![\"Drawing\"](\"newson_and_krumm_exp.png\")
\n",
"![\"Drawing\"](\"newson_krumm_score.png\")
"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"Clearly the quality of our data, by way of GPS noise and sample rate, has a huge impact on match error. Taking that into consideration, in addition to searching over the feasible space of $\\beta$'s and $\\sigma_z$'s, we'll want to iterate over different noise levels and sample rates so that we can optimize our map-matching regardless of the quality of the data we receive."
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"### a) Define the search space"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"Because this grid search takes place in such high dimensional space - $\\text{noise levels} \\times \\text{sample rates} \\times \\beta \\ \\text{values} \\times \\sigma\\ \\text{values} \\times \\text{# routes}$ - we use a smaller number of discrete noise levels for this section of the analysis:"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"collapsed": true,
"deletable": true,
"editable": true
},
"outputs": [],
"source": [
"noiseLevels = np.linspace(0, 100, 6) # in meters\n",
"sampleRates = [1, 5, 10, 20, 30] # in seconds\n",
"betas = np.linspace(0.5, 10, 20) # no units\n",
"sigmas = np.linspace(0.5, 10, 20) # no units"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"### b) Search and score"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"We've wrapped the entire grid search code into the `grid_search_hmm_params()` function since the internals are pretty messy. Behind the scenes, though, `grid_search_hmm_params()` proceeds as follows:"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"```python\n",
"for route in routes:\n",
" for beta in betas:\n",
" for sigma in sigmas:\n",
" for noise in noiseLevels:\n",
" for rate in sampleRates:\n",
" i) simulate gps\n",
" ii) match routes\n",
" iii) score the route\n",
" write scores to a route-specific file\n",
"```"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"Depending on the dimensions of your search space, grid search can be a long and arduous process. In our case, on a 2015 MacBook Pro, it seems to take ~1 min. per route for a given noise level and sample rate. With the configuration specified above, that works out to roughly 30 minutes per route. While this runtime does indeed seem a bit absurd, the beauty of grid search is that if you properly define your search space, you should really only need to do it once. "
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"collapsed": true,
"deletable": true,
"editable": true
},
"outputs": [],
"source": [
"val.grid_search_hmm_params(cityName, routeList, sampleRates, noiseLevels, betas, sigmas)"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"## 3. Get optimal params"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"### a) Combine the scores for all routes"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"Because of the slow runtime, `grid_search_hmm_params()` saves its results incrementally in route-specific files. We must first recombine them into a single master dataframe so that we can query, plot, and otherwise manipulate the data all together"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"collapsed": true,
"deletable": true,
"editable": true
},
"outputs": [],
"source": [
"frame = val.mergeCsvsToDataFrame(dataDir='../data/sf200/', matchString='*.csv')"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"The first five rows of the combined dataframe:"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"collapsed": false,
"deletable": true,
"editable": true
},
"outputs": [
{
"data": {
"text/html": [
"\n",
"![](\"https://s3.amazonaws.com/mapzen-assets/images/data-driven-map-matching/noise_vs_sample_rate.gif\")
\n",
"
\n",
"2. We expect most of our data to be within this bottom half of the range of noise levels we tested. According to the [Institute of Navigation](https://www.ion.org/publications/abstract.cfm?articleID=13079), GPS-enabled smartphones can typically achieve an accuracy of ~5 m. Even taking into consideration the fact that most of our Open Traffic partners will be operating in dense urban environments where GPS is notoriously inaccurate, we still expect to be operating primarily in the realm sub-60 m of noise. **It is likely not to our advantage to tune our algorithm parameters to optimize for the worst-case scenario**.\n",
"3. In the range of noiser data (> 60 m), although there is very _little_ dark green in the plots, there is also no evidence to suggest that we should expect to do any better by increasing our search space. In fact, our inuition about the limitations of map-matching with extremely noisy GPS data tells us that the opposite is probably true. **There is only so much we can expect the algorithm to do with data once its positional accuracy is less than the length of a standard city block.**\n",
"4. Performing a grid search over this many different parameters is extremely **slow**.\n",
"\n",
"Taking these ideas into consideration, it is clear that expanding our search space would yield diminishing returns. Instead, we'll use the results we've already got to make a more informed decision about how we want to fine-tune our parameters. In the next section, we'll implement such an approach for one hypothetical transportation network company (TNC)."
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"## 4. Case study: \"A-OK Rides\""
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"Mapzen's latest Open Traffic partner is a hypothetical TNC called **A-OK Rides**. A-OK Rides happens to be based in London, and nearly 90% of A-OK trips either start or end in the city's central business district. In order to conserve bandwidth and processing power, the A-OK app only collects GPS measurements from A-OK drivers every 20 seconds. In this case study, we'll see how Mapzen can use this contextual information about our partner's service territory and anticipated data quality in order to improve our match results."
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"### a) Generate routes"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"collapsed": false,
"deletable": true,
"editable": true
},
"outputs": [],
"source": [
"cityName = 'London'"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"collapsed": false,
"deletable": true,
"editable": true
},
"outputs": [],
"source": [
"routes = val.get_POI_routes_by_length(cityName, 1, 5, 100, gmapsKey)"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"### b) Customizing our search space"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"Since we know that almost 90% of our GPS data will be generated from within an area dominated by skyscrapers and other forms of dense urban development, this _is_ a case where we actually want to optimize for noisier data. As such, we will perform our parameter sweep over routes generated with 60 m of noise:"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"collapsed": true,
"deletable": true,
"editable": true
},
"outputs": [],
"source": [
"noiseLevels = [60] # meters"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"Similarly, since A-OK only collects data every 20 seconds, we can make sure we are only optimizing for data generated at this frequency:"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"collapsed": true,
"deletable": true,
"editable": true
},
"outputs": [],
"source": [
"sampleRates = [20] # seconds"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"We'll search for our optimal $\\beta$ and $\\sigma_z$ values over the same space as before:"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"collapsed": true,
"deletable": true,
"editable": true
},
"outputs": [],
"source": [
"betas = np.linspace(0.5, 10, 20)\n",
"sigmas = np.linspace(0.5, 10, 20)"
]
},
{
"cell_type": "code",
"execution_count": 42,
"metadata": {
"collapsed": false,
"deletable": true,
"editable": true
},
"outputs": [],
"source": [
"val.grid_search_hmm_params(cityName, routes[24:], sampleRates, noiseLevels, betas, sigmas)"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"### c) Retrieve optimal params"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"collapsed": false,
"deletable": true,
"editable": true
},
"outputs": [],
"source": [
"paramDf = val.mergeCsvsToDataFrame(dataDir='../data/', matchString=\"London*.csv\")"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {
"collapsed": false,
"deletable": true,
"editable": true
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Sigma: 10.0 // Beta: 5.5\n"
]
}
],
"source": [
"paramDict = val.get_param_scores(paramDf, sampleRates, noiseLevels, betas, sigmas, plot=False)\n",
"beta, sigma = val.get_optimal_params(paramDict, sampleRateStr='20', noiseLevelStr='60')\n",
"print('Sigma: {0} // Beta: {1}'.format(sigma, beta))"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"### d) Get scores with optimal params"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"Generate new routes:"
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {
"collapsed": false,
"deletable": true,
"editable": true
},
"outputs": [],
"source": [
