smd_supervised.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Machine Learning\n",
    "\n",
    "\n",
    "> __People worry that computers will get too smart and take over the world, but the real problem is that they're too stupid and they've already taken over the world.__\n",
    "> \n",
    "> Pedro Domingos \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Do not delete (math definitions here)\n",
    "\n",
    "$$\n",
    "% vectors and matrices\n",
    "\\def\\v#1{\\boldsymbol{#1}}\n",
    "\\def\\m#1{\\boldsymbol{#1}}\n",
    "\\def\\p#1{\\mathtt{#1}}\n",
    "% confusion matrix stuff\n",
    "\\def\\TP\\mathit{TP}\n",
    "\\def\\TN\\mathit{TN}\n",
    "\\def\\FP\\mathit{FP}\n",
    "\\def\\FN\\mathit{FN}\n",
    "% statistics \n",
    "\\DeclareMathOperator\\mse{mse}\n",
    "\\DeclareMathOperator\\E{E}\n",
    "\\DeclareMathOperator\\Var{Var}\n",
    "\\DeclareMathOperator\\Cov{Cov}\n",
    "\\DeclareMathOperator\\Bias{Bias}\n",
    "\\DeclareMathOperator*{\\argmax}{arg\\,max}\n",
    "\\DeclareMathOperator*{\\argmin}{arg\\,min}\n",
    "$$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:45:23.171492Z",
     "start_time": "2018-11-22T18:45:23.165092Z"
    }
   },
   "outputs": [],
   "source": [
    "%%html\n",
    "<style>\n",
    ".reminder-div\n",
    "{\n",
    "    margin: 0 auto;\n",
    "    width: 66%;\n",
    "    padding: 5pt;\n",
    "    background-color: linen;\n",
    "}\n",
    "</style>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# What machines can (and cannot) do\n",
    "\n",
    "\n",
    "## A Short History of Computing\n",
    "\n",
    "Around __100 B.C.__: Antikythera mechanism\n",
    "\n",
    "\n",
    "<img width=\"45%\" src=\"https://upload.wikimedia.org/wikipedia/commons/6/66/NAMA_Machine_d%27Anticyth%C3%A8re_1.jpg\" />   \n",
    "\n",
    "Build in ancient greece to predict the orbit of the moon and sun. Some suggest it can even predict the occurence of eclipses.\n",
    "\n",
    "Around __1690__ : Staffelwalze\n",
    "\n",
    "Leibniz build his mechanical calculator for simple arithmetics\n",
    "\n",
    "<img width=\"45%\" src=\"https://upload.wikimedia.org/wikipedia/commons/6/6f/Leibniz_Stepped_Reckoner_drawing.png\" />\n",
    "\n",
    "> .. it is beneath the dignity of excellent men to waste their time in calculation when any peasant could do the work  just as accurately with the aid of a machine.\n",
    ">\n",
    "> *Gottfried Leibniz*\n",
    "\n",
    "\n",
    "In __1938__: The Z1. A universal mechanical computer. \n",
    "\n",
    "<img width=\"45%\" src=\"https://upload.wikimedia.org/wikipedia/commons/6/6f/Zuse_Z1.jpg\" />\n",
    "\n",
    "Maybe the first truly programmable computer with a memory and a processing unit.\n",
    "\n",
    "> There is a replica of this Model in the Museum of Traffic and Technology in Berlin. Back then it didn't function well, and in that regard the replica is very reliable -- it also doesn't work well.\n",
    "> \n",
    "> *Konrad Zuse (1910 - 1995)*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Formal Description\n",
    "\n",
    "<img width=\"45%\" src=\"https://www.iwm.org.uk/sites/default/files/styles/text_with_media_large_desktop_1x/public/2018-01/npg_mw165875.jpg?itok=BMOpWRFl\" />   \n",
    "\n",
    "\n",
    "Alan Turings paper from 1936 brought the first formal description of what a *computing machine* does.\n",
    "Before, there was no knowledge of what *exactly* computers where able to do.\n",
    "He came up with an abstract, mathematical model for a machine performing a computation.\n",
    "Essentialy he (and Alonzo Church independently so) wrote down the first definition of what an *Algorithm* is. \n",
    "\n",
    "<img width=\"45%\" src=\"./images/computable_numbers.jpg\" />   \n",
    "\n",
    "\n",
    "> __ON    COMPUTABLE    NUMBERS,   WITH    AN   APPLICATION    TO THE   ENTSCHEIDUNGSPROBLEM__\n",
    ">\n",
    "> SECTION 9\n",
    ">\n",
    "> Computing is normally done by writing certain symbols on paper.  We\n",
    "may suppose this paper is divided into squares like a child's arithmetic book.\n",
    "In  elementary  arithmetic  the  two-dimensional  character  of  the  paper  is\n",
    "sometimes  used.    But  such  a  use  is always  avoidable,  and  I  think  that  it\n",
    "will be  agreed that  the  two-dimensional  character  of paper  is no  essential\n",
    "of  computation. \n",
    "\n",
    "__The Church-Turing Thesis__\n",
    "\n",
    "> Every effectively calculable function is a computable function.\n",
    "\n",
    "Using Turings definitions of computability\n",
    "\n",
    "> \"We shall use the expression 'computable function' to mean a function calculable by a machine, and let 'effectively calculable' refer to the intuitive idea without particular identification with any one of these definitions.\"\n",
    "\n",
    "\n",
    "To put it into somewhat more understandable words:\n",
    "\n",
    "* If something is intuitivly calculable (in whatever manner you can think of), it can be computed by a machine. \n",
    "\n",
    "* If something can be computed by a machine, it can be computed by a brain. (given enough time)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-15T17:17:08.234097Z",
     "start_time": "2018-11-15T17:17:08.210065Z"
    }
   },
   "source": [
    "## Artifical Inteligence\n",
    "\n",
    "AI is a large research field with various subcategories, philosophies and motiviations.\n",
    "AI encompasses many things including but not limited to:\n",
    "\n",
    "* simulating human behaviour\n",
    "* playing chess\n",
    "* solving optimization problems\n",
    "* assistive technologies \n",
    "* autonomus driving\n",
    "* defense\n",
    "* cyber security\n",
    "* computer games\n",
    "* computer graphics\n",
    "* statistical learning (← this is what we do)\n",
    "\n",
    "The idea of a *mechanical brain* or *living automaton* is not new. In fact its quite ancient.\n",
    "\n",
    "Greek mythology speaks of [Talos](https://en.wikipedia.org/wiki/Talos). The mechanical God and protector of Europe.\n",
    "\n",
    "\n",
    "The mechanical turk. A chess playing *machine* from 1770\n",
    "\n",
    "<img width=\"45%\" src=\"https://upload.wikimedia.org/wikipedia/commons/8/8b/Tuerkischer_schachspieler_windisch4.jpg\" />   \n",
    "\n",
    "<br/>\n",
    "<br/>\n",
    "\n",
    "The term *Artifical Inteligence*, as we understand it today, was coined by John McArthy for the 1956 Dartmouth Conference. \n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "<div>\n",
    "  <div style=\"float: left; width: 70%; padding-right:5em;\">\n",
    "  \n",
    "  __Are computers just as good as human brains then?__\n",
    "\n",
    "\n",
    "<p>They seem to work well together.</p>\n",
    "\n",
    "<p>The first computer assisted proof of the Four-Color-Problem in 1989: </p>\n",
    "\n",
    "<blockquote>  ...part of the proof was verified in over 400 pages of microfiche, which had to be checked by hand with the assistance of Haken's daughter... </blockquote>\n",
    "\n",
    "\n",
    "Easy problem for machines:\n",
    "<ul>\n",
    "<li>  Is the text in the image written in the english language?</li>\n",
    "</ul>\n",
    "Impossible problem for machines:\n",
    "<ul>\n",
    "<li> What does this text mean?</li>\n",
    "</ul>\n",
    "\n",
    "A few things to keep in mind while talking about artifical intelligence\n",
    "<ul>\n",
    "    <li> \n",
    "    Logical reasoning (within some limitations) is possible for machines as long as all relations and symbols are strictly defined \n",
    "    </li>\n",
    "    <li> \n",
    "    Mapping syntax and symbols to semantics and objects in the real world is something inherently human (or intelligent).\n",
    "    </li>\n",
    "    <li> \n",
    "    There is not really a powerfull formal equivalent for human creativity.\n",
    "    </li>\n",
    "    <li> \n",
    "    Machines cannot \"learn\"\n",
    "    </li>\n",
    "</ul>\n",
    "\n",
    "  </div>\n",
    "  <div style=\"float: left; width: 30%;\">\n",
    "      <img width=\"100%\" src=\"data:image/jpeg;base64,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\"/>\n",
    "  </div>\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "Interesting reads which lead you down the rabbit hole of computational theory, logic and complexity theory.\n",
    "\n",
    "https://en.wikipedia.org/wiki/First-order_logic\n",
    "\n",
    "https://en.wikipedia.org/wiki/Entscheidungsproblem\n",
    "\n",
    "https://en.wikipedia.org/wiki/G%C3%B6del%27s_incompleteness_theorems"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Machine Learning (Supervised)\n",
    "\n",
    "\n",
    "## Mathematical Notation and Problem Description\n",
    "\n",
    "*For more details see \"Elements of statistical Learning\" by Trevor Hastie.\n",
    "(Available for free as an E-Book)* \n",
    "\n",
    "I'll try to follow some naming conventions along this notebook.\n",
    "They are the same as in the book (for the most parts).\n",
    "\n",
    "* Uppercase letters such as $X$ or $Y$ denote generic aspects of a variable (i.e. the actual random variable)\n",
    "* Observed values are written in lowercase. The ith observed value of $X$ is written as $x_i$\n",
    "* Matrices are written in bold uppercase letters as in $\\m{X}$\n",
    "* Observations map as *rows* in the matrix while the observed variables are the *columns*.\n",
    "\n",
    "So if I measure two observables $p = 2$ the size and weight of $N = 100$ people, I get an $N \\times p$ matrix $\\m{X}$.\n",
    "\n",
    "One observation in that matrix is denoted as $\\v{x}_i = [\\mathrm{size}, \\mathrm{weight} ]$,\n",
    "while all observations of the variable weight are denoted by $\\v{x}_{\\bullet 1}$, analogous to the numpy indexing `X[:, 1]`\n",
    "\n",
    "Heres one possible definition of supervised machine learning:\n",
    "\n",
    "> Given a $N \\times p$ matrix $\\m{X}$ and some associated output vector $\\m{Y} \\in \\mathbb{R}^N$,\n",
    " find a function $f(X) = \\hat{Y}$ that takes a vector $X \\in \\mathbb{R}^p$ and returns a prediction for $\\hat{Y}$\n",
    " where some \"loss function\" $L(Y, f(X))$ is minimized for all $X$.\n",
    " \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "### The Titanic Example. Learning from disaster.\n",
    "\n",
    "In the spring of 1912 the R.M.S. Titanic embarked on a journey to cross the Atlantic ocean. Unfortunately it hit an iceberg on the night of April 14th and sank shortly afterwards.\n",
    "\n",
    "The disaster caused widespread outrage over what was seen as lax safety regulations and reckles behavoiur by some. New maritime safety laws were put in place after the sinking that are still in place today.\n",
    "\n",
    "What can _we_ learn from the Titanic just by looking at its passenger data?\n",
    "\n",
    "Our data contains a list of name, gender, age and ticket price for each (known) passenger.  \n",
    "\n",
    "<img width=\"70%\" src=\"data:image/jpeg;base64,/9j/4AAQSkZJRgABAQEASABIAAD/2wBDAAYEBQYFBAYGBQYHBwYIChAKCgkJChQODwwQFxQYGBcUFhYaHSUfGhsjHBYWICwgIyYnKSopGR8tMC0oMCUoKSj/2wBDAQcHBwoIChMKChMoGhYaKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCj/wAARCAGNAk4DAREAAhEBAxEB/8QAHAAAAQUBAQEAAAAAAAAAAAAAAwECBAUGAAcI/8QAUBAAAQMDAwIEAwQHBQUFBwIHAQIDEQAEBRIhMQZBEyJRYTJxgQcUkaEVI0JSscHRJDNiovAWcoLS4SVTksLxQ2ODk7Kz4nOjJjQ1RVTD0//EABQBAQAAAAAAAAAAAAAAAAAAAAD/xAAUEQEAAAAAAAAAAAAAAAAAAAAA/9oADAMBAAIRAxEAPwDWpAIMCDFAoRAMkHbiaCs6plHS+VIO4tXI9vKaCHcu5NjB4Y4xsuvKQgOgJBAT4R9feKCLZ5m/TdP2Nw40q8tmrXUrSNWtSkh2QP8AeEUFk1lLxOOxTq2A87dPFtakAAIHmgxPsPzoAMdTMN462cvkOpfUE+JDYESkErifh3Hv7UA85mb6xzDlqksJtloaS04pO6XVK4PsQFR70Fg31JjXG3lhTgQ3q8yk7EJgyN+87fWgOzlrZ/IptW0PFa5OsI8uxI594MHigqbrJ5qzfygdXYPNWjaHAEsrCiFEwOTxG/rQWV9n7Fu1JPiqJStYbLShq0zqTuOfKdqB1tmG1qQbtHgtvFH3byK1KBSknVttClATxQSL1pIymLkBKvEcInvDZFBZwD229qDo9oE9xQZjqjCtXmbx2Rv2kP461bcQ82pQSlJMELMxI9R8tqCDmcL0ebAXV9btWCHBqSoEtOKHskb79tqDOtdKYu9wj2YxtzeC1aKtLb6EydKooPQOn+nLLAod+6IUp10jxHXDKlD6dqDA/Z62211teIbUYT4+x/3veg9RklWwJ7b9qAqJCR5QkD0oOWJ2SJI5J7UHaSFEEkz6UEe+s2720ftn9fhvJKF6TpOk87ig8t676WYtbl9zE2jTVvbWrbjqEj1WoavyE/jQTOhstc5fp13BNXybW9Sf1TijKixPmSn/ABATHoPlQSGug7Nm/ZsXLhxbNw248kKSNTRTpnfv8XMfSg3mJx1vibJuzsWkoYQO/Kie5NBLjzCByOR2oAG7tk3qLRTg+8lovaB+4DE/jFAdJlJKT+J5oFEagAIMc0HDYjYxPbtQOSiCCY9IoPFftZ6Ux+AxWNdsQv7w8+tDzq1Sp0nzSe23tQabAuX1x0j0etlKHbtLq20Fw+VMIcSFn1gdu/FBgHbW/wAR9prjDN+s3qroNpu1iT+sEaiPkqg9u6cwttgcaLKzU6uVlxx1wytxZ5Ur50FqjcknY9570DwkBQkD69qBFaQoGPioEfALCwCUjSeO1B5B9h5W/ncy+pS3HSwCpxZkqJcmSaD2ZMAK2MGgWEmUxG3Yc0FT1SxdXPTmTax5ULpy3Wlojkkjj5nigx1p9nHTNxjra3S7covHGQsq8SHD2USgjYTtQUd59krjTpbxOTtXnknVoflC0pO0wJ9oO1BbdH9E3V3dW9/ncmu6bsrhYatw4paQtCtMyr3Ex8qCdb5vqE/aa7YG0WcYlOhTaeEtncPavXtH0oPQUk6pA7djzQP5IgcnmaBunmQIHFAokcEAREepoFSZSSRpI7UHNNpCBAA3nig4p2JG8+nag5QJdCYBHI+dA8J9APSDQNLQWsKVuoHaRxQFUIQDxG21A1CeSBO80FB07j7ZLNzcli38dd2+oOBtOqPEUOeaC9LcAAAARtAoKheYt02K7u6adtWUPFo+JpHmE+h9iKCSzkWri+RaIQvxPBS/J0wEnjvP5UC5XJsY1TQuAuHOClMgbgf+YH5TQK1kGHb9FqlC/EW0HdwICT9Z7UEVrNMrtbZ+3trl1p99TI0IEpUFESoEjbY/SgM5mbNNxdsaiXbVC3HUpjYJj+oigQZy2FxYMKQ4HL0am4CTAHrBP5TQWTilbEhIPpNBnEDeDE8/KgcZBJABPoKCD1K2g9P5ULAM2rgO3bSaCZZI8O0t43AbQI+lBBGBs0veN+sLxcU4XCrzKKlJVB9gUiB2FAYWNs0xa2gfKfAVrbBcGvv+W5oBu4C1d0buogJEpUN0hITp44IA4396CRd4i2unnXHwtXiFo6SdkltRUkj6mgjp6dszbBhwuraS5rSFEQkQQE7Djc+/vQOZwTDF3YOIddCbNsIbQY3gETMat53EwTvQEucQh96/cLriVXbbbagAPJoJ4+cmgg3nTjb763vvbiRDsoKAfj1zB7fGfnAoDXGDdvbbHJcvQly2CSpwMiVERunfynaPSDxQWF4NWUx0cAuq+XloJyEzO3HrQcEn6UHnv2lXJusvhMKpa0W1w6lx7SdyCrSP5mgd9o2MtMf0uoW7XncuG0OPuErcUN/iUZPp7UA8Yp1j7L1t27anXXHVNpCRP7fPsNjQb60e+9WrDxQtorSNSHE6VJPcEfjQeTdI/eEdYZFWPaS9dHx0p1mEt+fZSvb5UDen+o83iuoX2Mk25dLuHIXbqVCkuHYaT2HHtG9Bvsvmb7AdMqyGTZZeu0rALTJISnUfKJPMbSe9BiL/ADPUDn6LzpfUllZUttpA0sgiQpCvWQO9B6RgsqzmcYxfW6VpbeEaViCkjkf9e9B2ayKMTiby+dSVJt2yvQD8XoPxigzPRVxfZnJZRecZgO2raEoLZQlTairieeSJoMTkbVzo3qZLTaFuC2dFyySI1Nn3/FJoPUm7tq+zWIurY62XLJ5aT6pJboIXXXVJ6Zt2i3aF955JUnUvSlMRz3PPagyjHVGWw3V09RXKRbOpAW0kQhCFbpWge3fud/Sgk/aam+x99aZ/F3K0oW2ljWweOSPWQaAuU6uy6ejbW7tbZSHnAG371bf6ttfsn39eAduaAWR60zF70mq7x2OWy6hXhXNyANLR9UDn6nj3oJ/RnUWb6lFn4CLe3ZtHIvn1bl/0SlP7M8k/hQegITBBiN44oPAvtD6lyuecWzd233fDtXS0MuBk7qTI+Luedu1Bs+nL1Fh0R0o+PDWhLroh5wNgiHP2jsD/AOlBibu7VkPtat7gpbHiXzJ0tuBxMDSPiGxoPa87ncdgbdL+WuUsNrPl2KlLI3gAUHn/AFJ9oWWx+Uxl0zjVt4Z5vxEBwiblKu8idJHIH40GvsOrmshbvX1lZXS8Y02CXtB1OPEgBtCe/ME8A0GTx32i5NvMZW1y2JcU4gq8C2aH6xspHwK9fUnt8qC06C6wy3UBWxeYgpQnWXLxJKG0iDA0kbq4Gx43oM39juRatsjmX7xbbNsxaplUdg5+Z9PXigv2OsOqsjlruyx/TykoWvSy5cNqR4A41LPB9fy3oPSbRtbbDQeXreSgBakiApUbn60EHqHJt4PCXuSdTqRbNlYR+8eEj8YoMf0lY3mW6cezORyLnjZBK3ym3HhmE6glJX8WkdgI+tBQ/YOSrKZhS1KW6ptsqcWZnzH1oN10ZlrV3BZe68UKatr+5U4UjVCdZUDAnsaDLv8A2mP2PWS7PLYx21xrjaQgFEvid0rIHr+6Nx85oNLfdTXlviDcHC3ar59SvulkkanFoAB1uRsj1ImeKDN9F/aSchjrlnJ27r2ZZJU2zbNEm4T6D92O89t6BehftIOTcvbbMtLF3rUu2bYQVqWCf7oAftDse/figFgvtNuX85dsZXFvJt1kizYZRqeCxtoV7n17H2oN9d563xWDayWfAsAoeZlStakr/cEfEflQYKx+1UK6sSxfWirTDOpCW1rHnSSdnFf4T7cc0Hpt/f2lhZG8vblpm0ACi8tY0weI9Z7RQeU5r7T3X8o23ZYZ27wi0q1+KhSVXKeNSCOAIP8AOKDd9ALuX+lrO4uULaDqlustLUVKbaUolCSTuYEbn2oNQiJEcUDlCQYO0d6AaUyDpUdxyOaCs6fT/wBmpIKlS66d+/6xVBYkJChqBlJkEjiaCKMdbqYU2tBKFOqd0lX7Spk/maDkY5hq6TcoSsPJbDQAcVp0jYSmYPfeKCPlU29y4hm7x11c6AYKG5SNSYO8jtNAEKaF8bk2GUS4G9BKUmCBP7OqO596DrVmxDLNszjLthlLwcSPCICV8SYP40E53F2y1XBU3K30rQ4QYlKoB/gKBreKtwLZRC1Kt0wgrWVEbzuTzQGs7VFrbIYaK9DYhJWsqUR7k7mgpAnftQKYSSDsONu9BDzcHB3+pXhjwV+aeNqCVZkKsbdQ4LaP4UEg7AAAz7Cgyl8of7RZxehlSbbHIUdbOpUw5GlX8R3oJeNW4uxzqG7xwupKilGgSx+rBGn+VAPB5V5zJWlkp9LyPA8w0QoANoIcn3KiPp6zQQwHbe8yuRt1W7XhXa2dagoqUVFtMKGwgbke9BOevn7K+fDZYBS6VPEpP60gND12+M0DsclWQsxavZBLl02+47sfNpDikgGD6/0oL7HsfdbJi2C1L8JtKNauVQIk0BUJI8naNj60Ea7B/SlhseHe/GwoJwBJ9yN4oOJkGR279qDyv7WWnbTP43INk+ZrQkgcKSqf5igumnx19i/uynGrRlKkKf0r1PFSd/KngJ533oLi/wAZbYbpJyysysMIUgStUmVOJkk/U0FpYZO1vlXK7V0OsW7mlbwPkKhuoA942k8UHmf2ZPoV1dfLkAPturRPfzg0DepUk/adbok6fvFvqA4Pw0Gw+00BXSL6SQkKdbExx5qAHSFinJdNWib51D1q2SfuyRsVaiZX/EDYfOg1bRbKShkoIQdJCTsmO1Al3aM3tuu3umkusubKSRzFBEtFIX1JeQ4hRFoylICwSPMudvwoM79oOOYz+Cu3bJSLi/xijIQZIjdSD7xBoKf7Kry5uL9Nu6B92Yt3FskAiNSkah+Ux70Aftocm9xCRCkhDiiCON00GwTgUZp2yvs2wwoNNBNvbpGoAEAytX7R9uB70Fb9rJ8HpAISlKUm4QlISOICjx9KCraV4H2NBUpMp77zL1AuAcA+zLLuoIDZ8eIPOwFAX7GtJxuVKQQhV0nTI/w0Ho7ROoET9D3oPNft1DbXTmMbDSUA3SlaUiADpJP4zQTOhG23ulOkfEaSsB51YCgDBCXINBjuoh4f21MpS2kD74wdh3ITNBoPt1UB0/jREEXRG3YaDQL1e0kfY9jiUpUpDVsUyOJigvfsmAHQlluILjpj/jNBibYkfbusAEf2pW//AMKg9kuBpYdWBGlCjv6RQfPHRfTR6gt82pm4WHbVhLzbaBs6ZJhX4GPeg9I+xnqV/IIvMRdFbibVPjWylnfQTBSfkYig9OHf2E0GP+1lpb3QmQ0D4C2tQA5SFiaDPfZl1A5d9MrwzVmt6+twtCYUlKfDVwok+kngE0F59n3RCul2rtV1ci5dvEpbcQhEISBPHfvQW3TbFhYt3LqWmLd129et0lI0lcOHSmB6bx6fKg816wPifbfihPwrtgY/Gg9pUPIpU7R+NB4x9hSdXUWaeO6y0BJ9PEoO+z0Ff2vZkkjZVzuBx56AbQn7filMAC5J8p5PhUGm+3UaOkbUkCRdpH+VVBBt+l2OsOkunG0MO27ltbIQ5fLTEp7tpSfj9jsB6nign/aR04ux+zNvG4hq4uGbN1DitStawkTKvz4HFBhrbqLPs/ZvbW9pjlNY23dLbt4RIdQVToj0JkGOeO9B7L0Pn09S9N22QFsm2UCWlNpMpCk7be3pQXwAHHAoCkxwIoGpTBTB4MUFJZN3AxNobUpEOa1eaJSVKnt70EY3GQWlBtclYKaWdSVLG8Ep0/WCeeZFARDeUtVpVd5C2SI83iEQR5Z2gd5/GgkWF0wizDV9fMrfWVqUpLuwBKjCTtwAflFANCMjbW7Sk39sLdtI8Vx5RUoqCvNufbagAb2+bDTq8jj1J8iV6VbKmfhHYnbkxAoJlrfW7LJXdXdsSQkkoXMqIGqAN+6fxoAXCHzcqvLXKsi2dSrwEOKlOsDsZ4Gk7fOgC7dXjZdWrK2KQuSkEaglMAD5b7/WgjXN5esvpduspa+ENSAhqd5VIkAcgCJ9zQSAfKIAk+1AqYKwBuBzBoKjrTWOksupogOfdVkE/n/OgsLO6txY2yVXNuCG0Dd1PoPegOby2JB+9W4I2/vU7fnQKi6tFbi4t59nE70D03FuoSH2eezgoF1tKkpdRv3ChQKA2pRSQgzvAjc0Di02r4koUTzIB/12oGt2zSHVOIaQlZ5UlIBM0BUoPMHmd6BdJnYEfMUEO52ydjtuEu8/JNBMIJTIEA+hoEVyNp37UELO4a1zmMcs71JLavMlQ5QocEUHl9/0P1Bj7sO2KU3KUiEuW69Cv/CSI+hNBYY7pDP5RIGbunmbYnzJdfLigB+6ASPxoPQG8TZtYg4tplKbHwy3oT3Sed/5+tBmei+jhhMjc3N3+tU0tTVmSdwg/tfMzH4+tBn7hleT+09XhCUJukEkfutgap/Cg2WW6cVnby5Rk7tZx4EW9uydOlUfGr94gzHagzOG6NzWOy1yw3fvWtqU6V3LSv71Psk9+09qDd4fD2eIsxbY9rQ2DKpVKlqPKlE8mgreul3KcG2zaPi1+83DbDj5MeElR5n8BQYxX2bX1pkLf7ldpWy45Llwk6Ftf4onf2g80HpmOxlti7Fu2skaWmx33KieVKPcneTQZfpvDO4jrC7YZaWMatgutLjygqUJTPqN/pQB6q6NvM1ZPXF1dpuMk2VFhtACG0t/uD3O2570EPoex6gusWLbI3FzZ4xpyUpUIeVB+AK5CJ+vYbUFp1b03kOo3bhD2QTb2jQCrRltPxORupw/iBHagxvTXS/Ut3jL/DXJVY44rJV4okKcG40exMSRtHvQG6cwvUzjGTwxtxa2jp8N9dx8APdTccztxt60G86O6YY6asXGGrh24U4rWouACCNvLHag0TYglIEJnsaDzD7dX2nMXjbVtxC7gPqWplJlYSU8kelBXdJdX4vF9O9PWz/3ly5slul5pthRIlKwmO25UKCH01jMt1R9oAzz9i7a2Sbjxyp0aQAkQlKZ5OwmKDSdb9N9QdXG5RDNpa2aibNlRlVyrutSv2REx+dBjMU51NlcS90iLRbxZdQdb3lNsEn4VE/s+n5TQeo4LEZLpnoxFjZBrI5BCjok6GwVq535CZJ9TQed9U9P9QYHqyxyzTpyN3dPhSXm2ol6N2yn0jj2oPT7HG5d/CZFrLX5TfXqVAJaALdpKYCUdzHc96DJfZD0nf4bKZO7v1eEhoqsy0kSHSIOr5DaPWaDP9G4DPW/WWRtsUs2i2S5bvXKkyG21GRAP7REEfjQe2YyxGOx7Ns29cP+EnT4j7hW4r3JNA7IWTOQtX7S5Gtl9soUPUEUHgee6YznRuVF00q4NuhX9nu7ft/vRwfWdjQWWN+0rqtaha27LF+/MJItlFR+iSBQb3orp/KWn3nOdR638ktLjjFmg+VgKOpQHYKVx7etBgOvsd1E1ncf1Q/ZBh95xAaZaHiFhSf7tC/Un8J2oPWcNb9QXGMvH8rcMW9/ctQxaoblu1MGJPKie+8Dig82+xrDZrHdR5ZbzSWbZpKre4LiZl0GU6fyJ9vnQD+zzC51r7S7u4eaQCy4sXrigQg69/J7nYj25oG9Q2GYxn2yN3uOsjduvOC4YRwlaSnSqVdo3k9qDSddYPqbqTCCwct7YuWh+8KeSdP3hZH920n0AJGpUSRQUv2a9U9TG1cwNpixdu2w8Jp54+Gm1Ho7tuBvA57UGl686gzFu21hsLZO32U8NK7x1htQQhJ5Ajgq37yBQYhWf6j6oxlz0rb4NlCvIkBhCmhbpSqYVqMfjB+dB6F0tjMn0Z9nzrK7VF/kGVOPJYYMzJ4k8xyY+lAv2VZ/L9QYi4cy9vpDbsN3GnSHByUx/h4mg3QSZBgRHeg5QJgjtxNBW4RGvDWajJlqYn1oAuYq1tGn3sdYMrugApKCsjUpIATv8gBQRHG7x5hIfwyFKS3BQp0Ep2B0z380c0DsRYa0LF/iWmHkToUgApVKRPB23KtvQUCNOZRy28G7xDSpTrVDuxX5jx23CfxoIf3G5eWlTuBtgAAUw5JB7/zoJOKt/vV2+3e4hthVskBlZ1QsqEqhXpsJ70COMXDjabZzBW6GbcBbWpwFIUUnUB7gmN/WgjW9pkAHXDgLVokT4fibqIJHPuAk+1Bc2eMsnrcB6yQyo7rb+KFCfQ/4j+NBAAHIFA7QNyNh8u9Bymw4FpcCVIUIKSJBB5FBATgsSkpjFWKTPa3T/Sgf+g8TwMZY/wDyE/0oHjDYwiP0ZZf/ACE/0oFXhcURBxliSR/3Cf6UDTgsURp/RlkR7sp/pQMPT2GP/wDbLTnkNgUDT05iJEY22G/ZPFAiOncOoApsWiD+6pX9aDj0xhwmE2ekD0dXt+dAT/Z3FpA0sOCPS4c/5qA1pi7S1eDzCF6gCkKW6pcA8xqJ9qCeTtxx6UCpgjtFA4pgCRImgcBBMEjtzQIpMExHMUDQIHlGxNBXZXHXl4R93yb9kgp0qS2hKp95IkUAMB09Y4RBNqhxx9Y877h1LV/QUFxpSSSJg770DiJMHbbvQKBAE7H3oGuW7Vywtq4bQ6ysaVIWJBHvQBx+MtMawpqwYQy3OooT3P1oJBBPqZE7DigVKYSd4nbmgVI8giQSOaBFAbCOdt6BCEiQNvpQNmCZG0bUCgkJ23PvQKCCDEHagVscRv2oPH+vWOpunOr8hmsQl8214lI8dDYdAAA8igQYgjagoGcj171A/psl5Iz/AN2jwUj5mB/Gg0ODPWHT/UeIYzV049bX7vhlhT3jbAbn2jmg9f5SCBQJEgkhI33V60DwJjmg4p3CQAADPyoFUgyYMj0igclsxsBuR9KAqUpSVKAAJiduaB28CdiO9BwG+w+VA4gK2I7RQcw2htJ0BCCe6QBQEU2NudjxPegGpIkhWk/PtQOEEBJEEevagWAJ7CJmOaBEjQYAEnc0DY3JgH39KBwnfYEetAxLTbQWtppCFuK1LKBGpXqfWgrziC3nl5K2vHmC+Ei5YSEqQ8UiEncSDG23NBagJ1EpSkKUfMoDdUcTQcdiIA2O1ByEBA0hCUpB2AERPtQPSoAHywYoBLPlVO4Imgyib7Gu4awtLq5uUuW+hRLdu5upIiN0EEc0DEOdNtMNBt26a8JOkFDLoV8Wo8J7mZoJRzeFU6lQdyBUlIRDdq9uADz5fc0EW5zmAu1hbv6VKw2G9aLa4TsFA9h/rigYnLdOltLYfyafDSE7sXAVA23OmgNdZTA3qkFVzkkmNMIZfSCBtxpoBjM9OtB21U/kUBxCkKJYfmD6HTM0DG8v05a2z9uX8w81cCV+Iy+5EbwCpO0+1ATHXWAYeQ7b3eVKJKktqbfKE7qPBTx5jtxsPSgnWebwuOCyy/dfrCNQNs4RIET8HpQNTEHcxEUFLdZm5t7p5pFjqQlWkLHi+b8GyPzoGpz16YSjGqKp4JcE/wD7dBIVl7tOnVjVTMwFr/5KBycvcqWYxq5B7lf/ACUDk5a6SJNisgjbzq2P/goHHMOmAqwXJHZZ2/FNBycwsKg2S/8A5g/nFA45pBVpNssKG8eI3/zUD05tsQk2zoE91tn/AM1BwzTKCZYeA+aP+agRzOWwUAUO/IFG/wDmoHJzNsoEhL0+4Tv/AJqAyMmwtIVpdEDgp/pQPGRYgAJf3P8A3KqB36QYUkkB/YTsyrf8qB7d8wqILgBHdtX9KAgvWQo6lrIIn+7V/Sgau+tSILpg+qVb/lQcL60Jg3CBG0EGgaq/tCARcsj6xNBycjaBM/emCkdysbUCDIWYIUbu39vOKB5yNnqBN5bzH/eJoHJvbUAf2pgH08RO9A9N3blJCblkb9nE7/nQPN00U+V1ok+ixQKXEwSFonnkbUCpKIkEGROxFBxWCQAQZ9TxQOABUDPI9KDinmDt/CgaUxwDQIQpJ2BHzFA0plIIEUB22yQCUmedqAqkEpESN54oI12x47C23CsJUIOlRSY9iIIoK7H4DG2l+b1m0/tcaQ84tTigDzBUTFBZlMxsZIoOUAlJCuT7UD0jywBFAv7XB+lA4cCDPagfMRsDv2oFJJJ24PcUDgIEETNA7SAkifMfWg7QowTIJoHoCQdxMetA9Wyjxz3oAgJ1KIEkbmgIEgQZ7etBxSkp4iR6cUHEApJO/wDKg4iQDG0dqBYBjiPSg5PmJkb8fKgQCJ7nigaslI5jeKB22niT7GgaoQATHznmgdqCQUkTvQNICjsADPrQKFEqBUSD7nigKFEg7xv2NAiiQT5jEzE8UDFEg8nbegcFKncztPNA1JUVE6iPagUSp0KJiduaB86kmVEDsZ4oF1FIglRMUAjKkiZn50GcSkaQQPwoHpSBAUZigqOpLHJXSbJWJu126kPjxtK9Ops7HsdxsR60FDaW/VzF80p9x66tmx52y82A65Jg+un4dqDW5BVw5h3yw3cIulN+RDak60qPG52/6UGayCeol3Q+6NZBln7qtP8AfNGXo8ivkd5oNLaPvItWUPW1yp1KAFKWUkkgex770EBZyn3h0j714ClSmEo1pEiRG8+x9JFBbWhdcskOPoU29phQI/aA3gfwoM4q/wA4kvJ8Fapb/UKXbaf1oJmQJ2I7+tAXH5HLm6bGQsXQxqIKkMcenH8fyoNKtps7KbQY9U0CC1t1Ayyzsf8AuxtQIbO2JP8AZ2TP+BP9KBTZ28Em3ZPbZA3/ACoGixtARNqwCOIbG00DTYWna1Z4j4BQP/R9qAB4CBv2FA79H2aolhE+0igacdaA7NnVEbLV/WgUWFvpiFgH/wB4vb86BDjmCAIcHydX/WgCcXb6tw9zEh5f9aBysZbhKDL8cbvL/rQIrEWygkldyI9H1f1oGnCWqiDqeA9S5M/jQVWat7HFeC4tt5fiFSRJTpSRxPlOx4nsYoK127xyLV99yzd8Np4sqhTZkjV/h/wmKC8YxLD1g3cJZcStbesMw0TuNhOmPSgpkXGOcYadFlcoDg1SpDMgEkJ/Z7qSoe3NAbHvY29uLK3atrgOXKFLGpLPk0khU7ex43oJeWxdrZpbUGrl8qVohhtqRO28ge1BWvpxzTbTimLvS6rQmUM/EdX/ACmgY3+jFa9IuylClIUvwW/LGnciJjzCgfeMY0AFsXdxFybQhhls6XB2Mxz2oLpPS7SSVffFpIEmW07UFe0Mb92YcbvrlbbzX3htSGz5kiZ4Vz5TQI01aPpZcZvb1SHiNJCF6RKinfz+qTQTX8Wm1tw89fvqaUpKQUoWfiMDhfG43oB2Vi3dsl61yLnhJcKNRSuCQY/f496BLhq2ZUtDl/cBaVFMaHZJCdX7++xHHNAe1sUXDjrdvkXytuCokOj4uOV/OgZcNJtHSxdZC5CkI8QlKXyNI35CzQK2ttQbU1lbhSHVFCYD26hyPioDNlooaKMo8pK1QCnxjJ9Pi+dA124QLJd4jI3Dlu0tKTpDs6lGByf/AEoJiJXo05UlJ22Wvedh/EfKaArKXH3CyxkiXUpkoK1TG2/5igcpDjdwLZzJhL6wCEFxUn/UGgaUupKEnKoHiDWP1qoKRvI+k0Cq1vrQ01lUlxY8ulZk/lQORYX6CSi9JPBlxUUDhjsiSqL8jbjxF0Dhj8kVFJvxB7eI5sPxoEcx+R2Kb2BEHzq3oObsMk2sk3iVgjguLoDt2V+kgi6BBHBWd/yoHKtb8E6blBBHdZ2/Kgam3yCp13CCD+6s/wDLQOW1fBSSHGSZgSo7x/w0CFnIAAeIzPqFn/loE8HIBILj7QEby5x/loIougWw6nL23hEA6hcJgfXT7GgehF3coDllkQ82owFoeSoH5EINBXXGXtmXEsOdR2SHZ3CrlvUP8lBMbuip8WzebtzcqISlsXDeozvxooJDlnliBpvVkkRutO3/AO3QRAm6Wt1SMwgpbBLifGT5YMGfJ6zQOSHFEIGbQFEyEm4QCPpooCtoeU2SnLlaUp1q0uJVA9dkUArdFyXm1N5lCw78Dan0Sqd9hooFculKC4zCGylZQqHkp3GxG6O1AZDF6GS8MilTJGpKw8CI9Z0UEDL3CsQGjk88LVTs6AXANQET+x22oFTMhQKh8u1BDy1jcXiWxa3JtloVKt1DVPbyqFBBbxGTSfNlCrbgF3/noJAxeRKtQyKSj0Hi/wDPQJ+hcgEFKcqUncBRLhif+OgVOGyaW0JGSCjEEkuif89A8Y/Kgafv6CCP3nP60CpsssNzetHfutf9DQP+6ZYEE3TZHss/8lByWMuAQHWSDvu5/wDhQIWs0FAhbJ/+LH/+ug4t50HyraIHbxE7/wCSgRP6cPlKW9zJOtB/8ooHuOZoKV+oYIjYyn+tAwvZ2DNuwmOJjb/MKAoXmwZLLJI9Anb/AD0Di5lzBLLYPcAJ/wCegd42WndhBPoUp/8A+lBzV1lhHiWekn93Tv8A56DvvWUgpNkIA2kDf/PQIq9yQAAsCFT3T/8AlQOF7kSUn7jufUbj86Bi77KA7Y4gETuOP81A05DJBokY4mDtIVBoF/Sd+AkqxygqOwWf/LQI7l7xEn9GuEATshz/AJKAasg48pDj2IKlhJCStpwkA8xLfyoBMXrbClNtYPwkrOpWi3WAoj1/V/OgOc0+lsJVjHRA0kBLsf8A26CGLixShpKsEgIZSUoR4TnkCuQP1fHNAF5rEutwrBIR3lCFIUn3EIBoJV1nGHwE3GNcXpWlxKVBWyhwfh7UES5ucY6g+LhgsKUVmSqCo89u+9AFN5iQ8lSMOpCm1ap32/L2FA9vIYa1UFNY8j9YlyUqiVJ+E7ntNBLbzWOTYGxLLqreNCgbhMkHmTqB7mgJa33T7Ns6yyz4aVtqbcQm4SBpJkgefbcnighLuunAgjQsIJggXSQBuT/3n+JX40El/PYa7YTarCiwnSRpuGwBp43C+0CgfZ5fD2Ti/u6lIS5Hk8dspmSe6+d9/YAdqB5vcK6444orK3FKUom4SYJTpMefbbiOKBRd4NgLUSp0OkKIW8FgwI7qPagIjIYfUhSAUJ8NTfhpcSEwqCdtUUAw7hUtsJbDjSGllxAbcCYJ5Ox9qCYxdYltAWy64lYcS6SFbqUE6RO/pz60DlN4u5tl2zjzykOLLqgDySrUREQRI9KA3hWCnluJduG1FUpASfIZBJSI76U8zQOtWrK3yC7tL9ypxSdBCkmCNgP2Z20j8TQGfatbi9N0X3kueF4KYaP6v/EmU80EKzxGOYTpau7pKAFGNKttlAcjtrVA+VAT9HWrLlo+xfvBy3AbSpTX7HcbJ9J/jQWoyFoFCXSDE7tq/pQOGUsEkpLypAkjw1/0oHHJWgKYfjUJ3Srf8qBhydmkEm4QB8jtQL+kbQwpNw2ozHNA9F5aEbXCBt60DxeWoGrx2wmJ3NB3321KgC+yCTEaqAa7u3W5/ftn21jagcLu20gG4ajj4qAN8bS/sLm0cuEJQ+2ppehYkAiNqChZ6ZsGkFRv0F/w1IK0ISkHUCDKRtHBj1E9zQWOBs7LE4lFgi7QtA1QfKiJ+X8eaCif6NsH7qyuV5FCbm1bLSHAy3JRpCRq9SAOT6mgssZgsVjFWZauWNFupa0pWEk+aP2uQBG3pxQWWRyduqxfFjkrNm6KSG3FrlKVHuflQZ1dpdoK1WWUwKbl5JbfcVbai6mSQDv779pmgkHEWr1rbG5XilZBKda3AymA5IgpmdhBFBLsWDbWV20w9i2rhaSGnEt+Uf7wET+0aAVpYqRkFuXV5ilBDbn3fw2QlbK1bBUmeE7UDXse8454jeQwq3YUrW5a6lKUUgEkz3UJMdtqCRZW79qV2yr/ABxsy1pFu2nQGlbwU/4Y0iKCDn8M9lFrWrOWrcPFTKnWkOhLZSPIJG3mBNBMaB0gKkmO1Bn8/kM21l7SxwViw6FNqcffudSWmwNgNQ78+tBm8X1d1XlcVd32MxWNeat3vAIQtZUs+qR3G4oLr/aDLf7dIwDLVm6whlFxcOkK1NpiVAbxPp9KCqy3VnU2MtLa6uMZjm0Xdx4Nuw4Vh4ifKVDj0oNPgr/qJzNXNrm7C1Zt2mQ4l+3KlJWongE+m80A7/qgWvWmOwKWkKFy2VuOFzzIJB0iPp3oLPqbILxGDu79hpt0sJ1aHF6QR9Ad/Sgp8T1Wcn0Zc5tm1bQ/bhethbhCQU7/ABR3EHjmgBkeq8jjrHDLexDa7nJq0JYTcRoJjT29CJ9KCwzfUwsMhb421tF3+YdRr+7MqADY9VKNATC9RPXWZXishinrK+Q146v1iXG9HAOofhQaFMAyBuR68UGcv+o/AyFza2OMu8iu0SFXKmSkJaneN+THYUDbzqxpjpJrPixuFWzhENqUkLAJgK/6UAcx1m3iMdjby9xl0lF8CQnWnU2fff0g0Ftmc4McMf4Vo7eC9X4bXgqSIJEjn1E0AP8AaJB6pOETaXKngkOKdGnw0piZ5mOBQOx3UTGQyz1la2l06GlltVykJLIj/ED/ANaCLedaY23cfJYvnbVlZaXdtMamgsfszNBowvxAggbKEgEb770FfnsvbYLHG9vEvKaCwkltOoiaCic69xqUNuO2mVQ0uNC1WxCVTxuTQWuX6msMU4ll91xdysa027KNbn4CgNhc1j8y2pePe1lB0KQsFKkK9CDQR8f1XishlFY20deVepKkqQWVAJKeSSRAoDqz+NGdGH8dZyPPhhtR2ifiiPzoKxrr7psOqQb9adKtKlKZWAD7mKDV21w3csNPMLQ40sakqSZSQeN6B+nbyk/jzQVOM6jxOUvl2VheoeuWwSptIV5YMHkUHW3UGLvMkvH22QaevUlQU0gnUI57dqCI/wBX4C3ujaP5m1S/MFOvYHjcjagsr3I21lZG8u7lDNqACXSfLvsNx9KAOOzWPyDDr9jfMvstEpW4lWyYE/woAJ6mwjtm7ct5e1Ns2UpceS4ISTwJ996Cwxubw+SJaxuQsrlyJ8Nt0KUB8uaAmRfsrK28W9dtrZkn43VJQJ+tA2xdtbloG0ct3mwJlpaVc/KgM60mRLaDAjdPFAO6dtrZlS7hTLLUxrXpSATQELLfdpvbf4RvQVv6WwxkffsWY2Mut/1oDN3GJU224HbAsqUUpWFt6SRyAeKA6XsUQJesFDn42zQPUjFoKPE+4JkaklWgSPWgIljG3CobbsnYEnSEGPwoCCxxqggJYtFGdgEJP8KBTicdBBsbYCI/uhQOGGxoCP7BbiN9kjegRzCY9SRqsmYmZKaAZwePgBywYO+xKOKDk4nHLMJskFQ4G4AoHDC2SSQLJpJmNgaBUYexDY02iBPZJVsfxoHuYy2UQpVvMCASVSAPrQMRibQifBVB3/vFbn8aAoxNsIhpSZ7eIvf86BFYi2QA4ltYjeQ4vb86BqsdbJSZQvSe/jL/AK0Cfo22LZSEuQof98vb86BBiLTUZQsqn/vl/wBaAqMfa6DKFgAbjxF/1oAJxNsAT4bkE93l/wBaBXMdauaAUOSnbZ5Y/nQcrEWQIUpu4JUd5fc/5qBxxNkqEqYVtxDix/Ogb+hbDcFpwqmd33Cf40DDgrEgEMLEjgOuf1oHIwuP8MBLSwgD/vlmPzoOOBsCJ8BU8f3y9/zoHt4SwbQAm3WBPAeXt+dA84PHggFhYn0ec2/zUALrpyxdCfLcAjby3Tg/81BWNq8gJG4/OgouvMocT0hk7kEpX4fhoI/eVsP50Dvs4xxxXRWMt1jStxHjOT2K9/4RQUX2bxk8/wBT58iQ9cfd2VH9xPp/loE6gBzH2p4HHp8zNg0bt30BO4/gmg2PUWWbwuMdu3AXVg6GWk7qecPwoA9/4UHkz7rlvf4LJXdhkGcp+kPFvbm5Z0JWVEDQk+gEgD0oPRuuAq6/ReJbE/fr1OuOzTfnV/AUGJv0P47q7K9PMtH7tmn2XURwhJVK/wAtQoNFeuJyn2o421EFjG25eUBwFHj+KaAf2ZqN5nOpcndQbk3XgQR5kJk/9B9KDcN/d3Lp1xvwV3CP1S1pgqSOdJ/jFAZxaWpUohKUiTJiBQee9S2uTwWRuOpsA8h6zuAld2wTKVD19x7jcfKgd1u4nK23TWKtEeExkHUPKbSI0oAH/MfwoJ3UONb6izt5jyElqwx5Df8AhecPl/AJFBX9B3y8td4i2f1n9D27pUT+8ToT+CZoKa4vkvZe7yl394Tiby5Nq66wYVoG4EjeIA+e9B6rjre0tMey1YobZtQkFsN7JIPf60GBtW7vpq7XiMuym56fvLiG1EbJ1H+MxI+ooPSQkqHoeN6DBfaut24tMbirUa3rx8eSeY2AP1NBa4dy/vUXtrnMWxbWtmG0MszrClJE6ge4HliBsaCp+zBJvkZTN3A1XNy+UhZG6Ujfb8h9KAOOfaT9qOXfY2YRanx9I2lIT+c0BfsvYVcjL5laVBd/cEgq7gb/AMT+VBMYuQMr1TmhOiyaTaNHtKE6lf5iKDE9N3d2xgncMxjEu3maWpxD7y0pQocSJ54VG/NB6p0dinsH0/aWL7iVuoClrj4QpRmB7CgtL24Ta2b9y4QEstqcO/OkTQeT9A59zF2OQvXcJlL374+Xl3LDYKQByJ2481BVs3tw10l1DngjwrvKXf3ZtQEKSg+ZUfwoN9Z9L4jGdDPW11aMH+yFy4dUkalK0yTPIjtQYu3yTjH2NKQ6snxXzZsyJhJOr8twKDf2Vnb9O9GMl8IT9ztC64opG69Mqn67UGC6lYGJ+yfF2IbSm6ybiXXQEgEk+f8ALygUEe5tmOonMBa9FYx9m4x5Sm4vy34UER8RHfYnfc8UGj+0t84zqzDZLK2Jv8GhtTamymUpdJ+KDtMRE80Gl6IxmKx1lc3OCUFWWQdNykpEBIIgJ+Q3oNEozyTBPpzQeV/bFev5O1et8e0hdpiHEPXb5HDqtktp9xMmg9DbyLSenv0q4pPhC1+8GeI0zQeT/Zy7088wi2ymL++ZPJ3hDbarcFKEH0UdoG8xvQehDojB2OPtm7tpT9jji9cBl2PDJVuSod4AgUGS+z3o7E9RdM5W8v7RDYvrhYt1NiFMpH7vpv8AwoA9SvYBP2k29vl0NjE4mxDKx4RV4iwmUpIA7SPwoNzYYPB4C0yPUWPtPurT9hqNuoQlKdOrjeCdpFB5jg2sAOgrlTiFv9SPOLUwi3C/Ea8wCIjgd6D2vosZAdLY0ZkqN+GB42v4vafeImgt33EMMOOPkJbaQVrJ4gCaDyj7LchhrO0vchksrbM3ly+pLbL9zu21MjZR/OgidPls4nrfqR124XaqU+zZtl5elKT6b9yUx6UFn0V0crI9CYpF9f3KBcPi9uYdVrWNJ0ICplI4JoKvo3pe26kvep3E3N+zYMPm2sii6X5CJ80z5u3PrQaDqXpS2sMA5f5DIXTrWLxoZaQy8prU4JJWogySpRG1BW9HdEuZTpLE5C5yeRbv3HE3CgbhZSpsKnRpnuO9BM6XUrI/ar1Ddt3NynH2I0aPGV4ZcIAO0+yjQU/QuWvMb1wyLp+4Vis4l02SXnlLiHDp+L1j8CKCwfzaMBi+pepGHX33H75dnYtuPKWgR3CSf3go/IRQWmL6SyT2HTf5HO5VOedb8TUh/S20oiQjRxHE0GqwKL5nDWjWVf8AGvw0PHXA3WeeNqAucuE2mEv7h4qS20wtaigwRCTwfWg80wucv+nvs2t79x+4vc1l3fCtEXDhcSJMJMH0G59dqC7yHSmaZxC7uyz+Te6h0g61P/qVK21J0HyhPMUFb1g85dddYDEHLXVqym1U5eKRceCFJH5TsaDQdP4NGPyK8uxm7+9xLtqfDbffU6AeS4Ce0DbvQeeYm+ff6Qy+cf6mydtdpuHBj2U3eokDhJQZJkmKC66xzmWR9kePuMo69a5q6cbSFIUWlk6juqI/Z5FBZ4/pS0yLTAxfUeRXd2i2TeFN6t1tW0qTz3377UEDNOsZH7VLyxucxdWGNtLMKeSL0sguHiN+0iflQSvs+6gujddS+Ne3N705j5VbXT51KgSSNR+Lb/W9ATpK1yXXdq5m8xk7+zsnHCizs7J4tBKQY1KI5PNBr+kcflcYMi3msk7esJe/srj5TqSyByojud5n0oLdFw1dMoetnkPtK4W2oKSfkaDOpB2PpQUvUvS9j1IlhOSXclpqdLbLxSkk9yPWgsVYm3VhBiiu5FuGvC1B0+IUj/FzQM6d6dx/Tto8xi0OoadUFqC1qWJG200EXH9J4zHZteXY+8m+UValrfUrVPII9PQdqCRlOmrLJ5a1v7pd194tSFNaHilKSN/h9fWg7qfpyz6iSwnIuXIbaVqShlzQNXqf5UCOdM2qsxY5FV3fquLRAQ3L0pIiDIjee/rQSbjC211mrLKOJUbu1SpDcHYhXqPbeKCBjOlmcbmbnJs3d27dXA0u+OsKSoEz6DjtQRr7oq0dyr99Y31/jnrgy8LVekOE0FzhMNY4S1NvYoUlJVqcWs6luKP7SieTQGydr9/sLm18RbQebLfiJElIIg0GYY6NcRZpxis7erxIgfdSlI1JG+kqG8UEtXS61dTW2XGRWlNsAhm2DY0oQExpme++9BLwGFexeQv7l7Iru1Xig4tKmgiFD5e20UA7TpoWf6bVY3YZeyS9QWG58EHkRO/JoI9r0ipnpO7wirxp0PKK0uqZ+CSDxO55gztQDt+lMkjpd3COZdpTWtBad8I6mkg6iOd+0elBJHT2VvrqyVnsmw/a2jgdS0yzo8VaeCs/0oNSCDzufSaDK3eAylz1jbZhy7s/utsdLbBbVqCO/wDxTQadwA7EpkiN6DJWGCzGGF+xg37JVlcKLiA+khTKj8uRQE6d6YRjLK7F1cKub29SU3Fwe8g7D23P1oKnE4jq3BWhxmNVin7TWS1cOFQU0Cf3e/r3oJmT6Xv2+k28PjHWXVvqK7x55ZQXFE6iRseTH0oBZzpXIXfTmCFgLe3y+MKdEL8pA5GqPUA/jQbKxXcu2TK71pDVyUjxEIVKUnuAaCq61YyV7gn7HEsNuuXKfCWpboRoR3+ZoIBx+Us/s7YxllYJ/SBY+7lvxk6UzOperj3j3oKwdI3N59mzOHeQmzv7dXitkrCklYUTuR2IJHtQJmh1bn8WMOrFW9ghwBFzdG4StJSOdIG+/wD0oF6m6PvE9J4ywwC0rex7gc0LhPjSPN9Z3oOvT1N1YyMVfYZOJsVrSbt9T2orSDJSge+1ALqWyy2V67wi04a4Vhsc6kBYUnzb7rieBA+YFBNxOMyWC+0fIu2Vi+/hMmQpxxBGlpZ3nc9jq+hoALz+bctb+yzXS1/d+M64m28NseGpskhIVvtHrzQX3QWDuMD0ra2V4tJfSVOuJBkJKjOkH2/jQF6svslj8KteDxjt9fOgto8Phrb4lA/lQYfI9BJtejXn2Hc7cX76EuLskrELeVElaPY870ErIfpxz7K7fFMYi/ORUkWjidEaUJ3KvkRCR9aDZdIYxpjpzBofsiy9ZsgpQ8jztrIhX47/AEoAfaO/fjpq5tMVY3N3cXqSynwUT4YPxFX0kUErpG3/AET0Pj2haXKF27EuMlH60rG6tvczH0oMN0w+/ZXnUFxnemctcLyj2ototdaS2JISZI9fyoJvVWVzWW6FuGhhrxp6+fLbdu2ySpphMfH7ntQHzuOucLcdLZvBWd04i2aRa3DDbR1qZKf2k8z8X1ig9KYdS5bodSFJDiQsBY0kA77g9/WgzP2oX6rHpO7Yt0XLt1eoLDaWmlLMH4jtxtPNBQ5G2xGD+y+1f+4tPXLtmG2Sq0lxTik95TIgyd/Sgyeaft7H7LMDhGFrDmQWl24V4avInUdRO3YwPpQeo9SZO26b6RfW1JKGPAtUoSVFaymE7D8aCB9kFo1bdD2IaUSt5Tjj2pJELJ3Bn0gCgp/try7IwrODbdKX7x9BchJ8jQPxEgeoHvQbS9vLTp7pxT6nQm0tLcBuBzCYT+O1B5Z0rk2cR9luYyK3v+0sk+8jZJ1BwiAPoCTQW32jYcI+z/EX+FK1O4hLTrDqB5vDgSr8YVQM6s6cu0/ZngksW67l/HuN3dwykSpcyV7d9yaDvtA67t7jAWp6ZyrjN848FKbZR50I76gR+XrQelWKZsbZQccd/VpIcWPMvbk+9Bh/tly7Vl0m9YB5Iur8htKO+gK8yvlwKCv66tUYi16Iv20rfxWLWgPLaBUAkhEL+sGg1Vr1pi8rfW1phlryBX5nnG0FKLduN1LUY/Dmg89YyHS2d6x6lveoru0NmtKbW0Lx5SBBWj322PvQXXUHWuNP2e5P9ES2ykfo6xUo7vQkBRSOwANBlcsvp+16Uw1t0pouOrEFrS/ZIV4gVyrUY35iKC566zFje9X9L4vO3LH3ewT4+ROkqb8YpnQY+X50Gh6e6l6ax2RurHp1sfc1Jcv7x9KSltoAAeUR32AA4oMbiMG11h0v1flVoSvMOXRfYlPnbCd0jf13FBuOnbxPWP2Y3NtbBLWQNoq3eaCNGl3TsYgDzRNBT4Xqyww32bfo26uHsZnLO3Wx93KCHQ4JgpkQQdqC+6b6gTb/AGVDM9SO3d/rSsXMt+eFK06QNtuN6C/6WsLDHdO2jOAYcFkoeKhtayFDUNUkn5jagiJASN96ByBCj7UBUzpBEUGZzHUF23nmcFhbVm4yC2TcLNw4UIQgfKSTQHxXU1u/05cZTJtiyFstaLhJOoBSdvKe4O0UEGy6oyNxlMSi4wi2LPJBXguBetaABILgAhM/OgJluqLhN1f2mEsPv68e3runFOaEN7ToGxKl+1BeYPIjKYm0vQ0tkPthzw1jzJntQD6kvnsZgr69YU0l9hsuJ8b4THYx68UGS/2rzmPwFlm8tZ2K7C4KdbbJUHUAzvvI7UFz1x1FcYLE2l5j2WH3Lh5LSW3pBVqEiI/OgRN31S1lMexeWGOVbPuFLr1upZ8JIEmZ/L3oI9x1PknG8neYvGM3ONx7i2lqW/pccKPi0gA8e/NBY5TqJjF4a0u8gwsXFwEhq1b8y1LInSPlPNBEPUWSslWqcvgl2zd26lppTL6XCFq7KG0d6CRedRPs9WNYVjHh5a20ueMHwnSn9okR23+dBNymXesMpYWTGOXdG7Kh4iXEpCNPxTPoN6Cvd6nuF5u9xtjh37pdoQFLQ+lI3770BWepSenr7JvYy4ZNo4W1MKWnWSCAfbvQMtur2j9zXf2F7Y292B4Vw8Apsk8AkcTQadAOoQZj1oK/O5NvD4tzIXDLrrbRGpLIBVBMd4oKi06xx9w5bIuba9sfvUeA5cs6UOTxCgT7UEzI9QW2OybNk7bXZceA8NbbflUSeASRvxPpQNyeetcdkmMe4xePXLydaEsNa9QmD37d/SgvmkmJgAkfhQRcvlbDC2qXsi+lpKlaEiCpS1eiQNzQQcf1Rir69as21utXbh/VsvsqbUrvIkexoJ+XyVniceq7yDqWLdJ0lR33PAgetAdJCoKRAUPxoM2Ouung+tlWRCVoVoUVNOAJM9zFBc3eTs2MYrJOvINklHieMjzp0nvtNBVWPWXT+QuEMW2SYU64dLYUFJ1H0BUAKDQIUAlS40wJOo8UEfF39llLZVzj30XLIWWytsyNQ5oIuR6hw2KvEWuSyVtbPlOvQ4YMetAS26n6fuAENZiwW4sgJSHkyonjagkO5bHIvzZG+tvvoVH3cODWDExHNAFWWx7d8LE39sLxRA8AujxCSJA00D8fmMZkHVsWOQtn3kJ1KQ24CQBtMUEldxbNXDTDr7aHnpCG1KGpcCdh+NAZaZTAO/MmgVIJlQI222oBX+TtMY2yvIXLNslZ0pU4sIBV9aBMflsfkNf3C+trkgSoMvJXH4UDLvM4tp8su5GzbfT5VJW+lKkn3E0Brd9K1Bba0KQrdK0mQfwoJDl01buIFw+00pew1uBMz6TQSNWxKyAJkydhQOtXkOiW30Ln9xcx+FARwqB5Ow70AnAlUhUHSPNIoOSII0gSBsR6UDlBQgkASI370DBEElKSeNxQKseQagk6huOaBgbSmQG0gz6c0BHE60GZT7RzQMV5UgSQR6d6BgaZCwoMNpXE6tAk0BJBEidvWgaWkOmFtNr9NaZgUDylJaKCgFsiNMSDQBbt2Gk6LZpptKuUIQEg/QUCfo6zUI+6W47/AN0nb8qArdpbBIQLVnQDISGxAn6UHC3ZaClMsoQf8KAKBE26CCpTCFEn4tAoCIZQk7NISkj90b0DglSSfDQQAew5oOSkhRLYHyigIphpwDxWkLXwNSQSPxoI+cxTeYx5srh11q2WoF5De3io7oPoD3ignNteEyltpAQ2gBKQnYADgUGUCR22neTQEQI3EARJNApEz6k9jQeR5SyuesBmc+1doxT2KLttbhs6XFJRurxFT33iKB2TK8n9nHSVmlhFmzf3bTTugQnkifr8W9BfdPvXuP69y2It8hdZDHW9mHVeOrWWnYEJB/lQZG2bucV0Kxn7XLXbeSvL3Wm2QoaHVFRBBT3O280HtNsXFW7ReTpcKElSRwDG4oMX9rFwo9OsY9kLLuQuW2ExyRMn+VBUY/FNv9Yr6d6gubq8t7VtL1ihbmlCkhI2KRE9/wADQWPV7BzfW2EwzbymRbJN4pxCQS2f2djt2HPrQXuGc/Qlq9b5vLIub1JVcvLWr4WyqEmOw4+tBluosdedIZZedxL+rG3b6RdWajsSs8j+XcfKgl3s3v2t2Vu+AWrS1LrKDxqImf8AXpQbhxLK3m23dClzrQhUTt+0Pl60GR6WIyPWvUORPFuU2jc9o5/+mguRD/VrrilfqrGzCSSOFLVJ/wAqRQZjpJvMXRucrYrYDGQv/wBcpSf1iWkq7TtHb1oLv7Q3S7irOyt1ICr+6S0I3kTJ/OKAH2iFq16UTYNoSpVwtDLKB2j/AEPxoNdatFq2YQ4QVJbSlRA5IFBkftOW45jrHF2wl28uEpCSdoT6/MlNBDvmH8lm8dieoQ1YtMw5bNW36xLxT/iO444iguOvmknDNXM6XrV9DrZA7kxH+vSgssHigy65f3S/FyNxClucBKY2QkdgPzNBcpAkQSCR+NBhMaf039p+ScuAFW+IaDTKFdlnk/P4vyoNi/j7d/I22RcRNzbpWhszwFc/P2oMF9oLznUNrlLKzeCLPFo8S4KSJde7Nj2SmST60GsweRDnSdnkHdgmzS65Jn4U7/wNB5l0tlHxibywTi1ryOfcdcbeeKUtKmRtO+x1bUF7m7a46U+ypVg66FXbv6nymQC4qSB9JoKrI4zIZrAYbAY/BXdr91KVPXd02lAQQNynk8kn8NqDd9RX7tqxa4rHrCsrf/qmiv8AYQB53VfIT8zQUv2SA2tjmsWtYUqxvinUO4I5/wApoJn2gossN0/mssm3QrIXTSbZLixrMnypAnjaTtQTuk+k8cx01gW7uybN5aBFyFgAKDh33Pcb8UGOwudsm+vupcxdWl2+FOfd7ZdtaKdgJ2VuOCYH40B+lLq1y/XHUPUvhKTbWNultsPI0qBA8xI7Hyn33oLL7H7cDp67zFwhCHMhcOPlREQgH+E6qCny19cXvXHTXUJSGsc5dqs7QKTClt8Fw/7xUY9hQeseWDG31oOMJSUgH396Dzn7ZLppwYHFuglFzdB13Q2VqDadjAG/c8elACyt2819pOLvenMW/Z4+zbIubk25YDoM+WIE9h+fagrrTKYFX2g9RZHNWirhpxwWlq0m18UOrGxjaNRge+9BZdJoyPSvR3UeWvrNdpb+IX7OydO6ATA27Akp29qC16B6Stsjg2sn1LboyOQyKfHW5cjVpQr4Up9No4oInS2VGP6P6vYedL1jiXnmbYuHXKCCAj5TQZzBWuIV0djWsLbKv+qwtDygzrbUkapOtQjy6dtvnQew5PJIw2D++ZElSmW0+IEjUpazsEp9STsKDFfZVc3qc51ZZ5Yab43CLlTZUV6QtJ8sn02FAD7UXcfcdX9LY++fbtbdJW/cuqdKAGv3SZHMGgN0Y08eub5zp5dyekvBACnlqW0t3/3RVv678flQB6VsWs19p/UuVUgqt7J1LLPnOnxgIKon/CaDedQ4hvNYW4xz5UlLiNilUEKA8pkehig8t6OunMr0UemLFtZyzjrjd68VK/s6Ar+9UZ+LgJT3I9KD1HAYW3wePFpYqeW2DqU444VqUeJJPr6cUD88XG8PfONXItnG2lLS+UBeiBJMH5UGd+y1zIXHSjV9lLp+6urtxTwU4qYQDCQPQbT9aDLdSO9R4jBDMXuTu2c87feGzZNuJWw6gqgJDYH7v1oL3rrKPOXmOxjOeZwDxYN1cPqUNuEpb3I5Oo/SgG69f9F9L3d/dZZWdyF44hqzWo+Uz8MCTtyokc0BbW7yuF6xwuOyGVdyH6UYcU+24hKQy4kTLcAQnkRQb0nSDBJkdqDF/a/fJsehbxxLriLh1baGFNuFCgsnmR7TQYzJk9O5Ppa36dzF5d5m4cQm9txdF9C07aipMkDv9N6Cb1c/jch9rCbfKXyrTHWdjruAbpTSXFcpSII33HG5oNZhunMZ0+nKZq2fuF2r1jqS288pXhpCSpRBUZE7e4oPL8QnEI+zxeSfy10rqN5a1WrTF4suoOqEJ0An5mRxQbnqO/v2vsZW9m9acs5boQdJKFBwqATMd4gmgj9QaekcB0pZZNd6MQ+sqybzTiitTmgcqnVEzsOwoNf0li7bFi/yFpk13eIuUIctUqfU4lpCUkqgqJ5P1HBoMn9lt7hxZX+UymWYTcXNy4llm5vN2mZ2GlSvXvQaxsbDf/rQOSme8ie1A8iDsSdu9BQX/SOByF6u7u8e0t906lmVALI/eAIB+tBZ3uLsMhj/ALjeWrbtoQAG4hKY40xxHaKBmIxFhhmS1jbdDCVHUogSpR9STufrQRm+l8KxkhkG8eyLnVrCjMBR7hPAPvFBcyJJBJoKXMdNYzLXzV7etOLuWwAhSXVI0wZEQfzoDZLp/G5PIWl9dsrN5bbNOIcUgjed4NBAvOi8TdZF+/Wb5Fy6ZU43dLQd+2x49uKBjfRGIbt71lQu1pvQgPlx8qUsJMgajvQKz0bYJdt3H7u/u27dQUxb3NwVtoI4Md47TQH6h6Ytszc212p1+0v2D+rubdWlQ9vegJiOnbWwvVXrr9ze3606Tc3K9SkpPZPYUEK56RSMtc3+Iyl5jXbky8hnSpKz8j9aCejANsYW4x9tdvtruiVPXCoW4sn4iSfUbe3aggtdJXlvjhYWvUN4zZhOlKEMo2B5E80HXvSLztxjnLfKrt02DaUW6AylQQQIKt+5oJFj0yRkmr/MX7uSu2f7vUgIbQexCR3/AJ0GjbgHiSd9xxQZjOdNZHJdQMZJjJsMi1H9nbVb6wmedW++80HY7pm4TnG8vmcib27aTDaUt+GhPbifc0BuqcJf5ldoi1u7Zm2ZV4qkOoKtaxxweBQaJQhw6oUvbccTQEkbwAN+/agxmQwOax/Ur+X6ads1JvB/aLe7JAKvUEfj+NBK+49SNY+9eF/bv5i6hCQQUsWyADugclW/J5+lBAHRtsx01CMVZuZwsFtTi3FbrOxUV/nx7UA7fA9Qt9Brwmi0TcKUpoOB47MqMngczIj0oGZXpHIXXSmCt7JLDGXxS0qRLnlMfF5o77H8qAvV+JzuZew5btLMNWy03Fwg3GynJ3SJHA3g+9Bs30vBpamUBboSVpQTGo9hPagxGN6TdzN5kL/rKyQbtxweAGblRS00B8CdMcd/WgH0RgsphOqMmsYxDGIvT5NNyFlrTOnbkzNAb7Q8XmM4/jLayxv3jGs3CLh9QuEoLsfsgH2nf3oNpm3b5jDXK8ZZrurspKWmUKSmCRAMnsKDKdA2OQ6Z6Hcafxlyckhxx1TAUnU8o8EGY9OfQ0GXscR1DYfZ7lbFvDvqzGRunA8Q4jZCgPNz8x896C3uLTMDp3B9P2uJumrJxppu/dQ4nW2ifOgb9+SfSghdadGOWVrjh0yxk724ZuEuBK7rW22lPYBREHiIoNXcZ3Is9WYSxTYoTbX7C3XtRlxlQE7wYgbCeDPNBplJVpAIgHn2oMVb4/K3v2psZa9xLzWNYYXb27inEHSYPngGd5PvxQbPMXFzaYx96ytXbu6CdLbLZAUSdhz2HJoPOsf0de3P2WXuMuLN9nM+Ou7SVkS46NwRB7jy796DTO2WR6r+zt/G5S2dsck4wGV+NHmcTBCwQTsSP40EOwzWbsummMXb9OZEZxpkW6PIPu4IEBzxJiO9Bns705kML0JbYCzsru/vL58XOReYb1pG/mE7b8R680E3PsZDqTK4C2w+Ev8AHMWD6XHLy5a8JSEJgaUwZ47UFhei86u6kWgLyWIssYUuWylW8eO5uCvzAiB2HO80Fd05b3OJ+1fIh9eRvmbu2DSr1y2KEl1MGCUgJgRE8UAW8qwr7Q8zlctjMiq1DCbWzBsHHNQHxGI7/wA6CZ0Qq/xFh1LmLqwvbSwuHy7Y44tlSknf4UDifKPxNBA6Ez9v050o+b3HZZzLPvOXDzTdk55lk7DVEcRQeinKLt+nG8tkLdbTvgJeWw0lThSoiQkACfSgy/2NIBwF/eXSXE5C7vHHX0rbKFJJ4G439frQb5tIAMkidztxQYv7Wsh906RubG2Q8u6vh4TYabUuEyNZMDaB/GglvZS26f6Jtry1tnrtu3t222W22yFOEiBtEj1MigxnSmcGW6jsXshjMjcZt5yA5cMlu3s29yoNjff3O5oLLKZrFXD2dY6hwKFZFoqtrRAt1OOXLceUhUQNzyDtQRBgrnDdIdFfpZK0sWN8H73bV4KVSUzHYGAfSgt8KyrqLr1/qJKCMVZMfdbNa0keMo7rWJ7bkA96DfJCCkqKflFB5v8AaC7Z5frTprAur/s7Vz492lTZKCY8qCeN/wCdBt2MVhsO29d29laWKUpKnHGmQnYfITQeWdKZDBX191Re9RWr7rmTuoaaXZOLIZSNtJCdj9RxQWnWXU72T+zq7FlYXTa75xVtbNhlRX4KSApxQA8s7wKAOZxw6TyfSnUGEtnHbNLCLK9SwydS2ykecpAmee0yBQWH2i37GUzvS+IQH/ui7tu6uVhlWlKP2Arbv+XeguMj1dhLt3K4vqK0cbtm3S014lutaLlI5KYB3BkbfMUGXR946c+ynLssMXSF3z7wx9uptSnG2lkDzAcbSd/Wg2OGsumcL0hhl37FkposIQHnLcKUtWmTyJ9eaAo3SEpgEHf3oCNaRxJk0DhCZnc/woKvqO7+44LI3aXvAUywtSXAArSQNtjt6UGJtepcxb9F4u9yV+ym7yt0ltD7jICLZsz5iBE8UGnwbWdZy7zeTvUXuM8BK2bhLKUFaydwY9B/GgznR/U2WyPWVzZXz4dx6mnHbf8AVpTqSlcJO31oHHqjLZTrm3xeIeQxjVgq8ZTIWXEpkKUJ7EggfjQXX2idQOYHBF2zcQm9cVpbStOqQPiMe23NBosY8Lqyt3UkEOtpXPrImgyeKz+Qteq8njOob+0Ta2bXjpWGQ34iT3mdon60Fh0nfZrKuuXt391TilKX92/VKQ66mfKsyTA/jQTup8ucNZMLaaD13cPJt2G1K0pK1fvHsBvQBxmRy4y67DLWNshtLPjfebdaikmYCYI55mgTrDP/AOzuMRclpLq3HA0hBXEzyfoKC3uXVNWjjzeg6Gy4nUYSSBIk0GZxfUmdyGFGSt8HbKtylZA+9EKUE+g00Eq+6jvLdeGYRjA5dZFrX4Ze0+GeYO350EvG5bIXOXVY32K+6BLXjl0Pa0kTAggfOgvf2Qf4UFD1N1G30/4Kn7R19LoUZQtIiO29BOzWbtMNiRf3WoNEJCEIAKllXYfnQVN91exb4S1yq8dei3uHPDQklOo+iueDBoJFn1PbryLWPvLS7x12+JZRdAAOfIgmgNmOoLfH3gtk2l7dvhsOrTbNawhPqo7UEvCZW2y9im9svELC1FKdaNJMGDsaCxJIkkfIUETJ5C2xli/eXznhWzSZUuCY/CgLZ3KLu0t7pjzMvNh1BPJBEigpch1lhbG6fYuLlweAdLrqWVrbaV6KUBE0EjJdS4zGKtE3bzk3SC6yltla1FIjeAPcUFcevenVJBTeuEKVoTFu55lenHPtQWGa6lxOFeQzkLzw33BKWkJK1keulIJoJGJy9hl7cv465Q+2k6VFOxSr0UDuPwoKdXXfTjdw62u/KVtKKHP1DnkI232oNLY3dveWqH7J5DzLg1IcQZSr8KA13kLXF2KrzKXCGLZBGpxZ2E7UAsdkLLKtG5xly3dW+qNbZlJUO1BCsc1i8heOWtjkLa5uUElTba5UmNjQSlXtq1ds2rly0i7eBLbRPnUBzAoOvby1x9m5dXryGLdG6nFmAntv+VBW2WTwL7d7mLW5tFoSkIuLtB2AHAKvagIOpcGmx+9qy1mLPX4Qd8Ty64nTPrQTcLnsTk3D+i8haXLgElLbgUR9KDr7qXB2N2q1vcrZW76I1NuOAKEiR/KgkYzL4/KF39HX1tdFoAueCvVongGgnhXl2BIn1oI9vlbC4vHLRm8t13aJCmEOpK0xzKeaBVZCxF99xN3b/fYnwPEGuOfh5oJDt3aWy0NPPstOr4DjqUkz86Ady400suPqS0hAnUowke5mgE3f2i0mLu2k9g8n+tAdN00EpV47aUEmD4ggxz3oHi4ZWsaLhpazwlK0kz9KDkuNKJQHU65IPm3NARMyIgKA224oOSASoqkxQIowASBEcg8CgYpwJkpWkgCPKd6BASOIHffnegVQgbmZ7TQI0qIBB4j60BSFKGiPKfWgaWw24sERA3NAiFawQATv60DykDTqnUDFAqQCoxuQO3agQKLaSkbqUe/agRB0uEqXEjfegcDIBB+LjegRKtYIEETB34oCkAlMbfuz3oHiQkyFDg7Hmg595LaQ45uFbRp4NBkWwJO0/wA6B4gbDae4oHggKMjcUGD+16+S10yiwQ6gXN++hoJKgITMk/LigsrtOBv2l9KX62v7LbtEoUoIIGnylBnkd/nQZnAZH9BdL9Sspvg/YWbyreweWoedShEJPeCRxtzQR77FXNv1niMTilhBXiE27r4P921J1rH5x7mgsukHLFPUfUOX8RtnHY9KLJglQ0pbSP8A8frNBFySMtmbLK5J7FNqtLu10W7q30hTDA3nT6mJP4UGw6AuvvfR2LcSolQY8NRmTKdv5UGdzFha5/7ULW2U2hxqztNV0Adl7ylJ/FNB6MEBKQNhA2gcUFL1JjrDqK1dxFy+BcISm4ToPna5CVR+NBU9GXWQs1ZXGZm5Dwxugi4UeUlM7/IQd96DP5y+RnLTK3l3ZZAAs6LAm2UW22xBLhV2KvXsKC8y2WSPszTcoXpdftm2UxyVK8p/nQTumMPlsc5jmX7xBxrNsZZQgIhwxsf3uTv60EG//SGR61vzi1tpex9sEIU4nUkqUOO0cneg1uKS4iwYZunvGu220h5UgnWRuSBQS9JBEc8bDigwfWrar93MvIBUmxt22UTxrWoKV/lgUFT1YV5nBsXRktK8O0sGyYle3iO/kEj2mgs+r2H1ZbpzB44IcXbJLxbWYSrSnaY9dKvxoFwbDvVuUt8xfXDaEY90oTZtoIKVDfzFX0/Cgk9Xqy2CyxzuMQLmxeSE3TKjwE7T7fMfXag0uDNq5i7Vdi14Fq4gOIRERq838zQWW+wAJjfftQef9eXac27fYVq+ZtrazYL10S4kF1wCUNCfxP0FBd/ZvcKuOjMUpQSS0gsnf91UD+VBkmn7jpvJXeC6ktBcdP5S6WW7lPYrO8kfSRyORQbrqZ23w+Hv8qEIFxb2qmkrA8wB2SkH5kUFN0RiWLXoCzTesB1Kh9+UlYmVfEk/kmggfZTa/f2ch1JfpC76+fUlLihulA7D+H0oB4y6ZtvtW6i8JQRbC0Dj4T8JUkJM7d+fzoMz0lmLi1tL5tOOW/fdQuuG3dfcShlfKYk8wSdu+1B6J9m/Ttx0102iyvHEOXBcU6rQfKmf2QaCV18lV9i7TBJJQvLXTbBIG4bT51kfQUGG6IyTtt0Y7iLR0Jvnr19oOA7sspSC46fkNh6kiguPsfxyMf0xd5B0pR95fW4HFiIbRsCT/wCI0FHkL28uOuOmOpHx4OOurlVpatlPmDXGtX+8VEgekUHrFww3cWzrDqAtp0FKkqHIO1B5J0RaZK6tsn0g2UsWFpduC6vEq3LJP90n3VBk9hQWn2qN2ttadN9O2VotTLl0HFW9s3qUWkcwO53P4UD8Xbo6o+0OyyeLtW8ZbYlOm4bWEt3Dh3gFsbhO8SeRQQLXO2rf2g9Q5m7x1zdstlGOtfDZ1I1yEwpR2TPafWg9bx+Hscfc3rtlboYcvHQ48EcFQTp2H+vWgm+HAgnfiJ5oPG82FY/qZrrlpSgwMqqzWlIABZSA3q+pC/yoNJmcm1j3erupGmmVvWiW7C2c0CVKCZP+ZY+iYoJXR3RmNX0wh/P2jV/kr9v7xdP3A1q8wmATxE9o3oKz7PWk9TdC3uKyTzztozdqtkqndxlKgpKSo9u3rG1BXYXp3EZT7T8ygYuz/RmNt0W3ghoBBcPJjuRvQB67xeJxCukunlQiwReOXTq1p1FLIMqGwOxmPkBQbLpvAdNXt7YdS4CyRahAdQ2ttvw0OAnSV6T9YPMGgzX2fdPYvqe46mzF9ZIeaur9bdtqJGhI5Ukg7TI3oLX7JMlcrez+IfuXbpjF3Zat3nDqWW/MAnUeYjag0nXzjTfR+VLiNaXGvCbQSd3FEJSPxIoMP0Rdnpmz6t6dzKvGVjG1Xbev/wBq2pPm+hOn8aDYdMWFj0r0YwblLbCGmBcXS/VZTKj/ACH0FBluir69e+0jOfpBtxj7/at3TDKySW0A+Ub8GDJHY0E37V2TkMRbYxorbfdU5cBYPmQhptSlHb18o+tBXYzNP2/2VYhq0fKb64snD4yjJZbROt0/LgepIoKRKP0T9iJubl51WRyzqPDccdVqTKvLvO0JST9aDXYvpXpzJt2buLvVuO451s3DzVwtbbykiSlUnSZ2O3FBT3l7jsn1xmsT1Dc3NregpbxhDym20oKdlJKSBrKt9+eKD0bpuyXj8LaWdxcLubltoJdeWokrV3O++9BQ/aewq9wFti21qbuL66baQpOykRKlK29Eg0Gd6Cy7tl9mNi1bq8XJXD71vb6zMHUSVq9kpkn/AK0FV9nrCLLoXqfqHJOLuluh9DbjpPmQBHBP7SjvQWGGx99gvsYeuMch39K3NuLhxYJ8SFHt8kTEfOgndC2fTmcvsdl+nFrtVWYIurQrVqUsphJWCdyN4VvNBXu3OLy/V+ew3VRftcw4+UY25UtSUoQR5NEEQe/vNBc/ajbqx32c2mJbK3bl11i1ZcKjqKxyrmfX8aCD9oHTIex2E6fwvlu2W1XCnFKUSEDSk7z+0pU/Sg0gX33/AA4oCIjcmRO3NA+Y44A70Fdk8Hi8m6HchYW1w9p0hbrYUQB2oA3XT2GyK1uXeMtX1E+ZS2hJIEc/gKCU5hsa7ZtWjuOtV27J8jKmklKfkKCQ5j7RQURaMnU193UQnlv9z5e1BXo6awyLZxlvFWiWnY8RsNiFaZiflvQSDibE40Y82rf3Ap0+CBCYBn+NA7FYuzxTa0Y62btkOK1KSgQCfWgBbdPYmyyAv7WxQ3dySXUlU78zvQW4PmEHYmYoKPJdNWF1kF5Bx28ZvVpCS6xcKQQkD4RHagB/svjHMU7YJFwlh5zxHiHla3lf41cmgtX8Yh/EHHKfuEslsMlSFjWUj1MdxzQUr3Rdi7jLey+93wYt1lbILo8pP07bx86DUttqSylOoqKUhOpR3J9aDNWvSCrJ+5ets3k0PXStbqwUSs+vw/Ogh5Pp11jHtWFi/fu3T96m6XdqVERzrUPbgck0G0JOkpST5t/lQZ5rppw4vJ2Vxknbg36w4XVNpCkHvEfIR6UC5Dppx+6sFW9y0za2RR4LJt9emCJ3kcwPlQcxgLlvq5Waev0ujSpoNFmClB4AM9vX5+tAOz6duMdmsjfWF6hNveJUo2xRwsjY6p9ZPHG1BBR0/nU4BGIXlbVdkU+Gt4snxglXxJEmPWCaCzYx2SZz1s4xfBGFZtQz91I3KgInj/XFBaZYXrli6MStlu7KQELeBKEn123oKDF9IsW+CLd7j8de5eFlT7iSQ6tRJBUoifSaCuwnTOdxvSN3iUXVoh9boWy82tXkBIK529tooJd9ic71Cu1tc2LC3sLd5DzirdSlLeUniJHlFBI+0TEZfPYpFhi0WvgLWHHlOOlJ24SIB29aDRWjZbx1uy60hopaShTaDqSmEwQD6dqDG4zE9RdM2r+OwjNje2CnFOW7lw8ULZ1chSY80e1BFHRd/Y9N3rdq8zdZzJOarq5cVoCQZkJ29yPqaAmZ6Svb3obGWDKLe2yuPcStkoXKQQdzqjv8XzoNxhl3zliwrKNNNXkQ4lpetM+oO3P/AEoIAt8q/wBe2l7cWTQxNmw42y6HhqK1gSvT9NP50GOt+j8phcDm1WNkH8zlXnGyQ6keCyVGNz6+g9fagsMhi8y/hcJhGMWpGN0oTkYuEpc0g+ZA+fJI5GwoKrrLoPwW8erpTGPruWnw6VKujpSlPAhZ77ccRQejrurpOLU+1YrN74YX908RPx/u6uPrQZb7McVkMXZZAZi08G9ubovqcK0rDk+kTxvzQQsjb59X2hozX+zz11a2tuq3tWxcNoJJP95ue8n6UDsPiM+nP5nqzIWKU3z1v4Fpj2HUlR4A1K42gUAOiG+ounME7ZOdLPXV24+t9bhuWtClE7SJJ2gUF+1m+o7G66Ys8jasO3GSU598iApreQEgdkpiTv8AOg0+cubu3x9w9j7Ry7ug3DTTZA8x4O5Gw5NBnmen/wBIfZqjDXVs4xcfdYLbkFXjA6pkbbq3+RoIdh0g+/8AZScG4PAvHmy6oLO6XSrUJ/IUBLzM5+46b/RVjgb5nMuMC2W64Ei3b2grC537xQTemcV/sd0yLNtpy7eaHirDAlTyz8UD+HtQQvspsLy3x+Vfy9pc2mRur1b7oeRp1A7p0+sb0Fb96fR9pl3mLvC5g2Fta/dLVTdmpeoz5lR6cwe9BKueoL44TOJxuCvrKyYYFtj2TaqDq1rBlWkbBI2oIfTOYusF0PZ4rE4LLv5gNEQu2LbaXVGSpSlQIE/Wg0H2bdMu9NYVaLpYcyN47490tPGr0+m+9A/rG88XNYDGOW94tg3aLh9xthS2xpnQCoCPj0z6ATQUf2jdLXOQ6x6cvrZLnhPrFtfFEx4STrGr2Ikb+1BO6jce6r6