diff --git a/gen_binary.py b/gen_binary.py new file mode 100644 index 0000000..98288e9 --- /dev/null +++ b/gen_binary.py @@ -0,0 +1,10 @@ +k = 3 +def gen_bin (a,p): + if p 0: + for j in adj[queue[0]]: + if j not in conn_comp: + conn_comp[j] = coconut + queue.append (j) + + queue.popleft () + output_file.write(str(coconut)+"\n") + for i in conn_comp.values (): + output_file.write(str(i)+" ") +search_df() + +input_file.close() +output_file.close () + + + diff --git a/kolchanova/A-2.py b/kolchanova/A-2.py new file mode 100644 index 0000000..72e36ed --- /dev/null +++ b/kolchanova/A-2.py @@ -0,0 +1,51 @@ +import sys +sys.setrecursionlimit(200000) + +from collections import deque +from collections import defaultdict +input_file = open('pathbge1.in', 'r') +output_file = open('pathbge1.out', 'w') + + +def function(): + pass + + +n_vertices, n_edges = [int(i) for i in input_file.readline().split()] +vertices = [] +edges = [] +for i in range(1, n_vertices+1): + vertices.append (i) + i += 1 +for j in range (n_edges): + edges.append ([int(j) for j in input_file.readline().split()]) + + +adj = defaultdict (lambda: defaultdict(lambda: 0)) + +for x, y in edges: + adj[x][y] += 1 + adj [y][x] += 1 + + +distances = {1:0} +def BFS (): + queue = deque ([1]) + while len (queue): + for i in adj[queue[0]]: + if i not in distances: + distances[i] = distances[queue[0]] + 1 + queue.append(i) + queue.popleft() + return distances + +BFS () + +answer = distances.values() +for i in answer: + output_file.write(str(i)+ " ") +input_file.close () +output_file.close () + + + diff --git a/kolchanova/A-3.py b/kolchanova/A-3.py new file mode 100644 index 0000000..7e7b708 --- /dev/null +++ b/kolchanova/A-3.py @@ -0,0 +1,60 @@ +import sys +sys.setrecursionlimit(100000) +#from collections import deque + + +infile = open('pathmgep.in', 'r') +outfile = open('pathmgep.out', 'w') + +dimensions = [int(i) for i in infile.readline().split()] +n_vertices = dimensions[0] +start = dimensions[1] +end = dimensions[2] + +scope = [] +def get_matrix (scope, infile, n_vertices): + for i in range(n_vertices): + y = [int(i) for i in infile.readline().split()] + scope.append(y) + return scope +get_matrix (scope, infile, n_vertices) + + + +distances = {} #distance dict +def get_dist (distances, n_vertices): + for i in range(1, n_vertices+1): + distances[i] = -1 + return distances +get_dist (distances, n_vertices) + + +#the main algorithm +vertices = [] +def dijkstra_alg (distances,start,end): + + distances[start] = 0 + for y in range(n_vertices-1): + + vertices.append(start) + for i in range(n_vertices): + + if scope[start-1][i] != -1: + if distances[i+1] != -1: + distances[i+1] = min(distances[i+1], (distances[start]+scope[start-1][i])) + else: + distances[i+1] = scope[start-1][i]+distances[start] + assist = {} + for key in distances: + if key not in vertices and distances[key] != -1: + assist[key] = distances[key] + if len(assist) == 0: + break + start = min(assist, key=lambda k: assist[k]) + return distances +dijkstra_alg (distances,start,end) + +outfile.write(str(distances[end])) + +infile.close() +outfile.close() \ No newline at end of file diff --git a/kolchanova/A-5.py b/kolchanova/A-5.py new file mode 100644 index 0000000..f260530 --- /dev/null +++ b/kolchanova/A-5.py @@ -0,0 +1,51 @@ +import sys +sys.setrecursionlimit(100000) + + +input_file = open('pathsg.in', 'r') +output_file = open('pathsg.out', 'w') + +nodes, edges = [int(i) for i in input_file.readline().split()] +dist = [] + +dist = [[None for i in range(nodes)] for j in range(nodes)] + + +def get_dist (dist,nodes): + cumulative_w = 0 + + + for j in range (edges): + + x,y,z = [int (w) for w in input_file.readline().split()] + dist [x-1][y-1] = z + cumulative_w += z + + for j in range (nodes): + for i in range (nodes): + if j == i: + dist[j][i] = 0 + elif dist[j][i] is None: + dist [j][i] = cumulative_w +1 + return dist + + +def F_W (nodes, dist): + for i in range (nodes): + for j in range (nodes): + for x in range (nodes): + if dist[j][i] + dist[i][x] < dist [j][x]: + dist [j][x] = dist [j][i] + dist [i][x] + return dist + +get_dist (dist,nodes) + +F_W (nodes,dist) + +for line in dist: + + output_file.write(' '.join(str(i) for i in line) + '\n' ) + + +input_file.close() +output_file.close() diff --git a/kolchanova/B-1.py b/kolchanova/B-1.py new file mode 100755 index 0000000..b8e33d8 --- /dev/null +++ b/kolchanova/B-1.py @@ -0,0 +1,16 @@ +#!/usr/bin/python + +outdata = open("allvectors.out", "w") + +def gen_bin(vector, position): + if position < len(vector): + for nextvalue in ["0", "1"]: + vector[position] = nextvalue + gen_bin(vector, position + 1) + else: + outdata.write("".join(vector) + "\n") + +with open("allvectors.in", "r") as indata: + n = int(indata.readline().strip()) + +gen_bin(["0"] * n, 0) diff --git a/kolchanova/B-6.py b/kolchanova/B-6.py new file mode 100755 index 0000000..27966e2 --- /dev/null +++ b/kolchanova/B-6.py @@ -0,0 +1,31 @@ +#!/usr/bin/python + +def nextvector (seq): + result = seq[:] + for x,y in reversed(list(enumerate(seq))): + if y == '0': + result[x] = '1' + return result + else: + result[x] = '0' + return ['-'] + +def previousvector (seq): + result = seq[:] + for x,y in reversed(list(enumerate(seq))): + if y == '1': + result[x] = '0' + return result + else: + result[x] = '1' + return ['-'] + +with open ('nextvector.in', 'r') as infile: + seq = list(infile.readline().strip()) + +preceding = previousvector(seq) +following = nextvector(seq) + +with open ('nextvector.out', 'w') as outfile: + outfile.write(''.join(preceding) + '\n') + outfile.write(''.join(following) + '\n') diff --git a/kolchanova/B2.py b/kolchanova/B2.py new file mode 100644 index 0000000..8f7ef1b --- /dev/null +++ b/kolchanova/B2.py @@ -0,0 +1,26 @@ +#!/usr/bin/python + +indata = open ("vectors.in","r") +outdata = open ("vectors.out",'w') + +n = int (indata.readline().strip()) +vectors = [] + +def gener_vect (vector, position): + if position < len (vector): + for next in ["0","1"]: + if position > 0 and next == "1" and vector[position-1] == "1": + pass + else: + vector[position] = next + gener_vect(vector, position + 1) + else : + vectors.append (vector[:]) + return vectors + + +gener_vect (["0"] * n, 0) + +outdata.write (str(len(vectors))+"\n") +for vector in vectors: + outdata.write("".join(vector) + "\n") \ No newline at end of file diff --git a/kolchanova/B3.py b/kolchanova/B3.py new file mode 100644 index 0000000..b51d2e1 --- /dev/null +++ b/kolchanova/B3.py @@ -0,0 +1,19 @@ +infile = open ("permutations.in","r") +outfile = open ("permutations.out",'w') + +n = int (infile.readline().strip()) + +def generate_permutations (seq, loc): + if loc < len (seq): + for i in range (1, n+1): + if i not in seq: + seq[loc] = i + generate_permutations (seq, loc+1) + seq[loc] = 0 + else: + outfile.write(" ".join([str(i) for i in seq]) + "\n") + +generate_permutations (["0"] * n, 0) + + + diff --git a/kolchanova/B4.py b/kolchanova/B4.py new file mode 100644 index 0000000..c11cee3 --- /dev/null +++ b/kolchanova/B4.py @@ -0,0 +1,27 @@ +#!/usr/bin/python + +infile = open ("choose.in",'r') +outfile = open ("choose.out",'w') + +n, k = (int(x) for x in infile.readline().split()) +variants = [] + +def gener_comb (combination, position, n): + if position < k: + for i in range (1, n+1): + if i not in combination: + if position > 0 and i < combination[position-1]: + pass + else: + combination[position] = i + gener_comb (combination, position+1, n) + combination[position] = 0 + else: + variants.append(combination[:]) + return variants +gener_comb ([0]*k, 0, n) + +for v in variants: + outfile.write(" ".join([str(x) for x in v]) + "\n") + + diff --git a/kolchanova/B5.py b/kolchanova/B5.py new file mode 100644 index 0000000..bdec0f3 --- /dev/null +++ b/kolchanova/B5.py @@ -0,0 +1,24 @@ +infile = open ('subsets.in', 'r') +outfile = open ('subsets.out', 'w') + +n = int(infile.readline().strip()) + +limit = range (1, n+1) +subsets = [] + + +def generate_subsets(x, y): + subsets.append(' '.join(str(i) for i in x)) + for val in limit: + if val > y: + generate_subsets(x + [val], val) + return subsets + + +string = [] +output = generate_subsets(string, 0) + +outfile.write('\n'.join(output)) + +infile.close() +outfile.close() \ No newline at end of file diff --git a/kolchanova/B8.py b/kolchanova/B8.py new file mode 100644 index 0000000..cdced4e --- /dev/null +++ b/kolchanova/B8.py @@ -0,0 +1,19 @@ +#!/usr/bin/python + +with open('nextchoose.in') as infile: + number, n = [int(x) for x in infile.readline().split()] + combination = [int(x) for x in infile.readline().split()] + +def mk_choice(number, n, combination): + for x in range(n): + if combination [ n - x - 1 ] < number - x: + combination [ n - x -1 ] += 1 + for y in range ( n - x, n ): + comb [y] = combination [ y - 1 ] + 1 + return combination + return "-1" + +result = mk_choice (number, n, combination) +print result +with open ('nextchoose.out', 'w') as outfile: + outfile.write(" ".join(str(x) for x in result)) \ No newline at end of file diff --git a/kolchanova/C3.py b/kolchanova/C3.py new file mode 100644 index 0000000..e1f9861 --- /dev/null +++ b/kolchanova/C3.py @@ -0,0 +1,34 @@ +#!/usr/bin/python + +infile = open ("tower.in",'r') + +inlines = infile.readlines() +#print inlines +def peop_lang(inlines): + lang_vars = {} + population = {} + for someone, line in list(enumerate(inlines))[1:]: + population[someone] = [int(x) for x in line.split()[1:]] + for variant in population[someone]: + lang_vars[variant] = lang_vars.get(variant, []) + [someone] + return population, lang_vars + +def thinker_f(start, uzd): + uzd[start] = True + for variant in population[start]: + for neighbor in lang_vars[variant]: + if not uzd[neighbor]: + thinker_f(neighbor, uzd) + + +population, lang_vars = peop_lang(inlines) +uzd = {someone: False for someone in population.keys()} +start = 1 +uzd = uzd +thinker_f (start,uzd) + +outfile = open ("tower.out", "w") +outfile.write(str(uzd.values().count(True))) + +outfile.close() +infile.close() \ No newline at end of file diff --git a/kolchanova/C4.py b/kolchanova/C4.py new file mode 100644 index 0000000..f6fbb37 --- /dev/null +++ b/kolchanova/C4.py @@ -0,0 +1,20 @@ +#!/usr/bin/python +infile = open ('harddrive.in','r') +_ = int(infile.readline()) +parts = [int(x) for x in infile.readline().split()] + +def function (parts): + turns = -1 + backturn = set() + + for part in parts: + if (part - 1) not in backturn: + turns += 1 + backturn.add(part) + return turns + +result = function(parts) +#print result +infile.close() +with open('harddrive.out', 'w') as outfile: + outfile.write(str(result)) \ No newline at end of file diff --git a/kolchanova/C5.py b/kolchanova/C5.py new file mode 100644 index 0000000..f4bc217 --- /dev/null +++ b/kolchanova/C5.py @@ -0,0 +1,29 @@ +#!/usr/bin/python + +with open("maxpiece.in") as infile: + n, m, k = [int(x) for x in infile.readline().split()] + xcuts = [0, n] + ycuts = [0, m] + for line in infile: + t, v = [int(x) for x in line.split()] + if t == 0: + xcuts.append(v) + else: + ycuts.append(v) + +xcuts.sort() +xsizes = [] +for i, _ in enumerate(xcuts): + if i > 0: + xsizes.append(xcuts[i] - xcuts[i-1]) + +ycuts.sort() +ysizes = [] +for i, _ in enumerate(ycuts): + if i > 0: + ysizes.append(ycuts[i] - ycuts[i-1]) + +biggest_square = min(max(xsizes), max(ysizes)) + +with open("maxpiece.out", "w") as outfile: + outfile.write(str(biggest_square)) \ No newline at end of file diff --git a/kolchanova/C6.py b/kolchanova/C6.py new file mode 100644 index 0000000..50d036b --- /dev/null +++ b/kolchanova/C6.py @@ -0,0 +1,47 @@ + +#!/usr/bin/python + +infile = open("olympic.in", "r") + +_ = (infile.readline().strip()) +participants = {} +record = [0, 0, 0] +win_country = "" +top_three =[] + +def fill_dict(infile, participants): + for winners in infile: + first, second, third = winners.strip().split() + + for participant, index in zip((first, second, third), (0, 1, 2)): + + if participant not in participants: + + participants[participant] = [0, 0, 0] + + participants[participant][index] += 1 + ##print participants + + top_three = sorted(list(participants.keys())) + + return top_three + +def find_winner (top_three, win_country, record): + + for score in top_three: + + if participants[score] > record: + + record = participants[score] + + win_country = score + + return win_country + + +top_three = fill_dict (infile, participants) +win_country = find_winner(top_three, win_country,record) + +outfile = open("olympic.out", "w") +outfile.write(win_country + "\n") +outfile.close() \ No newline at end of file diff --git a/kolchanova/C7.py b/kolchanova/C7.py new file mode 100644 index 0000000..0defcb1 --- /dev/null +++ b/kolchanova/C7.py @@ -0,0 +1,16 @@ +#!/usr/bin/python +from __future__ import division + +with open("airplane.in") as infile: + M, N, alpha = [int(x) for x in infile.readline().split()] + passenger_mass = sum(int(x) for x in infile.readline().split()) + +mass_on_empty = M + passenger_mass + +if alpha == 1000: + result = "Impossible" +else: + result = mass_on_empty * alpha / (1000 - alpha) + +with open("airplane.out", "w") as outfile: + outfile.write(str(result)) \ No newline at end of file diff --git a/kolchanova/Classifiers.ipynb b/kolchanova/Classifiers.ipynb new file mode 100644 index 0000000..e4d855c --- /dev/null +++ b/kolchanova/Classifiers.ipynb @@ -0,0 +1,451 @@ +{ + "metadata": { + "name": "", + "signature": "sha256:46d94e6567358407841491b8c7099f3035fa724076809ddb6bec796b36a00164" + }, + "nbformat": 3, + "nbformat_minor": 0, + "worksheets": [ + { + "cells": [ + { + "cell_type": "code", + "collapsed": false, + "input": [ + "import sys\n", + "import numpy as np\n", + "import scipy\n", + "from sklearn.utils import shuffle\n", + "import matplotlib.pyplot as plot\n", + "from sklearn.neighbors import KNeighborsClassifier\n", + "%matplotlib inline" + ], + "language": "python", + "metadata": {}, + "outputs": [], + "prompt_number": 1 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "blue = np.loadtxt(\"./blue.txt\")\n", + "red = np.loadtxt(\"./red.txt\")" + ], + "language": "python", + "metadata": {}, + "outputs": [], + "prompt_number": 2 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "_ = plot.plot(blue[:, 0], blue[:, 1], \"b.\")\n", + "_ = plot.plot(red[:, 0], red[:, 1], \"r.\")" + ], + "language": "python", + "metadata": {}, + "outputs": [ + { + "metadata": {}, + "output_type": "display_data", + "png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAEACAYAAAC08h1NAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCgoKCnji/wO6FCbrjnOy5gAAAABJRU5ErkJggg==\n", + "text": [ + "" + ] + } + ], + "prompt_number": 3 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "red_points = np.hstack ((red, [[1]] * len (red) ))\n", + "blue_points = np.hstack ((blue, [[0]] * len (blue) ))\n", + "all_points = np.concatenate((red_points, blue_points), axis=0)" + ], + "language": "python", + "metadata": {}, + "outputs": [], + "prompt_number": 4 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "x = all_points[:, :-1]\n", + "y = all_points[:, 2]\n", + "x, y = shuffle(x, y, random_state=1)" + ], + "language": "python", + "metadata": {}, + "outputs": [], + "prompt_number": 5 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "train_set=int(all_points.shape[0] *0.6)\n", + "Xtrainset = x[:train_set]\n", + "Ytrainset = y[:train_set]\n", + "Xtestset = x[train_set:]\n", + "Ytestset = y[train_set:]" + ], + "language": "python", + "metadata": {}, + "outputs": [], + "prompt_number": 6 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "knearest = KNeighborsClassifier(n_neighbors=5)\n", + "knearest.fit(Xtrainset, Ytrainset)\n", + "print('Training set score:', knearest.score(Xtrainset, Ytrainset))\n", + "print('Test set score:', knearest.score(Xtestset, Ytestset))" + ], + "language": "python", + "metadata": {}, + "outputs": [ + { + "output_type": "stream", + "stream": "stdout", + "text": [ + "('Training set score:', 0.89166666666666672)\n", + "('Test set score:', 0.88)\n" + ] + } + ], + "prompt_number": 7 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "print (\"Visualization of the KNN method\")\n", + "prediction = knearest.predict(Xtestset)\n", + "_ = plot.xlim(-1, 1)\n", + "_ = plot.ylim(-1, 1)\n", + "_ = plot.scatter(Xtestset[:, 0], Xtestset[:, 1], c = prediction, linewidths = 0.001)\n", + "_ = plot.scatter(Xtrainset[:, 0], Xtrainset[:, 1], c = Ytrainset, marker = \"*\", linewidths = 0.001 )" + ], + "language": "python", + "metadata": {}, + "outputs": [ + { + "output_type": "stream", + "stream": "stdout", + "text": [ + "Visualization of the KNN method\n" + ] + }, + { + "metadata": {}, + "output_type": "display_data", + "png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAEACAYAAAC08h1NAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRRRRRBFFFFFEEUUUUcSnyf8AVDmXEKhQFKEAAAAASUVORK5CYII=\n", + "text": [ + "" + ] + } + ], + "prompt_number": 8 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "def classify (x,y):\n", + " if knearest.predict([x,y]) == 0:\n", + " print(\"must be blue...\")\n", + " else:\n", + " print(\"must be red...