From ed36bd1136a5920791ca413bab1b4ffed6333680 Mon Sep 17 00:00:00 2001 From: Tuan Vu Date: Sun, 3 Jul 2016 15:39:17 -0700 Subject: [PATCH] LSH Implement LSH with graphlab --- ...h-implementation_graphlab-checkpoint.ipynb | 1851 +++++++++++++++++ ...t-neighbors-lsh-implementation_blank.ipynb | 1409 +++++++++++++ ...eighbors-lsh-implementation_graphlab.ipynb | 1851 +++++++++++++++++ .../assigment/week2/week2-assignment2.pdf | Bin 0 -> 659725 bytes 4 files changed, 5111 insertions(+) create mode 100644 machine_learning/4_clustering_and_retrieval/assigment/week2/.ipynb_checkpoints/1_nearest-neighbors-lsh-implementation_graphlab-checkpoint.ipynb create mode 100644 machine_learning/4_clustering_and_retrieval/assigment/week2/1_nearest-neighbors-lsh-implementation_blank.ipynb create mode 100644 machine_learning/4_clustering_and_retrieval/assigment/week2/1_nearest-neighbors-lsh-implementation_graphlab.ipynb create mode 100644 machine_learning/4_clustering_and_retrieval/assigment/week2/week2-assignment2.pdf diff --git a/machine_learning/4_clustering_and_retrieval/assigment/week2/.ipynb_checkpoints/1_nearest-neighbors-lsh-implementation_graphlab-checkpoint.ipynb b/machine_learning/4_clustering_and_retrieval/assigment/week2/.ipynb_checkpoints/1_nearest-neighbors-lsh-implementation_graphlab-checkpoint.ipynb new file mode 100644 index 0000000..a54a180 --- /dev/null +++ b/machine_learning/4_clustering_and_retrieval/assigment/week2/.ipynb_checkpoints/1_nearest-neighbors-lsh-implementation_graphlab-checkpoint.ipynb @@ -0,0 +1,1851 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Locality Sensitive Hashing" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Locality Sensitive Hashing (LSH) provides for a fast, efficient approximate nearest neighbor search. The algorithm scales well with respect to the number of data points as well as dimensions.\n", + "\n", + "In this assignment, you will\n", + "* Implement the LSH algorithm for approximate nearest neighbor search\n", + "* Examine the accuracy for different documents by comparing against brute force search, and also contrast runtimes\n", + "* Explore the role of the algorithm’s tuning parameters in the accuracy of the method" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Note to Amazon EC2 users**: To conserve memory, make sure to stop all the other notebooks before running this notebook." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Import necessary packages" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "import numpy as np\n", + "import graphlab\n", + "from scipy.sparse import csr_matrix\n", + "from scipy.sparse.linalg import norm\n", + "from sklearn.metrics.pairwise import pairwise_distances\n", + "import time\n", + "from copy import copy\n", + "import matplotlib.pyplot as plt\n", + "%matplotlib inline" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Upgrading to Scipy 0.16.0 or later**. This assignment requires SciPy 0.16.0 or later. To upgrade, uncomment and run the following cell:" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "# !conda upgrade -y scipy" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "collapsed": false + }, + "source": [ + "## Load in the Wikipedia dataset" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "wiki = graphlab.SFrame('people_wiki.gl/')" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "For this assignment, let us assign a unique ID to each document." + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": { + "collapsed": false + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
idURInametext
0<http://dbpedia.org/resou
rce/Digby_Morrell> ...		Digby Morrelldigby morrell born 10
october 1979 is a former ...
1<http://dbpedia.org/resou
rce/Alfred_J._Lewy> ...		Alfred J. Lewyalfred j lewy aka sandy
lewy graduated from ...
2<http://dbpedia.org/resou
rce/Harpdog_Brown> ...		Harpdog Brownharpdog brown is a singer
and harmonica player who ...
3<http://dbpedia.org/resou
rce/Franz_Rottensteiner> ...		Franz Rottensteinerfranz rottensteiner born
in waidmannsfeld lower ...
4<http://dbpedia.org/resou
rce/G-Enka> ...		G-Enkahenry krvits born 30
december 1974 in tallinn ...
5<http://dbpedia.org/resou
rce/Sam_Henderson> ...		Sam Hendersonsam henderson born
october 18 1969 is an ...
6<http://dbpedia.org/resou
rce/Aaron_LaCrate> ...		Aaron LaCrateaaron lacrate is an
american music producer ...
7<http://dbpedia.org/resou
rce/Trevor_Ferguson> ...		Trevor Fergusontrevor ferguson aka john
farrow born 11 november ...
8<http://dbpedia.org/resou
rce/Grant_Nelson> ...		Grant Nelsongrant nelson born 27
april 1971 in london ...
9<http://dbpedia.org/resou
rce/Cathy_Caruth> ...		Cathy Caruthcathy caruth born 1955 is
frank h t rhodes ...
\n", + "[59071 rows x 4 columns]
Note: Only the head of the SFrame is printed.
You can use print_rows(num_rows=m, num_columns=n) to print more rows and columns.\n", + "
" + ], + "text/plain": [ + "Columns:\n", + "\tid\tint\n", + "\tURI\tstr\n", + "\tname\tstr\n", + "\ttext\tstr\n", + "\n", + "Rows: 59071\n", + "\n", + "Data:\n", + "+----+-------------------------------+---------------------+\n", + "| id | URI | name |\n", + "+----+-------------------------------+---------------------+\n", + "| 0 | \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
idURInametexttf_idf
0<http://dbpedia.org/resou
rce/Digby_Morrell> ...		Digby Morrelldigby morrell born 10
october 1979 is a former ...		{'selection':
3.836578553093086, ...
1<http://dbpedia.org/resou
rce/Alfred_J._Lewy> ...		Alfred J. Lewyalfred j lewy aka sandy
lewy graduated from ...		{'precise':
6.44320060695519, ...
2<http://dbpedia.org/resou
rce/Harpdog_Brown> ...		Harpdog Brownharpdog brown is a singer
and harmonica player who ...		{'just':
2.7007299687108643, ...
3<http://dbpedia.org/resou
rce/Franz_Rottensteiner> ...		Franz Rottensteinerfranz rottensteiner born
in waidmannsfeld lower ...		{'all':
1.6431112434912472, ...
4<http://dbpedia.org/resou
rce/G-Enka> ...		G-Enkahenry krvits born 30
december 1974 in tallinn ...		{'they':
1.8993401178193898, ...
5<http://dbpedia.org/resou
rce/Sam_Henderson> ...		Sam Hendersonsam henderson born
october 18 1969 is an ...		{'currently':
1.637088969126014, ...
6<http://dbpedia.org/resou
rce/Aaron_LaCrate> ...		Aaron LaCrateaaron lacrate is an
american music producer ...		{'exclusive':
10.455187230695827, ...
7<http://dbpedia.org/resou
rce/Trevor_Ferguson> ...		Trevor Fergusontrevor ferguson aka john
farrow born 11 november ...		{'taxi':
6.0520214560945025, ...
8<http://dbpedia.org/resou
rce/Grant_Nelson> ...		Grant Nelsongrant nelson born 27
april 1971 in london ...		{'houston':
3.935505942157149, ...
9<http://dbpedia.org/resou
rce/Cathy_Caruth> ...		Cathy Caruthcathy caruth born 1955 is
frank h t rhodes ...		{'phenomenon':
5.750053426395245, ...
\n", + "[59071 rows x 5 columns]
Note: Only the head of the SFrame is printed.
You can use print_rows(num_rows=m, num_columns=n) to print more rows and columns.\n", + "" + ], + "text/plain": [ + "Columns:\n", + "\tid\tint\n", + "\tURI\tstr\n", + "\tname\tstr\n", + "\ttext\tstr\n", + "\ttf_idf\tdict\n", + "\n", + "Rows: 59071\n", + "\n", + "Data:\n", + "+----+-------------------------------+---------------------+\n", + "| id | URI | name |\n", + "+----+-------------------------------+---------------------+\n", + "| 0 | = 0 # True if positive sign; False if negative sign" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Similarly, the second bit is computed as the sign of the dot product between the second random vector and the document vector." + ] + }, + { + "cell_type": "code", + "execution_count": 21, + "metadata": { + "collapsed": false + }, + "outputs": [ + { + "data": { + "text/plain": [ + "array([ True], dtype=bool)" + ] + }, + "execution_count": 21, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "doc.dot(random_vectors[:, 1]) >= 0 # True if positive sign; False if negative sign" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We can compute all of the bin index bits at once as follows. Note the absence of the explicit `for` loop over the 16 vectors. Matrix operations let us batch dot-product computation in a highly efficent manner, unlike the `for` loop construction. Given the relative inefficiency of loops in Python, the advantage of matrix operations is even greater." + ] + }, + { + "cell_type": "code", + "execution_count": 23, + "metadata": { + "collapsed": false + }, + "outputs": [ + { + "data": { + "text/plain": [ + "array([[ True, True, False, False, False, True, True, False, True,\n", + " True, True, False, False, True, False, True]], dtype=bool)" + ] + }, + "execution_count": 23, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "doc.dot(random_vectors) >= 0 # should return an array of 16 True/False bits" + ] + }, + { + "cell_type": "code", + "execution_count": 24, + "metadata": { + "collapsed": false + }, + "outputs": [ + { + "data": { + "text/plain": [ + "array([[1, 1, 0, 0, 0, 1, 1, 0, 1, 1, 1, 0, 0, 1, 0, 1]])" + ] + }, + "execution_count": 24, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "np.array(doc.dot(random_vectors) >= 0, dtype=int) # display index bits in 0/1's" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "All documents that obtain exactly this vector will be assigned to the same bin. We'd like to repeat the identical operation on all documents in the Wikipedia dataset and compute the corresponding bin indices. Again, we use matrix operations so that no explicit loop is needed." + ] + }, + { + "cell_type": "code", + "execution_count": 25, + "metadata": { + "collapsed": false + }, + "outputs": [ + { + "data": { + "text/plain": [ + "array([[ True, True, False, False, False, True, True, False, True,\n", + " True, True, False, False, True, False, True],\n", + " [ True, False, False, False, True, True, False, True, True,\n", + " False, True, False, True, False, False, True]], dtype=bool)" + ] + }, + "execution_count": 25, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "corpus[0:2].dot(random_vectors) >= 0 # compute bit indices of first two documents" + ] + }, + { + "cell_type": "code", + "execution_count": 26, + "metadata": { + "collapsed": false + }, + "outputs": [ + { + "data": { + "text/plain": [ + "array([[ True, True, False, ..., True, False, True],\n", + " [ True, False, False, ..., False, False, True],\n", + " [False, True, False, ..., True, False, True],\n", + " ..., \n", + " [ True, True, False, ..., True, True, True],\n", + " [False, True, True, ..., True, False, True],\n", + " [ True, False, True, ..., False, False, True]], dtype=bool)" + ] + }, + "execution_count": 26, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "corpus.dot(random_vectors) >= 0 # compute bit indices of ALL documents" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We're almost done! To make it convenient to refer to individual bins, we convert each binary bin index into a single integer: \n", + "```\n", + "Bin index integer\n", + "[0,0,0,0,0,0,0,0,0,0,0,0] => 0\n", + "[0,0,0,0,0,0,0,0,0,0,0,1] => 1\n", + "[0,0,0,0,0,0,0,0,0,0,1,0] => 2\n", + "[0,0,0,0,0,0,0,0,0,0,1,1] => 3\n", + "...\n", + "[1,1,1,1,1,1,1,1,1,1,0,0] => 65532\n", + "[1,1,1,1,1,1,1,1,1,1,0,1] => 65533\n", + "[1,1,1,1,1,1,1,1,1,1,1,0] => 65534\n", + "[1,1,1,1,1,1,1,1,1,1,1,1] => 65535 (= 2^16-1)\n", + "```\n", + "By the [rules of binary number representation](https://en.wikipedia.org/wiki/Binary_number#Decimal), we just need to compute the dot product between the document vector and the vector consisting of powers of 2:" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Notes**" + ] + }, + { + "cell_type": "code", + "execution_count": 28, + "metadata": { + "collapsed": false + }, + "outputs": [ + { + "data": { + "text/plain": [ + "array([15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0])" + ] + }, + "execution_count": 28, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "np.arange(15, -1, -1)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- **The Operators**:\n", + "\n", + "`x << y`\n", + "- Returns x with the bits shifted to the left by y places (and new bits on the right-hand-side are zeros). This is the same as multiplying x by 2**y." + ] + }, + { + "cell_type": "code", + "execution_count": 27, + "metadata": { + "collapsed": false + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[[ True True False False False True True False True True True False\n", + " False True False True]]\n", + "[32768 16384 8192 4096 2048 1024 512 256 128 64 32 16\n", + " 8 4 2 1]\n", + "[50917]\n" + ] + } + ], + "source": [ + "doc = corpus[0, :] # first document\n", + "index_bits = (doc.dot(random_vectors) >= 0)\n", + "powers_of_two = (1 << np.arange(15, -1, -1))\n", + "print index_bits\n", + "print powers_of_two\n", + "print index_bits.dot(powers_of_two)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Since it's the dot product again, we batch it with a matrix operation:" + ] + }, + { + "cell_type": "code", + "execution_count": 29, + "metadata": { + "collapsed": false + }, + "outputs": [ + { + "data": { + "text/plain": [ + "array([50917, 36265, 19365, ..., 52983, 27589, 41449])" + ] + }, + "execution_count": 29, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "index_bits = corpus.dot(random_vectors) >= 0\n", + "index_bits.dot(powers_of_two)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "This array gives us the integer index of the bins for all documents.\n", + "\n", + "Now we are ready to complete the following function. Given the integer bin indices for the documents, you should compile a list of document IDs that belong to each bin. Since a list is to be maintained for each unique bin index, a dictionary of lists is used.\n", + "\n", + "1. Compute the integer bin indices. This step is already completed.\n", + "2. For each document in the dataset, do the following:\n", + " * Get the integer bin index for the document.\n", + " * Fetch the list of document ids associated with the bin; if no list yet exists for this bin, assign the bin an empty list.\n", + " * Add the document id to the end of the list.\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "def train_lsh(data, num_vector=16, seed=None):\n", + " \n", + " dim = corpus.shape[1]\n", + " if seed is not None:\n", + " np.random.seed(seed)\n", + " random_vectors = generate_random_vectors(num_vector, dim)\n", + " \n", + " powers_of_two = 1 << np.arange(num_vector-1, -1, -1)\n", + " \n", + " table = {}\n", + " \n", + " # Partition data points into bins\n", + " bin_index_bits = (data.dot(random_vectors) >= 0)\n", + " \n", + " # Encode bin index bits into integers\n", + " bin_indices = bin_index_bits.dot(powers_of_two)\n", + " \n", + " # Update `table` so that `table[i]` is the list of document ids with bin index equal to i.\n", + " for data_index, bin_index in enumerate(bin_indices):\n", + " if bin_index not in table:\n", + " # If no list yet exists for this bin, assign the bin an empty list.\n", + " table[bin_index] = [] # YOUR CODE HERE\n", + " # Fetch the list of document ids associated with the bin and add the document id to the end.\n", + " ... # YOUR CODE HERE\n", + "\n", + " model = {'data': data,\n", + " 'bin_index_bits': bin_index_bits,\n", + " 'bin_indices': bin_indices,\n", + " 'table': table,\n", + " 'random_vectors': random_vectors,\n", + " 'num_vector': num_vector}\n", + " \n", + " return model" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Checkpoint**. " + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "model = train_lsh(corpus, num_vector=16, seed=143)\n", + "table = model['table']\n", + "if 0 in table and table[0] == [39583] and \\\n", + " 143 in table and table[143] == [19693, 28277, 29776, 30399]:\n", + " print 'Passed!'\n", + "else:\n", + " print 'Check your code.'" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Note.** We will be using the model trained here in the following sections, unless otherwise indicated." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Inspect bins" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let us look at some documents and see which bins they fall into." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "wiki[wiki['name'] == 'Barack Obama']" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Quiz Question**. What is the document `id` of Barack Obama's article?\n", + "\n", + "**Quiz Question**. Which bin contains Barack Obama's article? Enter its integer index." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Recall from the previous assignment that Joe Biden was a close neighbor of Barack Obama." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "wiki[wiki['name'] == 'Joe Biden']" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Quiz Question**. Examine the bit representations of the bins containing Barack Obama and Joe Biden. In how many places do they agree?\n", + "\n", + "1. 16 out of 16 places (Barack Obama and Joe Biden fall into the same bin)\n", + "2. 14 out of 16 places\n", + "3. 12 out of 16 places\n", + "4. 10 out of 16 places\n", + "5. 8 out of 16 places" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Compare the result with a former British diplomat, whose bin representation agrees with Obama's in only 8 out of 16 places." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "wiki[wiki['name']=='Wynn Normington Hugh-Jones']" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "print np.array(model['bin_index_bits'][22745], dtype=int) # list of 0/1's\n", + "print model['bin_indices'][22745] # integer format\n", + "model['bin_index_bits'][35817] == model['bin_index_bits'][22745]" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "How about the documents in the same bin as Barack Obama? Are they necessarily more similar to Obama than Biden? Let's look at which documents are in the same bin as the Barack Obama article." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "model['table'][model['bin_indices'][35817]]" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "There is four other documents that belong to the same bin. Which document are they?" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "doc_ids = list(model['table'][model['bin_indices'][35817]])\n", + "doc_ids.remove(35817) # display documents other than Obama\n", + "\n", + "docs = wiki.filter_by(values=doc_ids, column_name='id') # filter by id column\n", + "docs" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "It turns out that Joe Biden is much closer to Barack Obama than any of the four documents, even though Biden's bin representation differs from Obama's by 2 bits." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "def cosine_distance(x, y):\n", + " xy = x.dot(y.T)\n", + " dist = xy/(norm(x)*norm(y))\n", + " return 1-dist[0,0]\n", + "\n", + "obama_tf_idf = corpus[35817,:]\n", + "biden_tf_idf = corpus[24478,:]\n", + "\n", + "print '================= Cosine distance from Barack Obama'\n", + "print 'Barack Obama - {0:24s}: {1:f}'.format('Joe Biden',\n", + " cosine_distance(obama_tf_idf, biden_tf_idf))\n", + "for doc_id in doc_ids:\n", + " doc_tf_idf = corpus[doc_id,:]\n", + " print 'Barack Obama - {0:24s}: {1:f}'.format(wiki[doc_id]['name'],\n", + " cosine_distance(obama_tf_idf, doc_tf_idf))" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Moral of the story**. Similar data points will in general _tend to_ fall into _nearby_ bins, but that's all we can say about LSH. In a high-dimensional space such as text features, we often get unlucky with our selection of only a few random vectors such that dissimilar data points go into the same bin while similar data points fall into different bins. **Given a query document, we must consider all documents in the nearby bins and sort them according to their actual distances from the query.**" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Query the LSH model" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let us first implement the logic for searching nearby neighbors, which goes like this:\n", + "```\n", + "1. Let L be the bit representation of the bin that contains the query documents.\n", + "2. Consider all documents in bin L.\n", + "3. Consider documents in the bins whose bit representation differs from L by 1 bit.\n", + "4. Consider documents in the bins whose bit representation differs from L by 2 bits.\n", + "...\n", + "```" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "To obtain candidate bins that differ from the query bin by some number of bits, we use `itertools.combinations`, which produces all possible subsets of a given list. See [this documentation](https://docs.python.org/3/library/itertools.html#itertools.combinations) for details.\n", + "```\n", + "1. Decide on the search radius r. This will determine the number of different bits between the two vectors.\n", + "2. For each subset (n_1, n_2, ..., n_r) of the list [0, 1, 2, ..., num_vector-1], do the following:\n", + " * Flip the bits (n_1, n_2, ..., n_r) of the query bin to produce a new bit vector.\n", + " * Fetch the list of documents belonging to the bin indexed by the new bit vector.\n", + " * Add those documents to the candidate set.\n", + "```\n", + "\n", + "Each line of output from the following cell is a 3-tuple indicating where the candidate bin would differ from the query bin. For instance,\n", + "```\n", + "(0, 1, 3)\n", + "```\n", + "indicates that the candiate bin differs from the query bin in first, second, and fourth bits." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [ + "from itertools import combinations" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false, + "scrolled": true + }, + "outputs": [], + "source": [ + "num_vector = 16\n", + "search_radius = 3\n", + "\n", + "for diff in combinations(range(num_vector), search_radius):\n", + " print diff" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "With this output in mind, implement the logic for nearby bin search:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [ + "def search_nearby_bins(query_bin_bits, table, search_radius=2, initial_candidates=set()):\n", + " \"\"\"\n", + " For a given query vector and trained LSH model, return all candidate neighbors for\n", + " the query among all bins within the given search radius.\n", + " \n", + " Example usage\n", + " -------------\n", + " >>> model = train_lsh(corpus, num_vector=16, seed=143)\n", + " >>> q = model['bin_index_bits'][0] # vector for the first document\n", + " \n", + " >>> candidates = search_nearby_bins(q, model['table'])\n", + " \"\"\"\n", + " num_vector = len(query_bin_bits)\n", + " powers_of_two = 1 << np.arange(num_vector-1, -1, -1)\n", + " \n", + " # Allow the user to provide an initial set of candidates.\n", + " candidate_set = copy(initial_candidates)\n", + " \n", + " for different_bits in combinations(range(num_vector), search_radius): \n", + " # Flip the bits (n_1,n_2,...,n_r) of the query bin to produce a new bit vector.\n", + " ## Hint: you can iterate over a tuple like a list\n", + " alternate_bits = copy(query_bin_bits)\n", + " for i in different_bits:\n", + " alternate_bits[i] = ... # YOUR CODE HERE \n", + " \n", + " # Convert the new bit vector to an integer index\n", + " nearby_bin = alternate_bits.dot(powers_of_two)\n", + " \n", + " # Fetch the list of documents belonging to the bin indexed by the new bit vector.\n", + " # Then add those documents to candidate_set\n", + " # Make sure that the bin exists in the table!\n", + " # Hint: update() method for sets lets you add an entire list to the set\n", + " if nearby_bin in table:\n", + " ... # YOUR CODE HERE: Update candidate_set with the documents in this bin.\n", + " \n", + " return candidate_set" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Checkpoint**. Running the function with `search_radius=0` should yield the list of documents belonging to the same bin as the query." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "obama_bin_index = model['bin_index_bits'][35817] # bin index of Barack Obama\n", + "candidate_set = search_nearby_bins(obama_bin_index, model['table'], search_radius=0)\n", + "if candidate_set == set([35817, 21426, 53937, 39426, 50261]):\n", + " print 'Passed test'\n", + "else:\n", + " print 'Check your code'\n", + "print 'List of documents in the same bin as Obama: 35817, 21426, 53937, 39426, 50261'" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Checkpoint**. Running the function with `search_radius=1` adds more documents to the fore." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "candidate_set = search_nearby_bins(obama_bin_index, model['table'], search_radius=1, initial_candidates=candidate_set)\n", + "if candidate_set == set([39426, 38155, 38412, 28444, 9757, 41631, 39207, 59050, 47773, 53937, 21426, 34547,\n", + " 23229, 55615, 39877, 27404, 33996, 21715, 50261, 21975, 33243, 58723, 35817, 45676,\n", + " 19699, 2804, 20347]):\n", + " print 'Passed test'\n", + "else:\n", + " print 'Check your code'" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Note**. Don't be surprised if few of the candidates look similar to Obama. This is why we add as many candidates as our computational budget allows and sort them by their distance to the query." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Now we have a function that can return all the candidates from neighboring bins. Next we write a function to collect all candidates and compute their true distance to the query." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false, + "scrolled": true + }, + "outputs": [], + "source": [ + "def query(vec, model, k, max_search_radius):\n", + " \n", + " data = model['data']\n", + " table = model['table']\n", + " random_vectors = model['random_vectors']\n", + " num_vector = random_vectors.shape[1]\n", + " \n", + " \n", + " # Compute bin index for the query vector, in bit representation.\n", + " bin_index_bits = (vec.dot(random_vectors) >= 0).flatten()\n", + " \n", + " # Search nearby bins and collect candidates\n", + " candidate_set = set()\n", + " for search_radius in xrange(max_search_radius+1):\n", + " candidate_set = search_nearby_bins(bin_index_bits, table, search_radius, initial_candidates=candidate_set)\n", + " \n", + " # Sort candidates by their true distances from the query\n", + " nearest_neighbors = graphlab.SFrame({'id':candidate_set})\n", + " candidates = data[np.array(list(candidate_set)),:]\n", + " nearest_neighbors['distance'] = pairwise_distances(candidates, vec, metric='cosine').flatten()\n", + " \n", + " return nearest_neighbors.topk('distance', k, reverse=True), len(candidate_set)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let's try it out with Obama:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "query(corpus[35817,:], model, k=10, max_search_radius=3)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "To identify the documents, it's helpful to join this table with the Wikipedia table:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "query(corpus[35817,:], model, k=10, max_search_radius=3)[0].join(wiki[['id', 'name']], on='id').sort('distance')" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We have shown that we have a working LSH implementation!" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Experimenting with your LSH implementation" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "In the following sections we have implemented a few experiments so that you can gain intuition for how your LSH implementation behaves in different situations. This will help you understand the effect of searching nearby bins and the performance of LSH versus computing nearest neighbors using a brute force search." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Effect of nearby bin search" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "How does nearby bin search affect the outcome of LSH? There are three variables that are affected by the search radius:\n", + "* Number of candidate documents considered\n", + "* Query time\n", + "* Distance of approximate neighbors from the query" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let us run LSH multiple times, each with different radii for nearby bin search. We will measure the three variables as discussed above." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "wiki[wiki['name']=='Barack Obama']" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false, + "scrolled": true + }, + "outputs": [], + "source": [ + "num_candidates_history = []\n", + "query_time_history = []\n", + "max_distance_from_query_history = []\n", + "min_distance_from_query_history = []\n", + "average_distance_from_query_history = []\n", + "\n", + "for max_search_radius in xrange(17):\n", + " start=time.time()\n", + " result, num_candidates = query(corpus[35817,:], model, k=10,\n", + " max_search_radius=max_search_radius)\n", + " end=time.time()\n", + " query_time = end-start\n", + " \n", + " print 'Radius:', max_search_radius\n", + " print result.join(wiki[['id', 'name']], on='id').sort('distance')\n", + " \n", + " average_distance_from_query = result['distance'][1:].mean()\n", + " max_distance_from_query = result['distance'][1:].max()\n", + " min_distance_from_query = result['distance'][1:].min()\n", + " \n", + " num_candidates_history.append(num_candidates)\n", + " query_time_history.append(query_time)\n", + " average_distance_from_query_history.append(average_distance_from_query)\n", + " max_distance_from_query_history.append(max_distance_from_query)\n", + " min_distance_from_query_history.append(min_distance_from_query)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Notice that the top 10 query results become more relevant as the search radius grows. Let's plot the three variables:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(num_candidates_history, linewidth=4)\n", + "plt.xlabel('Search radius')\n", + "plt.ylabel('# of documents searched')\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()\n", + "\n", + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(query_time_history, linewidth=4)\n", + "plt.xlabel('Search radius')\n", + "plt.ylabel('Query time (seconds)')\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()\n", + "\n", + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(average_distance_from_query_history, linewidth=4, label='Average of 10 neighbors')\n", + "plt.plot(max_distance_from_query_history, linewidth=4, label='Farthest of 10 neighbors')\n", + "plt.plot(min_distance_from_query_history, linewidth=4, label='Closest of 10 neighbors')\n", + "plt.xlabel('Search radius')\n", + "plt.ylabel('Cosine distance of neighbors')\n", + "plt.legend(loc='best', prop={'size':15})\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Some observations:\n", + "* As we increase the search radius, we find more neighbors that are a smaller distance away.\n", + "* With increased search radius comes a greater number documents that have to be searched. Query time is higher as a consequence.\n", + "* With sufficiently high search radius, the results of LSH begin to resemble the results of brute-force search." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Quiz Question**. What was the smallest search radius that yielded the correct nearest neighbor, namely Joe Biden?\n", + "\n", + "\n", + "**Quiz Question**. Suppose our goal was to produce 10 approximate nearest neighbors whose average distance from the query document is within 0.01 of the average for the true 10 nearest neighbors. For Barack Obama, the true 10 nearest neighbors are on average about 0.77. What was the smallest search radius for Barack Obama that produced an average distance of 0.78 or better?" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Quality metrics for neighbors" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "The above analysis is limited by the fact that it was run with a single query, namely Barack Obama. We should repeat the analysis for the entirety of data. Iterating over all documents would take a long time, so let us randomly choose 10 documents for our analysis.\n", + "\n", + "For each document, we first compute the true 25 nearest neighbors, and then run LSH multiple times. We look at two metrics:\n", + "\n", + "* Precision@10: How many of the 10 neighbors given by LSH are among the true 25 nearest neighbors?\n", + "* Average cosine distance of the neighbors from the query\n", + "\n", + "Then we run LSH multiple times with different search radii." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [ + "def brute_force_query(vec, data, k):\n", + " num_data_points = data.shape[0]\n", + " \n", + " # Compute distances for ALL data points in training set\n", + " nearest_neighbors = graphlab.SFrame({'id':range(num_data_points)})\n", + " nearest_neighbors['distance'] = pairwise_distances(data, vec, metric='cosine').flatten()\n", + " \n", + " return nearest_neighbors.topk('distance', k, reverse=True)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "The following cell will run LSH with multiple search radii and compute the quality metrics for each run. Allow a few minutes to complete." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "max_radius = 17\n", + "precision = {i:[] for i in xrange(max_radius)}\n", + "average_distance = {i:[] for i in xrange(max_radius)}\n", + "query_time = {i:[] for i in xrange(max_radius)}\n", + "\n", + "np.random.seed(0)\n", + "num_queries = 10\n", + "for i, ix in enumerate(np.random.choice(corpus.shape[0], num_queries, replace=False)):\n", + " print('%s / %s' % (i, num_queries))\n", + " ground_truth = set(brute_force_query(corpus[ix,:], corpus, k=25)['id'])\n", + " # Get the set of 25 true nearest neighbors\n", + " \n", + " for r in xrange(1,max_radius):\n", + " start = time.time()\n", + " result, num_candidates = query(corpus[ix,:], model, k=10, max_search_radius=r)\n", + " end = time.time()\n", + "\n", + " query_time[r].append(end-start)\n", + " # precision = (# of neighbors both in result and ground_truth)/10.0\n", + " precision[r].append(len(set(result['id']) & ground_truth)/10.0)\n", + " average_distance[r].append(result['distance'][1:].mean())" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(range(1,17), [np.mean(average_distance[i]) for i in xrange(1,17)], linewidth=4, label='Average over 10 neighbors')\n", + "plt.xlabel('Search radius')\n", + "plt.ylabel('Cosine distance')\n", + "plt.legend(loc='best', prop={'size':15})\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()\n", + "\n", + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(range(1,17), [np.mean(precision[i]) for i in xrange(1,17)], linewidth=4, label='Precison@10')\n", + "plt.xlabel('Search radius')\n", + "plt.ylabel('Precision')\n", + "plt.legend(loc='best', prop={'size':15})\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()\n", + "\n", + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(range(1,17), [np.mean(query_time[i]) for i in xrange(1,17)], linewidth=4, label='Query time')\n", + "plt.xlabel('Search radius')\n", + "plt.ylabel('Query time (seconds)')\n", + "plt.legend(loc='best', prop={'size':15})\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "The observations for Barack Obama generalize to the entire dataset." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Effect of number of random vectors" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let us now turn our focus to the remaining parameter: the number of random vectors. We run LSH with different number of random vectors, ranging from 5 to 20. We fix the search radius to 3.\n", + "\n", + "Allow a few minutes for the following cell to complete." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "precision = {i:[] for i in xrange(5,20)}\n", + "average_distance = {i:[] for i in xrange(5,20)}\n", + "query_time = {i:[] for i in xrange(5,20)}\n", + "num_candidates_history = {i:[] for i in xrange(5,20)}\n", + "ground_truth = {}\n", + "\n", + "np.random.seed(0)\n", + "num_queries = 10\n", + "docs = np.random.choice(corpus.shape[0], num_queries, replace=False)\n", + "\n", + "for i, ix in enumerate(docs):\n", + " ground_truth[ix] = set(brute_force_query(corpus[ix,:], corpus, k=25)['id'])\n", + " # Get the set of 25 true nearest neighbors\n", + "\n", + "for num_vector in xrange(5,20):\n", + " print('num_vector = %s' % (num_vector))\n", + " model = train_lsh(corpus, num_vector, seed=143)\n", + " \n", + " for i, ix in enumerate(docs):\n", + " start = time.time()\n", + " result, num_candidates = query(corpus[ix,:], model, k=10, max_search_radius=3)\n", + " end = time.time()\n", + " \n", + " query_time[num_vector].append(end-start)\n", + " precision[num_vector].append(len(set(result['id']) & ground_truth[ix])/10.0)\n", + " average_distance[num_vector].append(result['distance'][1:].mean())\n", + " num_candidates_history[num_vector].append(num_candidates)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(range(5,20), [np.mean(average_distance[i]) for i in xrange(5,20)], linewidth=4, label='Average over 10 neighbors')\n", + "plt.xlabel('# of random vectors')\n", + "plt.ylabel('Cosine distance')\n", + "plt.legend(loc='best', prop={'size':15})\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()\n", + "\n", + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(range(5,20), [np.mean(precision[i]) for i in xrange(5,20)], linewidth=4, label='Precison@10')\n", + "plt.xlabel('# of random vectors')\n", + "plt.ylabel('Precision')\n", + "plt.legend(loc='best', prop={'size':15})\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()\n", + "\n", + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(range(5,20), [np.mean(query_time[i]) for i in xrange(5,20)], linewidth=4, label='Query time (seconds)')\n", + "plt.xlabel('# of random vectors')\n", + "plt.ylabel('Query time (seconds)')\n", + "plt.legend(loc='best', prop={'size':15})\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()\n", + "\n", + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(range(5,20), [np.mean(num_candidates_history[i]) for i in xrange(5,20)], linewidth=4,\n", + " label='# of documents searched')\n", + "plt.xlabel('# of random vectors')\n", + "plt.ylabel('# of documents searched')\n", + "plt.legend(loc='best', prop={'size':15})\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We see a similar trade-off between quality and performance: as the number of random vectors increases, the query time goes down as each bin contains fewer documents on average, but on average the neighbors are likewise placed farther from the query. On the other hand, when using a small enough number of random vectors, LSH becomes very similar brute-force search: Many documents appear in a single bin, so searching the query bin alone covers a lot of the corpus; then, including neighboring bins might result in searching all documents, just as in the brute-force approach." + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 2", + "language": "python", + "name": "python2" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 2 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython2", + "version": "2.7.11" + }, + "toc": { + "toc_cell": false, + "toc_number_sections": false, + "toc_threshold": "8", + "toc_window_display": false + } + }, + "nbformat": 4, + "nbformat_minor": 0 +} diff --git a/machine_learning/4_clustering_and_retrieval/assigment/week2/1_nearest-neighbors-lsh-implementation_blank.ipynb b/machine_learning/4_clustering_and_retrieval/assigment/week2/1_nearest-neighbors-lsh-implementation_blank.ipynb new file mode 100644 index 0000000..fc7d20b --- /dev/null +++ b/machine_learning/4_clustering_and_retrieval/assigment/week2/1_nearest-neighbors-lsh-implementation_blank.ipynb @@ -0,0 +1,1409 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Locality Sensitive Hashing" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Locality Sensitive Hashing (LSH) provides for a fast, efficient approximate nearest neighbor search. The algorithm scales well with respect to the number of data points as well as dimensions.\n", + "\n", + "In this assignment, you will\n", + "* Implement the LSH algorithm for approximate nearest neighbor search\n", + "* Examine the accuracy for different documents by comparing against brute force search, and also contrast runtimes\n", + "* Explore the role of the algorithm’s tuning parameters in the accuracy of the method" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Note to Amazon EC2 users**: To conserve memory, make sure to stop all the other notebooks before running this notebook." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Import necessary packages" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "import numpy as np\n", + "import graphlab\n", + "from scipy.sparse import csr_matrix\n", + "from scipy.sparse.linalg import norm\n", + "from sklearn.metrics.pairwise import pairwise_distances\n", + "import time\n", + "from copy import copy\n", + "import matplotlib.pyplot as plt\n", + "%matplotlib inline" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Upgrading to Scipy 0.16.0 or later**. This assignment requires SciPy 0.16.0 or later. To upgrade, uncomment and run the following cell:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "# !conda upgrade -y scipy" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "collapsed": false + }, + "source": [ + "## Load in the Wikipedia dataset" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "wiki = graphlab.SFrame('people_wiki.gl/')" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "For this assignment, let us assign a unique ID to each document." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "wiki = wiki.add_row_number()\n", + "wiki" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Extract TF-IDF matrix" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We first use GraphLab Create to compute a TF-IDF representation for each document." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "wiki['tf_idf'] = graphlab.text_analytics.tf_idf(wiki['text'])\n", + "wiki" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "collapsed": false + }, + "source": [ + "For the remainder of the assignment, we will use sparse matrices. Sparse matrices are [matrices](https://en.wikipedia.org/wiki/Matrix_(mathematics%29 ) that have a small number of nonzero entries. A good data structure for sparse matrices would only store the nonzero entries to save space and speed up computation. SciPy provides a highly-optimized library for sparse matrices. Many matrix operations available for NumPy arrays are also available for SciPy sparse matrices.\n", + "\n", + "We first convert the TF-IDF column (in dictionary format) into the SciPy sparse matrix format." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [ + "def sframe_to_scipy(column):\n", + " \"\"\" \n", + " Convert a dict-typed SArray into a SciPy sparse matrix.\n", + " \n", + " Returns\n", + " -------\n", + " mat : a SciPy sparse matrix where mat[i, j] is the value of word j for document i.\n", + " mapping : a dictionary where mapping[j] is the word whose values are in column j.\n", + " \"\"\"\n", + " # Create triples of (row_id, feature_id, count).\n", + " x = graphlab.SFrame({'X1':column})\n", + " \n", + " # 1. Add a row number.\n", + " x = x.add_row_number()\n", + " # 2. Stack will transform x to have a row for each unique (row, key) pair.\n", + " x = x.stack('X1', ['feature', 'value'])\n", + "\n", + " # Map words into integers using a OneHotEncoder feature transformation.\n", + " f = graphlab.feature_engineering.OneHotEncoder(features=['feature'])\n", + "\n", + " # We first fit the transformer using the above data.\n", + " f.fit(x)\n", + "\n", + " # The transform method will add a new column that is the transformed version\n", + " # of the 'word' column.\n", + " x = f.transform(x)\n", + "\n", + " # Get the feature mapping.\n", + " mapping = f['feature_encoding']\n", + "\n", + " # Get the actual word id.\n", + " x['feature_id'] = x['encoded_features'].dict_keys().apply(lambda x: x[0])\n", + "\n", + " # Create numpy arrays that contain the data for the sparse matrix.\n", + " i = np.array(x['id'])\n", + " j = np.array(x['feature_id'])\n", + " v = np.array(x['value'])\n", + " width = x['id'].max() + 1\n", + " height = x['feature_id'].max() + 1\n", + "\n", + " # Create a sparse matrix.\n", + " mat = csr_matrix((v, (i, j)), shape=(width, height))\n", + "\n", + " return mat, mapping" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "The conversion should take a few minutes to complete." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "start=time.time()\n", + "corpus, mapping = sframe_to_scipy(wiki['tf_idf'])\n", + "end=time.time()\n", + "print end-start" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Checkpoint**: The following code block should return 'Check passed correctly', indicating that your matrix contains TF-IDF values for 59071 documents and 547979 unique words. Otherwise, it will return Error." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "assert corpus.shape == (59071, 547979)\n", + "print 'Check passed correctly!'" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Train an LSH model" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "LSH performs an efficient neighbor search by randomly partitioning all reference data points into different bins. Today we will build a popular variant of LSH known as random binary projection, which approximates cosine distance. There are other variants we could use for other choices of distance metrics.\n", + "\n", + "The first step is to generate a collection of random vectors from the standard Gaussian distribution." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [ + "def generate_random_vectors(num_vector, dim):\n", + " return np.random.randn(dim, num_vector)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "To visualize these Gaussian random vectors, let's look at an example in low-dimensions. Below, we generate 3 random vectors each of dimension 5." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "# Generate 3 random vectors of dimension 5, arranged into a single 5 x 3 matrix.\n", + "np.random.seed(0) # set seed=0 for consistent results\n", + "generate_random_vectors(num_vector=3, dim=5)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We now generate random vectors of the same dimensionality as our vocubulary size (547979). Each vector can be used to compute one bit in the bin encoding. We generate 16 vectors, leading to a 16-bit encoding of the bin index for each document." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "# Generate 16 random vectors of dimension 547979\n", + "np.random.seed(0)\n", + "random_vectors = generate_random_vectors(num_vector=16, dim=547979)\n", + "random_vectors.shape" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Next, we partition data points into bins. Instead of using explicit loops, we'd like to utilize matrix operations for greater efficiency. Let's walk through the construction step by step.\n", + "\n", + "We'd like to decide which bin document 0 should go. Since 16 random vectors were generated in the previous cell, we have 16 bits to represent the bin index. The first bit is given by the sign of the dot product between the first random vector and the document's TF-IDF vector." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "doc = corpus[0, :] # vector of tf-idf values for document 0\n", + "doc.dot(random_vectors[:, 0]) >= 0 # True if positive sign; False if negative sign" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Similarly, the second bit is computed as the sign of the dot product between the second random vector and the document vector." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "doc.dot(random_vectors[:, 1]) >= 0 # True if positive sign; False if negative sign" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We can compute all of the bin index bits at once as follows. Note the absence of the explicit `for` loop over the 16 vectors. Matrix operations let us batch dot-product computation in a highly efficent manner, unlike the `for` loop construction. Given the relative inefficiency of loops in Python, the advantage of matrix operations is even greater." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "doc.dot(random_vectors) >= 0 # should return an array of 16 True/False bits" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "np.array(doc.dot(random_vectors) >= 0, dtype=int) # display index bits in 0/1's" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "All documents that obtain exactly this vector will be assigned to the same bin. We'd like to repeat the identical operation on all documents in the Wikipedia dataset and compute the corresponding bin indices. Again, we use matrix operations so that no explicit loop is needed." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "corpus[0:2].dot(random_vectors) >= 0 # compute bit indices of first two documents" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "corpus.dot(random_vectors) >= 0 # compute bit indices of ALL documents" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We're almost done! To make it convenient to refer to individual bins, we convert each binary bin index into a single integer: \n", + "```\n", + "Bin index integer\n", + "[0,0,0,0,0,0,0,0,0,0,0,0] => 0\n", + "[0,0,0,0,0,0,0,0,0,0,0,1] => 1\n", + "[0,0,0,0,0,0,0,0,0,0,1,0] => 2\n", + "[0,0,0,0,0,0,0,0,0,0,1,1] => 3\n", + "...\n", + "[1,1,1,1,1,1,1,1,1,1,0,0] => 65532\n", + "[1,1,1,1,1,1,1,1,1,1,0,1] => 65533\n", + "[1,1,1,1,1,1,1,1,1,1,1,0] => 65534\n", + "[1,1,1,1,1,1,1,1,1,1,1,1] => 65535 (= 2^16-1)\n", + "```\n", + "By the [rules of binary number representation](https://en.wikipedia.org/wiki/Binary_number#Decimal), we just need to compute the dot product between the document vector and the vector consisting of powers of 2:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "doc = corpus[0, :] # first document\n", + "index_bits = (doc.dot(random_vectors) >= 0)\n", + "powers_of_two = (1 << np.arange(15, -1, -1))\n", + "print index_bits\n", + "print powers_of_two\n", + "print index_bits.dot(powers_of_two)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Since it's the dot product again, we batch it with a matrix operation:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "index_bits = corpus.dot(random_vectors) >= 0\n", + "index_bits.dot(powers_of_two)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "This array gives us the integer index of the bins for all documents.\n", + "\n", + "Now we are ready to complete the following function. Given the integer bin indices for the documents, you should compile a list of document IDs that belong to each bin. Since a list is to be maintained for each unique bin index, a dictionary of lists is used.\n", + "\n", + "1. Compute the integer bin indices. This step is already completed.\n", + "2. For each document in the dataset, do the following:\n", + " * Get the integer bin index for the document.\n", + " * Fetch the list of document ids associated with the bin; if no list yet exists for this bin, assign the bin an empty list.\n", + " * Add the document id to the end of the list.\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "def train_lsh(data, num_vector=16, seed=None):\n", + " \n", + " dim = corpus.shape[1]\n", + " if seed is not None:\n", + " np.random.seed(seed)\n", + " random_vectors = generate_random_vectors(num_vector, dim)\n", + " \n", + " powers_of_two = 1 << np.arange(num_vector-1, -1, -1)\n", + " \n", + " table = {}\n", + " \n", + " # Partition data points into bins\n", + " bin_index_bits = (data.dot(random_vectors) >= 0)\n", + " \n", + " # Encode bin index bits into integers\n", + " bin_indices = bin_index_bits.dot(powers_of_two)\n", + " \n", + " # Update `table` so that `table[i]` is the list of document ids with bin index equal to i.\n", + " for data_index, bin_index in enumerate(bin_indices):\n", + " if bin_index not in table:\n", + " # If no list yet exists for this bin, assign the bin an empty list.\n", + " table[bin_index] = ... # YOUR CODE HERE\n", + " # Fetch the list of document ids associated with the bin and add the document id to the end.\n", + " ... # YOUR CODE HERE\n", + "\n", + " model = {'data': data,\n", + " 'bin_index_bits': bin_index_bits,\n", + " 'bin_indices': bin_indices,\n", + " 'table': table,\n", + " 'random_vectors': random_vectors,\n", + " 'num_vector': num_vector}\n", + " \n", + " return model" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Checkpoint**. " + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "model = train_lsh(corpus, num_vector=16, seed=143)\n", + "table = model['table']\n", + "if 0 in table and table[0] == [39583] and \\\n", + " 143 in table and table[143] == [19693, 28277, 29776, 30399]:\n", + " print 'Passed!'\n", + "else:\n", + " print 'Check your code.'" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Note.** We will be using the model trained here in the following sections, unless otherwise indicated." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Inspect bins" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let us look at some documents and see which bins they fall into." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "wiki[wiki['name'] == 'Barack Obama']" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Quiz Question**. What is the document `id` of Barack Obama's article?\n", + "\n", + "**Quiz Question**. Which bin contains Barack Obama's article? Enter its integer index." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Recall from the previous assignment that Joe Biden was a close neighbor of Barack Obama." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "wiki[wiki['name'] == 'Joe Biden']" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Quiz Question**. Examine the bit representations of the bins containing Barack Obama and Joe Biden. In how many places do they agree?\n", + "\n", + "1. 16 out of 16 places (Barack Obama and Joe Biden fall into the same bin)\n", + "2. 14 out of 16 places\n", + "3. 12 out of 16 places\n", + "4. 10 out of 16 places\n", + "5. 8 out of 16 places" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Compare the result with a former British diplomat, whose bin representation agrees with Obama's in only 8 out of 16 places." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "wiki[wiki['name']=='Wynn Normington Hugh-Jones']" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "print np.array(model['bin_index_bits'][22745], dtype=int) # list of 0/1's\n", + "print model['bin_indices'][22745] # integer format\n", + "model['bin_index_bits'][35817] == model['bin_index_bits'][22745]" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "How about the documents in the same bin as Barack Obama? Are they necessarily more similar to Obama than Biden? Let's look at which documents are in the same bin as the Barack Obama article." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "model['table'][model['bin_indices'][35817]]" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "There is four other documents that belong to the same bin. Which document are they?" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "doc_ids = list(model['table'][model['bin_indices'][35817]])\n", + "doc_ids.remove(35817) # display documents other than Obama\n", + "\n", + "docs = wiki.filter_by(values=doc_ids, column_name='id') # filter by id column\n", + "docs" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "It turns out that Joe Biden is much closer to Barack Obama than any of the four documents, even though Biden's bin representation differs from Obama's by 2 bits." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "def cosine_distance(x, y):\n", + " xy = x.dot(y.T)\n", + " dist = xy/(norm(x)*norm(y))\n", + " return 1-dist[0,0]\n", + "\n", + "obama_tf_idf = corpus[35817,:]\n", + "biden_tf_idf = corpus[24478,:]\n", + "\n", + "print '================= Cosine distance from Barack Obama'\n", + "print 'Barack Obama - {0:24s}: {1:f}'.format('Joe Biden',\n", + " cosine_distance(obama_tf_idf, biden_tf_idf))\n", + "for doc_id in doc_ids:\n", + " doc_tf_idf = corpus[doc_id,:]\n", + " print 'Barack Obama - {0:24s}: {1:f}'.format(wiki[doc_id]['name'],\n", + " cosine_distance(obama_tf_idf, doc_tf_idf))" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Moral of the story**. Similar data points will in general _tend to_ fall into _nearby_ bins, but that's all we can say about LSH. In a high-dimensional space such as text features, we often get unlucky with our selection of only a few random vectors such that dissimilar data points go into the same bin while similar data points fall into different bins. **Given a query document, we must consider all documents in the nearby bins and sort them according to their actual distances from the query.**" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Query the LSH model" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let us first implement the logic for searching nearby neighbors, which goes like this:\n", + "```\n", + "1. Let L be the bit representation of the bin that contains the query documents.\n", + "2. Consider all documents in bin L.\n", + "3. Consider documents in the bins whose bit representation differs from L by 1 bit.\n", + "4. Consider documents in the bins whose bit representation differs from L by 2 bits.\n", + "...\n", + "```" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "To obtain candidate bins that differ from the query bin by some number of bits, we use `itertools.combinations`, which produces all possible subsets of a given list. See [this documentation](https://docs.python.org/3/library/itertools.html#itertools.combinations) for details.\n", + "```\n", + "1. Decide on the search radius r. This will determine the number of different bits between the two vectors.\n", + "2. For each subset (n_1, n_2, ..., n_r) of the list [0, 1, 2, ..., num_vector-1], do the following:\n", + " * Flip the bits (n_1, n_2, ..., n_r) of the query bin to produce a new bit vector.\n", + " * Fetch the list of documents belonging to the bin indexed by the new bit vector.\n", + " * Add those documents to the candidate set.\n", + "```\n", + "\n", + "Each line of output from the following cell is a 3-tuple indicating where the candidate bin would differ from the query bin. For instance,\n", + "```\n", + "(0, 1, 3)\n", + "```\n", + "indicates that the candiate bin differs from the query bin in first, second, and fourth bits." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [ + "from itertools import combinations" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false, + "scrolled": true + }, + "outputs": [], + "source": [ + "num_vector = 16\n", + "search_radius = 3\n", + "\n", + "for diff in combinations(range(num_vector), search_radius):\n", + " print diff" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "With this output in mind, implement the logic for nearby bin search:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [ + "def search_nearby_bins(query_bin_bits, table, search_radius=2, initial_candidates=set()):\n", + " \"\"\"\n", + " For a given query vector and trained LSH model, return all candidate neighbors for\n", + " the query among all bins within the given search radius.\n", + " \n", + " Example usage\n", + " -------------\n", + " >>> model = train_lsh(corpus, num_vector=16, seed=143)\n", + " >>> q = model['bin_index_bits'][0] # vector for the first document\n", + " \n", + " >>> candidates = search_nearby_bins(q, model['table'])\n", + " \"\"\"\n", + " num_vector = len(query_bin_bits)\n", + " powers_of_two = 1 << np.arange(num_vector-1, -1, -1)\n", + " \n", + " # Allow the user to provide an initial set of candidates.\n", + " candidate_set = copy(initial_candidates)\n", + " \n", + " for different_bits in combinations(range(num_vector), search_radius): \n", + " # Flip the bits (n_1,n_2,...,n_r) of the query bin to produce a new bit vector.\n", + " ## Hint: you can iterate over a tuple like a list\n", + " alternate_bits = copy(query_bin_bits)\n", + " for i in different_bits:\n", + " alternate_bits[i] = ... # YOUR CODE HERE \n", + " \n", + " # Convert the new bit vector to an integer index\n", + " nearby_bin = alternate_bits.dot(powers_of_two)\n", + " \n", + " # Fetch the list of documents belonging to the bin indexed by the new bit vector.\n", + " # Then add those documents to candidate_set\n", + " # Make sure that the bin exists in the table!\n", + " # Hint: update() method for sets lets you add an entire list to the set\n", + " if nearby_bin in table:\n", + " ... # YOUR CODE HERE: Update candidate_set with the documents in this bin.\n", + " \n", + " return candidate_set" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Checkpoint**. Running the function with `search_radius=0` should yield the list of documents belonging to the same bin as the query." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "obama_bin_index = model['bin_index_bits'][35817] # bin index of Barack Obama\n", + "candidate_set = search_nearby_bins(obama_bin_index, model['table'], search_radius=0)\n", + "if candidate_set == set([35817, 21426, 53937, 39426, 50261]):\n", + " print 'Passed test'\n", + "else:\n", + " print 'Check your code'\n", + "print 'List of documents in the same bin as Obama: 35817, 21426, 53937, 39426, 50261'" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Checkpoint**. Running the function with `search_radius=1` adds more documents to the fore." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "candidate_set = search_nearby_bins(obama_bin_index, model['table'], search_radius=1, initial_candidates=candidate_set)\n", + "if candidate_set == set([39426, 38155, 38412, 28444, 9757, 41631, 39207, 59050, 47773, 53937, 21426, 34547,\n", + " 23229, 55615, 39877, 27404, 33996, 21715, 50261, 21975, 33243, 58723, 35817, 45676,\n", + " 19699, 2804, 20347]):\n", + " print 'Passed test'\n", + "else:\n", + " print 'Check your code'" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Note**. Don't be surprised if few of the candidates look similar to Obama. This is why we add as many candidates as our computational budget allows and sort them by their distance to the query." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Now we have a function that can return all the candidates from neighboring bins. Next we write a function to collect all candidates and compute their true distance to the query." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false, + "scrolled": true + }, + "outputs": [], + "source": [ + "def query(vec, model, k, max_search_radius):\n", + " \n", + " data = model['data']\n", + " table = model['table']\n", + " random_vectors = model['random_vectors']\n", + " num_vector = random_vectors.shape[1]\n", + " \n", + " \n", + " # Compute bin index for the query vector, in bit representation.\n", + " bin_index_bits = (vec.dot(random_vectors) >= 0).flatten()\n", + " \n", + " # Search nearby bins and collect candidates\n", + " candidate_set = set()\n", + " for search_radius in xrange(max_search_radius+1):\n", + " candidate_set = search_nearby_bins(bin_index_bits, table, search_radius, initial_candidates=candidate_set)\n", + " \n", + " # Sort candidates by their true distances from the query\n", + " nearest_neighbors = graphlab.SFrame({'id':candidate_set})\n", + " candidates = data[np.array(list(candidate_set)),:]\n", + " nearest_neighbors['distance'] = pairwise_distances(candidates, vec, metric='cosine').flatten()\n", + " \n", + " return nearest_neighbors.topk('distance', k, reverse=True), len(candidate_set)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let's try it out with Obama:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "query(corpus[35817,:], model, k=10, max_search_radius=3)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "To identify the documents, it's helpful to join this table with the Wikipedia table:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "query(corpus[35817,:], model, k=10, max_search_radius=3)[0].join(wiki[['id', 'name']], on='id').sort('distance')" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We have shown that we have a working LSH implementation!" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Experimenting with your LSH implementation" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "In the following sections we have implemented a few experiments so that you can gain intuition for how your LSH implementation behaves in different situations. This will help you understand the effect of searching nearby bins and the performance of LSH versus computing nearest neighbors using a brute force search." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Effect of nearby bin search" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "How does nearby bin search affect the outcome of LSH? There are three variables that are affected by the search radius:\n", + "* Number of candidate documents considered\n", + "* Query time\n", + "* Distance of approximate neighbors from the query" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let us run LSH multiple times, each with different radii for nearby bin search. We will measure the three variables as discussed above." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "wiki[wiki['name']=='Barack Obama']" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false, + "scrolled": true + }, + "outputs": [], + "source": [ + "num_candidates_history = []\n", + "query_time_history = []\n", + "max_distance_from_query_history = []\n", + "min_distance_from_query_history = []\n", + "average_distance_from_query_history = []\n", + "\n", + "for max_search_radius in xrange(17):\n", + " start=time.time()\n", + " result, num_candidates = query(corpus[35817,:], model, k=10,\n", + " max_search_radius=max_search_radius)\n", + " end=time.time()\n", + " query_time = end-start\n", + " \n", + " print 'Radius:', max_search_radius\n", + " print result.join(wiki[['id', 'name']], on='id').sort('distance')\n", + " \n", + " average_distance_from_query = result['distance'][1:].mean()\n", + " max_distance_from_query = result['distance'][1:].max()\n", + " min_distance_from_query = result['distance'][1:].min()\n", + " \n", + " num_candidates_history.append(num_candidates)\n", + " query_time_history.append(query_time)\n", + " average_distance_from_query_history.append(average_distance_from_query)\n", + " max_distance_from_query_history.append(max_distance_from_query)\n", + " min_distance_from_query_history.append(min_distance_from_query)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Notice that the top 10 query results become more relevant as the search radius grows. Let's plot the three variables:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(num_candidates_history, linewidth=4)\n", + "plt.xlabel('Search radius')\n", + "plt.ylabel('# of documents searched')\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()\n", + "\n", + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(query_time_history, linewidth=4)\n", + "plt.xlabel('Search radius')\n", + "plt.ylabel('Query time (seconds)')\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()\n", + "\n", + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(average_distance_from_query_history, linewidth=4, label='Average of 10 neighbors')\n", + "plt.plot(max_distance_from_query_history, linewidth=4, label='Farthest of 10 neighbors')\n", + "plt.plot(min_distance_from_query_history, linewidth=4, label='Closest of 10 neighbors')\n", + "plt.xlabel('Search radius')\n", + "plt.ylabel('Cosine distance of neighbors')\n", + "plt.legend(loc='best', prop={'size':15})\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Some observations:\n", + "* As we increase the search radius, we find more neighbors that are a smaller distance away.\n", + "* With increased search radius comes a greater number documents that have to be searched. Query time is higher as a consequence.\n", + "* With sufficiently high search radius, the results of LSH begin to resemble the results of brute-force search." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Quiz Question**. What was the smallest search radius that yielded the correct nearest neighbor, namely Joe Biden?\n", + "\n", + "\n", + "**Quiz Question**. Suppose our goal was to produce 10 approximate nearest neighbors whose average distance from the query document is within 0.01 of the average for the true 10 nearest neighbors. For Barack Obama, the true 10 nearest neighbors are on average about 0.77. What was the smallest search radius for Barack Obama that produced an average distance of 0.78 or better?" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Quality metrics for neighbors" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "The above analysis is limited by the fact that it was run with a single query, namely Barack Obama. We should repeat the analysis for the entirety of data. Iterating over all documents would take a long time, so let us randomly choose 10 documents for our analysis.\n", + "\n", + "For each document, we first compute the true 25 nearest neighbors, and then run LSH multiple times. We look at two metrics:\n", + "\n", + "* Precision@10: How many of the 10 neighbors given by LSH are among the true 25 nearest neighbors?\n", + "* Average cosine distance of the neighbors from the query\n", + "\n", + "Then we run LSH multiple times with different search radii." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [ + "def brute_force_query(vec, data, k):\n", + " num_data_points = data.shape[0]\n", + " \n", + " # Compute distances for ALL data points in training set\n", + " nearest_neighbors = graphlab.SFrame({'id':range(num_data_points)})\n", + " nearest_neighbors['distance'] = pairwise_distances(data, vec, metric='cosine').flatten()\n", + " \n", + " return nearest_neighbors.topk('distance', k, reverse=True)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "The following cell will run LSH with multiple search radii and compute the quality metrics for each run. Allow a few minutes to complete." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "max_radius = 17\n", + "precision = {i:[] for i in xrange(max_radius)}\n", + "average_distance = {i:[] for i in xrange(max_radius)}\n", + "query_time = {i:[] for i in xrange(max_radius)}\n", + "\n", + "np.random.seed(0)\n", + "num_queries = 10\n", + "for i, ix in enumerate(np.random.choice(corpus.shape[0], num_queries, replace=False)):\n", + " print('%s / %s' % (i, num_queries))\n", + " ground_truth = set(brute_force_query(corpus[ix,:], corpus, k=25)['id'])\n", + " # Get the set of 25 true nearest neighbors\n", + " \n", + " for r in xrange(1,max_radius):\n", + " start = time.time()\n", + " result, num_candidates = query(corpus[ix,:], model, k=10, max_search_radius=r)\n", + " end = time.time()\n", + "\n", + " query_time[r].append(end-start)\n", + " # precision = (# of neighbors both in result and ground_truth)/10.0\n", + " precision[r].append(len(set(result['id']) & ground_truth)/10.0)\n", + " average_distance[r].append(result['distance'][1:].mean())" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(range(1,17), [np.mean(average_distance[i]) for i in xrange(1,17)], linewidth=4, label='Average over 10 neighbors')\n", + "plt.xlabel('Search radius')\n", + "plt.ylabel('Cosine distance')\n", + "plt.legend(loc='best', prop={'size':15})\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()\n", + "\n", + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(range(1,17), [np.mean(precision[i]) for i in xrange(1,17)], linewidth=4, label='Precison@10')\n", + "plt.xlabel('Search radius')\n", + "plt.ylabel('Precision')\n", + "plt.legend(loc='best', prop={'size':15})\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()\n", + "\n", + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(range(1,17), [np.mean(query_time[i]) for i in xrange(1,17)], linewidth=4, label='Query time')\n", + "plt.xlabel('Search radius')\n", + "plt.ylabel('Query time (seconds)')\n", + "plt.legend(loc='best', prop={'size':15})\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "The observations for Barack Obama generalize to the entire dataset." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Effect of number of random vectors" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let us now turn our focus to the remaining parameter: the number of random vectors. We run LSH with different number of random vectors, ranging from 5 to 20. We fix the search radius to 3.\n", + "\n", + "Allow a few minutes for the following cell to complete." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "precision = {i:[] for i in xrange(5,20)}\n", + "average_distance = {i:[] for i in xrange(5,20)}\n", + "query_time = {i:[] for i in xrange(5,20)}\n", + "num_candidates_history = {i:[] for i in xrange(5,20)}\n", + "ground_truth = {}\n", + "\n", + "np.random.seed(0)\n", + "num_queries = 10\n", + "docs = np.random.choice(corpus.shape[0], num_queries, replace=False)\n", + "\n", + "for i, ix in enumerate(docs):\n", + " ground_truth[ix] = set(brute_force_query(corpus[ix,:], corpus, k=25)['id'])\n", + " # Get the set of 25 true nearest neighbors\n", + "\n", + "for num_vector in xrange(5,20):\n", + " print('num_vector = %s' % (num_vector))\n", + " model = train_lsh(corpus, num_vector, seed=143)\n", + " \n", + " for i, ix in enumerate(docs):\n", + " start = time.time()\n", + " result, num_candidates = query(corpus[ix,:], model, k=10, max_search_radius=3)\n", + " end = time.time()\n", + " \n", + " query_time[num_vector].append(end-start)\n", + " precision[num_vector].append(len(set(result['id']) & ground_truth[ix])/10.0)\n", + " average_distance[num_vector].append(result['distance'][1:].mean())\n", + " num_candidates_history[num_vector].append(num_candidates)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(range(5,20), [np.mean(average_distance[i]) for i in xrange(5,20)], linewidth=4, label='Average over 10 neighbors')\n", + "plt.xlabel('# of random vectors')\n", + "plt.ylabel('Cosine distance')\n", + "plt.legend(loc='best', prop={'size':15})\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()\n", + "\n", + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(range(5,20), [np.mean(precision[i]) for i in xrange(5,20)], linewidth=4, label='Precison@10')\n", + "plt.xlabel('# of random vectors')\n", + "plt.ylabel('Precision')\n", + "plt.legend(loc='best', prop={'size':15})\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()\n", + "\n", + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(range(5,20), [np.mean(query_time[i]) for i in xrange(5,20)], linewidth=4, label='Query time (seconds)')\n", + "plt.xlabel('# of random vectors')\n", + "plt.ylabel('Query time (seconds)')\n", + "plt.legend(loc='best', prop={'size':15})\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()\n", + "\n", + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(range(5,20), [np.mean(num_candidates_history[i]) for i in xrange(5,20)], linewidth=4,\n", + " label='# of documents searched')\n", + "plt.xlabel('# of random vectors')\n", + "plt.ylabel('# of documents searched')\n", + "plt.legend(loc='best', prop={'size':15})\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We see a similar trade-off between quality and performance: as the number of random vectors increases, the query time goes down as each bin contains fewer documents on average, but on average the neighbors are likewise placed farther from the query. On the other hand, when using a small enough number of random vectors, LSH becomes very similar brute-force search: Many documents appear in a single bin, so searching the query bin alone covers a lot of the corpus; then, including neighboring bins might result in searching all documents, just as in the brute-force approach." + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 2", + "language": "python", + "name": "python2" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 2 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython2", + "version": "2.7.11" + } + }, + "nbformat": 4, + "nbformat_minor": 0 +} diff --git a/machine_learning/4_clustering_and_retrieval/assigment/week2/1_nearest-neighbors-lsh-implementation_graphlab.ipynb b/machine_learning/4_clustering_and_retrieval/assigment/week2/1_nearest-neighbors-lsh-implementation_graphlab.ipynb new file mode 100644 index 0000000..a54a180 --- /dev/null +++ b/machine_learning/4_clustering_and_retrieval/assigment/week2/1_nearest-neighbors-lsh-implementation_graphlab.ipynb @@ -0,0 +1,1851 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Locality Sensitive Hashing" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Locality Sensitive Hashing (LSH) provides for a fast, efficient approximate nearest neighbor search. The algorithm scales well with respect to the number of data points as well as dimensions.\n", + "\n", + "In this assignment, you will\n", + "* Implement the LSH algorithm for approximate nearest neighbor search\n", + "* Examine the accuracy for different documents by comparing against brute force search, and also contrast runtimes\n", + "* Explore the role of the algorithm’s tuning parameters in the accuracy of the method" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Note to Amazon EC2 users**: To conserve memory, make sure to stop all the other notebooks before running this notebook." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Import necessary packages" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "import numpy as np\n", + "import graphlab\n", + "from scipy.sparse import csr_matrix\n", + "from scipy.sparse.linalg import norm\n", + "from sklearn.metrics.pairwise import pairwise_distances\n", + "import time\n", + "from copy import copy\n", + "import matplotlib.pyplot as plt\n", + "%matplotlib inline" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Upgrading to Scipy 0.16.0 or later**. This assignment requires SciPy 0.16.0 or later. To upgrade, uncomment and run the following cell:" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "# !conda upgrade -y scipy" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "collapsed": false + }, + "source": [ + "## Load in the Wikipedia dataset" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "wiki = graphlab.SFrame('people_wiki.gl/')" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "For this assignment, let us assign a unique ID to each document." + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": { + "collapsed": false + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
