Movie recommendation by Apriori algorithm.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "##APRIRORI ALGORITHM BY Srinath Sai Tripuraneni, A20338902."
      ],
      "metadata": {
        "id": "SFWMtCAZq9-h"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "OpwFXACccol_",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "59db694a-4fce-4ef2-e6ab-f0ba401ef607"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
            "Requirement already satisfied: pyspark in /usr/local/lib/python3.7/dist-packages (3.3.0)\n",
            "Requirement already satisfied: py4j==0.10.9.5 in /usr/local/lib/python3.7/dist-packages (from pyspark) (0.10.9.5)\n"
          ]
        }
      ],
      "source": [
        "!pip install pyspark"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import sys\n",
        "from pyspark import SparkContext, SparkConf"
      ],
      "metadata": {
        "id": "SIieHIp_nlT2"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# creating Spark context with necessary configuration\n",
        "sc = SparkContext(\"local\",\"PySpark - CS5683 - assignment-2\")"
      ],
      "metadata": {
        "id": "VsKbcz6jopi8"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "file1='/content/views.txt'"
      ],
      "metadata": {
        "id": "HWJlTjnbox7X"
      },
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "fileRDD = sc.textFile(file1) #converting textfile to rdd by using spark variable.\n"
      ],
      "metadata": {
        "id": "9FgNX9iIp8ZE"
      },
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#maindata=fileRDD.map(lambda x: list(x.split(\"::\"))) \n",
        "#maindata=fileRDD.map(lambda x:set(x.split(\"::\"\")))"
      ],
      "metadata": {
        "id": "-2cDDIk2r1bh"
      },
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "***( We need list of lines stored in a datastructure for checking frequency of combinations. Initially tried in list and set. Generally 2,3 and 4 combinations must also be in set or list to check frequency satisfying support=4 . That implies combinations in list or combinations in set must be in key. But getting error as list or set can't be hashed and that is reasonable because they are muttable. Therefore found this datastructure frozenset which is immutable)***\n",
        "\n",
        "reffered link :- https://www.programiz.com/python-programming/methods/built-in/frozenset\n"
      ],
      "metadata": {
        "id": "IHoKWOB78Rnn"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "maindata=fileRDD.map(lambda x: frozenset(x.split(\"::\"))) #as we need one main datapoint to check frequency of all combinations, creating list of frozenset \n",
        "# due to above reason , for every line by making ids as elements."
      ],
      "metadata": {
        "id": "zKK00YbslIVh"
      },
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#maindata.take(5)"
      ],
      "metadata": {
        "id": "pwEAr79dM5E6"
      },
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "maindata.persist() # as we need one main data point for checking it is persisted all over."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "o0IhQcDOoIM6",
        "outputId": "b3f843de-411a-45a9-e5ba-5cbd9a81fb4a"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "PythonRDD[2] at RDD at PythonRDD.scala:53"
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#Generating set of single movie ids whose frequency is greater than or equal to 4.(pass-1)"
      ],
      "metadata": {
        "id": "kVHQsGsQoz84"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "movieIDRDD = fileRDD.flatMap(lambda line:line.split('::')) # Similar to word count problem \n",
        "movieIDRDD=movieIDRDD.map(lambda x: frozenset((x,))) # converting key which is one movie idd into frozen set because of 2 reasons, they are one is immutable and other is from single movie idd\n",
        "                                                    # 2,3 and 4 sized sets generated from this also must also be frozenset.\n",
        "movieIDRDD=movieIDRDD.map(lambda x: (x , 1))\n",
        "movieIDRDD=movieIDRDD.reduceByKey(lambda a,b: a+b)\n",
        "movieIDRDD=movieIDRDD.filter(lambda a: a[1]>=4) # movie idd with single movies that have minimum support value.\n"
      ],
      "metadata": {
        "id": "oBX5sgC-o8XR"
      },
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#movieIDRDD.take(5) "
      ],
      "metadata": {
        "id": "IcyN0gQ2GEpd"
      },
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "##movieIDRDD.take(5)\n",
        "fileRDD.unpersist() # there is no use of fileRDD implies unpersisting it,can clear if there is no use."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "VK_5eyCkrwSS",
        "outputId": "a7bba1de-9b80-4b3b-846d-5604763da5d3"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "/content/views.txt MapPartitionsRDD[1] at textFile at NativeMethodAccessorImpl.java:0"
            ]
          },
          "metadata": {},
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#Generating frequent itemsets of sizes 2,3 and 4."
      ],
      "metadata": {
        "id": "nuSXh0p8ph9w"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#We need a function which generates allcombinations of current itemset size from previous itemset size of frequent combinations . So below is the function that returns list of lists of all combinations of current itemset size.\n",
        "\n",
        "def allcombinations(x,n,temp=[]):\n",
        "    if len(temp) == n: #base case to stop recurrsion.\n",
        "        return [temp]\n",
        "    ans = []\n",
        "    i=0\n",
        "    while(i<len(x)):      #logic is same as backtracking which we use for or while loop reccursion here, where reccursively it adds element after element to a list untill size becomes current itemset size. \n",
        "        newtemp = temp.copy()       #so whatever we are adding that will be passed as well \n",
        "        newtemp.append(x[i])                                                                   \n",
        "        ans=ans+allcombinations(x[i+1:], n,newtemp)     #taking elements until n sized as subsets.\n",
        "        i+=1\n",
        "    return ans"
      ],
      "metadata": {
        "id": "krBzG-j-AYos"
      },
      "execution_count": 13,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def issubset(subset,superset):\n",
        "  if(subset<=superset):\n",
        "    return True          #created a manual boolean function to check one set is subset or other. And this function is used below.\n",
        "  return False"
      ],
      "metadata": {
        "id": "VmJ0U-AGa7YJ"
      },
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "***(For apriori algorithm we need to check in whole data for filtering combinations which satisfy minimum support value. To reduce computations below, I designed a function which makes all combinations from line to line  and these are filtered in main reccurent function ).***"
      ],
      "metadata": {
        "id": "nBtMz9EBR57C"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def generate_combinations(x,broadcast_preceed,i):\n",
        "  cache=list(broadcast_preceed.value) # taking list of current frequent itemsize broadcast values for loop.\n",
        "  m=0;\n",
        "  combos_possibility=frozenset() #since every element in cache list will be frozenset. \n",
        "  while(m <len(cache)):# for every element in a list finding all combinations with line.\n",
        "    if issubset(cache[m],x): \n",
        "      combos_possibility=combos_possibility.union((cache[m])) # the current element of list will be added to set if it is present in one line.\n",
        "    m=m+1\n",
        "  x1=list(combos_possibility)\n",
        "  return(allcombinations(x1,i)) # all combinations of sets of current size are formed by bactracking logic of all combinations function.\n"
      ],
      "metadata": {
        "id": "ZYneNGMSQuNU"
      },
      "execution_count": 15,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "***(below function is main function if you call it by sending single frequent movieidd rdds with set size 2 it reccursively gives frequent item set of 2,3 and 4. )***\n",
        "**(THE RECCURSION STRUCTURE IS INSPIRED FROM A LEETCODE PROBLEM THAT I HAVE SOLVED USING RECCURSION TWO MONTHS AGO. PROBLEM LINK**:- https://leetcode.com/problems/binary-search/) \n",
        "\n",
        "*** My code for that problem that passed all cases:-***\n",