\")" + ], + "language": "python", + "metadata": {}, + "outputs": [], + "prompt_number": 9 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "print ('Lets try to classify something!')\n", + "print('what will be your x?')\n", + "x = raw_input()\n", + "print('and y?')\n", + "y = raw_input()\n", + "\n", + "classify (x,y)\n" + ], + "language": "python", + "metadata": {}, + "outputs": [ + { + "output_type": "stream", + "stream": "stdout", + "text": [ + "Lets try to classify something!\n", + "what will be your x?\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "stream": "stdout", + "text": [ + "5\n" + ] + }, + { + "output_type": "stream", + "stream": "stdout", + "text": [ + "and y?\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "stream": "stdout", + "text": [ + "7\n" + ] + }, + { + "output_type": "stream", + "stream": "stdout", + "text": [ + "must be blue...\n" + ] + } + ], + "prompt_number": 10 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "from sklearn.ensemble import RandomForestClassifier" + ], + "language": "python", + "metadata": {}, + "outputs": [], + "prompt_number": 11 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "rforest = RandomForestClassifier (n_estimators=10)\n", + "rforest.fit(Xtrainset, Ytrainset, sample_weight=None)\n", + "\n", + "print('train set score:', rforest.score(Xtrainset, Ytrainset))\n", + "print('test set score:', rforest.score(Xtestset, Ytestset))" + ], + "language": "python", + "metadata": {}, + "outputs": [ + { + "output_type": "stream", + "stream": "stdout", + "text": [ + "('train set score:', 0.97833333333333339)\n", + "('test set score:', 0.87)\n" + ] + } + ], + "prompt_number": 12 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "print (\"Visualization of the Random Forest method\")\n", + "prediction_rf = rforest.predict(Xtestset)\n", + "\n", + "_ = plot.xlim(-1, 1)\n", + "_ = plot.ylim(-1, 1)\n", + "_ = plot.scatter(Xtestset[:, 0], Xtestset[:, 1], c = prediction_rf, linewidths = 0.001)\n", + "_ = plot.scatter(Xtrainset[:, 0], Xtrainset[:, 1], c = Ytrainset, marker = \"*\", linewidths = 0.001)" + ], + "language": "python", + "metadata": {}, + "outputs": [ + { + "output_type": "stream", + "stream": "stdout", + "text": [ + "Visualization of the Random Forest method\n" + ] + }, + { + "metadata": {}, + "output_type": "display_data", + "png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAEACAYAAAC08h1NAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzsnXdYFNfXgN+ld0ERURSxG7FX7L3XaDQaNdYk9hgTY0xM\n7Ik11hg1xth7i72jYO+iooCIdJDeFlh2935/rKwsdYEFze/jfR6eZWZum92ZM3fOPQVKKKGEEkoo\noYQSSiihhBJKKKGEEkoooYQSSiihhBJKKKGEEkoooYQSSiihhBJKKKGEEkr4j7MVCAee5FJmLeAD\nPAYaFcegSiihhBJKKDraohLmOQn+XsDpt/+3AG4Vx6BKKKGEEkooWpzIWfBvBD7NsP0CKFfUAyqh\nhBJKKCF79IqhDwcgMMN2EFCxGPotoYQSSighG4pD8ANIMm2LYuq3hBJKKKGETBgUQx/BQKUM2xXf\n7tOgWrVqwtfXtxiGU0IJJZTwP4UvUD0/FYpjxn8c+Pzt/y5ALCorIA18fX0RQuj0z8cninbt/kEi\nmUfbtlvx9o7Ms86TJ+HUqrUOmMfw4YeJj0/R+bjS/2Jjk5ky5RR2dsvp0WMXly7l/zt4+DCEzZvv\nMXLkEW7eDCA5WcbcuXOLbMxCCJ4/j+D48Re0bv03Z874AHOBeeq/Y8ee66SfpCQZo0cfBeZRrdoa\ndu16hJHRAo2++vXboy5/8qQXhw8/46efLqFUKnV2voX5PmNjkwkLS8i1jFKpJDg4vkh/s4kTT7Bh\nwx2OH3+h87Y///wo339/niNHPLlyxY+zZ31ITpaxcuUNRow4wtatD1AqlaSmypk06SSlSnWna9cd\neHq+KdJzLqo/hUJJSEg8iYmpfPXViRzLnTzphbn5Yr788jilSv2mcd0uX35NZ+MBquVXKOtixr8X\naA/YotLlzwUM3x7bhMqipxfwEkgCxuigT62oXr00ffvWJDw8kX79alGjRpk869Sta4eLS0Xq1rWj\nT5+aWFoaF9n4SpUyYcgQZzZuvIeVlTGdOlXVuu69e8E0beqAkZE+ixe74+8fR48e1XFxqZR35UJi\na2vGmDHHuHUrmPv3Q/j225asXHkTkNC6dSV69MjX5CNHzMwM6du3JtHRyZibGzF8eAOqVbOhe/fd\nxMfLsLIyZvbsNuryvXvXxMcnioED6+ik/8ISGBjHyZPeJCWl8emnzjg4WKGnp6n1lEpl3L4djLu7\nP99+2woTEwP09XU7HxNC8McfvZFIJCiVgl27PBgxor7O2t64sTf6+hLGjz/Ozp0qG4/Kla3w948D\nJOza5cGRI8+5fPkVUqkcSOHChVecPOnNRx+V1ck4ios3bxLp2nUXHh7h6OmBRCIhNDSBLVv6Ubas\nuUZZU1MDgoNn4Orqh5ubP3FxqepjFSpYFvfQNdCF4B+mRZkpOuinQNSqVZrnzyfz4EGo1nWWLOmC\nvb0FYWGJRTgyFWlpCl69+ho3N39iAwKwdnTMtXxAQCyvX8cxZcpp1q/vRZUqpWjUyJ7u3atRvbpN\nrnUfPw7jo49sMTJS/ex+fjEcOPAMW1szRo9uqLXAsbU1o169cnTpUpWrV19z4cIrTEwMmD27DbNm\ntcHYWDcaxMhIKWfO+GBnZ84nn9RBCIGLiyM+PtPw9Iygdm1b7O0tAAgNTSAwMI6lS6+zYkU37OzM\nMTc30sk4Csq9eyHMnHkBuVxJ+fIWDB+eVdieOfOSceOOEx+fikIh+Pnndjofh0SietjcvRvMwYOe\n7Nrlga9vNBMmNKVcOYtCtx0Tk8L+/U/VQh/A3z9eo9zJkz7pNdT77t4NKVTf74MlS67h4aFSWCiV\nAgsLI8aPb5xF6APqidyAAR/h6FiKoUMPExKSwMiR9Rk6tG6xjjszxaHjfy8kJqaSmqpg/fp7tGrl\nmK+ZRbowSf8sSjp1qkpqYiI9Xcy5/NNP9P7jDwxMTdE3NMy2fNmy5vzwwyWePHnDX3/dZ8OGXhw4\nMBhDQ32SkmQAVK5cHz+/GKpUUT0IZDIFt28HsXnzfZo3d6BVq0rY2ZnTrNlfREUlA+Dq+ppduwZq\nPe61a3ty9eprFi1yBySkpChYtuwGP/7YtnBfyFuEEHTrtpOHD8MA2L//GU+fTsLRsRR2dubY2Wne\naFFRUkaMOIqPTzRVqlizZElXnYwDoEOHDgWq17VrVZo2rYCenoQeParj7x9L5crWGmUGDarD6tW3\nUSoFo0c3xMBAXwcjzp6mTSvwxx93CQ1NxMnJulBCPy1NgaGhPnv2eLBq1e18TpKcAGjXLvdJTn55\n8CCExo0r6LTNzCQlpWXYklCvnp36PsuNxo0r4O09tegGlk+Ky6qn2ImLS2XgwP2cP+/LkCGHiI9P\nzbtSMaFUCvz9Y9Xbb54+ZUfnzjzZtYtz336LUi7Psa6pqSG1apVh4cKO1KpVBktLEwwN9dXH9uzx\n4MwZBb/8coUDB56iVCoxMtInMlLKrl0eLF9+A3NzQ86efakW+gD79j1FqdTe2MrExIDY2BSNfUlJ\naaSlKbWqH/LgQa7Ho6KkaqEPkJAg486dLDYBaurWLUejRuUYPrwe7dpVxsBAd5d2QQW/QiE4f34E\nx44NIS4uhS+/PEF0tBSpVKYuk5ycxr//DsXNbTTGxkUn9EE1O69Xz44bN8YW6q1MqVQyc+Y5AD77\nrD5CCKKipCxY0EGtyvrqq8ZMmdKcNm0cmTy5KaVKvVOZOjo2YN26Hkye3LxQ55NOUFAc168HMHbs\nca5fDyAoKE4n7WbHlCnNsbZWnYuBgYRZs1pTt65dkfVXVGQ2s3yfiLcLFTpj7lxXvL2jcHa2Y84c\n3b9CF4TISCl37wZz9eprZsxohY2NSnAfHDKEtKQkmk+dSvUePXJtIzk5DVNTQ1JS5JiYaN7Ar1/H\nULfun5iaGvL48QS1LvHu3SB27PAgNDSBffs+4dIlP3r02K2u5+BgiY/PVExNs3/TyI64uBRq1lzH\nmzdSAL78sjGbNvXNtU5sQAAJwcGcmTKFnuvXY1OlChb29lnKKRRKqlRZQ2CgSmWgry/h8eMJODtn\nf5MJIYiOTqZMGTOioqSUKWOm9XnkBx+fKK3WijISEpLAsGGHcXPzp0MHJ3bt+hgHB6siGV9Rc+6c\nD4sWuXPzZhADB9Zm3rwO+PrGUKtWGaTSNKpUsUEmU2RRfXh7R3HgwFPs7S0ZM0Z7tWJGVAvhIktd\nqVTG2LHH2b//GUOHOrN1a/98Xcfa4O7uz+XLftSrZ0eLFhV58CCUmjXLUKuWrU77KQhvVXkfkizP\nF0LXhITEa3zmRUxMspg27bQYMuSAOH3aW+t+Ll16JerU+UNUrbpGbNv2MNeyZ854i9KllwiYK377\nzU0oFAqhVCpFXFCQEEKoPwuKr2+UWLz4qpg711WEhiao90ulMiGEEHK5QqSmpgkhhJg711WULbvs\n7XjmCQODBeKPP25r1c+zZ+Fi4sQTwsJisejadZv4++97QqlU5lkvJT5e7BswQMwDcWzMGCFPTc21\nj549d4k2bbaKw4efaTWuoiI2Nln4+8eKdu3+Fq9fx4joaGm+6s+de1kMGXJAzJ/vmq96KSlp4s2b\nxHzVKUoCA2OFnd0yAfNE69ZbRHx8SrH1/fhxqDh//mW2x+bNcxVz517O9/erDWfOeAuJZJ4A1d/K\nldd13kdhvkcK4Bf1IT0l3p7D+6NLlx1cuuQHqGaYN2+Oo1kzh1zrJCSk4uDwOwkJqtd3PT149mwS\ntWtnv6YghKBTp+3o6+vxxx+9dD5jEEKoF/My/p8T9++H0LTpZtIvBT09CA//DlvbrItVmfn558ss\nWuTOmjXdmTbNResxXvrpJ/QMDDA0M6PNrFla1VEqlejpvT/NZGxsCqNGHeP4cS86dHBi375B+dKR\nh4TEU6GClfozL4QQJCSkcu6cLyEhCYwZ0xALC6P3+h2k06fPbqytTbCyMmHDht5F0sfNm4EsWOCG\nnp6EhQs7cudOENu3e5CcnMbw4fWYMaOlxsw/KUmGubkRUqkMMzPdLuqPH3+cv/9+qN5u0cKBW7fG\nF7g9mUyuNrBIZ+HCq3zxRZMCrSsWZMb/P7u4WxDc3QPU/ysUgps3g/IU/NHRyWqhD6BUQkBAfI6C\nPzVVwfHjwzA3NyQyUppju3FxKfj4RNG0ae79ZyajoHd396ddO6dcy6vWPt7VUSpBKk3LuUIGqlSx\nxstrcr6tMxpNmwmGxliaaP+g37nTg6FD65LyJpRSlYreZDUz1tYmtG5dCWtrYypUsMz3wmi6sNdG\n6IPqdzx27AWTJp0mNVWBQiH45hvtH65FyfbtH6tVaqBa6A0IiKNatdI6af/NmyR69NitXpe7fTsI\nb+8pLFzoRmxsCpMmNcui7km34NKV0E9MlGFsrI+hoT6VK5fSOOboWCqHWnkTHBzP2bMvGTeuMaB6\nYM2YcY5duzz4+++HrF3bk379ahVq7Nrw/qcPHxBNmpTPsCVo3Dir7jkzFSta4eLyLvRQpUpWNG+e\ns7A2MTHA0tIYPT097OyyFx7nz/uyaJEbS5Zc5/BhT9LSFFqfA0BQUDxHjz7nyy9PcuTIc7y9I3Ms\n26aNI23avLOuaN++Mo6O1jmWz8jYsY2pWdM2WzPF7FAqlURFSTl60pe9ez2IT1bty43ExFTmzXNl\n8WI3Ro44zIrxS4n29SUtJSXXekXB2LENKVfOnG+/bVks/Q0dWhdn57I0aVKeUaMa5Pn2Vlykr5+U\nKWOGv38shw8/588/7+HtHUl0dBIAvr7RubYRG5uc47GXL6M1jDGiopJ58uQNGzb05tSpzwgNLToz\nayEEY8cew9LyN0qVWsKhQ8+YObMVw4fXw97egi5dqrB2be5rcDmxZ48HffvuZf78q8yefRGZTIG5\nuRHffONCmTJm1KhRuliE/oeGLlRlhSIsLEGMHHlEdOu2U+zZ46F1vfj4FLF0qbuYP99VBAfHFXoc\ngYFxolKl34W+/nxx507+df5KpVJ8880ZAfPEZ58dVuv0cyIlJU0sX35NTJhwQtSsuU5s3nyvyPTK\n27Y9FKami4Sx8UKxebN26wJ37wYJmCucy/0gfgGxonx58eryZfXxy5dfiRUrrotbtwKLZMxCqNYb\nunXbIQwMFoh+/fYIP7/oXMsrFIpC9xkbmywSE1OFTCb/oPT8GblyxU+ULbtM6OnNF19/fVr8/fd9\nERQUJzp2/EcEB8eL2NjkLHWUSqUYP/5Yjm1GR0uFnd1ytU69UqXfRWJizmtBuuTECS91vzBPmJsv\nFnJ54X/LdJo02SRMTRdprL89eRImEhJSxZ07QSItTa5RXi5X5HktUQAd//9bVU9qQgLGlprec+XK\nWbBjx8f5bsvS0pjvv2+Td0EtsbY2plev6pQvb4mhYf5fyiQSCRUrWrFhQy9CQxOy6BPT2bXLg88+\nq4exsQHTp7dk8OCDeHtHIZcrs3VI0QWffFKHjRvvIZFIGDLEWatZbEKCjJs3x3Fq5xUqe3bErJQV\nVTp2BGDnzsd8/vkxQLUuc+LEMHr2rKHzcdepY0fr1pW4ezeEnj2r4+SUs+32m6dPSYmLw7F160L1\nWaqUifr/ovo9CkvbtpVp0KAcYWGJbN/ugUKhxNbWDT+/WEaNOsru3Zq+IbdvBzJu3HE8PSPx949n\n69b+VKyoqf6ysTHl6tXRLFt2HT09CT/+2LbYnPEym31LpWnI5UqdeFOnpMiZP78DtWqVITw8Ua3P\nr1tXFaU+s1p56dJr/PyzK0ZG+mzc2Edn3tbw/1TwC6WSSz/9RK+1a9/3ULIlXXfq6xtNSop2+vbM\nTJvmgoGBHnJ5VlVKXFwKa9feZseOx7i5vWbIEGe6dKmGs3NZhg51JjU1f6ql/CCTKXB1HYWenh4J\nCdr5VnTsWAWAOhU7YlVxJHFBQerF3m3bHqnLKRSCHTse51vw37gRSEREEh07VsHKKucQHS1bVmLW\nrDZcvx6Q7XEhBPc3b+bZvn0oZDKchw6l+eTJSD6ABdmiIjVVzvHjwzAxMaBWrfXI5UqGDq1DcHAi\nVapYa6gznz+P4O7dUCwsjHFyKsXgwXWyCP10ate2ZevW/sV1Gmr69KlJ7dplePEiCoCvv26hM090\nExMDeveuqVXZJ0/C+eGHSwCkpSkZM+YYLi4VqV5dN+soH4bSUMXbt5aiJejOHU5NmEDEs2dU79mT\nPps2YVHuw8kLo1AoqVv3T168UOnlrayMefZsUo43SEG5fNmPzp130L17Nc6cGY5EIlFbAQktrIE+\nBMLCEpg58wK7dr0LFTBjhgsrV3bXuo25c11ZsMANENSqZcutW+OxtjbJs15OyFNSWOXoSJpUyndh\nYRhZFL3394dAZGQSiYkyTE0NCQ6Oo3FjB968SdQQ/G5u/vTuvZvERBk1a5bmxx/bMmqU7jKxZmct\nUxDi41WxhGxsTPIVP0uXXLnymo4dt2fZ37t3DY4e/VTttAkFs+r5352K5EDF5s2p2LIlFuXLU3/k\nyPcq9GWyrB66YWGJaqEPqlfPhw+1jzOkLSkpcq5eHU2nTlXUQj7z54eMVJrGrFmXWLmyO+3aVcbc\n3JBevWowd24HrdtQKpUsWXJNve3lFcXhw56FGldyTAw91qxh4O7dJIZnCUJbLGT2qM4P4eGJvHwZ\nTX4nYba25jg52VCunAWNG6tUFpmNF+zszEhMVFnAeXtHM2PGBXWYEVDdDxERSfnuO52lS6+jUGjn\nOZ4boaGJDBpUp8BC/9o1/0KPwcWlIg4O6d/fu+/j1Ckfdux4XOj2/98JfoB6w4cz1ceHUnkERCtq\nMgqddMqWNaNixXdrD8bG+tSpo/sIhr161aBdu8p8/33+9dCvX8fStOlmzM1/ZfDgg7i7+7Nq1U3c\n3F7rfJzpZLQSuXkzEGfnDezY8Zhx4/7l8OHBJCb+yKlTn+WqqsmMRCJ5GyZBoJowCRITCxfaw7J8\neeoNG0bt/v0pXS3f0XJ1ws8/X9YQqNqyadM9HBx+p0aNdQwYsF9rIfroUShr1tzi0qVXuZZ7+TIa\n1fesmlhERyerY/y4ub2mXLkV2NmtoHPnHRphLfIiKkrKuHH/snLlTbp125WjKi4zYWEJGttv3iTx\n4kUkX3xxAm/vKKKjcza3zg4fnyiOH/fiiy9OcOKEFx4eYWzYcJdduzzy/UCSSFQe36prU/NBePKk\nN40abaRPnz0EBBQsPMX/vOBPSpJlmVk7tmqFvqEhDs2avZcxeXlFMnjwARYvvsbQoYd49SpGfczI\nyIALFz6nf/9adOlShX//HZrFPjo2NpnU1Jzj+RQ1Eyee5P79UKTSNA4d8qRdu3+YMeM8HTpsZ8+e\nnFIvF4yEhFRCQuL56quTBAfHExubTMuWlWjfvjJt2zoyenRDrZzNskMikbBz50B1XJ969coxdeqH\nYStfEKKjpQwZcpCNG+/RsOFGXF1zF8QZSUtTMHXqGRQKlZA5ftyLkye986zn7u5PixZ/M336Obp0\n2cmmTfdyLNuyZaUMDkoS6tQpq7aJnzTpNLGxqoeuq+trNm26r/XYy5QxY8gQZ9LSlFSpYk3r1tpN\n6EaOPJrFZ2XChJO4uwcwd65qUTU/ODqWYu/ep7x4EcXp0z7067eXyZNPM3LkUT799FC+2gKQSPRQ\nPSTfiely5cw5duwFjx6Fc+qUD598ciDf7cL/mOA/dEjzNd3XN5p16+7w11+5BwQrbmrVsqVZswro\n6UGrVpWoWlXTQqR2bVuOHRvKhQuf0737u9j2CoWSESOOYGOzDBubpRw9+jzHPjw8whk58ihjx/6r\n8WDRBaqZSFaEQGOxtaDcvh2k/j8tTcGECae4dMmPIUMOqqMjzpnTFje3MTRsmLevRW70718bd/cx\nREV9z7Zt/bPEy/8vUbq0GV991YTq1cvQokVFOnbUXlUhBGqhn07mgHvR0VLi4jTVSLt2eSCTvTMG\nyOjhmpkyZcy4dm0M06Y1Z+bMlri6jlLrqtNVQOlou/CfjpGRPkFB39CrV965IO7cCaJhw41cvPiK\nSpV+58CBpwDY2ZnTtGkFJk1qSrNmFbCwyF8uDmNjA6pVs2bdup5IpWkaoakPH36er0CRxsYG/PZb\nZ/V2s2YVOHhwMJMnNyOjOt/TMyJfY0znf8KqJzQ0gd27n7B69S28vaPo2bM6L15E8vXXZ4mOTsbE\nxABv70iWLOmq8+BNBaV2bVvCw7/j6lXt9YHHj3uxe7dqRp2cLGfUqGP0718rixu/ykJlO9HRKieZ\ny5f9ePFiSpaAbgVlzJiGfPPNOUCCnh4ole8uxPLlC55g4vnzCHx8olmw4Cq//tqZxo3LY2trRrt2\njtjYmGBnZ64Obla9uipQmi68RdOT15QubVrott43dnbmPHkyEQ+PsHyFLzAy0mfBgg7MmeMKqBz7\n+vZVWaAolUpiYlI4cOAZRkb6DBhQm9KlTZFIJFlCDJQvn/tidrVqpVmzpmeW/d9/34rJk08DEsqX\nt2D06IYax9MDE2aHQqGkTRtHDA31tUrC06yZAxUqWPL4cTjlypkTGSmldOmlGBnps3RpZ0aNakRM\nTM4OZrnxyy/tMTIywNXVjx07PNT7LS2NMDPLn+z5/vvWDBxYm/j4VBo0sEdfX48HD0JYuNBN/VDu\n0aM6hw8XaKgfDIVyjFi//raAeWL06KPqfS1a/CUMDRcIU9NFwscnqlDtfwhs3/5Iw7lEX3++kMnk\nWcq5ur7SKAfzxPPnb3Q6ltOnvcXvv98Qbm5+omvXHcLEZJFo02arCA3VLiBedsTGJotOnbYJmCfm\nznVVO3eFhydofP4vkpCgOwel4OB4Ub/+BgHzRMOGf2o4C+XF06fh4to1f43rSqlUir/+uieMjRcK\nE5OFYvv2R+pjSUky0afPbmFiskg0abJJvH4dU+Bx378fLI4efZ7ld37wIETs3ZuzQ6Wb22tx4YJv\nvvr67ruzYs8eDzFu3DEBc9X3iZHRAhEZmVSg8Wdm3jxXYWa2SNjbLxdnzmgf9DEnUlLSRGhogrh6\n1U9MmnRSLFp0VSQnp/3/CNIWF5fCxImnePw4nE6dnFi1qgcGBip7bgcHS+7dC2H27LakpMh5/DgM\nb+8oKlSwxMrKOM+4O0WJQi5H3yD3GbfIw4wyNjYZF5e/8fJS2RjPnNmSZcu6ZSkXGppArVrr1TGE\n7O0tePly6nvPSJUXirQ0ps+4QMWKVtjamqnjmfx/4O+/H9CwoT1NmhQ+kUi3bju4cMGPdHkwblxj\ntmzpl2udbdsesWbNLWxsTFm/vlcWg4L4+FR69tyNEIILF0YW27W0efN9li69TkqKnPHjG/Hzz+3U\nyWqUSiVr1tzmzz/voVQKvvyyMd9910qrQHbp95qr6ys6ddqpcczHZ6rO7OUjIhIJDk6gVi3bQmkb\n3N396ddvL7GxqbRuXYmzZ4erVVH/L8Iyjx59VGMmu2SJe6GfpMXBg61bhTQq97eOrVsf5NlObGyy\nOHLEU1y96pdruevXA0SvXrtF//57hYdHWH6GWmSkpcnF6tU3xcSJJ8SpU17q/W88PcXqqlXFPBB/\ndewiUhISRGpq1jeZ/0VkMrn48ceLolSp34S9/XKxZct9reveuxecZd/Fi74aM1iYKz7+eF+u7dy+\nHagRdtjJaVWWUBqRkUlCJpOL5GRZvkNSF5YaNdaK0qWXZBu2QaFQCnv75aJs2WUiLS3/oRUSE1OF\ns/Mf6nNv126r1iEa8go3cvGirzA3XyxgnqhZc60ICyv4G2v9+n9qyL2lS9/JPQow4//PLe5mtHFX\nbUdlWy6/gc2KCnlqKpfnzMF90SIOfPIJzw4ezFLG3z+W2bMv8u2355k16wKHD3tSu/Z6LC1/Y9Kk\nkxqBzEqVMuHjjz/KM+pmq1aVOHXqM44dG0q9eu/HVyE1VU54+LuAWt9/f5Hp08/x55/36dNnL+fO\nvQTg9OTJxL5SWaAEu17k1qpV+baoKAjSyJyD1+ma0EePUCqyXpOGhvrMnNkKMzND7OwstHrLiYtL\n4erV13z//QUOHnzGs2dv1MdOnfJBc/InYcKEJgQExHL9egDx8Vlt/F+8iCTjy/br17GkpGhajZUp\nY4ahoT4mJobY2BTfWkhMTDK7dw/k7t0vsk3vGBaWwL//DuXcuREEB8dn04ImoZkyv5mbG3Ht2ljW\nr+/J5s19OHdupNbhGa5dC+Dly+zlD8CsWRfVxgje3tGsXXtbq3azI7N5rmYKyPzznxP8ffpkdHkW\n9O6dvXv+8uU3imdAeWBgbEzjL74g1t+fNKkU58GDs5SpXNmaMmVMiYlJQSKBn392xcsrisREGX/+\neZ+9e5++h5EXHCEE0dFSjh17wY4dj4mKkqJQKDl92idDGTh3zheA5Oh3NvoCiAvOOcViXnifOoVC\nlrcNeGpiIhd++KHA/WhLWkoK/teucX3pUh7+/Tdhj7JaPUVFJfPo0QT27h2kleWHlZUx9++HcPny\na3777ZrGonSNGpoqiooVLTl82JOqVdfQps0/1K+/MYtVVps2jlhavlPdtG/v9MEYQdjYmNKsmQNV\nq5bOdiG/QgUrmjevSKNG5bPkM85ISlwcAdeucfGHH3hx7BgRz99ZxFlbmzB5cnO++KKJVgYQCoWS\nDRvu8vPPrkyffo7t27O3ZMucyjQ/qU0zM3t2G9K1wBUqWDJ2bOE8not+WqU98+bNm5dnoXbtKlOp\nkhWVK1sze3abLKv44eGJTJt2hk2b7nHrVhA1a5ZRpx8sCuRyRZ46xUhvb5pPnYp15crY1q6Nnn7W\nr93LK4qff26HVCrn6NEXGvFyWrasRKtWmjHoZTIFa9fe5siR59jammlY07i7+zN+/An27n3KRx/Z\nUqGCJUqlkkOHPLl+PRAHB8t8m6rlB4lEwvHjXowff4Jz53yxt7fAxaUirq6vNd7Yxo1rROPG5ZHo\n6/PixAkA9AwMKN+gAcalSmHt5KR1nzKplAd//cWddeuI8PREoqdH6erZm/a9dnNjV7du+F+5Qtij\nR1Tt2hVDs6JJ1ahvYEDgjRu4zZ/Pm2fPaDRmDGZlNFM3li5tirm5EWXLmmsVF0YikRAamkD16qUx\nNzdkxIgPN3EjAAAgAElEQVQG6mNNmlRAKpURFBRPXFwK8fGp3L8f9tYbVkJcXCqmpgZ06lRFXcfG\nxpRevWpgZKRPjx7VWbeup87i07xPZDKFevaub2zMs4MHub9xI1FeXjQaOzZLkEZt0dOTULWqDd99\nd56YmBQ2buyT7RtqpUpWHD36ArlciZNTKTZu7JMvB8OMNG5cnoEDP6Jnz+osXtxZQ6bNnz8fYH6B\nGv4AKLD+KyNnz/qIAweeCmPjhWLatNM6aTM3li27pvM2p08/o9bllSr1m/D2jsxSZtiwQ+oypqaL\nxLNnKqud0NAEYWHxq/pYmTJLRXx8ihgz5oh6n4PDyiK3kJHJ5KJlyy2ideu/1brNN28SxeDBB0SD\nBn+Kn366qBFu1v/aNXFh9mwx38BALC9XTkQ8f57vPv2vXxfzQGxu2lTIpLnroY+MHCk2N2smXpw4\nke9+8kvAjRvi1JQpYv+gQbnqhYPv3xd3N24UQXfu5NlmeipNqVSWbZs//HAhk2XXO73/3LmuBT6X\n/xJbtz4Qfn7vrIyeHT4sLvzwgzg6alSh2/byihR79niIDRvu5HovhYTEi1u3AnVqtZUZ/j+HZb59\nO4gbNwL5559HdOpUmSdPJvDwYViR9RcSksDcua7s3fuUBw9CmTWrNQ0bls+7ohasWtWDVq0qERyc\nQL9+tbI4eAH8+6+X+v/kZDnnz/tSp05ZXr6M1nCGiYpK5tWrGLZvf+dRGxycwOnTPowerbsAWZmR\nStO4cGEkxsYGapvosmXNOXDgnarrzbNnXJ0/H6FU0vann2g4ahSyhARSYmOxqlgxp6ZzJE0qZdDe\nvYQ/eYK+Ue5WJ61nz6Zs7dpEeeftnZoRr5MnOT1pEvKUFDrMn0+ziRPzrGPfsCGVWrZEKZejlMvR\nN8yqRvE5c4a9ffsiFArQ02PIoUN89HHOIcLTVTE5qWQyq3z09fVQKATOzmWZOrV5nmP+L3LvXjBN\nmzogk8n55RdXdu70wMbGlO+/b8XnnzekRs+e1Bk4ELkOkvjUrFmGmjXL5FmufHnLPH1bzp71oUcP\nlco6NjYZa+uiX0P5kEyA3j68shIfn5rnK1JKipy+ffdw8aIfW7f2Y8yYohNq6Wzf/oixY/9lxoyW\nLF+e1ayyoGgTZbBhw408fvwuCNjJk8Po3bsmMTHJfPTRH4SHqzIhOTlZ4+k5CUfH1RqpHk+f/qxI\n4tZrS2pCAutq1CDpbSAz09Klmfj0KZblVQ9PRVpatgLyfZISH89Ke3vkyW+deyQSqnbrhkOzZrSb\nMwcD44Krzw4OHoznoXdu/dV79mT46dMFbk8IwY8/XuLIkedUq1aa5cu7IpGoHKiKQo0j3mNE18jI\nJJ4/j2TmzPMsWNCRKlVssLExpWrVNVSvXpoHD77SeZ9paQqNCJkFITIyiePHvVix4iZffdWYhg3t\nOXr0BatXZ3Vwy43/2eicy5ZdIyIiKdcyJiYG1K9fjq1b+2FqWvgLOyIiSWMxMjtsbEwJCppB48b5\nm+lndHHPjBAi2+BtmTl0aAgdOjhRo0ZplizprI7znZ7EYty4Rnz1VRNcXUdhamrIvn2DsLc3x8hI\nj2++aUGPHln13z4+Ubi4bMHefgUzZpwrcJREbYj181MLfVAt8Ma+fq3eziz0hRB4nz6Nx+7dGovB\nxUlqbOw7oa8aFK/OncN90SJOT5lSqLbNymrazZvb2RWqPYlEwm+/dcHLayqnTw/H2dmOOnXsikx3\nf/jw8yK9XkC1npYdNjamnDnjw+3bISxa5E7p0qaEhSXw6NFXLF/etUAB63IjJUXOunV3Ct2Ora05\nCoXg+fNI9u59xvjxJ9iw4R7Dhh0usOewtnxQM/4rV/zYtOk+SUmpjBrViK5dqzB9+jn27HmCvb0F\n69b1pE+f7HNSCiEQQqCnp6dO0lFQ9u9/yoEDngQFxTN0qDOTJzfTSZxvN7fXfPLJQSIjpXz6qTM7\nd36MRCJBKk3D0tKYW7cC+flnV27fDmbQoNosW9ZN55mXwsMTuHo1AEfHUhq5ggFatNjCnTvvLGrK\nlTNn8eJOReJIlRIfz7oaNZC+UZkiGltbM8bdnXJ162Zb/vTUqdxdvx4Am2rV+OLOHUxL68bJRluU\nCgU7OnfG/+pV4F1MTwCb6tWZ5pP7RCE3pJGR7O3fn6CbNynfpAmfnTiBhX3h4hDlB7lMhkEe6rHs\niI9PYfPm+2zb9phu3arRt29NOnRwUt+LumT58uvMnNkaIQS//OLKvn3PqFy5FH/91ZdLl14REBBH\nZGQSGzb01Wm/GTl//iW//nqN588j+fRTZxYv7oSlZcHf9HbtegxI8PAIIzg4kTt3gvjrr3506ODE\nvn1PcXf3p2nTCrlqMAoy4/+gBL+h4XzS0t7NGsaMacC0aS0YMGA/zs5lOXVqeLEMJDVVToMGG/H2\njuL166+1Tj6eF1WqrOb163dhVLdt64+TkzVBQfHqhOVffnmc3bufcuDAJ1pn69GWwMA4WrTY8jZZ\ntWDFim58+20r9XF7+xVqFVE6enoSHjz4kgYNtBNCQgj+/PMuz55F0KNHdfr2zTl59Jtnz3BbuJDo\nly8JffgQlEoajBpF/3/+0VAbKOVyFhkbIzL4MwzYvp0Gn3+u5ZnrDnlKCg//+YeA69d5snu3+gZy\n/vRTPtm3r9DtKxUKPA8dou6nnxa6rfxwdeFC2v/8c4Hqnjv3kh49dtOxoxPnz4/kyZNwEhJS8/Q1\n0YZr1wI4f96XBw9CuXLlNX371qR+fTt+/NFVXaZVKwcuXPgcMzOjfMUnKigDBuzj7NmXuLuPKXQ0\ngPRJqlKp5OHDMOrVs+PFi0ju3Qth3LgT6nIrV3ZlxoxW2bbxn1f1ZI4G+M8/j0lISMXTcxLz53fQ\nSZIFbZDLlUyY0JR//x1KRET+YnLnRnrY2XROnvRm0KADTJp0mgULriKXK+jRozpBQd/oLKBaRnbv\n9ngr9AEkWXwdhg7NOttWKkWu0T0TE2W8ehWt/m1++ukSkyefYcOGe/Trt48TJ7xyrGvn7Ez/f/4h\n7PFjeCvUH2/fzquLFzXKSfT1MbbSzECW12w/LigIf3d3UmJjcy2XXwxMTGg2cSKDdu2ix+rVVOnU\niSZffUXfzZsL3Xbw3btcXbCA8999x/Vly0h68ybvSgUgo0om/OlT9n38Mdd+/ZXDw4cTG6BdLPuM\nyGQKTp4cRrt2ldm8+T6TJp1mxozzrFp1s1D3bHrI74UL3Th1yovk5DQqV7biyZMIMhqy+PjEqIV9\nUQt9IQTDhtXl1auvszi5FYT0tyI9PT2aNKmAkZEB9evbv3XEe8fZs76F7kujX522pmMMDfVo1Kg8\nZmZGNG3qoJOEx9pgbm7E9Oku9O1bSyexU9KZNu2