"newRoutes = val.get_POI_routes_by_length(cityName, minRouteLength=1, maxRouteLength=5, numResults=100, apiKey=gmapsKey)"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"And score the matches using our tuned parameter values:"
]
},
{
"cell_type": "code",
"execution_count": 43,
"metadata": {
"collapsed": false,
"deletable": true,
"editable": true,
"scrolled": true
},
"outputs": [],
"source": [
"tunedDf, _, _ = val.get_route_metrics(cityName, newRoutes, sampleRates, noiseLevels,\n",
" sigmaZ=sigma, beta=beta, saveResults=False)"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"### e) Compare to performance of default params"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"If we don't specify a value for $\\sigma_z$ or $\\beta$, `get_route_metrics()` will use the default values:"
]
},
{
"cell_type": "code",
"execution_count": 44,
"metadata": {
"collapsed": false,
"deletable": true,
"editable": true,
"scrolled": true
},
"outputs": [],
"source": [
"defaultDf, _, _ = val.get_route_metrics(cityName, newRoutes, sampleRates, noiseLevels, saveResults=False)"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"Compare the before-and-after error distributions:"
]
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {
"collapsed": false,
"deletable": true,
"editable": true
},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAx0AAAI1CAYAAACzEZd8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzs3XeYk1Xax/HvDF1AukgVLBwRFQELWBCQoggIWNEVWXtv\nqy7qqqCsrG1VsBdEUF9sCLqiq+hiWRXXgoLKQREQUCwoSi8zef+4nwwhJJlkJplkZn6f68qVmeQp\nJ09Jzn1qXigUQkREREREJFPys50AERERERGp2BR0iIiIiIhIRinoEBERERGRjFLQISIiIiIiGaWg\nQ0REREREMkpBh4iIiIiIZJSCDpEscM7Ncs5lbbxq59xE51zIOdcm4rU2wWsTs5WuIB1ZPTbp4pzb\nwzn3gnNuRXBcV2U7TRVZcIxnpbB8j2CdUZlLlVQ0JfmedM6NCNYZkbmUieS+qtlOgEh5FSNjvAn4\nA1gKfAI8D7zmvS/IwL4XA3jv26R725kW/FifBrT13i/ObmoywzlXBZgG7A5MBpYBG4pZpw2wCBjt\nvR8VZFAeA3p672fFWecU4Ing337e+9dKme79gQuAw4FmwGZgCfAqcJf3fnmMdXoA/wHe8t73iPH+\n0cAzWCHXSd776cWkYVaw/0hrgW+AF4A7vPdrUvlcuSL4znjcez8i4rj92Xs/MYl1R2DXA8Db3vvo\nYxRerg3wLZAH4L3PK3XC0yji/ge40Xt/Q5zlTgMmBv/GvLZS3G8oHdupSILfkV2KWSyp61MkGQo6\nREpvdPBcBagPdABOBc4APnLOneK9XxC1znBgh7JL4nauBv4BbJeJzAHZPjbp0BbYC3jYe392Bvdz\nNhDCMphnAyUKOpxzedj1cBWwBXgdeBaoDhwMXAGc75w7zXv/XArbHQE8DKwGBnnv300hWY8Di7HP\n1hwYDIwCBjnnunnvN0Ut3x5Yl8L2y6stQHfnnPPe+xjvn4kdsy3k9m/8FuB059yNcQpmziL3P0Oy\nXgA+AH7IdkLiuBuIVxM7pywTIhVbRbiZRbLKez8q+jXnXFNgPHA8MNM5t7/3/qeIdb4ruxRuz3v/\nAzn6A5jtY5MmzYPn7zO1A+ecA7oDM4GGWGa8qff+xxJs7jos4FgMDPDefxG1r2OxGpUpzrk+3vv/\nJJG+kcBYrJbnyOhtJmFiZA1PsL3Pgc7AMCwoKeK9n5/i9surf2EB2JnAlZFvBDVsfwb+h12DLco8\ndckLf44jgZcj33DOtQcOwTLrQ8o+aenlvf8d+D3b6Ujgropa6yy5RUGHSAZ47390zp0ENAF6ANcA\nl4bfDzchiWz6EJQ2DwfOAfYA6gI/A18CE7z3T0c0yQivE9nE63Hv/YiI198CTgLGAEcBOwNneO8n\nFtfEyTm3J1by3R2oAXyKNYV4LWq5UcANxGgCFNFcKDpdYYss3wzAknBTsVjHJng9HyvNPwMr1c4L\nHxvgQe99YdTy4WNwHHAzMBDLnH8D3O69f4wUOOe6YOfxMKAesALLLN0UBHGR+w27wTkXbj4yOlaA\nWgpnBc+PAY2AccAI4JZUNhKcp+uwplSDYgUH3vvnnXNNgPuB+51ze0Uf74jt5QF3ARdj5+dI7/3S\nVNIUi/d+pXNuGnA+cABRQUe85jNBAcDNwABgR8ADd2LNxmJyzjXEMvSDgTZY08mPgFti3APVgXOx\nY98Wu19+Aj4DxnvvZ5bk8ybwBdAUOM05d433fnPEe0djwcYo7L7cTlD7NBDoxNYmdHOB+733T8RY\nfhbW3K0mdp2cEuxjGTAJGBuj1ikZTwL9sOv45aj3wtf2I8QIOpxz9bDvgqOAdsBOWKb+feBm7/0H\nUZ83fK8fHnV/bnNPOucOBP4CHAo0Bn7Fjs0j3vtnYqSjDfY92RuoA8wDRnnv/xW1XDgN2zRVCjeT\nxWpFRwMnYud2KVZDeKv3fptmvMH9dTH2O7ErsBILzq7FrrmMNbuN/BzY999I7DraMfx9XdxvT7BM\nM+BvbL1efwfeAf7uvf841X1K+aGO5CIZEmTKxgT/Dgt+LBL5O9aGeWesDfw/sVLsFliNCVhJ9Gjs\nS/r34O/wY1rU9hpiVfpdganAPUAypeBtsR/vRsCDWDObLsArzrkTk1g/kdEEP4xYlX447Xclse5k\nLNPbFMuMPIQFdfcF78VSH/gv0A14DsskNQcmBG3Gk+KcGwC8h2XWZmLnxgPnYU3o2kQsPpqtGeK3\n2PoZZyW7vyTSUx0LGv/AMhxPYRnjM5O4zqL9GSuAesF7PzfBco9gNTeO7ftcRKbrKSxT9B5wWDoC\njgjhz7Y54VJb09MoSMfpwALsOpsDPABcFmedXYCPsczNz8GyT2OB7qvOubOiVpmIXcvVsOtrHPA2\nsA9Wip8JD2PX/jFRr58FrAGmJFj3fiyQehs7HlOwdv2TnXM3JVjvGew4voR9l4Sw4Ob5ElxzYM15\nngWODjKhADjnamCFL29h5yyW9tj3ZSEWsPwTaxLYC3jHORd53OewtQnsErb9zpwVsd+zsGtlcPB8\nR7DtnbBAN9ouwIfYsZyMXSN7A9Odcz2L//hFqmHNIo8FXsHus1pYMHN9jOXvxc5bPew78P+Avtjn\nr5bCfkvjOKymajV2f0QHZHF/e5xzbbEA/nxgIXac/40FIO8F37Ul2aeUA6rpEMmsd7F2yTthP06L\nEix7DtbHYm/v/TZt051zjQGCWolwJ+OYTbsi7IP9GJ7uvd+SQpq7YzUBRU03nHP3YIHIA865V7z3\nf6SwvSJBB+k2QEdSqNJ3zg0DTsZqXLqHOxI75/6GZU5Ods697L1/KmrVjsCjwDnhduPOuTuxZjp/\nJaq0PM6+62AZy6pAD+/9OxHv/RXLHDyE/fCHP2MPLCiYlWztRnAs8iL+n8jWjrTRhmIlsQ9579cD\n651z/wpe7wW8kcw+A4cGzwlL5L33W4JS75Oxpi/RTazqYJm03ljG9MQgbWkR1LQMDv5Ntm/IWKw0\n+C7vfVGQEXE9x/I4lqEc5r2fErFOfSyTOs4592JQm1kPK9H9GDgoum9CEPQUiSyZDWoGS1pS+zSW\n8TwLC6ZxzrXASpUf896vjqhFjLa3935hVDqrYxnekc65B2INGIBl9Dt4738L1rkWuwYGAH8ifuCf\nyMNYgDECO1dgNRuNgvfi+Qpo7r3/JepztMQCgTuxwQ/w3s8B5gQ1jovjNIfdCyu8+AMLlKObF7aM\nkYYeWK3G6Ijlngr2eyXb3x/xNMcKYvqE7xfn3Ggs4LrMOXdzuDbLOXcYVtCxALveVgWvX4Pdv81J\nUIOXwKUu/uh6//DeRw+C0R/o771/Nc46iX57HgjS+Tfv/d/DLzrn7sMC4cedc7vEGCyiuH1KOaCa\nDpEM8t5vxKq/wUomi7MZ2K5TZfSPa5I2AVekGHCA1aDcGLX/j7DmEPXJThvr04PnkZE/Rt77tVjw\nANbGPdo64PLIzKD3/kus9qO9c65uEvs+BssEPR0ZcATuwGqf+jjnWifzQdIk3Dl9YsRr4b+jS+KL\nEy5lTqZGIrxM8xjvdcECDg8MTUPAMcI5N8o5N9o59zCW0WyGlXC+UNzKzrlqWFOg1ViJfJGI6zl6\nnY5YLc7zkQFHsM4qrMlSTaxUGrZ24t+IlboTtc7K6NfSISiUeAq77toEL5+ODWaRKLNOdMARvLYJ\nK0GvChwRZ9WbwgFHsM4GbECK8L5TFgws8BXb1tCdBfyGjf4Xb73fY30neu+XYUHYninej+dhn/2m\nOM0Ll8VYZwlba7LDy/0b+A44MIV9A1wceb8E/f+mY7UZkdFjuHb27+GAI1h+E1vPRUlcgl3bsR41\nYyw/vZjMf8zfniB464sdo1sj3/Pev4fV2jTECk9S3aeUA6rpEMm88I9pcXNPPAlcBHzhnHsWK8F/\nP+iEWBKLfUTn9RR84r1fHeP1WdiPXieSqCFIs85Ypm5WjPfewgK1TjHe+zpOrUw481wfy5QWt2+A\nN6PfCEr/38ZqsTphP6YZ5ZzbHStl9d77yNL6V7AmDEOcc43DmbIgUzoiejsRJb7JXp/FLeux/gwO\nGO+cOz+6PXqKYjV/m+i9/3OS6++JjYL2Tpx7aFaMfXQLnuu52PN3hAsO2gN47/9wzr2ENbub45x7\nHmubPju6tjIDHsb6kpwRlOKfAXzuvf8w0UpBZvyvWHDRGmvKEyle5/O3Yrz2DlaTW3TvOecuxe6r\nSNOCGodYHsGC917OuSVAT6wvzIYEtTU45w7BMsvdsJrk6jE+R7L3Y9fg+ZUklweYE12zFVjK1uso\nGb9777+Jsx2ABhGvhY9zrJq+D7BzURKpDl+e8Boj/m9POP3vRPVFCnsTqzXrhDVVTGWfUg4o6BDJ\nIOdcTazkBqx9eCKXYW1cT8fak48EtjjnZgB/ifPDlMiKFJcPi9fvI7y9eiXcbmnUA36N1WE1yPj/\ngmU8osVrMhD+ca6S5L4h/mhf4dejM1qZchaW+Z8Y+WJwHJ7AOsKOAG4P3mpD7E7Fo4LnH7AMejIl\nw+FmJrGOxQpsqOg3sMxwTefcGfE6nCehp/d+VlBj0R5rMjPCObfQez+mmHVh63kr7nqOFG4O1Sd4\nxFMn4u8TsUz8yWztO7DBOfccVtpbktHEiuW9/8Q59wnWJ+cDrEnYRYnWcc7timXeGmABw2tYzWYB\ndp2chgWOsWz3Obz3Bc65lWx7713K9nM/LCb+0KuTsI7+Z2K1B3kUU1vjnBuC1WhswPoyLMTmcinE\nAvLDE3yOWML3bipDiCf6bkmlFUkq31Fxr+mIc1EWivttifd+ab5LS/p7JjlEQYdIZh2K3Wc/FleS\nFJSa3Q3c7ZzbKVj3JKwTeQfnXIeguVaySlrK3DTO6zsHz5GlxuEMZazvknRmwn8HGjrnqkWXkDnn\nqmL9G0rUzyTJfcPWzx+tWdRyGRNkwEcE/451zo2Ns+hZBEFHEn0H3sVKl3uTILPnbDjWHsG//421\njPd+qXOuOxZ4jMACj1NL0MQvcpubgc+dcwOx0bBGB/13Pi1m1fD5KO56jrXOJd77cUmmbz0WwI1y\nzrXC+kSNwEps22CjnWXKQ1gb+QeA9WydKDKey7HAarsJ34J+U4kGV2hKVM1BcE00IuLe8ymOnOS9\n/8U5Fx4a9w+sdndeMavdhDXh2d97/1VUmh4kzkAHCYQz/i2AXB56OXycm2ITQBaJOBdlMfdScb8t\n8d4vzXdpaWpNJUeoT4dIhjgb4vXa4N/oDs4Jee9/8t5P9d6fgFU574aNjBJWQHKl9CXROU5fhx7B\nc2RmL9zGu1WM5fePs/1wk4RU0v8p9n3VPcZ73YNtfZLC9lIR/rw9ot8IAp5wR+xM7T/SMVipssc6\nyMd6fAu0c84lm/GaiJ2TIc65DgmWOx3ry+GJ3dQGAO/9CizTNwcLmp8NOiqXStBc6a/YdXBrMYuD\nZR7XAfsFHb6j9YjxWnio1RIFCt77pd778FCwXwOHRncmT7OnsBL+lsCzke3849g9eI7VX6K46yXW\n+4dhBQ7FBYDFeRirmWhCMbUcgd2BL2MEHPlsvR+jFRL/Oyd83o9KYt/ZFD7OsT5jV3K/ILko/cF3\nZ7TwqF9l8V0qWaCgQyQDgpqKKVjG5jus+UCi5Ws4546IHnoyKNkON8+KbCO+EmjinItuj50O9Yga\nqtE5tz/WKfd3tu3EG25n++fIH5GgxDfWcI+wtWN9Kh09JwTPY51zRbOVB3//I/j30RS2l4pp2Hj9\nw5xzXaPeuxQbHWmmL5tJDcMdyK/33p8Z68HWay2pmdC9998G61QDXgxG8tmGc24wVgtXAJxfXJOp\noD9JT2A2NuLUtKCpYWk9g4081ru4YUmDGpInsfluRkW+F3E9R6/zEdbsaKhzLmbnaOfcPsH9jXOu\niXPuoBiL1Q72uwUrkc+IoO/VkVgtwd+SWGVx8Nwj8kXnXD9iD8QQ6TrnXFH/guB8hmvaUprzJob/\nYAH1EBIP9xu2GNjDOVc0oEHw3XkDNudFLCuJXTgCNozwFuwzxrr+Y41elQ3hfg7XRgbSQVCf8Dcm\nFwQd8l/HagAvjXwvuI9Oxgqyih0oQsqnXI+KRXJeRIfTfKxJUQesJKo6lik/JYnRp2phQx4uds7N\nxto218TalbcHXowq1XsDmyDt1aAj80bgM+/9S2n4SG9jo8kchDWjaYa1W8/Hhp6NbEoxO9h/d+BD\n59ybWNX/QGzs9Vg/8m9gQ0o+HLR7XwOs8t7fEy9B3vunnHPHACdgHe2nYdXtg7F5RZ4JSpjTznu/\nJsiAPgu8FXTy/w4brakv1tb4nEzsO5Kz8e17A7+w/ZwskaZg/R+Odc5d5L3/NYnNj8IyypcDnznn\n/o1NQlcNOBg4CGu+M8x7v12H+li896ucc32wYXSPAl52zg0KRhwrEe99KOg0/QI2T8PBxaxyDdZh\n+tIg0HiXrdfzDGBQjHVOxmoXH3XOXYwFTquw2oR9sRrHbtgEgC2AD5xzX2Gls0uxCQgHYE1IxsUZ\nlCFtghGgknUf1gfk2aDT+3Ls8xyJBXSJ5uH5Crv3nsNG2TsGq4F9mZINl1skGHDgxRRWuRNrUvZp\n8Dk2Y8M474UN1zwwxjpvACcFHf8/xoKMt733b3vvv3TOnR+xzelYTVUjrMZ2NVtL4bPGe/+Wc+4h\nrEDhi4jPPhArEPqeGKOoJSHRkLmzfNTEr6V0Lva7cptzri82Z0crrBlxIdb0L6P3jGSPajpESi88\ntGC4M2l9rETqKKBbkh3Aw0O/zscyUpcE2/oDG87x+Kjlx2A/kLthQyXexNZhPEtrUZCG37AfiBOw\nDFV/7/3TMZY/BhuBpiXWkbUTcBVbh7LdRjCs5F+wH8vLgrRfkUS6hgEXYCWW5wRp+w24MHgvY7z3\n07FMzQys6cwVWDD4ANAlqC3ItDOxvhmTY3Woj0jrWizwqEHiNvqR6xR67/+CBRdPYYHzxVjmpg42\nulA77/2zqSQ4oiR+JjZ/yL+dczumso0Y25yGZRq7Bf08Ei37C3beHsM6y18K7IfdU3fGWWcZFlBe\ni9XsnIIdi4OxYPMcbJZqsBL3G7DAsycWtA3F7qGTiSrNzTbv/edYOt/D5j04DwuShmLXciInYDWO\nA7F7Lh8LVo8t5ShlKfPeP4gFTz9g1/gpWMB3EPGb5lyCDcl6IDaz+k3YNRne5sNYYdG/sJqgK7Gg\n9BdsOOFccR52na3BvgNPxu6vPti5LEnftkRD5vYodYojBN+V+2PXm8O+S4/C5jc5JPiulQoqLxRS\n3xwRERHZnrMJIQ/3ERMbSu5xzu2BTRo4xXuf0UIYkZJSTYeIiIhIOeCc2znoMB/52g7YDPWg/hCS\nw9SnQ0RERKR8uBQb1GIW1rxsZ6zfUktscsOUmkCKlCUFHSIiIiLlw+tAR2wQi4ZYh/gFwDjgrrLu\nXyOSCvXpEBERERGRjFKfDhERERERySgFHSIiIiIiklHq0yEipeKcG4bNy7E7NqfD3d77nJqfINc4\n59pgczk87r0fkeF97QHcik1o1xT43XtfP5P7FMk251wIeMt73yOD+1gM4L1vk6l9pMI5NxGbt6St\n935xdlMjsj0FHSIVUPCDG20TNtrJW8A/omY4L+l+ugFPAt8C9wPrgA9Ku11JzDn3MDZZ4Hqgufc+\n5mzCzrkq2Ozlu2OzRi8DNgTvjcIm/+qZ5hmHU+acG4FN4BdpEzbD8lvArd77L8s6Xbks4pj92Xs/\nMbupMc65PGyiweHAAUBjbDbvr4CpwAPe+3Vp2tdiyJ0Mf1nIpXtWpCQUdIhUbKMj/q6HzcY7HDjW\nOXeo935OKbd/NDZL9nDv/Xul3JYkwTlXFzgJCAG1gD8B98RZvC2wF/Cw9/7ssklhqXyGBUlg12sP\nrOT2BOdcL++9Atoc