idURInametext
0<http://dbpedia.org/resou
rce/Digby_Morrell> ...		Digby Morrelldigby morrell born 10
october 1979 is a former ...
1<http://dbpedia.org/resou
rce/Alfred_J._Lewy> ...		Alfred J. Lewyalfred j lewy aka sandy
lewy graduated from ...
2<http://dbpedia.org/resou
rce/Harpdog_Brown> ...		Harpdog Brownharpdog brown is a singer
and harmonica player who ...
3<http://dbpedia.org/resou
rce/Franz_Rottensteiner> ...		Franz Rottensteinerfranz rottensteiner born
in waidmannsfeld lower ...
4<http://dbpedia.org/resou
rce/G-Enka> ...		G-Enkahenry krvits born 30
december 1974 in tallinn ...
5<http://dbpedia.org/resou
rce/Sam_Henderson> ...		Sam Hendersonsam henderson born
october 18 1969 is an ...
6<http://dbpedia.org/resou
rce/Aaron_LaCrate> ...		Aaron LaCrateaaron lacrate is an
american music producer ...
7<http://dbpedia.org/resou
rce/Trevor_Ferguson> ...		Trevor Fergusontrevor ferguson aka john
farrow born 11 november ...
8<http://dbpedia.org/resou
rce/Grant_Nelson> ...		Grant Nelsongrant nelson born 27
april 1971 in london ...
9<http://dbpedia.org/resou
rce/Cathy_Caruth> ...		Cathy Caruthcathy caruth born 1955 is
frank h t rhodes ...
\n", + "[59071 rows x 4 columns]
Note: Only the head of the SFrame is printed.
You can use print_rows(num_rows=m, num_columns=n) to print more rows and columns.\n", + "
" + ], + "text/plain": [ + "Columns:\n", + "\tid\tint\n", + "\tURI\tstr\n", + "\tname\tstr\n", + "\ttext\tstr\n", + "\n", + "Rows: 59071\n", + "\n", + "Data:\n", + "+----+-------------------------------+---------------------+\n", + "| id | URI | name |\n", + "+----+-------------------------------+---------------------+\n", + "| 0 | \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
idURInametexttf_idf
0<http://dbpedia.org/resou
rce/Digby_Morrell> ...		Digby Morrelldigby morrell born 10
october 1979 is a former ...		{'selection':
3.836578553093086, ...
1<http://dbpedia.org/resou
rce/Alfred_J._Lewy> ...		Alfred J. Lewyalfred j lewy aka sandy
lewy graduated from ...		{'precise':
6.44320060695519, ...
2<http://dbpedia.org/resou
rce/Harpdog_Brown> ...		Harpdog Brownharpdog brown is a singer
and harmonica player who ...		{'just':
2.7007299687108643, ...
3<http://dbpedia.org/resou
rce/Franz_Rottensteiner> ...		Franz Rottensteinerfranz rottensteiner born
in waidmannsfeld lower ...		{'all':
1.6431112434912472, ...
4<http://dbpedia.org/resou
rce/G-Enka> ...		G-Enkahenry krvits born 30
december 1974 in tallinn ...		{'they':
1.8993401178193898, ...
5<http://dbpedia.org/resou
rce/Sam_Henderson> ...		Sam Hendersonsam henderson born
october 18 1969 is an ...		{'currently':
1.637088969126014, ...
6<http://dbpedia.org/resou
rce/Aaron_LaCrate> ...		Aaron LaCrateaaron lacrate is an
american music producer ...		{'exclusive':
10.455187230695827, ...
7<http://dbpedia.org/resou
rce/Trevor_Ferguson> ...		Trevor Fergusontrevor ferguson aka john
farrow born 11 november ...		{'taxi':
6.0520214560945025, ...
8<http://dbpedia.org/resou
rce/Grant_Nelson> ...		Grant Nelsongrant nelson born 27
april 1971 in london ...		{'houston':
3.935505942157149, ...
9<http://dbpedia.org/resou
rce/Cathy_Caruth> ...		Cathy Caruthcathy caruth born 1955 is
frank h t rhodes ...		{'phenomenon':
5.750053426395245, ...
\n", + "[59071 rows x 5 columns]
Note: Only the head of the SFrame is printed.
You can use print_rows(num_rows=m, num_columns=n) to print more rows and columns.\n", + "" + ], + "text/plain": [ + "Columns:\n", + "\tid\tint\n", + "\tURI\tstr\n", + "\tname\tstr\n", + "\ttext\tstr\n", + "\ttf_idf\tdict\n", + "\n", + "Rows: 59071\n", + "\n", + "Data:\n", + "+----+-------------------------------+---------------------+\n", + "| id | URI | name |\n", + "+----+-------------------------------+---------------------+\n", + "| 0 | = 0 # True if positive sign; False if negative sign" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Similarly, the second bit is computed as the sign of the dot product between the second random vector and the document vector." + ] + }, + { + "cell_type": "code", + "execution_count": 21, + "metadata": { + "collapsed": false + }, + "outputs": [ + { + "data": { + "text/plain": [ + "array([ True], dtype=bool)" + ] + }, + "execution_count": 21, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "doc.dot(random_vectors[:, 1]) >= 0 # True if positive sign; False if negative sign" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We can compute all of the bin index bits at once as follows. Note the absence of the explicit `for` loop over the 16 vectors. Matrix operations let us batch dot-product computation in a highly efficent manner, unlike the `for` loop construction. Given the relative inefficiency of loops in Python, the advantage of matrix operations is even greater." + ] + }, + { + "cell_type": "code", + "execution_count": 23, + "metadata": { + "collapsed": false + }, + "outputs": [ + { + "data": { + "text/plain": [ + "array([[ True, True, False, False, False, True, True, False, True,\n", + " True, True, False, False, True, False, True]], dtype=bool)" + ] + }, + "execution_count": 23, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "doc.dot(random_vectors) >= 0 # should return an array of 16 True/False bits" + ] + }, + { + "cell_type": "code", + "execution_count": 24, + "metadata": { + "collapsed": false + }, + "outputs": [ + { + "data": { + "text/plain": [ + "array([[1, 1, 0, 0, 0, 1, 1, 0, 1, 1, 1, 0, 0, 1, 0, 1]])" + ] + }, + "execution_count": 24, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "np.array(doc.dot(random_vectors) >= 0, dtype=int) # display index bits in 0/1's" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "All documents that obtain exactly this vector will be assigned to the same bin. We'd like to repeat the identical operation on all documents in the Wikipedia dataset and compute the corresponding bin indices. Again, we use matrix operations so that no explicit loop is needed." + ] + }, + { + "cell_type": "code", + "execution_count": 25, + "metadata": { + "collapsed": false + }, + "outputs": [ + { + "data": { + "text/plain": [ + "array([[ True, True, False, False, False, True, True, False, True,\n", + " True, True, False, False, True, False, True],\n", + " [ True, False, False, False, True, True, False, True, True,\n", + " False, True, False, True, False, False, True]], dtype=bool)" + ] + }, + "execution_count": 25, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "corpus[0:2].dot(random_vectors) >= 0 # compute bit indices of first two documents" + ] + }, + { + "cell_type": "code", + "execution_count": 26, + "metadata": { + "collapsed": false + }, + "outputs": [ + { + "data": { + "text/plain": [ + "array([[ True, True, False, ..., True, False, True],\n", + " [ True, False, False, ..., False, False, True],\n", + " [False, True, False, ..., True, False, True],\n", + " ..., \n", + " [ True, True, False, ..., True, True, True],\n", + " [False, True, True, ..., True, False, True],\n", + " [ True, False, True, ..., False, False, True]], dtype=bool)" + ] + }, + "execution_count": 26, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "corpus.dot(random_vectors) >= 0 # compute bit indices of ALL documents" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We're almost done! To make it convenient to refer to individual bins, we convert each binary bin index into a single integer: \n", + "```\n", + "Bin index integer\n", + "[0,0,0,0,0,0,0,0,0,0,0,0] => 0\n", + "[0,0,0,0,0,0,0,0,0,0,0,1] => 1\n", + "[0,0,0,0,0,0,0,0,0,0,1,0] => 2\n", + "[0,0,0,0,0,0,0,0,0,0,1,1] => 3\n", + "...\n", + "[1,1,1,1,1,1,1,1,1,1,0,0] => 65532\n", + "[1,1,1,1,1,1,1,1,1,1,0,1] => 65533\n", + "[1,1,1,1,1,1,1,1,1,1,1,0] => 65534\n", + "[1,1,1,1,1,1,1,1,1,1,1,1] => 65535 (= 2^16-1)\n", + "```\n", + "By the [rules of binary number representation](https://en.wikipedia.org/wiki/Binary_number#Decimal), we just need to compute the dot product between the document vector and the vector consisting of powers of 2:" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Notes**" + ] + }, + { + "cell_type": "code", + "execution_count": 28, + "metadata": { + "collapsed": false + }, + "outputs": [ + { + "data": { + "text/plain": [ + "array([15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0])" + ] + }, + "execution_count": 28, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "np.arange(15, -1, -1)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "- **The Operators**:\n", + "\n", + "`x << y`\n", + "- Returns x with the bits shifted to the left by y places (and new bits on the right-hand-side are zeros). This is the same as multiplying x by 2**y." + ] + }, + { + "cell_type": "code", + "execution_count": 27, + "metadata": { + "collapsed": false + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[[ True True False False False True True False True True True False\n", + " False True False True]]\n", + "[32768 16384 8192 4096 2048 1024 512 256 128 64 32 16\n", + " 8 4 2 1]\n", + "[50917]\n" + ] + } + ], + "source": [ + "doc = corpus[0, :] # first document\n", + "index_bits = (doc.dot(random_vectors) >= 0)\n", + "powers_of_two = (1 << np.arange(15, -1, -1))\n", + "print index_bits\n", + "print powers_of_two\n", + "print index_bits.dot(powers_of_two)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Since it's the dot product again, we batch it with a matrix operation:" + ] + }, + { + "cell_type": "code", + "execution_count": 29, + "metadata": { + "collapsed": false + }, + "outputs": [ + { + "data": { + "text/plain": [ + "array([50917, 36265, 19365, ..., 52983, 27589, 41449])" + ] + }, + "execution_count": 29, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "index_bits = corpus.dot(random_vectors) >= 0\n", + "index_bits.dot(powers_of_two)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "This array gives us the integer index of the bins for all documents.\n", + "\n", + "Now we are ready to complete the following function. Given the integer bin indices for the documents, you should compile a list of document IDs that belong to each bin. Since a list is to be maintained for each unique bin index, a dictionary of lists is used.\n", + "\n", + "1. Compute the integer bin indices. This step is already completed.\n", + "2. For each document in the dataset, do the following:\n", + " * Get the integer bin index for the document.\n", + " * Fetch the list of document ids associated with the bin; if no list yet exists for this bin, assign the bin an empty list.\n", + " * Add the document id to the end of the list.\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "def train_lsh(data, num_vector=16, seed=None):\n", + " \n", + " dim = corpus.shape[1]\n", + " if seed is not None:\n", + " np.random.seed(seed)\n", + " random_vectors = generate_random_vectors(num_vector, dim)\n", + " \n", + " powers_of_two = 1 << np.arange(num_vector-1, -1, -1)\n", + " \n", + " table = {}\n", + " \n", + " # Partition data points into bins\n", + " bin_index_bits = (data.dot(random_vectors) >= 0)\n", + " \n", + " # Encode bin index bits into integers\n", + " bin_indices = bin_index_bits.dot(powers_of_two)\n", + " \n", + " # Update `table` so that `table[i]` is the list of document ids with bin index equal to i.\n", + " for data_index, bin_index in enumerate(bin_indices):\n", + " if bin_index not in table:\n", + " # If no list yet exists for this bin, assign the bin an empty list.\n", + " table[bin_index] = [] # YOUR CODE HERE\n", + " # Fetch the list of document ids associated with the bin and add the document id to the end.\n", + " ... # YOUR CODE HERE\n", + "\n", + " model = {'data': data,\n", + " 'bin_index_bits': bin_index_bits,\n", + " 'bin_indices': bin_indices,\n", + " 'table': table,\n", + " 'random_vectors': random_vectors,\n", + " 'num_vector': num_vector}\n", + " \n", + " return model" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Checkpoint**. " + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "model = train_lsh(corpus, num_vector=16, seed=143)\n", + "table = model['table']\n", + "if 0 in table and table[0] == [39583] and \\\n", + " 143 in table and table[143] == [19693, 28277, 29776, 30399]:\n", + " print 'Passed!'\n", + "else:\n", + " print 'Check your code.'" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Note.** We will be using the model trained here in the following sections, unless otherwise indicated." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Inspect bins" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let us look at some documents and see which bins they fall into." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "wiki[wiki['name'] == 'Barack Obama']" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Quiz Question**. What is the document `id` of Barack Obama's article?\n", + "\n", + "**Quiz Question**. Which bin contains Barack Obama's article? Enter its integer index." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Recall from the previous assignment that Joe Biden was a close neighbor of Barack Obama." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "wiki[wiki['name'] == 'Joe Biden']" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Quiz Question**. Examine the bit representations of the bins containing Barack Obama and Joe Biden. In how many places do they agree?\n", + "\n", + "1. 16 out of 16 places (Barack Obama and Joe Biden fall into the same bin)\n", + "2. 14 out of 16 places\n", + "3. 12 out of 16 places\n", + "4. 10 out of 16 places\n", + "5. 8 out of 16 places" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Compare the result with a former British diplomat, whose bin representation agrees with Obama's in only 8 out of 16 places." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "wiki[wiki['name']=='Wynn Normington Hugh-Jones']" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "print np.array(model['bin_index_bits'][22745], dtype=int) # list of 0/1's\n", + "print model['bin_indices'][22745] # integer format\n", + "model['bin_index_bits'][35817] == model['bin_index_bits'][22745]" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "How about the documents in the same bin as Barack Obama? Are they necessarily more similar to Obama than Biden? Let's look at which documents are in the same bin as the Barack Obama article." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "model['table'][model['bin_indices'][35817]]" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "There is four other documents that belong to the same bin. Which document are they?" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "doc_ids = list(model['table'][model['bin_indices'][35817]])\n", + "doc_ids.remove(35817) # display documents other than Obama\n", + "\n", + "docs = wiki.filter_by(values=doc_ids, column_name='id') # filter by id column\n", + "docs" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "It turns out that Joe Biden is much closer to Barack Obama than any of the four documents, even though Biden's bin representation differs from Obama's by 2 bits." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "def cosine_distance(x, y):\n", + " xy = x.dot(y.T)\n", + " dist = xy/(norm(x)*norm(y))\n", + " return 1-dist[0,0]\n", + "\n", + "obama_tf_idf = corpus[35817,:]\n", + "biden_tf_idf = corpus[24478,:]\n", + "\n", + "print '================= Cosine distance from Barack Obama'\n", + "print 'Barack Obama - {0:24s}: {1:f}'.format('Joe Biden',\n", + " cosine_distance(obama_tf_idf, biden_tf_idf))\n", + "for doc_id in doc_ids:\n", + " doc_tf_idf = corpus[doc_id,:]\n", + " print 'Barack Obama - {0:24s}: {1:f}'.format(wiki[doc_id]['name'],\n", + " cosine_distance(obama_tf_idf, doc_tf_idf))" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Moral of the story**. Similar data points will in general _tend to_ fall into _nearby_ bins, but that's all we can say about LSH. In a high-dimensional space such as text features, we often get unlucky with our selection of only a few random vectors such that dissimilar data points go into the same bin while similar data points fall into different bins. **Given a query document, we must consider all documents in the nearby bins and sort them according to their actual distances from the query.**" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Query the LSH model" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let us first implement the logic for searching nearby neighbors, which goes like this:\n", + "```\n", + "1. Let L be the bit representation of the bin that contains the query documents.\n", + "2. Consider all documents in bin L.\n", + "3. Consider documents in the bins whose bit representation differs from L by 1 bit.\n", + "4. Consider documents in the bins whose bit representation differs from L by 2 bits.\n", + "...\n", + "```" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "To obtain candidate bins that differ from the query bin by some number of bits, we use `itertools.combinations`, which produces all possible subsets of a given list. See [this documentation](https://docs.python.org/3/library/itertools.html#itertools.combinations) for details.\n", + "```\n", + "1. Decide on the search radius r. This will determine the number of different bits between the two vectors.\n", + "2. For each subset (n_1, n_2, ..., n_r) of the list [0, 1, 2, ..., num_vector-1], do the following:\n", + " * Flip the bits (n_1, n_2, ..., n_r) of the query bin to produce a new bit vector.\n", + " * Fetch the list of documents belonging to the bin indexed by the new bit vector.\n", + " * Add those documents to the candidate set.\n", + "```\n", + "\n", + "Each line of output from the following cell is a 3-tuple indicating where the candidate bin would differ from the query bin. For instance,\n", + "```\n", + "(0, 1, 3)\n", + "```\n", + "indicates that the candiate bin differs from the query bin in first, second, and fourth bits." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [ + "from itertools import combinations" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false, + "scrolled": true + }, + "outputs": [], + "source": [ + "num_vector = 16\n", + "search_radius = 3\n", + "\n", + "for diff in combinations(range(num_vector), search_radius):\n", + " print diff" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "With this output in mind, implement the logic for nearby bin search:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [ + "def search_nearby_bins(query_bin_bits, table, search_radius=2, initial_candidates=set()):\n", + " \"\"\"\n", + " For a given query vector and trained LSH model, return all candidate neighbors for\n", + " the query among all bins within the given search radius.\n", + " \n", + " Example usage\n", + " -------------\n", + " >>> model = train_lsh(corpus, num_vector=16, seed=143)\n", + " >>> q = model['bin_index_bits'][0] # vector for the first document\n", + " \n", + " >>> candidates = search_nearby_bins(q, model['table'])\n", + " \"\"\"\n", + " num_vector = len(query_bin_bits)\n", + " powers_of_two = 1 << np.arange(num_vector-1, -1, -1)\n", + " \n", + " # Allow the user to provide an initial set of candidates.\n", + " candidate_set = copy(initial_candidates)\n", + " \n", + " for different_bits in combinations(range(num_vector), search_radius): \n", + " # Flip the bits (n_1,n_2,...,n_r) of the query bin to produce a new bit vector.\n", + " ## Hint: you can iterate over a tuple like a list\n", + " alternate_bits = copy(query_bin_bits)\n", + " for i in different_bits:\n", + " alternate_bits[i] = ... # YOUR CODE HERE \n", + " \n", + " # Convert the new bit vector to an integer index\n", + " nearby_bin = alternate_bits.dot(powers_of_two)\n", + " \n", + " # Fetch the list of documents belonging to the bin indexed by the new bit vector.\n", + " # Then add those documents to candidate_set\n", + " # Make sure that the bin exists in the table!\n", + " # Hint: update() method for sets lets you add an entire list to the set\n", + " if nearby_bin in table:\n", + " ... # YOUR CODE HERE: Update candidate_set with the documents in this bin.\n", + " \n", + " return candidate_set" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Checkpoint**. Running the function with `search_radius=0` should yield the list of documents belonging to the same bin as the query." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "obama_bin_index = model['bin_index_bits'][35817] # bin index of Barack Obama\n", + "candidate_set = search_nearby_bins(obama_bin_index, model['table'], search_radius=0)\n", + "if candidate_set == set([35817, 21426, 53937, 39426, 50261]):\n", + " print 'Passed test'\n", + "else:\n", + " print 'Check your code'\n", + "print 'List of documents in the same bin as Obama: 35817, 21426, 53937, 39426, 50261'" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Checkpoint**. Running the function with `search_radius=1` adds more documents to the fore." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "candidate_set = search_nearby_bins(obama_bin_index, model['table'], search_radius=1, initial_candidates=candidate_set)\n", + "if candidate_set == set([39426, 38155, 38412, 28444, 9757, 41631, 39207, 59050, 47773, 53937, 21426, 34547,\n", + " 23229, 55615, 39877, 27404, 33996, 21715, 50261, 21975, 33243, 58723, 35817, 45676,\n", + " 19699, 2804, 20347]):\n", + " print 'Passed test'\n", + "else:\n", + " print 'Check your code'" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Note**. Don't be surprised if few of the candidates look similar to Obama. This is why we add as many candidates as our computational budget allows and sort them by their distance to the query." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Now we have a function that can return all the candidates from neighboring bins. Next we write a function to collect all candidates and compute their true distance to the query." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false, + "scrolled": true + }, + "outputs": [], + "source": [ + "def query(vec, model, k, max_search_radius):\n", + " \n", + " data = model['data']\n", + " table = model['table']\n", + " random_vectors = model['random_vectors']\n", + " num_vector = random_vectors.shape[1]\n", + " \n", + " \n", + " # Compute bin index for the query vector, in bit representation.\n", + " bin_index_bits = (vec.dot(random_vectors) >= 0).flatten()\n", + " \n", + " # Search nearby bins and collect candidates\n", + " candidate_set = set()\n", + " for search_radius in xrange(max_search_radius+1):\n", + " candidate_set = search_nearby_bins(bin_index_bits, table, search_radius, initial_candidates=candidate_set)\n", + " \n", + " # Sort candidates by their true distances from the query\n", + " nearest_neighbors = graphlab.SFrame({'id':candidate_set})\n", + " candidates = data[np.array(list(candidate_set)),:]\n", + " nearest_neighbors['distance'] = pairwise_distances(candidates, vec, metric='cosine').flatten()\n", + " \n", + " return nearest_neighbors.topk('distance', k, reverse=True), len(candidate_set)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let's try it out with Obama:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "query(corpus[35817,:], model, k=10, max_search_radius=3)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "To identify the documents, it's helpful to join this table with the Wikipedia table:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "query(corpus[35817,:], model, k=10, max_search_radius=3)[0].join(wiki[['id', 'name']], on='id').sort('distance')" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We have shown that we have a working LSH implementation!" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Experimenting with your LSH implementation" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "In the following sections we have implemented a few experiments so that you can gain intuition for how your LSH implementation behaves in different situations. This will help you understand the effect of searching nearby bins and the performance of LSH versus computing nearest neighbors using a brute force search." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Effect of nearby bin search" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "How does nearby bin search affect the outcome of LSH? There are three variables that are affected by the search radius:\n", + "* Number of candidate documents considered\n", + "* Query time\n", + "* Distance of approximate neighbors from the query" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let us run LSH multiple times, each with different radii for nearby bin search. We will measure the three variables as discussed above." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "wiki[wiki['name']=='Barack Obama']" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false, + "scrolled": true + }, + "outputs": [], + "source": [ + "num_candidates_history = []\n", + "query_time_history = []\n", + "max_distance_from_query_history = []\n", + "min_distance_from_query_history = []\n", + "average_distance_from_query_history = []\n", + "\n", + "for max_search_radius in xrange(17):\n", + " start=time.time()\n", + " result, num_candidates = query(corpus[35817,:], model, k=10,\n", + " max_search_radius=max_search_radius)\n", + " end=time.time()\n", + " query_time = end-start\n", + " \n", + " print 'Radius:', max_search_radius\n", + " print result.join(wiki[['id', 'name']], on='id').sort('distance')\n", + " \n", + " average_distance_from_query = result['distance'][1:].mean()\n", + " max_distance_from_query = result['distance'][1:].max()\n", + " min_distance_from_query = result['distance'][1:].min()\n", + " \n", + " num_candidates_history.append(num_candidates)\n", + " query_time_history.append(query_time)\n", + " average_distance_from_query_history.append(average_distance_from_query)\n", + " max_distance_from_query_history.append(max_distance_from_query)\n", + " min_distance_from_query_history.append(min_distance_from_query)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Notice that the top 10 query results become more relevant as the search radius grows. Let's plot the three variables:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(num_candidates_history, linewidth=4)\n", + "plt.xlabel('Search radius')\n", + "plt.ylabel('# of documents searched')\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()\n", + "\n", + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(query_time_history, linewidth=4)\n", + "plt.xlabel('Search radius')\n", + "plt.ylabel('Query time (seconds)')\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()\n", + "\n", + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(average_distance_from_query_history, linewidth=4, label='Average of 10 neighbors')\n", + "plt.plot(max_distance_from_query_history, linewidth=4, label='Farthest of 10 neighbors')\n", + "plt.plot(min_distance_from_query_history, linewidth=4, label='Closest of 10 neighbors')\n", + "plt.xlabel('Search radius')\n", + "plt.ylabel('Cosine distance of neighbors')\n", + "plt.legend(loc='best', prop={'size':15})\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Some observations:\n", + "* As we increase the search radius, we find more neighbors that are a smaller distance away.\n", + "* With increased search radius comes a greater number documents that have to be searched. Query time is higher as a consequence.\n", + "* With sufficiently high search radius, the results of LSH begin to resemble the results of brute-force search." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "**Quiz Question**. What was the smallest search radius that yielded the correct nearest neighbor, namely Joe Biden?\n", + "\n", + "\n", + "**Quiz Question**. Suppose our goal was to produce 10 approximate nearest neighbors whose average distance from the query document is within 0.01 of the average for the true 10 nearest neighbors. For Barack Obama, the true 10 nearest neighbors are on average about 0.77. What was the smallest search radius for Barack Obama that produced an average distance of 0.78 or better?" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Quality metrics for neighbors" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "The above analysis is limited by the fact that it was run with a single query, namely Barack Obama. We should repeat the analysis for the entirety of data. Iterating over all documents would take a long time, so let us randomly choose 10 documents for our analysis.\n", + "\n", + "For each document, we first compute the true 25 nearest neighbors, and then run LSH multiple times. We look at two metrics:\n", + "\n", + "* Precision@10: How many of the 10 neighbors given by LSH are among the true 25 nearest neighbors?\n", + "* Average cosine distance of the neighbors from the query\n", + "\n", + "Then we run LSH multiple times with different search radii." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [ + "def brute_force_query(vec, data, k):\n", + " num_data_points = data.shape[0]\n", + " \n", + " # Compute distances for ALL data points in training set\n", + " nearest_neighbors = graphlab.SFrame({'id':range(num_data_points)})\n", + " nearest_neighbors['distance'] = pairwise_distances(data, vec, metric='cosine').flatten()\n", + " \n", + " return nearest_neighbors.topk('distance', k, reverse=True)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "The following cell will run LSH with multiple search radii and compute the quality metrics for each run. Allow a few minutes to complete." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "max_radius = 17\n", + "precision = {i:[] for i in xrange(max_radius)}\n", + "average_distance = {i:[] for i in xrange(max_radius)}\n", + "query_time = {i:[] for i in xrange(max_radius)}\n", + "\n", + "np.random.seed(0)\n", + "num_queries = 10\n", + "for i, ix in enumerate(np.random.choice(corpus.shape[0], num_queries, replace=False)):\n", + " print('%s / %s' % (i, num_queries))\n", + " ground_truth = set(brute_force_query(corpus[ix,:], corpus, k=25)['id'])\n", + " # Get the set of 25 true nearest neighbors\n", + " \n", + " for r in xrange(1,max_radius):\n", + " start = time.time()\n", + " result, num_candidates = query(corpus[ix,:], model, k=10, max_search_radius=r)\n", + " end = time.time()\n", + "\n", + " query_time[r].append(end-start)\n", + " # precision = (# of neighbors both in result and ground_truth)/10.0\n", + " precision[r].append(len(set(result['id']) & ground_truth)/10.0)\n", + " average_distance[r].append(result['distance'][1:].mean())" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(range(1,17), [np.mean(average_distance[i]) for i in xrange(1,17)], linewidth=4, label='Average over 10 neighbors')\n", + "plt.xlabel('Search radius')\n", + "plt.ylabel('Cosine distance')\n", + "plt.legend(loc='best', prop={'size':15})\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()\n", + "\n", + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(range(1,17), [np.mean(precision[i]) for i in xrange(1,17)], linewidth=4, label='Precison@10')\n", + "plt.xlabel('Search radius')\n", + "plt.ylabel('Precision')\n", + "plt.legend(loc='best', prop={'size':15})\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()\n", + "\n", + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(range(1,17), [np.mean(query_time[i]) for i in xrange(1,17)], linewidth=4, label='Query time')\n", + "plt.xlabel('Search radius')\n", + "plt.ylabel('Query time (seconds)')\n", + "plt.legend(loc='best', prop={'size':15})\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "The observations for Barack Obama generalize to the entire dataset." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Effect of number of random vectors" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Let us now turn our focus to the remaining parameter: the number of random vectors. We run LSH with different number of random vectors, ranging from 5 to 20. We fix the search radius to 3.\n", + "\n", + "Allow a few minutes for the following cell to complete." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "precision = {i:[] for i in xrange(5,20)}\n", + "average_distance = {i:[] for i in xrange(5,20)}\n", + "query_time = {i:[] for i in xrange(5,20)}\n", + "num_candidates_history = {i:[] for i in xrange(5,20)}\n", + "ground_truth = {}\n", + "\n", + "np.random.seed(0)\n", + "num_queries = 10\n", + "docs = np.random.choice(corpus.shape[0], num_queries, replace=False)\n", + "\n", + "for i, ix in enumerate(docs):\n", + " ground_truth[ix] = set(brute_force_query(corpus[ix,:], corpus, k=25)['id'])\n", + " # Get the set of 25 true nearest neighbors\n", + "\n", + "for num_vector in xrange(5,20):\n", + " print('num_vector = %s' % (num_vector))\n", + " model = train_lsh(corpus, num_vector, seed=143)\n", + " \n", + " for i, ix in enumerate(docs):\n", + " start = time.time()\n", + " result, num_candidates = query(corpus[ix,:], model, k=10, max_search_radius=3)\n", + " end = time.time()\n", + " \n", + " query_time[num_vector].append(end-start)\n", + " precision[num_vector].append(len(set(result['id']) & ground_truth[ix])/10.0)\n", + " average_distance[num_vector].append(result['distance'][1:].mean())\n", + " num_candidates_history[num_vector].append(num_candidates)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false + }, + "outputs": [], + "source": [ + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(range(5,20), [np.mean(average_distance[i]) for i in xrange(5,20)], linewidth=4, label='Average over 10 neighbors')\n", + "plt.xlabel('# of random vectors')\n", + "plt.ylabel('Cosine distance')\n", + "plt.legend(loc='best', prop={'size':15})\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()\n", + "\n", + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(range(5,20), [np.mean(precision[i]) for i in xrange(5,20)], linewidth=4, label='Precison@10')\n", + "plt.xlabel('# of random vectors')\n", + "plt.ylabel('Precision')\n", + "plt.legend(loc='best', prop={'size':15})\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()\n", + "\n", + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(range(5,20), [np.mean(query_time[i]) for i in xrange(5,20)], linewidth=4, label='Query time (seconds)')\n", + "plt.xlabel('# of random vectors')\n", + "plt.ylabel('Query time (seconds)')\n", + "plt.legend(loc='best', prop={'size':15})\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()\n", + "\n", + "plt.figure(figsize=(7,4.5))\n", + "plt.plot(range(5,20), [np.mean(num_candidates_history[i]) for i in xrange(5,20)], linewidth=4,\n", + " label='# of documents searched')\n", + "plt.xlabel('# of random vectors')\n", + "plt.ylabel('# of documents searched')\n", + "plt.legend(loc='best', prop={'size':15})\n", + "plt.rcParams.update({'font.size':16})\n", + "plt.tight_layout()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We see a similar trade-off between quality and performance: as the number of random vectors increases, the query time goes down as each bin contains fewer documents on average, but on average the neighbors are likewise placed farther from the query. On the other hand, when using a small enough number of random vectors, LSH becomes very similar brute-force search: Many documents appear in a single bin, so searching the query bin alone covers a lot of the corpus; then, including neighboring bins might result in searching all documents, just as in the brute-force approach." + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 2", + "language": "python", + "name": "python2" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 2 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython2", + "version": "2.7.11" + }, + "toc": { + "toc_cell": false, + "toc_number_sections": false, + "toc_threshold": "8", + "toc_window_display": false + } + }, + "nbformat": 4, + "nbformat_minor": 0 +} diff --git a/machine_learning/4_clustering_and_retrieval/assigment/week2/week2-assignment2.pdf b/machine_learning/4_clustering_and_retrieval/assigment/week2/week2-assignment2.pdf new file mode 100644 index 0000000000000000000000000000000000000000..38ce4cd998dd861e339842a3a68474b838d8528b GIT binary patch literal 659725 zcma%ibx>Tvw`OoBcyI<7Jh;2V3=jf?g+Oq35ALqR;K9S-8X&kN!GgOE5?q2qa9Muu zy{+B9wzlftdv151*6Ql+^L>4oG!&FLdAS6znWq2luVeE9cmU33cGwaU-12UgrXJ32 z0M>WTzP1hyrrd&DJOD%1cc$jHP9DzgHsS!VlZT}P;3Wd^UJLL6z{_LI%WEuX$Oe#g zadEKJvouq+_23rd7v|y@dda8?(tfA*7T{oOZwXMgG`DwtQIxlFb9S`k7UbpP;o=kK z5#ZwG;Rk4$TA8}p{!d0WY;FxVXA4hr%a@jD+1r|Ozcd)&D9HPg@Lw&ob#_uP^{@o6 zDv0s%@Cxw=^YHWW@C)0E-zrj4*p^Nf|LHg1|LMPyt%Ju)rQAvm zFU2cZnmb!qVsoomI$3+z0C)vNMERwp0PY?ygXD`^n&qhu?C%T_GC!QN= zyPs~F>DFT-9WAJ)Ai%${e>s4*zUM|a&t~WCmx+2GjGUS z$8HB(;RM6zT6sL8T|a! zaOS0Hn?vs~%)8kC#`z>0o?pBZ-NU^0+@@Ye1i-(-(EA^R+<0<_{bS0#y;iR zN|DBQ^p`83742zV8tLaL+TJ1J!s)+lEMK4|*FQhfKX(QB_T2CMSqcr+w4S&W1&6>u z1vA?D{%^aZP*ue8HjK}Dv!q!Tn8mKwflJ>@ZS-zDh*Dl-v1UaA`%q*wOXgW*ZaYx{ z6zq){tsp6UG~&u|Hag=1YfKnRcuP~3LU=G#3>!84hXSn!{$mP}LINkiEna0^!!J1) ziaj7fj#pS-jNgdj7=nMqY+=WhUdb3bD2EnD&HZe-@}e;wz#i>_LCt}m-tz|9O zKbZxzR}B}7xp5Zydo;aeUy)cnJ#Rg@GzF3dF)e;RI6pybo}}G^Av@$_$oIYqQ+{T| zhRn4AxZ5f1TYR>pFxHI*_Ay09#Z5HOQo*+;^n#aF{T5N(S zm7Qt@F)(x}(#?~aQmIvEgJn=zfd;#rQ%8yc`$*+$kcg{O0X!IVhb5;nk7NrhjE$>J z3>P3mOMO-pS0{nqHC;?7e_dhVR{o0VGUZn#fVzDdO$j<9I__#yhoRivJ{H0m|H@BF zrCu;P1vfUyrxFaE$g-O`k6Ry&o)tFoet)BlSzHgErezODg>8r>V<$&(_m_CaUXtS} z$5qcWP4SaJ`OLNMd)Mv0FdLB%sizGA93vwO-*|bu;-m!g^Nx4clj7h~qgWG@^ z02a3ZFm`H*D`Xj6)OG8iBD#F%4S=g4KhxsX@J(CXbC|>nOj&3K(<;x(#gAsW!@LK}bX%L&`TZrAN$3 zZ=6Vw7@S2OSdkSVtWg;!u(5VE?7l)H*=-lw{0axK?Vo{DWS}cU*7R#r6}Db*A?x>@ zo(3eoLF(_hsiG8oEHoRYFyoB7-NVy#5j7Zi2RA9!{XOO%t;!BpqEa=nI)T+7H%XOZ zw4?B3B;l!l-R?8c%iJeYXZp4P%00PL>A;D{$6ve~`4=~Lh@_)uhoFbQPZvta92oU_ zr|xeH0#9#7UzYFrpqsnTOL1?9ikdu1!X^<}WhfY`z!2OD?K z>moNSw@+AH_-0L}4?t~lWMcx=^_ICpy-OrdB5BLp6x#mxHEU*82(O~=x1)YFD{{ml z-;H`ztl5!w4#(H-dc&Rz`&OSpjdx^LeT{{L65c7S6U-^UO*m(=){Wa`QYBixrVB@F zU=7x&SHTSlAarHB)^}pg1p9cEbgvU+zCuhYsvtxvN8ZO@)%MkD5m)m=jFVDery`ql z=~a9WQS%@owFv8^P5zN@I0}7a;No1X8|9#_#!=LAW7BIO)Q7atc+pPPuyyFiui{zKm0qtnc? z*M!uiGcYbdW1FTP>Np%HHP~4hloPfw@!t7*3>d z0C7+|AiZP9V)Sf7KKfCNhgJ7JqmhtBfA6DeB-9kIZ3#vXbW~wMCqq`qkyFf62Vs_< z&A9ltK{L;t=Dh=H?WtdrZg11r44x}*-=Cxkc{p_IZ4q*@1o7eDraxRWgf<&wk4MGG zy~o}tPfXr{_BBV=2{5t=q0H39vJRlbQCTGY5_(LjpnRE0!T?eWW}5uYV)2QS?Z>Bh zWiOQEH`K6>p2ChI&h6gvx#W!o611;bH)xh@vilbHwWp*fpkXLwOw)eXA`g*X(qZq1 z_q{Hk0fQ*(XrQUG-F=*R`~(8-Sn<7`ug> zWumE%2Lly5O=n~7K7NeI1`{VVYZn~SP~E|hJwRCSX9ec|Hkz{+Wp$Q@rTjkpRP--+ zsOZ8onuax}ae$7}hr!(&gPmw^M#PpvPl}q~0Y9lHt2Q{*3d8ydX?zj?686d`FFISS znMwDC_59XE?4P0R6$qd>qEzb{p=`3I9!P_#Et@%3M2xB(`M$SEgiT`FdE<`)CC{MC z*V1bf>I^BTTGVdu8W7a8EAepyZBTX93;JF(Z9|hf%a;& zYsQU7vM7?DoPzx`XFQ5w9ou)+s;Ba&&TlAlk`WR5*NT{qQFwtR2v!a79} zlOg_mYbeS)BxWRdGe8Zatm{OnxRHDr|5a+Sv^`r_G$es0N@^i?tS15g&Z1osvm)RA z-CS)UPQIXGSwI3eW{zB>m&Y-vu=M=*lpT~~QcG0Q8lP~#{Hi_B84M_2*aFO)v)%@HCsdFsG zsM*b-WiVP#Eev*d=w)V3-qVt+j$&bDhxJB$fTa5>n)#`{_JSlXV?R=XdGMp=)1@%% zeeuobE^aX8b!d-cdYj1-j}2Q@?DR%K{chn35u2LTz_Jz45rZJh&VZ-yo%ZSjZuv(g z-Nh$}=%>x6j48yD^Xoo5_{Njg9GnI;^v`_g%OAiO1P(O4u(@{A2%P8dtq*o}F#bCF zAXIidJa39Zk1GB(RA|Wp`ydUtWF4drz9lPYbHAPJf?aaXG^fD>uva?izaeCE0&sGs z+nia5_mEcv)WPn1qiogm$;rpO9Q((Vpp9_tnZzw2=XN#rV||ev>-r$pDD_!gD0 z)Un5ymNa4I9>0seJVjA~DA$l;R^YtSnKGo#uzpgJt8RL(YEzi{Fuh6Or|``{F;q3O zUCcQUQv;Y_{dz<>aYc0PL(SJzJeGX|KruT(J8BY^nIb7_y~R};yd8xNX?m`iNpiT< z^x(^3c-w&GUvcMtcC}LimTmv6vmH)lMYG*k5%MDnDmpmZKKdSQ)HD3IcV!-;z5;R< z!bMsF{lX>{9n%q7Crpk1u zjF80A#p1~g*Z@ocx|KdM@Q?XDb@hZSCDR;iwXZ6e@UMIpJQ!}_!5&AH^UMqLF~d_+ zJU`ST<>hondtULQeTaNTMG>F4q0r%R8UGe9+f$S~CWT(|h+Pj49wz=c?dKg!vwgy< zx7ZOz{B`bM_PkVsv-YcZ_5;(N*;yc>J*v`eL=}NOK+wLmIh#}_;?n2=Be$0% z_Gm0lN<9%jXz4S6?dw=ph~Cu^uyqIry7A|!DR&4d`|hKAQ_tUXPJ&J@N>CsTY=rpD z?THNJ9*<6!m$p}%R}4t;ezsO{(<-5WqN3~i2(B*wCW|g{>zhhhMu3AG5=WD5INOPN zvMX)d5hwE%akllF4adhd1elg3a!Ljg6Mlv8!%vDUu=}#Nok<9yV(Fl+u2VfW0_!l4 zcEZL!h)G|b=LdBEthlPI?nOhBCkh!U1?68hpJBA&U%sxI9tl$UeT4&MddQc40=R12 z=1T?A(sq-0bIRZ)q&%%>PF#t*R*@ooKOyWU54q16GW^Re{bUi%Id&nA*GhWN)a(TK zXHtMIEaP@0b9Yy|KhqWhOkSU*IyWfb%L7D33OC%BDyeW-ku)>dd{N3} zcvc?S6r0k38<*;8K3jhj;=k%bWc__vhC=K}qOw3ypLBO#_R3z-+J9c`5Ub=Vk?IjV z5GgvWjuT!gaheW`j3YoZVE^{+ZK)otaf>HZ!SKshWrI~v_MhmE%33WeI@nQbh8(i;7-N*eEJbk~QucnRs-^_-O}<_FlaJ$+b^(Mwf$46zca+2{cZC zECN}(DC8*}HxIA!t3O~P{EE-vvyx(-@Slmt+|1MTWl~z`4uzrKKD2_3SBtJ}`l2F? ztEm-bNZDw7V_u**x0R$9ZzLD2j&u|)q51bv^TzgfIT_N zPnYPsaCZZfM7JLzBGyq;xwFfIKxUQC5Y`Ph75<3E4`^^aMB5$Pf?CwaFBiS?NA!t? zcguZgkP~lJxQty^`z~xCsX9_JZp@`u+Inm7$DX|M`(+f*mUl{q}I>!KTLCar;PBDs%8_isGGbRC{0$ z3S<&?V-Vbi4k-Z~tc{V=EyF#y>t#v=4*~dnd|Pp8BAJG_WhDPxLajt$n!-6ocaRYa?faVWz`&1V4_`8k-@I=7(bKSt zH2dCxHS+U9hQq@6PqZ|AtO)@#ZB`MSWL_=&F|2PVB7;N>A_EFX?G%%mI}*?mOJeH5 zLPH%|n$YPrXjhV&S`^kE5-=dYf!X0$aaj?m96)wg)0@&ruWZ6le6dIU*F72>c_Lb^ zlbvZXJ%=4sFFz7~1+*k*7lvwg{wfS-o}3m~mjDkj+=tJI1NJ_6iMN#I!H3 z>*4JZGUPuJg$_YtY{U~=sI{lVAsN=!>p)`bgT0OV^-a-$@@oRX+8=FKiJpKjZro8b zn0km78Tqx)8yAcp$?~n$!0+#g*!iQTN<`dPMfX>=u%Yr4*i+KEM+G0QGzpHX!mqyc zY9|es+YhV5DmSp^enQ(|gpRaPX5}H;$wFBZ(`#1KSnuc(ulz;Q($ybG#5%KA){)^K z>v3jN*t3l4zlb{Q0al^kdONM(6~bE0>Lj`A?phi?jcSc%Fr|2liB&A-+Gk7#d3i*A zV+;0B^ZJdto6bn1g)%rJV=z6gfNXg#cNJMF2$FU&68~6~rA_x2IN}v`8>WM3m zOwtS{n-7rn9~!8{X-)z~RvZh?*fRC^OZY(nKi9rz0>o2;km)y1x!CVz-Hp<$bldLf z2ogri#J#CHsmfC79Co%P{uL_}Sb6w*g;jlPFmCc2fDWdQ4*z1bUb^XdUk0N1Ysie( z+nz2>9QIzV8zs)N?V*xJ(T4OVO{;A6RRuv}0mr0F&hA&=E#nYYPU2MRab1SShoWZ@p7r`;R;y#JEG`vg+z;a=9 zFp5}RM94h*_;Q_IoZb}wu*Vxm8zH`n!Z~PKV>7R2ka%>5m5^(o zuPF;Kwa7gXk)Uae*C&xPk>)AJ<$yd<)E;^w#Foq>7It!#1-Oy2>_`yVIMC`BkM8M_ zq<(|1e(J&kR@(=I0$QnZjwY%sTBr^ivhy7uTVb1KN!wv%KVry(p1V8ZH^B0z%oc$|g^BxOfpgvY500cQWgJQ^o+ z9Z?{oTu$v_sFwJQgOcpFJ>`)U4RX#NE4PTD;+yxERWSeZP!Es6 z_AS+7b!xb=d+)RnmaG`pjU9dwd?KUyXIFQ0_bgWnxYdcrV1wkdVzt5)+c)N|fKN(B z8C$0ra0uF^(1qz-;8K-b?ZU8E#XEYoj*vKb?$N|?X54QHvG0vv?lV8bDytM%|>N7{`{ohdk1 zks+HV3YdlNT!j?sRSamzbZ| zQs;<3tGf57wMuFc6o}t0Ogfmbw>A~-A5&2S2|00x8&kbH3_a#BXH$k zPp`MlJ-ugw z;lQX@f}BAcO}Td5DRTwkM+j}izdE;;TxanMw^K)=!)o`D(-lVF2u7ErRBC_bRJIj%sRYi^aC|1Pu_g+8jrmssl`Pi@Am-MQtjc;Z&r z((y5Jhik}z>FWNZ&3pZ__~iO8^NUl+F)jmTWR11he;0~APTaP52|j9;nsr_M&P!oc z56x6J?v=3aR=CfVZV%N|MjsTMB0QWn5e%47L%=vvX7!ZCexbuZ4*#hxt18K^#UJVh z>0U+=70r4`uu^}ndre0;?u98@creFgBpVG?|W`PK$FK#vxcN#Md5 z^W#?R*pm&InCSiW3j~yN?wIhQ$=&W(@kc9ur(eWYQ9L77Z=-Ol(~i?ep}*&HCXKL) zmaOLXnqYY??H!$qxYn)-3#U{F5mJvyN_wAcRF6=}gy_{K7&+cAa}W@EZW%S)>fjv3 zo@{;z9=00!>(EdNUI_8X{SC52G;85I`yQFjTyAMlGXWYDe)Pvg1t!doA#F6ZJS;8t zZmCIP0k;k^1s+WP5$S#arOCs-^f8TkqD*C(!VyVq5Md~nV7~tO3t$5AS)zgtp|%|N zpaZMRdF~_7Fj(?F@NmQUJ(yZ+irL~&zT|sE#h#~SHe=4!WBrF%IVYY#w5A>+n zt4RMy65^5$Q^MX$C&q?3GU@1q+2T+yF6zRh0_sV2QKRu(1wU6*m(Kl+VUaCNK;Q!% zVU&7I8J`QIH;NL5BU1#3pmyeiO;DTf>G z!H&Bf1DK|xN!qfd9MOI#`(?ePy`3Q*1~KEjm7ki9w7b?z$B&`H(8h|JavH@MD)lL! znYCbe66Yvtc7XV-XyXf#(&ixQmPj6+zF6{4ifE8-YTb>;D1A+#6Ix7(B{RbX=M%Rim zl$nbtL@w$HGM)F7C}hLXB%XgG@Zw&@nIYf!FU=kkIKV6OTkETnUxz#mNZ^00ig<8C zL5onDTLOt1vqn8C%IwTt)9UW@k zRbg=IeKcynt5j|3kj}8_EdzB9A2Vj1w3j&9tS|YBdQ0re&+>v~ES9SUJ?y8VADXXT zm&Te>uHT~CzbE;+;ayIXl{L{3|6TWDu9#Sp#cOeih`;b>8|swna)FSRWaMv=Y_ZB# zigArwGE8JO4=naNdVolJ+8H9@_Sw+nD3&Ady z^cowjdS*@6V$5nC6I){@^%)oJi5m(I|Gf9L>#UMhbapO<#>phLY<0I|sjkb;l{EdfTw?BR_1#jjrw=YY5%FFE~ zgSxj#i0^-k|AiiwQX*gYB^MzyAC3~PZY`l@(esgErUD(-5VUTSN3nzOn)>9HM+qt| zG)8*F+bbTJR-TQ+`??d92)RbQ%3q8b+MGn=iv;KaDl_s9(+t};V%4I z{Y#!!O8feQ6vJ!UZ)~)!-{G&Oj2lm>Pq9ebgII zcHrU|M~N|P61lxysmtrOVlIcf8^0?R>jLl~nA|c$ev{nXNw+1Z3h@2yZ?3--;FAmU zR-$>)+L0>On+i)G-#xN88BmuZf#c&<6E;KtBF|H7aq+m?eeyJYssECT<6 zapA+u0@D&!BLDzm0eVTG%lLAKH%c6D$@wOWb<6q-o8ABf6|wM^A`Z%MF`Xu!5)OU` zChkdQquuF1D~i4}yx{5)f|r$(tG$6I9Inf*Zgffmhq6H@lc^S3xV5MhzNt1@nmC`5 zCimV{`<{qx51C`#NGE-?dGhgoASx2DEBwn#`FMUfc=UPWgQ^F{PJYG!cUo(@!Aq9P z!piduV+wm=D3Rm-&<8~D%tWpJ@Z@K}J_A-mdXb|{g`-^8uRS4k!*;1(6+cp=2QSo zvEq@-cH9jRPy0)3tlb9cQWIYp#9>UOLJ?2NnUY-=XEtU3V>5l6~saIFIbc|+V22`eLJN4-6U4O7{4_u7M20(#(0s5H;_ftOg&r1<6yCbe* zRD|io(m?3d4?A(4o)LL4+D2RX$u~YC2cm*=@1+Wa3cE=_4ZXlsr@Y@E+e$IupZ={0 z$U!$2hNRGG&8U2Ts@X0-`%*jgWIQqG??~>l4_JzmY+q6mL*#;MfjvNKNj zLUdkx4L79sDCaD{l>Wy>C|!5&QA-IubSnGh{h-?D{=X-iGbX$;x4|3}SUF=NjOqj! zAOKmR{kD1u|FIB8Rm<^`2k7gtwbQ$xh(^lPW2?V}vn(01>0O+6$&m*um=^Jnh)Y)p zTJ(Hv_o{+D&JVt+NWB<@v>}PoM0Uaj!kj_CnWceWAD=*F-}RYJWsJsp7Il56&j(tt zcK?DLMqP==yu9s|1^Np`R_6$nmTFQD>Zo;R4$vWrP0L0$nQFr76&RDZ$Us3Yvoulg zFRM24yDvb~=@<-ufpAP3F~vsBu=eRqHQLkUh(? zhrcZR&&M+CKBRZR4G_MHwPDz6$%3x0lnAGdTP2+auuP@GW4{q*;7McyV*aOnrt3kv3)!Ky5y7I$R;ikLwtz1k>PWAF_OQVfT2Oq*|);{4`k9)DxJHTtca&lk-SveI@A9RC@6UwPq~(xzmE{Bg9AP zkOK{=P}hr4iGIKF+osN0mR>5WIQFWT zT8T>JC`i|(s|L{V;B5g{+Xl`iSD6^MhT_LF>i?870+xIIJ1#dGe!1_^ehn-7xYb?? zv-g{knYxp|iM&TaOIki9F0f%}SG6AxEqH)J{3W9iI6>+%PB2v8fj1Y^$jm(nBcS_fo<=V#r6r7k z0QXW3*{i(Ok;$r%8I{`usxALdU3ciB9t`Q(VR)5`0ph8b^{^4eUIK=gl;gRGub)R6 zFF?V!azr3K%k;GCeLy0GyPde@2M&cY;sM2MB2lSzFpCqY^ zh|DoZqfL~U%+ajh+Xg_Y5v!-W*#sL|92p?RY9JgxiVa_k?!g#Iy6&1%)!B&+&%%h` zdBss;b=9zYmOjnUOz6B_|JxZbD+{m#SSIIN;srNGd^p_aLpb(Mss-W(F9*ZRduV>4 zLC2tndOWA-Fu_pzP5@T|^Fn>Bx*O9%{j|F&B*#(TN$40F9A+qCylQ||{N8_Dk_hlS z&BrcYUQOUxS7vJGr8b+TI+V_YX6lGgeD&VNOsFGqq}vZ1qez0e z99zSPN-E>6HmRV8HaYF+5VWXXbsvxNC0$rF<2?r3cHScjl}P;R&Q*fk=r-XN&Rct} z1%yvbr! zm*|d=z1&K?0MAu_!{>9~a(A1LfJ~2AbT{KcrC`d)Fk9?)&JFWNXNYwfSShfU)nftr>f%tfAI}Ly5fMr*F0A z*ty%5?HS)B?^eqL-@7InxbKA*_Vy6&SBDqADXLoQkGT(iU2c=|N!O@@e%wi$0n@a)%a7ErCvcM+R2lv&=GzT zA;Yz;h}jWdH9I?-!js*V<9w8H`HhvayNf56?>t#{7`uLSv$dN#j`r(?*W01MAaOFw z?rxemb{C2HOFF8v&{WIT+@`gQVJ~j_zoL$dNTKaw?m}IUWj_M-vgD@a-@ zw~>aK4I|*OQYNAfn0U_02J}&xpp(>j!|WYrlAxH6<36fFl?4s3uPMh>+W$JI-=!pG z99cxC7&yk+KUw{FIh73nwCn0U|E!1ern-qc{)oz`O`!>E31SnA@#_V~DN4syxW0Qd z`d+rXzFZvpOPlEj6X|{29ozTF$YTS@x*omtMzSfjUG>if$3Z8*`I?DGXZ6^&kj7Ba z6WNtukNY@#$jr-BGH+*ATtp;G@dL-ZR)$@+bjM`1h)ZFm-s~?7!P;A+u=1uOxMTZF z=^rG{f8{5L7F+HqX96$7$Mt zAev!(IBb0U?_=vAP*0`VyG;+05cA8E=J1FI%jELCdshuy@L;u&_OF)somcZ<{iKq8&E}6@5l+)XPTJ#`no2EjF2*M1(lCzer z+!$i2!D$^g6!5+9?68iRYM%cdiZ16d*yQCp&IbxcYZ7&KONnY*o3Z)iUt&?Ui=R`h zrxMFu$GqSb@k)+V6W2Xr>JrE&PX&G-2Jp6U|F(qWn9>MC`K!!P>Z3dzWUlJ)sY)F` z3|x6$)g?NLzi2hVB&&M*ZK>wIIhMBdeCiQ%5OW2O?WMOM4GPmFto3)i!eD_cscbpv z2=$>1;VECbqCl!Q;?SpVl?Ev1*P(ynRT0a&FbLte(VwrrQT)8q<|7~0xIzz~X(o=N z;$Ay&{mG)0&p|x&@=NyLY&#ZbG+FUyco=|(^hPdi?u}Gh%)IV2KGKArvg77jp+LX_ zm{sS*_F6<>ft=cU{uGiLHuO%IN^kUDXRyKXGRw?oy)j>xE_{7XmhSs8&L0vsT){)W z;-xfu`c!-41^>JAC-TQ@* z#!RCWkVDQM-Xkq$LMthh593TCQLPXURHtLWC`H$s+ON~D^ueMAW=6eGm}-I zXF%wUhXCgeJ5^NX@44+nixqQHCujr^LE|deFJuFdl&Y%rDWrnNmlq1t4b!D)5_En6 z_X~9vX6DBbn>hH&^VTNc`}(|LuU_RIQ-M>Cta9VCP0k!_p^{s-FU_M)JBm|u%B@XI zk}p^p(S8|j3vtIQ=;SD9#x;21G0(>ThJ5cs+sX6C3BL*;0=*vXD;gJ-p@RCGOj7&I z5I7>rrzY##zApI>UbraZC6A(cn^@~&-m&jlwGUigP z(BsFcAQY8TsAT!~k)ehJ#Cxg&;pmKzEx-?b&p5Rwj)!s!!W3`wRG-TEScHZQ=MD$~B%|RqPNhPx0Ypt|{#0j)9Z$bi+7pP)0Oub(RMTtp zrq=7*c6vmxtw7u|?1pn1N-`(0J?s%^Ni#CqZ^k!^QZ-_TD1wQUu1|3}ROd^GrmSY= zxHX&eT4i(P!3to8j8vZr-xvBh27fw$a(M@jJM&zK(T_Kp{$K>l7s}gKS>y}MNB5fOSL$2-ha0ywbvvmoL zG#!ST71yRP6_1kAMh)WucEWkMh%rVS=b3a$0LNYbC(Jz^KO}{eg6iY;Mp=ot)Jb#^e1*K zZm78I^lRG3gHsM{2?L6)jo2!Lp%A(Q!B?IJ;OnT55lvK3_Wf>bM`ysw!j0WRV>R4q zPRFJlX^?_<00+9Y;bk=2f>4F3tU_aSWDcB^M}eabik_VGS$7hWMT zstSAN-84C}tgyTIB%W(vxVA_rZg4Q6k|VC?`2C6v-J?flxONYT&-o$e@HtuW$>-d>GEm zZ`e69T*Yr(C&6x*<=FWCNabzsmWrJPct)n}skzI61w1iLS!8BUjxbEYD^o4dLfO|W zkA6iz^QR7bTx*n3&c0HqD%Q0)KFquA#qB_toTAe`KK31p#ig5h?w(^*NX22KM~&!f z?NK?s+pebAF&j}2_*YjW(oe!r{0PUOTQcPA)iX!5LHFy04m?l-{C0~iPYAG2e4qy$ zRLU&vj?bKlY!MM_`=a`9MHkj))8;$kz>$XWGp+a)3Xb#IBh}Pw@6umZ*3J;^vP$*& zqo;B=_$ziT(G&(zV?vEDUp}nYH{&dKJ&{xa3Hndt3P>uUFQZQsNVlb;sPVn>bt;`&f8AyA%VFAg*u8d8?R35w6z0KiS#k zNzlkcW?!Zt%ksR0PyE>;R0+UZU~o{Lo-| z;eXxVS{qvuBUFe=19haAfWakY!Ko^4Z|9>gJj))x_yoOx#NDcfK0#OWOsRcIsGKhb zQLr#E-(ehdA}Oqsc5Jcm%3ef3A_wZ->sTRaR=<}lQ&DJ}*;$p#f#&JfRPV*QdT=0T z6_*#zo{-4k;ViCg8(-JuOD_!Ak{3LCQr$++om86F%FfL4TXV}eIMUSqR&}K*7Gq~9 zr&91@V3_Tm%Pmxe8Wm6De5;0A5p2zl3`>!rHdr~^MxpM#5%5(-TqoLhZ>h9h;S3H- zV6N_r;J)0q)9$UT)}2c{M!yyKs4xBzkku~HuzMOjG1@-7t5Y5u;6v}wi^F`coo8~g zm$D}T*M;a<_}ub4#J)NHI{butAN-Pt8`O^uR4tn#eOhbMI(mLx9(xcyKWi{t2ywEb ziFxh1O}%yCPa!-h8L^P;_+HaW=p4l%?rY-%Rq({4CQ^CX2jeXCD<&ED8Rro8J6{lh zR6pccz>UnoS8q=p%!8O(lRr#8 zYrY_8-d3{z`rH+tm0MwlNAWjHYKS@}l|Gk6-i6;*X|e0|IrD$(xGR=KxNgk&R%Ir3 zBk6|7$Ve-ePURgj-tNy_xD~MZO~+XN^U$#?TP0{#1?U%xz>{n7>0g-0j@HPeWf|@x zv$@k`N2ZlPU6lPeH96mSmL*+?=Eis?yI0unV2w{Y^gb9*TT_r zb0|Uc*1f86LTs;40>^w^QIUqM#Pq}#JYti*{V&HpyM(IwP}kHb$ump#ONm6qnMOAJ zOm27+2})v%N2m<-CQ*W}Jw@lrlSZJll$RUO>}d^AK}!R$ zm}hbZy&jC@eJ>0B>x@Y}-~|cc6<5!#1uf3U7OvbmiS*tugWf&Ez?my}) zZtG|{gPX9ytDGiF+&0O3g$~g#**}nMuLj5%2CwPI@K1<5?D&7G;h?%kSFyXp zV*k||KS(C&LI&4Vv~@>n`>AxM^jAL|%nc7f`~Beu)pj(?4fY*xj#Y_R6v@Y8vM#}t za+ZFkBMTTfT{%D>yfn>#W+~6#lX-R&=tlBEALgc_!!iGk>iON@8q-O$Kxoc-Rqy8q#OB7YJ)gROZ!rQ) zlu;mUS4EDYu$y*sCCrpWA8dbwCU3L?yH+#XkU;R&C?4(-=&-%0e+p+bsM|7rMKMRy%vPx z8N7c+C06@dHgJwu2taDyc}{)!9$Vr%9p&3WijX~Pwa7|SF19ZWT{`@^Y3g>gbf+mN z*ZK12f?!)hmP6j_l{D?&uZBznd>sY@_sySHgn5|x-TLl-pLGi7)eOsd=heR>OlHBT{e6k!^3HEEh(u1WdoT&{37@nR;i{0eA-6uWYOuxPkd457T zOW)BwJw(xU-yQusi|T%mG4b5*`NS6Za4Q}3(Dk?_XY-6A6Liu3qM#QX@Nj%By-dBd z`}#yW@Vb=l`ET2D?eq1($16vGaZkA&Rr}o+Ai@=I1TY5 z5;6Hpl>~m?rhm#eK;!Jz}Ob=HzpEQUZE^3d|PYf2G zGhlV1E)O%vqjE2^3iudTEOQc*(ba~I2)^~3Ep zW41rrGr}w)3tNY=iX}hSUr=S9-vTXf@dZ|Ln2gRRZkz7`__@w*r(lx@XL59JMeM=i zD&h~A`<}BN)0JFT1UI$qCr6aZQg2R;L)Pz!lX;fCjyKfINy!5fCrGS!Sh7z6IdVvpm9zyydpLO*EBW9k7N6KBzcM(^T zfpL-Wa7oF85cj7ii3NcJ(?dZ~yD>eWa*{&f(ukGXjLU?;aGe~Z`hfSg=oKf2wP$iS z$xI!IBnk=LXl(FJ%sesZkkJ2QH!jZBQ&z|yv$+838of{{%$ce1Fr+UeP1VugfPaMC zfJkk9u_CCcNm3(C4*iVoqQN)YD5OlY%-{Ar4iPe7(O~lLNO~{dRwRQnu)`|@;?uPL zyrCu+AsU`^L>d*AUjA`^tvxm#P0ShZA1*F$brc8wkRA}h8Vkl-cu~BdsD{2suFK@M zjO9AHy34WL3g(G0O*F3*hFDzta^U8Tu2_^xnX3~kR@@#mK&Q-cz6Zm+AQe2W-Y451V0 zGqz!Lv^>&xCU<&<^xTktpIOF`(U6dj2ZUhl_%e*HS)cq}fd$E!8V_&wMF;_Bj{jiF z3@Z##jD%Dgva~pOvLccls8}C=wR6`Hmj<_YNvFCf`mjz#L5vU4UF;D)i3EYKH3(1Y z`;Sv#SR}LK7f2mS1j4`{s3C+3I%KMQ^`=#3LmV1wm{K6LMXWbc z5=-tT5v%|*`SNN`ie8)$XZId;Na=T3OEzO6X>S_eF^8Ql0ZI(#E3b_PTSvvwL~nz*2ZFoX;O_2DfDjnm-QC^Y-Q8tyC%6Q62ol`gA;6ON`}VB-&YrV> zOx1Ixy6a3=cUO1Uy^p;WzDF#WLndS{I@%#dKRje0*N6^HLTvggnBgl1)8lyO`EVTe z1^}twj{8L!C-_pyIqY(0suK=jh_X1}zAafDtA&E=s+-UMDbxRO9hc>&sC0q~!6W}l-plkDi{y8B1Wv9j!t z3^megBvi*cpVSxot!=#m5~B8?-_#=05Ff9w>v~@gKIaDkW)U9?h#w!{?mn>fm1YBc z9Q&rPu?JDs?EYSl^nMJC`1Pgh^XRq*HUBQ638b@=M$cHk zwO!vB9l87++(*FyFNaw{v|aY+Q9lU{A`U@&iCe!d3MX#;06YtV%k%psI~^FM;629) zi{64PFO2qQ{O)elUtyQL-pRe+wI-saiY(e^F|DU`M&P{+^uP3)k&&S02HhU%N(Q24 zpDc1;A@2~>^>k3%inJ|^m=7_t-9ZPS5}^%TRUx6(os!~*`v~F;<#�cJLxR~v+UbO7L{dT|7`#GC&WjFWWWfA)*)wX!f*^fdVhyPQy%aOQ` zJ#<&XLs3~24)&{w4rD=jQkH;v8x4Ns>xjtF`kOXfq@LWL(?`sihSnu$1Y;uB)Yzd> zXzNL!rnI5h7~Q^Bx`!5aFJk^e`$E&GjgEwNw^iHhQFQON8%oO}w*EW5Y+FPCcc#~G z%{0gmI`D$$b9~50@cu9K$Uh=%Brm!UE}Clnw#z+5uR(t*IfrH3qUs23Ew9btKL@L+ zV#mx%vm&Zaf&S?7Z`~g-@bgeXBAm59H`)Z!zZ?uhy68*-IE6r|>XVE6=DuV=NFLJp zhH#>dP!?SWA
    um)Taq20{qVZEF4373NJ(;OkoYw&j`0k-LOxBmQi_ii;knqHSI z;lNo+3mrC;NC#^{8>ecFKc;#LLF7gbmsB{TXq2vGLV?Sg*riBN(dX%}(3LJW^dn?H zoXf^Q!Sf)~Hc^HigUN8x57&77q_3+&kEe=hzo%!NlJJucVAht0(DAh0dDz>#Tttt& z6WI{z@XTQx@B7DLhnVl&%R%VrXFGHq!7ZRCuE6&rO~E`DVzGXE>GJk;z67lf=qsjM ze33%JB}x_H+Kf#DB?I$5%${1*@I1}}C2VGwV!B|**FR~xKdZO%g-Lw6AnZqhj7plq zArF*++SbRFb}aLb@kv|x1l+!z;%o3LU+pgzyDF|B@ZThHt!X)r$^hEo*SK1eQU_$| zw<;6IZNK6GMc=X8anL!H9LaTUnB(CrHNb7I*RZFHJ;a;k3!i{TB-YPq z_%aSvi0OL<^gZrh>u5Y&+XqVnhmxK2u|+4~iWx^_`46FP(e|SM;$PaQv((^U$*oQ8 zz<^&ZuDWuC^yM-BV|T zzQb}*NQp8cd54VDqvD=aKTfS^Mu$HuQHkoD0L`RO*buzOkgTBq1j&_Z?8Ym3$W4!- zflEE;>yWVgERXc?L6-9|zNdUCItGS2ltzlSY+%(cCQ1lW(nAfq{#3O%KVKSxl_X1X zW;_gQ1cO4naiXT-hAnN56R`BgNY)yE`3l)6qT2_Ph_@*K(0fh{=s}))IiibLeBh~L z2K178?f#Bi0;@|Wa_F|gm`e)3FhjJQY=Nf`1CLYX`{}0;NiBCINg_P^kt8iMlV17e zKlaUu5WdnxvikH|x(QkXYJ&{`26NSsqF$tY^{$MOXsoc(yPw%LfZ1Xt=qEcDL$B;b z+|$gWv=&;}MUwggjPl2wO01;yxHSqI+ntrUBG6w(&gcEo=E9BzboRBX_I0uKClwki zm(q+P^YSWFWFe0xyGkufrjE2TA}C=>^mjH0!=i-suswCW&jnjTc1|HKQVGUWZFJUy z(7jkO2&LP!FVx}FeOe3JUSWcFz=uX4UTRQ_$0Z%LNgdKnnMQVyR~-qJ`fU)a&;1Qk zaj9a#O2!qI+g{!@lGJE2FCISxkoxA6O}i{OW|g@ zqw4jZs8{TI=F&0B)3bW<3x6~sl0Qc8wGSl$a`zotM6;o2zGV43zdGOM;PTK@vs9vuyeK76zj6#_?O_Z;`kF|zW8Y{gMAF!^d7#rljHdBlYVMD6pubk z`oR8XHNEL(?smh&_)`!H64K#ZLzNWBwzO z(dXaH_$in+1%e}*KtnFL*!HGX4@o z#f-~CNpID35kHa%&64^jIe&(qU|o1|C~Hy2M$kLX#B#~kV`hwav$~E*?9PxLbu!$< zG8b{cm1y3CSg6<2j0@!*PwwgY7dWNQElJUprS{UsiudhHuF6(dkLQAD?RdGa=hf+r zZ50|P_{eE8#3^A5typmXuj<=?G3po|<4g<=dkUBg|bb-&E07I}YoKdq9O9 z|L_9;4FjD1tcSlmVOn8p7r&O)b&u~~JICs&Op2%kRpsS+=gmZ5oI&MIJ;3cLG62Ff52|PXbx(M#M z=R)|SC}b?!QjIV6Gbo6D;i}26{Uk8-c@$V`g+b-!bjsdSX(ua}NH`0|nhNNjs(Tf! zFr5lVfJIKJ`Qi#bLD%*;ri)!ZRuK}wg`tO?Du~@g+Uf$I#5eK66tqA9_-O$+fEvnK zD0h*@Ag;<)B->^6 z787k}?b1xN$R1^)KWS~f8vYyD4<6<6$P#Qk3q`dYGyKC@a`^Wwdm7+6Mn7{p`fH9% zfP3~a{mcFpus~3haX1*WbKDtInEJ;=7-B&c`f#+qlN0h#%q-&VS5FGPt;aw9)L7Q- zlFHgSKQpHQ8uvm@syxw@(SpR^0Lo^^W*$KzTIoPoZ?ba&#cCqsw?%xR8PL-SXg*LF zp1l2s&<$>{95Dik-=(gyM9CBuGMJ5iFr?-;BvfR3z}x36j0oA>Z?|R3$)}45*&g`D z*bi>IfY+vLRX^$MbLXN>K`=AYH~`MLcWtA_tzS@W4<>@U4Hn4S|9O!KfxL5>wFkrf z^*IkqMnwh!=m>PMNlteF#HJ}I5)o=_`|z+hrwGUJu=L61 zXR|6Umq&%|jPm_8Kz7qw#62aO`voioZux$~0_X-$qQI&B;GHeFwHqYic-&{;q91|_ zu~GwnK7Sulj@nyDF6>9Bnm`K+ULTT=&I-18p|OXsdp`#6hLOE{D1WpIH0%yxV@0*o z-8Kv)0zSy*bm#56KgVPPw9zmyw;?2C2RD9Yk9~iIqi2ZaIr?Yj`snrCF;NtpY=zE< zwo$~OpWn)lsUncn@aIgS7QZGbm&fdEFh@5(CRK-aG&kqnAFcK@)rWR?Cll73WqGK* zZl<;#O8YDrNo4&S&IC{Zjw9lQlZ)H2%K)zesqP11^k2ALCl)*wL;GPEBX+|*DI7LK zmDACYCsR;Oqb5_yN8a2c{iz-eLNB+9Z;bjLz*+)M&^GuiuIkAJ9BPB86Jt@$pKM~GYImp2Rda)ORChqVCyLefH zSE`2pg&>Ne63^G~bdf+_r~=xuS4Q;j-cf0Ci38^PdSEo$?qe@k@zD$LHX5|@rlBVv zi6%?RT@X^*KI|4ylT>^w8=-c!!ZI5yJ@_F%`nJDKiACA4`%AijD=pV7@gOi9;)#9k z$Q9#~6*m^(s$=ISe!Y(I2SwRB@a)8WTK78!rRwp^Rc6-e{QN6{_zA6A-&BJ`Eh~4m zoyhd>(49y@#|R4q^knl8XkMM4vb>RrH6FSpzrSHn@?8Y7*KBx5nC^-2p7qUi8sM^( z{=H$4bkhRNMAps*O+5`SsRP_7)AQZYoO2`MjRl9DnJ;}jr{E(uG-gVNq{|-Q1n)K3 zd<4TJ*e5A_3dF;LX{j+K>AN%Z#B1{pEOpaYyMIuhDK8kx! zq{3x}*ix2&J`Z(t;_9|e_s_YL!$@=e&fDGvhPD@$E{(mU-Syt|O0t%x7hN@S`6qN< zU&E+aj82xS)u`Es_y^}LDYpPB&X*%Zi@m5d^%>f&N`Di+G}ZN&qmu!)r4$?4=p-Wc zwC>&>RBZxxTBJI3Yx-F;*$f*8)(O;>P8$8H#f+$hfs=)<4OL-WB$|L#jixm`%t+8+ zG>72OLon6g_pkRLWS8M6wUr|i6J~Vg10pZgUa^`HeLAOx9@Ot$riHoOq(2@EwE7A- z!((QvqTV>xWN-t~`QQ-Zoi&R@HZjfK^#5@7YCZ>!t=Ajm#TB@ikpQmE!6Q>@uTE}g zjH`=}5$ZtuxLQz#A}JCGPu0SFX*y&?CtV_8b?WRg1uR@snI7$E4}bGW>5SA9pgR^E zW$x|72PWd=xqU9lIYE=@GcNJgxq{ZuiP&*-^| z-d}(#hJH(d38202GJ>(;cp}%uP3#N7<~4|Seq3w6aQPDImI*^SQGcX#-N$1B-eZQo z6m*|=oBQd-0XD4!QD%v%bIKphEsQSIt-$z-XGFVq9e{?QY)fs2%BTsfNMf_o#b5-- ze88&M&c4o3e|Wxl>~>W&n8n;1@{{9adV1$MLs{y7Sb8_;8L19meynT$h0PX8537&lc&`HK#w1k8J*;RvgXvQz zt59i&z{fZSD(1`+WFvQ-O=62Yg@1`S^&yA`Dbec&(n_>T`rIJjScZWs)ZC7VU*lRR z;S&ooUh7>^>w9f^l4pL@pI*Q*9R)|Sr0uPVd8Mw1PPvpUVT~NM2-XDf`!)p5P#SA5;j+B}`%Ehwi9E7PWHfdaL27z-)uhd8aU7&Z!zIDFcaq?hCSzOCs+_^N$g? zuaE9hu(9R_4dI|s4oBXPJ+Swn=9P^K_=%;N@fe&&=c*e{gljXp^_JD)<&J$i}Y>g zmZ_$&jOVt^jkVsppy*wJv;D5#I*r421;&avK!o{V4M43{`=+27X|Wpj+SQ;90b{$! ziXh%W5ZXHz_fyzz9SBN@rvR!9j4Ef1kLh=`ceQ7Bf{Ye@dbpb%jIa``s-NifI#52< z8U<04FW@rUx%D7)N4JAl_sbpD`1W^dawhF9CZvs6YB(Ajhd9Wa8?=><;xMY{eLFtV z!iJSCU!1BFMSWGZDF8n^h(L|W%g~5gdbyd$hZ^l*r&;~s==0IuNie^d_)u^Hpk+32 zZ%$}A>GGrgNpsoXMh5~KOYeMjbGW14_YS6$DbvV+M#FC-Bp)cZ{@qupwvqU&pAC`A zWJY+SQXo%2a&PBf4HAprJ!|t1zETc!NOFZM&iyb|)A&u!6b0135|BXw+$L#)#HM?< z0d1L_N%jwvETq6(3WC6dZ!^5@8DzMA%$`2vEMDd?6Zc|$X9WOG;RDy;=e9}?EXkx$ z$h_zw4V5-zK*l;t6&cEDgEX%7iOhJR%j7PR9*Oc*$~O|_<2y@C8zf48I=W`tg5aS{ zQ)K);xL>k{!QE0f>sYcUyWdPMAg;k&uBU&qIrDjZrXbUa&~JkTvuGC*6ocKlueNKwG9Ks`$20QJNrX zp-j9K7#-F~V>$iAM^g#>!osN@-mQ)^?TMBQs0^o0mZu4L1L$mM0iIiS9VB&=!>$u& zUuE)8^MnX4cV|IuPh(Y2k=1hSEx;jlsG%6g-pIN;7X!{|cj~O#DA?Rx+8#21EzY8P z&W6#Q%a3XWqGLm2ddYV;?t&+HgRS)?2IE)B=U@+Sv;{cXz4Dz?ux`|FMLnP~qcZF4 z*Oj3zSrIuBVGZXvSJL6=0&f`1U2w|Iw7?5i1Zeok!#k>`P|s7;oC^?!4=3{H?e2i1 z9)t?9U;)7wt4j`LJ`uh>=p$!x%5^rfCFZ~9-liEr!?{a?6C|Ubx_y+TV7g~ACCuqa zuQB8zFg=dL#(A+QAUF){6UwkP*3nK8?x9fLNl3d|fIG&V0{^t?>vw6YfHxR9!!1Y)z+UH43hMu>Nv zda$I(3cSlkcKWX_r}i}~?HK$>%T*O;O^r8ul#``JJshkrB8aM^Q(k{wvCc$+Ma zT1XS_h;|nIeDG>@tmDpXKkAIEYwehw7y|l?!zPcS4OoXKXjRH}hd%?!;jGRC4vd;x zN$4xE1E~JOx#DH+=H!ww^;2(pg<2`Il}SfTijH9wiOVo%PIrxU7YfxFhBTe;O zkjvi9Y1v_ZO%A`Ru~zn3)E}f@)YTEd`okanS0tiZ-c2;32VoTc@?wkec5ey_nCQSUWKloGj;R}3?}dL0X4fRlukLN6NZyybw93$*Gp~To0o6| zvi!upkIxMSyd6w zWJ}h8Uwv>iTzjxcpmXR5q9l4;j=Gy4PD4(5WY8OZ8rIY;5Gb^*MZ|D)TTX>9zReDg zIfpt)?&X5a=AE|4@PME+WX542%!q#RktebeVH)<4u*K&Emn14xVS>`DbM znzMu$ddq)~%;xh0c6-}=?Y7B~5d<+#2oOnl8A~QOg-x?Ly#cF<7VG4WM#$Kh@|3Ev zkkp(-ue+Eg=5c(e;3FRmY_lttJC9#e@XgRpKJy*)(RF`UciNQ~)!ENr*S{G9?PUc& zO7k_G*IjjinVzZd>J6|}$1OfBUma9B$MLwqCc8_0ofz?Y^1gS$%zq_g$FYqlH06+8 ziFPUQQpw-kzuXR8x2&A#rai-w?O6nW{Nj=GdpI%Rbf31ZSv;@)ttNbIqlz&;x^LA- zuwd{uVELhy>^!{QC8QJ~DLW{d~6A2RnVB zU$(@FgTar9frQUYO_)AKbRj*AXNgblxC<`Le>5*PJYCStb@&H?g&(9R$3-Zi?=cKe zT)LFZJ*angte>?S@msit=16$E*$w_T<%g=~)zIA)h-dRP`BWee2l~Da*My+0y|+C-Gs-0I-W0~M`bZ~50gElFBV?Op>BQyDtg^g z7Qk2h)Zq>#yjZRPY+2yRkuc7f3L+)FOLabtfRSF=>k_K5sz)NItt6HQYM#7BDJ*w3 z3=|A&aL>u%3eku)!Z*vi-!vx_fac|Aq+)l;D6Ew2c;P8Lrc3M#-MRR>(r-V1UKYhD z6}Ukw0lTLMLFu#O(f+o?70JBrk}rB}2KZG_R!8)_?0j^5YKh6-osbNwI*P%`vA@d- z^$mSy&-Ps-trM&m?znk#?q^vD+d(mku%}&Gofz$Xk6C#%@9MGfHqz3)a-ZW{@|3ej6I% zjRO_fNt6`Vz|nohx0P`yXxTOxO>ljKiC}EF{44{`#RYh3gCS`&PhRrug%c4pv#Ks{ zRKols;Nr|O$kKS&?mp*d%Oiw~=jd_7O(hbW!PYSD{-W{_p&An+n1~F1NlV9(Fl;~g zkQ+q$*e;w(FNo)BDV@Z|>w$23Ek+6p2dwDVhdjVq6zsxSz!kT)D$EQ4+(iyv405oe z)8gZvFb@3BlObD5ugPii*kDTXqLuHx__%~wMJw|w&w)8Xo>s0p`TJ;ju*+k>9YXR) z{y-{jH8v!L89lG8tQ>q+wNAau1_CdhSNyXms{<+r2afZq#zc z;9&xQ6{Ac7#S!DQ>G6E0%%j^=<-P?qkafo$=eZ#j8gIp*`a9FyDxE?eXz9AL-#2=F zV_S+!WMzqRop9NR`*`F?7&@v*Aog?PY3ZS0xKW(|W_lh+qc18$&Jl^#D9*xxY+Oq? zaMJcIrIbby&b!{iV+b8jgZ42lB`@M;*p?-XZ^TcJy?zwl$T%zT){s;OntE;N zmfGOqhi3CW#QlgEtuq}8wC<>RV;^p9vW4EC9{rx4n1Da4ic24Ys~EEX>UlRqepFoX zl;m_lChhDBy+dt<|KFAv)j#I{_GNbh<6g}BxE;Cl^S258AR)~D2*(__WM+P-^F0-O zze*#<%;Or;j)0nPQVZgq{Gfb2A{R_rb7XTAErNeQO@bcQNch(qWO~@)zY)j|PqPod zd#^4EMwWtPBRDh|lY4vu1x5RFV`{+)?CwN-d*-Ev4s+u-m|tdm6-S=;kI9i=!1_K+ z_E3p$z`%V)ytJ1-|x-N z!l+ILwZr@F>wrjC{@0HG*y@NOEI-lEOC{57>U*eQLiq&bzGFOgMVH%Z;0PVw5H|qgB)w5DlPp^SSlRV9R_QZGY{L>Gk(|KrXt_@Jg-z zZD!@T1h4%UO^tsF3)*;W$0X#3ji^#sE8jsU_6tbA{YF6Z@MvU5x=}RnrHo5&KDS|4 zPYe&qhi10Ll$884@qC><=HNB|=!40#SNLEvBUg0n7Me_t88LZ~WWABu$)x09`-#}{ zeRU)7D?FH99qZZ>5b(zWB$Sq4hCFKm`^;?V0(ONq)a4$)>Qx2Rm`AlD&zP77JvGTJ?mmLEzcd3tst+ zGjD@%x%geQ9pn$?h@%)hB#hzl%guOAsS?DLP`tRVDw;3$w<4(i$}@jO1xSKXtP#=B zOFdmmw!cU&JG0O!T1QHs#^$XgFPvmNmD)40P+9*{KvkBMU%|?9jBK3}=PdXYFHOto zcFCx3Bq@K5bDkPw2Q?sV*-oaZt+rNE8(~1f5obCps?2%AmDn3faBShRhj9QW4~lHp|<#~fvxeFkfs_5b>v)amC7Bz@3B+AwIa zXOY6J@nX%oMB`Z^UNuD@&R5qkqBJ zj^M5Qnp<8(`w32{_-BPtbIS{73YFol8qJAv&GX^DEhy(+i4$V}0|{5+kP_~Nv9iI` zC*=>ED_I+~%jNy8%JwBX-v~IA^k)VWrcCqee-xf}BQjvLOr7{-IAgrPni&)YK)f}V zOs7{#1(~|)HYq)2OE4jGDglEpUm;;Xhk%XfWxHv?1vQR!{?Ggf^5#BagRsik+GSx( zZ|D*(od%_q``a%b1u|^t*+>}a>}+-mq6Q4#>`F_EW{NXVrf~sE>~%A*`S<{D55Bvw zaRvlXDg?V)v<;gzm%lGih9eW+{7;h!L81TqN!t=<52!Z#t zn{LEq-{JS5elwj3zfHz0kH{&LE|M^k%r27_moPGIGNS;bx-Uj&Bv@N{6XUTJT}IX_ zaDM7z7MA~MX{%4OC)087WnMZOMi7u{amFK+|3D~&t(szB7FBh~memkkgQWG`lqpo? zle{v)PVt;<>-_I9*x#D^@Vq!CX|u{7-bF9J+IQ{hS&p*rGEM%i($gBwi!B_^i(HEq zihXg^RePIf^j4S}tJy5V^Lf=)v~|N};<3x4o8UeZ@{Cy#SDF9wbLCeNASu5MH;kz& zHd`bRjUm;4y%seKxCO|h25*lsR5sGNws0@D!(MasxGSa>i>Q>0M=9zJTv@nuuuaat z7LeUSkzF!Ce__-Vdx`16K6%OY`dkSB)hC{;*O4~gX+Alhu%0#T^_yHo?dW&9;$`9e zh+R+2@trPPB*tn-_9o&zqyE|awiS?#|5?j`<9zWk|5+Lsj)a5V$9| zK;7X#bGbihfxC9M5*A9tA|>)khCj*Tk7c;Ye%|cv+!jC+kO$u1JV@VNQCAEuiOpuC zhJUUNINJWE$#z3%_YbH{w}lkWf10t!lgwTQ+pTl6n8$ge7h9E@6!HYVG92A*ahZS@0j{#?${Gm4Q55>i{`Ykry1A`>f ziY@^x#JD}R({ha+hWoCd9p;HUoxH9siwj&CkKO@W?9| zOP?SEa6-fp#S|-_9MqAcADb1ZVNt@pB2!bDoK9ydEtS`7U3cc!TN*LfeSb@ zVGgwAMPT@e>pZV){?9E|6n|!uuIa;cMnS6JjVtKWdftoc%%=3{t~Vpf@oERN;f%57 z$M$ba?^z;Tkj+H2TuSC2lmD8&GC2*M{Ga%Lc!mgOqyF0s7%lCx?Z@Y$<>;a}t`~)z zkjb>)e%U42)k37}aZ6iT+nfK8L^giSXBDCowv_kvqJK*C)Qt+NWhN!KLY=co@Jdv%D`qdrsE$QOYKvqu%&I zyKEes;Lc>3-7jc6&_tWBAukt%z?Du z$vao9Du$3elmb(IiTzN})*k$!ScJ#t-@{djJHa-!s)2GXalpG5 z|LwZuyIf;+*wfr&zQ&xG_wnm0{U&J-T%=G9UfL6@X|$a8v0J@HmzXzv=IkSr(9YQ) zdd~{OOr_oGYDAqK>U6E+30Em@m)Ji05g5bfT;AY3qbFV=neQj6y$8(785_Nn(dXK$;l%l&jU{d9eynfp^Y6`koKkZUuWq1;$|nPug) zeiKdS^gI~d(|2>XY(_r~YoeKpXrhsec&}d1pEu?gyWjpW+Q7e9w?#i4f)=1@=)t(q zDO}^#V5PYAFho=O?y#09u%o60RMc{K=oH{9DeRG&^d(eL-numHMablb2?jpGi&be-|NK*>DmVLI#Rl~1 zrTm=w`pWTE;#YTnHn$8lx%}5Mo_{lYBa+~suNCIvD+6x%5^d$<=Y}8W?Z$))FnL}W zZ~;wZQoWRrA!>U^t;Pl8B8h}X4gb1exl!0g!8dFi@~)!+U~+F>WRPFq z;Hi3XTCS!eux6fppKN!6B|*j2A%ROQ+981oZ|xjK1}M@>4j|hobmmd$l)$Z`{zPv9 zu2FOMuxkwCKb%J&K51N_cu$CTaqt*lN)+Q0_<0rORBdk&$`mw zRFmo|4yzLjjp683t>lPQN$SZjDS`$T;bKp!rl;&u_c7wXj9w10PLhA~zB!!$C+E`N zd0e70D_B0I0`~8pHe|P<{+LTkOohgNLhGadH=Js($^Me#q{V~JyQXrWdp|BY z6%p&YFpA8fWG)qRrUV)CzXXQ5X0tHO@*CqYyz-9RXig=D)*tQYorD~3+_aU;oQ{*b z=28kmv9Y3jmXe}hRCN-R2p1i-q$pISn%9?5;W?CAjh#%N6r^UavSJIjxgaI_?RXEMkWnf)I)$Sqrtz&Ar5^8nQ#-mgaDC?t6 zE8#F%T8x@Z{~LH#uld)^xsL)MX<0*_aCo{?yu^zZa}a1;{e93(l5)|-1>IDfauK%^ zKIyrmtz}(4y#StJfscV$0gv-W>y}>eu&O6rI=k-l_IVyVf?Mx?dV27=aymWY_D3G* z>)|QhX~9cx71!F@tr$o^Nc8~SRBv;Oz7}h5mZ~TWR=m52Ya1NVC7eNs-#`CJt~3}n z9+#(#hO~`1$bYt&x@{%<%>Y-*g(zQN7#rf~ZC3Jxg6iXkJbI=J7jWO$ zd#cP|-mH8xXMV5;@;+OLs?R#}+WZrn=8695bbGA;nFPJz)+>CmPZj(HE(Putn3Avb z(YKm#O)jTf{JGPinl^gMe=njZ0QQK7CfMSvNl|Qkz z#dTGvOuk@2OwPki6wltm$6rQA&dqZn@gHm%-l>#B-woOj((Bmono z+PAqGz?N>LJZ>*?%(s}cxB)FO%-d{$!&pAA_@>hxw>Bu#h}Nr|#B^~D{Huvgjc5{q z9%v}x!M;E<%nChPAHK}wgIz`uQDEixp`iU0db91MjEO5c#$j;`luI;#tDZwJbj*@H zf%5T$Gu3utT%FX-Qp<~bnvBc)uVQWG#fAJ;xdf;=MW)={HnAPV|PwPONPdgqG#0mY#l zVh1~z5rAY#_89MT8cJvJ`N8E2Q48_;Q6T?8)ISi7s|<)S$ID0<`441*YOz+&Wqrez z^xol7tB6ji3>S?y#H&54v%=r*K%5DDdpoysmz^-$jXBsE5s0)HZACfQA~~#?l6*{% zA%?C^w`2e!V+-Vl@zD=}snNGQ3V@&*dZU2P9Ct63sn<&}h>=!p8CX{8U9>?U;$Sda z-IOaRqqL4^$$=)|lUN<~Tdzj4uZ$Ut%mV(V&1r!uvaqSzTFF|!UuX0g-aM#1bUi*%v~wD{t8q%}%(BAn0ubH$OWLOO zR4|~J@44dQ2Y3jF&@GSf=O_F&nONyqr!3WJ3tM%uqr4V`i0F0{4Z9x6^KI%H_cm!m z-a8bU&D;G|5>Y}nN^H~nhWH8qmEhA;6h`uRCo^oE+X~5UZ8t-qRh)%5JyJRL`tEyt znd+W*&ehXEA?~&yYVb8_4);y-=JyM^EA+&MarfP)iE(%|^D{Ad>OI~2`*%D^XxF!K zp5;2y&beB7C&1hyGWzfAbcs>g?XPG1G>BhPWnZ&%#8`dzZBtGC4>OhRz<8d|U0?Plfy17&DjfVPs^{Ax+P1?+W8~E2-E_Wu8p=mE#FBgiX1VV;c zfP&liW|Bg>aEllIeUh@?KPGHQVp+oP{>8AMP1e3q!;@DPn*}RO6MQ2EaRF5|1;i~{r_jY zFAoa`$Nw|lSEGtT&hfzOE%KILiPk#?4hLpI^Tne$^cftR0qGHe;>{1|)jOh|*{PBL zi;JS$DW~)gjns?WgM${PZKI-(H<$d6#=Ps@v@PFYl)smGz8|w6S9%{0yLP_4K7W7v zeSrNz9hX{>eSfO+d%mtaPx_Moh644{DfIU#%a5Y|^DvyoF{j1elMjEwhh0A#t*74~ zZ`WZTZXa4}E4>oecGs9j)g_WJ-=XDoevZ1ts!vgd0CD6!`uE!WWgto!|Iy4^Bc?a@ z{T(1ix(nWQ$=2G}i|~8Y<{ZAFrCc`ZH5jKdcv$Yaq@ji(_Z5{F>D%*5X3luqCH-(H z3$lJi%?{eTv-CQf;`Q(=Nk49FbP4pQ5g9l`qw2=f#kpw^*QKfJ&UL2S)GZ4f3sf^w zKrl^$tM#k{pmy0YDbe#eMxas1#cNZJUPu1!5m2NrXCZRl2w|k-UG^zn_L#W|3+fBX zU41CFbU}v|XawdalP>cV1uEn4HwMbwwC3Fz6RG=3w-jxg4X7(O^NN`%&wMt0$|9R&WNCsOtV! zR(xN;M2`3fz4hk*M>YHYNj2-s3P;X=p|fEjh!O4mAJy#DKdRYHiPd{MCbZC%$lED& zAi~F^!u#tpuiGfsVYQbM^9i!6&$q+^3PS+LA~oT!EioiYn{N+w<~yTxkjCgakM>_L z9;mI<4|yUK1iAAm@OCgzz(O(_1;NLKWWja{M7`pp_H4d0gJjchZg4$e$UQ{E&9WjS z;NU8X{6)f(Mx&;-y-T3fjM%~nid484bO%X_`&k&*7K~0t6q~Hq7l+@Kl?Q5$c!Zzl zs&V4Gu8p9J?Zt&^ghKHFf4jj7!#wzx6Fe!8gTxsWxVeRs`iZE~=p))>h5bFc>y zvZS(|QNt~7UJU+={?(JYV%cFNm5_mi$QU$Wk01O_f;Cr=&GwINHb;6EQ`#hfIq^?F z%xCm(783=I9!UAh{ z886wbxuj=J^r8lJ#A126HAz#W{O;CiOZzwhyU1SQ_hOJIL7bI6Pt3-5!?4# z^6&t1#{Jxh!YA@l>TCq^M2B=lf!!5+q`ODbiB0S56-5yPJ7l8MqrLof27fPm>5qPUr`Fj&bKp6n!Sv9HgTkx#85qq0->lm^Xbmepz>7s8 z=Ao{xpj-=)8L!31B`cNIH&XPWWlH2t3UFyf1O zc6chJ6$`SNDcr4mX(`uGof!S0a&NG7?=3pVPDAh|{712og(ATcaK;U;JBx1_lSDx@ zH01=0uT`{KDtJ#U^!|IF$(@UFLcIemI z`6@&8F93LokJMki=4NP&&15d1iDI6j|{u2L~K^$Kq>|n#MHC1^p5yC@)CD9u-QK$&31MGJM_ow@^yhV#bn~+ zN60$ata3-*td zIItLL^$93vf6`I2lsM5*afZK=(n{X_$EYprNpt7Uec(6z>xBa&sp%-z^Z?GGl4xXZ zfc*J^;=`1LA%nAep%k<6#|0`LSXiWJGE^d$LzwSCL2{CR^1A8HRZ6k)rt`YM9_*^P z2Q3=9C8Y<|&iOO9%;jA?*z`jE&C~v9^Qi8E(pnKzRpnPmA7L33H3q~_ChYP&Gzt&ZpFmf<7eh% z*aP=-o%qu-?xBM~#9`soV3CFUAF^O5yGl1T38X2*B~AcLOW%<6)>IZpu?l8w${$ux zSaty1PX9yEI!dOTr*$5JWdd1}G;;?>HDAww2u-G`@%+63O5~OyQ&%C^WCD7oX-Y*D zKkx_>(W=&6P~8+|`(1mt7xy^P zVdB(YJhsw!x5ScK)aV(}1vy4Np|AQSmt?xihV?d|tYppr9X5a5NUYzZexE~M;C#zu z%p!S#kC8uDmI!4DRhuZCJ(9s!Lye&!_b|K6N8cwkcf4A_PMvm5%Jlp7nh|N zn80v!o({2^N772orMuUi3opa48AHi!I1TCAD@HzQrSN5P^Hp3}QtY_?OGUu>3$c=H z?E)&fg`zmIj{R10uJl>?XKn)HBk4H8l2A1tiVtCnrz4D%9i zI_Yj|y({j4!?ev#RvkqXI^(8yGtNA!EJD7aDqGBIp@7=-F4UZ=le0Wt9Msb*L3%) z_3XW?pRKG|c5lP%RQ2}FLJ$)cEzpISkz2M3@l8qI&ND$>*z)gK0GJSk50QqxfS`^p z6KQwKh2jlSz1kv35s3_Wt?e@i4Yq>_x-m^676Lz%C}0a4Oy&s)uF(15RT!Lj?0rCS zH?Rxw2CRS34lW^r^onKCpLwfJ1o3t1ltW-+bNkthJtH<%l8%+G{qy%al+103+!guM znDf4g|JYY)BbWEB8Z7N6kX7+qcg!<;-fCrz5__2ezRe1Ejgy7gsg;N}EdZ2zH`Z|U z2HqGudPsTd@`Mu+-CAl_(2RAHuhlUD3$D-lHi%$!{aQ70I zvG{1{#E^RHsDO z+A@ywy{GgEuLR+a9L}cTNDWk>yH+Ytc;%x+h0~cujUNkTjlBTFStF>BGTv;xy`pZy zZC(>%nX10my5XTvDKs2{Ry46UNfYMvKzwNXp!f$yj6mOs8-gs-y<#bn*GzH9Fk!5= z7^Vo5`?4!o#-^|Tb(lN>tQmG-ex`*>sr*ONi}SFYzeSIhC}DJ+Y&>V@M=m9~C6Uif zE0#*yjX}xie4;-pWmY|2g83Oh#?GKI;s%pK*nQ&nm-dCkwWxR^+k>6?XHFO;ZL^jv z$r5fTk6D<{1IXYTr{O{8r*ACud*z(pM%C|5vZB&fAkagG-!8fAC`nx^G)dy6s7qd7 z9Zeq)PU_FhQsfkKV~`7a?7bhz=s5U_OD5F*+34UK=ZtFCpihS1)YR;O{@4*SsjpD~ zK7q(o>Yr>aqPBz>PgG^;#sOT4TifhgI7#2FW`%`Tbzw&b)`B9+~E)Cbn1PIW@=3?)quIC$oqHW zo2%A8ZkrK>I|NUx`%5zezR0l3t{KV;26VjHjiQ&vu}a3p0K$%wC_YFgj`?{X=$2mn z$UIstj8i(YVNUM-I&$H4^G|cOl&@%#h}#}gWB@q@5yv5d+`Z#9gz?|*wEhk3Zt$eu zoBBRPyA-~$UV{9+_3;qQ*45%!B?4V-X2npKU|k}I`jv+ffACV5EK<4+H8ak zo&I~Nffe1lu_0D}*`~kf%=3v_G`~Bw%pP~^A}T_NoIxq`Vb&JreU3Muy1?HfhUEC$ z@-5fTBW{FRNxh42-c6{;{5|l(uKST3NAhm)jl6LJ=P~ zP-T`{L9g>p4uZK!L13oEyH`D3a-F}W!S(hz85<4pk+H5p8o?GI;ZLjeLw)=u3wH*x zTj&mx2i!c#t&tKIohd>@0GNF)&9p}|M|cFeqAp!R2!)L3w!OdH3et}Wl@eP4d(SaO_d(zygcTX2pc!CUQ} zdijsP%D8Hk9Bh1@d-iKz7z~{~`8|lruvlT+e+AG_Zk*RVz z8~crG6)F=okSUnxHFrpv!>0}|C^~4AxIR|tBOK`Im_arPdp3$|RQo_jB86ps@Dfw! z&Wz(C)}?l(6!&&QFLb~@7oHufUWXK4dXU1Us=c*L{bHewiP9{f4aOiHBGi_7WFTwa zPmWHP9UIiLv$iU8t5f!ilVyH7f@k?yX7?lr#pn!BZ$78015} zG<&jN@EEEB&NO}lCj0j0TlW?}P$Tfh*L*ZvF>_z+sMGT%a+egy&k;Ttl8)~{nT#>N zz{8o(N=H8zZ>4H_e6cD?v=lmVY>t0TM>;Kds6bVdB6-?)Z?y@q8HsqbTITzi?qX+p z-9lO@2osQYZsvFshceY4ASi`8jR__b{n>C~KxMy;ZZ1ZXDvp!y{FG<@^>|ow4-*_t zf-IpSjPyf-6y8pyFKzr_ul}3WnshTz_ocv* zH9&e6?QZ=21)B7Z&?+hAdqZ)0!}m5n2teb06ge)@KN=x-gr#yMz@=`b&_fj4%qMhu zjz&;R`-o}8a=E!&>miYfS-5v3t1<8>x%(uun@X*?=9(|t?;%SPWwo7R3kPl_^8Xx} z727{;&Eq%$9w-4KfqD)_av2>s$zsCcr#a~Qs}^tC{aC&iYhyT;@rMvFVuCBMz8A}n z*qg@%a?;bO3*%{>w#(rERl;$6Kkc`h-vZnlTb8Vh4^^2LY*`_dtqL3*qn3YE+h^rd zl{zt$wt1ZE61#9(NTDs`ZZOL+bN*`m?b3m|Z$1(D%}D0sUhF}~QrS5<%}kufDBw|# z4%CWA=|GqL#aZv?-J>1@snyzUXMBZry2N=fo}Ra8ic*~gO^4XWt_^Z>ZpKQ%`Fp-L zyQ-gZzpTHk*`fleWlKDq1Utx$shiEvZm)i>*~jRz&UWRixnSGIsu5aZ2Zk>~}ll9P-C-F}xs25Sq^M*{;>}6z7@|OFDIw3Md z;W6A`MXBvWU-=*3AX1=pk#Rvvj}>g5U#I@^8C1n%33r|A7d_cLFSy?-?SB`V-Z^=e zt_YcOnP0?7(wy~o4qowJGe!EaWL`^XEzkM%dp;3B&b6nmv=pW`XcCuO+D0}WB%C*k z6KDG1_5rA#Dc)-xYa+hOH|$YN5aex$pMJI$!BH706G{q<$zyDuNU8*5G>yX2$3lPN zV0hCfBN?8R4cavsOUS+~g6J~GkVn1HnVM1(H#|~Ic$V43c>b8Z^BL@9n%H&i8nvjG zBrvt5OXl>fP&jK;8>#D|q&jlt=Bx84f;k*as{ayS>Eetc0XI-aQKUPL$_opb>%TtS~5Wi zFtHEc0=EFqa^pD*8%8FIsU_2`6n*ZJU~a)q5>jWTJXdrYl`!DhuYIE}YWffDTb8EW z_PcY=D22BIY~!Zh42wEtYPi2=c+LThxycew(ogmG!oG#N$B3eKgZ(S$D*C!rl`-nn z{nn*(Mv9$<9CIX=BbDW`LFxrrXw8w2(`+9D>1uZ=p?X+=SvHkV_BlZ&{fZTcFV_)` zv2!{XfS&G9ej8OR0cLV~+2In}FY#!c$>KiD!4aV)(|hG<#{9y`H})_sRz5D_ zC6|d1t2>l*L0fWC3$EOD#srt`5Meu87w1_KIQCSfj>$s8z8z9J;>QG}G8!M=0(J>N zC78U?&j57g%~%<%MSVtn7qgBT)7}FE>IrQE(QzKo%`(gG4a!n=Q@s5$InNq=1C^zq z3pT!l;i~cW0Y33Te!eSqsS6M6Rw1BoF81Qee)jwAAzz88BX&ZJI)RKrbF82DRb++( zc_td;KO$; zXGP&obmRJ2&stirZu`3|%G)G_Q`)iIxzXgNgxYGAm=5DM-sge*1dX?6DEa8Q6b>|1 zz2n=;{^(qjD&h^bQ(GOREKgyIe?4$*HpR$xqe~74c{-j|m|wP#YbO@gr%dnYIb!1O z{F8z-&rYcyB2K?uxYA$13({y!hYhaO#>lJK(wm2w?aWg?VADZZ)b{f0^T*qLn*F?^ zf56Sy18<+~xs@3T2sEyt^n@qpW8{q zQ$7F-dytluiUZACM%RV08dgLkx3Qa2Y+CgRl@9ev7u33yv&#WdGmN5tVs+kLMZ0-W z6ADT4Uy_3ZP_8oyyOO^`K6J>}yCsA~7Eamp6_dE`vJ}pt2)1R^q0cYbK9a4yZz`wT z3{C7hfkJoLMp@~0mnG58r8g~XWd>3AH0O+|mtiK`_)td>r?xUXB=2fx$s61usPzqD zM@u(ICo`&9xt8kv`eFjcX}t{7EpK9vaIXgKo2FPOm1^qAXVs!6`!|h;ipf$U)z7|o zz%eN)yGK3982FI9)?Ub{<}f;FSxD$7y}gu2V)`vYiK38?07j5X(zoApc}qR-pkr1_ z_M;u|+&T4^X{luBuyd>IN%4tq`Sq81G}vp*+nP>D-?lIMq;esF5AkN+0+ADL7L?az z>UoS^xRRSDl6z#Y4%E{(&vk~}w8y15SVMFEm>ya31FA!4^9OWN5J3u6REQ`0`qO03 z82#bi%=e_D?u~j))|Aa&s#HR*>+#dqTXS-YU4zlyXCKhmT8R>mu5{XJ)w)PH1=FmQ zn?+K8_xMG%(bJ`a<}!@>ESh^p01{WqixmBOX7dCBDn0&~T)iQj_k)%5PrkoDN= zBWmRF5t-vv&!h(OsA=<9=;==MC|FXHbtrx8eg?qF+V{!c)qT^weRt;es}(P94Xe*T zZuirr`Ps=AJ8Q(|h|p;opT2I8MK%6b6WoHSBI_&kalwM46$JMudY!Z1eC{XNAGLQ|Smdko6CPd=i} zV>U`l9!AtIn=%ih9A@7e<)X*_pzk_y>A_{o1^cmp1|~}S+(G<)%l0XpbO$I#euc-R zHstw|Rr!7ommQY~_Hv&t9WBhD-MG6>oTDj~;KT7u|0CLa()P~moMN1E0q-MmPy2dkm#Ps&1yVk|a1kc;Xm zc9y;;*Ei0F_u0Pz`-;eKpMDd=05+C|4y;B6Lu02|!t_XA-Xek9lomz}ja~%l;-mf6KeKsJcg1!NY4&J?Q37C2#?&9Mi53FkMn=o7}CG zzx_+hlxLOV%W?laNA5ll$r6!4xMmN9j6CuMtNq~{24ZHtTuM0;ybS4qZ zGRRW{q+A#Tx)Ye%ih0NZ-wHjZwN>liWsx%umn9Z`QMIOgEH75QKuuoy>NvDJPCU@q z4vIFIkIF8}-Il1+#wC*}2|G^3{?%np?GLwJNrKs zMv_wa`d3Fx^-3U@Fq>odyx#Pd-U#JqOodm;5Y~omC8e@oST2a~XDwa#ZLrpN1g;Qw z4=3j`8ZRgF%em>p6>%(Whr`_0@|lquQT`k>y2hd}elA1_{3?UM$C^J&R;Jmi_Rl3R z&)Fow>0T_cbNi3PDXlB{uN-#g-{XK;{I+qGyK-^0ivQN71_z$CHEa(65T_qrf|P6v z%~x!=y*wCc$RM=!1Y_=ud_KVeK8&SK2W=6w-0pT)ImdZ{Y&SD+<@!6%5Mqpc` zds4yFEM>Iv8$0Ir>5Q68lRY;08mk9;S?Iks45#qb1o68ohRe8HR`s5t)4CGh%p7}k zC0GFBfKErAwOlQ;4ezHU{#v!>H6IjcMXJ#dQ7y9E(O#J~!>idEtP)8lxflD=aN}e5 ztZ}V_hfka?C=j~Xumsn}8pGXhF6~}L9v*|mEoT>C8PzwD?OQtnE`oZSUmo2gS}G+c zm^0C+(&Jo=DuW_8^$={~h=iMg;MQQ@-1KcHYwT2T_Saz}?Uuno0TqHL+voywTN!#Z z&DOx_PFAJq1!T&_hR7;H65wqYpTP;g1`$5ftmQhDx8Wb0B4%T@XANryO z+hWjsq)M%nZOE1w^$c`T0r=1!$6+lXQK^`IenE;gftC4Y9bt5C^K>V917nk^_X z$aO0##UwY|s_VEgj#MyNn&PXgj5ximyAOC#Vr@XEK(@XUY1xTA*({~Dp4=P*?W7_U z(I%JC2<@3z+;ScJSPc_Ak4DTs7dE%pUAqXeRP1fZFB1B4@?FRqFVt^=BI!nGI?cXq zvo(L*(X2HeivTw^bh=*1L2!;=SZ2{rPTE+otU^&MPnf2+hk;#e+K@!o#1_&Ni9&we zfDO|_As5aK<)FmjCD~SniZM_ICzhPjJ1~^I6>W zKt#=vV>@DCAEZyq$~yomhoXE}QYkB0;rw>a7@_18jg@}&dJY=-9j;Cia3 z$Ba|;d4`K(>eVBMiF5TZ@&0^Pv)$+{tER$eJiMqGr4g;A5v%TFr(&{0?#I}#^xGWW zHjCgds?Uu>eO2u^vEIjs*0f-v{iQpx;sZMs^C$+hjKjtT)HeOj_-u0ARJ-AsQ%9yp z6XxXQC5tu!dNTE5dL5`*KHmt>rTyf=cNyq(8K!`+ca%7?;b)G2y!~6`@rSs^s6LtTjSaD^UPL1Mfx3m zcYT>7$U2vbUdE|1Yv)+@5Cnk*y#F;h53ZC zMfLVGF_Ow46S?S`DX|0Hdq(K#8+1Enbt6}96jNEqfo`9tBsz5yD5KRc%;*DXCv%r$ z6kE_V^TXXOuuh|+DsP85|IZ-@7@P4UvA+T)_|JuNGfDmV52Y)+0ZT*5*(7&C#jG+V zSN{W9z2eZ=Uo|liJhLwELhb9m+U#RsEZ=R(WUPb^2=WS zXo{57Q&3J7-u{%o8teVoiA@Z;K*-W8E0(_^u4c2;p=iuDMewH;emnfV>* zgaeJz=p)1rzU;B7x?_Rs$3x#-K$V@wwsT&4^NK9KV*56s)J@Di#tb)t-br=WPSdo+ zL}dQ~<*eJhifx6Nhh`inD-`slXlZ@4mGPHm_=^%%16FI6D9B;v@-~-+eC2?ZH!4)G z)lM?A$JI%ag6Yx?^%@h0u2sgeXG&bJNgEQ-;$Ui^MV$*?3yEPLrb${B>KuNIy>J=YEbC{f zlvt0qYDG5&>jkF?mPEq9X`DjRmKMNN6F5S72)fo_<7raD5?vmtAy+@$apLxJN;4-7 zn!D`Ea>>6;^@XBA%%Wne^J{|&h@?)@OMY>R;1tnH3Wv}%wx_D&&KD$0lwCYALsyJ#?qZgV=37rQ2HFkUW zcKDD5Gcm;RCbXg3O`#g+F4++%lib!@^)*&vh$`0Ow6!-#Uw&6{f~jbejVYB5X6EGZ zObksdv0O#5Pfw85{sMtVqUY-4E5XiYIH^Y;TV&!^dT>~zM`?754}pGbBtXSp8pxE+m(~)g ziQZ|>p%0*pP9K&U279bOqbKC`040W9+U^^T!mA&Nhu+!r2-tMV|8S}ztu*RbYCzUY z8<*_LKWna1|7q2q(SriZh>W{v3;5DRH=XxShe;XAw7DLW)s0kz6U{6%em>XPl1}91 z_33N=1=B+H#Ubb5oG6D&@X}pk@O0Jq_Q<>M3WjjR%!|`;xn!>pE_Ro`;NXJF$8LGc zfQ)YsOqqqaw)*OP9rh7oe<$&wWBR~n4P1QF%7*GOUjZe{sM=%TR_EQhM~yWqPOLdV z#{kmwROh);KPgz6C;_iqp3`_i3i_QW^<7Q?X zBuW~5g#comzURPNckU21yeINpArtD}j_Xvcoz3TSiG`B*nvVd40i&3Ka-J@o1HYr4 zQ{%vi8Dph!A##|WT~4eE%bli_B!aGNGHgKZB>#Z4%Wb9QIiO(Pb+eQ@;{bDz3$mQG|mSuPN&tG91rAvZ8fBphCl}37FygS_$ zv}4v@dfV@ZF>7WE`BUnPIoq%RI>@j&LQig(aWmVO?7Yarj@US^bd|Vqef( zW1I!g$V|)dXI}88FvP=Ti46ltD^b*52PJ&@tMEs_@Ubo-lAEfL=Wz4)r0s!P zMf(es)(V7q;*)D>nxwfnGgr1?K^qdTBz~aA1eFCILsW9T=k^5dC~?cw8O&?(yqIS{ zB}XO`*~Hg1%bYJs^+KLWr>-35XX?o3d^=9cKyL2T4AIQ@zY0+sknAMx)CRyFh;j~U z2}@kcvTbG3*)H$pI8F|7c~%!9iyB0{`n#ksWa&(?vR&eeUMBqU$6J4)O>D<#zl{qw zkbM-8wkK2Fw9^TZ+98I^ic?bh5K{)SW#yGg!1P<7muBw^gDF?2yrB;@I|y7eT`b(0 zPaQIcvs{Hun;yZof}B(#MnCP-CmGiR2?MYyGpw)2jA;?StU0S6yf8kx%Z^ml=8_AUctaeYP_hTj;I!lfxS+? zzFZ=ct-#XJ7Xg!92i+$^qW+IREm}xTwtoh^Fun>MBXUwTn-)$i3e3ae++LmsU!Tk4 zUh@$0B+#vytZk7k1N!g2K6z}wjI8FrmU%zj?eadv+^vf~zZAZ1cE6kq{_FxAAUFzx zF3t^KFSZQ>;uC+o8UF_ZcW2X4#_vfjKXLvxrA|)S zuliJ(H+O&s@MwrHUM`FH--ObsWj@6?Z`f9t`n_nuK5yeXAydkJw{6K3pnMO`8~#M^ zl0ZBjqymLu1Fii;rHP^rWN`OjzS+5o^z2zND*Md-AM%ZGEI)(vFa*A)j7V=hoFGRU ztClgxy}mpx48p?waa{y|CjtIU+zW_iD)8Ku@JO3*bYXA|f5sz=K-x-@gAdK|6`F8W zm59~D7d`Ru=*gFZUh3JI!y1e!lD>m)ZAP(kM~#1TA0+pK+57jP#=B)Uk${RzL^GA4 zP^aM{qj~yDKw`Y$EqojZ-Il;45ljZ_-`ndil71X~DM|adkbru^K3tTc^10K5LbXt2 zIN0yy6TXqHEV1KXYUHEKK>@=rv_5p%a*8svt3Yd3h01aNG0$&~Yqom+ zyV0}^AL9^*U&SkQe+ToA{AX!P8sN{wr{(L^=#uZxh@BC5b~(YcIgk&5D5hIU0^BUT0#)_>g{Jg(YC%mE!&$_uteLl;FRZBLJbahOKZjHLVAR1bfkk!br zcs%bJ!lPmL$mu30qHXwienq0$xhpJFWkQnV7 z0fq5ro?m2^6R)=CxqJe?WY2{9fGgLXo;L!bCF~EPog77X$T6PduUYx3%%*(KjFS}? z0$;}Y8Hu1?A!P_RrGP5IbYp3d2EdBsO4pugThkhdNDv7>1A{33AZ&z34 zJgs*=PH&Bcb>DW+wwR(7lcqUI;UD&p%hNS9LM0JFfzJ0(>h2?1MoAtuL912N<%R`LD3Uf4+%O)~ts@7{=*lf;!6R=9X`GTpa#%O@W z5_20bM9s}2Y$JIxIM;HOmSmjtWqTSDmTafS!y+soP}fM!imVBsP;z6*WsW<2i?QpN z=CwV#h&939WhW+lSBHANF^YGn>dUXo-id4o58w$i!cRwziQLucjru1m@gcXe`Y#i? z%oY#cgt^$i*fYI9tnJ8amV3`(^nH$WFfr_>cH9oo-LUNNrV8kE3LtB{Dl*B8C6NL^ z@-cvJZBofvaf$UVok>j_Ly;2GqI%6Bm$z@oT%XxQ4zQ%*`R$#f1M1F-&fVPT>pq)p zANgs8$0Q95*86ft*8X%My0=jiW=iw;uoT@%n08D22cq!0d$khWn*?l9tWkQ9$u-KS zoec`fP{Jl;$QS7st7e55lV^n(C;_K7nXU{*P4o9+vFC0=6X&Tl?H%coGjk2>e;Z1&5&8+tB9Q5gYd^-spDrFZaG(p^hFb3^k04?v@4OOc~YXoKqBpW49~m zt2j}#?wLqJ{RN#=TpRTN>bN#|a#q_R#c$%sNUw8pNvaD59PtSxx>N7nUlS)#qE@p$ z{0Rv`y~ZJ=ofX%TE>%R0T0Jzo!Xa#56pTQRT0MU13+eAnVYoHvHQ`mwYpH3;7uiMD zHrii)`n0gNeujL{Wy~fzz#j=XQwckT1^!5;t662V zAB+x&KxMyon4Uu~^x&yz0kYi--rKu2U0Bz)T``u;S{r|B_1gnJF0563-OBBLIhEw^ zPk(>paNHTUYvNI?#Y@~lZE<49UuDgV0aSmPFGmqu?Kb+`X^$+ppgZ3=X6fq(E7xQ8 zYhr1~o1sEK(UU-EI`czkx7H)tDN>I<+u>AY-uh%uaq zrQq24<{$d;H=5T(oITz6f4}#jfUY!A7ZLnJr~iJaA)}8nUhIy$x#ONx}TvS^Ago2$dBST=w z8uHms5`WnV42#Qu2*lGAHfVh&e*{FA}dT^*|E$oEXxh|Ksj{YlYqzO zALQPtRy>Iudbyh(UlOhH`I<%r^zrO_IscG*+a(M@SEy(SdohD{XLzw zK}<#+Lbqay<|U!!--ROHRDYaYnZpDo$*yq+UM!x8ApaM~^-i}_)xg2Th`it!3S z#352`%)nvPXNN7CZzdSpczMS!;cJeCsJ~&#InFI&(^zPeGmO_)Or!^p7489^X?DpH z?1F%bMk_*P6~&1<8NPhR%QTJZmWb!fxQd};eVm-4uMS? z#9PAV*=I*;Yv`uYRt+6pkX@s#wq;h(zYK<_AmI{S*}alblaw1?reh!%y{%@w&v(#Q zD&tEn*X0G_cw)O9h2rhRHc_|`a0MmOWA4ZCJ>4GaHBo*Mvz2T(b@a=0i(Xm5ooe%V zgeJLas^X4{XeIlV5=I}<>(~ie`-`EIp1|z5 zxGHV08qAeO`XmD#6{p~y3w&Rn$P_r8kTcloYJEBq z)7cCc;e70M!xqVamB!rQU&0Vvnud+;(RO^;s^B;}f0F1>&TzNampgYhG?q}Uom#8Wxp!-?{dc+I%A04XE5-Yyr-KkVSGJ#kd=6Ib)9OTQi+S1 z--V&J9Q7b`nI7;IvQcdk;M#Eq_KOzhI*GR56CIVzpi@j1kEz#?b~R zZ#q^b;!$?p0yCjlZwE0TlRdb$O#eXp*;z2fZKR~YGCNte-i)`Ft~E0Q><5Xl#2TBn zK%N{bv9PWu{N4l6F0-^eI;lMBC@S~h=9{caYFloQcFUwtmwbb#D)GWubZCU;(pTo& z;cYSDAEHJsm%zd;UKa4K_>;np>H`S^vhY#vBU8imL1`1>yAyBzGJ8o3dq06Pds_I% z+o5XN~8S1qV>u+dQ}$XTstrc>7Dee5cb;BGHr zqEpTW=tRTo;E1bh{`28tzx^YZ*=9fS$@2G6NrNM4z-sNR<7<*Q_)({Us&*~kV(xo= z$M;dH&vp^iTv15^;trw?