dNUX58hasWtUdOztzbGxMmDHDBQMDVXRBGxtT\nOnfWfcq3jMm7gSwL5qtWdWfbtv44Or4TsmXLmtGyZfYWNpcuvaJChZVUq7aOli3/Ji4uJcsFe+bM\ny1zHlG7pkpE0qephe2XhQtKSk5FIJHy8ezfGpUqBREKTCROo2adPjm2+PHeOdTVqsK1dO/6oU4fo\nV9rHpc8PLl9/zeeXLtFn48YsD6aCUKFpU2JeviQhKAizsmULrefPDiEET/buVW+Xq1sXu3r1QCLB\nqWNHrAuQnKhv31r07l2TefM68MUXjfH1jcbbO4qvvmpS4Ht2//6n/PKLawaPYgmdOjnh6xvDzZuB\n6WcDCAYMyPmtUtdIJBI+/bQuFSpY0rZt5SLrx9lZc83no49sddr+B2XOuWfPIL777jwhIQkYGuqz\naVOfPB2NhBC8fBlNqVLGGqneUlPlOl/ICgyMY9y44/j5xTB4cB0WL+6cL0uG+fM70ratIyEhCXTr\nVg0jIwNOnfoMCwsjoqKSi9SpClTBuk6d8uHUKR9KlzZlyxZNXahEImHUqIYMHPgRGzbcJSFBxpgx\nDbG3z94cbdq0M+pAcHfvhrBx4z3q1CmLh8e7mWpeF6yxpSUtZ8zg1u+/A1DRxQWbqlU5MnIkngcP\n8vryZeoMHkzzSZOYFRODMi0tT1PNq/Pno3hrspcYGsqt1avptXZt7l/Oe+DluXOEP36MU4cOODRv\njkQioWzduoy4cEFj4VtX+F25gtexY/icOkXs69c0mzgRUxsbyjdqxLehoQRcy9uoIC8iI6Vs2dIP\nExMDwsISqVo1/+swM2eeZ8WKm5n2SqhevQy//tqZ8uVXUqlSKT79tA5OTjZMmNC00OP+0Jgzpz1R\nUcm4u/vTrFkFfv21c96V8sEHpeNPfwVNSEjF0FAPE5PczfmSkmSsWHGdsLAkLC2N6NGjOh07qnLI\n/vKLKwsWdMyrQ7777jx79jzF0bEUO3YMoFatnAVVkyabePDgnW/Ajh0DGDmyQY7lP1Ti41MxNzcs\n9BtU9epr8fV9pwaaN68906a1YNKk03h6RtC9e1WWLOmqVXKTwJs3SU1IwKl9e/SNjDg5YQIPNm/G\n2NqabwIDMdYyafkbT0+2d+xI0ps36ovbZcYMuq9cWZBTLDLubdrEqQkTAJW38vAzZ6japUuRJrxX\nKhTs7d+fl6dOMfTECWrl8tZU3GQ0yLCzW66hYjU01KN9eyf27x9ESEgCSUlpBAXF06VL1Sxvse+b\nV6+iC/SwKwwF0fF/UDP+dCwtjfM0DTt1ypuTJ725csUPX99YLC2NGDmyAZ6eEcyadZELF17h5RXF\n8uVdclyc3b3bg99/vwWokpyPGHGEu3e/zFLOxyeKV69iePr0De90i5K3OUT/exTUbTwzc+e2Z8yY\nf1EoBI6OVowf3xgbG1P27h2U77YqtWypsW1oakrTyZMJf/w4zxl+OkmRkWzv0AFpRMQ7a5tq1Wgw\nejRxgYFYVaz4wSSS99ixQ/2/Ui7nyZ49VO3SpUgT3uvp61OpZUtcpk8nLSl38+jCkt8H1+nTPtSp\nU5aqVUtTsaKVhuDfsqUfn3+ummCVLq0KrdGihW7HW1iSkmQkJsqYOPEUW7f2x9zcsFgcsQrKByn4\nX7yIJDw8kfbtnXIs06lTFX755QovXkQyaFAdzM2NqFjRChsbU5o2rcDVq6/p1q1qrhY5/v6aiYoD\nArKP7mdqasgPP1xEJktfqJJgYCChe3fd6+H/S4wc2YAWLSoSEBBHs2YVdDr76rpiBfoGBigVCiRa\nmgVGPHuGNEIzFV3tjz9mc8OGCKWShmPG0H/rVp2NsTBYOmhagxTEU7kgtPvppyLvI8rHB1lCAuUb\n520GrFAoWbHiBnv2PKF0aTNGjqzHjh0D+OyzIwQGxvPZZ3UZObJ+kY+5oLx5k8iLF1HY25sza9ZF\nzp9/xYAB+zhwYHChBP+TJ+FFGlX3w5j+qBBKpZItWx6wZ88TkpLSGDrUma+/dslWJSGTyVm06Cq1\na5elQgVLOnSoovamO33ah/btK3P9egDduuWcg9PDIwwXl79JTlYtLk6Z0ox163plW3b06GNYW5uQ\nkJBC6dJmNGlSHhsbE7p3f3/eryVoEh8SwroaNZC/XRw2trZGFh+vYQY6+upVKrdr976GqCYhNJTD\nQ4eqgsR16cLHO3cWWaC44kIIgceuXbw4cgR5aiofDRpEozFj8nxwR0QkYm+/ksqVrfH2nqJOtPKh\n8+hRGJ06bScmJgUrK2PGjWtIeHgStrZm2aaX1AZ//1hCQxOZNesia9b0oHLlUjmGwU5LUxAXl5ru\n6/PfteMXQpCaKqdq1bXExqYQEjIjx1lkxlfJwuhDHz8O4/hxLypVsuLzzxtk63QihCApSYaFhTGJ\nials3fqQFStukJwsZ/Tohvz2W+f/zMX6v47flStcW7wYib4+Lb/7jl1du2ocH3H+PNUy7StBdyTH\nxLC+Vi3kyclMe/UK80weyZm5fTuQUaOO8fJlNA0a2LNnz0Bq1cq9zofCsGGH2LfvmXq7W7cqnDv3\nOeHhiZQrp92aVGZCQxMYOHA/t24FM21ac1as6JZtaIhr1/zp33//27za8+C/ruOPi0th2bIu2NiY\nEh6emKPgzyjoM/4fGpqQr4TfDRrY5+qcJJWmMX78cdzc/Gne3IGtW/uxfftjAgNVaqHNm+8xfbqL\nOgl4Ce+XKh06UKVDB/V2i6+/5vaaNQA4depElY65L/gXNSkpcpYuvfbWK7yuzs12ExNTi9Q6LCZG\nipmZUY4Wc/KUFNr+9BMGJibIEhLyFPwtWlSiU6cqyGRK5sxpl63QT0qSERUlxcHBqthMurUhs0BO\n/94LKvRBFdStfv1yuLhUpG5duxzjAU2adPqt0P/vU+jwpAqFUgwefKDQ7WRk5szzGuFtP//8SIYw\nt6pQt0eOeBa6n4wJrHWFNDpa523+Fwm6c0e8cnUVcpms2PpUyLNPHfnppwfV15KBwQJx61agTvud\nOvWUTttLR6FQihEjDguYK0xMForDh7O/5jOGiM4rNWE6N28GiLQ0ubh7NyjLMVdXP2FlpQoz7uLy\nV5GGN84v3t6RwsFhpYB5wt5+uXj6NFwn7cbFJWt8ZkeVKqszyKX/B6kXc8LDI4y6dTdw8KAn7dv/\nQ1BQ3mnYtMHPT9PzMzg4AXt7c1RvVhL09LKGwM0PwcHxzJlziW+/PcfMmed59Cg01/JxAQHs6NKF\ntdWrcymXhTppZCTuixcXeFz/Szg0a0aVDh2KNdrn7Rz8Bs6ff+eBKZcruXzZT6v2hBBEv3pFYlj2\nocZfvIigY8d/WL/+Lt2778DXN+eUgAXhxAkvdu16AkhISVEwatRRjZSg6eT0Jp4bLi6VMDDQp2nT\nrOEPJk8+RXy8ylfk1q1gNmy4m2d7sbEp/PDDBSZMOMGDByFajaEg1KhRBm/vKXh6TsLHZyrOzrpx\nuLOyMtH4zI5Zs1pTAHmv5n9G8Nevb0/PntXo1MmJoUPrUrGiblQvAwfW1tgePLgOZ84Mp0MHJ5o0\nKc/evZ9Qt27BV98dHKyoVKkUkZHJJCfL84zpf2TkSPwuXSLG15drv/6Kx65dWcp4nznDxkaNuLVm\nDUdGjiQlXjcPwexIfPOGC7NmcW7GDKJfvcLrxAmOf/kl15cvR5FWuLygHxLRvr4E3bmDPI+0jm88\nPTk2ZgyuP//M8S++0Ig4ClC3rqZwqFcvd2GhlMu599dfbG7alHXVqrGyQgVuZPBJCAtLIC1NQe3a\nZenWrTrly1vQt28tqlXLO1Z8fsicElIqTUOhyF3whHl4cP677/infXsuzZmTbc7hvMicZzc5Ofdr\nSghBr167Wbr0Bps2PaBdu234+eUccqSwmJkZ8dFHZYvc+TIzX33VlAcPvuLIkSEFqv/B6fgzk5+4\n15MmNadatdL4+urOvn7YsHrY2Jhy/XoATZtWoH9/1YPA1XWUzvowMzPk7t0vuHEj7zgp0S9f5roN\nULNnTxyaNlWlfJwyBRMdhBPIDrlMxvaOHYn09ATg0bZtJMfEqFeZYv396f02ouZ/mdvr13Pu668R\nSiUVW7Zk5MWLGOVggWNXpw6lKlcmTSrFrl69LOkj9+37hKlTz7wNzFcvx2iz6RwcMoQXR4++2yEE\nF7//noZjx6JnasmePU+pU6csHTpUpnXrigysG0eEje6jevbtW5OPPrLl+XNV6sxvv22Z432ZEh9P\nUng4e/v1I97fHwEEuLlhZG5O29mzteovICCO2bMvqj3DQVCxospXJDMKhVKt+09KknHzZpD6WFJS\nGid3uDJ17kDtT/YDRipNw8srEkfHUjRqVJ5GjQqW/OmDnvGnpSlYsSJr0vScFjXSkzFnl5S5MPTo\nUZ2FCzuphb6uGTmyAU2bVmDaNJc8y2bMyqRnaEiNXlnNT5VKJR0WLmTCo0eq+DY5IE9JKVBQrnTi\n/P3VQh8gJUZzZuV34UKB284OkY1qQZcE382qRhBvBW1630E3b+J58GCu7ZSpUYNvAgPVSWUyUqGC\nJYcPD+H27fFMm5a7F5IsKYnnGYV++piUSpRy1b0xc+YFBg8+yJ6/3DB7fp5z07/G8MG/2SZ1LwzW\n1qbcujWOo0c/5erV0SxblnPiIQNjY26vXasW+umEPnigdX+OjqVo29YRPT0Jkyc348yZEVy7NiaL\nEYVSKTTUP+bmRjg6ZDR/VODz++xCZd16+TKKhITULPujX75UBRgsJgID46hbdwONG2/GyWk1ly8X\nPAbVByv4XV396NJlJ8uX3+Crr04QG/vuh/v223OkpeX/tfF/gZ5r19JrwwZaff89Y9zccGieNY2e\nnp4e5erWRd/QkLK1sz6shBDE+PnhsXs3D/76i5hXrwr0Gm5hb4+Jzbu0kHoGmi+QtnXq5LvN3Li1\nenWeZWTS7KOpxvj5cXvdOi7MmsWzTII7zMODp/v24TpnDq+vXMmqGsukq87LLr3+8OFYOTjgPHhw\nruXywsDEBENzc7XwTF/Fc/n2WyzK2jJ+fGOaNCmPs3NZRk/uhEImI+bVKwKvX1cFXtMxVlYmDBhQ\nm3btcg9OZmBsjEOLFtg3aQK8szPMr0WVg4MVYWHf0amTE126VGX16tsax69efc2QIQf57bdrfPvt\nOWJiVAH9Ns6pSjVe4kAQAzlCmXhvpJGR+eobVIlfbt8OYt2622zYcId794LVEQU8du3i4qxZuM6Z\nw4OtWwt0/+SXVatuqtccExPTmDPHNY8a/w2yrFwPG3ZImJktFm5ur4UQQoSGxouePXcJiWSeqF17\nnbh9W7cWEbokMfHDsT7IDp+zZ8USGxuxwNBQPNy2rcDt+F+/Lra0bCk2NWkiXhw/LlznzRObmjQR\nBz/9VCRGROhkrGEeHuLw8OHiVwsLcfyLL0RcYM6/+81Vq0R8cLDGvgdbt4p5enpiHqj/bq5apT4e\n5esrNtSvL+aBuLVuXRZrlHubN4v5+vpiHoh/2rcXsuScrS10zStXV7HY3FzMNzAQO7p3F8H376uP\nhYbGi9TUNBEdLRVxccni/pYt4tnBg+LyL78U2/hyQhodLZQKhbiycKE4OHSouLNhQ4HbunEjQDRs\nuFEYGy8UgwcfEH5PXqmPDRlyUOjpzReXLr3blxQRIZaXK6f+rTc1aZKjlVVenD7tLYyMFgpb22Xi\nzp13VkdJkZFima2t+NXSUiSEhRX43PLD9OlnNCwMXVy2CCEKlmz9QyLLiR454ikiIpLUgl8IIfbt\neyIaNdooJk48USxfdkFQKpXim2/OvO9h5EpaSorY1rGj2Nmjh4jNRZB+KFz44QcxX19f3N+yJdvj\nacnJ4szXX4ulZcqI9bVri6cHD6qPLStbVkPozwPxd6tWQgghonx8xKOdO8WRkSPFhdmzxfN//822\n/bigIBH+5EmBBUhhiPHz0/jMifQHlrZmlAUh8OZN4XvxopCnFu/EZty4Y8LR8XexZ9NlcXH2bPH8\n+RshhBB//XVP+PhEibNnfTTKR/v6ivMzZwrXuXNFckxMgfv19Y0Sn312SPTvv1ckJb075/iQEHF9\n+XJxe906EeXrW+D284O/f4xwdPxdwDxhZrZIXLyo6pcCCP4PznM3L549e4Ozsx1Pn4YXypqmqHj4\nMJRJk07x4EEYvXpVZ8OG3vlyKCsuUhMT0Tc0RN/IiJTYWEwzqGwyolQoiPH1xcTGJk9nHF0ik0o1\nFlA9du+mSqdOBN26pbHOkZG4gAA2NWpEWWdnxri5qfevsLfPEua43vDhVOncmdurV5McE0P9ESPo\nMH8+QqHAwOTDivj4oXD2m2+4/Vbd5ti2LZ9fvKh1AL3Ccv16ABbBd/lr3I+UTfQmwqkLn279gyo1\n7HVmwZcdSUkyzM2NUCqVpKYqMDUtPpPg7EhISOX58wgqV7ZWO4r951MvaiP4M5KSIi+STFUZ2bz5\nPuvX36F0aVP++KOXVra633xzln37nrJmTQ+GDMk+h+yHjOfhw7w8d44ytWrh9e+/BLq7o2doyIDt\n26k3bFixjOHC99/TddmyfNUJf/IEi3LliHj+nIotW6pzyD7dv58jI0aglMsxMDLCqVMnPt65E3Nb\nW/5wdibO358ZQUGYWOsmxaY2KORyPA8eJDU+njqffIJZmbzNLxVyOdE+PpT96KNiGKEmyTExLMuU\n9WzYyZPU7N27WMexvXNnfD2D2BTWmTIfOXPs2FBq1tSt6ep/jfcl+HsAqwF9YAuwNNPxDsC/QPoS\n9GFgUTbt5FvwL17sxtdfu2BhUTSzjuvXA2jTZivpX5OTUyn8/KbnWe/KFT9atarEvXshtGqV/4xG\n7xPPw4c5+Mkn2R4zLVOG7wuwSJYf4kNCODNlCt6nTlGheXN6rF6Nw9tFwsKQGB6ONDIS21q11IvQ\nKfHxBN++TSlHR4RCQVkdL0bnhBCCg0OG8PzQIQAsKlSgxddfU6NnT8rlsCgb6eVFyN27+Lu703zy\nZGxr1851ti10HM8/NTGRpdbWiAyLmCMvXaJqp0466yMjjx+HZQmlokhLIyE0lEu34zh36DZJBqXZ\nvVv3Zpru7v60aeNY5CG85XIlCoUyz4RRUVFSypTJOYDf+8i5qw+sRyX86wDDgOymI1eBRm//shP6\n+SI+PoUvvjjOihU3adNmK2fP+uRdqQB4e0eR8ft8/ToOmezdhZ+amn3OzQ4dqmBkZPCfE/qgygqV\nE5lTJBYFVhUqUGfwYIytrKjYooVOhD6ARbly2Dk7a1gemVhZUa1rV2xr1So2oQ+QGhenFvoAiSEh\nXJw1i81Nm+aYBUuWlMSFWbO4v3kzwffu5aliebZ/v8a2Qi4n9OFDon0LlrvV2MKCnmvXItFX2e43\nHD26SIR+XFwK9++H8M0357hy5bWGT46+oSHWjo506lKdzfvHs359wSJg5kRSkoxjx14wb95V1q27\nza1bgXlXKiBbtjzA3PxXzMx+ZeHCqzmWCwlJYNmy6zrvv7CCvznwEngNpAH7gP7ZlNPqabRjh3a2\nx1ZWJkyc2BSJBGxsTLCxMS2SROzt2zthafnuBlOlS1Rd+EIIfv3VLaeq/1nKOjtrbBunqz8kEjoV\nU2SHVIoAACAASURBVAgIq4oVmREURPUePYqlv+LG0NwcowwZxdLfc5UyGY8zJGjJiH3DhpRr2JD6\nw4dTvlGjHNuOefUKt0WLODNtGlcWLCDSywtFWhq7e/Rgc+PGrKtRg+srVhRo3M0mTWJmRAQzgoPp\n/88/BWojLywtjTl82BNX19fMnHk+W1VuepjinMIVFxRzcyOSk9O4fNmPbdseU716Vn8geWpWe/78\nEhGRxIQJJ5HJFCiVgl9+ucKTJ1lTbZ486UWTJpv4/fdbjB17lMTEwvedTmEV5A5AxsdiEJDZK0UA\nrYDHQDDwHeBJNvz002X8/GKZPLkZtrbmuXasUAjGj2/E8uU3cXH5m65dq3L69HAMDHTnmlC1qg3X\nr4/ln38eUqaMGd98o8oSde9eCD/8cJGbNwPx84tj+fKuhYrI9yHRYsoUpOHh+J4/T9k6deiybBnR\nPj6YlyuHbc2axTKGym3bAlCtS5di6a+40Tc0ZNC+ffw7ZgzSqChQKtUzI/Ny2RssCLmcoUeOYGBs\nTGpCQo5t21StilAqkUZEIE9JwbZWLZ4fPYrfpUtvGxJcnDWLFlOnYmCc/zADpjY2kIMhQF5oo37S\n05NQp05ZZsxoQVxcSrFHvbWyMmbp0s54e0dlkUGJb97wdN8+XKZNK1QfCQmpWcJdZOeU2qdPLRo2\nvEtgYDxTp7roNCxEYZVYg1Cpeb54uz0CleCfmqGMJaAApEBPYA2QnQQR0J5PPqmDs7MdHTp0oEOG\n8LqZSUhIxcpqica+kyeH0bt38QinqVNP888/j9i1ayADBhSNR28J//vEvH7NgYEDefPkCVW7dmXw\ngQMabwMF4e6ff2JXty5hjx7RYupUXhw7xv6MllB6evwklRZI8BeGR9u2UW/48DyD5SUkpGJpaaz+\nLE7SQ8QoFEokEtT5OTyPHMFtwQISw8JwHjqUzr/+mmPYjrwQQjBw4AGOHXsBQMuWFblyZbRam5CO\nUinw9o6iWjUb/PxiqFlTlQ/8ypUrXLlyRV1u/vz5UMyGOi7A2Qzbs4FZedTxA7KLqSCuXw8QBw48\n0cqmNTExVejpzddwaMhsy1uUHD3qKeLiktW2tO8TRVra+x5CkZKamPi+h/CfRi6TiR1du6p8GCQS\ncX3lyixlov38xP0tW8Sry5e1alMaHS32DRwoVjk5iaNjxoi0lJQcy6bEx4tLP/8s1lSrJg5/9pnw\nOXu2wOfyPvm7bVux2MJCRL58qd73+HFogdqSyxXi6NHnYt++JyI5WfP+9fWNyldbvAcHLgPAF3AC\njIBHZF3cLce7p1FzVOsB2ZHvL2/lyhtCIlEJ/UGD9guFougcVz5k3JcsKVKnneImLU0uTp70Um+f\nmT79PY4mZ6J8fMStNWvE82PH3vv3//L8+VzzDcjT0kTo48ciOhsnsEgvL7HE2lrt3HZt2bI8+zs6\nZoyGQ5zr3Lm5lg+8eVPMA7Gja9f3/l0VBLlMJrxOnxaxgYEi+P59ERQUJ+7dCxZNm24S9+4Fi7Cw\nhEL3kZqaJsLDE8XEiSeEu/trERmZpFU9CiD4C6vjlwNTgHOoLHz+Bp4DX709vgn4BJj4tqwUGFrI\nPtXMmNGSIUPqkJSURs2aZYrc/OpDIykigos//sjzgwfxv3qVtnPm4NiqVY7lpVFRWtmLv08uXvTl\n4kVfLl/25+XTAKzc/iDowimC79yh1/r1WRY2E0JDMS1TRm2znxGhVHJq0iSe7NmDtZMTg/buxS7T\n4rU2CCFIDA3FxMYGQ1PVgmLE8+dsadEC2Vt9e5uffvo/9s47LKrj6+PfXXqvKk1AEbGggmJHscQa\nYsNeE1uMxhI1lmgUu7G3aOy9Yu8NRUAUBRSp0nvvS9t63j8WVhYW2IVFSd7f53l44O6dmTt32Z07\nc+ac78HAzfV2WJMZTnExQq5exYdTpxDz8iWsBg5Ey4EDq5RTUFSEUUfJScuDLl1Cad6XvBPvDh1C\n799/r1KuIDkZ3tu3g89mI6OSOJkkldiKsFkszH73Dok+VUUX/w0oKCmh9TChF5GOmRkKCtiYN+8B\n/PxSsW+fL44dc673NZhMBg4e9MWRI/64dSscV6+ORd++lvVuV+K15NDGIwA2AFoB2Fb22tGyHwD4\nG4AtADsIN3nfyuGaIszMdGBjY/j/btAHAI0mTdBu9GjwSkqgZWpa46APAM9WrgRXSpXCkry8BlfD\njJKg3tmrV3O8fBmP9+9TYOvQEl2mjIOavj6a9+4tNugL+HywCwoQeP48oh8/lrjh+eHMGfgfPQoO\ni4WMoCDcmj5d5j6yWSyc7tMHe0xNsdvYGLFlttUQNzfRoA8AgQ3k5VIbyurq0LW0RKK3N2IePZIo\n2lcbapUmA+qGhlXK8MskuN8fOoSA48eRGR4uJh5nM1KSM98XrAYNgmnXruixePF/4ruqra2CDh2a\n4rffekBbW1lMRLKuKCoqwNm5NYYNawU7OyPY2RkhISGv9op1oNGqc35N8vNLsXbtCyxc+BCBgZIz\nHDVWGIqK+C0xES0kzPLKKcrKwqWRI/Hx1Ckc6dABCTXMujhFRTjTrx926Olhl7Exkv385N5nVloa\nPl24gEe//oqAkyeR/umT6JyqqiL69DHHuXOjwOXyoWthgd8SE6vKTzMY8Dt6FC/++AM3pkxB5MOH\nVa+TIp59iZWcLHNf/f75B4mvhX7U7Px8PF4o9FuoKLlMECqVfisEPB6cjx5F8z59oCTjhmN+YiIi\nHzyAspYWCICWmRlGnDhRpVxBUhJyIr/Ey/CKi9F71Sr0WrECkx88qLcS6b+RyZPb4/r1UBw+/B5W\nVgfw7FndYiQqYmmpiwcPJmPt2j44dswPx44FICQkAwKBfM34jT4RS0NDRBg27KIoecPZs4H49Gke\nLC3r5rL2tWk1WKiL3mFi9RY0DUNDdJ45E6zERDSzs6txZfDu4EHEvxIGlBRnZODRggWY7etbbfm6\noGVkBFZqKnIiIhD54AE6TZ+OwvR0KKmrQ1lTEzt3DgaDwShz/7MGALEE6oDQ26LTtGn4fPs2iAjt\nJEQbtx0zBj5//QVOoTCLU8c6zPi5lWSey4/tfvoJCZ6eCHFzg4KSEtpNmICMkBCpTEnxXl4il1V5\n0GLgQJH3iaR0iDVxc+pUJJTpGjEADN61S2KcgIaREdSbNkVxRgYAQEFVFV3nzYOuRc0Szd8KTnEx\n+BwO1Oohw5GUVAAXl6sICsrA4MFWuHTJBerqXzySTpz4gMTEAgAMlJTwsH69BwYNsqpXv8vdwtls\nPrZseY3iYg7s7Y3kltaxnP+XM/74+C/Lp4ICtljGHhaLg7dvZZ8ZNnb0ra0xx88PvZYtq7FcaX6+\n2DG7gdI2qurqYsSpU9Bv3RrFWVn4dOECQq5dQ1F6usgUUKtJgMHADA8PTHn8GJyioiqnm7Zrhzl+\nfhi8ezfGublhsAzaP9ySEqQFBsJm5Ehom5mJrudYludYQVERo8+fx5C9e8EpLETg6dNQVKs5oCg9\nOBgh167h0a+/IvzOHRRlZkrdn5pgVsgPwKwlV0BlssLCxI6zIyIkllNWV8fUJ09gNWQILPv1w+R7\n9xrloE9EICJEPX6Mz3fuiPTzZSU2Nhc//XQH796loKSEhzt3PleJoGUyGTUe14cePZqjRw8zDBnS\nCj17msmt3XIa5Yw/MTEfzZtXnzkqPb0Qf/zhjqysEsyd27mK7z5VEyhSUsJFYSEHS5Y8xu7dg2Fg\noA4tLWWYmWkhKUlor2UwABubxr0BWhealkkS1DYjtZ85EwHHj6MkOxtgMNBz+fIG6Y/9zJlgKihA\nwOcj8Nw5uP/xBxSUlKCirS212UCzLNipsl94UUYGHsyfj9zoaLQZMwZOf/4pU9+KMjNxxskJWWFh\nUFRTw6izZ6Gkrg4dc3ORlk5GcDA+XbqE0txcdP31VxRlZEC/Zcsa29Vv2RLPV6xA+qdPyAwJQZta\n7OJfA2tnZ9H+BENBAVaDq8+sZWxnh6mPH1d7XhJ5cXHwWL8e3JIS9Fy6FGY9as8yV198DxwQZU0r\nTEtD7xUrpN5XICL4+CRi0aJHZbP5L6SkiO8jLVvWC3fvRiA6OhdaWsrYulV+EhZEhPv3J0FBgSkx\n+1d9aUy7LMTh8MDl8jF+/HW4uY2DggIDyspVn03du5/Au3fCWbmiIhPv38+BnZ3Qxrpu3Uvs2PEa\n2toqOHNmFEpLuXj2LAYdOjTF9Omd8Ouvj3D2bCBsbAzg5jYOHTo0Q0hIBhYvfoz8fDYWL+6GqVM7\n1etGOByexH7/W2ClpCDeywv6rVrBRE5aOTVRkpODK6NHgwQCTH/2rN6yyJecnRH54IHoeNS5c+g0\nbZrU9V9t3AiP9etFx01tbfFLUJDoODsyEsc6dwansBAEoZlt5JkzUFBSqjU71+PffoOupSWUNTXR\nedYs6W+qgeBzufA9cAD5CQlo6+ICy7595dY2j8PB323aIC82FgCgrKmJX0JCoGvesBpWfA4Hx7t3\nB4gw09tbpoC4zMwijB17DZ6eCfjiJcmAoiIDz59Ph5OTpVj5khIuoqJyYGamLXcJCWmpi0hboxqd\nEhMLMGGCG/z8UjF06AVcvDgGZmbiM38+X4D377+YYng84bGdnRG8vOKxaZPQXpmZWQwXl2soLf0i\nLBYfn48hQ6xQWMiBhoYSOnQQzhjbt2+K589lt/9Wx19/vcaffzrJrb2vjZaJCWwnTPhq1yMA093d\nAYEAvNLSeg/8mSEhYsfxr16hxcCB0DYxka4/lezklY+jnz4V7RswAITdvAmXy5elanvwzp1gKirW\nSfCuMD1dtMqRFwpKSrWa/yrC53JrjbwtpzgjQzToAwCnsBAZQUENPvCzWSz86OEBpqIiOEVFMg38\nTZpoYN48BwQGpqNZMw2sWdMH2dklcHKyQOfOVT8/ampfxpF/E43Kxt+ypR6GDGmFH35ojeHDrasM\n+gCgoMBE166mKH8aKyqWHwPp6eJ23oqDPgBcuhQMKys9XL8+Htu3f9GBef8+GTt3vq6yKy+rfTAt\njYUZM25hxw4ffP/9Jfj5/ff2ChoCdX19KCgqQkFZGSra9ddmaTVMXLXxw8mT2GdhgcALF6Sq32Xe\nPOhZCTfpFFRUMGDrVrHz+lbiG3h6VtJv6JWrg1bOTywN7qtXV9lslie1bQyX5uXB759/AABZEREI\ncXOrUe1To1kz6Fpaio6VNDTQ1Lbh81OoGxhAVUcHyhoa0Gwq+6aojY0BcnJW4Pr18ZgwwRa//dZT\n4qD/b6ZRmXqISGTfT0jIg7l51R359PRCXL0ajD//fAlLSx3MmtUZixYJ7YZ5eaVwcDiG6OhcAICd\nXTN8/PhF9U5JiQkTEy3s2PGdKEGKu3sMhg69CB5P+KE/etQZc+cKzRvnzgVi4kTbKhoaNXH9egim\nTr2Fn3/ugv375Ssb+z+kQ8DjwffAAXy+e1fkoQTIlk+AU1SEjOBgJPr44PWOHWAqKmLovn1o5+IC\nAPDevh3+x45BzdAQHSZORNOOHdFy4MAG8VEvzs7GvdmzEX7nDnQtLTHq3DlYODpWKcfn8ZDq7w8l\nDQ00q8MAG+vhAS0TE4lifGG3b8N761bkJyTArFcvfL53D+DxoKCqiqlPnlRrIsqJjsaLNWvALS5G\nr99/r9WbiYhQnJX1VbO9/dv5Fnr8cqd8U1fSoA8I3Z34fEJBAQctW+pj3jwH0TldXVX4+s7GsWPO\nuHLFBb6+s7F8eU/Y2Rlh1CgbGBiowdxcRywr1sWLQaJBHxC6cxYWcrBhgwc2b/bETz/dwaNH0uv9\na2mpICVlKfr2bXweD5LIiojAKUdH7LO0hPvatXX2gmhMMBUV0XPpUthWyhYmS0CasoYGNJo0wbPf\nf0dRWhpYSUm4MXmyyBPHcdUqzAsMBLeoCE+XLcOFQYPwcOHCWlqtG+oGBuj0448w69EDFn37wrhz\n5ypleBwOLg4dipM9euBIhw54sW6d1O2TQADfAwfwcs0aUWxF5c9B21GjoKypCQ6LJdz4LzNV8UtL\n8e7gwWrb1reywtgrVzDp7l2pXFhT/Pyqlab+H/Kj0Q380qCtrYJz50ahTRsDsU3UgIAUGBioY86c\nLpgwwRbKyorYuXMwPnz4GVu2DEBY2ALs3TtELIGKsbG4/c/ERBOamsoYOrQVIiNzkJZWiKFDW0nd\ntyFDWkFfXx0uLl8vsUd9uDFpEhJfv0Z+fDy8t2xBaIUEIRXJi4tDXkLCV+5d/eg4ZQqMygdJJhOD\ndu6UqT4rNVUs45SAwxFLVBP58CGyQr8ojPsdPgyelJHRstKkXTvMfP0ajn/8AQ9X1yrnY549E0kv\nMwB4bdpUo3xzRRhMJjpOm4aUgABkhISgw+TJVVYuAj4f3RcvxvywMChUUvVU1qp/Tmkiwsdz53B+\n0CC4r1qFZytWgM/l1rvd/yGZRjnwZ2VV9cmuyI8/2mHatE7YvFkYrZqZWYQ3bxIxf/5DvHoVh9jY\n3Cp12rVrCl1dNXTpYiJKdZaSUoC3b4U+/AwG0LFjUyx1Fn7gWSw23r6dBWdna7lHzTUkWZ8/y1Q+\nNyamxmMAuD9/Pva3aIF9FhZ4/scf9erf10RZUxMzX7/G7HfvsCgqqlYvmqjHj8VWBVomJjAsc4Ml\nADqWlvA/ehSf790Du6Cgyn6EoqoqmFJufMqKgbU1WCkpcF+9Gu/278eZfv2QVkEvp/KeAYPJBFNB\nehNlUUYGXC5fRv9Nm0Qb1xVhKiigzciR0DU3xw9Hj0K/lXAyZNi2LfoLZYHrBYPBQKdp02BgYwP9\n1q3RY8kSqTeR/21EReWgR48TMDbejcWLH9W6yv74MRX29v/A1HQ3Nm70+Dqd/IqI1OamTLkhk6od\nl8ujJUseEeBK/fqdpuzsYqnrnj37gbp0OUq/zLlBfkeP0gFra3qzbx8l+/nJ1IdvTX5iIkU9eUK3\nZ86knOho4pSUEBFRUWZmjfXuzJ4tUljcpKpKaZ8+iZ1P+fBBTIXRFaDcuDiZ+laTZO+3gM/nix2X\nsljkd+wYHe/enZ6vWUPR7u5ERHR9yhTKiY0lnz17yHPbNvqnSxdyBcj/xAlR3btz55IrQJvV1CjE\nzU2q6xfn5NCF4cPpLwMDuujsTCV5eVL3PfDCBdqur09Pfv9d7HUBn09XxowhV4DWAxKll+tLsr+/\n6L0TCARUlJUlV6XNUhaL8hITiVNcXONnjM/j0csNG+hMv370aMkS4hRL/31vDPTocVxMTv7kSX+x\n82/eJIgdW1ntFyv/6JG4/Dy+gTqnXImOzsG0abfw5k0SoqNzcfbsSFHygZpQVFSAjY0B1qxxRH4+\nG/r60vnTpqQUQCAgPH48FenphSh6dR05kZGIf/UK3RYskFjn/v0IODt/nWQvslCYloY7s2aBlZQE\n/Vat0HulMC3Co0WL4HLpUrX1vj9yBMZduoCVnIy2Y8ZUSfYtkLDcltUV8dWmTd9EuVISfB4PH0+f\nRpc5c0SvqWhqQsfcHMll0hTd5s/HldGjEX77NuJfvcLo8+fRol8/5EZHw3b8eDGF0x+OHsWQPXug\nqKIitafO89WrEVWmLRR5/z5e/vknhh04IFVdbTMzLE1ORmKl3LwMJhPj3NyQExkJJQ0N6JiZId7L\nC5/v3IFey5bo8vPPMq0AKsItKUHK+/fwPXAAVkOHwtTBAUZ2dnJXelXR1IRKmetlTVHB7w4dwquy\nOIs4Dw8I+HwMl/L9awwkJIgHhpUrCYSEZCA8PAu7dr3Bhg394OBgDF1dVcTFiQu1VT6uC41q4Ley\n0sfIkTYQCAjOztZSDfrlTJvWCRoayiguls4u6OkZh2HDLqK4mAdDQ3V4eMyAkro6XK5cQUZQUJUv\ncWhoJl6+jMVff71GQkIenJ1bV7sB/S0wcXCAsZ0dmvfqBct+/ZAXG4ubU6ci+e1b5ERFYfT588iP\njwenuBjWQ4eKfOUVFBXRdd48sbZK8vJEGicmDg5oO3asKDm4/ezZVdwZizIy4DZhAlL8/GDZrx9c\nLl2CipYWcuPj8XTZMkTcvYvUgAAM2rGjTt4mwJeN2doCpGoi+vlzfDx5Eolv3iA3JgaOq1ZBVUfo\nTCDg8zHi1Cmkf/qEnJgYJPv6Ct0kiNDU1hZEBOcjR8BUVKxie1bWqDlNaGUKKu2VFCRKl9RbIBCA\nU1gIJVVVtJSQlpLJZMLQxgYAEO/tjXMDBoge0pnh4XUeHJXU1JCXkIDQGzeQFhiIaU+f1qkdeZHq\n71/jcWNn4sT22LNHKFKsqqqIUaOEKUxMTLQwb94DvH2bhA8fUjFoUEswGAyMH98ely8HAwC0tZUx\naFDNEeLS0KgGfgD44YfWWLnSEaGhGTLV09AQ6rFXFFGqiS1bvFFcLPxSZGUVY+dOH5w6NR1MJhOc\nESOqlG/b1hB7975BYmIBsrNLGtWgDwhn4WPKBtxy3f02o0eDwWDAevhweG7ejKAyP3bT7t0xw8MD\nShICpYqzsuCzZw++K/NdZzAYGHv1KlLevwdTUVGiR8nT5csRXyZXHHn/Pl5t3IjBO3dCz8IC1sOG\nIfbZM5g4ONR50Acg0nE3l+DGKC0tBw7E2/37kRcfj1bDhokGfQBoNXSomNCZWc+eKEhIgP3s2VA3\nEOZ6YJRNBupre7adPBlRjx4BEK7R29cgsFdOnKcn4j088PnuXRQkJ8N24kSo1hDzEPnwodjKLOLu\n3XrNinWaN0eP335DYUoK9Fq0EL1OUuTRrQ5OURHy4uOha2FR48Pz8717MO3WTRS8ZuHkhE/nz4vO\nWzjVHCxZWsqTmLT9W7Fr12B06tQMcXF5GDmyDTp1EqoO6OmpoX17Qwwf3gqmplqi9/XcudFwcrJA\nRkYRxo1rBysrSQkMZaPxvBtltGvXVOx3Q6GkxKxyzGQywWOz8Wb3bjitXSt2nsFgoEULPdy6NaGK\nZkdjgKmoCJUy74ryJbjNDz/AccUKxHt64mUFvZpkX1/Ee3qKlD3LiXz0CA9++QVF6ekoTEvDsP37\noaKlBSaTCbPu3au9dmW544rHOubmWJaWhvC7d/Fk+XKU5uXBYe5cqXXjiQhv9+3Dx1OnAAB2M2ei\n++LFMouRlWM1eDAGbt6MwvR0sdcrC50N2LwZhm3aIDsiQu6++Z2mToWWkRGSfH3RvGdPtBhQu8aL\nadeueLFmDVL