{ "cells": [ { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "# Solving PINNs Based on MindSpore Flow\n", "\n", "[![DownloadNotebook](https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/website-images/r2.1/resource/_static/logo_notebook_en.svg)](https://obs.dualstack.cn-north-4.myhuaweicloud.com/mindspore-website/notebook/r2.1/mindflow/en/features/mindspore_solve_pinns_by_mindflow.ipynb) [![DownloadCode](https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/website-images/r2.1/resource/_static/logo_download_code_en.svg)](https://obs.dualstack.cn-north-4.myhuaweicloud.com/mindspore-website/notebook/r2.1/mindflow/en/features/mindspore_solve_pinns_by_mindflow.py) [![ViewSource](https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/website-images/r2.1/resource/_static/logo_source_en.svg)](https://gitee.com/mindspore/docs/blob/r2.1/docs/mindflow/docs/source_en/features/solve_pinns_by_mindflow.ipynb)\n", "\n" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "## Overview\n", "\n", "This tutorial describes how to use sympy to define the Dirichlet boundary conditions and the Neumann boundary conditions based on the two-dimensional Poisson problem and train a physical information neural network model. In this tutorial, the following three aspects are introduced:\n", "\n", "- How to use sympy to define partial differential equation based on [MindSpore Flow](https://mindspore.cn/mindflow/docs/zh-CN/r0.1/index.html).\n", "\n", "- How to define the Dirichlet boundary conditions and the Neumann boundary conditions in a model.\n", "\n", "- How to train a physical information neural network using [MindSpore](https://mindspore.cn/docs/zh-CN/r2.1/index.html) functional programming paradigm." ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "## Problem Description\n", "\n", "Poisson's equation is an elliptic partial differential equation of broad utility in theoretical physics. For example, the solution to Poisson's equation is the potential field caused by a given electric charge or mass density distribution; with the potential field known, one can then calculate electrostatic or gravitational (force) field. We start from a 2-D homogeneous Poisson equation,\n", "\n", "$$\n", "f + \\Delta u = 0\n", "$$\n", "\n", "where `u` is the primary variable, `f` is the source term, and $\\Delta$ denotes the Laplacian operator.\n", "\n", "We consider the source term `f` is a constant value and given ($f=1.0$), then the form of Poisson' equation is as follows:\n", "\n", "$$\n", "\\frac{\\partial^2u}{\\partial x^2} + \\frac{\\partial^2u}{\\partial y^2} + 1.0 = 0,\n", "$$\n", "\n", "In this case, the Dirichlet boundary condition and the Neumann boundary condition are used. The format is as follows:\n", "\n", "Dirichlet boundary condition on the boundary of outside circle:\n", "\n", "$$\n", "u = 0\n", "$$\n", "\n", "Neumann boundary condition on the boundary of inside circle:\n", "\n", "$$\n", "du/dn = 0\n", "$$\n", "\n", "In this case, the PINNs method is used to learn the mapping $(x, y) \\mapsto u$. So that the solution of Poisson' equation is realized." ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "## Technology Path\n", "\n", "MindSpore Flow solves the problem as follows:\n", "\n", "1. Training Dataset Construction.\n", "2. Model Construction.\n", "3. Optimizer.\n", "4. Poisson2D.\n", "5. Model Training.\n", "6. Model Evaluation and Visualization." ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "### Importing the required packages\n" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import time\n", "\n", "import matplotlib.pyplot as plt\n", "import numpy as np\n", "import sympy\n", "from sympy import symbols, Function, diff\n", "\n", "import mindspore as ms\n", "from mindspore import nn, ops, Tensor, set_context, set_seed, jit\n", "from mindspore import dtype as mstype\n", "\n", "\n", "set_seed(123456)\n", "set_context(mode=ms.GRAPH_MODE, device_target=\"GPU\", device_id=0)\n" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "## Training Dataset Construction\n", "\n", "In this case, random sampling is performed according to the domain, initial condition and boundary condition to generate training data sets. [Disk](https://mindspore.cn/mindflow/docs/en/r0.1/geometry/mindflow.geometry.Disk.html#mindflow.geometry.Disk) and [CSGXOR](https://mindspore.cn/mindflow/docs/en/r0.1/geometry/mindflow.geometry.CSGXOR.html#mindflow.geometry.CSGXOR) are used to make a geometry with input and output boundaries, as well as domain. Download data construction [Python script](https://gitee.com/mindspore/mindscience/blob/r0.3/MindFlow/features/solve_pinns_by_mindflow/src/dataset.py)." ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "from mindflow.geometry import generate_sampling_config, Disk, CSGXOR\n", "\n", "class MyIterable:\n", " def __init__(self, domain, bc_outer, bc_inner, bc_inner_normal):\n", " self._index = 0\n", " self._domain = domain.astype(np.float32)\n", " self._bc_outer = bc_outer.astype(np.float32)\n", " self._bc_inner = bc_inner.astype(np.float32)\n", " self._bc_inner_normal = bc_inner_normal.astype(np.float32)\n", "\n", " def __next__(self):\n", " if self._index >= len(self._domain):\n", " raise StopIteration\n", "\n", " item = (self._domain[self._index], self._bc_outer[self._index], self._bc_inner[self._index],\n", " self._bc_inner_normal[self._index])\n", " self._index += 1\n", " return item\n", "\n", " def __iter__(self):\n", " self._index = 0\n", " return self\n", "\n", " def __len__(self):\n", " return len(self._domain)\n", "\n", "\n", "def _get_region(config):\n", " indisk_cfg = config[\"in_disk\"]\n", " in_disk = Disk(indisk_cfg[\"name\"], (indisk_cfg[\"center_x\"], indisk_cfg[\"center_y\"]), indisk_cfg[\"radius\"])\n", " outdisk_cfg = config[\"out_disk\"]\n", " out_disk = Disk(outdisk_cfg[\"name\"], (outdisk_cfg[\"center_x\"], outdisk_cfg[\"center_y\"]), outdisk_cfg[\"radius\"])\n", " union = CSGXOR(out_disk, in_disk)\n", " return in_disk, out_disk, union\n", "\n", "\n", "def create_training_dataset(config):\n", " '''create_training_dataset'''\n", " in_disk, out_disk, union = _get_region(config)\n", "\n", " union.set_sampling_config(generate_sampling_config(config[\"data\"]))\n", " domain = union.sampling(geom_type=\"domain\")\n", "\n", " out_disk.set_sampling_config(generate_sampling_config(config[\"data\"]))\n", " bc_outer, _ = out_disk.sampling(geom_type=\"BC\")\n", "\n", " in_disk.set_sampling_config(generate_sampling_config(config[\"data\"]))\n", " bc_inner, bc_inner_normal = in_disk.sampling(geom_type=\"BC\")\n", "\n", " plt.figure()\n", " plt.axis(\"equal\")\n", " plt.scatter(domain[:, 0], domain[:, 1], c=\"powderblue\", s=0.5)\n", " plt.scatter(bc_outer[:, 0], bc_outer[:, 1], c=\"darkorange\", s=0.005)\n", " plt.scatter(bc_inner[:, 0], bc_inner[:, 1], c=\"cyan\", s=0.005)\n", " plt.show()\n", " dataset = ms.dataset.GeneratorDataset(source=MyIterable(domain, bc_outer, bc_inner, (-1.0) * bc_inner_normal),\n", " column_names=[\"data\", \"bc_outer\", \"bc_inner\", \"bc_inner_normal\"])\n", " return dataset\n", "\n", "\n", "def _numerical_solution(x, y):\n", " return (4.0 - x ** 2 - y ** 2) / 4\n", "\n", "\n", "def create_test_dataset(config):\n", " \"\"\"create test dataset\"\"\"\n", " _, _, union = _get_region(config)\n", " union.set_sampling_config(generate_sampling_config(config[\"data\"]))\n", " test_data = union.sampling(geom_type=\"domain\")\n", " test_label = _numerical_solution(test_data[:, 0], test_data[:, 1]).reshape(-1, 1)\n", " return test_data, test_label\n" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "The geometric shape of the generated data is a ring, the radius of the inner circle is 1.0, the radius of the outer circle is 2.0, and the amount of data in the boundary and domain is 8192. The generation parameters are set as follows:" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAi8AAAGdCAYAAADaPpOnAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy88F64QAAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOx9e1hU17n+O14AkZENDDMKTLyANyooaOrlKGguTXqaYIhNToqJSWuCubZRjtTGmJiYi9Xjpa1NIok9xBLrqTFEk3Oa9pdEQZqYVkHEeomgJlyUi4COIuBl//7Y8y3WXrP2gEm8Jet9nnlgZvZl7bX3rPWt73u/97MB0KGgoKCgoKCgcI2g25VugIKCgoKCgoLCxUAZLwoKCgoKCgrXFJTxoqCgoKCgoHBNQRkvCgoKCgoKCtcUlPGioKCgoKCgcE1BGS8KCgoKCgoK1xSU8aKgoKCgoKBwTUEZLwoKCgoKCgrXFHpc6QZcCkRFRcHj8VzpZigoKCgoKChcBOx2O2pqajrd7ltnvERFRaG6uvpKN0NBQUFBQUHhKyA6OrpTA+ZbZ7yQxyU6Olp5XxQUFBQUFK4R2O12VFdXd2nu/tYZLwSPx6OMFwUFBQUFhW8hFGFXQUFBQUFB4ZqCMl4UFBQUFBQUriko40VBQUFBQUHhmoIyXhQUFBQUFBSuKSjjRUFBQUFBQeGagjJeFBQUFBQUFK4pKONFQUFBQUFB4ZqCMl4UFBQUFBQUriko40VBQUFBQUHhmsIlNV7mzZuHf/zjHzh58iRqa2uRn5+PIUOGdLrfj3/8Y+zbtw9nzpzB7t278cMf/vBSNlNBQUFBQUHhGsIlNV5SU1Px+9//HuPGjcPNN9+Mnj174m9/+xuCg4Mt9xk/fjz+9Kc/Yc2aNUhKSsK7776Ld999F9/73vcuZVMVFBQUFBQUriHol+vlcDh0Xdf1SZMmWW6zfv16/b333jN99umnn+qvvvpql85ht9t1Xdd1u91+2a5LvdRLvdRLvdRLvb7e62Lm78tamDE0NBQA0NjYaLnN+PHjsXz5ctNnf/3rX3HHHXdItw8ICEBgYCB7b7fbv35DFRQUrip8mAncMNj3c10HFv4NWPTh5W+TgoLClcNlM15sNhtWrlyJoqIi/Otf/7Lcrm/fvqitrTV9Vltbi759+0q3/9WvfoWFCxd+k01VUFD4hvH2fcCdid/8cW024LlbjBdB143PO98ZOHsemPk/QF7xN982BQWFS4fLZrz8/ve/x4gRIzBx4sRv9Lgvv/yyyVNjt9tRXV39jZ5DQUHBP0bHAO/9DOjrdXxe0IFuogFhg+Hwpb8irD7vKrz7d9VwAYCe3YG1PzFeMnx8ELgp52u0SUFB4ZLgshgvv/vd73DbbbchJSWlU8Pi2LFjcLlcps9cLheOHTsm3b69vR3t7e3fWFsVFBT847MngOuv87OBTZIJYJP81S2+tzBg2s4BD23o8JK8eieQOd67K7e/fgGwdbM+Drpi3Hhxw2DgwlLfz5VRo6BwZfF11zqd4ne/+x3S09MxefJklJeXd7r9+vXrERwcjLS0NPbZ3//+d+zevRuPPPJIp/vb7XacPHkSffr0gcfj+VptV1D4rqPTcI+FEaIDsAkjyxlbAN4dtBzBSenYX7wDv5o2lX33TG4eRqVMxq7CrXj+gXt9ThObkIjpWdnQAaxbtgQVZbvZ52kzM1FcsAXJqVOweU0O0mZmYuJtaWg4WoMlj2aybQFg9spVmJR2BwDg+r/djuhTJR0N7uza/OCCDry+HXjkna5tr6Cg4IuLmb8vqefl97//PTIyMjB16lR4PB7mUTlx4gRaW1sBAG+++Saqq6vx1FNPAQB+85vfoKCgAHPmzMH//u//4p577sGYMWOQmZl5KZuqoKAAa2KsFBJvyac1wXj7uiUmY6KibDdiExKRkZWNuP4joes6WoSB6a1lS+BpbsbmNTlITZ+GjDlzsW75UhTkbwQApM3MxKiUyQCAURNTUHnwAH6XncUMFQBY8eTjAIDNa3IwbPQYOKKikDYzk31O39k1DTqA9XtTcUvGGtRVV+KNhQt8jJxF3Z5DcHu9rwEjMWq62YBZ440Xfb/3GDBiWRf7UkFB4aJwST0vui4/9AMPPIA333wTALBlyxYcOXIEP/3pT9n3P/7xj/HCCy9gwIABOHjwILKzs/GXv/ylS+dUnhcFha5jTxYQT1x4fjToZGRoPQs8ue8GjM58CR9tWI/oQXHMUAGMyX/ibWn4vLQE4U4XzpzyoP+weABAXVWlj0eExx8+K0aY04Wmulr8bGwygA7Py8iJKejWvTsAoHBTPvO0WBlLNhiGkexcvBdmV+FWeJqb2XHE4xHylt+H/7iQh+7Qrb01BOH7d3YDP/6jdZ8qKHzXcdV4XmxdYM5NmTLF57O3334bb7/99qVokoLCdxr3JgNr7jaIqiaInBT6XzdCIv/zeRh2jFjUMamvzUHazDvh6BeFG++6B45+UQDM3g8ASBg/AWFOF9pbWwFdR3tbG9YtN0gks1eu8jEOAKC2qhJhThdqqypNRkjBpnyEu/oiauAg1ByuYPuuePJxZiwNGz2GGUanmpsx8bY0eJqbTd4XwuY1OQjRNGan0f7itfB49r9LsXPm70ztfvs+ID2xcyrNnYkd/Jl/fgmM/V0nOygoKFjiknNeLjeU50VBwQwKBZ0/D3QXjRbAnAUEoKF7NPLHbDSFbcg4aDhaA0e/KBS9vxmb1+RgelY2etntOOPxMA8Hz085ULwDN951Dz7asJ4ZOaVFhYiOjWPHEY0E4rGQV4W8I/XVVX73eeqNXIRHOlHi5c3wx7Hy8ojnlHlwunoM6iP3P59HcvUadOO9M7KR1vvZiTMG+Xdnld/DKyh863HVeF4UFBSuDGTcFSvD5Vy3QPwZd+PZ/y5F2sxM2DUNI/tFITl1CjNeyJMiTu6e5maMnJiCovc3s8md56ecam7GrEnjAAA7t3yMtJmZCNE0OKKi0FBTw45rBd47st9rCBUXbPHZrqJsN47s24uwSCezEcgr0xXw29I107WInBp/bQWA4oImpE79IyMX/+6BGPwAfwNwDt14Y8b7NzQY+OcvjLetZ4FJryhDRkGhMyjjRUHhW4LRMcAHDwIRvbkPuRX/Ba8GyoULwPP/Dzh12yqfrJwVTz6O1PRpiI6NMxkJVpM7Tdi8EbJ5TQ5cbjciY9w+xyBvzee7SnDGYmUlGgyLvNlHs1eugkMwqnisW7YEp7ykXwCWXhOZl4XOy28bm5AIu6bhYGkJQjQNsQmJfr0vfB8lp07BxNvScKq5GU+syUHazJXs2OXzgEERkIbognp2GDKKI6OgYA1lvCgoXOM49izgDJF8wa3uz7QDKcKKPtW+BcNGj8G65UtNk3Jy6hS/RoIVeGOhtrISg0cm+RwjbWYmRk5MMcJPI5N8+CixCYkI0TSUFhX6eGVkhpLs3BTuyX4lR8pfIeOI57cA8PGw8G0dPDIJAJhx1FkIiW+r6AGKW9wRYrrh3QSE4oQ0eM9zZJQho6BghjJeFBSuQVCWEEdVMUHXjc8P2uLw728GSz0Pdk2TGildMRLIaxGiaRg1MYV9TwYAfwx+HzJMCjblm7wehLSZmRglhKEIvBHgrx2b1+QYhoskNEXelF1FhSiUtIH32oRoGj4vLYENQFVFOWzwNXBk6ApPprjAMBzv+PtNrO/9pamTIaMDyPlU6ckoKCjCroLCNQTL1GYOnx4Nxl8Tc3wE3Qi06t9VVNhlTwJl/QwYHg/N4cC59rMICAzE/pKdaPF4WDaQLL149spVmHh7Gtpb2xAQGCgl3JK2C592DfiGcqgtvHFCJGDSiUmbmYmJt6ehoUYuUjfxtjRTG2QCeFYEZWoPAFMWFH/d4jlkxoysHXxfrPlRPQY1FfreCO891wHkK2+MwrcMirCroPAtwqt3ApnjhJo9gkBc3Smg73MdE3tSdAx0GGEOcQKlVX/hpnzLsJA44U7PykZSqiFr0NrSgsBevQCbDS0eT6cpycUFW/D9m29BQFCgyRNCRkMvux2D4kcgICgIN951DyP40gQfomkmIyttZiYc/QzDZd3ypT4Gkxiy4SHzKokE4xVPPs76SNSwob7MfiUHkdExAMAypwDDIyOew0pIT2wHYcYv56PM5ULR8eOI2/owxuuf+Nx7m254Y84vAXKUsq/CdxDKeFFQuErhV5pfBy7YbPh82Gysb0jG83MNUitN7E31dTi8b690AgVgChfJPAO0PanRDhlliMWdaDyO/37hOdx67wy4Ytwo3JQPABg2eow0CwgwODRkuPCeEDIabDYbdF1Ha0sL04ABOiZ2p9uNpJTJsGsann/gXlMGUlX5QR8DzF+Wkew7XnWXzkm8n+hBccxgIV2atJmZcEQZffzFvr0+ITCe+Jz9Sg4+2rAepUWFJtKvvzaebDqOcJcLPXr0QO0d6/GG1/tz81/GIaStxrStjVP2Vd4Yhe8SlPGioHCVwRQakoAEzhbk/hFJ8ZOhF24FYOZziOEiq6yg2IREPPV6LsKcToRoGsvsoe1CNA1JXgMDAA6V7UZB/kYkp07BEC8hF4Bfgq+VJ4SMBlEnhkAT/ILcPABAL7udGRCdeXuoP7qi0VJRttunnpI/7wl5ZXgdHNl13z9vPsKcLtw6fQbKPv2EZR/507WpKNuNptpa9B86HA1Ha2APD0fLKQ92FRXifw7ejVunz0C/Q3/CTdX/he6CgrkNhrF79tfAG58pb4zCtxvKeFFQuErAUmhlsAF7e4zG1DVnAQCzV2aicFO+KTWYsmOsyK48Z4Mm0NkrVyHM6YTNZkMwZxzwIRIA6G23o8VrYMQmJMLpdqO5oQHFBVtQVX4QgG8IhJ+UZQaGzGiQgVKg7ZrGwkg2ALu8GUlWRgpvcPDX7s+QsWozb8yQd8uf5wowKwX7CxMRR4fUganOU4im4bqhwxAW6URjbS2iB8Xh8L690FJ+gffwC2zb/C7mNDyO6FDz8bp36/DGqOrXCt9WKONFQeEKw6+nxQbsd92J/f/2WwBAGt4FAOmkLJsg+aycGfPmIzzSycIvtK0okW/XNEZ+rSjbjUWcWi1gTLZDk0YDAJvAZcaJyPXoqidE3I43pDzNzXC63RiaNBq7Crd2EGS9kz9xYPjMJr7StMif6azNBD7M0xmnhdr/Qd5a1FVWsnNtXpMjrbfECkn26ygkSdt+7iVEh7tcSE6dgiP796KkcCts3v1WlHUQfAc2Ffpknt0wGKhfCNz6hhK+U/h2QRkvCgpXCCw1lpfn90YCzp4HNgTch+A7XgYANHs5LMUFW5AyNd1nUgaMiVM0InhNk7BIJ2CzoZfdzr4n4wQwjIZT3hW/TPOEN5iII1JcsMWyRlFXiKs8aNJ3ud0YkjTaZGTxlaZvvXcGbNx18JP/wy8sRkBQIHtfWlTIjDe7prFsKVkb/GnM8BD5Krz68OyVqwyFYm/aNr9dRla2EWbTdRbuomvmicfUV5QyvuLJx7Hms2IAQJ/wCKzyVtMmzJg3H2VOF/7e2Iibt9yIvnqdqb0RvQ3hO11X5F6Fbw+U8aKgcJnx2RPA9dd53wgFEc/aumFV9yy8svavAEoxPXwrS+GNiRuMR19agoDAQJR4vQ7+whH858zz4nTBGeOWqsXyHg4+HMUfh4yU5znVWyuDpLNJXiw1QMZN8/EGAIYdR5M7FXjMmDMXjXW17Jj85P+zpxeiT3g4TjY2Yt3ypUidmo7+wzqMFU9zM7QIh2VZgoysbCSlTGZ9K/MUycTwyJuSMjUdoyamYFdRIUut5mEDoOs6muvrWLjLSkiPeDXFBVsQm5CI+qpKAMDaxS/6GIH1VZUId7oA6Pj0jh0oen8zbFsWY/Gko+ihn+84v02FkxS+PVDGi4LCZYJJhEzQaGkIHIBPhr2AE+EjMSA0FGm2/ljx5OMmTkj2KzkICAoCuF07q9/Df19VfpBNlhSe8IeYuMGmiVqmi3KgeAcajtag+lC51CCRtYWvAC1LMeaPQRP1mdOn0Nbaio82rMfQ5DHQdR0tHo+JL0K9cu7sWUYq1iLNxgp5jGTg7UiZYSH7LG1mJpK8adYHSnai4WiNZQo6cVl4PRirGk+U7XT/vPnQdUCLdKDovc0oyN+IqvKDsGsanG43Xt64CTqAkoItpqyntJnz8N5taRjx7g0YrJf7XOgNg4Gzi4Gf/hnIK7Z+BhQUrlYo40VB4RLDR6eF+3v2nBEeaoi7H/3jhqMPgDqO4Mlj3fKluH/efNRWVWLdsiWW55OJrhGqK8pRVVHuo3wr84CIxgWB10WJ8WqcUMVoq3140LVVHyo3FVoUeSG07bDRY+D06tZED4rzqWE0YtwEREbHoOXkSTTW1mLtr1/syLzaZs688nizlGRZP7xxQenQvGEh+0zkDA2RlEQgWHmirEJu/HXXV1eZ7pkOYGjSaJZmvqtwq+keUrtKHQvx1rIl+PWo3UbaPffsde8OrP0JMG8KMGKZ9FYpKFy1UMaLgsIlhE/dIY5R+c+AKch48zgqyv6I3J1zARheg6oKYaXsRUH+RumkKBohdk3zEV0DfLORZB4Q4rNYyefz2+gA24a4OMF2OzOOZG3kvTiyQou8J4N4IUsezWTGmOgFAoAv9u9FuMuF4D59sHPrxyjI34gFuXkYlTIZTfUG/yM1fRrunzcfJxuPsywlEXzYTGb4yIwNGWeID4ulTE23VB+28prxobDUqelS5d/SokKUFG5FsDcLDDCXLqgo221KJ/+xt78fjDuEYfuXm4i98X2V2J3CtQdlvCgoXAJY1anRLwCfdpuAvzoexaiJKUi7YBAyjx05jNDwCJz2nMSoiSmIiY3Dkkc7snv8hWNET8muokLsKtxqEl0DzDwKwMxB4cM0ZOCQsUQCbfx2by1bgrSZmUwkLjYh0fDCREWh6L3NPhwRGYnVSnsmRNPQ226HnRN1k6VU8yUFAJiu1wbAZrMhPNKJtJmZBmHZ6UKY04UvDhwAAB+isamtk1LQUFNj+pwyuqwypirKdqO4YAsefmExAoOCMGz0GEsVXn/g76dY3sDKW9MZT4n1bVY2bI48/OzQ/bjO7uXD2Dr4MA+ONSqOL/rQbxMVFK44VG0jBYVvEK/eaUwCMlSfANwvGP+L3hL+b8acuXBERaF0WyGb9Nrb2hAQFIii93zr5QAdBg6t9N/yhpX48BFNinwdIHEC9leHh6/zA8BUm4ffhibc2IREzH8jF1qkE5+X7ERtZaVfI4wHHa+9rQ2vPT1P6nFavW07nDFu1FVVspIC/HU8tHARImPcWLv4RQBgIbc3Fi5gfcG39xmvt+bLA/vgum4Aq8PEXyuJ9pUUbmUeF77fRoyfgHCnC60tLXjt6Xkmz8sP752BfgMG4eiRQ3h94QLL6+drN9H9ps9lqdb+7p3V8Wm7/7u/BYMhePp0oKUdSH1VpVcrXF6o2kYKClcApiwiDlR3iIdIXgVgItZOz8pGbEIi+oSF42x7OwJ79UJTXZ1lOi7te+u9MzAseQx62e2oq6w0hY941Vwr8TZRzI62p0rQKVPTYdc0FGzKR4imMe8I78WhfdNmZrL07NMeT8f13p6GhPETcHjfXmnhSGrX92++BUHBwbh/3nxT+IrqIZ05dQqNtbWmkgJ8/9ZWVmKwl4OyeU0Oyj79xMQJGTZ6DBxRHeRlWsXZwyN86jBRm6ZnZQPwreRNmUrHvjiM1pYW5K/+vU+Yb8a8+egTHo4+4eGmPhKNjYqy3VjyaKbpHgAdITVSOxa9UbzHRlbzSdwOADILtmDV/f+O+PLfwUY9YAOCA4z06r3HFB9G4eqEMl4UFL4mXr0TyBzvO6Fd0IHfljpRef0ziE3wdfPTJCzqilSU7YanuRl9wsIBmw1nTp+GPSAAX+zfi+TUKawitCwd1xntZn/fWLjAp2YPAJMyr6juyhN2CaO4MBKd39PcbHAqbk9DtDfERVL+SSmT4XS70eLxsKrTRDAmoyEyOgZJkU5p4UjAmGDzV/8eN951Dxrrak3tSUqZzNjPJQVbLHlApAJcfajcJ0tIZiBQG8NdLuP98qUmvkvazEwUbMqHh+O1kHFA9z4ouDcCAgMRPSjOpz09evQEAJxsPC7V6KHtrFSJN6/JwYhxExDmdKL/sHhTujtp1Jw43gBHVJThvesXJRXl48NIaTMz8XlCGj6JyMDNH9+IAb1PmzSH4vsCR+YD095UXhiFqwvKeFFQ+BqwkvQnHY3ZK5+R6qBkeKs067qObZvf9VkdFxdsQcL4CaitqsQHeWt9iLMyzkPazEz8dd1a3HjXPWzi5VfnMi9PccEWjP/hj+CMcePBhYvwq2lTGe/E5TYMId64ou/smob9xTvQ3tqGyKhoZL+SgyWPZrJJ3BXjRmiEg4WVCGQ08ARcwLdwJIVpZk0aZwqx3XrvDJxtb0fPwEAA5npHvDaLXdOYCvCNd93DsoREo0MsuniquZkRhpNTp6Cq/KCUsyP2JWUq8WEx3hBJm5mJEC0UdVWVzEiUcX78CflVlO3GSw89wEJKfLo7idrtKipkpGEib8uOF6JpmJ6VjQJvYc3Na3Iwq+w0AODIf8XgOlQxA+a6MMMLQzW1FBSuBijjRUHhK0BKyLUBreeASauMVSpfKFH0kAR71WF1XUf1Id/souTUKQiNcKDs009Y+IFXmbUKBfATPk3I9L1I1E1Nn4aHX1iMngEBADq8NpSpQpM4b1yRl2FUymQMGB6PgMBAtLW2wtEvChleg2RX4VafSssiH6OzQoj8/3yIjQySxrpafLFvLwDD0Pn+zbfgtafnmTxTRFouNOmf+BoHYtHFkZNS0KNHDzjdbmRkZWPUxBR8efAA2tvacOb0KUZeFtsrQlQl5v/yx+isH3hUlO3GuuVLkTFnrml/ImPzGjNEpPYIRF4WfoJBJF63fKnp3tz4Zjge/dlPcM+JFegXcs7YyWaERI896xsCVVC4ElDGi4LCReLki0BIIMxUdxvQGPo9vHj6EeysMibF6VnZGJUyGZ+X7PThNlB6a7du3XBLxgxED4qzdO0TMubMhTPGjYw5c02hktT0aRgxbgI+Ly1hnoUQTcMor6cAgNz7M2cugoKDcba9HZ6mJqz99YvsO16/RPQi0GXXVlWi7NNPmF5Lb7sdg0cmsRATeS2sjAYeoggeQcz0oTAYcWViExIx/PqxCAoORsacuSxDS/RMUX+J/SqWBEibmYk+YeGw2WysflLR+5uRMH4CgoKDkXrHNAR4vT6i9s3E29IwclIKQsMjcN3QoWiqrWWGK399Yoq6WEuqs2wkErDjU8xln4n9Kt7bAcPjDaPTG2Kia6oo243NhYMRMqcId/89FaE92tiz7gwBLixVXhiFKw9lvCgodBGjY4D3fuY1XADmVj92uhteHbDWO9F3iMfRJB8Z48bgkUkAOia8dcuWIMLlQr8Bg+BpasTE2w3vQf7q3zNDRpx0aMUtElQz5sxFuMuFc2fbmeehVMKJISOEQjYfbVjPQkwy3giFH4yQR4cxRCJxxQVbkDo1HbdMnwHN4UB1RbnpnKJHY9joMUyJ1yojhs+0IfAFFcmjQ9tmZGXj2JdH0Cc8gnmk+H7zV+SRN/JI+4YmdtJPISOJT8um+8OD3o+alAoAiBowCNcNHiqt8C2mqIdomg/vyB/8ealkHhtZH/DFNvlQF4EM5Q9u34N/PjUFv55Qhe7dOo55/XVA/XPAra8rLozClYEyXhQUuoDKp4HoUO8bIjQC+NOBMOwYsQgTuQmQULgpHzGxcT4THk0ax2tr4R48FMdrj8F1XX8EBQfj7p/PQbdu3RCiaT6puMmpU0yTGx2HN0JIOp4XdANgWvWP4jJWxBRjgqgdwxtDfBiHyLNNdbUsfZfXhbFrGkI0DSlT001KvHybxPM6+kWhuaEBIZqGQi8ng7gbvKAeAFNoS2aAURaQ2J982jhdFxOHk2RA8Zo3PBlXJNje/fMnkT7rMZQUbsWw5NFwun3rSPEGlkkjxxt681f1WtyfbwOf6cXvK4avZLpBYt+tW77UKHIZGAh9yjz0fPJxn2y6iGCDC/PsX5UujMLlhzJeFBT84N5kYM3dQM/u5s8PIg577vgYp47WoNjrCRFXveTKjx4UJw0x8BNnTNxgPPzCYgQFBwPo4MSI+wC+XA3iuRBECXwx9JIwfgLCIp0+2VE8eO8AqbyKpNjigi0Y4T2Wjo5aSOTNCNE0plND11p9qBy3TJ8Bl2RSp/MNGz0GjXW1GDUxxXQNPCGW2nWgZCfOeDyWHJFgu90Qq3O5TB4fUexN9PhYhW7E+0ChQaqAHT0oDgGBgYgdkQAt0gnNK5InyybieUTkVQu22zFqYorUC2NVKJIyqSj1e8S4Cfhi/15mTIrZRf5KP/DneO3peZgxbz67V2N/txtv3wejzADAPI/P3QI89m+KC6NwedGt800UFL6b+DDTqP0iGi6rPwX+/c1gQ7QtyuAZEFeAvA6kfSJLZ6bP31q2hO1XkL8RT98zDY21tazooGwfMVQgfiYjCdOElTbTmAzfXPwi6qqrsL94B2urFarKD8LT3GyoAc/MNB0vOXUKXnrwAdRXV0FzOJAxZy4m3paGgcPjsauoEDbAyPI5WsOuNXpQHMIjnRiaPIZppvAggw8AThxvwHVDh+KZ3DwAxiRbVX4QAJA6NR0jJ6agrrISzz9wr6WXosXjga7r6BMewfpABvL4yIok+runFBqkv8UFW9BwtAYfbViPAyU7mTYPHWN6VjYm3p6G7FdyTP1O193i8bDnSmwrfx9N7fb28brlS9FQU4MwpxOjUiaz7chTQ4ZM0fubsW75UhS9v5lxpKgt/DmSU6cgLNKJIUmj2bF+/Eeg21ygxeaNnXotYGeIUWLg1Tul3aug8I1DeV4UFCQ49aIh1MXjxBkj/bk5rKP2DJ8W+9DCRRgYPwI9ehp6HlQpWSySKCNR0op37a9flNYTku0jCx9kv5KDyKho1NdUs8+J6BriFZSjiTJ91mNS8ingG2oI8VYxXpCbx0I51MbGuloEBgfjzCkPThxvQKjDwUIfHo43Q54f3lsjgo5p1zRokU6EOV3oP3Q4q3NE7bLSuRGxbtkS2ACEuVywASxDR/SgiJ4Yf30BdHiZSDOHJ0qTtw0wNGN+Ov9ZVOzZzVKZG2pqfCp7i5wVUaBOtg39T2HCqvKDWPJoJh5cuAiuGLc0m0ksfklGIPVD9aFytLe1ofpQOXZu+ViqEwQAI98cir/+QsOg5kLmgaESA4Mdxu9EQeFSQhkvCgocRscAnz4O9BC8Lby0/+z/9A3hzF65CkOTxwAAWltaTJOQrEiiiM6ycQCYUqVl/A7yHlDqMk2OJHo38fY0xF8/Fvmrf4+GozXMyJF5GvgMHDF1mr8OSl+22WzoExaOksKtPjV2UqamM7Lviicfx9rFLyJjzlxmBMmKNpInoJfdbgoL+TMyZKBrp3uQMjWdhZ344/kj+vIhNAotiWEXvmDirqJC2DUNUQMGAQBCIyJgA3zIzHwfiefvLCMLMJ6HGb+cjx49eyJEMwhZNhgp76EOh2V1a774ZYk3m4racuNd9yAoOBg33nUP/vzbldKaUtSWW36TiLSZq3D/0SwkOtrYdzcMBhbcpHgwCpcWynhRUPDi3mTgD/8B9BCCqas/NVfb5SdQnv/hcrtZLR0+5dnldqPfgEGWPA/+mHxFYlHe3SpVWnYM0XtTXLAF/3ZbGoKCg3FLxgx8sX8vmupqcVpSP4QEz3gCMr/Cl638eY4OeVr6Dx2G/sOG4+iRw2g4WsMMBjGtl3gjLrcbtZWV7Pie5mafGj6dVWKWGTV82rcNvmnjsn1FY5Kvgt1QU2PyuvF9z/NKPi8twflzZ1FbVWkiM19s/SGr/SjLDAAaa2sxcHg8tEgnAKDtzBmT54Xfv/pQOc62teH4sRqWCk/HFDPaZMrHYubWCvhWT3/uFuA/RqrSAgqXDsp4UVCAYbjk3gN0oxlOB86eByasMqeCipMIr7Q6b9pUn+NSjZ0hSaNhDxttChWIx+NX8PzKnibEksKtSL1jGquiLDuXmMVCJNXk1Cmw2YDz58+j7UwLRqVMRlN9HQZHOHw8QrLwREXZblZJmkIm1Af0ORF1KXPm7NmzAICIvv3Qo2dPZqyIhlovr+ETGePG4FFJRkp1RblPFWoZyBvVVFfrk47Og/cGyUTbKB2bKtWKoTFZv1DKttj3ZND1sttRV1WJMx6PD5lZ1k5RkZdPEZdV5V63fClm/HI+6qorAQDDkkbjZFMTzp09Cy3S7HnhjbFho8cgICgIYc6+cLn7s5AcAJ96TPx+dk0zkZN59H3O8LY8d0vHZ/F9gTMvAw9tAPKKfW6JgsLXwiU1XiZNmoS5c+di9OjRiIqKwh133IFNmzZZbp+amoqtW7f6fN63b1/U1tZewpYqfJdhSoP2EhAPhaUg7sFCAPJJBTBzJUTZeR6b1+TA6XbDGe1m5M2uHE9UhW04WiOtm0MQDStRZ+X7N9+CwKAgBHozmuqrKrHHW6yQ399qxU5pxyPGTYDmcLA20+efl+xkJNDk1Ck4c/oUUu+YhoJ3N6JX7xCf8BSFk0qLCrFt87sdFbX7RaGxrhYnjjewVGMAUo8FeaN6BASYwh+8no0NkBax5O8PADjdbgxLHgNd16UhPj6k9dTruQhzOn1SsGk7ClXZbDbouu6TdcV77WT3i+f+WHF8eENjQW4edADlZaVYt2wJMrKyWdFMMeOouGAL7p83Hycaj6OptlbaFupDPnz4xJJlLHNLtj2FiZ79gXcRACCwh0F6v+N7BtlXQeGbwiU1Xnr37o3S0lL84Q9/QH5+fpf3GzJkCE6ePMne19XVXYrmKSjg1Ts5w8WLLyJuwIY+v0BsQrNpUgnRNADmWj80MYiF/3hUlO1Gi8eDMKcTKVPTjYKGFrLxYlXnirLdLH24pHArklImS4mYgJxkyyvHBgQGormhHvVVlTgiqegsS6MFOsIslFpdV12JPds7jB76vMXLTeENs4DAQPTqHdJpqjiFRqoqylFdYZRL4FONQzRNqtfChzlEj8EoC16HCDJKFuTmQdd1NNXVSQ063pgMczphs9ks082p/3t7xe74cgnU58/k5pnCZWdOn2JkWd5Q8nSi+wJ0CAfyHCUqoCk+j3yBzeAQw+vF6+Gs83rSXG43hiSNxoGSnZielY2ogbEAjKrbVhytdbWJOBWQ6cODuTPRqLquVHkVvilcUuPlgw8+wAcffHDR+9XV1eHEiROXoEUKCgZGxwDbn4BJNRQ24FD/n+Cjvr/AkH5RmJ6VzbJlALDBvKRwq8+kT4X/rCZIG/dXJJ2KxoqoHUIckaSUyVIJeF79lSZpmoyorRTK6D8snkn4d6b8unmNoT8zbPQYFBdsYWnT4kS6v3gHhl8/FvuLd/it5yOeRzwOz7WhbcJdLowYPwEnG4+b+pEghjn4cxBHh2obieEu0XvAGwAATAq8ZNCR+q6nqRE1Rw7jrWVLfM4NgKnYim3lISowt7e1ISg4GOmzHkNtZaWJWOwPMi+ILHtJ9PCRLsz0rGwMGB7P+pe2bT7eAMAoshkW6QRsNrS1tmLt4hdZynpxwRYsyM2DDWDhw4m3peHN94HHjj1uKlp6/XWGl5OI7woKXwdXJedl165dCAwMxJ49e7Bw4UJ88sknltsGBAQgMDCQvbcL4l4KCiJevdNI6QTA+C2eNuC5f8ZA7/FvKF5nkDFdbrfJiCC9kd5CJeOuZMAUbMpHdGwcq+Jr1zRkZGWzVS4/4ff3TiQ0ufFGRerUdIRoGlLTpzEDg0InN951DyvKOGB4PCOoAh2hDC3SYWlkiZwZoCODiQwmGcmVz1ARawvR9nymFE18IvjrTJuZibeWLcFTb+Qi3OlCjx49WWjJX3kB/lqIlyHjEckE5niI3iEy6FxuN4Z6Q0t12wovioTLIzYhETYAB0p2wgbgi317sb94B9JnPYbAoCDMmDcfYZFOFqIj7ouMOCvzgojEZpmHj6p79x8WD80RifrqKpMxxgsC3j9vPrRIJ/Z+9il7NkwqyzBziTavycGKMmBPlsF9IUSHAk2LgLAFXeomBQVLXFXGy9GjRzFr1izs2LEDgYGBePDBB7F161aMHTsWJSUl0n1+9atfYeHChZe3oQrXLEzVoL0z+zmbDQtDfofNh3JQsckY8KvKD+KpN3IBdBgRZIDoMGesXGwxveTUKab0afIShGganliyHOFOFy5cuIADxTtMx3C53fje2PHoGRiIAcPjoUU4kOD1SpxsbERTXS2bSLUIB+qEyUg0sroy6Vp5T/gJk4ijjXW1rE9E8JlS+3fukIYcTCE4r9JtfVUlwp0uHD1yyERoFosZAtYZR1ZZWHRf+w+LZ/vKvEZ0rIL8jYYBq+toru8ILfFGEHkfaD9Zejuvx9PW2oqAwEAm5Ldzy8csZBMW6cTJpuM4d7adKe8S+HZaVS4n+LvP0bFx0BwO1NdUmxR9RSOWL7LJg8JjfNYSf09HLDOIvAtv6TCkQ4MMQbuFf1Pp1ApfHVeV8fL555/j888/Z+8//fRTxMbGYvbs2ZgxY4Z0n5dffhnLly9n7+12O6qrq6XbKny3YVoFej0uu22JeMeRjYlCJkfazExjUK+uwjqvAUAGSJVQgFCELM2V558A8Kk/RJL+NKF269aNaW3QpNre1oaAoCAAxqR+rr0dzugYaJFO1FdXYfDIJNMEI05WYlFCWQZLZ4UMrbxNxKEQs6kIxE35aMN6DE0eY+oLHjKlW/GaABjFDC3KJdBnPGmX+D186GbdsiUsM0rWb7LrIGOQ7lNsQqLJU8bzhpY8milNb5+elY3I6BicbWtDQFAg2lvbmKIuEYpT06chzOlCa0sLrhs8VPrMUbhxpDfUFhM3GNmv5Ej5PzJjkQ93ympmyVLuY+IGM4E76ovOajEt+hD4v/3A3x8HArz6STabkZk0sp8i8ip8NVxVxosM//jHPzBx4kTL79vb29He3n4ZW6RwLeLDTMFwAXCoxzCMmr0bsQlLfNJniST70Yb1PmJlsoHaXwaRTDdFDFNsXpODEeMmIMzpxLmzZ9Ha0oJGryeFVrcRrr7QnE4cO3IYry80/O40ORduymchJTqvFYgPc8CbHcRftzj5mnRPhMrHsrCElU5NQf5GVJUfZF6Vovc2+3h/ADAvwjpOF4UHb0zxyrbi/aE2WwkEsiKMy5ciZWo67F3oNzo/XzsKgI+RS1wSOv6MX85HyykPFm/chBaPh6WGH967B7WVlag+VI4b77rHxG+RGcr8MydT5s1+JcdkKPFGCK+wbDKotxWatHR8aiX1i4Kdq1H1/ZtvQWCvXgCMMBHgq5vD9y+1L2VmJr63JgdlGbsR1BPM+rsz0fhtKkVehYvFVW+8jBo1CkePHr3SzVC4hmFKhfaitYeGtdrTiE1YInWp0+RBVZB5zoFouKSmTzMq8AYZ3CtxEuX/WqWkps3MRF11JcKcTpR9UsQmSFqNn/LWFyp6f7NpkuCNoOTUKZh4u1Fb6LAkm4jQWUFGfvLlJ0mazGTeFTIqTPwSrxItGTvkVTnRYKRAE4do5MSUjgmSM2wAa6+BeD76XjSorOTt+eP6y8yx6iP+L/1PbV7yaCa7xzFxgxFst7PsJF3XcaBkJ+qrq/CXvLUoyN/IxO8y5sxFVfnBLvGoRH0gwFdgjoxUp9uNcKeL8X1ONTczIrJI3Oa9MSTEF6JpcERFob21DYFBQWiqr8MX+/b6ELp58NlLfHbdfACPRf8dg774EzNgbhhs8NB4IUgFhc5wyVOl4+I6NCkGDhyIkSNHorGxEZWVlXjppZcQHR2N+++/HwDwi1/8AocPH8a//vUvBAUF4cEHH8QNN9yAH/zgB5eymQrfYshSof96xI6/xb2AzQJZ1sqTkJw6helt0He8AZIxZy6CgoNZWQDRK8FnFIkiZbEJiZj/Ri60SCc+L9mJbZvf9ZkUqR2ySUKUsR82egyc0TEIcxrKq2LGC2C4+z3eCUxG9OQnX371TmnMMiOMQjS97HaUFhWiYFM+ZvxyPiKjY5CRlY1FD9xrCvkQybOpvg67igw9HVlhRNGjQ+8pdFEgEZPj+91K3p4MG6fbDQAmzogs3CT2tVit20oFOPuVHAQFB+PC+fP48uABo/AmgCEjk0yifcNGj0FkVDSyX8kxebZiExLxjFfDhW+L7LmlzCsK8ZHqcdSAgbCHhaOpro4pDJda1IYSjSY6Hnm5xHRvkdBNsMquAwDMzIS75CM8OapDAmPWeKDmpOLAKHQdl9R4GTNmjEl0bsWKFQCA3Nxc/PSnP0W/fv1w3XXXse8DAgKwbNkyREdHo6WlBbt378ZNN90kFa5TUOgMYqYDAHx8EPhhjgeAvCAeQaZWS2nTZGxQlsq65Utx/7z5qK2qtGyLTNuEPqc01NMej2V9G57wy5MnxUlsyaOZrH1WHhYx9MJfOz9x8+0mr0vD0RrT53TejKxsJHkVZw8U70By6hTUez1JNsl5ARhy9lwRR5lnS+ZhAWDKcLEKV/jzWBQXbEF0bByc0TGAzcbSvKlPZeEmumYiC8s4QyLWLV/KvHJfHDhgGfJa8mgmC9fwnq3p3n7VdZ21hwxVKpb58sZNOOPxsPAPtXOXV/zP6XZjWFg4bDbD2OOJzmTo8IaJvwKgvIEiq2BO/Sueh+dUAcDvd/fFXQPrTAsLUudVBoxCV3BJjZeCggLYbNZO6p/+9Kem90uXLsXSpUsvZZMUviMwcVwAXNCBsb81S/0D8lo5sgmQz4QhrwbLQuJEvzKysqWTsFUYgA9trJNohoiqt6LBIhpfFWW78eKDD0gzQ0TIrv3BhYtMKrO85yNjzlwWThLPy//KSbeElHOtjKM3F7/IUnEvNuuJBOB4FVmg8wKX9H3C+AnQIp04296OHgEBcEa7McRbXsAq3CSShZsbGiyJxwTi+tA9lGmiUFt5jxdBB5hwHnlNyHiyAawopq7rzJATn7XU9GkYMCweoRGOjvR6r+cKMMJ2Yho53St/9yVtZiZGTkpBQ02NT/921v8A4H7ycTQtMrKPCM/dAqQOUhwYhc5x1XNeFBQuFm/f502H9mYU6Trw3N+AkOunYfWf5FWZ+YHaSv+DXORNdbWMU0KgsA6f0moVOhKNIqvQBp1TnAysvEX8uTrjbogZJfSXKiGfbWuThsAy5sxFccEWn8/5LJxCiZosfz2iV0PkrVhdE99/ix64l+1HkzbvCSCvBpFh6Z5Tn/UfOhRhThdqDlfgiwMHfDJrZGJ8vOcoxsvP2bO9WZp2LquULWqi8KFAK0NBFM7jw327igqxq3CrT+VtwDCwpmdl461lS5CcOgUBQYE40dCAEK++UJK3dMH+4h0mHRsZeRvwDZXSZyIPig/HPZOb51NYU3x2wxYYv9c7OV624sAodAXKeFH4VmHBTd6B0Lus1G3A939jeFxWbzPSVmfMm2+aXFmGhVdbhDwqOsxGTXHBFoyalIqAwCDGJ6DvnW43IqNj0FhX68MloG1IoZf0QPiUU6sQh9PtxqkTJ0wVqXnDgSfI8tlBPGRaI2I4iP5+XlqC1pbTWLd8qc+xxGrQ/LVtXpPDvAlV5QelqreA3KvRmTFGkKnE8vvxacPkCSPV2vu5e77iycfx8kajxtqZlhafkEhnxpTIC5K1zcro5DVRyBi+9d4ZGDAsnhG+ZcRkghh2Sk6d4vP8pM3M9BGOI3VlqiXVVF+HcC60KLtnvNeND5XyWjZiP1A21iiLsJ54PbEJifj76Ey07VmAnwxtYp/PGg/cPASIW+zT/QoKAJTxovAtwl8et+OWAR6vrx2ADfhNiRPNYX0x+z8z8dGG9bjxrnvQVFfrM9GIZFFa7fKTEAD0CQ8HAAxLHoPpWdksQ+ZEgyGlfkao78NzEEhunfRArHgbhLSZmSwsYA8Lk2b5WGUH8SCtkYdfWOyTzUKqvb3sdlRXlLMaPAB8lGxlRobYP/4m/diERGRkZQMwk09lE5o/sTl+opRlF4VoGqsKTcUhz7ScxqS0O5AwfgJefPABnPF4oOs6zng8Pu3sijHl79yU3i47BhkdfCgwYfwEE+G7s3CNiQd0u6/hKhpJZFCQujJ5ydJmZpqI6DKuiyxUKhpmog6QeH5/oGMVAfjL2sex9icd3w2KMBYjigOjIIMyXhS+FXj7PhiGC8DCRXt7jEbTzbPx4J12DE0ajV2FW5l8Pk0eRFYEzHwUGsirD5WzYnk7t3wMp9uNfv0H4ugXh9HLbkdkdAya6uuw1svfID4BX0SQNxR417w4wIuTFk0CVNzPaiLgs4Bk+GjDetz98zkIDAoycXJIDG349WMRGBSEbe9tYh4LngPBG2My4ylE02DXNBRsyvfRE+Gvi6o2A8AAP+ncsiKBnU2CdA4AGDUxBQOGx6OuqhJRAwYhsFcv9LaHwmazIcxb6JEyrmTHFTN9KCxj5SGjfSi9PSY2Dkse9TUkRW8XAHYPSMCP9VPSaIwYPwEvPfgA21dmzMnS1yvKfOsq8R4v8Rqt+oHO6+gXhZONjTh39iwKN+Wb6hqRwSIaNOL5ZRlcIuG3osyoPs2HkJSQnYIVlPGicM3jsyeMom/kbTl/Hvh/Pf4drWk5SALQ5K1KLg7cVitXHnzdnj//diV+NW2qKQwEGHVp+AKB5IrnayDJsjXEAV6cAGgSooGfaiH5CIoJ2igEaqdd02CzAfU11SbSpw5gxNjxCAgKQmtLC/PC7CoqNPFW/JEwK8rMFYx5ATcxS+fE8Qboug5d1xHmdEGLdPqIx1H/2WBkI03PyvbRgZG1g87xeWkJ2tvaEB7pRLjXWwAAPXr2xP7iHcwz1pm3h/qWspF4QqyVZ6S4YAu+f/MtiIyOwQvrN+K1p+eZ7ndxwRYkjJ+AllMeHCwtYWRjenYoXNXc0ABwhha892zE+AmmauAVZbuZCJ6TCyvKrgkA6z++zzvjR/GG9qiJKaYaV3x4rTNvFfM0ehM4Fj1wrynMR33+99GZSKifjcHaWbbvnYnAvclAXrFlMxW+g1DGi8I1jXuTvYYLANiAM2eB3r8CYhOqMD1iqw+BVKaLIhNeo+1KCrci9Y5p+GjDevYdn/ZcUrgVwXY7S1cl3ZGSwq0+2SH+smkA/3WErFJ3ZdooYjt3FRWidFsh6wsifRJps7WlBa89PQ8pU9MxyluJmibd7FdyTNWqu1LBWPa/XdOYsiyFIJrqOuoDidwZUnTVAZQWFXqLBxphD1lxRjqOXdMQEBSIpvo61Ho9L33Cw9G7Tx+c8XjwvNcYFPcXvWWMnO09Dhk9vAKteF+TU6cgIDAQFy5cQFBwsKkkAGBwhsIinQhzunD6xAkE9+kDHR1GrBjK43lBpN9DBhl5pEI0DWFOJ8KcTtMzzLeTIHtWuhqikqXVi1lN/oyg4oItGHfLDxEQGMiMexlnaeJtaXjtfeCeisc7ftcA3vwJMDBchZAUOqCMF4VrFrIii/mB9wH4I8viEQdnkYxJqqS8+Bs/8DccrUFAYCCiB3WILfKDLq0oKV2VVrekUGolBieD1QSweU0OXG43ImPcpnaKk4dsP76dvEckNiERwXY7XDFuvLn4RRTkb0Tq1HTWlbEJiXj0pSUICArCrdNn4GdjkwH4kln5EJwVAZlCKiMnpmBXUSGO7NuLYLsdp718E96DRF4wngiaNjMTWqQD7a1t+GjDeiNlu5/cA0PGI596/tTruQhzOqUFGGl/XlSN+oxPYafrmb1yFSKjonG2vR2DRiTCHqax41B/8xlO4v0Yd8sPERAUhIBeQabzidlfYh+Sfk+Y04WBw+ORkZXNyLdkKItcJHoORYNDJPdaPZuyjCkeXclqIySnTkGPnj1NxUJlvCH6u6LMnIVkA7DwB0aNJFHuQOG7CWW8KFyTuDfZ13DZGzkVz/5XqWk7WYYKz8tInZqOyOgYpE5NZ6vkDG/hvKa6OiaRzk8A/KBLxwtzuRAaHsFIwZFR0Xj4hcV47el5qCo/6LNq5dGV1W9tZSUGc6qsXRVj493+vCZJRdlu1FVWYsioJGTMmcv2KyncisJN+ch+JYcVgeTF9/gJRqyDExkdA0BOQBYNLTKCqD4Qyc/zXjCf0gRRUaxv21pbTSnGvGCcjg4Ru0UP3IuXHnqAGUdkIPBqxSScVlK4lRk8IoeFnoPqQ+Voa21FYK9e6BkYiPrqKh8CcWxCIoYmj0Hq1HRGkKbvX3kqm/FbogfF+Xgd+OyvJY9msu82r8nBiw8+wPrbBkhrHvH8maL3zN8TH4s4VAWb8n2eCx6iKJ8/z2Fnz2NxwRaMGDeBVR7n9xEF8sgz9ss1ORgVvRuDIoztbTajuGPQPGkTFL5jUMaLwjWJNXdzb3TgnZBZ2K1PAmAWeuNd8c/k5iHYbkdkjBuawwEbwCoC9x8ezzgDtBqur6601CshkIeHpSwnj0FVRTk0RyQLHcyaNM4nJMWTFzvT0+CvQ5zsgM7FwACY6tjwkwZNluTNoDRjKz0bMU2b9zKlTE1nHgCr6tQE3oDgCaCioUjnJE8ME8vzpmwTwbj5eAMjfj60cBFsNht6e8NUYmozhYRIrdiKN1RcsAVjb/khIqNjMOOX842w1dEaBAQForm+Dof37UXhpnxMz8pmniS6n0kpk2GDwTMRjZBZk8b53C++ECgZaC+s34gv9u/FYK94nihkJ3smKbPsxrvuMZ2HDM3I6BgmHEqhOLG+EYE3fDvzHHbmwcmYMxdhTifCXS5WLsKKwMwfK+7Jx3HsWcAZYhwroDtw8gWgz9PSZih8h6CMF4VrDnuygJ7dO94ftMWhtNckjJqYYqkr8UxuHpOvB4CmOqPGjOZwoO3MGWgOB1vxUyaKmEYq00shiMTGg6UlCHO6WOhArDpNHBYCrX6tJgF/qbmp6dNMkz6fhivjKNA5QjQN1RXlaKyrhQ0wZSyRN2ng8HjExA2WEkHFLC2+T8iYsypoSQZSytR00/d8bR6ZMVNVftBHW4SqcR/ZtxcVZbtNOj2y/uMNWjJcZLwh4rDYbDbUVVdiz/ZPTHwUMlRGcWHDmNg4fLRhPZrq6tCjZ0/mSQJ808j5Z4IvBEolBYKCgxEZ42YicnxmXEzcYB+C8fSsbJw55cGJxuM4c+oUFuTmsZAXhZHOtbejZ6ChJ0PXZOUR9Md3EdvPZ52JxGEyFs+2tUk5L+J9Fv/2fQ5oeA4IDzb2CwkEyucpDZjvOpTxonBNQZT9P2ZzYs8dH8Mm1AwSXdLBdjt0Xcc5rxw8eRRsgPEdYBp4ZWmktKqd8cv5JlVanhMgCoiRV0H0rhCfItjekcbN669YTSgEMazBEzOtCk2St6fQGy4YONyQjG84WsO8LjTp6AAGDItn180bJmI4QQbeCJp4WxqrbCwq3dL3fGjCtCL38mCqKsoxamIKvn/zLchf/XtTP3yxfy+roRSbkAhntBu6rsMGX60a2oeyZYjEKmabURrvgZKdPmnS5O2hCd2uaYZHL9rNDBDN4UBpUSG7Jln1ZfGZoOtNTp2C156eZzKUTdW6OVFBCi/xBnF9dRWuGzoM1w0dxjhO1N/9hw5F/2HxONnYiA/y1jJNH3/gn20yoFKmpmPA8HhoEQ72rFlV5+aNFPLO8b8zwJyFJ+PSOJ4FTr0EBAcA0IFBDuDvc4Lxb8tbOm2/wrcTynhRuGYwZ2oMbhhSxZbUe48Bjx2ahIyxNSjYlG9atZMHhFblB0p2mqTUaUVKRNKGozVwjEwyDbziIErk3kavyB3VxxFVR2WS97ynBAArCbAgNw+Abxq3DDIdGEAebhEzSijtV4eRsWQDEBbpRFN9HQtVVB8qZ5PTQG84rb21VUo8BczhBDHcRdeRmj4N0bFxLOuHDCF+QtQBNhEmjJ+A2qpKVpmaQkTVFeVMLTd91mMICOxQo+U1W4jcW19dhdMej5RcLOs/MRTDp/F2xt+heynjcPDHc/SLQurUdPYd760AwIyQM6dP4dGXluDokUOmkBqFlYYlj0Evux09AwJMtabIIC7clO+TrUT9vSA3D9cNHY5zZ88idWq6T2FJkSslE1skfpMNQB3H+eH7kzca+WeaDL/O0s5lWBC8CsvOPc44buOjWvD2fUoD5rsKZbwoXBP4MBOG4QIANqDtnA0jlul4JtdMuOXTmIve3wyn240wpxNnPB5mqPAeBpkBYDUJkR4HL7qmRTrRy243paXKpOt5Twmf7izWrvEHfvIgTwEdh680LfJrqG1NXp4GETeNon/A0GRjFX/r9BnQIp1IGD8BoQ4H2ltbUXPkEMtC4vuCPDm7LMJd1EchmuZTdgAwlyw41dwMzeHAufazCHO6EOZ0obhgi6moIXkvZGRXHjLPFXnUZDo6FFIUa1lZTcR0L0VVWXGS9hdW5NsgeitWPPk4/vSvgwgICkL/YfEslElhpehBcew5bj7egF3bCn2MKNn5CeuWLWGeuiOAj7eSL5Nh1zQkpU7BuFt+iFeeypZ6UN6SKCVTfzrdbtRVVvqQhvm/fB0xfgFgRfo99KMUDGoqZEKUdyaqOkjfVSjjReGqB8ss8orQAcAnQxYgNuFd9PeGNshzIa6KxZUkvw0A6YRD4nXfv/kWv/Vm+EwXx6gkE29C5kERB27ZqtNfgT8+tCDq0qTNzJQK7vEps/zn/CRGmStkjNVWVeJsezsio2MwYFg8BgwdjsSJKYyrQKRK3hC0IhSXcuE8fkKiEFzGnLmM0Mor8NK5+L5OTp0iFRMUC2ny/WIDMCplMoK9GTZUtJGv7uyPI+OvztHFZONQH2VkZZuye2TGVv7q32PaI0/g6JFDUhVowCgpEB7pxBdeno8/iM+UFelX1A2iUg4BQUE+xHMr44jvx379B2Jo0mimncP3q7gtWwBYiBAChvFcOjkNEe9+D6G6hxkwmeOU8fJdhDJeFK5qjI4Bcu/hPtCBuuCheP6P/2IhguZ6Q0FXjKPLjIOu6FIwI6Gfkb5bfajcNNnJVuBdEaGjCYzaxAujFXpDJI11tRjCZZeI3gKRrEogdVdRcM+fMUdhMD7kxh97elY2etntGBQ/gqnwWk26Vpod5L0S8dGG9Uif9Rg+2rDe1I+8VgwPfxlZMuODroWE8foNGAR7WBgOlOxExpy5prTudcuWINhux8Dh8UhNn2aamP1xkERviiy9WTQORgmeP5mR++ffrsSff7sSAAyiecpkViqA9ju8by+0SKfpmq0g89hRH/LeRfFZWbdsCcJdLkQNGISPNqzvtHI2gbyJTrcbfcLDmSEqA21r1zQ4RnYsAPyJIf727z/A/9y0B84zBwCbYcM0LTIqVCt8d6CMF4WrGp8+DnTjRr+PDwI35RwAcIDxAYjfwhdTlIUyePiLtVeUGbLrj7y0BIG9euGWjBkIczoB+GqYXEzMHjBPwrwwGnkiegYESLNLRCNBJE8OHG5UJZYp0Pp4aLgUar5KtLgthSH4EI9s0pWF2QgpU9MxSuBVAED0oDgf8b+KMt+aPASeH0ICbXRMWeiN9/xs2/wuXG437GGj0a//QPQJD8fJpkYc8k7iFWW7Ee4NV2XMmWsKVcmMC74/6F7EeMUJrTKL+HvoL0QoPk86YCoV8FXCjTKPHXmrRoyfAC3CwQwb0cP45YEDuG7wUEQPikP0oDjTdYmqxPx+FFbktXNkvxWrBYDM48V7w/5+ywrc8L/fR2j7MQBAaBAU/+U7BmW8KFy1+OwJoAeXEv2lpzv+N/43SE03rxYfXLgIzQ0NjHjZlXornemkpExNR6BXpK2uuhJH9u+FDb4cCNlx/BE3xTbx2Siy7BKr9lmRJ8kzY7Uvf35ZBowMfN0mEaI6LvF+7JqGUSmT4WlqYrwYq3Z0BRVlHTWUKBTF34vOSLU8T6lPeDgOle02cSzIC/XRhvVMkXfEuAl46aEHfLwMfMiLjD6ZR4w3QGUibLL7Lt47PiOOL3hpxbuheyKSp5c8monpWdnsGOSxqa+qxLn2dhPxVyQ12zUNLreblWqgaxRViXnwYUXACMU63W4MTRpt4hbx95e/h3xBVBF0/uw1Thy47xi6dYPiv3wHoYwXhasSb9/nrVnkHRlbzwJrBr2JSSmTMfaWH6Jnz55s22FJowGbDSlT09kqz2rVTBBVVgk08Afb7YCuo6m+Dm8sXMAmBlkWEZ2PIIpviZwIUVaf/lZXlCPFq8rqjzDKt/+jDesxNHmMKaTFt4lXXCWSKwvRcGJvXVHtlWG6V43Y09SEprpaVHm1YjKysmGz2dAnLAw2wGTMyYy6rkAMRZFXR8Y54Tkm1DdidotoKJDRGOZ0GhWouXpB/LZk6PBZWOKzxvNl+JRv0bgcNnqMyRCS3XdeBFEsZCmrwG3XNIyclIIR4yfgi3172XPBE8bJc0O6NXTfZOE5HcCQpNEAgG2b3wVg/A4KvHWy+Gef+rf6UDkajtaYFhTNDQ0ADH6LLAzLHyN91mOmgqjisem3u77HSGRc+CPjv8waD/z9iCri+F2AMl4Urjrcm+ytacIt6eb3WgU7AJvNhgBOlj0mbjAr+EbpvV0J61D2Bk3cBBpodxUVonDzuz77iZOLzGtBIY5wlwvNxxsQbLebUlJjExIx/41caJFOjJqUij5hYRh/67+jZ2AgdF1nE5Q/wiiffSKGWviJNPuVHDhj3D7pxTyRl66Fl9iX9aHM6KAV/LmzZzF4ZBLjc5AR6Z1TTJO4FRG6M5hItLen4VTzCZw43oBeQv8S0mZmsjIBpHTL943M8KR7F+HqC3tYuE/Yjq7dH2GVPxYAkyow753ZvCaHfUeGEJ/SzIdkrDxVvPeDNxKaGxqMIpCRThbqJC8KpetTX1oRr/nQ267CrSz12p9nj8JRzfV1CI1wmMKI9PzYvUVBAXkpibSZmUaBy/PnceaUh3mb6LyUMp8xZy6efTQT198fh8F6h4fmD3cr4+W7AGW8KFx1yPmx9x9vdtE/A6Zg8+sdgzZfLC9tZiZ69OyJ1jNnEMqp5BKsiJ5Wk4E4SRFEVVnSyCDviqgka4Mh8qbrOo7s22sKc9g1jVVW7ta9O2CzMdXTpvo6n8mU/oqZR7L284hNSERTXS16BATgr2+tNaUXyyapXYLQn9iHoow7ANMKnrwq5DEgT0BxwRa2DU3iVpWwuwLicBDplvqXjsffKxLj4/kefD+KoRcArN6Se/AQSz6QVX+L4RrqX75GFn8sK0NIDMlYnZ/XuAHADMTSbR0FMJ1uN57JzWPbkrDhkkczTYaFGIKz+i34e/bImK2tqkTZp5/4XC95+ChxUEbOpfvrjI5B/2HxrJwAnY8vfFlRthtD/9MIMVMV6h6qhMB3Asp4Ubiq8GEmENQREcKXiEHG68fZ4CnGysVVnejC5gvPyfgwIsSMID7lmJ/Ad3G8C1pN8qtRQ0NFN4o7eg0t8qLsKipEY10twiKdqK44iNMeDyuWJ6vITAO/6IXpjIScNjOTeUPI9S47rtUkJevj1KnpPiUJ+ImJbyPQ4XHhU5wpjGVVhVoGWbpvhremEHkExHsFGMYVKQtbeZj47Ymrc6BkJzOIupLSDvimbPMaO3SuroAKRR4o2cneWxnSYlmIdcuX4tZ7Z2DA8Hi8ufhFpHhF8eCteL5u+VIf8i7vzRLvFeBbisDqN0L9zRuzMvDEbFFbhvR3WBXtyA7lZP584vM89nfAqRe9CrwwSggoAu+3G8p4UbhqsOAmrlI0AI8tBDvv+ATTHYWm1aU4kPI6IDwoNbXhaA0AM89Flj4KcAXsoqJ9UkvpGGLohFaTfBv5bBAZWTcmbjBm/HK+yYtkBdHrY8WDEd35Vl4b/lofWrgIkTFurF38IgD/cvpAh0IqacSIfBNxVS5OlHwNH39VqMXrJxItn9lCHhLKNKPnwK5pJvG8kZNS0NzQYJRDcDiwa5vZw8Tf2589/SxsNht6BQdb9iv/Ge9xE1O2ZaE5mUdPNOKI7Eqqz3zIiz837wkDwIyPngEBLHOquqIc0HW0t7WxMKkVufj7N99iulf+PG5WoSNTaK+TYo6sj7ySBHyV8M1rjCraF5M5OGfnO3ht5MfsBtyZaEgt7KyyPL3CNQxlvChcNXj2B+b3c0u/j9FjaxDucmGUN/5PGSfipCkb3KoPlUMHEBkVzVJYaQAn17QjKgrTs7JN8vKOflFoa201qi1nZbOVZIimIXVqepc8BTI3v0nPxFtllyeFWkG8NquJwV8IQLZP2sxMDPWKwmXMmYv9O3d0OkHwBlRy6hQfjZOKst2s4OG65Ut9Jkq+jSlT032yaKyun0i0wV7dlrSZhijfiYYGUzYTn+XChyDIUCJNIJHcS/1qDwsHANjDI/xmrdH/vMeN9zqIae5Wky/Py+G5KdRHGXPm+ogSWnkbycj4vLQEZ9vbsW75UlSVH4SH84SIYRwCFaHkDS3xXovGjlWmWmxCIpxuN04cbzB9z6eYUzo6q2wexVUJ5zxA/vhJfB9OvC0NDaPH4NCx32DQF39iBsyWh1X46NsKZbwoXBU49aJZz+WvX9gxOvMlOPpFofm4kaUw0OsKB3xl1mVehlGTUtG9e3e0t7b6DLw8D8HpdjPjiIimvH5KqMNhmgB5TwGvskuhiM1rOrJcCjbly+vcREWhqa4OX+zfaxnCsvK48JkcPERCKy+axnsjCJvX5MDldiMyxs0mFLumMY4E33bZ5FtVfhDTs7LR3NAAR1QUsl/JMcI5XCoxr8rqc885zwm9540QnkczclIKQsMjEBQczCZW3iNgxQXiQ0w2GJkuZCyIRGcitFJ9IH+TvZU+iZhlJHruZBwrp9sNV4zb537KKmjz5wbMSrdWyrnUzpSp6ZielS01vmUcKH+FN8mDRro4ogdwqDc7iYqTps3MxIjxExDu9QiRoWzXNMbLKi7Ygqryg+y+0vPEe1hl4I25/alTMGb/B/jJ0CYAqgL1txnKeFG44vgwsyNWDRtwCr3x6ZhXMdLrcl+3fClm/HI+wpxOlg6dkZWNg6UlPpWggQ4vg+dEMwDg0N49bHCVeS6e8RZHtME3RBIa4WBtoHou/EQiI7uKWS6iy70zjol4bL6dYiaHjOzJvEpez9GA4fEId7rQWFdr2q6ibDfmTZvK3scmJCI6No4ZaXzbZZMveTlOnTiBs+1nmYdLlkosXhOttpsbGlBaVAgbOoTdAPiEKyrKdiMpZTL6hHV4RGiyFvlGMkODDBX+XvOeACLlWtUHsspak02qvFdPpp0itq2ushJDRiaxoo0kukjHCtE0H6NDFm6ymuBZKJS7ryKvRcaB4r1q/DNMoTkyWh9cuAjhThdLx68+VI6TjY2w2WymEO2XBw8gOMSOjzasR21lJYaNNgpMDkkazYQalzyaiSWPZjIejMwrKSNGi8bcT7hHb1CECh99G6GMF4UrDsZz8XpePpr2OYKLd6DovQ5egs3WsQnPZRksVIIGzBooYgE/kcQL+GZsEEQjw8pQ4Leh848YNwF11ZX4IG+tTxaOP8Kj7Ni0L0nd85kc/HWJhFaa2MMiDXVgXqVVNhkzj1B9HeqrKqEDTP/D3yRN1aqb6usQommoKj+IWZPG+VxPavo0jBg3AZ+XluCDvLUs8+icl0gq3gPe85IyNR27Crf6eLLIy/V5aYlP+ElWIZnXJqE0ctJYEdvqTzxO1ucEvv9lXitZdo1d09B/WDy0SIcpFCYLK4nGCB/y5NtC19BYV2sU5qyrg80GS6OAbz+d51RzM6oPlePW6TPgdLtZ20dOSsGJhgZ8vquElY+gdPyGozXoE26E33gjNkTTTGq9jn5RaKyrRVNdHXr07GlKJa+qKEd1RTmKC7bgmdw8hLlc6BMWgbW/fpHxrvh7IeKd3V65BS9U+OjbB2W8KFxRVNKA4jVOjvf5HnRdx2mPx+RJCXUYNYwoawSQx+KBDpf20OQxLDRBIMOns/oy/j4HzBOXjD+gORzYs/0TVJUfROrUdMz45XxokQ7W7odfWMwUfGWaKnw6KA3WJHUvy+QQM134CUgHGFfEqsIyIA8dUD/xIaxb750BV4wbby5+kSm3UuorT54V++vhFxYjKDgY5862s4rRtMJOTp3iI/IGdGQw0T3j1X75TJWGmhoMGZlkSsluqqvFYG+dKMoi4tVd/XmIKPT16EtLcOu9MwCY1WUB/yrNvIEqasWI+5GAnOZwoL21DaEOB041N7NjhGiayeMnCzvKDCTqbyo5IYbkZARi0WgjT6UW6YQW6UT2Kzn4aMN61s6z7e3o0bMn2ltb0Vh3DL2CezPhREpRp3tLWVR2TcP+4h2MSK85HNhVVMi+m56VzXhLyalTDOPNu3rhC3n6C7f+fXQmEupnY7B2FoARPtqTBYxYJt1F4RqEMl4Urhg+ewKIDu14fyYgEn8duwEHvPVQRM4HX8NI5F3wWTsyMiU/QZNQlz+SqBVkHBdxJS4SNUd5V84krJf9Sg6CgoNNhQ4JZIScbWtj1Xz5wZqlXHs5LWTckFeml93uMymRoipNcLRSp1pBouQ8nYv/y4d7nDFuAMCjLy3BU3enM2OADCW7pG/TZmYiICgQrS0t+GjDetZGUbbeX6q2mDVF5Go+rGfXNNw/bz7CnC4EBfdm3KCUqekADB4nfwyZhwgwCkf+x8/nICAoyKh0reuoqzbHHfyRVvk+458RWd/y//Pp6NQfdO9k25IRK7tfAYFGf7+5+EWTZ0kWOqU+SRg/AWFOF+6fN99UgsOuaejvNUZuvOseBAQG4kRDAxrrao2MJoA959GD4lhoV9RyAYCRE1NYSLK6ohxF729G9aFyTHvkCQQEBuKLA/vYfasqP2hwZSKduHDhgqmQpxWo3197H1jS9ji6dwNgA+L7Wu6icA1CGS8KVwSjYzpEpQDgAoC/3b4Lvb0Ks7wuCgCTG5sf9KdnZSPJO4APGB6PtYtfNHljxFBERdlu9LLbMSx5DILtdryxcIFl+MYqtGIl6MZPVpvX5LCqzAdKdqLF42HGFV9Hhz9+bEIi+g8zVIJrjhxCcIidCXGJYTGqIE2hF/LKiAYb75Ep2JTPJtuKso5aQbJUZfGc/IT50/nPIjQiAgGBgabwg79j+nh1OEIxL1vfmQeMfybEkB15qCjbhip0p3oNFyoSKBJSZfc+elAcunXrxt63eVON+eu1Umnm75PMuJGFDXlPGU3s1B9if1mlI1v1t5VBSEY8eVscUVH4fFdH3/HHJt0aMs5TpqZjwPB4DPHqCG1ek2MSnjMI8yloqDG8K9TnpYI+ErVv9bbtCPB6Iu3hEdAiHJgxbz6O7NuLtYtfZO2LHhQn/V2mpk/DjHnzUV9Vib/krWXXGOICnrvVe9E2oOE5wPGsT3coXINQxovCFcGWh73/ePXj3+09CweLd+CMx+OziuRd3OKgr3PHDHe6MGPefGgRRniGBnSxeF2//gMBAP36D5QaHOTF6Sy00lnGBq1Et21+1zSxUOhDnIizX8mBFulAfXUVVmVnSScdSrMODApCfU01PtqwHjfedQ8KNuVL9Waof3rZ7T61jDrzHPDgDYiq8oOmVTUP/phWZQV4QjGfLsvUi4XQhVUf822KTUiEy+1G8/EGfJC31lSriTx2fPp0iKZhwPB407MiKhjbNQ3BdjtOezwo3JTPPDvkEaF2Vx8ql4Zg/Bk3Vhwa0hiqr6k2XSvfX7Jn8WLrUlGoitfsaaipYXW8RL4Q/wzEJiSaSOB0zkWCgcO3WfQWVZUfNPXJuuVL8dCzixDQKwj//PBvSEqZDGd0DMIinTjV3CzlEPHGXMacuQh3uhDudKG2spL10aIyIMzlxJNJRop8eG9g+XQn5rxV12kfKVzdUMaLwmXH2/cZMWjiuZzqEQHbzfNR9/7mTgWvAJgGfaq6OyghEX3CwlFfVYk9AplVXGk31dciNCICR784bFqBkpcCMKftygZwEbJCi5R2S/vLiJp0fD4EImaC8KDt6muq2YDOT5Bi+0h7xK5pcIxKMsnyd+Y5sAKfvSMiOXUKHFFRTCCNL6EAmBVUxfAPtUOs2mwFU6XnrGwMTR4DXdd9+kL02JGHSHM4WH/IlF5FNeeUqekYlTIZ/YfF46WHHjAJ7jn6RZkmU8B/WEkM2wFg56d7y3OyKN2bN574quApU9NNtZBE40hmEIoGBW/4+AvNPOTNLgKM0gz8s8oXihRTt2XeIr5dpz0n0Ts0FKl3TMNrT89D6tR0xnHjjyMT1Fu3fCnzvIgG9Zy36vBkUsf7nyfVYc5blo+VwjWCS2q8TJo0CXPnzsXo0aMRFRWFO+64A5s2bfK7T2pqKpYvX47vfe97qKysxAsvvIA333zzUjZT4TKDZQF4/cxrhq1Hk0CGFOvTEFmSH7ABsPRWWWaJaCRQ6mdpUSG+PHCAfU8r0F1CAbrOYus8aNAeMW4CNIexmhcnPxlRkwbwEeMn4GBpCV5fuAAxcYOZyFtnngeZgcVDJO5yUjoX5fHg7wkf5hJX+3xIq0dAgFHgUuK1EPk1fAijqa4WQcG90VRXK+XOyAxZGwDoOjxNTT78Gb4PeA8e34+zV67qtOaSDTBVm+Ynf15Mjgwyu6ZZGodiiA2Aj/Eq9i9tT6nOvJFHvBO6v+K9lRmEshRj/l6LXDJCpJfz1N7aynSR+D6iv1b9znPOiJ/06EtLsPHV3xkZS0GBSE6dwn4/osGTnDoFAUGBpnvFE7mp/Xzf/Wl/GH4yzNB+6aYr8u63AZfUeOnduzdKS0vxhz/8Afn5+Z1uP2DAAPzv//4vXnvtNUyfPh033ngj3njjDRw9ehR/+9vfLmVTFS4Tyudxb2zA4f4Z6D90OL48cEA6SVEF4qL3DJe/uEoHfAcqWZ0WMhJiYuNYiIVAq+Cu1tnhQeeOcPWFzWaDp6kRe7Z/Ip0Aqw+Vo72tDdWHyk2f0wBOA76Ywiten8w75a9tpsnv9o7JT/S88BMcCYvRvqnp0/DIS0sQEBhouboHwCaW9tY2aN6smYL8jYyPwqf78gYA8TxYCONoDQaPSjIJldG94v/y/3uam31I3SLB96tyRMj4+2L/PlZtWgylieexKnQpOyf/mcivovZS+CgyKhrZr+SgpHArUu+Yho82rMfOLR+bwoXivSXPhMwglD0rfNhT7Mu/vrXWVBiRx1vLlqCX3Y5BIxLx8sZNjFPGXwctFmwAunU3pqCegYGIHhSHp++Z5uNhEvtJ/Euihbz3SAwH7xiRiX9r/AWus58HYJB3F9wELPpQemsUrgFcUuPlgw8+wAcffNDl7R9++GEcPnwY//mf/wkA2L9/PyZOnIjZs2cr4+VbgkERHf83tHRH7ul/R++SnT4ZKizOH2VeDcsGfKtB3tEvCi+s34jXnp7HJlBHvygmBsYPePwqWOTViLj7508ifdZjyF/9e0QPisPE29LQ3NAAXddhDwu33O/Gu+5BUHAwbrzrHlNhudqqSoQ5XaitqkRsQkclaErh5a+PVvlW3hEe/vqFNxzoL586LJ7ziaUr0L17dwDA0KRkxCYkWk6+dk1DL7vdh7/Eq/fq6FDWtWsanNExaKyvY9lTfAVqf0qr4v98iEgsjOjPS9WZETj/jVxokU5LgUCr1Gj+OfAnriZei0xin8jeD7+wGI6oKCSlTEZAYCBuvOse7Nzysd+MJt6AlGm8iKFVWdiTT5+3ytKqKNuNcKcLoRERCI2IkD5nvCfUHmYo7B7et9fHwwR0EO6tDHZezZgPTYoh2Ym3peE37wPLznYc55mblfFyrUO/HC9d1/WpU6f63aagoEBfsWKF6bMHHnhAb25uttwnICBAt9vt7BUVFaXruq7b7fbLcl3q1fVX0yLoF5Z2vEYb2mb67JWr9I3lX+qzV64ybR+bkKjPXrlKj01I9Htc2XaxCYn6n/51UM8/XK2v3rZdB6Cnpk/TV2/bri/euEnfWP6lvmZ7sf7OoSp9QW6e6Rh8e+h9avo09p6O+6d/HTR9v3rbdn1jhe910Iu2SU2fZtl+WV/w36/etl3PP1ytr/msWF+Qm6c/4237xfTLxfYpnZN/WV0jfz+pfXz/vnOoSn/nUJX+TG4eO/6C3Dx2H8T++tO/DurvVFTqYz/ZruP8eR0XLphf5eWW/fzyxk36xoov9dXbtnd6vbLrp7Y/k5un5x+q0vMPV+svb9xk2X9Wz7HV9/7uBd9X/PHoGPQcdfbMdfX3RN8tyM2zvIauPjup6dP0NduL9RV/+X8+zyd/DKvjdaUt4r2+++dPdnptsQmJ+p4s8xj06p1XflxUr46X3W6/mPn78jSqK8bLgQMH9Hnz5pk+++EPf6jruq4HBQVJ93n22Wd1GZTxcnW9RseYB43PngCbuBZv3KS/vHETm+SAzgfKrgykorFAAz9N/C9v3MQmUtmx+clh9bbtbN/X//4PfWP5l/rDLy7ucpvE76wG8c4MDt4A4ye3rkwK/l7+9qFzPvziYn3NZ8X64o2b/B6bjvWM1yjh22gyuF591dcYuZSvU6e6fN288UDP6JrPik2Gp2jY8gZuZ/ecfx75Z42O79NXMBtUnT1HXf1NyL5/c+du/c2dZXpq+rSvZQDLjLnODLyL/Y37O57VtZ5dDP3Cfxmv80uv/NioXh2vizFervlso5dffhnLly9n7+12O6qrq69gixRk+LBDYws6gLG/A2av7JA+r6+uwpCRSbDBzF0A5BLgXZFrF0l8fCjqVHMzCjflI9zpQsGmfCmvhCdxUkjFGR0D2GzQdR29eoeY2iQjKIr8BeLhWBFPZend4rUSD4ZPV7Y6Hh/KEDNKePBZImImEd+Pr82fZ0nc5c/BK6rqAF4ZNwZtP38Ui+igMzJ82nDJERwMXLhg+qjiwgX0PVqPNPgKyVHYZN2yJZielY2wSCduvXcGq3EFwJRN5XS7Ee50MT4M3ROqhk6EVb5oJ51L5DmJqcd0H0f6UYe2qjbOQ+Q1ifWRklOnoE+4EdudMW8+9nz6ibQaufh8yZ4rq7R5+k6WXm66N10gzFtldMVyCsNi1tqm0Fm48/RqAIZw74eZwE1yWpLCVYyryng5duwYXC6X6TOXy4UTJ06gtbVVuk97ezva29svR/MUviLevg8I7dXxfuFfjb88UZZq1lBG0MHSEmnlZAKf6UGDoL8JGJDXnCFSI/EBeOEykcRJar4in0OE1aQlFuvjr0P8X5wgZKrAfPqvVdVo3mgSC0TyBlC4l3BMZQQIIrF2elY2YhNGok9YGJPal57jpsmoiI/H89Ie+ho4dAiIiwNefRXIzOwoevVV0a0bfhntAn7+qPHSdSAnBxWPPGLKYiMFY2e0G2FOo1ZUSeFWk+AaqdOSYUD3hC86adc0Q1RR101CfmKpAro3A4fHIzTCwdSUAevMMn+p2XQdPMk3baZv3aTNa3IwalIK+oRHmNKO6W+stygjX3vJypDhScNiPSIrhV9/xowMVun+vMKwWP6htNckTD29Gt0BQDdqq6nCjdcerirj5dNPP8W///u/mz67+eab8emnn16hFil8E0hP6Phf1ztIcqJeCBkHNOAPHpkkzSriV/j84MmnaVpBTNElrQ1aBZPxBMgrFHuamzvNSOINCpMmjXeCopU5P9jLih4CHRoXvex2DE0aDQA+mT6A4QGgWjHiNdJxxDpQ/OTVVFcHXdfR4vGY2iGeZ1TKZNi8BoMO82S26O47gaAgwwi4GJBBcjF45BHj1RXcey/w3/8NeAnHfmGzAbNmGS8Ah9vbMbGqzlRXijwvfApxQf5GpvLaVFdrqs+zeU0OM4gzsrKh6zqa6utM90L0EtK9sQFoPXNGWv9JRFdE8RqO1iDAm9nDLx6oLRVlu/GHF55Dxpy5+EveWh/vh+j9kRFj6b2VQc3/pcWHzNPqz6DhxwKZYS8uPPhjrVu2BJOfewZTDj3PRDI/zATCnrHsWoWrEJc8VTqOG5QGDhyIkSNHorGxEZWVlXjppZcQHR2N+++/HwDw2muv4fHHH8evf/1r/OEPf8ANN9yAu+++Gz/60Y8uZTMVLiE+e8K8OM7Zbr1thlckrsRbPTg6No65nDOyshHnXfGPGD+BKaPygyBVHn5L0J6wgqi1wWerFBdswYLcPJNLXeZRscomoWNR5gYN0PzkIgt90XEeWrgIkTFuuNxuDB6ZhBPHGwB0GGai4TVs9BhERsfghfUbWRYUHwKLTUj0KeYo83yJq3qxfw+U7KTxHjYAzQUFeF4L7VJ/AwD++U9g7Niub/9NIC/PeIloawN69vS764WAAEwbFAMMyjCMLG/YjiAa1Ef27cWoiSkmzyFvAJBoYGcehs7ujWxi95dNxWeBVXOVwsW6SQB8svL8nUNmKFuFuPwJUJZ2Um5Dmh3llUOorij3qXkmM7r4Y2U+C3x0fwyu0w13S2gv5X251nBJjZcxY8Zg69at7P2KFSsAALm5ufjpT3+Kfv364brrOgrcHDlyBD/60Y+wYsUK/OIXv0BVVRUefPBBlSZ9DYOvX3TuAvB5yirEHpQP3MF2Owtd0ACaMjUdMbFxiIyOYSv+k43HmYouPwiK1aK7An7AFfkFMpc6IC/4CPimm5IRIxugRfc7j7SZmRiaPIa950MTfJiMN7yWPJqJF9ZvRFBwsCH0FRjIrittZiZcbjeGJI02VVUWPV+y1GtZ/6649x4gPLxrHfxVvCqXC94+YtizB4iPt95+0CCDM9PaCkyaBOzc6RMyo3vsdLsRGR2DlKnpPinV/hSa+TRqf/fGKlxj5ZkhMUbRoBCrkdPxAF8vpAwyrhjgn4vChyFJgJIWCKnp05hAI2+g814V+u001zfA0S8KjXW1fkPM/DXxRteA/wQuLAXzvmx7DAj+leUhFK4yXFLjpaCggE04Mvz0pz+V7pOcnHwpm6VwmfDqneb3f+55n9/BsMXjYaELGmScbjec0TE42dSE7t27o3doKFpbWizDLKIhQKqxH21Yj6HJY3yE6Mj1nZGVbap1w698eYOB96jwPACejMgP+nQcfoAGDHKlOJHw1+JyuxEZ48Zab0Vg8pykTE3HqIkpsGsaADDjp6JsN157eh671uhBcag+VI4X1m9EYFAQmhrqARgeEx5d5RpsXpODFY9mdi0ktHp110M6VxNGjOj4v7ISiI6WbxcUZHiQALz24UeApNDgyxs3+eUQUfVoIgT78zLwoNDUycbjPrwTf4UmAflvRBf+isaFWAhVNJqopAER0QHDgxrrLdeRMjWdCfjx2j0EccHBE4pnTRonLfRJXp3SokLsaW42fisSQ1G8HhmZ/kRAX4SePQbYgKAexpj1yDuW3a9wFeGq4rwofLuQyelYNdoiUNn/xzgj8TQQyKXO1zLJ8EqUl5eVwgaDc3HG47EU/IpNSGQiaOuWLWGD4bRHnjCq1gpEScDML6irrmLHXGQhTy6unkWiMIWeaGKReYT8ufgrynbj9YULkJGVjVRu8CfvTWlRIfoPi0dYZCSa6usxPSsbby1b4sObWL1tO4KCg9Ha0mKqts2XARBr4kjR0IAKf56W8+eBn/5UHpq5VuF2d/zvx5A5c9ONWHHTjcC0qSjg9iFD3BntNnkNeE9NZHQMALBnfsS4CXC63UhNn2aZGcYXICRV6q54SnhyNg+S9+9tt/uIB4reE1GAL21mJhxRUWhvNSpuE8E4zNmRdGETrlkUnaRjsfYsX4r7581HyykPnsnNM2VkidtT/zyTm8fOJV4ziRRSJlhTXS0Gj0xifTSr4h6sv24li4POGq+Ml2sFynhRuCR4+z6O62IDtt1ZiqEAdhVutVzp00BMA04IeRcKt7JBllfAtQrd8LLmH21Yj/RZj+FUcxMcUdE42dTkwx0I0TR8XrITLR6PlIwrkgvFtotEYSsegHit/lbYYthK5BNoDgfaWlsR5nQi3OVCbEIiao4cxhnuGmgiqK2qNFXx5dNyxZo4JjQ0+A8PXc0hoW8SvCFj1SfR0UZI6cQJICwM65YtYaUOMrKyTVlnAEyeF7qnYU6ncT+dLp/MMAJfgLC4YIvJUAdgUhcW5f7FMCjQwfui73Z5M6jod0jPyf6dO3wqOYtkcLumISzSyMQ60XgcR48cRi+7Hb3sdlbjSjTKxOsjJeBJaXeg/9DhPgsNmafwrWVL2G9EDP06oqLQXN+AQfEjEBAUhB4BASZuTfSgOJzqEY2QVq+8hk15X64VKONF4ZKAFV+0AS0BkWisr8ORfXsB+GqQ8AMRDTgNNTWGp2VSKtpaWxETNxhV5QdZ9ef9xTvQcLQGZ06fwupt25lcPhET+UkhIDAQ3br3gK7rKC8rNcXOn3o9F2FOJ0oKt1p6HkRyIWAedPnB02rfi4WYCWKVQTRj3nyEO13oEx6BPuER0L2eJXLR11ZVYsjIJJMnq+WUB411tVi3fCmqyg/6DPwrfjYDCAmxbtw77wA//vFFX9O3At6im5ZGTGgocOECKurqsOTmHyBtZibjSIlhHfKoAV0jTwPmrKTZK1cZRod3lbDogXuZ91Esi2DXNBwo2SlN8efPTWT3jKxsnDl1Co21tSzVmCo58+Ua+MKOsQmJCHO5EDVgELZ/8H9ISpnMuGolBVs6LZ7Ih9TELCgCXwD1pYce8CmzwP9GyXPUWFcLLdLBPJCi9pM969dY2Or97euGx1gZL1c/lPGi8I2j4TnujQ4suPAsir1hi+KCLaaQimgM8B6GmLjBSJyYwoSm9u/cwVaIVMgv9Y5pCAoOxsMvLEZV+UFUlO02VXMWV4f8JB2iaQhzOmGz2TBgeLy0YB3gy2kRuS7+iJJWkA3cGVnZCLbb0eLxoHBTvmVWilgQ8MGFixA1YCAa62rRVFuL4oItePSlJQgICkJQcG+20hTDY9RfrO3t7VjRw2JIOHcOCAi4qGv8VoOMmAULgGefBbp1M3/vdKKidBdW1NUh9uYfMOFFGV8EMJ5/kaArEzvksXlNDkZNSkWf8HD05rg1/AJAzPyRPaciOZgZRTA8McmpU+B0uxHUqxfa29p8qmfz7et73QAEBAUh9Y5pCAwKgn7hAmzdukGHb9Vy8fdvlc0Xy1W4JlJ/uMuF6VnZPpXbRQ+lo18UqirKUfTeZsvfkqe5GY2hIxB+Yg9gM7yQb98H/PiPkvuucNWgW+ebKChcHMKDO/4/dLyjjP3E29KYVgW5nfnsm9krVwHo4JUkp05BN5uNCU1tXmNU0t1VuBXrli9F0fubkb/692hvbUVQcDCmZ2X7tIUm54L8jWzgzn4lh1W1LSnciqa6WmgRDrYKpraQQiwNqnzb+c/FVFNx/9iERCzIzcMzuXk+x6R9ybAYljwGSSmT8fALizHx9jRkv5LD9pEdv6JsN854PLCHhaOpthbPP3CvUdnZm0lz9MghU3+XFG5FU30dK84Ym5CIwFOnjJCHzHBpazMmZmW4yLFokdFvXgKvD5xOVOwqwYrjtVi3bAnLHCPDReR/8BCfEfG5MrwOpSZ9nlgq6NjQwKo+87+zroD/nekwPKX9+g8EbDYcP1aDhpoaVB8qZ78j/hk2Koq34tiXR9DsJYnXVVeZ+GcZc+ay8/Dtovck6scfd1TKZCSlTEaLx4MTjccBgAkH+rsO0oySGS7Un8UFW/CbnvMM0rKXuXxnong0hasNyvOi8I3isyfM7+MWG3/Jde10u7F44ybGL/EnbU5u3482rEdy6hRUlR+Upo8O9U74YiaNDPyqlLghvGdl9spVsGsaRnKCWVYpm1Z8FnEFGaJplmnX/N8QTUOw3Q5njBthjki0tbYyI8OfZoWYMcKHzgo35ZtW77zk/H/3sKF5V4lcpVbXge9/H9i5swu9qsC0a2SeGJsNeO45VPziF1jhcGBBbh4io2PQVF/Hwi8y0H0kATaZp1IMWabNzGRihqQnJIYuO8sw4/VfyFPaf+gwhEZEoHcfDSGhobjxrnsQGRWNttZWk+T/sNFjGCl2V1GhyYMiqghTu/j0aFk2n/g8DxweD+g6zgiCinT9fB/ZACSlTEb/4fF46cEHTNcr/k7X7w/DT4Y2se+V7svVDWW8KHyj4HVd2s6DrWySU6dABzA0aTRs3tpAPBlPZggkp06BIyrK0C0JMjwJMrc3L/wFWMfSecIkP3CLnJZdgmCWTMyO30/E5jU5cLrdGDF2PHr07InSokKUFG71UTIVUzfJMOONqZSp6bBrmikDRewr8fr50Bldk8i3WDFtqnUq8McfAzfdJP9OwT8WLTJeJ0/68obCw4ELF/B6w3G8cqoNX+zbaylySFwOj/e5o/srGtHiM7h5TQ76Dx2KfgMG4czpU9KwkxXZnQ/riHL+C3LzcN3QYTh65BBqKytNac9kJNHvtUfPAJQWFZoUiAFfFWGCWG+JT9Wm9vPPc2iEA031dUzhWczic7rdGDFuAlLTpxkGvc2GcKdvmInOYdc0TLw9DQ2jxwDbO1IktzwM9Hna791WuIJQxovCN4byeeb3rwTONQl4fV5agqa6OrSdaUFgcLBJVEpmCGxek8P2lbnXrTgnVloUgFzynz8f/bXKOLJqQ/Whctx41z0snh/udCEgKAjnz59HwaZ8S6KibCLhSwcAhvQ/cXxk1+CPFEztdbndGJUyGZ7YQai49Wa5VH51tTmzRuGro08f4+9nnwHXX2/66pgjAneGnUfssiWdPgviMykzonlUlO1GrxA7AoKCcNN/ZDCnmmjg0F/+/LwRseTRTJOhJFMGpsKOTrcbC3LzUOhVenb0i8Ke5uYui0XyHhmxHIVVf1A5gVNCP1SU7TZSyV0uPPzCYrz29DwMGB6PsEinj2eWX0DQ7+ugLQ6DdSMDLyRQeV+uZijjReEbw6CIjv9bbMFov24iSosKTUUXNYcDDUdroEU4fGqwiAJZaTMz2SpQZlBYrSBlpD1/3AKCPyPA6jtSKD3b1oaAoCBWwXbd8qV4+IXFCAgMlBaNI28IAEuVXV6VV1abiAev4yGK8NHK+WlXBPYOuNV3Z296r8IlwNixwLFjgLeYI0P37qjYVYKU0jJM7GNkLfHkct5jYvVsW4F/9jp77nmNIt6IkBlK4vMvplnbAFRVlLPyAwSrMBVPxKXwWVX5QZ/r4//nDQ6rtHC6/sCgICSnTsFLDz5gGldEVJR1FGzNfjMYB2d0fPfJ40DgPOluClcYynhR+EbAuC7eld4/pmzA0LCR2Lb5XeYuJqJhsN2OxrpaHw5JZ6sucZCyWkHKZMppQP46kA3CtJqrOXIIwSF2Fs8vyN/o4wIniKtHK5Vd+kvfyVzu1Ca7pjF9G9mqfFHG3b6EXF0HFi40whwKlw59+xp/xfIDNhsKRyXis5YWxHDPrZjuy2fodCX1Xnz2/Bn9vEaRGNbpiqHE81EAX8VcQF6CgNrBazLRdTndbjz1ei7W/vpFv9dKsgkATBw12fV31mf8NoeOdyzEenahlqfClYP+bXrZ7XZd13Xdbrdf8bZ8l14XlkK/8F/G6+x/2fRncvP0xRs36Qty8/TYhEQdgB6bkKiv3rZd31j+Jfu7IDdPn71ylR6bkKjHJiTqC3Lz9JV/+VDP3blbX7xxk8++71RU6qu3bWef02v2ylX6xoovfb6bvXKVvrH8S332ylU+bU5Nn6av3rZdT02fZnldsQmJrH2yY/Hf8/93dizZe7FNnR2Pv74FuXn6gtw8/RmuvwHoePVVHRcu+L7Onr3iz8x38tXQIL8fZ85In5PV27br+Yer9dXbtnf6PHX2eVefWfosNX2a/oz3ufL3DHZ23AW5efo7h6r0Bbl5pu1T06fpf/isWH+Z+53PXrlKzz9c7XPN4rFXb9uuv3Ooih2X2tvZ76WrrwtLO157sq6C5+Y78rqY+Vt5XhS+NkbHeP/RAdiAX26LxvJN95pWkKeam2HXNJbp89GG9bjxrnsQbLczOfIVTz6OU83NrPpxW0uLWbyuX5RlBg7Pj+G/8+ed4WP8tOKUKZOKdYrIYxQTN9jkZqfr5VVIeXSlfg3fJlI1pe15Iq+MvOvjWZLwLQB8d5Rxr0aQPoxY0TowEDh7Fhg3DhU7d7L7feaUB+2trSgp3OpDvhWfJ94Lx3sirEppEHi9FgBM4G7E+AkId7qg6zoGDI/HkX17fUi4BLH8AN82kVAOGOTgh19YjICgQBzet9ckTnf0yGGE9+2HjzasZ9vz3htPczMcUVFoqq/DF1ybrEQkrdra226HDqOUg3hdHx8Ebhhs/B/f1/JQClcQynhR+Np472fcGx1YvslguIls/ub6BpRuMzgwlKlQVVHOOB2UmWQUY3Rj3fKlPgJxVtwPPm5txRlYkJtnquMjpm4CvmRfKqooEiajY+PQo2cAwl0uZvxsXpODEeMmIDI6BhlZ2dLsITouU0HliMSxCYlorKtFz4AApn5L17wgN48Vjvz+zbcwHRdLArKMZ6HrwP33f7tqEF2rCAwE7r0XWLu247Pu3YF//hPdKyrw8//9K+yahuuGDgdgpPs6+kVJ5fl5jpgsW84qbEMQw7W0wGiqq0VYpBPtbW0Ij3QiLNLJ6jDxoSw6hlU5C1mqNokotre2Ml0XWnycP3cOPQMDkZQ6BX/+7UoAxrqI/vKhYD7MZVXJWna9SSmTWdFgXdd9iL835XgrTnuhSgZcfVDGi8LXRl9OK6qxpeN/nlzXf1g8wpxOHNm/l6VUNtc3sPRhfgD91bSp7Biy1RRfpZavzEu49d4ZGJo02qfYIKlzUqVfWeqmSPblY/hUB+nE8QY4oqLw+a4SnDvbjo82rGfG0hf79xqqvdwxZStkWt22t7aZCt8NGZmEovc3s3YRB4KUcVvPnEFgUBDqa6otU8PR0mJUPubR0uJf8l/h8iMvDxg40OAdcVo752Nj8bvHZuFXb/0PS7FnBn9Uh2fRH5mX9yLQxN9/WLy06COvJ1PoLYRIafoHSjp0fqjaO1/ziJ5TMvRl5SxEUMkOADi8dw/TqiGjyBljZLy5Yjoy33jvDXlhSfuJnn0aVyjlGgAjBPOeFWoreV5sANPS4fvtRCsQ6v0ZqYKNVx+U8aLwtfD2feb3jmd9t6ko222a1GmQdXrTdwEjQ4YX5PInBEcryYTxE6B5C8HZ4C0ZEBWFc+1nAQADBcl/qvTbIhG34ttKA6RYDoCMmeYGw4P0luCuJuLirsKtrEZMbEIi7Jpmyigiobz21jYEBHVkIxUXbMGI8RPgcrt9+sDldiMyxo2/vrUWw5LHsAnJx4Nz7z3KcLmWQLowZ84Y3hgvznXvjkU/ucv0mUgA7ywcRGBFIr0Tu1j0UaYnQ8Z1w9EaOPpFmUoLyDyWfIkBsayBjGhPvxUyKviFyUMLF7FnnT+O7HcpekrJe5P9Sg6qK8qlhGCxHAJ5ZAGYPr9pNfDPX1zMzVS4nFDGi8LXAi+jzXtdxAGLVwLl03cBY2UoDqD+hOBo4q6tqsSRfXvZe/JkBAYFoe3MGYQ6HJielc3OW7gpHzGxcWx1KYKvecTzcAjFBVvw/ZtvQVhkJPZsbwbQIcIHyLOHqK7MwdISZL+Sg482rDeMmW3mKruxCYnImDOXued53k5F2W7UVlZi8MgkRA+KM/UTXXdDTQ1W3ZjqWyxQabdcG+jVy1fYrmdPowCklycj/ha6wqGi/fiQqizsyvO5pmdls9ARL1UAGL+R5NQpfpWBO9NZ4kUUZW2d5/W8il5X8biiIUPX+dQbuXBGx6CprpaVOBArcPNtD3e5YLPZMHRUsmnRIOq7NDwnX5wpXBko40XhK+PeZPN7x7OCATApxRSjFwdYkcjXldRMfj9xIOY/o5BS/2Hx0BwOhGgaW32KuisE0rz48sA+NBw16rfwKz+qGUQhG1k4SCQm0v9EfiS14IaaGtO5yRVO1bfF1bXoBRL/X/HIQ2YCKACsXg088ojfvlS4itCnj48HBuHhhucs2CgYZqUWTd9N9xb3PC2QUEXvhphGzfO5Pi8tYaR6McRkZTBZtYuea10Hqg+VSy+b13sRwztWf63aUVG2G0f27UVYpBOnPR7mSZm9cpXJC8OTlPuEG3nRvUNDfRIB9h7rIOzyNdsUrjyU8aLwlfGHuzv+120dAxiJqzXU1Jiyf6zc3LEJiZarIlmGjUwPgyeukmGSnDoFYZFGjaBgu91HrE5sD7EO+oRHIDTCgRvvusfkYhc5BbRaJYXRdVytJgJfvyVjzlyWZeXo1+HC58XqxOvnBe10GCGzAq/nKCZusFEsMvNnvobLO+8ow+VaRK9eQGWluXRDUBDQ2orZr73hk0kkemJGeYmoMhIqv52VuGOIpmHg8HiEOhy4dfoMaJFORvSVhUAJvNjdogfuNWX7jbvlhwgICsKt02fgz79dKc3oswrvWClJ+1voiB5WavfnJTtx2svb4fvgr2+txbRHnsDRI4d8xoapa3JwcEbH71Ep7l49UMaLwldGDxJwsgGH+/8E07N+BB1AaVEh43yIirmi25cRY72DV0xsnMklTftQLBuQS6cDchE72q/F42EZGFaryIJN+eg/LB6epkYc3rcXhZvykTo13cTDEQdTXmFUnCz49vBCfcOSx6CqopyFjboiVscXdyQp82Gjx2DWuNFmqX+VUXTtw+0G3n4buPPOjs8CAvDqY7Mw963/MWXnyYi35Hmx8mAWF2zBiHET4BS4VfQ8h0Y40FBTg8a6WmicrD5xzXYVbgUA0/l5sTseFWW7cWjvHgxLHoPaqkoAvr87ngPTmdfVqiQIj+TUKaaaSxS6LfUWiqQ+oMyk5NQp6NGzJ744cEAumHn2cePCdODDTCDsGb9NVLhMUMaLwlfCMYr9ekerj/r+AnpFuc8k7E+2n4ySdcuXwq5p6D8s3pRNwe/Dh4OspNNlxpG/8gKi8ZOcOgVhTifCnU70CglBVflBxi8B4FPbhfal1G4xRVO2wqWBlDKKyKARw00E3jtF3U3lFlY8/KDZcDl3DggIkN4vhWsMP/4x8OGHwA03sI9au3fHorvvBIKDpZpCxC2zUtYlfZOBw+OhRToR5nQi+5Uc02JBln5N//PpyuKzzWcMiUbVGwsXmI4j/iUODF9dWhbWZdowgdZFWmXHJ0Oll93OvFYAmIEj8+Lwnz18fxwGoxywAaG9pKdUuEK44qp63+RLKexenhevQHnuv9AlhVn+xdR2KzoUay9GUVammNuZGm5X2rQgN0//w2fFbF9SvH154ybL41mdqzMF3a91D9rbfRVzR4++4s+Fen3Dr88+81Xj3bNHj01I1Nd8VqznH6rSn+GUa3ml6dT0aSbV5dkrVzFV2v/ZV6G/ubPM9Pvr7CVT0BVVbcXfgriPlWJvbEKi/j/7KvT8w9X6Hz4rlp6f1Ib/Z1+Fj3K3ldLwM7l5+prPivWNFV/qz3BKvFI1aj/XfeG/Osa7DzOvgufiW/pSCrsKlxXru90HoLRLNUQIlBkwPSvbMiwDXFw9I6uYOL8dAMuVKaVQ8tsT4ZDE9GTeka4QjemYLrcbkdExSJma7rO69NFrsfgMZ86Y6xTpOvDTnwI7d0LhW4axY4FXXwVmzer4LD4eh3b8Eyc/LkS404Ve9g6hJT5USkJw0HX2/YGSnXDFuBHqcKD8s1Kpx0/6zMGXgyLjnck8N7wAnozbQtv19HpUKLwkglK0G+tqTRWluxKObqqrQy+vrkvK1HTLMC2vFExSCBVlu43p0gtS3lW4slDGi8JFY08W98YG9Ep7EWndNvvwPWQZBDwqynZDB6QaCwTRPc0PoFZy4OLgO3vlKky83XCxV1eUmwiPPMTMHsBa+Eu8DvH85J4PjXCwzyfelgbPiWbYbDZEuFw+WhhiSqns+tHQYM5GARTH5duORx4BJk0yFXXUe/ZEftJIPNnSjjOcbhEtCrJfyYEz2qjb0VhfBx0dRRM/yFuLjDlzUbgpX5p5ZFV2wCr0SvpMVgJ4/F+e20Ik9nXLl5p4L4Wb8rEgN8+nQjofZrUKQ/FtLy7YgoTxE1BbVYkzHg8zZJrr66SkY7p2USk4bWYm/lr+K9zS3+hnzhZUuIJQxovCRYPV+rABF9ANDUdrfAYCqwwCEVZEP4I/r4bVd+JKzK5paK5v8OtB4Y0HkRxMA7gYyxfBp4mTIm5ddZXpXE63G33CwtEnPMJHw0KWDcWXJ0BDg6+OyzvvKMPlu4ARI3zSqAudDkxeu46R44GOZ3Dd8qU+6tOiQm3q1HQm7BiiaUxqwPTMwUya5/kxZKQP8GYnxXiJ5IBZAI83gHhjhFfqnTVpHOPrpE5N91shHeioKP2WIHInGjKhEQ6UffoJ+5ykE041N0t/x5RxJSp/f3l0DPRPx8FmM8SQP8w0SggoXDko40Xhq8Frbfwr7jGpYFVXMwh48TpA7rGhgamriqL8AEYZEgdKdmLP9k8sjQ/eeBCFuej7ibelsUFedhw+TZxk3d8SdCsysrKxq3ArI93KVo/8Cpi5t++Y6mu4NDYaxE6F7wZ69fIxYJ6f/h+IXbaEGda84U7pzeLvic9go2ee6guRavWBkp0m3ZZho8f4kOl5L0V9dZVU0I7ODcAk/Eh1vHr0DGBKvZRufaBkJ/v9XIx3RCTxi7+p54WQsAyi+i5/7bZuYOEjFTq68lDGi8JFgRem0wH8/M0ay5CQlYqmuJ1MqwLw9diIWhJdOSYNnC0ej18+jjjQiXwUku6PS0hESKgGwDfs1FmIidpfUrjVp66SzBAzGTalu8wHa2npqFCs8N2BaMB0744vP9uOiTUNAHy5Xtmv5CDSGz7iZQhkooqe5mY43W5okU4MjB+BISOTABjPOSn0ihk5Trcbrhg3yxDin2kT56S+DqVFhSwbya5pGDIyCbuKClmNIqo9BshDyPx5Zd4RaivhYjh4VuDVifvWBGN8v5bOd1K4LFDGi8JFYfWPwbwubejFYtAiye2rwp/HprMQk0zQjlzmNu/3osubb7NsoBPTSwFjlekvrdkKVu0XPUqiWmnFzh3mHVpbVa2i7zJ69QIuXGBvzwYFITPcjke9Yoc8J8wogFoHXYeP50RGwOXDp3wIkxATN9j0rNZVVmLIyCSpajX9lilUs6e5GcmpUzDxtjQ0H2/ArqJC5vEBulZ7jNotekf4vzxETwvVbOL7qTNQP62AqjR9NUEZLwoXhV6ckOuHjkwAv+nUjeuP8Af4Tt4yj01sQiJ0mIse8t9Nz8pmg6TIWSHZcz5+LrbZXwVc4q801dfh8L69lgTkziCGyPhzWNVw+VPGf/hmFgUrnfLvPN55xyRid1wLxZr/+DH+g3uW/Wm2EGSh2CWPZjLDniCKRdICI9hutyS/ykI1MXGDMe6WHyI80om6qkq0eDz4vLSEVbS20jvyB96TJJLgn3o9F2FOJwv3WhllXYWudxQAzxynjJcrCWW8KHQZnz1hfv9PXA/AXGLe6XbjqddzoUU62HeyLBoCc21HRZsIgbwXhdUoinSg6D3f9EbePS2LvfNqmgRqc7Dd7lPJmgfvoha9SlZGWVdTTflzyP4WVH2JY44I88b33++zv8J3ED/+MXDsGOB0so9qAgNMfBP6/fhTpJWFXPg6R2TY878lUoVOSjFKERQXbPFrzPPP/fSsbCN9G4Az2g3N4UDD0RoMGZnkl9hvBVHWQFwEhDmdsNlsJv6MP9I9gc+E4jOc1u9ZgJ8MbQLQYcQoXBko40Why7j+Ou8/OgBbRxYDuXFnr1yFlLQ7AJsNTXW1pswGq5pC9H1ba6upDhK/0qOYfX11lV9ZdD6zIjl1Cmu3KBcutnnibWkYODxe6lURXdQ8rGLtVp9bwbKGy9mz5g1PnFCZRQod6NvXpw5SxT8+A3r18ltWA/Bf8JN/T3/5Z5Qm8wHD4xEe6bQM48qgA9B1HU11dVj76xeZarZI9PXnqeUhZhaKf8WFR1eNo4w5c1kmFJUYmHhbGko1DWi4l5HpFtwELPrwIjpA4RuDMl4UvhK+PNndJzQUomk42dQEe1gYI8paZdGIAw4NYNWHyrF623Z8tGE9q+HCp3xarRT5UFNnwlk8KJvAGR0DLdJpWv2lpk/D/fPmo7aqEm8sXNBpeYHOPu+yCB0A7Nljlv5vbQXCwnzar/Adh9tt4r8gMBDfKy1D8cKFAMB+P6J3UfwNPrhwEZzRbqz99YsoyN/Y6URfUbYbLz34gN/MHRnESvJE1hWJvl3RmfGREpC00R/xVwY6FxVQpUwo8t4G2+1AA1itowU3K+PlSuKSS/4++uij+uHDh/UzZ87o27dv16+//nrLbe+//35dxJkzZy6JvLB6df01OsZcEuBP/zqov1NRyWTASRZ8QW6ej+y/+IpNSNQXb9yk5+7crb+8cZNJopskwFdv2y6V/e5KCYGL3cZKtpzaQpLlC3LzmBy6KIvelZesjIBlGQNREv4qeAbU6yp9lZf7PC/0+wHkz5hYSoOe8z/96+DX/m119dXVshoLcvP0dw5V6Wu2F5t+r2u2F7MxZ/bKVfqC3LyLLgkivp7xnosvucC3dUFunr7rvxJNY+EVv//fotdVVR7g7rvvxvLly/Hwww/js88+w5NPPom//vWvGDp0KOrr66X7nDhxAkOHDmXvdSVpeMXx/s+4NzYgKDgY7W1tsHtXdKK3QZT951FRththThdCwyMQGh5hyoAgCfB1y5f6rA6/SkaTVWkAcXUnIwmvW74U98+bj27deyAs0gkt0smEuDpzx8tc3hTe4vtF7LfYhERUFAsy/x9/3KVrVfiOIi4OOHnSlIE2a9xozPb+rmRVpEU9FKfbjdjvJSCwVy9kZGVbeiy6KlfQFYhp3WKmIP1+bABsNhvCnE6kzcwEABPHjTKWSosKLStudxW68FfW1kVlu01ZRwpXBpfceJkzZw5ef/115ObmAgAefvhh/OhHP8LPfvYz/PrXv5buo+s6amtrL3XTFC4CLiqfYgNOBxkx9tbTpzFyYgoj9fGTOFVjtiLhrVu+FDN+OR911ZWmiTs5dQoj7VaVHwTQoeVAWT/RnF6FzFgQ66mQuNzA4fEIi3TCrml4a9kShGgaM75kgxwvR05iW4VecTkxTk/ozOUt9ovonq8tLDCHi9rbgZtu6tpNUvjuok8fc/ioe3f8+dMiAB3V0mnyF3+PFWW78atpU/FMbh5GeX9jViBDYsTY8bj7508ielCc1ODoCmSlPsRFAWUZ7i/egRaPh/3e7JqGXnY7zng8KNiUz7L4WDmQi+Cc8bCSVhDFMmHrqHf02RPA2N9d1GkUvgFcUuOlZ8+eGD16NF5++WX2ma7r+PDDDzF+/HjL/UJCQnDkyBF069YNxcXFeOqpp7B3717ptgEBAQjkFCftXJEyhUsAHdhw/XuoLd6BD/LWSidwoPNihQX5G1ldFQJN/HZNY4MRvzoM0TQMGB4vJfbyyreyc9s1zdBpsdnQf5hRI0aWQi2DmPIpDtaiLgt/btGY4Tk+Ylpn2sxMrAgNNZ/cm5mhoNApVq82FXCszvsjYvfu9RF1s0LBpnxEx8ahYFM++0w0vgs25SNxYgoCgoKQPusxBAQGSg0OPjValrUjGjoi9624YAsW5OYhLmEk7GFh2FW4lXHeqg+VIzo2Dk11taxOmae5mbXX6XbjxPEGU3ahP4htOtXcjIm3p5kWSfx2IZqGOv13cB7/DNC5RAaFy4pLarw4HA706NHDx4tSW1uLYcOGSfc5cOAAfvazn2H37t0IDQ3Ff/7nf+KTTz7B9773PVRXV/ts/6tf/QoLveQ0hUuD0THm9z8b2yGzKyvsBvhqL8gmfFEwigawEE2TknLFis+AueAbv49MfAvoqG2SkZUNACbVT9nKkU+ZJIEtkgsfMW4Cvti/FzrM0uf8ua2yNsTVYdrMTKx8bJbp3CpcpHBReOQRYObMDl2g7t2hb9yIQ0MGs4lfFJnjIcvKExcHdk1DN5sNrS0tyF/9ex/PC78PAIMML8naAazVcAvyN2L2ylUsFRswnBxU6uNsWxsCgoLQIyAARe9vZuNFiKYhJjaOZSfKhPN48MYI//slEj+/SOKvq7SoEL8+cR+W6Z995Vul8PVx1WUbbd++Hdu3b2fvP/nkE+zbtw+zZs3CM88847P9yy+/jOXLl7P3drtdauQofHXMTjG/fyY3Dzrk2T/iKkaWsknaL5FR0WhrbWUrJN7gAcBCOgBMx5QNerSdP57N8w/cy4yRYLsdQ0Ymoej9zcwooWvgwadMrlu+FMNGj2GZCJHRMQhzOrGrcCsr9vhV6i8BwJ8/LYL+80c7Njh/XoWLFC4eAQGm8NGhuFjTQoFqF9k1zYfnJfMKEk9rwHDD6N+1rRDb3tvkV4CS11UqLtjCfjvidwSZwrRd03CgxOB+tXg8KNyUjxm/nA8AqDlyCMEhdpM3hwyryKhotLe14fDePVIvE6+sDQAjJ6Ywrgz/+6XFSnHBFjbeFXo9UtTOZUvBso4ULj8uqfHS0NCAc+fOweVymT53uVw4duxYl45x7tw5lJSUIC4uTvp9e3s72tvbv3ZbFaxxoA64oAPdbEBd8FAkpUyGDphSHq3CJDRYfbRhPaIHxXVou0QZ2i4BgYE+KyTRdVtVUY5RE1NM4SSZccLzSeg8IlGXVpdUXZpSST8vLZHyX3gCMe0bPchwJz+4cBFcMW7sL96B6EFx0uvnIQ7S/PfV694yd/q4cV/1dil813HqlIm8e+yTT+Csb0ZbWxtONB6HFumUzrdWXkFPczO0CAfaW9tQuCnfb1ozYPbgrHjycdP2/rw7gHlhU19TzcI2s1eughbpQFN9HZpqa/FuzmsszZpfwER7CfV1lZWmcM/0rGz0stvhinGzMh8l3KKDthWvffbKVYwcHBMbxxIJNq/JMXgvAGAD9mQBI5Z95Tum8BVwSY2Xs2fPYufOnbjxxhuxadMmAAbZ68Ybb8SqVau6dIxu3bohISEB//d//3cpm6rgB7+cYhgu7bae+Pst/w/N9XU4sm+vdBIWM2r4CV+ULbcivdJ337/5FjiiolDtNTTIPSxT4hX5JjLSLgnZlRYVMjG62StXYeTEFDQcrYFjZJKJ/0Lt51dhdHy+rsuNd91j8irxfwmyMBnDggVmuc7qamCnkHGkoNBVCOTd072DgYYTCAgMRFNtLb48cMDSKyETrmNhlCiz0WGlsyLz4ADwObbsfKJoZUZWNk41N7NtKMRDRgpgDi1TEUXeY2KDkZ1EIaj21lYcPXKYnVPGveH/2jUN/YfHwxEVhYdfWIwAL8dy99EsJEa2AToQ3/ci75HC18YlDxstX74cb775Jnbs2IF//OMfePLJJ9G7d2/893//NwDgzTffRHV1NZ566ikAwIIFC7B9+3aUl5dD0zTMnTsX/fv3xxtvvHGpm6pggeAAADYgQD/rs1IRQZ9TZhDVIeIzikwEQIuYdHLqFAQEBqKhpoalRqemT0P89WMRGRXtQ9gFzHwTmRFBA1/R+x0lBuj76kPluPGue0zu7M5USmVGmFWYiLxN0roqImfL7Zb2iYJCl9HSYqqBdW8/B+as38CMdrEOEGDtNRSNggW5eSxaMspr+POQeXAAmLwrPNeE+DT0HWD8plKmpmPg8HiEOjpKjUzPysauokKW9Sf+lvhzJ6VMBmw2HCjegV2FW9HLbkfUgEGwh4XBHhYO9+AhPtmQsrA0kfXJI9TUUA+7pmHamqH4/L7dsNkApeZx+XHJjZc///nPiIyMxPPPP4++ffti165duPXWW1FXVwcAuO6663CBWyWEhYXh9ddfR9++fdHU1ISdO3diwoQJ2Ldv36VuqoIEnz0BUznkrqQe8loJ4mDQFel8fkXHa7qQQVNfU81WbnZNkxaGs6qYKxZ+4wc70Z3dmYdIJBnKroMMNQqfrVu+1GT4Rf7qKdTzXpdTp6R9oqBwUQgJMXtfAnqadFlkv0OZ5hCvrcQbBTabDfuLd6D5eAOcMW7MfyMXLz74ADv25jU5qD5Ujva2Npw5fQqR0THMQ8MTX4ve32wi229ekwOX240Z8+ajvqoSoREOnGo+gRHjJsDldmPwyCTsKio0/R5lWXvFBVswYvwEhEU6cdrjYfy8oOBg9AkPx8mm49iz/ZMua8LwBhwZXZ7mZtjOGn2n6hxdfnzr6EZ2ux0nT55Enz594OmktLpC52BiTDbg+NleeKnX0k5/6P6E2rpSt4RWbEXvb8aKJx9nJFueN8P0HG5PQ0NNjakYoxWRUNbOjKxshLtciOwXjfPnz+MPLyy0zKDi298VbQv+OqiQXQmX8rl5TQ4qdpWYR75u3Sz7VUHhovDZZ8D117O3kfOfRv3LLwHo+u9wUtodsNlsOH/+PP782+XYueVjPPV6LsKdTpxsaoQOIDTcKB66v3gH6ior2TNPmUbEbSsp3OqTMSj+VtNmZiJlajoA4GRjI3ZtK8CI8RMQ7nShvbUV//rsUwAG0bbo/c0ADI9Oe1sbXnt6HiPfk4dYFKX0nGhGaHgEDhTvwLxpU33Gms4Qm5BoKqWwZULHouXQcSBu8cXfJoUOXMz8fdVlGylcpdCBl3ot9dFUsRr4QjQN07OyfdRwxRRq2THEmPmIcRMQ7nLhlukzsOfTT0zbUUgnY85cU2gnbWYmJt6ehhHjJ+CLfXtRwLmZeTc5n44JGN6d1KnpGJUy2ZSRIaZVWoWSZNdB7m7AWC2YVr284dLY6OcGKChcJMaONXlf6p9/DvAaL10pUkj6MKMmpqB79+5In/UY/vzblXjpoQeMEIo3JfnC+fPo1r07ogYMxBmPB7u88gNOtxs9egbA09SI64YOY8JvotEkqv2OnJSC0PAI1Bw5xBYvD7+wGAFBgYzDwntbv3/zLQgKDkbGnLlY8mgmO454LQDgdLvRJywcLd6J0YqjZoW0mZkYmjQagDFWHG1+F/1CzgM6MCiik50VvlEo40Why+AHgCRuche9EXZNYwx9IsBapVAnjJ+Aw/v2SmPYKVPTMWpiCj4vLcG5s+1oqqs1ubrFWDy/Pxk2kdExCI90Yvj1YxEQFGjKWKLBeWhSMnr3CcXpkycR4lXuBMwuSd7VvauoEL3tdlRXlPsd9PhB+a1lS9h5CSseftC8gze2r6DwjeHQIWDQIOP/7t2BV1819GC6ANJWuvvnTyJ91mPIX/179vmSRzOZ6vSB4h248a570OgVjSP5AZIikHlAALnRT+fkMwV5QUu7prH0ZtrmtafnsYxAmWov/5sHYDo2v5h6eeMmuGLceHPxiz5hYH6MI+L/5jU5iPnZHMw+v/RbGMO4+vGt63IVNvrm8PZ9wJ2JHe+7zTX+LsjNYyGQRQ/cywaJhqM1cERFobm+AZ6m4+gTHoEP3lqLYcljmDgcuWeJAOeMjoEOo0aJo1+UyRVcWlRoSo3uiqubB4WFBg6PR2iEAw1Ha1jaNe8mpuO63G4MSRqNLw/sR6+QEJOOBB/7p/ouB0p2oq6y0tQeWSkBy/bycu7nzwM9e36Fu6Sg0An45wz4SqFJ8Tcge56t6oiJlaApbfmMx8OO1dXftugBlf2O6fz0OwTgs614zOxXcuCMMYjydVWVmDVpnE84i0LUfKbj9KxsLGy4l82iNEYqfDWosJHCN4L0BN/PSAyO5LoBc5iHhW9sgBbpxK3TZzBdhbYzZ0yCdPzqrb66ClPuvAvXDR2Kd3NeY8cVMyMoBEMieZ0ZMaeam/Hm4hdNXhkr0u6C3DwAgD0sHJrDwci7aTMzTVlKNhhW/6D4ERgyKgkApAZZ/PVjWVqlz6D5maDOqXRdFC4VBN2Xr4KMrGwkp04BdN2ynIaMJC/bxtPc7OOZlXlkrPhlVuR7/hjk/T1QshMtHo9JSZsyC2lxQpmAJxsbcf7cWaxbvrRD2iCqQwJBVN1Nm5mJkRNToL/r9QJ8q9wAVz+U8aJgCZ6OcczmRGxCXybRvatwq0/MOjYhEdUV5aiqKEe4y4VwpwsBgUE4ULITzmg3tMgOg4AGJlp5rd62HQFBQRgwLJ6JWxFEzRYa+E4Jmiz8quuhhYswMH4Eeni9GWKKs2wVuW7ZEqYpIYag6C8VimuqrzO8OVzqM2lUNNYbmXRhjkiWGUVgNYw4IiUApeuicOkgFm1sa0PsmOu7vAiITUjEgOFGPbD2trYu1wySHYeMEaq1RAbFmdOncO7sWfQfOpQJRT60cBGGJo/B6Mk3ILhPHx9VYLumISMrm7Wf/51S+Q9ntJt5fInM297WxjgyBfkbTfuJpUvo982HqPnF2rDRY2DzGi02G1D5NOB+4St1j8JFQhkvCp3DBnx6xw6k9djsUzJedK3yWQADhsWjd2gowp0urP212fshrrTWLV+Knz29EIDuIx3Op0PHxA1GwvgJqK2qtFx1hWgahiaPAQCcbWvzUc4loi4A0yqSXzmKGUd8HJ14OGfb202pz6KxQwM1fxwy/lbw/fvOO1/tvigodBW63rEa6dkTDy5chGHJY6DrurTyO/+7zsjKRlikE+2trejRs6ePUF1Xs+943SReOZcMioCgIPQfFs/OOzB+BAAgwFucVOSg0SIG6OA/kCFTuCkfMbFxpgxFAu95AeRyDqTJxBdmFCtLz5g3H+GRTpyy9UaIfhoAEC3UVVW4dFDGi0Ln0IGGozUoLtiCqvKDzF0rc63yfwGwKtCiN0XctiB/I1sZidLhI7mQTdrMTIRGOFD26SeWypiU2QMAZ06fwkivJoNYodpflV0rYiFtb9c0OEYmmdoqc5vzQl2b1+Sg//B43OM2l8vAj39s0fEKCt8Q7r8fWLuWva3+twkY1tKGpro6n9+A+LsmB+zhvXtQW1lp0kYxCTlGRVmqX1eU7TapZlPohc595vQpTLnzLtQcOcSO26NnT5w/fx4fb/wzevUOkdZc4hV0gQ5vbHLqFDiionDjXfeYDBD6ff75tystu2rzmhw43W44o92IiRvsw9khzk2Yt0r9UsdqPFd/r+XxFC4NlPGi0DlsMAm48V4IR78onGhoQIimATDHq081N2OtwDfhBzRxtScr2iYL3dBfMVREIFVfkTQLGFWi7583H7VVlXhj4QIfYSva1iqFkg+RidlDPOiYZ06fQntbG6oPlSNtZiY0hwPt3bt3bKiKiCpcDuTlmYyXVZHhmHy4Gmt//aKPp4TCn/S7LtiUbyLOE2E/RNMY743nu2W/ksNCLLynJTl1CgKCAk2hVgr3ZL+Sgx4BPfHlgQPsMzKIevUOYdtERkWz473l1Uui8YIygABBRsFbYqCrRP+Kst1wRrsR7nJhxi/nm8LcRBQuLSpESeFW2GB4e56b8Q3cI4WLgjJeFKQon+f9x+uPpZRHHvSeLylPnBFRd4VIt2L5eR6yom2iq5Y3Vp56PRdhTifzovDaLHxsnE97zJgzF2FOF8KcLmRkZbO0TFkZAH86GJ3pZNAxKb5OK8Bjzkhg4F0dG6pSAAqXC3zJAJsNNhssQ0CA7++awAlu+/DdegQEMM+K6GkRuSUEMpZEo2bJo5nIyMpm3BZHvyhc0HU4+kXhhfUb8cX+vRg8Msnk7SHwHBUraQexHSSGuW75UtRVVyLc5UJddSVrI68K3FVDSOHSQRkvClIMigAbpc7bfAccwDxwUUl63oXcUFPDXL12TcPISSlobmjAQYsKzp0JRonE3TCnEzabzZTv74/wH5uQiMa6WoQ5negZEMgGYVnl668L3h2eesc0fLRhPSrKdqPirju/9rEVFL4ShJIBz5w6gVYJb4xEKAs35Zt+14Bh1IuaRbEJiWwhcbKxEZ/vKkGIpiF1ajoCggLR3NAAu+CZ5cGHgHhUlHmry3uNhoajNUwYLyg4GINHJuHUiROmMBS1h/fuUhYhHV8MCdP2CeMnIMzpwsMvLMZrT89jMgjURvqrjJarA8p4UbCGN6B8YOhspM2MtdRI4FdVHiFbhwaK5oYGnGhogBbhwLn2dgwemcS2tyK40mc0YPADSEzcYIwYPwH1VZUsTCSmToptowyCXUWFJt5Oxpy5cESZK19fDBHRqj+I8xIQGIgb77oHO7d8jAo+ZOSt76WgcCXwr4TvAZLFg2isiCFS0etIJTYAoE94ONrOtGCItwZR0XtGaYxRk4xK0KKHhI7n8RopInlYNB46uDjd0K17d5w7exZF7202hZFF7y5lEVqFnmlc+Ly0BL1C7AgIDGRqvTKyrojU9GmAbSOzjl69E3hEcfAvOZTxomANr2viH877EHJ8L1LTpyFlarqpWjRP7ONDLXxxQ/LE7NpWaEpFDtE0TLzd8NRUV5RjJMnue4l/9Bkdmx9A0mZmQotwYA9H3CVjiQwN0VPjiDLi+ACYkRSiaT4uawAsKyhh/ARojkh8/+ZbkL/696baSjKIqzr++tNmZpqzjPr2/Yo3RkHhK6KuDnA62VtZ8cXOjBUesQmJiEsYCQA4ffIEqsoPIjA4GCeON6BwUz7ji8TExpk0UsTQDZ9+zJcNEc9NoSBRqwXoUNT9vLSEJRhYgcqX6AAT1jzj8TC1XtGb4w8z5s3H+b+/h+4X2gEdeGicMl4uB5TxomANr+fl8L69GDUpBfHXj0Vgr14ADI8JYBB5z7WfhcvtZgJ2/KAk6iPQpC8Oak11tTh14gSC7X1wqvkEHP2iUFVRjqL3N0srv8pCTDLDgd/GrmnoPyweoyamsPNaxbDJxVxbVYnefUIRFByMaY88IdWN4SGSjivKdmPd8qXImDMXv/mZYvUpXGH07WsKHfH1vWITRuL5B6abNrdSzaX/7ZqGPuHhAIDK8oNo8XgMmQJdZ3wamUYK/1vldZeoJEj/ocPguq4/8lf/Hn/+7Uofj2ZsQiKGJo9B6tR0VJUfZNlMw0YbEgmOKKPeGUke0CKJ2kFSCbsKt6Lovc2s7EC0N736xrvu8Wv88J7ZHj16ojR5CZJ3PAlAidVdLijjRcE/dINN32Fk1OGL/XtZ6Ob7N9+CwF69MCRpNNJmGkXRxBRjq5UbP7HrABsEG2trmdGSnDqFDWj8MWUQjRXxvDrA4vKNdbWoqig3CXTx0v4HincgJjYOH+Stxa33zsCw5DGoOXIIXx44YBmaqijbLSUd02cXeJVTUbJdQeEKYMWsmfi36np0794dfcLDfX7Dnf2/q6gQJxsb0Sc8nClP67qO5npzCrb4W+R/qz7GxPubMfaWHyIwKIgVgxQXJvw+FI5KTp2CyKhoaI5ItgCi3yafeQQAn3uVd/nyBNGxcSy9mv8Nyww4MnaGjR6DEC0UO1vGIdlrtHRXheEvC5TxotAprLwnDy5chKDgYJw+cQKf7ypmBo2Y7uwPNLFXV5TjQPEO9BswCPXVlR18mdvTcKKhAbuKClF9qByrt23HuuVLkTI1naVrLvJmFpH72apGSrDdbhB8bTYMGZlkKu5GYSbSiyBjLTl1Ct5YuMDkqi4u2GISvPPn7TH9//NHOxrzwAMXdQ8UFL4xXLjQUd8oMBDdu3fH2fZ2VOzZLX1uQzQNdk3D/uIdJr0n2oZCrGQMeLqQlswbM8UFWzBi3ATUVVcyY0IsBinjq4RoGgYOjzdlM1GF6VMnT5g8qjR+kdEh1jnyV+RVDD+T0VZaVIhgu50tgh5SjtXLCmW8KPjgw8yubeeMNtJ821pbWWoypT3KUjBpIOHfiyx+iluLKyYbgPRZjzFZ7+qKcgAdaZt0TBqc6Jj8eVs8Hui6jpojh3DG40FsQiL6hIWzFEo+60Esqrjiycexett2OGPcTFaczsH/lXmZ2Ge88ZKX1+X7oaDwjeK554yXF7quo+zTv7NFANBhlFAEhEIqMr0n0WDpCk+EkJo+DQ+/sBgBgYHYs72Dv/bn3640CcnJjivWLaso222qMC0bfwBY6jPx5+DlFcRQ8IjxE9DbbkeLx4PB3srZKgPp8kM5uBR8cMNg7z824Jw3ukGrj+xXDIZ+bEIi6qor0VRXi7W/fpHtu3lNjkkThtj8E29PYy5pOlbazEw2YPBcll1FhSy1csmjmdhVVIj+w+MREBgIXddRUrgVby1bgm2b32XEYTqmjg5NGv48gBH+2rb5XXyQtxbRsXHoExYO2GxsgK4o243nH7gX65YtMQ2IpFHz0Yb1qKuqZIXbZq9cBQCm9tPnxP9hOHXqG7o7CgpfE4sWmd5u2/wuE5sjkBcyKWUybDB+U+uWLzVx0OgZ5z2ePs+9APH3kTFnLoKCg9He1uZXokDcj37bpNzNc+lmTRpnMj74sQsw/167gtSp6YiMjkHq1HQkp05BWKQTQ5JGm8aa2ITEjpWUwmWB8rwoyOH9If65530A/mjKmsl+JQdVFeUsHTI5dQojzUnrhPQz2Pyk7SILrfCrI9J2oBVSTGwcNIcDug50s9lw/Y0/wGvz51nG0GlgkpFnKX2ZCige2bdXOnCL8fWJt6Wh6P3NmDXJqP7My/5bFZE0rRRJHAwAGhsv7l4oKFxCrLjtVkDwavBeSJ4XVpC/0efZF8sJWEkqPOgtltqTI70T542vESZDV8KzMlBttFMnTsAZHYPpWdkmAcuugNeQonCVDR3eWWofzj0Ok+iUwiWFMl4U5PBmGr2+rQVAR0yYBqlqbyaQqAkhgs/0oRpD4oDDay3w+9BqjlIZA3v1Qu8+fZjqJa+IybuxCTLyLGCOscsq6vJGT2xCIlxuN5qPN5h4PLRN9aFyzF65isXJaRu/g6rDYf2dgsLlxqBBPh+RF5KHqMBLz/j0rGxERscwki7/u6wqP4iMrGwMHB6PMKehBdPa0sL2LcjfaPptys4l+111Fp4Sw8jNDQ3Q0bldIVPf5XViKsp2s/AaGXHDRo8xijz+m/cgNuCzJ4Cxv+vkZApfC8p4UZDD63kRpfpF4q6oCSEaFLJaQDLPBinyipM+vXe53RiSNBqNdbV4Y+ECAIbLWeSgyPYVj5mcOgVhTifCnE52LVYZQ8mpUzAkaTTbj1IvSR+GMhOIm0PqpAoKVzUOHZIaLf5g5VUkg+Dwvr2oKNuN7Fdy2O9y/84dLPR0srER586eldZT4sEvZsSSHf4gVrgnYi2fuWi1qLASuAOsEwF4b3Ry6hSca9uIHt2NDrn+us56U+HrQhkvCnJ4PS9WP/aYuME+uguzV67CiHETEO5y+RgU/EpJFKTiV1U8aTdE0xBst8MV48ZJb6jlC+8ACcDkcpatmKxWZ7zrlx/oyPAQV3nkPi8u2MI8T6XbzIMiZSL1ttu7lNatoHBFERfXpXR9fjFitRgo3JSPmNg41FdXYfW27Sgp3IqklMnM80K/tbckXk4RZLhERkWjvqYa65Yv9Wt08OB/xzYAu4oKfUJesvPxHprSokLGuePLl8gMN3ExV98TWPuTTpup8A1C/za97Ha7ruu6brfbr3hbrtXXhaUdLwB6bEKiPnvlKvZ3Y8WX+p/+dVDfWPGlPnvlKh2A8Xn5l/rLGzfpq7dt1+/++ZNsH6vz0D50DHrFJiTqC3Lz9D98VqznH67W8w9X63m79uqrt23XU9OnXdSxAOip6dN89uWvif5/JjdP31jxpb5623Zpu+kc9L1Pv5R/qT+Tm+d73WfP6rhwoeN1Fdxj9VIvq2eSf65Xb9uu5x+u1ldv2255HHr2//Svg6Zt6Xf8TG6e5TjAn4s/ltVvkN/vmdw8fQF3bNPv2GIsEM/3TG6e/s6hKn3xxk0+v2V+f3E/q5c4dqrXxb0uZv5WnhcFv2BFFbn0Y3Ll8mEeq5RnwOyBsEqTFr871dyMUIcDZ9va0DMwEOfOnZPyVwjkTXG63ViQm2dKdZaFl0ibgk+Tnp6Vjeb6BpOMuXgOn2u8nYt5w0wYZuDrGZ0/f/E3QUHhEuOZ3DzoMPgdGVnZSEqZDJfbjaa6WvQICGDPNw+RA8NL9gNGzSMitPJ1jWThHcCXiCv+jsT9SJPpFBeO5tOhZcrc4u9e9x77tMfj4xkm/s798+ajtqoSbyxcIPUc8d4pwHdsUrg0UMaLghxezsuktDvwxYF9aDhag+pD5UibmWly5fIhGn4AsXIxi4OVLFOHNwZ4wl7K1HRpNWo6/6nmZqa6SSJzAPDRhvVIn/UYPtqwnm2vC3/TZmYyt/GeLuhA0LWRIUcpmwRS6+1lt+NX/EHGjZP3t4LCFcQL0/8Db1fWAgDiEkbCZrOh74CBsIdqKHp/s3TBIAul8Nosve12AMDZ9nbTgkAsDUAiePS7tspWEkn9FM6VGUFUTTopZTJGjJuAlx56ABVlu31+90TGFQ0dakP2KzkIc7oQ5nRZZirxiyNsV8bL5YIyXhR8IegVhIZHIDTCYZBTo6Lw/ZtvwWtPzzOtoooLtrCCZoDZMBHj5jRYpaZPY0YQVYmmEve8McDL7FMKtRWXRSYylzYzEwGBgYgeFMe25TMIYhMSEaJpKC0qNMmF02AGgAl28TF0inlnZGWbBl8aaJ3RMYBN6MydO7/u3VFQ+GbAKe1e6N4dB0p2wgagT1gYAODokcMoray05Jt0lq5MopCH9uxGLXcc8e+AYfEIczqhAyahPB581iFpQD0viOqJbbEBsNlsjJgvqzDNyydMvC0Ndk1jXpvk1Cn4aMN6THvkCfQMDGQGjwiee/fQBIuNFL5xKONFwRc6mF7Bts3vmrwfD7+wmKncFuRvNHlLRD0Xmsj5fWZNGsd0XKI578iw0WMQ7nShsbYWu7b5EuaAzgdLPr0zNX0aS3eW7cevrmavXIVRXslwnqA36fapGHvLD9HW0sIG9Pjrx+K1p+eZMrB4XRqWPdUvCk31daitqgT63/4N3RgFhW8QY8cC//wne1tXWYnigi0YMDweJxuPo8XjYdXX+WwdMfQDwCc8A/gq7xKoiOJTr+eiR8+esIeFGWU7/DRVlFwAzEq5vBowf34AzDtD5/Yn6eByu1k1+dAIB4re34yn7k43XasIU7r3BCitl8uEb1032+12nDx5En369IHH47nSzbkmcYELb3eba/6Orzmyc8vHyMjKRrDdjl7BwbCHR6CtpQX9BgzE/uId+NW0qWxF8//Z+/bwqKpz/XeAZJKQCTNDMgNJBpAECDkkkIBVEBIoWuyvGkypPZ7gAS0avNUqOUQUURQvNBTKaZXWKKfRAzlUioHU09YeETLkVD1KLkCRSxLAXCCTkAQCIRfC/P7Y+1uz9pq1J0EF0e73efIkmdl77bVva33r+97v/bq7utikTwOfX4xcXb2Qh0WsPyJCTMvmQVL+nrpaJiwnQtSS4AfZuKRkvLBlG0JUcblWTyPCIobAHBLi16as/IHGU1NZwV1QQ9TawDUELuNoxVuFMEEpBdB8skHhtUl+U4YNvaMAMP32DDQ3NDBeiwz8e0Lp1ABw/swZ1FYrFanFQqliBqGY0lxR6kZsXDyiYmIBKIstyn6kd9C9owhpczP7lfH08rYdSEidguOHDqKlsRFhahkAmR6UrI+XfuH7bsC/9fcmGCBczvxteF4MXBZiRsezEEzM6HhMmp6G7q4umENDAQBedTCMHnUdS4OmUA4VcyMkpGr5ImSA8EXf9JCeOQ8/XfNLDBw4UKrzwrty9cCnVgJKoUl+oPrt08uw4InlONt6Gq2NjThU9qnG2CKIq7nq/fvQrnpjzrW14ZcBrqcBA9cKJs1IQ4UgAcD/prAw74Wg3+R5lRHdCTwvpXDdGjy65pcYMHAguro64amt9b0v6v68V5eMIl43CgCuG5+IIZGRaFO9nOSxzcrJRUr6LHi9Xo1x0y6Qe0WD5IIa6mptVPg/CalT4PV6Nf3SO6dfPvaIxmtt4MrCMF4MaHDOV6YIXskLKBu0oqJj0NrkgQmANcqBzo4ONBw/xgaeuuoqTJqexgYAUUCKeCe8t4LXjpG5qrOWLMXAgQPR29srNVD0lDtl52KxWjEpbSZMJpNmoKI2yHtEGROAv5tcdKMTh6Z4Y762IKMBA9coLnb3MO2j6v372PtD3tK66irUV1f5kVoBHym+vqZK2jbJ9Feo7wS9N7wSL+BfjZ1E4ESjiMK1QyIjmceHxpX2tjYlBOX1orXJw6rQh1ksCLdacf/KVRibMhlOl4txcag/fKgpTCUcey9d0j0vsQQJ29nAFYdhvBjQICzY93f+R/7fi1lFvEgTAI2r1u5wakoJ6KVV8wRXL3ypj3x2gZilwHtW+jJS9MCv4miw6mhvl6r8kpGWuyEfLZ5GjEuZjHCrFYVr85CxKBsOlwvjUibD4XLBEeOCzeHA4fK9iFvzC1R/od4ZMHB1cezgAUycoVSPFjMKMxZl+/HCeMy+8y6EhIVh9p13aTKOAG2mEL+/uMCQVWMXxxe+zXCrFZV73Mzg4I0jwOcEqas6ilX33M0WIefOnIHJZMLwUaMxZmKK37GJi1dR6kZnR4fueQG+EiRpatFG9BjClFcLhvFiQBcPvuP7O5A2A734cUnJyosfHY1zbW1+pQQIesUbqVAiHYMvGSBbuVHdlPS5mf1S7+Qhxqr1shyov3kPZeOp1wsQFRPLeDDXjU9EVk4uJk1PQ09PDwAww8VkMsEZ68Jfp0zsd58MGPg6ERI2GF4vEDk8WgkRcQsHURVbfKdlYVqen6JX/qMv6BFsRWPq8fWvYKJgXIlEep6Ua7HZcPJ4jSYLitqNionFpUuXcLjsU7h3FAUMP/Pe2+m3ZQDbHzE8L1cJV8V4eeihh7B06VIMGzYMlZWV+OlPf4pPOJa7iB/96EdYtWoVRo0ahaNHj+KJJ57An//856vRVQMc+AmeF3eilQ7/0vOrKxqk+iqeRpAJU8kE4UiHhe9TSvoswOvVTZ8Wwzl6BlhfqN6/DycOHYTN4UDD8WOwd3QgMjoaJkAhM0ZHo6m+Dju3bsGcrAXw1NfiL5veArZt9TVy7lyfxzFg4OvCyITxAIBLvb3YuXULYkbH66YUA9r3pq7qKA7t/VTDa8tYlI3pt2egrblZ4yGRGT+XC720azHsJH4GAH/e9JZfuIjf5zu3zGHelsUzbuyXZ7dkRxHa29qQaRguVw1X3Hj58Y9/jHXr1uGBBx7Axx9/jMceewzvvfcexo0bh6amJr/tp06div/6r//Ck08+iXfffRdZWVnYvn07UlNT8fe///1Kd9cAB36C58WdZEYJ7ykJlHEAaLN80udmavRTRAVeUfQuXF3hJEyeghZPo0KuU6vZ9nUOADTiWPzvvvpavDFfk/oZGz9GIQrvKALgi92nzc2EzeHA8UMH/Qe9iIiAxzJg4OvE2ZYWRNjtGDBwIGJGx0uVsfWqpssWA7yI44G2toDK2/0F/z6KHBjeAyzTZZL1UfTAElGf97bIsgn1FkG/yFOknWR8QQNfLa648bJkyRK8/vrrKCgoAAA88MAD+MEPfoCf/OQn+PnPf+63/c9+9jP85S9/wS9+oeScPfPMM7jlllvwyCOP4MEHH7zS3TXAQZzgeVE3WYVV+t3XiorkxyfcOA02hwOAosZpczhx4dw5jEwYjxvnfB/HDh7AmIkpmoKJaXMzceZ0MyKjo9HqaURTfR0K163RPWZZyS4kTZ0Gh8uleEK4PspIwSL0QmQU6iL5c/qbtCoCaVYYMHAt4vl75iMrJ5eRdnn05anUCytROEk0ei4nfCQzFmhMkL23GYuyFaVtkwlhFgs8qpeFPza1SVy1cKuVhY6Ji0Njg7iNqBCsOR9jALhquKLGS1BQECZPnoyXX36Zfeb1evH+++9j6tSp0n2mTp2KdevWaT577733cMcdd0i3Dw4OhtlsZv9bVIa4gS8J9eXjVygEvQqrgVIQqY2MRdmMxe+pr4WnvhaOGBeGjxqNCLsdXZ2dAIDgkBBExbpQ+m4xHC4XUtJmYuKMNETY7Og4exZtTc0ICRuMqJhYpM/N9PNy8PF2a5QD1igHPLW1Uo8R1SeSeYwom6C+porVf3HvKNJmE6kQiYJxSckGWdfANwo894vXUZJm1cDfEyJ6VojQmpo+C3VVRwOOD3rtysoJUIo0HUf02E6YOg22KAccMS6M5Ui5vDDl9NsycOZ0MwDlfRV1o+i44jZJU6fhSGW5NDRu2C5XD1fUeImMjMSgQYPQqObMExobG5GQkCDdZ9iwYdLthw0bJt3+ySefxMqVK7+S/hpQob55/OAhFmfkfxP6Wp2Rx+Vw+V7sKd7O2h47MQVHK8vR2XEeO7duQUr6LAwfdR2a6pQVU1ZOrtotE0wmEwYPGYIwLgQTZrH4eU+oL5Wlbhwp34uoWBcbeMWBLpBGBQ2+s++8i2lFUN0kIgfyhl31/n0wAaxo3POXffENGPh6ID7/fM2eQ3s/lRZG1SusWFayCysKNmGwxcKMfHHbQCJ0svGG1H4BYMjQSA2ZP3dDvqY0yUv33eMnQMmD/iehzJIdRViwbDnsDicWLFuOkqJt0m2yliyFzeFET3c3gK+Gv2Pgi+EbL/f58ssvIyIigv3ExMR83V365kON1xIpdkbGHQizWFi6M+9l4Sfu4o35qCj1SfuLoNXIBbWCK7VT+m4xXl+5Aotn3Ii3f7UeT86bi8o9boyZmILcDflw7yjCnuLt+I8XVqK8ZBcOl32KcvduvPXzF7GneDvOt7dj+m0ZbGCjvpS+W4zNa/PQWFsL69BIFuKhQTRjUTbLJOJTuXlQO4Xr1qDCvRvl7t3K39x50jWarxpZYvE3Awa+CRDf28J1a+Cpq8XOrVsQbrXiaGW53zb0fojE3tT0WUhJm4mxKZPRrvJdxPGB3sPcDfmszSz1XQoVxhtqc/ptGTABKH23mHlKKYTLZzPRPiVF29hYw4O+jxmtLETS52Zi4CBlLd9UVyvdJjV9FgrXrUGrpxEtnkZk5eT6jTsGrh6uqOelubkZFy9ehNPp1HzudDpx6tQp6T6nTp26rO27u7vRrVrBBr5CeJWXN9RiUcTboPWm6IWONF4H1QVNAxUx8nkjQcY9AZSBtK25WbdII7Wbmj4L7h1FmmJrfLuAr2BjuDpoip6jQFlRfP/4Oi2p6bMw/XZFuK7V00iXDIC26CMWZF3WZTdg4OuCWDOIF2mcpJYMGDMxRZPZF6hWULjV6sefiVWlFAj0jpPXh1/g8GFoQCHrH6ksZ7WK9Ph2eh4d4vPwsgp825YhVnjqlPpOr+35iIWPxPbJiKrkRDYNXH1cUeOlp6cHe/fuxezZs7Fjxw4ASpXP2bNn45VXXpHu8+GHH2L27Nn493//d/bZLbfcgg8//PBKdtWACuJrwKS88CSXfUGoM0GDk1iEcWRCIgAglAvl9CctWcwKorToAx/pk4T12pVlEGjk+rnYt94+4neiW5oPN9VVV7EClnTOXySTwoCBrxOl7xazbD6eEMuHbWQhGBH0Lon1gHgPSZjFgnEpk3G2tRWVe3zcMT6jTxwTyIAiDgtP2NWEu9TwNE/CZSReQGp8kefHBODW+QtgczhZ2RGx/ctJTjBw5XDFs43WrVuHN998E59++in+7//+D4899hgGDx6M3/3udwCAN998E/X19XjqqacAAP/+7/+OkpISLFmyBP/93/+Nu+66C1OmTEF2tuGauxrwegHTAMWIyd2Qj8J1a/y8JYB+NWVrVCSa6utwQQ3lAP6eD9kLr6fTwCvw8qJZ9L1oQAH6aZti+zz04vF8TRdRZKu+ugp11VVskP6yaaAGDHydKN6Yj/tWrkJbczPCLBZMmpHGiOwyz6ce9BYV/DuYlZMLmEyIsNuZ/AKg7zF1q5IEZECRKJzFatWMT3xCwGCLxY9kb4JcaI8fz45UlqOnu1tXmE7P2xSXlAyY9hnx4quEK268vP3224iKisLzzz+PYcOGoaKiArfeeis8Hg8AYMSIEbjEVTb98MMPkZWVhRdeeAEvvfQSjh49ijvuuMPQeLlKGKAWFTOZgKjoGLbSEl2uTJ6by7ihjAS+Vkm41YrJs76L8dffgGCzWbfAmTgglJXsYsZTavosXYVOIs/yqyk9o0ZEXFIy7l+5ClGxLry3+S2NC5gvCkerRT4bKWNRNibOSENbUzPm5+Ri89q8vtNAe3qAoKD+3QgDBq4yMhZlY1zKZADA8UMH0dzQoCGyB/JO8tB7D0Qvx9mWFkTYbCwzT2xb5jEFfLWW2tvamKeIMP22DBytLEdTfR288IWwTfCFw3jBzee57CrRw1RXdbTf50wLLHx8o1GY8Srhqijsvvrqq3j11Vel382aNcvvsz/84Q/4wx/+cKW7ZUAPauC5qaGeuW9Fl6us1gmfFllStI0NPInX34CQsDB0dnT0Oz5MmQ4PvLAaRa+9iuaGBrYS4sNRkdHRONPc7OfVodg631++Qi2geJYcsS4ASm0WvpyBOJBRpgNPWE6YPAVRMbGwORysaKOfYdbbCwwcqPxNvw0YuAYRbrXicPleVlkdgIZz0t8wLQ/ZdxTWOdvSgnL3bmxW64P15a3hwRtCIt8t3GrFmIkpqKuuQoV7N7Mjpt+WAafLhVGJEwD42xcyFWHaj/olng+lV7d4Gpln2DBcrg6M2kYG/KESX254Nx3V+7tQVrILE26cBk99rd9gwvNR9ISoKNUwkJiciMJ1a/DAC6sRHGLG7DvvYkYRkeX49sOtVr+q1WL2AeOoREdjfk4uK752tuU0Ll68iMJ1a3QF6Vhl6dszkDR1Go59dhCFa/NQuG4NFjyxHIOCgljRRlErJi51MqorK77onTBg4KqBFiO88SAzJCgNmgqZmgBM5EJMeqrW1JYJCvcxwm7XZCLxxyD0VWJExkeZn5OLilK3hnNDRs6EqdNgDglBZ0cHM9BE6IWwAZ/hRRWpk6ZOg83hxKCgYJS+W2yUB7iKMIwXAxqcuQAMCQXgBZIju/D4+lfgdLmY5D3v0iXvBxkPgHaw41dHMaPjAUCTUSTqMPCTfknRNlZ4McxiQV11lZQPU1ayCwuWLcdRVTSKvuO3ob7QwEqF4ro7u/AfLzyniePLhLioLfK0pEQpqsCxcfGwRkWiYo+bha9ErQyD0Gfgm4K+Mmd4z0RK2kwlC9HrxeHyveju7GLPPy9oR+Fjvl2qceQF+k1y72/4JmNRtl+BRr7vvBCdXjuiQSQToYuKdWHMxBR8fvQwQsMteK/wLbz9q/VYK6fJGLgCMIwXAxrcnA988jP1H5Oyarpw/hxMJhOGOrVCgbwQHA18vBeGjBINfyQ6WlGsra7CxBlp+M4tcxBsNmtKAPD1js61tWGSUJBRFNKyO5y42N0tNaxk6ZOAL6zEi26JYS8eZPxQzRQTwLw7vIudH4zps19+tbfIgIErgr4Km/LpxeFWq8bzYg4JQVdnJ3vv+fdIbLd6/z48f8/dukR8GWQZRLK+ypSv+e9Jq4U4LRovaT8MJMqG4onDI8aMY4szA1cPhvFiQIO9deof6hKj7XQzLFYrAMBis2u2lXk4KF5MRFd+O5a5o6YXn2luhi3KgZYmD0wA24+UbPlYNk/M42PPO7duYSEpHmIISPyf57fw5xOI6EuDLgCN7gsAzM/JRajFgp88/SwsNjszxqbflmEYLwa+EZCpxerVEuINiLikZBaG5VOp9SQO+LYjo6PR1tQckFwfl5SM68YrEgyBZPfJwGlt8vjtz6v28mMTX/6El0OgsYq0XkTUVR1FSdE2pGfOQ0xcvNRba+DKwjBeDPiBOKa9JuD4ZweRMiMdnR0deOvnL/a5L7mMd27dgnGpU5jhQ4MEX98kKydXqTmkKlpWlLpxuOxT3Dp/ARrrfCXrZcQ8yhiocO/G4hk3suOT4m0oJ0sO+Mex9Qoznmtrw/TbMzAyIREnDh3UCFrx4F3Lj69/BZNUNzpBI8716EP9vfQGDFw93K31YPBcMpG4bhEWDyJINoAntIsy/+J+NFa0ehoxcXqaJmOQR8aibAyJVCQYNuvwVAAfl8Ye5cD8nFzmIaEFU9vpZhypLIdJ7S+/cBF5ckTmJ60Xfju92k39SSM38NXBMF4M+IElx3h9oRs+fVCmp0KDldPlQlRMLBJSp7BsIwAadVoaoKhtGuBK3y1GzOh4DBkaif0f/k0asybYnU5loBLUmDMWZbP0yD3F2wOGksRz4MXnKIuIBtRAK0jSowi1WGAC0NHezoyeXz72iGG8GLg2UVCg+ZdCv7KqyTJ1bIKMZ0JttDU364ZxSAIhKDgYtUcPY8LUaUjPnOdnBGg8txKVXMLmtXkspOsF58mNjsElrxe2KAeGDI2EyQSU/tHX17ikZFisVlSUuuHeURTQo6u3GKJwMXoMfaerBcN4MSCHOgJohNqGR+sqb+ZuyEdUdAwrWEYS3oB+yIf4K6T7wItRiYOdaDxY7EMBgP0m8IaEqP4rquTyx+E9MnkPZbNBUpMmensGvnPLHPz26WWaAVYMJwWsdTJ5MrB3bz9ugAEDVxgDuNJ2LS3Sd0JPE4WHXnYOGQ4Xe+IRGz9Gyp2hEE5ouAUhYWHIWrLUr/o0LTxyN+QjKiYWJoBJE4gLnFXce3hO8LxcunQJAwcO9JNs4I0vymak2kkZi7I1CzZxEaVJr749Ayh6xEiVvkowjBcD/uDquk+/TUkPtkU50MpxUwBfOvHj619B5PBodHV2IjjEjKb6OpamyId8aHUiVozNWJStCEqp6c798ZC8tfpFFpPmQYYEr9VAYlaRw5UMo/7Ep6m/RNB17yjCd26ZwwZYPRcxzxGgFeAqfoP//V8gJKTve2DAwNVEZCT7k5+gw9TaZqRaK4PMq0mLACLkkgHBK+JW79+HwnVrkLVkKcrdu5GSNhOF69b4GUxxSclY/kYBbFEOnD97FsGhoYiKiUVWTi5Wqd/LyPnkLSaCbn1NFeZkLYCnXglTixIPlCxwtLIc4VYr5ufk6oa8RJAhBsAQqbtKMIwXA/7gfK8VpW6MS0kFTCY01tVqao8QREG3+poqpo7L1wYhg6KCy07iM5YqS926g0Z9TRW6u7pw4fw5xlHhuS4iyAMzMiER1shIHK0sR3dXF4JDzGx1V71/H/OokEYF+0xCHv7t08s0BpNeJgagcAQofDWwqwu9ZrPSseDgr+QWGTBwpcA/1+fV2mZRsa6AatWyEHLxxnzmvaCxIdxq1bxvxBkJHRzOtnWOHAmTyQSb04n0zHl44IXVCAkLAwB0dXVicEQEYDKxNZYeOV9GzI0ZHa+pBM2TkdPnZmIiV4Cy4jILL9ZVVyl/eA3b5WrAMF4M+MHrVW0XrzJpD7ZEAFAqvfKrMlml1pKibXhtz0cashtpK+zcugUVpW7NKo53OQeKk8++8y6EhIUh/Y55CA4xa4wNGUha3BoVieaGBnS0tyM4xKzoUXBVbHnXNf8ZpYJeONcOi30oykp2sSq7BFG1l1acfCjMC2DU2jxDqM7ANQ3eYOGNgcK1eUxWQCwTwMshiCFkCs/S55SdE5eUzDSRVvxuEwYFBTFiLXlcLvb0AACG2Icia8lShISFobuzEwc+/hDuHUUaKQX+GJTd5HC5cOZ0M3Zu3YKY0fF+Eg78PoCPjEzGiliAsq/0afIMWaMc8L7jGzsNXFkYxosBP1DSjMnk82DwgwUNBOFWq7RSK7mCyUNBUv+Zix/GZ598zOqniBWemSERHY3jh7SDBrW5c+sWzJm/wM9tTEZUyY4iJufvcLlwprmZ1Vri9RlEjgufNk2D1LiUyWj1eGAdGinNJuD7y4vT0fWhvhgwcM3h/fc1/8pIuuK7RxO/nmeDhZCjfeFZ0SvCh5OoNAegkHdpvm+oqUJouIWNHwuWLUdTXS0LRcs0mMioeuqNAtgdTni9XsSMjtdkBMqypUi4LiYuHu4dRX58G9l+ovFmsVphi3IgbeetzBt0yTBerjgM48WAPtQ3USTqkZ7CkfK9KHfv1qYFA34eip1bt+DHjy5BcIgZXoDVGxHdsbwhwadmipoLCalTYItysIJujMhnMrEwEYV8TFCMJ6o7REJV4nHFmDYNQmdbT+NiT7cuT4ZPESXQgE0CfAC0Wi8rVgCrVsGAga8N3/2u5l/RYOHfBzEdWAwTy4i6Ms0XAr3n961chehRo9FwvIZ9z2vJ8MeffnuGtPwGj4xF2bCr6tdtTR6Nx6W+pgrNJxtQX1Plp2fDnx9ffoT3whL5HwDzLNF5VpS6Ue7ejblnDrB40ar/uZybYeCLwDBeDMihzt5vLgzH9H/z/8oEwBHrwpurXwzoXYhLSsbsO++CyQSmRhtI3ptfRRG3RtRc4Hk3pM/Q09WFYLMZnvpaHPjob4ycNygoCJHDo/FMwWZc7OmBp74WYyemANAn4dEgdbh8L5yxLgyJlHteKEuBV//VFG2MjmGqo5p06ZUrDePFwDWDoL//PWANIT2NJAC6StTi+yCrGP3kvLma7Dy9bWWhXR68J4T3EvOek+aTSoVsqpMG+GcclpXswq13L0BbczPKSnaxdk0A06EBwIj/fFgKAFYu9PVpldaxZeAKYEDfmxj4h4S6gph66W9+X21emwdPfR2GDI1E1pKljABHgm80+QM+Fc0zzc0+Qls/QAMkua09dbV+mUWAMuA0n2zArne2wlNfh79segu/fOwRxIyOhzUyEtX7K9F8sgERdjvsTieiR41mcW3qa1xSMl7etgMbPypDeuY8ZpTYHU4MGRqpKfAoO3aYxaIhAdIA3tRQj+AQM1LTZyH0/Z2+HTkxOwMGvm70JCVJP6f3GfCFfPS2Sc+cx94n/t0FfJ7I3A35mrEBULy4MzLuQFZOrmZbXm6AxoDmkw1SDyiR7rOWLMXmtXksXXpFwSaMGDcObSr/pbLUjRZPox+fjvqbmj4L41Imw+ZwIDV9FuuLFz4NnOKN+Wg+2YDgEDMLSxHJH0BgCWADXykMz4sBP3zyOXD9CPl3oiw/rT4YUU9HC0as/CxrU48UR0UaSXNBrFobOTwa18/+HmwOB9LmZvq5ttPmZmJwRAQGRwxBw/Eaadn7hFQlzXHhsuU49tlBVJa6NfwZSpnmPUfkbq6rrvKr7cRnWhRvzMeF/fuAS5e+8D0xYOArwx/+0K/N+lNPiEi2E6ZOg3Wokm4tvt963DBAo8qgEYsTFwuBlGzLSnbhxjnf1/DgMhZls+KR8Hox+8670OJpZDw2EUT0bW9t1YSxqH3yLss4cnSOaxfCSDO6ijCMFwN+uOHXwKVfKH+bBN8crXK6O5UwzbjUKSzNkMIkYRYLJs1IY2UC6Pe5AAqdgaTHxW1Ew6Sy1I1QiwU2h4MNhnx69qTpaagodWu0Zvisg8mzvoue7m50XbiAxrpaTFIFq4i7s6JgE1LSZ8Hr9WqML9GdLpND19WH+M1vgAcf7ONOGDBwBfDDH2r+1UuBNgm/ZaC5uqmuFgc+/JuuuGThujVIn5uJcKGGER8CzlINoXL3bqmKNf+bR2r6LIVbxqVPE1dlsMUCR6wLkcMVJV+TyQSbw+FnRGUsymaJBJ49teyzspJdWLhsOWxRDlisVmxem8c+F3VtDMPl6sIwXgzIwWm9UL0gEmuTidZ9fvQwvF4vzKGhiB51Hc40NyNyeDQyFz+MkLAwzL7zLqbLInpaqMZJIPE4GaGQDJPSd4vZoCIObuKgJ3qHACUNOyg4GK2eRryxcoVfOyYA8HrRqpIACXyMnjxLbaebpStHPyxebBgvBr52hPT2SnkkgJK5FxMXjxJV9VoG944ixMbF48+b3pIKN/KLDioXwi8AeP5MIGNJlGgQx49JM9IBeFlfRbVdMjhuvXsBho+8Dg6XT7cmPXMeJs1Ix/mzZ3H6ZAPCrVbct3IVElIm48Y530ewKippdzqlmVY0BhilAa4uDM6LATm4kSR3Qz5S0mYiJW0mUtNn4dhnB+GFYt+U7ChC6bvFiLANxYCBA2EymRBhHwpPXS2aTzagZPs2P76KGNemGidZS5ZqYuI8h4Y3FGib4o35LFwj+x7QcmeIf9N8soEZXbQqpD6K8XpAWR26i7fjpfvukVbGnX5bBsIsFnR3dcEWGYVzbW1yUvK5c1/qlhgw8FXjwVdf0zW0+VCNHsRtRN4b8cLKSnZp3lcZNq/Nw57i7ZriizIeHc+TiUtKRtaSpbDYbLDY7EhNn+W3D73TJUXb4KmtRYTdjnEpk9n4k7VkKSLsdoQPGYLQ8HBMmp4GR4wLMJkQZDbjUm8vAKUUCZ1D4bo1GqHNf0nuNvguVxmG58WAHOR5ARAZrdQF+fzoYTbwkHBVavosppXwyOpfYFBwMM6fOYPz7e0YMzFFo5xJqraiN0Qvm0BPOZP+FzMk9FQ+RfVb+s3LlOvJ/fMCezJeDs/pCTab0dRQrxmc+VUibppuiNUZ+HrR2an5N5DsPR+eFdOLxW2I7yXWDxO5KoGOJ8t4kglB8h4aWpBc7O7GhfPnpdoyvOcl3GrFkfK96GhvZ30vXLeGacn8edNbSJubiTCLBZ76Wlxob8ehsk9ZkUZZplXxxnysXvh3I2x0lWEYLwb0ob6Mgxs/xYXhNyAs3MIGLype6HC5sKJgEwrX5qG9rRU2hxPdXZ2sYjSv2MkPQGJxM1EIKy4pGeFWqyYzoK/BVKbWC8gH6ECpobzBwQvsBZvNGjIyf2wqKGkCEBs/BvNzchFmsSAqxgVrlK9ujEbvpbUVsNn6ugsGDHx14MtT9OEJFMt6UBoyr6jNv0dU44zPztPjqqRnzsOCJ5bDU1+LN1au0M1kslitaGtuVjyzObk419amqXAdGz9GqTkWGopBwcFSbRkiHhOpuPTdYs27X1d1FMc/OwiT+ve5tjYWjqbt3v7VetYnPqxM4phBTT5Cc83pgJfVwFcEw3gxIAXneMF094+w+aa/aUI/1fv34VxbG1PYPdfWhje5Yon8oBYo24AgrtAyFmWzAUQc2KgGiVjkTTwmAA1BN2vJUkRGR/sZIKJiJvFXAEVgL3Pxwyh370bchCSFjMwRcvnaSPXVVZg4PQ0xcfFMNM/r9aKpvs43eD+8GBg4UPl7yJAve5sMGPjiiIjo12a8d1GmqC1uJ75XMo9N1pKlsDudsDudLEMI0JYc8QJMR+nAR20a4UreOxscYsb59rMICjajvqZKoxW1omAT4pMmwgR9UjFlJtF5BSIH84sii9WKlPRZSi2Ad8AWe/Gr+3VZDXxJGJwXA1IUcWNN2MBLOLT3U9RVHdVsU7wxH4fL9+JsSwscLhfqqo5i8YwbUVd11C/mnPdQNkr/6NNX4XUhqK2KUjcsVivSM+exlMn6miq8tucjpr/C6y54AT9NCAKv3TD9tgxW1ba5QdFlmZFxB+YL2hILly3HjIw7EGaxsHh2zOh4BJvNiJuQhMjh0XDEuPy8Qc0NigAW9atw3RpUuHfjcNmnOFK+l+nbVO/fBwQFfbU3yoCB/qK19bI2F3VeCtfm4Uj5XrQ2eZhSLekk0XY874zeq/k5uXimYBNWFGxi73vhujXoVkNYYRYLaysrJxep6bOUNGeAkfF/+dgjKNlRpNF6Yd7ZPW50XehEcEgIZt95F+s/GSURdjtamjx4feUKqV5NWckutHo8OFy+V0MCzt2Qj/TMeZpted5OqFqj7RJJIBicl6sKw/NiQIof/SdwiRwtJmBGxh1wulxorK3VeDo8tbUYlzIZQ4YOxU/z1iE0PBwtnkY/FVuNC/p2VTo/xMy2IU/O9NsyEKMWbit9txiz77yLqevmPaQYKfzKTi/9WlbunlzKJIhFIWra3+FywRrlQPSo0ehob9d8x3tuDnzkI+TGxo9BUHAwjlSWazRgKB6+omATW9VJtTKM0JGBqwXe06dTOZA4XpTaTOEWIqufb2/H2CgH5sxfwHRdAGg0k0RJA95jQ5lGvHZTuOpRCbdacd34RHi9XrQ1eVhoioyjcKtV452lcFC5ezfe+vmLmnpqdPwJU6fB7nAGvCyp6bNgjYzEgY/+xt5fCheTqjcPi9Xq8w55vag9esT3pWHAXDUYxouBfiMq1oUxglFSvDEfE26cBpvDgWGjroM5JASDgoJ1swp4cq6oXMsbCmRo8HWNZFyZ/ujCiAQ7no9D7RRvzMf8nFy0ejywOxxISZvJBlpaTRauW8NE6ygU9cALqxESFobQcF+lbN5dLk3/7OnxeWCM0JGBqwGhEOP3/mcn/gr/tGN+0q6rrgJMJtiiHIxvMtTphMlkQnvLab8QjPg3H76hMJBM2Za8MdeNT8SQyEg01ddpahjRu1xZ6mbeWWrTBGDU+EQUrs3TJAbQfm+poexApQV4YTxRhLNw3Rq/ituTVEOswr0be4q3o3hjPh5dACYtYeDqwDBeDPQNL7CneLu0EFv1/n146f57WPEzeuH1sncAXzFD944iDfmNH0Rpf34lBPjXUgG0cXJarRVvVKphi6JY1KYsq4Hi6ycOHdQMtKIhRATGhMlTEByipFKaQ0M16p60PS/CRX3N+M3r+CVf66i2FnC5YMDAFYNQiLHrtdcA+D/bhevWYMETy9HiaYR7RxHCLBY4YlyM63XmdDO8Xi9aGhv9qjPzf8clJeOZgk1MlVos7iriuvGJTDtKLL4oI+K3t7Vh89o85qWdn5Or/K1mOpG3J9xqZUaNjOhP7z1x6+jdLn23mOlS8WrcNK7wattNz5sMw+VrgGG8GOgXwt99BCXvK8YDuXFlRNm9uz7QpEWL4AcLvoIrDTYjxo1DWLhFYwCJLlxxtSgS7igMJRPF0kNZyS5MmDoNYRYLS5ecn5OLzWvzpJkTFqsVYRYL6qqrEGaxYFzKZD91T4vVCkA7sJOrW5N1FBNzGXfCgIEvhyEXe/0qRNPvkqJt7L280N4Ou8MJa2Qkjh86yGqCiQsYGdg7qareBjJe5ufkwqaGdo59dtBv3JAR8endJ8OEQkrkzSU+mwn+WVN81qNeVpTMo0THFM9laJhqtZiAD47AwFWCYbwY0EXvJWDgAAAm4JlbByDswSJ0tLezKquAb2Img4KXx++P1gr9TYNN9KjRCA4J0cSaKeNn59YtiEtKxlNvFMAW5WA1V4o35sPpciEq1qVR6Q2UNSAiNX0WbFEO2KIcGJmQCHNoKABt9gGBDCMywgo5yXAy6oi/I2ZlkIEz7sTnODySKyC1YoVRadrAlcHZs5p/f1m+H39RCbfFG/OlWUMWqxWjJyRjiN0OL4CuCxcAKKnEovdTXEzEJSXD6XKhp7sbg4KD4YWWg0bGD40JXijckVaPB4WcQJ2sbRF86InPIJR5PB0uFy729CAqJha5G/J1ZRvE6yH7LD1zHhYuW47GulqgZi7zvNzc91Bj4CuCYbwY0MXz/wM8d6vy90BcQkLqFHi9XpS7d/txWsidW1Hq1ihPysSi+IGIBgVS1bQ5nYiwDdXEmi2qAFzM6HjEjI6HPcqhqWNSvX8fGmtrMWZiiqZwWyBOjAjyllDcvdXjgae+lkmFj0uZDIvVylZdomFksVqxYNlyWCMjpd8TaFC9uDEf4AXrVq40jBcDVwbh4Zp/bVEOzHvwpxgUrPCuxEUGGecRdjt7z75zyxyYAI3XgiATkxyXMhkwmdDqaWTGvZ6svntHkVQ7SdY2FYLk30UgcPkAaodqF3VduIDI6GiNAUPoy1iibYjrZnM4gWMwQkZfAwzjxYAuVr0PPDcHzF1w4tBnsNjscO8o8gsf0eRvgs+9Kk7gvCYKDRr8YEHejOaTDZrq0Ucqy1mKZF3VUU3MmXA5XhYZqBaKSM6bflsG2pqbAQAjExIZf0YU5yISX1N9nV/RNkA7KDKDiifumow0BQNXALW1mn/TKvYBtigEh4Sgu6tLIcRz1aN5L2K41YoRY8bC7hyG//uf9xS5AFWriTd4xHePqjwHmc0sDCQj4/enIKvYNtkIgWwFWbv8+FSyo0iXxEv7koaUzFNE2jLdnZ2Irt6i7GhwXq46DOPFQN9QX8oThw9h+m0ZGiE5nsgWq5LnQi0W2B1O7Ny6BeFWK+bn5KJkR5GiltmkFGykVY/YBi9mRwNOfNJERNhsSJ+biefvuVsaP78cLwugv8IS4+sWqxWhFgtMJsAaFSnNWBBJfOIAyAbq2zMwYeo0nPjsoEIsNpsB0ogAtMaMAQNfBQQ+Vf3CBeh+uwjBISEINpuRmj7LT26f3sdV99zNeCI9XV1MBbu+pgpPvV4Am8MXuiXPKqU0DwoOQlN9HVtg8O8VLUzqa6rQfLKBacbIvB3iey1mCsrAGzz8e85LFVAfxHb49G49TxFfkf7g3fuAgTAMl68BhvFioN+YcmAFSgEmHLdz6xZG4svdkI+o6Bh0dXYietR1iLAPRebihxl3ZKSaTXCkfC8u9nRrDBTAN2jUVVehvrqKDWTn2tpgsdkAk0l3fAgUT9eDnhdIzFryAhiXMhmHy/div0Sdk8Ar/YoDIJ0faU7Yohw+LozX6/O6kPKuAQNfBU6d0v5fU4Pq/fuw4alcxtegZ1XkdskIrPNzchEVE4tb5y+ANcoBk8nEBOZ4T0plqRulfyzWvIcyj2bzSUXccc78BbBFOfxCQbISAv1ZpIhe0b68OyIsVitsTifOnWlDx7l2tHgacdNtGRgzcRKcrhFsUQIAQQPADJcPjuo2aeAK4IoaLzabDb/+9a9x++2349KlS9i2bRt+9rOf4fz587r77Nq1CzNnztR89tvf/hYPPvjgleyqAR309AJB6px6V0Ir8jltk9l33oXFM25kNU26OjsRHGLG8UMH0dnRgZ1bt2Bc6hSYoKho2qIc6Ghvx+srV2i8Hprwixo2IoheDR4iSZhfJelxbPh2aXvSsJDJn9Oq9EJ7e781ZfRIhCc+Owh7lAPtra1Mq6J65Urgued8jR04AEyYcNn3yYABPzgcmn/jMn+IDFWfiAxxnkjPexUI/PupeCBNONNyGp66WkTFKun9oihdX9wVMYTkdLlgi3L4LU74EgLk8RRrCwUKz8pC17L+iGGjSWkzYVIXFBabHQBgMpngdI1A6bvFyqJEXfjgoxvZvgZZ9+riihovmzdvxvDhw3HLLbcgKCgIv/vd75Cfn4/58+cH3C8/Px/PPPMM+7+jo+NKdtNAAEx7BfjkZwBUKYOsJUsRbDajt7cXO7cq8d6ykl1ImDwFO7duQczoeL/BKy4pGfNzclHu3s20EWSGgBg2IkPgea7uCe9eFknCYhpnoJg6n2ZJNVMqS90od+9mvB3AX6eFPyc+5i/TlKHBk7ajtqh2EvPWrFzp874kJn6Z22XAgAKB6xL6/k4paZYqwj/wwmoEm81McJLCQTzIkO/q6EBHeztsUQ546moZQT+QV0RmRNRVHUVd1VFk5eSiQh0beJDmjKe+loWAqGp1uNWK2Lh4REXHMO8pAOV77tz0iLw0ZvGlBsh4Y6FiKE6VocOGY6hzGP7n94X47fJliEtKZiFyg+vy9eGKGS8JCQn4/ve/jylTpmDv3r0AgJ/+9Kf405/+hH/7t3/DyZMndfft6OhAY2PjleqagcvA3jpo5GFJeTIyOhoxo+MB+IoqxoyOlw4UGYuyMXFGGpobGiRHgGZ7UrGVhWdEY0RcwYlpnGLse35OrkZciveQtOtkOwA+OXBe/l/si56mjGzFyWdX3LdyFU6ePoNfRVp9BzS4Lwa+LASuywMHD0tJswCwcNlyhISFobuzE576WtgcDpjg78XoaG+H1+tVfqvtduh4JGWhXFkoB4BuAdaSom3sfWaGy/BonGluZjy47q4uxqGrq65CZLS/cjeBfxcBIHJ4NNLnZiI1fZZG4oEPXT2+/hWMnZiCilI3QgeHs8VJ3kPZeG1xfH/uhIErhCtmvEydOhWtra3McAGA999/H5cuXcINN9yA7du36+47f/583H333Th16hT++Mc/YtWqVbig6gyICA4OhtlsZv9bLBbpdga+BLgS06tGvIv/99AHfnyVcKsVTpcLzxRswmaOtErfiyEaXo23ruooW1EB/c88EMWn6Du9kBSJZgHaOkO8ESNT4KT94pOS0XD8GKBejspSt6ZUAAA/FU/eKzM/J9dPH8cR40LC+QvIt1nQSZyXgQOBu+8GNm36QrfLwD841Ow4wuDubj+PB2/kN9bVwuZwoubgAbzBhXRFA130QpIBrpeazGqYmX01zFgxRa64KaD1yOi1R4ZJXXUVeycbjtcgwjYUUTGxaPU0+nFtePDv4uGyT9F8sgGhFgsmTk/TZDSK+wDwq2hdvX8fZnWqxzABn5zo570x8JXhihkvw4YNg8fj0XzW29uLlpYWDBs2THe/wsJCnDhxAg0NDUhOTsbPf/5zjBs3DvPmzZNu/+STT2LlypVfZdcNyKAaMGO8VajerzUwiFhLlWBj4uJZgTQaSMQQTXdXF0LCwpC1ZCkO7f1Uo47JF4fjB9lA4Sb6zbvGeQ2H4o35SJo6DdYoB8alTMbGj8rw1s9f9F/ZCQYUT7SNsA9FhH2ocjm8Xuwp3q5RCS7emO/ntoZ6PSKHR6P9TBvOnTkDh8vFVnBv/fxFPPDCahR+fgo/vI5bLf/ud4bxYuCLwW73/e0Fhl3/HY30PaB9f9/Q4aAFkh/oixTLh4B5T0jGomw/T4u4cNBLdeZ/j0pIhM3hQEtjI1obG2FzOPy8QDICPnlIKexTX12F0neL4XC5EBUTi/S5mdIxh/fOEky0qPMCN/y6vzfHwFeFyzZeXn75ZSxbtizgNgkJCV+4Q6+//jr7+8CBAzh58iQ++OADjB49GjU1NdL+rFu3jv1vsVhQX1//hY9vwB/v7AN+mIyAsd3ijYrK7XWJExAZrVR+lSlX0iBQX1OFHyxchDBLBEaOS0BFqZuFZXI35PuKw3EpjeJqSuSdkJta5M0Qjn92ECO9gN3pxOCICE37orw43++X7rsHWTm5GGyxsEtgglKF2gSwom786pAfrCOjo9Hd2YUhNkX0y2KzsX6nps9C0WuvYk7WAtiGDUVraIhygIEDgY4OICzsS9w5A/9wEPiB4X/4Q0B1a0B/USB6JPnQCn2elZOLMIsFFZwXUlywiMRgsQ98SEhvG5G7cvzQQZw4dJDxZGS8NLFsCHleKkrdcO8o0mQmrihQFgq8GjDfb/EaxSUlA6Z9Bt/la8RlGy9r165FQUFBwG1qampw6tQpOAS2+8CBA2G323FKTOELgI8//hgAEB8fLzVeuru70d3d3e/2DFw+fvSfwCVfpXmpJgOvcqsXc6btaKK/2NMDu92O8CFD0NJ4CvetXAVnrAv/9/5fkZI2E4Xr1uiSbsUBD/BlPfCDJoEvvOipr4UjxoXCdWuYYufh8r0a4qG4ahPJi4+vfwUzMu4AoAjTAfJsC55nkD43E2EWC863t7PU00lpM9Ha5IEtyoGNjS344Sjf+SAkpM97Y8AAw29+o31menvhfOEFnOM2ofAleSL0vJw8eGJ8ZambEdN540DmhexLR4nacLpccMTEoqXJ0yfxl/rDe2+oMvR9K1fBBIWHs3ltnlKv7MZp6LrQgQk3ToPD5cLYiSkofbcYJUXb2MKlrGQXwiwWtHo8rFis3phD5/TcvROBS/sMwu7XiMs2Xpqbm9EsxFRl+PDDD2Gz2ZCamoqysjIAwHe/+10MGDCAGST9waRJkwAgIMHXwFWECXjGvA5YtEQq1ka/A+ms8Mq5AwcNwqCgIKWOylAlJJN+xzz89ullSE2fpRGEIpDhEhUdg6aGer/YeWz8GJZJIFvx8X1Lm5sJwJ94yA/MXkAjjMVnJYxMSNQodYqDNj8Qi5MDjXmDBg3C4fK9uNDejtAqJy7cPNu3UVcXwHG6DBjQxeLFmn+3nTiJUs77KCtgSkVPH3hhNQBovBE86RbQhmXpfeBVtQl6oR8RxIu52N0DLxTPqCzlOTZ+jMbAEj0zYnqz1+tlOizWyEgWogagKWvCh5ijYmIBAOnqeFAhcHLikpKx/I0CWFU9mnmn7lZmTy/Q2dPP+2PgK8UV47wcOnQIf/7zn/H666/jgQceQFBQEF555RVs2bKFGSLR0dHYuXMnFixYgE8++QSjR49GVlYW/vSnP+H06dNITk7GL3/5S5SUlGD//v1XqqsG+oGDp4DEYQC8wJDzNQiP1KYFA3LXqkyTQeSokAfjTMtphIYNRrDZzOS7Af/BL2NRNiKHR6O1uQl11VWaY/N1R/jijnqrOVGxkx+waWA2QfHqUHom9et5oZwA9U1v1TZfdbGfb29H4do8FK7NY+medkcH8lauQDSAar7mUVCQQsBUayYZMCBFj3YGNXV0+NUfK96YD4fLBSdXwLRw3Ro89FIeQsLClNpcQ5XnjGoU8UrR1AagCFXmbshn8giAfvVnEfw7RkR+Xo0X0L5HCZOnaKrKi+8yhYMovblD9WzGxo/BhKnT0NVxHiFhgzUcN3G/Fk8jOtrbAciznzIWZcOm1lTzAjBzqrphT/V9ewx89biiOi/z58/HK6+8gp07dzKRukcffZR9HxQUhISEBISpVnF3dzduvvlmPPbYYxg8eDBqa2uxbds2vPDCC1eymwb6gQlrtaGjSTPSEBsX71fYDPANTqRnkjB5Cuqqq9jfeQ9lawY5UYROTLEU26ZshVCLBalpM3Hd+ES8ufpFpM/NxMiERASHmNHZ0cFIw4EgDoRi4TcKH1WWumEC/DgtsoGU/00QV4eAkvGU91A2C3/RNdt2oQsHQzlvC0/ANGBAxB/+4KfO/NgbBVJtI09tLcZyBUxLirYhbW4mUtJmoqmuFgc4FWn6HS5k2vzysUfw2p6P4Ih1IXPxw5psIv5YfYk6JkyeopFG4McRCvk4XC7s3LoFc7IWoMXT6LdgomOJJUPikpKRtWSpYnBACWvJwmIjxyfCHuVAuXs3q20mKz/Aj1PzPlsEGGuJrx3fuoidxWLB2bNnERERgXbVkjbw1YA3XoonbcKluJnspQf8FW8rS91Kpk10NCr3uDFSzRDg9+HRn4qulNlASpep6bPg9XrRVF/HXL9N9XVSo6o/WL1tB8alTsGJQwdx4vBhZoDxlbL7CovJQJ6XuKRkRNjsKHfv1ojvUbtkPN0ZE4VeXuult9fQfjEgB18fC0rxxfR9BzS6RATiuPCCkoC/Ui0P2Xspa0e2iBG1lehzPuwre1eJU2YymVBWsgvn1Cyh0neLdY0inr+Tmj4L02/PwJnmZnjqapm3kz9X8vqaTCa0eBrx0n33+BWLlV2PS2ugmTkHLJV2x8AXwOXM30ZtIwNfCDdXZuOvcUd4/To/xVuRI/KTpIkwmUwYNymV6cGIrlmx3pAIcVUYZrHAGauszhJSp/gNlJeDuKRkRMW44PV6MXzUaLjGjkPlHne/FET7ajdjUTZKdhQxyZySHUUa4jPjxewoQkxcPKbn/Awlb2/xragHDlQImUaZDAM8RP2rDz7A5AOHMEn1lJwTxBdJUJKEJjWFQnXeGdlzT16buKRkJKRO8RNxBBQvZoq6uODFGykTiTyOpP0kEt0nTJ0Ge5RDw6nRSwQAFLE9m8OJhcuW48X77mHb8zwdQFvOINxqxXXjE2GNjMT8nFy0t7XB4XJhXMpkTa0lTXjYZOi7XAswjBcD/cYHR4HvjgFgAsK8HdhTvF2THikjxcYlJQNQiHARNhsAYPCQIZgkpDBSFhLFwIkA++NHH0Pm4odR9NqrePtX6/0k9z21tRg7KYUJ3qWmz/rC55exKBvWyEh0XbiA4BAzmhsaNIN6Xysyve9FWfbSd4tZZgZfebp6/z42uaSmz4J55y789Xs3+w5AhEzDgDEAKIUXeTJ3by9w880oVrNvRqmTMuCvl1JWsgtZS5YiKiaWFQrV8ywGeu4pHAqAGSgUao1LSobX60Wrx+OXRi3TfhLfhZfuu0eqP8OD7xuJ7TXW1fbJvykr2cWyCd9c/SJS02ex8FibmpDChyToHZ53sdAn2mnou3ytMIwXA/3GzfnApV+o/5jgp25L8XAevDem3L1bQ6rzApp9q/fvQ+G6NchaspQRCjMXP4yQsDD8+NEl2LvrA19No9szMOHGaWiqr0Vbc7OyggtA8gX0q0+LWRW8+i8/WN+3chUSUqfA4XLhyXlz/drXI+vyE0Zq+izU11RhzvwFOFpZjjCLRcOxEQfamPHvot4V6zvI4sWG8WIAWLHCv/Bi6mRUA0yMzTo0UqpbxD+bC5YtR5NaXTpQsUK9zCGeC8J7WimriUK4sjZE7SeRW9MfTyffLi+2x0Nsh8YtMfOK+C5lJbuQNjeTlUgAFN7P0cpyZOJvSiNe4NK3inDxzYNhvBi4PHCrjo5fBOPfXS6/tEJCeuY8JE2dhiOV5UpmTfwYLFy2HI11tdisxp9Fclxq+ixERkczAbmi117Fjx9dggEmExuYLFYr2pqaYXM4YHM4UOHejeNQQkh11VW6ruWsnFykpM3EhBunaVak9DkVo6Mq2WlzM9kgtnltHhwxShXd0YkTGHFQLOjGXwt+VcinS7+25yPYHU5c7O5m2VKhFgtbmWoG7JEjFOExXr+js9PQgPlHh6AqHnKhU/O/6AWV6bmkps+CdWgkDnz4N+bR5PcV2xK9JwSxLhifQl2i6qbIJA8A/4xEMh5kx5GBxoNwqxWAflq2bD8xzZv36oZZLBiXMpnJJLBq98fAxr/n/tqvQxm4QhjwdXfAwDcLHxwFM2BC0I1xKZNxrq1NOshkLVkKm8MJR4wLGYuyseAJJSadkDoFGYuy2YqI37d4Yz6aGxpYsbW9uz7AE5m3Yc8fd7CBZeL0NJw4dBDl7t2ocO/GZjXdeezEFN2+AL76kp76Wg0nxyR8X7wxH6XvFsMEICVtJlLSZiJ3Qz7eK3wL3Z2dCA4JwfycXAC+lV/WkqWYOD1Nc3z6LmNRNlMpjUtKRuG6NfDU1aJw3RoUrs3DnuLtuNDejum3ZyB3Qz5b7RHibrgR8HLLvOBgJWRg4B8TPT2+KuQABvX2orCxBRmLstlnonz/gieWs3RjAj3nZCwA8Hsf+bYo1MkfJysnFzMy7kCW+j7Q9qvuuRub1+Yha8lSTL89gyluy8JO1CYZD1lLlmLG7XOl74II8jJNmp7G+hWXlIxnCjZhRcEm3f2pj8/fc7c0xOuMVRYqgy0Wtihp/k/VMPIqQ+Cq9wN2zcAVhuF5MXBZuDlfZdurBsyl91aheOOH0pBM4bo1WLhsOcyhIUjLuAMnDh+CyaQUghMFoCgrwb2jCPXVVQgJGwxHTCzm5+SycIq4guO5KKLXQwaxsByt+MTPeXe2F8Co8YoQ3bjUKTh35gxsZjNGJiQiPXOeVG6c+sQXoOM1MwCgrroKdVVHfQN2Ti7OtZ1BVEws7lu5Cp7aWnYtw61WHG49i7ftQ3wn43AABw4AEyZ8JffVwDcEp05p06J7e/HU5t+jHHIyK4VYz7WdQYunEYXr1mi8MLLQrx5knhnR8BePLSu5IZb14Nukfbo6OxE5PBr3rVwFu8MZUAFY9AxZrFYND+dyuDy89svxzw4CACaq2YbPxH3KiDC92gQvA18DDOPFwBeHF/h+62v40X5fCjORUgGw1dqMjDsAkwmnG0/hse/f7NcMT/ojEbi25mZ44SPN8bFt3pDhy9mLwlIiAhWT06vtsooTorNYrbBGKYRea2Qk49iUvluMuqqjSJubyaTXRQlzXjMjhRtYSQxs0vQ0nDmtEAWdsS6MnZSiVOQNMaNyjxsxmzZjwA8zcCmW478kJioaHz/6UX/uloFvOs6eBcLDff97vUBQEDarz2ds/Bg/wUReCK6i1I3U9FlImjoNNoeTib71J5MHkGcdkeHPV1cnI15G4Kd+6b17vAje7DvvwvCR12HI0KGsr4HqDtE73dbcjM8PH4LFZkdZya7L4vJU798HL4BxKZPR2uTBW6tfZAubtQt8573qfwJeKgNXAYbxYuCywQo1AggNVlYw/OonbW4mRowbh40fl+G9zW/hcPleOGNdcO8okrbHk/7IgyGK1PG/xeydilI3KkrdcLhcWFGwqV+p0hTztlj9lYJ5iDF5GqjFPlJaKFRp8gvnz8ELIMhs1nBZ4pKSmUSEeG58u2QYaTKeHnsEqKoCRo/2dfCHPzQ8MP8IEA0XAFi4EIB/NhtBrPVFWT2fHz2M0HALdm7dAkDL9bhcDSPRcBAXL3pV4Ond+/GjjzFyPK+eS7yzI5Xl6LrgE5wURSTFdqko68WeblgjI6VjCKC8106XC22nm5k3l971MIsFJpMJdocTt969ACYAG1/8EXDYd12MkNHXD8N4MXDZYIUaVV/xr++Jxf/L+RMbqNLmZmJUQiIA4P8t+AnMoaEINpuZqqcIUSGTtpGVpgf8J3saeEWPhgy8MWICmMdHHAgJokQ6De5iH8MsFgBAT083ykp24YEXVmPgwIH4zi1zYAI02jXisWT1j3SracfH+xswiYnA3XcDmzZJz8HANxynTvkZLqbGRoyu3IcM1dsBaN8Hsb4Xn9VjsVoxYsw4JusPwI+0Duhn5xHSM+dhwbLlONtyGp0dHagsdaNECJ/KUL1/HxOd+6cbpiI4JAQ/eXql5hh6Xhuv8Ftslww1sc+itzbcasXYlMkAwMYlPjOypbERNocD1yVOgDkkBDdty/wWSrp+s/Gtux2Gwu7VAVPbNQGXMBCz//cONvE+U7AJKemz0N3Zic6ODkTY7ejt7cWvlz6uG7fuL/Q0J4g3w6diy4S3RIXeFLWitKe21i91mne/k3CVTB1YIQhuRoTdDq/Xi3L3bgx1DsOwUddh9ztbcf3s7wVUFu7r3GTbHfvk/3ApONj3odcLfOc7wN69l3E1DVzzOHXKLyUaNTVAfLzmWRaN9UDfyZ6zZwo2serqHe3tbGKYND0N3V1dCDabNW3FJSXjhS3bWMFDeu5FsTk9EO/GNGAAoqJjcP7MGYSGh/fZXwDSv/s6Hl2P7q4uFoqFeo7k7eUlEuqqjjIV4Pa2VvzrrmQ2Ux48pZRLMfDVw1DYNXDFwQTrAJjQy7IYSDIcUN71w2WfsvonaXMz/VZwehLiPEj0ikiBE6enAfCvpfK8muZMhR7bJR4Y0YV8Tl2J8rFvWUz+mYJN6rn6I2NRNhPga/V4YAIwYuw4dHV2oqXxFEsMoX37ErMTz03cPmNRNqbXeTBv5HAfedNkAv7v/5RQguGB+XbgwAFdwwUIrDqr953s2SNieoV7N7wA82BWuHej7XQz7A4nWj2NmhTmjEXZCA4xo7uzEw3Ha9DS2IjBFgvz3hSqvC89w4LEGNuam+H1enHx4kVGbpc+7xLv54qCTX7eIj3wIaXmhgaNh4jabz7ZgMjoaCxcthzHPjvIRC9/1vIzwOJryzBcrg0YxouBLwQ+68hkApzb70LWki1wxLow+867cGjvp0yS++m75mli7oBvcmbhHnWGlw1CWTm5rIZRuXs3KkvdCOe4KmL2Al85VxwI+eyewRYLcztWcplK/G9e2r9d0KQh8Jwdqp8y/vobEBIWhnkP/hTBZjNamzxM20Yvbq834YgufZbRtDEf1WV7tQbMm28Cn31meGC+6RBDgwBQX88MFx48UVcksYqQGcg8uZyXD9i8Ng9ZObmwRTkQZA7BQy/lYZBaX4vPyrmg1g2ary4wrhufiKycXEySLDIIfOg3a8lSREZHw6v2heqJ0b482Z14NXkPZQfMdBLBh5R4g4XvS31NFTIXPwxrlAMpUQ4Wfl7L1XTrutiPgxm4KjCMFwNfGJ09QEgQAC9wu/M4vquq4xK5Ti/mzk/OxRvzkTR1GqxqDROCyE2B14v21haNO5sGF94Y2Lw2D3aHE0NUsh5pUwBgfcjdkI+omFhW4dnr9WJP8XYAYBkTBBkXQISMx/Lbp5cha8lSXLzYg+hRo+FRJcsB/7i9KGZHmjDkxr5w7hwA3yCtmZiCgrQidiYT8MknwMGDBon3mwqZ4eLxAC6X5iMZUbcv4q3MMBflB/jnvHBtHmLj4uFQdU86OzpYBg9l5QCKl3Pz2jylEOvwaJgAjTEUCHXVVaivroIJCsm4UqiNxo8foxISERUTiyw1q48fT/oKu4oLF1qw8CTh4BAz2po8OPbZQWnf79/a5+kYuEowjBcDXxgzNgCf/Ez52zQA+F7UCRza+ykjm/JKuYGKvr3I1TAh8ANz4bo1jGhI1ap5ca1QlSwbarGwYm+kLcETF6ldJRuhB0HBwbjY04Oq/ZWa1diEqdNgi3IwTRYTFK2XQFlJIkqKtqGu6iieer0AXq8X57n4beHaPL8Qkixs1d3VhZCwMLQ0NmJP8XbdieC1vfvw4I2TcYnX/0hMBD7+GLjhhn7118A1Apnhcvo0Ht/yBxQLz58ecb2v0CNvxEdG+zKDyGgWPTh5D2XjvpWr4Ix1sTpAZGSUu3czlVrRu0FeUZlarqzeV/HGfD+1XpHvwodgRe+SXniJvwZPvVEAu8MJqF5cvq8ykvCpZ7W3YlNZv++kgSsMw3gx8IWxt079Qx1Qcsb9Hf/NuWJpYHokby2GjRiFU58fx4gx4wD0LeNN+0dFxyBryVLkPaSoZ/IrLTJUKkvd2FO8HRarFZGTUtDc0MAye7JychEVE4u0uZkaTQuHy4WE1CkYOGgQPLW1msHL6XLBFuVgYSBaTfIu5/6QBDMWZcMaFYmm+jq4hSrS7Wq2BV8eQfzNEwgDEZ13bt2Czfah+Jf4EYonhnD99Qrhc9iwgP00cI3g3DlAJcAynD6Nxzf/3s97KCs7AeiHHsVQpUxATs/LWL1/n6aWV13VUXYMGcGdh0yfqXhjPtOf2bl1C2JGx2s8ILzBw+9vsVphi3JoQrAEXhSSr5nGHzNjUTZsUQqHqKXJI62tJr7fDi7J65PP9W6cga8DhvFi4Evh4CkgcRgAExB06QJbQfFFFmNGxyMoOBjRo0az7/kBjwaopKnTFKKcStzNeyibGSgZi7I1uhGPr38FkdHR6O7sQsmOIpQUbUN65jzExMVrCiqKcXEaoObn5OJQ2ae40N4udU/zA1jeQ9mYn5OLcKsV83NypYRhwN9t7RfX5/QvxFWeSD6mfsSMjmeThZ5bPGZ0PILNZjz+m9fxy8WLtJWGHQ6gttYv5GDgGkNPj1Y5F0BIby8e3Px7P4NEz7sCyLkucUnJGDlekS4YrUrnl6iaS/yz1F8OiZ7HI9xqZSKT1LfijT5NF/7dCbdaERUTi3GpU/zIvTI+CoV7vFAUusWMI6fLhbFqNqB7RxFGjU+E0+XSHJP6Qly3MIsFRyvLNVpP/LFv2qu9jkYF6WsLhvFi4EthwlqOuAvlhf/lfuU7yij4/OhhDBsxCkWvvYq3f7UegFbhlrwsjphYWKMciI2LZ54T3nDgwzZ89gDpNNDx6H+qa1Lu3s2ItABYfSQx9ZMfQEVjgjwlJIhnkRCGKaxF/eONM8pyEA2lQBAnqX4RfR97xF/QLCYG6O5WaiIZuLbwhz8oQoMCQi50YuVf/gfnJQUHyWvBc7MIMgM3Y1E2rJGKMnSEzY5JaTOlmXglO4oQExfPDJv+tA34nj+L1Sp9zs9x7w6R7YePHKUIwTmdfs+5nmFfqNYwc7hcmJFxBwsPTb8tA22qOrUJyrhji3LAFuXA4fK9aD7ZwLh3q4SMxB41DRyApqp7WckurL0JjJjGlxYzcG3AKMxo4EuDr/Pxw4mKhgMRTptPNmB7/m/xL/80Bnt3fYDH17+iqQnEezfK3btxtqWFEfIAeeE1+jzvoWyU/tHHfSkr2aUh+pGRIhZrLCvZxQY0Ag2g/DF4FG9UitjRADqR6899K1chbW4mkqZNx5HKco03iQyvyj1uFK5bg4xFSpFGvlCjHoo35msyq/QEumiAZ+cYEQGcOaPdaNAg4NIl4H1DGvSawblzUsPlho/+D5tPNsPmcPoVHHx8/Sv4/t0LWChUxHy1UCIVDgXU572hAeXu3ejp6sKJw59JQzy88S8D/47wxQ8BxejYvDYPpX8sZgsPAv/u0Ls8dNhwAECEfajf+8g/z/x7Qp93qPwxL9f2W6tfxJ7i7SjZUQSL1Yoj5XtR7t6NC+3tfudUvDEfFe7daPV4EGQ2AyYTq+oeG6/oP/z0e0rVeXJD5X8kvSQGvkYYnhcDXxr3/h5461/AXvRX77iIwzepugmcJ4QPD1mjHKhw79YQA5+/5268vG0HhgwdCrvTyeLe/GqI541opMlvz9Co2NL2shWq6KEBAmtmUP8olMMXXASA6FEKwXJQUBBGJ05g7YRbrRg1PhHWoZEofbdYk/kEyOPy/KBPdVZId4MMp/5kcMBmQ/DJk+h2OrWff/e7QHMzEBnZdxsGrhxOnfLnt/T2AjfeiObuHlTm5MLudGrk6+n94b0MImQGLj3vEfahCA4JQVi4xY+rIhJWCXpFFDMWZWuKH5KxoVcjTFTH5vlcsveRji2SigHfe0BZT9T/kqJteHz9KxqvKp/hyJ/L82rNMl4/ih+bfvDO4+xCegE8+I7sJhr4OmF4Xgx8aWwqAy5xVRQTPTtQUepG78WL6OnpwYXz5zSemLMtpwH4exDikpLhiHHB6/Uiwj6UrfJo8KPJX/SOFG/MR3NDA+PGEPRWkuJKr7/KtoBPE6Od8+acPF7Dvg82m1mfz7W1wRoZieaTDWzwJ89QfU0Vuru6cOH8OeRuyMf02+VeH56HwBMKA3lsCA//fht+Xt+oTIo87HYljGTg6uM3v1E8YKL4XGenQrbeu5cZrSMTEmGPcrDnl/cyHCnfi7ikiXh52w7Ns1C4Ng9Hyvdi1Hil6jlvbBe99io8dbVMyiBjkVJxOndDvsazIcsK4t9DQOGsHFa9G3yqcl/eRGpj764PWGYi/17wkJGK+XdVNh5QW7TQAXxk3CzVK/XU6wXsfFfdczc2r81DqMWCtuZmnFHHpiD0sAEq/8N+3VkDVxmG58XAV4Ln/go8NwfshZ94YQ8GjlIG3fQ75iHYbGYKlm1NzahQiXXkSQGU7CGq2vze5rdYFgKgDFp8qElGqhUzHvS8KWlzMzUZSCJBkMoMXJCUGJC1+frKFZp9+G1IvA7QrkBzN+QjJCzMd210BmhRy0JG1gzERcgAkF6yCyVvb9ESQimMZOjBXD3o8FsG1NXh0ogRms9MUPSH2po8Uk/byIREhISFIcJuZ2R2QHnGbA4n7A4nHnhhNQ5+8jEToHv7V+sZ5wzQZgTybfCQPe+8qF1fVZpF/PjRx5C5+GG/zEO9YztdLgwfNRr3r1yF11eu0CXzEjTeWE7QDlDkDohnk5WTyzKqMhZlM72aE4cO4v/9/UHNysrwulybMIwXA18JVr2vGi8AYALuOJ+PP3pXoKe7G6c+P47WxkaU7ChiWTcH2trYyincasWohETYHQ50qbVHYkbH+w2MpPEiU+GUpVmKbmy+aqzaTQD+Fasnpc2EyWSCV60QrZeeSpCJ1NHnfNhHI/6lZmKJqaJxSclY/kYBrKrOzPP33K05nmzAFkX6SHiM1/549MRJ3BUdiR4SsyMkJgJdXdoMJQNfLZqbFW+XBDefOYekd4pRptb6obR4Is/ymXO8NgpJ89ccPOBn3BSuW4MHXliN4BAzRo1PxJHKcj/CO+CvOiuD7L3iM4h40roYThURl5SMf350CQYMHIgRYxNQsackYAiUDLEIu50ZaYGy9HjwhH7yzFiHRqK7sxPBZrMm5MYvMgrX5mHlgir23ZlO3e4Z+JphGC8GvjJ8cBT47ljl7wHwwlyUjbKhWZg0PQ2fHz7MhNuycnJhsVpZVoPFaoXN4QBMJpw6fgwRdjscLpdmsCX+SpjFohGqEwe/+1auQkLqFDhcLo02BeAb/CtUXRhZ5g8NZFTgkQbo/oh/yUJOJvVnwg1TkZ45T1ONWozvP77+FThcLtgcCk/FC3+RrvtWroIjxqXhQZAxZnc68cKWbQg2m/1UVwFgBPWxq0urBxMUpHhhzpwB1BpNBr4iSFKgAQAdHYibOg3TcnLhtVqxYNly2B1OJupIkvk8D4TXRklInaJbD4zeM5IZuNjdjbETUxhfqj8hUrGiNC0WSM7ABKXGGBn3vDdGr92MRdkwDVCYCgMGDPAj0ovHz8rJRce5dpxtacHJ4zV9Gix8v4s35ksNMzofNYl1VQAAcmxJREFUkT9Hi4/fcI4xLwDbCt1DGfiaYRgvBr4ysHpHqob/nEt/wk/X1jFyHQ0WlDoZo6ZEQ9mcSf+PGJcAa5QDuRvyGQGX+Ct11VUaLRkR0aOu0/zmQStGXhFURPX+fcxD1OJpxMQZaWwFTG3wEI0a0ZjZvDaP1TnKWrJUV2yO2um9qBRP6e7qgntHkUY1NIaTaV/wxHIc+OhvGmNswtRpCAkLQ29vL/PoiKRGAHj8N6/j1YcXo1ucVIcMUYyYDz4Abr5Z2k8D/YROiAheL7ByJbBqFaoBlup7tLIcANDqaURWTi7jevDvDb0DMaPj2X6J19+A3z69TPpctXoaMSg4WOPd01Nzpv8J5M1LmjoNQ4ZGorLUjeaTDUzOoNy9W7N4EH/LjHpeZ6Wjvd3PgOCRsSibeSxpoUHkWlm1eDomGWx0Pryxxgv6UVhJVOO9/0Zfe3VtOvfWwDUBw3gx8JXik8+B69XwvcnkH4MG/F26v3zsESZUVV9ThVHjEzFo0CDN96K7WA8Nx48hwj4UDceP+X3HNCdu9xlOsrayliyFI9aFQUHBjAicNjcT59ra/LYVB21xMqjev4/VOSIDSAa6JmScBAUHa/YJt1oROTwaPd3dCAoOhqe+1u+apGfOU8IFZjPGpU7BubY2pM3N9CuQV1ayC79Wyy6UFm5Crxgy+u53Fa2YiAjd/hrQgZ7RAjC13OLtO1CtfiSGLMXaPrzqLWkV0T7fuWUOM4oBxaD11NfiL5veQtaSpYiKiQUAjeHC1zESf/MGB1E+zrScRk93Nw6VfYpQiwUhYWFoOH7Mz+OjJ1wHQMPH4UOnzxRskuoWUX/4hQZvzMg0auiYos4MkZLFTET6noo9AsCi2kcwgIsn/fBN/1to4NqBYbwY+Epxw69V74uKP/wr8KP/hN9ESy5dWn2RwFvzyQZYh0aiotStSQvmB8dAoZo3VPIshVzEGin1NVXo7uxihpHMjV64bo1mIkhNn6VUxJYMguKgLaZnp3Nchrqqo7orTV6QL8xiQVSsi4UNyKNDHqz0uZlscuGPXVd1FJ998jHzYk2/LQNHKsv9NG1S02chKjoGD7ywGs4nnsT/5ixBvStWeyPDwxUvjJq+a1Sp7gMHDij8IRl6exFz97/iJ9+/DSkZd2DCjdPw0v33aNL9Af93BPDPNuPvN28UZy1ZCrvTqfw4nIgcHo3uri4cUzkxeh4Wvk3eG+EF8PnhzxA9ajSCgoMx+867mDF0QVWiDbSQ0CPLA76Q0NiUVJhUjRURoqFTvFGpFh89arRfSJnaJD0l3jMjIyWLoVgaZ9Yu0PaBlT8xcE3CMF4MfOWoOQ2MHqr8/cNkYHIssFeiWlu8MZ+5eSu4YotENAXgJwPOD8S8y5f/nq8bRMekfZpPNmiye2SDOqn1Tr8tA57aWp9Uvyp9nrshn+lT8AN4XFIyIyQTV4G8OFlLluLQ3k81BhAAP1c4rUDFAZafZKhv54QVKK8cTNch3GrF2Ikpfpo2tGqffeddeHuk6iprbVVCRzwGDlSqVF+6pBR5NIwYLX7zG+D++4EBctUJ+4VOvH6yGaVTpyv8J5MJNodDmt0j43OI2WY8RN6U6HmJjI72q9sViCDLeyMmTU9Dd1cXgkNC0NnRgcJ1azRGc18ZbzKyfJZqmDtiXLA5HKyq+wWuaKmsLbo2ntpaJKRMRoR9Cubn5Gq8NXqcGxkpWXznf/nYI3hfUCl4p3/1Vw18jTCMFwNfOeJXc94XE7DzAcD6tP92vJuXd0PTgMyHmsKtVqSkzcTIceNgsQ9F7dHDGKkKwCVMnoK66ipMmp7GXM0VQtYD/c0bR4EGdfFzGgRJNEusVUTnExUdg67OTubpoFUxeV74VSCgZCKZTCaEWSxorK3VHfwD9U32OZ8lIgrb6YaybDbEJSWjdu+n6B4kDA0DBihGDKD8/kevVi0z9DgMbTuDh4v/G4fKPkXznXehvqYK41Kn4FDZp+hob0d9TRU2flQGT30t3li5ws9D2J/ngN+WDBn6jDeu+6tjJD4zJCS3c+sWxfjdUcTIriKPra80aQr7UBZfq8cDT30tOtrbUbg2T1rrTGyreGM+q/guakSJ4S/yvtK4EihjLy4pGd8du4+lR1/oUbzFBq59eL9NPxaLxev1er0Wi+Vr78s/8k/zc/Be+oXvZ3Ks8nlcUrL38fWveOOSkr3pmfO8r+35yJueOc9v/7ikZO+Kgk3eZwo2eeOSkr0vb9vhLTpW73378DFv0bF673/9/ah3W/Xn3j9Ufe7dVv2595mCTd7H17/ifaZgk3db1efeFer/cUnJ/eov3y/+b73t0jPnabah/v7Hx2XebVWfex9f/0q/jrPxozLvOzV13o197NffH72+9/f8Hl//indb1efe339W7UVPjxeXLun/nDr1tT9nV/Vn8uS+r8mFC5rrS9fztT0fae7va3s+8hYdq/cWHavXbPvyth3ed2rqvC9v2yF9zmT3in9mxM/ikpK9r+35yPtOda33v/5+1O9dE58F2bPx+PpXvNuqP1feOa7tZwo2ed+pqWPvaKD3jZ71opo678aPyvy2o34/U7DJu6Jgk3f1th2s3cvpLzvOsXrvOzV1musg9o8+a/7FUN9YtQbeFTdfA8/aP+jP5czfhufFwBVB5LPazKM//gSIfl7rsgUglQUHtG5gAEx5t6GmCqHhFuzcukWJw0fHoKmhnoVdSN1z1PhEWFUJfFlqs+iBoX45XS6MTEhkxdoCufb5PlN/j1SWo6e7268kAc994dt86f57/PrzZaC3ahWvuyzTAtB6qB6n6/Ph3/yl7AFFJfbSJcDrBfLzgQcf/FJ9v2YRiISrYvDFXmRveE3J4OG4IwBwtLIcXgD11VXMM9DiaWTkV97bcEaV/nfGujB2Ygrz1IlcK0C5RxOmToOT44BQqn84l+IfGR2NS14vQsLC8MALq1FXdVSqHUPhFZ73YoJSsJEI9nwGFKXoe9F3odHq/fvYsy7zAIkhq+aTDRgzMcWPnNsXMXi+WlYBAFo5gT/Ze0FkXnuR2p4X6OlVNKsMXPswygMYuGLwnIMysgEYpnLyijf65LtFUSteXpzfLndDPqyRkWht8uB0YyPyHsrG279aj7yHslGxpwR11T5Rqer9SiFH69BIzUBLRg1JhN+7/FmNVHjxRqUI4nWJExASFoburi420dD+geTPqeRAaFiYtGgecV8WLFuuaYMG47qqowCA2PgxAY/XlwQ7XTdZSImKPFIBSy/gJ69O/Skp2sYm1bip05SwkVjskWAyAYsXK4bMpUsKD+SbjrNnfecTyHD54APETZyE7A2vacIRpe8WI8xiwaS0mYhSDREqKZGxKBtjJ6agYo8bb6iqsfU1VThzWpGnr3DvxpurX1SKGa5bw5SpqSwE3X+qnjw2ZTK7h/T8U0HH4o35KP1jMd7+1Tp0dnTAHBLiJ6fPt1+8MR9HKsuReMNUpKTNxKS0mUhNn6X0o6GBhaOm35aB8+3t2FO8XVOxPRBk5QfE79w7itB8sgE7t25h7z8VgJQ98+Lz7gXg9XrR4mnES/fdw45F5Tjqa6o0+970l1s07S16u1+nYuAawBXzvDz11FP4wQ9+gEmTJqG7uxu2