jGCbdvsfZWw+8OXH3Yw86lXlQCoaSkc+544oKJxq7xH2rY19+7vskHLU29xcG0gIKp0glIj037UF5cPJy3U2b1W10PumMXbMqWypKHFKkuFJ78IHvQYZ+xusL9lzdvdNXCsvlG22Q2hlSvu9qFTp2G3cnuSfag0HVlxZt5rpHFhm4Vi8bFw8tNstaTCP1Y2Htv6TQT2OqLBu6vmsNj3RapZcvrhwMLbSXIAASCNyTE+lBn+rsjiurOn0sNWdwvqNQQlhpLCkusOEjUCYHlG+52oPXsW2pu0YQ6oqWltKVnuVAQTQZdy/ayf2jW1oEXOjH2jwS44yrQp9UA+YiNkTv77UHn+JtLrp/oXK31w1cryF4XrXHMNtKUWkKPmVAG08kn0FBN6icNp9lGB6fx1terVdpaXcabdcobCpXIj97tyaDZ5XqVvBZPH2irG6XivusuXDNutfgxs2CAO4BnuNqDPWgTa5/PdVdO4q4bs02QbbbLKmxd3BVPlRsY4nYUD+s7ux62xdtaWGKvVdQeI3oW5aqaNpCgVlSyAIG+0maCd1Vd/pL7Rul8apu5VaWDpcee8BXhl4jyjVEdhJ43oLbpnJNZXrXOXCW3wGWm7Zgu260JUhJJWUkj98x8gDQc0QDBG/zoCBMSY2ntQORskkHv3oFV322I+IjigiKYuzcEs3qUtECEFgKiPeaDvu1+Dq/SCOeDbp/rQP8C+IE36I9rcf1oFFvfghRv0FIMwLfnf50ExSQQNgIFByQADO4mdqDgAOSQf40EA/fP00Ul1r7l4H93p8wVPM0EPI5xy2yLdgxYfeblxvxEILoQYBIUTsYHG55mgEMnk0kxgnQYjy3TZ3/ACoCJzd2CEqwV76GHGj/AOagcznLmTqwmSB55aP/AJqAyc64UhJw+UBiN20Gf81A/wDTkEzjMqfkwP8AmoO/TbfBsMmFHsbbj86Bf0yyCSbLIg94tVGgM3l7ZSgS1ep372q/6UBE5e1JCYuUkjk27n9KBF5azkyt/wBd2HBP+WgEcxYAGXVwdt2XP6UHfpnGqaJFxEEctr/pQc3l8cswLtBjiQdvyoC/pOxIAF2386ByMnZODa7Y9NnBvQP+925TtcNAHfZQoHIurYH/APmWTJ/fG9AQXVvKofZPqCsUCG6tiDFwyRG36wbUHNvsLCf16CmdjrG0UChxkqWVPNkcfENqDi41IIcQO2yhvQES4gH4kkzzIoJTahrgrBMbQaBXYKgCQDNAJ0SrYyfnQICmefMfTvQOUAXONzvxQIlICiqAVERMce1AYNqUJI2Jnig5tKiqAAEgxsKAjcBSkjUY9qDvuNqb43wt0C8KA2p4p82kfs0BjpSSnT5Tvv3oOU2CogbbcHvQNShE7aht39aByESCogpB78TQDKACSSSeflQFIJAVAGoTQKB+rSVE7n8KDnmlKWIgRx6mgZoUJKt47E0CQdJAiR70BW95AkfKgUwUgkHn8YoEWnUmVbb9zxQAO8DWRB4oOUUSRBJ5oHt7QrbmN6BXmRBKSRPIB4oK91JQQiSVD3oObgSYM/CY7UEzVpbK1TEQfYUEpsBxAUhQW3zIPNA0p0kkEkA8xQOKZUuSTA2g0ACCmQSQSfT0oHNiRvMzxQPJOoAFW/oKCS0mXBqkpAjigG8uCJUJG0UGNbT5YAEfLmgyWIfbYxzd01cRk7khNyXFlSUEuKGpSZ2I3SBttQWvT12h+7uXXVpbevC04lpR7+CknTQRX766b6ov1tElkWpYb8QgN+IgBZ78woz8uaBWsldm4LgfbShTTSAlxuAhZdUhSyNXbvv6b0Fhg8ld5J8lwW6LZDbZUgA6ypSZ9Yj+VAF3OXX369t2hbqLakpbUQrY+IlBnff4u0D3NAP9IXF7dYpLi0NgPNlQSSNail3jfjy8b/lQNzC3R1C3C1NJQLfzpeI06nFAwjhU8GeBvQSjk3UMMXlw0gq0vLbQ2tUaU7ebsf5UA7LJvsPfdvCVcOKuHC6oKJDaS5o2J7D0PbYUFzzkFhMKHgp4P+JVBVaY6sujIJFk3EDcStVBPSqHDq+HvFA7YH4QTzQEZMpJAI+dAoE787UHAxBBIM7yKBTEgck7UCKSRJT5fmaBU7iAQSB29aB42MncgUHSSTM7cUA1JKZJJP04oOQmUqg9p37UCElMiefU0CrCoGhXtv2oEIIG4lXvQJBI4Aj34oA2gi3QISdu45mgOYVCSADzxQD0JKRIAXHbvQJpCgBpH1HPyoFOnVqUhMnvFAg2cJKRA3+dB3hgKB0gEDsOaCY20glOpCfLvsKAbjTa3JISZMmRxQBuGmi+ZQkqAjjigGtpIMltOqPTegcUokq0AH58UHJYa2JCkkdyTvQcplsgSneYO54oHtsNAwAomY2Wrt9aB7ekCCXBqI4cVt+dAq7NnxFEl+CeQ85/WgZ91bEQ5cid9rhz/moOTbNidL16E9/7U5z+NBybQJc1Jub09t7lz+tAZbayoxeX4Kh5ouFQKCP4LigUDIX4BHa4UaB5S+lICcjf7Dkuj+lANDFyQIyuSgcS8k8f8NAZKr9s6hlbv6+Gf/LQR3TkVqkZm7TB7NtH/wAtAEfpdKypvM3JgRCmWj/5aBS/m9PlzK0kbCbdv+lANV3nw4UjMj62je35UA13XUIJIzKYjvZt8UEB7K9SNlZGUY0iTKrJP8jQQl9S9RI1n9I2qpEJ/sQ2PvvQNR1P1NpRN/YEHcRZ/wD5UEj/AGsz5Equ7T6WvH+agY71LmTOq8s5Cd5tOZ/46CLadZ5RNwUOrtlJCFkxblJ1BJI/bNB6olKbmwJdCShbUrSe/lmgfiktpsGE24Q02W0rCUmAJE/1oJyNSpkA7TtQOAllSkme8+lBCv7dFy0th9rW2rlKjsRzQRE4bHaikWjYA3MnntQEbw+PDqAbRCZJiJ/rQSE4PHeEsi1QDPYq/rQdbWbNm0W7dpSEFRVAUTv+NBmgSBsJP8aCuYu7FV3e2qmmGVJeS2rUEjx1lIV9diOaB4yFoH7dDobQ4ttTyiVo/VAAcn6jfigltLtHHChKrdS48QoSUkkEcx7iN6Cux2Ux9zbsFy1RbtPnwbcrCdKwZOkRMfCZB4oLdp6zSsJQ9bytM+VSdwn+n5UCqRZF8ki2+8LPonUuN/me35UCrRbBQlFuFNCZOmW5/h3oIt87ZtXVstTCHX1pWUOJQCoBKSr89496ANve4+4KGmmtekhAQEAhIWgLO3pBE0EjH29hcNtvMWiUpbXrbJRHO+ofPY0BMew3aXC7dqAhDSTttMrUf60FcsR1TfE6p+6Mx7eZdBOjY7DfegaQVA7yPlQEaSAPQH2oDJI0kAGZ9KBigSYAGkmIig6OSN4MbUDlABI0kiT3oOGxER6bUDkiE7b/AEoOIATKiQImg5JlUj8xxQIQCYEAH0HNAgAJO8bUDkpSBG+xnegQgcGKATiZbMbbT86BrAAZaUk76RIoDaUgj4dhPzoBqACpI2mNqBSIgSkD+FAzRCj6zG/egU6UniTFARtMkTGr2oApJJlv4poHtOEOcASYoCujUQsDgetAwgyuYMiN6Bh0Ek8b8CgdsqCAB7RzQKlAneSJnYUCzKSRsQO3egc2lJImSJ5j0oDKEpTsRvQIpMKKgSQBFAMgEiQqOI9aDglMFSVEE+9A9tKpMxHPNAMtgulRAJnkigctsEBM895oE8NLaDBMg9qAaiRsFAigaoyoACTwT6UDVnzbgGDyRQR3FJAPznfvQRwtSknw5Amd+aDlP+IkDzGByBzQUl8spB1FRKvUcTQZ25/v1az5R6GJmgOm4AQ4SFDUZEDiKCLdPqTOlUhR2kcf63oIKLpwnzAkRHm7UEpDYccWQZhpew7eQ0HuyPLjlpAJlmP8tBj3VJZyLRUWFKQ1ahKJ/XKHhr4/w/vD0oJ1nnbhtl8IVbtNW9uVFnRuISnTBnvJ+UR3oJScvkB9zBet1ofK4XAQFgLSkAiTuQSYG9AOzyDt/mbLxLxMoS8haEhJQ4YQRpIMHYn3HBHNAa6ft7Xqoi6fbQtdiEoStYTqJcPHrwOKCBbZq7s7cBsW4YCdDSAPggN+aSeBrP4c0F5hr++v7n9c7bpZabSHEISZWpQ2VMxG3/WgrrG3uvvmQeDyGGk3byC4haitZ1AgFPAAAP40AEeUbnYiKChusLdLyL14w4ykOPpc1alApSEpB2AgzpMg+xmRQNb6fuUKaKXGFpahxIVPmV+r8p/w/q/z42oLTC4kY5TynA0txekBxI8wASARPpINBAawN2LovuCzCisFegEB1QSpPiFMQlfmHHMc0DncC8VWiWxbIZatw0QkR5i2pKjsN/iB/H1oCt4i8F7bvEW6VfePEUQomEbbQRued9iD7bUCXGGuHco++6LY263EmJ+IBaFbpjc+U7kmg7G4W5tb+0cdSwptsErWFeYAt6NAEcd/rQBsem12R1h0KUW0oUD3Idmf/AEp+QigexjrjEOJukslYZbSlKW1gJMhKNERPMq9O9BeNpjKPq1A/qkCI43VQVhJPU9/A4tmOBxuugsFCUiOD7UCJBBAA3PvQESITBHFA5KIB3oFAkiRz3BoOWAkFRESZ2NBxAMAgAUCoSNOo/xoHxHInt8qBqh5pHIoFSnYTvv68UDVgECfWgYeAOPlQKAQogGAfWgeADG3agj3SNVu4lzdJBGxoEQkhtISNkgDmgKBtEgEb70HHdAnc8mO9Bwme2mZ2FA1Y3MgczBFBxbIWN+O1ARgR3TIPrQN0kKlII1ehoGxpSNYTzO9AgcAg8/XigXbWQBJ53oOUlKgQACD6UHAAKjcJiBQOJO0SDP40CkERA3iSPSgI2Nu5AHbtQEiBKtv50DIkGCDPc0DU7xIgR6cUDkpIQAeOd6By0kmAZFAFKTq0nsaBxBBSTuoHkigaZEyRIPegGoncjf5DmgAVHfbSJj50A1KgEbDfvQRnVJUCRA1evegCCEpUrZIO8DvQRX7lLKkoUtIBoKXKXkgpVKk8T60Faolx6CCZ35oBXb3hoCUSlPO34UFW/dCEpMpI5jvQR0PpcdQAopSRtAoLexcSSsGBrbcA258hoPeHSE49xQEQwrkeiaA2OYaNpbEpSV+EjzRuRpFBT5XF4mxWu6v3XgzdO+GEqUIStw6dtp/EkAdqC0xdpj7cPosnfGWHNTqiQVJXEDjjjsKBuUurXHFkOhpLjzigyAAPOUk/nG5oIttfWdywxcvqtwpYSE6lJJCiJ0z670ExDLS1EqbRp3CZSOCN/8AXpQBsM1bOPrbcYNs0Er0uqUnSUoVoMx8O8RPagZkM5ibaW1XLSFJcKVbjcwCT/mFBTEKKCkHST+0O1BXdQZJzFWQeZQ244omELJ8wCSoxHfbvsOfaghqztyLtTQYZhbhbZUVH94JlX49qCsu727bscA+2+8pS7VXiISSrWV6UJJ+SlAzQSrHKvN4a0tnSp4otUFTwcUHCsaJlXvqH8+aCV1BdXCMuwq1fcQ0llKlLQ7CEfrkpKlJ/a2kRQFus3eOLaTZNMoMFa9apkaXY7Hu2DQR7/K3bd1i2nXAnxLNa3ktgkFakHQZjsUn05oJbGccS0hp5gBbaUeMoucatMESN/iE8RQETn1LaUty0JQ02tbhC+AgxtIEzI9KCdhrp59q5VdCFi5cQEAzpSOBO1AZuFX9wUx/dt9/96grWgv/AGhyKlDfwmBv/wAdBYBMgzsmfwoOJJP+LjYUDfMImIG21A5J8wnYUDgrYEduKB5SDuoxFA0gK+EE/SgKJEQJE7SKB4EyVfSgEE+UyY37UHJEkajsPSg5wgiBIETMUCLSNIBImg5AA9gaB4gpEGI7+lBGvIRbuHUQFbA/WgINgfKNt/nQEAABIMGgQpChIMGfTmgQQjQBMgfjQLESQVTzzQIpMELkmfTtQJAGmJG/pQJrgEHmeQeaBiSDMiB60HFAJ4SNNA4iFQCQOT9aBwASQBv2AnmgckIIEie3PFBwIE7gJO9BwEAGAfmeKAiBEk7EGYB5oHLkpB7Ht6UDp2ACRA327UDUoUZkx8u1AsRM79t/Sg5YSneT6QKACEJU4SIj2oHHcmSR6UA1JkAqIG8796BqjMgAAAxQRyIJkkAqj5UDVJBBI9e1BHWkoBVPlA7jg0EG7XpSoFUwZ2FBncmqQFqC1EbiRzFBXl77wohSf1ZG2r1oCaEpbBkkjcwOKCqvVlIJKYIlMgc0FS2oOhUATO23MUDCVGYgrHG1BZ4lQXcEqACfCXsf9w0H0VelLeLuCobBpXb0TQEx0px9oSDq8JAJ+goG5S2cu7VtDS0JWh1t4awSDpUFRtQMxWMcs3Hy4+hSltoZToB+BOrTM9/MZoA9QWjlyzbrYW2hTLpUdQ+IFCk9vnQUbvTrri21OONlA0I8NWqAAlveARJlvvtv7UGpaAUSSZUlUme9BnbLA3jDrzifuQd8RS4SVQ/+sKxr27AwImgcrA3DDSWw1ZXOlWxcMEgoQN9uQUn6GgjNdp2BFA26s7a+aCL23bfSDqSlxOoA0DV46zLZbNswpChwUD2P8YoFTa2qENITbtpS0nQ2AnZIkGB+AoKy6tcOSpDT1rbOoKUKUnTqCQQdG/HA+lBKLeFU6y4o2KlMCGjKfKCZ239d6B33fDNq1gWYlfiEhadyJ9/dX4mgcGsUQGgq10kISlJWDOn4Rz2nagKnE2QCAm1QEtklMdp5/wBHjtQKxirJhK0tsAJcSpCpJOoHnn1gUC2uOt7Rwqt0FJIIJKiqQTO8+9ARsf8AaFyYj9W2JP8AxUEBhJPUGVlQUnSxA9PKqglJJSqQqRPbtQESomeASZ2FAgb84PEnvQckHbsQeCPSgVUSkkTQOSraBuBz70DgAkGBvG8Cg5BAJB3MSaAiSdj6j1oGqMLHASfyoG7mATHfY80CubpElUfwoGq2BBJgn8KDoIKdyO0+tAUp8ukjt+dBXZmRZHSJPiI7x3FBNUPNsAQT3oHFJBUABsO5oEEpJkz86BE8CIO+80HCIUTuCO/ag5UxuYVH8KBumT5SdM96DgBqExE9qBoBUfMYSf5UC7CUxJiD7UHNjsARNA8A/tRPyoFQ3A8p296BUpAJKt1RG3egdEp2Eb0BGxpHEADkigaRqUdIHE78UDwAlBJ+Yg70DG5JAJIPzoFXBAAJJI9aDnQS1IBn27xQDZBAkevpQCcSohckxPagYNpAJmO54oGaoVDikhQExNBCdMAHcyYEcmgel1Kk8hKYjc8UAlPpW2CRAIifWgpMr/dyjc8wTxFBmn3nVk6hH+6eRQVzDivE1FJSOBB3SKCYb8NJLbxSSJEgdu1BW3t0HUqIWIKoAjYCgioZCWoSEcHdJoEcSYJbRBG23PFBNwyf7bpgEBtYgnf4TQfRGSSRjL2QT+pX37BJoJNl5rZkGAA2DvzxQSWzsNQ53370Fb1Hk12DDTrRa1LUYSudwBJ9Px/jQQLZ1xzE5Fy+u0ICXXgH07eEkKI7z9KCuQXU44tvOOBpxRcaZS+fHMmEo1EzHfnbjtQSLB+5ayLCXXVLdUtbDqAqUkIbSoKj11Hn3oDYO/cf6iy7b7x8AlJYSoQBplCoJ5lQnagq792/VmcjZt3t02HXfHaUk7ISkBKkD0ElJigc66m3t3n1AlLSCs+pAEmgrB1EwWrdxLD6mnSoFR0+TSoJ339VAbUB05dlbbhDDvlaU9tp3CUpVH+YfnQOVkNGMF47aPI1FIDMpUpWogJ3BjeR3oFxmRbvXChi2WEaUqLigmASJjmZoBsZe3uXGEMWjy1LUpAlCRAG+rc8fn7UDnMvjw1aKDQUi4JQkhCdjrCN/wDiIoH3d5a2r62l2i1qaQlxS0tJMSSEj5kigc3l7ZxrWgOadTST5O7nw/h3oGpzlmlx1BLiS2op8yDBgkbfUEUCqz9mmdQuSkuBoLDKinUd42oJTRDl3crbVqSW29JTwZk0Ffap/wC28qZ7sjcc+SgmhPlJ9DQOEBJjeBG9ARB3Mk0HKMkBO5BoEJg8SSKAqQdgAf6UHEQYG0elA2dhyKB2qIG42/Gga4NIHb60DFQkjUNqByTz29/Sg4kEESdMyaByTpJHNAoMKBMn5Higg5gqU1bIBJK30DnnegnTKime/wCFBxIkiSZ2MUHFYIg8KHfmgbPmK+BMbUCa9iIiRwe9AinIAgbx+NB2ohI23J70CKHlM+szQOQCQNIO1ATSrSO3vQCBgzuN+1ByVSoq3IPagchSioiJHv3oHJVpSIKTqEc8UC+IIgSI9RQGQsbhBER6cUHadQBED5d6BikhKgRB7bmgcpBIJAB9waBB5ykEQeZHege+mEp1Jk8CO/8AragChUJVOwT+dBDXcjQSCkgehoBIuJbUVEpEd6Cnvb1JWsqJIIoIjWRUXACCRHrxQDdyAdSpKD370D7e6cISkkaOCCeKAeaeMoCe6fijigy16taUDQSFE6eaCE66pAEypZkH3oAuPlSgDMEQZ96CudBLoQpUpJiQaCUyQE7GQUzQOKXUqLixpEzB9aCXgQ2m+UtPwlpfB52oPorMEJwV6RJm3XuP900EtpsJZRJAEAce1A8wShIIMHkUDLm3Yum9Nywy8lJ1JDidWk/WgrL04y1bLd2m3Q3cKVrStA0uGN59e1BXA4B9sIULBSU8IUlMDf8A6mgksu4tu4Lrb1k24UhOtKkgkDgTQHaVilIaW4qwJaVqbIUnyyZnn1oJCk2jwU/bNod1LMqYIUSe8n8KDNhAUleoBaVbEEetBneobdAyWPbQghKULIQjZJJWgbj6z896CDa3bKmCpS9Idt32myoESsNtDSPqD+FBcP3TKul7d9q5Nwy0tkuPKPGlaSomAOKAPSCSpx1xTqwktMkIIhJlM6vWeO9BFtLhti9bcW792ZacdUBPCimS3JHqDHzigsukGPAwyGHQlTrDq0kxwSdff50AcsSnIXqy4pLbbFspQIGnZ4kk/IUEOyfZRbK8R1pJDtq5ClDdOr4vlxvQVuaZdukvpt3AHG1qUrSePM+sA/gk0Fwwos260OOolF2wNGwg+GiT9aC6wG1kgnaGGRv/ALk0ALAg5bLAmYdbH/7YNBYkExpHf8BQDcRBHqfSgcmQrSJI/gaAmsaZAB3j50CAyRJjvtQFTESDH0oHgeUgfWKBhSAkJgelAgOn0kjvQI5zuUnftQcRxAgA0HA6QCN5oOUSYiUj+FByTuQk7E96BZidzFBBy8arEEQlT4IJ9gaCwRAbAkFREzPNByGwFbGSR37UCQkJ0pAj2oGqBjfee3pQctMbgcCg4iEpgAbH60DkJLiTA/GgRSYmRBnaTQMSrQO/FAUjUkRO/JPagGokyACRH50CCSAIg880CJMKGr4SI3oFA2OxAmflQOGw3Ox7gc0BgEkGBJidzQPBJQQPhigQklJCY789qB5AJGnttv2oFCQVkpMj2oEf2AA45270FbkypFquDAV6HigzgfWlwKC57gH9oUDcgtSYCSkSmT7UGfu1rJKlL2mNu4oOtVqJMKOonfSe1A9KSFhKQVKkmR2oD24VrXM6jsIPNAVReUlJSkFJ23G5oKi/slqUJKQs/kaCGLVQaCSjSUjdQEzQQ/uxUgrSAFxqEnagAmxdDh1jbvHY0EsWwUSgIG5AO3pQByDOmYJUjkA9vnQGwYBvVbpCEtrVEc7UH0Fl1D9B3gAk/d3IB/3TQR1ZhbDqrRuzceeCkto84SlZhM79o1Dmgc7mtMlNo4qHNMBadkCdSvpB2oC3GZUDoZtFqcLgSgFaQDKlj/yE/WgiKz7SbhaHLZ5P6zTIKTCdRTq54kfOgJe5E2uTcYXbOKShsO+KCnSQVBPEzyR2oI/6XYLLribR1SW0heoBO+pRSBufUUEzG3zN+4UMWTikNokuKSkJJIkJ5md/SPegC3mUuMoNvjbxKipY8JCEavKdKid4idqCrQSSO3y7UD07gmPYSKBpTChPw8b96Dl7JAPEyZoIyrtDbqk/d7okbSlkkH5GgcMgiN7a7JJ7sK3oOOQbEyxdDfVHgK/pQDTetuupR4TyVr8oUtlQT9SRQSQy2pXmbQokaTqA4/1FA5CEaiA2gTzCQJ7UHOWNq6oqctmFrUAFFTYlQ7UArVR+93qCIKVIAgf4BQQ8cJy2XmCfGQN//wBNNBOMBR7ACg5RlIPG/pQdpVwSJO/PNA8FJSFBPO8EUChICQoR8qB0kGNvXnigIkyJJOwoGKICiT3HrQLpgSJJO4mgaoEDkA8bigSZOx8w/OgedQJB323jvQKlQCTIk8zQMJMgAQJgyeKBRI2T6x86CrySVO5fGAKhKVLVpB4IFBagKICgUnvvQLJ1JAHliZj1oOVAA2JExzzQIlQCoBmKBpMzBHMcUHOwEgEjcdhxQNZd84TJMb7HigKdKpI3SPXvQNMQCoEelApBiB+yPxoB6QZgQJ3k80DpIJjkjvQcUhSQSNhvJ70DVJ8RQ3IMwQDzQF+IyqBB7jtQFSNBECduTQdqA4BjnnmgUhSYMBJ9PWgUEH2gxvQKX0oBkwJ9KCmzuXFmQlI1EjaeDQUV9l7i6tynUNJEGeR8qDsdbh1ta3TJngUBchZLW4QDpAEb96CkvbBS3NCQSmRuR2oIjFp4DxbcSYBJMjmguLPHSoLAJ2kbzvQWKsb4aRq23iRyJoGmz0yEjUF9jQRlMyoIWQFEykkcCaCpyNiovDWJE9jANBFNmEHQqCnmT3mgf90VqV4cAgwCRxQSGcegIKyNAI/GgoL9lZWrwwSiYIPegkYa1SbgaPhUnSQTxuAaD3fMCcde6CUkNKA24oKF9xLGUuHbh15poXaCtwT5ANI22MT3oAuXwZfU8lCgz4rgSktqUkPLSfJxwYB9JVQEXespuEeO6u3SxdobcciNCip4wZHpH40EQuC3eWtuS6brU0pQKgqXFAtx6FMkeh3oLLqJtSss+krPntUhJTyn9cgf1oKlsgWTtubpaFraYCWwU61frD5+O34UFv0i42Lx5sXI8SEH7uqJjw0Sr1/lQAs7y3sHkG7fbaSrx9KlqgE+MqR/Cgjvu/d7Vby0yltJUqDvAoK7H9SWN3bh1JcShRVpOnVMfL6fOgK5nsegEl1YlOsgNKkAGOIoJV3kbe3aZWtZCHiA2QCZnf6dqAFpl7W6u7e3ZK1LdSpc+GUhIT6zQKvMMIu1pcbeShLnhB3T5CqJ7b7bdqBpzVmXUIWVpUsydpj8PXtQHYydjdPpat7hC1qBUkAHePp86ALWWtHL9+0VqQ4wfMVAaSY9jPpzFBIbyrCsci8AWptWwATKiSdMfjQFTkrMoQo3DaA4JSFqg8T/AFoEsily4vVJUFtqKNMHYjQDNBCx/wD/AFDLkgT94T/9tFBYrE7AmCJFA1InVI/EUDwDoJJHrv3oFAIAkbK3EHigakkgTwfrQPSQDJTBj8KBqUkq1HkUDgJc7Sr1FA9SNMJB54oEJAMEyCe3pQdAJ2CQTtxQcEwYmd+9BxMTM+lB2kKieCe9A4wYkkn3oKi/OrqDHJAJhtxRjt2oLOTICY0kdu00DwFFPYj2PNAjgJIkCAfTigZMKBIBIG8Cg5JClDSdp7mgFcOQBBAjbmgCghJAStKlEySD2oJzRKkFWmDxEc0DYKlTMHiI5oCLJShMntvtxQDI2AG5Jkk9qByYSSOD70DlDYFBJTG4NAwjYxye/pQOSCG9gDvt/r8aB6PhBAhM9+1AsBBgnbn1mgLJ2J4Hp3oKHN5dtllaGnR4s/CD+VBlU9R3C7loFZDaj5hHFBKyR/S182m2U4poCBB4jvQWzmGbDaCk+YJhShyaCyDLdtYjSgqiEnSOd4oI+RSVK0pST7DvQBbtJQNZWFTOw5oDHHpMKUEknuRxQEtbBFuoKTO53BPFBMW0lQMokERv2oBFtIEIAkDigq7tgOOlZnUDpgdqATtu04T4gJUTIjegjuWrMEIG4PJHNAqm2pQUgqVO8DegR5rxG1hSdIkTtQUNwzoWrYjbt3NA7GMhBK/KDAGk87rFB61mFH9GXJM6CO/eTQXaT5zBPrzQMIUSQkwmdt6BqwZBKe8xQR7m9YtUNeKHCVExobUuPnpBoIasjbhyXC6B726/6UCHJ2fxK17b7sr/AKUDU5Kz1IWokbRqDC9/8tBOtjb3TSVoCXGzMEtn67ESKDJKa8ZlbTgCm1pKFA9wdiKCDZ2OJ2RbtNBTSlQgEgg8K2+goGqbxCgyohsB9ICFEkagd/8Ayn5UDbtvHXSGUOeE592VobHiRpUBxz6AUCYg424fFzaFCbhsFCfOfhJMbTBmDHtQSnLLH/fHHHUI8Z0wqHDuqPSeY5oGNMYlDjFw2tE+L4LavEOkrB49zsaCQxZWNpD6AlHggq1FeyU78z8zQMTaWV42paCSh5zxtaHCJUBHP5RQSVY9hdgq1JcCO0L8yYVIg+xigjLwtggo1BSCSEpOr4oTpjegn2yUpfuSgaSVJB+iRx+VBW44hd7lyTChdAT/APDRQWCQd4G0d6DgJjfYUD4JkAxqEetAunaBttx60CpQQCVEED2oGbKUICgAY52NAUgAgnmaBCSpQ0q3j05oCK3EEDT7d6AagZEDedpNByQQozvBoOBkTEH370CK+I6eKByYngEmgVYggEg9uKCpeCVdRswfgYKueJNBYJTHxLPymgOkwQCIAG1AixKdKhBA9aAAAAgGSdz7UCObIVEggSDHBoPAc1fX95eLVd3K3So8lZ2+lBFYcW04fAuHGlIOqULINB6J0N1NkhkGLZ9xV3bOKCCFnzifQ0HrF02gthQAEmOZoIbsFI3J2jc0AZhUEgmfWgalR8Qgnmgk6PKAd55oGpEGDuT6H3oCK0qOqSUn17CgeYKQUkDf15mgFe3DVnbrdfOlA3+dB59neq3rlxLdk8W2kKJ53JoIIsrm9JcWpSSsDzjg/jQTMN0k+84XrpRbSDO/eg2GIw7NgysJTK1/tE9qC2bSlLUJ+KewoBrks+UEypOx7b0CPsgjUUQoHseRQd4QQPKAZM7igYryqJG44g8UHI51cg+/FARSk7gzMfjQV94dO4PA7CgrLgKhak+Yc0EBbqtXMGI+VA3xSkbEqMevFARDydQUFRqO/wA6Bly+QCUiNp3PNBW6B4qluLhUwmDxQTLVKG2wQNSgpsCTyPETQek5pJVj7j9pKkgEA/4hQDYzjKVrU+hbTYcU2N9Wop+LjjjvQSP0/YrOjxzqlIKQ0raRtO340DlZq3TaWrq23kC4ZLnmQfKBGxjuZEDvQAvM5bsX1nblK1h9vx0qSRp0+8kH+negOzkre4uxasO61qSVAiYO5Bj8Pl6GgZb5a2fW2krKHXVaWwv9on8aApzNih19C3i2GyrUVjbymFUDns0xavoQGnXlLRrSWwD5Z53IoM2jiN9jNBXWmIatsgbzxVqdKnFbgbaiTH0k0ERvAtG3LdxdvPthvQlSgkBKIUBED/Ed6BHulrNUlTruxB2QkQAPl/rigVrpy2aas1JuSRbrQ4lyEmdJV3/4iKAp6dZdvbh5x1ay6vxN0DyH2P8AOgE7gkMWDDH3hYZt3VXEpQEyqZG3G29BJtMIwwq5dYWR4zAY06RCQO+/4x6zQAd6fWphts3YWJGrU0P3pMb/ACigcnp2W3EquQFuByYb2Grb1/H1NAVrAKaurV4XKVJZdLitSDKp9N9vwoLS0Ou5vY48QAyePImggYgk3eZJEFN6Uj3hCKCySARHO9BydydpBoHNk9gDvHPFA5QkngE+goOJ07CAJ49aBQNIBAlRHFAjmw1TEjeg4DY7GaB4MRA3jfegUAQR27RQDBHaY+fFA5SZk+g9aBipERv86B49t94oHEwI5J7xQUwSo9QvkH4WE8+5NBZd0kHfg0D0CSP59qBzm5HmJPMGgHBKtyJJoId88W7K4cA2S0oq9tpoPm+7uFOuEgqMmQB6UBLRSUqUkBZChG1Buvs+KXOo8YkBWpCp3oPb7rdsEaQSdwaCvVJOnvE8UFBluqMLjHyzd5FtLyT5mkgrIPvFBFt+uOnHHABkAmRHnbUkD8qDTWt0xeWgdtX0Ptz8bagoUBEShRhRA/iaBxVCVEnUrjagg5XM2eISg3ayEuGAANzQYnq3rNm+sUW9ihaAFbuLI3HtQU/TWIvL1YcWyCwsncig9GtcW6gNtOqQptH7AHEUFwygNcgmBtvQGSBI+EmY+VAjhAaUO3qO1AxfmaRuE/rBO3EdqAyiFJ3HA7CgjQpKRIkEzvQDUCRJO/zoBhUiREDYyKB6VFTY/GghXiSoEElauDQQnAAAAo8flQQLohDggEnvHFBGUUlsqIUBMqB7Gghh0BogEADdO/NBDfula/C1jcRxzQRkuFVxAWQkHcDvQXuPfQ6oIUuNbrRSD6eIig9PySy1ZPFJKVSmCDBG4oB/oiyV4gLSzqUpwanFKCVK5iTtyeKCE9ZY62u2fFCWVatSUOLP6wpTyZPt9e9BIfGLNuww6pOgBTSUhwyByd59h8qAl0cawloPLbabebDKCV7FAIgfiofjQNtbHHJumrqy3dRLaVpcJ2O5Tv29u0bcUEe3scW3kUMNvrNwwrxiwXSYkgj6bbDigsLvE2LzziHSsrdBWqFmRJ7fUmgLbWlot1TikpcW2kM+bfSBvHz3oM0g7iQeJmgV9rx7R1sHZaSnc+oigobLCXlvjr22dfQpL1shhpAWopSQmJ34HHH8aAuNw97bN3ou7sPoeYDSEKWpQSYI79veZNBFtun73wWmH3mBbJa0qbbKgNWvUPQGIG+xPeglW+MySMPbWqr2bpt3Wt0rV5kydp52kGOO3FAr+JuTh3bNbqLhwvJcSH1qUFAFJKVGJgkH8aCK1iso144S7DITLTLDyh5tMd+Px9+aArmLyj0ouHUqbgAQ+qR5VJJ49x+E80BmLTM21w242tt9tCVa0LeI1qgRO3tzQX6SSkAbSPWgjWKQLi9jUB4/cc+RFBX4cS9ljG5vV7j/AHUUFkgEyeD8uaBw+EQT7zQOSfNHaNqByZRHegVfxbkgjt6UCIGoynhI9OaBo1SdwCDHFARIgjedtqAihA2Mie1Awg6RBiPWgaU7yIG+21A7hQJKSDQIEkKKuY7+lAxKu8AkHeO9AiiQNz+BoKu1UpWdvgdJKW0CT2oLMJIkGCeTQOlQImQYoGEkEmJTPrQMBMHygme/agrs+74eEyCiCCm3X/8ATQfOK30hBSGlB3kEf6+VA5Dy0JEtaQRIkRINBvvspGvqi2EKUW0KVt3gUHud0ohrygmRq4oIBUICgkgn2oPnHrB4N9SZMqTpl5wiBzvQVNk4okpUFAKHBPFB6B9lj7zvUsNvqQjw1FaQdlAD/wBKD2JLm/PlI/D/AFvQOU4mQCZJPxRxQeYfaT1NjrtLFlZFL9w26S4uPKkbiPxoKj7OGWclnxb3bBdZDZKgR8NB7Lb2TFqkIaToQTEA80EsQTAEQI3PNA1xxKGxASD6zQDt3EpcWTBnjfag5SgJ80j2NB2qVoMg/rANzQSzpLYKCApR3k7UEdZhJgdp27UEdSjBnTEgb0HTsEqHlO30oGyUqMAARp55oItyqSoQD/KghOkFUgjjcUEZwhSQFISY9RzQQrhpQdBEfUwSaCG80W0rUpAkCedqCjeMOBaU8d5mKAIuSAsykQYUfSgvenwF37CkiVKcaiRuB4qKD1bNtarJ1CRqMo78edNBYtkbydxtv3oK69w7F9fs3S1rSppGnQQnSd+TI9zQV7/SjDil6blQ1+JyhJnVP8J+tA/JYBm9srK3cuFoDCCgKSEjUdOkSOBBgj3AoFY6eZdukOh9QUl4PKASNyDPP4b+m3egm5OyTfPPnxy2FIQ3p0yBpUSfnM/SgDaYMNXjL6LnxNClnSobGe/PxDbzfP1oI2V6fcecWsXIbC3VuANo80GOTPtv9PSghRJg7giKAgkQTvvFAqSpQJIj+VA9ZJTBiOdxQQlt3yidF2yEEykG3mB/4qByG7zYqu2DPpbnb/NQctm/KpFyyNp/uT/zUA1IvkPI8R9hTc+YBohRn0Or+VBPCpIJJ9OKB0ERBj50CgSOIBoI1kgJdvZUTLvrx5U0FbhQEu5QmZVfOfwTQWoJSkkGDMb9qBqRHEz6ntQFBgk7CDNA4jeZ/CgaNgSfwHag7kcEb/hQNEzAkKG+/agMlOwJE7+vFA5QBG5Cd6BpHBI5NB0blJkDt6GgQbxO+/4UDVKIHlJJPoKAWkxz8O1AqgQmDv6iKCjwzinc7lwCSlCkp3PtQXifMR5dgaBJEEA8GgcEpSPNsqe/egGSZBJIEb+9BT9ThKsBkElQALR54oPn2+H61elZJ1RBH8KDU423Q61b6yPFJmSJig9A6HwjFhftXTKSh4FSDHdJHpQei3a0lsDURv60Fa6rSymBvPeg+b+t7pN1n7pDQlQdXJHfegp7ZYRIckAiNvSg9K+yBBObdUknQlon8YoPXlKmAIO00GP+0jqK4wuMaYtGUFy7SsFxR+ADnjvQeN2iVX15DTSkoIBUpM+X50Gw6PuBhco044FnWdKlE8A+tB7AVKHxK1Cdp7UHG5JEKAjtFBW5HIJbcSAkqKT3oOtb1DsQB7bc0Fio6mtwBI22oEB0usAgmVcDvtQSdQ0kgbdt9jQIRqT5iQON+9BykhTZOwEb0DRpIO5JAiJoFKAU6hrCojc0FfeOaVKJ2j170FYp0FYSCoGeYoBLUgFMFRSPQc0DVKDgCCoDSfKY3FBEvwVpRpTqkwfQ0FK/ZLAJJCTOyQNtqCGq1IcRAJTJ78UGj6ebSLphYBCS4yNxH/tEUHp+STFoNUD9a2mVDnzpoLE6pIABg6dqBW0yCtJIMRpJ2oECtSVagDJ3oKrO2r1zYFm3LalpcQ4nxDCTpUDHf0oK1zHZBdwX21obdcP6wIdMAGSRuPl23oFbx2ZSwEtPJCg2EpbU9MEHiY7Dv3G1AXHWuVsb5WkJcslkgoU98OpaiVDbfaNvpQaQLIAmVJjaOaDznqCzeyGHuLW3SFLdKAnUYiFAk/TfbvxQV68HeNEtspLyUtrQ28t7SpEoA1ae88R2oB/ojIoQUKt0XJFqGlLLwT4igsk87jUOewmgt/ud4nppFqQhV4hCSE+IYJC9WnV8tpoA4PDP22QXe3bqw4Wg2GkqlJlKZk+xBj/rQAusTf3eYuXHFpFo66hQhcwlCSPh95oOXiblm4R91ZUG/FfJAe4Co0Hc9h6cUB8fY3THTtzaFxarjS4G1lzcz8O+8UDLW0yDfSyLZxTiL1JBSSvWpIDkjef3e0+00AWrTLNfrCh9xSGUkkPABcFJ0gE7HZR37GJoGOWmWaRcukXbkto0ttvQSpOn1V334jvM0Gmx+o/eTp0qLkH56U0FdhioryRUNzfOnj5CgtE7AgiZHbtQKmBsAPlQPTJBIBHaD3oHEqKfXaIoBgSoA8ERtQOSI52nbc0CgEKM779zQKlYMSI3gA96DgVEkkARtvQdukGBO/5UCgSRtB9jQNJAnn6UDCogkCSB6UDZlJk7T2oGL0wAB5ge1BW4u2dtby7fcaQlL7gVKFTMd6C2WTBCYHegYonVEDc9jQIpUgkCQDQCkCAoyY1RQUXWRUOmb08EpHHzFB4o2ym4uzMpSFaiR696Dc4i3ZaQQhMlQAGs0G06SdDt6dS4hJHG80GteEtSrdQEiaCC+pCbLxSNRAPlHJig+XswA5k7ha3FBa1FfHEnigm2lj4iEnUUoI0qJE80Ho32UWi7a/fCv2mtIUngkEUHpyVebkb0Hmv2w+K47jkNBSgErJ22G4oMdiFm1cJaWomNKo2BoNDa2nj3DGopHmTueBvQejX134KdKQSD60FYvIIYgOPJQJnfk0Edy9YulBIfRJ942+tBY2ACyjw0cHkHmgvwzpKQTsBJmgBdApesoESVHn/DQSAkkcwY9OaAyZUNJ2A9aDlzBCSkJ9qCP4hbHE7wfagal4BuQSPlQV98oqUYSdxsfWgo7lsg6kpUTOxBoOcaKpWVEgqnagkNAgBKQNgDxzQLdSlpKAI3lW9BUXaFIXBPBkSeaB9qxrVCSNMTvzQWmISg3bCCCD94Y3A/96mg9Jy7RNsgpgkPNnf/AHxQGQpXiBCUnc96Bw+IgA7COKDklJCgSADxI2oIN1a3rrmu1uWWWgIKHGSsk+oII9qASWMmJH3i01c7sqB/+qgelWQBPntFL7HQr+vzoCIOQCSlJswY5lW9AUrAtEG8U2FAgK/d1R2n60GWZ4JIkRQSAQQJEke9ByZBmdx69qDlHcAnnfaggKXkCo6RZwFHTJVJHadqBUqyCQoBqzIAn+8VufwoOKslBT4VmTG361X/AC0DmFXhdR4zFshB+IocJP8A9IoJgAJkGYG1A9v2Plj8aB4ImJBoI9iVk3Qc/wC/VEekCgrcJKvv+5P9teO/fzRQWZkD4t4iJoOROoEmR/CgKFEyDIHG4oFcIBgDj8qBAqQEgCY79qBuyoMyATyKBylbH94idqDpkSD+IoHLIU3OxFAwjYzIHHPFAoMgbg9oPegE6qJjcxtNBxVsQTE+tAOQqQRECfnQcokwpIBAM0EXGZG3yTK12hWpLavDVqTEKoJbnlSdRj+VA0hQTsQRQdJUCQJHv2oBk+UwYVFBmuvHfB6duSF6ZUnt2mg8asHCi9KwslJc3TQbK2yLbroS0dWgCR3A+f8AGg1vSqtORGmAotyY7UG2bfU2JWjUFJI54oKq4eC2yIWgAEie5H9KD5zyCdV2vxOQCT3negnWm7bRJOypPtQen/Zzsq4cIJGmJG280Gy8WSSkEGYVQYL7Rbht26tmvFiWyoidxv8AzoMK6lZdUpAUEKGmEnigv8E6vW1qUoArSU/U0GxfuPHuFo2MHsZigos2pSL9aSEqOkRNBWNqKDpJkzuAeKDcfZ+S746lKMIAEE8H/UUGweVAQqTM96CDdlP3y0CiVJlf8KCwUdbcTJHoOKAYdKDxG3c0AnHxq1E8H8KCJe3GyoIE8H60EVL6gydXlE7SKBXU+KEnV5QYgd6CI4FA6hp1AQNRoK9TqwsjzJBMSBzQO8dbUFKpPYTxQFdeWpAUpA1e3JoIl2VuqQSFR2I70DUgMACCCO8c0Frg1TkLYqJIL7XA9XE0HpGQKVMITupPityB2hQoMonB5Ju1eS4EPqUoFKC8UwIUnn2htXuZoOfwWTL167qL7bhBQ2l6CQVpUrnbsfptQWGMs760yF646nWy61rZSpydKyTKR7QE/iaCttcXm2be1SVLUUpcS7D0hQVqUDv3CoHyoJwtcgjAISpm5D33jV4CbgeIGp+HXPPB59qALtplyu4OpwtwotoQuFcLjcnc8elA64t8ncPvqtmr62DiUkeI4ITukR5Vc/ETHagEU5VKXmnRfh0KbIWsnSRo3EzEz7CgrOoUXRxD68cFm8aKXW0JMa4UDp+u4oIJdzVuFJaStxviVIC1c87n0nagG4/1C6y+HEhqQ4keG2NQ50wZ7QPnNAXJu5pnJXL2OaU6wWG220qggqCgVHTtyCRPtQPs1Zrx22lk+CDCnHGwSPMSeD9PbY0Ay/lmLu5DTV26VXAHmCVNhsqO6O48sT70Ei2fzDL7rbjK1tC3UtLikSouA7CZE7dqAPiZd3H3byw8i6/VKbaCNOnsoD8zQNZVnEtQsr1KSJITulQJ3+REGKAjl1llYi8KkOsvNuI8JTTXnWiN9jPPsDHpQFN/k2SEBlTpKlQpxlUJlQ0yU0F7ZHd8HnxlTHqIoK/BqKW70g83r5/z0FgNgQRue/rQITqAkQImgOkfqzt770HKiDO8iaBD5QAYI+XNAgKgDpECfSgRR3AjaKBySTAkgRtvQITpJBG5oGlUJJBJUPSgapcATI37UCE7DcgjffagafMqNyO5NBylJTEAifbigR9wBlREEJSTQZr7O1T0+tc6i4+tXNBpVbJgfLmgRCk6QeJMGga6qBABjnmgA46CkgEwTzQZf7RdaumXdIKleImR2ig8YYKvGGqUnxDx3oNRhWSVLWgBJIVAJ53oPReikpK1rKIWE+nrQa66UUtAlO6ttqClyN3bos3VFUrS2o6I349qDwZ1hT76i3sACZIoJWObUHgmCZVG9B6d0glLTbyW0pHlGop9RQXNxdAQ8CorIjTPNBgevkhOQlJCg40DIG6pJ3/pQZ/HAu60pRO4Mk0FswglbRdBRuNKk8c9qDWYvWpQWptIjZY7kn/QoKfqpGjIvQAEQCoFXtQUqVTMIJQDwDyKD0T7O5LFyTO6k+U9tqDVPiEoKgCJ7d6Ctd8Q5+0SFpKA0tURueKCzLgJ3lIj8aBjjiS4ACmCPWgApOqSkGT3oBOsfCsgKE9u0UAy0qZLYCf2jHFBIgBshKZg9qCuviPDklEpG23NBUEagsqO0TuaBiW1BepJ+IdzzFA1LqkGAoyU9+9AibqETwZ2AM0Ai+FSACZEKk7mguenTOUtYEq8ZraefOKD0jKQG7fTBBfbEDg+agkpSFACJETJoEJSCqCTIjY8UAl3ADikgnTECRxQRXXbwrT93Yt1pCdy44UkH6JNAiHMiUw5aWhk7RcHb/LQEDt8FAotGio9hccf5aAgubtDYIx6FKiSBcJ3/KgkvyptBcTCiASkGdPtQZFKt1TvpHJoMzdZu6bzVzZseBDaUFLaknVB0yvV6bkcdqATnUtxbW6lP2iHXQ3rhK9M8cSOBO5PG1BOyGauLK8QDbpct1ttqA1ecKVqntHYQKBl7mr23GNc+6JCbkS6hJ1luSkJg7epoADqtxSUqGMf1lgvAhUpEAnSVR/hNBMuc+4ptoWrA1vMIdSpZ2GpQBER29fWKAl1mX7fJO25typlswHP3oCDHP8AiPaOKBjmZdW3ZXLTASw5bruXQdynT2mR+MH5UAh1HuHV2wVblGuUOAq+FJEj5kj86C6w+R/SLK1pYW02IgqUkzt7elAbHFB8Yjb9escdwaCvwGzF5Jk/fX+P980FopIKSomY9TzFA0EmCOOdzQGQVGDJmg4cCd1d570HJJ1EqJPbigQyEyD+NBygQCQZ2igUAgbwCe4oGrIUZ/ZoGgjedvnQIvkkBPHcUAoVAJMd9qAhJSCojeO3eg5YgCdzO896CNlCGsVdLkjS0skg8+Wgo/s/b0dMW6gSkKUpUc8mg0JUkJJggT3oBNL8qiJKY70FHlM222oNokhJ5A5oG2mQaWoNtOeIomTvwKAXU7rNxjEsqWk6jqg9tu9B5acMyHgUPhzzE7bCgtrJ4FthABJbHwg9qD0boUQXSrSfLBHrQaS+RKClJJMzAHFBmstbhjFXa0mXEtqVqP8Au0Hi1w6gKKhuqO3fegsLLQFAgqSZ1Qd5oN/0QrXb3SlKJ0AFRoLhLYVcKcUnV4k6Z/ZT3/HagwPXSlNZ8pStKUpZSEhXb5UFFZKTbq1BYkjsYoJzTq3btpttUJKhsSIoPQcba+E0l077eZI70GX6lJ++3SVEKe16YHbYUAbRjVoQgIUnVq45NBveh7VVqzcJSQptwjYH4VUGguTAQQElSRQVylRnWOxDCzAHqR/1oHuuqUSkBURA9DQSGGYAVIiOKAiUEqTI4PPpQKpIg7kn37UHJa8u5B3mfWgEfKCIBgd+9BWXQJA2hMTxQQnLcDWqDA325FAIJUdyEhPHPFBGuggMhRAT242NBAA1PrSBMe8UBwyoOEqTpGrgGaC86cb05ZiEpSfFa39YVQeiXYSpLSFFIUH0FIPrNBT27mVS0ltDFylxXxrf0r0qjcjf4Z4H8KDg7nHHHEBooKGx51NCFKCSSBv3On5UETHMZlS2mbgBKCsEuOM7kEqJnfbsAO1A/Lpv280p23RdqCLaGg0kKaUd9Wr3+GP/AFoBqyeWbWwkM6lLCl6EskqSAB3n1n58UHW97mJ8RTaiFpkteFGklI4+s0Fki+vbjE5FaW1M3LOtLKQ2QT5ZBgzO9BGYymS+7NtPta30SFqLZ3M+oAH5CggAcwY78cUCgQkDYkD0oIt7kLWxLAuioF1fhogTJoBWuZtr24YYZQ6S4hSypaNOmOxnvztQD/T9klm2N0FsuPsKfCAnVCU87igc51Hi21FKnlSEqWYbUYAEmfpFAPK9RWtgLRSkLU1dp1NqAifofpQFT1FjJKStfiDVqSltSoIIB4H/AK0BDncekL1lQCUz5UKVKZjaB8tvSgtUttlKdITBHMUDktpSoaQAD6CgBYCPHkkw+s7njeggYJIQzdwQZu3zv/vmgslmQQSIPp3oGJM7EiR6jmgJqKYgjSNuaBXFEKBVJPt2oEbJBMDTtO1A87ECNvbtQNUqEqjiPWg4K0ySZnt6UCgykSRzNA0BOw42jegGZKwAJHqTzQKqFAgkjaeKBwkiYgzQconcESSfwoKvqdwNdP5EjdX3dcbc7UEbolJb6WsEx5int2kmgt9iTAg+/egjvuEIUAidu3eg8lzWQWm+cJCEokp27d6ATWaLRQUAJAMbn0oI1xl3LhwkOL2HfuaALVwyBpLqQVCQqOKC2t8q/alFvZWjD8AJ1KRKiTQekdGLfKfEvkeE4psEpAiDJ9KCJ1B1au1vnLZWPeUlCtKXJ+L3FA7LXKHMPdPCFK8FWxPqO1B4q8pPi/CTqMDf4RQWdkVLbBGwnQkAfGT2oNz9niFIcv0OiEhSQrfkifyoNQ6wkOhxMnSNQg7Ef6mg8s+0V5sdSORtpSjnvtNBnbV9ThKlLURMcfDQaHpdIdztohZUUFfpM/jQerWrKm3ATAQRsPSg8/6mgZu9SoSpCtUz7UCYxTqkiAsJG8EcTQegdKJT93dW2Turgnnagtbpwh34o9R6UFPcrWeoEAEAi2Pfg6qCahs8nfUDuKCwYOwAgnTtPagcpJBIO/bagRR/VkAwY7UCJOgJII45PagjvHSVyT/Ggr3laSQoggHb2oArcGpYKgJEK9DQAcbKeAACI2PFBAuCAFdwfTtQQHNSTI1JKttqCU0oB1JKojdW3NBeYBQXlLUDb+0N8djJoN/dkBVsShJl9Bk8jmgsFJhQMSNpNBW559+1sbl+00quEgBCXN0kkwAfxoKNPVOlhT67RxWlCFbCFeYE/D7AGgl5W9u7fFt3rDbXilKFqacUYKVbESO+4oIVxnA2bgC1KvuxhxQM8KAMDmdzt2EGglZjJjE2jVy6ytbaiB+qOogkdvXfvQQ3uoFNrDngamlEHZUKAM9voKC0vr9VrdNNBoLC0KVqUvTEEbf5qCkQTq7STG/egKE6hMAGY5oIt9YWl94P3y3be8FRU2VDdKvUUCs460ZuEvtW6EPJSUpWOQDvQRE4nCJbLZYtNJBlJX689/agK5Y4hxsN+HaqEaY1CYKQn19ABQPessY4ww0pFsplhJQhJI8oO0D8BQAXisVca0KaaDiySShcFUkE8H2/pQHvsDY3Nstnw1NBYHmbMEbg/wAhQWFow3ashpoFKJKoJmCTP9aAxMx7+nagDZAjxtty+5yOfNQVuDhVo+SAP7U+TB/94qgshOoSJn07UCKE8AyD6cUCzJidMHsKDlGVAzA70CJGqQlX49qDkGCdRlU9+1AjhVoWZIMUDW3VFxaSiEoIAI7yJoH+YkQfxFA4qCQQIG25oGaVAglPedjxQDKtb6gdXl7x60EPL5ZjEWL1zcJUUoGqEnzGgDjc0m/trZ5plYZuBstUeX5igh9a3bbfTF1oWnUopQJPO42oDdOXbQ6cslSYLQJ0jgzQSk3RftlqCFlXngHvB/8ASgx6znHMWXbtDbLqQtPhhBkx8J5oMS1YX988+UsFRSmTA5P+poILrDpfIU2oQIjTEfSgkMWt84lpds1qbSr0kAetBYZnC5DMZVSrdhpppJACkyEkGg0uL6fyuOtAlldktex80n+VBoMWw/bKWp1wErTpVGwPzoD5oPAAo0lABOlcLB/H+RoMblMjcPWK7Z1hDSwSFeETB7AUFbjsSxfY862Cq5nStRMFGyu3zAoK0WNzaXQb0JWWXEp0JVySJBmg02JvBb22SbNubVx6Nl7eUp8xB/E0Fzi+qbK9csrfzJU8VNtQP3fX6RQYrrq3Tc9SZEpcQksoCoUeQEiYoMvakJbW4QNzCQOVUGm6QWVZ2zWoApQrVIHAAMUHsLTiShMqSAPintQebdStlWcvFEpADhkk80HY0JLIkgEngGJ9KDfdFFK7N4gArC957wKCwuhqcJBA1+vtQUaVB3q5aYVtajUAOZNBpgwktjkE7jbmg5s/qwdKh2oEWoEbAzHPpQCdC0kxIniTxQctUpAHKeYoBPp1hYB8wgkk8UFW6kgknfV7biKCvfOpJ1azvOx70CuOL07E/L1oHFoBEnck8Af69qCE8lBIgTO8ehFAdCElGoiEgyN6C06eU3+k7ITuq4RydxzQegXYGm008i4QT780FjMElJGmPXcUDFCZSATuSSBQBVpUlckKVEAkUFbl71GPat9TKnUOuaCqQAn3JP8AqfSgEcxZlRkqCe5LZgTESffagErLWXhtl1RBjdOkmPy+VBItFWd6V+GlKkoOlQU3EHmNxQMzjDHgoXcLbbhWkKUkKjnbf1/lQZrJ5BrF2JuXW1rQFpQdESCox3IAHHegip6itVXDjPhPJW2+2wZCY1L4Ox4EGfxoEteprK4UylpLxcdWttKSBOpIlXfsN6At31FaWjTLlwhxIdQhxOw/aSogc8+U/WKDshcYvHvsJurdhKrkSiW07nUkfj5h+dBXnNYhCmkmwjxSYBt0CAADPPcKB2/KgIzlsU8ttKbEDxG0uD9U3slStI4Pr84oLCwVaXZW9Z2rbamVAAqaSncpB2j/AHqB9tnbV9q4UkOHwA6XITz4ZAV/09aCbeXzVlaB9xLikKKUpS2nUolXG1BBe6isGg6orWUtlCCUoJkrTqSPrx89qC0spLS9Wx8Zw88eY0FXgj/Yno73T/Pf9YqgsNSirtPoKBZJgEE99hQORImI3NB0+beBAj50HJISZJ423oEJClSDz7UA7l4oW0kHdao3PFA1wOAgJcSQTvqTzQMfD7jTjaQkEp0pUk7igYt99ttsOMLdcSmTEQSPr33oGfpVsNBTzb7KlEjStJMAfKgjYvqCwyd6u1tlOB5IKiFtKRqHqJoK/qy6tCbOzuwlSX1kQe23PyoA4nL2DFiGVXNs2lklCVjypVHtQVfWeRtrrDm0tHm3nlOpJCd4AP8A6UHX76E9O2Fh4qw2NHjaEkygHcCgMjrLGWyi0yxcjSo8NRG/zoA5Pqu3VbhbbDoB7LA3FBQs51htl563al53bzbAUFbZ3LbTi1XiFqKhGx2VNAW0yxxZWoNJLClzpB3TP+hQWKPtBaGhKMY65vCdLg3/ACoLG164ZfQS5YPIV8MFYMUElvqNNwpfhNRr2gniKAeT6jdZQr+zqWnj4+I5oMPedSpKlpLJKlGfioEGbZLLttcKftG7sD9e0dSkQfQUD7y3cu703WLuE3LDehKikc7RKhQWQORwDFtdXt6hWPdd0KQ4jXpO54/Higt3kYVhNje427acWh1T4kERI3Gw2oKzOYm9vru8yXhEpeaUIQdWnaOfp6UGQbtbllaHUoWVIUDBTMUGt6Na8PqFrxAghJO4Ow2oPSvFEshRClOq47ADeg85zjv9vu1ApMuq5PO9AG1cWsIhOxPY0HpfSgLWKWUndau/bigtgNS/NsQNt+9BV4s//wAVXaiEiGEao22mg0ZOqdJ2jkUA1DzKQBud+Of9bUEZS9KoncHg+lA0KIMOHtyO1A4pOneD22oEVJJEDcUEF4jxdB/dmfeghOtq8QTyDOw2FAhYSUHUJEyY7UHLbIQQADB796CK8ChwJUkjbgHbagG2FIUZgdt+9BYYZKDl7BTUhQum/qIVQeh3qocskkAlVwlIJHHlUaCxidGk8j8aBFgalQeOPegbpAJBM7TtQV1+3j7p1tq/NqtxlQcS24sSk9jzQR7WzxSQpDQtg2tQUQFDcgyDz2MGgM7i8WsiEMyTqlKx+7Hr6AUAk2uNtLxd6laUOLkqcU9tvHvHYUEq4KA2nUZSd5HBoMm9btXbCW3wVBK0rEGPMkyP5UAX8LYXK/EfY1rKvEkqOypJn8zHpQI3g8chZWm3AWVeJIUfi8u/+UD/ANaAKunsYuzNo604too0ed1SiAJgST2kxQNu3cNeutl+5tluMDSmXhtuk+vqkUAra3wVqlQauWAVpWiVXAMBUAjc+wj0oOaYwjVq62LtpSS2lsqVdSsJSZEKmRBoD4xrGNXaF2d2PFcQE+GLrUlUJAnTPxQBvQDR01ZW0G2NygEK1hDxHiagNWr1mB9aCb+i2X8cmyL92ppKwtLhdJcCgZHmPpQI3gbJpSyyHGypTa1QvkoMp5/D5UE3FnUy7IIIuHODP7ZoK7BqH3JewCTcPf8A3F0FgkgKEjYbg+lA4GSDIJidhQdJ0ie5oHBUqgCO+1BylSmACe8+lBwBBKontBoKDra5Vb4RT7SykpWlIWnkTQYW76ry+PfLaH0vhPGtAVNAdj7R71oBV1Y26jEgpJQT/Ggs7T7R7BZQbuyuGweFIIWP5UFqx1lg31hKbtTMbnxEETQWqcnZPNeJb3Ns8FHYoWN6BXmmLhKC4yhYJgFQB00ETIWONYstK7ZvwSuSNP7R70GTu7HEP3Hg2oU2+4dMpPO89/pQTrXp66ZQCL4Gf2VDj5UEZ3py4cyDjiXEKJABJPBoC3vTSbi7QXChKUpKSETvO1BTZDp63xdkw2LlaiXolSfWaCrUlFy+i3aVPcK70E/O9PpNvbvKJKVp0L0p3nseaCldsG8EwrS6HnVQoAoiKCoS64XFKSDJVM0Gg6fLhuDAKQgyqRxQXV9ik3ba3UuOAmVKggJ3oK/F9H2F4+y2pTgBVuUL5n0oBnoZxzKZG0x7wCbZSFIU+D5gRxPEjeglpxeS6fxCClSTcKc/W6R5Sn9kD13mgssfmkP2YYy2OStKviBTt+B+tBOTiemchaoLANo4lQ/uiUD6jcetBHV0hf24ecxmcWptav7lfGk/4hPtG1BEbZzOF0G4xBvW1E6ltK1qT6e9AVgfec3buWlgLVwJWpRK4Kp4Ch+NBLL1+0Lc3jGkpJKigykHt/KggWtxbKyaPvFo2XUhSlBwTrPsk0DMhats3pNulbTOgKKFnudzHGwoNr0upKcVABQCrbVyZoLVskLUU7iYgd6Clx7pX1dlADKUtNbg8UGhad0okQUkbe9Aq1aTqA1R3PJoAkySFGZ4IoOSkApJJB+VAaJSAPN6x2oGBIKiCCFd57igA6gEkzEGBPagCUpMEkj0igQttqG+4HtzQMgIkq2M7QKCPdtAkkJmDMGKCG62uUEJUdo81BOwyoy+PlMzdJO42PlXQbjJJIVaK3IQ54kIG+yVUEZvqVlNqXTbPJAt1XJBWnUlInbY8yPp3oLW0vUXtlb3LSCEutB0JPO4mNqCst+ord1xgfd3UF4r+MpGkIiZ353459aCKOpce64xFota3rcPplKdUEwEmTMzt6e9BLtbzHXF6mzFqhL5TrLSmkymADv/AOIUBr97G4+4ZZctGzrbWsQ2jZKeZmPXtQQxeYW7ureydx6A/cJJDblsnaNUhUT+6fb3oFzeTtcTbsm4Dq9Z0oQ22Vnbkx2AkfjQZXKZZnE2iLh5C1NF1DSiD8Or9r5Dk+1A226mx6rdDjhdZXpSpTakEqSFEhJMfI/Kga91LjEaip86ElwKIbVsEfF2/wDWgm4/IWt8CbRankIATqKCAZE7EjftQQ7PL2t7fi3tWlOBSFOKdKPKNJAgz6yI9qAL/UeLtrx9i5QpoM6kqcUz5SQAeQP8W3rQTnb60Vjra8YaS8xcLbSjSkDZagAd/n86CO91BiLJbybtabdbCjJU3sBq0yCB3/GKCyyGSZsUMOPBfguKKSpO+kaSqT+HaghP9QYy3aU6m5QUgDUriCY0gzG5kRQEX1HjUMLcVcwoD+6KSFiSQPKfkaCfjmihC5MEvOTt6qNBVYIkYwGQR4z3Pf8AWKoLEbiCTz3PFByQSNjyKApMKAJkc7igQnzdjQduU7Ab+hoHKWQnSACqKCh6sY+8dN3yFFIKG/EB9CkzQeOlS1DzEJVO8mggXK3C0RtqO/FAexYuXGhp16dyQByaC4tbC3bbWq5W6HlDy6D8PzmgkWlu5BRbrKUAbFQiaCzYVeWtuYuVk6Y8quJoETkMqi3WxcXq1BIB88KkUEdDykXKHwlKloMyRExQXyeq22tCn7ValJG2hWw+h+tBIX1djA2FOLcZUocKQdvwoJdplse+pPg5G2c1Jn4tJP40DM9pctSAEqg6tiDIoM7j8EE3DSy4EaiSARuO9BaZS1ULdaBerUhaZhZkTPINBj81iLgNLuSpDk+bngUFOpkNW4WkJBInmgtcSlbjEp1IXIiDsaCFncPnFujw2Lnwk7iFRP50EZzDZ/HWRectrtrUtDiVtkmOd/L9KC26S6g6lu8yMa1feMgFSlG6Rq0hPqdj6Dmg2xy99avLTf40PyndduudgY+FVAyyzOMgN3GmxcI+F1sokfOI/OgvmLCwvrdf3YMqRzraUP5UES4xBCkqYuVNlICU6RxAigjLcy9hspabhB4k8/zoGudRPNuJN3j1gRAcA3H40Bms5Y3DwGvQqY0qH9KC6ZNvdp2DDyYkbAkUDVJxS3Ta3C7c3Cd/DUoSCfnQR38MrxS4m5UpAA8NtQ2QB2BoDrfv7e1Qo233p3xOGSAQn13oKnBXGrNXtzcWt3aqvExpcb4KduR67UFwErRiS00uLgIOmDv7RQFun1tWbRJJc1J1e4J3oCl3S6GQdUAnft7UBEqkhUkbRsKA6Fkp0lRMmNqByUgq33ERuKATshRJOx2IFAFKSEjRsONxxQMPJBjc96Bhb85Jg/PtQIG4WVHeeZoAuNbq1FRJ3+VA+xBTmcSUQEfegCDz8C6Da3wSp1hKwlRW7pPpGlVBHTg7BB0fdlEeGQSXFFKgTMKBO/c7igsLW0ZtbZu2tkJbZQNKUCYA5igrzgMc0EJYaWlA1cPL31AAg77jgAHYRtQDbwuPS0tlLAQgiNLaymOeI45NBNtMda27zbjSFhxAISVEnYgD+CRQFubJq4cC3GlKV4a2v+FUT/AUEdvDWjFy1cNIdNwyNlqWST8XP/iNBG6mwyM1a2wS88w80okLQuDpI3B+oH4UGYdbaeb/AF7SHQg60pIncfOgp7fJ2Lqgp/HKtXFuFmVoSpUgwfhJ2md6Aal9PPXTV2C0pQRq8oKUlCuSobBXy59qCwuMtZYvHN3IRFo6YSW094222MmIj1oItjnMI0pK237dlTikgeWCfEPl49Yn+NAXIX2Mtcg83d2qNRCC86UAg6gqJ7/sGgR/NYIWqmH3bZVmkDUCnyNpiUyI+XbagGctgFtL/VW7jepaNBbEqUkSQPoJ+UUFmnJYvIutW5W2pw+ZpCv2vLJKfkCR+NBHKsFcm7aurVhGh3wlqdSAlSk7iN+2xoD29pg8g+vwRbvOKSFq8NwzB3nY+/8AKgssc6pbbhJk+O4PlCzQVWClOOE/967/APcVQWIHJ5HtQIDsN4MdxQESoKTqAIA9TQIoqmEiJEb0HJXA3AJA3oOUY2PHr6UGR69yZt7NdmmT97Z0pIHBCt6Dy8kFZbUBpBiSOaAzjKnSgNtEA/umgmWql2y0oWFJ0gcetBMur9uSnQtZiRPM0EVvIEnS2nQBQWNtdBToBPm4k0Em7DYWVK1bCdNBTuZBpTpS6khKdgAePegIzetidSdSZ2ntQQb9TakKOkye570DEWbSmAQJdjaf9fKgVPiMNApKwBsNJ5mgMz1LlrVxKWrkKA8oDiQragI91Ne3iVC6Swof4Rpj/W9Bz+aZTYracZWhSxJWlcg/SgzKcqoqCVoSUDfbv/rag03T+ZtW0A3CVMmYISNU/wCt6AfVnVd0vIsjFvqS2E6QCjfVP/pQenWBcTZWyXylKw2CUD5bige6hanQQhCU/tEABSvrQL93bdUVlBPl079hQAXYMvA6kpI1hZSRzBmPzNBDd6etVNOKQ0GHIOlxk6D+UUAV2uVtUEWl+6EhIMOAOavx+lBGVnMvaFYurS2uG0BfmQS2RpiTvPrQR0dW49wD77Z3LBMidOoH8KCbav4PIJLjFzbBa90ydJn6xQW1rg2Qz+pdcaUB8STINBQdUdHXt2+m6xzqHllJDgcOmSBHP1FBQ3Nr1biUhTaLtoDb9SdaRx6f0oA2nXfUNoP7QEOaBJ8dqDx67UF459oYctixdWqmi+1PiMu7okxIBoAs9TJ8ZlxrMELSjTF6wqD6HyyKC5s8+t5SnQbG+CyAW2bhKCNt4CoNBa22ScL7rjuNukqHwuBEpHoCaC2xdw5eMF51tLXnUAgGdgYk0FglMqmN/wCFA5M6yoAADmTxQCcSSsGZEfjQDDZSnYGVHsaBsAkkgDfvQcRKZUdh+VAqQAORJESKAZb84nmZJHegW1bP6XxQif7XJAPH6tyg1uRUnxLZSQpWlzVCRuYSragis9QWDjJcfc+7NpT4v9oHh+UmJ396Cbe3zNkwhx1Li0LUG0IbGorJ4gD60ALjL2aAS4uDBC9I+GAD5vQ7iggl3C3BFy6bMOOQSXikKG0wZ9iKArdviHXQgNWxdglKREqA7p9RvQDfdwtpcFi4Q20Q2XFzqASn1J/Ggku2uIWA3FvuTw9BMDfg/jQSbW8trlvVbupWkRslW4Hb6bUGMQTEHkcUFVkLrE499q2um0By4Dr6AEzOnzrP+uTQAYyHTqW/CSbZsJSV+G4jSUjVpO3zgUBb3J4qyuEWDrWkW6UupCkfq077EH2maBl2308uyuLxLVo8iz0eJ4YCtGmNO3tsaAruTwNw2b51dstbTajKgCooCinb2JmKBXRhLe4Ql2zQlTqEuLc8PyoBnTqPaYIFAga6ccbQ20bAQsJSkKAOpQ278kUEhBwVnkQhTSGbi1EJdWghIlI2Cj3gj3oJPgYcuLU4u2KnXNR1OyCs7TBPO5oJ1nirOxcbVashspSWwATABief90UBMSmGXBB3fdO/bzmgrsIAMeDJI1uHn/GqgmkTBPf2oGEkExuSYEmgKUkEb7R+NA4JkAEn+tBwEkwN+KBjhG8DaYoMt1bjmMmtQdUtK7a2LzZSeCCeaDzG0aS6vSnYg8g8zQXC3W7dIaB86R9B3oI7bzVwkk+XT5ioncxQV19dBT+poABIgqJ5NAWyUVoICgpXMD2oJiElBBJGrmQeaAr10pwJbCyVzJJPAoKm5YDV0UqBAnkjgGgQOhsEOdxHmHIoI7t62AUwVKB7DY0AzkVBoqghJG09v9b0FfcZO4cVssmBHPNBDTcLKySokgT5jQTsZcFxRLhBUBv7UEjIpGgEiE9wDyKCiTBcGkgjmgP99Q2TBExt7UDG764adS406UKHmSQeCDQajHdf5q3VF0pq6RH/ALZHm/8AEIoL20+0m0de/tVk+y7MFTagtPMnYx7ig0WM6rxd0pJTkmwlXl0vpLRmD9OYoLpjIoccR4GlxlafjQsK7wJiglC9acSpomFKTHmEcx/UUCtutqaSkLQsRspJ49P5UArptDpUFISRBBnuDzQUt3grR9aC2kIE7BJoOb6Xtg6g/E2EkaFjaYigsLbp5lhIUwX2Vn9pl1SY+k0EpNtk7dqGsi6veAH0JXH5TQOVe5ZkAusWz+2/hrKFb/OaBpyjRYV9+xb4kwdSEug/h/SghLx/St+gJetrdJSIGtCm1J+u1BEX0FgrxjTY3K0DVIKHA5G3G9BXs/Zl4TpWi/S6jUCEONxwfagt8jhLpjwm8JaNWbYVK3UubxHp7maCxbcydqh50occ8NvhaAfFI+XFBOwOcRl2lp8FbFw2fOhQ4n0mguNwlQkj5CgGRBA5A596BdJABAJkdxxQDcAIGkAkH8aBiwdHp8u9AipjURtzQMDm4kGI7d6B1oArMYrdYJuFRA5hpfNBqrqQ/aAHdTh4/wB1VBBydniWLEIvmEKYVDSUEKUTEmIEnbc0Ei5GNyFohm4Uyq3BlIC9BSU+m4Iig5NjjAJAb1OkK2c5J47+goI7lni2LptLhQHXXNCUlZKio+aI7fDNAyzbwyr8OWywHrYlCiVKATyCDO3Y/UUDr20xmQukXDiytekoT5yEwdvx3I9aBHMbiAFqdMB5wgqK585/rQLZ2lhYLcurQz46UJKtUggDbgehoMykyolO47+1AC6x1rcvi5et0OPaAgKVyEzP8zQBRicaq7Nw3bNm6kErBOowQfX1AoCOYiyurov3NsFv9lEmRH8/fmgDa2mFYTcsNGzS1cmXkhweYjbfegCrHYN19bqlWh1JIVD8agTJiDtvJoJl1b4e/LLj/wB2WW0hCZdAIA44Pbt+VA1GJwVwprQ2w6EqDqAHNwpO+2/zoJj+DsLl9x95lanXgkKIcUIiBtvsdhuN6ByMFjEuOlFqhKnDqWRInn/mNBY2rIt7dplK1rSgQC4rWo/MnmgZi1QxMky6vbt8ZoKvEmbBEbDW4dh/jVQTQdxI9tjQOTyOBtI2oHgzG89/nQVOX6gtMQUt3C1LeUJ0IG9A/GZdnJoW5ahYQkwSoRNBLNw2X/DK0pXHwzuaCovmW1Ze5UVEzj1xJ2iTQeRWhUJIBJBG88CgbdF4PAwsyY1xANBHU6VNuIbJE+vaaCEHFJUUSDvO/egs8Srw9RKyGxIO8TQSXrxCn06TAHrQTccpp1RSYK5kH1FBNyNqX25hKfLAPrH+hQUCrQuhAcUImRp5FBHubJFuvWlR0p9fegrLtxEBEyEiJHagiJIElRjeRvzQMJCiopTA7zQNYcLLgOxg96CU9fLfCkrKBtGw5oIESCANjQRXEhLnqaAghQ9IE7igc2sCIO9B2/iEjifWgkpeWqBAAmY9aAlvcOsrJbWpCjMFJ0kGguMd1LlrZwFF8+YEfrDqB/H6UF211felDYdZt1COAjTx8j7Cgn4/rhhltbbto8hRAEod1RAjYH6UFrZdaYt1ADlw4wsn/wBqzImf8M0GitM3jLsAMXtqvc+XxQk/gY9qC8t3dbQVJmORuDQTEOIWkJSQDHcc0AHC2ZSClQ9jQBcDCVBt3UkuGRtsaAqGGUgawSFe0zNAG+xFrcIWUtNpWvfUEbgzQQU4lxkH7s6+1G8JeUAPoZoI+WGYsrB1+zuHXnWwFeF4SVqI7wKCsR1ffsZO2tnLJ1xhxWlTjluttSZ/Lmg3KXfKIQVJA3IAP50FO4+t2yNq3cOIfbcQCtKk6iQZjnuKC8SolI5JO+/agcowSoyDEAetAN4wCSSN+woArMIEHb5cUA1GDuVER+NA0gkEBWw255oHWSgc1i5IEPKO4/8AdLoNRdKIvbEExDqpBH+A0C39kzftoS6p1BaX4iFtrKCDxyOxBIoIi8JYuqQtSFggK+E7K1CDIoI19j8PbqtlXyy2UvJcYC1+ULSNv+v8qCau0sWLtd644EulfiKUpwaQYKe/zP1oGO4ezLRUnYlxThcSQCVFU77fxoI7uKxNnZFVy5oYDmoa1SAtR/jPFAO1xNvaYhFo/kUkuveM3cNhKFTE+87Tue1BIs8ccW6+/bguOPECNgAkTHzO8TQZLKX7WMsxcOpUpvxEIVp/Z1KCZ+k0FYvqu1TbXVwph1Nsw2hanCR5tZAAA570E+xyLN1fvoatyh5tJhZI/WICimRHbUDzFBHPVFvrbAt3yXXAhuSnzgyNXPHlVsd6CVlbu3x7zQNgp1K0rc8RCUQNIJMyQeAaAFjl7DIX7dtb2WsLBKllCQkAbTvQTLl+wtnLppdokpty0CpLad9fH4UETF5rGX1xbCxslKW4CQsNoSUes7yInegmX2fssZeO21yXFPIZD5CI+FStPr6/lQOT1Jj/AAkuEvJalGlam4SrUoJ2+RIn0oDs9RWDqigKeTDhbKvCMJIMST6cb0E3GJ0sJKvKfEX358xoKvBCcQ0RJBK/r51UFgBIJPPFB0QDA3I5oEVISTPb14oPLOog6L51+6cT44VoAUPiSJ3H5UGq6cvrC0w9t4lylDrpJM/tKP8AoUEhFozkspcO/eFQ22GyhJ4V33oHotk2mUQwkFX/AGetO5mfN3oPKWrxKX7hZSsp8NSUhPagsM7mba+6btmUW62nGn9UndJ8u9BlVK8RKlk6E7DYc/hQMcSsHUpUjsT2oHOOFtoNhZMiTA3FBHW8rWCDBj13oLvG3SEp1OLAIMiDxQPuc3CVtkqAjYzxQVKMm42olJCvNsT2FB11fKuCNUEHkTQQlKQoqOoEkRJPNABXqmeO/egQrIQQdPeNqAWqODG/pQOSopgKAifWgM8pKUjmQKCvKj4hUNqDgpQEnefag5KhO54FA5JBVwJ9jxQFTMiO+xPpQFQpSXQsCYPcd6B/ilKlhQGqexoLFL3ip2BMcQBvFBBU4VOngxv6TQc24onVyr1nigMHiSFEbzG43EUFkzkHWkIU0861B5Qsj+FBfW3VeVaQlTWQeSAI0rIV/GaCU31vlAFSu2dncyxBH4RQWjHWNwS2u4tLdRTx4bikz/GgtbPrdsgC4tbhPycSY/GKCenrTGFISVvJWkxCmufwmgnNdR4xwJKbtsJMHzApifnQS7fJ27qZZfYW5EpQHBvQTS+oBCkoK1K8p0nj50FDlM61b5Rdhc/qW3bcrSXPhVOxHzHNAvTuMsVMs5Ji4Nyp0+KHCBvtH8KDS26SESAfr2oFUowAQTOw9KAShIIgkx3oBqJUmFCIHYc0DQkwCPTuaDm4PIJ39eaDrATm8XIjzuTPb9Wqg0l0ub6wTpiSs8eiaCYoktjTBPae1BxEojn5GgrctjGMmm2buS4lDLniJ0EbmI7g/wCu9AH/AGesnEpIL0wpIkgyFGfT5fhQObwVs3il2CC8lhbqniokTqJk9ojntQO/Rlg1afdlAFsuFwyYKlEzJIigA50/aOBxZcWrxB3CTp8unaR6UEjG2Ax2vw7lx1lZKvCWEwFEySDH0igzTjLbyQHUBYSoKGobAjcH6UAWMXYJPjNWzSnFJjxFCSR9aAlhZWtut5NtboZWs+YoRGqg63srJh5S2rZpCyouKUlMbkQT+FAFzIYm5hLt3ZrIBCdS07SIPfuJoGzg1XaLnxLIvoMpcDidQJ+tA504S4fNw6uwcuNISVlxMlI957UHW+PxLrzT9oi1WthWoLbV8K/eD396CTf4bH3brjl1ZocW5AU5JCoEDkdthQNTgcYloluzbIWBsSVJ2EbAmP8AU0BWunsYS0tu3KSF6yErUAo/4hO/AmeYFBOxitdo0pUElSjP/EaCn6fJGEtN58pO3eSaCygmd5+Xeg5UiNwO+44oAueZQ2gDse9Blevca3cY5V7pT49uACRsCmeP4UGc6UxdtdB25fuXAhiUwOBA1UFrZdQMWLalrStu4dcnw1DmTyaDUXWlWbtlgHzWjoAI43SaDxV9lTanS2oGCdQHAFAZzSrDJTGwcjn1FBW6oEAlKFERt7UDiCws6SCU7iTOqaCvBIcIUhIAMSO1AN2EQobkieaBhUogASTHIoFcWXR5zJA2k0A0rhMEEE+poOCTBIAjk0DSQDqKdQ9qB7rkAABQ3jftQCWklJkHj0oB64GlQ+U0DXSCkadxNARtZUglYBPb2oI6p3kRQdBk9oEb0CgxA00HAQZM0BQQYk8gmgK27A08CZ3FBL1IWVRwTsSOKCUwGgh9BSRxsTwRQDfZaAK3FKk/jv3oIwZT4hQl2ADydxQFRaugrSlSIH7R2B96CTa+K2sJDaVmJOmN/egM024pRSGJGoeWOKBjymkvLUGFpg6kiTt/ragl2L7DpLZWtCokeY80BPEUVFYuSpzjcc70COOuIISFoSFgmeAaCc2QsAl9EpOnckSrn5UHJyiWVKWN1E6R5x5PkI+dBPZzKWiXFvlKZnyg0EO4u2r68YWp9UtE6QREz86Dd9A2upwvCVMNiEkbJn/U0G9koBkygDsKAAcQvZKwob/CeKBpWSklI77UAVAKSQSeeQeKDkrSSBM7d+9AiiJkHj2oHWBnOY0gzKnTsfRug0F0+hl1l1w6W2krUoxMbD60DRmbM2zry3FIbZ0hxa21J0av3pG3v6d6BiuoMadYFzBSDKNCpEBJPb/EmgcjMWKnWG239S3zCYQqCSO5jaduY9KAF+vEsXgRdjTdOpLnK9wO5jigjm6whAUtYSQ2HYPiBWkzBjneDFAdgYi6f8O3WXHC3q0Bbg/iR+HNAJxWJYuzaOOOJfTCtJW7Ak7eaY5jvQGcy9k4rwk3KVOABeyVGAfePegzmRvmsfaLfdOwOlIkDUTwkT3NAG9ztpaX1lbuJcX96QtaHGxKQEiTP8ooG3GcsLVRU66UoB0mEHYyBv8AUigdfZm0trp1h1S0PIJMBsq1ABJMR/vCg61zWOvH0W7a9bxWluPCOyinUBx6fhQD/T+NbunWlhbS2XfBOpvZSpIhPzigOc9i9LTmvS2tPiJWWjumJ7Cge/mbC0ukMBp0F1oXHiJR5AkmNz/L8KAljnbTIKfTYlb/AISSpQS2d4jYTG+4oDN52wU22tTymQpOqHEFOkHifSaCV+krQXSrZdy0l5CkpKCYJKhIHvNAuNMWLSSAAASAe+9BVYJQOGs//wBIce9BYSACSQAO1ANSoMk8+3FA0zrEjYDt2oK/qCf0LeqSgLhvXHrG9Bh7dKlY1uxxIKbu+Pj3LxGyE77bcUBMu0za3VpdXrq1XbJjQTssJSI2HqaDTXWYszlsc6h5JQ5buKHO06YEevNB5besX7a1lxlxPikqlaeQTQTbXDZB+yeR9wuShDgUSlvcbHt3oKq4tVsNlN6w4zcIGtIUkp1AUEFLfilfhypaU6lAnkd4ighr1qUYWAJ7mgjLJJKAuE+/egFqWlRgkHmgeneVbHeg7dRAIO3oKDiIBKhBmNqBFEbhO4Hc0CatIPmBn1oGrJT5SYkbUASCAZ3I9qAZMqkCNqAjJGk7ieflQIsQB2oObO8ET2+VB0FMxtvFAxRJAB9e45oCBRBB39OaApOpO8DaeKArKtBA39d/agO04YUolRPz5mgR47oUokNkdj3oBtOLBEKAV896CWl9xJJSdR7A0EoOlRCiUFI+cER/60BkvFFwhTQIJTO55oBuXC1pVOr4t0mg61cWpaCE61AcRMkUBby4bKQlxKUrPqiO9AB26SErIYaM7RB2FBYWtzartVakJQmNZAWd4/nQR1NpUfMEKA8xClwofSgGWlurBUARASkA/wCveg2XSeDTk70tKWpaJ1u7bgD39+1B61btN27KGmEJQ2nypAEQKAhWltJWolMJnVNBFQ0htEICUhZJJT6mgjm7QoEWy0LSO4O00BGnFBJ1JnadqBqQBsBMiNu1AqhLW4iPxoExhjqLGJiID5n18goNJfNouH2rZ5JLbiVhUGPT0oBnEWyUuJPjFLiPCcQpxRSsHuRO5535oEXhbBx195xgKcd+MknzbzuKBysTa/ekXADrbqFawEOqQk+gKRyB2FAt7jLS/fbdudZW2NIhzbf2oAu4CxcS1qS6p1pBQ24HPMARH/pQEtcNb292LppT3jpRolS9QM/EfUk/P3igfd49m7UtVwhSxpShSSfKQFauPnQU1x02wgBVpdPW5hKSkKGmANu0/nFBCSApACglQ53FAt6UtWi3m2PFW02pSEpSJO3A/KgpU59tLTYex9wpx1pTqW0o8ygInYieSP50Eh/KLBcV+jHyhOoEkSox3Hz2oCZO7VY3Vl93xi31PNrJcSILZCZAO3fYUBrTItXV6bZywWh1KPE1uIBAkxB9+aCDcZMG1umcliNdvDgUGR4msJ9oHPagsxfIOLcuE2C1JaV4ZZAGopSYEfxAoIrGasUXbKXLG4t7i40jZnjWSBqUP9zf6UBbh7HP34xj9qStSoSC35VQkK/Dt9KC0XZWini+phovKKZWRuY4oG45U2DGoySncig8Zvur8mmyt7S1X93t2x4ets+ZUfwoK5vqHMtkui/fkjTuuZA9jQTmerMy2Cr7y94iiAQsCBHtQWV51lk1JHhLQhSD5iludfagY11U8pp1laErQ8kpImYPr/0oA4LM/dH7guOJbbcAkqHIB7fnQXisnhW8vdXbl02+04lIaASVaY5mgg5rqOycydk/j1LJYQsKIRsCeInmgi5fq5m6QAywsAJKFKWOZ229KC+wnWtqpaWrpJtW/D/vlGQY/wBGgkZu4GTftn8TdWNylsEPNqdTKgeIn60Gf6mRdOtMn/Z9TLrawpb7KPKpPf4frQefXqfAW6EwEhRCd+KCKnzCVzJoBupKiSn4QeR2oFgpUEk/9aArbgCydMHnYcxQJclP7RCtu3egFqkAhMGPNNAg0qUCveKBzrgccJ2+VAEjSkAyTz7GgFIEzt7DvQc3GuCCB6UCuEEESeN/agY2YAkwY/Gg5RkggyI9KAqUgqBBG/rQMUoEQREbT6UCtqhXc7xQSi5qOxoHf3ZJSSTHEc0BtQ8Iagk6RQRSYcCh5D6ntQSG1zGoGSNo70BmndJJCZg95iYoJ7LqfDSQQTp5EbTQCuELbQpajACd4V60EnG36LUtqb0AgGZG5n3oAZG7L7y1J2SSSAVUERL5LRJBBIHHaKAqHdUoVoSYnfueaCYHU+GPDQlKo7iZoObIVcNgBCgSCokRQe39E2LVphg40jS6+NfmG4H7M0GgUNh2IHEUEa7VNu6lQJChp2HrQNcToYUkDcJ2HptQUmFxgbxduUhQklek+5oLUq0N6wJKU89iaCrxF8bi1DrpUlwmCCeaCa25rQRrgT6cUB8ToHUVipG4Af1T/uooNHduJZumnVkhpDayo8/u0CDJ2R0TcI+EFQUdJSCJ83p35oH3N/a2qmC66Ci4MNkAkK2nt7UCMZGzecaQLlsrc/u9/i2nagjqbxDt86yUWirsrhwKA1FUA/wIoBKXhG332dFu2piA4S3CUSmeeOKB9s1iH1eFbptnVJBMIM94P5iD6UEcqwpuFoCWPHaIKkQZQQY3+tAt7lrEJCRcNLJO5Q4I49fXcbUFY2d+QNqA6QCR3FAqhKgqYjvQUNk71Alspure2Ky4uXCdgjV5dgedP/WgH+kOpSyCjFWpUpIPme07+bbn/d/E0Ey4dygRZFhhsqUnVdJUv4TpgBPr5j+AoGuuZRvHsqt2ULuy7DggQEyYPI/wzvsCT7UDVX+ZbMfo1t3UYlLmnTzv+MfjQTsXf3dx4ychZKtloWQkhepDg7EHn8qCbrGs8kg+lA9TsEQQAD3oIePcT9yYgkbc+m9B89rbWp50oSFDxFkkc80A2ytMo8wKVbSdt6CU7eupcGp5anVmVGODsf5CgM2tx1SwsghxW8HknegRklJWCSgrBBIHY0D3HS+403tDQhIAA9/60Ea8KClHhmQoSqTwaBGm0qCQkp25J2G9A1slalgAhAM80B22zcr8NslGhIgE/EKAtpjHRrWlKXVg6dM+lBOYvMnYKAt8hcWrc7JWomPpxQUN6S6t9wAlUwqO5O/9aCG4AQVEcbSKBFtwyXGyr4tJTHHvQMehQTE8QY7mgDp0LEHccR2oBvBSgZIMmDFArCtOrcFXueKBVn9YNx67d6BAQJJBJmgQjUCQQZHc8UAQkkGI9ZA4oOUYJ35MbmgcU6gTM7wI70AXW1IUDBJA5ig4E7fyoCKJgmOOwoBk6gD70BWgDIJiB2oFTCSNJJJE0BA4QSNwSOT3oDNqPhkAmSPoaAR31pmTPftFA9B85lUiO3egkAqSAoGQCOO1A9pQS4UrVEGSJ2FA914KakyVAQJoBoMDzBJURImgCpUqB0gzuaAyCkpCCAUk8xzQOKR4iAPMo7cUE22tnFXCEgE/IetBsenunWXMjbBwa9XnV/hg96D1y3cbSohKEhIgbCIigqM31La4y9FvdJeBUkL1JEiOKCC51diVIhN1CpEApImKDOdQdRqyN0wbJ11lKWzO+0nkbc0EhvrNVlYsti1L6kpCFyqPkeKBB1eX4SlohCk+YKPw0CYHKJbUEgILSz5te2mg0yXg4Au3IjYz8QM+9A/pq8W/1ShlwIC7dLoOg/vJQR9dxQbG6Shy5Q26hK21tLCkkcjy0HKx1k7ahhbKA0QEqSNpA7fL1oKvP32NtMhZHIP6SlK1pbU3rSpAIEj0UCpMeonb0BpzeDTdMqgeKHPCQsNmApS9H8aCWm9xxz/hlkDIBWgKLfJ0AzPygUAH8hibgXbdxaqLaHFoe1N6gopA5j12igJjbvEtXZYtmkNOhpT+7egFBV5jJ23P9aCCnN4Zxty8XaqQhxKlPOqZiIn4u/Dc/KKAOUt8Pc45VwmxQ8ldx5iUKTqVpIKvfiKDgkACI4j5UBQoCCYBImgeFGdIHb8aBriiG16SCuNp7GgrVPZVCdSmrDiTDi/+WgRt/KGZtbQekvKk/wCWg5Nxkk+U2lrJE7XCuP8Aw0BmHrsuFNxbtIREhSHtRn0jSKCUlWwJAE7UDkkhR337UHEnxBtPbagiYwxjmN5SEx+dB5Ix0wH2mHU5BhBePiKSowQD22oC3nT7aHfLe20lUlB1ben8qBg6f1On+32hIM6d5/OgKrpp4IH/AGhaAzIkEDb6UDT0ncKSC3d2i9phLkSKBl70/d2TReR4C0IQCspV8I7T+dBnH29BWVBSlETvtBoOAIgA+Y8wfWgE8lTYKUrIB3EDtQI0HAUQsEpHp60BQ9cW75LTi2grfSeCRQSWsvdtJWXSh1CiClKh/r2oCW+fuLa7DrLDXhLH6xuNUgfOgIcvbKdCjaBrYk7bGgj5PPruGHWWkNIbWeyeY96CjbTqG5SRH40AVhSZSCkA7yKBmtRCySJJnYUCoEpJI3I+I9qBqSUjVwQO3egCtRJEgHT2H86ByFmYUNh60CqALZURuVRE8UHIQC4UxuRtFA8akAbyPagaqSANzvHNBHcSoBREATzNAiDPAJMetA4aSiZCfY0CNEa5JJ+dAdTelQJ2ETQMUSSJMgCgkIUCBqUJ9BQK4CXYBiRM0CKMSBB3gwaAjaiQd9iI+dB0kFQmVRwaBxStYJUZ0jt2FA9DLhA0hRVEAEc0A0sPgwGlkg77d6Db9D4Bq9xyrt1rxbkO6AguBIbH70Hn3FBrLXovGPeK5cPoF0pRKXWlgJE9lDg8do5oKLJ9OX9i0DjVsLcUrdIeTISPSg0fRjK2EPu5PQi4PkSkupM+p2PyoNhbKZUVwtMEchXf8aCmzPTNrlLlDzj7zZCNICI/nQVn+w1mlQCbl9Se6YH+vWgJ/sXbJZCU3b429BQAc6KYI8128SB2SNqBg6Rt2XR/aHyeAABBNByenrVF0StbxSe2mBQXlm4wy00zbIUUIMbCYigh9C4u7x3VD6r1epdyp99CwZ2IbH5UHoLpnKWwKhIZWInndFBJ8QBSJIknuOdqDgEg6lJB3iSKAKLVhtooaYbbbKtWlAABUTJP470DL1029o+800HHm2ypKAJKvw3oKwZq4Y8IOYm5WXQCfCT5RKkjefYk/IEc0HDNPO+EThrkFxJI1R5Ygwr0mTQWja27tgKAR5hJSpG5HuPxoEuAUNgNoTAPBiKDz7P59zHWpWGQvU6pseaIjvwd6CoxnWLxadcdtEKSlK1ABZBMJ1CSZk/SgnXXWS2UNFqzSFu2v3gFTk6fPpjjegb/ALZqcty99xSBKklPi7mCN5igh4brpWSbQ4vHpbP3gMgB6QJ7/DzQTch1cbTI3LBsgsMteIFF2CTo1xx9KCPieuFZGzW8bBLelWkDxZ/aj0oLa6z/AN2xz1z921FlrxAkuQD7cUCYzqM3iEKNqEFxM/3kx+VBa2t6X3FjRp0Kj4pnigiW+Z+85M2hYKd/jC/T2igm49RRjWhzCT/Gg8ks3f8As5tSUgK9frQMbujcHdCUmZnnvQDvbaNaw4vUU0Fcy+v4dSyfUqoJttfOusudgiQBPI2oLSydcebQ224ptK21IWD5tQkevzNA9jp5hS3FF98lWxmP6UAXMPapuktAvDWJJCgP5UAcjhWLe38Vpx3UUydRmgi2mIadDqitQUjgxQSG8FbhwJK3CfXagsF9O2oUYWsaRPagQYtn7sSSTEHgUFXcYtvxHE+IsRvMCaA1rj12DK/AunAFp1KGhJ3+ooJDf3tWhH3pvRzvbNk/woBoVdO5H7oXmNhIX90b/hFBHuPvLg8FTlpAXpkWaBQOTjFW9k6ErtiG0nm1TvHqaCS50uHvDWbpCCraEW6QKCI70wlTUquUyEkbMJE+5oK686dbtnGf1wWVTP6sCgfd4dFvIQ4CVbGW0kdu1BDtMckOq1r1aT+7zQaB7pS2CAvxnJKEq4AiaByOlbIuoC1LKCU7DY7nfegE70lZF+4Slaw22kEJ53370Farpa0StDgcc5iKCSOlLNVuSla0kmBtxQS8b0dZOOBXiuCN4IBmgK/0faIeQfGcVqMEEeooBr6PtA2mHl8T8IoJNv0rZi4SoqKiV6dx7UFyjpfHBBWtClKA7GBQPc6Xxram1oaIXBMyf4UA1dN4/wACUtAAQqDvNAVvp/HHTrYnUmT5jzQSbfpnEls/2ZUaePEPoaCXbdPYxdq094CgpSAR+sO1ANjA4u9aJ+7Fs6imQ4aALuExbVy02bUqK1Rq8RQjagludL4hRCTa+XTqIDihJ996BrnS+ES+htVglR5B1q2+k0Em26axTKkFq1CSJnzEzQW1u0y2vShlIGkHagkkjWAUgxAB+ZoAqU5+kEjX+r8MjRHJ1RM0BPvRK9GngczzQKpWqDABKoMUDlAKbBIE0ETwka1yJBmRMTQSENISwYSB8qBLFAHUmPUOU278fi2KDQOGcq2D/wD46v8A6k0BkqJSJoHAneg7gyKALyihtSx8Wnn6UCpQklKlAFURPtQORbswpYbQFD0FA1DTTYR4bSEkTBAigV1w6wBISNoEf0oP/9k=\"/>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:02.241930Z",
     "start_time": "2018-11-22T18:39:59.472488Z"
    },
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "# Libraries we use for data analysis and visualisation\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "# libraries used for the notebook\n",
    "from IPython.display import SVG\n",
    "from graphviz import Source\n",
    "from IPython.display import display                               \n",
    "from ipywidgets import interactive\n",
    "\n",
    "# ml is in this directory, contains some utility functions to not clutter\n",
    "# this notebook too much\n",
    "from ml import plots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# some display settings\n",
    "\n",
    "\n",
    "pd.options.display.max_rows = 10\n",
    "plots.set_plot_style()\n",
    "\n",
    "colors = plots.colors\n",
    "cmap = plots.cmap\n",
    "\n",
    "%matplotlib widget"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:02.241930Z",
     "start_time": "2018-11-22T18:39:59.472488Z"
    },
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "def read_titanic():\n",
    "    '''Read the titanic survival data'''\n",
    "    passengers = pd.read_excel('./titanic.xls')\n",
    "    \n",
    "    # convert 0, 1 to False, True\n",
    "    passengers['survived'] = passengers['survived'].astype(bool)\n",
    "    \n",
    "    # covert strings \"male\", \"female\" to 0, 1\n",
    "    passengers['sex_code'] = pd.Categorical(passengers['sex'], ['male', 'female']).codes\n",