5ZyrDzyDzYz9OzADm7G9IdAPm9X+IufcAO/9F2WQpPZYIUgmHZHh7afqauAf\nwPJsJ0QkFgUdIhWY935U9GvOufHAhdh47yNKuYvmwfP3pdyOJO9krBnbncAFwFnEDzrK2/mZE3nN\nBiXnj2GBx1igZ5bSJQkEk9U9C/QG/g2c4r1fGfF+VeBGLFP8mnOus/f+x0ymyXs/P5PbD/axMNP7\nSIX3/gesNlskJynoEKl8XsOCjiax3gz6aJwN7IeVpC/CmlDd5r3fGCwzgm2bwyxyzoX/LmpP7Jzr\nAlwDHIaVXK8AXgZuCn4gI/c7Ectc7obVoJwF7AHMjmyX7ZzrB1yC1drUxZoMTQX+Hq+ZUYzP2Bxr\nntQv2F9D4BdgVpC2r6KWbxMch8eBUVhpYm8s8z8PGOW9/1eM/dTFaptOwJqaLAYeYmtpfkmcBRRi\nQccuwFDn3EHe+9lR+45sYneDc+6G4O/RWLC5S/D/fyLOHd77vIht7IAd6xOxcxEC5gLjvPf/F7W/\nHsB/gu3PwJqBdAMaUMI25t77kHPuPuy6ODBiX6U5fzcDN2EBTGOgl/d+VnCtDsdqV1oBOwBLgReB\nMd7736K2OYKgeRNWsnw90Alr8vYv4DLv/SrnXKdgf4cA1YA3gYtjHQ/nXEPgSmAw0AZrYvYRcIv3\n/rWI5WYBhwf/Puaci7wXI++/qti9PByr8aoKeOBR4D7vfWEqxyg6vRFOxu6Hb4Gh0U2ovPdbgGuc\nc7ti19IY7DoO73siW+/9wUGa22Dn9FngBu/9H8GyPbDrLLxu5HVe1EcqVp+OyOZJQDPgCqxGZBUw\nBbjae7/ROdcLO5+dgQLsfF4aGUgF21scfL42UccwkaLmcM65nsAw4FBscr9qwMLgM9/ivd8Qta+E\n92yiPh3OuROw7/2OQHXgG+Ap4J/h7/Xoz4VdM6Oxc9YUux8expo7ar4FSZmCDpHKp3fw/FH0G865\nR4HT2ZqRXwV0xTIgRzjn+gQZiDnYj9Fg7Efs7mBZws/OuQHA81jzq+eAJUAX4DzgGOfcIXEyondj\nQcrLWOa1ICJ91wf7/RXLCPwE7ItlHvo757qFMyfF6A6MxDIvzwNrsEz1ccCgIG2fxVhvF+BDLHM1\nGcvsnghMd8719t5HZoZqAG9gbds/wwK3+sB1bM0wpiTIwHYBXvfeLw0yGUOxTNrsqMVHYxm307B+\nEbOC12dh52hwkI7HsWAoel/1sQxyJ+ATYAI24mE/4CnnXAfv/d9iJLMbVqL9brBOYyzzXFLhICgy\nk1PS87cbdpwWYOejFhC+Xs4ChmDHaiZQBct0Xg4cFQR2q2NscxAwALseHwAOxoK6ts65kdg18A6W\n0d8HGAjs5pzbJyrTvwt2btoEy78K1A62/apz7hzv/cPB4hOxc3gMMB27H8PC91814CXsfHksg7kB\ny3CPBw4CTk3xGMUTDiBuL6bPxo3Y/XKqc+6iyEx14E7s3D4TfK5+WI3sYUFz0A3YtTo6eB3groj1\nk20uehFwFBb8z8Im2rsMaOicm44FIC9jBQQHY00YGwfrJLKKbZu0RroQaMS2Tb7+CuwJvBfsryYW\nnI4CegTfKeHvv7so5p6Nxzl3M3ZP/oJdB2uCz3Iz0C/4Xt8ctVo1rICqOTbT+ZZg//8I0hnvc4rE\npaBDpAILSvbCdsQywIdgGaTbo5YdgQUcL2DNI9ZHbecGrDnP3UFfkDlByV5HbCbcxRHL18EyRlWB\nHt77dyLe+yv2w/UQ9mMfrTPQyXu/TYlhUCo4Gngf6B9ZqxFR6jwayzwU502gaXQm0jnXEfhvkL5Y\nGYweWK3G6Ih1nsIyiFcSUQIL/AU73lOB48MZTOfcP4CPk0hjLOcEz+GS7VeAH4ETnXOXRQZc3vtR\nQanwacCsqKZ2s4Kg4nBgYpxS7LuwgOOv3vtbwy8652pimbVrnHPPxegX1Bc413v/YAk/Y5GgedX5\nwb+RQVVJz9+hwFjv/TUx3hsLXBCRyQtv8wzgkSAdt8RYbxBwhPf+rWD5fKyJUW8saD7be/9kxPbC\ngf1ALGMd9jgW1A7z3k+JWL4+ljEe55x70Xv/o/d+YlDSfQwwLU5H8muxTPs9WCl9QbC9Kti9d3pw\n/qZHrZfoGG0nqE3pGvw7M9Gy3vsvnXPfYxnZ/bHANNIhwH7e+yXBtq/GSv2HYvfXTcH3zKjgno/Z\nhDQJvYEu4RqxoIDgEywIGwj0jXE+j3TO7ZeoH1zwnbRdepxzo7GAYypWABN2PrAoutbAOXcT8Dcs\niH462PZdSdyz2wkG+7gaq6U40Hu/Inj9auy7fgB2bG+OWrU5VljSJ/xbEHyOBcBlzrmbYwQqIglp\nng6Riu2GiMdlWIbiK+D/YpTaXoKVZp0eGXAEbgJWAqckud9jsB/ZpyMDjsAdWCldH+dc6xjr3hod\ncAQuDp7Pim5GFWS65iSbPu/9T7FKrYPS8TeBnkFJcbQlWNOQyHX+DXxHRPOfwJ+xZlBXRZZoB59t\nXDLpjOScq401xfgdyyyEm608iZWIn5zqNhPsqxFWuvtRZMAR7HMDVkKbF2efc0oRcOznnBsVPO7E\nMoLDsSZL10akoaTn70filNB675dEBxyBCVhJf784af6/cAY12E4hVgsGMC8y4AhMCp73C78QBEuH\nA89HBhzB9lZh929N4Ng4adhGkFG+EGvOeFnk5wr+/gtWcxTrfol7jOJoiDXXAcvYFie8TPMY790d\nDjig6Fheid1Hp6eQpuKMi2yCFzQvehrLE70c43w+EfzbMdUdOeeGY021PgT+FPVd8G2cZkrh2pt4\n11wqwsdtTDjgCPa9BbsOCrGmirFcHPlb4L3/CQuU6wEuzjoicammQ6QC89u2z68NdMBKgZ8Mmsdc\nG7y3A/aD+gtwaWR74QgbsfbPyegcPL8ZI01bnHNvY81IOmEZ9kgfxtlmN2AzcLxz7vgY71cHmjjn\nGkW3vY7FOXc0cC5W4tqY7b8PG7N9p8w5cTKmS4P0hbddFxumdqmP3dl0FpaRTMVJWG3Vg1HNUh7D\nmgCdjTXvSYcDsOZFoajasrBwhj7W9RDv/CWjI1szdpux4z8ZG+J5myFzS3j+Potuvx6xvWpYTdJJ\nWFv2emxbMNciTpq3a6bI1o77sWq0wiMLtYx4LXzt1ItzvMP9r5K9/9phQf/XwN/i3M/r42wv7jGK\nI6/4RWKqvRVvAAAgAElEQVQuHyuz/Vb0C977b51zS4E2zrn60QUOJZSOc1asoHb2Eayfx8Dowpzg\nO/kSrFlfO6yPWuTxjHfNpSLRd/EC59wyrClg9LH93Xv/TYzthYPGBmlIm1QyCjpEKgnv/VrgQ+fc\nUKzPxlXOuQe890uxH5A8LHOTamY4lnrBc7yRVMKvx5qkbkWM18AyUVUpPn11sFqZuJxzF2N9R34D\nXscCn3VYRijcT6VGjFXjZXi2sG0GNfz5443QE+8zJhIe8nZi5Ive+3nOuY+BLs65/b33sTJUqWoU\nPB8QPOKpE+O1kny2sKQmSyzF+UuUtqexzN+3WGnuCizQBus/EGt7YDVP0bYk8V5kTUz4ePcJHvHE\nOt6xhLe3B4nvl3Scv5VYn53qWAf8r4tZPpxxj/XdkOh+2QW7r9IRdKTjnCXknGuPNadaCxwd1BJE\nvh8eVOBAbDCKp4GfsWAb7LzFu+ZSkcx3cWu2P7aJvuvACiVEUqKgQ6SS8TaijsdKwDpjJVfhH9pP\nvfed466cvPD2do7zfrOo5SLFGxXldyDfe9+wNAkL2qCPxjIynf32o2h1i7liasKfq2mc9+Mdl5ic\nc/uytfnW+3FKrsECk3QEHeH03+m9vzzFdTM6qk0pz1/MtDnn9scCjplYf6HNEe/lA1eVNt3FCB/v\nS7z3KTe9S7C9F7z3Q1NcN6XzF9RczsYGf+hNgqAjyIg3x4K5WDUKTbFO79HC90us74uc45zbCesY\nXhvo52NPxHoMdk9vF2g755qRnsIf2Pa7OFata6LvYpG0Up8OkcopXDWeD+C9XwN8AXRwNmxnaX0a\nPPeIfiPINB4a/PtJCtv8AGjgnOtQuqTRGKtheS9GhrUOW5sjlFjQ3+AboIVzbrcYi/RIcZPhWo5Z\n2ChIsR7rgWHBZ0hGuJlYrBLLD7G23oelmM6ykInzt3vw/GKMzrEHYiM4ZVJ40sNUjnei8zefYOS5\nOH1b0u2R4Ply51yiYxUe7WxyjH5jEGNUN2fD7LYCFkc1/ykgB0vbg8//EjYx55k+YkS7KOFr7vkY\n78Ub3S7ROY8n0Xfx7ljN06I0NVsTSUhBh0gl45wbjP0gbsaGagz7J9ZEYkIwSkr0eg2cc8lm6KZh\nw9oOc851jXrvUmBXYKb3Pro/RyJ3Bs8PO5unITp9tWPsK5afsKY4XSIz6EHm7G4sU5sOj2HfsbcE\npeXh/bRla6f4YgWZmFOwDMcp3vszYz2wzEsdrLN5MsJN0LbrzB80BXkS2N85d10QKEana7fgs5S1\nTJy/xcFzj8gXgxLre0uUyhQETeLeweZcidlh2jm3T5CesETnbws2LG4zbNSr7QIB51wz59xepU68\neRIbuW134Dnn3Dbt/Z1zVZxzN2IDD/yADRsdyyXOhg4Or5cP3IbdR49FLbsS68OV6YAwaUF6n8AC\n1dHe+0kJFl8cPPeI2sauxB4lDRKc8wQmBM9/c84Vzc0UjGJ2O3ZsH01heyIlpuZVIhVYVKfU2lgH\n2fBQotf4iFmBvfcTnE2Qdj6w0DkXHpWpIRakdMd++M8tbr/e+zVB5ulZ4C3n3LPBtrpgQ6quYOvw\nr0nx3r/hbN6DscDXzrkZWAfNOlh778OxITiPLGY7hc65cdg8D3OdjctfHZu/oCGWeUrHzNd3YP0L\njgU+CY5nPWyegrexoVaTcSJWsv+S9z7RzOKPYCNOnY1N4FWc/2C1GWOdc3tj/SPw3odH57oQ6xNw\nIzavwrtYm/vmWAfkA7AAp7jJ0NIqQ+fvf9hQu0Odc+9h11FT7F7xlM2M7idjbfwfDfqszMZqK1pi\nc9HsjXU4D/cNeB8Lvi4NaifD9/J47/3v2IhzHbH7daBz7k2sQ/RO2Hk9BBsRbJsO+iXhvS9wzh2L\nDQfbH/jWOfcyNtpbQ2wUprZYRntg5ChKUf6LDcX9NNbcp1/wGT4Gbo1aNjwHzqvBwBQbsU7wL5X2\n85TCcdjwvithu+/fsGnBsLsvYbWhlzvn9sFqJFpjQ9i+TOzAorh7djve+/ecc7diTQTnOeeew/qZ\nHIVdU+9igZ1IxqmmQ6Riix4ytzP2Y9fXe3979MLe+wuwcerfx9pnX45ljuthP0x3Ra8Tj7fx/w/B\n5irox9bZfx/Axsj/NtUP472/BQt+Xg62fSlwPDbKy0Nsbb5RnOuw4SLXY8HPUKwvxIFsP5pWiQQj\nAPXGamiaYKPU9MCG3E1mLpGw8MRrjyRaKBjmcwFWO9EpifR9hc3hsQILNG8KHuH3/8ACuYuwUc2O\nxa6HnsDq4DO8nsLnSKe0nr9gRLJBwP1YUHUx1gTwEezazfh8BN77ZVhQfi1BrVaQjoOxz3QONht8\nePnfsHPyJTY8c/j8NQje34wFvcOxwGkAdsyOxH77r8NqKNKV/t+w6/1ELCN7BDbc7YlY7cYVQAfv\n/bwEm7kMuz96YPdLE6z2qpfffiLBMdh3yW7YPBQ3keSQwhm0Q/DciG2/eyMf+0HRwB69sMn6OmDn\nel/sc/wp1saLu2fj8d7/FSsg+Bq7Hi7GroG/YfNwlGbyTpGk5YVCmsleREREssM5NxHLTLf1EZOM\nikjFopoOERERERHJKAUdIiIiIiKSUQo6REREREQko9SnQ0REREREMkpD5lYgeXl5iiBFREREpEyE\nQqG8ZJdV0FHBqOZKRERERDItLy/peANQnw4REREREckwBR0iIiIiIpJRCjpERERERCSjFHSIiIiI\niEhGKegQEREREZGMUtAhIiIiIiIZpaBDREREREQySkGHiIiIiIhklIIOEREREamwJk6cyPr164v+\nP+uss/jjjz+SXv+NN97goYceKnU6Tj31VObOnVvq7ZRXCjpEREREpMKaNGnSNkHHww8/zI477pj0\n+kcccQRnn312JpJWalu2bMl2EpJWNdsJEBERERFJ1mOPPcbzzz8PwHHHHceIESNYtmwZZ555Jh07\nduTLL7+kbdu23HLLLTz77LP89NNPnHbaadSvX5/JkyfTq1cvnnvuOdatW8eZZ55Jly5d+Oyzz3DO\nceyxxzJu3Dh+/fVXbr/9dvbdd1+mTp3KvHnzuP766znmmGOK0rFo0SIeeeQR9t57b2666SYWLFhA\nQUEBF154Ib1792bDhg1cffXVfPPNN+y2225s2LAh5ufp1asXRx11FLNnzwbgjjvuYJddduHNN9/k\n/vvvZ/PmzdSvX5/bb7+dxo0bM378eH766SeWL19OgwYNuOyyy7jqqquKAqvrrruOzp07M3v2bMaP\nH0+jRo2YP38+ffr0oV27dkyaNImNGzdy77330rp1a1555RXuvfde8vPzqVu3Lk8++WRmTlwoFNKj\ngjzsdIqIiIhk3hVXhEK77JLexxVXJN7n3LlzQwMGDAitXbs2tGbNmlD//v1DX3zxRWjp0qWhdu3a\nhT766KNQKBQKjRw5MvTII4+EQqFQqGfPnqGVK1cWbSP8/9KlS0Pt27cPzZ8/P1RQUBAaMmRIaOTI\nkaHCwsLQ66+/HjrvvPNCoVAo9Pzzz4dGjx69TTreeOON0LBhw0KbNm0K3XHHHaFp06aFQqFQ6Pff\nfw/17ds3tHbt2tCECRNCI0eODIVCodBXX30Vat++fejzzz/f7jP17NkzdN9994VCoVDohRdeCJ19\n9tmhUCgUWrVqVaiwsDAUCoVCzzzzTGjs2LGhUCgUGjduXGjIkCGh9evXh0KhUGjdunWhDRs2hEKh\nUGjRokWhIUOGhEKhUOiDDz4IdenSJfTjjz+GNm7cGDr00ENDd999dygUCoUmTpwYGjNmTCgUCoUG\nDBgQWrFiRVH6kxXkO5POp6qmQ0RERETKhY8//pjevXuzww47ANCnTx8++ugjevXqRbNmzejSpQsA\ngwYNYvLkyZxxxhkJt9eyZUuccwDsvvvudOvWjby8PJxzLF++POY6ixcv5tZbb+Xxxx+nWrVqvPvu\nu7z55ptMmDABgI0bN/LDDz/wv//9j1NPPRWAPffcs2g/sQwYMACAo48+mrFjxwKwYsUKLrvsMn7+\n+Wc2bdpEy5Yti5bv1asXNWvWBKyJ1Y033sj8+fPJz89n8eLFRcvts88+7LTTTgC0bt2aQw45BIB2\n7doV1ax06tSJkSNHctRRR9GnT5+Ex6s0FHSIiIiISMpuu80eZckK2GPLy8tL+H8s1atXL/o7Pz+/\n6P+8vDwKCgq2W37dunVceumljBkzhqZNmxa9Pm7cOHbddddi05SKMWPGMGLECI444ghmz57NPffc\nU/RerVq1iv6eOHEijRs3Zvr06RQWFrLvvvsW+/ny8/OLPt+NN97IZ599xqxZsxg8eDDTpk2jQYMG\nJU53POpILiIiIiLlwgEHHMDMmTNZv34969atY+bMmey///4AfP/993z66acAvPzyy0W1HrVr12bt\n2rVp2f/VV1/N0KFDi/YJcOihh/LEE08UBURffvllUVpfeuklABYsWID3Pu52X3nlFQBmzJhBp06d\nAFi9enVRYDNt2rS4665evZomTZqQn5/P9OnTYwZLiXz33Xd07NiRSy65hAYNGrBixYqU1k+WajpE\nREREpFzo0KEDQ4cO5fjjjwesI/lee+3FsmXL2G233XjhhRe4/vrradOmDcOGDQPghBNO4KyzzqJJ\nkyZMnjy5xPtevnw5//73v1m8eHFRR/YxY8Zw/vnnc/PNNzNo0CBCoRAtWrTgwQcfZNiwYVx99dUM\nHDiQ9u3bb1MDEW3Tpk0cf/zxFBYW8s9//hOACy+8kEsuuYSmTZvSsWNHli1bFnPdk08+mYsuuohX\nX32Vgw46qKjpWbJuvfVWlixZQigUomvXruy5554prZ+svETVVFK+5OXlhXQ+RUREpLJZtmwZ5557\nLv/617+ynZSUhUfTatiwYbaTkpK8vDxCoVDS7cfUvEpERERERDJKNR0ViGo6pKy1aWPPEQNl5Jac\nT6CIiEj5pJoOERERERHJKQo6REREREQkoxR0iIiIiIhIRinoEBERERGRjFLQISIiIiLl0vjx43n0\n0UcTLvPrr79y/PHHM3jwYD766KOU9zF16lRuvPFGAGbOnMk333xTorRWdpocUERK7Nprs52CYuR8\nAkVEJNPef/99dt11V2655ZZSb2vmzJn06NGD3XffPQ0p26qgoIAqVaqkdZu5RkPmViAaMldEREQq\nuvvvv59p06bRrFkzGjZsSIcOHTjjjDP47rvvGD16NL/99hs1a9bkpptuYtOmTZx33nls2LCBpk2b\n8vTTTzN27Fjmzp3Lxo0b6devHxdffDGw7SR9c+fO5dZbb2Xy5MlMnTqVefPmMWDAAM4991zq1KlD\n3bp1GT9+PK1bty5K18iRI6levTrffPMNK1euZOTIkfTs2ZNly5Zx1VVXsX79egCuu+46OnfuzOzZ\ns7nnnnvYaaed+Oqrr5gxYwbnn38+K1asYOPGjQwfPpwTTzwRgE6dOnHyySfz/vvvs+OOO3L55Zdz\n22238f3333PNNddwxBFH8PXXX3P11VezefNmCgsLGT9+PG3CQ8dnQKpD5qqmQ0RERERSd+WV8Oyz\n6d3m8cfDbbfFfXvevHnMmDGDadOmUVBQwJAhQ+jQoQNgmfnRo0fTpk0bPvvsM0aPHs2kSZO4+OKL\nmTdvHtdffz0Al112GfXr16egoIARI0Ywf/589txzz2KT1rlzZ3r16kWPHj048sgjYy6zfPlynnji\nCb777juGDx/OwQcfTKNGjXjssceoUaMGixcv5vLLL2fq1KkAzJ07l5deeolWrVoBcPPNN1O/fn02\nbNjAcccdR9++fWnQoAHr1q3jwAMP5Morr+SCCy7grrvuYsKECSxcuJC//vWvHHHEEUyZMoXhw4cz\naNAgNm3aRGFhYUqHPtMUdIhIiZ10kj1PmZLddMSV8wkUEZFUfPTRR/Tu3ZtatWoBVjsBsHbtWj79\n9FMuueSSomU3bdoUcxuvvPIKzzzzDFu2bOHnn39m4cKFSQUdyTjqqKPIz8+nTZs2tGrVim+//ZaW\nLVty4403Mn/+fPLz81kcMWHtPvvsUxRwAEyePJnXX38dgB9++IElS5bQoEEDqlWrRvfu3QFo164d\n1atXp1q1arRr147ly5cDsN9++/HAAw+wYsUK+vbtm9FajpJQ0CEiJfbBB9lOQTFyPoEiIuXYbbcl\nrJXIlLy87Vv0hEIhdtxxR6ZPn55w3aVLlzJhwgSee+456tWrx8iRI9m4cSMAVapUIdxMPfxaadOW\nl5fHxIkTady4MdOnT6ewsJB999236P0ddtih6O/Zs2fz3nvv8fTTT1OrVi1OPfXUonRUq1ataNv5\n+flUr1696O+CggIABg4cSMeOHZk1axZnnHEGY8aMoVu3biX6HJmg0atEREREpFw44IADeP3119mw\nYQNr1qzhP//5DwB16tShZcuWvPLKK4AFIfPnz99u/bVr11KrVi3q1q3LL7/8wttvv130XosWLZg3\nbx4Ar732Wsz9165dm7Vr18ZN36uvvkphYSHfffcdS5cupW3btqxevZomTZqQn5/P9OnTi4KEaKtX\nr6ZevXrUqlWLhQsXMmfOnOQOSmDp0qW0atWK4cOH06tXL7z3Ka2faarpEBEREZFyoUOHDvTv359j\njjmGFi1a0KVLl6L3brvtNkaNGsX999/Pli1b6N+//3bNpvbcc0/22msvjj76aFq1akXnzp2L3rvw\nwgu59tprefDBB+nYsWPM/ffv35/rrruOyZMnM27cuG06kgO0bduWP/3pT6xcuZLRo0dTo0YNTj75\nZC666CJeffVVDjrooG1qNyJ1796dKVOmMHDgQNq2bct+++2X0rGZMWMGL774IlWrVqVx48ZccMEF\nKa2faRq9qgLR6FVS1sLNRSOap+aWnE+giIhUFCNHjkzYybyiSXX0KjWvEhERERGRjFLzKhEpscMO\ny3YKipHzCRQRkYriH//4R7aTkNPUvKoCUfMqERERESkLal4lUg498wyccQa89x4obhQREZGKRjUd\nFYhqOsqnmTPhyCMhPILe/vvDpZfapKzBMNw569577TnHBsjYKucTKCIiUj6lWtOhoKMCUdBR/syf\nD127wrp1cOGF8NFH8O67VtvRrBn8/e/w5z9nO5Xx5fzgUDmfQBERkfJJQUclpqCjfFm5Eg46CBYu\nhGuugT597PXvv4cXXoAZM2DDBgtEOnXKblrjyfk8fc4nUEREpHxSnw6RcmDTJhg61AKOP/1pa8AB\n0Ly5tQYaPRoKC+Hss7c2vRIREREpjxR0iJSxUAjOPRfefhsOPzx+86n994feva2m4777yjaNIiIi\nIumkoEOkjD3xBDz2GDgHI0dCfoK78PzzoW5duPZaWLas7NIoIiIikk4KOkTK2OOP2/MNN0DNmomX\nbdAAzjkHVq+GSy7JfNpEREREMkEdySsQdSTPfStXQtOm0K5d8k2mCgttCN25c2H6dBg0KLNpFBER\nESmOOpKL5LAXX7RO4Ycdlvw6+flw+eVQtap1MF+9OnPpExEREckEBR0iZWjqVHvu3j219dq0gWHD\nrF/HmDFpT1aJzZ1rj5yV8wkUERGpHNS8qgJR86rc9scf0KQJtGwJjz6a+vobN8Ipp9hwu8uWwY47\npj+Nqcr5aTByPoEiIiLlk5pXieSoGTMsYEi1liOsRg0YPNiaV02YkN60iYiIiGSSgg6RMhJuWpVK\nf45oAwdC9epw992aMFBERETKDwUdImVg/Xqr6WjRAtq2Lfl26tWDfv2stdC0aWlLnoiIiEhGKegQ\nKQOvvQZr11rTqrykWz/Gdtxx9nznnaVPl4iIiEhZUNAhUgbS0bQqrHVrOOgg+O9/4X//K/32RERE\nRDKtarYTIFLRbd5s83M0aQLOpWebxx0Hs2dbbcdTT6VnmyUxaVL29p2UnE+giIhI5aAhcysQDZmb\nm15/Hfr2hSFD4OKL07PNUAjOOAOWLoVvv4VWrdKzXREREZFkaMhckRzz/PP2XNKhcmPJy7Paji1b\n4J570rddERERkUxQTUeGOedaAZOAnYFC4CHv/d1Ry/QApgOLgpemeu9vTHVfqunIPQUFNmLVxo3w\n3HNQpUr6tr1pE5x0ktV6LF0Kdeqkb9vJ2m8/e54zp+z3nZScT6CIiEj5pJqO3LMF+Iv3vj3QFbjA\nObdXjOXe8d7vFzxSDjgkN334Ifz4Ixx8cHoDDrD5OgYNglWrYPLk9G47WatW2SNn5XwCRUREKgcF\nHRnmvf/Be/9J8Pdq4CugRXZTJWXlww/tuVOnzGx/0CALZh54wGo8RERERHKRRq8qQ865NkAnYHaM\nt7s55z4Dvgeu8N5/Ucy2RgE3pDuNkl4ff2zP7dplZvsNG8Ihh8Dbb9toVl27ZmY/IiIiIqWhoKOM\nOOfqAM8Dl3rv/4h6+xNgF+/9Gudcf2AasEei7XnvRwGjIl/Ly8tTWXeO+eQTqFULWrbM3D4GDbKg\n44EHFHSIiIhIblLzqjLgnKuGBRxPeu+nRr/vvf/De78m+HsGUM0517iMkylptm4dfPUV7L475Gfw\nTuvUCZo3h6efht9+y9x+REREREpKQUeGOefygEeBr7z3/4yzzM7BcjjnDsTOy8qyS6VkwmefQWFh\n5ppWheXnw4ABsGFD2XcoP/NMe+SsnE+giIhI5aAhczPMOXco8A4wFxsyF+AaoDWA9/4B59yFwHnY\nSFfrgcu99++lui8NmZtb7r0XLrwQRo6Efv0yu6/ffoMTTrAAZ948m8dDREREJFNSHTJXfToyzHv/\nLpDwhHjv7wE0xVsF88kn9pzpmg6ABg1s8sE334T//hcOPTTz+xQRERFJlppXiWTIxx9DjRrQunXZ\n7G/gQHt+4IGy2R/AxRfbI2flfAJFREQqBzWvqkDUvCp3bNgAdetaLce995bNPkMhOO00+OknWL4c\nGjXK/D7btLHnxYszv68SyfkEioiIlE+akVwkB8ybB1u2wB4JBz5Or7w861C+cSM8/njZ7VdERESk\nOAo6RDKgLPtzROrXD6pXhwcf1AzlIiIikjsUdIhkQKZnIo+nXj04/HBYsABmzSrbfYuIiIjEo6BD\nJAM++QSqVYNddin7fYc7lD/4YNnvW0RERCQWDZkrkmabN8Pnn8Ouu1rgUdb23tv6T0+dap3Kd9op\nc/tyLnPbToucT6CIiEjloJoOkTT74gvYtKlsO5FHCnco37wZJk7M7L7+/W975KycT6CIiEjloKBD\nJM3CncizFXQA9O1rc4Q89BAUFmYvHSIiIiKgoEMk7cJBRzZb9tStCz17wsKF8MYbmdvPc8/ZI2fl\nfAJFREQqB00OWIFocsDc0K0b/O9/MGOGDV+bLV9+CRdcAMcem7l8d87PvZfzCRQRESmfNDmgSBZt\n2QKffWZ53WwGHADt28Nuu8H06bBiRXbTIiIiIpWbgg6RNPIe1q8v+/k5YsnLs+Fzt2yBCROynRoR\nERGpzBR0iKRRLnQij9S7N9SqZR3KCwqynRoRERGprBR0iKRRtmYij6d2bejVC5Ysgddey3ZqRERE\npLJS0CGSRp98Avn51pciV2iGchEREck2zUgukiaFhfDpp9C6NdSsme3UbOWc1by89BIsXQqtWqVv\n27Nnp29bGZHzCRQREakcVNMhkibffQdr1uRWLUfYMcdYUHT//endbtOm9shZOZ9AERGRykFBh0ia\nLFhgz+msSUiXI46AevXg4Ydhw4b0bffHH+2Rs3I+gSIiIpWDgg6RNPn6a3tu0SK76YilRg3o3x9+\n+QWefjp92z3oIHvkrJxPoIiISOWgoEMkTXK5pgNg0CDr5D5+PGjiehERESlLCjpE0iSXazoAdt4Z\nDjnEhvX94INsp0ZEREQqEwUdImny9dfQoAHUqZPtlMQ3eLA9jx+f3XSIiIhI5aKgQyQNNm+GRYty\nt5YjrFMnaNMGnn0Wfvgh26kRERGRykJBh0gaLFoEBQXQsmW2U5JYXh4MGQJbtsBDD2U7NSIiIlJZ\naHJAkTQIdyLP9aADoE8fGzr3gQfg6quhevWSb+v229OXrozI+QSKiIhUDgo6RNIg3Im8PAQdtWrB\nkUfCc8/B88/DsGEl39Zxx6UvXRmR8wkUERGpHNS8SiQNylNNB1iH8rw8dSgXERGRsqGgQyQNcn24\n3GgtWsCBB8L779sQuiXVr589clbOJ1BERKRyUNAhkgYLFkCTJlCzZrZTkryhQ+25NLUd3tsjZ+V8\nAkVERCoHBR0ipbR+PSxdWn6aVoXtv7+lecoU+PnnbKdGREREKjIFHSKltHChPZeXplVh+fnWt2Pj\nRnjkkWynRkRERCoyBR0ipRTuRN6qVXbTURJHHmmjWd1/v83dISIiIpIJCjpESqm8dSKPVLu29bNe\nuhSmT892akRERKSiUtAhUkrluaYDrIkVlKxD+aBB9shZOZ9AERGRyiEvFAplOw2SJnl5eSGdz7LX\nvTv897/w6qtQrVq2U1MyV1xhQ+d+/jnss0+2UyMiIiK5Li8vj1AolJfs8qrpECmlBQugadPyG3AA\nDBliz/fck910iIiISMWkoEOkFP74A378sfwNlxuta1fYeWd44gn47bfk1xszxh45K+cTKCIiUjko\n6BAphW++sefyHnRUqWJ9O9atgwkTkl/vkUdyfLjdnE+giIhI5aCgQ6QUwp3Iy3vQAXDUUVCjBtx7\nLxQUZDs1IiIiUpEo6BAphfBwuRUh6NhxR+jdGxYtghkzsp0aERERqUgUdIiUQrimozzO0RGLOpSL\niIhIJijoECmFr7+GqlWtE3ZFsNtusO++8Npr4H22UyMiIiIVhYIOkVJYsACaNbOO2BXF0KH2nExt\nR/369shZOZ9AERGRykGTA1YgmhywbK1cCY0bw8EHw9//nu3UpE9BAQwbBuvXw/Ll1tdDREREJJIm\nBxQpI+FO5BWlP0dYlSowaBCsWQOPP57t1IiIiEhFoKBDpIQq0nC50Y4+2mZYv+ceKCyMv9zbb9sj\nZ6pzpVUAACAASURBVOV8AkVERCqHqtlOgEh5VZGGy43WoAH07GkdymfOhL59Yy83fLg9L15cZklL\nTc4nUEREpHJQTYdICVXkmg7Y2qF8/PjspkNERETKPwUdIiX09dc2g3fjxtlOSWY4B3vtBS+/DN9+\nm+3UiIiISHmmoEOkhBYtsvk58ivwXTR4MIRC8Mgj2U6JiIiIlGcVOLskkjmrVtmjokwKGE/37lC7\nNjzxROIO5SIiIiKJKOgQKYElS+y5ogcdNWpAjx6wdCm89Va2UyMiIiLllUavEimBcNDRtGl201EW\n+va1fh2TJtmIVpFeeik7aUpazidQRESkclBNh0gJhEdgreg1HQB77w3NmsFzz8Hatdu+t88+9shZ\nOZ9AERGRykFBh0gJVJbmVWAd5fv0sRnKX3gh26kRERGR8khBh0gJhGs6KkPzKtg6OeCkSdu+3qaN\nPXJWzidQRESkclDQIVICixdD9eo2c3dl0KIFdOhgs5MvX57t1IiIiEh5o47kGeacawVMAnYGCoGH\nvPd3Ry2TB9wN9AfWASO895+UdVoleUuWWC1HXl62U1J2+vaFL76AJ5+Eq67KdmpERESkPFFNR+Zt\nAf7ivW8PdAUucM7tFbXMUcAeweNs4P6yTaKkYs0aWLmycvTniNSzJ1SrBo8/bhMGioiIiCRLQUeG\nee9/CNdaeO9XA18BLaIWOwaY5L0Pee8/AOo755qVcVIlSZVpuNxIdevCwQfDl1/Cp59mOzUiIiJS\nnijoKEPOuTZAJ2B21FstgKUR/y9j+8BEckRlGi43WrwO5SIiIiKJqE9HGXHO1QGeBy713v8R9Xas\nngEJG7A450YBN6QndZKKyhx0HHgg1KsHTz0Ft90GV16Z7RQVI+cTKCIiUjko6CgDzrlqWMDxpPd+\naoxFlgGtIv5vCXyfaJve+1HAqMjX8vLy1NK+DFTW5lUAVatCr142X8cbb8AFF2Q7RcXI+QSKiIhU\nDgo6MiwYmepR4Cvv/T/jLPYicKFzbgpwEPC79/6HskqjpCZbNR07LPqCBnP+Q51v5lDnmznUWraA\nTQ13Zn2L3VnffHfWtOvMTz1OoLDmDhlNx+GHW9Dxwgtw5JEZ3ZWIiIhUEAo6Mu8Q4FRgrnNuTvDa\nNUBrAO/9A8AMbLjcb7Ahc/+chXRKkpYssVGcGjYsm/1V/3k5uz56LU1fm0ReMGxUYdXqbGzcnGqr\nfmaHZV8XLbvbfX/h+4HnsHzwBWxqkpluQXvvDfXrw/TpNpJXfj5MnpyRXZXeqafac84mUEREpHLI\nC2nsywojLy8vpPOZeU2bWtDxxBOZ3U/+xvW0mnIrrf/vVqpsXMf65rvxU4/jWd+yHRt2ag1VqgBQ\nZf0aqv+8nHrz/kuj916k2trfKaxSle8HX8C3Z42lsEattKfttttgxgw7FjVrbq39yTnh2chzNoEi\nIiLlU15eHqFQKOkZyxR0VCAKOjJv/XrYYQfo0gVuvz1z+8lfv5Z9R/an/udvs7luQ37ofzq/Hngk\n5FdJuF7epo00+GQmO705hZo/L2Ntmw58+benWLvbvmlN3wcfwNVXw4472qzsOZunV9AhIiKSEakG\nHRoyVyQFZdGJPH/9Wva5ZgD1P3+bVft256trJvNr16OLDTgAQtVr8GvXo/FXPMzPhw6m9uIv6HLe\nAbR89s60zujXubMFX+vWaaJAERERKZ6CDpEUZLoTef6Gdexz7UAazJnFqn27s3j4dSXqGB6qXpPl\nx17Ct2feTEHN2ux+3+XscfeFUFiYlnRWrw4HHQRbtsDmzWnZpIiIiFRgCjpEUpDJmo68zZvY59pB\nNPj0P6za5zAWD78OqpRurIc/OnTDX/EI65vvRovp9+FuPwsKCtKS3kMPted169KyOREREanAFHSI\npCCTNR2tnxpLg0/e4Pe9D2ZJGgKOsC07NuSb8//JulaOZq9MoP3Y4eQVbCn1drt2hbykW3JmSdeu\n9hAREZGsUtAhkoJwTUe6g47a385llyf+zqZ6TVhy8tWEqlZL6/YLau/IN+fdzpo2HWj6xlO0H3NK\nqZta7bCDNbH6/XdYuDBNCU23KVPsISIiIlmloEMkBYsX20i1jRqlb5t5BVtwt55O/pbNLDv+Mgpr\n1UnfxiMU1qrDt+fexppd92WnWc/QZuKoUm8z3MTqhRdKvSkRERGpwBR0iKRg8WLYaaeiKTLSouUz\n/2RH/xG/dunDHx26pW/DMRTWqMWi029kY6PmtJl8E03efLpU21u71ppYTZ2apgSm28MP20NERESy\nSvN0VCCapyOzNm60ifD22w/uvDM926z1neeAMztSUGMH5o98jILa9dKz4WLU/GERe4yz0azmjHuH\n1W7/Em3npJPg119tBKvvv4dmzdKc0NLSPB0iIiIZoXk6RDLku+/sOZ39OfYYdyH5mzey7NhLyizg\nANjQrC1L/vQ38jdvZO9rj6H6yh9KvK2aNe15+vQ0JU5EREQqHAUdIklK93C5db/6kIYfz2T1Hp35\nf/buOzyqauvj+Hcy6T0hkISa0IaOdMGCiCJeFBtiV1DBchV7vfbeO1wLV7DXq9f6IoqVoiC9Dh0S\naiCQ3jPvH5sISEICzMyZ8vs8z3mGzJw5syCazJq911p5Rw10z0UPQX7n/mw+bSwROzfT4YnLDruw\nPCLC3PrsFisRERGxnJIOkQZyd7vclu8/DsC2ky9yzwUPQ86g88jr2I/kP7+n2f/GH9Y17HZo1w5+\n+gny890coIiIiAQEJR0iDeTOpCN6w3IaT/8fRS07UNi2x5Ff8HDZbGSdfzuVMQm0fu12ojcsP6zL\nHH20mU7+449ujk9EREQCgpIOkQZy5/aqlh88CcD2wRdaPmGvMj6ZrJE3Yy8vpeOjF2OrKD/ka/Tp\nY26/+87NwYmIiEhAcM/IY5EgsH49hIRA48ZHdp2IbRtp8sN7lDZpSV6XY9wS25HK63Y8O/sOpdHs\nKbR652HWX/5wg573zjvmNiQEYmNhyhRwuSzPo/ZaudLqCERERAStdIg02IYNkJICoUeYqrf4+FlC\nqirZPvgC827dR2w66zrKktJo9d5jxK5e0KDnhIWZw26HHj1MYrZqlWfjPCTh4eYQERERS/nOOx4R\nH1ZeDps2HXk9R1jeDtK/eYPyxCbs6jnYPcG5SXVkDNkjb8ZWXU27F65tUDerjRv3thLu29fc+tQW\nqxUrzCEiIiKWUtIh0gDZ2eY9+JHWc6R9+yb2shJyTjgXV2iYe4Jzo4IOfdjdfSAJS2eRNmVyveff\nfrs5wEfrOoYONYeIiIhYSkmHSAPUFJEf6UpH6g/vUm0PJbf3kCMPykM2nflPqsKjaP36HYTm5zb4\neamp0LKlaZ1bVubBAEVERMTvKOkQaYCadrlHstIRs2YRsWsXk9/paKpi4t0SlydUJDZm69DLCM/b\nQeuJdx/Sc/v2heJimD7dQ8GJiIiIX1LSIdIA7ljpSP3hPQB29TrJDRF5Vs7x51CSlkH6168Tt2JO\ng5/nk1usRERExHJKOkQa4IhXOqqrSf3hPSqjYsnv1N9dYXmOPZTsETdic7lo99L1pg9uA3TrZppF\nKekQERGRfSnpEGmAmg5NTZoc3vMTF/5CxI5N5HUfiCvMP1q4FrXpzu7uA4lf/gcpv37WoOdERprE\nY9Ei2LzZwwGKiIiI31DSIdIAWVmQnHz4Ix9Sv38XgFw/2Fq1r83DrsQVYqf1xLuwVVYc8Phtt5lj\nXzVbrKZO9UKA9Zk40RwiIiJiKSUdIvVwuUzScbiTyEPKS2n8y6eUJzahqHU39wbnYeWNm7Oj/2lE\nZ68i/ZsD37z36mWOfflUXcdJJ5lDRERELKWkQ6QeOTmmBezhbq1qNOtrQovzTQG5D00gb6htp1xK\nVUQUGZMfwF5cUO/5GRkmQZs6FaqqPB+fiIiI+