fopfPffcc7j//vthtVrxv//7v3jwwQdRVVXV944GrjkMe06beXQgB5iw1pc6TSssQF6EjbajFV+9ymvhxba8APvsl489gvTMeZgwdRqOVpb7yYn/8rFHfNkbJpOfVHh7WxsGBQehs6MDv316Gar379PwbgB/oiKBNDjO6BTN27l1CzIXP4z2ltMBPSO8uJzseH1xC2QrYFqBy65VbFy8bo0nv2OpnJi4Nb/AX2+ZrRgtMixerPwASuXkb0rKtSjep4feXmw7cdKXEQd5Bg9VXG6qq8Xxzw4ywUNZenTzyQZYoxywRjmYoCN59uqqjkrJ6l4Ah8v3ooMrSUHt0m++HtZvn16GW+9egAk3TmOCiSKhtXr/PtgdTphDQtDd2Ym/f/wh25+8pGL7oqp1f6DHw+G1bNg4kTYTAPwI6vy1JvDFS/l2Z995F0LCwjAnawHGpU5hJHlHx2GfwBSUOm0Gvhm4YsZLcHAwtm7dig8//BCLFi3q1z65ubl49NFHsXDhQhw7dgyrVq3Ce++9h8TERHR1dV2prhq4gnhnn5JxBACJw4Hdy4Zjx7DlLKWYl+8Wi7ABvtRjIry2c4aLqIkRl5SMB15YjZCwMJgAFo7hpcwpe6OsZBcWLlsOW5RDWiYAgKaYZFnJLuYdkYV2ZIRgHjGj4xFsNqOlsREnDh/2E+gKtD+/rR5ZN12Ql+chu1aAf3Vt0ViRHStjUTamxyfgn36teFfyH1qM84MkCrIESrkmXEtk37Nnlf71E9fXHMf3S/+Gkh1FKA2QPk/gJ1K6tmLmEP1dX1OFW+cvQGNdrd+9FdOpxWrlfU3o/H21O5R0al4w0c8A4EjmsqrrYskAkjkA9Gsa8ZAtVMTrwf8mQy1QSJV//os35vsRdgvXrcG9Tz8Li83G9GJymxf5GBQADp7qs+sGrjFcUQLOwoULva2trf3atqGhwZuTk8P+j4iI8F64cMH7z//8z1eE8GP8XJ2fS2u05F0i1PKEXD3Cn4yUGJeUzIh9esTCl7ftYETJFSqJl9onQuDL23b4tSEe97U9H3m3VSu/9ciDgUiV4ra0jd55ydrhiZF6bRIB9LU9H/kdW0aMll3DvkiX4jYiMTWkLzJroJ/f/ObKPYO/+Y0Xvb1fqF9RJ0+y5+Cdmjo/Emig50PvuvH3nv5/hntG+zofes77Oqbs2IFI8v39ofbovdr4cZnue6Tb/6r+9z/Qc0w//PP/+PpX2L3iifu0zduHarz/8XGZMi6pP3/416s3Hho/+j+XM39fM5yX6667DsOHD8f77/vYUmfPnsXHH3+MqVOn6u4XHBwMi8Wi+TFwbeGdfVAeNxXfKZ4BE5QQRkxcPAD9eLiMw0F6JmJtFIrvP33XPLyxcgWL5ZsA1galbI5LnYJxKZN166vQcSnWT+nNPOeA+DMLnljuxx0RuSni+YnnRavR6bdlIHdDvia+7xV+89eBjlu4bg08dbVSFV/ew8IfS7yGfB/1uDXV+/cxHlJs/Bi2ffHGfDz46muImzgJj/9qA4b0XPTrR0DwnJmv+mfxYv0Ql4gzZ/D4rzZgW00dHv/VBkR8bw57Dg6X70Vbkwf1NVV4fP0rAIC8h7LZ8yHeN/G68deKfx6n35YBLyDlKslQVrKL8U/6E6bhj11StA2LZ9yoW6aiL/DPqQlQwl2RkQHrFPH7PlOwCQ6XC5Wlbl2Fa357DV+N00wSt2v1NOJsSwtaPI0oK9mFCvdulLt3M7FG/h2p+ft+ZOz7V01s10iN/ubhmsk2GjZsGACgsbFR83ljYyP7ToYnn3wSK1euvJJdM/Al8aP/BHrzfPNHzKUT2Lw2DzFx8YiKjsFTrxfgxKGDTEdCFKcShbBEiXMCH98v3pjvF8unffj6KmIbfCyejivjHAC+sc9TX4vjhw5qquwGqnkEwK9fWTm5iIqJRXdXl6YsAgBGiDRBqxpMLnWny4W0uZm6k4Go+svr6+jVlwnErREnkYXLlqOxTtErAZSMkZ/919saXRDz4sWBeTJfE0LPnUP0TdOZwSY+X6I44NiJKZh9510ajR9e/0cmPieiL1XdvpCaPguR0dFY8MRypM/N1K310xf6eiZl2/HPTsmOIqTNzWRZVH0hKycXKemz4PV60dbkQVZOLgvryDgwvLR/4bo1aPE0YlBQsJ+BPj8nF2NTJqO1yYOxE1Pgqa3ViD3yGXUlRdtQ8YQZ9iiVhmACapov+9IZuAZwWcbLyy+/jGXLlgXcJiEhAYcPH/5SnbocvPzyy1i3bh3732KxoL6+/qod30D/kP8RsHgqGDnuwMKDWK8ORjaHAzaHgynYAvpCV3x9JNlgK+NtiAO0WJdFHJz5NNdQiwUX2tvZBCHG5EUODqV2itvxA3F9dZUfR4Cm9GMHD8BTW8uyLLzwcSf49gEusyRtJgA5oRHQel7qqo5qZNQDCYIB8DNuKM13UJBS8I84RjaHk3meJk1PQ0Wpmwn5EQn5n379G2bIMbE/px1dsjTsK4FPPkHcfff73etS1eCgYoSyWjviNeF5ScSpyFqyFGUluwKKwl2OwSCqL/PGVcLkKYiKiVWqsQN+Ke/9gSx9HvBXzaX0Zf4a0PUTVW0DnbsJALxe9HR1waaSk+mZFVO26V2jZ4eepdJ3i1FStE3DcSGPZFNdLQ58+DeNIRUbP4YZ7tSf5MguxY2pjkXxq/t9yQxcQ7gs42Xt2rUoKCgIuE1NTc0X6sipUwpbyul0sr/p/4qKCt39uru70d3d/YWOaeDq4cF3gBnXAYnDAJgAMy4iw1yC3x9KxfFDB5mehEhW5QdDWZYGAJY5k7VkKUuxlOk7AJASLPmBky+EmJo2U/EUeJXhkSYIfpKXraABfxIkPxC3ehr96g3xROLU9FlKQUXOKNEj6hZvzIfT5UJUrMtPW4bfhr9u/IQjtit6H269ewHGpUxmEzql+Za+W6yQkEPM6O7sRMPxGlisVpTsKAIAOFwuOGJd6O7s1FQ1pmtDdZxyAYxKSGTaHBT6oucgNn4M8+zwsvelfyz2m2T7M3lXc/dM9EjRJDhyfKJf7RzxnophFzIQ0+dmKka4hIwKQGMgURkLUoUmQ5U8d2IhQv5ZLly3Bvcuf5bVBrpcwuz9K1dh+KjROFpZ7kcgFo0Wnuit5zUiBPLY8c84lRqg44VbrTh3pg1RMbEs+4/PhKJ9SL+G9/7lPZQtzTACfF5C0rfZ8T3ue69SSNbANxOXZbw0NzejufnK+NiOHTuGkydPYvbs2aisrASgeFFuuOEG/ObboBthABPWalOnEw79EpvfTNIMOCRxTpk+mYsfRnCITzBOb+CkQWr2nXdh8QxfXXt+ghLl+fm2KPMpfW4m2tva4N5RhAk3TEVwSAi6u7rghX6aNKAv0sVPrDQQh1utGCPUGyJ8/+4FGJsyGYfL96LCvRuhFgvzAukdt7G2FmMmpSBryVKNAJnYN5okeG0d+i49cx5yN+SjxdOoEUdrU993mthJGG3EuHHo7Ojw8+DQREtpwkFmM0t75a/J/JxcjExQhASPVJYDAJPKFw1OSiH21Nb61akB9CfMQBlYgH+2VeHaPMSqhgfvneEnWwrh8cZSeuY8TJqRjnNn2hBqsfhlzfHPAM9fou8tVquf94wPbzpcLqwo2ITDZZ8yo/f7dy/AkKFDAa8X1fvb+82Xob6PS50CAOjsOK8Rd4yNH8O8HJWlblSUulllZwAaL5wspMtn54ngn0PeU0SyCW0SmQExbEc1vfiMqEACeYXr1uChl/JgDg3Fvz36IyS2+d6NS16lkKyBbyauGOfF5XLBbrdjxIgRGDhwICZOnAgAqKqqwvnz5wEAn332GZ588kls374dALB+/Xo8/fTTOHr0KEuVbmhoYN8b+Objg6PAd8cAMAEmrxc/n7QPT0CreutwuTAuZTIu9vQgKDgY3WqtIL2BE9Cmd/LeA7FQomzy41d44VYrmwg3PJWrqUvEe0D6u9oXJ1a9ekO0HQ3gHe3tWHXP3RqdGb3QAO/Vyd2Qr8t9Eb0u/DmQ8RcUHIyKUjfCLBbUVVfBrfIaiG9Dwmh2hxNerxd7ircDgJ9RxPN0xEk1Y1E2m6yb6usYV4bCTqJharFa/Yw4XltEzyu14InlsDudePClPKlRx3OGVhRsQuHaPOYRcrpceOqNAlgjI1mIBlCMC0BryGYtWYoIux0A0HvxIipK3ZrQpMZjooYAy0p2+T2HvEFTvX8fu/8zMu4AAMWwUp/l4aNGAwB6enqknJdAz2dZyS5MnJEGE0zM06jRGeI4LfT+tEvOmwd5lFqbPLAOjZQa5tQvChmSB5TuQ6jFghOfHcRmieidaHgTf6UvlBRtQ7rqxVx45iXNd8/9tc/dDVzDuGLGy/PPP4977rmH/U+hn5kzZ6KkpASAwo8ZwhU3y8vLw+DBg5Gfnw+r1YrS0lLceuuthsbLtwg356veF3UJ+sNk4HR2vo8bER2Nnu4eAED3hQsICg7GsYMHlFXX7RlImjoNxz47CLdA7uVFvWjC5wdivXAOgV8VnhMIfgR+0iSXukxin4deWEbcnudT8MYCfc4bVTJdj8J1a/DAC6sRFROrW/Va5LHwXAYS0fvL5rcQMzpeWQk3N7NwxsTpaew84ydOxPARo9DZ0cEmYZH7QJOv7HyLN+YrYaUYF1s58/cs7yEt6fV5zoijPjz1RgHsUQ4WzpLdU099LexOJ8xms8bTxocYY+Li4YhRCm+R14PnETXV12kqKlMoiw83Fa5bgwXLlmPQoCAMETJvZJNu8cZ8P50TnmMjhkpJ58TNEaAnzVDuW82BfbqGql4JiNT0WbAMsTL+CP9sUDFGen6o0OiF8+dw/ezv4YgQZiIE4p3ICL/dnV0aD2h7WxuriN4fw/tysHltHr7XvAEh3k722cFTRhmAbzqumPFy77334t577w24jUmSefDss8/i2WefvVLdMnANgBeugwn4l8p5+C9sY96TyOHRbNLguQ/fuWUOrFEOpEQ52CoU8J/MZcRKPYJuIF6DDKLgWyCjQtZeoCwk0T1OIRg9bw2P1PRZCDab0dPVhaiYWCxcthxDhkYqXAKOS8PzhXguQ8aibASrIR7y5DhiYmF3OnHy+DEmra4osA6DacAAhIaHs0n9SGU5HC4XninYpCGY5m7I97tP1fv3wVNbi7Fc6Kx4o1Ix2BETi/k5uX6EWTJ4kqZOg8Plgi3KAZhMfunjPN5YuYKJlfHXjbxMmYsfVoQDmzw4/tlBjaFJnoAL7e2oqzqq6Q+vNktGM08i5UMmNOmSkjS/v162l/iM8McuKdrGDE9PXS1eX7nC73nmDSZZuFPmqaLndEXBJlbeonBtHvPopd8xDyFhYbjY0y01HkRlW94zRucTbrUizGJBW1Mz3it8S8OD0vOe8fc/0PeAft2y6v37MPUSWDyqu1cJYRv4ZuOaSZU28I+DH/0n8PFPgetHAPAC4Z31mPzeD7C46LQmLZn3fGQsykZwiBntrS24ePGiHwmUIA7kMtdyIFKhHojkOGzUdThSWc7Ixe4dRTABfhO3Xn+Iy0Bp1bK+8AO1zLiRgfZxulwYmzIZjXW12P/h3xQVWKHsgGjcicekMNpTrxfA5nDAPmw4zCEhmH3nXXj7V+tR9NqrmPfgT9FwvAZegE3OFFrhCaZiSjYds6xkl6aAYPX+fTj+2UHYohxSg0QxmpywORRiL2l49FVXh1e0pcmUvEwl27chdHC4nxFbvX8fNq/NUwyviSnMK8PfQ7p+vOFBHJoHXlDSV8go45WkxUy0y7n/ogeN54zwbfBeiuKN+QGNXrF9Vj4D2iyqcvdupKTNZGEmcb9ARjof/huXMhkA/HhQskWDeAy9558ys64bn4ghQyPZs05E6IdO/xtMl3xJIPe9LW3GwDcMhvFi4GvBDb9WtV9UmcRU62kA+p4PfgCcOD3Nb/AjyFIu9drqL8GR2iWSY+/Fi8xDBEATZqD/acAVJ6fq/fs0ac9ixos4YIsTE0E2eRRvVGTRy927NfoZ7YLnha6x2LbIJaEqwHwoAQDe/tV6vP2r9Yx4W1HqhntHEW69ewGcXNaTbJIWazelpM2ECUpoiF+9yybtwnVr8JOnV2LgoEFw7yhi5O4VBZv8stWq9++TZvYQgs1mhA4Ol06Y83NyMXJ8ImyRUWhqqGdCZ3wIhkI/UdExGu7Gd26Zg5CwMKaBw8JTqqEtnpd4/wHtOyC7/7JMNgAaA5E+F7cVvWFQz83hcsHucGLn1i0ag48MstDB4RoiPKX+f+eWOfjt08v8FgnkvQoXeEp25zAMG3WdpkCiDHFJyVj+RgGsamiQSOH888BnRdFz5OE8tharFTOTbHDsOsXSoju6jfpF3xYYxouBrw2i9kvFE2a8M36jJmWUwHNSAGhW7Dz48MN9K1fBU1urcWUH8mKIGhy8zkbxRl9KclOdkt3Dc2n4dFe9+kC8W5u8L2lzM6X1hWii1BPkk63YMxZl+/EGAqX3yvpGxqHIlXj7V+v9rhd/vJKibUzEjc5FllZLnB4KywD+ysF651dStE2TIkvXi1KKRZ0gWWYPfx1lxqtIJs57SCERt7e1aUKEgFKLq6uzE5HR0YxndOrz44geNRo7t26RZsCJxoiY8cRDlh0mg2iMUsq/LDQ6PycXjphYtDZ5NG1OuHEa7E6nX7aenqEvGmqi9kpJ0TZmMPL6Q84RIzVePD1kLMrWhAb1nncKf5InjjyfZNguu/eYpt10I3H1WwPDeDHwteHBd4DMJMCh1sZLjupCS5INbTYlM03UbqDBkdJ59arMUvhh9D8lYVzKZIRbrSikMAA3uYmrYDFFV9TZWDZvLgB/r4fIR+H/1ltFk/dFLJbYH4Iuv51sMu6vR4m8NbymR1tzsy5Xoq8+kBeB5PP5e8cfywSFAHy0shxN9XVwq9ow/PWXeSQAH7F459Yt7NhkOPKKvo+vfwXuHUWaZ0imdPxMwSZmRHW0t+Nw2adoa/IwxWDqvx7viM9oy1qylIXOYkbHazLg9K5ZoHvG82Vi48dI07ZF0jeFqHjPIH//qDDMsc8Oas6NNzx4BCK4F732KjIXP4xy9248vv4VTJg6DXaHr+ijeG4Zi7JZteoWTyNbiIjvEp0nEZX50KDsedfjtn36430I6QVbHH3yObC3zu9UDHxDYRgvBr5WDHsO6F1D6pvA9F0ZWBX5JnPVA77Bl1ayg4KC/XQteEXSkh1FiI2LR1RMLEwmE6swLWYeias5mQFBqb6B4u9iqKq/kz39LXqYKPxTwa24++IY0L794fDwExWRgytL3awW1MUeZeINxJWQHY+8CKJ8PvWfjMcK1WCzCHo3YphJ5pGg6twxo+PZZ2J2mCzUInqWCJPSZrLEAa/Xi8Trb0Cw2Yz9H/4NgL6uCbUdl5SMuuoq1FdXMbl8emZknCuZNor4DBD8jJHoaMXbERqqyY7i29YTdQOU55Q36Hjw2Xo8ZIrAZCxYrFYEh5iRfsc8BIeYcbSiHBe7u5kBpBfeIkOZUuNFvg79T6EiWVv8Z2KmWvX+fVhxMzAkRN3QC5w+b2i6fNtgGC8Gvnbkf6iGjwAM8vbi4WN346Z3kjWx97ikZHSca0d3ZyfeK3yLcS74jAbylHgB1FdXocXTyP4XeQD0N/9bHCBXcVkegSZE8hJERkdjfk6uNHVaNlHpcVgoHCNmqPAEyKycXISp9ZnEFXhf4FVHf/v0Ms01oBTe1PRZ/SYzE+g6yDge4VYrMx5FPo7s+ut5Xvh7lp45Dw+8sFojYqi3L/FfThz+jImo1VUdZeGr0LDBGD7qOphDQ9FUX8eKT+op5RJ4ciwZAP0tAUD3VOSNyIQNeQ/P2dZWXOzp8bs24jPGi7oBWiOBf3cCPTv8c0fckgk3ToM1MhIVpW5WlLK5oQF/3vQWK0Eh649MjkB2X+l3f4n1slDxs92+7XsvAbe+obu7gW8oDOPFwNcOMXzkCAcKZ+7DDYJ3Y+S48QDAVt08P8QL4Ej5XnS0t7OwROm7xQDgF3cH9Fe9ehOPzGvCD64ykbtAKdL88VimRGQkOwafoULkR/LEzFcL3AGKt0CvppEeSBMmOMTsZ6TIRPz4fvK8AhHkeUlIncJEzTRp2X8s1lxvfjLj6zjxmTvpczP90l6pv7kb8hESFobOjg5Nf2U8klCLBSaTCVHDYxAWEYH0uZl4/p678byqRZO7IR+DgoMYz4VPZw7kfRKfCzKozCEhbBu9CZjnjZB8vYzkTftRJh55j0SvVF+eRP7vyzEM6Pf8nFwAin7OgY/+5veOZOXkIiVtJsKtVqxSS0nIjiHzyMhSnGWcLJlByIeKo0ddh8dP3IUBA8DIVPf+3ggXfRthGC8GrgkMew7o+TkwUM0+un4EcHeqkhmQnjkPSVOn4cThz9Da2CgN79DqV5Qfj40fgwk3TsPIcQlMRVU2QRB4g4hq1OgJ0fH9CLSqjEtK1hgfPHiPUVN9HWuLNyJET4wXitHS3tqKqv2VASdXmZFGE17WkqWMn8JPHHpZXCIHSDyOqC1CRmVlqVvX4IlLSlYE51TFXjLEyHtCdXtk2WOiRLx4X/hU5gvt7fB6vejt7WXXkD83MlToGHR8sW3e20YhIv7cspYslRpUsntUvX8ffvv0MmZIylKpxe3F51smaMcrBvM8H17V93L5UYCvNpHIT6E2Ro1PhAk+ef/+HkOWiUbPPR2L93wC0Hgh+XDYwvA/IfjSBbbdJ58b2UXfVhjGi4FrBvf+HnjrX3z/F9wFjH50E/5JrTHU092Nx79/C/s+kMFASJ+bCbvTCbvTiRHjEjS1YwD4GRkOlwtnTjfD5nQiKiYWrR6PH3k1UNaSLDYvywIi8IRTvQle9MSIgmAiKN2XCufKsofIOzEnawFsDgcmTJ3GpPBlKeZ8PwORS3ltEd6oBCANU7CsEgCtHl8GDPWvvrqKcXFocifo8TQIaXMzmYKwWPiSP4f6mip0d3Wh3L2b3Ve9LCB+ouV1bWgSLXfvRvod81D02qvS8KCIkqJtGm0jQF/zhO4pGSV0TXkBuOsSJyDYrITR6Bmh8NeEqdNgi3LA4XJpsvACgU83F2X7RYK7dWgkPPV1bLv+crB4Y1MUUqTnUaxRRvvx3p4/PX031j6mbdvguXx7YRgvBq4ZbCoDfnG7L3w0YACQe/Zx/NVcjs6ODr9MCEIggawwNZPkwvlzuHDuPMpKdrEQSJjFgvtWrsKF9nZGDiQRLUCZ+E0m+JEcA7ncZSqfevwN6vsqQU1WdgzeE9MX6ZNP9y1379YoAYsp2A6XCzaHA011tbjY3e1XtJKHqKDKH1fmhUrPnIfYuHi24pdds7KSXZhw4zRWlJHaJi8OTYSyUJYe+FBVZalbozUC+KeMz77zLoSEhSnEU3Xi18uc4idaXtfmvpWrkJA6BWdOn/YjFBP0QpW8Ea7HQ+HvKR8mpP5YrFYkqDpE3Z2dOPDxh8wIoAKbAwcNggmAM9aFsRNT/O6FDHy6OWm78Lo24VYrHC4XAMXDVrKjSJfgy5+XHgGeOEPhVivOnG5mPDJKg6fCqbQfL6r3x5+oHeWyiwx8e2EYLwauKQx7DriYBwxQB6Cw7iYk77gZc/4juM9VoqyybdrcTHi9Xlw4fx7WSKVgXGr6LEXUissyoQGRT7vlizrKsiZ4ngOJkc2+8y5WdI7qNdVVV/mt4vuTaUK/A23Lr4xJRp7CB3yaaVZOrl9Ip72tDX/Z9BZMAIaPGo1WzymmKZKeOQ8Lli1HU10t/rzpLZbtAvhnhYgufgLvvdALIaSmz4I1MhIHPvqb1vgSPFWB1FXFVFsK0bnpHkZHIzYuXle0kMJPPNGYJn0xc4qfaEnXJn1uJkYnTgCglDzRq/BMk/+EqdNYAULeEKSCpDJRNv6e8m3zhk+oxQJHjAtv/fxF1FUdZQb6kcpyOGNdsAyxwlNfx55TmTHtd204Lx+V6OAN3HNtbSycuKd4OyMJ82FXehd471+gBUDGomxMmpGGjvZzuNjdA5vTycjeYRYLJqXNZN6jkh1FaG9rw9PnH4MtAiwX/FyX4XX5tsMwXgxcc7hnCxc+8gKjLx7C3NGxGJwjV1Il0ETe1uTBkKGRjJNQ7t4N944iljrq3lHE3OwAcKG9nbXFe0F4d36gVGmxVk5zQ4OvTlO0EvoofbdYw8HoizDJH+OZgk1+BgqBVsahFotm1f68kCk1aUYauju7UPTaqyykQ8e3OZyIsNsRYbfjxOHDqN6/D7kb8mF3OJkkv0he5X/rnYvMG0PgPVT8tuJ+/LZ6Hgkx1ZZCdKnpszRFAJ96owDHPzuoyc6iStmiYcN7WEhUT9QloW0cLheCzGZ0d3biP15YqRvKovBfVEwsbFEONpFTCnlbs1JRPMxikWY66fFNiLT6pKpDRPec51INiYxknB6SDchaslRabZsHbxxlLVkKc0gImhrqUV9Thdf2fISdW7egwr3bz6gKt1rZM0Mp+Q6XC5PUEA8Z1TIjj66TI1bx6ETYhjKyd5ZKGnbE+LxHPzj4CGxj1Z1NwKmzQPTzuqdk4FsCw3gxcM1hUxlwzxTgu2N8n62a0Yz3dJRUCTSRU10fqvNT7t7NFGAp80gWquHBcwyAwAUV+ZX7uNQpMEExfPIeytYUBuTj+BQC68uzwp8XTzKlfSik5XS5MCPjDpbpQSCy88WeiwgJC0Pm4ofx9F3zAPjCQGUlu5iXhQysnVu3YM78BczzIhqLvDESyEsUKMU4UJVs3psjZrHwkBlT/G/AJyTnUI0GAoUcSPdFLyMmbW4mJqXNZNoqYibQMwWbAAB///jDgBwcImJT1hbz8EQrhounvhbHDx1EmMXC+FayUCXPYxH7Lao4y4x9xqEKECIUQZ6opoZ65D2UjdwN+X7qwaLWklgQlLKVqG6SXsi1eKOSIn7fylVwxiqeJLqu5Ani2127ACxc5L0E3P4ffZ6OgW8BDOPFwDWJm/OB97N9BkyotxPXb78Jr9tf1AzGPEQiK00qgbIf9AwHkWMg7itOvuSlIYIsZeTwEunFG316FOlzM1logBdOk00k/HkRxAldPFc6r6Sp02BzOHG2pQWXenthDglhExYdixdTIwJo6bvFWHRDKjteX5OyTEVYpmJM4D0bekReIqJeNz5Rc24EWdsyITlAMSbJkAyzWBR+iNeLMpUTFIhPw/MqZM8Qn4UTCNRf3vND+/DkZjqep75Wmh0UKI2bfy5XCdWoNX1QvSF611+E6EWTqQdTRppJlfRfdc/dyvNwu89g9wIY6hyGgr37cOr4MbzOqRiL7xTvSSKI/JgDOdyXXmDlX4206H8UGMaLgWsWN+cDK24Gnpuj/B/TewLJO+/G80XyyVScuEp2FGHU+ESEWiysDpIsvi4r5KjHMSCI3gZaCVfskcv906DPV06mY1eoxFKqTK0XFuNBE1yYGi6i2L9oXB2pLEePqnoqZrXIEIhc3B/0ZeQRRHVU8XueiGqNjELXhQs4VPapZqLVu3d6Rg1pujz1RgFLNadjideZb0NMERafIUrHd7hcmpICIvR4QrKMOb6WUrjVKq1qLeu3HrdI7AMZEyKXqb+QZXqRcW6PcmgWDBOmTkNUTCwjOFNl8CH2oRrPT6C+i2nZ83NysbjtGQy/WMNckp98Dqx6v9+nYOAbDsN4MXBNY9X7wKMzgKFhyv8/TAaO/yIWs9+066YJ8+mu1igHrFEONkjKsmQSJk9BVHQMcjfks0J8GYuyNZkutDIGfMRNmoAovn+muRmDVR4NQZzs3Grpgqb6Olw/+3s4UlmOQtUDQ56ekQmJSvryjdPw1s9f9KtTQ9yLcvduAHIRNN6zkaoK2gUKTRH0UoT7Cm3xCLdaMT8nFyU7ijT6NrI2yEjkC22KE9XI8YmwRTkwJ2sBrKqQH6W7M/n8nFw2wcsMJlEMsKm+DnXVVeyeUlFNPR2gQBN71pKlLB2fn4wDZWSJ7cuE26hEhFgqI1A4rq/0ZM01U5+r5oaGPj0wfXG06Ngv3XePX9r3CdVgd8a6YFXDdj1dXaj5+36/4+pdOzEte+HgP2F4c43yjwk4eNIg6P6jwTBeDFzzuPV14JOfqf+YgBGoQ+H9Y3DDo/7biqtb0XsimzQohk8cAACagZI8GJTBpDnW7RmaYoYpXKhJNsgz5dg75iEkLAwXe7rZxEYchVCLBXanEzaHg62M25qaGWmUJ6UWb8yXatxoPBu3Z+DGOd9HkNks5Y0APo5PqMUirWLMh3ECpUxT6ADwcZMoa0gkHtO+ADBpehq7ZuI9oslPVHblxfwsHAFZ5j1iIQ0AHlVFNzZ+DGLj4hFmsbAU+UA6QHooXLcGC55YDk99bUBStsgTosriYnV0vg4UEb1j1JTzrwItnkYEBQdj59YtSEidAi8UPSS90GV/q1sDcuNp89o8eAEMtljQWFerKWnBe954b5R47Xhj/Kffi8R1JwpZ+95LwIS1X+aKGPgmwjBeDFzz2FsH1J8BYob4PpvSvctv0Af8wzSUdUMaGrLsFn4S5D/n/3a4XIiKiUXa3EzmkeCJjwfUCZ1W8HorWWqzvqYKc7IWoMXTiPTMeUhNn6Wp+UOGBADUVVdpuDTiOfZnpc0XqSSI9ZR4bRiZhwSAxkiQGRrFG/PhcLngjHVpUo8Bf+IxHzbTC7XxXiaeK8IXlxRVZ4kTwnuPxH7xHrO66iqWMcPft/6GUfgQijgZ879FoTki/4qKxUTiJU6L7Hz4+6dXVVm8x4BSUoGvfE2cLPEe8PtTSLT0j/4ii/1B9f59ONfWplHBJuh5o/SMx00zPkXM+RPKP+qDlP/RZXfJwLcAhvFi4BsB1wvAqWdVATuT4uavXHgE4f+m3U5vMpfpoYgDu8VqRVZOrkbBlCZPAj/5i0YPbyytKNgkzZDh+xczOh7Tb8uAXU1FBnyD+Oa1eazPh9WaTZWlbiaI5ohRxNEC8Tz4PvJZIDLPAO+lojCFqMpLRoJXvVbpmfMQbrXiSGW5JuzT0d4Oa5QD41KnaM5dJB7LuBu8Zg6vfVLh3q05L0pPz1qyFCVF23Szn/jr4KmtxbhJqfjxo0swwGTCkYoyNJ9sgHtHkdT4kD1Hl6PPIz6LzCulGpGbddKFxTb0JnKevyLLvmPbqIRZ0htq9Xjgqa9lsgHiPRD3jxweDa9XMbi/KGTn0B+hQ74fi+KqEXNYa7g8+57Bc/lHhWG8GPjGYNhzwMc/Ba4fqfwfhk6cfRGIWO7bRkzXJK0X8mLw6cbz1YwdEg2TKZjy3oE9xdt1wzO8Zyc1fRaGOofBZDIhTCUL04rbrZJxKZWVVElFWXS+r8NHXocIux3l7t1ITZ/FlFRp4ub7CfiLxvH6NTLPACnglnB94wmjgHZSvE7loIwan4ghQyPRfLIBYyemsOvGZ+fIrpXe/3ROvGYOaZ94oYUs2yVQu3SeN92WgYEDB6KzowMd7e0YMzHFz0MDgOmYFK5bE7D4YX+PTW2TRL9XZ9u+MqjEWkZ0XrLsOzomcVwGBQcz1WI6DxPAimjq9ZmKR86+8y7s3fVBv7lPfV2Xvng/PGq2r8f426o0D0L9GcNw+UeGYbwY+Ebhhl8DnauBYPXJDTcDf/hX4Ef/qfyflZOL1PRZ8Hq9iImLZy7yCvduP+PDCwAmE2xRDhz/7KBGbEsmohZosBZXwe1n2gAoEzgfkolVeSC0Xem7xZr6NmUluxiBlTRcHC4XhtjtuG58It5c/SIcLhccMS42ccclJWPkuAT09PSgvqbKb1LgQyyy1S0ZNElTp8Ea5dBwUkhTg/fWWKMcgMnE9HTEyZM8Cl6AhcT0ODIiRLXbspJdSJ+byc6T9umrrpEMqemzMMBkQmdHB0q2b8P1s7+H2qNHNF4j8gC8sGUbQsLCNAYiXTf6LVP37WtiP/HZQSaUJ0NfxpGMSExCezKInK4DbW0aL4eegcrv/9unl7Hnh46vV6wUkBtgemRt/re4P/9cvTFzH0yC4eJ6QXrKBv5BYBgvBr5xuOkVlcCrEkx+OBH436gw3LSuQ1nte71oa/KgcN0aaeVfgntHEa4bn4jGuloNnwLweShokOYRaCCmAdfpciHCZocXygTx+eFDsNjsbFKur6nSSLTzui28NgxNTsvfKIA1yoH0uZl++hcZi7IxMmE8AODW+Qtw7LODGnLlvU8/iyH2obj36Wdxz+RkP88MbedwKdkgNEfQdzTx0XnypQdo0uYnz+r9+9g5xAqCgn1NzrxRokfo5e8DednE+8uXirh1/gI01tXiL5veYueQuyEfdqcTQyIj4RozVnOtczfkwxwait7eXuzcukXTPz7LjPR56HpQJpPeufFEa71JPVCqulidnPYVM+FEyDhdvMeQQnl6HBq6J3FJyUibm6kpMyE7pszAkd13PU+LuBCYd7EQQznD5ZPPjcwiA4bxYuAbiL11Sqz7uTlgsYmp0R2o+EUyfibonQSqvpw2NxPWKAeOfXZQM5nMV2vCVJS6AYB5JTx1tTjf3s7Is4BciK2kaBvSM+fB5nDCBGXibT7ZwGorRcW4MGlGOsKHDGES7XpCaNT2mZbTsDmcLPzFo76mCj3d3ei6cAGNdbV+lZwHDRqktq0cRc9dz09evKeFJ4/yeil0zsTN4asV64U09FbbMoiEXuoXeWMopRyAlPQ6/fYMeL3AwIEDYXM44amtZefZ4mmELcqBILMZLZ5G1h/ieHRduIDgEDMSUqf4Ea95DZ3mk0q9nbFqtlJTfZ3uufHX5OVtO+CMdcFTV4sxXJFEWao6b6Twxg8Z2JU6ZFseeiEqMv4AnwE94cZpmpR0/pqKlcN5MrIYzuK9Ov2973xmU8mOIvzlpu0we3vZ94bhYoBgGC8GvpFY9T4wcbii+0KjbzL2If+mDox7rH/kSxk3gw/xUJgpNi5ekZZ3OOH1elmlZn7CE1VlZXWNUtNnKSJeqkhXV2cnIqOjGUFXJoRGqc4Xu3sArxcX2tv9zmP2nXchKDgYrZ5G/GXTW3DGuuBwuTA/JxcTp6fh3JkzaGlsxFs/fxGA/kQiE46TkWrFc6ZFMV9vhtdG4cMufWVHEXgvA4VYKOuFZU8B6O7qwrGDB6SkV1q5d3d24uTxGhYeyliUjbETU9De1oYhZjPOtpz2m3jJMxZmsWDijDR855Y5KHrtVRbKAhTScuTEFNRVV6HcvVvjAQrEXXl8/SuMt+QFdDOtCDxxO5AA4uWCD7G2t7UhzGKByWRCe2uLJiVdVvBUvI8iIb54Yz4rBioj4OqBN5BWRG2DmfO4VDUbhosBHwzjxcA3Fj/6T2ByrFYDZoy3ClXLgPjVfa/2ZLLufFiEBl2qURRmseA8p1HB78P/ptV7W3Ozph3yyDz0Uh6CQ0Jw7OAB5qkQib/UlsVqRXtrGyLsdrQ0eRifhAdPXuWF+Q6X70V3V5diBOwpQWr6LFaIr6/0atl14ydkPrxRV3XUr94Mvw+fHqwXLpGl9EZGR6Nyj1uT1kyF/sjzYo2MRGNtrd/kLYZKyNgib4GirzIR8HoxxD7Uj0fy+PpXEDlcIble7O5hdaGCzWa2HWVekdESGz8GT71eAE+9omUiC+WQZ+HEoYOIsA/FW6tf9DPuxLCeSFbmt+0P4VcPVJi0s+M8JkydhvaW0/B6vTjdeMrfi3VbBppPNkhTtgF5GrxYGVwGvYyjH32ei6lcbbPeS8C/bA54Ogb+wWAYLwa+0dhbB5zrUoi7NHKOHqqUFVj1fuBJWi9TSORPVO/fxzI0ZJMC384zBZtYyIlCRoAvfFVXdRQHPv5Ql6fx1BsFsEU5WAbUxOlpOHNaybg5wYW3aHvqExXHq6s6qkl5Djab0dRQH5CjwLcl8h14LwygFe+LHK6otOY9lC31sgD+NaLELC7qD/9/uNWqFCZs8rB+86ERMgRFg0fv/gL+Xor2tjaEDxkCT30dM/pEI5aydI5WlMPmcLLCm7ywHK/VkjB5ClPaLS/ZJQ3l0KReUerGicOHUVd11O9eiteGUsxJY0evHAIVoOTl/gMZMx3t7fB6vRgwcBDsDifCI4awCux8qEz0Rsn4OO4dRRiVkMiy60QDOFANMTGE+bOWn2FEdK8ms+jGXxs1iwxoYRgvBr7xiHhaTaEe4fts5RzgWAvwYU/fK1FyeU+YOg3Wob5Yf6AJhYeGOKkWpisr2cU8Ow6XCymqS72dE+sCFD0YMmQyFmXDrmby2NUMqNJ3i9mkoVfbh++TKMwn4//IQByR79wyB8EhPu8CPwnFxo/RkElvmPN9RMXE4qnXC/DS/ffo6p44XS5Exbo0k57I6+GJqFR9+ISalcN7dXjIvFWBMl/4+0aeI6r5JGubPFp/3vSWnw5MO+fBoX6XlexiSruBCMTEk5k0I40ZInQvSWW5giNc8+c5anwiHDGxmJ+Ty+6zSHDlCzbKnhGxInl9TRUyFz8Mc0gIzrW1serrtA8f7grEx7E5HLA5fKU4ZN4bWQ0xuh+Pr38FC0/mYEREl7KTSsj/4KhhuBjwh2G8GPhW4IZfAwdygMRhyv8mAG/9C3DQcR0O36Tlo/BhDECpnQMATXW1OPDh36QDf6AQFG1XWepGW5MHNrUwHQ36K9SKz16hHSZaBrDJ0OlyYfio0Th5vEYjoR45PBpz5i+AjUtl7k94p6/VN4H3NHR3djFDgzdYiEy6YNlywAuYzWYl1ZybsERU79+HxtpaPz0VkdfDE1Fl/B89Gfnq/fuY8Zk0dRpevO+egCt7Ak+MFSdrSi3vONcuVVWmVHbiFFG/ZR4hColRGKy5oYGp+tLfuRvyWdq7RSDl8vd0fk4ugswhgMmkIW7LyNGB0pHFbKC9uz5g+i1lJbtYRhFPkCb9H722KkvdOHH4M0SPGi0Vs+OfL/5Z4TPa/t+JZxBzsQvMbagaLjf3ze028A8Iw3gx8K3BhLXaKtQwAYlNO9DxSRTKSlrx+PpX4HS5ME4lSwKKN8KqFup7nasILNZz6Ut8jP8thjJ4ZVkxlEErbfqOJnricZBX4khlOVxxY2ASJi4ZxAmbzyIRPST8RMvrgZChsXDZctgcTixcthwv3ncPEiZPgSMmFl4ALapSa0d7e8Askr4qVfcnE4WulcVqxX0rV2FcymRmxHkBmEwm2BxOv+Ka/ITLh1dktXrobxLK83q9fiUV+DTwilI3KkrdfvWJeC/KOVXsj+fr8CEqMmoo/AbAj4dFxg+VeACgIW7zzxRfOZwMKvG5lWUD8Vwf8gySUSd643hPF3/vcjfkIzgkBLPvvAtv/2q95pg8d8zpcmFFwSYm2OhwufCjQ/fBduYACxX19hqhIgOBYRgvBr5VWPU+8PBNahkBdQU3pe4NrPnxfai9PgNtKn8EUAiLhZxEu7hi74twqBeSkIUy9LwSYpFEmSE0UU21HhwRAcA3cel5FcQ2TKDJ3d9DIrbBk1zjkpIxQE2zDh9iRWz8mD7JyzKwYpRzM3UF62RpzhRioM8oRb334kWmXgz4+Bbm0BBExcSyCtOiN+WBF1YjOMSsTMZmMypL3RoDh+5pi6cRkcOjMUBV4hVJ0ryngzgmfKo2T7aeflsGjqop1aTWy3tnKDTFeyREgm/uhnxERcegrcmDxrpaXGhvlxK3+YKO1A+Zx02m8cKfGxmJJTuKpN448Zmh4+zcugWz77xLqnpMx21va/MTbBz/4WOK4aLiYi/wk7cNw8VAYBjGi4FvHVgZgRHwGTD1b8B+/ggW/XcUFjyxHDaHAx3t7RrDQiwtAAT2BvDCcqIR0pcYG0EMMYiGDj9Rps/N1IS79DwWYhu84q24regVEcM5FqsVvb29CDKb8dBLefj7xx9q5OX1qmfzkK30aaUuI5dSn8Qq3xWlbjSfbIAj1gUAOK8acanps2CNjETb6WaEWSKkWjmU4eX1emEODUVTfR0jA9OxLFarUgk7OhpHKsphdzhRuG6NNPWZ54DwHBP+GhLnyGK1+oXN+OeDDBjRiOKzrJoa6jVcERlETZ6+1HDF54QXBSQdo/rqKgwKCoY1KpL1X2ZgT78tA6XvFrM6WulzMzVlL/gwFhHK3TuK8OZNH2OEt46FiQBg6iuG4WKgb1wx4+Wpp57CD37wA0yaNAnd3d2w2Wx97vO73/0O99xzj+azv/zlL/j+979/hXpp4NuKG34NVC1TMo/IgBnd6sa/35SMn/38RWktGJGD0tekTLoY5AGQCXWJ7n9REVb0MuhNlACkfIr+aGfIhOVoMiGvCInl8WEzJcXajcNlnyqEztBQTEqbySZCWqFTyCRQcUhxpU/p5DJyabjVihZPIwYFBaO+pgop6bNw5nQzDpd9ChOAVk8j8/rw11ivxk/GomyW4mwymdDqaWQeHZ4MzId33uBCiHzfAPh5WKg2FH8eYl9kKfn0m65F1pKlAKBJAxezrGTQK2WhZzTq8Z94/krpu8VMubey1I0DH/n6r2dgizyuGEFdmfYlovHdqcCIaWojquFikHMN9BdXzHgJDg7G1q1b8eGHH2LRokX93u/Pf/4z7r33XvZ/V1fXleiegX8AxK/298AkYx82fz8I6b/2H7xFDooIcXKmVNMOnTCOaFjMz8lFSvosQBUFo+OdaW5mxE0Ky4gVsPk29L7rC3oTME+kpAmIJ46+/av12LvrA9y3chWiR42Gw6V4Pvg04f5I//OTHp/tI5JLLVYrxqmKtbPvvItxPX708M8wcNBAlP6x2K99i9WKtLmZLIz1TMEmdp02r82DxWrF6AnJiLDbNYrKomGop8jMe6hEYyF9bibzVDhcLiSkTmEKynSMQKFE/h5kLVnK6nE11dfpZi3xfdS77rzRaIJixGTl5GLS9DTGx9Ej91bv34f0zHm4bnwiQi0WaXkN2XHCrVYcUaug854XEe9nA98do/3snX2+GmUGDPSFK2a8rFy5EgCwcOHCy9qvq6sLjY2NV6BHBv4RccOvgd/8EFg81fdZ9Nm9eHPhNCx80z9kI4Z/ePCGw+a1eTABKHfv7jOMQyAPzdnWVhRvzGdhp8PlewGApcB6ue1psouNH4OsJUtx8WKPxttzOZD1j4iU961chQk3ToPDpajkVgiS89X798FTW4txKZMxzjZZ42nQ+x0IaXMzERUTiwVPLGceH34SDLVY4IhxYefWLfjhgz+FOSQEQcHB8NTVsvbFFHXApyXDi6bJPE8yBPJkpav9TZ+byQi7FG7ir9VTrxcox/V62Wd0XDIIAX8jg7hGYnhQ9GjJUo77IkTzJGPyqlisWrKuWCoAULxytigHbA4n5ufkSrWOxL4RT0ym+0PbvnjhEYQIM49huBi4XFxznJeZM2eisbERra2t+OCDD/D000+jpaVFd/vg4GAlZVOF5QsM6ga+3XjwHeB/jyup0zQ6T/P+DZULzXhP4k4HoBmkadClLB8v/Ivs8ZWb9Vao51VPTdX+SlTv38cmigvt7Tj22UFWFLFQNYxGJiRizKQUhFutSLz+BoSEheFSby+8Xi/je+ghkDy9iOr9+2B3KOJqAHTDFDxfQRa2oomUUn/FzBe+XxRyszkczOPE99NTW4uxE1MQMzoev3kqFwuWLUdTXa0mI4wPc/AVwQFthhffv0ACffz95rlPqemz2L0fmZCIt37+IixWK0ItFtRXV2lIy2/9/EX2HNDETV6gI4K8v3hsnifD90Uk4Yopx2QIUkq37Hwou4ovX8CHsmRp+xarFWdaW1iBUZmHp79yArTti3gaIUFgYSKvF1i4BdhUJt3FgAFdXFPGy1/+8he88847OHbsGOLi4vDSSy/hz3/+M6ZOnYpLly5J93nyySeZl8eAAT1sKgPumaJ1VYd5u3Db9pH41f/+UDNBAFo1Whqgj1aWo6m+DofLPsW41Ck