        "![Screenshot 2022-09-25 011042.jpg](data:image/jpeg;base64,/9j/4AAQSkZJRgABAQEAYABgAAD/2wBDAAMCAgMCAgMDAwMEAwMEBQgFBQQEBQoHBwYIDAoMDAsKCwsNDhIQDQ4RDgsLEBYQERMUFRUVDA8XGBYUGBIUFRT/2wBDAQMEBAUEBQkFBQkUDQsNFBQUFBQUFBQUFBQUFBQUFBQUFBQUFBQUFBQUFBQUFBQUFBQUFBQUFBQUFBQUFBQUFBT/wAARCAFlAtIDASIAAhEBAxEB/8QAHwAAAQUBAQEBAQEAAAAAAAAAAAECAwQFBgcICQoL/8QAtRAAAgEDAwIEAwUFBAQAAAF9AQIDAAQRBRIhMUEGE1FhByJxFDKBkaEII0KxwRVS0fAkM2JyggkKFhcYGRolJicoKSo0NTY3ODk6Q0RFRkdISUpTVFVWV1hZWmNkZWZnaGlqc3R1dnd4eXqDhIWGh4iJipKTlJWWl5iZmqKjpKWmp6ipqrKztLW2t7i5usLDxMXGx8jJytLT1NXW19jZ2uHi4+Tl5ufo6erx8vP09fb3+Pn6/8QAHwEAAwEBAQEBAQEBAQAAAAAAAAECAwQFBgcICQoL/8QAtREAAgECBAQDBAcFBAQAAQJ3AAECAxEEBSExBhJBUQdhcRMiMoEIFEKRobHBCSMzUvAVYnLRChYkNOEl8RcYGRomJygpKjU2Nzg5OkNERUZHSElKU1RVVldYWVpjZGVmZ2hpanN0dXZ3eHl6goOEhYaHiImKkpOUlZaXmJmaoqOkpaanqKmqsrO0tba3uLm6wsPExcbHyMnK0tPU1dbX2Nna4uPk5ebn6Onq8vP09fb3+Pn6/9oADAMBAAIRAxEAPwD9U6K5vxT8QtE8H3VtaahJeTXtwjSx2emadc39wY1IBkMVvG7qgJA3EBckDOTWd4g8Q65o/wARPCllFPp76BrDTW8lu9nILqORIJZd4m83btOxRsMWep3dh2U8JVmk2rJptNp2fKruztrsZyqRjv5fidrRXOT+P9ItfFEfh+cajBfyuI45ZdKultHYpvCrdGLySxAPAfORjrxRpPj7R9c1y70qy/tCWe1MiyXDaXdJZ7kba6rdNGIXYNkEK5OVYfwnEfVa6jzOm7Wvs9u/p5lc0b2udHRXPeFPHmk+NfPOlDUHihCt9outLurWGVWztaKSWNUlU4zujLDBB6EZT4geMI/AfhG/1yWD7QlqEG1pPLjXc6oHlkwfLiXdud8Hais2DjBPq1b2yw7g1NtKz0d3tuNNPY6KiuJsfiMtj4RtNc8SjTbZb6RRYp4bu59ZW9Rl3IYdlukkrFQzbUjbCqWzgHGb46+I13/wrG98W+Db+xZdP8ySeLVdOnZmEZKyQNGZIXglDDB3glcEFPTohl+IlUUHG15ct/s37XV0ZyqxjHn8rnpFFZV94o0zTb6ezubny7mCzbUJE8tjtgU4Z8gY69uvtXF6l8dNG07xtpmgDT9au4r7Tm1BL6z0W+uExmHYFEcDBlKzZZw2EKhWwWFZ0cFicQ7Uqbel9F0WpU6kaavJ2/pf5r7z0miuLvPG1wvxW0vwxA8EdtJp9zdXKXVjdrLKytCENvPs8h1XzDvXcWBZOOtT2PxU8M6l4iGi299M9200ltHMbKdbSaaPPmQx3RQQySLtcFEcsDG4IyjYHg8RyqSg2mubRPRa6vy032DmSvf+tLnW0Vx3g34ueFfH15HbaHqE128tubqF5LG4giniUqrtFJJGqybGdVcISUY4bB4rsGO0E1hWo1cNJwrxcX2aaf4lJp7C0Vw/wb8S67428CWPiTW59NkGsAX9hBpttJEttaSKGiid3lfznAOTKojVsjCLjnuKyatoyno7BRRRSEFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFAHF+JfBmtTeKR4i8Na1Y6RqUtkthdLqmmvfQTRI7SRlVSeFkdWkk53EEPyOARW8aeCfFPiDxdoGr6X4i0jTbXR5GmjtbvRZbl5JHieJ8yLdRjbtc4G3IIySRxXe0V3wx1em4tNXSa1jF6NWs7rXTRX2W1jOVOM783W3V9NvyPHbj9n0XfxMPiyfUNKlddWTVop30RW1MFUVBbm9aU/uAoOEWNSPlyThg2/4d+GOo6L48vvEB1PS7W1uGnk+waLps1ktxJIRiW7H2l47iQAD955aOSB8wX5a9DqC+uksLOe5kDGOGNpGC9SAMnH5V0TzXG1oqnKd1y8uy27bf8ABD2UXK6Wrd/ne/6nC/DP4Z33gXUL64uNR037NPGscWl6Bp02nWEJ3FmkFu9zMiuxPLRhM5JbccEdhr0Oq3Glyx6Je2en6iSvl3F/ZvdQqMjdmNJYicjIHzjBOeelcJa/HjSptDvtYuNC1zTtPh0WTX7WW7hhU39pGqs7RKJSysN6fLKIyd4xkZIT4pfFE+HtO1W2003VvqenHR7uWVLcTBre6vxCURRuZnKxyggLn5l2knp01MNmGKxcZVo++2ldpWvdLVJW0bSas+zV7ip8jjeL0f8AwP8ANfeVbD4L6lpcT6hY67ptj4lbV21gS2ui+XpnmtA0EmbQT7yXR2Zm87cZMNnHyVPrnwl1zUvh1qfhy08S2MGoazPNcarqVzpLzJI0vLCGJbhPKAwoXcz8LzuJLVLr3jO/8UfD/wAWzWH9ueBdd0SKSR47iOzkuFZYBMn/AD8QtG4IBIJPDD5SKd8SvF+raD8Ihq1k2oRXslrFJLqenW1rcNZAqGeZoZ5Y1deCNq5b5hheK641cwnVp3qR5nNJXUdGrNO/K0463SV11Sva+Mo0lC7WjTf3b/PXft1G+I/hp4m8RT293/wlljY6hNpc2kanNb6MxSWKRwwe3R7hvJkUZGXMqnPK9qu3nw71W1u/C19oWt2dlqGjaa+lSNqGnPdRXED+SWIRJ4yj5gXB3MAGYEHgjU8VePB4b1Ky0qz0TVPEusXMT3AsdJECtHCpCtK7zyxRqNzKoBfcxJwpCsR08TmSJHZGiZgCUbGV9jgkZ+hrzJYzGU4QlJpLW2kNndO6ttq0rq3bY09nCTae/wA/L8dF57M4XX/BPijVPiRpHiSz8RaRaafpsMttHp8+jSzStFMYTNmYXSjcTANp8vC7jkPVXS/hbqtnrGmpdeI4brwzpeoTanZaeuneXdea/mFUluPNKvGhmfAWJGO1NzNht9TU/GWsr+0X4e8M79W0zRJdDv7vynsrKSx1WRJLUb1nE5uYnh83bsMSo/mk7jsFV/Dv7RuleJPFEOk2/hjxHBa3Oo3+j2esXUNtHaXl/aGbzraLM/mk4t5mEhjER2EbwwK1jHMMT7KMFa1ml7sb2Tku176vXfXe7NZU1vLb/Nf8Pf0d9jc8JfCv/hF/+EI/4mn2r/hGdIn0v/j32fafN8j95947MeR93nO7qMc9dr/h/TPFWjXek6zptpq+l3aGO4sb6BZoJl/uujAqw9iK8y+B/wAZfEvxPudah1r4ea14ZhstRvrWPULqbTzb4hnMawsIb2aQzY+8wTyyVbDYxn0Pxp4tsPAXhDWvEmqFxpuk2ct7cGMAt5calmxkgZwO5A9SBzXHiMTUxUVVrO91fptJuXTzbf5aWLV+ZxW6dvmjl/gPrF7qXwz0qy1a21m21vSEGl6h/bdrPFK88ShWdZJR/pCHgiZGdXyfmJBx6FXkd9+0loeg/CN/iJ4j0i98L6NJJEljHqWpaU39oCUL5Lw3EN5JahJC+A8kyKMFmKrhjir+2H4Mm+EX/CwraBrnSo9TOk3EA1zRYvs8+3dg3Ul+tm4KlD+6uHPzgYyrhc5Xu77q1/V2/Vr0ur7ium157fj+ifrZ2PdqK8lg/aO0bVvDfg7VfD2g634qm8V2k93pun6P9kkkYQlBMjSvcLbqV3n5vN8tthCuxZA126+OSNeaVp2keCfFWv63d2EOp3ek2sNrbXGlQSsyIbo3VxCisWSRQiM7HynIBUZpfa5et7f1/nsK6ceZbb/163079D02ivB7X4/3fh34reOtE1zSda1DQbLxDpulwata21utlpQurOyESTM0iSSeZcTtzGspTeC+xCprvo9d1nX/AIsXGlWVyLHw/wCH7WOXUMRI7391Or+XDuOfLSJFEjYAZmkiwwVXDltIvo/8r/ky5Rcd/L8TuqK8m8F/ELVbzx98UV1ldZstP0OO1lttG1CzslEUPlzZngmt7iRpVmMRYLN5bJgDaMnFLRP2kl8VaHqF5pXgHxWLsaRHrek2V8ljBLq9m7bfOhV7sbAuUZlnML4dcKScVDkkm+yv+Df5Jv0T7MmPvNW6u34pfm0vmu57NRXgdn8bvFfij9lLxH47m8Naj4J8QW/hGTVrW6u/sUsM0/2JphNbpHcXH7sOAQs4VsEAqea9BtvDWuP8K009vHmvnWWtxP8A8JL9n077cCTvwE+yfZsY+T/U52/7XzVpVi6PNzfZ/wCD/kFO1SEJx2le3yt/mju6K4T4C+JNS8ZfBDwBr2s3RvtX1PQbG8vLoxpGZZpIEZ32oAq5Yk4UADPAFZ/xK+Oln8ONdOkDwt4j8TX0emvrE66JBblLezRiskrvNNEvykD5AS7Z+RWw20rJUajpzeqbX3Xb+Vk36DSbWi/rb89D0uivP7/43eH7Gw8RXoivbm20RLGSaSGNCJlugpiMeXGRhxndtx2zXEXX7YHhm28Tvov/AAjHiuTZqraU1+tnALYMuojT5JdzTAmNJ5IQSBuYTDYrFJRHMfen7NfF2EtY8620/Hb7+h7vRXnnhD4y2vjjx94h8NaV4f1Sa20K5eyvtc+02BtYrhVBMTRC6N0jHPBeBQwG4EqVY+h0LVJrrqHVrsFFZviTUbrSdBvryx0y81m7hiLRWGntCJ5m7Khmkjjz/vuo4615P4R+M0vhn4B+BfEuvnXvGeq6ytvaIbXS7aC+vLqbdsDW8cphiJI2kiQov3iwUFhPMk7PfT53dlb5/mh20v6/ge1UV4R46/aS1fSfBOia34d+H2valeXPiGHQ9R02d9PWbTpPtccEkMmb1EMjhz5bRvJHkqXKrXs3h3VLnWtFtL680e90C5mXc+m6i8DzwHJG1zBJJGT3+V2HPWrXvR51tt+CfzVmtSOZKXL1tf8AFrXs7p6M0aK8b+Inxs8T+DfjL4c8I6Z8Odb8T6ZqOl317JdabNpyPK8L2yjyjcXsOFTzm8wOoJLx7NwD4l8aftPeFfAHxZ0bwBrMbWuparJbRQXLatpajfO2yJfsrXgvGy+F3JbsuTnO0MQL3uW32nZet2v0LknFXfa56/RXm2l/G2LW/GU2h2Hg7xRd2Ntqcuj3niCO1g+wWt0iltrgzecyEbf3qRNGC4DMpDhef0n9qbRNV8VWOm/8Iv4htNFvrfUry18UXRsU0+W2sSFuZtv2o3CqrsqgNCGO7O3aCwhSi0pX0av8rX/LX01Hyvm5Fv2PaaK8E8E/th+F/ilpvic+DNMuvEOr6PZi+i0vT9V0i6mvIS+zzV8q+ZIgCQSlw8MmOik8V0Pwq+IHi34zfBu01eTQb/wB4hvtLtrm1vb9LOe1nmkiWTzIoormZvI3cFZTHJtbA2t8wuz5XK21vxv/AJP5+jM1KLko33v+Fv8ANanrVFcz8NvGDePvBGla5JaGwurhGS6s927yLiN2jmjDYG4LIjqGxyAD3rD+Mfii80vSdN0LQ9aTSfFOu31va2fkrFNeC3NxEt3cQQyBlYwwuzlmR0T5WdSOCWfMo9W7DjJSjzX03PQqKgsbeS1s7eCW5lvZY41RrmcIJJSBguwRVXJ6naoGTwAOKnoBbBRRRSGFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFUtat5LzRr+CJd8stvIiLkDLFSAOau0VUZcrTQ4vlaZ5T4G+C4j+H1tpfirU9U1m7ufD6aNNDePbqLGF4lWaGEwRoMZVRubex8tfmPezL8CbO6tNXF54o8Q32o6nFYxSapPJbedH9juGngaNBAIlIZsEeXhgOQWJY+m0V6ks1xcpympWu72SVlqnoraK6Tt1aTdzCnSjThGC6f8D/ACRwlx8KftPhbV9IfxVrv2vV3Jv9axaG7uFMfl+Xg25hRNgC4SNehOdxZjHrXwnl174fp4SuvGXiAWu0xTXsaWIuJ4cYETf6LsCgYwVRW4GWPOe/ryf9p/U9a0j4P6vc6RHOIYyj311Y66+k3ttbqwZntpVtp90mQF2MEBVm+cd8lj8Qmmmrp3WkdHsumiXRbK3kOUYcrctkn/wTodS+Gc1+2mXcfi/XrLXbKCS1bWrdbPz7mF2DlJY2tzAcMqkFYlYYOCNzbu0iQxxIjO0rKAC7Yy3ucADP0FeUfFP4xeI/BPiSbRfDvg+18RvZ6JJr95dX2s/YI47eN2VkXEErPKcAqNoU/NudMDdgat8T/iXe/Gzw1pvhjQvDt94Q1Tw7NqsS6lr01pNKokswZXVbCUo6ecyrGJCrhizMhAFcv1iWL5YXTte2iVviv02vGVuis7WVxySopyfl/wC2r5v3o+djrvEXwXufEHxT0rxwvxA8UabPpkbwW2kWcWmmyWGQwmeI+ZZvMVlNvGWJl3DnYyZq1pfwT0PSf+Ed8m71Bv7D1/UPEVtvkjO+4vPtfmo+E5jH22XaBgjamWODn0GvHfDanw1+0Z4/hXUtXn0uTw3purvZXuqXN3BDO9zfrI0EUsjLCCsUY2RBV+UcVlGTirXslf8Az/z/AKZ0+9OMrdLaer5f/bvzOt8L/DM+EfF2s6vY+JtZOlanNLdP4blW0NhDcSFTJNGwgFwGZgzbTMUy7YXpjp9c0+41bR7yztdTutFuJ4ykeoWKxNPbsejoJo5Iyw/20YeoNeJr+0h4g07TtP1fW/AttY6JrWmSaxo0tnrZubl7eMRyMt3F9nUQy+TJvCxPOpKMpcfKzdLN+0FpS3mpCLTpms9LtLq5vZp7u2tmj8u4jiiAMsixBZlcyo7yL8gGQC2A6dKVSccNBe9sltt08rfhp5GPMrOondaee/Xzu+pV0v8AZp0XS/C+r6bH4g1saxqesL4gl8S26WVnfrfrt2zhbe2jt2bC4bfCwkDMJN4Jq9rfwPuvEGm+HftfxF8VnxBoN5Nd2fiRYtLF5+9iaJ42j+xfZiuxyAfJDj+9yc8vrH7SWneJPDEMujXMmhLc6rb6VLqUF1p+qS2wkjlk3RR2k9yplbyfLjV1O6SRMJJ906fgv4geOB4N8ZoPD9z4s8S+H9RjtbOwvr21s7q8ikht5wLiWNBAkqLO2Qi7cIo5J3H0a2WV8NTnKqkuW11/4Cr6aP7Ket7WvpYzVRTs1rfr83313v6PU6Xwb8E9D8Dr4WFheanP/wAI7FfxWjXtwJmkF3KJZTK5Xc5DKMNnPXduPNS+LvhQfEHiyHxLpHivXfBmsG2SyvJ9DWzcX8COzxRzLdW86/Izy7WQI371xkjGO7GcDIwaWvL+1zdTRaR5Vt/X9eup5/q/wU0PWv8AhJvPu9QX/hINZ0/XLry5Ixsns/svlKmU4Q/Y4twOSdz4IyMWIdB1jQfixc6tZQLeeH9etI47/wDeKj2N1AG8uUKf9YkqMI2wcq0UeAQzFO4op328v8rflb7ipSclZv8Ar+keW6R8C5tN8aeJPENz8QfFWrDX7c2t5pd5Hpi2vlhZFiVTFZJKPLEr7T5mTxv31vaD8J9H8P32iXMM15OdJ0AeHIY53QrJbZiO58KCX/dLyCByeOmO0opbf15NflJr5kOKf9eaf5xT9UeL698FdT8Ofs8+OvA+ha9rXiv7V4dutL0TTtXayT7Iv2V4obeOSOGEleUXdOztwCX6k9bbeJdcT4VpqDeA9fGsrbiD/hGvtGnfbiQdmQ/2v7NjHz/67O3/AGvlru6KJ3qRlGT3/r8b63NVKyikvhbf32/DRWOE+AvhvUvBvwQ8AaDrNqbHV9M0Gxs7y1MiSGKaOBFdNyEq2GBGVJBxwTWhr3w30zxFrWp6pcz3aXGoaLJoUqxOoUQOzMWUFSd+WPJJHtXV0U69sROU6i3v+Kaf4NolNx28vwd1+KueO65+zHouuXUn/FUeJrDS7i0sba90myuoI7e8azx9nmkYwmUMNqhlSRY3CgOjDIMtx+zH4WubiSZr/Vw8l7LfkCaLHmSatHqjD/V/d8+JVHfYSM7vmHrtFNSanzre7fzZKSjHkW2n/ku33HnFv8FLeT4p2vjvVvEur+IL/TxcLpdnfWunJFpyzDDpFLDaR3DJt+XZJM6nhmDMqsJvhb4Q8UeH9Y8a6v4m1drk69qgu7LR49QlvbbSoEhSIRwySJGQJDGZWQIqq0jKC2N7eg0VK0t5K343/Mb1u++v6Gb4i0u51vRbuxs9YvdAuZk2pqWnJA88ByDuQTxyRk9vmRhz0rz/AMGfAO08JeE9A8PXPizxD4jstC1GHUdPk1QWSyQmJSFhzb20QaP5mJ3Avk/exxXqVFTyq/N10/B3X4hurev46Hn2sfBXSNW8Ma1oy6jqdidS1ka+t/ayRfaLO8WaOZJIt8bJhXiUhZEcEZDBgcV2Hh3S7nRdFtLG81i91+5hXa+paikCTznJO5xBHHGD2+VFHHStGinH3YqC2VvwVvySXyXZBu79dfxd/wA/17nE/ED4YjxxqWi6rZ+JNY8I65pImjt9U0RLR5TDME82FluoJ4yjGOJvuBgY1ww5B5fW/wBmvSdc8XalrT+KPEltZ6jrFh4gu9Dt5rZbOa+tDb+VKWMBmxi1iUx+b5fUhA2GHr1FVFuLTjuv80/zSfqr7ju/6/r+lpsYXhbwfZ+EV1cWUs8v9p6jNqc3nsDtllILBcAYXgYzk+5r5X+B3wH8W+HPHUqX/hS80Sw1hL6LxjPfWvh/7BqkUiyBUtLmzUalKxlkVhJetuMYkL/vGFfYtFRGKjZW0S5bdLWt+QXdnbq7/O9/zOI8A/De88E2lzZXnjfxH4v014Etrez18WTC1jUEYWSC2ilkJUgFpnkY7Qc5JJoeE/AeqfB/wDeaZoGrax40NpbJBoul65LZxJaIi7IoFmigjYxAbdzymWTauQWbhvRqKp3aavv/AF9/nvq+7ErRa02/r7vI5v4c+En8DeCtK0Wa6F/dW8Za6uxHsE9w7GSaQL/CGkd2x2z1NZHxo0fV9W8FxvoWmHWdS0/U9P1NNPSeOGS4W3u4ppI43kIQSFEYLvZVJIBZQSR3dFO/vKXVNP7ncUUkuXpsQWNxJdWdvPLbS2Uskau1tOUMkRIyUYozLkdDtYjI4JHNT0UUAtgooopDCiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAK5L4lfC3w/wDFzQRoviVdTl0zdveDTdZvNO83jG2Q20sZkXn7rErnBxkCutrx39rKxvL74H68YV0m506AJPqOn6vaXEyXkCOG8pHguYHifcEIfLgbSChzwWvoTKSjFyZ2lj8KfDdjbtD9nvbwtpb6M9xqOp3V5PJaMzM0bzTSM7nLH52Yt2zgCoPEHwd8L+Jf+EfNzb6haTaDEbfT7jSdYvNPnihOzMTS28qPJGfKjJRyysUUkEiuL+LvxC8eaH4uvNH8Hv4cs7TTfDcviC5uNbs7i6eQxyMot0SOaIDeB/rCx2bfuSbvlwNR1r4k+JPjx4Ol0HxdoejeHtX8K3GpxaRqGh3N1hRLY+YHKX8SyS5lOyQIPLVmUq+7NTRcaklyef4Obv8AfGfnu+upU/dp3XT8+RW+6UfK1l00+iK8+sfgT4T074iXPjiE6+3iK5BWaSbxPqcttJHmQiI2r3BgMSmWQrH5exCxKgGvQa+XPgd4y8V+FdQ0SxuX0abwl4j8b+JtJt7SK1l+3xSpealdfaHuDL5e0m3ePyRFnlW8zqg1jHm1W5q7qm30utPk3f5WPV9J/Zw+H+hw6pDa6TeGDULCfSzBcaxezx2lpN/rbezSSZls42AUbbcRjCIBjYuOl8LfDXw54L1S51HRtPa0vLmztbCWQ3EsgaG3UpCNrsRuCnBfG5gF3Ftq48k8K6rqHw3uf2gvEN3Y6Nquv6fOuovJpyT6emoFNNjeCOXz7idUcKFjLpsU43bAc1wOtfEL4neO/hN8QbjXoLGfRbaz006bJP4R1LwzJNfPcqZYWW9nllMcYEQMqw7W835SSjLWKqXhKUeiu/x0++6+fmKaaSb1u7flr+T+Xlp9WeJPDOneLtJk07VIXmtmZZAYpnhljdSGV45I2V43UgEMpBB6GuA+I/wF0/xn8IdY+Hml3S6NpOuFotUurpJb+7nhf/WsJpJg/wBoIC7ZpTJtKrlHAC1laf8AFDxr4ftfGGn65aaT4n1/Qbu0RJNF0+/tYJobiIOAYYlvpgyHeCwBVhg/JzVLUvi54+14eD9O0LTNF8Navr+p3VlJNrlnqU8VtFDbGfzVhmispZGO0rt+Vec7ztK16PscQ8J7VXdLmtvpzeav9+lvkY88I1Fd2fR/1/w/zPcLa3W1t4oELFI0CAuxZsAY5J5J9zUteF+Dfj3qmq3nhCDWV0mzivvDutajqd0FeKNbnT7q3t2ePdIdkLeZM5VixAC/NwSfKV/av+J2saP4cvtLs/CNvHqmlWdw0lza3UuLifRZ9Sdgqzr+7Q2xULuJcSj5k8vMnBUkoQdVu6//AGv/AJBm9Om5KKj1Wnpa/wCCPsqivCbz4teNLXxf4UutSWy8PeA9Wh01Fvm0CfUheXd1gGD7RDeKbI72RFee2aLJX94WcIPZP+Ek0j/hIjoH9qWX9ui1+2nS/tCfavs5fYJvKzu8vcCu7GMjGc1o4uOj7tfNb/5+muxjGSkuZbWT+/8Aq3roaVFeG+C9M0P4a/Ff41XujaFZ6fbrYadq11a6XapAbqcpdvJIQoAaRyOWPJPU1xfh39oL4uyfDXxN421jwXYRaONEXWNDuZBaw2zuzKUgMttqd41wjI+fPEcAGzPlneFWFq2l0Sb+d9PnZ2727tIXOlHnl3a+52v+Kv29E2fU1FfPHiD40ePfAPhvxzZ6yuh674p0afSltbrRdIukg8u/kEQP2MXEs1y8JWV/LidWmAVVCMc1zfiXxj8ZfEPwM1nXX17SfD08PiC1hsr5/B+o6VNc2AuIEMjW01+J4MyM4IkxujRhtxIrhPRSa6f8D/P562vZl3Wmu7t+f+X+Z9V0V85+ObH4jR/GD4YWttr3heXxadC8Qedq8uiXKWCR+fpxG2yF2ZGbAVObkAZL84EZ9D8E/FfVPFnwt8EeK4PCGpavPr+mwXtxb6RNahbRnjRiCbieIlcscbdxwvIHGd6dJ1bcrXzaS3kt3ZfZf5Fyi4xUuh6TRXlnjuG90/4x/Di7i1rVFt7+8urWXTVuilpsWxnfmJcByXVGzJuKlRt2gnPTSeINSj+K0GhnUvDq6RLoz3i6a07DWXnWdUMqx52m2CsFLYyHZRnmqqUvZqnZ35036W5v/kf63MlK/Nptb8bf5nW0UUVzlhRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAVznjb4beEfiZY29l4w8LaL4rsreTzobfW9OhvI45MEb1WVWAbBIyOcE10dcJ8S9Q1iPWPBmk6Tq8uirq+pTWt1dQQRSyiNbO4lGzzVZQ26NCCVI45BHFdFCg8RP2aaWjeu2ib6Jvp2JlLkVzb0f4e+FvD+mx6fpfhrR9NsIrRrCO1s7CKKJLYsWMAVVAEZLMSmMZJOOai8TfDLwd40s9MtPEPhPQ9dtdLdZLCDU9NhuEtHUAK0SupCEAAArjpXknxY8YeL/A91oehP420vSV+ySznxDrd7b6QL2QSlViLSWVzCWVChZUETMTuXaoKrJ4087S/ir4P1vxB8RJfCVve6HcQ+dbyWCWIuN9mWt4Zbi3YsJcM4DEuQnylQGFe1TyWpNQm6kbTUnG3M78t3slfV3a6rW6voc06ygmuXZLTTZ2Xps/TTse+Vkw+EtDtvsnk6Lp8X2O6lvrbZaxjyLiXzPNmTA+WR/Nl3MOW8x8k7jnyXxR8YLrQ/EHiTQm1uKDWk8UaRbadp/lI0/8AZ0zWKzOE2kmItJcL5p4DHaGBCiun8GL4m8Sa5qOqzeK547Gw1q8sho62NubeW3R2VQz7PN8wEghw4XCgFGOWPPPK61Gj7erJRi1dX5tbpNWsvO3a99TolPl0Wutvz/yZU1j9mf4ePofieDw14T0HwPrmu6bdadJ4i8PaPbWt/Gs6kOwlRAxOcMQTyQK5b4X/ALOPh/4O3Wu65r6+BbKzvbOCwms/DXhePQNKZUl3pJcQvcTiabeyhXZhtHAGWJq/8IfFXiq8vPAk+ueI5tcTxRoFzqE1tJaW8EVtJEbbYYvLQNyszb97MC3KhB8tbH7U14NP+Aniy6a6tbFYI4JTdXwJgh23ER3yDcnyDGT8y8A/MOtcGPwsstk1JqVk9r20cotapdU1sXrU92Xf/LX8F9x6O1tpmjPqGqtFaWLyoJb29KrGXWNcBpH4yFXPLHgVDZNoniu30jXbQ2GsQeX9q03U4dk67JEx5kMoyMOjfeU8q3oa+fdP8fWeofDX4iy3/wAbE1TwTCLdbL4hztoyw+bIpE1msn2b7HKgYRr80bHNwybiy8UvA/xCuviF8MfgRpXgLx3p+jW2r2q2Or33ha2sJzavBpvnPbxRmN4Ld1dVUoYyEUkbB8pXlUdJtdOX8b3+7T/g3RnKcU4p9b/g0vxvoe+6t8LfBmvW+l2+p+EdC1GDSro32nxXemwyrZ3BYuZogykRyFiW3Lg5Oc5qSP4ceEoooIk8L6KkcCqkSLp8IEarC0ChRt4Ahd4xjojsvQkV8vfEH49eKLX4c/D2yi8Y/wDCO+J9SGpLJrl9qNhpVvqDWc/2by902mXqvdTMUdbeCFC2JMEBQp7Hwb4u8efG5PC62fjO58DtqHgTS/EUv9mada3DfbpzKGDfaIn/AHPA3IArnau14/m3QrSuktL29fdk/wAotGqbUFNbWT+9pL/0r8z2X/hT/gP/AISyz8U/8IT4d/4SezRYrbWv7Jg+2wIqbFVJtm9QEJUAHgcdK3P+Eb0j/hIjr/8AZdl/bptfsR1T7On2r7OH3iHzcbvL3EttzjJzjNeP+IfG3i3S/iQ/w+GtMNV167s9S0jUFsocW+mxhf7RjVTkEoYSA7glTfwDLFaxY/ite2/7Ur+GLrx4mqW09x5Fl4V8NanpNwbNFtN7tqdpJAL6H58sskE0inK71iX79x1as97/APB/F29U0J6J36Wv+a/Rpb7Hq+g/BP4d+FfFEviXRPAPhjR/EcrSPJrFho1tBeO0mfMJmRA5LZOeecnNO0b4K/Dzw7Jq8mk+A/DOmSawNupNZ6Pbwm+G4tiYqg8z5iT82eTmvAPgn8bvGnij4larJrvi7wzcwQQX01/4Ftdehn1jThCSFWPThpsFxG4IUMZbmZTvyuQyGuH8CftNeNb3wj8SdRt/Gdn4mjt/BF14k0u6TU7HVm026iGRBMbXTLOJJQJUMkDNOy4X5lDAuqX7xxUN+W69NXb52fkDjduL6yS9W2lf72j6o+MHwhs/ij4R1TSof7O0+91CW0luLi901by3vVt5RIkF3DuQ3EBwymPevDEZGTnmvhR8EPDfwh0HWvD+sv4VePxlesJNC0nRYtJ0eZhahGgt7BpZclooWeTLuXw5OFAAh8cTeOvhX8OPEuunxxP4iE7WbLdalpVsv9gxPOEvLlRAkYkghhcyqkoZl8li8kgOBw3iWeHxuvwfj8JfGSbxsZvGN1EPF9u2lXs1t/xJ74vHH9lgS23qudu+N8FwWDjC1UIayS+f4f8AA123Su7jjrFT8m18k5f11Pf9F8B+Dfh/plr/AGR4d0Lw3p2kQzi3+xWMNrFZRSMJJ9m1QI1YorPjAJUE9K2NBs9M0/RLC20WC0ttHjgRbOGwRUt1hwNgjCfKFxjGOMdK8Atde1D4hfDfwPpPie7bVrS88bXmgavdzCOI30FlcXyxeckaqh86Szt1kRFVG8xl2hSVrX+NGseM5PG2p6d4e8Z3XhPTtK8JT66UsdPtJ5bi5jkdUVnuIpAIiB8yqoY4Xa6YO7KrUVK/tH1f/ksXNv7r6b/eXySvyde3m5clvvv5Hp3iH4VeCvF2rx6rrvg/Qda1SNVRL7UNMgnnVVJKgO6lgASSOeM1B/wq/TJfikPHt3dXd/rFvpz6XYQz+UINPgkZHmEWyNXYyNFEWMjPjZhNgZg3hviz9pa50OLxNZXPiG30zxJc2Oh32g6OsKPcPHcBPtMkMWwtLEreZvkIZYwpJKgZrjdS+MPxQj8YTzr45ePSo/EU8K6YmlWm37NF4mh01YDIYy2GguH3NkMTHCVKESGXqUqtWrCg29G0r7LdO3y7dDGPLKm6q2bivP3vh/FH2pRXzr4J+K9xe/tI6t4Yu/H6eJTLcXaW+geG9T0m7tNLhiUf8f8AAIEv7aUMCufMmiLnDNGXSMdz44+JFv4n8E/FDTfBV7c3virwxBLZ3drZwSJcwXLW6zIke5QHcxupUoSMkDOQailD2rgk7c3for219HuU/dk422dvV+R6jVW+1Sz0z7P9su4LT7RMtvD58ip5srZ2ouTyxwcAcnFeA/AFtKj8cazqPhV9Dg8KSaSr3lh4WVJLC0uVkzEryxDbNdmMyGU9QPJXGArvc+KXi6x+MXw3+Gmv+BvEXk6frXiWwew1pLMl0Q+arOsUyjDjDACRSAwBZGAKnfHUPqdaNJO93BbWa5pJarXXe2tnunYzhNTUntZN+to3/wCH6o9t17X9M8LaPdatrWpWmkaVaJ5lxfX86wQQr/ed2IVR7k1eVgygg5B5BFfJPxm8d/2b+zN44i8YfES3tLvRtW1DTLa712DSceIDDueK1ngntjBIWXGVhjRj5eRt5o/aX/aOfR/+EWbwh8Q7TR/D+s6TeXFjr2i6toEUM94jpHEHm1R/Jlt1zIXW23TKQOACAeHmXJGa1UrW+ceZfgaface17/J8v5n1ZHq1jNqlxpsd7byajbxRzzWayqZoo5C4jdkzkKxjkAJGCUbHQ1br468aah4s8M+IfiD4t0nx7O174c+G+i6xPdWltZXCa5LE2pyASuYNggk2vnyEiY71KugGD9QXPxF0Cx8aaL4Ru7/yPEmsWU+oWVkYZCJoYTGJmEgXYCplT5SwJ3ZAIBroqU3B2/rT/ga+m4PTXo0vxUX/AO3JHS0V81eAbeLwj+05400zWvi/qUWq6peW99p/hTUn0iE6vAbMLlEFos8iRGN1BicH9z85Y7i3JfCn9obxN4s+NF/ZxeKYdW0C9i1fGjSarZ3d7oklu58lbm3t9Nga0kIR8RT3M7EBupRmHL7RcnM10b+7cvl1t5pfefYVFfNen+I/iL4e/ZxuPibdePG17Ur7wfDqY03UNMtILSzvWhjcSwvGiFI/mfcLh5FB2sXjRWB850P47+LZvhP4z8UaN8S18Q2n9q6fpsF1rusaBK/h+3kAE91cz6RaXNvDlydryrMiKFZ0VVcnWf7ubhLdb/N8v5/K3zMYTVSEZx2lt56X/L+tj7aqvc6jaWdxawXF1DBNdyGK3jkkCtM4VnKoCfmIVWbA7KT0FfIXi648UfED9lb+2dU+KVjfQ6X4isruXxB4P1XTtVt5LGK/tnkkurk6fFDut8SykxQxKBEm/eA+71XxreaRrP7Luu6zp/j7/hOYdN0+51XTvGIls52N1bM8sMqtZxxwsY5YwmEQZ2FWyS2bcbJvs19zSd/zXqtPLSK50rbttfd/X3HuFFec/ETxU/h6++G8l74wXwkmo63FYzWbab9qXWJJLabZZeZg/ZyXAkEmeTFsz89ejVNt/J2/L/P+tCU7281f8/8AL+tQooopDCiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACkpaKAMDwX460f4gaZNfaNNcPHBO9rPDe2U9lcwSrglJYJ0SSM4ZWAZRlWVhkMCd+vNPhvHeN8SPiZNBPDHoo1eON7SSIvO139gsy0qyhgEiKbF8oozblZ/MwwjT0uqe0X3Sf3q/wDX+Y3pJr+v+HWz879AoooqRBRRRQAUUUUAFFFFABRRRQAUUUUAcjpXw3tLHx/qHjC81PUNa1aa3NlZi+8kR6ZasyvJBbiONCFd0RmaQu5KIN2FAHXUUUbJR6L+vz183qHVvv8A1/XloFFFFABRRRQBieM/B+nePPDd3ouqJIbWfayywOY5oJUYPHNE45SRHVXVhyGUHtWvDGYYY0aRpmVQpkfG5sDqcADJ9gKkooAKKKKACsDwn4F0XwQmoDSbaWOXULlru8urq6luri4lIADSTTM0jBVCooLYRVVVAVQBv0Udbh5FHXNHt/EOj3mmXUl1FbXUbRSPY3k1pOFI5KTQsskbf7SMCOxpug6Dp3hfRrPSdJs4rDTrSMRQW8IwqKP5nuSeSSSea0KKACvPfH3whbxl4jh13TvGXiLwZqi2LabPNoYspBc25feEdbu2nVcNuIaMI3zHJOBj0Kik4ptNrb/K35Nr0Gm1/Xz/AD1Mfwf4T0zwH4U0fw3otubXSNJtIrG0hZ2cpFGoVQWYkscAckknqagXwLoo8Zt4re2ln137N9kjuLi6llS3iJBZYImYxwl9q7zGql9ibi21cb9FVJuUuaWrEtFZBRRRSA5/x74Ls/iH4R1Lw9fz3NrbXqBTcWbhJoWVgySIWDLlWVWAZWU4wysCQcX4f/CiLwPrmr67e+I9a8X+ItUhgtZ9W1wWqSi3hMjRQqlrBBEFVppWzs3EucsQFC91RQtG2uodLBWH4q8I2XjKGwt9SeZ7K1vI71rRGAjuXjO6NZRjLIrhH2gjJRc5GVO5RQHkYOt+B9G8R+INF1nUrea6vNGd5bJGu5hbxyMMeYYA/lPIBna7qWTc20ruOd6iijpYOtwooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKSlrO8Sf8i7qn/XrL/6AaqMeZqPcqK5mkUfBfgXR/h/pk1jo0Nwkc873U817ez3tzPK2AXlnnd5JDhVUFmOFVVGAoA3mbapOCcDOB1r5w+FvgS51LwCdQ0vwhN4bmvvBpsZ5rqa2M2uXUsKGKZmjlcsFw+HmKuPOxgAGvS9O+F/hvwn4SvXh8FQ6peX1hDDqmm26Qs+psvXzhM6xSyZZiZJGycnLGvcxWBw+Hm6brczTttH8Xz2Sttq+ztrbkpVnVjGoo/Fr+T7dW362vc7Dwtr3/CUeH7HVf7O1DSPtUfmfYdUg8i5h5I2yJk7Tx0zWrXi3g/wvrMPwr8NRS+Eb7RvEHhKeOaysZZrMtcBQyyRwtFOyAPDJJGPMZAGIJAABqh8QvhlrGoeCdC0yTQX8QLdalPquuRWdvpl1cpPIsjr5Q1EGDYrP5e7BcKihRgnFvLsPPEOmq8Yx5mt00kru9+azVrWd9W7dLlKb5U7dP6/rse032u2Om6jpthcz+Xd6lI8VrHsY+YyRtIwyBgYVWPOOnrV+vni5+CzN4T+HN34h8BWPjPVtBMlpqFi0NjNcSW3lTJEqvN5UTIjNGwXKAdQoI20/4k/CW/8AFXj+1v7jRdfl0hbS0j006GNFL6Q6E7x/piF4GB2MJLVyTtHQxoTqsswcpxgsSlpK7drXUmlb3r6qzV/k3ssvbVOVy5O2n3fr/wAMfQlYFp4u+1+ONS8Nf2Lq8P2Kxgvv7Xmtdunz+a8i+TFNn5pU8vc64+USIcndW9XgPg3QdT8P/tOeMtZsfhJqWgaD4g0+1srjxDA2kxw3V1BNeO93Mkd2Z3EiywqrNGZOMMqgZr5jr8n+R3W9yUr6q35q/wCB79Wb4Z8Sad4w0Gx1rSLj7Xpl7GJrefYyb1PQ7WAI/EV82/Bn9mGLwbB8PLfUfBWmxwXXg670rxstwILn7fck2Rgiuss32kKEuVTO9UXKjapAPZfss/C+w+E/h250KL4W2/gfU7NEtrvXrW206OLXdjyBJVa2laZgAd379EYeZgDOcXKPLK1+j/BtfitV6lTio7O56r4u8Y2Xgu30ya9inlXUNSttLiFuqkiWeQRoWyR8oJ5IycdAaPHfiz/hBfCGq6//AGNq3iH7BCZv7L0G1+03tzggbIYsje3PTI6VxH7SPhXUPF3gPT7bT/D174qFvren3l3pGn3UNvNc20c6tMgaaaGMgpkFWkAYEqcgkHyPxx8GZ5v2dPE/h/Tfgz9qk1bUbi88PeEbddJH/CNO0ARJv3lwtvC3mLJIfszuVNwcZ+Y1NP3lLm0s/wAPd/zf3NaGMm4zVldW/H3uvyj991fY+r0beitgrkZw3UU6vBfGnwqHxx8ceDNV8XeAmfw5beHtWhutE8RvaXCw3sk1n9nE0MU0sUp2xzMpBdUIByrba8P+LHwH+Kfjjwb4Ps7nwnPqHiXRfBNnBpur2segTzWutpGxnF1d3ySTxgOkBjeywS7OWdflZbjG9rtLX9ZK/wD5KvlJGkYtxTv/AFp/m7+h91UV4t/woPRPGHxU8VeIvGXhm11XMmmz6JfXTh5LSWGIF5bYht1vJ5ipl02M4RASwVcYemeGTcftAah4Hhs4pPCGk3EfjuRk+5Fe3BkjjtWX1M8c95njDbeOhrON5NJ+f4f8C7+SXUi/ucy/r+nZfO/Sz+haK+ZPhx8KNe8P/FTxvqGjeC28Opq1tqTN4s8S6bpB1Z7qaUNEsF7Y3DzSWoIDCO6h8wbVzKcLGvG/C74C+KPA/gXx9JYeEPEMHjLUvDz2Nxa+Io/DL6XrNy3