    "    \n",
    "    # drop rows with missing data in columns age or fare\n",
    "    passengers.dropna(subset=['age', 'fare'], inplace=True)\n",
    "    return passengers\n",
    "\n",
    "\n",
    "passengers = read_titanic()\n",
    "passengers.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:02.615238Z",
     "start_time": "2018-11-22T18:40:02.245847Z"
    },
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "outputs": [],
   "source": [
    "fig, ax = plt.subplots()\n",
    "\n",
    "(\n",
    "    passengers\n",
    "    .survived\n",
    "    .value_counts()\n",
    "    .sort_index()\n",
    "    .plot.pie(autopct='%.1f %%', colors=colors, ax=ax, ylabel='')\n",
    ")\n",
    "ax.set_title('Survived')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "__The task:__\n",
    "\n",
    "Given a vector $\\boldsymbol{X} = (\\p{Fare}, \\p{Class}, \\p{Age}, \\p{Sex})$ can we find a function $f_\\text{survival}(x) \\in \\{{\\p{True}, \\p{False}}\\}$ that accurately predicts the survival of the passengers in most cases?\n",
    "\n",
    "How do we know if that function $f_\\text{survival}(x)$  is any good?\n",
    "\n",
    "To get some sense of the quality of this predictor, we gather the following numbers:\n",
    "\n",
    "| | | |\n",
    "|---|---|---|\n",
    "|__True Positives__  | $\\TP$ | The number of correctly predicted events that belong to the 'positive' class |\n",
    "|__False Positives__ | $\\FP$ | The number of events falsely predicted as positive that actually belong to the 'negative' class |\n",
    "| __True Negatives__  | $\\TN$ | The number of correctly predicted events that belong to the 'negative' class |\n",
    "| __False Negatives__ | $\\FN$ |  The number of events falsely predicted as negative that actually belong to the 'positive' class |\n",
    "\n",
    "\n",
    "We can look at the fraction of correctly labeled observations in the data\n",
    "\n",
    "$$\n",
    "    \\operatorname{accuracy}(\\v{y}, \\v{\\hat{y}}) = \\frac{1}{N} \\sum_{i = 1}^N \\mathbb{1}(y_i = \\hat{y}_i)\n",
    "$$\n",
    "\n",
    "or simply put \n",
    "\n",
    "$$\n",
    "    \\operatorname{accuracy}(\\v{y}, \\v{\\hat{y}}) =  \\frac{\\TP + \\TN}{ \\TP + \\FP + \\FN + \\TN} = \\frac{\\text{correctly predicted observations}}{\\text{total number of observations}}.\n",
    "$$\n",
    "\n",
    "\n",
    "Now we try to find a function where the accuracy is higher than for just guessing the most common value"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = read_titanic()\n",
    "truth = data['survived']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "__The baseline__:\n",
    "\n",
    "Just guess the most common value:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def predict(data):\n",
    "    '''Everyone dies'''\n",
    "    return pd.Series(0, index=data.index)\n",
    "\n",
    "\n",
    "prediction = predict(data)\n",
    "\n",
    "plots.plot_bars_and_confusion(truth=truth, prediction=prediction)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "__One possible solution__:\n",
    "\n",
    "Women (and children) first?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:03.337648Z",
     "start_time": "2018-11-22T18:40:02.618357Z"
    },
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "outputs": [],
   "source": [
    "def predict(data):\n",
    "    '''Only female passengers live'''\n",
    "    return data['sex'] == 'female'\n",
    "\n",
    "prediction = predict(data)\n",
    "\n",
    "plots.plot_bars_and_confusion(truth=truth, prediction=prediction)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# How can we improve the classification? \n",
    "\n",
    "We could try yo learn more about what happened on the Titanic. Essentially using *expert* knowledge.\n",
    "\n",
    "Perhaps even by watching that 1997 movie where Leonardo Di Caprio drowns. \n",
    "\n",
    "In the movie Di Caprios character dies along with many of his third class passenger friends.\n",
    "\n",
    "<br/>\n",
    "<br/>\n",
    "<br/>\n",
    "<br/>\n",
    "<img width=\"50%\" src=\"data:image/jpeg;base64,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\"/>\n",
    "<br/>\n",
    "<br/>\n",
    "\n",
    "\n",
    "**Let's presume rich people get to go into lifeboats:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:03.997238Z",
     "start_time": "2018-11-22T18:40:03.340545Z"
    },
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [],
   "source": [
    "def predict(data):\n",
    "    '''Only first class passengers live'''\n",
    "    return data['pclass'] == 1\n",
    "\n",
    "prediction = predict(data)\n",
    "\n",
    "plots.plot_bars_and_confusion(truth=truth, prediction=prediction)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "## Linear Models\n",
    "\n",
    "Can we improve our predictor by combining more variables into one predictor?\n",
    "\n",
    "Lets presume a linear weighted combination of variables:\n",
    "\n",
    "$\\def\\RSS{\\operatorname{RSS}}$\n",
    "\n",
    "$$\n",
    "f(x)= \\hat{y} =  \\hat{\\beta}_0 + \\sum_{j=1}^p x_j \\hat{\\beta}_j\n",
    "$$\n",
    "where $f:\\mathbb{R}^{p} \\to \\mathbb{R}$.\n",
    "\n",
    "For a single sample of the Titanic data we simply evaluate:\n",
    "$$\n",
    "f(x)= \\hat{\\beta}_0 + x_{Class} \\hat{\\beta}_1 + x_{Sex} \\hat{\\beta}_2\n",
    "$$\n",
    "\n",
    "When we include a 1 as the first entry into our sample $\\v{x}$ e.g. $\\v{x} = (1, x_1, x_2, \\ldots, x_p)$ we can rewrite\n",
    "$f$ in matrix form\n",
    "\n",
    "$$\n",
    "f(x)= \\hat{y} =  \\v{x}^\\top \\v{\\beta}\n",
    "$$\n",
    "\n",
    "where $\\v{\\beta} = (\\beta_0, \\beta_1, \\beta_2, \\ldots, \\beta_p)$.\n",
    "\n",
    "\n",
    "\n",
    "How do you find those weights? Like before, we choose a loss function and try to opimize it.\n",
    "In this case we choose a loss function called the residual sum of squares (RSS).\n",
    "We calculate it over all samples $x_i$ in a matrix $\\boldsymbol{X}$.\n",
    "\n",
    "$$L(\\v{\\beta}) = \\RSS(\\v{\\beta}) = \\sum_{i=1}^N (y_i - \\v{x}_i^\\top \\v\\beta)^2 $$\n",
    "\n",
    "Here $\\v{x}_i$ is a row in $\\v{X}$, hence the transpose.\n",
    "\n",
    "We can now rewrite the loss function in matrix form:\n",
    "\n",
    "\n",
    "$$\n",
    "\\RSS(\\v\\beta) = (\\v{y} - \\m{X} \\v\\beta)^\\top (\\v{y} - \\m{X} \\v\\beta )\n",
    "$$\n",
    "\n",
    "Now we optimize the loss function just like we would any other function, by differentiating with respect to $\\v\\beta$ and setting the result equals to zero.\n",
    "\n",
    "See https://en.wikipedia.org/wiki/Matrix_calculus, for how to do calculus with matrices.\n",
    "We need the product rule:\n",
    "$$\n",
    "\\frac{\\partial({\\v{u}} \\cdot {\\v{v}})}{\\partial{\\v{x}}}= \n",
    " \\v{u}^\\top \\frac{\\partial \\v{v}}{\\partial \\v{x}} + \\v{v}^\\top \\frac{\\partial \\v{u}}{\\partial \\v{x}}\n",
    "$$\n",
    "\n",
    "For our case, this yields\n",
    "$$\n",
    " 2 \\cdot \\m{X}^\\top (\\v{y} - \\m{X} \\v\\beta ) \\stackrel{!}{=} 0\n",
    "$$\n",
    "\n",
    "Solving for $\\v\\beta$ leads to\n",
    "\n",
    "$$\n",
    "\\hat{\\v\\beta} = (\\m{X}^\\top \\m{X})^{-1} \\m{X}^\\top \\v{y}\n",
    "$$\n",
    "\n",
    "\n",
    "We just performed  __Linear Least Squares__ regression.\n",
    "\n",
    "Now we can define a function to predict passenger survival according to\n",
    "\n",
    "$$\n",
    "\\hat{y} = \\begin{cases}\n",
    "\\p{True}, & \\text{if $ f(x) \\gt 0.5$} \\\\\n",
    "\\p{False}, & \\text{if $ f(x) \\le 0.5$}\n",
    "\\end{cases}\n",
    "$$ \n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "### Linear Regression and Classification with sklearn\n",
    "\n",
    "Create an artificial 2D dataset with two classes and use the least squares method with a linear model to seperate them."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:04.852970Z",
     "start_time": "2018-11-22T18:40:04.000736Z"
    },
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "from sklearn import linear_model\n",
    "from sklearn.datasets import make_blobs\n",
    "\n",
    "# Two 2d-normal distributed classes\n",
    "X, y = make_blobs(n_samples=400, centers=2, cluster_std=1.5, random_state=0)\n",
    "\n",
    "# train the linear regressor and save the coefficents\n",
    "reg = linear_model.LinearRegression()\n",
    "reg.fit(X, y)\n",
    "\n",
    "b1, b2 = reg.coef_  # scikit-learn convention, parameters available after fitting\n",
    "b0 = reg.intercept_ # end on _\n",
    "\n",
    "b0, b1, b2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:04.852970Z",
     "start_time": "2018-11-22T18:40:04.000736Z"
    },
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "outputs": [],
   "source": [
    "# solve the function y = b_0 + b_1*X_1 + b_2 * X_2 for X2\n",
    "boundary_x1 = np.array([X[:, 0].min(), X[:, 0].max()])\n",
    "boundary_x2 = (0.5 - b0 - b1 * boundary_x1) / b2\n",
    "\n",
    "plt.figure()\n",
    "plt.axes(aspect=1)\n",
    "\n",
    "plt.scatter(X[:, 0], X[:, 1], s=25, c=y, cmap=cmap)\n",
    "\n",
    "\n",
    "# draw decision boundary and fill regions\n",
    "plt.plot(boundary_x1, boundary_x2, color='gray', linestyle='--')\n",
    "plt.fill_between(boundary_x1, boundary_x2, X[:, 1].max(), color=colors[0], alpha=0.1)\n",
    "plt.fill_between(boundary_x1, boundary_x2, X[:, 1].min(), color=colors[1], alpha=0.1)\n",
    "\n",
    "plt.grid()\n",
    "plt.xlabel('X1')\n",
    "plt.ylabel('X2')\n",
    "plt.margins(x=0, y=0)\n",
    "None"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "We just *learned* the parameters for a statistical model based on labeled data.\n",
    "\n",
    "Can a linear classification improve the classification of the Titanic dataset case?\n",
    "\n",
    "We have to evaluate our 'learned' model on an independent test set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:05.408375Z",
     "start_time": "2018-11-22T18:40:04.858869Z"
    },
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn import linear_model\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "features = [\n",
    "    'sex_code',\n",
    "    'pclass',\n",
    "    'fare',\n",
    "    'age',\n",
    "]\n",
    "\n",
    "X = passengers[features]\n",
    "y = passengers['survived']\n",
    "\n",
    "\n",
    "# split data in independent sets for training / validation\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.6, random_state=0)\n",
    "\n",
    "\n",
    "reg = linear_model.LinearRegression()\n",
    "reg.fit(X_train, y_train)\n",
    "\n",
    "y_prediction = reg.predict(X_test)\n",
    "y_prediction = np.where(y_prediction > 0.5, 1, 0)\n",
    "\n",
    "plots.plot_bars_and_confusion(truth=y_test, prediction=y_prediction)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "The model does not seem to improve the classification to a large degree. \n",
    "\n",
    "\n",
    "We will talk more about properly validating models later. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:06.364929Z",
     "start_time": "2018-11-22T18:40:05.427028Z"
    },
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "outputs": [],
   "source": [
    "fig, axs = plt.subplots(1, 2, sharey=True)\n",
    "\n",
    "\n",
    "for i, (sex, df_sex) in enumerate(passengers.groupby('sex')):\n",
    "    p = axs[i].scatter(\n",
    "        np.log10(1 + df_sex.fare),\n",
    "        df_sex.age,\n",
    "        c=df_sex.survived,\n",
    "        cmap=cmap, vmin=-0.5, vmax=1.5\n",
    "    )\n",
    "    axs[i].set_xlabel('log10(Fare / $)')\n",
    "    axs[i].set_ylabel('Age / Years')\n",
    "    axs[i].set_title(sex)\n",
    "\n",
    "    \n",
    "cbar = fig.colorbar(p, ticks=[0, 1])\n",
    "cbar.ax.set_yticklabels(['died', 'survived'])\n",
    "\n",
    "None"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Naive Bayes\n",
    "\n",
    "<div class=\"reminder-div\"> \n",
    "<h3> Bayes Theorem</h3>\n",
    "<br/>    \n",
    "\n",
    "Bayes' Theorem gives a relation between the posterior belief, the likelihood and the prior. \n",
    "\n",
    "$$\n",
    "P(A\\mid B)={\\frac {P(B\\mid A)\\,P(A)}{P(B)}}\n",
    "$$\n",
    "\n",
    "https://en.wikipedia.org/wiki/Bayes'_theorem\n",
    "</div>\n",
    "\n",
    "Different classification algorithms optimize different loss functions. In the previous section we optimized the residual sum of squares. Now we will be looking at the __0-1 loss__.\n",
    "\n",
    "\n",
    "$$ L_{0,1}(\\hat{Y}, Y) = 1 - \\delta(Y, \\hat{Y}) =    \\begin{cases}\n",
    "                                                    1 & {\\text{if}}\\quad Y \\neq \\hat{Y} \\\\\n",
    "                                                    0 & {\\text{else}} \n",
    "                                                \\end{cases}\n",
    "$$\n",
    "\n",
    "This loss is used for classification problems as it basically counts the number of misclassifications and works best on discrete values of $Y$.\n",
    "\n",
    "One possible Idea is to try and minimize the 0-1 loss directly.\n",
    "\n",
    "Presume we knew $P(Y| X)$ i.e. the probability for the value of the label $Y$ in presence of a data point.\n",
    "Then intuitively the best classifier $f(x) = \\hat{y}$ is the one that minimizes the conditional expected loss.\n",
    "\n",
    "\\begin{align}\n",
    "    E(L_{0,1}(\\hat{y}, y) | x) =& \\sum_{y \\in Y}P(y| x) L_{0,1}(\\hat{y}, y) \\\\\n",
    "            =& \\sum_{y \\in Y}  P(y| x) (1 - \\delta(\\hat{y}, y)) \\\\\n",
    "            =& \\sum_{y \\in Y}  P(y| x) - \\sum_{y \\in Y} \\delta(\\hat{y}, y) P(y| x) \\\\\n",
    "            =& 1 - \\sum_{y \\in Y} \\delta(f(x), y) P(y| x) \\\\\n",
    "            =& 1 - P(\\hat{y}| x) \\\\\n",
    "\\end{align}\n",
    "\n",
    "Now if we minimze the expected loss given a sample $x$ while varying the predictor function $f(x) = \\hat{y}$ we get   \n",
    "<p style=\"color:gray\"> See <a href=\"https://de.wikipedia.org/wiki/Arg_max\" style=\"color:gray\">https://de.wikipedia.org/wiki/Arg_max</a> </p>\n",
    "\n",
    "\\begin{align}\n",
    "f_{*} =& \\argmin_f(E(L_{0,1}(f(x), y) | x)) = \\argmin_{\\hat{y}}(E(L_{0,1}(\\hat{y}, y) | x)) \\\\\n",
    "    =& \\argmin_{\\hat{y}} \\left(  1 - P(\\hat{y}| x) \\right) \\\\\n",
    "    =& \\argmax_{\\hat{y}} \\left( P(\\hat{y}| x) \\right) \\\\\n",
    "\\end{align}\n",
    "\n",
    "This is a special case of the so called the maximum a posteriori principle (MAP https://de.wikipedia.org/wiki/Maximum_a_posteriori) which can be used to estimate the mode of the posterior distribution.\n",
    "\n",
    "\n",
    "Keep in mind that $x$ is a vector of $p$ observables. \n",
    "\n",
    "$$\n",
    "P(\\hat{y} | x^{(1)}, \\ldots, x^{(p)}) \n",
    "$$\n",
    "\n",
    "Using Bayes' theorem for conditional probabilities we get \n",
    "\n",
    "\\begin{align}\n",
    "f_{*} =& \\argmax_{\\hat{y}}\\left( P(\\hat{y} |x^{(1)}, \\ldots, x^{(p)})  \\right) \\\\\n",
    " =& \\argmax_{\\hat{y}}\\left(\\frac{P(\\hat{y}) \\cdot P(x^{(1)}, \\ldots, x^{(p)} | \\hat{y})} {P(x^{(1)}, \\ldots, x^{(p)})} \\right) \\\\\n",
    "  =& \\argmax_{\\hat{y}}\\left(P(\\hat{y}) \\cdot P(x^{(1)}, \\ldots, x^{(p)} | \\hat{y}) \\right) \\\\\n",
    "\\end{align}\n",
    "\n",
    "The problem is that usually the a priori distributions (aka the prior) is unknown and has to be approximated using strong assumptions.\n",
    "\n",
    "The __Naive Bayes__ classifier does exactly that. Let's assume that the features are completely independent of each other. In that case we can write \n",
    "\n",
    "$$\n",
    "P(x^{(1)}, \\ldots, x^{(p)} | \\hat{y}) = \\prod^{p}_{i=1} P(x^{(i)}| \\hat{y})\n",
    "$$\n",
    "\n",
    "\n",
    "\n",
    "$$\n",
    "f_{*} = \\argmax_{\\hat{y}}\\left( P(\\hat{y}) \\prod^{p}_{i=1} P(x^{(i)}| \\hat{y})  \\right)\n",
    "$$\n",
    "\n",
    "\n",
    "When $Y$ takes finite values (or classes) $Y = \\{Y_1, \\ldots Y_K\\}$, like in the Titanic dataset, we get the decision function for the Naive Bayes classifier \n",
    "\n",
    "$$\n",
    "y_{\\text{Naive Bayes}} = \\argmax_{k \\in {1, \\ldots, K}}\\left( P(Y_k) \\prod^{p}_{i=1} P(x^{(i)}| Y_k)  \\right)\n",
    "$$\n",
    "\n",
    "\n",
    "Note that the independency assumption is a very strong constraint. Features in real data are almost never independent of each other. __This assumption is almost always wrong__. (Hastie does call it the *idiot's* classifier). \n",
    "\n",
    "Even under that assumption the probabilities $P(x^{(i)}| \\hat{y})$ are often not known. They can however often be approximated.\n",
    "\n",
    "In the Titanic dataset we simply assume uniform priors. That means $P(Y={\\text{Survived}}) = \\frac{\\text{number of passengers that survived}}{\\text{total number of passengers}}$.\n",
    "\n",
    "\n",
    "For the feature distributions/likelihoods we assume a Gaussian distribution. For a feature $\\mathbf{x_i}$, for example the age of a passenger, we assume a Gaussian distributions and empirically estimate both mean $\\mu_k$ and standard deviation $\\sigma_k$ of the samples of passengers in class $Y_k$. For a given value of an observable $v$, e.g. the age of a single passenger, we then compute \n",
    "\n",
    "$$\n",
    "p(x=v \\, | \\, Y_k) \\sim \\mathcal{N}(\\mu_k, \\sigma_k)\n",
    "$$\n",
    "\n",
    "\n",
    "Below we perform Gaussian Naive Bayes on the Titanic dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:06.869949Z",
     "start_time": "2018-11-22T18:40:06.371164Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "X = passengers[['sex_code', 'pclass', 'fare', 'age']]\n",
    "y = passengers['survived']\n",
    "\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.6, random_state=0)\n",
    "\n",
    "reg = GaussianNB()\n",
    "reg.fit(X_train, y_train)\n",
    "\n",
    "plots.plot_bars_and_confusion(truth=y_test, prediction=reg.predict(X_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "\n",
    "## Support Vector Machines (SVMs)\n",
    "\n",
    "The basic assumption underlying the least squares approach is that the model is linear in the observed variables. \n",
    "This works for data which can be separated by a linear function (a hyperplane in the parameter space).\n",
    "\n",
    "But how do we know that this method finds the 'best' hyperplane for separating the two classes?\n",
    "\n",
    "And what if the data cannot be seperated by a plane?\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:07.257732Z",
     "start_time": "2018-11-22T18:40:06.874257Z"
    },
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "outputs": [],
   "source": [
    "# many possible lines to separate the data. Which one is 'better'?\n",
    "from sklearn.datasets import make_blobs\n",
    "\n",
    "X, y = make_blobs(\n",
    "    n_samples=150,\n",
    "    centers=2,\n",
    "    random_state=3,\n",
    "    cluster_std=0.70,\n",
    ")\n",
    "\n",
    "plt.figure()\n",
    "plt.axes(aspect=1)\n",
    "plt.scatter(X[:, 0], X[:, 1], c=y, s=50, cmap=cmap)\n",
    "\n",
    "xs = np.linspace(-6.5, 3, 2)\n",
    "plt.plot(xs, -2 * xs - 2, color='gray', linestyle='--')\n",
    "plt.plot(xs, -0.4 * xs + 2, color='gray', linestyle='--')\n",
    "plt.xlim([-6, 3])\n",
    "plt.ylim([-2, 6])\n",
    "plt.axis('off')\n",
    "None"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### The Support Vector Machine\n",
    "\n",
    "Again we minimze a loss function.\n",
    "\n",
    "$$\n",
    "L(\\v\\beta) = C \\max(0, 1 - y_i \\v\\beta^\\top x_i) + \\frac{\\lambda}{2}||{\\v\\beta}||^2\n",
    "$$\n",
    "\n",
    "Support Vector Machines try to find the hyperplane which maximimizes the margin to the points in different classes in the parameter space.\n",
    "\n",
    "$C$ and $\\lambda$ are two parameters which can be chosen beforehand. \n",
    "\n",
    "*Note that, to fit the definition above, the label encoding has to be $y_i \\in \\{-1, 1\\}$*\n",
    "\n",
    "Turns out that this is a convex optimization problem. We can find $\\v\\beta$ by minimizing (or maximizing) the dual lagrange formulation of the problem (See Hastie Chapter 12.2 for details.)\n",
    "\n",
    "$$\n",
    "L_D = \\sum_{i=1}^{N} \\alpha_i - \\frac 1 2 \\sum_{i=1}^{N}\\sum_{j=1}^{N} \\alpha_i \\alpha_j y_i y_j x_i^\\top x_j\n",
    "$$\n",
    "\n",
    "Using the values for the lagrange multipliers $\\alpha_i$ optained by maximizing $L_D$ we find \n",
    "\n",
    "$$\n",
    "\\v\\beta = \\sum_{i=1}^{N} \\hat{\\alpha}_i y_i x_i\n",
    "$$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:07.694436Z",
     "start_time": "2018-11-22T18:40:07.261108Z"
    },
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.svm import SVC\n",
    "from sklearn.datasets import make_blobs\n",
    "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis\n",
    "\n",
    "X, y = make_blobs(\n",
    "    n_samples=150,\n",
    "    centers=2,\n",
    "    random_state=3,\n",
    "    cluster_std=0.70,\n",
    ")\n",
    "\n",
    "plt.figure()\n",
    "plt.axes(aspect=1)\n",
    "plt.scatter(X[:, 0], X[:, 1], c=y, s=50, cmap=cmap)\n",
    "\n",
    "clf = SVC(kernel='linear')\n",
    "clf.fit(X, y)\n",
    "plots.draw_svm_decision_function(clf, colors='black', label='SVM')\n",
    "\n",
    "\n",
    "reg = LinearDiscriminantAnalysis()\n",
    "reg.fit(X, y)\n",
    "plots.draw_linear_regression_function(reg, label='Fisher Discrimiant', color='lightgray')\n",
    "\n",
    "\n",
    "reg = linear_model.LinearRegression()\n",
    "reg.fit(X, y)\n",
    "plots.draw_linear_regression_function(reg, label='Linear Regression', color='darkgray')\n",
    "\n",
    "plt.xlim([-6, 3])\n",
    "plt.ylim([-2, 6])\n",
    "plt.legend(loc='upper left', framealpha=0.95)\n",
    "None"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "So far the data has still been separable by a linear function. \n",
    "\n",
    "For many problems in real life however this isn't the case. \n",
    "\n",
    "Heres an example of (artificial) data which cannot be seperated by a line."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:08.082092Z",