{*KxGdZeek(VEyWKXr{DipaY9|Vq)Ty1*Iz#m`WK$`h?bY?)rjNYHkLkE>+ zi(fzAK71TT`r`#h$1Iu)AXMF5XE%pX)`kCGIFL2V#hlpVV3YPERae%Y;a_w2kvHwb z@E7hWMp( zjZuI)VA>e_M)UfIj)Ya4;$XW4G-xob)28s#&HQ(HqGnc{tcpihfRdvMtx8sAj^q*BP9mPpnQJ6UrBzQVcGuh@e7U;igGV71Zcw5m zL3I06g0z0U-&?Dkg^@|D6Ve(Yby62ovPTG*A|_QFV-~VWwHM-U{#&YVfZM%Im?FAY`6EoT z#l;^cH2GGNk|s#SWjcIwhh;-Fz)_wV=dpZ6n+{Q?E4#V1UJ0+jSFpSTir4;e$U{G0euau-hD`?SHk>z7e}KYHKBrDukPkLSA_uYs)f zs#s9E)OP<%*Rh6wv@-vF1&r_WCJ}G2Xthj!1fix78?EEWVi?IXchK@*g=6X3Y;s#;$^nQPE;4W? z4y(r{`yz%T64g@2lb}=rH{3ihnPXgXq(Y)F`VnqPCE<@>PO+|I0T z?34j^C>8tH38&@=y4pX*2h6mvX4;?29NH^re1(&g{Th z5|RA<8OCc8yKHr5#>pDoR5-9e*^nK^ewyW9A|)8@^cd(&%CQ>bWXX(yF@Is0qWx7B zFwhbMZ+S^KUvnNZc_TgOX7x7Tp8kV57m4biRBfm{Pq^)%a)oi>n}idgNbX~gm?3xa zev)yWv?Y#6ZekMu&-t;oKPu`8(|JvU!wNS**#^fHUY$C&B}9Tk%{zynxNY6zs$fR=v*dL`aA+r@q7S%zI`E- zHLh6gUF9v%B5rgw9zzBOcvj{PO@A6xmvpF4+baT`TsoGvSgDFG!x#ho1;A2|zQ~wk%)))o zek|6|9!%jS*m6a6Yd&0`>EywCg>$>t+D7Uf{`ja@CM@3OjSqz#I_y)N%2Q0PPKvK& zzMEQ^EaZJ7n0i?8h3PXgh0?6fcL^B$yt6~BWdw_PC*_S$0qMA+M1_|GZx&{PRhCE3 z?i?b;6yNO?;$xj|fz#;$I@~#LrRV(+7*mzM|KBPeKcq>b(r5zwwQFB>hY>l#Evh{uA-M&sE!G$6SbzwIT~0APCBTPyv6_m z=TlVSm!YNgR?uIFJp~h9I)}HlUk*w5DMCoc3*j4rDWu^Qs>`}eO+E}*LPYr9d#vAH zW(KC0XknW|ZTV|pSdT1Oj;9xC$GSd=)%aGz?hu^vV3P@;9Xo!~t1nbXmc4A>>iXM^ ztGXFBea$5G=ZOj5Z z5o9T9bJmN>@GLjSOkd^v5rUjANFC@{ zyEoxUuK%m_7(~H%YZ|kjkeX7G)_={z&%>ABCnWgt3vYymF7Lj_%T}r?wt9~_9?R6- z(?c}s`2DXNj2F5vbV&+TiOQNvhj-K2i5`DKLIT%)I)(Se73%(32{+x?E=i9^_wF$< zY^J(w_jsW@{geBKLa$!kX|HYR7Uw6wv#_pbC(wy~Q zd1{1x+3Q9#m=lgQXBqHos!hNL_;U##;7X{wXFzDe@%woH#s5?T-w6SY@VVp>RyJ9{ zM^Ci>N9*5OS zU0q|*;Shmg)4Yj5kD4zWAx3&JcG*dlY^TPcR^2V#4#7KIJM>zp25)Z@1?aw3!muBj z#&zA=erSj;qSm9g#Ty51yRY)i&K!1nZ_N8kF%ybhG#oeTF!}CSFS<^dY*HzX`mg_| zwez;G{hLs-n}zSg}7!=ZP#o>;?vJKyFle~`~>&EVej=ef)FPg1vS{Q0M%Qnljg_DZIqx_rG2EGWj&ZOeR zYP@`$C^c4^t{vyFfyR#BsEMPj+sY&*na|cV7ES%)uptJKP&!-A(ouS8ac~w6^t>A_ zbvjKjpJglcpoQJ@ZImo&NC)ct+0(rPKUtE^8I7Psms0Z@)YR`zhGw79AYB zHira=G*Re=oehbY!yb2$U!PutyG2ASyIQgmr#~{bPX>veC)-wi_vF+&c^DBs~@YR zM?YJ^SkSyEZZIh~e%RHrz;`u+IFbAB4qoL7Hq=;={VrWA(1>dLH;5L%G*^NV`$LWj z5ADG1cvIqxfbwC9xe2&&T`D~nVfJ;-;cIrjuC}9*&SPTu4GomXz}b@Rb@xj)GN&iT z&}Q!CPl5S=0JcUG6R8cv^-0E!hJo(B9Ifj_*jFNkb(IYS4FLnr@XME+w`KnOR~FimFocnXM2N&hOI{l>+%l5>XnoB`A>@; zs&c3mNnNPjPUL9UxA+f*DnD?qsD9@Lj5EHiKzm!hwE@EyIDE(PgiVy6OYwvXGD*&U z$+6K+fm)(whv=@opVl{QeTKB(^Sm67d<(zKElQ2gXl{<3vRvy@LSc*9dPVRnqq+Y- zsj;~K@2Igj*f`iZ|Nqoj%T?U1_IuB-1djq=Qe+3q-XbCNoo{;MA$a{m<;1p+g#&y3 zx%nAVMyIl~gn6|#41HtKrs<-}5K%CsC=?tQ^Ky2E@GScHLGgN`p>%svJ@G=JPQAwQCe zhF9+gUzeKxV!u9aE56>|J6fB*GDp4=8nO@Q*qCvtPtGF7?6`C|v(WeTT8+phZ3oh# zX0+uEbILCga#+(GVYS@fH z7J*fQEmW9Y{!NtJ=U!YmIo1-K94ooyKt6RHkhJ95Iz&(S4L^XAt+-A|X7 zkGsG67DWbLwz;jf`VTmk1!>;3>*Vrjql3Y*wKg7seQbS^vxG@o9wcgM%2VJ0UD zPlNE8ncf%cTj=03?kV_}l$Qo&w^1gLjZh&;!yI4wY#nFI_KM+3DJ`=Wb6Cr~VzZ~z zU|HZy!aIbF=RVJRrK`C!2N<+U7Su&yW3Axq=LnF*@+OUuw7mW-+MdSLgA?ycegehY zY*UxtUm0`~Tb_3-WQ=KG(>0}VI~0EWDH>z^A8afEU3{6yk=O4R&0;5<+ZWkctrk({ zTQT_OqtoXUnSV)aDe}qkO{)E@w^oda96}%$N3|d!rL&skpY|&gQK|YUY^=NX_cN{z zj++2gX4X}!V-DtKRtYK>4R@(57RP-Ipus9qkJP$7(=~x5IeTVm?1e3B;x*u9!TM}| zl7}Y?=d!yTd-)%c@Vp1djMnnymeg)(wW)N?dnGnphQqX{pwIt0-QTE=IMKbTOH^j* zWIvM|-lKRj{?W#UAf^*oW&G3W)c+VBv-S4Wn3a77DH+krWc~mUA1}=QrPPYM1E{)5>(NRWIgTBZF=5$)5^lmFHDwf zq^66;G=i^MgM!A&O>2_<4>T4V3K~ncI3I(DqQ{uMj8w?2UDL{`7_WD27lT)Y+M0yR z0hYw5t0r^A;`!|7TemMq6qzc{F27%_@lO5fc=}FU^9yT2lGU^rosQc$E-}IZ&aOGmnM&{q1!>U{ zt-BPCW!|?K;q>SuLtwuVTIfy z{{Sb~e+_`iP88X>oPqsMWKZS?mrAJ{I)8DF}D ze$K(-RG)HU@`iP}1F5I+TdTlI6_ChM?jCzp`W)qg+e91d5BBe`e&ju2IQ~$a=&R<@ z1?pAKFGl}tm@rb9wobR?YaUMh3o>j7@_gR0WdTvFXl*M_$xvh!s|xWGokL=me>=EI zh@wzp<+JV3br`y=nu(?sdUuhsVK`)7vspYcPE$W$AW(XnVgnjIp3N62CGJ;cHkYy# z$Y+Oy;`rxr=emZsLoR&^BKvK?st58!5iYWU9pEC=b*{Lw)(39(Zxew~2N&G&`)yEq1|iitk6TI#+iUZJ7UE`?WQ)OYqV9a$T@Q z>2G?Sh4SF~KviE*kW@3WceEW5J9sYymyV#~5||es4GMBzPow}tMH)&4;YPTK_oW!X z^&F|UkWa)Vsof<84!q9mCulu=&sHamp&{(g)41kObPapaAYWW0f6_RTs*;Ca>lDJ7*#qKXM^RCI*sz$%-m0#fB4$xAa*6Aji`SPkBSBInb8`y*puXVEv(H(qJV02{Z0BmfTJb?7-v&fV5i|#J0>9qf-o6AJo*EZb(!C zY^;?I$eoG)pwxw=ceqyvg1newle_s7M7>pL!Mx=O0QkF@R31!LQ+FgLZ`V<=Y77rc zgz@!_)EzTGlUks=eR|N2J)Wu`&L(6N(UcsKVvQ|>ZF0urLvB*Ocd zf-3MZ&RW<&@s*P7>&QYBU*&-fK*~m=YSk17wxE{&YVBVyyG(Ka$osQ?B%6;P2mXz@qbYe%p}v+x zHd(tw4f3_4ho8b38PZ1W4z&lHYgtVBKv5PMkOwY|!?Y1-NqE~9s{u2*V#;_-6B#Hh ze!_=Pv1G+W&%TuQXNx<>M{6i? z;4FH-Ju@SXmj9US`&=RhUwk2M3T>_=hd*aof!3_N<;5jR4(zwKPW$LXdTN^=u}8~yaG4lh zIRTp{^9{rdAXmwFXf6bNiMe1!1l`V;rYIv#))IgAa0%1kQ4QAc-$N=joU~gsP%`Q) zG9X_31njx)PW*dr&%o>qQDx7#@ymmLKh@~4lMSx8jekl;(+l-4eGfKSfo>|gLcEwS znZ|P=v%(`E2CQt2) z`f4BskHgYPJ{!x^Y_7yO)VEX8xR}RxfwA^#|4xuzM=X0-Ug-l|U@D8VW78g@3L@6E?62S zx8L*&-$&tUmIh1wyr_v{;(AzeYlCC^On6q=juGBi5pD5;uGrnGc|Sm|NAgbYrl9*T zlo9BsS}sddql)Rs`Q0SfcZhN{g&cB^3qrJnlo4sU^Lr)~>W0Ruz}JzKYpLnx*j|() zO9OG9ed)}{_U$huRCLO3#WyEL@Za_aI9q>LlL{Bi3&+5AK}NhNuq;Y#YDZj5t~Z|M zF3qf|P?L)H%(!FnmD6QhDvb`h^(gtfo8mP%Bz@kxs!(B!(m1HW@LR0^-mvcXO2QA9 zh<0I=tuZr@7GtITnt5pHn(5}FWbi$%5!<4#9xS}Ko>QKhBl#D! z2_cM8$dl751;tU~Xd4OT)iXf$p%L(<$T z$OFX3yIN#Cws!kkxv%&4VEt}m2IoVuSArI=%J=`fCfW1j3h@?A93>uOl7rx z5-RagO)LwTLItil)z*9o2fyEd4V2`b5ChBeX4jX+c;W6JrgdK|^rqn=%Hn>RjH@MN zUIP%v@tO9Wr0=bL`WjvyjGH{93p16I@lHq4rRDfmlXTH2?44^a z^GmLq7k~Fs51kQHV&4SZj}<>RHs0gccJ+Z4yx_$^Qt5=f^fJ{hpS1k= zC*2&p!WBxiA(fYWFM_^Li-)SM9`j-7z` z7~ZoOx?#%Fu|>~F8jb|vAXq7An%c*dC2`>w@b@ra-O3I@tWP_J%|e^P*6 zPrJMsfMSf$)bARoz{CfTe@3H)YLwD^AtSYm;&_%5j@T(;^bYDXEV6eXP~IKbp&3#T z2#B2JXWx2j0$ohe(X0BtcDl;XzEjO@>Y)MwBwwwWOT3TLp2YKG^rJ5rAOAi+<~^z_ ztEN!Al(0f0&$I6u!7kyRnk-UKHrq%p$2AuX5@}^jtMrx}SnYpNvV6qT(dgnwI{jx_ zg&^O6umA$Bahb?jY+0itE`XN`cF#&PcWx#9%ve@7%#@m(S{jfuf)kGB2pG3qe>WI& ztZnLxVIHr-<9r`OdOn&FxsPGrUFKIXs5`;uqdt(L(@>WGT}VIl{zNLjKS#=;3pQBZ zMIt0Qc-X6EF25h1D~+%sqNZ;pgDVA@nY9EmxpZZRE;T&RbwSL;&$vkkQ)v&{+*Mk& zEnW{Yjcl7wc?J#XnOsa4bZ*LW&{bTQAmlIwzAxjSzaZ_~Nm!;{XBW=VreY(p3Da2o z)b5+=B5Xc#dAWy)-M$;;4=UYyD_ztVTdMEaNXyELsl?PPRq;uGQ_M}V`5U%k1r(j+ z?dzs<;~ZG?a1b;8*C6{$-*Vpq1H={@y}o#aKh2$9F5JJ2(Y7E%^%sOcjKPa5!B3JP zpMP~{vl!{FSTb2XIth1}6O{T93kKinP-0ou$a7J7<_UX{67ReLKZVpE!B- ziu;Qy-@Z7h{u`|?`+?Fdt<}qG zANnIadU0d1;^{1E(HS7!VT$`bVFx2EO9uiPOl#^j73)~~@_a54zhh0=v#1YAN1Ul*TL){&3OEp(WQbT z{B(1ilQV7Cm?)v%)OWL2jg?|xpKrTB7cI}NUu_(p%1hOOEnYH>qWY}}x;l5148BH* zY{aP_C7-oW{zDiT+w*B2tWw_N1sf{s5NO^H>{koZ?}027O5NjUW}fO1f2RyLlSp7IDj1}h8VAD>Za*@1L8 zT?ofUid{RHGfLs*PpGL7sy>B*O$2Gg6>p+3l~_xXFpTx8=?RBJ7^&c(x`21h-|wQ2 z_#;SoxP=E{xbMCe1(WZWeXbsSu9nl_Vr|5@rT;U)Z;v8eDPjOY=*zq2ouBhpww{By7IgBYzF|ESV~+M z+>Vd+NMHLU_5j@IxOxFrJ8R?u!_I3WWh;M(*9cVwc^T)eSSZZxf90%8a9dhtk^Mjr zH>&?ja;+R-t(~ywUhXUMIl$1ti)I6e4=g(7ShX=h7%kyjl)mXe zJ-c>p3Wu?Jtw6xlyBzDYBivtqfD6M@rahOxVUD9y-OsLs;6J;qYG2qk4lMA2BDx1Kd4-rqn@rZv#P{48CKF51v@3uwlSbb1 zQkntvC+>ctAO5}Cn`slbkPUc;5D4h|_gRW(J>Nj~Tf}ytC&#J8f!} zyClD#bYW{iLEdo(W`60vu8fQQ!-l%RBVhxz)5l+P1Pdc&3nn})mDt>PjsR%BGJhkd z%b7`2|Aem6VDP->9+PCm=@c)=UPd9` zma-PXZ`*EW7@tXHoCWEqsu87Vj_0v)&3>z_6@jf((g2+%y&v!`pZDMbLGey~d+bqF z_d`I4mvG^%#RGbk0IZ^Cs4};jWLkl>^;e+TeiRt~P6gddla$wOT|3rPj~mS6=_Kx3 z>7NdO;}JuniVMFgzM~&0xtjoFso80ZqBQVJ?L>Pc(?g^549US4osU;^e#_9ELkB$ zJKf{iNgUW*ViTZrLsFEoY$1>JE)mwyTF)c{+!}=ONOsH;KUunH+mT0;;hHprzn?8? zzPl|Wg1;dmu!Thr8z5#?^Y=T=m`I&<=gr>MWuw{UX z7@4|N!VsSHq`T6~x?M6^90*Xjo~8I(wN8?d$dR1x%TuOrUM5CAM$@<;{evIxEeBgf z!Ixp36D@+4z)e1t>*+zrTWx!*BSgidsI5c_4=vOV;dkmuL$_w+t)%d<>5rFJ#1)~? zuvj6vqRSQ{w)lk zQv_ZmmoV0ZDVP3}v{1vyV!;Oyh{9=D8lbA{mscC4;?B?% zogNjvA!6gA$$89C<641L?G$bkszZ2}b%@8mk@WX}2I>QnKfe7W*$7%CCD3b>d}<)H z3t=Gb!<|;`w)@tM-@dS!@9Bo&)vN~FNle#(omNQO^KgMM)k-Su3yZd?UkMfBoTzoC zb#8_N#Y*M@@EuGK7lm+935A9VzjEl>@41tb9l;{)EIB-!qP2CndEs;QnLMdIKU3e( zb3S3@@`AJR#WcVwu`<3)S#iN_S#z_e?8e01jK^>|KS=R9w+LHa0{GQ^-<=PHie?v@iPIixqMHq5-?$BkiSH6|@Bgn7NlHH&ufzF(G6mE_Jbh+2Ar+E;*L0%vhK}nK zj5)L*y1T^yxM1e6biUNrCC8h4uLKxr92mL0y1ix`U7K;T*|_!?l6%W6l9v1RpQcr9 zpIg;liLTG~5zX!#n$_7O_)CU2!}N+WqsPJuVUh0Ou$$>h3t8dl;jM#ZEyn9q4~S#k z0=;4{Qpfx|EWnJ-8e&*yAjH@xcb&qeQ$u?a1NL$w^pR`go{G+%f>f4e9p^=l!7S7R z-d^ZBCKk>Cqi5dol2y~YXpSSs*)el2b*r-zCFYi3G11}4@Ka$we#bt z^jyl%1~dVz7LsSjQ&XAcRb?#y!lwUj`+Iito%_fMZlg3G$s=+eai}^BUUlmxerp3{ z%07frHLOs-Kps!cKie5tTPO-@0j3q?XH-CtECv!>Zqh?@Eo5v>8}ztSNNFTFA_XjR z#LDe+W2BChE@TGLPQBGCBTjayEEAlRm4JAcYgp|#DMfB9rQn{>$<2=1^_^53U^*;w z0mZmatoBlNEP`vRdX^BdBrIp>8%w5-CHH#qFn$SoT5l>G%?&*QipM4arAN(>G!G_Pp{J)0p;Etf;Iw3Q><#g z_!<}L+{FF;xs1<>8_ z8amc^N=j}crr%N&7O0(ebV9Y$LPSndi*IECDI^mx!9sN=;)e({%7G`j-L9^0{7Z7t zZH&&BvvP)(`9`18YGq&A8Sx#|N-JZp%G zqrh@pp3t@Lil`yyU)s%=+9D~NMgcq+|JIoD`@!s|& zb{hjfFJ@7sZ4R>AK=@w*oGX=%;^gOk92;w~xF9*JeanCHuEc^Pf>xz&5iYMdo^Krg z6+4~_Tk$R!@;c7w_>z(@r(kZRSb{EolsCP-sz26YP%3YY$KV4_6kr}+NN<}EiIcRy ze1#7a#}-{XlqCeLD=6dADwTj6J{v?yoY%yA$u@&t3E%$`OG(u=iZ2s?k^0npOTl8> z7x;VD5~BK@3FkdcT{@>2ZcW%%T~{0qew=zeboGJSfe`<65iw7CCG(X;Q>nN)n_mmb|(F1rn0s_NigXP%tT9m&mT^nAe`&$ajJ%| zOrs_LwZD%v-sx2F-t#hEybu^{!|HBQ{SkBxHxRDgiI{w6J>wv{6Eo+k|NPR9;>4`> zk(gm%4faqZIiZfMVAhvMwTZI2zTwaI{Hj4x$pu{ywk^dITX zv*IjmjV^E1>SK*%70nsnpAs!AupdqmHjm=C^mX-D?}{v2xwtu29~K~6r0iJtAKgpg z;kOVozp^ykAyAzt#D4XyMCr86RZ7P2RQGvRl$}n=iDtV+6~zBA7QFfT72hWvXLk`^ zqO5rK-|h^QSCaXC3d21~`AntiXtRY+Kj?m5A$$eVRkIt~!+M zIb16VSaQz7u}kGodS=UdUC^a;ljg85RLaUVc;Ok%I5X($5W5_=fY%gJnEj# zgXpbmGR?P_DZze~`5fh*Ysd~t9c(8jAG43#zw74=Ptb6X?*}|ZY?@wjSJZMa)6Ei7 zY?pUf+`StEZ?Zkmk(-}0FHhK4Ko16>yX(^N^`V1ZzGOwZvUJhmEBz5))S7lA_L47k zt)W+889Y;ASd(3ok=)^BdGj#MSS|S3!&-h2K5z;FiD-7;I7f|XOuow6*R&{ijLwS_ z2e3=Xn*Ln3oh6@3>n$xyz?`d?yHUhjwl(MQxoy;zqPXDuqqTPHrr~m}&at(R2y9zIj;Jyg#pkbu9&N z(b*_|yuT78iknod9~1slp+y4O+tQ^Sh}lJ#5+KeGdNIIp^D*HlB>foH5Ja!T81$wy z!%59kX(6WCz^U09fH{4<#ML4Vxhkmws4$dZz<4&%WAG)JB|ADnTo zM&Vx2Tw4M_Yn%XHz$@dwRHF?Nbgkfs1GnPkvQF_}OJ~8cF|i^SDbiwe9B{sP zJc|Dq2{+)lq$OL|uY!b*|L|1>#sjo%5j*gosQs>C z6$VG7R^NS9SzEdN%4eHOZC(`VdWFI0&SG-Gb8$;5XmMZA0u*%FR-j|O^p`SsgU9}Y z%97(c<$=HT;U`G<5-tG{l~J9uk42vHEz9B3HQoxUhc=pEYWnh~Uc3RX_whulDlWsb zXr|?h?h^5#>NiVBLoDZVztFp>)CVk0!I9yW0W8h9K5^qh`5=;GN_fSr5d}p2!!Bax z{A30V$C9tGVkT5>u|?^<+Vy$cfE!08P-pJ;Z1k$^%}rFHL4nw1R)PUo|)<(wC} zfX4;8te>)%{P`h9}?vY zmMVBI{A2rX_A<-asCn}Oy~dj#um7C*B?rPIbJ~C{HJMSfq!o4&`D-O?1~n4?sNr@kce|ydmDB7(pAJsVx@UjkiqxXZz!@dmY<@U zMCTSsJpQP^+f~}(W3{8L_N2S-#TX^7=J+lcc!fpAyOu@(eHAX+Tv<`40T?fpkI-bC zY|ZahhuoNHLq2oh=e$ILLgs1#o1?%vfv^7js+*27Q{Ipzj)I4zetoJ~7aK37BiLrD z9q1^Wo^mFAdHCo2#d9%*9|WY7SfRA~4WzOC>qOCayRLs-aQfakivAOEGOxLH<3|x1 zH|E9{y8RZdp`q%KX~1g2f`t!MFSFsxG^^}HXj&V>XcJGdIojmFH&%j~_GtoTYD-GH}kix_Z=t&@mI z@5Fd|AR;6)s#z#?0RlR-{NS2mt3dTj2007eo+p2ppLX7v<^#=DIUDPgr(fet9CcZp z2E}Q}gJB)>wuWP{{_tH?=$<8nD2g~`Uz;h`@#Bt|<4Y4!xvSiz)T(tYB+|dxNqk66 z(d}eY0C;B2S9rE@hk>@-uuMI(U@jmI;ZPHC`WXdAP4VdPZl%(IA#xDCT+>v`;9U4Q zdl=~ur12TIe)vPdm4oXW$Q34Oge;T12sYd?8mh#-X6inc#M+d4KC!1zMf+23TDjx{iV$|n;q;!5`R`&gg3&^-vr0O2c%NAVBS+-&di zQgp@6Y|F2bIH?;l+S`jfV*Ft&m2Jf?4MrAa9nMvHYL5Wht){wOy=+ks=PeFCC~rRI zH=jLDxle+?Ml_;YW0H!eywh!tiYEaUg+yE&P0HLH*DcK&!kwe_}bm~i!mKKwnL z=bkp7kLWP$J{KVKweNH4f+8p1@XRf4X`ih943G<$U0#FBeBUCzvOl+!UA*ZTpo05z z4x%bV7M>76z;t6Hi7x@2>cluWi3}EvXe812;H9vhPq9VVnk?g-7~P~frnZS>TIJ}d zJ|B|r*6~CQCWSkpPMI!NRC&<}(y`zHI(m`Vvp`G2O5(W{jOrj2mvL{!pe79?x(-b; zBj4XV-^Nca9mJpM8yBur%5!hc9<|~Y|4~ts{U%>p%^Bf^OvO{x8{tdv588j z;j=+@b-C=q8=Ay|CvqA@a+(IXS_v1|zvCUg(=Lb0tTmdmi8d;)HM%vgYynWNoYOe4g`G+b zWeoQCaxWP*=A|=}Ld?x;ye*02IS_G_4}YlZoVmL%Vq9w!e?tG0U%3qAM)=6w85vES zcv+6Ui)vfT?ID)+iIylM2NFv`pR6?B@t+u)lvas5B*t?`;Wyi8%WA7KOG%s1pNMy* zgn-3;zIn@jHS3v_8NQ1AGMA4^iN+5MCK1OUJ}P}swsg3G`FGj+zK%)L3O*o`xR(<% z*j|gWxF@maVtk>0Iqi*F;GNl^%rb*KfGX*_Ki0 zbO{|Nmsm9bHunmZncvoeWJ`K73C4OmZyh&?#3Y-Uoldj1^|*ed^X`noZs|^LhpW6k zb|q#tC)6!)HO>={N54K&dKT(j@q7!+oidR};YhbRif{gRduk(Hj%4ovcYSFf{cX-J zK|xRDX>gd&UIYjosI{?vIt`k8KaHjhd(pFVFl=Uj%9O+YtRe#5j6+!zegp<1EDyo< z&rpshC6q}ok+h>NyPYdVJ2O8SQe%XS;xY-TL3SBzHGZpa!Q;HD(Y3GP4pCW`qfW#7 zjh-aH>zP2Ab=yMLVzqer%t}oVufTNaqJ?Zeoq@AF?Nf9<(P`e`%p?bFow>*4uHu4N znbb0X!BGwOs4b>Z{#lBWnz=2sz|Hk^3^-ij=t=^B*D&j!VKW9meSAN}g2#hPX-@sw z8h*Pa9tN}NijVi39q_)0O0m9?h770KQgy{GFem*E%b^G5(cQ$hK9d4g6w=pD24)M! zm!bOyL=3&wDc40W`Z<3%R%v|$I~x|@<)pS&Y{om0QX-@;N?9CeDKzpbcI=sNLgM|6 zE`lvIA+PQwj?1EgeYvLO!j)t_1CZPX20Dicp1XX48g~xu9T^38+STq}061#f@eWH% zI#@9Y1q4(l^1()-?Rj=an96XpcaHB+*u(9iz~+f|(7n2-V}9d_8J2uD`WhbQ~{yj65}8prq>iG9n*@w@-u%-AouNsBEPl}*htjUNZ=efxKu_~9W`UG$z|+IQgzM{qZXg z#&X${;AMUAeqit}RG|=GjLx?lSlYx}M3RbwFTiAHLr#i(U}2M4IM7b-8s9Sehtm7r zf<*q*qTxecW>ZR(iFySS+rf%#o^K2IZUQSWSONP_fP1L;l|9m4R$_R38nj2{5m#n zuRrVucFU*}P4766LYB3-&z%0JRnulPR^o1M|M$On*LS!57`%T?x!E{!o|M!l-S}9{ zApaBd-66sVDK+|E-2+22$#jQnGQS9zor&3}{`WQIw+M$%>RV=y(w8V9#VP)gnUh|> zz1+k8?(wh}SjlT^Q}QZxe0KFqji_f(UU%5~_SE-bz~5OSF!C*9)dsutx0Wv_e}2Z< z0X?sUS*%@t0jV77zWq6`n|}Gxa4f~*-@l8ypr2$ z7OfWCVTtVHN3q}}6Cp)1YbzxTe6O!Vun13jP=q*8#u72j2lGPB=NOcD*0N{v=0opx znO)Y06o0E! zkKn#lx#!YFcv{+tH*CkOzQrMl@j$#2{S>=;pSD|4vWzaddv;nB`ZJdGbZdHA1LJ7T zaEcrkf-R}c`8)tt_CTQ|MQNJ3Ih zpWs*r{(%=;X4YXESZ;Ov7TSm%2a93fA@K<}mgDpohbMaZ;_`gPtMGRMP<4nwk&Iv4 zdQL*n_%OI1n#l$`cTzH}?zL)vQh|V{KFr1~`ZT^5c1m1!wIRsyI&YrVH{LbO_R7CO zu84fl^EZEMjFICkN>aY)!+oUpQHOt;{JRZ)0yFM7b-mzg+xYR)U#vyC(0!FnAjBiT zt4Z9f2u@Gh?_2HEG?I!lr|mJOM8oF;-m;;ouGs$8Fr{=V6u_DtFI+La_^{~IG}RuW z(L1T#Q8m)WL!!g~QeC5$3!B{{vRa@52 z+6AUG(J1$Ha6V!6p=SP8jX2e2;r$hvA{E!mPT@hl8)zv@4ee5pg2p^BU=HWQXc#u;buFqy!Z+KuCBqquIH($#+ zW*dIBR>YeJ{Wu@rq_5wxeKv=cEg%g|7_n04z2Sn&A3GzI4XkDPZ--RXmnfvxZ>>ktVLW&(@cS5qqq9+H^(Fq;xQO2Be1 z;ma$XsRpUQ3LggcmB3eGYnwB1VDcrd6}mc`S_j$EUmm_md(tTh{?0}BiP!M~f&IxH zTPKsj-K(k5SJVbe%(kydqV#cpHWwrZxk3r+-qaRGp0PRGK`m%P(Q}##pfT8Fl#9g# z)rE)Ke*1t$wFvsXd!*^_*9O8jg(LbpJWV`(8p68T!cC)AjBcXgO$Bo&zdbXEJ@juE z&j?Q(7)slQzlCboBFCc*d1tF>&*Mx9EBoy9r)=u*irmfMQN!U-N?vSzbI1lxrDRbs zwZ1vSX%FbcgVr$O@?Ns)>hBb92sHPtfMI`@P&icI_?A%L9RE?(u%BF(u(Iyl-3(xi zAy&0jMH~*z6|(rRP=IhDdgW^vVbu>`vAWolHrA>_Z(rQXMXvAWfB*)usvl+21eKB> zuHPC)(9xk@-Ehl zW&k#i4T8#7Z~o?(0fbx7E6b>O%mH5;X$|}5vuIJ@L0Dz6flsUNW;0<1SoyX6KaBzu z2rC!&KvB?>oakQAQNr6Fpm7m?!~Sk2l$R9s8kB6I5|N=2{pXPr=x$vsx8VQjYYU~SDnz?qY=obpCN59HsVRlWvc>-&{SqRIu8=}Ncpb;ANjrRD(S6e^>o2STBgwG+Ld35Lh# z|0}Bhl$mui^xO>4GWq!<)QNZpmA^dL`t^!Qy`XVwKEwXEK&<}?H3y2<;cf5sfZ{UH zD>-6}e}EjKGEf?mpfvuS+>6Tl8C0Q^qZV0XR#qiEEOeX@)DS-a>5y3SEaf8IBjzsJ z8fg`~GzEp^UZ}|PgPd!E1isnbt?6;c4Cgllrf`P?!{_%dvSV&xK|C(9d0%n!+Yy*w z@vX{_Zy*HqT7JEfk*z&t#n*(s7QXmS3RG%@A*hQM`u93pXUvRpfvE<)csXqdOk?L3 zoJ$QDK?gGw(S>PCOvKb(UMYJUP~O^VX8M-<|4sOi;zYb93nOq)P>ALl(Wa=2Yi<1% z7magyapqu$x?@7{DgQj*llnHFCD%P^%EppbN zj3KO()=mex23UFB#Ad~T6DX7Cbc{~5ZTq6D5?~gm1u2Y-b+|!}WgXcnN{L;ptwEw8 z8iG})7YV?+82$7>l3K<}&%$=+Bd?()6Oe(0vXL^csx7`|FIyUEQK#H|r_I-D9~zm< zpPG&r9~;m;t-<0As&_A^HP9uK1JDIlG#oGb?7FQg+px5D)}w>1OM}8^{EbTb*-CGl z@PafnkL>|YIl1#QphaP^>QWBk-%^^#oDN|@KkS9x*t#3-@=Kg?9L|m~bUhtazrkZ~ zA=Eox!kQa>26H^!OK2}yb3Qde(1O~kn|;Q-ls@Kk{ElplTc9Z|6stILT<>Td^V6`{ zP3v9CUC8$ywkZbPefK-HFO6G};6NL%BKg%CfW7Z6c_@Az=^>F4cj_my@6AG0n~-E) z(Yq9{ySHBL;}IOb%4&j}l;)enU%E`tY_}iFI))jlu~{>8{lG#CZsEWlMjh>0-xA?d*6i6&5?XCy*yBAX@Bv! z)b$(DEK)C(SPEOpLUm+>Ql0uSbDc8JBt5w5?^|S^SzW%@jVb%TbD%M!%=N>^?ms`x zo{6R%h27WKg&>o71gqkv4$lvIv;N$`*VpGhy))S5X!W_(INFRj_ZO4k^$NTTwL8CcfM%AZwud|=YudDpR)!1t-q;s?E2vjirVUwkxlKtB4bk6P z?vyg3eReNJ<+Htj_x=+vrQhyngGlamnaW9*=Tkxd9Q(=cI;#dCda<#Q}sg_Ib19s_MDz2BYX zbVa(SoCkdg>mTF5N`_?O_0?V*g4|RF+9+OnAlJ-=OGUcPCoUPjvGX`Cj$+SsdBM2h zP&<0i%R+MmZFHng5K!Sd?%qB+GU%p4e+z! z4%(b+iJn+!hNUHEk3(lD!@d*RJhjaxxPWp}uM%$|kK`-S`g8s09LTWa)SAHOwJbAJ zcC<BDM3+S>cJEdApK>R~RAHQv^%vKe3$R+3Y_X6iB}Oh&yjQ z=wOU9-zuw*BT1RQDbNvCWrU|JQcN|6VgX|;+a*m24A<;wH$HF))s5bt)y9XQ>9S$^ z=^$&_?4PQJ|Iw)~zp2>Z2&u0+`K-9~EffCOs0y0?{&wjNm)WcRbcDrs{-=erhV>R3#2`ZK(QuYFY zU$VP^_Jkbcck5;-%#VG@;>VlQEnFKb~nDLKG7W;YnfdV(7#yP)L}cP9F(UxQwf z4Uq?ngfcJvIA_yl!N0i-kV$@}-K!PCJfR%$W$;>zp2LFY8Y4E-SU8mVB^*1}Dqb-7 zIaxcZq5L#b5a$%SN8r8~K$PF8`wUCzX%6lyO8Ch0`^fo)`X^(mf1kRzTz?JFusw)V zb)sW;SA??^<^JkP6j2W-$?lb2e|-vfaN__vhw7+Ow6dDqEJB0}cji)f{GCACtR}1J zo=h)4db$~pD}K;yNz}NWP^M2@-dlRJ5#UtONOUpbQ=Uhri}n=g>GAY8)Q+%|yvoOq z-t`yUEXcS9I|Qx79NK~%5;`LOp{#qPAM=-4?S#QYTlM;5)i#QDhVm8OZkitNAhRi* z%nso#zWMnrU%rZbCpB0`bqvPPf5|OopV#!^ABwpb!BxA)aZQ*H7AL9{QqJGIDFdbz zwT$b8w6r>sNIF0`ZYA6(H)JuswI%0E3RMbJ_c{G2Kdq>3(EH=fw`uYdzuRJbFQGwV zf)%C6rPfG=uOX&XC+s9JN>`M@*~s_~1jWvpaapuU8Z) zJW9)9+>=s=g{9nee%z@%#r;j19ZBSR?^sP(?D-U_QbUVBCAa)SjewWK1bA6YPe~od$gmkS&-e)xY=Kvm8 zNJ>mKi!%|+zw$ENGoX~U8ZDu_@>e8JTvud$_Bn6KYY3W~0_!`72i;Hd`V+b$R-Elz z0KVHg>HF*!!CBx4=9JcxMPL)4EzDSd(nz>25E82+EW79(_0>6=fqot;^K*8xiRUX| zg`(V!SWM(SWTU^0dOX}&Xr*Zd!R}oyaJ6ja4Svq5@#NNZvrBZ{ofzi9aSSk>8_g8`I`!RH0qsVb($WDwq~H+QrtT{ad{QPg z$mTFq47|tPAr5?DCN+V69jA)nbMnDsQYyhvZ-{Ck$|HSE=c3|ni8B^H8^xg+UsF2V z^U{>XOJkq;gL!XZU;D#9joHPN6)m`)75=Z&F-L+w!g~4Z5pvto9AV(t*|#T zP+jS0Df;E@ENy?i6}`MSh}Sd`!)%*Ppbar7tFYIC7*zk4!T$}}RjDzkU0aclB_eIF z4silW{_4`HokO^{l1&WQ#dtx$id~;a+t{0J`uX2VyH%jh|y*PRHF3fDlHwWi4`j1u3F_-F$D>$vC%D}&U{2^OSy`RGVrR$;IYCFEeQ_1;bsRXQO^n;k& zTj{@d{{K2U^BEKWbQ15M(W(7k$dgS-OaB?NTCf9?cAN;ckL`FFWIQ41q|9aEUn;&* zwcxy4ADrH~c^&M+Z+1LZYD%_>YLC?W?Jl|H;*FZBj>L~;E_fA?vr_;$UBOu$bnv#2 z8BhV49sr2@0!W+Ukogn5+|dHlR7@}Pgr*_!OX z5S^6<@L4rQ5kFv>Dj|4C`lsdz;Egg*p2R$D-D#1s|LWIgW06SjD+;Q7xJEA4OT{Cf zqK`e|USvexp-(wh+GSdTOr}~0yH-dWyAXCMKnUGh2r~l+vsMT*7YMUm2(!_Jy}?u< zgt|6_I>eyN05Y>%m#hEF;C~`Nu&ewBwLh5puYYSPNQDN(5|aLE)Vlar2=Ce^45uD9 z8vSzdw}BMYltyn0uTp>|SEapZ5qc-G>8sy(LKnmN8+COa7Y9VvnfTD`8)mEuB zPunH|_%HOIPVjgujV|o)K&+NZ43vz1{J#y-|4xr{*R~PT;J@@BqxgSA{{I}Xi5O|k zW#i#uCZdI^mFbE^vpPTKNp7YRe(Iv=c(tm}X@(M=*-4=z765!|3HvNn%E7dKtz0V& zIglqP-Tn+ujWG2^Mx&evPb~*a^`ClEX6Joy)j(ROm73^^Gbx3bN_Tl=b#YwnbY({A z<8%K$edzlVZliGg-T5JD6svI-V{{+-kaxHKSP^!$Q@>>Cs9k*@I%LgupHLl>)ba8% z;YrK>V}gqbS=c;KTa8Osg1XcHcFiLy(x_6V=G+3WRQ1o$U!WGXnT}4ZttPdZ#+|7K zkQl4D;X|?BLL_0TKjDJ4trWxK!(Lye?%7;3{v>I5MrH~_4D0HvYzm8#$J#<*nE%N6 z-)b>t$q|b)!%KOs@3Wa$%#y8;by-&a-zxKOF3}AxQa~%l4|7Zk`%Hzy_+M^WMb5FK zW99dL{48TR;uv|tu3?OE&Y<(<3NxaMQLV{pfXJ~&8I7UoQQ0dAceJL`w(nqXZTu!Y zUEp_4l*tv_(tFE8%LYQou4{ecGzciJrJkWKjVm_&XQJLe6$&*+=|DJx`F*QY{B43; z$x@~-pqQ*nY~K-P7EUIRRYxB-iKFp+9|@+j+BFtc2XZ-TBlw^ z43X?~`Bq_vh&ty zt5AFcAFp+7HH>X>i)p&F66iWxtr15JdL-~SI>Zem^V8;6wY-d_wHoSdIcg22DR2{( zebC3%ci2VzQSS8Qu9@v?&#u;8d$(t_5;I>1EN)$boR@C`ITM|~@RZLff1Nfm>Ei3h zU+u60S@{gCC|$PNMpYROG@Ew!PTQDgYJtnEcD@wP?oE96PuEOmYM>wTvr2v2H z!d+ck9!)hJzJ|8Zayf(4vt8cYs)34xbP{4cmiaHlj7|%k+@ZLw?TB!5EvkqHrs?_0Q z)lHn_nYXIfftsxE z`BEIG^oD=n{ol&|%Lx{ZxjKXLRl_@p6yP=|u#R81j!oZL8S<3hNxG+F5HiwpAYEI9BQ^alOHE)GCvQjlzF=~zhK~}pzAt^pT~3vT1jE@S9!y?C0r@C9y70HvFn--#^AcYY zpD{Oh(Lkj<2V(vazssJIrOW3vbJYkGhgY$*`VqqB*sc$(3DeF4wBy1Aq+gk;;Z(BfYwJwz@n=F7*GW6?(P zPN7%4BJzI*&*T077CeuWkB^uCe-EB_>aU;R)@r}u`yL&Dk$5#Ta|;y%e^L#3y#?A1 zBAOXuOeU1an^=7bn=CG4gKcajG0m`F(3;`IY;la+t2LpkKsXsgN~j97q88M&rhzeJL1!Kut9HUD#qQ9utEOshf~H5 zX#QpA*Dr5#D#q`}{Hu=d_xJg4>o3`2!|Td-&DYL(*3pMF98tX`=x7w>!V-~xbCY@o z@aS?#2&YVD5$LKpE`RqJ;O*T9s!eZ?78}SG84J9BCA$+|qXs9+=W$$aa6Bft-u2=T z{tBNHg2s94F@%1J;}W6^aH7LeQcNa>t-#h>8E!fZ9~h0lM6 z6yZ)hWcaep#No%up|g=i#;h6hgTZNHvPBJnVqXwGwr2wmE_UCydt;PE)`(v}3Fngf zUN)H{r;~Jv+NUE&(4_;GhCZE@>lPykjj!*c`gt<{w3l<5*RbTiuUIF)Tf9Y}47BT8 zd@od8q-a})8_cz+XgoH%&jLI~@O#^_Uaj+nEA%@a2dNk|azE=n+vF!)=>9rTLU|C@ zLzGKx&lHta>;154Pz~YzDz!pkpxDED-B7A(7m?kBa!(}p@Kp}K(-=qa>bt496X%#9#uD>q3OL)!zFsI| z++Wp^Aes$c7aC}{C_HCOfao&ESOZalqBraSJ4FQpNFz$R+}szI))IE00=;DvAu@@i zK9w5M^ly?1 z)>3jZco$LzxsRZBoN$L-M++J{p4|Hnzo>$t00wO?l5{xRt64Dt>IULuD_JNyU(%m_#v}M7OeyLaNZ87m$}K60lK(X~Pr5GiCB% zAJD`zlk_b(%MosB2~gyrVmUJu{Gwr#8fd_&f7=_#6gPr`zBiBDsfC@EGcY4CCPlWP zfnj(t5>$pu{H-srlO)$Xye9_-gEDMhNQqQF#25_6y17QM6s6vynMV;V79xy$D9QN2 z1Gc5(ag=@87EBD_2+?4 zE~1De6f*bX{?H!-R>ATcXb<`@USk{vY=gOE<>cQyU*}R_&@3DrFO$UoV2Z<}_rMrz zz(kpUv}&dMY(GpNpQwm8Q(g~^O&M$oqvIvvL4`kg2{-y(rJ^O9x)5(5mFtYTN(&Pf zn*qvWD+?Cf!|$VwF$vew6elbT<(vhZw@tl|+lLGr!WXW|m=vF32Rm?qi*YXpJOw8Z ztTCX4tvlTmb7007Zty;DO}0a!3{sX9IJNA^V*jGyyHII{-m-(qb0r&3!A3<4rz*q6 z7(+nnaUF2W*@xQ^cryu!I-fL7a4Bpa_?QeOT<)zA>k9*wkn#|>ZHH0->$?5wStiS9 zMW4%YGAlKGn=fD+^z8AD0>70c)Z&kQV6p@&0y*I)(ktjQI9O{VD<*ieds8 zKG7le$PY(+Z?DJ3@0-tJCW`f>2`T=UvW5tcsc26L`{7}PeOPzUdM~yZ^ckQ2-cNt~ z+)Yc02sT=KO>+I>hwJ*v@c3NJO-fqTAby1vi}Ql~3A&@|#!+;#_-+5{{!LcrN<9Nh zQYOZ$5Po&~Fyd_(ef#CY(2y1&Vc7rso2D;A3Ls&uc)7!wNIvo>8RmUsNQ)@2Po^*| zP$2lHElM`XWBL!D`DF}J*}GMBs1P>JaM97AG1yFBF#vH;J0I}F23!VWYZ(&N{QmPK zDGG~Y*id^MtvUqcADjlOm)=xsr$gpV6jOuhZs}Q2ucO5-Y-^Hd$go~5BIp?umw!=4 zax9Ub1x#wTWWj*v+J%I|I*#oFeh-L1F{vRZ+n+EuGwEA^G7ubVilyz)SlY(slk*w~ z>Y+*LAz<<&3eBYJ3JSd2V9HhH_nVYOYaMVmT4uZTw}zTms<%I6xCXxB;7{NG$|=E^!oH^$RMX& zSSwq)Ng!VfofN=9r=G^Hroie|FH7iHSwvt&Th;0*au#km82ZToC?nDozneu44a_>5O8( z>0S_T8gT{;#DKycQ8f<8%55CjOe!YG4(6NgykPLsLO5S7WhU7I~yjvr&eUUEkdH-8zq5|E3ILet0!ep2KqnQU>ma4%+nUu$}T96XU9?I zPv{u5MxSlk>UBre>X2(+InAVjq;6;&3R-ZlTK8(ykg&?3!UQ+y=pruC16z)m>ya?k zIehPaI^)42@VL5099R(Ntcc(KWK(Ay_v95M;Os$-AowbQya?BEVfz^yWso`}Lp?*j z_^_J(R~Hxl6WXrM3lddEmUc7z4G|84V->!?KENp7WR`8mv#;#95GN`I*w3Kaj zUK2V_=~f{Zi>5{=6|gh99b1^vsPil-eO58~6JFj6=Cv4#`@@)y!Fi2wz5Z5kiIOy*_*2?h7i=OFyB>l4i2 zQn>TtnF96k)`-V}v4FZc@aSv%Xig%EQZ+sxLDd*#Plv#YCJoG?S)DiB2N2@57ETIB zFLjz6ZbFX8ESktJt2?gVjHk|_qZa@+E&It?u+(yx0!p3}(J-_#wXZ1;=^_6LUD$#x zI12j&T-KEGP*^56o%Q`N)e{X8L=%a^)2CFF9O4JvDW_)~{>fO6TiUt^o^kry*g%A; zEOW-ouWE&o*Mj)H_*CF%a=Gh~=FytbPxe$TP|0DUpZkxMMj{0bpQp-^L*UTXAWkDb zBkt5NIJtyaP=IBX?-upCgmIoxKD>v}w)J?C9E6}6ORFXxghHF57#f_dg~P3+XwmYu z6n{(4cK~fEE8ilIS_g%;@X73w6d*yt_>(XLp7o)=p159EC6>0+ zuh1HgP#8;lEFgz;kK);~?M(6P)FqDSa3$wYHe+EXacDWG2uMtry>Bw+X+aa3Y|c!nFjF4NM`oPRhgQEdagsvMz!YPmSD z**mlVs!I3ZC!1dgZE7QmF^9pyag5?k8bzhp#*bbLr0LxU4oSeaJ{b<0bxh`H`x%ll z3DbN*ijb(GK2R+aNOQ_6C=4>iaW&KwEGjigZWrSPP$ln;xd~NRoL3=F=h89Z^^JZO zEW$JdFQ@dbISAd7f-q`>T57D~<27%3Lb8a`h`iS`jxa6_$1(wZGf8~rQ5jTp$w&-D zF-+M+rN!H%rQx{1z8UBiu{LnYU%gmq&5t<1-=FuH49(TFjj6R_y`y630Samp2Ruky zQm`A0F~H5LjO@7LYisOCwXZ+{fm*diOJr?1OXndbOpC}UxFWaMS@D!EFXG5qYdDmS zH3C7kI{sW_CXs?4+a7ck!Z5jzA7F>OHeo}Iqa(Y>uPvk+eZ3lGP!FcUsX{p7Vlg!w zz5Mpo1XmGu%Ma+Z29yk zIGK@YJ4i3X7tBKwCypN50Q9)#QSPQLf3Jy{`**mb7nZQb`7T3jQme81UT|L!@H_a<~7V~WT}eGh?AfaJxr4vfO5i-l>Tn< zh06mO?)^?dzvlz{&4PKrEK+t=mWM<{-8m&%DjoJ6jX*d?Pbc!h#jQbpNSx}AWTZbrq0Wi6vq-j76XTDG-8a=TlRMWyTqi8ywVCO^(w1{bBz z<48V`AMo)=8GD8o3}-Qo;1x=Ls6K*_tUgusk&)QdXO~@c+CYKn5q@vL|~7mzltw z_oSrz-t%t~h>+2&Hi|SR ze`?qBIQeV;9CEQboO>UkIkamkfD2yWZ!K|X-WQYxPUq(t45k#hciH0 zM9vI{RnCxn+n=RV(kuXbNmQ;XGlx;eU>G*7dBVJqD>g$W`g0^TyoJQ&udpT!gIL-z z#*Glq<5Y3ZHJ+!;V3mTB{M*SMNZ73V^t^+v$>A88slQy5Y79vX9*K69TfnfqzPuVD z)yw!M*nTeRA72Wtg!NNk!>wuRk7Hb912lYe%5j=cf5*}q#5mE*!nAjNMzl@puH6V} z_4yDDzbuXF9QDaFfozO}XM&V@HO}g=O{yHUt!%<~f!BYELBo9of`qexMX#m;kA(cUJ%~R2dS+3rL{Oi!=P(`_AH6D33>SS&Mp%&o zSiSuoJi3S!SDTj*E?^B4-Pn@lsuhE-gdZVb9kO|mT&;)i)OutvUH9#_@fRBwLRUS# zHxuoS#Tf{JNg#!VoFSWNP)9tRgrRFFODj8@ydWRY)qX;V_B5cQ#%~J~3G#`hz0EO! zHP)e~1!Mb45e&=`0q3+P8JHAi;IUBK$%LD_+CH`!0k24nn=<-Z>uk!j~C5ezgFsg9&3}b{Q5_H1YyC(g0@U?m=|ve!B$(BcvFo|9lGZmmp06HR|B+T zo9kpO4h)ZXOAh*K=;abg2yt_~1~W39 zNZz7bB?g7#_Cm#)^Y4Pv2@esjHA-y3*b(uSB!BrVy(Oa1n$}|bp1)At;*3Vm<1n1H zLKe?T1zx^p&EZ?<(i?OF6j^E|wd8AfH8sR!(Vi%u+iN3&-Al~yK9dXU7OGKAFc9{fXFQjt+6ogGQyQOcC2wh>Hfpy})333SCp zT6z98D;5wNive*~nJBe~A0~kk!V(UzZdEN$u9Iy|@Klb5G#<3goV6<>$#K;e%eCLC zhhqyR4SQ;f>_br$xzSN&z@`*_bHleYN7+l-BUtQJd%wsY ztMDbV;YCn1v7zXso_X=Q-U)+?9n4egW!o`HoCu^Jb&m6gKrO6+Julw+=ocUl+^@y( z7hpuih)yCF|c6y!MImd;C?R#Al7cLU*B0xT)Fm z9Da6<4o*Lazn=$xE} z`v>DmQ(Row!dG^O$2_}jO42ut;O)!KfU(<4(mq!1CQ`AhPJZz7eVrJy4{Diek7xuj zB{P=7pFgY-d!l{M4S zV$@v4OPi#xT+|d}6veb{!Sl-_-9puFC1&SQg))T4D3@AaVu^yac6HQfqnhOOp(p_9Em-4EDqD7!w-(avS5QjWfV!4 zvKE{eMjlRt_o)G@w?Wo*SNje&{JB<~m>!8`FM5KQhx8Ud_V~-ZNNMp6kHi>t%ko z-5W|*WPifM$*^mIfS7wy=H88EruDF61qL$C-2H~-5-bntfWEKLF3vl>I2B60fygy3 zfO67LeX@*VL<(?;EZK^Aw&fG;!(ZC=>&mk!E0&CyD16^STWy|sDF8<7yK0Ot3jufzY#CzuFqwzJ$8;Mhr`&9ig%0f+D?+Tu7ObcX%giG8y1>)Z(Ixzx zKE0~-%>#mpf5Ax3r4*tZkaXTT;ro#dxk)u@*y*mm(nlhNPbdEdkg^CWj6J<<`e^0?cB2DgdRjDe&iS;jL6%tu1ralTdcc~h>N3xHR$!E0aw zL0Nq$b1+aez-sv0cxC!F5{tnj2FalTa^o%^fa;@$k&ZwYHmEiSaf?G;G`kJG>j=}| zOjyC{SNr8;eG|DDkSvy{zuMy?cY{zQ87xgobzfNa5$Sd^%>Lo;&&W!=CyRqaqYkjN z&nRK-T9VyZ(7DC8Bcw6Wux%o~r_jM$s%#o8zQ3n*mhfbO$go&FX%Lb7MdlmQvcA#* zVi@wh9L%7UtyGhm_-WRMZ&o{u$G>H(oOac#Awd(628~M|%p+1)pY)U5Iqxtp{geJO z^y`Q))n$Q*>akC`1 zDBfRzWnHE*A;c28M07(UTmsOvcaa9QMjsSz_P1AKX+&GXhu44p&hh*c$$q3i zo(v*n#+^{0U7EG!l)FlmTApxvw~O3Jc0>XtjBzt$i%E(^@%Q{!d(l!q*8A?9Rsj+a zuSFfwgWA}}cRCGWn6^02w>^-@iM+pu3?Y21QA*@A=Rfzknl=J~vFq zQ%`JNY-MivmDiTe{Bj)G#L1<4&6j(m2`puiuv(KghanI5Eq)?ao$8JI<@uPPtd5+} z)af)_=t{_g7>yh!;{sa9?K?DqitcO2TUY#jOt+`@#nO}H;X05R^k1pOp8)jN=Q&56GroBIM~Su%^9vFyu5 z$*%V%B__MSstRDMwJ7rM>WXb8dndENz9R=IO!|Jo)ZJw{W4ws=357E{T?!{3AP<_{Vt%=%^=@iqxRaFL>eCJ4kH+ zC*uoNMThhWvHI!cn*w;n>0K>O0+%~uE?f-%1vkb1ghVvpW%F2g^JV0T7gXaxmCt4z^&F~ZH* zM85fJ8vDuN%I}Egbh;}VfRm?iG(2Z^2^h-G3gH3ySz_r#((9s%8#T1kU~&*)hnR%i z&{ihW@8+I`B04*8)@R!KHebmq+FlZ!LB%kDvE_PoK+1Zg`p`IBYj$lxf3rBzWlwa! zX4mBi48;^G5`6o&{s#39rbCfXqI2xAKfw3U{7diE#&?OZy79bKFCxo!&r|)S3-)Qp zHAlnHN23{Nt^586uL4g*OxR!Fdc+36(D^WjK!ojH*#I*(KO}2-ul$x&ieiG6vrZwD zHK9N-%yyw@L;F(U-*`$l8k6X&sgH~|OkQ_nhu%=9t=>B{*^9EG=P6&KQSNv2dz|EB z>;ZeUe@orB!xch5o5Luae$^fa)kp7C#q**f9tK(G->X-9%-?A*$MnEgM>G9Q=9SlZSAvZCgF!U}_sA(%YOLH$N=zSmyOP`|=E)0fjFXVVE9x$))6M*FK{=cGQYFR_l7dq$l=(WwftFS(?s4CgWJp@P3i1p zcG>#|l8Ly{ON~K1So+U|s5lzO`-ttf`>dbj3uh78B}eNFzL#j;v!qjg6JNr0dnZX^+EXfwy zCF?_!tmh`f9x6(mRmPga9A&}zQPNj!pwE`|YtZSh`XUw%r)vtfcX0$SbRw z-(I5I)T5ntRh?1APu6wF!lQ9HLPpv1dk(;#jV{@@YFb?tT~TS>;=dFMD}0n?u5eaF z*_GNa;nGu4XUKaCSgm7IV^IBk-u~_sj9p_Me<+#*pc>5QU!|ktZF@Qu)wm2Fc~(Pn z24eB7ulQHgk$;+p2s1pM`DHEI4Bzpg;-A*jjJWJkJ@@*0oCV=gT-Xt;QU?3O5uh}8 zXdD}nWOTR9Iqy=Ddf7s$0HI&Yn+>@?w0oM%_2hFwgHCBl+YTpKuB*P%>A~|*G|t(& zIJWmpM!sKrmd(QyR+hbp!$SWY7mE5go^h_9(pHNsmXYZIP0gBmO7Vz){N4^%NiM^; zQN6@VFejX3S{tBN>DOH|bu;}$%IQG}YWCO|MJfM$0*MZ|VPzNbAYq7Ppu73IkLGgu zbUM=uD!Srzq@#66hg=Oc)$q^eC<5ExTrWYAvu_()h9*?~xp4*^F(NS1^OK_Gtxu@! z-eNfC1$41mLEzX!Li_e=5i>K-XvL0e)cLEna4&*{(M{OMXfJxe@1goKIbu0@7efE2 zYSzV~zGXvj3n`;nFTmc;v_`Zy)n0_JX2DvFxI_PeX{f@71Bp2DMm1RRFxfVEVK<~w z#Ek8ZOjN_t94xW7INsaCWTksA`w;8Ci_`w$*(Gb)93jiZGl_yH7Ft0?@p6j+{`SoI zSF>ETv}r=7)|d7rj&<8mt(cBjg4Y0MBi@hrtr*g*E_k+}VS)J-oTWo<0s zz09F6Yd2fYYJ8nuIeR*3&$qKxDfkIu%-LES-A_cn^J4x@5^FC1N8s~4<^$hM<$Fx<#ltI!ZiLa zDJ-h2vMi#q6l-ee%W3{vB8t#G2J|)!?E=Z?jIXHxTWdJe`fA6KiUqumsSkhE^Tnek z7k_{4nuo+jMKCP5N`A=8KT@vf7+A#C?2c2?|4L@zuE^tF%EO7P;kd(QL`|8Jch3Ea zypZ`eo*TfVnfo;=$IUZacwnYOWnW)2w{mcet_+ZSxJHMjSSUEh+=HBA!x5E0(~KDA z-le&a`Ik8~mR1lD5h0%Lw@Kt%_+1aL+BmzaQ{y~;aFrwO{HU|fP0GZi`fhG`k<-ki z+JoXjmjDLVlBDmcoPPbvR8l8j(WPCv*LrYlc4&L)<4G-^QA)wo!ve4c+ZJO z{MG6n<;ZwyjTu+ovksG?zf)pp7E(BJJ;UQV2A<)W&kb~%eamXmKG13N4VO4P!xJn% zGb%KPWKc@mwOJGu@1uo(=R#DSM@E$E-V2>(lzwg62WD>fqiJUOt=bH|ohL?|XDrTj zgh0CgfY6eu&+tMOFd@k0T;IWuid#mYJN`9hHeQsgsYutL~#>6IKxzMSNf_BvJg5PQn>Q0><56#w+;Xggtw*H9x98 zL`6)y#yUW@gEWbJ$txg3+};4`!GM+vx51NK0p`oTg0D!7xIHChO2CtOlFiptm3-|hI3uACgetgZ;<|3QR439-USKLMk$xa5Ixol*ZdnQH* zi>_^%=>d}rX9(BW=9vhB#-&IyMvop}I+U~NJ*-Y;I+WoV&c0)*6bz^rFWC>^FgRr8 zJ!rvskp5(-x&xiDJ0*fzbfBZF&hy}7%nMq8?8-I!^m&Mb@t3DWT(2*H)yp4=RF}&+ zQr%;Y&9I=(MB$lh^Fd{XJj0E9Pi(*Y>sF2oGFB}!<#D^I* z%Kv+qQmYcKw^%ObBC#}7sMu7epMR~eb@q|y- zf7MUUck&I?^5LjPveEZ(xQbi|F7&c!l(UI)+dWOViLxowdqF>fw}cm^*5_#Y`wgD~ zK!t1>({b4D$%a&RUVR9o3N$gtYNgk-NRIRI_l^lrBwRm4Y!CVy6eI~uvixf)X((19!*xVU)|_uwiBmRKTY_ERbkz=!lBO? z__jlG-PPfLlj>cb7EbB(n1rv^jqL%Lc0$>Tr^aRO3=|9AlEB`T3uMd`_h-P?gxA>b z5Tfa7&&HP*Ro-0ns0^uVURk`G(xzp^aDMzOjXv61_;t0h+akPTEt^2M z*fXWPK_*Maa;*NQmRe2Xx{SKKG^7wPQoU3g>{;C z=7*mzXu_02#P~m!MLS&Lo<{c@gl5#pe}kvx-54xLn)G}^8TE~=Zr&v(H5Ko`%X8s* zKZ2b0$UT*F;Wf{4+VVl1ZqTv;)_LCAiIRmb*Ey1hfRuj|a)tH_=CwjmI^&EPeM(E< zz=zM^WedHVA~wNAy160(_Zg5>`YeBR8v|~jC#+mKq=JYq6Th<}CVp@w7mEJq%#JI? zi;X9?hz(`1j)|klXkMBgH})W8kT0Pq7|BFx8=d-F`2*rQA6fW$BQ7V=HK7>?(^)LyYa7_U(H!|;sq`hnhs+Q zt2V6+??%y|W26Lc7lR19EY-)|ON9_>*QbbEZCUAR2GfaP`S1fL2wQ^Z?Z7?m%9^mB zMG$I@r->g9Ih}IV7kX3@f*}{{g7agzDJLs!H_CZF^8k!tKh+>t^;X%EmO_dc?Fed5 zz@G?oM?aXFkgJ^Ic@ovW-tN~|5kp~(Z)H9OLt}c6NZag7*jJSL;V{Hcp|ZILLMe+` zpX^{OC_FAs3GlbfmzMe+?~`6wUIPEg#O4|t!M2PD?my{$4MaEi*R)j_Q^gHIgC)yX zo3zmfm$HnkdzV%JLRZZhcdCVp*T)n#266t%?Hd-uU;i@2p)`JTe;+?uy~I7{biDM@G8X}%I?Cy6c zP`?!At+*GgO|Fw!Nl2#~w!GAl8U zvT*iWYZNhN+6k@IEHZR*q7hk;-eiikAoX9fp(mAW&Ie?}fm{#naH3a8P3Xf4SAa#( zr(vJmo(1rm1%fh0r%yG%M0@mcx;-N|eLe8HLu7R~)i{My5dM|J5MJ>&|6RrfIIQ$& z0J9qY+S{V@sN?9(@!e_7blSP_~_I7=oF$Klw{xu5Y8)6e596k1J2uKCqc4hi^k+@Ius z8*zS~G!_lWN|Ky>zgOuBhu-eF%rP2vj9Gdd)dUo#0ShEpwSy%yk750~>qW_;YgHnO zj&T6zjGU06Q5^eO9QdOw-P3N8HCI<9a=uW+y`CZ41FRivzTqO1l~ zEMj?HrNDpz6{+uBFH&&wGoxZt?q(qNxhjW(^;Zy&ct6Snkr8$|K( zeeSh0V;73|x7?IGX}KN#dQ~ha4)!rU5hmLW20kG#5L`@ZwLU$m_4Rjk0XhA!7skIeLvcgD&>%qZA z_(-Q`8*dGPIPQ=wQ^1N0%(UULw0~bHV-53@xIFx2Mt|??gjnK$MGL)5LOAXODC>#~ zVhz`+EKit@2*9<}AzXK@Y%w5}iu|PHhGO7UY03DXz_^r@rE^@RFO=&(X&KsfzXbB6 z?HaPWmS&}paS1ATf*~)iW=!S?t`kM39^x`svxLLkOV4{Mo{5;aWVrjTe0=sm5B$5} zcF#l$;`4_-7}?u@IFXvJo2V7^Oe|LSjnsYRpRN=cs2(k^!hw|VzSLB;#y|v5t7YUL zWLhVx$5iOo|E(a(5HEK>-!0EK zzcsiw3aD!YXN~to9qGbzl>N%-Mq#gc?dMbAbfy+qZ39KcuLv|81b-m zmHVmOw>zc<5z~miqRRBD49OS$dUp}oef&XQw-BSrriN+0sL7=wp^ay6Crw{+ZMf+5 zFQ8oR8|AX22J^3qJ73AD9IL8VnWowp+ip$OyQ76H!PEj${EEDmwQ zC@}X9ambdfJU0vL4(R@>^EHL!Q(KGE>B?)wN3Xi8#80!8`jUpP5Rf(mG{3;^fS%g2 z9CKl~o&|WZ9DAbkGeh%F?=tw_#j`ZX!{}$~f1sc9QFPnkOWi!f5g=wmtq{L=u+Ye* zdDWO}jTL9qrJ1G)CIKZ!AZ^!S#<~PXHa`g^r)2R(HMn?64sH@|*)6CWUU~Lt8bA04 z+vjv!;KJ)=*Pr9yrubNSQGDEi=n?m$a;)BV#ysWupNx3{8>~DH3%Ex#T4XGpWdQo4 zBfz2mQydXcvh?4h-S1SEWozASY zu_0Zj*+>suo1D)ew(gRQZG3d;(z)vGk7|;DBQA+@Wcn&Yk;IQx_9BTiRrKV=M-rQ; zURi39O2 zk!j}{iy+URkT=p-5oOK*CWdv|*n7m`R40Z3*?G2-Jltv0e$Zh2DE1(6qQO6K@;8^c zT~M%C0|W)QKC|LPerz_Ad;cHR@rFnQ~!gdg-1vZ`?q=k^6Wp#nVw zb~!*`q9_aTHygz+PS19PIvo`Qak5p^1=aT@pzaxFSZDQF+z%SgR=tSB4_Z*Gl3iSk z@(7h<5q@;!k2$;GOtB00laM?(wjihEEj~G5xgrNQm`!G)t0(SMFUwiHAUnrUfjbSp z?aP5)mNIrlc1dmN;eo_ji5shuEJ6M1&`i@OGAu%8{`NG;Rr|(S$t63tfY3kNm3Ch( zZaYGV<%--`Co(wuN+I6ZZt6chp6F$PYSGev-EYyDJNqrj5`8H1y16(=Lwa1EMm1g$~x(9>d>>T&%;~x(; zM?GA1%6kbnAo914b{O-@geljeqX4eK9|sH)po)`qyTG>%=KkE%z50N6^Y20sz4oM@o#^@aaIXH~o;LqQl{tQgHB8Zvb z7n%>+zxXbuM?~RU8ajg!Gm{EMrT@uAfQFF%(QsJ7B@JthCRz)zu@3uF+CDs$aapa>cqj#%e!+H?4VcTm z4nP$9wu5J<5w2IQ3b}M7BQQ@-@^^BimBX-FwOrGZZLwlS!?J5BS5YtjeEOloQsB=! z_=pVs0V$>4XKDyRfduIJX@eVFHGwr%=E7eS)XXD$qqLg=plUfKZz4Kx@X7pRY;s#! zNh(Np76cvqr$bJNSSa~1Ph?zDk23jsMZ}w0d8?G(5iZHqwMy_=e?^llv!++uuDy!7 zk>|nIUi)_24!w&%F!9pa?JlA$;+R{@I`ErMi;v&LpZqNHT5=QDyX#^vkDycZoC>3! z`zy{Tg!o_uG$yo_;O#3|$LW^MpX*e^k53d%2wSp#pZ1HGLE%q%Jsnq2@c~yi(4@JG zwmJQt?hX=BDNu^&r$?({%Vpm^+?9>Bl^`Wj0@j$ukI%J}mDNq9|5RJfu_pHpjtr6e z_acojW^6VVV9F?xb_W;j!Bf-q-${fxgl(OFNFeL&Y~nyi2~7^14)IdLzl~IFEnnwx;YD=$59sJ(A_iVr?E04 zYv={VjpyXcCOg{{*L&d&O<cC`P4vA2whvuVNx z3GVI$_rcxWeFla=aDuxhxCM6zgS$&0xI=Jv4;tJdc<`M(?|$d(p8c~wxU0Ilx~jXG z(^GSIcU|GI$7jqGyjG^_jz!-{$?}EJ1mtvqD?F{eRd@N&}y8rn5r3EB+99xgW5?H&VWv_&BYj;YPTxOZMu2z&gI*HiZTi zEqNTU9!3`kp*_0)^D{KZW<41k=b2HTN1LEhz2U`VuhN!%A~h4t&X2(oZ^*V#A+MFx5D~O%TvBd?U>ax#{`W|cfp$fkO$ZI( z@$aFY3T>hf_147>wLUumwKApXMtDPFGZ`E;2lX{MELsJ16S57RCIko#wCCFU(1wo` zPzAsXwDoNRl=Ftp8G=VNR1xEL=8v0qF=aMpqb?y}{?o@p;Cvl0pKuHsII^OVN9hqF z;@7geGui%EF`Zv#(UU)J6xK)6@{>-wZ=R+lU)DNJ4<2vy@IWTdQxF1Wk3X`E{hAK& zACWj26C}GNtE~2~tJYC1GrU|}xQ&XO_#YKotoA$soF>^vH&<8d9P$)&I=gp%ifbHm z5pOfy$xAzf2#(ei0;+$gqkR;kewuQ2N=bBN0osSs(EI8aI}#GSxqlY!m3X z=bfBd`hF!}uWD49nTeQnUS~(H54a z<9D&VXZQz^{1%U3oQ?d2N*Gt6Ch_-I<#Pj=Zc@+5QlF@i}lF)9bfvSmqjs#WmR{3pWGV9Yd>T8 ze!APJ^L9kjIjbx&OZWM{LPQ(y$1{+VQlb%YzWhqs5`^qNbhp%fYoYz+dkd`om(yQ! zvvrmU+W$9`_&)}$%KOD$N9$%m=ZpOmc9xk-Bh~`TrW@bEF^Lzn;YDl4W6WXPt(cBm zTq8;d0kr1UBVXYIC!u=C0_}-U_5S83$nM%Q`*X3!M8-Ry)fc+u0H^KD+C7E}GQE|_4yx-{eOo^uVEZ4ZqmJ5`t zJo&dIGn8aLhSl#7zVP3*89*E>Wd5I&yz($1wv%0F-OOY7bGCxm>2ZAQdV%&x<%3x( ze6oih9Sp zf~mI`w{9z16TzB|nyib797k80-xmbHX~M1R#pl>ctEM+6(&{K`bPK$7=NwS^JEr2_ zzZE?T3J7i2uP_yQr(~e;J8G#U-;X8gpt&RvNRNm7a>5E8Sq=YA;o6O zfm8To4~;Z=h3d|clOy6MzKfXM(w#U;yX5jsw9V;g)SpAc(N-yFE&dn9gzqALx4b8D zG4$Rr|1{;3JW9;)1o0_qpsP z>a3n83uqx*W_aU^z@6LcA2#?h-z%1@ZSeDN`DVDlb4=1Fb_HhbDJLO|OR_N*qD1bZ zf)EOg7AcF)OGrwZ@O-bgc3J4{F%bL>oKL8x8LG#tL>^_Egor;w=fPB=^siz*zl^~~ za-usY&>PNa<+n*+CLLze$VXVDr=dkb%+D54xJU(@b$~dw7MpQhf+51Sc`N%UGc?Df zT6hIP1?O)tkpj%sA}vY`yKP1oSWUl6x&}%Fpd{|GMBwVAt#Xj8N3E_}O&}I$jSgrq zRlPvU*20W@@baM6h+TxBc5o`qJ1?C;$h&Sx_p;9I)t{L`%oLDEcy*_4SGPxm^( z7)N&xVwngXNey(j!sG}DHYo>2cSFBJf{JB8K^BrY|IvT^|0xeZM^dwIkkSp%`&K7s zLQu8!i>4-OkFzj!t9)n{UN@fW(u}JzwBM59$CK6H`y~#O0-+K(Ome!hdyShf zCLc}8@)+-)hJN5*URsV%gCxQ)-i8OCE>AXpzTQ%Hy}>aA$77A7T`L$;J`l9LguW5z z@(owT2<39#VM+9S_(KU)d3vX~4f_o)me1Va=*j8r?$+x0a5Uc^f0gmV@p>6P;^(uP zuSRYasd40&i*zfj*iB=lFD}P(6aqUeOXNMBi*rUACycjXsVIrb-)MQVPaQez5-&RH zBKr!6kHbfqzGRqBiF!5k?>qr0aJ`)Sw2k*E97`MK{;+VRcbO77n*k3~99efAO49b5 z*$1?XMzAb1-=>H@A%nG z5Wii^GNRnLa$ufC3Ax;qYTS~wdBzyJM4`DY*D`3KWX+XbA(8fcPZ7~aj;}50vix}3 zQjL9h&^W_yJ-QT0gxyZM&1WD!(66OSlrTv2(s9S8B(B3CB{9@xFVz05X623|A0%Wd zU)S}(9FtwU_vLe0v;J26?H^JY1W^p%iRtDmoOZR<5FTUpoL$}whKHRiD z_#Zc)Z;%i+e+rg|E?ZBuzG3PdX*JV%_oU|f(L;#%@*W_#A&B!AG_6E=1Yd0~es~id z9qa|s>0*A4uSud!MfVNepjADPs3p2_jo&Xl5zEN^ac}Lf?$n*-@^E9c=^elCzvvUf zOvWh8mg2KpTYY~O$lTWTa0IP5kYM+Xj*;eqo=4bz3U6K;@O_;!eE#_5vdswmc=5x; z0Zr#I*Kc5Ra(c7t5#qcmI@Jb&ej1u5u8SV{$|Cwaa4$}M7^5!Q4vJWTQ|Lb2Hy&Nw zyy4y!)`CYwVGKYGVm9)}n;+<6`8kicP~4(^)}Z^!KonJJ>Ys(i-z zZ~9WD)sc=RR0m=3FNpE2;zS~ z1%=rBH~fEE(a*(&!0}0v=D+^Y&l7)!-V%RuwDOr#1~AsK4?AQ4jee>z622VZKRDQ4 zG&8m&o;U=PU}8&N>N(>1qCQEkl*|iA7&^^8`*x~byLn^n$BdQUKQ6BF-q&xG;Jt8- zz)_!GT)Z~@XtvpcCZ(6=68SgJV$CClxBSI#t!f;bB?-o@YSeGO+0i%4bz%EJ4vcJL zz1t9P!vh+*4};V2#~#_0SzD0&sh=_jG$RL#)2|oPt%_F#eD2R8?(9j|rWo08nAy?& zlvVj-MhA~?6v4L)|KEHq0{=hrweSgX^8CN^wX9b2wL0v-_(%DXQ;ikNl+&4Tb}z^I ziA4s9O45GjJjBF4cnP9r)+UgPGN=p+U>o}C0yna8>vp0DqKrq~BzwM(fA=~yvwg4e z(>K)V`63sZu_fT|$$uGJ&Y&4vZbL)(PB%?gukVfD&Q7LoEM#V1;Syf&{>#{s@L$H3 zEqmdV3*w(|H)Bh0P+X9ouLsbKEmJQ}HfC>x13w28Vroz6xie_2M}y^?v?zrxnESJ5 zPI=Vg&@$+tGKMMMr=~m`*=vme?i9v(1qc1D0Nelp0WvesQy=!*7MnY@+4u1~f|}|v z@~YG&dDd5ybZVAwx_c57oC#-TJtT_iSjRigDNNR4188>+zrqQW; z+h)!3lQIUH^jDX!g=jI|CG@f)(cMzt42 z{v_a~V!3Mw?_5StnSEV8-uMgBu|vKRXy7N`VSR#qsI=aFG&hgoJZoQ>Ga!f+9khv$ zlCPSoej0VY#d zR!giuCGtM?5gn2sjHpt_IUwRtGttRQy`z?=5fBc)Q$&%AWutP9L3J+cl9W{J=Hg%J z=h}OMG2*|oHG3;-*+TTj1a&14&?bo(>B!o~n%KT@uE}oQ2P0!zX0Il;;Qpad3ZnOlmo(DQKU3$B_zb70vHuk$4cHoSC*M2<=a9UauEt(RBHRVm^! z?<0OMjF?Y(53>>BCYA%=FjqsJx*k^&S!bXMj-yM29xUeU`#7>fQHy$6atEJWm_cVZ z_KpwxQbX8oXeXh$e5DDoek!T4IH~I$-If^N!^P0eh4KqLbG?UIke?{c?;KBL_HRSt zVK0f9yWHgj!Au36E7@NWCC#MhJf?MxsFX7qW(V=t!1xIp!lhS=+8bD-6~S-{iu;Z5 zX>^J;`KsiZB_K=4U&Sk2P$q0m@vw?4Mtk7lg?v^f6EjVM|0n9=BN%KYrtROAZ0}?X zBCaHC`XXsG-6}ITa2*Ig+DGaIqTbg^Baa^9@Fi~3i#{C-Dq&jRK9m%(ov6+5CqH5{ zO4_bcE2CQ6!$d5{J?QC^h~txL`B8@e6Vxj1kg4es@t07jWY3$3eX6uV#>&X_>&suhMq5cH$k}5KQEk*rZg%^d#=iI6e!>=jz`@AVcF9F_S_Jgcm0+pr; z%r>1nTulA8wc)QK?1UBm(f4w6UYO?UE5$`XdEypNl^m%^{{9ApMBUPkP=$04f!0oI zc^tZ{kV0~~gkgeFQ!WmhjxiV}eh)Oj*Re|`<|mW=zaeC#-_jydWQnxfQbG2!nn(s<1)abZ!=*3CnZ%LH`5Yx z@0Vw|+o0F&{}YU^z)!faC{)-Kn${S9EPP8tC5^ZfdvAxt_Ii{^TgJoLnt2ii8Ibkm&ptfhy6olXhmvJW(1Ld~BnuCP|-}07Z>8F%^ z6E?)YFgde!|7F9a*ahQ$Ja@i)ZyuCY){hdz!qXoEYkLfea2AYPZGmFQOGi;!A0Zl9 zyTuQ=&lj!)i59^T#y!O#x43R2||Kv`*8f60gcTL3x#PR z8J;=b6dOpHZ3#Qyrt~u-?T$;xt-jZ6(I+vw%!q9M5etzS*VoIo(`}?rN8;eLBKu+M?8~nyISbM@{{LpN~&7bc(r-vWeVCILH}ROrLRl0q^ersrM|7^`Z(+v zO+h)_AfaJ- zOiM-)7ukA`=HlzURVFCRGe{_$04* zfI$vFW_I$yGCXqzmXKS1!P6~fOShpIT_BI4Iqjrl0ryev=6qqcw?k~!H{8;FH)Fe& ziu8x8TyEh|FxOY#3m2{p?=u`1OcRqYXP|zI`9Ev7(o5c`fofueF?8FU_`T#rYc@72 z3d?KS^k$dr@!z?p^L|owYqk_3byk}5YPH&J9v4(;=b`mB9_QqT(5Od_L8KU=7z8u= z3-C1Rn<2tLgu7aP7u$uF>0#Ff!HXu;L6}%1p!wLjL;NC+u=P24h!h=pPYaj{nK^9m z%tc{AFyb476A=CSkZEGofmb1!R7G%Yy}Bk|E)d>=sqz-*-r&Gt{t~Yyw&=x6fWY=G zLm7b_29N)@tV;vLPMobQxeKMhIqe~9ZNynbuqo|36z3PFTp9X4>u~f`WepFFyxG5M!-FDq^; zY+-?d3kVTJN-GtA2$rQzKhE-lw@2JoF+2}13U0(Q|K!q7^o$q5!TZshrPhoXBRKWJ z{WR$qHO&LQTYWov93E5Q!V3;WY~s?Xy^b=Z1mN?QwhqT;a6S8iZ~Wj%lXizo3rcMKg6LZWiXe*?hGVSt08lDJii|)edlO zTD!&}WTz9Xi^b)n!H}z15Cv}A6eM_elXp#);Ul9RAaart=(Xp~fIhg$DsThXnAOol z7?XS&I3jMgv34G0>Af0vx5&5FBj=C>m}pRhTx|}RO1}<&9B;!y+OidL;6cKM-qw;M zVGb@C4^EWMQuc}Fh{i0`jOylc9vUp?JtQl+^0T!wU4}x?iG!4Ej6Q&s3-T6ht<&w} zKGEMboPzYKfJ;g*R6=0q9w&^iNF@}Zd>9kM8OkKGfq44yGQS;A5;eDy%0e*&lf)Ogv&e~3}vuaaGsj5u4@gQAjT#9(5e>Rh*ApWy{QbHnQU+z?0d z!5Vw&&*~-gnbC?3l8bd;4IBZvG@pW%f0;-Mv62+Vu0dc*6!pJGj~Su>pCYXHwQauI zkmVudt&8PI?(}(mS#2k8uF2B`(0R`xmehC=#9|#_IK@+3$A3Cj?&^%karltW@-W$g zi@b#))^PrQ#}Yb_DxE9Fd~{g5XZD9isO0gS|=-;XC@1riPDOp4i0MODY5L zFf=sli#t#_d0DLy3MHVq%WT8Pcb}~oJbe^oYZgm>T<{n3gzmb1Q=1jZKFlu~Gy^>^jX3en%r#F^z< zG;o{4q)>orCd<{5r4f0u??D+%n5w!e7Af}`sD7yFs_B!5K{`YZBM_L6WviN^Gt*(J zIJz6+unHHgiOh3d1Xp4F=272Gz#N`&vu9F?TZJBJ#oVQuK_M_Tu7^nM=^3c7-$O|1 zEo@KjU4KFU$2fz$kra-*F%P>`wGfD*LGKOPiDMBG=(#t8x|* zd|ClA-6Qk2TS=<4!iid9E)CT371vZ2yb^;;l*Latr+$)F)6ULUehKIJ-A9e{Kn8YA zc5D;u^%PWV^%fkmk924BgOb;>zdoKrx1wX-nN4)YA_vZS=nGT5cor*|@UIt=`bib4 z;Mo@~o^-DjpR?3H@CViu6mPD%ETCtAVSI%fi>XM$@W`*w6(zV!Jk48>_t!skpXmab zM{uvEH5mbrb3;j@BO`1d`rjr-zop0t5DT{vC{D?He{bW7E?1KB$fq!JE3G_-dabsx zAuw?yWm*SB7=?8nyOP(g z!9!y$I^T5OZ^=!hncPyOQUxH)x2e4o2NXA?yYL^eyh~AxLLmsQvXLqV;UgOlR0wll zqpIKN!q)f-we?MkH$=qW&60nakSgZ{s>f(fUQ{qwYSd@stOhSIoqN-MIJb4 zsWSWsPvO{JX0W#S9axIeyzb!Da#rB#k}33v(wttH>#Sd2FwXF=QZsHfgJ4x?i{GvF zUzw>MrI`;$;ttl)j)v1dc6Hn@&3UyGi^9?FL&Ijs>bM1p^V6{l8xJxclZ z+(kn`vtu<$N+V*}J1R}!G5BerPr1@Ya^gptgV=*YLOOr-H^086IJf21*D#F{fVB!9j@&reisF2-iT%8F4Vl$8C5xQ3iNfqEk8YYj)G}x&)-qZNR*q#URTx*= zSamh%o{6T|TQSqqx80WiP>5tyY@NGZEmlEenkrB?ZYQl>RO;aEVET~lJ^TbZHw$h5 zOst-L3wzc;KmO*jR9P&i4`?u3m7ji?PeJV}rvZMK=zG zB!ef%r09<}VV+B7yNbxLcmR7-x|$=q6%PH^p^?haV9_OOpDMgUbR`TksC+2&-ftmg zynM9MrF@tn2A?EHM!s>Fm1K`e5YoaB-$t7H-Y<5{MJcv;?Gnd6E;tBC^lPMP_l*Bx zLBYOIj2$c?nM-R-PnN1FF-y>B{R6P5C8RB2Wl#%4Rb{Sd?hxrmchO!jtE?>K_LA)v zqTND+P7JAF!L3GylP?}0gyM!z3_3@+P6gIX0VOqbPZMd>-2l4KR>MJKgd0P_mV0}sM5-mgezASJ~T*jK8fvlQEjb=rH zhT*owA^(N;CIt0KH0f8%Z3wDQ^A4t+u@7#@9Bl=qf<~~jym5`S{^Yfzg6TGOeQkl} zpvBGZ1sQ`U^lR^r^X`<_j_oC?qWO*{qG=#3!ikP|M^q?<`fUoRJnz-By!r017Phb4 z9=Zl0ZfQw197HtIH^zK0z07MlZfZ!HWk;A1U$x;t1MpSX8QK=j(J~R1K5W{cVbIFN zx+lOgRigpQN4pnFeZV7#c^f(EuvYXdwRy9bk2-wpR&i809M?Z5@XGuXs=l|0$+sk# zq3Qn-B;`4}4ynlr;~@qob7nVg&{Y3jT#WL!GHu zZZO?(u%!hyGY-fO-@`lNjFCe}hDZcWXy9^n=~TUxG7?gj zhFYoy@!H8W2QMQ+LL z=TdK4t#3MJ%mV5SOBB(@*=cj4HpG3t!h%^%zzhQXI(Ciu4qKI0tL<2Z387$d=Rzen zZM?-2Fj)8vnq7+Vu0>N&h3JyGV)og2)gB_fn~kY7tWC|}j?BY=_Fh`G4BIlMqGuR0 z^^Cb9Ea}6Vi)n|kBUFr>-MP^{-*q_C?i+02Lk7Yx$@zV3Py2~fu2{S3v&@!{0YJ8t zwX6XbfoS|g`>aizc;O~-y)nNNCD(U`1v<y)zHu*WMCtLroE%F{ywByb~eI98b$)1*%}cK(|M2@qI8cnbU>bXY|+_(b3l3D8Ev zUM&mRY6y*p?Nt|>906x5f>03P=F<5V0NFzJePDMP9mpiQ-KO_>g&)OzU?Wag`Eb@w zTs)EoMjyPYNJdm)2ucM#*eAGS1%e8ZT6RKZU}{9FdbmQ>&MB@FyUJ%*IQl64Vx5{= zNJ>O%?t1^qL-E{MWQzuXIL6h{m9u~!>wqb;UT5VlrX!0~KY(8RoXstXh}+qx()6j} z?E?lsg1yNf91S2y(PqUIvNjh$DJ_>nspKYM5by_W&u*1$zLkk`);)+pgeXSle1Nyp z#k6R2rZZMO-Wtf>2XDU7fS@+ul@_R)6WIP&O30PRJ=uvD95X#?gD7NckaknNid)(Q zbC7jZcBdo)gj61vKrB{kz!yyPJ!Sp}c1bPvfu4)H83#O@;0ken3+O?pV(F-~i;EJ? z9u5@%0jA3bMN=~!8soXLuVm?1m*;8}O%5&CJM+1B-t zYU}ct1I7l(dh#BNdh5T8{+K^Vi zI{`>r#@t!Gm4nhBmo;&p#gq@8!f@cRGATImFZcGbO2y@H(F zuOXZIrFYi?@?0LJLok8oHv{!ebP5TK>LBKex4Y%X9n}s6W*rbUIfajp_K_69LdxXM zCXDZmN_1l}06A=-C3%7ECpY_e$AW_r8L{P4Jn=hdb`p0@j_^}wKNy^jaB2FbtR_i# ztjOMsq#z=yAhT8GltuAn{4-Vb{aT_>GIS{`$^hRf87!5~0~dqf$#ZN6N2^9AAM)Y- zlCWawco4atu-tdDyw%xI#GOwD!mjxp^B?pZ>Dk;`Sr9rgg5Y9**|M!d|FrhQd3NnLObtZQ@HB#>*4HHxd`_Dn1DaP^mjvvCZcG9 zkyR|3RUEzH5$1O`n{K!xj}oTk=THsOEI#(-L+$(?&xID}LG3Zfu!#yxj6oT#;MzwN z?^SamR|C`cylstCyDimcl-2Oy_Mw>35#)uYdz@Iu4CTg9XyLF)hYxxLmZsJFY%Yl4 ze(IRuPH8O7T;y0&&>?ZAYN@!!izd}!A;lZ_vgKGp(5*i?4)oTupB2MQPJ6SQPRl%s zdk>(cCta66tnI^N`f2O$N>Lj|_2KCfhhY(06<=`Q1#=O59;`RTnku!PVk|P~QqcrH z5@n?adUvWoFY&R`=RxDmWs%^szp}6<#XWN&0b0yo8JSQ^T{fWVA9zUyy%Yl8(f{V1 zTx}-!Gwz8rW>aC0>`QN+lrCiPwJ?^8pS^|SkW~k)Odu^XzNk{avZ~AZL-C%;!6Z#b zmi}p-Vrx5ESxyTvDn#RsV%OZmdz_emF*^KLUg5M|xbB6Ot;(d!6dJ&xr!t!2Yu#@G z7@JOm^f^LtA7R+6EQlcIX(N*0FJ--+Xb$(CRKID|uv3T37W@Y9tVtrDY!~vsy_3|Y z`SWU+-j%x2sutEOlB%Z`nFOE*%h4e5@=JEuLfa_^8W^p`W`Sms4RFpx0!WBoMt*M@ zodjaTV{Rtk5GF8z#>`x7WFe*e*T5;PO$cV99~pR40w^W#^id5kjZ z$%_G|&nYpMl1Evo2ha)G3|}$e_M4Ux9H=W+_zsjl}Z`6U1DSXsFj*RTc(;Q@AE#=48kuQd=zOvA((FvZF3B!Oi zLW?T|-j)~G5a;*#Rx!%-{`Sx*njg{8geU9dH891EJO_1()EnnpsmC`OLcV1Q1+Pi4 zF#a-#*rOWBU+ZX$C$3nBeKDYGHPn_kn5+z zyzyuGpF6a{FE7aY?1{{l=8CKBu}XAS!1T`5&iCe60Jp)=rV~);$VZL_!+3z|A=*ZuCHtGnW?AAd^?na1o9o2e1`y0 z=d;mSFU?feW`dm*IuZYsC3C5fCZo9~Q~=sTz4JPlPkJ!#ul=<%n${xEf#p@lIj3bPH_QQ<4Vpo2 z?^s`GISgu)->Uwu0Abl61V82@0%?cfG@G&{ua(HK6oqyQ^cJMH0Vq5pKP$_?)t+s* ztoAlZIK3T`9;{`_E3xd=i6*9QlX1w1D!*H2iD)x_|4l>+1&wfTZTX(B{E}BC(2-r~ zcV&H?18LrJODrGS4MmKVI76g|q_9+e&H_2Ezw-Qb72`g)-~RV|gH3cWr;=hp+N9)k zQH)ky|F8{JIXxjWqG>Z|frtQuD+2){9P3|`xMPNvMWIZv*T_ha=6vv1 zP^=u#z0N?9I&)cc6}2^UZn<%k;w=p#tEw!sanfT3ZnX;E;|E^n#oUS)r8elO7z}hu}{n2Gh(+|u5BCv+1t*Ha9jWa(g1om zE}Gl(+uJC{!rAUI5RM5?k$U1%HxmkUZnLiwUq=fLw)OQowrmOvt2kC=Z64#u+|gx1$;+=O0NMAX_e6v{5tqIma<3l^3J{-JpK9g_pw>vIZcr^0+ejJ;t-dAdF*jALag>eGx@oOiPfg*%AI zlt2AiZxj(Rf7VjL(a|D9bQ?NXmS%cQ)b$fvVK9HfSl8H-j3oRHGr#pWF_gA&w~0gL z7+rq`24m}h9k;#Ek;;~?xWcN8?IpS#Hb9yc-jjqk(GBeuBS2K&dV1%hlrzzQ{4kP8 zr#o`aPL%f%^dbb@iI=E2i~h2X%T{C%-}RIVh9LTDK8SbzY0KKp-O=*3f9s-C1(9#C z?to2A*p-jy-SxnBrgExQ+bywtvGSv6?}PNVS}|+;-sh1Ohdtj^!}fCb{KTh^!iy>9 z7%^YU#Y~nch|=f7FlH%;Mt}1}&rlE@S0CVg{mynIf~4ekx<1!CNRP>JTy6Lxt*l4~ zTjd_pI897J5RytcOX1g@Yz>XYlZ7{mGt1}#KqC;biIZX~fvDrzvC(C__?|tF?hih3 zyQclv5$oBsVFDO}6W(CcoiY;-!L#r157a(Ve2a?}s@Vmz8Y7U(Q|Nw*#>l%Jtm(7w zItYWC-t*88u6C2UPCmL7<*JLA;859oG@X}Jac)wjj>(U8+6T;`uWoVXG&@RIr;YTI zbUR7`&XY%6>y1O%Go|f!f+ryD5p^AH7oTt%Kr@^zH-H2jJ9-va#ZNSd`-4;8r@27XhvDxNCR1Qga$hKF7D&VpRhn|rPhu$O=RETl z_YXsoY$S!{i!1mVPRj8!(MI%zI-CR`<5z0k#!rFalZVXG1${DMZsr}hq|XWoH4}Kd zRbYps1GSY;29I9f>I#y?Xc|Tn3rOE*$y`-_)qiRlrAzm2OZA)O{utl`Napd5@=E|e z_ISql9inLQ|FBFJ@(CE;kn&}6lW!=R5F#laalTl_g{cJPR?|F*79Ch4u)QS08xtK#0#eBXKlh7?IERFi%Ny!H~Uy(>AZAzb|O_bFG-MLPjJWbTLD_e2{Ca|>j_YyN?RZ!VI^8!O^zOkC zwncbIe)O*0G&O8mo|YG1M5s{+&2H$LdmKM|`E$F1m8Bx}_TGE^Dfl+MCL1mJNpi72 z5^nk4#%e~>h>6s3ALd%b=^)&#ACrjC1>RcPX$Se#+!=m6%ng1Oq&WbC(s%o3jHpC0 zzMtpr<>1H%fl3;L$77YY=_MkOp~atN)-O${OF#V^zd_^+x&J=?o!>S-tEI^r`n;sPYJ(EHKgJe&9*oOJ|gY$rx) zxbiQ<&n418b39q`n{G1l|EgtW__c2Nne*rMXMfSm(?;w}7f)-9I0jS}1zXOGX9hK9hW~wkIUp-B@;1Bq8j)W}^YF?{ zV0o8G*DuOm{io{YJ2}SDy{upT&;HHV^)s&`%kasFpT6D6KfLop=eJ3D23C6d$Kz+_ z>g7On*mXe7Z-o5iDU+`mv_0^Bk^t z-o?ax%Pj{^NVmh2tYviycgc8|gxLff^yl}c`Y^>;wSXOTIX}mv=5i3zpTDmN{9B`g zd|y^GDjOg_T3lf@AEpp`soUw=ssO1LKoSMqsNAoRvlEDutCdq_pyGc)Cq_n*HF_CT z_1?w6#v>6M%}z(Z>QJzuP!Q0={?1Bi-78wLMX;?_Uc>H76$q;%2)WrUaz}*w zL$P2c%CxoXqTS@zvZ%lZv?AN_-z#wj+>Tm_Qfkk}d{uq?fU_ji0oeaqv+>n?_dwA2 zN}qzR@GOH!|2qX9MkU*a`0k|v!jsXZr%%zhcH&2&Z1PCMpQ4NBV730v6RH>u>9ThY zB>g(#qiiG)-mYY8lBSO^4UznDxW_57*q={CE%1e&Q?5Z`=J=FWSkjhbxIBH`9LL?` z#=vY+C*aRpMD?Ee$HY>*UquMte=Qr8>2N9~R+JhGYB$KmV&( z1wZS3E?aJ7E#L5B+w}8Vx{@od*7RqCqPfZk8IK7-V!r%_%IFb>+}SoAMr^&G_0&^- zn6;%i0U)iu#$#W?0SQ5dP`wViheltm@>CkAErG+AB}#(R5>}qY_O{nQZCGGD>yaAy zRnHfv0-J2gSepIePt4japG9PZ$#V+y+lEktLy+q}dtdcQ*|G_sL z1G~?g0oqa)KCu3E+7pOOOf(z zfs?N}A;@?TKuoNY&y6B#lfrQ=8!Po%r6VUmd-{t{(IuGf1XO?~DQD&9wU`n$-rpZ3 z-?J3;g+bJ9(6g38`%1SRa9q>m|70GE#ag(uJ1P3a(k;@y?*+ka5{o-;aG#{vCLb*; z+rXsR5Tcy^1djG*cQ@IQrN3wmyab7j%2kzOz|hEdsaMkNPz41{hUXX`xWcoU;?Fle zph%&0wi>RbpFQ28C$rJw6lXya>^EeSXXR~N7E@#i12re%vGnC>l!fpf?fs>aT|W%f zX^0jidr{(7bT66Di?5Y!cXAUZ#&2jaSfU+?5-*|H)j8>2;R*kWQy#_c2AsY!C$3AB zB)Ip%oft3H$)GoW>4Ph^&22@W4k?g@RD4E9+8JN-eG$x;+xxz5kEYU4mUZE{Vy~hd zMvw*sA2O3bN(s_DCEB;zJpyDLWxmB5I1JdI`A?;T%0)Z*kH^%&dpxX>`Hff7tNsVq zq*9>vWqRmEiV36)xZqxj^e{?0h9WWglf&hoZ5T&1YG{Vnb^4b1 zB_X%PrShMClb!Y`>JmDMpZ2)8;I0(v+FX%K(6u-3ngSMKN5blf9wKta?~} zqQIEW6|oY-YqBrB0vKZd+#io~A-r?fDCb{xIasH#P>stqM*I8$A@EuM(@tRn7~~Q} zC%aX81AmnfdtC8@Qsb&%?wpx@_pvaeo^Eycnz7hT-@1GiR7qxdZC@SAQ9mDsN@KENRlDbpwr-i zp8RQ#6|R(n0yuJh=>!wT1tD`(!~@VQhLZ-a1X26Ekk-wN%F%bK5c%IAqcvD0&RSLA zFc487=pR7}19cW~m;Leu+Xy;Y`apnAf?^IyDj`Vjo-5}~;^hJ{pW1&0rvhCft^(Ql z!a){;M)b4q)>}kotH-ZJ=Z z9HlL<1SLr(URkHpO;ajafa~gT=n-z(grbgL^uw#I{Io_pWtQsoAX8`LLb-+QE+mxi zqi{P$-C&8X%tR*A_fvqaC7_IExuz; zcC1L6BDPX5&UwRFlEG2eSLm#m7E+buD|`Ib33WxbH2X$Ei*kbfo&teNm*2y?>PkKi!pb`yM;fXxZ4ut=mvo6h1ydQ^f~y#+`WZa&K{}3NVW-KcI=-Lbw;j!ijca*# z+3r_PRlpk9)wC516zh6{;VgIl@)|Xa{+R&2OVFq)#K)yW59NOQ`_-)kJ@>_=SR%=| zxh^}m6F|cpk*7R8dYnqjO9H^1s(;0hNMg?LtZBXh2BC5?aR0&_2CS<>mG&~C$4yEM z+!9C1cHtCM%~z0uqE(v?k^&r|woVKQOaf5=3H zWjDQb^2sW+n4LoU(dbQ)ASGvdjUWXU^0sC+z#V?TfwV~1;op-=tm)|djP{y5tNqB) z@BR80Z@T%K=*e2K&8Y(paGlp?SLxy(n}&-Gl9vdvrE8-1 z{A7E1>vC@QSpmW1PpOojonH3i$*Rl!)D-ZJukD00fVswVGa--98F38r2l;!KJ3vZ$ zZv)9^Mk7Ujpx0SmbIoE8Z3$rrX5Wh0DKW%*FD{3)lvrdT6zico&rtCh2eBo2&P1nL9=_S<2ULiM^k+yrLl6 z=-Vo&N(b3{VuhPIjMxW${>eqCmK}X=0;sGZet9CtPiFrGk0tK${M+z{Y|qepi9=S^ zt2L$}EMcT@DIJhs5vE+k;?=X{#>IwQ?rpmh7Yc}xl9rb8cNw2kz4%^87>#T;Y)TrQ zuo1lLN^jQmG*1!90)UbC2Hi$q5u|Q?N-}FZ!MM42#AT_eWEubI)e_vR^(f9Xrin6T zky4DGQ#tfbbiZ=AE(IwH#5Agqm-y-VVI?2ir<+U*pPS_Q6Q4Lbv#{NA?!{uH5KC}DLJ z3tFzGwE=BOwuYy>~03H(h=1EF|hPrkq0d29nu=KP($i z`T9F1@D~Fw`m%X47Zxc33j?i>C;%Jr(qIZ@V0oxuO0e>su1+vu*ukC($RuP*`f@=} z@cvmEa-Stv%vcd^z;UZiQX|VE1{ff$CsvouXSqfkDZ}@*$h#>^oG@84IJOVq~9oSr7BPs9}) zGx4#V#hu;?68xA1+82jI&^K<=mRS3r9K`w(!Mf^``!*j1QgV78laONr&F;&&07vC? zyUX9ymQT%jQK*MNr*k&-c2NU&r_iL3i zh|B4lBoF_xqn(2e5fgtgZ4a(!BK+yo?w>eHRGz-2aa4~FnefID@YzJkJln{@V<&@{ z>N;I{S@c^Ja>wvU{734~BgW({h^+|9o(+DW7s{T-{42^gD)wTC5WFrLT>F2c*y@a1 z{pFim3`J1T(ium>!^i%J`6Th}d{#7i*me0$43J<>JDv?v$_t-%4Qb`6kh^{H`$w@t zgK^BYE3e!+;zg)qE_2c2eteN$K3Nix(xdiSSj9lL4!7$SSAmBT>GeRKJo0K3vLu{; zgx+VPx)l#dXxGboC7FnfLiY&APZp2B)vM07)Or6BG8jIRW6~G+WTpkHbBy6^&cq5} zYdU?=rKCMC?)Uys91r`4{{60Lgdt&2(|v>=wbpi}3TZDMsZak91a45+IP^B&mE1N$ zfFSH2gNm>YCiV{iBKTX?Gq1fSD2I3CPKV@TdGiej<@yZAOaB=sd%zP^0BK2QHlSw_k@D&pe;U^l{t$@GrJx?Z9NrQE_`8`Q%KvdmueAPme!b*Q$`V1*x*|-#WZ>5hslZg|Ig`k2aY%as>R3 zmSC2@#Dq3l=>087io2xO&PY@c?5(o4kZHYv>+*bq%62~B($6ahuu~x#)~>f&&1}X7 zY!V!;Vrpn@mnT=b}Euex7 zQdha2&_`(1+rKJKFT}aNU^32+2c>@` zZohB*{2VI5*-Nh*`spVQX7|@g38wwo9TOiW&Iw5#stnKr@ccYb0&;)pt2tNh(I#?+ zY!TNP%#2lRK&*@RuiD7zy zN=uVu{XQ%GUBIWO<^q1o7_sP=-ja@q5<$%%#VQ$f)@qv?y4X+$|7qlJX!{}3YNvd3 zx;2HVX2^a23y$;p5uLGYmMi}?*#M3#Hpq_dV##DaRTV2jIiispWrf4GczA79@5aj2@UC(^%`@8P{&3Tr^ThW>*%Pi})h$Ad zHn5TK!20}9sg3YpwiB^_EygMOSd}VGlafjMR3ANJbnsMcd$~AOa(me~|6s8Q~h!23QklmDyio$)IGLkVGX^2h)HrG4E&R}hU| zElDgOL8Tk43E75Slb`|vl$is@k^n~%5prcqj3|T(18cf}H6$eAsxn9B_lB!C&{XlK z=tk{TXV1nN@w`#r?bwd03PtYK{Ac^gi2AON=&isPYb(k}`mOo>&?9xZf^$_s+L(E1 z3M_~-`muI)XiR$1NZBv;&6lmIg&n$IalaFkQNUYCo?xltOQ)U0=kn#yZ4@(bn}RaV zCb<-MxiX$#;*(>7bSi$?fd~3pZ>GAZ70(jZ*XY4g;5!i#zhBK^i%5 zU`nWYQKnL~BW=hrh<}SqTjNZtT0bXQlf+wtQdeHLIBfdU1V8?;C~-%`UhgYqt5>q+ z*T5^bZ&UIOrT0{*Ii&;LPzc$3Mvx&Bjy3)w+KZR|A}<8r$^{&8HMKE<$z(`c5F61b zlldKStpj9eo_>_A+DQM1r198gVxfzWK&OgZg~_CX`%m#Yga2<)`)~PN(Mv3A)zOeR z(xSoZ{>2#&Z?{&!-}ydoK7w5(M{V06JTP~V`( z(N7}q-@03rXnI6bB%dvTZIrrl7hJ^t{8ny$?jX?HQ}?&U@`P)?mI71^)5Whr0ZJ$!7zs?zX5m|W4lTHFsl6;chA;|86}0|8DJ{1 zL}FfKi+Ax54i5Fz+D(@{`uz&ETZjXV{cj)W{C80VHc8Q(<~yyC9EK|)bLa7lS!J=m z>8;V~=5>|=%8Ot&T)fO*v3yoHSHnO5S)X{NJR`i6l8`UTqP7TID^|?qmEM8rPS-cv zxuOQq-h1CuM`@JhzD0OnU%8s#{9wmXk`rd`y*&3B9fEm%At8mS-_E9Ey1u|C{`9H# z4*C`cYi}@mi+!%+Py4Ns)xJc`KlYhVw7bH*P9xW6MC2O z#NpX9=AxMS{Viv5^5!A9)Z?f^N1%5=-B7c7%G>*4A-D6<`}~3i?tE!uqd29*Q>W@d zQRo>!>8Vdxi2;EBI9Gftj=fM*<>Y7ms^iy7(r$6zLx=6~@-?H^yE>M*&O>-VDhQ^T zqep!D+PQkO6@J(D56b-8XH>?+!;`_<##2JGS5~LqPA2NSK<<+PcC(xC!hbJe0d2~N zleqll#Jl?Z;-;syOOf8|u37Kmfv9}>{iElZ12#$^+DOQQts;X%+sehAZ~j5Q&GN5} zamfQGX|1Ah-4PYu;Ex$ZrdK|{s{v24K~Lp+y7T)&i(%$dKX!k-90>W;`#rAccfRc_g2DRlH-s2^ zdEUXm-MFr|NxKh;E+VPrcStF}mjk=5CyIEV*Qbrn5LV}MQdZyBr>c$j6^CHI=L=H5 z_l=vb!l1V%ns;sQ0X!>E!W2MQmFtIN)=7$!G#QSBs(Ju zew_W7l?yJUDIe0WYrNLcTWT_u3p^GhWEK+(9yzqs5)^{}Q)*82DK)=J{J4Uah`m8T zq2TE-kDbA%&aJ{?_vRy{hReYCF-~hS9ktsVoK|=#(||1lkAwotzna~z&g=7{;h41S}7=PG8*!umPHk(HBC zpvGw4QYg)Tn?Gt7C(YQ3oHMqEuR8K%27QkdkTVLM2+= zwbhwsleMQ~;D;TPtSB)6H`-dpF`ky6u)>{EVxH_BnL|I?KvVpFPj1m&&i%rlC%yIY z7XR_)yGEU3hf-yx^urQap+jN%?Aa*_7W5`rh&(gp6n?$)ih zrB`C!NtBX{w3%ivA2Ujkx)V7o89Ct}XEhxmd7iC-x{)#z8h?N;*^M((bktFI_3Bo; zvcuVt?JKXN`J+^CY=8?qtqd_*FEk=Bq96+@Jziq(606%Z!IM0$5`yVdYmR`6avNjs zcM}l2g}7N4Oi|9K5XlDZEJ6_oqb)b>+;#)t9jA(zQ)BmZ^maA*qpA%81|E)8F;yR8YT|Uw*E3?egwdX(RLz zY%`^Ik@ow8)6)flZ(s_KR(1C}3Hob}I|5_UieTvVNYgA^1VRtiloqxbFaTIXmWcOb zUJ)R+I7M|M0hLQBWhNTJiy{x~rMpVmCCkTwSxu!w zMIUg5nb`>BPWn`vSNn>@vpGYMNS7Gxlu<#(iJ|Wa9d>`J%{N?$J*AyBP3kCRAaJw5 zBce*eaU~JSOk3+3E7F$TJ{|hW&L$l+yrGUA{V4I~d}7jh+a#Oh>j$3YXHcXggsM(? zhPT9{7wbpRF}LYYyhIj^oIputEN%MC`t!N8MKL9tNh!59!l7(4I9OZ?2YsppB5Iwo4o z0u&S~3FC3QTj^zE5Qn#LiHRsY!Bw>yakaGLeX7kJ!;Fc<o!U zkZCt+*r~}7LZQeByYSW%K5#UJc(7slvl*zUUrefbEO{-Q*p{=L!-J%KbzAipVx}-8 zrP&s?IT(2qLFPzjG1W)E7j0mjmniHKO)*cO;+2;;i0+MS0OC&nN7`Z$jZkM(qXeen?01| z^^3!*bwgP9i!)_>GF4t@;EQ{D(~t;E{EyAxA8MEBEaMHyn8y$q0G7FiB6QsWn~sVg z*6S$ZiZEq8m{WbGbFN-tNLY&2Ns@;{5U2(akBoJS$*yP04nz&SQKSC4^_M+_f@1kS&7XgeVZ76X6; zhQJH6CQ9DWV)t z=xZAO9+#2Byl&t>wWk>S%LbaLItx|Cs?7UFzFGJKhkmrumk8=_;r6h3+;oH}-w(PA zVwxH-zkV`zPp(70ALJ7VO#b1{(TVEw5&U^uD3e4%22~<9JYAu}KKLqgmHPE(pxCG7 zL&`RN3JO)$TKEu*4Y=X_Tf`S(J6^c!?_y2IA`AW6gVDBge}b7Q(^!GpYX3N{taT*7 zrBtH)o&RcNrrDJ@!eiCE51dT6rz*espr{qv7o-1kVo(@tMId+TP^y)6Lphlcq{3b- zr1U`<_JL2oRT2!ljTMIHNO>Jh+>YukpJK;^@yYWXJ);4O=$FiLtPEoC4+x9X05e*3+k&ZdviGo-6MLhtHneuN+f`z1U8g0sVG()(DR#h$?nNK}~>REY*(xL9EQm zIQ2b?WTIz+=Bc3yt-MK17*DRfQ8&^0QOS587nrvyM^Ec9kv1dE)rZ+9pc(ppNlAVa;0V~hE0U_%+6|hMO?xZFzCA5a&4O`bT zCIS{S2o!?pgi96g>goM$<^VE!n}nf>D*QUGA zO~J_eyvxd*-qNeV=#hK_q3Z`+rAFxftiskmVAxUnsn=3C|A0kem8^oM?==ll&&iwO zf9&J^E!h?JjFQL3iWUU*cVSj&GR|4T9w(_Ry?EWBFRcgWc)Z*Cl#*D|t6}^|D#%lGaJo9)d zk+t0c*cg!}d5P5O%Q1y{Cr)CHbX)WXP?h@hp@sm2sDQ1b1l_O~M?oGqL0~mM96Ebp zaR2m*O7$bn8*r|pmOwF_*#*bN1bzDU6N%djW+ zY8b_xb2POYZ=Jb}A~jTz)17A5jR0#kph{0=;$3DQG7K)1Ofve7C-IY2n_kd4ydn=N zhdvnHSNqJ16F2PW%${16@R{VQa8nq?9XHNZDl#9*h4nAx?j9@05IWh`jd9``iY%wB z^C$@=0b^c_fxVP`ylxKKUB{5PFJgeGvT^N!=Fxtzw-!v&WbaO3^#!dYIg)me?kRibP!YfPR<*Eg-FkR>tjsua0WcRaoDcJ4A{NW%`516NwE3vD$9 z$2E@$t(NSo96yJk;#IqWs@uP9t5KzSasS{l1_AGuqYF(?!+yhS((O!yFZnuNrs5a2 z+cGFfvP~Kle`)VlAK1GLS-_kmI4KQzHZsIZz_{3j(Jc&mxV=Y9omk_NAmpSsSvD#! zw8C!E^4^I{e5U9wY^q0Xr9@$_FnnsDhy1rIV=$faU2BGt(Wy!6zG?1&Jmk&cxvKk$ zPwOq}!3BWk7peL)gGD|5W{DCUB+aEqGSF+Lf%tE|5h?z_Z^~3}ZEHI|Gq1m><4w?n|-=$R?cmF|`)8whaG1}6r@PMgXwmsrCuHlsz zOpS&LNW5D~$&xwX5+PG&0Q73=XtYi5$+>BC%LO;3?In;!|MDFMFMsFC5H+Y%XSrT}i7kSq?;?2Mw+NvXqnRPSA-H*QB#aI0=;r^FGJTt5<6Qlx{)q1Z()~ zc5W)2xMdRW^`107s2%?M4R8b-u@tB9@X|^35OJh3b(QncsTE}FA4u;ZqCBAci!vMR z0;UrUtWe~s>OhcREJX5LbWPn~jb1P*_H?l4_EiqiZ4D`~ZcI|KQpyn+-CIi0$X%Kx z>t-!c1oDjRTjz*8H#9GhQ(`L$Gb7BfHFzjRJMF}8is}Spq)bRYtsT{z>DOp#8 ztN2=KpWn(C&+nNxR0)>>_>G|-*5@@5fV_}f6I4z7u?4E8M$6*FZY(2MPQ@;W&Gz%i zh~yBNJ>?Op8jsZ-Mlw3@D`#>;Cw_4%Wt4gId#p@~7eHl#(GL5_%im3oqZ+IS^Ue~} zP(F%`s?oIL`TXe5X*bCD0oa}dEr`(z?Se%C^L6r4Ks|dXV z6Y#q*VVBr-$lQ(zu)}U_aXxGW?6ZM5Ujwjk+KqO9%I0ltNu<%tlv6{SNJt%h61&L8 zXXRM|J~YM*!hu7NKb>cml|WFx%BKOJ4=+Pgl~D-1{mREj46alf7s;WrUlE4p zk^sE^%5LnX|HMcJ3fDEme^%kZxfDdA7+UG>mRM3jFZU)kN|(Cx6k(Todh2>{r6M>% zK*cZ??MXoin=`9dFQAV~+^mVnaefsza@0p;UCMBp-V>yYoWYw#{?pGAZd$0d^`Yi5 zm;eL#o^lLo1*!^PjubRQZ7t#+A0!tfz#68FGYvObeCm8@3i!&9GDfDW{ zko(t>YeDQD;L;5Bq1G{KaIbG7V5n7T7(%kY-KVUpS_9py$tMs&(Mr6bEZ>5)#I#%& zdmK)9$=Gnci379^$fV2gbk?8O%e?^|y11gj=xL=X*oC0AMBAr!h=l7%aF6aIvJS%m zmNJ5ng=6>O{)j5$BkS(nrx&vk(6#D)Urm{sRWC5iq@n}TR#f-uNzeg76;eSl4fH^Z z@pK~Le+>((kw2`}3azRlEj1)We+|0?0Px2lf1=8g>PpWRlSXh+{1Ry;IvGZ^I+)XJ zy@hGqaZ$cXr(mv6>&WWa%3&}pDhTQm&=s-TLz)QSWCrs}r0B-Fpe&0ORfl=Xbyi0g zxgk3_ljqR_0IktQfq&o}`9=sU{_0noPN-2kJILa{!3vyTm1-%f)ztuGqnu`S_%edz z({nZxK|B#md3V(b;|thkC7%k?FCvTndNGBS=0;ufFOi)s#oe+eOMB@r(YpIyrTEe+N2)ub4B7}0MhH*A35WAI(ma5`04T9bTA?^-^Lp$Sq# z`7Ni~2X0g_=q}bZxNq*l$RbW(y}3vG>Rq|Y?tjk}DDg1*Z=H8LyQwXud(}%|@E(6whFRT&^!D(p`f?Z|(yq3PkI)&j)FIs4ff(2=QBtf3!T%;Q@ zyum1mO;4d-KCTAHEp}%Is#SbwhV2=)`JVA;;UyDPlL>$ zfS=cE-N7AzJ_fw|X$ELLemSZb1goG1jq#~O%V>bdG!+{4R%u|PrsX;dh9OZmA)S^y z-M-j;2%3^Dd0Gaejg>63?!lz82*opND>fP&YsM8uN?z)zp0I(ziQ)v#-@?A`*WRVe zzyS}<6H24``lkpck_0%n>=x?M@J-8HPjmblk7W)B+3{NRR4>NTdxrKp8RbFSZ%dvF zQj!@hB9K;xcbaT$!OBUqP0fsRCyQexFa82+`owR&^XEsWq204{;T_CIZiw>P-Hs_Z z47Z)=qiROTj;J;{DPBe)`wBpRsjfc;Rh(bNx@&CyuGV9^k2ive<^LFA-cX-&A0%y2 z$QedIt`{V_dH<7(-dQZlZ)-`VGyGfZ4jAB%S}|CxPMcOr=%;cF6c6LGSmzIu=|9^QUd1U%2(lBp&m4e zcX-%=B5+>A^l;qb<26W7)Mw&n%-Fr)P|>In68Dq3-q+?us0`obFHpg4@?FA{5OFG= zi(;+($U(~qxMx4cUC~lg<6LlDGh3(P!8KomxaiY3>fHG4aAGw4X_O3ELT|d7Zdo+4TUO!Of%5|W*o;y-hJ;U6j{j9 z&6Q4!69jBYzt0Avuf?P$np(qmoQdd=y>vM>ah-}e*#-TA@#xl;g%)LI-CqfVqQUE2 z+jj*C5p3A4>mJ1zr8c;uB!8}XV3eU)HwsdL@V zQ`t%yv83LQCNB?iQ=eX{QW(SEV-I%W027HiZId1>0Tf@{5>P#GYIEEY%B(|$Xi;Nj z14^g3J`=sFBW3%k*^$0K*yb=Rlo^E_7%EE?CML3_M(X^dYHVt&CP0emjmbw z>wwca1MbJcs8o{#-iG-L*1vCt*APfGu@Zc**kdU>$LFy1Gh94y(N16&OukBGaG-pfcM}bGh%zlpNuod0WXurqP%)7)f_XbwqF0r=PQ37C*_g3nrLupuKz>8iL zq^qeq6GyMzRsvlU#%I$B#hHBsE@8*fS|>3uH3qe%ZzfFlb>-|H(MCtRcPH@^`&gv3 zlN@7?}2XT?dXy#MWtQ1>Vj=LYWVod7I@CTv$CF86H zRTB1#9^9dtceG5%@Fa2yLZ4B(&Sc5@_i<*MEdU;l_R+m&Oh?txxmg?vkGCO9$3{tm zn~>i7EMWOM^#F$fn5Y2X2|sJs!@FwiV(3-cpWwq_DoItPXf&zYO0`|#S}+k@p)$1| zbE|0v4r}xkKP_6B%Aj~D<0t!KCQ)939~(jY5{%8!(fTh{Kpi^N5sj$UG%luL*1OaY zN3%IoOXViP#GFv{JW?jR_Hb1{t(uCC8_9RrTx5F)vh1rDCWcqV)t{j$o_;;?_DUWC zkHzJbe{B~r{6rqgC(o{)w#O-M0TS&c5LP5jc~O|5Tvp;^`UV0 zqJK$8tTTF@hU8*v25tVgN$WLFe(RE3Ai*29%mMVc^NVO~0i|N2>@(r3xY7klN@l6{}^_>4_ zuBU1AhlcL45adL`H2xa?(_dioY#>CHcLitsV&4NWm$QhfL$)8=DcMbVp>aBgC3V=z3_x>O6U?53bFJ;CiwMBjI#OsH8gC?L1e zzP}aGS5#r4JBm%S5-lVH%b9Xt_X0bpaqu265Voe)Z`3CcblDen$J_;~`aM>^umq^( zR|A1OT!OT^adZ2JLjn+zp56v|%Mh5}BQAS-ssU{p+bxb&nbUuPKDsB-2oV02Bw@4W z$okz%t1>_;{msxE+(C#{d`!K1TMJ*XgxP7#;`Hlh%pko#h;1OPlSs|Y8|3yHt8vyt zqK#PNWq)ou5_?*cr0-QXymJEeW?9AR&`BBJmP0+tlM=}Ud@5W(C+@7?kn1d5KjDIe z483nFh?e1fuy3}aBVKN+vB3*qUhy^5-$eIqcuMO&o$pz=FVd&O4Whly|-h%7IL~40I@%t6Npf@f2EQD zev84a(Au9$70@$}l$5Rf{0P+)yqETVOi$(;Pbo$D@#SD^B_-3o8u=MdY>* z51;fHDI>ChG_g9x2gSSLeo(CNXPl<3g2a?PwX_|+rZ~O>zXtbwD15m!vI;6Tx`Yw_ zG^FekJGgyU3MxbSPl?Y7yuE;AIIO`+rw7}X%P;)VwhS&GfKsfaxzKwUt5D3Y>fpNc z=}~6fRMK^5cN+K5@cHBsNgkI^5_17+7pT?QZGI=y5^#%hq|n{LgH@9(SGZ3N`7BM( zvSS3Sx4es4GAhvBZKMKFkRN+hjV<`50L9L92eNA428xj;3=!>+k7^2rK2Ml8*rZk~Mw7^LC=H+tcd(75;?B`3XA)0Cwzd<2PxK%5sWccQ;Bq&I_OVg?zFIg>51AM(Sj(I7r&082-yG6E9I z%&7aO5+ET{$-*By=Nd@3=W2zEOI3K4-tfBpgA!(wV>|%R%KYdKVy@w|F-V6|b9|W{ z>z>W%FnE0&C8wRn`8o%InVgnxZFkQ-njWZPr;`)s(6EO+-Og`tBG#SR)3TP)P$+|J ztkTAD;!WY|J_9H<+3_x&$hL##h?e}4F%fOHssys@|IXauogPydNUjO9yjj9dvA(Yc z;B{v6y5#*Gc(-Y~T4pc#=}>xxZ_G^+Wqu)f3WI6HJO5VFN4)77e7}oKDBB?l_jA+M zrp_iNc6Kj@gkV20enV>VU4{-@{jZsyb{sEf>X?a9V`I~TODB5n+u51RnLt0`Gw+7f z_d1x&&zX!H92cO}sKoYna-|k!5^c)@Wa{e=*c?EqHot)OdTnvzDecWluDufA9+N*A z{(G@p4ZL5lVhE&GsOrN`3KaLx1Lcc2jmoRB%|=qvkJsY3rXPMUQ5-y*rzERQd)fb( z9u(Y9+f44UNhL3`FRcZn%+pqGCSyq`j~BmzoEx8)$4hd4mnvQetV|fJY)*Mi-jiLb zq!p>Cncw;H-hSF&c=G6(S^qMWEM~Ho;>~J!aDuvEbLFh8`=VyX zFv8LVrSH{H!#w_*0Us_8$>~e!F`)xuf6zOuS@%1s17bUnwF!cQ;9SkV4cYt++y7lJ z)GAkY)}32QU))E?wXvnJD&K3~zw_3+w1b*W&+04=aoH=7cZG7#Nv4Z{Z{-#il34yP zFA+R~^CUsL^bB1gkx$R}(>D*nm;Rn;w?oC_B~D7Ar+)bj;?vV{Vb7?TP3qD9T?oUcLB zoDW3bYw%51RKF%IMrV2v6sAEev#XSSkl0t7wHtSZXXDVRg?jRHnc}C65Fb!S_LTsx z+`aR;C!!s2M0webz7*>1o680os7T&^dJ2IOb7p?Cm*UQ!+}MXRs*SP4K}kpyK64}^ z>qmN7dlo#YwCvpOE}vzGidRr9*joMQEl41BdxT2=;dmAcsBteBWs=l`PSV<6Szk#v zB>Cl>87ZUhC9$4^?gZZ9r2r=W`>Or>zAT}UgDl;-enqX3!-5553bCZ=D8WE-vX|C) z3NroeS2o?)379Pf0m3q$>;0VzB3VGV-n@X>{nkAq;`1Y(?;NR6st>;zkc*DZeix)J zSxTfBcIG&KzPX;ng{I%B)sze@JH2{K3cRAle|KQbeO^nL!Yz5L87q3XQovZg{`B*9 z8?{^4y=fH%RJ!uKr4M;$BO=mWQHsizYq>%TIGeXMcSvubEeW@SXP?I<5=uz}0;$v5s)Yfrsw zpUF4Py|NJE&mz1)fzRL*m-z0mqGkWMtFI9iT(`&4$s|eV!K@oX5ABCL9(OAwh#NM< z%q}ad#yFnsvo=+z4Rh@%EkS@vg0~J=i{4ZpGC#$fS+z2lh>s- zMCR+4>L&@m>LBU67R?W#&5!pT!p?85-@?VP?yfZ6FR&3iwH3$pyCCaf>~j!tjPSc8 zUmslB^}F7x>^`1RNk3394QZg?MxD`uPW377NxGEf1>aeLURgDPuRCd7y-k}RZ#TJu z%R;Yb%o|Mn~ z=wfZiBi@YZ)6&Jsr{70oW5vleVHfkxtG!DJY;$V~3hNucPq%cdV1GRgw6+qGMekGm z4YFhZ;KTwgJ~9N1r4p_~?Uw04|dr5156{40-(_(8D%g zM7`CFBu@KCFc*Mds;0bn5tToSncLh0k<1+sy^-eN8<-CWF8oUap@b~YgN_L6&4kn$ z{}JmR&YM%7Ve~EtewxGkFf$)X$q*jCJD^1vo5H5T6;({e0$+Tqh@x9l7hv^EQ<>fQ zz!GRtVW(jcR89IYs>FGRy>4Kc;YO0|tu6j?GVmxQfyc7O+Bg#L_5;WU%^Md3Ms+7X<6ZoTXiSz&TxZyb$5mT1!Fy zt*n1V(PzQ`0k=T&Ya5Wy@ugEsS`p$CfUp*+-qLfUxnL&6uPb_39I-waBvj6BgsrUVBuI zyq;SW2^PPY^)YBaoM8){ST-o)Z+L)fEIK;S!e;=ekKU@Oe^tF787X-npi4DhGg1U) zMBWhLLml9C8|}_|YH<_9R`Fu_6bumryb!o$5uan4->&5}zr+T?mW2*vl-Nq3pDIVC z`G!Rd?s_UWDpKY2u%0nn@Q{P+J-V3-EJ_52iXo$3b7 zdL?T<$g&!lC17^qF18YS8gI=>kTLx612-AZiqOp4)e9yc%r|ZpUpTg7R&o-o?D0?& zJqHHJmN~<6b*gm|ehh#9?MtOcWV(4;j6_ZHJ?Du~G`NT5&n6C?l)D`f64+7XGJLH6 zZI?7^=I1oA^>!yKYvGxi5UjVEsaHSoZSFSUt;4CO#DjjwSJe0>IwnHFLysiWjLnm_ zb3b7Yc9Y%jxIGz^is@B4a^t9xwR-W+_Z3Eegqz7luXIf48%UP-l}_7lGgAscy89?j zwVM_IrMWH?{rnS5AE@TJ#Su4$P*OPs&rGN!s%bYE?ITvwI+Z}X|K=6Cyd>)W(%#g5 z_q+Vjo>74Oh=?TlL8HXtq1n{$bn@4UqoFECsi_MJs%K;E-Mc*li=fm}S!#Vxv4A9G zG=dU-HEPBZMWBoPczX`x?N#hxNUNJWLH=kKYfx*njKB#?-g$vaS{nYWsc}RQ@?XX? zEgdQe7hm*t{>So*EF5THpz<0o5>Z>{+v0)LVc@?yJ0Z^0R;s9uFDv)P#?o}bIX*A) zr8IdYBIBKJu12oazlBHynZHBju9cUys9!FP^8%NhU%J|UHDM=qn{k~qmZCqN{?;9y zdC#D?>D;eCLd%*br!1E3?0F~~?`kmpX;6`Y zt3k`!|NR3$cXpwS$z4_4^8KR~ic{5lBoPIV**NWwG_!ja}P1%g@cr z7k7yQnqR3tcxcp(WkIn5-?OC12qyi}6|9L*Z^J5@iVF zQ4PoM=^p2KUsr>|WxBdnrpb#0>I+liQJJ~gKN*e|b+z>f315mTMyB_<0$BT-7aq>k zpxaf5BJqE=*U)unoTc{km8o3J&B^^N`m1S&q79wWu~u8hho*L)hI^N1YRqb#oF5zQ z<|s1zCq;RNW6hymsMvH9*muEE3w^x)i=o$fWawBDxzC*Ag6KW?2t#-9VNgU)@*uBR zIrwPVI4bKH^_?1X@a23aoJBtw#Z;w=V*4danxp|?GT_Oai=G%C?HFjKkkj&Qc0mjC zsOGdv7^~#JHyfK|T6!vbv^wSI=vwD z_XY#0W*Nkcj=Y-B98}KlrUjiRgKfl&JH~C@Dd#Ya(WpSrd@-Si_%cn>*K7h6kWKT6 zxav;*JVX-Wp)WhjUCoD{6Jqgrl0q$Z~vY1}*9v<3=@) zf@wY=;td~shp)IkTiQ7CEr3pi)a|hA0IgqxB*lG03q9JRzfJWc$&)I1;D^VCcW9u=oK?4cB7d%m*dkmQ_vh(iIQ&Q_ zM>3Q3S#Fg^4^_35^id|#Cas?>-X4u04lXX)y$Q?@51M9y8DQ=I7bSxyd93v?0}HN! z!!kY;r)sQJlFQTeSU(<#8v`k`h+MF>(77!R{nXX!s)|9HMa}@8npArTNRR$>E8rbe zEsTxH|0!(czI_!k1?7ohf-l>EgzD*ZbdYpDzOUjo&&pcCm6laq^tr)qWTa)oG?#&h zIpgUBYE)ZvYMR6578zJ|Zd=?*#~+cWM=V-VpTp^YROw;6tol}{3X+Cb;# z>`J)q5+F6FN7=JrJ`yg`aqN`TaE^PqrlkBHb4)jZfx$U9jDfMQ)dY_A6zQE67F%8x zpOVki6zNUIAwQYD3E+GPFOZ#tUG`3i=dw*7fgSw!#xhAOLb>L(;=Wqk>d=SP3%gss z#u9P~DywkpqeOtVuXgC8e3(@Lv>Ngil&@A-(V6}1u0Y9f6Z{6Xcj4+!V^L;ZKjv;U zp}um+XG=AGNa6li|2Kw6;9i0Y_=B*G|ABGIJAcExxJ%?sWvMM4P#@eC*!87r zb%)gNDYa%t`~~;XIJo&zsJ&>-;^>)-)b>^BC~n7KXW@yNI#xM0&h&@(?;D`J#0Q+8 zob|u1Z!4eLWpnJvuBVv*wuWe^`(&0Cc^$>-VAbKM_dWPj`_dOH zB*b+@Yh;rB)r&C%S8t0mF}$u`suuc3oeqJmf@z6m3kWCk8%WR;bN|cGb;xJZMn-SZ0msO9fBHEGT7j4Zz(iMU-HxqV z+%j667Ek0Yr_!D5LSE97m!mie#f!z0=#xWYOMTx2guD;Dy-t4L4-SH0+Y%N1xdi0b zi7ULS3WP%|y$~d ze&qJp8&oU#%%k>Ln#6`rf1eaI$H-R0^lM1pIeMFnspy?<5?b-Haj6)kbQ3PCu43Yk zs`MoK~J}x4mA4)v+%n9r9u!l0FyxiC@FkaK}fUvZFzQ|TeJa`n#_~}dO zkORC9Qw1T0c>vz+OS$lsNh)*v?p(~EVR}@-@E z0$VrT1?&=8N}1^jejer*<|o2E=C`GeWGV@WVCQY+@?Zzfr5OCDmHt0kH^a|8%~4%K z*!!C6AhdE{w9(I25PKA4Fgc_MqpJ7W8L{HQrqs{n#qPaxpe73O#NIyQwMo0or<{+r{9pt=%3`u!)Yt@81vY3`iF?1EutkqH(i${rM?>V=M zWpVDzgu0rBuVKb$KDrSh*^S44FW)YmVh=^r*;eA;A(+G=BuQ03EdhJ275(g{@_PMM z{n)`kUP+i*NpVSZwQpVk00PCGu&OTbb5OoYI!>B%$!0R1gLK%jU`}&^q>G?~bS6R- z!hcV-A#_&N?E!e$(OndKG?2C0+E)%Z>Wn?_9hUEjW)rlWC1Ef&w#i8@K#7a;1bIUX z4v1ub!=>ZFN!Rpj{_ej2RsI;8P0eO$|EnDov4G+5U?TVt#@6x#bSS<3m8;?7*okM(+=M-3rfTuK>A7gaM# zR-IRi%h+&xx`Z*RSoWc>X!lz)Y(1$GqF8G0XRFMH25Dh<(n}SQJI>2}s4Z*FHRsyZ zqpG{BcI|@js(OkhtCZ~Q8TA)pqi1;@)=!vZ+gN{U|1l@!>pNyZx?ev*CREcA;{mBd z?)j}fZyjKT)8>-@oA43UR*JfXPmL9|T+F%g_2uL*b9KC9SViQ5V1QDwfk!aO(ju2< z*7p$(oisCZf)bFFZ_qFSavHdfXXI|+i4d`3dyLN_4(c>c$}lIEX5x;vn13gWc$9Kk zBxvD8GbZL)yfte;Tu1K)${8Zzaze#J93c%8w1|`K0v~59Ff3$e2wJFadm+5x3;Cu1 zGg)JY@%+KtTgddrlv;@INqNRiNrcBh4zY@^Zse|7Wbu)n;mq`dPn#&pU=d)P@`_d> z@F*YTBHL1%wAS*C>W=oBA$RQy4|@DXiSeL+!bK(BcnP;vl(WpSyGwdhZQ_YBI6=*8 zit+QOW<(@QGest7;ILEwKAuL}jfz741%%9XlPjCjP^nfPi(TcM2@xW$x)MdMMZ3fp;_mU<4?H*mL65?Y zQAjq{hamE}P~z^!QrSo&Uxk&tm*?{aM5o2$oZ&~!hQUa}-3B>_%COV;N+pEWhPKj8 zl^pYVgPFeoo2vaL^&i$tWdfu)9zk}fFWiGdTKvDL?-H~xgm2*$qTR|foFS_Pd-34h z1vZO|q#-?aEmLHPyKK1sCeKm`^wB(y9xMDP%IG$*quiP3ZLK~y1|>OLP~SC&q21ar zT+{+nLD6sy7bYpP^3PhgORE;HMl=lbgp{2frss-9Savsxt5nbsTln^^{2Q$S*orrzRe-KgLT+MX!ToGp(h4UnP#1<5 z!ytR@m+1~u#XEE>jSc->|NMorDq&0Sy=hj;aT^;iQq+gu+gR%6kL8>7BsOT=$?sd@ z0R4io{7lVG+`onsB(L3dtY$IOj28{#6GU5(q9n>gn}q$JP7t8Hz=ZZ`mTcp)AKN6@ z=n3l>2WEHsF-Z~5SdZoF{gDc zkfp06Pe6s}Mv6MfJu|mOaculHm9lkd|5XintIVb3NP-!9Jpc8MXZ8=fRXtTwVdX=+ z9bx5pZAEl!se<2KDZJ5~fRR;j(TN3ipCvA#1H^SSiE zVC*t)>;MBSqsN?!GRU_BVea~`$}N+AOmD1zgG|!TZ#P8vuX0ohjNDAmTJ|j*R|!4i z98omBk*~!{3nhA=78;z4J!4KtG%R{XNF-lcF|vEh#*;WT@23K>=^F^HCMdwI35&O3 z0$D|16lJ12ne}{E8w+MH9Wgk*Wz*Hew%@;H>rL#~n^yG#m2lhc+J4$OG}N??2+Adz z96Mm?sNJO_CX75b7KDZc)toj+L?nH*vR{ln?1kT7x`qG(gLjOqm(ozIN6ku%BE#tD?)^o7)J-HQH|5YzD zD`Gf?g%-XKIXWk~t2Ir}QPKVjX~E_Q2Mo)66l&> z9Y+RT{fRseapqdC2#N}^R>M_Lxt$_Gj(V9odEcYy>ZF`&rdnb}&d~ZczSyC!YfE;l zhT|B#jzpssX{)^x-nxZ^aE!}vsYzKV!ANa|pAauhqqUpzGfI^Q-db%HzuL)}+1#sd!b}e&|i2FOrYK@Ukt@kYVe)NSAo)l<}#8E}AQ?A5py>$+(uCNVg+kl-1 zn-NrojKR*!Szat-QnT1plb};jDUIzFx3Sw5dniFMMyrXsW4Tk`F|nlqr)yD@8;7!> zIV^)Ug1;D%!M5j~a)o}_jJiI4GNa38_7$4~Ep{yk{TJ6YIAjh!d7HHVuW?8FtokmU z=GTMcQaGsQeZ^*LJ@by-z#SHj-VZX7fO7d#1W<}Pa!pN%HbRm@FJqGpgFz=uxf+Qj z%1YCoU%lLaD90E_4-Kbm?X1`gAxXTKp^2D5vlFH=g2WR17b5XL2OomSQ{9bYvuu*$UxS3-x*Y@-g6 zrd>Vtnis*1{WIDwjr|(cbInVer9Tgsj~p}%Tp?FYsnH3L1zB0J4wOiv{-bu`jCCNk zZltdP&yqx;vCcx`GL7BP&_-LecI=U}PnWR@PEw7wqo^`YQiHZ*x#JO2;ui8yQ(dp> zm`jn8cDZI>0aK#K^W8FtX%LTb>F$9eOL|<6g0X9K^EXp(`5QIXQW!|&Y|f6nDps{k zd>eQQ)tB|{l0$#${+(m$-&-T-zAC80vh9B)>+SU(CF`@0n>w$yKuPbJusta=)CN=g zZ>4}sqA#`)k(5g5c)0&H7~E+Bu*raX%UiZ^_S4b!7dKW>xx92X=QE^|b~|yLxFVsy zbJ#WwT+S+4nKd5mwLG$aa8mv?{$PF!Kl}=*mH?K9^CB4$+x6z9GgtPHZoKR~v_KNE zUz&u!OVWrIvOsWQtsJ5nBV*huIBM`|6pV;r5pkmm_YzXib|Ib zMp%6Lm#**SFQIHUBDpKHzl|Tvp8u(Vw=vdYcwSgB`ANS*S`zr`#~(IQjHL8}1)c8D zi^8JEudCrV)r-GA&i9{5m6K0dY6bu0VAgX@E{NSw5cpTo(}dPf{3L?m<@P#_pcmX{ zQrnugS`0RH-US0m_~CAa1}(bI1;J(0U{*izZUxsZm;QdBMDS~F9c&PwGyvej6!Y6j zA|Dxc>+O8%|*mR~LMTOAm#e|kBtIDHmdjQftGNu-F) zVE+uWy|a4vbLC0zBjGL z3xn?diWJrjV3(pUdc}k_Hv4(l2q}w56tLXG=kD|Y24GI1pybW+%?x2UKiT)O2*Zj_ z{7t!!nqIaKvR$XDYb04j9r+bnVu*en^?0=WXD|v~cn581~uAA5^IF$be&2XyIt*4jR%d@&Uz;GDJ=y(mURwPyMCfsT&!kO>pio+3c zRrBL9pu9KB0W>v^*Oqd3(Aq1;mpk2>%+cC8DV_c-gqLR93IJrooo+WC05cO{Lbk4GA>E}ubAc=D3vR6?ZUU3p6qBXiaG|c;W3oam zyd`^@+1{Wr&r#`TTb?6AeOn$#d(#3}K})hiYgcmG2!CQ|rhKiX<~|UIi^rShN+8Ia zR%2#o$8*G>1w?ol+VUv$f`$tRAB}l3tiWbGBX)Yj+gozNzq^JD>F;}O7QGy5E$954 zYAuJt(iC(Jrfgd3p-~HU?mK2v=`rt@kaWo}LQd0mHnH0I!-B;P4~>AOwFQ1ptTcg%C@Z zS$s7LE}lc5G0>+5I5j;<@f_M?07x&uc|X1BfD!D8i)W)2p@hdKv-%wWif+ zgz%>C!jw^AY=}AqWaj@%=h#}nU{NVyxCpCC7`b0A(3<0L-p5kT{aGaU3+DV;>V!WX zO+ql;&&jP2I#LEam1J%-+v+45=K6ofUF^DE*rRm9I27(Rt(#We#xKC#@#`}Xvq3Kk|z%Te{t;nXYT^( zrw~+e#gpt)6Fb0?okWe{-%6z$#gwc5~0A2*Z`jP*On1Woq3lzo$LI=5} zvEw|^RgyVbz8edSGpr8&=b*U;q%LcL!ljK^C0X`+LQj3;S7rRpeiB&SJA{4`9-MxX zSSzh>y2ob>{iwQT%w6CPR>`Kz9%Pw$ZFA;^dLP5STLQ@@95V^gHZKhfmh$2d{~b(r z2J;Y_1VmI?jp-;;(Q$iKcgd!3Svuz6pm14#k=LjMGuttTaHx=WKCU|I&Ybmp)ejM3 z;!@NY7y&_};V6VP)jW{3*IcHnbaE~PT+USNIXT_90YTvv3a!0xs2OhwxJ=de#Y*3z zE`%#$R8|-e6E1FSjE^V62M3`Uq7W{w4&>xij04uR0CDlYB=~5+b^zEY-@ZmAkr5gn z=aC~O6gX$M;46v!CIC3CM{;tQ)zDBOmJSvcPGdO~L~<{60c<-ypw#ttb9_96_`61 zhBVBBVAo_ay+re9|briMz7~PR+@7#4B|-#q0gB6 zvwwa+vWu-dCZ~RLGIw5c{F&+CR=IZxG&a-{P+h(QULgm~PHkEL?dt>m{MZk%Re%*0 zG<#mda>I0B=`*o=Rl;w7MNe`N*FbuEv%x8bT$o^J%p%0C|!Zp_<$~?iOKBRNZpw=u+&? z$?40d)n#Jj$;R7_Ppe$tfQI?j-(gXDPkmwjui?~(u$M+-4kSGjZ~?)Lyil0M%eSiq z(z9#@@@;v8Jej3lx`Zu$S4g54{X{$@ez$#W7~Q`2mtZ*@P_2LgnK8DLKQ=FZGywX% z&L!lMS8l<ZoSgp^= zWu=Y72j~qO7B&WE3~VCxfHPl4>{q=brLBRPh7BM>Go4aP_lV!o<;HkmM)Pf?9z+8+ zFp~ld=ze$`h@k*1?R<`4)r4^iaxL9vp!td)Hq?ORxMhVLMybU!Tc~vQp%OS3G%s3p z$CPyoivxS-_7F+D5;zq$FX{$RYI$1QZ&{UbN>^s43CA!3r?&UNsNC7NtQgUxE78sY zlmIZiHeWtR4e5p`#aPhT$V7-j??i}9eRiA-ECdx3 zJHR#U1h`@|1*K`@JaY_?1=!8?#s`Uq_8eh<0MU|f$O0)Q{16+0gv0B-o zet{HDXsUjs;$b!4f1{0`^H&^p0kYpRz)cnbMjx4nrIrRnTb=tW${PWs+hKr(;=fD= ziRISOf)s7qf#{@Qz`h3{p*{e$_3#AXq&CJEV29ObW1U~@?Dm*t=3!0!Vc9iR{X6zS z@k{#U7$rFyVz+s9Vz*><GKqvOFPo^>M4c)Qb7d#~3KL&mzt)T%2^?w~4k4}&EokZ>lt#NETHkJ-^5V8Ywam$&8;ve_(TL!X*AKJDDl-V6ym5j@2@&C$i-AKZ=_ zevWy91-pE>8OcJ!&O1_sLpC`Xz@*$(RA2yE4X)7hB&zLp7pEVv*huZsvAaPtV_Ma& z!k06QIZIz0ITy`qZ>R{d*_xLieqs6iM@)aAWA4>EJPPuO;O-lerzb7;-{`BY**o^J zUZ&U{a-u$2%h~#h;O*4?%v(>?L|NAz4F^p8kfBgZJq^d_ugo!E&n_UITZ}i zXBnbR%vlrII_U8gGN2~aE)FhxVZllBYZYhUKw~onIsEM#LbQu$Tb=AiuDUYHW2o{kb0EeY$_hG@M)AnX4s8$;bCSpEE>pf3}AW6<8X6FugJK zaq^yF8QhqmoccZ{O79}BG4u(7NFClk zL(Yqc53AEQyg~;T6vFo-#J0uLP+>_qv@@P$&jcO~`WHc&>PZA5jGNj{SQ667Cn(XpX`v^e)E_ z+v(>0ceb@+pQ%S?|CuTYpADrgWXr4@tXudl+NXRAUl~Y|q+R)C!$p7_E;k_2)(kzM zdqGzz5BlR##F%B+k&!4>?zQyib*T_r4^gEKs!l~vrKpdf5s~JyQ2*^=5oi6QdWW-U z$yNEujOI$b&ogcL+GXv12&`1JHA@@!*Kgq65H5cd*1bsl_9$HnJPO{dJ<%Q)SsfNp z^BYr(&HHfp?#R4jv^?9q#wwKJNK_DBDtSKZ?TWEQ%qF z^XdOy9+dz8-}0dB{A?T?A|m8Y&W>Os2-27IBX7eKre+B2{t>%l=Y?(3zAs!Df>7kq z&LQCdB7VD5c}+%5L>3wgUx8_*pz#^znAd z?wRgKmFP1}w7T@E;x_b>pvF;JWCAkJ~dJ zulo6H#Ko&)`W%w>3`}eKKDS)&IUG(tklNcPfqCjwkIamLnx(>Y-ZgTR8aj~FrKAE!RGA3He3_%fWQWN`sup+X!^ovh)?D3_7g_hbSo&~~TW99xa zsqSu00w-uDNvCz;Xqz{+H|=JEcthL$?b_kR37<2RN>%@yAl_g1Z_YdWB3rD)b~o|Q zKL~*WIYK^yis}4gfzwYxt}H=Fvd=aj zYS^wDNUXI&h3O_^d4pd#%V>+osL13(yyVcF`9hf_!lP`RNv{4R zmkt5jFgEI@SfY}Q`alShg{*N6uU^|{myAs!SH5d6NV70s zYPWdw<&LWCJ(%#d305t5Of5LuRIC=Uc_N}^&+PFvbBaEv!c@2N7Cw299J_5c=iDfy z`AfqxxWxBu&Myp8cs@L+_gYI%k@5mGNiP{^WTSbb-}pIG7E(hNR* zIjX;j=yH2oB0bN$CKvKnK%!T4?{JXL7@;+U*tY+_9#Q1SNhh57c|>#j^f!B&G?$~9Z$HS( zDrwm%wmEUT-Gw7+2r|Cn+RiB$oAX6GnG_wdXXfBm(iefGc6cCiarl_ClV@VF1=^E2 zjMT*Q8%TRKh_dKf3qjZ37@HotO|ewl>)?o?(bgmU>z zCOpDYqg&WUrj6V>enDI1${%H*`opx0z$-Sme(n|yZR*c$kQsh7Q?R1*Xq~pgx<6aZ zX{D2O{LP2q2{;0g5pj?5iW9RUt!}#4##&8%DkwdU7K#}T3u`M(uY1>A#qjJGNXyQt zD*{9d(FP`gzu5TUjIcL&0mh&?_wQbQAm_+~(fxFg%P3cJ#9*YY0O@_sMC^x8W)uA8 z64ctXFTa$R={qDAf^0q?{Cviiice&|B@d5jleBp>|4KN~PTlTL?&(NdHWpy}^7&3- zw#sJPu)}MrdD{0%?D>J@`DciKYHxF<)NQ4xJ=q!S)8;q z=*PScQ0#nOD=Fh8y2WMDl2E7|rvl8v^;xG^?g>-$gXzg#9qRq9rTle~LO(K{!}htT z<^BG1k@Ew#QwDsQ;nHw%v`c=-E!AGkWVRfh%BbDc)TG+NXWdypoL4?%m+Cb8D!jKQ z!(ne#p2C}h9T{QOHo+~0sF$uKhf(Ksn7SPIemC33Iw!{HJj?{8vn9bb6V?mJYM>4{ zV8%d@<*aaR&Ja^Qgtc!5^HeNj-DpN6k2EUh#z|M=kv$BfeV1QsewGDo%hC{j+WOE& zdx}F5;{;qFa%FQR;Y_`N-(ZP_N;DYpx=w(OsQevzwFKyW3EnFQ!bmXD`x2%11_WGz z2l7m<4$A%1Hr5Md{8u=#-a23hoBeeu&iIR-`(^U?zr*}S!lc|j{tnB`thtO-XvY7Z z?6!e)WHRtwe0IEi$+7271trO7Ir*DrvgnNGScoRL>ed)NonoY4~#MYAUeMUYzQj}wPs-&dNL4ys&-Vg*;(wTYV zB)oNnvz(U|s><_F4{kl5=HBa?fE8=0{hor=P)Nmq8~ef)2_@xgiAa^A=$ee`tcOWl zLz8hE5%+4Op0j-_^i6m^@r3xKp|`Sm<$nF+Q;Xb9LNG+*DPO8a6n~j(8h1&@E9DX= z-N!*k&dsZ#}2U zXWJK7hFdbMrfkQ`$rp?TA-8$H4tU%8%j=(B?Z$W_f-AC(qBB&?YX~;Um}SSVXD&kd z*wJf9<7jFnoJ%@(RGv)r#uRB(K2Nk)`lL$pci1mh#g%?;m~n|M5z%RGTnfThtMQH0 ztkUsePEAY7+^up~sBxgm>qzHr4Q))UD2={@V6~M}ghk)!ACh01Zbw(@YeKH1DMJ!T zs*)ElPVi!?=*je@bZjvOF?4rBo-lXhcs2+3t44Br(9RmFT{L^&r(c_dn_CcEGv1I| zG)r=s=_KW3P_#CHUxqmltFjuj+YMFbNTQ<=Lo>g#rViViMM{`pgM_!;Qwe<{%o@O1 zZIRx&3z`wl)~{vwqH@{2qarKy$E+`~S++_o|iOKBkRi^2Ey8O8!6$)jUcmbwvBxTkNT-u3zb~0 zMrsnH_|F<`Emo6~16AO|>~aGfbLCf@(l%-3WjB2068wZFiZ`_o{(WABHZLb)ISPZWu__cLy^f)G;&eP>R2!OblfHp?r^$~joex)^R!8E3RLl_9ZdvDNGZIW<+oa_nV_o{-cBc&sRMha?+}{j zmfIO?Iy+FEo?i+amN32k755^34z$Q6VVe!aaFBUT)RnXNOEMXBUH{K6SEv&e4<}IaKeVwY~+ALS0 zbfu%{5pD@3dYi2$4`Fc&;gPugq0q?e`MY^z^{(70X4>l`#%27;Czx7bagk(!bqWG* zgu=N!6oPg*k&0Xr>`3Ubce%^qTjbc&!bMl+*{hR$1}XlIRl^g+2#2D4MfdsVb!oQr zNS!cmS+TT6doL<7Urf*vD)N`Al)852BDSfcAW7lKMxr1{bGf>L;{sO@@dUR%y#}0y zF_m4O86Ow33#M#s?_wP|Vv7oG#hKVCwHqV!ok2)6+|j|Z< zx~|>hb^hWbN%r0I+k)U%%C3kwBs(g$;klav!i!i@`cAs0gu(XV!5&y1A&p`S7DsFb znjh;l_EL76pAl7ylkkR~$cC@-wc^Jf$yBZ$#PWXt3s+y}ZUhl4+budeJYCS4Kg=f% zeP0r!^IcbZ7|u`n*hSjgZ^+U8esE&SkK6J|VL&6Mt7JLs?e1a^<1#ll+u|bk=!uGL zP~>`xGu}w_z+Af(p%j<6abYMgyN>YsaW9E*`~*#^nahfHF`WrkBuSb**7GL(60*Bx zhnma_*z~+k?!H<;jm=b;Ilpd$*g_5zMV`8(&R>4Nns2!~S?vo3xRJ@FNlqC~g$J$B z_SYWs=O`&$?%K^!`f>;XXgSjA0?x|QIAP4~2d<%0};~<+-#{4AHBvzR^cr71;qd+3%R(5D$~hD&3<#mbIUt> zA8^)f3yb*>`mRH*v+FZ=@9@)PiNw{9&s3 zY{B7a3EPzu0$biEaZ0w`0BON&I%qS zR>BuH^!S_cxTnAH^4Kd53Jb_5%Rw?R^745!#F?6?hEWSNX^KPvV|k0Fq}+>CS4{jZ zLaA-t3#K`rVMMR^$1ySECHr#qBxu8iYwDp3E7$K5oL`@bOAXq(zCKGM)!%~yO^2-v zFHbbvy4S7@dm;7cpdk>p*u3;f6`)Hh%$ikk)`^V9I+z5GI-ivTZWV?Jr{5x3 zkf+*tXIAAbUXWay9C$FT+>A(S=?-NgCcA%A>H`p6_mRouaHwms^hM+GUfU`;bKu-SC3W z91^V};>)Q*iZIMhyF#*d-l@_?^DCzOql?-T1wl~*;+lS@jp(39?=yP~A(0_*%nH<( za9AF@E)ifyYX>YZB3JwJ8y4!h=eu^naa)ZTWx@D;(H^pJUWhv?Zw8g}0r008& z|L|hOIP{zvMNh(BP>y1vAF63nP_2u5PhYc+DKp`u?f`Om=)GIn)9Xz&f*dwP4K+obLL?9T5$JGf&%cJMZ)wjOnet2Vg!wkl32 zX6|G6_aC%2(VdlvFaMA#8b3*lYJYUw#U=ozj??>^FYO8>c&k@%FrGfYSQL#?n&}`V zC7HW?U}`LTc#4tubeWj>>745Lv#haF=G=MK=gNcIuI6gyrfdQqq9LX6<&5;l7E0&} z(LC9$I>%0mEmN4-W<|F<+SlUs8h^5>Xj8X2d$OmZdB$$%@J&$A8bMH`6W;Sqi*aL0 z-u2MQnWz8@EvCC^xu?SD@>r8lWLcaaPsR@Zd}3=D3(KE9diaOgm*H-*+$SnmnVT&W zGUk%VwA=W-gt^;-UZK;iHNYZf{3xxVl(O#WiOuTgyNbh}1%`1`>FPJjb(<8=f@)b6 zYh4&cG%ZD(=Iyxx&6Ybdz5~Z5P0&4QV|xp@LC1`l(9R{i>0+W~Q0k>tc`>`zel^EN z|1*)6e{b+2>xBgkl|4x%gYPDr~Od zFb2jzXt^>R>)2nTgZhx-7r8WxrL47Td;GST`stG}oOjK4?r|39>TIhO$#>vQ4i`MJ zc+|rgeR$=oWsRx}eF`IsA%kOlSsO2_MGogd2El`qC{*Y+k%j-M`yRY!odD?KUgHC% z^0901%?W*&Xl^2HsI$#l_Q&4RVI7C(Nx8Tw4pI)oC^f-Kb&o|cTdIB zJbouGRvOJ8#P`3&a=sy{!IIb#*iR~dpJxv_v(F;4;PslBExBL#%2^feEklzTFLG3 zWcIBUL!!rJ{j>Af{E|1n@UX}65FGF%#)d~#F^u!Pk-7J)Uwf8-<)#MeyFOc1;u1=& z($$xxJA(@5gNmd5Q*nYyDa4^wvX_U%8D!^9(La^eb<6=AqF-y*0vm zQH!s_Et#dt!A!CLXjj7)_y_CG(;N6)Y=rIQuTYBDp!IOkAgO?*%mBBUW$9*i#<31YxS7mX4TX zpkOo9UvS97ShwMuCZ=S(U6FILzDgbqRu=fFd4B@BDHkL;q>2UME$Y)_DgC|@qowC- zeXR72=J%rh&b5+HO*wiQ7m}zK*qhQmTvr(F<8QTHVJ@Yx9G*VKbsD zv2y`aqCBte$EnIWPXd(+X;y5Te`y6cx}`}3#9zxc(OIGHc`XCK3r2MCpjOtpjdJLw z%koX}x)oE}$6od$RXAyKGDDwbtY)fhKG|k~@|{Eva8?cJLuxg9JIRt6jlHqH=<7Nh zru`xgq+yffW~q{FTgfi#Cm3%=q%pxfCW=pH8E-iIi_enZ4Ro6jZ6m(dcwl{+pT94_~IXcqprg3jzo??0uN zZhFjf=E`Q2pD}#=kRU?k0wVd4)H*^JcFR{~@{>ht-`Qmg>cOUF=OB9dd3$=I_PD;F zjc)OiU2)pXL~Z4Z-oGYWBcHaJy@kiY&pi{40S;c6x2FN_P}&NI7XD_Z`n-Ccfp}KM zv&$l%xteCk(4Cl!mKS}Kgw`Gtrb|3;7Q4x>|Mz$E%5#x^~Z%$bZ^`WzO5b6#fEDrYSkqojE(uT3zpn zOH=5K-!U{w&24koXf+&HX{ULF)WlDqz)P^I*uqpfSNYGVp=YHQe^cgBnfk za1WJAYB7amgMNGTWzTWR)g{?Ql+D631Z%?Kt*3x_DedM|%9v1gibU8~dvkoX!@5Ju zHP@Q<-9XrOH{*9Lu8LU*C{8~uuO0P4i2MxFub^SOXgIYqW`S1=vMQ*$^rc3O(j?rM#kq%@ zuQrfYOQb{0chH;$g{_n`@^cw4zfUEs!5v6ANpg!g7LX!9fGhD%Gf8QL$J2Q|PMMc& z^RXWACUsxsnZ?r_MVK03NHPV|SD8fnj{&E_%30?|dF-|a8edUTYYr!KnG53Kv-K>; z)*TZ3k-yK$#lc$mvuuRfuXFdkd?RAhPtXy>hsV4k%#8>~?;Z?pwULfg_qmZAg*i~^ zu7i?9+%%1;g>XRWiwA)!z&t^u_t5?D1bd$f9mcXASw30`D(Db#=f-%PS=9@V;F0EI zNGsnVK6-`YqHIzgLlP%_H71lf!v7F<0ptZOV3aNY7O4y@xJzN(VtP(FSKy4W6Cfo# zt9lqbq)=JQzk_rH*LY#Ez^Gwp?$j1W`E~y&opvmV%ps4)jibC*JCsai$*~}#?n?KR zl22bJ@a~Lq_Yy9`M2{8ZR5CV3Vdw)7OX_21X-%#%=*HJthx^bhAXj6$q6o_jp4`|@ zA`>IXF$%vuqbx4o7LZFK>ny>GnP(OCXF(JGK*}QO&tMT@5*}$7S;@SDK-4^MO}yz* z#kclNgrqtXNB71ty;rbRHR=|n|3C)KEIA*0$CTegk|47NSCO>)9ox29)wg4NWXh_6Xr4b(wuLG5%S1TG90pM z+G}73uf1L>kH1&gLoxBl3SIZ8u4B3 zc274f-@~tU+wV^V=ODfC!f$yXA`7~jyJ3shjX)Q_9S0g$K+rU=2#N+~1bX-0N3x*G z?7rajhmT~1krRb@dFb6G(quvBq~7=f*mqz);QuzExf$Sf?vN>WSQA+3htxWu08g0>GOGSP(LQy6UCr0S95UXZe zMCSAUhyTj~xPGdbAbpF;CHx1K)uIf4@SsJa>|hA(`c$>d1r+SuUD_(g6~gbNM5M6$ zpjDsn-F3;sA4KTClu)+yelytrbi$sKBrr)NyIb%vs?BR_qagAE4cDV2{c^V;@~S~? zPA;%PplVm&fZ@mQoYEvD$O37n94t4~fFXCl>b$8kA?co^Q75G01BUzM3@a2?q36$mFyNc7Um`)75@UTo`Lxu zCBJ6OEkUw|&T}=3v=jP6R$0pGCRz5_i?qB9bxL%l)k=IwM_962CRx*EabnoQ7RX0& zvRbB0=MCc`CC-w4j*_6(w?|k$Gnhwlp*5~p^Rzfyq_+t7TT%+!5q*6kCE;&!cgIbl z-*QcQ=0c0_HoLK?@yvi}N{tu;M39hTR8V*1PC(C)>9|d)&aX`&et^4_9qapvTJqa1ZNju%LGOGd1(mSSC7}kb z#r)#Fp9R1MqvNu(Q@8l7s&4Yz5j(XpzOwIUn@Zcq7me<^Dzk5-&ab3y8J|A?o^8;J zMjKuVMT(X1Zy5gxo6d9BEop>&b=OVX7fSX@Ypkcg#^y+F^ztA7SrG?0u*q?kTe2F~ z1wOOilk7a(%wQ{%-?O9vR2VGZBK}g_OP)96=KdkVnE{eW|8Sn@^m0UZG{oxnY$EaN z;iW~@XsWK;@#Mz3ejw4XYIZw%C8G00=x3APn(hs6X)S<^1g>xG&KnwVs9Mf(rjr}% zrLDVwtW=IsP1x*8&H!Zl?;VFY%R6w{U7z_ zJ!Q{lZAn*V3yu9>7yVD8X;;n4n)C&)UmEB2=+GT_+0tP0Fvr&~;yjk@&u_A=CZv6h z8nXh2P-Om8Z%semqzF>}4Qj$1aB1AG9VXaOt!&lE{N~08FMqh6L%pb41~0I%SR%wS zf+Jhm%AExseKFJQ&@ybE!ug?jU$381Gr4(t)=^Uc*m9BDqE(^ z7Uswta|hLEOgztV`ygBvU!c+SQ@zbtZMaJBc&FHGClA&4Z(uLbE!Gv%Gp({MKL>~MvInrl6 zeoYg@@mIJ=3$0pnQyC+kS6A?8>80y*-p?y~F840CQ2IJmm;V%7kkp#Q_}w8`*!pp5 z$NX_wWTd=E5qCjjv_I$f7}-h!NrV#F!IXXrJcSWL3#Z9KL4R#fmRjRa$Vz?050TdV z=<7dHk4?On66Z-A`Fh>k%wBY9oOW`Je{yU_z5cWWqJy0(Aqxlk|5^?AuU6AtintNhc0k}ec3?*5{T}LX;U$|jx};HwTll+%Ap)m*&VRGsL*vb zWh~XIJ9ad^#qW$+*~R*~JD|C7lp*i@|1kEJVR1FV!XWMv+y{aV?k>UI26uPY;0YRn z6BwM}?(VL^onXOTg1hbHz4zOF_SrxCqpG^gx_YM1^yxEwx-59_xLkIAxz$bosPyoh zx|%Yl6TO|hn)5luoi4O~JQDg0xj0J8QhrO8PFX*eI#hwnADKS=)YbH_yVm_8P+t}d zxYm73(}Z`_)esr`1gWzh>%MtgX~KIP{YRlklh@|zUy^k@AGRRV6wD}x0_N3NqLn}P zwb&P*StgTYz-|iNL=N-G_-_6SXE_Rj&Cg(e@K-C%OTaY7DsYle10WVNBbuj1M1+~9Fz+Idnw^eK5>x$n;+w88Wt+= zT|_KK=zx&6gX|fM9)aCL79%O zt;#Tl?p){*_|c&S)szT z@T*rsweXqAW_5HMJY>x+FPo-Ex99Vdr0j2Uk_H()a+5ScGp^8;HV}mlnT9&1OI9)h z54e5zVR)~Dcd}u4R#rT;tGU92kjs;_Boxk>R)1kfYb}PH3ZF!BKY@BR=-Yv>6I*{7 zY1I+$PiqYN|NPFvPX5pikl9xd3U96!4r?xiKC@TZg73V_^;lN-)_*iZQ$>P9FK_iy6ozst7G zJIa?!MvI)@809QbjVD3rmBYqr{ zlZuOZnOHpG{zTZ_$bS>`>nfc*r-B(w{=LEAs=6BE-u?@gUAyr$tj762S@w>%`MyE* zBfR6DP$#D=`AS_={{9@1?UANstI>7Ho9~oF^6wX@g#f^n;@Uyx)MqiDiMY8W^pJmU zzYH?|$TrS;Sg~OH;_Wy)@#k6Pe!-=dVfS(C?MtCzPm2JzRh3L-%R2u=>RdD;xKjZ9 zuW)dcJdv7^Pkteo!ERjxu1=bZ<_{^H)}rJ2ENnMudCz4*+U;Admz*d&Sv?mI5oJF8 zN2;rn=W5vP_jvp_G8{7|3Tb9Tmo*rc6aA&@N;T@}VBbpqkz8iKXCtNj1mPo(^XI8* zZm+;`_)q7b4Es!WpPKe9Zrc^sr<=Ly$~Vr+#+~gamAhBja=I${1z-PG<=JahFPwL( z@s}EmZ>Q6BSL(7@U23LnEM`_7yNf9X=+_&JA1_P29L`S;aOMkq>a@kr7DoT=#XE-8 zd$Jb6mi7wgRuWrlpO;@ipV%Zdem`4%HbHxER~==b5{ z?Z~Begz1;4*_R@rP;K@nY|He;DpDk%zGNnE5i6)@0f-fs5SGo8FG(tIoTu0h zqQoeKu>1dGf5MQC=-Bm9uGXl4Tzc1#o2=_yU<=1E$lEeD|6u3OO?>uGLJ4F?aNgDgGcSY}!UVFA zq=%_NLmR3xP~@P>!$TQ1K)mBYyeGFk!VgmJ0fBfQa&k1FicuBdU7Soyp(dFj5jyf? z7AK=XwN+#z6~eR>n3X~uvX{&w&$s*PHfo4YQ-M)7%dsE^F(DA~P>>pQCKJTUVp-i2 zhzC`Yqh>9N4^c^sDTINt^M8>+=A}|cVAT^XfXmXjB!u^pul+(v zh=uQg)<{&SBVu?EuraW6B4ccy{jHbKD0`eVJw+R($yH+~0t5k9TwgQX1@HfWG%PDs zDqs<(yi+ENP$rjdXgDfsDW(qdg$F`VglkFmHpmcFD)NHU2kZKjzVC2{sJm z(`kYJqG1ZvM%JYb97cfJh*c^$xPdDOz>~2YI%(YvZ!^dWeBjW6H;eG+&=5|!qwben zI;~Zdbwwn;pe({-RAF{y?b`vq=4qUKw>fV!Jj(!tn*pF^44rfU&5?w`cGr#}@t_(#u3$e@l??lXu zIQ_3FRt((`QBao7PY1un;H*@<>M0%R6GlsG{<7L=$YLy-4~3a3(vImrh!r3EmDlHy zznZ73o7*!4!N~t%+rEG0f1FasWed^<8_1f-=7De~Z)-k8sy{*|swhw3QMy^GzV0|G zMoJyj%cz?908uU>xQemj-H*e=0ga6y9~LL{xdZ0Jl3g?T0tV5KGd)L-ix{0$t@E`f zP9jM(z>RKPJBN_Me0_t&S#YJYk1>ZJ9$(}!TJ4_~7)uU{NJ~T$-2K~7K z-mZVaL}j3n$gGuiV3dQs$gGxwZySgE|DkU`%TFq9!M32F&Z*Qoyw+;RVI;@{p4 zOb%hU_<7m(*dVW!916>d1~5QgC|zZE1dxf;89Q=UHpRY2%u)LOR zC3?n+9H_0G&h1YBw}P3g(}f zFRmn7RbRCzN3@+#F!HEq+drV z-ZVUKBh>9U(*%uFj6o8VMaKwf^?b>GiB)1Q1h#-{7QnLQ|9}sLHw_9t16pRwst43U zjacEO;d-HU%=UGl4a!$oP3Z8N>yLcFm zvalB4*1UNbj_54_wX4nwQ*V}lArq?Xnb4S@`CwiGA`qU#Ax2WM3QKAx=(CG9s>*6p z-|-9}`f{_|>UPsBVD@fxK5$E}dcFt}9x+f`SX259k<=9GIQN~Jc< zKBQ_{uTHbi6Ucw-8*b5v;&^J|`lmUbq0B1^Qkw|dxWyl_!b=x!nrN2gDW7(Td&D0} zdXs2Pcm82;^r?fnQ)>3l{3q;O<5@!d5M$w?7L6D8FD{i_GG>Rf?IEvEZ zUZJ%nU)k2G!c-X?aF;}i+aL0r%9%9G|3oJ>~<4}anvnu%8F;Y@k9{%h`a<(M=hrBA`UaK`r^s=aZ6$V z2V0q7-`{1?d{SnbYS&*>zsZDW*B{P9Y!dOtM#=5hBa}ul?pV@wuTLC5IkaK?b#K@Q zcM+H`fB4A*k=Jwg8_XvWjjd84Y=FB6A356rRSv3~2uvHr0o5Uvn@FH>uS9@B7W@Cy zT0Bhury01H;7m*w-qdSo;PaKi#N(!t=_Dcn^7!NCXB?>oerE}0$M@3K)CQTabVE<%ezE< zpQ>#M5WFvgY#j9Aq0^bl8sVSg&osF`-sSIgv(i>Rpa>qMPd(vhX8LYjsN&|DIx=3P zP&YAXI;q|n_}74RUuXeVqnU5qr)<~~lHtcK>-ZA>M~Ks&#?(X&avQqUyJrL||KaCF z<6ZM{G2Z?iBg_5rH_CI?@HP2YDVBczU9DqQx2*km{nGA3n}b`WhWMzY6ao&L+CU@bc+-ORtNt@x5n?^rxq}{_1=+{3Y4MN}GLwU3UWyjPk ztmh+koW5SRp{7f9B&=~c&eCngwFfAw4Z-;o(a7stgcVsMa?zOIsr~yUTP&$CGl^}W4&Ka~tLu2Bw zAS%ZhNF+Z8Y5RsA!yvBnIrk{~q<`z;nC=Lhg~r@}$Ko0Fi>`spc@R@WNB~|C&N`L! zQb`-6DGlm$B$8rOh!>BY_+brL3xOEtncDF1hL7i?UL$YEpf)0hQDQQ z5j6=wU8h=X4s3F$M1wF>&mq=R9~PT8#W8Q0`A#6pqWJUwx3~uNSpQ;z91_SpYfk+6%idVl8)z%JQ zQXu-Q&||5cqpiJqT!?0D^vCqL^R81sKrBXEW_Zcl6Kou#JzOeNUDagyAKqiI3)I#I zEsjB=aEIv>U?|&MxNMcKj-6E5SpZG)8j+pVULFGvAdTFAHzNC)Jyf0_^^cC=WIRScY2K)``UN`FAo~ZW z?KXi8#4xxz=%isE;ZrRSqpGCX6E5Dvz^Zhx8q>JP0RF3-*R^K0grf(-BmOAOgu+ z3nv`B@Ie(*OJW*d2Oav>gxjSmZ{fo6qE7Hp0H>yUvKZ1fL3bZAWz?hCBE z1*q3$;#P>}%8cEn0xZN&6vU5X&_WM?kd}m~unv0W_5V7K#;QPIDTP$7ksuNRh=iqv z_@Pg_JP1`-4?VE%wu%COJp6mD3E}pO7^;PT;Sik+H$IXS1Y~mI#X&LuK_AE_NN3)f zIex{@BIx2r4!m2w|E+S3sI*pCWvwXW>)wD>{fp13L%}<$c`k_xdE(6!|6>8w;s4;# z@0(qSITgZz_(+zL=0VXF{>J7(8q!NY0nG7{$3~tk5WXFRbL{93#r#28SdStT$0lgL z>6CB=2J)L+JmU`_K@D}!)1zpl@mr6@M=C6<6=fLBddhw)wm;jB}0ul5ZDBIFayh6{{uc0L)54zn~2VpcFvob#uCWh zA&xvJwG{4-MI#QJL;r*q4{!D4aUOSi-^1Cy)<)>PA$h0I6!f*NuL9CL?FKHs;1f16m30&5KBT?_$W;)c^oehoX#~?Sb zSz-|A1+`eKYSjM+F1UY*piw*oWpbSkfM_JHxZh%6a6@C9z|6i2p)q_Y{nF4VebUKQ z_u=Ohq?&@wGlFO`u8XF8$I{C%dP;nd?d-B}{UAI%FgY|>6PqwLZ=QFngv8Q*f@=$? zEc`9sPJbpyM@s7dR6u-@_usdGlbUEUmrZtjWh{YTn+TC)X0Uj+fDBkl5Vx&@IfSuo zpiYu1uYcaJkRU$dKzwCQcm}bM71CD8N-{czDPN>|+*XNvXS+Bw#2YJ`j3yYBON=kF zIcy6k4`~*7lHr1D8GsX82J{Mp=>s*&WVD2A0oSR%<@4P_qNJP&g@krz$NVk-5RMxX z9l^g)?_eOU{U?XyD+aDWc39_iIQN0TRm8D*ZY!br3`aaU5HD?J87>-*W5OrMTP2ai zOelSzOmv9P@@{0A%cEWwTti|dzDQ0$KS)Cr5=CJbvW#m#X+Ov#8p4jGFp5PgsZODS zEKS`cMTHDv12N8SVOmJB;^9-MV2b+okgS)#(6TglscrwrCnk@|gFycDa9+vKf3OX_ zsfEFP&>;;I!I#O68w4Gi>uQ{Go>zkf_DZZo=*!YnqI~_9WOjFBt^WL6)mf`eHqqiV zAb7=LAHmQKaq~gdRgCo11~pYO(Lzo^_poFq`d7{N#iI(>7393*8rRc39Q`Jq+L$LW zcyZ2FPWE}{Xu}jT>nl(NvDr)MA`&2hmDc|32GfwNbF_iq8v=73W`nY0@ySX+7y%a-~r+?o&*Yz^1v&cdE&Ht;X=(BgQA$#XBd??{{ac&PM7{Xc{4u0U4UqPg?DGd(n}dWrm^?%KRI41iqx1 zZe7iw^<>$b&KQ_cRaR_aBNq5#qsGHVNR>r)Chs>7%Dg@vh2G7IG`}GO99h*0z_Pxx zX(mSoGJ@SjXs~moH(mZ=rmwMJv zGCnuH9hxM!Yg_N$W)xevyX_z|&Mj+VRS25umR74iID2?0U+KzfhMgzhnJFQ@&o7M& zvw2Xo_DX5h6qI{lG~=@vO6*1%Prz3Y}linJ7is5jQYSZ z|8Tm^9rFj@vd}<<08Thk()b^7T7C5pV>^F-VPh!Qh9ogG_lW93o^qn`?nNxwTF!@K z<=f#F%8aQsuoNTSZ6wI`^ACAduw3-OkTXdOB-66@H~;%i3`mi z&*YS){8s2}d0;8^CQt?($dqrodY!IUqauLbC z&Wm2MAa23QyVPBU^BgpxrjVUJW`63jcVqy_qx=cFE}aubk9ST?`sCM9jCEzMH?DYx z%`?{^!j8l(E~1o?HewX?*B^mRI#QWxcU6dRt}vy>XM7};rBh~k&L3ef7%!Nn5r9FB+z$tv0)vVTulGkgU%mnR5~qXOZfq6o}9_9ov{ibQzxZ;n*q3^Ji4#x zn#76`^FtqXD8|M3JhFUEd>-jh({-Wo_OoWH@US_}nzPVw$RBlk2sS){;FVeUX*1~T0l0gdq+AYzY=6%HFe1}gCYF+ilG|&%%9A9supa6 zs8I}_(9HX(;CzRx5b?Qdr9Mu3O>Xr2>2?zKSti!ohx>fWST@!XcYNJ1G zCr52V1of^W>Z8jN%i=*Je@E~LbAvPgBr>vzFN6P)zU6{{%+eCN%jVOn1_>XP3+8qq z>8Yh@{m#=xwn~y52(14{Bt>?XCm15W`*%+| zX~{|MepC2dFHcFyYfzW3iH+52G!grc>tV}~kb!b=iRNd`P|-E5STN{E&!c2bdT#wa zh{-G9Z%g*azmBixjZ579Bkb^Kcf2y}++KEd`m*UF^*P)$Cn(YLnqHq|uhxZdJg4pl zC`}KuV2>s4_Xf_o6&F_?IXxSgn}5RLW@vWD4Jz6VTjFnh!R54-05=B4R;J3?&n0`F zhd(wY@^y0~xh%_x>DzXnQ7d=SZjnBxd0@`3HVsG!{8Y9n#er$5Glp95;kqCF67*I9 zn_+m(`Al;NGCWLp-L{D{wBbyS3nQ&FU-uARNnANi%1+Lc=={WxyZ(hA=d9j(^!%3y*k)c9&eb!l+?zhOR$4Me zHn!Ij{zJwOn9>hOG+ew>WeFFiZLc`XHNfy2|E#+z(Mo!kdRF%~4`CD&X6fMYA*z$A zm?U-ht#c`IuADJ?8lf7D{;UmLB~GUI8iI6y`2y%QQVfphtcG2qvSU(kSIJV7u1 zo}HGf*4D9Hs`mF5PNx{sq>~^1w)?+8tpMZ_<;MO}gE+L6*L&hZ-oxjc=5yq``~&%3MG*AZhI-TjhuUPl=JVgUV0%^Bq` zCtQ9Or(z~e5fV1+VaLq<#KSDXOdDNR@iYjce+&X0%&iJb@J**g2xhPU^+8busUKL) zYE@cfEf$X&w)r}LOBgtM?@Nc!SzF|AP zN)fiWX3<80t($LV)JA#yQ17f=g?k0~kF?poD6%mG|uRLXbV&p(^S`&!A%!xmY&UTEIqZ+HLjb-DEG;yoO%z?-tV zHj|TDrO6_nyTsSFOvX_VQ3#b;)R-li&kL|J6nF?Q9IVe699lRrdkFCR?${WCjYs0Xwg|#??tr*7 zlmj^OFv~Uxwe5bLd$>J+zg21{b8ZZ%lY*srW*Pc(!yU;k7)Lp7&!