9/aHRtGmNgz4g3AwGvhh/DST45suCWffusHRygoDHg4DHA1NRES/+/BM+O3dCSUMD\nI0+fRhsJE6bqyAwPx7mBA1GYkgItExNMd3eHYRvx3NV8Lhe++/cj6PJlaJuaov3EiegwaRI6z5oF\nAZ+PWHd3GNnZoZeEpDEVcXX1EAvYLEhKQm5sLJp26FAv9c66wmAwMH26ncRz+/cPg4qKopjnoaIi\nEz//7CCxfJ37INfW6kfZPsXXITAwDYMGnUdmZjEsLXXx8uUMIO4j3NesQVZoKNq4uGDwzp1Q0/t3\nyDPXRGl+PnYaGorNAGe+eYPmEgSzLv3wAzJDQzH++nWJ8ryS+HDmDO789JPww8RkYuKdO7BxFs9I\ndLJnTyS9FS5vlTQ08PPHjzBoJZ2bLJ/LxW5jYzAYDCxLTa1T1Ot/gcdLl8LY3h4q2tq1CrzxuFzc\nnzcPEbdvQ9XAAMMPHYLVoEFye4jFe3nhdN++ogFEUV0dq3JzoSAhE5okrk+ahJArV0TH7caPx7ir\nV6uUY6WkYG/z5tBr1Qq/BAVJzLRWHbGxuVi8+DEePozEoEEtsXfvECjE+uHq6NHgs9nQNDHBTG9v\nsRXMv5F/vVbP16RTJyPExCxCUlIBLC31hEtBy37QtbBAemAg7GbM+E8M+gCgqqOD4UeO4P68eQCf\njx7LlkmcwRMR+m/cCMM2bSS6dVaH/Y8/QtvMDGkfPqB5795V9P65xcWiQR8AuEVFSHrzpsaBv7Sg\nQDSrLUxLw8Tbt0V/i2SSvyJZERESI1q/JkN27waDwZBKc19RSQntXFwQfOkSSCCAkqpqnQZ9Pocj\ncTAvzsoSG2l4xcXgFhdLPfBXdhCQ5ERQXMxBUUYGpjx6hOzISPDZbJkG/hYt9DB0aCt4eyegb18L\ntGnTBMenrwefLVS7LExJwbuDBzFkzx6p2/yv0Cj9+OtDQUEpCgs5UpXV1FRBmzZNxOx/thMnYkl8\nfEN175vRZfZstHZ2xsr8fAzZtUviIMBgMGBsbw8lNTWZc9Naffcdev/+u8QkL0rq6iK/b0Ao/9uk\nhvYFPJ6YaUqneXOYOzrC3NFRbNDnlpSI1atsvpEHJbm5yI2Lw705c5AbF4finBy5X0Nayv9n0pq5\nNJo1g5G9PZo7Okrcm5GG17t2gc+p+n1qMWAA9MvMSQDQfsIEqMpgNnFctQoqZeVVdXXhuHp1lTIb\nF1+EoW0HWA0ejG4LFohMmbJgY2OA1NRlcHAQau1UXi02xtVjVFQORoy4jD59TuPqVeGmLhHBzS0E\nu3a9Rni4fPI4NBbo7NmPIt/UuvgHu7q+JGA9MRiutH27l8z1vzU8DodKcnPl3m52VBQd69qV1gP0\nT+fOlPn5s9yvURuZnz/TRWdnOtm7NwVfvVptucS3b+lwx460UUmJrowZQ4UZGdWWfbZypdjxvV9+\nIVZamuhYIBBQ8LVr9ep3uc+9K0BnBw4kVnp6vdr7GggEAvp4/jydHTKEjjo4kPfOnTLFChARFefm\n0p05c2iLhgadcnSkmBcvqpQpysoiv2PH6NPlyySoFBshDaz0dIp79YoKK72nyTEpNPWHv+knZgty\n6rSF3O8HyNx2dcR5edFWbW1yBehA69bESk2Vqp6np2yxK3VFIBBQq1Zf/PaZzA3k55dMv//+RPSa\nhsYWCg7+8p7h3+7HP2PGbSgpMTFhgi3c3EIwYYL0XiAREdlwdX0FgAEiYPVqd0yY0P5fk0Ix6ulT\nuI0dCw6LhdYjRmD89etyi1zUt7KCzahRUNHWhoWTU40mC25pqURvn/pi2Lo1Jt+7V2s5s+7dF2lK\nEwAAIABJREFUYdq9O9gFBbD78UeJYl05MTF4smQJoh49QkZQEPpv2gSf3bsRdOkSwm/exKizZ6Go\nqoqI+/cRef8+8uLj0Xn27Dpt5Knp6cGsZ09omZhAo1kzqdUeC5KT8f7vv8FgMtFt4UIxOeX8/FLo\n6Mj/PY568gSthgwBg8FA+3Hj4LlxI3JjYjD53j0xDyZp4LBYYBcUgFtUBFZamsQE7OoGBmLxELKi\n2bSpxPfTwFAT3XLvI0cQi2apF+BgN6PO16iMhaMjlsTHozA1FbotWtT6WS8s5ODFi1js3+8LF5e2\ncHAwRrduDWdqLC3lISrqi/KAQEAICcnE+fNf8kEUFXFx82ZYvdIxNqqBHwAuXvwEX99kPH0ajYSE\nfMyd20WqL0nlLDVEwjeoPrDS0hB+8yZUdXVhO3FivXzIa+Pe7NnglGmrRNy9i08XLsD+p59qrENE\n4HO5wg9xLWnwbCdMQN8//qhRRhcQJiH5rpIM8dfGfuZMfP/332JJ2HOio0XxA/otW6LloEFI8PaG\neZloWZe5c5EXGwuNZs3QasgQCPh8eG/bhqzwcDRp00aqQb+6h57DvHnQMDREkZTKnmwWC6ccHZEf\nFwcACHFzw7yPH6FUlprx99+f4tgx6T1gaiM7IgIJr1/j1YYNcFq/Hi3694eqvj66L1kCPUtLoY6/\njMngdZo3h7aZGQhAk/btqwT2NSQqWlow7NwVplZGyCwAtE1N5dq+mq6u1JMATU1l0eCfm1uCCROq\nJvRJTMxHRkYRbG2biuRg6tw3NSX06tUcPj7C2A51deGxsbEm0tK+SGmYmtZPH6nR2fiHD7dGSEgm\nwsKy0KlTM6lnRvb2xhg27IsdedSoNmjXru7SrkWZmTjetSseLliAm1Om4KYMGZzqQsXctoTac91m\nhIRgr6UlNquo4HjXrsiJjgZbgsZKOaJBsxrt+PzERFwbOxZvd+/GhaFDkREcLPtNyInmPXpAQUkJ\nJl26gMdmIy8+Hm/27EHEw4fITxJqKzVp1w7LUlNFtmvt5s0x8/VrDN61CwKBAEwFBZj16IHJDx5A\nUEForSY8N22S+LpG2WxXQ8KsVxLpQUGiQR8AciIikBMVhaSkAgwZcgEnTnxA167H8elTmlTt1Ya+\ntTWS371Dfnw8MkNDoWtpCVVtbXSbPx/Ww4fDqJN4Rjl2YSFSAgJqfZAZd+6M+SEhsB4+XC79lIWR\n6xZhzJlTmLi//jpA5ZBAgLf79uHOrFkIluBBVB3a2srYtm0A7O2NqqgCnD8fiBYt9sPB4Th69Top\n9f5iTTx4MBnLl/fErFl2ePlyBlq10se5c6Nga9sEOjoqmDu3M2bMkOwO+m+ENm9+RQKBgFxdX9Kj\nRxF07164TPYxDodHDx58psePI4nPr5+GyKeLF8X0adYDxC4qqlebNfFqyxZaX3adzZqaYrbqivC5\nXAq+do0OtG5N6wFaV9a/vwwNq+jsyErAqVO0TUeH3NeurVc78kQgEJD3jh3kCtAuY2NKeP1a7tfI\nioigSyNG0EZFRbo4YgRlRUbWXqkG8pOSaJOKiuizs0VDg4qzs4mI6OTJAGrT5hCtWvVMHl0X4bVt\nGwVevEg+e/bUWC4vPp72WliQK0BbtbQoztNTbn2Qp24PEVFxdja92rxZbu09/+MPse90TXtNFeFw\neERExOPxq+g8NW26U0xH5++/38mtv9KCf7uNf80aYTKH9ev71am+kpIChg+Xj8udRqUUd6q6uvVO\nCVgTff/4A+oGBvDcsgWFqam4NX06XC5dEtNDIYEAl0eMEEV9lvvlUFl/67Mc5xQWCoOtUlMRV5a8\nhOoRlSkvGAwGWg4ciNYjRoDPZqO5BK+h+mJgbQ2Lvn0R9/IlWvTrJ+ZmWpCUhHs//4z8+Hi0nzix\nSmCfJLRNTTH+xg28WLMGDCYT3+3YATV9YbSlvb0RQkPnIyAgVa730HvlSjAYjFqVHl/v2oX8Mq81\nDouFl3/+iR/Loq7rS9TjxzDq1AlaUqa4rImIBw/wdNkysFJTkZ+QgMG7d4t0fMopSE7Go4ULwUpJ\nQacZM9D1l19qbDOmktRE9LNnaD9+fK19UVISahwpKAgNJPfufcaHD2lwcvp35NyQRKMz9XxLKsov\ntxgwAI5//AEldXVomZpinJtbnaNFpcWkTCxNwOMh+ulTRJYN8OXkREWJBv2KKGtqwnHVKpTm1V28\n6dnKlWg5cCCU1NRgPXQoACDk2rU6tydPtM3NMfH2bYy9dg08Nrv2CnWgqa0tlqeno0k78TwKN6ZM\nQdTDh8gMCYHHn38ixM1NqvZaf/895n38iDnv38Oqgq6Ovb0wEK1Ll/oPjhXdWcsf0LU+qCs9GKQ1\ng9UEn8vF61274LF+Pe7//LPUKS5rovX330OvbPO128KFVQZ9AHAbPx7ht24h2dcXD+fPF8uVIImm\nHTuKHddlonT48DuMGHEF69d7oH//s5g82RYKCsL3vEsXY0yb1rGWFhoH/xv4K3D0qJ8oGxeDwcDA\nLVvwR1ERliYliX15Gwr/EydEwTkE4P2hQ2LnlbW0wKiksOjk6oomtra4PW0adhkZIURCIpX04GD4\n7NmDiAcPqpzL+vwZp/v3x/vDh3F2wABkR0UhJyYGXtu24dHChfDcuhWZ4eHyu8k6oGFoCAaDAVVt\nbShWSgJSGxU3iGui1ZAhUFJTQ6shQ8Rez6p075WPa8N3/36ZysvCiwqxDtLSc+lSaJXFQihpaKB/\nNfsa0pD8/j189uxBoo8POk6ZghR/f2SGhaH9uHF1brMcEghEuY+PdemCC0OHojQ/X6xM5f9tbf/r\nYfv3w37WLJh264Y+a9ag26+/ytyvcrE0QPgMzc4uQXz8EgQEzIWPzyxoacn2+fxWNCpTz7eioICN\nzZs9cflyMJ49i8Evvzhg0CDpE2jLCzU9PdGTmAGgQEXcE0PL2BjfHz6MR4sWgfh8DNiyBWoGBkgp\ni4rls9l4OH8+2o8dK6qT4ueH0336iDJDfffXX+i9YoXofNSjR7AeNgz5sbGwdnYWmTn4XC6KMzNR\nmpcHLWNjkEDQoF5N8qYkNxc5UVF4umwZ+m/aBG0zs2o3tmvC+vvvEXj6NABhsI9VJX2j6kh8+xY+\nO3ci3sMD2VFR6L9hg9zyyGZHRuLRwoWIfvYM2eHhGLp/v9T3pteiBRaEhiIrPBw6FhZ1SkYOCF1H\nLzs7CyNwGQz037gRE+/cQV58PHhstswP6MowmEyEurkh6c0bAED0kyfw2LABQytE2VoNGoTwW7cA\nCP83Lfr3r7FNZU1NjDhxol79srDQhbf3FzXVoiIOmjXTgKlpzdpJjY1/zze5Grhcfr3zxGprq2DE\niNZITi4An08NNujHeXri45kzyKsmMthx1SqY9+kDFtSRYdwbYebjkZJSAA7nS3h7l7lz8WtkJNaU\nlKD3ihVVoip5bLbY+xF85YpYOsCPZYNYSU4OHi1aBK+tWxHx4AFGnjsHswraPeqGhpjm7g5tU1N8\nvnMH8Z6ecnkPvhYq2toIcXNDgpcXnq9aBSUZpZPL+eHoUQzevRvdFi3CNHf3GsXqKtK8Rw+o6euj\nJC8PVoMHyzV5uIG1NVoMHAh1fX20HDRI5geaipYWTLt2rfOgDwCfzp37IrtAhNgXL2Dj7IzuCxbU\nKiAnLaxU8X2QokrHYy5cQN9162A3axamPn0KEwf5CplJYu/eIRg40BKKikwAhJs3wzB+/PV/Xa7q\nf/3A/+xZNN6/T669YC2w2XyEhS2As7N17YXrwNv9+3HWyQl3fvoJ/3TqhMywsCplVHV08JOnJ4Y/\n/YAbglE4cSECp09/hKKi0LxDRGCzWHi+ciUYTCZKcnPRYfJksfSA/TdtErPzahobi12jfONNTV8f\n1sOHoyQnB3oWFmjRt6+Y3EK3+fNh6eSE0rw8PJg/H1dGjcK7w4e/6geclZZW5ctfDr+WZDBMBQU0\nbdcOvVasQLNOnaAlox97OQpKSui5dCmG7d8Py759Zapr7uiIJXFxMunLSItx585YmpKCpl/Rv74i\nGpXeT1njBKTB7scfhTlRAYDJRIdKLtVK6urov2EDRp44UetsX140aaKBVav6lJmEGQAYuHUrHMnJ\nNbtfNzb+teqcAgFhx47XOHEiAMrKCpg/3wELFnRrUC+U+ni5HGzdGjmRkaLjXqtXY1A1gVJEhCFD\nzkNJSQGbNw+Avb1w8E7y9cX1iRORHxcHq6FDoduiBZwPHwa7sBDJ795Bs1mzKho7fA4Ht2bMQPjd\nuzBs3Rrjr18XzRCjnjxBk7ZtEfvihfBLJgE2i4Wj9vZQVFXFL0FBUt0/p6gISurqEssKBAKhtr2E\nczwOBxnBwVA3NERxRgbC794FiGAzYgSMu3QBk8lESU4OLo8ciQRvbxjZ2WHS3bvQad682n4oa2iI\nfv8P+VGan48bEyci9uVLmDg4YJybG7QqTTLkQdyrV0jx94d5795Sr7YaGn//FDg4HEP58KmkxERm\n5u91jsZ+8SIGAwbUPblKXdQ5G+3A//FjGuzsap5FsNk8GBvvhoaGMhISljS462HY7dto3qtXnZbI\nJ3r0EKX2A4BBu3ah17JlEstyODwUF3Oho6OKzMwiNG36xaPh9o8/Iu3jR7BSU8HOy0P78eMx/O+/\nqyT+lhf5iYlQUlcXDZ7SpNsLunQJelZWEr+ocZ6eUNHWhrGdeAAKp6gIZ/v3R8r792AoKKDn8uXw\nO3wYIMLEu3dFM7qHixbh/cGDonrtJ0zA2Aryvv/j/yckECA9KKhKsJo8qThpWb/+BbZufQ1lZQX8\n88/3mDat9usSEW7fDkd0dA6GD28NfX01vHoVB1fXV1i7tg86dWoGW9tmtbZTmboM/I0JIiKKjMyi\n58+jqGfPE+Tjk0BZWdUHTUVHZ1NAQAp5ecVRWhqLiIiysorI2fkiGRvvokmTrlNxMafeARI8Doe8\nd+ygo5070+URI+jTpUsyt5Hy4QPtNjen9QBd/P77Oicgz4mJIYFAQOeGDKEDrVtTvLd3ndqRB+yi\nIro/fz6d7NWLnq9ZQ1w2m16uX09/GRjQbhMTenfkiCioR8Dn05t9++hY1650olcvel/hHBGR37Fj\nYsE1W7S16cqYMVWE2q5PmiRW7uzAgV/9vv8r8Dj1/27ICp/LlXugV2pgIPmfPEluEydSgo9PgyVf\nj/P0pNSPX4QkuVw+sdlc4nJ5UtVfu9ZdFOilorKR3rxJoI0bPQhwJReXq8RmS9dOZVCHAK7GBBER\nZWYWkaPjKQJcacWKpzJH4M6YcUsskm7NGvc6vZmVyUtIoPUA/d2uHXHZ7Dq1IRAI6jzgl1Oan09E\nwgcJl82m9ODgerVXH+4vWCA2CL/etYs4RUW028SEDtrYVPmCc4qKaIumJv1lYEAXnZ3pgLU1PVqy\nhAR8vsSBvzxKkl1YKGoj2t2dNikrkytAG5jMeqtv/n/Ga9s2mevkxMRQXkJCna8Z+egRJb55I1Od\nZD+/Gs9nff5Mey0taT1AH86eFTvH43Ip2d+fsqOiZO5rOXwej3z27qVj3brRyd69xSY07u7R5O4e\nI1U7FhZ7K4xN68nF5QodOfKejh/3p9Wrn8vUp4oPCdRh4G9Um7tJvr7Q11eFrW0TrFrVGzY2BmAy\nv6xg7t6t3Yc6Nja30rEwqCk9vXodG2kozszEtKdP4TB/PqiWjcXqYDAY9XZz+3ThAtICA2FsZwdF\nZWWZdfPlSdqHD2LH6R8/ojA9HbPfvcOEW7dQUkm7npWWhrFXr0KvZUtE3L+PnMhI+O7bh/eHD6PD\n5Mkw6doVgNA1b9i+faKAuYr2+ZYDBmCuvz+cjx3DrLdv6+0zXpKXh4yQEHAreD79V0ipJgl5sp8f\nrowaBe9t23B39myp8xg8mD8fB1q2xD5zc5ljCAR8Pjy3bcODBQtw+8cf8fbAgVodBQrT0hDv5YV7\nc+Yg3tMTubGxEsvpW1vDuHNnOPzyi5iHE5/LxaVhw3C8SxcctLaGz+7dNV6vuuBApoICusyZg6yw\nMORERqLzzJkAgD173mDu3PuYM+cedu16XWuCHBMTLQjHaOF937nzGSwWG7Nm2WPjxn411q3I/fuf\nYWNzECNHXkZERLbU9RordPH77+ndkSMU5SWcERQWCmfWnp5xtG7dC+rU6QgdOfKe8vJKyNs7nlq1\nOkD6+n/Rn39+mdUfOuRLwHrRU/XGjRAiInJxuVpFZ+PfBJfNpifLl9MWTU3a2awZ+Z86Jfcls6y4\n//mn2Cw94NQpqer9bWsrVu/Br78SkfAeUwICKD8xUax80OXLcu87kVCbfZuOjkibvSAlpUGu87Vh\nFxZS3KtXdPH77+nTxYuUGhhYpcyDhQtpo5KS1CumlIAAsf+ZK0B58fEy9YvLZgvNgKamUn12eRwO\n3Z07l1wBuvTDD1RaUCCxHJ/LFZmtKq4OQ27cEO8zkylxxZ3g40M7mjYlV4CuurhINIFlR0XR5/v3\nKcTNTfQ54fMFZGy8i5o23SGVuScsLIM6dTpCGhqbSUdnG7Vrd4hKS7m11qvI5cufiMH4YtEwM9v9\n79fqiXzwAHmxsZhapqmhoSF0g+vWzRTr13sgMDAdRkYa0NFRhYvLNaSnFwEANm3yQp8+Fhg0yAoL\nFnSDiYkWPnxIQ58+5mjTxhB9+pyCt3cCOnb8B1euuNRpA+Vbo6isDMeVKxF85Qq0jIzQuRbJ5q9B\nf1dXqOnpIe3DB1j271+tZ1BlbEaMQGa5+ieDgdbOzsiOjMTtH39EfkICOk6diu+2bUNufDwCz53D\n+4MHkfX5M9q6uKCZrXQ5Gkrz86to0Ec8fIjWFZQm3VetArssGjQnIgJv9uzB4J07a207NzYWNydP\nRnZEBGxGjIDzsWNyy50gD5Q1NJARGorIBw+QGxuLKRJkPpr36oW+a9cixc9PqjYlpUasnD6xNljJ\nyZju7g5eaSkK09Jq9QJSUFKCQevWcNqwAdyiomozcFXMolVxdSits8fdmTNRnJEBAAi7cQMfTp+G\nw9y5YmX0rayqxEukpBTgwYPJYDIZSE0tRPPmNec8aNOmCT5+nIf09EKoqChAICDk55eKOW/UhpKS\ngpjqRlISS+q6FWlUA3/r2UsQ5R8KD38WRlSQ4FZRUYSDgzF+/rkLiIRBWxkZRWJ1k5K++NGOHt0W\no0e3FR2PG9cOysqK6NzZCFlZxQ1+Hw1FSV4e5gUGoig9HezCQon6JV8TBpOJnr/9JnO9AZs3Q69F\nC2SFh6PV0KFo+d13ON6tG1LevwcAvN6+HUZ2drCdMAHK6uoozsxESU6O1IM+AHhs2ID+GzdCRVMT\naYGBSPTxgc+uXShKS4P18OHQNDKqMnDxJQxukrg3Z44oh/DHM2fQtFMn9FyyROq+yZOPZ8/i0/nz\n0DIxwaCdO6HZrBm4xcXICguD1ZAhUNHRga65uVidwMA0dJo4EQDEHoQ1YdK1K9q6uCDsxg0AQOc5\nc6DXUjYXxLokNe86fz6U1NSqpNmUhtbOzrAaPBjRT58CDAYG79ol0dRakituHq5soqwOMzMdmJmJ\nD/ZsNq9WTf5mzer+vXVwMIGKigLYbKHGkq1tE3xDBXW5QCoqm4iBdQT8KZY6USAQfPEOKfs9adJ1\n0XKnadMdlJycX+3yKDIyi168iKH58+/RwIFn6eLFT1RSItsSSxIhbm71buN/CPnLwEBsWf5qyxYi\nEnr7xHt51So3XE5Jbi5dnzSJNqmo0EEbG4p58YJ4XC5dGz+eXAHy3rlTVDbi4UPapKoqlHw2MaGc\nGMmbdJySEsqN+5J676CNjVhfHy9dSkREsXWUOK6rd020uzutr9CPk337EpfNpuPdu4te+6dzZ2KX\neblER+eQu3sMDRx4lvz8kikvr0Sm6wkEAkp8+7bWzdbGBJ/Ho7TAQLH/X2U8t2wRvV87mjSpsawk\nysekqKhsOnfuYy2l64+vbyJNn36TFiy4T2lprDqZehqT7ycBrqKDdu2aICRkfrWFeTw+Ll4MQnZ2\nCcaNa1frMisnpwROTmcQHJwBd/fpGDCg6uxDmqc1IBQ2C7t1C2/37kXPpUvRbuxY0TKwtKBAbiHr\n35KYmBy0bKn/1a53/5df4P/PPwAARTU1zPb1RbMOHURBcyRD8FyshwfuzZ4Nk65dMfbyZQBCQbMm\n7dqBU1goli4wPykJ+XFxaNqhQxXTEBEhLz4eCZ6eyI6MhP3MmdAxN4fXtm3wKNvcVFBWxg+nTkFB\nSQleW7ZgwObNMO3eXaZYD88tW9Bn9WqZtZB8du/Gs+XLRceK6uqY6e2NY5USq8/x94dJ585ITi7A\niBGXERCQhrVr+2DDhv5izhONCSJCwuvXKMnJQevhwxs8KXrsy5coSEpCc0dH6Mu4Mnn1KhZ+fik4\ndy4IubklmDGjE9atcxLJOTc0dfHjb1SmnooYG9e8HFJUVJApC42OjgocHZtjzpzOqG782LrVCxs2\nSA795hQXQ1ldHQBgaGMDdl4eijMywC0uFrP9vd27F11/+QUalb74RZmZYKWkNGiAiTxgs3nIyyuF\nq6sHlizpAQsLXRgYqMut/by4OOTGxsLY3h6qFdLfff/33zDu3Bn5iYloN2aMSDJXarnhCqjp6WF+\naCgyKqg19tuwAUwms4oniY6ZGXTMJOdQZTAYyI6IwKPFi8FhsaBraYnOs2ah75o1aNKuHbI/f0ar\noUOhb22Nqy4uyPj0CakfP6K1s7NU/SzKysKzFSsQduMGYt3d4bRhAyz79JH6Ppv36gUwmUCZN4ml\nkxM0mjQBQ0EBVCa3zFBQEOkEmZpqo2PHpnByMkf79k0azaCfExWF6xMnIic6Gm1Gj8aI48dxdfRo\nRJTlaFYzNMRvCQmi1JUNQXmA4PPVq/Hdtm1S1eHzBTh2zB/XroVCR0cFWVlFyMsrxdKlPUWDfl5c\nHHQtLRuq2/8JaNKk66SltZW6dDlKUVHZcl0e8Xh8iX8TEcXE5NDYsVdJSWkjjRx5mUJDM8TOl7JY\n5L5undhrb/btoxh3d3p78CARCb0J7s6dS1s0NGh/y5b0+d49IhIuAyMePqSHixbR9UmTKPTmTeLz\n6hao8TXgcHi0bNkTAlzJwmIv+fom1l6pjNruK+zWLZEP/p7mzaX2B/8WgUblcNlsOtO/P50fOrRG\nE8CDxYvJZ88e8jt+XKb2o548oU2qqnT7p5/q1L/P9++T28SJ9HjpUirJyyOiskxqurq0TUeH/E+c\nECvPYgm9WgoK6hdPIk9O9ekjZjrz3L69igfR0xUrGrQPaUFBdG7wYNqopESXR42S2tyTkJBLioqu\nZGKyi27fDqWwsAwKDk6nUhaLCjMy6NygQVSYkUFsFqvB+o7/QgDXt+Lw4Xekrr6ZNm9+JfZ61NOn\ndLhjR9qqpUW3Z84UBVBV3nMgIkoPDqY9zZvT+SFDxNrIiY2lnUZGtElFhdIkuNY1Np4/j6aRIy/T\nqFHSu1FmRURQ+P37NZY50rGj2Jf52cqVtbabGR5OwdevS90PeVPKYhGnpIT4fD4VZWdTVlYR+fhU\nfWCVB/XJ+pCKefGCirKyKETO91hxX6zebTWwG/S+Fi3EPhcPlyypMvCX76M0JC/WraNtOjoyPbyP\nHXtPv//+hPbu9RFTGSjMyKALQ4cK91569arioixP8G9356wrfL4Aa9e+wMuXcbC3N8Lu3UOgri6b\ne52lpS7S05fDwyNO7HWrQYNg0KoVClNS4DBvnkgTR5IJggQCLAgLQ2ZoqJh+vaquLtqOGQNVPT0o\nyBDAVVjIBpvNr9HUEvPiBVoOGCB1m9Jga9sUt29PRGZmkTBxeQ22ZyJCwMmT+HTuHNgFBcgKDUXP\npUvBVKhq31SopFJZ+bgyASdPwv/oUVG7fVavbnBbb2XKPaeICM+9M/DoURSSk1kYN64dJk3qUCbP\nC5ECp6xuneUmhnYuLlKVDwpKR4cOtbsj12Qa43L5SE8vgplZ7XtRnOJiBJ47h67z5knVv7rQcdo0\neG7cCABQVFWF3bRpyAoLQ/STJ2AAUNLUhOPq1Q1y7fj4PFhYCE2OzXv1guPq1Uh8/brWegKBAHv2\nvMG+fUL9rSVLeoglYtdo0gSW/ftDTV8f2ubm0K7GnPg/6jHj377dS0ymYcGCmmeesiAQCCj6+XMq\nzc+npHd1S6TM437xIJLWzJOXV0Jnz36gw4ffUV5eSZXgs8zwcAq6fJn2W1lR0OXLMnsiyBMeh0N7\nLSxos6qqyNwgiVgPD9qqrU2uAP1ta0tFWVk1tsvn82l/q1a0UUWlxna/FunpLDI330MKChsoJCT9\nq147J6eYAgJSqHePo+TlEUVxcbmicxmhoWKfseoQCASUmlpA164F07p1LyglpaDGwKOIBw/owrBh\ntKd5c3ry++9U2oDmipDr18lr+3axYLOwu3fpVP/+tMfCgi4MH04Fqalyu15OTjF5e8fTokUPKSur\niLhc2Vc1PB6fTE13k4nJLonvIystTex3XUlJkRy4Vg7+v874P30SDzkPDBQ/zs4uxsOHkTA0VMfQ\noa1k2igsT/YNAKZlkgKyolBhlippJiyJGzdCsWDBQ/D5BD5fgAULuomd17OygtfWrciNjkZqQADa\nT5hQp77JAzaLhYHbt0PLyAislJQq3jHlWDo5YWliIgrT06FraVnr7JhdUIB+GzciLSAABcnJEtul\nr5gQXlVVEePGtYWJiTa43JrD8+V+bSXCqpHr8TrRFJP7b8fVCz/A1GgIsj9/xru//0bzMtliQxub\nattgMBh4/z4Fc+feQ0F+KRJuXcKi33rBvppgQOvhw/F23z6wkpPRYdKkBo0bkbTiKYiPR8LLl6K/\nr4wahTll8RP15cOHVEyefBPp6UXQ1laBq2s/mdtISyvEs2fTRH9X9unXbNZM7HddWbLkMS5cGCNX\nL6FGpdVTV777TjyQpGIGrezsYjg4HMP06bcxfPgl/Pbb46/dvToxaVIHdOzYDPb2RpiF/mCfAAAg\nAElEQVQ2rVOVwU1BURG6LVpgxMmT0DQ2/mqDnyTU9fXRYeJEWPbrhyZt29ZYVkVbGwbW1lKZROJe\nvoTvvn0IungRAcePSwziCS5z12wIqJIHkJaWCnbtGoqlS3uhY8evG/0deukCVBLfowdeoxmlwnf1\nQigoKyPF3x8Bx4/DffVqcItrD040y/eDiXIeLCgWukFuuDNzJkJv3pRYVsDno/3EiZgfEgI2q24R\novUhJypKbCqb4uuLOA8PAEBpXl692h4woCU6dmyGwYOtMGZMWygoSDcU5ueXYvToKzA13Y1ly57C\n3FwHbds2aZCJQGoqC4MGncO1a6Fo1+5v+PunyK1tecz4hwLYB0ABwAkAf0kocwDAMADFAH4E8EFC\nmTrz00/2YDIZ8PCIg52dERYu/KIDf+9eBOLiviRpPnToPXbtGizKatVYYbP58PD4EUwmAyyWZPGo\nvmvXQkFREYIy173/Gm1Hj8a7AwdQmpcniuAsJ97LC+G3buHznTvIjYuDw7x5UNeXb9zB53v3YNG3\nL9TK3E4rPly/9oOWW1yM9giBKrgohRK4xTpgMBgwsrNDh8mTUZqXh2Y1uAqX5ufj/ZEjeHfqPMaV\npEGAXJRCDQwI89m2GzOmSh2mgoJIkMywTZuGurVqaf3DD3hbKVl9zPPnsOzXDy/XrcN327dDSb1u\nrsYcDg+XL7vAwEBdJgHHVaue4/btzwCAq1dDYGysgTVrnLBtmxfmznWAlZUe9PTk43ZqbKyFyZOF\nsSw2Nobo0sVELu0C9Z/xKwA4BOHg3w7AJACVp3zDAbQCYA1gLoAj9bymRGbMsMPp06OweHEPMf9k\nXV3xzVQdHRWpn+7fEl1dVaipKUFFRRGGhpKzR2WFh+P5qlXw2bULHClme9+K2tIkVoeAx0P3JUvw\na3h4lRmeuaMjcqOjkRsTg6bt28t10Oex2fDZsweemzbh4fz5CKtmRiwNnMLqBxUBnw+vbdvgNn48\n3pcFr1VHhylTYNxSmGlMFVz0LQsg07OywpgLFzDh5k2JejrlqOrowMjODoWRodDSVIECCOooBgEo\nTE8Hn8NBVkQEAk6eRGJZgvNvTcuBA2HaTdzEqW1uDrfx4+H3zz840rEjYstWALKirKwocpqoTkKB\nlZYmdsxm8xAcnCH2WmxsHnbs8Mbx4x8watQVhIZm1qk/1eHkZIHnz2dg2bKecm23vjP+bgCiAMSV\nHV8BMBJAxYSyIwCcLfvbF4AugGYApNOCrScjR7bBrFl2OHnyA3R11XDxoku1s7XK9uKM0FBEP30K\n/VatYCNlUM7XIjc2Fqd69QKnbAke/+oVpjx8WK82syIiYNi6tTy6J4bP7t3os3KlzPWYiopoM3Ik\nAECnkt4Mg8GASbdu6LZ4sUhoTVp4HE6NeXAVVVTQ2tkZz5YtA7+0FKPOnJG574Bw0PfcskUsICgn\nKgpK6urQMjGBx/r18NqyBQAQ6uYmkv+VhEaTJpjr74+E16+hZWICY3t7AEB2RASCLlyAepMm6Ll0\naY394XM4mPr0KeK9vFCal4dEb29kff6Mz3fu4PZPPyH89m3wiosBBgPOx46hy+zZdbrvguRkJHh7\nw6B1a1E/68rk+/fxcMEC5MbEoK2LCxzmzoVey5bICA6GkZ0dWvTrV6/2a8L/6FG0LQsmTE8vRL9+\nZxAenoWKQbITJtjCwkIXISHCAb93b/NqWqsb5dHzXzOKXhrGAjhe4XgqgIOVytwD0KvC8XMAXSS0\nVa+d79ooLa0988/x4/6iv1M/fqSNZcFGrgB5bt1abb1vIfcccOqUmJ/zeqDaBDG5cXHk9ddf5H/y\npJj3B4/DocgnT6gkL4+yo6PpdL9+lBURIZbxShb8K/k/p336RG4TJtBmdXW6M2tWvRJ4yJM7s2fT\n/pYtaYuGBj1ZtkximQQfHwq/e5e8tm+vU0anqGfP6KCNDW1WU6MbU6dScU4OuZXpBbkyGOSzZw+d\ncnQU+x/emDJFpmtkhIaKtIZcAbro7Fxj+YqxJ+V/n+nfn0726kVuEyeK9eUfe3uZ77m8T9v19ESJ\ncgJOn65TOzWRHhREfB5PLBuWPOGUlNDDRYvIlcGg7bq6FHTlCq1c+UwsiYqi4gYaNOgsFRayKTVV\n6HWTmVlUJThU3kRHZ1NYmPj3E9/Aq0faC1aeYkus5+rqKvq7X79+6CfHp3lNGjzx8Xk4dOgdjh8P\nQGhoBpy7KiD+6BbwORwQhJ3/eOoU+lTjSxx85Qqshw8X2YK/BgbW1mLHepaWEmexBcnJON61K4oz\nhTOSmGfPMPbyZcS4uyP2xQskeHqi/eTJSPDwQLyHB+78+CNcrl6VqS/ZUVHw/+cfBJw4gazwcHT5\n+WcYWFujWYcO0G3RAnw2G+aOjtUmRf9a5Cck4PmqVaINYQLwZvdumPftizYjRoiVbd5TuLS2+eEH\niW0REdI/fQKIYGRXVTrE6rvv0NTWFrySEjj88gtS/PwQcu1aeWU8W74cXebNQ4K3t6iOcRdJ86Hq\nifPwAL9CApnIhw9rjLsoX80GX72K1s7OUFBSgsulS9Bo2hRPK2j+AIC6oWG1163pGgEnTqC0TO2S\nBAK83bMH9lLKdUtL0zKV1oaSP1FSVUW3X39FgpcXlLW0YDthAji+FZ1CGODxBHj2LBYrVz7HoUNC\nhVNDQ/lJm1RGIBDg5s0w3LsXieTkT9DTS0O7dk3qvNdU34E/GUDFb3NzAEm1lDEre60KFQf+r0VR\nEQcWFrowNtZEfj4bampKcBrbF0c2LhI9rRgANCVoh/PYbLzesQMhV64g5MoVdJw+He3HjpXp+nwu\nF95//YXM4GC0GjYMdjNmSFXP3NERQw8exPtDh6Cmrw/no0cllot59kw06ANA6LVr4J87B9Nu3fBs\nxQqkBQTAydUVxONBRVcXCioq1WrXVIdBq1ZQMzAAOz8fytraiH72DG/37UPLQYPQrGNHLE1OFskY\nf0t0zM1h4eSEoLKBv/z/y0qp3VuCz+Xi5bp1SHrzBmY9e6I4KwsfTpwAILS/jz5/vsqXsNfy5TCy\ns0NmWFgVcxQJBOi7bh0UVVVF+Qy6L1ok0/0Y2NiIJiaAUEOqpmA7Xmkp3u7bh+DLlxFx9y7aT5gg\nMqX137QJGcHBiHn+HAbW1hh2sPLC/QsBx47BoZqALpVKAoUq2tqI8/CApYRJHK+0FIqqqjXe47dC\nUUUFc969AystDdzSUixc2B1XroQgNVV8z0beNv3qYDKZ6NbNFD///AA8njICA7fC0lI40dywYcNX\n6UNFFAFEA7AEoAzgIyRv7pYbn3sAqG4EaNAlUnXs2eND8fF5dPp0AAUFpdE//wiDtJ4sX04nevYk\nV0VFOtyhA2WEhUmsnx0VRa4AHevatU6h7Y8qhafLO9tU1NOnYu3vNDIigUBAfB6PHi5aRH7HjlHU\n06eiYKragqqqw//kSUoPCqIL338vZn4KuXFDnrdTbz4/eEA3pk0TyRnvaNJEqnD6itnGKkohl/8k\n1SJVzOdy6dx334nKu69ZI5f7eXvwIB3p2JHODhxImZ8/11o+LSiI1pd9XiUFffFrCAQrTE+nJ8uW\n0Q5DQ7o9axYl+vpWKVOSn08ne/cmV4C26evTi/Xr6XCHDhR05YqYCZHLZtPztWslXkcWqYlgNzf6\nx86OTvToQQk+PlLXqws5OcW0atWzChn+XMXk4xuapKR8WrPmOa1b90IsgA/fSKtnGIDPEG7ylttC\nfi77KedQ2flAAOKasV/4am8gEVFJCYcWLnxAOjrbyNJyL12+/EnsfLndvqL9Pj0kpEo7CT4+lODj\nQ+8OH65TIvUjdnZiA8jduXNlbqMmBAIBua9bRzsMDemgjQ3Fv34ter1iGXlxtHNnqe4n2d9f4utf\nAz6fT58uXaK3+/dLve/wf+ydd1xT5/fHPwl7iaAMFRkOREFQ3HugVuveWK3WVW3VOttaO8S9R3Hv\nUffEBW5FQEEZCsjeU/YmO+f3RyASCJCEgLS/7/v18hVz86ybXJ773POccz4XK9zQpP2Txd7MZbMp\n6fVr+hQcXGvZ+iLe05Ninz4ln927q90TqomQK1doPUB3Fi6stoxQKKTSvDxiFxXRP199RS4Avd67\nV3ydxXt60vFevWiLri7dmj2bSnIkEzIGnT0r0yIqKzKSNqioiH+D7QYGxCkpqbVeXbl0KZgWL75H\nJ04ENKj8acW5qOL/8YUidz3K/lWkst1hqRL6USqammr45Zf+uHYtDK1aNYGzc2eJz5lMpli+j5WX\nh6K0NHgsXYoRe/ZA29hYbA4ptwWXv8pLC0dHZLx///m9nHbe2mAwGBi6YQOGVnocZDAYKM7IwMPl\ny1GQmAi7GTPkNjVIo3nHjkgPDBS/rxzQVZyRgfyEBDxYsgSjXF1FaZHNlesJURtMJhOdZ8yQq47l\n4MGIfvBA/N58wAAkenmBAcDx++9hYm9fY/28uDh8+vABHSdOVGTISsNy4EAAEEejy4uKmhqWhIUh\nwdOz2jIMBkO839W8Y0d0GD8eatraYlOY5cCBaG5tjbzoaDjMni12xeUUFuL13r0I+ecfJL56hc7f\nfFPjOPPj4sTppwGAnZcHVnY21Ov5epoxozNmzOhce0EFEAqF4HIF0NSsGuBY0YxXk0nv30aD3TnL\nCQ3NoOzsEvL1Ta4ieiwUCune4sVERMRjs+neDz+QC0AnevWigpQUpY2BU1xMD5YsobODB5PXtm0S\n/dc3Z4cOlTBfhLu5KdROdnS0+P+lubl0Y8YMOty5M3msWkUJlVSpeGw23Vm4kFwAujBqFLELCign\nJuaLC8cTEeXGxZH3rl0UdO5clRWnUCikN/v30/Xp0+nN/v0kFAopOzqasiIjaxy7UCikgFOn6FTf\nvnSwY0e6t2hRjeaU2hDweBR88SIFnT0rFhbncTgUef8+xTx58kU8zGqi3JxU2awU6e5OnKIiSvT2\nljge+/Sp+Nqo7VxKsrJot6npZ0+kLl0adcpzWfD2TqSbN6taFmoC/3YFrpISDrS1a87YWF88eRKL\n06ffw8hIG4vGN4HXLz8h4/17tP3qK4w7cQLR7u7IDAkBn8vF2FqCbepK6rt3IKEQZr161V64Duwy\nNkZpVpb4qhmycSMGlQUGyUJhaioyQ0IQee8e+v/6K3RNTcUZN5N9fZHq54dod3f0WLIEbZycxELY\nr/fuBSs3F9ySEvRbvRpv9u6F1dChaNGtW60C3PVFQXIyjjs6ojQ7GwDQZe5cjD99WiltExH+btMG\nhSkpMGjbFl/t24fm1tZVxLtlaefSmDGIdncHA6KnxdkvXuDKuHFILFuB233zDZy2bkVTCwuljL2h\niXJ3h5aBAZJ8fGDn7Fyro0FeXBzeHj4MVU1N9F21CloyBvJF3ruHCDc3GLRpg75r1kjV4m1IiAiH\nDr3Fzp2vweMJ8OOPPbBu3QCZgk0VUeBqTNCqVQ8l7mTFxfLbIBXh/ft0UlPbKN6wsbc/TOdmfk+u\n7dvT+/PniYjEefjLX+VFljztQqGQ/A4coJO9e9OJXr3ozd9/1+sK7rqzs8RG5d3vv5drNZro7U27\nTE1pPUBeO3ZIrHwL09Npf5s2tB6gkGvXJOpxiospJiaHWIWF5L1rF60v22SNf/lSaecmL++OHpWw\n2W9UU1PaUwgrP59Crl6lyxMnkgtA16dNq7IfxCkqoldbttCTX3+l7Kgoqe3kxsZKbC6vB8hnzx6J\np7b1AJ0ePJjYBQUK7Tk1Bvg8Hn0KDSWP1auJz+Uq/W8g+tEjievebf58pbZfF6ys9pOx8U5isWTX\ndYACK/5GZShydX0LJ6dz+PgxE3y+EH/++bxB+vX1TSlLsiT6DoODM3DxgxEYP51H8zIbdbmbWmV3\nNVl5vWtXrWUYDAYcFy5ETlQUcqKi0O377+vVljfhzBkM/PNPqOvpgaGigqGbN0vkuxfweCgpWwFL\no3XfvjBxcECHcePQZtgwsQ2XiBBx+za4RUXQMjREyJUrEJbJA2ZllSAmsQQrVjxEVr4QrQY6of2o\nUTBxcIDloEFyn0NpTk6VZGryUB7yr9dSMg+Krqmp0vLxaOrrw27aNJj17o3hu3fDoH17iRUmEeHC\nqFF4/vvv8NmxA6f79kVRenrVdpo2BVRUJJZ2lcfNAJD08iWuTZ5c50RmX4rYhw9xbtAg+O3Zg0cr\nV0LAkZ6rShGEAgFerF8v8R3GP32qtPbrQlZWCW7fng4fn3lV3EaVTWPKVObSrNkoTJ3aCWZmTTBh\nwhW4u0cjOPgThgyxgo5O/ZmAhELC6dNBKJ8/GAwG8gW6uHBhEgzN6xZ0lBYQgHuLFiHkwgXkREej\nVa9e0NDTk1o2/PZtBBw/DsuhQ9FzyRKo6+pWm+JYGTBVVdGiWzcYtm2L3itWQMjnS6SQjXv6FClv\n3qBl9+5S6wu4XHScNAldv/sOTDU1sSYxg8FAsw4d8GrzZnCLi0E8HhgqKmjVowdSUgoxZco1+Pik\noKSEi69Ht0fPHxah06RJICK5hVaCzpwBCQRyB4elv3+P2EeP4LlhA7SNjESbnkwmsj5+hL6FBaZc\nvowmrVrJ1WZttOzeHRYDBsC8Xz+J82Tl5sJj6Wf/B15pKcz69IFRp04S9dW0tKDRpAkSX70CAPRb\nuxZ9li8Hr7QUya9fg6miAsuhQ9HC0REWgwah3YgRdRpvcnIB9PUb3s++mbU14p4+hY6pKQb9+adS\nA/98XV0RdOqUxMRvOWgQ7OTc7K8PdHTUYWqqC0NDbXGitw8fPsHERKfGRUiZH3/DO/MrCeJy+fTs\nWSwRES1ceJdat95L9+/X7pusDK5dC6Xhw8/T2LGX6PTpQLp+PZRSUhQz61TmwbJltFFdnUIrmTwq\n8vbwYYlH9tDr15XStyIIhULyO3xYlA6AyaRnf/0lk9BHRfKTkyn48mXaoqNDu1u2lPhs9uxbNGvW\nLbp8OZhCrl6le4sXk+fmzcRlsWRquyAlhSLd3cl92TLapKFB25s2pYCTJ+UyzbCLiuickxO5APRm\n/365zk3Z8Hk82mlkJP79NzCZlB4UVG15oVBYZfOZlZ9PnOJiseNBXRwQ+HwBlZRwac6c25SUlE+l\npQ2reczncik/OZkEfL7S03y4zZsnNomtB2i7oSEVZ2UptQ9lUFLCodevk2jatGt07lwQffhQvQgN\nvpAfv7KQOBkvrwTicPhyiX03VkKuXqXirCyK8vCotsyZwYMlbMw3nJ0bcIRVEQoEdNTRkY44OFTx\ns5ZGSVYWZUdFSeS1yYmJoU/BwZTo4yMRvJOZWUw8NpteuB6X+CO8Pn16rf0kenvTFh0dcgFoi64u\n7TU3p6Ndu8o06Vcu4/7TT+R74IB4H0dZKGKTTn33jk707EkHO3akgFOnlDoeeSkqYtOcObcJcKHW\nrfeSt3fiFx2PMgm+dEnCvv903bp67a+ma0EgEJCHRxS5uYVLtemfOhVIgAu1bfs3RUZ+vjkVF3No\n5sybZGGxj6ZMufbv9+qhOthq65vijIw6K+nUxJ358/G+ghdJv99+w7CtW+utv9pg5eeDBAKoammB\nU1BQo7dNwsuXiPbwAKegAKU5OZhy5UqtSmMRd+7ghrMz+BVyzWgZGODX3Nwa612ZMAGRd+4AEF3t\nNhMmYJSrKzSaNKnRLPbkl1/gd+AAtAwNMenCBVgNGQIBjwcVNTUI+Xylavm+O3q0XjVqGwIPj2js\n2+eLpk01ce3a1C89HKUScvky4p89g3Hnzui1bJlYG1vZZIWHoygtrdpYhG+/vYULF0IAAL17t8KL\nF3Mk/Pc9PRNw7dpHZGSU4MaNaeLjv/76BDt3vq7Qkgsg51z+n5BerG9S372D58aNmHb9er3lFhm+\naxdKs7KQHhgIi4EDMfCPP+qlH1mpmHBOvRaxC+3mzRFw/Dg4hYVgMBi4OnEiRv79NwysrKqtYzN+\nPJqYmSEnJgYMiK5aY1vbWsdVUaCdAUBdV7dWG3DMw4fizfXitDTcmDYNP2dliVXAlDXppwcFwf/Y\nMYTfuIHc6Gj0Wb0aTVq2RGpqIVq1qtkpQCgQyCzL2RDY2Rnj0aNZSEkpbFB5y4ag84wZcgfwyQMR\nIeDECYRevAhuSQk6z5yJ3suXS9xg0tKKxJM+APj6psLLK0lCPbB795awtGyKt29TEB6ehY4djQAA\n0dE1L45koVF59TRGPJYvx4mePRF9/z52mZigIDm5XvrRNjTEjLt3sSolBZMvXap1sm1MNDEzg834\n8WjVvTvU9fTQbtSoGif94sxMJL15A3VdXfT86ScYWlujw/jxmCSDjOLgjRuhV7bpqm9hgcGVEvuR\nsKoEXnGGpPQDKzcXghpESxSlRdeuYKqooDQnByb29tAyMkFYWBb27fMFlyuo0fso7OZN5MXHK31M\n1RFx965UKctyWrcWKXyVv5ZTmCo1vyIA0e/6P0TODV2++w45MTHICguD44IF4kk/O7sEAKClpQoV\nFcmbqZ6epANLWFgWOnc+gmnTbqJr12N48iQWAKCmxkRdzfr/m/hrgIjwvkyEgyAKKY+uRuykMZup\n6hs1HR1MOHsWQ7dswcqkJDSrQfAbELk3Bp04gYz375EfFYXv/fzg7OYmDtZh1eCGaGRjg6k3bmBZ\nTAyWRkaKg6CyIiJw0MYGm9TUcHH0aAlFsvajRqFJhUAgh9mzZdL8VQTjzp3x48ePUNXSwuPHsRg4\n8Az27HmDVasegsOpqkQm4PHwwsUFL/74AzecnRF45ky9Xks8Fgu+f/+NV5s2wWP5ckRVSENRE0Kh\nEAIeD17btyMnOhrCSqpqQqEQD5cvV3hcQqEQEd7+EsdiHz9WuL2aaIgbLDs/HyP37cP027dRXEHJ\na+XKRwAAAwMtHD78tXjyX726N3r1kgxWc3V9i6IiLgCRFOvmza+wcOFdPHoUW+bl+K/f1wVQjykb\nMjOLFaonFAppu6GhxKZrwMmTUsv6urrWZYj/WrgslkKJsXx276Ynv/5Kj3/+WeK4gM+nC19/LTWQ\njFNSQok+PnR18mR6f/68RLKzypvjnlu2SNQtSk+nt4cPU8iVKwplUa1IiRxeIMOHn6d+/U5RQEBq\ntWVKc3Nps6YmHejQoUFSV6QGBJALQMccHWUO8uKUlIhTbew0MqLoCo4KeYmJdLBjR1oP0GF7e8oI\nDZVrPFmRkRR85QqtsBpC6R8+UFpQEAWePk37LS0p6MwZyktIkLmtmr4/TkkJlWRl0T8jR1LRp0/E\nLiyUa5yyIBBI7z80NIMcHY8S4EKDBp2hxMR8IiIqLeVSUZH0QNVFi+5VEH9xoQkTLpObWzhpaGwi\nU9NdFT77lwdwKZOPHzNx5Mg7vHqVgO3bvZGeXiR3GwwGA18fPCi2K1sOHgz7mTMlyhSmpuLBkiV4\n/scfuD1vHj59+KCU8f9biH/6FDEPH9ZesBLdFy3CsO3bMeivv8THEry8sM/CAjEeHjhoY4PQ69cl\n6qhpaSErLAzhN2/i5fr1yImOBqdMerK0UqBZRQ0CQBSQ1eOHH2A3fXqtm3n5SUnIjowUr7zzEhIk\nPvfZvbvKMWnw+UKcPTsBXl5zYWwsXdcVAApTUjDv9Ws4bdsmPp/6hFNYiGk3b6L9mDGAjLZ7dW1t\nOM6fD/MBA9CqRw+0GzkSgEiTIviff8DOz4emvj6amJvLtFdTkeTYT7g0Zwmaxr/A9yO34ZPQCIle\nXshPSEBOdLRcSfwyQkKQFREhareSdjC3pASXx41D7MOHuDhqlPgaibx7Fw+WLMG7w4flftrKySnF\nokX3MXLkBZw//x4HDvhJLWdra4xRo9qhe/cWmDq1E8zNRY4IWlpq0NWVHqP011+DYGPTDABgYaGP\nHTuGQ1NTFSkpq7BkSQ8YGDROLQN5qdOdtrCQTTdufKSnT2PpzZtk0tDYVHY3/IuYzA3UosVuevBA\nsZiAkqwsyoqIqDYBVOCpU+TCYNCjNWvqcgr/KgQCAb3es4e2NWlCW/X0yHvnzjqvVt+fO0frGQxa\nD9A2fX3iFBVVKRP3/DldmTRJnI53n7k55cbF0dtDh8Sr/S06OpQWGKjQGF5t2yZ297s0fjylBQXR\nre++o9KcHCrJzRWteplM2mlkpHBSuy+Noim58xITSSgUUkFyssRTU35yMrkwGHTI1pY+ybnaL2eb\n/QBa0Lw3rXYWxVS82LCBgi9eJN+//5apvkAgoKCzZ+nSuHF01smJ7i9bRke6dKGoBw+oJCuLhEIh\nXR4/Xhwns7tlS+JzuRR2+7bEk+KT336Ta9xOTufEcozlkozz59+m0NBPVcqGhIiOhYZmyNw+ny+g\n9PQi4vGqPqUWFXEoMjLr/687Z3ExF336nERoqOgObmtrJBY/BgiamqpwcrLC/fszq2+kDny8fh2t\nevRA/IsX6Dp3br30IQvhbm7oOGFCg/Un4PFw2NYWRIQfQ0PrlOiqJCsLe1q0gEAgEF/FP374ANNK\n6Y65JSW4NHq0OC0wA0D3JUsw+uBBxL98iZzISFgOHozmtewzSINdUIAdBgZAhetQTVsbvNJStOzZ\nE7OfP0debCxuf/stdE1N8e2jR4qebrXEx+fByspA6e3KC7e4GKnv3kGvZctav8v0wEBwS0qQHREB\n+5kzoSanYwKfy0VkeAY62ZshxD8O9j3agoRCMJhM8assFH36hIPW1mCoqMDYzg7J3t4YsnkzBv7+\nO3Kio3HQ2lqi/LzXr/Hh3DkEVFCvM+3aFYsqpBWXGCdfCFVVybFoaW0Bmy2537F2bT9s2zZMpjEr\nSkEBWxxVrUiStv+EO6eHR7R40gdEZp6K38PgwVbYvXtEjVqhdcF2qsjP+UtN+il+foh//hzvz5xB\nYUoK7JydoVOmmfopOBjPfvsNAg4H/detQ5uhQ5XWLys3F995egIMBth5edA1Na21TklWFnSMjKoc\nZxcUgAQCMCCyPxIg1m6tiLqOjsi9UErbVoMHw0qKxF95nqBaf3siiUmfAHBLS4YS9XwAACAASURB\nVMEAkPr2Ld66usJ22jQsfPcOBYmJEPD5UFGSKyiPJ0BBAQebN7/C9wsc0L6DMQwNv4xnV2l2Nk73\n64ecqCgwmEyMPXECXefNq7Z8C0eRtpLFgAEAgJyoKDSrNMnWhKq6OmwdRC659j1Em/Xlk708PvZ8\nFguDN26EkM9HVlgYbMaPh1YzkalEXVdXfCMpR6NJkyppMSrqR/DYbKhVcN9eu/YJ/vprEJo0+XzM\n2toQwcHl3kyia8fCov7SrACAQCDE2rVPceTIGIXb+E/Y+Cu7QamrMzFiRBswmQzY2RnjwIGRsLEx\nQlZWKW7eDENAQO0aq/8mWjg6IuXNG+RGR0NDT0886XNLS3FhxAjEuLsj/tkzXBk7VqnuqLomJtBr\n0QJ6pqbiST8/MbHa8qzcXHhu3Cj1M4M2bdB25EjxhN6yW7dq01IP2bQJ6mX5jvTNzdG3klB4ZVJ9\nfZHu719jGUCUBK3/unXi9xoVYhkAID0gAIZt20JVXR3N2rdX2qRfzroll3D69HuM6OuKo98sh7CC\nyEhDEnT2LHKiogCI3GNfyJiquzgjA1nh4bi/eDFyYmLArqQzLA/Jr1/XXqgSTS0t0WfFCvRbswZj\njhxB3zVrxHtyei1aYKSrqyhmg8HA4I0bYWxri55Ll6LvL7/AuHNn2Do74+tDhwAABSkpCDp1CgCQ\nmVmEiROv4ODBd3B0PI5Dh96K+3zw4BsYGWmhfNJfuNARNjZVFzbK4t27VDg6HsfJk0EYN+4yMjIU\nS+b2n1jxf/VVOyxY0BUnTwZBU1MVZ86Mg7NzZ3HgyZMnscjNZWH8+Kv49KkYDAZw+PBoLF4sPfnY\nvw0VNTUYde4MhzlzIOByxceL0tJQUsGHnVdaipyoKKUmvSqHU1wMbmEhHq9ahXGnToGhooKPV69C\nwOWi8zffIM3fH3fmzUNRWhrYeXn4+tAhiUhbJpOJGXfvIuzmTQi4XHSaPLnaYDnLgQOxPD4eBYmJ\naGZtDXVd6RunQqEQfq6uCLlwAQwmE51nzRJFatawoem0ZQvsZ84Et7gY0Q8fwnP9egCi58fW/fop\n/gXVgpqaClSfHkZ7dAADBN6jq/h43QmdnZ3rrc/qqOzqWjForiYEXC7uzJ2LVD8/eG/Zgq8PH5a7\nb25JCRJevICfqytsJkxAi+7dYdazp0x1K/6u5WbHimPvuWQJHBcsAAmFUNMSJUFjMJkYvmMHhu/Y\nIS4XevUqPDduBCsnB/nx8Ri6ZQv69WsNN7dIpKYWYejQzzEqZmb6SExciZcvE/DhwyckJORj0yZP\nREdnY86cLlBXlz7FpqcXITQ0E7a2RmjZUvaMvz16tEL//q3B4fAxe7YDTEyqdxqoif+Ejb+coiIO\nNDRUoa4uioDk8QQ4cuQdrl0LQ1ZWCaKicspKMmBp2RTx8Yr7HTc2SEp0JZ/DwWFbW+TFigI/NA0M\nsDQiAjrGxkrv/1NwMG5Mm4acyEjYzZiBotRUcRZJo06dsODtW7jNno3cuDh8feAAzPv3V/oYpMFj\ns7GnRQswmEysTkuTax+ChEK83r0bqW/fonX//uj100/1liabiLBWsxm0uXkohha0wcaYw4fQ44cf\n6qW/muAUF+PCiBFIefMGqlpamHrtGqzHyGZWuP/jjxDy+WjVvTu6ff+9Qv2HXLqEWzNnwsTBAbMe\nP4ZuPVyvtXGyTx9kffyIH8PCoG9mhlOnAvDqVRIiIrLh5TVX6oReUMBG9+7HERubh+DgxbCzk57i\nxdc3GSNGXEBRERdNmmjgyZNZ6NmzZsGZivj7p8LRsSWCgzPQpYupQjb+/9TEL43U1AKYm++HhgYD\nLBYf5afcpo0hYmMbZuIvzwlTE/UVsp+flATvbdvAZ7PRe+XKKpul1cHncqEq40qvHLfvvgO7oABW\nQ4ZUCeb59ulT6LVsCcP27ZEXG6vQ5qsi5CcmIicyEiQUwtjOTiKQS9b6DaVm9eTXX/F6504Aok3l\nNsOGoe8vv8C8Hp80qkMoECA/IQFahobQMpB9s7k0Nxfahobi19rIT0qCdvPmEpHq0R4eSA8MRF5s\nrNJU0OSBx2Yj2ccHzTp0ACsnB6YODmCxeNDSUgMRgcsV4OLFYLi7x6Bjx+b444+B0NBQRUkJF5s2\necLKygBduphWCcgqZ9Kkq7h9O0L8fvLkjrhxY5rCqTH+N/FL4f37dOza9RpXrnwEEYEIYDIJDx9+\nK5EXoz7x3LSpWklDHouF61OnItrdHYbt2mHG3btobmPTIOOqDhIK8WrrVgySM19QYVoamrRsiczw\ncJxwdPycgI3BwA8hIXL7d39J2AUFKE5Px6vNmzHy77+hrqvbIPJ80R4euDt/PorLhFhUtbXxY2ho\njSkw/o0I+HzkJyTg/blz0Dc3RxsnJxhYWYHBYHxOnicQgMFg1FsSNUW5ejUUzs43UD59Ll7cDUeO\njJGYuGuaxKdPv45r18LE72fMsEWvXma4cuUjunc3xYEDo+UajyITf+P6RuuBLl1a4OjRMejc2RgW\nFvrYtGkwFi7s1iCTfnZkJK5NmYJXmzfj+vTpyI2Lq1LG7++/EfXgAYgIudHRePAFHu0rkujtjfPD\nhuH1rl1wmzdPrvD2JmVqUMYdO2LSxYvQMTGBZtOmGOXq+q+a9AEgNzoaF0eNQsjFi3jx558NlpKj\nZffu4kkfAPilpfgUFNQgfTckTBUVJL9+jdc7duDZL78gLzZWPFGKk+epqDS6SR8AvL2TUHGe9fJK\nAiC5x1DTyn3jxiEwMxM5J5ib60NHRw0rVjyEr28KDh70h5PTOeTklFZbXxn8JzZ3a6O0lAdv77ll\nuVIYaNpUs0EyDjbv0AEtunVDtLs7zAcMgGGbNhKf89lspL57J76ECF8+0ZVF//7Qt7RE8ps3MLaz\nQ+KrVwqtNjtOmoSOkybVwwgbhpbdu6Nl9+4wsbdHh3HjJNz66hMtQ0M0tbJCftkNV0VDAyYymuca\ngsi7d2ExaFCdleEYDAbaODnByskJAi5XIdlNRSjNyQErLw/N2rVTuI3u3SXlLnv0aFlNSel06NAc\n0dE/ITW1EGZmTeDkdF7icw5HgGbN6teV9//FxF++862rW/+P6pUx6tQJaz59EgccVURVUxPm/fsj\n4tYt8bHuteRxJ6EQYTdvgpWXh44TJtTLRq316NFo3qkTXrq4gEGEvNhYDFq/Xq49iA8XLiDt3TuY\nDxgA2ylTUJyRAYaKitjV9N/AqIMHoWtiIpFkq75hqqjg2ydP8Pz338EtKkKvlSthWIdJSlnwORz4\nHTiAsKtXYXj5MmwmT4btlCl1alNNWxvfPHgAIZ8PPpcrs/eQoiT5+CD20SOw8vLQacoUWAwYoNAT\nxezZDsjLY4lt/Fu3Ss+3XxOamqpo21a0B+LgYAIfnyTxZ5aWTaurpjT+8zb+L4mAz0dcmZBzm2HD\npPp9J3h64uO1a0jz94fT9u1oM2RIjW3emTtXnDFU39wc3wcEgKGqKpE/XxkQEY7Y24NXUoLFHz5U\nqxMsjdd79+Lx6tXii8tq2DDx9zBs+3b0//VXpY71f4gC4/wOHAAJhei1dKlMwXTykhMdjQPW1jC2\ns8P3/v4NsuehTNKDgnB+2DDwSksx+8mTBvMsq43SUh5Wr36EoKBPGDrUCps2DYGKiuw3pP9t7jYi\nhAIBrkyYgOj79wEA1mPHwtnNrcoKg8/hQFVDAyQUQigU1hgUxGOzsUVLS+JHm/DPP8iJjka3hQvF\naY0rQkIhIu/dg8348aL3Mpq4SnNzwSkogEaTJuCVlIgTZZFQiBQ/P6hqaqJF164SY1PT1ETymzdw\nmz0bOTEx4s8kemMwsDotTaGJKT8hARr6+nJ5mfzXKEhOrhKHwWezcczREdnh4QBEwXCLP3yoNr5B\nUZJ9fcHKyUF2eDh6Ll1ab6JEyiD++XN4b9sGFQ0NOG3bBpPOnRH/8iV89++HdrNmGPTXXw3mrVXf\n/L9N2dAYyQgJEU/6ABB17x4yQ0Or2GvLV00MJhMqtTx2qqirQ7NpU3Aq5KsPLZORe3/qFEYdPCiR\nqyf++XNElUXtFiYno1XfvmAymWjRpUut49c2NPzsjlcW9i4UCHBl/HhEl+Vw77ViBUbu24eitDSE\n3bqFXkuXoomZGUqyssBANZmjiGoUAAGAzI8fJTaDhQIBUvz8EHbjBjT19dH2q69g1qvXf0oVqjaE\nAgH4HA68t29HjyVLYNCmjXjfISc6WjzpA0BeXBwyQ0Nh1ru3UsfQuqw969HyeZ00NPmJibg0Zgz4\nZddZmr8/nLZtQ/itW9AxNob1mDHQ/gKxAY2Jxrdl/h9BTVtbYuIjiARL6gKTycTUa9egbWwMpro6\n+v78M0bs3g29Vq3Q3MamSoI28/79kerrKxI8SU6G95YtePHnnwg6e1ahdABJ3t7iSR8A/Pbvh/+x\nY/hnxAh4bd6MR6tXQ6tZM7QbNQrGdnZoamGBLvPmwWr4cHEdhzlz0NTSUmr7rPx8ZH78CPelS5Hq\n74/ClBTReauoQMBmI+DoUfj9/bfIxU+OSZ9XWgqPFStwfvhwvNm3T+7zVgRlP70KuFw8+fln+B8+\njHNDhiDJy0v8ma6pKVQr+MEz1dWhJ2e8wn+FtIAAZEdGiid9ACjJyEBTCwvEPXmC2IcP0bpvX6iX\nRe4qQmEhB0uXPsDYsZdw8WKwMobd4PxvxV9PNLe2xpCNG/GyLOS/1/LlYrWoutB2+HCs+fQJRAQm\nk4ms8HD8EBKC3JiYKoFiDFVVmA8ciN4rV0LA5eL9qVPgczgYe/y43MFiAh4PjEp1CCLvnQ/nzqEg\nKQl9Vq6EipoaJl28CCaTKQ4CEwoESPT0hIq6eo12VTVtbbw7cgSJL1/CY+lSTK+w6W3Yvj1sJk6E\ngMNB8wqJtGTBY/lyBJ08CUCkH6DZtGm9J9TLCA6Gqqam0gLV1LS0YD9rFrLCwqCmrY22FW6mOkZG\ncHZzw5Off4ZQIIDT1q1SzX6NhfoIVmTl5yMrNBQv/voLDnPmQENfH5yyXEGG7dtDw9AQQzZtQml2\nNpjVBFNmhYfD/+hRqOnooO+aNdUGoM2efRt37kQCAO7fj4axsQ6GDrWq0S4fF5eHNm0aj4myMT0r\n/6ds/OUUJCeDXVQEj6VLMemff6Cuq6uQK5wi7qcV6xSkpiL4wgWoaWnBesyYKq6lAJD+4QOKU1Nh\nYm8vEeFamp2NiDt30HXePLh99x2Cz58HGAwM2bQJfdesQaKXF5rb2ICdmythyhLy+SjJyoJeixYy\njznozBl8CgoS3aAqpMvllpaK86vw2WwwmEywCwpkCuc/2qULMioI5HRbvBhjjhyRKMNjsxF06hQ4\nhYVw+PZbuSN8yyEiBJ05g4hbt8BUVYXNxImw//ZbpaR6KEhJgV7LlihOT4euqWmjEmeXRkZICEw6\nd65y/M2+feizcqVS+yIi+OzciWdr16JFt24YtnMnQi5cgIqGBgasWwe9Vq3Ev4G0LL3Fnz7hsJ0d\nWDmitC4tunXDgrdvpf5uxsa7kJVV7mdPWLduADp1MsKIEW1hZCT5VJ+bW4qCAg6WLHHHyZNjoa+v\nWSabqDwUsfE3JmQWJ2gMlObm0v0ff6QrEydS2O3bUssk+/nR7pYtxcIe/4wYIZdsXzmc4mIKuXKl\nTuOtSXxDwOdT4OnTtN3AgNYD9BeDQf5lEpOh167Rqb59aZ+FBT1Zu5Z4LBblxMRQYWr1UoJCoZA4\nJSUU5e5OPrt2EbekRGbBD1Z+vsSrNDhFRfTxxg3y/ftv4hQVkaAWKcUHS5dKiG0EX7xYpcz5ESPE\nn+9u0YKKMzNlGq80ijIyaKueHm03NKSS7GyF25EVoVBIyb6+MpX9cOECHbK1pRM9e1KKn1+9jKcg\nOZlS372jE716UVpAgPi7TAsKopuzZtG2Jk3o3uLFNV5DivDx+nV6+ttvdPu77+SuG+7mJnGNuABU\nmJ4utey4cZcqyB7+Rf37n6T27V1p6NBzdOlSsETZqKhscnA4QoALzZp1k4qKZJO6lAcoIMTyP1OP\nAhRnZIhWB2Vyf5F37mCutzda9+kjUe72nDkoTksT34rVdHSgXY0fOzs/H75//w0+m43uixaJ7eCR\nd+8i9MoVZIaGIi8uDr2WL5fIayIrNUUVMlVUUPzpkzj/PYMIj9esgeO8ebCdOhW+e/eiOD0d3Rcv\nhqqmZq0mKwaDgQg3N9yZPx/E5ULA5aL/2rUyyfyVPw3V9FQUfusW7i9eDAGXCyGfj14rVtTY5ld7\n9kDTwADZYWFoN2oUOn/zDQBRugw1LS2wcnMRV0HYuyg9HfEvXsBu2rRaxyuN9IAAOG3bBgGP9zlt\nRT2R6u+PNH9/RN+/jx7LlsGif3+oV7OXlPnxI9zmzAGV7e9cGDUKP8XGKt0VWE1bG89++w2pfn54\nd+QIRh04AABo0aULdExMwCkqglnv3tBrKV/gU220+/prdJoyBdxS+aNeDdu3B0NFRfzd6JiYQKsa\nU8/58xOxbt1TvHyZiNjYPLx7lwYuVwgdHTVMny4Zod6+fTP06NESdnbGGDrUCrq6GuBy+dVm7YyI\nyMZvvz1FaSkPv/zSD05OVZ/MlcH/Jn4FuP/99+JJHyhzcfT1rTLxl5ZF4RJEz2FGdnZS2xPw+Tg/\nbBjSAwIAAB/OncPi4GDoNG+O9qNH49WWLcgMCYH1xYsKTfqyIBQKJZYNjDKzG5/DQe/Vq2Fqb4+S\nrCyZXOAyQkLEZib9tm1hPW6cUkPvbZ2d8bYsb7r97Nm1mlHYBQUYKkUH4MX69RixcyfU9fSgUclb\nKtnbG03MzGDet6/M42Ln5yM/IQE+O3bAads26LVsiSatWslcXxF0TU3x1tUV2eHhsJsxo9pJHwBy\nY2LEExsg0kcIu3ED3RYsUOqYtAwNYWxvD1NHR+i1bCkR9Wzq4IAVSUlIe/euzv1U3tMq/9tQ5G/E\nuFMnTLp0CT7bt0NdVxdf7d9fbZJCfX1NHDo0BjyeAPb2R1BSwsWVK1MQE5MLLlcITU3J63HLlqEw\nNtZFZqYod/7OnT74/feBEguwjIxiGBhoYcSIf5CcXAgAePUqEWFhS+pFke1/Xj0KkBMdLfGeALTs\n0aNKue5leXcYALSaNZPYUKQK+xlFaWniSR8AitPTkVYmHEJCITpOmYJvnz0DS4oiFREh6OxZPPvj\nD6T4SRd6loWey5ahVY8eoqcTBgMj9uwBg8GAqoYGbKdMQTNra7TqLpt+gUGbNvhw9iwELBa6zJlT\n7VOOovBZLMx59gzzvLwkFLOkwS0uxvNKyeayIyNxcdQo+O3bh0tl4jTObm5o1qGDWHwl9tEjCZFv\nEgoRcvky/A4cQGFqqtS+1HV18f7cOSR6euLRqlX1HokKAPpmZmjh6Ij+69bVGsXduk8f6Bgbo9w2\nwFRVxbO1a+G1dauEMpUiZIaF4UTPntjXujWe//knhmzYgOE7dsBx/nyJcg7ffgt9MzN0nDixTv2V\nZmfj/blzdWqjMnbTpmFRYCDmvnqFlmWqYjWRl8fCixdzEBCwCFZWBpg9uws0NauupY2NRfEURMB3\n37lhx47XGDv2Mvz9P19HP/30EJmZJeJJHwDYbAHCwrKqtPdfQ6r9KjIyi7p0OUK6ultp5sybxOVK\nFzxvSJ6sXSthC3z255/Vlo188IDeHT1KeYmJEsf9DhwQ/59TUkLbDQzE7W1UVaXsSNmE4R+tWSOu\nt0FVlZLevFHspIiIx+FQqr8/5SclKdxGOR4rVpDfgQMUULZXUJGYR4/kbk8oFBK3tFSuOvGenrTf\nyoo2MJl0dfJkygwLE3/mtW0bbTcwIL+DB6vUef7XX3Tr228lBMXd5s6V2AMoTEuT2qf/yZPkvmIF\n3Vm0SK6xKgqfxyM+j0dERFwWq9byOTEx9PiXX+jFhg2038qK9piZEY8tu93Z//hxurNwIQWdPStx\n/Ii9vcTfxMebN+U7ETkIuXqVTvbtS3vNzenRr7/KfV18Sa5fDyUNjU3000/uREQUGZlNjo5HCXCh\n3r1PkoXFXvH+gZ7eVkpOrn6vqxwoYONvTEg9qUGDzlTYSHGhvXtfK/WHUAShUEj+J07QozVrKP7F\nC7nqZoSE0J0FC2irnh7d++EH8Q0h0dubjnfvTodsbeXayN1nYSHxB/f0t9/kGo80SnJyKMrDg4Iv\nXlR4k5PP5Uq8EhHlJSRQ4OnTtN/SkgJPnaKs8HCZ28sIDaX358/LPY4b33xDRx0dKfDMGbr/448k\n4PNJIBBQzOPHxGOzKfbpU4nyPA6HiES/sYAvWmQIeDzawGRKfM+Bp09L7Y9VUCDx2lgpysig7Kgo\nyktMrLIoqY7Xe/ZIfAf+x46JP9uqpyfxmc/u3fU1dCIiOt6zJ21SV6e8hASZ63A4PLp9O4zu3Ysk\nPr9mh4D64tGjGMrJKaHr10PFx9avf04DB56mHTu8KCkpnxYuvEOzZt2id+9SZGoT/8XN3fR0SU3J\ntLSiLzSSzzAYDIXtosZ2dtBq1gzc4mK06NIFTcvMCeb9+mGhDHbP3NhYvPjzT/BZLPRetQpNraxQ\nUEHntmkd87bzORzcnj0bMWWBWnqtWmHhu3dyuWQCn1PrVrTB6pubIy82FvkJCUh+/Rpdvvuu1nao\nzD3SZ9s2cEtKkBMZicEuLiLtVBnqDvzjD+QnJuL2rFlg5eSAz2JhpKur2A++jZNkgq1yuy6DwRDH\nLTBVVaFtZCQhY1n5+yjNzoZ28+bQbCKS0St/bazoGhtX6wpblJ6OG87OyAgORtsRIzDx3Dmoamoi\n5uFDiXKxjx+LVbY6TZ2K92WiKWo6Omj/9dc19i/g8RB48iSKMzJgN326hMh5bQi4XPT7+WcY2dnJ\nvO/E5wsxcuRFvHiRAAAYO9Yabm7T601RrTpGjBA5RkyZ8nkTeOZMe7i4DEFUVDYiI3Nw4MDX0NCo\n36m5Udn48/JYEAgkbY1z5nz2C9fSUq2ya15XuFwBcnPrN/d1ZZrb2GB5XBw05cw5I+DxcH7YMIRe\nvowINzdc+OorDN28GZaDB0Pf3Bw9f/oJXefNU3hcRIS3Bw8i+sEDsR24KDUVH69fV7jNijAYDOgY\nG2PSpUtoamUl04Yvg8GA47x5UNXUBI/FQr9ffpFp0i+va9SxI9qPHAmzPn1gPmAAusybBw0FcthM\nu3ED+hYWUNfVRb9ff0W7kSMlPvf46Se522yseCxbhqRXr8DJz0fYtWvwKVMFM67kk29UIa3GmGPH\nMPrYMQxavx4LfH1rncjvzJ0L9x9/xKsNG3Cqd2/kSdGqqA4VdXV0mjIFRjY2Mu87BQamiyd9ALh3\nLwqRkTnVV2hA2rdvBjabDy+vJGza5Il1657h0aOY2ivWgUa14j95MhCrV0t6xvz22wBYWzdDUlIB\nRoxoW62OpaI8fRqH9PQizJ9f+2aOsuhattKtLnVBdZRmZaEgIUH8ns9ioTQnB3NevFDKuBgMBrrO\nn48na9eC+HyJfpRFzzKxc6plU7YoLQ15cXEw798fnKIijD1xAjqmpij+9Akacq6mhUIhvj54EPrm\n5ihISqq9ghTM+/fHigrffTl5CQm4PWsWknx8kB0RgYkXLsC4UyeF+mgsFCYnS7wvKHs/dPNm8EpL\nkfb2Lcz69cPA338Xl1FRVUV3OTR2wyosJjiFhYj28EDPJUvEx0jJehlNmkhmEmUyGdDTq//Nd1nR\n1FRFp05GePUqCXl5bKxfP7he+2tUK/5