     "start_time": "2018-11-22T18:40:07.697517Z"
    },
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.datasets import make_moons\n",
    "\n",
    "X, y = make_moons(n_samples=200, noise=0.10, random_state=0)\n",
    "\n",
    "plt.figure()\n",
    "plt.scatter(X[:, 0], X[:, 1], c=y, s=50, cmap=cmap)\n",
    "\n",
    "clf = SVC(kernel='linear')\n",
    "clf.fit(X, y)\n",
    "plots.draw_svm_decision_function(clf, colors='black')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Above we saw that the solution for $\\v\\beta$ only depends on the scalar product \n",
    "of the rows in our data matrix. We can *transform* our relationship between the datapoints by using a kernel funktion \n",
    "which satisfies \n",
    "$$\n",
    "k({x}_{i},{x}_{j})=\\varphi ({x}_{i})\\cdot \\varphi ({x}_{j})\n",
    "$$\n",
    "\n",
    "The computation of $\\v\\beta$ does not change.\n",
    "\n",
    "\n",
    "$$\n",
    "\\v\\beta = \\sum_{i=1}^{N} \\hat{\\alpha}_i y_i \\varphi ({x}_{i})\n",
    "$$\n",
    "\n",
    "\n",
    "Here we are going to use:\n",
    "\n",
    "$$\n",
    "k(x_0, x_1) = \\exp(-(x_0^2 + x_1^2)) = \\exp(-x_0^2) \\cdot \\exp(-x_1^2)\n",
    "$$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:09.726556Z",
     "start_time": "2018-11-22T18:40:08.085200Z"
    },
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "outputs": [],
   "source": [
    "from mpl_toolkits import mplot3d\n",
    "X, y = make_moons(n_samples=400, noise=0.10, random_state=0)\n",
    "\n",
    "\n",
    "# add a dimension by applying a transformation on the two variables in the data. \n",
    "r = np.exp(-(X[:, 0] ** 2 + X[:, 1] ** 2))\n",
    "\n",
    "plots.plot_3d_views(np.c_[X,r], y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:10.195610Z",
     "start_time": "2018-11-22T18:40:09.729968Z"
    },
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "outputs": [],
   "source": [
    "X, y = make_moons(n_samples=200, noise=0.10)\n",
    "\n",
    "\n",
    "clf = SVC(kernel='rbf') #use the radial basis function instead of the linear one.\n",
    "clf.fit(X, y)\n",
    "\n",
    "plt.figure()\n",
    "plt.axes(aspect=1)\n",
    "plt.scatter(X[:, 0], X[:, 1], c=y, s=50, cmap=cmap)\n",
    "plots.draw_svm_decision_function(clf, colors='black', label='SVM')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "The same approach works for other linear methods as well. \n",
    "\n",
    "What makes SVM's so special?:\n",
    "\n",
    "+ SVM's have proven to perform very well for many use-cases.\n",
    "\n",
    "+ SVM's handle large number of dimensions relativly fast.\n",
    "\n",
    "+ The kernel functions basically come for free.\n",
    "\n",
    "+ Easily extendable to multi-class problems.\n",
    "\n",
    "+ Generalize better than comparable methods. \n",
    "\n",
    "Kernel functions are constrained to fulfill certain criteria. *(See Chapter 12.3.1 in the Hastie Book)*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:18.798755Z",
     "start_time": "2018-11-22T18:40:10.199569Z"
    },
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.svm import SVC\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "\n",
    "X = passengers[['sex_code', 'pclass', 'fare', 'age']]\n",
    "y = passengers['survived']\n",
    "\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.5, random_state=0)\n",
    "\n",
    "\n",
    "# Use linear kernel\n",
    "svc = SVC(kernel='linear')\n",
    "svc.fit(X_train, y_train)\n",
    "prediction_linear = svc.predict(X_test)\n",
    "\n",
    "# Use the rbf kernel\n",
    "svc_rbf = SVC(kernel='rbf')\n",
    "svc_rbf.fit(X_train, y_train)\n",
    "prediction_rbf = svc_rbf.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:18.798755Z",
     "start_time": "2018-11-22T18:40:10.199569Z"
    },
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "outputs": [],
   "source": [
    "fig1, axes1 = plt.subplots(1, 2)\n",
    "\n",
    "fig1.suptitle('Linear Kernel')\n",
    "plots.plot_bars_and_confusion(\n",
    "    truth=y_test,\n",
    "    axes=axes1.tolist(),\n",
    "    prediction=prediction_linear,\n",
    "    vmin=0, vmax=182,\n",
    ")\n",
    "\n",
    "fig2, axes2 = plt.subplots(1, 2)\n",
    "fig2.suptitle('RBF Kernel')\n",
    "plots.plot_bars_and_confusion(\n",
    "    truth=y_test,\n",
    "    prediction=prediction_rbf,\n",
    "    axes=axes2.tolist(),\n",
    "    vmin=0, vmax=182,\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Extending SVMs to more dimensions\n",
    "\n",
    "The Titanic dataset we looked at in the previous excercise had four observed variables or dimensions and 714 observations in total. \n",
    "\n",
    "        > X = data[['sex_code', 'pclass', 'fare', 'age']]\n",
    "        > X.shape\n",
    "        (714, 4)\n",
    "\n",
    "Now we take a dataset that has 64 variables. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:19.929142Z",
     "start_time": "2018-11-22T18:40:18.801741Z"
    },
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from sklearn import datasets\n",
    "from sklearn.svm import SVC\n",
    "\n",
    "# obfuscation, so you don't immediately get what we are doing here\n",
    "dataset = b\"\\x64\\x69\\x67\\x69\\x74\\x73\".decode('ascii')\n",
    "load_data = getattr(datasets, 'load_' + dataset)\n",
    "\n",
    "\n",
    "X, y = load_data(return_X_y=True)\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.5)\n",
    "\n",
    "#SVC with default settings.\n",
    "clf = SVC(kernel='rbf')\n",
    "\n",
    "# We learn the kernel on the first half of the data\n",
    "clf.fit(X_train, y_train)\n",
    "\n",
    "# Now predict the value of the digit on the test sample\n",
    "y_prediction = clf.predict(X_test)\n",
    "\n",
    "cm = confusion_matrix(y_test, y_prediction)\n",
    "\n",
    "plt.figure()\n",
    "plt.axes(aspect=1)\n",
    "sns.heatmap(\n",
    "    cm,\n",
    "    annot=True,\n",
    "    fmt='d',\n",
    "    cmap='inferno',\n",
    ")\n",
    "None"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "__What mystery data did we just classifiy?__\n",
    "\n",
    "The SVM can easily classifiy a dataset of many observables and target classes.\n",
    "\n",
    "This data set had 64 observables and 10 different classes.\n",
    "\n",
    "Lets take the 64 numbers in the single observations and plot them into a $8\\times8$ grid"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:20.203589Z",
     "start_time": "2018-11-22T18:40:19.932972Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn import datasets, svm, metrics\n",
    "\n",
    "plt.figure(figsize=(9, 2))\n",
    "plt.imshow([X[0]], aspect='auto', cmap='gray_r',)\n",
    "plt.grid(False)\n",
    "plt.yticks([])\n",
    "None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:21.217537Z",
     "start_time": "2018-11-22T18:40:20.207290Z"
    },
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "outputs": [],
   "source": [
    "fig, axs = plt.subplots(2, 5)\n",
    "for x_i, ax in zip(X[0:10], axs.flat):\n",
    "    img = x_i.reshape(-1, 8)\n",
    "    ax.imshow(img, cmap='gray_r', interpolation='nearest')\n",
    "    ax.grid('off')\n",
    "    ax.axis('off')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "This classifier just performed __character recognition__ on raw image inputs without any feature engineering.\n",
    "\n",
    "Currently there is a lot of buzz (or even hype) about image recognition tasks and neural networks (Deep Learning etc.)\n",
    "Neural networks and SVMs are very similar in nature they just use slightly modified loss functions.\n",
    "\n",
    "For further information please checkout Stanford's computer science lecture CS231n, especially the chapter on linear classification.\n",
    "\n",
    "[http://cs231n.github.io](http://cs231n.github.io)\n",
    "\n",
    "Also, checkout what can happen if this goes wrong:\n",
    "\n",
    "* David Kriesel: „Glaube keinem Scan, den du nicht selbst gefälscht hast” https://youtu.be/7FeqF1-Z1g0 (German)\n",
    "* David Kriesel: “Lies, damned lies and scans” https://youtu.be/c0O6UXrOZJo (English)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Local Optimization\n",
    "\n",
    "So far we looked at loss functions which optimized some global optimization criterion.\n",
    "\n",
    "We have seen that not all problems can be solved by a linear model.\n",
    "It is often not possible to find a transformation, or kernel function, to transform the data into linearly separable classes.\n",
    "\n",
    "Neither by hand nor by some automated procedure.\n",
    "\n",
    "The idea of local optimization methods is to split the \n",
    "parameter space into subspaces where the problem is easier to solve.\n",
    "\n",
    "### Decision Trees\n",
    "\n",
    "Idea:\n",
    "* Split the parameter space into many subspaces where observations of the same class live.\n",
    "\n",
    "We split the space into two regions \n",
    "\n",
    "$$\n",
    "R_1(j, s) = \\{X |\\, X_j \\le s\\}\n",
    "$$\n",
    "\n",
    "and \n",
    "\n",
    "$$\n",
    "R_2(j, s) = \\{X |\\, X_j \\gt s\\}.\n",
    "$$\n",
    "\n",
    "These regions define half-planes in the parameter space.\n",
    "\n",
    "In the example below we define 4 half-planes/regions to select the region in space where the blue points live. \n",
    "Now we can optimize a loss function in each of the overlapping regions.\n",
    "\n",
    "In the example below we can simply use a decision function that is constant in each region.\n",
    "\n",
    "$$\n",
    "f(x) = \\sum_{m=1}^{M} c_m \\mathbb{I}(x \\in R_m)\n",
    "$$\n",
    "\n",
    "where \n",
    "\n",
    "$$\n",
    "\\mathbb{I}(x \\in R_m) =\\begin{cases}\n",
    "1,  & \\text{if $x \\in R_m$} \\\\\n",
    "0, & \\text{otherwise}\n",
    "\\end{cases}\n",
    "$$\n",
    "\n",
    "and $c_m$ is the class we assign to each region $R_m$\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:21.961809Z",
     "start_time": "2018-11-22T18:40:21.220973Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.datasets import make_circles\n",
    "\n",
    "X, y = make_circles(n_samples=1000, noise=0.10, random_state=0, factor=0.2)\n",
    "\n",
    "plt.figure()\n",
    "plt.axes(aspect=1)\n",
    "plt.scatter(X[:, 0], X[:, 1], c=y, s=50, cmap=cmap)\n",
    "\n",
    "plt.axhspan(-0.5, 1.5, color='red', alpha=0.1)\n",
    "plt.axvspan(-0.5, 1.5, color='green', alpha=0.1)\n",
    "\n",
    "plt.axhspan(0.5, -1.5, color='blue', alpha=0.1)\n",
    "plt.axvspan(0.5, -1.5, color='yellow', alpha=0.1)\n",
    "\n",
    "plt.xlim([-1.4, 1.4])\n",
    "plt.ylim([-1.4, 1.4])\n",
    "None"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we choose the constant $c_m$ to be the average target value in that region. \n",
    "\n",
    "$$\n",
    "c_m  = \\frac{1}{N} \\sum_{x_i \\in R_m} y_i\n",
    "$$\n",
    "\n",
    "This works extremly well in the example above. The question remains on how to find the splits.\n",
    "This problem is again NP-Hard and therefore takes too long to solve. \n",
    "We can however employ a greedy strategy again. \n",
    "\n",
    "Algorithms performing these steps are called __Decision Tree__ methods.\n",
    "\n",
    "We perform recursive binary splits of the subspace using the split which yields the lowest value for the loss function. One advantage of binary splits is the straightforward intepretation of the results.\n",
    "The image below shows what a decision tree looks like."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:22.150340Z",
     "start_time": "2018-11-22T18:40:21.964818Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.datasets import make_moons\n",
    "\n",
    "X, y = make_moons(n_samples=200, noise=0.10, random_state=1234)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:22.150340Z",
     "start_time": "2018-11-22T18:40:21.964818Z"
    }
   },
   "outputs": [],
   "source": [
    "clf = DecisionTreeClassifier(max_depth=4, criterion='entropy')\n",
    "clf.fit(X, y)\n",
    "\n",
    "from IPython.display import Image\n",
    "Image(plots.draw_tree(clf))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The loss function this time is called the Information Gain. \n",
    "To find the best split in a node, we maximize the Information Gain, $\\mathit{IG}$,  over all features  $X \\in \\mathbf{X}$ and splits $s$\n",
    "\n",
    "$$\n",
    "    \\mathit{IG}(X,Y) = H(Y) - H(Y |\\, X).\n",
    "$$\n",
    "\n",
    "where $H$ is the entropy function\n",
    "\n",
    "\\begin{equation}\n",
    "  H(Y) = - \\sum_{z \\in Z} P(Y=z) \\log_2{P(Y=z)}.\n",
    "\\end{equation}\n",
    "\n",
    "for the random variable $Y$ which takes values from the finite set of symbols $Z$\n",
    "\n",
    "and $H(Y |\\, X)$ is the conditional entropy.\n",
    "\n",
    "\\begin{aligned}\n",
    "    H(Y |\\, X)  &=  \\sum_{m \\in M}P(X = m) H(Y |\\, X = m)  \\\\\n",
    "       &= - \\sum_{m \\in M}P(X = m) \\sum_{z \\in Z} P(Y = z |\\, X = m)\\log{P(Y = z |\\, X = m)}.\n",
    "\\end{aligned}\n",
    "\n",
    "\n",
    "In the context of decision trees we want to know the entropy of $Y$ given some new information about $X$.\n",
    "In this context $H(Y)$ can be understood as the entropy of $Y$ before a split and $H(Y | \\, X)$ the entropy after the split. Probabilities are assumed to be uniform. Hence we get\n",
    "\n",
    "\n",
    "\\begin{aligned}\n",
    "  H(Y) =& - \\sum_{z \\in \\{\\text{Dead, Alive}\\}} P(Y=z) \\log_2{P(Y=z)} \\\\\n",
    "       =& -\\left( \\frac{ n_{ \\text{ dead }}}{n} \\log_2{\\frac{ n_{ \\text{ dead }}}{n}} + \\frac{ n_{ \\text{ alive }}}{n} \\log2{\\frac{ n_{ \\text{ alive }}}{n}} \\right)\n",
    "\\end{aligned}\n",
    "\n",
    "\n",
    "\\begin{aligned}\n",
    "    H(Y |\\, X)  &=  \\sum_{g \\in \\{\\text{Male}, \\text{Female}\\} }P(X_{\\text{Sex}} = g) H(Y |\\, X = m)  \\\\\n",
    "        &=   \\frac{ n_{ \\text{male}}}{n} H(Y |\\, X_{\\text{Sex}} = \\text{male}) \n",
    "        + \\frac{ n_{ \\text{female}}}{n} H(Y |\\, X_{\\text{Sex}} = \\text{female})\n",
    "\\end{aligned}\n",
    "\n",
    "where the specific conditioned entropy is \n",
    "\n",
    "\\begin{aligned}\n",
    "    H(Y |\\, X_{\\text{Sex}} = \\text{male}) = -\\left( \n",
    "        \\frac{ n_{\\text{ dead, male }}}{n_{\\text{male}}} \\log2{\\frac{ n_{ \\text{ dead, male }}}{n_{\\text{male}}}}\n",
    "        + \\frac{ n_{\\text{alive, male}}}{n_{\\text{male}}} \\log2{\\frac{ n_{ \\text{ alive, male}}}{n_{\\text{male}}} }\n",
    "    \\right)\n",
    "\\end{aligned}\n",
    "\n",
    "When using continuous features such as *Age* each possible split has to be tested. Using the nomenclature from above where $s$ is some split we can write down the total (local) loss function\n",
    "$$\n",
    "  \\max_{(X, s) \\in \\, \\mathbf{X} \\times {S}}IG(X,Y) =   \\max_{(X, s) \\in \\, \\mathbf{X} \\times {S}} ( H(Y) - H(Y |\\, X) ).\n",
    "$$\n",
    "\n",
    "We call this local optimization since we only evaluate the loss function in the subspace of the data defined by the decision tree. It is local to the current node of the decision tree. \n",
    "\n",
    "Here is another nice explanation of entropy and information gain\n",
    "https://stackoverflow.com/questions/1859554/what-is-entropy-and-information-gain\n",
    "\n",
    "Outline for building a binary decision tree:\n",
    "\n",
    "```python\n",
    "def build_tree(space)\n",
    "    if stopping_criterion_fulfilled():\n",
    "        return {}\n",
    "\n",
    "    variable, split_point = find_best_split(space)\n",
    "\n",
    "    left, right = split_space(space, variable, split_point)\n",
    "\n",
    "    left_tree = build_tree(left)\n",
    "    right_tree = build_tree(right)\n",
    "\n",
    "    return {'node' : (variable, split_point), 'left': left_tree, 'right': right_tree}\n",
    "```           "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:22.594221Z",
     "start_time": "2018-11-22T18:40:22.154795Z"
    },
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.datasets import make_moons\n",
    "\n",
    "X, y = make_moons(n_samples=200, noise=0.10, random_state=0)\n",
    "\n",
    "clf = DecisionTreeClassifier(max_depth=5, criterion='entropy')\n",
    "clf.fit(X, y)\n",
    "\n",
    "plt.figure()\n",
    "plt.axes(aspect=1)\n",
    "plt.scatter(X[:, 0], X[:, 1], c=y, s=50, cmap=cmap)\n",
    "plots.draw_decision_boundaries(clf, cmap=cmap)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Below we apply a decision tree to the Titanic dataset while varying some parameters like depth and which loss function to use. The results are plotted into a heatmap."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:24.110499Z",
     "start_time": "2018-11-22T18:40:22.596487Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import ParameterGrid\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "\n",
    "\n",
    "# load the data\n",
    "X = passengers[['sex_code', 'pclass', 'fare', 'age']]\n",
    "y = passengers['survived']\n",
    "\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=1235)\n",
    "\n",
    "# loop over every possible combination of parameters \n",
    "perf = []\n",
    "ps = ParameterGrid({'max_depth': range(1, 20), 'criterion': ['entropy', 'gini']})\n",
    "\n",
    "for d in ps:\n",
    "    clf = DecisionTreeClassifier(max_depth=d['max_depth'], criterion=d['criterion'])\n",
    "    clf.fit(X_train, y_train)\n",
    "    acc = accuracy_score(y_test, clf.predict(X_test))\n",
    "    \n",
    "    perf.append({\n",
    "        'max_depth': d['max_depth'],\n",
    "        'criterion': d['criterion'],\n",
    "        'accuracy': acc,\n",
    "    })\n",
    "    \n",
    "\n",
    "# plot a heatmap to see the best paramter combination. \n",
    "perf = pd.DataFrame(perf).pivot(index='max_depth', columns='criterion', values='accuracy')\n",
    "\n",
    "plt.figure()\n",
    "sns.heatmap(perf, cmap='inferno', annot=True, fmt='.3f', label='Accuracy', vmin=0.5, vmax=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    },
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "### k - Nearest Neighbour Methods\n",
    "\n",
    "The k-NN classifier is another local optimizer.\n",
    "\n",
    "Lets assume that the decision function is constant over some local region in the parameter space:\n",
    "\n",
    "$$\n",
    "\\hat{f}(x_0) = \\hat{y} = \\frac{1}{k} \\sum_{x_i \\in N_k(x_o)} y_i\n",
    "$$\n",
    "\n",
    "where $x_i \\in N_k(x)$ describes the $k$ points in the training data $\\m{X}$ that are in the *neighbourhood* of $x_0$.\n",
    "\n",
    "To put it in words. We assume $x$ will have the same $y$ as other points nearby."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:24.824971Z",
     "start_time": "2018-11-22T18:40:24.116125Z"
    },
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.datasets import make_moons\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "\n",
    "X, y = make_moons(n_samples=200, noise=0.05)\n",
    "\n",
    "\n",
    "knn = KNeighborsClassifier(n_neighbors=100)\n",
    "knn.fit(X, y)\n",
    "\n",
    "plt.figure()\n",
    "plt.scatter(X[:, 0], X[:, 1], c=y, s=50, cmap='winter')\n",
    "plots.draw_decision_boundaries(knn)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This works pretty well in this artifical, low-noise, example. \n",
    "\n",
    "Classification on noisy data will not work as good.\n",
    "\n",
    "Real world data always has some form of noise."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:25.466347Z",
     "start_time": "2018-11-22T18:40:24.828675Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "\n",
    "X, y = make_moons(n_samples=100, noise=0.2, random_state=1234)\n",
    "\n",
    "fig, axs = plt.subplots(1, 2, figsize=(9, 3))\n",
    "\n",
    "\n",
    "for k, ax in zip([40, 1], axs.flat): \n",
    "    knn = KNeighborsClassifier(n_neighbors=k)\n",
    "    knn.fit(X, y)\n",
    "    prediction = knn.predict(X)\n",
    "\n",
    "    ax.scatter(X[:, 0], X[:, 1], c=y, s=50, cmap='winter')\n",
    "    plots.draw_decision_boundaries(knn, ax=ax)\n",
    "    ax.set_title(f'Accuracy for k={k}: {accuracy_score(y, prediction)}')\n",
    "    ax.set_axis_off()\n",
    "    ax.set_aspect(1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Classifier Validation \n",
    "\n",
    "So far we only looked at the *accuracy* measure of each classifier. We did not however compute any uncertainties on that number. There are ways to compute uncertainties on numbers like accuracy. First we try to motivate why these errors are needed.\n",
    "\n",
    "### Overfitting (Bias-Variance Tradeoff)\n",
    "\n",
    "Assume the target $y$ is generated by some function $f(x)$ with added gaussian noise $\\epsilon$\n",
    "\n",
    "$$\n",
    "y = f(x) + \\epsilon, \\qquad \\epsilon \\propto \\mathcal{N}(\\mu=0, \\sigma)\n",
    "$$\n",
    "\n",
    "The mean squared error ($\\mse$) of the predictor function $\\hat{f}(x) = \\hat{y}$ is\n",
    "\n",
    "$$\n",
    "\\mse(y, \\hat{f}) = E[(y - \\hat{f}(x))^2] .\n",
    "$$\n",
    "\n",
    "We can actually calculate this expectation value for the k-NN classifier under the assumptions above.\n",
    "\n",
    "Some mathematical definitions up front.\n",
    "\n",
    "__Variance__ of a random variable $X$ \n",
    "  \n",
    "  $$\n",
    "  \\Var(X) = \\E[(X - \\E[X])^2] = \\E[X^2] - \\E[X]^2 \\iff \\E[X^2] = \\Var[X] +  \\E[X]^2 \n",
    "  $$\n",
    "  \n",
    "__Bias__ of an estimator $\\hat{f}$\n",
    "\n",
    "  $$\n",
    "  \\Bias(\\hat{f}) = \\E[\\hat{f} - f] = \\E[\\hat{f}] - \\E[f] = \\E[\\hat{f}] - f\n",
    "  $$\n",
    "\n",
    "The last equality holds since $f$ is a fixed, deterministic,  function and hence its values never change.\n",
    "\n",
    "$$\n",
    "\\E[f] = f\n",
    "$$\n",
    "\n",
    "\n",
    "\n",
    "Using the definition of the variance, we can substitute our function $y = f(x) + \\epsilon$, which gives \n",
    "\n",
    "\\begin{align}\n",
    "\\Var[y] = \\Var[f(x) + \\epsilon] & = \\Var[f(x)] + \\Var[\\epsilon] \\\\\n",
    "                                & = \\Var[f(x)] + \\sigma^2 \\\\\n",