~9OiB+U&uqI}$ zdv9nH9^kz4xJ9y%yh~Sykz(1(%Ya@oT5SK=!N1>tPo%ielBQ1tXSl%jJga3U(c{%w z%iLLHJ{}anP8rwGXw1vpkTKiA(by32V*A(!fSj0cfGaUyc%iwn zezU8>mwD0N5Fj*UxR`lrbzxZKpUQP|*aOI@<_-Atwg22xXk+!<^JbdL1?8~Z_$%$p z+g2YhYw-=g(yLc%i%6q; zSfvd9fobgaP3{jyHr$E@VkgIe%$H6_=HF0v2)7l73|x@3?p%(bYOc!PAG3JTOAY|N ze{Nd3XFa5Z?@9j_9ULVUF&?*uyYhCDR&yDGri6dAA31(0Jc{(KW2iZLr72@P9ytyw zIxzLgwG|cCwOqR4>wp-)t`j17CT4Myxf${xa$79w{iknYE$Ky#nzOG9T#w+910_~2NX96mm&@hr zqTgppMa;*AMfaDG&e-i8TauFx*Eo;a%#Y;~{4K~`C`pmvL|hEuBffRBu~b#!wwqUY zT`5a>>eOVt6=e9fVt##FRISPCDOBmZ+*p%Z(UlKkZO2d4lym-n6sJ+#n(mhN?)~P+ z?mm4PhNus}joOu!{C2FOfy9FTuVCpPW+&R|F+UO&NbJub8K7I4=isanluk4K2AE=AkH&A+0Qmgv3!s z4)0<&c*D;7{k6Qp z&J*#k{>*kX44?N|^_~+O2f-dXCP>i*Q^CLwFI+V8e+fUIxD@mc4{<-Nee|9CP1@b> z-POEt3szf8hxOV}igO+k9@Y-*diuXro%#Ikmk#drV>g<2km5Gpzu=#SFPbT3`SIf+ zsP%sCs!5hRHzJ=i#a}GTdZ?nSI_txRRvgP6WCAtp+IPXV@Sv7|aU{@sU1zCyB1@4x z`>$D+3W2Q8rp!mRli@mU6McFJ0hgPEDORQn)n|QW8>?ZO-MpEdey7*Z%&7MF#f9(W zEkBFr{=cVEv9taEJeG=&pP&8zeJs_VN$PgT%~zp8FS4F9wbB9x=w29Cvu6Q&Sf9WU zzFsrsK!8mv6Ps8LWFw#limG#hY59ed(iEV z8gMB5_3Zg~_Sd(&`&In6t#|*|$M=}8Q6$gJhA*pcOZURX9h|95CRZ@hvF^gLRD(5i z2#Hpy=Egme(1`?W!)S?3@QNQ(H{O>;RayYr%N*|`Z*kW(&j0{D*`olW;tKWnBi|G! z;79Lw=2LcH>JBFMzRA1y$X6lLkE=fL)*og3r%8)BCtXFhzx1px^NIBo)WGx&{L(Poj)dsPJ=PU@zz}L&cSquMW!SZ*)EYNA%ZBU4p2qx7HHi!!lY_=T? zCca0F73?)Vn{+k41hEjlxv_coa<5vIY0awL)Bw~;U45wY7s z2VeiL+yhlD%N_=Hgo$DoZeqmyvVeyX^!`n)$lm`~Yv4NrVMI18GL_;%)#ObJZgj^G z`x%P#7Tr-Rh7bPHyq^j-!cwIB6jdGW+j+*z{S}a|%Xw07D#v7=DDx?kS|XxH)oq-U zZ0r?EB4Ty?4Nu8nct^Z$`RB6>^xw&}#`0TSlSCXuuq14z8H{Qwy$^!y*LbKNiLe_A zME)kF-c#Mag7B$nObcqx(3U2|8YJkCmbexl%xJE1;#2|bl;yQV1GVS!o2u^6A@MLW zf;Q^0BgS;ob*@?@&oWZz64UO1vHEppqE`dec=3H$4wXjVpmE>ipn3;8_FW4Mkgq6I zQH3!uREukDrGjaon`fx#bMmyP#Y7)U`W46Ug*dWNDSRG9CFKY@b^HZtc|WM(iw3U* zmNNyywLSRV^0Q$+N)^0cu`MZ{D32e{foWLVk=64!xxQ^ z;EmEF&H%12f})HnzkxXMfk{nh43$RXVA|!1o%Av1LtB2Gi8*~} zdDOSbby0`b!$UsRc+qeKwf5D^8Jf2tY3ypN@0TC73O+yuE8sXSS=eRjyVW|&^QjKq z6~SoDu3JAR!%c^C#b=`6<|%QojL^Z_NdCy_$X9w4gKaHFAPtYrR-r4bi#L6iSuZ-8 zpMi8a2-h{FB0q5cMwgZB|?V_Ltc6dxYqJr>)RPxJ&^N_l- z%{~zIwKPl`wXCT`ng9=a{y4)k)u@al8OnENT@gyVWg?xg+}mh5C4qs@$ddAN(6ZDY zqoS%~;i$3UJzk7RlyT!%RvV9~W~_b;{s6)mTmTr6#fXGIqZim9pLew>C%}x}zNl`0 zK9rc@0iS$y3Y>51XZZ;J11mY;HaWF2dXZbe&P@>cLEa*IqWl|cVpJ!K@7_5HhfXPe^}^!|wcmsM!r7;;xxKk#0K z{tijS^a_XI8R$ml^$`0>mHd$4{q4;7?eEDa6UAxLX8$2Zrv|h;ZJtBt*(5RPuFW7F zL7U6sgW3>z9CP=HJ~;#U&%s^4&)zi5o^hXwQ9b0J;69yai&Ca!yJ>wAn&J!aJtcW1 zn5@1fHyuxvZJ>Wte?Iar=ryvhIUH(~3KKSGC$hzmvA^*$pCY=&+e*68OZ|5#-6{@f=mz!aWNO~3M z-#3J5INDsnyvK&4Xl`(%XlXxzaXa~D!K3vKEyIQr39K)zt%&#=!Vg)#Mu% zI$}{Su$n6`m%*8ZQ-g}1jFMbj;W$nXk~Ei*Z~2B;82=qCaHIbiLjbKrNpyBjhXWt_ z5lpO53CQP8Pnzo;z|u08yOTTj;tu04SHW^RlU5TVJ|3E%u z%zjR?T0?abfAW{aNs!@p3D8o-#-YGLe2tgg`|zN>9AEEu5@{noMl-Z~#va^2td32p zPRuxSNshtg=*@+-MYg669O;Y;gx};ZN$@=&407X|XVJvZBkcgpt$w^uHRz*!{UDj% zPxCF6l>#C1=;DK?mB~%+xP!LXR#bR1WjiqNx1xlt3Rk@1Ku?Y>Qz?|gqwDP4A$tZD zTG$*dlLy?1r>!=d0Ib*){rWM8onA1#U&x>A(l=8dD@6v<-NSU6-#@AYV{CwX98-bM z_9@zmW_S$`FusZ&Y3Mn}uF~{iAUYm(CT?AenLK$P8{H!dgA_)R6*4t@HqVC_)F+kB zN6`WQP}Ike@JBcgb!G3BKlRFlF{v~9N0Tk*CFMxaVjQEPD2{Lu1>UJDqP#tAqPw4e zhf#Wo(nZoaP&a&ap2|9boA)$4ZdJxUKF*C@KjuX8Wk)rdh1+3KWSvh#Eh$mxQX3^Y zMhj8n7*@6B19S8=eHcxYpf4=0u)@tyCGKUPSi>0qHi=s|{H0bQZ#2%W(E1r6?8yNe zG86?SF|1`kK6AujxM6v=6?c+|;W@1E@%Vc>!J?qRKV_vE=6W)KOpk2LbD>esVLCp} zyl)=!s1w?(g{IT-_^PA7zv`FrgRgc%V#NTL$nsTD4hM&XuEUXplh9C{v;|FgGa3PM z>8c_R=Fo*d*L%b{h#BaBr7}S|GaXETzwV(j@S$DXRRD9xc^y586?}7BM^rAX$jQCW>Y%6y#*i=J;sc?Yr9aIc3`7AtmG&D=3pxqxF1GTeVs(aF8TTx4LmO;(tJ7p%siwfp#4kOJt&T*)5_04l>R3w zKxQ+|w(H%_5SUk53(g7+LM9O?{6iFXrGqpxs~J8MMKDC?UPVYM~~fTt%3`~?+j0)9hX}RLJ70GP2D9><+@ZB3|Y!5ON1oC;?Ljo zlbxgyU7(K$X%ftr`^sEBG4_acjkOO*-m}j^GIJPUn{sE$Ir& zE>k=t1Zhv%w4*>1Y8>CH6Hgl(qbrii%dItQgC*$8zj*a?Q2adk+^$Rnp5)OJqd%b$ zYR*;AhP6xdH)-|ukqIm;IA-$v5QG)O3>6|Fhcmb^-ZB?`#Z^6@x0ArTsF_aGtY{t~ zC=MOvTOqX%djVYPfulh;BD9t5YiNFgVkIeQrYp^dmAB+s%U#4!g7Zi^tOZBF;9jpO zD(xeiCX3dGUDX!oLdzB}i}QmuTE(2~np80g*Xj?6S%d#3Nf7Q{smUj;J3?}1EIH%? z{;EWWQWE5b`!(dMA7^928&q&rIjbx8y3`KcB7!|a@^G3n@DLz=g1WfCJdCGNL`hCj z%4=)uM2UG`yNacqoU}?NYxmMB-`iPCluGK2irZx@l!)!H>H-iNnA;q<@ur_A>j>hy z*vP)nS!cBSMjL!|I*bkb!N?txy`w;EA2?yi?!>tVG)FHo&#@Cc2Ys_NKR%#YO7Rt7 zL)eq#j2@Hf!)Vp$;;#hf2r8KETX{0G$cImSM!btuR>xqUI}93!8EkdvUwE8 zXdhnMp)x6mk>N}I(rK8BGIs}HR}0%N>Kp^97rWyno_RtuI>?z%?)7jT!-L&Whap#G zC%=#yuShByMIR%v7Z}e!m#R9az#dzYd$!M@BKY{NXTzu}CvYenpLH*SKljF)-dP%Z zwuvqexh*x$DUZ$e_InOsahDYn2N${<)hcgMs)7y4DxshiR@YQpc4C+;1{XJHZ~Y^e zS_2{7!Dmv6u_V~nLqN7DmDKX|(s*vx<2IWEPK#K510Tj9bm7Lg;e|pn1R8ZFwxcu+ zWsJfdG9w~XSn6AyQ8yQn$+x;?PAgo3yE3ubTDUYzG!42w6E?L2F%`n~eB@$|Jn|3; zF0k9~*+s>Ihm`eM&T)$DjPrWiCLHl?WPqIh3>Cz8?f_(f1q2Upq>an|4wy0i-D0O1aA*18NRgPL>7NY0d45Ml73Kxk^YQO>H=2%elgHKdkK=!E7=}^gN4xd!EBti+ zz#`#kvNxwRBDP4dCxCvU(}aN-;xrfFpYuRNRyga99;TyKc3Bb`r(S5XVuJQAzr&nC zIr1t1O6($T8ecgwMH5=fQu(eBU;n5oit46uY`8|&B#NcbZ+#}r4Xlvtr)((P$@uHn zc7|%+GO^3^@CT$n%oDOst5ErAr9E1i!YSFf!I714g`TV`OOj}W4@qzt(^#n#Q;u7T zEYIS_n_|!fe|36LBwByrZdSbU`I^!ngsQT1r&CQ01xRY8?wJ=_9<>C{Zl%-7j^K_S z3Q6d+9exMK`tqXx4E}RyRsSqoQPZ|S2==gr**o;Ky)3L)ESy~sRin^v313LJmEP+m zP3KdUwr3GQpsC*wm1dlfvfkj5$D-=X&MDxE1Z{Am_V+SaI$EN5dnhCIbL;;NY95|v zoNCb-BvnCC{mc%XNgr+})iia<=RReA!4{NxB(N7?!1rRtFPHIID=w&O;-mRSZs=!t zcpAe>TZv*(TvE~S?`D~(@X`VdqASgVI`Z@DQ%3lq)ka~S07l@Ejzs+-2yRL@%(H16 zSp_8fY{o5cb{7h9b%F5{Sr&x>F=c8&SL|w`|L6uST0%JeB_rpS~|U2zH5FVWVqvMmtr@kNToJWm^-1oVg4Il+nz@O|M)M~*SC~yf+ zVB{ei)7b06J_tzG7p@}P=93K&cqe8Qm{XQ7Dks8~kfjfwg*HUcnmk&=7oySCu2sn+ z(3Be$PDkat0$W)&zj2p*4OYXL5K_e4w>8i7pv|2Ubc_xZiM@+mdFbP9EpqhcH1Zl$ z63OQ}1j8D|2}ikW3^q(_KAysucG)hkWC$(NIAT7~dI^4;kiAfH!kWGV+(WrB!@IO| zQl6{%%?~z?+QpRO*S>vIOHTA|7bdH~hHalra|3sp^pF3g1*Q7Kyk>3LowxIwQ?q!< zisW8_Ic(UuE!Mhhckex%i5`W_@rA@I>DLRpAJg%hk~-n#R=HHx$MJ7YDx%7o0V;pF z7kLF@nf^9jQNxl8*;%oN`1VSGs|BUBM_Cm|iBzAqV_S-#3nzk@TR-opa@;bQ>Abs+ z)@neKkAu~&*(=`p2BO>KC-NH-R7mBCVdLu;?f#O=kMjG#7i!Y=NS5srf72}WAJ~i5 zv(k$$%zF|a!}(NPYw>4&_G-Xd(WbWav{5^>o~7c~^v?X5XRf`eeU%wcGJ~^UL$8p3 zZg@G%xITez3%>bCF$vk92~N9eK>i>GirAhiM^!fVirt8)NPaYZzx;hGf*Ko;UKBv6 zCSGX%04PNIraYzEPn=mheY8wE3R||&ez!E}^JkG-(?0aQ+>QD?Bn=HnMmW!L>xEHh z^tX;!f$j!#XdtJ6h9@Rap~=^lK8k8HL9Ddq63fN58Zc%6nd8rPbz*3AF;H-c<=F%C zEsqhMZ-JjRx&$guydt4h*APY}c(hQ}0){8M55C`4HXGnzI6R4W9@)jg zkSH7r;L6eDg7<69)ufbYOjNJ($n;hEz1YM&eSyHm{OWkobbFubJ7myLfTy6wI^(et zid|xy;IWcSqroa2?yqg#Mv;s>`VSg0g||gDgbI;R&}V-eO+uSxm}+u!=8xG8^Wd3R zabqS=PRi)H?)dYr+ZNIeRHNDktx;o5%C#5#p>QU4&D{gZ7*!f;5Hqyyk|>JO16H8*dhW>G)kv+%T%v<1EDAI#tjaQTi3W zjDBxzn2f$HIr>8Dt7(rS`moCdJmeY579j>+a1p*;tK|LOSPuuQZ`0W-R;}pw;91v- z12gH$5YyJb$HgWmYHx1dLS_AF&~j1;#`m47YI*2G>>HY_$A z%&U3FuQ>STkh9bCz4|Vv!@hkHPS((!2~{y2np|P`uYot`yM_)~kHqHJIMR;y(?4wr znM2(*Lh*C^4Ii4W$A*A|ped)-63txGKSYo3WLTkKA#NU(0lE7~4)O4(|0okyY&MT` z#itD%nE%4CeTGa+=uNDz(^Stgp5yp&{SLmbXYTd8IMg&zgZ4Sm_B7nLVkhQ-)if+3 zQ`F0*^Hb>P8f%+XujFIug}h{dRxOMu^0Okp#KD90!^YPy{qa5MVSTEo<#pZW02`nN z&dxJ$#fF}!d2ohRz(#GwvCujuYklN`W33P_1rSQH(l@dY6vPr3R{8#2Qzg4sHEtqM zn1}VaIR5B;wRT}YV(z7Mi%}g9g+mXgYiXl)LbZw}hruYYuLi2(_U8p>+`K?Xlp=0V*A$(fNtBXBXfRCl`5fP;EXq+It;j-i`X11CwOJ=s6zSSVi}?t(rTYjq z4!majjb*;Yd)rKMW%}Qcztw}(HiZL73OA`^)I>|zfMLh%3B+*TLhP33!Ms9O=Npgr z(Rft${_~Pp{~0OKACmK#Iz4<99Y!d$V6sxbwxg9d-Jp}_=e-HSOZ6*4I(g=rj4Q@X zZNlH2Jf{M?pD^voBK+)`0z{N0O^}8KWb>g6*))9P!x7fmD0clKBooA}73)TdkXEyO z`blMD20_+{LyUIuzVz3#+lyEHNEn|C2mgi)Edf4Br9}cd2T131gB|G6-v$m=tT@Mx z@_lKJN512tgW;B+aOi+9HVoP<|ivkqGI2M_H38dxj?r8X!E$~kHYLBkP+ zt@w$U)sy<%aR;~=rcG(v*OZp?L!gQJl*MdjwNiEG{Kr%3L5IN`xE7E_I8Szbx86To z6-Uf?Dl2M=aS0dv+UMd&O5!~xok;v{F%>3uUTKVzp+KBiyRBFrB$%U7jHLM4iDqya zS+;u9^)o0%qSP&6c*jBMl@QD^W9}{LxYm>Ng*BHgo=^4B@0k*=LR{m-;({~f#=W;* z53bF_2R_=eutWi)P!F3DwahZrv2MM2o+{<%pt(;NzWR0!YB`TqabmOY_}%eq%#sen zDr)wj-qD(fbTsENx+0!W=g;;J1l+5AN(S=|yPU3|`uEuZsR?|=p0g_a=7C(10jDZKRb^$oeS^hjo7pmq_Q$rN1co-!FRR zrmJSe^Xm-OiF5PCOab$G$2{+UG$7NNHHb6M98>^y$nO_7MvMXq*4Y{dArnID!eqE) z@{OUXn^w$&SF8ZwkT++E#TjJtYq_+dDTGOpMA$9>0r3*dBkyUoMifNh+r9=(Lsdyx zSf|Qwg{iCEmf^u)K5T?uI}&pTN|_Z7p4364DdlQ)y@|(RaXQ_YS><0&?P&_`+#(=R zxc%5~a!<>PQ4$u%NUr1r?mSDqxBF^_+J0}M31}#Xd$zJG zO@c+YwzPvitE4)~cq7l-I%6c9tn0u`MhLdsz&gj^zl%nWdM$!D1un`L$RhCF`>f~Osj@}o}9%b{YAqIr>*(<&7TMl zz+s@qLA2AwXdZ<+kv!VWMBp+bL5(r!!t_eEg@hQIlSqz*1hJ8}JRKYc zNa}6?D!8R{Fj26)I+-gn?V#LPGLrWmj$C0uyf1aR@=k|XRn2etTEbk@){pSlZ$F}C z&OC_>Y<&FkbJx`sJ?Y!`J*QL#GFAN{Y(it5Rr)@6=+LW@n$~_7#i#oDH4aS|`&q#2durmYk!C&|F>ITJ{^} z91t>=Q;-)n)I4P8cuccHvb@l-8XIYQ(&9h`W3H*-oNrGFb33br*e?cmT&SUEk_EzJ zR?25yCsY^=HFFcUm$Rax+<>-PkETmckUyKZ<=~oEIA%SDH$-0yAh0T$x{$&fLb+>|t{rvF>%9Q|X{H$O)7H2gG5<8XxYylc-MK>cNw2W5PZp?gfcNct*#JXC>NdMEK z$cp4pg^NZx?O8UE5I(Haz8_bV<_EyqC z>?OMiK@$!_#V?V6f*w7`NSyDP)TJs8p_3h85lxx=Xe#~aLNK1ZC_qU^PWih^ogCy- zP&8WdCEKn@^alFw%F;zbIAXqQE|8i;%Vc{wT**gt&u0P>V;-CNui zWop|w@dMav!>H~|Kai9!Z52M;+-sqyL9UKH55&_TmApqWt~ZFXXGe^`^Pasn*bC&E z)#c1gs`Vu)FdgjUs*0NRysR2cgjM~W zRkxl`$>$tG5Qqxm8D>%di15!H{XF&@uwhXj`<9GUf1w{70Z~A+VEI4=iA_0~Ulbi# z!3(SQ!;@5_cx9jMyNSU@K^-27CcTb~x6(J}H_ zNZ~vDns;T|E!;`>H*uWCUiSEENXE-OMO_>6hY%-01*qkfX2pRg;`Vr44rxJKY6m9) zQ*x|XR=?Q7@toFe5TEd0jaC)A5AhX=9%m4qVB@e{US<`V_b6V;Y>wE71ziiXw2I=Q z$%{H2{Xi7Ho#xatV~$peOPz2m0KMrT0bk+~8PNrv;u5RpU3 zY^g_AcX4nMWU3)%c3@mC?2m<}&osYZ3-@t#*{M+0n!(Jm4osQKA$#!cLXOcBF6x&D z&*B8J&qAF#acI;O z%SAn7toV?cF*gsbR~b3UzJ5J$;@;0tnGaVnc@eoLYj9+L5<_1#Yc^XmvEZp(GG404 z@|@0uQ|(&nhE4wiXwbbkr!#r)xm++@m@O91)orO?!Du(w!hIv>gY91EWMg*9K|bVm zXvmdmMq<7JBmPvaApF|QR%>55lKg^GCftC(x(w()+5IKqvOP({-S9luplEu z6|q)Ar6K(Kb1*y+=duu-xmfbPM~v6_BljIb_QyZ{zdB-9CO#W8~zId2~uA0Xl>w_DWU!+dGkDV2Xydx<^DCH}$2|2J@^x zf-vDh%ydTIZXfK5sGH|CRl%IW%p5&+xWsYQMGnga^0h zgl^7dS%D$7{Euk8{5vd&DH9)}yoM zE5`*4tx8z)OD9Gs(f;}YQ@-$4FQ4?>FrdeY5zOWtO-uwHop-#kaj<6@HLI-#kut=X zF74uB{<|j%8+F=wC}&wCBxcTYHMzFT45XSbV2F9chQ>#DX zLoC{fp|~(+D5}aPBxa;=Gsb;bV*O62KL>UHc5B>}>+k_YJg!QPcn|6vbo+hD!zGs>qhtiN3XQyPe!h}n)EPk_P*&86y1bZc& z+zs%F;cFYHLTTl<;ep8JNWbiI;m54w~*}~lxKCWkfISPASmy9sY#LlP`k~&hX6RI^$ z`|eK1&gjlABgdS*Cq8=ShQA|-&=^O5e`jAg@nT?*?rNdq|L`pJP;XGL$TLJ~4U%W@6jIQ>jCO~YfIN|>*Q7Ao18NHXYipW@)9I(44ziOo0P|`q9iGlL*(3rvHJh2H-Y#HQ?*>$ z`bV~5qgdH0M%x}4{?lyto+Vi%#xtd2K%*{~z* zp?gk{E~bBE``z%{`x{Dhfv)OU)!6nt&-Tr7+Y8@qULall?gHPpmCh}@ztb&~2$*4i z^KEmG;1-Md)1w?DzBwFP;yj+_42$MFq>nz{Ud!?ghg3o?0L95=5;%Ayhye}>1F#I6 zXX>TKngl#v-v-j6WyTmZxt_fLP9?UJr?}5 zNRJ{bEs!5)JCzsYJqT{>a*fb_kLX;be>s2$q<{Nm^6Dnw^Sp_Y-%qcA4bYXy5`B2; zk*MFQ)$;67&MfuODzich&7)h{|gW1+oBxwb&ffJUBwTw{5JI?O|W+f=C;X)5L@PsJCiEd zGkb8MP&~T=!|RV#@ZXQWnOcIJ^0scVV=2mcbUn)kKR{;8D=l6=l&M>t*AQ3MpJ=}O zy~ku;b$h#qowprHk>p?Hv;Ps#KtvszjOn2q=p?HzyQ=XU*6*Z7#_sdS8s&4&pzpOp{1ls@%6GJ3lkyYe*G&Pii9w^2HoL-|l} zD||CZ`Xb21Z>xm1YCT`z-rgY6xop?w(%kZv>; zwunc9DBEwF$0LDmvU}{tD1$q4^=NiXAiNaY1;fLc7!KpF>^xDV_49>3FZayREXEU2 ziBJx8>U6?n)1o3ppz%V?b%$!`+{h$LeK8s%Q@y*yQS#bE79S%!P|7+gCD*;#^vO@1 zuG}up*x%b8w6pfb7z$h`Ef=NatW=21R{hLf$wNEq?IY4BXb*<@puY*Ea znN8{48P$2KR4c2Y}lRbVlc&*anw zzRh>6TUkz(0C0}vBY;l#0JVe}_0R(Di>O4R%AS98%}GSeavK#Sn>!b;S@O~2S}!lH zN(d8L?b2Tb>7&f6^r+Kqk3I`()O;G}`9&p|y0yXa2m;oX4pl7uA}9 zpHZ*)@U7gS)pD=`V3uP-!>Mt~Sedb4(*h^CH+_mAz*^W3hr1}t1jE?@`7$8a*L=De zQSt9I#A_ld3Z8A2Utt<{g>1y8sx|usKQW5QA#B7k^*a(1Qo%N(H+o!z;AoTxU9REO zxp6p$iS|FvhzP2R*_^V^392++xk%8%b`(v8!OMeZh8z97+?J#)236QtjEZE@b*aUeQ* zAx?|#Z|Rk;4Dv7r7A+($ZWtNRf~rH%W&SN$+cy(>_Hjq?c_Er0AAPupRA3`~N04_C z>u_NuS}v+UCyc>B8~F7iQVggZ6RGq8lK8qjK>xn^0I~!JvuGux;l?Vy1LIUYo>ZUb z46{%%RtT5N7@BSu7leH*s*$a!s7a$|GUl`0wMNfv458@Cuh%6lbjQNdmt`#}$3l9E z1|ou-TQ82H?=rZ&UK})8%8?|S^`{5MSF(7!7^A5hB#7ENG>moE zD?CtE{en#)3K#Z*XqAs_w=V@CjmZMhbl4V%)3N-3u&w^n;c}-LgS_4+_Va3V9Gq4{ z#>6#gMl>96Z$hHv4-yEQn-SbTYa7(U7r1jgIB8AFZ^ZQK!{P-iMgEKq5(`hlidaAY zOaVad+k96n*YK_^KB{yAd_Tu z?~@h}7cXxMfI1c9`+sz*)<%*?+gNAakNsS@x>zrkK2I*CsKy#3I~(-NM&)=jzFPbS zOB~G9%{86jf|qRNRXBYx9%0*4W?m^s&9h(N?nW$!|7qY>@0zB~#K@a%q^n= z2x5|(>HG;Of5i%cpJsLm;}bFjI=TvFK)V!hZTj(G{0?{S&ICfUY-z{t1j*>tPJO+Q=dawvv8>w| zy$w_c!EGmF!1$}y;-MjXhEx}SJtOQ-`aPx ztLaov2Ixht>{7Si@nxrAMIs+hy;&1B_9VISeQ~OkP*-uI*&-yyRyHwpG={q9MgBQA`61n(t1OI8ML9C-ga^JV2O@_-DB6XQ+IGBLn$0&lK%}-*$ zBVR1r1@|AUr@mOHoiQMlbMFJuKxmUHVdszu7>GDuwZcNv*j$eYQdLAunc6^^?xc8p z@OLt&xukHUJ8d4e7gwO_LFJBo#vU~^bc%Y7xybD9(};f%EIz;Cca^#jCQrB`!~d?Tq>x4O8jpQ<`vp&9XyV z_->?CTl~8zTbDl#qn(#3ax%`n2u#m;lW+|6*#ex~K209oQ+|8&kR_mT-c~vU&h_>Z z_1x=a&j(?Bv#$V9`>XD5tsn3;XG|?4wC3sv(f!lnq9mrYhMjrsC%y)UXN(lbGgXB# z&*z8)Uf3pFj+SfYQN6WJ$nibn_JqA(uAY2>-*kZbP}TGVTQ?kGyl-aO`IQ;ZQ5K3lpB3Mr4h#$Fs-Cw}D?ND9>^Bg0J!L{y*jTW>R97__ zED_#usHw>_H7Mm3(O3)Rx@9ZOpEjEd=8}h2Tj?K;G}ELhW6dTk<*JPtXhy}-s`gLQ z!dq!NJ^qd4R|4UY7?H67O5reBysP-Q(wc|#bebX6MW1oFJx&#iiX#J?KiBJH;z;hu z#awWN{H=UkDY>L+Ejd655HVcDcqzpAs5cN~^{5tj&tG%k{0Y4UHYLTf6ePS$KGRrq zsCv|XfN&%xNtZ^21%|65;_`%CUMBi>4#(FJ|J)HtH4doBq_UeY#r`n8PmOY2p2dJz ziQD5@9H+5j0HkTkZ-d!LIEpO?G8KwIG2R&uEId&!gBfNP-UHcim+>L~H1Bpnxd3wW zXAscC(%KSJ<2@}cTN5_}3o{AI3b9TP4bQ^p!t^68dzn;c@5YxABJ@~{YC~wF7Z*>H zAJFC|y5W9;WHJm&#kYK3`8o>2Z%+1xcgmLpteori(0EHX{AQl@C5DLKK6A%*=TJhk zU`!yB%lJ_7d^P6sS1TG^v_f@Y<5aZiVC7&8-+m~ENhPIq63s|Vbr?&QGVQ~;JBIcX zTGF`In6xtk-f}#xgY|_}mJ_h06LKN-RFL#la|e}K9>%2!VMvKZh3G1I~{0fF(ne!}AHHf6j3)7O|&`c-;P~D}I_fV?IIgyrUB!qr-l(1a2+*q3Rnsj0M%0 zPEWLVtWDjsyl=G>W=_+;#mh@EdR}eWY6ObspbImTm8N6Y{b^L1jb@h;wUyptJY6*A zocbF=gaqJN^h+qi^sTHnQT`{1&mj(^^|X1tMl-dY9iVf}4CFX;Q=Hx?%PboAoqhHzz&`%69Up%%tG zTYPKKYc_M_d!Rx$S0tAKPGU2YaMT|S$#h}Ffy@@$Fx1HLgXP^D^4rc#B^=jB!nQh+ zLY$aEZosG48I=@>-zviHfH=Pidk-s#bva1Arb)KoU(WH& zT|aiKIrJ;6E8rMmcb?4%JS#RC)JK?% zL%LU$`z{=Vfh-je5`#dKXT@e3W9BJ7NH|uGHU`VqQ?X!SGhM3O7Y~b7gE!S{ykC8# zi>();>=S|khY2}BoRkC{gE1;Tf5gK8)C3&z^77(?T4`MTPLzFvi>#*UG5}Pfy@@{v zi<9rQe>69W;X9m;4K+tG0TVhp^H z4}uqZk;r<2T4Zn)_;3}$Uic0Zbji!5Rc?_Ad|hslFQ5Oe(0%~DRe;0ok>@G)6XAb_ z!PN>a&wr$aew7>ia9w`Gq*?AHdeAzJ@7TCmE*l44;3~bydH_D};sM7)GHKZ1%c_2w zIp)oqZ5e}$HH2cJ3cVm=@MYj0_FTH-jPpe*?1ZTOE%zW)1uG5qK3O$3c%rQ091C?+ za*;-CIeH&^F5q&U3kh^P_6J^t&gA^pU2&9nwa_Vs6BQoE9~cBh-s$Y>_bAhErOF1Jvx)>J4d=W1j@lJenw9X^H`;*}duFxs|1}Ike+H=m zT7eap-S_icH2YpU0;ZBeaTGw>+J(uKJNGmX@lp@yoo0LI3}f!xi~j#GwRNd0pPtYm z`y5fw{7}bhCSzs~xw{ZzKK^Ryzm3FP%g+aN_Elm@&3bQwb#SMntMvjhFF0xMu%t?q zsCGSwKDKbuo|4DUrXXdtOk4j>LAsits`b^xl-GkUb~7XIwI}=sw^|3y=c9q`UYGE6 zzCV1rM(k^+i~E;e;(#rN9!{%po@wz;3tR#~-ErFK^qJti^K{p|VGeK7x;`WABsEQa z!~rS|U(^!AZM0Tu@J(UthR7upK$M(x!Vl{*RpLoaIvIgwn<~LaWb8`i{_KI)_necF zFefNr89W84tJlD+yn51O4rhO=of+M9UD>TB&EB^7-^I!Hel_WwmLHocwj@qoGXVu) z97U$Ri4YE>KXR3Q@Oe6pi2YC2zmt592EjMle~Qrnh^22BK> zNE*T#9Pg#idGeF9<#5>L{eqwzp;WmW<*=fd9ns8S06u6`w^3xJQYO;P`c@{1H}UPN ztH+fs6>5w39CC6L!vLZ)Q0P9Y1RQiq?&>TJ{njbtE9R+aoUk%cQUS7$Uhh(KP{2+_ zo9#!lj;hNo|D@)s04S{G!bFI@!J~v5R6-CDqRO1<&1$4uQn#blqo+>Fbq{EW(*_)e)V=B?`dH0dryXW7SPz-WMd>*0+){X zN$6&vb@R?V8dJ+(%Tq(Da~s(QXLmb=3To;NCl+XC9{jw&DE%O^_USDEjvi@j zz1#M2{;7F{OjkQx+QbRnfObKq$YV3fYnJ{LbZgfV+2UB7zP~-0O-Q){r4J>z-%u|)KP4un{_tb%y3g;X=?NiG)*QKp%NzGdA3b{2E8+4fu zA)M)%{rYkpD;Ba1I{%S`4ZVAcLF!kLUbG%N zF`dbnbAw|@*znS2whd__ZNn~qhKuNfxXhR{$6)|=SX}D9X`B?Q#C=?6WoZay#mccV zI_(ECl}xn^-t^BddaGXq1xE?7>GK<+zKY1^^#7`&8Yftx{g-z0i_y^1zTN+J;&^FyyIo}X5C0QaV@LPUzJBWo~`9BY*`Y}u~dWrqCc7$perqLDInLk zQTbYo6Dp_jH@x70Q2~frdt&w)Xv{#%%b2Jcm7lym=W3ipvAa!3+|Br7A2%$_z8jlg zU!t#diV;cVi2aJreoVBdcxXIJE%nPE8G!lgHa5*)Hdo!;$H38xbicpjYhHGn`RPU9 zcmpPfT+3yo>OweohTKMEq&Uu<gxflbsv^==2DezN2jH$3KD-RxAr9Nkv7IAKnS9#IM{XFMPSky6Id#7f0 zuBU(TV;2T@P9iG&6LPue-lY<_7?!zUA+B(o1kS6X;HOKgWocc-V3@KS8dR}cHEJ~h zE4uo4>#x}j!GzU}v963#>CI`A?5gJib&gE&Pl_=~B~_P)N2`gFnZ|I=RFE;zX3AIF znmMTj3`tteO@K&@0=guvZYTkudkwsRiMBd)HWXFv1^}^{S0(*^1=YCOjjvRC1~Z>C&y8fAKYKnU zSo3FRik#cTj&9Bwzk!=%4_XM2z76~Aky3sitpbRbVX{-ISK-g3WAQR#%@FrC{%HqDbbRN@0eosEAQ+v}1jD z(>jQh*zBBStyY{m-LcFfcsoz^WZrnQ-6E@J-67adU@s*MHBtE*sO0T^728q|KCsF3jYAPB!pLiO~bxL)Yb#To|oO#?TS0* ze)tXL3t{2lz`y%>F}c!l}n&8R+`lBkp!q?>L2HS=&$$vljr>uf+w`$osXvX+m-*F(8&E@2R&p;OQ1h$=ES^4h$#MdlG8H! zzH@Bz4DjzQ{cS0QxyHjO1#f@(z|m2%^}u`F>(wZ@(|Ytau}0A^7GCEPBPoFiMBYVV zmc;!6!ZB|m5sq2*W6t)b>?AgFWo`#WuiAp0sT99O5QESLaAyLQBUZIQwS2 z4(*!p?sBfm%SK&BU}9FrlJ!UcTQ{busAyYk9ttK3XNuJGv??X)A$R>7E6K#GIGP2MuTk-4vF(UA0(? zt$S^pXXV|9_)171G6CHro;8&!v~Y2B^RffL6YW`$R+wun^kEIj^>sl*hJ< zl!6M^Sv_0v6znX}^-RGO5LU%`Q=LBr&U;!4B~tmXL7YQJ98f@|jU@})uaoW=9{gpr z_GRB?EsiIm=rOL&FC=nOUByESQ9mk%%sDzvt0OvnXUaO{nYZlVf6=ifsCVsekYcC=z z%7~h?@+ds`U>Ea`Jm(Q{86UT}@@UT0f4Ew8gX0+|VSnoR{J_n+q{!-J!TtWloqYrA zx6FgUlwLqu@uh};d#$u7=``#CA^!To~=Ukm!f^4)`A#Dy-S z#d8p^r$`_o`t%zRlPKNLMG6}i;tm-~vhT|wiS!%Ed&)YTTQY_{4;x2hPp6>pTKbPt zOVoLek6yLsrS>O+t5a}AVXQXgakCa5c0G2G@SFYXZ|ZABd`+H=ekXof>$LbyKfb|P zbMf*kz8mk;PA-OI@W)sBwZ|#F-{_`g%Pp7B#-W1u4nnd=#;1e^jzK|A5z+4iCoSoF z`UiqGk)oVeUFzMM5Z>-MJsZc?UJVfc-jC+)zGCJw&+gth&SJw~oq`K|fFb$rqdgm+ zB240)F+3;M+WX|N;XeuMv)tyGK*rk!?{wKTmU{mgR2e$e_Zoa4iWvdpEL5%QENwQV zU%U{Qu##FTVHkCI*(Xh4)0a30%ee;5Qj9nt?x#$e(9zxcQqi?Ztsni(i8t}h8+_Po zSbld$P_*}juR)h*3}*YYbZXfA;E|TNe^`Jqt2s;1V?N|$vRd@L(eHBhM|_Fp9^;id z@k_s=2WyWa6FCA$h1c&wEl85STf-;bC0ye%g>^(T%9>wuL$6eB@dkzKX;+Ah|NJ(x z(K|-9$hyD7S&U-K=(X4}9D4M7nLcTCNroc?x{>hB+~w*8A2~cxJ_^6t;|CrGIA(o^7DQKZ%*RognqYX_rA&j z{BPuwo$ZM>f0+~g5a|>DRjub5zV-ktmp^xDTE@rB67^4Y&@#261nnS*=nG<^K8hS8DUOa?z6N=#_QCS9RK;A5IsGpN)8ahVqX; zbAK}|T@3KeGeEk{xp3s6{wJ`&`~AI*$nX7gt+3NK^JzWraq+Yz8Xq197Z16={qDJ1 zOpEVj=QG#adI}$VMOI#a==m~g-OzKUcD7FU{zmViJ3Q_S2AQER%Kfx^_#5|w>!aYG z;pDsi-}A33vCt{J2p_xhYkRg3jQM*TTu>=G;CYy*WfQN%%)zxHqAWwhBs${NxO0aP zUCr5!UZI3v_pfrbZfHx`CH)b~zLqaP;62vI73-m>dSYG-Sbi5&07}!T|rUG zwa5Z__AgDo4^`H3;qK3@{e%=G?|+Wc`mFhLw@0$Nf29{Y7Ia6}J=>;G3Ob;X#x~_W z`G>^uyLj{eBkmOY|IfHnyqsTo|L<|9)_qswSo!UCpB|!o$bWQ7+@XB?8ZhSTnd^vj zh48L3R&4%F%RdT2l_(U}M6oj#7zfGrFHz0C$!6MjQi*c711JF?Z8s_Y4GkIon|jt=`D3sA%-U@n(>#Zg&h+&LfqHe>kUtIA z7(rf}onQDH^v4{syaBtJ_YYFCtI{6gEz~D#k(MaK$bU^dBqkmy(Blod5*zGa2Gz~U zIN$oI`*CsU{%IR?JUgcO+i)dRpZ+A0Rjl2Rab6iYO))O92zLwmkql#3FrKX(dl!;c z{1@ce>@{8G;Y)^=6k0QwiEEyOSkqC<@NTVzz?$VlyCmueHncM2q_(ZumZ*Muufh(cGR8O7J^acPldNPufHKxq$v@HAJ8YH>QX{0z zD|i}pVgvn}GN1_?k7}I6$-;epqWkC96*u@gL1c}NR=5W{^n|{rO%y9Hukm8Z3;zIG z(YUI0F`lZ;nWqytHlkl{o^PtWAT^~UKxaq^V;MjML-~);82nS;G~@>XhNUlC80D+5 zRSmkurfg(2LbO^kSd<9gqV$4Dx=U=A_U@^H5V4zn}$=08#)QKAm<&(Ot}Q^dL~ zr&nUUmNgk+1CW#L&EI|I-+sc89f=^fWeGz?^T++Hlm4_4x0BM34Akv#b`VVqW2MHX zi>!*Z0|sJpQ=w8`vS(_;rxH_Jtk3QD5e3QLy;)8ag-8;UkrSqQ8kn#?=i8ZT7%L-w zG@^rx3#iP_Uw`~{Y|bnF3}a05k&T}mj=C1OhqwJepr;>1Wi)A zC|7n7FQP#L^1n$ z8P7I?$EvL;=R*@Uh-_EEw!C&DLD1UN@$uTk9hwpfiqAW4aVafHb(g{}O3rGFLo8>s zlHBOn^AZ5s?EeYVbF~hCZ^kIUK?+PD{v~V~Mm_T4F`%E$%T3?JzmlY8%MS~df6khW zyM@eVRaglg95$-iwgZ1hK))%4KYQ!G*0kiAnzejlsY{cg*}%z~=4d}LVW{?1X*I*SiCZNDro!#^OhBCaf8 zwfm59#wanIQ5kV*gn~H?7DTQboXmEyqB#bH{HX8AOKvg)Fbae<^z@@8be?{s6(0aZ zP?6iZS}O%&4GgLo%2VQ+CFy#Xe8i@iaHT>^lwlc6b&cAIk%kTIxkh?~!IL}KKx+BK zSzC(bGjbLAZTab_u%T9mpMm@tO`R!Js9f#8p{$>l!Jld6I1ni1!f9y1ndC{cJS{l7 zHEdGy+#>MUUO$@O{oU!`lf=*)jU1M)kY8I*^&p@DX2js#=qJ2k{FK0tO8sVAHoK5H ziwn`Ib%IAGuxK)I0w|?}h8IFPq6CB#Ps3+E{?Hs?O$8%pP#C<{&T-^ z^D%P?bf zRjpfp_X$F=@w{p6?XZRN80YBj54s!zD&r8xGSYhDkhjXlGcHPk$(TcI<4CpLKXvOE ziCU=z8!678xO$w(lS->r@Q4l?M2yfZiEUj(KcW;*d2;bq***A0gmINy{}=eH zdvT9ZvFw7D3+!d-)k2^xtz%~A{=x$IHR>QWN*epD@22Vc`8fCSFqQfJtXSq5p-(eE zU(r(ek6pWXpUY1Vf#Y$c14$2SBWN=S_h^diAxt zZXzA8WDB}ORN6d8i;gAamL}RgA9`v1j(I(>aTunsnhj~G7MNqUotvs3D64K|gHoGr z^`$#bu7+l7VqZDzDuxoOg_I;`Pd>7hVDcv*Y}I7zqqw=>pV(Zb*FpFlBF)e9Gq!+t zL#n@HI|0z>7~lIE8m5_1#Up)SLiCzd{Ip)I`BRay(rx+ew?;v)1|`~)VpDrPN9t(4 z+f_wf$I2Z-OOvfpVQBQ6Eu*f?q`hEAv}AIVtBnhYM!YG;;QgbDr#LWGI({AuQhq3fh@^qcf)P0O#*`UPlf#Kr6z*}m>&Vmq?lc(bd6*Z@m zY>$WOWMi~^dRza`KUzu`{Icpol&2)})TKY| z-|Gn=G#pF3-zUiMZ6$v3d0ehQG*RZXgMADLn+!&fbZ&da(MC(BA(PaeKdl&N&Mp%W z)!EDhb2FiAp(Xryj5AjV-vO3u)zMf@jLR}*7|Nq|WTwtvURJ8*Cx=l}sWk#@fg?OC zp0TPLqUF?BEBbq)-e)Faz{MnkRoj$t$4|{&1(r*xYg3mGOCx&Ut~p{RNJuxRST3QN zQ~ZKyznjD|v?PsdH(i%xvEmam0vU{stoCPHFm=6flN4B7;~{!Z4e^p)#h6}5w&iO4 zPG;092NB&vZ`>7y^}Ui4#S6-bi+Kzjy>kNfBu85K0u}O`1Xm4YHmQY_3{I}$zc(Ah zFJdP7lXY52+ykDf78;PRD8M#W4olEWU`E2df|YiDkm#kWW1XDegTa~25Yw#?6I9R40^FGCS7$My*TVDERSafm@tOzp zA-V7VBp0Uh<*fNiH^}QAXy$e~&O5wg|K7s%l(7)9Crhw=7J@jHaK4H6Xt#D+|7+i` zE{I{*j`?DFwh#MOsKmDOGLZD;$den{;&Z#LYZkRalRWe10aQTx0f5R7Mc`3-L^rOS z1P7C@qYreN&-&ee&u8kBTnKpzR;&`kGBvzqt8Fmtu8XnG*>DBAs5z`lRt-dgaR@Ng zn@sv$;bzv>aV@$yo=N7%`Z{6QZ&>Os*NS+IX7j2%eB5c!;3c_40$pGT?~pPY8|c;x z<(u9j?em`bY|oss&mXUp-^FC4Dw$F35M7LGc12Nrxr_-QRX=(CeVQ20Kc!VZG-YX) zU9Vy(x7XVAHw1bKc}n>i!Z1SO?u252J4d7`g0WuIo_OWaM4#NzDlb{4VU%1{lPVh0 zBKebTw+)RVY>$B@lEyL=uB?&VLhmKLF%MCIx#e%ov_))jp5yy^!ujb3oy(Hpo|8Cx z&&n3H_)vSSCSLTc1p*?H-CFlZ0RwN%Rf`pY+!dA8ST}w5TbiF? zbq-7ODFBPaST*XQU3H$cNavs59o4&k_jWeU{~4wU|KTatWq&uQE_iONa5JuQ`ZNd+ z-4IiGPV&$-<2;bEW+5WwaL1N{Q{kd(OwLRq9t+(z#wq17xU6b&jM3qEORJ_*yM|f6 z4i4UbhAcdq5u0UKII$Gr)+Z)7R*d4GB?wW~w3{gzg-hbtpH%hcynoA}?-ejK(`D6|99w>>@dKnLk#=ln$V)}zh&RJ7GB)qTD+JMCvFw}HQAcr%VInJj-n z$-ItPf-V=|n|{YnK`@%#; z-kPweJo1s%Fav%2>rAQ;MSQ6VDIFr{OatgBkXM!~kA!o{gSqktb$7;b5{tq2F94}~ zIyDM`6pMij>2v6$Pu`fmMfOgix0+`2*)>_MOdKvvoCq7Bf<)gGt5er)sF^~O>0^P~ zBj%xklFyB@&thEbUG52gF8C@!GmvKI=bhVH_<^Es{(AU9i`qlkcTRVA?%TrKB*K-3 zeQU-2LufT0sSEfs-tO>4z6f`P9>t?-AZj;%E5)<)t_GsQ{GkB6DUXYc7f;l_ zL`^D)YfkLyRJyUR=;4RP?R*)$T^OLmPVRRB@NT4gyMa@cS@HNzdB~SxD<;?HRzg>$ zBIP!Hd!eE6`RT!b$G2LLjg15qT4tM)8`&tm$b0h`9=3YmQ_<~&qTX0CDVcL>g0*<(1 zsTnAw%e7F^*YQe~ZeqKB_98CX(52-cMBMv_Y3H;(eS4FOMT`l`SiK|;IDZlF^ z6SHpHrpTe(+UC=Hb9}7r0>)|pN|=ZGmkpq$0AjTw2qZ=K~gbqUbI_Z*ruJ4)Wy5+%81J zkyzb$&y@x&y)7F-h4^)9Tjn#ZiC7g>*Bys*cm)~=TDj7H=UT^FewS^Ms90zmUr7-? z+M>1=o-AJI5+~(ZY+K~A`Z$lir^j9t37u+Ngug?J5)8d;D<@9z$-$40Vi}2as_5Qy zVd96!$g44w8P*YnWcqp|jdY}EOLxwNHlQH%&0P0Vvgnaw2f0nA!ZFSY@z#fOR-zAO ziQ})bRN*Z<>R~P41-z$fPTK!~gl+*HfV0hc9P+5rP@fl=t)rxt!Wiz8#C< z)`~%V_PmEn>bVsqy|xfc4MUSA@CL;SdQoXztZ1YQ{LGKx z8a?N4oEQ-9%?*7w4U8A8wmrn7VFUq6bc;7CkC91##j0qk_YR7g+Uk9BHZMz~Pi4@@ za4A&xwtF{zlkRo9NC#+Oy3e2hckH|~oZfeqwQp(19Y`THngX9KS7yf@0!+?I6ShB3 z8AxO(QBLiklvd|b^FXOjmL~z;&A;Q__ShR=APQBscoB?4>h=TLRvCxWp4DSUR@rfm zB>{p{vN|69->W%`qCy>Jr3pP#Tcn+5rPrl%i>V{2*CO#-T1S0C^$XP)3O5c3Z#h#Z zt0up;mCvL%ZC;}QNm~4jX2bdtGB}$Pm#yRqcAi79Q_JDm0$vuV3wkp>UtJX7E@*Ls z?Im1dDx6#TEn`T{SCG*t73I|@2wH&JrxfIbxYQg1Rs3X?Xf8BVV+p2cV}m!X6!GJ} zsjP<>huYX)r%HC*-t{&z(t?0x*D(TcgM)Ra0cwCRAXzu1S0lWA(<>f~7_%9>S-8`0PdDZzNEJGmgse^PMvvU8e!c&>P`w0njO2yE7FYK4>zj*9^tZo~hQA}2&!^kym@yx==DN=8> z8{Vy|lx;ID2ica->r;RgjRYiaR7&a7aRMcXxvO!rk2o zu7TjPg1b8j8r(HF1b26Lm%JhQ_SyTKckg}o{n0biJ>AvSCBLd!GizqOp2?Von!-SP z3o4EHu9Vqq57Iu1I-IHbel?NUgs@tThtJ~2QA9&C6xVg@25g*{m{JH0`~;qNWOu03 z%2{W-obamm+Q@a((nbs7&9Cwj~U@+k8W zvpTtVcUpHU)!M5<`Pf8Oi(x-;SWxOU=v^c#HmMD?5sZs|$}Wx{8LY>w408@D(MOi1Wfpj?M@<)nt=- zVha^1sYl{QRsO(@uqFq~vebs>J2ftPPZlL7FTp*}cZ6S~@AE$hY2t!w%Uxvii%&;< zNp?!3nyD#DA}>_Uv>$!IEK^;z5MJJ*xyoRkOyW47v#IH+0qnBgL7X%;Ww1w~+3fvP zUlV^8Ke~;CDIbnz(^YdGt70SEg2h#rzfxkl-WduLltlhQhM2F=9_H7kOPUdotQnYW_d=hQ5>Dd3wfH`qmH#|+ZkjwEw<^Kqg9*2dQlA&i4w7ipY=QK&$XD1DeA8i3nsH1}= zzotdR`u*rARbmre+(M`KD{Q*gd#~<q13`^01^$Tfl^AzBt^#IU|C zmDw(fa0W9fXpW0FbJjTP7i}2=HSq8%wJkircOUk%?*Gb|04l%=%=Gwo(w4C#!ZQ7o5(aLnVNojl{vh+_YXpT?FlUHlbwdkA;C(WZSWzJ@ z=72_-DkvMt_ar2W=e(9pl*8Q;s9WPjIQgFy`)T?dGGc^$V~D3>zbTiC`&g{RuUkon zA9#Z;tp@9rsgItd_+EWXU8JFepsa~a?m*JQcpMqN=z$ASV&mR`mQti^C2-ED7t%T( zj>un99*kC_>+&No!&Cbl3n^NjrRa zTw{DlH}VNP)XbgLOnE5h+JRugy>p-j{iltL7Kkfa6)au2a4f%}RO#pDBxv^LuJ6g% zwbsR<(Pr*P1=&4k*a|iWnt{G{$tfOJln>-5q5e~7Jzce0wHekyFLZs(c`~rtx)8f9 zmH_0WKnviZtyh-b1{7L<$@91mL%N!l4UT zq1msOqT0frNGNWvvsT%JZ~ewGuRfe_%Nh8d$rEyH3)bX8#Y}6f>03|jYkX)d4Tj2j zWm8NW_3x=~kavp??fdh7O3eDlDkHw2Og+{d5>k}a;2nt4#@!h6Ul zn5mV{BVK^E1Yy@(lsWnQxP)#Sx@tz+#7&LF6~wK3g2>nXDMM^Z@N)G2%^D(^xA@{o z&g}K>Llbxunei<{KbNDIxpX3IvFn~{Z`SrkG2PWQ`+BdA zAnV9(1OK?jDz*Z(eEo7V7nL{759PigGd6KwP9 z7iO2xB_~f<>{3I=KX;a|5-`b%E2RfyG%ph?rz1*@@YW>*Kc=EM{raI4%EIyBphoQ= zCSS{(nU9c7mrUFd3}r?oY%ZPC7ejnZ@YrL`8>Ddnu1Zx_ zHxEmGbk@r@aSoD^SaMNMo;BJ!a^~)=vz5>}1D|tbP_}T4Yep*Wc8@UphLd=2E=hk> zs6_Ir2-pbWL8x>GeA`clg!jDEw%`6Pa@-dlZWpiktIx$<`?YcNyD^YwnU!XlU@aGY znN>k76DYpTU76_upX$dbpYGHB4Gttf6LD&V+s@;GO!8 z8NlIIqJa!=ciN6QF!8#+1DaB!)^N~uaXF#Ojc#zZMP{=3@=jpS_xCgD70%f>IYk7= zkFJwIkdjE|s-Dj&^6IM`r)v2M{*x0Ey@iLRm5GrPVzdp^!-n*7k(CFFw3E#_QOcM($5~!fCAj`)KzsFBq|dhB;G!N$8RkH5e>`}$yPF_$RLsYhj??f^ z^L0P5Z!*3DMIpYZqg56>BUltHz6P0~zwuthbD49++vgeCXQcBzyR50qZot{y_{d2q zNn|FIgDMpX8y?a)rx+*uSljweHBM_MX?TiyrU*|P?vd~30>bzq<2oF@BC5rH6Gjt{P%;ZGY3SZWUb;O~*A`4*WJ_kIu{?_^8zPq!c13Vj(U;GAiH_ z=oXNS)M0>d{VsXu?w+jjN3$=YSrl^HT_C1vU;KFCLim(fo6lqNY3~)Xw|n#Nz*occ z)@V5m2heh-Ek6B~zITsXQl~na8#h8e5hM5D#S5RO&aNQEjM?`osiQ-8Y`-5j;%ncf z+qW;$i`|(=?EvtUPtMBl7m{os>Rnn59fbNnBsi`J z0l$r{u(ME08)&^MruU$pR1Md=%q|d~S&*OM8EYORnOUGCVKwDK~vA#_K1jd-pj)pP&8_XBeY0*6WZYts+5+ zx6du6n5s?Fe z;vXkPTAFlMd@H}Kk+hjzU2kwEPHuYfN^XLmdCf*@Ns8%;78X+N#)zpu?v+l4n3wys ze!AdLCoGFCM2*T4^CEYRu*Np4JcGTufVGXfpyp$FCzX;Pb$tB3KBYS46-apc73r3; zp_^NhtK<4aeU_I-hSjCCw?_7`+@;qXj#kJsx!Bh5X3CboP(l&Z89LN-N^$QD zqfK!B`gwCt`RqITK@uZB!=f0fbMWBB0mgd;48s^|qP@!SeE9RuccP8#X@#6fs z?OjXO#l<&-1svrNcgZ+dFPnk)YoG~DrHn3~-CVBh1dB}IHZx@ z=c0+!xM1{hy~n#I7BqzHu;1|0m zpkW`0rrO|-c$ZGj8jsSC_Wo;oV&ch(=5{h`c5?E*pqI%NI@9HXu&qg>S<;X*sE#Aw zm^lQ~9po++DeKt7)a>>#Qb5*_v!HPJOkOYBg0THt42guz2JUh)^5HIe@8NN*B8Y-Z z3`ZQ5)ea5+NaZ2AFJLhtX|afTd(H*8QMEBl5S6)_wa8w^WT3?y>abuM(scDEm2#im)T{m9i4Qw&|q??4-I_I3ZjyAPM-T&xBE*C zd{+wE(Dv5yL|8}b+yv+1nRljUZLbe=OmAm}#*X)Cz4o%r&Eg&+#c+B~R&mQEJ}Dk~ zImL#*zjtd584|e)mulG$3STFVA6Gtl2pq`VT3 z0<<_bb2S8W+rFF=cW`v9sTf`Y?O<)uf1$98dsj%Q`nfd_lvET5nx)*t>Y?7Kvkwc6 z(eUr18Y0G#Tg&nTC7o+>1}?pEL?Q@G>lA`4tWfNuvc_^mf`?7(OyILea>B=vV+y=i z5Z~Iy>M8kMZyz?7Tat~MW>}j~-3C>dWW(1Qz5|dhNo=r30u>?uAwY;_|A(H~yFz<~ zcB-MCN_(Uxx=k#0OMozRTBj&oS|@PsWv>6x^<`=&8``u^Z8KzmxI1tl1ke-*|6z{t zg}K-NOI`tTUy&Dfu&&Sw+#SLQwtiB{Om_ZR31*Ew(o6Ygfs9+g@gHW8xB>&90O{JitR@!3iMxtLZn^fq>-x`B zFNFTb%%Q^nZDwGa7rrd%qW>`aq895p=t4#n z!@t}XV;~nY2G_ipN2q-tl}wMupS2s5Jatmp#`%mg0NQnUBWEBD#TCdk2lU1ESi{<0 zSO3R&*nky5${FMUJ^n4vp2)EC6jH~qVsvI=zcfQBT^$5iIfJ@DfKB2u$7Ns#@^mkR!oTwmakQ&=vZss0*P z@D=@FAZW4Zt#r=2!X!$A5n%NICrpYlbZ>$BUc^^W(r7)f0A0Y87XB^T3%r&e;83M2 zMnBN3(nmRi8-z4N{}*As+G4c*!)RyPVIUyjOW?imGD3h=+9GrPah5*Ex_{gNu>kuyfMmabafelI$p|jcdqY@`sW%@<0HQF7 zk3TPU!*+58U}CPGxG%g7LUIQDq=40b`QtTzR0p`pe>m)B=QU?fWuSQy8x4IZH`d}CSK#I_QDhOu3ZC$JZC|qVFAsFOtYgm|MlfExo z5CmZSfFp0S(5H3A&HSS*paJmx9xpL4py^&J@_`MkZQl2VNm;>F=^TLh^z;68#O+nU ztZD&+{No2VA!(h;yMX8Oy)R_UZ~UkAAD{h~UxX2m-E_25*+$TlL8KW*NgWscWdM`x zW(j~`3T@7w$QZj6kz5MUco-M`oe~$oFA5!iw`3Y9j!0T1 zeN^O9RWzGe1s?#b4$$N5A#850j0BiT3j_@jlT87}QnKGkRq)mr3>a-TeVQ=6btY)j zIO%gB$oL|ksSj{x?kPdQ%CB2qiT2N20ZXN9P!s#Rw*Q74|B3@~0KFE|-iz#CV77r1fC&Hw zIst21(+6VEKk*5`#hysV4J?lR;!oGZz&cyB-vHx*AVACq2*ZO%9MiM<#QW_)IB1d5 zIm$oH&@wSyeytywt4FjSg9->VhyY?!O$PoKaQ+8eI~qKs<#~;_PA-*^`?j!x>@#3T(zP!Tdgz5@cL`Y0 z902~2VU9sx0rhW%Qvu@a(ge&|)Lq#BZ$$SLl`}vBVE#{p6PA02Q#(qf`HWInuSgFR zTmF|dGAX)vBZ<1-VVPq~FD3mA;?|NcSo+sTtc#%vD}(^10AIn}+5#o{ z194z(qmycA>?H;r63Ui~1B(I9yx>m3I)I%2cc}n?k0x;d>P0Gzp#bi1{$iqYu9B7F zdu!34p;}{FIIch80))Yw@)4o3M&A$ zGkIQ~Q^^22c@aO59&KT5!U%NyqqZI@h^hgE zHnHTuK#tKP^j{(1um@0?1u|z~01zjFlm9T4L>d1PfX19jdR!oE<$+%aPynTgA2vss z6^H_YTwEzp&3*v7G;ScJb+U8zQKb=o263{$q;*E)*aNpM+NmHucyj*&?}iOgOQ>;FsnE}$3X zD2cq7_NMuKYzu(;Er6E*FJ=CdX*yj3zC-p;uKRyd{(mOGw#sp>1VEfmzy!qXb@Ud< z5w>2e`fvILq+37=P+|y~W_Z233G(}pMkxCVK#@2A5D^`ZW9DBx?|dR))^`^|S ze3^z(7&nx&S`R_~9?C*nK_2zCJq3b#9%vqsq|Q3NnG|5iuQZAt^}=_#pANSc3nwGt zDIMY~zyQa%N5`xvMe#;;R0IEpzGiO)=*Q>a<-x2TX{o zezV&bTbt*)#%vl_gRw*0#?W&Voyz?#`XJj{vhcXv>{Ln$6Bv(A=9ty_GPt;nhAIO( zC`;DT9&W6k3O>9MF#{8$E{RG8_y>&;^Wisgd5r#=I~NvF`ySS7zB|;fi;|vZ4Kit3 z1A)$T-~v(iTk=kp+lI5gO05*EtxZ@7d~Cr0X2mEWrYS2X%Ino=uHo8Atyg|rvT?i)AGeK>oDJj(-ZIuA zQ&N;K`G`dvK@Ic^{g@&)YO0+#vYt`(j5wN)jjYrW*L}U9a$Ys83BEF|JZ+5xJECap zoz*pfH~M#*ok$2vTXmM7N?nghqeoWCEcd6)7JF?~;MIZc(ykt`)lmet)`LzYO)3`p z4kX+4Z*G`+w;I4pCE^wJ;H5_Ww)$5fjPM8Gs`uC^hxf)n%GrSg6X*a4(_1EPgJ0%K zb1|6D?Gwolx|fC>ejvL$?!xL`=O-U>^cI7Qufrc0DLfrWYMXy~ zOSpTFcrUKpH`RliZrZ{h+}y6^>%k9NZT;RdOZQ(MR-O9IoI(H!n`9q8k^D-)9Pz77 zvk&5<|3u>FSfz60kPUcLwZSSs1X(&t{VPGqI=Ex#4{N{?mxGp;?k;~89fcP5ypj-j zkDstu?q&@Edl5b>je4tgJqs9#SqB>zt6>NxUV^VCJ2~?Qnx2AG#fx_NttDU1e_~$c zPcwtLfu!7 ziK7J{gXy>FSOPt%JC(dMuY+F_3#pnI7feP+zpo=j=U^K?#^~WgSvLr5^F(-N^f%Cs zl3SdGb4XFIuH(c~=ALh1etf>ha39D_qd0F(?2c9WwN)g_;_*EBK0BBN#C*~vGFnS) zn_0kj6M=7j06es6);6GcPj=p7W5<^ZH?Jg)I23c8p|}A&9cniV&QQ!XwLNP|8>k+T zz8mapB+=xU$IRc?i>~@%vI6|`Vx0*7!P8ZH(-WF2&9?f;?88oKR7%rW2BkA$V~fYY z1h)V$jud%C zneRv}kQ9>>23;=Yq!?hPD)(dl>xd@hR0b3PDZ*^JZenT{azuH6^YaB)ss>5J8JHNZ9vn zcy~)1Yy-i%69<2jT8~)>cYKCGW7hLoDka>B?-XubU0ejMXB-^q;~D0|is90<`eSpb z=1aM5!`SlJWWi!=`=CS8v}q)O_ghpFG(vxe%hiFIzbL$d22RrlZr7dxUySWTRORz| zigu)ZrF2IR$4l8=oKaiN6G%yl#h6b=L;*rr&iT%AW6zX#Xrs4dj8AKn=6t^MApNfS z0fPx8-|l=aNq^F5#a4g-Gs5L~;mFL$e{972VJ9;Zv3Kiv+wi{sx2il`{o?fFG(p$6 ztY{5=C<8eVi~khi9PS-F)6D0~urArcFRe{Z863EtT@VGPF2~9E7ea49PPs9SkZ1)3 znp*@fl{lgGz6JI&5&+2Kx})oCv<3WXGcZDH9iH0vm_{Q{Vfb|R=6_y6i)$0QEsWT2 zN5<4?d#>6L%)AmTa%>udw%fRetiDor_0<-z|Jf5=$5=+v&caW$_{HG~LO~v$1LuT^ zqDg#Nm?)JDbjf%Alc&VlKLgFSlZjn=n3R2!jdL&$ml zW@46hxE3p(%;@bGGI!eGrS%5rN|LW#w#|&HTQTpqDzrAO_@%*NIOi5f1cR`%j+Q{@^Q2{6VNsmqB|7(#fQ&5xU!O$zf`z&&b0;QXSzYya5~chduom2dof35 zn>0FnqN{kJ4*6E$oi2O{eW>m=I;qVvBWjHv%JcCsdV4YJN*rr}j5Q4_yZblk z3AM&L0zb#-fP6KvPw zfm=gt$i>HG(0biQUsxABNcdY>{(`=KY$NldWh8xD{czGFVf1Eucic%*^WCQ><(v0Z zm}Cv)BpbRPdg46Zr?(lMXX;W7r3GM3@~S=Wt8;_St!5lt8BQ!%NZX9|Fv&7AaowBD zFGc&SlJZh}Kju>GujON}<&0ro>5O^i+)Gi8TQN`gz8*-`=VSR@Fz=JNo*=n(5&-Eb zT8)XoXR}?0OT3KNM^@_+#4zZ|C8&SgO# z)(K|ZM6^#l{k3;~bqC)tC|DGUaUZyiOeM`@ep5;hS4Z>TDjuOy~`aEyQH#`<7y>+%jIUM6mRlyID1gsk?1hr|O){Ord5yCW85 zL;ei}i_Gkp(3tA7;ObF@eo;OZ(5P9Fag=mS8}YbLQdc>JCL!?e@`I*zIKJCshGJs( zwy42FL>hZ-Q~JLZnUDL?HwX1Y+Dd%qK17Mh(k0@{sYQ_s|w{?Q_#;*xTC*nn+IUFQWdL`GH-(8?!BU z1o%L`88;8!gENH*Jp4&{T$8_{Mxg=rMtMPfwpZ$vJ5Eztg1ij0gGO1^oJ0lds<8Y= zxd5{@m0mZBIL|j@g%E*@#I>rX1SOH<7~G zJ9oY%M2C`=$(cHrdYE~)R$n|=?Vw|7Nab`d9 zsqsI=dstmIN$fqv<7v4$owBt#YS&nIrdBr}(Dc43^U&P!%;i6ecU#g(j~;N<>z}E~ zI;H3I@!j$X_I7i%yK{46YeK4-b{cJMt$kc_fMvC5h4^HlnZrN9+U(wEq1gg|2$CID zGjWq0#If`$+0)pbw(`hXHfLhDnXc4Hp%=1fqDE^>k{oQ-*nZs-cy?#U7HeQvE;!UC zIaS5^9tJ@bTQGI6Sz@#A6CdiMG?D^=9;r4{%F21o%^$tIk#)nb$-M4 zY!?P!fZ_=un|xYr-A?q_@!8zKGk<`o@|996-B>ND_3hrMfMqVb6moaKnv-Q}d{gxD zo5!?Ccj?HC)pp`U`IED!N29w(pA5ZIr|~_U4pOk^wR-=TJjEKKpYWOE^*a~sJFLfd zm$L(LSl73dCz-w{J<0Clq#7BgdL6f3u{xi%NHX)qSvHP1ev$5cz<=DI$ew;E0FPEc z267K~{DwYDzo+_NyyJuU|MDFlESyYC|M`xODc~I+n|+bj-h3Z*2xhCXC?wm#u+zGp zWi7<*uNY4u1$g1C+XO0%C8#p|;~am*N2bFamRVAxp}^fmcjymgVm{u1@t)1*#&_fe ze1|e`cYJ+r7k{6Z1JB{y2k%{!|9%$ibS}%hP#)ZAdwx7id`?A_ylI2jv-5epy*l}P zc+uPDH8ffPdXU+>=*<*(USS>neU%9u_} z8;r$r3%;IPn-83@$r}8w-}lwle>+h%(jh42bIG5Om0rc`NX_UvJ@Var-`e7hY%EDS zLxlTBmh$P|ad!3iqVnNiSY+RdO+;8M;5Zi4#!++&;H78@9A#7T%wxt``|l+tm_yNN zEM#bCT?2chIvmUIDocMo;@jH?+OSI&v(r>|vP-K-1=*TO2mQ3SrwJlIXt1j=>5eJa zES-(IjM*vHz%SJ;I8tP>mCNLEha<0Z5A9|h?K7T5|D3elkPKUzsu~)r^NH^!ZaBw^ zd`B5o_C7KPQb*A#-Rc6CcIQ56y78>k)(zp_b(y^~GW;0%vOIh2k%|)chq~*J0uwe? z4}zT(l*l#t8gJ~DH_nw#H`dH$^l7Tvz_sdVyTo8l+)s3{Dj3&E|N z4ZkFjOfC|eQcw(5&&E;y5F>1}07WgVXddB@BpfNq@F}@Ye5Aroj4uK;wIVv`X~*2^NItj4BtGh^|6^Ja$p;7g24u3 zDLMaRxdbNO8r*lIkf9h_h~>sjF=j+_v#uav2Q7AKWmc9ku5nRQnFBNyD7mr)(o(6G zWH!hh^I^^ED?fIHGn{mrL8GzhN!kUt9}Jq+HL?7VAJ+J6?pJWmPJTYY(6JnKK>2!p zXy>y~R4Sb>Y1bf$2qihK)F!=NR>s}mu@_X9x))tQEQXhab2 zlT+t!O(h|sadXaj2soKUO>Ga16T5>x!buzAR`~O=BXq#<#gH*qeeqnerl5md@MxiQ zr*=`aMlta&BFbtx;<7-(ynzU*ryKts2}J@IncVtX7`LxL$MehU6p3#KD6ex4D{(Vt zxIQz@L(}n>4f?U5L^~K=oJ)8X^~}UIv>!%{O&dYI&5lk$jhTd&3@*d6*B^UMutjVS zddJ)hgUyIz^OY5T->}y>5A~{BSrY->i3a+;(}&Yfknj1kxtXA6yr3$g37~E&$S(a^ zDJLp(j$x_ZcR?voToC&qaz>5COVU6d+2oEeiE3^O?rfH@1tNC(t=W9^#~=uqgA)yt zH%TFpy|?vnXNt&?xJ2va40}IDNPVHN%k8EHp&;`UDaCK_-27+aB$`hOyr_pvmQ$@{ zgOceg1f`{wNtl-(6nF2ZLT5Rl^jHd$q|3weYjuu@1Q=l@!JR|&c>PR>j$F2ujmsn? zef^{6-{cXLQ?>l$n1N#yw#b}szg=&lC=$IUlZ9E5l>((E=*Md2I1>^s_-@|9!FZ zO55?#F8y?;?IIKBml3SjY1-<|?dQ9Xzt`7oFuuNw$46R_BRUbsN%fYk8^s7UKRO!<}Ib`734haeps%w~{vRnJbc+7q@b}F%` z|C@iB84@bqxDiy1i})&ZqtXDu#};B6Hd$4IZ7Oq6KknyYmKm( zN~me$#zyfh4cJ@H7m7wsviVRZBsDYQraDkDerXe@%%L*=`n77xU3Xm$L>xK%35nAM zI!&`q!2|M(pVG&K?$Xce+?_dw*7IG|2O&e8d^Nq6bssBTSIzxNAirTBe~$=5)|ZKl zr@oI~Vy|vIQlmg&f1&1lwJeP;IfQk5*YIF1Ntj z6Oxs+zTYays(AUGHlA-7&Kssp2h_JXeubVv_~pH>;CvMHrUVJ_-($S57G}wUt;%$IyISuo2STBi)%hpfc&Uk!Yu>nV$b4~qhwj~I%LoPj335uZTKqkoQtOT!bd_(}Du($+?3qB-!xKPnJc11G@^GeDl~{xAPXg8` zg$n2+`0gymZ(A@0^=d9n3rhKG`SLKRH(yZVt2Mw$RKCc5eG!fA_qao?Q60Mk0T1}r@IcNgMRz z-Hv!7>1EdrALUQq3?XXM2tmUk{(OMr{Ivc(r6~Xt{oHO&M z0-q{+eS0`(uw~oCm^Rpu>?u4c+#JzPvp0ziA2vG5P5e=7pp=Gq+HOZ##55G`NW6d- zuWMcSf*+%gyq$c6G6{;XWuIp)tZev(+ekcr{~+x87@1ZhfQDE3d2>y{HNx%u%NOZ zS4Qv^Ns@#-tSCfH4Xdb5L_0oHq*|%v=GfR4W9+f)kbt| zWsrv*D_^}D{|%$?$p#0rD{t6mhBOT~2}4-gV8cu z5NZ5=b0@nf@8q$a>wuPwvUB8+HVD6f@FNkd&bN~*H}Z3{XB;~2Glm50sH4X-_-HIX zOnJJ)dZ-7rTCo?sd-tMO*nX zH4#Ga{WwlX<2Pm3YI`M<7N}kPD`?wFC~`%_3+Yqx@GSg0k%y=6=4|MhSG2m7a`*>t zqvl^daK*O{1lL>08yP9zV>tH8E+pybG2kNfqTwWS&rnH(q)c}SqUBKt^2mReqBSjtvHMNNLPh5!M9D^QY0a zBT9H0`dhtLMebKKYe@zpYgov)ixvL#@qbIo{@`bZI{(QA3~~e5rBDPKHzl%Cod#^) z;_s!#s;7C(9M7f*Rr zb49Q@#DGsM&FfGf4Oe;Z9p!l{xM1^WG7)Ovc@vxi1Z>p8$; zBWC)uoO}v4;RYuL^|qh|8sE<{xNQUbCX)1Dxr>VJp?nXQ?cg%O%_<@`l^kbL3f6dN zfqN%ww~hUF@MNwnvQ;0@(AXI-M6NnauN98yVl6)!+Kp_7T+qTT!s^zHsP2)1Vkccl zY={@@1IuCPy;nzdI^Je(_4R{?cG$K}@(0Bso-~f)rXji(?M#jIj#hd%^tNDz_tEVK zwoAbpq{}$p-%X!9e$?*i9^F7<;`qT0a_X3z$_OUL-SAgwyh{64fuUGEdajyc&n zpF^%qQ-FjH_2)l@ZAxo{*=)x^JE$n9en=Agr%6#MCB2C93MoB{dJF;jn@nlmPGN2Q z$q(G5w37)`Qr+)qMNP`wDiM&ciqN^h6)2p&sOML_9T!*a<}wPgoe4e3aKJ?+^GK69FDEn0>YPlfZ|ZatVXiggjODU#D}NeiRFz+MO)TWs4n?+J zU%KcJ_|4qO+1(s16t5p(x_q?D6SsAW$M@VdxmR8syezIyv?y9X+u9qs)u#S~rShPb z3+*6h;%Q1U!j|MY9bkxy(Kk>4S{aAlDn-M1ug1 z9{W9Qa(x`0L<5^Ww7CYEU!=fcz>FF1p*FdBc=}*PtqheKcjB4&=%~pf@m)j23aLQi z8BR#%Ak2p04)Zh8nG>F#^4$yJD|vv_?wLEi$cOu^s7zpzSVD;%W+7LA6fQSU;@QnB zVCcuMnS&L?&hsQdgXbndLFodZ00U4SW$>ZjxCjR*=bG&>xBdd!#jw)=GTuqTJIoP% z%{4+m^U#G;vOeBiLxe_FU3X>Oy6)umo5g&LKSdj~)ZiyXlDq>^zeTs~>}cQ~?%J!V zzPzqOj*~D$lvvR$n(wq#oy*{&EOjJjaNq;u)bW7x1ukFQ&6$Y}g{B z9xk6azo*?wm@$K{YZi6JPSDM)2@S&E3}?vB`aGj`pu^$kQRE8#xj2*V>9RsAZmlTF z4Z?g@!#s~UNqIgH)9xz(U-GElNZS^)d&xKhibB} zg3p|-EV~wom()mL#ZhY)3e3jZUZMoVp8?r$e$;P`+so)3oXZ95fj^Zs{#wJm5=i7Y z`)<9{|N7vmjlm5r)r_U_F=nDP4bpddM#T@0v<^-*1%))CfjIj?ke_MWQYf##RnlyC z*_yyLjStQ$?y)D@@m2r6jx$Xe!fz+&t0rZrKgFs; ziW#-|lW2ZAN7TK=Wx!3m+@!oRKDyK7Nm!!b)A26seqk*fj=@XocImsIdz~_*$rDH6 zET5c4sM#gtq^X~t1f8}!5Axq{>Vyv2mADgU^6!T;a(-H?=B^xaby__ST&MEM?&)C0 zA8~j_ou<%8^IQvbT21Ck9?6+0B`PxT_Z^S;rJdmj9e$crH$cugWAnfir%o!5ZQ^jq z^S|S6@O%YS+aP$31>U!m$sO1G@&S9L|Mz|+Bbc0R|d8#&eRn=x*Q{#Y+a(# zI{YuG&Y_O_`{rZQJJG=;>xHO%-DA$*)V8l`Z+dO)?Sbn*!0;7=U5sCh6WWov8|5+L?O2K~jU$Vw75T!>B}-?`dvTGk zdyYh#7NTB8Wn!%C+g73|1|valFIWO#k{v#ndV~Z-Y_mi(>@<~@xmT-r^*YN+r67wR$SmWE&o$f&2G5NG z_q*SIL>-h*X^j3qjC}=ERo}NRB}hw`gmiZ(jdVzNcXyYBlt{OLNSAN`k?scR21$`_ z>4vw^-+lM~@B8079%Gzw&R)IOnrqH)esk^3?`Cxif$j_KjI-yFo!ryfh{8favM7BxJ=)|(5u#%$w8~gH@C@6 zjG8G?7JR51EwU>$q}8;G5*X`($hDD~RL?&&$sn&sztN^8KRL_}Ho?NB_c5~$3F+6j z(y{`Db=0+s_S7Dda6k0DUY-E=r#PABPbyxuVIUP=$H0_0pl6l+ey}wG^LyN>KsY1# z47Mg?$_EZNh}qnCV~cDo$RT>kYN))nA6>*2DOmJxz$KJVI0!*vCw9q-$R3k0r`1f)2f2q5-4%`f6t?IvLb>l1Y_#QO}UYA9uBFZH(mCl17Fc^roy1v@I|4+RJ)q zL<`H*Z(O@eA8}nTREk!?l#mUX_HZOIi+{|rjxnpL5uGy8<03Hl_&_jLaml~~dR?yo z3@UYFwGM6HqB~QzDpJaSI?X{d!@qokHk8woqU=l$FTu?)bG#e(KzQlP?FxZXc>%j} ztWn6Iv4vlF^398J4(*MDMYPk+?TRiz1B@-5)xtO3XnPahM)c>RdF*;0KcKB|rZLNj zhaF#@JK56nCT6C9Ja<}uDhs~&T4(c})55qsFGyM3yzl*-3f$K7WW|l}^O!60AMI?^q``3>sEZH?Ee8KvEfGao$`#Tl;TeGn|I0{5#WbC26!#ntAwn{^)R^n zvSQFt5(jAHKt^#V4GB?Z@eQ*`aS1ymXydU45*Tv9>wns))Bta`dV!g^H0_OCE)E4u#tB=C81PTjeibYwb>7kr`M_tzAXj-<=eCJ=|ZZbh>9* zjs0voy{|nvoPF@V4_@8u6zD6ljvsoU(QFg)`=-O%dT;a^9>E^-bbnTEg&HHZC7E`C za(XU=O?M(SS&URqB<$>Js}es}Ic)x9%#7X)#Z;tMj=Xtr2~t~46^RKIRVTvdgP zIENS&UdGSrmn&8k7{Kqwgh3h;Uj5pg(GOiBrRIpgDw`j@>;C5M=|ORHxf79c$5x$o zcDv~zkRDQADUb_ed*A9me|r>IYdY;uqI+JfFlbu!+3|e}jk&R@gZFAB4Hiu<7xn&# zF}oAH_NuzA{nCbxgvANHrgZMNMJA)}sA2PgC0!z7kA?BAy47TfFl`Qum-epq3gtvE z_@-q!0wuOpNz*l=zYt}+!-e7hAXBOei>0I)qZ^VdF@~w?X>ZC#R(4`0(S7IF1St-Bjr=E1TjFH-<4#RPmbEP@1gortq{Nu*r5^8qC7>*9&}_P25`4yDVwQjYy3XVR3u zlxB(W9Cn+t&Xh=|u;z}~5r6x61`%4`gm5t0%jr9^si%EJO~q$M`6tO`vfIgJeBX`r zBq7aNyTdX1V{VfZ4itKu(K~tTtE$;ZdpYCQB}8n4-7tKpy2Wd4_xm};Hm|>}?u?SjhJZxbZV`xByuGDapOiN2(DTP;pJ=~?h>Y~-f_EQ0 zfryus`ka6Brt-+UKqI3Rj3BupCUwH8SYvvTwCgmPYoVYZ0Dvk^xVMC)aCa15nQL3c~gFB;L1)C-Ol4>ZMSi4iVP#O0`$lP z>(3?G_G5(h*d?)Hp;-!uBIa4)H%@UXTKb>6{?Zjb+pP z)34jTd=&CbBlcfEIpoBi`-XI+1%A=9m0&EfnO;So9Cxu@EC8;mEkdV?Xbr@A6uXedhgT``xS+JI*t1y zf6)6lRO_J(0~g|T_U z#|im?K@>bQdv+yBrQUvya@+7@JM(w5c(hRBJ?1gKre5&H z*3?@Uv`CdpxEZ9!Uh2v_OT{Z)Nb~H69=i-#Wn&jy|JZs>qp9Y$D*@8p%;uJ%dc6Ku z9_nqwngn+4Lz%I56#8#i2eTXe%g#^38LNg$t8Z#j1-;w3zkSZ3B=%qE;O3~@`k59d zjxZC+$Xmy4(zy9-+`{2yop}aG62qn5hKpRWH?1ICh%zOAMzvir4XR1tQf;$_!b$Vq zk9MU72`M=_4)y($T=7+0nm3YBW}FS?c$kuS`ie;}rBJ8gfHK03xcdBE8K021_>7W* zQhcH#nx14v2q-@gBtxCIYdQ8>Z}UU}s12q`x3WAb zJxLBOnaEvUrWF(LbJSWAZa}YTu;&j*)2ne`*U7rp9Bxt z=S|Gg36U2GC%W(AQhA6quddPvhJ@BM6U7h zI1JW?DwcAZ2n-s;mQYs35_zy#BR7ZeU#NQD_N5y^nqzrH4M2Tc6oX8j&K~Ubp$bGv zaJdEArHD8-zwLhJ#UX*B#<^>iQynY2>zHPlW);iRNeJ>HKS`|`s`cHRnB!lxNG_r5 z3ihWPnJ!Z}gf^g`n9uSqRc7u_3pkQT>|3bJ+}i&bImy-7`?HOQ8oO!t!8R_2jb!Xk zKkB94w>V)wmbuxk)Spuz2Tjk;f+0)%xeHo{jTX&{CW7lNza^BLjpW~%4qUw*{_))% z3Xo>9+fB*xr$VT-k3DDq#?qp_<|iLI&Ak(;;K^LIQC- zD)vH%%*W;*ZV0QvbO*6d2d@udUvOE+@pL|R|L*G%xKDFitbugIo}Lp>PWP2j*{_yU zrpG$=b$#h{*LL>B7MWlXm{m}dvExAL_CInoY^6s=V99}9@ z(jQ*V%d0Mdgwhd)LpYT!r_I$Tdedq8AtWAx`lDv1xs&+0jXfTsF?gPm|JT+Acj!c3uJ} z!*|0Ze6Qh`b|^ixtSQSp|Ik;?y*s>1uS zJCs%v(lFe^njZfXUqfz=b$1ZD^!aC-=}@mzR;oEpyN~`4B-x!G)>_)$cIHN<4VMbk zSX$4UJUMwql$3aigk2hArxx#5^bKZ|K7}4}d2o*{l1lXy$Gum*RA$^nXY{CwUV6m! zhog^ml4(!PyJMzlIa&?ab2Gc&(i`?{3*Z zxAU1ul-}qZ1-b4PaKv>V1)U<>^G-;XccCvh4O|BD*%Js!9dzqs2-OJ)fSL z&k9tC#cs*2G(LIV253(dz9ILEUD%x=_Y*dqdxLirw5HFAcLYnA9&HzK8EF1M>-W<| zv@LMXEYy)xLc`~6f90cE-N zb0aQAj+y&G>IfLcZvygP2x8oWIzBzKyW%!l)%v%!v8}H_ncJPhhdl(pO>YB?Oz5t8Zy(K}+ z*wNO`!4_X#;Gko)^)nTJI|dSl_IYZ+_f2CM+BCzWW))n^0E>bJnxNagL^;2P3Fgv9 zyva{7RJ_p=eBH!p%Z+9roz)Xd%Qg6 z>y8z@?bO#MjJDcr()SbPs2+Jr6n==;eeP{r9~&%!$soS)1~6-QJ+=;f$iqZ%;t?|6^HB# zd6h|$%HK=HT>IrnhZ)+)%^u|7<8RyHQ}{JxsnPtQ&!>V{Yx3y8TWb}DQsNR{++%;Jc^759=jMlRDP0UU znkSA!miVHiZ>lojVc&Z*_YF$yXoMIgD(`vOVxdRON~OZU;tUt-iLmL z9X>Qk*{d|*i+wfO<#2Id6;#EE2hA&jAq_>GE``fAu7ft6qQU4;$(ubq-sE2#kYj=2 zkkd7JG>b3)Ojbxl5@}vO?%#!YYl-3dte@dc4&;Ef`IB&8m#?AkFg0Los)Py*>QpS7 zNuviwq@a%|D#5gts%U6mhqo5A$>rHsL$1p+Hfk79Y+75DB_dAztP)@JowwHgs2Bv= zKPM*?;DH@8w#p)XTq2!61-MKj1Dd|+>dn;fftIM$K}!~7ZW#mU#k;2)F0L62yUH{5wh0Z<+7TqxC-n?FW>*V1XO+IM zFwZJA3waXtx18P!=jv z4g&I;C(-R*{>c)t>Cnf-s|v+EcUFr%ENH$piz&H8Nl1?A5#;~+%~xPI^{vy3&>ojL zN%W7tvV)_Z@Hacxl9nkW6|`T)g6OISEHo!mUi|s;Pln$=CcvE{k|*5oTXEnFxxlt( zEipq{%xxBOurKZ3KSo}p@yqC$GF-gRoUvNKi)pwMDxau3UfzH3%8>B5UO!%T3Q>YS z{31HO^5F-%fvNyHTg;etfhWo9&sV8bs|U$wUcaxSluvRy=Mv+UPYO<-zb3KS3-zd0 zv62~TMMeKf=g9ut*Wj$)uZ3Uhq-=L&f8lIt_e{SRl1etk{Dhgcw8N*`%Zv#fnZFK-mx7ERGfo}Z+gdT38o;6Hz zbCF5yZjSugtQXYQk<}A?xc@3wA#9VXeeU1dF;dyw3jQDNszU4aErKgt>*qf#OZ5I&&%M7||27Afp!Nro|8SbQDxOaoPln#fAjmH%e4#Y~#Te2R z{dn=7z$OhTUh)l=TK)2A<${&A%A&QmtVbI4m-$AF#2HtU*>`*Ae!=Bim3p%;;})0V ze&gh$ghpZg*cW)UY8RpN_)8DbKq+1fWM|ws*QSpj-0jUt{w^ggUc<4CxVdpNlysPV zBZq{*7g0n8O|negBpzaZGIl;{4|`lwQ`PO z3gUkKv8c(|Qn(%Ax;fxUby--HSehFlvhKjW<*rQaiZM~v71?yacjuh$3_4o_)y zKkA;Nz6hAFNffHbpS2PJuhIB0gVEW;^`p7xu2x5ERm^=p=R=BgM^E_Plk~I{_Yx;^ zpZ0>v{%xbxAeh~ms2i%^VylrLZ)bTl1xyDr_f)F|84*q8yrzb!W7uN~A9(E}Y~)@D zald?WM^CknONkS%PXr;4=>=itE(A2Zkj*FalPoM3ZluA-GN5|(?p`gCbye@HnVuoh z6b$6FWZeMaGhtxi!xLOtVwQjN>=?&?K0BuBM6?D{7~+Rn(r z42ea<%GA}8jPun?Bo-+%D+^0kvX`&8kXS?=Y#p3c9F0ssT`@CvD-$zi$=66MudQ5N z6wRDP9qb$(?9J>!VI&p_D_d7HXEGKETO(I9F*6efQ_!rOnY{&Qm+RF_PWJ!$aN1r5 zqOR)H%>#mBthllm9BhXwE|=9;I9*2r22>^t4vfhPPKE{cBrz&~^8+Nz0?e}dSca^= zocwx9^R$}}ev24mrsoe77l-S^D100)=Ak1uT;*Qi|62_Yb~^bkTMWOth*U;|`geat zw2kijJF{PNGcq!ooA2fwS~KF}+NU_;`8e=H}dYr}rm|RXQwiaB-tag_gRz zyFu#mvJfAi&g>UOT4Q5let!O)$>P=B-A1zk+-jbvn3#+6^P0-abBmGCe*Xp4uoPo`Qk`czF2uC5NS^??P;plzD!4 z*9;nE3$6bCii*$b@7BI2`0ggssq}50Y>tc!4o1Fz&&tEI{LKP=Sw^NX&1#}RiD6NN zz!(84V8`WbN3HVcS9b(gdVE5Hbd8pl)ivvBG4vWE$?XB7~WqNcbo9OP&&V+)#FP8~%ai=ScM8C|OFC&0kLGDl3 z-`ndl{%Wq`=~-7&Qc__)TVX^(LZbVMmp85HZtn+23l}~8hj^E8vDfQeU1Gcxx=o&O z4tmwrI8&q2!Tktm&$HOya{bidU}rB1J$8rOJ3&Dyu5OLz%cgVNsh8_7R#{DGS6g%0 zOxymbG}BGU%rq$1Zw^Kn$>2*_xH;Wcp+`VK@bU3M*5R?66N6X`f66`v6SU0`P;u55 zN0B&xxYVqp#knCWxzzGD-64PBvvvxLL0Nh$CKQx4lBqeUN<&ka#Gs+var?s}GfuA_ znm|Eeq&)}$8C_mVii~J;V?()KGKS>-{=P%KA83o8?Ln#}T-J#Z5fK$h$L`+id3lud zyRfl*$pNdJsjlwDtMuw@QF>$G4~Q^6fV=A zAmciAQrOMcFqy!ak7kLOMYhSMaTZ5ne4L$~{Sedk9ttV|Lp!!~iGa;m%t{>{8QI0f z1&ic4A)$8hk8#-)<|9bQkMX@dTW3#iZ{Fp!)KqCMs^Mfp?cVXs@4Rq0&lG-;eP$n( zVMPjw$85g!?d}Zw(CC3vYfo>ltD#$8TIVU^cW!VcRJcatvep~(J?56~Vb$X*&{j*L z!Pemmqgk!(ruz?vc~c#~B)d5>ofJx0U8?NUE@*@mP%J=+@VTwK(< zGB%^|w^Q=-^9zDUy-bRtk^e-7p3qz$#-2LZ+bdaDIW|5XKRD8A3#3q|)_y1{U)icN z6m`3_z1De#SE<=NICT0e{8_7VxMx4?QKASzd*8ZDC1+}?tsFe{Zv3ms*o1#Glx*B%nuX!vUu#)y1KgB+Jq?$E@kSp*wE0>rkAKh z+-s3cjJ8T}DDeXRY&q;bloIBO_zo3Lcs_-`V>90jJOW}^atexiV%Xc; zTTTy@j{-j2ttM^L5S^(d9znsSfq@8Z0Re%5x{AqB-sENZc41+Dvmj4TQ}*>EljAEP zspVa2ua*7g(&}P&jw(+SSwU)8UfckBiX{3NlDLQ8%~cL~;gI1y3ua86h%R?>r?+vl z(v2An9VPkRm7VM5 z$-D&xMbT8}mvs*E%mU zZ(RD}zBj!ZvEWFk?XUc1`tT`!xuN)3=|@&BGE(Be=qFjal$h(;LTk$-iRXJ{#VTx8 z^ATN%^f7m;m}d~zG2d;LlKe*ZzAf?LjHKnpLRzc!*HN&?V;EbSP%>hS>LmPR=>np2 z0>Z{xYzZ@1;WroYv7IjRee2HC{sHZ`EM5d>jwQ4N*mSjGeID8_ZOtQm)7uqx_p<`v+ci@ zF_##dbFoL3ef!DHmFwx>?uN3ngxQWtFntf`F&*Z~J3S*l2l~3g@#0Rx_06Vp_vy-{ zVP>$49Iqh;4@3Q_68PQa+`x`hDD$8pdFJr?7}#2OlKFauZiXF-nD;}M zn7&Jzl=p5Fp}P2;VuFEzff@QgfBiBxH=oXv#u5?|YAAGG?TM1!yN;1*Y};w0d|~3S z$mqgk)n_(76aNn8X@-F9`Zv3IZV!)(&M@>7E>*wvBn@EJ2qf}xKqub1FIf(!z6N7A zs}Oxpoxpcy!=CLgR4Y-pnJ!f-(S0;I9O04?pyfo*Yt>i#M8)-?BShg{yEDdp-4%um zqznuUE9LK;_vWfgn%;Iaxs!4ZwkUHbip$mBVj~Y3S=;#dII!#o^VK%B4&ZIx zFRY$y5F-n_kvYUXFDN^8%B6}S5!nij4(yX;PFb#Z=}q)I?h*#O`3uGwM0M)~_pIv` zjofByp_K{QF&!;!o%h)nF>)i~_tUJgU}qqeYL?}=YQ$ZdGb_912G;w!@68_d5IKCr zbGp5B`htj#ZqVq?G+*=u{REV@rgTU`OuCPkimcx1d2&(`E48uM=<%=;MhdXQ%aV~a zt`cA%QF6zTz>>ohJKT|K=4KOjW~sq|;NGn+am$H!X?0lB%k5Qtr^{UTUODqnjeqm|2} zRXttZQV8Vk?(W;SZy^w*5OQkjp3cr(Py}dOof=yupWVw#PX`AF-a}6>F97h8a9@<7 zMZRz17iW;f$@k~Mk>OFF?WdunT&teZr$BU3e*kwURqxPxcMj0TfFHTm;=+P2Kq~$% z0elpWRW>ujLqm4pj*E+nfxzcor1Ll+EhoxD|BryKu=f4?_gNl4{u`XD=lt5a)28Wp zF8AXIH7S2P_cc}BaG$3ULndfRvEg8C4V#cqCiHeIH-;9qfKi)v1#_kJ^XJcHKN3HE zl6&)SaNpKxy91qs1(r52;051Bh$ zonRf5gh`j6->N0v>+^(Tdo{3i>I(AMFOc!H+#U~9+swGTyHhtC0Eu6)aH3$P(iY(7 z*R_MP_!mGJS($(*CN8ezJj|4&ASai{(a{}&cQp+*snXC?iAJ&S&o*d!WR|x>eSKMk zND>?1tSj5LnEWHERmqt*k|{;KQ}jLN;Ftj z+Fdq>sV)SWt}EaX5K8QbYpf=ytp!{*P?BgAKHJ*aG1&?@uRU$XsQqfHtNWbT>*n09 zCCS;xTCc^o-aJ@h?rW9=n?-W}l!U(eF9dYfW8+i&Vqf`BFWHPbf;0F$%q=XygY$jg zzlR^6_*@()B@`BJ0z?>s0AwOI3IMd?gw5w*$1knOjL%<`E-XquKbWD>ZzR@;yUJv`1o1K{s2>^t-kK z__gGuF&~^W(tv1J7bs+w1y4a2LHW2B>k3;_U_QLX)F4WIYP*hwhbO6SJDLTZT9ZJZ z^tKq6N!t;Y<$pkzGEr?}&14i5xt0ECPz|enna;9-<3r-VB3*UuOc!EJJy$BKy-zlV z!cYjv2mMM=9L|6k@bK^$saRTC=9|m1=rwfJB3a0SZ8TM2S-dfMPApc+PSqW3)9jeJ>V0`tcZuu52QXPfoTQp;{}Cy27#6j4F996sZ%g zE-#;IaBFL8$6sA(9L!bQ>`p`KD^=<3W-DG~a9ye&Xwz6p@3sLgAF$8@`K_pS{p*Ql zx1}}E1z~8!zgAaEes{$P-G6kp#k{<~yV)E{m$c5ey#+A7KPN&I76xWvagoP%Ha0#U z6CYnneG`7_)w`e|;^)sXF)?X(Nmva7s;k}P<gjSpT-Lc*C3{_^U|>HweB=iJt>Hvpnj`=yyHBebch zsc33q0*enG%0z@w-5p&~>6#|U5xUr_Lrk!&tjwOJqP7++LhZM2E?B}j)BdV+Pt*}Y zgLKPoJ+nTH`&9AdFBd`6v&M~77I#<58N4>zm>_C2Bw2&vA0JeUplGPJU;tFyG6|+#+1xwJ= zx5u8Mqodt+r-}JJe+MICNo|~;pUcb31J}>|k4=HbqQo?l4KvQKb?+IOntJ(hAgoK! zbxYC0ViO2oV_h8z5qIP~+1RYn($_3=_6LNdsci&w!_capc6GrR5&?QfMpOKN1E3V| z>r+cAaA^fxS-GDSg|jBcsM1$gRLr)#B{OIle+S5#XC%x6y&(cxTm zgb`E&tU7`>HHDs8yzBSZk%cpgv;7voy243)e?xadxgvmy;bJ+qO68i{Em3UJ+Ag5( z`anu!LjzD~<{N{lV4fJButv%-wCxCI-d_6CC*-w~b`3*u;nf_oF+2g9=? zlLBxLm?(e{ye|%j@$ofF)FB0L(tml#Ler)!5P@!tL|LVgK6k1M0{_pXH#+INsp;8j zZ;Z-Ny~`#bQn>9G`dH~9P$jT>97V1Dl-rAqA1R-VfO+sf+sV(*r^80-pMAJJQ7Cgh z+nEBo(fBV~ke6OAVV7?3MWmK)e*mKS{Q2|oiHYLTBBfjd|GT6T%8}LDSSVW9ubO7@ zr>J1O?~G^rdjX9jDEJxaEieZ&Ws6`-0*g5R%|d~cl=Po@jIL(`gATiIqokt+)0oVx z$Ln|NN&lSBy;Q&1dqW%$IO!wvQDDn$y(gN%NruS}^N*BP&U&}fw+U7~u@=~kt<&v^ zQj;;5(21{YU6;S_fMEE52}6|=6%_C}=j7l3qz9ovLP`oA3CY3MmL>4}z)(~Q08 z7#Z^ppE5HuPft$+F5<_JA06r;uQ9N&R##S(3gw9n43?R6YKoePGu}2eH$zHGfjas1 z>z545cTQehd_o@;HEL;-6ytd8BCvvaqsx9bOziCJoScdC_f#AlK-Z_^_=kWU6PGzO zG!&xlFjJP1o<0u-uA-s>3kxfK*fE2Gh*)gUdf%VKaB;Xyij6I1FXVMhxy)Rh$-fPe0wQWXmLvYez-X;6PQd$A z5!Y^KKqo_gdTL6|6X2LmAeME`uCAP3Hvpgc+Ni2xQ*B~?KhY}uPiiaQvnce2#OV>o z9M0FqIQ&5}{qs{WK~sV5^?q-_e(zEQ{l^Og zf_W%@4sgJ~a$UGYrD#7rP%=J z!v#G={o?fWIlpJc@1KD2sC2ToXHKH4y|}o*t5H6913^G1$vXzS36Lhiki^=F2pTFX zrB~F{q0!h8-M$QXSEX9tCOg3G;!4XZrBX)egi0$b(%Y^BT9nVI11|r|6ySM^%oj0J zU%8l16ud!3My?M7o`c#J$I?Qn=C%V@I~XJ~@SeUJivy-qIvl8NYMz3^La^IMxVgDM zy&m{^eYOjx2~3!es+hh$i9m)tf@<=>`u~7;0Gsv~?~LogwjsA<7PCuxuPK0?=O)G8 zcr1PWCKt~LL$8t2l;D_h0E=1msdqa;Tte~J^M^aRBa^&J;OkaP$-EwtyC(z*&squ~nfG;$j?i^|BS=;Na>S8yx^{ zqN1W=WsNNzM*ZRfhkyV!5G5t0;IbUS@8Rx(n9tpD9f&}ztq$<02?&b4loaX} zorlZAlv-y0g{#`qtNRbCDvJf}HX0gGcxpc7z`Rk#eEswkw8{9O? zbfwn;lUf8(Vv{g*QzS0aQ=7!3y_;A8#sVkgN=<5Es~5mPpwtc*>P6RqTPZX1*Zou^ z7TkgWL3j|PmKG5Qb_JLxX06I#q(|M@I!yWMRjRY+U*)fwnz#s*|It^_Xi@;gw?IBa zyVhRM9;bhy-le`D>>7==Ib~P`bp*BC3dJ9)S*Nw>vTOK?2uXnuUsLuUU`H&yy;vmR zF#o-)4@8&8|K9hXmZN@fs8UPEkOWSMtOxwzSyZ8`rN`g10v#dq*9I5CP=w+t#7|>| zlkx-gS;YC5!s&NjX?m^^|8D1>QjDTP^zxJGAG!#Akf9Y{{AHh#el2s*hkGlNe?t56 z^e-!<>^oWhQt6_Yp(mLos~^xAy3BY7M(OeK>>%RMdVd{ZBsB-4j)1dym_Q$XU4ZJy zTf}R-eljLWib~9z$k5x<rEU z%+PwhQ$isb?F$$MM@Pp(`3(F&JdL7AG^vX8WI`>u62er6u=_=W;|h$~Ht^2!oUN^` z0TcWm<*r?s)?V#s)mY3{XPQ2gfb+ zB(b*_@{~JS4ZtsWxWAjPv1-Sw84rF zX9_I?;L5@A4M>50y>m#(9#~pSOWMC!sh^5Z)~t8&t(DYdATBH16$r){T>d_6eQ5DG zBq=eh3VGqm$ZAT>gdS8&_}%wD*4MvE@T_`Tiq;=6?ktq~ORPCv$bI37c2NCP%=1?{>HrwY zmj*%?3>{>Em01z3J9uh|N`3sJ!EJYxyl))s zw6L%MB%FrEOpM_5)?g}!uZ_OGK8P&+?_^H>ezzw!lDM2)WLTJF2Ctj%+qbp0KXT5e zkB*L(mX=OiKqF30PWj?bOJLu`W9D~=-7Rx~g**p)zOAi|`c0bcQqvV!ZVffMoABM4 z^30b`VSd44#9MGR$Hh9e zW#;UopK&1tTH+wq2=~=qlG%Bg-)QjdE-e};{IK<&iOy1b(s1W z|HSIFq+2E6L1%vV(o{;VjwIo{qlmy~4Zn^i=KJ!%dvUNxp~0e6DW+wrt)Zb|wEI89 zoGv;xwz%P^cr5yBI)IE2=UVUxgGE47^jdCt>uVz+0Y9aTx|>($nJ9+1c2jwnaikg-qIR=gX#C zs;}5uxw~_};ujFea2OgK%p)93e#icir|_jxIThx^Augs{md?ik$!x{j^LdB3xVSZa z`OlQnDEn+|(&~&c&eJ@8xMq)T^92^wr1^sScs-Y3Eo9@kiH4{1tLKpn_%7r=uZOcbe zSNBtGZEa9cX9zN0qIh9=KmZir{{P?59Pi5)F9ItMpkd)PfuMNcuO_$u)_9?%sp;6s z^88d^mBVdE^?)8kX?uE-@d2Ux{^J(VtjoEQIb1hQOH7KXLnPJK~h z(ypSt2?uPV9l)q}*JpNicGBC4`7SE?&ljShDP64hmzo=zsU66lxq)=oM0446W`?}0$U0;nG8@n4~Ow2`Vdw0ijpzh_nj2IUS%K@#f~OaLwC~7s3LHf^=}5!)lxoU}O6Tw}x`Hl4)@6U^X-l z=c5OsVc(@wcX&gPl&)d`7%94F>gO_~-K;MWz@Cl@@su$PJ_)}b z7YF;L1t@C}^9I@?6|enpTJuPN2$evy*&Gu?{ZmmZBPJehCEC+TegGoDl7Ukc`c0lr zfKM!-|1+)LsB8UZA-@aRewl!2fCuxRk+BGTUIO?mKEpFYsoB_HVvC}b$%bK2Z7?;* zN9X6@xB%ZDB1RwlvmJe4vNg(q^v(j3Dm*+KcqKj{0M?1`3RkK>ap`I3l(V!Tr#pT= z2(<;fah%F-_5rMK<6#S*Y8wFj9o%1bI-5)a1+O*}WpY}Wi$4fepb4LWI|9UaYHBKw zwi0k0g9Ijx)2ibSnwg>Xb_dErXA)~*y%ZLP3%};r$-@I338{Zzzyv=iI$+8M&By(F zaQOA~kkll!EC3SI)9yeIqri?5S9qg}(5qcfGO2rm_$-p`HO$8j^+$q1x6%~x?>&iX z)@ahpO~#OR{zNuJ(Yg90;8O$KPibjs;M%8@{4)r+<74bII2m-}NSvJj5Pa)A#LNl} z33>7d6X|1tsN)f3alVu<2VM}%V?GuR&Y2$FUs8&Crt-p^Dbf$3!6S_(-8d^h_y@!sxtYj!Mg6Jgtn%~UdfDH(6*C225j4Cv@Fl`&UmGX^&ohld**=5}`b>+5Gj zJj>YF*uNoOUS2CJD}aS`b>EtX%onQ^29Gc?GCmjbYl?_K7MS^$b3?QNyly*}*Vpzx zD#iC9U=KC_p>Vd|0t#9tVibxGS2gn|`UfmGnz+j&6Uv>opMbqs`GP{g9xP^UZ!hSw zF<6Afq@M)n$M-WGVhdJI(L-z_UHcmaAI{Hf%{9n^~&5}9L=Gp zC>Vgu0dd&i`7bJH3jjp|k*4dPKR*C8d`B4AE0nX5ktp;uG&Cxywj*5Y|3c-Mn48y_ z4-p<@%TebhHiFpELT97Lp)#hRNe{~XqfwEeA-p#Pb$M@oLyU}!9#y~h&{HEfEKz2S zmQ^hP;JK3%Pe;(8d;o>^aDT&6&-)l*0W%K>^Nwji{Tzw%j*X42?E|fJ1W{5_O8g}I z!SB^aKE6K(k@8hgrgWRVll~k;s&w2nhj5Y(YGTxmu8^AqP0uA6qb<{_7be2m}%Wk&J$xebW8!i5Y@; zD_6#!StG*(F!`;^#0hzAjzA^6a@hcJI0=X|u!n0aA6ljC?aLApQ~-jk4+F*|2mu|J zBsMlyn7g2}=rPu`@|0hso~N3G-!ry9S0c*U$qA!Id1=@m`F(hJHbu8F+cJH^_aZlN z#w$&!eW##nkTyx|?)DXM)=&Udd<2YrOpHc}ZoPAUOpL-4gB1X`9GcIPOG+4oNRdil z_nkR3jOY}vHB12F4+6v|2%y%}$-htqhg?+R3Wa>mO94NLO33;0?KP*tj~csqb_qjx zR6^CBr-Oi?@5c{Z>G7f$guyz48EVDA$LJLOKqI{y-i1zT%tgWa@{xa;)y@bW%x{0~ zceS1ZfOg7_Zn5fIp#mv5n}u0kA;`!$I5hM|6YGCF_wgxRK%1nF=3=!wg+;($D&;^oWP_VyS570WC~|KVs%jf~n+TOGT)yOsV6Nuy&N48~C5_N||Kr_tkaI$=Hz zYyxj`Z!{ZrJC5;k6TR?g2P&2?tsTIb0}tCL&3xJd$Qyt3rhpX&Cqj@Z_2A z-n|2E6tc$_yZdAQRGp885ryLQmLpJImWIrkiHV7|OWp$=9UTh`pAG=uvR_rj`OhJT zFR)ghG#VgMa`C`LhQbMf_)2Y?I0C!YaOvd`OFDy~d(8Yl-UEfNLzVjAU3z^$I7FAm-L8-_9wBUkdC$7X0=C5k_G zzeWH$*v}3ewtVwwYG!mB-G?S7P5^o}{1b+Vip!5sRp%fP3Z05AIJ}^T{9lCs{Vt2#}rRH!ws5)1+^EE?@i+6zS@I_5B0LEZ#&GZkuwWukjnO*sRczX|U zF8}|3R70s0AsHbvD>Eq>Sy?F~yNs-oQAYMi$|{?vtPnyOMF`2JC?jN*O-4l7&f~4` z@BjaubN=U?>s;5lPS^LkzMre^_xpX{ulsdB$74M2$8(_rdDs}D)+22wbRgx-(0Zx= zCg?xwnsSu?;tw{pm81>A(lZv$=jG>9Y_yWsVW~jARw}-;8zO^3&#PR11O=q{YpFnv$|@# zI$yr9K5*9jg$V-{(YMO8!>o^*fGV=~UhC@;gs$eX+65?G1k}^t#XG4HUE9CKweF9( z=%2;u`PKRDC`l~U7Ia6)kwtOILUCZ4TQHeFfJl7zmg;QBaJA9A{`NG{gMq*21)d%_ z|DVr4^d~#?iX`fSTg&M?Cm3Rno*#Aq9*ul9vFW+EC)-x7ex$Mg4tLeW4X;|%L8`I! zb-lf72m`4y=p!MxSzz*E^y%0iG(LfYQmSrM9u&~SL0`X~6c&!$i|c&auS%xdceATN zWyCkdS)6cyd2D?;z;qk%3_(FGS?Im(@9#n)9VtgXnu!%OU0YvA*D&sc72>e|)YP<( zjLg~5F;!UcKav~k|55A&zj-54CUg6CVXjuPT6J}`i;D|@=DN;KXXv8-HA>UdHsRG!PT9wOzu;~3gM#1KQkjPqZa{aR1lH0VsGS^?+L=T z9Cv84BRE8-!0jJG1Lz5O5*>(sw+h$a90PO|6dKTLLF79#P$9+18PMxBKde;l^j|1V zn}Hlan&8>)0)IR9G|f+t9RG*9GI5gq-Yq=xgb@q3+Ikb~a*L)BimS-?n@dAvdP0+b*VV$&lCY%xL+TBJ!(sEx+5byr>O~1=+xc*?{ zbg}1ml&RzOJw0pUWw*_@ogW>X;JHH5o{s@2xk_JgU6tSpa+f!jnYd%4yrgU9jlss}D!yf|hsu}a)wT-biL z>y7M)FRK(&;CJ^!L?Y-xulK&4^CjNb&J8Y55vLU1s(o_4lNtr||KyD31-hRQh)@p? z4?iAyq?3D(q@B>5qLsU>#~pnVikD_O4<6?n%8o2~;;f&EO^=5EAi|IbI@lTCr)odd z)j0OMMGA@%)=GQr)Kz@0yQe3V;Y=aR#evIsDt79Lq3UhyrE>z;OiuoNW~Ll9HPZL{ z_h<9PL^GNX9yv1g{rlPU9sebaKHeL0|Np9HId7LkOP~GHa%1)4a!FR>O}BGmSU3?v z2LX~`_0M!e1L!wItY_}r&f}jUhfZR-IQCNJP;0S+<;(x~Qs_KWA7+$j{{9)9p0=bG ziqiY<6?-Tqp8?c<>K0*ZjC%nw-5dE<+YeiabI?@FLQ!h`;l7ZwpNO|Ya@noni_!Z+ z9zOwm^21ThVM_?BE!A6^e)AA#p9b5;7~#%~&}{|!MW6G}rx=rjo|bzZ$8V|ri|P^r zP7#eaSKfYk`<(=GS)C+Pr%xdRm-tTYy;gJO+^)0#Ijk>PjZ>#qRsW84F3_zB=vIB#c^Bq z%PY`HN9~#9DoIm(n!8v=Im#o2Q?9{rq~|HpO0-;Tl+{wdv7G zW$&sh_J53cRd=6d(4KciK)rqY514dC4nTItib{)%zaTApdyh3Ua4$9vcn$Od#SeuFmxiPc)OeExZE}kBUbM!hb_N*eS4lAhbvBYWfP==&!Ery{v8^vPqxniRz!@SEu@if~w z(Ika8jYK6vf!E38wPXS@XTu~KF(XKxpjQ2dg^2JOhE)t@N>0v~WDO4p$I=id$p}0m ztH@X9j@EGf;I{_=V#<*On@b)3u61+_QU0%%i~Rz>%nh;KyO(`;$9orJJ|^G^=uvdI zXmoTWHxJk@oN>clA7bIZD40><7SylEc0r}Gnv*EH>L#J?IO8(jia9q1@ih%PZ~?ct`;M;#hB9;!cjRcnmHC;uIOy|>~O32josGw9PZ-30{q`cQ9C zmD&#-EJuEiDxN7%o^dEyoMtI@KU_6VbnTyTJz738Id0WF+(_~9lguYi6KSp#(%_e_ zQR0`{heJL!X=WSo&>vyT<8K&Vn|^aH!ux{BRVt$6nvo+ZnpVbjhNt)wH}AgId_qyH z+irWDp<3`ut7VI>0;vy|#*UE`CM(4{<+K)2ou|Ljs+(RMiEPWsN;HG4e?+CeW!qv|&O{dV?ovVn+EH1l2xzD^WiZlDhQM_87!_b zJP8kHe#A4|tN!$3c=}XKjv?2&*K>&88vJKf#NP2;=cc$}IH6bhq)hRM(Aj*GcwYWj z6sEjXp)(s@+&(UiwMm158OBlf4~Hf<)cv}3`*uh`0QpysKi!tl###BR*KK&IdHwQv zpfNDTtEc^(*P()!^-7_4BCqEwYoo^mPtbAsm=c9^A$#`r}G&q8laK) z>fuo7=k6(#|MT-)cL`hf9u|`qF9PyONvfa)l|A+8JKg;sw4Hvx18d*Q}g@wtgCBW z*D4X@=6=z*xwC|5F6yCljj38C=YD=h-LiaLeSNRrpF}@A2GWKu%V1jIi^J>dU#tdR z8!x)C2*z`behN=GW6&o^ExQ`}-`9dH+feU(*W@TmUY6q8ZHI={KK8@|k3M(ad26n* z^7R)iD2q4t=IK7^yLN5!AUstxm&A7+deHi7@JhC5uN{>a!~aS-<1-_ z@z2t_ZJ8w6bFdj^9N2f5FB@n(YR|X+my-M+^U!aD35@t+qNGj#J68ae-aH>VJBuI# zzSJPbhmU9t6ZY)x1H;awZ*Nc2{e%mQXxuNz`e2Hv7a|Z)HH20@0|R%|E4W@_Vq(&v zgRM}ipk<3{IYT(41`&w$cIF^AD={70-NYS(XesKuYjOEJ9yYY)|S+jeX^++6th%GPV?7qcf7#KiB#bt>Ezx%C|3uOu}kmoa*di}Slq6n_tT;{_%q85!Tl z?69X~-f!J8W8&pi4kGLAts&i-lt0}ZMQwH00sChU{($rh;>E|pmLdNB{y3j=;tM%X z%jZ0Cv)zI#FJ*qG7R+8kCY^``Y*KylK{%^|MvOf1sCK$@`fKIvzNV}fm9REoEAeC} zA9+xIb3%$jRmIQWA6n79p|mHhX^Mlj#I3~L|2&YqfO`2VealpLfx+u*%#jVT2Jz(; zYN|jO`>zUUale%(vNyeztz^>UlptBB~;3Q%YpxTO~2_JFPI)3JqZ}N_{ft??Z4^_YiFq!Dq z$TlRDmY%6O%k>pcY~4e81*QZP9k?~D`6cH$e)bgc$Hz}R31MX6I-URt;yo(2*?x%` zW{0f5<6rJtQS)7{?|i^0@p#^^E|lr#aqhMx!ebPNi+2Jhf}SM(Vxk32E&MKAu16Il zxLNr5L$e9IA9umZ)$!Tci-{g$BXc5AU231Edy3+!g+UA;nJEz)(38;$rJ{yk36*Q3 zK#L>t{vZu%dkPAfL>`G(+th;nC1n-{Jer@LUbB4C++0~%c_~KQCS`K6M)osTtd*Z- z2|lF$=|`3SB6%p6iH||`Xa5%nXHgA+2>eWF8lP$Dj>^ZD`S7$67A(&1#MNq&yZI&X zOhL|-YXN;F70kQN3rCa!hhG0zYZTWK6LREqM0vCIK=~CZlPGEem$=+g@Xd0yXuzw@ zTf7?k&Dbf&_?khr8pm%~kG^$v2^f9K;ML#vC~dFjl+^NYB&@L7tH)vA3YHv$ndiri zad#ebz4N~6)FVVz#bV}{7h2L2JuWI<=b@iq3U*iAe7rCn3Y>QkB`$1Px}~>sysmPx zCVcoH^QNV+pg`C&CZiO%deWv#l)Uil!sgF9Chy@J=N{6@C$dl3Xy}pJ1k=n2#USi5 zP#d2Lj*kt^;t4J{1kP7mQv-?etxWf;S6Q-*PulgL-?lK-9VsbJ)q)4+0vYb=t1%jq znMg`{dKGh9_1)$qK%E>Ldtk0sR#rAPKB{+xEHSNhFizO=K*wjPxL*aHXdSiZGLnmK zEeZ|NLLiruGmXaz8YodTr|?M2D=7SAW^1cTDhx;tjVazGZ1K^UCW3tLeIv4}ARw}W zrnPEdhRo$HO*E)`Tohw*7vSrD*3v?6_f9&UzBN;F$M@R(ern>EDG8>9rM4P&(DeCw_ER- zez=}szA-a{hSB3!#52F{Tg}J^q7|6TNwjB$<&Y5rCHWbM* zyvH=ESpeg8>@N6|5Cvvg+5 z{`T{jR%D}aL*|_A${Ttp+Ml7Be&Fw25>VKSfB!n)yqT1~y1a}MPi2Wpxdujd&?(Tj zoq%fU5RW!V&PF58t6$K5sH;;w_*MKNksb1Q@GR($xXNCoq?jV-64aRXZnQ*13-XWn zt|&T6t^KNk=19cAuJ^69q&I>nf-D9MP%|=K;9z@Xp99S}jOQrzo__$G4BeOF_caGu zSy`kG6(P26?2TLQ-xSWBix2?l@&<4&5*?w275>KA>B{^EdwX|J+a$Ux_d=LM550avV?!GM9(Euqcq#|%18=OY;woF zp-F&0T_JwqsatqW~3k#d; zv)=6YRkhO}t)P-nvDV_GqPhrQ0@$IJjWJ=A(9OuuJe;BW3!d#t*96FLj_+Vk3W-(F|xkK~!wtzdIxsIzYPy8ym< z%ONIFHs8qe;V$MbsEcH^QBKURLzOeZdrm@4IUVODeYW@(?i4^CCJ8q=G8B2o*D{3+ z``Ap(zs$~t6`5YW3PFE;2GUS_34v5itM?<2poh~neO$SuW-?D3G_I2v2UdJY;>~vh z*m6&VaBHEN$A)GL9JXd2ql{1Rm$Y*+OeGyoS>_0 zb;ooAngW-(tMeUSQWP?$rl!^)4@JddRHyaB{!-&GHc*XJc<)OKr#DQ0hdDS^j=X>6 z+tJ~5=x=(W*Rvs%3N@WOckP6(Q z9UjrJp#o<5bmu32Dzc4pQHwebAXRB@Zf^D1*H=bHZ@nyl6)jW?WqVDursYFYZ|XEt z^a7c&_ zXM@=Ygl52{k2&MtqaG#^hoq?pACZ6b>36YFyqGWuD?^T0b<2w%@ zs+yUZX_YV%5jnmOI>e)>Ieg;AjT>jqlp;Dg-~sqh=?}*#Qfq5RM@M7h3#d%sYxwS6 zzk#xZ?Y2J)r@9o-u*V&JeSIn(3Xl*%i12w&8XkUsGA3>gd_}R$?zr-yM8!_y45Yo6WPMX`=8Xn9tDsgORZVpKx&e}obQyCE!rY4>B z7lTc3w-J!HW_s`*dc6YM7-A6yy%;+oe3@}yb&7CBcZfzp)?Al?h=?>UNEeRScUa2H zwAvV+BK!Pu7WA$LdV2Av9qzkBn@Z%vtogD7mH8Ydk!;m;uK)m`F1SP;Tw|pWL`3v{ z=}VrY(}iMfQDNbf9Um_xKgNd6WVoJNcJR=l8Zo~JY2#?(J6hB`we?89njT8WFJ#WI z{CovYM(oU)RS2G}6zHt&Xr{(?|2v9Y*IZCHRLk#^JiuPH83sZ%5^=I4@(0)(i{__# zg9VK=I-*_u0$nX++p~B{vh$DTHeStT%p&PM8x#iGKyw|1TILZ3hAV>w^^h+p?jq5h=?9AxSS_(+d*VJ_1;wGGd&TJ>25|! z-+ly@GrwZV7j_Z7@48dxmn4dWen6aKXW2F)f!p5&ng9R)3PHorkLgy@S>-iEA|t&y3H(G&D38$jQjE4-g4>G^&+QFlrK)Z9D5!-=xVKdu;#4S=dN3qukC$E!kO|Bz7&Ac74X`olGWaM<_xBfH`6_XiMZ@Uw-?cyTesY*u?g3Id&dE96mVOb*ytolJ$l2MMWA6PV(KQS`(LG`G?KM^0li=z^(ugAx z&xC~Ce)*IHX#71prInnotdsvoS6MkIIyxG*iasdu?V};BfpOK60=LK~4A8Zy)N!ZW z8YQ-n1imuQ(&0Vo%2W4t8w6GRlbNWVANhP~<=yEKGE2d~U2xt4QSz7@lrQJtz zJnw7LPftX%-M-kqg-`EQNZNtI*NZXPDiMJ-BkwL~JiD}%{YY+5IGy=VkcLPa&ksSZ z&-6+6e=p@-ZIHR{P@vKI+CyAze(Z;Gtm}_Z?yGjZ6h@^%H6bIBWq1eH2Q6JdaO)G=yai80u!J)*>NN@68|q%8OvTK=wg64RH*OdK*3XyHc9M=968cS>v7A7Xf9z zdnj%j+@;~`&?$*{QwL@m&BjY@T1YJD=o$%IrJZx8U&}==vxzYL#xrQa;$@7(b$b^b zsY7`fl~cd)_S@WsXU#XG)YZQYq_=wnbGG&@fJBI|G_%gDs^%C~NJhHA+2!D}pg4EG zG{|DYWRN>^Mq}J}yZn>q+_R_bVkpJgBnX*bVf5{Fsdz5%rYg#Syfm@ERp5iNiqP%6W&b-KfUNri|U~g7&5a>#Ow1 zJQ1lInp4H1REebAm1Bi0x60;82U5z+YV0TC&xlQsPp(y5Z1fI_yzk9>eL~HisN+&e ztAIG7d1@N%lvUHElDk3Z@(MW5qn@<>vY> zWtuA$9uZcypG}bA_LS_M>n2f3nP2Gt;neVKS!2aQ{ZrV6vr70otv{b!l+sV8&6Fj{ z|7Y93e;v2*CaBfuP+-<=lW^^>C)~n147(bdUzBu>*t4^*@v&!O(c~P`w~KB3y*ppe z&o(m>_h))EV)M!IvYX}4g2fXhG&kl-JvPW%LsyiP4Mg+@ZGMD)TmEqMC}SxqRS~UN zKa-cz$Zo`~{_ zBT^ijx1Y&5gr|Q4`9!h8a&ywI_*)5sn_}}t<*3%qZExfyNBZB{B=FY<8gP`k$=}$d zww8#2X_=1B%Esp54V7)5x$;!Bq6`O&Pkz(SyUZ@25WX~cPkHC<y{rNqtosV>vJU8JQT)Zv1M> z@yGX)_aFYR0fhJZLlPq5wV0$QiWzS5l z{>n0!m$EC_F+^`5GXRqIa;E~1J_)zf*rn*|%d$_PZfMt!597NMg@Iq~E$Q}w&euhi zzt?%KcnlGXnM{~?N*aU$zyg!;+NuSn5y#uNXAlK2bdR%$zh@F56riIKsvR^CAk^xo zVccsMC)M8lybiQ>l%Wb+bK{9@rre`nJeP}uq>OH`aj}M`kv^Tj=^wGC zzZ&lqw(0ig_HNGn`g%$f4)Ph30FL1z+U2NHBYAvBgH-SKZIwxo3X##%x8->7ZBOs{ zoljO?y#H!yLHFy)OPw|O#y4(2H6Y5mCxX^KB^~x{wP{6M!mdaB4e?sUOWpSth44KIzmhw$3#SkK-Ka{6!=+Xn~!dWo{fx=(?{Vz0EXn%Ef)8>xP#4rk{iJmg6_&kJ__XL^jW zIjevmr_DsGw)}*r*bNK5+Ug&&^Iw}=ihj^B@O}Bh5yY{_AoRFyu8+r=tJMLhmC>8G zN<=Gt0V43CzD@Kn1S~HIY!N}Vwag0I0P{cp zP4de%B@$qVuQJ#8P+G+Vb3=RqzZ$>l&aen--|D>C(C+@o^-{P69M|4rT#vz1Tys6T2B z)D{j(%@CbGRLJPQwj}tslwg40QsB%utL)pmcdJTd9UM!%6lKd#m^P%mdrIQD8Yc#Z z3Emp4KFE6R^E%{gR8!B(;_x2&LuS7@qSu|EN|p6KJMsPWRyQZGH-ss2pZ9U!Mt9Bq z*W1p%rz&Jj;q`2YV7H0}9-8(be(Q8RQZ>@PUCg!VBTc(|73UuB+>fp%M1O=#Y7Nt! z>Imfojil5@BIGnLb`EttI5@t)kg=es=pvz1=mjvS!e3Usb<`=HSDd`W%9|O9g0mz< zjynuOfqg>7t(6mDWo>P}by2eClGT>q96b}he92F z0)M+Ob`^kK)AT3#`|0zTXsSnC^-h$C^I?DhB&X?{aC|q=_|`E3bOWyg+Xgg&lP@$5 z$fC_XvAmviIb$I+QsB>aW#X;gcvpbjGpdbiS2V+Qf0Y^%WmNT=0@8pArNh7unN2AY zXarZVm+}yFxc{otYZR0@`DZ0O`Cp#iZeID-lrSs>jVA*G!vIh=2Z!J2NO^+2L`;FO zZf&3BTXVkeV$M!f`R1Kvh{CEZ0_Qk-6f&%{+|iHY;%;dVyuMC3GQD-M^xS6G8M0i? zkdq;8Pc-DVbDrE;b6?~()lT=oT>^z&Dbl9FbPsF$ukIGHnkqBdQNy=Arpq)~gqzc9 zieWfpXlAv^gU>yB(6#jas=JFz;m4Uz&3)^meZ}3h#g`AN9UD^6*v6>mieB}Tei^z&`kB+_i+fkXqX-9D+_;ftyNGCuiM@LxnDWJ# z855))9UkN=Q&;6u{dC@p*hGSwocz~x(ExW;PCvn>Z1aue-+^woo3FVG3P$vnC^57L z@-k=FfeJ+egeNQCs)L)7G7|QE@PwZ}eDGa)Prm##_F7uyHjdEr!ow_CBr|zv)*RC`Iut? zr2rn5{0=a++4t|?w>s|uBpug>cI%b_0SU&|AhWtS9+-0?z@vWghH^pY;R{uWseu{Vx2QVSqL35~*{x z$uQ~_2n?(#CgC26;#8OUdk5_apt(Ui+)&cyq~q&oFbV!vuQ2)Een<))HehS{h`X{B z4RMvZ%1$O-7~WZF`Bru~ThWf~PKI7Z6mS%MJ-rD@0RwG??2xkzKUUBOKsI^OviV3m z2c9v_i%QzZcBX0?pX&*}7tXf9cRId)&j+_`&~#Yknh!umTiyg=-a_%UFV9NI9BoZw@fsA88UR(!hFe{5gAxPdy33CE)(fuw<#tj_Ts-5Xb}WCJM@_Mc zZWbneY`cB@@>>CK?NWR`&-M3c3E9Hmzjhb}=N)VWw4rK~TbBGXP|U6_7bvc|xwvp@ z+e9@!-YoGdT2eJ7G@Zb)M^C8O4QdHJ+ZJh#>h z5byS(!TqJqE$!`&P_A2xFf&(zHEC+P@#(S1+lv?j^BgrSVEXopPgGRy)^}Vk$vENO z_ns;9uGpL&$zUF)OPUL&g@pWsoeYF@Z@@Nz(?$E}6gVYePPJ{q4L*a}U3jeM>FLKk zs{;;V{)D#FF}-*J0c>r@=-0 zcMuNWH^$?nHc%WdmO+^qe3TK&kf#+DQXeKjOCC6K>gMwN2yD{Wc8G&*&@Y3~1l%5c z^hnbr*`l{PwhgoE(plRoO}8G4_CGS?+qOyUD#F-c#02`^xrK#)e*ZRj=Q4+SdVeB& zzL<&Pb_0voWxDiA6tkK=`;vE8f6qfXefjOpA3%Em+D0hfBh{%^qmz>~(ecyRy||p* z{wVx2*|4DGB|1M7m%kTgYBk+H32D%ZWKg>Kir4xp-xVO0_0pJ3@|xR@O|r1d^#NJ* zE)$~Wc&*8#jV5x$$3t6oBwUXbukILV6Rvria_qO26q?jSPaLt z&^-{8mg+dz#FAZi?g7n@;bBLpmn{|{MkOHZkTg&pyFkSzg>Ymf?#B5_TSjT(PqQXuqg0u;t2u&^EE|-IA1J$hhbU7swc*SQL;9~sJ8drBj05rF*-he z3bF9Wk$|HEZEbBmJ?=VUy6?G9TB&&dnbNDT0QrT-E-x<+;cMUi{aP2#X&=j1S-axu zYLG_WSE*>fM#B|PPKrVyRWTe=x6$wK{|{--is}uECu!c^L)=GmTNYlMIk6Z#m53Zo z?VZ&4rfC0@W6_w+Pfj-!gHv9B5IucL03&$o>ijEeIy*aCS}vLnVi+4TByOykn?5eX z>9%}5Z&b8l`n!d0VJoF$JJ0EcHkjoDaK;TX11roE^~EzOE9-`$B8j&ODAV}(coB|- z6W1xSy+BriWQ%?BgtZ%CNYY~|DIp^Jf?2rKXy$F_J+{X?`gbal-g3Gqx4V3>mp!`!^VR4W858dAzyw}G z!W|fV6jF=%%dc8WIvm1iS)^_bx#@FZ@7E~WZyXURC4Yuhm>#I8upeoRD=f;Oq^7pO z#$#e?(a!VHvXP)SdglUxm&ncUXKkAhk-sk^U}vPVWl_?draM-ksqH4O&``E>az-ZL zAa)si)svG}CMN2c8fm-4=}uZ_tiH#PcOmn}q;mo1)0LyS3Bwxd|NPHXb$JEFm&_{C zJ#G_W+>fn|O-`Ny7D>%=$Sp1~zjeo?qEQ7$8YPRoZfM?xQM!hOF38N$+1pkV#Xjx1 z_nj$u3AOps#KDFXmfY8`d9_}&V<4rO(Ed-!W=_7LmtV%7RC=@9H1vW-=B4qc2mNl> zn|frNAdw8-#E{d}TPKVUGymtWLU2xA^B zr}fk;nXAUM&qc&uo2*nEKCp8xHZN}u%9L*3Ksr6w!i9a(+Q(?t-FOyNj4EQDIkI#f z4ItFHdfAMWl$2S-Rb$FHptZaBO0AAt=j?c$I*X_@ zxr`3Z&J`|m&l+pN02`McU1rW2b1)NhU zM?zLZHHJ9r_YfkWlO8Xr%oaU9X6ejd6;C@ur?lZN_h~}zr5t%ENX!Mr-Fa@^;s3n)@aF<0S9mFN=v;DZ zy&dUHdi&FIds-)RHhYP!5WF1J7p{DT_RQwN1LEnZ9U-Ltb{a`{eMdh3$%-yG(S6aL4ghqX~y5R^x= z0|O|$+S^}4rvSX;`u6?rneVqAE$6iBuUDCivbAG#4O7GqC&-91!fd5q?%@h~kLvmJ zn0r(k%Cw2$13Si~#nYrc_6*z0r=Lw5U8s+zdQR!Of6iqJv>k?b+26dGo1ZUJjdvD6 zp-_o&aWq1NoHMABYmuhmuyWD!ni&xb8Oz%`Hi`zQDi8;J$e}*)Tv?caO{?k;hW%l3 z^CNC8<(@mJ_oStz0pLTs1(P)Du9n88b@W7_VqH!@bVuzu*mUcs#p6R4C!(LSH8c?J zR5{THH){l0g3#&tbLd2Y=|%8J?X(AEL8Ci#2k95jIo6@HwQ{?5r0{A?f3d^vxe7Sm z^&~2o!#Q6(YD6a}9C3NwzeI?_mC{7wPlt}tLmELDgU-V{r{T%_`j!W-bPQ-eBOocs zm<3|6#ONi~6rBp(bJ$U}zbdh#y!leGqY~uLhAO4C;zWkCMNQA1^?``OJNVTre#e}l zF!Us+&YS@<6qKDXa4PrWbu%A3gU|Iua)!t#m|z1@CEM=ZB_$;+lcRPMurH?O9G5Vg z863<9GRhm-@%vCHQ&YT)va@i+I~VB-$L595WkBEt)kkyXos0wM&T0mPEI!h}G-)#d zf}sxz!@~30EmEmzT#rT6tlUr2DO{lq6c5Q|G{6jiqN2a(h6>z`3$wiz#?@6+nvgT{ zwEDD}HUlAno*E4zr+2XAaxy=%NiUL4=}Ng2e;MFx zbNIxiOpO}MU#9Qw$2!G5etaQ8TyH3E;l}j@1pBm?4?IiSaAPE8Zibe@lL>VV_GZ{L z2DIjy!y4K7%BW@G{eKVs(P3f65M`l9dANJ<-rEfUZW2Sf>M@7E7as4wI5A?)mT2Yf zTYnhWjE(}u-M0@PwyTSZWBi7z^%QypkebHunCxU_^#X#Hf)$W&jWNf02!XViu6lU?3vQA07P2%$ez|hQ;nSy1K@x5%%@KJ^ zI~235)slR+LbjoX29>I3R8+s1&$(Sk=UGu9*VbA<|3avc&z<|$)3Xooh#Q|gx7KT$ z=cx;i%~S0LBH9Ai0w4m>Om7A@m!wwcrl=Tz`?d9r=gfpUXhYszNj=fEFWe60F0RD? zBl91Wm&PU_uhCqhtF+wq?gs*%7{W;6jIuJnDTnL4U(|(Rqn%s08L0kZOR4FpsTf{8 zmP-~p(PJ8%oj_-!;imNX%H#UK0hOk+cpBigz;*wb9oYOcZLrOO1;H%8xl3L7uJa)v zIc2izG=Xae2tCj_zN;%#?1Xgj(ug&U((gTEa9haM`BHDhrl&6WF1$V{?SWD!FOO1W zs}e8v)oO~y>*&}{J4HlYAm3Eb4XxoW0ELc`C;&mA*DfP;1$I$o5bORWQA{%gst( zV&V8ALPkG)Y^;TAe_Tifd$8if&Scuk)%O|3=H?w32#8%txPc`k#rCH^N$X8GScc;) zko|b2;~+xb7fXBZbDDI917c`Fe(SN&V^|x$trA2|Uglk|s@?>ve{0yp&Tya?Qy=c_ zgW!Ws!1z51SBsCaw@v?Mg}oZ{pvQ#f{=PC`nVT<B7G{DUqa)uu5Xy**Sb_PT@l`xVJRkUwqIKO4+PvLZ{I>2oH1Jh zoAqtNKqcc!U&Lau3lXugG*nb(U|K9K57W@NpgXbQLZ<0K_lnXFQ|bW#h&`g5LCbaL9iXU`8p!l{59PTO!DswaeDT)fr*ZlB-l6Y?$1cx>%> zT>m`4({WmwJe1+oO(H{vlh(0v%a#JbCj$fbP>f67y?3t$!-AIxor|fhZNh{~lUY8V zSr+ZiQ=HrM`hF><1|E%AMdq=novo>96{GZr z?b^Ty3+#D)r;ogT{{TuzpimgQ@{JqNTKyp)fAZ=ig_Hpoc0#HNXobDOHK-1p!4_o3>U?X^<#Bv|D}6^) zxvMM!$$FT(L<~0mks}`;(hx;DlwwRl(ybYtPDTm0*o-fb9WOeJ_P5p_p9+bJb!Nm9U(|ic4{X)yEoM4?_()|A;qL@Ew|+aZs?q#K z3+>|O9YzLr64|Zq2^Yq7EpXwwPJ|HtcH7adihdig!^GVgl42!h2PMRm$v>PWTx~eq z(x`mh9HWlP?cB;Ng)V=$KK5m+Q3{tw4d40ld8WA7BsP=FodjVPk~{yduX3nv{>s68 zK~_!r&gd5ligiZU3Exw_Z40GKOl7P2a;kf3BC|3lt)}Bd^JbQ|FX4Rh@WYWUDzY4! zm&>CHJ(R2xXIdLyzUb3`y}sPJwE%L%4xyQ01InPo(Webi^U=>0yfzWZl>d67d1Eh) z5?yEXh9JKd%fS^P65sFP1LDu0tf|ojWRWpNsXmEXQKigOxqTv}f^McYaYeB+`+V6g ziQ^CHzU+~S8r~gh9{*0A?$7l#ji823z_0`RnE%rRBt+l!kg}(p44`7XEPjBnCJP4OI;c{`~oK zY%E=Hjv(?peq8F;w|ht}cIM{nTEhCx*|uR7vfau?0$2Jn^h&&r)$}QhPqiO^Vx}-; z_LfUgxpj;lKu=0b+XPvo?(O6U=7+yn@IKP->*x?+WAm#>^WNORYdq*EoiV=*jmX+@ z%q>Bb#h?XEC9bPkq3YdN?btHE|H`v7DLT+<|Cr>~6rg>+kKl|g-qT$dwwSv-+F+AMmiAJLNP#6mvtTja)`bac!w*{E6s% zRFw5*E%&z;GJK2Lzb_ls$O6h?R!00B{80DxJIo}$^?l==MfWQQ1}dPhSUc`Daqaxm zMI)6xzDG%<|4-BmT$%zWPr}M^+yEI1dd1GJg7BjxzT|%Y`kPDtjg8^(oO}(NhlHkP z8ytEL6QF*^$H(6S=EnaeB^?Dh2}pl$lrw%wX2*Z$TzJ6TR9Bn93EUflAs{AI_wx?! zfVMNmrkV%`vyg@5Fq@Os!gquhm+4>vXFG=T8>0GydN=-aoJJaYWRwxmz#kH0r3HnM zj;{c;UtNlSPz@L7wqIMC11kWrde66SFi3b}K%vaYFedfF#-D-t$nYumpYb}i6TCwQ zMmXZ#o!g1`!Zo#oKJRLM7m6=HfKxLwA+x?1292pd36h@N_aDqH0}aY6LN7bm(1nXe zdpz^~yN`1LkVWKDLEU~o_EOnK8q!|P?-ph#4^X=S*Q_!~X_eBviGe}^JxsmSfpwEm zFsDOLI@9$14uEF3Bq<{fi)jX72C=Nd#>U2y53?yifTaU0S-xhN--|a#ya7P7R6bOu zsV)s-@Sx)NpWb*mgprZa_&Uzk*D@O|MFyo#w(gx$>osQwhr2zI3#+k?Awsox^y%;H zEKj}vx@6{oyu3WnfQ8A98|clTM&3uw`~941!*8=2inOxxYs(HBlPJ;VY6IEkcVZw} z^)V}){$OASB?ZON7VxS2;ty)wOf<*TcCtfi(s)zW#rM2-uuZZuj+a8mcvc~RGO9M4;G`H4cNl!N*r=-`N7-*|JC2UFIL9<%=7;wjW zKNss!G#YisTycdmmcyHP&B^W`l+@LxPG&piBIG*j9^#6*lkrmChT{)BhT%RA2m%XjbT2l0W_Fe!vVvfZu4wG-kNJfK@M+1P<5ueN zcJ8~A7T@3BLBesvnL?6r>&6jAMn3D8?OHxTyE^tueF7#KUTbW_@J;CX2Mm__mDz(C zafH)}5cJpwkiBbvh)Z@qUNnZChYN6XcONQM1EmL?cr>d4L?!;RKlIwB=?pWG@08s^c$03`a&6S zd$L_g;^7@x$v$mwJUVDT(v+h~XE`6zLiKYC%wSi(OLoXN+FLrwGWRfANPXx#{obOn z#9g#=2J4{8MXcI=C`td6M%!{+On6`n()?adP~Bu0g=7zKzup*B{8U^Ta_@Ns$(y4& z)wjnCk<+L#K>)WM28i4HfHYPMZI+$%s3o$pwEofQi9k>t15pgEkf}^K zAuf?h(gTE(GqZ;yAo=&aH*jy&R>6&(2DRP~BX`k7V7iO+Q*C~{QUd;4Zvxm9MxbL! zqS^1o)DqM@zuRE@kQJ|pfiz+F-&3#<3ElKMbx9~DEbP7KhF1ALuvFVENKwDDOBqOD zo6=CoaX^g%$I-!ycIV{X?21mA{}PN|q=&X2Jy>$}}N1jjzcRoP6yW-e*(>#el^?hLc}PCtn(A+Y?K;&NSDO5)R#K?`RH z+Zd03jPcGYJl~9` z?Nj68XuZj!#Y01IPCFCHDMdIgJmqvOza;$<7@6;Ha2lB8ebg-f^=moe{b%;Xr)=&} zI%(Bj-`!1{m}k}TBDp`@(s`0!}NkgLjL5SMz7W%+~H_ODtcyefLK=r zQ4I5{niu^KikrTrlzV$mjBbD!b0m>(XoO66N|XOT{*T3StT`SCEi|yOo#5RdX1L`5 zo9w=e=N^=SI@csuhwxre@6kq$G-cX(W(MD*0sPh2I7f>>1>!}AHPzL;M&)bJFEZf5 z?-ZE0WZ{u`EJaa|_Z}82(Ic7L^Rp{Qk?h?o^nTw&g2JqDPQqQKl}p(d$ajW|a_Z(9 zgFJL%3kVIxpuyidMxF@adVn>t!uF-M1Lb|lvoJ*{G2O>J!nX=JJ)L)*0y*>5((;`NT(s82z#LcA9!%20L^n%dkX zi3-GAa8c04GPzPyV+sV}P!AD89__|_3B-*@1Ff(0k46DyB!JkjRT~|E*a2_TK#f?7 zp&z(6jdlcx8W8SwzRP?3A!tpn)Z}6@txjOOK_HwVeBx(DIm%Q*UKi!Y3p8(Y#7|5> zdgy-37zn!jOLJvMUZh`v<9BnV&U1%l0*7K3bwS?S&Jkc+=ZA)tZd&&y-vO0kVH+TH zRr6|7w^R&x2hkq%EV(S^fwKT`63z8Zd{SdlR4!lTL=?WikAm<&&S#H+<130Ah2`x) z`ZKkIG5aXh90!aXugkpkV-XajQkQrIsr7-p_XTlqjett0?9=FtJnp}U?35p{SX;_> z69ie2mmu$JeaEd3dpck*{XV80oF#_eTj@@J9x+hUJ6^z&k2}fm&x=B>@Uo6A;6Z7^gw- z=>-lMD9)$Aw${&x=K3d=H?WzD*Mn)Y*vu_aRAP)IftGID)x<6Fn`_ct{jfh!9?VyR z8@CH5A`bnbUQiD^n;Pv%e^LUnkI4d&kK$fI*MQXLwmzzEehYh`hwm^Q0epBA1M?CBun%%@Q!p_xf&4G9$$}rmC%&x>g;$2z@TcHAO_1K~ zwjkdF2TlVjXttLotSl_Kk3S4TZCp+L4dyYWKrn+!75u(|hlQ!GXK6|yXdgKGf*mlx zzCmct2<{@f-WBSf!e9oT7leLUOQ^hwRDhM~#M#+@8R6_2Xb3UkbUDTG#h*Z^@v6Yd z1`L&B9>aT51U$Tmy*kY*`w4d_O*jY^?+OSE!3+WRrHS70#ssHuY;>yrX(v)7}*BOD=|b48@XoEdNk0GVf9C%Ydf3p~2@)}o*go-F?V z|NpQYKoAUu3rLn*S z{Qn^e{8ipyh1{FS)ip!Tzz{17Z1S)r2>j^r@#K~RsK+BVAc`G_@;`=P`9XrIq^*ht61_OQM_HdoJf>PW@*# z`{3in^f#<&?DOMWTma~|Sbc&$2z}pkNbvFV4eH8|yxNfnmJ_u|J{5u&kPQP0dwW077dr5+U1^gD z4KWvVe<_V@dfRyai<<=@nSZ|Ad-lsqoDbBB%VY6f6H!IPA5u7e0gh{@s-*OVU+U!P zqeob;C%StX;7EdDz4ACsERpV>3EWy*V`AXQw3!>MFRncL-tybcA{vn4%!-UC{EBS2 zQoJ7RVKUQVmnt+-Ul$|ut?M1>JCz!;Bxgt48?2eGaI-R_N)b}P_J^`!Y6^n1Pe(@m!JRyGcuq+L&>mB zr`jI;89g?`R`c|Mw~^Ow+>bieayiH;L^J)bn3_rdiK7C#V=h(IgXcZ$SG=CBQ2<6a z0OXb4q^12l*kr(#QD~X^UjO-1Q6h(Q-OB#ut)~(qcV93>J|_K(t4ew#zp#QFC8H^!ITiq_Y<`5wej2cqhVl_kdMiwy+2l^A7Vv-6xNrY)gnfpKt z<7JjXLFnVwExkYpYrW}4(~Wk5Vv>9St^gf8yVJ2Bk^{DxmX?pC6*>lwLDvX5>;cd-wL-C&PTbj8-0KQ}s7bdbId zW8p(2v~ld$t|8O9RU-N^8U+F@Ec2UF2EXp}{VG4=+Ifb0@|AXKQzY-Cuti4x-*X%P z;KHYi?Cq9rus0h*0NKU+ukV)YB9cBDY?d}P{A3e$tI_Y@zQOolNq#q)H%(iH&a4{S z@xO|RImGQ>Vt@1IjTrE3fm;dj#G(FCECda=t?4NUqdws3RZyN;9O8$4FFPY`{kA^@nfkGG{o})uo z^gUtdNOU0iK?JFmVf{#q~ggxj+jq72`ys!Q-nX{hM;#CtShj( zDVyQKg05Eo6E!@JzxcH^%iX#<1s4Mq zV`WtpR3#1_QB?c>E@ym0GCOS~QbX`g{(2VSrd@b8`xO@!$~jz5fj~9SkS;M{=6TIz ze+cL|m4)u<6q)Fj^D`f<{#q0PKK>V%5mgWuJb|ap+GVlNy*f7f{?YaO;#%#^tLaXs z%dfjBo%QHv%wRVfNb|7#@Ie6V_$Po2SHd4ir~%{8SM(3S5fV5Y;LBFMTIo1*wzJS> z_rR6|eB8c(A4o7hjY3B!D9OKw0;5I*Tl4?BU&=N3AX^mRwz0AO%_$!kE{UYlO^LCy zFTp~jJFvG4QZwY<7mrtgwX7?oOqzs8E{`TF!YAO=uXR))gq z_!til&sO~AisGmg5JbUBi6Ck3m#S2mHRTO#!}udJ7{Xrp2snBH3|GFvd6y4=3R2I_ zn>U?;UlNiK5p@_9o5Tm4C|Z9qfv%n8a*` zI2?4~Ak7gt2#Sd%r*y7CO&OX6vV}%~76F0K%2&&T%9NX?LQMP`9gsm%#l6Vs05n{I zTEx^15*fS}LHA&}Q12j6i$`D<3GGGqH?XWyPT=^OUU4B(os1h{7Gr@K(^$Qicz->QY2{50t1M|hfltPSH#Q2JQfJEzP`6qokr>>VcqH;Aat zS43QairS5VJ3Y#iaF-w(ww6F=6Z|F9z`#=Z=-@-lZV%G$M~3tdNgF(^Xe@%FZt$d? zpcD05;ar zhNi%z!Ne_rZuh?`;`r~JpEbM>KoQ>_07tYrLpo@JJJ+Z((XQ013p`48uh?erR9M}+ zUEDQ&mgkV4?$E?+SWu(0V(*2?r6Qo0rN&X3^fs@90?#*+S&dsOpOo4OiX7@9LkKih2B z8d4>s-_c~L51S|bQ;z1Yewwe}1nzY}7+7%pYJYiiN01>F6~{dMErWxoSer zFx6-2FHK3=%j0+6`;w0C8RM_WnHWQyUU=c8H~gk8wbPOGA4ma3Q48!**YEZOP>B?Q zog;=iD9PQ;ZOJh(y)I+i1hLYC2cCnp@k07x(_sAB-a;A>5HOSnOrQ#(cM#~DZ)5kU z93r~-sm*cahCq?ERIHwJ$q5ze$v{{e@HOy+Q3RkIe76b<-6rhJ0@s-ohDl=-q{crwgyJ2 z?t<0Ut4uj&yvZ|c56*k3DP1aHx}yUd)4uYwzfGNpjzVdd?d`Q`Ym}&I7U(Nbva+6Y zekd}|K?DTTk#2WLu~>f$e~qtC*bhF%ogs)`j5+EYvsDFC4IJ8X-fu-d3Lja39jAVh_;IK5xOh?7T9+zK-4U>OkmVuf`3G zWNJR0@rTH7&!R?I%P;!n@QCd@n6Vw^M!q;59~vpC!kl(iq~x|VxSDxfE&9W~`k}f# zq?t4Y?OdVfX5K^l#hr9d-f1k8zM!5mIr&Hf6ibQ7DR7UyXR`xgGYU z2!974!!~`U)D%~|vv6P>O?JUG5*(DiBI&EUP_9$*FCh0_OmuhiZBXATMc>p<+qaVo z8{=FX!$;eS8-$}lS`_0XA+PCyx9{$@DU48q4yjN-ZN@tcNA72)q;`H43KytD2^B|= zhQ40acE}Zxd+r9JhJF7$nE6njzmEZyfVrYY@bRPQXSoN6vq3bMhH=wi#=*Aj#Nh_t zli`{i#kb7@W)))3I=1A#Dm{79q4ToC^A-Fe>brUgGrX6XK8CC~2ZP8#Xw(v>5~Tzw zuO$#0WZl54{t}aq@5D>emIM&J$3Y(^O4ZKVe*DY_TD;J^0Vg|( zkKx~*u?0Q8p|P+8yS*3>4S=i#CA*)WpL6tA2~N}m21~HjwwtO)1EY9mCjE)P_l;8+ zj4@krI716nThK++T%Gs&h3+Y!C}7A3_#6*LF+l^2&N%u1Qf4cE?}QgmdL8(#z*vMt zj$2G0+U8zBT^NN%GTvxx2ixIU`-kfWpy2SdB%K8QD;r?Tck`Z+1)#dpgI4tAjEP-;KBUS;uP(S*8~oBHwPqaBkdl>iAZJn$?Zq*Rk*^C92w)?#55UK82diSqaGfYya7AYTM!z)!V~sL~g<4;O zJ8P_;2P?=J6b@l0*sRPI5{6fAXPahs3mzc41i{s=0#d7IB$h{{2}Z&f_b-nap8}|V z@^-s{iWwJr?+!fN1LtljNw$-8rGHQ#Yk$^~-p(i|^Y{c61QnYUn2{h{>gM}aE6Jo( zR4j83ZtZ7Jw4O7*Rk0Wx=feGGHsf)e5v#*2Nl}J0Oq~_V1XioeiVQdG1>g2ji6S7= zqp#7*&rp93?h+3-u*wRUw`OrJj{lI8zursWEfv6HKjdU>o#rr8x_Q1(EakcP!ySPf zzzE&%o%AU}@lx!FzD5fMsgah6Iix!0z4|@W*Y`N}lPO&o@PpxHWfb@1h|0W~TlK?*R`1ttGpJh-t0)?R~fE$e9IyyTE-gDXiMin;OrjQkn@Z4UC zyXd<4)9Y;1_#>uSv){fA(p(sOZYkpt>Y|0XM*$~?ngXN(htLcu$Yb}Z zfhFQ}ZT!7bFUV|qmCv1%@q`U-GfKSE?p2_f^L|KwCxH6qv|tvA@w#c=`Hzt`^8ae;0Is_xC=^B{Bw`@;7Z(A8oj~aKJDT^1BBn_;9aDhO7#?Kr>j3`S!@-p zoF_lw?T?%it(^aLP-pJ06AwDWUGn)dZyg3`1uH5mL*9au=!26k7l1DL(`ts$=)@HB z(4WwQg5O_EBvjxi{!7wO-2V%1w8nki+}GbC)mvR=ap|ZU?A^V zJ73a5d9cM*s~B1SEzpnOzxM1q6*~54qIheurtM#e?Gy6=KS_WcKvblqz-txkZss;%_kN=Z%U9A z#6&8-x)Io&dd$|nW+M071|-A4OY5 zwGl*#i^pZzr)4*qFXFA`d{8X2LYe1MmlfutPT5;j0H(@xP7e59O{%tWVsN@uT@9+OWnzRs#%CciMYt^ zY_&xFak%@S`cCMYhXX2Xt-!iqA0fCDErBscGxJ2+F}IR5_S6Q6f~Uf~E0+rje1LZX z`g)F3Zc2Dl241HQ#`>u0THq4XG2rT+%DP+14X7YM-2)2oYcT0ij0de*NeB2A)Bb!A z@b5Vl_%uc~VfOB27T&r+Q_-T!`pb^)R>n4 zbxkm+^Lc7!8-_91K4cQVk<6bJUyXQEU`=cjPm`rIzI_MNV?&((#@ zF{g>_=c4a`SLg%t*f1`v$b_vX}0RxJc$k$f1&exaV z<5OD-zw#RpHFr!GKMDYp1((-eCtaMVH%?uym>Eb~?Zb3< zo*oIg36Nh}L7Nb+eBOcy8_krp`x28>-XZDt$!abO-!V84x~UC#Lh%ZnnK-at zg!H*)F!?W>$a!!S%UDg8_37`x^Zbq=^EXditAINaLqeU-D%Mdj+Hm#l-t89IG<-rL zXbOs2G5HT_3Fy(6i2HiJa{+p=sg|0Wy50OqIZ9jMAdtf*!4*(6)?d=-NRdl)V@AHI z+x|~Jihn2qt=!qk-uHY7^|k(VsTLR;{!_~K$20_yxI?N^E1Q=udv$O*lo1U6&a%dM zxHyi_PU8$MoT^1#rQAPSPjj|m7|yrs>Yht0)H;~5sej;(45W(@N@-a@nfJ5bBbMg% zZ?|^y43<*ir@oRd;*8^Q1XIrCkhqi=)r5>%iHt;odQHGVoW#JasP?*bNb)vGGyfzJ zj&EqGX8iQomvL3nY1d!O_g?5X0T{tqAvhP7O!%3|avur?C;etfxy1>j{@}v#uEFTW z1&`}+jx{;->6c7k#@b>|KunA%6B~Jf3-iE3$0+4E+eQy~rypF0MlAecS_>hX$Ybpi zYR}469*z4A7&fFVG98(pYN7mT42B}Df2<7us@gbsPn;ay9P`yScuPB-0Yt}&$Q-5s zN?s{q>1X61eQUzMzrEDj#1OE2N79>vOAbLZS5JT3PwC}6AG)>>@M}_g7agv{zO2(< z+BNhhiP89Q@?=p8ZcNMurWQBU5D?L3D4W|oOC>13x{;dzcJSu|(8992l95a-=7OYz z_l&Xm!PD?L<5wXWUY_w0dpBQ??I(+lUxgD=cpr2jxvo3j%CmIvL zx+DflO07~8>Bc)8viAK5mP&6(KR|)uHMvqA%FshVE9}xodK_?G)_*_GB4o{7;Z2RB zAjyf#+>U&AYK3CKW%NvT66MgE{LeYOvG(vGU;Zfz9E#y>Mc*L^a4 zgKB63T{oi(bBL1Z;^H>UiMyLnhfw9KCGg12uS8YoDV(xlpxq|!+EtWZB#lGs_?kiH z18ZR*l|mb@m-%YWHSx~T9hA@pe_^vJYmL|B?^jM!V@g?{dTrvmNt;JkQd9H74VLI! z;?Y1OkH{c1ii(XTCL)r5Nw5IF1XvkleQg&2zBMO$BUV_fw1tb^Hvm>XsZyMrHlWTa ztZ@P}ad3}=K*LB)9ey4`J}2vrU3v^_Qm!E9*emw~3v|oohhLPw)GoF#s=<;=A^(JZ z0-C|gXS&NX9YEo^VH67MwMGLk-;iB+^gpL~%Ed($# zIeYN#WRs6DCr-jk*f>0qSb#2z!oP{^I0Pcjza`_mr}XQPd+u(&bfOOK+dlrXT@*_f z1!+qzi)!^}PIQ$S+M`!Ye{>WEQ6dqh<*aksF+C?kDF!l@qyNaa96{x%ovc zM}UGso!=`~51ivzx&T%4u)|V@;FD&jli_T*zinTLSFt;1;8@YBX88xgaiv9(v3*hk zZ69_~!gU?gA3F$!?40~Vyz)_$AAs%-;0^GgzAb42r87!GNT5Q7HKB>>bhnJi+Uxm2hOFXjsrex4nOg(~!5b^Kq8Y-^ZCG;_-JNmVJ~8I0H2naFl)m zAr_j2GuMo)pu)2cZ~;8V1Et>1XvMsMb?BDC_A0x8_}Zs-jLKTfEvkj+Ce@(}v?J=m ze*$mwTfpE1fRM7Ps;V+9EtuFx<54PrBX*t&RN~^|;Msw#?znW?(Tw_17W;qimh^gG z41N5l)x&|SKex=)=T_9TpDk}o3G3N-sK?RkU^Z)w=$Js?GVF+LcEI%X*#p&w|GQV2vf-dW0do}K_1%L-L4Tj6f?ja0jZt0z zOg;Waq$>T;M{Z1|nC>6ZUZK4FGd$r|OM@q~%ADBip6n@>V@@}VbOYZ7DSdt^{(%4e z(RvoYgiPpJ6G2fuK2CcMyer?0aY56HY$Eu*`)f1^kDq)J;VgbmU0@B>x=*}OPFtME z$i?XVPdL#kq4nu`W_sU>%cb)#w&w5-LjZ;rJ;D{l)mWFQzolNV*b7WGptimC;RBe! zqfvq<^KYrPch*2H868a+AmAAbOc$^rs==abxw*7m>4F=`oZhFB8HAf#5 zKbFalJ<4DY=k&L4k7kg)SLZ9ed1M?d5>xxu${qHdU5K8@ZE6vWI8OE1)DFCNb@>=I zswf}#DO|PVFfhE$RP;A2L&vVEf%&2pOt8k3rtHcu;Oaw6(w*22`w)b=Dk+zP^*`VL zsRCHfn3E|`%+vxRfrd9WItrV0IADlLRMwH6F2f7UwDboMM)`r{Tq8_#-rHJdXUyg@3lKzZ;%v{CYto_j-kF4iCFb^>Oe? zOn$&vMrcqwCf(fath;HJ!Bt#E4oK_aR!7P_aU!^Zc8ivii8*s^*sI1TV7&{KTNYNjV8k;x}s zb>BN662Mae&QMi#6Q_3|fzxrE)5zdpIdqU@pvcu( zVZh_=U)Qb`ho_vNa&Cgl|Kx_*S3%d>oD~uChs=BD!c>qxXXbQEp zwM@W4Loak%>f~2_f#o1ql0cU{0$OA&?~k|lJ{LAuVp+HOw%eQ>?wcHz``|jHRboWX z8%W0`l*KFCyU363@1lAo0QlcmKT25-~@s`l$xY-vT* z6s*E98*sTlu;@xmMC1haCD~Co60FWyGV|~_z%d0Qfc8#!DKoMP)IW6sOJ-I!#{7%> z<|NIn5QZHWFjW8=;`+U9NXX~gZ6rUqA0QODKRdW-$%>G)#5+N01`zy^rwI2&ym@cz zTV^-L&xWO{kMcyyX%y)RYH7enrC@m-9nE{&Z*%&7fV*?haL_`n@%3t5as<@_xsZCS zO(qDF%-&ex&69#}9OXZFLZ7m-oS09Ne(W%O#Q<{zzzo4}cM~8Xsbr?A~Z{Rr!#Equb zkfor<>~>)+f`0eX(Jdq+qOHz)m}+(JK?0Y~(;a8uvlXSku#+)A{bCtKmb8N#^k{tK zPeU!bVWcM|=RtUFi0Bv&Mr=`Q8_+RY?Q!(`|7#|J=0%0R7(htZC$9bL{ZMIrUY^x> z-p&@-I%#Apk!^BZ`j($E$~pLDyH?f|m5MH-4|l2{yE8}#Rm5w)8tRLp!fSXixE9rc zU0e(`N_Giu^jq;PQy7tLxF3^7lvA+stnoH3^7Q^sDwVI13ma7gf|`;wG5a_z0+YdLhGXwc-?_A@>ELvZAHbB2JoqdaXu{wOE;rU z?fwOZj$Qjrt3OJOhZD{@CMUoB9pZfH4y>9mkQELnzPF5mi6hvmQaIldV!0j@8*7<| zw%-RM%KU3fUwymB^ucZ75x!qTqj>kPxzT>q&C{_oB}J!gk~-P+vy}lq1Xh zj@_wHdjWStU)RyZJH>FazkJ^U*l1MWH0yBjfYdgO+JN-iX^3k?{9$qaoSQ`k#VP~5 zSa5ffhoj{@2mX6H*{XTnEvA$tN8G51n)_^O`R_L?!B~afbM9#Ct?P6oJ2H9Fa?e9q zf*~()+~@e=vf?t;B@*)?=~sF72<6aRT`2BfJ>d4g+5PYZEd>ScIZX8K4C<>m{R6t* z09V^cEK5_1F<{P_YSo6Xb<c_A399Z-@R^8#TpB`1e{qy@*pbS48P z)IgS{yeEo-^~43@Z54F*>#qSXnRWLB9UPFlf#b6ciDH&UZoKxO1PooW45FCv!b~Sx(dQ;eMeuI(nZc8<94{OR(rd+>_Q({?JyOk_Xz;R ziNU$nGx)*uGk&S>fXlk~+{a8Q9PNA+RB6DEYK5kanIfSRz}L3cR$b+NNj_)@6-|k6 z@MQqvEb_?AEX%u$Yp4)d>KRw0bJo|^K-+K(^029`c$v3oKVM(a61l=$pH1OO_}=pr zPd!$^xs?B3*P3};R+Hk&%$4%bTTnUvk@?Xdkq$kM)BSl>myfc`X-4@he~pP_2KGqO zqkdtq99e*MAlSoZ9TSNqEe4l%x=T4uewlSnWPf2AM3<7tz|2wDyWalN2l4^F`a(+R z5za+E7H!`!wK%3%Axd-KVHEAuPYGj}a1>wyg+(o5;^F|S6t^5fiCb$}Uo>E6opr56 z|G_i-hSqCpOF1e@m1(feJZ)__ORc(oej&u%5ld57zG?-o+w+|CYzqgAj1Szo*Gi_L zZU?oHIq%v*@tIvg6xulkNY9EJk>n5KcaKrFK5>-Eg|R50A@lRA<*PCKUMPQ{`Wn9| zRxR+!bbk$a^WYx@SLMU;a@&YE-4BiCe{wd%3du_6?noPJR?68`;benU-oHC#nv&<0 zW(t*H51R)~QSD_?G3qfO-*okpJtHn-6m_{o2zNu4GEo1gA2Py_DDGpg&E&spD3Wvb zy*~u<=ZyTV`}~&VA0(I4Bt3=47d#c*P^%4+|y@3TSJPH3cs%{Yb_JG-0RbC}@VP64AxBK;@I$}Y~%t3bKX$7&{QY|d0=U_(yCDEj-S zmGXMn;ekQ$AjN-=Yv+91`n&N{wU*d@GD0AEu@ijQm-kY|*KfJFtm1uS5@30Xf;m>2 zxR(oH4YpuYN!bpfum4;DqL~cL=dqY(SF;G{FbOF+zFdhniPM@~CJJu=<&@ncB_%ac zceO4|~ zb)bc2a|k&(SWGhSuq%-HaNp39<1RB_Z{J1fnne?tZ<$^EIZ7oj&1Y7ap>VnV$~*kO zy3eyBE>m5qIWJjHN*lBAK$>`LGHtmouYYLzWyHI1-(z#SRR#r++6=T$#Sbd7=iXv) zi;9Y%F#mlAGzg#{)T+g{2;U zzpk99_xIEBckHX|&~g>yx@?T5Z2sO}jjoSNA-JeVfB4QM6P@{TV|LaFMiKA==6%aL zdI&@5Z}`?4*SU2;qLbmY0#lZU-wej(_1MV4h$jeLVtlUWChmXYx%%9ZJY$j=fo(D- zy#&&A(i${?Q!jJ6P+zCn;kFf<%v~(IeB$DZW4Qhi&)fC#lEcmKv_CN2HJ1oMw&(3; zLG)BhlM@AaN+XQlpiJ@xY9|DEAVn9Jl#n38`Rm>nqPt>06*6c=6{qbu^H*4>N)~*Z z7gPPRN4%y;eO%ho>3hEkw&bph;O`b)IB?xbNtGSKsHsT-rb$CwSzfNuaz(*l$XS)2 zleW7NCQ)kMN$x}+A9xfW3GzjsvkD1SUo-8kFk`pax4u6lMNB!-d^L{eudHb@rH`a= zO9u~0J)n>}g81S=OwTT|KH88Q#(-TjmJpWE{NM>p5wgwcReo!QM9F4U9{&Ex`ly1& zlSL`r`Hitkj`?~t&2bf_$PEQ#lj^w!{m}vP!@L-Gija+q%;3nOsC!Dvua4$BWs4ui z$SyGxBT>8twWLvN(5*qvHd2xuduXo#OKyM1P?=tC=CmE-P(&c!fkFO@vl3VlDl6Lt ztmsD`+kgJpfnsI{S1?^ir$htcIN4aK7~HwkE*VhWsaJ}tIwOH<9{9lXw~eGn!!V2q z|LyCl`dD&hJKVj)JuBdSIc@E*fEgkzyvs2^&d$qQTU~vG*S5HK2-|jHnb#-;>bZ>! zAcsH$`R6Kvt=}ZO>GkaBJet?cei7Z9R5^CNuOmm;W%Qy~SseXWO#!#ZG<7?>xqZ)u zx$G8xQ3$xZUy`o8fJ7yQ1FeRUk#RNR;vdr)wuPcc(K9%Kv#>t%%`0BK-EiuQM=#Ik zvwA*&RgkjcHL&8%Z{@O)EP1$aQ&V|;tCwH)&)AK?%)-%Q&mCpjk|s(yDT!Vr!kE)N zSdeJ58NnxdD~O3Zb6zjatMZKwQA=K-k-#$x8l^^_LT0YN#KnyVTY?Dcii4*dO$?p< zrOutmH|&tiYGsVAU02*-J=nBPoz(dm2V^UkKrjX~8u4TBw$?3T;IzcwhP zn~5TH!Nu*@YUHK zSdANgQC3n{F3oBM1~;svW%b3y4zSI{FGnULu`wu|7X56h@Y<7MDxACzQu;>g(sTl> zfB%rsT&JV^*};%ZUL1d)6<@<2z8df;XmsbZjFjFmS94y{`-UocGHk11#+!^V(xlw- z#Ym~#Pg8wsocFuL`eCG>*J;J~0o&O|{GbGI&Po@uPtL|I;Q#aI&-=K~a+QfXF_Ro9 z;0Uxl%TE9(K(CAAfOSU1 zk>_L7j(pBl;*U{5N}RmzR5fngTUGtocz#8PgL>|OWp$rVaJWG{ApLXT9t*w(ad&lK zSxrrg>k9GL3(BF@u$ad>0j*Os+)4F@JXFx4W4bxgQY_t}wmOo>{kZ?`w6DMB*)RB@ zWvRudBlsSnD;fw!VQH=;4vepjX5rpy^kbq3zs|9`7iBf$Q-}9eDI9U&cV3-G1i(lh zkI%j>O#iwv?N3aPVhUFSC|`93G_A-LHSf@($rgT4J_!;f<$qEO#MUcUuE3T*TK;GFzO=NoU^4z_Int7Dya)Bb zg9A0u@O+8vVggO9sG?))W`2BD9Sv@gGd7l(Z~C1D_~rk zb|PaS5K#;+Y3FA-DK{c$;At>Q{zxv=p&%6e_0^c>^5uC+qun&s?{Y~sd#mBCBr}eV z5RFfwqXo(Nf|zoG3em7-f|&k|<31S!1DK@{vFzk4M0@2Jr;z*P4&&7=?RA(db|0vV z-O!sBdt0c=TNIr9qoa9F?@sP0_7p;d0yq*-@kMet&DLgSGHeX!=l4pItFsF~erbOu z%M$OoJFzwYF(xigSw)=2XynK8Jt=))O;`5tq>4J)6+rplN<`}=_Esqk@H!1Y+3DIop6Y1kXelZCe~K1f$c=X9h$u3Mr0 zJ@r(SYs0Bp^OEVc`MY$|>(HQJNmH+we}MEHSevhj9?|Iaq5AbXP;s=MiclPLlQev4HCj)@1<6Ky0}zxVl$Y-vy48k*LpFe`f)?a;Oi={BW^l)>v>JEJG4}#@VHm(^PD&PWyL=yyQF?mrO=4^hndk-iz>1yHW)rZ zUMKc<%ht)&wU_H};?;{U%0?!J1lE43#K=W6MezT?U_Ny;1|?B`{#IY6%Nor2?jw!n zGrf)%4*d%*y>2q`N^HU(zj#ded+KaImEB>%&$>`qfAn0qs&Wzo(ZUZH3$>Iy@*V4J#b|p7!(0!+S9` zHF?o~E`P4wym(DofY=!Yg;fItP!O07{`%G6wst)mls{eV?XP2$naBH48xJ^8BydH!}da|w{&4Pw%O;{(1z-Q8dlr~p)96c|9Z zV_-J#17D#cE_DAs^m>6{mIY8cxBzu`cdM1%x7**}hiMR4F@m9r5eV{|>>FW>Kulz4 zI0fA#wN#}09%fRs2V*}~<;h%icwZxz%!fYqishK+nIzg*mbHE~_-(Zfo8Bd9w%1kO zQ2K+S6}M#0SJK#KAF6vm1krfcDJj7br@p@aBL@y*mWz#ZlIGIp;^Ghp8$tdBoi`wI z#FLWGx=B}wVaN%=R4IE0V(a|MPY0vZV_<{sk)IvDD)Wzd%=$NXxDl*+x&AhkY|#|k zD^VYRZ6KqKye>_8&l(kM$ewr?C-9-oJ}j1pjwWA{kAp+W*!cV~;O+%rgUs1Nb=jH* ziXprX;K!foC-XZ_i{s)3x3b>aa_@Z`FlMnH`$&brvzbsw1f1wLjA+mWKAdboy&CO`<#1$m zXF6I%z83YnRLb>XM1YJpNXxxjuL)X7z`!XfFCrkx3bA>6c@fczY3&CB$_@sde-a+O zy@k}%rz>Yu`Z*%}|O4kH=hc;s%2>wQPL z)89c=SVvTx_^PYm(Ug_wW(7`{n0l~6I0&<)%Y8O;_uQH+`;F`{1`EA_lL27*pv1k_ zu?j$Tpbw}K&!7xN<3S#P{d&JN|5wM`okPj!>8liQjUEKden2$VB>8XqhlQu}Z`cXD z4o38d6E?Vjx#k3Pjs3`GmGUF^>TX1J(L$gfE$Q()x`|aY;h7)5$9(utZO9f=0zG>> zAuPJB4$2!FGa7u(%fmw?C5&oAMK6n_e@fV~4}+wTkPsk4y$9prR1x*L0h!EF&^^KN z?hv+?D7ns)RVZmy>|Ubn%^*pd+!J*=f4D1ThWD<$Bs+RasCEq1xW1cyU7GW5dl642 zP@=fFR8HePVRkuG5(uPe!i4YXzpm(vRoPMl?JJvOQi_S`Lbat;Z^PwXo1FkKS^i98 zb7!<7sl5KXNq>6FWP8`u-!Y>om)e8PcQs@UN^HsG_uumT?VOlcxxQ=z@uGn3Wqh_f zi>(_&Tv9r()DMd6gwZ~nr~`P|*?BKGW#$HE17PCot?c5$-$*N`pujn49E#Rf)Ha%Z z#9Zaw%UJ1CR8H=%1}~-`@qq=xD?p4YDnvFmHY>WxFk-}t0fjfANIEPtjA?Fe?g+R+ z5BxV6LTEE$?OaVvUMDAW9ujbEYa@F*Vbbw3Ctbql)|0U^bG&Q=SP%Sf04Fd57-R5W z{&)Wd2RTqWRl}QAJzp_P{4t{`6R@khhs*ur%N6=z=wd#GUy;dDz)^Uud>M%Xjia1g z_rb$JmP%+SF}UpX(aA#g-i`Pma)iAh)|a1pd)I@9I6&m=#=iy!!3x~KzyN45BwKgX zD1`+D?SJLQJX-_kXI7LPwhd8J@Bd2G^^$ucI5&b4u9qWVqjtTn-4IeK@DF($R11H0cH(1iT)36w zfD;q^0Zc1yJ@NXtJqNx4bEc+dW;ovIusE&q;5SGI;hlw_Ktx0Y+=11lCHbuqx$wxw z&m*tPkD7e+S`3gE!#&Q`71sUD1%5=F`h7iCm>7pTdB2Yv8|alU@01?_EWSw23`U9d za`Wm$6kbbnkC3N}kBW?oZ~7jw*&vJ+)@^lcGKv;@RZB3AHpE+qKP%?1cDx(O^fF-i=5QB6&{UyLwb zF)&y!t=)JbHj671xZL9*v1w*R8Ur<4bemwHGxz&PW161%xNjAJ1Xq(^x)btDMo*P_w6Q z%IkGhxeJYOxFjCK;gTRe7=98{B}5%5x3hY+iPazR+kK$?NK#RO&5}LQm*vvoFjRuj zB(H)MV0}Fd@LnWvU;3nRMb>L+o8DqS;LH?Q+d*9v9`?6c-TjD#Pcp9fi|6e9+i0P9 zs=8EIshNM^&TOQ}NKcI~=&OFI;33qdX=ziyhPrx1jq^qiSIK4q8WuUtMINV}zo z^gF$Bt2kf%pYP6}Z@Xq&XRN%!I(z+OgomLG=VL47RWUL(ba)~IEwwkg_T@S0o)y8A zs03)if{u*7p5t&Z{F`D+*~HMcG& zEl_MWTW)O=;%6b@&;bw!E;68c3fNnYQM(?bekVAyG3r%z;dF&Ch0YN1NHIatvkr&W zxsPI*9kg2y#=fz}#Vl~wsONi>1p86&7eC*^;J1=@>lunIrQQSu#sA&f#P>?bsSfPr z#K7?%e!5$A50$i6H7*R*Wv>_Ybp&!kEpx($fi_~veQ_eJ5DnR@QSR#hKKxb|nWiag z@YSb_`g&9V4GZtzucw|isy^BZyR}|lAL-i%O;OhSwU?c@-P-J~A=vqgp~4T_B@iIy zPl}SrYbDXaKhOBG?<-m+jQXyk`s)Z+g^BW}*S`@G9|@}A8da$WEB#hNu_7T%OBaiZ zCHs~Vqq!{hvmLXo`e&>zySzoF*gogrbE{48M`lj!j|k|xTf0dnTz8xgNm!u=%qlwVrud|U~{MOQ{fUa{1{kswH+(gUWjWklOiy{35V(40V zr+3m_;pzn^)9#v!NR*X@1$fsh`ST?nFD000^_cLK<7=@> zPqX4Czw}e2-KsYF)hkMZP!n)6a8xFTiP{V(4X`Jpu2IZ8p}bzl1O6JfA~E4Liw>i> zwkE&mpDr#Arp4!>@&w@ttCZtwoxLfKuLLtCLk|cpSRm|YZP zyO_WBJs9GddoR>GA2qX9CQ>y>hWl{5S`Dl#mu$M}VMAbrJe_x7{I{|lFe7Oy!f@%% z?Po=sAjt%a2VG}Z*Axk#YBpyWyaPQSdvIL65>r%c%Z&{CdOlXu&h^u&=OK2{>wEsc zPM({v@4^c`>lA8!5-YX81_Av&F}`#U0TQM9p3f9BR3e)O;x}zY#jf676l|@8gA7P3 z3p9&2`ne$(le)!&T{`e zRE_mhO{5uF7q?LR?>|d-*|2OMFlW4d8M8)v)-1v9hVYL z5x`^tuN%-5PPXS{>*P3^WX0k=wc4H=^}m0F^c!=5Z%Ls3kcA+k(yD0jQMBZ{S^Sa^ zM@L9Fd^l-I2%6C^=rurPiZuF(P){$0-=~`G8PbXhXD+_2?Z19meoQ$y?qKI4 z9|ox8&hA-!P5KP^H;xc(QpO74HmdXN0A(eyiJH9h&inGC-(_%L1prGRliW2EUUoXQ z+L>--UU26Sg(4?