycBYvAlsP7iUJAQDqAuYQIiL+ki41cnRDhgkwJCPa8ns+lTd024GRTSlNbl2wG2HdV9hNG7fYQrafErV68reC099D2Fuubt1o7XAJwWy1COtog5BW0htMutFyXJetEqVSUZQzkyFCM6c9VxWSYPbSkAk2xcRW6bSsARG643SHWZLpyAVHhxVezUaldJGMxvsRRga8l9uO6ZW+STqXj+GnPbTAJjQuWiTrjx/ZbSG0TLPUXtZcKjeO9HEke5txEcm3NC5J3TrTDxv36gInQJWmUAckyTtKueaV5aTi3xMhUg8cMkw2I6VnXysmt9zuEXe5kZ03Y4l2RkwnXD714VJpRVPEvVV17IY6T5OCY2uNWjpK4SuASubMeYFuzRfChItULiHRzusOiSzaKPMvdF9hq4j3QmW5IzAFHeoAiinlx4KqddS/oqB0psRoH3rQZfG4isMmt60bROtFndkYmCEK/alTghJmqJJ6MjGLcOHcXrKm7psiMezdVVbUdK3cJCTxMtAQ3WkJuvGXaFplpMyEV03EooqivHwVJelyVJaZjCDrlpymmy0h2zESvLK61err6ySqif2IblvSXZ77jpWG43GbcGOTpiFwANuoGk0iiKqrkKdXVVIe1LiUUPZbYbR9MODYq33NNiYiYEWoxjRRt8ogApWYknrJUcLZjDCIzj3SmCC8b2u5prAjquOPvDCQ0g9atkqonHLrqEbK7Mb/EWots0Cee3Z2tkyLAkLNhNEbhKZCZvZoWSZMivbVAsLCm5mHzSwqUQyBAu9viOoh5hMS7Qp4k4pV2rFJZRSkm3jXPeM+J9HNrPdjRXDui70WYpaZJraIUJF6k4Kna4KnhrOShNWlc4WrltERG3Vzdf9ldZPwxHDBAR3pC064XK0LFpFYFxZ3io2rwTJLvBXM20cQd5oISuNwdN1olzauCL5XHKqUost1Wmb9jIrkyfGjOiTrZnraErDISHXcYcQBEG5VVcsh89dRRcCgsRu5mWGRYaNsuUrjMSuuv5g0CYoiKiqhLxXiiU57nFgm5Msnm3rTZHvo2kFwGGhu5e0J3ePSniq8M2LftnN/ortI+/mI6/ht81X5LGhi6s8sZ8XYYajSXR3jQg0ThENrvKBiAABmK8XTDrJc1EOrKq5AG3PsTjLpeSJWO+qLL6CZl6iFVmbSSILUJ85fdJtGTTG6bACMydd3ukt4KCYhHdNM1yzBM6hDWzWFYk/JLD3MRIWhEm4rUYSLWDpA0N6qpkO6PNfDaS55caiWRmKOW83W7Le3CO6ttO8uULC43LcPD0qx2jx3dYnfCNsRgD3Iwe7ZeEhaK592x5CAxOTvTTNOCbtOylTvCJ2Aw8OdYApcrEYrBd93pSI8NwhfIHoYk6sUpDLosBe3mNzwKikiKqRCdtDhRYc3FFvERd3UZh1/dQLjjQXZb8cB77pNSlNCpcFVIDXFc1oSIviuIFMfdflWkUgriLUQgQ6QsvUlFoeq1OCJwREyTJQ7c3AjWja4GJzrdOoTGLg4jaIpzIQ9VbIrGHnduhxh2wHHXLWohWNNDcZla5pAR61Xh1eNK2vbSRmIzbUBuSMkHnzGZJ3QmwL7UVku5gZVUGRbFBEeVUUUJckuVHAjKKksMRbjqWZCcIgG8bHbbXW7hLdOjwdaKxVQTA8xUetFFUXJUVKUVX3RLiP2WKXaTetj6toOj+QuXmWrBrhW3bDoV3On1ZzHweq9OB1vZF50q2jPrxh+K0ZGOkjD97EvHmjld/qy0uj+SXsVW8I2x1EG91cpEQh6VwjkRfhT4auuQ0LgEBcpiTZeqQ5F+KqUdjE2440fM0ZNl7BEJfk1vfIa+5yhKg+MHleD/DNR5WWm5VjVX71h+K/sS3bvbj6n9H2IRWjyfxbGyw9u0tXcowsPemOZF1tk2G6XLP/Kk+Cjvc/wDRPI2qxF6DHfGIMeKUl2S40ToD31poGrRJNZK5mnmAvFV77Gjsi/A3G1ESTNdZnyHYm6cngDTEmPBac0xJLY3k5F8KKuQD71WL0a7abC7PlIXBoc2Es1GhfW2bKVwWN7uk+u5Dm7RN6fAcs7kz6ky3tmpFa/uH5n/fkX+Zu/49cxYzCk4Ni7rF27mYTiBgjo8LZMGQqA6F3gvbQk+CvqBsz0pYZiAGUIn3nW+Kxd1ZLMB5jZaNU7oFPCjakvmqn9tsK2AxGbJxHEsKluy5Dn1y5ficcidbAWtTDMwAA0EERUREXPNV4qq1QDl7pPaZvF+i+TirOSBOi4Q/bndYZ4rh+9ZVfKbdvBfOC1wx0Vbav4Di0XGIjbLz8Pf2NP3qyW/jPRSv3ZiXAHiXgqcRSu5IWP7FFgpbMtQcQPCnTu+p6fVBxSLugJfe3u6VeEe6AE8hJEzz8aosQnbL9GDX2TCJif8A7GNLb5i+v+BaV4eamAQP92xj3/duC/EnfrVWrsviMvpK2JxVqa1CiTGsQJuCTAuiwEmLHiyWSdV03CG8n3WlJOoXVyTNOMbTBuizq+pE35fGv1+ry6I29n8Iw4xwiHJwqFIfKWLcrupXZJk200chlqW849urGmRuyEVuHLNS4gfOLD5WJ7OYwLoo7h+KYZI5XB1tOW2kJguYuNG0ap4UIXfChV0fA927OFgRkYJEdft1Oty3mWSLyhjk0ZCnmuX366I24wnZnaJtfqrh6StwCkj5tmxJaERV2xuVHcF0c7CKxFyWxc0yqtD9z7sPpPcYognxEe6nbeZQt4rddcPUq56kXqVKA426H/3xYJ/LWG/7cxXc3u3Oih/HcJanYe2T2IYQrpiw2Kk7KhuiPdDLQDxdfEmgcQeKraaJmpIiyDo46MtkMNdadw3Cm1k3CQSZIuzXmnRttMDluGkdxFJOLdvGrQlY4y2Nx3223Z28C4CWnNUVevqoD5e9C/SjiWy88pmH2ELo7qVDfQ9xJbEs7TEVRQdAs1Q04pcvWhKJWr0n+68xXFMPdgQobGFd1ATT8kH3JEjcmNpgwRACMKQ5ip5EqIS5Ki5KnTG2/RtsntBJN2XhRFLXU5Kjb2E6XLc7INlwAcyQR1vIqp4PDVUSOj/o0gySAwn4hul1Wvyno12rSLjBNo7y9YqSL4140BQXuaOiaRtLjDLatmmGxXQdxGSolut0BISxAcTL64dTQiJxRCIuKDXXvu+k/wDw1zL/ALwhflOU87OdMOzEFhuHh7DsRgPsbEaDumhuXUVgKiXKvFS8PWtJukDpB2VxqIUDFmJcqKToOk1bIj98a5D3kZ4D4XdWdAc1/sdqf/lcn+Q5X+14fT7+yAdF/ck1raSIGTGIELE4RHS3ObDvL/iEXmQyXJE1MKq8XKtTo9xDYPAZZTsIgy4kg2CYJ27EJFzDhtGYWSpRjztAuaJmlvnqR7adLeyuKwncOxNqXKiyEHesEw6F1hiYEhtOCYEhgioQqi6aAqj9j/6V7t5srNc4pvJOFERdn7JKgj72t9E4/bvAiJVefshP77w/kiJ/Wy6t3Z+H0bwpLE6Fhs9mTGdB9h0JOMXNugtwlks5ULziqKipnnmirTp0h4nsBjUvu7F4E2XJ3LbG9zxJnvTakoBZFlAHC9eOWfGgHv3Af7zWv5Qm/lhVi+6D/eptB/ImIf7I7UF6N+kXZXCIzeF4LGnx2Dkd7jC1KkKUiSQpkhyXjK4itTLPKnvpB6VtniGbg+Id0yGnWTjTGmmnbCbfaydZ3zJiQlYdqqKpkuaZ5ouQHz56Af317Pfy7hn+2s19YK5Rw7DujfDZrEpjDJjMmDIYktO7/FnN0+0QPsHa5OUDyIRXIkVFtyXPqq9/2zMOCJ3dL7ow9gvsRTmu53ZH8Wi5q89w45iOSp1Z0Bx5+yO/vmw/+QmP94YlVufscP72sQ/l17/d+HVv6R8b2Dx6S3MxiDNmPtRxYBy7EI9rAuuuiFkWUALredW5Uz1deSJk7dHXSBsdgEZyHg8abDYdfKS43ZLkXPkANEd8p8yTQyCZIqJp6uK0As93Z+8md/GYP+2NVyx7gr9+cb+JTv6iupNvOlTZPGYR4dirU2VFdVsza3Uhm4mjEwW+O8BpkaIuSLUX2Cm9H2DzRxHCsPmxZTYG2Lt2JvaXQycHdyJZguY+NKAqD9ka/fTC/kCN/vDE6un3I+zbGK9HJ4ZKHNic7iTB8MyC91bHQRe2B2mnnBKw6RMb2Bx6WE3GIM2XJajjGB3/AKyj2sA466AWRZQDwN91c1TPV15ImU36OdrtmsIw1qLhDEqLBI33G2iSQ8t5OqLq3yXjPiYrwVaAo7ZP3FslqdFdn4nCkRGpDTkmO208hvsNmhmyKnwS8RtzXxrXv7JYlp7OInDIMU4e1hldKO9K2GCuS9052iVu4XhnqQV1cCyyXLz1Cek2RshtAsZcahyZvcm93GqbH3e/3W9/ySS3fnuWubPLLh1rQEM/Y3f8w4p/K/8A7KPXvuyOnrFdn8RiYVhCMME7CGY/KdaSQZC6++w000B6AFO53VUlQs7xyytW6edG+M7K4DHejYLGkxGXnd+6H13IudsEL85T5kOgETJFRNNIulB3ZHaEWhxiI/IKNduHUF6O+2h84b1l0VIFtztLNPDlnxoBy9yX0pS9pcFdl4g003KiTCiOOMCoNSEFll4HbFz3R5PWqKKqaUVMrrUTdPLgpiLQk44Au4eLblrQOlb3UboW3uD1m0OfVlYnXmqI47B7V7OYNCbw/CmHokRoiIWxaM1Iz1G66664pumuXMSquQonUiIkR6YMbanS40iPvN2sIRzMbc1F6Rnpz8BXIvqr5lWqKFL7VYc20UkjtFp511wdIgNplpLdBpDPrtTgnUmdSzoVYaYZYEvsl5Ofmjd7PGte0GCuSbXxETYZZb3+oezdps6yz83gpugE429oK24tP5VeVPGYmyNKcIyz1LPoX+9AaksEwYif93skPpD1U2rsM+PJJtH/AEjd5fGvS74ax2FxRwWREzbMvauH2hqUP4o5bpt9YhIrfi5XVbjGL1ZDenF4iR9Nk902Tjz5ENpXEVto6eyIpTHsXgbbU/fiJERgLQ3cwtCREI29nMzIl9ZPFS7GsXIiEXXS1FaN1ohd6IDwuTV150v2SbEnnC3l9pDaVVjq8IvOMlFykLOkPCRkttjaJEybb7V33Vorx9XO3LP0qY0wA3xF1oiHTcLg6vZcDykKprj3N6VNGGTHGHCtESG7vgcvtB5OfXktVlpLBai3u+yMa4JOu1iLvpC4Y+jqAw8Pvr4KS7Q7GiQtSTEd40ekdNpXabdKZ1YzOMtlytPe0A/lXrSeeZOajGwQK4Q7RF5Re94E/wCFJRXaR5+eeGCutqMJd7gliRfaSKweURt5PGWmobsVsvDw+EM6VG7rlmNzbTbQuugLhiACN3ACUiFLlVKtmQN5EBcp83q9qm/CIm4ZcjHrEBJy8hG87TuC7q8kaZfBFIYzvPqI3Httu3AxyPU402Vwifa1CiXF41y6863OlpEfR9LtEXlL4Rt6q1Z05JCEbXZREAEI7sBtR6TaIiRBwyaYuEu+knvISotewwU5b0m+0h/SDAckxmAEm2WBkOOyZjtwxYwsNCAXmKKpvH3aaCyCERKK5Itq21/im0YtMuwcM3jMZ4bZcly0ZeJaSAhdEVUYsS0zRGAVc0JbyJckGQdN2LE7IiMDaDTMQnG2Au3TTjsiQBFaSre6oMNZuFmq3deWSJWplqqpAeMOW2NMPywjRvadkBLH5sA09pabSSlxLu4A/wDippXf/ox2t1b7/wBVHePo023UAqjJ3t0riHQLem/mN0DtIhNLhUWj4EhJpXhmiEKdPSrai959d30ftQepcP8AlHgXj4U4ItaFoVHHZ/ERYlsP8ogY39nQWh35hFV21z6dXbslM38KM71kTQiXrhoP8JAS1zjl5Z6U7hd8X9V9zduSFzrOi+6S+jHWqu6Q44tTyLl34C6Pk3W2H7VwZ+1VoZ1COlmP3th3yTJov9YN4/1RfGrA8j7rmdoRj1TTj918zMcpbfnLOUuuDT+z+RAcRZteIez2e1oLUPLl13eb+ytcFrtF2eX+9/xpwxoRc3dgiNhE0VufKBXCZEWSkVhAmeSZ2p46yYaG27sjaNvslq6v7eP4a7KcxHTZ9shtIdJCQldqEh3dpDaQ5Wki2rw451Opm0bUtnd4w05IG0RHEWCAMTaEbbd4R6JoJbyu8dS5FTpgHRndEbdOTY6Y3WboSAbtVplnndpHUnV56j2K4A6xcMgd0IjzEWgx7RgfUQc39qJUIzi+DLs6M4ayQ/8AR9DjYa9vwksTWpWmNKbuB0bOdqTHd1xXeVbVzRU4oq+B/wBvNnImNt6y3T4fYpIW83kOj2w/GngWqMxfFt24JCIkQ2k56AdgPXW65fCi5J1otWdhrUtiS5GeLcOR3bHBu7Q9oS7QqNq5147jfpy309DMWEaNek4S95fPvFew2yeH4bJHdMFNnNFvH8QmNiLMUhHmw+JqEMisydc3hZlwRLkSpCcVx24yJx0nSHU79ldPUO9dLNbbRIkEc8hTPLwkTtFRoxEx1OAOofutokI3emNxKi+BSXquVaXbPwSdcvIRtDl5RHVdcZ+VkhdfgXx5ZV66VRTWUYm4oSoz3X5EbYC18mh1CDMvUI27wiiSNVpdnWKJw6qUYPF3okwRaiLeN2iXMI3GOnjaQCPnzabTw077VNiE1q0h1wpYkdvkxZBePxW+bIUyypoacFu3dFaQuiVw83aLTcmnK3P2fDnVwsD/ALOMOb9oOUQId4Nw6SK620esi055Jx0qnDKt/STtbh8Ed1IcJ5+64YjFu9u16XXSzCKK73inElTJURcqje3cl2JgsmVENyOb70MBcaK022nScKQy26OoMurMVFchFc0zqoBZEtRDvSIy0cpGQkN3MmfEfPx4+GhQVbdbW4hiY7o/reGXLBjCSMF5JyTJbpJ8vFxVTMUyRM6iYQLtJfFHm5rh1Fw7XwW1MsSmNtNiA7u3V2R5iK4CIc0uLT158OHXlUIxiaQuCQaLeXsldbbdaPrCv04gKZDLbWnTdbqL6f0eMaSyMQ1aRHs/3dPHyRypAT12rm+n/wBq1kXL8XyuX6FQGDxlzfTmrX9Pp8aszjuFyiReyRURmdWryqAVw2bRuLmLs/TtUpZb1ctxFpERG4ri7Ij2juLgleqfZHUREIiI6iIitEREe0Xgy4rS9760bK0h7rISFx0SEu5r9JNNcdT1vBTTq4onhKgLk2Z2VYaYiTAaZaxIIAiL5DeDT7rNt5tCaCbqXEil15EqZ+FK4e6O32CJ/GJcfD2LrhkukciRM7RnDhBk7KzUuJkgJ4c1SnMuld8GGouFMDE3Udtrup2x6V3poGy3AqKNR+okzyMlTjmOdqQfEd++ROvG686fO664brp9rW6eal2uuhIluNbVsQZBDg8RsHxIbsXnCEqaJWDacOOWbMDLysjVULivhqDYnMdkvOPyHXJD587rpGZlzWiRmqrampETqTwZU6463dJIbdRE3q9GwOz1/DSdqIPN5Xo+t5VANatkXxawCORcvaH/AJqeijF4OyP53zqw3OofWL0R0/oQaERqjwiIrbSIvJpyj4Vy3D+cXN2rc/plSp18W/zbea38XvcaRv4kfKOn5xe1/wAc6AUnh7TXOVtv5vo9n3qm2zzrZQGHQ5WDlt2ly70HWHRH0h+uwVU8KCtVcRlU06Ope8Zk4f2iMZjA9oyYAwltBxzIlZIHfegH40oB4IvK1Xaru0RFqIvWWva8SvaAKKxVa3pEc5l0D5TmgS9IbuJ+yirUQa05fh/NKnTDY1zZBu96+VrjQ8wNc32e9bO+CQkgLnmoN9aGoljEaaFkn/8AKCB1ttsLSFo3SEyG4S1OtII5qmQ8bUXgWdbcZw6Y1IcaLeG4w9bc1qtdcK4Dta4g6ZEJIRIirci8c0WpAaJvMRi4RuEPfDG4RLybSLUeXjVE8WXhWOuXCWq7m5vyql2KRt2V27ca+1kDgkJA+2AE6GrVbrA0zy4OonHLNUW0DRdwMOWlb3U+N43WammSEHOGW9W0lTjnkPVxzq3OKepkrS5ksU3wHjo+xXvlhait+bUqlY3vXCYHS2H2Qh/IuqutkztlxOyLpE0Xrcw/hp52secw1iS6Y3A07vCL0brhIi7I6hTOvLLiZVfMdNv8LcmQrY9wkBCTZN3aSHTb8PjqqcB2gxPCn7HRfO4uUhI7vVIU1Ut2e6ZXZLhBFKIAgTYkTrrTQjvDsHUfhUrU4J2qnzOK4oUfusmIUhobtTToEQ2nYWrJEuuu4Z1VQZeVVY3W4+o2Jj+KzBEwdKEIly7sXTL+EuXIPeTNfPVjYA3czrK53tOFpuL1eyPmqFT9thw236pxG44ndbcTWrTdptNer3qcIe3mGy274r7Yl5FwXezauqqbhGpLK0xjuwTZqRbpKtTsyo9gsl11wrtTVl2rsldy3U4hUUzyzF8NSIi9IbfJK2kWPyNQsNCJOHaG6ESIyHsjp1ES9SIlJCV83yFkrBABvMitaaEu26ZcB7KJ1qq8ERVVErS/iQt3BHuIj0uzC0POjpuFr73aVOHhVU61yWxPVCnjVmNrXOjjHwZtGM0wXZkP3fYisNmLdzK7kqDKdTyUW1F5s+VGqXJJ1wnTIicMriIiIi+MXHh1J5sqI9t1wkQkIkXaG0hDTqBF7VqdSJ1cU60VNx7rRER3h23doNREIkAhwutHNRXhlmqJl1XjwFK9J0jeYrJHsgLDVvkk1EZAx+VF1ffJaitmqnTadXHX35xDa1NlSXWjuG0xdedPRavKPUuXUvBclypmUu12eb2foJUBIMSi94hhcIj3O46Rem7LkB7/ABZYjr1U0vMD2SEvmlSvaRwhkboi1Mx4jBei61CjtSA+CQLudI443UBlKZcFtobdNjjntE6YkXOvYaa45AvUmSoiESPMhqVyomm0ytJoGwtL0AESG0wFQzIS4KmadWa9axqSo3eqVAalSrQ6I5N0R1r7k9p9UwEk/GJrVZgOki8nm9If+H06qnHRA7a9Jb8tps/iGQ//ANa1rldQ53Z038G7L54+hnuTlXcvo/zZXy/JY3gqPdIsfeYc7lzATbg+yYiXzSKpEtINomrokkfKju/NAiT8Y1yTZlbmrqnPsmn8zo1/T5y3nDti/oVK0hOEQ+WQl6o2226u0to+fQlbJBiOkebVdy/ncRFOv369Ae15I29n1SG3s9r8XDwUMsE68Ij2iuu9q33v7Px13844W5sjiGJtQo054C7hkXCD+kguadVkhdFMya1gqJdki5cM6lk92NiDPczvb5SHmaPsmBdVyF8C9S1NOh9tgtn40Z3dGAhKFwCt+x92SOJD5NvHP4aqjFoDuGz3QISGM66RRDLUO6IrgaIuoTQfB/xrwVqbh7cDPWlxGuuaqrXt7f7lWbX7CS4bxA7qYuJwZhXbkh7TrpDmoHqysyVVUkREJVRFujpXwp0Z784B71vt074+yW91Llcl2VqZcBTrVFVXCOrU5kokhsXWnRt1fNL0STx1MnC37hNm2JkQlcNpETtw2mfWgiHKiquSp5s0zvU5qtFxZ5KlOVpVU48P80IdsjCEdRld9PJqZQRK0m2hG3UQlpARIriIfSJS45da9XUi5V9gcm14mN43uAdJsHxcEmCa5hEHjVAO0eGaL1iqVYbGIxnWRaiOCfKRODmQCYrpvMckPPqXLhkS5VZtsxbzouB6NoNVUnHV8URzGGyIry1kLMy7lK0TiSCErR4CK+ZVTWnvUzYWyTj2rtWEVpaRuIRLlRbRtu6vB46fYqXOOMOjaW5faIivEd6UczIuvSCtXZLx4AfHimSfAMZFhxvT3wrR03DaPkWlmRFqyVVzXT1+BMgYQc9t4zT8QcI3thZtG66ArzCNw2gKjqJFFcuGXnqm9uMFLCCt3m931xNO26iusvEhzyEkIc8lz4F79TuHL3sl+STnfCfK7s8pWj5vgqM9N88hZiGTbbvfXG++CRDqASu607Il4erOvHSruVRx6tcGVuLKNOgp9emfMq518iEtRXCXNdyiNw+z7yfopnxBPp878z+ipKk8iu7xGEbb9TZdq0fL1arqTHMHm3ES7ybC9n7Z/T+mvYYkYHB03Vga9m220va+MXreDJKcJWK6rdxE09ndF6QkXP6NYHi+nTGiXfwR/wB+gEKlSyC3q0+qIiNxERFptHruXqyTrryNiX/honlfYi/xKdcNxAiGSQtsNEDIk2bbZCYkciK0RCRKtpKDpjmmSpfwWgNkgu5htArpJ3CRjqGNpsJpoh4b7sqadSEqJxzKmnK7T2foIj4fK/CNZOjyj6P/ADfFurcwxdq7P/Lq/N48KFWEQO1b2vp826nYVuG0fJ9LtCVxeOk7Sj7Ijbbyjd/dS6tLmIjdbzXFqLTdzdnqQeHg6qFB3x0LZLg2kZkI9kbbrB1XddqdeS0jbjlcNxEfo9gdXk/8OFSrHsPaabGcbrY90Ae4aK/fumxY0Q6UVLUuuzXJMsk4qqZN0DdOtldOhRXCLU2+OIEd1vX9aw3BsW7JOOeleHUqgIrrdI6vp9FrfHjEQ8vMX5X/ADU6bO4Bv57UMX2SKQ7um32xdJnUPOIuiBkPvoK0mx8hiPEwEluVutLpNC6DQPiRi8zc7lvbSD7IPBUJMs6ASP4c327fp+P8FIZUEbeW30i5f6V1fgqT4rs1iDUQZRxnwS+SLok0oow1HCKYvuGS8pb406ky3Cr7yZjALsK+qr0xmOG+djNgbbrrrskGBfaYbsz59WoskRBVVXqqIIa9F+nKJadPn+D0qIqG04D7LjjRgYuNvjpscbISEhuTUSF4P+KV666RauX+8JVqed7P0toCx8MxRucNzTbJSbbnIlpCRF2nYZAolIaXrVpVIh49aJemtJI/cmfjSC//ALZVXA+1p+NcPL6uqpLE2rd5ZbYzf9KTm6m8o/8AagRUe9Z8XVy6lSgJF3afZIQ/gwBovjgiF+Fa0EV2otRF2u1WiNjEF37ZJj+i6wDrQ/6+O4pn79iU4JGaLkmQiut0k47H5uzdNaaS7zZ8KkB8jy2G4TjDsa9wXWCF9t0mHmjMCdd5kNu9BEWs7c9C5r1KOGGq2682HdZbt3vRty72rRdZKJcDrSuARAyeVxq1wHLgnV5Kw111uSbW7dEpv2h+PI1ELxW94dK0vDb19fCmiXBfa+ytPND5TrRtflolAS3EJ29fdcnxCGNIduJ9gRdADETELDaXckREZKooSLm6q55ZiUI23bbARsGwQHvoXEZAe+etEiJNbosmwKqmWSiSZJxpXDkG0W8aMgLlubIgL1bhVFt81K8UmC+zc803IJ3eNvmTZNOiRcjovR7BMjEuG9Rxc2TVeCpRd5chLdkmiMsyLGxdHmadbdG3VbaQ3fNq3Zm7mMleIm1IYJtwStISAh7V1UTg7hCL7B8wCTf5vteOrB6Lsd3kYQK4iAbfV3ekvPy15akMcDOqWcMpbEOhUWMQdsITiHqYI7lsLtNOe94F45p56f4nQzcNndZAB6iba3ohd5Vl6IReerQxYSae3lu9aLVb+UPrJTpA2ow8RtISAht0k0F46uyRJq+FFpCehkowhu5gvEpLajoPi7m8Jb5OjzEdphbd4s7vx059BfQgLUtZmIWvC0Yk0HEU4ahIhXrVV45eC2rbWXGfc+t23DLyj5fiiifgqUYezumbe1SUi1cKnFZS9owFBBshEbbi+n6K1GdMrktwnCt1W83rfTjWTrhFpHtE2N1wjqddBoBu6uJmKZqqInWuSZrUYQbZjak8RbMJc5x3vQ/YxIiEGx5i5d6YjxM7eFy55JwTJOFJ1jlqIrQEeYjIAER8or1RRHz0y7RbSOMF3GTBATVwub0gI23N66Bd6aNQMkIc8iIkW3Lhmq1D8XffctJ10nm7u9l9qEu0ItCiCyXjHJPHxTJV9hhHqTV/HGB3gtEMpy0REGtN5EV1oOuolxaOwh53ZeFKj8ra2W4JiBCwJiI96uEx3doiQuktwlaOSqmXh4JnTK0y4TOgHDEndVomQ96DTdot4C8XhzS7jkipm9bNYeMl8RlWk2IkROCRm7ptC0moqG67aRtKpZcEzVc0G2hQjOLvE2y240I7oTLvese5X3byI4wiaC007ca5ImSFciZJalNjT3dktgXmxO91pi250RNs3uQrDRRFSMkVRVFyLgqZJk/ymmxcdau3rBE40RDaRONXaSG7JCNLRNFyRMwFckpgwNgo08RLUUXul/TykUOK9LaMeq4C3IEi+FDSgHfHHMLkszpzXeZzs9xwY1x7pxh+Q66T7RF2rXgGzjkjJrklyKjLhrYkQjyXEI3XCNt2kiuJUERS7PNVRE8KpTMDu70+TUq6Mou/m7w+WOG9t/0twi0XwXEvvgleS+vI2tCdaXCCz+F5vQ9NpbSuK0aceLeB0XZ6Y+Jm0LbQu3F325q0SK7SAISjp4cUSozj+y8yNrdbE2+060RGA+vciEI+dUq58qDESEhIbhIbSEuUhLgol5l6q5nT5cXaq70lHc+FJ8PHPE3qfJS35vCct7t7/wAFCC382pb0Zpu5/rsuelbqAvzKa9oIgxpb7QFaIHpHyQMBdAfgE8s/Rpb0fyhLEWA7Vr/xdyZW/NzrfNq1I3GzKk1wnTcl4YyjUtnwlRv4QlxU0n64LVWtcoLmzHygIfjDW2iuGRlutM6zJZWCu9pNlHWIwyopt4hBvt7uiCZgHozGSTewHbbEscREzJERVrHZgRaJsnRvsISc0ly83i8Gn4viWtexmNPxHClRHHI7g6dJaXB5bHQLMXQ9E0JM8+HCp/CmQ8QbtMWcIk8xOtAJYZJIvuoZE5AItOpu4US5VRM6+jTiUXhpkyi7TxnWxsdbEhtttIezaQ8q07bQ4g3iEYWHg0kQ8o3C47dptIEzaK4uteCqf4KZ2x2JdwWNExW9vdSDcF9gX2nbS3zqNOw7DXuhkmUA1tTgh5rlxRJlsVju9bHdam7eyWqvA5SpPD1TM9zVO5hv09JL5G8HSwx8YcohQyHeNjvBJ0Wy5RdAeIH7/XUkmNNYlGciukQi6NtwkQl83mHxiuaL4agu1fRa5Jcbfwx9lkiMnHAfJ24TdMSM23ciIsy45Fx8+VWLh+CjEbbG68xEbj7V3aL41W5xUWp03p9CdOTqRcKq17e0j2z3R0MRndSpPdoCd7YE2QA2PZGy9Uuqc4Xu2xFsBEBEbREREaZ5Mgi9mk7UwhL2qjKrmWWXI0cQ3VwRIcTHvbpiN5dzugVo6zA2jGwS8pCIVRPxdWUAwxsnScfLdtRmiFx+S+W4jseVe86iIJahS3ivVwWp7AMu1VWdOmzEuWISYr7jzTIldhwloaIrs3ozI5IRqRcVyVfOqcK9lGqvdfkYm5tnrKPmMcXHRbluFdcw664TblpCJCRkQENyJ1jauS5U/wC2JtSYBaRdsJtxvtEJDdnbbmtyhciInX1VH9k4RFEaAx12DcJD2re0PZqWP4O4OHPkGhwBbcbt02k0YHdpTwW5+evNSSjV82Za6lvWqzx3UVLMUiEbB5tPKRadRDzdm3j+Ckiwt0N2knB9rmt5bvEPXw8XV1VKsZiEI3DovHeDpERaLTvQES1WAZCiKvHIR8yrGnY5fbXLRu5dQl6XgyLO0eKf/GTNYIjPaISuLtFddq1ej4x5cq0oWrtafWpzxMdJctvN+SQ/h1ceNNsZPm/m/lUBubTs/Tm0/m054M73mSP/AIcfK7M2F+cRUgGOThCI6tP5Pa9Xlp4w4BYGT23Bij6o/XsL/lXx6aEjSyFxatPN2fjW+Vy04xU03Dd5I+z6Pa5KYGXiu1cvk/N0/Grc7KImxHlu/vf81CIpxRwuW0h1atJDy6vF6q+9SBt36eyVYPFp+nq/m1gyn5X/AC0BNdrccJwWI1lncQujvbiLe91bl3UFiIFtuXWWfXwpPg+1MuM3ZHdIBIiK2xk9faK51sl7I+Hs1r2sw91h5wnmJDV5btvetOtDe20wRiJGiIRiJhmnHJDTPrSmhiO45qASMR8lsiH5qaffoCV4BtQ6xibGImPdDjT4uk2RC1eVpDzACoPwJWt99pxh9wt2Lrr0lweXeiJyMINoR8NqAU9EXxb1fHTAguXCFpbwrRELSuIiIbRtyzIlu4J4bqVzsPfabvdYfaC5xsTNp0BJ0CISaEiDInUIDRRzVUUV8S1EHruIXMtxrW7GXX3W9Ou+SEcHbiJfJitZdWXHrzpwfxwzw5jCt22O6xA5gvm+LVxOx9yTRC6gg0Kdd6lxUssk61z2i2JmRBbLTIIj3RhGbedNiTcQmyVreux0d1eK5KYuCmatqtIcXwl2MW6d3G/IxaJpuXEdeArCIbmWXFcDPTkqoiah8aZga9oIW63ZCUZR3MQbWpMWR3/uRkpFwsumQ9+3uaqmWY5J1pTTYXa0jb9LbuYveqTQNmXSbkvukIFFZN/cEJi6bTH+UEImCfYiIM+C5ITi5ru8lYZLguF+bd8bV9PeoDHIStEbvWtt+cS/D9OO9uIPkl2fIHV5Po1ky2I9m38r6fTitSgMDtwxvE7tJytxuLdVo7210jzysV1h0LcusOvjUgNMSII2lcN3MOkiLSduofPqXLxDx68q8myx0kJXdotVxXDqIS4adV3D0k4r1rIcR2adbjQ3xudfxAyaahi3e6TYtCYH1qlxIbRoKJwQxVVTwMr2AyWLTlME02R233ATThDqtB0FITztLNRzytyXjmlAPOKxS+pU4fLx1q277mMeeI+biPH2s6S7NtOtaWHX2beY2DdatLTcNwGmpBL+jxLU1DZ51/AylEUbdOkxMu3hWAEOJIivCREme9TQqj15kSdaZrFml7zeRDqHm0iIiN3Z4+Aff/ooBRiW0lvODEge0ToCLpetIj2umXnJS86LRg2KRH3N0DpRykWtuRHyaMnG7wK6K9kIPOiQiSNkjZKooiXKq1EJW8IisFx0iLvTTbZGRlykdg5qWSD1J56XbObBTHC7pltuRGwucHetkLr9g32A0WVoaeJLkiXcM1ySmOslDV4EmJja8Tgai5S5tQj+dWjYbHu5MRtPS28V3a0l2qWTktcKmjF4O9G4dLg8pVbi4vRmelB40Og8PcYfERJsfmlS5rZWG5qJvmLyi5q5+2W27fh2sSB5O2XaG7mEvPVhQ+lJu267yfSK4qtOk4sRm8YTLLawliNyAI2+tTTj08Rbc7JW6fjW+Lm5cvHUPxDpPattHUVvzqjEvaN2Xq1bsC9LUXZHq8Gr36qqfWReX4kvgvDylqcLmHtXF2fnU74tALuB1sNLp2kRjd3rdkLpERCuYtAAEql4LVWmvYjC3Ps72ki+xiXMI+l79SLaRHe