{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "# Inequality constraints with *refnx*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Simple equality constraints can use the mechanisms outlined in this notebook, but are better expressed using the `Parameter.constraint` mechanism, or by sharing `Parameter` objects. It is sometimes also possible to implement different parameterisation of the model to use physically relevant values.\n",
    "\n",
    "The following processes can be used to make inequality constraints with *refnx*. The dataset is reflectivity from a clean silicon wafer with a native oxide layer."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "from importlib import resources\n",
    "import numpy as np\n",
    "\n",
    "import refnx\n",
    "from refnx.dataset import ReflectDataset\n",
    "from refnx.reflect import SLD, MaterialSLD, ReflectModel\n",
    "from refnx.analysis import Objective, CurveFitter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "pth = resources.files(refnx.dataset)\n",
    "DATASET_NAME = 'c_PLP0000708.dat'\n",
    "file_path = pth / f\"tests/{DATASET_NAME}\"\n",
    "\n",
    "data = ReflectDataset(file_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "air = SLD(0)\n",
    "sio2 = MaterialSLD('SiO2', 2.2)\n",
    "si = MaterialSLD('Si', 2.33)\n",
    "s = air | sio2(15, 3) | si(0, 3)\n",
    "\n",
    "model = ReflectModel(s, bkg=3e-8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "# model.bkg.setp(vary=True, bounds=(0, 1e-6))\n",
    "# model.scale.setp(vary=True, bounds=(0.9, 1.1))\n",
    "\n",
    "# sio2 layer\n",
    "s[1].rough.setp(vary=True, bounds=(0, 10))\n",
    "s[1].thick.setp(vary=True, bounds=(0, 20))\n",
    "\n",
    "# si/sio2 interface\n",
    "s[-1].rough.setp(vary=True, bounds=(0, 10))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "objective = Objective(model, data)\n",
    "fitter = CurveFitter(objective)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "-22.303297934635626: : 23it [00:00, 76.34it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "________________________________________________________________________________\n",
      "Structure:                \n",
      "solvent: None\n",
      "reverse structure: False\n",
      "contract: 0\n",
      "\n",
      "________________________________________________________________________________\n",
      "Parameters:       ''       \n",
      "<Parameter:  ' - thick'   , value=0  (fixed) , bounds=[-inf, inf]>\n",
      "________________________________________________________________________________\n",
      "Parameters:       ''       \n",
      "<Parameter:   ' - sld'    , value=0  (fixed) , bounds=[-inf, inf]>\n",
      "<Parameter:   ' - isld'   , value=0  (fixed) , bounds=[-inf, inf]>\n",
      "<Parameter:  ' - rough'   , value=0  (fixed) , bounds=[-inf, inf]>\n",
      "<Parameter:' - volfrac solvent', value=0  (fixed) , bounds=[0.0, 1.0]>\n",
      "________________________________________________________________________________\n",
      "Parameters:       ''       \n",
      "<Parameter:  ' - thick'   , value=10.1028 +/- 0.412, bounds=[0.0, 20.0]>\n",
      "________________________________________________________________________________\n",
      "Parameters:       ''       \n",
      "<Parameter:   'density'   , value=2.2  (fixed) , bounds=[-inf, inf]>\n",
      "<Parameter:  ' - rough'   , value=2.6702 +/- 0.439, bounds=[0.0, 10.0]>\n",
      "<Parameter:' - volfrac solvent', value=0  (fixed) , bounds=[0.0, 1.0]>\n",
      "________________________________________________________________________________\n",
      "Parameters:       ''       \n",
      "<Parameter:  ' - thick'   , value=0  (fixed) , bounds=[-inf, inf]>\n",
      "________________________________________________________________________________\n",
      "Parameters:       ''       \n",
      "<Parameter:   'density'   , value=2.33  (fixed) , bounds=[-inf, inf]>\n",
      "<Parameter:  ' - rough'   , value=5.37419 +/- 0.961, bounds=[0.0, 10.0]>\n",
      "<Parameter:' - volfrac solvent', value=0  (fixed) , bounds=[0.0, 1.0]>\n"
     ]
    }
   ],
   "source": [
    "fitter.fit('differential_evolution', rng=1234)\n",
    "print(s)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Inequality constraints with `differential_evolution`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "*Simple equality constraints can use the following mechanism, but are better expressed using the `Parameter.constraint` mechanism, or by sharing `Parameter` objects. It is sometimes also possible to implement different parameterisation of the model to use physically relevant values.*\n",
    "\n",
    "We see that the thickness of the SiO2 layer is 12.45 and the roughness of the air/SiO2 interface is 4.77. Let's make a constraint that the roughness can't be more than a quarter of the layer thickness. In optimisation such constraints are expressed as inequalities:\n",
    "\n",
    "$$t > 4\\sigma$$\n",
    "\n",
    "We need to rearrange so that all variables are on one side, we do the rearrangement like this so there is no divide by 0:\n",
    "\n",
    "$$t - 4\\sigma > 0$$\n",
    "\n",
    "Now we create a callable object (has the `__call__` magic method) that encodes this inequality. We're going to create the object with the parameters we want to constrain (`pars`), so we can refer to them later. We'll also store the objective because we'll need to update it with the fitting parameters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "class DEC(object):\n",
    "    def __init__(self, pars, objective):\n",
    "        # we'll store the parameters and objective in this object\n",
    "        # this will be necessary for pickling in the future\n",
    "        self.pars = pars\n",
    "        self.objective = objective\n",
    "\n",
    "    def __call__(self, x):\n",
    "        # we need to update the varying parameters in the\n",
    "        # objective first\n",
    "        self.objective.setp(x)\n",
    "        return float(self.pars[0] - 4*self.pars[1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Now lets create an instance of that object, using the parameters we want to constrain. Following that we set up a `scipy.optimize.NonlinearConstraint` for use with `differential_evolution`. Note that we want the constraint calculation to be greater than 0."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "pars = (s[1].thick, s[1].rough)\n",
    "dec = DEC(pars, objective)\n",
    "\n",
    "from scipy.optimize import NonlinearConstraint\n",
    "constraint = NonlinearConstraint(dec, 0, np.inf)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Now do the fit with the added constraint. Note that you can have more than one constraint."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "-22.201646155358837: : 32it [00:00, 102.48it/s]/Users/andrew/miniforge3/envs/dev3/lib/python3.13/site-packages/scipy/optimize/_differentiable_functions.py:552: UserWarning: delta_grad == 0.0. Check if the approximated function is linear. If the function is linear better results can be obtained by defining the Hessian as zero instead of using quasi-Newton approximations.\n",
      "  self.H.update(delta_x, delta_g)\n",
      "/Users/andrew/miniforge3/envs/dev3/lib/python3.13/site-packages/scipy/optimize/_differentiable_functions.py:317: UserWarning: delta_grad == 0.0. Check if the approximated function is linear. If the function is linear better results can be obtained by defining the Hessian as zero instead of using quasi-Newton approximations.\n",
      "  self.H.update(self.x - self.x_prev, self.g - self.g_prev)\n",
      "-22.201646155358837: : 32it [00:00, 60.63it/s] "
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "________________________________________________________________________________\n",
      "Structure:                \n",
      "solvent: None\n",
      "reverse structure: False\n",
      "contract: 0\n",
      "\n",
      "________________________________________________________________________________\n",
      "Parameters:       ''       \n",
      "<Parameter:  ' - thick'   , value=0  (fixed) , bounds=[-inf, inf]>\n",
      "________________________________________________________________________________\n",
      "Parameters:       ''       \n",
      "<Parameter:   ' - sld'    , value=0  (fixed) , bounds=[-inf, inf]>\n",
      "<Parameter:   ' - isld'   , value=0  (fixed) , bounds=[-inf, inf]>\n",
      "<Parameter:  ' - rough'   , value=0  (fixed) , bounds=[-inf, inf]>\n",
      "<Parameter:' - volfrac solvent', value=0  (fixed) , bounds=[0.0, 1.0]>\n",
      "________________________________________________________________________________\n",
      "Parameters:       ''       \n",
      "<Parameter:  ' - thick'   , value=9.9294 +/- 0.387, bounds=[0.0, 20.0]>\n",
      "________________________________________________________________________________\n",
      "Parameters:       ''       \n",
      "<Parameter:   'density'   , value=2.2  (fixed) , bounds=[-inf, inf]>\n",
      "<Parameter:  ' - rough'   , value=2.48235 +/- 0.394, bounds=[0.0, 10.0]>\n",
      "<Parameter:' - volfrac solvent', value=0  (fixed) , bounds=[0.0, 1.0]>\n",
      "________________________________________________________________________________\n",
      "Parameters:       ''       \n",
      "<Parameter:  ' - thick'   , value=0  (fixed) , bounds=[-inf, inf]>\n",
