From a515caa7f2a8bee0434f5f7592f655b3b2d9ff1b Mon Sep 17 00:00:00 2001 From: Radev Date: Fri, 24 May 2024 12:44:02 -0400 Subject: [PATCH] Add working notebook with coupling flows [will later add splines as well] --- examples/TwoMoons_Bimodal_Posterior.ipynb | 750 +++++++++++++++++++++- 1 file changed, 749 insertions(+), 1 deletion(-) diff --git a/examples/TwoMoons_Bimodal_Posterior.ipynb b/examples/TwoMoons_Bimodal_Posterior.ipynb index 576aa8b07..352095b8c 100644 --- a/examples/TwoMoons_Bimodal_Posterior.ipynb +++ b/examples/TwoMoons_Bimodal_Posterior.ipynb @@ -1 +1,749 @@ -{"cells":[{"cell_type":"markdown","id":"009b6adf","metadata":{},"source":["# Two Moons: Tackling Bimodal Posteriors"]},{"cell_type":"markdown","id":"3ed81254","metadata":{},"source":["## Table of Contents\n"," * [Inference Network and Amortizer](#inference_network_and)\n"," * [Trainer](#trainer)\n"," * [Validation](#validation)\n","\t * [Global Calibration](#global_calibration)\n","\t * [Two Moons Posterior](#two_moons_posterior)\n"," * [Further Experimentation](#further_experimentation)"]},{"cell_type":"code","execution_count":2,"id":"d5f88a59","metadata":{},"outputs":[],"source":["import matplotlib.pyplot as plt\n","import numpy as np\n","import seaborn as sns\n","from matplotlib import cm\n","\n","np.set_printoptions(suppress=True)\n","\n","import bayesflow as bf\n","from bayesflow import benchmarks"]},{"cell_type":"markdown","id":"9525ffd7","metadata":{},"source":["This example will demonstrate amortized estimation of a somewhat strange Bayesian model, whose posterior evaluated at the origin $x = (0, 0)$ of the \"data\" will resemble two crescent moons. The forward process is a noisy non-linear transformation on a 2D plane:\n","\n","$$\n","\\begin{align}\n","x_1 &= -|\\theta_1 + \\theta_2|/\\sqrt{2} + r \\cos(\\alpha) + 0.25\\\\\n","x_2 &= (-\\theta_1 + \\theta_2)/\\sqrt{2} + r\\sin{\\alpha}\n","\\end{align}\n","$$\n","\n","with $x = (x_1, x_2)$ playing the role of \"observables\", $\\alpha \\sim \\text{Uniform}(-\\pi/2, \\pi/2)$, $r \\sim \\text{Normal}(0.1, 0.01)$, and a prior over the 2D parameter vector $\\theta = (\\theta_1, \\theta_2)$:\n","\n","$$\n","\\begin{align}\n","\\theta_1, \\theta_2 \\sim \\text{Uniform}(-1, 1)\n","\\end{align}\n","$$\n","\n","This method is typically used for benchmarking simulation-based inference (SBI) methods (see https://arxiv.org/pdf/2101.04653) and any method for amortized Bayesian inference should be capable of recovering the two moons posterior *without* using a gazillion of simulations. Note, that this is a considerably harder task than modeling the common unconditional two moons data set used often in the context of normalizing flows.\n","\n","The two moons generative model, along with all benchmarks from https://arxiv.org/pdf/2101.04653, exists as a standalone object in the `bayesflow.benchmarks` module. So let's import it using the `Benchmark` helper class."]},{"cell_type":"code","execution_count":3,"id":"0b9a9817","metadata":{},"outputs":[{"name":"stderr","output_type":"stream","text":["INFO:root:Performing 2 pilot runs with the two_moons model...\n","INFO:root:Shape of parameter batch after 2 pilot simulations: (batch_size = 2, 2)\n","INFO:root:Shape of simulation batch after 2 pilot simulations: (batch_size = 2, 2)\n","INFO:root:No optional prior non-batchable context provided.\n","INFO:root:No optional prior batchable context provided.\n","INFO:root:No optional simulation non-batchable context provided.\n","INFO:root:No optional simulation batchable context provided.\n"]}],"source":["benchmark = benchmarks.Benchmark(\"two_moons\", mode=\"posterior\")"]},{"cell_type":"markdown","id":"2d4c6eb0","metadata":{},"source":["## Inference Network and Amortizer \n","We will use a neural spline flow (https://arxiv.org/abs/1906.04032) for modeling the posterior, as these are specialized for locally weird posteriors. By default, some weight regularization and dropout will be applied during training. These can be modified through the `coupling_settings` keyword of the `InvertibleNetwork`."]},{"cell_type":"code","execution_count":5,"id":"5c8b3187","metadata":{},"outputs":[],"source":["inference_net = bf.networks.InvertibleNetwork(\n"," num_params=2, coupling_design=\"spline\", num_coupling_layers=4, permutation=\"learnable\"\n",")\n","\n","amortizer = bf.amortizers.AmortizedPosterior(inference_net)"]},{"cell_type":"markdown","id":"86117817","metadata":{},"source":["## Trainer \n","The `benchmark` object already contains the two moons generative model to simulate pairs of $(\\theta, x)$ and the configurator to arrange them into the corresponding dictionary data structure expected by the `amortizer`."]},{"cell_type":"code","execution_count":7,"id":"d09fcc90","metadata":{"code_folding":[]},"outputs":[{"name":"stderr","output_type":"stream","text":["INFO:root:Performing a consistency check with provided components...\n","INFO:root:Done.\n"]}],"source":["trainer = bf.trainers.Trainer(\n"," amortizer=amortizer, configurator=benchmark.configurator, generative_model=benchmark.generative_model, memory=False\n",")"]},{"cell_type":"markdown","id":"16a225fd","metadata":{},"source":["You have probably noticed, that we are not using any summary network, since the two moons \"observables\" are single 2D vectors. This is the reason why the simulator outputs will be going into the `direct_conditions` key of the configured output.\n","\n","To make the problem more challenging, we will generate $10000$ simulations from the model, which puts us into the \"mid-range\" of training data availability. \n","\n","Depending on your machine, the below training should take between $3 - 5$ minutes."]},{"cell_type":"code","execution_count":8,"id":"dee531b4","metadata":{},"outputs":[],"source":["offline_data = benchmark.generative_model(10000)"]},{"cell_type":"code","execution_count":10,"id":"ed932905","metadata":{"scrolled":true},"outputs":[],"source":["%%time\n","history = trainer.train_offline(offline_data, epochs=50, batch_size=32, validation_sims=200)"]},{"cell_type":"code","execution_count":11,"id":"7e28d37f","metadata":{},"outputs":[{"data":{"image/png":"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","text/plain":["
"]},"metadata":{},"output_type":"display_data"}],"source":["fig = bf.diagnostics.plot_losses(history[\"train_losses\"], history[\"val_losses\"])"]},{"cell_type":"markdown","id":"b90a6062","metadata":{},"source":["## Validation \n","We can use simulation-based calibration(SBC) for free (due to amortization) checking of computational faithfulness.\n","\n","1. Talts, S., Betancourt, M., Simpson, D., Vehtari, A., & Gelman, A. (2018). Validating Bayesian inference algorithms with simulation-based calibration. arXiv preprint arXiv:1804.06788.\n","2. Säilynoja, T., Bürkner, P. C., & Vehtari, A. (2022). Graphical test for discrete uniformity and its applications in goodness-of-fit evaluation and multiple sample comparison. Statistics and Computing, 32(2), 32."]},{"cell_type":"code","execution_count":12,"id":"45b1b211","metadata":{},"outputs":[{"name":"stdout","output_type":"stream","text":["CPU times: user 135 ms, sys: 15.3 ms, total: 150 ms\n","Wall time: 145 ms\n"]}],"source":["%%time\n","new_sims = trainer.configurator(trainer.generative_model(500))\n","posterior_draws = amortizer.sample(new_sims, n_samples=250)"]},{"cell_type":"markdown","id":"aa6f6be0","metadata":{},"source":["### Global Calibration \n","For a good calibration, the ECDF trajectories should remain within the simultanoeus confidence bands."]},{"cell_type":"code","execution_count":13,"id":"f76289b3","metadata":{},"outputs":[{"data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAA90AAAHqCAYAAAAZLi26AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOzdd3gbVdYH4N/MqFqy3O24pFfSCSkEUoHAhl6SAEtdYIGFZYFAaEtJWCBAgADLwkfJBgjs0mtYOukJhHQgvbh3S7aaVWe+P+wZ1C3L6j7v8/iJPZqRribS3Dm3nMsIgiCAEEIIIYQQQgghUccmugCEEEIIIYQQQki6oqCbEEIIIYQQQgiJEQq6CSGEEEIIIYSQGKGgmxBCCCGEEEIIiREKugkhhBBCCCGEkBihoJsQQgghhBBCCIkRCroJIYQQQgghhJAYoaCbEEIIIYQQQgiJEQq6CSGEEEIIIYSQGKGgmxBCCCGEEEIIiREKugkhSWP37t249tpr0b9/f6hUKvTr1w9/+ctf0NDQkOiiEUIIIWmL6l9CYosRBEFIdCEIIWTp0qV48MEH4XK5IJfLoVQqYTabAQD9+/fHtm3bkJ+fn+BSEkIIIemF6l9CYo96ugkhCbdkyRLcd9990Ol0ePPNN2GxWGAymfDll18iMzMTFRUVuO+++xJdTEIIISStUP1LSHxQTzchJKG+/vprzJ07FxqNBhs3bsS4ceO8Hn/iiSdwzz33QKfTobGxEUqlMkElJYQQQtIH1b+ExA/1dBNCEsbpdOLmm2+GIAh46KGH/Cp8APjDH/4AADAajThw4EC8i0gIIYSkHap/CYkvCroJIQmzatUqHDlyBHl5ebj55psD7pOXlyf9XlNTE6+iEUIIIWmL6l9C4ouCbkJIwrz22msAgMsvvxxqtTrgPk6nU/pdJpPFpVyEEEJIOqP6l5D4oqCbEJIQFRUV2LJlCwBg/vz5QfdraWmRftfpdDEvFyGEEJLOqP4lJP6o2YoQkhA//PCD9PuFF14IhmEC7ufZ0j506NCYl4sQQghJZ1T/EhJ/FHQTQhJi48aN0u+NjY1d7l9UVITc3NxYFokQQghJe1T/EhJ/NLycEJIQe/fuBQAsXLgQgiAE/Zk+fToAYNq0adKxZrMZixcvxtlnn40+ffqAYRhcffXViXgbhBBCSErpSf27Y8cOLFy4EOPGjYNOp0NBQQFmzJiBTz75JBFvhZCUQUE3ISQhjh49CgDo27dv0H3MZjN+/PFHAMApp5wibW9ubsaSJUuwY8cOTJw4MbYFJYQQQtJIT+rfJ598Em+88QYmT56MZcuW4e9//zvsdjsuuOACPPjgg7EtOCEpjIaXE0ISwmg0AgCysrKC7vPZZ5/B6XRCJpNh3rx50vbi4mJUV1ejtLQUNpstaOZVQgghhHjrSf17yy234PXXX4dKpfLaNm3aNCxduhS33XYbDUUnJADq6SaEJIQgCAAQNIEL8PuSJueccw4KCwul7UqlEqWlpbEtICGEEJKGelL/nnzyyV4BNwBwHIcLL7wQLpcLBw8ejEGJCUl9FHQTQhKipKQEAFBXVxfw8fXr12PNmjVgGAaLFy+OY8kIIYSQ9BWL+re2thYAUFBQEJUyEpJuKOgmhCTEpEmTAHQMYfPV0tKCa6+9FgDwt7/9DWPHjo1r2QghhJB0Fe36t6amBitXrsSUKVMwePDg6BaWkDRBQTchJCGuuuoqAMCPP/4oJWIBgA0bNmD69Ok4fPgwTj75ZDzxxBOJLCYhhBCSVqJZ/1qtVlxwwQWw2+145ZVXYlpuQlIZI4gTOwghJM7++Mc/4r///S8AQC6XQy6Xw2q1AgDOPPNMvPvuu9BqtSGfQ0ykdtVVV+H111+PdZEJIYSQlBeN+tfhcODcc8/Fd999h/feew8XXnhhzMtNSKqinm5CSMK89dZbeOqppzBq1CjIZDKo1WqcfvrpeO+99/DFF190WeETQgghpPt6Wv86nU4sWLAA33zzDVasWEEBNyFdoJ5uQkhKo55uQgghJH7cbjcuueQSfPDBB3jppZdw4403JrpIhCQ96ukmhBBCCCGEdInneVx11VX44IMPsHz5cgq4CQmTLNEFIISQSLzwwgtobW2Fy+UCAOzZswePPPIIAGDGjBmYMWNGIotHCCGEpJ1Fixbh7bffxtSpU5Gfn4+33nrL6/GTTjoJgwYNSlDpCEleNLycEJKSBgwYgIqKioCPPfTQQ7S2NyGEEBJls2bNwrp164I+vnLlSlx99dXxKxAhKYKCbkIIIYQQQgghJEZoTjchhBBCCCGEEBIjFHQTQgghhBBCCCExQkE3IYQQQgghhBASIxR0J5AgCDAajaBp9YQQQkj8UP1LCCEknijoTiCTyYSsrCyYTKYePQ/P86iurgbP81EqWXqi89Q1OkfhofPUNTpH4aHzlBhU/8YXnaeu0TkKD52nrtE5Ck+8zxMF3YQQQgghhBBCSIxQ0E0IIYQQQgghhMQIBd2EEEIIIYQQQkiMUNBNCCGEEEIIIYTECAXdhBBCCCGEEEJIjFDQTQghhBBCCCGExAgF3YQQQgghhBBCSIxQ0E0IIYQQQgghhMSILNEFIIQQQgghxOl0wu12J7oYCcHzPFwuF2w2G1iW+sSCofPUNTpH4Ql2njiOg1wuj/rrUdBNCCGEEEISxmg0orm5GXa7PdFFSRhBEOB2u2G1WsEwTKKLk7ToPHWNzlF4Qp0npVKJ/Px86HS6qL0eBd2EEEIIISQhjEYjampqoNVqkZ+fD7lc3isDBUEQ4HQ6e+37Dxedp67ROQpPoPMkbmtra0NNTQ0ARC3wpqCbEEIIIYQkRHNzM7RaLcrKynp1gCAIgjSstTefh67QeeoanaPwBDtParUamZmZqK6uRnNzc9SCbhroTwghhBBC4s7pdMJutyMrK4uCA0JI0mAYBllZWbDb7XA6nVF5Tgq6CSGEEEJI3IlJ02KRtIgQQnpCvC5FK7kjBd2EEEIIISRhqJebEJJson1doqCbEEIIIYQQQgiJEUqkRggJShAEOBwOr9a+QL939bjv74QQQggJThAECIIQ8LHu1qdU/xKSeBR0E0KCMplMaGxslCr+cILo7gTiLMtK23z38dzm+7h4nO/vkTYCMAwDjUZDNyaEEEKSgsvl6tZc0mgG4kqlEjNmzMB3330X89djGMarHickXVHQTQgJyuFwwOFwQKvVem33bH33bYkP9Jjvv+EeF+jxYI95rrHIMEzAbYF+BzqSZZSWlkKtVvs9NyGEEBIL5eXlGDhwoNc2mUyGwsJCTJ06FbfffjsmTpwYldcKVecG29836O/qGE+rVq3CDTfcEHKfyy+/HK+99hoUCoVXUL59+3a8+OKLWL9+PWpra8HzPEpKSnDSSSfhiiuuwKxZs6R9Fy9ejCVLlkh/syyLzMxMFBYWYty4cTjzzDOxYMECaDQav9e/+uqr8cYbbwQt36233opnn3027PdMSCgUdBNCgnI4HJDJZJDJ0vtS0dbWFrXslIQQQkh3DB48GJdddhl4nkd7ezt27NiBDz/8EJ999hm+/PJLTJ8+vcevEUnPdE96oMVjZ8+ejZNPPjngPmPHjpWG0TMMA57nceedd2L58uWQyWQ45ZRTcO6550Iul+Po0aP44osv8NZbb+Ghhx7CQw895PVcF110EUaPHg0AMBqNKC8vx9q1a/HBBx/gwQcfxKpVq7yCdU/XXnstysrK/LafeOKJEb9/Qnyl9500ISRi4nxujuMSXZS4cLlciS4CIYSQXmjIkCFYvHgxnE4n5HI5BEHAo48+igcffBBLliwJOcw72Z166qlYtGhR0Md5npd+v//++7F8+XKMHz8eH3zwAQYPHuy1b3t7O/75z3+iqanJ73nmzZuHSy65xGub3W7Hs88+i/vuuw9nn302Nm/ejLFjx/ode91111GATWKOJlEQkuJ4nofZbJZ+LBaL9GO1WqWf9vZ26cdms8Fms8Fut0s/4lByl8sFp9MJm80Gt9vdK4JuhmHgcDgSXQxCCCEphOd5uN3uLn94ng/6A/yeNE384XkeV111FQBgx44dfq/7+uuv46KLLsKwYcOg0+nQp08fnHXWWVi7dq3fvuvWrYNSqcQ//vEPbN++HXPnzkVeXh4KCgowf/58lJeXh/VeBUHAnXfeCaVSiauuugpOpzPi8+b7vIIg4PDhw3jyySeRl5eHr776yi/gBgC1Wo1FixbhwQcfDOu5lUol7r77bjz44IOwWCy45557olJmQiJBPd2EpLj29nbU1tbC7XYHnLcMdJ30TPxdEASYTCY4nU4pENXpdLF9A0lAJpPBZrOhvb1d2hYqKZt4o+V0OsGybFjZ3SlJGyGEpBee57sMPkNd+8VjxZFlbrfbb950oOldt956K8aOHYtTTjkF+fn5qK2txWeffYa5c+fi3Xffxbnnnut3zLZt2/D0009j5syZuO6667Br1y589tln+PXXX7Fz506oVKqQ5bzuuuvwzjvv4JZbbsGyZcuiWqfxPI+VK1fC7Xbj+uuvR0FBgVcPuC+FQuHVSAEEz/bOMAzuuOMOPPnkk/j666/R1taGrKysqJWdkHBR0E1IihMznAarRLqTvEwQBFitVigUCgAdFVtvyCoqk8lgNpu9gm5RoOAb6Jgz5nA4gp4fz/21Wi0KCwujWGJCCElfgiDAnQKjj9zo2dznUKt2vPnmmwCAk046ye+4Xbt2+SVgq6urw0knnYR77703YND91Vdf4a233sL8+fOlbddccw3efvttfPbZZ1iwYEHAMprNZlx88cX47rvv8Mgjj4QcKh7I999/D5vNFvCx+fPnY9iwYXC73di4cSMAYPr06V2OPPNsnBDzsTidTtjtdgDe51Umk0Gr1eKEE07Ahg0bsH37dpxyyilez/faa6/hq6++8tqmUqmoZ5xEFQXdhKQ4sVc6mO6slc3zPDiOA8dxvSLYFikUCsjlcgDhZ2PnOA5yudwvE7rvvjabDRzH+Y0+IIQQEpjb4cCHV1+d6GKEJAA45+WXwXU2UvfEkSNH8Mgjj0iJ1Hbu3Im1a9eiqKgIS5cu9dvfN+AGgOLiYpx//vl48cUXUVFRgf79+3s9Pn36dK+AGwCuuuoqvP3229i2bVvAoLu5uRnnnXcedu7ciVdeeUUa8t4da9aswZo1awI+NnbsWIwYMQIA0NDQAAAoKysLeP/hWccGGj0WaJvnFLmSkhLpPflasWKF37asrCwKuklUUdBNSIoL1dtKwhesRzsQsXFCJpN1ee4ZhoHL5ZKOIYQQkh6i1Zh69OhRPPLII17b+vTpg++//x5DhgwJuP+yZcuwZs0a1NbWSj28orq6Or+g+/jjj/d7ntLSUgAdK3j4amhowKxZs1BdXY333nsPZ599drffF4CIescDCRRk+/7ruw/LsiGHqYu2bNlCidRIzFHQTUgSsFqtXvPCwpkjLBKX9SLJieM4KUEdBd2EENI1TqHARa+/HvPXEZOW+faihnus4JPTI1Jz5szB6tWr4XK5YDAYsGrVKvz973/HRRddhE2bNkGr1Ur7Hj58GNOmTYPRaMTMmTNx1llnQafTgWVZrF+/HuvXr/cLwgEEzM8i3jsEWjKzvr4eRqMRgwcPxuTJk3v8HrvSp08fHDhwALW1tRg+fHjUnlecOldbWwsAKCgoiNpzE9IddKdOSBLQ6/VobW0Fx3EBk6GF+l0QBK8KmSQXsaXd5XJBqVQmujiEEJL0GIaBLA7XS57n4XY4vKcRhXtwjKYLFRQUYOHChTAajVi6dCkeeughPP3009Ljzz//PAwGA1auXIk//vGPXsfefPPNWL9+fVTKMW7cOFx++eW48cYbcfrpp+Prr79GUVFRVJ47kKlTp2LdunVYs2YNZs+eHZXnFBPEms1mbN++HRzHYcKECVF5bkK6i4JuQhJMzH6akZERMnsoSU1ipd/c3Ay9Xi9tCzQkzvd3oCNoz87OptEMhBASI8k4Cunuu+/GG2+8gZdffhm33HILBgwYAKBjaDkAnHPOOV77C4KALVu2RLUMV111FTiOw5///Gecfvrp+Oabb2IWeF9xxRVYtmwZVqxYgVtuuSVkj7Tdbg+rThTr3yeffBJWqxVnnnkm1Gq1lKhNHHrudDoDZqFnGIbqXhI1NBGUkAQTs4/ThT19ZWRkSJlV7XY7bDabtH66xWKByWSCyWSC0WhEW1sbWltbYTAYoNfr0dzcHHCoICGEkJ4JtsxUMlCr1bjjjjvgdDq9kqn169cPALBp0yav/ZctW4bffvst6uW4/PLL8dprr+HgwYOYM2cO6uvro/4aADBkyBDccccdaG5uxrnnnotjx4757WOz2fDss8/i0UcfDes57XY7nnnmGTz22GPQarV4+OGHvdZIF//vA6237nK54HK5kvbzQVIP3eUTkmDiBZ6SoaUvuVwuZUfvrra2toDz7QghhPRMsgdU1113HZ5++mm89dZbuOuuuzB48GBcf/31ePPNN3HJJZdg3rx5yM3NxdatW7Fz507MnTsXX375ZdTLcdlll4FlWVx77bWYM2cOvvnmGxQXF4d1bKglw4qKinD99ddLfy9ZsgQ2mw3PP/88xowZg1mzZmHUqFGQy+UoLy/HDz/8gJaWFjz00EN+z/XRRx/hwIEDADqWOSsvL8fGjRvR3NyMvn37YuXKlRgzZozXMZ4jygLdg4mBOa08QqKBgm5CYsBms3mt+RwoAZr4u9VqhSAIFHSTgBiGCTjsjRBCiD/PNZyDEetfnueTOqBSqVRYtGgRbr/9djz66KP497//jfHjx+OLL77A4sWL8cknn4DjOJx44olYu3YtVq9eHZOgGwAuvfRSsCyLP/3pT9Icb3EZrlC6WjLMM+hmWRbLli3DJZdcgldeeQUbNmzAxo0bwfM8+vTpgzlz5uDKK6/EzJkz/Z7r448/xscffwyWZaHValFQUIAZM2Zg7ty5mDdvHjIyMrr9npN5JARJPYxAn6aEMRqNyMrKQltbW8CskuHieR61tbUoKSmhwC2EeJ6nlpYW1NTU+K3j7Pu7+C/HcdBoNDEtUzh4nofBYEBOTg59lkKI53kym83IzMwMu1chWdB1KTx0nhKD6t/4CnaebDYbjh07hoEDB0Ylp4kgCHA4HHC73WEH02yUMpBHg8vloqlmYYjXeXK73VAoFEk55z8UQRDgdDq97kGJv67OU7SvT/TNJiQG7HY7VCoVZRUnPSYuOWYwGEKOmBB/Z1k2KRpwCCEkEQRB8FoJhJCeCDVyItDa4dT4RoJJu0+G3W7H3XffjZKSEqjVakyZMgXffvttWMfW1NRgwYIFyM7Ohk6nw3nnnSdlifTV0NCAG264AaWlpVCpVBgwYACuvfbaaL4VkqLElvZUaxklyUmhUKC9vR0NDQ2oq6uTfmpra6V/a2pqUFtbi8rKSjQ0NMDlciW62KSXobqXJAMxOKKAm0QDy7Jwu91SdnPfH4fD4fVDdS8JJe16uq+++mp88MEHuO222zB06FC8/vrrOPPMM7FmzRpMmzYt6HFmsxmzZ89GW1sb7rvvPsjlcixfvhwzZ87Erl27kJeXJ+1bVVWFk08+GQBw4403orS0FLW1tdi6dWvM3x9JfmJiNAq6STRwHBf28Fe324329nYaokjijupekgxoDi6JJnH6XzjEzx4lXiPBpNVd2datW/HOO+9g2bJluPPOOwEAV155JUaPHo277roLmzdvDnrsiy++iEOHDmHr1q2YNGkSAGDu3LkYPXo0nn76aTz22GPSvjfccANkMhl+/vlnrxsCkv6MRiOcTmfIYb7ifDKFQpGQMpLei+M4aekTQuKF6l4Sa4IghHVdE9ddJiQRqMGHhJJWw8s/+OADcBznlQlRpVLh2muvxZYtW1BVVRXy2EmTJkmVPgCMGDECp556Kt577z1p2/79+/Hll19i0aJFyMvLg81mo8zCvYQgCNDr9aitrUVDQwMaGxvR0NCA+vp66d+6ujro9XppThkh8SYIAg1xI3FFdS+JNfG6FmyYr/jTnQRqhESTZ6JcQgJJq6B7586dGDZsmN9QzMmTJwMAdu3aFfA4nuexZ88eTJw40e+xyZMn48iRIzCZTACA7777DkDH2oKnnnoq1Go11Go15s6di/Ly8ui9GZJ0XC4X3G43dDodsrKyoNPpgv6emZmZ6OKSXophGFgsFlitVlitVrS3t0s/NpsNNpsNdrsddrudbg5IVFDdS2JNHLbLcVyXP5TIiiSKOCLD7XaD5/mQP6T3Savh5XV1dQGX1RG31dbWBjxOr9fDbrd3eezw4cNx6NAhAMD111+PSZMm4d1330VlZSWWLFmC0047DXv27Am6FqB4oysyGo0A0OMvIM/zEASBvsRd6Ol5ElvSFQpF2p5rzzlJ6foeoyGZz5NMJkNrayva2tqC9viI89SKiopilumcrkvhieZ5SlSwkex1L0D1b6IFO0/i9q7mYlMDIUkVoUaaeU5NjNdyXvTdCU+g8+R5n9fVNT6c+jetgu729nYolUq/7eLaau3t7UGPAxDWsWazGQDQp08ffPHFF9JJLisrw6WXXor//Oc/uO666wK+ztKlS7FkyRK/7XV1ddLzRkIQBBgMBgCUsTOUnp4nm82Gtra2tB7SKAgCLBYLAPoshZIK56mrG1ibzQae50MGKj19fboudS2a56msrCwaReq2ZK97Aap/Ey3YeRJHkDmdzpBTssSew3QPIHiep+lBYUjV8yR+fuORZT8Vz08ihDpP4pSVxsbGLpPThlP/plXQrVarvVqyRTabTXo82HEAwjpW/HfBggVerRrz58/HFVdcgc2bNwet+O+9914sXLhQ+ttoNKJv374oLi4OOztxIGLrS3FxMQ2rCqGr82Q2m6X/b8C71Uq8MKb70HGxIsjJyaEbyBDS4TyZzWZkZ2cjPz8/Js9P16XwpMN5Sva6F6D6N9GCnSebzQar1QqGYQJeSz3nyfaGNZBp5YnwpPp5crvdkMlkMf88y+XymD5/ugh2nsSViAoLC6WG4J5I3U9sAMXFxaipqfHbXldXBwAoKSkJeFxubi6USqW0X6hjxX+Lioq89uM4Dnl5eVJLbiBKpTJgiz7Lsj3+4omVUbpXSD0V6jyZzWY0NTV5/R95Lv0gCAIUCkVan2Oe56Wbn3R+nz2VDueJZVm4XK6Ylp+uS+FJ9fOU7HUvQPVvMgh0nsTfw+m5pPNL0kkslxbzHBGSqh0D8dDVefK8z4vG9Setgu7x48djzZo1MBqNXi3XP/30k/R4ICzLYsyYMdi2bZvfYz/99BMGDRok9W6ecMIJAOB3g+FwONDc3IyCgoJovBUSZ4IgwOFwIDMzMyqtWYQkO47j4HA4pCXwAi1/J/5OlTYJhepeEg204gfpTcJZU57q3vSSVs2G8+bNg9vtxiuvvCJts9vtWLlyJaZMmYK+ffsCACorK7F//36/Y3/++Wevyv/AgQP44YcfMH/+fGnbrFmzUFhYiLfffttrKPLrr78Ot9uNOXPmxOrtkRgSs01SSzrpLWQyGWw2GyorK1FRUYHy8nKUl5f7/V5fX5/oopIkR3Uv6Ylwgg9C0gnDMHC73VKCR/HH4XB4/dC87PSSVj3dU6ZMwfz583HvvfeisbERQ4YMwRtvvIHy8nKsWLFC2u/KK6/EunXrvC7yN910E1599VWcddZZuPPOOyGXy/HMM8+gqKgId9xxh7SfUqnEsmXLcNVVV2HGjBm44oorUFlZieeeew7Tp0/HhRdeGNf3TKLD5XLB5XIFHH5ISDoSM6cGuuEV/3Y4HOB5XprXREggVPcSQkj4WJbtsqGJ53nqCEozaRV0A8Cbb76JBx54AKtWrYLBYMDYsWOxevVqzJgxI+RxmZmZWLt2LW6//XY88sgj4Hkes2bNwvLly/2GrV155ZVQKBR4/PHHsWjRImRnZ+OGG27AY489RjemSUwQBGnetvj/JF7Q7HY7XeBIr9NVIhqO42C32+FyuejaRkKiupeE4nA4YDKZ0NzcLNWzLMtKDXvBiFnLkxHLsvS5SwFOpxOPP/44/vvf/6KqqgoOhwPvvfcexo0bh+HDh+OKK67Aa6+91uXzlJeXd2v/rnQ1dJxhmKT97Cej8vJyDBw4EFdddRVef/31RBcnoLQLulUqFZYtW4Zly5YF3Wft2rUBt5eVleH9998P63UuueQSXHLJJZEUkSSI0+mE2Wz267UTk1mkciZMQmKB4zi43W4aBUK6RHUvCcVut6OtrQ2Ad7DhdrshCELABm+3243a2lo4HI64lbM7FAoFSkpKohJ48zyP//u//8Prr7+OgwcPQiaTYdy4cbjttttwzjnn+O0/bNgwVFRUBHyuGTNm4Ntvv/XatmLFCjz77LOora3FqFGj8Pjjj+Okk07yO/bbb7/Fueeei++//z7g4+HYsWMHXn75ZWzYsAF1dXXgeR7FxcWYOnUqLrvsMpx22mkRPW+knn32WTzyyCOYPn065s2bB7lcjuHDh8e1DJHwHIkW77ndYgDrKyMjA4MHD8ZFF12EO+64A1qtNq7lSnUUZZBew+VyQRAE6HQ6ap0mJAxipe92uxNdFEJICnO5XGAYBpmZmV4BtsvlgtFoDHgMz/NwOBzgOC7p6my32y310ve0bIIg4I9//CM+/vhjDBo0CFdffTXsdjtWr16NefPmYfny5bjpppv8jsvKysItt9zit71///5ef3/00Ue46aabcNJJJ2Hu3Ln45JNPcPbZZ2P37t1SvgUAsFqt+Otf/4rrrrsuooCb53ncfffdeP755yGTyTBr1iycffbZkMvlOHbsGL788kv85z//wUMPPYT77ruv288fqf/973/QarX43//+B4VCIW13Op3YvXs3srKy4laW7khk0C0aPHgwLr/8cgAdn9OmpiZ8+eWXWLx4Mb766its3Lgx6b6byYyCbtJriEE3ZYPs3Xi3G3WbNiFryBBogyxlRH7HMAwMBoPUS+V5wywu10RrgRJCQnE4HBFP3+I4LilHokWrMfLjjz/Gxx9/jJNOOgmfffaZlLH/H//4B0466STcc889OPPMMzFgwACv47Kzs/HAAw90+fwrVqzAsGHD8P3334NlWdx8880YMWIE3nnnHSxatEja76GHHoLT6cSjjz4a0ft46KGH8Pzzz2PcuHH473//i8GDB3s93t7ejhdffBF6vT6i549UbW0t8vLyvAJuoCOvyYgRI+Jalu4SBEFaYcSXOEIzlve0Q4YMweLFi7222e12TJ06FT/++CPWrVuHU045JWavn25oAivpNcRhbKR3a/ntNxz+5BNsf+qpRBclJWRkZMDpdErZVdvb22G1WmG1WmEwGGC32xNdREJIkrPb7ZQzJYjPP/8cAHDXXXdBrVZL2/Pz83HLLbfAbrfjzTffjPj5q6urMW7cOOn89+/fH/n5+aisrJT22bZtG/71r3/hueee81r2L1yHDx/G008/jby8PHz++ed+ATcAqNVq3HHHHXjwwQf9HnvjjTcwbdo05ObmIjc3F9OmTQv4ntetWwelUolHH30U27dvx9y5c5GXl4eCggLMnz8f5eXl0r7/+Mc/oFQqpZU4lEollEolhg0bBqBjCLVSqcR1113n9RputxtPPfUUjjvuOOh0Ohx33HF48sknu5xfvWHDBlxwwQUoKSlBZmYmRo4ciYceeghWqzXge/jHP/7R5XsQE67xPI/169fjoosuQt++fZGZmYmBAwfiwgsvxMaNG/3Ksn79elxwwQUoKCiAUqnE0KFDcf/99/uVJRJKpRKzZ88GADQ3N0vbHQ4H/vnPf+KMM85A3759oVQqUVhYiAsvvBA7d+70e561a9eCYRgsXrwY27Ztw5w5c5CZmYmsrCxccMEFXufBk9vtxhNPPIEhQ4ZApVJhyJAhWLp0adD/nw8//BAzZ85EYWEhVCoVSkpKcNppp+HDDz/s8bnoruRrOiSkBzznjYkVjLjOsNVqpUqfwNXeLv1uMxigyslJYGmSn0wmC9rLZDQaaUkTQggAwGw2w2q1SnUu4D1FhUaZBSYuy+jbk+25be3atX7BqhiM19XVITMzExMnTsTkyZP9nqOsrAx79uyRksVWVlaiubkZ/fr1A9AxCvDGG2/EeeedF3D+eDhWrVoFt9uN6667DkVFRSH39c0Pcvvtt+PFF19EaWkprr76agDAJ598gj//+c/YtWsXnnnmGb/n2L59O5YvX46ZM2fiuuuuw65du/DZZ5/h119/xc6dO6FSqTBjxgzcf//9+Oc//wkA0lD8roaT33TTTXj99dcxYMAA3HjjjbDZbHjuueewZcuWoMe8/PLLuPXWW5GdnY2zzjoLBQUF2LFjBx5//HGsW7cO33zzjV9P+7Zt2/D000+HfA/id+mf//wnFi1aBLVajfPOOw99+/ZFTU0NtmzZgg8++ADTpk2Tnvell17CzTffjOzsbJxzzjkoLCzEtm3b8Oijj2LNmjVYs2aNX1m6w+FwSAHz+PHjpe16vR633XYbpk+fjjPPPBM5OTk4evQoPvvsM3z55ZdYv349Jk2a5Pd8P//8M5588knMnj0bN9xwA3bu3IlPPvkEv/zyC3799VeoVCqv/a+//nr8+9//xsCBA3HzzTfDZrPhmWeewebNm/2e+6WXXsJNN92E4uJiXHDBBcjLy0N9fT22bt2Kjz/+OO6rXlDQTdJKe3s7GhsbAwYJPM/TMFgCeUaG9Lt+/36UTJ2awNKkPgq6CSEA0NbWhubmZr+gShxhloxDxJNBfn4+gI6e16FDh3o9Jvb2HTp0yO+4+vp6/PnPf/baNnHiRLz55ptePc3XXHMN/vjHP2LOnDmYOHEiPvnkE6jVaikh4dNPP43q6mqsXr064vcgBqSzZs3q1nEbNmzAiy++iBEjRmD9+vVSQPzAAw9g+vTp+Ne//oULL7zQK6gEgK+//hpvvfUW5s+f7/U+3377bXz22WdYsGABZs6ciZkzZ2LVqlXSc3Zl3bp1eP311zF27FisXbsWGo0GAHD33XcHbNAAgH379mHhwoUYM2YMvvrqK+Tl5UmPLVu2DPfffz/+9a9/4fbbb/c67quvvuryPQDAnj17cNddd6G4uBhr1qyRGmLcbjfkcjkaGxul4/fu3Yu//e1vGDt2LL788kv06dNHaux6/PHHce+99+Kf//yn13KMoRw+fFgaXi4IApqbm/H111+jpqYGTz75pDRqAABycnJQWVmJ0tJSr+f47bffcOKJJ+K+++7zS/AHdMy5f+edd3DxxRdL26688kqsWrUKn3zyiVfizLVr1+Lf//43xo0bh02bNkn/P/fdd59XA4Dotddeg0KhwK5du1BYWOj1WEtLS1jnIJqo24+kFZfLBZlMBp1OF/CHEj4QT/p9+xJdhJQmLilGCOndxLmnmZmZfvVuVlYWMjMzqac7iDPOOAMA8NRTT8Fms0nbW1papF7a1tZWr2OuvPJKfPXVV6iqqoLBYMDWrVtx2WWXYdu2bZg7dy5MJpO070UXXYR//vOfaGhowKuvvoqioiKsXr0affv2xcGDB/HYY49h6dKl6NOnD5YtW4Z+/fpBo9Hg9NNPDxjsByL21vsGXF156623AHQExJ490Dk5Obj//vsBQAqaPU2bNs0rWAWAq666CkBHD3Kk3n77bQDA3//+dymgAzre11//+teAx7z66qtwuVxYvny5V8ANAHfccQcKCgrw3nvv+R03ffr0sN7Dq6++Cp7nsXjx4oCjIUo8ctO8/PLLcLlceP755/3Kctddd6GgoAD//e9/A76PQI4cOYIlS5ZgyZIlePjhh/Hiiy/iyJEjOO200/yy0CuVyoD//6NGjcLs2bOxfv16OJ1Ov8dnzJjhFXADHY0PQEcvuCdxysGDDz7o9/9z6623BnwPcrk8YIeb7/mJB2p2JGnFbrdTYN3L6ffvh8DzcFos4JRKqPPzgyZMaz14ELzTCZZGQESE4zg4HA40NTVJw+DEG2tBEJJ2qR9CSHS5XC643W7qzY7AJZdcglWrVmHt2rWYPHkyTj/9dDidTnz++edS75zv1DgxIBWNGzcO//73vwF0BI4rVqzAbbfdJj1+/fXX4/rrr/c6RhAE3HzzzZgyZQquvvpqvPvuu3jggQfw0EMP4YQTTsDf//53LFiwANu3b4/Z1Lxdu3YB6Ai8fM2cORMAsHv3br/HAvVqigGfOMUwEnv27AEAnHzyyX6PBdoGAFu3bgXQsdzamjVr/B6Xy+U4cOCA3/bjjz/eb1ug9yAG4IGWWuN5XgpkGYaRRhx89dVX+Oabb8BxnFdjl1wux/79+wO+j0DOOOMMfPXVV9LfLS0t2LRpE2699VacfPLJ+OGHHzBlyhTp8V27duHJJ5/Exo0bUV9f7xdkNzc3o7i42GvbCSec4Pe6ZWVlAPwbm8TPwvTp0/2OCbTtkksuwV133YXRo0fjj3/8I2bPno1p06ZFlLcgGujqSNKGeJNPQXfv5TCZ8Msrr/htn/HUU2ACrQPrcMBYUYHsIUPiUby0o1AoYLFYvIZpiSsE2O12ms5BSC/hdrvhcrn8hpaTrslkMnz22WdYtmwZ3nnnHaxYsQJZWVk499xzcfvtt2P06NF+Q2ODue666/D2229jy5YtXkF3ICtWrMDWrVuxfft2MAyDF154AaeccgruvfdeAIBWq8Xs2bPx3Xff4fTTTw/5XH369MGBAwdQW1vbrTWwjUYjWJZFQUGB32NFRUVgGCbgknKBgiaxwacnWeXb2trAsqw05N9TsP8DMRv7448/3q3XCvc9tLW1gWEYv2CVZVm/9yqWZenSpd0qS7jy8vJw7rnnIiMjA3PmzMH9998vDRnfvHmzlMn89NNPx9ChQ6HVasEwDD755BPs3r074Mi47vxfhvr/CZRL4M4770ReXh5eeuklPP3003jqqacgk8lw1llnYfny5QFHDsQSBd0kJTmdTr/WTJ7nwfM8tbT3Yu4gPavmmhpkeqxH6vsYBd2RYVlWWt7Gl8ViiUqmVEJIcrFarTCbzV69n+Ka1ZSsNDJKpRL3338/7rnnHq97mHXr1gEAJkyYENbziENmLRZLyP3q6upw33334f7778eQzvrv4MGD0rBe4Pfe5AMHDnQZdE+dOhXr1q3DmjVrpMzW4dDpdOB5Hk1NTX5BbWNjIwRBiGuvZFZWFnieR3Nzs19DgOfcaU9i+Zqbm4PWhz2RnZ0NQRBQV1fnNXybYRi/TiaxLA0NDdDpdFAqlTGZ1iH2bnsO/3700Udht9uxYcMGvzn4P/74Y8ARC90V6v+noaHBb3+GYXDNNdfgmmuuQUtLCzZs2ID//ve/eO+993Do0KGolKk76OpIUpLNZkNjYyOam5ulH71eD6fTSUF3LxascmnZuzfoMebq6lgVp1fjOA5ut7vLZVYIIanFbDajrq7Oq/41Go00siUG3nnnHQCQkmp1RQyC+vfvH3K/W2+9Ff379/dL7uXZEyn+Hk7QdsUVV4DjOKxYsQJNTU0h9/V8DTGwX79+vd9+4rZx48Z1+frRMnbsWADApk2b/B4LtA2AlJH7p59+ikmZJk6cCAD47rvvutxXLMvWrVtjukSuwWAAAK/6/ciRI9Jyb56sVit27NgRldcVPwsbNmzweyzQNk95eXk4//zz8e677+KUU07B3r17cfjw4aiUK1wUdJOU5HK5wLJswGRplKyF+AqVMM1cUxPHkvQeHMdBEATKbk5ImrHb7dBoNH51b4bHyhCkewINof7oo4/w+uuvY+LEiTj//POl7fv37w84imj//v34+9//DgBeGZ99ffLJJ1i9ejX+7//+z6uTYsSIEfjuu++ka7Y4l3fEiBFdln/IkCG444470NzcjHPPPRfHjh3z28dms+HZZ5/FP/7xD2nb5ZdfDgB45JFHvM5BW1sbHnnkEa994uGPf/wjgI5eW8/RAjU1NXjhhRcCHnPjjTdCJpPh9ttv91r7XNTa2irNXY/E9ddfD47jsHjxYlRUVHg9JggCamtr/cpyxx13BC1LoDWzu0tcxs1zLn7//v1hMBjw22+/SdvcbjfuvPPOLhtiwnXFFVcAAB5++GG//5/nnnvOb/+1a9f6NT44nU5pGL7vcmSxRl2CJCU5nU4KrtOcwPNoPXwY2tJSyD2yVEbCVFkJp9kMuVYrbVNmZ8Pe2gprQwOsjY1wWa3QxXl+TzpjWRY8z6O1tVW6GfdcuzcjI4OGohKSYnielxq9SfRMmzYNZWVlGDZsGDIyMrBt2zasW7cOAwcOxH/+8x+vYcTvv/8+nnvuOUybNk3KNH7o0CF89dVXcDqduOuuuwImlQI6gtnbbrsNf/3rX/0SWP31r3/F5ZdfjtNPPx1jx47FqlWrMHr0aGmebleWLFkCm82G559/HmPGjMGsWbMwatQoyOVylJeX44cffkBLSwuWLFkiHTN9+nTcdNNNePHFFzFhwgScf/75EAQBn3zyCaqrq3HzzTcHfS+xMGvWLFx11VV44403MGHCBJx33nmw2+344IMPMHnyZPzvf//zO2bUqFF4/vnnccstt2DMmDH4wx/+gEGDBsFkMuHYsWPYsGEDrrjiCvzrX/+KqEyjR4/GU089hYULF+L444/Hueeei379+qGhoQEbNmzA3Llz8fTTT/uVZfz48Zg7dy4GDRoEs9mMo0ePYv369bjqqqvw8ssvh/XankuGAR1zxjdt2oQdO3YgJycHTzzxhPTYLbfcgm+++QbTpk3DggULoFKpsHbtWtTU1GDWrFlYu3ZtRO/f0+zZs/GnP/0JK1euxJgxY3DBBRfAbrfj3XffxYknnui37N35558PnU6HE088Ef3794fT6cS3336LvXv3Yt68edK2eKGgm6QkylKe/pp//RV7X38dyuxsnPjggz1+Pv2BAyjyuMlQ6HTgnU44LRb83JkAZcLChcjszJpJekYMsA0GAwwGg1dWc5lMhtLSUuoZIyTFiFnKFQpF3F6zJ4mxYiXaZZo/fz4++eQTbN26FU6nEwMGDMC9996LhQsX+s1pnjlzJvbv349du3Zh06ZNsFqtyM/Pxx/+8AfccMMNmDNnTtDXueeee6BUKr0CKc8yVFZW4oUXXsC2bdtw8skn41//+lfYDSwsy2LZsmW45JJL8Morr2DDhg3YuHEjeJ5Hnz59MGfOHFx55ZU49dRTvY5bvnw5xo8fj1deeQUrVqwAAIwcORIPPvigtIRWPL300ksYOnQoVqxYgZdeeklajuqiiy4KGHQDwLXXXotx48bhueeew8aNG/HFF18gKysLffv2xd/+9rce99bfdNNNGDVqFJ599ll8/fXXMJvNKCwsxKRJkzBv3ryAZVm+fDk2bdqE1atXIysrC2VlZbjllltwxRVXSAlPuyIuGSZSKpUoKyvDX/7yF9xzzz3o16+f9NjZZ5+NDz74AI899hjeeustZGRk4JRTTsHHH3+Mhx9+uEfv39Orr76KYcOG4dVXX8ULL7yAsrIyLFy4EAsWLPALupcuXYqvvvoKW7duxeeffw6NRoPBgwfjpZdewrXXXhu1MoWLEWI56J+EZDQakZWVhba2th4liuB5HrW1tSgpKUm71me32y3NHQF+v5EXszl2J1Mqz/MwGAzIyclJu/MULcl0jg59+CFqO+dQzewcytQVm16PnzqHpPWZNAm82422I0dgb2tD4fHH47grrkDzL7/gt5UroevfH5xKBYPHUh59TzkFg84+u8vXSabzlKxCnSOj0YiSkpKELduRTNL5+p3MqP7tms1mg9lslv5mWRYOhwNtbW3QaDTder/BrgculwtGoxH9+vXzG+rpdrtRW1ubtEsPKhQKlJSURLUDwOVyUV6aMNB56lqgcyQIAgRBgEKhSLvrVSQEQYDT6YRcLg/YCGGz2XDs2DEMHDgwKkPR6RNLkprdbkdzc7PXNrGFTusxVJikH43H2tr2tjYos7LCPpaVyzH80ksBAG1Hj2LXCy9Af+AABJ+kXtrSUq+g21JX18NSk3DRXG9Cklt7eztqa2u9erUFQQDHcXG5Yec4DiUlJUmbjJFlWRpxR1IKwzDgeT6mSdZIcBR0k6Qm3phTj1jvw3nc6JlraroVdHvK7N8fMpUKLqsVJp/EIlqP5TcACrrjRewxI4QkL4fDAaVSGZNlkMLFcRwFtoSQtEBjC0hSc7vd1CKXptpbWuAMkIHV0tAAe1sbrB5rLnaVYdxmMMDe2gpTgOW/WI5DTmf21RafLOZan/nb9tbWcIvfY4IgwFxTAz4J5yvGGsuysNvtsFqtaG9vR3t7O2w2m9dSMoSQxLLZbBTwEpKGeJ4P+ENii3q6SVKjhGnpyWEy4eelS6EpLsYJd9whbTdXV2N7gPnbodbS5l0u/Lx0KfgQw5VzjzsOTbt2Qb93r1eiNHVeHjilEm6PYM9tt4PrRq6ASNVu2oTDH32EPlOmYPjFF8f89ZKJXC6H2WyGzWbz2q5QKFBaWhrXJE2EEH9utxtut5vqX0LSDMMwUkJETyzLUt0bYxR0k6ThcDikdRrF+WJ2u52SPaQhh8kEgedhrqmB02qFvDOLdVt5ecD9Q/V0ux2OkAE3AOR29nSba2rg8FgLlGFZaEtL0Xb0qLTN0tAAnUdGzlipXrMGAFD/00+9MujOzs6W/hZHs5jNZjidTqr4CYkzi8WC9vZ2MAzjdVPenWSlhJDkF+ieWhAEaa43LccbOxTNkKRhs9nQ0NCA5uZmNDY2oqGhAe3t7XQDnuY8e7FVOTkB97Hp9QGHogfD+6y7qMjMRGbfvgA6lg7zpPVI2AYAltrasF+nJzyHtvfGKRTizT3DMGBZFizLQhCEpFweiJB0ZzKZUFNTg6amJjQ0NECv14PneerpJqQX8FzSk8QOBd0kabhcLnAcB51OB51Oh6ysLOh0OloWIs11NV+7u/sFk3vccQCA1oMHvbb7zuu21Nf36HXClVFUJP1u0+vj8prJjmEYOH0aTAghsWe326HRaLzq3szMTOr1IqSXEJcTI7+L9vmgoJskDYfDQUPJe6FQ87U9Gfbv91vyqztyR44E0DEc3ZNv0G2NQ9Dtdji8yhHuOUh3LMuivb0dTqcTTqcTLpcLLpeLErwQEkPi/O1ENHAzDANBEGgJQUKSgDjMXBxq3tsDcbETIFojfqgLkSQNSprWO4Xbg121Zg1c7e0YtmBBRK+TWVYGuUYDp8Xitd2zxxmIz7Jhu55/HmaPYezmmhoUjBsX89dNdjKZDBaLBRUVFV49bHK5HKWlpXR9ICQGxPnbiQi6xSXBjEYjNBoN9awTkiBiLgdfHMdBLpcnoESJJQgC2traoFQqo/b+KegmCWGz2WD1mKPL87w0vJz0Lu1NTXA7HF7rcgMd87sZmQza0lI07doFAKj78UcMnT8/6I1ZRmEh3A4H8seM8XuMYVnkjhiBhu3bvbazHId+c+ag9eBBGCsq4DCZ4DCbodBqo/MGAzD7zBvv6dD5dKFUKqUbf7F13e12w2az0fWBkCgxmUxe0zgcDkdCM5Wr1WqYTCYA6PVTytxuN/X6h4HOU9eicY54ngfLsnC5XGnZICYIApxOJ9xut9e8dqfTiba2NpjNZpSWlkbt9XrvlY0klNVqRXV1NeRyudcXmTKl9j6CIMBSWwvdgAHStsy+fTHh9tsBAPa2NinoBjoSnWmDXARPWLQIbIgbx9zjjvMLugFg4Ny5wNy5+OmRR2DT62Gtr4diyJDI3lAEaHh5B4Zh/G64ZTIZHA4HXC4XXR8IiQK9Xo+2tjav+pdl2YTdVIvfa6vVCpPJlJY39+EQh/Ym8v8iFdB56lq0zpE4vFwmk6XluRaTt3Ic5/f+lEolSktLodPpovZ6FHSThHA4HFCpVNDGsDeRpA5TdbVX0O1J4XPB0+/bFzTo7krO8OHSHMJANMXFsOn1sNTVITuOQbfDZILDaPR7r4TmfBISTWLvl06nS6qVQZRKJZRKJdxud6+dQ8rzPIxGIzIzMym/TQh0nroWrXMkjjQrLS1NqutFtPA8j8bGRhQWFnqdp1gNqaegmyQEzd8mnkwVFRBOPjngY76tj/p9+9DvtNMieh25RoPM/v1hDLIeuKa4GC2//Rbzed2BAn9jRYXXsHje5QIToPW1NxKzmtMaooT0jJiYMFnmaPJOJ1iPsvTm+wK3ywXwfMdon85rHUNBpR+e56VRURR0Bxatc8RxHGw2G1iWhVKpTLv6l+/8vqlUqrh8lujTSuJObGnvzZUr8dawfTv2vflm0MflHiMijOXl3Vqz25e4dFggmj59AMR+2TAmwGf/t5Ur0bhjB4CO7OZbH3kEe158MablSBUsy6KtrQ0VFRUoLy9HeXk5qqqqaHkxQrpJTJqWDPVv486d2HjPPWjYti3RRUkKe1euxP7nn4fTasXuf/0LWx991G+1DULiSQyym5qavOrf+jgtrZpuKOgmcdHW1gaDwQCDwYCWlpakqfRJ8mjavRuu9vaAj429/noos7IAdMzBMRw4EPHrFJ1wAlQ5OcgbNcrvMU1xMYCODObxGuKYNXCg9Pu+t94C0DGP3W40ovXIEbQ3N8elHMksIyMDDMNIS5m43W5YLBY46IaUkJDcbrdU9xoMBrS1tSXNiBFzdTUEQUD9Tz8luihJwVRZCZfFAv3evTBVVsJmMMBw8GCii0V6OY1G47WUmM1mg8VigdvtTnTRUg4F3STm3G43WlpaUF1djfr6euj1egiCQMOCiFcPNoCgwbS2rAwnPvQQymbNAtAxxDxSqtxcTHnggYBD1NUFBWBYFm67HfbW1ohfI1xTHngA/c84w2ub2273+rsn7zVdsCwLlUol/ajVaikBCiEkOLvdjsbGRtTX16O+vh5GozFphpaL2o4dC9rg2hsZPK75dP0niSYOvxZ/NBoNZY+PEEU9JObE4WxZWVnIysqCTqdDZmZmUrS0k8RiGMbrc6Dfvz/k/uLQcP3+/RB4PurlYWUyZBQWAojPet1AR4OCJ8OhQ15/6/fujUs5Uo04z5sQEpzL5YIgCF71b0ZGRqKL5UXgeerR9WA4cADoHGml37ev1yaWI8mJZVlpmV/SPRR0k5hzu900nJwE5ZmJ3GmxhNw3a+BAcEolnGZzzNa29hxiHg9ynxtg356N1iNHaF5fACzL0vByQrqQKtnAqUf3dy6bTfo/s7e2wkrzZ0kSERPB0kiz7qPs5STmxJZ26tnuvZp27cKhDz7AcVdcgZzhw70e05aVwRTmOtWsTIacYcPQ/MsvaNm7F5l9+0a9rBmdydSOffEFVDk5KJwwISrPe+C//4WxogLH33pryP3qtmxB3ZYt0t+8y4XWw4eRN3JkVMqRLjiOQ3t7O2pra6VrS2ZmJi1DSIiHVFkppH7rVjTt3i1Nr2HlcvBOJ/pMmoThl16a4NLFlqWuDnv+7/+CNjpvW7YMWYMGYdxNN/llM6/ZuBFHP/0UvNuN4qlTMWz+fAAd9VfD9u04/m9/gzI7O9ZvgfQyDMOgtbUVZrNZGrGYm5ublsuKRRP1dJOYsNlsMBgMXl9K0nsd+vBDOK1W7Hn5Zb/H+p9xhl9vbyjSEPMY9YyIGcyB3xObRYN+/35YGxsDlnvEZZeFPpZ6gfyIa/paLBaYTCY0NTXB0sVICUJ6A5PJhNbWVrS2tsJut6dM/hTPfBZ859SR+p9/htNsTlSR4qJuyxY4TKaQ+7QdPYrWw4f9th/+6CPwnT2OdVu2SHPj9fv3w97aisadO6NfYNLrqdVq2O12qf5taWmB3ScfDfGXGldiknKMRqOUOK2trY1av3o5ZU6O9DvvMyRJmZWFE5cskVrouyIG3abKSjhicDOmKSmJ+nN6agkwR7vohBMwfdmyoMuZ6ffuTYkhovHEsiw0Gg00Gg20Wi0yMjJouDnp9XieR3NzM6qqqlBXV4f29vaUr3/1PVitIhWo8vPD2i+cxlffZKTUYEtiQS6Xe9W/LMvSHO8wUNBNYsJut0OtVkOn0yErKwsqlSrRRSIJ5DkM3HjsmN/jLMeFXD/bkzIrS5oHbugi8VokVB4NBAC67IHoLsP+/VKSHE/BzgErk8FmMKC9sTGq5Ug3HMfB6XSCj0GCPUJShZhDxTNxmkyW3DMJZV3cH6R7MknOp1HE83ywHlMDAgXQ2UOGeP3tu0/bsWNw2WzRKCYhQVGOlfBQ0E2iTsxqmArzyEh8eGUoD9LyrszOhjbMXuZYDjH3nTMX7YRtTqvVr7dfFCjozho8GAD1WHSF4zhaxoT0euJqIakypBwAckaMCPm4/sCBmKxWkaw88554/m5tbITDaPTal/NpsPBd2UNwu9FKmeFJjHEcR8PLw5A6V+Uw2e123H333SgpKYFarcaUKVPw7bffhnVsTU0NFixYgOzsbOh0Opx33nk4evRoyGM2btwoJRFobm6OxltIeWKlT0E3EQQBv73+Omo3b5a2hQoew+3t9lw6DDG+GTNXVUEQBPz6739j/9tvd3uYd3tLC7Y//XRYPebqvDy/bXmd7/XIZ59h27JlsBkM3Xr93kIMupubm6HX66HX69Ha2krD8uOE6t7kICYuTfage//bb6NqzRoA6DLRl8tqxfo778TmBx8M+1qaKsq/+goH33vPa1uWR++1b524ZfFiHPjvf8G73dj94oto+fVXr8cdJhM2LFrk1Vj82+uvY+fzz1OPN4kZcaRZU1OTVP9SjhV/yX1VjsDVV1+NZ555Bpdddhmee+45cByHM888Exs3bgx5nNlsxuzZs7Fu3Trcd999WLJkCXbu3ImZM2eipaUl4DE8z+OWW26BRqOJxVtJGYIgwGg0orW1FW1tbWhtbU25lnYSG26bDc179nhts9TXw92Z7MVXwfHHg2FZr2RmgWT26weZWg1XezuMlZVRK69o2MUXS7+ba2rgsljQ8uuvaNi+vdvDvOu2bOlWb3m/U0+VftcNGIBcj6zllro67F25sluv31swDAOO42A0GtHQ0IDa2lo0NzdTz3ecUN2bGC6XS6p329raYEqBgFTgeTRs3+61rf+cOV0eJy4Vefjjj2NVtLir+OYbr79ZhQL548Yho6gIisxMFE6Y4JdotP7nn9Hy228BE6sBCNjQaCwvR8XXX0ev4IR4kMvlcLlcaGlpQUNDA6qrq6HX6xNdrKST3BN9umnr1q145513sGzZMtx5550AgCuvvBKjR4/GXXfdhc0evW2+XnzxRRw6dAhbt27FpEmTAABz587F6NGj8fTTT+Oxxx7zO+aVV15BVVUVrrvuOjz33HOxeVMpwOVyoampCTabDRzHSa3sFHSTYMy1tQG3a0tKMPXhh7uc48dyHHKGD0fTrl0xGXZdPGUKOIUC+1at8guY9fv2oaSgIOznUmZldeu1B5x5JspmzgQjk4HhOHByOdT5+Wjv7M0Ld3m13sgzCON5HhaLBS6XC3K5PIGlSn9U9yaOzWZDXV0dgN+DrVRMnNb/D39A6fTpYOVygGHAO51gWBZNu3f79QQHyguSDnKGDUOfc86BPCMDExYuhOB2Q6ZSYcqDD8JYXo49//d/0r6thw55HTvo7LPBKhQ4/NFHQZ+/+ZdfMPi882JWftJ7sSyLzMxM6W+bzQaXywWe5ykW8JBWZ+KDDz4Ax3G4/vrrpW0qlQrXXnsttmzZgqqqqpDHTpo0Sar0AWDEiBE49dRT8Z7PBR8A9Ho97r//fjz88MPI7uVrILpcLrhcLilpS1ZWlteXjxBf5hCBozwjw29edSDSEPMYJdnJGTYMQMfwcM9hefpuJm/jlErvv7u4IWYYBnKtFjKVClxnsJjrM+cxnYZXxgrLshAEgXq644Dq3sQRh5OLdW9WVhbUanWii9Vt4nWPUyrBKRSQazSQqdXIGzXKb197W1sCShgHDCPN0ebkcqnxmVMo/BOm+dR7wc6VZ31j0+uD5hMhJJpkMpmU1JH8Lq2C7p07d2LYsGHQ6XRe2ydPngwA2LVrV8DjeJ7Hnj17MHHiRL/HJk+ejCNHjvgN2XrggQfQp08f3HDDDdEpfApLlTlkJHlEIzmZGIjGao6zXKOR5hp6lrf1yBGv9WS7Swzmu8NziDngvywMCUwQBKr044Dq3sRxOBxeiSpTkW9yME+KIA346ZJYzXO1DFNFRdD9GJb1Crx96z2B5/1W3gD86xtTDKZjEeKLZVlKbBpAWg0vr6urQ3Fxsd92cVttkCGter0edru9y2OHd2aR3LNnD15++WX873//61ayMLvd7pXdz9hZ0fA836NlbniehyAICVsqR1ymJ9mX6hEEQfpJ9rImSrTO0bHVqzvmmwVJYmWpr5deL9LXkWk00JaVefWa8zwPRPEGVFtaCntrK/a+8Ya0TXC7YTh4EGxpadDyOy0WHPjPf+Bqb/e7ydENHozmzuQ34X5vMgcO9Pq7Ze9eFEyYEMlbiptk+b75Bm0ZGRlJtYRSNK/fiWr4TPa6F0jf+tdut4Nl2aSv0wRBQMO6dSgvL+8YQu7BcPBgyPJnDRqENp/Eeub6ein3R/WaNTBVV2P4JZf4PXc4mvfsQd2PP2L4pZcGDfKjqX7rVtRv2QKGZb2CZ5fNFvKamTNiRNB53MaqKvTheWhKSmDx+L5lFBcDHsnWWvbuBZeRgSMffoiyU06JqBE40ZKlbklmyXCOeJ6HwWCQrrsMw0Cr1SZVI2G869/kufOIgvb2dih9hnICkNaIbg+SvEncHu6xf/vb3zB37lycfvrp3Srf0qVLsWTJEr/tdXV1MJvN3XouT4IgwNB54Y7nh9nWWUG0t7fDbrcnfY+SIAhSNsVk+tInk2idIzErra+iWbPQsHYthM7PinhRjpR6wACYPIauthoMYKKYNZ/NyQmYlKZu925paGug89Ty889Bh6HLBwyAXKeDy2qF2eFAe5jvX11aCmtnA0PLvn3Qt7SENQw/UZLh++Z0OmE0GsEwjPT/mJOTgwyfxESJFM3rd1lZWTSK1G3JXvcC6VX/8jwPu90OQRCksif7cj2CIKBh/fqA5yj/pJNC1gN506ej9cgRr22NBw8iu/NzU7VuHZwmEzKGDoXOY4mtcFWuXw/zsWOo2LwZ+Z2jM2Lp4LvvBtyee8IJIa+ZisGDg67GoBo8GAaDAQWzZ8P81lu/HzNwIDi1Gi6rFQDQ+MsvcAAwHDoEh8OBQd3IT5IskqFuSXbJcI7sdrvU6C0IAmQyGXJzc5Mqx0q869+0CrrVanXAisfWOR8z2DwncXs4x7777rvYvHkzfvVZpiEc9957LxYuXCj9bTQa0bdvXxQXF/sNy+sOsYWmuLg4bj0dLpcLVVVVsNlskMlk0Gq1SZ+8xfOmmy7UgUXrHAU69vjbboOmtBSGbdvg7KwMWJZFToAhceGSTZiAJo/syNk5OWCjGHTzQ4d6Pb+o/dgxlMydG/Q8OfPzUeezXde/P4ZffjlUOTnIueMO8E5nl0vleDr+5pthNxiw+4UX4Gpvh8xkgm7AgO6+pbhJxu+byWRCdnY28vPzE10USSKu39GW7HUvkF71r8ViQU1NDXieh0qlglarTfrPjiAI0hJvnsb+5S/QDRoU8hqRk5OD3HvvBadS4eB//wvDgQMQWluluoPtfF5HdTVyTjyx22Wrlss7GuYMhh7VR+EK9F6Hzp+PghNOQJvRGPyamZODrAcfhOHAARzqzHfAyuU4/vbbkVFY2LlLDvLLysCp1XDb7VDn5aHgvvvQ3tyMXc89B3tjIzi7HQzDoL26GpkqFWQplgMgGeuWZJOM58hkMiEvLw9arTbRRZHE+/qdVkF3cXExagLMFRUze5aUlAQ8Ljc3F0qlUtov1LGLFi3C/PnzoVAoUF5eDgBobW0FAFRVVcHhcAR9HaVSGbBFPxqZvhmGiWvGcHGuZFZWVsqsx83zvFTpJ/sNSqLE8hxpi4vBchy0paUwHDwIAD1+HV2/fpBrNF5BfDTLrevb128bK5PB3tYGR1MTmLy8gK8nC9AAxalUyOhch1sZwRBGVqWCvLgYuSNGoHHnThj270f2oEHdfp54Scbvm0wmg9PpTJryiOJ9/Y62ZK97gfSrfwVBSKlEcsGGb2pLSsK6h9B0BpUF48fDcOAALLW1v5/vzqDCsG9fwMC+K+L+Xs8ZQ54jb0Ta4mJwMlmX10x1Tg4UEybgyIcfgne7wcrl0PossakpKvL6W6HRQKHRILNvX5iqqtDamRNE4Hm0HT6MgnHjovjuYi8Z65Zkk4zniGGYpMxmHs/rd3K98x4aP348Dh48KM3VEv3000/S44GwLIsxY8Zg27Ztfo/99NNPGDRokJSNu6qqCv/5z38wcOBA6UdcsmTChAk488wzo/iOkpfL5aK1uElEtFEcAsuwrF9m72hSBFjuS0xmY/IZ7tiVUElyukPK2h6DpdLSHcdxcDqdQYdokshQ3RtfTqczaXqv4k1bWgqgYwUM3++xva0NlgANOOGy1NeDdzp7VL5waEI0DoWDUyiQNXhwt48T6w5rU5O0jeoREk+9PbFaWkVM8+bNg9vtxiuvvCJts9vtWLlyJaZMmYK+nb1WlZWV2O8z33LevHn4+eefvSr/AwcO4IcffsD8+fOlbR9//LHfz8UXXwwAePPNN7F8+fJYvsWk4XK5ImpRJkS8aYoW8UYiFgJ9vsVM4g1r1mD/qlX47fXXsW/VKux7+204O+fNBeK57FhP5HTOWTTX1KTv0jkxwnEc3G43jEYjrFYr2tvbKQCPAqp748vhcKRkg3fAJGfd/P5p+vQBw3Fw2WwwV1dj/3/+45X9vCdBpMDzUpLPeOtu/eC7okU48gIco9+3j66BJC44joPVaoXFYoHVaoXD4Uh0keIurYaXT5kyBfPnz8e9996LxsZGDBkyBG+88QbKy8uxYsUKab8rr7wS69at87rQ3HTTTXj11Vdx1lln4c4774RcLsczzzyDoqIi3HHHHdJ+559/vt/risuhzJ07N6nmCkab3W6XviQ9STxDeicxwVk0e7qBjoyunEIBtnNoXrQVTZyIBo+AQAzyBUFA0+7dXvu629sx+rrrAj5PQZDevu5SZGZKWdvbjhxBYZJnMU8mHMfBYrFIQ5cVCgVKS0sDDjsm4aO6N/YsFos0RNtut6fMtK6usN387rEyGTR9+sBcU4ND778Pk8fqFUBHENnv1FMjLo+5pgaZAaYVxZq6m5/fvFGjcPSzz6DoxvxYbVmZ13QsAHCYTB3vOUFJGEnvIZfLYTabYe3snNBoNFKDbG+RVkE30NHi/cADD2DVqlUwGAwYO3YsVq9ejRkzZoQ8LjMzE2vXrsXtt9+ORx55BDzPY9asWVi+fDkKUjC7Yyy0traiublZamFP9sRpJPGOu/xyZPTp0xEQd35u1Hl54JTKHq117UmekYHjb78dDMvGJJv3sPnzUTRpEuQaDTiFAuq8PGQUFsLS0OC3b8vevV5/q/PycNwVV8DtcES1sUHXrx/M1dUw19RQ0N0NLMtK82AFQYDJZEr6VRdSBdW9scPzPBobG2G1WqX6V6PRJLhUkRu2YAF0/fsDDAMugkzG2tJSmGtqAl6DjceOdSSajDA5mDlAboJY6XfqqSicOBG8wwF1fn63li1S5+Vhwm23QdaNzwHDssg97jivRmSgo6GCgm4SawqFQoobHA4HHA4H3G532jQghiPtgm6VSoVly5Zh2bJlQfdZu3ZtwO1lZWV4//33u/2aixcvxuLFi7t9XKqx2WxQKpVJtdwOSW7ZQ4f6rXvKsCy0paV+6672hG/imGhi5XLkDB3qtS1nxIiAN3yA9zBBeWYmMvv1i3qZNOK8xjjeIKYbMZlRb59jFi1U98aOy+WCy+VCZmZmUq0xH6nsoUOh7kwqGQltWRmwdavf/GuG4yC43TAcOBDxyCKzT895LGUNGtSjuiuShlzPoFs8X/p9+9B/zpyIy0FId7EsC6fTCZfL1auC7tSbFEQSwu1297oWKRI72h4mkkm0UPPIxczssST2SgRKJkTCxzAMBd0k6blcLvA8T/Vvp2B5QcTVHHoyr9tcWwuhGz3OqSZn+HBpGlbWwIEAOpJ8hspHQki0iflVelv9m/pNpiQuxKCb5j6SUBwmE6p++KHL/aI9rzvedJ03K4Ho9+1DXgyTuwFARp8+YBgGTqsV9tZWqOKwtmw6YlkWVqtVGqork8kosCFJx+VySetcpwLDgQNo6Vy+i5XLwXeuGMBHKet6sEbb7GHDYDh0CPr9+yHwfETTjXinE9amppiMnuJdLlR8801CRyjJMzKgGzAAbceOQZmdDU2fPrDU12PvypUoPOEEFEewzjkh3SWONGtvb5dG78jl8pRMENkdFHSTkJxOJ+x2O9rb26mnm3Sp4ptvULtpU5f76QYMAICI590lGiuTQa7VwuWRkEak37cvphnVAYCTy5HRpw8sdXUw19RQ0B0hmUwmZVJlGAYZGRkojXJ2fUIiJda7FoslpUa07Hn55Zg+P6dUIqOwENbGRq/teaNHo/LbbzuSg9XWRjxP2VJTE5OgW79/Pyq/+y7qz9tdeaNHo+3YMcgyMpB73HGw1Nej9cgRtB450uOh/4SEi2VZ6PV66PV6MAyDgoICKedKuqKgm4TU1taGxsZGaXmwVGlpJ4nhDDOrfUZhIcbccAOUOl2MSxQ7Q/78Z6C5GZxCAU6hgMNkwsF334XDaISltjbmr68tLe0IuqurkT96dMxfLx2pVCooFAoIgiCtzsDzfNq3tpPkJwgCGhsbYbFYwDAM5BEkHEtnOcOGSUG3MjsbwxYsgKaoCNnDhqHl11+h37u320G3TK2Gq70dpurqtE5QWTp9OpRZWR25SWprUbVmjfSY/rffUNpF8kNCokGr1UrJA00mU69YQozuLEhIdrsdcrkcWVlZ0KVwgETioztztXOHD4emuDiGpYktmUaDgvHjkT96NHKGDUPRCScgZ9gwAIDeJ4t5LIhD9CmZWs+wLAuO46BQKKRpNIQkmpg8TaPRICsrixKY+vBcpzqjqAi5I0Z0bO8cZRTJvG5xrniskqlFkqk9FliZDIUTJnQMNR840KszpaUH8+EJ6Q6GYcBxnFT/2qO0ok0yo6CbBCUIAhwOBw0pJ+Hr5T2E4g2f79qxsaClDOZRxbJsr0zsQpKT+FlMxfq3u2tORyJr8GDpd2tdnfS7GHwbKyrCHnkl8rymptJw/p5gOU5qLAaA1oMH/bLCExJrHMfB6XR2a9m8VNTjO+S9e/fitddew9KlS/HZZ59J23me7xVDBdIZZUxNLja9HpXff++1JBVJLrGey+1JvEG0t7ai4ttve81NYqywLCsNM3e73XQ+SUKJ9W8qTXXg3W7UrF+P9ubm0DtG4bvl2WtsNxql31U5OdIIqu6uJKHp0wcMx8HV3g67wdDjMnbF1d4e89cIh+eoAUEQ0HbsWAJLQ3ojjuPA87xU/6ariK/mVVVVOO200zBmzBjccMMNuP/++/HJJ59Ij7/66qtQq9X4/vvvo1FOEkculwsWiwUmkyllW9rTUdUPP+DYF1+get26RBclbGwarCnbHarcXGj69InLa8lUKqlHqfzLL9GwdWtcXjfdtbS0oLy8HI0+SZoIiQe73Q6LxQKz2ZxyOVT0e/fisMd9YDDRqhdKp00DAL9rrtj42dLNaT6MTCY9l7G8vOcF7EKyJBLN88kJ0nroUIJKQnorlmXhcDhQW1uLiooKmEymRBcpJiIKuvV6PWbOnIkffvgBo0aNwl/+8he/XoEFCxaAZVmv3m+SGoxGI6qqqtDY2Ai3251SLe3pzN05ciQe84V7SpmVhZFXXZU0NxXxlIjebgCo3bw5bq+brrRarTTMzWazUW83ibvGxkZUVlaitbU15epe31FY+aNHY/ill2LovHkYdM45KDv3XAy/7DIoo5SheNB552Hw+edj1DXXeG0Xr8GGzqXDuiN76FAAHZnGY2novHnIGT48pq8RLlVODkZfe21cp0cR4ollWan+tVqtaTtSOqIr+hNPPIHy8nLceeed2L17N1544QW/fXJycjBmzBhs3Lixx4Uk8eVwOMCyLHQ6Xdqn709FpqoqOJK8FTBn2DAUjBuX6GIkRFyDbo/svKaqqri9brpiWRYKhQIqlQoulyuth7mR5ON2u+FwOKBWq6HT6aDVahNdpG6R+yR7K5k+HX0mTULJSSehdOZM5Iwdi8Ljj4/a67Ech7IZM/zmkOsGDIBMpYLTaoWpsrJbzyklYosgYA+XtqQEJSedFNE64rGSN2oU+p9+OoDeNaedJA+ZTAaFQpHWSdUi+sZ/+umnGDBgAB5//PGQw58GDRqE2jgsnUOiy26305DyJBfrVngSOd3AgeCUyri8ltZnTeku51KSsHAcR5nMSdyl4jzuZMRynNSL3N0s5lmd12+n2dzrklRqiovBMAycZjMcHvPkCYknjuPgcDjSsuEnoit7RUUFJkyY0GXFoFAooNfrIyoYSYxUzpjaE+3NzWjcsSMpvuQumw31W7fCabUG3SeS5VBizVRVlZTlijffbLCx5Bt00/mPDspkThJBHF2RKvWvvbUV9T//jMadO9Gydy8ad+zwelxIYKOVNK+7m9dEViaTrt/dnRMeik2vR8P27VF7vljgFApkFBUBoFUxSOKkc6N3RNksVCpVWJPcKysrkZWVFclLkARwOp2wWCxwu91QKBSJLk5cHfrgAxgOHoRco0n4PKu6LVtw9PPPUXLyyRh60UUB9zHs3w/e7QabRDdnO5YvT3QRkkbuyJFo/uWXmK/LqsjM9Ppbv3cvSqdPj+lr9gbiCK6mpiY0NTVBp9MhNzc3waUi6cxms8FisUAQhJRJoLb3zTdDJxxL4PsQlw4zV1fDYTRCodOFf+xxx6H5l19gOHAAA844Iyrl+W3lypQIZDWlpbDU18NcXY08j6zmhMQLx3Gw2Wyorq4GwzAoKChAhs/UlVQVUU/3iBEjsGPHDlgslqD7NDc3Y/fu3Rg7dmzEhSPx1dTUhIaGBrhcrl43vE1cuqMtDhlLu+LuTEbT8ttvQXveXTYbTBUV8SwW6YaiE05A/zlzMGDu3Ji/1tgbb0ThCScAAFqPHJES7pGe0Wg04HkeZrM5ZF1HSE+5XC7U19fDYDCkVN0bKuAe8Ic/IHvIkPgVxodCp5NyXugPHOjWsboBAwAAltraqM3rToWAGwAyPdYqJyQROI6DSqUCz/MwGo1plVQtoqv7vHnz0NLSgoULFwZdyHzRokWwWq24+OKLe1RAEh/i+rQymQxZWVkp09IebZYkqmjsra2w1tcHfZyGEicvVibDgLlzpZu3WMoZNgwj/vhHqHJywLtcaD18OOav2RvIZDKoVCpkZGRIc20JiQWXywW73Y6MjAxk+oxeSWbyIIneMsvK0P/00xM+EitPTIrWzboyo7AQrFwOt8MRtTwZnsukxSpBWzSIDRVmymBOEkhMaKpQKCjovvnmmzF69Gi89tprmDx5Mh577DEAwJEjR/DMM89g6tSpePPNNzF+/HhcffXV0SwviRHxplImk/XagBtIvtbdQDcL2pKSoI8lC1eaZp5MVgzD/J51N4k/F6koneeXkeTgcrkgCELKzOUW+eaUECVLPeq5dBjfje8vw7LQFBcDiF7wmdmvn/S7JURjeqKJ/6c2gyFkXhlC4oHjuLTKZB5R0K1SqfD1119j6tSp2LFjBx544AEAwMaNG7Fo0SL89NNPmDhxIlavXg15jOc0kujobQnUjOXlMFVXQ3/ggNcQbpvBAGcCh5JaGxu9bliadu9Gy9694J1OaZs459xcWwt7ayucFgsMPu8jngSe9xu+13rwYELK0ptJQffevdJngXe5oN+/n4ac94AYdFNSNRIrqfjZMpaXw24wBHwsGRKSAh2BrjwjAy6bDXWbN8Pa0CA95rLZoN+3L2ivc6bY49uDBgRXe7u09FgyLQ8WikythqozfwX1dpNEY1k2rUaaRZRIDQCKi4uxceNGfP311/jiiy9w9OhR8DyPvn37Yu7cuTjvvPN6dY9pKnE4HLBYLL1mqRKbXo+dzz8v/T3kggu8HjfX1MQt+7Svnx9/3OtvU1UVfn3tNRSdcIKUlEau0SCzXz+YKiuh378fhgMH0LR7N0Zfey3yRo2Ke5mbdu3Cvrfe8tqmzMmJezl6u+yhQ8FyHGwGA6yNjdAUFaHuxx9x+KOP0O+00zDwzDMTXcSUxLIseJ5HU1MTZDIZMjMzU2oIMElegiDA4XCgvTOnSKqwGQxedWiyYlgWOcOHo3HnThz++GMAwLTHHwenUODY6tWo3bwZw4NMgdRGYW7z0c8/R92PP2L4pZdG/ByJoC0rg02vT+i9ECFAxzQvm82GmpoayOVy5OTkQBmnJVljIeKgW3TGGWfgjChldyTxx/M86urqYLfbe00jib211evvI598Is1jAhIbdAfTsH07iiZOlP7OPe64jqB73z64OoeANf/yS0KCboNPL7duwICgWddJ7HAKBbIGD4bh4EHo9+6FpqgIzs5VJtqOHk1w6VKbSqWC3W5Ha2srGIahoJtEhc1mQ21tLVwuF2SyHt+OxY3DZ/UaTqlE4fjx4N1uCIKAogkTElQyf7kjR6Jx507p79bDh5E3ciScZjOAjnozEM+5zZFmlBdHzbV4vIYsIwPDL7mk288VT9rSUjTv2UM93SThZDIZWJaFzWaD0WiEWq1O6aA7/bs1SUgulwsOhwMKhaLX3EiyPsuhCYLgNRwuWSsah9Eo/S4u5WE4eFCaq6bfty8hw/q0fft6/T380kuDzvUjsZUrfi727/fabq6pSerkPclOqVRCo9FAo9HA4XAkzfBZktrEBGoajSallsRhfRoIhlxwAYZdfDFG/PGPOO6yy6SpLskg12cJUN+cF4ZDh7ymb4ky+vQBw7JwWq1+DfXdZTh4EHznFIKhF16I/NGje/R8sRaNXn5CoiUjIwMajQYcx6XkVBxPEQXdW7ZswTXXXIPNmzcH3WfTpk245pprsHXr1ogLR2JPnCshl8t7TU93IJ5ZwpOlolH5rAvs2VupLS2FXKuF226XGgkcJlNCyt6bPzfJRrzZbT16FK7OpecAwG23w6bXJ6pYaUOs9CmpGokGl8sFhmF6xbSuRJFrtVDn50t/+zZO804nTFVVfsdxcjkyiooA9Lwh3u1wJG1jfiBi0N3e1ET5QEjSYFk25TOZR3Slf+WVV/Cf//wHw0IMwR02bBjefvttvPbaaxEXjsSemDW1twdOvEfrmbWxMWAl3J3nMlVXw9rU1O3sn57/D76JVzzLyLAsckeM8Nvu2Yrvam+HtbGxW69PUltGQQHU+fkQ3G60Hjrk9Zh402czGLxGTZDwUSZzEk0OhyMtAm5Xks9JF+tKoCOni2H/fjg6h5cD3nWoJzH4bNi+vcfBZ7DXSEbKrCwoMjMhCAIstbVRe16bwdDjUQOk9xIzmafySLOIrvabN2/G+PHjke/ReuiroKAAxx9/PDZs2BBx4UhsOZ1O2Gy2lP4Ax8qO5cvRduxYRMce+fRT7HjmGfy8dCl+evjh7h3scQPGdTFvJdAQPs+ge8ezz+Lnxx+Pf+83DWNOKPEG03cYpam6Gm6HAz/94x/YsngxDTePAMuycLvdaG5uRnNzc8olwCLJQUygZrPZ0mLFkGQPuvN8hnP/8uqrYeW5EDOYN+/Zg19ffTUmZUtWYoODKUo99LzbjZ+XLsWPDz9MveckImKjd1NTE1paWuAMMC0k2UUUdNfU1GDAgAFd7te/f3/URrGVjESP3W5HdXU1Wltb06LSj1Smz3xkxuNcVHz9dUTPWbtpk/S72+GAO4I1BjXFxRg2fz6Ov+02aYkwXznDh/uNUDBVVEjJW9qbmgAA9XGc4tFn0iSoCwvj9nrEnzivW79vHzyb08w1NV7L4bU3N8e5ZKmPYRjI5XKYTCbU1NTA5JNQipBwmEwmVFZWor29PaUSqAWSN2oUSk46KdHFCCl76FCUTpsGtptL2HrmJmk9ciTaxUpqYiI5S5Qa7XmHQ+rtN9CSoiQCcrkcLpcLer0eDQ0NsHlMoUsVEQXd4Y6rdzqdKT/pPV25XC7YbDaoVCpoNJpEFychlFlZmHD77V4Vq6ZzDhcQecWg0Om8/jZH0PA05s9/Rma/ftD164exN9yAwgDZYOUZGdD5NH4JguCXTdxzbdJYyh8zBsMvvbTXT1VItKzBg8HK5bC3tcFSVydtN1dXe01ZSJbcBakmIyMDOp0OGo0G9gga1AhxOp1wOBzQaDRQ+CT2TCWKzEyMvvZavzov2TAMgyEXXogT7rijW8dpfBKCJnuPfjRJPd0xqCd8R2EREg5x5RCdTgeGYVJymldEQfeAAQOwZcuWkAG1y+XCli1b0L9//4gLR2JH/L+TyWS9PkgSewYB+GXd9pz3FS7f3vNoBDd5HmX05DnEXEy81uJTodG87t6Fk8uRM3QoAKDVo+HIabF4zeVOpcQ+yYhlWSkRJSHdYbfbIZPJevUos0RQFxT4bfNNWOpJplJ5/R1pXR7qNZKV2NNtrauTVkiJFgq6SU+lalK1iILuM844A42NjVi8eHHQfZYsWYLGxkb84Q9/iLRsJIacTmevC7adVit4hwOOtjav7Xkegavnet2A99JLboejy5Zup8XilzAlGsFNsCHmnkG3+Lth/37YPd5jLBOXuJ1OryHLJDmInwXfuXMtv/0m/R7uDaTDaIz6TVc6oKRqJBLifO5UCLjtra1plfOFYRi/JcQy+/UL+/hIg+5kWkItXKrcXMhUKvBut7S6i8NshtvhgL2tDQ6TqVvJ4XzvSayd098IiYSYVC3VRDSZ6Pbbb8eKFSuwdOlS/Pbbb/jzn/+MEZ3Je/bv349XX30Vn332GTIzM7Fw4cKoFpj0jCAI0tDydMiaGi633Y6fH3sMrvZ2v8aGzH79IMvIgMtq9WuR1u/bh6KJEwF0JFezNjTgpEcegTzAmqoumw0/PfKI3xzuaPR0yzUa6Pr3h7GiAvAov6akBEqdDnajEVkDB6Jx+3Y4LRb8uGSJ1/G82w02Bjd5v61YQfOzklCwm7yKb76RfjfX1HS5coGpuho7ly9H0cSJGH7ppVEvZyrjOA42mw0Gg0Faw5uQUHieh8PhgNvtTvqgu37rVhx45x0MufBClE6blujiRE3e6NHQe0zBkmdkQJWTA5vBEPBamNm3r7SaSaQN6Jo+faDMyuoIPFOks4NhGGhLS9F65AjMNTWQazTY+uijXg2w2UOGYNxNN3X5XOJnyZNh3z5kBBh5QEg4xJFmYv2bKtN0Ioq6SkpK8P777yMzMxOffvopzjnnHAwdOhRDhw7FOeecg08//RRarRbvv/8+ynx6DkliWSwWVFVVwWKxpHwCl+6wt7bCHSTpAsOyGHD66dANGADdgAEYeNZZUuWrP3BAyvIszo327P32eg2DIWDSNEt9fVSWC+l76qnILCvzWv6EYRj0O+MM6Pr3R9bgwUF7xG0tLT1+/UAo4E5OqtxcaY1ZIPBa6k6LpctREO1NTRAEAU27d4NPwUyhscSyLBiGQWNjI1pi9P0i6UWv16OmpkYaXp7MxGlJDdu2Jbgk0dVnyhTkjRwJZXY2Mvv2RenMmeh3+ukddeigQX77j7zqKun3njSg9xfr6YEDI36OeBOn25mrq2GurfUb8dR6+DBcYSSzqvz2W79txsrK6BSS9EoKhQI2mw01NTUwRzANNFEi7uo89dRT8euvv+LWW2/F8OHDoVKpoFKpMGzYMNx666349ddfMWfOnGiWlUSBy+VCe3s7VCoVlF0sSZVOmAC9Cp7D5kpnzMDxf/sb5BoN+p16KqY9+SRkKhVcVitMPpWDpXOoVVevMfSiiyBTqyG43UGP6Y780aMxYeFCaIqLvbaXTJ2K42+9FcqsrKA9nPFKpkaSh+dnwTMA9xTuTaTb4QhriZ3eRqvVQqPRwOVy0TBz0qX29nY4HA5otdqUGWlmqqyMKLdJsmJlMoy+7jqc+OCDmHD77cgoKEDxlCkddWh2tt/+qtxcnPjQQwA66tFIl7sqPvHEoK+RrMTpduaaGsjU6oD7tB461OXz6AI0NFAiT9ITLMsiKysLCoUipeZ29+iqX1ZWhuXLl2Pv3r2wWCywWCzYt28fli9fjr4+yaRIcnA4HJDJZJDL5b1uTrcvz6RSvliOQ05nj7JfYrJuBNBSS3GcKhjPXnBP8Qi6ae3N5OIZdGtKSgLu053hkpT8JjBxbjet1EFC4XkeLpcLSqUy6Xu5fQUb3dVbKHQ6yLVaCIIQlQb0VOF1/xJkbn849UKg7PbtjY10z0B6jGXZlJrbnRpNrSRq7HZ70s8li4VIhneLQYt+716vobWWMANYt93uNTwrHoIt3RKroNuz1b718OGYvAaJTNbAgeA6R7ME66UwVVSEnShJv39/REPMeaezy9fgnU4IPO/3PU2FIe2UUI2EQ/yMJEvA7fnd6up7pt+3LyW+i7Eizm8GAHPn/O6u8C5X0EA1VagLC8HKZHA7HEHve/T79nU72Z5MpepowIhgOVVCPMlkspQaaUZBdy8i9sb0tqC7dvNmbHvyyW4fJ/Yam2tqsOHuu6Xt7WFm3bTp9V7Ds+LFd/g5EJ+ebiFFLnq9BSuTIWfYsJD76A8cwJ6XXgrr+ayNjdhw99049uWXYZfB7XBg66OP4pdXXgm6j7m2Fpvuvx/r77wTWx56SJpnrj9wABvvvRfV69eH/XqJwDAMBEGA0WiE1WpNdHFIkhITqCXDsHJzbS02/f3vOPLppzCWl2PTffeh/Kuvgu7fuHMnNtx9N5p27YpfIZNMd0atuR0O/PTII2j+9ddYFyumWI6T7icOvf9+wH3sbW2w1NV163kTcV9E0hPLsuB5Hq2trSnR4x3x1d9kMuGJJ57AnDlzMHLkSAwaNCjgz+DBg6NZXtID9fX1vbKn+/CHH3r9Lfb+lZx8csjjFJmZyAySCLCrpcOU2dkoOflkqaK21NZKCdlibdSf/uQ3b8za2BjzpV+GLVgQ0+cn3Vdy8slQ6nTIGT4co665BnKtFgC8gvHWw4e7Ncyv8ttvw/4sOdva4DSbYThwIGjCnWaPJG2u9naUdwb15upqCDyP+p9+CrtsiSKTyaDX69HQ0EDDzIkfp9OJ5ubmpAm6rZ3JPet//hmV330H3u32WtkgmL1vvhmH0iWnzG4EiraWlpDT11KJ7zKqQEeDLuPxOe7O1KPcESOg698fAAXdpOc4joPL5UJDQ0NKJDSNaJxTbW0tpk2bhoowhib29nnDycLtdsNut4Nl2aQZ3hYvmuJimDuHMTEch5MefTTsG5/ckSNhCjA03FJfHzQLqUylwokPPggAEHgerFwOt8OB9qamoAmtokmdny+9Pu9yYeM993SsrdnaClVOTkxec8Jtt3VrvVMSHznDhuHExYulv/NHj5Z+X+exnKOlvh66bvz/WerqoA0yTzzoMTU1yArQCKv2WTZG7zN/1FJXB5vBELPPbjRoNBq43W60t7fD5XL1umssCc3pdMJms0Hb2eiVLFxWK5weozN4lwssfXYDEoNPS11dl0twBkrcmqrEjgOROi8Pk//+dwBA7aZNOPThh9Dv24d+p57a5XOVzZiBweefL42YiNe0O5K+GIZBVlYWbDYbHA5Hl8ugJlpETa733XcfysvLMW7cOLzzzjvYvXs3jh07FvDnKGW8TQrifDKVSpXoosSdZ6XR3SHQPc0GzrCsFJwkolWXlcmgzs8H8PsSMIQA3tMQuhoeKPdZgzqSpGrBPv+MTwOYw2SC22f+qG8gnow4jpOSZRHiyeVyQRCEpGyMafeoFwIlCfX97vfWud2q3FxwSiV4l6tXrQbiO9qv3aM3Ubw/Mh471uXoP0++DRiE9BTLsimRWyWioPvrr79GUVER1qxZgwULFmDMmDHo379/0B+SeGKigd42tBwA0IPhfJlBsvAncwZzX2Lvem+6USBd82yMsnYRdGv69PH6O5pBdyBtR470+PUSQRCEpK/0Sfy53e6YT++JlGdPd6DvaM7w4V5/96bs3Z4Ylo24Lk/lhoqMADliRKrcXGQUFUEQBBgOHAj7OaUGDLfb674kWb8jiUTnJDxiQrVkb/SOKBoxGAyYOnUqsrKyol2eHrPb7bj77rtRUlICtVqNKVOm4Ntvvw3r2JqaGixYsADZ2dnQ6XQ477zz/Hrqq6qqsGTJEkyePBk5OTnIz8/HrFmz8N1338Xi7USN+EFM5mEXPdV6+DA2P/AA1i1ciHULF2Ljvfdi3cKFPZoT6tsLJwo3gznwe3BjCjPrabRR0E0C8Qy6A91IH/viC+xbtUr62/Pa0d2eDcD7RtXe1oYflyzB0c8/D7jvL6+8gmNffCH93XrwYErcuDIMA4vFgvb29l53s0R1b3Di1K5kcPTzz7HvrbcCPlazcSM2P/AA6n/+Wdqm8BkSf/jjj7H9qadiWsZkFWnQncp1LyeXh3xcTDi79803sfWxx7wacYCOa3nVDz94bfNswNj+1FPSPduWBx/0q4uOfPopfnz4Ydjb2nr6VlJO444d2PLgg2ijEcNdYlkWgiDAbDYndUK1iGqBvn37go9TUqjuuvrqq/HMM8/gsssuw3PPPQeO43DmmWdi48aNIY8zm82YPXs21q1bh/vuuw9LlizBzp07MXPmTK/J+Z9++imeeOIJDBkyBI888ggeeOABmEwmzJkzBytXroz124tIe3s7DAZDoosRcxXffAOnxSL97Q7wxcuMILHf2L/8xW/+Vnd6ujM7R3sYKyoSsi5lRmEhgNSu+En09ZkyRfo90PDyyu+/9/p7/C23SPM9BUGAqbKyW69nqa+XAmdLbS3sbW2o3bQprOX83A5HStx4yGQyGI1G1NbWwhYkcVy6oro3MIPBALPZnDSjzKrWrAn6mLmmBk6LBbU+/2fD5s+XfjeWl8eqaElPyrrdjbnIrFyOPlOnxqpIcTHgD3+Qfu8zaZLXY55T8Nqbm6VEmKJgU4N854oDgNNiwd7XX/faVr1uHeytrTj80UfdLXbK2/fWW3BaLNj1wguJLkpKYBgGLS0tqK+vT9oYNaIJRvPmzcMrr7wCi8UCjc98n0TaunUr3nnnHSxbtgx33nknAODKK6/E6NGjcdddd2Hz5s1Bj33xxRdx6NAhbN26FZM6Lypz587F6NGj8fTTT+Oxxx4DAMyePRuVlZXI75wnCwA33ngjxo8fjwcffBB/+tOfYvgOI+NwOGA2m6ELsoZzulDn54dcK7r/GWdAO2FCt583Z+hQnPz442BYFi6rFZsfeAD2tja4bDbIwpgjrykuhjIrC/a2NrQdPSq1DMcL9XSTQGQqFU5+7DFsuu8+OEwmOM1mKbt5ILoBA3Dy0qXY/9ZbaNq9G+aaGr+hp6EIPA9Lfb3XlA23w4G2Y8cAANlDhmDk1Vdjy0MPBcy9oN+3r1uvlwgZGRnS8mHJPswtmqjuDc5ms8FmsyHbZ0WJZGaqqvKaUlI8dSpYuRz7//OfBJYq8Tx7ugWeDzoSTiRTqTD1H/8ImXQtFfQ//XT0PfXUjtFOPqMlswYNAqdQSB0K+r17gYsu6vI5A2VFB4LnntHv3dvNUpPeJjMzE06nE06nM2lWivAVUYkeeOAB9O3bFwsWLEBjEiVn+uCDD8BxHK6//nppm0qlwrXXXostW7agKsTw3g8++ACTJk2SKn0AGDFiBE499VS899570rZRo0Z5VfoAoFQqceaZZ6K6uhomkymK7yg6xLW5k6WlPVZ8l8nywzBdVpLBsBwHhmEg12ig7Gy8CDuZGsNIrcGJqDjEoNtpsXiNBCBEplJBlZsLwH+Iubi0HgBpuS+W43qUoyDQMZ7zteUZGX6rAogNW6mQTA34fRh+b5rbTXVvYIIgwOFwQK1WJ80NoGcAGCrLtt5njm6g/CaOJDznsZRRVARWJoPbbodNrw/rmFQPuEUsx4FhWb8piqxM5tUYajMYwkqOFqinOxDx9Xi3OyEjBRPJN5cK6RrHcXC73Unb6B1RLfDXv/4VgwcPxpdffokhQ4bglFNOwdVXX41rrrnG7+faa6+NdpmD2rlzJ4YNG+bXozt58mQAwK7OZQp88TyPPXv2YOLEiX6PTZ48GUeOHOmyQq+vr0dGRgYyMjIiK3wMORyOpKnwE6m7c1CDyei8EHZniLkYdLfs3Rv3uZ6cQiE1SFBvN/ElZjH3DbrVHgGOZ7AcyRBL6XkCHOO7nq3vigGyjAwwDANrY6NX5txkxrIsHL3oBpHq3sDEm79kavD2XKJPcLv9spOLfL+X6oICsF3M7013LMdJgRAtd/U732u2sXP0kq92j4aKUMunes4LV3dOjwOA1kOHIi1iSsrwCLodZnMCS5I6xLndyRp0RzS8/PXXX5dan8xmM9auXRt0X4ZhsGLFiogK1111dXUoDpBpUdxW27lWsy+9Xg+73d7lscODDG08fPgwPvroI8yfPz9k5Wq3270m+Bs7KzWe53s0/4DneQiCEPA5BEGAzWYDy7JJO8chWroKZs1VVcgWhKDnKlzqwkIYDh6Eua7O63l4j9f3ff6sIUPAsCxsej023nMPBp51FnJGjMC+N96AvbUVYBjI1GppWG1PPxO+MgoLYW9thbm+HpkDBgTdT+g8P12dI5vBgL0rV3aUHR3vPd0/X57CPU+pIKNPH7T89hvMtbUh34v4mJjN1trUBGd7OzilEoIgYP+qVQDDYMTll4NhGOn8eKrdvBlcRgZ0QVa14HkeOSNGAB4J1mx6PbIGDULb0aNo+e03lEybFvDYhp9/RvW6dRh59dVeDQbhajtyBIc//BD21lbkjhyJ4ZddFnHiSZZlYbVa4XQ6uwy4Ql2/I3ndREj2uhdITP3rdDrhcrkgl8uT9jqRM3w4GnfsCPq453vTlJTAVFHh9Xi47ytdrpma0lKYqqthqq5G3tixAfcJdS8QSqqeo+xhw7z+3v3ii5Cp1ZD5NISZKip+f18hrq2m6mpkFBXht9de8+ooaNm7FznHHZey5ylcgiBg77//7TUKzFxbi+whQ7r1HMHOkeHgQRz99FMMmTcPWQMHgne78esrr0DTpw8GX3CB33NVr12Lpp07Mfr664M20vkyHDyIIx9/DIfRiPxx4zBswYKwy95TgiDAarUio7PBPpR4178RBd3JmrSkvb0dSo8hkSJxber2ID2d4vZIjrVarZg/fz7UajUef/zxkOVbunQplixZ4re9rq4O5h60YgmCICVK8/2AtbW1wWq1QiaTBX0P6aLdag0ZeOuOPx6WzuHVPcniLmi1EAQBbdXVXgnq7G1t0pc3UOK6jH79YD52DG6HA4c//hj5kyfD7HEz6jn0u62tDfIo9ogzWVkdn5OKCqhCzIsVBCGsc9S4YYNX76fRaISrFyTrE4V7nlKB9HmuqvL63IprCwNA3sSJXo/JNBo4zWbUHTgATd++cDscaNq9GwCQNXEi1H36QBCEgFm8K7/9Fv0uvNBvu9PphMFggCCXQ5GbC3tnr3af004DeB6tR46gYc8eqEeNCvg+DrzzDgBg9//9H4bffHO3z8Phjz5Ce+cNXuPOndCdcALUEQ7vc7lccDqdsFqtXebSCHX97q6yIPMkYy3Z614g/vWvWA/YbDbY7fakuU64PJYvy5s0CRlDh0LYvh0yrRauAOfBZrNJ70+elwfBI5GadsCAsJO0pss1k8nO7vg/P3YMWUHee1f3AsGk8jlSFRej3fN+xmr1y2SeP3261/nImzwZzQFWl2k6dAhMeTlMPqMJ9IcPd9QRKXyewuE0m9HiMxWx+ehRCHl5YT9HqHNUu2MHLPX1qNy4EWXZ2bA3N6P18GG0Hj6MzIkT/XK71G7dCltDA6q3b0f2mDFhvf6Bd9+Fo7NTpv6nn5A1aRIUccprYbPZYLVaYTQaoVarQ+4b7/o3oqD7qquuiuSwmFOr1QFTxYtZZIOdfHF7d491u9245JJLsHfvXnz55ZcoKSkJWb57770XCxculP42Go3o27cviouLe5TkTGyhKS4u9mppEQQBzs5swTqdLi0vTp5Mna1ahSecgCEXXgje5ZKGw/FOJ2QZGTAYDMjJyenRuWAHDUIdw8Cp1yMnJ0fabnW5wDAMWJb12i49PmYMLB43LGzn/oFkZWV1PUe9G2z9+kH/88/gjcaAZROJN2NdnSN7nz5o9Hhcp9MhM8Tzpptwz1MqUAwZgmqGgaO5GdnZ2dL7kclkYBgGwy+7DAXjxnnlQ9D16wf9vn1gOz9Pbo+gwlVTI/VG2JqawHSO4vCc3uFqaPA7b3K5XPpsTrrrLvAulzT81VJXh4Y1a2CtrIROowGnUPi9D/H5nK2tIT/jwaizsmDz6FUR30ekLBYLVCpVl/VCsOt3Kkn2uheIf/0r3vhptdqAjQqJIuvMTzLiiiuQP3YsGIZBnxEjYNPrseu55/z2V6lU0vfJPngw9J294v3POANls2dLKxp0JV2umdywYaj7+mvYm5q8rpeeuroXCCaVz9EJt90GR1sbfl66FAjQYXDCokV+Q8qz58+H6+yzO+aKcxzK//c/1G7YAN5gQN6YMajzOQeOlhbotFrpM5eK5ykcjs6615PQ1ha1z5JepQLDMHA1NyMnJwdWh0PaR2hoQI5P/gbxXgAmU9hlUGq1cHos8+aurUWOT76WWBEEAW1tbcjNze2yvPGufyMKupNVcXExagIk6qnrXA4nWMWcm5sLpVIp7RfusX/+85+xevVqvP322zjllFO6LJ9SqQxY+bIs2+P/bPEC7/k8LpcLPM9Dq9Um1ZyyWBEvGpxCAbnvjZpKBZ7nwTCMdK4ipe0c9mhvbYXgdEoJp1iPC1ug588bNcprXWJTiKVXovGZ8KTt7LFrb2wM+bzhniOZT9DD9vCcpppofZaSgaaoCAzHwW23w2k0QuVTSSk0GnA+N9basjLo9+2Dpba2Yw6Vxzkw7N+PAaefLp0jwH+9e0PnsDnPrLfA798bVqEAPD5j2pISKLOzYW9thenYMb85hACQ2a+ftIyZ02jsdqOVtrQUhoMH/d5HpBQKBdydvYpdXX8DXb9TSbLXvUD861/xZq6rnpa46/xOKjQa6XOpzMwMuhKH5zXOM5maTKXyqwdCSZdrZmZpKRiGgdNshstshjIry2+fru4Fgknlc8SyLGT5+dD17x9wWTmFVhvwPXEevaq5w4ejtnMUXeHxx/vtK/A8bI2N0HT+H6TieQpHoPdkra+P2mdJ/HRa6+sBQfB63LB/P4o7c3FI+3d+nrtTBl3//l6jIQ3796Ns+vSwy99TcrkcLpcrrPLGs/7t8Svs3bsXr732GpYuXYrPPvtM2s7zfNwTyYwfPx4HDx6U5mqJfuocvjJ+/PiAx7EsizFjxmDbtm1+j/30008YNGgQMjMzvbYvWrQIK1euxPLly3HppZdG5w1EmcvlSrokLulArtFA0fl5CLa8RSCeCWyAjnnR8SK2MNsMhl6XAZSExspkyOj8bAZarzuQUBnMTRUVfsMKPY8Bfv/sqwsKAt60+vJcAcB32J3Ic66Z51y4SBnLy3uUfFHMotobsphT3evP7TGMOxVwYSRJ03j0VNp70XQiT5xCISX3imQFh3SX14PRQWId0d7Y6HefIuWR6qUJ7Cz19VG/nvBut19CYMOBA0Gzz3cnebDvnP3Ww4fjeu/JcZw0WiqZRBx0V1VV4bTTTsOYMWNwww034P7778cnn3wiPf7qq69CrVbj+++/j0Y5wzJv3jy43W688sor0ja73Y6VK1diypQp6NvZSltZWYn9PsvPzJs3Dz///LNX5X/gwAH88MMPmD9/vte+y5Ytw1NPPYX77rsPt956awzfUc+IczLTsSVQ1N7Sgj0vvYQdzzyD8q++itvrikGsb8bnUBiGkYL1rkT74irXaqWgJNIM5q72dvzy6quo+/FH/8eS8OJGwidmMC//8kvsfO457Fi+POQNZWbn3CVLfT2MlZXY/eKL0mOCIODHxYux67nncOytt6TtAZeI8QimuyItu7dvX0fitrffxo7ly7HrhRew/emnvQLtligE3QC8er5FbocDv772Gmo2bgx5LMdx4Hkedrs9LZP9eKK615/T6Uzpoa+Z/fr5bfPMXt6bA07xWvbra6+h7ciRBJcmuYR7PQ9EodNB3pljZN+qVV6PiSOXmn/5Bb+8/DJMEZz3pj17sOell2D3GPacKtx2e0wauny/x672dmnEmC+bwRD0Xu/o6tU48M47EILUdbzTibajR3tW2G4QG70dDkdSNX5GFI3p9XrMnDkTP/zwA0aNGoW//OUvfm9qwYIFYFnWq/c71qZMmYL58+fj3nvvxV133YVXXnkFp5xyCsrLy/Hkk09K+1155ZU4zufCcNNNN2Hw4ME466yzsGzZMjz77LOYM2cOioqKcMcdd0j7ffzxx7jrrrswdOhQHHfccXjrrbe8fhqSZEkmh8PRrQQeqarqhx9gOHTIL+FGrIlBd7da/oCwMjiyMlnYGSK7QypzhJ9RY0UF9Pv24dgXX/h930Mt/0GSn7QMTk0NjBUVMIVYVxkAlDk5Urb9X1991W9/3uWCqaoKLo/kgKVBhpYVdA4jFNcLDyZn6FCwHAebXo/WgwfRsH07TFVVaDt61O/GwXj0aFQq2kA95uaqKrTs3YtjX3wRcj1ahmHA8zwaGxtRW1ubVBV/tFHd681sNsNsNqfcKLNB55wDAOh32mkonDABgP/3sqBz1EKwVQR6g0yPhEm7/vWvBJYk+WhKS/3WJucUCnBBpi94YhgGmUHW7+4zZQoAQL9/P1oPHUL5f//b7bLV//gjDIcOoT5A8rZU0J1OnnAFGjkQaqRYsHve6rVrUb91a9CAHQD0QUapxYJMJoPdbkeVT4LYRItoTvcTTzyB8vJy3HnnnXjiiSfAMAxe9OjpADom748ZMwYbu+gNiLY333wTDzzwAFatWgWDwYCxY8di9erVmDFjRsjjMjMzsXbtWtx+++145JFHwPM8Zs2aheXLl6PAY1jw7s4MvYcOHcIVV1zh9zxr1qxBURIEIHa7HSaTKSnXLo2mYPPQYk0MUrobwOaNGoVJ99yD2s2bUbN+PQAga9AgjLjsMrjtdsg1Ggg8HzBRVE9lFBWh7ejRbg2J9yS2YDotFpg7g6zMsjKMvPrqsIYIk+SVEWDJplAYhoG2tBSthw97Zd0PShCgLS3F5L//HU07d+LY//4nPZQzdCgm3XNPl0E3p1Qia/BgGA4e7HL4uNNqhb211W9+ejgKJ0xA3siR2PfWW9Dv3w9BELx6LMXg2W23w3jsWMhlXLRarZS92u12QxZm0qlURHXv78TMuZEk9EukslmzkDtypDTdJGfYML8G1RGXXYYBf/gDMjzWT+5tND6BodvhiEmdnYoYhsGJS5aAdzjgstl+v6cJc413bVkZ9AcOSH9nFBVh9HXXQXC7ezyaUbx26/ftQ/8e5OuIp0l3342Kb75B486dsNTVIW/kyKg+f6ARK/p9+zDwzDMD7m+pq4Mu0LKzHufW8/E+U6ZAN2AADr77Llr27sXgCy6IywggjuOg0WhgNpuTaph5RD3dn376KQYMGIDHH3885MkbNGhQ0PU5Y0WlUmHZsmWoq6uDzWbD1q1bccYZZ3jts3bt2oC9DmVlZXj//ffR1tYGk8mEzz//HEN8bqgWL17stf6d78+sWbNi+fbC5na7O4Yzp3lFkKhgL5Lh5dKxhYXI91h2QaZSQZWTA02fPlBkZsbsPfW0p9uTGPTI1OougyWS/DTdDLqBIMPFgxDneKvz8qSeMk8ZhYVhZUH2HGLelUiHvyoyM5E/ZgxYuRwOoxGWEM/TVTk4joNKpeoVc7up7v2dw+GAWq1OuZ5uhmE6EiuyLBiWhaZPH797PJbjenXADfhf+2iIuTd5RgaU2dkR3dMEqlfUeXlQ5+f7NWw4I1zqz1hREfGx8ZZRVCTVz+HmXOkOc02NdE0WRyiYa2pg71zuy1dXZfCd2qXU6TpWP+kcpdbe1NTzQodJJpNBqVQm1RDziILuiooKTJgwocu5wgqFAnq9PqKCkZ5xeCwB0NvY4/CZy+js6bbp9RElh9ANGBD3XvpoBt3WOF44SeypcnK63VOjDbImZVcBvDo/H+r8/G69lkgMusP5/PUk4Q4rlyNn2DAAoQPrcIJ/lmXhdrvhcrkiLg9JHTzPw+l0plzATcIn9xlBGCy5I+k+33pFvF9hWBYan5UMLD24l/HsTU92Yp3a3emM4XA7HGhvbgYAsEqltEKB3if3hqirjiZzdTUcPgk1ZSoVsgcNCvm8sZJsCU0jCrpVKhVMJlOX+1VWViKLhp0mhN1uT9uhjE6LBfvffhv7Vq3CYY/kfaJ4XEwVnonJGhthqqryS/wRCstxyBk+PFbFC0jsnWhvbg45FzWQlr17ceA//4lFsUgSYFg26Lx8R5BrfbCe7nAS6USabEddUBD2yIqeJnqSsqWHCKwt9fWwddHIJy7bQkF37yDe4FHQ3XuIyR2P/e9/qFqzJtHFSWmhru++dY61Bz2/8ZxfHErN+vU4unp1yJ7YDI/pjN29dwvHb//+t/R7V6PJLHV1cJhM2Pf222g9fBg2gwH7Vq3yKv/O559H7aZNXsdJzxvn8y4G3clS/0YUdI8YMQI7duyAJcRcvubmZuzevRtjx46NuHAkMjzPh70+XSqq+/FHNGzfjsadOwM+3nf27LiUQ5rXXV+P/W+/3e2b/PzO70a4Gc17SpmdDU6hgMDzUstmuH597bWAy0CR9BGshzrY9ozCwoC948Unnui3TanTef1dEOFnnwkj27n4/Y9W0B1sCTRROL3dDMOgtbUVVvoOpT2Xy0VBdy9Q6pFIzqbXw1hejsrvvsPRzz/36+kj4WNY1qteyB48WPrdtxe8J4nF9AcOBM20HU+HP/kEVT/8gNbDh4PuI45E493ubt+7dVfuiBEAgNYjRwI2BDgtFtRv3YrG7dtx9PPPUfntt3734oEaosVOprZjx+J63jmOg9PpRFNTU1L0dkfUFTpv3jwsWrQICxcuxEsvvRQwuFu0aBGsVisuvvjiHheSdI9Y6afrfG7fzJhAx/Imw+bNg9NqRdbAgXEpR0ZREVqPHIG1oSGiSrZg/HgoMjOl4TyxxjAMMgoLYaquhrWhwWvNVUI8g+vSGTNQPHUq3DabV6ZeTwzLInvYMLT8+isAIHf4cAy+4AKo8/MxcdEi8DwPfUMDcvLz/YLrrMGDMf6WWyKaG5o3cqRXK/q4m24Cp1SClcngslqhKS1F1Zo1sLe2wmk2Q67Vdvs1AEh5Fiz19TDs3y9lc/al37cPJSefHPK5MjIy0NbWhuzOZW9I+uoNS3USYNB55yF/7FiUf/UV2o4eRctvv0mPxXsIbbqZeNddaG9qgiAIXiOwotnT7bJaYaqsDJwULAGMx44hZ+jQgI8xLIuMPn1gqqyEta4uavduco3GLxGqprS0Y2RWiGSkYoO2qaoq7IZzdWEhWI4D73R2PG8ccwFlZGTAarUmxbSfiGqFm2++GaNHj8Zrr72GyZMn47HHHgMAHDlyBM888wymTp2KN998E+PHj8fVV18dzfKSMIjD29K10pcFWE5LlZsLbVkZcoYN81pLNJbEIT+W+npkd87/BMJfs5phGGQPGQJOqYxJ+QKJdF63LM2z4JPfR24AgK5/f2iKiqDr3z/kMZ69zuqCAimI1hQXQ1NcDG3//tCWlQWsuLMGDoxoabyswYOlpGusTIbsIUOQ2bcvNMXFyBo8GDKVSpozHq3e7kC92Uxn5W04dAhupzPk83AcB5lMBmcX+5HU53a7kyZpD4kdluOQPWSIlBTV8xoRzugXEpxco4FuwICOOsLj3sOzjgI67r0i+a6J1+5QU4firau6KhbJ1HwbMVxWKzi5XLq39c2LIo5s8yyDLczluFiOg7rz/iAWS5+FolQqIQhCUgwxj3hO99dff42pU6dix44deOCBBwAAGzduxKJFi/DTTz9h4sSJWL16NeRxCoDI71wuF3ieT9ugO5B4ZkQUiQGGtaEBsjgGzj0RadDt27JKQXj68ezp5sOsnMShaADCWzosCjiFAtlBegRE4s1E1ILu/fv9hsSpc3Oh1OnAO51oO3q0y+fiOA52u71H5SHJz+Fw9Kq6t7cTrxGegUgsskwT+K1w4bbbg2bZDkVK6pUk87qBbgTdUQxYg00dC1Z/iveP7R7Lzvp+1jP79fP62zPRsHT/GeegW5QMQXfEmbaKi4uxceNGfP311/jiiy9w9OhR8DyPvn37Yu7cuTjvvPN6bfbsRHO5XL3u3Pf05joS4gWrvbk55vNsoiVU0F29di0cZjMsJhO0Z58NR2srGrZtA8MwaDt2zGvf3OHDg86pJ6lJ7jFMLNxGGc8ebMPBg1EvUzC5xx0XsjdJW1aGpt27YepsqXfZbKj89lswLAu309mR3IzjIHTO8SqeOjXg8+gGDgSnVMJpscBUXQ2d5w1F5/zyup9+gn7vXuR6JEZ0Wiyo/O47FE2aBG1nxl1xbpnvut8kvdjt9oQPYQyEdzpR8c03FBBGmbqgAOq8PLS3tCS6KL2Spa4u4EgqAKj64Qcos7NRcPzxqPz2Wxg6k+zmjBgBw6FDMNfUwGE0QuGTcyQeHGYzqr7/XvrbptdDEATY9HqU/+9/fvuLvfyWujoYy8vRtHs3BsydG3LVEZfFgqMbNoCVyaDQ6VA2YwZqN29G3Y8/AkDQUaHasjI0bNsGY0UFjv3vf9L9taakBKaqqpCjC3JHjICpslL6u9VjOb1orqATiZQNut98800olUpcfPHFOOOMM/zW4iSJY7PZYLFYet1Nnee61/ES6VzRRJIueo2NEHgeTGePjMNoxJHPPgMACIKA3xob4TAYYA8wV11dUIC80aPRuHMnZGp1/ApPYsozEO3OGtx5I0eiZe/egOtvx0reyJE48vHHQT9/vi31Ddu2hcwqXL1uXcAEjCzHIWfYMDT/8gtaDx3yDrqB34PuffuACy6Qtjft2YPqdetgqavD2BtvBNARdLtcLrhcLhoBlqaMRmNSzBsMpOW331DpcZNPokNM7lizcWOii9Ir9J092+tabq2vR97IkX772dvacHT1ajAcB4VOh/KvvpIek2u1yCwrg6m6Gvr9+9Fn8uS4lN1T9Zo1qF63zmubTa/HkY8/DrgEnWcnz7EvvkDrkSNQ5eV5JfTz1bR5M1q2bv39OYqKcOiDD7yftzNviSex/jQcOCA1VIj7dqVwwgRUfPON9Lfn/404YjLew8uB5BlpFlHQ/ac//Qmnn346JUlLQu3t7dDr9b1iqTZVTg6GXHQR3DabtKZuPDEMA3V+vlcvd86wYRh07rlxL0u4VHl5YAIks/AdOmuqqAh4fOGECRh45plQ5uSAYRjo4pS0jsTHlL//HabKSuSNHh32McddeSVafv0VuQFufGJFlZuLsX/5C7gga92LNw3tTU1w2WwQetDCndmvH5p/+SXgaJrsYcPAcJw02kWcSy6+XuuRI3Db7eCUSnAcB5vNBrfbTUF3mjIYDLBYLMgJ0vOWSDTPPHZyR46koDtOBpx5JjQlJWg8eBD6n38OOnJDnCIluN0wHDrk93juyJEdQfe+fQkJul3t7X7bzFVVQUdNKjIzpcRnYl2k37cvZNDt9nmNxl27/PYZd/PNMBw8CE6phLIz0WewRnd1fn5HMrQAWcA1xcUYcsEFyCgsxPF/+xtcNhtcVqvXvUSGuN54Q0PcR3xxHAeHw5HwkWYRTTzKy8tDbhwzz5Hw2e12qFSqXnFTl1FYiLyRI1E4YULCep0zfFr+socOlYaTJiOW45BRUACgo7c7FFmAgCZ/zBiocnPBMAwKxo+Hshc07vQmyuxs5I8dK42ACAenUKBwwoSAn5dYyh4yJGhmdUVmprRMmaW21mvofDCec888icvU+CaVATq+I+JqCYGGuwtutzTsnmVZaTlHkn7EBKZarTYp53THM2Fnb5M1aJDXfOPeNtIwnliOQ8Hxx0PbmeQznF5TQ4BrszgX33DgQEzWvu5KRoAM5ObaWmR5LJHmS+ztFpP1th46FLTeAgBl572eKFAdJddoUHj88cgbOVK6d/VMRuqJ4Tivcnt+zrWlpcgeMgQAoBswALkjRqBwwgSv4e/q/PyOKV4OR0Rz8XuC47ikqH8jqhmmTJmCPXv2RLssJArsdntSVvjpym/phiRY97ErngngQgk3CzshyUgKlsPM92CprQ38PGKveXNzwO+ElGwtSFIezxsdhmESXumT2BCD7mQcWk5iyze5Y2YXqz6QnlN53Md0FTSbAjSYZvbtC7lGA5fNBmN5eSyK2G3m6uqQq+/4Du/mXS60ecyZ9uPT+OO7tC0fYjWNYL3dnh1NvknTusJyHNRip0+c53V7Tu9KpIiGl991112YPXs2Xn75Zdxwww3RLhOJUG+o9I3l5ahNomFcvj3dxqqqBJUkfGJL5ZFPPwXDcWAYJuw5Ng6TKZZFIyRqtKWlaNm7F+bqaq8l/YIJFpwrtFoos7Jgb2tD086daO5cl1yUe9xxOPr55x1DyR0OmCorpUQ1ANC4YwdUeXkoOuEEgGUTXumT2HC5XClV/7qTYH5jOvFM7ph33HFJE8ilK3lWFli5HLzTCVtzM2QZGajfuhWcUgmGYUI2ttr0ejAsi5wRI9C4fTv0+/YhO0QPc7S4HQ7UbtoETqFA/U8/+T2u378fyhD3YoGyjUvJERkGnFIpfa8ZuRyGAMPJPYVadUNMRupVfpvNqwzyzExk9u0LUzfuezVFRbA2NMBaX++1+kmsiSPN3AkY1eApoqBbEATceOONuOmmm/Dhhx/ioosuwoABA6AOktRmxowZPSokCY9Y6SvTeBjZzuefT3QRvOgGDPD6WxtkCYZk4jk86PBHH3XrWEUKJo8jvZPY022sqAgr6A41TE9bWgp7WxsOvv++32MZRUVQ5eTAZjCg9fBh/Praa37Pe+yLL1C7YQPGLFoEi8UCtVoNTQRrlJPkJTampMrQ4mD5EEhkxBEvnEKB3FGjcOzLL2lIfwwxLAtNnz4wVVXBUlcH/b59qP/557COFecu5x13HBq3b4/byhv1W7fi6Oefh9wn1LDrQEG3saICxiA5eLqav6wLMSIjUE83p1b7lSFv1CiYqqrC/qxn9OkD7NkDS4IymLe1tUGhUEARIut7LEUUdM+aNQsMw0AQBHz33Xf4PkRGTBpOFz88z4Pn+ZRpaU8H6rw8jPrTn+Cy2eBub0dRAhJydJc4vDyY4jlz0LhunTT0SDdgAEpOOgluux15CcgST0gksgYPBsMwsDY2wqbXA+iY1zronHPAO52QazTgXS40bNvW5Trb2rKygBllgd+zF9du3hxyGTO70QiVSgW9Xk9BdxpKdA9KuBiWxdCLLopLz15vIt4LsAoFtCUlOO6KK6AII5cEiVyGGHTX14fd2zrkwgtRNGECgN+HR1vr6sC73WBjfO8caKSgQqfDkPPPx9HVq6V6Cui4Txt8/vle+2r79vXqzQ6XUqdD3ujRqN28Wdo24A9/QEmIJGy+Qfew+fORM3SoX6NA2axZ4JRKFIwbF1ZZErlWt1qtRltbG3Q6XWoF3TNmzEiZ1tzeRAy6e8v/TbLkYk3EcmU9oe4i6M6fMgW8wYDG7dsBAH0mT0bRxInxKBohUSPPyIBu4EC0HT0qBcO6/v1RctJJXvtlDx2KrY8+GvK5ArX6e86Hk4LuIIG5iOF5aDSapFi6hEQXnwL5PABAW1ISdF160jOe9wKFxx+fwJL0Dp5rV2cNHtzlNDmlTueV7VuVlweZSgWXzQZrfX23lsqMhDpAAurswYNRMH48mnbt8gq680aP9ht+zXIccoYPR3M3c2plDR2Kweefj4Zt26QRXWUzZ4bsnfZtMBKvGcqcHHAKhfQ8nEKBspkzwy6L9H9WXx/3TOIKhQLt7e0J7QiOKOheu3ZtlItBoiFVKv2eUGZnSy1t7V1k3yaBcWG08OV2znUiJJXlHncc2o4eDboEHtCx/Jh44xWMNkCWdJvBIP2ePXQoWJnMaxsAaS64yNLQAHl+PlwuF3iep6SXacTtdveaBm9CkoG4BJWlvj5gtm1pv855xL4YhoG2tBStR47AXFMT86A7ELH3W1taiqb/Z+++45uq+j+Af25206R7D7ZsFJHhQIaIgjhZ7gcRRR8cPCKoOH6IuBUHKD6KCrgVnI+iIshUZIMgG4FCW6CLJs3Ovef3R7nXpEmaNM3u9/169QW9Izn3JL3nnvU9LpVp5mPkTEaXLlKlO9D51A6jETKFor7CvnNnwGnj5HKPdHAch+T8fJ9D2v1JysoCx3HgbTbYa2ulof6RIpPJ4GgkgFzY3z9q70xCriVUul0rjA0fcEnopHXqRA+QJO6J8ywbW6NYfPBqjDotDQqt1ud+uUrldakX1wo3ANQePIjq7dthN5niZjgyCYzT6YzZe6alqooaUUnCEXtNrZWVjcbkEMsBr69x5t5vLCnByS1bQvpcaamsxKmtW8EYg+nECbdKtej0wYMAPBt2fU15cu391hcXB5QOcc56Y/ngja/lb73NLQ+UTKmMWgRzoD6KeTRHmoWk0m2321FeXo5ql6ERJPJa0tByANI6vKTp/N00xaG5ABpdwoKQWJacnx/QWvLeerJdcRznc01wUaaXBxpdQYFbhf7Qd9/hwBdf4MAnn1CskwQTyyMXds2fj1PbtkU7GYSElFKvh1KrBWOs0ZGPmV27AgA4hefgXvG+Xvb779j78cfY9NxzIUvfgSVLsOejj1C9eze2vf6615gfYtmja1CB9jUN0LU8C7RnXqycixV2juOAAO5VvuZpJ5+pjMuDfDYU53UHumpOKMnlcjgcjqh1UgY1vFz00UcfYc6cOdi2bRsEQcC4cePw/vvvAwC+/vprLF68GM888wzannl4J+HVknpOUtq0QYeRI6OdjLjVY+JElP32G2QKBZgggPE8OJc1FAGgw3XXofLPP5HZrVsUU0pI8DiOQ0bnzij3sjyLq0AeXnSFhVKPQVJ2Ns4aPdptf3qXLsA330i/Z3TqhHbXXgtlcjK2vfqqWw+K4dAhqnQnGJ7nY7bSbaapWCQBiUOdTx86JFXgFElJKBwwAMzpBCeXIzk/H6nt26PdVVe5rdwianjvF5zOkM01dlosAICqv/7yCH7WauhQMJ5HXp8+AOpXhul8882wVFRAcDpRePHFPl+31wMPoOqvv5Bz3nnQ5uaiet8+MJ6HQqMB73CA8TxkKhXMRiPUKhXy+/UDUD9iq/NNN4HxfEAV5qJBg8A7HEh3WYMeAPL69IHdYAg6boE2Lw/YuTNqPd12ux1OpzMqwdSCrnTfcccdWLBgARhj0Ol0qKurc9vfsWNHfPbZZ+jVqxemTp3a7IQS/1rSnLK2V1zht+eJ+KZOTUXbK67w2C4IAmrOVA50hYVRmeNESChldOkSskq3qNWll3o8iGizs5GUlQVLZSUAoO2IEUg+85DXY+JEbHrhBbfjozmvjIQWYyyme7pdNTbVgpB4o83Lw+lDh6S50Xl9+6LN5Zd7HFc8eLDX85NyciBTKCC4NILaDYaARkgFqmEPd0aXLmg7fLjHcbnnnRfQ6+mLi6Xe69R27ZDarp3HMeKzXHp6utt9qSlBcTmZzGteytVqr8+PgUqOYk+3TCYDz/NR66QMqoT4+OOP8f7776N79+7YtGkTahvMWwOAbt26oaioCD/++GOzE0kCE8st7YQQEg1pHTuC87MUTFJOjt9pFG5D0H0MTWsYbVZ6/exsj9evo97HhCEIQsQj8TaFuDQSANSVlkYxJYSEVnPmFwP1EcEbvkbd8ePNes2GGsb2sLbwqbjaM3PxzSdPRrwRUCaTQRCEqI00C6qG9s4770Cn0+H777/Heeed57Og6dGjBw4fPtysBJLAJfKcbsYYavbv97rOISGE+KLQaJDqZ4qTtwevhlyj45orKrwek3Fm7mBDnEzmFpSGA1BNZWPCYIyBMRaTjd61hw7BQeUmSVDNrXQDnjE9wt0wFY1h1bEkKTsbHMfBabHAbjCE5DWdViuqdu8GC2CuNsdx8VXp3rFjB/r164diP5HzMjIycLKFf7kihTGW0D3dtYcO4c///leaI0MIIYESo7Y21uMtTlfxdQzncm/11Sue2r69tK/h6zR8sDMcO9YiVpxoCcTh5bHW6G06cQLb33yTVvogCUuMYN4cDacXhbvS3dj62C2BXKmEJjMTQOgaII4sXYpd777rdyoZUF/pjtb0rqDmdNtsNqQGMN+hoqICcj/D+kjoxMucsmBEY+4HISQx5F9wASynTiGnkTlzRYMGQXA6kde3r89jzrnnHpzassVnkBu5UomOY8ei7vhxj6A9Hg92x4/D6XRC4SWiLokvsdrTbWtQ2VbqdOh0ww1RSg0hoadISoKuoAB1ZWVBv0bD+EChHl7uSpudjY7XXx+2148X2rw8WCorYT55EukdOzb79RwmEwCg8s8/UXDBBY0eG81lw4Iq7QsLC7HHS+h7V4wx7N69myKXR4hY6MdaS3uohDKoBSGkZVFoNH4fdJKysvxWSNLat0eal/W4XeWed57XgDhuPd0cB3N5OXiep0p3AojV8leu0bj9ftaoUdLySYQkiowuXZpV6dY2GKJuramBw2yGUqttbtI8dLz+eq+Bz1qa5Lw8VO3aFfKh9rWHDoG32yFvJDJ5NJcNC6pZdsiQIdi7dy++/fZbn8d8+OGHOH78OIYOHRp04kjgxGAEsVboE0II+Sdiq8huMMBEw34TAkUEJyR6xOlDQHB/i96Wzwplb7fWx5rbLVm41uoWnE6cPnCg0WPkcjl4no/KvO6gKt1Tp06FWq3GTTfdhNdeew1lLi1M1dXV+O9//4tJkyYhOTkZ999/f8gSS3yL1Zb2UODtdpjKy9220UMOISSeeJsHXkPB1BKCWP7GOoGWqSMJSN+6tfR/Y0lJSF7zxKZN4EP09+LaKEDquVa6A7132mprYa2uhvHYMViqqmA7fRoAYKmqcguyXLlzJ4xnGk14u10aBeG0WmE6cQJyuTxqEcyDGtd21llnYdGiRfjXv/6FBx98EA8++CA4jsOiRYuwaNEiAIBSqcTHH3+MVi5LVZDwidVALqGw96OPULlrl9s2wW6PUmoIISQ0ag4fBvzMPyPxIR7KXqfVGu0kEBJyMpfYUVyQcRUyOndG9d690u+ntmwBb7Gg+x13NDt9KW3bQr5+PXi7HYiD+0QkiL3/TrMZjro6qPR6v+dsefllae626IKZM7HxmWfctp3YuBEnNm5Ep+uvx6nt21Gzbx+633EHSteuRc2+fThvyhQIen1U1uoOOurHmDFjsGnTJowZMwZ6vV5q6dVoNLjqqquwfv16jBo1KpRpJX4kak93wwp3ds+eSAtB4AVCCImkXlOmIKNLF+Scey44ALUh6pUh0RUPvdxZPXogp1evaCeDkLA478y9tcO11wZ1fqebbkJm164ouOgiaVvV7t0hSZtMLke7a65Bds+eHkHbWiq5SoWkJkYwb1jhBoBTW7f6PH7f55+jZt8+AMDRn36C/cx66RU7dwJA/PR0i7p3747PPvsMjDFUVVVBEARkZWXFXATPlqClzOlOad0aXf/1r2gngxBCmkxfVIQed96J6n37cGrrVtSVltKyYQkg1ivdSVlZ6DZ+fLSTQUjY6M7cW4Ol0unQ/Y47YDcYUPbbb9J2JghB9567KrjgAr9RtVsabW4uLFVVMJ04gbQOHYJ6jRqX0QmNqSstlYa0V+/ejaz+/WGPwojZgL5JcrkcEyZMkH5/6qmn8N1330m/cxyHrKws5OTkUIU7ShJ5Trfr0juWysoopoQQQppPXD7MUlkJi8tcNBKf4mVONyGkcaqUFLffLVVVUUpJ4tOeWWO9ORHMaxoETdNkZHg9zvX+XFdaCt5kit1Kd8MC5cknn8Q333wTrjSRIHgr9JkgwHZmOIXIbjBAiMI8hmDZjUbIXJbU8Ta8hBBC4olKp4M6LQ0AUE3B1OJeojZ4E9LSNQziGyi70QghCsOX44nYoXb6wAGf9RLebofDbPaoy4hYg5FiukaG7zstFun/xkOH4HA4Ij6vO6BKt0ajgcFgCHdaSDN4K/QPLF6MP2bOhOHMvEFjSQnWP/kk9pwJdhfrrNXV+GPmTNSVlkY7KYQQElJib/fpo0ejnBLSXFTpJiRxaNLTpf8HU+m2VFXhjyefDLrC3lIkiz3dp05h53//6/WYHW++id8ffxx/zJwZ0Gsq1Gqf+1wr7rX794Pn+YhXugOa092+fXv8+uuv+PXXX9HhzLj7uro6lAQYBIYimEeHuaICAFCxdStSWrWSApJV7toFwel060GORcaSEo9WrLNGj45SagghJHR0RUWoLSkBT70hcU8QBMhdIigTQuJXj7vuwqbnnwcAmINYR9p49ChNNwmA69TR04cOeW28NB475nGeOjUV4DhpyTCRvrgYRZdcgrSzzsKRH3+EtabG53vXHjgAp8MR8WBqAdW6/vWvf+Hhhx/G0KFDpW1ffvklvvzyS7/nchwXlQhxLU1jf+DVe/ag/bXXun3Baw8fRvpZZ0UiaUFT6nRuv3e7/XZkde8epdQQQkjotBo6FJkXXYSUlJSoLF1CQidRl+skpCXS5uSgx8SJ2PnOO0H1Vjd8diXeyVUqt9+t1dVSRPPG9JoyBSq9HuZTp6TGEZlCgV4PPAAASM7NRW7v3tgye7bPkbJOqxV1JSVwFhQ08yqaJqBK97Rp05CcnIzFixejpKQEhw8fhlarRVZWVrjTRwLUWCAXc0UFLJWVbg8F1Xv2xHylmxBCEpVMLodcLofdbqde0jhHw8sJSSzi0GdLRQV4hwNypTLKKUp8dcePB1TpFiVlZ0OTkQFrdbXX/RldunitdKtTU2GrrYVh/37wffoEnd5gBBxqfNKkSVi5ciUOHToEoH6d7sOHDwf0Q8KLCQIslZVuQQJ4hwNwqYRX79njFtShes8eAIDA8x4BDHiHA4LT6TG0O+piLT2EENIMcrkcTqeTlg2Lc94q3QLPS2VoTJanhBCfVKmpUGg0YIzBcupUs16Lhpr7JjZuAPBaQW7YG+6K4zhkdOnic7+vfeJ2w8GDsNlsgSY1JAKqdJeUlKDapSVh4MCB6Ny5c9gS1Rw2mw0PP/wwCgoKkJSUhH79+uGXX34J6NzS0lKMHTsWaWlpSElJwTXXXIO///7b67HvvfceunTpAo1Gg7POOgtz584N5WU0ybqXX8buOXNgOLMIvN1oxB8zZqDWpcHj4NdfY99nn0m/m0+ehLW6Grvmz8eGWbOkCnvF9u1Y9/DDWPvQQ9j84osxFencRsH8CCEJRC6Xg+f5hKh0t8SyF6h/oG44vJwJArbOno3NL74I3m7Hpueew9ZXX6WHb0LiBMdxSD4z9Li5AdHsPiJvE0h5DNT3dDfkb430zK5dfe5Lad0aCq3WY3t6x47gOA7WkydhbGaDSlMFVOlu27Ytpk2bJv3epk2bmB1aftttt+GVV17BzTffjNdffx1yuRxXXHEF1q1b1+h5dXV1GDx4MFavXo1HH30UM2fOxLZt2zBw4EBUNVin7+2338Ydd9yBbt26Ye7cubjgggtw//3344UXXgjnpfmkP/OltZy5MRhLSuC0Wv2eZywpgbGkBHaDQer5PvLTT9J+86lTMMTISIWkzEyaz00ISShipTsR5nS3xLIX+KcXy7XSLTidMJ04AfOpU6jcuRPWmhrUlZbCRCtxEBI3xF5YUxDB1ERJWVnIpGdXn9pffbX0/0BWKkpt1w7K5OR/fu/QASmtWyOrRw+PYzmZDEUDByIpMxPanBxpu0Krhb51awBnRgFHsNE7oDndDecLL1q0CBzH4fbbbw9bwoKxceNGfPbZZ3jppZcwdepUAPVB4Lp3746HHnoIv//+u89z582bhwMHDmDjxo3oc2aM//Dhw9G9e3fMnj0bzz77LADAYrHgsccew4gRI7BkyRIAwJ133glBEDBr1ixMnDgR6S7LDURCeps29Wk7c2Pw1rLjjesXvGrPHuT06gVdYSHMLi0/1Xv2IO1MxPpo0ebkoM8jj0Q1DYQQEmpiRS3ee7pbatkbiOq9e//5/549ja4jSwiJHVqx0h1kT3dyXh56P/RQKJOUcFQpKej//PNY98gjsBuNsNXW1kcnb6DP9OnQZmd7bJcrlTh38mSfr9966FC0HjoUf3//Pcy//iptz+jSBYbDh1Gzdy/yL7ggNBcTgIB6urVarUeLcyxasmQJ5HI5Jk6cKG3TaDSYMGEC1q9fj2NeQs+7ntunTx+p0AeAzp07Y8iQIfjiiy+kbStXrkRVVRUmTZrkdv4999wDk8mEH374IYRXFJi0M5Vuc3m513ljrq1CrowuQzlq9u4FEwSoGzy0iD3ghBBCQk8cnhzPWmrZG4gal0p3FZWnhMQNXX4+gOYPLyeNk6tUUk90IL3dwdAVFrr9ntGlC8BxMBw6BEcE53UHVOnu0qULli9fjvfffx9r1qwBAJw4cQJr1qwJ6CdStm3bho4dOyIlJcVte9++fQEA27dv93qeIAj4888/0bt3b499ffv2xaFDh2A0GqX3AOBx7HnnnQeZTCbtjyR9fj5kCgUEhwOWykq3AGoAkN6pk9fz6kpLpREMDpPJ63p4phMn3BaUDyVv89sai8JOCCGJRi6Xw+FwRDsZzdJSy95AOEwm6f/Go0fhMJsBNF7WhawMpLKUkKBpz1S6badPBzRlkwRPrBSHrdLdYISRrrAQqpQUCA4Han3EDwmHgIaXT5o0CRMmTMCdd94pbfv555/x888/+z03kut0l5eXI//MH4krcVtZWZnX86qrq2Gz2fye26lTJ5SXl0MulyPHZX4AAKhUKmRmZvp8D6A+0IxrpDzDmcBggiA0r6eD46DJyYG5rAz7Pv0UhqNH3XZrMjK8nuOoq3PbVLV7t9fC/o+ZM5HZvTu63nZb8GlsgAkCdrzxBuQqFbpPnAhOJsOJP/7AoW++geB0Iue885Dr0vMRip4g8SEnEXqWwoXyKDCUT/5RHgWG4zg4HI6Q5JHMT9CZcIn1shcIX/krCILH99zX6zHGULVnD5wmE/7+7jswQUD+hReiw8iR0jGWykr8+eabKOjfH8VDhgSdrtK1a/H3t9+6pTOa6H7gH+VRYCKVT3KNBqqUFNgNBtSVlSHlzKjSQNIHAAzR+7uLt+9ScmEhsG0b6o4fd0uvWCNhzbxPq9PTIVepwNvtUr6kd+qEExs34vTevRCGDm3mFQRW/gZU6R4/fjxycnKwZMkSlJSUYOXKlcjJyYm5COYWiwVqtdpju0ajkfb7Og9AQOdaLBaofISw12g0Pt8DAJ577jnMnDnTY3t5eTnqGlSAm4IxBkVWFlBaitojR9z2yVQqqDt0QFpFBWrO9DbkXHwxavfsga2y0u3YUzt3Qte2rdeKd+XOnagsL4f8TH40l9NikRoHyvfuRVJ+PvYvXiztP7l5M9StWoExBifPo6amptnvyRiD6UyvA62p6h3lUWAon/yjPAqMw+GAyWRCaWlps9frLorSfOFYL3uB8JW/PM/DbDaD4zgoz6zlKzgcPnurK/btQ9WmTdLvZb/9hoyBA6Uovaf37oXNYEDp+vXQ9eoVdLoOffON9H+n0xmSMrQ56H7gH+VRYCKZT8rMTNhqa1Fx6BB4L3ONvakzGsEYAx/Fv7t4+y6xlBQwxlBbUuKWZ+xMo2atwQCrIqAqq08pXbuidu9e2DUa1NTUQFVcDGzYAGNZGcrLy5udT4GUvwFfwYgRIzBixAgA9bX54cOH4/333w8+dWGQlJTkdc0165lhIUlJST7PAxDQuUlJSbDb7V5fx2q1+nwPAJg+fTqmTJki/W4wGFBcXIz8/HyPYXlNIQgCjrZuDcOff7p9aTSZmej98MPgZDLkdegAdvPNAOoj+u01mVDRYJ6+9cQJpBYXg+M4FFx8MbJ79sQOl+VYlBYLUr30SATDoVZLaXWWlSG9a1cokpLAuwzhsZeWguM4KOTykATIER+C0tPT4+ImFA2UR4GhfPKP8igwdrsdgiAgLy9PqrTFm1gve4Hwlr81NTXQarXQngliytvtPr/zfHW1x74kAElnyjhncnL96IfTp6HXaqHw0iARCNf3UCgUUQ8yR/cD/yiPAhPJfKopLobp8GHAYAj4b4irrATHcZBH8e8u3r5LOrUaRzgOjtpa6DUaKM7czzmZDBzHITUlRbpHBiv9llsg8DxkZxq3U3r3hq6wEPKUFOTn50dkpFhQzQbjxo1D//79Q52WZsvPz0epl/kA5WeCIBS4rAfnKiMjA2q1WjqusXPz8/PB8zxOnTrlNszNbrejqqrK53sA9a353lr0ZTJZsz9sbxFRrVVVkLu2DLm8R0pxMSpc5sApNBo4rVbUnFnrW8ZxSD0TUl9kOXkS6SGKZO56vTV796LNZZchrX17VP31l9t2b8cHS1xLleO4qA3DjHWUR4GhfPKP8igwrnkUr/kU62UvEN7yV6lUQhAE6XWYy+uJZauorrQUae3b4/ShQ9I2y8mTSD4Tmdd1XVpLeTlS27ULKk2p7dpJcxUtlZVR/27R/cA/yqPARDKfdGfuK+aTJwN+L7GSyyF6U37i7buk1umgSU+HtaYG5vJyadUksbmAC1H56PoaMo0GSVlZcDgcESt/g3qHBQsWxNxyYQDQs2dP7N+/X5qrJdqwYYO03xuZTIYePXpg8+bNHvs2bNiAdu3aQa/Xu71Gw2M3b94MQRB8vke4JeflNak1K7lBJL/0M1MF7C55x8lkSGvfXvq9OWsVNsZ49CgcJhNkDXp57A0+R0IIIbGnJZe9QH1PsiAIELwExEt3mYYnUyjA22ywNxjO7is6cp3LCiNNTlOIpoIR0pIlu0QwpyC/4SV2HoYrmFosiP3mjyYYPXo0eJ7HO++8I22z2WxYsGAB+vXrh+LiYgBASUkJ9rr0oornbtq0ya1A37dvH3799VeMGTNG2nbJJZcgIyMDb731ltv5b731FrRarTQEP9KUGg00OTkBRyv1Gj7fC9ftoVg2Yc/HH2PXe++5LW3GGMPvTzyBCh8RbgkhhMSullz2AkDFH39g14sv4shPP3nsE8tQjuOkB3jLqVNux5h9NGg39eGz4s8/sfGZZ7Bu+nRU7d7dpHMJIZ60ubkA6lchaBh8mISWWC8xHj8O88mT2PziiwkXNT6g4eWXXHIJOI7DokWLUFRUhEsuuSTgN+A4DitWrAg6gU3Rr18/jBkzBtOnT8epU6fQoUMHLFq0CEeOHMF7770nHfevf/0Lq1evdmu1mjRpEubPn48RI0Zg6tSpUCqVeOWVV5Cbm4sHH3xQOi4pKQmzZs3CPffcgzFjxuDyyy/H2rVr8dFHH+GZZ55BhrdI4REgl8uRXFAA68mT0rasHj18Hq/UaqWhHACQ0qoVFFotnGeWMxHl9euHv7//HkB9TzdjLOj5IUwQcGrLFgD1vduEEELiI9BNY1py2QvUrzPrtFi89kxndOoElV4PVUoKdIWFMB475tFj5rOnu4mV7r0ffgiB5z22N/YsQAjxTa5SISkrC5bKSpjKy6E6M/KGhJ7+zJTW2oMHUapWh210bTQFVOletWoVOI6D+UyFbNWqVQG/QaQfJj744AM88cQT+PDDD1FTU4Ozzz4b33//PQYMGNDoeXq9HqtWrcIDDzyAp59+GoIgYNCgQXj11VeRfWaulWjSpElQKpWYPXs2vvvuOxQXF+PVV1/F5MmTw3lpjeI4DinFxag6M0+7/TXXoMDPvHtdUZFU6YZMhoxOnXCqwVqnyuRkXPj00/j98cfhNJthNxigDjCCY2Oq9+zxua/t8OFQ6vXY/8UXzX4fQgiJZWIFLN4r3i217AUA/ZlhkcbSUo8KtVyjQd/HHgMAnHSJWg4Ahf37o3TdOphPnYLgdELWIDqv6cQJCA6Hx9QrX2RqNYQGDeeBPAsQQnxLzs+XKt3pHTtGOzkJK7VtW8iUSthqa+H0sxpFqDDGIjrnPaBK98qVKwEArVq1cvs9Fmk0Grz00kt46aWXfB7jq9GgqKgIi12WrWrMnXfe6bZueSxIbd0aDPWBB+QajRShzxddYSEqd+6Ufs/o2tWj0g3U94prc3JgPnUKphMnwl7plms0yOzWrdnvQQghsU4cPRTvle6WXPbqCwvByWRwmkyw19ZCcSaKuUh+ZqmzhgFPtfn5kKvV4G02WCoqpOHnIiYIMJ04Af2Z4fn+pLVv71amA4E9CxBCfEvOy0Plzp0J2fMaS+QqFdI6dED1nj0wNFj+OFyaM3o3GAFVugcOHNjo7yQ2ZLRpI/3feWZ9vsY0fADI6NTJ57HJeXkwnzoFc3l5o8cFyuqyDp+usNBtGJ3gdEKl10NfXAzjsWPNfi9CCIlViVLpbsnkKlV9w/SJEzAeP+6zNyw5Px8cx/0zuuHMNsORIzCVl3tUuoH6IeaBVrq99oi7xE8hhDSdazA1El4ZXbqges8etzpCuMVcpZvEB43LXJNA5oI1DKam1OmQ0ro1DF7mW2vz84E//wzLTSejSxe39IrzvTO6dKFKNyEkoUW6pZ2ER0qrVjCdOIG6RirdcpUK2txctx4zqdLtoxdt/xdfQKXX+xz9VXv4MI4sXQqn1eq13E/kSMCERIL2TKXbfOIEmCDgwOLFkGs0aHfVVdj78cewVFZCrlaj0w03QBPF2BKJwFdQZ+Z0hu09Izm8PKGil7d0crkcqWedBQDI69vX7/GqlBQk5+VBplRCpdMBALLPPRcAPG4c4Wzpyz//fLffc/v0qU/LOeeA4zi6iRFCElak55SR8Eg9M/2urqys0eMaLtcZSNm6yyUYXUMHFi/G6UOHfFaus84+u9H0EEIal5SVBblKBd5ux+mDB1G+YQOOr16Nmv37cWrbNhiPHcPpgwfx1/vvRzupcS8pMxPanBy3bRzHQZ2WFpb3i/QycAH1dDdnTW6O49yil5LwUSgUOOuWW8BMJui9DFNriOM49LzvPvA2GxRJSQDqA7ukd+woLZMgSs7LAwCYTp4EEwRwIXxIVKWm4vz/+z9wcjmcFov0B5ecn48+jz5K0SIJIQmLeroTQ9qZ6V3+1tbWFxZKq3gALmVrg0q3KiUFdpd1z5tS7hYNGIDiIUPgtFqhbRCMjhDSNDK5HGkdO6Jq1y5UuyzFV9Ng+UN/DW4kMBldusB8ZlnFtA4d0Pmmm6Q6SjjE3PDyhQsXet0uJrRhS4Hrdqp0R45CoYBCpQKnVgd+TlKS25eZk8mkhwBXSVlZkCkUEBwOWKurkZSV1ay0yhQKCC7DRcRWrIYV7KTMzGa9DyGExDLq6U4MYkwV2+nTcDQSU6VhLBWxvLVWV4O32aTtSVlZbpVuX+VuSps2HkPT1enp9cuUUYM1ISGR0aVLfaXbJQhwYwGBSfAyunTB8dWrAdTXUcLVyw1EftWQgCrdCxYs8Ni2adMmzJs3D3l5eRg7dizatm0LADhy5AgWL16MsrIyTJo0CX3ODBUm4SeTyaBUKmFzKbhDhZPJoM3NRV1pKUzl5c2udKeddRbdsAghLR71dCcGjV4PTUYGbNXVjc6j9hZLRaXXw240wnTypLS9Ya+28dgxr+Uu5yUyuRDG+Y+EtEQZnTsDAMwVFdI21/+LGmtwI4FJbdtWGs4fCTG3ZNi4cePcft+1axf+/e9/Y9KkSZg9ezbUDXpWX3jhBTz44IN4//33cdddd4UutcQvtVotraceasn5+agrLUXp2rWo2L4dHUaNgrLB0ii+WKqqcPiHH6TfxQiFhBDS0lFPd/xTKBRILiyErbq60bmdiqQkaDIyYK2ulrYl5+fDbjTi6E8/4fShQ9J2Ti4H43kA9QHRcs7EXBEd+eknlP32m8d7GA4fbu7lEEJcaNLTkZyf7zeuEQUubD6ZUon0jh1RuWtXWN8n0vO5gSADqT355JPIz8/HnDlzPCrcAKBSqfD6668jLy8PTz75ZHPTSJpAqVSG7YskBnw5ffAgTm3bhn2ffRbwuUd/+gkV27dLv6d36gS5SgVlcjL18hBCWizq6U4MMpkM+gZDx31Ja98eQP28beCfIebVe/dCcDik49pdeaX0/4YP87zdjqPLlnl9/axzzgk84YSQgIi93Y2hSndoiPewcE6RicZynUEtGbZmzRpceumljbbOy2Qy9OvXD8uXLw86caTpFIrwrQLXcA3Rqia0QrEGa4UqtVr0euABcHJ5SIOyEUIIIdGQ3qYNjqB+/e3GtL/uOuT164eUM9PytD4CnxYOGAAmCPj7f/9D3fHjbg00DcvUrLPPRpvLL4fDZEJqu3bNvBJCSEMZXbrg2MqV3vd16oTqfftQd/y4xxQS0nQ5vXpBnZoK/ZlVIcIl0pXuoGo7RqMRNQEsXF5TU4O6urpg3oIESaFQgOM4CA0K5FDwFmAtUFov52pzc5s9N5wQQuJZpAt9Ej4ZAVZ2FRoNUtu1kz73hg3awJlo5RyHgosuAsdxcJhMsNfW+nzN5Lw8JOfnI61DB2rIJiQMUtq2hdxldK8YT0GuVqNwwAAA1NMdKhzHIa1DB8hVqrC9h3DmHhvzle4OHTpg1apV2L9/v89j9u3bh5UrV6L9mWFUJDLkcjnkcnlYKt2q1FQoNBq3bbzLULjG0CMlIYR4R5XuxKBNTw/qPG8N2oYjRwAAcpVKWsKzsQd6IcCymBASHJlcjoxOnaTf01wa2cTebfOpU/S3GEfiotI9YcIE2Gw2DBo0CPPnz3cL3GU2m/Huu+9iyJAhcDgcmDBhQsgSS/xTKBSQy+VwhiF6KcdxHi3y5gZLlXhzautWzyE5UQhgQAghsYgq3Ykh2OldcpUKmowMt22uw8eTzzzQ15WW4sTGjTj03Xc49M03bsefPngwqPcmhAQuo0sXr/9XpaRI84930TLJcUHs6Y6koCrd9913H6655hqcOHECd999N/R6PXJzc5Gbmwu9Xo+77roLZWVluOqqq3D//feHOs2kETKZDAqFIiw93YDnMLi648f9nrPno4/gtFrdtsnCOGSEEELiQTSip5LwUSgUyOjRw61R2duSXt40LFtTWreW/i8GaKsrLcW+zz7D8VWrcGLjRrfjXSsAhJDwyOjSBZxcDoVGg4xu3cBxHBRJSQAAXYCBFElsiflAanK5HF999RXmzZuH1157DYcOHUKFy3p17dq1w3/+8x/cc8891IIfBeFcNqzh3Oymzl/RFRaiw3XXhXWeBiGExAMxMBYtGZYY5HI5OowZg5ru3aFNTYVcrYasCZXuqr/+AgDkn38+Wl9+ubRPHLrqq5G70w03ILtnz+YlnhDilyolBedMmgSZXA5tdjZ63H23tHSurrCQlsKNI4IgQCaTxX6lG6hvGbjnnntwzz33oKysDMfPFAaFhYUopMh9UaVSqSLW020MoKfblb6oiCKrEkIIolPok/CRyWRQa7VI794d2jMP4oFyndeddc45UKemSr+LlW6rjwC22T17UkM2IRGSembVAQBIP+ss6f8No5bzdnvE0kSajjEGuVweH5VuVwUFBSgoKAjFS5EQCOeXqGGl21ReXh9lNcCeGrvJFI5kEUJI3GGMUaU7wajVapiCKOfcytYG0w4USUnQZGTAWl3d3OQRQsKk4fBy+nuNbYIgQBXhxsrwLepMosZ12bBQD1tUNmi9FxwO7Pv0UyQXFECuUoHxPAouvtjtIVKuUkktfjV794Y0PYQQEq/E4eVU6U4cKpUqqLn6SdnZ0v9tp0977NcVFdFDPCExrGEwRBLbxJ7ucI0M9oYq3QlIjGDO83xE5gqe3LIF2LLlnw0yGQr79/d6LAV7IYSQetFYJ5SEV7ARzGUu54lRkF3pi4pQ+eefXs8NNFgbISR8Gt7HZUHeC0hkCIIAhUIBewSnAdA3IgGJa3XzPA+lUhny1+/76KOo2bcPdcePo3zDBo/9ZWvXeq10F/bvj+IhQ0KeHkIIiUdioU+V7sQhl8shk8mCGml23tSpMJaUeG2cTuvYEVi6VPq9+JJLoM3NhTI5OeBgbYSQ8OozfTpO798PxhhS27SJdnKIH/II3zup0p2AZDIZVCpV2CKYJ2VlISkrC+UbNnitdJsrKqRhk64KBwxwCw5DCCEtmTi8jSSO5ow00xUUQOcjPo6+qAjK5GQ4zswXLxowAKqUlGanlxASOtrsbGhdpoqQ2BYX63ST2KdUKsHzfFjfo2GkRlfmEyfC+t6EEBLvGGNBD0cmscm10h1KnEyGjM6dQ/qahBDSkkV6uU6qdCeoYIO5NEVygzW7XZWsWIHKnTvD+v6EEBLPBEGgnu4Ew3EcVCpVWBq9KSYKIYSEhrh6SCRRpTtBKRQKMMbCGiCgsSARp7ZuxV8LFqCaopUTQohPVOlOPEqlEhaLJeQV73SXnm6ORkgQQkhQxCmwka50B3TXNhgMUKlU0Gg04U4PCRGVSoWUlBQYjcawrkN37uTJqN69G0wQINdoYKmowImNG6X9ZevWhe29CSEk3kW60Cfhl5ycjJSUFFitViQnJ4fsdZVaLbr+619wmM0ey3cSQggJjBjoMiYDqaWnp+O2227De++957Hvgw8+QIcOHXDhhReGPHEkeCqVCmlpaairqwvr+6S0bo2U1q3dttUePgxLRQUAoHrPHsjCEEGdEELimTj9h+Z0J57k5GRYLBZUVlaG/LWze/YM+WsSQkhLwvO8tNJTJAXUxM4Y8zk/+LbbbsO7774b0kSR0BCXoonkwu+A+7wzxhgEhyOi708IIbFObGmnSndiUiqVYY+rQgghpOliutJN4pNCoYBCoQh7FPOGMhsEe6EHD0IIcScW+lTpTkxyuZzWXyeEkBjE8zzUajUtGUZCR2zFiXSlO7Vdu4i+HyGExBNBEFBXVweFQkFzuhOU+NlGeqQZIYQQ35xOJ6xWa1jjXflCpX0C4zguIut1NyRTKtHllluQ0qZNRN+XEELigcPhQFJSEvR6PfWGJihxvW6n0xntpBBCCDnDarUiLS0N2igEo6RKd4JTq9URr3QDQE6vXuh5770UYZUQQhrgeR4KhQIZGRnRTgoJE3GkGfV0E0JI7OB5Hnq9HklJSRF/b6p0J7hoBnPhZDKkd+oUlfcmhJBYxfM8LcGZ4DiOg0qlop5uQgiJIRzHRTyAmijgSveiRYukllvXH47jfO6jIDHRF+1gLhldu0btvQkhJBYJghCV+WQkslQqFfV0E0JIjIj2Up0BvytFoI5PYjAXm80GtVod8ffPcO3pprmLhBAS1ZZ2EjlKpRJOp1OKVE8IISR6or1qSEDvunLlynCng4SJUqmERqOB0WiESqWKeK+3UqdD2xEjYKmshIbmLxJCWjjGGDiOo1FgLYBSqYROp4PZbIZer492cgghpEWL1vrcooBK/YEDB4Y7HSRM5HI5cnJyYLVaIQhCVL5orYYMifh7EkJILOJ5HjKZjCrdLYBWq0VaWhrKy8ujnRRCCGnxeJ6HSqWK/TndJH6JS5dEI4o5IYSQf0R7eBuJLKVSCY7jaIoeIYREGc/zUZlqKwq61BcEAUePHkVVVRU4jkNGRgbatGlDa47GIPEBj6KoEkJIdImFvkxGbd4tgRhXJVojzQghhNSLdhDTJpf6v/32G6655hqkp6ejQ4cO6NevH/r27YsOHTogPT0do0aNwoYNG8KRVr9Onz6NiRMnIjs7G8nJyRg8eDC2bt0a8Pl79uzBsGHDoNPpkJGRgVtvvRUVFRVux+zduxcPPfQQevbsCb1ej/z8fIwYMQKbN28O9eWEjLh0ic1mo4o3IYREiSAIsFgsCRm5nMpf78SRZjabjSKZE0JIlDidTgiCENVRZk2qdD/yyCMYMGAAvv/+exiNRjDG3H4MBgO+/vprXHjhhXjiiSfClWavBEHAiBEj8Mknn+Dee+/Fiy++iFOnTmHQoEE4cOCA3/OPHz+OAQMG4ODBg3j22WcxdepU/PDDDxg6dCjsdrt03Lvvvov58+ejd+/emD17NqZMmYJ9+/bh/PPPx/Lly8N5ic2iVquhUqlgMpmo4k0IIVFgtVqRnJyMpKSkaCclpKj89U0ul0vDGQ0GQ5RTQwghLZPJZIJer49qozfHApxoNHfuXEyePBkA0KdPH9xyyy0477zzkJWVBUEQUFlZia1bt+Ljjz/Gpk2bwHEc3njjDfz73/8O6wWIvvjiC1x//fVYvHgxRo8eDQCoqKhAx44dMXz4cHzyySeNnj9p0iQsXLgQe/fuRatWrQAAy5cvx9ChQ/H2229j4sSJAIAtW7agU6dO0Ol00rlVVVXo0qULOnbsiHXr1gWcZoPBgNTUVNTW1iIlJaWplywRBAFlZWUoKCjwOWSRMQa73Y7S0lJwHBfVOQ3RIggCampqkJ6eTkM7faA8Cgzlk3+UR54MBgMyMjKQk5MjbQvk/h3rqPxt/PPjeR4mkwnl5eXQ6XQtchoe3Q/8ozwKDOWTf5RHngwGAwoLC91Wkoh0+RvQOxiNRjz66KOQyWSYN28eNmzYgPvuuw8XXnghOnbsiM6dO6N///64//77sWHDBrzxxhsA6nvG6+rqwnoBoiVLliA3NxcjR46UtmVnZ2Ps2LH49ttvYbPZGj3/yy+/xJVXXikV+ABw6aWXomPHjvjiiy+kbeedd55bgQ8AmZmZuPjii7Fnz54QXU3oiRVtpVJJAdUIISQKGGMJObScyt/GyeVyaDQaCmhKCCFRIAhCTCzVGVCl+7PPPoPJZMLkyZNx9913+z1+0qRJmDx5Murq6vD55583O5GB2LZtG3r16uXRUtG3b1+YzWbs37/f57mlpaU4deoUevfu7bGvb9++2LZtm9/3P3HiBLKyspqe8AhTq9VU6BNCSJREu9APByp//RMDqlH5SwghkRUrq4YE9O6rVq2CXC7HQw89FPALP/zww5gzZw5+/fVXTJgwIegEBqq8vBwDBgzw2J6fnw8AKCsrQ48ePXye63psw/Orq6ths9l8Dsleu3Yt1q9fj8cff7zRNNpsNrcWf3F+lyAIzQqwIggCGGMBvYbY0t4SA7q4xh9oidcfCMqjwFA++Ud55M7hcIDjOHAc55YfTbl/+xOtYYRU/gb2+SkUClgsFiiVyqDfL17R/cA/yqPAUD75R3nkzmw2Q61WR738DajSvX37dnTt2hW5ubkBv3lubi66deuGHTt2BHyOSBAEt+ApjREz0WKxeC2UNRoNAMBisfh8DXGfv/O97T916hRuuukmtG3b1m+jxHPPPYeZM2d6bC8vL2/WMHzGGGpqagDA71wxi8UCi8UCk8kkXVtLwRiDyWQC4D+fWirKo8BQPvlHeeROvOdWVla6LR3VlPu3P0VFRc06H6Dyt6ma8vnV1dXBaDTCbDZHvccl0uh+4B/lUWAon/yjPPoHz/NwOBzgeR4nTpxw2xfp8jegu/6pU6fQv3//Jiegffv2WLt2bZPPW7NmDQYPHhzQsXv27EHnzp2RlJTkdd6Y1WoFgEajxYr7mnq+yWTClVdeCaPRiHXr1nnMNWto+vTpmDJlivS7wWBAcXEx8vPzmx3IBajvFfDX0sIYQ0VFBaqrq5v1nvFIjBmYnp7e4m9CvlAeBYbyyT/Ko38wxqBUKtGqVSuPxs6m3L8jgcrfpmnK55ebm4uysjLYbDZotdqg3zMe0f3AP8qjwFA++Ud59A+bzQaZTIZWrVp55EWky9+AKt1ilM+m0uv1MBqNTT6vc+fOWLBgQUDHikPS8vPzpWFqrsRtBQUFfl/D1/kZGRkerex2ux0jR47En3/+iZ9//hndu3f3m1a1Wu21tV4mkzX7w+Y4LuDXSU5ORk1NTUw84EWSGEhBzCviifIoMJRP/lEe/cNut0OpVEKlUnnNi6bcv8ONyt+mC/Tzk8lkSEpKgsViiYnPOpLofuAf5VFgKJ/8ozz6hyAIUiBLbyJZ/gZU6XY4HEElRiaTweFwNPm8vLw83HbbbU06p2fPnli7di0EQXBL64YNG6DVatGxY0ef5xYWFiI7OxubN2/22Ldx40b07NnTbZsgCPjXv/6FFStW4IsvvsDAgQOblNZoEwO6NMwrQgghoScIAuRyuc9CP5ZQ+RteKpVK6oUihBASXjzPx8wyyQlT4xo9ejROnjyJr776StpWWVmJxYsX46qrrnLL8EOHDuHQoUNu548aNQrff/89jh07Jm1bsWIF9u/fjzFjxrgde9999+Hzzz/HvHnz3JZIiRcKhYKWLiGEkAhgjMFqtUKlUiXsMD8qfwPX0uZyE0JItAiCAKfTGTPBKzkWQJOrTCaDTqdr8pIclZWVMJlMEanc8TyP/v37Y9euXZg2bRqysrIwb948lJSUYNOmTejUqZN0bJs2bQAAR44ckbYdO3YM5557LtLS0qTlzl566SUUFRVh06ZN0kPDa6+9hgceeAAXXHABJk2a5JGO6667DsnJyQGlWRy2X1tb2+w5ZU1Z3J0xhpKSEpjNZnAc12LmdguCgJqaGqSnp1MPvw+UR4GhfPKP8qiexWKBIAjIzMxERkaGx/6m3r9jEZW/gX9+VqsVpaWlsNvtUKlUjc53TyR0P/CP8igwlE/+UR7VMxqNUKvVyMnJ8RpHI9Llb8BNrnV1dUFF+IxUy75cLsfSpUsxbdo0zJkzBxaLBX369MHChQvdCnxfiouLsXr1akyZMgWPPPIIVCoVRowYgdmzZ7u10m/fvh0AsH79eqxfv97jdQ4fPhxwoR8tHMchNzcXRqMRlZWV0U4OIYQkLJ7nkZycjPT09GgnJWyo/A2cWq1Gfn4+Tp8+DYPB0GIq3YQQEmmMMaSmpsZM4MqAeroXLVrUrDcZN25cs85PVNFqaRdZrVaUlJQgKSkpLuYaNhe1/PlHeRQYyif/KI/qGQwGZGdnIzMz0+v+ROjpjkfRLn9ra2tRXl5OI82IhPIoMJRP/lEe1Ve4jUYjiouLfTbGxmRPN1WaE5Pr3O6WUOkmhJBIE5cLI8SVQqEAYwyMsYSd608IIdEiBvaMpTgaLbP5gwCAFE2XAqoRQkjo8TwPjuOoUZN4EBu9xXViCSGEhI7NZoNcLo/PSvemTZvw3Xff4cCBA36P3b9/P7777juvS4CQ2MFxHFQqFaxWKwwGAy1jQgghIWQ0GpGUlEQ93cSD+DBoMBhgNpujnRxCCEkYPM/D4XBApVLF1PD6gKr/lZWVGDJkCPR6vRTIpDHp6emYNGkSzGYz/v77b6SlpTUzmSRcMjIyoFarUV1dDZ7nY6pFiBBC4hVjDHK5HBkZGVCpVNFODokxCoUCOTk5qKurQ01NTbSTQwghCcPpdEKtViMvLy+mpu8EVP3/6KOPUFdXh5kzZyI7O9vv8dnZ2Xjqqadw+vRpfPTRR81OJAkfjUaDtLQ0GmZOCCEhxPM8ZDKZW/RtQlzpdDqkpKRAJpPRSDNCCAkRMVZVrI0yC6jSvXTpUiQnJzcpoNqtt94KnU6H77//PujEkcigud2EEBJaYqFP87lJYxQKBWQyGZW/hBASIjzPQ6PRRDsZHgKqdO/atQv9+vVrUouBUqlE3759sXPnzqATRyJHo9FQoU8IISHA8zzsdrsULIsQX1xXESGEENI8TqcTTqczJqd1BVTprq6uRl5eXpNfPDc3F1VVVU0+j0SeSqWC3W6HwWCgaKqEEBIkQRBgMBggk8lisqWdxBZxSRuj0Yi6urpoJ4cQQuKWzWaDxWKBRqOJyRhVAaVIrVbDZDI1+cXNZjPNZ4sTWq0WhYWFqK2thd1up4dFQggJgjisLT8/n8o/EpDMzEyo1WqcPn1aWluWEEJI0zidTiQlJSEvLy/m5nMDAVa68/Ly8Oeffzb5xf/888+geshJ5KnVaqjValgsFlq+hBBCgiQGUEtKSoqpqKkkdiUnJ0OlUsFoNErfH0IIIU3D8zy0Wm1MDi0HAhxefuGFF+LIkSP4/fffA37h3377DYcPH8aFF14YdOJI5CmVSppbRgghQeJ5HiqViircpElobjchhDRfLA4rFwVU6b755pvBGMPEiRNRW1vr9/jTp09j4sSJ4DgON954Y7MTSSJHpVLR0iWEEBIEQRDgcDhoWDlpMo7joFKpqNJNCCFB4HkegiDEf6X70ksvxZAhQ7B7926cd955+O6777xWzBhj+Pbbb9G7d2/s3bsXgwYNwmWXXRbyRJPwUSgUYIxJQV2oAk4IIf45HA4YjUYolcqYnEtGYp9KpYLZbIbRaITFYol2cgghJC6YTCaYzeaYDaAmCjhln332GS666CLs378f1113HdLS0tCrVy/k5OQAAE6dOoWtW7fi9OnTYIyhQ4cO+Pzzz8OWcBIearUaWVlZ4HkeZrMZDocjZudGEEJIrHA6ndBqtcjNzaV7JglKcnIy8vPzYbfbYTQaodFoaJoCIYT4wfM80tLSkJ6eHtPlb8CV7szMTGzcuBH33nsvPv30U9TU1GDFihVSgSD2iMpkMtx0002YO3cu0tLSwpJoEj5KpVJqSDl69CicTmeUU0QIIbFPDOBCKz+QYGm1Wmi1WmnZG57nY7rXhhBCYgFjDFqtNuandjXpbp6SkoIPPvgAM2fOxPfff4/NmzejoqICAJCdnY3zzjsPV155Jdq1axeWxJLIUqvVsNls0U4GIYTEPEEQYr7AJ/FBoVBAJpNBEIRoJ4UQQmIaYwwcx0Eul0c7KX4F1YTatm1b3HfffaFOC4kxKpWKCn1CCPGDMQbGGPVKkpCQy+VQKBQ00owQQhrBGAPP89I9M9bFfgpJ1CgUCmluN8dxSEpKinaSCCEkplgsFin2RTwU+iQ+qNVqGAwGqTEnlucpEkJINBiNRnAcB7VaHRflb0DRyz/44AOfa3QbDAZYrVav+z799FNMmTIl+NSRqFKr1UhLS5PW7nY4HNFOEiGExBSn0wm9Xo+CggJqmCQhk5ycLJW/FMmcEEK8y8jIQH5+flwMLw+o0n3bbbfh3Xff9bovPT0d99xzj9d9y5Ytw+uvvx586khUqdVqFBYWori4WKp4E0II+YcgCEhOTkZSUhJFmiYho9frUVxcjJycHJrfTQghDQiCAI7joNPp4iaeSkCV7saIc9lI4pLJZFCpVFTpJoSQBuIlgAuJTwqFAnK5nMpfQghxIQgC5HJ5XJW/za50k5aBeroJIcSd2NIeD3PJSHyiSjchhHhyOp1xE0BNFD8pJVGlUqngcDhgtVqlnm9CCGmp6urqIAgCBVAjYcVxHFQqFU6fPi2NqqDvGyGkJautrZWGlstk8dN/THduEhCVSoWUlBQIggCr1Rp3QzoIISSUeJ5HRkYGUlJSoFQqo50cksCSkpJgt9sBACaTCampqVFOESGERIe4RFhWVhZ0Ol20k9Mk8dM8QKJKq9WiVatWKCoqoqHmhBACQKfTUcRyEnYZGRlo06YNsrOzwXEcxdEhhLRYYqU7JSUl7kbdUqWbBEycuyiu300IIS2RWOjTaB8SKRzHQalU0vxuQkiLFs/lb8DDyw8ePIgPPvigSfsOHjwYfMpITBLnl9lstmgnhRBCokIs9GluLYkkhUJBy4cRQlo0nufjdonOgJ8YfvvtN/z2228e2zmO87mPMRaXmUIaJwZVczgckMlkcdnaRAghTWU2m+FwOMAYQ3JyMt37SESJDT0Wi0UKqhZPQYQIISRYBoNBWqY63oaViwKqdLdq1Yoqz0SiUqmg0WjgcDhgt9uRkpJCBT8hJOE5nU6kpaVBr9dDLpdTuUgiLikpCU6nEw6HAxaLBSkpKdFOEiGEhJW4PGdWVhY0Gg3UanW0kxSUgCrdR44cCXMySDwRgwfZ7XaUlZWB53mqdBNCWgSdThd3EVNJ4sjKykJ6ejrq6upw4sSJaCeHEELCThCEuA2e5oompJEmEwOqiUPLeZ6nJXMIIQlNbGmnedwkmmQyGWQyGTQajTS/mxq9CSGJTOzci/fyl+7UJGgymQwqlYoiqRJCEp7Y0h7vhT5JDAqFgiKZE0JaBLFzL94bGOnpgTSLUqmE0+mUoqnG+x8EIYS4slgssNvtEASBgqeRmCFWuh0Oh/SdpPKXEJJIjEYjBEGAIAgJEb+CKt2kWVQqFeRyOUwmk7RNr9dHMUWEEBI6DocDKSkpSElJoWjRJGZwHAe1Wg2j0QiTySRF9NVoNNFOGiGENJsYqTwzMzOug6e5oko3aZaUlBSpkDcajaiqqopyigghJHQYY9Dr9dSYSGJOTk4OMjIyAACVlZUwm81RThEhhISGGK8iNTU1ISrcAM3pJs0kBnTRaDRITk4GUP+QSggh8Y4xJq2HTEisUSgUbuUvze8mhCQKnuchl8sTqvylSjcJGfGPgwp+QkgiEFvaKXgaiXW0bjwhJJHwPC/FrkgUCVXpPn36NCZOnIjs7GwkJydj8ODB2Lp1a8Dn79mzB8OGDYNOp0NGRgZuvfVWVFRUNHrOxx9/DI7jaN1WUDRVQkj843ketbW1MBgMMBgMUCgUVOkOAJW/0SV+R2mkGSEkXlmtVqn8tVqtUKvVCdWYmDBPEoIgYMSIEdixYwemTZuGrKwszJs3D4MGDcKWLVtw1llnNXr+8ePHMWDAAKSmpuLZZ59FXV0dXn75ZezcuRMbN270uhh7XV0dHnroIWlYdUsnl8uhVCpRV1cHq9UKxhi0Wm1cL2RPCGlZHA4HkpKSkJ6eDgAJN7wtHKj8jT6FQgGlUgmDwSBt0+v1FPiPEBI37HY7UlJSoNfrwRhLmLncooSpdC9ZsgS///47Fi9ejNGjRwMAxo4di44dO2LGjBn45JNPGj3/2WefhclkwpYtW9CqVSsAQN++fTF06FAsXLgQEydO9Djn6aefhl6vx+DBg/HNN9+E/JriUU5OjvSwWl1dDZvNRpVuQkjc4HkearUaqamp0U5K3KDyN/qUSiXy8/MhCAKcTieqqqrgdDqp/CWExJXk5OSEWB7Mm4RpAl2yZAlyc3MxcuRIaVt2djbGjh2Lb7/9FjabrdHzv/zyS1x55ZVSgQ8Al156KTp27IgvvvjC4/gDBw7g1VdfxSuvvEJDD12o1WrodDrodDpotVoaak4IiStipZsEjsrf6OM4DlqtFjqdDqmpqZDL5RAEIdrJIoSQJknke3rCVLq3bduGXr16eQyl6tu3L8xmM/bv3+/z3NLSUpw6dQq9e/f22Ne3b19s27bNY/t//vMfDB48GFdccUXzE5+glEolzS8jhMQVxhiUSmW0kxFXqPyNLRzHQaVSwel0RjsphBASEEEQwHFcQle6E+bKysvLMWDAAI/t+fn5AICysjL06NHD57muxzY8XxwmLfZ+/PDDD1i2bBl27NjRpDTabDa3Fn9x7pUgCM1qkRYEAYyxmGvVFh/AYiVdjDHpJ1bSFGsojwJD+eRfPOURYwx1dXXS7xzHRSzNobx/R2v+LpW/sfcdVygU4Hk+ZtIVT/eDaKE8Cgzlk3/xlEcOhwMWiwWCIECj0UAmkyVs+RuTlW5BEGC32wM6VoxsZ7FYvA4J1Gg0AACLxeLzNcR9/s5Xq9Ww2+144IEHcPfdd6Nr164BpVH03HPPYebMmR7by8vL3R74mooxhpqaGgCIqSh/DocDJpMJBoMBHMeBMQa5XB61oZuMMZhMJgCxlU+xhPIoMJRP/sVTHjkcDgCAVquFTCbD6dOn3QJShVMo799FRUXNTg+Vv00Tq+Wv2WyGwWCA0WiU1psXg61FQzzdD6KF8igwlE/+xVMeiVHKNRoNGGM4efJkxNIc6fI3Jivda9asweDBgwM6ds+ePejcuTOSkpK8zhuzWq0AgKSkJJ+vIe4L5PxXX30VlZWVXgtvf6ZPn44pU6ZIvxsMBhQXFyM/P79ZQQPEFpr8/PyYilTKGEN6errU2maxWFBbWxu1AAniUPf09PSYvwlFC+VRYCif/IunPBIrdaGotDZVrN2/qfxtmlj7/EQOhwPZ2dkA/nkAr6urg16vj0p64ul+EC2UR4GhfPIvnvLIYDAgNzdXCsIcSZG+f8dkpbtz585YsGBBQMeKQ9Ly8/OlYWquxG0FBQV+X8PX+RkZGVCr1aitrcXTTz+NSZMmSWu4AvVLlzDGcOTIEWi1WuTk5Hh9H7Va7bU1XyaTNfvD5jguJK8TamlpadL/LRaL1KMQjXSK80XEvCKeKI8CQ/nkXzzlkSAIUKvVUUtnLN2/qfxtulj6/EQNr1cmk8FkMkUtjfF0P4gWyqPAUD75F095JMagaAnlb0xWuvPy8nDbbbc16ZyePXti7dq1EATBLeM2bNgArVaLjh07+jy3sLAQ2dnZ2Lx5s8e+jRs3omfPngCAmpoa1NXV4cUXX8SLL77ocWzbtm1xzTXX0PIlPigUCsjlcvA8H/M3AUJIy8EYo6WVzqDyNzEpFAoKbEoIiSmMMchkMsjl8mgnJSJistIdjNGjR2PJkiX46quvpHVCKysrsXjxYlx11VVuLb6HDh0CALRv317aNmrUKCxatAjHjh1DcXExAGDFihXYv38/HnjgAQD1a1B//fXXHu89Z84crF+/Hp9++qnXYDCknmulm6IDE0KixWq1SvOWxekviRwxNdyo/I19CoVCClBEjd6EkGgxmUzScsLiEp0tpfzlWII0ffI8j/79+2PXrl2YNm0asrKyMG/ePJSUlGDTpk3o1KmTdGybNm0AAEeOHJG2HTt2DOeeey7S0tIwefJk1NXV4aWXXkJRURE2bdrUaPCv2267DUuWLGlyMBaDwYDU1NRmz3MWBAFlZWUoKCiI+cK0tLQUVVVVUKlU0tAXMXhRuAmCgJqaGqSnp8d8PkUL5VFgKJ/8i+U8qq2tRWpqqttcY51OF5XGwHi6f/tC5W/sf34OhwMlJSWwWq2Qy+XgOA5yuRxarTYi7x/L94NYQXkUGMon/2I1jxhjMBqNSEtLk+7rHMchNTU1KnPPI33/TpimBblcjqVLl2LatGmYM2cOLBYL+vTpg4ULF7oV+L4UFxdj9erVmDJlCh555BGoVCqMGDECs2fPjlq07USUmpoKtVotheivq6uDw+GgPCaERAzHcdDpdFEL6phoqPyNfQqFAunp6dISaQ6HA0ajkXq+CSERw/M85HI5MjIyWuSUroTp6Y5HLa2l3Ztjx47BZrNFpLU9Vlv+YgnlUWAon/yL1TwSW9pbtWoVsV6+xsTz/TuetfTy1263o6SkBCqVKiJDO2P1fhBLKI8CQ/nkX6zmkc1mA2MMrVu3jol53JG+f8fOJ0FaJLVaLc3tIISQcBNb2lvKHDJCvBHneFP5SwiJFLH8jYUKdzTQUweJKqVSSRFVCSFhxfM8TCYTgPq5rcnJyS220CcEqF9CTKlUel0fnRBCQsVms0n3GZvNhry8vCinKHqo0k2iSqFQgOd5mM1maVukhrsRQloGu90OjUYjDSNuyS3thIjUajVqamrctmk0mpgajkoIiW82mw16vV4KXKrRaKKcouihmg2JKrVajfT0dCmwmiAIMJvNFOCIEBIyPM9Dq9UiIyMj2kkhJGZotVpkZmYCqJ/baLfbYbVaYyLWASEkceh0OqSmpkY7GVFHlW4SVSqVCgUFBdLvBoPBrdebEEKaSxCEFhkplZDG6HQ66HQ66fcTJ07AYDBEMUWEkETCGANjjEavnkFjiEhMUSgU4DiO5nkTQkKKCn1CGqdSqSiwGiEkZARBoMClLigXSEwR51ryPE9/pISQoFitVthsNnAcJ22j+wkhjRMbvQkhJFh1dXXgeR4cx8HpdCIpKYliqJxBTyEkpigUCsjlcpjNZqjVamk7tZQRQgJls9mQmpoKjUYDQRDAcZzb/YQQ4kksY61Wq1vlW6lUUnA1Qohf4kiZjIwMKBQKMMYocKkLqsWQmCKXy5GUlCQV+OIwc5PJREEYCCEB4TgOqampSE5OjnZSCIkbSqUSycnJcDqdAP6Zj2k2m93mfhNCiDfiOtyZmZnUUeYF5QiJOeIafq4V7tLSUjDGaOgbIaRRPM9DJpNRyzohTaRUKlFUVATgn/K3uroalZWV0UwWISROiHO4qfz1jirdccjhcLgFOxEEAU6nE1arNSGHgImFv81ma1bLmSAIYIzB6XQmZD6FAuVRYCif6slkspi7frGlnVrZCWk6sWFb/FelUlFgU0JIQJxOJ3Q6HXWQ+UBPJXHEYDCgsrISNpvNbTtjDDzPw2w2J+QXXazcOByOZl2fOFTOaDQmZD6FAuVRYCif6jHGoFKpolrIms1maTgsUN8oqdfrqaWdkBBwXVGkJd/rCCHuGGOoq6tza5Sz2+3IyMiIYqpiG1W644TBYEBpaSl0Oh2ysrKgVCrd5j07HA63bYnGbreHpNAXe8GIb5RHgaF8qq/wig2BGo0m4u8v3vtSUlKkQGnimtyJei8kJJLEoaINR5pxHNfi73+EtGQOhwMKhQJ6vR4ymUzqjKBYKr5RpTtOVFZWQqfToaioyONhUowOmMiVbnEZMW+aMrzV6XTSsFM/KI8CQ/kEaDQa2O12GAwGqNXqiN9/xPljWVlZFJ2ckDBQKBRQqVRwOBxuo+wEQYBSqYxKYxshJPrEjofs7OyErXuEWst+YowTYmGXlZXVYr/Y4lJirniedxtWSgiJPJ1Oh9ra2qgMPxULfepxIyQ85HI5CgsLpV4soL6hv6KiAmazOcqpI4REC8/zbqsNEf+o0h0HxB5epVIZ5ZRED8dxXnv4CSHRJc75FAQh4kHVqNJNSPh5G9Gj1WpRV1cXhdQQQmKBIAg0wqyJqNIdR6g1yZ1YEacAL4S0DHV1dRAEAUD937/VaqWhbYREQUufWkNIS8LzPEwmk/Q7x3HgeZ7uA01EuUXilrfeb0JIYhJH/KSmprr1bGu12mgliZAWS3zYpkZvQhKf3W6HRqOBTqeTGr4BUEyHJqJKN4lbYqVbEAS3m4C4L9bWDyaEBM81aAu1rhMSXWKANaPRKG1zjV7ckqfDEZJoeJ6HVqtFVlZWtJMS1+jJhcQ1hULhMbeb53mfkc5DSa1WY8CAAfjll1/C/l6EtHTiUDaav01I9KlUKhQUFEgN3mI5XFlZKS1hSghJDOJSnKR5qCuQxLwjR45Ivdrij1KpRGFhIW644QZs27ZNCqYUb0GVPvjgA6jV6kZ/7rjjDq/nbt26FXfddRe6du2K9PR0pKamonPnzhg/fjyWL1/uduysWbPcXjMpKQk5OTno1q0bbrzxRixatMhtvo6rO+64o9H0PfjggyHPF0Ia4nme1t8mJIZoNBpotVpotVokJycjOTkZGo0mIo3ehJDIohFmzUc5SOJG+/btccsttwAATCYTtmzZgsWLF+Obb77B8uXLMWDAgCinMHiDBw/GRRdd5HXf2Wef7fa7IAh4+OGHMWfOHCgUCgwaNAhXXnkllEolDh8+jB9//BGffPIJZsyYgUcffdTt3Ouuuw7dunUDABgMBhw9ehRr1qzBV199haeeegrvv/++z3SMHz8ehYWFHtv79u0bzCUT4pPZbIbT6XSrYNtsNmRmZkYxVYQQf1Qqlcd0L0JIfBAEASaTyWusBqp0Nx/lIIkbHTp0wJNPPum27fnnn8f06dPxxBNPYPXq1QDiM6r5kCFDMG3atICOnTFjBubMmYNzzjkHn376Kdq3b++232KxYN68eaiurvY4d+TIkRg7dqzbNpvNhrlz5+KJJ57AddddhxUrVuDcc8/1OHf8+PHo169fE66KkKZjjMHpdEKv10tDVMWhqxQ0jZDYJpfL46bcJYS4czgcUCgU0Ol00t+x+CxNy4M1Hw0vJ3FtwoQJAIAtW7ZI28RK94IFCzBy5EicddZZSElJQV5eHq6++mqsWrXK43VWr14NtVqNWbNmYcuWLRg+fDgyMzORnZ2NMWPG4MiRIwGlhzGGqVOnQq1WY9y4cXA4HKG4TMnBgwcxe/ZsZGZm4n//+59HhRsAkpKS8OCDD+L//u//AnpNtVqNqVOn4rHHHoPJZAr4PELCwTVgWk5ODnJycpCbm4vc3FwkJSVFO3mEkEYoFArIZDLU1tbCYDBI/xqNRlit1mgnjxDSCDF2Sm5urlv5m5OTQz3dIUCVbpIQXG8GHMdBoVDggQceQEVFBYYMGYLJkyfjiiuuwMaNGzF8+HB89913Xl9n8+bNGDJkCFQqFe644w706tUL3333HYYPH+73gcHhcOC2227D3Llzcd9992HhwoUhDybz4Ycfgud53HHHHcjNzW302Ka2Sv7nP/+BVqvF8uXLUVtb25xkEhI0sdIdT7EZCCH1NBoN8vPzUVhYiPz8fBQUFCA/Px/p6ekA4BH4lBASOwRBoB7tMKJmizjHGIPTZoPT4QAnCDE5rEsexuBH7777LgCgf//+7u8pl2P37t1o27attI0xhr///hsXX3wxpk+fjquvvtrj9X766Sd89NFHGDNmjLTt9ttvx8cff4zvvvvOY2i2qK6uDtdffz2WL1+Op59+OuCh4qIVK1b4rNSPGTMGnTt3BgCsX78eADBo0KAmvX4gdDodevXqhXXr1mHr1q0YPHiw2/4FCxZg2bJlbts0Gk2Tr5WQxogB06jSTUj8kclk0Ov1Hts1Gg1KS0shCAL9bRMSo3iep0p3GFGlO87xdju+HD9e+j32qtzAqIULoQjBH/HBgwelOd1iILWVK1ciNzcXL730ksfxrhVuUUFBAa699lq89dZbOHr0KFq3bu22/+KLL3arcAPAuHHj8PHHH2Pz5s1eK92VlZW45pprsG3bNrzzzjsYN25ck69t5cqVWLlypdd9Z599tlTpPnHiBAB4DWgWCvn5+QCAqqoqj30LFizw2JaamkqVbhI0i8UCh8Ph1ihntVq9PrQTQuKXONfb6XTScmKERBljTAqYJuI4jhrFwowq3SRuHDp0CDNnznTblpeXh7Vr16JDhw4ex//999947rnn8Ouvv6K0tBQ2m81tf3l5uUel21sAMbGC623I9cmTJzFo0CAcP34cX3zxBa688somXxeAoHrHI23NmjUUSI2ElN1uR0pKikfAtOTk5GgmixASYjKZDHK5nCKbExIDeJ4Hx3HQ6/VSJVssfzUaTTSTltCo0h3n5CoVRi1YAIfDAaVSGbPDy0Ph8ssvx08//QQAqKiowKJFi/Dwww/j6quvxsaNG6HT6aRjDx48iL59+8JgMGDw4MG46qqroNfrwRjDmjVrsHbtWlgsFukBQLzZeOthE+eLe1t79MSJEzAYDGjfvn1Els7Ky8vDvn37UFZWhk6dOoX89cvLywEAWVlZIX9tQlwJggCZTIaMjAwKkEZIC6BUKmEymaSy1PV5heM4aLVayGQUaoiQcHMNmEZ/c5FDle44x3EcFGo1mEwGRYxWusMhOzsbU6dORW1tLZ5++mk8/vjjeO2116T9r776KmpqavDhhx9Ka3szxmC323Hfffdh7dq1bsGaxJtOU4O8nHPOObjllltw991347LLLsPPP//sN8BZc1xwwQVYvXo1Vq5c6THnurnq6uqwdetWyOVyrz3+hISSOIyNIqIS0jKo1WpkZWVBJpOBMSaVt+JQV7vdTr1shESAOHebKtyRRblN4tqjjz6KgoICzJs3z21Zr0OHDgEArrnmGo9zxGBkCoUCSqUSSqWyWQ/+48aNw/z587Fv3z5cdtllOHnyZNCv5c+tt94KuVyO9957DxUVFY0e23A4vT+vvfYazGYzLrvsMqSmpjYnmYT45XQ6KUo5IS2IWOnOysryWBJQo9HA6XRGO4mEtAgUMC06qNJN4lpSUhIefvhhOBwOzJo1S9ouztVet26d2/EvvfQSdu3aFfJ03HLLLXj33Xexf/9+DB06VAp4FmodOnTAgw8+iMrKSlx99dU4fPiwxzFWqxWvvfaaW340xmaz4eWXX8YzzzwDnU7nMW+ekGAxxlBXVyet0yv+1NXVwWQySWv6EkJaNpVKRfO9CQkhnudhNBq9lr92ux2qEE39JIGjcX0k7k2cOBEvvPACPvjgAzz66KNo37497r77bixYsACjRo3C2LFjkZmZiT/++ANbt27FiBEj8MMPP3h9reYMz7/55pshk8kwYcIEDB06FMuWLZOigfvT2JJhubm5mDhxovT7zJkzYbVaMWfOHPTo0QODBg1Ct27doFQqceTIEfz666+oqqryWnn+6quvsG/fPgD1w8mPHDmCdevWobKyEsXFxViwYAG6desWxNUT4kmcN6bT6TymcDDGaC43IQRA/XxvWsObkNBxOBxIT093K2ddy1+ayhF5VOkmcU+j0WD69Om47777MHPmTHzwwQc499xzsWzZMjz++OP46quvIJfLceGFF2LVqlVYunSpz0o38E9UR5HrTcqfG2+8ETKZDOPHj5fmeBcUFPg9z9+SYa6VbplMhpdeegk33HAD3nnnHaxduxbr1q2DIAjIy8vD0KFD8a9//QtDhgzxeK2vv/4aX3/9NWQyGXQ6HbKzszFgwAAMHz4co0ePhlarpSF+JGQEQZCCtbSUeBOEkKYTp3iZzWZpm+s9o7nTwAhpaRhj0Gq1yM7OjnZSyBkco6bFqDEYDEhNTUVtbS1SUlJ8Hme1WnH48GG0bdvWa8sUYyymo5fHCn/5xBiD0+n0WcluKesXOp1OergJAOVTPavVipKSEqSkpLjlhyAIKC8vR0FBQdjWlU8EgiCgrKwMBQUFNNQ+ggItf/2hzy8w/vLJ4XCgoqJCimzuOtScMQar1dqszykeCIKAmpoapKen03epEZRP/gmCgNLSUnTq1AlpaWnRTk7MivT9m54YCTmD4zhpveCGBEGA3W4HY4waNggJEGOM5o0RQvxSKpU+R4UZDAaUlZVFOEWExC/xWbUldBTFE6p0ExIAqmgT4psgCKirq4NMJnP7WxEEwWdDFiGEBEIcQUON3oS4s1gscDgcANyfU8VpkjQaL7Yk1LiM06dPY+LEicjOzkZycjIGDx6MrVu3Bnz+nj17MGzYMOh0OmRkZODWW2/1uSzToUOHcNNNNyEnJwdJSUk466yz8Nhjj4XqUkiM4TgOHMdRoBdCGhAEARzHIT09HRkZGUhPT0d6ejpSUlKg1+tpWZIWgspfEi4KhQJyuVwaek4IqWe326HX65GZmSmVvWlpaUhPT0dycjI1eseYhGkCEQQBI0aMwI4dOzBt2jRkZWVh3rx5GDRoELZs2YKzzjqr0fOPHz+OAQMGIDU1Fc8++yzq6urw8ssvY+fOndi4caPbEMnt27dj0KBBKCwsxIMPPojMzEyUlJTg2LFj4b5MEkUymcxtzrcv1BJPWhqZTIb09HS3mBOCINBaoC0Elb8knORyOeRyOaxWq/RdcC1nxf/TcFrSkgiCAJlMhoyMDI+VQMTyl+a8x5aEqXQvWbIEv//+OxYvXozRo0cDAMaOHYuOHTtixowZ+OSTTxo9/9lnn4XJZMKWLVvQqlUrAEDfvn0xdOhQLFy4UIoeLQgCbr31VnTu3BkrV66kJW9akEB7uxljdKMjhLQYVP6ScJLL5dBoNLBYLB693a7lsdPpRFJSEvXukRaB53nI5XIaQh5HEuaTWrJkCXJzczFy5EhpW3Z2NsaOHYuPPvoINput0R6XL7/8EldeeaVU4APApZdeio4dO+KLL76QCv1ly5Zh165dWLp0KZKSkmA2m6FWq6l1tQUQW9sbci30eZ6nJbdIiyI2RpGWi8pfEm75+flSRHPXMtd1tZGysjI4nU6qdJMWQax00/0vfiRMd9y2bdvQq1cvjx7Gvn37wmw2Y//+/T7PLS0txalTp9C7d2+PfX379sW2bduk35cvXw4AUKvV6N27N5KTk6HVanHDDTeguro6RFdDYpFYuWj4I5PJpB+qfJBEwhgDz/MeP4IgSMPXCKHyl4SbOHRc7NkTf5RKJZRKJVQqFdRqNd2TSMKwWq0wGAweP0ajEUajEXV1dVCpVDSyMo4kTE93eXk5BgwY4LE9Pz8fAFBWVoYePXr4PNf12IbnV1dXSy31Bw4cAFA/dG7YsGGYPn06duzYgeeeew7Hjh3DunXrfFa8bDYbbDab9LvBYAAA6QHWF0EQwBiTfhpDgb4CE658oko3SSTiVAlfhbrrVIqG9zHxvtXYvY2ENp+i9fCVyOWvP/Q9D0wk8kmhUEiNgvHI9TkvXq8hElpKPtlsNuj1ereYFg3Xr9doNF7zgO5LgYl0+RuTlW5xTeRAqNVqcBwHi8XidfiaGNjHYrH4fA1xn7/z1Wo16urqAAB9+vTBRx99BAAYNWoUtFotpk+fjhUrVuDSSy/1+j7PPfccZs6c6bG9vLxcel1vnE4neJ6Hw+HwOYyEhjQHJtz5JP7xxnPjhyAI9H0KQEvIJ7FS3VhjktgbfurUKbe5ZYwx1NTUAKDGqMaEMp+KioqanR4qf5uGvueBiUQ+WSwWGAwGaQklUbx8LowxmEwmAPGT5mhoKflksVj8ztc2mUxSXrii+1JgIl3+xmSle82aNRg8eHBAx+7ZswedO3dGUlKSWyu2yGq1AkCjAVfEfYGcL/574403uh130003Yfr06fj99999FvrTp0/HlClTpN8NBgOKi4uRn5+PlJQUn+mzWq0wm83SMCpfaB5TYMKZT2JlO9BWs1ici+N0OikwRwBaQj7xPA+lUtno91ScV5aTk+MRvRyo762k4W++xVo+UfnbNLH2+cWqSOST2WwGx3FujaEN53+LwVBVKpXb/SoWiGlNT0+nilIjWkI+McagVCpRUFAQVMBIui8FJtL5FJNPjJ07d8aCBQsCOlYckpafny8NU3MlbisoKPD7Gr7Oz8jIkFrhxdfJzc11Oy4nJwcApBYTb9RqtdfW/MaGb4r7XecQN+RaqCTqDSgUIpFPHMd5VOq99Xq3hF5Skhj8BUprGNug4T5/9zcSW/lE5W/TxdLnF8vCnU/JyckoKipyC64mcv1/bW0tDAZDzH1egiC43U+Jdy0hn8RG/ebM2ab7UmAimU8xWenOy8vDbbfd1qRzevbsibVr10rr1ok2bNgArVaLjh07+jy3sLAQ2dnZ2Lx5s8e+jRs3omfPntLv5513HubPn4/S0lK348rKygDUR2wlLVsgf7gcx1HAFxI1gU6BoMjkLQ+VvyRecRwXUO+10+mU5vQTEml1dXXgeV4qW8URGK4cDgd0Ol1MjoYkwUuY5o/Ro0fj5MmT+Oqrr6RtlZWVWLx4Ma666iq3Fu5Dhw7h0KFDbuePGjUK33//PY4dOyZtW7FiBfbv348xY8ZI26655hqo1WosWLDAbQjxu+++CwAYOnRoyK+NJB7Xm20scDgcmDVrFrp27Yr09HSo1Wp8++23OHLkCNRqNe64446AXqepx5PIEj+fO++8E3K5XJqyIv64RgVWKBSQy+VU6SZ+UflL4olCoQgoMC0hoSY2eGdkZCA3Nxc5OTnIzc1FdnY2srKykJWVhczMTOTm5iItLY16qRNMTPZ0B2P06NE4//zzMX78eOzevRtZWVmYN28eeJ73CJ4yZMgQAPUPoKJHH30UixcvxuDBgzF58mTU1dXhpZdeQo8ePTB+/HjpuLy8PDz22GP4v//7PwwbNgzXXnstduzYgfnz5+PGG29Enz59InK9DYm9prH2gCyTyULWUicIAubNm4f3338fe/fuhUKhQM+ePTF16lRcffXVHse3adMGR48e9fpaAwcOxKpVq9y2zZ8/H7Nnz0ZpaSm6d++Ol19+GRdddJHHucuWLcPw4cOxZs0ar/sDsWXLFrzxxhtYt24dysvLIQgC8vPzccEFF+Dmm2/2OS8xXF577TU8/fTTuPjiizFy5Eio1Wp06tQpommIN0ePHkXXrl09tmu1WrRr1w7XXXcd/vOf/0Cn00Uhdf7R+p4kVFp6+Uvii9igKAgC3QNJRPE8D4VCgYyMDIrD1AIlTKVbLpdj6dKlmDZtGubMmQOLxYI+ffpg4cKFAVUeiouLsXr1akyZMgWPPPIIVCoVRowYgdmzZ3vMA3v88ceRnp6OuXPn4j//+Y/bg0A08DyPsrIyaZ5LLFEqlSgqKmp2wcYYw9ixY/Hll1+iffv2mDBhAmw2G7799ltcc801mDt3Lu69916P81JTU/Gf//xHeg1x+GPbtm3djluyZAkmTpyIiy66CCNGjMBXX32Fyy+/HHv27EFxcbF0nNlsxt1334277rorqAq3IAiYOnUqXn31VSgUCgwaNAgjRoyAUqnE4cOH8eOPP+KTTz7BE088genTp7udG851wJcuXQqdToelS5dCJpNJAcIcDgd27NiB1NTUsLxvImjXrh1uuukmAPXfsYqKCvz888+YNWsWli1bhpUrV8bkg12s3StI/GrJ5S+JP+Ja3+K87obDfF2H+4r/T05Ojsn7OIkvPM+7PWORloVjNL4magwGA1JTU1FbW+s3evnhw4fRtm1br/OV7HY7Dh06BJVKFVMtZ06nE4IgoHXr1s1O15IlSzBmzBhcdNFF+OWXX6RojpWVlejduzdOnDiBvXv3ok2bNtI54v/FHhXGGBwOB5RKpUeF4/LLL8fRo0exe/duyGQyHD16FO3atcMzzzyDRx55RDpuypQp+OKLL7B79+6gIt4++uijeO6559CzZ0988cUXHpV/i8WCefPmoaqqCs8++6y0XRAEKUp0OHTq1AmMMezfv79ZUbmPHDmCTp064dZbb5WGfCaqQ4cOoWvXrhg6dCi+//57t302mw0DBgzA9u3b8dNPPwUcDTrcxM/nlltuwQcffBCSirev+5MgCCgrK0NBQQENkWsE5VN0BFr++kOfX2BiLZ/EebUib0HXxH+NRiMcDge0Wm1Y0yQIAmpqapCenh4TeRSr4jmfTCYTtFotCgsLw/o+sfb3FqsinU/0SSSQhvMxY+EnVL799lsA9ZVW1+UTsrKy8MADD8BmswUccdebY8eOoWfPntIfXevWrZGVlYWSkhLpmE2bNmHOnDmYN29eUA9pBw8exIsvvojMzEz89NNPOOusszzyS6/X4+GHH8asWbM85tZ+8MEH6N+/PzIyMpCRkYH+/fvjgw8+8Hif1atXQ61WY9asWdiyZQuGDx+OzMxMZGdnY8yYMW7DOmfNmgW1Wo0jR47g6NGjUKvVSE5OlgIf+ZqjzfM8Xn75ZXTp0gUpKSno0qULXnzxRb9Lpa1duxbXXXcdCgoKoNfr0bVrV8yYMQNmsznoa2j4+qNHj0ZxcTH0ej3at2+PsWPH4rfffgs6LcFQq9UYOHAgAKCqqkrabrfb8eabb2LEiBFo37499Ho9ioqKMHbsWGzfvt3jdQLNB3GtbPHHbrfjxRdfdPt8nn/+eZ8R87/88ksMHDhQWvaroKAAl156Kb788stm5wUhhMQanU6H1NRU6SctLU36SU9PR3p6ulTWqtXqgJcBJS2PyWSSIuKL//r6v9ls9rqKAmkZaHwDiQsnTpwAAI+eYddtv/76q8f8QZvNhoULF6KsrAx6vR7nnnuu12HhxcXF2LFjhzT8vKSkBJWVlWjVqhWA+l77O+64A9ddd53X+eOBWLhwIXiex1133eWx5E1DDW/K//nPf/DGG2+gsLBQiiz8zTff4M4778T27dvxyiuveLzG5s2bMXv2bAwcOBB33HEHtm/fju+++w67du3Ctm3boNFoMGDAADz++OOYO3cuAOC+++6DIAhIT09vNH2TJk3CwoUL0aZNG9x9992wWq14/fXXsX79ep/nvP3225g8eTLS0tIwYsQIZGdnY+vWrXj++eexevVqC3VuywAALJBJREFULFu2DCqVqsnXIJo7dy6mTZuGpKQkXHPNNSguLkZZWRl+++03fPXVV26fezBpaQq73Y41a9aA4zicffbZ0vbq6mpMnToV/fv3x+WXX4709HQcPnwY33//PX7++WesWLECvXv39ng9f/mgUqkgl8ulRiPx82nbti3+/e9/w2q1Yu7cudi4cSMA96kKb731FiZNmoT8/Hxcd911yMzMxIkTJ7Bx40Z8/fXXGDVqVND5QAgh8U6lUlGlm/jE8zzS0tKg1WobHTEB1PesJicnRzyNJDZQpZvEhaysLADA4cOH0aVLF7d9hw8fBgDs37/f47wTJ064BeIBgD59+uDTTz9F+/btpW133HEHxo4di0GDBqFv37746quvkJSUhJtvvhkA8OKLL+LYsWP4+eefg74Gsbf1kksuadJ5a9aswRtvvIHOnTtj9erVSEtLAwA88cQTuPjii/Hmm29i5MiR6N+/v9t5P/30Ez766CO36L+33347Pv74Y3z33XcYO3YsBg4ciIEDB+LDDz+UXtPf8PLVq1dj4cKFOPvss7Fq1SqpAHn44YfRt29fr+fs2bMHU6ZMQY8ePfDTTz8hMzNT2vfSSy/h8ccfx5tvvokHHnigydcAAH/++Sceeugh5OfnY+XKlW7TDBhjbmsAB5sWXw4dOoRZs2ZJ71VZWYlffvkFZWVlePbZZ92WS0pPT8fBgwc9hpbt3r0bF198MZ544gn8+OOPHu/hLx9GjRolzRNbtWoVFi5ciHPOOQe//fab9Pk8/vjj0vJLrsPK3333XahUKmzfvl1a71jk2ktPCCEtEc2/Jb6IFerU1NSwTz8g8Y+Gl5O4MHz4cADA888/D6vVKm2vqqrCa6+9BgA4ffq02znjx4/HihUrcPLkSZhMJmzduhU333wzNm3ahCFDhsBoNErHjhkzBm+99RZOnjyJ//73v8jNzcXPP/+M4uJi7N+/H7NmzcJLL72EvLw8PP/888jLy4NSqcTgwYNx4MCBgK5B7K0vKipq0rUvWrQIAPDYY49Br9dDEAQIgoDU1FQ8+uijAIAPPvhA2i4WAv3798eoUaPclkYZN24cAHhdEzdQH3/8sZQe1xbbwsJCr8HsgPrI8E6nE6+++qpbJRcAHnzwQWRnZ+OLL77wOO/iiy92q2j6uob58+dDEAQ8+eSTbhVuoL6CWVBQ0Oy0+PL333/j6aefxtNPP41nnnkGb7/9Nv7++29ccsklUqRmkVqt9jqXq2vXrhg4cCDWrVsHh8PhsT+QfBAr0uKUg//7v//z+HwmT57s9RrEJcMaapg/hBDS0oiVbqPRiLq6OtTV1cFkMkk/ZrPZ7cdisUg/VqvV57QeEv/EWDvUMEMCQd8SEhduuukmLFy4ECtXrkSPHj0wbNgwOBwOfPPNN9JQ7YZBEGbMmOH2e8+ePfH+++9DJpPhww8/xPz58zFlyhRp/9133427777b7RzGGCZOnIjzzz8ft99+Oz799FM8+uijeOqpp9CnTx88/PDDuO666/Dnn3+GLQjDtm3bANT3kIvri4rEAF2u7y9Wvnr16iX9X1waRazw1dbWBp2eP//8EwC8DtP3FdFdHNb8yy+/YOXKlR77lUol9u3b57H93HPP9djm7RrEimcgS60FmxZfGgZSq6qqwu+//44HH3wQgwYNws8//+w2AmDHjh2YPXs2fv/9d5w4ccKjkl1ZWYn8/Hy3bf7ywbXneseOHQDqK+oNedt2ww034KGHHkL37t1x0003YfDgwejfv3+zgksRQkiiUKlUyMjIcGvUbrjOtzj83HW7+K/JZKIVQBKUWOmmyPYkEFTpJnFBoVDgxx9/xPPPP49PPvkE77zzDlJTU3Hddddh6tSp6Nixo8fQWF8mTpyIDz/8EL/99ptbpdub+fPnY8OGDdi5cyc4jsPrr7+OSy+9FI8//jiA+mAs/fv3xy+//ILLL7+80dfKy8vD3r17UVpa2qQ1sMVlTfLz8z2iTRcXF4PjOBiNRmkOsthjmZ6eDrVaDcYYbDYbGGNSa6xr1Namqq2thUwmk4b8u/L1GVRXVwOoH6nQFN4qft6uQax4NqyshjItgcrMzMRVV10FrVaLK664AjNmzJCGjK9fv176nlx66aUYOXIkdDodOI7Dt99+i507d8JisUjXJj7I6XQ6j89MbGQRe1HE70Zjn4+3WAJTp05FZmYm3nrrLcyePRsvv/wyFAoFRowYgVdffdVrHAVCCGkpFApFwM8XDZnNZpSUlLgtQ0ZiC8/zqKurAwCfS8b5+r/D4UBqaipVuklAqNJN4oZarcaMGTM8erBXrVoFAF4DUHkjVkZMJlOjx5WXl+Ohhx7CjBkz0KFDBwDAvn37cOedd0rHiD2Qe/fu9Vvpvuiii7Bq1SqsWLGiSfO6U1JSIAgCKioqPAr+U6dOgTHWaK8kx3FSQRIKqampEAQBlZWVyM7O9kiPN2L6KisrodfrQ5IOV2lpadLcbX9LcYQ7LSKxd3vLli3Stueffx42mw2//vqrx6gAsXFHqVRKDShiA4NcLvcY/i3u4zgOCoVCehho7PM5efKkRzo5jsPtt9+O22+/HVVVVVi7di0+/fRTfPHFFzhw4AD+/PNPeqAghJAgiL2gPM/TEOQY5XA4oNFoPBqqvQVF8/b/5gRdJS0LzekmcU+cY3zDDTcEdPyGDRsAwGPub0P33HMP2rRpg6lTp7ptt9lsHv8PpAX7tttug1wuxzvvvIOKiopGj3V9D7FiLzYuuBK3iQGyfAnl0HcxGre3Zbi8bQPqg9cB/+R9qIkNLsuXL/d7bLjTIqqpqQEAt6i3f//9NzIyMjwq3CaTSVoyTHxIcx2yJgZJ87Ykn7hP/A6ec845AOqXRGvI2zZXmZmZuPbaa/H555/jkksuwe7du3Hw4MEgrp4QQohCoYBMJqPo5zGM53moVCro9Xq3n5SUFOmnseXlKBo5CRRVukncMBgMHtuWLFmC999/H3369MHIkSOl7Xv37vW63vLevXvxyCOPAKifJ+7LV199he+++w7vvvuuW+t0ly5dsGzZMmlI79KlS6Xt/nTo0AEPPfQQKisrMXz4cCnquiur1YpXXnkFTz75pLRNDJg1c+ZMtzyora2VlkgTj/FF7OluOB/N24+43xcx35555hm30QKlpaV44403vJ5z9913Q6FQ4IEHHnBb+1x0+vRpr+tUB2rixImQy+V48skncfToUbd9jDGUlZVFLC2i119/HQDcosq3atUKNTU12L17t7SN53k88sgjfhtiAnXrrbcCAJ566imPz0dMk6tVq1Z5fN4Oh0Mahu+6LBshhJDAiUG2LBYL7HY77HY7HA6H9ON0OqUfnuelgKiu/3cNktpwLjlpPrHSTUi40VgXEjf69euH4uJidOnSBRqNBhs3bsSqVavQrl07LF682G0I7GeffYZXXnkFAwYMQOvWrZGcnIz9+/dj6dKlcDgcmD59OgYMGOD1fWpra3Hvvfdi8uTJHkPWJ0+ejBtuuAGDBw9Gz549sXDhQvTo0cMjSrUvTz/9NKxWK1599VV06tQJl1xyCbp37w6lUonDhw9j+fLlqKqqwtNPPy2dM2DAANx3332YO3cuunfvLkUk//LLL3H8+HHcf//9Pq9FxHGcR2+3t4K7sUq42FI/YMAA3Hrrrfjwww/Rq1cvXH311bDZbPjyyy/Rt29fqSHCVbdu3TBnzhzcd999UiC8du3awWg04vDhw1i7di1uvfVWvPnmmwHlY0Pdu3fHyy+/jClTpuDcc8/F1VdfjVatWuHkyZNYu3Ythg8fjtmzZ4ckLQ3zzXXJMKB+zvj69euxbds2pKen49lnn5X2TZo0CcuXL8fgwYMxatQoaDQarFmzBmVlZRg4cCBWr14d1PW7Gjx4MMaPH48FCxagR48euO6662Cz2fD555/j/PPPdwv6BgDXXnstUlJScP7556N169ZwOBz45ZdfsHv3bowePRqtW7dudpoIIaSlSkpKkirVrhoOU2aMweFwwGKxNDolrGEAN61W2yIqjWLjQ8P/u3IdddiUOfTeVu8gJNSo0p1AnE5nTAXqCPUyGddffz2++uor/PHHH3A4HGjbti0ef/xxTJs2zWNO8+DBg7Fnzx5s27YNa9euhdlsRlZWFoYNG4Z77rmn0fnX06ZNg0ajcatIuaahpKQEr732GjZt2oSLL74Yb7/9dsDDt2UyGV555RXcdNNNeOutt7BmzRqsWbMGgiAgPz8fl19+OcaPH+8RhXvOnDk499xz8dZbb+Gdd94BUF95fOqppzzWIfdGLpdDJpNJBbNcLncrpMXvjUqlgsPhkAog8V/Xc4H6tZ07d+6M9957D//973+l5ahGjRrltdINABMmTMA555yD119/HevWrcMPP/yA1NRUFBcX4/7778ctt9wSUB76MmnSJHTr1g2vvfYafv75Z9TV1SEnJwd9+vTB6NGjQ5IWnufdAqkA/ywZJhKXBbvzzjsxdepUFBcXSw9bw4YNwyeffIIXX3wRn376KbRaLQYOHIjPPvsspIHd5s+fj44dO2L+/Pl44403UFRUhClTpmDs2LEele7nnnsOP/30EzZu3Ij//e9/SE5ORvv27fHWW29hwoQJIUsTIYS0RNnZ2cjMzPQ7R1gQBJSXlyMvLw8ymczrMQ23VVVVwWazJXylWxAEGAwGKJVKCIIAu90Oi8Xi8/iG+dRYMDQxLgoh4cYxGqcSNQaDAampqaitrW00EJbVasXhw4fRtm1br0M9nU4njh49CkEQYqrSDdRX2oqKimIiEJPYiqxUKmMun2JFMHnE8zzsdntMfMbhxvM8lEolZDKZW+OEN029tXobjRAPfN2fBEFAWVkZCgoK4vK6IoXyKToCLX/9oc8vMJRP/gWTR5WVlaisrEz4JR7Fofhig8SJEyeQl5fn0QgezP8BQKvVJtQzDP29BSbS+URNOwlALpejoKAAcrk85iqTMpksoW5kxJPYUtwSiBVj8Zpb0rUTQgiJLS2lh5bnechkMmi1WjDGoFQqoVarqUJJ4krL+GttAcQlhagCQKIl0dchpUFBhBBCYolCoWgRZRPP89BqtSFd/pSQSKNKNyGkWcTe3lAsiRKKSnsoK/6ur+U6/4sQQgiJNoVCAaVSidra2kbnLbv+31cDeWNByKK9z+FwQK1We6SZkHhClW5CSLM0DELS1FZof8c3d39zjm94rFjpppZ2Qggh0aZWq5Gbm+sz0Fpj/3dtKHf9v6/lRRvbH8j7+VruLJA0i40LhMQzqnQTQpot1ubth6siTr3chBBCYgXHcdDr9dFOhpumVPybcmyiR2gniY8q3YSQhNPUyjFVpgkhhJDmC3atbEISHYX9I4QQQgghhBBCwoQq3XGE5pESQmIN3ZcIIYQQQhpHle44IM6XdTgcUU4JIYS4czqdAFrOerGEEEIIIU1Fle44oFQqoVarUVtbS71KhJCYYjAYIJfLYy6YHiGEEEJIrKCuiTiRlZWF0tJSHD9+HKmpqVAqlW5rLzocDvA8T0ErGkH55B/lUWAon+rzwGQywWAwID8/v8XmAyGEEEKIP1TpjhMpKSkAgMrKSpSWlrrtY4yB53nI5XJ68G0E5ZN/lEeBoXyqx3Ec0tLSkJqaGu2kEEIIIYTELKp0x5GUlBSkpKRIPWwiQRBw6tQp5OTkQCajGQO+UD75R3kUGMqnekqlkoaVE0IIIYT4QZXuOKRUKqFUKqXfBUGAQqGARqNp0RUAfyif/KM8CgzlEyGEEEIICRQ9LRJCCCGEEEIIIWFClW5CCCGEEEIIISRMqNJNCCGEEEIIIYSECVW6CSGEEEIIIYSQMKFKNyGEEEIIIYQQEiZU6SaEEEIIIYQQQsKElgyLIsYYAMBgMDTrdQRBgNFohMFgoOWLGkH55B/lUWAon/yjPApMqPNJr9eD47gQpCyxUfkbWZRP/lEeBYbyyT/Ko8BEuvylSncUGY1GAEBxcXGUU0IIISQR1NbWIiUlJdrJiHlU/hJCCAklf+Uvx8TmXhJxgiCgrKys2T0TBoMBxcXFOHbsGD1sNYLyyT/Ko8BQPvlHeRSYUOcT9XQHhsrfyKJ88o/yKDCUT/5RHgUm0uUv9XRHkUwmQ1FRUcheLyUlhf64AkD55B/lUWAon/yjPAoM5VNkUfkbHZRP/lEeBYbyyT/Ko8BEKp9ooD8hhBBCCCGEEBImVOkmhBBCCCGEEELChCrdCUCtVmPGjBlQq9XRTkpMo3zyj/IoMJRP/lEeBYbyKb7R5xcYyif/KI8CQ/nkH+VRYCKdTxRIjRBCCCGEEEIICRPq6SaEEEIIIYQQQsKEKt2EEEIIIYQQQkiYUKWbEEIIIYQQQggJE6p0xzCbzYaHH34YBQUFSEpKQr9+/fDLL78EdG5paSnGjh2LtLQ0pKSk4JprrsHff/8d5hRHR7D59NVXX+H6669Hu3btoNVq0alTJzz44IM4ffp0+BMdYc35LrkaOnQoOI7DvffeG4ZURl9z8+nzzz/HBRdcgOTkZKSlpeHCCy/Er7/+GsYUR15z8mj58uUYPHgwsrKykJaWhr59++LDDz8Mc4qjo66uDjNmzMCwYcOQkZEBjuOwcOHCgM8/ffo0Jk6ciOzsbCQnJ2Pw4MHYunVr+BJM3FD5Gxgqf/2j8jcwVP76R+WvfzFd9jISs2644QamUCjY1KlT2dtvv80uuOACplAo2Nq1axs9z2g0srPOOovl5OSwF154gb3yyiusuLiYFRUVscrKygilPnKCzafMzEzWo0cP9sQTT7D58+ez+++/n6lUKta5c2dmNpsjlPrICDaPXH355ZcsOTmZAWD33HNPGFMbPc3JpxkzZjCO49iYMWPYf//7XzZ37lx21113sQ8++CACKY+cYPPo22+/ZRzHsQsvvJDNnTuXvfHGG2zAgAEMAHvllVcilPrIOXz4MAPAWrVqxQYNGsQAsAULFgR0Ls/z7MILL2TJycnsySefZG+88Qbr2rUr0+v1bP/+/eFNOGGMUfkbKCp//aPyNzBU/vpH5a9/sVz2UqU7Rm3YsIEBYC+99JK0zWKxsPbt27MLLrig0XNfeOEFBoBt3LhR2rZnzx4ml8vZ9OnTw5bmaGhOPq1cudJj26JFixgANn/+/FAnNWqak0eux7dp04Y99dRTCVvoNyef1q9fzziOS7jCq6Hm5NHQoUNZQUEBs1qt0jaHw8Hat2/Pzj777LClOVqsVisrLy9njDG2adOmJhX8n3/+OQPAFi9eLG07deoUS0tLYzfeeGM4kktcUPkbGCp//aPyNzBU/vpH5W9gYrnspUp3jJo2bRqTy+WstrbWbfuzzz7LALCSkhKf5/bp04f16dPHY/tll13G2rdvH/K0RlNz8skbg8HAALApU6aEMplRFYo8mjlzJmvVqhUzm80JW+g3J5+uv/56lp+fz3ieZ4IgMKPRGO7kRkVz8qhfv36sW7duXrf369cv5GmNJU0t+MeMGcNyc3MZz/Nu2ydOnMi0Wq3bgxMJPSp/A0Plr39U/gaGyl//qPxtulgre2lOd4zatm0bOnbsiJSUFLftffv2BQBs377d63mCIODPP/9E7969Pfb17dsXhw4dgtFoDHl6oyXYfPLlxIkTAICsrKyQpC8WNDePSkpK8Pzzz+OFF15AUlJSuJIZdc3JpxUrVqBPnz6YM2cOsrOzodfrkZ+fjzfeeCOcSY645uTRoEGD8Ndff+GJJ57AwYMHcejQIcyaNQubN2/GQw89FM5kx51t27ahV69ekMnci+i+ffvCbDZj//79UUpZy0Dlb2Co/PWPyt/AUPnrH5W/4RfuslfRrLNJ2JSXlyM/P99ju7itrKzM63nV1dWw2Wx+z+3UqVMIUxs9weaTLy+88ALkcjlGjx4dkvTFgubm0YMPPohzzz0XN9xwQ1jSFyuCzaeamhpUVlbit99+w6+//ooZM2agVatWWLBgAe677z4olUrcddddYU17pDTnu/TEE0/g8OHDeOaZZ/D0008DALRaLb788ktcc8014UlwnCovL8eAAQM8trvmc48ePSKdrBaDyt/AUPnrH5W/gaHy1z8qf8Mv3GUvVbpjlMVigVqt9tiu0Wik/b7OAxDUufEo2Hzy5pNPPsF7772Hhx56CGeddVbI0hhtzcmjlStX4ssvv8SGDRvClr5YEWw+1dXVAQCqqqrw2Wef4frrrwcAjB49Gj169MDTTz+dMIV+c75LarUaHTt2xOjRozFy5EjwPI933nkHt9xyC3755Recf/75YUt3vAnlfY00HZW/gaHy1z8qfwND5a9/VP6GX7jLXqp0x6ikpCTYbDaP7VarVdrv6zwAQZ0bj4LNp4bWrl2LCRMm4PLLL8czzzwT0jRGW7B55HQ6cf/99+PWW29Fnz59wprGWNDcvzmlUunWQyOTyXD99ddjxowZKCkpQatWrcKQ6shqzt/bvffeiz/++ANbt26Vhm6NHTsW3bp1w+TJk1vEg2WgQnVfI8Gh8jcwVP76R+VvYKj89Y/K3/ALd9lLc7pjVH5+PsrLyz22i9sKCgq8npeRkQG1Wh3UufEo2HxytWPHDlx99dXo3r07lixZAoUisdqigs2jDz74APv27cNdd92FI0eOSD8AYDQaceTIEZjN5rClO9Ka8zen0WiQmZkJuVzuti8nJwdA/RC4RBBsHtntdrz33nsYMWKE21wppVKJ4cOHY/PmzbDb7eFJdBwKxX2NBI/K38BQ+esflb+BofLXPyp/wy/cZS9VumNUz549sX//fhgMBrftYmtUz549vZ4nk8nQo0cPbN682WPfhg0b0K5dO+j1+pCnN1qCzSfRoUOHMGzYMOTk5GDp0qXQ6XThSmrUBJtHJSUlcDgcuOiii9C2bVvpB6h/IGjbti2WLVsW1rRHUnP+5nr27ImKigqPgkucY5WdnR36BEdBsHlUVVUFp9MJnuc99jkcDgiC4HVfS9WzZ09s3boVgiC4bd+wYQO0Wi06duwYpZS1DFT+BobKX/+o/A0Mlb/+UfkbfmEve5sV+5yEzR9//OGxHp/VamUdOnRwC+9/9OhRtmfPHrdzn3/+eQaAbdq0Sdq2d+9eJpfL2cMPPxz+xEdQc/KpvLyctWvXjhUUFLDDhw9HKskRF2we7dmzh3399dcePwDYFVdcwb7++mtWVlYW0WsJp+Z8l1599VUGgL3zzjvSNovFwtq1a8e6du0a/sRHSLB55HQ6WVpaGuvYsSOz2WzSdqPRyIqKiljnzp0jcwFR0tiyJWVlZWzPnj3MbrdL2z777DOPtUIrKipYWloau/766yOR5BaNyt/AUPnrH5W/gaHy1z8qf5su1speqnTHsDFjxjCFQsGmTZvG3n77bXbhhRcyhULBVq9eLR0zcOBA1rDtxGAwsPbt27OcnBz24osvsldffZUVFxezgoICdurUqUhfRtgFm0/nnHMOA8Aeeugh9uGHH7r9LFu2LNKXEVbB5pE3SNB1QhkLPp/MZjPr1q0bUyqVbOrUqWzOnDmsT58+TC6Xs6VLl0b6MsIq2Dx6+umnGQB27rnnsldffZW9/PLLrEuXLgwA++ijjyJ9GRExd+5cNmvWLPbvf/+bAWAjR45ks2bNYrNmzWKnT59mjDE2btw4BsCt4uF0Otn555/PdDodmzlzJnvzzTdZt27dmF6vZ3v37o3S1bQsVP4Ghspf/6j8DQyVv/5R+RuYWC17qdIdwywWC5s6dSrLy8tjarWa9enTh/30009ux/i6UR87doyNHj2apaSkMJ1Ox6688kp24MCBSCU9ooLNJwA+fwYOHBjBKwi/5nyXGkrkQr85+XTy5Ek2btw4lpGRwdRqNevXr5/HuYmgOXn08ccfs759+7K0tDSWlJTE+vXrx5YsWRKppEdc69atfd5jxILeW8HPGGPV1dVswoQJLDMzk2m1WjZw4EC33lMSXlT+BobKX/+o/A0Mlb/+UfkbmFgteznGGGvqkHRCCCGEEEIIIYT4R4HUCCGEEEIIIYSQMKFKNyGEEEIIIYQQEiZU6SaEEEIIIYQQQsKEKt2EEEIIIYQQQkiYUKWbEEIIIYQQQggJE6p0E0IIIYQQQgghYUKVbkIIIYQQQgghJEyo0k0IIYQQQgghhIQJVboJIYQQQgghhJAwoUo3iQtt2rQBx3GN/rz22mvRTmZQxGs7cuRItJPSZKtWrQLHcRg0aFC0k9Js4veopTly5Ag4jkObNm3C/l5PPvkkOI7Dk08+Gfb3IoQ0H5W9sYnK3vhHZW/Lo4h2AghpiosuuggdOnTwuq9r164RTo1/t912GxYtWoQFCxbgtttui3ZySIxatWoVBg8ejIEDB2LVqlUhfe02bdrg6NGjOHz4cFgL93BeAyEkuqjsJYmIyl4SSVTpJnHljjvuSLgCdMWKFXA4HCgsLIx2UkgLVFhYiD179kCpVIb9ve69917ccMMNyMrKCvt7EUJCh8peQkKLyt6WhyrdhERZ+/bto50E0oIplUp07tw5Iu+VlZVFhT4hJCZQ2UuiicrelofmdJOE4zpP69tvv8Ull1yCjIwMcBwnDb2pqKjAnDlzcMUVV6Bt27ZISkpCSkoKevfujRdeeAFWq9Xn65vNZrz22mvo378/0tPToVar0bp1a1x11VX45JNPAPwzV2fRokUAgPHjx7vNgXOdV9PYvDKz2Yznn38evXr1gl6vh1arRbdu3fD444+jpqbG43jXOUKMMbzzzjs477zzkJycjNTUVFx22WVYv3691+vauHEjHnroIfTt2xd5eXlQqVTIzc3FVVddheXLlweY+4FxncO1YMECXHDBBUhNTXXLh6NHj+KFF17AJZdcglatWkGtViMtLQ39+/fH22+/DUEQQnr9vvA8j3//+9/gOA49evTAsWPHAjrvwIEDuP3229G2bVuo1WrodDq0bt0aI0aMwIIFC6TjBg0ahMGDBwMAVq9e7fY9cR2S1tTv7MKFC8FxHI4ePQoAaNu2rdtri38Ljc0rC/U1+JtXtn//fkyaNAmdOnWCVqtFSkoKunbtikmTJmHXrl0B5TshJDqo7KWyl8peKnuJb9TTTRLW7Nmz8cYbb6B3794YNmwYysrKIJfLAQA///wzJk+ejMLCQnTo0AHnn38+KioqsGHDBjzyyCP49ttvsXLlSqjVarfXPHbsGIYNG4bdu3dDq9XioosuQmZmJkpLS7F27Vrs3LkTN910E3Q6HcaNG4d169bh0KFDHvPhevbs6Tf91dXVGDJkCLZv346UlBRccsklUCqVWL16NZ555hl88skn+PXXX33OFRo/fjw++eQTXHzxxbjyyiuxfft2/PLLL1izZg1Wr16Nfv36uR3/6KOPYuXKlejWrZtUWB46dAjff/89vv/+e7z22muYPHly0z4EP+677z7MmzcPF154IUaMGIG///5beiD48MMP8cQTT6Bt27bo2LEjLrroIpSXl2P9+vX47bffsGzZMixZssRnAJamXr83dXV1GDt2LH788UcMHToUS5YsQUpKit/zdu3ahYsuuggGgwGdOnXClVdeCblcjuPHj2PNmjUoLS3F+PHjAQDDhg2DRqPBzz//jNzcXAwbNkx6HdeW6aZ+Zzt06IBx48ZhyZIlMJlMGDVqFHQ6nfR6eXl5Eb+GxnzyySe4/fbbYbPZ0KpVK1xxxRUQBAF///03/vvf/yInJwfdu3cP6LUIIdFDZS+VvVT2UtlLvGCExIHWrVszAGzBggUBHyuXy9m3337r9Zjdu3ez9evXe2yvrq5ml112GQPAXnzxRbd9PM+z3r17MwDssssuY6dOnXLbb7FY2A8//OC2bdy4cX7TLab38OHDbtuvv/56BoD169ePVVZWStuNRiMbPnw4A8AuvPBCt3MOHz7MADAArHXr1mzfvn3SPqfTyW6//XYp/Q0tXbqUlZWVeWz//fffWUpKClMqlez48eNu+1auXMkAsIEDB/q8Pm/ENKakpHj9HBhjbOPGjWznzp0e20tLS9k555zDALAvvvjCbV9zrl88T3T8+HHWs2dPBoCNHz+e2e32gK9v/PjxDAB7+umnPfaZzWa2evVqt22B5GMw31nGfH+/RGKetW7dOuzXMGPGDAaAzZgxw2375s2bmVKpZBzHsTlz5jCe5932HzlyhG3evNnn6xJCwoPKXip7RVT2uqOylzQVVbpJXBBvXr5+XG824rG33357UO+1b98+BoD16dPHbfs333zDALD8/HxmNBoDeq1gC/6jR48ymUzGOI5jO3bs8Djn+PHjTKPRMADst99+k7a7Fnzfffedx3nl5eUMAFOr1U0qyKZPn84AsDfffNNte3ML/qeeeqpJ54l+/vlnBoCNGTPGbXtzrt+14N+xYwcrKioKOo1XXHEFA8C2bt0a0PHB5qPI13eWseAL/nBcg6+C/9prr2UA2H333RfQexFCIoPKXndU9lLZ64rKXtIUNLycxBVfy5Z4C0YxevToRl+L53msWrUKv//+O8rLy2GxWMDqG6IAAPv27XM7/qeffgIAaQhbOK1ZswaCIKBXr144++yzPfYXFhbi8ssvl4Y1XXjhhW77FQqF2zAjUV5eHtLT01FTU4OqqiqPYU5VVVX44YcfsGvXLtTU1MDhcACon18EeOZJc/n7jGw2G5YtW4ZNmzbh1KlTsNlsYIzBaDQ2mp5grx+oH0o2ZswY2Gw2fPjhh7jllluafF19+/bF0qVL8e9//xszZ87EwIEDodFomvw6DTX1O9sc4bqGhniexy+//AIAmDhxYshfnxDSfFT21qOyl8peKntJsKjSTeJKU5YtaWxdxAMHDuC6667DX3/95fMYg8Hg9rsYFCMS0SZLS0sB1Afg8EWMvCoe6yo/P9/nMhQpKSmoqanxCP4xf/58PPDAAzCZTD7fs2GeNFdjn9Eff/yB66+/HiUlJU1OTzDXL7ryyivhdDrx0Ucf4eabb/ad+EZMmzYN69atw/LlyzFs2DAolUqcc845GDBgAG644Qb06dOnya8ZzHe2OcJxDd5UVVVJ37lOnTqF5DUJIaFFZe8/qOylsrchKntJICh6OUlYSUlJPveNHj0af/31F6688kqsWbMGlZWVsNvtYIzBZrNFMJXhIZM17U97y5YtuOuuu2Cz2fDCCy9g9+7dqKurgyAIYIzh7bffBgCpVTdUfH1GZrMZ1157LUpKSjB+/Hhs3LgR1dXVcDqdYIxJrcq+0tPU63c1btw4AMATTzyBw4cPB/UaWq0Wv/zyCzZu3IinnnoKQ4YMwf79+/HKK6+gb9++uOeee5r8mpH+zobjGgghiY/K3sBR2fsPKnvDdw0kNlBPN2lx9u7diz///BM5OTn4+uuvoVC4/xmIw7kaatWqlXR+uBUWFgIA/v77b5/HiPvEY5tj8eLFYIzhvvvuw0MPPeSx31eehMuaNWtw8uRJ9OrVC++//35E0zN//nzodDq8/vrruPjii7F8+fKge1j69OkjtUo7nU588803+Ne//oV58+Zh9OjR0lIf/gT7nQ2FUF2DL5mZmdBqtTCbzdi3bx9FSSUkQVHZ64nK3n9Q2euOyt7EQz3dpMWprq4GABQUFHjcQAHgo48+8nqeOE/p008/bXQYmCuVSgWg/obZFAMGDIBMJsP27duxY8cOj/3l5eXSPLfm3niBf/KkdevWHvusViu+/PLLZr9HMOkRH7Ya8vUZhQLHcXjttdfw+OOPo7S0FAMGDMD27dub/boKhQKjR4/G5ZdfDgBur+nvexLsdzaQ126K5lyDL3K5HEOHDgVQ/9BFCElMVPZ6orL3H1T2+kZlb2KgSjdpcTp27Ai5XI6dO3di1apVbvv+97//4dVXX/V63tVXX41zzz0XZWVlGDNmDKqqqtz2W61W/Pjjj27bioqKAKDRuUDetGrVCmPGjAFjDHfddZfbe5lMJkycOBFWqxUXXnihRyCXYHTp0gUAsGjRIilQClB/TZMmTQp6qFdz07NixQrs3r3bbd8777yDzz//POxpmDVrFl588UVUVFRg8ODBWL9+fcDnzps3z2tglRMnTmDz5s0A3B+yxO/JgQMHpAA6roL9zrq+dlO/g6G+hsY89thjUCgUeOONNzBv3jyPoYtHjx7Fli1bmvSahJDYQmWvJyp7PVHZS2VvwopMkHRCmieYtUJ9LdPAGGOTJ09mAJhMJmMDBw5kN954I+vVqxcDwB5//HGPdSNFR44cYZ06dWIAmFarZZdddhm78cYb2YABA1hqaqrH0g87duxgMpmMyWQydumll7Lx48ezCRMmuK1h6iu9lZWV0pqYqamp7Nprr2WjR49m2dnZDABr27atxzm+lqDwlz81NTXS9szMTHbttdeyUaNGsZycHKbX66X8GjdunNtrNXfZksZcc801DABTqVTssssuYzfccAPr3Lkz4ziOPfbYY16vM9jrbyxNb731FuM4jiUnJ7MVK1YEdH3i59a2bVt21VVXsZtvvplddtllLCkpiQFgl1xyCXM4HG7niOvQdurUid18881swoQJ7OGHH5b2B/udfeONNxgAptPp2MiRI9mECRPYhAkT2N69exvNs3Bcg69lSxhjbNGiRUypVEppGT16NBs5ciTr2bMn4zjO6zmEkPCispfKXip7qewloUGVbhIXQl3wC4LA3nvvPXbeeecxnU7HUlNTWf/+/dlnn33GGGu8YDIajeyFF15gffr0YXq9nqnVata6dWt29dVXS+e7+vrrr9lFF13E9Ho94zjO48bXWHpNJhN77rnnWM+ePZlWq2UajYZ16dKFPfroo6y6utrj+OYUfBUVFWzSpEmsffv2TK1Ws4KCAnbLLbewAwcOsAULFkS84Lfb7eyll15iPXr0YFqtlmVkZLDLLruMLVu2zOd1hqPgZ4yxjz76iCkUCqbRaNj//vc/v9f3/fffs3//+9/s3HPPZdnZ2UylUrGioiI2aNAgtmjRIq/rtB49epTddNNNLD8/nykUCo/rCPY7y/M8e+6551i3bt2k9WUBsJUrVzLGfOdZOK6hsYKfMcb++usvNmHCBNa2bVumVqtZamoq69q1K7v33nvZX3/91Ximk/9v546NAIBBIIaR/YcmS+BOWuLPBcA522t7ba/t5cbbPX6JCAAAAMyMm24AAADIiG4AAACIiG4AAACIiG4AAACIiG4AAACIiG4AAACIiG4AAACIiG4AAACIiG4AAACIiG4AAACIiG4AAACIiG4AAACIiG4AAACIfN4UM5+3mG3WAAAAAElFTkSuQmCC","text/plain":["