vvuks8egVFpaJoiIuIiJysGpVX6VO+kSEgwf9MGnSVSxa\ndB8bN76s8rQhD1yuAHPn3oGp6W4MG3Ye6emSJqnaJjpZ0DE2RtMKEbeq2towdXCoc7sVEXC56DBu\nnITRMF5JNxbgcwxBTX/UoVeu4PHq1fDcsAH+x45BVUsLZr17w8DSEs2sreXuk8lkoqmFBRgMhtIz\nMhpYWsJ22jS07tMH7UeP/tdP+gBgN3Om+P8MFRXYTp8OQJQ2YsyRI/g+IABfu7oq5LVUUsIFAOhX\n+h0qLoKICO/PS4qT1BUbm+bYuHEwAIKKCgN7946AmVndxGSUDYfDx/HjYzBsmBU0NRt3VLYyoYyM\nIiIiSkzMJ0/PBOrd+wS1b+9K+vrbaMGCO1RYqNyoNx5PQF27HiUHhyOUn1+7R0RNbN36SmITeuLE\nK1RaKtrY5LFYFHHvnjKGTLlxcXTz22/pyqRJlOjtrZQ2K3Nv0SKJjbqTvXvXSz/VwS4ooL2tW9Mm\ndXWleBjVN+XeQhW9hmRBKBDQvcWLaWfz5nS8Rw/Kjoqqj+EpROT9++S1bZvMEcGysn27F6WlFVL6\n+/d0rFs32mtmRi9cXMSR3fGenuS+bBntt7Kilxs3UklOjlL7Ly3lEovFk7uerJHndaFiBHpt0egV\ngQKbu40pv0PZOQBRUTkYP/4KIiKy0b69AfLyOHj8eBa6dpVvg7E28vJYICJoaqqhsJANU1PZxUYq\ns3jxfRw79tkXv1u3FhgyxBILB/MQdf8+PgUFofPMmXBcsECcc6axkhYQgLODBoFXUgIGk4mJFy6g\n84wZDdY/Ky8PAceOQcfERCR/KCXfy3+BwJMncW/hQvF7s759Md/HR6a64W5uVbKxSoNbWgoVdfUa\ndR4aAhaLhzVrHuPChRAYGWlj585hmDRJ8umIz2aDqa6Oa5MnI9LNDdPv3IHNuHFfaMQiBAIhsrNL\n8fx5HKZNs5NLIKWh+M/k47e2bgYHBxP07t0S7do1w5IlPREbmwsiwpMncSgs5GDkyLZ1llI0MNDC\n9esfcfDgOxgYaGLv3q8UzqA3eXJHnDwZCIFAdAMuKuJg7943+PjBDF+leCMvPBSjDh5s9JM+ALTs\n1g2LP3xA8uvXMLazkxBcaQi0DAxEKR7+41Teb5Bl/yE7MhLRHh54vXMn8hMS0GHcOImEe3wuF3e+\n+w7ht27BsG1bOH7/vShVSLNmMOvTB3r1oMwlC1paali5sjdu3YqAhUXTxcGKsQAAIABJREFUKpM+\nALzcsAHDtm2DWe/e6LNqFUorRMcrApXtE0RH5+D583h06NAMgwfL7vXG5Qrg7h6FEyeCEBeXh9ev\nUzBpUkcMGVI3z7nGQKOc+AHg2LEx0NfXRH4+C02baqJbt5ZYuPAOTp58DwCwtzeGj8+8Ok3+gYHp\ncHa+CaFQ9KQREZGNiIilCrU1fHhbeHp+h6dP42Bra4SEhALs3OkDp+HtYVswHiZ2v4NbXFx7Q40E\nw7ZtlZJGWpmUZGVBx8gIRenpcruXNkZsJk6Ez86dEHA4AIDOFWzrgCjVhNe2bSjNzobjggVo3bs3\nmllbw//oURSnp4OVkwODSu67AceOIfTyZRBE0bSPV60SR+VqNW+OBW/efDENXxaLj/DwJYiMzAaf\nL4Sqqmj1nJ+UhEcrViDy3j2kBwZixJ49MKkmvYk8XLgQDFtbIwwceBYlJTwABFfXUVi2rJdM9a9f\n/4hFi+6X1QVatdLDwIH/DdWuRjvx6+uL8ns0bSpaIRcUsMWTPgAEB2fi6dM4TJjw2W1MIBDK9SgW\nEpIhnvQBIDIyB2w2X2rYtSz062eOfv1EfvnPn8chLW01Xr1KwNChmxVq739IEnDqFCwHDcKbXbsw\n9do1MJhMpeYAamhadO2KhW/fIur+fRi0aQM7Z2eJz6+MH49ET08AQOilS1gUFITmHTpAr1UrzLh/\nH7mxsVU2yctjDcqPCiukYmBlZyPgxAkM37Gj/k6qBjp3Fjln9OolmfK6qbk5rMeMQcLLlzDr3bvO\nk35uLguHD7/DiRMBMDTUKttQZgBg4NixAJkn/hkz7HDgwFvk5JRi48YhiInJBZPZmKzjitNoJ/7K\naGioQl2dCS7384Vc2UXrxo0w9OnTGubmVXfri4s5uH49DCoqDEybZgtNTTX06dMaWlqqYLFErot9\n+7ZWeNKvzNChbSReayI3NhZNLS0bfX71LwWfy8Xz33/H6927wWQyQUIhjnXtiinXrskV+KNo39Ul\n6yqHlZtbbSbH2jCxt68ix1neb4Knp3gC57NYSPbxQfMOHdB39epqXWJtp0+H7/794JWUAAC0jY3F\nyQKBuqvA1Rf6FhZYnZ6OhJcv69yWoaEWOnUyQlJSYRWXzcr58muitJQPD4+Z0NFRw717Ufjtt8Yh\nzq4MGtPti6RdyBW5dCkY8+bdAYcjwI8/9sChQyLtTx5PgK1bvXDhQjCaN9fGDz/0wOzZn90c2Wwe\nBgw4A3//dADA4MHmePJkNlRVVfDmTTJOnAiEoaEWnJ1t0b17/WZTrIiAx0NRair8XF1h1rcvWvXs\nKY7k/R+SFCQn49Lo0WAXFsLIxgYm9vYYXhZYVJ98+OcftO7XT6pwTTluc+Zg/NmzNbqo5kRHQ0VN\nrYrbYk11DtrYICdSpPTEYDIx7/VrmPWqfbWaExWF2MePYdCmDZjq6rgybhz4LBaadeiABX5+CgkB\n/du4ceMjTEx08fJlAt6+TcODB1Fo394Qbm7O6NjRSOZ2Sku5cHOLwMmTQRg61Ardupli1Cj5XYrr\nk4YWYjEE8ARAFIDHAKpL6p0AIBhAEIC3NbRXq9vSxYvBxGSuJ2A99e17kkpKOOLPsrKKSV19I3Xs\neLBKPR+fRAlXS8CFwsM/C6IUFrIpMTGfhg49Q7GxOZSdXSJR/+XLeJldq2SBzf7sTuZ34AC5MBi0\ny9iY4l++VGo/0hAKhRTXAP0om7zEROKx2ZTs50ec4mLKi4+v1/54HA49/f132m9pSUe7dqX3585V\nKVOQkkLnnJzIhcGgIw4OlP7+vdS27syf/zmh3++/i48HnT8vzgckjZyYGLo0diyd6ttXnKcoJzZW\nwrXwU3BwddWJiOifCgIzLgwGJfn41Fj+v0pdcvO4u0cR4EIODkdo7txbcrlaNgRQwJ2zLgbStRBN\n/NYAnpW9lwYBGAygK4CedegPy5a5QygU2epev07BhQufhY7T0org57cQmzYNQXExR6Je8+baqPjd\nMBjA27cpFd4z8PPPj/H8eSK++eYWSktFmzlxcbm4dSscCxfew61b4YiPr5oWWRHOnHmP5GSRHqjV\n0KGiwB97e5gPGKCU9qsj5e1bvDtyBJ5//YUINzcUffqk9D4KUlIQfvs2ssLClNpuU3NzqGpowKxn\nT6jr6Mgd9VxOqr+/TMF0qurq6LtqFYozMiDgcuEwe3aVMk1atYKdszNaODqi7YgRUoPpUt+9Q9Cp\nU+L3Xlu2ICs8HK+2bIH35s1wX7oUcc+eSR2DYdu2mHH3Lub5+MDO2RklmZnw3bcPiV5eyImNRXpg\nIB4uX474ly+RFx8vtY2kiu6hREh+86bWc68LfA6n9kJfgLq4YaqoMLBx4yDk5rJw9uwH9OhxHIGB\naUocXcNTl4l/HIDyhNjnANTkVKyUx5CKG7GV39vbm6JLF1NMntypiqePtXVzuLqOgqoqAwCBSIg5\nc9zw/LkoP4iurjoGDLDAvHld0Lu3GVq3Fj0Km5vrw8MjGtHRuXj8OEbq3oE8sNk8/PzzY7i4vMTI\nkRdx5UoodExMMOPuXcy4c0fs3VFfGLZrh/dnziDx1SvkxMUh5PJlic0/WWEXFsL9p59waexYfPjn\nH/HxzI8fcaRzZ1ybNAlHHRwQdvNmlbrCCkIgDQm3tBTxL17Ae8sWBJ06hagHD/Bw5Uo8Wr0aRenp\nUusUpafj+4AAjNi9G9wym3k5sbG5AACzPn3wvb8/HObMkdoGSTlfNW1ttOrVCzlRUciNi4Pl4MG1\njp/BZOLtoUN4e/AgbkybhvyEBITdvImEFy/wZM2aaqNoTSrcjKrTjVAmVyZNqtf2vwTDhrXFpEmd\n8MMP3WBh0RQDB1rC0VG56mHyUB79/KWouPxlVHpfkTiIzDz+ABZWUwaQwdRz/Lg/MRgiU03Hjq4U\nHS2fnmmHDgcI+Kvs33pq124/ZWUVE5HIVBQTk0OZmUUSdTZteklHjrylLVs85eqrOtLSCklffxv1\n7HlCKe3Jy50FC+jGzJm0y9SUthsY0NkhQ+TKx05EdGXSJInI3ih3UW7xytq2x3v0qFLXZ8+eOo2/\nLhGU78+fJxeAXNu3p92mpuJxurZrRzwZctkTEeXklFBiYj45OZ2j5OSCWqPJhUIhXZs6VdyX+7Jl\nRCSKjI17/pxebd0q8/hT/f3pwqhR9M9XXxER0YeLF+nhypXkNm9elbICPp8ujx8v0nsuM0XVVbe5\nJmKePqX9bdvSeoD2t21L0Y8f11tfDc3y5e5lJuL1NHnyFfLxkT0VdH2wbJm7xHvUQ+TuEwDSIj5+\nh2iVXzHaKRciu39lWgBIB2BU1t4yAF5SytH69evFbwYPHozBUlZCMTE5mD//Ll69SoKKCgMHD36N\nxYtly9A3btxl3LsXhfLvaenSHjhwYDS4XAGys0uxdasXJkzoAHt7Uxgbi3b/eTwB1NRUxK91JTj4\nEwwNtREdnQNbW2NxP/UJj8VC+O3bYKiowGb8eKhpauJo167IjoiAipoausydi55LlsicB2dvq1Yo\nSvv8qDvwzz8xZONGPP3tN/hs3y4+bjlkCOY8fw4AyIuPx6stWxB6+TI6TZ2K/r/+KrdHDre4GDGP\nHqHT5Mm1n3NpKW7PmYOEFy9g2qULJl+6hIyQEMQ+foy0gAAkVDKv/BAaCmPb2jO/JicXYNq06/D1\nTcXkyR1x+vT4Kt5llSEipAcGQkVNTaoHjzSyIyPRvEMHiWPFGRnQMTYGKy8P6rq6EHA40NDTA7e4\nGOqVMo5+vHEDN6ZOFb9X1dLCb0VF9eo5duHrrxHj4YF2X3+NWWVpvf/tJCXlw8JiPypOlW/ezEfv\n3mbVV6q3sRRg8eL7ePToKVq1ysWYMe1hbKyLDRs2AEqO3B1ew2cZEN0UPkE0uWdWU678OToLwG2I\n7PzSJn64uLjUMhzg48csvHolinAUCAg//eSB+fO7yjQpHz8+BgsX3kNQUDosLZti+nSRv7CqKhNn\nzgTh0KG3uHkzDNeuTRVPyOXtKmPSB0QmKQAwM2uCZcvcsX//yHoNA+dzODg/dChSfH0BAO1GjsS0\nW7fw9cGDiHn8GF4bN0K7eXO5kp+16t0bEbduARDdQs3KtIb7/fwzYh89wqegIOi2aIGR+/eL6xhY\nWaFFly74cOYMDNq0kTrps/LzqxX+DndzQ/iNG8gKC0NubCx6LVtWYxS017ZtCL9xAwAQ/+wZHq5Y\ngXGnTqGNkxPyEhNxyMYGgjIhdA19fZm1cVu31kefPq1hY2OErl1Na530AdEeUstu3WRqn8dioTg9\nHd7btsH+22/RvEMHNDETTTK6JiI/eO0y19FyN9PKkz6AKmZDAZcrCuSqx4m/mbU1hm3bhqAzZ6p8\nlhUeLkprXU+a0fWFNK8rab78pORsotIwN9fHpEkdERKSiblzu2DDhiEAUD7xy0VdZpy7AMoNm3MA\nuEkpow2gPAGODoARAEIU6ay4mIOkpALw+ZI2aYGAqtj+q8PUVA8XLkzCxIkdERWViz//fIGnT+PA\nZDIwalR7jB5tDXt7EwwYUL/ReZmZxZgw4QqOHQuAo+Nx+Pml1F4JgLd3Yu2FyihPpZsWECCe9AEg\n5uFDPF23Dk0tLWHYti2WRkWhSevWco1//Jkz6LF0KazHjsWEc+fQftQoACKR7YX+/liTmYmVSUlV\nVrc6JiZYmZwsNXKUXVgITymC6DwWCyl+fjCytUVWeDjSg4JgPXp0rakvCqWkQyivY2BhgRl376JV\nr14w69MH39y/j/Dbt/HPV1/h3qJFYFcQXZfGunUDcObMeHzzjfJzCKmoqyP40iV8OHcOd+bNQ3Fm\ndeupmrGZMAGmZak2CMCg9eslRHHqg1H798PUwQGjKtzwhXw+Unx94efqisDjx5EWGKiUTLUNRevW\n+li79rP//uzZ9ujRQ9K+z+Xyce7c+8pV6wVbWyMkJq7AsGG1xwfVF4YAnqKqO2dLAOXPeW0AvC/7\nFwrgtxraq9amlZCQS3/88ZSmTr1GkZFZ1Lv3CbFb5ubNVW3vAQHS9VDLefMmidTVN9GECZfFx7Ky\nSkgoFFJeXmm96/pmZBTT7dth1Lbt37R4ce1ZO/PySunu3Qjq3/80nTkTRIGB1Z8fl8UidmEhXRw9\nmjjFxZT+4YOE3d0FoPUA7WnVikqysqptRxoFKSn07I8/6OWGDVSam0vBly/T419+Edv4FSX26VNy\nbdeONqmr082ZM4mVL9IZZeXn0xEHB9G4mUy6Nm0aRbm7y5SVNMrdnVwYDPE5vz18uPqyDx5IfD+X\nxo6t0/kQUZ2yWsZ7etL16dPrPA5uaSnFPHlCaYGBdWqnrkTcuUMbVFRol4nJFx+LokRHZ1NYWGaV\nPSYPjyiaNesGtW37N/311zOlZxCWBfwXNXdv3w6jNm32k5raRgJcqE2b/eTrm0ze3okUGpohUTY8\nPJPu34+kQYPOkJdXIuXlSd+w8/SMp5SUArp8ObhB0q1WRCAQkrPzDfL3TyU+X0B+frLpah458pYA\nF+rZ8zjl5kqKS7OLPm9IZ0dF0dGuXckFoLNDh1JhWhp579xJG1RVRRt9Ff6F374t87hZ+fm019xc\nXHdXixbi9tYD9PHGDZnbksa1qVPJtX17SvTyEh/zO3BAYrw7jYzkajPh1Svy3LyZIu/fr7HcCxcX\niX52mZgodA4FqalUnJlJMY8e0TFHR4q4e1fuVM1ERCXZIqcFVn4+vXgWIxH38W8kMyyMbs+ZQ5fH\njyc+h1N7BTlhs3lV/iaUzdGjb6UeFwqF5OR0jgAX8vaWTbtY2aCB/fgbhAkTOsLKygA6OuoYNqwN\n+vRpjV69zNCvnzlsbSU1Q5s21cSGDZ7w9EzEy5fx0NeXbn8dONASrVo1gbNz53q3y1UkODgDdnaH\nceVKKNaseYz09CL07Cmbbbl5c23s2DEMDg4maPJ/7Z13WFNnG4fvsKcoQxFF3HvvvbfWrXWvat1a\nR1vbasVaR7XWWa3WWfcq7r0HbsGJiCjI3ntD3u+PSCSQQALB2n65r8srJrznPScn5zznnGf8HkvF\nR/ZL338v/79NpUqU79iRqn36UKlHDyxLlqTF118zNyZGsXOVRIKVBlXCwe7uCu6ThKAgeTRJAnge\nVebpUw8hBC3mzmXSs2cYZ6kqza7DI9HQP+3UqhWtf/iByj165DouM0aRSZmWmpXmCyGQSqVcX7SI\njLQ0fG/cIOjRI+6tXSv3z2uCmY0NyclpHDnpy5JlrixadJ3Df17k3a1bSLN0Rvs3cGPZMo4MHkxy\ndDTd1q8vFDfPhQveHDv2UuvzgqzeZ+LEE/z44zXmzr1IaKhiWq8Qgu7dK+LqOvYfT7PUhE9eqyc9\nPYOffmpLsWKmpKVlYGtrpjKQYm9vSa1axenUqTwVK9rIx3h7R7Jy5W1A8PXXLShXLn/SywWldu0S\ndO1akeLFzRk6tJZGHYB6966KoaE+qSlp3F27luazZhHk7s6piRMJeviQ6Ddv+GzzZiwdHGgyfTpF\nSpeW+/kBjMzMGHL8OKcmTiQ5NpZW339Pyfrqt5u0KlMGPUNDpGmy4ras/wcoVgDFR4lEgsP7bcmq\nvW/p6CirtntvLHKTapZmZOQ7Y6Vi587037+fF4cOYeXkRDsNg2UpcXGcnjyZp3v28NLFhZpDh9Jh\n2TJiAwIwtsxfjwcTE0McHYtw4YI3Lx+9JjViOc9Jpmy7dgw7ezZP/aCsqDpf3ly6lKPZvDZ5fugQ\nl7+TeXdDnjwhJTZWnuWlDYQQrF9/j2++uUB6uhQ/v1h++KG1VoXUGjcuzebNDwkNTaBOnRI5svD0\n9PSYNat5rtuYdd8HB8djb695z+f/MuL6dfXzY7/66owwNl4kHBxWyiUVpFKpvOtV5mtMTLJwcPhV\nHhMoU2aViI/X/uOmunh5RSi8asLbq1fFnp49xXI7O3FyyhSREBEhzn39tVhZurRw37VL25uag2eH\nDon11aqJDbVrixcuLmJ/375idbly4ujYsSJVzTx4TdjaooVY8N6V5Azi1JQpSselp6YK199+0/r6\nNeHt1atiW6tWYl+fPnJ3RkH3ybVrPmLFsquiId3EPD7EKzw16OaWGBEhXhw5ovBZwIMH4v7GjeKP\nevXEk717td7lKpPLCxYouNB+tbfX+jpSU9NF7dobRI0av4uYGO0fg0IIsW7dHfHkSbA4dOiZ2ssk\nJKSIzz7bK/T1F4rq1dcLT88w4e8fI6ZMyd3tmB/Ih6vnk7rjr1hRPYXDEyc8Wb36LiCTahg8+AhB\nQTLFQlNTmRsk8/Xly3ACAz/o4L97F8OrVxHUq1eSjPT0j96ZKPM7qvtds1K2TRsebt5MYkQEVfv0\nwczamqq9e9Nx6VICHzzQ9qbmoMaAAdQYMED+PnsHqOSYGK0KgOkZGCD4kKCsp+S3envlCq4rVhD4\n8CHRPj60W7gQExUpoQUlPTmZF0eOINHTo3r//ugbGZGakICRuTlWTk6MvnaNaF9fJO+309DEpEDr\nq11On/hDh+nBGUSWNG2Jmsfss4MHeb5vH1Fv3xLh5UWzmTPRNzLCrlo1Ln7zDcFubqTGx8vTQ5WR\nHBPDndWrSY2Pp/748dhqkPZboVMnri9aBO+rwyt06aL2suoSG5vC5cujMDLSJz4+lSJFCrbPlTF1\nqkwYL1NWOi8SE1NZufL2+5ohePEinEGDDhMUFE9oaAKxsals2NADC4sPT21SqZSMDKG1tPG8+KR8\n/CVKqPcIFBys2NAkLCxBZaP0cuWKKkizWlkZ4+QkMwz3N2wgNTExn1v7z1Chc2dmvH0rNwNlWrRA\nT19fLdVGZQgt+Vwz0tO5+F1uSVsazpeaSq0RI9A3NEQCWJYuTfOvv84xrly7dhhbWpIcGUm1fv1y\nNfqenvnv6JSRlsbODh1wGT6cv4cOZXfXrqSlpnJ53jxA1nRdIpFQrGxZ9LTQIyAtKYm/2rXl/vr1\nSAC99zd11QYMoEKn3MprPlC9Xz+i3r4l5PFjagwaJJd1MDQzo3jt2vT4449cZZqlUim7u3ThmrMz\nt3/9lW3NmxOvgb5TmRYtGHHhAg0nT6bDsmX03LxZ7WXVxcbGDBMTAywtjSlZMv+tU7XFpEknMTdf\nwsKF1xQ+FwKaNClFixaOfPllfQWjf/KkJ9bWyzEx+ZkZM858lO38pAy/ukbos8+q4ODw4UceN66+\nyiIoOztzTp0aStu2TrRrV5YzZ4ZhYZTB2Zkzcf3lF/Z99hleZ/K/s1NSPm6wre6oURQtU4byHTsW\neK6UuDg8XFwKPI//vXtsrl8ftz//ZF/v3iTkI/c8OSaGu+vWce/330mJi0PfyAhjCwukaWmYlyhB\nnx07lBZZCSGo2KMH0729ycgSc8hKWFgCr17JKr5fvYogIiL3i70QAt/r1xU+C3Jzw9/VVf7e58oV\n/mzQgAcbNnCwf3+tC95Fvn5NlLe3wmf9Dhxg0KFDSmMZyjSQMtLSqPfFF4y4dCnHb9L5119pOGEC\nNT//XOU2JIaFEXD3rvx9UkSEQk2IOpRv354ev/9Oy2+/1SguoS5Pn4awf/+zHJ+Hhyfi6Rmu8oaw\nMDh/3ps//ngISLKtV/DFF3VYu7YrN26MoXTpD0kWUqlg2DAXYmJSkEph7dp7nD/vnWNubfNJuXrU\nrWC1t7fgwYPxHDvmiZ2dGf365V7636qVE1eujFb4rPaIEdxbv54SderIC5Dyw5IlN3B2bvtRs4O0\nwQsXF14eOULYixe8OnWKIqVKUa5dO8q1a6fxXKUbN8axZUvSkpKoM3Ik5sVl2VbS9HQSIyOxKF48\n1+Vf/P0315ydCX0qq+17vHMnY27epFiFCtQfP55ILy+C3d2VBiIlEgl13ytnqspS0teX8NVXZ7l1\ny485c87z11+q9QT9XF3xc3Xl9ZkzNJw0iYrdumFkbo6ptbWC2wk9PUo1akRyVBQ1hw7Ndy/b9JQU\nDIxl2Wex/v74Xr9OsQoVsK5UCSNLS1Lj4gCZ5ELZXNRb7/3+O02nT1f4zNDUlCbTpikdn3nxyC0g\nblKsGGZ2diSGhb1fSI9in1A7zh073PnxxyskJKTh5RXBTz+1w8jIgEOHnjN8uAupqRm0bFmG8+eH\ny12/hUl0dHK2T6To6+tTsmQR6tVzoGxZWVJJ5ivI+vrGxipWWed1Y6INPqk7fk0oWdKSiRMb0r9/\n9XwZ3fTERKa9ekXNoUPz5e4ICopj+PC/WbHClW7d9uDmplzh8VOlco8ehL14QZCbG27btnFj0SJ2\ntm+vVFEzO8ruLuuPHcsUDw+MihThyLBh7Gjblgtz5/Li0CGEEEr38btbt7i+eDGXvvuOkKdP5RGq\nwPv3iXz1Crtq1SjTsiXBbm5cnDuXc3PmkJ6qecqctbUZDRs6MGVKIxo3LiVv56kMmypVeLpnD28v\nX0bPwACj964Qm4oV6bpmDfrGxhiYmtJz40bqjhnDjLdvsVRT7iE7QghuLF4MyLR5Ntauzd/DhrG1\naVOeHzjAkJMnKd28OaWbNWPwsWNK+wwH3L+Py4gRXF+4kNPTpxP/vvViVtLSNFdElUqlZAg9hp05\nQ+lmzSheqxZ9du5UyLr6pxk9ui7W1qYYGuoxd24rjIxk97HTpp0hNVX2nW/efMeuXU9ym0ZrdO1a\nkapVbd6/k9C9e2UqVbKhXbuyKtUATEwMGDeunvx9+fJF6dKl8C+un9JtqtCWv/ljsX//U7744jiT\nJjXk11+1H7gqTNJTUri9ahXP9u8n5PFj+YFQfeBABh48mOuyrr/9RvNZs5T+bWe7dvL2eQJZulur\nefNoPX9+jkB6enIyuzp35t2NG4pBXENDZvr5ybVptjRtSlJkJLbVq5OelETDSZNyBJbzIjo6maJF\nTeSvmfjevIn3uXPYVq1KraFDkUgknPjyS2yqVMGuRg0qde2qMI+QSgl5+pSjo0eTEBJC/S+/pJ0a\nGlPZeXnqFC5Dh5ISG0uRUqWo/NlnPPzjD/nfrStVYtqrV2rNdWLCBB5t3syAQ4cUgu8gu7j8/PNV\n5s9X/0nuwgVvBg06RExMCqNG1WHr1l5aiVtom7i4FN68icLe3oK4uFR5woSt7XIiIpLk49av78aU\nKQVqBaI2MTHJnDnjRbFippQsaUGNGsV58iSEevVyXrQzkUqlnDrlRVRUMt26VcDOTrN0z/x04Pqk\nXD3/NmxszAgKms2VKz4FnisxMZUFC67i4RFOjx6VmDSpcHXTDYyNaTV3LpFeXoQ+fiy/2zYvoTpz\nIT4khKsLF/J8/34C792j2ezZlMqm7x706JHCe/MSJWgybZrS7CkDExNKNmxIvXHjCHzwgOcHDqCn\nr0+X1avlRj81IYGBBw+yp3t3PI8dQwK8uXiRLx88oOR7LZqI168JdnfHvm5dbFTUE2Qa+6xG/+2V\nK/zVsaM86yTCy4u2P/5Iz02bkEgkZCgplpLo6XFo4EAivbwAuL5wIaUbN6ZS9+4q95synu3eTWps\nLACxAQGEvVQsQMorMymzbiHW3x+/27eR6OtzZd48bCpXxv69RpK7ezALFsj0qF6/jmbp0g44OBTJ\ndV6A4cP/Jjpa5n7YseMx3btXYuDAvJVLPzaWlsbUqSNzsWU9bJcu7cDEiaeQSgW1ahVn+HBFzShR\niIJqVlYmDB6s+FSUm9EH2c3RZ5/JlFhdXDxo0qS0QgyzMPj0LuMfiaSkNObMOU+PHnv5/fd7+XL3\ndOpUgSJFTOjdu2qBt2f69DP8+uttTp3yYvLk0+zdq1rLzs8vhsmTTzFu3HFevAgr0HrbLlyIkYUF\nsr5m4Hv9Ohkq3CkWJUpQoWNHkqKiMDA3z2H0gRxNRfrt3avSPSOEoMvKldQdOZKuq1fzdUgIswMD\nqTlokHyMkbk5Fvb2hD1//uGWRiol2F0miuVz7Rp/1K7N4YED2VizptrNuoVUKgtsZ2lE43H4MBI9\nPblRUJXqG+OrKJYX7eOj1jqzEh8YqJB8bWlvT7mOHZEC5vb29NiyqCAPAAAgAElEQVS4UelyCWFh\nbGnWjEUGBmyqXx8kEprNmoWFvT1V+/WTG32AunXtqVzZhowMQZcuFdQy+kIIYmIUfc6RkUkqRqtH\nSko6V6/KOoR5eoZz/LgngYFxBZozN8aPb4Cn51Ru3BjD3bvjsLJSTPHcsydfOpGFSmbf8AULrjJq\n1FH27HlSqGJ2/7eGf/r0M6xceZvTp72YOvUMEyac/Ee3586dAIX3t2/7KR2XkpJOu3Y72bjxAVu3\nutGmzQ7Cw/MfDIoLCCA1/kN6bOiTJ4S/b/CtDD1DQ3r+8QdGKtQx++3ZQ7Ovv6bWsGEMO32acm3b\nUsRBebciiUQiN7K5uRL0jYxwaPzhUV3PyIhS79/fWb2a9CSZYcpISeHumjUq58nK2ytX0MumVlm0\nXDm1lq2e9cJUpAjlO3dWa7ms1BoxApBdbCV6epTr0IF6Y8dSrFw5Oq9ciZmd8obgx8ePJ+B9Zk2w\nmxsXv/0W2ypV+MrXl8pKnjoaN3YgMHAWxYrlrmaaiUQiYerUDxd0R8ci9O2rWd+ErFy44I2z81V+\n/vkGM2eepnr19fTuvZ8aNTbw5MmHTKjs7VILSsWK1rRsWUYhqOvpGc6PP17mq6/O8uOPl+Vd1JRx\n9aoPZcqswtJyCfPna6/aWBWGhvqMG1efFy/C8PePYciQwpWT+b919dy+rSiF/Oefj+jVqzI9e1ZR\nsYRq3NyCuHz5LTVrFqdLl/xJFzRtWornzz/cvTdrplzj5d27GLy9PzQ7Cw9P5MmTENq3V89oZcfS\nwUFBfkHf2FjuZsmOVCol4tUrrsyfD0Jg5eRE8zlzFA5QIwsLOi9frta6hRDEBwdjWbIkD//8k8jX\nr4kPCqLm0KE5fOtDTpzgqrMzKVFR1Bs3Tt40JbskglEeEgnSjAxur1rFk1270Dc0pHTz5sT4+mJb\ntSo9s/jYc6P39u2UadmShNBQqg8cqNK9lBsNxo3DqkwZQtzdKdOqFaUaN+bY2LFEv8+7rzVkiML4\nkOfPifD0xPfqVfmTggRIDA/H8b3WkDKNoYEDZT0nunWrpPa2rVjRmU6dKhAcHE/37pWws8t/s6CG\nDR2YMeMsHh5hhIbGv++ZLYu5rF9/j82bewEwb95lVq/Of3ZdXiQlpVGlii0ZGYKIiCSMjPSpUEF5\n4ZoQgv79D8qfdH7++Qbt2pXL9zmmLkFBcVy9Ogpv7ygSElKxtMy710N++VcY/ocPA1m58jbGxvq0\nbu3E8eOvsLU1ZdGidtjb588X1rx5aQVDCwJXVz+NDb+rqx9t2+4gLU3mMlizpivTp2teTLV+fQ9s\nbMzw8Aine/eKDBmiPHvCwcESOzszwsJkd/lmZoZUrmyjdKw6WDk60nfXLi5++y0SPT26/PabPB0z\nO3p6ejSZNg23bdsAaDJtmtp3JQnh4cQHBSlkhbw6cYK769ZRvE4d7q9bJ3cxPdmzhy9cXRWK0iyK\nF6fnhg055m23aBH+d+8S+eoV1pUq0VaJpr/Cd9DXp/GUKVz/6ScMzcyY6e+vcfW2voEBDSdM0GgZ\nZVTs3JmK758W3t2+zbvbtzE0NcXn2jUSwsIU0mCtHB05P3MmKTExMvE6qRT09Gighe3ITlJSGmvX\n3uXMmddUqmTD0aOfU62a8ieQvLCwMKJ9+7JMnNiA3bsVs2vMzY14+zaKSZNOceHCG549C+P337tT\npYqtNr6GAnv3PqVVKyccHCxxdR3L48eq6y5SUjJyuLcK0zWVSWa8omXLwu0HAv+CrJ6goDiqVfv9\nvd8x670ONGrkwL17ubXxVZic1Ph4+R1icnI6dev+gadnhHzeEyeG0rOn+iXpAFOnnub33+/L39et\nWwI3t4lqLZuYmIa+vgRjY/UMjxCCceOOs22bOxIJVK9ux7p13WjXrvDuRHxv3MApS/54fGiovKGH\nNC1N5UUi6za/PneO1+fPkxQaKqs87dIF1+XLuf3bb6TGxqJvbJyjY1SnX3+l+ezZam2jEILE8HBM\nbWzUyj6J9PYm1s+PtMRE7OvXz3cOvrY5Pn487jt2MPDgQar17Zvj72dmzMCqTBlZJlDp0pSsX5+S\n9etr3SWwZMkNfvjhg3ujXbuyXL6svJl8XkilUrlL7/HjILp120tQUDy1ahXn4sURFC9uwbp1d1m6\n9CYzZjTh2281U0bNi9TUdH788Srbt7tTpIgx8+e3ZuTIOnkuN2zY3/I4W6lSljx6NEHrbVLDX73C\n0sEBYyUd1DThP5nV8+xZaLZg04fvd/9+IOnpUgwM8j7Zg93cCH78mHpjxgCy/Nn798fzww+XePMm\nmoEDq+dq9JOT0zAxyVkEkj36rm40fu7cC/zyyy0MDPRZu7arWlk85855s22bLKgphEx1tGVLWdFS\nQkIqGRlSrWmVJEVF4Xv9OreWLaPBhAkUr10bh/r18yzGyo5EIsG2cmWOjxlDYmQkTWfOxMjUlAbj\nx3PV2RlDU1Osq1RBZGQQ8vSpPOhUok7eJ2fWdZir8Ikrw7pCBay1WIgUF5eCiYlBgXVWnNq2pd2i\nRfLAdXY6LV+OgbGxQtHXs/37qTl4cIHWm53sMaPMp8v8kPVCXKdOSXx8ZhARkUSJEhZyFc26de3x\n85vJrVvvVE2Tb4yMDPj66+b8+ecjLCyM1DL6AH/91Ydu3SoQFZXMgAHVtWr0M9LTCbx/nye7d2Pl\n5MTzhFLc87egYUMHJkxo8FFSZz95w1+1qi1mZgYkJmam1n3I+G7YsGSeRl8IwaOtW7n644+kJycT\n4eVF+59+Qs9Apu+xdq16aXgHDjynadPS7/2EUqZNO83Jk15UqmRN9+4VuXrVhxo1irNhQ+7a7y9e\nhNG58y4CAuIACenpGUydepoBA6rn6UuNiVGsDExOTpc3gT9z5jXp6VIGD66p1vfJC9NixYjx98f/\nzh0k+voMOZn/4LeJtTWVe/XCpGhRDN73XI0PCaHvzp2kJiRQplUrzGxsODdrFnGBgdQZNYoKWpCk\nyCQuLqVQ/KVSqWD48L/Zt+8Z5uaGHDgwgB49ct48PD94kJcuLhSrUIHW8+ZhoEK8rc6wYQBYZotv\nZJJp7A2MjXl38yYvjx3jpYsLUT4+NPjyy1zF1jRh5Mg6bN78kIQEWdxn8uSGWpkXZIY4u6ZOZnFT\n69ZlNZ7Pzy8GR8fchQFDQxN4/Hgifn4xJCSkYm6et3SEvr4ew4erf/OhCfoGBvy1zxPpn3+CiSUb\n4wYQhAPbtrkTG5ui9aceZXzyrh6AGzd8+eWXWxgbG9CxY1nOnPHG1taUxYs7qCXMJIRgY+3aJEVE\nMMXDQyMFybS0DJydr7Jjx2MsLY2YO7cFsbEpzJhxTj6mR49KnDw5VK352rTZwfXrWdMBZReyt2+n\nK5RyKyM+PpXmzbfy9KlMd2X69MasWtWFVavu8N13lwD4+ef2fP11c4QQBb5zeLJvH3EBAUR5e9NT\nRXqhOmRVQc1LETX0xQse/fknxpaWNJ01S2XzdU1YtuwmI0bUplQpxXTG2MBAlRlH6nDgwDMGD/5Q\n6Wxra0ZYmKKQnNeZM+zNkm1TZ9Qo+uzYke91ZiKE4ECfPngeP86gv/9W6hoqCK9ehXP1qg9VqtjS\npk3ZXMempKSr7a7UFgkJqfj7x7J8+S1WreqCqanhR1O2LAiPHwezfbsbf+1wo23pUKxDH7AjrA0Z\nyC5GHTqU4+LFkRrN+Z909YDsjiBryfOkSZpV4SXHxDBg/35MbWxIiojQyPAbGurz9dct+P33+9jY\nmDJ6dD1mzjyrMObNmygVS+dEmQ7HwIHV5YqhuWFhYYSr61guXXqLlZUJbduWBWDq1MbysvSpUxsj\nkUjYsOEeU6bkT7EzkxoDBqBvaKh216f09AwMDHKefJmG3uf6dSJevqRCly4UdcoZwIoNCGBbixak\nvG92/vrcOcbduZNvH3ZiYipz5pxn9+6nbNr0gJUrO9OvX3WkUikIwelp0xiwbx96Bgb5ukhm11iJ\ni0vJURzkc01RpVHdOoO8kEgklGzYkEbTpqlsDi+VSslITc2XPHTlyrZUrpx3kDU8PIHdu5/w1VfN\n8hyrTdzcghky5Aj+/rEYGemzalVXCrmXfIFxdw+mf/+DvH0bhRDQcckwwsP7wIIL8jG1a2vmSs0v\n/wrDX1BMixZV+84xKCiOEydeYWdnRp8+VZFIJAQHx+PmNgF//1ji4lLo06cqa9feldf+DBigfp7z\nlCmNmDz5tGy7TA1YsaIjkyY1Vtu4WVgY5ygYi4tL5fr10UgkEp4/D2H37qds3+7OixfhTJvWhKpV\nc57A6lQvZgZxlengZ+fJkxACAmJVpg3eWbOGs199hQQwtrJirKsrxatXVxjjf+eO3OgDBN67R2JE\nBOa2+cvyMDMzYtasZhw79oqyZYvSr59sfQnBwfw9fDg+V67wR506DDhwQKHwSV3696/OL7/ckqfX\nzprVNMc+LdVQ0U3i0FA7bhMhBK3nzUMikags9Am4e5f44GCtPw1k4uLygoULrxMYGMfbt7LKYDMz\n7StwKqNlyzI0aFCSGjXs+PzzmpiYfPqmbMyYo+9vEmW/1/37fmze3JuYmBSuX/elUSMHlizRnosz\nNz79vfURCQyMpVGjLfLUrSlTGrJ+fQ+54cxs2dimTVmuXh3NuXPeVKliw4gR6vsCJ01qRM2axfHy\niqRt27KUL1/wNpC2tjK/eUpKOo0aleb27QDi4lKxsTHLYfT9/WMwMjLg3r0AjTOYVPHHHw/YvPkh\nSUlpuLsH8803LXIord5euVL+LJoSE8ODjRvpvm6dwhjrSpUgM1URWQVrQZuqJCen4+ExBU/PcHki\ngKWDA9UHDULP0BC76tXzZfQBrK1NefDgSy5efIOdnRmtW+d8iqk+YADd1q3D4++/sa5YkU4rVmi0\nDi+vCDZsuI+JiQFz5jTHxkb2W2e9wCi7gN/7/XdcV6xAmpZG6LNntPr++3y3plRF377VWbv2Hm/e\nRDFtWhO50ZdKpYUeoJRKpWzc2IOSJS0/SqqlKl6+DCMqKpn69Uvm6e76ECSX/V7lylljaKjPypX/\nLp0vbaP1lmSasmnTA3mLRnAWRkaLhFQqLfT1Zq4jPT1DvHsXLZKT0/I1z4IFV4QQQuza5S5evQoX\n27Y9kv8tNTVdhITEi5Ytt4mePfeI+vX/ECtW3BQpKflbV/btd3T8TVhYLBYJCcrbWv7ZpIm8haIz\niJvLlysd93jPHrGxbl2xrVUrEeTmVuBtU0XkmzcKr58iYWEJonjxFfLjsVatDSI9PUPt5VeXKyeW\n29kVSltMIWTH1OnTr0RgYKx48CBA/vmOHW7izZvCaeeYF48fB4kVK24JFxePQl/XihU3BSwQ4Cya\nNv1T3u5VFT/9dFX+W1pZLRUvX4apvS6pVCri4pSfW+Sj9eK/Irj7sXBx8aBfvw/KlCVLWhAYqF4u\neUE4f94ba2tTRo8+yvPnYRQvbsbZs8PzFHfKJDAwjq+/voCLiwdt2jixZEmHHMu6ur5jyJAjvHsX\ni4WFIfHxqfj5zdSo4bsqwsISePQoiKJFTSha1JgqVXKmVga5u7Orc2cSw8JwbNaMUZcvq8xu+ZT5\nGHezmZw9+5pu3fYofObr+xVlyuT9myWGhxPj74+RhQV6+voUU1OOoiAkJaWxdOkN9u9/TmJiGuvW\ndVMp9yCE4KefrrF9uzsODpZs3dor30VimTx8GEiLFttISZFJMv/0U1vmz29ToDlVIZUKTE0Xy+Wf\nAXbv7suwYaqfHoUQnDv3Gl/fGDp1Kk/58uplYd2960/v3vsJCUmgS5cKuLh8riBFkZ/g7v+tVo8y\n+vatxtSpjTAy0qdkSQv27+9fqOvLyJDy22+3mTfvMgMGHOT581BAEBqayJw559Wex8HBkp49KyGR\nQK1axZVeMJo3L0PLlk60b1+WwYNrcOrUUEJCErTyPezszOnSpSJNmpRWavQBStaty9chIcxLSeEL\nV9d/pdEHcNuyRWtzvTx2LNe/V6hQDEPDD6eojY0ptrbqae6Y2dpS8r1a6ccw+iDrc928uSNeXuEE\nBMSxcuVtlXo4x4974ux8DV/fGG7f9ufzzw8XeP0HDz6XG32Av/5SXg+hLfT1FW1tXqnlEomErl0r\nMWFCQ7WNPsCECSfl5+q5c95s2vRQ843Nhs7Hn41167qzdm23j9JRS19fj5EjazN37sUswSnZepOS\nNGvpaGNjSnDwnFwlopct64CjoxXv3kVTpoxq33lKSjp6ehKtp8dJJJJCab/3MYh6+5Z769bhtnUr\noc+f02jyZGyraK7rBBD6/DneFy7gumIFMe/eUbV3b6XdwypVsuHgwYH8/PN1TE0N+fXXThoFT2Nj\nUzh69CVmZob07VtV7Q53BaFIEWMmT27EyZNe9OxZSaUeTvYLgiaZcarInq5bqlTBn2ZVoacnYc2a\nrkyceBKpFDp1Kk/fvgVX6VVGdrXU7PU8/3bU9nf9l3j+PFScPv1K/PTTVVG06FIBzsLQ8Cdx4sRL\n4eX18f2kixdfE3p6C4WBwU9i/fq7H339nzK3VqwQC0BccXYu8FwnJk4UziCu/fyzFrYsJ/HxKaJm\nzQ1yn/KAAQcLZT3KePo0WOFVGS9ehAojo5/k29e48eYCrzctLUOMGPG3KFJkqWjQYNNHOX8CA2OF\nh0eoRrEXTVm79o58P9na/iK8vSMV/o7Ox6+cLVseMm5cg3wtq64kRH5xcfGQ+0GDg+N49CgIJycr\nzMyMWL78Fj17VqZOHXuFBs2FxYsXodSosYHMw0JPT4K//0y1iuSUERoaT/Hi6uuQSKWCzZsf4O0d\nRe/eVQpVrOrGDV+V7fBU4b5zJ6WaNMHP1ZX6Y8cWaP3XlyzBvnZtot+9o/Hkyeze/SRHw5CCcPq0\nFz167FX4LDh4NiVKFEwXRhukpqYTG5tCyZIrSU//0A/B03NagQQH/8vcvu3H27dRtG1bFgeHIrx7\nF8OKFbfIyJCyceNn8F8s4MovYWEJrFp1m+3b3Xn4MIgRI2rTvLnyhtzK8PGJ5s4dP3lHHW1eBG7c\n8OH8eW8OHfLA1zeaUaPqYm9vSffulkilMt//H3885PjxVxw5MuijGH7ZI+WH40cqFe9P0PwZ/unT\nz7J//4A8x/3112N++OEysbEp8qKo1avvcvXqKFq0UP/3Ugcfn2gePw7mm28usnx5R2rXLiFP082L\nuqNkQmV2VQv+SJ/W5HOs69iTUCqG77+/xNatbnh5RfDllw1yuCzyQ2aKbyYmJgZYWPzzbraoqCR2\n735C584VSE9XaF9PaGjCv8Lw/xOVys2aOdKsmSMgq1pu02YHPj7KC/fU4T8d3LWzM6d+fQeCgxOI\nj0+V77i8EEKwY4cbgwYdYt68KwwffgRr62UYG//MhAknZJWfBaRZszI8eBCEp2cEpUoVUWiUoaen\nR6NGpejSpQK1ahWnaVPl2vzapkGDkrRu/cHQ9uxZKV8noodHGPXrb+LAgee0bLlN4QANDVUMKHt7\nRzJ27DH8/WOzVMIK0tOlnDqlXs9ZTXB0LMLJk6949SqCkydf4eio3QuqqgY6maSlZbBx430WL77J\nrFnn8faOIjo6idDQBEqUMNeK0Qdo3LgUCxa0wdBQD0tLI/76q49aGjWFydGjHnTpspulS2+yYcN9\nmjf/0KS+Zk07GjSQJSX4+8eyZctDzp9//U9taq4sXHgt70GFiJdXRIGMPvzHDT+ARAKPHn1Jw4YO\nagdsJRIJo0fXIyEhDX//WM6ceU1UVMp7V8Qjjh1T3aFKXQwM9GjQwJ6jRz8nu4vuxAlPunbdzblz\n3sTGJhMVVbDWd+piZGTA+fMjOHx4IEePfo6Ly+f5CnJXq2ZHz56VadHCkf79q1G2rCyQLJVKmTr1\ntMLYgIBYMjKUu/gym2drE319PcqWLcq6dV0pV66YUomJ/BAdLWuyPXv2efbte6pS793QUJ9Oncpz\n7ZoPt269o3PnClSubMOdO19gYaG5kNy5c14q/+bs3JakpB+IiZn7SfTM7dOnGmZmhkRFJTFlSiMu\nXhzFli2fsWFDd27eHIupqSFv30ZRr94mxo8/SZcue1i48Oo/vdly/Pxi6NfvAL/+6kqnTn8pdBD7\nmJQpY6XQO/rfTqEFR/JDbGyycHF5IR4+DBASyQKFwq4tWx4WeH6pVCov3MpeJObo+JvC+rIHWf38\nYgq8fmVERiaKR48CRVxcshBCiNevI8SSJdfE4cPP5WNSU9PFwYPPxP79T3MtNPPwCFV4ffo0RNSs\n+bsAZ9G69Xbh6xslhJAFIatWXa9Q2FKx4hrx9dfnREZG4QTM0tIyFF61xapVtwU4ixYttoqIiASV\n4x48CBDbtz8SCxdeEbGxyflal6dnuNixw004Of0mdu50F2/fRua9UCEQFqb6e2YnPT1DHD3qIXx9\no8S9e/5KxyxbdkPh2LezU17op20ePAgQX311RixefC3XQqzt292EldVSMW/eJY3mX7fujmja9E8x\ncOBBERgYq3SMh0eouHTJW61jwtX1nWjXbodo3XpbvoK7/2kff0GwtDSmTx9Z0HXq1MasWydrtlKi\nhLlWpA5yK7nPWhQCyHOTU1PTSU3NYMqUU+zY0RtTU0OlPQJy4/nzUKpUsc0Rq7h7159u3fYQFZVM\n6dKW7NzZh/79DxEdnQDoM3t2M375pSM9euzlwoU3ALRqVYbLl0cpjXtUrWqn8FqzZnG6d6+EjY0Z\nAwdWl6eTmpsbcfPmGLZvd8fQUI8vvqhf6L7ozO3VdtDe3t6CpUs74OMThbW1mcpxDRo40KBB/lVB\nQfY0tHKlK76+sTx7FsqIEdoLDKvL06chnDz5iu++a5X3YGRPW5k6U6rSibPHJrK/LwxevgyjVavt\n8hTq27f9OXFCudpumTJFCA6ezbVrvkr/rowTJzyZNk0m7HjnTgDBwXFcv66YHLBp0wMmTz6NVCqo\nUsWGmzfHYGurWqa9WTNHeXMciUTzRINPScfU2dnZ+Z/eBqV061aJokWNKVHCjCJFjKlY0Rp7e3ON\n89wfPAjAwSFvH26RIkacOiV7hK9UyZq1a7tiZmZEdHQyw4e7cPKkF6dOvaZTp/Jq90NNSUnH3T2I\n9evvExGRiJGRvkJzibFjj8lbUcbGpvL4cTC+vjHyvz97FkKfPlWZM+eDkuC7dzH06lVF7eYz5coV\nY9asZhQtaizXnAFZ+8jmzR1p2rQ0Rkaf0iGpGZUr29CmTVk6daqAvr5E4YJ++rQXlSppL3Ap62gV\nzNixdZFI0ChpQRvs3/+MgQMPce6cN5GRSbRt64SBgT5SqeD0aS8ePw6mXLliGp8jdeqU4PXrKJ49\nC6VUKUv27RugtbiHKg4deqHgvn39OpL581srdXNmugc1cUMePvyCy5d95O/DwhL54YfWCmN69txL\nfLys9WhERBL29uZqxyQXLlwIsFDtDeL/wMevLfr1q87du4GcPetNaGiCRoU0YWEJXLvmw6RJpzh/\n/jVeXhG5jv/yy4a8eDGFS5dG8vDhl/Irv42NGb17V6FXryq0beukUYm7sbEBnp4RbNvmxo8/XiEj\nQzFAnT1gbW6e+SQhASRYW5thZWUi75okQ2jka8w8WdSR+1WH168jSEhI1cpcBcHLK4IjR17g7x8L\nyPZ1ptHw8Ahj06YHTJx4kk2bHhQ4KJeV775rxfDhdZgxo6nW5lSXwYNrUqNGcapUsWHGjCaYmBgi\nhGDYsCP07LmPQYMO06bNDpKS0nKd59KlNwrv9fX12LOnH6mp8/Dzm0njxqVULCk7Zs+cUR3jUJcq\nVWzI6i2pUsVWq7Ic7dqVfV/lK/PKCCGYOfMs8fEfCrOyVmgDhZ419CndXmnljj8lJT3H3VYmiYlp\nLFlyg/37n1KsmKlamieZWFmZcPPmu/d59eqnAIKslH3DhvscPuxBUFA8Y8bUzdNFY2trRrlyxXIc\nAMWLmzNhQkNq1SqOlZVmAZ6MDCmpqRkULWrC9OlNFIx4+fLFOHLEg5SUDOztzVm4sC3v3kUTEBCP\nqaks6Fulii3Fi5tx4cIb9PQkLF/eUePm9NogPT2Dhw+D2Lr1EV5ekZiaGmBvb5FrIHrbtke0abOD\nJUtuYmNjSsOGBXO1ZHLhgjctW25j377nbNnyiDZtyii4MWxsTNm2zY3Ll32oXbsE/ftXz2U2zcj8\n/RQvxh+H5OR0unSpwOzZzYiPT6VoUVOCg+MZM+a4fExgYBytW5dRWr3r4RHGhQtvmD37PHZ2ZhQr\nZkqRIh+C23p6evLfUwhBSkqGgmvuxg1f/vrrMZs3P0JPT0KlStb5NpblyhXDxsYUf/9Yata0Y9eu\nvlp1MTk6WsmfZmNiUoiJSeGPP3oq6GRVrFiMY8c8SU+X0qpVGX77rYvaT0v5ueP/zxRwZWRIGTPm\nGLt3P6FoURP27+9P584VFcb06rWPEydkKYJGRvrcvz+e2rVLqDV/errMz25goK+y/25u/PHHfUJD\nE4iISGLNmm4aLastMtvOpaVlIAQ53CqZYmtpaRksXXqL1q0dqVmzOAEBcXzzzYd2cBkZUoSQua6K\nFjWR+/E/JsePezJgwEGKFDHm5MkhNG2q+rE4MDAOJ6fV8mIhPT0JXl5TNdJLyUpCQip37vhTvLg5\n3357kTNnPqQdDhxYnYMHByqMX7bsJpUqWRMQEMf06QVrjvMpExOTjJ3dCtLSPjw93r07Tulde1pa\nBqNHH2Xv3md8/31LFi/uoHTOy5ffMHDgYSIjkxgypCa7dvVFX1+PxMQ0OnXahaurH8eOfU6vXoUj\nl6BNpFLB4cPPady4FIGBcTncc9HRSTx7FsrLlxE4OVnRsWN5tbLq/rMduNRhzZo78i5UUVHJjBzp\nQnCwYhu88+e95f9PTc3g6lUftQ1/1rS/TKMvhGD//mcMGVIrz+VHjaqLqalhno++hUlmHreqOwk7\nO3M6dSrPzz9fx9XVD4kE5sxpoeCPB9mB9uefDzh82AMTE1eqoaIAABViSURBVAP696/GmDH1lM4Z\nG5tCWFgCZcpYaVX7p2ZNO/r3r0ZCQhp16tjnOjYiIlGhQlQqFYSHJ+bL8MfEJNOq1fb37S9FjuOn\nSJGcLsBvv21BamrGJxO/EGo04ckPVlYmbN7ckwkTTpKams6337ZU6aoxNNSnQoVibNvWi7i4FKVj\nAEaNOkpkpCyded++Z3TvXpHhw+tgampAo0YlGTeunsbf5fZtP5o2LZ3vfRAbm4y5uZHG2kd6ehIG\nDZL1xFbWZjUqKpk+fQ4QESH7vgsWtMbZuV2+tjHPbSmUWT8y16/78O23l7J8IoiOTsnRmahmTcW2\nZjVq5P9O9erVt8yYcYb586+wePG1PHPtM2VUs8qpforo6cny3J2d21Czpl0Ooy8bI2HgwBrcuePP\n/fsB9O+vXHr34sU3lCr1GxUrrqNZs605WhUWhOLFzdm7tz8uLp+r7ECVSbVqdrRs+eGJoEmTUtSt\nq6hg+uZNFGPGHGPYsCO4uwepnOvgwefynscgwcsrkgoVigGCqlVtWLgw54kqkUg4csQDD49wtb9f\nYXLw4PNCm3v06HrEx3/PyJH18sz2+fHHNowZU4/JkxupHBMdnaLy/erV3Rgzph49eijv+pad+PgU\nTp9+hbPzVTZtesiDB4FqLZdJSko63bvvwcpqGSVK/Iqr6zuNls+LAweeyY0+wMaNBVfhVMV/wvBv\n2eKmcEcHEqZPz9nO8PDhgXz2WWUaNXJg48YedOhQPt/rbN26LG/eROPtHUXduiUVKm//7QwaVJMF\nC9qyalVXlWOiopJYtaozNWoUp23bHXzzzQXS0hTTUGfMOCPPVHj4MIiNG+9rbRstLIyRSCTo6+vl\nGWg3MNDj/PkRbNvWi61be3H58iiFu++kpDTatdvBjh3u7N37jPbt/yI0NF7pXNnv2g0N9fj++5Y4\nORVh1qymOZQT09IymD//Mt98c4F+/Q6wZcvDPC9UhUVERCLLl9/iu+8u8cMPl/KsMs4PUVFJDB36\nN3/99ZiaNTdy7ZqPyrGZT9G5FdHNmPGhv3apUpYMGCCLkWQ9t9UNxFpYGBMZmcT582/4/fd7lCql\nmRTJ5s0P37v1JEREJDFhwimNls+L7HEFG5vCsymfrKvn1q13pKZm0KpVmTyrK7NrlLdo4cgvv3TK\nMa5s2WIcPz5EK9snS6ErzezZzbR6J/spkCkRndvTScWKNmzd6iaXgXZzC8HMzBBn57byMVm10ZW9\n/5iYmhqqdEf5+kbz7l2s/H1UVDIvXoQrFZgbPLgme/Y85cKFNxgb67NpU09evozA1zeOmzf9GTu2\nvsJ4Q0N9vvmmBevX36dIEeN8iwVqAxsbM0qXtuTt22g8PSNYtEh1xkx+KVbMlNGj6xAcHE+ZMla0\naVMWkF1c8/O0u2hRe9q1K0dQUBydO1fQSPQPcjbOsbWVJS54e0cq1aBKTk4jJiaFEiUsSE/PwNMz\ngho1ZJ6CwpZHHjWqLhcvvuXgwefY21uwbVsvrc6flU8yuDt27DG2b5c1UejWrSInTw7NNXMhMjKR\nPn0OcPPmO2rWLM6JE0NwcipYr1YdedOt2x7Onv0Q2OzduwpHjw6Wv9+79wkjRx4lI0Pg5GTF7dtf\n5FvwrTBITk7HxMSAhIRUKlRYK292YWFhiKfnVJU1F0IIfH1j3nccM2HdurtYW5vi5RWh1Cfr6RmO\nnp6EkJCEfGVjaZP9+59iY2PGnTt+zJ/ftlDW4e0dSfnyxXjzJooKFax5+DCQN2+i/hHZiI0b7zNp\n0gdXUmasJSNDikQiUbAr8fGpnDv3Gj+/GFq0KIO7ezCurn5MndqYWrVKEBgYS5MmWwgNTQQEq1Z1\n4auvmml9m9PSZBlMmkjM8GnZco0QQgjx5k2kQsk2OIubN33VKovOXoJ//bqPWsvdu+cvQkPj1Rqr\n4wO//KJYXr9mze0cY169CheXLr0R0dHa6fsaHZ0kIiIS1R7/9m2UaNLkT1Gs2DIxapSLSEvLEImJ\nqaJ79z0CnIWDw6/iwYMA8fRpiOjTZ7/o1m232sdbJpla7Nk12ePiksXt237i3btojeYrTGJiksTs\n2WfF558fFCdOvCz09W3ceE+ULbtKODisFAsWXBZpaemFvk4hZDr5M2eeFcWKLRNffnlcPHoUmOcy\nO3e6CTOzn4W+/kIxa9ZZuXTKrl2PFebdv/+pcHV9l+tcUqlUrFhxU3Tvvlv88MMlkZJSeN+b/4Ie\nf2BgLKVLryKrG/TRoy/V7j8LMjGlBw8C+fbbiyxb1oHate2VVtqlp2ewf/8zDh16QYkS5nTvXone\nvat+lO5b/wWEEGzc+IC7dwNo0cKRL78sPDeGVColJCSBY8c8SUvLYMCA6nnm7gN06rSLixc/FAmt\nWtWFjAypQgVy7dolcHEZRLlyxbT224eGJtCixVZev47CyEifffv606+f8iB4VkQhZdxk0rv3fo4f\nl1Wp6ulJuHFjDM2bq1chmp6eobGonRCCypXXEhaWhLv7BOztLTROhc7E1dWPceOOExOTwpw5zZg5\nM/e77Z073Rk9+ih9+1bl778H5zoWZHfaLVtuA+DPJTWZt8YHOzszJk1qSMOGmrnFNmy4z5QpHwQJ\nZ81qysqVXTSaQ13+Ez13HRyKsHhxe2TfRTB9emONjD7IWrBdufIWL69IXFxeytUhs2NgoE/jxqU5\ne9abU6e8aNbM8aMa/ZSUdJKTFdM7xT/ccF4TJBIJkyc3YufOPpQoYc60aafZsuWRVmSrs6Onp8eV\nK2+ZPfs833xzkYsX36j1W/n5xSi89/ePzZGB9eRJCBUqrGP8+BNa295Nmx7w+rWsnWBqagZz517M\nc5nw8IRCkaLOStZgq1QquHFDPc2ZzB4RmhIRIZMnGDKkBs7Ol1mw4Eq+jg+pVNC79348PMIJDIxj\n1qzzCsHpqKgkkpMV25V6eUUyfnx9Xr4MZ/fuxznOtewkJqZxcHNLdv5UlvPf/8ja78qzfmUbqlbV\nvNL8zh1/hfe3b/urGAmvz57VeP6C8skZfpCVooeEzCEgYJZCsVNISLxKudus6OlJKFPGig0bulOh\nQrFcxbjS0jJYvLg9w4bV+ujNFS5c8JYLnmWyc2fhNojOJCwsgc8/P0TDhptZvPh6geY6cuQFffrs\nZ/36+4wff4Kffy7YfKr47LMq1K9vT6NGDnKxr7zIKl5mZKTPwIHVGTmyDtbWmQkBH56Ut2514+5d\n1SeoJshyvEWW97lfpA4efM6kSadZtOgGq1bdJjVVs57L6lK/ftabKJHtvXJu3XpHr177+eUXVyZO\nPEFERKLa67O1Nads2WJs2+bOX389Y8sWd7755gIpKZp9v4SEVMLDFdf77l0MUqlg2LC/sbZeTrFi\nv3D48IdU1VmzmnHjhh8eHhEEBsaxa9cT/P1jsk8tx8rKBGs7K67P+56Eexd5ffI4xpaW+ZLLbtZM\nsYdG8+Y5e2pEenvzaMsWTk6YwKMtW4h4/Wn2HyhscvWXnTzpKSZPPiEGDz4oDh9+nqevMDU1XeFV\nFVmlfwtLBljZOpcvvyGKFl0qihZdKpYvvynCwuLE/PmXhJPTKjFhwglx7Zp68YncyB7zyCr32rPn\nXgX//O7dMj9mYGCs+OOP++Lvv18oLPv2yhWV6xkz5qjCXE2a/FngbVdGZGSiSE1NF2lpGSIqSnnM\nQCqViuS4OIXPjhx5LpYuva7g5333LlqsXn1bwAJRinEax5Py4s2bSGFltUSAszAzWyxOnXqV6/jk\n5DRRrdp6IZE4Cx+fKK1sgzJCQuLEyJEuon37nWLbtkdqLzd6tIswMFgoTp3yzMc648WgQYdE0aJL\nhaXlEhEQkD9Z8R499sh/J3v7X0VwcJz4++8XCseepeUSucx5amq6mDr1lGjXbrsAmbR68eLL84y5\nuIwdK87Oni0e796dr+3M5LffXEXPnnvE/PmXVdqh0zNmCGcQp2fMyCHProqIiESxYcM9sXOnm0hL\ny8iXj78gDASeAxlA/VzGdQVeAl7At7mMy/EFT570lAfMfH2jhb39r8LE5Gfx9GmIfMyzZyFi1y53\n4eERlo+f5p8jNTVdVKiwWpQvv1qkpMh07U+e9BTgLPr02Vfg+aVSqVi8+JrCZxMnnpD/38lplcIJ\n8913F0VQUJxwcFgp/2zKlFMiOSZGPDt4UGxt3lzcXb9e+N25k2Ndy5ffVJhr1CiXAm+/pry7dUsE\nP30qfnN0FM4gtrZsKZJicjcwsUFB4qtei8SXlBAVGCoGtP9NZGSod/LlxvXrPqJ0aVlgsEOHnWo1\n/Y6PTxG//npLnDjhKe7fV65V/+hRoJgy5aSYP/+SiInRTrBcXXbvfiwCAmLE+fOvNV42MTFVJCam\nisuX3wgvr3Dx5InqJuy5kZKSLn7//Z5YsuS68PWVGe+dO90Ujj19/YVyI5tpSC0tlyiMWb06ZxJC\nVjJvHLLfQBQGt1asEM8OHBC3VqxQa3xMTLKoVGmt/Lt89tnej67H/xToC2zKZYw+sB7oCAQA94Hj\ngEduE8fHp7B371O2bHHj1i1fOnQoT716JRk4sBpWViZyJbuzZ73o1Ws/aWlSjI31OXlyCB07VijA\nV/p4hIcncuXKaI4e9WDKlFP06FGZ8PBEVq/uUuCOWxs2HOLUqRRu3vTD2zuKcePqMXfuJa5f98XT\nM4JffulAly4V2Lz5ESCrSejYsTwnTngSGBgnn2fz5oesW9eN9KQk/FxdEUJQZ8SIHOubObMZ/v6x\nXLjwhjp1SrB6tfaCWBkZ0lxL40OePSPi5UtuLl2KND2dWD+Z39fv5k1cf/2V9j/9pHJZi+LF6VY5\nhtuEMLN5MGP//kJp2vDVq1dp27at2tvcqpUTrVuXwd8/lq++aqKWhK+5uRGzZzdX+Xdv70hat94h\nL4i7ccOPK1dGqb1NBWXYMJnLTB1Z8exk5u+3a1cOkO1PUE8qJStGRvo5qnz79KlG9equvHghkxT/\n+uvmcmmQzBiQnZ05cXEfVFyzS5k/fhzMiROvcHKykrl8LWS1ApmvhUmz2bORSCRqx/auXfPByytS\n/j5Te0xTCmL4X6oxpjHwGvB5/34/0Js8DL+FhTFOTkW5fz8QPT0J8+a1xdhYn7VruwPIJYU3bHgg\nF4RKSclgw4YH/xrDX7KkJWvX3mHGjHMAbNnijoGBhPR0ga2tGYMG1aB6dUWJCT+/GBwd81YUDQ19\njqNjA5KTfejVqwrNmpWhe/dK+PvH0bKlIy4unqxf353y5Yvh7R1Jnz5Vad++XA7NFFtbMyQSCUaW\nlnRds4YwDw+Mi+Q88Q0M9PIUnjt27CVz5pwHZKqeffvmneGSkpLOli2PmDKlscoxRZ2cODtjBkGP\nHmHpoKi4mRyduwSyRE+PomXL0mHpUhLDwzG1Um7UNDX8QggWLWpHuXLFePs2Su3lcuPmzXdyoy/b\nJh95HcK/jaz78/XrCGbOPEd0dDLTpjWWa9moS5Eixty58wVXrvhgbW1CixY5+xLs3t2XAQMOEhKS\nwMiRtRk8+MM6Hj8OpmnTLSQny4oLHz0KyrViXdtkXpzUTSqxt1e8GJmZGZCofshFTmEfNaWArHXh\n/oBa8oQZGVJ27OjNkyfBmJjoK1TfZd4BZteCL1pU8yDMP0n23r3p6bKrfnh4IuvW3WPjxp6ArLDE\nzS2Is2dfM3duS0xNDfJMq2vbtiyLFrXn4UOZHkm7duWoVs2OMWOOERubQmhoPGvWdFNowN27d1Wm\nTm3Epk0PsbU1Y//+AQBU6dULPX19hFSar3TDkJB4Bg8+Is+6GDz4MG/ezMi1wcaZM16sW3ePJ09C\n8PGJ5scf22BpmfP3Nba0xK56dSp17074y5e4bd0KQmBkaUn9L77Ic9vqjxuHgbEx6Snaq76WSCRy\nAbj8KoBmR5ZZIsjM2itXrui/0uhnp0ePfbx6JetPcfu2P5Ur21K3bu6ie9mxtDSmVy/V8uDNmjkS\nEDBb6dPjsWOecqMPcODA849q+DWlUaNSLF3agZ9/vo65uSFbtvSiV695Gs+T15FzAVD2K3wPqJP7\nlu+gQ7dulfJ8BFq8uD2PHgXx/HkYtWoV5+ef2+d3df8IlSvbKHTmyXpim5l9yHW+etWHkSNdiIpK\nQk9PwoIFbfKce/BgmWJo164yAatMlcSGDUsSFBTP9OlNFIx+JuvWdWfNmq4KF1o9/fePzvlsThEY\nGKeQapeaKiUwMC5Xw9+tWyXWrr1LcHA8n39eU6nRz6TzypUYGBmRkZZGw4kTifD0xLFFC4o6OeW5\nbQbGxgqvnypNmpRm+/berF9/H2trU9as+XSNk7okJqbKjT5ARobg6dMQjQ2/uihzGZYpo3gMOjoW\nbrcvbTB3bku+/bZFgVLPtZG0fgWYDTxS8remgDOyAC/Ad4AU+EXJ2NfAv8NPo0OHDh2fDt5AxTxH\naZkrgKqSTQNkG1UWMALcgbyduzp06NCh45OkLzL/fRIQDJx5/7kDkFWvtBvgieyO/ruPuYE6dOjQ\noUOHDh06dOj4h9B28df/O9bIAvGvgPOAKk1qH+AJ4Abc+yhb9u9CneNt7fu/PwaUC/zryCSv/dkW\niEF2PLoBmqen/H+wDQhBVjulin/FcVkVqIwsPqDK8Osjcw+VBQzRxQdyYznwzfv/fwssUzHuLbKL\nhI6cqHO8dQcyJRebAHc+1sb9C1Fnf7ZFVtCpI3daITPmqgy/xsflPyXS9hLZ3WluZC3+SuND8ZeO\nnPQCdr7//06gTy5jdZrTylHneMu6n+8ie7LSvAT1/wN1z1/d8Zg3N4DcKgE1Pi4/SXXO9ygr/tJ+\nr7j/BiWQPQry/lXVjy6Ai8ADYPxH2K5/E+ocb8rG5JRd1AHq7U8BNEfmnjgNVP84m/afQ+PjsjBL\n//6x4q//KKr25w/Z3ucm2tQCCALs3s/3EtndhA71j7fsd6i641Q56uyXR4AjkIgs++8oMhewDs3R\n6LgsTMOfs9u5ZgQgOygycUR2Jft/Jbf9GYLsohAMlARCVYwLev8aBrggexzXGX4Z6hxv2ceUfv+Z\njpyosz/jsvz/DLABWQwqEh2a8K87LnXFX9phOR+yJuaiPLhrBmR2OjcHbgGdC3/T/jWoc7xlDaI1\nRRfczQ119mcJPtypNuaDmKOOnJRFveDuJ31c6oq/tIs1Mt999nTOrPuzPLKTzx14hm5/KkPZ8Tbh\n/b9M1r//+2NyT0XWkff+nILsWHQHXJEZLR052QcEAqnI7OZYdMelDh06dOjQoUOHDh06dOjQoUOH\nDh06dOjQoUOHDh06dOjQoUOHDh06dOjQoUOHDh06dOjQoUOHDh06dHya/A/PreP4M3PIqQAAAABJ\nRU5ErkJggg==\n", + "text": [ + "" + ] + } + ], + "prompt_number": 13 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "def classify_forest (x,y):\n", + " if rforest.predict([x,y]) == 0:\n", + " print(\"must be blue...\")\n", + " else:\n", + " print(\"must be red...\")" + ], + "language": "python", + "metadata": {}, + "outputs": [], + "prompt_number": 14 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "print('what will be your x?')\n", + "x = raw_input()\n", + "print('and y?')\n", + "y = raw_input()\n", + "\n", + "classify_forest (x,y)" + ], + "language": "python", + "metadata": {}, + "outputs": [ + { + "output_type": "stream", + "stream": "stdout", + "text": [ + "what will be your x?\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "stream": "stdout", + "text": [ + "6\n" + ] + }, + { + "output_type": "stream", + "stream": "stdout", + "text": [ + "and y?\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "stream": "stdout", + "text": [ + "5\n" + ] + }, + { + "output_type": "stream", + "stream": "stdout", + "text": [ + "must be blue...\n" + ] + } + ], + "prompt_number": 16 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "print 'Which model is more accurate according to scores?'\n", + "if rforest.score(Xtestset, Ytestset) >= knearest.score(Xtestset, Ytestset):\n", + " print 'Random forest is better'\n", + "else:\n", + " print 'I would prefer KNN'" + ], + "language": "python", + "metadata": {}, + "outputs": [ + { + "output_type": "stream", + "stream": "stdout", + "text": [ + "Which model is more accurate according to scores?\n", + "I would prefer KNN\n" + ] + } + ], + "prompt_number": 21 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "import numpy\n", + "wrong_rforest = []\n", + "for X, Y in zip(Xtestset.tolist(), Ytestset.tolist()):\n", + " if rforest.predict(X) != Y:\n", + " wrong_rforest.append(X)\n", + "wrong_rforest = numpy.array(wrong_rforest)\n", + "print (\"Random Forest, number of inaccurately classified dots:\")\n", + "print (len(wrong_rforest))\n", + "\n", + "wrong_knn = []\n", + "for X, Y in zip (Xtestset.tolist(), Ytestset.tolist()):\n", + " if knearest.predict(X) != Y:\n", + " wrong_knn.append(X)\n", + "wrong_knn = numpy.array(wrong_knn)\n", + "print (\"KNN, number of inaccurately classified dots:\")\n", + "print (len(wrong_knn))\n", + "\n", + "if len(wrong_knn) > len(wrong_rforest):\n", + " print(\"RF seems to be better in this case\")\n", + "else:\n", + " print ('KNN classifies more accurately')" + ], + "language": "python", + "metadata": {}, + "outputs": [ + { + "output_type": "stream", + "stream": "stdout", + "text": [ + "Random Forest, number of inaccurately classified dots:\n", + "52\n", + "KNN, number of inaccurately classified dots:" + ] + }, + { + "output_type": "stream", + "stream": "stdout", + "text": [ + "\n", + "48\n", + "KNN classifies more accurately\n" + ] + } + ], + "prompt_number": 33 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [], + "language": "python", + "metadata": {}, + "outputs": [] + } + ], + "metadata": {} + } + ] +} \ No newline at end of file diff --git a/kolchanova/my8.py b/kolchanova/my8.py new file mode 100644 index 0000000..6aba6df --- /dev/null +++ b/kolchanova/my8.py @@ -0,0 +1,19 @@ +#!