      "________________________________________________________________________________\n",
      "Parameters:       ''       \n",
      "<Parameter:   'density'   , value=2.33  (fixed) , bounds=[-inf, inf]>\n",
      "<Parameter:  ' - rough'   , value=5.76572 +/- 0.889, bounds=[0.0, 10.0]>\n",
      "<Parameter:' - volfrac solvent', value=0  (fixed) , bounds=[0.0, 1.0]>\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "fitter.fit('differential_evolution', constraints=(constraint,), rng=1234)\n",
    "print(s)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Inequality constraints during MCMC sampling"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "If we want to implement that inequality constraint during sampling we can add an extra log-probability term to the `Objective`. This log-probability term will return 0 if the inequality is satisfied, but `-np.inf` if not."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "class LogpExtra(object):\n",
    "    def __init__(self, pars):\n",
    "        # we'll store the parameters and objective in this object\n",
    "        # this will be necessary for pickling in the future\n",
    "        self.pars = pars\n",
    "\n",
    "    def __call__(self, model, data):\n",
    "        if float(self.pars[0] - 4*self.pars[1]) > 0:\n",
    "            return 0\n",
    "        return -np.inf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "lpe = LogpExtra(pars)\n",
    "\n",
    "# set the log_extra attribute of the Objective with our extra log-probability term.\n",
    "objective.logp_extra = lpe"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Lets check what happens to the probabilities with the specified inequality."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<Parameter:  ' - thick'   , value=9.9294 +/- 0.387, bounds=[0.0, 20.0]>\n",
      "<Parameter:  ' - rough'   , value=2.48235 +/- 0.394, bounds=[0.0, 10.0]>\n",
      "14.602783132959756\n",
      "Now exceed the inequality\n",
      "-inf\n"
     ]
    }
   ],
   "source": [
    "print(s[1].thick)\n",
    "print(s[1].rough)\n",
    "print(objective.logpost())\n",
    "\n",
    "print(\"Now exceed the inequality\")\n",
    "s[1].rough.value = 5.\n",
    "print(objective.logpost())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "Now let's MCMC sample the system. There will be a user warning because some walkers have initial starting points which disobey the inequality. Normally one would sample for a far longer time, and thin more appropriately. However, the purpose of the following is to produce a corner plot that demonstrates the inequality - note the sharp dropoff in the probability distribution for the roughness. The roughness doesn't like to go much higher than ~2.5, which is around a quarter of the optimal layer thickness of ~10."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    },
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|                                                                                                                            | 0/200 [00:00<?, ?it/s]/Users/andrew/miniforge3/envs/dev3/lib/python3.13/site-packages/refnx/_lib/emcee/moves/red_blue.py:99: RuntimeWarning: invalid value encountered in scalar subtract\n",
      "  lnpdiff = f + nlp - state.log_prob[j]\n",
      "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 200/200 [00:10<00:00, 19.94it/s]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsoAAALKCAYAAAArlndAAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8hTgPZAAAACXBIWXMAAA9hAAAPYQGoP6dpAADxWklEQVR4nOzde1yT5f8/8Nc2YGzjJGx4BFFTI9PUTEutPJsdzKw8p2alpplpall5LPPYwdIOdrCDZd8sK09llmZZmefyQGoZgqKyIQjshgHbfn/wu+/PgHtjgzG28Xo+Hj6abFzXhXFv7117X++3wm6320FERERERGUoa3sBRERERET+iIEyEREREZEMBspERERERDIYKBMRERERyWCgTEREREQkg4EyEREREZEMBspERERERDIYKBMRERERyQip7QUEE5vNhoyMDERGRkKhUNT2coj8it1uR15eHho1agSlku/RiYjI/zFQ9qKMjAwkJCTU9jKI/Fp6ejqaNGlS28sgIiKqFANlL4qMjARQGghERUXV8mrqFqvVit9++w0A0LVrV6hUKr+ewxfr9Te5ublISEiQrhMiIiJ/p7Db7fbaXkSwyM3NRXR0NK5cucJAmagcXh9ERBRomChIRERERCSDqRcUFIqLi7FmzRoAwPjx4xEaGurXc/hivURERFQ9TL3wIn60XHvMZjMiIiIAAPn5+dDpdH49hy/W6294fRARUaBh6gURERERkQwGykREREREMhgoExERERHJYKBMRERERCSDgTIRERERkQwGykREREREMlhHmYKCWq3Gli1bpNv+Pocv1ktERETVwzrKXsQ6sUTO8fogIqJAw9QLIiIiIiIZTL2goFBcXIxPPvkEADBy5Mgaa2HtrTl8sV4iIiKqHqZeeBE/Wq49bGHt/3h9EBFRoOGOMlEVHDlyBBqNxun9er0eiYmJPlwREREReRsDZSI3paenS7e7d+/u8rFarRYpKSkMlomIiAIYA2UiN2VlZUm39+zZ43RHOSUlBaNGjYLJZGKgTEREFMAYKBNVQfv27etEXjEREVFdxvJwREREREQyGCgTEREREclg6gUFBbVajc8//1y6XRPEWsdLlizxSgvrml4vERERVQ8DZQoKISEhuP/++2t8DgDo27evdLs6Y9X0eomIiKh6mHpBRERERCSDgTIFhZKSEmzYsAEbNmxASUlJjc0BADt27Kj2HL5YLxEREVUPUy8oKFgsFgwZMgRAaUvo6qZGyCkuLgYAPP3003jssceqNYcv1ktERETVwx1lIiIiIiIZDJSJiIiIiGQwUCYiIiIiksFAmYiIiIhIBk8QEdWQlJQUp/cVFBT4cCVERERUFQyUibxMr9dDq9Vi1KhRbj0+PT0dV199dQ2vioiIiDzFQJmCQlhYGNauXSvdrgliCbd58+a5nCMxMREpKSkwmUxOH1NcXIz3338fa9aswZUrV7y+ViIiIqo+BsoUFEJDQzF27NganwMABg4cKN12JjExEYmJiZWOt2bNmkrHIiIiotrBQJkIQFpamssdYMB1zjEREREFHwbKFBRKSkqwfft2AED//v096nSXlpaG5ORkCIJQ6WPVajX+/PNPtGvXrlrd9EpKSvDLL79It4mIiMj/KOx2u722FxEscnNzER0djStXriAqKqq2l1OnmM1mREREAChtCa3T6dz+3kOHDuH666/HunXrkJyc7PRxBQUF6N69e5XmcLXePXv2oFu3blUeK1Dw+iAiokDDHWWi/y85ORkdO3Z0er/ZbPbhaoiIiKi2seEIEREREZEMBspERERERDIYKBMRERERyWCgTEREREQkg4EyEREREZEMVr2goBAWFoZVq1ZJt/19jrCwMDz11FNYunRpteoxExERUc3hKzQFhdDQUEyePDlg5ggNDcWQIUOwdOlStrAmIiLyU0y9ICIiIiKSwR1lCgpWq1VqCX3zzTdDpVL59RxWqxUHDhyQbhMREZH/YaBMQaGwsBA9e/YEUP320r6Yo7CwEBMmTAAAFBUVeWV9RERE5F1MvSAiIiIiksFAmYiIiIhIBgNlIiIiIiIZDJSJiIiIiGQwUCYiIiIiksFAmYiIiIhIBsvDUVAIDQ3FsmXLpNv+PkdoaCimTp2KlStXsoU1ERGRn+IrNAWFsLAwzJw5M2DmCAsLw+jRo7Fy5Uq2sCYiIvJTTL0gIiIiIpLBHWUKClarFYcOHQIAdOzYscZaWHtrDqvViuPHj0u3iYiIyP8wUKagUFhYiM6dOwOo2RbW3pqjsLAQo0ePBsAW1kRERP6KqRdERERERDIYKBMRERERyWCgTEREREQkg4EyEREREZEMBspERERERDIYKBMRERERyWB5OAoKoaGhmDdvnnTb3+cIDQ3F+PHjsWbNGrawJiIi8lN8haagEBYWhvnz5wfMHGFhYZgwYQLWrFnDFtZERER+iqkXREREREQyuKNMQcFmsyElJQUAkJycDKXS++8BvTmHzWbDv//+K90mIiIi/8NAmYJCQUEBrr32WgA118Lam3MUFBRgyJAhAACLxeKV9REREZF3MfWCiIiIiEgGA2UiIiIiIhkMlImIiIiIZDBQJiIiIiKSwUCZiIiIiEgGA2UiIiIiIhksD0dBITQ0FDNmzJBu+/scoaGheOCBB/Dxxx+zhTUREZGf4is0BYWwsDAsX748YOYICwvDE088gY8//pgtrImIiPwUUy+IiIiIiGRwR5mCgs1mQ1paGgAgMTGxxlpYe2sOm82GjIwM6TYRERH5HwbKFBQKCgrQrFkzADXbwtpbcxQUFOCuu+4CwBbWRERE/oqpF0REREREMhgoExERERHJYOoFBbW0tDSYTCaXj0lJSfHRaoiIiCiQMFCmoJWWlobk5GQIglDpY7VaLfR6vQ9WRURERIGCgTIFLZPJBEEQsG7dOiQnJ7t8rF6vR2Jioo9WRkRERIGAgTIFveTkZHTs2LG2l0FEREQBhoEyBYWQkBBMmjRJuu3vc4SEhOD+++/Hhg0boFKpvLE8IiIi8jIGyhQU1Go1Vq9eHTBzqNVqPP3009iwYQPCwsK8MiYRERF5F8vDERERERHJ4I4yBQW73S6VgdPr9VAoFH49h91uR3Z2tnSbiIiI/A8DZQoKgiAgPj4eQM21sPbmHIIgoE+fPgCAwsJCr6yPiIiIvIupF0REREREMhgoExERERHJYKBMRERERCSDgTIRERERkQwGykREREREMhgoExERERHJYHk4CgohISEYM2aMdNvf5wgJCcGdd96JLVu2sIU1ERGRn2KgTEFBrVbjgw8+CJg51Go1FixYgC1btrCFNRERkZ9i6gURERERkQzuKFNQsNvtEAQBAKDVamushbW35rDb7SgoKAAAnDhxotLH6/V6JCYmVnk+IiIi8hwDZQoKgiAgIiICQM22sPbWHIIgoHv37gCABx54oNLHa7VapKSkMFgmIiLyIQbKRLVsz5490Gg0Tu9PSUnBqFGjYDKZGCgTERH5EANlolrWvn37GtkBJyIiourhYT4iIiIiIhkMlImIiIiIZDBQJiIiIiKSwUCZiIiIiEgGD/NRUFCpVLjvvvuk2/4+hy/WS0RERNXDQJmCQnh4ODZs2BAwc/hivURERFQ9TL0gIiIiIpLBHWUKSGlpaTCZTC4fk5KS4qPVEBERUTBioEwBJy0tDcnJyRAEodLHarVa6PV6r8xrNpu91sLam2MRERFRzWCgTAHHZDJBEASsW7cOycnJAICCggJ0794dQNmW0Hq9nm2fiYiIqEoYKFPASk5ORseOHQGU7tCK2BKaiIiIvIGH+YiIiIiIZDBQJiIiIiKSwUCZiIiIiEgGA2UiIiIiIhk8zEdBQaVS4fbbb5du+/scvlgvERERVQ8DZQoK4eHh2Lp1a8DM4Yv1EhERUfUw9YKIiIiISAYDZSIiIiIiGQyUKSiYzWbodDrodLoyzUf8dQ5frJeIiIiqhznKFDQEQQioOXyxXiIiIqo67igTEREREclgoExEREREJIOBMhERERGRDAbKREREREQyGCgTEREREclg1QsKCkqlErfeeqt029/n8MV6iYiIqHoYKFNQ0Gg0+OmnnwJmDl+sl4iIiKqHW1lERERERDIYKBMRERERyWCgTEHBbDbDYDDAYDDUaAtrb83hi/USERFR9TBHmYKGyWQKqDl8sV4iIiKqOu4oExERERHJYKBMRERERCSDgTIRERERkQwGykREREREMhgoExERERHJYNULCgpKpRKdOnWSbvv7HL5YLxEREVUPA2XyK2lpaZWWTUtJSanwNY1Gg/3799fUsrw+hy/WS0RERNXDQJn8RlpaGpKTkyEIQqWP1Wq10Ov1PlgVERER1VUMlMlvmEwmCIKAdevWITk52eVj9Xo9EhMTfbQyIiIiqosYKJPfSU5ORseOHT36HkEQcM011wAATpw4Aa1W6/V1eXMOX6yXiIiIqoeBMgUFu92Os2fPSrf9fQ5frJeIiIiqh8ftiYiIiIhkMFAmIiIiIpLBQJmIiIiISAYDZSIiIiIiGQyUiYiIiIhksOoFBQWFQiGVW1MoFH4/hy/WS0RERNXDQJmCglarxfHjxwNmDl+sl4iIiKqHqRdERERERDIYKBMRERERyWCgTEFBEAS0adMGbdq0gSAIfj+HL9ZLRERE1cMcZQoKdrsdJ06ckG77+xxVGSslJaXSx+j1eiQmJlZrbURERFSKgTKRn9Pr9dBqtRg1alSlj9VqtUhJSWGwTERE5AUMlIn8XGJiIlJSUmAymVw+LiUlBaNGjYLJZGKgTERE5AUMlIkCQGJiIoNfIiIiH+NhPiIiIiIiGQyUiYiIiIhkMPWCgoJCoUDTpk2l2/4+hy/WS0RERNXDQJl8Ii0tza3DaFWl1WqRmppa5e/39Ry+WC8RERFVDwNlqnFpaWlITk52q7GGVquFXq/3waqIiIiIXGOgTDXOZDJBEASsW7cOycnJLh/LhhlERETkLxgok88kJyejY8eONTJ2QUEBbrnlFgDAzz//DI1G49dz+GK9REREVD0MlCko2Gw2HDhwQLrt73P4Yr1ERERUPSwPR0REREQkg4EyEREREZEMBspERERERDIYKBMRERERyWCgTEREREQkg1UvKGj4olGJN+dgYxUiIiL/xkCZgoJOp4PRaAyYOXyxXiIiIqoepl4QEREREclgoExEREREJIOBMgWFgoIC9OjRAz169EBBQYHfz+GL9RIREVH1MEeZgoLNZsPu3bul2/4+hy/WS0RERNXDQJkoyKSkpFT6GL1ej8TERB+shoiIKHAxUCYKEnq9HlqtFqNGjar0sVqtFikpKQyWiYiIXGCgTBQkEhMTkZKSApPJ5PJxKSkpGDVqFEwmEwNlIiIiFxgoU7WkpaW5FZiRbyQmJjL4JSIi8hIGylRlaWlpSE5OhiAIlT5Wq9WyEx0REREFFAbKVGUmkwmCIGDdunVITk52+VhfHB7TarU1Or635/DFeomIiKjqGChTtSUnJ6Njx461ugadTgez2Rwwc/hivURERFQ9DJRJFnOPiYiIqK5joEwVMPeYiIiIiIEyyfC33GN3FBYW4t577wUAfPnllwgPD/frOXyx3sqwMQkREZFrDJTJKX/IPXaX1WrFtm3bpNv+Pocv1usMG5MQERG5h4FyHcPcY2JjEiIiIvcwUK5DmHtMIjYmISIiqhwD5SDh7k5xoOUeExEREdUWBspBwNOd4ptvvplBMBEREVElGCjXEnd2gN3FnWKqSayOQUREdRUDZS+y2+0AgF9//RU6nc7p40wmE0aNGoWCggKvza3RaNC+fXskJCRU+tjc3FyvzesvHLvc5ebm1kglCW/O4Yv1VpdarYZGo3GrOoZGo8G6detc5rWLP7N4nRAREfk7hZ2vWl5z7tw5twJVorosPT0dTZo0qe1lEBERVYqBshfZbDZkZGQgMjISCoUCQOluYUJCAtLT0xEVFVXLK/QM1+57gbpuoPK12+125OXloVGjRlAqlbWwQiIiIs8w9cKLlEql052yqKiogAt8RFy77wXqugHXa4+OjvbxaoiIiKqO2zpERERERDIYKBMRERERyWCgXMPUajXmzZsHtVpd20vxGNfue4G6biCw105ERCSHh/mIiIiIiGRwR5mIiIiISAYDZSIiIiIiGSwP50VydZSJqJS7dZR5HRE5x3rkRL7FQNmLMjIy2JmPqBKVdebjdURUOXa4JPINBspeFBkZCQAB2VWNqKZdvHgRrVu3lq4TZ3gdkTNWqxW//fYbAKBr165QqVQ+Gcdb83qD2AGzsuuIiLyDVS+8KDc3F9HR0bhy5Qpf4InKcff64HVE5ByvDyLfYoITEREREZEMpl4QEVFAKC4uxpo1awAA48ePR2hoqE/G8da8RBR4mHrhRfxIjMg5pl5QdZnNZkRERAAA8vPzodPpfDKOt+b1Bl4fRL7F1AsiIiIiIhkMlImIiIiIZDBQJiIiIiKSwUCZiIiIiEgGA2UiIiIiIhkMlImIiIiIZLCOMhERBQS1Wo0tW7ZIt301jrfmJaLAwzrKXsT6lkTOsY4yUfXx+iDyLaZeEBERERHJYOoFEREFhOLiYnzyyScAgJEjR1arhbUn43hrXiIKPEy98CJ+JEbVZbPZUFhYCK1W6/O5zWYzzGYzdDpdjbToZeoFVRdbWPP6IPI17igT+Ynff/8dw4YNQ1paGho1aoRWrVqhVatWaNmypXS7efPmCAsLq5H5zWYzSkpKpGCZqK5KS0uDyWSS/l5QUCDdPnLkCDQaDQBAr9cjMTHR5+sjIt9hoEzkB959911MmjQJxcXFAICMjAxkZGTgp59+KvM4pVKJpKQktGrVCt27d8djjz2G6Ohor6xBp9MxSKY6Ly0tDcnJyRAEQfb+7t27S7e1Wi1SUlIYLBMFMQbKRLWoqKgITzzxBN58800AwODBg/Hqq68iIyMDp06dwunTp3Hq1Cnpdn5+Ps6cOYMzZ87gu+++w8svv4zZs2dj8uTJ0i5XVdVUygVRIDGZTBAEAevWrUNycjKA0h1lMUDes2cPNBoNUlJSMGrUKJhMJgbKREGMgTJRDXAn9f/SpUsYPHgwfv/9dygUCjz33HOYMWMGFAoFYmJicM0110iPVavVsNvtuHjxIk6fPo0TJ05g9erVOHXqFGbOnIlXX30VzzzzDB566CG3DhopFIpq/Xw1PR5RbUtOTkbHjh0BlKYlidq3b883lER1CMvDEdWC/fv3o1OnTvj9998RFRWFzz77DDNnznQZcCoUCjRs2BC33HILJk6ciMOHD2PNmjVITEzE+fPnMXnyZFx77bX47LPPYLPZPF6T2WxGSkoKDh48iNTUVOmPIAgQBAFGo9Hpx9FERETBiIEykY99+OGHuOWWW3D+/Hm0atUKu3btwoABAzweJyQkBGPGjMGxY8ewYsUKGAwG/PPPPxg5ciQ6deqErVu3urWzDZQGyX///TeOHj2KzMxMnD9/Hvn5+cjLy5OqYYgH/YiIiOoKpl4Q+UhxcTFmzJiB119/HQAwcOBAvPHGG2VKPNlsNigUCo9SGdRqNaZMmYIHH3wQq1evxooVK/Dnn39i4MCB6NSpE5555hnccsstsFqtAABBEGC322EwGCAIgnS632azwWKxwGw2Iz4+XiqHJX7MzIN+VNvUajU+//xz6bavxvHWvEQUeBgoE1VTcXExTCYTMjMzpT+XLl2C0WhEZmam9N9Tp04hOzsbADB37lxMmzYN+/fvx9GjR3Hs2DH89ddfOHHiBJo3b47t27d7XCM1IiICzz77LCZOnIhly5Zh1apVOHDgAAYPHoxGjRrhq6++gkqlksrLFRQU4MyZM1AqlQgNDUW9evWg1WoRGxsLrVaLpKSkMuPL1XY2Go0wGo0wGAyIj4+v2j8gkZtCQkJw//33+3wcb81LRIGHgTKRhy5duoQff/wRP/74I3bu3InU1FS3vzc8PBzt27fHJ598goULF8o+5vjx45g1axbeeuutKq0vLi4OS5cuxdSpU5GQkACgdBf5ypUr0Gg00p+srCxotVqkpqaiRYsWAIDWrVvjypUr0Ov1bs1lNBphsVhgNBoZKBMRUdBhoExUidzcXPz888/44Ycf8OOPP+LYsWMVHqNUKmEwGBAXFyftvFosFuTm5iIzM1NqWFBYWIi9e/dK39eoUSO0bdsWbdu2xbXXXgulUomxY8fi008/Rf/+/XHPPfdUed3R0dEIDw9HYWEhZs6cCY1Gg4YNGyIyMhJarRZZWVmwWq1o1aoVoqOjoVKpoNfr0axZM7dTLAwGg7SjTFTTSkpK8NVXXwEA7rnnHoSEVO0lzNNxvDUvEQUeXu1EMo4ePYoNGzbghx9+wL59+6T8XlH79u3Ru3dvNG7cGBcuXMA///yDo0ePIiUlRfYAXWhoKJKTk3HdddeV+RMeHl7hsdOnT8eKFSswdepUdO7cGY0bN67yz1BYWIiYmBi0a9cOsbGxUKlUEARBOqAnCALq1auH+vXrQ6vVepyDbDAYYDAYWB6OfMJisWDIkCEASltJVzVg9XQcb81LRIGHVztROR9//DHGjRuHkpIS6WstWrRA79690bt3b/Ts2RORkZF4+umnMX369Arf36BBA1x33XVo27Yt2rVrh3bt2uHqq6+WbT2dn59f4WuzZ8/Gjz/+iMOHD+PRRx/F119/XaWf48iRIwCAxo0bSx3/gNLOY7m5uSgsLJQCY+4IExERVcRAmej/s9vtWLJkCZ555hkAQL9+/TBkyBD07t27zMG2o0ePolevXlIKxr333otu3bqhXbt2aNu2LeLj4yvsKufk5KC4uNitHdvQ0FC8++67uPnmm/HTTz/hjTfewJNPPunxz/Pnn38CAJo2bSoF6YIgICcnByEhIWjQoAHCw8Oh1WpZ0YKIiEgGA2UKau7WERYEAU8++STefvttAMC0adPwwgsvQKksLTVeWFgIm82G8ePH4/PPP0dxcTGioqLw8MMPo3379gCAjIwMZGRkAABGjx7t1rz79+93et/UqVOxePFizJs3D7fddhvatm1b6XiOKRBioNyjRw80a9YMMTExyMnJQWxsLGw2G1q1agWgNP9SEATodLoyVSyYUkFERHUdA2Wq8woKCjB8+HBs3rwZCoUCy5cvx+TJk8s85sKFC5gwYQJ27NgBALjuuuvw8MMPIzo6WnY8b7jnnnvwyy+/YM+ePRg9ejR+//132ZxmOVarFX/99RcASDnORUVF0Gq1iImJQUhIiFRH2XE32bGKBdMxiIiorqvTnfksFot0292dRwouWVlZ6Nu3LzZv3gy1Wo1PPvmkQpC8ZcsWdO7cGTt27EBoaChGjx6N6dOnVwiSz507hw8++ABTp051e/7jx487vU+hUGDOnDmIjY3FsWPH8Nxzz7k97unTp1FQUICwsDA0btwYMTEx0Gq10Gq1yMnJgdlshtFoBFCanyxW6jAYDFCr1QySiYiIUIcD5RMnTuDee+/Fjz/+CKA0KGGwXLekpqaie/fu+O233xATE4MtW7aUKcdmNpsxZcoU3H///TCZTGjXrh0WLlyIPn36SCkJJSUl+OOPP7Bo0SI888wz2LlzJwoLC91ew9ixYzF//nwpaC0vNjZWCpBXrlwp/b5WRjzId9VVVyEiIgI5OTnSfTExMRAEQbYltcFgwDXXXMNAmYiICHU09cJut2PZsmXYs2ePFPD07t1bCpbdzcu0WCxldqVzc3NrZL3kXXa7HRs2bMCUKVNgNBqRkJCAb775BsnJydJjzp49i8GDB+PEiRMAgCeeeALz58/Hpk2bpMdcuHABL7/8Mi5dugSgtJby9ddfj969e2PJkiVur2fr1q3YvXs31q9fjwYNGlS4/+abb8ZDDz2E9957D+PHj8fp06el3GlnP993330HAOjQoQPi4uLKHOY7ffo01Go1tFqtbOoIkb8KCwvD2rVrpdu+Gsdb8xJR4KmTgbJCoYBOp8PVV1+N0NBQLFmyBDabDX379vXo8NLixYuxYMGCGlwpeVt6ejomT56MLVu2ACjNNd6yZUuZTnQ7duzAQw89BKPRiAYNGuC9995Dr169yoxz8uRJvPrqqzCbzYiKikKvXr3Qo0cPxMbGerSeevXqITs7GzabzeljLl26hJ9//hkAoFarYbPZXAbKCxcuxEcffQSgNMh2bD197tw5lJSUICQkBDqdTrYtNZG/Cg0NxdixY6v8/WlpaTCZTACAdu3aASitYuMoJSXF6/MSUeCqk4EyAHTv3h2JiYno2bMn5s6dixUrVsBgMOD777/HsGHDkJiYWOkYs2fPLlNHNzc3V2oZTP5n+/btGD58OHJychAaGopnnnkGTz/9NNRqNSwWC4qKijBv3jy8+uqrAEpfSL/88ks0adKkzDh79+7FmjVrUFJSghYtWmDatGmIioqq0pqys7ORlJSEpUuXyu4mX7hwAY8++ijOnz+PpKQkbNu2zWWzgyVLlkhv3mbNmoU+ffoA+F8HvejoaFy8eBFFRUUwm81MsaA6Iy0tDcnJyRAEodLHarVat9u4E1Fwq7OBcmRkJDZt2oSnnnoKM2fOxOrVq3HnnXciIyMDY8aMAYBK0zDUajXUarWvlkxVIAgC8vPz8d5772HOnDmw2Wzo0qUL3nvvPVxzzTXS4/7991+MHj0ahw4dAgBMmDABS5YsKVNlwm63Y8uWLfj8888BANdffz0mTpxYrd+BPn364LnnnpOtYXz+/Hk8+uijuHDhAlq0aIHvv//e5Ru4lStXSjWgn332WTz++OMAIO0aa7Va1K9fH0VFRVAoFG4FDET+pKSkBNu3bwcA9O/f36MOeSaTCYIgYN26dWjZsiV+//13AMBNN91UYRy9Xl/mWqvOvEQU2Ors1d6qVSupLbGYU5qdnY0uXbrg9OnTqF+/PmvIBgGTyYTp06dj48aNAICHHnoIq1atKhPcfvrpp3j00UeRl5eHevXq4a233sLAgQPLjFNSUoLp06dLQXK/fv0wYsQIlykQ7njxxRdlf8/S09Px6KOP4tKlS0hMTMSPP/7ospX1W2+9hRkzZgAAJk+ejEmTJgEo23FPp9OVKQPH3WQKNBaLBXfeeSeAqreSTk5ORuvWrdGlSxdpnMqa7XhjXiIKTHW26sVVV10FtVqN9PR0jB49GidOnMCKFSvQoEEDTJ8+XcoJpcB1+vRp3HHHHdi4cSNUKhVef/11rFmzRgqS8/PzMW7cOIwaNQp5eXno1q0b/vjjjwpBcn5+PoYMGYJ33nkHCoUCI0eOxKhRo6odJAOQDZLPnj2LCRMm4NKlS0hKSsLbb7/tMkheu3YtpkyZAqB0J3zs2LE4c+aM0x1jVrYgIiJyT518W2y321FSUgK73Y6bbroJSqUSW7duRfv27dG0aVN89NFHZVoWk/9xdfgNAHbt2oVhw4bBZDIhNjYWa9euRffu3aUyaUePHsVDDz2Ef//9F0qlEn369EGfPn2wefPmMuPk5ubivffew7lz5xASEoK7774bTZs2xd9//+1yfrESxpkzZzB79mz89NNPAEpz49966y0pSC1/+C8lJQWPPfYYjEYjrrnmGnz//feoX78+VCqV7DyffvopJkyYAAAYN24cZs2ahaysLKkttWNDEceDe/y0hIiIqHJBv6MsVxtZoVAgLCwMEyZMQJMmTfDVV19JbYhvv/12vPfee24d5iP/9P777+P222+HyWTCNddcgx9++AHdu3cHUPr78Pbbb6Nfv374999/0bBhQ3z99dfo379/hWD04sWLeO2113Du3DnodDpMmjQJrVu3dmsNgiBgyZIluPXWW/HTTz8hLCwMM2fOxOeffy67k5ubm4vnn38e3bp1w8WLF3Httdfihx9+QP369Z3O8cUXX2DMmDGw2+0YNGgQZs+ejYiICDRt2lRKK8rMzERJSQkyMzNhNBqZl0xEROSBoNxRNpvNsNlssNvtLqsRDBkyBHfccQdiYmIA/O/wXmX5auSfiouLMXPmTLz++usAgMGDB+OVV15BREQEgNJgdNKkSfj2228BALfddhtef/11xMbG4vTp02XG+ueff7B27VoUFhZCr9fjkUcegV6vR1ZWVqXrOH36NG655Rakp6cDAHr16oUXX3wRzZo1q/BYs9mMN954AytWrMDly5cBAF26dMHXX3/t8tT95s2bMWLECFitVgwYMABPPfUUioqKyuwcl5SUSLW+TSaTFDyzJBwREZF7gi5QPnHiBKZNmwaj0YhLly5h2bJlGDlyZJkKFlarFSqVCqGhoYiJiZHq0vLj6MCVlZWFYcOGYefOnQCABQsW4JlnnpGawPzzzz8YNWoUTp8+jbCwMCxcuBAPP/yw7P/z33//HRs3boTNZkNSUhLGjRvn1punnJwc7NixQwq6GzdujOeffx633357hXkKCwvx0Ucf4fXXX0dmZiYA4Oqrr8a8efMwePBgl/nP69evx9ixY1FSUoJevXrhqaeeQmhoKPR6vRQE63Q6mM1mREdHw2w2w2q1wmKx8E0gERGRB4IqUD5x4gRuueUWjB49Gp06dcLBgwfx4IMPok2bNlJqBQDpI/ZNmzbhpptu4qGmAJeWloY+ffrg33//RUREBD788EMMGjRIun/v3r0YMWIErly5gkaNGuGjjz5Chw4dZMfat28fvvjiCwBA+/btMWzYMISGhla6hr/++gvfffcdSkpKoFQqMWnSJEyfPl02MD116hRGjx6N//77DwDQvHlzzJkzB8OHD3eaiyx6/fXXMXXqVACllTdee+01pKenw2q1lkmr0Gq1FXaO2WCEiIjIM0ETKF++fBnTpk3DyJEj8fLLLwMARowYgUOHDuH999/Ha6+9VmZXecuWLZg8eTLGjBmDhQsXeqWCAfleWloaevfujTNnziApKQnffPMNrr32Wun+HTt24MEHH0RBQQFuuOEGfPjhhy7zfsWW5DqdDiNGjKg0cC0pKcGOHTtw+PBhAEBiYiJuu+02zJkzR/bxu3fvxsMPP4zc3Fw0aNAAc+fOxdixYysNxu12OxYuXCg1ExkyZAjmzJkDq9UKq9WKCxcuICIiQvbwnlzQTBSIwsLCsGrVKum2r8bx1rxEFHiCJlAuLi5GTk4O7rvvPgCQ0imaNWsm5X46fvx95513Yt++fRg7diyD5ADlGCQ3b94cO3fuLNMZ0TFFoW/fvnj//fcrDRg7d+6MH374Afn5+fjtt9/QqlUr5OXlIS8vD/n5+cjMzJSCUbPZjCtXrsBsNgMobRfdvXt3pyk8H3/8MZ566ilYrVZ07twZa9euRatWrSr9OW02G5588km89tprAEqrWwwaNAhKpRJ6vR4FBQWwWq1SpYuSkhKYzWYGxxR0QkNDMXnyZJ+P4615iSjwBE2gXL9+fanjElCah6xUKtG4cWOcPXu2zGNzcnIQExODhQsX1sZSqZrMZjNOnjyJwYMH4+zZs7JB8ptvvokpU6bAbrfjvvvuw6pVq9xKoVCr1ejVqxc2bdqEr7/+2q31hIeHY+DAgbjqqqucrvfFF1/Eu+++C+B/hwwdu/45U1JSggkTJuCjjz4CAMybNw89e/ZEYWGhdDivdevWEAQBWq0WCoVC2lEmIiKi6gmaQBmAFCTbbDYpKLLb7dJhKQBYvHgx1Go1Hn/8cXZXClCugmS73Y4lS5bgueeeA1C6+7p06VKPPjXo2rUr9u3bh4sXL0Kj0SAiIgKRkZGIiIhAaGiolNYgdrozGAyybaxtNhs2bNiARYsWSXWVZ82ahenTp7t1cLSwsBCjRo3CN998A5VKhffffx+33XYb0tPTpTU5rkPEnWQKVlarFb/88guA0k9wKkuN8tY43pqXiAJPUEaKSqWyTD6yGCTNnTsXL7zwAg4fPswgOUCIKQ5iMJiWlob77rvPaZA8a9YsKUf92WefxbRp0zyuZhIaGooZM2bAarVW+D1xpzwcUHoocM6cOThy5AgAoGnTpnj++efRv39/t74/Pz8f9957L3bu3Am1Wo23334bgwcPhiAIaNiwIUJCQmQPoTprMEIUDAoLC9GzZ08A7rWe9tY43pqXiAJP0EaLYqAcEhKChIQErFixAsuWLcOBAwdw3XXX1fby6hy5xi/uPM4x59ZoNKJ3797477//kJSUhI0bN0Kn0+Hy5csoKSnBk08+iU8//RQA8Pzzz2PixIk4duyYW/OKDUkq065dO5f3p6Wl4bnnnsOLL74IAIiMjMSzzz6Lxx9/XHbXWS6Iv3z5Mu644w788ccfiIiIwIoVK9ChQwcYjUbodDpkZ2c7rdTi+O/FF3MiIqLqCdpAWdxFDg0NxTvvvIOoqCjs2bMHHTt2rOWVkSe0Wi0EQUBWVhbuvPNO6eDexo0b0bhxYwBAUVERJkyYgC1btkCpVOLVV1/F8OHDfbpOs9mMl156Ca+88goKCwuhUCjw4IMP4vnnn0eDBg3cHicjIwP9+/fH8ePHERsbi+eee076OUX16tWr8H3iTjIAhISEMEgmIiLygqAv9yB+1P3bb7+hU6dOtbwa8pQYKDsGyTt37iwTJD/yyCPYsmULwsLCsHbtWp8GyTabDevXr0e7du2wePFiFBYWonv37ti/fz/eeecdt4PkkpISvP/+++jcuTOOHz8Og8GAF198Ec2bN0fDhg0RGRkpPVYMhAVBkNpSizvJABAfH89AmYiIyAuCdkdZ1KlTJ+Tl5TFwCFDOSsBdvnxZCpK3bdsGtVqNDz/8EL179/bZ2nbt2oU5c+bg4MGDAErzkBcvXoxBgwbJplnIsdvt+Prrr/Hcc88hJSUFQGlHvxUrViAuLg6NGzdGYmIiBEGQAmEx7cJoNJZJs2C6BRERkXcFfaAMgMFDgKqsTvK0adNqJUi+cuUKJk6cKJWPi4iIwKxZszBlyhS3Sr6JTp48iYkTJ+Lnn38GAMTFxeGhhx5Cv379oFarER0dLT1WLhB2/JrYVIRt2ImIiLynTgTKFHjS09NdBskbN27E559/DpVK5dMg+fDhwxg5ciT+++8/hIaGYvz48Zg5c6bLbn/lFRUVYcWKFXjhhRdgsVig1Woxbtw4PP300yguLkZeXp5UfkqtVkMQBBgMhgpVLNhxj4iIqGYxUCafq6yE2alTp3Dbbbc5bSaSnp6OWbNmAQCefPJJnwTJdrsdX331FV599VVYLBYkJibik08+8Tjvfd++fRg/fjyOHj0KoLRm8xNPPIGkpCRcvHgR//33HwRBQJs2bco0EiGi0sPZy5Ytk277ahxvzUtEgYeBMvlcZmYm8vLyEBkZiaSkpDL3paenuwySbTYbHnroIVy5cgUdO3bEE088UePrFQQBS5Yswfbt2wEAd9xxB9555x3Z6hPO5OfnY86cOXj99ddht9sRFxeHWbNm4f7774fJZIIgCCgqKkJxcTFsNhsAVGgkQlTXhYWFYebMmT4fx1vzElHgYaBMfkEQBJw6dQr33nuv0yAZAFatWoUff/wRWq0Wb7zxRo3v7vz777945plnkJqaCpVKheeffx5PPPGE27nABQUFeO+997B06VJkZGQAAIYNG4Y5c+YgMTERQNlayhqNBkBpvrLRaJQ67xEREZHvMVAmnxPLlzkGgEeOHMGwYcNw7tw5p0HyiRMn8PTTTwMAFixYgBYtWtToOrdu3YqlS5fCYrHAYDBg0aJFeOCBB9z6XkEQsGbNGixfvhwXL14EADRq1Ahz587FPffcU+bwnXgoTxAEtGrVSvr+kpISCIJQ5t+JnfeoLrNarTh06BAAoGPHjtVqYe3JON6al4gCDwNlqpaqdNzTaDTQaDRSHeDLly9jxIgROHfuHJo1a4atW7ciLi4OgiBI31NUVISRI0fCYrGgX79+GD16tFtzd+nSxa31hYWFobi4GH/99Rf++OMPfP/999i8eTMAoG/fvvj4449hMBgq3Uk2m8148803sWLFCmRmZgIAGjZsiGnTpuGuu+5Cfn4+cnNzAaBCAFxSUoLz589DrVZDpVKV2U0W52XnParLCgsL0blzZwDVb2HtyTjempeIAg8DZao1giDg5MmTGDlyJNLT09G8eXNs3boVTZo0qfDYF198EX/99RdiY2OxevVqr5RBy8jIwB9//IF9+/bhwIEDOHDgAAoLC6X7FQoF5s+fj9mzZ1e6g5SXl4fVq1fj5ZdfhslkAlC6g/zggw9ixIgR0Ov1AEqbhRQVFVV4oRV3lbVaLaxWK7RarWybatZLJiIi8h0GyuRzgiBAEAScP3++TJC8c+dOxMXFVXj8b7/9hldeeQUA8Nprr6FBgwYoKCjweF6j0YivvvoKu3fvxr59+3Du3LkKj4mJiUGXLl1w44034o477sD111/vcswrV65g1apVeOWVV3D58mUAQGJiIsaPH4/JkycjNDRUNrAVd5vj4+OlNAxxB9lVIMwDfkRERL7DQJlqjCAIUkDoWAdYEAScPXsW999/f5kgOSEhoUy6BVC6UzthwgTYbDaMHDkSd999t8dr2LJlCz777DPs2LFD6m4HAEqlEm3btkXnzp1x00034cYbb0TLli2hVDrv7J6fn4/ffvsNP/30E3bv3o39+/dLY7Zq1QozZszAHXfcgejo6DI5xI6BsdlsRn5+PgBUyDVmbWQiIiL/wUCZaozZbEZmZiYEQYBCoYBer5fykocMGeKyuoVoyZIlSE1NBQAkJCTg1KlTaNmyZaVzZ2dnY82aNXjvvfeQk5Mjfb1jx464++67ceONN6Jjx46IiIgAUJqjXNl4S5cuxapVqyrsZrdu3Rpz587FkCFDUFhYKAX74sG78oGxTqeT5tXpdDygR0RE5KcYKFONunLlihSECoKAvLw8PPDAA0hNTUWzZs1cBskAEBUVJd1esmQJlixZgubNm6N3797o06cPunbtCrVaLT3GaDTizTffxNq1a2E2mwEATZs2xfDhwzFs2DC0bt3ao/UXFBRg9erVWLx4sRRwN2zYENdeey2uu+463HLLLWjbti30ej0KCwuRlpZWZj3ibrNjYKzVasvUjzYajdIBPQbKRERE/oOBMlWZuGMM/C/XViTuHDdo0AAqlQp6vR5WqxVjx47F4cOHERsbi23btrkMkgHgqaeewn333Yft27dj+/bt+OWXX3DmzBmcOXMG77zzDrRaLW655Rb07dsXJ0+exMcffyzt+LZp0wbPPPMMBg0a5DKdQo7VasVHH32E+fPnS7nMbdq0weLFi9G5c2fp51apVAgLC4PJZEJWVhZMJhOUSiWuuuoqaaz4+Hjp30sOD+gRERH5JwbKVGVmsxl5eXkAKubams1mqNVqWCwWJCUlQaPR4IknnsC3334LtVqNr776SqoZXJkWLVpg0qRJmDRpEvLy8rBr1y5s27YNP/zwAy5duoTvvvsO3333nfT4Dh06YPr06ejXr1+ZHWl3WCwWfPrpp1ixYgX+/vtvAKXVK55++mlMnDgRISEhMBqNKCoqgkajkapZFBUVwWQy4Z9//sFVV12FgoICxMTEICQkBFqt1uWuMfOSidwTGhqKefPmSbd9NY635iWiwMNAmapMp9MhMjJSul3+PpHZbMbLL7+M119/HQCwdu1adO/evUpzRkZGYuDAgejbty9sNhuOHTuGHTt2YNeuXdDpdJg4cSJ69Ojhcfm4K1eu4J133sHKlSulDnqxsbF46KGHMG7cOMTGxiI7O1sKaBs2bIiioiIA/wt0T58+jYYNG0KpVEKj0SAjI0OqZMFdY6LqCwsLw/z5830+jrfmJaLAw0CZPCIeThODP8dcW2eP/+qrr6TdmKVLl2Lo0KFeWYtSqUS7du3Qrl07PPnkk1UaIyMjA6tXr8a7774rNQKpX78+Hn30UQwYMABxcXHSmwGxU55Wq4UgCCgqKkJJSQlMJhO0Wi1atmwJQRCg1+ulA3xWqxVms7lM1Q8iIiIKDAyU6xh3O+k549gZTi7wE1sxC4IAq9WKH374Ac8++yzsdjsmTJiAJ554Alar1en4jgfhXKnuzuzff/+NFStWYN26ddLOcOvWrTFhwgTcf//9CAsLQ25uLgoLC9GgQQPpZxNTTHQ6HYxGo9RcJCwsTGoSIv4b2O32MhU35KpbeKNxClFdYbPZkJKSAgBITk72+OxBVcfx1rxEFHgYKJPbxEAPAKKjo50+Rqz0cPr0aTz33HMoLCzE7bffjldeeaXWAkO73Y6MjAwcOXIE7777Lr755hvpvhtvvBGTJ09Gnz59oFQqywThjkGt3BuDmJgYWCwWhISESOkVJSUl0Gq1iI+PlwJ/QRBw4sQJWK1W1K9fv9KdeCKqqKCgANdeey2A6rWS9nQcb81LRIGHgTLJktv9FA/oiQfUHIm7qCaTCZcvX4ZCocCjjz4Kk8mEjh07Yt26dQgJ8c2vW25uLo4dO4ajR4/i6NGj0u3s7GzpMQqFArfddhtmzJiBa6+9FiUlJVAqlWXaRpev4uFYtcIx71j8ucS0FKBspQvxcVartUodBYmIiKh2MFAmWWazGbm5uTAajUhKSpJtryymHoiVH/Ly8nDmzBlERETg8ccfR2pqKhITE7F582apjnBN+Pfff/F///d/+P3333Hs2DGcPXtW9nEqlQpNmjRB165dMWfOHFx99dVS90Ax91hUvopHZmYm8vLycOXKFcTExCAiIkL6d0lNTUVmZiauXLmC5s2bIyoqSnYXun79+gD+F0QTERGRf2OgTLLEHFy1Wi3lI2u1WpjNZqSmpsJgMODkyZMwGo2Ij49H+/btYTKZEB8fj1deeQVHjx5FVFQUtm7dioYNG5ZpHe0tZrMZjz/+ONauXVvhvsaNG6Nt27a4+uqrcdVVV6Fjx45o1aqVdGDPccdXrVYjKysLVqsVWVlZaNSokcfNPwoKChAWFoaioiLZj2XLNxkhIiIi/+dxoGw2m7FkyRL8+OOPyMzMhM1mK3P/mTNnvLY48j3HlIukpKQKJc3EGsJGoxEFBQXIzc2VqkKIwbS4wywIAo4dO4Y2bdp4fZ3Hjx/H0KFDceLECSgUCvTp0wc9evRAkyZN0KVLF6c1mq1Wa5nDiOIuuUajQX5+PoqKimCxWBAdHV3m30IMrMX7xd1ho9GIrKwsREREIC4urkLjFSIiIgpcHgfKDz/8MHbv3o0HHngADRs25Kn9IONY1UKupJnBYEBaWpqUa9uoUSPExsbi5MmTsNlsMBgMmD17NjIyMrBx40aMGDECZrMZo0eP9sr67HY7PvzwQ0ydOhUFBQUwGAz48MMPcdttt+Hs2bPIzMxEfn6+05rF5dNHxOBep9MhNTUVVqsVgiBI/xa5ublIS0tDXFwcgIpVOUwmE1QqFdRqNXeMiYiIgozHgfK3336LrVu3olu3bjWxHp+5cuUKrFYr8vLy0LRp09pejt9w1hjDcXc1MTERFy9eRHR0NCIjI6XOc2KAGRUVhdWrVyM6Ohpr167Fww8/jNzcXDz22GPVWlt+fj4ee+wxfPrppwCAHj16YN68eVK3PKA0BUKj0cBkMkEQBOnncDyYKLfjK6ZGiMGymF5isVik35PIyEgp11ocV6/Xl8nTJiLfS0tLk0o1OiOWdyMi8oTHgXK9evUQGxtbE2vxmePHj2PSpEnIzs5GQUEBnn32WYwdO9bjcSwWCywWi/R3Mf81kDkGko7BsbjTnJmZCaD0YFx8fHyZKhFAaeAoVoRYuXIlbDYbPvzwQ0yfPh35+fl4+umnq7Suv/76C8OHD8fp06ehUqkwffp0PPfcc/jrr7+gUqlgMpmQmJgIQRBQUFAg1XE2mUzQaDRQq9XIzMyU1lZZsCzuHCclJUk/M9MqiGpXaGgoZsyYId0GSoPk5ORk6Y26K2JqmNw4ns5LRHWDwu5hB4p169bhm2++wYcffhiQQUNKSgq6d++ORx55BG3btsXBgwexZ88e7Nixw2ltYGfmz5+PBQsWVPj6lStXEBUV5a0le5X4v1uu/BtQmnNrNBqlADg3NxcWi0X6u2OJuPJBcnlGoxG5ubmYO3cu1q9fDwCYNWsWnn/+ebdTdux2O9577z1MmzYNFosFcXFxWLBgAe644w40bdoUZ8+exblz56DRaACU7ig3bNgQKpVKCpbF7wNKUycsFovTgFkQBKSmpqKgoACJiYmV/owpKSnIycmB1WpFx44dPb4m6lLqUm5uLqKjoyu9Ptx9HBEAHDp0CNdffz3WrVuH5ORkl4/V6/VITEz06rwHDx5Ex44dvTKmO3h9EPmWWzvKHTp0KPOC/s8//0hNE8q/uz506JB3V+hFxcXFWLFiBQYNGoQlS5YAKN0lPH36NCwWC06dOuX0EJic2bNnY/r06dLfc3NzkZCQ4PV1u8PTjnty5d+A0uDWYrHAZDKhadOmMJlM0u6qwWCATqdDZmYmcnJyYDabZXdZxZJrWVlZAICxY8ciNDQUH330EZYtWwZBEPDqq6867W6VmZmJgwcP4sCBA9i9ezd27twJALjjjjuwevVqaDSaMjnGzZs3x7///ovi4mIUFxfDYrFI6TSpqamIiYkpU/fYYrE47S4o1jsuKSmB0Wh0uvssElMvYmJiyoxXlwJgIn+QnJzs04CViOoGtwLlQYMG1fAyfOfUqVPo1KmT9Pfff/8de/fuxa233orLly/jgQcewIoVK9waS61Wu91y2d/IlX8DSgPPrKws1KtXD1qtFk2bNpWaaIg7rYIgID8/Xxqr/CE2s9mMvLw8GI1GaDQaxMbG4pFHHkFiYiIWLVqEVatWIT8/H++88w6ys7OloPjQoUM4cOAA0tPTy4wXEhKCZ599FhMnTpRyoh1/DrPZjCZNmiArK0tas/gYx8odjmklZ8+elQ7xlR8vMjISFoulwr+NHPHNg7PDg0TkPTabDWlpaQCAxMTEarWw9mQcb81LRIHHrUB53rx5Nb0OnwgNDcUtt9yCTz75BHFxccjMzMR7772HDz/8EAkJCbh48SLuu+8+XHPNNRg3blxtL7cMseubuDNaXWI+bvkAT6fT4aqrrpK6zYnBpdFoRH5+PrKysqDRaJCXl4fQ0FDZvEAx2FSpVFJliquvvhqtWrVCgwYN8Pjjj+ODDz7A5s2bpV1nRwqFAklJSejatSuuv/569O/fHxqNBpmZmSgoKJCtLhEfHy99XRAE6XCfuB7HvGvxZxGDfTHYFX/WpKQk6HQ6nD17VkrTkOvS5/g9gZiGRBRoCgoK0KxZMwDVb2HtyTjempeIAk+dazgybNgwAKU7y//88w8WLVqE++67T7r/5ptvxsGDB/0yUBbTBbz1JC0X4DnbHdXpdLBYLDAajYiJiUHDhg2hVqtl0y7E74+NjYXVapV2bgHg9ttvh9lsxnPPPScFyS1atEDHjh3RpUsXNGnSBImJiWjQoEGZgDg1NVX2Z3A8ZOiYXlFSUoKcnJwKKRFih73c3FzUr18fgiA4TcNQq9WwWq0V7pObk4EyERFR8KlS1Qu5/EuFQoHw8HBcddVVGDt2LB588EGvLNDb2rZti7Zt28JisaBbt24ICwuT7rPb7QgNDUXDhg1rcYXyaurjfXd2R8XHAP9LzQgPD5cabDgSg8iMjAxYrVapOoY4jslkQrt27bBs2TKEhoaiU6dO6Ny5s+x6HMXHx8vupos502JnPVFGRgYAQKPRVDikqdVqcdVVV5U5qCeO6/izio1U5P7Ns7OzIQiCVDaONZSJiIiCj8eB8ty5c7Fo0SIMGDBACnD27duH7777DpMnT8Z///2HRx99FCUlJXjkkUe8vmBvUavV6Ny5M37//XfcdNNNSEpKwssvv4xjx45h9erVtb28CryVclGe40613K6oIAg4e/YswsLCoNVqERYWhvDwcISFhcnupDp2urNarWVyirVarVROr1OnToiNja1Qf9hVoC43n7hbbbVaUVRUBKC0XJRYCSMuLq7M9zgG3OXL4Dn+e4SEhLgMftVqNc6dOyfVkiYiIqLg43GgvGfPHrzwwguYOHFima+//fbb+P777/Hll1+iXbt2eO2112otUBZ3MivTtWtXvPXWW+jRowdat24No9GILVu24KqrrvLBKv1DZTvVZrMZYWFhKCoqQtOmTdG0aVOprnD5XViz2QxBEGAwGJzmPzdu3BgmkwkKhQKJiYluBf+ugnlxbpVKJR08FEvChYeHVxhLLhAvn9Yi/tdZgC4ehBTL0JXfVSciIqLg4HGgvH37dixdurTC13v37o0nn3wSQGkealUbS1TXqVOnsHnzZowYMcJpCoXNZoNSqcSoUaPQunVrnD59GqGhobjxxhtrrbxbbalsBxcAioqKoNfrodVqpW53jsGjGGieO3cOERERUh1mx5JzYvc6nU6H8+fPw2q14siRI2jVqpU0rmMus+M6gNLKF3JBtWNdZ61WKwXNLVq0kE0NcSY7O1tas+O65QL08gchmZ9MREQUnDwOlGNjY7F582ZMmzatzNc3b94sdewzm8218nH0P//8g5tuugnZ2dnIysrC9OnTK3y0b7fby5T2ueGGG3DDDTf4eqk1rvxuqKv0hfLfIwafYpAaExMjPUZud9exRJu4o+zIaDSiqKhICpYbNmyIlJQUxMXFSV8rKSkp03LacS5XzU3kdsQbNWrkVkMUR/Xq1XNrbBErXRAREQU/jwPlOXPm4NFHH8WuXbukHOX9+/dj27ZteOuttwAAO3bswK233urdlVbCbDZj8eLFGDhwIG644QY89thjKCkpwaxZs8oEy+JBxOXLl6OgoABz58716Tp9pXxAW/7vcoG02LpZpVJJO7hmsxk5OTnSv6Gz1ASDweA0MDUYDFJALAaXbdq0QUFBQZmdancrcIgNTQBUaKNdlUOPzr6HwTCRfwkJCcGkSZOk274ax1vzElHg8fiKf+SRR3DNNddg1apV2LhxIwCgdevW2L17N7p27QoAUgqGLymVSlx//fWIi4vD0KFDodfrpVJw5YPly5cv4+DBg0hNTcVjjz0m7YQHE7mAFoBUAcIxcAZKy6+Jub2OnfpSU1ORmZkJu91eZkyxtrLjGM52rMsH0WJgDVR+UE8uUBUbmojNTxzX65gOIs7lzr8V66IS+T+1Wu2Vw9aejuOteYko8FTprXG3bt3QrVs3b6+lWjQaDcaMGSMFPEOGDIHdbsfw4cNht9vx9NNPIy4uDlarFUqlEm+88QYsFktQBclypd4ASB34xDxeoGwgLaZalA+SxTGtVqs0tuOudPkxXFXPcMbd73P82SrrnlcTNaeJiIio7vE4UBbbeDqTmJhY5cVUlxgUicHw0KFDYbfbMWLECCgUCjzxxBNYvnw5UlNT8dlnnwVVkAw4DzrFWsOOXeacNRsp/7WmTZtKh/PKpyiU372VO3DnuJstziF+j2ODkMoCWsefTayqER8f77Q5ivjzOgby3uxsSES+Z7fbYTKZAAB6vV62pn9NjOOteYko8HgcKCclJbl8knBs+lBbVCoV7HY7bDYbhg0bBoVCgQceeACbNm3Cv//+i3379kGtVtf2Mr1GDAIB+WBV3P11tnPrKhe3fNqEs9Jqzg7PifdnZ2ejXr160jodg97KDhc6BtRA6Q55+aobjmsr//OW3/FmoEwUmARBkCrZVKeVtKfjeGteIgo8HgfKhw8fLvP34uJiHD58GC+//DIWLVrktYVVlxjM2+12DB06FGvWrMGRI0dw6NAhtG3btpZX512eVoeQywt2pyqGM+fPn5eCU2cH8sR1ld9Rdufncgyoy5dsc7WL7jhHTXU2JCIiouDlcaB83XXXVfhap06d0KhRIyxfvhyDBw/2ysK8QaFQwGq1YubMmdi1axeOHDkS0EGys/QBd4LAyuoDe5JjXL7OslqthtVq9Wi32lWbbPFnAv5X39jZz1pZxQrxDRNTLoiIiMhTXqtz07p1a+zfv99bw3lVmzZtcOjQIbRr1662lyLL3fxZZ+kDnu4CywWXnuy4ll+HWDO7OoGosxSN8vWNywfeztIvqrMO5jITERERUIVAOTc3t8zf7XY7Lly4gPnz56Nly5ZeW5i3qFQqjBs3zq8PX7ibP1vd9IHydY8deVIzuHyZuKSkpGqtxbGCRnVTNLwRKDOXmYiIiIAqBMoxMTEVgk673Y6EhAR89tlnXluYN/lzkAy4HwBXd5ezsqYj7vJGI47ya3E3RUOON/OPmctMREREIo8D5V27dpX5u1KphMFgwFVXXcWORVXkq4/5yweBZrMZubm5MBqNFeon+3ot5XkSxHsz/YIpF0RERCTyOLL1dWtq8h4xxUGk0+mkZiTlg1a73e7WmO7u1pcfr7JdaU+DeG+mXxCRfwoJCcGYMWOk274ax1vzElHgqdIV/++//+LVV19FSkoKAOCaa67B1KlT0aJFC68ujmqWTqdDUlKSR6kG1Skj58m45YP4yuZiygRR8FOr1fjggw98Po635iWiwKP09Bu2b9+Oa665Bvv27UO7du3Qrl07/PHHH2jTpg127NhRE2ukGlKVCg+OO7feIAgCjEYjMjMzkZubi9TUVAiCIB0SjIqKcrk28fsBuGxeQkREROQpj3eUn376aUybNg1Lliyp8PWnnnoKffv29driqHoq2/2tSoUHb+/cimsAAIvFUmYH2VV6hvizmc1mt3ediSiw2e12CIIAAGXqpNf0ON6al4gCj8c7yikpKXjooYcqfH3cuHE4ceKEVxZF3lHZ7q9Op5Ntee2KVqv16s6tuIb4+Hi3dpBFjgG2pz8DEQUmQRAQERGBiIgIKXD1xTjempeIAo/HO8oGgwFHjhypUDP5yJEjiI+P99rCqPp0Oh0yMzNhsVictpf2VYDpbHfbWQORyog729HR0dxJJiIiohrhcaD8yCOPYPz48Thz5gy6du0KAPj111+xdOlSTJ8+3esLpKoTD8X5QzUIT6tSVJY24o1azkRERESueBwoz5kzB5GRkXjppZcwe/ZsAECjRo0wf/58PP74415fYF3j7RbKNVUNwjGQdWdsd9fhae5xTVXhICIiIvIoUC4pKcGnn36KESNGYNq0acjLywMAREZG1sji6iJvt1CuqZ1XT9fp7jrkco9dBcOsn0xEREQ1xaPDfCEhIZg4cSIKCwsBlAbIDJK9qyoH7GpD+XWazWZkZmZ6pWxcdnY2dDqddGjQ1aHEQPn3IiIiosDjcepF586dcfjwYTRt2rQm1lNnOHaqK79jKu6MutsdD3C/Q563xiufclHZDrMn66tXr16FucqPK47HltNERERUUzwOlCdNmoQnn3wS586dw/XXX18hSGnXrp3XFldXuJs+4M/5uN7KhZYbhwf3iAgAVCoV7rvvPum2r8bx1rxEFHg8DpSHDRsGAGUO7ikUCtjtdigUClitVu+tro5wN8j053xcb+3sMigmImfCw8OxYcMGn4/jrXmJKPB4HCj/999/NbGOOs3d4LCmKlhUh7erdBARERH5C48DZeYm1x7HgNpf0jC8XaWDiIiIyF943MKa/ENl7al9xdtVJwRBgNFoZJtYIqrAbDZDoVBAoVBU67nP03G8NS8RBR6Pd5TJNyrbMfaXNAxvp1z4cx42ERER1S0MlP1UZQFjoB16czeX2V/eABARERExUPZTwRYwupvLHGhvAIiIiCh4VStHedKkSTCZTN5aCznQarVSZ7pgwA56REREFGiqFSivW7cOubm53lpLnSIeDFEoFGUOsDl+3ZM//k6n0yE+Pr7SQDlYfl4iIiIKfNUKlD1psUzO+UsFCyIiIiL6nzqZo/zPP//gs88+w/Hjx9GvXz90794dLVu29Hgci8UCi8Ui/b2qu+vBlo9MRFQTVCoVbr/9dum2r8bx1rxEFHiqFSjn5eV5ax0+c+zYMfTp0wddu3bFlStXsGjRItx1111YsmQJwsLCPPpYf/HixViwYEG118SudkRElQsPD8fWrVt9Po635iWiwFOnGo6cO3cOQ4cOxYMPPoiNGzfixx9/xJw5c/Dhhx/iwoULHue+zp49G1euXJH+pKen19DKiYiIiMjX6kygbLfbsXPnTiQnJ2PChAmw2WwAgOHDh6Nhw4ZITU31eEy1Wo2oqKgyf4iIiIgoONSZQFmhUKBBgwbo3r07kpKSoFSW/ug2mw2CIODixYu1vEIiInLFsWlRdVtYezKOt+YlosBTpw7z9evXD/369QNQusOsUCigVqtRr149hIaGSo9bv349WrZsiU6dOtXWUomISIYgCLUyjrfmJaLAUmd2lMsT85EVCgV0Oh00Gg2A0rzjRx99FPXq1avN5RERERFRLatTO8pyiouLkZWVhaKiIrzwwgtYuXIlfv75Z7Ro0aK2l0ZEVKelpaWV6f5aUFAg3T5y5Ag0Gg1SUlJqY2lEVEcEZaBstVrdrnWpUChQr149PPPMM0hNTcXPP//MlAsiolqWlpaG5ORkpykP3bt3l25rtVro9XpfLY2I6pCgC5RPnTqFzZs3Y8SIEWjYsKHsY8T8ZKB0RxkAjEYj9u7di3bt2vlsrUREJM9kMkEQBKxbtw7JyckASneUxQB5z549UsqcXq9HYmJira2ViIJXUAXK//zzD2666SZkZ2cjKysL06dPr7DL4BgkA4BGo8GECRNw4403olWrVr5ecq0ym81lTnMTEfmb5ORkdOzYEQDKVJxo3749n7eIqMYFTaBsNpuxePFiDBw4EDfccAMee+wxlJSUYNasWWWCZTFIXr58OQRBwLx58zB69OjaWnatMpvNKCkpYftsIgoISqUSt956q3TbV+N4a14iCjxBEygrlUpcf/31iIuLw9ChQ6HX6zFs2DAAqBAsX758GQcPHkRqaioee+wxxMXF1daya5VYE5RBMhEFAo1Gg59++snn43hrXiIKPEETKGs0GowZM0YK+oYMGQK73Y7hw4fDbrfj6aefRlxcHKxWK5RKJd544w1YLJY6GyQDYMoFERERkQtBEygDkII+MRgeOnQo7HY7RowYAYVCgSeeeALLly9HamoqPvvsM8TGxtbyiomIiIjIXwVVoCxSqVSw2+2w2WwYNmwYFAoFHnjgAWzatAn//vsv9u3bB7VaXdvLJCIiD5jNZiQlJQEAUlNTq/yJmKfjeGteIgo8QXsqQaFQQKFQwG63Y+jQobj55pthNBpx6NAhdOjQobaXR0REVWAymco0IfHVON6al4gCS1DuKIsUCgWsVitmzpyJXbt24ciRI2jbtm1tL8snWPqNiIiIqHqCdkfZUZs2bXDo0KE61UzEsfQbEREREXkuqHeUgdJ85XHjxpVpMlIXeFL6jbvPRERERBUFfaAMoM4FyYBnpd/YeISIiIioojqRekGu6XQ6hISEMEgmIiIiclAndpTJNaZcEFEgUCqV6NSpk3TbV+N4a14iCjwMlIMIc42JKJhpNBrs37/f5+N4a14iCjx8axxEWOmCiIiIyHsYKAcR5hoTEREReQ8D5SCi0+kQHx/PQJmIgpIgCEhKSkJSUhIEQfDZON6al4gCD3OUiYgoINjtdpw9e1a67atxvDUvEQUe7igTEREREclgoExEREREJIOBMhERERGRDAbKREREREQyGCgTEREREclg1QsiIgoICoUC11xzjXTbV+N4a14iCjwMlImIKCBotVocP37c5+N4a14iCjxMvSAiIiIiksFAmYiIiIhIBlMviIgoIAiCgBtuuAEAsH//fmi1Wp+M4+rxKSkplc6n1+uRmJhYpbUSUe1ioExERAHBbrfjxIkT0m1fjSP3eL1eD61Wi1GjRlX6/VqtFikpKQyWiQIQA2UiIvKptLQ0mEwml49xZ6e2NiUmJiIlJcWtn2PUqFEwmUwMlIkCEANlIiLymbS0NCQnJ0MQhEofq9VqodfrfbCqqklMTGTwSxTkGCgTEZHPmEwmCIKAdevWITk52eVjmdtLRLWtzgbKmZmZyMvLQ4sWLWp7KUREdU5ycjI6duxY28sgInKpTpaH++uvv3DzzTdj+/btMBqNtb0cIiIiIvJDdW5H+fTp0+jVqxdGjRqF0aNHIyIiosz9NpsNSqV77x8sFgssFov099zcXK+ulYiI/kehUKBp06bSbV+N4615iSjw1LlA+e2330a/fv3w6quvwm6347PPPoPRaERMTAweeOABKJVKt4PlxYsXY8GCBT5YNRERabVapKam+nwcb81LRIGnzgXKZ8+exc033wwA6Nq1K0JDQ5GRkQEAWL16NX777TcolUrY7fZKdw5mz56N6dOnS3/Pzc1FQkJCzS2eiIiIiHymzuUol5SU4MiRI3jrrbcQFRWFr776Cn/88Qc++eQT5ObmYtCgQQDc+3hNrVYjKiqqzB8iIiIiCg51JlC22WwAgHvuuQfp6enYuHEjbrzxRsTFxSEuLg5dunTBvHnz8O+//+K///6r5dUSEVF5BQUFuOGGG3DDDTegoKDAZ+N4a14iCjx1JvVCzDnu0aMH3n//ffz8889o0KBBmcc0bNgQVqvV7cN8RETkOzabDQcOHJBu+2ocb81LRIGnTkWEdrsdiYmJWLNmDdq3b4+tW7di8eLFAEorWPz444+Ii4tjCgURERERBWegbLPZYLVaK3xNoVDAZrOhVatW+Oyzz9CrVy+8+eabaNSoEfr164dVq1Zh9erVqFevXi2tnIiIiIj8RdClXpw4cQIvvvgiLl68iJYtW+LOO+/EHXfcAaVSCavVCpVKJQXLb731Fs6dO4dvv/0WiYmJ6NKlCzv1ERERERGAIAuUT548ia5du2LAgAG44YYb8O233+LAgQP44Ycf8Morr0ClUqGoqAhhYWEAIB3ku+6662p55URERETkb4Im9cJut+Ojjz5C//79sX79eixevBi//PILBg0ahJ9++gnjx48HAClI3rRpE9tXExEREZFTQRMoKxQKZGRk4OLFi9LXIiMj8fjjj2PUqFE4fPgwlixZAgDYunUrJk+ejJUrV/IEMxFRANHr9dDr9T4fx1vzElFgCYpA2W63AwA6duwIq9WKkydPSvdFRkZi3Lhx6NChAzZv3oyioiLccccdGDduHMaNG8dScEREAUKn08FoNMJoNEKn0/lsHG/NS0SBJyiiRLGL3u23346TJ09i2bJlyM/PB1AaRNerVw9z5szB77//ju+//x4AsGDBAjRv3rzW1kxERERE/i2oDvO1aNECn3/+OQYMGACNRoP58+dLH5WFhoaiXbt2iIuLq+VVEhEREVEgCKpAGQB69uyJDRs24P7778eFCxcwZMgQtGvXDh999BEyMzORkJBQ20skIgpKaWlpMJlMLh+TkpJS5fELCgowYMAAAMC3334LjUbjk3G8NS8RBZ6gC5QB4K677sJvv/2G6dOn46mnnkJISAhUKhW2bt2KJk2a1PbyiIiCTlpaGpKTkyEIQqWP1Wq1VToYZ7PZsHv3bul2VXk6jrfmJaLAE5SBMlB6sG/Tpk24fPky8vLy0LBhQ55YJiKqISaTCYIgYN26dUhOTnb5WL1ej8TERB+tjIio6oI2UAaAqKgoREVF1fYyiIjqjOTkZHTs2LG2l0FE5BVBUfWCiIiIiMjbGCgTEREREclgoExEREREJCOoc5SJiCi4aLXaWhnHW/MSUWBhoExERAFBp9PBbDb7fBxvzUtEgYepF0REREREMhgoExERERHJYOoFERG5VNOtqd1VWFiIe++9FwDw5ZdfIjw83CfjeGteIgo8DJSJiMgpX7SmdpfVasW2bduk274ax1vzElHgYaBMREROsTU1EdVlDJSJiKhSbE1NRHURA2UiojrKX3KPiYj8FQNlIqI6yJ9yj4mI/BUDZSKiIOPuTjFzj4mIXGOg7EV2ux0AcPHixVpeCZH/yc3NBfC/68QZ8f5ff/0VOp2uxtcVbEwmE0aNGoWCgoJKH6vRaNC+fXskJCRU+ljx/19tcuyOl5ubW+UKFJ6OU5158/PzAQAHDx6UbleHuJbKriMi8g6FnVeb15w7d86tFxyiuiw9PR1NmjRxej+vI6LKVXYdEZF3MFD2IpvNhoyMDERGRkKhUAAo3X1ISEhAeno6oqKianmFnuHafS9Q1w1Uvna73Y68vDw0atQISqXzpqBy15GvBdr/B663ZvnTet29jojIO5h64UVKpdLpO/yoqKhaf4KtKq7d9wJ13YDrtUdHR1f6/a6uI18LtP8PXG/N8pf1unMdEZF38O0oEREREZEMBspERERERDIYKNcwtVqNefPmQa1W1/ZSPMa1+16grhsI7LWXF2g/C9dbswJtvUTkPTzMR0REREQkgzvKREREREQyGCgTEREREclgeTgv8of6r0T+KpDqKBP5K15HRNXnST1yBspelJGRwY5iRJWorKMYryOiyvE6Iqo+dzpcMlD2osjISADAyZMnUb9+fa+M6e5OgLtnMrmzQLVF7G4mXifOOF5HDRo08MXSiAKGp9eRP3QT9EdWqxW//fYbAKBr165QqVR+MVZtjF8XuXsdAQyUvUoMQuvXr++1JyYGyhRsKvsdFO9v0KABX+CJnHD3OvKXboL+6I477vDLsWpj/LrKnZiIh/mIiIiIiGRwR5mIiIjqlOLiYqxZswYAMH78eISGhvrFWLUxPrnGhiNelJubi+joaOTk5DD1gqgc8fq4cuWKy+vD3ccR1UW8jrzDbDYjIiICAJCfnw+dTucXY9XG+HWRJ9cHUy/8gCAIMBqNEAShtpdCROSU2WxGZmYmzGZzbS+FiMgnGCj7AbPZjJKSEr74EJFf43MVEdU1DJT9gE6nQ0hICD9OISK/xucqIqpreJjPD2i1Wmi1Wq+OKQgCzGYzdDqd18cmorpJp9O5DJLNZrP0vMNgmoiCAQPlGqBQKCo9NOfu4buqHtJz/IjU8QXL22c3eTiQKPi5+7zh+Lzj6g06nzeIKFAw9SJI8SNSIvI1Pu8QUbDhjnKQ4kefRORrNZFGRlQT1Go1tmzZIt32l7FqY3xyjYEyERER1SkhISFeawvtzbFqY3xyjakX5BJrPBNRZfg8QUTBijvKAcTblSzcGc/dwzlEVDfIPW+4+zzh+L1MDaPaVFxcjE8++QQAMHLkyGq3sPbWWLUxPrnGHWU/4O5ujLeL/bszHg/nEAUOX3TOk3vecPd5gg1LyF8UFRXhwQcfxIMPPoiioiK/Gas2xifXuKPsB9zdjdHpdBXKvVWHO+PxcA5R4HBWFtKb5J433H2e8PZzGFFNSktLg8lkqvRxfI0MbgyU/YC7Lx7eDlodx2ODEqLA54tA1NnzkDvPIeL3so4y+bu0tDQkJye7lXev0Wh8sCKqLQyU/YA/7NoyF5ko8NVm7i+fQyiYmEwmCIKAdevWITk52enjUlJSMGrUKOnvR44ccRo46/V6JCYmen2tVLMYKAeJ6nbc40eiRHVPVTt/yuFzCAWj5ORkdOzY0en9er0eGo0GBQUFAIDu3bs7faxWq0VKSgqD5QDDQDlA1HRqhD/sahNR4HL13MTULgpWiYmJOHTokLTrvGfPHtkdZXHn2WQyMVAOMAyUA4S3P9bkCxcR+Ur5ahcsEUfBJCEhQbrdvn17/l4HGQbKAcLbH2syn5CIfMXx+csXlTmIKqNWq/H5559Lt/1lrNoYn1xjoBwg3EmN8GSX2N3AmzvPRN5hNpvr1E5q+ecOx+cPBslU20JCQnD//ff73Vi1MT65xoYjQcSTYv5arRYGg6HS4FduTLarJfJcXWu2YTabkZubi9TU1DLPFVqtFvHx8QyUiSggMFAOIjXRRU9uzLr2gk/kDXWty6VOp4PFYoFareZzBfmdkpISbNiwARs2bEBJSYnfjFUb45NrTL3wM9VJdaiJyhVyY7IMFJHn/DXloqbSq7RaLZKSkvhcQX7JYrFgyJAhAID8/HyEhFQ9HPLmWLUxPrnGHWU/483dWm+nSIjjAXArbYOI/F/55xxvPm9otVrpjTVTtYgoENXZtyXnzp3Db7/9hpCQELRs2RJt27at7SUB8O5urbOSTFUNcDMzM5GXlweVSgWDweC3O2RE5L7yzzmVVcRxdShRbnfa0wo7de3QIxH5tzoZKB89ehR33XUXDAYD0tPT0blzZ7zyyito0aJFbS/Nq+kTzkoyVXd8QRBqrLScOx3AiMg1TzrulQ9IHZ83xOvRcTxXzyVy98m9+Xe1PpaPIyJ/UucC5bNnz2LAgAF44IEH8Nxzz+Hnn3/GuHHjkJWV5XGgbLFYYLFYpL/n5uZ6e7nV4u2STOVPqvNFjLzB36+juqaynVxXn3rJ3efpm3+egSAif1LnAuXt27ejZcuWePHFF6FQKDBgwAB07NgRR44cwd9//42EhAT07NnTrbEWL16MBQsW1Mg6vX3Apio71XJ1UJmXTN5Wk9dRXVcTB/VcPQ+4+xzhal18niEif1LnDvPZ7XakpaXhyJEjAIBFixbh22+/xYYNG7Bq1SoMGzYMH3zwgVtjzZ49G1euXJH+pKene22dtVmCTTzMk5mZyTJwVONq8jqq62rjecSdw4AsMUlEgaLO7Sj369cPH330EYYMGYLrrrsOGzduxFdffYWBAwfCaDRi0aJF+PDDD3HXXXchNjbWZc6sWq2usXaSvvr40dXhGwBS3Vdv70yx4x+JavI6qutqI41BPPQbGRmJ+Ph42etcp9MhMzMTFouFzwFUK8LCwrB27Vrptr+MVRvjk2t1LlBu1qwZ1q1bh/379+PEiRNQKBS4++67AZTm4DZq1Ai7d+8uc5ClNvji40dBEJCamlomSBF3eEJCQhAdHS2twWg0VnqAz5Pg15uHC4mCVXUrQPg6jUEQBJhMJuTn50t552LDEcd1iGXj+BxAtSU0NBRjx471u7FqY3xyrc6lXgClwfKQIUPQpEkTFBQUoKioSLrv0qVLSEpKgtVqrcUV+obZbIZarZZ2dRx3ksvXSXanq5gn7a7rWpcyoqoItBQFs9mMevXqQaVSoV69egDg9Dov/xzg7brvRETeUOd2lB117doVM2bMwMqVK9GgQQMcO3YMa9euxc8//+yXAZy30hXEcQAgKiqqQtqF3M/uzs6U3Me8znaOeWCHqHI1kTrh+DxS1XGdPReJ40VFRUl/d/fgHz9lIl8qKSnB9u3bAQD9+/evVrc7b45VG+OTa3X6X/uaa67BV199hUceeQRKpRKNGzfG7t27/ab5SHneeiERxwkJCYHBYJC+Xt3gle2uibyrJppueKNOcU28AeZzBfmSxWLBnXfeCcA7Lay9NVZtjE+u1fl/7Z49e2Lfvn0oLi6GWq1GTExMbS/JKW+9kLgzjuOus/g9VXkB5M4xkX/xxvNIdcZwthvt+KmW49+JiGpTnQ+UASA2Nra2l+CW6gadji9QBoNBygmUC4LFHaOcnBzExMTAbDbDbrdDEATpII7cuL56cfOk8xgR/Y/j84i715EgCMjMzARQeuhZ7n5vHOQtfx+vcyKqbXXyMF+wquwwTPmDQa4OCmm1WoSEhECv1yMkJARarVZqXV1+fFfj8IAOUeAzm83Iz89Hfn6+VIlDvOYFQcDZs2eRm5vr1qFDVwd53Tnky+cUIvIl7igHAXE3R6xiIbdT45hKER0dLX09Ozu7TJ6ySC43UhAE5OTkQK/XV3iss49heUCHKPCU3yHW6XSIiIgA8L8De5mZmcjJyUFaWhrUajWKiopknwM86fDpzqdmfE4hIl9ioOwHqpK64Pg9cg1CyhODaHF3WCSWcHJX+RzuytbOAzpEta/8dVrZdVs+GNVqtUhKSiozXkFBAfLz82G326FSqdC0aVO3xqouPqcQkS8x9cIPVKVWavmT6yEhIYiPj69Q/1gkV7PUbDZLNZTd+ThTTMdwVtLJ2fc4WxMR+YYnaVdA5SkQYvc9i8WC+Ph4NG3aFAB8UjOdzylE5EvcUfYDVdkhcfwedz6ulKtZ6rjD7E7nPbl0DHEdACocDGSbaiL/UP45prLnHHefU+Lj45GUlCTlKYvtdatSMo7PF+RLYWFhWLVqlXTbX8aqjfHJNQbKfqAq1SyqWwHD0xdOkXh4xzEYdkz/cAy0MzMzkZ+fj4iIiDIf2xKRb5V/vqju80d8fHyZN85msxlhYWHSJ1Qid4NfQRCQmpoKtVotfY1BM9Wk0NBQTJ482e/Gqo3xyTUGykGqshcoxxdKT3Zyyle+cPz41vGgnyAIyMrKgs1mkw4BEVFF4kHcmmgu4kp1dnDlAu2ioqIKKRFiikZkZKTLN8viJ1xioM0De0TkLxgoBylPXmjk8hcdayWX30XOycmBRqORvj86Ohpms7nMQT/x70VFRbJ1V2uSN1r0EvmKNzrlVXdebwSjcs2aBEHA5cuXoVKpnAbm4vmIgoICJCYmlmk8Ut1/j9p6E0L+z2q14pdffgEA3HzzzVCpVH4xVm2MT64xUA5SrnKHnT3WcSdHTKkAyu4im81mWK1WZGVloV69erBYLBXGEP8u/rf8i6I3P1KVG6+2Ag+iqqitKg7l563OtensZ9BqtdBoNFLDErnylWKN5oKCAhiNRmk8ubKVInfXyucCcqawsBA9e/YEUNoWujq/H94cqzbGJ9cYKNcSb3eSKt/BSvxo1J1DenIH/RwvxPKBNABoNBoUFhZKL3zlP3ItP6a4PscXLsdd6aqS2xVzXK9cZ6/qBATsAEbeVhO7ne78npb//a/ODrOznOf4+HgIgiDlHovVLxyvS8dqOoWFhWWeH5z9HHId/OSu6/LPBbx+ichTDJSDnDu7VXINARw5fk0sJSfu9phMJpfzi2M77vYCQFRUVJnHiDtJBoPB7QOFYoBR/uer7KAS8x/JH/hbWoC3drbLP58kJSVVCGDFwFY866DRaFBQUIC4uDiEhIRAEAQcPnwYGo0GiYmJsjvVjulggPx1Xd1Di0REDJSDXE10unJsUiKXlwiU7RZotVphMpmg0WikknRAaZAt7nqnp6dDo9GUCX7FF0K5DoHieqtST5UNC8gf+FtagLeCSneeTwRBkDr6WSwW2Gw2qNVqaLVa6PV6/P333zAajbBYLFJQbDKZoNfrpTfTzkpV+sO/JREFDwbKdVz51taedtpz9sIkvlgCpafhHT96FV/4SkpKYDQacfnyZelx4u6yyWSSvsfZblJVXxC5y0T+IFACO09TleSeIxwDZ0EQcPLkSekxCQkJ0s6yOL5er0dGRob0d5PJBIvFIr25Fj/J0uv10jy8romoJjBQruNcNR4R73d8wXPk6oVJfLEUK2KIO8aOO8CCICAnJwc2mw06nQ6xsbEICwuTguQLFy4gJycHAMoc7NHpdFAoFNJ6+OJIgaimUy68ldpR3VSl8geLxXWJ5ST1er20YyzuGgOl6Rh2u116nhHvEwQBeXl5ACD7iRMRkTcxUK7jXDUeKV82rqSkBNnZ2ahXr16lL5qOQbQYjNtstjIVOPR6PcxmsxQoGwwGCIIgvVDm5ORAqVTK7iI5pl+4U/uZzQuorvFWaoenO9/lA+vyB4vFMcPDw8u8aRYrXwiCgIKCAhQVFUGpVAKA9AZb3HmOjIwEgDLf7+oaZ8lIIqoqBsp1XGUdu8rvKDvu7MqVnXNM5RCZTCZcunQJ4eHhUCqVMJvNuHLlClq2bCnlG5ZvgKLVapGQkOB0F8lsNuPcuXNo0qRJmRdyuRdJHt6juqiqqR1yh3s9uW6clZ0DAIvFAkEQoNFopDfLQGnJq5ycHCgUCtjtdpw7dw6FhYWIioqCyWSCIAg4d+4coqOjUb9+fTRt2rTMnJVd4/6WD061LzQ0FMuWLZNu+8tYtTE+ucZAmZySC6JFRqMRubm5MJlMaNq0aYUaxjk5OYiJiUFOTg6sVitCQ0MRHR0NADh79qy0Q3z11VcDgLRT5Fiz2WAwwGAwwGw2w2QyoaioSJpfzFEWX4RdvUgGSi4okTdVdfe0um8sy7/pFVtTR0VFQafTIT8/HwqFQnrcyZMn0aNHDxQWFmLmzJno0KEDioqKEBoaisjISFitVpw7dw5FRUU4efIkVCpVhZSLyq5xPgdQeWFhYZg5c6bfjVUb45NrytpeAAUO8aCdmGeYk5ODsLCwMjvIOp0OISEh0Ov10n8jIiKg1+uhVCpRUFAAvV6PHTt2YMOGDUhPT5c+khV3ksV8aZHJZEJmZiays7OlKhsGgwFqtVrakRbrs4prFANvoPSFuyrVMYjqIsfrqSocr0Ex7UpsTGQ2m6FUKhEREQGtVovU1FTcdtttyMzMRG5uLubMmYPXXnsNarUaKpUKeXl5KCoqQmxsLARBkAJnx+sbqPwaF+9noExEnuKOch3grRxdcacpIyNDqkhRfpdXbh6DwQC73Y6zZ88iPz8fH330Ed544w0AwFtvvYVJkyZhxIgRqF+/foVdMLPZjKysLGRnZwMALly4gMjISGm3ufycYtCdmZlZZqzy6xIDfnEcb/9bEQWq8jXP3emsKXbeE4mNiMSdXPHazM/Ph0qlgsFggNFoxN133420tDQ0b94cQ4cOxdKlS/H999/j2LFjmDNnDrp06QK1Wg29Xg+1Wo2cnBxERkZK6VeOAbOza9bxWo+Pj/f6vxcFnrS0NFy6dAl///03AODqq6+u0BY6JSXF7fGsVisOHToEAOjYsWONtLCuyfHJNQbKQcJVxylPPkot/0JY/hCMOIbVakVhYSHCw8OlQ3aCIEg1kxMTE8vMVVRUBJvNhm+//VYKkuPi4nDx4kXMnTsXX3zxBZYvX45evXqVWU9+fj6io6NRWFiIwsJCREdHw2azwWazSfdbrVbk5+cjPDxcapcLoMJBRMef32g0oqioSHoBrcq/FVGwcnYduPp6Xl4eFAoFVCoVLl++LOUh63Q6lJSU4Pz588jIyEB4eDhsNhuee+45HDt2DLGxsfjmm2/QqlUr9O/fH2PGjMHZs2cxefJkjBkzBrNnz5YqYMTHx5cpGWe1WqU0L6PRKKVzOL7ZFtO2TCYTA2VCWloakpOTK3wqIccxj96VwsJCdO7cGUDNtbCuyfHJNaZe1AGefJSamZmJixcvSrtD5V8YDQYDmjZtivr16yM8PByXL19Gdna2lDrx33//SWWeyjt8+DBeeOEFAMDjjz+O06dPY+HChYiMjMRff/2F/v37Y/DgwdK7fFFRURHq1auHpk2bSvmJjsQScsD/ntj0ej1UKpUU4Jf/+Q0GA8LCwqQg2bE7YHU+diYKBs6eM1x9PTIyUtrtFd8wixVrQkJCEBsbi+joaBQUFOD555/Hjh07oFar8eWXX6JVq1YAgJtuugn79+/HiBEjYLPZsHbtWowcORJnz56tsEatVguVSiV18wMgPQ+JxDfKNpvNrYCHgp94OPTdd9+VvrZnzx4cPHiwwp+UlBQkJibW4mrJH3BHuQ5wTEGw2+0ef79Y71Qk7tqYzWapkoWYVyy+EJb3zTff4KGHHkJJSQmGDh2KxYsXQ6lUYubMmRgzZgxefPFFvPvuu9i8eTO2bduGBx98EAsXLoRCoUBYWJhUNUPMVRTnEAQBarUaWVlZ0t8BSMGyWF5K3B0XK3WUT7kQ3xCEhISU+ToR/U9ltdPF+7KyshAeHi61qdZqtdKnTIsXL8aWLVugUCjwwQcfoGvXrmXGiY6Oxtq1a9G/f39MmTIF+/fvR7du3TBv3jzcc889LtcgPleJaRmCICAmJkbq6EckEg+SA0D79u25OUJOcUeZyoiPj0eDBg3KfETp2Kba8aBOfHy8dChH3Mk1GAxITEws00Rg//79mDJlCsxmM3r16oU1a9ZIAaw456uvvoqDBw/irrvugtVqxbvvvovWrVvjrbfeQm5uLsLCwiAIAlJSUmA0GpGVlQWr1YqCggJcvHgRNpsNWVlZMJlMSE9Pl93RLl8X2lF1DzARBRNX14oj8fkgMzOzzG5uXFyclCbh+PUffvgBa9asAQAsWbIEgwcPdjr2sGHDcPDgQdx0003Izc3Fk08+iWnTpuHMmTP466+/kJ6eLlXEuHTpErKysqTnKrF6DoAKh4OJiDzBHWUqo/xujbN2tOJhOQBSEKvX68t8TGUymVBcXIwxY8bAaDSiQ4cO+OyzzxAWFiY7d+vWrbFx40bs3r0bM2fOxOHDh7FgwQIsWrQIjRo1QlxcHBo0aIDGjRvjuuuuQ0JCghScl5SUIC4uDtnZ2SgsLJTaYgOQXrBdlYhi+1ui/3G3nJpjq3rH9vRhYWFl/i6WgXvssccAAJMmTcLUqVMrXUezZs2wa9cuvPjii3jhhRfwzTffYPfu3RgwYADuvvtu9OjRAwBw+fJlRERElAmKBUGQStIREVUVA2VyyVngbLFYyrxAApDyER0ff/nyZZw+fRoA8P7770sdtVy59dZbsXfvXqxfvx5z585FWlqa9EdOSEgImjRpgh49emDy5MmIjo6GxWKByWSSdsSSkpIYDBM5UdUmI46t6rVaLex2u9SaXq/XS+MIgoADBw7AYrHgmmuuwYoVK1weQHYUEhKCuXPnok+fPhg7dizOnDmD9evXY/369ejQoQMGDRqE3r17Iy4urkKreyKi6mLqBbkkV5cY+F+qguMJ9MzMzAopD1qtFvXq1QMA2dxlZ5RKJUaOHInTp0/jzJkz2LVrF9auXYv58+dj7Nix6NmzJ5o3b46QkBCUlJQgNTUVH3zwAfr374+ffvoJTZo0kdI7srKy3DrhTFRXuZtq4YqYaiXmBZe/r7i4GADQsGHDKpW36tq1K44fP46vvvoKgwYNQkhICA4fPox58+ahd+/emDJlCj7//HPk5uaW+bnEuu9ERFXBHWVyqXzVi/KH3sS6xQUFBVJtZfEQDVB6uCY2NhZZWVm4cOECOnTo4NH8SqUSCQkJSEhIQPfu3aWvi7tRVqsVGRkZOH78OJ566ikcP34cU6ZMwYEDB/Dss89Kh/1Y7o3Iuap2risfYDvLC9ZqtSgsLJTmqqrQ0FDcdddduOuuu2A0GvHpp5/iww8/xF9//YUtW7Zgy5YtaNiwIcaMGYPx48dLpenETp5EopCQEMybNw+Ad1pYe2us2hifXOOOMrkkd8gtJydH2qkBSp9wEhISEB8fLx3iE18w4+LipIoZFy9e9Pr6VCoVEhIScNttt+GPP/7Ak08+CYVCgQ8//BC33nor/vvvP1gslgrNCYCKu+XOds+Jgp1cZzuj0YgTJ05I17kcx+eH8p8ylXflyhVpLm8wGAyYOnWqVMrrscceQ2xsLC5cuIAlS5agXbt22LZtGy5cuOCV+Si4hIaGYv78+Zg/f77TczPuCgsL89pYtTE+ucZAmVySewGNiYlBQUGBFAyLpdjEgFgQBBQVFUmVMJKSkgCgxl+wwsPDsXTpUuzatQvNmzfHhQsXMGbMGLz++uuw2WwVPn4tvxtmNpuRm5uL1NRUBstU5xmNRlgsFpeBstzzg/hG2fEaEvOWAe/nDisUCnTo0AGvvfYazp8/j88//xydO3dGfn4+pk2bhiNHjvB6Jr+RkpKCQ4cOufzj7DwO1Y46nXpx7tw5qRpDXeHuARpn9ZbFj2gd6yo75iU7nngX7xNLzbmzo+zurnN4eLjT+9q0aYMff/wRCxcuxNq1a/HRRx9hz549eOutt3DrrbcCKM2XNhqNKCwsRKNGjQCUvuibTCaEh4c7TdWorA61py2w3f3/QeQJud9TT383xdQqV/WHHceUq4Ah0mq1UjdNnU7ndj13cRe6MmLwrVAoMHDgQPTu3Rt33nknfv/9d8yePRuffPIJGjZsiOLi4jJBc/kufiJel8HPZrPh+PHjAIDk5OQyJUurMpbY8trZWOLh1lGjRlU6nlarLdPsxJ3xqebU2UD5+PHjGDBgAO6//3506NABNpuNv3weEF9sjUYj8vLycOHCBemQTlFRES5cuIDs7GyEhoYiOjoaQM3vKDuKiIjAsmXLcO+992LChAk4c+YMbrvtNjz55JOYM2cOBEGQWmFnZGTAYrFI5e3EclZVwRbY5K88/d0s35SnsjHFYDkqKqrC+GIXPcA31Sh0Oh2+/vpr3HbbbTh8+DDGjh2LnTt3IiwsDPn5+bBYLGjUqBFzl+swi8WCG264AUD120IXFBTg2muvdTlWYmIiUlJSZGv8O0pJScGoUaNgMpmkQNmd8anm1MnI8M8//0Tnzp0REhKCTz/9FJmZmQyS3eRYR1n8SNZisUgl2cR3zfn5+SguLkZxcTHq168PwLeBsqhv3744ePAgRo4cCZvNhuXLl6Nbt25IT0+HTqeDUqmEzWZDXl6e9KJpMBiq/ETExiXkr5z9brqTi+zOmGKqlVwQLggCLl26BMB7OcqViY6OxqZNm3D11VcjIyMDAwYMKLMGNiIhX0tMTETHjh1d/klOTq7tZVI5dS46/PPPP3HTTTfhiSeewL59+xAXF4d33nkHdrvd4/bOFosFubm5Zf4EO/GFEfjfCffExETUr18fiYmJ0gumwWBAfHw82rVrh5YtWwKonUAZAOrVq4f3338f//d//we9Xo+jR4+iZ8+e+Pzzz9GwYUPExcVBpVJJHyOX50mJKbmcTXKtLl5HtcHZ76Y7ucii8gde3fl9FwQBJ0+elNKqfPkm0mAwYNu2bWjWrBn+++8/jB07Frm5uVKjInEXnCXkiMiZOhUo//XXX+jSpQumTZuGRYsWITY2FsnJyfjmm2+gUCigUCg8CpYXL16M6Oho6U9CQkINrt4/OAbB4o5M+V1YnU6H5ORktG7dGgDQoEEDAMClS5ekElG1YdCgQTh48CB69eoFQRAwe/Zs9O7dG++++y7S09ORl5cn+7GY4+Ekb1XGEMfhi3PdvI78icFggFqtdppm4fg7X5V6y4IgwGq1SteMWEayKgRBQF5enkff06hRI2zbtg2NGzfGyZMn8eCDD2L37t1ITU2VquGUP3xIRCTyOFA2m82YM2cOunbtiquuugrNmzcv88efWSwWzJo1C4sWLZJykl944QWcOnUKb775JgDPDnHMnj0bV65ckf6kp6fX1NL9jhgwu9odMplMuHTpEnQ6HRo3bgybzYZ3333Xh6usqEGDBti6dSsWLVqE0NBQ/P3331i8eDFuu+02jBgxAmvWrMHhw4dx6NAhpKSkSPmc4i76iRMncObMGWRmZlZrHd5o8BAs6vJ15A8MBgOuueYa2UBZEASkpqYiNzdXykOWS99w9QZSq9UiIiICzZo1AwBs2rSpSussLCxE9+7d0bJlS4wePRrfffed1MSkMs2aNcO2bdvQoEEDnD59GmPHjsV7770nXX9iRQ4iovI8DpQffvhhvPfee7j55pvx2GOPYerUqWX++LMbbrgBCxcuBFDayMJut6NBgwbo2bMnfvrpJ1itVo92lNVqNaKiosr8oYqUSiUef/xxAMDy5ctrPThUKpWYMWMGUlNT8cYbb6Bnz55QKpU4duwYFi9ejOuvvx6jR4/Gxx9/LAUHYhBhtVrd6jBY2c4zc5n/h9eR/zKbzVCr1bBYLGXSqsqnWohv/MqXhBM/odHr9Zg2bRqUSiW2b9+Ov/76y+O17N69G2fPnkVxcTG2bt2KkSNHol27dpg7dy5OnDhR6fe3bt0ahw4dQp8+fSAIAhYtWoQxY8YgJSWlWrvcRBTcPA6Uv/32W2zYsAFLly7FE088EVCBcnkKhQLR0dF44IEHsGHDBuzdu5dlgbxEDIYjIyOh1+sxduxYJCUlITMzE2+88UYtr66UXq/HQw89hO+++w7nz5/H6tWrceutt0KhUODEiRNYsmQJpk2bJgXGWq0W9evXR1JSklTyzpnKdozd2ZEn8obqpAvpdDpERUUhKSmpQnDsOG75ZiOCICAtLU1KZxIEAQ0bNsTgwYMBACtWrPB4LVu2bAEADBw4EJMnT4bBYEBmZiZWr16Njh07onv37nj77beRnZ3tdIwGDRrgu+++w6JFi6BSqbB9+3aMHTsWFy9e5LkCIpLlcaBcr149xMbG1sRaas2dd96Jvn374s0333Rrt5AqJ9ZTFm+HhoZi7ty5AICXX37Zqx912mw2LFy4EBMmTMC6detw7tw5j8eoX78+Hn30UezatQvp6emYP38+VCoV1q9fj+7du+PIkSMAgKSkJNmgoTzuGJO/qE6aj6vDeuXLzTk+Tmwdb7FYyuQVz549GwDwxRdf4MyZM26vo6SkBN9++y0AYNy4cVi4cCGOHj2KdevW4fbbb0dISAgOHDiAqVOnIikpCaNGjcKvv/4qO5ZSqcTs2bOxe/duNG7cGKmpqZgxY4ZUvo7qhpCQEMyYMQMzZszwSgtrb41VG+OTax4Hys8//zzmzp0bVAcfwsLC0LNnT2zevNntAvfkmmNeb0lJCQoKCjBkyBC0bNkSOTk5WLlypdfmeuONN/D6669j48aNmDZtGjp06ICbbroJTz/9NLZs2VKlwz9z587F999/j7i4OBw+fBj9+/fH1q1bAfxvJ0384ywvk9UvyB/U1Js2x3HlqmGoVKoKv/+tWrVCt27dYLPZPHoO+O2335CdnY3Y2FjcdNNNAEqDhwEDBuDjjz/Gf//9h+XLl6Nt27awWCz44osv0KdPHyxevFhqdFJe165dsX37dkRHR2Pv3r0YNWqU08dS8AkNDcXy5cuxfPlyr7Sw9tZYtTE+uaawu5GU26FDhzIpCf/88w/sdjuSkpIqvLs5dOiQ91dZg+x2OxQKBbKzs9G3b1988cUXUstlT+Xm5iI6OhpXrlypM3mW7naqs9vtEAQB69evx7PPPouIiAicOnWqwgEid3e9zp8/D6C0kskDDzyAkpIS3HnnnTh37hyOHj0Kq9UqPTYkJATt2rXDTTfdhK5du6JNmzYVdo/i4uJk50lPT8e4ceNw/PhxhISE4KWXXsKQIUNgtVpx8eJFhIWFQaVSIS4uDlqtFhqNBgC8vjsVDClB7l4fdfE6qi2O16+nnfscGY1GlJSUICQkRLqmxa+J+c1iKbYdO3Zg1KhRCA8Px7///usyjUksKTdnzhx88MEHGDp0qGzahhig2+12nDhxAh9++CG++eYbAKW11JcsWYKIiAipC6ejn3/+GYMGDUJxcTGmTJmCF1980e03FbVxXfI6qp5Dhw7h+uuvx8GDB9GxY8faXk4F/r6+YOHJ9eFWZ75BgwZ5Y11+SXyii4mJwe7du/lRuZdptVpotVpkZmYiPz9fOrV++vRpLFu2DMuXL6/y2Lm5uZgxYwZKSkpw2223YcmSJVAoFMjLy8O+ffvw66+/4rfffkN6ejoOHTqEQ4cOYfXq1YiNjcXAgQNxzz33VFqpJSEhAd988w1mzJiBr7/+GlOnTsXevXuxePFiaDQa6XCfeJBJDJSJAk35FA1PgmYxCBZ3l8UxLBZLmfEBoE+fPrjuuuvw559/YtWqVdIBa2dsNhu+++47AMCAAQNcPlahUKBNmzZYtmwZbrjhBixYsAA7duzAf//9h9WrV8sGyrfccgtWr16N8ePH4/XXX0fTpk0xffr0Sn9mCmw2mw2pqakASnsBVLeFdVpamlfGqo3xyTW3dpTJPXXxHby7vz75+flIT0/HhQsXcPz4cUydOhVqtRr//POPVGcZcH9H+dy5c5g+fTq+//57NGnSBF988QUiIyMrPM5msyE9PR2///479u7di71795ZJxbjuuutwzz33YPjw4bLf7/hzvv3223jhhRdgs9nQqVMnvPbaa2jQoIEUSHBH2TXuhPkfZzvKYtDsuEMsR24X2nF3GShNvcrJyUFMTIw03uuvv46pU6ciJiYGZ86ccXrtXbx4EYcPH8bAgQMRERGBw4cPIzw8XHYdco4cOYIpU6YgMzMTkZGR+OCDD9CvXz/Zxy5duhQvvPAClEolNm3ahNtvv93pzy3ijnLgEXds9+zZg+7duwOofltos9mMiIgIr4wlt6PszfGplCfXB9+WUI1xzFvUaDSIi4tD06ZNpfQHi8WCVatWVWnsDRs24Pvvv0dISAiWL1/uMshNSEjAkCFD8PLLL2P37t1YuXIlevToAZVKhT///BPz589H+/bt8cQTT2Dv3r2ywb9CocDEiROxdetW1KtXDwcOHMCgQYNw7NgxKe1CEAQeBqWA5ZhX725es7ODguJhXXEcvV4Pi8UiNfi4//770axZM+Tk5FRaW108xNezZ0/ZINmV9u3b48svv0SHDh2Ql5eH++67Dy+99JLsNT5r1iyp1f3QoUPx+++/e6W5EBEFtipXvSj/