333WUccl3JnzNsl+q7hK2/bvkpKWl/K/z10/BfeU9Hbzt+Wv4v7j7Nor46+Hf7PvinQvhz8QvD1p4a1PQ9G1i/wBHu9N028bRbC6Ux3CfbHZNJWK3ifZEpBQszKEO/d8ieofEf4B6LofgqDTfBHgizm8OtrkOqa74K0nybS31mARNG8KxO6W4JbyZWRiiS+SQ5+YkuXuysvL8bfkTD3o3em/4X/Pp6nquoeNrHTfHGieFZYrhtQ1eyvL6CRFUxLHbNAsgY7shiblMYBHDZI4yeNvGtj4F0dL68hubyWeeO0tLCxjElzd3DnCRRqSASeSSSFVVZmKqrMPFLj4C6D4+174cLefCWz8O+BNK0nWobjwlqMFgLa1nmntGhWS2tpZIJA5illG0uqsFY7XAxa8O6fqfgDwp8A38bo0NxpMCaTqE11MkotL6Wz8iF5JQxBJcNAGBOWuFGfmrRpJLvdL1vJr77Jd909dnpJLlUl2/z/Dv/TPY9Y8Wf2P4m8P6N/Y2rXv9sNOv9oWdr5lpZeVHvzcyZ/dh/urwctxW9Xg3xS0K/g+P/gXxZovwk1HxBdaWs0V/4p006TFK1tLBLGtvvnu4p2VHcOV27eSV3NxWTJ8AbiPx5qHjWLwzbP4t/wCE7t7601qSSJ7qHSDHDHOsUjMTFEym43RLjcWYlSWyYj70lF7Pr/28l+Tv6J72OeU3BSdm7befuuX5rl9Wu57zpPinS9d1TW9Nsbrz73RbhLS/i8tl8mV4Y5lXJADZjljbKkj5sdQQOe+HvxQX4i3XiIWnhvWdOsNH1C401NSv2tBDfzQTSQzeQsc7yAK8R5lSPIIIzzjzDwF8D9A+HH7QnibV7T4PacV1i9XUtO8ZabYaXGmlA2SQzwsTKt0jPJHKxMUTK32nJPL47H9nH/kT/En/AGOPiL/063NVFJxT68v46X/X/g7nXKMVByj3ivvUm/yOg+GfxOT4kL4hjbw9q/hjUNC1H+zL3T9aNq0qyeRDOGVraeaMqUnjP385yCBiu1ryj4M/8j98av8AsbYv/TPp1Xfjp8Oz8TtH8NaRLpw1XS0161udStXlEaPaKH8wONw3oQQrR871YqQQSKVT3fZ2XxKHy5lG79Fdv0IklFy8rv7ru36HpVFfNdv8O/E3wxl8N3Xh/wACzapo3h/xVq01n4e0K4srdoNPubeVYmgWaaKJYxI/Me9WUNwpxivPbT9lrxP4i1bRrrxT4Rt5jB/Z6TxrqMcsMaf2zqU16i/MpdTa3SKSUBeOZ48fNIlSnzRjJdbfK6T19L6mV3FVHJfCm1bW9mlp66td0u2p9rUV8b3XwO8R2fhP4Q27fDV/Enizw34dsdOX+3LLRdT0HTZVZd5cTXUd1FKmwfvbQsMBCUmKKg+h7z4rmP42aV8P7HTItUE2lXGo6nqFreqz6SyNGII54ApKicPIUZmXPkvgNgkXJJOyfVpfK7v80rlyTjJrdK2vr/wT0GsHSfFf9reKte0T+x9Ws/7JW3b+0ru12Wd55qlsW8mf3hTGH4G0kDmt6vmjw94P1jT/ABp8Y7bSPg/qPhPSfFmnlYNQjfSIba6vEiuVeaRILtpN0zSoQ7R7j1fZisKknCEpRV2k3bvbp/X3AtZRXRtK/bzPpO4uI7W3lnlbbFGpd2wTgAZJqn4d1+w8VaBputaVP9q0zUbaO7tZ9jJ5kUihkbawDDKkHBAPrXg3gv8AZm0bRXuNFm8F6SnhrWfCdrba3Bdww3Md/qkcpbzbpWLG4lG4t5z7iTj5iQK6D9lX4f23w0+HNroQ+GMPw71WztbW11K5t4dPSPWbiOII9yrWkrs4JDENOqP8/KjnHQkve122+Tkn+Sa8nf0y5m+Wy33+6LX5tPzW/fvPip8Qh8K/A2qeKZdB1XxDZaZC91d2+jm2E8cCIzyS4nmiUhVUkgMWPZTXT2lyt5aw3CAhJUWRQ3XBGRmvOv2j7fXNV+Cni/RPDnhnUPFWr63pd3pUFpp09pCYmmt5EWWRrmaJRGGIztLNyMKeceffEDwzqPjq38F6t4m+C+oeNtDtLC6tLrwRqk+kzzWt4Wi8m7aKW7+xyDZHMocStIgmGF+eTbC2f9dH/kvmzeSSjFrd3/8Abbem7+S0V9/oyivlOx+GHxU8EaKsGk2V1e31r4fs7omDVozFdX8fnW7WJaWRGl220q4eYBC0MTZDDj1S18cL8OfF/wANPhXpemQ6w02mNHfyW9+v2jSLeC3xDcS2+0s0UrxtEJCVAfaPmJwGle6+7z3v91vn00s3m3az+/y2t99/v03PWK5/S/F39p+Mtc8Pf2LrFp/ZUFtP/at1a7LG887zPkt5c/vHj8v5xgbd6dc10FfP3w78Pal4f+OXxFubD4R6j4T0DxHbwIdYjbSY7e5u4WuzLcypBdtMTN50W12j3n+MJipbsm10Tf8AwB9vVf8ADn0ATtBJ6Vl+FfFGmeNvDema/otz9t0jUrdLq0uPLaPzInAZW2sAwyD0IBr5z+Dv7MsPhOLwLaar4K0xbK+8Dy6d42S6S3uRqOo7rJolu8ljcldt2FY71RcqCAVB6z9k34X2nwl8F/8ACPD4YQ+BtXsLa3sr/XLe306OPxA8QdRcK1tK8rjq3+kKjDzemd2NpQUW03f08pST/JNeT26lNWV+un5f1/W3rHinxhZ+ETo4vIp5P7U1GLTIfIVTtlkDFS2SML8pyRk+1HjjxV/whPhe+1v+x9W1/wCyhT/Z2h2v2m8my4XEceRuI3ZPPQE1yXx0+HR+J2j+GdJl04arpkev2l1qNs8ojRrVA/mBxuG9DkBk53KxUggkV4/8Rvgtd2vwL8ReD9P+FCeJymv3lx4X02wTTFi0eNm3w3EX2ieJYAC0gAj+dd+NoFc1OTk5KWlm7elofrKVvR9iKl4pOOun4+9/kl5XXc+qQcgHGKwtL8Y2WreLtd8OwxTre6PDazXEjqojZZxIUCEHJI8ps5A6jGa+Uv2o9F1b4weJvD1g3gO41T/im7ufUPDOsaXperTaYksyxi6VZdTt7eG4AjcJNFLcEBmBRQPmu2/wQ0j4jeE9S8ZeEfCcF3b3Hhjw/N4EuLtIIpbN4UkkV4Y8hLaVQ0eXUJnoCVGK0p3nFz6J2/GS/OOnfXXQzqScGklfS/4J28t/y7n19WBB4s87xzd+Gv7G1eP7PYRX/wDbElrjT5d8jp5KTZ5mXZuZMcKynPNc7ffFfy/jbpXw+sdNj1RZtLuNR1LULa9Vn0lkaMQRzwBSVE4eQozMufJfAbBI4fw7od/oP7TXiXWtO+Emo6RpGuafDY3nia2OkxQ3dxFJM/2mZUu/tDqVdEDNEXzwVAGaz53eNldPm/BP9Vb/AIBu1aMu6t+LX6f1c95rL8N+JtN8X6PHqmkXP2uxkkliWXy2TLRyNG4wwB4dGHTnGRxXzl8E/wBmgeBV+Hcd54N0+2S58G3mk+NvO8ic39y32PyYrolmNyFCXKrneqJlBtUgHqv2Y/hPpvwjj1nRoPhXaeD7+G4uQ3iaxtdOih1eBrqWSBVaCUz4SNkG2aNAuMDOK6JRUXa/f8G1+Nk16iqJwlo7r+v+D+Gp7H4n8Tab4N8O6jrusXP2PStPga5urjy2fy41GWbaoLHA7AE1pg7gCOhrzX9o34U2nxm+Dfifw1Noml67fzWUz6ZDqsEckcd4I2ETqXUiN8sQHHI3GvKfiJ8M5/EHwT0bw34W+Ch8KeH5dbEuseFbXTPDcl6kKxORcW0M8kunF2mEILSMXCbyAG21l0ffT+v6uLqu2v4K/wCOlr2PqGivln4a/s76pe+HfhtoHxB8NW+r6Hog1+G703UzZz2wgkvFbT45IIVSCRRCqEIkQSMouFQquPOv2oPhjpureM7vwvZeFm8S+EvDfg21ivbP+zLLVZtCs2luv39m+oahbC1naOFlEirc5EEe5FEahyTUZ2e19+iVm7vy/wAy4x5k+6/F3S09b/1c+66y7LxNpuo6/qei29z5mp6bHDLdweWw8tZg5jO4jac+W/QnGOccV8p6j8FbvxX8bNK8Xrofiq88PyPpV14a1bw7B4cSHSLFYof9HkN7Et/aqGWRnjtWIZJSAAzMtd9Y/BHw/wCEf2lNT8UwfB7TNTTWktbi38VafYaYjaTdqbj7RLKZJEnDyb4yXhSQtk7sYq+WzSemtvNev9W82YuWicddE/vX/Df5I9l0LxjYa9rWt6RGs1tqejypHc2tygVtjruimTBIaJwG2sD1R1OGRlFXwv46j8W654gsbXSNRhstHufsf9rXAhW2u5gP3qQgSGX92flYvGik/dLjJrmLfy9e+P2qy6dLJH/ZHh4abqN1AAQtxPMs0EZJypkjjV5NpBwtyh6PzP8AAtli8O67YG6+33ena9f2l1eyFfPupRKWM0wQBFkcMrEIqJz8qIuFGcXdpP8Albt10ko39Gtf+3la6TLk+W/ql98eb89Pk762PR6KKKoAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACmyRpNG0ciK8bAqysMgg9QRTqKAIbS0g0+1htbWGO2toUWOKGFAiRoowqqo4AAAAAqaiim227sSVlZBUDX1ut6lm1xELuSNpUty48xkUqGYL1IBZQT0BYetT1wXxH8A33jLV9GltZoLe2jhubS9keVllWKQxSKYgFIZvNt4vvFcKWIOeD0YenTq1FGrLlWuvyuDvbQn1j4ueHfDvicaZquraXplibYyf2leahHFH54lMZg+bA35DcZz8pGOKs694/g0nxz4c8MW/8AZ93faoZXuIZNUhhubaBYnZZlt2O+ZWdNnyDjJJ4U1xuh/B3U4POu9S/suXU7nRdQs5ZIyzBLm7unnk2sYwfLy+CcAnaMrVnX/Cvjhb3wD/ZWnaDf2/hwLNPNfaxPbyTym1lt3VVW0kG0eYGDFsnBBUda936tgOflpyTspLWVk3Z8r6dbaX7dE788pzV2lp+K+H/N/NdUdTdeMNSsfiVp3hufS7X+zdQsbi7ttRjvWabdCYQ6PAYgFB88YYSN905ArUh8aeH7jxJN4di13TJfEEKebLpKXkZukTAO5ot24DDKckdx61yfiPRfGV38VdB1uw0vQptE022uLRpLnV5orl1uDbl3EQtWUFPJOF8z5sjJSuf8H/BzUPD3j0317Zxajp0Oq3mqWupHxTqQeJ5zK3/INZTbBh58kZYOMglsAkrWKw2ElRU5ytLkvaLTvK8t7y005bxWuui0aNJSa+H+v6/Dqdvo/jqfxL4i8S6dpGmCez0TFq2pXE5iinvtu9rdAEYlUVk3yDIDMVAYo4Wj8M/iRd+LPBup6x4l0/T/AAxdaVf3tjfx2+pG6tIvs0jI8gnkihJTC7ssi4H51lfDC4uvDtx428KNbbtds9UvdXs1vJGjivre8nkuIZBKFbCK7vA2FLJ5X3SChbitH+FfxH1b4d+L9A8Q6X4T0u7vNb/4SLTjY63c38E8325bz7Lch7KEpETGsbOu8lXY7OMH5uN+Za6OO/ndfpzabq1n72/S4xvZP7Vv+3df/tX8+x7B4W+J3g7xzYW194b8WaH4gsrm4a0gudL1KG5jlmVC7RKyMQzhAWKjkAE9K17jxBpdpHqMk+pWkKaau+9aSdVFqu3fulJPyDb82Wxxz0rwnVvgv428YX2p+J9TtvDmj+KL++sVFjYalPc2ttaxQXNvLMLhraNpLhor2fCmJV+SNC2MtW14Y+Ad7b6BrFvqupQx6mdbXUNLmt2e5tkht4oYLMXEEgVZcR26MyH7rsWjdXCyDsoRpzqKNWXLHvv2/wA/PZ6aa43dr7/53dvwW/mu57Dpepwa1ptrf2pkNtcxrLEZYniYqwyCUcBlOOxANcFqnx08N+G/ihf+DvEWp6V4dMdnp01leanqUVv9vnu5bqNbaJH25cfZcgKWLb+gxy3x78G7T4uWOiS+J3t7TUbGFhJHa6fp+oRB327wjXtpKwXK8bQme4OBjh/Fn7Nl5fWnxFg0mXS1fWvAlr4T0i6uIkgkgkiN4WLiGFUjjJngOIlxmM/INq56alHDxgpRqXk27xtsr6e9s7+Xf5FQvLmi1baz+cb6eSuei3/xVtYfjFpHgGyOk395Pp9xf6go1u3S+09U8vySbInzpEk3t+8A2rs5+8KpWnjb4QfC3xXP4Stde8E+EfEusXwu5tChvLOyvL27uCAJWgDK8ksh2/MQWbjrWV4w8PfEa6+MHhLW9H0Pwxd+HdGtp7WWe+8QXNveSC48kSOsK2MiZTyjtBl+fIyUr520nxZFp37TFzavqNnrMf8Awnkstv4Lj8SwW+q211JH9lbURpf2D7Q0SwtJKWe9MTRFphGMog82lzTqqn3v8/eX3b7vTRd20S0puo+n4e63+mvVJve1n9S/Cv43eG/ilax29rqulQ+JlhNxd+HYdSinvLSPeVV5Ixh1VhtIJUD5hW54I8bL4uXVraezbS9Y0e9axv7F3EnluAHR0fA3xyRukitgHDYIVlZR5/8ADv4I3/ggfDsL/ZcA8P8A9qtqAs9y+e922/cnyDcS2GYtjJ55NX/Buk2/xD174h64yXltoWtfZ9HtbqzupbOa6itlkD3EUsTJIgMk0iLIrAsIgynayk503JpKW9v8tf0B30stL/o/z/4B1Hwx8Zal4+8Ovrd7pFtpFjc3Eh01Yb5rmWa1BwkkymJBFI2DmNWkAGPnPQdfXmv7Pfi3T/FHwx063tb2zuL/AEYvpWpWlqqRvZXULFXhliUDynAAO0qvBBAwRXpVXF3in5L/AIf5h1YVW1HTbTWLGeyv7WG9s7hDHNb3EYkjkU9VZSMEexqzRVAFFFFABXPaD8O/CnhXVtZ1XRfDGjaPqetSebql7YafFBNfvljundFDSnLMcsTyx9TXQ0UD6WMDwZ8P/C/w502bTvCfhvSPC+nzTG5ltdFsIrSJ5SADIyRqoLEKo3EZwo9K36KKd29xBRRRSAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigDlPHHwn8EfE42Z8Y+DfD/iw2W77L/bmlwXvkbsbtnmo23O1c464HpXUxxrFGqIoRFGFVRgADoAKdRR0sG+4UUUUAFFFFABRRRQAVyfi/4S+BviFqVjqHinwZ4f8AEuoWHFpdaxpcF3Lb/MG/dvIhKcgHgjkZrrKKOqfYN00wooooAr2en2unLKtpbQ2omlaeQQxhN8jHLOcDlieSTyao2PhPQ9L17U9cstG0+01vVFjW/wBSgtUS5uxGu2MSyAbnCA4XcTgcCtaihabAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABXm/hb4lXXiT4zeL/DCS2yaTomnWcot59Mv7W+8+SW4V5PMmjWCa3IiUI8LN8yyZPSvSK8p0r4Z+OLP426p42ufF/h+40bULKLTH0eLw3PHcLawyXEkIFyb5l8wNcnc/k7WCgBE60tb/J/fbQrTkl30t96v+FyXwv8AtLfD/wAa21xPoWpalqkUentqkRt9B1BvtlupUSNajyM3RRnRXSDeyMwVgG4qz8Dvjhpnxw8K22r2Ok61pEzQRzTQano99aRLvLYEU9xBEk+NpyY844zjIzS8A/A3/hB2+Gp/tv7b/wAIb4eudB/49PL+2ed9l/e/fPl4+y/d+bO/qMc6/wAJPAWv/DfSZNCvtf03WPD1p+60eG30mS1uraHexCzzG5kWdgpUblji+6Tg54uSSlo9LP8AN2+9Wf3lTUV8Dv8A1/X+Rp/ET4o+GvhTpVpqPifUH0+1u7g2lv5VrNcyTTeVJKI0jiR2ZikUmABkkBRkkA+b+N/2n9AuPhF4k8UeAtYt7+60+YWUN9f6Hqlxp8U7RJOks/2aBpPsxidXFwo8ohlIfmu7+KPwz/4WUvhwf2l/Z39j6hJf/wCo83zd9lc2uz7y7cfad2efuYxzkeZ337MfiC1+GXirwhoPjXTbH/hJI7W2vLzUdAkutkEOmW9iwjRLuLDsbcSBiWA3Fdrfeohyvm5/K336/h/T2FKyceXXRt/4r6L5rX5Wvrp6j8QPitoPw2tbQavLeTaheRyPbafpGl3ep3UoQDfILe1jkl8tSyBpCu1S6AkFhnmfh9+0J4Y1L4ZeCNb8Y+KfDfhvXdd0Ky1eaxutQjtMGaJXJSOWTeE3bgMk9MZJFM174T+NNa1LQtbi8c6bpfiSDTLnRtUvLTw8Wt7m2mljffbQyXTmCdPKG15HnTLNuicYAveE/gLpWg+BfCfh+91jXruXQdHtdIF1p+uX+mRziGMJ5hgt7hUDNgk9T0GSAK7cJ9VvL63e2luW1/td9LfDf/gGT5uRKO/n6L/g9+99bG3rHjK/l+IGi+GNEt7eYPbnVNVvrjLR29nkpGkYUjdLK+dv8KrFIxydqtm2firxPa/HW58Mahd6TdeG7rRZNVsY7bT5Yby2aOaGJklmM7pKGMjMCsceOBz1K32l3/hv4zWWuW+nzX+k67pcWi3k8O6R7KW3eaa3kcf88mE86M/VW8rOQxK0Lv4b+N7j42W/jNPF3h9NDhs30waM3hydrk2ryRyOPtX24L5m6MYfydoBwUJ5rgj8UX096/3NLv8A3Xp010vYVVPlko7+7b705f8Aty+4sJ+0d8P2uNQRtYuoYbK2uLw31xpN5FZ3MMClpntbloRFdbFBY+Q7nCk9ATXZJ4w0WbWLbSotRhn1C484JDCTJgxBDKGK5ClfNjyGIPzCvJdH/Zz1WOx0rQ9c8WWereE/D1lc2OgafBozW9zCktu9qhu5zcOtwUgkdBsjhDFtxBOMXvCf7OVrputR33ibULHxbA+jtY3djd6UoguLuWaKW4udryOArm3twIiDs8ofMeMXBRckpOy72169L+S0v9pa6M0lo3ba7+7S35t+i22v3sPxG0hvAmo+Lp2ms9G0+O7muXnQbkS2eRZWwpPH7piPbH0rz7xX8atc+Hnhn4faz4qsrPTP+Ei102d5Y29tPdT2tu9pdTwwxpEWaW6zDChCK25mZUQ5U1o2P7POg6D8K/FHhLRrTTdKvdcs7y1l1Wy06OBz5zSmLeF5cRCUAAkjC9BnFXYNB1D4o6b8