(Xg`>5`HL2?4*!eZSB5K6{M4$JqW*hAggYvDP)`I?v-g;_80Ky4~C^B01N<zb|W>?&69$UDL~}j{2C|&=RuT7gRwywZWBBCoP1h$ro+0*U;3+EwpJN6o~VtW`q*8{ab}@jlan#Z5FRx> zs<@bAeZ51h9X!FF3HMr=?v32nZD#(8iRmVk2SI$q{add_1VC;8s{lbe<9&L1I)Wi2 zBqa8!@98QJT7BIa9rHhZA*A%5VbH=*P~?L-qf8@MrSGHK@4QkcO2028FKP2IqpC!m!(PK(nspJII+A}en!!12cc0dH6aB0nc73*z1am@*Ey4Z^?^))u9XY=i? zDmUL*Y5edMiMs9P;Za-+bb4KEpwI`Y%&$vl-<^Z(wDRKOx3nvO{?xMoD48;iz<5kp z<^p{HymyL_638Agzn=L8`K-|$?!f^XjX)PH75`%&ycNxVH49D)G?{o!-SH3#2N~5^4XtWSA+31S#7^9atWq! z`zF1MiW6Txa6NegJ^m86=tC$4!d5ay>i+!G_nyb$|F2({kPoq+!{0AV$^ogr)kbaQ1=>^0n^%52Fn;^igK3ePo8cpbt z{h1t%m)PH+MsU-OUPxTvJM_=-S}=4ilbN|?>(7#)1u^N_k3kK;)Q`n#ZLj1yTab5c zrGt9keS;eX_QARzYiVt9j-->HjLKmTPAD+jEjE5JGH837It1)XjZVq(B&P4gxQ zaM=!*swo7-XP8y~yL0vAtFe+%8m=q^K;|cV^U=z}coJ2OO1r^>!gp_82XT|@O)cqF zulTraQ5i@!ei_S{bU<1Ei05Lm-6a%ZV}^gef8LQO>K%%D&Of!{XL!r$)55;8Ts_^; zfZSMD)>&eG@^M4caR0Hsc|wLSY-eRhWgzU^JRnAN%9SKuSu?22hc7`=VHeGB?&fEK-Y zU;aZoyv7gL&uLr`a!N*=? zlwHk{^%kkZz{vUTE`DOgnmnYR`E=;7y*ON<%x_Nkz)$o884;?txFqqtYvJjLX}$zjpD zl@t6>mq$I&BS+;wSp9As62EAZZ*(Kjnf>BkHI;;3vLw2P5iTu74Qlg#&a7#%hPS&z zziD>XURu=%T4`;*A{Ojgh?nXQkDQnM(g%FWUWP?*yVyBQKk-yeipQ^mk}-pX+gu1b zi}DuYYl|)f{rgn-RFug~+yKMr&tHJYhJ#KU1g`<0)whSYfSa59Wmc5nRavqFO090l z*ATw5Z2?RVZo$FtTd5*i1DFgW``rs(zi)t{CL_xQOurXLJKTK2T$LpL6U$WDtoXG%tZC&zhU*_ z_<&@7Gp-8M2Ka3r!UDgN(dpEUo7(w5%dlA#Ij?JTPrlQ8wK zDVs7Gd8mb+09+WPnu4bVSo(n4k(ZypRSTvg01pOJ?UzbAt*ym1s>XXZ1?u&;vc61k z!)d-H;LpK=cPk6`^tpKvfh;W@JVpQqJ3tp73NJun_4=G9t-YpnQAZk?qO?GSjiZmu zpw6S`>E6_O;p@LGk3Xqf9olKb_qUHlL?sU`n~3{?RRtzeS6`3a%tS&$0!$lyYrG_M zSiPvQqIuM5s{nbOos(Db?(fzZ!beoeqePI9YI(2`f zG?r+A%jSb4hb1U8P$o^$be~9`KU*ck6fKn0_E6i7p^Z2;>ASHSK3>e?a((46=D5D= z)TTY_iTHF6DubrDPCLYa4PyUYvhIscsQn2z3`ThSb+q-)x4ysC;84p}S1TQ$=o#h6 z3SmII*mpTn3myxmgF}6RSZP#RGpR=ysvv0@nz7QOzZw*4p*7MH(Aa&nN+tQ z#iY)yhlsM6-D_vk{j|kr0IR7trvpYfqV$y-6?tXP$Ly>uq?Bj~{X9H*$!1%i(}n$g zu~@k^nLa>p&o)^qfI8BJQ&1j*sEb?I(t_(yvZj`{)Mf0#lCprcYd=QF7tOpK?dgx!-K5>t z9#nUsC%9e!5do&8pbe<2sybL}ZLFv`1+x%RCon`2ji_UYQw68l6rdT6l25Rm3$Kl{)oD05I$mrzi!H zC4l*5$~iLIj%$7+^p_F`CYZD=0tCr|LUZ*9*s=kp_{%~ZZcAqmt9khG%EXuyuE%>w zXKi-277VArXakr$G=bmm-=h$;B(>KZFshl+po|5~%87Atiz_S7_V%zbF!D-Efw{L1 z$YlVW9$QF?$wZ1|5TJ&E2q}WLAlhcW0|u-wtPhF{Kx!A0l>7u{l9X2n>m~&n4tE0o zz9K;0w*XKC{FJ<|4-y|D%q%QGYWNgDhk?&qS0DsFKK{t>muI#D%Yo1k5Zq65b?f#5 zomNavgqcCRkEBhFeYE@n7(v2-5u^?R0>anOv!f&LJUFeO%7kx@`@jNM>^biTW=*XzX>s>qyQGL z{RtePe)-kaj`^f^&HvMo&0BeJI)I(Yzx?^T$rn%?>@6#O< zXokl^`{G{*BmiyPz=honXbTK*dU~+1!r^rVtT4@Qs&}OR^GgBjlt{q44hc48`N**! zNY}qFiO^sHGV70tiHbYw9t|P?%o=Z8rX&!j(ql4|PDf6j2Z}2aA%FUlmvYtoTaYwx zFn7KK?YjCI4jS64z58FkY@1z&O8ft!6AUsp*y@TwVVGS@fA^D~+=$FT#UHRaa4GATZ zV3NS`xgA4>Fp7(YW~i%+l!HV45?f=l^eE*cKQXATb&`xN)kx9cfc**t3)Y5M8}_%R zXlyF?z4i(;sFJY+F7>p#$NO4fvI=C<@wi)HAsrmRMJ1IV9v*^r5!~rw5)5qY!u)*I zHos9Y$Os4C{r@}30HiVCpnegPG+FwyoVIju5EBJP1Wn`px|%r^8{6#Qpd_!!t5m%f zS_TH>G5W>NCBTXkv^+nr?w$9WLCjxqaWR*%(Sk=;4GQw&&o&dsACpuk-z^uk;WWSLmxY53A%%NKWAet*lK|t%}ZU=HV;wjoxGGINJWGU&LO0%1SsUHKw9YV3r+?M0ly?AZ37Es zpj>!499zR-HoNBG{;D7!ytW&-r`6R?zRC{ifcr&LDEnpr*hi{lvACs0U~7=DX4X7R z=>O+Pb-?DPq__a~`(S^OWBnvQFgAIazf|=cDG7ejAsRXP_h5Aul_Wr!tEgb~r3S+qj6YyE&!I~f0LSf5 zs=l17D?3QTdn(%8k0 zzpXC>vo2`p0?7Ahq55cC>M}ATz;+*mIspdLrVB8jbp8}w1H0{YBUJW_wp7ShiuVoJ zy`V3=-2kiB#pMbR1HjY=RnTVsySa&pGuTC5t{#KN*`|oL0Co63XN6c(1hQTLI(Y+d zhXUHFbzu9!#Kbw#FoshaPbjVpJ{0u(82)?PgYmj{gkVAE9fACdh5e1^xgf8T1GgBe`@yu~gsr#3~FgYO#zS7$x_Kj$GM zC&$jo2?BwCf)5f%;@ANUb`X_fViIi1`JG%KtHvAq-OM!DP8cII=y-U%0KEZVvtXii z&I5)xKrL58MjPU3@Pry9JXRtTmG!#_eByg>RMnxhk~KLMlINEn(8SO};l$*qun-h8 zxpS);(Xr-?F|WuC7WsM~$K4$#*^s1qwe`%JP|6Ch5@}lp$jiW|*I!&l2?G z{W4lueC62&H~?U`0|yG z^SeE7`=D|49D~BXmTd3r0Fz}v?;FZV1(*!5k1F_C4Lrcm z5yFxxF@}Zw9@JVy0r$V4ps;Y5Xo#`&k9-o1KwQ3Lsi|duPfr3fUF`QuG4WM3qs2vW zIH=b*@&ENyhL3)rB2ERa5MCK?Fv!T5o(rGN3Su*y2Qen!;Rka{KL^Wl z^Y)#QH;09h8LQ!`??pxH7pV|{8Y&zpL0w!Z6f zUV63x#>mYXlh6klaEo$tw}9@bfc9T(W!`fP?3t0Qf&tIJc6a|50HL6iX(;-K7R+GS zk?Aj`rj{HRH&Jam7@Z?OgCy|-5k8K=;OQk*!0Pk=CHNXS9~RZsv4c;a_XWWLkUN3N z_PwwFg(`^2{$G0ULcZin+(_7mWV9|s65y6ttl>0lde@7+Y9P93eu>S+Z%z&!G^7d_n zrXFE$Ur`lpMe8muw^0*pZoILmNsBAM3Nt>CPy_Vo0C&odAw zK?QK!jV2*SDF7%LurfE@qL4LSm}Nmh-GKQCqrmNPswXV`N-En{3dBZk7J?aB7ZJ2F zSb^OE(vhFGn76mL%gW1vrWX3|#a2AkekhnF30Fh#JO2Kqjf#po<)fj2v-}OvN1@_- zfZ2v#iz~d~F+h;*zbv#SMav-Ym*nog`iqZW+j8^`?0n`{R@}h675o7*9$Nqwg7}@6 zb01;bjQy>&6dQpouCxjG!j8q_Vq#UfW!WhyFwaORDAWYugy;6=W@chc1YM%dKu27# zz1;58n28BIF=1$xEE0U5?k@71oA2waXNM%HJXS(hwulko;mHU-LkZ?=fG9@LO)agg zuvxfaD|e6bwZQ`~l22PwbWnTvj^QlCg1+CXRM!^p@onIJk( zz0x2n`?WcP=U;McMW?sG3A_PdFV}wl_j|bK*vsE`Y3S)geQbd}oo(a?fQ$igR=Y5% zQPljFK8y-*)9QAl#13Xf0f-G})P;Le^5eN&gJc4U$uRseFbJlCNtXMw}77AD~ATFw>ssu}79t0?Xrf-;6whRT;oW>9s zys_V17$!xNPKBg}Hiit$+Zi ze+dM?BfP#^S%P$;L}BG45Elk2>kvLV@Vr&y$DTL>bR_TuI&TYaBp%pS#+eU-gcZ&P z5ip6q=27#2KM6y7-AwmlpS%|Vnj02E0z}%CK(;X@Ev+TcdV-y3tur9Um3})m7Kw%+ zFkKKUF(U&}0}y>s$}oM;!1^RwT{-{tsYt5GW_V(P8XE&-eeN6_NS*Pr0F56A2mBOW z+=S_Ww*|Yf&=o0`2O#|Jaft%2Ks_}_n;>K$tEOl_+42$qY(_23C64g4Z?8OQL=m1! zZo%CZ839xx{FvzaWV^VoRtq{^$e8mFF57+Z)FH|^>N+~< z*BW11Pk<+a0v(+Z_`F)VnI{{+yyf-?U|oXQo7POavfkVsg8@!cBdETa^jrGOAQa<3 z_E>M0Tn-Qg%jL3*(-Fff(Ivbq6&*%+PM1qLJ)i-CkgQas=caQRsWx*8W?*|GAkYq2 za?}*c==0!a?%MkR5V>iP^ca-Kl5})V(@dA=B#vOrF_R1$^j-mpb0G>JC4o&OTFNI4 z97a9QESMD?mg-yseZA`jb7Ny5$^8u^lG4$qz#Z32l-GF|E)^ zo;>kY(uy@$VpWXRLlFJppg-x|VFIZQAms$;MU31A?3vA#%e8A^R5s(KI!_Ms%Q2#4 zP{NPr)x|h~mHrD49ttW`D8L^KLqIX%9P&-gLJf&q6rst*HAW;yDFazpYUVdH)bQvi zh*3mH{u&z_yP1oWPEItfDf}7C&>Da{WyW_Pc|MNsGXIZgoU-k|5F&`x zg$U;MW-hMI=Eim}Upbh3Kwx1fWhQ<3f}fvB!rIo=+?kX~!q(W;T+H0m!OR?iNzUBf z($$KTm4gi=_K^P14|`^2>8wOk4Wj$zRB}(k%xf&Pcj2`qX-u*r{%)Ou-l?yC{JS}H zl$};FB-lK_#+=DRr2nN&Ox&W8pWMF_{qc}Z!2jWr9QEmG6g7UQ_pGzS_l``^|7kMj zb;MjmpAK>7U*JpdKuvdj_v&D9l8HnR;`iO-XiOVQ)(?up!F?vCfLP_{$Kt{L&Q8P8 z3ulJudMp*0@&?y#_MSQdqdYEapl%nsMd^Fjej5){Yrh~!0ck&@%OQhpS35o zTHir>Ti|pHu>x_!-kW@LBr0_pDNq&vm{uQoF;bfjF~YH1lQup z{cKk(82fqIU1##v(45mNY}kiLaY1&P^OAQn6+N#UYc6!!YVPWU>HSUn2CwFR?XC_f zWQ%=*13$J}vEBvyxA=+s*=Lro*l#CmRYq>ScaPb|D)uvK7Z_noeW%KEQk7WY3ilcY zSnt@5_?4|u*_7L|cypSvbVd;((|N!GTIUB1_3EQ5k5$1(>ZVxzo-gi|)5cJIh# zUau@AB3Y$lvp{X{;92N$I+-1FktzsECG##sxn;%uhDmgW%(H?lT@9pJ6@FLp268IB z>s?ItLxW>FN*`;XBJ%3~to+&<2K!=SuO(2Ysb&|%KauyGy|7iZ+^7rs5QIfTrUfz@k*>kib+$W#rl&B@~HN(EoJIehLv=yo72h8es$yDchhQC+Tu^w2FN*dQqvm%@ApluPdnnE#DSm);cy| z_DR^&t*#~3A($u+wCAZqqYtINawV699H?$`*ChY+N^@m|2d0k&c4GK*S`WLi<9S0( z48}@i8W(AeY81u~gVt&vK!ik_a6HLJ3uRE-wJ=<5dxwTKof%xTH6zMaPlsx-GCbNwOk^#sc-%RUw$C>|PZukGOv%-Kg_QbK3CxTm9)EQj{|f0M(pHch#{Y?5y3u5*jQ?JLzW(0cPN@A* zjj_vTsF=Q^XKc3h4JEexEs;g~pU9D-kGaSeKkB&+ACAcEnP-cMH;vM`=T_@;&lUGP zR~zMJXhRZWn-6GP;s@;BH-}x|JaC(9uSxDOk~aNJE+mL;@)^)gJ>2-1@MDVlx8NJi zIrQSwBbd{Cr|#I=L%azNrW)?Pe20#1xk|(j#bRSe>6)$jTvfH2$ODPyN|~;P20|8M zG?%a~e3dzFmc4mb_glrdYI}B1V~aL5Yjpdm(Rw7N+}$gF7j9$UhWrE{%U8;8?8Y{P ziEf*}q~z=Hd8BDo&e>eHcaK4&*&9>qUu(-~`u~vJDe`G0`UpvQ=sbK^{E0J@rMPEB z;?IKF{F)V^*wA+`PN&A{COZQMV-JA~k94mVRWtR9Qu+GcCTLuO>l+N*>{@nxd!r(b zaMv`RhMyUl>!Y1^Die3?N)6-7?u)hG!}1M$6i$D1Ov2pwQd#1sN7d5|I`)}EJouL?tpu1*~>UlItR?H9h8-$p!W0#*hSwmOGVdV1B zr(ygOd7GnKbw*Tbpc=mT!w>pOmoKh0HGGQ16sape_EzHyJYl|)K*m}w3v8MTbfKVZW#`nfF;<5d_6X7-K8DXe zdu`W7=+^P206$_Ed92AuPRYnTo5e3bO;krwQ_sgrk}~dww_MU*oaBIbQoLQd{$$!l zl;h_yv2|{_#YZ99;Bqrce~)q986>BaY~N`mVSCdXuBUozLQQ?$)#o3M$qVx|7m~BL zZ8DHo58l-ix_DfcYRTf*g3mwd=F&8#&{n?8OVNY`ul9U}DPXn1zf$-XL`CmsIgTPR zKBi;a7+uA`k$Vx@IDqf?hKiRe?frM>%_QXt-0vNYu8=#@svTL$f7my>+#c>F<;EQs zYAWCPekf^VzI;EgnHXK|Kn^<97({_Br2K#00>i78;P zy=^vl=8iY-izRpxU&*qnZL40>^jyQ`2*7{mInBbSQGvtfG|h5ly@#|uaRU8OtD6rY zy=&ipHc`9baoa@Py|0mX(rX8+vSvvc$D9C3OaeS1Ll8 zHkohqXAj6uotqLS;j)|fCH;DL*)CO_>9fsFsS)c!Osewp4E6cxsYP+sVL(gqtyLd! zrgKh{Fg&G_^UynztNR3D_(8g7XhppXw;;PVCt^mORu}9O6Mn9!Ve!|_MP$;|Xpx_r zuVfOQhv22cSx?3ByHppSvOt zj0|k}^eQR+5}FG2P+*jIs?5dYC==@v8Q}ADVF*7s zx0P2uWa=*C6$ok za65RPzYFIRWh~|P@vc-TzuG`5VF`Vv;@$k7l=&8W=GH328r;5*eI&e~BySAgs`vJt z3NI|xT(8!Y=5TXqN%+93lwI@4!njx*rnoev#lF)a&OK*+pS?V8Ps~Rt`d74!>dk4q zm?y5|;zG2H?*~E7tg_0P+9klFb{!#T)?NH*F#2 z=+zoN_P>vb;;MdYMyMZD%SkdNPyQ@#d3(=$Z%5Z~QnV~N@ncm=NQx@Wr{(^3|FWG~ zNP2`2RqHDWs-kzi5dN17(=&bV{tOrVMg(cB&+Re;A{64LqJ4HsIYJ^>DAJVOShjRf z1@#kTojv_$utDFAHkF3UQl#EG=`Euv4U$8 z626^)xV`aL?Lprwl0w?^uEy*tx7+6l>k^O>Z=1Z)S%DyVPk`ws5{5=cz;%-Q-jSTK z0tmfP?-NXJzgpL|?F?}x8kfs~`XY%pcXhnXT; z@pq2SJ~54ABj7+AY4R>{f3gc~sba!Uo18;Y6c(!DNIdqDeV>HcAj;f=xzQo_kHtnL zhLl#vQ;eXU3S&*h`w>jUOyH$5n_vT3Gzbqyy2D;IHjueMq3`n~@sK%5lL_Osc)I9) z#(?izB_MB5xNxj&`(f|v1Ch^+;olPk{_c1`ZTGXkyHyIIpUhqaqA9bI-d5Uv)2ESa z+mt-bg=8!l1M+Pc`5qzaC`=@!i|dEC_*kp0&$2`$Df5LPi~2X;H;GN#k-F(=^e;%tsu`-`MtFLoSj>|u^-{$J1O z8xh{*N&CN%F*(`OpO!J-`*pabmEE#U|NhkHB>8k5EFG`kQ(I1uvnCcdSs^0Hb#j%K zBgE%-PakoSCadT(7jU-hWUf9G=Ni_l2G_0>}j`$;2v9l7U7kq!Bj z{k+gljQw?pt<&S*eZ`054#b=I9KoTy;l0jQzx|Y)mLJqKqd8&3eWPc3HnTy_C#d1K zm$P##j|B3anOLu22z{Zam$7Us@V0(Ie|?9TmN4BR*p!W!#(n0s@un3=MX%#$7q2z+ zqT9|7=E3Aeg}baoCf*dnc_y|6Faah4gr^z4R1G_?zeF(` zML$^B1?OjMyEZ9~F+hjcQ<&-^TuK9bZaw}OTh+(~Jp1Q$3r z!ao*Twv03AtN)DsT#!T+SiW+htHV@G%5-fri|NLjiIe01`{_vR5?3N^nYJpev`Iqe zi?ZAk*%)teA|CDAsxtzCu}&4Jx67)e646yi3O=@dQt?3+DCMcb7cC2zy%6zIzI9}n ze2Tmnv@H>`Kapmf66R6&2KC}yr_=+8AK9<+6A*+G?yYgMde*kR;%k6?LWk#eTR{L*)mM8>)hwcl35c zO>U0cX$W!ee`7JWWD-qLsUj5du@4rc#R@IH9V-75>5wdf5dZ>+Y>I=*3D!m{`F55U zuu_)0;gQaBcxL5(`bH}yjp+}ar|_J^v?Mvz!c*r^VzPy;wlq`qqH0jboOOh*lge?J zG|HR)QW^4OdG2#YBd|*ml?%kW#LHxkxtp43)7Kkly+_D7!a7kc&aV(dP&;7ZItU zlcBq!HBX0@3zUwAfpSF*!<%|vvYp@C5S5|cE{9ndK5cP&4vc<9{eExzJ$+ofh)%na z(}b~xPQBJGEo$f{qXb6lou;v$W&f}_OtrqtMx6;4!t>Vba2Zw9x2w(bC}^~wYRwxD ze}EM|8}YEdw?4A#5IL*x41j4TFr)V78j`!V|6muf{;^X2+Wsgur>N;I=8To#p{bl) zDJ2;(F-ys;-g&}p-0cwVF+xVs(&m||=#`HXWxm6QF)`|i^nBxPCp&#u|HxO{8^b(4 zh(y*^i(!k)^{QKYUmD7tJcs`frG+A0HK@ilLQC1n&FgUqhwoEqeEAb9&aIIX&uD!O zpVPQKVLABP;RP6Sc10x(^n_ey3O?CIA4IA{~>F4z0;r9d6REimHTSx5pP=Z?3P}FHjy_l`{ z?*vJcJBkHch?cwcZ6h}bmriOH8Is&GMEPsI+s0esIi zOfKCvBSw&Zp97g*0mPDN2;Sm!y$W%jGyvDBl_>=Rt*M zqkk58$6L-;^&D&d%|V>dZ(Grr7a2^G9MlnL`RxdoH$|$jIfJr8Xy}oUB)t^+4gL@t ziFvd7KeYo*hr;FRH2O$?``dl&Z9OMPtzwr$AzSw}?`r7gzd>Io&JF%Gu%z(nJ+>KF z%9lF^;^JF}-^tfqulrZXdDmZs_JlOzn~Lgvuuu6`!H0VvPKxYujH%x`Eh`^QVdi?W zkg8;J{OMMgb^Q%|#h;WB)Ri(ds4r7ZP~-%c6&&!E;Y4X#jd1zBF7HG(Tj5Wo9PL=e zUEh0(I*ruwjpsfnOVVPhQ(Hfq(FK3wm z5^pEN%u3?aYL8>M;r?`!7(cd`p_T9}f+?(;^nN|b2@d0f!IfreL&pvcbiz4d0!^u0 zw8cDIs7J5uoCj;}$(DQz_cEi|%u_}kdRyjRJqJz#eg#8X!k0e8st%7nX5MNMUsS^3jud1019P}Kp&W@6jaH1p z!1n7S3-olPrG0u%3x%WJ$X=TB8oK?_EPUss`dJ{09RyR8z70qOTS|JE%CF=wxqAN% zRbT*%+?V%AbaL)Z&g-mmc^ebkw7Pe&4Xn-3cRk^9u_t19&VC9k76?y&6x5rJM`gKW zmP32ZJT+H)jz_~;IBRa-^KN7zdQzSYHMyV-XZ5Z{5!peACWHa+3>S&*qK;03{%Nvh zGYA^guUIZ3{Fx=Cxs{V_Kamq(A;&FhzmjvYM)&YC7 zGHs|4M3wwuN>4Qv=FofJtfy*K{YEu2s)lvFMzzpQug+yJUZvg=30!t~ZzC2_P1fuw zXm6v4-*Y>XFwe(FscWP=t-`-j_$-QvQ$yM9N(97?e4;nI;G_cd4NxGb*Z3cC=Ja3H zY_=&eiO~-R4UIQ?xVWDuVtUk`vyuN{Gt1?Y8VYpJ?SstN7R02uk6SL$1Zva9V7V6k zwN?dzS}(SO3`$HTS(Hk?KGz=}auVP4sz07oXGMiV+U~g75*Cv-yGfRyokIt09?@iY zS9hx)(GmapiKPHnk+jxlR)85ix;OaICOtSH-g}jjS18CEi52q(&y4zrrQRZ!IAlyr z;F^~HTZCX3^i67!Q7*Bn4a^aXvi7%#4qL>3kHK6nR*t_Qz`U9AFcT3dBHEJZfLt}G zWbu>K8$R(CK)u1lKf}`g=-lp!#QK%E#DM>SoP15`qh#an#X=iNYSIuSl16VK{Nt2X zGP~GVPr9`mcUFy$O;qghpQGd`$9N1R$9#9FQXyz{1*nU?F1xZJRT|8kYl=1#(ayi& z_RpoCmwcU7#keUmdp*W2r79Z_+b9TK1LINdM2IMeow3ljBr`Md=1uMuY$g?_8OBs9 zIgKmN)}c(>q<0klXoj?a-2L>QSOvC{2;~viEmanArGr?g)t#neDvMa;ra^V4YHKfX zEDc7+(8lrIp*@U@Ps89ngB@RX)M1JZU*&5z$gk5UQ_>m;9-ARpEk`k|iSSI3$Lewd zy6jlOI~5|Id3Adb^L&1STW`eh#bz^41g07r{L^tW^j4+k`j-#9mvE0h1;$&}#7>Hn zp->c&ShY(C78G=4EvYP5MEo8KpRaz!kEgFBObQTNipmK|(XF}K==-VKRIV1VaBo5x zNY8!^qkat~JtYn_lEY&DV12R{=d{?EtH70t1A$1;Q=JSpF>5POX<>njfo;vV~-H#gr1# zdWG8pUvOr985vh8T-Ko-+4Cbo+7V*~OE5oU+hNY>`!4er{$m$uvsL-_RI!d|nU;sIObxWYKbY8j${! z7qD?Ob5YzFI#2qMsl5;`+V1gWx~bfIun zeUeI<$$b{%wBCqj~M@lF)u!tw#- z%9*XN4{V6X~VkfaYKG04%R+yT$1|lh4l%W8p z=2%EZ;a?2j{62hyjlW|^L+W{-^5R+;T6!aEVS1%pFiA z$jaW2bYkO-!DCFR!}uLc>Jm~~$s7{9qBDf@EAF9m-mqYkf`_&|5@An!PyQWN>)~2V*qK0I)vFr?QDOtChlSqm{I`75jl^yrz2BpY ziWReCMB|yhbFPA@&gV!>3;)fYrNJ-j$I zt1|3Zd?@+Dl2ZKs#1_1MRRt$M2Mpt|ctSYeA-86uY1RhXWung4FCC<>_+Taq<3F+w z2Dy|xhSya??yKA*-y6VA6jtieeo|)(!!woL4|O>-9iR7;B^vW8)bfuaxv4_Zj>dPm#J%XeZcS(fyS= zDMg&iRoB0-{cx>>q}f^(JeIP^sXL+R4;$H*19e$ZOi8a6$WB$O?a@LS)!^@1_Lwg} z;i+=3or_LYt_5dhZVO_kS&+fFjPDeZ2phe%Ec0rkX6-<{T|Sc6%A+QVWOT{&F*h3$ zxO6(mfc|ABN}ZHPTJMe+FvEoyWj}%Q%s#}=bt-#-?6eq@Cc%|S5}WqEV^{(F#G1@! zdefb9s!&bY@t|HjrK^dfE9=;gcMJ`1jQRDL?moNJ@lqKfZ+r+?Q`(86>_I0j^7UuE z8CeLu<7802WqW;7qopZl9(~BQ(>Oy(Mw;Lm(dCes{6rWA@Y$4b03o=G1UOjQhvWBk$7A zAIpOzoKDnexKy1r+f}!>Y*wZ7IsEV91@Jw*V>k+qw1y~kbU%&1by?H67R!2XJW=|9 zrC|0;M?e=rtD)_ePP-1mlAG;gAcR1ae*PP?JgWs3jBg5TYjb@^(lR#!Zf2Z>mLE?R zxJm=*ljMGm3ZHx}qAPjS5Kci+JTgV};(~^F)z5ud_Qo`ZQl{d~Yx^=?8Kw@^ng&%o ze3_5LE=+{B6$8=;O5ZLvREyy&-mX%}R=UF1Xnh+Ds27j!(+eRXcDc(tl+*dqf)*tF6d&)HR zR8A}w>8LFP3XGOeyP>fgYgQy?gkA4ta&aiZUNHWCz+v%}Ap}YNG_(#kD{1Ks3h9H~ z*)sE~8TrYqQ+xuVzFK(RUd%#YJIqac#`2NHbkfHISU-fm#7ec9+OqNsTGGx6Tf6-l zmRNW7a~#k2NyFdJ?=*_ueyU=bmHY5bF-lggqiyK!WcBb@j363C)~)v1%5}B=1r?)^ zwrjs#WT+Qdu9Y1crVS*|56#Wd~Bbn-R(d zR?M@O9myi@$G1{daqu!Qse`J-tuN7dUdsUQHBq;cL#h_qj}Nf11E8~S%EyQdA-Grn zxYtyqgl1}XTb&V@x-U=|nKdK^vrda8wul{Mhy*9mZiwJdHm{$yO#W+#(%&hrE6LjS zb*Cdml~^bF)}KUQpL?Y*T-bhP17rlI?;e)A&v2s*Gb!Z6H|U!pubHt;Z5`{!Il7jt zkSEoMbd}S%^qpj>&k-9bKHZ6@XI2hi)1JrhV^kk~Pz-lHHc&s3tG&g1lP9xIC}ThP z)_KW_p=YNC=JJS5yiaPF#W=&i|4+X8;B**dk{lreUmQz=id1tJQheH4%QTc^c${+PtH~Sd)X8c|U!OpSJ zTNRYI*|9!v6D01gKd@mOIW_&ZFJQ&+yQLFEaTB=eG`zPulRcU1ba_*rxEXE58q zfK${Z-IR%=L@^;1P33C_7|F9b{HLU` zlH1)0)Sf;S@k8h@v%3A-Vc$q%&36r{5-~y6FS(o2 zBVvY{B~z_s=5(1k#TN!Wqfb5uO=}*-+cGn(#ZTt;xmozu6_ImIyi2;i#zbexY+_9F zV|GSlXlA&i9LEO>R^yKLiS%FgMcb$a*4!`?kUJL3?KZKk2Z@3;xfpLkH{x<^45Xes zh)ssYyc+JJosIOw3~!br6RXQRcl^zpDK`{U;YKO!7UiIz!zby}UF;SY6`HZcV5Fz1 zz95<1%)PE(`0y&sKHhF&C}Ag8PnBJ5l^peZ;Q^NgF>3nDVM|ZyozqP(S&MdW ze6}pV@Q-N|g1uKuOF>PdH_+}DHiaFcJ1rvpwVnBIROYXK3f?!T&2^{A4HbaGT7hgN zdJdMSlq29|sDSrCXr7iZL1>z`>p;=<4M2c8+LRd0L5Dhg`{U!)Qd59Oc*ZWKrEz_@ zX5G$ktUx1mQhT$V4S9$F{KAJf<7Rm>RPXI`9Fc6+PV*P-*%jDAjo*z}RTWAyI7(By zO|^pm!|7|C^83a>E48}vf$Pn&hRWdlkBi+nCoM-i3-Mgx*_CDRam!6;;#|xo3 zP@dGp*nyZ}f~~c|ui?E6y_2ei1f=dW-DRFm9)%yGV6fum9G?)SaaOq8x4x6 z6A<6-Ai2tz3WN3(we7tSgN>;v9#EBxQws{2I!~|Z)oJ!V)4|xDzLGDASqEjuXV=*( zbU?=I`|-LShvNztE}T#=B#Ra-?Ocil{Bzs=HLeEeA6kC{;MZb9v8cGTHauY7mjG)T|Hc*&xga%hC1Q!Y zHy4V7+DQ9eTPSsJ2u%Ln58Q#;D7D(-<-2a4^^12X8GGCw!@hH4LsfhGnySJM@ZAmE zX#lUns}=!`3aON(x&l?@jJV+8nh%YtCZ6UkSrenDNmiLk{LY1D{C&alNc zI3cWT0^S^m{g#ZGko{9>=)Tj0{VWH#)xV98-hPj}w=X;q6|R}iRnh|H8vSnAwN1x_ z<1VGlrjbZuS(12awApCfQ$J4g%_j+RACy=^MY)-uU z6%sC#9VbLoZE%gPLgIH}*P0>s zmv3%uKrGBPM1$hs9v(E+6+8%F@R_xm=48SU8cG z{loHB_R>L*S_-^Y6PVg{)}u0}9;R`2N&{UE0FyYodl^?UQ!ZDA0J-?E;oG>hnRJMm zKN{S;KpP2rGv5<8N^mY&{>`A(b36=w!-0%LGV#3;=ACngT{a45$C4+@a>!Bv8h97s zAc2{QHEUEHZutO}R;ws-%_;2i-PYQitCe3we{L~bY$5X01xG8>cHnBBt&XULlGG&P z<@Y}M+jm>%X#-}<%Ui4r2K9j^BB*wnfiIO4qKIBD0?nY1ps~J8)QwN!Mqs+4>GBPl z8=wxV_nxk)ypqUF023xJL1t5Rk#IC>H~yH;7I{Ae3Tk1l?=&iF(jDAJtNPzACHNNT zNkxI2>=Z7A5YtmMsVavCm3qTA{7bns@h6AqooH*qYx+ zTbzpV^94q`yj|ZlI69gv*sj$CI-022#(-XA7<`L+je>D-X~ZG?4{Pa>3SSFtVfR6!V1mPs(+0}jb}+ZIi*a^^->U=mLl^dc z+7Jdmi~30o=n+n0zdbs87ol^CKnOny;NxbIVZ-y(cNROD{{Y&=A%DJOQrCe(v}jt3 z*`|zPH0LUr)86@@%9=w?>-gLz0c#nD1p9agif**8T!6LNk^xaTHgIzN!=OJ$m5e=M zn$|5B(6h_H9Q%vRcZhFkkx)5r5Y%kZbZcXaGlxc!jY^emasuXRuX87#u?w6sy&_b1 zrGmKKz&-|sd0G;GC;!c2jUkxf>z1uKGe%k+zJ+o{Vm+6a%ew!nM#ThUjx$?Z3#ZYJn?egDsEuwbd;wUO^k-gTVES_IUs zf8D}m!S#PdEewR+2r>@KQ|Ud;RB@83qGc{q-jQHPnA|4Iw_FkSBTT2#?j@(lGeO0# z#>q4vwih$7F3qI8t(&ZiBO!B(-JWP)4R5Z$!Q4Y%Z{zp|@`XOd_wox=@4U3-q02tv zbMQe2IdFI>CrL|}Prt)N)q;Bo?%ta#$&?Zx{{}7PrgDY~i%s=2w_#H@xq>U164ERd zgH^Ymj^vsvfE&13+;`#u6cb%ts$)%|<6)MmLM=;W98iQgj3nn{DJa|8OyTD&TCf$RRAFV6LXSf?6S zaGw`o*nVy4kb|gY;7{UGj^A+I2pD?A#2n`#wcd{T;`xVL`ZaqEoO}+H2aTW#r;LBP zC@I&^w-yWiuGOY&457fdFopK`gN3EcjeyIuC4!kbk}Qp!E*HS&q+~ih?%0T_P6)@T z>*GM!Fcbn0&B_0#H`(7(*A>-NDUP5a$-hzU=T38S%Un9G)rG%HM`ab}RipVtO;Z<6 zwvr@@SV@t|&jl?EADdDrXAUP3LfrsW^=}%XSv}yQ9gOo(*ciCgrYX7IE(8t@AYdj5 z6#1WWg`)o9_64YPghBaeg$n!|)SPo9cN@VR+I9b-d4oiCUoqn(eIVHdtDWvDmPTo- zW@W8Dc|v(_XQ@^=_m6+*et*Yk1Y5_W(#$&s#Deq=k4hfVotj#kf;mTW*~_RwFlS z4OCh0GkfCs325EVyR{UyP31i*dJ0R8^mV>{<>#f)=2!R~Pyc2e5`~NK zI=%Wt$~nJN{Xm~r6R(Qa2Ldf0iS$|v*+8=*eBC>LXO`*KS@Ua{_svo&MDeB*cAH6KXaXC97eQp6bY)7$LAOx z;PhiKl$$s8q)Q!s8XXc$i)_EVcRplw;KgeO)P=KKB{Wuzui^PwiMY4>!H-M;2pU6q zLj&}!NN))&q~trxyO09q1!@U~;jh$|(?`S`yxp-*2P*-7TY$r#ir>-(me+qO7Sr0N zNb|L4ECB8VONk_M^hTg6g3GqC-q}DQvSk?yT%^U7CkGw(WaRWWXKF{%o2n%8_QbF@ zr`)lRA@VQqZ3daMI?b;qgS>~d_M41}F24*AJd25vvNC)HUFh`Sl}~Ox-}Z|YdBepX z^cJh{RZ&FdT(nGWwvXU;BJ05ndy{7-co43zY-E-y6qQ?QsEpa0h`ItQ8~$>tg5_{h zc9L4hZZ=Z391cw13uvOhtxsEwtYPb4;+CR=SR z(LwnU@m@prLnGz$aRb;4Zja!OEqMR=kSPs_^Z8b_U_}OWRAT;hppV6Uq~i@e-~Eo# z4OCb-@ANX3lY6yi;a2oLFe9N7@acmb!7~kHF5NRo+a`})_q=6ic3j5xxWioMh$d## zVr;Xr*B!4~?B+m5+@X(o5C@Fk-gSQb{mLe+ucq;D;s~%Xv{oVn5v_H7Z)yO!9vUpo z;Na(TdgYU6{Xl;!u_kLyliveKR7ri&<5BeWORXB{VkTan>wC3Th?h_iG?bHa$6XT{ z_c!~46{u2VS%HJ|eJ538Z}bNkDq)K)!qS8>)wKYEz4Q&H+OcFI1PL_ChXGPR;;Gi+Y5)aS~= zSIzg|jqL7+pS!=zjqGa6&OMEkj~F|Vku0CaDr`SwBnAHQe)e?i^?c;9_gUoeS=qLn zpS>yGX6t@&Rnu~$)yxhThy#RqLcEp=`rqDt_9CZA2L{z>A*p?XVJ=!JH-JiFS5-Ea zjkQ7%flqX}u5JRYRABkDy#-sEMj^W;xq%u?61~b>AY8GPMxa^R9@1}RGf88IDBTe8=A;5tg>U<1%<#mYdgNn z+RI*`qWz}Xc3y0^ai$}KP-ZpARI)nCsji3iI~t2nD`^c$s3HI?L4loCs*pSiW5_T_Fe{n3<5_Y-KA&h-125C{pYxpT8 zz}g$7uTvhx-OGX(j@;KO_`Qh8ijQ9-FPnkFekJxCohPo&w+<*~B=E?pqKIeOzO8Mr ziF22hPjIe&Q&s^G@UIx=V&$^xH(ad^o1q^L(3uTC_6i@Dmtv%}y26JaHaPE`z`=Ly z`^6;fOFt`tw<@+GPIWRqGt!p+uQY~nmT8YVr>a)Fa3aCp-SPb&pxe$f?t=@mrEcX& zLv&okoN6F*xFRSuP#y%+b&(xw87t!!NlSV#hlf%Pzq$~VelAJmQlR=2dViO$T5Q9P zx;6a{psrIOx+h}HWgd>l5FuXO0plO$5vb?sQPgF|%+M2jTwD+*3LF8~Aja1_=dOfH z6@233aas>Yc~@W8DiXxfrtUA-OR@RRbuwJV88@`G)YRnUFPMYgx3J{j!GS7nA?M!Y z`bARjv28JJP!yTg2hOD)N@!Yi*lDXEu%?(@3Boj=Tqx!A?I-*S#QMck$Ywbl@x>Q( z;Iy3nlLS?6ZVmp<<<{MMqX9HXl4dHX5X1^CV8p3rX>Gd;=9Og!BEF1sFd^H0^a;qw z^JAijh`cM*=^R`7X&7HO&q33M7?fbM=o7V|IG2sz%jDVGhI{j$&+4 zc59XVI1<2>z2Fj|y2T-J({*u48|EsdtuE&)K01Y=UdAjhc>j8DjXZvh1VIEnK3G?r zBB0A&G~r1BA}jZPVW`F?{RE3p`^3xY-=@Gv6zuF#go$`vF+q%<#u&;iL16OYNW}A);}j|q#z_}Kcgx(Or-mdyaX8Q;Io>gYW*xT^M_@a#ksze zGj9ABrfd@krrP`KNhL%(EvEluKPV|EZB+Ou`T7EJXt|WgjG`6jmAR+=b&@D0PI^*8 z8}zaumT@AR@-O*kNgKSyIHrME#==wE-ws;^al+udL0r@Rek~`Gc?qQa=r+jG+79=> zcF2XZOHfydu7G@1al*gh;0gdeR{w^I>X{cSNrRGtWG^;xJ2<|TSO=uxn6jS>e@~^< zX@EAWVWg(n#+bohdv8~Q~@H5J|V3nGt%O z#2mN0Y4#zfN)kSKyZ!6DJT|Dy@G)AxC;pLlr_1?Sna#b~2K@z>5w|1eblTg672{go zu;Wbr8BH)#$6C_tT}{G$FX!8j>voea!ReG~TGV<#cG91ew+DaINQC2`MiI-exM)lp zMD|4v4&C2L_El~%x;K5nl-pn8@Xg8$dDbMa5hq~%)}6C@rkceKRB>@V z6+BDwccxT)wu=jgyMR>?B3oh)4A0T&@3-Ab>=uXgIPH$487`(&f}-3WktPKF6`Viq zQ-2Etp%nR=Xe8wq=k@z{T+rf48FZJ~BN&jsTlSm$Q@!BQH)Y-r)a}VpS}il}}jWTsb}Mw8(I(61grMtZ-<8g{P; z&&lq*Q|xv2u_#V+{LHAK#316425HUkEc5$4rAwMqVi*l8=L{{>!dX(z4RJZuqt47Y zz zhfc`C`k-5-eQ{#+e06haLX`Rt9{{FGtr_?rj%JXX3GGW@*V_554>{pIuU_E)&? zhl`F3@wSB$`A!gxlw}gkhR?e;Os2T{?k36aTTv!TtG|r>kfTlA zXW*TcTRLFAJp{NnwF8W?&u>e!a*`gpgS3aqXvfjJ617WGMj1rRuxsP*ot|#6e;kX` z+xHDVgi9fx9DeaKzIF@NW3tL};~m4;=WLJPY;FncO(Bc-dg9mf#PJPv#BJAcLMn8tY;AOe(nqj&&85NR8iY66bEbO?`RUc;+MpL z`Ce*A>UO}ig426{47~8P)|{{*e1J;9=P+bVzi98A>FGIHd?!n+wnvuR(NE+h|+^W8h3${cO+~vcFZ(7 z#gC_sTgI($xk$WO5H!KPP>izWY+B0DwV_3)TE(5*_IE#o4;X+no$m1168GZ$7U|L3 z*;>Z~NvpYlgZX%+RmI>#nM^My=)NCLmrVi$E@sds@yp56-i;k5iEXbZ*ui+7k#oEqz@@|zFsv~@;6I|Rbg%M_0%u{UA`=Vy|By_CfHsvrV zJ7C4D#=FH)uXBmaL|i>0#wHE>q=R6Yxfl0)!FUsGS3le2<~ZSE()>Kqr}~jo<-1<$ z(E2h2nQCvHS?zm2V%_;N?{;#3qAAF*P;U6Rty|@L&+FCt;?g|6f3nzloLp|`;-p>e zyB_D*`Z6zdd4IA<-KO_`BDd)@zc{|23e?*q$#Zb_686yo5uRi`7Ik&BaC+%e`V_VB zcwK}0e7ARaC{D}m67?imV6;}YFwjfvdVZG&2!w-Md9I$uvvRfJJm+6qn5~5izTS(2 zY%4KPek=T78UB}I2iFq$dzukPH<>rf#X&R4eRXw@Y>iekhU%wB9nTf#q;n@H&((dX zvJbrdRzw~)_|*P42r@?}zu8AYRIfzXkbZ$ea9#`@1D^K|k~ya!juV9VBZeWJblo4Y zO0*zslOkqVhhfTOg>kUy;M%r)T=te3dneYHj$@kU;$Nou+=Tn?*Zk_{ZsX)z`WC2# z#wW`*cWIKCIC71`B|={T5BF<3l0bVSyL-#Xg+C&)X5lW?;wKaWr_FS6FE{xE`F8Vq z&M+b$7f%9?-Fpodu3ecv{6sZwJJ^R3XUBznPJ9^op1 z$hmhiM(8%6*gow6`z6f)3{>szK63-XavdT-rvu4%`>uBz|wt_GyL&g6Z~Gs;@I| zx;<1$5#>;MnR5!QH3QnDhQ4%$989W{Q0%&{1pAL~j!2T?1jMq#{o)0p+TFxVhu4Y( zc+LlGaW%M_>miUfCQLqPHn&jA1vmu_ruLwslU_*^6GWBNSzJFCBQCB2?dLdXt%=pS zsI`biUvfrDAX5C-ACUd#lZ(Gzn|I*PAu-d1|Kmx5xx%OR6MB?2rwK@8 zC+>=cE1i6)!IuoS6{Qag17x{!SV2+)vOai+XKiR^*;;)({XZ+CJZf z=S^9^`uKVdI`a%jMx=VBA}d~Hr5YoWCuDa`!^t%3(AwyOD0iw>)xXia-xUj}x34pv zE^gSu27T8YbFP06;nSx*F)DK^IkcBO7l#=RRrb`?3`Ij-M0h%OY7v!ch(sD$i6}GZYN5PJC@AOaBJeQ7In8O>FO^-b0lUa-4Q486n%gr-2pUC zanPlmsPN}VKRAc0j*7+poU5dcWFL<8dqgc(gBlsq^iW!&yFGzOI?2Na9!^k(vgHj) zGJar^q2CmvqGj2Lh^y6Mp=|@~K&flRpppbeQ?-!=R>!&>`lr=_Q6GqR8J+W4tS{`H z|38QefLKVICU&u*e_ZMpl4HR$v;)VF({Se8|EYh89`4GyZyy_0hT||Q zUC0c8lfYQekWin@*mH}PlKuq zx8Bt75)DGCONb;lgu0ndggI56*rW^n++R8(8f|XB=`7*v9GeE*PiPs3Wy^!u8$m-b zfZ*%g3THG)uqdt7YN;WxMVls##AX3E-B)aJdQ@b)H$uDb;?y_67-A5)vwDI*eiYt^ zX=3q1Zyo#pZu9yQ)td#J$`NPj2Hp1yU{`mGc0M*>&YqIC_W(@P;^)n9#qKo@rmNV* zwFHN^!8~bb_Zw018=J!~*;rOldjxHX4qxcr!7QvgR??<>SM6ei<3G=Xwx0WJm$X6| zn!NhA(5=!#scYea84f=m`!;YbY`p>x@D@$#WtFO~3Y1@MdjouJIINt3h#4je0@|bb z2j0;gHBz_AvqLcbAnrFavQ>)SKS+LQ!%skFT&~t_irFP`1J=3Bs;3?C2c}r=icUXW zMCLTE=;}wePV-))ol{zYW=RorYmfq z-#QW|;kw@@tAOOd^REZcv4UgeCe6V9Qt4U2C*n8OKb)Q^_yf~u#~YEm7TfD2lLErm zR2{B!mg3f>0!nMe?EA9xk+I*7*WDB|N^!b#%r?87?72_=vN_Wpk?+8i2Z9tke6EvoBY!ta8lIdW;mBtLs;nR$!q4xuy3v1o zMcMS5_O?o4pPV-eJe%}>Ujw@JI$EN`Ie1-|_@em935C&UUailf$JVT8(V6hP6+V^G z!&BH4*V&kGd3$N)=&TMv3_a9TU2?_NtPw3l)#mwkmJe1?)hP@NPRoWLhjJ$Fzy%m` zGfDFM`j|3<^{?2#KNceoq-H3L*j?00op;PV&Pkkia9G5vpLaBb&U_&Uak%=q+E7zj zR$wImK9hAZUz7Sd0ibAwcbIFWaOjs_x)=CHq3amEbtCaCT5H3hb&1vPpOVu9+4xsi zai)o4aO))=vXp~l<+AJt^1hTswf2{|ystNqRZo@MUBPCnO+ciR;{>*38Ef{yu! zKii>EtPhXtvxqYH#`yNhoQ*Zc_;)#-vkQ5_lpdf>tJA6M5L*_Q=YL75$WxVdtBa@J z&Z{j^Q>dK!v-VmUT+zVOlG;`@yv@Ir1@0b;NIxc%rmEAM$EuB(k960$o<6eZgOTFd zPSiK8FXcrGeIu${2=?5pP9+Evcg$5+DvNohFF*2;nlPzf2sIjd4JO1o`@^}u0+o;G zM5keN!xPpC9|l*yF{+OCHLrMPtK7%~dm8VZWJgRZj(Gp>u{K?&tK1w}qi=|M$r711 zU(!;>2yqGcJ|qgX%>)lkX2(nYsF)_^_gxRFzj(UOTJ)4|SJ+xZ$;ZzxbV6 z)Z`0{Yb$T^t+zCj(Cu#YKULQy22e?R1C{H)w7|I1|8TqJKV%&#Z~FIrZ(UBYI0f1v z2F;anxNE!jkib~8;MH!{;P=Nvh8MdRxoNl$*hvkVBlwN5H&;0AzOU{@(`> zNES>(&{KN3+jw-Gl}*EAN0MyPwJV5yl_Mm@3tayq1abE+PF_srvD&&ZgY487A&}%q zC;t-`Oe#$vVLAOWyIn{m%}IHENLr9^HC^$AtMZEOiHfPs&l%m5nJ{@ZeIdycmG?qy z$}HwD)=8Y^wHIXb5D-(dc+dFM(tfhHtiyy5{dgBm6~;1tseHJNrah>7qg5`!VD&z} zl(}@_i?pJL1cSucSMVCQm{+8vu}Yoc>rYkENVyqqsJrSv)p>I!{-7A+e;Jti)V@#0 zgm90X_Oc>8*ZDTe(W|f}KEjV-Q-T+2^MS$IVE1%8G9BUAL|qXeU&O4jPs08VS-eoeP2yt{q0Z}K-3hKxX$z6lrq0_4 zike(5M~t*g~Z^xfOE)Qi6O?8Zr1UG{sH$&pYX_IIZ1kOb=!e^7ww zajKgwO|>~4xd7znUpdx&Tjox)X{lZ{KF#V@sB(!Z|26&@MKWNvA^z_uOp_3#zG2s< zgi@k+?nuNxP)0qLy6_Sus$b=58{D;oh4j+9Y9%6QDZ^c4<)c|aQ<(Jq?Mh$LJWfNQ zIUCzbHU7$`tx>B`y@!7Pm%JePS(5XP1j5 zyCkFaE9SF|<2i+=e{)|n*v+<=@#8qrhpsrrV)GZM)T z`R~?YeE%WB?>2?2d?!!{*fX|->H*jcdzdLTIQTWO^?XDZicu!-koZZJ0Q_(0gR-)TG2iHUqgzJ;lPv8%!uGG1Hi+5cJ~9x#z?ma9j2kiH0Du3-y0w z6$-dekrMORB^Ccm2FU-Y-7l@D#GOqk>u_nJFwL<@zgcK)dh@|r&HkP zrx>Qk4yHg$o0&=c_i*bS=SV;l1Y*nl1-N7}ci$lQ3T0A=oh-MwOx4*{ZR;Q2$cw0v zKbx(0JJLkF@wh)D)U~~q+eY1kTP5hZjKGun%0s~tH@F4I9rirYGo9{FBhwA-+3pmiGj<_^1I6!cW&fCq)-7SYN3TzKX#2O5N4ShYtzw*<)net zFdoSQPB4pacN?8jzQeOf0ni?riOt4SN#KVKaWj2$eN3Pe8fLvXloO9;#$meH;MexUS=GGt7X*=ZV^zvpDcTiphAa4Oy({hUTrHI==`B?!AsPM;#8< zrLj72gaPhd*)+lt=*FxFyLE~^8N%B2^xmqsVN%+~%yF5zk<(oNIrELu68in#!={qv zov?WEu(mN{W>c^}&nWTI!|E7sL2GPsZAp0)CvG93>gZstET_~_N{ZI>No8A6Q3Z6> zjI=f)GmKMv)sm2q$V#>qQ)m^>7Q%0|`(LjOky<8zQ2Bm33NKzYC$rz4lA6?XfH9n& zR|dDA7|fo!8880JJ{Jb>K{_UfexBn@X(x#>TKH=DPsCCCJvus!kptXN* z<x0&5;wZRlJe*({S9uhbe+rXjtG|yoHl@?78(H*DnD_@4__{ zRXcIeGQDCGtz^0z%t!6CmfC9*jW3olacyPU?#(wrz|F__>{M1}Bi}$VRa#cZ%0wrfiMU+=LOywlZrKAP88%-;R5h_T9kyFa%lg?bmlQQB3E*ljq_lp8$ANU0^a8b7i(+g`+w8z)c@A%qp$Bqkf$ zlx!D_`FKkZ0VEDG<&inO&}MPAf-V)*tB_@n)(< z&v<`rx>}{K7;t#r+j$)C?kM1ZwMzda-V^{NM$(9ug{OD`n5-1j`J|HMDe+53opiQ3 z3v?~M4vFW8La-Bjs;!+JZqBFj#P}QhZ7y8HTuGzcT-T*US*$$3uFGLt0%i}{jJb1 zdv#jKaOF7)xGX##1g4kuaiv6?p_0k6wqO57IB)y5uGeCGeuZHO{H9-hD^Nc<)V2Y~ z%DJ5LSf^MRO|3LWDCpza*_Y-du&7izdE0S(^>)YVr|4%9fAH-t!Un1}f{6AK^9>|sN9+!=g&weHdOnS!n~0_$Uo_sY=Y?9khD>g43nJ5Zm4#RC!)|+Z80VHwV)IvK@1CynOp1_h{GE!=gR{D`%DuDhR!_7n?l69k z-OjXAO=p-;x-Yx6%4W`o0<)|D{9DKSM9J_oKb4{6c}?&-2iNqHwYAji+hU!rKrWw< z{&RUL%vRHz0O>UD-yiI1W=ftKzIPR?f^AFqcsosgFA1O5i=S`8&0P_17TOPg7kriF z>J^HAzIxM66eT&n1^|(vSEZ4V&eLCa8gvBJkqTjBmW_6?10Z5r;0F+qMoSOKsy37G(r-4knLj2(~Gr zZnM4ggwA(gk4-Cn%qkl#6a7g|x!EmAo93vUrovWKsqb2@RfME!@&;YfpDf08XZ1p~aS2 zYnZAil)6W2^<+Z2HDX|I`L8=vF%hh#ZsfQBj;_l8d&rVQC8#qtSe3O{zF}u#m z#Y*3oIefZ?i;iz)%H#!^avrW3>#WlMd*aB7vf#Vfo{==xh(kHrTnaf%byhFlaSBfV zXO5h8#}qF8trF!gehJQmf&Y){d|dzks`Ih3aB*|~U)A~abz>fyqPlVnSr?&CR^l0S zdrW>8DR{2i`hWQ6eq)G-u>Jf|Q|OI~LUztQd@9lm9-j7WDU)0hia8jXEhP1Hxrp($ zsnU0d^20~|t0>;*+qTo|w&9l_cY+u++5P?s3B73bkqP4U`8Z`J_N(;6Rd#*_7w>-l z*yY~*y>$5K_nsPEB=&*0efydt^Z_I&-u!qxeEeMfV2s||40vj02|~Tegp3;zrE&uR z_)lYJ2#Q%S($dBXr{co?)1jxN4MoeBvTgd>$xk*Q1991XDBhyE@%uQ&okmY*00+vW z4E=QstP*782BZTrBZZfR@?@VaqkeiuB)fHr|AbWttcfFL6?H;z&^!38EYR6*3(c@; zF+%y60XOlP9Uer5x1$^rOQhH*W106!%SE@vPmZ5qvHXHT-R zT$ilYG~ozkjOhX!x&er79F4nJkk=tV=HKj2sa2+Zys;FIHC}NAwlfntx>V=U&AOOc z1QRuHi1tX3T@a$tAi<&4zovje9+X|a?``w*1P~vJe`{dKi7!7_sSlMaal8sjIsD%7dD~3X7$ZcE%SlQ960%o5Xw+d?$#UnmSK$*C zT?JGB!gVl>z6vVgNI3hyeY1TqOl^=xw6iKd9nLS{yYyWwE#)AF!1Rl73X%VeKm|>U zI%ZS($105C&O$Lo%wbryQQ2!3Myj<~iEqL_a@$ISZ9)l+aVUBZjg)3mIg3IPVV`NI z7v0+$CNGNF$6?Z3%P8ZpHKP_J+}D(ArrY3e1J9sY4Ewjw!V#7%V4l->VX7 zyIEUZt5F0eItEr31kJ@WY;~wl+&iOpWCb9#o-|=HChh@twx!`qSt~O#w#WB$LTcdM zRe=zp1)Q?;$&hql*Wa^}%(MzQ)yFip&J)yZJ*0_yDf0xeo-Ywz-CW+F5^B& zJP8`3=}iuAl}{$@c}sAMs%b^6 zrUv~jnkj7&{fQ#MnMW>bly>1-`ne`HGt{=hgEnUoRU4^%1RTO?y?0j&Y6&#$zQpST8^>)CMOkSB&0}lQ(6&}?L}m`>7dkw zun>bQZH0nOjpvYdggKCik+H(x8Az==B?D#hiJr)A@j$rW2HEGsFdf2WMx{f6*Pp?Bce)rEPmKcgDrUpBBLwtxg^j3F^W+4)P=F41| z9j>P-p?Mw#S`0Uv-~2|<*w~^AGmoo7e5uoEIy@<4pk0X2tWl)x_f-# zj`Bqv^cnX1Y%ofxsSmlK9;FUkR7Gn#HC`TcU~lp^YAG}R{qqmk3u-+b^hFTE_OS4@ zBZ_|!Qo-tqC(;0RbW~6kj&Ph1eYy!J-Yv|%S+R(>8Yc^8CSEL{oWlPkYQ`n+R&uvC7#$7an(cmy zoRITL+h=FyF*DAnJ6Au-&u`ZxPZNb(*<9W!@*rBkLW$RdtLUzG-N?m+HbPD!i`tHo zRc6sHpo4xAP@_FEnZ!9;vxujYdga%3&a4#-8c*zh;RhZPj;aXD{M4S|Q89Z)nSCWd z85}Yl^`;qAyJ<uD^9lTCRe3x_H7Nv4;1@6O!aD`Lpey>nODyDr8TNSIoV|nV+Cv z|Ml$l#ZJ9te8(qBjlbUu&Q8mO>PO)Z7%euQ#Gr}1mUyc$tNChS{-L*UJ!XVDvm|x{ zU3pGmKL+F^Ym_eH8bA6lhDh` z13T+!wU^Yk1lZe)W_woL<`7dki#GK=p(rsIJjso8Lrbsp9cTksZO%JN7OOQJo2r%On zC1P%dsMW~koo?!CIAWokY%ePaeIGQszL2{8c2z{WF_z!G&fGps@q+>gJ0bIma6*5o zMv9d%f=K2tcTugw9bg*=ae`!PGf+K+itL7HgR)Idd}Bo{fePzVs<5^uSqO(Pd6lXz zA27q2SJyv6OxcG+@cQv+Q*N|V zuo*rR`)H+<4lW48Vvzn|MOnruOJ=A$S40~I*pom6F50CS|q z5vQ`PvCga-dDjI(&oRY(dsFs9?CFoC*LhG2K8b%Tk<~a{lolOrv~xc zMRuEz*}Yg=h~VmJl0wMhpWFO7vt&2e+{-bQDlS|D%nROw4TxRB%?3rsgi;DMp6zqD zdik@y^j_M(kFuIg>?XcS8Ycx+&qh+2aU5(k9%ej&+>ODb>zdpgU>Ts+esFI;`nNXFzt$*>nRLYz{Ehk?Gv?AT_MaFctOK)cUt%B2X{%a_Tq)w1bEzksnqBwz`&>b@$X?oc zRha>Vc&EgZ^N+xHS$DJ*a8q96(>1TZV{nk`vsx)#J}}3J4g?L&QvO0iD|c)Ve~LZC z#e)=k^rfFrZn3r6FdbhELvpVa){Pacq&c4hBL*$bruC}jipvQFzKxqiIes&w1q1EQ zzl^=J74rmD_z>_a5ATnS4yAR+nIM2#3uDLy`+0MgE3+Skn86K%pLrY~C**!>mT<5? zVl4<1dZO&<;PX)4FNmkNAik_?W27O)4&86Lx??~Thm&SXc{|zg29YBTdd(8RcR)Qva|A;RTJWW)^1EmXZugnW5Wg#t&pLlIkO`hf|Y^(a3{kzTW%=fuJ`vXz(h#O$t z{-Qwu0pm)IY#e^g6GRKzMQy@~5KA=?_+JV}&&hC7Qqgt>Ng;EsAZ*o zFcb@uLv|pash`so7Ti#=oxSU))|E?9Xv`~p%JFsqV;EwWx#yFm99dvBf}nT9f|?M& zptrP3?NE50KJQHV5)kR`g;_JO?Y4n`@?gwXd96lYVwt%)*ulZC`_7TGVs-gK^S37A zEr`NxgRw_fB>1Hk;}GC30GtTz7NS}~ zzLy%QBYvTyot}*9k?yOhNJ(4IZlGo-r00Fa zFsd$)M<%Z9>9%jBHpiof#5No;>n#x{q0oDeg`OYo$&G~~fjOeHjag4Ygaxg~+JQJK zyUjrj^^E+bV+6p=EyBAQFR;3{7YHrqXx5~Tkrc~D?x+(P>ot?mq9yR4ywoNnca99cNSStyx^i1F+Nqshw7mXhds0~oD15-n zWgSWC*y)Bi8cIJf>u?W-^3sNEV=G`k+uuDI$u{WNlEeR8@d?(Uvx$UQ$oVt;D2-pK}nUA6p-$2knTo6x>G>9q(o3!x{>aXT!AGe zq>;|0yIVlu9zK5Wy}x^3U;lvh+2@&a&di*d_q^w@8(glFbCW6?qc{21G;Pn740|OL z6~i;5R*Qkn$fIi2is@XVct}yPX4aT->KaL3F=nezEY#9eeC9}izOu%3-du?UrmVYb zpj}yaxu`f!pE9Lah@cVi4MmaNkLY{Xv`riC>c0!`*xiY~n(_u&ZJPBRS{i=9iSy$< zT~T2d^_=X~bf!9ANlPXcvNLmO_H5$b=z>n>ttvJDp>e~5RN90I4t!HFK<(-_nN)Cn zhEtkd`_ssw(g*K3mE6RBv36{UoKXfw4=3iYN$y0={V%`sZ1tww3233{A-T2vOjoeS)HI~p(04BjEKB>ypG0c+9xr^r2d_bdZS@K6s`>8q6XJ>` z-PfpfB#q}P7qiA>v_|-Hay__1$B#d!6_gabi<05fH^moRR$BFYQX=5n%FxlvZXFkE zZb*f1&4QL&k<;!OUn^7&HF|7}R=m?B#98jSGoj;^!11EW_^A=9muKh7pPEr&H_uO} zj+qGu=cB4qZ?`iAwtrrE%%hEXOA`&ZW;dc#bgW)MLp(b#)Vg%pi{-oq7#sp78Wk_Y zGyHqXgw5R)4MVu8>->y$1ry+{o9ec>%hKV_5W^ZubvH#W-WVY|RH=3ZFe*9M_A}dka8$FP` zKXl0RD<+lV$kp7ZN|!sH#7O(-!q;a!65!+8Cc~a8++3l0=A^f_pYC=;jS6U5lyiDl zu}7$c{QsilUd8<7v47_SH-21q?|c_=Te<1T=O-|4@Bv>b_R>GK&8W?3hx|f^q%c%& zGrG%3){bCkk!a}D&1Y+8pegaxQwil2%U87YqKnyZMMV_SrE!A>hnVo*oCX{cwMo)< zzjlk&hpcY-X(ml_Saz#Pz2pkDQ-s-T_0ukTf#(yNpYg&fC?~NQ1f^!Iz*AYYvU5zkG_mJ@N3T*sn_-#|(E6d9%05z&4 zq7a@e%mLr=p25bY_bJzQAd2`6$D0g!ZYvgH8>NDotYZNOO*y+!hBvEqi$%|?K40MG zzHrsv7<~L@a6fcJ>2gfsFxv9375@RRnB-oGa=WB>=zdk{4JVh|W-zqb=l6I90UXU>fkD_=GU?J{8v`B_Wu8i! z{#|9G`XxfVlWAx5SQSmPfhP)mfV3a5_`-fr37c3{#t*gv% z270hq-Ni>qzj3bEsi~~>;Fb+ zT2kcWaHU}`qWR=}`lK*E&}s@F$l$c^A3Z7D8`NSw|LceI$1)eKsZjZ+CZfE;!NMZ>oW(8!%y>SCp|-9hm@j{2XQZ> zf}9EI+bY^;Jd0QP=X%DiWu#Q}@3x;)rfbE=#1or@t{w-M4^0E4V8uSa*oWCJ-dsS4 zeV%H{q@o|%?5l13^-o0X8H39G$KH3`lOJ?ewd{0!`|u?OWvuf-H`NfldRy@k1NGfy zT7M#mhB?25Q|sIMPhz9GtjPLfTb_2!Pu~xUu9==F*{!F1&yN-~`}y>}+89mEfgR>? zZQAh#N(mFPp9a-o4};Gv+re8=!Ch?3+_*W~^aYzzxor!oK!HkCh3zQfC$*t+pDh^k z1?^v{QWntVXD8rQ3ziRzYmr4LQ(8| zMOA}8MoRrjc4R(m`(Pr#9&;vaJM3*7V`IXQa76GGW7D+g38d|jNws*R2>4p=(Fei z^=aGGn`UbtuF!Q^Gvh}Eb_1_d;+?~!&bu5^<4sjC6y&S4gQ)Yl@V=I|$ExevIlp6~ z&kyybS|(DxEee*U$qU*c-H#7YFFn~yNxTT3?9YmFzYysi`8@=ueNpRP!YW~A7&NK; z$UWIV05z+)@FjMhXj#zYke<$zZBE0#AGEu53muJfZp(C&b(wyfCK>z=QEo~DU5zvw zrWs5xt&B5B=9J2I$5VXBC+mjv%LKbVhP&DS7QpK#n&f8ZMX3Bijj2+`XgMe!;&ze4 zUdDKZHCfl~?S`&W#<>_^mcdU>vb(my`F&{IEy``9)~U>{oL*K^k>MsGoxdsJg%`?L zUSl&kf6R;Gq(Q;}4!~Z?pE3 zzc+Q?YV;U>`>FiQ@=>{FNO;;As2OudXLwI(71DK?gVMfXwBk-_};V9W4&Hnq{qrqi@w=9=F{e`Vx_}+Z8Zk2 zbg@JX={mf~&=}zTUx5UF7lNUuur5h}&;<~?Rg|#;aV<$-fz)yr@2RbixgT#xve5CTx)4Ydq{5N-=#;<@ zPO)$#RdzW(+Af5m#MxboL< zU3|dDM{=Jk6Of1O`sTLO7||ni=g3yH%(^~Gt%eKbN{`H{*G~4;Fz$Qu0wrwTXBm1kAn^A9fsb}y4x^%;q+ zk9;Etqw#PAw*HcE`pFC5tEjF1be&AON9=iFCsf4FU@G+TcVk6gLlYpI|Eed6yg|Pw zLZ|yAMM_%not%-}x=^Vb^Sg-HP+z@5Sr_6X?naV@^5|gX?sZu)4qrj0S5NUt&0~T& z(fpcr!Ihqn38a&1UubNY+@zoux=#I}*3Jm(6RN>xDX$2M=r@mRw|&1-MW15|Y+e;w zpXl(m3k~zKm!*A;x~ZM>$(d9ei3)y4(obhyNs>0O&M)g>sBaYUTOeHg{ z*N3emuF&*^EaGTwds<-8UC{GHvn@Irv$b zbQbt{emW&ZFNsLwhsJU8mxd&wytAc$63vx*_Veywj(zlIfz);sr4I2S;fgeg9>m=U63-etIU;d zr05?62w5|Lx3;HA$hh!aJV$iJ$kxQhh8#@{(j83}=r_y^?Dij(36gt1+SUCxbWfvo zVx5T&LAd}G$!^YD!4Q-GlA9p8590312lQQK_VqCKvT(~f5|GROvX|MJKJO%)lq#jf z1W)3*F6&r9rK?y{6D+~Snlr4Q0u7I`*OueOZ04BxQXJKS(vq=hPtQ1));?_E@v)J5%&+bdkA;3 z;tz0(#)?`HDQV8QaWxhhw~jqulpZR#PIE`@xbbKcc-yp6HOL;>Y}}Y~LoLYu)@s~% zVmo(ysp48x8k6Sm&fWHV^CcnoG@LntNbw@`VMwoWV;@>h+oMyTXtkhFw_74xwif{p zlCU*`^oCiulgn0_dl(jrwWT^g^f0sHeXubTeD-ImQ}|B>x`pT2yQ zICj`KBE{}CwV+;fpK)Uj*JvWePH=NGVFXdx_*X&PO%q)Pe(2v#lg8#^2~3(uFhD-- zytUn9uS_CE%R|PZ_B@}5XIefI%{O`!x0{nX&J63-*@q@Oiuz02>z#}+UT9pHPpj0x zX`$bg1~3(;&s9=xxq7b>#D z5D-|Rhukx~2(4z9X( zWI*G)>_DY^K&JUPn3-b2|5*6JKs2UCVweXN199WVIXB=X&EY-wxbgYj{MT=f7dl%4 zfF`#e*?v#eb9jYq%l6DCm}s8NCzR;jp`kRU)|Ph=(FgMGC)~-dz)X0G-(@G{FzYEI z)Pg2n&4b0Xt)OrxpKW~LO%LnU&I2w*1H7#J18_Q1Fb`LTCt1Mm15{0J{Kf5g+%xpq zD08V|t;BHdmskWktbNqMguQKQ_TLn+*sFtr5T7bh8issmbbvY)_<88_6R4?CI6It0#9MCw}8{N3)Xa9lrp;U z?hhnx-tNfH*I};DdDY0y-%=(u`iUB3AG^ex{AFp{RhUw#wv+UWV2+Sniqd16sg?aJ zEPmH#E)W5C8tHkWL}1v4j#YU*pYXVB7$J;RKeVkc!mfqwdEgss@#A2|XP$|f=_}~k zl<&1N)Bh&BQzNqW#c6ynd4*Ux`tLV6@6Y<5+E1Ehl|`fpSc2-Qv8#h)DoU9S%*>bj z5VcgxwqQ6vehSVV-Gxl+w9S$xmTTjG=ZFtu(tq1FtIAN#RvnO#t!ZxZH`%jC)$j7_ zl! z$I||$MnFb=q}a*%;n1PvhKlr@$AF=MxjT2QfG{_fvC2p%n9d1yKA~XDr_rCPbBT3X zge;{SC+CjiclqbBdY-lQ(~x0yof?}F`U^tF7e6<>Z4V5J2B5a(JnZ-47WjDl4hBW( zpccv=^**ugz4&nz-1lzc^AZ6=%^UKMiz~xgzK{LFS$o`1@$WzEA(W7959ID}vnM6) zDPlC;K(&JwZh7&Jtb6mgqZ2{2(B(*Z`Z@6(b|iJ!ENfqpx8y)BH4Zy?t9#>jY-qBK zf)?7zE9MkeKlI&1*N1RUz6+``;{^cabf*4IfFBes1iPup;cnqZ%WESvpa#?N5V=5} zWVQ?@3~?(=c2(e?tdnx78p;$6aSMOz2y#R*J<0BD8pG3E9a?T!v|IR_gbm;EnzP zQpz;*;J@_8I%`o_YWVzbQKO|Qcu06TQN#5?Dq(USZR(+uK5L0 zNqBnMDrpaz>r#~2M5lgFSgreFl`VGjl1@yf5V#ZCL+NM=ZmI>t?N_n)^w~sKFUi`G zg{OaM?x1WHy#nKSpX#jDyR+l-r#y+4I#49Eh+(A0MP18~Qr7hel!9kEzg;uOIJAn1 zV+V6N=>X4SfQ17;kaA!}F^gg3Uc>5ocB3jk#C-qk=B4A#_g`{mWa0YTPk3jE#52Kj zGdF1X{M&HwOjV_E4D`SNXs1W-;?&S$$+ME<{a^lsz}4Ymj52SIy~4N533J0V8;@vm z41^PBR9xQ8`_)eh;Wr4M&-=};yQVb1GXv9=959QKsp(@fi&@l0@1S(VUa*SkQ7)bq z?eM?<3!!yQIZZBnVSIQ>dws*jVk}(nJ#m;^6o25Y;aGC+m51k@%x`!!x&Qs2(@K$h z1k(0zJ+MyhC^yeFZN+!p_vLKXd{JbZ1LS@mdE4(H4E5v?Y_t+II-IN*=}98@=Sk76 zlda{(PnI1=>~6=J5U&4TYqen-8=I3&dN=AcOn1rGbZejPbr2by&#X(Yc9lKrAYyHl z{^hp1Yd(396Nyu`*fPS$G?dYuHjOtG!sF)`zTgN8H8^gtUOi3uMN7|Jc1(kR<-R_* zPwS%hQ9=})fk0*kXCQp~_I=EnXxesc2IvRBEs1K^AnCdN(&M8KdS}Q`jPJ%ynH6Y~7-Eyi}8tUQ{MLo>5*YJAq^hGj?0oKt; zN(e3ip-iR?(fx7F%cR+a%=F35o%f4rrPX|qkLE9!Ban;H%)oM%v6h~ zgVCzH_hVzrj;|^zz^fd+vf~fh#cx-4n`0_H>b`Ls9uR6WxG&;mQM1CjecYrSxExO2 zc&;uVbelyELwTM(^D+4xn@81ymu1Sjr}})*oj9rLST8c}wB_F*Ke(MK-48|v6|yq1 z&!02xI(Q-Fk?tN=dOsQa7?|PqV)^e!5pj!I;T54<@*!D<@&D!QC-47z_EXdAy~Rsb z9Tihs3o})9X*RqMz_v(;&-l$&)JsPg|fW7X_9gZM{tC&|J{dKOHWTedwU>bb`c2zVN=(t&@Y~u zOJXs|3HY!-TQ~5|&#$Gj(h=&gkUVp9d8qgH?O-x{&>;Eq=i+Kmtz8T-tYQ^BDlCkW zv-QJOtWpY(J<+3S_2T$~sd7D;()-h4(Y?`$Qf&=+TiX(?QcMU$TA>PlFjlCNA>j6d zUJvS!|E5B}L3}KZMya~0YMDhzQ8Dk&heLR?ZV4L%Ldf6c-Q;_7fq;N8AOG_+_80@L zN@m`6w#4Vg#zyBP^*tpx{JTj|A_JI+G!_hqzm@{x7n-0U?y;dbw29v{I*}m|N$l;? zy}u>QeQO@QV8|+k@N0uuEZMhH%@|p}zuoZueEBYqrZy?hlBDuY#9*p!DM&Yg#q*eQ|M68y%HZ&Dr5=<8N}iRUPu-C5po7sYF{Fw8wmE9aPjA$}RBKTR?cx

    r)8pA-{<|mY834fZECOF>|*w59+XQrmg zkIg?7;N|t0|KT*icZtueTNLH;jo*bvaL5;!-!>el5W2d$>Spjrgt9CavCbPuwKSHh z{$XN>RLYLP*yyl9BsdD>3yEhMDvX&-NJz+adKe87v$Vb4E6BZMPeyZ8Z{6mB5TGmd zb53Xp?Kt2N5#rhOsts=iVet{<$Xzagu@k8U9kR3^3(@|9`hfrfG=Wg@6#EYgBmk;_ zs82yAd1Em0b`=QpQGY>~=(0KFd(unW($eC(g@R2spqC2VdqUfYpPwJ>6u0fs{DDQO zgxFZ|^#h=l>Gb2{85>@n~*zfPn z)C^||GxE@S%?%8c(K%uhaTR%;ZdmBB4}x`w)mE^HcU~Ec(4Te@7-<%%QoMYb<#&5+ z(c*hh^W{7V1`|qAEO38D+E$>Hb^^aLS8_E#W($WH>j(ST?Py|4^NysOj zFYlW^D`bqoaq=U$y1HUvVY$7zG-~<4pRv%}+pBoJKL@2p+!Qg7bruj51Un-kB|kkw z(4#{2b){i5@aCR8PLFj|ERwC+I=icrbOvCo$*m-A2pVZ`9EgP`RRGIB!=1Zxk5`5dH-W%Bqp`Av~p@ zF;VBV+FcuTbFuGpyb?T%RJ3C+pTcH>xE+(m;k+g*Dk@r6I<|arbd)8bwAckyrg9_e zpOSx;KN6EC@NTX?l3}vs4P&2Nm=>;Q+C_Nzev*-&90Apdr~&)U*|vn;Z}kISaqavR zPOH%*Sppf|v(0b*42C{w6$Gy9*B}FV>`jX$8i`)#Br(0M_P@Uaf6_hb z|EpM|aGdO$)o|wT^=h!_Ggha)Y4+>JZBP8VR%$se%O(5cm*S3NSK=Mngb4>OHjN5N zEc$b*1JWjBI5+H=^#e`(*4L}iq7@*%N-ZfWkMg~UeiHIR;lu6ul&wNulMULkY{~9a zr8aKok?TR6%k;G>#;|vqzS0=2a=LL+b>DjXuarT#Dy|0WKS#@9AN4%Z#5M;~elq2J z|LV*!EL>na#-)K{d4pTUN$|2;!CV||HC#BP1YXdD@tlY^_<+7dyRyRFO+bI(XL`yE zFm2rJbe*+KrigDJlkeX@>{8>3wF8k*a*+?RzxfOhW+GvyLxMt%HN*1-tnu}|`dKd= zHHJ8?hS(Ig5oHLl8dXQR400QKKDRVopYM9?jLYpGXX7>^j!M-`|9s>~ppfF3R?+

    3Y7h%03 zXZQjoh>r`~XaRpKM)$#DMQoGw<+ihyx*$N)F1JN4Z4g`=-|Fc_lL@E3?dyc% z>W*cI_A-m{8D?`WkVw_XxvD)UbXoXL4_$AT3TAm*`6Mmb3Ul!qG9&l zKA1&9$?aM%j(d&h-y@{qxxK^Kfr|5;35Da?Ihe#{J1Y67kdRTXg}f2&C_TN668R+T zx}MKb#0D(eWRW{FMQT*=y0irmm;saHwtehHlUrUH{)h5c*krF4@+IHdYq?gRZI6BS z+J3H$4uWazm);uBV^aYa`jwp1n7z&l*%j(HhCkjf3!xS`Nz`2;F1Tvw^oQ8v^2A-d zaI}kpf^8!kLtBX}?k2<4ITtq7+8mpo0+p}2JQ2M;!!dZYCD1V^GG^P^S3do-=(=gG z2t*zoP_~y*yLWm;n`HQ>asDA z5=APgJ_I0Z=>hA1>p&Cm{frgsOx0*9GD^Ujbujvp$I;5d0;GO5KHrF+A0MG_X$pgA z9Jkqee@AuyZvID=dcHhv2Ycwi@Eb8PF|ub@3P|Jk9o#ggl)&^ z-1`BXG@XNDlkmOHuMwu>`tE=4XL!R??AZC`*`8(VpB41cKOia&b8T>LPgNLjg1G9D z*stfEwv8&=6PaRhjc|9;&uXhZwX{#KQ7~R%kY+Qn+-|p6UsYw|KbVKN3xdA0qYH7* z)~IyP2ato&5-#lf6nnk{Ohe5tQmFE6V?nNpI1n1XzHboot#a%4Te*cpRN0AJ@<7w4 zY0%zp?^+S@tL?xhPkyhU_FlaC{_X}hu!hCrWm0VKHZHv0UQb*c(LPMse*<%>B=12w&%4p-?zo4X{!@XTO*gR9tsF?k(rUa>CT_c16aMcc1NkBsLX;dW9U)Z-8iu;qBMhRpsCVobPY*wN9^qO5mZ zRJJT&{j#qz+UoC*ii%nwDhIan){ca@md^(eYUP{cplq(Bd4fa&FM+nTn^yn~%U0AE z>uIs)q;~Tke{bTVT!2^j7_e%mD7~uCI4_r^%-axG{|R`?LihxUP)t?6JKrLGg;H zSLx3Bm7a5))A&AaMdWG?r{Ei>T+mdyc>dQYS&Mm|7buPKUWZ&pd=UdGZ$Vx*&bT{?zbPajYUy3#5K z<^%_K{k=MMGoZf>`nYr0auanxv$C>6M`LwU6pjh1fcSDob&a)(KGI=)mVibDG+24! z9Mn<6T$cBIK-NXJ$@LWtO}+K+j`sHU+1c6Q;pkg-wRaB&9WXeavPfyJZsfIpt>6(F z07ZBJ^gQ~|ZMer`m7P(OliWZOG3qLCS&sA*HX zLG43OlOYg!G}xd7lpPcZ?h`Pcrz4Kl@DaIjwH2>l9x|i|6-*7s5eM~>QMj#hUefslc^ng>iUjV| zhSSvlEy+7Lp^064CDeutX?wU=!VyQ5S)J(7;XBF(Uvwaq4vaKX4h?;wkuYvBCV8^x zzcqmgp+g4ux;&^tO(LOtm8*=!_W}dl{g)gpW2jl_mw6a`Az_?!92uJp;ont4n3C+P zu#0lAWXo3}>WC1-hc$#kL(7=FPG{wJ;H?D0NC=Q49`a+)G|xY(hC!?Lv^X$x1PH|O z^jQKl6sIRaz|H?iB=g1NfGR3yM=NvlGf%`H3^aLHfrluyY_l`Vurw$s*T^g^e&>bxXeGQ4gW%rru>Y^4BN>Af3?R*SyMqNsi z+wSCd6A1_T`@37f@;J`_Xdj4$l9G_Tv$roVE7PsDenw1O>ModbUavpopLVnyw1xjT zfSuCPQBPM_Ia4Sl?fJ83pjgf3IsNnmhpy2N09;OmoAv#DCAf&o21eY;dXiCvQS04W zZyeatj&t?Q`{&(}#BZQfx{f%}tIHX3O%L7o@3=R(V&V`1CmZVs4yaf>tP!E~} zp)+%m9cuXwfeV5;dM_c;pCND&r-piWMLWqg{h zGW%Rw$~e|jX){)U8#pE@EgdpD0CIAE{^h~KYJMEFu$e|gq(#z)jw!BH%a-|V1I7X( z0vp;d*@#q1Dyq8u{Y$t{C8+pZ1Aq1Rn>9BBo`8tk*4mJrkB^Uy?Q7a-0R*#Dl^$RQ zv&O8irxViS7(}r01}y;@N;R%y22E&w=-M^6D>`!9m!~h`4{#w>^5pdNts}Xh`@rn1 z?r9i1JNtZvFPB)2)H%{xOiIf3))t@D5J{avF-;^f@5R}frok&eN1Ru-2RC%YKJs!K z5D1kR#8JCMGc?BzU~FDT_}<=LnOE1|-aEi+c}l*Gjg6(HCGf=o zrbnk{JyDU7kzns^bp)d?%+G)P_>nH=?e%VD>&4-s|7v$6D6qK}Kp}`$53Et5TV&{& zrqiHUthcmQ`J4GE#0-q!dc-n+)I#G`}cS6?8b7X zSGSF4r|YmiBIt;teV3TxB7U$@*X1RL42j1?^lXxj zBmy94gW6%Hx@%#yr6%Em^)!^|?h(|-6H9dsVMW8(D2(LSgF0f_pH zO>USkpOBaLiia9-XT7~P)89>WnW3&*mC(>zpM=Io7yE&h4iJo|2(a6wb|gUOxVW4K z1O!BjTo-`?oojz}RbuZryABzG1YA%? zN=ogSpPwJti==rGTK&wVA!vz1AUCWIXxR9kmu)6>P zHZ(LGP>F*r(mnVu+dD@_O~2fC8&re6zuhuGCjJP4A52vmoo)<*?Ol}B+1VK%9}no9 z$qEBOZtge#P5Bb+TS5`ZZQj1`Eb40Ry#R*DBwxnc&s0m%#HopaGQ#Y6p<>Etot<`h zqMiT-d`{AktktJC{&)+dcKFKxsJ*>?G`VO69pRDR)(8d7@zL^6kmLW;RuD9_P-T7I zDgNwj^K$?gMzE2OK**_s+RsVj+}+)+_C#Afw;D`cUt7z2*bya#KRtL)_p1F$^q9P3cNK4perrgg z1Ee#G?eI~H->m>=^ZVtFhZLZeD-HM+zpIn{hbN@h?M?eD<%N;#Qe??QB*8e6nB5ED zBbgm>7L~PkUJy>JGQ?~(DnkNtfF}Vuq1pSKfi;Hm{eN5gCjb#ij3^gcJBBB64!x>{uvT39yWQO z9fJq!%n4&-66B?H?dHCb-xc+DNLVQ-djY+@y zPax^+=>?!(W^rH70fgX03^txCp`iO7pkr&*4)jb1;55&p7~EKk-%-jnL_Vk_7_$HB zmF>hLwN!YFPnu4%da;k3Flz5up(-zgQk?b{0_ZQg_o39NaGHV$?d!;DUq%};zkKA^ zTwtrL_a_Z!2qvW{D=SxMm8TXL+W_|R?)K8?!6TOJrP_T4I~wVVef~T>!$5w`^mxf$ zRJ2tp9G`s*@cVayhp?LW%i@lXhr^=x`VUU$nMOe?^BYzl+i12yv4Qd$7d+wP*E@`-I=3hOhv?y3<{9JaA=hQR>B578Z!-mKvF^+st_DYEzaKOFVZ!+wK38ojJ)uS zV#M~;W4^iy8?sLr7xu$2lzS53Fp>!!8y(G4r{`3|eK0#98&Yh1+xrgAN{mkXYT~RTjq2UTZj+D1(g*NW(A0OFr z%E-v*>FB7ES@H-9&dtm;09ImP4F7dq(P0CHhI{uz}tA3uNb;C#6rGz62_ zS>yWhP{p#vX%$6DC6%kUyE~+b3v6lkhi2x3OzPz=ll-qYM`98FsIU2IH|?>pv4^Js z%v7ZZbwr-&wZd5Z%KAF^NJjvhKrGrQkd%}h*|FtRNoEVo*#$%zXHBgUQ2N?FpqfYa z01sebV9;e=+^Q!O1vL)ujL7$U+`EQDD(<#awZ5Sy6w>en_~!}_)k zHwM$}#*3VG#_0zcTVeoR*)j;b*_$la11crucg|~Q;P*=Ns-uP(Y@`cQt%l4C`pf*n z40sLpI7r{jZ>v8g{Qy*>{1hywrpBdrPVNh-p!?VK6o=2i-1(2wX*f7Itbccg;wZqC zIrh0h3$N=SgMF-A!sOR&RPHNsu$R-m#t?E#NlV+#&EcYu^SuuMKujp6Ix`OaOU_jMzB`YI%te zjAd#N{0rkNyt)8G=rARcnl+Pijd1d7V8GSYHJ#H6i__=+&R0R9zo>|gM?zLsHV6e% zQ~MKi@k{hiGzkY4oKRc?1;~g9qG#Hff|TpQe2Vxyh?S#@${7M`WlbPD4Ki962k z^VPruf9de>aDRVLs6|hW?E`#ZfJs0gHD$2BpIFfS-Me>tpxJKI4|rKXT6(2VQF0i_ zPdo9}R@hR7BEtP7?iz}5o$q1D6ub~!ndG~NURoJQKMhkqnwlzg>q=fRTAG`G3J$JX zoo|Tdl2cNmzN(>c1zQb(Wqp19vhs4d!Vj z?W?o-DVqG+H>UKuP=rlcKrkQ=jbQ^cGLd)}BME+C#E1C;o&doyDe8p=`l5w2mj4k1 zT%1MxcX4xgJ%Nwnaw_*X)l-N)U;_ZLGYI1krhqntt|_B{eN>D1&~s+t=I2M#R{kZt zr}1eXV*SVj<_j=Hm01su&1h$T|L)0pe|)uZJK~0%uPGATk2vu*zK1l#xGrNSFAG z1$M!Q;}t1GL$ZV}PoLe%@-7j7Jz8*cZ}+)8g@^NVFvUJX1em>*71VBuMXFuZb#e{_ zYQ_t%009KJpMp7u)GdRj6B|TV5)^1fs$vs!phEU=bsYzQ9!%B8y=$eb9b)|q0Y>`# z`N{HUa)$Wkg(i_|z!9UNN{r2!6gb@ZrPeBTb_W5}j@bZCDG+;x{_e-x86co;A z*^Q4@j3Kq3*=`}AV+pv+nav;GUS7q%7j=XnsiDF62qDZAYd_KeJk1nv+W|e{4qR=4 zrTyxUmn%!tq`;t`Q9&H(2*_eb0D)~p}Zyr z8yf^0trE@RbhLtkf`96>KsO}8c%1@toyKE7Tcnl?N@}@R&&%;OY1-!g?>7fLdxd>2 z9^f@HoWa274pcaV=R0LSqelek+ExFe{-)8)$mM{R1F^KV&@|HgW# z8$m#Q=&%=n&gLSBHEU~Y0Qm-og!}^!=#c+LeSpvJ-T_D~02(XOgnwK#IhYFYJkliDhJZ#DY+i zQn^C!zqcu-FyT_K5G1bOF2KN^G9S;g- ze0@gdF$12$MSL#GA|sVRnU8`&$Y#> z{j7l8SY2HmV7-j=^nt;_?L!%I-}vbLs& zii>wVoKd*g5JBu38>`VMRJP(5^1aps2(F%w|A*+J;oAjJiVHXWp2v)Mw(sQwI6%O$ z($mwIgww^~+t1d@o#KNn1(y1NU@D|)(D9j6)2^1g_ZNbm~|a)sZL!speY{&g;hd+5EH(bJ#}w#-v)44({f5|t1UFEbXlrX91P8(PmzIp} z?CdNoCUT_0r>Ax504|n_(yvJ7{S1`cqvd={)Zt*vWIa3wfFbDI(%I58=U{X>n4GNj z&c}cG@)#XGJm+Mw)n614mbWubTO--?`MArK{~ro>eddQ#?vC`5ee=>$`v+7Dx4BeV z_e0=`Y-j+A^72B#`5@f7_;k4!T#n0;Fr{OjM{~H&j!i0 zC|f|mmxM3b`%u8NAsC=6Z5bJJs3KV^={j!aOq*<|ll$gSAUJ*gxhr2pK^%eP)Q<@NjQ9H#awV9pZ$g`7>5i-n@v3l)ISAp<1H#^HAN&p0riTc4_?Q%%i5OG<5uYvgc z`vX$W{js^OF42|f19}8hez{InPG#jGNa))RA3JP(d~QH6;#BazJ=bnD2kmfn3W~r2 zy*gVddC;sawaq@N&x}o_^w8Gw=#=D2g8Y`#<^GXQ8R{#CUjJ z1^Uh2bpYuf92~g1y$pP)EJ&Yu0rc_p>r(;(0sxNzt^#M|q6Ec1qr=KuLHKh~*Sfr{ z$#yUf`11GaRR@yeE_bgS2mJk?bp>$Ix(yr+wD$!^xH%Lo+#KPb{1)1>0-#=Pb0}@U ze_j>z?R4W51yKA41~G#&S~p#q?Hwrj6CrL$NC*$Gkr;z`lKyx%dG}BOF_QnHITI$g7(6tNZBH}Gxz>s;)9VP!s80AzHQhMdV z$)5Opi~Zl#ul;z-L7>4kEOv540R-UvC*Uh+W3>p!k1y2X!();D=VFm;e%OdHwHqpT z%>TvOd>h|NY}Tce7^Vny?_6GuYazG(s>@o`*^=!<2hc(o~bA+06pXyRpMezt4$nVW~u7E z2(RsaKL6Z6JaUx!b>7U=@6(TRT{T0$eD){}#l|ls7~N zpVaL*5-Ya7fm6=z?~R?V96v9@XyMZI=;Q7SH+?{b`4JxmRIG;`H@f9({`rwhJ*a!w1CneH zn*RN_4teh3VI>0U99P!fUZ%XMbd&rzPJfzCRnhP&1Oj$+hYO|f-=gVZEbAK`ZAtHs zkHTc!ZwVihGyeNc?A&H-7kcpK)lWWk@6Wj%Jn;8pL)oMGPw$AHZdYgZm9(nVdd++f z&C=#FTBV-zl0yatT9V!}qkQV>C-bSB|2_uhBLf3GPuXVb_1e{WLv^O}@-EsXiIP7{ zqCe~j*HAf6z@n~bsC8FJaQYUCguiQX69V4>$$8z&lM5>;Lu(2bP978^-zjs3BA3o{ z<0u0|lV=Md!Kv%LiUYKFz~FAMqoh6V$;B#rTD{-qGTWk$>l1~iV@n%;5XgUb+pO3} z{WGV+!*cIBWrVwDlFd?oKQ1D~+jDxXGs{Y1WG7GgwB)|>n|uDQQ_=wTLyWuoV28Lz z@Bu;ZmA#>o%`|`iO&RaPe$_&(yI*YxLT-$_Y-NMJgrbmrq9qOak_4Yk1jZ>1^q|6-@DQNPT=xew3a7<)dlbF%ptK8}L z%ev~TXr6`a;KM&^`!qGKnPmArmVsdF?cG+5C8hK;139allOLv5&|fl9t?vD2skiKi zo&QPuwJ2v%hkM9{IyB4M;#93{+9vi1>5VD@J3<>1A{R(eD|k!o-i_u;e>dPkssks& zs#$9<{65v2^hVp4DEY@*=Y4fw=FOa4?F$n@n)C#odZ$R1OkZ&Lr-qqf+V z?wG0jMS4fzKl@@ATPRPMVD0A@wej=ny?&;x*iyL`mmFzX&+on;e*eRJ1?{Gjtp!m9 zF6>s-pIZ+kr0Ooc)VMI`87pGCG(KCl*f+FV#$h*CSN3phrpUH2xMynA$bMLE?bV%? zwsiBMH7^&%*B*+lW2}a!OICVM9P?ilGOe+?(L6b2^Y^gJU2t2}b{NmH?t8M5xM$G& z#8RX#bFu1p;jfvz)pK!5b&SV%`U#$z*wi5XM*oG3i}m_?=J{_O0l6%*?$uN4iA#E> z-Ybt&XRZ3DM^7#6s!zW(XTU18eb(xi)mv7m12*+iEj5O7Pj4NvAo+e;_nFV`JHdOF zYj@nJQOlo*5pKrbpA>eKw-vZ_%zHx&pIH;)vtGv_ZoNeL?t;U2h1dN!Y4`k?Vk_Pj~K0 zPW;T5=XJ|gTD?2^1{D>5s_)}wIbpo?l`!(Qx3#zDf_S?#wY!VJRnjlLtzI)-FX9f5 z9xJ=HYATWj0pa?j}?=fxaf z)74w(nb-t2NmCR*^#l5yGZQKxS!>#{>lJ&G6GrZoS8psn3f#_jjP%s;^Dtm{&g!w(-mZ2HD4({1j~z%o?g zqL%d+qwn6igBJmout2bK(D6fe;+^)c|GW0&=HklZ#X|wBaryfU6aX!)zH5sCCM6zo z*Hi!z4ceS}-1{gv_{+b{4DF-paycrGIql}=1ok53!p`@QQ@B>mIp3<+k? zIe&2x_>-j1dFD3PLCReKDgWiRxO1hP>`PEM$A5cDqKZUsty5EujtBsGeY|87z-4H8>$TlV>;L)g z$SqACo371q-FvBTHed7)__#;NR83f7nVsvESrtuJA=K}2PmVg60hS~Xa?*ODwFa2h)8ydT{VWgi-T0xJ=ZPVewZp32`Oy9yVe z&FE&FUyPMCpg@*;&6&{ip`qnyH-{2>t-Ali-Cn z&u5HS&>X;Q1h#S3?gGpY-`)qHA-OPAXqNoj4&?N#oh-sfX!|H+tAHwkz$mZ1`0&!;_wRK3_Nk#8iYU=c&gag%=boGkupoe>?NHL=ZyZD`ay2t^mFzRU39q%zsr9y3#1)3 zr>`f)T`z>75*0GI;7Jm35*W_|P~;H29${!ZW;c6~{+-*O--m8QjnJ{*J)*6(GreWK z74>`7PP_E9w6H@EQ)G~?*&kYOUJWfQ6!)AXJnFX+*^m#8Nq^GA}?ES2nJ=B!g*0bNApW=%G zM_J2lP3J27YR7i^R%%GPk8M&EJ49utvx#S8m=Z@x4i#hG?3W9^Hgd9`(0XNjj z@}G=~!Yu;H%S$Gn!Eg_7{sqKfPBiArL4GfI)GX-LYYSiHto5*`;$=f;x z9tgdGKvmE`54ubsMO|oUaz~$yj$=ukN89hmLqc*a;!MnG>cT>83+(AxJ0W`^io(JY zhaYzn5?HF%orgd4+hl!2$4_?nTxmK055!VadiL)iCQ);d*tZOA40cNg>Axkb@|m2G zavn|^3oI+qO9^2Y5ebWBxJ?{u&pZ_>z>)}gYE^Z0n|Sm!&byZ3Ai$x?65Ub%s;KC5 z?PS>DtSjDNqnGE#(&Nt5{87&*Cd&FwjE$*k({(Cl6+V4Gx3)alwz@@BfX=RsWxt{( zU}>Y57dMz_4y$o^GHc}DuPIN6jO>8eUc~Q61(OvQdcDf7?rvjC{cXKm>zYb?n3+GN zY3{A{-2d>mrrXbniL>I#_ojwho`J4Q^UcTZAIXyO7#c4tD{F3UMp9mb9phFSlCg!W2(%k+->UCo7ZHW=pxY1WWz!q!9N19K>OJ&@cZlSoc|o0Ct`(b1`G+s~VFf zp%M|ze_N{?Ao74l_K62iM%F^Fb9_sle3ZLf<`(~yU0PcD5WAQlL zdv}CvZ#TH$ZDDH*#_#l@>ra#QAu;S;cMmiLM>Fgjw7mxp&{>aP@ts}YY@)1P?{Ukh z?LEkdBfe?@+|NR;`s@tMg+uU)QE-#}&hv?^!UfQlPQ(C+pGKHtJ?<7lT>%qiI*D&} zo@A%{5Db2^FMY=~NUp`CeOBDQUC+xQSoSyJ09wV3|@Z?1NTIR6oz#7RZkug2o$YB>uqZvk8>N2adI&^bDM|ZHw9C!GxP}1wb zx_asNRX1E)(mJgU!|N9kdnk1|i|p=icc1!R1tx%pCND4V=_kSaN?*i1Hr2ia0uC^G zR-*%+g_7sZ_qps@VmAMpS$i-nQhnIsW3f(*`jE_~kXsL(gtGMOy#q%pph&vya*n^l z8redvk1pYh3SXplzru~^*~H1QvC^w8Y?rQW`kw~b|Ms)@H=u`hJtc+V;o%Ux0iXmV zzA)J*WX}Q=R83voAjgRF-2IA`>8;C^CoOj-9{Hg|AoL3?LS6MKl7C@wD6txvwd!hM zantX2$UZ*m@bqoR&R!Bw|DQ51zhg;%mCmCis0SrC^Q0`DQ>=n!T%g<}aN+sXUze#o zfOD9&k`GZb6(a8;h!p1KUF&@L4RVG?!6op?KyCX62Q4mN)`7wi=}^{v`i{v`Eq9{s z3)5BPx2s?CI4z*aI2;a{(lqpyqA&JYL2GHDSvD^r5EvBX?Be2LZ%?-TX$4z_eHsLr zhzx}v4Iomus#D)K$tYyHf9KBr;bEn0;?_O35SZGH_K{ zs=$$II>9~y2$h22F67WDrK_N{ETaHvd9|h}LyIK#%5~i6!i5XkVx<#FCvWrVX23+M zEOo;5*EusYHfnnaP*8m^OF3T!6@bFBHspMfwN{yrifHjGUl<`4Cew35>GTp+KD<3H zcYjuuaVt%{)enR%%U&+EpR$evTlD6TFb)ck*{m_XP` z<=6sSa@yGhm9x-_9`e;zsy&QCWFIg8x;V$FR^G3Fsva-~-m1Qc^p_9nx!$OrQZ)9D zp;o^QKR^G(djoL5Dl%n0t0mW1VqEFEj>gMtNm`J(f-ahHVtIMF`w04iD_5^(wwve2Af|$fKCc;Cll0*&WwoJH1i!q1?t@742- zPO;Ip?AScX57Xd+YZ-(ln>Pob;F^Tunw_NMjotO3E-DSRi;1nzPd4p9SGp3P2Gnnl z==5I8`_mMBhP_+W%WnfA>6hVa`$J6k%?#pP9UMkFvcEuI+{pj_LoEG6C`ltsLe)Oa zxWd!=`n;tNjV>v~l!ok2kBfipwdh#qyD(uQp`od{b8I}UniP18__3L7;(V*jpMJ{E zq{Hj03w`K%urrXJ&bPnYNbnOmQ@Ft$okmJ?P|!BTo1Ul(!sFw!U3-9pIxIg0_u{AD z-an32J#%V#9*!Y*hu;XVv0qWr$HvF!=UF^&EGnrn&%;tP42~mTw>$;nEs{@Gycrkg zWf5Z;;JnBC3Df8QpwgE($WAuMn17mmbTOrYDg@koZf-8*kk7WIG_DvSfm<&3h<3KJK1}3>WeB%a`c+ z1=xfwgfse*UZWMidlwB(OMAO+J`rW50XivU53=wX zE(P;RXnNmeVLpW4C+t&k&t;?{`j8hUrBUSh0}FF=>j*twVITibPv;wZavP@P~eE9HRWq8#mhPY-g2G z%f^aoFBf7W030KD+^bK|*&&^l-WliM!q*H%@Ee@v&%*1tf8S8DC{?+t|M=c7@{c3g^T3j{obC zcI<>mWm}=y*q4DRe%Jjd_%Z~Lqva(S}^ zoqxpH?UJIR4<0?TbOIYx-O>`e4eGv7RB;%MVO-*@>~j0}e7so*!>305*A6K?q6PI! z4+0m>&s6np5jId8=uI1qj0vc4GO z%cwrczlXtauXTK!5*WkX%2fq4eDi7d^3A?4xW>W;LHVwxjGZ{Jk@TbbqWzv`?tYt# zcf7h@{2r=N;llL6fs))^1Uo_*ZO}ttbKQyT(TD5XTpkusBy@o}llqlGRY#+tq0u44 z*h}^CwsG#&wt^#_xtIBCMb&JWe`MN(WUL-|`1_MT@nLKy%VPIMPdnqUv#Yxgop!s% z%<)@-vdpFc-W#L+R>bSPk<7O5i*`Nfq5>dDuwxbwo{6C9GCnK#>5Z@1`z;zaw>S1$ zf@26j6Fe}#-)hHl``hI;rf>L;x1?>qiyTBe+`cu1N^pgFE^H&-Pa&`$FQlb&i{H>I zA)k9eXLNa?he*|R|IY2B8a9>#ZqO(d7o#Tt9h<$4I5{$M48(L^S3k)fq$4Pai5f^^s#vJpD2 zBzo(=@p6Acw6a2Mx9am1Nv4Vid1Gqbewa5In*AX6IRvg3|u^W1u6 zTt?{^as-{JpM8bSJi~;gS06;nX`eTheK|)uZMD9HEU!828p_NWwFxXxHiQ%SJ*uNP z2^*DnLy){)wy>x=3ivDsgdBY$m_o#bDS(&V4AA;be*1PZw!EzD&EijHe--IksCN{v zw~T@W$0=GJNl)zTKyfnw9sihzr>798%$#1R%u!2te5>?F7p<8LcL^Lc8;NVWi8qjU zQMUXT-xsNe>Jd31dqSj%6hX(tH1hN3NA=|R?OiN|XSV&lfsa{(D&FgBM{R1n*I!>f zS~yZyb%M)o(Y?)RstozlH=HPG-6_}0Rk66VWZV5Zl&uMQ0KqGiy7n;yrKJ2T8feLD zTInZ5NLF8R6Uxg^P_Lt*h0KLj!a7F&D=BE3?4SPI$f7z=F!-b~Ry2Uu`D>nqvwcl? zz;>265sN@THu?854~#&L>WkcnC(7=LJ}h6zJUOmeoQ zMUNeeZD889>tymIkj>Z#EWOTIE=#UwVdGF==@OefmXU!Y0xB<>9*9)~lu<|?=wM<2 zU-1C)Yo%^*R%l4b+rwyp*=z>h-lSA(L=rt`EUPiSFj zQGumPFqhdJS4%KPaEe!I)qeO4hgR4#eq2LdCFOVIbET018=&;IjmU_=iUuoAqqz!_ zGOyWbB!&i0yQFJ8zn4q4jjGRcvK)fA?h4<;hVZDU(}Q7)U+MG2isORb={7;UYN8;L zUXSOfUAB+n_r2Lhv9iRY4wYS9Qe0ej*%px>+V}=+R4%+SQ-lxAT~CD1ZfIb@4V?Aj z;y9ll*VUqozJ!sGl7HLoo@LvkgBX@lGH~j6{JSVLOlotipyCF+MvfF&KYSQ0Rab7|o8AWJjZP4-nJUaQFt z>v9bP@ogLUt{}X*u)e!{1$nXluz#D_r?p#e*`=hqDl2bIAIW#-fLM6siJ~}nnGJ{6 ze*HgA<_x=QXNFGGTl0VLxRxCls?Kc(f!pKw>50-wZ64KICz`&Ycj62p8uI9;XT%87 zU>_KiJt8WlmjOpRk0Zr=SjfPM}SnU>$kS(bJ0DM>thh zP}mqAAaU;5Y1%aJ_%SEfty!xCAkONM#}Md5<6~l!Qk$z3$9J^}{hL!5{+;FXV;j=J z_0E@vBvaAMIrHEJ7@w%^n2$RaKebq`yjbePp!`P5`K!MC<8;^B@*JK&ms;Mxzb$X9 zJps*Uk!_EZ<3LSAyrk$N>qRP)0X)L|X^!5BdSMu=;Wk47pg#7_tGxiwfCEXPDv&*(53ItwV3pCTwmQ z%dWQhv8@exy>AZ@1ym?)wzC|Ox&C0T-G|+7H(5S?-0g_JJ$i588kAJqUONubbPO!Z-F0VmL+Y8Y$viHLLI8 zERP8JvzF36;l0W)rz!64U~LpD)=@!W#>?C@hjA2nW7zI-3r0>^+@(mi=>^y8kKX7K zy3I7ScDfgfCgJVNMC--yB%%r*$JNTS^#oY>`|4nbLBBI{RA&>iUU?QCrYH$BAe&ACt^vp{Uj zyS02G?8ROOv>bu+pJ>sMY}+=FR=_Lwp06PIzwDOKh3d{6gc@U_FPh|-a^hM>C>0|6 z`S>!gMG-!}EL^qqq=K*SI?nmEXNSulkV&}LCd=^sg1)I?7Zq(op~&*NZWBJ^v={m0 zy6>)iAMmkbA^1TIilc923PI=ssHWukK-l}P{FQiv+qrmUgR|ny{Ct{%rInEphyp`r zXED+E=x9moCH#kSM$q_mb3=mw4-bz|H#+R%8&ysxzyOY-4JUp^|ihoNq zU*~$rzjCY<=5p@$!|_Uf$&_}KQK6BQMz z{$Ds{U8hQthx+%gm9HcuxajUHd|{xFv;XjmJe>s9<4?|tz*`Hdt=3wT(Z}5|;yaT( zv92>hIh@Pz+_{W5b5y;bucw_A{H5F~O8=1K?OuHIDS3H}>8nzwrlH{*^`3BWa2Oc> z@IgVsV*b}xI196;j+Nzn5qoah`MP|bahvt#kN0BA$3bu%IMArh1fLs_(5`wQ&95R-K256$a`JJBaCoap}rSfCOSV6NlrF4_q67u4IjA!NPSagX*&9UU?M=HFFFyh7AOSRMZ<#v+F5-nj zgL3bad6=f$a&%d)1~l4iWmdL4^DK1ZUd93b)5ehS!HoXDFt#enXV3l|R4B_yH^0JP zf9xSgbhvO4%U)iyM@0CO{{KXMTZOpsFMh3*+sCAo-rs0vz4`RNJo=X|g@%Py*8fMh zM{fe9+yiyPB5S5p+-z9%?%vNjGR&HtO4C(G#B8C1t?#wO9G<`D9d5(k-riq3YYGa) z{+ZNg38{GD|4!iQNp7_pV;707jBQhLyvTZ2nS%UT`t(6gJGO4UTwNTV{Z*0U8cq7M zr%yX004vW=^pJXbdVs2IltIsgfiYROxR^6*YKQl8S=q4FHgvf76Pj7r7K$YNh;yMe zy#6N$p3bM)sgGR4hG!IW6teC%g+-EZQws-j%8C1wRm_0zdNTjtUpVOFW2)V$)cRC# zpR(U8@-OV(XWXeI;zaeYpF`g>bHzTsb1Cd?=S||xny(D!kE-QS;j#&Klr5spnwQvX zYnobK)1qlrpjG@L`1o|#s;96T9l?$HUYwKfp6%fUAC4#!XmG4!Vh?C_Qoi&&BNj&Z zO@qIW{rJTHA*tfI+zvVN6bQ=9`b+4(RcO?J2hebO#T0&^mrT_>mKzUl4U@ZXf&V2 zVIj8;d;Z)Wu2zCO#8(J=N^^02g!JCK!%jtoreV+A6OjWje2AjcBibd7vbmiXx<}H( z670hgyggapyPcZasw0a*l9PugBV&OgJ9lh!lrKD!x)qAqi=FAXB6gWmKO_kLFqQpx z2N*-C8Kw@z#(o8N+#%>stlFXYB^S%*?Ri9v&QJcsuYXq*fZQPf2{du|{M-@565;*Z$SFQ|c7lcEH z95sKn<*A~g#X&m_PjxRZudnS45xnYUG|WT;fivU>aq^AX2L>i4UL(8B_>YRpQKySO z$z8y)6m%rc?1MT#xAOzt4Z8cd7AH-dIXB2r3s;jctN>;-=&b7xt2aOY>Ej2=8N4Cz zzT~^de;3slPB~p-R~BW(qkl_&HjRxC-B>jzy=$acSzQH6$s1}I7ZrtwwTFeJ!NdZM zTY)qff_)NO?H)Wv-*3%vKTrWr)(iKdcry46CJVYgmM)GniHM84cJ10eh*TT;P}pPQ z)XzaHsjuH`gu}z1y&Mn=GcNPPv;DoRm{_=>Ou%yfm*9S`k-PY#42-dS%I0N%iJe+~ ztVD`%Zx{CD=D^&ZqYpXgJ;zE;|MMpS4;aZ0*exq*gnkK6gjE1m|}_C0Fk#XND#y%f9-#5KE;lFV)rt7_{ku!+>mnw2QzLbv=J z2bl{$BfKk6}t?-H3!o??>^#&=8YeF2nnZJ zT=33f zDk_?0SPBU^21(MVT}1}%{Nw6zX~$c^LsvOAB<-a3#dpA7fRX`g{D%GnVFM4< zyRsgr;?X2Ic&@gHrEc7~k-kvnzfX8^3Vicm4V0};H zOB-Hab~~2?wv>y#X?Tad;G<~ra4Z#}YmrFUG&N)k!C%Q1Ibo-;WE#9UuxY63^G5vZ zisw8CT)*z0rXvKNIbw%NG}e&BlP5=*w)6cT16>&AWEOw{VJZz;j;zkOx1^^}J*)9d zM|P2y#SPmDrTg`ctSYLez*~&V@YkPzU-|`pzRLISi+R~k2iZG1YR4bLNR2XYU}*qw z&cX0n1WUqnY6-rkrn`ihz!I(1 z&KL1-Ye3hnu6lE~u26u<=;T&`qmsq^g@xlS&jPNUbvu9jS055V>!$Y{8+e%H=PAoT z`~wn=leC+_JQ!b~AsW2;5Fj}1xDUM#Pc4@9hSV<9)dm+(_V^xAkwWoH2U32i8M2HY-Jzcj z6aS}JtL)NyD!20$N@unAamD`2wRDyAUsik3gahDS>Gl+64?w(!YkM_!Fe`EXU0Gcf6NkfyGx)e4)^)t0Ca87KHpTNfHoUDM0a z)-zAeVYp+zu+Bd;`E#OPlrC`r=EuiFkDZ#ryStmbC!7+89;f*ykRC9aZ9FLFKwX@u zV0fenEs?Q!Q$n6mA+M-tcw8GQ#nEM*>yMxIwzlSB&@CYU$l6)wST>@6f5p03nsRUB zJ57W=Iyc+=4aEW*v!fVY!wOs#xvAx-?W0KKp1atQGha>Jf|ZpOqo)y&CxnIryZ7AE zkiu_%I4l3J--PKVR{Z*&{BE>9z+b)RIt<{Q*e7YbfAbZCfbO3?pQ2Ux3(tn-Upb;q zm$dQdr*TlTfW0&f?IUIR`LiL;Gcql*HU$27^k}KnfW+kXU;K}`My%O>u6%ALFfd1d zp#*iG>W^i{9HA3FD=|?h4|v}n`{%b$)AIk~U}|MeAQ3p<8`?@<|7?xqUvLD|ZU8!p z_=8e*aCQCJ|7$daw4%)uk$WwHLz;Lwjzg{%_gP<65W5^9fMLwM`z39Sk(d7Y*b2c5 zm14?%BHtYM_rnX0hO*GIVV_>@%S-QZMn5%t)x)PPN#xpZyf`H197c%P`_YcF+AE&X z)mF&JZHfee-qw_(6Rcy@Mn=rc!0N3;OCU;1!dCGwmY)Y(FZ zfkl**9wNFYYw|E>xP<-cyi5Lu4~DjgGZ>?*n;oK((cOFqv<}Q9Ks-umnkjO$lap6R zHs22lI-c_r8RhQ>OjQq7T7|Lgp31+ncf%G<72_h}?%lf!X);p~)$eYBV z)FkPhT8P%&Gcb?IQxBd~JH|7LrZYYw{G#_eOXt`6<1i75i*Lij8NHx|TR9B&|ikF7wHD-hJW}>I!1)eVd?gS1(YW3is zaj3Vh?SKCWDE)*76iRrqA;Z1 zkQ|U$`276vFAl4bTevBaUDIXpVLyvXpzZ<%kL)&V`?UMXw2_L6%3?c15Y76_m*E_i zY*NYumD|&S)VZ*S2|6u9i-o-K3FtC;ByF~kX;}#K6Yp<;X`cGf2^1i{7up`{BJFRB z<8mM6_(>u}P+u9hk!nGO3cevWv;%(YXO-}jeyplulH$O4wZ6U!IX$_ytb)**@nk@8C@e%8q_>lJfH&$X4s$?}CKfAw1M-$5P{qi2+)$Nq!Ral(BX z9jBgM4LkAJeAPW{xp&?z#k89Gj|Yq5<4KSLsBRUBRf{Cy_+>*Gn0)F9)O6pzwP0#+ z!X|Xy=#UaC-n`J$*N?&o5ihUf`Abk$C@3l-`0$F0y8!vbR% z;SA5r%m89gVA~~LHWCEns_5m*Ei^QA532S*Heqp$7u|8-s#X9vP6h;$v#ksr-E~2v;)wPTf|OBaA!Rf7xpYFHosN4&Cgy{zYc_m*_utyun|7 zB<-_uo{MOjGDQS4A`Zwv+BpX&ryqag^Zq|GSs%q}(>2IvwAvofhp=wK0i`9nIBvi@V*0vY92d8iTtik>oo9%zx5rvyH-XeZae?UA2 z33fpC=utNarD14BhXu&TeA2nx#yz~&>(NtTTtY0!Jl?x8C>ey)MIp1ifUhlkiE7x z0+9~jSI2mv3y=7aM4Oor$_9hSyG>2D-F@WZ;E)`Z|4$H%LJ)KS-dlzwXz?zgFO*bv z^*J6SCsU4%-*olh98tBdvXV_o%BUOzKG9)!H?INSfdZml1o8UIT10j*;e(hmK|(jK z`>cirv7i8CEnj{hE_I=g;SpiwEhE1B`A$b$(0Pm%i7C)>6wEo?#8bsYATYKqc@yUx z>c)9rkTslo%aYo1W0j5L?oPreDvoQJ3cq7uX6Eng244L6n&@VR=2Foamn#hFx?wRf z8HL zXHbnxFWPeKugVt;>{MyKA7)9GmP#{-oQF&q^$A|8Fn8d=n$5)ATFT4tfp`bx)Qb;7 zoX-Ex+DwOo4jRLe6CsZ0#fucSI>yt|-EY;$qsFwiFv3g4{^ACz0SRX9NAzdIpWj6I zvX-|!RlHX+4WF8CI?S=*;x0$#I{!0~2%`=0&0SP~f1Y@89S=P5_&Z$ao5Wcjal+BfC`CH`yYGs$A_vS+ohDGc+*(X)QQ%|pZOdEB4wPtq zE;Tkh&ty)}NGo&s_&3@RKI7vKr)Z}p7;)FWatIU4& zQMQ5Z#-=6^n+wy^U3z|du13<#O?ZLKJ9O&qb4-mF7msvZLo+!%)CFnl#FLtF$I$p} zVfE>cedKe$>(G!Z-i`v02GBxG2b5k$;?BOx0&p{<6axG5YJzLPwYc;38`OC^%;cMW z6yhZPugZes$dTse)ozzUATHp!O9t`jyLRb!Do@XybGD_lKT}gohGH2QQccOj_%WrE zJ)k-Gl)$(z#+Rf2g>4&%O3Cs;;-|CMfsJ!(G1Aim=@9L4EFgYrY|DEYZvMlA{76)4 z#>xJ3_>t$rk|?T9dBSdA)1gam! zMCs}vrhLfKfwnGH0B|nASQx{FOQ9r-_O%`y`D1-Q_>~E{)W?XCn`XM9fzzAm=FQfN z7ZDXzWYaP>-k)EBHWd6VI2*JVpbTBdK98ex^kBvA*%QoqT`hcNSqcfrUj6S`b2YT? z(mYDh{C*pDl$&4oUAhrQQzXc>t$2vP3B|Nk+~K_AsJ2_Rk(T^6<9NKeJc8ZrO5Q@d z2PEkG?g(Ltmu8Wf39{JyUJY3Y`_@Wmk^cUyCTWMv?KQgvU-0K<_gyoUKd-R)`=!aN zLhHyJ$onDvrPFe?;U5(`KS_N4BW?|xmNgez@)h0gzWZINH$y@_j|Lb0n;I9?vfhT% z?{K-fnD#9@9<$Y7nUy= z7kbJQaEpY9NGn$FdPSqca$15w_hcxx+({+Ol zsOI?EOJ`6mdV71rw-OT-RfNtGGcrJ9&VVB?@nj>aX1UMd+ENn9Yjs7was1AmjX=#I zN)*m+_en!3&s*xd{$}YKYB?nivZjN4D(Q+{cH>^}DAp8XLe6)HB2DS`?mf4%Oo4Q; zgGoezxWCwh?Yfylfr zJvXx5uNS-H!Xx=12|^;zi-GZ$T80%_V zwCA3TXJHQ>+>2#~5UG-MRL+Z5DlCo{DoG7#0a0tnc9$)C;|74Pt zMg0SF*zZ+#xhlVIAEy-;^#kpvHAFrC9Hh?8qSd^4kl43T=mKB5`sbO|%QK_je^ejw zWJ$Vp($>JkDpN#5^+cbn4>LEz4;o(r!%oLiI?q3vyS{2@RM%{N6UOZOWBH69$o+V{&Nf7W&-g~{0(T&3}!UT~W3RBqk#sLQZnWq0hobd9QoU_vB~V<}shO5~Xx zpFS!mkB)r0)=|-46jt$q{Oxt8|3{JxO?dX~85yR*Y@Vh2pbjoyB?iWmniBbmc9FEh z0W%7=YTCWfI-(wv-%$(p7|g%bC%&*kbU7?a}f zZJv#??&I^E51M3Jn84?gXrV>eE#|-&>vrVSU{VfmyjK6_px>-PYu8P~ps( z>}c4!V&dbOWdq~8b1&$=jT*fWl&o&r7=~xl*Hr7in57jnl|g^4>duPUnJN{98irH0 z(G-%!<3j?A4p9*co=jUr4@Ao6#DrYU&r~-&8Qo6!eYdW*7PDmjb6n}~D**V*3QI~l7`k2N8ZCh^ zn4}Ob^{pO`Lzpv6F=Gb>m4-$WW;VgN_l7@J0LnO-N;S;?o(!b#f<{kH&JVWI6Y(%@ z<;roL*)Q$-9P^{$<$?Or6V5-t^R))*wta8ZS&tzPU{JNqe}``)CA)j8Y^P@9n*r;ipx zXJ>|G#oLft_{l~hfMIEShf1~Upe)hq2Wr#Cx>xToV00fVT|HC2Q+DRa9w7Wv4GJ*Z zap^%qK_nvu*bkT0x4y1T10Q8S<%HLi-iq#?}Do zYj5o@Fxf^-XmfWjgAn081pgSmY^@JuS0=T>*2-F&EUH5H4-};g8&hYI-_hx{t7hE% zu$<1F!zkUsCnA!EdN_+E!yK>1KuNlyuf|wT*?OS|X{`8pWUGA@#%)f0K5wU-s{L(1 z!kaHS47h~Te|_nB%$_e4hNK2xYMQL!-=4RM7ivYy*OA?z7$o=Hv>g7;kF&}5WmJTW zUx!SpluuW)`ZhDyK^_W|r|QE_PSq#?aCoE6%z6HvZjg&3-NpG!GQeJ*dLJDNZ9Xq- z0*WH_OC@U0*SI3NDvYW(Exe|;v_-B$?&g3*hdID`GEG*{X-hF80mGsi zCTM2S0|fg%<*Cfj!{~|Gw%~G|7rwpzYI&SUHtfRg8$BNbK_XcmiOHuye zq}iK~jP%8?wZ}-`5G+q`=8tGC6>*8*B1=q!&a+W^vN`)nzTLf`2f@MVRO94XReD&} zs_o+MAu>wlOa6Qr+ur>--EEne*Ci#|78%KIFm(yN`EzS+^z(}gO?EkaX11!St=MiD z)zEfkuVA)$h8e127{^|t_|m>u-1F-489rqtCCx~Tp}Mti+iY&N_gtrDiE4s98KH$f z8Y8*NAXV#gap#nlj_eZVqBosrm7y25n>m3#n3V!kdyH(_c zKia33QYWm~Rl~Z$s))H0k-D^eTD~_RZe@1-?pmMj)d)v20S}oZPfF%Nsvg zvQh!Rgym=L$S*M#OY4B9S3cXoBvV;G4wmz-zkep9_NjZO3b6OEUMSSwH%y;8pfxVX z3H*n?W+6;B7V?|cp$LQl2V>S#44-OhG*?%%EB0JX#GvTubHWzy0>0NJkWx$AsW-=5 z%06BsIO{Pzpd80#nS0X1L#jGQ!~{K8GGilXHa*T+E3Fgv+)g&Ji&GYKk=plX85Q#3 zR|M{U+|a61`&_#;RA!j^(*A?1AIYPHQ4_<{@reI&iWw$$>iQni3FrzWF9zE$p%qGef=P*R z{&_+Cn6_9BPY;57cb>)84)UH}dz)cKc7Esh;MHlzKL;FX^97Zt@Toq=rr5X_`vfR6 zCP_I!vx~V1v!4d0H5xUm-Yic@FNr?m)?F>MpZTfY%b)frdd=n(!?U_=`^4_u@_+^j zBj+($fvtQj^TN_Ix+9Lbw3a@<32{9nd!^l#l?(084_wFkb5?p4<`-7qm!TDCa1cF5<1gN~28Cwwq88A32%WX`XHgM&Xm9I?X~>r_?P_n1MOAb^;`%$wDnw%NKl zX+xWa{3qSbp-L$}QgfxJRZ?4HVq&0P#=BmRNl8&{+(>rEZme*CQILB3_B5VlINmW^ zF1J*in>&DxgpP4_X%1Nrcbg)&jCY5`z?Dr}&@3i5Dp20I?F5-Ug5qEdrufcL_Xgavs%G?~i6_6}6&;ol`3GsdO0(i*=> zJbvUI`Q;AX>M>#K!1#5^Z?@GrOg4rdyR&bgK(stq^TbFQ*`RWv;*E!G_7c&`9cV*= z@~ZP}e)G3_@&pQ`y`N2q8nWW6^ac5&j1+{o8v?JRnff_BM;{5@LUOZ?O)AebH(fi$ z8^g^v<6f5pvxnEh$#pi^>09Ndv^#lcPYgtE8h5HYrDD+YKDHa`A+5;f^VH)j@WTyH z)DK|h1lBuxc1^Q@vM>($@WefX|KW*_JHvtSwR!J@yT z5@0BmmM(=gZSRr=1tCJZXZrxAeruNegkp7T9ueRbwqm=dHO#2()lX+9`x3W{g^_SK zV!ah6QQeld`!W3cgykHJ9|$~IFDy+dKSADO%R|I_7c}A<6J?v8%XrLY09`35 zZrBfD@eUXr%PUw31PM(xP(2SSqnKFU-6_i25c|)>6_|OVU6ZyTvsYs;Lp&UtQ(x}k zi}BYOiE>&|vADF9xyeR8Sv|BSi2EyOBlv>A07Ozb#vjP}Q@o0b3(Zi#-MiWr5TGj_ zLpr#u*BMx4W|U*_q8EiZlq$<`$}{iNkau))!c*)`Q)kW^z!o4Qe!yz=+RB0RK-M*z z4YO#hX!XaBrzfR%5eV4dhVmyBGGQyXxqf{aChT==?@^+91qKMC2<4qREFb`fZ5yhR z4uMJxIgo1^*Yw**!q$n`w;d}NKc3J6xUjA+uvesuXy=PLAvx9`~T4dXK zIh2CKo2^=&Rw46bP(*ZO<)B%`6lDJL7R+OR$RZuZbm?gwMp{aCEblr7&<94BGt!C++&)7L7y!YOJ@F2X_ob%5^_#git{s2B_L^Jj4At}^}WPh8E-%`vklKL`n_-ypO& z;mH$QD=YW8Q9VnBzrVufREqNhK+VheAf#D%wKX~E9NzPwa9(T!A)_CntV+k>sM~B# z@jf(N$$?ya%4%xuUEx0FQT*ps;4S%9O0$$~#`#o}=fCyb+53wV(lTM#AU87_8hGrF zJ_Gg(lP7o8Law^c748@d-yE8cq5z_&+k`hjw1`xRMC#ndST4Xz?7Ty%vD1&nk)(8J z%jPEs2@C^8q4GPoA3sUptFdACKe*rR7C|?JbeiVm(cc?4WE>ZV8F-_ z8I<=x9FgCA``_N2HRt5u_tq zai2~>b%8Ivq7Q`d4`#?@7v&li20m#3peUM+rhcE4GlmA_1zwPLW0>q#MpykI(nL`#tA9 z@z2@U{=;i;U2Cp6?|I*2{A#>$O@-Y2s%2=9R~8m_z-m1;J^c|^{nj(3ki$CXpK+&C z(9w>SIw@WM_(0==u2KZpG-&StBnH>TP6(B6#uwP&p&$&kf*3Tz17U5*jDvvK9uRu9 zw=?kY>;o9Uy+4EZgAlEi1aAMAiPWzv2c^de0ND~%$NSc&!WhOk@Sbz`=+r6h$inDX zj&H^wkA;}1M)24nB>lg~VgA}y%hjdvVGd&xNJ;>9$Ww$;%M0LMI7FQuo3?{DbR4#@ zVzm)t=oa6-)@yX%`H9l~^3qnQz|#))VNn(q@Bv90c<)>*R!ZWe0TLOMUXbtf2}JS; z(Gz1+@!&hK<{!c)Xi@n7{d>^OWM1A{=KE&>E&Ns#Sw4hocs$SpA?MEgDtD8o+j~F7 zf{f8yFlZUvP05Omb!psbwDH=x7!%#ZhA6#Fba#qyTnrXTF=Jbw$emV9v~d$L|D2ybg@YWn_$C3XhM3bCdP>lN5tyFGw{YUo^E~RY2VO565s45V zi`4sbo)B|qNLP~GN0tHa572_VC<=bIm=aw$BPU+z(K*~&kj zYq0fncjxpP{sl>Qm3O9qf` z73;nn{^;3RC~1fK&ug_CLH_=uuT27g7~}aW#_&q?vTXi=vFrKOfF)=p%s{p437QjM zC>5Nnm|T({R1GKWfMNsOg?cT%>D6~c+?DN%ySZu*_R}UWw_@_)se^Y~{_xrvAa2n~ z6+{npw*jzLKqB+#k=k=;>MBlQ2(Rce~W>Qx(KY@dUO6W+>F>gusqDmUg?TP7#LZNqPnXT_LHF z6#}6F5Tz&JvAYb*EpIfLL;HNz7kWL=Pk8(IAZiAjV@QgJI@24R&z=LUy^NTF3~U;F zOcJ|&0$s@?j;M;!-AJ&(DPsay40N@zwRO)dp$9Z^zjSj|eLy)d;r&{ng-Q|DFVW5->_TJ|?@bYX z(?9BH4A}0%W!$(UG6_U?cXxMRU$NiSC6F^?UaFrTic{kmmB%gWT>PffBHWoUW;OS} z3%e3_H?&uX5&(j1>Ynv~*UqOUlo-A>Z@nYtuhAZZ9tAW5h~i>9d_9B_(-|m@bE`_%`!nC9X4%w@G$WBjz4Efq z;d6C^e!B6-&FG*qT0EJ4PZm{ltxgYr0Skq##;$wbI_vgyIttyJI+0^GsO7ssFiDMBYHI%1I?8ZfBBwBA~*Z$nO^_X zoiYRW*_7SmLj%^|k8cO6h)ZR6nL!jE0~M$$DBk2ro-u)1+DLd*v%@EQmrdz=HX#TnB_#!-zR0En2n@Tep@UF( z4$E9YtaTlWrACw?-y;8;Fs*7s@}?E-Lzhq9tK7=|j2HLVl>DtP3hij^Hd#(`Dwcax z;=T$Fr_Y#@^!Gs4hW5LWv;OQy*g_5yxKrn7%uT$56jC#J1m+LSMNEtc)c$PyB{0{( zR%2(ukP(ihV~0fC0i+e1UZrGeR@|qO*@1&)IcC&nzYR19xyDYiZ_)wVupegl+ic3- zeC24H^-~2=$FQk$`1WG1$}9SGRQkTch~mdi*yA%ZkR_1w))vBd00#ldVRZBW1kb|= zf)ju@7e!@XI3i|z`Mw1xjNFe2R9VvHcEzLe_dhYI6sU9`CI&0`yKU1IXjLPbZ&F3n z!LIiUz?;FPnW?FhvolCy%?A1w5Gy+n!v&SW*>>$1goMn;tz1bxKMSE$Yx?x#PE{o(2}#FB`!m)j+wXK@W{^7x&G=TK`T_$k^P#OT-p@d|%{d%A;U0;EeAG3_XzRn$g1quo5)B5m%Rfi3 zhx=FYFzO&gY;xUG$5H{EJ#&up6c*dT@L4G6n zGU&5<6dpIaZNu$OUV*Y3locUkd*m-?(SPXsHX|6ws=^$?W^dm4ubs8);`njq1B3m1 z!LMekYii8xt`xOxjg9Zd%En*2%>a?KMeio?midroa#UBwSa+ABJ^@AcM5u-QwpQAL z!!3qIUe;Xii6)dE)HO4Z6W`)@1Kh zZCd4%n_-DgmaAbSDt`LX^Rnd*F*6{`1$7t@S^Iu4O;n8(AAbmr?!P}bBl4@!%EGHp zg7c3tjuIMw)08uL$lt$8nGS6ejaNuL|ApH(veh^|@%z&*t zG%|9Og%Y?Y$ePf8P)gHj>rZQH%JOhCkFk7C;VZnVgY~N<7SZbC*8GAHw%L}-xxJVy}7kC-PsivdM&}4xvKE_V(}7((b+KiZ5+mjR9qD82JJ|KK{R7P|}mQ{Mq)vIkM67nh>W>!4-lfrZtYp`gF}rZTwsbu5<(LAO&gE9U9*X1dnYpHJ(H zstVtkxrRnBNB58}F6*cYXtfBgJiZn`KtZd#bI?Ra2oJG8@c~ zLBH4fa60pWV<_5$cJUxcQ{g&+%(58F5C@#%VOGKMO;r;+Xn7j|3HwD4poDWLG<;Qu zi0^$leNSgehEnmX`G8@hK~M(>ForVgncQ}$ayn+r;-LHM&FPhQz_qNu=Bw+K`U9?}>*DVuY} zz)|?16PR0z~>1m`PI0NX!vAG_(u>;wt5){peS%O`=U;MT_95U zyYH~TBO>biXc<`T#|pzoF*m!TVV8G)3-eD+LD&u1)0tP$P-g$w&Li)%%|~MUD9Bg>g_Gq&P1Q}ntdX+8akTbL2FP}Iy7hZd|KwVRi3lpBai@(5|r(Br&&j}ff zy}620S7>J8hVwRGktn!38ukNdG|l?c3N(&ECjcgozdMxj3%#qy`darZ8{FbTzjN3duP%iuf&;c0X|89Wl%37tJ>AaE+YaR6@KLQXRSzT2%S8bcBNs|B`=YKyB zUk~1e(HongoLF`|hi33t5%@KjuZksD#6_{hxYDom^d;|CVjRYHIy{=^C&s8#RJ0b7 zc6Ud}79d#A5qd%$Fd6|6YkJTaOkI;{B4PAb@Htmoi67CzA?Lad??xeva zbejz!kl%a*l|eA}@^b0!$CS(mR-_VViXBtOn@|u#EItP_bC}MYcVD~-d|sBoTlR|a z8=D-Dn4^ayN5AZP79#~>yCZ^`r= zWLXPTR)MY)hP|O^4`9!O;b2c5o=6oWBT$G_kQhJx=g`HYu6rhEl_X^b&Oqcq!D663>6Uy7?V)y#U7s#siW;M~?&kyHa$xVGlGFuw62bmFhR0 z!es#9JzApY1(6sC39C3JY#uspl)9Bj#+MmSYwMnnjri<<>;{qH;o1s1B%pZPZ&sF- zg}#2R$pMO_%txG+R{$`A*FCxPVZkO$p5|-l3VSvUQz$3>2^xd{r_G-qdDYJkoi2?|W$hqGp0r&49DBf|-y9cyRJ*;nx z-Fb-lhBgxoRv!@Neqgk$90ZI~V|yQHS%6NA&&41HVDab;VuIE5mmw-o z$4k}ji3b}!eQu5@u9&uLZub)*>p<5UWrgrSqs&I5m#lnzZ#QR^U_RD4q`KyvHg!R? z(Dwp0#w85&|E*gEQw4qF_}3Gtr-!2Iv%>aA1xM!b>hq4tDXWF>(;#=5YoD}Rg!{|oRDA&!2ncL14VFN zfX)S|9Eg<5Cp4#x2mKtgaRV4&ADabB`1mS9r(Gq=)q>tlWnKrFf)|KW z95_vAqjSbU#L~HPelB;nCV;+4lzC&kniK!MP}m^4CUl3+;eg12)s(v?Q_yXTk{DhV zm5X88XK=G9lqeGmk7DkfU`bX5uu(1_0vTRZRJ4AD7>it&Gz*@>=I;>)FF?`1qQs!q z{99#m^04HUieD=&G&+5KoA7)znonom7C3rA=>+;GlfMV(Y_WF1j*Ce*ehyz*C>+~m zWl)6YtKP4PrW90NzM$8JlXEvA@q>fP_02na$BmJwjaapo_5+X!3C9B%<1L{s{Cn7} z$Xd5ncS?Ehp*O!qbA1srL&%`u;Oq?ixxkwlAPUYP&rt(vbSRw5|JF&bE88!9?&C(; z8s5gJOMm}<8x|QsSW;mfAIzBp%M}xo?7dt6)>N-OT{zB#hP$NBm$A&Ikbc*awJ5S} zLgxVtQ4JhGutPzezt_vn%3AF*pyTSk`qwW7Xu(k%(A z*!w66#-l;Q`)AXok`?*UKc)l`K!^Y;dq-_^hykB*K4at~S2S_u1WeC+sS%leLm z4eAHjAAwnigQ|=O14`7SO{XRJY(u_bvIO%Ke_aH-h^R z{%RI{pFkTpFyr39!rIx}Q{%w8apMLyHk=u;cOOa7B4WD3y)y9X z(I4a(2uorx?(J!iltC)W^B{&LjQZMIu&5R!Mp6n6cXi2a_k%Pyeg+to-JKl>zW6W) z6w98(yxs*^Mywp`c5l~wX@@$S^bA7;N+Nix5dmw`7~hI^P$AJKzqkb92$D=Lax<5A z_k$*T7)HgtWA0 z;iwSZ1f!tF9>2t1pYwnx@@-&KRUQ4iLoyjQe*0wrr^Dd)?=60Q5Goz9K|GWY6Juy+ z_bxW}bI~wLF*!K~LKO#o9_Szn<(n@Kb>(;AXIUZoL)G;YF8Bat*d8E{9?p6&T|;v3 zA1y*WJj3(Wzs?Bd)$M=J4yB;pahR>DrZxh>RAZM2WsBi^Sggyh>oEEQYZmgk=(7&u)>@4o_{`i7&o`+4+^EAbl+u6qMUJhqP4p|gKB9+? zFrAnUWRcpOfl+(65KbTdpVqKC$!h+%(EW*Sn#~J&;B(4TK|QRJb`p0_+)k29MZPfl zBKyW#gA7tB9Bm+GR`3FeJ^dIGU%xILc!SiAL*Ri*1YX)AH4Tq&9^OQQRqpEK1hdOu zYA!%!UXb>;>7ngf=^if$in7&dpA^Nz#dXvj8!Fy{>S%o){d3khTz7yWNbe!1dx+&( zr9~_xasA!Y+8_9Qo^u72n=fa#}s{mGFk#wzLL^*2-?HGOG`T`U>_SZWB6>} zJ@Nuo+?b;VrWA3NAwBQ~)f&b}vWMR8FbVmW{9Q^(yI1%?iHvnzUf83M&rl0c?!Qr1 zF5KE=MeDTXl(60~(7%KDpGh$=GveOD7u%0H!bj4;3BYN>ZCdCV6t!7H0dF%Vkt2nF zxluv;WQ*}n#F`G8MOVHRdEumfee+4f!f#nTc0e>w%$_~J8I~L-a39s+(Mt!|l3}hU zNnx~sYNzdB7Vja487*wp;!9@DCJqh|#_~y>B{UJ$k)krth0VrB*ciU&4n|#*)ffwC zI)ox1%?ttow~gy!XEN;e7~t6zYGG!%ARD5YU(6r|GYC57Kp_t;zJ2a()dg!7+Vw^#6li1fq09zl#R*u*1%-t{yAXNsn@1w`GF{;H zXfU?_(rhy%F(q*sf7B|qGc}bi-U3Dn7k9W|Tv`T8nM+k8&gY)hL@yJ0II9j%079=3?Q`ycIL#zRsjHdI8wcX)Yt zY_f=5uapmJf8qu@QmZx?Jw{z4UP>=&8FJWukVK9}j0La)^ha2P45Z{T5;)s{7yX`# zhdsc|EKi6N?1h?;O<;ZQ4B3^S5CM5IJe?|scdi!pa!}M=YjZ_-+(9LnV7Fl)D;o+- zyr-O@A@#LSv669daS&t)H4{7%))%i`Ew(3v2ZI5-JFEsG3x40@7a0)|*Q?UT9?QI@ zlq=I6<5D-He+4V>vly z%i@eLVG(1PwqSbtd;xdX{lB(QL>@d81^?Yqzl0(5f9I5bdT)H1aC4JCL?SaXJqLdp z@cHZz{AuO#7lYsJEc`)Zx`Wt_tEPL!ukCLshz7~heas4zhtmat(Y4SpLEp?aguPj* z9>}NeF4266Fv-GL3mF;o3i<1mRu+_oDTwh+RW?)H4<1}V5sNrH=Fy0z4Wi)}SQGB- z7{f2P0k)yP#+HoZx&O)T=^6d)D9)Z6d$at#bzpT=*zs~{?IET*!R+!O_&4Bk+P-{Y zQUXz_5~JR&CjsX#u#U*GiRm)=II9FxdqMvF&;F>#415-H z6R$)=UfA(2J={_W0NSFW`oPi=pHc^5FG^lsUjLw64ZE@O3HZ+ebu^)E_JrCzfe@o^ zb7=`?apC2f**%h$h6dtqv6m1jm33a3`VS~tdJ#4lEFbY|_ItexYx>KqGoS=3|G@16 z0Un?T_oE!PXlhsq%b!kVN61)A@Zw+j+=kC4<6lC8uy7N5`_hyF3v=@tlP-elCsvq_ z4Z|>?^5g{u{6PDEU{ag}O7lHCFpauG1{K^0AlF1;CdjM%02-B0>NhI8T9)EZ<0yVj zCJWDsM$7*ZC}I#t-?=GIEowma3;wStg|AcEn6~bB;l^49|1E8a_4MD-))vcHc&X#h z_lsM!rGhEYtfc1y$=}e*1wf)Wad z0^*7Enef#yUOT)KXc}9t@TUKmEfo5B!LVg$qJLaA$}hVHx$FfuX0Gb_%-b>5`BA~g zQ8qiC0YnGy`SL%10Xz8qenIU4lPOWGuL7g6Qx)5{C$c@dL zl5~LoYhCwegf4?z5h|9m#Rv}2cSkg~&^?CB=~;NieZ6mRi-bK^_PFF0l3y*^lPr)z zSb=>8T1m5i`U=(6d*3I&+ z$of{;?5z}xn?$zc0I8r7?nJfVPa~a{C(ew{f*3a>v$(Pglx@d%1OFp`V_QpQZDXT% zHu~vxM_pyRHKr7e5mF(ELN`0(cs_sejTKio?_m48VGEILkKma`ioX1hRHF2>v~Z_G zXrqP)c;Tmm)!ud?L#C>5)xE{-0o~~NYJ)z~%XWP`g#P`G2Pk=s%AYo+0pLeRFofGq z_4&VLbu@OW^_YBulgrBfce;EuB>hB?EGwmC&Ryyukai#v%Y=Lpda(EbFi?euh9;>a zk{0?&Nl3Jo7jpYIp>iH62Peol)CO#5E$?Dxtbzes>WUzNmn4JM(5rbFeuS{gFB4E<5(foDZ|2qLQgA@v> zw8u9zEgPGA7!!=KT_#t=89|Fw#FUg9+uNo31nRAwG!WDO;nwB3QA=WPg`r!RwBzw{!RbrxU4I5pMQKt;3r(YV5NeT3s{OKY(AyM zQ@z#P_*8;{gXl~TmifxePY4Re=V*YqQYRDhYD6@OyCUzBN)e1|)V&bQ!F$r-a!)r~O`g=k_$x3Oe8-7E?wKZcmeNWJpqE zh(PYgv(Rtd3$H5AWP~c;hN&jx>L@Orc`^Gvvs(U+dNYs>;EqZV3N8F=u* zznv}Yr~Nz84X-0YLAstyCsJC5o}HkSLAR5;x0~y@j;1Uc==d6Jf6`22J0dwqro(8!~ZPZ$o1<0)!vU zMb(*%#07oyolHpze>FsO!i-?}Q3_nY7#>_xj`2_KN6r?gn1IJcDl^54_DyUJ+L_Q^ zE^!N0BBB-06Bk&OiKcp+X&7BI|n*m1Psw z68+fs59vj{m8)$dZ2IJX7Lg6pT+kyI(^oydGgN;C&js;Y1)$XWzecV>+hsMH4>NGX z0d7?7RP!`wbXRfX2REd2iIjxsQlb9J?6YqdXW@~$;g3mH7+?1sZ70y%#}_|kQ9M>5 z7nMFecq%t)C?}P)+=eIjLpZL6LSE!!XgW5!OZ(tApXpwm7G-J1-Gc49K7?~5kP$VpG3ahmeA5JO9tIj=J)8w=^ z+jr<{tQ&4>F!eJ?967^8!h8?075^2{!U+i3KhO)rtpI1PoB+Xka1_gD5yEVXnU&$I zjJ<8V$M|z)KjeL~gXIX;WRK2i8!olFXoUbS-)Q0D%lJn>9?!|jusjQ+$HKzHRXNC@ zkArOO@9E$=orAEGgE*-DfcyX{eer`H`kLPr1Z7;g0^L~|a@#7>=dO}+$`oCXs2YE% zWQHmo1I>`QQEKg2URxx3uLV~>sg znIV0!H46p_=ra%EK(Vc62PgA0>jB5E&ES>zv8Yiy>ac@hNzo? z&*t%Z?BY+gZ+mN{cA1Yo3I~%LX?uxOvyABJ{?~^l89G5JUI$|#Z1_(BYzF`jUcjQ_V$gH8s?8xIeODF~6kJ$X_~@|( zei6mwW8lX=ttHGhgJ>Bsq{YZK&i&>fDLvwG$N ze(C@qEzy#c0V>><`@ki21kxyt{8iBKuoKdyq4`B^jVd z><>-p$;om%0x(atnKdiGDC8kxhIs)ws=9>z=lGa&O&bUm(BVJ;;c8e_gh^u%%Ln3$ zNKoz`gS9Du3?>H>9&@8$EC;fiOq zf+^_hN3o*evZQPy0s7w2i`80Jqp7k%4A97pafDw|3*ZOTgP_dj^0CN@9tJ<^a)W9q zOcA2m59WsR27npWyRI;02=+mcq9Q3#xukx_wQB3E-T6R>AyM zKExF%mm?B|i**tnq6~5*al3cF)jziiCy2`BrLWpG)|TPpiKXhXPxig1n#z#H zDo0M6tU#4A$=PM)D-~`^m~iIf>*XgT{NAfU(KpRW^w9HssNdsCsC2J+?ZUg|`*y!a z%Yt@tQWB+r1CkH64%6t_lN7!o-?>W=D8SVY{y)ZOG8D)eQIh$fFwx2U|Pm8 zwC(Qhf+qvowGX4*bB6~9(w*RMtcw8|kHC%fgFyamH_nV)rNA{PQXf8i2=phY0tjZ_ zUX4snhWb*b>}7^jSM#kOf~D5s-YFBDwS@IicCW7oR0PC;3o+D!I0br|PQVO>ypAv= zWIYl_}k@8%W|$ zpJ3g``?5U;=ub$@uU}R)IA=aLED8t%K9_E9Z%3Dl%1jx}?A`=t0@Z9=hW;eis|9HkB1YTH{GOfIqc}GZfrC|&FMj0P)wvn^iLDc~H zB}(r^n9H~>NXLhUKHFnDSb|dHPiH0yn6fyso@bPl;MzV1dT@h?n3Rz4)gz^SnbX2` zzXC%1u&5|gwT>~7%0J)kY$nfZ9!9?V8!*`2)02GGSIo#j3nTK%Ae8pzP0|}cix4Zp z&Ed#;;O}$!=H0v3#)ti&`=|kpcyaM2M4^B*VA)ZSMXmFZ(wpbtz4~GtOcg(W>(}q! z()^LpOnWmaH(qf;w~Y4-m@q&XAtEYkVu|Bo2V-zRm67(PxnUVOyy!+Vx!6UE-`Vif zL5l>iYTIV4EZzFcOi?4<{hKkLL3G}2!YzUbnp@XyF0hZL!L-ryNm#?axE*0SU)R>83+!Dx&&Ac8f(yBprq0?+5wsWM=L+SxOk=dJ6yt`8O$YB*Zh8_ zxZgrC{p8o-O`EE`tQq*2D`KGZKz={<)-~$TZ0Om<;DhZ1I?+~*K|6zCo zqo+Y3;ZPri0!G;>5vl!N!GKYelpZxD763Iig56$9 zDljv10chuIf=w@$&~@>BfSMaz1Ft{e=3pLYG7%RK_@q^n83jfrctr4lF;WK!p`&>-$EC@wao85C;nWc50CTBjM+cpyBo3GccS&ANQdF$k!Cg(chl zUSCJd*%?TcKym*D_Zb$;J-Yk%33FdGSi;PoXbS!<3}=cC5W`)^!uq*a4{PHK5?ehe zhoGFy=u#~pI>kzilLqJ*J8o?3`d2(hY+gHid#%GFuR3s!0TTT-qBy4O+}t!8=m&z! z#X{(IWdAI85xP($*=0Mhqv`X1-GdF%hy2DI6rsvkEihj|Osq8&|8DJVjkNUigeHm+ zvE&na&KH-+b0G6-zW@^l)&-H3AvLce9@@_D06$^yRDf>O`}hy?hx+>ZbG=-F7s0)> zd^9x3-HlKSgolTxq>u+m1WZC!swPooKbWKq1f=eGoXr|ZUR=d75iS-qPd39N)Zpu} zwz1*i<9i3zb$IVH4$I7;I7^#coP@z^?;iq`-?1t;=$D(9m-W_x1Ky3!F;D={5i20B zM9}+~lL)Js^)>V)w;*Q(+EWT}Tm+s7dYH<*@v0=ZF^QPz6fz5zr;$|%AM-_hF~5Gq8BzHA}9Y?z>QeXg1-nm zfh<>^fR$_-eF#Ofi?j27^2FD#fmiJE?(PEIhuJd+2L~4e_E(X16;ZEXL+(+6cBm*j zyDf}O&Bbm!x1j-g4ed8}YinzFe?iK}MrBnM2=Ns3YP8#7ax{?h``V)L#9L6Ah9L-m zx_V<1lVLgyb!u|b!qQR@;=}0t16hc|L0=3*VQ=*u{sK{h1uTCXo1#~*B(LgmsHbD3 z4j*>G1_CyEM6mB@#thTunadgAVeq8j?c;>NA`BrLd;5W~P3ZI;tV~T^KtnM+JdD^N zw%_=}F%7!~3{-3D{PINtlJ|gg?{~S0jSWHH5*n@50JQ)9{rhN&5FbtaoB<*(X{40d z3e>F(4S#HHH3Oe?S^X1;LTE#R7v<>Upu;&mJ$2L#1P2AwR$l%UdTMIfSu;L%IMJ+d zNsvdB*^?ybWClZsKRv36brZX+e#Wka^@s$a1(@xG=%?%0*oH`zDG@%G=N{k=6CzJm zsD`wS!(glrGe87B^F#ai@gsP*P_C{9NTmsab)QCTbh;YbQ+Blm!Qgg`IJa;yg zGBvh0F~#OkFtsyxwxH(b;^%t!kow<$?4FjcGxO0z4&SG=baLQ}iMh!#u~b?5Pmep6 zgKjTmzG_llQ9X-oPS`L{lOg!QN<)9~lacgB+}*`B;m>}yIZGS{cN`}NsQg6w z4SX+Fb{0C^&v)O4AUDKI8g+G)`uUuT7=)oe@I&5TTI&dT;(z`9-D9P2+bJ^zKnP?|(w z&uMJrdwl*yWLT1_TwgTqvWz`7E>P8RbMv+ut+9?pvE}&n;F1`1N(Mh8h33YolfK|r zU&l?IVv3$`4e|Y6zQtoYd-3TtUS4PiPrF5uYaJ;^#g05yF|NC5&4fdeLV3zZAq5ju ztrjgK)Ed%zMjVgW!)TIe>bP@4h;*MRDGVhE&FcLqs(j%#!7221o8r1_j`p+cX^*VC zeX81bUrms8Jx{y#$dz=Z;bu>cpPQ>)Q96H?z1c&zoRdtgWU`RV@xZjzRO3CNc(=xtQIL! zwx48ph;n5zVdh%tEhTNEN(|Kww?u@!ldLI|OZ8qa)6*@OOE>ptAVFWwKvuhlWS*>| zC3f-^$$_Z=L!Yi=owI;DFYrO8`Q12Tam!S3{6NDfBQ$Zw=a0!}FzKnAXZ9TD#9bXJbzAt0_ChizrM`#j8hc36r z69!*}zSXAW%c#NBRg#KFtKMN19CTnb!u;A%*XwAr*g0({mgXJ9lsgdaGWU85J=PI5 z%L27<)}LusjI}@jbuhV=_079!M~7+LuiQB}*JmeYa2n=_cqS4;J6hSZ!i@bRv4VG~ ztKaxbD2=WvpmcqC@g?^$wKi!RGUHHX?8=b~4ei4KvsfnU_Ht%vg+K=DTl%_kIW2N+ zeRI^hr_OW4bT_*1EZZd(ttS!jVCZu?7S2!~ie*jNpWh)>mqh z*+C)PFTz4jxVYkx2KEjev~7l(r)=aCs0J7 zyOtG=SkWLsF-zQH*1r}*@xTwsa-UjMf$Iag_|`@duf^Kwk31Gvhc_k-2^n&M--0Pz zq@rIKClJL)ggRlHJYJz)e61&bv4*#HI|mo-@UGRcRycarw4vAT+ZDBEdpS7(CRIT*E$Pg7TREq{eq z&8mI#+0g2JaoMurUCFpz_2C&!>{}MI_B--;6u8SaSlBVW*TUG8Q$*iH@GXTU{nn(l z>f@(AlKoM9ZE~t>{p;o~F7+ z?U5#mazywt?s!jTFJ01<_*AY0RzEs8Ij+wB9Q5bm;5|yzoV`P%lHz`nH)*98foc{( zGC^-Y_l%Gdv*CDNaj%;66ZM5+(wX$`X{0T-yAIL~GtGs#_ISOwt?pT`0z@uf7j{P|`QtidBe-XS*UvO4ZVG7}iEA;p`Al zzHR*QlP-Vmxy%Loz#TvBVn>DfFUQ!cmwZ$YiVKJGI@|N7K5k>*J*_fn2;e%~-;H;2 zzUHXOU(8qZ;F(r7HV^M%pB4W|JvpnaZP3nn>++5)C5@fgD+wo5IAPRg}?Z%M)6=QndhuR1Ixnw zbp34X7v>93m!NVIou(iaYm23K6iN4GOP}+If72BVY~s{eA46}M5mC(Z{G`}i&0i&m z72Z;X88J6oBSvi_vmoM9IQ1-RsoB&^EmffseW2v)N;6q(W1fXw4qejK|-s+PU%_nR+`@Lv@&%rOa1u&OYP# zF|?b&lxU6uZH_w!^K|863NL}XWwX)%C&*EKv&yz@_k_vMs|VY*1T}q-Pe?(2NB%tR zwhNI@q(j=y5QVp(O4*BstA%)hO?&yfxI-Z0Oomxrt#q`lva3~-U?Vo<9tG!wbrD!Bw!$)VPuFGH; zKOyNUSux(QEcDUYU0zku;O_ZR4dMB`Yfh=?#i{Y>m#fPOQJ-|bjhxv5%oYSiT+2OeR*`8oU>0c3YSS zW9WQ|Dt#sS@VRJvuc(|KN2;PF`M68h*|^2H%7{%{we24FeRAF)o+rU(C$I9fCq=hT zsrX4dYd6(6V@&VlN3&U&PX!CIak`E7=U9w)4cnxc-LcS{Tx9dy8F!i7MQ$KCa3M{R z-m&Vg$Rs;JF`v;IzTOyx>UfW0H-)DwOYUuv+#yZW43kVzE!Sq*9IK5ccD2|05aW>5 zBg1QM0%qCERuzQ8logfo{ezp7U4u!)BDW;(|E{BQF!}m+Tc>F9_5Rxd<7ytA#>kGc zes_V3Z^?{Lmql#xQ~L?oxzj@Z71x+karADo5j<-6>A7Xems&#Np;VYD*P_Kk%{uOp zKE82(Qlzx4wiHA3ml=-urBVmY>z%AKloM-<{nLA$q;DqODn9hsS?qJYn=rBHJuRv( zmEQhVKUG(ng8o(Bq2HW{`F#G-#de2-P5%7La=SN~TwY~UKEo)c=mBwS_fDP;V4!u2 z$nV&@8yN)Oo?^JR_B^C&CdYz*Nl;7y8!JBlBEk0s=gkIDV=FcM1*HW=b4@OlXR6#I*Nn(ZG7bC=cg#hnK-Ylv-N}- z{A^o$`t_qWsBrjsBsr*cKAL_D-G9CPfVM)0vucl%C+jB_5k6(FUUl}Sd>KZ7uB;%M zR8ZU3(0ExCOJiMbsSvVQ2IQqB8<`urU7d|Cq5BJ{gr;+H#yeZ1v#qi}@tEyU3x;V_ zH_(mfi_MTnlf+75+^4^`YPeB9eQhZm^WtI)H|;hVPa{2jhgiLW)38X!Eke3T`r+R^ z0hodXFS!&AKUq8+`e=>vMtZ~|I8~-H>W}wcaQu(dRaKaeCjTSHC7yOtC1JGIPaBVDI7|j~h-I5xHq4Wrp19P!~xto^1I+hy7A6`Br6l zc8OU0v(qV2dctGLx3X^6aEJ1*CBA(_=q`0%MOwFEHGmx34TlcbR&-lQ*HkU|Cx-{o zTyn8=sVpJcD!Oh5UNYNMtNY#lD0hOQbW{rLDO=%tLs=w*>30=P+AH6i$&|&NapmZR z@mDbZpxmr5vwe_(Ds`@C-oa<1lbT{HYap9B@5^|yCuOao zcd$0zRWh@WbUal2T}GMGQwfUG&_gR6Tyj$K+q9dn^3MG&?y$vPQ*%&L5tI2XviAIY zl`m&LHZG;&r+oS{IR;xB(tP7e9&}B)wxWd4DI!kYPnxu5lVnm?Sc^PaQ+NUUYS#-g zQOjiWig25!2+Cw{4DCkW65x_Dvpch>`TB)6osH{y*{g?qw_YMYh@*i44pp?$D0Yiy< zjJK9b57NZQ(#0a|C;AM{Nokkb`)6X5^1mg<@1a!Om)DKPx`AbDsU2g|bnoY2!b{@G z%9dPR6yr9$(w2-vLEA$XQ~J$ z&6_%5ac{~+yC;bAg^)HJI(&G+`P;6wF#DNA#2p;&s$10ebV9LuKmSe<)Hy3y99Wos z^7z`1Hl>}{-VE6^X(-25IlSs0$cd2KYv;PKT)hQ%aEIQ?&=_{Cjr~c|#+^Lu*^hF4 z@VUgGU8R!JvN!RBm3$+9O&v^4t2(JHjUIy6+|y?AQn}Upfy&V^x|| zpSx=#ORFr3>MSHV;lCyikUB`j54dG=E0}aVik%fLIhjr1riW0g=Pi@fUEBJiGL7oI z+;wC=4rv8hZZ;XG#9qwK7o;;RHRD97uOv~hahfRP@}z%a_x}zP!s0f4ljXuipeMj5 zaDQ%mba%_pVH4v{s>9rCtjV9wAJLMVe?ARyRG1q}l5sF56cSoE{(xWfyuKZi#1_R4 zHN+4ZH>zC}xkgb~pDTFjQKKItKhGVCcHTaUq{g_msvkkGQr?*IJ-Ei>$%r;&YK{Jy z%FDzo&$>xERDhCe7dJWC@?r3c`J=|5XMqAI<(a#$&FiC4lip{fE-AS+V(Ewxw5Pvp zH27TnbhMt??S&BcI@`nfEGh8=(y>msMLuB8+I2KrbJ=~FGg~LZkE!J?!(dWYeCJ8AkKTOo zkj?5zAG^){{MIf^e#Po1ZBp_K`uv7D#^N)K-$#tukF7_iZybK}re3DlVs5wi?9boy zfuYgr-UQE(wiuJ+WA|?aHRzF}BXL4o=#-~nDE!xwadnasEveTy-!7BNxfBKJ2CE^J zni%{U4D9gG--{WT*>uQTUfI$SNGe}xdS)Oqd%e1q^G;75Y1dPd3`I8-&81=tUb#g@ zt~eP5F_r!w2P`t&JG}074;T6@Yd>I$>y_6Iy)=}1(tn~HY4V(jvuE7sxSrfWyyAvM zO{$W(3PX5P)Q3UpoOLAjADzoSXpHo;Bc{?t_@$0k9g4dj+})vgi@UoPr?_i@gIjSc#R>;^=g#|n_p^1^ zx<8VgWRjiPnau3T$n!uPpbrF$5`;o;xKHJf;7$)bTO zazlu%LLa4Gk4+dd4Bqp1i?5sWIIv4i_zz_x5xD8p^A~IkdIo*yGK`6R2_(2p1bZEm z{4huoJao^3Mw@~kY%)(7Be~4O;&DFsiiuJ>L}Gl<`&CVzxHyRbzvKHEI~n7HrTs2H zOJ~yVYNLLC3Lzb-cy0D7QK`ac zgLi%2ORuJ>dQ`UK{iORUcD?pz(X%U5>Of?7Glrs-Dodn$GHUxjOoIyJN=eV9^?Fcag~`@bE0OAy~!3+LZ=X-0$SYhdj3 z1W)}k1>KHZ=w#w=&;$dBgW8JLMwt96E0*WF-gD|xuH@40V=xos%ooW+NU5|(Nk9jvz~KvmII*TOQ=mYdT~HvBmyd;k?)$O32wPT!jfO;ywY|>hPt&W zUTA<>ac_Sr6&2YB`n!@uxD*>F3461n!oX)WH&DzK3iDnY_U}0|{Z&78Rt_nZ-+gVJ zGaWiJSOqg{2}~H-8Sy@vw*4Th;BQ`~sgiz3iod8hkfEQs_Xl-^;SPD>)H;3)wSB=~ z$8}f#qwJ+ozBgAT_faUE`_KL99aNFa&M6GKsM=Wc7sHtUdkC8k$gXGeEneX%R`Vm0Q36NHJ<{X=#C)7<9)vA`r@I^j)ws2{P6&$ zKO#FxMzRNiJO+VfwqacL7Tu=CsZTz3@8M2lA>EDgq6q6>y^P!rGu1^q#ZvY9H(u(V)lK80Q-Tt-azwEC9#l?E!gufHiY4)I=Esf>jK~PG`2=N z&|E(K0n0s#t$1&fTZiXT1SRtj8>QQ~8xn2k$|(O8mkQ#8GHyZrFfL%;G~7WDyH-KR z#+07A%feEo6?0|b23s@fY*Z$pPymVz;t(72Sp#4%Slh;%kb ztVZIOLN12Ug5@X`XuYaUQa>y?uMyxhH;LK!cBKB*Q#NwK+E36(5|5e4Qa9a>c(YUj*N37?wfw758CD^J zCzQGiG9!WenAnub#wWIPN>On+iPTZg`mV+E8TFQ9*cU1D!Rs`IWcYD~zon^N-I7c0 z34`cgnV?zXzlG4=qi~i*Cgxd1$nvU6QeP$gG9q}!nG8ueQ|1^JSJhCPApad1T(9`Z zNd#924wcKd)Hk%Nst2*cR_AM82awy;3u=9f|Ip$Nx6w5ln_xd_t-cHkZ952?d1z}% zmKw(Gi3!UWyUQxsK@gf#>s)IsRzHW8Py=gVaOW()?tmtjchs^dt^1VNW`x?F6QjmCyUIdO8XNh)?p~=^ROP9s!3{FG!Mv4U|JGH)BCbaUW2+jK~PAlgw-`C%t;0A{u*n0=#8Kgk5C;Ex!4&Z`JjXNXt-#*5Yc8H)((`eVED>N z;C!X67VScVtRQRULiOmD#k@Qt>-RNH2xN(R5BfyFzdf1{K8=zn9%R0~BEON8y3(k; z>eW|~oO#VoIdvtuYa@o7GB~xRK-yCxWONqK>&$kbgm5s#8~P?NufNmWB_j(S$R~EbR?#N+NGO zSX8wY&oVPkFvN~0=;lp!!c}5EBRtBkPk7WmY-$c}9t}S?2lmd+|f~%ZVb)eG;l!>2O_@%p*tO)+=o9 z!C*JoY6F!0mT&U@Z4W+cr%CxC`VGUtku^(=Kl$OzTJOSJ+}-q@9C=6P=YuXjI4obe zp_yF}zco+JQE4W4EefE9j!@XlY!U7_cE(Zy&Us*;!+^XFn(Fs)1D znYH{L+?o*h_aCbtr}H|UZ?WE!hop?Z_4EsQ{LUI{_SzboUq`a1XZT%KM&3}yKTTQ2 zX2t#ufP3W6Y@hTBUB6`271kThp7wO+6Bm{*`O5LtL&p`A#s(EXL7Qw}j?`Ox!GoC? zI4%QMFMJ+SV(_Wc>Kn7e-7!LXyc_j?aU9<6J!-9D0ZOgwz?LaXn6;hf@{@qf-3;~E za(W12RY7XTZJzhjd0GJ)am_5}{99iFaY3QQP)X2VP$}&4lDhon%zbwq#6+dxLwt$f zDdzEUm89D7ArgZHa_yBpPuh(sZjKmm+vAR19mn051!Kd<;S)pp`g}{$eA~O+tCwd3 z=BTJgy|o<9g%Uy=Ls8xid8^Np-eF1<_(CwAu3N$1%^B2nac1)%(#0K=phO~6_x?MZ ztVzSAPnNM>nGeA=x(rs8Aga4MwTf=fY=jnbhlO zeY%$fnbhNetuHra(WS2Daytn?aYt^rOXJ}@sQg`Jyh2IJp-z%@d}Y+fcFX<5#qzGo zjBKBHeO15jT&b-2rJJpOw!=<1(sE@65Hc^vQ^=wa$7ZW%Kvq>h5IZFBsC}qu8nkGg zUQ5d8Si>NE6egaiSV2PFfdy6fnuZVS8@*fXs^dOztCj4m!A^LB^D89L4v4$+`fxIM z)Y;~mbKIr|H}Lh>^cuYNP^MVV9&7=|5(W>yF`@#WiE4PoNb|s)yuvpwFOOwK6-dgm zu6AUhsynU=ic(mk?L_MI*1Rd%W)BVnr4HX)9^RKdxoVAoN#s7HIFiHXJiK}bAt9%( zch%+&dslWweH17BJ5ab@%V*H9S_5*d-JS+W36E63Kt?h^Qh+I zfORUmD~MW|@0a37HetDCwvM1x^2j$8X zGAfvwJcg`GQgIGl8>>QcSlY{FjgjBN;A}NbS#5uRDl{&`$Sf&`!>EzDY`XY+9hgsO_;8&7{_LzR|SLX}0|A5SCgu z7cI1L6(Uf?_bgS^$3|(xcnk%_~sK9#i zEYilUdHud7%Z}|gAP|kH35oB-8XcpdlMunm`|n;qID*EdOj(F8Pq$pv5PXFvRC~Cv zFnpQ0@$?}A1w<4(lRNBh9AOFCjnb{wg5qOZiW@~22(Z(ZjzE^UfNHmRny}x*PKl)5 zs$sq?Vi;cLgMCeNv6j)L;ml&C^|M>;N2-c~i|LhU(eXd~!IWJ0H3 zy9`=!B)fLjf~G6d>+VPe+IZ92*f1z~oSWK5Fu!p*H(M1DESanY?Uafy5E_B}-&Y=! z#X|Fj?!SHHadxFg`phUga)wNAS+t?S8^C2CfPis#@rT0JrA69D*d9-_@5BLOJ47oS zv@|4BD+$Vq)X0P51ev$WVOaBEPHUE&4Cyk3zeu!YO8qG_d&-Jb9Kqb{-)wRtlw&%{ z=f1gE(dXj#s&D^nqI(wJ)OWRwnyNAGjyR_>al}31`YMvn!}8>Y*5@_utpco{x~Ha< zjY!G*;bV_2;fQ(|HE+8Dxi&x|ef`zEh>-EkB!}j)ASwBQ>}ZEiH-UGQJ^5nQMC#O{yiQ^^5}aUF@Yok=zxR4Sf!O0@D($s2l|rXY}WvKe2afpj(1?Fm$jpg(2VK zd3s1TK4hqyjnA679GitaLcaVC>c)0$Pwjj4r7J;;hqxgxtt zhuQXU`hakD2_!b$6ij?>);3@zoS(m^$Hx=E_g_q zykNAmyH6w?K2gZmwc{PkzD};;DxB(qz4a16XM-pN2^xGll$zYoy+vjO&WQy+# zfgS|rx|p3OJ`?!JJ$g=rJjhVTjp=&x{U}%LoMh8Dj5^_QW7w4zT$+DxwV2n^EysJH z^K(;S-{*z`CH~#Po5_~>%|=TuP4Ec>b!2JZ7_`zgb0JBNSm2sVv!dR0y8*jyh^c+h zm6pjZ$vAi@%3%n4%?@hMxARg}x0|`qC=IUP{}vOEq{l%$$us%^7uu071klpTy$NoQR0Z-#=qvQR)CqdMxX!< zzFgaiB@5T>_P)so35%STogmPgLHSG0yc$Y6Q&PmMtDoF3bCSFAAr76K<)(K3-O~qO zxTNI6bplsm8Q(f)LL$EW`H&BJsZetdU8JUaHHYZOKe`0y^*KDi^Cna0`g`It#at+E z2bo1843SmkVLf(q1dwP0SECV#!~4ao3m3MP2aPXX%>p8r#jgTMr_8@mi*FRfc5hb< z`_?MLEYCLBD#8rd{|RyYdJVMSsh;FzU+WDO`Cz_N#8n|NQ$aR6;!`5~MkknH zj97A<{9V|YaL#>A1@oN_)c=NmnD4wxn&@H)UqU>1GVhu<`g?v8Ay!X{&nK7{5x}V$ zL)?=m9sT$dybbZ86Al$X{g(K1Pc`v&rC9ODpyKRb9_;wiE@7glje^exr@o4PqckZU^K0@W& zlAK)=X->`@+eSs3PR3Ntb+h0gyw9yY(D1Ji&9L%;8%#$Eb5W)pz^dUh%~^_O?YS3V zo5I@O20hy7{{#%4dM~Dee*n7%hb*44pSm|Juz<$;vB?6j#?Uw3x<)%qpH*EFrj=nl zHeQL!92qxhy~&deyqu-)^I~G}Q_J#M203KN`oyL)I(zs)dD3jlg_|;rgw2Vx_ub0U zEXY?C4PA{y$&z$!@}^Kyxh%2TruJ+T&MyAlwkl~a@H5iLzFJlK6>7Ka7t;|l_UVI= zQCXV#)@w?r2uoxx_ek&U9y>lat2oTS`+!mfZ!>P+X%t(#1)CnzAj}?X!n1*B2-6b= zO50|`7UI%I1DTlTL)r^M!}ykEFV+CZad~+BxpSa+z7wb5;U7x!cSVL--nEfxr%khdm@c{RYcPr3L($l#y<6StPU zp41eIIrAs90sA*tYj$Yq=_^=FXW8-xk?D0>a+B+b;gK4?vKil{Lol?u^8qDWxO|#B zlyWeg20PIT7|2-~I0Vz(n`32!Y-ROfuP(bkBaL7`6Cfz|ll_yQ@eSvyqTctCUja+u zJYsFEaDe)kSxJ2OS&)6~_F=&e1$Ym;*!BhdzR_-@qmmBDlI48!YSUljqZ zn3XeZd{7;JW2d7{zcvUND863q!;`}?cLeR{+9Y7&n0ww~cQ33-y}z}NYPk8%5b@=} zr6+{w4xv_B{t0%Kn>{u$w5lUz1S@pmoiAYII1Xx%sXDvNS!(5|FisJY;B^4Q(W7l@CYHP zdZ%)39ALaoo8{S;&y*nyZ*x49gNaOdtb8Rk-+uWL3*iRpRnpboPwCVE8 zY{sTExJZ{?9AzUfut@_wKE|r~*W&$HDe_YHue~GHwR?0xmOor!o@|_+lH)fvSYer* z2m8tiUZ=DjH@P9s*n`q3rgGZMsJ!t1aA($ZCX@B7JIcBE7f!sN%M6FHo_csb;oel0 zTz}nObLe`~nV>x6^zMM@AYqr>85=pK(0*J=vOAG&MIV~+K!fPAb>eq)ro*YUz3QPD z1~_igc@=S=HGgcrA{6&{!-eZ(rntRp+HyC&7HCpmd2&x?wW?l;qGV)tGFTb8f<2f|EKS-^9OVG#Ak4)cWdAX9<8xArm ztcTyeALh)wCK9)Qe&E^6l;#HG)AfY)x?0B^>$-jaF*Sp~63sFTZt{>Gmbsw*&>9G> zhX^F}>{j_DIJQ6aNZKjI$BKW50_yl4ibfi>7_l{;0L4@gi}xf}7nxtiZ}UmVe7pn5Sj=AD#FPEG zwBy2uW4)f@i6bFxt6Q61K1A}c?>ZuxQ`7eAqHU5qRjkaMT&1HYnA%i<$LKIO6wS zl~_;WA<}%Ue4_t={&im6BYy}AT1!atPM7#gYncao#c3n2D7muALU#^ZnDARprJs;F zLV%oerDGMZ@3z0EisQ{?!YOH0@Wv-e zkVwo>Cr1=0(4dMbd2$R}1aL~HE5^-IC!7BQ)loXscm@3U6uT*pOVmAXE4dztD%@3H z#kis?f7aN|w`q@hY9uMX)4@AY&Ep`Xs@lU)xkD5xzO%sT=*Bav)h+I@9J007Qf*+~ zJ>>?%(!QF^Cl=!K=#R2@bu!#Y9jSU`xL)z=d_=%C&Nte$(!}x&D0whi`1uEClj^Gb zqZ=!`LtSCFQ4-(tk6gMeZT84Sm^pElQQ~T@fa0&XzH<8}HJRk4w_PN;ZH1sOg*4TqORg(aaU&OL{btNTbOUi9lji6sL1LT$1mGOsg4b92{n2ofzvvU=#v}H_P zwsp$EB2Xy50I^SMn!SI7-3I@~?50Y+08R9Zyc1JE(TF8kPx?sCyYbpOO-)|j?-%z8 zz%rk}lWlb8Yn9W;1_hNt_aBi{IIP+pmB)k?`^TD|+`iG%hLM9(p3=4J3kx&eSicfP zt2$mKq7>;144Y}~TP0R0$;GSGfI+Z3E*vFfEj0j(QRH8F_52%s&XM!P@R1j4*R$Q7 zxUavag1e;i*d?q*VAZ(Xma56{OrCzl;nlnHqQQRGYh|b6XFJ?Z=8vD}sTSAI=w?wb zwo7Ud8pVpbp&|dp!IDQkg9*Ho3`yhY=m#sDn#ohU8UbhjHj7S4Wr2N7(SlA?<%FI_^)GS&4E7w$ym%b3SAYXbuhIIo3no55ZN@{k4C!AQ*nB zG~0nJbSdY)A4z%;rgyxIx{!x*dbqrU^Aoe+Y7!Vl>l=9z?UC#xce+y}*+a4!1IKQ@A-K!9|cS zBvQe;QKTIUt%ttpF!f36` zCu4zZSAEuW%qf&D0^TglDdzK`t#OX+<0#Jaa#?)uZ=C1a`K;$j+kJZTrS6}$kH_Hd z2>vSac8-yj)=PUa`RzIZFJ{ua6P`PMBZQH13)Cx>+wmeA~4+Qoz^Q1-i=9HJAQ&{bxdMgzHa|oZU}ati`W1e%7(S zrknZ_7iFpDZ$s;)>2>X6IoaKurfU+VarLMi?8}ZxTBV;C!?@=3=37u(r3n{8M_8&h ztBg1F378%>u1oC`T^EUTS?HTT_Hr~BY1qmIT2+;5a;gMc)!0yl()K?2QzUH#^q`NP zRnKq6`a?-Q=L$L`GSCmyd*<+1a_=pI!8J!(Pe+Tx%d7fI)hrHX1=MOt91k)NO_DT& z?+g__62SdUSzY6%;)|`NQ2?=_INO=khpE&2(sPK5hl@=PpI-sxGH^HkTrjD?O>0|C zCLn9U{}jDJ9J#cZWHtDtrUx2?x08#Q)XeYu`enL~ntGopi;Z+)Cb{F zOSrs)2vL!RrCL<_D^($$;7+lX;;54!Ntd5EkukW#W%TWv8cuW5GHl}_sfLn=28 z?l|;*ci7%JNJQa&(mv_Ql%9pw8p($3w+@~}km68}!CkKY{43Y2sn0U(kD&u>6e+8q zk5#WaYridiu{uj{k5|8RAjiijrh0RB>SMN+f}cS*%5RZQ`)rnC3g77#rbGhR;(ws8 z{$0Gl#79u2oWswQ#%4t?`qj-%<4-jx`c>_2wdwXj=Uq%#ecCVwr-i1JCJE;db+UDo z0?jOTzG49#bEc#^a6Kuz2SF>%g3xAD7OD>DftbmeapphcDlUdu>>zt+CEaey227=k z5t^A4Y1~4(*vhA0qAt!vy0nE+oBN4=R_Ejs#j=cfdU53sm(*3WjhwG6pz^hTWEVsc4;0;j0$fv2-cZFw} zRfom7tjyooRtR%TxhKgVK|J^*HMwGNy@&mWcyBxZtGyH2+r7SYTmKJ2xS4lYf8b=e zQJmpf9Z>Wf%wDJBw^kL(ditoHdc$zwu_T>(e@)#D;rKk2diDf8g4bA@uXM}1-M^>_ zY&m=9eE57OD*(OEbF+)Vu^gkt+9SySyujw13g7-egjcA+t8e3xJc8>aWU|t)9?ZqU z)+QXE%wS~Z9B?R=2^}Gz)B;Tdr*rC$+$3t;-q`L1{h|9{g;ZzwOa9cS44pze8q@Uj zG^QWoEY(lhoTS%p@o~W(t*gH=l@@cl)oH`fh%I9k=b-+)=b|6B&=`QX)rj$nKqvct z!!to}ITGdF--`3DAAbMq;aF`Lu6x`v6)mF(y~0_t9PI>kvT}wBO%7|iBEKnfru%?D z`*xU3L!`3KEj@(-Eu&v*LMc{Sw>)?baomU+bQoq)=lR2ZfH1SdWh|F9sB8eyuEypr zO29N$-cNe~@rnpbIYZ?AHc(Br(S4vnn?R4>w1z9$ZYr16tF8oV5v<`F@7}TQA+Y7w zc^MM=qcB-hUKkEJL|KdO0SxxVaKB2$n*j6IC7)^NX(Vk3cS4N>EMoQEaL>YX*e_P8 zZ<;n5%lG^5QqpXXe@d&h%ew|GTWF4EEgIpQ#{{h2rZy%Rb4mj} zAP%niDx>> z%t4?3GNogoFU6@hnhw81U^8Gc&>RZ=d8TsHeW^0zp{wD?u_)lPA&xvyo+!kx#B*% zlBKc=Qo5(P4yL$b&9{LGbsU4--;^7hCf$?We?B1VQypvidWACnW$}5*`0ZpQU*y%A z^LE+xSqj&eY92qPe#Xi$baun=cG>*tukwDw2X>REo$g7&s^#+I_a3 zRiLO5QmfY5QtxAzQ;i~bf2dcJf#CMo&WOU-VjYK7&x+6f?# zUua&4a)s;$tk93e*Z!<49N(iFNF;|$#eMGf#1gyAAw>kT<=Kys5wgAF=W*TXAYVu@ zP0S?|E^+poUp%JSvorfs9de`kv-1zz*k4$+IyLFjU?T?Y@yflaj*Kej3tR7eifGfq z_EZ&=RAq*IeEQrG5ApUn9|Nf*1`@AzfXs6zO}@O@GVjB_{DBXFhEcu$DR$zsvx}|K zeNOBuNV|UEz`lYX$J~yF->&A)+hkM{aDp|GV~~z7xdIk^B8EjCM@*A`m|4FZQfFfC zwzA%a2g+8Sjgx0U?*2T9si(2%fNfBe<(<9gr z+p1RblSX}S#_iJnB8!^-5z&RcKPNw<> z>IGm(b1f<8ROyDXs#v!A9*xd_yK$w$$Hp~ep-8d+%#Xdd99i0Rcv1%=p41M=aeu{} zmGSJ&*Jbo1VnFt=%E%_ELAeUH5Uk(ajQz0$<=&-NTQ7nUhkbsvX-Sujlhyis*Osvt z;(ws1(D=rUWMmSPGSe|M-eOwX;O>)pYATW7LQ8DeVkyp9uSU=lviDtHhy@3%zP5W8G-PbYJ#SJp0DB zpLk}Pmb}lW9`4Yvd8Erfo}k61Uhk@AWA#=n`iPXj(^320Kn25J7`enMpy%3p5<02S z+%L;FdWTyPQshu)B#$Wc+8D`>F}vf@edPPZjv*o6a#IO8vCF}U+x=nW(Y@tG#)@t6 z_-*t3aNERihTphP-weUH@Y%x${ABYQ*AxC~A|o1npJ%;?g%fKrW$AgY2BFq#zG9xi z_59`TzOHom+p$Xa}ce6qp> zvz^jt1WKeE3tMOZ@*-lHbNAiJT4u&2^jZL%!&4f?;$IDg;m>*T`=SmutjfdDLvq2> zldAr7#)M@KhO%zxais_1nQ* zPwph}%U_m#yhprQ#H{U|hZh&^X3h-rHDt$tvIYwW(%-vfy46{pIpM1Y)*~6$u#kKX z6rY6Evyqr%We}UxJqwZhQ}i^+kYMIHuAP6EtxZ03yckG>&<3M6;do*d@J-=S^=4A&o-7<=zYF{_38q9b^n-yXSLPW%05FpZG>Hiu8WA|hK zmCLSrRiv8FEILT$)_x(7E@SvUlgur}3h=4#Jx7^Dl)kP z0PBjE>zp|TTygB#{i2jap5)kr_@<5=BwVQyHQUmF67%Zm#s-*3gW!m8yHv>@rneBx zu+3Hdw-`o+TIoBtm5i$zh&Hbek^15tA4#K-i zw~yrR??dgN%|K;|8Y&q)=@Bb>DOAkqdax4d z%2Y3ve0IY&!hqzK`g~kU*RvK#fy6SVK&qP$T)P9VQQN0{)mp+5Ag^uV23DZN{|Y?j zel$yKRxDtTMct1}M7mMC{A*3h4yW`>1iIMv;6)ZBZcvQ}X+gEC&@w0U5?*hhP%@tQ=Sl835F@9JD|?ClQwpy&HYJT4ZUA3{2-I|UqZ3G$uMYYoFh8H7 zM+dtBo2*tJYZ%Q#0h=tJ`sT005wB;r5d^n@iD7Hs&*fGS!~E`(K4y)Msx;m^&fgR% zv#{?ho@VG8)SVC~gOkNO)-ki_y0G9JJjaKFf57t-tQdlEf;NWJO-hjE-7%=<;nMk=C?tE?v{w31_6I2jy(>OZ~%)mSD8%Jrc&6(;%N}Uih}lo zhM(MZjz=6vnj!ri*q)4xK+BQ9*+!?`BDmD_rj3!6rlN$$K)o%F zkU;9EEYm$Llk|wxH~A3O>gPpp0=;Fgenmb?x$P~4eI~ej$Tm<2AXH!6{F7!Xv@y;= zynevUZaBv9*Pfa`mF|e)Z+}SPp_-8xkK*_B@v2oQyOY~*-)jcFihGbSPA!QBty0id znKYl;7HUVGE4baVozD^@CW#5m^$H5r65qF`u@cbZ@F9@dSc8N0l~YloqKMm?r{u1* zcl(%{$Bj)*Vr^d9_cpa|ZIL`B!BsjXX$5t3Dh+HgD6^GVOfi_t{a_>dJtbV}W_Y_h z>7XK#&o9MisS}3l^pR>o9XIzSul@${@sR0}z4o?OcZMaom%}AJoX7b(T!w2as;6$6 zoyU2S>|gnTs=9?D8>Q*MmS~1+3nG1%rA~uL)&|mNy;=$UOx5CQ>3z2r-kVahX#IB) zeY!W3RQvE!k6{lCx)aYe%}%UAFMkClw&1XF#7#gHS3C^~4{O)r8nubx!{(4{fpGLy zt1XGWb=c(ZIlU&(saJT4>qF9*)b z(w%tyPb5L>;)nH%MdAq4<+Ml?P1Aw$6foRJ-_fD;YKLY9>yu*^C~-G9U>wL6}COxDFc%6-VeP!wg6Fr(;5*xM!dM`ee^Fm_`PK zq4XC#aTD|TnhS9RF}p7GkUD!Gim-!h$#4vmpe+*Q%4Cz0bHgH@R3%n3G1?dPg)mp8 z2Ls>FIN`IdYYINQI=QBemr_sPITou08JCW_1>;QlQI>krZbkOtIE2!F=qwbPw{Gme z4cZ(M;5cc%>*LQ~EMb*;x8F?B*nM@g&7`Y_YWcTz*2?Z30zC1zd@4puRY6a%Fn&zaE%#ot?K>Tp{>n`}2$Ts^WSq#GmfX8)jP1v( zSQI5AX_$qW`iO8xVg+fPnu-OAv=|+JJ#|FZ$}^MZGCE+{hR?5eC8zYxU-|YU znjQ2#2=m5$R8r}k#I3E;m&K<8JLsmKfvk72)nv43H_Z6YNlP%+AXBAx(q(7;-v1@I zxi<*w5c|NsP`8mr+)tP2Rfv6_4R~-`w$PmHBr=q?(5wwcx7HPiCN0;nFpkI76O9V7 z52_vq(zwR96P0jN#R{>HvtGNS(y1S*r$2e9PJcluiW2l_wexAvj z!hRGy2_Cq6aO@4AcoGt&__pkwigTP=0#&EHX6Sl@ta7FKwxG)EO_UaOQm!Tmqg{Jm zZqVC<0SyngOA@7LPZ0)nWC1d@t$Bcgpmnuc7RPoR-4`26REwYO*0|P;BH9;d?Q_!`Yizxt!*Pab~D8Ai)yc9i!dE% zD<%uR8DgsVxZANAhG!NR*YCVW<&rc{<)Suf<}wHL%v%BUoYVF`=}$?bl+8>GBT^fR z9@d|d8u1jkNxdg!LieX*eW>);5V`)DVbW5u#>GHsLU+h(IHz^XZ&gB0{iIK*{J$yk z;3jV=GtML*c5at2}F95gq#9?fJQqe`OsUH2HDWSx8rY|)@6o4%7mOmqQLHejg&(jg(7Wc z{&h)f#j6IfjNH{NZ_%p;C&g#@S~caLP_tqM0b7N=0QD`Hs)X+KCW5G|tc90twVc+At2}cBYubPUqQQgtf|ZOSPHXOM zKr!mS<^V?Y5&%Zj7jk~8=-deGNg>VAf%c8-qn2-hMpB zMa6Osn4yb=H0pCI_T6-Q1hAp<_Zsmjmx$dh18>)?J#&Jj2|2UhrsYB{uj1zlVt*xb z`^tq*)h*d$+05u_gOXc<*wuD2=Q87E7%(fDe9`g}59uyBt%In6iN|s!MaW47%nsdh{*}~nF`}DKj)uVOwNvsM zn=1&ILO<)j!~#3dLD>61Ek|RJU*~VUt_Q ztGsh9iO1cTH7uT{kyC$c)J)=6$ePQqN&2Nbm-3nu3jsBhu)REt{kh<}%6#>=xVZP^ zDy3u5#+`oh<-4s>9KQDpKH8%ujt50EkwlS=vEB;m@JA#!;_Vir|mGB zxo$1*J2jj%A77$AcV)@QmrN8ZH{IE?3)H6(O3yak`dz#Ws-iVkh!uvbQZSl`8U6(- zb}?3nDT}e9ic&hdgxNn&)qQvt%^{dxBwnmhA8u4lp<&Bn&#Nio{5G3$VUJxZ7c(sL zMizevlFp``>z z$w$y%A~fp&g^?3uLTIdJL@VEuwpwJm+NDCH>YA=tGD+01l@acp-#J*PN;kL zp;e>Xtmcrs6LG7NnS7{&mpJU4FNBL0UXwlfGV_bW2((&to(1KMU37->zB%Q zS7V7_ftpDM8Orx`+;zm?P)cUZ{qSW~Mf@oAlRR)O6aeppoR=u2dk%OfMsL77;rh>1 zAoN=2O6vkBf%Q6ufOmqOCX8aHW=)lpkDuf7}Jx&^7aW0K)5d&{gve(;hC)p}wCuz4n0x*_OlWO@5e566iRV;U% zBU| zjv8LFDsN8Q09(y^dm3Be#PT!{i7^vyZD`L+?R3WWwYtLD>+@2Y%*pX_6mua*ex#+d z+Q&dqW|bI89nbXn*~cAb(4J%3i^Mk=i02ydGW8J#qWY{iLc6D!9iPH65~Zn+M$D7U zac;+7EZUZ1(<{$6ADHeM(LA8taYGG>8w`I)TWxY+9hrN9ks9=)2j!m*SKOOfw=6OubOI}jCq55C{LPT zg4@mnjxL;@hMJL6-yrF!KE5|6zAGg@jy*X-`l(=B(AK$d8Iy>u|2W!<-t2uMDOc?% zDR6soBDWKBOq8Z0A1^%x?lJvV`sf(24(#hix75zbw}<)*>08vAlm@)lPee|j-sFaW z4Nq3zeG*fFz+c@|M9SJ zd1AIB?r;qoGktPzok8-(Z?DE8^=rF+?nGa&EwX4_w;zBH9-0uhB2+PF9{%IuJ8fT^ z2OJ`rk{q6mU6t@vG5pi%|(qNNRJjG#N4ANhuj$C9%~**WvZ? zgy+OxL~8T?{nD1`=5jJmBHN1JwT!AZAyjI&IlEDNMeA*-QER8;J`nXbZkQEFO-G8!pBa>`^y0%5g!999}dhfr<~88 z`ed0fG4}g3{clj6hO5i}FCG&+>wn`hY4|u>khAEhnAuvGyCbovdziWZTO#LRY6XBx zYTJMSY)MXDHY65V3mYqIcXBp%4j?1pWbfpv?rdrfG?ujRv@y3(lMzQ^5w~%7`)uJV z;pE`#euzl)3b6eQZRidzCWYj(ByIFM z1|Db{>o_tM-`ixwAT3(`LJCaG*kVJ$Lr_zz>@YuSkTqZ&PaikmCan;>pfKBlyLuxd zX#q|HK3bl;be@B}k7{)P3B^oE!^6WvPF@iB@@Q#k848CwHae=B%*Vih%+6yy)A#dd z3BY6lut&NQ0N~>Q*N-^><_e&U{{CI>yghJKEf7L3Ygh)rBTpXyiL{ijb~$7TQ?IA+6+uS0*1 zYPkVZ@Xi6$*Yop&a&I3WG4E6AWMiYq%}GowEG=ygdV4*pVgeeJ z*SEBM`0$~u{2$g20M6KEAL>0EKv5o?$L@dc?BH-htCH8}uBeEjNG9MpF+Dxq*Vnhc z?!OC81Zd}Kd&U4D>0b^27&;!Kj;IR2ia0+%ub#!n$%$>Y4O&JSp8A%Hi5{tZBoH9Z92jH{aeZHWT_Pe*yVzoS*ossHf<3MbTW zV||^xHsTuq`K!|u@Z>4G)ntQ}$|oX1rbgb8#t;@sj-W#Jb;^S-o6J=R;8-qzoT6;(j9vq;T`=ONzWvA()qrvG#zbqOK!6XJ7XTEx*8kz{tD~x3yKbd7-Q6HagVIX3w6IA*KxrhTRHQpZx{=t@ zQi6o2G)hUMAcBA(Atj-J#9gTG_s0G1z29GVj5E#{=M47Vzj&Ut)|_iTYgXOR8i-b6 zLsWw+p{k99+w8f{39wdj=dH<~i>}^6SXreM7(eX|tSgh?E2pH_+=F|8amw|spHIYT z!5YF>hMBCV5XR+>4-ZF0MzYw2Da?WW0L;b*cb3LGUsPZOuArcxT7i1Zqes{m6zv(A znOk7;HO#!Wu(0Tg{cnFpLo)^J6liGFrq23AiqTb$jxI_RUhs3 zh1BLMSLl1f80)&{&o5oNBx0WMQOVJ96?{ctuyS2p-P?z>>d*~DH?@G-MB|-X6T}dHI3h>Ral<5YE-sirjlTK` z0