    "                                & = \\E[f(x)^2] - \\E[f(x)]^2 + \\sigma^2 \\\\\n",
    "                                & = f(x)^2 - f(x)^2 + \\sigma^2 \\\\\n",
    "                                & = \\sigma^2 \n",
    "\\end{align}\n",
    "\n",
    "We can now simplify $\\E[\\mse(y, \\hat{f})]$:\n",
    "\n",
    "\\begin{align}\n",
    " \\E[(y - \\hat{f}(x))^2] & = \\E[y^2 + \\hat{f}^2 - 2y\\hat{f}] \\\\\n",
    "                        & \\ldots \\\\\n",
    "                        & = \\sigma^2 + \\Var[\\hat{f}] + \\Bias[\\hat{f}]^2.\n",
    "\\end{align}\n",
    "\n",
    "For the k-NN classifier there exists an analytical expression for $\\E[\\mse(y, \\hat{f})]$.\n",
    "\n",
    "Using \n",
    "$$\n",
    "\\hat{f}(x) = \\hat{y} = \\frac{1}{k} \\sum_{x_i \\in N_k(x)} y_i\n",
    "$$\n",
    "\n",
    "and once again assuming the data is subject to normally distributed noise $\\epsilon$\n",
    "\n",
    "$$\n",
    "y_i = f(x_i) + \\epsilon_i\n",
    "$$\n",
    "\n",
    "one can calculate \n",
    "\n",
    "$$\n",
    " \\E[(y - \\hat{f}(x))^2] = \\sigma^2 + \\Var[\\hat{f}] + \\Bias[\\hat{f}]^2.\n",
    "$$\n",
    "\n",
    "\n",
    "Starting with $\\Var(\\hat{f})$\n",
    "\n",
    "\\begin{align}\n",
    "\\Var(\\hat{f}(x_0)) &= \\Var \\left( \\frac{1}{k} \\sum_{y_i \\in N_k(x_0)} y_i \\right)  \\\\\n",
    "&= \\frac{1}{k^2}  \\sum_{y_i \\in N_k(x_0)} \\Var \\left( f(x_i) + \\epsilon_i \\right)  \\\\\n",
    "&= \\frac{1}{k^2}  \\sum_{y_i \\in N_k(x_0)} \\Var \\left( f(x_i) \\right) + \\Var \\left( \\epsilon_i \\right)  \\\\\n",
    "&= \\frac{1}{k^2}  \\sum_{y_i \\in N_k(x_0)} \\Var \\left( \\epsilon_i \\right)  \\\\\n",
    "&= \\frac{1}{k^2} k \\sigma^2  \\\\\n",
    "&= \\frac{\\sigma^2}{k}\n",
    "\\end{align}\n",
    "\n",
    "Now the bias term\n",
    "\n",
    "\\begin{align}\n",
    "    \\Bias^2(\\hat{f}) & = \\E\\left[\\hat{f} - y \\right]^2 \\\\\n",
    "                     & = \\E\\left[\\frac{1}{k} \\left( \\sum_{y_i \\in N_k(x_0)} y_i \\right) - y \\right]^2 \\\\\n",
    "                     & = \\E\\left[\\frac{1}{k} \\left( \\sum_{x_i \\in N_k(x_0)} f(x_i) + \\epsilon_i \\right) - ( f(x_0) + \\epsilon_0 )\\right]^2 \\\\\n",
    "                     & = \\E\\left[\\frac{1}{k} \\left( \\sum_{x_i \\in N_k(x_0)} f(x_i) \\right) - f(x_0) \\right]^2 \\\\\n",
    "                    & = \\left( \\frac{1}{k} \\left( \\sum_{x_i \\in N_k(x_0)} f(x_i) \\right) - f(x_0) \\right)^2\n",
    "\\end{align}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Putting it together gives \n",
    "\n",
    "$$\n",
    " \\E_{knn}[(y - \\hat{f}(x))^2] = \\sigma^2 + \\Var[\\hat{f}] + \\Bias[\\hat{f}]^2 = \\sigma^2 + \\frac{\\sigma^2}{k} + \\left( \\frac{1}{k} \\left( \\sum_{y_i \\in N_k(x_0)} f(x_i) \\right) - f(x_0) \\right)^2.\n",
    "$$\n",
    "\n",
    "The expectation of mean squared error depends on the choice of $k$ and the noise of the data $\\sigma$. \n",
    "\n",
    "When $k$ grows, the dependency on the data noise decreases but the bias likely increases. \n",
    "\n",
    "For a smooth $f(x)$, the points close to $x_0$ will have a rather similar $f(x_0)$, but as $k$ increases, the distances grow and the differences will become larger\n",
    "\n",
    "This so called Bias-Variance dillemma is a universal problem in supervised machine learning. \n",
    "\n",
    "There are two error sources:\n",
    "\n",
    "* High bias might decrease overall predictor performance.\n",
    "* High variance can make the learned parameters prone to noise in the training data.  \n",
    "\n",
    "If the parameters are tuned to the noise in the training data, the model will not generalize to new data. \n",
    "\n",
    "This problem is called __overfitting__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:26.444566Z",
     "start_time": "2018-11-22T18:40:25.469827Z"
    }
   },
   "outputs": [],
   "source": [
    "k = 25\n",
    "\n",
    "X, y = make_moons(n_samples=200, noise=0.7, random_state=1234)\n",
    "X_test, y_test = make_moons(n_samples=100, noise=0.7, random_state=1337)\n",
    "\n",
    "knn = KNeighborsClassifier(n_neighbors=k)\n",
    "knn.fit(X, y)\n",
    "\n",
    "fig, (ax1, ax2) = plt.subplots(1, 2)\n",
    "\n",
    "\n",
    "ax1.scatter(X[:, 0], X[:, 1], c=y, s=30, cmap='winter')\n",
    "plots.draw_decision_boundaries(knn, ax=ax1)\n",
    "ax1.set_title(f'on Train Sample {accuracy_score(y, knn.predict(X))}')\n",
    "ax1.axis('off')\n",
    "\n",
    "ax2.scatter(X_test[:, 0], X_test[:, 1], c=y_test, s=30, cmap='winter')\n",
    "ax2.scatter(X[:, 0], X[:, 1], c=y, s=15, alpha=0.1,  cmap='winter')\n",
    "plots.draw_decision_boundaries(knn, ax=ax2)\n",
    "ax2.set_title(f'on Test Sample {accuracy_score(y_test, knn.predict(X_test))}')\n",
    "ax2.axis('off')\n",
    "\n",
    "fig.suptitle(f'Accuracy for k={k}')\n",
    "None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:29.857163Z",
     "start_time": "2018-11-22T18:40:26.451832Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "from tqdm.auto import tqdm\n",
    "\n",
    "X, y = make_moons(n_samples=400, noise=0.2, random_state=1234)\n",
    "X_test, y_test = make_moons(n_samples=400, noise=0.2, random_state=1337)\n",
    "\n",
    "n_neighbors = np.arange(1, 201)\n",
    "\n",
    "e_train = []\n",
    "e_test = []\n",
    "for k in tqdm(n_neighbors):\n",
    "    knn = KNeighborsClassifier(n_neighbors=k)\n",
    "    knn.fit(X, y)\n",
    "    e_train.append(1 - accuracy_score(y, knn.predict(X)))\n",
    "    e_test.append(1 - accuracy_score(y_test, knn.predict(X_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:30.482619Z",
     "start_time": "2018-11-22T18:40:29.860644Z"
    }
   },
   "outputs": [],
   "source": [
    "fig, ax = plt.subplots()\n",
    "\n",
    "ax.plot(n_neighbors, e_train,'.', color='#FF6B6B', ms=10, label='Training Sample')\n",
    "ax.plot(n_neighbors, e_test, '.' ,color='#FFAE6B', ms=10, label='Test Sample', )\n",
    "\n",
    "ax.set_xlabel('k')\n",
    "ax.set_ylabel('Prediction Error')\n",
    "\n",
    "\n",
    "ax.text(150, 0.026, 'Increasing Model Complexity')\n",
    "ax.arrow(150, 0.018, -50, 0, width = 0.0005, head_width=0.003, head_length=3, fc='k', ec='k')\n",
    "ax.legend()\n",
    "ax.invert_xaxis()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    }
   },
   "source": [
    "### Validation on independent test sets\n",
    "\n",
    "Validating the fitted models is essential for avoiding overfitting.\n",
    "\n",
    "The predictions error has to be assesed on an independent test dataset. \n",
    "\n",
    "Models might still be susceptible to noise in the training data.\n",
    "\n",
    "#### Cross Validation\n",
    "\n",
    "A $k$-fold cross validation automatically splits the training data into $k$ subsets.\n",
    "\n",
    "The model is then trained on $k-1$ subsets and evaluated on the remaining set. \n",
    "\n",
    "\n",
    "<img width=\"80%\" src=\"data:image/png;base64,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\"/>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Below we use sklearn's cross validation method to estimate the accuracy of a Decision Tree  classifier on many independent test sets. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:30.707080Z",
     "start_time": "2018-11-22T18:40:30.491620Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.model_selection import cross_validate, StratifiedKFold\n",
    "\n",
    "X = passengers[['sex_code', 'pclass', 'fare', 'age']]\n",
    "y = passengers['survived']\n",
    "\n",
    "\n",
    "clf = DecisionTreeClassifier(max_depth=3)\n",
    "\n",
    "r = cross_validate(\n",
    "    clf, X=X, y=y,\n",
    "    cv=StratifiedKFold(10, shuffle=True),   \n",
    "    scoring=['accuracy', 'precision', 'recall'],\n",
    "    return_train_score=True,\n",
    ")\n",
    "\n",
    "\n",
    "print('Accuracy on test sets:     {:.3f} ± {:.3f}'.format(r['test_accuracy'].mean(), r['test_accuracy'].std() ))\n",
    "print('Accuracy on training sets: {:.3f} ± {:.3f}'.format(r['train_accuracy'].mean(), r['train_accuracy'].std() ))\n",
    "\n",
    "with np.printoptions(precision=3):\n",
    "    for k, v in r.items():\n",
    "        print(f'{k:>15}:', v)\n",
    "        \n",
    "clf"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Other Quality Metrics\n",
    "\n",
    "Accuracy is not always a good measure of model quality.\n",
    "\n",
    "Imagine a classifier function which simply predicts a fixed outcome.\n",
    "\n",
    "On an imbalanced dataset this classifier will have an accuracy equal to the ratio of positive examples to the total number of examples."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:30.848343Z",
     "start_time": "2018-11-22T18:40:30.710164Z"
    }
   },
   "outputs": [],
   "source": [
    "def f_fixed(x):\n",
    "    return np.zeros(x.shape[0])\n",
    "\n",
    "X = passengers[['sex_code', 'pclass', 'fare', 'age']]\n",
    "y = passengers['survived']\n",
    "\n",
    "\n",
    "print('Accuracy of fixed classifier {:1.6f} \\n'.format(accuracy_score(y, f_fixed(X))))\n",
    "\n",
    "print('Ratio of survived to total passengers: ')\n",
    "print(data.survived.value_counts(normalize=True))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Another view at the confusion matrix. It generalizes to more than two classes as seen in the picture below.\n",
    "\n",
    "<img width=\"60%\" src=\"data:image/png;base64,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\">\n",
    "\n",
    "The numbers in the confusion matrix can be used to calculate a whole range of qulity criteria.\n",
    "\n",
    "Lets build a classifier which randomly chooses an outcome and look at the different criteria.\n",
    "\n",
    "```\n",
    "def f_random(passenger):\n",
    "    return np.random.choice([True, False)\n",
    "```\n",
    "\n",
    "\n",
    "<a href=\"https://stackoverflow.com/questions/31324218/scikit-learn-how-to-obtain-true-positive-true-negative-false-positive-and-fal\" style=\"color:#BBBBBB;\">The nice Stack Overflow post where I stole the picture from.</a>\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:30.933660Z",
     "start_time": "2018-11-22T18:40:30.851402Z"
    }
   },
   "outputs": [],
   "source": [
    "def f_random(x):\n",
    "    return np.random.randint(2, size=len(x))\n",
    "\n",
    "prediction = f_random(X)\n",
    "cm = confusion_matrix(y, prediction)\n",
    "\n",
    "FP = cm.sum(axis=0) - np.diag(cm)  \n",
    "FN = cm.sum(axis=1) - np.diag(cm)\n",
    "TP = np.diag(cm)\n",
    "TN = cm.sum() - (FP + FN + TP)\n",
    "\n",
    "\n",
    "metrics = {}\n",
    "# Sensitivity, hit rate, recall, efficiency, or true positive rate\n",
    "metrics['TPR'] = TP / (TP + FN)\n",
    "\n",
    "\n",
    "# Specificity or true negative rate\n",
    "metrics['TNR'] = TN / (TN + FP) \n",
    "\n",
    "# Precision or positive predictive value or purity\n",
    "metrics['PPV'] = TP / (TP + FP)\n",
    "\n",
    "# Negative predictive value\n",
    "metrics['NPV'] = TN / (TN + FN)\n",
    "\n",
    "# Fall out or false positive rate\n",
    "metrics['FPR'] = FP / (FP + TN)\n",
    "\n",
    "# False negative rate\n",
    "metrics['FNR'] = FN / (TP + FN)\n",
    "\n",
    "# False discovery rate\n",
    "metrics['FDR'] = FP / (TP + FP)\n",
    "\n",
    "# Overall accuracy\n",
    "metrics['ACC'] = (TP + TN) / (TP + FP + FN + TN)\n",
    "\n",
    "print('       P=0     P=1')\n",
    "for metric, value in metrics.items():\n",
    "    print(f'{metric} = {value[0]:.3f} | {value[1]:.3f}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:40:30.947487Z",
     "start_time": "2018-11-22T18:40:30.937239Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.metrics import recall_score, precision_score\n",
    "\n",
    "print(f'{recall_score(y, prediction):.3f}')\n",
    "print(f'{precision_score(y, prediction):.3f}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Precision and recall can be understood in an intuitive way\n",
    "\n",
    "* __Recall__ How many of the wanted examples are found.\n",
    "* __Precision__ The percentage of the found examples that are actually relevant.\n",
    "\n",
    "So what should you optimize for? Maximumg accuracy or precision or recall?\n",
    "\n",
    "There is no clear answer.  \n",
    "It depends on your use-case.  \n",
    "Can you tolerate false positives?   \n",
    "Can you tolerate losing some true positives?  \n",
    "\n",
    "We will always have to make the trade-off between recall and precission. \n",
    "There are several metrics which try to combine both into one.\n",
    "\n",
    "The $f_{\\beta}$ score is one example.\n",
    "\n",
    "$$\n",
    "f_{\\beta } = (1+\\beta ^{2})\\cdot {\\frac {\\mathrm {precision} \\cdot \\mathrm {recall} }{(\\beta ^{2}\\cdot \\mathrm {precision} )+\\mathrm {recall} }} =\n",
    "\\frac {(1 + \\beta^2) \\cdot TP }{(1 + \\beta^2) \\cdot TP + \\beta^2 \\cdot FN + FP}.\n",
    "$$\n",
    "\n",
    "But in the end there is no absolute truth to what's best.\n",
    "\n",
    "\n",
    "Imagine you devise a new, cheap and easy cancer test.   \n",
    "What should you optimize your decision threshold for?\n",
    "Precision or recall?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:41:19.582828Z",
     "start_time": "2018-11-22T18:40:30.951327Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import cross_validate, StratifiedKFold\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.svm import SVC\n",
    "\n",
    "X = passengers[['sex_code', 'pclass', 'fare', 'age']]\n",
    "y = passengers['survived']\n",
    "\n",
    "classifiers = dict(\n",
    "    svc=SVC(kernel='linear'),\n",
    "    knn=KNeighborsClassifier(n_neighbors=5),\n",
    "    tree=DecisionTreeClassifier(max_depth=3),\n",
    ")\n",
    "\n",
    "results = []\n",
    "for name, clf in classifiers.items():\n",
    "    r = cross_validate(\n",
    "        clf, X=X, y=y, cv=StratifiedKFold(5, shuffle=True, random_state=1337),\n",
    "        scoring=['accuracy', 'precision', 'recall', 'f1'],\n",
    "        n_jobs=5,\n",
    "    )\n",
    "    df = pd.DataFrame.from_dict(r)\n",
    "    df['classifier'] = name\n",
    "    results.append(df)\n",
    "\n",
    "df = pd.concat(results).drop(['fit_time', 'score_time'], axis='columns')\n",
    "\n",
    "means = df.groupby('classifier').mean()\n",
    "stds = df.groupby('classifier').std()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:41:19.582828Z",
     "start_time": "2018-11-22T18:40:30.951327Z"
    }
   },
   "outputs": [],
   "source": [
    "fig, (ax1, ax2) = plt.subplots(1, 2)\n",
    "sns.heatmap(means, cmap='viridis', annot=True, ax=ax1, vmin=0, vmax=1)\n",
    "sns.heatmap(stds, cmap='viridis', annot=True, ax=ax2, vmin=0, vmax=1)\n",
    "\n",
    "ax1.set_title('mean')\n",
    "ax2.set_title('std')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Decision Thresholds and Classifier Validation\n",
    "\n",
    "Classifier objects usually provide all of the following methods\n",
    "\n",
    "* `classifier.fit(X)` takes training data and finds some parameters based on that data.\n",
    "* `classifier.predict(X_new)` takes new data (one row or many) and predicts the target label for each row.\n",
    "* `classifier.predict_proba(X_new)` takes new data (one row or many) and predicts 'some notion of confidence'.\n",
    "\n",
    "In the case of binary classification (i.e. two classes) the `classifier.predict_proba` usually returns a number where higher numbers indicate some measure of 'confidence'.\n",
    "\n",
    "The `classifier.predict(X_new)` is basically a wrapper around the `predict_proba` function which simply applies a decision threshold at some value (usually 0.5).\n",
    "This is exactly what we did in the case of linear least squares regression.\n",
    "\n",
    "\n",
    "$$\n",
    "\\hat{Y} = \\begin{cases}\n",
    "\\text{Yes}, & \\text{if $ f(X) \\gt 0.5$} \\\\\n",
    "\\text{No}, & \\text{if $ f(X) \\le 0.5$}\n",
    "\\end{cases}\n",
    "$$\n",
    "\n",
    "In this case the decision threshold corresponds to the distance of a point to the seperating hyperplane.\n",
    "\n",
    "\n",
    "<p style=\"padding: 2rem;  margin-top: 5rem; font-size: 2rem; line-height: 1.5; color: crimson; text-align: center\">\n",
    "Despite the name, <code>predict_proba</code> does not return a probability. <br>\n",
    "It is merely a measure of certainty of the classifier.\n",
    "</p>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Example: Fisher Discriminant\n",
    "\n",
    "We find a separating hyperplane to distinguish between two classes using a set of training points.\n",
    "We create two random blobs and use the LDA to transform the points to maximize class separation. \n",
    "\n",
    "What do we know about a new data point that we want to classify? What's the probability of the point belonging\n",
    "to either class?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:41:20.234965Z",
     "start_time": "2018-11-22T18:41:19.587898Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.datasets import make_blobs\n",
    "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis\n",
    "\n",
    "X, y = make_blobs(n_samples=1250, centers=2, random_state=12, cluster_std=1.4)\n",
    "\n",
    "\n",
    "fig, ax = plt.subplots()\n",
    "ax.scatter(X[:, 0], X[:, 1], c=y, s=10, cmap=cmap)\n",
    "\n",
    "clf = LinearDiscriminantAnalysis()\n",
    "lda = clf.fit(X, y)\n",
    "plots.draw_linear_regression_function(lda, color='gray', ax=ax)\n",
    "\n",
    "circle = plt.Circle((-2, 4), 0.15, fc='black')\n",
    "ax.add_patch(circle)\n",
    "ax.set_aspect(1)\n",
    "ax.annotate('Green vs Blue?', xy=(-2, 4), xytext=(2.5, 4), arrowprops=dict(facecolor='gray', shrink=0.05),)\n",
    "ax.set_axis_off()\n",
    "None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:41:20.684122Z",
     "start_time": "2018-11-22T18:41:20.237423Z"
    }
   },
   "outputs": [],
   "source": [
    "X_prime = lda.transform(X)\n",
    "\n",
    "hist_options = dict(bins=30, range=[-5, 5], histtype='step', lw=3)\n",
    "plt.figure()\n",
    "plt.hist(X_prime[y == 0], color=colors[0], **hist_options)\n",
    "plt.hist(X_prime[y == 1], color=colors[1], **hist_options)\n",
    "plt.axvline(0, lw=3, color='gray')\n",
    "None"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Obviously the distance to the hyperplane seems to correlate with the probability.\n",
    "In the case above, data created by two gaussian distributions, the distance certainly maps to the underlying \n",
    "probability density. In the case that the distributions are known, one can calculate actual probabilities for the \n",
    "new data points. In this artificial example, two gaussians with a separating plane, this is quite easy.\n",
    "\n",
    "If $lda$ is the function projecting a data point $x$ to the space of maximized separating hyperplanes and \n",
    "$\\mathcal{N}_{\\mathrm{target}}$ the distribution of the points of the blue or red class, then\n",
    "\n",
    "$$\n",
    "p = \\mathcal{N}(\\operatorname{lda}(x); \\mu, \\sigma)\n",
    "$$\n",
    "\n",
    "for known or estimated $\\mu, \\sigma$.\n",
    "\n",
    "\n",
    "\n",
    "In *most* cases this will not work\n",
    "\n",
    "While many, if not most, classifiers can return numbers between 0 and 1, by no means do all classifiers return the \n",
    "desired probability estimate.\n",
    "\n",
    "The function `predict_proba` is a slight misnomer. While the numbers indicate some notion of 'certainty of \n",
    "belonging to one class or the other' it is in no way an actual *significance*, *confidence* or probability.\n",
    "\n",
    "Still the number can be interpreted as some level of 'certainty' in many cases. And in many applications you will \n",
    "find that the number can be interpreted as a probability. Just not the one you might think of.\n",
    "\n",
    "\n",
    "### Example: Decision Tree\n",
    "\n",
    "Below we train a simple Decision Tree and draw its decision boundary. How can we define the *certainty* of the \n",
    "prediction in this case? Especially when the underlying distributions are not known."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:41:21.400198Z",
     "start_time": "2018-11-22T18:41:20.688230Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.datasets import make_moons\n",
    "\n",
    "X, y = make_moons(n_samples=2200, noise=0.30, random_state=0)\n",
    "clf = DecisionTreeClassifier(max_depth=4, criterion='entropy')\n",
    "clf.fit(X, y)\n",
    "\n",
    "plt.figure()\n",
    "plt.axes(aspect=1)\n",
    "plt.scatter(X[:, 0], X[:, 1], c=y, s=50, cmap=cmap)\n",
    "plots.draw_decision_boundaries(clf, cmap=cmap, alpha=0.5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "While building the tree the parameter space of the training sample gets split into binary subspaces.\n",
    "We can color each node of the tree according to the amount of examples left of either class in the subspace \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:41:21.668672Z",
     "start_time": "2018-11-22T18:41:21.404451Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn import tree\n",
    "\n",
    "def plot_tree(crit, depth=1, min_split=0.5, min_leaf=0.2):\n",
    "    estimator = DecisionTreeClassifier(\n",
    "        random_state=0,\n",
    "        criterion=crit,\n",
    "        splitter='best',\n",
    "        max_depth=depth,\n",
    "        min_samples_split=min_split,\n",
    "        min_samples_leaf=min_leaf,\n",
    "    )\n",
    "    \n",
    "    estimator.fit(X, y)\n",
    "\n",
    "    graph = Source(tree.export_graphviz(estimator, out_file=None, feature_names=['Red', 'Blue'], class_names=['0', '1',], filled = True))\n",
    "\n",
    "    display(SVG(graph.pipe(format='svg')))\n",
    "\n",
    "    return estimator\n",
    "\n",
    "inter=interactive(\n",
    "    plot_tree,\n",
    "    crit=[\"gini\", \"entropy\"], \n",
    "    depth=(1, 20),\n",
    "    min_split=(0.1,1, 0.01),\n",
    "    min_leaf=(0.01, 0.50, 0.005),\n",