5HdzEcnnZ/koOjHddG7XunSVEvQCLIVXJeKcc8qpGSUkWq0HuNdZUuMd9jMSS+yNWxHS9G0zhEXlGrQSG/eiD4VWm2E4QkVpW3AftWiRD+AuKL4FFFqf4Hgf1Sw438KavveaF2HMbBmVDkxglb+PJA1FBdQpA5EPBb+pEW6opiEd+M441KiNx3N05pcatuGwtQlnkY+dFyr1OLRh3o8MbHXN4LV5EegiK4ryuvPw8V6rOC8evwKlSHB40lhhp0YxEGJOk007rDXpEBIDC0xS0XhJMxXjmqqHe27GJQy5dz+5iCWkiFt2xuwCHWIoZlm6NqqiLlwXLhbT7izzTXeopMyLTEWHWGwGyMTTTDzx7pxwrzjtMtoTyof2XNBtRKoUIkStXEPfC1W33B8YQyW4fDzcfGnXSfGIe7ZfmCQ2lA7mF0dQ73uuI0XMmklgvG2grkuTRLxTr3uRCEiuJsLSK65wbh1cpAOZiXwVumPixhxEbhE27PYEm22BfB0W48s3WTGQ4GhSKOVyIqorQKiouRCBW5jq5r6lXRpiLcaWW9IQbda3VxdkyMCaIvR0mir4OtaYdot22+Qx94DRNRnWwJzekAyYrMiy+wbxEnSRFVM8hTNVXNV0NL3nV9tf081veA1du0iRJA8yKqXLkqXKheW9tIXVCdKXCax/fy4nptbiVvVjUjxTydA1g+6LYkZkIiA3ERaRER5iKqewvamZGEQB25seVtwRdEfJtIkuEfNnlWvFdoJMnS86RD2QERAPijld8Oa6q5vT5C3HO4lOO52rOceGOJvE+VtHm8xjLe7NMZ8c8DTtLP7rkuvjdaZ6bvIERACt8qwR4e/S3o3ZL6qsF5Iv/AOzmP51R6RI8m2pL0WFvMTH0WHS/JD86t52rCNDZtSEeEabivDGEals+cqt9CcuLmm/XJbtFFa5BWtmXiAi+KOdcKjHLwdbbwslbbOs6fJbHm9IvJ8Hve14OuleJ4mIN3do/sQCVpf8AKXN4l8FJ2nxG1rlbAbi9Iv8Aj1e1Udmzd45dq8kbuyPpCPL4/otfRxxAe9mcDk4k/wBzR92TlhE444VoNARah0p4TIuGXFSVad5mDYhgL4kNpCXklcwflWEWSgforUX2f2odw14n2Ru07tzl5dJEWnmyK34KtzY/bLu5u6Ru3WzC0g0kJCXlCVWKtXc0nHMX/mhlbOg5Q36csSXp4NdjHfo92mKZ6JANxB2vJqeNldz1EdjdkY0ae3OiuEDZAYuRi1CN46d2WedqKPUufv1OHo+q6vFuxWdx5RkpVHL31h9fZnuGySzbdTO7puK3tVKDb7XlU2zYY2kPlVZkisKuDHDMUbtG4rfWp3R8THdmIl6WmoxHY5hKt2HqLfLcPtFbUk2JKL1PJOACL+89Lm8ofJKncgAW92VtpCQ+yQ21qem6dOqkEhxwmyIea0rR9K3SNXoU0uBZqzcl7XAqyDh/2UCc5nScaK67lEAMRuXSZNDlkiLxaby5uEYBoX5O6usG5zeOWkW7baAzMhHNLsgAsurl8Fb+kSRJhvNMOtbmSyYg7qAnQPQ6BaMwIlGwkJOu0PNWRYyHcz4nynHdfw8REbGH5QHCnxLstLS90Gearn9aM+WtZE15kW2hgvhuyNpxpt66wi7dlhFq8sUdFVReORjwyJKZFcFv0rfi+T+aS1JcbltdxNRgJt1xqQ67v2hkAJNPtM370ZABfIWzK5E4I1lmqZZRB8dX09UvyS+N+GpQXMYk422QiVtxc3atLTp+b9OKKcLLvcv+KD/t8KmcNXN5JU6YafeZn8VH50+FQCVo+b1f+atzZXEPkiWn5vN7VJELSXrD9PnVuaL/AJfjF/doDa+X0+d+dWUdbfWt/OrU65bzai06fJ02/m1qNzm+bQFsdLs0sTxjEjCWTsaK0DsYdZsiCQQN0GgI0Rq51osyROKlmueXGL7I46LDbgk7jDV53W4fiQwGi0iNzoFGc3p+C7NOGSZU3zSklNdYib4jkCLe6aG8jDuW50bR427ojVV8CXKuSJnTYydo/nfOqIJY3iDbmKxpJOSSbGXGInZkkZTwg0bRETsgWwQhEQJerggonHrqWvbXNOFi+/k75r6tYW/EAiIhcixsVJ18mB8hIzQZ9WaZddVcZF8Yv7v5vH6cd0aG44RNtNuPuWmdjYkZWgBmZWDxK0AIly6kFV8FAWZtltI43/0laMm4s52bEbaJhyaLslgJZ2kG9kOCDXchgSI0jaKMjqXPhp6Q9p3XZEG51t6JKaw1/fuCBqLsN1QkEw6aKTJ3KYll1oKeDLOtpMlx0RJ4yMmmG2wIiuIWGhtaAS67BHgieBBRE4IlO20clzuLChvIhKK7zehPlstfFZAATxIAp1JQEpi4mLuNTJInv4kSFirrh3OlvWHcPeiDpM7BJyQ9HbRAEc1dTNF8ECiHb8X0fSL6fBSdiaYtuMA4QtvWk4AkVhkwRk1f5VhERIi5oi+fjW6N5Nvxi9UvzaqgK2pI9kSL6dr53g/BVgw8ehi23FkH9bNYVGcIG77zxBqeOIOxBLLSRi88wrnUlyqirUCYG0uzp7PKP9OovD4FrbGguyXt2yDjxlpEGgIyL0REEX4V/wDiqgnkbaAXxE3nxEjwrFSfk7t4WoczE5G6De7oCJmOENqOyhohIO+bSmnasxjQMLhtOb28JM4iEXRAu6nxYjkAugJ2WRM0UkRVRc8slSpPsX0VY2VroO/UrTbeTpg8QFaRjumFzICtHMTUeXq4JU4i9E0PeE7iEmXiT5/ZDcdJoS7PZVTttEURFJckFMqnGm2GynMH2kfHDnMI5QkPNutahG10Ru7l8W6dMGsslTIslVclWpT0f9HMucIuyicjxC1Fpted5dLQGmkdPEiTxZJVx4DsrhcQrosOOB+WQ713tFpdduIebwLUgceq7Gjh6lMkYw3AomHj9aMNtWjbfzOl6zpai97OortdN3jzg+TFErvKJ83RIfisgtT/ABDUJD9Oaq420C0Rd/1DnkiVxGwXw3Op7+VSuaeaTwXrWSVVZKgxlrvhU2GNSfFmLqjspu2sazPrgIXUFzSYiQ01zcLt1NCJesRDTsglSzC4xFzCVva01JTkkHFPiNmzeGOmQ94G4i9M7vWuXKrm2Z2YFu03tRDyh2RrZspEaFtsgbt09odX/LUpBi7Vy1CUm+JbbSehk2FLsM0uDSUG6UNOW1baLcnk3YY5biz4gNu+gMOvl/pWnjZYP1zaIxzXwRRTwU7Y/gzE5ncSm7xISES5TC4eYDHiJe9TLscROlJnFyynRaY9KLFvAD9U3nZBoqcFQwXw1KGk5azNCnikkzDXEs1GUhtd0ZYhGInYosSmriISbFpp0BIu0LqIIZdWhcvEidVQDFWX2yEZTbweSLonaXpBfwIfOPCuvGipnx3ZiHLuIxJpwiuI2nDaF0rbfrloFQJYeg6hJ1cKtSo9hA5gYJrciUjfFc6TbZtkF4C0AEdwmnfhQXWkQcxy8eXCm/pEAWsOgtAW9GRIky23SbsG0Bai223laSEBLlnml6Z5dVXjO6MH23G7e55sQHSdMIbY4ViRCXO0Nh9zPXbpoMkSPw8PDjVm22AFLtYMygFh5OtRmMSEAlzN+wEoiERc786Mx1plSaQ0tJTzyasKy4tAgGJzibJsRaiXDEw/vrkZqUZ/9WxObutHAHxaUHgKeHNaeMB2a+qTbbofWgiRE4IjewRERAXcxFkYF3kFVolcFN6ioqZqyDdjEFjdtySkuOt7mI0ZRIxvg07Gisxzaf7qcYNlSNk1FSHUmapnkqJauzcRGojDQi4FrQ6HOcSPWYnwTXcS58E4+BOqtY5UbXnYWydP35vCfZ1t/Yz2wNmwvKzU/dgsvv7ER17o7jW6X5IllzFujD2gFsV/HUEx/CXIbxNOj2bm3B5DDVqHyfeXiij4fDdyeeoN0wtj3PGLti+Qj5VhASn7NwhWqcmuUd3Uu40asnOM8rXGU8N5RsO3NiW8LaVSnHccOzg9cFXvHdUy6F2bpb7nkRbfaN0PzQKoSSFVndDMbvct3ynW2/iARl/WjW38qqvN7Oqd6S9Wvsa3yfp799Duy/RMn60i2hd3cSSfkx3S/wDSKl1MHSI/Zhkv0gFv5UwD+gq49s+lzlzTh2yivVnS7ypuUZy7It/Irdyfp9Ii1fF+nwZ00OvW3F2rrR/vej/zV4w4REJel63o8vxq0YilrhegPa9Ic7vnV9BHGjdgOGSZ81iBEG9+Sdo3FkAaSI3XTFNDTYCZqXiBevglTuTgT8aWZYK1JWHHVtoFkuAC4gQCovzhB0kWODjo3I2nDIh4JmuTv0W7Ndxsd0ncMmW1afZIGHCE9z49RCBKnjEU7OazCtK2pyr5uo6VGKmlo284fgso3DZfJ6TgqlSTg3qkvvkUbHYy+OqU3uCt5bhK74iqn46mbG0bGkSd9bSen5lQSisHHlLcL9sfn+TPT2RRlxb+X4LKLaOCVvfR0+g7/h0mnY9DLldH5N3+5VfUVJ8prl/tj6P8ln9Eo/FL5fgkT+ItXaXLvZP9FaG8UHtF+VTJRVFyluF+2Po/yXf0ej2v5D83iDQ6rvml+ikO0G0JRozj7NpOgNzV1pgJ3WiZj1EA9aovBbcqb6Kv0eVdaMk5Qi13ZX3ZYrbBpzi0pNPyZV+KzCdJ9x0ifMjJwjIr3TK4hK4y43as/HpWmvDHN4JRu0BE6x64h31rq+2NBnw6yjtInWtSvbvBha+umhtbIt26HZEnLhEx8kV6lTxknjWoG2642VwkVzRCQuarhNorwLV2htDLLPlrfrC+p3dFVIcH1dafYzQ7yznbVXTnxXzXaYSD02+z2fJ/sAhptfW76fG/J+dT7jgDvBfAbWpAb0RHlBy615gfUdvRPRsXtUxPL2fa/N/KEfjV7DyFgbRdyRNmsDIIMIpOKhiovznGyKQDcHE2gAmiE0QXVExC8kJUEFRMrlpp2ShiLkxqU3C0R7XAxB2XHii4M2FpN2A4DomhdSCuWeWfDOos3IuFsJBPutNXi00L9m6vtIt1vW3BazIRJUQeK9fHikixPEWnxnPtNONE8G8cFx0Xh1YhCK0CFlu0ffuVeHVlxA82sbaAWtwODhcR3fUuTisrlJq3un6pvHZ5t3lnrzzySs4KYVuWyNzFt/YN1rEDcb+zUIl3Shbq/wAKpnllwRc0qMp2fp2h/u0piqOn1vm8tASnYWO043ibsphuR3Jhrj7YOXW3tPBbqBUIR1ZLkqLlnkqU9zoUHuiDJkMRIsSbgu/xAB71badu+woM1IcQU9wgtjci2qhJYRlVdd1ODcIG4Au97cFtwgEwuEiA7VS8F67VzTSlZuySIRvccOwd22JEZiA+QAkugE8SZJqoC0MeBpscefi7gRHDMPYjE0Qk73HOmssyCmeEcQMB3JoqJkj2Q6FFVcMBw2G4/hEE4MYvqlgYvvv2nvxdGPIMXWiE0Fo7mizLLNbk46UqDFixRpsm5sXW5EfuaS0WkXWHWmituHkdB0GnUNOomQXJUzRWdqc+JNEL74k0G7bMXXRJsNQ2NFfm0GoksTJNa+NaiCaYKEEsMbJ0YH1T7kfKMJOCMQmLrhOd9rHGEDuvdiSoioLSml1iHjjswYY4McKMyMkomGzikiJ74nwMjCzjaNxASrwVV0ovKlQkeX435pUrwnGXGH2JJD3UUQR3APkZNAbWqPcGeplt4hPdCootuXBFWgH7aGC03tFMitBGJgcSmNNtPlIaiiIOvALRlCMXhAStFLFRcwROrNFU9IgNDAw3dDhQkLTtvcMnE39PduIXC13a6aGzfzEeaoWSIqJmlMmKyybbadK7uuQBE6ZDaY2vOnvSuy788ItEpdeQl4HOMp2c6Pp2KxsKGO3Y0EeWLspzS01fic0/fM1EhVEFF5k6uuqpN8AMsrcMYHh7oxmDfmniTbksxLeg2xKC0gtVE3usEQ1RckEk7SrTh0f7L92N6Y2LPHeWqHGjnHs088iRIC07rsx8WXnyv/Yjonw2Cy0Lzf1QdauIXJeZtNkRCR7iKSq2yJEIquSZr4VWrABRERERERHlERtER9ERq6qYKY2P6FBbfF+a6W6E7m41oE6Q+TJMVUB94Lvfq19n8AhwW7IjDLI9q0dZedwy1EXvrS9TrBSq8oAVAtJpTFaTctraxNEuapbrWqDYjRvVW4ma2vWc11YX1PLZHCGyYxq0/Tm/pS34qeOontLEHURjvWnR3brZXa2/zSTrRU4ouWVTZ8bqaJrHZIbruz5XpD6XjT4a9FN9pba60UxjmDGwV4d9jGXe3e0F3K0/5Jp1Z9S9aeJNcHAW3faqw52HmwRG1qEuYS1CQlzCQ9r3qYm47F1w/WpeSI3MkXqczXwKqeavJX2e3rT9DJ0L9YxP1G1nYlrm00tiYa03pERGnxHd2NxahH7mJu+1aCKVvwVXW2fSXh8OS00ZOaztM7bNwREIiTgO5OEPNnYhZW8cs0rGyo1IvDTPdGtF9ZPoVo6Rp7ZXTUC2a24w2SVkV8pTgjcQRmJcgxHyiFppbR861K+6nSHvTRB5JPluR9oOLg/FpGhUlwiyE6sFq2hZJetppioU4tBEMEStdfG4Sk280WKQ9leonk4IlyIueahm3gQuldNMpfkxhEmon+tDNSleLI1tXyalUON2j5R5R5RER5RER4Wp4qyFCxxrU9Dw1rxPSHqLMObG3lEGwERER0gADpEREeAiicMvRpxj0mZau9X6fOpcgV6ps8KM869FKwRaUtJVokj1ka8xHDGJLe7lNNvD5Lgidvq3dXwVvAqy3lWpakioNv8AoXwwW350InMNdADJ0B+uIskOs2XY8hSFQPqt6s7Vy4VGMqtbpWn2Q90PXIdFPYDWv4xBPaqqK5Fy8ut+6jRX/wCccvxl/ZHQ+SVtuUJVH+9/Jf3ParnpRxdsZLUU296AtbxwhKx5onCJEJo+KZ5BxAkUVQupFRDCxqZOkno7xdj/AKzjssSWnWgNzcRmDmxu9JwMnWyc0pwubVETxJXn5EWnO3rqY0pxb83ovuXOVdxzdqqfxyXotfrgp2Vhm7b34F3RGvt34jbY4VxCzJDNe55FoktqqqLaSipIl1Wr0YRt3hzZfdjdd+MVgr8RoagDEiTLjSwN9x0h7mtGTLABHvxXCJy3UECXSlqLmtvBFyq3cKibhhpj7k0DfxBFFX4V41sPLu53baFLrnLPlFfkw3JGhvV5VOyOPV/gVVCOmORbAbb+6yA+KAGZfjsqb1VnTVMukxGB7DRul6zpiI/gFovjVpnJW357aNNdjcvRNr5mz8oKvN2U+/C9X+CJi0bRWmJCQkPo6btQ+rU42KwFrfHiuJDnBimwDbX/AHhiL9pQoQ+U0lu+PryEOKZHwbMOw5zFZLEaKIi+6RN6rRaABG8njLwMgAkSlxy3S9aqlSXa3EIzj0SHFL6xw14WInlSTN4O6MQdQUyF150RVPAg2IiJkqL2i4eKbfczl9JZmk+1E3WihaK4hLizr8eB6Nc/9G70qdGIpE/FVJ3FXIe+HGXYyx2RYbdA2Y7pr3Y7eapuxzVc09+r/FaqbZ7ovmRGSYCThkge7ClgUiE88bEggALw+uBHggDlmi8a2HYlzRowqKclFy3cZT6s54f54mG2rQq1JwcE2lvZxjuxxK5g7T4mOz2