4UlnO6uVQem3WkqV+KrMEcUIVyxPwbvx2NYW7uaEBJgDmkBD09vZiwIAB6LpwAW6VUAzIJ18ZGVavAjOgLTGg13+er6AH/joc2vup7mRXUepGeckujByf6Kf9AfhPhKS/Q/1Oz5yHCVOn4WhlufR8+jJUSPiPUpfF+y3yNipL3Wj1eGBzOJCmel9keix8VWwCeYHOt7f7KTqLiriigUAk3LamZlisVgDwqxAt1uiStUHcFZGPw4eM+JApGeeVpW5dvo54nwJ5ribHQvG4GIaLga8Il2W8vPzyy1i2bFnAbRISEnD48OEv1Jnf//737O8DBw5g3759qKmpwcyZM/HBBx/o9mndunXsf4vFgvr6+i90fAPfbtycr+jArPweoMplIMjrxR9u+gDu6F+yCULMNOK5GGMmpmD2nXcpxM6TDRg7MQXn2to0kz/QvxozMuKjHuG3XSWWZi1Z6qdyKvJgiAB6prlZU4wwRVX/DbVYNDWVAO2kK+t7fz/jr4PT5UJ3V5dGtEw8r/k5ufBYLH5aKWLWk2gELVi2nEnaXw6IIE3Vph94YTV++7QyppWV7MIKtbzD4fK9uKDyNtLmZsIEoKm+FjaHg03wJFaXnjnPzxjgwRutvCdlwtRpsEU50NbkkYbe4pKS4XC5cKa5Ge2tLSw09vw9d/tlk/EGht5vh8uFSWkzceOc72NQUJDmeoohU9EoCVSLSlZig78WGn6Larh09gAzNhjkXANfHJdlvKxduxYFBQUBt6mpqfky/dHg2LFjaGpqQnx8vK7x0t3dje7u7q/smAa+3Vj1vvLTm+czYIZ6W3FH0Qj875C7YJ2Vp8k0EnkS7Vx2ivibD3305WKXrVBlhF9x0qirOoqsnFzNZG9zOpUQjBr2IQ5Eq6cRk6ansYlzwo3TYHM4WPVnh8vF0oF5b0F/+y77jDeCXtvzEULCwjSiZaJRQho2kWqVZ9l1kYUj2ltOw+5wIths9vPaiBCzwCjs0tvbi+AQM5tsn3q9QAmdeb1wF29nbZJnprLUjT3F21FfU4UVv1OMnIEDB2JUQqJyXV0ueGpr/SZw8ZwpnbnV0whblENDIBa3HZcyGSaTCQNVvZ1QjgelFw4U0//p3Ofn5MJkMiE4JETDHZJBbFPmzZH9Tzwg+rziCTOSo7o0dYrqzwCuF3QPbcBAv3BZxktzczOam5v73vArQkxMDIYOHYqTJ09etWMa+MfAwi3Af/wzMGgAAJPicr/pzBas/6MLxf/xHtuO13Jxq6Jd5EmoqzqKkqJtzDPAkyhlE25fnADZ9+IkUb1/n1+WzxD7UABgvyn9ua66iinAxsaPwYlDB3H80EGmbJo0dRpsDqdfqEuv7+RpIKOpr/MhL8zOrVuktYdoPyKp8m33lRLe0tiIkePGo7XJg7KSXXimYBPTDiHDgUjOvRcvYvio6+BwufDkvLkacixtOz8nl3F+WjlPiHg9qvfvw2t7PsKQocq17u7sxKCgIJhMJoxOnICxk1L8JnBAns5M91cvs81iteL82bMIHzIELZ5GVOxx+xFt+wKffh5mseBMy2mcPH7MLxW8L8i8OXyqPH8v6Zr+5odQDBfAkPs38JXjinFeXC4X7HY7RowYgYEDB2LixIkAgKqqKpw/fx4A8Nlnn+HJJ5/E9u3bMXjwYDz77LPYtm0bTp06hbi4OOTl5aGqqgrvvfdeoEMZMHDZ2FSm/HS8BCUObwJMXuBnPWtgSnagdpEy2fJ8AuJzZC5+mOmGkMKorG6LCL0VcSCNlP4YQW+uflETsuK/p8mLuBsVpW42uVAYSlRE1eMuUBuU0UMGjN55kBdGFvYRj0PejcTrb0DRa6+y0Jy4D107L5RMr5IdRcp9UUsWxHLaIgmTp8AR60Jvby8AwBHj8jsukYpDLRZ4vV60Nnnw0n33AAAziArX5mn6ULhuDe5d/ixMJhMajtdg7MQUdHZ0INhsZiUARGNMz+smS6Gm6uL2KAfOtrYq/WpsZErBcUkTMXnmd/Hyth0aIyQQUdcEYKzqxSFvUF/QMyDpc1I8JiOa976dexEIC4amTlHNacNwMfDV4YoZL88//7xGcK6iogKAkk1UUlICQOHHDFG1EHp7e5GcnIyFCxfCarWioaEBf/3rX7FixQojLGTgiiHsKaD2aSBGpU4MMAGPpXjQfWkpsGiNhk9AK/mdW7cgZnQ8MxL4VFd+RT39tgw4XC44Y11orKvVTDR9ueEJeiqo/eWrlJXsQtLUaeg4184mMf44euRc8XgWq1Wj3krg+11WsotJ8PPn0ZeHhr77zi1zEBIWphiHIWa/Y/HH5OvvRA6PRqsqnQ8AddVVGuOs3L0bKWkzUbhujd/14TOW9hRvZ58/vv4VjYw9HxIsKdrGjJ6ykl3w1NZqCMAlRduYMUaGnuwaUF8cLhfGpUxmSs0Zi7JhU6uLnzxeg4o9vhBPavosRNjtAICE1CmakNDIhERYo3xV0fnrVLwxH6PGJ7Kiof3hZInPpNjfE4c/Q/PJBk2K9m9+CGTf6AvJAjA8LgauCK6Y8XLvvfdqxOZkMHFPeGdnJ2699dYr1R0DBnThegH4w78CP0xWP/ACwZcu4OXuRxDuBFZxpMZAmTiybJlJM9IQYR+qhGfUmjvEveDDMP2Z4AHfhMKHqGTfA8qEk5o+C9YoB6xRDnx++DCKN+ZryJ18OENGOqX2ao8eRndnF3Zu3aL5nvg19TVVeOCF1QgJC0NnRwerSCxbtfNVvXneRsn2bbj5n7NwpKIcw0aM0Ejz8/vw14qqXlNdncjoaJT+0b/Kc+jgcGZwiHV5qC1Kec/dkI+dW7egwr0bIxMSFfVb7t7JPClUKiBNrd8khodkngvSfTl35oxGu4fP/BEzqYo35iN15ncRPmQIzp05wzxfZGg11dcxsq/T5ULb6WaUlexC9f59ePG+e/z0Yqj/fL/oHMVnkvY5o9YHG2IfiiFDfWUD1s13YHGKR8NvudhryP0buDK4plKlDRj4uvCj/xTE7LxA8EClQvXE4cr3gVarejWKXt62AxH2oThz+rTG61G8Md8vDEOr2xUFm3RTmol4SwUI+QwRJunPVUsmI8kLrYy9mFIt42mkZ85D0tRpOFJZDkesi3lF9u76gPWLPB+z77wLwSFmdHZ0oOi1V1kKsKjkyk+0Yl2klLSZGDhwIMZOSkGw2cwmRX4f2o9vj44vqzMEAPevXIVxqVPgdLnw+soV7LoA/kbFgy/lwRwSgjlZC7DoxlSNocFXihaNE/pN21E9IdkzQkYLids5XC5YbDam1BxILLF6/z4crSjDpLSZOFpRxowMusdkED6+/hWMVVWK6Try58pzVihbasS4cRiVkAiny4Vl8+b6XRsKQZH2DXmbykp2Kd5Lq0fppBomaukAIp+VnoYBA18ahvFiwICKB98B3vg/YOOPgeThvs9/mKyke/53otyNLq5kizfmM02Vv2x6i6UlAz4BsPtWroIjVkklpomQVRBW5fBlKc3V+/fhxKGDLGWXR8aibEyckYbmhgbWXsaibGYE8dwTPsRDPA2akF7b8xEj29ocConVU1eL8IghCA7RZvfIiJpkUJBGjejp4IXuePAE35jR8cx7U1ayi+1jAjS1oGTHFw2+KLWQY1Ssy4+nQ9eNPiMu06CgIMZX4cXjnC6Xphgnz1chTkpMXDzCLBY/NV1e44YXAuxL+Vc0jMXUaJn+Du+9Eb1s1A4Rv8mIvtjTo7le4rM1STW4+Pt8U9vvcd9N22Dywo/fEr9aejoGDHwlMIwXAwY47K0DJq0Dmp8D7GG+z787BpjZ/Qj+uuNPKC5QfOB8Jsdgi4W56EW1UpGgCQBOdYLgPQz3qwaN99IlwGSCM1ZJaRarAouTF4ERh4dH46nXC2AyAUMifRwI3kvw1OsFCAkLQ29vL3Zu3cImpNwN+XDEuvDAC6tRsn0b0u+Yh7MtpzFmUgramptx4rOD0kmWMq+offodGz8GMXHxrCZQVk4uAG3IiCAKvPHGFsn7Z+XkslpQZESJ9YnEzKa3BEIzb2gC0Bh0E26chkFBQQgfMkRjpJHRs/yNApigLwiXPjcTjphYmMPCcLSyXJPSLoaoCIHE3QB/HZ++tidwhZp9fb09A9+5ZQ5++/QyjeGXNjcTdqcTdscweOpqNZo71GfA51lyuFz419PPI/rwXs1BvF5g4X8ZwnMGrjwM48WAAQkin1UE7Z65BRg4QPlsgAm49dKfsHsu4NqvFf8alzoFXq+XGQFE1NWrN/Pm6hexcNlyNHJ6G7TiNQ0YAE9dLfOG8FWBKctHT+Y/76Fs5r0BfBwI+p68CCRSNmDAAE32VOG6NXjghdUwh4TgOzd/D+bQUABAd2cXrJGRuBgXz44Xl5SM5W8UwBblYBOrjBBLonqp6bOYUB6AgGq9shAYv/oXpf8JYv0p6sviGTeybfQyuDIWZcMaGanbfsaibAwZGonOCxd0hQy9AGAyIcJmR+/FiwrxFmBZUYHKR+jBK/zWA88NMgF+4TmeFJ21ZCkWz7iRPQ+xcfGIjFaEDcelTMb8nFzN/RE9UD+N2Ibos3s1x+/1AvcairkGrhIM48WAAR2QoF3VMmD0UN/nMUOAnl8MwI/e7VC8GG8UKOmsHg+btKk2T7pK4BRX2zIZ+bdWv4gFTyyHp96XmSRWO+4LZMDwpQAAaLwRGYuyEW4dglZPIwYMGoQhNjvOtrYg3GpFXdVRPH3XPJZVYo1ywGIfiuAQM7o7uzQej4xF2bCqWTE0sdLqXgzr0G8SyutrIs5SjZCzrS2Yn5OLkh1FzJjhCxaKXhZ+oieviOi5Eo2/spJdWP5GAS50nMeZ081w7yiSFhQkcnJUdAyylixlBSU1adRqXSovgKFOJ+wOJ0ItFixcthw2hxMLly3vM8NLhKjQOz8nF2EWC863t2s8WDw3qNy9W1Pvic6B0tD59Hg+zNfiaYQ1yoFQi0WTKk7HyHLuw7Pdj2DAaWhcO598rhRBNWDgasEwXgwY6APxq4GPfwpcP0L9wAQMxCUU3VaFT4JnoT7KgVaPBy/df4+mVgwAP69JIIgGjShs1l+IhE9RZ4U3KMhTcfHiRabGK6r7Ep8kyGzGd26Zgwvnz7H9eaIofUahKzGsAwAv3X8Pa1MmXEcwQclGHGIfipS0mRh//Q0INps1tYRkGTNi/ahwqxWjxifCOjRSsx3vISJujw1g3jOZ4nDC5CmMlxMZHc0KUIpquuSBWq4atRfa29FYVwubw8nSuWXQI4TzXg+eE+X1etm50jUVibviM1D6brHGC0X3jH7Hxo+BXeU5iZ6bAzlA4jBuRy9w8RKw6n8UI9+AgasJw3gxYKAfuOHXShjp2e8BAwC26ry+exeSdyTgZdtvdScc0WvSH40NwKfuO+HGabBGRvpl7vQXoveD9xaQZybMYkFTXa1GWE0Uc/t50bsYOHAg0u+Yh//ZUqghA/PnLWreANBU3abU4kBG3ea1efACGGyxwBHrwhC1SCXvRaDJmvoMwO+6nuMKXPL94Q2fwnVrsHDZcpxpOY3WxkZ9LlF0NCsDQfWJqNaUeB4Zi7IxJDISTfV17BoHIuWKfeJJxbyxphRp9MBTV4vz7e1+ZHG9UJxeKr5emK+uuop5bs7vfQftvwjBYG+nZl/POWDYc7qnY8DAFYVhvBgw0E9QGOnsC0C4Wf3QC5h7z2Fl0934yTMD8e/2f9dwNGReEz1BOhHkfYiw23HuzBkMtlgwaXpaQCOGNxLIexBI2bd6/z5WIbm7qwvBZrNfCjP1ecCAAejt7UXRa68GPAcZJ0esuq03mfJtkPdI7DNv+LSrGTPn1LRksU8iSZZIv4MtFlSqfJpA+j3UF94g4//n61vxhQn5VGq6R33VX+I5PrLrq0f47eta0jn0VU+LPGkUmgOAnQtbMGKiKtLChYkOngImrNU9nAEDVxyG8WLAwGUi4ml/UTsAGBHRi9W9/4bYn/wUdaaR0kJ1gE/YjTdsxHTZjEXZKNlRhPHX34CQsDBE2O24eLFHqpLLQzQSxPZJY4TflzwLUdExaGqol06CMiOAzkWvmjAPsep2oIwZPc0cWV/Ez/i/ZR4uPhNsT/H2Pj1YgUT1+CynFQWbFC/Z1GnSENX9K1chKtaFt1a/6Gco8ZWmKSwmO8dARG1ZCYX+ePdkIcQK9248nHwKP0vxaNPxvUDvJeB5I0xk4BqAYbwYMPAF8KP/VH6zGi4A4AWC0InHetZgddUUNKvKr3wpAcAn7MZzK/TSd3/79DIseGI5BgUFsWyfwnVrpCq5xRvzNUaCTIeG1xjh9xW5GwSxIjMRgSnbJn1uJiZOT5Maafz+ZSW7/n975x8WVZ3v8fdIMIYeGWgAHSBLFKg0FLWsVUYybz6PJZFbec0rtrW2Ke7NVckyuxaVu7YYe1PbTK92M/Ou11Ruz73VWv5IXSvJwDJNIIofCk4IjiKgcu4fM9/DOWfODFDYMPh+Pc/3GZjz63u+4pzPfH68P6gsKcaEadNhT88wFOATtOWZ0j+svfUK0htqwrtx7FABzrtDLkZGo/h56rxsXH/DjYrWjQnewzLiIe+s/REXm5s1humkR2YiMWWE65w6o1JsN2q90BZGXjb1Odvy7ukNHJHsnO14FKHDVCEi9wZK/JOuBI0XQn4GvRe5BOzuGOR+Q3b1dXkq4SDw6SjcdO3tGP1EHgDvnYXVP4ukSSG5n2JPw0u/nQEAyrfzFHuaR36FUb8ivShd0vARHpU0bT3k1Nuj4uKQlDICkGXcdOttCDabcexQAfa+l6/xvAjUHoWk4SM0iaZ6/Rs1beVnGHl52mOoTXpkJpLdvX7EtdVrBLQajb0tFqTY0yDLMpobGxFsNkOGsYcsf+1qRXtHsliQPChRY5jmr12N6Lg4RMbGeTTBFNuFoFx7/10A7142X2uoRn+Nx8umo6+jpbVkSxUmev0fLhFHQroKNF4I+ZmIb6N6LwxMwO3yfpx+3rVPajvyROKH3IzMhYtgiYzChIemI0wVghAl0GpJd+ERATxDOGojSSRiqkt8xTb1K+DpbRHbF61Zr9xaSM+eAIDzTqcy7xR7muaeWhNMHaitqcbpmmrI7mN8VRsZ5eiIJovWfjal3Fqo3IprqXM3xKs+N0RI4hsJx6n3e2heNiDLqDtVgzf/+KJGVdbI2BFVXFPnZaNw7x6NUvHurVuwcHI6vKFWuwWAqyUJtusG4HjhIZ/Ghz4U520NBd56F8V9/jwu/AkI6tGiFZORmdtCui40XgjpJHovAobHAvuzgOAgKA+CsKuBz34P/OXz57EXMHxwi0RSdZiiuqIch/+xX7O/XtIdgFKOLLwI+hCOCJfU/eiA1dZawuwN/Tdyse+bf3zRJWDnFq6rra7WVNLoDTPh7Wk460TisOE4tGeXYmyIHJHeFgs2usXkjDwq6jLlSFsMmhobceb0jwiPikIvSVLWRSSb9rZYNPet1yoR6yc8P0b5NCLfSK0NIzwb6nyliuLjyjFAa3XYoT27MO7+KV69IuLfW+3BEYmygKtbNAA0Npzzma9ipBUk/o7UGj9GLSxeeSILltNFeOl8FsxDPc/NSiLS1aHxQkgnUlABmBfqEnrhCiU9MbQGuJCF478aiK/ufgVA64N+6rxsxXNxprYWxYcLscfdAC81PQND3bkb4uH50eZNSEwZYehF0IdwRLjk28JDuNjcjMrSYuWh31ZVC6B90D4zZbImgRWApv+QutRa5PbUubsQm1TnEh2UQyVJE1oSr6JjtvqehOfldHU1fjh2zCVVr5K7lwElGXdj7jJXq4OYWMhobeaoV+3V01a5sjpfSRgvgl6SBJPJhF6S5OEVMSpJHn3PJAwedburjcM1ViXEFSpJiI6Nw0ebN2Fu3gqlAeLG5S8rnbF9dQFXt6YAYOiRKn/GJbao51ILsOZThohI14fGCyGXAZHQezoHCOup3TZILsaAbf0h95qJfPeDXp3vUFVWipwZ0xRPSqEqd0OEf2IGDEQvSUJiygiESpJHF2DxjVxdgmsCYE0epnRgFoJmomRY/YD1yKlRqeY+byCAJ1ksHl2yjYwpUdHyw7FvcKrSVYIrklU/2rwJGY/NRqQtBg/Ny1Y8H8KLUlFSjMqSYiXZN37IzUo11tQ/LMDpmmrFeBAhq+amJpQe+UpZO5Hz4s2joX7At2XYCU+LCF81OJ2QZRkNTqfGK6Lu3i2O7W2xoN7hUBps1rjbOJQcLsJT7hCTWNvmpib0DA3FnJdfwfHCQ0hIHqbozgwedTu+P3pEWRN1fo3eO1RyuAgTj2Qhd7rx3+y7Ra1/t4R0dXr4ewKEdGfCFwPT3wEuXdK+HwQZ9517Hcczi3DwmWsgw+VxgVuRNX7IzYiKi0P9jw7s3r5VeXj2tlgUY0PdLVmNqLYRxkiyWzn37dxl2PtePjYufxl738uHDGDo6FTEuPsViYf1pEdmas73xe6dSmuAp99Yj2fXb1BKpfPXrlbOZbXZ4DjRWjEjjKndW7coIRqRUiFFXANrPxvOO53Y+z/5WDZrJmIGDERITzOaGhsRKkmauYi+Rs66Ok258l+fWaj0gWpwOgFZRpT7we2oqsJVwcGoKS9XHuz6SiuxTgJ11+kvdu9UkpHFfuKe1AaneH07dxk+yd+mhGsEwuCsdzjQ22JxtT4YnYrvvjmCQ3t24eihAlSUFHv87Yj57t62BbIsIygoCLbrBsBxwmXoOaqqEB4VhaGpY5V1KjlchHNOJ8Ijo5CanqGca+3EU2j5syqxXEVlPdBjAQ0XEljQ80LIZWbDF66haTEguucBSLH8iISzT+Llw3/C2bo6fLF7p6a54oRp0zH1DwtQW1ONhORhiudAdEsW4QVNSMJL8qraM2PPmIybbr0NkTGxmDovWyPxrybFnqb0NgqPikJ4VJRHzohaSdhXnoaQ7zcKbYmQmLWfDTIAx4kqpRu1UbhHJAmLEBMA3DDyVoRZrbCnZ6C2phrBISFK4rI+EVjtDfFVveVNDVjd3duXroqYc2+LBUNHp6Jw7x7FEyaaRwqhPfX5xXzn5q2A3NKCpqYm1NacRP/EG5CYMkJJ4A6VJI3K8MAhyTCZTAiVJGz87TUY0ZwFk9V9UlUFEfNaSCBD44WQXwjRuO61+4CZo9y2i/th0ruxEs+dn4bSH4GV9hWu3I6aGnx/9Aj6J92IiOho9AwNVR7oQGvCpq/eRb7E4FLsaQgxm2EymWCCdxE0fZ6NCZ4Gjq/rGFFRfNyjpFuyWBSDA4CmG3XymFTUnXLgoXnZSohEb2SI+3FUVUEGkDhsuHKf+sTWR5fkICo2Dmdqaz0SqPU5Pz+l7FifOKs38MT+ar0eYQTpRfHUpdRXSxJMAK6/4UYArbk9iUNTcOPIW1F29Aik8HAkFyxAetlGz4nKwMVLwG/+xu7PJLCh8ULIL8zj77qGptGd+9vwgGuA3AtZaN72r3jwveuxfVexIkZWW1ONhKHDPMqdv9i9E4NH3Y6ouDhNaKMt1A/Ft3OXedVRSU3PQGRMLCZm/gbFh4s8BOb0D2q950G9XYSqAM88EqEjI8syvvvmiCbEI3RihNdH5HbU/ejw0M1RGxl6Q0vco+26AQCASxcvIMWepjE89OvXXml99as+cVacVx2SEvurzz83bwWGpo6FyeRyy+XMmKapkvpy7x7UVFYojS8BV7iuRZbRMzQUv/9hOq7+rt7zH1sGZBlYfYDJuKR7QOOFED8h9DM0InduQuRL2DqxGJgIvFu0Bb8es0Uj+qYud06xpynhHKMyaF/dio26T4vQjWSxwFlXh1B3FU0fd4dnfXjD6EGtRr39yz27FKNEnyD87PoNiraKviuyCJEIY2TSIzORoPOs6I2MHAM1XGFwfFt4CI0N55QKHsCzuqqtlge+PFxqw1BtPLUlPpe/djWG3HY7wiOjNEncRsaR+rwP1i/HLb1KtTotqhAR1XFJd0P15909kCQJZ86cQZ8+feB0Ov09HULazX//C5AxxFVWradFBp77ENhY7WmI2DMmY/qTi1BTWY41SxZ7PGifXb9B6VljJG+vf1iL194WC4a6wzU1leUwATCHhiIiqi+qykqVa7Wnd4/eM6Pv5yPCKkbqtUaGgwivXC1JOO90as7r6xy+JPUFc/NWYPQ9k9Dc2ISQnq4wlDqvxtv6tZXvY7Sv0bHe+ikZnefZs1kIM3tsBsCcFhJ4dOT5zWojQroIv34LCMp2qZrq6WECnrsLOJ5ZhGcvLtJsS7GnwWK1KlU1AlFNIzRVvH1LEd4A0XZAVAdtzF2GulOuct4GpxMLJ6cjtLeEPhERSEoZoalwqS4vh8VqxdQ/LNBU74jtOTOm4Xl3CERc02pzacCotWpEeEVf+aSvDCo5XITnZ0xDTXk5ksekIntVqydHHGd0jql/WACrzWY4T0H+2tVwVFUhxOxKUhbCft7wVqXlbd7iHgHgoXnZSJ10L55esx7xQ25W7kGE14zOufhO4FjmYeRezPIowxd8/gMNF9K9YdiIkC6GCCctvhN49k4gKMi9we2RCUM9jmcWQW5x5TD82YuonKhu+XLvHnySv81rwqnPhFST67LCGbRx+cvIXLgI1RXlmkTX/LWrlXCTWqPFqPeQOilWzNGpSmQV51PPSRgIIpSlPo+4rtrjopf7F+eItMWgRZYNlYaNGlwaNdb0di9GHba9zVsgA4DJhPDIKGSvWo3KkmJFUNBxogrnz51V2gwsjtyi6qEle/jNL7YAOez4TK4QGDYipIuz+E7g2fFAUA+0WhG6/7W1DYD131w/q8XtjB6YgHH4RB/eEQ9exwnjsIm4jj7sIwwStbCe0Xv60I64nrpxokCcOzouDgnDhmtCYO0NN6lDVXUOh0bcTX8/ABSBuL8+sxAVxcc9ko718xThubpTNQizWuGoqlK6dUfFxSktEkQ7BDHfh+Zlo/8NN8JiteLLT/bgbF2dsl5DDs5HfMXfXP/sXj6pK+uBuBeMtxESSHTk+U3PCyFdnJwdrd+mPbwxbiJCgRZ3f74Pip/Ch+95NiZUo+5ILCTne1ssmsRbfUKqmvghN2sE84DWkIg9YzJi4wfii907FYNEqARXlhbjhU1bENLTlaih7p3kywMkzr14/QYA0HR4FpU7sQMHKWEhcR594u2yWa5w0rWJiRhmT1PUiY2uf8v4uxT13qMFB5W1OXaoQFOyLhC2RXVFOS40Nyul3vlrW5tamuCZtPt27jI8NC8bZXDluAy+pgF/ndSEqG3lMAk9IAPDhUm45EqGnhdCApBpKcC6BzyNGDXNl4BHveh5qD0volS4cO8eRTvPqORZjWisqG62KFB7MNRelpLDRXj9kwOIio1DY0MDnpky2ec1jKp+ACjJrCYAyaNT4ThRBWs/m+ZVGCnCc6RODAaAtZ9+gYioaJyprcWXn+xu0ztVUXwcjy7JQXRsHKoryhWxQG9hJ0DnUbpnkpL4q95WcrhIWa/4bXfjZrkImjIj3aczy51Jd4aeF0K6OUK1FzAutQaAkCDgP//ZNQDtN3V17x19/xuBryoavTS++hi1R0ZfQqxuWNhWZY66o7S6S7bTrXlS53CgcO8e7N6+Ffb0DFwtSZreR+pcFHt6hqZppFAnrq2p9lq6XFF8HEcLDiqaOjXl5UgYOgxBVwXDWVen0dUBPPVg8te6ejmFR0ej3uHAR5s3eVQRVS3pgb4Xs4BtWa2Gis7bcukS8DBF5QjRQM8LId2E4bHAjsc8G0Ea8fkPrYq/Am85L75yUPSGjT4P5ueg9rzY0zM0nayFtL/IEVHnlBjpu4h5fbl3j6aNgS8DTRxzvPAQwqOi8dHmTRh3/xRExsQqInKHdu80LD8Xx4+ZdC9MJhNkWcapygpExsQi9aMJiKj/StMiAoDHJ/GRk63J24RcCdDzQsgVSEGFqxEk4AorrX0ACPYSVhp5bWuODOAyZjbFt4rJAdB0nNaj7xOkrhLqbbFAslhgz5hsKPIm9q8sLca4+6doKnr0InpqIT7R/yd/rasqp6KkGCa4EmvP1tcpHaWNEPcgWSw+1XSNjhk6xo4+ERGY/PgcrHo6G/b0DAwYcjP6hEf4/OaXv3Y1JIsFD59Ziv7NX2uNFS9eFiOjkhDiCY0XQroh6rDS4juBJf9kLH4nGHktMPJCFvBuFmQAfXreh/oJeUoTSF/oE1BNAIamjsV1N9yIsGusyvuAtiFic2MTeoaGIuOx2QgxmzX7iX31pdOiaaXIYclfu1oJ4UjhEa7O0irUeSuiIkqdjOwLYdgs3bIdfSIiEGI2Kx6gE2XfoeRwkeIJUtNaHVYEnNJ5ZQysnSMn6GEhpKMwbETIFYa3HBkP3J8OFy4BL+zwrh+iT1R9+o31CI+KwrFDBagpL9d4VEQoxnGiSgnDCM+LXo5fKN2qFW6N3gO8l36/sGkLeoaGoqaiHI+NGdWhdVKHrVLTM5QPS+Gd+iR/G155Igv/lyXhruuc7f4kbZGBN5h0S4gHXSJs1L9/fyxevBh33HEH+vbti6qqKmzYsAEvvvgiLly44PU4s9mM3NxcTJkyBWazGR988AFmzZqFmpqayzVVQq4o1OW1PsNL7odxcJBL3fe5u1o3nXQC9/yHK1SlbyxosVpxqrLCsFWBvhw5ZsBAVJeXI2bAQKSmZ2iaNuoF6Iw6ZgtS7GlKaXJF8XGXdkrSjTD37InGhgZsXP4y2kKd87N7+1ZX2bU7UVjk0RzOicFN71YCAO6VgdyXAZjcH7I+vgrWn3ete0FFm9MghLSDy+Z5ueuuu/Dggw/inXfeQXFxMQYPHow33ngDb731FhYsWOD1uFWrVmHixImYMWMG6uvrsWLFCrS0tGD06NHtui49L4T8dKalAGsecFUq/RR+cAbhLxF/aVefH00PIbPZUFSvPX2D1L2A9riNjsiYWADAqcoKn32J9PMZM+leAID9f3+F8MbvXRv0zQ4FXpogAm17qwghnnTk+f2Lho3mz5+Pxx9/HPHx8Ybb+/Tpg1OnTmHq1KnYssXl+k1MTMTRo0cxatQofPrpp21eg8YLIZ3Hp3Nc+TAdxuCTpekisOhALOS0hYoHReS/tNX80Ai96u6FpiaUHvkKCUOH4fzZc7gqJATvvvYq/vbveZr9hSE0LQV4437ArPY/e8sL0hkqQp1fwMogQn4+XdZ4ycnJwYQJEzBy5EjD7Wlpafj4449hsVhQX1+vvF9WVoa8vDzk5eV5HBMSEgKzubWtqiRJqKyspPFCyGVC067ACB/lv2KbDLjUY8V7vj6F9OczwbPM2Bfq63ibl6/rqva91AI8/F/UXCHkctAlcl70xMfHY86cOZg/f77Xffr27YumpiaN4QIA1dXV6Nu3r+ExTz31FJYsWdKZUyWE+CDHSzhESQT2IWkvtplk3Xve0BsoanU8I0PGKMQj9vE1LwM+/pby+4R0VTpsvCxduhQLFy70uU9SUhKOHTum/G6z2fD+++9j8+bNWLNmTcdn2cZ8li9frvwuPC+EkF+Wth70p3PcAnqyO+zSHs+JkZCb2gDRvxod68NQYn8gQgKTDhsvubm5WL9+vc99SktLlZ/79euHnTt3Yv/+/Zg5c6bP406ePAmz2YywsDCN9yU6OhonT540PKa5uRnNzc3tvwFCiF8QAnqEENIZyJdr2Gw2+dixY/LGjRvlHj16tLl/nz595KamJvm+++5T3ktISJBlWZZvvfXWdl1TkiRZlmVZkqTLdl8cHBwcHBwcnTs6+Py+PJOw2Wzyt99+K//973+XbTabHB0drQz1Pt988408cuRI5b1Vq1bJZWVl8tixY+WUlBR537598r59+y7XzXNwcHBwcHB0gdGR5/dlS9gdP348Bg0ahEGDBnnkoIimZsHBwUhKSkJoaKiybe7cuWhpacGWLVs0InWEEEIIIQDbAxBCCCGkC9CR57c3pQZCCCGEkC4JjRdCCCGEBBQ0XgghhBASUNB4IYQQQkhAQeOFEEIIIQEFjRdCCCGEBBQ0XgghhBASUNB4IYQQQkhAQeOFEEIIIQHFZWsP4G8kSfL3FAghhBDSTjry3O52xou4eX0/JUIIIYR0fSRJarM9QLfrbQQANpuNfY1+IpIkobKyEjExMVzDToDr2blwPTsXrmfnwvX8+UiShKqqqjb363aeFwDtunHiG6fTyf98nQjXs3PhenYuXM/Ohev502nvujFhlxBCCCEBBY0XQgghhAQUNF6IhqamJixZsgRNTU3+nkq3gOvZuXA9OxeuZ+fC9fzl6JYJu4QQQgjpvtDzQgghhJCAgsYLIYQQQgIKGi+EEEIICShovBBCCCEkoKDxQrzSv39/rFmzBqWlpWhoaEBxcTGWLFmC4OBgf08tYHn66aexb98+nDt3DqdPn/b3dAKOWbNm4bvvvsP58+dx4MABjBw50t9TCkjGjBmD/Px8VFZWQpZlpKen+3tKAc3ChQvx2Wef4cyZM6iursbWrVuRkJDg72l1a2i8EK8kJSWhR48eeOyxx3DTTTdh7ty5+N3vfoeXXnrJ31MLWEJCQrB582a89tpr/p5KwPHAAw9g+fLleO6555CSkoLCwkJ88MEHiIyM9PfUAo5evXqhsLAQs2fP9vdUugV2ux0rV67EqFGjMH78eAQHB+PDDz9EaGiov6fWrZE5ONo75s+fL5eUlPh9HoE+MjMz5dOnT/t9HoE0Dhw4IL/66qvK7yaTSa6oqJCffPJJv88tkIcsy3J6errf59GdhtVqlWVZlseMGeP3uXTXQc8L6RBhYWGora319zTIFUZwcDCGDx+OHTt2KO/JsowdO3bgtttu8+PMCPEkLCwMAPhZeRmh8ULaTXx8PObMmYPXX3/d31MhVxhWqxVXXXUVqqurNe9XV1ejb9++fpoVIZ6YTCbk5eVh7969+Prrr/09nW4LjZcrkKVLl0KWZZ8jMTFRc4zNZsP777+PzZs3Y82aNX6aedfkp6wnIaR7snLlSgwePBhTpkzx91S6NVf5ewLklyc3Nxfr16/3uU9paanyc79+/bBz507s378fM2fOvMyzCzw6up6k4zgcDly8eBHR0dGa96Ojo3Hy5Ek/zYoQLa+++iruvvtupKamorKy0t/T6dbQeLkCcTgccDgc7drXZrNh586dKCgowMMPPwxZli/z7AKPjqwn+WlcuHABBQUFGDduHLZv3w7A5Z4fN24cVqxY4efZEeIyXDIyMjB27FiUlZX5ezrdHhovxCs2mw27du3C999/j/nz52tKUvW5B6R9xMXFISIiAtdeey2CgoKQnJwMACguLsa5c+f8PLuuzfLly/Hmm2/i4MGD+Oyzz/DEE0+gV69eWLdunb+nFnD06tULAwcOVH6//vrrkZycjNraWpSXl/txZoHJypUrMXXqVKSnp8PpdCoewvr6ejQ2Nvp5dt0Xv5c8cXTNkZmZKXvD33ML1LFu3TrD9bTb7X6fWyCM2bNny2VlZXJjY6N84MAB+ZZbbvH7nAJx2O12w7/DdevW+X1ugTi8kZmZ6fe5dddhcv9ACCGEEBIQsNqIEEIIIQEFjRdCCCGEBBQ0XgghhBASUNB4IYQQQkhAQeOFEEIIIQEFjRdCCCGEBBQ0XgghhBASUNB4IYQQQkhAQeOFEEIIIQEFjRdCCCGEBBQ0XgghhBASUNB4IYQQQkhA8f9xP15sns4+lAAAAABJRU5ErkJggg==", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "in_disk = {\"name\": \"in_disk\", \"center_x\": 0.0, \"center_y\": 0.0, \"radius\": 1.0}\n", "out_disk = {\"name\": \"out_disk\", \"center_x\": 0.0, \"center_y\": 0.0, \"radius\": 2.0}\n", "domain = {\"size\": 8192, \"random_sampling\": True, \"sampler\": \"uniform\"}\n", "BC = {\"size\": 8192, \"random_sampling\": True, \"sampler\": \"uniform\", \"with_normal\": True}\n", "data = {\"domain\": domain, \"BC\": BC}\n", "config = {\"in_disk\": in_disk, \"out_disk\": out_disk, \"data\": data}\n", "\n", "# create training dataset\n", "dataset = create_training_dataset(config)\n", "train_dataset = dataset.batch(batch_size=8192)\n", "\n", "# create test dataset\n", "inputs, label = create_test_dataset(config)" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "## Model Construction\n", "\n", "This example uses [MultiScaleFCCell](https://mindspore.cn/mindflow/docs/en/r0.1/cell/mindflow.cell.MultiScaleFCSequential.html) to construct a simple fully-connected network with a depth of 6 layers and the activation function is the `tanh` function. `MultiScaleFCCell` is imported from `MindSpore Flow` [cell](https://mindspore.cn/mindflow/docs/zh-CN/r0.1/mindflow.cell.html) module." ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "from mindflow.cell import MultiScaleFCCell\n", "\n", "model = MultiScaleFCCell(in_channels=2,\n", " out_channels=1,\n", " layers=6,\n", " neurons=128,\n", " residual=False,\n", " act=\"tanh\",\n", " num_scales=1)" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "## Optimizer\n", "\n", "Using Adaptive Moment Estimation (Adam) optimizer." ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [], "source": [ "optimizer = nn.Adam(model.trainable_params(), 0.001)\n" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "## Poisson2D\n", "\n", "The following `Poisson2D` includes the governing equations, Dirichlet boundary conditions, Norman boundary conditions, etc. The `sympy` is used for delineating partial differential equations in symbolic forms and computing all equations' loss." ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "### Symbol Declaration\n", "\n", "Define `x`, `y`, and `n` to indicate the the horizontal coordinate, vertical coordinate, and normal vectors of inner circle boundary, respectively. The output `u` is a function related to `x` and `y`." ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "independent variables: [x, y]\n", "dependent variables: [u(x, y)]\n" ] } ], "source": [ "x, y, n = symbols('x y n')\n", "u = Function('u')(x, y)\n", "\n", "# independent variables\n", "in_vars = [x, y]\n", "print(\"independent variables: \", in_vars)\n", "\n", "# dependent variables\n", "out_vars = [u]\n", "print(\"dependent variables: \", out_vars)" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "### Governing Equations" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "governing equation: Derivative(u(x, y), (x, 2)) + Derivative(u(x, y), (y, 2)) + 1.0\n" ] } ], "source": [ "govern_eq = diff(u, (x, 2)) + diff(u, (y, 2)) + 1.0\n", "print(\"governing equation: \", govern_eq)" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "### Dirichlet Boundary Condition" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "bc_outer equation: u(x, y)\n" ] } ], "source": [ "bc_outer = u\n", "print(\"bc_outer equation: \", bc_outer)" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "### Neumann Boundary Condition" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "bc_inner equation: Derivative(u(x, y), n) - 0.5\n" ] } ], "source": [ "bc_inner = sympy.Derivative(u, n) - 0.5\n", "print(\"bc_inner equation: \", bc_inner)" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "The following `Poisson2D` problem is defined based on the [Poisson](https://mindspore.cn/mindflow/docs/en/r0.1/pde/mindflow.pde.Poisson.html#mindflow.pde.Poisson) base class combined with the governing equations and boundary conditions defined above. Download `Poisson2D` [Python script](https://gitee.com/mindspore/mindscience/blob/r0.3/MindFlow/features/solve_pinns_by_mindflow/src/model.py)." ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "poisson: Derivative(u(x, y), (x, 2)) + Derivative(u(x, y), (y, 2)) + 1.0\n", " Item numbers of current derivative formula nodes: 3\n", "bc_outer: u(x, y)\n", " Item numbers of current derivative formula nodes: 1\n", "bc_inner: Derivative(u(x, y), n) - 0.5\n", " Item numbers of current derivative formula nodes: 2\n" ] } ], "source": [ "from mindflow.pde import Poisson, sympy_to_mindspore\n", "\n", "class Poisson2D(Poisson):\n", " def __init__(self, model, loss_fn=\"mse\"):\n", " super(Poisson2D, self).__init__(model, loss_fn=loss_fn)\n", " self.bc_outer_nodes = sympy_to_mindspore(self.bc_outer(), self.in_vars, self.out_vars)\n", " self.bc_inner_nodes = sympy_to_mindspore(self.bc_inner(), self.in_vars, self.out_vars)\n", "\n", " def bc_outer(self):\n", " bc_outer_eq = self.u\n", " equations = {\"bc_outer\": bc_outer_eq}\n", " return equations\n", "\n", " def bc_inner(self):\n", " bc_inner_eq = sympy.Derivative(self.u, self.normal) - 0.5\n", " equations = {\"bc_inner\": bc_inner_eq}\n", " return equations\n", "\n", " def get_loss(self, pde_data, bc_outer_data, bc_inner_data, bc_inner_normal):\n", " pde_res = self.parse_node(self.pde_nodes, inputs=pde_data)\n", " pde_loss = self.loss_fn(pde_res[0], Tensor(np.array([0.0]), mstype.float32))\n", "\n", " bc_inner_res = self.parse_node(self.bc_inner_nodes, inputs=bc_inner_data, norm=bc_inner_normal)\n", " bc_inner_loss = self.loss_fn(bc_inner_res[0], Tensor(np.array([0.0]), mstype.float32))\n", "\n", " bc_outer_res = self.parse_node(self.bc_outer_nodes, inputs=bc_outer_data)\n", " bc_outer_loss = self.loss_fn(bc_outer_res[0], Tensor(np.array([0.0]), mstype.float32))\n", "\n", " return pde_loss + bc_inner_loss + bc_outer_loss\n", "\n", "problem = Poisson2D(model)" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "## Model Training\n", "\n", "With **MindSpore version >= 2.0.0**, we can use the functional programming for training neural networks. Download training [Python script](https://gitee.com/mindspore/mindscience/blob/r0.3/MindFlow/features/solve_pinns_by_mindflow/train.py)." ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [], "source": [ "# define forward function\n", "def forward_fn(pde_data, bc_outer_data, bc_inner_data, bc_inner_normal):\n", " loss = problem.get_loss(pde_data, bc_outer_data, bc_inner_data, bc_inner_normal)\n", " return loss\n", "\n", "# define grad function\n", "grad_fn = ms.value_and_grad(forward_fn, None, optimizer.parameters, has_aux=False)\n", "\n", "# using jit to accelerate training\n", "@jit\n", "def train_step(pde_data, bc_outer_data, bc_inner_data, bc_inner_normal):\n", " loss, grads = grad_fn(pde_data, bc_outer_data, bc_inner_data, bc_inner_normal)\n", " loss = ops.depend(loss, optimizer(grads))\n", " return loss\n" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "Download the calculate function of training process in [Python script](https://gitee.com/mindspore/mindscience/blob/r0.3/MindFlow/features/solve_pinns_by_mindflow/src/utils.py)." ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [], "source": [ "def _calculate_error(label, prediction):\n", " '''calculate l2-error to evaluate accuracy'''\n", " error = label - prediction\n", " l2_error = np.sqrt(np.sum(np.square(error[..., 0]))) / np.sqrt(np.sum(np.square(label[..., 0])))\n", "\n", " return l2_error\n", "\n", "\n", "def _get_prediction(model, inputs, label_shape, batch_size):\n", " '''calculate the prediction respect to the given inputs'''\n", " prediction = np.zeros(label_shape)\n", " prediction = prediction.reshape((-1, label_shape[1]))\n", " inputs = inputs.reshape((-1, inputs.shape[1]))\n", "\n", " time_beg = time.time()\n", "\n", " index = 0\n", " while index < inputs.shape[0]:\n", " index_end = min(index + batch_size, inputs.shape[0])\n", " test_batch = Tensor(inputs[index: index_end, :], mstype.float32)\n", " prediction[index: index_end, :] = model(test_batch).asnumpy()\n", " index = index_end\n", "\n", " print(\" predict total time: {} ms\".format((time.time() - time_beg) * 1000))\n", " prediction = prediction.reshape(label_shape)\n", " prediction = prediction.reshape((-1, label_shape[1]))\n", " return prediction\n", "\n", "\n", "def calculate_l2_error(model, inputs, label, batch_size):\n", " label_shape = label.shape\n", " prediction = _get_prediction(model, inputs, label_shape, batch_size)\n", " label = label.reshape((-1, label_shape[1]))\n", " l2_error = _calculate_error(label, prediction)\n", " print(\" l2_error: \", l2_error)\n", " print(\"==================================================================================================\")\n" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "epoch: 1 train loss: 1.2577767 epoch time: 6024.162 ms\n", "epoch: 2 train loss: 1.2554792 epoch time: 70.884 ms\n", "epoch: 3 train loss: 1.2534575 epoch time: 71.048 ms\n", "epoch: 4 train loss: 1.2516733 epoch time: 100.632 ms\n", "epoch: 5 train loss: 1.2503157 epoch time: 65.656 ms\n", "epoch: 6 train loss: 1.2501826 epoch time: 137.487 ms\n", "epoch: 7 train loss: 1.2511331 epoch time: 51.191 ms\n", "epoch: 8 train loss: 1.2508672 epoch time: 65.980 ms\n", "epoch: 9 train loss: 1.2503275 epoch time: 211.144 ms\n", "epoch: 10 train loss: 1.2500556 epoch time: 224.515 ms\n", "epoch: 11 train loss: 1.2500004 epoch time: 225.964 ms\n", "epoch: 12 train loss: 1.2500298 epoch time: 220.117 ms\n", "epoch: 13 train loss: 1.2500703 epoch time: 221.441 ms\n", "epoch: 14 train loss: 1.2500948 epoch time: 220.214 ms\n", "epoch: 15 train loss: 1.2500978 epoch time: 219.836 ms\n", "epoch: 16 train loss: 1.250083 epoch time: 220.141 ms\n", "epoch: 17 train loss: 1.2500567 epoch time: 229.682 ms\n", "epoch: 18 train loss: 1.2500263 epoch time: 216.013 ms\n", "epoch: 19 train loss: 1.2500005 epoch time: 218.639 ms\n", "epoch: 20 train loss: 1.2499874 epoch time: 228.765 ms\n", "epoch: 21 train loss: 1.2499917 epoch time: 230.179 ms\n", "epoch: 22 train loss: 1.250007 epoch time: 215.476 ms\n", "epoch: 23 train loss: 1.250018 epoch time: 224.334 ms\n", "epoch: 24 train loss: 1.2500123 epoch time: 217.322 ms\n", "epoch: 25 train loss: 1.249995 epoch time: 217.106 ms\n", "epoch: 26 train loss: 1.249982 epoch time: 217.612 ms\n", "epoch: 27 train loss: 1.249983 epoch time: 223.639 ms\n", "epoch: 28 train loss: 1.2499921 epoch time: 226.419 ms\n", "epoch: 29 train loss: 1.2499963 epoch time: 212.248 ms\n", "epoch: 30 train loss: 1.2499878 epoch time: 225.295 ms\n", "epoch: 31 train loss: 1.2499707 epoch time: 219.656 ms\n", "epoch: 32 train loss: 1.2499548 epoch time: 218.019 ms\n", "epoch: 33 train loss: 1.2499464 epoch time: 226.645 ms\n", "epoch: 34 train loss: 1.2499409 epoch time: 222.399 ms\n", "epoch: 35 train loss: 1.249928 epoch time: 224.290 ms\n", "epoch: 36 train loss: 1.2499026 epoch time: 221.664 ms\n", "epoch: 37 train loss: 1.2498704 epoch time: 221.300 ms\n", "epoch: 38 train loss: 1.2498367 epoch time: 220.392 ms\n", "epoch: 39 train loss: 1.2497979 epoch time: 221.848 ms\n", "epoch: 40 train loss: 1.2497431 epoch time: 219.831 ms\n", "epoch: 41 train loss: 1.2496608 epoch time: 217.333 ms\n", "epoch: 42 train loss: 1.249544 epoch time: 220.116 ms\n", "epoch: 43 train loss: 1.2493837 epoch time: 214.985 ms\n", "epoch: 44 train loss: 1.2491598 epoch time: 220.717 ms\n", "epoch: 45 train loss: 1.248828 epoch time: 216.047 ms\n", "epoch: 46 train loss: 1.2483226 epoch time: 218.554 ms\n", "epoch: 47 train loss: 1.247554 epoch time: 221.158 ms\n", "epoch: 48 train loss: 1.2463655 epoch time: 218.594 ms\n", "epoch: 49 train loss: 1.2444699 epoch time: 216.152 ms\n", "epoch: 50 train loss: 1.2413855 epoch time: 220.306 ms\n", "epoch: 51 train loss: 1.2362938 epoch time: 211.876 ms\n", "epoch: 52 train loss: 1.2277732 epoch time: 213.732 ms\n", "epoch: 53 train loss: 1.2135327 epoch time: 219.945 ms\n", "epoch: 54 train loss: 1.1906419 epoch time: 217.820 ms\n", "epoch: 55 train loss: 1.1573513 epoch time: 225.694 ms\n", "epoch: 56 train loss: 1.1058999 epoch time: 221.004 ms\n", "epoch: 57 train loss: 1.0343707 epoch time: 225.404 ms\n", "epoch: 58 train loss: 0.9365865 epoch time: 224.163 ms\n", "epoch: 59 train loss: 0.83171475 epoch time: 211.383 ms\n", "epoch: 60 train loss: 0.77913564 epoch time: 229.284 ms\n", "epoch: 61 train loss: 0.74204475 epoch time: 223.518 ms\n", "epoch: 62 train loss: 0.80121577 epoch time: 229.029 ms\n", "epoch: 63 train loss: 0.8549291 epoch time: 223.576 ms\n", "epoch: 64 train loss: 0.7383551 epoch time: 235.727 ms\n", "epoch: 65 train loss: 0.72710323 epoch time: 222.646 ms\n", "epoch: 66 train loss: 0.6702794 epoch time: 226.154 ms\n", "epoch: 67 train loss: 0.6987355 epoch time: 221.565 ms\n", "epoch: 68 train loss: 0.6746455 epoch time: 234.406 ms\n", "epoch: 69 train loss: 0.70462525 epoch time: 226.131 ms\n", "epoch: 70 train loss: 0.67767555 epoch time: 229.177 ms\n", "epoch: 71 train loss: 0.6821881 epoch time: 224.217 ms\n", "epoch: 72 train loss: 0.64521456 epoch time: 223.717 ms\n", "epoch: 73 train loss: 0.6368966 epoch time: 226.217 ms\n", "epoch: 74 train loss: 0.592155 epoch time: 219.816 ms\n", "epoch: 75 train loss: 0.6024764 epoch time: 214.097 ms\n", "epoch: 76 train loss: 0.58170027 epoch time: 224.460 ms\n", "epoch: 77 train loss: 0.5691892 epoch time: 223.964 ms\n", "epoch: 78 train loss: 0.5925416 epoch time: 226.897 ms\n", "epoch: 79 train loss: 0.61034954 epoch time: 221.710 ms\n", "epoch: 80 train loss: 0.5831032 epoch time: 231.325 ms\n", "epoch: 81 train loss: 0.5364084 epoch time: 222.517 ms\n", "epoch: 82 train loss: 0.5502083 epoch time: 215.709 ms\n", "epoch: 83 train loss: 0.5633007 epoch time: 209.054 ms\n", "epoch: 84 train loss: 0.52546465 epoch time: 219.471 ms\n", "epoch: 85 train loss: 0.53276706 epoch time: 218.961 ms\n", "epoch: 86 train loss: 0.55396163 epoch time: 237.759 ms\n", "epoch: 87 train loss: 0.5206229 epoch time: 219.588 ms\n", "epoch: 88 train loss: 0.5106571 epoch time: 225.651 ms\n", "epoch: 89 train loss: 0.53332406 epoch time: 224.282 ms\n", "epoch: 90 train loss: 0.53076947 epoch time: 235.400 ms\n", "epoch: 91 train loss: 0.5049336 epoch time: 215.371 ms\n", "epoch: 92 train loss: 0.48215953 epoch time: 239.344 ms\n", "epoch: 93 train loss: 0.4843874 epoch time: 218.674 ms\n", "epoch: 94 train loss: 0.51292086 epoch time: 221.709 ms\n", "epoch: 95 train loss: 0.56979203 epoch time: 225.018 ms\n", "epoch: 96 train loss: 0.61994594 epoch time: 219.800 ms\n", "epoch: 97 train loss: 0.4962491 epoch time: 223.785 ms\n", "epoch: 98 train loss: 0.4802659 epoch time: 230.708 ms\n", "epoch: 99 train loss: 0.54967964 epoch time: 221.209 ms\n", "epoch: 100 train loss: 0.46414006 epoch time: 223.953 ms\n", " predict total time: 124.87483024597168 ms\n", " l2_error: 0.9584533008207833\n", "==================================================================================================\n", "...\n", "epoch: 4901 train loss: 0.00012433846 epoch time: 241.115 ms\n", "epoch: 4902 train loss: 0.00012422525 epoch time: 239.142 ms\n", "epoch: 4903 train loss: 0.00012412701 epoch time: 234.900 ms\n", "epoch: 4904 train loss: 0.00012404467 epoch time: 237.946 ms\n", "epoch: 4905 train loss: 0.000123965 epoch time: 236.818 ms\n", "epoch: 4906 train loss: 0.0001238766 epoch time: 255.728 ms\n", "epoch: 4907 train loss: 0.00012378026 epoch time: 225.175 ms\n", "epoch: 4908 train loss: 0.00012368544 epoch time: 241.107 ms\n", "epoch: 4909 train loss: 0.00012359957 epoch time: 248.310 ms\n", "epoch: 4910 train loss: 0.00012352059 epoch time: 239.238 ms\n", "epoch: 4911 train loss: 0.0001234413 epoch time: 229.464 ms\n", "epoch: 4912 train loss: 0.00012335769 epoch time: 228.504 ms\n", "epoch: 4913 train loss: 0.00012327175 epoch time: 238.126 ms\n", "epoch: 4914 train loss: 0.00012318943 epoch time: 236.290 ms\n", "epoch: 4915 train loss: 0.00012311469 epoch time: 221.079 ms\n", "epoch: 4916 train loss: 0.00012304715 epoch time: 238.825 ms\n", "epoch: 4917 train loss: 0.00012298509 epoch time: 243.784 ms\n", "epoch: 4918 train loss: 0.00012292914 epoch time: 235.416 ms\n", "epoch: 4919 train loss: 0.00012288446 epoch time: 221.510 ms\n", "epoch: 4920 train loss: 0.00012286083 epoch time: 244.597 ms\n", "epoch: 4921 train loss: 0.00012286832 epoch time: 245.989 ms\n", "epoch: 4922 train loss: 0.00012292268 epoch time: 234.209 ms\n", "epoch: 4923 train loss: 0.00012304373 epoch time: 223.089 ms\n", "epoch: 4924 train loss: 0.0001232689 epoch time: 234.764 ms\n", "epoch: 4925 train loss: 0.00012365196 epoch time: 246.427 ms\n", "epoch: 4926 train loss: 0.00012429312 epoch time: 233.304 ms\n", "epoch: 4927 train loss: 0.0001253246 epoch time: 221.859 ms\n", "epoch: 4928 train loss: 0.00012700919 epoch time: 230.224 ms\n", "epoch: 4929 train loss: 0.00012967733 epoch time: 254.334 ms\n", "epoch: 4930 train loss: 0.000134056 epoch time: 242.256 ms\n", "epoch: 4931 train loss: 0.00014098533 epoch time: 225.075 ms\n", "epoch: 4932 train loss: 0.00015257315 epoch time: 235.392 ms\n", "epoch: 4933 train loss: 0.00017092828 epoch time: 243.934 ms\n", "epoch: 4934 train loss: 0.00020245541 epoch time: 244.327 ms\n", "epoch: 4935 train loss: 0.00025212576 epoch time: 238.360 ms\n", "epoch: 4936 train loss: 0.00034049098 epoch time: 233.573 ms\n", "epoch: 4937 train loss: 0.0004768648 epoch time: 242.150 ms\n", "epoch: 4938 train loss: 0.0007306886 epoch time: 247.166 ms\n", "epoch: 4939 train loss: 0.0011023732 epoch time: 242.486 ms\n", "epoch: 4940 train loss: 0.001834287 epoch time: 237.257 ms\n", "epoch: 4941 train loss: 0.0027812878 epoch time: 242.192 ms\n", "epoch: 4942 train loss: 0.004766387 epoch time: 236.694 ms\n", "epoch: 4943 train loss: 0.006648433 epoch time: 223.730 ms\n", "epoch: 4944 train loss: 0.010807082 epoch time: 237.647 ms\n", "epoch: 4945 train loss: 0.01206318 epoch time: 233.036 ms\n", "epoch: 4946 train loss: 0.015489452 epoch time: 227.594 ms\n", "epoch: 4947 train loss: 0.011565584 epoch time: 227.099 ms\n", "epoch: 4948 train loss: 0.008471975 epoch time: 231.612 ms\n", "epoch: 4949 train loss: 0.0035123175 epoch time: 276.854 ms\n", "epoch: 4950 train loss: 0.00091841083 epoch time: 229.893 ms\n", "epoch: 4951 train loss: 0.00053060305 epoch time: 231.533 ms\n", "epoch: 4952 train loss: 0.0017638807 epoch time: 231.814 ms\n", "epoch: 4953 train loss: 0.0035763814 epoch time: 236.456 ms\n", "epoch: 4954 train loss: 0.004056363 epoch time: 244.743 ms\n", "epoch: 4955 train loss: 0.0039708405 epoch time: 221.469 ms\n", "epoch: 4956 train loss: 0.0027319128 epoch time: 236.732 ms\n", "epoch: 4957 train loss: 0.0017904624 epoch time: 231.612 ms\n", "epoch: 4958 train loss: 0.0009970744 epoch time: 244.820 ms\n", "epoch: 4959 train loss: 0.00061692565 epoch time: 221.442 ms\n", "epoch: 4960 train loss: 0.0007383662 epoch time: 245.430 ms\n", "epoch: 4961 train loss: 0.0012403185 epoch time: 231.007 ms\n", "epoch: 4962 train loss: 0.0018439001 epoch time: 233.718 ms\n", "epoch: 4963 train loss: 0.0017326038 epoch time: 224.514 ms\n", "epoch: 4964 train loss: 0.0011022249 epoch time: 237.367 ms\n", "epoch: 4965 train loss: 0.00033718432 epoch time: 242.038 ms\n", "epoch: 4966 train loss: 0.00018465787 epoch time: 230.688 ms\n", "epoch: 4967 train loss: 0.00055812683 epoch time: 218.956 ms\n", "epoch: 4968 train loss: 0.00085373345 epoch time: 239.294 ms\n", "epoch: 4969 train loss: 0.0007744754 epoch time: 234.656 ms\n", "epoch: 4970 train loss: 0.00047988302 epoch time: 243.434 ms\n", "epoch: 4971 train loss: 0.0003720247 epoch time: 214.474 ms\n", "epoch: 4972 train loss: 0.0004015266 epoch time: 239.108 ms\n", "epoch: 4973 train loss: 0.00037753512 epoch time: 246.952 ms\n", "epoch: 4974 train loss: 0.0002867286 epoch time: 235.668 ms\n", "epoch: 4975 train loss: 0.00029258506 epoch time: 229.418 ms\n", "epoch: 4976 train loss: 0.00041505875 epoch time: 239.248 ms\n", "epoch: 4977 train loss: 0.00043451862 epoch time: 235.341 ms\n", "epoch: 4978 train loss: 0.00030042438 epoch time: 236.248 ms\n", "epoch: 4979 train loss: 0.00015146247 epoch time: 218.208 ms\n", "epoch: 4980 train loss: 0.00016080803 epoch time: 246.693 ms\n", "epoch: 4981 train loss: 0.00027215388 epoch time: 238.106 ms\n", "epoch: 4982 train loss: 0.00031257677 epoch time: 236.889 ms\n", "epoch: 4983 train loss: 0.0002639915 epoch time: 222.054 ms\n", "epoch: 4984 train loss: 0.000204898 epoch time: 231.757 ms\n", "epoch: 4985 train loss: 0.00019457101 epoch time: 246.356 ms\n", "epoch: 4986 train loss: 0.00018954299 epoch time: 247.128 ms\n", "epoch: 4987 train loss: 0.00016800698 epoch time: 219.524 ms\n", "epoch: 4988 train loss: 0.00016529267 epoch time: 240.068 ms\n", "epoch: 4989 train loss: 0.00019697993 epoch time: 235.206 ms\n", "epoch: 4990 train loss: 0.00021988692 epoch time: 237.431 ms\n", "epoch: 4991 train loss: 0.00019355604 epoch time: 219.045 ms\n", "epoch: 4992 train loss: 0.00014973793 epoch time: 246.853 ms\n", "epoch: 4993 train loss: 0.00013542885 epoch time: 226.068 ms\n", "epoch: 4994 train loss: 0.00015176873 epoch time: 248.399 ms\n", "epoch: 4995 train loss: 0.0001647438 epoch time: 217.880 ms\n", "epoch: 4996 train loss: 0.00016070419 epoch time: 237.407 ms\n", "epoch: 4997 train loss: 0.00015653059 epoch time: 235.727 ms\n", "epoch: 4998 train loss: 0.00015907965 epoch time: 247.844 ms\n", "epoch: 4999 train loss: 0.00015674165 epoch time: 226.864 ms\n", "epoch: 5000 train loss: 0.00014248541 epoch time: 244.331 ms\n", " predict total time: 1.7328262329101562 ms\n", " l2_error: 0.00915682980750216\n", "==================================================================================================\n" ] } ], "source": [ "epochs = 5000\n", "steps_per_epochs = train_dataset.get_dataset_size()\n", "sink_process = ms.data_sink(train_step, train_dataset, sink_size=1)\n", "\n", "for epoch in range(1, epochs + 1):\n", " # train\n", " time_beg = time.time()\n", " model.set_train(True)\n", " for _ in range(steps_per_epochs):\n", " step_train_loss = sink_process()\n", " print(f\"epoch: {epoch} train loss: {step_train_loss} epoch time: {(time.time() - time_beg)*1000 :.3f} ms\")\n", " model.set_train(False)\n", " if epoch % 100 == 0:\n", " # eval\n", " calculate_l2_error(model, inputs, label, 8192)" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "## Model Evaluation and Visualization\n", "\n", "After training, all data points in the flow field can be inferred. And related results can be visualized. Download visulization [Python script](https://gitee.com/mindspore/mindscience/blob/r0.3/MindFlow/features/solve_pinns_by_mindflow/src/utils.py)." ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [], "source": [ "def visual(model, inputs, label, epochs=1):\n", " '''visual result for poisson 2D'''\n", " fig, ax = plt.subplots(2, 1)\n", " ax = ax.flatten()\n", " plt.subplots_adjust(hspace=0.5)\n", " ax0 = ax[0].scatter(inputs[:, 0], inputs[:, 1], c=label[:, 0], cmap=plt.cm.rainbow, s=0.5)\n", " ax[0].set_title(\"true\")\n", " ax[0].set_xlabel('x')\n", " ax[0].set_ylabel('y')\n", " ax[0].axis('equal')\n", " ax[1].scatter(inputs[:, 0], inputs[:, 1], c=model(Tensor(inputs, mstype.float32)), cmap=plt.cm.rainbow, s=0.5)\n", " ax[1].set_title(\"prediction\")\n", " ax[1].set_xlabel('x')\n", " ax[1].set_ylabel('y')\n", " ax[1].axis('equal')\n", " cbar = fig.colorbar(ax0, ax=[ax[0], ax[1]])\n", " cbar.set_label('u(x, y)')\n", "\n", " plt.savefig(f\"images/{epochs}-result.jpg\", dpi=600)\n" ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# visualization\n", "visual(model, inputs, label, 5000)" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3.9.0 ('py39')", "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": "3.9.0" }, "orig_nbformat": 4, "vscode": { "interpreter": { "hash": "57ace93c29d9374277a79956c3f1b916d7d9a05468d906842f9921d0d494a29f" } } }, "nbformat": 4, "nbformat_minor": 2 }