"]},"metadata":{},"output_type":"display_data"}],"source":["fig = bf.diagnostics.plot_sbc_ecdf(posterior_draws, new_sims[\"parameters\"], difference=True)"]},{"cell_type":"markdown","id":"e9ee69a1","metadata":{},"source":["### Two Moons Posterior \n","\n","The two moons posterior at point $x = (0, 0)$ should resemble two crescent shapes. Below, we plot the corresponding posterior samples and posterior density. These results suggest that our spline flow setup can approximate the expected analytical posterior fairly well."]},{"cell_type":"code","execution_count":14,"id":"065384db","metadata":{},"outputs":[{"data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAA90AAAGGCAYAAABmGOKbAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAAB1ZUlEQVR4nO3deXhV1b3/8c8+JzMkJCEhDKIIQjCAqEEQK9oqlcFa7K+mjpVart4KeG2ht2prHWkdqq2tMtSKKLdaucHr0IpYJ0QUGQIqEAmzQiDDSUISMp5h//4IOeTsnIxkJznh/XqePLL22WvvtVdiVr57TYZpmqYAAAAAAECHc3R1AQAAAAAA6KkIugEAAAAAsAlBNwAAAAAANiHoBgAAAADAJgTdAAAAAADYhKAbAAAAAACbEHQDAAAAAGATgm4AAAAAAGxC0A0AAAAAgE0IugEAAAAAsAlBNwAAAAAANiHoBprxwgsvyDAMHThwoKuL0i1RPx3n8ccf18iRI+Xz+bq6KFqyZIlOP/101dTUdHVRAKBdukP79MADD8gwjC67f73uUBehrLu0z7TNoY2gGx2q/hd7/VdUVJRGjBihuXPnKj8/v8Pv9+mnn+qBBx7Q0aNHO/zaCG2h9LNRVlamxx57THfddZccjsBfyzU1Nbrrrrs0cOBARUdHa8KECXr33XfbdZ/WXusnP/mJamtr9de//rVd9wEQemi/Tx1dXfddff+26E7tM21zaCPohi0eeugh/c///I+eeeYZXXTRRVq8eLEmTpyoysrKDr3Pp59+qgcffNC2X9w//vGPVVVVpTPOOMOW68M+dv9sdKTnn39eHo9H119/faPPfvKTn+iPf/yjbrzxRv35z3+W0+nU9OnTtW7dujbfp7XXioqK0syZM/XHP/5Rpmm2+7kAhB7a757HWhdd3T529f3boju1z7TNIc4EOtCyZctMSeamTZsCjs+bN8+UZL788ssder8//OEPpiRz//79HXrdY8eOdej17L5uV6n/fnd0/XcEu3427HDOOeeYN910U6PjGzZsMCWZf/jDH/zHqqqqzGHDhpkTJ05s0z3aeq3Nmzebksz333+/TfcBEJpovzv+uvfff7/ZHf/U7ur2savv3xbdrX2mbQ5d9HSjU1x22WWSpP379/uPbd26VdOmTVNcXJx69+6tyy+/XJ999llAvvLycv385z/XkCFDFBkZqX79+um73/2utmzZogceeED//d//LUk688wz/UPiGs5Zys3N1U9/+lOlpKQoMjJSo0aN0vPPPx9wj/o5V9nZ2brhhhuUkJCgiy++WFLT86BaU/bmrhtMc89a7+uvv9bs2bOVmpqq6Oho9e3bVxkZGY3KV3/vXbt26aabblKfPn2UnJys3/72tzJNUwcPHtSMGTMUFxen/v3768knnwyaf+fOnfrRj36kuLg49e3bV3feeaeqq6ubfIa21HtrnjeY1tRBa342gpU5KipKP/3pTwOOv/feewoPD9cvfvGLFp+7Pfbv368vv/xSkydPbvTZypUr5XQ6ddttt/mPRUVFadasWVq/fr0OHjzY6vu09Vrp6elKTEzUG2+80Y6nAtBT0H633H5L0rp163TBBRcoKipKw4YNa3IIcGueq2EZ9uzZo5/85CeKj49Xnz59dMsttwSMOmhNW9qwLpqr+w8//FCGYei1115rVJ6XX35ZhmFo/fr1TdYB7bP97TNtc+gK6+oC4NSwd+9eSVLfvn0lSTt27NCkSZMUFxenX/3qVwoPD9df//pXffvb39ZHH32kCRMmSJJ+9rOfaeXKlZo7d67S0tJUVFSkdevW6auvvtL/+3//T7t27dI//vEP/elPf1JSUpIkKTk5WZKUn5+vCy+8UIZhaO7cuUpOTtbbb7+tWbNmqaysTD//+c8DypiRkaHhw4fr97//fbPDdlpb9rZet7lnPf/88yVJmzZt0qeffqrrrrtOp512mg4cOKDFixfr29/+trKzsxUTExNwzWuvvVZnn322Hn30Ub311ltasGCBEhMT9de//lWXXXaZHnvsMb300kv65S9/qQsuuECXXHJJQP4f/ehHGjJkiB555BF99tln+stf/qKSkhItX768yedobb235nmDaU0dtPSzEcygQYP0H//xH3r22Wd1//3364wzztDOnTuVkZGhadOmNXoxIUlut1ulpaVNXrOhxMTERvPBpLphdpKCPvPWrVs1YsQIxcXFBRwfP368JOnzzz/X4MGDW3X/9lzr/PPP1yeffNKq6wPomWi/W77utm3bdMUVVyg5OVkPPPCAPB6P7r//fqWkpASc19bnkura4TPPPFOPPPKItmzZoueee079+vXTY4891mI9B2tXmqv7M844Q4MHD9ZLL72kH/zgBwH5XnrpJQ0bNkwTJ05ssh5onzunfaZtDlFd2s+OHqd+eNp7771nFhYWmgcPHjRfeeUVs2/fvmZ0dLR56NAh0zRN8+qrrzYjIiLMvXv3+vMePnzYjI2NNS+55BL/sT59+phz5sxp8n7NDVGaNWuWOWDAANPlcgUcv+6668w+ffqYlZWVpmmeGP51/fXXN/k8Da/f2rI3d91gWnpW0zT9ZW5o/fr1piRz+fLlje592223+Y95PB7ztNNOMw3DMB999FH/8ZKSEjM6OtqcOXNmo/zf//73A+41e/ZsU5L5xRdfmKYZvH5aW++ted6TqYP2DF87dOiQGRkZad5+++2my+Uyhw0bZp577rlNDiv88MMPTUmt+mqqHPfee68pySwvL2/02ahRo8zLLrus0fEdO3aYkswlS5a0+tnac63bbrvNjI6ObvU9AIQu2u/2t99XX321GRUVZX799df+Y9nZ2abT6QwYXt7a52pYhp/+9KcB5/7gBz8w+/bt60+3pi211kVzdX/PPfeYkZGR5tGjR/3HCgoKzLCwMPP+++9v9j60z53TPtM2hyaGl8MWkydPVnJysgYPHqzrrrtOvXv31muvvaZBgwbJ6/Xq3//+t66++moNHTrUn2fAgAG64YYbtG7dOpWVlUmS4uPjtWHDBh0+fLhN9zdNU6+++qquuuoqmaYpl8vl/5oyZYpKS0sbDWP+2c9+1uJ121L2tlxXat2zRkdH+//tdrtVVFSks846S/Hx8UGHZf/Hf/yH/99Op1Pjxo2TaZqaNWtWwH1TU1O1b9++RvnnzJkTkL7jjjskSatWrQpavrbUe3u/t22tg7YYNGiQbr31Vj3//PO68sorVVVVpX/961/q1atX0PPHjh2rd999t1Vf/fv3D3qNoqIihYWFqXfv3o0+q6qqUmRkZKPjUVFR/s9bqz3XSkhIUFVVVYcvoASg+6L9bvt133nnHV199dU6/fTT/cfPPvtsTZky5aSeK1gZJk2apKKiopOu56bcfPPNqqmp0cqVK/3HVqxYIY/Ho5tuuqnZvLTPndM+0zaHJoaXwxYLFy7UiBEjFBYWppSUFKWmpvqH7hQWFqqyslKpqamN8p199tny+Xw6ePCgRo0apccff1wzZ87U4MGDlZ6erunTp+vmm28OaDCDKSws1NGjR/Xss8/q2WefDXpOQUFBQPrMM89s8bnaUva2XFdSq561qqpKjzzyiJYtW6bc3NyA4W7BhlE1/ANAkvr06aOoqCj/cK6Gx4uKihrlHz58eEB62LBhcjgcTc69aku9t/d729Y6aKtf/vKXeuaZZ/Tll1/q448/1qBBg5o8NyEhIehcr44SHR0ddD/O+nn1Df/AseNa9XXbHfaZBdA5aL/b1n4XFhaqqqqqUXspSampqf6X1O15LqlxO56QkCBJKikpUVxcXLvruSkjR47UBRdcoJdeesn/gv6ll17ShRdeqLPOOqvZvLTPndM+0zaHJoJu2GL8+PEaN27cSV/nRz/6kSZNmqTXXntN//73v/WHP/xBjz32mP7v//5P06ZNazKfz+eTJN10002aOXNm0HPOOeecgHRbfkG2RWuv25pnveOOO7Rs2TL9/Oc/18SJE9WnTx8ZhqHrrrvO/8wNOZ3OVh2T1KrtJ1r6Bd+Wem/v97atddBWv/vd7yRJHo9HiYmJzZ5bW1ur4uLiVl03OTk5aN337dtXHo9H5eXlio2NDfhswIABys3NbZTnyJEjkqSBAwe26t7tvVZJSYliYmJs+38DQPdD+23PddvzXFLLbXZ767k5N998s+68804dOnRINTU1+uyzz/TMM8+0mI/2uXPaZ9rm0ETQjU6XnJysmJgY5eTkNPps586dcjgcAQtGDBgwQLNnz9bs2bNVUFCg888/X7/73e80bdq0JoPA5ORkxcbGyuv1duibzraWva2ae1apboXLmTNnBiwaUl1dbdtel7t37w54079nzx75fD4NGTIk6PltrfeWnjeY1tZBe94A/+EPf9Bzzz2nZ555Rv/93/+t3/3ud3ruueeaPP/TTz/Vd77znVZde//+/UHrbeTIkf7PrX9wnXvuufrwww9VVlYWsMDKhg0b/J+3VnuutX//fp199tmtvgeAno32O/h1o6OjtXv37kafNbyXXc8ltb0tbal9vO666zRv3jz94x//UFVVlcLDw3Xttde2WA7a585pn2mbQxNzutHpnE6nrrjiCr3xxhsBw5Tz8/P18ssv6+KLL1ZcXJy8Xm+j4Uj9+vXTwIED/cNw6ufyWH+hO51O/fCHP9Srr76q7du3NypDYWGhrWVvq9Y8a/39rT3STz/9tLxeb5vv2RoLFy5sdC9JTTbkra331j5vU/doTR009bPRlNdff1133323Hn74Yc2ZM0e33Xabli9fHrBNjlVHzBmrXwl28+bNjT675ppr5PV6A4Yi1tTUaNmyZZowYUKb/kBsz7W2bNmiiy66qNX3ANCz0X4Hv+6UKVP0+uuv65tvvvEf/+qrr/TOO+8EnNfRz9XetrSl9jEpKUnTpk3T3//+d7300kuaOnVqo2lpwdA+d077TNscmujpRpdYsGCB3n33XV188cWaPXu2wsLC9Ne//lU1NTV6/PHHJdXtPXnaaafpmmuu0dixY9W7d2+999572rRpk/8tanp6uiTpN7/5ja677jqFh4frqquuUq9evfToo4/qww8/1IQJE3TrrbcqLS1NxcXF2rJli957771WDztqT9nbqjXPKknf+9739D//8z/q06eP0tLStH79er333nv+rVw62v79+/X9739fU6dO1fr16/X3v/9dN9xwg8aOHdtkntbUe2ufN5jW1kFzPxtWWVlZuvHGG3XjjTfqN7/5jSTpV7/6lZYsWdLs2/SOmDM2dOhQjR49Wu+9916jPUgnTJigjIwM3XPPPSooKNBZZ52lF198UQcOHNDSpUsbXcswDF166aVas2ZNo8/aeq2srCwVFxdrxowZJ/V8AHoW2u/GHnzwQa1evVqTJk3S7Nmz5fF49PTTT2vUqFH68ssv/ed19HO1ty1tTft4880365prrpEkPfzww60qD+1z8Da1uba5rdeibQ5hnbxaOnq4+m0pNm3a1OK5W7ZsMadMmWL27t3bjImJMb/zne+Yn376qf/zmpoa87//+7/NsWPHmrGxsWavXr3MsWPHmosWLQq4zsMPP2wOGjTIdDgcjbZ+yM/PN+fMmWMOHjzYDA8PN/v3729efvnl5rPPPus/p35bjsLCwiafx7qdREtlb+m6Vq191pKSEvOWW24xk5KSzN69e5tTpkwxd+7caZ5xxhlBt/yy3nvmzJlmr169Gt3/0ksvNUeNGtUof3Z2tnnNNdeYsbGxZkJCgjl37lyzqqqqxfppqd5b+7zBtLYOTLP5n416Bw8eNAcMGGB+61vfMqurqwM+u/32283w8HBz3759LZbrZPzxj380e/fuHXS7laqqKvOXv/yl2b9/fzMyMtK84IILzNWrVzc6r7y83JRkXnfddU3ep7XXMk3TvOuuu8zTTz/d9Pl87X8wACGD9vuEtrTf9T766CMzPT3djIiIMIcOHWouWbLEf52GWvNczZWh4XO1tp6D1UVL7WNNTY2ZkJBg9unTJ6Ddbw7tc+M2tTVtc2uvZZq0zaHMMM1WrJ4E4JTywAMP6MEHH1RhYWGrhpTh5JSWlmro0KF6/PHHA7Zza4tVq1bpe9/7nr744guNGTPmpMpTU1OjIUOG6O6779add955UtcCAIQej8ejgQMH6qqrrgo6GupUcbLtM20z6jGnGwC6WJ8+ffSrX/1Kf/jDH9q9wuuHH36o66677qQbdUlatmyZwsPDW73HPACgZ3n99ddVWFiom2++uauL0qVOtn2mbUY9eroBNEJPNwAAp54NGzboyy+/1MMPP6ykpCRt2bKlq4sE9Aj0dAMAAADQ4sWLdfvtt6tfv35avnx5VxcH6DHo6QYAAAAAwCb0dAMAAAAAYBOCbgAAAAAAbBLW1QXoaKZpqry8XLGxsTIMo6uLAwAAmuDz+XT48GHabABASKqPPQcOHCiHo+n+7B4XdJeXl6tPnz4qLS1VXFxcVxenRT6fTwUFBerXr1+z3yi0HXVrD+rVPtStfajb7unw4cMaPHhwVxcDAICTcvDgQZ122mlNft7jgm4AABAaYmNjJUnRkujnBgCEGlNSlU60Z00h6AYAAF2ifki5IYJuAEDoammKFGPsAAAAAACwCUE3AAAAAAA2IegGAAAAAMAmBN0AAAAAANiEoBsAAAAAAJsQdAMAAAAAYBOCbgAAAAAAbELQDQAAAACATQi6AQAAAACwCUE3AAAAAAA2IegGAAAAAMAmBN0AAAAAANiEoBsAAAAAAJsQdAMAAAAAYBOCbgAAAAAAbELQDQAAAACATQi6AQAAAACwCUE3AAAAAAA2IegGAAAAAMAmBN0AAAAAANiEoBsAAAAAAJsQdAMAAAAAYBOCbgAAAAAAbELQDQAAAACATQi6AQAAAACwie1B98KFCzVkyBBFRUVpwoQJ2rhxY6vyvfLKKzIMQ1dffbW9BQQAAAAAwCa2Bt0rVqzQvHnzdP/992vLli0aO3aspkyZooKCgmbzHThwQL/85S81adIkO4sHAAAa4EU5AAAdz9ag+49//KNuvfVW3XLLLUpLS9OSJUsUExOj559/vsk8Xq9XN954ox588EENHTrUzuIBAIDjeFEOAIA9bAu6a2trlZWVpcmTJ5+4mcOhyZMna/369U3me+ihh9SvXz/NmjXLrqIBAAALXpQDAGCPMLsu7HK55PV6lZKSEnA8JSVFO3fuDJpn3bp1Wrp0qT7//PNW36empkY1NTX+dFlZmSTJ5/PJ5/O1veCdzOfzyTTNkChrqKFu7UG92oe6tU+o1a3D0bnrnNa/KL/nnnsCytCWF+Uff/xxZxQVAICQY1vQ3Vbl5eX68Y9/rL/97W9KSkpqdb5HHnlEDz74YKPjhYWFqq6u7sgi2sLn86m0tFSmaXb6H1k9HXVrD+rVPtStfUKtbvv379+p9+vqF+UAAPRktgXdSUlJcjqdys/PDzien58f9I+JvXv36sCBA7rqqqv8x+p7JMLCwpSTk6Nhw4Y1ynfPPfdo3rx5/nRZWZkGDx6s5ORkxcXFddTj2Mbn88kwDCUnJ4fEH4KhhLq1B/VqH+rWPtRtx+roF+UAAPRktgXdERERSk9P1/vvv+9fzdTn8+n999/X3LlzG50/cuRIbdu2LeDYvffeq/Lycv35z3/W4MGDg94nMjJSkZGRjY47HI6Q+cPKMIyQKm8ooW7tQb3ah7q1D3XbtK5+UQ4AQE9m6/DyefPmaebMmRo3bpzGjx+vp556ShUVFbrlllskSTfffLMGDRqkRx55RFFRURo9enRA/vj4eElqdBwAAHScrn5RDgBAT2Zr0H3ttdeqsLBQ9913n/Ly8nTuuedq9erV/jlj33zzDT0OAAB0A7woBwDAHrYvpDZ37tygb8klac2aNc3mfeGFFzq+QAAAoBFelAMAYA/DNE2zqwvRkcrKytSnTx+VlpaGzEJqBQUF6tevH3/MdDDq1h7Uq32oW/tQt91TfZsdI8no6sIAANBGpqRKqcXYk788AAAAAACwCUE3AAAAAAA2IegGAAAAAMAmBN0AAAAAANiEoBsAAAAAAJsQdAMAAAAAYBOCbgAAAAAAbELQDQAAAACATQi6AQAAAACwCUE3AAAAAAA2IegGAAAAAMAmBN0AAAAAANiEoBsAAAAAAJsQdAMAAAAAYBOCbgAAAAAAbELQDQAAAACATQi6AQAAAACwCUE3AAAAAAA2IegGAAAAAMAmBN0AAAAAANiEoBsAAAAAAJsQdAMAAAAAYBOCbgAAAAAAbELQDQCdyJWTo7ULFsiVk9PVRQEAAEAnIOgGgE6UnZmprUuXKjszM+jnZYcO6ePf/56gHAAAoIcI6+oCAEB35crJUXZmptIyMpSUmtoh+dMyMgL+a/XNJ59ox7JlMiRdcu+97S47AACngmA9iE5LOryFtCT1sqTdlnRFkDzVlrQvyDmARNANAE2q75WWWh8ANwy06/NXFhYqJjm5yUC7odO/9S2Fl5c3ee7JvggAAABA5yLoBoAmNNUrHSzwrT9WWVio7Fdf1YE1azRx/nxJUtHu3fpi+XJVFhZKhqEvXnxRRbt2KToxUTJNjZs9W0mpqSratUvffPKJxl5zTZMBdXteBAAAAKDrEHQDQBD1QfSA9PSAAHv3229r1dy5ch87pkqXSzFJSf5e7c/+9Cd5amslw9DXH32kskOHNGj8eOVu2iR3VZWqSkoUnZAgn9er/R9+qJrSUknSoY0bddqECTIlHdyzR+Hl5bq0iYDa+iKgqRcAmxctkgxD426/nR5xAEC30NJiUtZh4VLjoeBRlnRskDz9LemzLOmRQfJYW8pCS/rTIHm+tqSLW7iG1HiYunUYuzWNnoGgG0DIaOvQ6racbw1U63uUD6xZo4Lt27V12TJNf+YZrX3oIR3dv19h0dH6/MUXVXvsmLJXrtTljzwi0zTlPnbMf82inBwV7dolmaZkGPpm3Tr1GzNGUfHxqi0vlzMyUp7qauV/8YWKdu7UiO9/X1EJCSrevVu7335bR7KyGpU9KTXV38PtysnRqjlz5Nq5U9KJnu/NixYp69ln5QgPV0xSEj3iAAAAXYigG0DIaOvQav+c6gY90tbguz7YPrB2rVzZ2XKEh+vARx/JNE2dPmmSTp80SUe2bFFFQYH+edttcobXvXP3VFbKU1kpScr/8ku9dvPNMgxDMoy6ILte/b9NU0f379fR/fvr0g6H5PP5/236fNrx6qvqlZamA198ofwvv1TpgQN1veQVFZo4f76GT5sWUO5Vc+Yo7/PPFdmnj3qlpGjtggUakJ6u3E2bZDid6jd6dKvmkQMAAMA+BN0AQkZLK3835MrJUWVhoVJnzJBMMyBYty529sXy5ao9dkw+r1fhMTEq+PJLmT6fqouLVV1aqprychlOp8oPHaoLqoME1lUuV9sexucL+Le3pkZy1w0qM71elR08qNqKCu1//315qqpUsn+/zv3JT7Rp0SI5wsLkc7vl83jkCA9XeW6uNi1cqJrSUmWvXCnXzp1KOPNMDbrgAn9dbF68OGD+OAAAHck6dLw9Q8Xjg+QZbEmPtaR/ECTPBcMtB66xpL8fJNMES9qyNPnMJUHyPBGY/PxwYPqjIFm+sKR3W9KHguQ5akkzBD30EHQDCBkNh1a3JDszU9mvvqqkkSM1cf58xSQnq1dKip6bOFGeqipVulyqPB4on37JJdr//vvyVVbKCDvxa/FYXp4OfvqpTI/nxIVNMzDgbsgajLdTRFycqoqLJdOUz+2WDEMV+fla9/vfy1N94q+A8F69FJuUpIr8fJUdOqS+I0aosrhY3poaFe/dq9KlSyXDUExSkr548UVJUkxyMsPNAQAAOhFBN4AeKS0jQwfWrJFr504dycrSJffeq+cmTlTuZ59JDodSxozRgTVrVLxnj6ITE+UID1dUfLxSzjlHhzdtkjMioq6Xu6ys9TftgIBbUt28cEtPuruysm5IutMpeb1151VWquboURkOh6rLypS3dauGXnGFKgsKVH30qP8FQVpGRt0LhuP/BgAAQOch6AbQY1gXTpu+cKE2L1qkSpdLu99+W9UlJZJhyHA45KmuVlFOjiTJU1OjyNhYGQ6HqoqL1W/UKIXFxOjAhx8G9nJ3AsPhkHk8qA5gmv5gu+Gxivz8un97vTIdDh3ZskUX/uIX+urVV2VKik5K0hu33KKw6Gh5KiuVPHo0+3wDAE6adTh5a4aKp1jS1lHg1hHeknSTJd1rjuXAfwbJNOYyy4EplvSgIJksor8VmP6Fdd1xSb/4JCB5rl4KTL+xtnGeOwKTKw8Gpt8IUpQtlrR1VXTLSHhJki/IMXQdgm4APUbDhdbSMjK0efFi7f/oIx3dt085b76powcOSJIievc+saCZJNPjUc2xY3IYhlxffSWvx1M357qDeq7bwvS1vpk0wsNluk/M7PLW1Kj80CGtffhhRfTqJU9NjT55/PGAFdVLDx2S8/gQ+vo57QTgAAAA9iHoBhASrL3YwbYDa7jQWnZmprYuWyZ3RYUMh0Nlhw75g+ia0tJGAbXpdssryRkV1bhH2W7tnAveMOBuyOd21w0vl+qGo1v0HjDAv//45iVLdGDNGk2cPz/oFmUAAAA4OS3tUQ8AncaVk6O1CxbIdXzYd0P1vdjZmZn+9OYlS7RqzhztfvttvXbzzXrjlls0ID1dSamp6pWSIp/HU9dzbBh1C5IdZzidchzv7TUcgb8GvbW1Nj5hE2zqUTfCwuq2MavncKj88GEd2bJF2//xD1W6XIobPFiunTu1/sknA+oXAAAAHYOebgDdRsOe1+kLFwb0uDbsxc7629+0ceFC+TweFWzbpvVPPqmDn3wiT02N1j/5pBKGDtVHDz3k30e7PuB2RETI9HhkNAjCTWvA29rh3SezUrnDoV7JyaouK5O3qqp912jO8UC74Xx0Z2SkeiUnq/zIEXndbu199105nE6dedlliujVS6OuvVYV+flKy8gIOooAAHDqCtZLZ53DHWtJJwfJM9KSts60vsY66VuSrBtuTLOkewXJs+aDwPRmS9o6SVpqvHeXdcL5zUHyTJ1uOTA7MDljXuM8M/4SkLzmrsCyjXu8cZY/WdLWrcgsO5VJkiqDHEPXIegG0CWaGh5ev+J4dmZmwNZWDbcLe2n6dFXk5clwONRnyBCdcemlKtm3TxWFhTrj0kv13l131e2prbpttbxut3y1tYobNEiRcXHK/+L4LpkOhxxOp3zHg1NneLi8bnfrgmnTbD7wNgyFRUaqz5AhKt61S6bPJ0d4uKITE1VRUKBKl6v5+dsOR+tfAAQrm4XX7dax/Py6hdrcblUUFCgmMVG7V62Su7JS7ooKzXjhBUnSqjlz5Nq5U5WFhYpJTib4BgAAOAkMLwfQJazDxSX5Vxwf97OfNbu11cV3363Y005T4vDh8tbUaM+qVSr95hu5jx3Tnrff1uGsrLoTDUNRCQn+Xu2j+/eraNeuuoDWMKT6lcJNs24YtsPRYsBtOBwyjs+TNpxORcTFyRkREbC/d2R8vCbccYfS//M/dd3rr2v4974nGYZM09SgCRPU75xzFJ2QEFAOIzxccjgU0auXevXvr+iEBIVFR7e3ehvz+eRzu08Ms/f5VOlyqaqoSJ6qKuV9+aU2L16sVXPmKG/rVjkjI1VVUsKQcwAAgJNETzeATtOwd7vhcPGGGvZoN5d/3sGDcuXkaPOiRTqwdm3dVmBOpxLPOkvn3nKLPnroIcX07auiPXsCruFpOJzb41F9iG36fPLW1Pg/qg+srdt31fdOOyMjFZOUpIr8fEUnJem0CRO0Z/VqSdLEefN06W9/68/z3ccfV+nXX8u1c6ei4uN15plnauuyZUoZM0YpY8Zo/4cfqvzwYUUlJMjr8ai6pERJI0aoLDc3sLwdyHA4GvW0f/nSS6otK6t7EWAYik5IUOqMGaosLNTut99moTUAOEVYe+UaL8nZmHVsVrClPq1bWzVatrQ0SCbrHlr/Z0mva5zlHcvNN1g+t44klyTLzl3qtTkwPfYfjfNM06qA9KSBgWm9H+RGI/9fYPqx2wOSQ85f3CjLL64LTBdbPi8Pchtr/QdfehWdhaAbQKdpuKXXJffe22RwXc86BH3zokX6YvlyFe3erb7Dh9cF7Iahkn37lDhsmKITEpQ4fLh2rFihq559VqvmzpWnwrKvptMZuDp5g2HcDqfT3/sddK9s1fVih0VFST5f3dxwj0fVxcWKP+MMnX7xxXLt3Bm4eJnqXiQM+fa3Vfr113U93KYph8OhIZdconGzZ6u6rEyGpPNuu037t29X2ebNkmHI3bDshiFnRETAi4E2azAcvuFcdiMsTD63W1Uul/+Y93hPd3RionLefFMFO3aoZO9eSWrx+wYAAIATCLoBdJqmerebYg3S6xcIK961S9+sXVt30vEANjohQcfy8vT5Cy+o9OuvVZSTo8QRI1RVXCzT55NpmvJWV6tXv34qP3z4xDByn09R8fGqLi2VERam2IEDVXP0qNzV1XULsTWcW20YOnfmTI27/XZtXrxYJQcOKLpvX6WMHatxs2f7yxzs+cbdfrtikpICPzMMbV60SN989NHxpKF+o0ap4L33lDJmjMKjo5W3dasMp1NDJ0+WaZra/a9/BfRQh/Xq1fjFQlOaGDof2bu3eqWkqHj3bkXFx6u2okKe6mrtWLFCSWlpOn3SJElSv9GjW/29AwAAQB2CbgCdprmh48FYg/T6wHVAerp/qLMkxSQn+4/1SknRukcf1bEjR1Sxbp36n3++Ths/XluXLZPD6VRMUpIqXa66rcGOL4bmrq6W4XQqKTVVF8yZox0rVmjUtdeqcPt2VZWU6Ot161T29ddKHD5c426/XUmpqYpJSlLOG2/ovFmzAp6pqeezPntMcrK2Ll2q1BkzNPbmmyXDUNo116jk2DH1djg06vizrbzuOrmysxXZp48m/frXqiwoUN4XX/i3Ohs0YYIOffaZf6X2VmsQgNdWVKh23z4ZDoeqS0tl+nwK79VLnqoqFeXkqKasTN6aGo372c8YWg4AkNR4aLh1w82yIHnyLGnLCG5VW8efSzr4XmDauqlodpD7FFqva0kHG2odfHzbCZ8FOWYZTK5zLcuI/+jsxnmm32QZH7/UcsK1lzTKM+SjtQHpaZYR6AeClM36VwHDy7sWQTeAbssaqDZMD592Ys8Q67EzLrlEb9xyiw5v3qwjWVnyVFdr5NVXKzoxUcOmTNHed95RVXGx8rdtU8G2bfJWV8sRFqYh3/62vvn4Yx385BOF9+qlQRdcoHGzZ2uSFHSl9Yb/basB6ek6sGaNhk2Z4i+3z+eTt6BAZ/3613I4HHLl5NS9IHC7Vbxrl5JSUzXjhReUnZnpf8mw66232h5wWzTcw1yqm+/dKzlZFQUFik5MVM3Ro4qMj9eA9PSTug8AAMCpiNXLAYQ8V06O1i5YIFdO3fvvpNRUzVi2TMmjRsn0+VS8e7f6jhihcbffriNZWRp3++36wfLluuaVVzTmhhuUMnasRt9wg8bdfruK9+yRp6ZGeVu3auvSpdq8eHHQPavrXwC0t+f3SFaWSvbu1ZH6ldaDPNOqOXPkqaxU/Jln6pL77gu47/Bp03TJvffqvJ/+VM7ISEmSIzz8xAUMQ5Hx8e0qm2maOrp/v0zT1Pm33qr+552nmtJSrZo7V7vffjugjA3rHQAAAI3R0w0gJDRcVE0K7HluNPdbxxcvu/RSlezd65+LbD0vKTVVP1i+POA+l/z2t1r/5JMade21qsjPV2VhYaNrd4SWesqzMzPl2rlT/c87T9MXLmwyuK/Iz1dUnz5yRkbKERamo/v3131gmqo5erTN5XJGRUmSvMfntH/2pz9p8mOPqWT/fh09cEDrn3xSCUOHKjszU5WFhcp5801JLK4GAD1ZS0OvpcbDl4ONwdreQjpYb6D1utayWFdND6Y15W/pOsGWMd1nSVtXQN8bJE/x3wPTN8VYTvirda11SX8ITN7wTmD6I2tBJFlGurdqiD3sQ9ANICRkZ2Zq85IlOrBmjfqNHq2cN+r2ELnk3nubDGAbLl6WlJraqiHhw6dNCxi67srJUUxycocvINbS/PaGZW2uN31AeroShg1TbWWlinbu9B93hIcrsk+fgBXJT3zYYHE4C291tcJ79ZLX6ZR8PlWXlOidefN02vjxioiNVezAgdq8aJFy3nxTqTNm6LxZs1hcDQAAoBkE3QBCQlpGhg6sWaOCbdtUW1Gh1Bkz/MFeUwFsWxduC6Y917BuddaR97Ve+0hWllxffSWf16uwyEh5a2rkjIxUVJ8+ih86VIdLShpvf9ZEwF3PbVkN3X3smL755BOdfvHF+ubjjwOC7YbP1xHPDQAA0NMQdAMICUmpqZq+cKFWzZkj186dirnyyjYHdsGGodvBzvtYr52WkaFKl0syTSWPHq3Ply1TWHS0PFVVOmvaNJXs2aPKoiI5nE75PJ7W36h+T+/6/5qmTNNU6owZGjZlStC56J1VvwAAe1lfzQYb9t3SkO0gC5F3mdYMQe+I61qHoAdbWX2FJX3Gs4HpSRcFGcg+c2hg+leB48mn/6xxFusg9aNByoLOQ9ANoNuy9pzWB95N7YXdkpNdcbw73Md67aTUVE196in/5xX5+Vr36KPy1tTIXVUlT22twiIjZYSHy1de3vobHR+CHhYVJU9Vlby1tTq4bp3iBg3Sqrlz/b3hl9x7r//7VL+6OcPNAQAATrB99fKFCxdqyJAhioqK0oQJE7Rx48Ymz/3b3/6mSZMmKSEhQQkJCZo8eXKz5wPo2ep7TrMzM/3HTmbV8JNdcbw73Kela6dlZCgqIUE+j0fFu3frzO98R+m33aaw4yucN8nhqOvVruf1SqYpT01N3WeSIuLilL9tm44eOKDwXr0CFrXbunSpjmRldUr9AgAAhBJbg+4VK1Zo3rx5uv/++7VlyxaNHTtWU6ZMUUFBQdDz16xZo+uvv14ffvih1q9fr8GDB+uKK65Qbm6uncUE0E2lZWSwUFc7nPmd7ygiLk6eqirVHjumcbNnK+Wcc+q2FmsYWDfk89UNI2/qM0lVRUUKi4xU/JAhuvjuuwP2LOf71DPwohwAgI5nmGZTf2WdvAkTJuiCCy7QM888I0ny+XwaPHiw7rjjDt19990t5vd6vUpISNAzzzyjm2++uVX3LCsrU58+fVRaWqq4uLiTKn9n8Pl8KigoUL9+/eRwsG16R6Ju7UG92qcj6nbtggXaunSpTp80SeWHD2vi/Pk6kpWl9U8+KZ/Pp9iBA1WUk9N0gG1VP6f7uPDevRXVp4/OvPxylefmauL8+QGrvXdX/Ny2bMWKFbr55pu1ZMkSTZgwQU899ZQyMzOVk5Ojfv36NTr/xhtv1Le+9S1ddNFFioqK0mOPPabXXntNO3bs0KBBg1p1z/o2O0ZSE6+DAHRTLf0mtWsedahLtKRvsqR/PzZIptWWtLXyBzTOcpXlG2Cd4x1sSze0nam6umwp9rRtTndtba2ysrJ0zz33+I85HA5NnjxZ69evb9U1Kisr5Xa7lZho/fEEcKphZezWCbbVWMLQoSratUvFe/Yo8ayzdCwvT56qKoVFR7e8l7clOPdUV8uIj9fuVatUVVSk2ooKf9DN9yi0/fGPf9Stt96qW265RZK0ZMkSvfXWW3r++eeDvih/6aWXAtLPPfecXn31Vb3//vutflEOAMCpwLag2+Vyyev1KiUlJeB4SkqKdjbYS7Y5d911lwYOHKjJkyc3eU5NTY1qak6s8ldWViaprlfD18K2ON2Bz+eTaZohUdZQQ93ao6vqdUdmpj5ftkympEm//nWn3ruzdETdJg4frouP10/9dRKHD1fiiBHa/fbbKvzqKw1IT1dVcbHyt23zz9duLUd4uCqKiupWQnc61bt/f/99gn2PinbtUvbKlUq75hr1HTGi3c91skLt90Fn98bzohwAAPt029XLH330Ub3yyitas2aNoqKimjzvkUce0YMPPtjoeGFhoaqru9NmBcH5fD6VlpbKNE2GPHYw6tYeXVWvA6ZOlTs2VgO+9a0m14UIdXbWbex55ylqxAh53W6V1tbKFxGhuPPOa/uFDEOGYSgsOlremho5Bg3SB08/rdO/9a2g36Pt77yjfZ99JndsrEbHx3foM7VFqP0+6N+/f6fer6tflAMIPaHxCrP7se4jss2S3vdF4zxDt1gOTO8bmL60qFGe4R8Gpq2XDRYl8T21j21Bd1JSkpxOp/Lz8wOO5+fnt/jHxBNPPKFHH31U7733ns4555xmz73nnns0b948f7qsrEyDBw9WcnJyyMzpNgxDycnJIfGHYCihbu3RVfXar18/JfTureyVK5XQxb2mwXREj25H1W3Rrl3a/Ne/SqapcT/7mfqOGKGcrVtVvWuXPDU18tTWynS7W3/BsDCp4R7fDodSxoxR+b59yj18WId8PoWXl2vSr3+ts84/PyDr2ClTFF5errQpU9Q3yLzgzsLvA3ud7ItyAAB6MtuC7oiICKWnp+v999/X1VdfLanuj573339fc+fObTLf448/rt/97nd65513NG7cuBbvExkZqcggW+E4HI6Q+cPKMIyQKm8ooW7t0VX1+tXKlfp86VIZqtsfujvpqLJ1RN1+tXKlvnzhBUlSr6QkXXLvvRqVkSFD0u5Vq3SowXDh8N69FXfaaSpqpjfTYZoypbqh6Kap5LQ0pYwZo7DISBXv3q3IPn3UOyVF637/+0bzuZNHjtSl3eR7xe+DpnX1i3IAAHoyW4eXz5s3TzNnztS4ceM0fvx4PfXUU6qoqPAv0nLzzTdr0KBBeuSRRyRJjz32mO677z69/PLLGjJkiPLy8iRJvXv3Vu/eve0sKoAQ0HCRsO6ifvGwAenpkrpH2dIyMlS0a5eKdu/2l6veWdOm6fDmzfId7+kOi4hQ8a5dTV4rPDZWMk15a2pk+nwyfT7VHjumb9au1emXXKKKggLVlJZq3aOPynt82HB3eyGClnX1i3IAOFV4LenDlvTnQfIMPWg90icwObzx8PJ4y/DycMvnziD3YXi5fWwNuq+99loVFhbqvvvuU15ens4991ytXr3aP2fsm2++CehxWLx4sWpra3XNNdcEXOf+++/XAw88YGdRAYSApNTUDgnoOnKV7ezMTG1dulRS9wk2k1JT1XfECH3z8cfau3q1jmRlqdLlUs4bbyhh2DA5o6Jk+nwKi4lRVUnJiRXKLduDSZK7oqJun26HQ/L55IyK0sV3362K/HxVulzy1tQosk8fuY8dU78xY7rFSwe0Dy/KAQCwh+0Lqc2dO7fJt+Rr1qwJSB84cMDu4gBAhwbK3bH3XTpRnkqXS1uXLlXq97+v82bN0oD0dP3fjTfK7fXKfexY0EA7QP1q38f/6/N4VLhjh2SaGjZ1qmKSkjQgPV17V6+uuxZCFi/KAQCwR7ddvRwA7NKRgXJH9b53tPpyuXJyFJOU5H/W7MxMnfPjH+uLF16Q1+2Wp6rqRKbmgu/6U7xebV22TA6HQzHJyUrLyFB2ZqZkGMp54w3FHJ9DjtDEi3IA6FzW1czzg52023rAsmBlWuMsTS9pia5A0A3glNNdA2U7NHzWtQsWaOvSpUoYNkwRsbGqyA/atLcoKi5O/c8/X5UulzYvWqTsV19V3ODBSp0xo9v1+AMAAHQ1lnAF0GlcOTlau2CBXDk5XV2ULtXZ9eDKydHqn/9cuZs2qfeAARp17bXqc/rpMlvRs+1nGDLCw+UMD1dlUZFKv/5aX7z4oqpKShQ3eLB/yHlSairfZwAAgAYIugF0mvq51NmZmV1dlC7V2fWQnZmpL158Ufv+/W8VbN+uHStWKPGss+RwOuvmYTsckjPYOqYNmKZMt1ve2lp5qqtVXVYmn9erot271Xf48BPXEt9nAACAhhheDsA21lXCu+uiY52tM+vBlZOjysJCjbjqKklS8Z49Kti+XZUul+KHDlWly6Wq4mLJa93EJDgjLEymxyNDUmSfPirYtk19hw/XxPnz/c8zID1dB9asabRdGQAACGTtAbVu7SUpyP5elh0ikhtnCXqdZu4Le1HfAGxj7fGsn198stt0hbrOrIfszEzlvPmm+o4YoR8sX64Zy5ap3+jRdft45+SoqqhIrV1z3HA65YyMlCM8XJFxcaopK5PP41Hxnj0B268dycpSyd69OpKVZd+DAQAAhAiCbgC2ScvI0HmzZp3yPdtdaUB6unr176/Dmzdr9Z13SpImzp8vw+GoW63cNGXWbwvWAtPrlaeiQmHR0bpgzhyNnDFDMcnJKtm7N2AoOd93AACAExheDsA2p9Iq4d1N/dD+ol27lP/558ozTYVHR0uSCnbsqAu0HQ45w8Lkra1t07XdlZXa+vzz6jt8uGrKytRv1KiAAJvvOwAArdPLko4NdlKjgXEjA5NfbGyUZZ8lXW1Jt+51OzoKQTcA9ED1Q/t79e8vR3i4EoYOVcrYscrduFFFu3fLERYmh8ejiN696+Z0t4bDIUN1Pd5HsrJUVVIiX22tEs86K+hQeeucfgAAgFMRQTcA9EANFzXbu3p13cripqmSfftk+nyqLS+XpNYH3JLk89Wtcm6aMk1TNaWlckREKDoxMejp9YG/JHq+AQDAKYugGwB6oIZDvI9kZWnr0qU6/ZJLFN67tzzFxXJGRspbU9O6ix0P2KW6Xm5JCo+Jka+2VjFJSRo2ZUrQXm1WqwcAIJB1VfEoSzrouLCbrAf6BCZXNc5y0JJ2W9Kt27MEHYWgGwB6uPqgt9LlkruiQvL55GvlFmGS/AG3JDnCwzVw3DidNW2aPn/hBbkrKnQkK8sf2EsnerWZ2w0AAEDQDQA9UsOe55J9+3RgzRqdcemlCu/VS9VHj/p7rFvUoJdbknqlpGj49Okq2r1bFfn5ik5M1ID0dCUMHSqJXm0AAAArgm4A6IHq51NXulzK+ec/dfTAAR3etElet7tRIN2shucZhmL69tXWpUvVu39/mV6vqktLdSQrS8OnTaNXGwCAFjgt6YGW9LnW8eeSFGHJVfF0QDJ3e+Ms5Za0dXg5q5d3LoJuAAhxweZTD0hP14E1a1Ry4IAq8vMV2bu3PNXVdUF3awPuBhzh4Rp93XUaff31OpKVFbBAG73bAAAATSPoBoAQ13CV8LSMDGVnZqrS5VLJ3r2K7NNHPrdbUfHxcldV1QXcbenpPm7c7bfrgtmzlZ2ZKdM0tWruXF18991Kv/VWOx4JAACgxyDoBoAQl5aRoUqXS5WFhdq8eLFy3nhDqd//vlK//33t+fe/5fV4VJ6beyJDO3q6P3/hBW1dulSmacoRFqbasjK9d9ddijvtNB3JymIvbgAALIKNFI+1pC9o6QRJes6yDsungcmVQbLkWtKsVt61HF1dAADAyUlKTVVMUpJy3nxTMk2dN2uWxs2eLRmGSvbulUxThtMpR0SEJMkRFub/d2vVlpXJXVEhT1WVBo4bp6iEBBkOh9Y/+aS2Ll2q7MxMOx4NAAAg5NHTDQA9QP286gHp6TqSlVV30DRlGIYcYWEakJ6uQePHa8uzz8oZFaWI3r1VfuhQs9cMj42Vu/z4UizHh6RH9+2rKxctklQ3rL3+fszrBgAACI6gGwB6gPo9sdcuWOCf313f2y3TVPLo0Xrvrrvkqa6Wp6ZG4b16yXA4ZPqCrF9qGHKGh2vQBRfo0Pr18lRVyREWpl4pKTrzsssC7idJw6dN67TnBAAACDUE3QDQg9T3ONf/NyYpSaZpavV//Zc8NTX+IPxY/RzvIIuqGU6nks4+W1XFxfJUV8twOpWclqbw6Gjtf/999R0+nO3BAABoQbB5vNYp22WWdHZx4zxDLWuWWqZ06+0g9ym0pNkirGsxpxsAegjr1mHZmZnavGSJ1v3+9/JUV8sZEaF+Y8bUzec2jLpMpikjvG6pF2f9cZ9PJfv3K/6MMxQ7aJBGfO97SjnnHBXs2KGY5GQV7d6t1XfeKVdOThc+LQAAQGgg6AaAHqJ+67D6Rc0GpKfLGRkpR3i4ohISNGzKFFUVFSn57LOVMnas4s88U47wcEX26iU5HPJ5vQqPiak7FhenyLg4OcPCNHDcOEmS7/ge37vefFNfLF/O4mkAAACtwPByAOghGg4td+Xk6L177lHZwYNKTkvTNStWSJJWzZmj3I0bJUnJo0apd0qKSvbtU+zAgarIz5e7ulrOsDB5KisVnZio82bNUlpGhjYvWqSwqCiljB2rId/+tmSaLJ4GAEAD1t5MZ5BzrMO891nSf2/FdfdY0sHGndUEOYauQ9ANACGu4bDyS+69V66cHK2aM0eu7Gz53G5VFZ+YINZv9GiF9+6t3A0blP/FF4o/4wzFDx2q4dOna9PChaooKJC3pkaRgwZp3O23+/feHjd7tmKSk9mPGwAAoI0IugEgCOv86O6sfli5JF1y773KzsyUa+dOJaWlqbqoSN6aGv9Q8OyVK+WMjFRtebl8brdK9u9XeHS0Inr1kruyUo6wMPVOSdH0Z55RUmpqo4AeAAAAbUPQDQBBWAPZ7sy6YvmA9HQljRypifPnK2HoUH/QLEkH1qxRwbZtShw+XDJNxQ8ZovgzztCwqVO1d/VqyTACerhDqR4AAOhOvEGOHbWkrUPFD7biuoct6aLWFghdhqAbAIKwBrLdWcM9syXpSFaWSvbu1ZGsLA2fNi3gs4nz52v9k08qduBAffPxx4o/4wzFJCcrYehQTf3znxtdO5TqAQAAoDti9XIACKI+kO3uQ8uDGZCeroRhw9QrJUVrFywI2NqrPiCvXyStqqRE6598UpsXLw56rVCuBwAAgO6Anm4A6GHqA+sdK1aoZO9eSSeGhjfsuU5KTdXqO++sy2SaXVJWAAB6KneQYxWWdJ4lHWzFc6ujlrR1RXR0PwTdABBiWlrkreHc7iNZWQFDw61D0RuuSt6WewAAAKB1CLoBIMQEW9zMGiTXHx8+bVqz17IG4c3dAwAAAG1H0A0AISbY4mYnGyRbg3YWUAMAAOgYBN0AEGKC9U6fbJBsDdqb6gEHAADBtWZutXWet3WOd7A53dY8weaKo3sj6AaAHuBkg2R6tgEAAOzBlmEA0AO5cnIabRfWHLYGAwAAsAdBNwD0QPXDxbMzM7u6KAAA4Dif5as1vJYvhB6GlwNAD8RwcQAAgO6Bnm4A6IHqVyDPzswMOsS8rcPPAQAA0D70dANAD1U/xLyysFAxycn+7cAafiaxDzcAAJ2lpR7PYCuTt3YYOrovgm4ACGHW/bUbqh9aXulyNQqwGX4OAADQOQi6ASCENddjXb8iuSsnRzFJSQEBNvtwAwAAdA6CbgAIYa3psSbABgCge7AOFa/pklKgsxF0A0AII6AGAADo3li9HAAAAAAAmxB0AwAAAABgE4JuAAAAAABsQtANAAAAAIBNCLoBAAAAALAJQTcAAAAAADYh6AYAAAAAwCYE3QAAAAAA2ISgGwAAAAAAmxB0AwAAAABgE9uD7oULF2rIkCGKiorShAkTtHHjxmbPz8zM1MiRIxUVFaUxY8Zo1apVdhcRAACINhsAADvYGnSvWLFC8+bN0/33368tW7Zo7NixmjJligoKCoKe/+mnn+r666/XrFmztHXrVl199dW6+uqrtX37djuLCQDAKY82GwAAeximaZp2XXzChAm64IIL9Mwzz0iSfD6fBg8erDvuuEN33313o/OvvfZaVVRU6F//+pf/2IUXXqhzzz1XS5YsadU9y8rK1KdPH5WWliouLq5jHsRGPp9PBQUF6tevnxwORvt3JOrWHtSrfahb+1C3LevKNjtGktEhTwEAQOcxJVVKLcaeYXYVoLa2VllZWbrnnnv8xxwOhyZPnqz169cHzbN+/XrNmzcv4NiUKVP0+uuvN3mfmpoa1dTU+NNlZWWS6v5Y8Pl8J/EEncPn88k0zZAoa6ihbu1BvdqHurVPqNVtZ78Y6Oo2GwCAnsy2oNvlcsnr9SolJSXgeEpKinbu3Bk0T15eXtDz8/LymrzPI488ogcffLDR8cLCQlVXV7ej5J3L5/OptLRUpmnS+9LBqFt7UK/2oW7tE2p1279//069X1e32QAA9GS2Bd2d5Z577gl4015WVqbBgwcrOTk5ZIaXG4ah5OTkkPhDMJRQt/agXu1D3dqHuu0emmqzAQDoyWwLupOSkuR0OpWfnx9wPD8/v8k3+P3792/T+ZIUGRmpyMjIRscdDkfI/GFlGEZIlTeUULf2oF7tQ93ah7ptWle32QAA9GS2/eURERGh9PR0vf/++/5jPp9P77//viZOnBg0z8SJEwPOl6R33323yfMBAMDJo80GAMA+tg4vnzdvnmbOnKlx48Zp/Pjxeuqpp1RRUaFbbrlFknTzzTdr0KBBeuSRRyRJd955py699FI9+eSTuvLKK/XKK69o8+bNevbZZ+0sJgAApzzabAAA7GFr0H3ttdeqsLBQ9913n/Ly8nTuuedq9erV/oVXvvnmm4BhfhdddJFefvll3Xvvvfr1r3+t4cOH6/XXX9fo0aPtLCYAAKc82mwAAOxh6z7dXYF9ulGPurUH9Wof6tY+1G33xD7dAIBQ1tp9uvnLAwAAAAAAmxB0AwAAAABgE4JuAAAAAABsQtANAAAAAIBNCLoBAAAAALAJQTcAAAAAADYh6D4FuHJytHbBArlyck7qHAAAYD9HkK+OuE645SuyFV8dUQ4AONXx+/MUkJ2Zqa1Llyo7M/OkzgEAAAAAtE1YVxcA9kvLyAj4b3vPaYorJ0fZmZlKy8hQUmpq+wrZgdcBAAAAgO6CoPsUkJSaqkvuvfekz5FOBMYD0tN1JCtLaRkZ/l5ySUGv0dpguqXrAABwKvC1I0+woYvhlrSzFdfxtnCNYGWz5mlP+QGgJyPoRpvUB8YH1qxRyd69klruJW9tMH0yve0AAAAA0B0RdKNN6gPiAenp2vvOO6osLJTUMcF0a3vbAQAAACBUEHSjTRoGxkeysrR16VLFJCc3GywTTAMAAAA4VRF0o90YDg4AQPdlnVttTQebBx5hSVvndAdTYUm7W7gvAJxq2DIM7Vbfgx1scbSW9v1mX3AAAAAApwKCbnQ4V06OVs2Zo81LljS573d2ZqY2L1miVXPmEHgDAAAA6LEIutHhsjMz5dq5U0kjRzY59DwtI0NJI0fKtXNno8CcXnAAANrGF+TLa/lyWL56BfmKtXwlWr6sn8dKirJ8OS1fAHCqY043OlzDud5N7cudlJqqifPn66OHHlLRrl3a/fbbrd73GwAAAABCBUE32sWVk6PszMyggbV1tfKmzj2SlaWinTtVmJ2tbz75RO5jx3RgzRpNnD9fUtMLtDV3bwAAAADoTgi60S7W3ujmAuGmeq7TMjJUWVio3E2bVLJ3r5wREcrduFF7V6/W1D//OeAaDa9PTzgAAG1nXZk8Mcg5sS3ksa5ULknVLaQB4FRH0I12sW4XtnnxYn3x4ouqdLk09amnJJ0IlAekpwecWy8pNVXjZs+WFi3SoPHjVVVSol1vvikZhv+c+mtUulzKeeMN/3UqCwtV6XLJlZNDbzcAAACAbouF1NAujbYLM01JUlVJiX8RtPoe6fq52psXLdLqn/88YIG07MxM5bz5pmSaik5I0IirrlJVcbFW33lnwDVkmjpv1ix/T3pMcrJy3nijydXRAQAAAKA7oKcbHWLc7NmKSU5W0a5dWv/kk6osLNSwqVOVvXKlsl99VUW7dmnXP/8pn8+nQxs26LQJEzTu9tv9vd/1Pdm9+/dX3uefy3A6VbBjR8D87oY92taedgAA0LwoS3pIkHOGWtLFlvTBIHms5wAAAhF046Q1nG/98e9+J091tQq2b1fOv/6lsoMH5fN4dCwvT33OOEMyDBXu2KGinTslSTFJSf7AOSYpSUW7d6tgxw6Fx8Q0Ob9barxYGwAAAAB0RwTdOCmunBytmjNHruNBdL28L79UlculsJgYRfftq9rycpVWV2vEVVfJU12tSpdLe1avVu2xY5LqFkSrX5Ct7/DhKtq92z+/m9XKAQAAAIQqgm6clM2LFil340YlDB2qot27lf/llzKcTkmS4XTKMAwN++53VbR7t/oOH67ohASVHjggT3W1asvLdfrFFwcMEa/vwa4PvlmtHACA9gm3pK2rlY8Nkud8S3qfJR1s9fIDlrSv+WIBwCmHoBsnxzDkOB5kZ//v/8pwOtUrJUU1paXqPWCABqanq2j3bpUdPKgRV16ptIwM5W/frtwNG9Rv9Gj1Gz066GWTUlP9AXdTq58DAAAAQHdH0I12qR/yPWzKFMUkJSl30ya5vvpKiampmvzII1r/5JNy7dyp2mPHVLx7t7xutw5v3qy0jAx5KivlrqzU0a+/VtHxlcyt24zRww0AAACgJyDoRrtYA+LVP/+5HOHhKs/N1Tu/+IViBw1Sn8GDdcall+rIli2qLSvTrrfeUvyQIUocMUJHtmyRz+ORz+uVTPPEftyFhXVbiIkVyk9FzN8HAABAT0PQjXYZkJ6uA2vWqFdKitYuWKBhU6Yo55//1NF9+1Tlcqlk716FRUerLDdXnupqOaOi5KutVcn+/fru44+rPDdXh7OyJMNQyYED2rxokXLefFOpM2b49+PGqYfRDQDQPo4gx6xbhFlfZV4TJM8YS6a/VwemnUHyWOeOe4OcAwCnMoJutMuRrCyV7N2rHStWqGTvXlUWFqrf6NGqLCyUt6ZGkfHxks+nssOHZZimHBERMn0+lX79tZJSUzV94UL9z3e/q7KDB7X7rbc0/MorlTBsmJJHjVJFfr4kArBTEaMbAAAA0NMQdKPNXDk5Ktq1S5Hx8Qrv3VunT5qk3I0bVZCdLXdlpUyvV5UFBZLDIUOSKclbUyPD4ZCnulqvXH21juXnK3bgQJUdOiQZhioLClR68KDWPfqovDU1kgjATkXsvw4AAICehqAbbZadmald//ynPNXVKj1wQH1HjlTBtm2K6N1bkXFxKs/NlQxDhmHUBd6mKdPjkQyjbvG0Xbsk01Sv/v0VHhOj6IQEnTV9uj5/4QXVHD2q/ued55/TSwAGAEDLgg37Trakv2tJj5kcJFN8YDJtZWDakpQk1TRXMABA0ClAQABXTo7WLlgg1/GVxtMyMjTiqqvUNzVVI666SuExMXJXVammtFSSFB4To74jRsjhdMp0u+sCbkmmzydvba1kmpKkqD59lHLOOfLW1irr2WdVkZ+vxBEjNHH+fGVnZvrvBwAAAAChip5utKjh3Oq0jAxtXrxYRbt361hensKio3X0wAHJ55PhdKq6pESJZ50lU5J5PLgO4PNJkhxhYTq6f7/6pqbKNE0dycqS4XCo7/Dh/u3GJOZyAwAAAAhtBN1oUVpGhioLC1XpcmnzokX6Yvly+bxexSQnq2DbNnlraxWdlCRJqi0vlysnR97q6mav6TNNqbZWeVu3asi3v63CHTvUb/RoRSckaP8HHyhp5EjmcgMAcBL6W9I/sJ7wH0EybQtMui0fW1cql6TmW3wAAEE3WpSUmqqY5GRtXbpUqTNmaOzMmZJpKnn0aL3/61/LXVWl6qNH/fO2FayHuwFnVFRdUG4YOpaXp9xNm5Q8apQu+e1vlTB0qGKSk/1zutm3GQAAAEAoI+hGq9T3Og9IT9eRrCylZWQoOzNTtWVlks8n8/iw8ZYCbknyVlfLcDpler0yvV4V79kjQ9L2f/xDP1i+PGBIOduGAQAAAAhlBN1okSsnR5sXLZIMQ3vfeUc5b7whqS4QL9q1Szn//Kdqjh4NyFMfVDfF9HoVFhOj0y68UIXZ2arIz1fxnj2NzmPbMAAAWhZsZVzr6uXx51sOXDu0cabYfQHJPMvHxUHu03RrDwCQWL0crZCdmakvli/XFy++KJmmzps1S71SUrRqzhwlDh8uR1jjdzfNBdz1PFVVqioq0nceekhnXnaZzr3lloBV0qUT+zYztBwAAABAKKKnGy2qX0hNhqFxt9+upNRULZ04UYc++0wHPvzwxNDyVnJGRspbUyOZplxffaWK/Hzd/N57WrtggbYuXarKwsKAed0AAAAAEKoIutGipNRUTf3zn/1pV06O3MdXJ29rwC1J3tpaOcLDFd23r4Z997uNhpBXulxtmsfNYmsAAAAAuiuCbrRa/dzu3E2bdOzIEUUnJanK5Wr9BQxDkhTVp49GXHWVJv3mNwFBcv1QcldOjmKSklo9j5vF1gAAp7qIIMesW4bpO9YDv2mcaeisgGSM5ePyNpUKACARdKMN6ud2e91uRSUkKDwmRlXFxVKw3m6Ho/Fx05QzMlKGw6G+I0ZIktYuWNCoh7o++G4tFlsDAAAA0F0RdKPV6ud2527apILt2+WtrW16i7Amhp17a2rkCQ/X4c2bteutt1R28KCkk+uhbmuQDgAAAACdhaAbTaqfK91wb+6pf/6zf5h5/vbt+mbdOvlqa1u8liMsTM7oaLkrKuSurNSe1avljIjQoPHjNSA9PWiPN4JjDjsAwKo1W4ZpgvXA9MaZLOPJnZaPg+1NwpZhANA8gm40qX6u9IE1a1Syd6+kut7u7MxMJY8eraxnn21VwC1JPo9HvvJyhUVFyeN2Kzw6WnFnnKFR116r9U8+KdfOnZKYk90azGEHAAAAQgdBNwI07EWtnyPdsKfbH/A5HPIcX8FcUt0iaceHmjvCw+Vzu4Nevz5P9dGj8rrd2rFihVw7dypp5Eh6vFuJOewAAABA6CDoRoDszExtXrJEB9as0fSFC/2Bdn0gnJaRoaLdu5W7cWPgYmkN5nY3FXBbeaqqlDxqlIZ8+9sBAX3Rrl0qP3xYE+fP1/Bp0+x4zJDGHHYAQGs0GnLeaLx5fONMNYHJ1gwdtw5Bb91fAQBw6iDoRoC0jAwdWLNGrp07lZ2ZKUkBQ5mTUlNVnpur4t27FX58jnarHd8yrD5AN30+ffV//6d5xxdTq++53b1qVV1QLxF0AwAAAAhpBN0IkJSaqukLF/p7t+s1/PfE+fNVtGuXf+XxVjNNRSclyV1ZKU9VlWSaijvttIAh5Zfce68GpKdr/ZNPauL8+R31WAAAAADQJQi60Yh1+HL9v105Odq8eLFkmkocPrzNQbfhdEo+n/qcfrqKcnIkw1D10aMBw9mTUlM1fNq0NvVws5o3AOBUF2yjzkrrgXLrgdzGmfoEJntZPrYOJUcd61D+4BunAjhVBdthAggqOzNTX7z4or5Yvlz5X37Z9gsYhhJHjKj7t2lKpqlKl0sRsbH65uOP9fHvf9/ucm1dutQ/HB4AAAAAugvbgu7i4mLdeOONiouLU3x8vGbNmqVjx441e/4dd9yh1NRURUdH6/TTT9d//dd/qbS01K4iohVcOTlau2CBXDk5GpCerqSzz9aIq65STFLSiZPq52o3wxkZKdPjUcG2bapt8HNQW16usoMH5a2tVf4XX7SrjAPS05UwbJgGpKe3Kz8AnOposwEAsI9tw8tvvPFGHTlyRO+++67cbrduueUW3XbbbXr55ZeDnn/48GEdPnxYTzzxhNLS0vT111/rZz/7mQ4fPqyVK1faVUy0oOGe0JJ07MgRDZ8+XaOvv16v3XyzqoqKFB4T0/yCaoYhn7du/VN3RUXduce3GDMcDpler4ywMKWcc44/S1uGjB/JylLJ3r06kpXVYxdeYwg9ADvRZoe+2iDHGk0CW2dJX7W9caZ+IwKSZ2hXQDomyH0Yct4Yw80BNGRL0P3VV19p9erV2rRpk8aNGydJevrppzV9+nQ98cQTGjhwYKM8o0eP1quvvupPDxs2TL/73e900003yePxKCyM6eddIdie0GkZGSrZt8+/GFqLK5ibpgzDkCnJCAuT6fH4VzD3ut0aOG6cTpswQeNuv92fpWGw39L2WKfCvtVtqQ8AaAvabAAA7GVLq7h+/XrFx8f7G29Jmjx5shwOhzZs2KAf/OAHrbpOaWmp4uLiaLy7ULBF1Vw5OXrjpz9tHGwf770Oxud2K7xXL8kw5G4wZDEmKUmX3ndfox7qtgTSp8K+1afCiwUAXaMz2+yamhrV1JzYCLqsrKz9BQcAIETYEs3m5eWpX79+gTcKC1NiYqLy8vJadQ2Xy6WHH35Yt912W7PnNdWA+3w++XzdfzCPz+eTaZohUVZJKtq1SytvuEEVBQWSo/GSAM6oKHkbfD8acldV1f3D4ZAMQ4bTKU9Njdb/8Y+KP/NMSVL2ypVKu+Ya9R0xQhf/+teS1O66CbW6bU7i8OGtqo+iXbsC6tAOPaleuxvq1j6hVreOIL9f7dKZbfYjjzyiBx98sN1lBQAgFLUp6L777rv12GOPNXvOV199dVIFkuoC5yuvvFJpaWl64IEHmj23qQa8sLBQ1dXVJ10Wu/l8PpWWlso0zU79I6u9tr/zjmoiIxV3MouWGYYMSaZpyhkZqQrD0LtPPKE+p5+u3I0b5Y6N1ej4+JMua6jVbUfY/s472vfZZx1Wh8GcivXaWahb+4Ra3fbv3/+kr9Ed2+x77rlH8+bNC8g7ePDgky4Dgs8ZbvTK5DVL+rF3guSaEpAaNDBwTvfAw41z7LOkrXO8Q+NVV+sF+w1ifWZvZxQEQMhoU9A9f/58/eQnP2n2nKFDh6p///4qKCgIOO7xeFRcXNziHxLl5eWaOnWqYmNj9dprryk8PLzZ85tqwJOTkxUXF9f8A3UDPp9PhmEoOTm52/8hWLRrl9z79yusrEzFu3fL9Pnq5ms7nXJGRMhTUyO1ohcpPDa2boi5aSp20CCZkZE6vG2bEq67TqkXXqi0KVOko0dPusc2lOq2o4ydMkXh5eVKmzJFfS09Vx3lVKzXzkLd2udUrNvu2GZHRkYqMjKyVeUHAKCnaFPQnZycrOTk5BbPmzhxoo4ePaqsrCylH+8R/eCDD+Tz+TRhwoQm85WVlWnKlCmKjIzUm2++qaioqBbv1VQD7nA4QuYPK8MwQqK8X61cqV1vvKHU739fVenpOrRxo6qKilTlcsnjdp84sZm53c6ICLnLyvyfVxcXKywqSmHR0aouLtZZV1yhr1auVGVhoXLefFOGTm7hsFCp246SPHKkLu3A+e1NrZp+qtVrZ6Ju7XOq1W13bLMBADgV2TKn++yzz9bUqVN16623asmSJXK73Zo7d66uu+46/yqoubm5uvzyy7V8+XKNHz9eZWVluuKKK1RZWam///3vKisr88/PTk5OltPJhhRdreFiXpsXLVL5oUMyTbNujnbDHu4mAm5J8tYGbmrirqhQdEKCyvPytO2ll5T3xReqLS1V6owZOm/WLBYO62Ksmg70fLTZPUOw4cxfW9Krdgemp2ctbpwp/c+B6ZcCkzO/0zjLHku6NZP73C2f0m1YX9O156c72Ku+njbsHkDTbFsW/KWXXtLcuXN1+eWXy+Fw6Ic//KH+8pe/+D93u93KyclRZWWlJGnLli3asGGDJOmss84KuNb+/fs1ZMgQu4qKVgpYJdww5Kmulun1Sg6HHGFh8nk8rb+YYcjhdMoRHq6yvDzpeN4ql0vOyEgljxql9FtvteEp0Basmg6cGmizAQCwj21Bd2Jiol5++eUmPx8yZEhdL+lx3/72twPS6N7G3X67Dnz0kQp37FBknz6qPnq0bRc43kPuqV/R/Ljqo0flqa7W58uWqSI/v9Gw5p6oqSHcXXWdhk6F7dgA0GYDAGAnNsBGuySlpuqaV17R5sWLtWf1alW5XG2+hs8y1FySPFVViklOVuJZZzU5rNmO4LIrddQQboaCA8CpK9hQ5VxLepklPf2KIJmK/hqY/vYdAckpP3q6UZZN/xuY/rvl89Igt7GyDo/vyqHX1qHg1uUBO2vyBMPPgZ6DoBsnJXfDBpXs2yfD6ZQMo264eXt6P44vvmb6fEoYOlSjr79e5YcPa8DxRX0aBto9LbjsqCHcDAUHAAAAuh+CbrRbdmamCrKzJUmxAwbokvvu09bnn9fhzZtltmV+t3QiUHc4dNa0aVr70EPK37ZNax96SAlDhwYE2j0tuExKTfW/TDiZ3nuGggMAAADdD0E32i0tI0OVLpcObdigsoMHVZGfr0vvu0+vzJihVvd1W7cX8/m07aWXdHT/fvm8XhXs2KHNixdLpqnUGTP8QWlLwaUrJ0c7MjM1YOpU9bNpv+qOdDK99z1tuD0AoGPUWNIbLOk7ihvneTojO/BA5pDA9IrGY9Lv3fPvgHTplsDPPw5StkJLutKSbjwBrfFw62ArtltZh4Jbh44H2+jOesw6vDzYsG/rauzW8rdmtfbWPA+A0ETQjXZLSk3V1KeeajT0u02CDEUv2bdPjrAwxSQmath3vyuZprJffVVJI0e2+rLZmZn6fNkyuWNjddb557etTC2wY+Gzk+m972nD7QEAAICehKAbJ61hz3NaRoaKdu3S9hUrTiyU5nDIqJ/v3Qo+d9374EqXS3vffVffeeghxW3cqNyNG/Xx736nviNGaEB6uo5kZWlAerr2rl4tGYbG3X67PwhOy8iQKWnAt76lol279NXKlR3WE9zeILdhkC1Jq+bMkWvnTv912tPDvXnxYlUVF/tHAQAAAADoXgi60aGSUlP1g+XLJcPQjldeUURcnGrLyuRtZcBdrz7wrsjL0/v33CNHeLh8Ho8OrFmjvf/+t8J79ZK3pkZJI0cqd+NG+dxuVZWUqO/w4f6APO2aa+SNj1f2c88pa8kSHVizRtMXLjzpwLu9vdINg3VJcu3cqaSRI/3XaaoHPdhxV06OVs2Zo9xNm+RwODRx/nyGlgMAAADdEEE3bDHp17/2B8Db//EPZa9c2WhP7taqKiqSJIVFRammrEyGwyGf2624wYNVWVSk2IEDVX7kiIp37dI3a9dq91tvybVzpyqKinTeXXcp7Zpr9PWaNXLt3KnszMyAHmVXTo42L1rUqKe8Oc3NKbcGyA2vnzxqlBKGDVOvlBR98/HHihs8WKOuvVbZmZnqlZKitQ8/rEqXS7tXrdKMZXUbu2RnZqqysFA5b74pqa5HvD7gLti2Tf1GjdKgCy6glxsA0CLrFO5/BTknZWVg+t67VgUeeGx640xZIwKSf7hwV0D6Y+tkckmvWdI5lrR1zrckVVjS1tf58UHyDLSkx1jSFwXJM9iSPmBJW6asS5LWW9IHLemjQfJYy8+cbqDnIuiGLRoGpkeysmT6Tn63SUdEhKL79lVNaanCY2JUlpurqqIi9U1N1aALLtCoa69VRX6+cjdtkufzz1VdUqKyQ4d0ZPVqjbr2Wu1YscK/BVm9zYsWKevZZ+WIiJAkxSQl+QPYhkPBm+uBbjjEvWT/fu3797+Vu3mzBo0bp0qXS1uWLpW3pkZxgwfr2JEjqnS5VLJ/vzxVVSrMzlZ4dLScEREqO1jXROd9/rl/bvzWpUuVOmOGzps1y1+WzYsWKXfjRiWPGqUZy5bRww0AAAB0YwTdsN2A9HQ5IyPlramR4XDUDRV3u9sciEf16aOL775bax9+WGWHD0ter2QY8tbU6EhWlvqNHq2pTz2l1XfeqbCoKEXFx2v3229r+1NPKWnECB3Ly9ORrCwNnzbtxEUNQ47wcPUbNUoyzYDh38H+Xd/TnJ2Zqf0ffqiv16yRIyJCptcrR1iYfB6PfG639r37rg588IGGfOc7kiSfx6PSb76R6fWqpqxM/dLSlLtpk2rLyuSrrVXK2LEqO3RIMk05wsM1ID1dCUOHSlLjueiGIYfTqdMmTCDgBgAAALo5gm7Y7khWlsKjouRwOmUYhvqfd55M09SBDz5o03XcFRX68L77VOlySfUBu2nKW1srT3W1Sg4c0GszZyp340b1GTJEw664Qrs3b5bP69Wx/Hz5PB6Zpqm1CxaoV0qKdqxYoTMuvVSDxo/XxPnzlTB0qGKSk/09ypUul4p275YkpX7/+wG93hufeUYVhYWSzydvdXXdYnHHh71LkqeyUjIMHcnKkruibkBc/UJy5bm58tbWKiwqSu7KSnmqq1WwfbuGfOc7yv/yS9WUlmr7P/6hSb/5TcDz1wf7w6ZMCeiRBwCgPYIN4V5mSTsfD0zfU24Zbi5Ji84JTH8WOGh70gefNsoy6Q3LgRWByQ/yG9/GOgTdupXXxY2zaOQEy4H/sqSnBMn0RWAy7ZHAtPu9xlm2B7lMQ9VBjlm3ETv5MYEAuiuCbtim4fDr8XfcoV4pKfr8+ecVO2iQRl93nSJiY7Xn7bdPrHLegqriE7PRwmNj5S4vlyQdO3JEps+nXW++Wbfvt88nGYb+dfvtGv/QQ0pOS9PhDRtk+nxa88ADcoaF1fWQu90qyslRdWmp3pk3T9EJCbrkt7/1z8U+tGGD8rZskSM8XKddeKG+XrtWq+bM0ahrr1V4r15Sft1fBIbDofBevVRbadll1DRVnpvb6Dm8tbWNjrsrKlSYne3vKd/99tuKTkxU9sqV/gXg2BoMAAAACD0E3bCNNUhcu2CBXDt3qmDHDpXn5ip24MATPdaGEXTP7qZ4q6slp1Py+U4MUzfNE9cwTZUfOqRtL70k79GjMuuP+3zyHg/yI+LiFN67tyoKClS8a5dMn0//+s//1I/ffVcf//73OrxpU91QbsPQN598om/WrZO3tlZHtmzR8O99T0e/+UbyeGT6fKo9/gLgZFTkn3itX330qPa8845qy8v9c7xPZi9vAAAAAF2DoBu2sQaJaRkZqnS5tH/NGuVu3KjYgQPlq99KrA0Bt3RiSzEZRrPnVR89qvK9e+UIC5MzIkLu473RvVJSNOyKK7T3nXfqtjU7dkzemhody8vTx7//vfb++991w8EdDrmPHTtxQcNQ9dGj2vl//yd5PG0qc5McDsk06+a7O53yut2K6N1bJXv3yud2K7xXLx3evFmVhYUaN3u2JGntggUdtu84AAD1rKO6n7OkKxY3zrNg25eBB161nHBZSuNMl10emP7zzsCPGw0mly7Ls6xfbr2ssbDxfTTbks6zpL/VOMs7+wJzWIaTB1mMvdEK50WWdLDh5QwnB04dBN2wjXVrraTU1LqFzn7+c33x9dcKi4qqW3zM7fb3dIdFR7dta7HWBOumWRdwV5xorJPT0hQdH6/wmBgd3b+/7qBhyPT5tOOVV/y94bIu9uZwyHF88bWT4YiI8A+rdzgc8nm9ioqPV01ZWd089ZoaOaOi5HO75a6q0p7VqxUeHa2Y5GRJYpg5AAAAECIIutHpxt1+u2KSkvx7eB/auFFVRUWqOXo0IOA2HI4O2WrM9HoDAm5JOvDBB40XcjNNmV6vvN5mdsr0euWTTm44+fEXDIbTWXdbn08yTbmrqnTGpEnK3bxZteXlMhwOGU6nDIdDSSNHasillwYMLWeYOQAAAND9EXSj0yWlpiotI0PZmZmKTkhQZX6+vG63nJGR8jUYsh0WE1MXLLemN7uNc8JPltlcYN4Sw5DP7ZYzKkremhqFx8TUrWJeVSUZhq555RWtfeghxaSkyJB0LD9f5/30p/453/UjCFw5OQwzBwDYyjrcfHmQcw6vC0z/0TLsO35ykKXIv/dyYPoyy+djbmmcp79lSLpqLGnLEuiS9PmcwLRlNfaD/2icxXroY0t6nxortqStw8kZSg6c2hxdXQCcmvyLrBmGRlx1lSJiY+WpadB4Op11w86PB9JGWFjd3OcW5nCHhOO994bDoUETJmjKn/6kc266SaddeKEmzp+v4dOmadb69br+9dc1cNw4VeTlaceKFdq6dKmyMzP9l6mvw4bHAAAAAHQv9HSjSzRcXC07M1M7X39dhmEopn9/DRo/XpKUu2GDqoqKZPp8MltatKwTe7lbzeFQWFSUfz55/XByn9er+lcHp02YoPRbb1X6rbcGvUR9PQ1IT9eRrKygw8sZZg4AAAB0XwTd6BINF1lLy8hQZWGhcjdtUuk338hdUaEjWVnyeb0aMG6cju7fH7CdlnUouREW1jgoNww5IyNlODppMEeQ4e2RvXsr9eqrtf/99xWdmKhjR47IW1ur5NGj1X/s2Lp9xVt4WdCwnoZPm9bkZwAAdAbrMGpJ+pclbV13PNWy+rcknW85lmb5/LLByxpnsh6LCUzWBrnPSkv6I0u68Rrp0mFL+qgl7Q6SxzrpjOHkABoi6EaXS0pN1dQ//1munBxlZ2bWLbD2yivK/+IL9R0xQuf99Kd691e/Us3Ro3IeX/HcU10t0+uV4XQqLDpaETExcoSFqaasTN6aGnm9XkXGxsoRdpI/4kGC6bDoaMk05aluMGPr+MJo9fuBh0VF1W2HZppK++EPVVVSor3vvquK/HxFxMRo0q9/rb7Dh9NLDQAAAPRwBN3odPXBtXUBsIY9t0eysrTrzTdVsm+fynNzlfr972vXm29qxFVXqWj3bh3JypKpugXNnGFhcoSFyXA4VHvsmMJjYuStrVVVSYn6JCScXGEbBNzOqCiZXm9dsG2ackZF1X3g88nn8yksMlIRsbGqKiqqK0NNjaITEhSTnKycN9/UsO9+V+WHD2vUtdcGfX4AAAAAPQ9BNzqdfxE1Nb3PdFpGhipdLuVu2CDXzp3qN3q0Js6fr7SMDG1evFiF2dmKHThQ0fHxOvf4yt65mzapsrBQgyZMUGF2tqpKShQ/ZIhir7pK+99/Xz63W1EJCaosKGjzVmS9+vdXbXm53NXVCouO9m9nlnbNNZKk/G3blDJmjGrKyrT7rbfkrqjQ4Isu0rjZswOeKSk1VWsXLGjy+Zt6IQEAAAAgNBF0o9O1ZgGwpNRUTX3qqaBBaP0+39bA1JWTo+wLLqgLzBct0hd//7t6JSerd0qKjmzerLjBg9V3+HBJUv6XXyr/yy+bnVMdERur2ooKRScmatgVV2j3W2/JGRGhC3/xC+15+20V7Nih4j175PrqK0nSkEsuUfHu3XJGRiplzBhNX7jQX76GwXVzz9+aFxIAAHQnlZZ0tiW9O0ge6/TreEs6+WDjPDFBjjUUZGMyFVnS1rJat/aSmJ8NoOMRdKPTtXYBsNYMQ7det3419GFTp8o0DPkMQwPOP1/jfvYzVbpcyl65UkkjRypl7FgV79mjiN69FR4To7JDhyRJPo9HYTEx8lRUqPfAgepz2mly7dyp6Ph4Xfjzn/t72gu2b1e/0aN1yW9/q73vvFMXvBuGSg8e1GkXXhgQcLfl+VmRHAAAAOhZCLrRbdX3+lYWFiomOblVQ643L16sL158UWNnzlRMUpJyPvtMvQ1Dl957r1w5OSrYvr1uuPqoUbr47rv9wW39Am5HsrLUKyVFO1as0MT585UwdGjjwN805XA6NeiCCzR82jT/quKunJygPfBtwYrkAAAAQM9C0I1uqz4grnS5Gg25bnLuc/1wcdNU2jXXyB0bq7QpUyTVBbTTFy4Mmu+S40H5kawsnXHJJQH7ZluD4HGzZ/tfAjREwAwAQOMttYJtsWUd5n3Ukj4QJI+zhftah4VLDA0H0D0QdKPbqg9iG/Yg12tq7nPDgDhx+HCNjo9X3379Gl3TypWTo1Vz5si1c6ck+YepB+u1JrgGAAAA0FoE3ej2ggW5Tc19bniurw0rlGdnZsq1c6eSRo70B9wsaAYAAADgZBF0IyR1dG9zwyC+fkG2hscBAIB9WvOanKHiAEIVQTegxkE8Q8gBAAAAdARHVxcAAAAAAICeiqAbAAAAAACbEHQDAAAAAGATgm4AAAAAAGxC0A0AAAAAgE0IugEAAAAAsAlBNwAAAAAANiHoBgAAAADAJgTdAAAAAADYhKAbAAAAAACbEHQDAAAAAGATgm4AAAAAAGxC0A0AAAAAgE0IugEAAAAAsAlBNwAAAAAANiHoBgAAAADAJgTdAAAAAADYhKAbAAAAAACbEHQDAAAAAGATgm4AAAAAAGxC0A0AAAAAgE0IugEAOMUVFxfrxhtvVFxcnOLj4zVr1iwdO3asVXlN09S0adNkGIZef/11ewsKAEAIIugGAOAUd+ONN2rHjh1699139a9//Utr167Vbbfd1qq8Tz31lAzDsLmEAACELtuCbt6aAwDQ/X311VdavXq1nnvuOU2YMEEXX3yxnn76ab3yyis6fPhws3k///xzPfnkk3r++ec7qbQAAIQe24Ju3poDAND9rV+/XvHx8Ro3bpz/2OTJk+VwOLRhw4Ym81VWVuqGG27QwoUL1b9//84oKgAAISnMjovWvzXftGmTvxF/+umnNX36dD3xxBMaOHBgk3nr35pv3rxZAwYMsKN4AADguLy8PPXr1y/gWFhYmBITE5WXl9dkvl/84he66KKLNGPGjFbfq6amRjU1Nf50WVlZ2wsMAECIsSXobumt+Q9+8IOg+drz1rypBtzn88nn853EU3QOn88n0zRDoqyhhrq1B/VqH+rWPqFWtw7HyQ9Eu/vuu/XYY481e85XX33Vrmu/+eab+uCDD7R169Y25XvkkUf04IMPtuueAACEKluC7s58a95UA15YWKjq6urWF7qL+Hw+lZaWyjTNDvkjCydQt/agXu1D3don1Oq2I4Zrz58/Xz/5yU+aPWfo0KHq37+/CgoKAo57PB4VFxc3WY4PPvhAe/fuVXx8fMDxH/7wh5o0aZLWrFkTNN8999yjefPm+dNlZWUaPHhwi88CAEAoa1PQ3R3fmjfVgCcnJysuLq5dZelMPp9PhmEoOTk5JP4QDCXUrT2oV/tQt/Y5Fes2OTlZycnJLZ43ceJEHT16VFlZWUpPT5dUF1T7fD5NmDAhaJ67775b//Ef/xFwbMyYMfrTn/6kq666qsl7RUZGKjIysg1PAQBA6GtT0N0d35o31YA7HI6Q+cPKMIyQKm8ooW7tQb3ah7q1D3Ub3Nlnn62pU6fq1ltv1ZIlS+R2uzV37lxdd911/jVYcnNzdfnll2v58uUaP368+vfvH7Q9P/3003XmmWd29iMAANCttSno7o5vzQEAwMl56aWXNHfuXF1++eVyOBz64Q9/qL/85S/+z91ut3JyclRZWdmFpQQAIDTZMqebt+YAAISOxMREvfzyy01+PmTIEJmm2ew1WvocAIBTlW1j7F566SWNHDlSl19+uaZPn66LL75Yzz77rP9z3poDAAAAAHo6W3q6Jd6aAwAAAADAajIAAAAAANiEoBsAAAAAAJsQdAMAAAAAYBOCbgAAAAAAbELQDQAAAACATQi6AQAAAACwCUE3AAAAAAA2IegGAAAAAMAmBN0AAAAAANiEoBsAAAAAAJsQdAMAAAAAYBOCbgAAAAAAbELQDQAAAACATQi6AQAAAACwCUE3AAAAAAA2IegGAAAAAMAmBN0AAAAAANiEoBsAAAAAAJsQdAMAAAAAYBOCbgAAAAAAbELQDQAAAACATQi6AQAAAACwCUE3AAAAAAA2IegGAAAAAMAmYV1dgI5mmqYkqaysrItL0jo+n0/l5eWKioqSw8E7kI5E3dqDerUPdWufUKzb2NhYGYbR1cWwVX2bbXZxOQAAaI/69qu+PWtKjwu6y8vLJUmDBw/u4pIAANB+paWliouL6+pi2Kq+za7q4nIAAHAyysvL1adPnyY/N8yWwvIQ4/P5dPjw4ZDpISgrK9PgwYN18ODBHv/HVWejbu1BvdqHurVPKNZtqLRjJyPU2mwpNH+WQgH1ag/q1R7Uqz1CsV5N01R5ebkGDhzY7Ei6HtfT7XA4dNppp3V1MdosLi4uZH64Qg11aw/q1T7UrX2o2+4lVNtsiZ8lu1Cv9qBe7UG92iPU6rW5Hu56oTGxDQAAAACAEETQDQAAAACATQi6u1hkZKTuv/9+RUZGdnVRehzq1h7Uq32oW/tQt+go/CzZg3q1B/VqD+rVHj25XnvcQmoAAAAAAHQX9HQDAAAAAGATgm4AAAAAAGxC0A0AAAAAgE0IugEAAAAAsAlBdycrLi7WjTfeqLi4OMXHx2vWrFk6duxYq/Kapqlp06bJMAy9/vrr9hY0BLW1bouLi3XHHXcoNTVV0dHROv300/Vf//VfKi0t7cRSd08LFy7UkCFDFBUVpQkTJmjjxo3Nnp+ZmamRI0cqKipKY8aM0apVqzqppKGnLXX7t7/9TZMmTVJCQoISEhI0efLkFr8Xp7K2/tzWe+WVV2QYhq6++mp7C4iQRdttD9rtjkGbbQ/aa3ucqm01QXcnu/HGG7Vjxw69++67+te//qW1a9fqtttua1Xep556SoZh2FzC0NXWuj18+LAOHz6sJ554Qtu3b9cLL7yg1atXa9asWZ1Y6u5nxYoVmjdvnu6//35t2bJFY8eO1ZQpU1RQUBD0/E8//VTXX3+9Zs2apa1bt+rqq6/W1Vdfre3bt3dyybu/ttbtmjVrdP311+vDDz/U+vXrNXjwYF1xxRXKzc3t5JJ3f22t23oHDhzQL3/5S02aNKmTSopQRNttD9rtk0ebbQ/aa3uc0m21iU6TnZ1tSjI3bdrkP/b222+bhmGYubm5zebdunWrOWjQIPPIkSOmJPO1116zubSh5WTqtqH//d//NSMiIky3221HMUPC+PHjzTlz5vjTXq/XHDhwoPnII48EPf9HP/qReeWVVwYcmzBhgvmf//mftpYzFLW1bq08Ho8ZGxtrvvjii3YVMWS1p249Ho950UUXmc8995w5c+ZMc8aMGZ1QUoQa2m570G53DNpse9Be2+NUbqvp6e5E69evV3x8vMaNG+c/NnnyZDkcDm3YsKHJfJWVlbrhhhu0cOFC9e/fvzOKGnLaW7dWpaWliouLU1hYmB3F7PZqa2uVlZWlyZMn+485HA5NnjxZ69evD5pn/fr1AedL0pQpU5o8/1TVnrq1qqyslNvtVmJiol3FDEntrduHHnpI/fr1O6V7ydAy2m570G6fPNpse9Be2+NUb6tPvd9QXSgvL0/9+vULOBYWFqbExETl5eU1me8Xv/iFLrroIs2YMcPuIoas9tZtQy6XSw8//HCrhwz2RC6XS16vVykpKQHHU1JStHPnzqB58vLygp7f2no/VbSnbq3uuusuDRw4sNEfTKe69tTtunXrtHTpUn3++eedUEKEMtpue9BunzzabHvQXtvjVG+r6enuAHfffbcMw2j2q7X/k1q9+eab+uCDD/TUU091bKFDhJ1121BZWZmuvPJKpaWl6YEHHjj5ggMd7NFHH9Urr7yi1157TVFRUV1dnJBWXl6uH//4x/rb3/6mpKSkri4Oughttz1ot3Gqo73uGD2traanuwPMnz9fP/nJT5o9Z+jQoerfv3+jhQI8Ho+Ki4ubHHr2wQcfaO/evYqPjw84/sMf/lCTJk3SmjVrTqLk3Z+ddVuvvLxcU6dOVWxsrF577TWFh4efbLFDVlJSkpxOp/Lz8wOO5+fnN1mP/fv3b9P5p6r21G29J554Qo8++qjee+89nXPOOXYWMyS1tW737t2rAwcO6KqrrvIf8/l8kup62nJycjRs2DB7C40uR9ttD9rtzkObbQ/aa3uc8m11V08qP5XULxqyefNm/7F33nmn2UVDjhw5Ym7bti3gS5L55z//2dy3b19nFb3ba0/dmqZplpaWmhdeeKF56aWXmhUVFZ1R1G5v/Pjx5ty5c/1pr9drDho0qNlFWb73ve8FHJs4cSKLsgTR1ro1TdN87LHHzLi4OHP9+vWdUcSQ1Za6raqqavR7dcaMGeZll11mbtu2zaypqenMoqObo+22B+12x6DNtgfttT1O5baaoLuTTZ061TzvvPPMDRs2mOvWrTOHDx9uXn/99f7PDx06ZKamppobNmxo8hpiBdSg2lq3paWl5oQJE8wxY8aYe/bsMY8cOeL/8ng8XfUYXe6VV14xIyMjzRdeeMHMzs42b7vtNjM+Pt7My8szTdM0f/zjH5t33323//xPPvnEDAsLM5944gnzq6++Mu+//34zPDzc3LZtW1c9QrfV1rp99NFHzYiICHPlypUBP5/l5eVd9QjdVlvr1iqUV0SF/Wi77UG7ffJos+1Be22PU7mtJujuZEVFReb1119v9u7d24yLizNvueWWgP8h9+/fb0oyP/zwwyavQcMdXFvr9sMPPzQlBf3av39/1zxEN/H000+bp59+uhkREWGOHz/e/Oyzz/yfXXrppebMmTMDzv/f//1fc8SIEWZERIQ5atQo86233urkEoeOttTtGWecEfTn8/777+/8goeAtv7cNhTKDTnsR9ttD9rtjkGbbQ/aa3ucqm21YZqmae8AdgAAAAAATk2sXg4AAAAAgE0IugEAAAAAsAlBNwAAAAAANiHoBgAAAADAJgTdAAAAAADYhKAbAAAAAACbEHQDAAAAAGATgm4AAAAAAGxC0A0AAAAAgE0IugEAAAAAsAlBNwAAAAAANiHoBgAAAADAJv8fyAOifXYmR80AAAAASUVORK5CYII=","text/plain":["