IuuzsRSSzisVkTclSRfaRQf3reZHc3mScR8Y+an3pQOZFiPvxZ2LtFGfiBcmLvTN81KjkbrkhkCuw+1ywEU11X5ZeGpGXRDJPD50N6e26/iGINTDkbhRS8N7vLgEuYidz4ZJSjaTo1nzGXmimYZH7p3QyXWILoPSRj7tWwcddeJUBCACyG3NQT3q2FbTsuc3lOKW829HjGId3c+P9zC9Buubw4yzu4Wq45l3+HAjW1GKShMpjs/FAiwcDwl9yPFmvMOSpUwRaC881QEI7iJzIl0+FVzpD0WbbOzdoYDbTuJtNGMxH4sjEn58dwggvm0Yo4KLwJM8iu4gKoqeCeYj0bzFktyGJsZGxwyPAfjyIqvsSgjgIrvWyNEtUxQkyyVFRMqy2b6NnY+KQ55OYa03CGT9bQohxr1kRzZVTcNwyNUuRdSr1KiZZ1ZntKydvOO8m3BpaNa7uMYxhvPX39xcjZXfPRlh4303w1Wc5znOMFmUUUVoRuA2bUt3QpI83eXC9oRuH8BCK+zVSCnLboIe0Xa/vEvX7NW5tOVsKSXkx3PyCqmMQkFze18UdPzSyronI5/7E1/N9kaLyqX+9F/y/djjF74y7Ft1Dc+x64j30B/hWR6utVjtInXUYdX8r6fkjSxmU4JNuAVrg2kJ9oTbK8D1eLr4+Eazxxkbt6AiDUgN42I3WgVxA+yNyaRB0SRPDbu1XK6txNUG3PV630L81Pfp1w9e8Sf4r6P37hpfnfTKmi8fp8a34134qcYBjuX9X/Zf/AH+FavjDQGhU5fa/OrdFTUPrD+UNJ8+X1vKH0/71boy6vaH8qgMHC8nyq2thp+dSdxdI+z+TS9pu0dWn0e1/y0A6bSp9cuf6svmBbSH6fT6eKlu1C9/d7PL/AFQW+esdm8ElznhjQmHJDpWlaA8o2lcRmWkBTxkqJRLIEbZ9n6doaX4FgsmY+MaI04+6d2hseUea4iLgAp5SrVz7Ge58cLvuKybNV3c0a0y5uU3jTIe1yovN11dezGzMPDWdxCYbZG3UQ6nT9I3S1GXvrU40+0FZdGHQq1Etk4vu5sm37BaJxWittuK/hIO0etUyTxcEWreK0REREREeUR0iPqiPLWRLSeQVemMMDgZ7yvFcpPfWKnV1RI5FCOVldSNDrcB0cRk2lTfMa7Q0tzrEhuqsXhlGImD8qlIPUOM1qUKuaMjwFN1aX27hrFCrYhVRLBUaZoW8vxS1Xf0fh6/fpjxCE2fZES+nw/hyqWvN3U1SI3ZLs8vo+qX05kq9CRBoipQLeWq090vEKTgLoKJG5HMZbZESpYLQmBmCpxvQHTXhwyFc+C1cr8a36Xfou+Gqr6cNksTxQWosLdg0A70jIhEHHLTAmnQI7SHkVM0LK5fHwuyeYiHEj+0u1OFzMOw+dFfZHHCZjGwMYSOWMkgC+K+TKZgBGdqtuKiLxROurrgM6R3ttwiN3k3VUXRb0IlBktTp77ch9rvjTQiZMMPlba6BFal4jciZCKavMlXhCh2+l9PR4/jWqJtLUSw3oZRmxutEdXxR+dxL4EWnZiP5X09Xyf6ferXCZHmER1fk+V8PX+Cl6JXnnIlFHop5NZ5VhWYLVlkjMW6zQqEKvCSokjai14pVq3taZEgREiLlESIi9EdS/NqEmorefUSit54RXPSlOvlo0nLHat9t3Wf4t0nwVEaUYhKJ143S5nTI/VuLO34E4Vor522vedLu6lb4pPHhwXyOw7Ot+j28KfYlnx6/mLsAhb+Swx4DdG71B1H8xCq9BKqt6LYlz5vr9qCwf4Q/0Aip7dWQh11DkJYc1ZOs+NSXyWi+eTRuVd1zlyqS/Yvm9fpgrjpX6O8NNs8RBvueTv4zru60hKJp64UcDquXPNSTiqJxzqHLU+6WJ/BiN6ROkPvZgH9LnxagNafy4vOevubXCnFLzer/AAbDyWtuatd98ZvPktF+QqiekObvcVkndpAxYH0RaEQL54mvtVduKS9ww++XLHZddL1WmidX8Q1zarhGRERXEZERF5RFqIvjVkeQVpmdSu+pKK89X9jw8r7jEIUl1tyfloizMRkFhGHdyjpxPGGBOX5WH4O5kbMT0ZUm0HTHrQAaRUS7Oo6GKOuS4w3WjvmB06b7nQ57OHny/D4qY8fxF2XJdfkOE6/KeJ10y5iIyuL1R8CInBEFETgiVNth9i3ZLbWKzXW8Mw8JA7uU+OvEH7xVqJhkfgUoiUMt4nBEzXjYop0ev/Cl4P6GkUP4i8UWytFFFcQlxOvrgFFFO2yGF91zGo5fY1W4/UbG4vw6Rz9Kp0aUqs4048W0l5luvWjSg5y4JNvyNuBbMSJQK7oZYTrkPLY37PhP3+rz0vTZSMpWNYtDM+y2uhFLxIW8XP4Ep8nSWHwdmyxVcPiOdzwoQaQfMNKGqJwJPF4EQV8XFlZ2pimVkjDYfc66e9AiPAPjE8kuVPNb79bK7OyobsajTzwbcterOI6JPqbya30u9rtyp5W7xUVHTrxl6ya68Y7Bkx3BZEM7JAZXchjqaP1D/sXJU8VN1T/BJMeUb+EIZvRXAvhOuoW8YcEM7BvyLIdeXmBU6lqBOtkJEBaSBSEh9IStL51YjaFpClipSeYNtcU8NcVnrWqaMvs+7nVzCqsSik+GMp8Hh8HxTXU0Y0UUVjTJDlsuIlNjCoiQk8AkJDcJCS6hJC4ENRX3TewEaCLWJwhFlt17dOxhHvQkQmYugPYC0TzHqzyyqVbKr9exv4w3+VUn6f4ov7P4hcN3c7Tb4+ULjToFp98c094lrpPIqObep/V9kaLyr/jR/p+5xoLnMHkFpH0eYfwgX9NPmDTnd27GadcAj+uWrdNzrQd+Dw/ZGQLJOtSjsp2qjE07XLu0Ij83tfCJVvZlENpgRCTRCTZdoSbISAvg0qnnrbDVBd9X5d32d7430+hU6wcal7mT39z/ACS7m/8AG4eP5JElM2Ltjc2+0Ii1IHeCI8rRXEL0cfRbeEkTw2k2q81Sno72Pl4qUliK2IikTvsp0iaiRh7rhO3yXssg700ZIPFVQCyRclyAb4OITHXG2mjkOuGdjbTdxmZkJWgAClxF5kpzxpjFYJRCm79kZTQvsERCQvsXc4GBqJdacM88jBcsjRVsBp/DcIIoOGbuU7ui+q+OPhrKM4PfYWHNZ5xGnbbVJFuVPCqoJjLGlg4zhQO4wRRQliUSGRWvDgpRhzjusGSIbW+AwIwO+5ADMkQEyAoEcYl2iQvuW2jqu1cva8mnSHib7bYvvOuEJamGLiHf2kQ3naqKERDEkUkVFJRIRVMiNrPafZOTgzxDiTXIe6iCN3c+IWheLwPCqfWlpCS5KhKulERUI2kGAxHcSmttXETrxiJGNoiDQDaRCApk0ANDkgiiIiACIiIiIhLLwOBJOj/YiXjchx95zdMb25+SQ2kRkVxNMAOSX+DhwRMvBkldPbD4TBw1gYsBsQDmIrrnXS5b3T6yL8SeBEqBkAxIzcWKNjTIWiI9q3tlbzGvWq+FaZY20BtOXXFzezWWp2WI95DeL9Bwa9JKgpz5Jxhkx9em4h7RD2rfS81KNmtqxfG3tcpD2rh5htLlJKtO3a1RLfJWaUlfWs0furA1qiWAzUlYFWVCpVwgzUtetuV6aVr3XaqQFglWV9JEWsrqjujIoU6xVa0XVldVcDJkSVjlXt1YrRIAq0nkBW4qxqSKMQPt0lIadVD6fT8P4a1qxU4yICJgKWCnZHtfk9ovi8ParMG7a2MB2vK/J+nGjlklg3tDW5FrUNZZ1aZIyzrJFrXnWaVQIzQq9vrWq1jUGipkq1HOkGduohAPNILdj6vMf4ht9upElVp0hz97LsHljpZ7ZanfzR9itY5XbR6Hs+WPeqeyvPj8jO8nrPpF3HPCHtPy4fMjdeLXuVK8Hh7+Q0x5Z2l5h5jL4BElrhlClKrUjTjq5NJeLOo1akacHN8EsvyLH2Fh7qE35TvfS9vk+YI/jqRCtJgHsjpt7NIdqJ/c0R0x0lbu2/KvPShD72ovgr6GhGns6xSfClDXyWvmzkM5SvLltcakvq/sVvtbP38x0+sUKxv1G9A2++txfDTVXle1893dxK4rSqS4zbb83k65b0Y0acYLgkl6Eg2H2eaxBx9iSN8Yozrbo8RzSSBM5Zp1aFdVF8aJVTdLHQrLwpwn4Tb03DyLSYjfIY7VjoBxIfTRMvHlXRPRtF3ELelzSDv/ANWGgPZ6y9uneTjm77Vds5JbMdvs+OeM/afnw+WDmHKG7Ve8ljhD2V5cfnk5Nc2ew3Z+2VjojiWJ2iUbAG3B7ni3DeD2NSAzTzowF2aEmdyEqgk2VcxDajGhk4hIyYw9tZTpiItRcPhxyE2o0VnKxkTNGm0Tiq9aqSgq1EtpGhbecB5pwzKx0j7rv3oymgltPXkzme8afaczLjrTPJc0qR9HHSG5h7f1PiwYm7ly2XH33ScdkEf2Fm4kUQIGbycEVHJCyVc1rLV/4UvB/QxVH+IvFFxUUUVxCXE6+uAVKuip8RxEELhvGXQH1rRP+gCqK1uw6WTDoPBztmJj6yLnaXmXqX1q9NjXVGvCo+Ckm/DOp5r2jz1CcFxaaJXPjqWAgHagTXWn08klIxzy990E+GodVlYhiDMeYjx2rh2NMDvbuUHbclP0dJjn75L2ahm0mFHh0oRuRQzF2O6WSo6AkKp5iVNKKn9ipWY2ta5xUXCCUZdyXuvwa4GH2TdJJ02tZNzj3595eKZNMKgmEzCI6mqlHiuvOhkCboTBRRLhRFJLyt458vnqvcXdE5Dxp1HIdMfeVxVT8VTRjEzGDKxV5R7onr3Kwo8EQG0ICUEzW1EsdXr+1p46gKVDbFSChCEOD9vXjjCUc97SyT2PSk5znLq9nTty5S9G8eR7RRRWAM+LcActksF5Lol86rD2nwV3EMKmQwcFp2bHcFsnLrAItQby1FW3T4M6r3Z9frtj+Gb/ACqtmFJG4fWrp3IdZtqn9f2RonKv+PD+n7nDfSPsViGEPC3iDBNCRbtt0dcd3yd08PAtPZXJUt4olRho+z/9fpld8WvobtVhTUyO5GeG4HRt5QO30rXUID9UkJF6lRU4VyjjEGDgeLOjKhwO62Sujb1ufLwvlvZmycPHdkF13WyUgUUFRGxULk2ucMamqifo+6PxKEMzHXChYa6e/hgNqYnibgNXOtYdHdTU06yLWbpZIqsMqiKi3I74vt5vYjsOAw3h+GsMiUbD2tVx91xB30p4tUqQt2akS/hXMlgm2mJ4q/J+qc18pBuujusRacB6Pc0ZG01Gda0R7S1IwqASXcRRVWthrpfdaG1t6ELjYjytl3fFB9kfJFt0SRPDaTa+FKtg3i9da2Wpu/evl2n3fJL0E8Xo1MMPxm3DHxeG8RdalxmiIh3tolFdMrciGOmhFVFRSUCRFRUUgW7DdFE6X3+QIxGBK5vf3CT/AGguZBUcFlNKrxBVTgipncNhYf0G4eW8KViU9510bXCbGLHC20BFBa3R2CIgKIKLkiCiIiIiIklSk9cDJBdn9qRxON9RZssYr5acMxBwWijg6Y6cNxCOYE1Iw1wlyQDEkFbck4Cgyjol2LKI25JlRG4kx4nG3GmiMmmwbdLUAm4dhGWrQtuVmSCnCnqP0A4a3IafGTLebArijP7gwdIeW4wbHR5slz8dWSGFWlXot4qL3mUkmyLLg93Zupkxzo9ccbccacEHOYQIdPqkXXVoMM28tZOBXr6S86FFEpLYfaR3DXyg4gJAJFpIuUbuGnwEC+NKcOkGC5EIcXha2NJS2g7I/fAj5KD1/AvjqcbV7NMTm7HW9Q8pjpMS9EqZtncNkwy7mMt8wekSLsj5JD57qvxmnquJFjhsljjcllswK4TG66pEJVQuDS/qNtBJwi760dtkwx8hp/na9UXb0RPAmVXNElXCJVCcOsrwF6lWQrWgDrJDq3gobt3dWJVkJVgdUB4lepWOdZpUgYolGVZ515TIMayzotr1UqIMFoyr1UoyqQPCGvESs0WsXKrkpgwtu01vRK1sJ8atqLUclTxUr2vUrzKqAKySvESihVIKK9WvKpowJMZnCww4+XYHSPlHygPwlbVOuGRERFxIiuIvGRaiL4VqZdJeJ3EEUex3xz1iHQPwIRL7SVC64ry52r0m85mD9mlp/wBT4+nA6XyXsOZt+dkvaqa+S4evEKmfRpA1OSSHl7036xZEZfAlqe0tQ4Ru0jqItIj6VW3gMHcR22h5gHvnpGWoy/CXDzU5C7M6Te89JezSWf8Aqei/I5U33M23NrjU08lx/AvyqB9JeIXONxh5WhvP13Or8Afl1OJL4ttk4ekQEiIvREbi/Jqn8Qkk68bpczhkXq58o+8icE96ts5fbS5m1VvF+1Vev9K/LMDyUsucrus+EFp4v8I0VtgRydcBkeYzEfjL1/B11qqRbDx++OSS5WhtH+Ec0/iC74yVzDYtg767p0F+56+C1fyN32ldK2t51H1LTx4IleMzhabFhoeURbER8kRtH5tNMaC66WrTdSmGG8c1cxfNHyalERgWxtGvpKEVRgoRRxyU3OWWcN4tLYkvE663LN07bi7pjjpEbBERGFaACAiKAKIiIIoiIiJWvZ5hhybGaZYmm6UhshHumOXIQumVvcqXCIARKmaKqCtLpcBhot2ctsSK1z7E9qErSDVl5P04V5sY7GiYi2Yvi64YmwFrUhogKYPcpGJE0SF3l10Mly5880yrB3H8KXg/oXqP8SPii9VooWiuHy4nYFwCiiiqFSf7PYvITCgahgbz7L5CSbknARkrz5iS1FzIeGefmyp22fx6Y6TkSaAxpToKUR02kQCUU4goLw4ZeDwZ+Kmvo3db+p8polC8pIkgFKOGpDY1q3zWsU0l1deSp4a1bXRxbaGU0LQOsOgYu/VSRMcyE00A29w4nbnkvUi1vdtOpG3hXUnJKCzHXGFlPqeuOGpo1xGDuJ0XFZc3iWmcvVda9nPcxB0h41vxisEvfY4EMgLFAUklYhpb1LkoFkqZpx6+NRKpJ0kG05O3rKgYvMtGdhCqIdqplmPhsAF/+ajdartWcp3Um3nXGe5LC+XE2nZUIxtobqxpnHe3l/Ph3BRRRWOMgK8F/wAoa/hR/KqdlMt9moJg32dr+FH8qpxu66jyC/8Aj1P6vsjQuVn8aH9P3JBh+IbwRuqDdMfRW1jZNywccakx2SbsbFq6SAkRiFxqlpiRHkqrlqVOHXUqw4LaeYT3LW61IJmqo4mNyHhr7oE5jkR0bgfaNiBYYjd3qSy68rchnm0OIQqnHKrx6GtlIJud0uxnI5A6X1s/uhApIEF7oxwcdRkr2mkUUW3NkMhGzJbP2+2Gw3FbSmxGXnWdTRleF3+idNhUI2l8Irmnm8bFieCuPtuuYXbCxOKZETDpEUWSZXOqD5alEHSK5HgTPMs1ReIrYpUlnLKvJJ8Xw4i1AWr8qo53U62VpVo6KOkdjFxdYMXIs6E6TEyC/wAHoz4LaYFbwMMxLIhzRfxU+bRxe+Db2q90H1MjJGcbGCHtU5x8Yu7VRp/BnC5StKmebh0xrUBXVV04yKKTLJbmXVkR3VWUDa0mnLJAkBekNtTHDcZbcG4Sq3KjjgS3h6UKRPDqrcMwa0vFdUYpooyn+m3ACcksYi0PfIvezt5tw4Kld7JiK+0tPGwW0G