    SfXK9O%Yk{9!B+&P@sl)lFvUE47+Sc%rZ>FBEW^34#K6edaC=fAMQ0;-VqyZk zKxygdeD2+o^4*U&HANyZ?cBE8ZpMSn#f^6-XqMzdnC$HAYGE&r+%SB?H@?0MWmy3?Ok^%x7@3O?* z1xZ8tJoZoYY9Z!qX&71Qu4W9K5 zI3t1sO86wagXzzonVY-d?XmBi?th89eED(=wP1KA*z-+*p@1QmvZ8JEA*@;}?g=Jj zd}?aZb5>|rgvG`ZN>F{=UQDxSj|y2cmG$^aK%usY7hd?;cZX0?8U}`UpE<(VPlIDk zK4PC!a$P&6J}D zCu9wQU}CHv=f~zIp54;DxOHKTS;G0MtV}^d**dJfwz)hX2}vT_OPKG23wL!ewdMNK z1Y6wZ(6e!@hM5D77OJr*M5SkBVjab(+A97SsI{)v&{hncJ9OS~wmt zkiGWl(rqk9C-SE{U#B*F>+Pd3M655TH>BW|n?Y**FvdN48O;0;pDo;dIwMI*JfoCO zWhKZT61uZ1n&}~6g8#K9NkkW`5C$*&a9Va)m1L&_5<9Y=TRB9>@~}6BcTBIzd?lW# z$X#I{?<1>4Bz8(@Yod>!;*A?j3!$TUCeG=cP73IEB)lrou!44PH2a!t$RLrsp}6Fi zz1P%qf_THgqLmAtD)q6b90m>%4EZm9AKXh#K@ndI_GoUl=Wkj1P#hobIU&sE^ul-w zoTuvDqr;O}B=K`a*BDmK2%?K_kNM07brKY_jA zSRz_iA|q#k-E!LCRK4>P3c{@Fy1GJFEGdi+E762OS{+>y(pcG7l`GubgDm5+!*_^6 z!>n2G*n+D`+HX0MKlrg0q=6f?Mvsn{_@FS8r>xT6@hv*|0)^6~XJZ=?GhR+^#_{#_ zoh|3#e`;UDdp3q1$?B#oXC$JjKP-6e*+#=>e+}&r)S@?|nY4TX|M5OL{$UlaMk*f~ zanfk?FPvcCiKS{{@U=s(^EynAqjz2^4TuzXrFgU-KI>$#i~sF?@V*T49{1JG<+vWo z!b*RtMR+2N^z`@^9*w&E@d;U3PWJYQw0>9wrr;k*MDF$&|FdBkF>OC~t*MA5)vNQz zT%o^yZt91%Qd5;LFD>EFqZ@s0^6KmDbp+>vYDBPjx9`#bBaiSxJ_R9`BlkHxgWKH; zJxP$@b#%~FgJybGU~emAkO*tojD%>MyTsU8Vl1-|>XEd2%fd<7`^m^7q8Iv6hmRgX zY|*cEqYJKOm7K1z(darRc%Wx$I2=L580;DOtW<#om8vgtyrf|Naked_$3ge`r%DL! z%w~P9m}Wk$II0#(IdVPtyj{mTU`RGW>~;;C^4m~;$UOcRwbh4Ivzj?ONcV+!`vFhX z^_O^imhl1gQf-Z-opT}MGjZ1hyuQ|!@YO9YwS&`xeP15^IX3H_xA^?JIJ6$^Mw{T| zTWpB=Hki^C_F4?~ih(jKxQSsTYh&FaA#CR>wcoq83)`C|CwU4lZ-Qj+n#e)YGGyXB zNfIqLlRR{?@oHAyN=IT?Lj ztozA#f3S%iary36e?7d|8^)1~g&6@ZXk#(Y#w^0XIz;{}HX$7ti+FC%Jf@>z!I9hw zY+^}yV#sv}ZgxW|2yVgQ$mr-`*F~^r=$3+N@Z31~EuEF)U`aC8)Dcb?a_+jrvuFhf zXdyvCXye7N15W)Xyb4e~&GwN(rTt*_RzyL#8eXA)i)PGiaSYkMIEN(r! z^|tjKn52ccUZ%NGwsxZ)09;U2>#Ex#ep7vIErw2FbgDnYFewugkpdAeE^7o^t4WI4 z>gFafH{6uErsfg2(6K~gIB`1>ZSBX3XO<6L2myIu9y0i_(zhLTKJ!)B% z7(*dTDyf{re%X>7@J^s0qFlqkYkzBwWGbHhGRvzV@^|1^oV=0S;QCoX7>W?|VyXR& zRw9Ryf~b)s{HbO{S)?@U>YO%>(VHF;BSlZbi{2y3q@7clF7gM}bG3ohBuDY!6*r&| zPf+W2p7$LCDJdE&<^ro7D@_QmkHqX^jaIr4TIHA7iOTWl4X|heTt_k5X1W{10F{xj z1UV*G8li&jet6=Sc-9eoA|lS0Zj`sgO4lOPB!m^TtKUH=fgI1JjPbin6gZr6ZGi>M zTdqAVf-`>3M*e^#Lh%(LYj#=wuw^Y6Hz$0FnA=u;CW{HFo8PAx4>w!%W1W*7L`JW3 z>$>vnn!JqC3i4$-FZsswN)ECmJ@PagXEvuAsxs+Yt@sb+Y?@F(0VfAI+Pn+h))TJP zG2M6K+(S0x%C_3t#+lcGz*A0+b&FMR56P^2cXe##gGGRsyL+(KJs+PD4=QTvkfYK& z!%Ks?#ncL)Wp#t1$S<^$Vx`&Y21qn5x2O(TO;O8}O8h9+D${K(?;!V`Q|QmoW{~FV zz0YH%L#%>j)nPXWCM9!o@;X5zF*0wWh#XlM`-497fJ^B2{E5Eb>jnUCep z0nqi(`ha?m5%Ib##OzD;QOwqMVds?A3ud(gdOXpV`9~W@K_oAVllaxtC z47ZnRW7 z2i2HPY?*XVU%ZeKt2_P=q|au(YG>c*64d8$*m?pa#p->D5_^^I-rm)j*H_GPk{T?Z z5U(yz!pa*~we=8Zi_jzAt5t%F$5LMC*QS1D7@(=3);hxn+uy!b-4HWyrFAp} zn=lASoH#*&GwbW?J-c7uW%W(3DrfpaDecZ-NdXAga-4c@4&x4f*uYY5gYcF@k|`R! zObpoN97U0d*;wfA>#MD<&dX^{N=#H}>3afY6iKWFRBqpGI7;j>uVFoCMqIe1 z9B#AtDg8pDcS?$Q29g+_k&N&D8JL7J{FJj}d^}}vNuPMz?{K&D(IagoB}_?P3Ov^H zk78qA-}^rQCYB!9Du`PpS;Gsh&GwiFm=jydBr%1ga|$nJU+8Pho47D;`lf~u=7u4? zG45GfvUp)g91BRgAITm>!p{9p3RYblIIAH&)cmFeUkh3D?7_ORFeX{-)67gx1WR_r zs^l2O=4TD;y@X9)Ch6o7azR8HgOoQ{)7Lp51wt3CkN90asLc%0`%}E1OYm}>EiaXr za>KL|UK)&A$f^7sc%kv0e8Gy31HnLMbSl|kmYzaL=* z$W8jW`>qSYh7JXl`^P~&qO9K^sZZEJZode5m20Hp_I$^L=s5XK(NS4SD^Y3@W3VGjippKx@+1f(GYf=ExCEHmdcua~~FnML(m?4~z zfimn5B~uX2&67A{Av4iytgDAwxjfGF_eLGlHx>InE&459A_}U+my~mMhd(+=a}2B znZOUgRx&rAhG5}N6>Oaz`|u$$Qxg0Ub##cg<7=5^e5-3}@bK{1?Br*6`=K}#6&Gg~ zbxgh~dD|Wym9v;L*J#(BBQEbdNIJ3f-VYx>baaH|%r7ot%{?3Ghb8YkT0|(#AtW>e z+*kR{cpIS1pvuTk9|#X;S7YQ9SG}pW;}IZ>&OCgITiu2Pi)ye?+f9aYz}VE3=be;g zYakKh*{x%d4=`>=soj0r1W%;SiC5!@3b%#nlsEg3e^|!8t__$ zl>T>oYB~M}haCqQZtftC^5@1oiWhgPtvXmdR8+#}*~CN`kcFIv=){$%I*C(KGJROop{oSjHmD#0S9B+;3Dn(Q3QN)^@f2m zI1WU^7^||ku`&4mh-kJmid;-gtQuFS8c4^^aUsAl8iTc7-weB}-51pj!sI$SWE0cj zAPW^6LJw`>MzOmfBbK>jt6K778ssk2TEi^?0Z~e96nNaqU|RMH-1}CIcE5iIL*atr zdwsgKc()hNmJAEY&`Y2P{2YLvR-+%QD8hV*A}zSw=>9^JT6IEdZN7wzAhWl-J92H5 zBuXh6C1~DcZEs)V-}T=_VB^<5qABPX>w$qaGpQ%NV zWX}UEYh*^D4W-}m5GyY)?;bCZ@%c7=sqnNtb51$w=Pn?0LdMI_KY77LU|?XB(znX@jiJ`oZUaq!X0leRdk^qW zMMZ^tm*uu{3>D?*z$l=*_lB))ZSr?wAe&gFMi6(Jipa*%2mSOBJ{MBd^x$V@RMb^~ z&mA2d@(+b=`{Q3;$uh`8k%r25-sCnIZ5{Lx?eTzL=50Ct9KE^~>8^O3$+5G(ct5;HG!WoaxzKMG~=juZT|BX?`%{^NJ zY%OIHB`d3ZGqwV`%lcWCBuJ59p{_g`*aC-AZKJm-L~aS$mw{6o&&eeT;21C7dLvYa zi!HCrj$rT1;Swgf>?lnXYDo$K zpp(yd51n$eMzIVtq{!!BnBZRGtGofbd!$eUNOG|B1P|EDO?;Zp!M}eV7n6{0eST;W_mZ&g&^Ufb zQB6ZbuA~R2exZT*+0*1J_80W1`uQ;hu5KM(pU>VZ$NBR@2Buyi(JJxk-%nl_aK9U5 z%p_9~@!W$)d31anaP(cZNd2kHS&-qj)#mr8y1UAJuj;N=YhEeDEkSDPwB_Yxa0QI+ z)7xkQL(V@?+1=fJ%o;o#gZs})$%_|Pz>XQ1U%yf%Hj`&Jm4W0ZBqTI9H}Cnt%@Yj2 z9ql1p*#COQqv(s=GF`B|zC(|$uU8&;^Bq#gL`MVUTs|%!NDrK3qr960R;~Zmt<(4Y z`5ufxR;ovA@b^lE4gN8+XG)sb-&YS3Ea|UlIc`Mcvbz?=i?X9h?E*{;M zn@D7GWu=R)EvF>D{H1=#VmxOkD!;fG=YIDjF|&-Zi3t`KRzfXccp)Jn;~Hz+l+PbN zP~UXu`WXirppL(CZfY-<924+u+r&TVhBn$R)u`e z9Th7-EXj_~$={W(MiqT~dof)!MUEvhHr5=NeKI>*;f^=^ z$i_Sg5g{xK=&&&`FmziGEc#zYw%q)eJeHe~@EtA~T!B|tSKB#ROjA)P(gppV{p(WI zX=IX9-!b3DMQ;hXYPIE+LPmW5!ea?qgSu=8VP4*E5zPK>?(TYPIH?`Iy=C)^jI99i zT00iFTYsa5^uYCY6nLH*yn4cZ#16GdiW5)p80*aIr2bq*d+u&1`i!3;jm^qe4(|{P zk=b0ZwRi}4-0GPzfqKCk9znCl>a%w1za=ne5Z;F-up|O=7aC+JtXp#9nEDYCWv9TZ~Go>+mx1&PRdKAMtu%+n+ug6~;sV;T8u-MaceY)i7PTcIuF5dp7Z zZ1gNmF*;gBqQBOxXG@O5y?!3|rZ+yXo4Amk2kaGSfobK+Fml%4o_2*yecq|y>IpoHzIb4+#2tkdDgcEz z+&qV$ui)%aER??^_hshsJ!N_zW{gu?m39n7VUI(kYs>z8%p=At0>;&O=_xJZkfXGr zAdh1to#g)J%`_I#`c|fv>Q9+1(LH_T=HV1}X^VGkWDNC|i;E$IFDQpkmWSD2C`CSH zef%fwSn*5ha6rqlk~2o%B~t&Cl47X+@Q;7Gtsp zJd$eLY?<@>R2Ns4poRk(2~;xtGuGKX7!KUk&aWWh%&D8Kx=qC*7s!3W(ROvXWSqwf@D zaIv+pv<;75#Mi-VsFy!ACl2K+JJQ7m*#m#|_Xy(C2iq$C%?snaxGDp6nU12+^{XnY z+B%Ev#>z9oxYA@ZwfH#KSx!Ub_1B>*^e$6$!4tv_tS&IL1wv{EsGo?x?})OtI3;|@ zTM8{`Q%v3($~j0fx&DVy*&D-8RG|nGclp%4v~xi-z4cFuu8v3(fxQrR1uV;f9BFz^m4`I)(($K5`HJ=R z^?$RlhH-O&NT^FIIR(``f}j8j8_;q1L-7X@t^E{~3rSIM*8t3@|CCkj*mB)ppOai> zV^Fj&An4Crg$@!BsxLiP$Wm(ThuLcMGo9g~i*0@RI70F<{X+NPTM{~M3_Go^$#C|y z`~oFME&P1yfvyDaNGSvHaj0o=Cs_uH%dAjE1d!H62I}7xvOZ_h*_y`+A3FP!6tL?( zAkc%-$eA+vy4t^^NvoCa1ori{wTBNMW@l%&x3`y2GopitBuB=t>B$pzJ});n1TOCV zgYCkbtR?nJwQ^>P0XV(=rSM?XEf+4TFT~D3gL38x`W^=bCFS=6UwC&(mn_k3S+61oCSKDlY^te{3R7q%DS6`M-eJ6K7M{ckp3$pg;4v5IgWKtPn$nuK#HJDPyyV8 z`E~b#?Oy!;g^<#JYHlXW%lZC-Nf{ZYcK2;2+C&hed|O#ba8>i&UcklN0r5&fs>@V8 z;kIJa?>yUX(s{EqN!|e=?LYue!+pY%2jm#wSnLy>mPRdFWBRDEQ9@VXcp43w(qqHJ ztWP$eaT9_9zjFnC>3dr&3js&_&;EYZBPOchTR)qy!= zR8-hcpZ1>4cj4>k=BsA|rQvyCVa;OF(r`FZzFjz`jq<3pH5yx9-&82wP zZ~K7i@pMl$W|)gkzH+V#qMug^K`(f z&k}K{@mSI}ZUP^12@@_qVQL<_VFZc7U1xJT1)Tj`h}aU?@LsE94p8yAM*$%b?77El zYGMLe8Uru1JLAxq^c=n7^X!})fs{RDhv&_kY;|WfO}sl$7%$w<_`dul)p*Cn`LIi@ zyX)6s5o`x0a9z1rVy~#!0lwiq917712?>Xke!jj^($Xv33r~qtY_T)08ygSlgGxlM zo#b`LZkfSTxmh3;pQOSwrb?~6@!GVK0Z?e{>}a7#=*D^&tB4*EL1r1$4`3~yq`@F; zOHlCC;vx@^x{i(@FK(1$E%rT%v z&wwPyOtQqj;Oh&2EluM5Do%I?$rvF;=%c{|zjjw@p}PS)wBr=&;SeDRWF`K}+b+rf z$`6VYfE)vF+Ox`(DtC%+2ba079M*8htT! zb)wnq%MHwzX527KSx#hN$2ZW252+mh9?T6mc>%9;%pMQ4Yyevfth+!O1J>QE*lvfn zq+5Gt4*-|yai6tt@bH4$V`F36#=*57!splsz>!#H*_>Bs3@197A}g@JqOU6{vy1jm zU$IeG#{ZrY&>`L~O(UM7#MaZ(1M-`o)7$>iQap5}R@grU3G=U9A>})9XW{#+JBr=f zzm`FAXw3)L7WRj)Q(}Ker$?vI(C~(Yu9WK!nKmW`xylFOx6cB9r~HS$EBKlrRSAK$ zl$RtwKLPI#cc(^}^ivZB0r2kWL&11u9<)DUQM7-}p^(*6I47`mcu390@;wrgV-~c0 zmU06&W@p_tXQb7l9zJ|^=L>d7mL(oiL=WSYFZb}1u%YYz?Qj_k1es@M>tDZqjinP^ z{ZN%#U;hIl6U&{i{TYB|Z5+u7AhS6JR^KmlZK;{ZwXVWeh*yW&7qado{Nc~d&-h^q zziAiMM-1FQ?+??VQN|+0v9QR|F_To|85|W^7SyyNhR`GrSN)M$c<&q2~b56AvFh>4B)#jTYC+w0iVR;{TLYyA zG@zJ3{PkbzP(3~z%2=UCN^I*x&v)~$vNyNEisI>?F~7GuR;pus7F3Lk#D#^Evflwf zo&@jt!i+CpzOeGi2AoKxTnaexm6m=z9~URP3`!T^GCSwpQPfD0b_qc5bX00dcz`SAZRgR!Q)>Ff+_jbps>b!o}Y($dn! zMaF%h8*&6Mq_N7vSNGXzPx-&wsNP(2V{?2B1;}t;S~<<7gmln`H2dKz+J~Y6nGs3K zzP7&3&&LOJLrWoABC-Yb@0he~yXD^Z_R-O#Lwyuh9>=DZwjcavNAHXa-=fMnt)=}-E9rRC_MsI2X<_(^aI062BfER=m|FYUVwmM z=a$^fcX)G?qo}3(iL3X%h-D*yL)-X7m3xIf_+C!)-qRjfy8W%#`X;|K-YrP7d1k8( z%I6ephTMI8E^7imz*xjeTqp7m$tw;W(+O#01FaK*o^!SYXIWk@)@wkrP(f2%wnAf<9dJ>_WC63q8LaoiDO0cSUuaS)33#+KCVR26qqD;y; z_G{iZLs0)nk29B&Av^X-PXj4Tc%ERm1}P@?^0$0eEq3dt#m&Pn-=PEJWQ3=RO8hWwJrL0&dm#oIf zR>mOqkPND{-QC@qnwoBtKOHnG!5akEJXKV%Q_vHH3j(tXYnJjpPD~6j2l&jkq8Vq# z$Fl&V&_s33%*^)H#Ut|;!;eYlR6id-n8)~~w6jF=`ZZ#z47WZy0N(2AGE!bbV_XXh zT@&&R?)KX&g(ZSPujmT_KezO_d3h~Dydn&-yvNA+7&@KqVvzSh&pa?NFg!f$CRuI! zj!sy(cHr?9{X+T6w@HK7_xJaq+YQ0x*Wsa|YDWUyt&#Lg{*O#82EKf^Xh}V9PT~BJ z;?Zg9@>t?fg&3>&0`5mslIbg?@)%JoZJj^(6@2UP+WDaP&XIlq`VuVi-z9LcJAMyP z69fbWF6WAIW@fFSu9Ro~lVxilREddU8v)ruML{v7@kh!V16jACV*Mel@PZ5R1z5~f zKf00+I9zmd)gwyT5_WrX=WnSh#Zckl?4Hjw(rgKr3c}DCsuLal9G5PX2qLA%In}wg zV^ru4>cPtV=}A&#wJb)d|8KVOmG}4AX5+XbE!CX7yucu#z3D4Gbx-G=v%>V>kh3?l z+Ul?y@{^y`=+oTQ^AuLje$2uB^QRJKDQ(#x5`wdY_nPsI8#hE#ER=PJ*OrbST;=_u zC+oC>dDrEOrC8{72q8f0XlIH<)mh9!<@IT4Y0aBF zuk!p_C!V)+Pk+1Fa$32{N%1Z?+E%zhuj`|TY=eUw1r;ifRG>Ytluqonn#@B_AlYFv zUtq=o<_)oG`#WXZ@*fR<YH+3;%#A zCG(-Aw{trJB-$P2rI3mA_4QZc9zsn11-u6BCrtf3lu7rv^~LGIT+s{a>LTXgW%;M< z88E%koIJ!_buI2fcYkXxg0vF8VFb#3baXmV#{%H;PjQHue4%*3?z00z{wbtLi~Gqd z!6)lRa5E&~zfiyh0>)5*MFweP|8upK;;ma>pmz8TWS89&)C#WqlRaWmXOQ9I*>rww zV6%dP>-Dg^Ni7n5M2xMq`0d*iyhlKA*SLkIUwGKb%F6!ztjTsr{0x7#TXR$UNBr+L zIpe2c_y?ZXG3C=mQXmY^)3-vO{YKRd=|7-(?^ENAuW47dt3ISD`OtNI_nZ;Fs`!zU z?+nUKV8ybW5<^i;$!qYuw3Pcd5PnXFhtKh-EO$iGmL?&qKZ9EfcRKf+ZW0@Di~v}_y?>31#CJa2|Foh% z+^3OzKi(HJA;d%@YOw52c=O?3k^A$7B#knlOInb zPj+ort0Z;YmIF&_;_~}?lZPx&&K=~fMNbQ_BJ{GXqDtL8JlfVlN%or_4=*G^0VyFr zw6U;2K}A)xJ8lH8cWh+D@qKy5h=TEmn_>RG@55wz8-E50yuvH>c=WvA4&9vO&vPP0 z5CYr@{u{csiakbK@>db4(6TIbKB`=)`+6(DKo=8Fi(K zzTfoYFT)$nam*=_q!T-s2M1hmn^(uR#$f}AzNW24Tq^lDTyL=H$j!}t$J-k_NY;11 zy1H73m9-5dEdGy3@WMciYYx2v%_k_{?&w13@6k^(~MaEbc}A*D<{}KH<)ZNS-@OkdUoGJo_&KpHFd7dM1CG&Lz#^ zUg;}sHf+};L!aLG`s-(6mz9FwxcG(3v+8~1)X4P$s9xx8 z9E)H8H0tV@ZVi-t#5dQ}^3V@l=rs4tzOIF#RF-feHwBwz0O><%XOdkqis4bd7UYLj+bl)_9}u!6j)~TprL@ zM0riNQEtv$WS*4qKhHGr^kK;@?^_UskrYLmSTbv)#c>d-iaxl*N~EV~*9VB(w4_GM{N>M|r3{*}l_zn~cZ%5z1JXA#&|P&0aa zd%Lln`&+4S@l#O{P9_V?H==%aCDYn8%wF*gmkF*W{msMUgojE&2AhRX=mNme!fK&K+qEE1Zu=d~RC>${3Y*wy%h^ovj(rQvY|JfmF+ zZcHyFND91Lx%G$Vv^(OM1V+0qKA`iy!=hFZbfqXUgq?8k0cWYsj>mO;g}GE^P7Z ze-aH;4JFjK78e#udI&Utg>bO50~EOMJJ+bZUfzNY4k!GeJy~j0l@lAQb$W6P)l4!~ zdt$h7q%N0vtwPE5A6`>-+H4`$Yzui>tXULgYw`Hcp(3O%0gwBKYo+@$q;4 z{pamK&}8}NEG5Jp;LV|trq2!k`Y<}mn>sD}6E=^}SD=eCv$EpoMAe@$h>2vALZ44F z>%5@d{QzHNWcp!;0{!E)*OdrVXydSw1cE5C}kEtK-#qg#CqY^5>K3L{fIKpCF53@qc> zm+Tj_l3f*7!6T$x_>_G?;x2XogHrruEP71?Ze>LZLRJyX5m3B0x-a&^F*bO9Ab%)N z_vnhIWZ8lo$;A(%tW(U58PdCsa2+|aI1h423y9#+{1cM0+yF>w2}i!-6LWI32jLj3 zVdOKTO0!bOCqr9XapyTX4J)uj(P$(*Q^_RnIw%(u&+U98DcCr7rnSUY{snIdB)Iyo zY`ESLTxtN7WN|@ZVNfj4$(n`m{sW&wBW3({G9h7=1SnM3C>L(`li!}<>WpA#Rb%}7 z>PJW6Fa(>O6}qhfVbk?j_bx7{&Afhfb|mzwwl=37@ErM|+M!P6Z4VDO3Weiu1P7*9 zx1>sl6Mhl|P{9KsqSsWOi5-Ew_Sw3zscHMbcQw2X5Gj-aN=i!cwUd*RfClb*djs+a z?+XGdFg+s!7>^7ypOp~~xI%I91`xhZ?{~v#ISTRK@bIUqe5o6sILnJ!mN=!_dW;Dk zlZKL4cOpUK5EK+NMmU>i<47t&8@%u@%PBI_YKP=_J}WbHi_$Za*KD0 zSH@IXkXyXuY0l)+Zf0wCH}O13T-HnET0tR@zTbnn=3Xr;yjnhzymn+V{6ZLYu_Jl+ zHk^Lg3gbq!L!U9Xf52J2iJj)ish=DGL%XUo3nEQkIpC74Yn6(!4xQq{(i3<~?xQ zE3k%3Iff^rmPewOKW4hQDtbDpf-oB>j2y;D!7F4pa>-|0lzwW5>QJ{g@Xl>29#L>m zAZQ$z;TSR+2$EP3PJgAH@{Xi`WtjhQ#Pv*V$x?^-tHvy#`m5m#9%K8f!HGA;{KFYT zjQcAf{kJdUC<1nn5P`_M@en%jPdJ777=bK({P;0cxNu<6*qCMyr3hzU%Z;C%$q&NA z!$F${M6Ho2LIZ?@2o$zhItU>DW*C~LqPwR%4ed$4`_x@3rob{@H_opUa zcwA7Ekd|gFT6&jd1j0#7d6FL;5AB`?{_{}!NI#1QK0dyWx~VCh8i;c`M&e|tZJ`L% zFgP}ggjoT|P!KZ$tqUzncdFkf(O-7OQ1@NE_g`%!FMT+D0Mf@~(^bIW5XV-_WzMD* zgukN4Qaa2cFfuX%sl{oNJ)zq}UJ{2au5ka(*?w}_Lo^1qrxNQZxlJ22!)FiR!= zGiX`+SRO48Kgr8ufsfz1b<54m-rinb9<2_^mcVDxoZWQ#a!GUXpx@}7Q53rNOA?Gg zVkgbvRVUc1#Z^_Z-fI)E)6btjf3!QQ1H@Qn#o2LT02^=&`Ym)OZH=Od0CE5aP2I<4SZg`h`0{ zZ5<|+&jzl%ugJoFx(j4fpI;WtD`3SUa&U(RWCP$MP?iqet^f*CdGH)6yyG@Bi0I16 zqI9`#Lu!O3YNojBja$)%`n-{3b9Bzo>k>61->YML%X9?IA7pPlt0D`kawD^wQEZ`G zWLqzj@a=s8m$3j86&)VTVlW|5gPb%-Re0UCMbKrfso#z9BQR}2FzscAk=BKwog4z zPy8gDG_%AAD}O=f6J7!ubpce$!5~#@G$^Ej8Ev$b^*ekHvltkDxUsfa_mDj^E}t}o_!&Slbw-|od=2D& zaFB`G3WW2bCcKugNOhk<*wUE1ll_4PQ~tU0l=m=7d4W%Z0vk^sS{Qj0Xp!G$TFG)01A4UEG1-eIKFd_w{#>k60TG+O$l!PC)f#P2OSoiem4p)U9 zaDXb07_Ms9Dh=l`B2xR`y@SK+bDz*YE0K5+{(QhSZ3NmNapPdA?UfEcCjyDUHf(r8 z=j|rl;DV|fSoQe1xmoSt>FpydBi_DwqsI-rQR;&cH9`#ew8xJbkr=_*kYxtaFDmNj zY(krau3WD||GC=v?Vmp*k_62X$Bg}w4IQBCZrKj{(8OB4Yr94kepuYsu3ba+!lq)S z37jwpsSAFMdCd>nT>i&PhC}=eDE3txc`=41QAKTSqU^dh zVXR9KMC6AT(pTx4UnO=-PcwQ;_=+I$Iw|9~ff;rzQ->$yD`UY$p~nsC2N-JdtZ0?p z;D>YYX2Ki6a|Ya|YzYMPVg!#GGMS~gjW{GKS!*ljH zddo|K)ocZyK*%3bcewF#6I#KiKi04FEdJVTi6DOqu-hq5rmL%KW@ZMCI{R+TqNQ|) ze|crArmjAcn!iWtPIppVSm;oF6J$SjSt$4wQTrhlpI(px<*9=VG`Fa%;JI)uM5tXU zh*q{egJI;Hnt|UQK>&PjZIf)s4f8E7k1Z1Y`t1KpOTP{sMv%NwP!1uG{k10S!!)50 z4~o~qCSDA?eh>|`+E5?+ap-x&v%<)x6->b*8Uum$+4=I?r6&-y+b+rl5DJ)KD3m;D z<7q236NnB^e>*YJ$oWM;5=qz;5*L>pHb~u09827UVGAZRmPK&oZQX$ z0|b6FTGTbx+yVyG-MaE_7o`l)d!&hP_k+eagqu58n6n;S){Meaq7~czV_h#Pf}W@jtnT z20I3cPk;s=A9L(azX;uRw@LrG_Asc+R#sO>a(D6)Oxq%WW%~>qH&DV5a6`kw+HOPZ z((6M+Iqq_#|3o}at!hRWT$1p#+;6)!ie$mx7oNIrCeQ&Usj z7RO;qg@j*gJ}MzI-e}P?@8$yA_8Vh~n{!#v&oD|7e^XXQX%=YDdrn})GG)Qd%`FeU zwB8VoRqLT2v|0gYb2O?W`z+rO*u;~Q6A+O>QaytqpzSO{!yd=U7N;bpUlpCClNum} zRQ!McC-fNLmhR8m28U5wh%dQjV*q(TM|Sh(O;qWHrxR0CANu{c#^pha+U!g#PaR%MdVMSyw&7R_cn z4-#?##$ISa*W{xLnFAF(Z+gq+8PPX3%C)Q0&WK2i@d+1Z+c!aO{9+cb1^-f%7yYFHSCU^xRy?Ru^~>cd6p#~CXxe#9x|%&fKE}^R#8cGH zlewYO5cm7SF(05Z6~4kJZ}2=&-iD)Amfr25xGXBw@$;)^+#-Rm3~PC>^e3Oh-Xc z;Q67A8^aqCZCnXO$%~vErh|{8pwf=71!4#a>{(3A@sNHNy-OF?7$j(4G&OBZH4uBZ zgJM!aL4oBMx~^_HtJ5#VvsGDiY;B8>U+o%&f`m=Y%sxRGXZ69v)Kn(mB%7~yYKq>L zk&!XKzFK&5`eg%1C?)jyuGbTIbC$pK*>j9-xkxAVk>p{vpP97a+U=&>NfhcdI7u6XmM3(Lypny$jIhYQs1bOTsGqpC9x6ip0& zKUm7WA}M(-jgln0BkJ4$`vM;q9j5kU&vVt%NDdRIa|zE&$! zB&c1PIcORy4Oe#z)Q@lgL$1uTVKhO17aofx{|nWkkkq-Rz)Y866qcU@2-A{ffkuS; zZ>>*u!2)!I5k%`i3=O5m%i$|f0n6*En<|f+k_dGl7bBi1Ol$-T9w* zQBbYOX(6T8nFqAunNK=)VQsUPfB;=WYv@5r>0N`f`ghC92b72EeaTL1FcOGm2lD6g z{cr->`Q>Fbm{LY&yXnf5TmcWt-ndW? z^4bSE#IzOV;mbZaUW=4hHaZ`j`arDNjKZ8!S;OX5^;|ZV9j!!HD`pB&ag=NG!Pm2* zx)(j@@Pg8?yK7y13<^MO3YCGbuG>O4X^Alm$}t#cmM3|?4JQucu|os2wY3X0i2b?Z3&sildLfBWeiW`6us|EX><|SE#^TI(I8#^* z$2|3q=Sau^e|8kB43u3H>4C8QZTx95Cq^*KZQIsBG??g=*n7zPoOcWTjvJOlTYoq9j(6)z4}eN7S$o570@(u((7Z{5Q1j)i$?UWzZ_O^LaDl5k&vi9l92 zZ;k+_IqO?;aw?F+no$urgwa41<>jphpxUy==>niRSA(Uk*m8PuC7X867mxPjV2Sxs z3kn5cP{MA2VIl*H&>84xLr5>pRfQIUc_0&P22frV*0enzTdeh&X$c@!&wrx~AwI|y zGa9?N^gaeR7oEV)N_1*3y2BQMF@7j56O*=Sn%?vIn4<-;XE5hYr~ncQ*LTo!#|OYT zld+L3ao59x19@TA^iW(~<~jrg|4+KRFK&dnQ}U-Mzt~NpPzGHZv_M;Ds?#hj{PxC; zcBvnIe*FMK+zAK{0d}z52YI~pWr@>1!Itx`xy9Nr%@t4^{X%Fhu>^xaK#9BDaC~m_ zGz=kuFdgz{Zsg7IFsUbDZd&Fc5Jf;Y{QvxCi~~|O6&cz5#s<9G^1{MBU?@b-RL^H1 z50W>)4&y~o$B7#1K>b3Ln4D}4%+~*6?ajlfeAhN$B9wVbt;`~`h?2~6QkhATu}BIL zm1LgEkSV0FGK(mLN|_TCGNoi5n~*6DzH?Rk-TSxS_xQelzTY4F*vHCR>silp-`9Oz z=XIXv_2tW8p}7ztpPlIoH1&z}z-s)GB}9k6+=ewJ%6Q4Y&O zEQft*R7?yC3>59_sKtxY zvwPh|$+(Y&Rnx_chT=QRfmX{^bw&FDF7(~Pswbr*72e6_-(h_V_M+9m`&6dPtg1-3 zvS?EU@wKf>zdaJxTKJH=u>OERmHSbB#;nWlCA&PMm|hr6GI~X4ES$Yl&vTmVG@-Fw zEjz%|{?^dI!x=95s1jC5xMq>DA6@oImFUP^-)Z}GN9+W@m3&tfnP~OF`Ad1RMzT|c ztfFZGpB%jo!=eHs>E+Kiv!o?rxi|S!%*s>$eYJT0>p8xk-t`Gv+upSVc5SLWo7d}V z(BZ{?O?Sr5boHU_gLA=@7xz|D?qysA0_bw7p{HjTFE6ivKv3;GLWi>VXDce&9}YO| zR6M5)@)*lJTjB&g_Jhr|7letkXTKuLL~G>-Qu~7EporN@He8tsSh%JkubZk-^3{M^ zE9g21N-1%19jHokT(5=MF2^1is%9Mj6tx!p^JtnAF(tVO|-cPRT!H& zlNum%7;{^gn>Tg0X7zbT#~et#AcVtdWp%z)%G8yX{pf?~4@U^PZ@*GrRiS3ImH4RG zrPcw7V9s%fXf~=%fc`suzQIn~Wl%wsi3SRpjtsRRgy-hQ#(_$A3eLT7w=*_=jq4XX znhmVSBeLe)Cqh~3g<+bs5}RFod}`A(c~jLdMs*xNI~;z@Nj>Dg{v7v=MXv_a^G%9&)jsslW*<`-f5luia?dWFgxqEV+K7SSpRI?3aQd4xFl1HVXK~C z$Kr5_ci}iyhgLy`&b2op>uK7M?J#5R2U*ZFx8a@YAC$SE>qlrbXEq=2>< z7sukE1r8SQAh+*WF976d-NjQ5n#=k?l;a0>euyF>0BF~8DK!vb^a(rc1&g@y=#e8g zEGfGwc0NP`{^mZl8KvP7OG~3K8v~qpo=b^~C%7bNF70Ft=OVf^T03fKUH>9$QYF5;zx$W;0NqT$u0-g{ec6r7dLXPkGDq9fpRP79i%hM9Y3J*+)ej#At!nZxZwygZRwlUv zlElgsUqmF*jb!sS6Hm{+2YUpK2sKBtsHZuv()MU%TUC2*IH^vcD1h0Kt^_~LI+-v1 z<+M3`1=NpYZ;nA0K}AIc(2FS@9AhRkKK94Rh-~XaTTJtdMXUeO8Y-N*eyp+*{&=nf z)r%;xhTqL#4WD>`ev;UPgU0KV+%!y_&c_OG-P)|%dZ6c65#talD($6xALNP|Hw z<=K-R{%^2(H*SJ(gF85*%=dy0wdOQ1F%n?Uz83TNtG~;LFOLsRaJK)(b;Lj?H&fLQ zOVoh|-XF;B+;_Se-fwAsB|C7axJTad0dYO`=OTGB9d?}}OC49oY8^vn*`TpC+2v)A zlx=Q&*GIqkt*lgC{POtWa#YLmEyeR+-otws5Qdtqnc1d0b#GdZspST^}k)qfD20`fDHi%NT zl;|D!XuMBfe{9ajAk0ESQu5r{vq4f^Gjb{Ia`7c6u6L3>x*g_D#@%W=apdSg0AlL^|6oS9h2SwJB! zJ|1lworL(v8sxlScEBZ}CyXXr16Rj^$TR^}jYhAUcR`r^b4Jv(}Oh!P6bX8IFCH>yVTbO zzF#bbAw&05s*+h0LC|KjXQ0r6xiIhX+wZ+Vv(U4%k~OhFQzCjIK>Hbaw~dFJh;+kg ze6r>R&YJ$^550_vgE$>fD#z`G&@sO)n%-65L5<%}LbdigGUZc!JI+2Lf5%H|&(yE6 zJ<{GjPh#uD(kkW*tM7; zvCgCVE3fDwR|JhciL$k$l-8HL>f*AbG&J$n-8qh3@ej^qe@;-npvpt~(3`cWs zk_!YjcO_IA3-M z9$e=s=$H$98-JHSoRj`Hw;x)yF+#b|!pw*!PY7 zi-Yaf6|)o6`etR)HtxN2*_RJ#@gBV{ko6^hhDP8iK3i4G{P3@uT+i*TX+DqQLrjN` z+~mq27Rne4iL7&WiJqMk;Ph|io-KOGV9;TjQm5`rug9yjN^4L5G${Pi4`IXW6~s?R zq^FPW!?tv1`kGrn&wbpz_j&Wi00cRe;ts_M0hL1vLPnZOjf*O7E9frMY`S>0@u(ON zXb7cIb?Hn~dd|(r;0cQ*7` za<4lUq@lRXtQPB0XBSgi=`v1LH_^@zX%!rkXkhYQe)-dp=|`0AUJ@Qc=z!E-l?_bb z`TeDZG`(vE1pf!V`Lz_Y@1Ir3h|NWut!374} zqh|*%FCN%;rSaaPOT(gU%bOGdHgRUw-50D^4or(eCB!hR@vp03ADa`%r`O6f7E0ur zN=kZOaVBwALiSk|K}X)ZqB2i+*f!s!C)~((T;`E6mD=+SSCpDYicL9VMNRK7|LRmw zVDP=@_N3v}(JaXm+&c1St99*}l($>E|EN$VUDe;8$_@)5?HA@wt(()8MVtpsg3=`9 z>+UvRxuw{VS_=x7(TBmRD2goREX9&Od-2Z%iBkv62>#dIEuE(t8)>uoKFK{xsoW_o z_2A=zhuhZ1&(g1V#A?URKmGla%Ww?z%b8h}B~GrX=&d#_cD_>X5T6YAsd;&zdpGy{ zGV5y}W`}6z_H=`syLqG(3l}Nq{P_;}2q}ox_aFq}yh17ndpLCA;$E%g*2NY}rwqOn zqkpV>m6egb^fTxJvc<>Oewut$mdR07Hs{YY2l8gpe?Z^j%PMC zut@+OB6Vw_l=FkP3)3QUr7jnlOqaIFuQjPR^?!Lei&f7a*&{HN$s8G^!_Z(w9k`i} zRZ)nE=FXiv;Z=5y^HsFArf=VcI6FIANiKsX;Lzedc&T}Gu4~fTR=qd*(7_6mpKk~a z#~wBj$>(Lm9mu=-lrkpd-{)u93o<{LG#M7I5 zN&ri#lpS48#J#|GbEQHJb-_J$Xt?_-5?hW7# zt)+}PeB=mxVh#4AC4OGdmL9$GrOvRDe{MHm*ync1uNwuLmuWVK$8^!}2N@I7Cz zzj86RQ!4J;d8{QS3S1GLk*sD1lez{k;LILWB9e#iHFR{i<7}9b%XD$)!IRvr_nKMB z=~EF+)})|KF@@T5GGDFgD;pw_&Ae9WD8=~_p4KwBDF7d!)`F_aeUm>s<8!Rtct)}U`(kiu_5Vpsm!xS2&G+d=te^I3=B#mTZpwf^x3cd`XU2#ej>DI z?9`Yp*#b7ReNAOvp`!vIS>_LksFH$bhD% zr=8E7VJyA{GbxBtyJA{mi}_@}X-Y68ST(;N@YyLJIJN%w4o?y2xXoS_T{A#>co%Aa zZ?uVLjkJ0cS?Max0bDdjyTq3^UwL>uS~pu}=8>iRkxhY96o1|-()FgYU$zd*I)VQ0 zE9a~5#xsL;XB}dBtAEV6n^VEK_jDpFaZ6t#ahJ8~4kk5MBk_(q@Zya`&|zEAG75>h3Qa z=oHe(ZSUUO#x-*);%L7emP6T};5+g1(-FPhL6hU8*JS^Eym-a|fiwwPwdbLH++xDQ zG>nYfd9+j0)9)4?_w{`gLO_FIMs7=|oTh*nWh8m0vKi?|JB4oG784N}{`w6L$<$*{ z*WFRiYA_zyL3$Ew}z`^23K0(S$7Drl8Q@-Cf?s1q&@$K%mqsb{E6a zogz#7sM+G>UwUAGpVF2SZzwNEr-crRws;yb#d)*h|Raa~UbM@N$Tjzy{Zk@ZuS@2H`$ znsydb7ciIF)VH@S;Tdgtux%g^F_AoRQMFX%27s2 zOeSy58b2WDGzJz_w(a&VdX52uv@6S^#BO>zxI z6x>YA%xdp3Dnw}a_{4-q)jKeePyPgTtf%%g60 zrPE18x zoKiAO$soSCviE^|g*c^Q$hnyF=g*VGw(sqe7){FFboA&j>%O`fRmDJ=IcN`}mx|X6 z+v8?&o;;`f_Tw_7<)4r-V2?(?uj$BQH?JV&nekmqZdA8KJ(IyY2u;3R>1IyTOem_qQKVKdEeiE{y0#ynfH@F z?vp8ZzY+|EIqt5%aL7a`&*`kDa6)&+n8LX)4L549iT;Tir8gq%5rUMDA_SQ@U!5$4 zXAD{6u=$L5z(XjDF zlBp5@_fNk0HW5C$m3hp9yxZ6Mce9T>*t@ zf%AX&?icbsC1iyLc}2hA<5W9_;}z<|= z&h;@9lNL2+4w|hMHZkX%ZXV=k&e3?-%zH=4b1GJE7#V)1)mzDR0c3TM(Vw!D+rFLJ zto+4M;Y2;apbUK8DH3X0H{>$8*m8cJi#*9Ik@)=Aw}+lbBFAn-5ZAs%`dpAdJUW=H z+Yz>sRUl+NK>v zxW-qiudH$kEhMkDo==}YbF;H+Nkl|PpY!tSw%EhRN6vN$Em@EqW_=X%mb%T4VWl2? zEF{{0$1p{r^^NNG-9t1gU&JgQsrm{TuM6X6q#a(qm)@}Hd14jku#{NCiY`-d0DcNA zL0cOe*N<)4ymH1JLBsqEW$$h6ziF2yiJH}R_0)#&ZMj#wN zJ`pFlA6j~-61rT7#1}V*gBG5P#xsP`oL^!bxg92m?O^+M@NUw)gh@>M-LR?g{avzk z;#Y@D(f0@MrX0;$IEeka2|x3Msk}3>M66m`L)pdixuk&$aR^BTUBShTmv}*Cee_sO zzbT)Uh3jC&I8D&o@Z-J${fVM2t%(Ezg@%)oUCE4KhNOV z1=R^&hZ(6j6#YS!(1m>ZopogKCnWvOl)C~2PJIwa-JxA}{`Sg}#D!a}iI22S6$)LZ zX%w7t4&H)54o)fXtgbklJp?gKeYs66WWXQ`^EL6%C#~1@Ley`kjif)>f8=6e=CA}}yFJEqN4@Ct`C&kwpPjUjEPF7^xsoZ+goSR-0-CFT zl%;AO>gwytl^gnOym0OFQ0UHQ7qH1IfxBA5B~pRyx- zPFNyQB3Zz!R5nK7T_pqcK>;EEb1~>hpTX*FkfD6a1(zQ%J$-z9##Iv8*rO6|&^}4L zP#~y8waD?gv{%Y9||&@+!r$@!G7}rt(vC|PPpycAwF>j2JFy+GJrHK z==UL~?X+Fx5Mk&p+d=bDnz{Fz8hrkok#_10c8A|B>rv#+DZBsAiyloeMh%_y%FP=Z zneElCH&>L9K~`ap+8pzGJNv796_YfBPj-6+F(s*AZyp&PC3OsF z-QC^iEHlP6SK$wJHB6KL_RDUsDKexUxBonE8rE<^{xdH+p5Q8m2I2P4(${nsz*J0B z7Y^|FbJH90()5kT6bjA;Yp1%C<&WUz0XnxslSniKasMaUJJY|9XXe}dn_5pAd2s14 z*cMY{^7Jd20UKM~QviZ2iGFW?_n$TD5#joA55&h}YkK^K8>IqlRnbymC~in^J#KG* z7K!HX>u`MtCp|by9y>Nj;bB8=mT7S0Qu{pEfcJzSz4rV{CQ6Xr<=o6x#Xuch(iwlA z=DV~u+v8JjWbo#>+1d6ATK88z=g#daP5qsaPDm!Su>Y)QR7rTlVj^pI>S>Fda8cl3?B(Dhv_W-D|E4VGSuUE#IPwvgna0mewUf8e5uyJVxJ4+UcX9dfujn1q{Mt+f zgU;mS7EyVDa0m20I5j0DWv1JpLsyXOV2<)H$xoZ{j1^G4(V5C7!l)bF7qpNI&jkL7 z!Z&YK@P`wQF&K_1G(|t%0RtrxWT$Pk0CX6XkmQ5};9>jUTt1|E*}UMA5Fdn|K`dON zs*a%jFM0@BU5YJudQBtJO6Ned!Bxgs?J3-Zec7GPuXEn?^)U$~ky?5ofOe6`wO8PO zbYVJQzRe=~)(MzE6|8s{*%D$_p=xzD&qmBXzuOvph7 z0R~exHd{EP^0NyJvcFMI zX3J6)e=oBV0qa2t+l@5A!tEjkx~O-cllUFj@sXpO$UNR)wda$?J`$B|ZHw0PpF z$a0ZfsIpjO1q2Wa|s!}_WakJ&`loO`9UA9kogWvpUuV;-pQ+f z@^;}1ui1P?GpvPlE&BT++Ct+#$X}#7%cZN8IbaV4rD?iX<2HCoJwIWiqE$RvD2yk@ zjwIgoU1bOCr|?S_xGOk+|H1M>t>W7Z6AG~#k@JD>Y=&P7Y#GQ;-8=^m$e&runvU?1 z5>1a|Ns$xmb^pbwK*(ppqiJm|3=qN^_#2GkcVxuy_s+f0Bz;%i=KIA?X&=kyafL#* zlMMI{Nqk2nLh3e_eegpD#jVmxGNu{y2QhK+tnp{#RP0g7!6SBCV&Xnj&Tds8h^Tm; zKLr9M)DFlZAb3C_4IU(?v>l~{jwEc78%%2cH~! zd;uQS-f)6aU0pW86&`A`qBc6{WU_!R$>+;tBPWE(2jAyrxE}jcxhuaBAm!ApX`2ht zL_zR^>wTV6aKNekh!%<{TM{@3>@Q5UA8Wp?`HyUQ;B{+GRu(jrxI-kXsQHPg1S~YG zs;jskitXd2+mzZX?qUp1L;de)xTB$O{u5mSr3ylrXbNx)Rni0ZhGEuH+Iu=)fQ}+-}ss?`-xT|tR+6|kox}v zC~Gyzfy1t!pC6q0&%#|k76}8=FfFQp^6l3za|Er*n`fevl zu$6yYE$qQH_-aS3Vr%%}dZrk5Ha3zzb+>;|Qp*8gX;zZE!GIWz&T#VZ4BJRIO)zTQ z*PrpsYgPhxXfxGsz>4p9!;>}9afQ=X&G=%n-}QTFGKZ3amiPv<@J+SUe0jmYu$Vgy|Lu*fVckp*V;Oi+%DPa2@c>#46z!CWR zqos_9PPu+Pq@0@e2SPpR)$$bmvJyU2t})`Kkf0(2UteXoeJ}fSTKJm^cUgP`V+5RM zV4W|YJ@bE_erYKsB%~2Gj@=~ioQ8PNW(Qz{JkChH(pxlEM1zORa^VAW_?t%*RPpr;3HO}x`qZx z%P)w!p#M0Tc`o#E$i`>J0on!TeVq+3DyDd3IcKkYK{$$Ag217@&yZnyA)_r`Zw*^s-~}{ zR%29L-_GT*x;jtflLZksjWo4KFv&{^Q)+ehog`g2R96w(71T}p~ z5lFH#VI-tflbT5}z92Zm^6|Q7-};zan(1Yqy@?Y@TPmOWc9tRzv-$Km+L-s^J9q0O z&lhH~=}by9hWgz3-x(5s`-T}1{;@nk$HH_AP>o5H10V%Xis+bz^7dOiQoZ(Js!Lri zR_UU)`eVjxtGoRYrqtB->}+B3Sew1sr~TSir9iO%fewEon_RrjZ+twMYivSzTtWAP z^<$&{+xL&}(P3cgn74>I@hZz$D55g;AV5OT5&6hekw1Z;j_@=rYwPu&+iETq%07B< ze9!V1k=3t4FH*t-eWxp>84VJ!FOW>egn+)UESw=`_G965nnu`z>8O}dddGW$^VXEU#m@Q$X zxe@eMSP-_lyS#2&(6TB{{1f~;F$iD<-8Yhc_jhxI(pHd>N7u!W?NcptjZXk>4Gx2U zv>2^Z_UdhmLouiEnt?$^j`hSRNNg(8*=X*=k&L;+4*k; z_j6pyJ+z6Qfz|7@;7MzZuG{yWy7i+OA@Dorn=JjPT8MGU zM*Qe0+SJ?}*{m6PfhyvMl4e)cQO_UcF5;=_!Zl}agm~Y)|C`qkm`oePEqWl+G)zGM zJ^wHF8lQQ}4u56hK?%)UoS7kh+&z=8n)T3u>HzB@Tj-vgRpMzYseuwR-_PPo5`P1V zv_Lf1wRT6FpKI^r>+-Jg%a@nxIM$o{e_u;DJt&$YvyW{^>fWz*|5AaeEhnG-?%dMY zS+}idn=lXdOwg#D=3V{B+Oc9tY>=O8@){Y{5#;a_yLp}0EM-aS=c`)+oL$Gt>~91& zRZ-niyvF@bDn-utosxlxM(g<_JsMiq$*(*N<(&x8-R3fAa(%z+p1P)8b+JoC+w?0n zH$*M@n7+Ihsq)Dz<2@!S`C)ex_4>|ZX5kOdiPm?67Zvl8DEB0sM@6QuJ7f9vc0t<7 z$odNrGt-wGGsh2K{)5c=_{e{QWy%jTc2Vu z;7K#Y`-=k}cavshT5G5CPU?=fY2yCF*-V9N=kI&IzA%T|+Veg%+i@t8smoIK(mi3X zhue1R=zV!)^N*m3S;#h~&F?}eF_(c+yMrch3I-=P_a0IBz05FSO?9;6wpPZ% zk;_Sh#UxH4AMwAy8Vdx17pdG*_J}DcDGhWTXs6rNW>8`qYF7}d=0K}9pXO7#MDs29 z^!DB591J|$EUWa>yh;~-et+(==YPf2$Jf7!Dq#5BzdxleICOk+pE*qf-zQTmuELSA z3!JtVQXO?wBd@nlbJE)h|97W&`aLcCfxUe@G-^Y)Y`jfa`tqdnSp2W8S$yeU(iK!X zeM$zhE;%bhegNk6tMbz}aBFDoq&$_JX;V)Zl0wfK;qW@Qx#Rmzp1;jwZ9PvYYm)aa z_*AUJoOMj~G1?zEKJ~s(A7ifqdD(UTv!hTx67(5@-PlNDE31#l$tEWs9E*PPLs#5`jfjDk~=S|A8)(EqL$ zQpBRW^RAhLrM(M@ZdrXinus32i_y89gEu$w z%#@Xtq3{U+B|!KMe#6jFW)>C?SO+^hS-+c;_Hn)Jca2!E{VF;6@KLqBr$llzrReRlaoA8aQp^5oU z1N3{*rvy(Fgi-V|F2K8+UChj+KSoc1u!NsVGB`WgeG*TONx8A9Nj>CC&G%v!%Wmk@ z!o$M>bR$^-Xi$53la=PPA|c<&tr75B;P}N+ac1ykny|v|-5mO`1N1Q7t;t?#{kE^q zfX8zEm#Bj`6;ev}Jq5$wR9P9Vu^<*73f4jrx^7z711*N@d?;px23nG9(toK2B$x8Y zmeQ`Y61N7@U%-B9j-r-2^}Nc#cTd5tEdNK7f#I3H-|QYpiiP&0dqm0XB`B*sezeT7 zb(7PEIf)y5UwkhBnxv!LRq~WmOJgT*95dk}6-|B0F4xqcAc}HIB5h^PjRGag86KV$ z(lb#yc@wPnLWlmXK!D$2+0;Mqnh++`xTv-SQ8S7@kv8et zS^;X4aOOllXl&HA*5eO8VG+zq>K#0oBt3ESkN`=^;FDO%CZp5w$CYlMo?hUYTJXPp zmDk|fkF@L8DAd`(F3sqH4;^_AG!8Te+Cf!gW8(vFMHgcjt>lLK)r#(8WJ;c~v&ec^ zdSmjMs``hgLF7t&Bbt%~Kd`5mgG%!>`Suhf26g6(umD(DU9CGoUQ>rZdN@BxFgeo) z;VIuE9!%*)rH@YZkP5Rh0mAFzu~2f1P!6p4300*`)1$A`>f2bbrgczr@|&gHoEI^^dZca%$FQ@2KM5Tc7YT#h5OgB$5Oli%RGe3pu6bty{O+ zH;12ujXz1z{?+FZXfml4rW1Ki5t^MaqpD=8=ZJrPDy|~*C*+kGwdhu|7Pc9BB7$US zLa!kRAB(U)vUTRKchOmk(R?H^?YzHOSDr&1T)SJ3=CIg(Q_{?{foi#6{V~Qs(q*ap z`6*l7RFnFR)&&wzv?Y_b3f;K(lLFBv*qq(^@;yp*8sYQvV-XvUtOpi`^k3T1uM+zR z&-ZQcfjH;Ym8aBU8;&iPiCBszGHj8deD9Y-(9Q4M;ghSjt-_{u-TZwrjuEvWs@i*% zZB=AZZ+4IcQwVm^ZV_hIgQC!>KNK6~MN7*xNXZT@prAAMMgy=nI~b4E%fZPB{(1SZ`&Cr{hw_TYL_|D(e=|&@uBgME3f%MLg;j zmX&JifT#^OG|9|e6NL$ZoEI;C*NZMUY2P))@JlShbmd8?P0)s*&;Brufo(CMToVub zygb>D7=a|3z8(OwPirl$g|l>KQ)tWB6N7r{vk;j5Zf*`soS2u=%uo&|4qu@<%)-iQ zyhmjH%BEd5Psx0kyPkwdkKH`8Ejj*5QVX?LWo6|Z>&FTb5>Ye-Xo}j+Xt%(jVZ8(f z{q3csYcJWrD)dy;cXX%iMY0iwROsbvWu$q@-`69w2VUt1xJyzBv$0X0>3Xi;?CjEq z5p)@pzci$bD#7Ax*_xw1K1uYF{Rjo@9kU{WudNCe$b&&Y7ckN;)dl4fsMg#e20%3k ziI5WJPJ<0`M_W_VL_I#d$^TB?7ygk?uJhhChU9N!Z|~JVXlU30qAFi9z$>)q!#Lh` zNvvKTT6+xPB)Frqf+QCaH|EUq4E7qlHCK%TDwuDCzqJ|uxgq^CWT|HJt|k41oVf_; z4Bq(N`s>e}IRkXIWh>QShpJ8n*NV8TbEg9=7wMcW5U_Cm;KVC1zxy6sNRW^o-%U8b zad=(9FUeFm6_j?DeewN&_wgY8t1SV>M|@vgVZQMV#!JsHfd9O6PUX&-0kSqzU17V5%5j4 z--MADZXPYjU=W7Qh0>qB>hG_9_c|akDJgIm74pKJGgr?y7u+Lmk&oGm%!cyO492Fw zrp3bgp2I%y0KIxIqbCj-v!bG6c(a^1jLU=%>FE)b559&I^QB`YyHBk`A_m3?D!oNG zfT-Exvd~V%Ick6A%pg&V)`?=#dDx~1N+?&1K^Xap_K;e*AKGIx($ETYl|lstBMLN? zOxfsbeVv?y(`c7M*3WPROE(fuh+fmn9-`dAo%{`J4o*Iy&%sE0tgpO%{>+n~RE73USIMOVL%O`t&J5cK zI}ezCOixcYe0^?q*ZL>S{{p!|&L#~^IDHjelU<$YhyldAxb!hVfn?|Ljcpl0r=A?i zmKUhSBx{d^Be%8bjZVWIjoLUMXXzPvSp5vnneZY3?W_K0Nv9Oa&Te41kkaBb9spDl zdpyQy*y?(K-<|ZG9*QHE>OdLsjKyhBOtV0af`qUz*&Yzb;o!@*s9G+g;@LkfXGIqn)95@CuJl(-tO7gt=xN#)VF}4%Y zwz#!ph3w=ko$$3)gC2ZT(3|iRg#FuC`t+2)?B_C&eWurZwsVLEqPil}!cB+QMTVMo zwlFQ!iKv>%d|M=9?J)(!CwhPEQ;Sl8QBCdS*pCAxWMP5ocI<5QQGE6uKI?4;u}|V+ z=!4Ch`uiC$a8m-P_}Hk7TZ+43qiRaI%Ds zd|MSefsC4uo&J`IG`G{6#>Nn*Ocu?`t%X=$Om!E0LZHa}0ey0UWflF|7<1*a`F9I?MGLQW4nI=_?i zGK}}oCdiVI^^{a)8CbEkw zOt%BwV1+<^BFrCWS|P%1m+X}bHv4Yzi;2-Dk*25(%s_B2N+=d`JCJ$$Cp@*D+BC>m zm^iKzNZcYA7Z;Zy*new@u~#UMajkBCVfr;+o)Cq7rM|)a=VL#*aG>(~S^x=5m*Y0h zzE7%RoB3L#ZycDmg8yQW!xeH*5Bil zWW`4Nh9uh0bPG*aE3;OizDFjD7Bv;sNyRQCqejNYnYQ3mwfFc@s%>rs@lTb&^h@sa zyX4pnG_KgnOx?Aq^)R*?+=Q4n-J+{l{ZoLmJC;mL(tj*y=S`?VS%$rlL;P+z%(O0diiLO&$GC(0uvOl}~Yww4m~~uF8=Ho8Q)EJ|YDNDb$z{ zCt`%~UBtfj5C}JzZ>%nV`7$v<_40|w_{M7)7&)@p*EOKE9Z2u+bDCb`yxEmGdtyyB ziIun)XTFa#*tHkgFNAl-JnTxS+4)e8TW4$K^pnMN7dgk7N6H8!s0I}9t zL5O?3R(-E;0pwq*1VxITb+%27jm$lQ>_y5U>CZ)yq>eF!VvrPtAjqQJGgz^4@8A%e zjVwTYH!MBY8}L+ZU@MZzeyC@bsjt;KW+=JHuQd!e7p9&>eUwsp>DHt6r`@?ApBS(fD-XG?tZHymv(Yc+PhyrEQj(*p zqqE&Ph?N;GD{Gw{SWy2T&R_q(|G_3>u*+(6#)Tz5{c-#WKkOxa0?SI%5?#WiKhQ1 zbKMd97wuPY7ZWT;XIKaX0uz&}#0CgP7#MmizCq0c?EyvuCU(jKRVj%6hlaq^ejIp= z!2)by6+EY&j?QKD!`p+DFzY#sz(zH(9y6_kghDM@;7+sJZL!7Sw5=^9shm7K=_x5y zaBM^x4W|0P{+d^dl%Eo(0Di;^{)_ucc#sIp#gyd_1%dZ0(>#ilXK^I4v(;mGRVSu$ z;z4{3J2y_y`!~kk^c)KCRk7+3*Bi=*8-{5jR6uVlF+j%_3(xciZxh+pwonOUE2p*uaittFvMIlsqMyY#;i-w(kG^td43w-?_ACp8TusUwHCXIko_Rt z-_HQx2Cf615nL}-RG$3wfTsVOuQD6U5x5LMWx>kI+Oi695>*&>&sX#nKoxs%|Nd6E z8n4<;&-ADHNas6o>#VF8YEd{h)ocNJ7f8R84XD@-wGNqY^h3pPn{yjIx+F^NMQtU%s%tp$jB^7C`0R#BdI#E$;)J~p8Jmv)9h&SZz?Wg z5?<%L@^I~t$Wa|_Cdv+SQA?g+^_1;e9U1~o&vda^5;llh9;!HHv@z?9RkLcUAViTV z4%^rR^T^IBeey?1(+yc zn#BI{No~UpSs5WCX5EKpgZp#Ebdnh;RMr7oT)zc@FUfM{(}O`w?L-`pZ(`jddz&*Y zwRPNx{pg;yv^~a-r;VwIF~niZiJ>9YOR@Jh<~b2Wo?Z742>`iwv+fNsx*s%@)YLdA zED98Jz9M&(JK42)_i4`3ew^CVuXE#%>~`&^G~1reNIGAmx9n_s@!~~KPX*#5(}y^@ z%zn`7DCf?DcnmHkjD@cwvxwjOe2kGTPywHsaZ3qc+D5gGO&&G@clHS>*qlee#+$(# zq;iz7ezbXGd|Vu|0~!e)EOGClv|!9~O?i8Do!gMNFqD8ZA!&h7;@1J($^tTrB~93<`5hzFUTa3yt-BTAxnt5K34>IRqTM!iVR5Wu6lJTecwo>xUWt7FCkV9_Y%J?$u!}X>$ z?=7OKr_p}vH&zZR9mIA`+Hs)c&@DysJ@5gvy2Kh>!$ir_s=!8nkc=lT-YNlAaB*!? zgD9s}q6B> zc3l=X;iY#Do$Omh4lkIM&j$vSeY3TW40VdN=*gTy>f}AswQruh)It%OgYe^cY_mLm zD*+e!78X3iRU({YNKOuFa$EL+YTJPtO7MaDR(Rj&6~hmkRdJtq(1(m`U7OUu%0|EO z7iQDnUmW<2@BOdAwO`cIUvY{&^Wg7;<)NaGC$%2lArSv6TEVqnb}^`a3&6oc64;5Y zu-GXm-P^SkRGs;AOsFz2HZt@&L(P5Fv?svD<_THzgj;*P4?dQbF)3pf7LDs6`qFKW zsvq{ApnlC(w6~+9(%s(fa2QeT>Za#Ug>b47h3>${-Dv!ouX1|wIq374><&iJzt-tM z)Z1Cbu1Er`Jmi#>;VabNAR`;9?-DPOwVLLDL>fCfE6G1ffS;eYE9_`SIksF-9S%Ta z05bKBJ4*l->up=~D-r+fOFnMT4of&b*sR*$W;3KirmiP5sXssQVrwjE+-0~a z`+gE)VnUiO390Nd?1YbHs|rz@v}uqDEi5g8Jj#9Q!2bQO`$P%s^rbGC$z)=g=Wa28 z$j4V_E)_<*g>ntM2Zx5_9hlltzhpQ@UM#AyO~F&M!{Kv?svybEgbPl7@NZ<~QlR7z zkWpG^1t)TA>8Z*K_6~VYqqDyr-%-};xJDrzyKCp>6=o<6d3=2+na$_$E zkM{h~6ZNkDwhp6F2NE4r*gd#ia@qNmXDuK=4={RW(+IJRyrEbWI=II8FFvmKaLWY~f_>~&YB6FmFx2Rid3QJ*xmwnk*p;{<%Mb)}EtY+wpc0J=J$j~Smi zO3q4l>OcH%iwe z0R|W7jjJEReI71P4kx`iX8vV|qUKYD<`qpVx(gUIUxMszFpzM>-+=v5v&wcbaKZTkZKtTWp{lZ(A+tm>-|6JtoVW~N+%GI|3U?)yMfpiQF z>SU0Er41MBATc&C&9Gfs7$U}Gy>aCpg}Y2zxB_t$<;3vtFis`01u+;rBXooDH>NhE zQa3nw?5)_<$XzVRc~XMx$gI-YWBZaY?!*Yf8`Q&bhGwlet}qIg^_2P_si+hf2%1 znar4-TG8)^&)BnTIQLLVW>i8iGJ>+2aA;$1G=Lb^R&0_hSFe&L>p^R{aA~TYG&I`b zKxGmpt6-@yd7xtgg^t$y=iWGPfVQ*RCk~q`U0RhjycHk+8VW)?D^Cz1^9IbPqxHVm zKIJ{)q?kryeWXtlzKhhj%jVp>YQ=cCxDNdM#C{I;KV3`nL}C=w;ChTAk7u`1bsneq zshcZ$}2o zPPL*r=e6TOhkYs0C7^L11zH#7?iNlO%T2S?cmG6;bo4Q`ETepA4!~g7ABc@)!HB!> zwg!1lw>?&2_6oBO<*Y76n>iiTLr8tej(r7Ua%v{jNl;A81tn+BX)Ue8Wlqd;61N#d z`}=%54ViTx%s1rzyg~gphwSgOs>xc2?d}UjJgI6p@m>@Dcd|1WP#Mx7Q6p z4M_(H6G?SepX13yAQZ^#K8)&e(Y9Z7$FYuA>HA%gDKZY)LQ|}JKBLhdo=XyzWn^d3 zO@nb3a$|Vhm!Ut1CJuxdkhu??oY62gb@TR)iHsyBY3k^}~h>9AY^XY%N8RH{UQ!k*Mj$SU13iSg~_gQY3b)0!Gtn_K^!`$(;U~)<|(e%iR z5PnbX?Ea2a$Q=NA)Q$g9>PSV!T$VncdTH8Di`TW5d1Ar_KQY`k`JKJYw=^0mK?M{4 z=OoqsczbyzV^(hN>vjW*09tb;xWU-%pz1{FLt?mIpJV%pS{i5m)Xq}%f7g|_ws&Sa zDaF_*?kYCE(7#rsGFJ+d#LZ+;-23(Qe*jnuA8%{Zr>3TsyiW2qiZYHfWTv51-9 zw-Dm-q%KsT8ZbQnbF~~(1+~+6s&&|v-pPNg-tqIo@5>}_pDKl@?hev_i_ZXaRaJ3T zQG*k^6YZ^mz0gPhljPO%O z$9o)~X|>k#+|T{o*L9uOd7c+~4-ld_J3B+Uh`1q8dxY2ZmkA{_Iz80nhb=_`nIiL+ zv}nRn60?|NH(YxJ&e?i7%V-Y$BgR`wN<4Y?EUs|N=FNmk49KKgLePE-p=L)%F=*5e zm3+P^=#nkUuN#YZqErw2E*K8Q4jn}iKv z)FaCY&O#pc`RQ_|qOZqeiNp>2#VG~~a~sYp+&EJ2Cr#gPDKR=7k7uvdp_WMd|ZwC34M^4cSV7M85ykI^7`2NBXSpAOt7lotlqhA z6hT0D#_^86!98@LaLKLB7<3kS5u;D~`mJ}!x0~t*Cx85a$;p~Is2AW zVZr!j0W9rCezoz5iA6N=%6O2mt?QhmtdI@4#Z*H^je6U+4JAcfbY$#U0d!U{Ov%Tm zLU&bQ!^(X)88Y1f>qPsNG+Xr3B);UstHQVfofNYR2e{zHdK?#k-ZUd7v)b7~YYnfo z#r2seq$tLG|Naca)`MpJ5^6^zR*xl$Cd$-=0s4>^6LUq!WM&#L(S+JvEd@t%%c40B z$Y}8gVT=iQ6ePMiHJfAE?`zw%%UQRiavY1fV5<##2ltaxC{#~%_MKrl?b1Ywy7APY zy{O)|Es!xg`4}&XT62oefZD%oEzx$?uY|OJ-<2)~XA|L=`+qXFBRVt5@yhL3y`xPcn zc>V~{q0?H8IY_(`%#^_5Q8WPsnz{`OF~TZ2+nNn{7{__xEBJDtUc^^qnUsyv{K}^- zCPUiePW=6&6Z3<3eUW`Ej%7LHP@7BNPI+YAhGfl8njO>eHiE7uyC_R#EySy%ac=Tfip~%2JMff@Q zi79*U^6QqY#D+o+y~T0%eXvvsf_e;)?_StA;vgMywk!EH!r+>c#BglbEc8MDZtehI z#%L?2iJyZwnEj3%F)Yfbrqoi!KNcBWOOhCx@<>Xo-==LTOJrq^)rP!Ks|%$plo)s6 zMYh7)mH!o78#3Nj1!rROY1td5=7CrCDYMG|{k+{J31iAwJPW2RW(UWckfI>Df{+94 zLZnsjS^AwD0D%6Xc=^s_JDrKN6nm^otEcw-yMsp%;su<=Gc-3bC`+ zNM}KmL>vRi=}}2yoz3tl5E|r+*sYr?!-yxU;7Su)i%u|C^4hWu#~RA?7cdk;n%NM- z2A^Y$CCfx_GAb-+3ayh5U%uGe**$S1(nfE=_Pij{k&XqE$AW>!l6a3*jDaDdty!0J zx2#P6R|>9}z-ehnDl8mjdh;bF2PA&VWPq+I8zslD_Af1qR3GeVyWOt=<6V|eHHi!Z zbX%0syF}vG-V7)p^Gg6HeDwZrzAWCA8a@MMXee`2zQi8}7}I<$fdGXecja|kC53?N*xPUG{VC6t*K2;m!(Oo9%%mN${U7!>garQeL?XjiXz;gz z2_BWErY5B4E(fG|{%b=p)H1A)2o3r%9%9I`ILgbqi?(Ug5V4Six~d-Bkc>72=4uF> zv0@k=d^R>_%6Z zQR%u-dbzv-a|c8>S$TQ(%+{NL084NHwW<@rFD_YNA@05S=I^`8unak)l{>Sil`&;D8R1EN!#2W(cbsZ!uYh(2z0V9qaIeqI|H{tb$D z-h6MSmMIUtimxuz=+W>^wTP`jB)WdJ7|L_5;tN$`0u5hlr7G7OI98;a|NfE*mH?eMSj-r&U!T>hzI?803^CeTBBE#w-$RKgRd!^; zF2+UKej$RD#;It^;uHAQUQ)a_FiVD)_iTq7B$?V|h+qij&nY3jniQp4Z+I_VQ{{uD zpu~Amx!8Rd(Aocn6sJlwO<$kr>EZF_^=rz3;!;nFqCb(|FLd8RaVQEZ@qHKeC@YuA zhNA9=YblAcjG;uB*}WW23QX8B%$p*59iM@iU}5^bbIJ7y9J^;0UVBnbTf19cBaecL zOZRi0Z$ad!V&PZIOiOb;eE0=a#z=1i?!ANA0$4xx4h|m^ISRKgGZqIvZvB(}U6Zh& zdFwG!>4%@^o_fZHhJ^vjjD7S`K$q9N#!PB12(RnAJpjv$LPVJiAqS>-|0Cf7xS6#e zPWhCQ|IZ|g23nEm1kvlQWx`>wR{X(nP{LY|+|oU1aLKb+sGV|2)ck6gDDnn;G_Eex zL1h3Kvbrd4wAeqCwYHKbHiwnY6CQ*Lj$YYKxv)D4`z~P4>ctc5&&)oRq|iK(tuBSH zW-Gc=(g*wj+V$I{id9K^?hUkB-NK=gk5mFa#7;F&nmPYCx!29+b$U7vwBVF*V{J@s z8MuYT-dIXm>~Z!+Zg+re1`;B#T#p_#?)DAB34)UU?`O@(GZ%M*r+-C!MWJy6NEpoL zg#qclNSC&Bam5JLPb)(>X@1Sj5=nOOr* zt5DOw0?6_Ai!I45533C(Q{MMvU2Y&E72^(73&yZB_%CB$T--uo{9WwNFwL+IG*7Dm zhI0OB?DTL>gE2n(PwXgriTiOzztDT*4yZ&W#se7D&MfPhUzY=n8>OF*PY{ z3jJ3=vW{Ep5%>__?2&&04F={)nF#|36?OV62!6p~$$W2p%%CPomR~vCOnQjxSL5J- zyf)#X!o=)`L=_nsOj>yx9SxIHfBzYP#{kK6x(=|>@UrW2SQ&&kuF>LHbTff`9rj!Y z0b8K9!`A0MYqTWGE5mNK)d7!Lm+6EoC4vPF0PCXM{n6fYk={D5V6Hk2I@>k9Q zO%SG@>`_rcK{xjK^O4G&tgKhaP`7Q{2KWrGHpQst4ue9`5u=g0CcH}YdMB6pN>m9C z_d=NWKZ);VW!1BQzV=-Z21b;dmPVaHnTN&7nv7Sz;?*JTR~y>yc=lgL)^gf1jZUO0 zR&*SmdH);?{~LS3wkua(h6D##{;$dBdZ)R$8y*SP6=SPr$~~k!H|qc4YnB*Ju*|6e zyHrljs4$(;i=hhGS>q5Y!xgJ0`$rLmI`%OI@&FP{ixr9$`hi!BnbT?02OY{3_? z52));6)1fQPtKd!=dGi3pfX5BT(PgQz{qGt_MTLbmYzgo_4b;CZ>X((Ce#0m?ovGd zpsI?BkCASd8TqpQ8_|*yRhzAp*H?j`UyUJ6KAYRkUO_Ue3*kqGCm7U5NQ!@@jG4`Y zMmV~N*e$H=pMT@VjYkQS1*8{H!9Y6PFZ7a<0-6LQigw5CyYMVeb-ge1+Cf-b%Na|B zg4~Ao%k##J4lDj|Pqp4T?TMR;pB_os3qUYz5OUq&{3HjRs}$0vH*bD} z{jvQp5!Ou=-S=2Q2~*xBDVbmwM{oDyL|>kj3|M2QpO>sX9zGRG40v#1O*ttrBZY=! zA)o)=i%vks?b-WielCINxXt{RT-J9)KOz*mSTT_XiEgc14Y-ZZ3AqmpRzF(D+S>MW zO?LUjYuViN(ctNm8l0THQ{BToQ^&&>hF<3#*fzGwD)5x%;JezR2SO60P5sw|k<+#j zpX?(eSxJ2=8;Tx}Oqf_UEa@H%4Gleh>}e{Et{qHh{L~+8ptPWW|NcF%TLJ)V8Egme z^UeS!SQG;Ju^3&^oef7C9)EMN(Ag(Cq)K%kyC|73KM1m^udgo}CR)wekS=DWKgk($mMq#$I=BKKO9>T~^t|h2!7+&Z>*23iq*6uH0dw>T%%7 zS4^`Xf8iIC*C+Zv5sUUu;GfOby?VLqLm@h^(ROl2zHa^Hk$EsL&h(m_!gufwhIwKK znhIN3Td`OLcI>G#_1)^{y?FL&&xbrZI;~-yK6haG5UCc`*_<{bXOL!62AXrsTKm5_ zFirEhl_J(LXkEa0B#b~h@Uj9E9d2uO?+v@G-`Y=bntl_NC(P8wL8rD0a}kh!SvXXx zmrYcpowo9=9GLtuW;c32#J8SJ%SVJxO@c2O=Gd@$s0p0xuboY;AX&ifxBHT$#dvP< zfT>}aSW_0CmJ*xR1;nz>;o*a(%oyeR?;bhUo%Nku*I~k`bJb;tpNqp!i*R16&mE`R zv6t6h12F~l|DY6Sxr_{&)4y}ow***!nP=K4B`2e8ES@MKI-!VueukfhE4tKNEssVH z(Hrg_XYR>;WWq^T9>5fO9chh>B>iFhXXzq~*9|*2=@k3_E*0^wQH~yvqf-lBoN+}< zrLJLTx)$-$Sm)vWhG8B4{Cdga`scO#+|jrT^!HcTzTF3)vLpjVYGh3;)+BO&{^syH z%bi#Ld|?My(YWYDb!24Z)-lb0HzgMQ%}migd_hT2K1K8s+4aKY%*|?!6h1w|v;219 z=c7&^X_V~+wUjh}VtD3YHMj~g=E9rp?Y+&&{8WE`+T!#Dyc<}qB!iiZ7IK`5>8Zkw zZEJCypTzWo%9|boYAT~>D0bfx)UT=O&nNAS*WI){$%ut)?B7`WzUlD&04CD2fBW(7 z*-r9@^CzQem#aefwJMmjE`XMGH$PufYz~!V>c1!Pa<7bDa(i#fa9dc&4_7Rn5}n$< z5Q80HD6%aOU&&zc>f~jSwj9n0%uBGF#Fhj_uI^p5L>4K?Oz{N0O6$?%2v zqhdi+9(reRa<9m;8li|}_1+zL@QJPTq{CrhVV(@TcI`^t$j@(3RGvZ4`%L9?kHMRU zNJ_`Yf-8p1$HpR4CV2ZC)*_xjJ1?`wplGklyJ*S9 z!#hpV6GTO#@*lT6#p!bMjpsBdH_(Eh`vJ}^10y5+$ZtK^3AzV27uP7p^!WIE`S4+T z_~H!A(8?USd7lYEOP9i7&SetuMQm>b$IkPjV{P;oz-fUB_68D+FvBQ1xhDC~QEx!a zMD23RzV^#twL6$*a|;V-4V*#Ka`v+0;Lg~m)q4Z(gv#iuEwP55XN@k4XBCFc>}~~x zd~yH8B-a4g5z5$}S(w^*;s;K~aCHp{iG8Pxw@&esBF&_d9j<@)MuoTijg4XxFP9e2 zNc3aq@zGm((p@mV^Tx&;a1vy%K(`bZqS_yEeOklM0I0+? z#HZab^V_AQv@*fi;TG_@Cr>Qt%+D_w$EDW?vzzx`*)8Lt6enVPb!6mBcsRe7e&`gK zXV%Lu8N6y3a=YkjW0TowVO3LyRJ-Vi1MeT<*3A zk*$8Z0Rt3)hjcnn2g{Gsg4&qjvySA7eU@qLCScwEC~Ir0>qq@>G~vYi$@wHSqW=8( zqcxE4n3yEHdQK^^5ujyv?uDiw3NS_O{S4sJ{+Y@K;t?T??T|D!<1(qQ;#v)2_7mS! z%k@F{{>dMk9=4$3iX+qmONY@K=&oXvN28=vq?fNr!`Bry!2xtRlDicqf&+ehyf=z- z)x^D@TbU1x@1-{R9yOrt^TZv@UEuTHFDiH+Hc$qQ`Ts!^&c`M=B;?nRAFnNwZr!?d z<%)os?+Tu-%QC7Ni)8GQn4=4m&p&fA*xqy`58f2x*ObfLa??t1*Uet;sn_Q-_xY^f z`KcOAC-lo+!Dep9u3hnVZ*b7Rdl!z-H+CZeDl)XPpVPuNIMCeHRZRI5xTt zWm_B6S!C=PQwX4B?+d0%)LdsH2d7~5`gC;JFUgRR0X;*DiqW`J46Cb2Sp0#IGA3&pz~+&HVm`uk?vdpx|HbHyC4LwY6Vj zl3uEK+IHz0ZCed;+{e&uRzvA`VNZacn*y|nTZL(Z9PSJ#iv63#~BLH^+26}mT@NL+@G?^x{VW$tz z-ymb3aO#i!n>&|si_f2mo+?zXq9^ZU3hk~HzUT$nJ+&w*ZJuf zimSwNgc&7-uou7|*4$Z^Lt|x+aDvFDhM0f<+7;2AW@Jp(dO zG`Pr#{taKRcC9K)mrMFha>&PRJLr%&o_^wO#Os6wzxX%mACVM2)VKY4zvxl3-EY5Xr&eHV zfbRqFJ>#jY%J%KIaSRkyU~uMyeOL1O{*DRxoEtcVSIGR*kE<$@XKQC*hGN7&YDT(#-78>fe>_oyArqC(7 zrGE(z*X4Po>Z>w_vfXX-8n&Z%&D$Pkke-)ro*u3q@}d*_qKKz(X3u{NA#qePBql8K>K4cSnPX$8=;?bRr=MLFLsugbr4exm=Ge^0hw5?Zn zS`Xg2)K_t_AzoygrPhS#sHsNc@jD&24i;2%8Jj&9UIw=rt?Fd{8>GFsKmMZrUonmg z9*Lic$?B{;DmKG|*xO{Ww)=4tK1B@W-WP<_LI)s#*;gsW2^T+UVU0{-714uA&OGn* z2FtdS9(ik~UM?(Aqi7f%9Nzx>TK%AttB#k0bOr0}lma8oM&s&OG}KgBI+P_C$~Q8F z`o`=9w*jJ@+xPD6X}V7!Q-cn<^!VWqjkEj0xBA+c2IzLrPsjT1CXkJvsFMpKp-h6T z6eeFk$-S6F7y&IOj_cZ*8a56NikaBu4A7@v1Eaw7`7)Xw`mr1FN&9?KKDc?SZrfvC zdbC5IUyM`h0&nLWUJaHrGvFMe1_43gode5emmf!$?o`8|0%NT9$jP1S{S2F~x|55V zKHD@1B_lLF1F?kvhU&Ygmv_~k{M;jC2gb;~-=?7baTj{UI+?6)bxlbO7-zn!*6xtN zv>y*2V1Nch#ps-atnd}YqEV2u%0?K`iuoJ0nVFa-P+a^rEg*4it&-S(Ebvs;uEieR z*v-0oDJohJz52P|nYVn9ycT{OMvXps&1I%Qf9{N z?wL&bHMQZ#6)UEzAw@lP@}%8?1DI4Cm0)NPMMu~FvN*Cuj8s=$E$zn&APrQML*wWW z3!v?MYd!CIf4V5mk{Jf-+*uwm;!m|*?vL0&`~_|{Mgmr(L#9jvJ9g}5$w%xqFVzbo zh5`ya&puP~Y5XR2#k3-Ts-5)<2-^&sgcRLk&FsFv5b=5tucG3no_gg<=DK7lDX*Sq14wHQ0JF3AI2o9)NuMKe-K}N;1i=xUhEShoex(vziIh9 zndw1LqE+&jFp9`4hJjt?mXfmaU98TR`WaA;Ao&B!S*;jS*M@udv?QznT{@jU{RhFh z|Eta+j15bnXoz^!0zNzxDwaazKkp@+rvI|`MsD#X?jJA3EP}Au=p~ySwz5jsTqSYv ziCyab+|D90^aRZhv}BDqsTk#-6XADRm4?yZeS0X_$`c1#xVFmf4U-5;yVm-#p8Moi zLir}$y>)<4Qj2cfIO^p!aos|cIywJXiTzGCSiL1jv#!&LCmlT%oxnMRpXJp*B_}Ca ziW|X9qJg`-`Ymo?jHjKqWrqjernmKHbE4);g|)*sZnB_9te4}{kS2CVEuAnYGu`h_ zy!bVpLu6~Zo78!$bwUf-9Mk+HUd`NoN@TN@7`dTSbo^g+KsY+u!RpPu)Mv`cC#8@(@4J}qD;zv{`J=2>Gb%DNx4_?#2Bzoe`(r6@GKQ&&J*=3?kT+WnxDO5FIo zh}M5TFa2sYGh30zDNFKh5clJF@h^RUYbUWcT~Vix*K;@wQcPJ6dKWnc4!Wy6IFe{{ z=Va8o8||&vMsYv?{g<@b!UMM7UyA8p`L<6?kocePR_*BM=i7BWQ@x54KPmH{Z&qo! zaphtI8}Z5yx>u{%>)z3TEyX;%0@x4UCyKIYvG zo$(9erhWBS{J+bQzvq2uQ?)ohWWjTGSZ^^W;h`+I?)&g9v3DN5N^1)`qWifQl5V8| zhS+`oE@*;HbhuBeZC38tN_iJ|;1S8(-{<|mdi;1HyKd8t+vrOOyxNk$k!UNoJ}M!M zk?G2r>l@|2#TCZ0Wc`uqe7a5}qDQ;T*-=jZuITHutto5!=@LFH$Un<_VE3QJoLJ}g zzB8VqEVVdCW8dDWh#jL3e;{~Ge{@({pHM+%b9bTsQ%ln?j~LURQ=5GqpYmI;D*vEc zyS2P*7V(N>wl4gKcIfZ)LvQ(hv~3_y5L}n}v>mLER*f?M{>8Ier5`9~x-xoW;D04K zWHtCGt*%}<^n!&vnAY}kt9E1O+uIqJ6vFTNAC8`==e{VLwFQNT*vmi1uPc-_R&g;4 z*nWDPE;%t!(Jv$}y72fyao-gCSKSjHiP{V-ZyQ*@VnoNS@cj`3vb)20Uj>8rvw>QYi~;yA#rTLRWfj3#O~ui?V|UazOkA1N|yM`e)HUakG!ak+T;GhuVuqk@AoYCms!82 zUCNd>?IYVNOIY3YYLZejW-IVBH@_!C`#9RWwcf?a-;cMVvh79GtSR@^d_6r%$nUuk z7oV{t>Ug!~a7W2c(v{sc5-l(P?crL%zOKgm{c86+6z{x$T+c2mA*FYfDwj|3!)dE~ z;!bSlp#cH4dtp#dTdY&I7F1MaDO(^txgook`(zFa!uA5%R+zF5!UY6f>P&7LAY8Q3gzxPy&=j4J>=%=d11R|+-nRSSru4k`S)~@M~ zMo(VX>a2Vwo@J1ulT3MAUu5RQtFrIhNg5GGSAUUF#N?-Er^ARNf}6`%%&wN27KiP4 zxTBQG_1Afs=Tv^Iqou;r%jqLAAbx(65FB)1Vm$Hi^K{svczYbmfx;%fp2T%MIb!x3 z#O>F~MpvG__L10pov(Gj&)h!p!cn)uUw^_n@N(q;x#RhSO}GEw+9cZK-*$hPE9v#z zN;xo6A+*8kk3IS9$BPldFP_AlAhgM1LE;5dndLN{Wh5O855m+G5yJFuN8ZWe)BsCp@=EMr3^w;r0~b5>XKyzT@yf z-04Bkos;b<tGF&_K))SOwCJQ{>+Je+}9YP5@0v8%*eD% z&m0!S$MCzc`4qppV+hY57*)eR6hZa7JpH<=#k!SOg9-8R@Z668=!~Yp$O!^vSZapt z$2QnaMO!g$Jqhgu1`Up&%Q5f)sFce;U}3A2RNwRGF!rFWmEucA!U?SO)&mJt@6Z&f z__w+PA_b&%Xq*nNp_Bc@u}-*Vlv$ScpdmVu3u9*3G!&TqxE;-UzvLo@+Sgsl=l8|< z-gs+1{AD!yv7n$JIP*DsKu0fza=0@?M_U`VlG%Gu+ry{=O&OLW$6a4p(R}9lKmP3h zgcz*u9Y-<_ZBEV}xB%o&a7&sfPhra-ZDIo2k*@d~h?J}(R z3vL1+FYolk29-ds4pkK>!)k1Tx!iRE7LU*(RSEg-d+Zp8CkgGV{r1RFyMA@oOOObw@^0Ek05YUrhZO&koxQ6MrzU<0|(o4#7#E@KeCJ zG12}HS7D^3f-b$Nj2~1$euob62n%DVRsno1@UuBr`=dJI!TfpxNewCr3R8JT zT1AE51|8aX275H)c4IOw3I##cU^&dNShefcOFpnN{)({e;Dg9!ZOAS{NE67B6-JYI zWe1NXZt`!sX<5o~qF-k57%T~{c1Tdeu$W~{9jH<&t+R_yby?Yy&Q2ewlCJNbg7*FJ zO)65npOcS|%vbh8!dYF=JMC^CI$K#;DZzWA#FZe@w&g8Gh?OoqQZPR@xUG4GW)DTz za{_jpqNJCa?mSIIf!)VqGtF#Y3oeacI*hrYgh9<{)ueb2&CTM3)KgrvTF|$;xw^J? zM@`W5O#0R)XfM5I+&Yi;VD{H@!fO0A?V;^FCx3kx3jYdy>(19Q2bS0!ATdENcf=Ep}*RHsPfR565A@Pk+E?()j8E94o=Pl!=Kps zb-u6Ht*#$gcLrpoi!u*NQv=(t?ma~_`Uh-!3Z)Ax-73Phx-%2h+|hVy*rNBrnn&~-LN$P?r!(MNxyJq z`s3)RA|_Uj1t|%hE4Nj0Yy&A034Anb8+Aj-v^umbGT8Gm{-TK$&dsZ6{EyC4uN8>; zJo^<^w$}}SHC>e18Y@fhZy&ZrgBU3awNn20rz@r`%J8bmOiwpUerK*E=&4+RtrKVh zoJUAV=z#A!dOCYdd!|^`96x>>)Gc2xue+e*2Rjs>2N|#fpm-d&qGPt#?fqjreSy2> zw*rJQ$9DzdOcj-cuC8YZ!#zcJMYd!N#kW>Rw=}^)0f&722|dh_ax~X|Qt{NL5$c*t zrs`3tXPd;Hr1`0H0kdTE^z7@Zv^dxl&YX00*HQkWogy!!My97Z^>|Y>@@U2AK7kXT z-C;wG=Yg4R+-G$Id%)I*MDu{}MWYwZoGTywM~~myOEfI*@9Q&L{9XN1v{<`+VyCCCE?j}C&n@CAjTm6h&iSo8|?2=!jm zcx6QKsd-gwxE{j;0Fpk>!Arm`wr$%5=cHlAGNl!prExjq?Jy2FCW8kntE`-zb15~I zial)h6SITy#f;}slT%aOj*_VRRqI*Ns$#Sifg0k4f0!2!zm1)pe#9E@>Ut0y+TGeC zFJjdKxUXmBnact$Cf=8syhIK;b^(_`1bgEk8623>m zGRwr-IRc##{$p(1$Q;wtrB=2nD=W9ya&Y87a?{RWzmzdt5FgPEvEA98jj%@uAG2!~ zrTMAF!@izBum}9BozIJ@Y+ZHY*`i|d!+9U2`JG{J&1>>HFsUO11Q%E}PT_)RpU~}6 ze6OqB8EcBLQwR#0?M=VG$lArqs`I_2H52!O*^QX7S}N81QDhYD`#}}`hNlYq{Tbqm zLm`f**d1;kV*b=3GSA5XFC@1{R~E0PZT80PkIrp$_wZm?KZO_$sU7+pED>$bWOazJL4NRc+wm#Oe%A&n)wsE_8qJVcFq|2 zTdw{a+iYog9^#bqdUsdxPaxx0_Lw6|eOz9QtLq`Pfp>vtiqaIGC0?GRFoC&u?^H6M zJeb->8ewLIEzDL+4oVTL;T7pae(DI({~!PNKM#A6e)qR98pzDfwi`31#8O)t00|Q2YeqX+NWt+`cVgYov-ut=4wr$h6D7^XN zuO20V!C29lvAffuRlk7~>=d1}I*0eGKEi@TcJt-v^?zR`@YIXR+U44rjkT2jiBY zbLY>avk|tk7(h?dQ4j%)9@Oe7*7`tm1N>tkZh^8BkdmL98_lwI9 zkjbA$ZKt9R0M+KcJzw#vvL zkemUg6l)|!nNZZJ!&u6utcx_Q_(c8~@UMYN6I3po$B(_fJ%-m}l#cFMOomp`z-52` zkwcjJfw6BM%1qEqBZG`xfpov+R0Y=4UpOiVv!3N;f9`P1ZrOMg$ttBUrQ_-8iGhaV zK9+#2qB?*!YaSVO0qeyZhmwCuA_Sc=%-8;mCWzdf8HVpb1=R+I%?m>z@3koYG)nr? zVq&9i(lV0y484)-axUG5F$G`uzGvQYe8~mrSEmg(QtWA-?e(sT!-m2cLEe+^WmlBP zrnSwA`u4?)tJ`w(^2BILZ7CC&cdQCRKEIfICW3pji)HG5`DjvsVxSR+9+mpnu`!~l z0F&M$bT-H2qVcxGWCy>zejtz|wv%9QISGP-!e@kr$$p#%%Ja| z^Lw$)0gdNNq_qAD4mu(^rwVf zTUV(Bx8601G=_OFX1DScbq1<((-WId%`vDKE5yK^<|gYzAR>KEF)TBkgm-*~q`c(g7l>rqa zOqH$%?^|+8ibwDrDaOn|UYs=1^Jp;U6vI!ohBWGn`psh!)5Tj{toY?BW@yCYb{l~9 zB5ZFl?7QZFEc)l@?jrml7OkB^8`{02UsoD6Uu&qUBE)h|f^bp(uB5F;{p1Xl4VbDs zPvoVhrebHAVT=o!*Sx!EJTpPDY;9_PVGM4CRmJVCIPKV0LMkT{ZI*-y2lkjNg%??v zE6p7fjr~5uH7?MNt98&K+HOv=$>wzyOS+jtI$YfA(!PhaXS;t=6kQ%?=~<}8SHKVY z{3%Mak;S~G&2TFZX#)J$@^=(>8uYaaT+xUcs}<2A^c?^5J1CVQHUfS==|-Afc)``S zaF2%t#k>-B1fo)8Gz z8@H|f`E?C?ZzO#c)?7#;gbMCK^-;8i);i6H$3 zMkvQV1h_q8Tgdl9>hg#C(@i^e@3y@|6D{gkf^IK{#^1PJ(;V#_^(rw&wuKo}egJ9ed1zRV~`=RcM3IxHEB325$VU0O>f8A-iB~?~e=1kG4gMmPx zr*vv0ahzM|g_Y8ar?hDmZOeIL4?+S1>FysJtfC%ZU*;7M5CFA0gJQYY-Ay6Rpaq9% zKygmlO5W?Nm7se9cf)@b2YWe362+14Fl%l+@&!iEeahF`SSV(>wN1B4k1vRb@bc0# z?$FSP07VI%;rVaoIE7jJ3gPnr=gC9GoX`8SdcR0--O5nh*GrV#yt$8<@8y5_^7;$W zq|xoy%DoPFi?Uy(J`$w>lVPc^{Yf@w;;w1KfAl()&gJav$aN7=aN23?-J81=V#+|M z>82#;E~QfaB;$MHY#U_fWx}n?VrZRCZAeD*VZr(vLq}?knAD(fS95=aB|(;o&OWZ z^2{_S?gxqlb{xNe0L;x$GFcXUftwi`>MB8BPKJg`Y}xYSo%$xGXfNq}FZFB9uz;`> zfrGA~)0+YvR0f~XQ$tr6BWEHjA!*dlLOhD@1w#QNa=8Ts*+@-bSvaY{N-A}ZViTE! z?l z_ZzH=vgb+eAh7+Ket*qA{P(|>O3rS+7SeYH~K?Yz7p3dy=3r#}*hCz6d3NZah$ zUXTkc$XF4abltg7Yv6`0CefCOj0GE9FYD36hnj2ymqIPV5>C$^Cvi?IyySjL-+M~A)5hkl6>#DqoZKr zJiaBBgPHlq?5ta6NolFhByJ6M8wALV2ocVAvu>&qiCFjTfGY zZEDbB?fdChbL5qYvEE}ms272OM%UDxjP>-~+}wb!L2V~0r)6~gQ$DW{?2u4K$$@cM zIq;p=e2fP~$11*+oW;(f}EV=XP3;N{84 zgCBVG=qQh7mIdu0QRjR69aRk+m-$@724cRM)0Xm!e>@m=SM(Z^)D~B20ho}$SAb_P zbHFTi%HiU6TBM-UA3j{n%mmSI6nK)E(hDwQSSPz3KAZvN#*dl*!gv!FUkW%vCs?wt zeHqKwhx=|WPFkM{mLfbt>$+4{y|c`(nu!17&bB`J4CjJh*cZs}=o($z{iW|DXY*5W zcELmTCU7_~#bW zi@gTzjI5{jQvc*0I4DKjZ_BWfw5gWPDlnsb+}yxXn#9YlQ7$bb+_k$%iEYl8yKBz0re{X)`+El>FtkU-he3Tpz- zI)>h)96p1leeSCs|Vk{Iaep>oySq^^F@ z)#g#s3A|uPC@~W7$I=SGiwfsGGBQb5_Zt}kRxErUT*?df2WAjjMW})}m38&%!BfjA zNjJ9_rSv;zZSK}o2~16W;t?c}5<&bnHC5)#2*Nzx$#WG)Ki%n7pZoFS>9c3>PUYEY zTF4LhtRF$5!ulbXW=MBd((!9Z|1osdCPFpeYrM;jSCeSFFTls{qB_dkmJ zFUkH!*3lC24uu1wxBnE%-HJLPkTMg`%ii5C(;(r}<9(e?i|CG-2ZX%_G*Q>%jKQH6 zH=X&y8yr#=)8$MC*lAuTCoQcA7TWL@b76{&_{$qt%FDf>Ttz{*`spe66WA`st{s?p za8Mc*ruFIw@)?LJJ&qk4#^yy{EiGj~<{_=A3Yyo^fiiOx(Ieu(^+>Y?Bm{ACrQ)p* zV9yoIBR!&I`A6$~Ua-`y!59LTZ3AH10;6LoEdV^f>*$*e0s`P37+u+Q{%Tj&8;S6j zy}g`u&jCI||6)mmSM2<1zt~|Q;P?T+-y@c0wVmS3kHt1^qT!&i`V3kn4xJ;2D`KvLhmNqGR6I3gX^M>TX_3acy{mfn+_`Stq-k@z8@ zFJ>oZauTX9Yh)bF7-r0_D>#EVqIK#I>Rd>WZ(+TG{J>={`O>9Jw{C4?G8QG|f|LwV zA!ci&-VOa(G3yDBH%xx7MX;x|_vz?bJiUoo^#>}Hq@;$)-Y6^EoKhNRU6VG62flso<kCXPmT z<||fd4RT~Z>*+~JvciGk@(eulJelIVca=3XL`?iOR%U4L9>H?>%j`KdTQnOG`8*b5 zWME0e^0mFkBwHb>6D5K~t~$U-4n}G^)(XbQ{f{NLZ@;+8^yenww{ql+>okhcQD0fv z)0|9)+HyEJcGi4)>({In@;&$WP+CUt4V~rB2+Rpg3pBdMA_tY@S{X(M{9$^5G%u!d z=X)~wr+bTZ(`D-_jvw5wYy9GjG(H`24bi*jZ%JNn=naQjfKX!8X?DobkpY_OH-|4- zoce>1r<1;nDVIK619WTmpeG#IM3-6fj^}Rp()R1j7Yq1YS@VzBlYMrvN<=A@?K^el z3oknd^)<_6O6{8=7F^7ioRri;oUdslDE1RuumfA%fyd|9F~2Q%?@DJM|1FZaoEVqF zp?p?T>gMrZ6lsbZ<{EY``3J-34Kt~xuT6mS$g-o|vW2Dj&w@bB7YKyr16*FhYY#<~ z^2r&OF`0oWj@U8(s5}0n+M#2R4YlpDwf}7=pDQQIQ<^;Tu;>)A-Mg`iG2gZC_~L5I z0fsY{lF}Kkj=iA|py+Imbk4-cn$G;*iPt&)N>WLr*9-uH)Ql#3?LSsCgfccEmGm6i zHBmx@h4;>+lH8}^oZEd9Y-%n)i*D=+V;U_?D*5;cw6yk4LPWo8dxTPKa_U_;%JQ_H z)m(`JJ6V$>XpX1O+t=49!a#~go6C{OESV)KeuTS-MNBr_=Gl;iS#@%97s!JJ*4^QK z9Jx}H9xX&&6pE9tV~^TVH{?F zbcWHJmNo`Gy|`7;hhOA|x5CkMx1i=10YhqU=hhM0VP2J!M2iAw@(6)xD&q?EkBaLx?i=|=bLlrx{8e6j z)12lg%5h!7!7~zL)f-QPyI!z0Wgg@ZyoL%zhyS5|1k0`-H)kcsc&T4)#U&-$>g6(n zr1;P&$&{?G-<6l{7fJY#gOyL-r&F>%s26yOQ%3VHYuc_I306E#!yYem$ppey=ce3) zbko*aWUFrGv`>3ZO{An4NMHVdiqhD`1cRUB%IpqzNsoL|hWXUYo)FEQyYH!#XdH=wasddc-{t;ppak@~=KxiXPT!H_Zj0R1QYcRi# zF5}uOM0MnO*%O?t|RjRRvoa8a1*FJ^7j52DX1_42s@2*sj&=oHxE z=tCM9By}>>pwvavaeh;{tDRL%R#w(*)VEi+L_S@rh?e*1S(lUWsX>)g`f>BgB!4q9 z&&+ggMh5*^_@;-?mwS96C*#1J#uJ7~<#Dv$7&S`2=h1-mt zO_|*Vx1n1&AuLP-HKqSB(GAes902g(xQcXn5H=YsNtqeyWY?eczz3h83~Y5G3}4C4zHV2oy`&sCHTVO(xX$kG*gKp` zZ`<0UzwCh8P$joC{|lPwExTMzF>qdry-mICyPSOu(`AsS-|sy7#_BwL5*+H=SWvc8 ztsIz?R2i5uLA9W}>R%Tf-zsJdvbiO#HJv1v9d~p%-?fSS*L6n)1~^9P1pSq_qfqi! z)DGF`y=|w@S;YeAiRjl?-PhqbCi5G`z01Xm{awO~k4uGLTbN=nBE}Nmr(4yxOL=|Gu_222%etq_+p2CqcTvr!mth~CgLk~Cc~kT^Ac{jhJscd; zx`8acyw5coMS~3u4F@e#S17iA>yf9hlbi>|&5Getdg##R!+u_Dg493{yo9e)*>3C} za@}>7q3BJCvA`GRZxl`Ql4Ibh`{J_S(&A0EzWHzeIXCCNAWrTjVuSkA8_p{wC2hTR;QT)x8oUf=O2cBB(F7q~= z8kT-F^y80qVBki~4bIG0B(7h-zWpaz&ohR+Ny+7`*q_b@)y2K7?`?xV)7cth^t8Md zZPGCI7n<5j#b!fOs3p5U57+Xr+kJS-?X3o(T2q`^!cFZ8ddv^r;^l&xVULv51nkrY zo`d#4z^Y?d;%tvvxb@@>RvS{F>WtiA%%l?Ms1Zo|Yy1wi!-``)lyq#uJ zQzN`X3|ApPP(wlw4!s$fOSbdK4$=74w{GRTT=Q^SAoE!!n)h1g~J?G_3hLmt=3D}UCA9@wX}P2{NVoZf!9%-T7buiwjh zx%Zw88DA3oz_BmzY>pU9Grd$JgP;BG4NL2yU1b^ zNTHLH6Brn$PoKu&SZ?NY{P=AQC_o~P&eNo=fNYCfNZ-!JC6aMD^vP^TW{- ztkh7Q2Zpaz2P41o38>pat^f-|ZO4usqM}re&%#r;DJj(es6v|q^E)v4h`}5VFKJuz zb{^q-atewb3;+lGS9N%N>g$MS#a5}f_o+7C-oas0o+fEm<}|)CQGV@cRDL)B9=b=B zJ$wfroE#k)NzKfBbdCg<{p@LJDH)mDw{KsDLbuk=%#2F*`K?-B3hMHF`g7=C%%yw% z0r*g9%NgJD_4M5B(Y>KTY46^5FJJP_^722rNqXOS>C}L%j zQ~AH(s?c8yr-D>VJ3FRHR2@2(H!gjqpfS1wEbQ)o?q$6nPY2{up7=>T8qOHmP{nWJ-=)0c~w7k5$!w>@)`_YQg42AIy zj5WF~63bja;t$5pW81hIkN+|D2~I=PIc_{ai`>y+u|A&k=Rts#a`c3{V=8AAa3ORf zET%0p3rJmJUlZTSa?D*AJ*`7W4gl+>@b;YE(*LZQke6e&|th{6GD+{{Y<2o$U|PE1grVo@56VvJ!FbA zdU7@?$zrYb!GpbDf)YYS^806&ak+Sj5sSNjBHYOY-JikM4V0b>2Cl9?a4|3MnJ@Dv z|CwMhN79N&@?LYg3 z)I{lR027ypP^<`Pt{c6QGi8&qTUINWTyhLwnLf>;yn=Lb*+_C1KZd(~u|XY*dUpuH@tZ#FXZwqFA&s zL@qBa;qRScg{MYEQFO5c!Jrlm0qydH1V%>2NLlf;Cl8Fec>cUlX2Ys^V%30XMKb4v zycZn;EP>W2Vg!9D4|`Eb`m@MVoQyC!Rf?XFIOKfpFi3l@uAr$@m0OlZ=TLh0Q&Z^m z86#4RTqe=e9H8Cj{CLW{_786j6R9#agVQb)^&L7rfN!xS@j{kd&Fc%kTjJ%i)PGy^ zO|GPrOdH(E0twNxNa%Rg7IYtv%nCh4U_-A$m;!*gKWYyMaG zKvec?U_)xA`Guu=bmcDy>85Bta|kMH`W(rPzLIoi!N}(cQ_-u<6sNle2CXHa3`a*= zqKfJsNh5T-%&tG&Zg+W>&VbtqMpQ$D^j?%TNJT09BD@B$p{r%1#I#)N`5|rT6C1j2 zy(N>GNS1gI2lrrV5@m%vdD+i#NV68LeNNhuzA4cJM|<3u4rL*3leVpG`K6s%n6p7& zJn?(W;U#~@6p`XHT8Ew}er1+@YhQ@Rqw{a2*&R-Cq`s-NKZlp+#@qOOOdfTq~K zfBQCMCmd~EMe~@KJ@A+lpFZc8WXdyOq8p=7Kh&?iN7PqSYXn&Z`)tF84N6Df)NGpG z`bqhn?R)|4@g3<0mjz(NN2X`$eB9gX2N$D(uWJaWjC-&pP6P!Fp$Uo-Bzie1)8Mc= z$Kj~?!8FRnZq+nX2Nb3XOEN2&IJ@5fimk@s+5rBR7UkC9$92@<6U`(eB#5eX}K8ltXhC`1H$4+Y6!P^-yUE znIGBqhBQ&{clhNwO_g9_lk2gfZ;aWWQldY>96cV2m`0-RmTL`>-131QJB8PDW}N*k zyEdVS-zzQLPB%HFzbnhaeV=IKc52ExxjpIBlgapwi4Za}G9d3(6qTFqLPwbM_!(@;`J?XN@94FkU`)7?1q25#orG< ze)NcfI;FaBD0KJq)Kyi9BBAigBl$UIU9t0CJD6L*qgAsmdRB4QF7?p)GNXfJMh-!R z-Mc0KeJl!zCJh94)f)VNKL&~^5`+qT4AnXk>%Si>B=zQHuu{DF&SnsFmqn$np(^j_ z>x0I10R>V11jsbtFv6Mf7)HoYuJe2nL-d<0`u=P56yv`)j0ct|0KT*HwcgK`x8#(7 z_9Itl{dOlVQ$|J7@R(b%VFcxa_6z|_JbZjN5u~APN1}w(ant>Px;_gqw{F*c$L34y zRm7nxuVXk?=vwnj>f0N!_4@0FTOUsobg$sT(4wR`P+0v&(-nXhIE{%(N$ZkpSI=(y zW?iGZ>BMd+%lRT^$_Czr*AVx5dAFZ%hVTxLbC~?-?_X73K8dMUlsEcKrI*Xl<=ba7 z*iJ^5no@rCM$}Ya2GBPhZ7CjjdZDq&hrO}w*X-;j%4*i)RLc=Byyf1*Cdip0YES#R zKo*UNR2aa~UF(ei@2dz-Fbq*jls#O@%PzNEEtgt1!o0f7NqO=dOr51DkkB#SvDu^d z>bH*{bq;%=5fw-|GW&v2AQ;MThys^};sp5y=?oN9(2lY3D?NGmaP!u!DV@2%X7u#d z1BZmY$g|eg0(6=I>4VgmU@&-$ zfLTLAM<0zvF~1 z>;G?dDP$x}AW+@(Fwqp^=cm}_zp$S=rYBnP+?k!AnP9Lq!i?D3IiUY!Kp($3kX9 z#<8yox)xT)T)B5b3y zBAaqOU<2=ST>7qA#zZfEAFJRe-|8=HHF#*nfN#2L4}{{k z;{b(9HYvnWVII@{S7qe+*nlr^io*H98|IPnw_F(Omd^-mQxw{}ZKyZiRem3bxYM)3 zAMD-akV$@>nu=EC>E;GT&@labgzR_Q;I`te`kvi+^bR{AmK>lM0Yd&a=6*|{2hecG zJUjrs)Pb=ekiuXr2qC3Q*itrz;=-HAO?(w|`i7@8R$u|N)3dW<_%2u)0Ue&+BSywRuo#Hge(Sl)paz16gLeu-uP3LcJ6Hob3(tl(s>93GHJ{){>j^?+93M{(t9;zo>3hNiI;9a?{6eugCnN1QQ>q z6yTC~$LRyu;&g0a zpMd@WvF(eL+Rk5zmENU~Kia5fN8o8cvn+`oL;2V+R0d=@{`U zD4=l$&Z#?aIKv=>U3&GYn^80_JLF=LJNK?(RGD;d7lZLXYuEs67N8DC zod3A&WMDeAoqf+S6H`}={1pv4R=h4O?qU)in7hZRd${Cqyk16of6p}Vl{~&0be0Hv z3oF*Q>rm0fS#iO<0JO63HxSve#jH2y6>3cdno1WHUN-|Ej@}{uBTXIno(`;j;xIHB zP6;Y5UIQV;MuuGdoos!+o#m)Hahc)DMjr`XSZ>bo5#}Tn8otDrTNz`Sx$CJ% z(%mX}kR8JZg(#>&K<_G^Q_YMg?os?`YnpcA7l9Zh^va?9Q18&tyoCu9NFZR!V3aFb zP-Y%uRA@`%5f8sVV-)fGx1iVrGAW#I4E6kKnm>szFB-?*k5zzxyt^5(~z{ppEwNS4;eBVr&@@}-JAb#_Ue$UtV1*uY=uW(c5i72Q0 z7ee4SXqq7aCEH{p_8eu+*Ks2an(#&DSO0wR)%3lbXzkZ=TAB%CQMv0(qV2yYv_6oB z(bS{X?m-+O+`G`x0(BvuD&l*8zb}mB+Sv?p1=z`YMd)C4Jww1Uk>Rjl;YdC?MdT40 z8X4{Vht~p7p%DQ@P#l!3fE_CZtjC13{E)#vH0ZaWRfD`^K->_HgLDb#vJY8B`1$Lh zu7Zn;rQ_jRGnSIOE-!j>`Q0xTl;Ia}%l$_ki68T&F7N&O_rMTD+zlnR^9FC=ZMZ{# zxKjMMQl~JBg^U3mrs{P7+tXjaMgZr4lqa>u!+0Vy3KpFfq6>xE0XhQfJp^Ar69TrO z=0PHPxV!qY0z+n4O;k#g009EwNNb2gR9sAMt_741u#BXJr>3R=;^gJ$N92?J`(5sf z4WV}x<0$|6;UV!UgK&Sq*s_#0f8k?LP*ixq+pq7KE0APs;!MNoOVs&Hb)G31;{r9$Q2T`+kQ1%GAG@?Qm z6r2Mtd}*aCSZ6)Iptx^oBvSHOBlJGD0=^quQE=vi*Q4b4Qu}RQzYJT7Tx|UB#*b*O)FAXpiu>OitoQB*7ePKx`0kS+BV_!u4itEQS1cSqVB8XAJ;5qP3tC&w%# zL@N6NtVeBpCBMVB!qN7BEc!}p4@G@gj#FjC)m>KkW<_N2k)Xu~&uU0j4hn(`k7II| z4?>PEZviLLY?V%S$-EzmEJaWPNRo{`1S3)s7>PfG!JW}P7M4IVh*6Y8Nbu0ZL8Ykv zZ#}pK&BW;csUQ?g&(D+S+t4NyS9cQamxu$6e6Y@);63041PL7hp8aFyJCra|&0_=L z5&IASYXmylkfYKAXfz}{042^>0YV|o)7zMHUJ+Vlh}rogOlE@KT+>_bKx9{ZVr@NK z=%N)!kusPAW^Ene&p>s4`5Hq7N^hcITVa%y$2K z!A^1u&~-_vFC#;g4&m=R`IdQOCFO~HPG zN?>KbrI<@BV~0|ThO#mWhZnRIjhW;KX%<9GJdHLw*x4Z!cuMLX2H=$jdj;ACWV5pj zcwnxfvw(pCN&(C^|E{n1-V2M&d^HH08MXr^3X~w3GW-GpdX2gz>zH(eUz^VR4SjxK zf^7ngZp2$Z*t(p7Y=u7$RS$bYaj-Eg?L%;+SX^F~ox_&}PbEkxAm;@qy2dFX0SbhL z*y&J63(3&LM0IZuu!AJ7DC)WGJv(CH2jv#j+j z=fL1#@$3W1G_Veq!ky)0knZMeyhC{=1l|DhD)fql3&Fr1-T+M)AnYl~I-9uUR=P(h>%6#nK?{whLr)q1-i)GvCdB9TY*Hy4PYoG z|9!-l_n*54Y4IZ#$y+}aIMRUFlmRO6XLq9oRuKhM%>ufXroEatD6i(;S%k#J4Z>mJ zP=fgq90_r!6jq!l1-{r75)^cd)%k&Bc^o<}vPx-WAbWiUqs5Q7z8=J7nV6_j_;U6} z;;<`1kIZirh6cZes8mOQNe7nrrl3$IU$9?RX6hhhTB}HOSA2$VS67Ov$WF^D6m;&o zC5pdH4*($P^Kd-s?ccxq*Y@F!Q~c-E5-k9VgDf+A_Rr724&Jka36*=MEwEU@O$D4X zlowg}-#Mun4*(OoRmoEw6j5aq&2+t{7g)59IoVO9mjC^l)v0Hm zN%290CSrB#$Y_W|22zHQYo7Efq&$L1@!#{gHd3}$QbJgz`U-IpL}J2GK03-PENn>Y z4vQX|j(-juKG!!AX-$3V?Ad7{amw_Obg*eyTIPxBQZA=wW+G~3jbZ+vy}FSwuXnA; zQp+?f)Zl2d8OEdvox2V&dJ#A#TT}StXQJX1dHiQ^86XH1WKk-l&M(3}AW;T(Z-sv@ z#RSFv+I$^2%yXiwKwzc$zrJ=b9S7Q1yR#yI(A>-}<67|hB4p{vCylqq7lIqXhePTL zuXaxu5=4Q*53JV1gM)c5$hQFbA+jPo7{Bg7RGbVK0LA3ou&fT-V2e+a>Fn zt4ADmV&QqVo^xM23O;-CbEE+XXLXV~Jzr_qCVMh3I@8L@9+!ZfboDFw)QL5MLt zF|oFPy7LKIEf{OkHoSo4=MY|?3+v`VeMilLq9bfaZG&&E)zt@qfCt6Im#!{IJiJRv zx*k}lIZU?HU*ncNvedbA6R$uI-YDSo1H6%{*w=(liz5tkzJCw>ZhTDZN7KEAKGsO$ z^Fd>#y{|ZDz{X5^6@X5f9Kg+jqQ@CvvvXvS5D-0QK|DtbKp~JT0IyZIHMi)4+|@)p*)wRb6x};{auKX3uT56&eYPf& zA5r8$0)prcOHJ>kua@!Yad?h+*hk!iY3ak2(Zk1xFe^oELlly4oM2<%>)e2>zO!=^ zj&gfc=w85HJ`GPuL}U}XHK&+MgRic?68;kwV^3B9j&A2zgD(1)a3P-_%8gA0zyf08B6Y~HQ zk@nv=U=;}xSM;94Soagy;P5_pQ@^FQiw4|3k~~8g*b%yTSu+==wdBmG=OL0WJ@`RSW00W5bx9v+a4gdnH}pya0e zXPvsgJjq$6+Ce`cihgHIHY{f^ElwB(CbP@~QTY;mK=1ya%6y+zh-G;HUorFF*A*04 z^(C{y+=YpW>0KBE7TJ{Qusv$_t^qkX#P|GXB(O}o_!XepT3U&HtGRiN;W}(s)SO69&UL5ejmVMm?A?KblVTObqiF0whpa0 zq?JztGhlw+{2xP!jS*;5g0j-U%}E#RraHPIDor*mIENOTwuCh>TWKptMks`p(UTGh z4bN}8rJ5#%Ch+AsxI6?DgcUOyj?Ta=a8J^Vfsr1V%_cFc7u^OHBh6+J^%2oOGh{Cg zILk2?c7b}jQHlCrO_B8~_;Sc5UJ**~p48OkUMyet?C;*hlP#Xl^h+k81i#MZViRmY1&J?!754YOFCwkV7*uaek1{heF@=Ga5HF4Z`7%DQfTv_U zGC{p`d4p#%e&+rfLpZja@h(c37o35458%(}i$3}fST7r3Jp+UT1cJt|>0NSPOpH>w z+N2R_<1ViaOLOXPXoT>eX%9Mszd?H@B)t`>&~Ap0z?x#Oi*``Jut;!6dr2mz zk@e<}zv|q<#sGGbX86Y4leyxY=Vj#%P1mJWr<>KAn`~edhC@jS;^>i~VQvsu!Y2TT zPJ4L_Tql9jG70kUd0;mJ>-WSCz_(+yI4tL1Ocv-iRZ`euoP1h)&i!N; zx_LGHyabd31*Ui%2CS_oSs>Dd=?T;b@h$e7lFdg?*d2x_kEN9%TLXOO8l({AVH+dO zm#r#Ux*+ZgDK9`mE=2Nx?F4(mlTOQ;NgzGx-tbC(Athh2mq(^lSEI3`At0(TmDVjF zkpBEA_6;pzcSKbe%&nx^;O(W=j1YJ-Sz!*yjgUGRfplyc*k-6g<+`%9l?8oMLc+rD z-&j#L5AqCyLqeb%$8wC25Cvyi5N7Os>Y9?GoGE87Q9c>nt%Up_ZYUPF%GuGEM#?@U z$dccA@Ii?}N2jUE@|R4eWhNJ=2MglH1SX*G-|F{Q0@O~8w~tU=Sl)wY32YF$W{7FD zrP&SWgFbCy&BJChgEfyh@0zy3_#QZbE@MA_q?q2Q<|Dq{*TGcgT8!+Me({rGG<5i_ z`|fjaU4SDG_V3-OcTVb@#0Kx1~twnn-dGJX*>XkwOC4 z<;E9I!-5&6y$Pb0xy_=>%Ns_sU-PUD4wr~_3=QX=fIo{&DkKcXr=&dlGtKWuhusL{ z_OTy~q3}(Ujs6e8iZjzcE`uP}a)(KZltqM?8mw}FNm8iTMPsU;hL4%Jt{kYtcI{d? zX(U(pIz5ty*H(wh;_W&(^~2fBB1ea|+td`Q5ihGxMn%#aj8^JeEJ5{E^o4R90Y->S z#NyKYJY+Uk6KBN72YNqxB+a6j$A0&dry^N-iH9SR<|Y~U%RYK+Ov>m%IErB|0T*B@ zV&z0tMFrAdaPaUZ{&}rUd%K${TkpTx&uoZ`r5m>nKbGy#7|K0qhP2tAKR+4O+d~}; z#x*ChlRZ7f5J2yvi0p)+7+N-`(b~j~WI6>GwbH+~ET+y#C$gOn5eW__|Q08H4Pn#(=#|Jrd^r$!i z4NhIup?MWk#t)jBc27;W2dXF&y$na+(0CyK`zR$CJ-#H%>(1^TFA_*4Z}6?HU^io z=yI7nluo}eLr?(h%5({eRR+RgDKLFNY6Ln?7OpDa4aGbR@%v_dEm0&tQ)tKhsRpiI z)FIu;CGWeB7`Jc(K=1nVmvG?gdtQ1Xul?+5`tw2A2t)|U(@59mZ;`- zuY!PUL&4R8){cKya9sl$XI9ZQs8B&h4ZIY^y7F?w!3UMZvu7B&G*L{mA?L?d2@9%~ zwPDi(vc#gvQy&$y?wT>tc*w{d$n}$`RMCinGZ09LJ3oytFrox8WO_Tx(w%Z4Y#4sC4E7_!KEx)Ro10B&RXT_ zB;-u8>^6h=v}B{~dQfmgEu*XD50bAbALNOzB8d>B=!8YhFaE?9d4c)13dgacD@rXph70xi6C+r=Q_nEptj1UL+Xh>9IND$@1Ys2}xd3zzzGZgs_vWvD zC7N0YNwnTwj9Mbid8JB}9W4-0ZcS%zZ732q*V{huBq?JOb@7Kpq}QMLtnUfnmf?G@ zw;szSUDj}d2%Z3vH&EUlkkADF5@gGE)fG*P?2FK03(%nnadSGAOMIt8acH=K9=~0A zHv&`J+4H@YYUZC`L3gPPrekVd-gvKIy82|IRGrj3Roa%byK)5L`EvYq8XKoLBq+en z4ZiX=UNEn$A%~+PxgvUvP_qy}X61N0)iXs>xb+c3IIn1d#AxBvgV@pWuQv8{We#-q zMe8IHYt-w5enPPs2N=A(c?SzjWhpuS#Llw}Q?2T5d!Yw?)~rwc&vf*37|N~avCA7y z4z_5%!(wxrCRJn|abr-VWIHvx#W%} z5r_J{KS_o;Y*%ixxsJvRpFERAj)ty7G7{+{lyg$^zKcB5@;Qn<=y+~t>SR~w@~7-z z6=&m*aqTBBE4=M~DhbX?`l}NeJp&eY&eW9T-r?IU*IKsWOyWL8FfoUnaka9gj{i%qsX|d! zlTn=C4AJ74+1*GsXxakpa~UXDx^ z+Gj)#;Hc(}jwfZW=(2Sq09sZsG&#|X-Etj~c&}O>0x;}r%-VKkDqQN+ln%ffP80iG`++iOg z;b(gbA|W9`m3H(u;M&-L7tD$P999S5NQVU$KFLvbBrpw*e@_eo4AtFT1Qy{Emj_AW z)mMNwLoP0W#1KHgIL`8;5;z{D68?UG@K#n-05#Md$SF7fK-)bJtuOf5=z+2iY-fPy zyaE<7cZR6MK`hWQ9B}#DEcmB86tX}zff+9OM{Jy*Ac@Dc9R$V;Fikh-=Wjfo{^0lt zls-ywE%k`lOYFNmUBFvr?E3c25n4)U6K6Cp@7Uzib8u8b=q{Z0cl7eW6yMIt352p? zt$tu#06f#&N$I;uX=xFQ%YXm2f+hGQFK|eb-3>|vk%6iVzU@G+i@fG{dO2dYMSXp8 z3sdxJ6=nz^#Fg9iKiSFEID$~XTuv1=HK40@BgQqfV){^~H?0#q@=ShK9fY6%-aO#TsL2_YCu>eI=p!QWXif zYBpK#cskF8Jqlu702pgNbtoesRnVAgy3M=2MLpOk&gr1CI2VKE8glkIGZ=DS*H2w( zBR0(je*8G>#FNl@m`^S7@V#ul=3g)g*xu%RixLD}CgA^s<6vdx+XF$G-EJG|y?f$}fOFs$)rxL^!2d1Gx2M$y0n0j9gUippp!qddbfVH5)w7rrjX zG>y-@f_uJjv2KYW%o0AaH^{<hLChoEn{Ci2Xqw>*xbH!i2!k< z$el-Ud0~hKX$149E_f(PdHsuUZA+$dh3}Izd+#CMVBy){HLJ^W&gU{&#H!a3XK5$A7f~ezAP+$i9XWqpwEjd;vhk zuBha?j$r-Laq+2C(Dw!dI`@S=)^ub#2$(S}2jNPB@q{iUd43sy6fWIvkn*#SK`tDj zghZ>);pU@8x3CyuSm8V{PW(Zlb50}>I|zfu%b&cELK2d=u2*RUT9a7nr1W%jUvLP_ zd!jOS|=1OQ}?dS3?45K=>RLFwxpy3gZ|E$KD*u4?z9m4U{#Qas-c=0mnd!7Bm@H zcS2TaFQ_k#K`3VWH75MNm7QHFwR)_&VaT>i@#G^&#bH9iZ9)d&2CN29q7KF3kZnCb zKdRJf2P+oZKEm4ayw9@rN4+F*Z+6#=z-j6Zuj`_0GGnfh*pQ_6u&tg8Hx${WLoKLv z?+^j#NxH@c9k*?kq<21IK}ZS$=yO{Ey_)KHsoKMbBz@_CYjr|E(G4NecTYCQN+;#8 zGzSH4993q37CCf^7L#QM_NJ0nfIOk)fKh{H+2N3veySKOipX}32Va(8RY_0fht8hm z@lWo5WzGT9`CN>Eo%wdyvS7@>VbY+%;_KoCN_7Y$X##E~2(9H56drrQ;vji~;efmx zW+pvJ$gX1{AWtJTvyLkTvJ;?VgbZo}L0|gAY?w|q!kq?+s@(VtFODD-uY9xL8o?wT>7_ zz3*uz0wv(l&JKXd5N-k>&2aY1>5)1+?WGi3F?D%m{7}vLr6nSAa(4ZSl_wS!yR#1d zN`+c17XX~rtC|tS#OPW_31oelMh_k$8PbIyJwVz+;f9^Ad93hl15y+CVVnakn z$%(gsI3r`2M7<3h)isDWB>2qcw)! z`vrPOiCyrd+5HLJ${tfePENI}wJrRau7a6rM0__qHNF;Z=aX?bVH)?r!5@-8|4!GJ z!%oqA^Jw!#*0d(3nC8=ogF0}kWF?6j?4pHT=Zo&?eKLqrWeQ=E+ny})X> zrNm^CYZD31!urwrQcn-?u>Jy&rGs9Zu!3BmPbg>rW!f5*Z%O{yCHcWigkn(ug!f)`>|HZ_N8d zf-z6gN5Wo2i+wBc0Vc8wp;}&Ul5+GdLx}7VzXK5)3%K0aA@6Gp+mWtsr@E+2u-=>_ z6?ku)Z2?h};8-Cr34g14l3Zx(}o!DoLJM>eJUk6$ta%>E`$Qb)$tX`kAhzJC8ctXpUB$x)9TF9ot2h2d0koJ<*l#fe$*^;CL|?=?7%U8ssKk!uM(d<(HQGRKF~>o ziDPkO&pSJ|L(;neh=C5 zM?|uw;dfqEHz(cs7C?qRL20&7tuGF#m>i5o0%?9 zh?Ygc%?SLOs~NMP{3|Ncx{*y1Wz)L2xVUNU&wi(FPy=yQfcnjPN*zD6-3^jcBGG0f zfCjFoN@EKb)bdMct2bv3Ag3&>o4yBf`-~8lcgHL!{oN;g?z)-!WLw*OfZ5=-X^?@M z{i0sLDbS)J?63s8&OKVP-Jzrv*C+jFoskv z7YEF`9wrJV0YWB-jO7`APyG+n-%JW<_3>$eNDg7#zRn{fOZDGK`QtEvt_49-+OO!8`K zYay&|3H0%=D1?J^@TsW2Q;~srI`pk?Pr(<(`T+|M56{q3DSq&Q;v9=DQBRl$+SF4Y z-KaN0u9iAF#5}wY9-M=`x$gb@C8{_=6blncD#xc!K^L_W-Ck$1Z^#RsT|BXqiHD^n zofkSG&co4nWULHi-~3u+`$=B$d7zb{wu9$^czIKnY&qvzu}2jAu~M|aRKQTW0HcTm zaR!Fa2Ou}yfcgC5la3788zViG!Lsa?KdZ*dcRs@E@NT?Nj$Q)s4{Y5U&dY>*2%pRg z{vK#&g&^w>^sR6E0+XOxwG^ProP*x@=Ghf$^Q{m=yhJz%M3| zWs_dcj{9n~J)m`eN1{Gr(9HjGC+B3Y^p6# z5?7)vuIRt6Z>%3n1o#~jwTTB@?MFz+MeN-Lj2NzIzWdej^}*?W!1uX4`c~9k<4I|B z#xzMTIrBao`93>iW#O}U3?_F~Ou=*(t`)Q-A*N5{s?_P!0j4EAA1j~fuhTE!nLI`J z?-x$dV`h2~Rcc(xU!=%oiB9VirX-=HN+%(3`Z(-W!*ug-0`bRTdH&^srZ4g4Te7e2 zq;K&TDXY$en?~*>>X?2v=Ag9sVqea#O89|HS%;LF#Kgl`imjmuha=TP_!fo|VPbS> z9Y@2aL0lMrINE5oMQt$mFAk^AK_@o0@7j6XXUcx4@)_9By7$Ms`Dl<9w!a}}>Z|cG z;Cb5jx`5x+JZJ8MS5!y$<=)u|csamwK&7U^~eau02{6A825cnBZi z|0qyl?1?8~^1IK~Mj~vJ(?8v2Iq=${akQ|?1O4K+?~{I^uO9*fNaQz}b1l3D{@SZ) zYm+v{KE}n&KS+9Jtn^*3%4f^aGEgt}8C8PT`d_uH*^j2ft#@Bknp%ue=!mH&YS7^; zM}_^Yp}@me_Q<+Qd01D+&-dEAA_A?cJZoAvKxByFv7{Ye8!`4gZ3es3Owz0wg zog~?5*-<0KZ&qXblvOO{3Dcx*J1BHM5}a`Lpnsd2h!S`6(+Oz-4!Si`r)~UL(k;v> zP5B(r(0e0l1;6+3YZYVpg#K~~bR0fJ#aP_OfA76^8l6;jKov3+_&TgUwb}WJt!*q$ zN2);4tKeo7MjWkt@JPMYN}$x?a7HRPOka2d-Q?ALIA77SNkYF*;Yld>habcmUl(#F z%DX!K6+^7F-_g1yHZZ9BOcHzvsx1;ob42HPfvj zX9@Bb3EPT^N@eG=qRP(oBtFMRB`^HMq12{}&$V+uj))Ap|9lNT+(XAd4rQxQX=F!XBXduA8HN_rjib0NCG#AKI2&!l3R8`Zu%K|w;KO7=fQBK-mQc7=Dnq1F zGq!+ES*KmBa(O6+GIJgiYDX$vcI2k#&CR(aAtGWD^htFVts-0+x?Q<@e*BF12`n0C zn#^j%iZ3gL#TZ36+;|z4Z{l3kMC0p>(A&n>GMb#*3oBu*|Jf~&O*<%9r%?Xcg&F_J zk(xnzMT|yvE9F!Dr>=5}+|PJg&V<7t#82&5zH)-1=HUfs+Zeq(Fc^Ne5- z8&&J)q`}8`#YW?*cZp)vgdL}t(HX1?3X>R4BMZJ&65GD*yG%_+X%9z&*%|5y17+fv zWL1-T-*h~)37btMsY=t$Qu-iwEb|SF4cTWVn;5|_4Y*tc85cMe(X4s&`8=hN3UqFG zEFPo8=DH442`M4DeSDdfFFCRr=x&*+nw!Wq@2#>Z6{Xg7KpsD&?dH$!_m`>o&L|5K zoxaejCW@>*%js*^ugi~u13Osnmk@G3W4FK0DtOX97ig6e=-xzR%=*I>oyCpytD$co zN>+YmQdUyDdmhm%rB|FsmCpuJN#zp*-H&3UYt(MMOQF-x+R_dtHE^13735XEt5>Vs z(Tn_XlV9FO&0J!KB?B4z#j<%tw|SWT5P#(LX&^r??)ClzUY7I41cm<<`=v=x7va;k z$AN!$ZytW}UX&2OI;0lA#(p5H|IoZz-6Z1Ikt2?AZJS7Rnk#!j#d0 zv-1*@RerPA7=%=o-3d)S^v_%{ec2mcBJB^@sdnbzz6^>htGr3w_2}_3uYI>~UPe5f z?TgdPA^f!CZ~UphNmVcIk?7n$uc{>+taeQZ(~hI^bgw@@d49b%w1hk`M(=c@nN%sF zF?xt0ayD_w`q+jdeA%I0qOCH{xrbux^!WacY^!+t{yQR4-Zkdt=d^nlNV@y3FUp)m z1N`?`Zprcn@y8yW9}0e$t5Tl~nEr+QCQOm%%FNv8`k+Sc6C0z_VhNl%@eMS@jWWfL{g=nBNrsJJRBtqtWC9>kJd)0Sg`#1 zpI@m`yU$5@V}D@uA8atj>|D04Aa&kp>7V&yj#oM1=82S!CB2>HgUtSv)@E{-iBECo z)!SC_*dv0~=CFcWAsKyQzG|=Ww(Y5cDT!w#`S=3aktTgiT#PQQ_|U%FOGH{?rgdv_ zn=B>e30lq}sWQ`v@z{SPCMG#;T%QGZ9Kw>rKy(@W9G7?;X0R zw=H?2XQ|FNjTIYRoqkqf<@YGa(J?;IG>6F9IPAE%bV>SAU7L@NQW=Du=$Joi-)gg0 z_N9H-MCHTyE|cTyW3Sr=-~3gsyO7GUZin6WHs>$=MWHd+WKap~6%8NdtCl znY2rqm2ULs>pW@_jde%n%b#Z_vx!c>5`4w(TzYXrwie>rzw*K+r{K7~Z6!KmEpUZ^ z9gR(w<1UGj`CF}h0wdLQ(X&oT>%k4*M}G1v7bxUP!rCQ3Ljt4lYdiSqqJ=- z#NUu`Xnd{w40GIxb=k3o9Su2AYHn3jAvRY?wn;6PtNWQc?(e_DLy=rn-aGcDr|PlD zRhPqAJ;fJKw$x7tv24xzON2Hfwojvel7 ztWK=xd3@l#UCy-i@M|=|`K>QxYT*l{h<^u#vp+P=b!UG^eV?t?9}qFxNagcoy3=Ofek+|HC)fRxZAoQS z67~}Tryv*I(s-P|uH*E*_T}Z0ozMR;U#n&wD0e=8i4}l-RK))!E_vH#HRf(KHSfcp z0TU;~FU_#Udaeb+scuSZsK~VZlAr)xKhf>E^K6t@?O3y<$tl?A+35 ze8pDGw2W%Y)v}~(hL)7381l32F4gdrR<(Dg`wY<2qOdb4`gAGFS;GR}A+=;l#8mxEpug+5& zwm*_mTcpl%{wPO7CeLaP`W0wd^|IUNd(UfbHYADFi~dJmp4-jLSJA$xo+5M6FTUKX zRcFXKJ<8}euvT>DTb>?#hrK4e`11OEku~bXvpBk_mm{-!24{s>&=Jq@9ZLhV8;9Lj z9pQ<5;-C#Zo*l{0uIwG$i@GQNOL}9DC%6x|R`h1;pY&Be9mWi4@fp0~nDoiad5)LF zJlUu7yYo+8DSah6+r}qs;z_bYKaW)zLm%!ItLmi;ecXCx^7mMXwy$^&$WEtt6)YNb_fJ zW@D_-r#`;&VwcFA|{@5!vl_^d7NF?4s@k|T(qd*wgW;Kp&M&yb3NQWcA zqs7&K^y9I~?}@V04u{LN?w6`_zv{1eujS15-g(TtWRKJ9@fXC6le(*#%^Ca2CBGN_ z^)1~x`kN?_PSocV3EAHT=wG7sYH1bwX|x@|rA`^-S5ZN{u! z(E7DqO8U8jmV@&r(VT`T(PtjhehW%_VOXEdSBF^>jxL5b{MN(rwMP=uH9b9=wgn%E zFmgSMc}`43fdBR__W#S6GM?AJM{k+cgtdlCP5 zwSMKG=J~i;MA5h+bHMCOGi)p_pMQ{wXLqoT|F>D;sR3pqp<1YtNU?rL6xYgZNA@$e z<8V_AXghJJ&Z)mpci>Rzv*Lu! zTW`a|KFU3g@GoP~ug}5!uNwcRcKABEJNyvs`Mq2A&VKI<_5Je~%DzDp)e)u(hf|4@ zhwi3kN3ZwKLNl7hEtsm+)4%?%bynXyJDhyr{X)QI@4r7V*vS9<#pdAh;j{eMiQ>AS z`T=UDX2&o0&WM>xgLmQoN(sAXp-LS#%`JQnzW>&q{)H;(n0@w#yze&=lR}^O3t^0` ziHQ-?>T1L%jb*~sgYt}*yryP@8}oCuIctRJe*IWxzEck03}r~~w_smp>S4s3Ilb_| z+`c*It+~9DT&{G;s_)+ZRG+u-^SpgkgsCfkKd!4juZ12hF4Fs|3-Xcl@(iPQn6CJC z)x?YGai2X_0xG$-`E2GYr72{moi!sL3tWf8hYRoxd4%Qo5f(Ca__s1{!AC|eGsW3q zLSy)|Mv)#3GOvGPy2jT8Jg0qiIQ=+G zXuLcOp5!pd)+JS72%oZNkSuy%BOZzLx^#r6;u?erp>KT)6SC>Ip~dBiQ%BWXDV3%m zvbm)v*X((%W3KzA5>*Phd@4~lGUVVVTNk$lJdf@hTUX6CIbOy1$a+kdY$iO`X3E;N z+q?&_WCT8a?D(hx|JHD*Ze*thDMwdstj1P%9O4hf3KLq@hMzDJQGLHl>dXpW9r6cw z9aYBgItJjUzQNXul*ypQW2usj;d9$Qt(NVlDAkM%L4S|w%JN`A%~-ql9Lp8|&*kkf zA;o>f)jj9HbRAgzCQV^Qm9v*~Sc;iT1hokp=56s630O$GIkkmh*A?ZNE;9jY>ELCC zxMWSv@QIxs?W_9D#_KHj%B_vph(Dr2ngYcJF462gMcP-VL;GPuy}W6=W~fyToK+`& zBn{#c?`sFjdA`?*@c#S-q+0&J zI+xlxTS|{Ro_80%7hp<)X=ldwqRF==|~Ae*ro2{Qp6Y0$c*z|10E}{m8hSQ<&fd0&>JU zTK2z5Aj<3Rr-!@#Nhn$<`s}*h=u{+oyG-Xr9x{*nk;q1Y*0IWmYmbA$`(l@7e-`%D zB?3;=FZR#p{9CS8<`7upM+DY*dgI;c#l|DGoa?Lo9`z-4inrAPwCick)N?NwC)Yma zh+mk9BoE??w_aZCvaek@`(K{^asK^e_?Q3f$-xc+X_m%N@**zPDf1JNKZ>jh)O$Ed zIqAd#>AhUmAxi41p8sHtt*pztFl|UVH^Z2dNW*ET7A_vy?6B(LJ}jy4CUz$E48{N4 zR!~im`7CFxrtYd zf^(f8FUzFjz)Vv4mxp|W@8ff_+9x)o%u!H%3DbKAXO*m|^k*B45~vsTsU9VeR+*Ye zxP`iGc1693H1VsSW|lk8VeG3DcYh>gE^ABqxGt~?M@eS6bw~+aSw`%B%5*2^;rl6F> zP`aWEQ&(G4{j?FO9m9#lqHbhq#EE%!P4R$Sn3WuVNnnJH9RIPP7CS8#(mSyP)4Q9c z;sW2r6`~8zr`sFO)hCxWea)%P ziPhs0zK`2NuzcQvukH87@Alsd?L{Hh&k;mq`1jUz_F5iJ3<;4payXj&qFfecQHmFO z6gW(I_U8lP`%LlfkUxs}ffQ+@jg2oJZM?h_rkC~2twx}WqdzClp3CRTKpZhNCf4r*NsS<8edjK15#9?V2%J^{3$?r6+>azYj}h zGNT=Sdz^I{zP`2b$|UhTP2~Fdiwl#`oC50^cEh(Z3N)?EI^EVLTOorAn_2zdI*Mhs zGr9sOFDYK)ME1PRy}*0B&Y;Mo?NX`{e>?lJu;tQ`l$Gu~*9NO`<>e&r?Wd7C5lN&8 zdKe1J^0FqFjT~X*1T}x8%dS(p4BxUy7F?3jKHm%!B{WN)k71{clMNh+<8h^vTzvZN zD|>=n!nYK3B);074jFuTDtlU1xP_FeNZxgSkQE}EZ#71K z7XLQQxl7y3z01*n-(kI!D`tX@ySOQmh?QX4NSZy)9nbBNf-O_5WQL_ zri%4(-Y7=?w`18?D93eHi3~Y!>h3BNpt8k%JW*bLiXQ)#vGhhxgVZ8%ukN;z3;98) z$`=oU$4cb!EEog$ZFxjvQIkT&M8f1XZLWg)(!^caJa5>^H1Ux!2;)U|t5IU5T&?se z^pQlT!@m>Ie0MK!w_MGXX0YhpNj>mT1|GDtWUSiP(|pm#g8rWs^SRK|T$VLSL$z5> zzcqUAaVE$tgpv?rr1=mNsO;U4IKheh8|9(qbuXEaoH>FnEisD(nPEhD(cNT?VLfX- z-D0sg=F*mOiZDcpa3f1!{!2U&5=Eg))RM3AGV=cXWlV-UE2Gty?6t+DbO-}TI+bS$1A`iN8-Z$ z;_!NnRKaB_pv8atIz{c|HgEq3dM))rPw2GF)&E?-)MUasJNDr_rLe8``)^DSZnX@>h9{Q>guY^OjU2x z)>M`EZ2JMGx?{DTuF6EQYH0H7bS+;g7x?kWUn41q%Fpj|jF(pGZ)aXC!*^AV$;2NKD46?+NI?-W0lVn#MNB_R!z-Cp3`3VbO9P#iGdL^I zmJI`rbpR)}jm`avKZl}JwMEj*Fbh-2w%4XJ_$XpH-d0n#SJ}DK+cNPZn4$@7GNo4Hv6LhfSgMYcxA{W7$ko5GEDCA7 z1By*(!~Chg-gMaqSV#^c*lA*Bp?p0y479^CI*RR5`Z%N^SIX*${ds24;C+jMT4sWj zbO(@ftPZ_^c(R+=$2VDrQ%ZFea;SdDeh@0t3|`JrVpS%|i$psgHE?OFmPaitHzWX8 zPa6A`x^~;np8tyUxoMFgu6*#QUpeGg)Q428xe^2cEldrM<`Bn=L;>4&XU824grctCCuyi~-zB1k-#`DJgl$5j}j;44@7jaP+&*>T5F7oKg^{xKR$FHvk{jE_v*!y82wzk#sq{M_HdCFa=PBYLRF zhNUG^kXEYBc7T9an`!7ctVF;7exP-$-%mg^2eP zTk2u14)Gn$l>dvOv;9oMAonjGc}Y^#E!;kM$>vjHo@O6Sa&nHBLie^GX+nniZJ9ac z;hw~$p=-;BJ9D&4r6R*~C29hOQ#w4xZA;CK_)70T{|Fl%tZ0WvR%m;q&DHiQV9ve5 zP;}KPcT};|Nt6B(T+e@2NWjHPEj!Dcq;PYHZ zm+J!6s&JthFYitya)p?1?f$ihp*0WB5)zXbS$uVE3>faIEvO1O8S4%1XcXrxB6Ica z;nd^io5w-1ma1lxD$en5(4lh>$tjPBhBSwn!pk~v(qhR6vsmny+S$blYvDqGoZ9w3 zfD+N_430<-osqUaM!rXMyDu#x6}2DIR;fKB*COdFS*nflzP^dGE! z5@(-|k`j|XGr4JM*bf@?CLA~7m6C~fYbiAKA`Qay$1qCrdsc~RM^=HW-gV>{72DN8 z*f-HXG8qTBEfixGZ-5DA}MCV zOVX$OEp(?en72GA&icA1WQcCrj6uX9T+Q&jhV79A?XaX8J!~*KVrd1k`pMqT-AG4p2lR}jG~JNODKLP5Ug%* zhI>S|xM{&x;>`6>tm?1s0LZp|>O}1KltxTmgz1k1JLYVInWmz1XsT*$=->i2a~jE- z4eqh4=;88`PBGty2;UUD_Uz~TnZgXWZ2l*2N#X-3sh=by?!G^d}`*JL<-m#y;}RF)Ab>qtQ{}k1JQIoc4jsIltIlK|1dp5-)j>O zX^LyNcMP&$NOlD)M4*M&%ieSqhW7VoiIrbGaxebpzo70v*oC0Uk zUG9x*;8CR8pfI0u3q~W`n6PW2-~1iZ$lXc5KD+N=sKdbT5e|#)@dQBuWA(s;JmJ%j zheT>JMkwNQfY3-~y-?8wqYq z&A&saXphJXYtvC+w~~@aH63S<8EG*Y>j^W*grW&ckhr)(Kb!3tLQ#dPkN}ur?UUSY_wAch(d!qHGj?iX`g^(Ci8-nJu@>w&!iD< zQ(?)MyeHs>m4hJ06lmPAKD)40?CDPPU~-*|UlWYI%(}cvG}5>(H?@d2t%nuC|y?-%AD2aQ20ei?F1)Uh)^?>OoMMz|AkjGKoVc?>-bFe z56(v@VVcM@Axtnt%B7jj86$&iW|fSE9&wt$jpQYB|7Js7zZNyfMX!R~=dS?^QW8uf z*pxa#Cn?X3_?Yk*2k+&@H_x*nK9rxkss0uas5Nw&FoMMt@#?f-Q4fr^^j^!BS_98M z+|?Q##AsFjjGaGLfI24s*q$@90lm2&5xBpDr&;zCO2w*)y%{*P%lW|aBfSw@ z=>YArseE1go%yPYNLJPeFVA4(!jU|`-`)l`Wk2DG74EepO)h`GdJGo@__%|e;YI;2 zQD+?oty(N+0G&<&1{A8$UAcl3mGps5TKIy|kdIN%2tnUKJ+kWZ@Gc`J%sua>-gyUP z-_g)GlD{GvyGO^$q6eBd)Fu!Sx%xIu7LAF-#*oy(dEGMeM;{wZEYfM(1p8>pz@^Fc z$mU@`>b*2PrU(m03`=$?vqsK_b}uIafKrJRl;A2kR>(kl&R$e6&k!W?lxA^P{Sz_r zl_2d1T^Pp19A5n<92qYf@1;5}QtLI`MXKu&W_D-uJ@&HlOPmkP(ZP!H%k7+U2JJ9B zbRzerUJ76jqhzit>l4`|MUA$CB87OHQg4WQ|3+bKRb;!bDiEIAAUy!RT;#ppdq&zY za#bmUG;XVt?{Pml!P`(YL+U9+T}Mbqi413&g-!hge}8h4CY$@l%jnNqcPiDw8>vxD zl9!lh4k1h-?Ybpmu`UsjhGGA$u&XaUjAj8=m18|C(>xCQch|N$`+Cx0Q;@AB$-_*4 zgD@T{`8%^!Rq11>7Nsctg2}QS z1Dhu~jBUDkeESWhB#AO%hMHE{0s|cHGrjwPmsmm93~Patbl!YJ>cVK*18vBiEo4#w zns3LT<$io%ppy-Me*803+Ls&($CLzw*0IeabB>L5U~s&vDKmU#d8i<-tQc5uyH zuz#5tEsy=O2BY>9+`E^>6o_57PZr|g5`<_6`zDj-%MV?D$ND$oMpIg@nE^`{pq#Jr1qr!{izGo z%SQ<9Gwy76E~Us~p`g>+4LP>=WGUw*zIq3_HU(%-Fq&}UA9_dhHWzJerT*u`*o_Qim znvp*>#rwJh?^D^jhq5|3VoQA(U-&4mIO2$?h!f3qs0K)SqCYMxgHe)su+~)9;Mimt z+8Fy4@2T3xS_?y|ezB1f!NHt|LEi66kfwZ!Rn1=&0G|a|s8)e{hU&Fr>B|321#rC- zbXH?}WO_iMoaQ;_^sCieY%~3~!*oT~a5#sb{!4<;nZu&=C^dIwY9zlLpI4+^Z&dq0 z1~Xd))%imG_NOxc;pIA-#>*6S_9%brNqfvjdhq!S+CXx;omPq&zL@1X;ph}RYv@Gp zr|%U?gFT(``w-(v*YR=ds@>?9X+6GurGHbKsNc*W;ILlQ?^#XSjzw z^(a3wPk3bU)t;_4vx>P(^8lqC^Zm+h3SUd!mOcB&mZZhm-ogWh#ahD+qcx!1%aSJM zWn?eVS^U05YCTGZ01pje)oJ|28oKld74sy^E6x zh-Fa7wX&1FiJBP@#0HbHxFmo{-OL>bV3M%~iA418yV&1%89f9h33s5B8W0qn{WeyL z8Nm97X#^%I762>v8;zhMD}a^ftq2kUlcYU}u(!x=W&qnC#y}tz0NWo1Kt)yn`@f)U z0QP@D*#R8?f^q;j{y+sm9|R@^GgB)gQG0iQ&RaVt3oC$&hZRJXh@GAN8w;R`{#vO& z00bsA7h~X`rIxp{vqoSNwRbWFg_r*s%nTZ)YGwii=)9d7!p6)2qJ$m5#=`Z6?v3KV z5t#t3?Cn6908ARHG6-)M0JS9$=-|xD#N-STgp~=SsS(hg(Zt>sbf&3|nKP3KNVv{s zPDXzh7}%K{%W5m zm&L%&45U%cKn6QAD+^0wdnacG8)r)fD_am3wjfOdacysBU~FS#XU%Bk;9+O{haC_l zVxZLqwS)H3U#jpg#r%5-|Gl%`l;%Ga^WSQP@VDIm0T2@bupls*7`5NgoCpBuP8rl3~1HLKd|B2~91pMzz_on>+U#4SZW&Z2K z`iISbR{wwT`)8y52LW%p@E_{+rrIFUi5fYZ{n?*P>Q=U9&J3#dwniXTm#_onU0B&! zfOfqpXfdsT9t_e5OyXwFCZLQBBRk;R5YS#@d&}EU1ueR|y@s9DTdD?t?SHVKf4cln z2zkTshR?~$0ch_8VEbEJ{ulK8dlQSvh`)_|qaGwFP)*%lN=974$l<>z{R39>a0Z&$ z%GjCPzwIzpGYczF(ufCuS_G7DVn&0&r1WM5-Wa6*7lR<6nv28V+6`cSGc{UoWfQ1vpJR2J$Gcy-A$O3_2AS=kl$j!#e%*Fz0=6SP(9Lyjn z56CzCqlJwdpaVKhfRTrXm4%y255Ug$9~1lo=U>*18HAIE8`O)Nn+HS%GYbc&9ssoa zbvQUU8Cf`3+1T0i03gb^m;vnUY>eFOT$~)704^3#2hi;Q0R02=TgxBX{!JH%t^WZ4 zdLQ?{x#4C3aKD*jW&k(Gf3VH=Z<_wW{hy9+o&Q5_Z-772V`gS#;o)FoVb%MGd;GKd z|4m8$OFjPbA&N$}AdmeYdJ0MsvTz1)g3R@sW&%N85v0?9sVT@8zE#vfYM=>VVdiG! z=H}#jbJ$`=4$^Nn4G7@m;$VDRKMn*Y8PH>Zm5GR*g^k&pMS0UP(LYXx0koowpe_Ig zc2;giP|DYvd|9~u#Rh;JMpf9ij8`2UvS1Uh7$ z2V~3s%69@eNj7FK)_;Z{9Jj-It7~90`wn|u7B!EUekw9IXSHG*DokQ6z0!w-fky)f z?*U4%^b(AJ=;L%ZMN)%<2c>8neQW)$iDlFEaj^9}#CO{C`?X!{hoAP%WH~((-QZ@(FG^iI9| zu;W@93p>qait!LB?K1u!7+8>^*Qz$MHjxKaE16rUenN5qkD}D>1i#3o5w(PZgPACp zNKuSl-ffvv$RIQs;n@n0cdY>k6q0Sk=qR==vGl{u6KyS7h2aaUH+_T3a-Sa0_#w^) z3JF1BKCpKXxd^%G3TV05x$pqQ+@Nid1z7N=TF%7W;<-GR9@AgYn~rCj@i>5<{(c{D zTY5>#T*P^{tKA~JLAZvvZFoFmq99u8U31)nI$jf!Rg`h0r5+b)n6#PUARtFWeMa)Y zEu>v;r<6-3_w5}#?toz~bQeh%w10gr5&>UEGf{sQ^ff2MDEKkt%C`V>Whj7&OKuxd zKR4n9gG}~a?y60fJ{Vtvw)m=4`MI9u`yWfVj?uAOOJp|sr98Fg^(@aw!SiK@k$lf>C@hPJz9S3c-5k&`Fc}`XZeZet2uDD1CmxPn}|XU%{U9DvQCgEj9yK z=uN#Xtm6jyn$D>b-d$)yMnm(If*T&Lg4>RpyL|AmlpB4Jt_lgqW4?!+fZ)c<(k-AJ zqW+cru^KDfx)yoXNarbr{3L)^5B?>O)Ljtf@z^I7)|Sn~iMa)O0dLGSc?doU-I3V; zDWVSX($I@br@Z7I zpEpWN+*F-m1ZA2EgjFqez!@#rg^}k&`281{n_wmap26suKclT7m2{?UgR9qig3*r- zUF5cZHP)v3bsO;h9>To^w=%dR(4Xo@CKcG|)cmCl?xJkhh;K4P2V;wGpD-$98rXI* zc`cCx*h5QrV+!FRRN5)8xpmqR>jDW;1D+vv2lO@@UYLh@oj{I1mcjV?W7%t4OGGhn%=6V+@sAIS&;0fDWTlfFm2_85>{jiCvlkH8Ay)AKFO z)(L`X>j>{A1pF?3fVm<)pizlf7+n*k-{+&h?oM!BH+g5ePTU7E?s~H6Rb~^f8v=6lU_2W}I;=mxey60$8-dn49op92rw`bixpL2Y#4Pz2G4o@IAz#FhTD~Io`H&fhAh&2RT^Hej6OfEj z|Bfm=?eWGe>P$OHM~JK{k8B}VS$E{tOxo``+G`_^DFkN z>i7N|>Py7Ov#)Usj~?^~2~H~Z+~BVhVL7{N&bB&FXZLAC*gB_|=i>`)f_JMIT>CnM zs&_p%Xh*sbulM}l-37DHFdtn|Lfg}NUzxU&VunBW&^+GfFfSI^-yIu2-Gt22D}KaG zEBYZvSu9qw^ zz>))4`7sw*34``%Ea*uZ3+E)4YgeZ@1#1)^bK4sSd1NjKf8N^+Noch1W5pu0e#?qi z#;=qbZDB_70+yVS6Q6#Uj`-Mgo&`F%pshXSmxa$3hcEp-{~pee8G_$cRLwlqCb3S= zYw+WqkK=oZSDK;SUtbEnHFiMn7p1urNgr>Cx%WS}Ct=7CV9QavV(T7|fA|bOXfWU~ zRc(4KkLEM&2bt(`eu@hxs7`~`-z@iKQ;NJE{3A&_#moD{O;Lb%2qzVwLjYb1m)vJ1 z2YO$mfd~BjScW?hEO``chc#%F3hxyX;D^zG^o_A09oztOH9MFL9_$$|KJ!sYcpk~5 zeaoh^=a6~D&ao}P4|YmM*#hHNI+RH!`N10eD++nFm=H$EH1Po#9OK>q(i)^dmH#=# zefEY)B-4m5`5X(!j_@aBzad`P+rCZ{z<2%*y(a937u0uzwHbgH?pAk$j!Cubj_f4-p z$gW5P>pOuVFjG1vGI+;# z)s16*a3l@TD8~E{h4Fz&$uJ1K#$UM=(g8+vhrSnHON4~S6w&z5sW328<3kvm16>DH zbwAfy#+l|)gkPLczrmWFH^M+f#C&*?xHOFw<#@c-pwWfaPZf$8<@X%Qn2nS4Uz8B7 z@N5fz?5ldx=t*!*S?^P}rU;7i?Oy_gd}3Y#daijPy3$!$lxY)3w6tW3vnxW z3xA7qLU4k0+rFtdupEo}-hexn=jG&X#R_Q7O^_v)wV5?KCNqY;E4ypA%Qgl%7BnWr zC*oMigkLO&f3-ftq+rdNES8HBl{gcD;>F<*wyL7(gW5>s3o|74(jRbdg zqV|CUM|UTFNPabb+m5-{cdrS(n-RiVTOo+H-JVBXYUfVg0%yjED*?BgIOj&7R3+rs zD1b2~$_A{yJU}Vw^AKzoqPRx_jurcR_w}zW>_~0l5KO7zZQl^w9R=DGR+O42v$~>I zr|0&CxO=Tyn7J|+dzbr{2L#5Bw{M*7orxWCoDH0)zWMQ>of$`M#&4Eve%mBg++yd` z8v`Fhm(kZf(%oD-1Vu{MT`KH)jr|x?%32=t<5uGH(nZkvGJI6`U6+*Ur-VNL?=YN& z6B-%)y@xa7)ZoJisrKkz%GF8Kb;Zvkmwm6Dx7kC0#)=UnILmdLH_zNq}q z3iS0tv*3FYSE5Jf^Co+C!N&18<#BpAkFbTi0}la5wxT7wIReF#F#@aE_a(d4mLB8f z@fWJ&lf71AOG#a|0$_Y#L@@$R4VUxu!Lb^I%fJdGhgXxBfrGdiqH-QhpNfMI+qJ#} zKl`pk{o4xaOhmfU2io9#^dI)yBW4Bl9`|j*hK1e-O<1_(X!txVQPm4+gYO&oegtlH zkj;T@^VDsrjZzg0dF)ExO*Hr9-0$KM*RbDzl7Vm_Bw@{<$5};Xz&w=$)4Ud-Bl*lG zU6@e*P*@E+pIlgtb2+9?YIBhLgb7@hZFoAdm^Qk*XYjJitH0nB{^4MJz+mdwg>)z z`#T&ZZ#6e#QfR_$j~B@I5!J~7>iNB`X?5MUP zmy@CmU<{&Bld26+3}T;it0$d+G#>*P#poutOAu|LJ@lYiJ)wC6ULv!5pwvc;h~#Sg zJ(ZE5Pd8?vK2jaERwIHt0r_!>l(R;S@R*w^Fi(0n2$9%Bc#XN{`C{6dp?A{ z3*4}~8+7OPdz;+BKirW!Afz178T%>roUfo6`^@fW+fZ(QqdxQ}wgy%9iL6^(lzKfPntCrqT%o~`XKcb^ul|Z>;+1Iv zN22|fL_pl2sCwddANES42EKZlYftLRqz2hO;r$g?F3sxK(|{Pb=|f?Y6)@~AfKL+L z8p1w+H*J$U8zoO{pP4r^$pPZMh}4t{ko~>bz?379pK99_2w_`1o1!>>+KM1c#KP&5qL{^`>-;-sVbsGt-QVDkCW6JgQ+p7t@r?OV`>dYu zdWu>O3%V$K=fitOPlN)ZLU}oXmAs?RNxmX+Qzp8ZQv5sc&(XdiIbU-&EhhA@$a(!; z3k3EkfE1qKm7Awx%VN>ShGTZWk0nje#~xac;};>Zn$ze>Y2 z=>{Yl{`e&gS>=ziuY%7CAd{7%G?m0QmEbgmMcfrutZ}m2646hN@E(kVphkpN;ruDm zLXEmzfQAtCsuE6G0{5-}Lp}1RGEbTCvob}Q=u8QWgbG8M*mVgwM?U2*Vq{^+NdP-8 zvWVCuGdp37FxezEJ3jyCT1!IwILcgec60?%oIGrHJiX|mTuoe|X~=02*b)C_LByLI zoB3sDBfs&W_I5B=uEj+;L$~_&cEX$MuQl~dVeRb*Hyn#IT7muGK2TpQ-|uB0=LS`Y z%mv=GWMc*RwxBOVJ?vov=`70^W#G?V7=c`+U@?t&%ZLvqbe4sWlaPfN0^((pKXM0} z@}xut1Yqz#!7k=l;&_VcN;GBbSrU692*hNIjOXu8Tv`e}2vi(J)r>LaeHJNiM`%5E znXrUcxJfPPdx82LyMuBXI&U5lu)7!(qT9@r)|IAvvc8e#=~oSSCDg z&QN760V7chJUwngJtIxU!sS6rSEk8_8j(3%p1-xSg1ej9Q_IY1PF{!=ceD<>6r{A{GGZi#@T)_`I_H`UjVsxoS(D|#0&az?D!fH z;Sa|^ILct<9`tPK@G~_TfGR1OQ$FBlGWmqmzH5B&=kUmqh)~OOK5?ll@)+FkghEVA z7EHui41(Guqz!=L2C3Z!iGHGtcdrrzc@&fk=MYH?dE8bK8hjFjDjdf3=@{kCNO9CuxRd+^Tkx9-L40?${(9aMzq zW-{Vk$K^wY5;B4!g6CZa**+qw2l&l%@hg<C>OOWBuF^440uV|!HIoC72Hpyt2T49?#5i!R}1w#@mz!Jr2 zPbQbnx;R5~IqP9zn<)V89xJIZZx-{BpuG}#G|RF#STD6i!h3Ld*+PG299e$hvsQ-I zu!B&uhX%y!Hdn-z(6LtdHi|cb`&o&8hv96J8vv<(nnU5J7tw^lq`8T;!SN#=R#1*P zQoYS-@{H0#Nu~eCtTv@X7hC=i!@Ra-sxbvCg(7M1-)i9rHauPw5En^69E2^D6VWSIx4yho6Yi=q;9#l{S~fhj>o?bekZbcIaGI{Xx`8D_{^K-KWWX9UX4qG4fHu8D(7OPG(R;=Q7cg$Y{&Gv7jwO)a!