"]},"metadata":{},"output_type":"display_data"}],"source":["# Prepare figure\n","f, axarr = plt.subplots(1, 2, figsize=(10, 4))\n","\n","# Obtain amortized samples\n","obs_data = np.zeros((1, 2)).astype(np.float32)\n","samples_at_origin = amortizer.sample({\"direct_conditions\": obs_data}, n_samples=5000)\n","\n","# Plot samples\n","axarr[0].scatter(samples_at_origin[:, 0], samples_at_origin[:, 1], color=\"maroon\", alpha=0.75, s=0.5)\n","sns.despine(ax=axarr[0])\n","axarr[0].set_title(r\"Posterior samples at $x=(0, 0)$\")\n","axarr[0].grid(alpha=0.3)\n","axarr[0].set_xlim([-0.5, 0.5])\n","axarr[0].set_ylim([-0.5, 0.5])\n","\n","# Compute log density on relevant posterior range\n","side = np.linspace(-0.5, 0.5, 100)\n","x, y = np.meshgrid(side, side)\n","obs_data_rep = np.zeros((10000, 2)).astype(np.float32)\n","params = np.c_[x.flatten(), y.flatten()]\n","lpdf = amortizer.log_posterior({\"parameters\": params, \"direct_conditions\": obs_data_rep})\n","\n","# Plot the density map using nearest-neighbor interpolation\n","axarr[1].pcolormesh(x, y, np.exp(lpdf).reshape(100, 100), cmap=cm.hot)\n","axarr[1].set_title(r\"Posterior density at $x=(0, 0)$\")\n","\n","f.tight_layout()"]},{"cell_type":"markdown","id":"66248a2f","metadata":{},"source":["## Further Experimentation \n","\n","Feel free to explore the following settings:\n","1. Change the `coupling_design` argument of the `InvertibleNetwork` to either `affine` or `interleaved`. What do you observe?\n","2. Try out `mc_dropout=True` in the `coupling_settings` of the `InvertibleNetwork`. What happens to the spread of the posterior samples?\n","3. Can you make the network overfit by removing all regularization?\n","4. What posteriors do different \"observed\" points $x$ yield?"]}],"metadata":{"kernelspec":{"display_name":"Python 3 (ipykernel)","language":"python","name":"python3"},"language_info":{"codemirror_mode":{"name":"ipython","version":3},"file_extension":".py","mimetype":"text/x-python","name":"python","nbconvert_exporter":"python","pygments_lexer":"ipython3","version":"3.10.13"},"toc":{"base_numbering":1,"nav_menu":{},"number_sections":true,"sideBar":true,"skip_h1_title":true,"title_cell":"Table of Contents","title_sidebar":"Contents","toc_cell":true,"toc_position":{"height":"calc(100% - 180px)","left":"10px","top":"150px","width":"165px"},"toc_section_display":true,"toc_window_display":true}},"nbformat":4,"nbformat_minor":5} +{ + "cells": [ + { + "cell_type": "markdown", + "id": "009b6adf", + "metadata": {}, + "source": [ + "# Two Moons: Tackling Bimodal Posteriors" + ] + }, + { + "cell_type": "markdown", + "id": "3ed81254", + "metadata": {}, + "source": [ + "## Table of Contents\n", + " * [Inference Network and Amortizer](#inference_network_and)\n", + " * [Trainer](#trainer)\n", + " * [Validation](#validation)\n", + "\t * [Global Calibration](#global_calibration)\n", + "\t * [Two Moons Posterior](#two_moons_posterior)\n", + " * [Further Experimentation](#further_experimentation)" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "d5f88a59", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "c:\\Users\\radevs\\Desktop\\Projects\\BayesFlow\\examples\\..\\bayesflow\\simulation.py:28: TqdmExperimentalWarning: Using `tqdm.autonotebook.tqdm` in notebook mode. Use `tqdm.tqdm` instead to force console mode (e.g. in jupyter console)\n", + " from tqdm.autonotebook import tqdm\n" + ] + } + ], + "source": [ + "import matplotlib.pyplot as plt\n", + "import numpy as np\n", + "import seaborn as sns\n", + "from matplotlib import cm\n", + "\n", + "import keras\n", + "\n", + "## REMOVE ON PRODUCTION\n", + "import sys\n", + "sys.path.append('../')\n", + "\n", + "from bayesflow.experimental.networks import CouplingFlow\n", + "from bayesflow.experimental.amortizers import AmortizedPosterior\n", + "from bayesflow.experimental.datasets import OfflineDataset\n", + "\n", + "from bayesflow import benchmarks" + ] + }, + { + "cell_type": "markdown", + "id": "9525ffd7", + "metadata": {}, + "source": [ + "This example will demonstrate amortized estimation of a somewhat strange Bayesian model, whose posterior evaluated at the origin $x = (0, 0)$ of the \"data\" will resemble two crescent moons. The forward process is a noisy non-linear transformation on a 2D plane:\n", + "\n", + "$$\n", + "\\begin{align}\n", + "x_1 &= -|\\theta_1 + \\theta_2|/\\sqrt{2} + r \\cos(\\alpha) + 0.25\\\\\n", + "x_2 &= (-\\theta_1 + \\theta_2)/\\sqrt{2} + r\\sin{\\alpha}\n", + "\\end{align}\n", + "$$\n", + "\n", + "with $x = (x_1, x_2)$ playing the role of \"observables\", $\\alpha \\sim \\text{Uniform}(-\\pi/2, \\pi/2)$, $r \\sim \\text{Normal}(0.1, 0.01)$, and a prior over the 2D parameter vector $\\theta = (\\theta_1, \\theta_2)$:\n", + "\n", + "$$\n", + "\\begin{align}\n", + "\\theta_1, \\theta_2 \\sim \\text{Uniform}(-1, 1)\n", + "\\end{align}\n", + "$$\n", + "\n", + "This method is typically used for benchmarking simulation-based inference (SBI) methods (see https://arxiv.org/pdf/2101.04653) and any method for amortized Bayesian inference should be capable of recovering the two moons posterior *without* using a gazillion of simulations. Note, that this is a considerably harder task than modeling the common unconditional two moons data set used often in the context of normalizing flows.\n", + "\n", + "The two moons generative model, along with all benchmarks from https://arxiv.org/pdf/2101.04653, exists as a standalone object in the `bayesflow.benchmarks` module. So let's import it using the `Benchmark` helper class." + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "0b9a9817", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:root:Performing 2 pilot runs with the two_moons model...\n", + "INFO:root:Shape of parameter batch after 2 pilot simulations: (batch_size = 2, 2)\n", + "INFO:root:Shape of simulation batch after 2 pilot simulations: (batch_size = 2, 2)\n", + "INFO:root:No optional prior non-batchable context provided.\n", + "INFO:root:No optional prior batchable context provided.\n", + "INFO:root:No optional simulation non-batchable context provided.\n", + "INFO:root:No optional simulation batchable context provided.\n" + ] + } + ], + "source": [ + "benchmark = benchmarks.Benchmark(\"two_moons\", mode=\"posterior\")" + ] + }, + { + "cell_type": "markdown", + "id": "2d4c6eb0", + "metadata": {}, + "source": [ + "## Inference Network and Amortizer \n", + "We will use a neural spline flow (https://arxiv.org/abs/1906.04032) for modeling the posterior, as these are specialized for locally weird posteriors. By default, some weight regularization and dropout will be applied during training. These can be modified through the `coupling_settings` keyword of the `InvertibleNetwork`." + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "e8d7e053", + "metadata": {}, + "outputs": [], + "source": [ + "offline_data = benchmark.generative_model(1000)\n", + "\n", + "data = {\n", + " \"parameters\": dict(theta=offline_data[\"prior_draws\"]),\n", + " \"observables\": dict(x=offline_data[\"sim_data\"])\n", + "}\n", + "dataset = OfflineDataset(data, batch_size=8, batches_per_epoch=1000//8)" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "b1c98fbd", + "metadata": {}, + "outputs": [], + "source": [ + "inference_net = CouplingFlow.all_in_one(target_dim=2, coupling_layers=4)\n", + "\n", + "amortizer = AmortizedPosterior(inference_net)\n", + "\n", + "epochs = 200\n", + "lr = keras.optimizers.schedules.CosineDecay(5e-4, decay_steps=int(epochs * dataset.batches_per_epoch))\n", + "# TODO: Experiments with clipnorm\n", + "optimizer = keras.optimizers.AdamW(lr, weight_decay=1e-3)\n", + "amortizer.compile(optimizer=optimizer)" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "ecc11536", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Epoch 1/200\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "c:\\Users\\radevs\\AppData\\Local\\anaconda3\\envs\\bf\\Lib\\site-packages\\keras\\src\\layers\\layer.py:1295: UserWarning: Layer 'amortized_posterior_1' looks like it has unbuilt state, but Keras is not able to trace the layer `call()` in order to build it automatically. Possible causes:\n", + "1. The `call()` method of your layer may be crashing. Try to `__call__()` the layer eagerly on some test input first to see if it works. E.g. `x = np.random.random((3, 4)); y = layer(x)`\n", + "2. If the `call()` method is correct, then you may need to implement the `def build(self, input_shape)` method on your layer. It should create all variables used by the layer (e.g. by calling `layer.build()` on all its children layers).\n", + "Exception encountered: ''Exception encountered when calling CouplingFlow.call().\n", + "\n", + "\u001b[1mgot an unexpected keyword argument 'kwargs'\u001b[0m\n", + "\n", + "Arguments received by CouplingFlow.call():\n", + " • args=('tf.Tensor(shape=(None, 2), dtype=float32)', 'tf.Tensor(shape=(None, 2), dtype=float32)')\n", + " • kwargs={'kwargs': {}}''\n", + " warnings.warn(\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "WARNING:tensorflow:From c:\\Users\\radevs\\AppData\\Local\\anaconda3\\envs\\bf\\Lib\\site-packages\\keras\\src\\backend\\tensorflow\\core.py:184: The name tf.placeholder is deprecated. Please use tf.compat.v1.placeholder instead.\n", + "\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "WARNING:tensorflow:From c:\\Users\\radevs\\AppData\\Local\\anaconda3\\envs\\bf\\Lib\\site-packages\\keras\\src\\backend\\tensorflow\\core.py:184: The name tf.placeholder is deprecated. Please use tf.compat.v1.placeholder instead.\n", + "\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m17s\u001b[0m 23ms/step\n", + "Epoch 2/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 22ms/step\n", + "Epoch 3/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 22ms/step\n", + "Epoch 4/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 22ms/step\n", + "Epoch 5/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 22ms/step\n", + "Epoch 6/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 23ms/step\n", + "Epoch 7/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 23ms/step\n", + "Epoch 8/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 23ms/step\n", + "Epoch 9/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 22ms/step\n", + "Epoch 10/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 22ms/step\n", + "Epoch 11/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 23ms/step\n", + "Epoch 12/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 23ms/step\n", + "Epoch 13/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 23ms/step\n", + "Epoch 14/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 23ms/step\n", + "Epoch 15/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 16/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 17/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 18/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 19/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 26ms/step\n", + "Epoch 20/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 21/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 22/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 23/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 24/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 25/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 26/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 27/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 28/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 29/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 30/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 31/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 32/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 33/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 34/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 35/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 36/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 37/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 38/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 39/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 40/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 41/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 42/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 43/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 44/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 45/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 46/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 47/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 48/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 49/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 50/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 51/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 52/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 53/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 54/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 55/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 56/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 57/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 58/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 59/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 60/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 61/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 62/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 63/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 64/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 65/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 66/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 67/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 68/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 69/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 70/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 71/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 72/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 73/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 74/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 75/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 76/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 77/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 78/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 79/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 80/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 81/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 82/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 83/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 84/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 85/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 86/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 87/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 88/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 89/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 90/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 91/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 92/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 93/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 94/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 95/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 96/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 97/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 98/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 99/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 100/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 101/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 102/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 103/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 104/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 105/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 106/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 107/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 108/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 109/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 110/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 111/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 112/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 27ms/step\n", + "Epoch 113/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 114/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 115/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 116/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 117/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 118/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 119/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 120/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 121/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 122/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 123/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 124/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 125/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 126/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 127/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 128/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 129/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 130/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 131/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 132/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 133/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 134/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 135/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 136/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 137/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 138/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 139/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 140/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 141/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 142/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 143/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 144/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 145/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 146/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 26ms/step\n", + "Epoch 147/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 148/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 149/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 150/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 151/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 152/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 153/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 154/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 155/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 156/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 157/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 158/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 159/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 160/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 161/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 162/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 163/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 164/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 165/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 26ms/step\n", + "Epoch 166/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 167/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 168/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 169/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 170/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 171/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 172/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 173/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 174/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 175/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 24ms/step\n", + "Epoch 176/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 177/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 178/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 179/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 26ms/step\n", + "Epoch 180/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 26ms/step\n", + "Epoch 181/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 28ms/step\n", + "Epoch 182/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 26ms/step\n", + "Epoch 183/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 26ms/step\n", + "Epoch 184/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 26ms/step\n", + "Epoch 185/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 186/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 187/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 188/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 189/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 190/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 191/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 192/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 193/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 194/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 195/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 26ms/step\n", + "Epoch 196/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 26ms/step\n", + "Epoch 197/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 198/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 25ms/step\n", + "Epoch 199/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 26ms/step\n", + "Epoch 200/200\n", + "\u001b[1m125/125\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 28ms/step\n" + ] + } + ], + "source": [ + "h = amortizer.fit(dataset, epochs=epochs)" + ] + }, + { + "cell_type": "markdown", + "id": "b90a6062", + "metadata": {}, + "source": [ + "## Validation \n", + "We can use simulation-based calibration(SBC) for free (due to amortization) checking of computational faithfulness.\n", + "\n", + "1. Talts, S., Betancourt, M., Simpson, D., Vehtari, A., & Gelman, A. (2018). Validating Bayesian inference algorithms with simulation-based calibration. arXiv preprint arXiv:1804.06788.\n", + "2. Säilynoja, T., Bürkner, P. C., & Vehtari, A. (2022). Graphical test for discrete uniformity and its applications in goodness-of-fit evaluation and multiple sample comparison. Statistics and Computing, 32(2), 32." + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "id": "f76289b3", + "metadata": {}, + "outputs": [], + "source": [ + "# TODO" + ] + }, + { + "cell_type": "markdown", + "id": "e9ee69a1", + "metadata": {}, + "source": [ + "### Two Moons Posterior \n", + "\n", + "The two moons posterior at point $x = (0, 0)$ should resemble two crescent shapes. Below, we plot the corresponding posterior samples and posterior density. These results suggest that our spline flow setup can approximate the expected analytical posterior fairly well." + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "065384db", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# Set the number of posterior draws you want to get\n", + "num_draws = 5000\n", + "\n", + "# Prepare figure\n", + "f, axarr = plt.subplots(1, 2, figsize=(10, 4))\n", + "\n", + "# Obtain amortized samples\n", + "obs_data = np.zeros((num_draws, 2)).astype(np.float32)\n", + "# TODO - replace with amortizer interface\n", + "samples_at_origin = inference_net.sample(num_draws, obs_data)\n", + "\n", + "# Plot samples\n", + "axarr[0].scatter(samples_at_origin[:, 0], samples_at_origin[:, 1], color=\"maroon\", alpha=0.75, s=0.5)\n", + "sns.despine(ax=axarr[0])\n", + "axarr[0].set_title(r\"Posterior samples at $x=(0, 0)$\")\n", + "axarr[0].grid(alpha=0.3)\n", + "axarr[0].set_xlim([-0.5, 0.5])\n", + "axarr[0].set_ylim([-0.5, 0.5])\n", + "\n", + "# Compute log density on relevant posterior range\n", + "side = np.linspace(-0.5, 0.5, 100)\n", + "x, y = np.meshgrid(side, side)\n", + "obs_data_rep = np.zeros((10000, 2)).astype(np.float32)\n", + "params = np.c_[x.flatten(), y.flatten()]\n", + "# TODO - replace with amortizer interface\n", + "lpdf = amortizer.log_prob(params, obs_data_rep)\n", + "\n", + "# # Plot the density map using nearest-neighbor interpolation\n", + "axarr[1].pcolormesh(x, y, np.exp(lpdf).reshape(100, 100), cmap=cm.hot)\n", + "axarr[1].set_title(r\"Posterior density at $x=(0, 0)$\")\n", + "\n", + "f.tight_layout()" + ] + }, + { + "cell_type": "markdown", + "id": "66248a2f", + "metadata": {}, + "source": [ + "## Further Experimentation \n", + "\n", + "# TODO" + ] + }, + { + "cell_type": "markdown", + "id": "c9944c66", + "metadata": {}, + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.11.5" + }, + "toc": { + "base_numbering": 1, + "nav_menu": {}, + "number_sections": true, + "sideBar": true, + "skip_h1_title": true, + "title_cell": "Table of Contents", + "title_sidebar": "Contents", + "toc_cell": true, + "toc_position": { + "height": "calc(100% - 180px)", + "left": "10px", + "top": "150px", + "width": "165px" + }, + "toc_section_display": true, + "toc_window_display": true + } + }, + "nbformat": 4, + "nbformat_minor": 5 +}