D7/ewck4mVZWeb2cJOOmq1V/tC1qjaVcclsP+E8\nwndvp8XHz+732DXXmGNfNptZ7cjNhblzvRhobfr23TsqXURERCyjpEOkHkdSRG4vLiD5j28padqG\n0vRM9wbmRTmDRlIRl0yLj54hLHfbX/fv2mWOv/OZLVbbt5tDRERELKWkQ6QeR7LSkTT3B0IqK8jr\ncox7g/Ky6ogotg65BHtpES0/fKre83v1MjvJLE86RERExCco6RCpx5GsdCT/8X8A5Hfs58aIrJF7\n9D8oT2xC0y//TXju1oOeGxcH7dvDH39AYaGXAhQRERGfpaRDpB41Kx2H3L3K5aLRH99SGZNAcUuH\n2+PyNldoONtOuhB7WQktPny63vN79IDKSpgxwwvBiYiIiE9T0iFSj6wssNvNnI5DEbNuCRE7NpHv\n6A0hds8E52W5/U6lPLFxg1Y7jjrK3P70kxcCExEREZ8WanUAIr5u40ZISTGJx6FI/uNbAAoCYGtV\nDbPacREtPn2BFh8+zSmnPFvnuV27mn8zS5OOyy6z8MVFRESkhpIOkYOoqIAtW6BLl0N/bqM//g+X\nzUZ+hz7uD8xCuf1OJfWH92j65b8Z+/7tVCSn1npeVBR06AB//gl5eZCQ4OVAAR580IIXFRERkb/T\n9iqRg9i8GaqrD72ew16YR8Li6RS3cFAVm+iZ4CxSs9phLyuh5UcHr+3o2dP8+/32m5eCExEREZ+k\npEPkIA63XW7S3B+wVVcF1NaqfdXUdjT57wTefWFHnedZXtdx663mEBEREUsp6RA5iMNNOhrtqecI\nhFa5tXGFhrP9hJFEVJXQ4YdX6jyvc2cIC7Mw6fj0U3OIiIiIpZR0iBzEYc3ocLlInv1/VMQkUNzC\n/1vl1iX36GHk2pK5vPhl7CW1D+OIiIBOnWDBAsjN9XKAIiIi4jOUdIgcxOGsdMSuWUjEzi0UdOhr\nxnIHqOqIKN6MuJZkVy7p30ys87wePcDlgl9/9WJwIiIi4lMC9x2RiBsczkrH3inkfT0QkW+ZFHE1\nRcTQ/ONnsVWU13pOjx7mVvM6REREgpeSDpGDyMqCyEiIi2v4cxLn/whAgaO3h6LyHbtCGvFuxOVE\n5mSTOu39Ws/p0MFss/rxRy8HJyIiIj5DSYfIQWRlmVUOm61h59uqKklYOovS1FYB1yq3Ns0alfBl\n+liq7aG0+OBJ0x/3b8LDzZyTJUsgJ8fLAWZmmkNEREQspaRDpA7FxbBz5yHWc6yaj720iMLWXT0X\nmA95458LefSGHezqdRIxG1eQMuOLWs+raZ3788/eiw0we7q0r0tERMRySjpE6lBTRH4ogwETFplq\n6aI23TwQke/afuIFALT46JlaH1ddh4iISHBT0iFSh5oi8tTUhj8ncZEZvV3YOjiSjp8WN+KnxY0o\nS21JXucBJCydSfySmQec53BAVJQFSccXX5hDRERELKWkQ6QOh7zSUV1NwuLplCelUpF0CJmKH3v6\n83Y8/Xk7ALYPOg+AFh8fuNoRGgrdusGKFbBlixcDvOEGc4iIiIillHSI1KEm6WjoSkf0huWE5e8M\nmlWOvytq3ZWilh1Imf4/orJXHfC4ZXUdIiIiYjklHSJ1qNle1dCVjsTFZmtVUZAUkR/AZmP7oPOw\nuVw0/+S5Ax5WXYeIiEjwUtIhUodDnUaeUFPPEWRF5PvK63YcZY3SSZsymbDd+/fHbdsWoqPhl18s\nCk5EREQso6RDpA4bN0JCghkOWC+Xi4RFv1IRk0BZk5Yej81nhdjJGXgu9vJSmn4xYb+H7Hbo2hVW\nrvRyXYeIiIhYTkmHSC1cLrPS0dCtVZHbNhCZk01R624NnyQYoHL7DqUyOo5mn79CSGnxfo91725u\nf/3VgsBERETEMko6RGqxa5cZDnioW6uCrZ5j0rj5TBo3f7/7qiOi2HHMGYTn7SBt6tv7PVaTdHht\ni9X06eYQERERSynpEKlFTRF5w5MO89F9sNVzpCaWkZpYdsD9O447i+rQMFNQXlX11/3t25t5HV5L\nOpo3N4eIiIhYSkmHSC0OtYg8cdFvVEVEUdK0reeC8kH5xaHkF4cecH9lXDK7eg8hOnsVKTO//Ov+\n0FDo3BmWLYOcnAOe5n67dplDRERELKWkQ6QWh7LSEbZrO9FZTooyOptq6SBy/jO9Of+Z3rU+tv2E\ncwFo8dH+wwK9WtfRo8feXr0iIiJiGSUdIrU4lJWO+GW/A1CU2cWDEfmfstRW5HXqT8LSmcQvmfnX\n/V6v6xARERHLKekQqcWhJB1xzjkAFLfs6MGI/NP2E88DoMXHe1c7HA4ID1fSISIiEkyUdIjUYuNG\nCAmBlJT6z41fPhuA4pYOD0flf4pad6OoZQdSpv+PqOxVgEk4OneGxYshN9fiAEVERMQrlHSI1CIr\nCxo1akCJhstFnPNPyho1pSomwSux+RWbjZwTRmJzuWj+yfN/3d29u5mF8ttvFsYmIiIiXqOkw8Mc\nDsebDodju8PhWFLH4yc4HI48h8OxYM9xn7djlP1VVUF2dsO2VkVuXktYQa5WOQ5id7fjKWuUTtqU\nSYTtNi2rVNchIiISXA7sdSnuNhl4BXj7IOf85nQ6T/NOOFKfrVtN4tGQaeTxf9VzdPBwVL7p+mFr\n6z/Jbifn+BE0//xlmn4xgQ2X3U/HjhAW5oWk47HHPPwCIiIi0hBa6fAwp9P5K6Cd636kpog8NbX+\nc+Nq6jlaBGfScWqv7Zzaa3u95+X2O5XK6Hiaff4KIWUlRERAx46wYAHk5XkwwAsvNIeIiIhYSkmH\nb+jvcDgWOhyO/3M4HJ0b8gSHw/GAw+Fw7Xt4OshgUTOjoyErHXHOObhsIZQ0b+fZoPxcdUQUOwac\nTnjeDtK+ewswW6yqq2H6dIuDExEREY9T0mG9eUArp9PZHXgZ+F9DnuR0Oh9wOp22fQ+PRhlEGtou\n11ZVSdyqeZSmtaI6IsrzgfmgcW90ZdwbXRt07o7jzqY6NIzmnzwHVVXeqes47TRziIiIiKWUdFjM\n6XTmO53Owj1//hYIczgcDWjUKp7S0Gnk0RuWYy8tDtp6DoDVW2JYvSWmQedWxiezq9fJRGevImXW\nV3TqZLqD/fyzBwNcssQcIiIiYiklHRZzOBxpDofDtufPfTHfk53WRhXcGrrSEez1HIdj+6CRALT4\n8GmioqBDB5g3DwoKLA5MREREPErdqzzM4XB8AJwApDgcjmzgfiAMwOl0vgqMAK5xOByVQAlwvtPp\nVH2GhbKyzAC7xMSDn7e3c5Xa5TZUWWor8jr1J2HpTOKXzKR79wEsXQozZsDQoVZHJyIiIp6ipMPD\nnE7nBfU8/gqmpa74iI0bTRG5rZ4qmbgVc6gODaM0vbV3AgsQ2weNJGHZLFp8/CzdTxvA+++bLVZK\nOkRERAKXtleJ7KO0FLZvr39rVUh5KTFrF1HSrC2u0DDvBBcgitp0p6iFg5Tpn9MveZXn6zpERETE\ncko6RPaRnW1u60s6YlcvIKSqMujrOQZ0yGVAh0McQ2OzkTNoJDaXi7ZfPU+HDvDnnx6q6xg6VEso\nIiIiPkBJh8g+GlxEvkL1HAD3jFzJPSNXHvLzdncbSFlyOmlTJnFchxyqquC33zwQ4KuvmkNEREQs\npaRDZB+HnnR09HBEAcpuJ2fgOdjLS7k4fwKgLVYiIiKBTEmHyD4aOqMjbuWfVEVEU9a4ueeD8mGT\np7Vg8rQWh/Xc3H7/oDI6jj5/vEKsvYSffnJzcABPPWUOERERsZSSDpF91Kx0NG5c9zkhpcVEZzkp\nadYWQoL7f6GPZzTj4xnNDuu51RFR7BgwnPD8HdyW+jbz5kFenpsDnDDBHCIiImKp4H7HJPI3DVnp\niFm7GFt1NSXN2ngnqAC249izqA4NY2zBM9iqKz1T1yEiIiKWU9Ihso+sLIiNhZiYus+JXb0AwKx0\nyBGpTGhEbp9TSCtYzUW855ktViIiImI5JR0i+8jKakC73DV7ko6mSjrcYdvJF1NtD+M+HmL6TxVW\nhyMiIiIeoKRDZI+8PMjPP3g9B5iVDleIndL0DK/EFegqklLZ2X8YbVhL9/mT2bXL6ohERETE3ZR0\niOzRoHa5VVXErl1EaVorXKHhXonLl0WFVxEVXnXE19l20kWUh0RwDw8zfVqZGyLbIybm4HvlRERE\nxCuUdIjs0ZAi8qhNq7GXFmtr1R7/vXMO/71zzhFfpzIhBWfXc2lJFmUTJrohsj2WLjWHiIiIWEpJ\nh8geDVnpUBG555SfMYJCYjj+t8egpMTqcERERMSNlHSI7NGgpGONko59LVofz6L18W65lj0pgU8T\nx9CkcjNFT7tptsbvv5tDRERELKWkQ2SPhmyvils1H4CSpprRAXDn25248+1Obrvekl6XkksSYU8+\nDDk5R37B8883h4iIiFhKSYfIHjUrHSkpdZ8Tu3oB5UmpVMW459N92V+njnA/DxJenAf33GN1OCIi\nIuImSjpE9ti4EZKTIbyOplThuVsJ37VNk8g9qEurfCaFjmVVWEd44w2YP9/qkERERMQNlHSIANXV\nkJ3d0CLydl6KKviEh7ro0rqIaypeApcLxo0ztyIiIuLXlHSIYMoHysvVucoX9Gm3m2mcRHb7E2H6\ndPj4Y6tDEhERkSOkpEOEhhWRK+nwjgGOXABeTH0UwsLgttugsNDiqERERORIKOkQYW8ReePGdZ8T\nu3oBlVGxlCeleicoP/DM6CU8M3qJW6/paFZIUmw57y3rgWvkSPPNufXWw7vYf/9rDhEREbGUkg4R\n9q50pNaRT4SUFBGVvZLSpm3AZvNeYD6uU4tCOrVw7ypESAj0d+xiy84Ilhx7NWRmwmuvwbffHvrF\nevUyh4iIiFhKSYcI9a90xK5bjM3l0tYqL+m/Z4vVlAXpcPfdEBoKV1wBO3ZYHJmIiIgcDiUdIuxN\nOupa6VA9R+2GP9qX4Y/2dft1+3cwScd3c5KhbVsYPRq2boVrrjm0blbt2plDRERELKWkQwSzvSo0\nFJKSan88Zu1iAErSW3sxKt9XWRVCZZX7f4w0iqvA0ayA3xYnUFwaAuedB126wKefwrvvNvxCFRXm\nEBEREUsp6RDBrHSkpJh6gtrErFuMyxZCaWor7wYWxPo7dlFeEcIvCxPBboe77oKoKLjqKvjjD6vD\nExERkUOgpEOCXnk5bNlykHa5LhcxaxdTltIMV3iEV2MLZvttsQJo2hTuvRfKyuD002HNGgujExER\nkUOhpEOC3ubNpkygrqQjfMdmwgp3U5qe6d3AgtxRmXlEhlfx3Zx99rz17w833GCmOZ56qgrLRURE\n/ISSDgl6NUXkdSUdsetUz2GF8FAXfdruZsXGGDZu22eFafhwuOACWLUKzjgDSkqsC1JEREQaREmH\nBL36ppHXFJGXpmd4JyA/ctHAbC4amO2x69e0zv1ri1WNK6+EE0+EmTNh0CDT2ao2N95oDhEREbFU\nqNUBiFitvpWOmHVm4napVjoO4MmEA2DAPnUdY07bsveBkBC44w7TcmzqVOjbF776Crp33/8CSjhE\nRER8glY6JOjVu9KxbjHVYeGUpTT1XlACQMvGJTRNLuGHuUlUVv1tEnx4ONx5J4wZYzLHY46Bzz6z\nJlARERE5KCUdEvQOttJhq6okZv0y0yo3xO7dwPzAAx86eOBDh8eub7OZ1rl5RaHMXh5X+wkXXggP\nPQRVVXDOOXDaabB8uXn8iivMISIiIpZS0iFBLyvLjH+IjT3wschNawipKNPWqjrMXpnE7JV1TFR0\nkwF/b51bm+OOgwkToEcP+OYb6NoV/vlP+O47mDbNo/GJiIhI/ZR0SNDbuNGscthsBz6mzlXW69Nu\nF/YQ18GTDoDMTHj2WXjkEUhPN0nIpk1mCMsLL5hvtMvlnaBFRERkPyokl6BWWAi7dkGbNrU/rs5V\n1ouLqqJ7Zh6zVySwNTectOTyuk+22UxtR79+psD8xRfN9MebbjJHSopZBena1XzTGzc2R5Mm5jYl\nBcLCvPeXExERCRJKOiSo1dRzpKbW/nhN5yqtdFjrxK47mLcmkf9NT+Hq4Zvrf0JoKPzjH/D221Bd\nDRdfDHPmwNq18NNP5qhLUtKByUhmJnToYI42bZSYiIiIHCIlHRLUajpXNW5c++Mx6xZTGR1PZXwj\n7wUlBzixWw7P/K8tn/7SuGFJx75CQsxAweHDzdclJbBunZlqvnt37UdODqxebRKWvwsPh169zIrK\nMceYepJG+u9DRETkYJR0SFDbsMHcpqcf+FhIaTFRm1ZT1Lpr7QUfQsfmBV55nfSkMrq2yufnBYns\nyAsjJaGiYU9s3/7A+6KioFOn+p9bVWX23+XmwubNJkPduNGslsyeDbNmwTPPmKRmwADTNev006Fj\nR/33IiIi8jdKOiSorV9vbtPSDnwsesNybC6XtlYdxLOXL/Xaaw3ulsPiDfF8Mb0RVwyrYwL53z30\n0OG/oN0OCQnmyMw0qxo1SkpgxQpYvNhs25o5E6ZPN3NDOnWCiy6CCy4wzxMRERF1r5LgdrCko6Zz\nVWm63jj6gsHdcgD49Nc69sJ5U1SUac976aXw8stmKOFdd5mtVqtXw7/+Ba1bm0Rl0iQoKrI6YhER\nEUtppUOC2vr15gPt2rbk/9W5Kk1JR12+nG2yteF9G7jycASap5TSsXkB0+YlsasglKS4yvqfNGWK\nuR061LPBJSTAkCHmKCw0qx4//GC2YM2cCTfeaIrZx4yBo47ybCwiIiI+SCsdEtTWrzcNiuy1DBvf\n27lKSUddXp2SwatTMrz2eoO75VBRGcJXMxtYuD15sjm8KTbWJDnPPAPvv29WQ8LD9w4v7NsXJk40\nyYmIiEiQUNIhQau01MyNq21rFZjOVeVJqVRH1TKqXCxxUnezxeq/vrDFqiHS0mD0aPjwQ3j0Uejf\nH+bONSseTZvCddeZ2hAREZEAp6RDglZNu9zaZnSE5u0kYucWStIyvBqTHFyrJiW0TS/kuznJ5BfV\nsjzlq+x20+HqscdMAjJqFERGwvjxptvVkCHw5ZemY5aIiEgAUtIhQetgReQ1W6tK1bnK55zUmyBA\nSQAAIABJREFULYeyihC++d1PZ2M0bgyXXQYffAAPPADdu8P338MZZ0C7dmZbVm6u1VGKiIi4lZIO\nCVrqXOWf/G6LVV3sdhg4EF54wdR4nHaamQdy223QvDlcfTWsXGl1lCIiIm6hpEOC1kFXOvZ0rlIR\nue9pnVZMRpNivv0jmaKSAPkR1qYN3HILfPIJXHMNJCbCa69Bhw5w5pmmG5aIiIgfC5Df2CKHrmYa\neV3bq1whdspSW3o3KD/z6R1z+PSOOV59TZvNrHaUlNmZMjv54CdPmmQOfxEXByNHwjvvmK1XDgd8\n8YWZ/3HCCTBtGrhcVkcpIiJyyJR0SNCqmdGRkvK3B1wuYtYtoaxxc1yh4VaE5jeiI6qIjvB+8XPN\nFquPf25y8BOjoszhb2q2Xk2YAC+9BP36wS+/wEknmYGD339vdYQiIiKHREmHBK26ZnREbNtIaHG+\ntlY1QPaOSLJ3RHr9dds3LSQztYj/zUhhW25Y3Sdu3mwOf2WzQdeu8MQT8OqrJuGYNct0uzrlFFi4\n0OoIRUREGkRJhwSlsjLzXrS2drnqXNVwYyccxdgJ3p+wbbPBecduorwihInfptd94s03myMQOBzw\nyCOm1qNXL5g61QwbHDXKDJwRERHxYUo6JCjVzOhQ5yr/Naz3NqIjKnn1y6ZUVtmsDsd72rc3bXWf\negpat4a33jKzPl59FaqrrY5ORESkVko6JCipc5X/i42s4rTe28jOieTLGf43s6OkLITVm6L4fVk8\nX89qxKT/S+Ol/zZj9aYG1qD06WNWPW66yQwVvOYaOPZYWLzYs4GLiIgchlCrAxCxQk3SUdf2qqrw\nSMqTD7JtR3zCyGM38fGMZoz/XzPOPn6H1eE0SElZCM9/0pwnPmhJQfGBP4JvnuDi4pO38a+LN9Cu\necnBL2a3w/DhptbjlVfg55+hd29TA3LDDRCiz5VERMQ3KOmQoFTXSoetsoLojcspadpGb9j8QJu0\nYvq03cWP85NYtj6aThnFVodUp+pqeO+HVO6emEl2TiRJMeUM77uFpJgKkmIrSIypoKraxvu/Nuet\n79J45/tULhq8jXsu2UD7FvUkH40awf33mwLzp582dSxTp8LkybVn1iIiIl6mpEOCUl1JR1TWSkIq\nK1TP4UfOO24Tc1YnMeGLZrxywyqrw6nVrKXxXP9SO+aujCM8tJrRgzcwevBG4qIObDd8Zr8tTFvU\nmNentuKd79N4f1oq79+zjJGDcup/of79zXTzJ5+EKVOgWzd47z3TaldERMRCSjokKNXM6GjceP/7\n1bnq0Nx97kqrQ2Bg552kJpby1nepPHblWuJj9nkjf9NN1gW2x1tTUrnyGQeVVSGc2nMb1w1bS9Pk\nsjrPDwmBk4/KYXC3HKYtasyDHzm4+LGOJMRUckrfXfW/YHIyPP44fPYZvP46DB0Kzz0H119v2n6J\niIhYQPtHvMDhcLzpcDi2OxyOJXU8bnM4HC85HI7VDodjkcPh6OntGIPNhg0m4fj7jI6azlUlaVrp\naIhjO+ZybMdcS2MItbsYMWAzhSWhvDP1b1uJ+vUzhwVcLnjorVaMerIj0RFVvH7tAh67ZPlBE459\n1SQfL1yxmBCbi7Pv68LMJfENe/GQEBgxAp5/HhISTH3H2LFQXn4EfyMREZHDp6TDOyYDQw/y+KlA\nuz3HWODfXogpaNXM6DhY56rSpko6/MlZR28h1F7N+C+a4XJZHQ1UVNq4/CkH90/OpGlyCZPHzadP\nu92Hda3ebfN48tJllFWEMOyurixaE9PwJ3fuDP/+t2mzO3Gi2WaVa22SKCIiwUlJhxc4nc5fgYP9\npj8DeNvpdLqcTufvQKLD4VDrJA/JyjKfQteadKxbQkVsIpVxyd4PzA+NfqkHo1/qYXUYNIqrYMhR\nOSzfEMOP8xL3PnD99ebwovwiO8Pu6srkKel0apHP2zfOIzP1yArcB3bZyYMXrGB3YRhDbuve8La6\nAE2awIsvwgknwG+/wcCBGiYoIiJep6TDNzQDsvb5OnvPfeIBdbXLtZcUErVlLaXaWtVg23ZHsG13\nhNVhAGZCOcC9kzL3zsjLyTGHl+wqCGXwLd35/s9kju+0g4n/XECjuAq3XHtY723cftYqtu0KZ8ht\n3cgvstf/pBqRkXDvvXDWWbBkiWmxu2aNW+ISERFpCCUdvqG26s6DbhJxOBwPOBwO176Hh2ILOHV1\nropetxTQJHJ/1S0jn5O6b2fW0gQmT6llGcvDdhWEcvKt3fnTGc8Zfbfw7OVLiYpw74TwC47fxKgT\nN7JuSxR3vXGIzQ5CQsyqz6hRsG6dBgmKiIhXKenwDdlAi32+bg5sPtgTnE7nA06n07bv4dEIA0hd\nSUdN56qSpupc5a9uPXMNUeFV3PF6a3LzvdecLzc/lJNu6c7clXGc2W8L953nJNTumc8Brjl1Ha1T\ni5jwRTNmLG5gYXkNmw0uuwz++U/YuhUGDTIrHyIiIh6mpMM3fAlcuqeL1dFAntPp1KZrD6kr6ajp\nXKXtVf4rNbGMq05Zz468cP71H+98H2sSjnmr4jjr6M3cO9Lp0bmS4aEu7jvfic3m4spnHJSVH8bn\nDSNGwK23ws6dMHgwrFjh/kBFRET2oaTDCxwOxwfALPNHR7bD4bjC4XBc7XA4rt5zyrfAWmA18AZw\nrUWhBoU6Z3TUdK5Ky/B6TOI+Fw7MpnVqEa991ZQ/K7p79LV25pmEY/7qOM4+ejP3nLvSK4Psu2fk\nM/KYTazYGMNj77U6vIsMGwY33gjbt8OJJ8Iq3xys6DNcLigoMLUws2bBjBlQ4Z56HRGRYGBz+UJ/\nSXELm83m0vezfs2bQ2UlfPjh/vcPOCsVV4id5fe+b01gfuilr81qwrjT1lkcyf7+XJ3ImPFH0Sd5\nDbOG3I/9qivd/hrZOREMua0byzfEcE7/zdw9wjsJR42iUjvnPNmH3MJw5r/+J50zD7ND1qefwvjx\n5n+M336DjAy3xun3tm6Fl182gxZ37Nj/sUaN4Oyz4dxzzVa1UM3bFZHgYbPZcLlcDV5u10qHBJXy\n8tpndITt2k747u0qIj9E405b53MJB0Dvtrs5tec25uS2YWL6vW6/vnNjFMdc14PlG2K4aGCW1xMO\ngJjIKu4esYqKyhCufMZBVVX9z6nViBFw1VWQnQ2nnHLgG+tgtWYNXHkltGoFjz1mVjX69TMT3s8/\nH4YPN6sfb7wBQ4ZAz55aLRIROQh9LCNBpa4ZHX8VkSvpCBg3DV/Dr0sbcdfE1gw/ZifpjdwzjftP\nZxyn3tGVHXnhXD9sLaMHb8RmURuH4zvvZEiP7Uyd34QJXzTj+rM3Hd6Fzj8f8vLM8t/pp8O0aRAd\n7d5g/clbb8G110JxsVkBOvdck5BF/K099LhxpgPYt9/C999D797muWeeaU3cIiI+TCsdElTq7FxV\nU8+Rrs5Vh+KlrzP/2mLlaxonlHNXq/fZVRDGCTceRdb2I58nMm1uIoNu6k5ufhj3jnRy+UnWJRw1\nbj9rFfHRFdzzn0x25h3B50hjxsDJJ8Pvv8N555k9iMGmsNB09xo1ynT6uvtumDzZrGr8PeEAUxx2\n1FHmvLvvhrIyMwvljjuC899PROQglHRIUKm3c5VWOg7JlHmpTJmXWv+JFrl1++3cGvEyK7OjOfb6\nHqzKPoRJ3vuorLLx5Act+Mdd3SirCOGpUUs5u79vNJhrFFfBmJM3kF8cyhMftDz8C4WEwG23mU/r\nv/4arrnGLAsGC6fT/N3ffhs6dDA1HCefbBKLhjj5ZJgwAZo1g6eeMjNRgunfT0SkHko6JKgcbEZH\ntT2U0iYtDniO+C+bDR6NfpTrh61l4/ZIjhvXg0VrYg7pGkvWxTDguh7c+Xob4iIrGT92EYO7+Vbd\nw7nHbCYtsZSXP2tOds4RrOiEhcGDD0K7djBxovlzMJg/H447ziQeI0bASy9B06aHfp3WreHVV6FN\nG3M7frz7YxUR8VNKOiSo1CQdqft+OF9dTcy6JZQ1aQl2lTkFostP2shd56xk265wBt54FL8vq3+o\nXkWljUffbUmvq3oxZ0U8w3pv5b93zKZPu91eiPjQRIRVc82p6ymrCOGByRlHdrHoaHjiCUhPN0nH\n66+7JUafNWOG6Ty1YwfcdJMZnBgWdvjXi42FRx6BpCS44QaYOtV9sYqI+DElHRJU1q83u0j2ndER\nuXU99tIiStMzrApLvGDksZt55KLlFBSHcuz1PRh0U3ee+agFyzdE/7ULJjc/lC+mN+Km8W3ocnkf\n7vlPaxKiKnjhisU8ctEKEmJ8d5/+sN5baZ1axKQpaazYeIRF4MnJZotQQoLZZvXVV+4J0td8953Z\nFlVYCP/6l6ndcIe0NHj4YbM1a+RIDV8UEUHdqyTIrF9vEo592+nv7VylIvJAN6z3NuKjK3jtuwx+\nXpDEzwuSuO3VNmSmlxAfXcWitTHUtBwPD63mjH5buHn4GuKjfTfZqGEPgeuGreXmN7vyr4mZ/Peh\npUd2webN4fHH4eabTWH5tGnQv797gvUFU6fuTTIeftj9f7fOnU2NzGOPmdeZPx9iDm1rn4hIIFHS\nIUGjvBw2bYKuXfe/P2adOlcdrtTEMqtDOKjyxMYH3Hdcp1yO65TLzoIwZixPZvqyRsxyJrMpJ4Je\nbXbTu81uerXdTddWBUSEVVsQ9eE7octOurbK47PfGvPHsjj6dSo4sgt27Aj33w/33GNa6c6cCe3b\nuydYK/366962to89Br16eeZ1Tj7ZzO745BPz7/jMM555HRERP6CJ5AFEE8kPzuk0TWmGDjUdLWt0\neuh8mvz0Ecvu/YDy5LS6LyABq7LKhssFYaH+///Pn6sTGDO+BycctYsfn1vonpa+33xj3jC3bg2z\nZkGTJm64qEVmz4bBg6G0FB56yPOrN2VlcMUVsGWLaUfcp49nX09ExEs0kVykDjXDgps33//+mHVL\nqIqIpjzJd1u/imeF2l0BkXAA9G6bxzEdd/LzgiSmzklyz0WHDYNLLoG1a82KR1GRe67rbQsXmiF/\nxcVm9cYb28UiIuCWW6C62kw4r6jw/GuKiPggJR0SNGpLOmwV5URlOc18DqunvPmh6cuTmb482eow\n6hS3/A/ilv9hdRhed/2wdQD86z+t3TcqYvRo84Z99my48EL/G363fLnZ7pSXZ5Y6Bw703mv36GES\nt0WL4Omnvfe6IiI+REmHBI2apKNZs733RW9cQUhVJSVpGZbE5O8e+6Q9j33iu3v8W3zyPC0+ed7q\nMLzO0ayQk7tvZ+7KOL6a2cg9F7XZzCf2vXvDl1/C1Vf7z/C7NWvgpJMgJwduvBGGDPF+DFdfbbqC\nPfSQ2espIhJklHRI0Kgt6ajpXKUicgk0V52yHpvNxf2TM9yXG9QMD2zfHv7zH7jrLjdd2IOyskwN\nx+bNcO217muLe6hiY83cjrIyGDfOmhhERCykpEOCxqpV0KgRREXtve+vzlVNlXRIYGmTXswpPbaz\nYHUc/5ue4r4LR0fDk09Cixbm1pe3C23ZYhKODRvg8svh3HOtjef4481K0dSpGhooIkFHSYcEhbIy\n2Lhx/1UOgNi1JukoScu0ICoRzxo7ZAMhNhcPTM6g2p3dfxMTTbLRuDHcfrtZ9fA1mzfDCSeYTxsu\nvBAuvtjqiIyrrjJb1W6/HaqqrI5GRMRrlHRIUFi71mw/r61zVUV8MlWxCdYEJuJBmanFnNpzG4vW\nxvLZbwfOLDkiqakm8UhIgDFj4M033Xv9I7Fpk0k4Vq6ECy4wXaN8pVFE27amoH3hQnjvPaujERHx\nGiUdEhRq61xlL8onctsGStK0tUoC1xhPrXYAtGpl5nfEx5tZFK+/7uYXOAzZ2fuvcIwZ4zsJR40r\nroDwcNO2t6TE6mhERLxCSYcEhZqko2nTvfftLSLP8H5AAeL1axfw+rULrA6jTquvfY7V1z5ndRiW\natWkhGG9t7F0fQyf/Ozm1Q4wn9w/+6xZ8bjqKvj3v93/Gg21fDkcdxysXm22U/nSCse+mjSBc84x\nRe4vvmh1NCIiXqGkQ4JCbSsdsWsWAlDatI0FEQWG5imlNE8ptTqMOpWnNKU8pWn9Jwa4MUPWYw9x\n8cBbGZ4pI2jTBp5/HpKSTIeoRx/1fjvdX3+FAQNg/XpTNH755b6ZcNS48EKTqD3+OOzYYXU0IiIe\np6RDgkJt7XJrko4SJR2HrbjMTnGZ3eow6hRSVkJImbavtEgp5bQ+W1mxMYYPfkz1zItkZprEo0kT\ns23okkug1EsJ6QcfmDqJggIz+O+SS3w74QDTQvfiiyE/32xRExEJcEo6JCisWgUpKRAZufe+mLWL\ncIXYKU1rZV1gfm7Ek30Y8WQfq8OoU4cnR9PhydFWh+ETxg7ZQKi9mvsmZVBe4aE35K1ame1VHTua\nIulBg2DrVs+8FkB5uUkyLrzQzBB58kkYOtRzr+duw4ebH0wvvwzbtlkdjYiIRynpkIBXUmK2Tu/X\nuaq6mtg1iyhNbYkrNNyy2ES8pWlyKecO2My6LVG8/rUHt5wlJ8MLL5gJ4L//buZSfPON+19n+XI4\n+mh46ilTrPXyy9Crl/tfx5PCw+Gii6C42Pw9REQCmJIOCXhr15rbfbdWRW1Zi720SFurJKhcefIG\noiMqefjtVhSWeHBbXHg43H03jB1rVjpOO828uc7JOfJrV1SYBKNnT5g/H/7xD5g40Wzv8kf/+IfZ\nkjZhghlmKCISoJR0SMCrrZ4jZrXqOST4JMdVcOkJWWzfHc5zHzev/wlHwmYzMzJefx06dID334dO\nneCVV0ztxaGqrITJk821xo0zic2DD8Jtt0FUlNvD95rwcFPbUVoKTzxhdTQiIh6jpEMCnjpXiex1\n8QnZJMWW8/RHLcjZHeb5F2zd2iQa115rko3rrzf/M954oxnedzAuFyxbZlryduwIo0ebvZJnnQWT\nJsHxx3s+fm8YOhTS0uC118ycERGRABRqdQAinnawpKOkmZIOCS4xkVWMOXkDT33ejkffbcUL1632\n/Iva7XDuuTB4MHz9NXz5pZlP8eKLZrJ5jx7mSE01iUlBgdmK9eOPsGGDuUZoqCm8vugisx0pkISF\nmY5bTz9tWuiOH291RCIibmdzebuXuniMzWZz6ft5oEGD4OefYcoUiIgw9x19fgb24gKWPvRfS2Pz\nd1/OTgNgeF8Pdig6AsmzpwCQ29ePOhp5QUWljTMf78uOggicb88mI83Ls1YqKuC33+Cnn8ynAnV1\nboqNNYXo/fpB376mSD1QVVbCqFGwfbv5N2mlrnoi4ttsNhsul6vB7RCVdAQQJR21a97cvMf56CPz\ndWjhbo49PYl8R2/WXv20tcGJWOSbP1O5572OXHLyVt6+e4W1weTlmSniRUUQHW1qNGJioEULs0oS\nLKZONSsdY8aYWhgRER92qEmHajokoBUXw6ZNfysiX7MIUBG5BLdTe26jXdNC3v0hlfmrYq0NJiHB\ntLs9/nizstG5M2RkBFfCAWb7WcuWpl6lpu2eiEiAUNIhAW3NGnNb2yTyUtVzHLFb3uzMLW92tjqM\nOmW8eR8Zb95ndRg+KSQEbh6+BpfLxlXPtaeqyuqIBLsdLr3UbLV65BGroxERcSslHRLQDlpE3rSt\nBREFluXZcSzPjrM6jDpFZ68kOrueDklB7GjHLk7tuY05K+L595fN6n+CeN4JJ5hVnrff3vsDTEQk\nACjpkIBW64yONQupDg2jtEkLa4IS8SG3nLmauKgK7p6YyeYd4VaHI3Y7XHYZVFXBQw9ZHY2IiNso\n6ZCA9veVDltVJTHrllCa2grs6hgt0iiughtPX0tBcSg3vKLVP59w/PHQpo0ZqLh8udXRiIi4hZIO\nCWirVpnByE2bmq+jsldhLy/VUECRfZzZbwtHZebx6S9N+GZWALel9RchIaZ9bnW1VjtEJGAo6ZCA\ntmqVmSMWvmfXyN56DiUdIjVCQuBf564kNKSaf77YnqIS/Wqw3DHHQLt2ptf3kiVWRyMicsT0m0UC\nVmEhbNmyfxF5jCaRu1Xf9rvo236X1WHUqaB9Twra97Q6DL/QNr2IS0/MYsO2SB58K8PqcMRmg9Gj\nweWCBx+0OhoRkSOmTe0SsA7WLlcrHe7xwPlOq0M4qKzzb7c6BL9y5ckbmDq/Cc9+0oIhfXZxUi/f\nTSiDwtFHQ4cO8OmnsHAhdO9udUQiIodNKx0SsGrrXBW7egHlCSlUxSRYE5SID4sKr+aRi5cTYnNx\n3oOdWLcl0uqQglvNagfA/fdbG4uIyBFS0iEB6++dq8J3biFi5xZKWrS3LqgA894vzXnvl+b1n2iR\nxr98SuNfPrU6DL/SPSOfu85ZRW5BGGfd20X1HVbr08dMaP/iC5g71+poREQOm36bSMD6e9IRt9L8\nwi5urqTDXZR0BKaz+2/hnP6bWbgmliue7oDLZXVEQWzf1Y777rM2FhGRI6CkQwLWqlWmK096uvk6\ndk/SUaKkQ6Red5y9iqMy8/jopyY885EGaVqqZ0/o1g2+/RZ++cXqaEREDouSDglILhcsWwZpaRAW\nZu77a6WjhcPCyET8Q1ioi6dHLaVxQhl3vtGa72YnWR1S8LLZ4OqrzZ9vuslMKxcR8TNKOiQgbdkC\nublmqG+NOOdcyhNSqIzX8DORhkiJL+fZ0Uuwh7g4674ufP5bitUhBa+OHeHkk2H+fHjrLaujERE5\nZEo6JCAtWmRuW7c2t+G5W4nYuVlbq0QOUddWBTw7agm44Jz7O/PCp75bwxPwxoyByEi4+24oKLA6\nGhGRQ6KkQwLS4sXmNjPT3MY6VUTuCaH2akLt1VaHUSeXPRSXXeOIjtRxnXOZeN18GsWVc9P4ttzw\nclvt8LFC48Zw/vmwbRs8/rjV0YiIHBKbS21JAobNZnPp+2lceim88w68/Ta0aAGt3nqIzMn3s/bK\nx8jv3N/q8ET80pZdEVz/ejfWbI1h+IAdvH/PMmKifDfpDEilpeYHXH4+LF++95MVEREvs9lsuFwu\nW0PP10qHBKRFiyAiApo2NV+rXa7IkUtPKmPSuHn0a5/LlzNTcFzaj9e/SqeissG/c+RIRUbC2LFQ\nVgY33oj6GYuIv1DSIQGnosJ8AJiRAXa7uS9u5Vwq4htRmdDI0tgCzbKsWJZlxVodRp2ispxEZTmt\nDiOgxEVV8fLYxVxx0gZ25oVy1XMOOo/uw4c/NqFaix7eMXgwdO8OX34JH39sdTQiIg2ipEMCzsqV\nUF6+t4g8LHcbETs2aZXDA26d1IVbJ3WxOow6ZU66n8xJ91sdRsAJs7u4btg6vvjXH5x7zCbWbY7k\ngoc70XNsb578oAVzVsRRWaXVD4+x2eC228xy7nXXQU6O1RGJiNRLFZYScGqKyGuSjpqtVSUtlHSI\nuFOThHLuHrGKS07I4rUpGXw7L5WFa0yf6vjoSgYetZtjuuSRmlRBo/gKkuMqaJRQSWR4NWXlNkrL\nQyirCKG0PISiUjv5RXbyi0PJL7aTXxRKeaWNisoQKqtsVFbZqK6GhNhKGsVX0ijeXDO9UTldWxcR\nGR5kyyzNmsHll8O//w033ADvv291RCIiB6WkQwLO39vlxjn/BFTPIeIpLVJKeeTiFdx0xhr+XJ3I\nnFVJzF6VyFczU/hqpudne4Taq+nauog+jgL6dChgYPfdtGte4vHXtdw558DPP8MHH8B558EZZ1gd\nkYhInZR0SMCpa6WjWCsdIh7VKK6CU3rkcEoPs91n664IlmfHkl8cRl5xGHnFoeQVhVFWEUJ4aDXh\nYdXmNtRFVHgVsZGVxERWERNZSUxEFeFh1YSGuAi1u7CHuLDhoqAklN3FYeQVmett2RXJ0o1xLF0X\nx/xVcbz+tYmlf+c8Rp2ylfMGbSchNkD7+9rtcPvtprD8mmvguOMgWcNPRcQ3KemQgLN4sfm9m5ho\nvjZF5MlUJmiasog3pSWVkZZU5pXXqqiysWZLDIs3xPPzkhRmLUti1tIEbnilLWcdu4Nx52zi6E75\nXonFqzIyTAvd//wHLrkEvvoKQlSuKSK+Rz+ZJKDk5cGGDSoiFwk2YXYXHZoXcu4xmxl/1SL+777f\nuX7YWlITyvjgx1T6/7Mn5z3YifVbI60O1f0uuAB694Zvv4VHH7U6GhGRWmk4YADRcECYMQOOPRbO\nPReuvRaSf/+WbncNY+uQS9l66mirwws4i9bHA9Atwzc/QY5evwyA4oxOFkciVnG5YO6aBF76ug2L\nN8QTEVbNjSOyuevCDYG17SovD666CrZvN8nH0KFWRyQiAU7DASWo1VlErnoOj+iWke+zCQeYZEMJ\nR3Cz2aB32zzeumEej12yjOTYcp78oCXtLunHRz82tjo890lIgAcfhNBQuOgiWL/e6ohERPajpEMC\nyt+LyOOXzQKguJXeeIoEM5sNTu25nc/unM11w9ZSWGzn/Ic7c83z7SgtD5BfhQ6HaZ+bmwtnngm7\nd1sdkYjIXwLkJ62IsXixqaHMyACqq0lYOovSlGZUxiVZHVpAOueJPpzzRB+rw6iT44lROJ4YZXUY\n4kMiw6u54qSNvH/Ln7RrWsirXzZjwD97sHpTlNWhucewYTB8OCxcaP5cWGh1RCIigJIOr3A4HEMd\nDofT4XCsdjgcd9by+CiHw5HjcDgW7DmutCJOf+dymaSjeXMID4foDcsJLcqjOKOz1aEFrJJyOyXl\ndqvDqJO9vBR7eanVYYgPymhSwts3zOPsozczf3UcPcf24pOfA2S71bhxMHgwzJxpZneU6v8BEbGe\nkg4PczgcdmA8cCrQCbjA4XDUttfnI6fTedSeY6JXgwwQWVmmljIz03ydsGdrVZGSDhGpRWR4Nfee\nt5JHL15GZaWNkQ925sG3WuH3/TjsdrjzTjjmGPjxR9NZo6LC6qhEJMgp6fC8vsBqp9O51ul0lgMf\nAhob6wE1ReRt2pjb+CUzASjKVNIhInX7R6/tvHvzXJo1KuGByZnc+EpbqqutjuoIhYbh0GNpAAAg\nAElEQVTCffeZVrpff222XOX7btMHEQl8Sjo8rxmQtc/X2Xvu+7tzHA7HIofD8anD4WhR30UdDscD\nDofDte/hroD9VU0R+V8rHUtnUhURTWlahmUxiYh/aJ1azJvXz6dteiEvfdacy57oQEVlgztB+qbw\ncHj4YejXD6ZMMRPLs7Lqf56IiAco6fC82n5r/T1B+ArIcDqd3YAfgLfqu6jT6XzA6XTa9j3cEKtf\nq0k62rSB0LydRGc5KWrVCUJ8t+ZARHxHk4RyJv5zAV1b5fHu92mcfV9nSsr8/NdkZKQZGHjGGWY5\n+OijYf58q6MSkSDk5z9N/UI2sO/KRXNg874nOJ3OnU6ns2zPl28AvbwUW0BZtAiioiA1FeKX/Q5o\nKJynjTxmEyOP2WR1GHXaccwZ7DhGuxml4RJiKnntmoX0d+Ty9awUht7ejcISP//gwm43rXSvuQa2\nbDETVCdPxv+LV0TEnyjp8Lw5QDuHw5HpcDjCgfOBL/c9weFwpO/z5XBguRfjCwhlZeB0mvkcISFm\naxWonsPTRg3OYtRg392usX3wBWwffIHVYYifiYqo5oUrF3Ny9+38uiiRM+/p4v+zPGw2GDkSHnrI\n/Hn0aLjkEigosDoyEQkSfv5T1Pc5nc5K4DrgO0wy8bHT6VzqcDgecjgcw/ecNs7hcCx1OBwLgXHA\nKGui9V8rVkBl5d56jvhls3DZbBoKKCKHJTzUxaOXLGdglx1Mm5fEBQ93pLIqAHaxHnssvPEGdOwI\n770HPXrAnDlWRyUiQcDm0vJqwLDZbK5g/X6++6750G7cODh7eCXHnpZAeWITnHdMsjq0gPbIx+0B\nuGfkSosjqV3zj58DIHvkzRZHIv6qrCKE61/vypzVSVw6ZCuT7lhBSCB8XFdZCZMmwfvv722xe++9\nEBFhdWQi4idsNhsul6vBn8YEwo9Okb+KyFu3hpg1i7CXFms+hxfMXJHMzBXJVodRp/gVs4lfMdvq\nMMSPRYRV8/wVS+jSMp+3p6Zx4yttA6MUIjQUxoyBZ5+Fxo1NsXmfPioyFxGPUdIhAeGPP8w25TZt\nIH6pGQpYnNnF4qhEJBDERFbxythFtE0v5OXPm3P/pAyrQ3Kfnj3hP/+B0083n9707QsPPADl5VZH\nJiIBRkmH+L3ycpg929RzxMbuU0SulQ4RcZOEmEomXLWIFiklPPxOBm9+m2Z1SO4THQ033wxPPQXJ\nyfDgg2a2x8KFVkcmIgFESYf4vfnzoaQEuuxZ2IhfOpPKmATKGje3NjARCSiNE8p5eewiEqIruOq5\n9vw0P9HqkNyrTx+z6jFsGCxYYL5+9FGoqrI6MhEJAEo6xO/NmGFuu3aF8JxNRG1db4YC2gKg04yI\n+JRWjUt49vIlAJx9X2ecG6MsjsjNYmPh1lvhiScgMRHuuQdOPFGTzEXkiCnpEL83fbq57dIFkuZN\nA6Cw3VEWRhQ82qYX0Ta9yOow6lSSnklJeqbVYUiA6dUmj/vOc7K7MIxhd3VjR16Y1SG5X79+ZtXj\n+OPh11+he3f4/HOroxIRP6aWuQEkGFvmulyQlmZuP/oIOj5xGWlT32bFbRMpbdrG6vBEJICN/zaD\nid9ncEyXPKY9u4CI8AD8+etywddfw/jxZgrrDTfAM8+Y7lciEtTUMleCyurVsH27WeWw4SJp7g9U\nxCZSmqZPt0XEs64Zup4hPbYzY0kCVzzdITBa6f6dzWY6W736KmRkwIsvwtChsHOn1ZGJiJ9R0iF+\nbd96jugsJxE7N1PYrgeBMb3L9/3f3Cb839wmVodRp8S5P5A49werw5AAFRICD12wgm4Zebz3QyoP\nv93K6pA8JyMDXnkFBgyAadNMa92lS62OSkT8iN6ZiV/br55jz5vLgva9LIwouLz8TWte/qa11WHU\nqek3E2n6zUSrw5AAFhFWzfOXL6Fpcgn3T87k/R98Nwk/YjEx8PDDcMklsHYtHH20SUBERBpASYf4\ntRkzTIv51q0h8a8i8p4WRyUiwSQ5roKXxiwmLqqS0U92YMbieKtD8pyQELj8crjvPlPj8Y9/wGef\nWR2ViPgBJR3it3bsgBUroFMnCKWSpPk/UdaoKeWN0q0OTUSCTJu0Yp66bClV1TbOvLcLazZFWh2S\nZw0aBI8/DnY7nHuu6XQlInIQSjrEb800g8fp0gViV84jtCiPgnY9rA1KRILW0Y5d3D1iJTvywhl2\nVzd2FQR4h6deveC55yAuDq680vxZRKQOSjrEb9VWz1HYXlurRMQ6Z/ffwqWDNuLMimbE/Z0prwjw\nIaUdOpiOVo0bwy23wEsvWR2RiPgoJR3it2bMMCv7nTrtHQpYoHoOEbHYDaet5cSuOfw4P4mrn2sf\nmK1099WqlVnlSE42czxef93qiETEB2k4YAAJpuGApaWQkGC6OL7xUgnHnp5EWePmOG9TpyJvyi82\n20fioystjqR29uICAKqi4yyORIJNSXkIV75yFMuy4nnsyrXcddFGq0PyvPXr4cYbIT8fJk+GSy+1\nOiIR8SANB5Sg8OefUF5u5nMkLJlBSEWZWuVaID660mcTDjDJhhIOsUJUeDUvXLGEtMRS7p7Ymo9/\namx1SJ6XkWGmlcfGwujR6molIvtR0iF+qfb5HNpa5W3bdkewbXeE1WHUKWx3DmG7c6wOQ4JU44Ry\nXhqzmJjISi59vCOzlgZwK90abdvCU09BRARcdNHeCa4iEvSUdIhfqvk91qULJP/xLdWhYRS17mZt\nUEFo9Es9GP2S73YMa/fS9bR76Xqrw5Ag1q5pEU9dtpTKKhtn3BMErXTBFJc/8ABUVMDpp5ve5iIS\n9JR0iN+prjZJR3o6tChdRezaxRS07011RJTVoYmIHGBAh13cec5KcnaHM+S27mzNDbc6JM/r29d0\ns9q1C4YOha1b/3rI5TJ1eSISXJR0iN9ZscL8HuvSBVJ++xyAvG7HWRyViEjdRgzYwtgh61m7JYqh\nt3cjr9BudUied+qplFx4OWzYwMauwzhnaJGpw0uAqCho396M93jnHdiwwepgRcTTlHSI3/nlF3Pb\npQukTP8cV4idvM4DrA1KRKQeVw9dz4gBm1i4Jpbh/+pKaXng/gpelR3FdS+2o/F/X2MiV9ByxzzO\n/24Ua9e4SEmB7t1h82YzyPzSS00Nert28NFHBH6LYZEgFeDjUiUQfW4WNzih3SYSnv+dgrY9qIpN\nsDYoEZF62Gxw5zmr2F0Uxg8Lm3DBwx355IFlhNoD5132zCXxPPFBS76e1QiXy0ZaYinLB9/C9kXz\nODf7U3qf9zAbLrsPgKoqWLsWFi6ERYvgjz/g/PPhtdfg5Zehc2eL/zIi4lZKOsSv5ObCTz+BwwEd\nVvwP0NYqEfEf9hB49OLl5BeH8b/pjbnq2fa8cauTED9f9CgssXPHa62Z8EUzALq2yueigVmc2G0H\nYXYX2/rfR8Lz15A5+X6KMjqzY+A52O1mdaNdOxgxAjZtggkTzM/47t1h3DhTjx4fBE2/RIKBhgMG\nkGAYDvjWWzBqFIwZA0/OPYmkedNYev/HVCQGQQ98H/TT4kYADOq60+JIahe/2PRWzu96rMWRiOyv\nqNTOmPFHsTw7jjHDNvPqzSv9NvH4cV4iVzztYP3WKNqkFXH3iJX0bJN3wHmRm9fQ7sXrwWZj/isz\nKGx7VK3XmzULxo83SUjz5vDNN9BNzQlFfM6hDgdU0hFAgiHpGD4cvvoKPhy/k5HXp1Lcoj2rbpxg\ndVgiIodsd1Eo1/y7Oys2xTFq6BYm3urE7kf15QXFdm5/rTWvftkMe4iLUSduZOwp6wkPrfv3UMLi\n6WS+eS+lTVoy99U5VCQ1qfW88nJ4911TZB4fb+YMDh7sob+IiBwWTSSXgFVQAFOnQmYmdN/4Fbbq\nKvK6amuViPinxJhKXrt2IZ1b5jN5SjqXPt6RyqoG//621MLVMfQc24tXv2xG2/RC3rphHtcNW3fQ\nhAMgr+uxbDn1ciK3b6Tz/edgqyiv9bzwcLj8crj3XigpgVNPNQmIiPgvJR3iN779FsrK4Pjj97bK\n3a16DkuNGd+dMeO7Wx1GndqMv4k242+yOgyROsVHV/LvqxfSLSOP96elcuHDHamo9N3Ew+WC/3yT\nxtH/7MnqTdFcNmgj7908l84tCxp8jW0nX8yuHoNIXDyd9i9ce9B2VSeeuHfA+aWXwmOPqbuViL9S\n0iF+47PPzO2JfQtJnvMdJen/3959x0dVpX8c/6TRa+hVkHIQ0BUB+4oVC67YBRcVFRV7wbZrQ2wI\ndgVdVxTUVVTWgigq9oqCrsJP5NBb6CW0ENLm98czkTGGZIDcmUnyfb9e80oyc+/MM2cuzH3uec45\nbclp1DK+QVVyGeuqk7EucRdlrLpuBVXXrYh3GCIlql09n9GXzaDb3pm88UVjzrirC1u2JV6d1dZt\nyQwc3olBD3WiSmoBjw+ayXWnLCi1d+NPkpJY0u9mslp2pNn7Y2jx3ydK3Hz//W02qyZN4LbbbIC5\nEg+R8kdJh5QL27bZYMIWLaBHxjsk525nowYHi0gFUbNaPqMuncFBHdfz7rcNOeyqbixaWS3eYf3u\nt8U1OOiK7rz4UVO6tN7Eq0N+5Iguuz+BRKhKNRZefC+5ddJp//QN1J/2UYnbt2kDTz0F7drZz3/8\nY7dfWkTiREmHlAsffQRbt1ppVfNJzwKwvsfxcY5KRKTsVK9awJOXzuSswzKYsaAWPQcfwFcz4rsG\nUWE5VY/LuvPropqcc/gynr/6fzRPz97j586t14iFF95DKDmVLnefTfUlvsTtGzaEkSOhVSt48EF4\n4IE9DkFEYkhJh5QLhaVVp3ScTb0ZX7K5Y3dyGrWIb1AiImUsLSXEP8+cyz/OnEPm5lSOGfIXnnuv\nWVxi2bglhf73dGbQQ51ISS5gxAW/cusZ83a9nKoEWW06s/TsIaRu3ci+t51C6uYNJW5fvz489JCV\nWv3znza1roiUD0o6JOHl5MDEidC4MRz2q/VyrD3k5DhHJSISnLMPW87owTOoUSWfSx5yDBrpyNwS\nu/V8p86qw/6X9OC1zxqzf9uNjL9xOsftvyaQ19rQszerju5PjWVz6DzsHJLy80rcvnFjSzzS0+Gq\nqzSrVXkWCsGGDbYq/aRJ8Mwz8MortlbLihUau1PRaEVySXiffw6ZmdDv1GyafjiO3Fr12dT1sHiH\nJcCxfwnmJKSsZP6lV7xDENltPTtk8vL1PzLkha6Meb8Z701N58lr5nHGEWtICmiCq8wtqdw9bi+e\nfLMlBSG4pPciLu29mNSUYM/+VvS5mGqrFpM+fQrtRg9h3tWPl7h9y5Y2q9V118GFF9paHn37Bhqi\nlJE1a+D112H8ePj5Z9iyZefbVqtm0+SfeCL07w/duxPYsS/B0+KAFUhFXRzwssvg2Wfhg/P+w/Ev\nDWDV0f1Z8bdL4x2WiEhM5OYn8eJnrXj2wzbk5CVzyqFreeraubRqvL3MXiM/H56f3IzbxrRlTWYV\nWjbYxl39PD3aZ5bZa5QmOTuLDo9fRfWVC/FDnmXFyZeUus+sWXDjjXZF/JNP4NBDYxCo7LKsLHj7\nbfjPf+DDD+14S062CQKaNrXeq8aNbdzO1q2wcqXdVqyApUttMhmA9u0t+Tj3XOjUKa5vSdCK5JVa\nRUw68vOheXPIzYX5rXpRf8aXzPrnyxrPISKVzuI11bnv9Y5Mm1efWtXzuPLU5Qw+ZTltmu7+oO5Q\nCL6aUZfrR7Xnp7m1qV4ln4uPW8yAXsuomlZQhtFHp8ra5XR87AqSs7fyy0Mfs3H/0nsrv//eptKt\nWxe+/Raci0GgEpX8fBg3Dm6/3RIIgI4d4dhjbQ2WBg1Kf46cHJg2DT791D7f7PDhfuaZMGwY7LNP\ncPFLyZR0VGIVMel47z04+WS4/MjfGP15ZzZ37M78yx+Kd1gS9vTkNgBcfuKiuMaxM00nvwDAyhMv\njHMkImUjFIKJPzTl8Xf3ZsPWKiQlhehz8Dqu6Luc43uuJznKkZqLVlbj5SlNeHlKE/zSGgD06b6S\na05eQON6xa8SHis15/1M+6dvJK9WPX58ZhrZzdqWus/kyVZu1aaNjQdo2jT4OKVkU6ZYL9SMGba4\n42mnWZlU69a7/5zbtlniMWECzJ5tvSUDBsDQoVaGJbGlpKMSq2hJRygEhx9u/8HMPO4Guk55lIUX\n3MXG/Y+Md2gSdtKwgwF4/86pcY6keJ2H9QNg1p3j4xyJSNnanpvMRz834o1vWjBzcR0AWjfO5sB9\nNtGlTRZd2myl815baZKey8r1VVi+tgrL11UlY20VPvghna9n1gOgalo+vbqso/8Ry9i/7aZ4vqU/\naPDtu7R64xG2tunCT099S37NOqXuM24cjB0L3brBF19A7drBxyl/NneuLeD4wQc2/uL44+Gii6BR\no7J7jVDIzg2efx4WLIC0NCvFvvde6/GS2FDSUYlVtKTjyy+hVy/ofWAmk2a1BZKYdddrhFLT4h2a\nhCnpEIm/WUtr8fo3Lfh0RiM2byt9fpikpBA92mVyUo9VHLPfGmpXz49BlLuuxZtP0Oirt1jf4zhm\n3j+JUFqVErcPheDhh62HvHdvmw0pTV8XMRMK2fjLG26wMRwHHACXX27jMIJSUACffQYvvAAZGVaO\n/dRT1qsiwdvVpEOzV0nCuv9++3lf+kOkbckk42+XKeEQESmic6stDO3nuescz+qNVVmwsgbzV9Zk\n/sqabMpKpUGdHBrVyaFhne00qptDh2ZbaVKv7AahByWj75VUWb+K9OlTcCMvZvat4yipfiwpCa6/\nHtavtwVlL7wQXnyxxF2kjKxcCYMGWcJXuzbccQccdVTwM00lJ8Mxx9jCwePH2/TJp59utyeftCRE\nEoeSDklIP/5oM1wc3XU13T5/jNw66aw9/NR4hyUikrCSkqBJve00qbedQzqVvMheuZCSwqLz76D9\n6CE0nfIyOQ2as+CyB0vbhTvugJtuspmS0tPh8cc1zWqQ3n4bLrkE1q61KW1vuaVsS6mikZYG551n\n1REPP2wLCn/yia3ncvHF+vwThfJ/SUgPPGA/H0wfTkr2VlYdN4BQlWrxDUpERGIqVKUaCwbdT3bj\nVrQeP4IWE0pevwOgenXrKW/b1q5233NPDAKthLZvt8UZTzsNNm2y30eMiH3CEal1a3j0Uevxysuz\nZOiUU6wnRuJPSYcknNmz7SpFr3bLOOC70Wyv35R1B2sF8kSUXjuH9NrxnemmJLm165Nbu368wxCR\nPZBfqy4LLhtBbp10Ooy6jqbvjSl1nzp17AS4WTO46y6r85eys3ChTfQyapQld//6F5xxRmKUsiUn\nW6Lxwgs2rmTSJNh3X3jrrXhHJhpIXoFUlIHkF15oM5BM7zGY7tP/xZJ+N7P+oBPjHZaIiMRRtRUL\naTfqelKzNjH75hdYdcIFpe6TkWEzKa1fb+VW554bg0AruHffhfPPh8xMm5nquuts5fBEVFBgycaz\nz9p6HwMHWrldndInQ5MoaPaqSqwiJB2LF9tMFwc3XsAXqxw5DZox++YXrFBXREQqtWoZ82g/eggp\n2zYz+x8vsuq4AaXuM2+eldtkZ8Mrr8BZZ8Ug0AooN9cW+RsxAqpUgWuvhZNOindU0Vm8GO67z6bz\n3Wsvm2DgiCPiHVX5t6tJRwJ0hIns8NBDkJcX4rlqV5Kcn8fKEwYq4UhgU319pvrELV+q5X+klv8x\n3mGISBnJbtGe+YNHkl+tJp2GX0CTKS+Xuk/79jZOsEoV6NfPTjhl12Rk2AriI0ZAy5ZWVlVeEg6w\nRGPUKBtsvnQpHHkk3HyzjUuR2FHSIQnjl1/g3/+GW2uPxi34gE2depK5/1HxDktKMOw1x7DXXLzD\n2KnWr42k9Wsj4x2GiJShba06Mn/wQ+RXrcE+959Hy9cfKXWfrl1tVqOaNeGCC2wMgkTn449twcWv\nv7bZoZ55Jti1N4KSlmaLFD7xhE2lO3Ik9OxpK6ZLbCjpkISwZQucfTa02T6bYdtuJK9GHZb0u1nz\n3ImIyJ9sa+2Yd9Xj5NRtSPunh7D3MzdZAX8JOnWymY3q1YPBg+132bn8fLj7bltoccMGGxtz112W\nuJVnXbrYBc6//Q1mzrTE4777rHxMgqWkQxLClVfCwjk5vFd/AGl52Sw9+wby6jaMd1giIpKgspvv\nzdxrniS7cWtav/YQnYZfQFJOyfUy7drZQOKGDW3l7GHDbCVt+aNly2yQ+NCh0Lix9Q6cdlrFuQ5Y\nvbp9/vffb4sZ3n479OgB06fHO7KKTUmHxN24cVZj+0SDYbTb8CPrex7Pxr/0indYIiKS4HLTmzL3\nmifYuldnmk55mW7X/pWqKxeXuE/r1vDYY9C0qV25P+cc2Lw5RgEnuFDIZvnq2tUW1zv0UJv5aZ99\n4h1ZMA45xGbL7NPHyqwOOsjGemRlxTuyiklJh8TVb7/BFVfAeVVf59L1D7A9vSnLTr863mGJiEg5\nkV+zLvOueJj1PY+nzuxp9Lj0AOr/8GGJ+7RoYQOL99sP3ngDDj4YvI9RwAlq7Vorcx4wwKaXvfFG\nuPfeij+9bK1a9l4fftgS0ZEj7biYOFG9YGVNSYfEzbZt9h/ccVlvMzbnXEJVq7PogqEUVCvnBaMi\nIhJToSrVWNL/FpaedQMp2zaz360n0nbM7SWWW6Wn24nmGWfArFlW2//22zEMOkGEQrYgb9euMGGC\nLaT33HN29b+ilFNF44ADYMwY6/latAj69oXjjtNA87KkdToqkPK0Tsf27bZIz8bx7/NO0qkkp6Uy\nf/AIstp2jXdosgsWrqoBQNsmidkXXXXVEgC2N2kd50hEJFaqL/G0GTuUqhtWsnWvzvibnmNTl0NK\n3GfKFEtAtm+3Wv+777Yr4BXdzJm2uN+nn+6Y3emsszRT/eLF8PTT8P33tsL5oEFwzz02vkV20OKA\nlVh5STpWrrQrS3W+nczbSaeRmgoLLh3O1vb7xzs0ERGpAJKzt9J80rM0/GYioaQkMk69ioUX3UN+\nrbo73WfePBs4nZFha1E8+qh9V1XEq/3r1sGdd9r0twUFVl52+eU23kV2+OEHGD3akpAaNeDSS2HI\nEDs+RElHpVYeko7p0+GsvjkMXn4HNzESUtNYMOg+trge8Q5NdkNunv1fk5aamMddUp7NgRhKTYtz\nJCISDzUXzKTVayOptnopubXTWdL/FjJOu4qCajWK3b5w1fLx420K1d694cknoWPHGAcekLVr7Qr+\no4/aNLitWtnskQcdFO/IEld+Prz7rh0Xa9ZYj9B559mAc5e4y1TFhJKOSizRk46XX4YHL57DCznn\n0oMfyW7YgiUDbiNrrwo6LUYlcNKwgwF4/86pcY6keJ2H9QNg1p3j4xyJiMRLUm4Ojb6YQONPXyV1\n2xa2pzdlSf9bWXniheTXLH6U9LJlNrXu9Om2kvlFF9k6FeV1Fqd58yzReOEFG09Zq5YtknjqqZCa\nGu/oyofcXFso8dVXbVXzpCRblX3gQFvzo2rVeEcYe0o6KrFETTr+9z8YPXQ1e098lGt4gppkse7A\nE8g47eqdXm2S8kFJh4iUFynbttDos9dp9MUEUnK2kV+tJquOOZflfS9nS4duf9o+FIIvv7QSpJUr\n7b7jj4drr7WfyQk+Fc+2bfDBB/DSSzZAPhSCJk3gzDPtZLmGvn53S36+rc4+fjzMnm331a8P/fpZ\nInfggRWzJK84SjoqsURKOkIhG5g2dugien79CJfwb6qTTXbNdFaecSWZ3Y6Od4hSBpR0iEh5k7Il\nkwbfTaLBd+9RdYNlE1vadmXdYX1Ze/ipbO7Y/Q9njfn58M038N//7pjJqGNHO3nv08dKkxJl4HVW\nFkyebNMAT5oEW7fa/R062Elxr16JE2tFsHAhfPihTUSwfr3d17y5leX17m2zXzWswOscK+lIQM65\nE4DHgRTgOe/98CKPVwVeBLoD64BzvPeLdvV14p10ZGfD998V4CfMpOCtd+i54h268xMAm2s1I7P3\nWaw/6CRCVSphH2QFpaRDRMqtgnxqz55Gw+8mUXv2NJLzcgDY3qA5G/f7Kxu7HMqmroeyZe/9CKVV\nAWDuXEs+PvvM1rIAm3r3hBPsJLNbN+jUyUqyYiEjA777zm7ffgs//bQjrubN4cgj4YgjLEmqLFff\n4yE/30rxPv4Ypk2DjRvt/qQkOyYOPNCm5O3WzaYmrlYtvvGWFSUdCcY5lwLMAY4DlgHTgP7e+1kR\n21wB7Oe9H+yc6wec5r0/Z1dfKx5JR0F+iHcvfov8KZ/RZMXP7Bv6hTrY0q55pLKmdXdyDzuSDQcc\no8G8FZCSDhGpCJK3b6O2n0bdmd9Qe/YPpG3J/P2xgpRUspu3I6t1J7JadyK7cWs2127OjLXN+Xpu\nE774uS4L19WmAOtCSEuzsR/77WcDjZs1DdGicS5NW6bSrEUy9epZUlJSEhAKWXnU+vU209T69bBi\nBcyfb+MzCm+rV+/YJyUF2rWDHj0s2WjfXolGPBQU2GczbZolIv/3f5CXt+Px1FQ7Ltq2hTZtdtxa\ntLAENj0d6tUrH2NtdjXpKAdvqdw7EJjnvV8A4JwbD/QFZkVs0xcYGv59AvCUcy7Je5/wGWHmgvX0\nHXcGAPkks6JGOzKa9aLggB7kd+tOQfVKMNG5iIiUawVVq7NxvyPYuN8REApRZd1yai78lZqLfqX6\n8vlUXbWYGks9fPPO7/t0By6MeI7tqTXISapKXkEyOTNSYAZUZxs1yCKVfL7gCLrxBWDJQPXqO26h\nkA1Uzsuzn9u37+ixKE5Kio3POPRQ6NIFOne2E9nq1YNpH4lecrL1LHXsCH//u32OCxdaIjJnjv1c\nuBB+/bXk56lTx8bdpKX98bZ9u1WWbNtmtxo1rPyvQ4fYvL89oaQjeC2ApRF/LwOKTk73+zbe+zzn\n3EagAbA2JhHugfQODVj8wqfkrt1Ilc7tCVWthsamVR7XDNwEQP19E3PS8s0Dr1TkH0AAABTJSURB\nVAYSNz4RSVSt4KiD2ApsBQiFSM5cR9rSBaSsXUXKutWkrltN8oa1JGdtJTlrC0lZW6ial0fVgnxq\nFhSQlwvZSY1YlVSDrdRgeZM+HN/YxllkZ//xlpJisx+lptrvaWlQt65d8a5Xz35v0MDW0dhrLyud\nSlPxQLmx995wzDE7/g6FYNMmmyUtI8Nua9ZYWVZmpv3cuNGOjdxcO2Zyc62MKy3NyrPq1rXEs3lz\nO0bKAyUdwSuu26loD0Y02/yBc24ocNduxlSm9hp4VLxDkDi59ff/RBN0mdZjbgMsgxcR2TMdgIOj\n3roK/OEiXEegfxlHJOVb9+7xjiC2lHQEbxnQKuLvlsDynWyzzDmXCtQF1pf0pN77oewoyQJsTMce\nxioiIiIiUuaUdARvGtDBOdcWyAD6AecW2WYicAHwHXAm8Gl5GM8h0s/GaTM+UcdpJ3yAIiIilYOS\njoCFx2hcBXyITZn7vPf+V+fcMGC6934iMAZ4yTk3D+vh6Be/iEWiNzUxJ63aIeEDFBERqRw0ZW4F\nEu91OqTyadPGfi5aFM8oSpDwAYqIiJRPuzplbnKQwYiIiIiIiCjpEBERERGRQCnpEBERERGRQGkg\nuYjstr/+Nd4RlCLhAxQREakcNJC8AtFAchERERGJBQ0kFxERERGRhKKkQ0R226hRdktYCR+giIhI\n5aDyqgpE5VUSawm/DEbCBygiIlI+qbxKREREREQSipIOEREREREJlJIOEREREREJlJIOEREREREJ\nlJIOEREREREJlGavqkA0e5WIiIiIxIJmrxIRERERkYSipENERERERAKlpENERERERAKlpENERERE\nRAKlpENERERERAKlpENERERERAKlpENERERERAKlpENERERERAKlpENERERERAKlpENERERERAKl\npENERERERAKlpENERERERAKlpENERERERAKlpENERERERAKlpENERERERAKlpENERERERAKlpENE\nRERERAKlpENERERERAKVGu8ApGwlJSXFOwQRERERkT9ICoVC8Y5B4sA5F/LeK0PZTWq/Pac23HNq\nwz2j9ttzasM9pzbcM2q/PRerNlR5lYiIiIiIBEpJh4iIiIiIBEpJh4iIiIiIBEpJR+V1d7wDKOfU\nfntObbjn1IZ7Ru2359SGe05tuGfUfnsuJm2ogeQiIiIiIhIo9XSIiIiIiEiglHSIiIiIiEiglHSI\niIiIiEiglHSIiIiIiEiglHSIiIiIiEiglHSIiIiIiEigUuMdgMSGcy4deA1oAywCzvbebyiyzf7A\n00AdIB+4z3v/WmwjTSzOuROAx4EU4Dnv/fAij1cFXgS6A+uAc7z3i2IdZyKLog1vAAYBecAa4CLv\n/eKYB5qgSmu/iO3OBN4Aenrvp8cwxIQXTRs6584GhgIh4Bfv/bkxDTLBRfHvuDUwDqgX3uZW7/37\nMQ80QTnnngdOBlZ777sW83gS1r4nAVnAQO/9T7GNMrFF0YZ/B24J/7kFuNx7/0sMQ0xopbVfxHY9\nganY+cyEsoxBPR2Vx63AJ977DsAn4b+LygLO9953AU4AHnPO1YthjAnFOZcCjAJOBDoD/Z1znYts\ndjGwwXvfHngUeDC2USa2KNvwf0AP7/1+wARgRGyjTFxRth/OudrANcD3sY0w8UXThs65DsA/gMPC\n//9dF/NAE1iUx+HtwOve+25AP2B0bKNMeGOx79WdORHoEL5dil0AlD8aS8ltuBDoFf4uuQd4NhZB\nlSNjKbn9Cv+tPwh8GEQASjoqj77YVSjCP08tuoH3fo73fm749+XAaqBRzCJMPAcC87z3C7z3OcB4\nrB0jRbbrBOCY8BUrMaW2off+M+99VvjPqUDLGMeYyKI5BsG+YEcA2bEMrpyIpg0vAUYV9v5671fH\nOMZEF00bhrBecoC6wPIYxpfwvPdfAutL2KQv8KL3PuS9nwrUc841i0105UNpbei9/zaigkPfJUVE\ncQwCXA38Fzv/K3NKOiqPJt77FQDhn41L2tg5dyBQBZgfg9gSVQtgacTfy8L3FbuN9z4P2Ag0iEl0\n5UM0bRjpYmByoBGVL6W2n3OuG9DKez8ploGVI9Ecgx2Bjs65b5xzU8OlRLJDNG04FBjgnFsGvI+d\nvEj0dvX/SimZvkt2kXOuBXAa8ExQr6ExHRWIc+5joGkxD922i8/TDHgJuMB7X1AWsZVTxfVYhHZj\nm8os6vZxzg0AegC9Ao2ofCmx/ZxzyVhZ38BYBVQORXMMpmJlLUdiV0e/cs519d5nBhxbeRFNG/YH\nxnrvH3bOHQK8FG7Dyvwdsiv0XVJGnHNHYUnH4fGOpZx5DLjFe5/vnAvkBZR0VCDe+2N39phzbpVz\nrpn3fkU4qSi268w5Vwd4D7g93MVbmS0DWkX83ZI/lwwUbrPMOZeKlRWU1n1ZmUTThjjnjsWS417e\n++0xiq08KK39agNdgc/DXxJNgYnOuVM0mPx30f47nuq9zwUWOuc8loRMi02ICS+aNryYcL249/47\n51w1oCEBlWlUQFH9Xyklc87tBzwHnOi9XxfveMqZHsD48HdJQ+Ak51ye9/7tsnoBJR2Vx0TgAmB4\n+Oc7RTdwzlUB3sLqSt+IbXgJaRrQwTnXFsjABkcWndGmsF2/A84EPvXe6+rUDqW2Ybg86F/ACaql\n/5MS2897vxH7cgDAOfc5cKMSjj+I5t/x24Sv1DvnGmLlVgtiGmVii6YNlwDHYG24D1ANm41OojMR\nuMo5Nx44CNhYWBIt0QnPoPYmcJ73fk684ylvvPdtC393zo0FJpVlwgFKOiqT4cDrzrmLsS+HswCc\ncz2Awd77QcDZwBFAA+fcwPB+A733P8ch3rjz3uc5567CZnFIAZ733v/qnBsGTPfeTwTGYGUE87Ae\njn7xizjxRNmGI4FawBvhKyxLvPenxC3oBBJl+0kJomzDD4HezrlZ2HThN+kq6Q5RtuEQ4N/Oueux\nsqCBugCzg3PuVax8r2F43MtdQBqA9/4ZbBzMScA8bCbJC+MTaeKKog3vxMZUjg5/l+R573vEJ9rE\nE0X7BS4pFNL/CSIiIiIiEhzNXiUiIiIiIoFS0iEiIiIiIoFS0iEiIiIiIoFS0iEiIiIiIoFS0iEi\nIiIiIoHSlLkiInvIORfNNIBHee8/DzqW0jjnpgLzvPcDitx/BDYlagPgfOBpIC08XWo1YBtwiff+\nuRKeeyXQJPxnDrAW+Bl4GRgfOYWqc244MMB73zLKuA8Fjvbe3xvdO00s4fUXWnrvE3aV5PAU6tOA\nQ3a2OKxzrhPw206e4j3v/clBxSci5ZuSDhGRPXdIxO/VgU+Be4H3Iu6fFdOIdl0f4DPvfVZ4jvvd\nNRZb7DEVaAacCLwI9HfOne69zwtvNwp4bRee91DgdqxdJf6u4c8rtm+IRyAiUj4o6RAR2UORV4Wd\nc7XCv87f2dXiBNUH693YUxlF3vcbzrk3sRWXhwAPAnjvlwJLy+D1JD5+25Xj2zlXBVusraDI/UlA\nFe/99t0JwjlX3Xu/bXf2FZHYUtIhIhIDzrkmwDKspOi1iPtTgCXAOO/9PwvLjsK3x4BOwK/AlUVP\n8pxzl2NXnPcGlgOPe+8f243Y9gK6YKsilznv/STn3ETgcsJJR9HyqnAJ13DgDKAxVpo1FTgLuBRb\nuT6ylO1D7/0Jzrl9sZWIDwHSgQVY8jS6sJzLOXcCMBk4HLgR6A2sBB4oWi7mnDsaGAp0B3KBn4Br\nvfczw4+3BUYAx2Gr+X4OXOO9nx9NWzjnzgLuA1oD32Mla3MiHr81/J47YCtTTwWu994vjNjmSOB+\nYF+gAJgP3O29fydim1KPDefctcBNQH3gI+Df0byHKN/nVGx17alYstkaaOqcG4Id2+cBDwNdgb9j\nyWkH4BFs1eQQ8AlwQ+F7jyjzuxr7d9EP+xy7llXcIhIcDSQXEYkB7/0qYBJwYZGHegPNsbKkQnWA\nF4AnsRPQbOAD51yDwg2cc3dgScnrWC/FGGCEc27QboTXB5gVeWIbgCnAXs65pjt5/E4s4fgndkJ/\nA3bSnQS8ibVFDpZcHAJcF96vJTATGAychLXbgxGPR3oeO9E/FfgO+Ldz7i+FDzrnjsdOvjdjJ8X9\nw9s3Cz/eGPgGaAMMCj/eEPgofCW/NB2wZOFO7ES7Mfa5pkVs0wJ4HDgFuAyoAXztnKsZjqEB8C5W\nrncacDbwKpY4FL6PUo8N59w54W3eBE4H5gLPRvEeCiU751KL3JKKbHMMNj5oCNAX+zwB6gLPAaOB\nE4D/OecKyxL3Bi4CLgY6A5875+oWed7bgHpY8nLjLsQsInGkng4RkdgZA7zjnGvhvc8I33ch8E3k\n1W6gNjDQe/8mgHPuK6wU6WpgqHMuHTs5v9N7/2B4n4+dc3WwE9qdDvbeiT78cfxJEJaFfzbBrk4X\ndSDwovf+pYj7CnuEVjvnlgChor093vvJWC9GYanO19hJ7SXAo0VeY5z3fnh426+Ak7ET91/Cjw/H\nkoyTIwa9T47Y/ybsYt2x3vuN4ef5DliIJSljSmoALMk4yXv/Y3jfGcBsLAEZG34/VxduHO4F+wTr\n9emDJRH7ADWxnq/CkqQPI/aJ9ti4DXjbe39N4XM455phJ/LR+LCY+/6BtWGh2sCJ3vt1EfEB1ALO\n9N5Hxn0ddmwcGi69wzn3IzAHS0IiP8tFRSdCEJHEp6RDRCR2JmMn3OcDDzjn6mNXtK8ssl0+NgYC\nAO99pnPuU+zEHOCvQDWsJCXy//FPgJucc03CPSulCl9hPgorGQpS0avgRf0MXOKcW4+VTv1fNE/q\nnKuBnUD3x3o9CnsN8orZ/KPCX7z32c65BeF9CH8W+wOXRs6yVcSxwAfA1oh234AlLT2AMeFE4ff3\nGjFwHmBJYcIRfmyuc+7/sM91bDiOw4Fh4VjqR+zbMfxzDtbzNd459zzwZWECFFbqsQFsxEqSHiny\n/t4k+qTjSuCHIvctK/L31MiEI0Iu1vMV6cDw9r+P8/HeL3DOTcPK4iKTjqATZBEJgMqrRERixHuf\nD4wDBobvOhc7OX69yKYbipysAqwmXOaDlfSA1fLnRtw+CN/fahfCOhorW/pmF/bZHS3CP3eWDBVe\nhb8WmOmcWxIel1CaR7EeoFHYTFk9sfEfqUVOugEyi/ydg52gg00VDLCihNdqCFzAH9s8F5tZq7DN\nv4l8zDl3cMT+q4t5zt8/V+dcO6wHYTtWvnVY+P1sLIzTe78aOB7rLfgvsMY5NzE8LqcwRij52GiK\nJUZF4ykuvp2Z472fXuRWtAdrZ5/1mqIDyrE2KG77VdhYnWieV0QSmHo6RERiawxwa3jdiYHABO/9\n5iLb1HfOpRZJPBqz44R4ffhnb4qfpnRn6ygUpw/Ws1Bcz0BZ6o2VxRRXWoX3PgsrC/qnsxqcq4DR\nzrnfSlnf5EzgEe/9w4V3OOfO2I34Cq/INythm/XYwOgHi3mssLdhIJYQFIr8LBoXs19jbDwF2GeR\nApxaWDoV7omqE7mD9/4r4LjwOI/jsMRrHDYAO5pjIx8bqF00nuLi2xM76zEq7v4V7EhMIzXBBsJH\n87wiksCUdIiIxJD3fr5z7gus9r0HxQ+ETcHKrgrHdNTFeiQKT6y/xq7SN/XeFy1T2VV9gDv28DlK\n5Jw7GfgbcGs023vvvXPueuAKwoOJsfeb5pxLLrxKHh7DUR3rGSh8rVRscPUu8d5vcM79D+vJ2Nks\nTp9gvSkzvPc5O3me2SW8TGvnXPeIMR0dsDKnwtKh6lhCkB+xT392Uprmvd8KvO2c64bNDAZRHhvO\nuV+xwd1jI+4+vYTYg/Y9MDJyvJNzrg3W03NTHOMSkTKipENEJPbGAC9hJTBfFvP4ZuCR8DiD1djJ\negHwFID3fo1z7j7gaedce+xEMxVw2EDcqE66nXNdsTENk0vbdhe0CJcUpbBjccDzsZm7io4hiIzl\nPaw06WcsieiHnXx/Fd5kNlYSfK1z7msgMzwm4mPguvBA841YeVZp40d25hZgsnPuXewzysZKnL7y\n3n+EjXvpB3zinBuFXZ1vivUwfOy9/28pz78aG4txB1bydC82XfIr4cc/AR7Axoa8CPwFm/Z2S+ET\nOOdOD8fwDjaGohU20PpT2KVj437gFefcE9hncyw2tida+zjnthS5L8t7P2MXniPSs1gC/oFz7m7s\nMxwGZFD6AH0RKQc0pkNEJPYmYiUiY3cyaHkTdiJ5HTABmzb1RO/92sINvPfDsLEMp2BTqP4HOIcd\nJ+nR6ANM896v2Z03sRMDseloPwOewE7Kzwf6llLC9Q1WKjUeeAvrATi1cH0M4GNsKtlbsKviT4bv\nH4wNaH4G66H4gR09Qrsk3DNwPDaG4NXw7RDC5T3h0rCDsNmqnsDGXwzHPp9oBr7PZceq6q8Aa7DP\nNSf8/NOxWbeOwBKBM7Deh60RzzEHSyIexAbGP4AdT5dFvI9Sjw3v/avYVLZnAm9j6178/hxReAL7\nnCNvr5S4RwnCC/wdDSzGel+eBzxwpPd+0+4+r4gkjqRQSKWRIiKxFL5a/QbQJnK2nvBjf1g0L+A4\nvgSmeO/vCfq1RESkclN5lYhIjDjnWmBTn94LvFU04Yg17/0R8Xx9ERGpPFReJSISO1dj6xNkUvyK\n2SIiIhWSyqtERERERCRQ6ukQEREREZFAKekQEREREZFAKekQEREREZFAKekQEREREZFAKekQERER\nEZFAKekQEREREZFAKekQEREREZFAKekQEREREZFAKekQEREREZFAKekQEREREZFAKekQEREREZFA\nKekQEREREZFAKekQEREREZFAKekQEREREZFAKekQEREREZFAKekQEREREZFAKekQEREREZFAKekQ\nEREREZFAKekQEREREZFAKekQEREREZFAKekQEREREZFAKekQEREREZFAKekQEREREZFAKekQERER\nEZFAKekQEREREZFAKekQEREREZFAKekQEREREZFAKekQEREREZFAKekQEREREZFAKekQEREREZFA\nKekQEREREZFAKekQEREREZFAKekQEREREZFAKekQEREREZFAKekQEREREZFAKekQEREREZFAKekQ\nEREREZFAKekQEREREZFA/T/aqcPyeDS1iAAAAABJRU5ErkJggg==\n",
"text/plain": [
"\n",
"