9bEDLvg6SHyql8uKJuOCY3CYWl86qaACw/Fdx2L9PpAZaP0V7I+1HBHiXjHPTSkaZsJf3giX0+i9VOoOVYksFRSi14q0AVeGtRAJWWda869RaA217nWtFr2ogzrKsM69yoD1a8SvUSsaA9oIa8zrJFoDSS1mK1ideItSBvGvawRayqJU9KvM69zooApPic0WGTdPlAbvWLsiPpKtqe1Siop0mkXczYjy74bviHl/bWK21dytLOpWgtYxbXjwXpxPds23Ve5hTlwb18CATJBOuG6fEjJSL4f7E6vgrXQlehXztKU6s3J6uT9WzsMVGEcLhEkfR/hu9k70vsce0vWdL7EPwWkXsp46sXOmzZnDu5ozYfbC1OeuXMPwaR9ml5FXeuS+yls6xipL2pe1Pz6vJHKtuXzvLptcF7MfLr82RjpFxDdsjGEtTpXF6INl/advxVqA04bRzt/Jcd7OdofwY8B/Dze1TfXIOU20+n306ifsx9mPgvy9ToWxLLottGL956y8X+OALU2w2PumAa7XO565/oG0fZqI4YFzw3corcXs6ql8V3ePVu/wD6c7K9+8mv5Y/Vv7Gt8r773beL/ml9l9x5wBjmMqerqSs2iNbEOunyeWaMcJ7UzRdIrfuuktXJyjbdxtXz/pphCS424Dg8wGLg+sJCQF8YaJL3ZLmrTbpGsQ1lYLilg6kwmcMlhp8NIvNNuWlzDeAnaXpJdSqqMwja6Th8i0BF1oouHkbDhEInbhUK0hMUVWj8F2S8ByVFyTKe4Z0m4e6LhFv2iaavcEm7tIkAlaQKt2ox4cF81cw2nybuKNRulFzjLhjVrua7jomz+UFCrTSqSUJdedE+9Mm1FQn9tHCvukj5A6zPpMwwRuJx75B2sX+jXn/FP0Zkf1W0/wCSHqiztjMYaivkT7AvNGNq3AhmGq5DC/h76e94qmhbS4cFzinHfTsMBh5tPeZCccW34ckqhMR6Q8PYccB1x4SadcaK1oy1tGQFb5Q3D11rTpJw3dke8kWiTYkW4d5jEyHwZXWtHw69K+JcsvZvadtT5uNFtZysp6emM/MxN3T2fcVOclVSfXhrX1zjywTrFZW+fdesBveHdYCZCA8BRE+BOK+Fc18NJahbXSfhRX2uvd6DeF3g7rSMGtI9ornQ4e/4qVYZ0gYa+VgOuCX+kaMPyqxktk39STbpTy229H+DJ09pWcEoxqRxHRaolVFJ48xtzkK6nfCMPF8rd4I/lEXZHVlbn1Z1cp8ndoVJbqpS81heugqbZtILLnHyeX6G7ZiMRPbzstavaIbRH1vD7NTZgaQ4VHbbERDSPzru1d6VOKuiI11jk/sn9OtubbzJvek1wz2LuS0Oe7Y2h0yvvpaJYXh3+Jmy5q9mnATppw5biu8oqdFSsyzGIXMv3F7NM2JuWk3LDSTR7h30mjLQXpELpDl4kdPx0rw53mL0q1tiLgyWi5XSIPVuAfnIvGoKJIi2PbIRn8THF4TowsQsFuSP2mcAaQJ0R4o8Kad4meaZIqLkijLVBxxgd9bvQK7SV2ofzVqPYlBJyMLrWiSzd6pE2WtovRXUnm663bI41v2xu0uDpIfJLlIfWSrm68adRFsewc5ayIqTkuoh+lpV6J1RlBuxzBWpLZCYj6PlCXlCXWJVXTqu4e9uzIiaMrW3fzS8k0T8NWvTTjOFC7dcIkJcwlVyM+pggmPDi4CMrC5YyLdRQ3WxtMf9EY5Ld6Jfh8FJdk+mVtxzubEGiiP3bsrrrLu0PgUC8xIlSvDsJKNcIlc32R8n0ai/SFsSxiQ7wbWZYj3t8R5/Qdt5x8S9afiqrSJJotCHMbdETAhIS7VVZ7oPDt0LGItfajED9UiuAvgMcvaqsMC2zxDAZPcksSJoS5C1aPLaLtB4cv0ZVP8AazbuJiWEyQEhO9rTbzCYkJDp7JXD1Uj7LKuGNScbHYjvYzZjykI2+rbUqjOVUXQhiW8hCBczRE35XKWn5ttWmw7SayQawPDS1kS0njnW8qtAxzoQqwKvM6AU517nWgCrbQGxKEWsEWvc6iDaK1jnWOdY0Bmte51ii0KtAYrQKV5Qi1IG0VrLOsBr2og9Ss61517nQqZKtJMVhNvsk07ymPN2hLskPpIvGlNYqtW61KFWDhNZjJYafWmXITlCSlF4a1TK4m7HSxLvQi8PZITAPjCapaXvZ08bMbJk04L8ghuDUDQ6kEuybhdVydaInv51Ls68Va1a15F7Pt7hVoqTw8qLeYp+muO/JnK/KS7rUuabSysNpYb+Z4q1H9tsQ3UYhHne0D6pc5fF4e+SVIFqs9s5+9kkI8rPex9btr+Hh7wpV3lftToNhLdftVPYj58X5IhyfselXUcr2Ye0/LgvNjJXqJXi1myNxDdy9r1e1XC6FGdepGnHWU2kl3vgdRrVI04OcnhRWX4IdsNYtbEu07q9UB5fw/op5wBoicu9Kkcdki7NpFyj5I9kfgqW4JBtGvpXZdlCws6dCP7Ul4vi35s43f3UrmvKrLrfy6l5I3DdWaFSvuavUjV6cnkSOHtvsKiQR0tsuuFPkxhHeT7R+p7UXu28hkiokkuVuUyVc0jquaFmtRkCG0SKIwWkdROT9Wkhu/yrwkV3BETMUyyTMVgsva+Y7bvXN7arhDeTzlpOle6Q3uLapHqVU614rWwts5NohaxoG3lPPx/dKxOS5gtDFe+SGwCIybrsXDxARKaRkRYXF5QGTrIbDXjn580TJPMIitOP9wtDGCTLB0X5LRy3Y8WM0JT5AtXyTSbIRmISIo5Ct2SKSmjoV8/0izCG0W4zVzbTTptg8hvtsMtMNA6aurosaRVAbUVeKotqWtv/AEtkbwXURsTAhISHegQkFqgaELmYmiiKoqL1jTIwTnEYItkwQA+AyIoyR3pRzIgN0waMSjmtoEACuRoC5kvBUyM9LgWtkReT6OnSY2/GHrSok5ttKIlM924ZCiE44rzhlaiCNxm6qlkgiiZ+AUrW5tjJIbVBjj6J+lx+yZdqmRgtDHojrs+YINiX17MK4iEAEReeJSIj4CKCOWarlnkmfVWyHGY3Etx4hdaiNNy5LpFIDeWS2oDLUUQcaIGjPEARSdRVRLlRBVEEq+xHpLnPERK3EC90niFtt1EJ0idK8iV1SMkV08rlXJCVEyRVRUcXbyY0VzO7ZLVrZ3zRiJcFETB1FQbeFCpZz+GNsP4kwMNm6JHHeg4/IIgJqXACQJk1iCgdrxiHe1z0CuQrclNuE2uvboW22tDjm9bKUR96jmdtrrxDqIc14dpcskyRIB/02ld9IhYInxtcIhNSLvrTylnvM71NoVVV8a0YfttJac3otxiK0xtIXcsnAUC5XEXqJfDVVLDyimC5Njdq3BIWnSL0S+nZqyMN2gttMC/vVyrL24lOEJbuMAtDa02AGINAi5oA98zXxqRZqq8VVVVVVyh9KU9vgjcQvWB7815K98LuONSq7ztzBNqW3xE7h1Fu3fRd7B+ofV7/AL9PrknTprhnDumvEWLrGMPW8VEhJqTkqL1cslOpeKU9RPdI403b3jCzy8pmX+O2UlS6VAo4ncmCM2t3FW7EJFo1xUHurseHh3JgvyE/9erTI91Pjp5XRcH0/wCgm/rtQ6TDJTdZ29FO1m4tOmkeGvd7bLtOmTnxi0/NrjOR7q7HjDdlEwVBXyWJ2f8At1Yt+6rx4bcomDaEtHvE7qT/APeqvSaZTdZ2VFmiM2TF7Vjb9voukYF+EmDX2qh0qR3HiZW6WnyuHyb+V0fyS9pa5Ua90ljQ4k5im4wzeuxW4ytbqX3OgNOuuoYh3XdvVV0kVVVUyy4ePXjfujMYlW7yNhYqLu8Em2pYkhe1LXgtTjd00N1nckyR3tt/s8peqXa+Ara0uSbdXZrjGJ7qTHAa3PcmDmNtupibn82anGsGvdP44I29zYSqZW6mJi/+8qPSKY3Gdn/VEfKrE8XGuK190pjX3thKf6mX+uVqd90djJdcbC/kpn63TpNMbp1pju04t8v96ow/tcWq1pz4pW1zYnuhsV+9MJX32Jf63WRe6IxfwxMJ+Ql/rdOlQJYLw2wbjYozbIaIXAuJt8eYPKuIuBB40WqchNbtx/dOXCyDjmm210QMRLTxS20s+KUxYx04YnJZcaONhoC6Nqk21JEhHw25yVHj41RV8VRljbyUJEW7jLc040QkDmSi60bSrwc5kQ8/gSrVW7+AqsnRnQZtHc+61bbcO808parS/N6qvuBIurgTAuk2bEcF1lqJcKKOSg9at1uakgvJmWlKmkP3TGNN9UbCS9ZiX+bLSrlO7ju+3xIuLO4IjlOArXDjfuqceH/suD/ITf12t6e602g+9MF+Qnfr1Hc0yKizth1K151xYvutdoPvTBfkJ/69WP7rPH/vTBfkJ369VOkwJKJ2w2tbRKuI091pj/3pgvyE79er391ttB96YL8hP/XqdJgN1nb2de3VxD+63x/7zwP+bzv1+vf3W+0H3ngf83nfr9U6RApus7eoVa4h/dcbQfeeCfzed+v17+642g+88D/m879fqvSYDdZ25XqVxF+642g+88D/AJvO/X6E91ztB954H/N5/wCv1TpMBus7bKvEriX91xtB954H/N536/Xn7rfaD7zwP+bzv1+nSYDdZ28i17dXEP7rjaD7zwP+bzv1+j91xtB954H/ADed+v06RAbrO30Kvc64f/dcbQfeeB/zed+v17+662g+88D+Qn/r9OkQK7rO30WsSWuIf3XO0H3ngf8AN5/6/QvuudoPvPA/5vO/X6dIgN1nbyV4tcRJ7rnaD7zwP+bzv1+hfdc7QfeeB/zed+v1TpEBus7N2in7iM472rbQ9ctI/g6/eFaqhVrnrGfdQY3KERejYSggSkiNszBRSJMsyumLnlxy9Zaa/wB0Hiv3thnyUr9arnnKvY17tS4i6e7uQWFl41fF4Nv2BtS0sqTU870nrhdS4I6ap32bw3eleXKJafSL/hXJ37oPFfvbDfk5X61TTE6bscZeN1iSLV5ke7ESJkbl5RAyXIE0oieZKscmOS9WyvFXucNRT3UnnXt8kXtt8oKVxbunRzlvXKxp/c7+hYeLeo6cBlNj2hr5/v8Auhtoz5pLPwMAn9FN8rpvx5zmlCnqgg/2103pUHxNK3D6HrijflDWl3Hmh5jb+MNfOOR0qYwfNMc9kjH8gkpte21nHzukefZJ2Qqf1tFcUyW6iM0UUVjSoUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUAUUUUB//2Q==\n",
      "text/html": [
       "\n",
       "        <iframe\n",
       "            width=\"400\"\n",
       "            height=\"300\"\n",
       "            src=\"https://www.youtube.com/embed/ZnBF2GeAKbo\"\n",
       "            frameborder=\"0\"\n",
       "            allowfullscreen\n",
       "            \n",
       "        ></iframe>\n",
       "        "
      ],
      "text/plain": [
       "<IPython.lib.display.YouTubeVideo at 0x7fa498359df0>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "YouTubeVideo(\"ZnBF2GeAKbo\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "extended-correlation",
   "metadata": {},
   "source": [
    "### Rule-Generated Sequences\n",
    "\n",
    "#### Fibonacci Numbers\n",
    "\n",
    "You may have heard of this famous sequence of numbers, named for Fibonacci.  Below is some Python code for generating the first ten terms in the sequence, starting from 0.\n",
    "\n",
    "Don't worry if you're not yet familiar with the idea of a \"generator function\" in Python, which uses the ```yield``` keyword instead of ```return```.  We will be introducing features of the Python language as we go.\n",
    "\n",
    "You could call this short program \"an algorithm for generating the Fibonacci numbers\".  See Note [2] below."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "presidential-prediction",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0, 1, 1, 2, 3, 5, 8, 13, 21, 34]\n"
     ]
    }
   ],
   "source": [
    "# Fibonacci Numbers\n",
    "# https://oeis.org/A000045\n",
    "\n",
    "def fibo(a=0, b=1):\n",
    "    while True:\n",
    "        yield a\n",
    "        b, a = a + b, b\n",
    "        \n",
    "gen = fibo()\n",
    "fibs = [next(gen) for _ in range(10)]\n",
    "print(fibs)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5a342dad-6e68-4ea1-8ee0-7d9e15a65d03",
   "metadata": {},
   "source": [
    "From the above, we might return the nth Fibonacci number as follows:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "6295f492-523f-42c4-86a4-c04e39ce5ba1",
   "metadata": {},
   "outputs": [],
   "source": [
    "def fibonacci(nth):\n",
    "    gen = fibo()\n",
    "    return [next(gen) for _ in range(nth+1)][-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "724e84bf-b571-4ce0-9b57-0b66d6d8e001",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "8"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fibonacci(6)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "02e557ca-8e17-4365-a446-7548a20c9cd6",