    ")\n",
    "\n",
    "display(inter)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The *certainty* of belonging to a specific class can be simply defined as the number of samples belonging to the class divided by the total number of samples in the node."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T16:49:54.101678Z",
     "start_time": "2018-11-22T16:49:54.070357Z"
    }
   },
   "source": [
    "#### Varying the Threshold\n",
    "\n",
    "Varying the decision threshold is extremely useful for modifying your classifier output to create more/less \n",
    "'conservative' predictions.\n",
    "\n",
    "\n",
    "In essence this is a new classifier/predictor with a free parameter. The old $\\hat{y} = \\hat{f}(x)$ now becomes\n",
    "\n",
    "$$\n",
    "\\hat{y} = \\hat{f}(x, \\alpha)\n",
    "$$\n",
    "\n",
    "where $\\alpha$ is a parameter which can be chosen freely (or optimized according to some criterion which has \n",
    "nothing to do with the underlying loss function of the predictor.)\n",
    "\n",
    "The example below shows how the number of true positives can be influenced by varying the threshold $\\alpha$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:41:23.170409Z",
     "start_time": "2018-11-22T18:41:21.673079Z"
    }
   },
   "outputs": [],
   "source": [
    "threshold = np.linspace(0, 1, 50)\n",
    "\n",
    "N = 10000\n",
    "\n",
    "X, y = datasets.make_moons(n_samples=N, noise=0.3)\n",
    "X_test, y_test = datasets.make_moons(n_samples=N, noise=0.3)\n",
    "\n",
    "\n",
    "clf = DecisionTreeClassifier(max_depth=5, criterion='entropy')\n",
    "clf.fit(X, y)\n",
    "y_predicted = clf.predict_proba(X_test)[:, 1]\n",
    "\n",
    "tp = [(y_predicted > t).sum()/N for t in threshold]\n",
    "acc = [accuracy_score(y_test, y_predicted > t) for t in threshold]\n",
    "precision = [precision_score(y_test, y_predicted > t, zero_division=1) for t in threshold]\n",
    "\n",
    "plt.figure()\n",
    "plt.plot(threshold, tp, label='Decision Tree TPR', color='C1')\n",
    "plt.plot(threshold, acc, label='Decision Tree Accuracy', color='C1', ls='--')\n",
    "plt.plot(threshold, precision, label='Decision Tree Precision', color='C1', ls=':')\n",
    "\n",
    "clf = SVC(probability=True)\n",
    "clf.fit(X, y)\n",
    "y_predicted = clf.predict_proba(X_test)[:, 1]\n",
    "\n",
    "tp = [(y_predicted > t).sum() / N for t in threshold]\n",
    "acc = [accuracy_score(y_test, y_predicted > t) for t in threshold]\n",
    "precision = [precision_score(y_test, y_predicted > t, zero_division=1) for t in threshold]\n",
    "plt.plot(threshold, tp, label='SVM TPR', color='C2')\n",
    "plt.plot(threshold, acc, label='SVM Accuracy', color='C2', ls='--')\n",
    "plt.plot(threshold, precision, label='SVM Precision', color='C2', ls=':')\n",
    "plt.legend()\n",
    "plt.ylabel('true positives & accuracy')\n",
    "None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:41:24.700943Z",
     "start_time": "2018-11-22T18:41:23.173982Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.datasets import make_moons\n",
    "\n",
    "X, y = make_moons(n_samples=1500, noise=0.6)\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y)\n",
    "\n",
    "clf = LinearDiscriminantAnalysis() #probability=True)\n",
    "clf.fit(X_train, y_train)\n",
    "\n",
    "predictions = clf.predict_proba(X_test)[:,1]\n",
    "\n",
    "plt.figure()\n",
    "plt.scatter(X[:, 0], X[:, 1], c=y, s=10, cmap='winter_r', vmin=0, vmax=1)\n",
    "plots.draw_decision_surface(clf, predictions, label=r'$ \\alpha $')\n",
    "\n",
    "None"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Classifier Calibration\n",
    "\n",
    "There are ways to transform the output of a classifiers into more reasonable probability estimates. \n",
    "\n",
    "This process is often called classifier calibration. There is a detailed guide in sklearn's documentation\n",
    "\n",
    "http://scikit-learn.org/stable/modules/calibration.html\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Receiver Operating Characteristic \n",
    "\n",
    "The decision threshold is incredibly helpful in validating classifier performance.\n",
    "\n",
    "The plot of false positive rate vs. true positive rate while varying the decision threshold is called the Receiver Operating Characteristic curve (ROC curve).\n",
    "\n",
    "It is a very popular tool for classifier performance evaluation.\n",
    "\n",
    "Wikipedia Quote:\n",
    "\n",
    "> Following the attack on Pearl Harbor in 1941, the United States army began new research to increase the prediction of correctly detected Japanese aircraft from their radar signals. For this purposes they measured the ability of radar receiver operators to make these important distinctions, which was called the Receiver Operating Characteristics.\n",
    "\n",
    "\n",
    "A classifier which assigns random labels to the data will have a ROC curve which lies on the diagonal. With an area under curve (AUC) of 0.5.\n",
    "\n",
    "\n",
    "__Problems__\n",
    "\n",
    "There are circumstance in which the ROC is not a good measure of quality. \n",
    "\n",
    "See https://en.wikipedia.org/wiki/Receiver_operating_characteristic#Area_under_the_curve for a discussion.\n",
    "\n",
    "Also in the case of imbalanced classes (eg. imagine you had 10000 surviors but 10 deceased passengers) the ROC curve itself won't change. The intepretation of the ROC curve however changes drastically.\n",
    "\n",
    "In that case its better to plot the precission vs recall curve and the corresponding area under curve.\n",
    "\n",
    "See https://classeval.wordpress.com for some very good discussions on classifier evaluation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:41:33.181397Z",
     "start_time": "2018-11-22T18:41:24.706457Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.datasets import make_classification\n",
    "from sklearn.metrics import roc_curve\n",
    "\n",
    "X, y = make_classification(n_samples=10000, n_features=10, n_informative=2, random_state=0)\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y)\n",
    "\n",
    "prediction = SVC(probability=True).fit(X_train, y_train).predict_proba(X_test)[:, 1]\n",
    "\n",
    "fpr, tpr, t = roc_curve(y_test, prediction)\n",
    "\n",
    "plt.figure()\n",
    "plt.gca().set_aspect(1)\n",
    "plt.scatter(fpr, tpr, c=t, cmap='viridis', s=50)\n",
    "plt.colorbar(label='Decision Threshold')\n",
    "plt.xlim([0, 1])\n",
    "plt.ylim([0, 1.02])\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:41:33.683670Z",
     "start_time": "2018-11-22T18:41:33.183884Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.datasets import make_classification\n",
    "from sklearn.metrics import roc_curve\n",
    "\n",
    "rng = np.random.default_rng(0)\n",
    "\n",
    "X, y = make_classification(n_samples=1000)\n",
    "\n",
    "prediction = rng.uniform(0, 1, len(y))\n",
    "\n",
    "fpr, tpr, t = roc_curve(y, prediction)\n",
    "\n",
    "plt.figure()\n",
    "plt.gca().set_aspect(1)\n",
    "plt.scatter(fpr, tpr, c=t, cmap='viridis')\n",
    "plt.xlim([0, 1])\n",
    "plt.ylim([0, 1])\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Ensemble Methods\n",
    "\n",
    "We have used a decision tree to classifiy artificial data as well as the Titanic data. \n",
    "\n",
    "Theoretically a decision tree is not limited in its depth. \n",
    "\n",
    "This quickly leads to overfitted tree models.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:41:34.775716Z",
     "start_time": "2018-11-22T18:41:33.687711Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.datasets import make_moons\n",
    "\n",
    "X, y = make_moons(n_samples=5000, noise=0.30, random_state=0)\n",
    "clf = DecisionTreeClassifier(max_depth=300, criterion='entropy')\n",
    "clf.fit(X, y)\n",
    "\n",
    "plt.figure()\n",
    "plt.scatter(X[:, 0], X[:, 1], c=y, s=3, cmap='winter')\n",
    "plots.draw_decision_boundaries(clf, alpha=0.5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As mentioned in the discussion about decision trees earlier, the tree building algorithms try to find the optimal split criterion in some local region of the parameter space.\n",
    "\n",
    "Finding the best overal split in parameter space is computationaly infeasible.\n",
    "\n",
    "This means the decision tree algorithm can run into a local optimum. \n",
    "\n",
    "The idea of _ensemble learning_ is to train several weak (high bias, low variance) base classifiers on different subsets of the data and then combine them into one big classifier."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Bagging\n",
    "\n",
    "A popular way to build ensembles is called *bagging*.\n",
    "\n",
    "Split the training data into $B$ subsets using sampling with replacement (Bootstrapping). For each subset $b$ we train a classifier $\\hat{f}_b$. Bagging then combines the overall prediction by taking the average.  \n",
    "\n",
    "$$\n",
    "\\hat{y} = \\hat{f}_{\\text{bag}}(x) = \\frac{1}{B} \\sum_{b=1}^B \\hat{f}_b (x) \n",
    "$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Random Forests\n",
    "\n",
    "Random Forests are a modification to bagging in which a number of *randomized decision trees* are trained. These randomized decision trees use a random subset of variables to find the best split in each node.\n",
    "\n",
    "```python\n",
    "def build_random_tree(space)\n",
    "    if stopping_criterion_fulfilled():\n",
    "        return {}\n",
    "\n",
    "    random_variable_choice = choose_random_selection_of_variables()\n",
    "    variable, split_point = find_best_split(space, random_variable_choice)\n",
    "\n",
    "    left, right = split_space(space, variable, split_point)\n",
    "\n",
    "    left_tree = build_tree(left)\n",
    "    right_tree = build_tree(right)\n",
    "\n",
    "    return {'node' : (variable, split_point), 'left': left_tree, 'right': right_tree}\n",
    "```\n",
    "Random Forests are a very popular choice for classification tasks since their parameters can be easily tuned and they often outperform other methods."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:41:37.298742Z",
     "start_time": "2018-11-22T18:41:34.781672Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.metrics import roc_curve, roc_auc_score, make_scorer\n",
    "from sklearn.model_selection import cross_validate, StratifiedKFold\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "data = read_titanic()\n",
    "\n",
    "X = passengers[['sex_code', 'pclass', 'fare', 'age']].to_numpy()\n",
    "y = passengers['survived']\n",
    "\n",
    "tree = DecisionTreeClassifier(min_samples_leaf=5)\n",
    "\n",
    "rf = RandomForestClassifier(n_estimators=100, min_samples_leaf=5, random_state=0)\n",
    "\n",
    "score = cross_validate(tree, X, y, scoring=make_scorer(roc_auc_score), cv=StratifiedKFold(5, shuffle=True, random_state=0))\n",
    "print('ROC AUC Decision Tree   {:0.3f} ± {:0.3f}'.format(score['test_score'].mean(), score['test_score'].std()))\n",
    "score = cross_validate(tree, X, y, scoring=make_scorer(accuracy_score), cv=StratifiedKFold(5, shuffle=True, random_state=0))\n",
    "print('Accuracy Decision Tree  {:0.3f} ± {:0.3f}'.format(score['test_score'].mean(), score['test_score'].std()))\n",
    "\n",
    "score = cross_validate(rf, X, y, scoring=make_scorer(roc_auc_score), cv=StratifiedKFold(5, shuffle=True, random_state=0))\n",
    "print('ROC AUC Random Forest   {:0.3f} ± {:0.3f}'.format(score['test_score'].mean(), score['test_score'].std()))\n",
    "score = cross_validate(rf, X, y, scoring=make_scorer(accuracy_score), cv=StratifiedKFold(5, shuffle=True, random_state=0))\n",
    "print('Accuracy Random Forests {:0.3f} ± {:0.3f}'.format(score['test_score'].mean(), score['test_score'].std()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "rf.fit(X, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "np.mean([tree.predict(X) for tree in rf.estimators_], axis=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Regression\n",
    "\n",
    "Regression and classification are very similar in nature. The biggest difference being that the target variable $y$ is continous and has a natural ordering associated with it. \n",
    "\n",
    "The same basic rules for classification apply for regression as well. \n",
    "\n",
    "* Models need to be verified on independent test data\n",
    "* There is a tradeoff between bias and variance. Overfitting can occur.\n",
    "* There are many quality measures to pick from. \n",
    "\n",
    "The loss functions we are minimizing need to be adapted. \n",
    "For classification the loss function for the decision tree was based on entropy. The one for the Naive Bayes \n",
    "classifier was the 0-1 loss. Both are designed to improve classification. \n",
    "\n",
    "To make a regressor out of a decision tree we can change the loss function to minimize the variance of the \n",
    "target variable in the node. \n",
    "\n",
    "To find the best split, we minimize the variance of the target , $\\mathit{Var(y)}$,  over all features  $X \\in \n",
    "\\mathbf{X}$ and splits $s$\n",
    "\n",
    "The algorithm stays the same we simply changed the way to select the features in each step according to \n",
    "\n",
    "$$\n",
    "  \\min_{(X, s) \\in \\, \\mathbf{X} \\times {S}}Var(Y_S)\n",
    "$$\n",
    "\n",
    "where $Y_S$ are the values of the target variable in the subspace created by splitting the data on the feature $X$ at a value $s$ \n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:28:51.658023Z",
     "start_time": "2018-11-22T18:28:51.622417Z"
    }
   },
   "source": [
    "### Example: Boston Housing\n",
    "\n",
    "A typical example of a regression task. Given some information about housing in the larger Boston area try and predict the price of the object.\n",
    "\n",
    "There are 14 attributes in each case of the dataset. They are:\n",
    "\n",
    "- CRIM - per capita crime rate by town\n",
    "- ZN - proportion of residential land zoned for lots over 25,000 sq.ft.\n",
    "- INDUS - proportion of non-retail business acres per town.\n",
    "- CHAS - Charles River dummy variable (1 if tract bounds river; 0 otherwise)\n",
    "- NOX - nitric oxides concentration (parts per 10 million)\n",
    "- RM - average number of rooms per dwelling\n",
    "- AGE - proportion of owner-occupied units built prior to 1940\n",
    "- DIS - weighted distances to five Boston employment centres\n",
    "- RAD - index of accessibility to radial highways\n",
    "- TAX - full-value property-tax rate per \\$10,000\n",
    "- PTRATIO - pupil-teacher ratio by town\n",
    "- B - $1000(Bk - 0.63)^2$ where Bk is the proportion of blacks by town\n",
    "- LSTAT - percent lower status of the population\n",
    "- MEDV - Median value of owner-occupied homes in \\$1000's\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Our regression task is to predict the price of a house from the 13 given variables. From a plot of some of the variables we can see some weak correlations in some variables.\n",
    "\n",
    "None of these variables alone would suffice to build a good predictor. Hence we try to use multivariate regression methods."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:41:37.324296Z",
     "start_time": "2018-11-22T18:41:37.302090Z"
    }
   },
   "outputs": [],
   "source": [
    "try:\n",
    "    from sklearn.datasets import load_boston\n",
    "except ImportError as e:\n",
    "    print(e)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:41:37.324296Z",
     "start_time": "2018-11-22T18:41:37.302090Z"
    }
   },
   "outputs": [],
   "source": [
    "names = [\n",
    "    \"CRIM\", \"ZN\", \"INDUS\", \"CHAS\", \"NOX\",\n",
    "    \"RM\", \"AGE\", \"DIS\", \"RAD\", \"TAX\", \"PTRATIO\",\n",
    "    \"B\", \"LSTAT\", \"price\"\n",
    "]\n",
    "\n",
    "\n",
    "data_url = \"http://lib.stat.cmu.edu/datasets/boston\"\n",
    "raw_df = pd.read_csv(data_url, sep=\"\\s+\", skiprows=22, header=None)\n",
    "houses = np.hstack([raw_df.values[::2, :], raw_df.values[1::2, :3]])\n",
    "\n",
    "\n",
    "data = pd.DataFrame(houses, columns=names)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:41:40.561938Z",
     "start_time": "2018-11-22T18:41:37.328310Z"
    }
   },
   "outputs": [],
   "source": [
    "sns.pairplot(data[['RM', 'NOX', 'B', 'price']])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:41:42.530980Z",
     "start_time": "2018-11-22T18:41:41.522108Z"
    }
   },
   "outputs": [],
   "source": [
    "sns.jointplot(data, x='RM', y='price', kind='scatter')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The question this data set can help us answer is:\n",
    "\n",
    "> What is the price of a house based on the given features?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:41:40.945743Z",
     "start_time": "2018-11-22T18:41:40.565078Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.svm import SVR\n",
    "from sklearn.tree import DecisionTreeRegressor\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.model_selection import cross_validate, KFold\n",
    "\n",
    "X = data.drop([\"price\"], axis=\"columns\").to_numpy()\n",
    "y = data['price'].to_numpy()\n",
    "\n",
    "cv = KFold(5, shuffle=True, random_state=0)\n",
    "\n",
    "reg = LinearRegression()\n",
    "score = cross_validate(reg, X, y, cv=cv, scoring='neg_mean_absolute_error')\n",
    "print('Score linear regression {:0.3f} +/- {:0.3f}'.format(score['test_score'].mean(), score['test_score'].std()))\n",
    "\n",
    "reg = SVR(kernel='rbf')\n",
    "score = cross_validate(reg, X, y, cv=cv, scoring='neg_mean_absolute_error')\n",
    "print('Score SVR {:0.3f} +/- {:0.3f}'.format(score['test_score'].mean(), score['test_score'].std()))\n",
    "\n",
    "reg = DecisionTreeRegressor()\n",
    "score = cross_validate(reg, X, y, cv=cv, scoring='neg_mean_absolute_error')\n",
    "print('Score Tree {:0.3f} +/- {:0.3f}'.format(score['test_score'].mean(), score['test_score'].std()))\n",
    "\n",
    "reg = RandomForestRegressor()\n",
    "score = cross_validate(reg, X, y, cv=cv, scoring='neg_mean_absolute_error')\n",
    "print('Score RandomForestRegressor {:0.3f} +/- {:0.3f}'.format(score['test_score'].mean(), score['test_score'].std()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2018-11-22T18:41:41.516386Z",
     "start_time": "2018-11-22T18:41:40.949642Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor\n",
    "\n",
    "reg = RandomForestRegressor()\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.5)\n",
    "\n",
    "reg.fit(X_train, y_train)\n",
    "prediction = reg.predict(X_test)\n",
    "\n",
    "bin_edges = np.linspace(0, 60, 30)\n",
    "\n",
    "plt.figure()\n",
    "plt.hist2d(prediction, y_test, bins=bin_edges, cmap='gray_r',)\n",
    "plt.colorbar()\n",
    "plt.grid()\n",
    "plt.plot([0, 60], [0, 60], color='gray')\n",
    "plt.xlabel('Prediction')\n",
    "plt.ylabel('Truth')\n",
    "plt.gca().set_aspect(1)\n",
    "None"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Regression is not Curve Fitting\n",
    "\n",
    "The output of the predictor shows stronger correlation than any of the other variables. It is still noisy and shows some smearing and outliers. The output of a regressor should never be understood as 'true' but rather as a new correlated variable which can show noise and bias. \n",
    "\n",
    "\n",
    "There is a fundamental difference between *curve fitting* and regression.\n",
    "\n",
    "* __Curve Fitting__ All data is available. Some known (or presumed) analytical function is fit to the data to estimate free parameters of that function.\n",
    "\n",
    "\n",
    "* __Regression__ Training data is available. A model is fitted on training data to predict the dependent variable on some new, unknown, data.\n",
    "\n",
    "There is a lot more to learn about linar models and regression. Check sklearn's user guide on linear models for more information \n",
    "http://scikit-learn.org/stable/modules/linear_model.html\n",
    "\n",
    "A very popular regression metric is the $R^2$ score. Read about it here\n",
    "\n",
    "https://en.wikipedia.org/wiki/Coefficient_of_determination\n",
    "\n",
    "and here\n",
    "\n",
    "http://scikit-learn.org/stable/modules/generated/sklearn.metrics.r2_score.html .\n",
    "\n",
    "\n",
    "\n",
    "## Example: Curve Fitting using Decision Trees"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "rng = np.random.default_rng(0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "x = np.linspace(0, 2 * np.pi, 1000)\n",
    "X = x[:, np.newaxis]\n",
    "y = np.sin(x) + rng.normal(0, 0.2, len(x))\n",
    "\n",
    "x_test = np.linspace(-1, 2 * np.pi + 1, 1000)\n",
    "X_test = x_test[:, np.newaxis]\n",
    "\n",
    "\n",
    "rf = RandomForestRegressor(n_estimators=100, min_samples_split=40)\n",
    "rf.fit(X, y)\n",
    "\n",
    "tree = DecisionTreeRegressor(min_samples_split=40)\n",
    "tree.fit(X, y)\n",
    "\n",
    "plt.figure()\n",
    "plt.plot(x, y, 'k.', label='training data')\n",
    "plt.plot(x_test, rf.predict(X_test), '.', label='RandomForest prediction')\n",
    "plt.plot(x_test, tree.predict(X_test), '.', label='Tree prediction')\n",
    "plt.plot(x_test, np.sin(x_test), label='truth')\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# More Material\n",
    "\n",
    "Read scikit-learns's user guide. It's detailed and describes pros and cons of many alogirthms and evaluation criteria.\n",
    "Its also full of code examples.\n",
    "\n",
    "http://scikit-learn.org/stable/user_guide.html\n",
    "\n",
    "Read the book by Hastie (if you're a crazy maths person)\n",
    "\n",
    "http://web.stanford.edu/~hastie/ElemStatLearn/\n",
    "\n",
    "Read this book by James and Hastie (if you're a normal person)\n",
    "\n",
    "http://www-bcf.usc.edu/~gareth/ISL/"
   ]
  }
 ],
 "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": "3.12.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}