P9Z1e3n8M6t4b1dtQvNNng3+ZMltc2joj7gDGTOZElGdyheBu46MRGhHEzWHk3BPS6tpftd/5/e0TTb5F7TdqX38vu/jfr+VzL8f/Hi0tvhvo/iLwjeRO+ranb2Fvc6tompyW0J+1rDcR3SwQmS1kH7yMeeIwsoCvjBFdJ8S/jd4Q+Eag+Jb2+ib7NLfSR6bpN5qUkFtH/rJ5UtYpGiiXPMjhVzxmuM8TfAXxNdfD+88L+HvGGk6Wl94hu9eubrU9AlvSfN1A3scSKl5Dt2OQpYltwHAQ9PKf2mvCfiDUPG2hzax9suZIvDTW0+oaN4R1/UrDVZmmJmtHttKvkaOIhYyY72SaJw4C/dlLcDk+ZJLRv1fw3t96t530Rtyxu7O6Xyuua1/ud/Lqet6b+034eHxA1PQtUnuLS0ZdMk0110a+8yNLuMFGvT5ZFoGkYIpnEXIZeSrY9prxmx+EGqeMPBetza3JB4b1nxVZ6TLe6ZawiaLS57eOMyQowcCRdylQRjA9a9Kl8C+G5vE8XiSTw9pUniKIbY9XayiN2g2lcCbbvA2kr16Eit4xjGc41G1a9rWd9XZbr7/AMDmjJyipLsjkfDfjLxIvxp8YeFtcvNHudEstJstY0+WzsZLaeBJ5rqNo53eeRZSotgd6rGPmPy03Rv2jvh/rzzi11i6SOOD7VFcXekXttBe2/mLH59pLLCqXUIaSPMsBdAsiMW2spObF8JfGc3xe13xVqHi/Qbnw/rGnJo02jQeHJ4rlbON7l4gt0b5l80G5bc/k7WCjCJ1rGT9nXxJq2kyaf4l8c2WsxWHh668P6E0GgfZmgWdI0+0Xn+kMLmVRDFxGLdDmT5BuXZK+Fen+f8A9r663a3N6lud8m2n5Rv8r8z8uiex6rJ4y03VLx9I0bV7KbW5ra7ktsK1xCjQSLDLv2ED93M6oyblbIYcEHFX4c+Oh418J/2leWy6VqNnNNY6pZtJuW1uoHMcyhyBuTKllcgbkZWwM4rmvA/wrf4Z65rXiW71iXWll0yCM28Fg3mrMilrqVAruX891R/LC7gwb5n3DbX8O+B9f1n4b31pqsMGmXvijVJr/VrOZiWtrKaTLWuUJDS/Z1SFiG27i7AnAzrRjGc1Gbsur7e9pbbTlu35pbXsZyel0uune3Ld/NOy+fXc37LxZ4q1bwdYatpPh+z1G91KZ5be3vbxrCKCzJZonmfZK+8oEyqxn5nwQoBIyrv4vXMXw/8AC3jpdOji8P3jQnWIZH3S2cMpCCeOQHa6RyEFjjDRlnUjaFfa+KPgrXPGvh620rQdftfD0PnqbxLnT3uo7u3CkG3ISaFkRiVyVcEhdvQkHA8faD4g1L4d6f4IeG11O/1qVbC8vtL097Kws7ENmZyjSyFCIR5aKHJaRlIAUMU66zoSpc8UlJy2966XZ/Ztts0+++mfvq9uz+/879t+is9jtfEPjvQ/CuteHNI1S++y6h4hu3stMh8mR/tEyQvMy5VSFxHG7ZYgcYzkgVv15/8AEJvHN14z8C2nhPzLLR49Ra68Q38yWz20lisMim1wzecJXkaJlaNdoEbbnH3H9Arzlqr+f+X9f1rq9Hby/wA/6/qyKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAK4LRfE+p3vxY8TaXONSt7Kx0y2mt9NuLa1EUpaWdTcRTJMznf5e3ZKqbdgP8Rrva4mz+Gs1n8QrnxafF2uTzXES28mmyR2X2UwK0jRxcWwk2q0rkHzNx43Mw4r0MLKkoVVUaTcdLpvW6elk7aJq+nqTLbTfT8/8AL+rXOX+G3jDxR44+IHiCfUbDxJoOkabdPaR6fP8A2U1kNsMTbZWjeS4MxMjNlG8vAUdcgu0L4tT6X8K9W+J/iV7ifQLtlvdO0zS7YTy29gzrHARt+aaWRSsrAE43hEB25f0XQfDNr4fn1mW3eaRtVvWv5xKQQshjjjIXAGFxGvXJyTzXA+F/hbdTfDK++HXiFWi0bT5ktdM1CwkRWuLFHSW3BUg7HjAWFgRhvL3LjdhYzCvDEVG6EFBKKSSVlotW73u79Xq9SaceVPnd3zP7tbfhb5k2k/HrT7y3can4b17w3qcesWejS6Vqa2rXEcl1s8iUmGeSPy2D5yHLDawK5GKxPHH7V3hbwFqN3Z3mi+Ir57Oe8huG06zjmESWotGnmP7wHYq3sbAAFm2OqqXKK/SeMPghpfi+TXLgaxq+j6jql5p+oLf6c8Hm2dxZlTC8QlidD935hIrg5PArmT+yf4Znt7tbzXvEeo3N5HfrdXt3dwvNM94LTz5CfKCgk2UZVVARN7qqhdipx09W/aab/lHl/Hmv6q3k5cyacf60l+vJbfZ383yftR6T/YPh27t/CfiC51jXNVvtHtvDrTabbXoubSR450LT3kcDENGcKkrOwOQuFfbf8car4i1L4yfDbQrDxDqfhXTLzTdS1XULOzgspJLpreSxCQStLFNtTE8obymVjnh+Aad4h/Z7tNa8I33hm08Xa7o2i6jfajfajaQ2ul3cd8bydp5YpUvLOdTGrO4UKAcH5i3Bql4i8J6p4N+J/wAKrrw74V1LXfDei6Pf6DM1jdWoawSVrAQySfaJ42dAtvJkx73+X7pJFVC3Om/691/hzW3+dra9Hu87t8Pvfrb52/Hu3pR8WftdeGfCPivVdCn8N+KbyTTZJo57y1soTbZgNubgh3mUlY0u4HLYwQ21d0g2V0ll4y1m6/aHu/Dtx/a2maLBoDXMFpc2lk1nfyCeMNdRXEc7XClRIIzFJEgPLDOATT8Rfsz+GPE2pave3V/q8cupi/EywzRBV+1izEu3MZxj7DFtznG585yNulc/Bu6ufixF47/4T/xPHLFCbRdESLTfsH2ZmR3g5szNtZo1Yt5u/qAwHFRH/l3fez5vVxa+69vxe6RzT5mp8vdW9FL9Yr73ba5zOk/HS/8AG3xU8E2WiaJrNl4M1WLUJo9cvEsxaauscamNoFEzXCrklgXijDDBG4EZ1vAevav4d8XfEuw17xHrHirTtHmtLi0afT4ZLmCOaEu0Ucdnbo0ig9Mq746k1J4J/Z40nwPr+i6lB4i8RajBoSXFvo2l31zD9j022mGGgjSOJC6qAoV5TJIoUDfgkGPwH4dv/E3i74l3vinwZcaVouszWlvBY661nci9ihhKM5SGaZdjHoHIbHVRUwSVSPPe19e9vyv/AEtLs0e77X/r5f09Tq7X4oaNeaTfajHZeIlt7MosiS+GdSjmbecDy4WtxJIPUorbRycCuL+K3x5vfCnw5fxH4U8Ia94hmXUbWyeC6006cYhJNGhdkvpLVmBD7VZNw3suflDEdpa/CHwJY6TfaVbeCfDtvpl8Ua7sotJgWG4KHKGRAm1ip5GQcHpVfU/g34UuvA+q+FtL0m18L6dqBErtoNrDavHOpVo51AQoZEZEYFlYfIAQRxXpYn6j7Kf1fn5r+7fltbS97db326WM4e05lz2t13OZ+If7Rdl8KfAeleJvFfhbUtA/tC6a0/szVNZ0OynhYByC0k+opbtuVCQscztg8qMHEl1+0VpNxY+FpvDfhvxD40n8S6I3iDTrXRIrYO9opgDFnuJ4o1YfaIztL5OGC5IxUms/AmbXZPDd9c/EPxYPEWhx3dsmvxLpqXVzb3LRtLDIgs/JUHyogHiijkAQYfJJOv4K+DOheApPCjaZPfsvhvQn8PWSXEyuGtmaFt0h25Zx9nQbsjq2QScjgVm3fv8Ao/15fPfybuXw+7vp+av+F/wOT8cftR6Z4NkcQeC/FniBLdNON+1jFZ2x0+S+kWO1hmS8uYHDszqCFVgmfnK4NR6h+154B074xQ/Dh7tX1hr2PTZZ11LTgIbt1BWE2xuheMcsg3x27Rgvy4CuV8z+M3wd8b6x8dtT8XeHvDN9Lr6LZp4c8R29p4eurCz2ptY3r36NqESq7SFksiAYyCn713Ne86b8JZtF8bXWvaX408R6Zp99dm/vfDUJs5dOnnZArsPOt3niDbQxWKZF3ZbALMWIdJSWl369LfK19bXvbRq4pXUeVfFbfpt273t1ta+2hzXw1+N3ijxp8UPGfhnUPhxrml6Zo+qLYxau82neVAhsoZwbgJfPIzO0hK+VGQFkjD7WD7eh8N+PJtH8SeN9B8VahAkfh+KPWY9YuSkEX9mXHnFGmOFRDE1vcRk9CkaOTlmxetfhgNN+JF74t07xLrOnRaiUk1LQIUtGsL6ZIRCs0he3adXCLGP3cyKfKTIPzbm+CPDeqL4x8V+K9ahWzutUaHT7OyBVmhsbYy+UXZSQXkeaaXAOFWRF+8rEitypdeX8dNe2u/p8zWTje9u3/B/C/wA7EHwdvdb17RNR8QazrFzqEeqahcvp9o1vDFbWtmk8kduYCqCR1liWOQvI77i+5dqkKO/rz34Nafrfh/Stb0PWdDm0wWesX09pffaIZbe+guLqa4R4gjmRdqyKrLIiEMDt3Lhj6FTla+m39f0zKN9ebe7/AD6eXby2CiiipKCiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKAOc8f8Ai5vA/h0anHYNqUjXlpZpbLKIyzT3EcCncRgYMgPPp1HWuP8AFHxa8Q+DdHthqng4HxBdXjW9tZ6bc3V/ayxLGJGn8y3tHnVRnZg2+d4/unfXoeuaDYeJLOO01GD7Rbx3EN0qb2XEsMqyxtlSDw6KcdDjByOKpeLPBOkeNre1i1WCZmtJvPtri0uprS4gk2lS0c0LpIhKsynawyGIOQa9bCVsFBQjiaXNq7tXvaysl7yW976bbMiSk/hdjzfxF8TPG95p/wAO7/QvD1rYNrOpta3tjrV3PZyblguT5WHs2dYiYvMWQqjkKgMY3tt6DXvitd+GbXxZ9v0WE3vh3w5Drs8Vvel45XcXO6BXMSnaDbHEhXJ3/cGMHotS8AaPq/hu10O7GoTWdq6SQzHVLoXcbqcq4uhJ527kjdvyQSCcEisrxB8FfCPiqNE1awu75RYDTJRLql3/AKTbgNtS4/e/vypZmDS7mDEsCG5rtjiMrnyxqUmkm9lrbmur++r+7ePlvdkxU1Zy10X3316dV/W1oLz4geIbjxVf6TofhWDVINN+yveXE+qC3cxzDJ8lPLYO6AMSrtGp+XDZJC52j/F/U9R8SW1tc+GobPQ7rXLzw/DqH9peZO9xB553+QIsCJhbsMmTcGONpXDn0Cx0Kx07Ur2/t4PLu70RrPJvY7xGpVOCcDAJ6fjVCHwJocH2XZY7fsupzaxD++k+W7l83zJPvc586X5T8o3cAYGOaOIwPK4uj0STvK9+XV/Fb4rNaWte6Goy5NXr/wAB/ra/4EXxK8Yf8K8+HfijxT9j/tD+w9LutS+yCTy/P8mJpNm/B2524zg4z0NeS+Bf2kPFHjXxvc+GG8AWek3v2XUWtZLnX96S3Nr9kcRPstm2I0d9BukG4q6yqqSKqyP7Z4k8O6f4v8O6poWr2/2vStTtZbK7t97J5kMiFHXcpDDKsRkEEZ4Ncw3wT8H/AG66vYtOurO+uI7uJ7uy1K6t5lFykCTlJI5VaNmFrBhkIZfLypUkk+NGyk3Laz+/o/6+43bjaFlrd39LaW+e/l8jz34a/ES+8H/AHTtSuLXUtd8QXWuXWkQWeqa4b1pL6TU5oFja9aCM/Z1fo3k7kiUAI7KFZfhh40+ILab8bb6/0C31rxbpOuldO8L2OvtNasw0qxkjtobueGIRq7uWJMSqrSMcHknrNJ/Zu8B6L4P1LwvDaazPot/cpeSQ33iTUrt4bhZTMs0Estw0lvIJT5m+Fkbf82cjNVta/Z50O3+FvjDwX4QS28PReLN41W6v47jUmuTLCkFxK++dXad4kH71nJL/ALxxIS258yiqllq1ZeWsX+FnrrutN7qn0U/n+N+3lpprfXt6jZySzWcEk8P2ad0VpId4fy2I5XcODg8Zqas7w5odv4X8PaXo1o80lrp9rFaRPcSGSRkjQKpZj95sAZPc1o0StzPl2M435VzbhRRRUlBRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAV5TpulWXhH9oDVToujjHiLTLe41UaascKQTJJc7by6VmUSNKAsIdBJIfLUMAihh6tXIaD4L1Gx+IPiDxRqerwX3262gsLKztbI262ttE8rgSM0jmWUtM2XGxcBQEBBZkvjT9fy2++z+XoP7El3t+a/S//DXOvorK0HxZonipbttF1nT9YFpMbe4NhdJP5Mo6o+wnaw9DzWrWk4Spy5Zqz8yb32CiiqmratY6DptxqGp3tvp1hbp5k11dyrFFEo/iZ2IAHuTSjFyajFXbGW6KpatrWn6Bpc+panf2um6dAu+W8u5lihjXplnYgAcjknvVePxZok2hQ63HrGnyaNPs8rUVukNvJvYIm2TO07mIUYPJIFPklye0t7u1+l+wrrY1aKKKgYUUU2SRIY2kkZURQWZmOAAOpJo2AdRWD4N8e+GfiNpJ1Xwn4i0nxPpayNCb3Rr6K7hEgAJTfGzLuAIyM55FaeratY6Dpd3qWp3tvp2nWcTT3N5dyrFDDGoyzu7EBVABJJOABQ/d3Be9sW6KitrqG8torm3ljnt5UEkc0bBkdSMhgRwQRzmub1P4reCdFtNCutR8YaBYWuvOsekTXWpwRpqLNjatuzMBKTuXATOcj1o20YLVXR1NFFFABRRRQAUUUUAFFV7vULWwa3W6uYbY3EoghE0gXzJCCQi5PLEAnA54NU/E3ijRvBeiXOs+IdXsdC0e1CmfUNTuUt7eLcwVd8jkKuWIAyepAoHZt2RqUU2ORZY1dGDowyrKcgg9CDTqBBRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABXgPgn4jeK9e+MU+n3niXw/aQx6jeWs3hafVolvVt4/METx2Zs1m3sqxSbzcvGys5CgFQnv1FejhcVSw8KsalJTclZN/Z81o/0fmZzi5JJOx4/4M+JHhnVPH3xJh0nxXoNxf5hnjVb+KQKI7VFeRgrZ2Iwwx7YIOK5zwp8VtYb4c+NLyz16Txjrek28MjahHd2V9paM+7cbeayt0chFBd0lh8xcLhWVgW+g6K7v7Qw2t6F/g3ab9yyevKtJJWf6k8km077X/HueXfAfxJq/iTSdVk1LxVofi2CKdFgu9I1iDUnjJXLpK8FpbIv8JUeXu+Y5J4xF+01rFhovwh1ua+8QWuhHyZDBDerZyRahII2ZbZo7mN1kDYztUBjt4I5r1aiudY2msfHGRpJRTT5Va2nTZr8AVOSpuHNrrr6njnxUji8faT4J1Pw940uYdJn1q0jjl0M2VxbysJDiUPJDLlkZeMHbkcqSOL/iTThe/Fr4b6Fq95LqdnZ6df6qjXiIDd39ubWOKZwiqm9FuJnChQNxDKB5Yx6Zdaha2MltHc3MNu91L5MCyyBTLJtZtig/ebarHA5wpPasnxV4NsPF39lSXTT213pd7Hf2V5aPsmhkXIIBwcq6M8bqQQyOw9xhWxLqU6dJRtGLbW19fNJNtb+q6DULNt6txt+fro72+/qeC+MPHXjrTvHHjnWLbxhcW+h+GvF2haNbeHY7C0MFzBeLpy3AnlaJpjj7W7IY3Qq2dxkXCrJqH7Qlxb+J7jwoviWE+K4fHB0+XSY4I3uYNLaMvG0sYQmOJgU2zOAGLKockgH6UorzJR5uVX2/H4H/AO2v5Sa9emclKNkrP/gNfrf1S7Hxh8GfjH8S9Z8UfDu41/xwNS0zXDYm7sW0u1giH2nQrm+fLqgcKklvGVwwPzS7i4ZFj7n9nr4o3njqy8Y2GpeOx4/1K208T3F7omp6TqWg27N5n7u2ms7eGdGIwTHdKSABseTazn6PvryHTbK4u7h/Lt7eNpZHwTtVRknA5PA7VTsL+w8YeGre9sp2n0zVLRZoJ1UoXhlQFWAYZGVYHBGR3FVW/eRmoro1bte9v+H3dgU1GcOm3zs9fwaT+Xc8Et9c8QaP+zH8F4PDetNoF7qdvouny3yWsVw8cL2nz7FkBXfhflYggMASrDKnzW1+LOs+Iv2ZfHGo/EH4lW+lvcaBpuo295eWelwwPLdab5xsPKuYHiljkcEbHVnbcwVgMBfov9nKLQ7f4T6VbeGfF+reNvD9m0mnWWoazbwwzRpbObYwYit4AVRoWUMyFjgncwINem08TH2zrcsvjldPsrbffr2v3Td8cNejGmpLWCSfm1f+u+nRnzL4p+N+kappfw7t9J+J9n4Q8Hatpdw48TeGxY3Ru7+DyI49Nt2ling3t5kx8lI2lYwhUxtYGHQ43h/YU+HscokWRbDw8rCZNjgi7tc7lwMH1GBj0r6grjfiZ8Nh8TLHT7CfxFq+i6fb3cN3cWulra7b7ypUlSOVpoJGCBo1P7oo3X5qdSXNGSS+Jr83/n06Lqy6fuRUb7fp/XW/qloZvjLVdTg8daJaapcT6N4Rklj8q909iTd3m5fKt7mQYNvGzdAAVlYKjSLu8qbyu1t4/Cf7XWvLrnxd1TRpdastLl0rQr9tJgj1cCa9DWcO+0EsqxZUYifzR5o3u2VNfTFFdNSspqmlG3Krevn69279lZaGco8yavv+H9eVtdbnyH4T/aF8R61+05ZaPa+I47zw3f69qOh3Hhq71SzlvdN+zwXLrNJZxabHNbK72hMbTXkm+N87Dn5Oz8H618RB8Fbn4lzeOX1m+m8NXd5HoN9pNuthHdLGWt5EMEazjlP3gZ5A29iqxgAD6JrE8b+D9O+IPg3W/DGrrK2l6xZzWF0IJDHJ5ciFG2sOQcE81wzi3BRTs0rX6+vqdEZR9rzNaX2+d/y0+XmfHPgv4/eMtR+HvxE8U6T8Rv8AhJ4tOs9Nt4rPVtX8PXUulSTTlby/lbRre4WKCOMho3l8zHlTNJEVXJ29VvvFnxU/ZN+IE1x8T9N8QnT5JZI9W8E6zp2uNNAkSPJb3Uy6bBECCxfEUCOFCZc5O73LSvBulfB68uPGPirxprnijUJktdEi1TXYbXfbxy3KpFBHHZWsK4eaWPLMjHplgor1NmCKWYhVUZJPAFbyaacrf5K3/A1fra7VjnUXKNmu/wA9Evz2+Tsm2eQRr4Z8d/AHXNvxG/4WNoht7ib/AISv7TYyNFJF+8WRZLKKKFXgdFZWVQysgJORmoNV1LWNe0/4N3OvWtvFDdXNneXlxG5My6gbV2WIQ7dojJMhMnmFlKoAjBi6drdW/h343eCLV4b2bUPDF9Ksx+z7o4tQijkPyMWUF4HKg/L8sqY5aNyGt/EHwHD8QNDgsW1XUdBurW7hvrPVNJaJbi1mjbKsoljkjYEFlKujKVY8dCM2nzar7UHr05Xq156/fFXuirXVr9JLy961r+lvubtqdPRTY1KRqpYuQMFmxk+5xxTqYwooooAKKwfCvjjRPGzayui3v21tH1GXSr7908fk3UYUvH8yjdgOvzLlTng1asvE2m6jr+p6Lb3PmanpscMt3B5bDy1mDmM7iNpz5b9CcY5xxRuk++vyewrq9vl81ualFUNd8QaX4X0m41TWdStNI0y3AM15fTrBDGCQAWdiAMkgcnqRVyGZLiJJYnWSN1DK6HIYHkEHuKrldua2gx9FZMPinS7jxTd+HI7rdrNpZw6hNbeWw2wSvKkb7sbTloZRgHI28gZGa+seONE8P+JvD/h6/vfI1fXmnXTbfynbzzDH5kvzBSq4Xn5iM9smpB+7o/66/kb1FFZfhvxNpvi/R49U0i5+12MkksSy+WyZaORo3GGAPDow6c4yOKBX1salFJVXStWsdd0211HTby31DT7qNZre7tZVlimRhlWR1JDAjoQcUDLdFFFABRRVVtVso9Ui01ry3XUZYXuI7QyqJniVlVnCZyVBdATjALL6igC1RWbF4i06bxDcaGtyBqsFul29syMpMLsyh1JGGG5SDtJ2nGcZGdKjzAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKK5jQ/HcWseMtd8NTaVqOl6hpccVwr3ixGK8t5C6pPC0cj/LuikUrJscFclMFSTrYOlzp6KKKAOM+LUerT+EYotFkvobyTVNOjeTTd3nJC17CJmGAcKIy5YkYC5zxmuD+KnhPWfCvh7TNP0HWvEn9h3GpNLfzt/aOs3Nsnk4RFNvcR3piMi7iVkchmAP7vIX2+ivWwmYTwihGK0Tb6a3SS3TV42vF2dm/UiUFPc+f/FXgdr7wP8O9U1fUPFGvJo+oh7y501NXsbs2rR3CK32RJGuC6GSFS7bpdoYsxDPl/j7XNb0Wx8U2+g2njC7t9S8GwReHvs9tfTTJdoLreXdxugn2vASZisjFQPmcYr32iuyOcSuvax5km3ZvTWXNslZO+9rJx0SW4ow5bW6JL7nf/gHmK+Db/wAU+PNdm1HVvEmn2dmLCTTfsV/NawLIELSnapCTAlVDJIHTj7oLMTz+g6H4ksfEml63Ld+JJ7m48X6laXFtdXVw1rFpf+mGL9wT5Yj3LCyyld3zKobaVWvbqK545pUUXCys0lbsuWz+/d+YKmlDlf8AWjX4Xv6oyPGEEl14S1uGGNpZpLGdEjjUszMY2AAA6kmvmv4D2gtfDKTSn4lLocPg9LbxLFrNrrVrP9vVIgo061ZEkjZUFwD9gjVeY9uWAx9V0V4kVyuUu/8AlJfd71/VJ9C5e9y67X/Fxf8A7b9zZ8peHda8SeKf2do9C0eT4h6drlt4ptbSW81LTtUttTXTZdbADrPexCSULZnLOS+xR8+MVX8eWvi7wr4P8TeFbSHxg+gL4yjtLLXJH1/Vryw086dDcGYGyuY7+7jN0ZIRtm2oZPmO2PbX1rRV3Wul7/5xf4cun+JlRtGKj2/4P+Z8tfDvw5468aeEfhjo3i688X2Ful1rttq72lxfaZPPbRTyLYmebz3uI9yJE6sbgyNjBkcM+7ovicuufCB/DFhomt69rDeJrUeDLSXUL57qW3viHktb1t3BKRC5Msv33EUZYuRmvoOucuvh/ol943tPFtzBc3OtWdu1tatNfTvbW4bIZ47Yv5KSkEqZVQSFSVLbTipaUpuT2bu/ut+XXpe62BNxTt8vLW6+SdvVK11ueF/Eiy1Pw78Y/B8Onz+MPFUFpBp9lFosb+IrO3j2zHfezalbbrK5bbjfBeL823mZAdj5bWfjq+/acnm1TxR4h0SG31lBp1hb+GNcutNudNEanabqC8/sxd4Mqs9zbiZHJx0iavquinGTjJSvtf8AFp/fpa7vo2rW2iUVKLgtFZL7k192vS2qT3Pkb4U33j26/aMi1Ka38T2WkX15qtvq+k39jr32PT0Bka2K3F7ePZy+YYlZWsrdEUNs3gMqyd18O/h74u0/4TjxTa654ol+Jdz4euYhY+ItVnks3vCn+j+ZaTExxMjIg3IqMwLGQuSTXv8ARWbjeKim1ZWNIy5Zub1u76/qfJ2qeH7PxZ8F9V03w2PifJ4jm1Tw82qSeKIdZa4ikXU7d5ZbZb5DbjYFldzaL5ShVJGwJXQ6tomq+Ffhh8bNBuZfEFx4YsLtBYT6xeXd9cvpr2VpJeCO5mZ5ZVDNdgHe205UEBQB9IUjKHUqwDKwwQRkGtJSvGUUt0/T7O3kuXTsnYIvlioPW3Xr019dNzyX4oeE7rxt4q+HulWWoa/p3hKRbx9Qbw1fz2CSRrChgR57cq8a7sFSjoTtxnaWVvOLPxh428H6b4UbWNN8barLd+Eb/TIobHT7u6kfUkuIltzcFRthlaPkXFwY1+8TIOa+jvDPhnTPB2h2mjaNaLYaXaKUt7WNmKQoSSETJO1BnCqOFACqAAANSspU7zbTdnf8Y8v3pO67MmOiXN0t+DT+5ta91ofE3iLwv8UNck0WV9T+Idsr6dZ21xBY3d5bqv8AxT9zJMzFMMJGvI7ZWcncr4VSjSPv7a3uteh+MvhG4uV8YeI7+a302CfTWi8Q6XY6X/o4NxO88YOmXhyxZopgjhty+YzKkQ+o6K109rKp3d7fOT/Hm/C6s9TNQtSjTv8ADHlv321fnp/npoefeIPiVDceLvEvgGzsNWj1+18Of21Dexwj7PKsjSRKkUitu81XTJXaOGUgnnHgPwT8O/ELRPHWiXniDUvH1/p/nBLxdWuLySHy5NBt553MeAMm9LqMD926mOMJllP1ZpfhnR9DvtTvtN0mx0+91SYXF/c2tskUl3KFCB5WUAuwUBctk4AFadKPuqXVyVvxb/J2fe1+ti5Lma10X/2v+T9OZrzfyr4L8YazB8LfifpXhW3+IxvLPUZLzSZPEWkay18dNItxJ9mm1KLdLKB9oKRFmfdjC4Izc+F/hXVV1P4oXPgu38SaLpmoWunfYr74hW+p6jPfosVysojW8uVuY8MVULIVKkE+XtdWb6eorajUdGcai1aSWtne3e+97det36QoWkn2k3999F23fyt21+fde0fU9O/ZZNj4osP7e1o6AbTSbXSfDV1Jc2k72DRIjRq07rINzo0uI1G4ghc89nqXxf0v4f8Awb07xbe6Rr9xYQfY7GW1j0uWC8jaSaO33GCcRuEV3BJxyoJUNxn0+szVvDOj69eaZd6npNjqN1pk/wBqsJ7u2SV7SbaV8yJmBKPtZhuXBwSO9dmJxn1mL5o2blzdLWe6SS/UmFP2dOFNP4U199v8l+J87/HT4fa/rfxC+I/iPSJPFVpeab4BtpdEbQLm5tkutTil1GSND5JBuGUtH/o7FkYSjfG2UI2viF4sTw38cPh5qLW/j6SdoJItZj0nS9cvdJEL28gh3xQRvaeYJiuWx5ijG4hQK+gaK89S0S7f8G/5/cb1n7bXZ6fgopfJct/mz4qn8R3+s/tLk6/4i8Q6Rotj4wEEGpW+p+IYtPuUEYht9MkhggXS13XDIfMa5Z3KrFJHudox3vwY8B+IPBOseBtQD+Kd2qajrsGsWd9c3Jsra2M1zNbn7KSIYTuWPbKEDuJCCzbq9Uj/AGf/AARD4uPiNLDUBeG/OqfYTrV8dM+1k5Nx/Z/nfZfM3Eyb/Kz5n7zO/wCavRKmFo0oR6r/ACSfzdnd9eurd86keaq5J6dP/Am7em35bKNuA0/xfo/xavfiB4N+y65p0nh+7j0u/mWeSweVpLeO4V7a4glEgXZKvzBkYHPGOT8+fDfSpvhv+y3q+iaDo3j6fxnp2nWOlata6o3iN47eTf5Ms9hukVpI4lMsu3TXUuqIFZdyEfV/h/wzo/hLTzYaHpNjo1iZZJza6fbJBEZJGLyPtQAbmYliepJJNadSl7rXVpX+X6avTT5GylbTpe58g/DfR/iHqXgmbQ73VPGDaW/jezjttTW21XTbyXSJLSM3H/H9cXF5FEZTMpaWXehOUMREe3W/aGsdS+Fuh+B/Dehah4r8RaZcajezz6dJr+ttqE0QjLIn23ToLrUSkUkgI3KyYwskg/dq31RXI+PfhV4d+JUmmza1FqEd5pxkNpfaPq13pd3EJABIgntZY5NjbV3IW2kohIJVSFJOSa7tfgor/wBtvfu3bclatX6Lp31d/nfXyWtz5H1SHxPd+AfhzaaX8QfFd74Yg0eVZvEllofifWbu41hJ2S5W5SyvYb2JkYECO6V4wVdCqFAtelfEL4d6fpvxm8EeLfEsvj7VILnQprG71Hwzc68sf24SWRjL2djIfs0cixysylRESv7wlgpP0N4U8K6V4I8P2WiaLa/Y9Ns02RR+Y0jHJJZndyWd2YlmdyWZmLMSSTWtW3Nyz54929fO6++z0t2je9jOUfaJqXVW8tLPbTS6u15u1rnmHxISNvi18JjaSBNYW/vzKE++2nfYZfPB4/1fn/YSeg3CPvin6LrunJ+0R4o0MX3iaXVm8PafftZ3UgbRYYfPuY1a3UcpOzK3mbsbgkeM7Wx3UXh3TofENxri2wOqz26Wj3LOzEQozMEUE4UbmJO0DccZzgYTS/DOj6HfanfabpNjp97qkwuL+5tbZIpLuUKEDysoBdgoC5bJwAKlPa/S/wCtvzv+HW5cndW9Pzvr59Pu7WenRRRSAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigArzVozqH7QcsdvfSafJp2gWtxdR23lsb+OWa7RIpg6sVRGRnUx7GLZBYrlT6VXJaFdeBNH8Za1pOjTeHbLxZfOL7VLGxaCO+uH2qBNOi/Ox2lfmYHgjnGKqFOU5c0Y35U3p00t92v6dQbXK0+tvzT/S349DraK5Twf4s1LXNW8Sabq2mWmmXGkXKRK1nfNcpNG8YkVyWijKthuVwQD/EaueH/AIgeF/Flte3Gh+JNI1m3sv8Aj6l0++inSDgn5yjELwCecdDXRPC1qbacdrXtqve1WqutehPMtDformoviN4d1DwlqniTSNXsvEGk6dFLLNNpFzHcrmNN7IGVtu7GOCR1FS+A9Z1nxF4T07U9e0uz0bUbuPzmsrC/a9ijRuU/etFES20gkbMA5ALAbjzyjKE5U5K0o2uno1e9vyY7ppSWz/Tc6CiiipGFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFeNeD/g5qHh7x6b69s4tR06HVbzVLXUj4p1IPE85lb/kGsptgw8+SMsHGQS2ASVr2Wiu7D4yrhoThTek1Z6td+zXd6O6fVMiUVOyZ5p4e0Dxivibxhc6xonh/+y9aUFEt9anlfKQrEqOptEAVwMlgxK54DdaxbT4XeL9S8B+J/D2pahDp1ndW8MOlaaur3GorabCWYNeGGC4KSHapDM7IASr4IVfZaK6lmdWLvCKT9227tyWs1dtX0tfe111d17NO1/P8TxO5+Fd1F8MfGkV/pVzaatdWxmjSx8WalrL3Lwxs0KM9wqMyliVMBDRyKdrhgSteifDHxrpHxC8A6Jr+h6nZavp93bIRc6fIrxbwMOny8KVYMpXgqQQQCMVqeJPE2meEdKbUtXu1srFZI4mmZWYBpHWNBwCeWZR+NYEfxe8MzaPJqKT6gVjuvsTWX9kXn24TbBJs+yeV5+fLIf7n3Pm+7zWEqeKxs54lQlJSaWnM0ml3d3dprrskHu04xje1r/j/AMN+LOzorzfxT8dNE8OR+FJ4bPVdUtdeujbrJZ6VeStAojlZtyRwM3mq0WwwsFkHzkgbGroZviR4ftbLUrq4u5rWPTdLj1m8S4s54pILVxIVdo2QMG/cyZTG8bcFRkVMsBi4RjKVKVnto+/L+enqUpRk7J+f37HT0Vx+tfFrwvoGstpV5fzi9jaETLBYXE6WwlIETzPHGywoxzh5Cq/K3PythNM+LnhXWPEh0Kz1CaXUBdTWJ/0G4WD7TFv8yDzzGI/NAjdtm7cVG4AqQan6jiuT2nspctr35Xa299tvMXPHe52NFZ3iLUJNI8P6nfQqrTWtrLOiyAlSyoWAOD0yK84+Fv7Q3h34geE7e8867l1eHRIdXvYLPRr0xzq0YMj2J8o/bEDnaDbtLyVGSWGeFNNtLpb8b/5O5T0t5/8AA/zVu56xRXkngf442cXwVt/HnjrV9LsrRr2e2kvNPsb23hQfbXt4I2huIxOkv+rR1ZBiTdj5cVvWvxu8L3/hC58SWq69dWNrdfYri1t/DWpSajBNwdklitublOGVstGBtZW6EEva6fT/AIH+a+8Lrp/X9WO9orz7Ufj14K0vQtE1WbUb14da81dOtLfSLya+uZIv9bEtokJn81MNuiMe9QjkqAjYu6P8Y/CGvWb3Vpq/+jppC668lxazQBbNi48w+Yi4KmNw6ffQgB1UkZTko3v0/wAr/lr6FJN7f1rb82l6naUVws3xq8Lx61o+kxnWry+1WC3uYUsfD2oXKwxTkiJrl44GW1DYb/XmPAVicAHE158YvC1j40j8LTXV9/abzLbGdNKu3sI52XKQSXqxG3jlYFcRvIHJdABl1Bvld7W1/wAtyOaNua+h2lFcbpXxa8Pax40uPCsa6zaaxD5uP7S0C/srafyyA/kXM0CQz4zn907ZUFhlQTWd4W+P3gXxlqH2TStYmcG0kvo7y6066trOaGPb5rRXMsSwyGMsA6o5ZDkMAQcRdWv8/kXZ3aPQ6K8U8Z/tMaLafDfUPFfhw3Rs7C/0qGa+1zQ7+xspLa6voYHmglmjiWdRG7sHiZ1HyE5DDd1tt8bvDGpeAdf8WWDahcWWiCRbyyuNMuLK+jmRFcQtbXEccquyvGVDKNwkUjgg1VnyuXb/AIH+aC2ifc76ivONQ+Llt8N/Dnh9viLPHZeJNY8wRaXoFhd6g8kg+cwQRwxySzNGhGWVAWCO4RFBC7emfFTwtrEVlJZ6qJ0vNJbXICIJRus1KhpDleCC6gocPk/d4NS2k2r7for/AJa+mouiff8Azt+bt66HWUV5Lrn7Vnwu8O3EMF94lkE81pFfJDBpl5O5ilgNxF8scTHc8SuyJ95vLkCgmNwvSQ/GLwxdeKrPw9aPq2oX91HFKs1hod9c2cSyJvQTXccLQQsUw22SRThlOPmXL6uPVf8AB/yf3MXMrKV9Hr8u/wCK+87aiivNfh94211ta+Idh4x1LRJLfw3ex+VqFhZSWESWr2qT5mEtxKNyByC4ZVIGdq1Lko7lJN7HpVFedaL+0F4G12SWOLUr6zkSS2RY9V0a9sGnFxOsEMkInhQzRNK6J5se5AXXLDIrrdN8XaNrGqvp1hqEN7dLbLdkW5Lp5RkeMNvHy/fjdcZzlTxVdvP9NWLv5f1+psUV5p8ZPi4vwl1DwK9yqnS9a1mTTrwraTXVxtFjdTotvFDl3laWGNQqq5bcQFyQRl/EH47Wtl4E8Na74UvIT/burWdjBdavoupyW0ateRwTxz+TCXtZvmeNRcCMCXCvjDCnba3VpfNja5Wl3V/lr/kev0V598Svjx4L+Es7Q+I77UEmjtDfzx6Xo17qbWtsCV86cWsMnkxkqwDybQxR8E7WxzsP7SGhaf8AErWfDWtTTWkKf2abBo9JvHeNbqMbWvGCMtqGkYIpn8oEhl5KnE3Wnm7fPX/Jifuuz9fl3PY6KK4Xwj8RdQ8RfEPxR4Xv/D50b+yLW0vIJnvFmkuIp5LiMF0RdsZzbMwAd8q6k7Wyo3hSnUjKUdoq71W10vnq1sS5KO53VFZ3iDXLfwzot5ql3FeTW1rGZJI7CymvJ2A7JDCjyOfZVJ9q84+Ef7RWjfFb4YxeME0XxFpix6dDf3dnJ4e1JiPMQtstma2U3uMHm3V88cfMM4738t/nf/IvldrnrFFedr+0B4HHhd9fn1K+sbJL9dKaC/0e9tb1bxkDx25tJIVnEkishRfLy/mJt3F1yyT9oPwTH4Zstc+16rJBeXUtlBp8OgahJqbTxZ81DYLAbpSgUs2YhhSGPBBKukm30/4H+a+9dw2tfr/X6M9HorzKL9o7wLceE9D8R211rV5pWtQtc2ZtPDepTzmBW2meSBLcyww56SyIqEEEMQQaz/EH7Svh/wAPfFTTfBj6V4hv1vdNfUF1TS/D2pXsB+eARhGgtnSSMrPlpVfZGVCtgsKpRblyddfwu/0Jk1FNvp/ml+bR67RXHat40uvDvxI0XQ9Qhh/sjX4ZItNu4wQ6XsSNK8EmTzvhVnQgDHkSBuq1pWHjvQ9T8aat4Str7zPEGk2tve3ln5Mg8qGcyCJt5Xa24xScKSRt5AyMnRP+v6/4fYfqb9FFFIAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiivONP1+/8NfGLUND1rxGbjStbtYrnQ7fUkghcXIM32i1tmREMqrFHHJtbfIuXYsVwFXVLuPo5dj0eiiimI5/xv4V/4TLRoLD7V9j8rULK+8zy9+fs9zFPsxkfe8vbntnODjFYXxY+FcHxOt9JL/2U9xplw00UOu6Smp2UoZCjLJAzISeQQyupBUckEqe9ortoYyvhnGVKVnFtrbdqz+9IUkpKzPP7r4Yzr4P8O6bpNxouh6poV2t9aSWOjmPTll2yI4Fos4IVkmk4EuQx3ZPQ5Hjj4O674yXVyniu002TXtBTRNXeLSC+/Z5xWS3DT/ugWuHyr+blcAMp+avV6K3p5niqcueMle7esYvd36p9dbbJ6olRSsl0SX3O6/EwNE8K/wBj+I9b1X7V539pLbL5Pl7fL8pCvXJznOegx71gWnwr+y/2V/xNN32HxLeeIv8Aj3xv8/7T+5+9xt+0/e5zs6DPHfUVgsZXje0t7LZdFyr8Hb/gjUVGPItv+A1+TZS1zTf7Z0W/0/zPJ+1W8kHmbd23cpXOMjOM15X8Pfg/4w8IaZatqXjHRdW1zR9C/sHQrm38PS21rbR4jzJcQG9d52LQw52yxDCkAAsTXsNFcS0ba6/8Ff8Atz+8bSla/T9bf5L7jw7w38AfE9v8PR4V8Q+MtJ1SOHxHbeIbe50zw/LZFWTUxqEsTq95NuDvlFYFdgOSHNV/iN+zDL45/t1k1zSpY9U8Tx+I5NL8QaF/aelzhNOjshb3Vt58XnqDEJ1O9drqnynbz7zRV8zs13/+1/8AkY/d5sqLcVyrb/O/+Z438K/2dIPhbp/gyytdWhlt/Dd1q9xHDbaalpE630zyiNI0bbEse/aAoIIUcLVH44eH7n4hfEbwT4c0i21OGZJXbxBfnT5lsG0N1JntTcsgheSWWG2Ty1YyKpZsKvzV7lRU39/ne97/ANfr3V11F387/ieRfEL4Jan42+JOkeJbPWNF0CKy+zrJeWOkXMWuyxRymRrf+0Ir2NTbucZglgkjJySrHBXFtf2XLDT/AI0XfjiG08D6hFean/a0kms+DIrrW7ebYoIt9SWdCihlDLvikZdzANt2hfd6KUfdakt1f8Wm/wAUgl7yal1/S/8Amz53+H/7JQ8F/Fiy8b3GsaFe3tneahci8tfDKW2q36XYlyt9fGd3uGj3oEKrGoCkFD8hT0vQ/hJZWfwXHw41O7k1HTpNJl0i5uYk8h5YpEZHKjLbThj3OK72ip5U0o9lb5Di3CTnHR3v8zySf4T+M/Engt/Dfivxno+pwQXul3Vldab4eks5lWzu4rhhPuvJFkaQQqu5FiCksdpGFE/iP4PXlxpPxPfTtShk1PxVdwanZpcQlYrW4gtbaKJXIYl1L2qsSAOGIwcZPqlFXJuUXG+9/wAbf5L+rjjLkjyrb+v8jgtP0mb4gal4I8Z3FveeHrnS4rvztG1C3AmWSVBG6MwbA2Mhwy7lcYKkqQTws/7OWvWFnoMHh7xxb6Q9rpF7oWoXFxon2l57a5nWYvb5nVYZl2kBnEycgmM4wfd6KzlCEpc9u/4qz/DQmN4/h+DTX4pM8Am/ZT81NLX/AISjH2GG3i/5B/3/ACtGuNMz/reM/aPN742bec7hpL+z9q8fjvw7rMHiDRdNsdKjsopptJ0W5s9Y1CO2iCiG5vUvhHNAzbj5Utu6qrELhwJB7bRWl/fdTq3f53b/APbn95KilCNO2kVyr07fgef6GPHF58ZPEN3ePLZeAIdNgtLLT7uO2LT3wdnkuoXiZpPKKMsZWbYd0ZITHzNy2mfBHxXfal8Q4vFvi/RNW8PeNbaS2vLLSPD01hcwg262ylJ3vZl4jXnMfLHIKj5a9poqOVWXlf8AH+tO3Q0jJxlzRdndP7v+G+b1Z4hdfAnxb4kt7+88T+ONK1HxJHpi6bo1/YeHXtbeyxPFOZpoGu5DPIZLeAnDxqAmAoyTVqz+F2p/CfSvHGr+HHvvEOqahYW1lpVjp8Fsl1aCMSYIa5nSKU+fcTzneyDDFcMfvey0VT1TjfdW/G/y1e61FG0XdLrfye2/3L06HmOlfDjxJrmh/CW78Y67b3XizwpIl/q1xZ24MOoXZ0+e1m2Y2BFL3DOCE/hxtXPHM+Jv2fvFF38OovCXh7xnpGlwN4iuPENzc6n4flvWd21X+0oYkCXkOwI/yMTu3ryPLNe6UVpKpKTvtrfT1uOMpRSV9rfhf/5J/f6Hxf8AtI+EfEt/8R1uNQtLjULpfDFpbebYeE9e1DTPEFwstw0lpLDp17GkMO4x5TUHnjInOMASl/d774P3ni7RfFMuo3sej6h4qh0yS5tYoRMlhLbqheMMHHmAlSM/Lj3r1misXFOPI9rt+Wt729b69NFZLrEvelzeVvwS/TT1e+lsP/hBfDf/AAlH/CS/8I9pX/CRY2/2v9ii+1427Medt342/L16cVynh74d+KNK+LOteL7zxLpF3Y6pawWMmmwaJLDKkMD3DQYmN2w3g3Db28vDbRhUr0eiu2OLrRi48101y62em9le9vlYlwUtzz74XDxzdav41vvFplttHvNUB8P6XdpbC5srRYURxI1uzoweVZJEJdn2Ou7afkXi/wDhnnxFcfBfUvhld+MNKl0COK2tdGMegSq0drDKH+z34N4wvI5ERIpAgg3qZBxv+X3WiuO3u8vkl626+vn5vuzVSad13ueC/Dn9le3+H9i8NvqOj2Rm8T2viaSz8PeH00vT43itkgMMFusreWrbA24s7ZLZ3ZrK/a+0HUNavfAclt/bFha2txeSSa1o2k6xqUlpIYlVEMWj3EF4PMVpBvEnkgIQ6szRlfo+ilKPMred/uSX5RX3eolo7+Vvzf5tnyhpv7NU/wAV/BfgjXtT8NeA9H1bT9KfSRovizwP/atitmk7m1mhtZbiKeylaPDNG0rlfMKOCyBh7N4l+F+st4m8K694S1rR/Dt7o2nTaRJa3mivd2ctpI8DlIoo7mEwspt1CncygMQVbgj0mitL2d46at/fe/5v7yHHmXva6W/L8rK3XRdTz7xRpl94q+KnhCCOykj0nw08utXN/MjKklxJbzWsMER6Ods07uR93bGMZf5V0BvHN58Y/EtzfeZY/D+HTra10+yuktjJcXweRprqJ4mZxEUaJNs21tyEhFHzP6BRS7eV/wAb/lfT5PfUp+8rPy/B3/4f1fcKKKKQBRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRXn3x20m5vvhrqt/p+o3mlavo8banYXNtqj2CCeJSyiYh0jkiP8Uc26M/xDjg9Q6HoNFUNC1q38RaPaalax3UVvdIJES+s5bSYA/34ZVWRD7MoNX6ACuA8PWGs33xg8Ta3eaLPpWkRaba6VZz3U8LteOks8kksaRu5WLEkYHmbHJVvkAALd/RS6qXb9dPyv94/suPe34O/5oKKKKYgooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigCj/bEP92T8h/jR/bEP92T8h/jWPRXf7GBze0kbH9sQ/3ZPyH+NH9sQ/3ZPyH+NY9FHsYB7SRsf2xD/dk/If40f2xD/dk/If41j0UexgHtJGx/bEP92T8h/jXDfGWx1jxt4FvPDmhQWJOrEWV7dahcvF9ktnyJJo0SN/OkUfdjYxhs8yLjnoa5Dx94h1nSdQ8M6bohsYrnWL6S0a41CJ5UhVbaaYOER0LHMQG3cMgnkda3oYVVqijH117JXf4L17ETqe61LZ/qej/2xD/dk/If40f2xD/dk/If414V4s8d+P8ASdQ0zw/pmk2WqeIvsb3t3LZQRyQyIJTGuyKe9t2TPyliGl2FgvzZDGTVLnx1efFbw9Fa6xpuiWt3odxcvpV5p8t0InWS0EiuyXSJI4LkLIAAoLDDbs16McnTs51Iq6k929I+STettNCJYhxvpqrfjb/NHuP9sQ/3ZPyH+NH9sQ/3ZPyH+NeNa18VLzSZPEEZFgj2PifTNDtll3AyR3Isy5I3cuBcSkYwPkGQcHOj4a1zxd4i1u7uFk0WLQrPVLnT5bUwTfapI42ZVlWXzNqtnaChjIIBO9c7Ri8pnGn7WdlHzfkmvzRpKq4/16/5M9U/tiH+7J+Q/wAaP7Yh/uyfkP8AGvEPhn8RPFHiG68Kvr66R9m8SaRNqUEOm28qPatEYBhneRhIHE27AVdhG3L/AHq9SvJ3tbOeaO3ku5I0Z1t4SoeUgZCqWKrk9BuIHPJHWscTl7wtT2U7X8npo2n9zTX/AANSlUkbv9sQ/wB2T8h/jR/bEP8Adk/If4151fNe+NPB1+l1Z654OkO4NGbm3S6ZAMnbJBJKEVuV3K6yDkgqcGuK8N+Ltc0/9ljw14jg1rRrbWx4esLqTVvGF06WQZo4zJJcyg7uQW+bP3iM1EsFGFKVRv4Wlprum73WnRk+1k2kut/wt/me9f2xD/dk/If40f2xD/dk/If41jK25Qcg5GcjpS1zexiP2jepsf2xD/dk/If40f2xD/dk/If41j0UexgP2kjY/tiH+7J+Q/xo/tiH+7J+Q/xrHoo9jAPaSNj+2If7sn5D/Gj+2If7sn5D/Gseij2MA9pI2P7Yh/uyfkP8aP7Yh/uyfkP8ax6KPYwD2kjY/tiH+7J+Q/xo/tiH+7J+Q/xrHoo9jAPaSNj+2If7sn5D/Gj+2If7sn5D/Gseij2MA9pI2P7Yh/uyfkP8aP7Yh/uyfkP8ax6KPYwD2kjY/tiH+7J+Q/xo/tiH+7J+Q/xrHoo9jAPaSNj+2If7sn5D/Gj+2If7sn5D/Gseij2MA9pI2P7Yh/uyfkP8aP7Yh/uyfkP8ax6KPYwD2kjY/tiH+7J+Q/xo/tiH+7J+Q/xrHoo9jAPaSNj+2If7sn5D/Gj+2If7sn5D/Gseij2MA9pI2P7Yh/uyfkP8aP7Yh/uyfkP8ax6KPYwD2kjY/tiH+7J+Q/xo/tiH+7J+Q/xrHoo9jAPaSNj+2If7sn5D/Gj+2If7sn5D/Gseij2MA9pI2P7Yh/uyfkP8aP7Yh/uyfkP8ax6KPYwD2kjY/tiH+7J+Q/xo/tiH+7J+Q/xrHoo9jAPaSNj+2If7sn5D/Gj+2If7sn5D/Gseij2MA9pI2P7Yh/uyfkP8aP7Yh/uyfkP8ax6KPYwD2kjY/tiH+7J+Q/xo/tiH+7J+Q/xrHoo9jAPaSNj+2If7sn5D/Gj+2If7sn5D/Gseij2MA9pI2P7Yh/uyfkP8aP7Yh/uyfkP8ax6KPYwD2kjY/tiH+7J+Q/xo/tiH+7J+Q/xrHoo9jAPaSNj+2If7sn5D/Gj+2If7sn5D/Gseij2MA9pI2P7Yh/uyfkP8aP7Yh/uyfkP8ax6KPYwD2kjY/tiH+7J+Q/xo/tiH+7J+Q/xrHoo9jAPaSNj+2If7sn5D/Gj+2If7sn5D/Gseij2MA9pI2P7Yh/uyfkP8aP7Yh/uyfkP8ax6KPYwD2kjY/tiH+7J+Q/xo/tiH+7J+Q/xrHoo9jAPaSNj+2If7sn5D/Gj+2If7sn5D/Gseij2MA9pI2P7Yh/uyfkP8aP7Yh/uyfkP8ax6KPYwD2kjY/tiH+7J+Q/xo/tiH+7J+Q/xrHoo9jAPaSNj+2If7sn5D/Gj+2If7sn5D/Gseij2MA9pI2P7Yh/uyfkP8aP7Yh/uyfkP8ax6KPYwD2kjY/tiH+7J+Q/xo/tiH+7J+Q/xrHoo9jAPaSNj+2If7sn5D/Gj+2If7sn5D/Gseij2MA9pI2P7Yh/uyfkP8aKx6KPYwD2kjT/sX/pt/47/9ej+xf+m3/jv/ANetOiuX2s+5tyRMz+xf+m3/AI7/APXo/sb/AKbf+O//AF606KPaz7hyR7HL+GNS0Txto8Wr+Hdf0/X9KmZkjvtLuEuYHZWKsBIjFSQwIODwQRWr/Yv/AE2/8d/+vXC/DP7VD8SPidDbW0DaM2sRyy3byGOcXhsLMNEsQUh4tgRvNLq25mTy8KJH9NqnUnZNPdJ/ev69dwdOKk1/X9d+z0Mz+xf+m3/jv/16guPC9tdTW006xTTWzmSCSSEM0TFSpZSfukqzDI7MR3raopKtUWzF7OPY5PxL8MfD3jOGCLxBo+ma7FAxaJNSsY7hYyRglQ4OCfal1z4Z+H/E1vaQaxpOm6rBaMHtor6xjmWFgMAoGBCnHcV1dFbRxmIioqM2uXbXa+9uwnSg90cldfDHw9fa1/bFzo+mXGr7UT7fLYxtPtRg6DzCN2FZVYc8FQRyK1LXwzb2KyC2WK3EkjSuIogu52OWY46knknvWzRUyxVaSUZTbSH7OPY5618EabY/Yvs1pa2/2KJoLXyrZV8iNtu5Ex91TsXIGAdo9BV7+xf+m3/jv/1606KiVepLWUrj5InOa94D0rxVpcum63ZWesadKVMlnf2qTwuQQQSj5BwQCOOorj/FP7OvhXxH8P77wTawR+GfDOov/wATCw0Gzt7ZLyJhiSFx5Z2rIoCs6bZAANrqRmvU6Kr6xW5PZ8z5d7dPuD2cb81tTJg8Px2sMcMTrHFGoREVMBVAwAOaf/Yv/Tb/AMd/+vWnRUe2m+olTitEjM/sX/pt/wCO/wD16P7F/wCm3/jv/wBetOil7WfcfJEzP7F/6bf+O/8A16P7F/6bf+O//XrToo9rPuHJEzP7F/6bf+O//Xo/sX/pt/47/wDXrToo9rPuHJEzP7F/6bf+O/8A16P7F/6bf+O//XrToo9rPuHJEzP7F/6bf+O//Xo/sX/pt/47/wDXrToo9rPuHJEzP7F/6bf+O/8A16P7F/6bf+O//XrToo9rPuHJEzP7F/6bf+O//Xo/sX/pt/47/wDXrToo9rPuHJEzP7F/6bf+O/8A16P7F/6bf+O//XrToo9rPuHJEzP7F/6bf+O//Xo/sX/pt/47/wDXrToo9rPuHJEzP7F/6bf+O/8A16P7F/6bf+O//XrToo9rPuHJEzP7F/6bf+O//Xo/sX/pt/47/wDXrToo9rPuHJEzP7F/6bf+O/8A16P7F/6bf+O//XrToo9rPuHJEzP7F/6bf+O//Xo/sX/pt/47/wDXrToo9rPuHJEzP7F/6bf+O/8A16P7F/6bf+O//XrToo9rPuHJEzP7F/6bf+O//Xo/sX/pt/47/wDXrToo9rPuHJEzP7F/6bf+O/8A16P7F/6bf+O//XrToo9rPuHJEzP7F/6bf+O//Xo/sX/pt/47/wDXrToo9rPuHJEzP7F/6bf+O/8A16P7F/6bf+O//XrToo9rPuHJEzP7F/6bf+O//Xo/sX/pt/47/wDXrToo9rPuHJEzP7F/6bf+O/8A16P7F/6bf+O//XrToo9rPuHJEzP7F/6bf+O//Xo/sX/pt/47/wDXrToo9rPuHJEzP7F/6bf+O/8A16P7F/6bf+O//XrToo9rPuHJEzP7F/6bf+O//Xo/sX/pt/47/wDXrToo9rPuHJEzP7F/6bf+O/8A16P7F/6bf+O//XrToo9rPuHJEzP7F/6bf+O//Xo/sX/pt/47/wDXrToo9rPuHJEzP7F/6bf+O/8A16P7F/6bf+O//XrToo9rPuHJEzP7F/6bf+O//Xo/sX/pt/47/wDXrToo9rPuHJEzP7F/6bf+O/8A16P7F/6bf+O//XrToo9rPuHJEzP7F/6bf+O//Xo/sX/pt/47/wDXrToo9rPuHJEzP7F/6bf+O/8A16P7F/6bf+O//XrToo9rPuHJEzP7F/6bf+O//Xo/sX/pt/47/wDXrToo9rPuHJEzP7F/6bf+O/8A16P7F/6bf+O//XrToo9rPuHJEzP7F/6bf+O//Xo/sX/pt/47/wDXrToo9rPuHJEzP7F/6bf+O/8A16P7F/6bf+O//XrToo9rPuHJEzP7F/6bf+O//Xo/sX/pt/47/wDXrToo9rPuHJEzP7F/6bf+O/8A16P7F/6bf+O//XrToo9rPuHJEzP7F/6bf+O//XorToo9rPuHJEwdY8caRoPiTRtDv5ri3v8AVy6WTGynNvI6qWMZuAhiSQgEqjOGYK20HBrerzr41LPLp/hiKwkjttYk123XTrq4jMsFvP5cpEksQZTKgUOCiuhORh16j0OPcI1DkM+PmZRgE98DJx+dYQfMpPs7fgn+v9bK5aSS7q/4tfp/XV1JS0VQHL+AfAcXgOx1CM6tqOvX+o3jX17qeqmHz7iUoiAlYY441CxxxoAiKMICcsWY9RRRTuHVsKKKKQBRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFAGDrHgfSNe8SaNrl/DcXF/o5drJDeTi3jdlKmQ24cRPIASFdkLKGbaRk1vUUUbaB1uFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFAH//Z)\n"
      ],
      "metadata": {
        "id": "2lklSOgMOtvK"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def reccurent_item_set_generator(i,preceed1,freqset):\n",
        "  while(i <5): #repeating the logic untill frequent itemset size is 4.\n",
        "    preceed=preceed1.map(lambda x:x[0]) #taking keys of current RDD containing current item sized of frequent items for every iteration.\n",
        "    broadcast_preceed=sc.broadcast(preceed.collect()) #broadcasted the above keys.\n",
        "    usefulRDD=maindata.filter(lambda x : len(x)>=i) #main logic is for taking combinations from line that line must have minimum current item set size elements or we can't make combinations ,\n",
        "                                                  #so to reduce time, filtering only lines that have minimum current frequent item size.\n",
        "    usefulRDD=usefulRDD.flatMap(lambda x : generate_combinations(x,broadcast_preceed,i)) # from every line getting current itemsize of all combinations by function generate_combinations.\n",
        "    usefulRDD=usefulRDD.map(lambda x:(frozenset(x),1)).reduceByKey(lambda a,b: a+b).filter(lambda a: a[1]>=4) # using wordcount logic for set of 2,3 and 4 sized set of frequent items as keys\n",