   "metadata": {},
   "source": [
    "That's using a Python generator, which is interesting, but how about using recursion?  Since Fib(N) = Fib(N-1) + Fib(N-2), why not call the Fib function from inside itself?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "e068a6c5-6e75-4c0e-bb02-4571d6eafe16",
   "metadata": {},
   "outputs": [],
   "source": [
    "def Fib(nth):\n",
    "    if nth == 0:\n",
    "        return 0\n",
    "    if nth == 1:\n",
    "        return 1\n",
    "    return Fib(nth-1) + Fib(nth-2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "974aedf3-324b-476c-9c9f-5d42d0eea1c8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "8"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Fib(6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "c0c4be51-94c3-46a3-871b-8c79e906cf64",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3.39 µs ± 1.12 µs per loop (mean ± std. dev. of 7 runs, 100,000 loops each)\n"
     ]
    }
   ],
   "source": [
    "%timeit fibonacci(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "c1451c27-b51b-4679-ae33-d4f39d1c7b57",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "349 µs ± 42.8 µs per loop (mean ± std. dev. of 7 runs, 1,000 loops each)\n"
     ]
    }
   ],
   "source": [
    "%timeit Fib(15)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "efde683b-22be-4891-9281-8754a90009d0",
   "metadata": {},
   "source": [
    "Fib is very inefficient because recursion redundantly calls for the same computation multiple times.  A solution:  memoization.  Memoization prevents unnecessary recomputations of the same answer."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "d0e7034a-85b9-47f1-b2fc-efb2de9b257f",
   "metadata": {},
   "outputs": [],
   "source": [
    "fibs = {}\n",
    "\n",
    "def Fib(nth, fibs):\n",
    "    if nth in fibs:\n",
    "        r = fibs[nth]\n",
    "    elif nth == 0:\n",
    "        r = 0\n",
    "    elif nth == 1:\n",
    "        r = 1\n",
    "    else:\n",
    "        r = Fib(nth-1, fibs) + Fib(nth-2, fibs)\n",
    "        fibs[nth] = r\n",
    "    return r"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "709d1ca7-1fad-407b-9fcb-441d89279aa7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "8"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Fib(5, {0:1, 1:1})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "4a9f3846-2664-4425-97b8-be8692df1804",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "7.75 µs ± 1.51 µs per loop (mean ± std. dev. of 7 runs, 100,000 loops each)\n"
     ]
    }
   ],
   "source": [
    "%timeit Fib(15, {0:1, 1:1}) # much faster with memoization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "3695de1b-9578-4733-a0c3-1fec5ce3b2b9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "573147844013817084101"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Fib(100, {0:1, 1:1})"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f1382f09-a12b-4c63-981c-9e92291f3c3f",
   "metadata": {},
   "source": [
    "Links:\n",
    "\n",
    "* [Fibonacci Numbers in multiple languages and algorithms](https://the-algorithms.com/algorithm/fibonacci-numbers)\n",
    "* [Fibonacci Numbers at OEIS](https://oeis.org/A000045)\n",
    "* [Recursion](https://youtu.be/IJDJ0kBx2LM) (Youtube: freecodecamp.org)\n",
    "* [Dynamic Programming](https://youtu.be/oBt53YbR9Kk) (Youtube: freecodecamp.org)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "alternative-professor",
   "metadata": {},
   "source": [
    "# Data Structures & Structured Data\n",
    "\n",
    "When you cook, following a recipe, you need ways to store your ingredients, as well as places to mix them.  Containers of various types come to mind: liquid measuring spoons, sized cups, bags and jars.  We also need cutting boards and other work surfaces.\n",
    "\n",
    "Algorithms likewise require lots of containers, for organizing all the ingredients in a numerical, semi-numerical, or non-numerical recipe.  We might call these containers data structures, or refer to their role as one of structuring (organizing) our data. \n",
    "\n",
    "To quote Wes McKinney, in *Python for Data Analysis*, 3rd Edition:\n",
    "\n",
    "<blockquote>\n",
    "    When I say “data,” what am I referring to exactly? The primary focus is on structured data, a deliberately vague term that encompasses many different common forms of data, such as:\n",
    "\n",
    "* Tabular or spreadsheet-like data in which each column may be a different type (string, numeric, date, or otherwise). This includes most kinds of data commonly stored in relational databases or tab- or comma-delimited text files.\n",
    "\n",
    "* Multidimensional arrays (matrices).\n",
    "\n",
    "* Multiple tables of data interrelated by key columns (what would be primary or foreign keys for a SQL user).\n",
    "\n",
    "* Evenly or unevenly spaced time series.\n",
    "</blockquote>\n",
    "\n",
    "For example, in the Fibonacci Numbers code, we stored the sequence in a Python list named fibs.  The Python list is a type of data structure.  What are some other data structures in Python?\n",
    "\n",
    "We will begin our study of data structures by using them within algorithms.  \n",
    "\n",
    "In an object oriented programming language, such as Python, we imagine objects as a combination of data, and algorithms for working with that data."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "featured-criminal",
   "metadata": {},
   "source": [
    "The Python type defined below allows for the creation of any number of individual Pythons through the \"birthing\" process, or `__init__`.  Each new Python starts out with a name and an empty stomach.  The eat method, defined at the class level (along with the other methods) is a way to append foods to the stomach object contained in each snake."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "public-conjunction",
   "metadata": {},
   "outputs": [],
   "source": [
    "class Python:\n",
    "  family = \"snake\"\n",
    "\n",
    "  def __init__(self, nm):\n",
    "      self.name = nm\n",
    "      self.stomach = [ ]  # <-- built-in data structure\n",
    "\n",
    "  def eat(self, food):\n",
    "      self.stomach.append(food)\n",
    "\n",
    "  def __repr__(self):\n",
    "      return 'Python named ' + self.name"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "ordinary-pulse",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Python named Shelly"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "snake1 = Python(\"Shelly\")\n",
    "snake1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "thirty-nursery",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['emoji']"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "snake1.eat(\"emoji\")\n",
    "snake1.stomach"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "digital-packet",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'name': 'Guido', 'stomach': ['eggs']}"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "snake2 = Python(\"Guido\")\n",
    "snake2.eat(\"eggs\")\n",
    "snake2.__dict__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "martial-science",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'snake'"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Python.family"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "tribal-release",
   "metadata": {},
   "source": [
    "Python's data structures may be classified into two major categories:  *sequences* and *mappings*. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "mounted-greek",