/usr/bin/python + +with open('nextchoose.in') as infile: + number, n = [int(x) for x in infile.readline().split()] + combination = [int(x) for x in infile.readline().split()] + +def mk_choice(number, n, combination): + for x in range(n): + if combination [ n - x - 1 ] < number - x: + combination [ n - x -1 ] += 1 + for y in range ( n - x, n ): + combination [y] = combination [ y - 1 ] + 1 + return combination + return [-1] + +result = mk_choice (number, n, combination) +#print result +with open ('nextchoose.out', 'w') as outfile: + outfile.write(" ".join(str(x) for x in result)) \ No newline at end of file diff --git a/kolchanova/python3.ipynb b/kolchanova/python3.ipynb new file mode 100644 index 0000000..ae7ec23 --- /dev/null +++ b/kolchanova/python3.ipynb @@ -0,0 +1,375 @@ +{ + "metadata": { + "name": "", + "signature": "sha256:b6bc5b1ee763aa8c5d4520e4d96ea1d850639a84a4a9ec46e31991ace23aa207" + }, + "nbformat": 3, + "nbformat_minor": 0, + "worksheets": [ + { + "cells": [ + { + "cell_type": "code", + "collapsed": false, + "input": [ + "import numpy, scipy\n", + "from math import sqrt\n", + "from scipy import stats\n", + "import matplotlib.pyplot as plot\n", + "import random\n", + "%matplotlib inline" + ], + "language": "python", + "metadata": {}, + "outputs": [], + "prompt_number": 2 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "x = numpy.linspace(0, 2, 1000)\n", + "y = 2 * x + 1\n", + "noise = scipy.randn(y.size)\n", + "y_noise = y + noise\n", + "\n", + "X_full = numpy.array(x).T\n", + "y_full = numpy.array(y_noise).T\n", + "print (X_full.shape)\n", + "print (y_full.shape)" + ], + "language": "python", + "metadata": {}, + "outputs": [ + { + "output_type": "stream", + "stream": "stdout", + "text": [ + "(1000,)\n", + "(1000,)\n" + ] + } + ], + "prompt_number": 95 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "TRAIN_SIZE = 800\n", + "from sklearn.utils import shuffle\n", + "x, y_noise = shuffle(x, y_noise, random_state=1)\n", + "x_train = x[:TRAIN_SIZE]\n", + "y_train = y_noise[:TRAIN_SIZE]\n", + "x_test = x[TRAIN_SIZE:]\n", + "y_test = y_noise[TRAIN_SIZE:]" + ], + "language": "python", + "metadata": {}, + "outputs": [], + "prompt_number": 64 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "X_train = numpy.asmatrix(x_train).T\n", + "Y_train = numpy.asmatrix(y_train).T\n", + "Train_set = numpy.hstack((X_train, Y_train))\n", + "print (Train_set.shape)\n", + "\n", + "X_test = numpy.asmatrix(x_test).T\n", + "Y_test = numpy.asmatrix(y_test).T\n", + "Test_set = numpy.hstack((X_test,Y_test))\n", + "print (Test_set.shape)" + ], + "language": "python", + "metadata": {}, + "outputs": [ + { + "output_type": "stream", + "stream": "stdout", + "text": [ + "(800, 2)\n", + "(200, 2)\n" + ] + } + ], + "prompt_number": 65 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "betas = scipy.polyfit(x, y_noise, 1)[::-1] # 1 is order (max degree of x)\n", + "print (\"betas:\", betas)\n", + "print (\"original model: y = 2 * x + 1\")\n", + "print (\"proposed model: y = 1.99725967*x + 1.03406303\")" + ], + "language": "python", + "metadata": {}, + "outputs": [ + { + "output_type": "stream", + "stream": "stdout", + "text": [ + "betas: [ 1.00597503 2.01020888]\n", + "original model: y = 2 * x + 1\n", + "proposed model: y = 1.99725967*x + 1.03406303\n" + ] + } + ], + "prompt_number": 66 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "_ = plot.plot(x, y_noise, \"r.\", label = \"full set\")\n", + "_ = plot.plot(x, betas[0] + betas[1]*x, color=\"blue\", label = \"model\")\n", + "plot.xlabel('X')\n", + "plot.ylabel('Y')\n", + "_ = plot.legend(bbox_to_anchor = (1.05, 1), loc = 2, borderaxespad = 0)" + ], + "language": "python", + "metadata": {}, + "outputs": [ + { + "metadata": {}, + "output_type": "display_data", + "png": "iVBORw0KGgoAAAANSUhEUgAAAeEAAAEPCAYAAAB1BHz8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n", + "text": [ + "" + ] + } + ], + "prompt_number": 75 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "from sklearn import linear_model\n", + "regr = linear_model.LinearRegression()\n", + "regr.fit(X_train, Y_train)\n", + "print(regr.coef_)\n", + "print(regr.intercept_)" + ], + "language": "python", + "metadata": {}, + "outputs": [ + { + "output_type": "stream", + "stream": "stdout", + "text": [ + "[[ 2.03151967]]\n", + "[ 1.01171718]\n" + ] + } + ], + "prompt_number": 68 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "regr.fit(X_test, Y_test)\n", + "print (regr.coef_)\n", + "print (regr.intercept_)" + ], + "language": "python", + "metadata": {}, + "outputs": [ + { + "output_type": "stream", + "stream": "stdout", + "text": [ + "[[ 1.92562103]]\n", + "[ 0.98433975]\n" + ] + } + ], + "prompt_number": 69 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "slope, intercept, r_value, p_value, std_err = stats.linregress(x,y_noise)\n", + "print (\"Full set r-squared (determination coefficient):\", r_value**2)\n", + "print (\"standard error = \", std_err)\n", + "\n", + "train_set_error = sqrt(numpy.sum(((slope*x_train+intercept) - y_train)**2)/len(y_train))\n", + "print ('Residual sum of squares, train:', train_set_error)\n", + "\n", + "test_set_error = sqrt(numpy.sum(((slope*x_test+intercept) - y_test)**2)/len(y_test))\n", + "print ('Residual sum of squares, test:', test_set_error)" + ], + "language": "python", + "metadata": {}, + "outputs": [ + { + "output_type": "stream", + "stream": "stdout", + "text": [ + "Full set r-squared (determination coefficient): 0.57897951659\n", + "standard error = 0.054262016436\n", + "Residual sum of squares, train: 0.9613567770303205\n", + "Residual sum of squares, test: 1.1002000473349025\n" + ] + } + ], + "prompt_number": 70 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "from scipy.stats import pearsonr\n", + "print (\"Correlation coefficient:\")\n", + "pearsonr(x, y_noise) #correlation coefficient x1" + ], + "language": "python", + "metadata": {}, + "outputs": [ + { + "output_type": "stream", + "stream": "stdout", + "text": [ + "Correlation coefficient:\n" + ] + }, + { + "metadata": {}, + "output_type": "pyout", + "prompt_number": 71, + "text": [ + "(0.76090703544549931, 1.1166799486254262e-189)" + ] + } + ], + "prompt_number": 71 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "print (\"Correlation coefficient array:\")\n", + "numpy.corrcoef(x,y_noise) #correlation coefficient x2" + ], + "language": "python", + "metadata": {}, + "outputs": [ + { + "output_type": "stream", + "stream": "stdout", + "text": [ + "Correlation coefficient array:\n" + ] + }, + { + "metadata": {}, + "output_type": "pyout", + "prompt_number": 72, + "text": [ + "array([[ 1. , 0.76090704],\n", + " [ 0.76090704, 1. ]])" + ] + } + ], + "prompt_number": 72 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "_ = plot.plot(x_train, y_train, \"r.\", label = \"train\")\n", + "_ = plot.plot(x_test,y_test, \"b.\", label = \"test\")\n", + "_ = plot.plot(x, betas[0] + betas[1]*x, color=\"green\", linewidth = 3, label = \"model\")\n", + "plot.xlabel('X')\n", + "plot.ylabel('Y')\n", + "_ = plot.legend(bbox_to_anchor = (1.05, 1), loc = 2, borderaxespad = 0)" + ], + "language": "python", + "metadata": {}, + "outputs": [ + { + "metadata": {}, + "output_type": "display_data", + "png": "iVBORw0KGgoAAAANSUhEUgAAAdsAAAEPCAYAAAAK1W4NAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n", + "text": [ + "" + ] + } + ], + "prompt_number": 74 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "model_function = lambda x: betas[0] + betas[1]*x\n", + "ok = []\n", + "too_far = []\n", + "for tx, ty in numpy.vstack((X_full, y_full)).T.tolist():\n", + " if (model_function(tx) - ty)**2 < std_err:\n", + " ok.append([tx, ty])\n", + " else:\n", + " too_far.append([tx, ty])" + ], + "language": "python", + "metadata": {}, + "outputs": [], + "prompt_number": 99 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "ok_points = numpy.array(ok)\n", + "too_far = numpy.array(too_far)\n", + "print (ok_points.shape)\n", + "print (too_far.shape)" + ], + "language": "python", + "metadata": {}, + "outputs": [ + { + "output_type": "stream", + "stream": "stdout", + "text": [ + "(174, 2)\n", + "(826, 2)\n" + ] + } + ], + "prompt_number": 104 + }, + { + "cell_type": "code", + "collapsed": false, + "input": [ + "_ = plot.plot(x, betas[0] + betas[1]*x, color=\"blue\", linewidth = 1, label = \"model\")\n", + "_ = plot.plot(ok_points[:, 0], ok_points[:, 1], \"g.\", label = \"ok points (dist. < std)\")\n", + "_ = plot.plot(too_far[:, 0], too_far[:, 1], \"r.\", label = \"points, that are too far (dist.>std)\")\n", + "_ = plot.legend(bbox_to_anchor = (1.05, 1),loc = \"best\",borderaxespad = 0)" + ], + "language": "python", + "metadata": {}, + "outputs": [ + { + "metadata": {}, + "output_type": "display_data", + "png": "iVBORw0KGgoAAAANSUhEUgAAAnUAAAEACAYAAADV8IC7AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nB3kKz8S3Ej8hAAAAAElFTkSuQmCC\n", + "text": [ + "" + ] + } + ], + "prompt_number": 108 + } + ], + "metadata": {} + } + ] +} \ No newline at end of file diff --git a/kolchanova/python3.py b/kolchanova/python3.py new file mode 100644 index 0000000..2fda833 --- /dev/null +++ b/kolchanova/python3.py @@ -0,0 +1,101 @@ +#!/usr/bin/python + +import numpy, scipy +from math import sqrt +from scipy import stats +import matplotlib.pyplot as plot +import random +%matplotlib inline + +x = numpy.linspace(0, 2, 1000) +y = 2 * x + 1 +noise = scipy.randn(y.size) +y_noise = y + noise + +X_full = numpy.array(x).T +y_full = numpy.array(y_noise).T +print (X_full.shape) +print (y_full.shape) + +TRAIN_SIZE = 800 +from sklearn.utils import shuffle +x, y_noise = shuffle(x, y_noise, random_state=1) +x_train = x[:TRAIN_SIZE] +y_train = y_noise[:TRAIN_SIZE] +x_test = x[TRAIN_SIZE:] +y_test = y_noise[TRAIN_SIZE:] + +X_train = numpy.asmatrix(x_train).T +Y_train = numpy.asmatrix(y_train).T +Train_set = numpy.hstack((X_train, Y_train)) +print (Train_set.shape) + +X_test = numpy.asmatrix(x_test).T +Y_test = numpy.asmatrix(y_test).T +Test_set = numpy.hstack((X_test,Y_test)) +print (Test_set.shape) + +betas = scipy.polyfit(x, y_noise, 1)[::-1] # 1 is order (max degree of x) +print ("betas:", betas) +print ("original model: y = 2 * x + 1") +print ("proposed model: y = 1.99725967*x + 1.03406303") + +_ = plot.plot(x, y_noise, "r.", label = "full set") +_ = plot.plot(x, betas[0] + betas[1]*x, color="blue", label = "model") +plot.xlabel('X') +plot.ylabel('Y') +_ = plot.legend(bbox_to_anchor = (1.05, 1), loc = 2, borderaxespad = 0) + +from sklearn import linear_model +regr = linear_model.LinearRegression() +regr.fit(X_train, Y_train) +print(regr.coef_) +print(regr.intercept_) + +regr.fit(X_test, Y_test) +print (regr.coef_) +print (regr.intercept_) + +slope, intercept, r_value, p_value, std_err = stats.linregress(x,y_noise) +print ("Full set r-squared (determination coefficient):", r_value**2) +print ("standard error = ", std_err) + +train_set_error = sqrt(numpy.sum(((slope*x_train+intercept) - y_train)**2)/len(y_train)) +print ('Residual sum of squares, train:', train_set_error) + +test_set_error = sqrt(numpy.sum(((slope*x_test+intercept) - y_test)**2)/len(y_test)) +print ('Residual sum of squares, test:', test_set_error) + +from scipy.stats import pearsonr +print ("Correlation coefficient:") +pearsonr(x, y_noise) #correlation coefficient x1 + +print ("Correlation coefficient array:") +numpy.corrcoef(x,y_noise) #correlation coefficient x2 + +_ = plot.plot(x_train, y_train, "r.", label = "train") +_ = plot.plot(x_test,y_test, "b.", label = "test") +_ = plot.plot(x, betas[0] + betas[1]*x, color="green", linewidth = 3, label = "model") +plot.xlabel('X') +plot.ylabel('Y') +_ = plot.legend(bbox_to_anchor = (1.05, 1), loc = 2, borderaxespad = 0) + +model_function = lambda x: betas[0] + betas[1]*x +ok = [] +too_far = [] +for tx, ty in numpy.vstack((X_full, y_full)).T.tolist(): + if (model_function(tx) - ty)**2 < std_err: + ok.append([tx, ty]) + else: + too_far.append([tx, ty]) + +ok_points = numpy.array(ok) +too_far = numpy.array(too_far) +print (ok_points.shape) +print (too_far.shape) + +_ = plot.plot(x, betas[0] + betas[1]*x, color="blue", linewidth = 1, label = "model") +_ = plot.plot(ok_points[:, 0], ok_points[:, 1], "g.", label = "ok points (dist. < std)") +_ = plot.plot(too_far[:, 0], too_far[:, 1], "r.", label = "points, that are too far (dist.>std)") +_ = plot.legend(bbox_to_anchor = (1.05, 1),loc = "best",borderaxespad = 0) + diff --git a/kolchanova/regression1_1.py b/kolchanova/regression1_1.py new file mode 100644 index 0000000..2c4ddb7 --- /dev/null +++ b/kolchanova/regression1_1.py @@ -0,0 +1,51 @@ +#!/usr/bin/python +import numpy + + +yz = [1, 3, 2] +xz = [1, 2, 3] + + +xmatrix = [float(x) for x in xz] +print numpy.matrix(xmatrix).shape +x_matrix = numpy.matrix(xmatrix) + +ymatrix = [ float(y) for y in yz] +print numpy.matrix(ymatrix).shape +y_matrix = numpy.matrix(ymatrix) + +from scipy import stats +import numpy as np +#linregress - This computes a least-squares regression for two sets of measurements. +slope, intercept, r_value, p_value, std_err = stats.linregress(xmatrix,ymatrix) +print "slope:", slope +print "intercept:", intercept + +print "standard error:", std_err +print "Coefficient of determination (R-squared):", r_value**2 +print "p-value:", p_value + +from scipy import polyfit +coefficients = polyfit(xz, yz, 1)[::-1] +print "coefficients (same as slope, intercept):", coefficients + + +#let us use the CHI-SQUARED TEST + +def Sum (y_matrix, yz, xz, coefficients): + for i in range (y_matrix.shape[0]): + a = (sum([yz[i] - coefficients[1]*xz[i] - coefficients[0]])**2)*6 + + return a +a = Sum(y_matrix, yz, xz, coefficients) + + +#count p-value + + +from scipy import stats +p_value = stats.chi2.cdf(a,3) +print "chi-squared, P-value:", p_value + + + diff --git a/kolchanova/regression2.py b/kolchanova/regression2.py new file mode 100644 index 0000000..9573877 --- /dev/null +++ b/kolchanova/regression2.py @@ -0,0 +1,51 @@ +#!/usr/bin/python +import scipy +import numpy + + +xfile = open ('x.txt', 'r') +yfile = open ('y.txt','r') + +xmatrix = [[float(x) for x in row.split()] for row in xfile] + +print numpy.matrix(xmatrix).shape +x_matrix = numpy.matrix(xmatrix) + + +ymatrix = [[ float(y) ] for y in yfile] + +print numpy.matrix(ymatrix).shape +y_matrix = numpy.matrix(ymatrix) + +add_column = numpy.matrix([1]*x_matrix.shape[0]).T +new_x = numpy.hstack((add_column, x_matrix)) + +#matrices multiplication + +step_one = numpy.dot (new_x.T, new_x) +step_two = numpy.linalg.pinv(step_one) +step_three = numpy.dot(step_two, new_x.T) + + +coeffs = numpy.dot(step_three, y_matrix) +errors = y_matrix - numpy.dot(new_x, coeffs) + +print coeffs +#the model is too complicated (multidimentional) +#are errors distributed normally? if yes, then the model is accurate + +import numpy as np +import numpy.ma as ma +from scipy.stats import mstats + +x = np.array(errors) + +z,pval = mstats.normaltest(x) #Tests whether a sample differs from a normal distribution. +#This function tests the null hypothesis that a sample comes from a normal distribution +print "Z-score:", z +print "P-value:", pval + +if(pval < 0.055): + print "Not normal distribution" +if (pval >= 0.055): + print "This seems to be a normal distribution! Our model is good" \ No newline at end of file diff --git a/kolchanova/regression3.py b/kolchanova/regression3.py new file mode 100755 index 0000000..40d5f53 --- /dev/null +++ b/kolchanova/regression3.py @@ -0,0 +1,57 @@ +#!/usr/bin/python + +#I chose to perform the classification task using the KNN method +import sys +import numpy as np +import scipy +from sklearn.utils import shuffle +import matplotlib.pyplot as plot +from sklearn.neighbors import KNeighborsClassifier + +blue = np.loadtxt("./blue.txt") +red = np.loadtxt("./red.txt") + +#Visualize data (this doesn't work outside ipython), but we get something +#like an ellipse.... (pic attached) + +#_ = plot.plot(blue[:, 0], blue[:, 1], "b.") +#_ = plot.plot(red[:, 0], red[:, 1], "r.") + + +#Create the classifying function + +red_points = np.hstack ((red, [[1]] * len (red) )) +blue_points = np.hstack ((blue, [[0]] * len (blue) )) +all_points = np.concatenate((red_points, blue_points), axis=0) + +x = all_points[:, :-1] +y = all_points[:, 2] +x, y = shuffle(x, y, random_state=1) + + +train_set=all_points.shape[0] *0.6 +Xtrainset = x[:train_set] +Ytrainset = y[:train_set] +Xtestset = x[train_set:] +Ytestset = y[train_set:] + +knearest = KNeighborsClassifier(n_neighbors=5) +knearest.fit(Xtrainset, Ytrainset) +print('Training set score:', knearest.score(Xtrainset, Ytrainset)) +print('Test set score:', knearest.score(Xtestset, Ytestset)) + +#Prediction function + +def classify (x,y): + if knearest.predict([x,y]) == 0: + print("must be blue...") + else: + print("must be red...") + +#propose some coordinates :) +print('what will be your x?') +x = raw_input() +print('and y?') +y = raw_input() + +classify (x,y)