# zE-b?K4hjtRy_7sIPoA?vP$k>%2_$;(`voCm(Ous)V`sMHpNH>h)>4jlw$d(*Mm!dT zgE11w8D9+s z)+3#HG9jHZF=qx>@Pjx^97Y#KyiainD>VWL0#rV>-m7EmfQEOi5Ggrp8Up+LfI#2?~EP7^VoE)xsh6Xs23|GmJCnpAX{YT!Q2pe}DBv zF@TiR3)ziyBh-jf_F)#)5E9j36V+f5y=MO~K|h$QB9?lvK(C>ddf=vAHe9bhT&M1# zt?Whh4VLza@7!1J7l-uqxJa26Z0QMZ;YU7cmwmuYv^jQ*&? z7X^HBNT*xM8HEhM@h?nh5(`_L7dxfi^dy~)qaFg+m?7w^3_TMfms$MtchrMMX_NR^ zt|5WSw&zs`qyE?@Wr7GtX+al#BMaY2rAQiW=F*17KH|la zFVc%6tA<{>YOEqHEPQub1YsH(Z=1U}u}r=M`r^RS6kmKV^cO^QO^|ZqK@9_wP#B97P0AwHM0Fza$JQrw{&nwd zS#?>eTiZj1JYF+9`D1Nh>752QG1BspH=v>0!=N#(or1KzIEBkUSvDx7@!lP$L{eD)26h+?&zGRFB@r6aFO@zd8HMxmM^4+P>UviWCcY;B~ zxW!^Fiik*mnh)z|hB}VNz%o7dA50DJ<0Ov^$$dvX+l5BwiWSmpOyZ;&_sQPE?-FI$OdaSR>rPJmtOTTA5nwu;0E)nd5tal5&7~_3VhqJeA=Oa$6 z2-%Ew1tyz9^^)?F#KVpw%Bywe-4zW8JRd*0FGJ&hkL75JzjkrigsvmdzL}}0mYmfh zJ3Kttn=v-9v5Pq_IK(zsvlyEC3Q7>jj=rM^KMPWU@lRhFQ_SyU3Ty@Mh@Bzny)Yi{ zUoAkHXZwy9|3QI5+|M<<o|L0scy*`< z;e0^3GUW?qZ%B+KTdyc(1H@=pox+}EIIUYj@dlIdNBCY6jYCJSHMSsmI3AcXK zT!uY4AZ-W73y1h;>%89Y=i?+Rx{xhUr0T>CJOsv_{B`6VpuBJ+=DOku^h~1aQ0XSP zjd2(jP19XTEs>S_J-sWzrXMyJ0Q1T~mgq3tGjkx{fjARLr;uJ#llm7&oyHF2*)#Qib1{Aq*lXSxA{l{dhnQ3O6(hBI{!1=uqs-mt zNs77}yh?&(r5)}QP*}bdyZ-{i6S7JvcaB~+a9{%(&DO5BI-d7SKm1e8T}rr5{Z5#Q zrbsY9Dnn?3ooF)zDzRi-_ciGQn}dF6dtqt?2~}yVWJ|(R-QB@m?49#@v!R50`0;R_ z7PbYG$Z%iUE#~vYVEYzt5g7T{e6;bj(LTDEXFKmcU))^Wd^BRW!$sCnbFE{Bl4OI?bbx79-@2Bh zlw;1O0&rSP-V2^roB*Pa+0KW*vUS0x55lrJkVX%a6jhrWoC`k*qAg{wyCYSMyYqtw zrY*Qr=!aDI+;&~-w7uJ@C8+YbvwDK@CFn$*Q}KqV^25l6>y>ZoRmwMI*4B&x&GN8@bMx6Sh3n8( zj_#|pUJw4d(otfn%qpMP5a`wHavcgC3Mxh6G+gCZWzz@W*$H)-EN6Zb15x*mAA%l& ztbQPA{@CsZZi?*_Omlsy>$9G(p8M~AuMddE6kOud&0_otB#Nc03TMLlQy~Kq$KjYT zs^q&Z0m&HU@}x+;gQO`wn|AnSV?%hZbxxK_+sw!NKl6TP%c`p_sFe!9HwQY1@6JMl zEj`s$5&NydzSa~+`2zc?vTb9!=8|5-k$qb7ozx|2Q>3tCPkZ>X9bU}|t#3+hxDk+# z8R3dsVouQSa#-+Gs?iU9312aUP+3YL?hCBEn?&8M3qZ*L!l?(A0l>QG1PM?dPbmyf&7=pKJCO1~S^ zAwDhQyQe|h=ABSb8Fzh(xkwA>2wo8zKw}W@^KOYmNsw*gcE3&qzSFKgC7q(uy||ub zBs~K1h`}nkS}VB~d^y2SlY5S|#QUlIFe(^ia8Uo)C^%NlIuwP@nrKXK3f5=sulnIv zA6#*_f_V~_Nm&p9pMAH_SV~nFqK>j$k-iodKp!;)7rE#vJiEQ(UHBsA*@@$T&GoWGQ{uo=f3 z-vm$NpZRh30DU$-pd`6(>HDj0#Gx_Z4w(Wb|0tdI`IlH{4z2eAhj-aQ&!y{E1rFf4 zhDw&K3_KObg{p{Es0&2{V%RNp3T@QT6Oum4lOI}hZsAfu7scn6I!Sw4V$ZTiB*By< zH8e@)T42Fgyd7(6RJmhq)!^y1bg;2OuzX(u_7$Nn%V4J2`>cpvx_fi|o6Y#Vh0J9V znsjNdFhsV7o8pD}a>FD0noj88)_gHxb>#m?%hh&RzUjvrCEl>7lXOYO9n3~WEa8=e;b>5Gn`$@PPX*InJ~ zfxM!s@gR2r;jySl2Ue|{*Qcc;Q|0$uLp6Nc>_g(W8&^fTYk1lZX6d@6Ya9)@!^t%b znb>yW4Ao7lI_V#OhkRAl9Vudz4i@IXZU*Z<=}A<@bkSP(_)ur|5?Y&=nb~8{#eXp{ zZlblxRfI%WGbVJI4sQ@!EaWu2JS~Sa6f1OoqKDeM604WGS`qD&a+hDg^WxSTof9Lv zH*2{qJWL({m%g7alBbJHKZ0#8rv>*}hPbvZvy{1I)EqOvWxhZoU4dLdMiIJzD8;K2 zY1lNcp@FfDyR7y+?^-0qH8v=xzAx^19OEXcm{vE?DZ<7=1lJXexpaG;FG2yV0=`jQ z8?_oL)XGbBiPoa6?u$VX0~{NZOdM7!77lc;Sy*Y#I5c27J(sLmN*(3&xU|L9$7d{7F_q>S%dBVz{WqCyvR}^(8ZbfVoz4$yIwGKcNK& z-tEZ8UH@?K>B7+Jady{O3dUULM})neJV2=0{|K{C)Vg478|#u5o{CI8gjv5rtCU)k zgkpOOPA%fVstxL@3S+fHO3|Z5btUAz3ZADh_Kx5-++2uUHc%!%kpY9&QLm*DkF)s9wy?p2uO+9v|WRQzb%gxm-&cO=OW z)gl({M*LCZ_EGaj@V@izB}QFbmpfC`d7Iu(wSs6Itu7y9GjB@oc(U7dqO&B8O{a3= z0s+tW9z*L;fjb*(;FupPm63M9$u z*%dFYk2yGFbTc(x<{q=IQ#uYu)-LyZz*9@jW9J_XcWfYJVxd%`iua&GsC=P3ffI>o zx1)SV2@BJG76@UQcwX!i64gAQxa6VHDk@9O6HM&vas1mzw}_?5=?GFxGZv4T`etgi zSW|I)$m=4K%dQ)!f}A4rj$bhGMENx7obM?0Nafj0k&o~ZFv`=D@Hsw)a_!jp$tfP4 z_&3(I;uM&hnNyOIH5*cn{f{O}f{35XtSj{)CEkY#@EKjyM<|wAhaTqdAbjUB1?W)I z0=!5sD>Upsjy_zNkvPjWGu9G|`>eZ};a7&&-&IN)u(+yOuv{4Dj_fxXA8qjWQDb`G z`X#2WN77XZGr(1VPg(g;*d^gZ>I5gMPI^B;Zcwt~Uq$;}wb*CxNBabh3azpFZfCzh z&ZU)KN=CQc22s^#5OwfveA{V~+cCC1LL&;k{?@qrzS+wC6|{#Vc{)egWd;gmNi+oD+Odi=1 zq;m9~JJ@)P?BsBnRQw@BlM)U2I`QvaZArnr9JS9~!7@4XB>8B0*h_&ihV{$v5-PIx zame2hR`Cg@c$;Q*4jaybmWA2|+cC28mRY|kKJivU&e{nGAwQwAMXqJ;HO2Jk;WhKx zxvXY8>#StmH%k%$`4cadSAr5wFVwyDJ&A_a%f2DVNQg)nwHJ1l(oB4q=yV{pUEgs! zellA!-e#Sdy6P+Xx=XWt*I)wC@3(fDqCg5nS{n$+f0|9EQoVr!Gaj&C*HaG}qpP69qE=s0yi2y#cVQ5tCeV{p1;HJS?jE;6D7)$#C%D*?2(g zLObR&YIgg_!}sj%)E0|$%a77CXys3l70oQo*1HO?`7iUcs3+m=-V4_;;3laHniHMJ zk1VsS1sA8CUN{8ef`YiO*!H<`-M-9u8zot$a?pBBGDhI1qw$rRrk8EMjhou`1lfeP zn_R5NTX#oFB_hBhA%q`z54AXULdy&8TVgV-3-XlJ^k_6@*td4(W`oCovmqA;)|bv1 zy)5DzQtk{D6puH~UcVmlp1Rf7IZu}W{~PWM`O@%o9ZXEp82eK zIb>JSsOgeXS>uryW!dbd-inm7SY2-|oA48ot5VD$@)x5S&B-#}#s!OxI=;+hBBB>;>5{z$~B_QeGkl(N;8~;&vQ?1 zK5Mn&!L8ikP9OFLf$CLy2;!$L!5r3BYy;wqZvfg$iVcH0eD@z6Oxb9o#?eEU%3O6{ zw|7Sc;z{1)d+s6e+`8{CBwoKvuOno|K22-gwm*6BWG#9 zq@+@W_^yS5-;gM8ew+R1Fi1t$8N5d$G)ywen(Nn6{_5__-hS1@w=j* zL`l@O5fE*;qw5VRXo=gFUsED1H|Z_9eh4ZlcQ;Om=px4z%|L!(9BAGUyQA`bZb8&E?$R{SxVyW1b5Cqnzg7-VPj`u*b2ysl=&k?M5WQ+R4P5wKbZA#n& zTBy#oZM?lg)txVy|1jlm7enH58a)hK1`9?aC~fCETe?9@T@3tG0n+Ru`-f6fpo&wh z^I>XklCVf4I`_5w-^s2OqR@-!vDY`U02W`R^M5}VM&ZQg@}2}z>{)SQqE^!GM$AGcDUpNmdQmkrR_!j+3%JlB zb^vBRws(~E3_Yb6)}mEUsL=4ESWRr?mJ7e4x}6SlMn9P(%bnZuYi6J6Aw^Cd0e?{5 z%zkX2RgA+)GYSKuYIHjY@6 z26pyM(}#p08o1oog8MDhy?SY8RRp>1*Bj-gSNudi`asStnZy(YlE{ zP8LRas!>|`&3t*PAyy5AP_#k$=lH8+9npDZ zom+Ud>>j7BVHQ}uyk}==i@dA-;EiK@wOahi5~35{QX$&~S0wGsFs@SmS~{hEi|ku; z#c&7%rMEJ#UV>`Qeq<`n$R-nP+zRc~lQpi}LYS%~f}i9N9q}m1vAw-6V|ExVcr2kCL*I!1Y@CWsbIq{S4HIG zGWJiECxtVhjOIa|%+}BD< zPn)m|(iyBEVVcdHchIWR-640?qqMNU4Uu2pHlaNjm*q+YIW-Gy(z7S?!zhnwZcwi2n5#Q~kp7;lI3UBYnMC-A zNt;ZQ+F>$lPuStPH1Z1X?Kv6ceFel#>CBLE*I{1JC743ND#g|JyngIqha5Uh;nej7 zP22CBL)ya!aBjY`{0^s?k(iC;ttS}uB0X!`V04l{lva>3e13n}rDq%>!hNk#ag4wvZlUEk!DE<@K~e%63iw4XPk;%s=Wb=MB$5>I z;PN7SH_D4{IZ7e@#oX!`5Y#p_9)d9^lT7Pc@j#12`^$Wx zl)b!e4ttKe9b$tPM{vcs!BS+ zFgBqb@j}~)&V25yh|m)IEIZc3qKFZuB0e5)S!8!2^#}7%hCGBz+g7x5*8BnYzJQ~S zxB)_YqB8@e*gl2K7KU4z@SGHmL#>q(_Ndwh zpmKoZ=k~#B|Na2HNM>O7lQg8&$dYL3dLBrI!3pX-iCc>=@`1?-7U7;%1yCb+O&$Ls!d{aH-Y^c37{-R zR>z!@=1RoAj7J5BzU+7W23wcZH5Oc|HP!?tp~Odosmr>JC_7!rZ*aL6>=8{WR{ZBU z@};>etw8-vc)AXj3nrRwDc(nwu4%r`foEa@5x{Ne+oYa|QPY zPs3Q&eq4`uRE(3vpE^A}-@M-wlX3!Eo+ERyDNR`ksU6vm)Y+lu=G(7x3r@#;3vnAy zcJAd)wCCFH8SntBe8eub5_}^8{xEPG2;guM#Jfh*I(_d8@gbsLGCRpU4<{a(eJ@sJ2)(>z zXMc4o=G^}=Go`|_@U~k2JC!h0XE`=}@3h&VP1l5t#d(lw=rem!=j|?7EBLw0Ot&Ai z((xEmUu}0NXS7{4u{`fM-erC`p$Veywd$tO5&?IY(*%`@^RTRIDmET_uBw#EkWRyH zb^+6yAXss6LDUaC<|#Fv?Y(H-`@22O-w<2xp;*4}tD@X`#pnn1S=h66G2{E&_pu}` zLf!;&>?5Mv|Jqy)^^bYHV*(;|uvD?Tg^y<{8>6&!Qa9n=AMQhTso^I>_GK-V2?tsg zq9SwS-^$uYKm}UZ=W~35e)70#rL{u$jd4*qZY1#Ms&z(m**=~SBAY=2Rn-r``N{Wt zSs0FV-;w}`=vZN;ij+ssivIV!bSX*#r3^pN5@mWGA;u_`@^{rXb&Znvj37AH%d;bz z?_Y5Rx@+BbYcr#UhCu5WZ-=vQrh~st%6*i=XPlS8?Gr8}mRPh8t37%pIwHhq#d4;HMrI^?}*h245n zsZk=fQe+%y2_4#zt=Vl#Uk2mR_SN_rDzRz2G~kpA`Jvz|i7Ev+hh( zQU}G*g`9|SuJ~VEJ*dg~f?}!crav9%S@mpvMxI>{QPRPWZaZEtzpDS~Se?PkImy89 zo~Ls6<~#5T9=2xl;8&cs=<4`uKnIK49~Kl;ySxW|HE|^w5d+O_<`Ddkdx6Ofc2q!4GLGf_r8gP5v+V`ioN8KJvlgh1%U@-PauK{MJ!)Aog!jt8< zw~&=dj)t(*Y|q=G$znF@Kbs7QTbj2m)fhpGKWmk$B$ao0jg~osxQ+Y?7b(Z)4^VMY zqS|93mPk8BlhZHfXi*$7;5(d>^oA9Y;WOtBFyjk1d3+IOAhTa+*r<7meFuw+ z8kxxHR_R^gMU&TTHJ`GCXeo!!-86v~pkW@F<{!ILl`1%Xc-#UcD~h`tl1e`a#rtPa z6ul#hN5ovJ0e?oXx38rOE=uuF@x-mA?@ljzkFMjk&&7KiS_-8#28ju(!B@lESXgP# zH5u$Je|6=K^VaqWEF7q|Kcg`fXRmk0n;A0`GL3oNQ&nO@-&R@eVu$LGOpeo{99Vmu^4)aK$ z?H4VW(sXmuj7|>)Qt!5w1r8)px#6u=&8O!2US!T z2LI9)|0p#;*$0&R^X-YrT8(JhLInl(eAko3c0N8`m0>S$F1pl~-f<}=O7#~bJLqsp z+?V_SYcPtDfu;3Gg?^mYE;d;^v@zx;gz)$Djb1~!aJ3$1UEQcYFINMIGq`y`>qVkS zt)tO|ZEH3|kY-p3N0f9}p+a@F7BlmzJWWe~&e=l}s<~hw9qTPffZ_e;YtpP)v};-N zMv0#!QrO>y#_VIHVm}#6^$Z4G5lz$>nkjyG>S%>o6*=uY3+gPe{>E!S`o$G}dX>d47 z@@?$IL~YLhMjf~MspceF7~3$tuQ1Sd!xpW;^R8CVxr~`|#yMj^hIM`)$3-1<>*%zC z_H)_jqf>UAUWb5>z~_=uRD5NHW)h`H^PB#o60;0UrP z780JB2o=K#{YhAjo9Aw|Jn#}^k=%BB>_C}yYt|4pNtpe+b&FBB_2elwpslrVH#VNe zdw2T2V!5Hi?0bfUptj+f6t@vOZ)7guLGA{t4H(Rfrtqz&b-PCC@kYc_W5G_06xjJ$ z1WC<11afZP#J?Sb@6bV3_`DxyGbBN76aqiwZ4>z5R^0rg6lTu zLiN@;G3cOAU%PB6UC#Ncv&DC&xDJa#!Us((z~T47-KVeQa*18 zHwy6V2SB8H)v`vc+93vMRhe6!s=EC!FYG35T*mGnM}Vicl<}zT*#4!zEm@r zV3lM}T3wa8mCHCo`8~$SCg}E{q|Yc&oH0_o07~Fv#^ovojlPbprO^P=&v8v7#~HWY z)Y7c&G4Sw271&!}5}I6B?)X3kBs83ga!*U#Hk?ubsE2B}LRBEnJqf(<< zrach0T+zFhSV>VRMF2cQ&KMAHDE^}S+$Viywrc38`kBT*9xOpRKJIv?D#;LvRzd+#2CHKtCp z$#~T6qEeTd5oQ*&{KH%c8g7l3`w>#Ids(vAR+Mimpr9P8@T@)@Sm3HIuT@JeQyt;E zsKw%|GbjcDVr;WswbkSy--M~R40JsiwuMR(6v&{TH_QzGF4@*k8cDNp`!3QXj+Wek zk**ky>nS;6z$)R|1`UgX>}TV<2y$Hf($IYLSD2{jsuyYTp3ZyN%5PE2S;5B_oB$U# zb5J+)?v?di5>dRv`UepIPD(($Sw z+05}xF>eOV+W7j23C+#{yr|~~FW2J9!jtD!P`8;s<>Ah8ZEv5!(GSCd2GL`OYSc^2 zOSiEsn$ipt$pZ|xrt!0y+N0J^=cYByRZC~qNxx&1>> zBM$xdZhjzaxQ|~+%?cLd=rR%=>SC1sg$wKau+faOO7d2c6xZny-!_&PYa~{Y5Su4= zBq>i8*M3Mcn{B?mh+#Lp{_(QwZ>m~H!$!yI;`USiJ_okVW7i4N}qBs8mK3q+PU*?L!#y!v)4bTsi5mmu9y zmqZkgM~=;?&g0(W{QwdJI@RAel#;{N!)HU88LS!j4MnU?tgftb4J-@^x{2Kzbw8UK zd_S|BTqrrDp#I3JRl|Hq$(VyOYfTc#_}8vF3go;p>%-q%c)DMD7g(=5PR4Cj8@f9QBanZz z$`f!*3>zqq+XiZUtV&7;uOgaWD9rZdSI`sQFvc4XTMHpXsmzbjjB<>;c>d{ z%wjckKd+Zp*Syy}#JNnE^5w>G7H-&MA8O5;)b{bwaJE%^rneHNW*S~yeJN?qe z6;C=ni3ZS2Ih(|6(4oMsnx32-O{Si}97ziE`C(r~?2VJ?;e+JnssUSi+s6W-W&Afx zR9J8yPsK*iRpf4<+4%hX41>D-# z?NuS4TfZ6v!Y&_ct~6rAGYJkH zMMo?$bWc#$w>tYdgzrX55@|y0XwmSE{k5Ku<97GMF`&~!`wD=Ox%|Sp$F#cxNBPm5 z7J&KmPS4HR&L-u&9S+DY@+1>OA-9Ap63i)r`W^R3Zqb@PtDndQoy(C^)H^*GMJ8t` zi2WnuZJL6cOSxzjcI?_A!3;2SwChST`9=4$RI??)vB+z(WJ8Zbii`~jeQkS@;6gth z487EdRYD3_6@|5!7MPZl9j0Y^nqQ41?aBwyPXctyVvR%1gr%sHn#e|E3vQr7(Sy;8 zeuPVg#vyG2M}%t+)`^R_vR)B=C1W}I2_p06sKpsNY(K--Ea^voG8};DJ@sFN&sg)WthvgSnPQ_N=T8Cni`lZSWA&%4mvp4iS!Hj1su>os*6%Uh z{ngpLsh`mp6Jk{eW=-`hG!frj*EC)8e&FB6bo_>R*Arx#zcj>~OtzAn#se(urx z?#yv4-=ASk==<<*H+R1ki8l=_E(yf-PH=pmn`rio=jnD;T`Psu4|l`#Klj`2=SpSfP!EB0OQrmvP2%;|<6XeKL|b{pFHt&OJjvmQ>mPV>&+ z>)E4)7k{vF93+DOZp!@+@aX?Sc41}W`gh1hH&M>Ip9wMS3Wy=#7XJ0f88<{I(=NVW zK%<#3eofk318=W2{Pks3DWeyegD+9h9xU+W{q5zeU75AYR-5FEP?(By! z;xn-~W!32+z3yHolh%UZ(TdIX#nZGc=Us<2yUxsimG(n6BBc6kyuLIqV@1$Rc`7wT zVTCuw3XS^%6IG6ceV#BY8Wbj#^+9o8_CSQB4j}0zr z_1O9t={RfD&egtwv!>_4A7KTn=4}7JeDwdAV*iZ|`adC8!cLZkHgYQe_rTSE@_PRU zT>YnO_djcwv;H?>?;i!|KP;es2}VS$oS$R;l=1!rT>aNr{(r@-{)e*nzv@;0n@ROA zzUqGnPX7*1{R2_`yJqzXQT-cC_1XMC*r$JW`Cpu<{}QY|eZT+2tcX5^C5C_3w*Ga> z^Hx$eKl{n?(2i2ENdkLlmhsefkVzkPfDHD&)Xh5vz0{nuRm z2ROz2X^;I+af+LXoq>h(li}k0H+YJfnSq;`jfvy4*?;&`T%Vd2_viViH1#ijij$dx z;qzC4otyRlo?*ei~y$|9Ekqtkvh$|9OM|4WIf)D*HFZ z@)I5VuRHxe;ZvU^8QUk2^`H0@I~yn4r-MfHDW{nl+QNHeo&SSRi7gBInb`q4GHtx3 z?Z)hAJ}YyRY7)UoUXi5<1q7&^L?8u8B8dGE`%xSjkXIjKB$NYVLq!z?MMsopBnFA0 zC|%#f+Tbo40lMa!et8%8xE*Bxc$*#mzVod$Aa?>9cALl*0QXfD1@TIZt;hMCpET#?#t8LqSj&yuZ3LWq ztoKt+K{8mV`Ir5UFu6=4ecZPj?f~ZA?^nDw8}V?C(^>yiHHJn7;A;TLKXxUaY;3q! z1dlk`rJ)`*pH!7-3hf2hUH&dsExKz7vm2}QzlNCqt(a>r;dQC)TdY#QY6?owd#OV- zp!Rr!a@xjL2nfQ+5b_7RfntUx%d1xmX)X`V^g|3D7)1?g)r}17D1KLvJnR)}$G$X2 zHTp~54VJ`oWfs#S+8&-f`1*!gipCxB9wA9ug=+*`fD&!IwOL8G#2mQTe`4~zY|Wr^ z0igWrf6?_r=}v>R>b9tpduL3C}AzV2l#F= zCn)Vbmv(%Mm~hu~hv(2dp)W!nCVF4mG#JqK8kq4S-{KQMq`Ro2_fQsM!G?@;6kP9L zwj4>ff)Z>dnvDD2aeYCooxbNvSSyEusBn$RN7hAJ+1-o7=U%P4BSB6@#;A%BdiP8* z=pEp~e*DzFNZ=6C6npd31GYxkF2h{Cevb)mf2A=cT=vq=|6t~ZvG7&~QeP(;{z^!yl$aD*JeP_HGrRee2$aBpU{5+7Ktj==z!&o}ULif^Rgr zYP#yG&SZ^ieF%R@WTiR{Xe<%h^uXQ574pb=N12sm)F20S4f@T0+EqXk>UJmb_MGeghf%t9r^*g`bZI(Uku z8`h5=K$wbpA$Ugk?y;QV_tOaE>Sb$MFAT0lCg}$-L+@&2%j&0180j|yPtAA+O@<`e zeBySTKG1O~lm2K$wNvH<$5x`~z$iBR{64{dOh4&&T0=P`)s1xG%N(UU%Z+2{X42f_ z6V7I`_?ca2=729u14j0}USM6qx3jF6)R2EklLkdBvmTl!QaLZ_8Ws_QQ6wiG{8ure zZ2ygc9w$Fc*Wz1zsPihsp8QfF?f2itCXJ1R!);JNPsAMIoS)y~W<74cClt}=T#*yO z6w|vvqkKdz!K_|&4O5A3dVg4g{2{z_)`B>kqmN)t(uRZgG>2S7xFuLn-3SNorGIdf zsLGMjE-)8KpTMYLFLMld%tBR>V>O~a?Y)NNWlQM(z4x2Ls+HG3L>Ms6k(c8BJ#WmU z4i`x5!o9|36HO4lidgAagnl|QE61X)pW?GXS@S^M01T5|jUOxXzryCS-e%e5{9VXE zm4s$vYw(y}gJMyL=G=9A+NMRD3$w#rgy4ItDgSeLAxYlZx2;khxEA!dSz5Xh2i1MU z`Z8jJ*N`wV#mu`muAjR^tuM@3h{ww7^Bo$;3*SDE_3=hV^l27(&j;n&V;_84+%w!f z3H5lD%-rZ1J!F41VlPLpqXg5-D&~Q0uqDN7Uihi1Pzf`J^?cRK&fYI@_?c%d_enmf zrBC(*XNq@CXbZp|hD;(@2QVH`+6PK&gzQD29_9-e@f4)pgiJ#TacN_%WKi+N`j}bI z8l}-^3Hb*mPX>W?>+Sb}oxcd@gjv%R(#hd~o$WivQ$LDocMy)P=+rsZzSp*?mj;O; z3G+H4W<*UJ=b$nfcC%-}FszIm1SR7KGhp3{U_--?*|<;ipK6bfw-U{44|`1^=n?p$ zQS`@PX_VrDceM#RBCE-~8GxP6XcglvjlC%xi5Q?S-QcC14TFLXNPu2%TpNKPc0${E zpGNM#2mX0S8MRQutoRo^B2*Ftwt)_R@I1M~;`rt7*H%1b{X-_VO;|_-W8Kn#=y9T< zY0$vi?;0_d(d3OuE0-d&-mnjx5U%cDQd+RxaBDB1(x%1^S@gqzu(JB!upmbdHc-H} zU=B+SUizMD)s7MG6l2%_#E}vn3Xy)ZJY*Xp$?)pkV@0I94Jy^_`aJ-r<`GalPxlO_ zJ@bXn25FOiDoD}Df8<+6!6md1>5cEgBJK6>ZIs|a(d*wDP|+GXh(RX3zxDuu-Q+TY z;QX3AZ(w`026<`02S@$7y9Q}=HlnL9u(G?bS=Mxv^upXy1GhZ}UC<-yObW+w?d(|n z>U@l~KApetbG{%NQCbX)FMheIKzRVkc5IjN>}47GYd)8}8|hoWoSg0H-^axzEk++` z5ZqATuk|{A-MOC2&~st75w6V76^6;j-4NyQd?eCGIL%kyY+Z@jh_j&M2u?+7V9JWU z&yA?s;sV)CZbKR&FTK&sUD)XV61|5NhY-@BJ8(0xrn zZVZo)%Od{g-&=gxJ1&=I_!>6*QrJhk{Rh2+pE-#%m%o1yY7Jr_A@?PBWTPNAvYH{y z7!mm1FD8Qhy;A!erZ>eLJ^ObBwSb#WKd(2BH_tBBEW7Yt*PWzbEl&+-TkHXyiMTRC zbrcb8XrNr@(_!x3?vSHLLWz?ppOkn|0{B90&9}G3GtCz!JPR55Ln3tokfZ+86o`hzRH5f~C!m5gf zhlggrjOmnD)f}((gd2C)_Yb%--f&cVV&qqVT{PGjOJd;wu%wBJ7Sz5HdsD%hfh zOf{#ljS%kw*2!>MScSVBd|%KGm~0372`kvLU;j0ktl(j2c$8x8&^0Y7lWnS^8vE(l z9=-pCZ9qjTs&>~wjjedh(DgK|&UB4$?7eCcEz<{z#wBr4jpvm*%Fr|0fL=^)|Fgyq z;;>W_B;kx$SP@7g-~r+UA#h&c-aq^O;e8KKeB6N8@3*#+jy3mkc18Sddzv||Jp=Ma z0=xpEmbh8VB)WY5^3SIpKee4*`o1p`3`;l)Sf7fVAN>)}ott>II2>pBNPmfgK6v50 z95(6z>lGODF+c2K0i*+KB$sblAtIkj3D=MA{IJi&b=RdCdTBy1&`f8T(~jWfZac%9 zhY3G>Zs1m#|7|5K=)uY%ZFK|gX|Tg9oeq3IN8p zBtnPi=c?C@4>mo()pOk5!|PZ5PJwrqp6+wtYr14ar5JzLi+t}R=I#`3cTWwwN>0R#c&PrfybnrQG{a1oy1|Yk~^+3W>d61gU;+| zJ;y3WFv79IBkhQuO~H-UvzoDGLh$IF_%>#}x480S267e}En+3?`%$KKbNE2BU~%hd zy>B(w%w6U)K|v0Rhs+ux1&Q83tKiVvEGEqO7mF6%+Jx+3Qg&kYeBUeAy_jibdcgRd z%c1$Wavu5`uw%gZ{CxZlT5!JNi~rz6ZsEib`@$h_X}O z7ud4suPXc%JO6iCiW z0;L53?PE&=J)f0w&C?T_%rl14&R!rp{Rd}~vxgUKm(_pL|82U*@~jXzO!M+z;T+|V{$lsjy2#M~$m;&?1;^?DIO0v(47le){CY%!NU(erT|>Vw z^8f&f2mm1aA)YmJpFB}`gW=2X@TvpufZhW7Ee+gBbY=bTHt+l$Vpn%#$D5s`vMU<=gY8kE$RHc zw_gr70*)$IP-wWG+cM1Yksi;*c_O>}bD2T7)8M27Vl%0^mRlYa-x(jd@N?8ro5X!kJ@gN=Qc3_Rj@LZp2;^GWqe*HsC@SOnw zc-Xs|%YFzvnfeIzd)4iN`+N!aQQ)t83NhdD3iI~ND?qsfQ$qRvnRZ!&B73FRLuirf zK&b5u_Ef@1^fsVz)g!TnK`pI6iTi!%Ez~>&?PFl%2@?FNCx;TPRsFG%Gk{~d|MBl# z{CmV--=@-Kqy@YmI2XS1{mZdOia%%ydRpk!nh!DT*YU(N+1I-cD)fJV+`2rdc5&^! zJ8jLppNQmyD%plO0T#;_7v7VFoVUwAyoX1|+wA|65_uVW;?4T;fg>tLuCOj1GpZQ{{bJulw%ok?EbAmyx5MQ`PQh39beJ`!F} zg-EZ5dW$#DflD7l-7O6-diJlH`bzY^-_DACxm z0`1Qb?8@3P>4R+e#YObmJoz;p)!U5)|1jOn;bSn$GicDhu1CI4V54C&6+Y0wQ`0i$ zrnt4bCiG;^*P(_7`EeAhu4dQ(vRa+Plc_P0AmnFvR}~*J+h~kmpztCFldRP6{iFd_ zPoRY@0mxM%pYpacwB};X{0Sb?7zBLt`_TvCqVnxLT*?$S95*l6$1oQhVU51p*@i~AV zFK;#Xo%NZ3RNe$fyeL8d+bXWam*~LPwiHJK1sPx0+9TV8+K4V%ML^9BP_aoM)KOf) z`*0fONhf^E;anq*-#z3R(nvl1);<4h6~gEn;qnQaco$ueD8f68l%TU z!U>klQG2B2BJHpH*@zdk!p^5v{%KkJR!~rjpWl_A{kzqID2g|<#97v2)??6>=(+2g z%ARg{n|}dk&L%66x|RHmb9yEXxxu~seHYu`7eqCe(9H$%ELD_9HPqx4c?8Zr{5kd& zG!0920>sZAA%f_V*bmPZOCJFGz>X))_fW#KMOMNtU>6o4zKMeQg*SIgo+}`!O_-m6 z^z!K{XA|4pEF)glz%C0|2+){8%8+(`ZV>*E$Gd8q@VYQUk#%OZ71#z z{V33@FFceSB8=_LiYWMDPb?U7fLdN9z}LfKkZbO_<2m-18{+IcWRSOI$g`xchu^{T z+0)tWzIm-~$l3_PK_1q|#|M@WK|l~rCA1SwL({cBco!n{K~P|*fCnFqudKv1{ z))-rZz{Z|sm%ATe=66oUXx05t0ys{ceuVem=BgHk^zT_1d|KuoRBSCJghK1%?u>f} z_ue8y40N-ZM9nC)AKYq7x9emapCb54{LmNVb3$!XUI;jU!+F$03hGJfKRf}e{zCWh zMQ^L6NB;0g#&{++7Yx_`3Fhj?-K!S)u^Rm@8gfHXl3#(r%Wz9C=P49Jk{A0K!V&AS zMKGoSMVQnO@-jTGpa(tZ1RZUU883E$P%*>cM;HJrJlUfjUAz zO%p>1n)m+FeY6jqtQVp!nSa zU_zXqLCA$=>xE#)S0GG)>NMxrSDszd2id684{u#`k*I}FmJpO@TgLBg`nzj>$YlH) zs;=w^3iM@t*-NX|9HWuaTR%FrtG|Qsq4w@{81;PO_Jn6TH-+_8@Xe><-MOsCThFD! zoz?E4dMJ1IV}~^vbY-`3svE04bjI)8ws^<#TlJ6>YrDN(e(Rh&>?Rd0F3(qYnkZbt*8ow7!A+ zFcp61)g3iq=n6;UE&JD<_IpR<0RKB9qxOJ=`C$ zj$jR9uR8Vok9GW|e6q?&nR5he8KK6W3kbac=lFFp!J5ghU}qIZkBH0qHC<6?Bwdf6 ztA{4x`)`LgJ70HU9xKBxN`W~UT_`tv`kB>U7i2A==XhWHU!SP9-jW38g&%p-w1F~V z_HD%kzJn6>2Us|!&EHMmvj84$!PZi!`bmJGZ;y8N7u^U%(LK0tNPrrbzLyD@ww_Da5U-(eZS5ap2>P6u_}dqrtcl*{uNWO&MA9oXbfZ1Oa9{dTVTfKoYemEmzjlXRN|`}5UmKe ztkKaf;<-B%;%ketfQ{MJVt@hB+Kbiirmx1|;mg2hqngv%%msIX4w-^I%Ca$Mu|gs9 z!ci{2y z{P#KjdBj88N7_frr}UZlrXSrRh@E)q3-T!wZ)L23L@Vi9!m*^g39l2LG5$epINlw8 zG+rFf5|5sP3eO|gBhQvY=NZ@T^TN$9*5MPUX7?1HaBc~%VXooQIGLUf$CaVhU(waG z){`{{I@aD*n@ye8>(##glg_>tu67TMi_6wut*X7Nz4^{typQF;e-%QU$cPloD5sHZ z(DH@&Xp`{K;$iqAQ!fH8S}&Z}8Jox5YGFtKN*C#mYE3@4zkIZ<^o_6S7q+!q zS_K^4i2x}}xAb5@Of*94>ngVHn7he$PM+AKIyVib<>aTwrbZ_HHV9S>*X4ha=US$Sq-(h&z(=G0Q_a@0Llzw8&X2w zbf$P+#RwaOZDo|@a-5#OgjV!@Fm4ak>mTmFP&Xo+0nV(sAZR^#&KCxhEfjy6W;tb#@g#fHZ=1Z2 zW+heqJ8Fa^LMh*Ux=UOy+Ui9_U8@;0x5-i{zB&b2mDR zd_}*IZ{$tzAbJ0>5xLSJU(Z*nxAVR#sEC%UmZPLbBkv_aG79lgsct_oTcR93iZ%}LJbKAg+(g(kwecwF))?x`CeST@2FmI2_R(`&h5}wA5%-h?LFy3mwq>!UnG)s~sSJ_CaYtW$vgy zLoxL7czoFkgSaL^3^lC%Hnxq57~Zl?)CqI-x5MFk_h8q_ivwhApNs?LD`ebXiR&*u zu%tWQTrda*Y%xL1Ovpohy0y?l{XfzIw%$$B;FS$SnwZed zgt#Yy2)89xllw z>QDviGtY_N2c*FziLwG2B~jrhYT~=dLP8S7_6hfMO$b-z^~h@yn7$%w_sWq-l6hu# z3KvMSMH1KT%Q=Bn|5bhd+D*b?MCnMD(^1GRwNuAWM30z!On!{k zF!=TNYvXH7r?eh@O~Ufn*}nO~!>#izv}Ylg79Od6Vs0wuxb{BQKGs3%epS8p7KQrQ z`T_o~hGSNz@@Ezxo%(3U0sk%Stsao@R`9_0)(@x(WCywek$`f*fY+4QW*-FsvL4bN zTEY}SQrw8d;F(dA2B{ZWe8dlO;jiRUiR5BR)h?XTUhfv@>KvH}eLbp9m%H2xI1 z@$zH2sB{SbmOjbt!# z1vsUkWK?n$IOV=%SaRt|HKX4egC#4;T;zHvbHNX&oBD!PxwAbpJqtZ^)2I4pfMdWZ zAl;u;pX0;sz29fhXVGWUXH&Vq-5 z17XwV+jtg<^oj}PLt2F2@TU3olA4utiUevYG;vhpza^4~e-H0)WvQj~Q;wmJ!5>I9 z>dWWjdy8upSE?;hStq-WHH>*Ze-j8scYqfNn@&FZjWi~1nsnP)uen;#nb%p?S!AQS zF~(7>v|4?x;C8y>6+-Yw;$+q4lYfQ}mkZMTwibcNzbpo|S!5+?x7D>7zPKS*TS4k3yFyy_(E|@TozQ zijR)`tnvdJzpi}|=Thb|Hs}e20^)v?c6wxm0tKuu`tD{>j zkYBBktD;^m|3|)}f<8#BQ4&>#AwGKF+$r%@#E(36#PpUI7_39AE``%3UY%s6Bsw=c z3@%Mo436>VhiSP&vem*59L4O6KcOH+f2-mzZ1K>u26Jg)GYWGdCNoSpKS?dfeih8J zg$*g;f~=xv&LSw~q$uQwG}#h(%91E%O)^B3Ei-beIOpY$DIcaEa_K zXNMZ$CiCPhDJBb68?kMFYM0qpz?^j53wR0o22AC#T5@NL5f(L^h`PjjMNAdnEx0eZ zFZQ2$pQ1gKfx1C~pnOn*mn?s^U!`AxU%6jNj)Hu7Zbj+B{3-SW!h_5M?L#G~3G@q8 z1%9>cK46`KE8cnMRvslcUS*y}){#+{yvxZ#5WHO(-iHu%TW>%?* zjb2y!H=Y)0Ro&8$@tF*(bgh(>D)#vdYx>3%E!wK4{ojd=99pq%nM-Qpi#l4oRb5NM zQ#kHqXil&(Pqq=F43ADx@VQHv+Y)!jh9*|Ri#TZXH^fgAgi~=_cgGp z8Kb&q%~yK0DoNb3&3p`N@--scTD5Yd*)XgAMi$#v4v`LJ+&{nO7%dB%nAbFHv)ocS z^fL407AP!}nyA-QY;kmDQ)j|Ylt8e$;)KOEoE9jXWzjp_PFDO}&H!bNe=+CvBe#WbLHR%<4PqY>mtDmrYWvk8OoT3R4fnz`l z7YV{A{-!IIu`R;;Q?=hWcTvpuN9dHyHKkpqtN49RsFwmMF-)4YXm(C_j?6M6o@|qL zGJb59*Gn>kb~2TwNNnz_<)21G{^El0oyxNqI8#_d%$!QH=vxN5l(g?b>*Jn5G)Lo5P7-ZNJ{=pHmn81V z9iy8E1Wxj8sTt1 zCl}~4<;(<~^SBrAEMwV+GqVp(oXfd4vka#=TG)=#bPxHQDV^1w<(>7N^V?Q*tQc71 z*y1?i7Uebj=+X9_ouaZR%tx^HWhRaLkm@5dAd@T-W$UGw|4>Ytu$)jdj2hjt0y8U1 zz?Ac-N508+lb8>V?_TW^?qb}Od&v3A3l$iSdc`iTP-wi6%(9U9(-=OT%kSdH>Jt=6yi7 zHlVPl;A-(B|3ev2dbO-yY(Lv_BJ35>Eea?OKk>eY<(CGO63(D}Mqp;-qHsSh zVHo}>gb-D*;CG8C_f>$Gvp7fu^wpIZik$?GokWf%o^`*Zj2L+#R=f<|b{`MOK`kbY zmH{^zsvjzI9_P97Ov=)#;)T+5=C{(IpqPlE%6eiJo!j#n$VbIR#zx0S!a~DCv1Aq( zEidW9m6WKgG(U&SZkIUA#!B0vd51}_-DH>Kti7(I-tmZkdyBy1*52iuTD{r+bi7*O zp4j_tJ}N85j5N_8dWa&+2IZ1f7){YfM7LC znf7vc6m>qF%IwKA+9?>6AkJ8 z22-2cTU6};bzAl`EHKhO7(euHUHTnN}yaM<$9bJ!?Te4|eE^N7NwOXuJ3pCx-%;sY#e&H6E z7MGAw{y*H^W0WLKzbN{)ZQHhO+qTVVd)l_$(>-n5#>xPdxu-%A*W5C=fljUCQQl%sC_@?GF z5fL74j~RL)u2(~~){2OW0j6bzc?|^snR>x8=d;5xl{1{8A$CmQnGN;vHHsd?>^y|I z#Yr(u?Tx_?ki^ttsk`$cYAyMXykC!`kFVOuVMa)|uRhjJLpYj@mZHE)K<}i*pz_?R z=j}6%ZF!a}ytiEhpf_%Qe#Qo@FSaxIcEO9cyP{bteWf*tv{)Ftrbw{aBgR-vw z1-phIF;Ppg|Fx@zoAy_B#+Zo6iNAA%kInerFRWtu@&ZUm(6Ix;hBa`Fw@20i|#GcAS6vR5g_+;79v(VGiswyZX%g|I()ZbM+#zM)KDX{uiC96M%+g9TsVwDBW%!xTF;zl3{e~L6+s-42|#Z}m)nTP zh;4&?HieujqKWUHug??4|*9% zm&E}w6Ha2XDHHD(LLmw`CKPVoKzfiqz;BR%*cFIEJ<)lAHbpJLJs07V6AP~(h-s1X zc`!CP??7b@ASO-BitwEP0S5iFBOX_{p1$q+0*9w=*(a$rsI~BfdDZu%o-+D9GM?XW zaH?Cw7Q%eOh{K>EFyDVNAM0j|;h&*A%}wP)5}`9<)MC}bI|AlUPu*A*lc)RaemN4N z3Z6*SYMwx$u0Yiv_slRpV*LG8J{l_ps)u%X64pi_HgLoUXr>@Q!p!Bz;K}Mr3I$ml za}v_Bxu!*aru20((N z^7k;f;10ry!&o8Mfr??O)q=C}ggPQQ;4mT33&@3v#i?u4sXU8jS2h>vP;_US31B#V z)4G10wBG~#xSD3?byWXn@>IJuY8E2DZ&{19ZW^o&3H; z7N#QtKVUG>50QRZc#~|}Bv7%B%25P#SP^{1xdCvd;t<``nuV(XaB;qCdrI`sR}MJbpLehOK?J>@$3Ih% z6K!ZcqX-||KJM-IXE#QQ+vBC(gCwrZ^jnDPx_DL7Ga_)bPo$(J85MVD=Tf`nQrnJQ()sEu-vQ8Ia6;qD zZ|6__Z>O<;`tU1P<|}ui-5F5Ouw!Phvcp(5O778s@rBEG?*?0ZPls{Q!aAcd@!Emq zCYF@fh_|)@G`LP2IRMZ~0JZk~W#e}E3T=AT@>~Ph$bl#Y>wag9nz4zhu3+{#qEZ*a z1IS1;iE9=rYapP3ROax;455-K19rv*dUj&$_(Nl&dp6x}P54ctOt*=8#@u#GTwM|7 z>(Xq*g|Sz%1eNVDgXuWkn)nQ~4S1$XSz6tarfA%G?OjECq4ugC&p`cTA2v0$Ei=Z7 zsTV<8p{)!{H{-gO@w7=)>zi$7Hn44WO;^WpKHHK>g>xz`yJsBjaAr9sb^UlsfUd+H zm~w4Bv{hQs9EX!q1m7vxTCcKPw`r zR0UzHZnGSVYmZiXnLcBcm*oi{!gy?o&&T|xLa@H3O=S1tlDOpSzBG{QuD^x~rslgn z9`He`4g_yTBu(eE0JjUIc3o`H&b6sf=4@V~S?aD!O>sjtBi0o`C~vs%Rn7W9mxqOt z{t7pG4crir=*b#Xj3S%T)KLN1$)9uTcLrVqYoHo=jg&5Plx@F(?54!o)7vbT&-Zec ziRVrP#z+E=aEmtCA;HE z>_2zC!*Mde3k^Dgqzx73naZDbN*UfrF%OR!qN$zXfesJ0$RW)TxutpkdP2I9iKarP zDWJV10sF0p*PX`X)2+s&K~96;Ks;u{wN|9QlkJ%8)Hfnxb zo?LA~3P7FGI-Trb#vwt#xU(oW+&maaW8bV7u4V3H(G^l`=5kdp?s$5~GE zNn=`o!0bpfQm~h#VOkkZScNxM)1n{EpkBlgIN7>oRoM0FTZZn9&sQeyBS5BsLym z3eEz^+R;!OI+XDWAPvxqASu|ZD>R)!{0Tw&7a;V2F!wR!o+0DOX*$%PgH{$SZ65VH zD6&NBkLmh>`iXkHoS==qui<6Wgh=0a9&3)sPmrf1ErLn$LdkbnG)pc((3_NEICBFe zYm_Ju!;j>qRL=!*)}wt8^09HqUM;Z}`Ea^OclLerffsKphA%vTokfKqqk@xbLRpA$ zEwfb4PYyX@@|TBPDt=3r2J1dm2Kx1w576k|S3K$jLr8?KIr4CC^VaE?mgT4>M&cyK zNmEWkT}6fZ^=UgKk;JK=d>f@Pm7+V*n-WKWkPAoKjtd7Z&`kq@8&HdnrtW*` zB={09QB*aklqjB;;rb#UliV}l7s}5OztKF2dx}CJ?}o7Cr^gYaK zibaTvd|k#?)}pISVUyG@S}KaH)AT@CdUCJk*Ur}R`3YuqxI|N3Rb3-t8t~`+FIItUms0a!-1j(?4L9|6&gRUjSJ!C3` zK4iSl_03Zx(E@o-^hX+*8wz-&@}TN&^KP?J?S0IBqfNr_sPh5Uv-!Q%Div?Co>cg4 zO`Y_!1g)t&gaLVn&SNs1lUN$*J~X!>J z!lXAJlkc*G!ZOI40aL+qB8W?UkTW8EBOYyv$KguX@U@jH5RQFw*9i>qJOJF5eEugH zZ4d`7+$<`ShbCeh_#rLG6VojR;+VD-7R#fsPbokNS~|WIB*j1khUXyFy?~Q7LD2 zoqpm7!Obcffx;4T`$tUGZ>=Ya5J~i2BTyo$jiDd^U_gxtBjX_L@bBao(no3wM=#zq zA+hQKWV=!Z3Iqsu#T5h>s=eI^{EB0~f$9H*2S1o;fhjNh|W@ElrS^ z1^l7MpYO=e6S-y&s9;Er&AI9E`Wa(l zGK>>UXkG()o*xnnB9i@_Zi(;?9gn>Zt82`n7<$Z5&({o2y&>xv5fA18GvV`}SN zCLl|_Kf6QcPr8G!PC_1X&|ioE?l1wm3jx&bC4s~?1q?6+u8jfmfdTsp0f{ERAwI(j zoFlEPsCy#1--96~Vo&>nvE%{5S_lAR{eMrsbps?_<0na-S2I%sI08BX!gU}Z?6m1F z`)v$<%-8C!PkpfTdQM={UjX9rt-zv9`Kp^JNHkFlSe3Ar6p~pbP%NmbDqTbnseAaY z?p(>g8g8Sf+SMA2Sh3wbVXC)OjlY{1vF$Y$WxfB_-jG|~x-ILIlGUJGu2frvFHH3y z<>@)?=}}FcJOxZ%2)NmDud!md4yB$q>Qx9|u_JVomX|Iw4HUIeM#)iQn(oULXU{OMoD5jcTF+7=m1StP1d|+x8WDms0VUaXeYd_pwXRkjpUpXib=h73YU^yn zeu#gk@2`)82I^=}R8S}j?%dqGZL*Q8uj_13{GLko@9bNvE2O6EzT>0tuStr*{-{e zI9>MFq_`2#)Z+nh2?gPaPCuo+X$0~1*?t)iZ#`x-99aRzICA=@MHnX;WR1&EOxA+@ z)Fm-O-8jBgFWt_r@?(_d{>H8$b#b58F-P1?Sv5z_hP#!tO`~#zksSuD`GNHgzl~KD8}J8x+`yAokI*QZIgu<%0?x1ZdgNQ2XIZe- zSMq!3s0DHDp{HmFJUri6-|7RsPw)z5WOD99DAX4q>5xyOv#f!Y7}J!4k8xq6=F#&j z{Z_0&PT<0<_zdO{4nQR$;K7L!ISe=yn&``LEdzJZ>bPQ!@a%h%ViO=uDHA2p^{PCv3du#5hnD)oV3Y)OjFzBV(%10^D6N0|Z4k%5E_sQYkXf=q9}6!IGS#Eh_#{!0k?GrLZe*$bM89*?u< zOBo!gw(M?~9QcWdI*$}xx>`ig;9A5cA`BX<3_wz3`!_@&B__sFQJP<4Hk%!qP0XX| zY5Oj0<93@o)XXd_GLBrxy9aUGH42P0EG#4%DagCX*@N`jVBPOMPrrOT(VELJ*R1EW znSDLZSE!wfc|Yj+uJqP^l0G-oylu8F<#?1IE|HH8FSet>Tkvzze@+J15L(ja;PQPe z)G1y~-fFj)EaLLk)eVs}uW;I)#*Zp@nSa_#noqeKH7uhgfS4V&CKlcV4OJNG7!s(W7IhI{x%^K`flRq*YP3>ZK&#US9LucR9J{=kWSQEcSN4f zR0zsz8s-WyvQEIUitQPx=ip&> z$#ru4_%hka|u5GyQn2xa{>!84(FRsS;~Om4(S|sfo0fI2xLx*UNO) zmcI3HYG7^EPLwNlzi+v)Q<=}nxeC6cn29&pW*hsZ*0d$#w_A*ssIzUn0#%a zNbXA*F}7UophjgjBsDmuQRaY$^KtC2-n>OIYGRRrcyL;p9!}dPtPonmC+O2B6(rT4 zQy>pXO?y^(M*Qn1X&*~Q43e0%@-JkXgA~}q>{Yrsxq8ldo~1Wb)VE#m*VxD06Z$7) z6rm+vOz%OOK|9(((tDVyEUc&3Gl#<1ndRRi{K?V|-Y7Bjit_b{fzU*bwJT(FWQ0Y? znYKuyb@6M3!bq_c110-ZDmTXID9GbQ4xlO-Vy4}O6{Z}6dN&m5CU1`Y8v@Jtt~~rF zN~_P3-sco>fRz~<(v!RVcB9k2sPPPJuLyCtBNJ+s3r2}F`!=F!)xh_Y7Ozuyz&KsKa6tsZ^|^rh)q@X4;zs}LF0YhRlVPwzC?Ett1! zJF>P;s3+OkTo}sDX!0o@r^!qZ$`}JH>PkYYYcM$`*tb>0B|`%CO6gE!QX(DgQ&GdL zCVtFaURl${Z6^+g;kSNdmpM(=h%V&L=pClKe~r23Il+H8hchAuVoco6P-a!Hc&PLe zOBg4IpnUj)c1n9*Lb`4cx0YWh%HT~R zronW$6s`9h-#<7a3&ERP__E12?y%$Pg zNlIt4Mi$A^1V|pov5#A&Yy{F?loH7Z3)KlB0a2+}!$oBAR$4o%j8ANOh{r%sqs!tX z>M{^qV=4=tOp7jw=ZQ5`pOC)!VPOiS70 zt!vUlTvnr;68|nQ*%ErgA}=Cul(op5XSK9^$#5z5(78)mzX$3hpyMyL*V#%^9II(D zq%1x;ILRgTL4Q>43n}uiQQ$0!uv?4A!7XwqH)#|GYnD8y!H?MmE05sY9!n}He3yN` znK4tgkVqjO--m$b#}cbBtoq_YZW$PHHRWE5+m~U2LF>|Y&eQ=ivl-HV$!Sn@ht>yt z{~9@Pkf8ZSUL}apa(Oxt%x4q9?zMnA_&N81=Qnehd$t7}O}2#bDZ$qKaZ|ze zX`z4#b?ThqAmDk&*CgY0V6c5tzvFQe zwqL&D`3PR#?%VcOc=kZU)y0*^qBTc+uj?8{r~H;FN(U0DA*Pdib@J z0$meV@>}aN%)MCZ^ojH7hDsICGr?%n^@CPWEGD8M4Wi~+unR@&ys_3OuPJ8&SCT9y zelKXYXM~qQw|7V0sGZ@BxiHeSdo*$18&PB)79}qScs2u z2E95ql{2yFsOuW)NQ-W%rG>!4cAi2VWJtDZDRrHk;?33`Ma!aRHnMcV=cKt&R#xT6 z$#74x4IDWGE3iuDsdK-Z@DJ{gXBLKuigtpiOf+cvOI>WxIoorCxKDUv9mM&2aGRay zJ}9N)Di7c5WCdj&>8B)D__cX5=yG3m@k!&A`hvPT@yV&cZ3$C5EVD{jPl`cvmnd-e%45&B$yb231T9NCR?`~C!j-Hg}AZndlN-* zg-#!VBGUd1#ke5Mx#yW|@m6-<8(FqOHC~%-#(afAp?`@uo-#iCM&Z0_@d(c@ccluG zb1)SGjO;d%tOS&G911ax`p*4)nJ|ajXQW$*CY0$n(K8a_ya#R)IM zh}~f^rfet6rkBCCQ{K~yG@E6kix!gC_CW#3auK9b>wdgQ5*#vW2T;G&ZXj}CiW6(x z`!Uc=+w9#i>`8bTF>e<1cOP}wy-QO-FrKc9_Fm{?y3GmbA(+D_Pk2O#*jo6+s1@!f z$3S&*Q}WeqkXtlv&bEPa_3|hHt3mQAyEf7_!W`5oEB#;}+te7z)C(4wMskZCMK@Ul zB%{Iu1hWDZB61Ni4OwyA;CK6P^N46@Cr27s0ze{c9sUkxWWY~kL9k2&X|PQGwiSKL z0>Bj57XIn5EdKWNKpj6&d9FiuP~(wOIoE4hB1CRZ3#`f`cNYxb$khSL=_0IkQQV;z z8;8oz70uKflkB&U22Y_uGm^In!~+;UlU51PJ@aU4m^N2xZc&%Ky!#&s5UecjUAHp4 zMo}kJ*SJyFa|A^S3L|8;@Rn+;ve|-BcyX+=lkQxTI*Z!8KCV0L$qLLPA{ir`*s19z zTb9g#E|k5U9padmaS%j-XbJfS*>woBwh+2z;Ed6$U+!F6s|YnRTMh3U$ZeS6Dpxyr z^II;0huu(8T(cLnT=+J9uBZEWwcS<8XUvD-?B1VVPRT6}?oW8t`IO$vroQ_H-6Bch zIe_Osoj}+A$sdT zpAcMc5dgttrY^yOEs4N=w#^5oJUU8)(rXVwRRZDsHK=J-pHAs2NxaOesnONoY}#z+ zd66zaSLm)kve>AyxsjxXq{01(!!k?F3>Te)u7T#w4yj-?;Q%hbdVa|z)T4qy2g?r2 znk_^$QVJj=YI}gk6ep>{QhEsR!7*y9dYy@0@iGYrDOVjc z3Q&Y1#9q0~k3(?Zzf?^Z;Vl)HW#%pa?%P**>5+3(1&Cn901RK|U0_HM(k?l};LF z7e>r;PYfCkDi$s<9x5(68X6ZotntHq@VwoPz#C72;_NEn0Z;V=jdwR-~u|94^JQ8*YP8(i(+MC^8%T*4+beJ-zlTe|eH`mni? zd&an2u-Xu0GO*_;D+>t*Q#dCje+Bc0!%F*3oBg-K!XG(Z_S^l@z-%&^U8Z|Uqia@T zlAoo=3)b=GgY~RSXl}N;KC&8qPtoo%ozBO`X3|G_RCqm{F#bVRY_fr2sw9v`P0NDk z#4!jz!s^iStbUq7LQT^Ul!_cpfG4JOsE}H9@P>iRJo$G&N?LbD@aq7&%4AH4ew2KX zrzvAPhpN@uSj#uj{H?|2m87X;*`gl|mwDJ@>P2b>yc|L_Yt8n|6dYC!fWi(y)KwFV^+3 z(Oy99zIxt!=H)+nrmoW8C26+fTjy|{WwAU}LPOV@fn60+W}2M|z~&yc#_isY4Hj6@ zny>WdvW?;wBbS9~;y(_~Q-CT~h#(W#i-lvc^n>^#cxnruDtmA*mzt*+ZGhjZp#&YVzs;jznFQvtSW?QbzHb%Uj$27JLf?PRnQ zBvZrM**~|CPc=!lxpI78qLpzyjk)AC{bc)6)1}%i?I-7_?r!b*HS!4mbdv3TxUDod zu*(vpJ5M`924nPHNDSe<&ns{Qq?=&3@>i{cNl#m+B?JtF(CK7~P~rKr0tv&6^5T*Db$!&+P$wWlu1g}Dgb zcY*Vpu&S}cvsbRr8_Q6O$3ug_D+oPU0F&Lp$!(dXWyQK1!0wOTGxTteS7+$2H$!FF zqLtaGtvJmUFo_UAC;)7pjdWr;o61=Eqaou2yYGNuJz6ZwS5G^13n@ zHnzv#Z1a_UEF^}{T1DjenG)UEP8`2^2>ip3&8p+Qr`->!D)xf%fL*7XW_9NC*Soe| zhI|7l`vyPehTBn7zG-br+-#TC`zE(L&&`FoLsk>LH=SEMTs`U2P{nq?<&nRPyC1;A ziNlMsPsFwXVt3X&iUtXe-YeSWRLkG4wgD;A&ou%c&Ts~}D|-|Gu4?-%JZH9C#@lp5a)7~OaQX41 z$h303`#z)Jly{#91*vpiX7VFrZ7(f_uvvFtL+cYM`#fIte8b24QDU_ns3%ChbOMj9 z#6fjCN9)bs%%=qQIC6KD=H71eqz!Haw0bcI&?UxeO!2M_2>LX*z-kSKq437{7mI;u z00ArWnX;rbL4YJWbD#BySbI1?GTO38_vJ``fhEhHlbNbupbqo6k0AF2TYQ6g)?*`| z2*1nmI^aCWb5MB8-!nWRoFcxZU*5Yd9AiGGJV5c%^wjeR?JVw?`Y81t<5TJ~(x>3f zi$O9)B<LVwj)G}1(B?DXFCNQ-|2gT@*6iqZ%90jzwXatU| z#Zy%<+KVYYszV;_09okn1_ozb6g@9$VlWEfG3OK5b2x}u=zHwY#+bPm4#woM*msZiPrRmYbeuKEX-6(7GjI{nV-9eUHC=^chs8wXcFYk<7Dws^tLuO;C25oH?5md@&epTaG#6^ z7s3@Ng@Kgu6LcP~8b~@5sqaXs8XL1Fji2mcNN#vw8Pf5o1*~Z#SUzlxd2`2rjA9d& z2NxxXJQ!PVHke~=xxxTZle&L_&$axtMO9Zp6qO?8x0M05PgX)8GcM=Z*=DzwH}uvpPUP zxN5prUAwPX8l-~EO+NN7{jOMRAelb7)yw#}STX6KeV&O-g*sb%eTae6$>7N54f09lwJ`n? zK|Zct3fbI%-}P;1Y7@_yvAs0#lQcBUVj<+$mdQWF&VOuG_D|*AO|gCNJ%L)*FiD^` zG6p|$ZCqZ~6yG!YY27nqm)AWwgtT^oRa+Y>nvK7rVTP?MZhMYw=k5cU1^NScRjJq> z(GBoHx-fVniC<8o;0?(Teg_54F6t)X_O^yN!Pt6 z)o7(|2|Q8u`6kJFrZ_-2D0oIYw8V79KfUER6w&GbXv0UaBg>G7Y@VNKSlDqZ_(b2a z8^8;>hU!xs$l-eCb+s*XeYTQKC3Y}40gt?|W%X~Wr%Va`#>}jV*oDeAWkzu%CXoza zoEG98M4_ZmmxH<;TW=f34*8q9LZ*yoz#}gwgaJJ13=mtPS->>00*V6KL%q4NDJDT* zv^^ynX{v66&pfE7HMSBg2|x}EJ5&jIOqLB0Cw1v&6=rq8nN4Dn#+$lG8ID6*J7a^R|%|W8Yn9oU`%PusuiNOb07fuigGqO2#(A`holvzKu%XU@FLOz zc$7ihR$j=HdW27~sb?0|L^ScD&U8(PR5>n)jkX?dq&<0`N-#bn=P9_lRC*ga~OX{}_M1mdag zRV0l<0SdtTaYC`!Rnu`F+s(mG;m7fH=SUxYBEPcVroZT`iv>HY^(dEIU44QxG}=^3 zJ2Qrf{iK(0_9x;G$)XFY?}E=!J9C2G0m`A|)7;5|0 zX^o^y(PJt`?lzSM}GwLyoABvcAD%x(^A!3{693(5TrRGkiJUBS- zc8~Q&6IBbWT!8VANR!vH*D@)6y1~&+{_&iH7YhRn3 zEb41Mdv9}kyZ;)RvG*8_$vE%ZB8>_*d!09lHtnH+T^CbliQh!U?a4V$b;{i|lLVI` z8g+-!ZI)sVr2rtyq`4HBNsO>0-mXph)V}nb7^vtvt2_{T-Y=p$f1rKEg7x%-^zs;W zh4NNPgrX|Yu;|@y`5D06cNqv~wG30pLryulo1h(MrIGq;T=7Q=uS#7}S_QQ9>;{?( zhpxDmNep?(QodNQ`b>EXS)A%LX)Ei=Vk25VgI3Tl?A5557oesN%x5M9n3L%zbS(2x zCd$d8muR9vFvSlSn6 z{X$fHu%D?p5UZ)jB5gKTPIXQ?Ry-2#JR#=P9jJufX$9D&jTf~vOM(mg5^J&t(m~Y6 z5;~zwKVdlxp8rpz^dDmQ|B94m{F`UhjUTrQVn7gm{D7jL6AdO5K!PMJ7Kzum&!6=L zsQsED#hBuV_U3buPLkXPa59lS{99(WVR+GED$$*N6pjbKK@slo?;HYw7d&lMc-_3@0hOlM=(4Uf|g;Pz};x32VJ@W5q?7 zsFxr1Wv4tNSDS_pWM+{;rNpM>MIC~xNfLu#lvD9zNns5M4MwpIwBc9@=>O#@~Ye}=P{K?|u`xDNepF_!-hSNQ8T#n*oSME#`@f32z5i%W{g8an(NZtMRO7` z|IZDXzW8YlW;#xeFOK<-4gQxa&B(yWO2^2|#?DCauUP$elKo$4`mf~QWf>XhIGF#` zO~B6n#b5ROwWGgs|Em430&M?A!ti$!I9ZtKI9S-( zSpSgZ|CZPNN)xbg{@okwUx($7k^NT({gwQ?A^#HFe;4?-um1A7|3SC^{p8;!|G)LR z|3_F~QU2Qtx&)kzUnc_NKOM}-@pZia-M{W%w)a2ZkpK81{r|!C{yO{q=TF9;_5Yjy z3Kka5zie-IW`_T;y}zl5KK?(p_r`tB#xF8xLIEKm@&r(!D_%jVFWb9}O$1^V7$Sm* zU{>S=Nr({z6jiVXAR-S4h^inE#Rn1EhYO>>T{GX*Q;bRo%3JZ zTZqpgj?42rZzlZ5+7Q zIlW9}Gn~N$fiXaJ$)PgnKo5rqz5t74vVwm*KoH` zPH((`4!Fw=aMo^c(dqEG`tn|Z**!ADR5NvI|5L^IK1dzx{>@nWgqSVv7lvvL&kXdo zs@3@gz%QHY;S_$d zIKf0((zVyY)4+A3GRLCqJtAOqXgqIZ7?Gyz1|5*I8x?dZS1~RABLepVBAAvJi0_EH znmFT^My@w_g)2TTTyUPFZWyHZ(^9paqJ4xIQ5rEj3vcW-ay9_Oj9p-77khxGwM-i3 z@hyO8#xG10_+ggg@n*)}sNcPeJ5`7)B<=_j6obP~B){cy4t*MSI}UCDEuc+-6kA1MOvDM)+>dL*x{L2qvGF657B|6cK;1OwIy5)x1c{O#bU=9! z7O@iLKvX>;vk(wApr1v-@srer#kSZYQXTtU zNmQaHU`>a68xJQgLd-RXgfZF-9QdY1be~;h#MbDPti_eXfO5nSs;X5w4KK5fE#`GQ#W;%Mtq# zVG(T+Rcmc~+&Fdh6wT6XPHzM12yuk)jJ90Te@7d(^5ybnYnrY{HtwOY_c{Q}xyhH% z%b3;DYrjAKgL})nKI46jSafBRBU3g-agFVUqSeHA!4+A#RW5N~HZOsXv~W)gaiJW$ zQ6a0Z@beve>NAl&f;qy6KmL|w0`W^DsF$+6B_s6#C{t3nh zO7xLDqTq7rI{^SEjZIPF{dKN|Xxw`jj~M3I(ff~+PuOEiL*mtm=vCC6q>`O)dInsC zU!#a1LUWBg5sB~>W#z~s4|a+)PT|-Pzi@AgO<-71`1!p*xOc%nxc6nBdQEI5c>S~T zDmn@N?FnTXVTh1l8-y)Zlc;>Z3YeQ;QQ`RLj1QB#V#?b*eKqQT;ok8PrGCROI!q4O zH_&5>+DW4lO>7LW760Jg%-**@-xbMnLw8(@oZoIW1?583&w^T z*T3(XKg6$am=ULE8@mnZcM;L5cl+B3aGJTbfOum5!M$N0Zmj5b;2-PV{Yx@b;sOMT3yI{|3SHn zmPWJQ>miO^`siMqA2cF={^-Dm?$L#ExXVG*nqC^47IXaTTfrCZ4e|%~9_bAJgM0I& zjb)4a1uV#h3ix!61m`S2_IqMgl||LDc^T2^#&iPo$_4wXz9{vZ0sC65z-e05ads|^)Q|Lmfv&XK`=eliXDaM;s2NJ z{Y{<`rS14*Kvjd#B{Zj0xV*?)gVaqt8#=`GTkKM;p8Fzt1$uGy5`Axfseb)5+b^L^ z;I_BN*-Hm(M;`eCU2%u!k-}&M!hK@+&@pgIga}&yEX+ExK)?Zv|Mx2CmP#F{{wT zdRIkWEF9o}t>n=(86uDihuw1oJE$`7!kwvxLC=)Ra}xvrUo3(DKFRgZopeJ(Sj@JWYk3NcbogMRF$drou7dc7pBv!? z=u3xaO?`wgD{hZkmJxq4OoN~EnErl2vl#7hNe!?Ye#a`H>4sx5 zrx;#pi>h8%((m%~eY;a~^sVv3`@m=rI~7|6uv9OyeQW5LRGVLmu=LfwRX3$N7CYVA zxcihB*x(H028#s z-h04zb>)lq4o7`N_+17Bo38&f5Zk8eBYYYOu#)cBRphh^;E502c; z+wFGi;$!8Ebo^i5nuq*nABBRy-!uQ=y$OB);=R>R%)tV`6j__x>c?>n_OUfJC+HMn zn=p$YxM8O5IjU90241zKD}4w1hVOuTfMIxi7Q>Dgb-VxY-t-dE40w_E=${#f&6R$2 zweT8qAI*E8Cs@z=6=jxSFRDFT#jlE60@mLntZ)1xz$-UCe`|fPTf02FzW8B1R^Rzj zL=#2}-deO{QoR)W-2&qw_{HVE<_=N6;C?!G1nTd`st@GeQjg_q<%S2AGO0|rf1=pf z_g=Ley}rIGD8`Q8{|Y0H%}51b4fr{P)rNHe`tqqo1GGeO0VH4tD|U_?pO61I>NjGM`!RJiXuWRf?z~# zRODB+f`_#B5xDzv{GrvG{sR32qaYCeNk9vHh=9#_J;ITCbr*pV;i3tAvF`3$5m^yg z!&UZG_Kjun*GyCy>~|Xvo(3WAPYSQe78?d+#%tWmAHL_ zBT}teN4`|*kEk;Pc^!DvCO3n2-g|SQeh;^pSLTk9vN+Q$!qBIwgq3F$?dpLHgSjcj zv0P$5;KrxTxR8;3pQ*C8`aASX?RU*zGH#cXuKt}vca=hOU>a38fd@`_4RPdWQ<-Rv z;|5eEw23)Erbcxi-o@>Hx~A5<`dRO_^%JFa%(BK4s6IfQex3Qa+_98fRoyx3Kufc^ zznv+zI1honv32&K5j?t%P**Ry`L5>qm?n8cF|6PAOs#zfonJPm(=&r#`99Onu_fe% z^?kwmwjBq%p4NoAl0LHO_a9{$)|L6s?s%^Z%@ukNa{3#mpWlM`^eb1gAkpRxsWbg- z(XYJtmXiFC#+AeN?FqciV_)t#-PodzCA)iW#}&o#MRf%+_t8(Mu1hd`>Q=EmI}y8% z%#74U$R4a27T0=maGwF^t_0f$+(UtIUs^MoGQL~1$Hf$j|2B_O+tV~4;YV*fG0%9+ zffTz|{q21A)0-V+l7E*niu~A^y>yZqezAVHD<(QV`_}?U4^O(aNme|TX>oI2$iAS5 zYJNx!EK-Z&are2nHhPfr64=oKlyYk|VN5kVaI3(GF#&CZ@jjK({XFI|s>~{wdVg zea0K+H76^<&%XGI{{@*x%pURj8?NhH1Vuj`iM6I8wtOz65I9s|#9q4Rs~BhoD>2g=@Lwb$xj)=!Jih z3E@WPIdY5+J&gHgVC);Y@g=SJ^FI|^FgpP{RKmrxJyCqJG|bsv-n=gUoN42I16w?K zkP}`@1+*;q!oAh=DgWT!=wK%YM-Uz@Y=AH1y!S^Q{m+pE*Z2ZO*R=P0_#rfQ9FIUVUR|IOh zC>92ufOMABdlUQS7zFVkUk5Ummy{VWmZ;m8;jtj5gs49^)o>sq48;ByXa5*uOV_r6 z+HKpmZQHhO+wRr2yI0%hYTLGL+jj51pEqCf{oBb-Mk425uh{w)N#`HlF&6ITm0Qq-2TrNYKQU}S>+~IZBV)+_UYZ|2nR9SvPK5f)dw+3m`z4bO^FCr18`IAYu>$4|T&-LG&M!A8qPs=h3@8k+F=vrBx zzIw>~z|@VE#>w<(wAW;&skJXk;D@G#W1LTzd#;=4w?2m6e0yz+=)Wl~3disU6&_7n z(Y@x~2;&_6kN3v(FL?Qn_qMzfc(>2VXT&wTQTZBT{*H7ZoG1ML<-O4r#PxgI;et5*Y@5G<^Dmd0`+5C%)BCNjDrj@P9?U!26gxy_CI{ekpra%s@K)kN0MKclLX>o955E#H}$` z1iKc8|3BV)@R#?-a^(LX?>#-@1+k`wxZZO>MMID8LHEys=V$Kp#Lsf;Px;M}VQ1k<>E#yh9sNhlL*aow-|5Y~%^w3`qI*|( zBjWw=fDo86YQ;-kQagn41%fZ%p?WU=#30XBc2kD^;SctL@SJRvy#?Us>p1xM2ldYc zu(w0`iv5G6U*3D>1sooA<>J&tREqHn))@g05l{jmKUKMD~ibw>Oz(C5yG3wYVp zjJ_q&oyacdhoZY~8g6fnxJ7X2`(oslnk>JD7wc7TR*0``95#7yM_7X+`17Z)kB^G? zw|R%g1M|L`o{u2UcWzz}`2X?V9Gi*|xQcXK#%t5-Bb!D=)i9r-9w;9hz@hV&3PIS> zAA88BH^D~l7=sir;si}ZP6!hSal*8`UvT4N_!@Y9cNjkq<$t`r0T=w(8CJmm{`=*< zh49`7Jw5pK+myX<@kK_7Z_aFw9A)qPG-CJVB81}tp;`Nhj^#(ZK72mh-D-NrFFe1Q z#=DsMSuG5)U}%4J3LD#&S&;O3;SoXok*?veaFQ`!Z#hS;r51Du6Rig|J4HP)=Wyh0oA!9v)#evZ zs!-;Ysl;R}rlerO9jTuMpRT}NwpIEwW9Inhh1%>t-#fP}@~FV)*Y`ee5PanMdv}1g z59O^k@VxN;pYL6#r=R(KyC%7XR_`nUbI~Jyca>Wu^Ban0fOklb*?lw&lG}3e5bS*D-7HUa z4ZZq+^V^xg+jX-!`C-YqJ_GKXZRYa{_r-)<({c64HamPa<2`Koo0n1lB}p=VmQ~-C z>ra2OV-3KvpE`hbjrbU%w8ky@?$y(>`RxW9^bI#==%MSMib+|LV$i16?JBG<%qNf~ z;1k^%%W=hBO7#`^qkv9^7{g@&w>8<;8oNnx&1i?D-aB=q z<~Y*2(HU|>0RmmuPq-K9AbmKfz^K(21v}JTKo%2?i^k3K{IyD4r?q+5s=ohsE~q?zecy*nhZt%dgjO-f!7&z;EK(RRi}9u7!9>abs>aVb9`n)alVR z+DXQ(^K^Bu^$aDeZ`#4UbK$D$>~i<=`0^s_*k`PbE@wFEZuvFub@sK)!`ny6$H}Mi z_4_VJXVKJ|6PYl9ECbm#f-nLxf-wRzVhP!r+>9hvVp2ZqF9)@w-G2X|YD6Yd6p}mn zwJcBktK$CMAUzU&2lFPLyRv(Ycb<2xcdNI`Q^Hbi8D{r#o7*+6SLt_&9oBgOoXqb-Pa;6`1J2&r*C+L$j5xR;1-VO=R*NqsGT z&psb=AA%p2AF0o|S1q`^5JKIm&PAM9$#=FG_qGAX#eFRdLH1YKy4mo=J7~#wv8;FM+C!;N)rast*c%eQM)qG%YXx(nhxK8BH za=wR|CvpwkwAWdk&M$U5^TS*_INLtM>{r;2iVi)4{(*jD!r>%eot6|G_cF}0!{cdg zFN0f!{ZvJ{n;)WkPwfEguh>_BHFCqN?-LZS>caoP^kF-#x~$r)x-0ZpbyaJm)Q2kj zRj;Ws;;3zMcK2PrxUTkZ^s75w2jd19K9}Q8iNDf(>vs2<0nkHrJBh) z9C{ZF^Dbu^@V|bq#ws=nffBip5C>$xK+jvIU*3DV4^{&XdsqLKc{6aS7LqSGs1_5f zS3m>6(*XpmkB|dqWS7DRg!C5oT9!ZbzyWAGSbUd@2R^r-+yN=H7v2HYY=8j+YQT^r z`j5CF>3_~QR@ax~Ek< zKQiETuWo%B4WaZHwE~nL5mLlxUV`bNYJFA}0rePF0RpmOM0yq3zdqa7lq$l1N$jIq z^|4jt*rQ_g5jJG@iL^!WHl(5^*q>mf5`gK#%}KZ>hKcP6&Qs{JYk# zp{o%=9FjboB77uaV~9@4J|a@1jQd28Bnho!j!uEUDmEecSZQsh3ZbebQaM=auC|jD z4Z=yd^N{Dh^S+N`z-#AgV5@?UqK~8wNjD808qy%?zLJNqkEjk=Hx(VadN}p3-eGme z>KOCd{uz2zFo0F{xG4M|qP7EmC`k$T71u8^13O#XX#RsAbT|F|}2D zllU5S2Wy?2iBl zl>-nxL?DGkAPq&Jq{2|#ko;iMKIL8GYx-;3Ylw#!4|NW556Vmw`e4+)`u@s3`#$NN z_FesJ_A9T|$1mpkTIeC$DJPdSGkS6e?G&1&8bp$bLIEn-rxcDt6(YH(6pTVCD$S@= zg+jhY8g`hxzhuBF0;^hm4*FE^p7Ni|>?Qf7)>Y+K`kVDLKKIuMH zZ1xY6y$Q`IKT%pL9*ML?MYMv7RC1}vgisMyd7#+GGV-yCMKyy=Hq}gmxma`Frd%L{ zcq-*saz^r@7m~dPE+;KZV;9BDlUX)AmNbe|B{xfn&a+u|Gz!Nk5HC@%YQ-oT zFL5l@aH*7H{cB$YVO4akR9&q2g#Hw2qvRK~BQqbHIzn@j_Uqh~vZHND*ciV$e0BnF zE8LX4rg2Z`7~eX)c7ks!&7wUS?K%uQ+IGTk`?o1SeWb6Hv}k2juccd^zpSFXsPY8< z6zQeY3iz->>Nr1>krH z)+XskbTT?~sDFsoChJGqldv;}Pm;b#=q^clmw0|^3`m-+6nn3dN0d?~gK9;@S|&BM zC)X zrp%1#kDnYt-;4el^VVNX{7!bsoDezJ3!q9N<*pyCL557K&tj9JCF)Rs1zhy{6cD~ z`zRgz3RzN}`}N^c1^a3h6sjpC#TkiJD(FbTBE_Lf%94soDOyq)i8YIA=M=6fUQ)ir z0rLtD(5lH|QlaoXk$d$C{betT9Z=N#WC}0 zmZgpATokw|a#H5SQS+*2B`)e-6uc=mQA0=uby_x3vGdZN8fVpyitkik$vVY7Oaho? zhm=9dedEOA2&QFT(wHSM>S7ecDdH)L;})jTOlp{=(du~0c%=`g&Mthq0=cF1vmuY7 z??}2u-$ME&QD;c)$=yP!bLwYK?LwYHUDe+e-zDF2{wg~P`sync7mH8lV6KJjk`n4a zc`IkIc2MjonbT8;X3h_-Z<|8bq;`-mRM%j7C(F**ZBv`SoLNiI#!%N(*Mx5`?y=OV z?6vVV$_wZV=?m%$?F+DbOh@0Bh<^e9g8!WMBf3}Sx0FBfIW%l==(*XGtZS5)cvt1$ z3O!ZkgV?{{f;Y1rjmci+!+_!(6$-i*Bqk6spzzCfR3?zL!D>R-`9ZUy zr-YCDmqPl6e;70{$atutQSnYHAdrJ4h4{S07lJYSY4QcqsCmd~5cCA(EfZ=rG^l6} z5<`yqr3_k=w+63`;Ok1$Xph6Y7)}gb zE8C}L@TUjMv(`rN>sr#Nk0XQ!^UvX!S)6C$}coj<=6>oO3+GJ<~jcy|$4IBS`cU9}a=qBz`X^>I_KsQ8KH2 znNCTD2(q!?pY^1%qKrwhD9Ti_%XzkD%z^8gN+6jW#!(K@1a z{O5V^)6z)?lFscntJJA9k!WgZuWPJpt!q|jCekj^veK~9QvM>}w9@9%=+f%a?9%T4 z1HT96x99uj=jJ7G==sR_1mP<>7PrpvT>rf{zsI)AZ~g0*hM$8!GyIKyD8`>d`i*`l zB|jZ`ME;F@C`CUNeMkL|^roM)KQn#D9e)?5*w78J6LO!P=Hwtpk3-^97)1ulUvx*| zg3MMt{(x)~@?Q*U7UIt5mAVGp1KuN$x)B$OaDZVo%wo{v`wERbzV{w&pEMjU~NO8OQ{?xc0ALuO@7nvr|-8-{=pBi z?srZ0!XEcUutqNClfwy3cE#;-x!5MXVz=e}I5~ax3qV=?%(%7Pvdh}EJ{zWSC8_y ztKDa4T(weP`{5nidZl`&HesXjZHhKTJF3}sr|zM8U$^7s()T0V_hTpIcY)C1b~Y?5 z4(F@vaPAaqom?0%9UB=J6%!E;4GRee#jF`@ZJKHl zk#s??bF#qGaSd8qtdP4M;f-69m1@wC?iXXee_QaT5I9*=YmZDmu{!;6YflllG+d3s z`~meDCAFkYEX-(`;?~^e5e3bd6KAC4zv(rMoYbRnM-;t;&DV-^>bSssxYBJOys!*r_-NKf{+hGy-+1q2-7NvOF z)$)D{b+w0n33ap!LBiD4PJcba(&@FShJx(rEUYDH$%cR$z>r-lEeqM@C;{fKbfG%U z{37-PjBznwXB_Ry8~QYDn;^ipv11-}y`Sy15S<^KWibP$=1f%#zCESEE{biQGR^17 z`?fh9!H4jCI5b*{v@;5wHO>y7nj~Fkr&~OYRI=t#T8jYL8Bwz4tMlIF)5ok!Atu~R zm)c+3`)!zgbSbO&5Y8Ees-~od5Ol!F(W(FnDt|(Gx}c0=hVF_;gw-9zWx|aUnz>My z^Q*zAIaEL}wX|8(3n$^_WU;J!3GLMQav7i!kwJSL94zJNjIfIXy6H-3ShL^j&?*K< z@f72FtXFS12 z9`b#&zuW+dNzcLHX8FDe)2}{iLsZHWYrnh zd=mSW8C79uQ};FgN?MxMxA@Qyl1u!0@VEO7H`k`CP3n!gPJ)$P+YP&yA)mLWiTh{; zLx%<%x(qBrtf{j$isY^ds0NAVws`qL{{p&2&!NO5Y#LE`s`bYO^Pvd3SG?(z24FJ3ej;MmKZ8cZ=|1?RX{;sfynJd3i#?}Cic#8A{i-8) z2Tq%sK;1{)cfG#T<2JuO%4N5i2s#@^c5qMDU>1?$yjhpZsLAurC1EqytCX$@u-&vJ z+75^pt|_O?V;kAt&P!c5mTRy7v%PU>fNUY*>LJ{Pt9%~m;03yKCVm5ad<+-?d1aqV z5ivAmd@b4hv4&W-Q;+8sR>RiP%(m($q~mH5IavYR1leoO+1!J_2KDG5>4+#&Rg!V6 zcS5MmY!rz*O4FQep`m0O*vk7jYo}1Y&U-H?2GIujK;XYxIb>S)T$zaQ>p*{XYal|P z-IOCPghB*r1awU}^ob!9&dX*p)mDir> zRi~jmPObd9P%y&sEjok7@*^|Y(U4=5;quj@6UhspD)~J@WG2ixDFsSuxL{~szT-KH5*D#e}qrXOy;s+H1@vL#KZ1bkB#H$!Qg|cuUdR_8NW_TKj z7wB;40r{m+ia!^`#Vf@>duCYVC=oXsB4OrZyH&OI9NN(rP5|c;h8t=6k#UD5~M|K|wVGqqc zjlB}QmjY4@Q}h>SdV_G?&4pv5-z{>523&2_h#4oe_PxRLkRjImp5%l z)ypR?IPzFBwRaGwGswevZ{@8NB}NV>{CK9r2$_ zB>o6vxgz(a@t-h>JE@1{NuL;r_~Q4xB>wQ_aaE(_Iae>D*d9a@--wAl>4$N|Z+3~f z68Fdw-9d{FD9YxF7cSW>D`U-MGH3C8ix|~$DO?GKa><_L>%qPrys@*7Y2eVD`@SK> z2N*FsYP1XJ|JE36Rg%;meXCyi76GGm7lH zF@9dGpj<>^eo6)X*}?9UPVW zU}mJf8Zds07VM7hK43VmTkmM~Uen(Dsy6~#=HY8#! zxlp*)ZGet|#Nn&QQm9%wGcgtrUOU1}%oKzL&SsJDgr2H%kl}NK?5NMt?4Y<&vlfDF z27$Yb5NC))jU)brU!>}NKO-m z^pK<{Hyo0l0vv{OLfx{cPKS75+y)|6n^hoCphu^u_~^uAc+?y#ZIRQ&N+?K=eErAK zE8TG~_TDs>#3+4Gq)SGF4`*%y2XeyP6bk!7)X1n-LMK`;SVS;dk0ey&Ow?_y+?#cVA`M^iiB{lRWc+uIw=6qWU! z#p?PzH>IK4v~>OH+S=MpQ@e@#YiNqAE_2DuwyEj`eeFrZO?p|4_kZITChek~HSJV> zR>RCV&WroEV5`EK-+3&??VJ(1CdR%T(vrYzr1WU_sJ0vssg1{MmrD!rSG7Y`b9M>$bjc?fEd! zbKWE|kHLnYZLZZ?X%0z0ow7?Ik(n|Trm{JeS6(_k-S|MVHZl_aJL+f(s;LDgt&CE* zn|t!Z7{q3*6)*o{L!86Ggg1QT5aMXtp~wQd6-_&7_gK7o&p-zb&JjzJHiAfUl>1Wh)f=g+j-p?8m3vfq^9`8RfkNmF(cT4wb(?6iv3#Qhg+BKT38=lk1x?) zVOBd0K4NF6EG|P}0Ly)5Oekyvd>(O|AkQGq=r-WTK&E|6X%O`NGCIp%hT}G;f%d1nmb9%@_25(2#4 z^Bl1G0O%)BBnojQ;;}@;7EH9&5g11`-DZLdC+tMsU7sjMyD2gAPInF4AKrx5ZAWPV zOowCZ1*q@o8ucY`jdI-6aLRV)Pg(H4Lw#jAyOFS7vQWBDjzp+GKkAW*?ZzbHsxVG)4^y|1kcrEXTF^qlfV&CxY{iu6Urz74tDJq`^_3IvG|)a#QI&>@h?S&8B@YqWS~6!x+qWcbv# zM)G2|q>ZG0=q!xIa>PHH^ZB%wI!r!k5ksW!?%cvajB!xh>K^#@hb(B3S8`ZAW7ijJySA`*1kk+yLAb&Y+!mMwIhf zfGtn=RBV3KJ}cDD0X(_DYm=P;HFRNS59|Dn#v}dkF0R)(UXHBa{T9qhH1W1fLl%ZK z>`bBi4_MNbz}wR$`wK~cgAjxav2SbfKI}Uv57>w&a@Gl(;Z58Ed)xD;Y(6w*Al5C+ z(+L5GgtCgU2YZ_uv!xJ^XyuXwPp{5psT@X^YmwHT;{$WqBiA75FJ2!?2U@*YQdpi zXY!~`tel&aDYjJS`so4yPiyu174=5Yv>ZPOw>ARomb4rS9XrwU8k*G_kda~L#011w zq5nw(J*Fl)#zK#gb8MFyB3u>3oPc?T5#j%4qVtSakIa5gvT-R^1IhicO28Ol*8Yed z;<;h-M@`)Z{J_GVmI5TA3%ucE2fhOw=K=!pfzzBG91MV-Pfk1qAU0??s{p zgku39e1Qwt*#2paXk~mNumk2W2N;|Ib%+6z7-hR<6;Ndsn3uccs__|P2ei@wlDGni z4}Fo)KM}^g^zPE*%ZvYEwfcxB1mX#P86D(fvjm@=1c_fl&l3gz^OxhVS}LH{!RtCL zki`#>z{m>#*9_39a6c4JgQ|)DXlZz$f=r8t3a2m&CF~na9U&oQsuaDT+1b(LX*Hu)4Yo67IN1|L zDGSQ^1EvE0gd=7$7X+cWESYPh^U{9!;UQeJwlCBA?_Cbv9xdHkjj)-SxAxglvu;BR zBoBhoFYa=9xF`jC4732uStI9l>i$Hx=y%0|+4uX*?mn}n-$nm7xt70hHflqx;?&pG zTS#y>+#wr?eMk@uK`J!Dcv+0(_Yu7fEPX}7Kp_b1@~;QGo!V5QN)_hg1pW4fM{30O zG>bY1B%8;mafK170d%5CW_4PTeGn=jlahR1BT*i`1KV_;<7P>(`pG#vU6=2a8}?03 zLb@V#$|Ip2mPUu(me*N9@QwmDA-W#da`(zMrM#-=Xo^f?SN2QdjzsShbFsVJAqvg< zL-WeD>*mTgL0lVEwN+u{m>7YCAumJqF2wv*ZQupALef`&N+*$g?c5WBbvs2IuPdXH-K_DEr422l>kkS0u!7_PTngm!HMHnK& zg)}U;5I+WJk|F8zq*RJ5m)u*mbh9-_0|%zZ&>H;`d9j)mmz-ocg=)EW%l34l9Pa|w z;*pfp@T6l)+~p5L^F}brf}snEl%1>mp#@h);gzJf%{(ZJ6h;gv(c9bukwO{WQ=M*T z^7D#SaX$u*39h*Xa)wBx`=vo*JvKda-3tzh7ZJsUP3XTU?LYbvZ`7V$sVyI;^ZOTz z*GOchDDBoecLDu=S-BeXzSO2zdyJw(cR~_bLr66!);kx@eLy)9Y?x96X_`>M=~Eu8 z4J;CU5fpvWklVRU2v-Z$A2%-0Hzf7ky2F}Fz$?576;$=VtJ+YjlV)zFmZmM8?lV^f z#PeuDMoJ>M=iKX}5W*mYRqLWN0D4)H=H(4lVZb1wO2ifqyKecmM46)Q(AT)@-1X9r z4wkK43?3qBQ@8Mimwcax<0DbS?6_`DOQrI>J=R}?5f>}+8LK7mypkUGvn75T zo$HwYRzALuhwKy7yXbdXB}3y4SHh+Z9{d2wZ{eo1A)7sLvPO_JygB&Or%Rws046;#0i8*xU!ECEcB6A7$+vs59@38N{D?ZsYFlSf|1k$9>70Jy0rs|Dpmw( zCCJ_Url!?zOz4+JpXA!{%zc|p5Pz+%uHNX7;%MV=nv5nQSu}`%fJbf7$|M5;KlQrC z88sME(20Z_6Jbq54Gz_9wx$70Mld(zjG!*?;xV&^gmGUs3$U!aElwxPORqM8DE;;( znblteu#j~&RtY*QAS_`&QY;Ic+n*GO!$Os89SCV)b+{vLXOu@)%wO+KYMtqPRxYRU zG(2lqdl~1n#{`S*;~D0*igS_R$0lfHI!|r^g=Rhgk%7oU{E1n7NdtB_nfG%+$(HeK&m{YkJ-Ks^{H#rJ zIfmJ+J-PeW%tVj+%9)ZOe2PZT3~W&b&STiE*Zi&bAK+6y{24PFIXUCvSxp$TUW;A! zI>mdtqDYs(LpXFK#ohdt7Yol_Q!)E8t>^oisfqfnJ$`QImerMarT2*K_9lt(?zbd( zH!hFQUVY&1fxgaKR+r&ThUS-be$nGB-Gd#Y3B(T=pCL~^WB3vozz;yW&23{it- zOXA1trP>)OSyroE-P@xqc?FdmffY4o60#0`^cs!tevy}{;15o{Wb!Ji&fPWM#uYtm zZl~!J^VyY9;G*LYb0};!cKw#;(G+CV#VVmK)?bZ&Mll?kWseR{x* zT;}nr<7q-cr{9FtaM#zS-wl0g-GZV8u+1r8dOW9{{8paL1HyySgV%#$!)&azxo5&j zpu%t;jZ>n55+~$j*`+uKU+{s)G|-~W9;qGt7BCZ61T`<&t|twUxQq`Y?&n{EG@iQW zb+R5(mNzsqUI$(Me3_x5oDm|=ech&FA9GyGfD|T2C0fF!fc3HQTF|e+L?k52Avw zvxh$CiVN5dcdS?rJ~lRB9inh@V1Ms|(|E0bo&8F|%;Gu%IQhUVW5C**rUXY?;3VN? z;9WtT0)pwNvc>duJ#Kg1>~VjYK)aJIwtwF0`AR)1TsWX{)n9UZu#I)RJ{b+r5V14; zU?N69GLHHAjDnLYWokKbQFR~)@k9mNL?5IQa-Ecmh)fkqIb88+w&~ zw%AW9HiDN0yG?yJnev%4ViQshA_7ema+0V@OIsEr9E)(;wUjIr9Iq08SV;&tNP4XV zQSU$+Vm*M=3J%Z>o`R7ImF+b>of6C8wOHeVTg z9!4=-&#DT7*-hJ={Ky+jvjZbboNZ3LdLt7jkyw5APoA zd^M!*Z9$#ks`RuL8JVNM8^=C8>~F5(;Cc035EtAn!$XfdN2lUPrPIxXbJF@_p;2&RL{+}J2wh&! z!{R}6KEXtosyMdd$o)ke^;#Pfs>_`_G~s7Uuy{oA&vn#S*g*#2MjBarj2Pq%f^9+( zcYL5dPWbO{uP!}`7?#H{eebJZJVYMuWMDu;&jZ7nt}~6yM*5l4vMgd{oNTVvqrG^5 z5sP_vF-)@qhiLCQpv|mcFj^ZyJ5T2h!J=|R1{ac&y?8U*7I8@%{${S|17O|chIm6d zmxJ+&`leXj+`s2-V%`fk_kzJ~>b#WPX146q_M}fI_Qe+Vc2FApxKKZ!e&>yLQ!Ka zRtQ8cS;sXf@cik-ZSqOaT6?8E%KKTr;f?%1Zi-F=AFVvRu zV4_q`fcdD1Unh?kuC?DJ6QK_rN_^WVjAta*6wd$?{bylsP@mvk8Ko#=EQJ_OEQ**F z$EJ+$ej<)pcoYD$o3W{^QtZOPGo=V+Nt}nx*g+K@_&MD5do>$%6I#-Vdx>Z?^brPtcU7#M@R{nLD1c1zI+?f*=l8wjeQkUW-P=qdCR=+1;}WKs+`~Vn8xI=xWKJ{ z{ve!+gkOtk*|b&?*c3;Y%ds4c2ORQQwj)X|UA3O5mEHAL8VI#6Jog=OUmznP$i%msnV>RFx6fxg)Z(X``DU{1tBp+L+#8%>JrEa*4 zMw!O|3VN<>#y%%XrjlAbmpMT8eDf|-+Fo>>lfZq0bhi6xEieCS`BcLEczSV=OQXeq zm#$h=Y1?0TMdr`9jx%S>k-*<0U?bhN^zjtCTaf3~nE;OG-+rBPThw z)kI3_VgIEf+2al%^HLP8Tv`FWw6d64@u$-z^E_QE%_|Ko-7T$~O|pVgz{pk#vw#BR z_FN2^JHQDNvvWI!5+IPm^tR;PN|-QY=CC}7{FAD(M;fz*F5i7DHYY7iQ_E}W7pk7; zhHM^~935C?{Eq#LE17~3ru41%5m!?CVmwCGO-O&9WqDRk6ds(4B5^1|Bwj3%gO2Pd zs3jPuEcF73P#iDOW={VF(Ce-e;$y@+I?Fu=fR1g(p5>Yrc7BMxT|MJF+WVxd{vKJ= z)#dQS3{9@+ePEbD4ob?NAs3y2Aqy@78(A_i^#yD}zlpKX4yXyp!lm@rqgE9X21Hs3 zA_CY=;=*Z|oB5K=Tg<1Ec;GBLsM|3v<2@8!y7tL+_t`xy>M1Teecg^Zf6_qQt@u4* z>k3AfiXQZQBZu!bfX^2qqtVOtf-%3dWn~ZZ2B>VXRf3P)(Xa{Wc~Spiu(cSmgUFxv z9Wy{rx#2==ZXbz>h0sFdBQz#aAAi>2G!j(0Uus&%li2;~70%X4+(Uo)$T+#B!9|J# zls2FI^Zaa#+cuV=l({aJlsC^);3AyHJSrHiTKoAFje(GofQB)mHN~p#x`<2B6`Mu+ z5-|b!qqad@DQ=mQ7fcjJ3McE)V=;KO4swCGM06@9`ylHq7N*t}3ZRDC8CPbRs@#e^ zr9x>HZyhQ_P`lmSyCF1dY?|jRbTEO3E1m3y>?$8>tAV*#HBgmEi81@UZk>xYG&nD1 zIap=G!llP8=?OtL)qbW*pB8X}dbePLH?k0}965c$edksiPL;*cEt~G!`Uz-FXc4hPhNOA*2T(0cLbN;XBiPK>nnwp{s}W;xBtZr!-Gb0;loEukZ74yd_Mt zU2ER4l*>`FH2IywIKcC-$=74_L|H9BRYS8iV_PB8SK_9nVjWn?_v#WKju`tM-)~bFNL#k6PQYt&> zLmLukl$cxc&(iac)CbaQS%veR^L6)hm-{mtaKsbB!Z zF}O(UGU)39K6h=$a-*esP?1K~;>G3a*>+VO?Y`>ydzxY>lY9P?O>V>4?w5;?FJR{z z_tdtRtQ((?mu%+2gwK){z;4pW$o{|bz>Heqfc=TK#d^i`sDC4%pRiA1->EMrbaMSf zQbC&mr}IlLCJt%e`IG(2ejFmA`^%!E{*oq`Vu-pz;>=J(pmBtR0pzV{U`2eUckYX( zQe+ID_`KrSGDV{MN=MgKAskrPp)c7*2q<(n-rCPxg)%9P)*q?K%uX&amoR^%#bJ5@ zBgu9UG#;Kw;s4YL*(>A)#<|Aqj3}!=@bV1p6n8z;(6$Sm27QAgx&{k#%_H@_w=)!= z_2-1GGzc$P9%@(PW080hxg^kN{=3<_SnqJ>wZ1`Ik=rJk$?7C^`|fP(iWY37Z*lv zHlp;f%R8g*bTPHM-UU8ZLO&;gIe2i-1CPN%hU}1)VK|IFL#d~W_NRyWS1k_ zQ?6J3WK_dhPX>Im%0wz+-<|@gTfWa_Z@`^$FETJqvcJ6qd4X5#5(bFlZu6kaB;NK< z192x(#5r5Z2n__Bm<@J^)map`)rkne~!xTE7qVm1b%zQO;*c>y=Pa-Aq#<-l#VTvO4@QemGjCe#UjL&q9 zAXgzkXPYr%Q|F(+W@f?nmk?LPP#J}4I-sk(jy!N3?OH&dUOw3$72lj|RsYIsaB%2v zMT6E&P%%M3-RTspBSmID!R&*sUUu@vXkA5zwhyxx;K#2T@Pf?UgGI6f+f;B5L5-Jl zLoml<@eL{FK@J?Hj5$_-^lk$4m^o;AQY5MEJ76R#y_KDEJ(=^ za~2mKMa}nu#DAC}Mlr8~0f}xA_BSDmJb`?mE(mGh>j(!QuQ`c7E`-AxZunX;a~e*7 zH5iYJGkD#3pUwhRb9H}K^(m}YcjP~zn!oQ5vbvs@=1*t*a4Tg3&nwP%#4bg!&xf_b+$igwOkLp1eO<(DQ*qX0L=Y z^AKo*{l+LjAfnD(IN!Cwu^B_aJ^}fiPANrl$NH_v(PBEk`g414a&p{Nh22*cC$2pX z+Ar18+73zwhGpB_j~2}v{xx*?f}ZGRZMD3N**uq*oSTcoV80w8=YCb<{D@5yN-*uZ z=9@aS938Bt5!~|RVMa?xMvn>XGP5Fmmq}N zW;s0?ynpy`RY5(itDpMdnIe(<3zf=u%j!>D`>|>N$f)_EIU_Rlq+KZBi#SxKkomUa zmjCFmE1=^sQ7V^|)qdhPZX{lH<_!d|;j`#tH4yxJ^St7>2qUd^zx`^qFW~3#3}2t; z$8GU`Z1|E^z*ZyN_Q$*PL-ML}7{tC)gqGe72?TFH69GbL2UbBEr0CBgvjYy+Avi8b z$fXfVqmm?($q@UZ(v;|L1(%c$e|Ft(10RJzxWew+DdqB(GLRG;PtFt35v5GlV)2hp zDLsuD$@HS%{j5CcSo^xP9Ddr?CCigA(kdk03-#1a9r2z4p zQVzF!_L9NFMXehX!bm|h^@eSXhzJQG5G;h%gv3?kV0#8%VI66VrK#MLJYYH7gTED~ zB;j*-#=}+#e}{eNv0A!X`@;T7_HA|N&ieiyK-f-S*YUWUZoYWdQR1-TExK+!A5B;J zVy^v3nyxw73-R{?cYC7SdeT?D;#HpRaa)IlznSveSYScIvf#lJM=B0?=+AXP>6n4? z_GyB_MgDCV@QX;)h)e_>>?50T=IEDW!Ru#-LnI0ZA?Czchd`PFk2toz?CYVS$&{O} zpGlk*{sF#6#<`zJ;_wUemO;%pcX7>MuVFIA3OpN*?K3H+4Y+iaSxIo<-7k?bCPfrR zpyU)uJXMZvj=RZ^y;)nBsq6U2%{#`$m$$vi4P1XkGAeV0fM)UV1wY1xBA0H7;|d~? zZCP1d{~%O-?rxmiMkl}W6`b4#A8<=-(pj>Wbp!557j2A;ZrbP*s|969=>=No8`*_0 z*dpU<1DvhOPVd!#;97m-3d`;}s&$^6x0Q}FIQf4Vd*|@XnqXfvwkGD}jcvWLZQHi( zOl;fEBok+1+nm_8Cg#bv&)Iu_`?=>n_x`uKS66ja*RT3n{d83=#lxIXo*nP0pw>07 zc+REenOV)Ht#n~nIZXq#DkiuD zjDe$42fA=1p1v7d2qtJTgrXa$U@>V1Nl`wxu!9c;4j94qUD|w-y)!T8ino?0G}6z- z$NQ7l$HyH6LoV6#6xNXx?*UICP5hNPlks>u>EB?74@L-zBD{dvu!Emx$@U}(ba~Fc z5Ye?jNxOOlCNlS5V(e8y;KxE^ssdV=?)dyMAp1lN3eEv|J65qJTwdYh1$jJ}+bFg# z7Gz3-!M$haQ>?3owwYibdn;#acU=($nUPrkYX;Dov3XD3p{e zjW{zxQ^|nv#Ax91!Q2gy&gHR%>e0>$O<||jwu4b!tHep6eW>sY}Re zhz_HoaBz9knZlOnr00|gRy{8*D%?Wp1r(+ekxtQq29}76SO+ac7&I~xJ8>R-f#CEj zmRK(KL+7Htqz*%)u#U^v;0vSzlpFxp2fDTH3I*Bv786pJMs%t6?$fPOy3af(`B0l1 zo~1fcC~swcgY?C;TcT|*9M{@r0e4ai$e^(=RCMYNvS@?n*74X-Xug8Abah5R!=kHQ* z1~XI%A;6A|vU!5Gwu24MHv*W^$!@`fv+#!xNh!F?vo*4k;<@Vjg0t~taWZdRJPX$8 zA4jTa0qx=C6d0Z7{4DY5gjVH(;8gOoWq9)<6t4KZvJ?r}aJNA?%tGw1ym8Lu(^O)= z3y|QpcW2c1tEUEMF{j;8!nvq0bfWXY} z)eWk^0a%EQgw9w3n&3%Bk=$)n=9AN=QJyL+XoCE#Wx*>OJ?>a?(OT7-0Ge27gzqFq0Ks-+qij= zROXjGw&z4Ku_ZrnY%2nw4ed z^mBjSn=7IHKr7nY-HW6sJA`qLS?The^@McZ#Y)FQZ>p8{p;1aJvy0kH_2PdLq$Y(f zK$dK_O+e@|#V;bpAwNb;Z79+VEqpIw)z5@`(d@yP%Vm__0q-Fm`jI?n^4tLC}!oq0W_5S*E?BDjS1Wmf#v4h)jjOA1kQ8Qoc$q`jL>+ ze@}Y1?LiTUz-V=+ey5I`qXBU|z}#BI5`iDC0-JWEXn-qjE)E>9HIO3Apf>zy?PB?x zM9BY;wPmKi-T@iw4T0H{2|9@3sWbb^Ya^D%XW{q!_B3> zslS(j_`{wy4IW`ohOI&y&8r>eM)te!DE1poX&j*SgSPMuKp8~wOQ`ug=sY$}HE>PWZ9uT9o!Ns`0j@u&P(nk$1&o;6p!gN*QB z+)y)%$m3Xr5X0_)R!3hVJakci0DdH7?F-X1=_qeVf)%0*F2Xwr5I_3 z9(>>rWBS4TvwzPGGp%H8V{~a8`+8B`&j5k}DF>n$G$d1m;-vNCz4}6&&XFiS$2%Mq zW<<}xp#c7Es?sneiC}azMijP82-AnGym6xu>8 zCS4bfep2bs`HW-EJCTo$Ajg}!h#NAN6DU8yUdR#-FP%Z4L%>bW2RhhuOO;qbavet|8VD&73W2T{4|A z?MnJN$5MLoTclg%o5dSH_0w9&J`J{f;8;lx zmBOlG*0BqSCA+yu)e?gx7pIMSjgIh)gaF0LhZAZ0)2t)-qMw%xx!Sg#p%SapoNSv|S zw9_bWQg2oZyOKt5$-T@~zjf)Y+;HqNBfHfxtHpeSs5sbHuF2eEM>c*~B0kbiz!w3GR}8Cuh1;?o}u7>;P(~@I`tDk6<6+IQX=KddCrzm1Bj_e0L9U==d@mjDN77h)=X z4d!piK_qxxqpd&vM`0t9I1A#Na>(CA43FNQ^=NX>3Lgn=-ef483C}I)B(GoK5XX|} z4D80lH11c-<37Qg)kK(mc&d+r&AHgD?J&UtHE2)`L<2+6wV!wJUc&6L+1QRf?@48` z>rhi@X`xG8lzyuGOb!`~>5JKgDv{?b94TeiEx?w-q%7MO4u*3=m?b$oldOjE)kYOn zHs}oM0bB5myjMb}DOC5hlp2}0!ZctZi@%n%G;0I#4n_py3Y}4+s$_Nk25DdRy_6Cz zk4eUjOInra$_jjK_`9LzzTrl#u3pRNLoJuX1@{Mp<0%2`sKc>CZ5Pt3q8uE&d94k@ zSQJl*t$?uJnZG^S2Fg`@QG`mDuEr07pGq!kFuMw%w|l1pBLiesA+$ zm>PUs23#ftRX`a)XcG)wvYiRDAnYE{k z`{JJU=eB4In201eR@G1xVFD6CN%+NBMz#(-f_-4A2ozl{Gk5P>Hu(Gj*dk_18*Z2o zvK{K{P@CF+eMWM?Xmhn7J5YiEG%!s1GiQGxHzFyAeC#0F5Jl%@Hm*is2j&EI=S37zC>Kie*KDx`V4%Sa-)X<)BFp6G|3dbq?dkjLo(YshgS!aguy0Dy zGX}Dj5Vj7#?zBflyMNf zNj}_e`DMezB^qp3^2^c0B0Zlktt?{I*Q8v7+CN~pJ&wM3Uwv2yeU$TOzrPT2JkIfp zBWdl?nf(n$i!;O;+0lTc-=YdHvNVfsvEKfBaVbD&ls-*|1(uGoI^#iN#7(6U&hS2p z`vasD-FBU-RE9E*T6coEC0d)@e}IWrcs{!)3ZLRDDT@l&ZX#8$o@f$#D1uwb$ImC7 zg)TAy$0>RUK($ra6b+$OOcco!V(xE>g3~G_o~kYW9+BwQOOQ%E9tnGlGhe<2JAPR! zA|kk%O=dDGvQ8Bn{u)JcSH3Oxr>P93!SN;i!h8>`#n1Z@C)LIkhTM^u_53WXbRKNi zB$c#;squSRp=)DC2U!3VWn|LKS5p7PWnj?=4ZoCB4(@p*v<+wkalHe?ond@lCtGk# zmU4HdmPNLa{SVyv&()<`hnJQHhTz%mJAUzBg=Hqg`i`PYP%Y2<4eQoRyZ;i^PGjXWg~ z)6yyq4scw9?M+Nvo(FEln(BO7SCzVIPs#6J>Y?9|L-Uk;OIj^#VKdM|3n+y6X*O`Y z3S8_h^iK$V2~**W-x))j8Ut0s*L;847y^H5nHy7n4~L$|IB3ElS7u7!Y&MFWiPs}; zEMB5KOHoh0TDY1U8qvZsxf>I)&2O8*>nN}K6-^zb)^@7f-2UASv1K_-H%r&weU^SO zgYHSy?qgeVJj@)jRYEr}s^8V0UOgGk?LSXhDSTN{L@B5@ffT6_6(YUMy*1M7r} zs+Du(q(KeD8r8^zw>br9sHwyloWc|&B0 z&d=}p5R1Ykhudc`u%y;(9Z|i8ZW+~P!+#HvKdU+_un?;#FzWF)UAf+$`twGKhxwYc-(j1EX{R-C||rp zW`Vb1=FATo$Uxx49h6#Vyxgnbw@{A6yiea;cmBQ|h2VvVxmGV8 zcQbbAXjIV8YrX&4>3Z@W3+xG;(dW^J?n3g_x|Z|-d}m&f`ZcFLxM+>(oy(r1+xMOF ziNaY&-6bz-hY}@%*pcSBke)MrnL$|h_kM^|+{7}cOU6t&zDv;Muv6}Fg3e}tEd+59+kUr5fr1_L_D?;n@eKh3!pxQqmS+y#$>nlyW9`ffl^#@t?;N+ct%e_ zA?kP&-+a6zIMQM;M@oEU?AAYV`iMVBC8eZf4Jg?SO}akC%L~!W*=KP~$n&^PAW%*v zdBf?pxwr(y4eJz*0a-w?XBPz>6(N$OCzXqCpOI>oj@)X*!@3^nG{d;9wHAHA*B!_~59hbYu zNx?YJ!csfaz<3R~2fPQow0wAId|QZb%(f~>Yp6_dYZCVrS|PejcB72cZ`FHFF4Psp zvb0_OgZoR6)TpXSm6X33h0Pj#8HNv3JpQE8k{S4Ch|51 z7qH>SzvRi}ThADm#>zVI(r8+wX4J-X*3pbk)v+%{7r%LxIkFB3Tp3}L4IJeQZP@vKwVt__s z>JPS-4=w4_7eDaRP*Yk~FNA^2ulN@zw)J#pgu+I~N?iO>zz@bMK~zxey-;}L^%2ud zgmcqe*~sDcjfk`--?!m6km+rr%%ORo=y?2ud7lY#QDz$vx*L(#>jI67yEm)YSgNqs zzZtPI*0Eo;Is9_@8BlJgV4TapY-s9+qX% zci$zP`8$nA41W*3lRe54ZHZUoc|op`vyTT&=Vr*GILu@Q76egdyzcqM@m}&`nuWHChlPqxJITZ5VeoEs0$52_ z_w_LBBK5{}Pr1bi;7@yy+1OChFZGY@ALl>He+zDvE4NLF6rv)#Df9QEXC;|^<;J3H zx(*Pyd!sx;&|+BX?w0>-9rd!JXQ$NcBu6Jt8YnW06+3~%sdUNM$jG1Y168jCFi9eiWa`G(5`wRva%!S8 zo%b;0dpQG#y$tbTy&+T0_{aRpva;M6ez@!VPt{Yzqu=Bcof~dJfR<}9)xb0tK0&qFpr;|3uT5%Mdz zi6m`p3E6+97PrjJdR#z(2xFQh#g`DXJ0hGb8R=jd4uN9CM!d(MS>16J4Aq3M5{;TM zcls>#I5OwK^I#%$!npqsazKy*n1>xL2WafO7N@>B6xqh^tqkMz)0f_yLQ&ji2hfu0 zx`+q#T7F9pJ-Kx~Gr+x(G>=i_>3~vqCSeyeLX>7B%q3~P*a_p^v0lt>1~C8h8d~;! zJ%+YB(NN_60DFupxq)ZsI@m`Y_-1ZCUDhpxA84M^s0nYHaj!NadF6QvWT3zdhNIq3 zZY~6p#9hlJ4tpP6J0Fdd$wjw|Ej3C~!?XjAgnpsbGBFJZD6-4hOkdMJ%L7BBHXnt9- z`NMZn6~Xqx+_A<_x@iDdJiIhU-A5(|GdlVcn+4DGqk(RuU|ip zo(}#Mvunj!^z|SNZeCwC<T zDZD`_9I+|$s;kbQ_LI{TkJWli?~7a}sD94odwaQct`@TQa7p>RZM5AfX>--M5VGtvK!;@$0^s*oQ5Rg8=p8QR793Wj>2h5@D; zlg7%|2Y9BhjMx~yq|0Mh$8q*41Tba1ABP*-XlGvSad z%8(l3Gb@K&8P_z#%0SjQFJ3y5|6y};D;`B42u{KC^7hW?zFc$f8!nc^?`eSiKADRw zp2XMQyS4}nm8~~U$9?chMTDUpXF(XO{SKmG*T+1BiP3i2jFvAXr_znZ8R!K^1 z6byB2)ypNHZ+RQWjqu)22LC25NX}P+Pm&uE#7zc2Kta?&7sj(`4o|V~%(i2sa}QJ< z?yW~s;140i*pO42MN0UyP_+cvp1DMJtR*#50)wvW1F?)AnATKrBk1i4s#6kT?xwSq zrK`qD%;zLxKdJdBXVUf7PuSJ|=C?G!odHR~ewkB`{g0sWN-s08YO)rX zPCYJQ9fP$@nzoap@$(& zMpzdCLvaT~S%L^gzn>04OWluS!?8-_KLLu}Qk*ES2$z}7ap z@b0(!#c*G48=Kl9Lg0_)J!K5DW;NSSoxhKQE{(_2VfnF6w??F!eGGIZ@>X2I|ZW zC+HebJGY(p^}&b5p4Tf##9k8)O|W$q3oI-a=)kSW3|=?4yq#H8XI@#@9hC8FDw{O1 zkA7A1iu6aqXI^PuF($`AnJ{9)W)jV$j}_%4WT)~N!3Yy>YvhOJAmt#hAWbx`-iHFm zMYvg(s0bG9a~UxHoA}&lLI+q^{6Xx8_stjzb_S=F3zI}0+2TAXCKXl2F1!cEQ^(}m z@!M?+Omzk8^&YW^v{?8aL2rjhva?7pa_s(B>f{>_f4)^@k^_GL5K3SW^pHRETE7?h zDNaIX+nx1#G&{y;;TBcZ203KhY*fSvtHb5>k*t;=?}R@ct9}&zzTk&n;>fYUCO>m- z!YPrA4aBYWoNnY&(k-ozGH_ur`x)YD-a23+6Az4u3@HTq7M7jXBx?zDzh zKHi^*R6m?W^T}It1c#KN zxN6bKi_S$H|AuJdL>Mp3><%+ASr?oNi9;t(K|iuWX6xZnd3OaqC}EI7bd4 z`Sd2|E2;>>R4<>vb3sb!qmcJka%<>f!zIAm>A_Kinopw*D5Z<=g1p= z&9}n4!cX54pbMCq!P|)W!0A;zr*^FC8tWYE2718WMc`LXm6_Ky6hDtiwPo0*s!=Ne z#;&1IaW+MQpZLBBLA`B*ne)IN0t0(=j#EJ75c50te<)K(1n`X~Ni3?}pj>m=-?H+b z8-Wct?OKg?cg$^1;@YCg=lbW6zCVPqg716N(&u&7lAb7{R-UaP+$>C+AcCIF zM(MHe9n0M)m$JPbQ$>qrDLR(Q>reP8dJ7JY^uFVsG&`kE*jD4O(D3R-un zffxyVe`T-2dXf9>hmgFGhxnMFW&D|v7U#QmCM?vHR;2Do|M?Q9LG zy~ZuRI~WK+vt|RIU?|odv{af3-!@meM?N|)CM*5sZ!G+GSkl~h>pUk^pAGwz}@8?CV0?kq{@xEmRP$BiTylz+Is*S(+v|CuGcE=O-Ns ziNb@*y$}>38$nF_Pz6w>IbakhnR#b~F>pQ`%=sz4u_sl#3(dSN!5fZIRw@U=m6;Au_)me4*HxNTN`!ss%h1|l{?9B z=?A2eVljcwO{W{8g;o zMg^V)%opImfN*nI`3%F%@)L?~yhiiEVDp&=bvV)^i8%6U{6y*M<^9qslS7&!{LTy{ zBn^CdPj87^&pO2a6H3-XWffElxe?pF(_JS8>*cZv51}=571rBP+#mXP?M`nvpTd1H;>&eP3tl zkUzeERD50K*;Hfp9$--#FLZ^RN@Gk(SD6Cv3=;2i97raD#UN_MGRrbzgMoG#B6jgB zLUQSK$9vh@c({iUN8gf;q_Sqk#oMPUqDi1>~3uP zOa|EUYLW>X4OTT1n;zS36wT`-ScwH_jEMudA#9EGTr`fUV*?6}RKJZC!_d-wBcq=b z5zt}iaZBZuPO=%1iMYLMb3GHt1@_(K4fT@kNhy&DSL`&Ah<)TYqP+LB-I0Q6b<4Kj z+|tk!trI=5YBg%z(#L6iDDggO?C?40dfdjD#mJd7o~A@1 zucWBX;xe?guEY(G_ARx zUl&Uo$MIC=^?m)T$mVU;g-Vs_|u0vXZN5_j`=d#rJ@cKvS9623V{ z3gfHb%du(^Gg&Bwi|$ShaD0tJw)fFmFm!Jm#G#gZN}f*7K+F zPB`ZBooWH@#RKp{O359Cl%Sb~iEbt2G?F7GM5bx@3=ymWHYH#w#J%;jb>hVXx{USx zM&25cqVZ&z$&gX|H6P&>Cl0`ct>rCjr0;yo$Fx;!*y^mzW%8l}DK2una~1sU^r^{$ zURHoNPqzaGWb(bBhsPx?Bis* zV}uT#ZM+w5`(yiriYB=W+7p70{s|O8{Ns=VnG6wG=0m(@oJPadsClk&8)nX)9wYH+ zti{t=Fm5|rUM%?q=4@iGmT+vR1_T43X0mROHuU$^aq*CFodBL?QhoYj-6jJdzZUi$ zuC)9%tzRUgCZBkgfsE7il39!TbyP^nWz)K;PxCcTv-aA_2Zg&3ZAy%i!q3zesU}qS z5_fd$lP-}Ky_UExbx%~gLiSPluPRA$cFkw*Uk6w3U91VL)uYRzbw?)ZBu*N*mD6n9 zQgIcol{SL<7zz3qM{wco5Yiel-Q^hPn?|^l;x_AT0&CJ#CF;U1r`^hmnX?_z6$wd8 zV@wudErTl7YDPO|Qeh($y4{^x=)+gjv6Pdd6t|sjKa|ggCjJ+~x(-$<{X6m?>9`{!wt#<$4C(-4KtApzkSa?!6RrquJ!W>?c+_XHt{D z2P~m%SgP46!3jsX;IR^cOfl(aqXgIKN6V;2s3E2Jwl)aF?NIN-GK<)6kM|QuOE@O4 zgvl&_={u498Ke1=ng`6ME>vX)i}moIAXXalK$dzw^b|@m>_m!wSc|dCY$|alO&Dx- zrwmtu6rzQ&P}o-xg-G@FP-uJ3^PRbw;q8qIzPb}DqwwNMR=%%2QNMk~FQGTB`ORe|9(Rt)#z?$e7S7g!9}$NGP)D z3iXe!V7c+DPA~A>zxDS|FSl~8z4j@aUJ----4@(a{n_;SRWT$#H5(|PTer8*M{3Dh z9QJ`HiA&(OeN2=GWg1%`61%G+l7L$`(lo|OBvlWku+KS(*i?uzYNRnQoZ-33JHgxc zl^@W^_dD)vPk&xy^47esQT(pL&=-hLSM#HJRH)YT##y-klVQUFl@)F;96wU!pOF=XgGFR8Em9f|{Kq-(^5kJE$S`sXRs6E*qcKEB} zX=MDKh)!S3IaBepYPsc4dXl0bB~z_gGPxpFNq?LUS|u;ka6J%_OeKA_f=m^Qn!=ex z$~o3SB6$JD0M<#w4YWU%$;6V@y91T9u$G}mw02xe!l`)N!%i7@sw%Ur!Q&A_7~VOm z#}V}cZ9@FTguF6pazN2(4Ruzoh1Fx)@$A{gd$0l}?U3_LM76EJ z#6@Xn{HZpKr}u5ckL{Q~K{^@LHJDk5;IeHiUV5V>*ZCds5=#C@&EUqa<;|~y#d7PI z@s70TYGr$7JIAfv)>-eoMoB+K|K8<`cd3{4SNSIW&na${m;=hOtAwI&VtAFhm9ttF zwa;!Sy}Up&lj2e;uEdlPCm2uO++yw9LU38ZK1M3(5yVuU789MrYj6RTIHi=?v_cgZ zEY)sd;g;Ht_HxZsovYvL3!;~4>&f0&Jlj$b)uCauAe7_op+;OEiDIb~U%-##>qhgoj0pdBeL( z5{9x!YEX(7oMlb~n2XCGgI;z5zfEhE$|#>*T@xHtF52gGx7VS83_x>KU#5 z+*kc^Ce}>uF$1X{K^#19upu(8)xY>d#9}3#0A+FTjDtYe74bFJXbAJiLFa^M$|n+Y zgFPEZ|# z+MT{f>l5JnQOaG~HNpc1;D;T}J^3eXi1YWr?U2cWXM&{q2 zfsBvcN>{bhgTAfV4L+ZrNatWce)S!b(&npJk7`x5hOlU@j zw+;SkZHzG9Z%2!Mb|l;7p?}^F)Fvm1EA%H% z51gVC-MWI(-7m+v$eP`-o^fMp`%4SaoPSkKx|AC92gXY1sOz(O*3&(6tKFS7T{w5; zpkoW7e>SPI2L>XgkfJF-^waFpsMosG>K~!Y z?=SXRjMgZ)(yI=zo{vf{Jcv!cu)UC7T)_r)h&tzuw$!cl2sypL3owl1b(gtnIj544 zZ4>gUJT-Nx-3h<_ZS^+uD_SI6peM*O&;Go z!(!s3CEA(9K70w!zkUQueE0hg1W6R^fj=SI6CMq$O5`Pju(ymN7Wc)+wkQiu z8Wru`C_i8C;_6moF12O5BDCAh;xjLfwz<^{xj#5?abJ=(byeBMvYV<_vga087dwA< zuJDxoYUupcP|;rU%eCBdaV+I9I<;iHt$hM#Qg<@PIui?-x3Zzc@*w)Kq*9&T^{|UA zGn+N1q3DR!-L2p%t3OOFhEck#k@nQ)%u0JsSL`yAdAzH&n%OS8L!#laNY95cMOIZJvg?TyuqmAQ2`mut|5Ycm`3 zc`~|R(vz0bsg1N&>|yCFtMp>>S*!IHnrmaF)o{*UXfLaUspQUP^QVi0TbQq)X}S)=_a^I8_SdXvi@ zxn4&qs|T5vIgKAgTR`5dF9A~hT6JQ9&3aXMN4=uY4^X>(N9yDcZ-9z zX^qu3)>=MXsF~;0B#p?s-L|goucvm-j(>5Z&0MDI4^uA6j4db=D{X`IvtTFyav{`( z0|()l-K2@!?vsmP>FtbIbftv(&2!!P`a*d}CmVMxp3|1Zb9C-+<}S&fv4``yBDt=s z_iDAN5lsE*qZ1H^2#u5355;PC?-j>gY{BNXll2I8fpGw5RB6;IDsBS0g}YCXI_J7< zI8!?lXBQ_^L)(9z?2WA8SXlv#fPWr%co@ViZCp&901RR_hAyTerpERrrf>{0rgr8o z7629&W9`!l#HX}5xvQp-}Z7g&8jJU(Albb4+Nc<@JK$A5k8{nWPNDD z2^geYQ(GN=^<43qB-wMW>mjWNa9Axcu80L%Phb>AJJUO*-Z5=jcW5K+fNluaZ`D4ffc;0P%O^&WWa{dc z_9Y z%yjdLRokBg%Un6*}hiuzXtr9!~cEie@Xm5RR32$8zZBx?!SQf590nG z!2EAW!u^NK$QjyxInV!ej<3DV+!?_11wX-m{J{ml$;d{}%<%;Rj{o!#7XUL86Fnyr zDPo6h+o0p_^Hi91&UJo0YT|*2ko5nJ}6f2OC*Ooh0d?8T)@(Opi4V*0YkVDUVOE zr(`^O=i!7~yncR+zxwfgUFltAx95C?>tD7WZ61L}d`po0D^XZQ#qkx@(@&-^f~u=1 ziT*HkUsQwF{V~DvH$2u~4jt8P;~1}u^?rmULW8wseM#;+cjlGvn|Yz8K3Wq)uEWM1 zUc<&T^~hE5(={yGP#jVo8ctljdEjAq?CWuaewVNZ^32YuwSC)ZECvG|saHM2SWTDJ z*q@~n{VORF>MC+2K5TC1wpX9#VLsx_`d|q>riZp!Wz?l?50t9$@Iv9?d{e{#O~~tCu745pbLF_)FV}4-Am`8c zVhe}|9dU9BblSpHcK3_fSrL3%-yoI3D+Mt(Im0Jas5%9e_bt(p?q+u@Bs^2ZYHN<2 z_I1JUl6NaII#NOl-Vg@L0yl@bU#%7VfVD^#(09;EfeAgPf(W3uJq^gq(OlK&_ZoLB znZ*}nQ8SpJGy$_c?_&;LAfCuSb1b7J#I>ar#F_gl!E7 zn1)1&5uPUmUz`jWYq7CmpJ6^|uUk>{BWol4aPr*%B16SRGW6Zy9r+m+-R2JKCNX$> z<#=c>_V*_>Ofq*hd@~eG)%F4JJ-uGO*5M`oJ*=l>O$~?g(JP=X2d@>-`@3j+N?q5V z#2}MA(T*2T=$`mZe@44;-p{LT%2LA3!CNs)za^>9o1D-d8gNglsm~NiP;6p8WhgyK zMye{8lkBU9zc7Fg6-{B$`C?PNZ1cmOCU)m*ET|&L2(7(Z9gv#*@X(*ICK}U2p@DMh zL`#Rkse-UWc^7`Q4W3}S>aN}CqEPsad$A9nFPjD#&C6KB3jU_7h}EzPJ0TnZHj)RC z!3W6$&BO4b053fqRC~HAgSvJ8`K=Ka1^yg572X#-GH|L8VYRuSV{%UYIC@65fKZ4d zS)ERofcWJ*_Kzk^Oz5Qv*KI=B0T9bmwKkq`)?|!Z91<6`s4u=O`|^XLYDbV;q#0`K1J|? zR3UF6Fv{~D^GZM6d$!bn!qSK{BkGUa*?X3ogf()Q>|C+HzffY03{{8s7?Rki9L zsW7q!Cm;WQTIf+;q$9d{`?I9!Wi=iy=bIar3TRAk+ZyCK6d&BK@;30Uj=`aB886uP_XG5@ zg@L7j{(IT!oipJtsLpmZe!OE#FYxyT+|#e$D?DHtZvE<{ ztb6>EYAul?n>*H0b-s6>; z`ZI^c(v!Y&^Q*e?I^EgB!p{IYw6h8~1`H?ESECEQpT%>kg5D4Ds+bqrug4DKNbz21 zi(90)KfJckUw>+Z%~@4(kD%Re(AyVUAM77u$%%hjVRw! zOPCs-w?2lXxMius+^490_+H9mNwTtExy|n*-TarmQ2-aDlRJc0xKLYE$4=Dpe{uF! zF?D!Pv@lMQLMiUW9e%jGTXA=HcRe`8-QC^Y9g0hFcXtjB=Uo2Z!}ofVo9s+xCfToh zc4pSBwLmkG=HK#^2OAl~qkTl$5C@Qlbw+!svdq@~eLyZaINuLFCwWVg!X|=?BzS?i zhz@fDK@^O*IP$IS&H?kNn#2E567esmQtOBYs=9#<$y;}=ejaguxqYy}9}uZ-UteLJ z+b8nVUEP~SjLIUCK-Co5L~wR26b)7MzA~bxE>Rv7jKgXcd8@EIBSFy z2qlzyGDnu5k-q~K9}~O_T%b9&D-y< z!JEfY9NpTYRzO>jyu;%dU)2X$GTF{I-t40tieJ;@Spj$=4D31D_XkqMD7|n_6fs1} zR7GFZ+MpT*FpO{G`S`pFAiCuN67SWZf8KkMfqGe^bTto5_4$4S@xY7SzMcr6X=!!+ zR;FM3YGCSfil}}i2=7yi2(yo2GyZ^vSd~Ln%+$PqKSk_X5*+U}E%`ziNjBme^pBY4i2T)<84WT% zSg#w%_dsRtSbOBR6y`tAD7VYD-|SNx{MZ%jgYd-JYlE;{d&L&Mf&WA} z>e;1(3Pp9?bDtv^MuIpU3T;EXf%}>=&B4CF2?;Xw$8^lWeZ)bqh6>(epUi>v-Z)R< zz_i~tA?K(kPs9NDxg?;;W_JN_C`|I|{-Y z*4p*)O6Y?VC^3JXzcpjX(|L&#wtn3Rrw1g$-e54$`daMuA?|3s)a6&zxNJiLE%5Yc zXI|C&te+FSTT|K!?}SW$)cY)=n{5k!DdIha7y~{LdXHEFcg0UJAzya-w=og3-U&7i zzzD8_`**98(d+y*%dMdCQIMCyyI>5s%;ykbqCkiSRBDiK`zbvk@nK~s*?;^J1?7_mK-fSrc*8bw_I*HGIs8;# z8~haslF!3GueqIN0RLcVdh11YaY|i>&Ra*-LysvM{oXRGRNd90Hef;Hv~MrkP$Q7B`#yr(B15NI$NWglz(j435x3j8XDK_hh1izYtW!=nnS+zW(9kh6KNZ)XM z?Xlr?$=r(cebJXs!!eAuqTg!v8V! zGrEh@)1fT%1AYR_(BnPM1=q+`Oo((MbuVM+DcVP%L<1j!&m<^5 z!1E4Su@|}Vjm(<^#uL`=-w!Elwi1${vxe`1uEIWSFbh5(;)0;s&7H@Cn*|kosIv+% z(&tN;chZ+T4xa1#2ov%K0p{ZRl3X6X+y}tvSU{N2TR_-nfZsYCUpVXeKV$H)z8{$- z?x}cwQefGS(T){>+J7MJ`pu9X5&?gWA3YFGsiXdvf(X@caGxmiseM)fWvzr@_HTKR zJzR6CyN_2W&2}WKr4FL?b+Z8iI!ugjv8{e%&{tY^MZ~>&kbT)CiwTfVmp3Z@9)IlU*vix^66J#*J}p2=ZT?B;U2I- z_SmQF+30m){LK+P+9zT`h&ncFn-LgF4t1HBzUH6#^&Yg|WGBkqh{Pyvt06nTY}?&( zL$=yxl87~Dm&UX#WBm^KhXut6OXh}ttWNXN$Tu7Dr)5xXTin&53xorSzW=QEjiI%v zn?ZjMydtGl?#frNpuu_^vvQdK`G|PuAJKA>u?{m_7q&x^n_$JJ0tiX93{PNEBEHn% zxA|u?+j{QXXJ936&aeb1CB6Dyydym>Lr)R))cmw>M`$p%-g=R{<$!sjYqK4CP5*sY z1kSNy$pVPYhRjy`)B`C?`R+)WR-W=Wb;fm%%e_clU4@ifihIlCJ|`0VkM>ID;&NL} zUzYoE8~ZQDY~G?rd5SYUdkueA6|r#biV;e%>SoO1HaxY>DakAQVoTWOvP;dbpAcb9 zjVPZ3uP|zq2L;|6dyBtn4gkHItlWl-i%586Tg%}CtLDT3<{D$-l;zG=j&H2#BNnc? zRbihy*E2QDySDzL&fYbY1K;N&X;IQI!kd*u8)zc6w&- zpZ#qebCwnad!HBLv6gz$9+*^XpL3@S=E3J%Igi~$?v<&a*U#YH99_ciZ%5yEhF){F zkp!a`N1((a}69PQTPU`Y0^L3#To|3`QPm4E;-ebGO z`P{e5mroba3?W+>A3A_vQSN`YHBk`Sq}F zyAHp)R=DeHWAMLbs_l-=aL^#4<_+5i^lnp9n4KoENz9#O12%|kTZH!(3)nR`qqoZ) z?mnXhlnm7FSea;0XT;giSoU3d>S+Bm>;eDt63dYVApiuKK*X@1w$V=oKMnoIzVE*D zgPO9E!6%7DfgaC_W5i!gkfj?YW(lD#dWZ|X%xMs-@W;0fe@Fr}4%$E1K{xb%46d?4 zDBtpPe$D+B<$q!RW5{<4is_U1TmWcx4TBpr;rF+QA@^iIQ+%86Ky3xV#5Qz7J5aOa z7t!RtOm`nzn_GW8Y$f6C3tpTlha|pAof>+E@ChXku=r6yUV{lh{ zL)DLan;SCAefjKRBStt#jNzXE#6A3$MOrh2i=Ep9#5t{pFhu}jWzRbV#CandF~tV} z$>+5CW4@II&A>@O=>(r~sm``Oa% zsjx$0y?W4if%Go%W#YplpTj>;AXv}PPYNu8{4vz40Kabkv;#i>`9QYJjt8rH#qiKQ zE4SiFx6v^t$2AG%;zjzk$1fNHhTm6)pe#u9xHjZ}+@ME!M3jgI`rPb(wqvTJD%(JV zxbJCRW*Wp_v4px41*g_|Ydp>}Zns-u_*gDJt*TqJ1vThOOV*2SDTqrFgmu_>JXbI& zpWH%tbv#5EDkOIaX$8w~q!k;8!umNDw|Wq1x2@1SF@dn+uQkfWo+lGKN_h;ud{C&7%2xy!c=T!5!-t-}9SX?9YuZT<}xvXx=_}k2NLi#{OfY zivWxT{>PdQ@Y1IQ90T-U<_2S}??^8NVY=x_=ppo;$Zp6^l}RVI1{6%DiAF$4Js zlb7Ylp7F*UKE*mW)CkAhXQkn7s%;(O7-ui=F%#_Pq&D$GN!T6W0ZGZ+z>jUJ&L`|4 z!gGZfG+2d!`;wEk+d=nY-@HDW<;?}Dpw1LCLhJjphhULi#cmXCP<*CA9mZftKe z3yl9qH{c;_eU<{L`Q99VppjZ*$~IvQ&B0jiQF*lT=OKrS*A|5@>Raxo^{jJ(@RrQb zs=6VuF7@^U>h-hF@I&e^BJBEYk~PVy8jlR&FRmG9tpR^RsLFug{yo9MZ0sL>8oz$!VrvN|n%$TD%7oRAa!w@+Q`Dpw zpTV4)K-phuD~Fyo^1nV7|B$B9!lRZ?HJ5TJ{%I8fEC;3nn=a$wY2}kWkJOHIj;d)Z zRk_@f$22R*Los!X1wa1c_Wfopscy+Wz~?KuJSF-L_xsl)AYI zAZrWa>0pqhkzf0|u|m=|6t9!U(-EVP1Ji-omWkLQZl~*+n-eFvgW&v-edEd_0R2&gWprqEl_M-i^Rv>>=ia(1asb(j@Y`FGxW zo8HyFOldmvG%sGdxQ!UjZ4i|9K3uGeJf!|Q{-7(BkVY-sI;7vEyd2Z{Lg|YR3H=zK3 z$>Yv2w>r7)?f22J_HzuJujvJ*237D*QJ5d)A<6=Awxk~#Zff(H&KNthu>RqB2N8uE z7(f+l3qQ|Wy#=N&1ULT>xK?`0B`TN(lI0XMHOICt*Z>jvrA;j98H${MJTYY&R>VEx z4s%(}dA)-jKsz2G_%oW$%--M@|G4b}4=X4?Q7TI``N9a-?|PCe3y-QYNmjJ-1^AYD z)A?s?N#^q_u4L1tS*PYN<85=SKEDm5fnWuJx)lyr*wElQL^XQ-&PoEWA2nL`8XL*iT$KvmEs?w&*@VcIG;(>QCf8 zk#;8f>Z21{u%9yFng}{mtxaWXC{M^$rYKm0?u~l%$=q5vNM+V<$?QgGwwxP7nT=Ad zAy;OzjVGFzKDj;pg_>-3WK2;;4VU6iKch^hHTh%>nVK|CIe8?<&RCC)?Nk3Iu2x;L zx5o9%Y#TZ>eqJ)9PGvS>q)wpJ2VdGfq4=crNOox=rH*Phy1r%i8ip_abZh7o)2rOB z*)C#KNvnFUe6Dz|LM^9TgkMULadAs|`(dX@tgcaIvnaXPVqNi8-uc^4&QDjb&R&so zBI(d(QT`nL+}tgyQ_WA=PiCvsdx@n*w??sTa%EzrLZOOx(PB}?t*ldatL9qixpKC0 zwnFXv;T-8a>Rk4G@|^Cx>fGjB;L+{z`=h*D-&?A;5I?m+yp3tzrG}Idq6WFRI)=C! zo45w@NNM`{#yRhK*Ez|f-y`&6{ae9Xg0~odrGGV%il9=}lETu#xyAX(Ire$lxyJd5 z=G9Im)nd{)%_HxVz|){+rFqEOA~L%mbJeuM^ue*k@yRjvaoVxQ@yb))rQimj0c;JP_Mh?J_8-tUd0unM zcl!^1C9&mLJ2JJhWp1H4Rl75_pg29jJefnGOTJ65OSDTRt8S}=VGheWs`;l~L#voh zVJsx@7UTD<}4P;AXS(fqRY*OzCV=Zg8t=`CKM zkjyn&AV03UmQ0ld7gGr1f&DfJsQ74Hhsa`&C_e9O)0ck#3qa$PL8p6!x6(0SN1v3nIYY=7 z67Eceda&1x5NBB9r9w2mUQbUnQsjjkcjVNK8+U@R32R`7&>c&@oND%9x*W@SLc^ER zwrkZ*p>Tq09=o>9XH3K{$kE5z$ElDdcv^R_&X?47Z0CvPFxlkAn>)1efYMzA%IuAs zJ?4DSd8zS5$(ua6+jzP9K=G%_tLvM&*+&9$0y!T)-*ny}Kl4GYpjc2XDA!*KU;q{p ztOk?=iizjr&WSI^fC@nCPnFMhPzI<26bdQ5}- zk43Z>HhkC`5|=jK)kIoF>3h~DGHfqeAEAe>E>mkNpKqr za4HLMJB@HShS)A|Yxk4QZ{^W3()RuniB=q{XipKNke22cjCL%FKr>=vCyk*dTmQps z9K%(Vf#yJt!#HF|jwdbAD9WxJty&bCVBXBO5QAKlvY;!?%o4LKfloGRX=93i6-`jo zGe2-@>B_=`ua~S^5<0*1_Y1briM#bZF2Mr$pEq7l^tJ@?L-cErQ^Wj90k>=TUOl&5 z_|~a~uJIv$8s29N>fTd_|&oP8cgz$IKUmT-oHbzbfTn!uz zypd5$lW68;9E)hqT9#T4N*wg*rcv?(yaVk6o&#@_M&^-p>rF#B)>3@d=nP+)zO0?0 zw7a7ZeShM1|pFgGL_7lm+KZ=W~tz_s*suHVQRrq$V3xLXR z)P5k zXlfMIX;EpaN|2AxN>sgc*NSb`x+mPt!W(7RYWT{!|FF*BPe2~o(WrQ7yb5s_`WE_9 zwy5m*9S12J(wTQz`bV9<(_fFo@ky%+D=}$AEjgXTxD~!O=WM8TO%oZTwn?s=Tp@hb#!CL_ zTziNA)w)&QHJ2+QA0{7z_mU^5cSWCqPiLP7zihuCzr;`Nhjg-av>^fWu~n+-Ouva( ze1H74^q39uEL#=ZO8KTUjv@1=434g8Bcz7V_3YOFblPho7=rvp1=0YifCNAaAZd2| zWV&q%`dIkE?t$<@?}7gz{EG`^2jVcW^%tu;#pn^bIMZLwxi<2$rPP_ZHhXT+QXRKv z{w~=~Hn8X1P8bnOF4OXr=|JGOU;>saw!_MNF_$vdgLUwIJ# z@c@GR#-AVBULgJB`=)pHZthJ!(DkPUkH%gtK3qVs;6wndUf;J)elgriSM;$sU6B;i-KDA zC$qGIG=idF1=_OYoBYS$r+)30=-PStzk`kqiuCDn;(K9#i%dS`pYS)>SEeYW9B3jj zG>~c=iknjVEz-9zs7oQLOHwY2X|QYbNLf`hC$GyA7892=aUGQRfgy2iQv#8$-iW{0 zFbl0ajaaW@;}tD5J(K>FCMeA_H#kl2qw8J|7atzeR4^=Pms&JW*ltl)*|~RRKN0e{ zpKq4fT#IjY`rLa1FPFMzNBrLP5&>3VNN6Ou??f!FtGAPgQ4-PJzKfU|SaGGyCSx&u z=jRCMRC+E~ZYv78ERSJU|174m-45iN9D0wIxm`}S+}{a3X8LRf2E{3&(&&5MbIm4i zdTv1Q>U{0aOQOwR2vZwbV!h|pg9}dEP2lh`!@@FE-uf{EUEL}O7;8c6alPJSanm%b z#;?HC3%sYR>utSx|NjF3TXkpsU(ah_ zPuAVx_ea+ocP0B3#vEQZ+q{2_U3ThO%NaE+3p%AOQu0>0*~zEw6vLcRQ85v*k+yjA zDEP9HJ|#;;CFk>OD#C7+pOTBl_Iz9S%EkXbMCzA@h$vBYW)0A74#y>$+BH zk3lBVs51IX`NdR<3o06k+*=*VS!)wpylYomJ9KM$;}XcUVxCrYhoo)K$(Yv;p=nx1 z@PiQ%hK|p%xpR|OW+T_z+XAdq$w0>I(DMm7o!P_O1G|ix@+ostOVYy_nOx=3!cJzo z{vvM=nuB8AIT{CB=eq6f^h5nZczv3Z=ToQcAmPv+pAia5EVI`8#$1!6F;AL~a89-% zS`q#l^KNT}I&=1eTxto?s4cUaN|{tu9xFMgFhAY<_W3ky?~-}(!kBvG=8Z{w_V6_CCdo%_bt=%W}@3hr)WU1Jp3G1p7g^g%LkJE~|0PGcX;R849btn|ND8{_8;`>)H1aSYNdLaX;llLgy< zH_8F2JBTp`FNiKA6cCe`OMma%_)^j0ZXk@@(irdF#|}ly0rB{wWV7|eI(MIGX=Y44 z#ud0kF{$!+UdNK>2Qb)2UeP1+=aP8TME&hnvl8~?6|iXwABVFG1}B>Ab=Z zX?}Bl8?P*GrI<3o|cnFWBrfK*jsY69pl$Opj z^~g)yPn=Hyd6^_`%^W7?_>%2tvXpo@U4^K>c`(F2JZUM12+JAjHqDqf;ooV2!IX{NxJL7x`qBJauMGa+ur=!yLt1> z`FO2Z(vq>iB%C^#*}32;1@pHoM-{C|mZ{Hu_5!pjgyX%;X9ePJxjXl2D!Sy|$Nf}m z>bD0vTavZk$p+ECnIt5!ipG_FS?R7J#a>4+Y99I;l6{ z9+3pPnSf5gB(dnGs@6=F@Jnn*3tRr$*&}+k0i))sv z{d4~2T83$vo7$rinNUcAb7KD|k%YZ<@EaO`6eFsGF=kY4eYmN|4xM~(u7YR|k(IZ& zs_WC?mU1V1vO4CxQ;?Ai_o7=J24*_me-e>2Yjo_vzd{W6OG55T869$woC^Lj8s8eC zB9aU^@ROJ7yfk3>kx4znpXorJ*X_eYM+NnfS-R=7f?#R85&?2HX#2jLfP}{WTeLci zo?YVmoJB(5fBOeQ9zm_pTWYy4nN^7z>H+yTsA;^UBZjY0N4R!DkQy@o{lLUV4v{oJ zDY?add00eN^MzO^0D9?*OXtr9N(8IlodNMq|8?`b!EO;1Gu`ovjJ&e?!n;A$P@|&h z-KN?E+0tpidR1maH3V_g@_*3uh8imY$L|+4i|s3=?5+{FV%~5JX3NsXx&|L?Ka|w} z&J-%r5Mm_{zINWn7ABe<`E1taa1Ag$hs5Xr>egv@{Z)3U7v36&bZun z83`>rR~z)=>=2Gm7w`wbYmQ>4OcM5ZvEwA+V3&ElFt&({yyO~(e%qPp-2A6r&YPzI z4H?UG4B-;!)o(|B;KSj_>WK6u#>GOI4mlOXo&gn0vzz{9u*NVPG)2b7RwjiZWi2u60P;UpVINZxe*zXSvU>@GEgWZ;MrL@bmeMnF;j{#OXeg-DNr z%Eh;P8!%+zza*9<7FT0`0&FaO6d4kM*6v>@$<&emg^s3L1on>Y z1AB1c8~cvIL*G$2xi+!F@&A`(K%wHw-gOKbvU1`Y*n&6)VU>x84JLu;JJQ`cXqjC`?@4ySLf|`i`Ch zdo5ft14@Xo%8tp}1SPxFu#z-fG7PcGvQ8EdK?i~w6JSs47rSc3G5yJthE%K^St#oLs1O94Z@LI_aLZ5|0VfdC zh=dRVkRRPWRMIF*aMm_0cIz&gjq+8?p5;7>dR6u+c`7M()-4lTUQU}vYf9-@onvpg zQ8zb`wxloD_Mx2*1kG8{yc`?xaO)x42~WS7LT3me?lvwg zDD}=IVN^S#&Ou4RY!)I!VknRg4Y|(#zBa_C)pw64S+`1n>)kqHXcVHI)Io;vD(xP@ z(a0TeS47&wY$#)P(GhA(=@O89N`h=`*=)f#Rq3FEH&ykDTK+{gsdMRZ#J&2T^LUV( zWjxA2J1n8@HD0W7@Prop3 z&~(pI<TOgJ7bf-6x}N^nE2GgJ>bSQfDjEq!la_3{ImwhPjR@3c=@Mxm zX;?fHkrhj?)m~IGUg%Qs7cLlFjsI|17b{-|oD@&&>$DF3X6&umVHr^$as68HY>P;q zeFz5{zHV1GBnysSeR@ebQ#9>SmykajZr}P&5U&qy(2(^hKcGIjPyKzih9yIq#Odn# zR-xbf13sl7G-Zk@@pvqPH;P5#A#L=EeAM`uiGjB8^l!#iQCGMJ(#2-kih(aCd>^Yn zSZoD#)JNkGl?G@Z%XDuZ6(HD4z2lSBKEFCH+EL_Dm%}F(UpSa^k3qEVQ!kZP2BZ~f zn5{9MJN|)DgcX#Zz8$<-h;u28s)X!C zhh%}KrGr;)zD{z|58^4Dq-Oix7G znoormzIFGlxvf1(lFzi2r@1=m&*80DOD$^Y@7O6vkrRmGI$IAQQeFj94{W=mh}L^C zoh|;YnpSTG3~vKP%~F-P!rU}?OIS|3xxM}N;X8JevSk(kb{lqk4L z)T$#FxSuO79P+-^mzL-JUyutDfh{Qi^R!|sz-tCbBM+C=Tn^oUm8BHz$md`9p${*z zhyvSpdCX^85$%h0HtEKN3FIP=l0L)?+*6n=#!U1ArFJe_xw=0w zv$@^ayf3;CCtLzm-7#*y$4#_}%EPZ_-r`wLkdX(K`9}>AYnBCtWSX$@2#N}tlu(n( zsb-89nwiR(TI)*U=ds_VTv|WGcVVy?FArSh33k8K3T#y z@gR89Cv?^*n)#zFML}vQx)4!dAsqO-O5e)+WJtQqaSakrqH}$Vc49ITM1eLFqv22RicS67S zzWXY?Z9fSf0Ps7}gWgmsP|3aVn~nMTC2tC!P|r+w7=iGQ9%f{#n3iUEI)AS%6dKdk zHZFCd6RY#cw^N=)s*+y6$N#%pP69rD1UwF3lYy882LOo}G}7n44 z&RFi?xBuX33vO`v;!yXBUOoM(4EodVuU^(hp6{Mtjr}nVJn3hq^&&|Y-G}XAWlCxj>qr(Txn!yOZta0%tjVe-q#wy4# z$Y%TbkOwMX5;|oCrey#8mE6g^$LWaJRpoEg*wm`uTFmW{b5XPZOsTY{wUG2``5^84 zgVJD(MdpxPtD%1q(G0t6iMuo8(L~gob8^4u+ah*}bQx(NwBu0xzQ%5gXJ=HViS15z z@%MPahff)A@VZzR%)`hU{Af~!z@NWy=(N1E2o6WRkdr*a-yv%h?S*}+NK`kJXj`DV zMBo|20849fA}m|7RZsmqefRyT?trnprdvIRqZx0r1l_FHomA_l+BM#E#`Y<(L*13H z#BV6wcQ2nzoFMq%0kkj_g(rM)s;=2{ zxOwj6Gp_KBuCsI2X730HTu}Ki`A$$X>H~`cN|y z{%mD)+R5bU3HKu{voGc5h2A8hFP@)pj*b6<8h{*rvVH#%x}b~JP{V&-h4f-Cv~)|Y z%kpc{%MWDA2NT|)5sb4fI`bB{;azw!Est8QMogY*@l0_=4gMJ|^do#c6rIs11j0@2 z5>`Xe92mCzJA)Y|9OT&(+K%Kaw}P!!$9!{5?hR@FRPWOsU<4WuDXlQM;>L&~5tav} zXaU~4S?O(<&gB1gal0s^rVbOq8!)R|ej8D-IAR^GNO1`0L(ikW8X-E6^rmsISqa4?Nid~9Q^X&!%vQ2t0^av{${^q-vtI^1DtotP1wSh4@kI{F0=YNLZ93Y**l%o;=cF#bU=IL{`waInhI>DXo8^ z3H>`0OK1fX%+HH4@5YI5IT5AD%SQZ!@W4|l?7x5{|CCGS3!|n;tjC2Yi_p#{mwAsZ zKjIb3C@*uPD&(kF)H;=YYJ_kW83x;I{*hoSM2oQ$Bg32je|Pu6CXRPw-!MPo#qb_i zXEGF)XuUiX(3MT|#5m-im?P)aM#f6`=~Ix8&E_$3_CC1VC_MrmZ0*6By63U=4^X}9 z<`iKRVI{j!33Aw*%1f|uFTtPo9Wo!dQMkQ1w*2UQkZYe-qMT}3>f=B8oX)M6$D7ye z4J))C6rB^Y8v6-N`n=6u;WBZSzz>-!i#_0qCU*L2)=;uy zC+oEB3=ef)972LmPr;V9`LTw1&)^#s0~UQ}-@kF?H@G+4gW2=fmP8Ex!U7b0EL7gX z*N&mB2uB4(!y=sVKLMWkQ-`a?e}>HE_PdxgKQ92*0S_k?pCmylxGJ#g<_$`0))s@MD5Li?DGMLOJv>7NXpcqwep+2C4yyz)so5@L{HvixV+KGq(#Gld-` zxS>C;t*%?~UrD0DU9>%Q^d;Ptd1yTDK@lHHsA<||YEL8m*s6O&Om#9;(99K6Gu-(C z|MO+JO1cDIN;*!d^%XDM&FXKNFI2(Y#tLihV@Gy(bz9$SZ`UxrZkU#oBx)gb#*=sF zZx)MFBp|#A2QB_1l?Vst8g3(*twT~}J+wT5d|6@;^aSU4+g_@MgXxKxnbFec zbBw&*Rxn!sB(56Cv_jGwpSVd?P#$wyzh7y&Ho5Fu-+1lR!N<0ii_Fs+shJ1oznN6^}XYKdJ?{go2*#Qw7-f- z;$c?1BHxI%`*~tb>q4i6sL^`%-uB0hPlkSbGp|5zs{t<4@Jsb3PZO42!ugx`-q<=j zN%e@!1mmKOi5IW*1Mw3MA@51K)yqOiKuq%!cO|^RAHpXL@>U_DUTtL_bLQ#g(d`Hx z+O4Z3%b&zddL?pmb*+93q56d^wfJ(4xM9_}(H+uJzZO(QQ@0}V3aBgXABIOgnj$Et zR$mBo5Yx!jumB|E=!Mq031$+V{Is;Gw6B(NXU5A+`?DrLs}866!zK;2M)`vAyF6VI zl?EL}0VY#)qsy5AI8z&IN-9M*a6qD+3)xfZkOlrWFwK-f)@EjO^d_v#MKo+%Z*#s| z^RG-+PLBaurj&~AlexhwiP*ir1<|B(uDvS+i8t~%Y=w?g(*6b=j2rH+e({T;ENa)} z0^+vk<^4NxQgX?+LqD^m67REb*1Xtx?eBXAyBcTC>iP!X-f5Hduw~l2O0`!T=RtMJ zbjcnxsoT%U?UR}q4b;vIHL3>e+ezE3fW&AFvgzDoOWb)D=P;I`cG>wBN^Mour1J7q zK0O(wW;U~w`}_t4?={uVra;Zz;GE&gGd4=KkFD7HQ-Y}V|6*l02GKEdXf8-{<8QO$ zKj=i)`W3Oxt?+WtwZ|9ur1vs7Xn^nWdq93IwGnf6lj+P>>Ltzc38Vka$?s67_yk z`APaAt=Xmy5oC>;y~hsM=Dn`s*QLyCP8~?(g2J|jBNP0_c42dD!~?U>-$eWP{#fSZ4@i7{t`Hr zIc}wUazUWC#AJug#ifpL;mXQPqiGS_*P!PSNLXU>@K5g^341*}1Qp=*a7?HX8vc#Y zD?myhCLyzV$|a%pZ)R>`SQkcWiS1C32D0?;K%+KX8p^ZGOs2bD4weku3wKj)zJ2iR zf17f)v5E-MfAs-k%P_}Jf>}NsZT50TIUCWLv>-zy8c^l^%ZyA3K;Nrcd?M2|%pbcT zzSyut)5qrD@GqkHE`ygajg|Hi_la5qzb?R?{3mP=;73m%dJpF*oPq~XUrm{{EVvbW7Udd>vWawB) zXo+~)NGZjs-*m#gqDct1bgxz}AwR#n#=xE z(IiN%Jrn7#}PbibS0d)Dy-wox%8|#zaqLxhQ3Pq?Qg_ zEsk+Yb-n%U7J#Ok@$F1gS*f-=ef?Yg%kNNxDh_rs?C}BEzw0J+sYywhqPRo6?v^L1 z!8Y|3th((OS0&fJGt@ilWqm-qZS5A9GhX`W7Mr8@zb$6lNCv(=7K_)peAj826KGMn z78-r;u=pNkeN6f{o>w)3d3QO_IX7udNC&!5O7>&OvTSP$8IhOS1mg6yhZs@boswcl z@%u`M$zC!oh3(14W|7X?2`4mzWC>Cy3*on#J*x7`DfN2Bt*E4w{~nz;)&qBbxXN^$ zM8DzS-G zI;*SwraXHvtxbk$P+71c!1{aLbgh0Z^VYoapW|Ah7u}j&bA#SA)!%~{=~k@|iB~6+ zrYkFxv@h+)YzF6=?kPxr73M#~Nl30a zL(43c&KF0r*95F8?p8vMbH)3ls-8^DWRpWlMRM~*Q^wuh(YgKQakjdCir z#2Pk-HyQJjS8}5|SG~fEck1bluc?Qq=yU2`3Bf?H|mX@RzD)G2<9*4}z`J$}|U1=9ld z+6`g`aWA#8(Z5h7q@tV17Ms1fkk!(?2DFBGd6O2(3=}d-l_XLf+a~5lZoBLLnLQ_q z5ug32gR5;t4S#Ms59F0*FRAMERR_EqUVFKFdwO$vxvQ_z%*^uqKBicgK6=hmR3=zE zUuvKE4|NW*ki+c<9R&Jz0DCJBA}E7hub5Rre~nV9#MRs-E~8T!)mN7$-VsqzWt~Do z6th)c;0uZ+LE|n_t|Jz(V`anthq$|r$)pJv1&_<%4ucQw?moE7Ft{_gySux)4K~0G z@Z#=ngS)%E*o*7S@|}}=&v*CS-R$l^n{+zqR9B_aPp7()s(yX~;y~Re^8nY)Q{U#F z#aHd`;fNtFpRM$T%ST7qh{Ul>LoV+j1p2ni&l8` zTniqHGFG$^PBgs2X(mb0WbHMsKfL4Tg1lKxo!!>k^45F}9*8(Dj9f_iQ#?ZokRXIB%+RhaT+#abS!^?l|U+%Tr-&hbt; zpAkKb`_Lwe>$H&vihlkfoN&c?>Q(bO#ccXo`4i{1^!>vmStcFMgn1b?FKdn#JSO_5 zQspvKg)z>SL@i4#)DPWFuyEEVJSqD$+N$@axwQW3CucwOYVw6Fi&8QPEca;!d4PCUdN43ZtXkT`T1 zD2hbcuV0M*{8S^chSkTBU|entPN)*ZsnM|3pEcwzi)|*b{+{N_*^`7-Q-=mDVe9rg?&e8PW4UE4XiGE>Ae@d!W@>U-W zeMMgG()^7$AC{-*KE%QT_A84Eq|7XKU%B>+%s0zTq(MN2)3!=$Z=v(c?zyjz6QD-w zoXKC=i>Ih$zcsyeNBu|E?2guFIXo67-MnVD%cR8JNOYkB!$h)*#S1h_^Kc6(e*&;s zC=k^{`T}{2;EDIuNg{H*)Y`a(HCY`iiMmzR4k007y{Jig@pp4rbNNGPLGr`zmp&!g zbaiq@iPt*AyU;;L2@Srpyra3-UWuUyj;;;iA&*=uX+E-cY*r>0N2S&n$rM%$^AzY_ z393a$4ujC<9U)H<;ULAwZa${O2B%C0mNef2G|yVij}(!c)NHsYv4Kww=kV{p5CiO& z@M5}O?MzEK|K~6XZvos7S1Fe)_ITfmzugZ(j;k@_v_0ev5 z%}U=^7ldY$dQcujN%nB@VVaMi6&L}cC_|CqOeG3oTC80=A zwdvy$9UeV>bTr+kdpfC!)lmyU)k+$UY#KOrP&Aq6`lgA!&J2hs2If2#eRE{3@tIqM zug=Nlz2A!Au6lX3I_ooFdl5J~Z{Ff-Ewbg@eAn3WgFa3>D_ijZ!MCE>8VGQ(@eQ+? zoOLvc0H41P7x0T$@ za?4_~PGLJSQhZ6-zk_RWlaSwYR>U13tCpbb5U+DzoSkrG17_JF4?iltO%c?yyMQ6# z%)`9--`pH?rZ*i|6MwkPTf(}16MrRoHFp2}P<-Zku`r&|-9^%tCwPU(6WmTBIW4nZ z#e@a@OHX(TD>Ipc&ck`;(vEbsR23}s9yf#fK6^ZL90CqN5X9k4cVWMhivNP}dyV2z zh^=rS?;2pv&900n9l8X2kT&4MO$~7Tk@m4yb^c>C3Dar8w+DAf{JN4O|F)bd?O8u- zqNtYl^u|*x12FFg)runjwTorLvLyLd#B0MUs=G@ zF&3V9-W&#tK!8#L;Vn&IbNf^yx&wXnkrLKBl)?ZnI&LiS7-+um15fa*`GV$AHi`0C zAk1iUf1m0Qy9tM#34|R54(u>x;`Wnk(?zfu3dE3x0@gAo@JN^ypd)?*SHd2}L<=c2 z#OH(}6kiVL1l^x;Pr&8EL(ycl*e;kKa=HJu9V=56cCASkO3MaEfCpW2}?%6W6+yTV*}91=Lar_lM^p zNGVJ|8V#FEk4;93J|4r&j;B<=68*zr1p9bw_^AmqJWgj6zMP>2sfs3<3Sdy?;jz)+ z^5zzhkE)}!)ccbnvK_V!vN^T@kR=d`+S)(&F)z4gG7?waiXTuHQ-=vKas_Jj*)MT)A0!aF9DUkqh82wt1*$5d`sS z7q~z^cxxg0-_~-wba5TXc%^I4#Wx(eZ|S_do3~%Lox5!XEN|ZB;Y?;v@=vv~AWhCp z&Ro#N)8$JU%E>b9;!9VREe=;7YU=WU2&;5oUx<8!hKX1_DUlAQb75o?c=ZtF8|CC) zpAZ+OJ5QY^$O$YDYjTp{6qDf2RqnXA=l00Nc*{R|ODf1YA$%`~Q4#0p!TPixNuD9X z8nE8D>^k1y;-u0va0fqSn9wfvd%XXmx196fw`yYtnO>Z{%lDOTt@H_*tuA{nmk7f4 zT|Dr)?LEt~%f*5STR~i|&{2?u!KT}{U%PmKUE0H*5V}?Q$_75sc8w5+CaZ(^)Wqa( zsCl~X5H@=0CDpHYPZRjug|SXFuiuv8%%aY4|Nq}X<_ z|KPwjVQwO3y?WgMt4)^uRFClK4_xK3F&LmcB^8mZ_|B(WB`syczH-3@8>M*nUXYB@ z>)G4`o#1ghGT-=LPJ7#Hy0=tk9Cd+*DDrFE=WTRhrazgOmfz7$+h^}4<|hk;pLVg@ zA-9|Y9gsZ-!<<1PB;wp!G;xRf#j{Fx7;D1*fY;>PWN%I*_SP#(^P*|msgbR+F0u}E zUb^nW4u*H7Rz?H01wET;U1-FeMaKKkg}3MzCLZ9g^+ZGEZ!vgkhtk<5jl9`RF^sW; zM`Sv0umMU?ngW)`+$`l0j{Q{LpIMf^88Rz!F+|P@?f387v7UvY!areKOrBj~IOAnY zV)xSN$;7`)Y$j?7BC|}TIukSbonAMdk>0`Mu1G8(d9k9SZ$Lipk0aov7$vNfu&|7ff-cxB&sF1M&)%ozcd5m zy-4Nx3BP2J{;CD?9My;)FDRf8B$`jL zk+(S3hJ1C8Z^u|Ll!lUO2jsIvWIFrYrAs71zmnHE3k4VS&UxL-PF}0ra=MdFWEDm# zj>h+=LXs16!r=?f`oaZ5&p1H=sKR4V5Y^ajwzgNqv0lUGxeG<`3`{Ikv4Y%9h9BG6#8X9Q=kd&jut` z22!}fsz^02)1?rxGkELh{bmW2Ny^utx(y?1bt&I99Kj0uW)zn+@FfE)IA^11iX}BY zvlx;TgFktT6v4Kt?>kv=yfyo8W@NtCG|kc}SaI<38V8 zCec{e(mIKA6!qHAXy7mB@oe|fnDlhlfjJzg8E1pdJ>vNnF0wvkxckM>GiD?##Nfhh zqPm0@+yssToGD->lk;znT+fI5LHzFPDPXOq-;%s;z2PfpY$GOpv)fDiWh1V3=HfN6 zjNbXibc%T88I^mlM?}ciYiVAIarmEt&de4Q86Nd2*LkS>Ifbx!!v%eOv5Vkl8svZKxM|gJWa{B*&|HD zk3WYNWqp1ITEJp}h>&y?irK7j8v_+7uYyNLV-fgMu1(5Wd!H2H43@6S%f?gem6nky zYH5U9_+HSS4;dez4(l|#^7|ya{YlL&ktU|Uw!PelP&(iDk}oLj1+SiiuR}*xwaU;_ z_(P8lH1~N*4973WuWj8i#>jx7VZR`+k0ZH0-NDOeAUo!a+;N#l(V_9Bq!nQC=C7 zGw$`E#2V-l6XXP;aJlUhTu(ep@XpUaa-X(O^U|Y#iTPxMk=j074 z7-j*Z70C&5jO8UoB?R;abWb%Bv@=kxQQjkfva}BT-2qV&RYG_(Hi<@f zn$@#=EhT$3;Bau-KX{CS6nNzDuxfky>V>g&?gh1DPuIOX=Iz(RRG zcQ-mi-31S$`%Nt31a)rkzK@>>U3g<2UJjr8O5|DA>$B=!dYZ+Q;|zR$K8MFeZrGO1 z@=SpVF4$9Iqr``FlwD!Z6lv@;kKk&69H1 zdvOR!yLmcOj*m30nt54x#e>lD8e2d2vLco`DnM58#vRs{}?5g z*q*T8sKBe@?uAFy?ka{6oGaf@%Ja7RgR%w%+m_|(gaF3t3i9t*BNNS59w(k`;5ok> zfrc__7mXv@z@pMXRnl|$wldnN%q%$Tfo86}eyv4pUVjv@yjW`{ZUYQi!HKolJ^R~U zixLvA?KrphzBWPv9^MdHd4Dw(p}>8n4qevae|c%xJr_)8^yJ>^eE`{pnRc0Fm5%q0 zVn!Kp=%zu2RcM;i#yY2JSEWUMsdP-o*f*hEDUqv?-!19kAGU|DpRSz*b?l{vxGbw+ z*QY|`n+t7*5S%qHk@mM*ver1ci5c30l;=>K$lxk3_hs@3+EnNKFKw^6#E}k36Q$#o zX0V+3S&pe&jfQ{#xblm>1TFWXmtS?E6*Dic{~71)qY;nb|V=hYZBcL}?^ROcJJM zGS%%~(%`vocAvQL?3hfVISmbpXg?+;PGXmw89QU_daB{$*WzhCdJDXv$QvaEhRHy; zzY6Z?Zy+vd#8E(CS7EJyczE(h_F?V4qeKkmY7AD~8hGjc;;(Uqe7pyC{Z^NfjHWF8 zuwRXTEmzPhI2n;gv<{iLxIn;v&imr2cqNVe*!!ncAOfa2K+!jTT2%;TZu?`kHo&#~ zr^?g|G$-HyqDF79xG!#p_*8snsjnLasj#NVSr3B10OdC|_%FGGmk_s}Q8pSLm8GBv zT7yWRw?u(%=+fRzrex0pW!Kb9x)NH$fL2a8Z%3lQ6dMd|0-V0j@6OM+z=faR02<$N zu9BB4H5urG2ybofaHm;mkManoMFX^olNLx~VIMI?dS)uwo{&#b*8$!=xe`N5{H3B4#~^PJR2Zz!REW2(|xkv&Z~kMD^GMlR(8P7 z&BL>kr0?ST?t1jkH8Y1c|IOlu*H^I5pX%)yM6%7mTO66Xi_!$6+xB zaE=D1#CFFR)goIuU3%*G+xDb{yOyM-9am#Oh?6lXLsVV)BSLG{Nc(X6);M$UerU#B zaEkwQT+}RY==#2^8^Bz3LM==$s7KiVK`poiG$_n{>NerF)b8G%?@=402h?BFE7LpD z7j$0H6Voh>ae0Eaxkt|arubqbrW9LxgscxuBQx?IO#*7x|0xXS^QI@v&W95pGq8Jn zWWp!%#w)GJK##KxTBhJy;F5GdY{EL!vbT+P0HOWbOy`vM*o=bKyDdJ;ZQpI>yUfs- z%2l~z^#>Mj31)3T-9BN+WUc67GnVLQKS;f(y-)};Ao>@^SJd;cDc*C$%J_*0SlpAX zoI8&uZ@Wrv*h|2a42qvq3QdDN&L~G+7ev{2Zm4aa?eSDJ*|%yEbZ!F;u@jL`EE7IU zin>2%X3UKjWJSNh)}PYcrJB57w8l92A7Yp*FFvTgEED*Zfe;;kLnw1-1tWc&Cui1u zoqm2HPtFmwr*?+M1A0V|m0swvvQpQk^h4QjXZ;vN7G~Hpf>sknEq1<D@bM1V-75F9k5C4wYcZIj!DHYWZDMn*i|xj#fjxW_`-d zLyKglGi+jrrkETGi{u-n4wF5jG=^^%{|R!jb6b%hE6m#SRAL{`O%8#|jqs55U#X(S zP@folbgs~fr~6)>-fdtlft3~YO6gSB@VRP7)P53YjwhImq$M5Kyl3HSm=Tn zOnf5rM11h=md+zvgvpP-K%?ucp%dQI_^#v$!&8x({X`OC$c(Z_zxZEx}4=V7$W}f~vRvOk;elj0jbanYA;E9>Ww7vJB{QM}~JbrPwZ5B)z(OfPZ}x z_QM6_>(mi+yNXhRB`!ylq-Fh1FL=LPKi6n~O9?zY2(iYj=|JvVt)wg27euhpP==}V zu%T{EfRp4&dmG3m*V~~C;2o&rP1{H~r!gE@gzg@?IICY_F%U$~WsBmIs|IXY(^or=Lc19a- zg(>cWY%Vhylp-@uoW|jo55rIPuL=ePHX3{Dtn(VZM479e-znwHq47ij^5L&Z zgOO#?z8YLIl&0u5)EEg3AE(E|4%cTmQMQ};&O8z>aZTl3bQhX~{`!-3hCS_ma;Y=} zxf#EZ&VQ$?WA8U2?M9k!-m!`xc5uc&i-fEiGgltLxBtRDIzi>DZEpP+>mX{(SFk{K zkF$Q#_&HIa)OXUa#L*e{A@mV*E=ECS$fbt_+F%H03`?Cr%Z{7U;lMqF4l_0$A}MMw zf~s}ZoJ)a^;cW`vWL(>B@n_@gZ|x;*SQ`k-04PS=A@c0sGu~!|Y-6L4?8oBj9h4D! z@N8hz7!}Bv>8`_<5L|M$e%)Q6CZwN&BVJr;&V?6x9TAm@U>4>6HzHZ52crQ*wA@+7 zWR{>Rnd(q=T*V*Eqqw16M(Au_HU7BSkcccfDdss)8FHEupS~|`t2PFYEQPLX5q&3i zjsU&IF#OK0*Ik}GW72(b1{H4Y@wC%xP2>t?G5IKTNPrhixL@eLyk|-J+ZTF3g^+%A zfiVx9=!BandOC)OQ|KB8`9mKjp$S>_K=pKU8DS7h7ZB7Sdaj!EV79@FnDx-#AgkgG zJvc5nUJ|lO2|92Y!8vt|2WqPiy&z|bvy-wLk-P-d1grFTJ{iqigcly5N~BP|ENmh#1sZgi#JdOH$fip$Qma3mB$GH z30*aCL}wf9Ecu!S=`upoC22hRvl|t7yJVaP=C7PHjBtAf)+hxKFP;GUKUM~`%Kk?o zOgi0#1#;##m@4eA$_1-7N~ZYJFZ3`RcJ6PTko}$t{2Poa_8c2OdWgojn96G6?(0Vb zWknLymSrt1y=MZ0!hdDZoqgG3>on_OC7FZAb{B6QcO4Dn!%m&RzE8YYTMsPkOf9J? zXvi=pWVPx}?c@8QRH_Jv>_RUjx+BP6=7W+sQ6cOB+wf)pO?{50(vmn);W`^=tW>(C z(vmtY{Tu@c?N{<%=*V{>jf;Z+oOBMujZ)Zch47{*e#MT`D^D9mMyk9-N5wYB-HS9TLVB-h^br$>|`ZIveM{7cC{ zmes_w?v{QG>ubjzdH13*8kRoLg6&XaT~Fz3;;S%#rq3aVGyWO`m&iq*kE<}_pvv)| zWs0d(a)O?bPXoK;!f(}t_w`-E@A7{Y+lKwgFqwzD)IzjE(?K;WDSP*@PBaj(N_cj0 zAJ^DJhn|;ndCn0iYzV$bFY#k3)`Ek~K|T#xHQ?8o^&V)BHJbch9u#lP<@caA zG8I_sq4GO%)tMiUpkWz*%~hv9R;?80nK`7+oWGhxC*>Ds7RrM#|JS^0G4-(I9&!rR z=y+xst47>eVg7OcNq$9vV=W&mZMX}&hK`snhx6|fAu6-t!}B)V>#|nGN4i$Z`nTr{ zbSUr@Et&UhGCMeM`FWbsg(7Uc@%iL-?X(#Uy^$Hy9>=#@io=Su*XO*)UbF6q9aRZx zC2`kpJal6DT4IGJmWEJtJLQ8zE!~}$lkaHjfV)k1w+kQIZsf@WGDCktIB_qx$Ce?~ zEQ;W$czG7TEwbShu$$i6&mUZag5sJpTYI64#`T-_j$odr0+mL*)w>y*QA>x;vA^CT zB?q4@K@Xmm19QDeX=%kFF6oW%WoG;woi^>OA$uG``dRslo36iHBplbHaeB2+O2@r5 zVJ#W~)|ah~opu8)`L4idLUx8#I8p*}`QQ@EiH`gH?^i!+%q0%WjX({z$@vq%6{H;y z?bMVCvWt0?0iT-Se0+4)RyCXM8PUx7yIk;lT`g+xP7mFG`K+x4*Zb)DtnKVfHbpBa zvqEzT^-^VZO>XA9zWm0}6vz~}M-i&}Gne;4$Mf^g%CCkT^lE1fPOshGCYxp9RQ(wX z{#$jmoa}1j=x$?ZL5M7@l}04Qo_2y%ZL3dsrn`XZ5T4}zgWo5lGs{TV`}@+T^1uxE zcJEo078vVQayxyTGwVz3tP3h}y^eHLPuPz76jw-`k^~Z+)rAx*pzf@EjZP;r9Q!^> z;xB=W%3~X7c6fOBq!YbbZmnEq#3q2=2WDZl@#L|kl+%oH=Ci7a!HCEN=oBscJ+q`0 zt?Kl7b9SvO8NG`J#ldd8@NCiI6G=ZUSU4ep%D;`XsBOFvizifZw@;hKTv7z>S$$eC+$*C~yB*YyS^Or;qjDe+y3f$%-HF9(ve= za9GC4m%brUY(#R^WI4v?;#rcwW$|eBShJDfHN2@3^;4+IDSwB~OT7XIa^tyF`cba*CGdpvCG1j;!(cl| z60jL~d7gGD{Kx#@@Nk_6*~o%fOVts2L+7N`jpZCd@atL^H_%Ij`{y83SV38 zv(V>Ey+6Ip_n#Q?zK#SRePqRgg|-3bk&d{X+-d?ISADa2zw);tj?zxAfgWZ_$=^{`V+w zvHn*S{w>`7PqOZRZ|(iRVD`A!{-M177iN#^A9L|vy4Qc)^4~PG|MbBA%QPe7V*jVw ze`EIk>C6A@?*Bi+-bcUwZQp(Pc^^96zx2F+idp}!q&$5xPOc9V@83)u4+oj>=3wRJ;`_kd{%>G8Zcb({u7AkF4_J=Qa3^=8z^sOaAlxob;-n_3gs{=Ooknka7EE^Fw&@@g2ynwzzraA zV`KM1;i9Ybh9l4yrS89OW^JI5pB`+!yn*xIFP}tK4qP^~_&6*MJ8qh?T(}`1A&AFB z1I<%@#b)pG&zkb|OoO!^r`94A@u^DlSDkiPP@nLs1qFrTkUtVEz$>(DK8QZ3&w^a~@?cD^EhiqV@Z}V_=0V%6UFd>@DA_ z1`g!+3^8lNi`2P@0}B^(x90J2mMzYg=&WM9?)I;OGMd%5tf3cTwNZdr29ODIFWLFi zqi`9`dR@bC;Oo@{GctzWJ(fgh&o*pmOLJyl$$+ouUZSUVhiph#$F`WA@Bj@f5~eqcZi!{Z;Hl#IDBh z1u{vlAmRhkb0@cpeh``V*otg6Ms7b^_2~n>4BuKj4lP$P*Gy?