   "metadata": {},
   "source": [
    "### Sequences"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "taken-stadium",
   "metadata": {},
   "outputs": [],
   "source": [
    "# List\n",
    "\n",
    "L = [5, 2, 7, 0, 10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "hearing-metabolism",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[10, 0, 7, 2, 5]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "L.reverse()  # some algorithm knows how to reverse any list\n",
    "L"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "impossible-integration",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0, 2, 5, 7, 10]"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "L.sort()  # some algorithm knows how to sort some lists\n",
    "L"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "applicable-distribution",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "'<' not supported between instances of 'int' and 'str'\n"
     ]
    }
   ],
   "source": [
    "M = ['a', 10, True]  # but not all lists can be sorted\n",
    "try:\n",
    "    M.sort()\n",
    "except Exception as e:\n",
    "    print(e)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "stupid-legend",
   "metadata": {},
   "source": [
    "Any data structure that has a first-to-last ordering is a sequence.  A list is a sequence.  So is a string."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "urban-chorus",
   "metadata": {},
   "outputs": [],
   "source": [
    "# all of these may be indexed\n",
    "a_tuple = (1, 7, 9, 12)\n",
    "a_list = [1, 7, 9, 12]\n",
    "a_string = \"1 7 9 12\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "equal-homework",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "7"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a_tuple[1] # tuples are like lists but fight being changed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "coordinate-contamination",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "12"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a_list[-1] # negative index:  coming from the right"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "artificial-paper",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'1 7 9'"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a_string[0:5]  # slice notation, still indexing"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "failing-kitchen",
   "metadata": {},
   "source": [
    "### NamedTuple\n",
    "\n",
    "![Periodic Table](https://upload.wikimedia.org/wikipedia/commons/thumb/f/fe/Periodic_table_large-es.svg/800px-Periodic_table_large-es.svg.png)\n",
    "\n",
    "The namedtuple type, importable from the collections library, allows us to access elements by name, not just by index number.  \n",
    "\n",
    "For example, consider an Element from the Periodic Table, such as [iron](https://www.britannica.com/science/iron-chemical-element):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "numeric-beijing",
   "metadata": {},
   "outputs": [],
   "source": [
    "from collections import namedtuple"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "textile-defense",
   "metadata": {},
   "outputs": [],
   "source": [
    "Element = namedtuple(\"Atom\", 'protons weight symbol name category')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "demographic-party",
   "metadata": {},
   "outputs": [],
   "source": [
    "iron = Element(*(26, 55.845, \"Fe\", \"Iron\", 'transition metal')) # the star \"explodes\" the tuple into individual arguments"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "dutch-avatar",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "26"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iron.protons  # named attribute"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "involved-sheriff",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Fe'"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iron[2]       # symbol"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "successful-olive",
   "metadata": {},
   "source": [
    "### Numpy Array\n",
    "\n",
    "The numpy array is the workhorse of array-based computing throughout business and industry.  Computing is array-based, or \"vectorized\", when all elements may be given the same treatment, as inputs to some operation, without explicit looping and/or indexing.\n",
    "\n",
    "For example, lets import the numpy library and take advantage of its powerful random number generating capabilities."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "occasional-privacy",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "divided-banking",
   "metadata": {},
   "outputs": [],
   "source": [
    "rng = np.random.default_rng()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "improved-creation",
   "metadata": {},
   "outputs": [],
   "source": [
    "# rng?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "separated-condition",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[87, 88, 97, 49, 21,  0, 31, 23, 76,  9],\n",
       "       [91, 58, 18, 72, 33, 83, 18, 35, 77, 66],\n",
       "       [29, 82, 82, 13, 32, 48, 44, 49, 53, 50],\n",
       "       [99, 46, 11, 53, 15, 16, 83,  8, 12, 10],\n",
       "       [44, 37, 10, 37, 95, 78, 37, 27, 86, 47]])"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "arr_2d = rng.integers(0, 101, 50).reshape(5, 10)\n",
    "arr_2d"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "unable-harvard",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5, 10)"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "arr_2d.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "considered-hybrid",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([70. , 62.2, 43.6, 44.8, 39.2, 45. , 42.6, 28.4, 60.8, 36.4])"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "arr_2d.mean(axis=0) # averages for each row"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "diagnostic-kingdom",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([48.1, 55.1, 48.2, 35.3, 49.8])"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "arr_2d.mean(axis=1) # averages for each column"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "contained-diagram",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 3,  4,  1,  1,  9,  0,  7, 11,  4,  9],\n",
       "       [ 7, 10,  6,  0,  9, 11,  6, 11,  5,  6],\n",
       "       [ 5, 10, 10,  1,  8,  0,  8,  1,  5,  2],\n",
       "       [ 3, 10, 11,  5,  3,  4, 11,  8,  0, 10],\n",
       "       [ 8,  1, 10,  1, 11,  6,  1,  3,  2, 11]])"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.mod(arr_2d, 12)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "whole-activity",
   "metadata": {},
   "source": [
    "### Mappings\n",
    "\n",
    "#### The Python Dictionary \n",
    "\n",
    "A mapping holds data, perhaps key:value pairs, as in a dict (dictionary), but has no special first-to-last order such that indexes might be used."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "meaningful-trance",
   "metadata": {},
   "outputs": [],
   "source": [
    "bio_data = { \"Leo Fibonacci\": (1170, 1250),\n",
    "             \"Al Khwarizmi\": (780, 850),\n",