        "                                                                                                            # and filtering that met support.\n",
        "    freqset=freqset+usefulRDD # for every iteration adding current rdd of current item set to it.\n",
        "    return reccurent_item_set_generator(i+1,usefulRDD,freqset) # calling reccursion for next itemset size combinations.\n",
        "  return freqset"
      ],
      "metadata": {
        "id": "jpttOsB_Kio_"
      },
      "execution_count": 16,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "y=reccurent_item_set_generator(2,movieIDRDD,sc.parallelize([])) #just passing single item set sized frequent rdd to reccuression function which will return frequent item set of size 2,3 and 4\n",
        "                                                                #at a time.\n",
        "y=y+movieIDRDD #combining frequent single item sized to pass-2 rdd."
      ],
      "metadata": {
        "id": "_EOLwB1h5p1v"
      },
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "movieIDRDD.unpersist() #unpersisting frequent items os single set as they are no use after combining to pass-2 rdd."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gWlRytFLTz4j",
        "outputId": "30f45846-811f-41b9-930d-1a52e06a9fe9"
      },
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "PythonRDD[30] at RDD at PythonRDD.scala:53"
            ]
          },
          "metadata": {},
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "##Association rules generation"
      ],
      "metadata": {
        "id": "2qO2bMDBhmgd"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "cache=y.collectAsMap() # collecting frequent items with count as hashmap for computing association rules."
      ],
      "metadata": {
        "id": "7EQr6g4PhuXC"
      },
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "output_subsets=[]\n",
        "def getsubsets(main_list,index,temp):\n",
        "  output_subsets.append(temp[:])                #As we know we will calculate confidences of all subsets from super set for the rules\n",
        "                                                #This is the function I have developed by using backtracking which will give list of all subsets. from a list.\n",
        "  if(index>= len(main_list)):\n",
        "      return                                       # if [1,2,3] is given then this fucntion gives all list of subsets as output which are:- [[],[1],[2],[3],[1,2],[1,3],[2,3],[1,2,3]].\n",
        "  for i in range(index,len(main_list)):\n",
        "    temp.append(main_list[i])                   #Logic of backtracking\n",
        "    getsubsets(main_list,i+1,temp)              #reccursive calls.\n",
        "    temp.pop()                                  #backtracking redo operation."
      ],
      "metadata": {
        "id": "Qk4QYIsTlqp5"
      },
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import math\n",
        "def association_rules_formation(x): # every line of frequent elementsrdd is passed here which gives output as list of tuples of 3 parts.\n",
        "  output=[]                            #In tuple first part will be left side of rule,second part will be right side of rule(only one element), third part is confidence value.\n",
        "  union_set=list(x[0]) \n",
        "  numerator=cache[x[0]] #getting numerator in confidence formula which is support of combined elements and this value is taken from hashmap created above where set of frequent elements\n",
        "  if(len(union_set)==1):         #are taken as key and their frequencies as values.\n",
        "    return ''  # As we need rules from 2-set we don't need 1 set elements, we are not proceeding with it.\n",
        "  else:\n",
        "    getsubsets(union_set,0,[]) #from set size of 2 calling my developed backtracking function to get all subsets.\n",
        "    filtered=[]\n",
        "    for i in range(0,len(output_subsets)):\n",
        "      if(len(output_subsets[i])!=0 and len(output_subsets[i])!=len(union_set)): #from above backtracking function we can get empty subset and full set so taking all other subsets \n",
        "        filtered.append(output_subsets[i])                                      # excluding these two.\n",
        "    for i in filtered:                             \n",
        "      denominator=cache[frozenset(i)]                   # now in filtered list, we get all subsets whose frequencies will be denominator and taken from hashmap.\n",
        "      confidence=round((numerator/denominator)*100,2) #calculating confidence.\n",
        "      if(confidence>=50 and confidence<=100): #filtering confidence values.\n",
        "        copy=[x for x in union_set if x != i] #taking all elements except in filtered because we want only one element on right side and their combined set excluding to be left.\n",
        "        output.append((copy,i,confidence)) #appending in tuples for easy formation of rules.\n",
        "\n",
        "  return output\n",
        "\n"
      ],
      "metadata": {
        "id": "13JV0e9FjBdm"
      },
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "rulesRDD = y.flatMap(lambda x:association_rules_formation(x)) #getting rules that pass threshold confidence in form of list of tuples from frequent set rdds."
      ],
      "metadata": {
        "id": "Ep16TBKTvDsp"
      },
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#rulesRDD.take(10)"
      ],
      "metadata": {
        "id": "AQdm9EpiybqP"
      },
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "del y #deleting frequentset items rdd from memory as they are not use in future."
      ],
      "metadata": {
        "id": "GAxDMccEcATs"
      },
      "execution_count": 24,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#Writing to file"
      ],
      "metadata": {
        "id": "Cc4paVhqcCE-"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def regex(x):  # A function that is useful to make list of frequent items on left side to string for readability of rule.\n",
        "  rule=[]\n",
        "  string =''\n",
        "  for i in range(0,len(x[0])): # as first part of tuple will be list which is left part of rule converting all in string form.\n",
        "    string+= x[0][i]+','\n",
        "  rule.append((string,x[1][0],float(x[2]))) #this string which is left part of rule will be first part of tuple, second part of tuple will be rightside of rule, third part will be confidence.\n",
        "  return rule; #returning rule.\n"
      ],
      "metadata": {
        "id": "SwSt8dWrDcUz"
      },
      "execution_count": 25,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "rulesRDD=rulesRDD.flatMap(lambda x: regex(x)) #getting rules in readeble format.\n",
        "rulesRDD=rulesRDD.map(lambda x: (x[2],x[0][:-1]+'->'+ str(x[1])+';'+'Confidence'+'='+str(x[2])+'%')) # converting into required format with making key as confidence value for sorting."
      ],
      "metadata": {
        "id": "AQmaFgCwD0Xs"
      },
      "execution_count": 26,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#rulesRDD.take(5)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XkK1_g1ZGmzC",
        "outputId": "2e89e718-c051-44d0-d006-e41ddd91e73f"
      },
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[(50.0, '1196,1580->2687;Confidence=50.0%'),\n",
              " (50.0, '1196,1580->2687;Confidence=50.0%'),\n",
              " (50.0, '1196,3114->2687;Confidence=50.0%'),\n",
              " (50.0, '1196,3114->2687;Confidence=50.0%'),\n",
              " (65.91, '1196,480->2687;Confidence=65.91%')]"
            ]
          },
          "metadata": {},
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "rulesRDD=rulesRDD.distinct() #taking distinct values for fast computation."
      ],
      "metadata": {
        "id": "xjgtGZMkJI3D"
      },
      "execution_count": 28,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "rulesRDD=rulesRDD.sortByKey(ascending=False).map(lambda x: x[1]) # Sorting in descending order of confidence and in above our entire rule is stored in value and making rdd with that only."
      ],
      "metadata": {
        "id": "IbxTsZ91jG5F"
      },
      "execution_count": 29,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "rulesRDD.coalesce(1).saveAsTextFile('output of rules.txt')#writing into file."
      ],
      "metadata": {
        "id": "XfaEmtfAjQ9p"
      },
      "execution_count": 30,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "del maindata # deleting maindata from memory after writing into file."
      ],
      "metadata": {
        "id": "_cR_phrVqqNB"
      },
      "execution_count": 31,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#Bonus"
      ],
      "metadata": {
        "id": "EsyvzfX4q0O2"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "![image.png](data:image/png;base64,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)"
      ],
      "metadata": {
        "id": "__BBa4TU6bN8"
      }
    }
  ]
}