*$Tq||eXh{3u)SmZ zLt>9x_Y?WVQv$gNO$Y*O;6n6cqkv4An33q{4`2TH7Fi-hAvR^l+_w@rt={WM5~6Q$ z)*w19CcVH#&kGVSg3V!>eGiubNlUqa(4jgc649rze&Ccu<$#U<@PRSq?Jzip>@TLW z3qf{>R-|7Z(p3})?uNt3;KjsHGi%7`#XoYAm&Gh{ZP~Nt{a5Uf?dE7;Hn-LT_9DSV z;6W0R=cNL*HsK-(;NW_EXVnIK(IiiDWGO{PQi7LI971pMwa3K=#P1#&8!65ymMc{$ z87ti&=_+1gvK9FDNzXvlJk!?sNpvM~Q;KAFxn<({!kDT+xnQyqWEgXoLMDQ?_aT>``I}Bw4{)X zNoH|qhwC#Z>Oioxdq^oy%OHGZ44eHfN;RvrzY>XcU%EUs<~Ed3INx?Ez}=Y<{$(%*T0N|*e#To zzOZ11vQCc9OE$|?0 zHls#Y3+869f5@T)dF(%Z z+NHNxw%+{H5AB#P-qe2^oaOg}JAOsR{3tUIaUP%yY+xzNQHjqY7KR9U_0Bt%)+E}~ z^xu#ch`kY-X5EJ}4bii*#r_%_=S=-eL{er5uH&-Z+Rr*kb?>1p4ytQj6V#=1aO~|ELReUF6ADd6({XbK?iy5m4qJxT1a?Q`5n#u%kLV(-B&aw7$|c8`Pw@!tD&Ap6l5S1No)usPsJq{A($ z+JL55(W`wi+B0?mY(nC45uFbhTQUnrxb^jh-f6gycp5=xAT7EJhU<+WnZeM~)(h35 zwr_sk9D*kI_pjXR*gj{&HD;gi3|dOaTM&ew8gr7!k`N)O)g<#(AnJ{|rw(2xH{ zsg*I^mCPcU^4_0)7EVr3d%G8;Ay6nQ9o^@=Ql z``j~bD=dczL%2RQW`*k;DRc}0cT*@Zkj1J&1d;XCWq!lRlF>h~%^NWvyOLFX9Zfm< zdrjiOpo$Dmhv$15zHZ%h$aGHH92#3U5nL4CvOmc>Giq2xpGInDLN8_fQ0tKUtT-*r z=rS0uqndQz2A>jHbR#?DM1JU|lH{^@7Jr1yJ!q&ypZKa4PQ_uOfpGJ?X6`p47I{M+ z10tJqUTD+T7!c^?$I_sCW&9(h)VUFnc#ChtFEo9a9E?9FN7J&$kRB;FX)N`xL52Z)W0|q!C19}8Nmd=k7G5n zmPo*N13b0Z+MBjl`C}XO{ovY^LVv$~7x^nf=utVr>!On z%ivGpPMA?B{O&DFh4)X7RvHlr`E3vQnP?>#?{M)8g-TLyT|3<*YXL~PTkz;qx!odS zEz03!kwk3q5AX6QjrLz!qOR~)R$T3`eg08aodLyxCHcZYu$xMP!6xw&_` z4PivAN{}(^9@&I>QF?yr&&yAtkXPJ%AK;XiQ$GqT!fYsxtpaTbUBKjcCs|k!^`2I- zHsL&bCg=%qLjLWk7vweY zuw!0gUt*VdFTTnp3Ou#->#Q=wt)U9|UY2jq`Osm?5X^?6wxGuDIz%?qIT4EK0>dY$6C0Tf(>KzZjSv~b zhdcmD^wUYCex^vs(;B*Dra>?#krmt*Y>eRZ47M82eP!VEyW8^;Fd-nrWP%-P_{7M^ zK<4sKclcwom65Q6C=kE}brV3LV=?uUgepJ?uZ%MoLkRK4Q^he@hTu6uOj^z>W+aTj)&svs_RP%i+_quq5g_gjapNQtfb`f9 zX!n*9*BPYzvNic$CVbs}RgvY7TouvHfcCSIl!s&fVep;Ke!u1{a~jhF{KfIYq?c+>Or~YiFSFP^L=aC6Pa9!k%xMEew=%HuEUG8C?apey}P{- zfz{QGWYLd#gElP(3^zU>PEjOW1EJ1Ee{V%pOZo(Xx-lX|NzE7s>j+N2kEk04Dok{T zigs(e;GW31qn^;$MV>nlE@ZyLAKL9g8Z?h2B=IC4Jv)!@x{#25nQ9@{>l;E-#5hMM zkF1JB3~DvEkvV(eh&ITmav2mGa@c)|^k&;eu|mp>^@P09n9!ZUjTr5S<4TyeXXhmFp z6qefX+Xoc3^5S2kw-#OUeJ5WFz)-re;Br~%YM;h(e=2bBQp_76_4Xbi^2Un4!iGQe`8?EZ zY4?>M?UwK%iV5K@=pZtw>j-CPuzT~8bMxlXCm!_jc+x1`4HkYgI;NQDaWgbhb{CqZ z_?XLvT-~&QwV?dS4S&Bg<9-^T&LkOmc=nz0Ste1p%sX;~&7Io1d1}l0oApDCxT)@m z$33#^$c1>Yy+#gzOC&2#_G|BzA)}Q}GQI>GzpNtF);+I=8)5&CO0 z1yekr-sO--e=oiLN^vpd%iirFa?o*YB*0J&To$V9Y^8?qRKi((dXRZ@OUoz}1^w(e zRUxwr`CK}ufgE3oJrQcp&l!$R>J<4Fp79;|q(9SQ z%3maiJm|=}>&>dAua?vlHuPPaxt|uUEsFB701NkJ&Z&zty)o#VD9FY(_|7fxE*5~U zd0f2gA&oElrzT;67xzrMrVzXU-9&I9{kywha#wV|7C2KOT*FkX2!tLda9fDz7Hku) z$(@P63pyZvHz5kpyu2I&c^q6k_c%Yi_wj-++GIb9WZzAEH%_6)+uz{#9tDJHmf+KA z-ea6@E72G)wF8JO3ml1c>+zO;O;NnfUtV@guM*SA8UD@59J&i*TP=8>H@tL$sSb#v zL2uFE=6??2o-(r8Zj}j;*lt@rc0u8KNgR6KoI|)Z6%3XUN4~lnTwUzg1?Kc`S^VWT zd^yES?E1z+8uU1{M0%OCvY987g3j>zz%#W4SmL?Qmabj^MBQyrI2?7tiHqJ2-lLYh zRTmyok;d{Qc10S!6gj*M56*)8{cC*uv-E-4`hnxSpBe~CA&&p_rz#55Zz659{uT1f z17r`J?=00MtqDNe8KV9e!;KZs507WGH{#a&JP2e-lqzz7idBZYI`Y~ZOd*SM2-4?A z5g%M5S6+^HaUWN}~4p}mH&lx1@^F0pQo}z!^AiqD`XU??wy$^~(0ft6x?Rpjj zZnT)^(BMG!-q+2-+{jM@072~~K073CP-p2=ds=AymbaRyUbGEKU`Tx^8(l9b%VdH* zD3R|O{RmXh<2*s%Ug1m{Tk)5OgEvR&HG;Zq3N{$f!w-OdFAREMbj>-8ji*ef4*ZCy zmL^oB4j!^JGG(c2(Cz+IZay@|(lJ?{_?90>3H|3G14ZL}T-ZMcF`lICLlq~lt*@3M zmpQdZ%41a@#sbns&#Jy~nC}JJF9@@#J*XE*{Prr0;VVzJP?4qUOZfTO9RiGan;?t;ZA9$%kNKCHqh5w7<{371jxWnebYk*AMJ z`YWiijr-Htu6&}#q@^YW_?8&GWK57>h@hAZJbDb9U z=J(F<`*6|tZUfORHh34_nHO?JW^AS|vR<**6Tvw?mI@Io0^b8bvE9OcFejO!BzqlL z<;C)~RaoTkyT{M!x19y-{nB2xBgggr&&VlUxWI+7yDh_~_6KdSBoT0~9-1bxf0DROl_s$YCvu&C@tSAd27>ckkc)qZ{S zdiKG|1+99tfP>f_h2U6g9uUEw*u62~4w`^F2{-S!*Cw_RD#UljuVaCx`6nTu5>ks- zt#LARJ$Z_O2w?wX+8$sFFrHE{mSOUwK$@i@x&tW2 z@mJ_ng@ueR!H*Dw0fVGwsmkC~Y|?^9AAzoAAm?Vp;CuYJW%s%KqIdu2cb_^`bKU7n zb|DAqF1TVq!c(k&*Ih#`?psOdT}Y{wis%+se$s_lj^n1Mv^Rt)B+UKYPtuoHt@D_- zSSIiTnHgC4irTZ9+0l-0s^`=w;2Xx&^PU~7WR0 z@!Zsxj8S@ZFhK;pdCwERN9&o46IW`_`w@LeU`I=sfDL0#Ri-M^WZ+xm$#iapp%`Cl z>>WmMFoSR=qbo^!y!Rw)gm z_u5&^$m7eLu>PJ?buxUtN&*tP{_CY*!o3h_KR&jb@@;2XHoY*CQx|W1RiSmFF5Fy% zza^pdg8_lrWdqyF^>Fa=yl1}u$YP@u!+mA&z4@gLsHypJu@r6%;E&yV&sgyL=GBUz z|9xqy`BWt0x&26z$D3zQ>eBj^?wwTlTgCguP~7H0i*eT`M+Tt~i@3HKn=|Bd+Qv2Y z9_yi@lwj<=8*vCic<+O>A86HAOnM_f+-5q5e5IRa>|S@ZTC%j zYqP^?;avc}t6=Q@P)mVgAz8e$euIi2j|mFxMXGN zAjQ7+^+61`^R!-UpB0`}ci@3aU>9G&d2EjJufsR~V5#~m0}_x}&F7aE#ZT8OjF4B} z-zV}5-|trb@RF}ubRQ236G|kQ=0Px%`5R4HFMDSLDpp?5r<}H2S1f2+nSQ(^W*&A+ zgQ^ND72sDcjqI^mo=nI8-W@YQdKb4$Z<*J3|1AK%6m0xtK|BzCva=l6Jg?ICfcydS zDDK_?rZIcw50WkIzV`SF6wu^SY}Naz1?czWAJw#?RG#>~m~)XX6^i?9`)7kdy0PO5 z#~$rB!qY>bBcX3`NMNVByC~uVSF!U|mYQ^=N-l`iC}5c|#tPNrDLI2!S|bwqZ>7O;0p<-PDw=ihYo-RC{! zJ!;u5chbp=QfjWx;#gZ*%-XZPoNz&PQgRjE_MLx9zVQuCMbAdhO3d6=vP_=lC=BRJ z428!ckzdY~!8(wzF_AIxG;uAWs8u!bHpv(;jw(*JOlG3%g=IcvZ?~;x)l6krOgkHM zG;4R(Yi=FWLu)&1#cd;Ts&W##{Ox$|dhXnL;yvXyeU}9!^oU>EyEtZcOK?nZPw-Hl z$V_yyNEtyvB*Zl+)YLH;ZJh=t*bw4qJ+G;qZP~k zM07>>QMw$!t358E`Ktr5DhzQ*pB&%)4W)fl~PM#bm1CBI%@`>G#l*(Q5igs<_ zKA7t%3JzqDZ)ps_??OnY_MaT@9-kkrsYqv;wwnU>BKMGnS>j`q9K*@BW^=P9xf^{6 zBYbst_0ErULkk>)FrTT`gMD??#+!}ybA;G7r9uE>nIX z!2#5-3ddR79E%+39ETjw1ak!EvCdM$6nPkvKc60h{$l0k{!Kv3@3=h}O7Q^mqRz3r zN3zW(NYaceZ=yMxtVgvgUkJKv17lOr^0fU;_M*&&Gl&g+h%{Y>m+I4Lz9c9M$KxJ& z*6-!-X%D9);1G?Aq`Or9txX@Ui{}zpJl5K+%``P09+@eK&*yvIn32jA7<eq%Lo3rai4J?cmYk9ZWHl_aB z%p&ayx;}V8_aH)F1t!He5bq#ovC@kE>oE&k)!SMV5`hBRlgk0m=^v%>onJ|fC$rrZ zb90>G3AXV7nt&5`ze0n%vN0$@XtX|CV}zlitsWvz1l*pvDZH7QGaXB>v%& zSG?Z5P{Z3qm7$7P46jglMofZempulakm3CZT2K@p>3@TpjY&JuF8h2OA-V<|Pv9NV zQF~(?8D3$~`iHN`ePK+ueR!d7jEQ2yK&+7VA#2853ZbK{h{bOej_?X0Cyq3}@E*NF zjy&r~5B+{u?3F?AMsTUU3Lkb#DDe*huUwTc=t(k;u+_;2`-pYHf`|($;y;SeageXE zy|ODQ|BP7IgVF8Wh;k<%jZoKR2`U?qu_QAM$TuZ_szU(> z_Lt-`lBMla)#Y_4{EiHRNh1D1Q;SK6NKNjD*+p4M8QbyIJ;@PM3^!SGRAkh?E}}!q zfF_CZQ&V8ZvMi@pNQbP63}X}pXrHMLkGa?E@lzKy>MkWBWm1RYCi%;V|24B$0uk9| z3T3i_-hd)%viLsqHThGdFS3B_Ct0dEna{bwbNe-V%q)p`V=?;?`-9h*PsUyquaU3i zubCYR{(1(?RO8Ac3`U)=C9kQk4IOF*RDkrQ5t{>xeYflSYj3X*UL_(Hhtz@;p3$6x zqXWEsypIgjnw#{RBWwF|Aa1AN*W`~Bm)h0I9wP!Dzc=Y$|G5@9*tjNn;(Bs>LU~er zV)6pKHoeAoC~vZKGj!9CCT?Y*jYSOnF=^DI_>3tdQIt#;^+72mf2l~O&Z5)@i-J|a zGQx}>i6N6j(nL~3G7`py_gVJ$t`VO~UQ1uoUK@SogqZ>;0vG}!R>2{@QeahKhIfhh zMCFmPeU58BzTXkF+*DeKs$bJssq~^$zNFC(Ru9nD%H?7Lc(Kcpe^Q)v5K z%f89KF}&HlrTXiEBl=CwOr@g&y&@2WP=ph zM|vH2TVAWZl5L@wslY;>mT;7S8jqSl9jdZYsz|mLFL6N3Ezxu~;}(unHgs~+D(HL) z;g*}3PkKhPQBibUvMYas(2(a6wJJWLaJTTYp^i=ZB6qXgNvW-5V*$rZwjt>s=hHvN zr;i1Jf3rQwhb{1*QlI{55W>!dRgxSSx)*Uv&8##&j(8w(Q|L(f?D+Ke68VzjEn{6` ze4uiSd5vyW)ym~5r%}R{!&SmnNGgdc8(ScqM?Q@^RXC*x`-4CY=#R7C=Xy!OBJ7IR z8Tn-w^Dgs=mo#h`D^nT$Z4#Z2h7Pas0Z5ea2~)_WP$(8)Fvkok78#l4@Xt$FG9~|= z{*57$I&Fi&pBih#6e}Xf8Bs{jlsGu!a@wtGi7$d3lFvSpFHRIjqJZi+hm1cn0jm!jU&a@Tl~v^r{4G{%OJg+fNAmAoxJ|p#MCUNc_(-9$d*xXV8CGZm)Fc+0v}IZC@qQRlbj!%nYG zaZiaK7&1hUBw5H+J94!X1oD43s8_`F=G%}9getdcr5epT zOoY!Ug(HR4n6~dvQ>jNAEm*`VNJIk)h^SSeW&|4Os5Qf;2^z>IDuvDI)yq+e2Fzh6 zN{CHD)MJGg|Ir}M$Tx_nrzTb!ijBldnwK-rq?%3CAF3B778sgSe;W=dD;gF>%qd$6 zkSa0~^EUDumh70L7WvPL9cORrJu3LVguj%|${lOSP!@94q);~W7OfUr%{J;U)#1M1 zgO@z3c$j%L`>6j^&tc)iEJjwNO2i9W(3=Zw;9sR){Zk=?nG3TZ(V)9(V;R6Ak6940 zXtnTdhz#dUkx4ck*u)^A=gpvYt&zs=XS%B65?C zBN;)|1-A5h=*Cp5x(K;Qxd^ymr~pqvPC=vvq6Sd+D(?{M23?n4r(P#k%d+uJLE(Z2 z1x^Lf2G9o1^*;5g?WA8bex!T~co(1^Ti+9WWPIZPqqVYpK)pwOpngUEDAlf5I{u5U zQh(3TAy74IrY;Br1VZqeA71D+4^O=0TZSbw8nY^iW3eGO8V8wABd=0YG_Qhz2Lr&G zg*R)XT4zQm3hi&IHUBB+Qzb zX4K2MY2QjD%sQDC-G|m|xB)M*T}glyzy%D>iq4YG%AEw?V^N4g8XL4>1)brOiadtc)t)CR95As9S`n}!st>2wHLhbUNnFsMUXdL^ zl%~_74rd(D+)mi3w4|+TUE)8*yN`4n@Yw0JWUcGa#Dfh39}u*t;sh95%`}~^{d`A| z@}o)ie2)PeE4USN4d+t!9+8@;T`_=V2uda#pE-1~&t0*UVJ=Ex9G@6pIb1x%svAl- zzSO(4y0pBsreT0%NlFz>9ZDWbv7U?4$fa_+{lzXL6CJlNd7EIAY~yQ@OVUG>N)%@y zFI6hem?G6WVo{syC#4)Wfn)(mADG%ZytE(hBKNNKF4#__jZ$@dbq4kz_zLzK^eXx~ z^s4(2{2B6D@`?Tt=_Tu(ZG7MKhf;Cc?u9zsNXPHu8k8MoEgP^o|J2L7&74em-;Zpf zqbd0?!|U5LOmG|h+4!3b8Xgo{~H}LV@Wlb}=Gif%vQezOWcF283xb z=99#e_*0(txadW4*lNv>h65mctzkZFS5*@#CWKVTiGV|a!f0ZHym6gquQ;0A_b;-H z1TTS-rkwSlY96`My#rgam%W3v0LEUpU|hLT&g^t&fev71&-xs9v49~OVn!Uj-mYv6t zI1=S{~a?x#)XuW+zYqfh-s|GJ~`4Aw4>Z!a7iwf61xtSFAB#me4TmMV}e zcZc1=DYJQ&+SBnnm`^}+-`<9pS z)~E9p=k;fg;B`+BuBXTIFzNNmk!X~s?YmIlrqqA5@hxwRN2C91`@FiGaN$6HyxJbI z^j45*YqV5vGK<7Y}OhP4}p*o_0Y7Cs^_L)AmE~6B3d?rsZDl_ zE2O9;f{f=A@sj5+nLDF-2))dw(ZZZ;drieDXK&S( zj;bmwY9y&)x5q28{maNZa>5;*v|mo(|B9sRCZ{c5Gv}UIfe(bqc9%D} z&)$(gpK?W=%u~qK#RNy&q{QkdamqA0&j#F>>L7KRd@DjCy(&SlhM1Wh^J3W~|C^eP z%-TU-yK^Zsen?~+S@We&#LfMQd~_;IT#U>reTO(hgb3zIKwh+pIYY%)rX90teWdK# zDqVkNqxztuOB=00&1Z8q!KubT^3$e*)NF{XvbzqTwXx-s@z@YbI&h$60wNBiU^!7ko|y6gGvmb|who zme{MZzwS!T7rbTE1Wevi8nhwLs70bLlg?*J^A%Is1jb&jC0R`yuR2M08P;0vlUcWa zohSf$#k|7$AYLeeb~cuViR29K!l<8dEx)76{yb_0c3tyX+x-`dSN9{ry8z}`G3CA*pKBI{WhfeI{6 zp?Oy^JU5;AjkWiHH^xx|N#@;2;QiPdO8@CV@0IMpGD*fKsG_YMy-!6(OO2~+7!Mywe+C9FlV~Sx+}S1L3y-cIu-B+_@=?H%D<`~$=EnP3|`9nzqojFLx0dJ z02_jcD9Y)iQ2!RGk-ATCbwPil!<;MemeV@fp-@{#`vZ25D&N$ShZW=m5wT6=2N^*z zyNAU&tv{$#Z|_nXkZ*~9w-`R^fQ5_4hfL6d|6Bx1 zVEBiNfpJD~NU&0H+<7=WX}>ZXb|WmgANJbn4NkcLQMw=+I1rYLMd)!LTBvHcYGAZ0 zw=x65%?~}+T`t}|1Om!Lqn%|!N)($~%XP~XF3$0!?}_iaHBgi{lPDgZROh-Xbq$?{ z!{l4?>T*H;j+GFAruX{|(^^Ka!6z0Bi;`XI2n-SsIyJb?5T< z7~VJMcnv-TirWUk!(lwKLsK;bgT&UGn6<|E@| z)6@Z>;G%k1|v&Dj~aK4Sh4*SoN3F&)z zxqz$SQk+YvBO_y5f6AQ6#;JOeFY9O<+XH_&&lC7_03Z|sfRB%6?H zjElb@F!U5(QI}^4ji10XWE59}CC3yPofTK3A`c@q#F2l5G>s@HBg`B-g0`p#X0rt<}}@mpu3b@-yX0@MB((EsPO zfZN6DA6Xn6!J`?50m(W8pfPQsL zXU$OVgJ+(`P`%r>WfiO}wlX349ky`dv?flT$c(GyrFw~hSglJ6KZBw;l?{aN5!~yg z?MRZTDz}r|Gwd|$oyby^TN*BREt9p&D z$ow?tDez;K^^=Ze+LF*@I`x=X4uG{7>+xb&UX!C+6UD&snPtM}MA?03fe9r(9I}2% z(xf2mMptntTDgf6v)&XR9Mz!8x%F9E~0BS6!Xr1Ag_jc@kW0BA>C-93A+80l?mT8`<=4K6K z^zJj?xcBgf-+Eq$4~h=sG}==I-}C@6)s6~C7p<4GKG)MH}=Fu)(9aX-99Y$h@lGRe?d9 zEqc$>EzfIB9-7#@LK)!WL+P`niL|hpb7kS4@0$wA%|>4at*N2qO_gO3sI^)F{^E+^ zQ9IR?!WYU<*Uzi#ITs*(>nqk<(n3k+@e{ER=MLXw!>G_5OF z_0W&*Cs$|{=;j_nlZ+DH-}eLuVBI1vzTttWe^}dPCV}1kT7W5h1NmI9k=DVYdNm!) z{i_Xl?gy#@4x#{UauCwGFdU7{Hb2f_1JJ7x3B=-WxVe7R1x8gbSXA>7IQVSRA)v%u z4r0@x$pqJBBRd4G=>o_ZL&Jl@3kEq{P@KT8&=#@~?xr?kT%wrH+)obBGy}1`9v6BhZHEaG1Gq-W#E;7t?#Eox^ZcbVfemC^g;~DEyyeX3 z$s~brANZL^2!gMJW3>LaX37Z0;WIXbe$3X`&FPR_c%uR% zHyIb96-w2xEi-8ZUOECR7ldq|dB30YDUrem?=ZqG^Mz2*Ea{0#Z{Du`ZqO)Q`x!nR z5K52Hb4Eb)Vm@3;U^xC$Ov}FREeJawY|NDj*lJc}(NO?>ufdK|4WZXth{Ax{ES9P7 zyOy-vL>e>h{JWI_hJ7~NXrC0;hkiBS$^r~4VCvfaXD0z*MjYO2W$r@2j@HXKTG)Wv zdr>>21B*Z>=4yucL!|QeeZ68K{doztSKKjDI2~?*?2Nw$?X#$xcrX$m&><%I61uhiymR&)eq#~9@BCTMl;I3Bf1?=_tsgg$_L)o(()a-D9s;s2A zgsC&YCSC9zr+Is{X9lkCiY@)hkjz$Nn@~?=Q#g)kHq$|L5}J5#*uW0e?w0?10%L6G zlVBxK^EDv7k6y!s3fCTS)z2xv6qS5+-poL1v7^)sIWGGS+*r zxUgeZ<0LPhEG*$-bjXa{UEl|8&zPgm2_9)+ia_TzfJ>-Oh?7QlQ`8j)k}wbli8#TE z=fX6Pm8A3er88jW)d8eZz zN~KehB@X>Q-Qq9g?pwy2h5#eImJ2EmstmlaCa|b5VhEA(31|+hL{FKXdkHuoplDlI zm!Sn$JNM;BA?2w+{9dL3;%eY0QQSK|8XD-65V;w#2Ta``t`DTiT%0NYp%JkgWf$<; zVb5#oa?uJCq79m&ksSkEY|6yB-|`TO9+4l1nM8>Ekp5ueA=LsrqP+f`+u0_5d%bsT zum1PKIYBPkr;>x&pzCnsC~OgC=l?7*9LMti7hH(3m6VsqVS}8e(3-jY}HjX@7a&ndHQ|DHGn-^Vb() zt2QyDcb_d(`Z;a4U!w!E*|_@{P!WA(8zm;XSw1^x)tS9*?-1_C1n0ixshJLCa(Gg` zZ`?qh;V^&wlmCRDMevhe_ zPyYOmE(nHOw`nLI@GNbAws&t_2^Jp|JD5VjABP4$*>pcB-IUhuyhVnQeCRAtKZD>9 zn19Ti0qH#g_eZqgL~_HJ@Bw2?`+zcK1N}d&e&YTheg`W6yq%3h@(I`oR71gm2sNJ~ zJ!(Ew^E#d*Jyr8c)2%)$r_y+W*j~>_5*|0RNtA|PQ8;SRfJnLgt&K-ypeFnSkP4&p z2ir^RhBU3aE;kJNZR)ya2QD-4fq{yJ1VF_?CCwix=`pR55j7t3srKCgKG)B`v3>eW z1+IqAg-}o6M5rdp3W^I08Auo07`T!vtZs-^C1&aGd3nPmW2s;F6G2~KZ$dz-y?AAP zv2xo9Adrh4V^t)f$cPzX7KKaCM$u7pA>bw+RmW|KdgHP~wVye`jfu5V=*k$IsV&cd zE-^bdf&?HB6Au}w7j_b}Q!Yd+Y2FYIVSs^y3K<@?AXo16QkcLjnp*U@-A9={dC(`y z+qf)W!gako>p|>Hxf@rBXW?}|V77Y;DiYsWjJsmd?r^Z2XbI9OdFk{(-CGCQUx(dK z4-0V1)$i5oH8q7q&;1()e(*p$wi6$BCrYKge#X$_4oxTJ>YYm!a>c+G&Q zI0C$0hu>Y8Uf$)u?#GvK4oAyIMVBNrDRZw{$g4&1z&)lmold0kj{a>R=wZV>_YTb@ zzYt3Cz9JUjF96hfX`H3ivQLaY&J0d?Fg3(Pk+gU)(QJz5<+4BIt68kbn1<8~_o@ka zszaCcKg@pFT^uu{F@#nu;!h@`JQ(P25VNGDH1C&}E&Srl7u9dX zY!zECa-(#KGeA+5;(_pq2jYtM6_Y^FQh-}pCejRl8UMQS(pq(G`Eg~7?vwoYiL3`C zD&@Caa1x(ms8D$>CYN&H?(dhZsoSVMy+2r}z%KY!$teQWCG-QEi4FCJCox9V}FKIq=V z6q#xL`WbuRjQpV?bVWzTBdM3OG|3IfMe_+MxCyD-B9=+^jA|qYe>AXp)^o6Uwz^ij zX1&B7$h+B;o#PR#rji(o8j%DD!&oxaAXp-;F_E#F*iM(?5sIlJ*ApLWM#CEzX+oAV zcoiFXK5euMJV|wg&tCk%hDSTLb~PYHtOQp zv$O$XQ=qHEY64{0Vuc4^3<^(LClOKEQ0Z*P^Xe{RT}6|-Mvq8ZLN|9IByN1x4xL)l zTAG>*Sjfy0`1$LnwG6RVCJ`spQicFo@f+n6nJf;Dn6o6Y5f#8~dij(THKw;#FWFMq zc9l{GcuqyNDG{QPdV4Vw7IX}(<#zCGS}AKNEv{!iWcf)_B5$cylqu@f`o-gT7IcP2uMyqGb;n^L@v>q3{{RwwA91D-j!n$c&~ih_S)r|a8r4$ z6|B=xQkp4Q?4HvQ!Hi|lIE^Yet(0|p4p+F1AEBXYg-F|^npitaTX_Bc@DC*AEBnr}VMX;U zlr4sq#UqJfTP#gvgk&6^f-h%?%|Z%>)dBQGZs~4mhT#v7!E1R>Y&?wKI4><9Wz)EX z-u2$t&a=8~=Q1TF4W_D$9#Y;VHT9_NO7*f zfav$mB^B)It|8OOBxtl1dq#W#NDD!6}TXsvm zQ)NEWMW|j#Ure0iyb-;{9Trhdsm+^6mJXP2>7!nRhr?jrmc z$g@v?ybBuTwQ=n3XVpYVDf3Ke3o_0Z(Of9EG z*DNd(6l5Go`b(u;9T|h$-0i7{zJl@X5SxpvT$Rn$!>Ok`x=8w0K_99%b_B*uO<~kE zu$d57eoRQFyjD^#gM*5yGdHi);_kUEdMhUtVM)X$LwT7tJPRNB_oRN3bXOo4v-qf( zY1TYXf%|A|_yW4bSU_Tg9V6tuU%BiU0SDbm{PLa1LUx?2_GiMPFFm9=psw^%RjOj-? z389Lh3PWOyKB%YjU+Wu&XJK5Q`&G9R62EdTByw%fhKZ9g0aZp^_b#jHPmoe`IvT)Z zmzvOwHxtvo<~HrtYo;n3#JsdRMxoj=sG7I3u8z2}P+oH`p^O8m+A`F%I>%T@Z16NV za*qBxb;JV3Zlba#7E9xyBn+)DgL~lq2q}+UO5etmft`VYVD@NP{C;pgB*9#V=+xF{Pp;NeIj*C_1FyG>-p9w2rN>TY8w-l?ldWMS z*8FIj&%Q@ZdUW8}SY_nf-nLd+U_IR9xz|-hhjsPCsHP@tZPyZ#$lnVeaHRL^VGG|t zh4qWJeiVVu>N$ej+{T#4b1~*+EG!Z+kn);XKw;WA%|zS+!`k#PPK1yPS{?zH1xCH1 z7qq@Yc8w2dGQG?mk7$_D0}~3+9r;Y}a=Z!E<-A%~YDftbXII9J@)>%M13g2VVI%9V z2S^4|4M#7gu23)n)A+!{f_@);y0N$7R7k8nV5y#}^Z~3ri91G9((_WrP|N>#X?l!@ z_g8!w@j+#^vCGlslAktBE81=HEe_83T?|>4z{&aCo!?yz0{3Ni}7jG3vNT_ z1M+mAE4>)2^RL{tk8$u~>;^xrWk{p?otb+*;;O4O&_s^(>}b8xXKq|-P6`$VP;Ri| zw1F-tkXo#8TzNAQyh=!!a8ljtcnR%3H=!k}B#929xWWA|n&PE|-6N`2M3S-=<0^5p0Ic#9#W+2HK^mtj zY$(36La>hBik&Be#Dzip%5~x=c)BFDGii_g(&mluN$Mtxg5l~jjRI@%}iav z_k$wG(IW(QVy8++*auvcE1FKhHkEp}%%C+~l%6wSRM~Casy1O$F|qaz5kGXq>XOP? z4s3YfwBlA(X`-1?5sGRp-^M_xDJUSA&--) zeP{UK%lL(N(|^0!cIA1!*!E`2XKy|j3c==&4=?-I9fx5JW6i48u`Z-VrYepjJz`wk zHf1Np7JH9`J&A3hF~(BARm3TRT-ll)E)oliQ3l%@$Nbwd%wrUFmsd@j+Zp^x_NlmB z@*F4AhaN*zxHj9T5o~GsOHXADZ_V^WN;5(l4z~BlH^*=vT=S7Hn)@X>WLXMFp`>J- zh~ou`USg1T4xLYQ`d+Kb6~u|3;IErH8PWU-BKfNn)_JvBQ|FvCIE%DbxG=S z)=5y}XnmG^0hf_xA7cwbBL-?BKy0c*Pft}(qoOP^i3`tGao6)(jORacCdg{sPNP~s zd|W(iw80Q5Btc&vEw4dBFASp$X6`TT|Il;TlW2R&W|?l9&NIn1DSuKaRa4j;rx1(T zAcbQn;mT6#Fw| zpU-beS5xA@uM<>`vRy)iSTl-A+onkC)z4wmoJ#NJs^MjrO#K;mUbesJI2XLoI0w6^ zUAU~4t`;q% zTfLddUY2BHf@&;3#R544x?Q%~Hpc$7Ib0-wc=?BFufDOr4T(lesm9B7xeU%JqxZRi zr@FsO_7&7CqU?{6Ky4F(i-`|O!|*$=@I&63uGv@|$VgiWbV#eX=Z1yyCpLp3+I2=eXYj} z263^jx>)OIP)sgxw7ss6?{vHZC*~FxWS^-s>xjg?oC86Y$DSj&y*#pjwr5C|XF_v+ z91(<6^OF9;6L0pDZw{=t$uUVAu7K?H{Pfg8l<3XtW%Ntqi}d1GMc4#vBDepPgtDjr zl-K)gBYHSeR2xFG0z+@7A(&Sg8G0ug{&2mJ_G$1&sJtB#4?W}LJ=5&*ky1Ujt_JRV zkX_-A>Yy?X^T==Ge3xvznmAaE4eyHRI8VnpiK?!1O_!Srnu9w}CAZ@CeC;ietkHD( z2%8G(%5LvJnz9kNSyV8#;$f|2o?vssTC9azCcU!LR}o9?0P@qvl+_Qx32|M+j(FPF z(=4M+gl|zL01tBoF(KFpX;h50)SPRkl-c!SLmv>d`$%vABwZPJlBS2R4H>>!BqE`s ziZ>#{((@?Ct!s$Wu%{YL6~|cm&3Q}S72HFeB;1>=X8Z-#OJYds3L~- z=C)ln?)#Amp5ir*>b3r@0x>znwN$rn{v%rbwSFl8+Vd3DRzB_+~P*51D`P$)ds&Z0D*QUGMH?rlyKWYQeI7VggUwMxIMjir!DZw)VGY8-wq#S#1PGBp*n}8L4NMJf^wf_Gv;zUZYgb7%pvCZZ>hlL5GxqDdPQs` zt*{|w1mEScKv;<8ohMC;!ZA8v-}W@PK0vz6#t#=ljHtW` zoSmb2ZY7?pV(Ut@bw6~<6 zFJUTIX_ z{%n!pX{?EI5C{wW*cZPv#0+}Qy4HI3ls_+8`T)bU@81hqPA2su)zwEmxi0^w^B#U~ z(a?8i@9)m+hbh%HQTgk*tIM*FuF4izyOAzDz!T|fzZ!*hB!jtq`g)P{XcwmlP^vfi z&IRAm^tvW;0;yrCEo@D6!vvw)uo+N-P#W3ls|Wt&fUe%!G1lmGRE z{qls>$j#F(AFzXkh1Tj8BSrtgBz_q3^9dmr5$01$o|w421y=&}k5fX9ws44(2|9`J zbAbWY>0x%4_^7J2) zy7=o4L;g64+v!rlHW?%SJ)@vXgy6{%HDO-EXGZ$MpC;OTl%&~_4_{j^f7F@hO&&OZ z5+h5514)ql7*VaIGXK5ii$QjpIkZgHpDoyr@mi4U@P^HmT%Ol)+y_|A^uI-W<_XiR zUbGy}l`*o|5~(r7F+(ddtVKSB&CvmEEFK;K(WE@i8DJ$|8^H^Yp<#&Fl-n9t|*CR~+BnTlk!EW)L5E41MUvm{zeWZkh9 zA;0$BJhJ=#7SAVo8Og9xHR+vG4NodHw^S`Sz!_8U;>bkB_~=JQy2^2I<%v;4{`8w; zUOr}R$*DRgHVw*2GY6}6;z_f-a=388G>=8`H(Vs849PHW(Hu-!zRhVUUN1g}_rmh( zJZ}Yo+4|Vfb8$Ufo-i)4TD_2g8KxzhYIqQ$bm~)ux42XeaW&GUv;^rmfIJP-VG$C$ zmEX~X-bA~~q%eo#;IY+D+-b#pShAEwbjZkfpA0a7cvtX^KZo56*n%$$VTQT_K=0RfYbtvZ8Z@wi~vPv;gX$Gz4#aogvGgBNaTUDR7(u1&spVw06GRNvQ5_P9WrBvfS zRzdYnDraQce`YWxQiMmE-`9NxN-NAsMdFouLQAOVc43gqQSON=ipK&Qh=!Gq`P?(S zJMxjtdW=f>GgR8Wr!NWbBq?=1e)2%oi9XtxIi9ZE&qmZO(m;HW4H+uTg&M z1e0o%b>sGw>tWRy29ywE@qaS{_q~CibL3P0H{jQQ7_9#l@FhF*zj7|MVy3?FyU>GI zu8_EGg2LH+DiHWe4smSfQG<^^qLXQ6!R5r@+CO5OFY>X$Uk5&UIGtwei!k9W@@)AG{6JpPQN zC72}e^c|L+a5teP68m~>Sa9FP@OW1PR$X(w8s3%%MLJl7gN6x9YB9Tj_tJQk}bH4BCnni7)E z_tSe5EO&LBvTnGnIA9M6eWsi%p1`%1xv9nLdj{A3?RX)v>V@8Ex;?EV8-{#VrLKVSJ@psxQ8 zNBvK;L+33(fvn89q<^Kzc&;d3@w%M{&S!I1DckNj)sAik%^J@ z|A?lgW2a$bVPyCRP0PeU_l>w^`|g3qM#n@$&(6+3&xS|O@XhaKVg3f+{*QoKw*To@ z`)B*VGEP~Um{=fbr3|f%9Zcx)n7?_rbpJYV{{hq@DLv77O;Iz}br7W)uo^nok@NY0 z03`5;B(H=)$T1+W$sxhXg25qkk@!=Ik&r={6L^Wci#w5h7Y|j zƑUlWhsKC-H~&g~92Q{0cfAJItB74(7EBdmiU^dynNVW)TQ%Yek!jewW)*qeqR zeDZ$i<`8JyW!K@*J+45!z$J$j$R^%IM)RJ#&bjWJoCt7$LENRpVALJ0;HA-UM_B{g zbseME7lCC)CC}P%<$9dzoE>Pa7+=`zMsnP5FC+V~#VaGi0g*}BWt z`bhn{kv<9Qh&;d)`~W;C*Z%mZF1!&l!ZwPao~d$-o+8|?U?$VJeuQ2j{O)G}I@_1c zp4*$GxleV?;=rIADdBoZ^+@s9Lpk%J4VrGNZB;u@ceCu{8!-Q66|`;$y=;5UqK|LT z)3fd={o>N>F(T8WYw=>H5xV&+U`Mq?hm+ctcb<7P4djsq@+jo zoxr!NbDF~;+}GEbz}BszrU{L=9W_I?3x(ydIQTZxt7=455SkVaKU~-J~i~+}7uN-1YqYh@Rl`wy|WofxDQr)A)UuSnfuwhgX^uk=w;{CJX7}=j0 zwDQ{P*mB;R&xL|QwtA4PZ_r~7`rU%XhIMKi!EC#@dpT(T&YgAwhK0(z-|pnVVfKb? z!l%eAf_q{+W7TkayQI}{qce7i6tlVcj+d+m{o^b#4hAa$^UdV9c|?OGxiu>$7v6#A zYZsaRo}(aHg(FUyes@-U>{LO!jIc}5qARpYjr^_EN>;pGB=MIp`NQva{;_Ui4jrnq zNz^_Si>TUpUVgyP4M2pG@fF5w7rbqQ6${&2y7G zZ8v`*=EjLTNtDaZTwUBeMy$*F`+m&9Ofpt;M!)$b>)Kc;lRfyV*N)z2ueB1oa)UN` zwyEQe-LR%#|3aA(k?RvjA!Mh< z>gDQQTN0Sm@f$1i)zxP$#iDO0we_x~regf5^rF&w-?dQE!vQ~6@l`p_XoE!gC4o9L zYNxMtBhuAXoA71%l69+69)2^{m8YTg5_wT&^MLKvG`;PCZ zE;Y4CZKzZ7rdL~aXM*j{RM6P5E0+?w(z?o_0amZ#b>8%244O*cP3$)5UfTYJ_yxjN zYf2ldC>yr_dAn+#h^44TCB#4MIp!=&4Qh$rLo5USZc6{tEoBBL49sk-FQ-W3#564&ZTv;uCpH3 zZIw$~rL{_u3h|}cS!`=<=ZWm)3#Hdr8H0!@Vdt!=YFN?d$WRHlRgyZ75-DM)%|fWB zK2zj{wPl5?2F6>e;kji9f0plitaR*MuI7TrP##Z2zoTw^)y^K7rbtnfoRdXEH-MV< z-RGb|^)h^Pkk>c=tpRkZ-@?k=Ka=#iz~wNCZ)($QnH`=)51@3~>GW$52JI zMm;m->_Bwa-?r|*I1`Ie3HM-llC7rMMV>|maoR*r@~-%!c3?gwSWHNt9OHFn`6pbM z6319wc#KFt4~Di+sJ|+8GTcKEGMRit>^a=DMJLx*wx7z?=~zprCTaV*T_W5>s8^oR zc23cTKlEP$b@gHKs3I4Mtxw!)5;Ze)FgmpJwq1z@cm-7~1o+fCs-bl$1%CC4jj1=f&ifHQ5)-BOjHIW|ORQ%dGEOie`C9 z_oudJ%TVMCZOglx_N6ga78LCP0PJ>S;z5!VVM~)!EyGQsAdB8eM8XMBMn9{#m>PLD zg^oHUPT7n3?DOX;#bptd~^I^WAm51bK z{(|qvjN_?mJ~OBow=;;YMupvI!5tI;jt1E5XbcpswsetV?OO6PC{T3I#?_NhK!=c& zYOPWrntcxi6nBOL&((h+_39U)_d?6!yGUARk%xcc;HUZGU0BV<0P}#I1+_UpRb^)v zEZ3iK9Znu<*2x0T&;x`|Drfu`$y+z6K|S{H)t5c=L9W~P;u&5RomwDzUgku~cW2kL z)t0k_hC`_x)Q{!VAy>uC{Ed)3Ih~wwG<}Ay{UhvyB)bGgvk2_EkskHC$8upf_NRt- zc`pyX3P&`yD%X4zntWMYhQ|x$NrmTwmjjZ9Q--QXre0g%PKIrldC-$aiB;Nq@m8wA zQBRRC*>Lj%kuKz!H@unV&R*4{s)l2{Cn`tLJaw1IgOL4JWG~lEUEZ!^|C8xU+Wn?u zUjH!r$qHcl6}m~?(=JfWvm(?3$Ez3CG|U zsG*58nFeS&&7}FXkQURnv>z>{!{|6Vjn1PB=n}e&K15g3^>h<`mR8a)SZ_9%4P!U4 zDQp&-$L6y~*=F_x+s7R|m-pgB_;5a&kK>d0EqpnDoIlOC^F90({u)2Xzu>>{-&568 zM`|dwX=+hwo78rx-BPbj9hrJ->K$nhrIn}s9xw;44-6076qppaBXD2fiNG^~=YkMq zK_zGkCI-`j>A~z^PB1svG}t=WA=o`w790`0AviHOB{)5JM{s#MO;^)>>1pYq^oHpz z(%Ynm)7z(aOz)CDG<{_Hy`dK}Y#AAuB$H-ZGo6|7nMs)`nHiZ4GYc|{GVjS;mib!d z!A!$fsnKbt>6i_WN5BnfFbr;mI}x)B;XYUe55a20>?XwQ)9@T( zb{`yu&)_#oXcA3B%x2Im+Jd%1%w9#Wp+o3MI)UCw=hH=q*%gS{wR9tWf>zL7^f>Fo zu4Th7V)lNvUdJrqPM*hm^I?1xAIm55$$Tka#h>KQ@SS`gf0e(%-{D{J-%?4cDb<;p zm70&3El%y2+B0=n>bTSyX(a98v_}FYUsWUc4lxHVpd6aq^BZgv(pRGix9Kzbj+5ek4V2Wv^&G}Z!t?Y zX`3vY+&X3fF^h~_^Gi)t%?T_oAbcxPO{dbAVje3ShA>FKicvOch!BKjH9hd^&-B-t zYP|k6JyzGxWhfo=nrNLakVlw_FonEB`jBp%3|sVESwVD$rqW!CuR4uJq zRE1viiRsZRUzL4zo!%S0?62a!!npG*FNPk$OxM#zv3Lz>y`>U-iP ziB&Lm59m|V0)$HXx{i%E>HG9Uie5YY65%U)3Xf+HV~T;fJfmzJJ+cDUj&))E*eEue z-Hlfen`{|d&DLVMnVm&F_tu}p83*-aRjn(32qMyCO$4GxQ7lG?QCOQS9;|(@*dSiS z<7)!ftK2HDs6PFeq&86VV!dQd%~K21=K5K4wIv>l^m?mHW2r6GR%#zC`>2EUbDaXR zt`1hm{PVj1`J;Bb>=XzJHO4sW$y_oIrjk|UTXHwKk1Ql>$P;7(yM`-!}vX(qaej#VcIr12Jn!G{&MxG(V$Z)chj39@|Nb)*4NZuxIlEdUJ z@*NpP-a}qJLY^g~$r*A#`H;L%J|JVr_vAFWmy9Jhka1)J8BcB`kC2S!6buLw+DXlHEXo0S5sJsHB=iKm!w)!9r?) zfE8>o34r`Yeg_{U!x*>${1AX3q(cZYAQQ6CYMg^?s7L-l51w#f|ARh|gMks_9&=U1oebxY4L2D?2Hc$*Dq=IaPFtml6U?OyYj?f7@ zLl@{u&XWtIiku{SpbRb9Fc?n0Cf~4Z)PP-hSD^_L@56FfL!~`0L!bKs?R7WU;DFsY)i@o7hSdMUwCeA5DTZ8l>ss8JHq6MHX7cLA;s` z;HP+m4P=c)2c@IZLFvSM3dZl|Kk{#RjUfCy{~Mn#to%-~on6OOK9A4lbNEke2>(Xd z*j#SncX5T!q{CoZbKh0+I-}!IsTHb^I%FnP;c0K=&|IFsFJK0^r&ozFQ|H1zu z>;eQ89YrV6MRXQjMK{q?^b*}g4_?SyvG3R^p2;(K7Ek9P_C5Q7on}9hkI`4m=Jj}e z-hemejd%`kDEf*1Xvgz-6W)~Pi$Zdg93!8RFUcq53(=lei9X^g(N|n828e-TkQgkk z71xPUULbnYCSsTvDy|npM47lzOc3M6IB|m*DMpFWVvHD$@;(-(p;vM%OotgP7jA=@ za63B-v)~Sx&3<9$SRUWYnzEnR8TNPfGrN~9VCUH%{CRd2Z_a*azp-D@lenG_VF4Co z=`b1Qz@2K9i@)qn6dy{?3 z{=p9O*ZFR=o%?mq)ehDt3~6&Awr+S&`CODPmu;&9MYudw~h z$__9adzIPQYs|r%%*DF19_(-Ib>?PqES`B-0!w5`%*){Q=E!Raf?@!cqLIu68psg@v8WncwMn835rd;Ar6W|;!W|EIILL3+lobThye~cwABu0qcjA=zUeOejVirG$(@InEqxeajQS!w%N^_-!(o!ib8NqU&R z#m(G8-=^=-BlKNv{rTA>fn4t+#Frk~JHxt%-cQF@GiMn9)t&@a*3ana*wolejy zdXj$4-8_zdL%*fp(NjF0d+7J{2YQs5o?=h4 zXSk0ivu9Za+sd}lU+Hf=g{QLZYzKRe?c{0P&z@(MY!};2f2V)Y^Yj7_@F1 zwR|0agg?q3Ca;MGyfrW4ZFn&+;bHy@d6B#%=JNIYF_9yhix#4xSSadajIy2Y5cNb; zkuRFCp?oW^5RF7*(Ng5{ZK6OlWr#9ZxmLMODOGM$W-7CkJCq(uFQq3@s5xp#>%MuX z+t%oZl7Ib)-pL3jf+!gIXvBn(kA+y#N4FCPaS|7ClQ2QIUPbznt1-^&Pp%;YFftoN29s;a zb)=MBPlk{(G8ES@t^WwL{G-suk3ow*7H#@?wCXpaMW2W^U0U=hXw#)dzZI?d47BGn z(W1{nYc6g1o#tav=7jJ6H*vyu7;~J)*y4x(0ZwpV z;e`JIPWa!U6EJ3!F`*v2{pUJ?`~@dyNFzdOu5iNtI47LH$O&KnDJNXG$O+%n;e=}B zgqk{>@Le5Fco#X1piYpVrN}qiYEdNeCxTt~0wLuWYmev}!GA>~T(J#0CuCjW1oHo@ zzX%*g{3}iX0VDrc{6!cAqyI7Qte;50=1ExF43m za##T?VHG?855hz6FqFe;SOaTe9XtY$!eg*rzyElQ6JSG>6X0>!h@7wqHpe&todVmbj5ATg>F zAil0n0FSN{Zjjl2$kaI@3&!g@;l#x}zs&i|{C>XV1So(OwK$4 z`ghQELPweXCsn#mI4Pe1kT;+Q^hBNTHTeemN}T{#!!_i67ytud5DbQE(VG1L6JQwm z3WmcN)CnZqv`JoW<3x(;t1_dfk>-C;`ooDGMHarOi@s2>pq%j== z6;KxFSkY@;-bf=-C5gnK?U1rHo-F7)eY-at z@Lt~1sNt@pS*?SJxLu9fHfk%&*1@D%(ej(SqSxQuXVLo|_!7(ReJ|n;W#exX^ThSMsNLld2Wiey9;=`#@T8q_Km zksS<5!Y!-}lVNyLG5e~4(enTqmb{IGb8`k&&@$PwFV^DeFI#5ES}xvR7DArfp({{N zg_d>kpTn6D*Ktg%3P|`D%_H^pO477p_ex^wGjJ=weS<2YW?m)fkh%+XFdK3`Hs?#G zzo7%}A5RJz;)Mo5tTk&G=!AIhgxDSwSP)pyb;N=|r@$CwYo4i>*f4Uzpj_Pg)dO$9 zW50pHitwQ1i`9{X2DQR*n#*yxKF)$cIKo(Cgt7VvI6yUaYSFMe5;?2Sz^evU%%3A<0;CA+EV)1FW=)j&e1QfkO;hupoMdT)1z+ zg5(8K+@t3xGj%V7jSH1T9{^>c5@z?&n`VcC$?`%d7z*OD26e!xS~u)|b&Np~!#Uw# zZm&CfJ=^%%(L8rt?#jGzZ;yL><(Y9xZkxO|Z?}1;&6zf$62@+8)ZG7qe7K7S%@1dx zYFoO$Pupa;8A*6XKbFB1G$vEXY=kcnI0@i!DZ&{9M*85f0)fKM;Y6*e|0>Pnw^{Q2 zs*>+FX-)i$Ht|=&{+$_K|2~9D$lazFugIdU4PAmgXo~#kAnefq%e}_=o^7q>`1hjs zunR4rA9ii?H^aLsw`I-2;`wc@T4Li)Y>@Xnx6R)kPo-xf`|jA5(HJ{z-IgJ5so0S{ z$FGI=z(!e4hdUUO?s&8}HBv!-c^#x<%DExxkm z%W#f#iQ*MKA*;x@*9h0^P8##CQKBhm6up#f@qu`>K}owXjIWuuIAK~zQi;3RRoJP+ zpB&1JW6q^d(xobN?=yWbB7Ho%V1BUeRQ=zn{%)#l>V7r0EY(|<%9f?ysY|0RNvRd9 zkku=mOdV9wR90$I2X(Jldv)NtfxBQcZ0y)!7y7YS3>vtLZGg=k`^xKV0~(@OH=JxU zb|Z~sA-k<3e%X!q**e(`Q5x-+0h?sMd|6;WZT^LOm9)$iB)B%+ek z`wUBr_R*q!`ZmHojT-5Fn$9MD^*()@&eoSTRPPcBVV71Ry$c*6A-zioj_AWMxYQ>* z+Q)uRtdISkf9g~Cr+sq%{byvm?QL!!w^G{MvQWp72xS!ur=edldsrZ_b=>WSA?|ejpbaUf(-a9oy=UNy&+34}xzp+ro;*x*H!ju$UR7UdZE z?naM(CVtB?$}QKwyUBL+onNV`lvj8$xu5(7+p$i`w3-_9{&!%XEvSMUqi-IM@y9FR z$1AI;A0v?<#_;9P6}Cn?V1&|=44}4}Nf<@ELEd0*VQVRR^oucaUxU843q9;k7zw^s z(*x(+3zOgoKUnhw&f~|Zq%(O)U&|}lr#DQ(sK<{pcEb@KfJ!!nFRB?{b7#$hns+en z@naNHO2*T zhyE}Fck>bT)3&rH%VwR}Hg=Hvc`hHxZ&M~ze;K*C#$Ho_QCvQ*pdT4YZZdKn_v;wl zjq?-4N+LACu?AxF`viPP2~DI8X&&uLd+G0G{?6j{w}*!EMe4Rl{m7cgZ#8O7=bDFV z9@VuUM@(rL0j6Q3S07`$<`}_sLHx+I4cY+U@UwDW8M?+9sEYS(+BAmx|g0|8XM21utjVO zJIFp^$MO3cyMP$*^K5>Nxaxw4JQ#UB@>9*rNVR&nw$Wq)nTrwp6G*Y$dcH;GR=y&q za1Ce3FTgMsbmAJ4a24_`t(N#Lfo^aO##m!8x|#|zVGgW-a$M6kjLeU~&s5MP+Kn!v zE9jH-Q+gcNl+8+z7M*Zi>)Cl+OR@f@!$>}X&*cmF27XXX5kDwSb*_3GW#b@5?XO%G ze{Uh{)mLzxYmiFY5JOwY!^i=9bc&59Tlj0_G_KDNlOY8<(dOhR1jvol2Mgf<^rGoh zrz70TG?WneIY=tuFfA3U$scqA^rjT~`v^_Lk>()>uRs6Ig>sg0;b>$oyYoVy>T;nW z|4qke_cOMP`^i%n3Fl$_9{WtZ9HcV~Whm{hM}hYsM82sgE}*7BEuf|lG@}UP5#B)` z0}ZGN`uG{E)bXbAfVK_8toz9$N~BMf+xkh}~8yf5!#LYklp&;_jw8cHP!+!0)C8Pt6(-GI&3Hx=% zYh7@yT?Z1<4cFQo*WRNc!WjA6QCHv!T!AZa1+KsqxB^$;3S5CJa0RZw6}SRd;0j!U zD{uv_z!kUx|3?6oel@XswD5(^@+HPnF zK$3ELInNgLtS;*1JSPiMQsk^Is+NZ)dG)~+RD-S{n^2uaKV22+3(qLdRp(RmcUaRP@c`#|xBw+a#NY83k&^*6sf+t>~uExLRc5L0hf9qEL zuYukCn)6Ra68pDq-S6t6qJH{%T-?w0D!ugm7lf?>Iw*|tt5k87S_z-+RLE5fu3}i- zMon7oo5xO9IjcUG2b{V{ZWlrJ!@t9{I||QoKh3Q^LUZ+ZdOH)s=iz>niJ$x$ zoh}3maHo>EO1SGekAtV$1=hlu zs?}oAG?n_i-h>2?*iqs)KZ(!~-NjY95ZBoxF@7W?maj7gx5|S6c z(IXPRI(nEd8t2^ZJTSj;k}SaGE-aMbsPk!(7uPI5F4!y>m(L{RdqOPe39?X}gciZL zVCMrjM&5${Qui!wz6lq+5ri=sZa3YXl$2aAnDT4mV7iv%HMxVz zuht-;zGRlgneO$w};7-$<8@4P2Pfn<~qCgJY9X-i)61p zecCB2xK;coaZ`dV(W8NpCV5Hir-%DOew^^6flGCykwn0n&4RVuV3Rax6v{~>`Kef| zpN+M6?oK044G#&_ng(_=9`XW>(@4V%JTy(Kmj(^8@IWrUgcPgHfl|;pJ*0DeV!Gmq zPsnfDyjejO3qk&0wLz=`FP~O!e=V)n!RoFlHNV8 zD{eQqyW;%#si!~s=;^0EzP)_>g!1wU}ZdAOjE~5Ty^||V4uEN};)1*YlwYxCaUFa^HZ*QD4 zU)+9RzWu;~1M{8s14TteQS3Kw);znh$;E{&t2r-7BlVnFh(vqCG}74AFpV}7Ez(G~ zt9}~fSn<%voQn;)CXGlIzl7mA_{p?GC_ST@GqW)?D=3DjVo{-(wt;+Eiwi+9wq%8L z{3a%tz`!$`H2md>A4blDyYJP*o*J`Y_oQtw%wbY?r*!)9$Lhb0oOboV9%V_rRG1PO z_u?}%+ueQVn67`{3u@Kq?LA8?%2q5Gd2oKDN7L|=@lA5u{kkx6jD6Ypi4JW`CtlU4 z^agXa8UDWE#XBF-e*g(!FFVd$x*kgkTi_g}1laXd@Z9%5tS&4>tm&F6NVnX51N3Ld z=>!5~Z)6eAQzj4>35OdfX8HRWiqm8EI6GQ8ItI}}ikZ0LOqrU~l}X&%4BOS*>ux#I zAc;WO#rM4lJj3783dI}mHi7b+@wS4;H zkMJoZx>v{jBXZA2A4Tr@dwlf&H>|<+xb*dO2sbvFO%{{G_efgT=IB|R8DG|+@XSkac(2VRg;DMPZPDJKLzCQwu zo_;#=-VtSjJMvv5>&G7>S&{EtN zR;Srz0+nbEyQ0){3*oR>DXr&b%;K=4@!&42MG;)H5!Xl?ABjwb*If!F%xS zu6i25+D^f6@iPr;YUtIN8b$MRw&(($CKgK2Io&Qf9o4O@T-3KwJtR*G3n5NqH-Z*| z<+IF08>pFaRP{tomO_5ya495v_WB@redJ9jxIXfQk1mI6UwmxUgX>;^>mnOoSoh$n z$6ky)j{DYa7MHp-D;#egNz_zOMshP5X=bS=K2otdKa`8DA|-l(xb_5HAs4E6SH6qS zi{!43j`$u?)yYd}fC|5F|q^U4;dDXkxg1vgUO+?H(he6Nf&1 zRpY?74MiP?+&}Ba{R<=youJw5d-@8> zfiG-@RMhc;c=-!Y$)T&9Uz6OP)zTmZJtvq(v&+lny7t!`VJ1k#{d&VzWg@Ydu4i6r zY&oZO;SsgHic#&$hPG=rv@Ex*eTN}KIsuvZ9|cHt>sk@sV@$r2 zN;&fh>gAQtd+UR`p~VI1q(&ps(^>5*yl6iXQFnj*JUhBISH896c}eKpB#YEC@T!Snl9cEGTNvy%mnE)@q;OvuNz!~_VyDh9 zxXsS8s3n9P)YVd5&p|UtN+Cd8P7>6Q3>u@>rmlEY$7!gu-L3*U3QUj?ITiUe@>}FP zNX(|*{9D4Cp3&%x>hE(Vc6{*>r8v?)GCy*6q&>U+cncLn=C=4s2Xm6j&R^!`)&vTBL)u7B^FibQinoU zf*3S-g-j~VblE{|*1Sc&3mvOwv}M1w3ywaz8>U|9U(TOte|?k9r7QkesPj5miR%s{ zABD)C@YVHA^;4cmVp$?f$x^elPH|m>1L6h+$EJ);9h)}RKR!5B+@jo~PEVehGBb5X z+6@2n;DbpIrEE%jE%`fBvR4bFxJ^Q&$0gVUCc)BEd=`5p?A@LqLg|&TCu|{Lwe4@5(cZ+9tLTY z=gdJum)6%!aS0k>-SEoYOUs-FD4h<9r$HXR(7*YE{l^AA?)b=v!o_26nN{6)`k~=> zR7@_rcA`R0ZE5$w-T~u-XJ!|VnBWUFf4rzs?~T{raO~+x*I)P4Er_T4QQo>Ep0dc1 z@Xg&^1N_&dk9EDl58K`deZ)Vty&w9~_J_@C(Na;1*b77}wZPt8?dTaCKaPy>+(vFw zr+TJ(7m&H0x!#A#!|F=U7P7&!!CR@lN?wIGJV(V*&naVz{-5Iclop9hR!S9+WCtmV|RD0H)ZXx zPa{!`5*3oATi{oG35x4}bvg^gNmwqa;AIa8^vy@nCzo#8}c z<=jT7syfV8PRTG~c3@~|$C$6k`F>G#Vv!TQK)I6ww>#fQQ!#lGnLG*O z&Ss#Cfs46-?ysIem+t|8}&F+a4`(X9EUggBB6@(z`5t?X-J|X-Jz{6LUNsU85Amw z(>IVGAV?1$O3!X4RhV=#GlF~B$m;$9=nZ3+Mb52Sb*lfng9bf$eWWH*$+GEp+tIc8 z_VlVB=XX8Sxb-!UUvvK(k`AK`+mS>v$*bXRtrMFkw@&MB@0Hj+t$X?iG6JTOsj!l) zgpJy(%B$Kt))Q8DvLnZmlbqu(OzxL(o&N^MsLXl%P0LYwlz--^%5;!`H_oForRk=? zvjG|im_3qHox&?GhRn9O)W@7^K+RK`>dPh}XDATL3$+XN4UGuR4r!szNvILRW~VJ+ z%d;_?7d`*Ptubp+J?+iXBt#L`q6P!8Q;PNx2`?Qx)BrJQC!wg)Z<^@IlKLK+*7C(6 zqss@gr++dza`KzVnaDrj8pv3)V&n36s$fF*2L~*_XYq4e>Br;dPkV3CXOUg-Hy8qQ zVSnSjy(4!b(J=P);3h;s@?OlGMeGp!a* zu)!oavcA(wtv+93-#Y-{D`Iw}=QEL*&J=Zb~8-!Tt)=NjZ4pNy$N3qRSCNDJ4M3?>6|2Ir1Yo=88O_ZqOE7O;ioRwY3IwW^W>6F?jvr|@w>`pnIa{DFsOX-)@ zFS~zE|J;$uBU8ru$A(5`jm#cbZ$kYMIU{n%H<{3MT>cdEq~u8{ld>jd&!{)E{>&yb zo6gLiYhKW#+_A~Fi|m43beHcF=f_Py&Oe!QGWBGWlTFX%oXagvCiNV=Q7|a^Q5B6O zKL!$Pq#kXMmh7N;8EKDMrghHi8|Q15@fe@hxj|pAx1bV|w;JMkdfM5fsJkL7G5*#H|e`@*Bs~;wjaypW z{kFTibeh?->d@93mM$1Q@|}R5V?r8@GUWS>@PMqeW{K_7`lgK~V_^oF0jtSscv5>J zZjbU@;<5Bo8InrpGvY)#xE;xvnOdeLGr74|klf5a$}-jWn)a$`zvb2Bcj>35cO0ip zr!wQ>n1^_y!d@15L<+l8O8GXkh|0W8m*r4SsCB4ks5CS#bX#a)Xhmp4$P`j-zU(+u z1xr*=y;(bUUGip57v>_8U4^ARPhS-A)9#BRj%rZG-4e?rKPuy>2NR?x2WamvWjQuS zWgL>f0VmiZPezVKWjwMxCgZH-xcO7xUvvyQM+zg4MrKCZ#XH|y zrOP;t#sPe$o!+R3RbL($us;hD+Sd&h*vN}v z1C7Q+IH!xGl9pk=5f!;7CaJ=bJkb-4jz(yj$D4L(b{Qeb1(e`2{xnRAtF@5JLZ%kR zV(NxY{F94ox)lEj(kq;Zb`0Yf#tHCf4FhU;9evafo)=~&qZ&~OC$|w*k%`%&q9PZ{ ziudfvB4=be9r7oUHHC6<>*)=O8n?mjN9 z{?d*_bRU5G5M27*C%1vVkE(yY4-+)Qe{xRxO|_T$tB*eolVIu=`K!xD4y+kz3vuBl z_i4aMQbF@{4{KTCX`T*&lFXzecV@;p-ttTTkcL4EDF|{)i{`2RrGb}Fv?}4{aAJG7 zsb0|TYhRSP4#SFLc|pzFE|v?cK3A%1r|Pe#ohwZ$s@5F{PU|*Y%B7cW+|D@tgOuRtOi+)Gh1cmi@e?>(^W4GLUDmK67NY2d(xn7^8`<^wEJ$gEg8j)JOJj=?~kIf3)t-%!^^L41yp)!|*lFnSAi$c#H>UU@NcOt*8 zfnJaTUg+F!!LS}>Nhwo?kDmYXg_msgld^l)pK<1{h40P3yK2$8?itg@6y-H&ecS76 zT7(Y4I8b0byc(Gp*}Zhvg!a?h-nQqSO^_DpHlS-(_j~(ng4TC$iflW0D6;VpdID_S z)(+{}?_Q*95P9iowErrxLi6y03NZw>f6e?_{A->M%^${p==sKcGXA9JymsE?v}hI+y^FmV|4G{S zY3I{LUoAb=oEe`SmmVKy=7}k(T}W~l$n4^X1Ap90abLv!6vyM@P=OIGF<%)Q$hWP` zPEN^?ila?-R_=l7Qu#ou>X);qXQgqPbXsbCaK)n3LRVC}TCtFp<;mokj;!>ocvE6A z^RmQZNOJ~?A!G>@vjjD+82!EYxFq2#2ERSEm;|jE#b7b(vY|@}z^}ZYNs2=pU-|^Y_Gd|%56?VT@_YJh?gu57yTlTO{^-NOeR?; z*(Q6RTqn(z=W5_5=Kb;!L#Og-qE&s3y_Ky{FjANA{F zYM=!VKeJ$GNn@kGb3KhdKZF8D_uMwz70XWdH6LBKy{r zO1pxVfX~!pP!G8|oV(xgxj3T!Bu=O&>_53x#$lm-p_}Gt))U0jMe>3(Us;*xv1A|@ zBqrn@_(f)Ae=%&WzMV7{r)AlZ-tH`$$(aLj3=a;~nnPS7zV>Efglw81)6hyN8wux* zQmz~(Mt+@n*O|!1`G=qFK5)aH2NdVl$iu&U8+r7FEz5vaeQ{<^G^cn7*YCphTgZr= zeQZl@RFg8k(`+`Gs8HlD1)vfj7Af~WpxRQ?Bc=+|J`*>M6(z)HQ!E*bdaacZ+PbXv zQ`??rt4gDpmuT!6eRK<@nHYBl4dAW|yIJ$<6|`~nyIZy>&ehdFmscN`IaWh9O{8bm3rRpPUg}P7W>R9Tta1$cZYu>y3vdQJ#mrb01ancj4fAv(_vwFKR zU-MdgnnE54ciqW%YHyp4nZ8HPR1^;!b3=}n(xtZ#N_>#(1HTV^J`!Wtq>==kT}w#v z8gEiY3h|{$rcJ$!X>nCmxHwW$U8Qraly>|WhP|fm-;K#Mb34F zngw}y!@aLWjy#3j`+Sspr+@uhph-E5guH!l*V#+_yZ1t5n1Hk8Q689u`0lOqKuOro zS+>dC*(@lkd`)N>s;c;GRPw9W+UEAlpUgY;NAo2`h0}7AilTT%Oe2KwIO`xSe&dbm zR}^RU2)g?BGxU?{?C5+__XhO&x`ea2pu}vkDJsua1T{mpg=oqt&4he)ijbF3#JtvP zm&Ib#qKgSigo}?TaN=e%GZQ89Q_w%Mg-wZUL3PhRS-tvvwz1C}(f{77>-^Gi+V`$M zTsXEyoWL|ww#7_Yq0^+P8CFZ@I1HZiwvjlubk&^GK8(3RSh!+NkCQv>St64=)M8L< zteEJ8(zAzp{tI0U{fUL{f!z4%@}75ekF2NJlipmnE&autixz*w2CTlI{r2fp24a;06FvP<(j-r6i+mRkUf6w5 zDu_P}I<~~|R^oVWNDB8A%q$X^m-|E}dPWUIOWsmUx8G|o=l6>Tt!r%M_AUHz@sw?g zeGlIu_S*K^53obFL-v%pwuN@;wS$BEc_TiH-^cHz) z%)x(T3!ykKP!{?94EyY!d#VS~P50b$xjwLy`QiGDtPk*qt?M~o$)8}4^DVaBHeszV z^QyeQ+hVbDtNg_jMv*6HhBMHs)buMBRbrriXS2!YWcDREYW>dW0y-|L5V_H!1G*Z) zYS96g!FMQSvZ_iKkUMjt&Rr0<1Bv?pl!}p>}t+M<9e~4eyUo5}I{hDZ>YhUPG7(X{* znSG6OP5iQi{q{r7L-G3)-lfO*yY^$wWAX1MxLauMLWNliiOG>~@_9Y^%$qcB1mTEjsV9jnuBc4F<;MCkX5YOFF<*pxe%^_+%V#;jlb~&j` zIy&8SRP>pW{wHemXH3iX?_Y+H8+(!zd$P0|xHrO=Z#2DD@h2kyw1h6q+Qn|C=-g}W{D8JkquR=(1jwLm5}jw!HNj{w>Yv8klN1qe z&T?idnGTEGgK!sv%Vf_XajK{IlC?tp0MZKR`76q$Z=$EIf+3i35bpf((jlyUkSbyde45;c{Ki5@_VW8 z1&@V(ASdl7ohRdeO86o1horOiKa$U;W(&JDS)?jKYe+B@cKIhq%$nKck!{f4mL z;x8oLLN_BOizz)n-~o>>DBfq2?%bCvR)?SeU?$kXg^ZZkb$jG9 zgAG%s3~pH7XXBFieX#5s=nHQ~Iz;B}h&=zhSQg{M7+)e?t}tz>})wldPx#l0;@jc7df0j-4Ne;Am|>gZpBR{LeO!Az z>r|FSb7)RAT@7U=X9l|SL1M6NRPw063(jq!qfR?U`-!Fylh2A%)BGs`GqWY6`287f zv)SrTv6^`Ti$4*L8|zLebTK#)CS%>KFvFxbsowaz)cl~Yo;Tl)ph4S-CW$SwVB-IgCtw`&mNt zIpUf#2l}q+Id<8o(UCL%0J^b!%VTra%^1>aXy2O1ACWJv+Z23pa>=xgS1s<}y7i+! zJ^NGM)9vml8#1d|ZtJ|&KkSG^{5lR}&k|Kqv{g{_P@a=5X(H9VJEbn>s*0b!X3zkoQnK}Mr zkbj3WmY34VYvGLT>U-)p8mmXLX#!}S~S@DIg=JB1>9$HstS68R_PM$&9VCNuLf6ood7zlR@*iq$E^#ju}>!-GF*eT_t_KoQe?V8<}=#i}iTTO2RfH$GFp-j(4P1@PhE}bs5-Zm>Lk|6VU=(q}*@U~0O z)u>@*^2M5-1lNf*dD5hOmLHdx-y+U9>dvgn-q}$()-K}LSMH7z)~RdvzVy0B#Gz5S z(0k1qw)H~&RS&Qae?P-h)~-EYCFQ#SY5I`P$6LbJa6VW6ldPY@r|NfbO4ZOPF^6y{ zv*~QML@ZI3(xq&jSf^~HPbhZubTix-b^#UA=T$jVFAlFlzDLPJcq4VxeU zBtr|!EK1v*0gBvuN0eKAXk~m!D8xQ5S`vrv@bdnBKu}$(4(I9l@#pTg=0mDLLVf{VY zw&9d`63;S71}h*1tOM!5MvxKgHn@$>r8c@7Uc>HZ!e+LR{uG82K_Z+Vw(*@#@?EGV zB^UIUVo{N?@RN}zKNcEvipD*oYL4^k`2-SA0^~&ay4O4>)w3!i_wpxv|6h0K9v@Y4 z_y3u5Kf9aVv%A@3cXQd33n2+4*+?J+QaE6M5C!4|@fHOkLKMl$O=u}DG{C}xHAPfv z1ZjmD5G^8Ngj(}Zq(Zf}m8#YD5nG^CMPt1*LbCaNXXc#UjoRPu_5A*L{&*G$GiTEbkoRu>_nYa19t-a5Wyi?PUlpf7Un>%yg z+3z>b`opEoKm6yHM{(sdc8>4+@Z{>7@hbdYF#hG)pY7RCSl&0#BisaSMb=JVG_Kev z?#p~L^8@jM{FQhS#33TGf9(aB0gN{>nfud%89_oRLj^2B!C#3pWnU=@h*_0iqm&IDB?o6U zh}oQ{0ea;(DPeP8IVTn54=oxBv*`dPYu#)k0iJT*$MZIyez5oXs)d?<_;B;gJ%3-_ zIP3TPZ4^-abplA5-d7#Oy=72bTetU%yF(zs9fEt~?k*t^+zHyaySoS1kRZV=5UgSMe-r3t*UR{Mj~_sbh~N2S zbP-eIP6n>LZ@%7CJu^%>KH`Ghn<#E5wBDV6cRQzUoI5q_@%1uBJfN5B^sNjV zVn7#d8$rpE>PXD!Ap?0f^<~!E4+o*;V`a9cKa)lB&Xx4fUQq~N7l3@WZ$g_pye@=B zD5eKJP2AN;0jgd%5xaoX0RHDXUxK>Ges$wto+6AyBG$^nG=XnjbHaKLfiq$|ew8s) zG$aq&twiT<7FOMw2L*!Uns(<_`U$*12J76H+z9&5< zE+n_G)b@$h9MsB{=Ezvi{Ik z8jR`z^AXbZCz`ccjk8i?vMwO@A%h*VD!NVx*UIOjm)RF5hlu|M!v{W=pzY($y_Cya+}a@>RsN<4K$P-*K90#nDFlf>`^40xpC_d>C`| zvtcYGZ#Q4RNnFa16WO)R9rTG(e_XAhc)oj{-@|@-nC`sGB^yItJHU3bJfiVDb!^$9 z%DtaYbawIuD}zrqzpgastTWW}J-Y#pXw-t%a8rjoKl70Ye2d_1+nqdVHBaOFkW(&F zN~-lvVF1foa4ozo&Vb?$U$r5!%ig%F2%}~tXZ$Mwsl^3!^J5m7H=rp*G_k}E)o~!? zk1ds<$b=z>3xaQAU*k2Y@_-vy7kJvfcBgOu=+gP=~J|9)h?@ZiZ z&KQ|YjIg)8Kj@r>;!|-=+~x@645#Av##i3lk>Kelw#_VJcSVbJz=pEaO!8Qh0c0b`k8V<|J4-WYQZJs-^e z;@&la=klUWn_RXGl9PHg;&angHELB;ziwAIawKPeE7Kd6qL{Wd!rp(dT4by6mg!Zb z&gb^4W6gIgM4x{7FK3EC80pxTY8^G-C;V^`ZkJ%(LL+wa?w&)u%qfncA0Ez2DgI{4 zbRBpAVDoea0~rD@+uj5k^SkQH$UHYE04Vw<{MWpo-?=i5Yqgq*!X=^Rz*M|>t33B= zFnHGDbCv8}CdL;!p!^1R*Xcz<^0snm{8eFNV^qh=_(CWw8+RD*J zV{qjX!34%6ZJVvmar-vg#Za|ktrC!K1;)^p3vT?g+8JNeftf7)g(ZsLx3nZVh?_DM z$@DVfAEtUTqm&i0ND_$8ZPk(#?8L?-@2BPtG41Ttf+nkxTm_m_f#tPvhr3~LbR^)G z-Mka+e&v8&3p5cs@W$x&gV2CMa+gwyE-g)t%j`n97%%BBZ_d6vx!|i^{N33<&)fyK z48r0MhZ^LrF#hw4pAg0()%w*o$qiQ4Tn(otkE%)Z+{gkOi-87_^V|W;{C7eLX{N^Q zxnlBbs6Qpvxc>0v6njgip8O(ca%zHi%Z1eX(9V=8I;ZG{|oVV4R^62s`e^`O~v1 zb)^!voMF_}P+4WYH(xgHp*W4#3ab0~PLVKPaZHA0*{vDiJ|<7Io1yo~KXG-T(ZjYq zYN_=(qwTbt#0rZ{FclmN5H=LuNY1#^7kp6MvVP{$7tJmsp6{_UkrIO^sV|a}eV1OH zkq&*TXgzD`GTHAVI9GBe&P{j=`oYRWTPchYwSKg|G50qPD$2vW(0 zfBiRkzAWew8Nc2~XLw1ZOOYdnq5IjK1IdGl-jNTOX(XA@n`m{dG@_zeyE<(H0FnFb zU!CWfjh<@g>x&nwlJ_LCn)(!IH?Qj}qkDOp(w|F%K<$}B^PBW0_49frKYRboVYvpc zK3^J4!5jus_~DW;+!!C8Zdae4esMQM{eESvuJgvwahu9kHYy7@6FZQ6SBy#+zUlMr zP;17o@Im2AtNNUe6gCue$I&P@J-Gp7VqI@r6ok+A`)$aEMT{#Zan7n~w*BKO!IdE! zU?)?r=d%#+3t`KAyL=Dpv++tH^gk>HOzhg}t?9NEArMr!J-O>Ezr5frBoB zj$j>{+w~WSmUYfF&%TDtg#@Aoyl#KpuFFx$F*1NIbul3S%&E!{hxIEdH}}}SW^5_E zfSdH%aU~QJ+YvFa9UbIK*(MrEs#c-XaAj|Au{CR-tAcmS@u#_QoJ{@Bt-u$fyko*t$u-R0!&lltM}RkT_+GhcVXA6UV)FO0k@_2%F+1Q zqQ(thEl+jUCYwL3v$FwiegAM;iS&t33x5{2z)#rPT6W%O@sboR%B{_-Esz)55H8Ad zqPQhj6U%oNzJ%)(Cn|q%D!9u6XRR1ld7;rRXapTs*P{%naI7=Oo@HX+l}qiy4O~mEmYK+EsRQJ+@(2WqA?aEw(abYAa8NG^xZ72TfIb` z-6|FHc7i-3Q!Iig1b=tIdhf{3ZBDQtv^oha{X>M4wEB(K&~S{ z&9p4Ek}HIxI9}MeO%U@i4jFf+H{zy$=7#>gE`Y=zvEa?U!Su*GNI34+V!VY+kTQuH zy7b3tS!wjGWvf21ay`4fNhMGBw9XIr4zEQ8vu5g8pnGguTAtg79E{YrJyk1M*uL#$ znf~M9v(nZ73gbRY|I_1`P56Lczl4$YAvi^u6*tS`S2-PZ>S!Q0h+qhe7w8&rui|+C z5~;X5)C!oUgzQm!C&vBJH!{P3myn;v0jf2;#j|dpV#5pzPQ}6aNu{kFVl>8Fgr@kPWUsO_#A-~8 zHqhqK928blz;uX8XcT9nTvx+Iks^kxp>(A)THhNEKP>GX?tM4l`j|qL`kRzjr)U1t z#!rmTre*xa{0)SL!Sn~I7mkv1hjY8E(yTY>jfB>*6#_Grfkah}HExy4`UeJCKYoE& z8kN_=$R!>p(I3#g2d7uv+Oyuz6mCc6@=;OpuY6bc=*<1GH5+Fc9OCu{oN3=nPgztr zpMSqh@B^Jfkl;;7Btm4jxXVzjPD@=IBBJ6*0oR;Iw(^gnwItZ-Bv6EO^LWf@cID#k^b|h34xdA<~hCgMc1moQo7&>5r-m!4rPV zQ!QTCPPzA(tl!pSBI?QZXMNPIWKpawxID|bU7TnAD_mUCU)u!$zC9!s2)8vHm5$8C zmw#w@C?8H;d)yqWV^<^d)iTVg)7+N#4H7w>DDj% zkJG4|8ryOe7^n1&5`G9GjDKXz#3kB|$K;QCcqbNknuaWoday~p$HA>8=(%NXxo=!Y zFpuH84DCtv31KJ$l|QgPv`4oGD1#=pm0j?g=Sst$kPG{>%Ko}a}FdDT?|ef7V+WImtdiB_eZ;1XR0`3>Iri$G8?M4vG+x*6`~~+_O_rDKfzP9Ua z?pUMbG4uh)s^sY9AMa22*OZa!5RHPY_yMmMS+-b-|x=T#Yu>j6~O(il#)#9n~8`OFn?K zS!dcJ{#+Y;YYvONz27_BUxp4z{K}a1(EN_`a^BTnHO9Il=Cq##YmhsOl>tX1@+7UON2w|*>3KEe%hwP!w( zqjU>4VgXahT{?~*#KGc%d%}AUieSZuj7#Hczy9(Tde_u@fCF%0&3Pz9|92PUys5was%(aQ=)K^v54soy=*b+#ssqXfD zD4?KL=Xlc?#|TFag`v+g+*76i4U6^!0fBPDTR5b5i{}yczK3?028Ak7ew9hCeef8A zJf}h2-J7oXVCvS((2 z-$B9o-kQhBg_ljC77x-V-zxk6d%rLGM5 zwCrftPg@F$?}2>)bnY*0ldl_BGj(rVw7n}w@9T33T^~B?8TmEF%f4+=Nq%4>)}+?a zMt0%b-i?qP->}RcL3KQs!n0l&PC1mhzwi0nrA>m(3I=2t- zRo^Jm&tKr99>I?IllcLDu#p|-2{oMl)TC|yb|`5~Oc3Mi*1d)sGMcYxTL?q3-6{$N znZBt7osf1*NXduZU+@Rbjwkcv5Uqn|dS zyEmi_N4$x;?MSJ0$mPTrk>Ty)n&6{(m+q}92yec0ljZxk@2RQr6T;fJLn_`+`0nDc zEw(vGtIPrtW6?on92`3&rSAxwuGD0giW5n+3B|Qb>=#>It-GGR3Tk%a*pw^y@D4r7 zl-|Wve-L|@U^|gENmN+b4{N0r%klVeJb}y)q|#dhyVDpl|NhV_eSshfF;GJn0{8JtB2WePhR$^ zJ(lcJ<4%+D&2XIjc`lH*>5<6J@k~X|9=krjHT3Yq-xVXhmrOa#gNGU6%yMbr6+hW3 z%oMEPVht9!0>?EOH&RCQPg^V8lk@QsD-D*p0=s_!2hW7ZzZopw!R&!p+g0)_duvL*I3s$j*?2R8FP zE)H-lMu#=i{M6sAd&@8UaJ7Di@FSVncLO^`p|-O@19)toI5~8DlP7Ea7=n^05T0ur zq*g57YMNAAq{O6_ASKT*NbAGe!TqcJ2q%r!@HCtt7@objZ7J$dS2|>7X$;pG7Apu) zrWo_t=IWTSEs*wm)#WGMzNBapQ#X<;^Ag5v$18ej({KwKT-N3J)e0MyUnD2ckGGRQ zsmMo1HX2zTrMJ3T)Xo2GHDM!G?Yii0!N{8BS7=TI0VMjR_$g?`)9WX=O8QxrHxV1go3t7#UUw)(M=f%rrvUK)STKK8?2 zU!@X5_V1r1T?|=O{?@nS9t{XA8Qur;OL=wD3uZNLj!M$hS6=n@jR=a0F?b5)>dJL! zM6iEYBp1mkFBzSW6&6DRz6CoEsWo}$jjli6zECa-kKymgc=yG+0I_TV5)Ee+`w1UDCx?U~5g zi(4)O;~}#1C&qC%a)hzcL%dzC=qqm=Iip>sW%QI@jzr;D$aZ@C?v$*gGT&uNb{)R2 z0C(HUj0ibV&#D+md(cSrU!BQKO7_xmpIx$blpz zxbeSF4((YwJ(yo3MUOcq5(_mCq^Au2QMMz68wwrhi{JtAi6=^zQ>jlyVab1CEaayY zAg&6g6EM(pxg!`Zk||A*Vyj@T9Kq{{J>w~vSXI!g1=yD0jz!a(n|@|xvHF%UVD_bW zfZ%Ine-9EqIXA1UjB0Z=UM%f&=GDiB;X6n%zAdxn*-cO@g;rFdqEaEX^-=?|&?0MZeadVV3ogH!prg^ESzfA$0j5?7qMX3HVlXMiq8{c8}-a z=NCj`67#+X9D<xXNBzyxCRkct|IWWT)%CaVu55m(N)>-hU4~uGt zUsYB}KvX)6luN73D49|l+(`t~q;TT{R15en>yv|D>wh+E<@Eurxoi4=HudN)`+!v; zGrBcNt7j`f;PJfF5Y5ZDE6~8Zzjc+b{PnPxMJFy_3mAdwz^t6zjnO+bpxco$YA(I7 z(9vj4?4eXyrg!jM=1^;Jf0C|uz(N!^t^3JWMQUH9*v#k?Yj;fkhPUdcXtQCi`r(6v zif9uV?KkCIDDOMo>n^GjY0yQDJ0hPBrLg!!=b13^ExcnfquASr-KErE*vpl697jN> zB^q(jthLapt!3oyrbZ^kdz$_AB_qt{yof6EV|NSuP80acbVSxSjs->OUbO4zcc`DG zn;)X6YTx}D8GDDv`)MR5TNHnJsTD{|$BW1PJJv^)ON&KDAtHX?{&hZqz=43Nf>l0# z=}`KfG!l%r%u=^jY3pa^tG-V*By}WNcu2j?pMFZSBoiwsgcEbVAt8;|+tVa#7>}m| zrtySHTgh-XjgDd%(klvL#_D;Gk$k|?9&N2Z(E3`dxC)^9XwqF?4)Zok_0i??%)1%g zMnDlDTCujFyr|kxUq?sB0E0eN84+F ztl6s0JI}K(y6F(a6;l2Cte*!=8NXUrjypTjMDls5=7< z$-ZiiCmJAcxsz^eub9lXIIcK{_-b7o9UT_l&@_z)s%jD55P^F0hiO>-)Em2UN=w*qH2B~CSrPgy%gWrcb{ff-&F_4j&09&( z;RNBk2_-AzWg)BJWu~w*r`S)7Zk1g~tf7c1F*0}6*U$6D^a-GND(CK1M3G__;?Uz1 z+h+Tsy=Zn)3IDw+oYWxZK*sk&LepR(#af%^X*#RjT(+lxhy5Hhu9_WJmj|%aTOHC| z6&RicgCIPR2c)oOow7!(=lX9cyIunX2i;h99OT(BeOhz58XB5}I3Y zIYP*IB+0L9y;CTuB&G^`+`-T!cH=g!JXJ3u*_nMS3&u7>TAd=VeJWq%mDCz&}4Px`2!p+nAsc9?WPws328RU%z#&$gu9brxHl z$n_?iSAS^D+H8qNnxNz^4u_s=3Ot%5b7lIhD)*{|FrT4+IJtZRvB@hg%`~Y~`yz_G z^q;MWv2}zNnD$?%2;U=RB<}|eNnkN`OqGdG7)=r8oe(X)N8x+jl_rhY@PVmqV)tty zKTZRZTwZ$g_r29t!LtIY_;0Q>acmwbe9mS8rmY2xb2i~dWr+t01-og5xDD4~bEwUS z58(i-U`?K3LE!Iv01wctQA?fYK$JFInB-xIJ*-fH> z8-Rk@pIB_xPVs4-NaL+&bi_vUq>~@grwug?frMhT@r0gS3T+puNy}x+j|c#AZD*a& z$;=w@u1ZrM9WlRc1Vjxuk(H$%NpH36O?v%#&DEmy5$+bill+KcIPM>D*dodk zoiSx0N~W?QRhCbYP;&9REVxEW=S-)F`CfmL=UJ5}55kt+4HSLdS?nVZ|%I+_anY)BY1J0iIWhEp4s77uLNwL{sru&4Z1g+YcVOHpKm>4Ky zq4(&lZ^3z8VkCFi!85pBfpBL9u;3cxXHOW2b`lK|m;&Z`DHy`G4#K{pfPH2Ng6MZa zSk53A_eFW|^N3*AAPC30F+8{w24WWkAzcUI+_A$yYpg@0)**~%Yv|y%Ac*sNAle<| z)!i)&gr8}Pv@91L4q~=0hITgx3#kZ(*sp_7?rM;pbAll>T_F5B3b^ONAP7Ye#AY4x z_KZRt-wRV^yJ2$2PxRZCQZR&Z-54265)5JOf?%zK;O?MW^@Aa5!4TN>HB4|II(QBC zc`pbeHq~U(%8kg-s*3XYgCuUNQJEwF1?dcgcV~2hr#X#5)5K-c2Ar#|A@)*VnMXfpFkk`2TYV;=!d!FB8Wz%dSAuvjRA9DLmvu z7gVV+23Q9k@)!h33=U*C3nagzKzQ}Z5S%jz?~6DPMA3nf`LfSp0E(8FSyU3#@xa| ztb+pqXAqGy2s1QfFn18%joz!H>Vzr&uw92v0}$pV%uZn;V_ktlXMx0bAOvtAG8pP| z<3f%0uy6OuJlX8ByndL(y?{c!XZn}uk_(0;203V4j!F|)4tT#Cm3_~8BNY~PUPp#fJy1YZliclD^HOeo|7i+Uy>J& zR;xwf5=^E}SvtMAKBQl%ZIC2g+KQbnqis_7ots$+OW$L1zf}s_?zJSc zu{d1BjaPUY#JvK-R>@g&RJqjLfmmgQb=T9YEa|u0zCd8a+3(kTGP>d%-GRnT?I-dR zixuE|#3S$6wEyZ_K92u%?Y|It>ZaC~XzZGnu0VGeb4yoBVPST47ofS8(W#teT9Mo1B)Lshj0XTR9F&uD{uKa-5W0e{t5(*yXq= zx&95xV}F6^y;POupyYm$1I6IUa#Hef{(Z?s$@A}0Zc3gPq~A*g9!lQ7E=8d~XzVJM z7B;4mKrc%Dm#Tc6T$BI-E~pdPtym>fKm0PYuHoWzUFzdJH_vjI9nd!l65)>J@y`9sP0 z(ap`tRgj(C6>7bWIh%#48<5Q$=)i7eV{hroZVvURtEG$S-xsVr>`sf=o&0RhIm!i3${^jF`P9_I0C7%G_{|(i~4yI1VHjWmS zUdC=fV^8SR|L=Xz$-zg-$0q=dUug94bN>HihOH~m@qalVT>Q}afG#p_egGvOFW~=* zA#PU2HWpTFj!r)Ri|FU&;iBY*3g_kHrsV$rB;4A@?W4QdU*Z0;U|FH$PfK+cNol2%`~r@vCxe*t6uWBI@X-dJzCs^bV-)X!D{JI!iA_|4HZktIB^P;$^`9WOPzO-Sk%? z{cp7VM{<->kbY_PqM8fp1O@4LKsg0z6;mf@vi}>4`d6`*w<|P1C^%XHUow}brL~Q# zn~OIkqXg6|mT%G6)n2me3r&^r-_4+fTJBEJ+~@G(M<_BAja}y@{ptgFUlRDso&jLv z=jP(z<}{?_S(aUOqNXUM?t2)R2;w3)lqk#V(#zIl1FX{0=$K~h!r#Jr9^gs3gqvA`a z|M0zua&{c z$tD0Lx$;4){R`{L51mf|=shnqVoaUnEp4nnx={i+`PjHQ_<04OTv!D+Q+pe82}f&t zOG@Y(c^O&quN|G0kB^6qm-l7zDOsU_RW<+z02&vRTu_4S3r3cg?WKd4RQ1wE7P{4P zK~orXCjYr8Uhe<46k5X#y4kZs4RN!SwtQJc|0cLXS2Q%{{(T4KxeD;}{l9sxerf%) z$bMG&ofD28)4m>B9$I`a_3`vy9>R+SWj&Y>*7^#`im{g>{3~6Exb(ly= z;Ib6L!7Ld`}-cL8p)YAd7o3YpeV$Ypd~G*WJ<426P;l$s$sO zkNh6s$-0|j!y6ck{rG4|yXX3Tk><}I9WV^+aIeSX$h{!*g{;5c!r%pewZDyfBb=0f zXN%a?vH=r60zL4+D{!j(NbWpclt^}}Rhb!g<9FkPo;5)LWO zw$xAk)^U~-8_H|54XyD9tf3AdpP>IFlz8X!A&k~aRYa@EX{HVAAeq$Wc6PN zdJ#?+3_gZ(I=uO17FW~f6v#3+Ni>^wk2B0DIdJvPZ`{B*Ff|-d7|q{(`Re8?Lcszq zC;riXLL<%;>Lmh=LujVVYWTvo;RI8AB0-7yQ-iJ`^_tw$maEnJ(|cVu)U9bw((sd{ zrs3=H2ZLz_dr*x^U_^6#o-Oq z-CA^_*vw%glNhHWjO+se7!m;hNrPbJ2YibmkIa~hPl*!+5xNl*;QWzZoc;J~fOs}* z{tg!pJ}353>-rx47Un?dM$1iKt0PBz_eBTJiIc%k+}qk4lc%h>J^N*af#i#&tYe{6h4&Gk2CWzTU0M$5!?iQ5MK7*-G;wi%F0LbIa8sPEiu$S2O3L<86H|^; zzo(L=%BB&d(yVpt@e^ zew?LjoH=Ty&-Yj`<`y#?SKx9_-oNu~J*!5q#Y^C>CX9TdOVWb%+5Z0Rw2(dbbYy=c ziWQAvjEivIFpFJ50aqu-hb>`=ytAoxJ%5KGcZ717r2W+EgMh<58B8*)it{YX3(Cf~ z%!M8NxQpm&U6kk4+vw$!;(;Bdnsqh>ilu*|>@Oio!h3D;pnmHrNX(V~xg8bll& z@(*Y?D3_)QUVWE{ZzElN5UiMrIFi_8b0XZQiaGLvMJNvD^<3OyM}L2!6T2Rk+0PIv4V}7ZzDgs;2s(< zR`y%Bhwa4Z>em_&6x0+%2x)+ z>fVwnTU+XL8b{otC>s+HA&l%SW5cW>?UtH6^W5#mD|=NY1Yl zc+*n}3v+2r@0IQhD24!&YWtzBy!gF3A|$$^MuU86#z=J{7H2L^T}y{@d3@8V^A!pF zCG$pvonJfVXR|#XmhOm7ww^2|`dwHO??`=4)&+uVeQn&a1u14}-cOoOzTbz#PHM_G z9GdG9{Aeu1#z&7$LmU;q!Q01!QA+c~V3rC<=+rmoPsj|O2!KC$|DDaG(4i2uIP*Kk zIm@O$q+`+K8m#T(AU~CF*xc8=We`f4A=38xa92(g(EX4d<>}?YS;*pHkyI)f#q~xq z%tAS_*TsEPljm^SLmsc|^6_NxswOB9)9!pqyDjngT%7m@xGv3VYd!-2fae?@fx%ls zuQYz%KXQlpSfaoUX-uGg@gLHtMM+gh!hJo_dBD}^4{(CK=I=QAmhaVO@3M0Raj6RIW0{e<%m z5?by_T6;SJ=1{@NvSQz@TzU*-v&9v8W0;m_ zl88a`2V8dHcX1J~L_u75(;jy}zjf|iU!;Sc$-A0HqEuEd=H{qm&F+X0#`oqx(ho?# ztT`cxx_>yo0DVq%FuU$7`W)j5dhbM{%rMxciDP)kaN|&VE@Izgid~et??VKij;;qm zN>kQv;!YatMcBV!&lyUAmuAHBn~U7%h@a0vWK~Vp>XAP8L3SlIQ>U(TMvveM-h_fp zulI8iy4&TF1;2*CA9qK%AX$%zqCe+=%ZdFyu#(`k`|;60%TDe~eR0A*@5_`Oy~4$Sx?$4w!~F_RYy(OP!=^IWrc z0m>@2o0x`7w*=m`rH_3Ne&O4<2&y+hl)cM#f6eKbEI^NI>aGs__19AKO0b{ zOUeM0N-HQIUauu|c{e-=|Drl3|FvTwy0mul&J?T>r{5PU46MVJa%7lr-$}pUTNabf zcNS_%YNFu6WyO(*8jG5GuCua}uY5$0X<*aei+K_TMN?28-UT&(jW;DVqZt@`4Jc;C zN`{|uY+JYdPLVhhG!t~wF*qh7s`d#?IWe7l%BmK2^KtL=Rbp)i`^1u;S`G1NOmac7 z@}R)X;aX|07vmC%roBC=PwQZ*fH{?YLwKHOPbz+lOczKIK+mIS@z|2F{#GK=Z;jj~ zDv&+vM6K>#(GiBQGf~5r;-2Iq?MF!X^Ye2Cd;;)$i~?_-;X2ZK3=^^+s) zyoCVe6Q8G@8OdBF_H%#6Q_)=qKa$)nW*n*Zy;u3nfgNiH_g&4yNhi+B|&?_y=AOqGIMe zkK=GO{oa!~r#7p?_;EN|cb)Dr^CZDriC!>U3hk$&KfQN)T&xV{>O|{1ejvekslC#> zyOH$YJRg`d-HLWL8-kt8`0xlO8h zHS=aaG4mPm7X!!mcz{@F8+gxl^~t7X+!@9nG3=Bj!;}%;P;S$IFO06!sS*%p^P@Jj z7ZT@QTSCZDuxHoj{y;Fhf)p{SSZR*<9q|a4r7X}r$P(#3X|tjYr_BX30q!>M&_U*fTXvj z^99`fs;1@#meT8#RNMwA7eW zgZf0w`cJLRj#$Qdzx6wID)^}VYgC0X&a0Y5jQ|ZDZAtXh=?@QpC7qh3rJF+r;BHOA z`Au?$>spwc_*LJ<`!5}ipMBbh@|%-7V@w`DpFS-1_c)6N)QW-z^4+J|x_F)6F2-@e zf^7o2PuC&YA!LtOC- z1Aj#p?8W1$h0=DyC;$5JnsOEWz6a|ncB!bNU-0e+hG@=$!>%ieYax0_f%XMAm4}Cm z*eI9%)^jZMpQorbQ{Rj(a)XAk^cd(4^uZ?RJeAvD! zeR{h}i|Nz8K3`S}UvYUO=Q?+?)H zzKJHkM_Aq&2!_Y%d#ua(HazzyVi1M)me$=3itP<4_(|gD?nL(U<44xO-nPgx z1tIM~=K#Fy8Hzgkag zz$=d|keoeUht=4E)t%ym>{}N{DLW8@YpRcMr&kxK&2w+bHTGf}Le1w<;48JQG1K}jK(!ii}6#Wxo2+=!n!?J zM=IGg8@{K5WPi!SEIda2|*T|eeOijfUM4qCv# z>qq0J4nDur{rk7Ro@&n{9$R-#;6uc~3@b8(M@kxFW4TH>(z6W(q?u2QH`OT|1Uoj2 zfyQ+uqXpY0qSY#I{G8Vw0rheDd3YPY;cxyV5~*M4fHfj~cGMJZE}}urTh|@qW@5^n zp7xRV_Qol*WbYqVQ=7P!LN3{+FC#WdIF8>JOiYio<=<5Hcx(26gMUdgemV6uN+Xws zo?3F5&-2P37Gd@}6Z_F$cjNj=ILGf);wRXD-Th3_VSSM=A$X_zfu7Rn;(lo~k?d6F z+H>_P=j39^XaoCr&^>DVG^6$Ip|y=4ZtjYPQjuH`4h|~gR&+bN1As3mqC+Qi{d61pv}L3#%m^qk+bm%W8>Cf_s_Mb zZeMU7xgV16T=cyRTs7ktN9Y&5Ie^#`ayWTb?ajRezK{lb{5kajyxzE43E0$yzPwio*0}& zmRZ5?I4v)UVwb3#OEkADv)xXQJdk2<#N_p@iL3Wq%(TDvlVCYq5ErvL(o@s&crVe` zaNFdyAk=ZBiM>=4fYyD(Qc&am+?crmIk~KA-BS*_!BYoy|_gpq-oI|r;wuJADxd@x$ z=lj{dfCuYDCzn6*^kCvj6gy;!uj%~Hn6Kkq=U&&X*1{_3rBXP`vq%)<+AZ>aYV`Z$Q<4hM%K=BA2X>V3+<%yFet<9cf(H^JRu#=gx!RmAhC+p7B0&!>Z2i z1oR*b0DlI;rNF(RyyfZ5Y-Jlv{e9ci+Qqe^Z~KH_Qcgf;ASf>pKOyYS z@B1@vhv!=5!?d1T5RqTLl^+=Ei3@(bEzg_m2D0ptm*JvU3VQ4it3F@#1W%O z@VPDj)@iavX69G^0h`tO6#Lz2nkpseCb+~SLf$)L&KUQnsqhk!waFT@$@*H}g*lmT z4H5!{S$mD_hwmmBQo)}n^(OgJ>5rKWCJ{<-u$!jvS~cVeT2alz6nY4XBr_Q*2}B91 zqv-2$2w3_RPZh-!Gb|;rJ{4P5EE{wx-7prL%MB#N_%q*#S28%-F1tF9Yz9QP*AG#PvpKoo;V-vAB7)NZ$8~%LKC;)%?k7& zABuVG1yF6^(U5BksROBHLU)2`f<^-8LGkpd9nP~0c;vEr`bedE&k*R{aeWOw%Bg`k zt+<-2Ygc?z(gu9PuK;>EdI=x5`R=R0#g{(qS4gHz8fG#IGYUue=~K2dk`e|G!T0^q zOuGtm(`{u8t)yAo-1b_a%XqtD4twvRcDz*g+ZjhE!A~F86UuW!sTf;aU0fM#hMjde z!a5Q?(m2YPRhnhyt3{+*`(^*(`!aE7ymTjPW7XL?>*!!J+j+<4DfGT_L!a`3m5XY8 zu(>{rDi`=2w4r}X46RpFTT`*I1iQVoH`Kh#>qm0e2aRhCs@yf){_zy`^w z5_58Z-z7=77y!0~(r6#f$&qQm=$;S}VzB>X-8Fb`bWeFNZtu!90(ywnPOhJ-AVaRy z7xy8gz+P)|CfRoHJ#i5g!P#rE$rM?_dn}^gHobg-EV2GT?^a!)JN z5!Va%KLt%Zq$-YUFnk`H%+pjAVHJt># z-$1NFiQTnbL=`QDhE0lCTo~o38BZ$t)e#%*mfRh@_nMyjmsdZ0`72DBeB;h{UYtA6W4J_BXl+#@80AGsco3wAw=|IJ7Mj1$-jmzHsM)@ zzwJSL4Zmf=;`4}af~Ood>_p**ecKy)1|y7+*wuSRTp3o*f%wyu7>pv;ty-TcEX9Z$ zB~ROe1sH0uK>eVYAIell&!rKCMt&je9n3;AGK^NA^080c&5X8UDAoxvfT3bo$pX_m ztCFrJ$=yO^L7Xo|XPADIK4+M){?&63&6ki-Iw&8$KD1qyDLNVFz6 zle(KWtXNt@o6>^UUbf;P8q zcNd`>9p9IVaKYxCc+7x2^tcW0lH@0kr+`<7k1QaKm6|v-xQ&1D_LzE@HvCHp4?n#1 zFzye@eCCPNy%F^7!JY7n|A(}53eK#H-h9$Q$F|e4ZQHhO+qP}nw)Mt-V|VP1jc@Xw zskxqusaj{(x;=GH)!u8b^*le+`#*0|Z&YtqUQM5|pS7u^3`rToIB#5UUT+XyS-raY z#0{w$lMGWe#~DX#_tOWGPN}`>{$lzR4H+C0xW^jDa?Lusv<^ufQ(FKZz~a5TSHx$w zF4aw{$7I8F&vC%Sf2l7MKF1tK9LF6;9mgI=e)m21KKK4_fuH@KLB5LLiu|O67!HwY z#%*Uwf`3rbz@<`2rP9%)(pV>n&hI7e$L}fbk>4smvp%~%!+d4==?fAEQWd5uj+X8< z?$_=)?y=u$KkGj8KKZOa67IR)DBgNMS-7P6NeWUJCYYRc5tV7B3sFkJX^5pOQOZMU zkflpe>Hw5Y(z(WJw$i=Hg}*O}y({fk70xT38Xg;-93ENTwZ3$|c)olN9mZ(J_TmDt z8gLr08*m-4YGPN!b(gavqo|{#lZ`d-q??nAOR24*Rhlo2ygQZaWmO2&%4rg)SW>g5 zXwBN0C8`ups2@__rhXkenHAlKHcNUbdr5oY#1B)MG5OZ>mhjfl3Vl4`Z8l-1noT*I zg*7W{mGxHEhBN%d!iz|`#rCc0Ev;43S6`)Wpu9|dnu0Vdxvy@Pa-zveqnlDU%W2k2 zSRi?#Amzl^}V%Q(@)$_U9Xs4tw5zf zBO?bq#WI;FUY7`SlHmxvOVpn%V-o$I=Pg8+Y%MZHmx4nQy^8$&%mkDyRXHeHwU}A0 zQku=02wSy6+`1TaS(c795pvNGM;K>85l5U^N(4uq#F8|wc#$=H49ICMW?)C)jSpRl zJXst?nnSL{H>DB(5-d($nXx@05A+ge5G?V6NnOW&I8Qg#@Yss)$t)D+$X9 zD+L7BDc<^D zT78B3ssTz2S(}vu)fp>$%QP1^7lapTPdIsUr;->{ihcFFiNU3YZPIn=!ljB-vM?Ha zIF_O%%Asgg(AkwzKTaCi)mGU;OKc?G8jH$s@8wR(YAjZu*;*}<>{R;Fi70J?bds>E z!mWyQvaoA^S|#bEV^)n@HR#9~|r&qFN5U8|lijJ*3Ea}?Rvt?}R+7>sh-?W3&D-D1FE1Oo= za?w(3R^%a49N29L+7fjst6G=EZY$kZJ8cBo3UwK)IyAs{OTRYblu0y0<{Vn>s8xj? zmCJ@SYY!{1R%W|Gc1he?_{DJRk=CBJ{Y|K^W_D%VvYnE+)qg8vmL_dZo1S*T>`FPM za^(-_zur0ZBnt{-epswj@2q;Y|A`kA%>&6?sr=&k6$}ce&z!IMzoNf;z9an8zDpYx z4S)u#YZnb?myb{HkW@MY;sNEDp;p7LM&XPynItoq1ep0~`55U?41cZvlI^1#^))K3 zliFgm!}5d$Ksi9&Kz)RTJQm+3If}jtzbd~?vI<;{)n+j(HhyllqaGN)|{K zNR5}N06Bz7&UwyrU4`D&-u1fay&NG*3mF{J!Hx~NbU7&`rkXklnu{z7h$tozxWwdB zmX0BvMT(ko$xbI`9qX1+`o7tn`wOQaoQt;Q?HN20d8YD?=A1FRYPRM6(Z3}6O!Xb@ zIwx&Q{G-8~CUX|(b*bpr+PgSSFhp`4Iy+9-HlR<5m>4-mes?Efn-ULU`hm5>v4~?I z#zKonFpXxK%W|-@o?zu*Kg80T^f&c2wKVlKbu_K7?`fLj8ROaGS@lW?SVc36B{NJ) zIY;M`8zeQbC4gd1Kw2Zg6$>#>K#msTu!am3;>JRz5Dyx%Yr+ME6Nxkuhx&@lMQQ-{ zMg;&h2W)rU-$nkC0<{690kr|e0o8A>8qwQ_w%++Y(RKs1gf(=7o&jyDh+iz85i=C9j~-!viV_wyC^5XD``}j~6A*53 z-{dK{$>iLGFwga!V6Il3fGg-b#lLX`SP7U3*!@LMFUAd_b0avM$oeB&gU6+8n6r3= zFp?-H{S>7LR@qc)spd@CHe~P0!BvW>CQUiGWGBnfRH~~&s!FO-s#>UKk66ocm}TK* zv&zHE#mfoG^UL+icgz2%q*HAzk1vlduPo0l>*Z4_B;S;LJm$UAf4TW}d?S7pyc75} zd_($W^h>-ffj#=YgYroVlt!Exy(52R`U@)nRUTQtqW$FtiY(5w-s!#&%)Ss(ul1E3 zsw5j>*opNe1)&LP;giCPmpxLtW%v|NeoLAG1(!qQl`X1ZrT=XKd4T+>yNIPBz=%p{ z)Y2#*>2eekP}3?mk+u}7N*L#2OawRrB&o!7QRFc%a-R{i%cABCb6rlN2!$&ON~B^% z!-3WkZchwh4$yj+>khvEFU3{=9lLBd*qIP;AzGD?B2j#lkbHwY2QBc&Uzo}v&UqK( z>7?~B?UC(5nA%#D**a8(Iu&;+PEq8l5+Az~mF|f54S`Bc8XX5#F4Q1w`f8ll{yP#* zK5bX-dNbgPRyEaBYz3}+5zVk2917xm&0n8zSS$Xi%+N1rQKnbW1zTR?-voD$L(>xS?80G-{+0_bgrWdyqWth^MI$s zg$jX0#OA*Gw@)G--)-^OXEGL#D`0d`I{C8m-(UzL-{(YX)$tNR-uHTJR*U(3RVI(a zcknnSf797yMZW&$Op8qV@B;&_PQUSSHKxWu>&bCjL#+;j?fUBl`O68NPKVvzo%ZpZ zUc=4zxWLPmAK0$e9n(7uLtXQs0R7GY6Bc9|2H09 z0-k(W{-5z*hYw4AyI7x2k9J!xNrTwKwz(Ycsi_5;>U$xc>!cJfst1sOSDtJbNB zX;CQ&dEcmrct|J+!zFOC5|>piiA<>w90-d7J8f$(k)`8G+U`&=bPxr`WMo(u%G?2; z0^=}y*q(N0+pgo%&3M|#&84@STc4c^FrE(Sjdj%#DxURY6j)T3dViuEyZ!tGpyYh; zB6{rahw<7DJNoisG@^^b&3AN3Wz`Mpae8@*32E{3+qc+T=B&F{5|iMw!1)6?nBJ}Py^)FDLqt;{eyUznF`tl>DFN(5u7}3lynmb@d*u*lm3z0{iZ;0 z7BR-{*qCe))X|UAvA_`j_+u}Aw=ZNDW3keYOQ(t^8r#@pDK<)*<0F=Eu<3EQ(N+px5M)->*H~MORFfYe zd$l|r-nNKMHfqldR@2&)->C2P!NJ-TSe6>s+SLYNj~by7Pv2hu?I7Kx_8(DG)jD`9u$RL=lS#Tx-wTAu%3Y4c*?o!Fk+0yPi;Ic`VvM70ttXG; zT9hi(mqtaI(6N(F@Snx^;%fqVQ8x$)=OI1q%U~gm0D8M6>$n|5g=chsr(DxJt+0Dq z)&eM6BX_7?_Nu7#(<766<8ErgyVWFS1zN~xI}JTuIG{f;J0iSqJTB7Kt0V@Db%eyS z5<2p{_fQ^o9n@XsF`a+^2t|sk7(`1T{4v9t1LVvxsMx#3c)nA*6cMXbRa2$&g>A#^ zWwd;5+5kv||2be(KkAy6adv6VtjT(~oYvi8bO0Qz@|%PQA!86Wt2_pxVMgd)6qkF^ z-EWaoJIH5vE$Ukg8vmIsiFn#xPRfAS!?@HukMOj{4B|)|^)H(n3YEC!)+f z zSJ2dlo<9nI?}pOAJ+m(qNEkbly_aaeS);9rIc7(f)(Kmf?0hiS(h2I5$*@6qiaHXT zSCEbPdUYhjmY}Fh?VZN7Jq}{Ad&oHgCBho1mM16AyzB zBP`K*=0#IV*a}|>;fYUxc$#_GMKVPmM=o8(BhmbDMzrYaQoSDWfG|eT&~^yxnaM zQ5o*%`(SUuLu7bUo%%sk3Wh_(vD z3DQbj3F!(_KFaiJpFx`!yrzZe5L5C6X~6Qu?{LR!u<(T!nLJ{#U#HN?xwN?72eK?- zHg5U_-C7zDQPS2V`}ZDOQ%k3UN#KxOxDkVD|7>}Uqq`H@RHC+%xJGrQd$w&{bu`)> zr3JybL-A$&quBxwxWzoha7-A)CTEo4@xolj#KsWAae`DMH4sK6zmW)u;)O**VHfd> zfFnAM+0x#YF(q+rZdl#LH^Q9GpG1BMONeLu@_Q9Gj@^V-ct9}My>w2R*3~cA6Xd>#egqh`;T+v9ScQ#Y z8ccnXdPWHyOktcJ-t5gFl79MDI#1PBixBk8`V)E9=TKz!;!@=dRFTWZQg(hHBuBk6ukip5ln z%gL^XEc-H===itfo47K6P%4jPS6(Vbk~6N=MRn<&4!5a`^5Ep^s_c2#f_IK_8dV}k9E&dp4pAMH253F>SRx$A5wxjm zvV)Ht4bnU-NpR@Gv8f}Sw|DBbx5Jmt58^~Fi$4kuSx;gQf#%)CGFBCLAquaT$ZjJm zN5o)pa4OaDNCHu}j(D=3!M;RfN<$0T8}GIMno*}prduV+Z1#_MTFT^F{mM$CF28bD zMIr66<-MHCG+&MT?YU~@X>yKni6-u@Nf>2ifsa1o0Qa=pY4dAL*W5-yLcqqySKohz zsK(|Oze_}$+Kk>RuS5MftL`OtyORe!Z(%lX9)AN;ajI zbT>+aW|}6MHndWRY79;xj#@}{bu)FfDg;|8&}H-3^mK!m`M(5}bqyk@NM75btEqwc zFq|LJ2L2E}t!r_fcN=luw$E>o3L*BFH2tVFK;Uk4>Uwx*2)9thuHE)*ZXPwdf`W&N z1w{w&Ywln+l$LTxl_M{%#r5Vu`oKtvrFf&;kOB#X)5M>(go@=PBUMmz{YV|TE?67# z1mont(Xts2x82W}PBOep{p~7^CLcZ9D4rMy-X^lVZH4Br5qak*UD}B~DSxBVQqPKo zWL@gBNnlv|8yQ(1wG?4KqU{mpOa#gU?0m%mVCFtZF7Dw(=VI1k{63Q*3VEi&%JB5< zDtqi9D)e}Zm$-^3z(pBl9@YNm6V8V^>Rj{wzRDc8KlK`(+6S3;A*#(icIr6NGB#$E z(S@5AE+z`Mh$2?vl_|*4V)oNISvJiYu|b)bvn?E*w9kiKvnd^|bs1$OU;K&dp z_kI77|0N`s%LQQl>*J>Q+#%eh85DE=4vhlI+qUZC?$RmZ=+RPQY>Rc9+}SQDzl1p3 z@=EA zLWc3uNGQtHU+@D`)X(`_wykq;2RzM!uT0KK%=6zi)(tYENOu)S56V&|66@SU21gmQrKS*s@oGG|%DBmr47t|HF3xzG% z$*<#)hg(Koh!#if%Lpo$LJEIU*??$K?{*>bT0oQ?86Ke5&j{u!m=R_F4YYh1ZXcDF z4NNu>q^^i3Ht4Oi#RBCqLgm2K5+-k7A4_J6Z!{kwUo3y+7Uc?U_dzuXL5UGd;5VOu zOo5NQAY{5HPWm8Kv*Z>6eQg<@OFzh=QI#S!%KT2g&3dqBC*l^7FeazuvoZz0AEmFv zpNSD+9p9-Kf{s(W8Fa8RM!fW4dSGU2$(?e50_MCB)Cl*#@e2ANx<)X8v|B_@@mQeP zy!E_TWT@*|;MWEuVRE0a+e+(&`UDenS1L@Grv=<6>TcKwX;QDyxTH9hjI$J?H-8wC zCAir7W=7Ld$B_qjSROg_2S8W>7fP_7cc>HoW>AAP+BDg|6j83sl=BaQCxU(;M+R=WGpD2UlQ?Uc%|@G zgTh1suUOEWK$Q+_;g9h!p>R%O#^rPnWSNdjjSF#WpH=qg_6KbLiJf z*KljWue8))YB*Hwp1i>bJHa_U4?<`VwEGkJiywG0h(~RK7a)<`(?^#krI)fen}c9W zjU{~~r6eSk9g*!mYfRc70WFSxwQq=;c;?_PjJ3jFhZppmDRLv^HHo?6@<}&9M;F4H z6RS$1ZbFw0D;gEw1~zIs!V*Fmz!<=4gyO#5MM5r8lSd*fQnjQ0j!XfV9e5mgC(_iq|K_5TMTH5yZVWV!N$L!J zLpt^XfkWAYnzz8~$*VS~-8;dzi-cYaJmDh1ff4q!yPCb1>qIFV64t zUfkJwhQlG>ghM3C5rp3u!dr_~yEf#KM3LKoO#kQvEN5I^%oI%v_KJZE?cB?E5(+3HiHRccLf5)=5 zVm=GIyJ>W9s8Gg^e&+`h~UooZ0o-b!$1bq#C3-KAiMLUg9 zpLZRb3b@;&d^QwNTglU0N%7;jmh_TL&NYMFiS;PSq6AAyn=*-&@puG=H&&d>m+$FJ z-MRKEiDOC1hT;xIbe2?=yZm@Q{Qfi3G!U@D%p}H`>t?o72dc+K#g&9YY9GuLhV$6s zGg`rb^G7X;dy!Vp=J!YR1If8h{vg6#`w%aHl80Xc-{u?O1TU z{=V4N!a;sJV#|`lrH!?n5$?C@-udTru{Kc8W76{>)-jGPR2psl)|Csj6iZW+LRV9Y zkBy8jY<}s$gy3SOC8gz*O)Lq!^+Z7n%z$De&kx1^bvSfRDsO@FEk{ze5p<`uZ?tbD zR7A`gNWOX4lFe!dEOm+}*+{|~h}G5GbC8|{Kuj&nw{fKS0Vda};VQTFW*llM(!#sm zac$q-6)KXDC!Q9tL3+IgE3*aoBNP5FVwhI~v7A1+5GCz~Vp!N#EtV zs=6_pP1Ujo7G)@x25nQJt3Brrq5bll9G449`0n7U@#NN#9vbl`K5hPoUByEs%c@q& zRtaUF7XLKEqX3ZMT~a&G>0fv6EW^^Zj%wjf)C`goxwhf}v~lRyB9rNqC6zF_hDpO?y}TOHCzC)P0w{0HTbWk!-Bo(Z&S zjwC?(y?MG;aeq&}oF;%-@V)?Up`%Ryr=R_Y-}Nr%(ZqJ50e8_F!v{n1=it)pu>Dg( zRQ%>l=dm;uT~e$bk5p%3;wA^+BvV#MVeiXJz)~^vc9pa&bM=Ols8UPQy2mCFpb(^< zKtGgL7*#dmtinqCm0Z{2#Pw(YLoMzsoV_fJB2y-Y9D^@`SIkE{n}nKvd4{w|Tt29s zxIVckL5DyKbB(#g(4NrNQ0FiVWu8RS$n3DHb&=UyZ{G9?U9Dk}OpT17T0`Nd+qA>>$(p>HO@;7hIt zR%d5i+#g7K2a{S8(2Tq$mQIqm=CZVk#a7P;Y2=YEXo;_+wDSK6s%NSw0zj0jr6bduQJ)kys|Pze_bjne2!fLjC#)GV?ZeBCMN|Y z6_1Q5Ii5-m_=%!~@h@;~BA%Km60J|W)z@t^hThk?o|PVsE~%mK7peVspPGI>9`7$O z0p2^Yr!~@d!0opOf?v@x7tKo%&r9<6ct4Sdd!N!5u<3fG;>04IU%|NpX(VN(K+WXj z0x?BLv{>`$#g`&Kg+C~qbvcb!6N^?Ji9EWPud68$cS-%M&^4Z}@q_}M`~^ii^}ESW zr!ZSSiyaVdDoFLIBYQq2y8U#q1UQ`WI+~K;n#RRbrU3Y&>Z;Dd|>DrVGl$ zHAL2ztshFv)yo-)T*Ea5QzaaE&yhpZ&lo&Jf|^nb^QLA{>0}#nL7?c(qE<;L~E~>+zGpfghDddEEwoD6D=FyA@Dy2B z@zfZp)k7uH3tuc#By+`J&FHXoNe^Z5vE(pO@-)CT%$%WGb9K`wKHKloIGy zMMQ>20Jn{e3rJa$jV1rtl!M;zF`Mf({QJ^T-CHsN?5NQH*sQH&0qGycG`4A6a*`OL zrwMbtzQR0e0`!FWTef)y9CD!FAUr6-5nb8+oX#>i!ed_zwFkE$mi}F(=kZaOjXqDN zAjoCDg5`|k7Uyk;xhn^8(Wi(WChq?jZ)w()qP9DSNKM!@$1TVPL1FF6`#x;H(C6VQ ztXOa{yuqKW`J;M|IB?Z;J%jlacbiDrZAD!HxUOW+JJNY&@;&GsmXS_x#%v)ghE|%4 z#oSs=H#_e(8)m?+&))Q7k#&S&Sj({MU1v+n#IlQl%5qj#UFGCOawRzxZjv-?Y2|UO zagsDPnnu}K-KX7FiTXS`pIQl}?`TG*I5Ae$49p%~QnE!F;v+21(vd=;j z6_c@0QSj36lNt+NWU6N4q+l;l|Gf)52F(Yt@)|4;gas82OuC~Bj3PBL0r5zn!|0+M zoXA8c%9-_qk3=4(Iq$|tR)_!qCT4h$eFPo?l7H0sN@w=_pK$K^zv*4*4gZ4U3c%e(Cx4ctsY>0rLXh?h8ab`w!QX&YZrsw<)#Y2voS@QKv5jM~}2#eI93b^-p;{)2ei`%&DK&%oQe^y+0&8WT)xc zD`x_o>~JH%$2PM5II-DG3B8RFv8W$E48*fD;h-g63LN;cW9B7Z=H@KXC z02HX390YDx_Q`sTJ{Mq3{~~=5=d=3id&FQB*6~sK&#sxiTEr8u`lpb2`YIj2V(XW> z!@)C)ldr;kSpHFBV=~#>Q>+cf!~Yxb^c%yUoyeeKN>8QFx}N@b3&+bW+#7?=Ki>0(^dFm>eJEuoxq|~Ho zAZm?PT*fkxQYA7z3Edbf4tjwk8Q~I8-1ErXr;FKF!e@S$6?nh`?oVOF$& zWRW^?5;g_xgk`YvRrGQ9XAXN&5tfvk6xK3a328yPq%tu!qa$vQ*K&zdu~3XcT)5&{IrNUalj|$g4iG6xhK?`yu-H#d%CDEmgJv9RUqv0 zt2{eOi(-b0Acmk53$26(nWR>@AaRoks&oqAhF>-MbS z*cLbgMZ$u^AP-m%sx5}(md3h>tVaYs*z7jM-M1BsE^C9ftNZ4-CXMG(pwt}-(2MH> zq&rq*;P}o52Vh4BB}@4UQfP!J2uwv%5opG0+El90hFNG})1{6N)vVETURxZ5e=9xH z23&h@B*Hy@841+;`hS|8zTNfSR9*#jWu*gIUK8f4Iqk_vbI;kqPr#d~TFoKNNGOi% z;`?qdnlARn|6{Z`guX1FsQNN*(D_=QfG`z*wTY0VWwIr5-7HUAEcxeknNs4wKXsBR zjMt)4nG{r+Bv!uyNu7OFj;GQuWsL{qEN8g3JHrOjoY-@V;#9*UP{nN6e5rDye3~6? z{B0q|9G(};KhDN9Z{se>>~@bf?yg}ilq!cbnzb@fq0ly@hC4@ET3Z-*?KrRf)LUnB z<{A_h4g=XS7LQv_6BJBCdyKUG^(@%3W6XsoNgGr@*gSYIs47Q}st*AVXPk{XK|xnu z9mJecC|KVZ#xcZU-J~E#Ob_jEvDH?3d@q(TK6x%-AyFnA`*w z%_Sw~c$2f@zCSOMq=@@-_t`p~tWs}Uc+Gs12d>}17eV340w<P zkL^k-S1YI+w5_vWO((SADCfr3DCf*Y*@93n1}dHZWe=kL2Kwb+lI>y0{jfQHWa8I`e|7lxo34Y_ zI&q(kMUU*8VGs6Um;^(YyhgQT;Mc!r%pPr=>Kyx1xjMXT6aM=RY`{76QbLo^<@~S!@?9=dT z@bd0mpNa%!$l>pb^sX0hIKdlgi`{9I^H$%AN|=dSYfX6m_x88W&!Nd}n-(PK znkphOh=&*sLV%T5!f+bnpZ?b9m~c?zH0gcm$$k>ozI7oJqi7;jrU@yTCwP{I7`7u~&n8psskuu;_j#Oq*|P$I-{9O=VL|4s^CUZe^$NF z_^+as0yLqT`z5!>fyEWlxhx72L>Pux zU7NK5&2!s)w=ccjz!VCS5{*H@`-N~>-KK^P+K)lY{^?~&D@IE1f4M#saF#!sIUi`3 ze*trRmjT~jxs0fuhm^aVVX;$6lViuExET+NT*IO;L&P0|+?YNa=OxP$4tzpt;BQFD z^Ysdl**GwogHv=QBqZD84gKETfi;|8t{FMS)->#-#Vt-CjZudgV^WT z#vOLr7rAQ~K2;7XqvU-)(zGO=53&CWXcT%{W^ke2x3ke z#vkk^dBBVNlPXQBVXQ)v+1+h?C9n8s{fi0uHVVmJ%ubNF5q?fp`VewO)Sfc;A67GE znMw<`5-Oe;+a3>Di_ZfoIG-ZDxu88NXXqmkWBl~feTqxJHH z9WpBJF!#pW!8wuL|Ga5q2{%=egXMTzDMSezu9=jBKMJ|x_}y#l4^4R*v8~QDCAerI zL>RRhIrGRDE|&A5tUf3Ic}~HWwe{%pLsu*H1K_bTjF2gj|9c&2Vtd`_82dMnB=9=A zHOJ5*!h zbP@46E(CBrs2P|9^UUFA>_HD$JTyF1*pkz3m^qW$Pt)qKS1D2wRjUAmwP>v=B~UVb zN7z!JB22ykdF1R|7g-A>pef-cl@=(ZF|5!M5tG7l;qMeUPYRYqQ!{z`C=8=5t4OtFHJ%5}C%sHM1;SsOm-Lyr4nk+A$i+jiqNv_P z3{!E;nTxEqZ|_^@AxUs#nJldJw0vhz7%5mt>fmhb@$?3wibP&i{O689$ixv+rymIb z<^%fj*ArsHhNHqAJj2&V!b`fFF6(UriA)c}*6OgDugA~|O*(s}4I=LY8FrX(&hk>M+?uLRh*7o~qv%1*ojce0g}La+j|5098N6#>yXgvtG{K5B)=v_E#L;oSsB&cBb&rJ@Jto5=!jA6Chhv;=C7Kc&H2?8-Cq!dJ_ z+z1u)DsGYoNV1CJ;vG~ugv1gZXP{(Y8wPLzKLLF$lsrqBIMk&dGJ0kSHdUWO#N8tZ zu2KNdkn*v}OSl|w^V316f4nOjSsfXyv=GJV0Jg}@WX!vmBlJ(0ERyaYx(3#4qtyi8 zl+fGQe=`yGe&2|ku3zJOzCKrg6d(8PUW*$!o@Yi-U9a)*CJ%AMfKNAG=ns9bw=@HR z|L$cNi||dX>cQO?gi~%|P7!7~=Q|fNEgb8f>2s*sDQwKdo*DIW1@$cAx+H`icunlp9%>lT)I9xa%*QJL;hbi$JhwhcJGB}gLZcjH1d;)JGY5Slx?;zGiQ2Wkw6>$e4 zBc?mW|Do6;uT8@{1=~tDlYaavzY?hpRg3m5olx_T)}(F&;~Y%Q{%>6)yOH|)B-87U zTp?*d`akC#GUEAMUBg-be=`4oP9+bf>D$f3>!2@@)}?67`rG?^_jp5U>D$T9$j4)p zqDWF4e2z-pZf|OsdcJoM8W0pt=r0WgOZSq;SF1M7B@M{5t?>ba{~&3Xn~YAq>sd87 z|IDXZq%%-mQ|-|15j!vhaNed+4H)`4b7DZR&F0*TXC|;{Fvlt-q)A3B6MInNTSu(! z>yXqXtx?)6vu)z+NsdaE$;YcqQxq8V%UWsl&~(>?PlZo)qz6d61-W~y@Tv6cyd>Ud zJ{um3sQJ+3H$bT!+vWrD6w|?`Tm;iDBRt)p{!fmq*%;48)h1glYYBtdLIcCuR!%7% zOkB1|#j1~M7uPP<5YJEG$3}g&cwkBX?w(2$@6R(OS`gGn1=3x(C_*7wjk`V+8JThbnA%detHT5v{lqN9;^ZjUiT`U6gcN)(7n zmeM%wHg&tWEh9rN#)0LDkxUry+1eigqk7#*JO~SXOKr$mVAxk z^i<2>(H1nnj$(CbMAVoqCB*mMQ30(FOl)CE{@5#J>h*6b%u-`);1En1KTrG-HB7ILyK9y+d=XbD9W#|b%LO-h}j!)`159vtnik_T*- zSYugLU3HaFF7A{g;5-=QQ{Gs&O?k0fV&Lh0c8#5^|6#b#j-^~}ug!9rWqL+$)6n}N zTd|V8|7*i1z0_d+{^I5?EX<~bkA;CYhV+7Ar`_~ZEP6?LLY=t347#vp>O(ChK8i%wJ z)w32fd_oM6)@zmRAh1)dSLF!l?2Is~2Ox5Qtb`P$P1G#yy7CL+TQ(s}Z0-I91ZT(H zj!okOl};oWDj)gpCc}Y{Iehfr2alKhLxiA00ij_2aRvE`3gnWds6-B2oLJ1LkW^Xn z;eSTpQEl0?*+-TJEvFHVF1UD5H{SgubF<2%$J`%v+E&>zQzWovDz3jzp!wj2xhHmBgu$p;7xd zCF)L@n)eav8tTitI@H3YUS%g8HN0|(JO^jL306S;17av?K2B|ZaeZqMj3{eG}1++XBZ z^05FhBM-FCB#&RXCuF6PzcA1M(zv+Lk|CBn%qy>`a=hYY=F-IX zI~L`pj+QuN$&5tj&%>xb3P0`H`_g1_RS4)d$0ML935?LME{;6hYHOS%_^5h0aW&_# zWMR1}JXRg8;0(rVpmG=H3$qm!ok0xyLEH@axwcnHGuVycj{!5fLy~<;K&1PW2bF*r zBISG`1z=|?=FIC@?y>tG+FOGwPM%uDkV_goER|Zdh?T}}LdkTuNjQ!-QMv?kBgu%* z^gI&FBwoHVv4QExFY~1Onj+umT@XilmAkD%v5O2dAPq*6t z5m#7_!gz6UZgEcY2yiF=LPX7mJj87EYCWEQ>b@*G5wcx-Z||A=CX{zg)=3Of@90tp zdRP$m^!4dk%$fXoG}+`|Z|wR}J#zp*Luc=*EEpN^t8(NZTKB>eMeArGNSg3PJMOPm zOGBI`f3>lIqw24$_LF{2)&q}`+Cpt;Q*PFVOj>aqO%crnP+WdFBDk`w%zd;u)n3j> znlsr2?g&I&2RwFt83OWU2_)Hu#!i_(Gf2M?;0~@``3F3|f#djY3U|(KTX!X~+Q9ROb$& za>g;PK>_4EI*x!hAny~3Upv>Dc0EJuYV;0}+9t&@kCT*-3UB#ECjeP~TCG|utlBs2 zo4OBJ4KFV;yDENeZbjjf@C@aj>ZR(ZxfSsY`}=F)Xcx(!Az;w2@V`lJs!u(7RS0I2 zxbpD7`o6`~dFAO6?M|Z&E*Q*YO459_b;0V7uA%2U6WHRSq!5h6f> zi?Pw9mUd<8RJvyRotQtct1zJ>SSHy<^e|Z{?nupUZXQi)zr5!=<$iW=5`K^75luf3 zHHzx1qe-o>jxx$6-aI_tPyEU`R<6)k{CW<^EH;fPHzV_6ns_KCeCFDJ`0d9(0IqMZ zoD{r(cZZrAj76)RTbh%-CsAP$ghvs8D5yCObQ>U~8d3>IOsV;+<5znF;tx=|Vu!Fu z#YGqA8q__Xup^(#ExEcW2ewopR|d~Mn({-MVyaz~{LOMDEW zh4u?+<8P3xCK0?8TFnS~lt#@_Y=&9!xO}ebkDl0y>U%EQ0=H~F>d>-{7i;Ewz(^~% z^IcF={&DuB-eev^mBh@VW0VhkX&?XTM)INeTG<>truJ zhk@y}8n}4uc1MLV5c6-eEt*&gFmo{*eNtcqj&|W;@Vj#< zPi*yiwi+4lH$yyWc^9ZNpw}4}GT$1p92;xL1|Q#sy%m~VC(3N5S>Ax(;bDTMIR%2< zfmnpfXySzd`ht?b07zbdJbfWUF{$(>q6yyd;K_;*{ud*?7i?-Q^jHNmHW6f3agA8a z1z2Vi(nLH_NptKySpF*+OIyVyO|f*XbmT$Q1=iA5t@Z9@j}Z0hGreHnkdi=i&tYVG z%yyVs$G|SY-PRSntQGFxdu10EH&q^m?x|m{VR!jX_uzeD0pa;k^=FTjZ7wl)cz>4G zp72r1lsUqF6Yrb@T@xt9Ka+}6rElvn%36Vr(P4;_H&jyCj|n_5%AKvGWISP2xnoup z`&`~QpfDB0vZ@L!ToHyeP4OKGJ0Ei)Y+))WJ^>u=nBOOh96DqCrq#FWr!ANC_MCj> zao;fWc7&IqZfUW4L;f(VC?f$+mZg%=7qwFu;%+jEVSy(9i@CcBt|M5`1TB`u3>Kp$ zS%*@Pep@kMRGt&_>Gcz+YjeL7{``(?N*zTE)jftp-%8Y!-tgI-Ctp7V| zI5uqnP){uzo|@8pK6bJW25=qywCa604pOCvP*q7Tfu)-wR8bu`=nP&|a*BldjNWe$ z0){FTE1M?(Z|ZNnWZpew>W52UM`7wU&iwBE>k<(2;MibrJyS?RN zEo=*~i=OMQ;@ub?W3LN0bv$$X54x_pB{;)jA!oZe3MXb8Ps#|=lYzJ*RXl2N{Q(sg zP=k_HcsX4HM0!+$gNHa@dIC#)amyeE$u*UT@Uw(+V~S}#8)B$^2N@$o*cV&dk|`Ghg<-Z5+_F>G?gRimG`-WTq{m;hn3s&0G91X{exg|V6W%}(Bu?zh51~>59`&)X#jwUaZ zC&UKK{e#fg1fnN3VZ}smy=ox`2;&$+1UH6=3L%{bUAjudr}JSg0vdrg0|J_yfI~&3 zM0xx+JH4-K!Gw3Zgf|8S86qWza%c|#Bq+sLcHU$`62fa;%(-h(g3-OXt=N8c_bQM( z`Wsyor|SqFRUwpfEu_RL8t)|Nm&=4#oh%c0Hjn@oJGvk@Uj#dtAV~FfDZSW46Dk^o z#}o8KUc8p72w^N|yTfieM6KkJ(w*(eAgzC?TbQsNal@nzd8Xs2)UG^Ndd_;=4iB%;G+0=B;R z<+nXaxfz(b?WtSSMib-Cfr}kaviM@=&J{st+$ccZ^_YCI5(2N=MIAH({bq(0UuYJI zOlTHrDw9e;h3sSc=(`MNy3&V6B5&Z!(N}6#qI!a~f8zK=8 zIfB>nTrvGF2^u+@>&(TSx^0fe6LQ4YYT<0REOo#oGQt$&JaY}owO$aGup%Fq?|^eX zzba?9v)^he2kv<5gdW04Q<3Gh%dmP-UhAWeX&tf3Q{(0YO=;!W^Gju#QPa~h@8%o! z(=lt4+YcA=Uu@Eh(=t>an$j_tUg#zM&%d`MD9wY|FDlPlKR&AOncH3DPQ4PRrK~^A zp(iEk!NR%P11G_Cg4l#6i?2C5dz2Zs7Zvrml5EonCX`rj%#w#(77HFc!~99_Fv=lW z>kL*B#;tn(f7ZOk$M}k_CQOU=8C!uB)_>YvouYa9XEMJovrHQ3#0?=D)@TVIN&7_d z=2lKNCQUs*)>y1HZ>pqDwU`rvv#P8!l8|#oUS5%5bvjs4 zx|yM_P3e-pm^4(kc-L>Vcb`1338~Rq<2`Aqu{e{y;7a(jj*wttUa4nX>ln5^Ig_b8t)8gGGFBF_ft#jQ zF?nu*DqKu*5v)x(qVrH#u=QadUCH?KM|mUTYr}$*RZR%Ahpu6=nN5MuAi?nB+EJGu zpDSp<3iuKWdPN{R@8u~@9khGR$+WOxI5pc8%tKf5TOORN;Pc@0w4Y^xhwC1)+Fbt} zLu?z$X3DwF68kEGYlFk@!P^I`n$qdtXBke^Sr=`~Zy%11>g}mzd@Ic2$`ZD2#voXw z+B#U17)tqi@!1(p5LZhqzn0UN7B`ZPjz8&dt@CNnGqlGEcq9aPq@^`p+%JE!$B#3B zezlB0!R=XvpXwv*!EC?ZgOXQ(s=vlap09GNz#*TLXwHs#XReL$_3*)EQY;0W%H2F4 zhb0uqi& zkVu3UjJm5Ao%nz%ec&bj2Ta&sWadBNXD~A`{Z9n8MFNsh=JLQ8)M-h)XL&Pz-$g0& zV(bx)Pl4^8OCq0u;m=aal{>pxt8igL!0y_34gflLdU~5Fvs{2cY;4Z6BIq7LmgOCx+uW zFR7~y5lu*_L59LhtJ!IX*C%{pj>s@@W-&6EoX!~86xvHj*G&no)<5<)Y zA8^I1`r|-w%%u&6at{Uni_t`2c%r;E@^{7IcCw!fZ7X+6iH1uCx0sA(ghOm`NQ_k^ z@c`izG&S@E?w(3X(kI6RM*Ux6+y4(X8{?-J#=_3_ ziNJ6Yv2bwyC!MOlG@HREw&qA{X{2XwO=n|o^4Dt9w=|-+vN3e}i?o>-S$`63jz)&` z4i^8`$k<5F(aGLO=M%lN*ZYV1M(6msPxsUKQ>{||l;Ujc&D@Rb=}aB1EdMtEt54zR zA0uMs;P@|$=&v?sWKH*1khA@3;QeDre@mZ2l!cMCnfoV1XJf6SNoi^1YG$BksiSAD zXX)l(=J2^P_b)Z~&$!h;%p6$y{~2!eccT9XEa?9gx1#+w*XO@tT>Zlc{1@Yj{U2x6 zKN(kloBRiT`3nI3J74q{Z2FIktAFa!|8dk6<6rIS-?6LztabITVAiLh^-reNr*ZX3 zM*TaD_5Y1!rA5TfO2qN6Y}F@K#r(fQRcR4%a1wDc{k5#L{%Ml`N&Wn{%iq6F@Bg4v z{nNbu8&bv2_FqJ*n3&n=KHCwdaAi;0Vur}rB#5FVX_QNB;7 zDuwx`1q2!7(pRmMsJSgCuk8fn!ru4%9b`HfCTK^B&F6Mo-nTFll%bJOha-QL`cFH8 zfPZzom8O~kaWz{zG!X?}*3i?`PH6#pUG#6pc7))GC<3h?rofd_LMh3)UV&$OVE{cseah>eKzbq_^N2SN zSkAUGU;nGiLa%M}27@$Q}DwVE7i;ZjwGH!V`O}%W%XO-8JZ#j^L~EDgydm zsl^MoQ{?^fS>x%kysz_|5V1j4KS}UIumheRIn-}l?a|Lk0IZNRtwfo@XD;(MDng9& zc0nWXM%^P1`!&6nG`kD>{K{}yb0pFFA+9YL{XcvbYw?OsoQ*^xdr*^N3GbicPg z-?*DpjT5IIjQwBdZpz4J_Qv3p)1I`P%6O_T>qa)yQC57FYz4=Fw6QPSy z8zX#+53&xcI+H$(yBaU}L>`7W(4_PCnlD>2n~XXPzb?;YVyKQT=^J9Ac=aXh@q%vI zheWS(nL|wlTpt7SNaN(m6RJ?G4`SC!l*35a2$aeYmmA8cSo6jqgU^a6m&g3cxAJ9-Z1>TyzBK$=CU`0%W7OlR$n zDLqLB8i$7N1yp5FAk&L$=yB=8Z9|Saq*P*6)V5`ww=`SZrQ0cLbqsXVl~F`S|33MN zzh{Az)n<*5W8F-p|jXu{<5Rh z!Z#wbR#TKKrB_9FaaM))6e|SzOBYF~;gSzsk>Y2Bl`Y+XIrG!vAx4u5=a<#E`>xC1 z2#08R*CB?#T#tOx@)lsahF7;(RT_>k-#DXJyLrY{K1#Q z5XG?&g$j^;ok55&cc8WG zhnm7#_KJ-_w!GE-R!7M~&CF(mcjmeAooB6~2Po^zyEjm)Iz%mT>)YJ7qq4uYbWuQ2oI0VUb}2jE8^A%rW&Xh;~O#SbaFpz(|L+l zwiFq9>l+A#X#nVY>Uj4zBo!-1tBHCVgyLRV;#^?4Y8<~}f^|JM1pVT{^f=BhPVQ$t z>Mk^2%%~rggx_xd+odw))AQJ3&dy{RbDw2Rd$8N zj2#GK!%znS>TPWCCTg@OZGK@v_2BU$5Cv#0$ik4;jd;o#jyU1|q$!)EtClMAXglRy zcD>?t6*EWbbGp0Y`AiPE=(xr4q}{~s!@A}5DPAGCX;#@c-xxlydkxu?(nH!0^R4V_ zd{mW(D!*ggJ2F}@f4FJc#j}dOL_iPoU_{JLYL;u=jswVgsU32S8u>X%_T2Km_&(AV zV03YE5<-oTR;`6y*5d`*wUvuOTb_L>1ugX; zGzQqM@n)rCTnKZ)b`4Z0*1a@&JNke-iMW4#V?LLo8Q|rEB1oG5guC#QzrR8$_R%x= zSPU92*y6(sx=hivo;tJrW_VYj+PbI)Y+>&}o>*-z;z8izo7oJdK&m=c~2}z1up2o($f(-d)FTJKhgh zODr@v1=Ocip!c-R^ue+pyqFAe>p|S2t(#-50_k~y9G2eEm+c@e=i}uXFOuiS$OcKG1PUNZ6=N zpSa@|SO_`+zw9PbCYb(qJk|LDc5tGATgXv(@(3gv6G?S}veF@VrUANIgHdEynRAoe z>jE|NHuN+-cJ1C-c%V6ze)YIHp9`3a{pxic&BgC;q9KM~2-hntFj&QwuIzG6J6JZr zmKAYF-<($pToRzo7_+7oUJY?YpMnI1YJV!8A}pt$PQebjTT}Nl^6s;Ry_O^K251VG ziP3a2eS&hFIncp{L}CtW@4t0!I%GfG`Cn|z+P%sUDtbl`9`Zi^v>e>%nD(@f@P2%V zojheB6cIx|a2%iIe>9o2P$|c15}KxRXGIQgI;X6NSw}m0?B=u(>Mr)#JQ4Ui(k*x4 zu3|cPgf=|N)P>Z>%Gy6=g~Z4@Kfo<4Iy}2!$!(iLV;;YEs;Lg1)sMFxDX1z=xP1NY zG<8$A;I$9y^{Uon^9vm~`NvKt=nXM?u*92BT3GFu?F%8#=mXDc>QvHv+?sdwE$KOR zfP$=v8(|&w^Lre)L)EA*@L(J04e#|ws%fKaY9hAjViV{gz;Dn7dvSc$FUM+_W-6} zODIuymDZ4mrh+3XY+F5f^o?>2t10}lk6A_|GP7L$1UtwpyGpRBAhZ0FwhDSS5_QZX zbm&m?w?~488By0(p*`yR*eG!tC4&P9}5gV$#YY z9p3;r`pyT-irfb9xJye3E|C*;S4YV9}Ac_J3xPu_w$dg`2j8alv*t z&pwMANA%BEE>G_jgg#h=3&_}n4;_=y9i`cu$?swWyoApYn*oDjp22&~ z8_g(JguPCq%-~nBU|w_-4;USSo_@AO-3)xNRL}O~>?k^n56XUE-hrsUjsJscnycFvY9g^YI-%Tc$c?S zj9|FJFecU95lMQFCe_TsUI$t}W+LZI>!;mO!%glP=4BG#F2mD8u1Y`C z+Sy_YWU&i8FC2k;)~)O5h8pn)$UE?0MWmCkUT$f>HF&kWGV)(cU4c#;A>mg|??3L| z(T`pVS5#g>ftdR}KRIFTdke4NTfw~&c|TLYKvs?q%pKCsHMR(-8b9y81@dDPH_s{0 z=o}(GRV`Ex?5p0zl)9-`lfO1F4iD~fu?d3ol6=%vpgRDawLBf*)`tsCZ$+QMGUui9 zPr&mlqP-WBwMVU6x%Yk&?Cy!zS8t(*G29<2nYS-t^o~zXJMSTFuq_fAG3#>ZRCo_9 zHlhA!^MfY!)YE!rWvJoLO$1C%J_rw{S3IV}jgwywE)dcRi&a@OhO+S7G%d#4`_s$_ zi7qV*_1YY``;~G-;)1teun9u+cq+TJ27Qid2)XSL-Uq5sJn^}&+75f`A)EGnE9=dI zc|mw`jCu*4Fl{O6GFb z-yL6M6-T}S()nB8)EWubVqp7u_xGS!ma}9ldW4$@*z*uB_CC%t-U(k9H6Ol;7P>qx zEJq`luCzQLd+GJienhfF9F~8!cM!Y|1&pEM`<cKDLT) z=eLAJv|aX6x)J(8ccI{Shh|YL4<0@UV!EG=s^I5I5odD?onG%edi2dO8d_mjWP0yW zt2+k%Zg`*g5GW-$Uq5(wC0{@NqxNHTGL!))I$DE4Hxq7EIszB(#8zUfMwE4P(}2o? ziOWlc`-j|tLUJhE2jQj7&2vjnyR8ZmYJDdnaWXc_Mm$vb890(1{)33krWB zY=~~&$Og5@CGUt+#ymQ)poeV7?^skU zDb5!%<~CB%-rh9H$8|7ssJOw@R&_exGxfZP1$;JYz;fMl4 zy>CLV((7;REWIin4rc_IamkV@H`_wa*SP5796`+-D#P=KWiiN53cKUkfn9O$Uj;u| zvR&h%jBSBwKukoiJz~DC!h|N=g4gL)cAImdE$cmFV(Rd56 zXKG@q?jxpIopt}A*Yi!arUCk9wddy^YPp^{Ul&i zMC+o|i_S7g;qZnhu(MYB4C~bazKC>V3dxR69u$42mV_%`1Nn7rlgNvIn!i*yqdSzg zpTg^Q!UUUZgV1Z*W>1?z>RvE&98FRylX~kSf}p>~1xCj-q-q^ZB&~|z(o?Kx1ZiZ3 zUo?m-#KJM`1uChYFD zy5*7HNS2-r!53@7$7`jLjOVj;_=g>GZ0myN6o(*@FIu&MdduVm-89cImLk85W%R%o zuxbbhC{F5an%5_ptI(VZq4fjdcn1$SXmK0`@Lf}*2ADCfLx~;Zl!II5F!!bjoy>9d z(RjZc6;bG9Sv&RBvC-gL^^z5dNQ<;Lr&a`Ka8;1r3xnPJb@+7EmNj7O3Gw(kXa&2U zM{=#9o=A)F-U_T&_G`1L*8rcX=2t$QK!4Z%>XwEy z%e~;?Hk9`gpPzgl_9oMHJ6ACwVszL5Uwm$XFh$ULW2+~;5xLQP!zqgYA-|ZDF8Z`f zs!V!`SGkoUXc27c@9s9lk;L&b1lJkW(BEV9l2iSjxLA0hWiPe;9Nve^Afk(4H4DfIKk1Le(Y#*T;n^5Jk`OY|4_VGHz<-(tw-MX#-Q zdL;>;7@puMfMcb1 zN}O@%ezG^tLO;(bp#$;iV`!!dd~ggiJ9wAWR;Jf3Az`3yvIL8HuG_7sv5A-J*VJC) znkP$QkEr%BZKVgif~3>%SOVc57=V!awD>bq~TasWQuC*j?Tc@T?0)|KTE zuEjql_~v7c<5lN%1PwqNUXym?c0H%%Hmj)y4viYxx%u%#y%XJ$nl%IJ+BF-G(S{Lc zUddnA5wPLmQQ;BhD0AHAy0g3rNMhxo%jhz>`KA}98;8D!lcpAqGlzkP6;nY|iTa60R}w=2KuQ31^vdoQ4#4FFS@5~i$T>R2AL6SIIX=#>!H?(wxr{ z3y#=zEhr}`xdMT``l>q0eFfFf3QQ48H>WHwN(^vbee!lGuIg~~`y8VgOel4puaO}F`6m(XNbM{oNfcve^>8h;_$)9iiK%Cmb# zcHWCYndRbea$An3a+&92^oGH+p|c*OxggafN&nEadUm~*R7>vyTs^WM7>OE5NxEoB zvul3aYe-trlxjt>@NzrnZ<}pqpTOsMSmxBsK8Q-w@HyXeSUTWJKXIko z*Su{ma){0B$UXR4aM4(~we2S6t?n)9{lmN4(BRp`UV@0O`N8%UaTFz*64UN=+wkhE zr#5tqQ1xtWf7BSikooCZf1y``@+G}axXh97gx>I>yB+CmA+?y=W1|s;1Er#7c8QNg zsrFEYmLRyV3PiaJt_qQcfw1o$geK(V z5}d|=K#zdfmY|}mw(6_5Kj9WsSx~txe0w~W-2<(EzC65*APip6&Lvx#?{8aJpcwzV zmMgSukJO=p7t+WkxGU0n$kZi*rXN5TKE7KO8)-)utW5BQ7iBmb?7JT7&o09zEV>@_ zWGF>Fu2J6?+Yc|q^NY<%LR51vWZkZ<03?jL>!oeANI9U>ulubW> z9%1EJlOgk(U>?9veG-D+-(}@;J~14O~mOp4{?Mm662brQ_}TbA6 zGN)w0y>T4Zh@3L({TvH9;aUSpf54$dW6-rrv~>h+ympaa>c*RPr+V(X&GkLepJyf zlHZfRgnNzA)!HmJ!ei^HEV`V2f634?SXe?j1>B>xc6l3jROu|)oVwgYw+>lXtM42B z0RclP>VyzPh76-M8d=P>|hNVzEGVNjzwtcXt$aq}z5XElVw9+BT zDy<_%3(Q&&fskBw(xk>A=`Gqb9C)B#HxV%an`Za>TKu^r61m2c(v!@S){}5^@#+tA zjKZAaEWUyYd6kmrQn@0z61ifz(utzdS(CZBW1KrqVmWLQc#L3}r*7VE+iU?!Ll3*L zJ5*;lk3bx~$|^Zm(M9^U?`_}kkP%@+0vCeV1ADqrwq&-7K7$MC+Q`#=}0%y8gyFI#Jx}m#6y1#CbZk1hDU&468XLfQnfe>~Tc5`mET^8F4K1w%HdeXfD z9)+67JgGd%J!v&Dfha(X3d;5bK1rSY%xRskS@1uhs5pmp$yJ6FgDHjW(hsN~syOCF z(z-{tf7<=w&}x%Q>q56HaP8obN~0Y|Aw(ubCqyO0JP5z(1@t~`gV|*_iEGl(p(Jfb zJry-+dP-{&R~X79Y)Sh4DEcKI9p+p+vr^46#U&@kw?7cnPC|5GK0#RWD zeT}_n+Y(ptj}4EA=dBmv7h!}Y7IO!4PRJ4!u_wFG_8G6@-X!dU^t&q8xKHV?f~bQs z_7StF6lPICM0p`9qn}wM2}RNvB|p$4@(5;syDKo?6<{6Y@2yO(q>Q12 zVPWR*M_gl)?2oj@v@%Oj!Sp;!WI=?iLhj6=qdQ@`pvm;h5e--Bcs9}`^pP;?WXO?b znD|fqIHIsPvY|MV;y8)x7#?!^AsRO+%pt&a!Yi2%l>lt~r|d%|0-MmCXGIJWNg3WPTtitM6vXDlb1#6X^~=~lK7H5iD|}3FrQ{|1OkGRZmxva>$%mX4 zGK;NKT`K_=0rP+~w6TejBYA-Qy=jAKShM`wrrW^VyxaKOy4zoMlJ3gtK#xX6(jm!j#%wB*mF0xrx9cET$KTc0_s3lZvm5uS^&jW*=@G#@OTDE4gmBmuaqe z6`-H7Ic&TQzlO97ICJD{mH45DbsEe*m{bKm(HN2-#Q=@irTL8i=>*MO9I@`<;1Nd* z>hDQCqqPhA6m(1y9!5)+Osh)-WDYGzwoC`-QClTA-xA#fy-9kI3pYz0<*b$`7VjO4 z)ce!Pdk0c%eQ!fN*;0sukrD?mrO=Q@$L;NE7ZE>Dr5tn(}FbS$F0}x0-5dG zJ6Z>Lc0Y#H8@;rlKjl!k`Ss;3lZS?<=i4hdSb)E)Z&sHN7RooO>}KFn8>`Ccfw%T|4QJfW=gaMd z#uq@9`;O;jzkra<$R5BtkNfGP28C?Kr@S(AdG2xHbGelgy}2Vdp8j#$Z4z17)jy-V zV^(#=^0?u}ngN9?7w}JYi>^49JP7X{jyR5vc!z_hwkl#T&Q>dVjj}Rkb4;AYqWU+1 z#xgspy?ySW>{VP&@GLkF2dYC z)oT5AUDb|E3VR}o{&R()WF+0K#0t`#Z6%!(KiR=BAPteUFqe*^dXb5Uw7Qvg#k!O) z!KiUAhWO1sRwv$QqE<;~?n#@AzV`@}9$rLL4@SfJ@{mS@m8!}}#U+Ku-P>6ZbG~t` zXLf15xi@|huOWpe=1!o3Yfu{)3E@P#A z)~I9ctgQO;r?&o$`E>IjUk=V`dPzIlEzP8wO@aEcChf`CqU4jUuAi00O~4J-ysEB< zpub6Vdy}oIH7oV#j@2WXZmZgy;0zt)2BQ8RFx}x`UJk+FrZzyziigAJT~p)kk9YH( zV(ud|J<>^M4NHSm*DKxPtq@(`CfVti3qHbd#v)V?wrDDBgHq+Q^wZpfXsS*am~u>u4Ah>f{G%8APJqxJOnVba{TnMxLg7Cvqw942 zGo^frKavU@+?~ho4s|GGyFCmqg3aZvsB84(?9Vu2*j^Q;q4u%ryDT3Q098Mo7-fac z>X(~!f%*8+2YaP`syw*kA78zC!CvMr&VNxk_K8;~X8P4?avr0@&#ZYm2?;sX8f((; zj&sB=4j{TWIqBaFv9gS5TUQecUI}h!(j+d>!59%>^qDvbMuLm!X1ObOXp&k6py;i1 z#m2{|$iTS{3%W>l5@j%rr2kN9>x?>XRqSYoDN^1(_HLu~^IpnbQTpBKBTmOS?Ezo) zaL7)!kjUH7Nxnh=e(~k&k=60Ere?j)fhMMy79N6u`*-v0s+Linm=iVH-tFi-3wyiB zvaIjNjJ@;u%ReUPT=!FQa^6RTSeKo&B`F#snp@P?=(fgfU%0zn948$3dJy64>IOQA zZw^w=#ZUa5P2OMWoHdf3AaCF->eNV2*&@8&wOcYOGIcVvHZ!JhtTz0sj_KfTY>{&% zxM!7N(;JqC+Pqir4A*^$n>}bjK7So>Xv~e6YtB+!Z@`uwS+fw9Djk?^Q6;FkLcS#` zI$D{6+mX=oxZ35BlEb+D!mIgHSRh())ea)uya>f)YUh#~MR%ohSLM)H+G$c|Wu<__ z{ICE&(N{UmI?3%a2tNZ@etvN%%-%T?3La|l1vKtz37QeFtLicMOv?2gVI}0m5OCpL zZfv()t3u6tdT1{HJ4tZP8A?WW>uLUMFr&NY#JSGgBrvs}rJJ#Ou`Q^ghfjMZbNx|T z$dKe5@MZC_1s@bejBzd#%flOBs!DwUe(N@R#gpiveX)Imi1zK}KK6O=R(-(H9@0Xi zod04Wv**sF;9Yt;bG>5L#_i_m9_bO@$%54u@-eS6os*x zvLlES$b`)dFhRj4YcDT!0K!ts0Ovxfa|KqGNqlp=nw$k7p)SA+%!Ur(3SDS%y{!8g zC$Zs3aF(mZsP)cRA7N<8`mIMjH|A`9qntIKC9;pAew7lltwuE`2J$g6m zY!O*6)~RXnXKs=f(dGoq!sm3A`~>}k;{+l#gM|(@tfxmkFesP>SAR_OCb;~z;i!Do zs?Xe6qn44eR1JZ+wH)z;bUt3B-IQ9}v9UhCzI(a5gnePvw4!*b(N)%2BURU6%AI;? zUQoLv=@0O>t#27CD^H67cfs_3=b z#mVTRvFL)?u$1qt%c20Vj-LA@nly1RA6PD%?AKzS9OTO{dGd#dbKYgpXN*~BM+#n5 zU)+CHDSa??Y70x7a*KFS{5V?ycZWYk*IfIS4TvnetCW>HRG@vp|I->7)8~e*A#<=b zjhCz;EaW;XtmkR79#M;PJ(ofyigoYWUWd+j%G5^oEgUJD?H` zRD9T`IXI6gcNxOQ^XY>S-8(lL8yG5WR~67J9x&UOd6dumOc>C7=NkbmEb5TK1jFM zGR^Li3YLCtdoXR5mERV?ccD#;KOb@y@bE}+VSB@t{cxan0g;l@1(OILN9Lw1g-uTC zdVa6a#oTibBfxgl%*(huk`TYhSWPM?bham9d@j}!=Wo&w4Z8whYmTGAJ`1p#Juy+5 zBYXAVJe@9U$&_4JAj2y6Czu<9x-A;-m2pU=G08PaNC>eTI_E%&KE*l9LRkVMWp*Vq zaQb?({c|n>CBM2=&S82$a=IGM4JKg`!?FknQTqLkf==F!Fz|2@BwGj%qSZU(im7n? z$C*dQKF!R#@KgCM9oDbPH(LdPfRk)Es*`h$RD#Rz1 z!O}Xs;7*z^{`RimaWFHoJ<7>~Z319A9EgQP(1~VJp@gmiUmSyB7e}J1qw6jtu$Vx{ zM6nG8eHO=nA+;h|E@55z`H5dhiY~vUfVz+b_|hc5oO``0YbIhx~&~ z()hW27Us7GQw6>|23jTu=5oSQY#KbGdXokWivJOSuCW(H7c=u)mOXWVs1w?m$9L}{ zqV#>)^p0Qrz9^Py0+$|=-jM@p_?RBrE*s*^@lH%xs(1-=MNCYV$U>|&1oA_2J|^{N zfz^>^OVs@g z6QYaP9YN}1;t8q@p?wX;?s(|pL4ar=iC(b*El3^n94C>BNE_!cxmvHrjz!JY&qeSQp} z?!L$gX;Rgs<0JN_@}}}Lj6IKgmdwq^_z`(qJ-X|CE$&v}*}(%?k8I7+-}4T@Y`HeA&i_>DLi z>3?-ea~FbkIL8Ztw&moi{Nw0H+5R^z? zImKNFu{b|7_Ff<3S1-O7+mfJ1E!r+>McZ2uJ@26Da^f#v^~dlsa-Au`$p}7>3Ao%| zq`^WmzgtYe&BhcuU5&LR*(SWgNEMn&H`U=#R2RcFWyQN93uwy}lWLzf`9`O# zeI`SAWWTm-Hkcd8Lbvi_R|_n`B#d)S`X$a%?|2M^t6H+qx0pS3XDhx}8Y^UN{gqx8 zVoi?`w~R81!g{Y;Xz_)0{eZOGL1v*Z6(cXUcgk6TJi2g+K2Ws%Q1USmeRoaRcdQ^* zso|w2DC?>}Zu_5X_NO7W>+cP3q*1SGhsBbCEtzQBu;|6D_7^dAqxNhq=Ax7xG z)_-lMWoL4qGZy%Eov86>nu527wucsh1O~QRQytKV_d3D57#Q%X@=e?E-k`njp$py& z@j_`IYwTx(Z>5P%$?La;m`oS>8yk6H>(dhlwFKrzx_7v7v- z-AlFb`0~HithPbTC#NjsJgzTQ^vu+zB7-9^(O z^!YAT!;9dlp@FUTbo%B1b)oZ9G*2R7`(<2aJb|ARPIio&`*)j!vlh{sMD}~C2RtWr*s2*{D5bFZrNj2GzM7ZQTEHLb87 z{MZ*a6#x%JsUdRa96r7sWBf}U%Q~0e*4au>qiv~S;2M`dFEcVg4KlU}a@0`%;dS!} z60(dR9Sw;^t~T;QsZ?({cE4(R25{r5BIy+N0+HbykFe5W;$vdvs-!q<@x1Bdy~TuR zHHW1Kt7e6QIT}T~;YYR-hhn0;lapg&gmTm&MW;c0hJDJ{2~wwF3%_*Cs2BFLhSdTP z$jb+TbKr%R<6mF4OKSl%cHWL|b9EVJE{!Lgw?O|^6)uZDFTwe_4gQ9)r=vnap>k`uyD)LeAY zL_ZjXt9bd}I2ae8;jdicA|=^951~0(H49Xt>nPcC1WY)E0WlOc{oWW}bOR#)q+(Y(K^@N63xInIWg?g*|!|HN@!!C{Zlz%gCc z`#q3~CMaJ5r=1GMb2n#O?|CJ=^orvghsi7?2r0b{sDVxkhKR2Qs3KCGJyP=Q*P5}} zcHmde^zb@vhP^H6^WEzwhoeY(Btm1_j3RDv^h1h`R?s16gie-IuWr9rj~41bZ8XQa ztR}ZE!w(`&4K(M)1oUq*hM1#-NivCkpEree>1CCgGpo`njQaKtS{%=~fNaK-4L=}S zJiLhGwcfd9F#)mZ<|UM?NUQo%>(3W5!KZD5c!T!X%bF%?AOdqAl^-Ngr|h|SX)Vi#z%d!=io!FKsQCS}FP|_akBWLAF7q;4ls+MCh#ljFQ?Qsm^i(zB2Ba;l5SQSYkQLz*3 zD##=fjDY=RfpS=c);}Ff`1CTiVv8kW;OAcDtQTTPTQR)06EAh9?I+Ur0A5ezwn4qk z77R{boknQUOTU2wz(f=s0xL=^!6wc9&R&$??6N+?1;(jy@TOA{x)x>;r^!+x8 zgUI=eAVXGNpo-LLVtM0iTAfdpU(Uo~{xd#ixh*9m2A%MruZ@^(njo|DwN1KQXa^1T zPiot%%(u!;Yzt(V;|mjngdOSfvQbjX1@0J#`51P*SP3>Ny0H~Zm}X3m=Je%y@sMbn z?;`3XBP43O-dvE=#E6-m-W|hohj&tZ=T4sVMWRu1cKdV7uTlg)|Z1O2aexq@oKU+yS#a| z(^4M>L3HFCL>>}T7|SKEX(y~(3ZdrXzAC2jDXyO?g zb-Z;$G@@d&*=ps?$$Y*jcjy||`nN&v4&QegZG2fs2ruoK#PS;U(d%l6^yjA@M z((nXJDDz96LA=6%YQA;qmnlo!M?fo9&8-O6KN$z3eUXEmfWJWVrqzT;sw9TS%gEYR zF1+6Ob$3%rY!-#Aq$86u2_1Z+6E+eyg+W`G=a*0^uye_&!VqCiNdwKvKPH~K=%#ue zb!{7$5t9kHUD|?G^*CyH+C&q1>>Xt#+}Ej~jk}03ZLs@p`)w4Sq9p8;;>5w%LNSw) zDxz+gmCr*8&XNY>G2rqbR^T~RnPwH*ml{KW21s3_q~)u z)>&*q;}sdE{b6f>K!+7Z#h84U3;B){9+OMjO0D>#sA_p zll?zf{D0&IpJ@jgeh+)}SAOLpXMfsOimUhYDCk1l;W9;zhaVGY!#*1Xlf+z^Pm_=A1kwoch@Sy=p~ZnfMwYjvB_q ziZsX=$r5%ZPOT^=fHXZc%PRnTe}8{_d8xunJX$imZv49=!QC~T<-k3o&K2o_?nZlu z)^#B!c})Q`7270YMM}Txa>+4Z_2kDHIo~rpgt@@9W`tWX-@}t3N#>hs62znWtrW3y z4z4j!QUTj|m`1MS^6_eJp`cBT$*Az=EOLBz-=IZ~-utm@JIX++m9PG-7D<}|e*rJ+ z?Bh^Kn>Llzbbg=Tn9jO!6JAHpK%M~KZeExO2k|_JGCm+2ta|wpWiUT(B%h=YK>tPi zGovkJ6<^Y55yPG=u&;Wb;30S%0mJtos7hWJ*gL=trcaXxGDcq=104Kf=!=ZGylPT> zfunPLLcLg)VUKJcY=hu6i?VN{xP6MkW&?uIp);7gh7 z8Qe`dVGtnv+J(#DP=LwCat`#z@k^)6jvW6jYO}|k3x&r+V{n-&Yos*XLt3L{JjO`&bf;seT~HJ}xr$VE|8P20_NCFzX8h)Usf2 zy3DD;H{(>vS2KtG5LA?8^w9ai%Hd2UwaK&Va@a5O;NT8|7?T}8o-h$vXMId!-1Ujo zaUZ;c@ArPRubrY+q>?oT7;SCT7s;YE>)EAxZ!{$C)R+zp#?S^AvZB!XglF>dl7A-G z&BZxMkrWO(C;FIL^{QLgN*XELgDdoK)yX~}c~ftf3!fGTMa$8gV_#X?p5%lFHHz6I zn4jU%@Py)^BT$j))AxR*E-{J})Ud-J-olD038BA`!Md z)83RJb7m=1hHc6`WTryOkSUpE%uME)WDFUSIfTkQPZ9Z^?Q}ZXy?x&~*Y$npJMZg{ z=x43T zu-sNUKJRz=?EPJ7VV^|pS&1d7yseg+rOS#H-7ROYvsQR`bO#1%(5%E4yp8+f@c_}M4F-O zI*t5~&vnBsU1`n{5=q{@y!BF6E~hx5ODQ~p!DMEFnc876J<7No0~091F4$>?jJeWf zxy_GrkGyuU*KNHIUH74)cs+1&b>>b}US`GgQcJ__G7D3uyK>&-f_FUzn3}FzuZ^Eh z+6=L%vQ=^(?e2`3Utks{_LE!awm2L#@A`hKJgl;QVvfh9Zjz}$Civ7d|5@2}_?HM0 z&W#Kw_Cm?I#5P5_jMjAaw2bbs?nCL%q#O$9j)drI3}1+C&eH^aML&D6kQvPJYUv({ z=B?|opj6tIL-4m2S~_4=lr)N4zGieZhErzDE%Nx|uK3~g2yl>dUy1lU<*CXW`qTAq zTT5hWI$2aH<|a=V)~wmGMyJ!1kUO3wQBN&-kz`oL z_?Dh2dSy_s;Mj$wf+lKRnvlBgv6is|3RGBhXy43LwmT`5A#R)z1nD^}0w{t7f*7I{ ztFlu}lN_{GAQ>kgcY&;oJ(sr3JuY=#nx97}}Z2ql=rYx*U z3ZE3iT*8NrnpoB;WvoxT!P7&{T!QC9#{~^&=|zdqbc94!k|(^A&G<*%$5hr<3!JZS zbJHv6uxM$~3aw75ky#12aIh&>7zT#XWm`RWxc-(Q&#>Gun+Uk3zROpwV#|SSgL8$l zLMt9>Ig+S9vL2_QD6-DKZGO>#zvlRdk8k95&KxQ>a=RxS{30vETpvmafr9F+NwNqY zF@&Lt3z~>nXYxwl1wl1tsjI2`KfZj`IzKTGp#Q{9M>8b%3#lP5MNIiEg-gY=M9e9E zlw<3Uc(qVzJ3EEYNgLMS8}9kvKFaZ2Dn?8Z8!;<{vAyx(D*JtPqLZg#6c&sq$`C~s9?rVM zd15?a5gI2?UsG&{mb=Om!u*WLt@PcV*y_lLPrHL9Va%JgHGBvID_4?~%)UA5tG4O^n7 z)e}-pD;L4zMDD1B{4D;Lm|>BF4Nksu2XCn7dua_19jsG}_p%qhR2DeOeZ!<+=8>C$ z+wBH%twm}G{PT%r)HW@f7h$7g5%9ShN6X|*$%kjE1RskLDljwm2Is4_o=An3EGL07 z`fJZIwWf1Z3wwLN>0n^^_+a!LL3bl@@u=gL$%l=D9ZFu44poFK1GWhwm1Z9FgTh}= zX8@9%8tvjv-D)77lo8^}ia|jk6B5pcRnl=?VeJwz>GJbC2%T*Ino3^6z{mbQ+aGtH zq7XW-%c2h@D-~^eDz7?xdOv$__7z)9bn;k60L}X6)j6`ZTZ~TCTF!Uu%8SJstMos| zmbR30JGZNOnQ}DuGn%8GWz6c>mFCINoXxqEL)NS?$~a2FUf<~uPbQhD`AT{DNYLxy z$ByzA7GbN(wp$pDu@%IzFr>f+c|My;wbD{y+$HLu7Q$n5n1eY5R)mHQ8r|7W)cEx3lqYQyR|i=7;bnm#rl~v$$A(Zpw$Y-dp=T1; zG>gNB6tA$|<`t_Q*C`?FJ-e>mEtDm<|6?O3&)*1cU+>qU}s4a_zm$US>*e_+9mjwi4h@S)6OYCrG#Eifj*BT)oFd`_ z;`Xtn>vhdrDD{+QHjVwfF;@2dC~P2C(CGz+2?_F;%CowXTNihV#Lc6as0T0TZR!bT zjpxrnji-dqv{|-ksFi1+J81pKqr9Bp*B{hsHcf?7CwQiEp>Ma8%W4z5oXp(K9U#Uq zlWVc$fX#GxC8Vg7oEG&R^U|7a6raD1K6`y&%Xa0&LE(E=Hph-FQPfo~x<${wI9|o? z<5L~%7nLY1s&Ihqth`Ex&3c2LvBY!9WBwG!6hx)v%|RpU6&^NFkq2x^a!reYdB<6j z?zi=wxX%^65I!1Pq#qPb%j?rC_T=ocbWb3NPvRK7kpBb3y>c@FK5Y3xPBdX5z=*Px z(T>Kx$?sTg-4a>w#YUSTl26iaG&aaVBX%!k#h*6Y>5lUffb|Voh#$$iqn&!P>#K(5 z=SwqToZy6;yhWh1D%A55;V+Ut z-P2XzkYXi>Il3b_E}(a}o@?Ec!y%Lnlrr$b)!X*q4VDMGcXg;x<(w*1xW>Gru)3Av%C~* zkYDUgtfCUqKd8=^L3yHr=DLBO6uT`^g;gAeO(}?YBVO4?ZcLL!paT zTB=6}ieBhbc&D&hW=kq#Fcpg&#wMRS8>AJfyIV}3+$FrjtPoyYAABZi2irf3MZBW@ zl#WSL%cECSJ&h%I?Vi&5 z=a(|+ogn8rS(g4esr;%>S>$_Asq30imjI^B(WKmL@XO&Zfi^ikH$H2p4Jv|%zb-kHjotu8wpYo>POf|vcPHU1kO0Gs}w}q?0J?j*4zD5N> z&0>r%UC_wF7~7F^_k6;mF^#TH<>N}xK=^o=~;U7{yrlW>lYkG+$6FKzPfN_ZQ6`lVQ* zqkL7$?VSge1QO=`MFyD1ujp)YroI(nlaK~)=)2JBwLNMI)9ttFAX~A#K`Q2MlBQYd za)mqbCgrjh{jKuEv%4)8Y+Jhr__lOCditawNi1{$Ys+YwaK(r84?ji@E7sy>SM^=xbFGx?4)kWrp*W6Sl&nK739DZ*TJS?K8YT7|Lkf2(44_#@G&{S0ly7~rQ zOP|lu738Fyv04Igp`BU!=;W+4bH%)CcR@Oh%P=TI<{_6xYOiv3v7lC1yt)tTs=zx+ zrU`}#&oV7~ws$+JsbR$$y>Ys;Rzbt}SpuBb+-x3huhIK~+tod@WZ$qUX6A(64}Q}p^7xYhM+ z3u)mRitiHi#EeHv%vHR`rF`Fc4NO^uW8WrOdC?$Pb2YC@im~(NOFQovycp?D8*&wRg~`Yax8esi$>23%eq_SQS^qSh*O3U7GIE_)Q5;m zf1XZwRF2~+XU6kT9VxG@RMR(}={|3Qn`!N6j(rHo^m*wlea(Nv80AsM!hGcQJ$YtPs*czY)A6Gz1MDhKZnpGw1!1Yu zO;xZ(V{e+5t-|m44>og2@#P&+k_x0%9cY&3YzR|b;&~LFZci_+4!L{nE=k2BQcO`r zLZ*YF0u6I_WnwPt+Ko#$1mca}CFw>JM3qxvsTyPp>%$O*p?=JF?Rr}FxzEI^zmlpUl!kWXyr;0%%FA|R?1vK|@k&Gv@PeNM?=qgx&< zEFm4I>+|8WG)19vH=X;7wjAdl!9nESBrXu8@>6(ffp(sOP``4336oY?^@+le#T zN9Wd;%~UCwuPK>J=Ms{&X}5%>Y`c4u7E-Edu<5Q5i5~vY;N9WF#$3B*d6M=#tsyNj ztvKybT2_4(jEDlefCa6coU;?)PUWhw4;`;TrGH`;5dKOnDueRh7;#ToO1%(qk4M?bY)*afkZ)&Ewo212p)8Myc$6G%_@%2x4E!AQW?VIepGXyI z_~oQj7kAEcv4W?y1<&;+hm2&2tQn}T9O5;ULUo>yRF;@xGG={nRg_=x$(iJ&cfk;e zK+fo!Q-wP-+w`j^vC$Q};+VATKBfF*7b)AgGLpmjp6qcHeO2<{L{PuY<;y0I6%t*F z{lq+4FGgB83f-yTr@I|M+?d=x({p%!*You3Y~)kzsLc4%e-Qx)zDC?XP(2Y)X4_N# z0#wghUVrLwl*MjG@AD}8MHm%2jM*#WAj_2yuudN z`dI7VCWC=#ihwd9Ap3>{@xwq67@#W$M&K3${jtoRmLaeXh*w2Xk`(t3h{MR%*7Dby zg2Kk;I@XTb=5}VfSZh{Mte&x%&TpEBz|tas5*pSV*y)~4ztj;#LjObJ^4(sXOsC?n z?ZwF|?$av#$MEo5psqkb-vvGYvWn+-FMQJh{e^lYAQt#f^}?R`9w6igc;R=40ABcA z))e-KtSKBPr1*cEt`Gzi1mS~&pl}EX0zv+Tgw?muv*EFHv^BE$P1Y33Yig`(4P+~? zu`Slx*22QnW>2A%^^gDRS(xbp35^p7wfRX5bWeoy*QEY;kf!}1oPQ>8-cAW*|`l)))w?tR-6G6kVs8DzHhq8zp#CeI{(bL|6%)IGkA~w--(VNY!uEPe~ZDm@Tke@n&4U`pf6Gb zgZTDR5{Ki2U2)(QFbMj)oq)#b-j9Dau=^F2f3ox6p+o(lh<@kypVRmLgWund)nD=Z zKdvDJB%Fcl|24n=wvO;$@cZ||`oF^O{Y}D8{Km!oZ^>g)UU5DUD=aPqzQd2I+g300>9_)+M0!;sF9Q-vP$_`t08u_W$h4{e=?ptq}gI zQ2r%nLiV44sJ;9F!iP22#~HonoxjEUpA+VPvhyz;2$1~;AHJOq)>s2lFd&9b3jY2D zLcrlrILH9>9Rpqu_I?X8|BgX$A2;E?zGHlF2o&(*J`4(m<9dmG7>pkU1}L}>LxRD8 z82f$!0_HLm~mubvz8vVaN9$6pBFNPPrfFLZN_;JCKw6FgOxWf!~h-lIsv8 zp1m+AKOEmLfC~^fru?`b4g+-W@i8Q>IN^;$;a(W_V+c45&o&qgfq>)f8NiUZTM+Q( z0>ISF78F#dW71Ongpz&L%pMFUx4k$-~g@g{DJ^tiO+i=D}j0V zb72tt+y}nhEzSPtB2f6;1xA5^+eYxlA%Sb4e#Y=~3z&=evE~4`>lmA2tx17*E+ybX9o#X< st7u_i3&I5h*G@~C8(4sV;NxDbNr7v7tZnxWOkf?L&XcmSiOF60FYDwERsaA1 literal 0 HcmV?d00001