    "             \"Al Gore\": (1948,)}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "reflected-bennett",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['Leo Fibonacci', 'Al Khwarizmi', 'Al Gore'])"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bio_data.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "transsexual-ordering",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_values([(1170, 1250), (780, 850), (1948,)])"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bio_data.values()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "celtic-abraham",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_items([('Leo Fibonacci', (1170, 1250)), ('Al Khwarizmi', (780, 850)), ('Al Gore', (1948,))])"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bio_data.items()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "ahead-regular",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Not Found'"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "default = \"Not Found\"\n",
    "bio_data.get(\"Fibonacci\", default)  # get method allows a default"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "vocational-october",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Not Found'"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bio_data.get(\"Joe\", default)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "emerging-promise",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(780, 850)"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bio_data[\"Al Khwarizmi\"]  # direct access using the key"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "casual-property",
   "metadata": {},
   "source": [
    "We will talk a lot more about the fine points of Python's data structures as the class progresses.  \n",
    "\n",
    "We will also look at data structures we need to build ourselves, but from built-in data structures."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "injured-intent",
   "metadata": {},
   "source": [
    "#### Notes:\n",
    "\n",
    "1. Politician Al Gore took credit for \"inventing the internet\" which many considered too much of an exaggeration, leading to many jokes at Al's expense.\n",
    "\n",
    "2. Fibonacci Numbers are derived by adding the previous two terms to get the next term, starting from 0 and 1. The closely related [Lucas Numbers](https://oeis.org/A000032) start off with 2 and 1 instead of 0 and 1."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "naughty-navigation",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2, 1, 3, 4, 7, 11, 18, 29, 47, 76]\n"
     ]
    }
   ],
   "source": [
    "gen = fibo(2, 1)  # reusing the same generator, changing defaults\n",
    "lucas = [next(gen) for _ in range(10)] # OEIS A000032\n",
    "print(lucas)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "secondary-basin",
   "metadata": {},
   "source": [
    "#### The Python Set\n",
    "\n",
    "The Python set is a lot like a dictionary, except it only has keys, no values.  Like with a dict, the keys must be unique, and therefore immutable.  If keys were allowed to change in place, like lists might, then originally unique keys could be made the same, and that would corrupt the entire data structure.\n",
    "\n",
    "Sets themselves are mutable, however if you \"freeze\" a set by turning it into a \"frozen set\", then it too may be used, like a tuple, like a string, as a set element and/or key."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "original-crown",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# a set comprehension\n",
    "\n",
    "with open(\"wordkeep.txt\", \"r\") as the_file:\n",
    "    five_letter_starting_x = {word[:-1] for word \n",
    "                                in the_file.readlines()\n",
    "                                if \"x\" == word[0]}\n",
    "    \n",
    "    the_file.seek(0) # move pointer back to the top\n",
    "    five_letter_starting_y = {word[:-1] for word \n",
    "                                in the_file.readlines()\n",
    "                                if \"y\" == word[0]}\n",
    "\n",
    "# file closes when with-scope ends\n",
    "len(five_letter_starting_x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "legal-basket",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "47"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(five_letter_starting_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "alleged-netherlands",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['yearn',\n",
       " 'yogic',\n",
       " 'yella',\n",
       " 'yecch',\n",
       " 'yoked',\n",
       " 'yawls',\n",
       " 'yucky',\n",
       " 'yards',\n",
       " 'yummy']"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(five_letter_starting_y)[30:39]  # sample the set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "distinguished-suite",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'xenon', 'xerox', 'xored', 'xylem'}"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "five_letter_starting_x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "suspended-hartford",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{frozenset({'xenon', 'xerox', 'xored', 'xylem'}): 'must start with x'}"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_words = frozenset(five_letter_starting_x)\n",
    "test_dict = {x_words: \"must start with x\"}\n",
    "test_dict"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "neural-porcelain",
   "metadata": {},
   "source": [
    "### The pandas Series\n",
    "\n",
    "pandas is the premier Python data analysis library, although it's not alone on stage.  \n",
    "\n",
    "pandas shines when used with additional libraries and packages.  It depends on numpy and matplotlib.\n",
    "\n",
    "Inside of pandas are the Series and the DataFrame types.\n",
    "\n",
    "A Series is like a one-dimensional numpy array, but potentially with string labels for indexing, making it more like a namedtuple or dictionary.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "collected-nightlife",
   "metadata": {},
   "source": [
    "### The pandas DataFrame¶\n",
    "\n",
    "\n",
    "A DataFrame is an arrangement of Series, left to right, serving as successive columns and sharing a common row index, i.e. a tabular, 2-dimensional format."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "frequent-transcript",
   "metadata": {},
   "source": [
    "Resources:\n",
    "\n",
    "*  [Official Python Docs on Data Structures](https://docs.python.org/3/tutorial/datastructures.html)\n",
    "*  A whole Github repo devoted to [Python Data Structures](https://github.com/okeeffed/cheat-sheets/blob/master/Python-Data-Structures.md#data-structures-with-python), by Dennis O'Keeffe \n",
    "* [List, Dict and Set Comprehensions (Notebook)](List_Comprehensions.ipynb)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "criminal-poverty",
   "metadata": {},
   "source": [
    "#### Collectible / Recommended Books:\n",
    "\n",
    "<a data-flickr-embed=\"true\" href=\"https://www.flickr.com/photos/kirbyurner/51836242767/in/dateposted-public/\" title=\"Old Fave\"><img src=\"https://live.staticflickr.com/65535/51836242767_e9bf8daa1a_w.jpg\" width=\"300\" height=\"400\" alt=\"Old Fave\"></a><script async src=\"//embedr.flickr.com/assets/client-code.js\" charset=\"utf-8\"></script>\n",
    "\n",
    "Used at [Phillips Academy / Andover](https://www.andover.edu/).  Introduces the cryptographic algorithm named [RSA](https://github.com/4dsolutions/elite_school/blob/master/RSA.ipynb).\n",
    "\n",
    "<a data-flickr-embed=\"true\" href=\"https://www.flickr.com/photos/kirbyurner/51837924600/in/dateposted-public/\" title=\"Sigma\"><img src=\"https://live.staticflickr.com/65535/51837924600_93b7a663be_w.jpg\" width=\"300\" height=\"400\" alt=\"Sigma\"></a><script async src=\"//embedr.flickr.com/assets/client-code.js\" charset=\"utf-8\"></script>\n",
    "\n",
    "Used at Stanford University to help make *The Art of Computer Programming* more understandable."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}