Jy4uDo0bN8att96KtWvX1sRaKcA4vogWFBRAEAQpn3fy5MkAgLfeegtZWVkejZuZmYklS5YAAKZNm4Z27dq5/b2hoaHo1asXXn/9dfzwww+YPn06mjVrhoKCAnz++ecYPHgwunXrhldffVW25XafPn2wd+9eXHvttcjMzMS9996LVatWIT8/n929KGi4exjVWUAdExMDk8mEs2fPAoBUClGtVsNsNsNgMGDGjBkAgFdffdXlG8wdO3YAqDztwpn4+Hh89NFHGDp0KOx2O+bPn4/BgwdXqLqhUCjw+uuvo2fPnjCbzRg8eDDOnTtX63Xfiah2eRwoz507F0qlEnfccQcWLFiABQsW4I477oBSqcTkyZPRqlUrPProo3jnnXdqYr0UQMqfig8LC4NWq4Ver8egQYOQnJyMK1euSCWj3PXff//BYrHAYDBg9OjRVV6fXq/Hgw8+iG+++QabNm3CiBEjoNPpkJqaimXLluHmm2+W7SLWvHlz/PLLL7jvvvtQXFyMadOmYdiwYTh//jyrXFCdIhdQi9e9IAiwWCxS0xHH54OCggJ07doV9evXR0ZGBl577TWnc4iVKKpTFissLAwLFy7EypUrERYWhh9++AFdunTBsmXLyuRRh4aG4uOPP0abNm1w6dIljB8/HhaLhTvLRHWYx4Hynj178MILL+Djjz/GlClTMGXKFHz88cd44YUXcPDgQbzzzjtYvny5yyc+qhvKv4iKpfc0Gg0MBgOWLl0KAFi7di1++eUXt8eNjo4GABQXF3vlUINCoUCnTp2wYsUK/Pnnn1i5ciXat28PQRAwceJELFiwoEKr3IiICHz66ad45ZVXoNFo8PPPP6Nnz5547bXXPN6Bqk5DCCJfq+z3VbzuExMToVarodfry3wdKK0mk5eXh2HDhgEozQ++dOmS7Hi33norAGDnzp3VXvu4ceOwd+9e9OjRA4WFhXj++efRo0ePMp9qRUdH4+uvv0ZMTAwOHjyIGTNmoLi4mDvLRHWUx1HG9u3b0adPnwpf7927N7Zv3w4AuP322z0qJk91gxjginr06IHhw4cDACZNmlRmZ8cVMfE+Ly/Po5bj7tBqtbj//vuxadMmKT3k7bffxtChQ5GZmVnmsQqFAo899hgOHz6MW2+9FYIg4LnnnkO/fv3w559/Og0mHAMNQRCQmpqK3NxcvhCTT5nN5iq9QXO3gYnBYEBycjIMBkOZ33mz2YywsDCYTCb0798fLVu2RF5eHhYsWCA7Tq9evQCUBsreuN5btmyJTZs24f3334fBYMCxY8cwbNiwMukfLVq0wPr166FUKvH555/jk08+4QEqojrK40A5NjYWmzdvrvD1zZs3Sx37zGazy8NVVPeIDUgcP6LVaDSYN28eDAYD/v77b7fb2ooBt1iarSaEhITg2WefxXvvvYeIiAjs3bsX/fv3x549eyo8tkWLFvj++++xatUqREREYN++fbjpppuwZMkS/PvvvxUCEbPZjNzcXKSmpiIzM1PqYMYXYvKlqnbsq0oDE8e5dDodIiMjkZiYiIiICEyaNAkA8P7778s2fLrxxhuh0Whw6dIlnDhxwqO1OqNQKHD//fdj69atUsORhx9+uEz99f79++PFF18EADz77LM4ePCgV+YmosDicaA8Z84czJw5EwMHDsQLL7yAF154AXfffTdmzZqFefPmASg9fCF+XEYk7pw6lk8TNWrUCM8//zwAYPHixfjnn38qHU/s+gWUlsqpSQMGDMC3336L1q1b49KlS+jbty9ee+21CjtbSqUSEyZMwJEjR9C3b18UFhbilVdewfDhw/HDDz9IwbK4s3bmzBnk5+fDZDJBpVLBYDBIFQGIfKGqHfsqO+jnrHNl+bb1TZo0QatWrTB8+HAkJCTAarXiwIEDFcYLDw+Xynj9+OOPHq21MsnJyfjss88QFhaGTZs24emnny5zbc+cORNDhw5FSUkJ7r//fqSkpJT5GfkpUOBSqVSYNGkSJk2aJJUyrKqQkBCvjVUb45NrVaqj/Ouvv2LVqlU4efIkAKB169aYMmUKunbt6vUFBpK6WLfSnV8fua5d5cfo06cPdu3ahVtvvRXfffddpfVJGzRogJycHHz55Zdo0aKF08cdPny48h8CqPQQXmFhId5++2389ttvAIA777wTixYtkg0y7HY7du7ciVmzZuHKlSsICwvDQw89hPHjxyM0NBRWqxWFhYUoLi5GgwYNpE9f3KlbWxl/rrfM+q+By90OnGIN5ry8PFy4cAENGzas8PvteFvMXx40aBA2b96M559/Hs8884w0bnZ2NgDggw8+wIwZM9C5c2ds27atwvzu5i876wK4c+dOqfHJxIkT8corr0j3mc1m9O7dG3/++SdatWqF7du3Iz8/H2FhYYiJiXHZWbAm8DqqHn/vfOfv6wsWNV5HuVu3bli/fr3U7Wz9+vV1Pkgm5yrbuVIoFFiyZAk0Gg12796NTz75pNIxHYuv+0J4eDgef/xxzJkzByEhIdiyZQvuu+8+/PfffxUeq1AoMGbMGBw6dAi33347ioqK8Oabb2LEiBE4c+YMdDod4uLi0KxZM0RGRkrdC8unphD5O3FnNTMzs0xqhcViQXR0NC5cuIC0tDSYTKYKtZUdf9fbt28PAPjjjz9k5+nbty8A4MCBA7h8+bLXf45evXpJKSBvvfUW/u///k+6T6fT4csvv0SjRo1w6tQpPPDAA1AoFCgqKmK6FFEd4HGgnJaW5vIPUXnu1GRt06YNnnzySQDAU089JZWUckbcpXLsslfTxAB43bp1iI+Px+nTpzFo0CB8//33so9v3LgxvvzyS3zwwQeIjY1FSkoKxo0bhzNnzqBJkyaIi4uT6suKnc0c2/+Wx+oY5G/E3GMA0pthrVaLpk2bon79+gBKd27Onz+PmJgYAJBKRIq/7yaTCV26dAFQGijL7V43btwYbdq0gc1m83r6hWjIkCG47777AACPPPIIdu/eXWb+jRs3QqfTYc+ePVi0aBEaNWrEQDmA2e12KT2ouodEvTlWbYxPrnkcKCclJaFZs2ZO/xC5yzHw02q1ePLJJ3H11VcjKyur0trKVdlRzs3NxcqVKzFr1iy8/PLLWL9+PXbt2oW///4bOTk5bj8BderUCd988w1uuOEGmM1mTJo0CS+99FKZg0AihUKBYcOGYe/evWjfvj1MJhNuv/12vPnmmxXmEwTBZdOSqh6+Iqop4u5wfHx8mTfD4pvj2NhYqFQq1KtXr8KnSoIgIC0tDXl5eWjbti1CQ0NhNBqRmpoqO5dYbemHH36osZ9n0qRJuPXWW1FcXIwhQ4bg2LFj0n3XXXcdPv74YyiVSqxfvx6rV6+usXVQzSssLER8fDzi4+OrvfkgCILXxqqN8ck1j7PCy+d8FhcX4/Dhw3j55ZexaNEiry2MgptYFk2tVgMofWEtKirC4sWLMXjwYHz88ccYOHAg7rrrLtnvF3eUz5075/acL7/8stQpzGg04ujRo2XuV6lUiIqKQlRUFKKjo9GkSRN069YNLVq0qJD7azAY8NFHH2HZsmVYu3Yt3nzzTRw8eBAvvvgikpKSKsydmJiInTt34tFHH8X//d//Ydq0aUhJScGKFSukoF/cYXO28y7mf3IXi/yF+Lsqvnkr/7vbtGlT6PV66XdWfJxGo4EgCFLFl/r16+O6667DgQMHsHfvXtlNl759+2LlypXYuXMnrFardKDXm5RKJZ555hmUlJTg119/xd13342dO3eiadOmAEoP97700kuYNm0a5s6diw4dOuDOO+/0+jqIyH94vKN83XXXlfnTqVMnPPLII1ixYgWbjJDbzGZzhbJoOp0OXbp0wbhx4wAAo0ePxt69e2W/v0OHDgBKaxwfOnTIrTkLCwsBAF26dMGYMWPQr18/tG3bFgaDAQqFAlarFdnZ2Th79iz++usvbNu2Dc8++yymT5+OjRs3VqijHBoaimeffRYvv/wyNBoN9u3bJ+0Wl29QApQGER988AEWL14MAFizZg3mzJkjBQ86nU5KxZDjblthIl9y9kmH4wE/rVZb5nFiJRyVSgW9Xg9BENC5c2cAwM8//yw7T6dOnRAREYHs7Gz8/fffXll7VlZWmWYjAKBWq/H5558jOTkZGRkZGDBgADIyMqT7J06ciAcffBAAsHLlSq+sg4j8V/Xbmv1/rVu3xv79+701HAU5nU6HqKgoafdVLLUkCAKmTJmCnj17oqCgAIMHD5Z9URwxYgR69eoltZB29nGtozvuuAMAcOLECXTv3h3Dhw/H9OnTsWzZMnz88cd44403sHjxYsyePRuTJk1Ct27dEBYWhoyMDPzf//0fpkyZguHDh+Ozzz5Dbm6uNO7AgQOxbds2dO/eHUVFRXjppZfQtWtX2etBoVBg2rRpWL58OYDSF9qFCxfi7NmzMBqNOHv2LM6ePcv0CgoYzg7rioFxZmYmUlNTYTKZpDfGgiAgLy8P586dQ3p6Oi5duiQd6HNWwSLk/7V333FNXe8fwD8hQCCRHUBBhgtFnLiqVXFr66gD96yzah1oVbTWUbfirNZt1eK21r2qqHV83VIcaBVFHCgJyJCEmfP7g9+9TSCMIJAEn/frxauS3JycpPeQJ+ee8zympqhfvz6A3Df96eLixYvo27cv+vfvj7CwMI377O3tcfz4cVSoUAEvXrxAx44dIZPJ+PsnTZoEICtdHe3NIaR00zlQTkxM1PhJSEjA48ePMXPmTFSpUqU4+kiMVF6bz9RnR7kPVC5YTkhIwMKFC+Hr64u4uDh07tw5x+Y+oVCIBQsWoGbNmkhISMD333+f7274Jk2aQCwW8x/Q6szMzODg4ICKFSuiTp068PPzw/jx47Fx40aMHj0aNWrUgEAgwK1btzBz5kw0atQIY8eOxV9//YW0tDS4ubnht99+Q1BQEOzs7HD//n34+flhypQpWtdRjxs3DtOmTQMABAUFYceOHXxZ36SkJFqHRkpEcnIyYmJiPumLWW5XOrgAGsjaS5CZmcnPLovFYiQkJEAoFPIp4GrVqgUTExM8e/Ys1+CzQYMGAPBJkzKMMezcuRNz5sxBWloaUlJSEBgYyOdI5ri4uODUqVNwdXXF48eP0bVrV756aOXKldGiRQswxrBly5ZC94UQYvh0DpRtbW1hZ2fH/9jb26N69er43//+h/Xr1xdHH4mRyuuSbGRkJCIjI6FQKDTSRSUlJSE9PR1WVlbYv38/KleujFevXmHIkCE5NstZWlpi9erVcHV1xevXrzFx4kStSx44QqGQX/tY0BLrYrEYLVq0wE8//YR169Zh6tSp8PLyQnp6Os6cOYPRo0ejffv2ePDgAQQCAbp27YrTp0+jT58+UKlU+OWXX+Dr68uXd1c3Z84cjBw5EowxLFy4EDdv3oSVlRWfMo6Q4qbLBlFthUQK8mXYyckJZcqUgVAo1CibbW9vj8zMTLi4uMDZ2RnOzs6oWbMmgNwLi3DZMW7evFmo15uamooFCxZg27ZtAIAePXqgTp06UCgUmDJlCp4+fapxvIeHB06ePAmpVIq7d+/ylfoAYNiwYQCArVu38pk/CCGlj86B8oULFxASEsL/XLx4EY8ePUJERAQaN25cHH0kRiqvS7LczGlycjL/gcqVtuV2yzs4OGDjxo2wtLTEuXPnND6kOPb29li3bh2sra0RFhaGtWvX5tknrjhJRESEzq/HwcEBI0eOxIkTJ3D06FEMHToUDg4OePXqFXr37o3Dhw/zx23fvh1Hjx6Fu7s7oqKi8M0332DkyJEagbxAIMC6devg7++PtLQ0TJ06FS9fvqQgmZQYXarzcUH1y5cv8e7dO34mOrcvw1xAHRMTA6VSCYVCAZVKBblczgeWFStWhFQq5X9atGgBIPflF/Xq1YOJiQnfB10kJCRg0qRJOHfuHIRCISZPnoxx48Zh4cKFqFGjBj5+/IjJkyfn+BLt5eXF/10JCgriqwd26dIFjo6OePv2LU6cOKFTXwghxkPnrBdUmpoUFHeJFdCsGMcFxNy/ufvUb7e2tkZycjKqVauGoKAgjB07FgsXLsSXX36JDh06aCzzqVKlCjZt2oQ+ffpgx44d6N69O59KKvsGvIYNG+Lo0aOIioqCq6srf/uQIUMK9JrMzMw0fre0tISFhQVSUlLwww8/YO7cubCzs8OuXbtgZ2eHzZs3Y+vWrTh48CB27tyJ169f80VLuP5s27YNcXFxCAkJQY8ePbB+/XrUq1cP5cuXz/V5c1PQFHeGXMGPlBwuh3dBj+W+2HJXd3LLxMIF0B8+fEBmZibi4uJgaWmJlJQUvvKkpaUl4uPjIZVKIRAIIJFINDJbcO1n70ONGjUQFhaGsLAw/otvfhkwoqKisHDhQshkMggEAlhZWeH333/H77//DgBQqVQwNTVFYmIihg8fDgsLC7i7u/OPL1++PNq2bYu//voLAwcOxNatWyESiTBkyBAsW7YMmzZtQpcuXXJ9fhpvhkcoFGLw4MEAUCQlrIuqLX20T/JWqM18ERERGDduHNq0aYM2bdpg/PjxhZqhI58niUTC591W/yCUSCRwcnLSyILB/YEYNmwYGGMYMGAAn+JNXefOnTFixAgAWcUC3r9/r/W5q1atCiArrVxRFCsRCoVwcnKCjY0NgKwCKO/fv+fXXYrFYowbNw6LFi2CqakpQkJCsHjxYo1lJNwu+/r16yM2NhZjxozBw4cP+UvbsbGxOq1ZpsIkRBcFXafMXfnx9PRE2bJl4eTklO/6ZEdHR1hZWfFLLxwdHfkfiUTCFyHhztlKlSrBwsIC7969y7FmmPPFF18AAP73v/8V6PXduXMHM2bMgEwm43M6m5ubaxxjYmICW1tbmJqaQqVSYcKECXjz5o3GMRMnToS9vT0iIyOxdetWAOD/5pw6dUrr3yViuMzNzbF9+3Zs376dT1NaWCKRqMja0kf7JG86B8pnzpxB9erVcfPmTdSqVQu1atXCjRs34OPjg7/++qs4+kiMnEKhKPCGodwu5c6ePRs1a9ZEXFwcevbsyW+qUbdo0SJ4e3sjJiYGo0aNgkqlynGMjY0NXFxcAABPnjwp5CvSJBAIYGtry6eZS01NxfDhwzWydTRp0gRz586FUCjEmTNnsGzZMo3+WVlZ4ciRI6hSpQrkcjlmzpyJFy9e4OHDh5DL5YiJiSnQulCACpMQ3eh6vnDBMYB81ydza5QtLS01guLsWTC4PohEIn7DXm7rlLlAObfUkRzGGI4fP47FixdDqVTCx8eHL3yiDRcsC4VCyGQyTJgwQWN5h42NDaZMmQIA2Lt3Lx48eIDKlSujdevWYIzxwTMhpHTROVAODAxEQEAAbty4gRUrVmDFihW4ceMGJk6cyO/iJ0SdLh/E6msm1R9nb2+PX3/9Ffb29rh9+7bWc83S0hI7duyAhYUF/vrrr1zXK1erVg0AiiwXK0csFqNs2bIwNTVFTEwMxo4di5MnT/L3N2/eHLNmzYKJiQlOnDiBlStXaiyVkEqlOHToEGxtbREWFoaZM2dCLpcjIiKCr9qXnJyc7/upy7pTQgp7vhR0XHM50+Pj4/nZ6+xZMNQr/HXo0AFA1n4Ybbi9MKGhoVAqlVqPycjIwKZNm/Dbb79BpVKhVatW+Omnn2BikvdHHhcsu7m54d27d5gwYYJGxp1mzZqhffv2UKlUWLBgAZRKJT+rvG3bNtrUZ0QYY/zf06IoYV1UbemjfZI3nQPl8PBwfrevuqFDh+LRo0dF0ilSuujyQax+KVf9cWKxGF5eXlizZg0EAgE2b96MnTt35ni8j48PlixZAgCYNWtWjvyowH+B8j///POJrywnc3NzlCtXDk2aNOErDarPNLVq1QozZsyAQCDA4cOHMX/+fI3He3l5ITg4GCYmJjh58iQOHz7Mv37uvcjv/dR2OZyWY5DcZF/ypMvj8joPuXMO+G8NcWZmJmJjY/Hu3TuNYkPq52zr1q0BZG3oU89dzPHw8EDZsmWRnp6e6/KLbdu24ezZsxAIBBg0aBDGjBlT4HX+QqEQq1evRrly5fDmzRuMHz9eI1ieMGECv4l34cKF6Nq1K7+p79SpUwV6DqJ/KSkpKFOmDMqUKVMkJayLqi19tE/ypnOg7OjoiNDQ0By3h4aGwsnJqSj6REoZsVhcqA/i7AEft9ln+vTpALJyEV+6dCnH44YNG4Zu3bohPT0dP/30U46Z4wYNGsDExAT//PMPrly5UshX9Z/s3/BNTEywaNEifoPg9u3bsWfPHv7+9u3b85dwFy5ciGvXrmk8vm3btnwAffjwYcjlcr5qH7dBUtcKfbQcgxSWeko3dfmdh9w5B4Bfj8wtmZJKpfy5nD1dZL169VC5cmUkJiaiT58+OZZZCQQCfP311wCAuXPnap1h42aaa9eujW+++UbnzXROTk5Ys2YNnJycEBUVhe+//57f92BtbY3x48cDyArIhUIhunfvDgC4fPmyTs9DCDF8OgfKI0aMwMiRI7FkyRJcvnwZly9fxuLFizFq1Cj+EhQhQPHNYgYGBqJz587IyMhAv379cqw15macmzdvDqVSiR9//BEvXrzg73d1dUXv3r0BAL/88otGlT1dMMbw7t07vHv3Lsd6aBMTEwwbNowfE7/++iuOHj3K39+5c2d06NABKpUK3377bY4+TJw4EXXr1kV8fLzWmXNd0XIMUlj5pYDLbXyrn3PqlTjd3d0hEokglUoBZGWmiYqK4vcxCIVCbNu2DdbW1rh27RrGjh2bIxieMWMGJBIJbt68ib179+Z47m+++QZA1lWjwm6yK1euHNatW4dy5crh9evXmDBhAlJSUgBkLaOysrJCdHQ0Ll26hLp16wIA7t+/X6jnIoQYLp0D5Z9++gmzZs3CL7/8Aj8/P/j5+WHt2rWYM2cOZs6cWRx9JEaqqGcxufaUSiXWr1+P+vXrIz4+Hj169MhxidbS0hL79+9HtWrVkJSUhMDAQI1d7P3794ebmxs+fPiAjRs3Fqo/6enpSE1NRVpaWq5VAQcOHIj+/fsDAJYvX65RKGHixInw8PBAZGQkJk+erPE4oVDILzM5ffo0/vnnH43ARFvxh7wUZhaaECD/EtW5jW/1c079346OjvD29uY3BAKAhYUF0tPT+TL2vr6+2Lx5M4RCIXbt2oWgoCCNtl1dXTF16lQAwI8//phjrbKnpycaN24Mxhj2799f6Nderlw5rF27Fk5OTnj9+jUflJuZmfE5n/fs2YNatWoBgNalXoQQ46ZToJyRkYHff/8d/fr1w+vXr5GQkICEhAT+2zbliiTqinoWU709Ozs7HDx4EO7u7njx4gX69OnDz/ZwrKysMH/+fFSsWBEfPnzAtGnT+LzK5ubmmDx5MgQCAf76668cjy2ItLQ0/t/cRovsBAIBRo0aha+//hoqlQqzZ8/my2dLJBJs27YNJiYm2LlzJw4dOqTx2IYNG+Lbb78FkLUuMiEhQWNDX2JiIkJDQ/HixYsc+aIJKSrqy36y367r+NY2C+3k5IQKFSrA1dUVIpGIz9Xcs2dPfpnVrFmz8Oeff2q0NX78eHh6eiI6Opov9qOuV69eEAgEuH79usYVJV05Oztj7NixAIDg4GD+S3nbtm0BAH/88QeqVKkCgUCA6OhoreuqCSHGS6dA2dTUFN999x0fVHDldgnRpqhnMbO35+joiMOHD8PW1hbXr1/XmhLO2toaixYtgqurK2JiYhAYGMjnOK5evTq6du0KIKtql7Z0cnnhAmVuJ31sbKzWEtoCgQCTJ0+Gj48PPn78iMDAQD6obtq0Kb9eeezYsTlyt86fPx8ODg4IDw/Hr7/+ym+AkkgkSE1Nhbm5ea47/wkpToVdK5+YmIjw8HBERkZCJpPxBUucnJxyBN6DBw+Gv78/gKy9B3fu3OHvs7CwwOLFiwEAR44cyfFl0d3dHU2aNAEA7Nu3r9CvE8jahFuzZk2kpKTwV6Bq164NFxcXxMfH4+rVq3zxE5pVJqR00bnES8OGDXHv3j14eHgUR3+IkSnoVYSiTmljaWkJAKhTpw62bt2K3r174+DBg6hQoQKf9QL4L5VUzZo10b59e7x69QoLFixASEgIRCIRatWqhXv37iEyMhIVK1ZEt27d8nxeLqhVxwXYjDG8ffsWADB69Ogcx2VkZMDU1BQvX75Ez5494e7ujkOHDmHAgAE4fvw4Hj58iIEDB2LLli0aaawCAgIwc+ZMrFmzBr169eLHnoeHBx9USKVSMMboqg4pFkU1ziUSCWQyGVQqFT5+/AilUglbW1skJyfzQTc368yta16xYgWio6Nx9epV+Pv74/r163BzcwMA+Pv7o1WrVggJCcHx48exadMmjedbuHAhWrZsiVu3bkEsFvMVO/PTqVOnHLdxGxNPnz6NK1euwNTUFEOGDMGqVauwe/du1KxZE8+ePUNoaChatmyp8dj8KgcSQgyXzmuUx4wZg8mTJ2Pt2rX43//+x5cS5X4IKQncTnyZTAZfX18+OF62bBnWrVuX43g3NzccPXoUUqkUYWFh/EyURCLBL7/8AgC4du0anj9/Xmx9NjMzg7u7OwQCAZKSkjSWgSxbtgwWFha4evUqX1aX06NHDzRq1AhJSUka6eTEYjHc3d0hlUr5dcuEGDKxWAxPT084OjqiTJkykEqlOWaRk5OTkZSUxG/CE4lE2LFjB6pWrYp3796he/fufGVLgUCAFStW8LnJr169qvF8lStX5q8aLV++/JP6bmpqyqeYUyqVYIyhT58+AIBjx47xaSdpQ59xMDExgb+/P/z9/T/5i4xQKCyytjjh4eG4e/cu7t69i7CwMLRu3RqtW7dGWFgYf3tUVFSRPBfJm4DpONWnLWG7QCDgZ7PUS/N+bhITE2FjY4OEhARYW1vruzsGpahnlGUyGZKSkhAdHY1y5crBysoKmzdvxsyZMyEQCPDHH3+ga9euOdYNHz16FP3794dQKERISAh8fX0BAK1bt8bNmzchlUoxadKkXHOuaptR1qZixYq5XpL+8OEDv8Ri1apV+OqrrwAAu3fvxty5c2FmZob9+/ejevXq/GMSExPRqFEjqFQqHD9+nH8MAERFReHjx48oU6ZMga/06GPmuaDjg8aR8frUcc5V8YyNjYWdnR2EQiG/GZD7UhwfH4/9+/ejR48e/OOGDh2KHTt2wNvbG6dPn9aovvf8+XO0aNECmZmZaNeuHZ9tIy/qhYLUqVQqPkONWCxGQkICqlevjqdPn2LChAlYvXo16tati1u3bmk8rihnlGkcfZq7d++iXr16uHPnDv/335BERUXB29u7QBMfYrEY4eHhcHd3L4GelS66jA+dZ5RfvHiR4+f58+f8f42JQqHQ2JBFjIdYLEZqaipsbGz4PKvDhw/Ht99+C8YY+vXrp7UYQZcuXeDv74/MzEyMHj2af2ynTp1gbW0NuVyOs2fPfnL/Xr16leuXRjs7Ozg4OAAApk+fzud57tu3L1q2bIn09HRMmDABHz9+5B/j6+uL7777DkDWUoy0tDQoFArI5XKaSSZGRSaTITw8PMemN4VCgZcvX0KlUsHBwQFWVlZQKBR4//49oqKiIBAIMHDgQABAUFCQRlD+ww8/wNbWFuHh4di9e7dGuxUrVuSD6k+d7TUxMYGFhQWArFnl1NRUPtUkV8Do4cOHOfYaEFJQ7u7uCA8Px507d/L8CQ4O5j8DSPHSOVD28PDI88dYPHjwAL169cL169dzJLQnhif7bnmJRAJ3d3c4Ozvz36YzMzMxf/58dOzYESkpKejSpYvWD8Zly5ZBKpXi0aNHmD17NoCsNc9c0YBLly4hIiLik/qbnp6eZyaKsmXLQiKRQKlU4vvvv0d6ejoEAgEWL14MV1dXREVF4eeff9Z4zLx58+Dk5ISnT5/il19+4ctacwVdCjJTRoi+yWQypKam5viA50pdp6amwsnJiV+zrFQq8fbtW8TExKB79+6wsLDAzZs3ce7cOf6x9vb2+OGHHwAAS5cuRXR0tEbbEydOhFAoRHR0NJ91prBEIhF/FXXDhg388osrV66gdu3aSEtLw8CBA6nUMCk0d3d3+Pr65vnj7e2t725+NnQOlEuDhw8folmzZihfvjwqVKgAkUhUqHZSU1ORmJio8UMKRtdiJOo5W9VL43Kbf4CsNYTW1tbYu3cv6tati9jYWHTs2BH37t3TaEsqlWL16tUAgHXr1vElpn18fODr6wvGGHbs2PHJKde0LVPiCAQCuLu7w8bGBq9eveIDeltbW6xYsQJA1k5+9bLwtra2WLhwIQBg8eLF/OZAqVTKVzrTxtDLV9M4+rw4OjpqFBzhSCQSWFlZ8RMu3BhPTExEZmYm0tLS4OnpiQEDBgDIyhKjfk4PHDgQPj4++PDhA4YPH66R8tHDwwODBw8GkBXQclX2CkMgEPCfGTdv3kS1atVQqVIlZGRkICAgABYWFvj7779x+vTpQj8HKX5KpRICgQACgeCTc/0nJycXWVv6aJ/k7bMLlJOTkzFp0iT07dsXGzZsgJubGx4/fozQ0FCdF8YvWrQINjY2/A+3E5vkT9diJOo5W7nHchvYuN3ojo6OfOq0/fv3o3bt2vjw4QM6d+6sUegDyFqCwRXImTRpEsLDwwFk7aJ3d3eHUqnE5s2b+SwWuvLy8oKzs3OexwiFQj4rh3oZ6zp16vC77leuXKnxmIEDB8LLywtxcXHYuHEjFAoFoqKiEBUVlWsgbOjlq2kcfV7UC46ol6/m7hOLxRpj3MbGBpaWlihfvjykUilGjBiBsmXLIiIiAnPmzOHbNTU1xebNm2Fra4vQ0FAEBgZqzOrOmjULrq6uUKlUuHTp0iddsua+BHOz09zsXlJSEp/xZs6cOTSrTEgp8NkFyqamplAoFBgxYgQyMzPRoUMHDBo0CM2bN0fv3r352cWCmD59Ol90JSEhAa9evSrGnpcueRUrKGhpXO7YtLQ0jdnU5ORkpKWlYceOHWjYsCESEhLwzTff5NgRP3XqVPTv3x8qlQrBwcF48+YNzMzM8O2330IqlSI+Ph5r165FaGiozq/P3Ny8QMdxeV6zr6ceP348TE1N8ffff2tsDDI1NcVPP/0EAFi9ejWePXsGuVwOmUyWZ7DM5Y4G/tswZSiBM42j0k/bmFYoFAgPD8eTJ09ynI/cGBeLxTAxMeGvmsjlcqSkpOD7778HkPVFUn18eHh4YP369TAxMcGBAwewatUq/j4zMzM0bdoUzs7OyMjIwMWLFxEfH1+o18MFytwXaS8vLwDAv//+i6lTp0IikeDOnTsaZesJIcbpswuU4+Pj8eTJE8jlcj6DwZYtW7B//340a9YMM2fOxMGDBwvUlkgkgrW1tcYPKZi8ihVomwFVv417LJAVkIrFYo2AW6FQQCQSwczMDPv27YOfnx8+fvyI7t274+LFi/xxAoEAa9asgZ+fH9LS0rBt2zbEx8ejTJkyGDduHLy8vJCeno5du3bhxIkTOhckKQguUA4NDdXYvOfh4cEXWli+fLnGzFTv3r1Rv359JCQkYPr06RCJRIiJiYFcLs91uYidnR3/b0ObYaZxVPrlNqaTk5ORkJCA9PR0jTGs/vehXLlyOZYWtW3bFt27d4dKpcLw4cM1NmU3b96cT6MYFBSkUWxEKBSiefPmkEqlSEtLQ0hISKGW+qgHyiqVik8N9/jxYzg6OmLcuHEAgNmzZxfL3w1CSMn5pEB5zJgxRrfj0snJCa1bt8bRo0fx9OlTBAQEoFatWujQoQPGjx+PNm3a4Pz588jMzKTLZnqibbZZ221isZifdVInFothZWUFd3d3SCQSbN68Ga1atYJCoUDPnj3x119/8ceam5sjODgYzs7OSExMxLZt25CSkgKxWIxhw4ahRYsWAICLFy/qdLWhoNzc3FC+fHlkZGTg9u3bGveNGTMGFhYWuHfvHk6cOMHfLhQKsXv3blhZWeH27dvYvXs3/8VBm+zvXVGXFickP7mNaYlEwpevBpBj1jn7GJdKpXB3d4eTkxPmz58PqVSKBw8e5MidPnjwYH7WeerUqbh06RJ/n5mZGVq0aAFbW1ukpKTgwoULOmc/4tIrpqWlQS6Xo2rVqgCAJ0+eAAAmT54MGxsbPHjwAAcOHNCpbUKIYdE5j7I6a2trhIaGomLFikXZp2J3+/ZttGjRAgqFAkeOHEHnzp35+3744QfcunULFy9e1DnXLOWtLDkFPW25fMtJSUn48ccfcfLkSVhYWOD27dsaeYpDQ0PRvn17yGQytGzZErt27eJzKf/555+YOHEiFAoFnJycEBAQkG/eyoJu5OnWrRvWr1+Pv/76C506dcLQoUM17t+5cycOHz6M6tWr48qVKxobBP/88098++23AIDNmzejadOmcHNzK7KS4booyFih/K+Ew41fhULBl7DmylunpqbC09Mz3/M4MjISBw4cwLRp02BmZoabN2+iRo0a/P0qlQqDBw/G3r17YWVlhZCQEDg5OfH3y+VydOvWDVFRUfjmm2+wZs0a/j5dMjhdunQJrq6uqFy5MgDgzZs3EIvFWLZsGRYuXIgqVarg0aNHGrmdPwWNo0/D5VG+cuUKmjZtCgD4+PHjJ00cJCcno0yZMkXSVkHbN/R80IauWPMoqzPWGdf69evj1KlTAIBNmzbh4cOH/H3p6enw8vLiN4gRw5TbOubst3P5loVCIRYuXAg/Pz+kpKSgf//+GmkB3dzcsHv3bojFYly4cAHTpk3jz+9u3brh5MmT8PDwQExMDObMmYPr168X2WupXbs2AGitbNmtWzeIxWI8evQox5Kgbt268YH19OnT+debH0PPgkGMU3Jy8ietfZdIJEhNTeWXEnGb/LhNu9w5y/1bqVSiefPm+PLLL5Geno6RI0dq5C43MTHBli1b0KJFCyQlJaFLly4a+Y2lUinWrFkDoVCII0eOFHo98du3b+Hg4AB7e3sA4FNLjho1Cvb29nj69Cl27dpVqLYJIfr32a1R5jRr1gwXL17E3bt3MXToUAwfPhyDBg3Cjh07MGHChFwrsxHDkNs62+y3c/mW09LSkJCQgICAADg6OuKff/7hg0tOnTp1sGHDBggEAvz+++8aG4F8fHzw119/oWbNmkhNTcUvv/yCPXv2FMn6wxo1akAgECAqKkpj0x0AWFlZ8SV4Fy5cmOMS8cKFC+Hj4wO5XI5hw4bh3bt3fGnvgmbBoMCZFIXCrH1Xf0xycjJfxAPImjn7+PEjH4C/f/+eD8QzMjJgaWkJa2trTJw4EdbW1rh165bGrDCQtf79wIED8PHxQXR0NAYPHoyEhAT+/rp16/JLNH788cdCZbnhHlOlShUAWRv6gKwrruPHjwcAzJ07F+np6Tq3TYqPiYkJvv76a3z99ddFUsK6qNrSR/skb58UKCclJRndsgt1zZs3R0hICNq1a4eXL1+iTJkyuHLlisblO2KYcltnm9taSAcHBzDGYGFhgcDAQABZ5aOzV+H76quv+I1ACxcuxNSpU/ng1M7ODlOnTuVTtx0/fhxLly7V2IRXGNbW1qhQoQIA7bPKnTp1gpOTEyIjI7Fz506N+ywsLLB9+3ZIJBJcu3YNQUFBkMlk+QYs6gG5oW3uI8apMGvf1R8jk8mQlpaGuLg4SCQSmJiY5CgGpVQqkZyczBcl8fDwgKenJ0aNGgUga/Nc9mJBtra2OHbsGFxcXPD06VOMHDlSo91x48ahdu3aSExMxOTJk3X+8ssVN+EC5WfPnvH3DR8+HE5OTnjx4gV+++03ndolxUskEuHEiRM4ceIEX22xsCwsLIqsLX20T/L22c4oc6pWrYp58+bhzJkzWLt2LQXJRiK3rBnqGTFkMhkf/EmlUnh4eMDNzQ1+fn580YJBgwbxJaQ5I0eOxIwZMwAAv/32G3r06MFvWjUxMUHfvn3x/fffQyQS4f79+5g+fTquXbtW6KVIGRkZsLW1BaC9xK6FhQWfoSV7XmUg6wOau33lypV85TOuOIs26lkwaHMfKQoSiQROTk46nUfq49jR0RGZmZn8WHB0dISLiwuArE3Yzs7OsLS0hEgkgkQigVgshlgshru7O7p06YK6detCqVTiu+++yzEW3dzccOzYMZQpUwbXr1/XWFplZmaGVatWwdLSEteuXdN50272QJmbUebek0mTJgEAXyyIFJ+oqCjcvXs3zx8uZz4hBfXZB8ocExOTPCupEeOiXrAAyPrA8vDwgLe3N6pWrYr58+ejatWqiImJQYsWLTQq4AFAQEAAdu3aBSsrK1y/fh3t27fXOKZx48aYM2cOnJ2dERcXh3Xr1mHu3Lkas0kF8eLFC0ybNg13794FAI3NRuq4gCG33fm9evVCo0aNwBjDo0ePIJFIIBKJtM4SZw+M80rVR0hJcXR0hK+vL5ydnfmMGNx5yp2jTk5O/G1csZKoqCjY2NhgxIgRsLS0zDVDTa1atbBx40YIhUL8+eefGldnKlasyBcgWrp0qU799vT0BJB1hRXImUO9devWAID3799TqrhiFBUVBW9vb9SrVy/PnwEDBkAsFueoDElIbigyJKWSesECbfe5u7vj9OnTqFGjBmJiYtC1a9cchUXatWuH06dPw9PTE1FRUfj6669x584d/n53d3csXrwY/v7+EIlEePr0KWbPno1ff/0132UMmZmZuHfvHqZOnYoXL16gTJkymDBhAnr27Kn1eO6Dv3///rm2aWlpCQD8pevcZpULExjTOmZSXNTPLfVzU9t5qn5bcnIyZDIZ5HI5EhMT4eXlxY+fadOm8VXz1DVt2pRfejVv3jz+CyqQNbbatm2rU6q48uXL85X4bty4AQBo2LChxjFPnz4FkHX1kiZjio9cLodCoUBwcDDu3LmT5094eDgcHBz4L2RFUcK6qNrSR/skbzRqSanEfaDmdRnYw8MDly5dQqNGjfDhwwd07949R6lrLy8vnDlzBs2aNUNycjJWrlyJo0eP8pdtzc3N0a1bNyxfvhzNmzeHQCDA1atXceTIEYSGhmrdwCOTyXDixAk8ePAAmZmZaNy4MV/4RFuatbdv3yIkJAQCgQBDhgzJ9fVws1UikYgPLpKSknItQqILWsdMiouu5xYXWANZVwItLS2RkpICJycnjB07FrVq1UJiYiJf9CO7ESNG4Ouvv0Z6ejpGjx7NL6sSCARYsmRJnjnJs1u6dCkkEgnS09P5L9GNGjXSOIZbiqGejpIUH29vb/j6+ub5w6X35LKoFIWibEsf7ZPcUaBMSh1dZj/t7Oxw9uxZNG7cGElJSejZsycuX76scYy9vT327duHb7/9Fowx7Nu3D+vXr9eYebKzs8OoUaMwb948VKtWDZmZmbh//z6OHDmCiIgIMMb4oiJnzpxBQkICv/Z4ypQp/LpMbc6cOQMga4abu8yrDRcoc8s3FAoF4uLiiuSPK61jJsVFl3NLoVDg5cuX/DIHb29vfkmGUCiEt7c3Nm/eDDMzMxw/fjxH2XogKyBetmwZKlWqhHfv3mHChAn8F18HBwesWLGiwH3/6quvAAAPHjyAQqGAjY0NX6WPwxUh8fb2LnC7hBDDQYEyKXV0naGysrLCjh070KhRIygUCvTr1w/nzp3TOMbMzAxLly7FkCFDYGJigqtXr2L+/Pk50rlVqFABM2fORPPmzVGmTBkolUpcu3YNJ0+exPHjxxEeHg7GGCpWrIguXbqgcePGefYtNTUVFy5cAAAMGzYsz2O5D3uBQMDPuNnb2xfJ2mNax0yKiy7nVnJyMkQiEVJTU/m1yx4eHnBycoKbmxssLS01yr8HBQVpbadMmTLYsGEDLCwscOXKFX6MAVnZkHTFXYlq0KBBjuUV3IwyBcqEGKeiKRVESAnLqxIct44rrxkq9YpgYrEYNjY2WLt2LSZNmoTLly9j0KBB2LNnD7p3767xuF9++QX+/v4YOHAgIiIisHDhQuzduxd169bVOM7b2xtpaWnYu3cvtmzZgri4OACAs7MzZsyYwVeEyi84OHToED5+/AhPT88C59BMT09HRkaGRtvJycn879wlPO6150XX6pSEFERBz6vs2Su4sc2txweyxpD67xKJBOPGjcPevXtx/PhxPHr0iM9m5OzszB/n7OyM7777DqtWrcKGDRvQp08fvl+HDx9GQEAAXrx4gXr16mHu3LlISkrCyJEjkZSUhOHDh6N79+58qq5bt24ByFoDrZ6+izFGSy8IMXI0o0xKnYKkqco+6ywWi+Hg4IAtW7agW7duSE9Ph7+/PyZNmoSUlBSNx/r5+eHChQvw8vLC27dv0bp1a/z000858imbm5tj0KBBOHz4MPr374+BAwdi//79fJBcEHv27AGQlY81vyCZCyosLCz4PLFc3lluhhnICpRpvTExRmKxOM+MLtyXwJo1a/JfcpctW5ZrexMnToSFhQVu3LihMatsbm6OqVOnwtzcHHfu3MGxY8ewdetWvnbAN998wx/LGMP//vc/AMhxhejNmzdISkqCUCjkS1wTQowLBcrks6G+dll9XST34ero6IgqVapg3759fGnoVatWoVGjRnjw4IFGW5UqVUJISAg6d+6M9PR0rFq1CvXq1cMff/yRYxbM3t4ekyZNwsSJE1GmTJkC9/f+/ft48OABzM3NMXjw4HyP59YoW1hY8MGEUqnUemz2JSOEGIr8SmFnX9OsUCj4jAcKhQJJSUl48uQJBg0aBCDry+bLly+1tlW2bFl+rC9atEjjPg8PD36509atW3Hu3DkIBAJ8//33Gl9ao6Ki8PbtW5iamqJ+/foabXDrkytXrpwjbRwhxDhQoExKNfXgmJtF5j6EuaUH2XMum5qaYsuWLThy5AgcHR1x//59NGjQAKtXr9bIg2pjY4Pdu3dj//79qFChAt6+fYshQ4agU6dOOaqDFQY3m9yhQ4cC5fzkAnRuo5OpqSnc3Nzg5OSUYye/esERQgyJtj0GeW3QVR+/YrEY8fHxkMvlsLW1ha+vLzIzM/PcoDdp0iSYmZnh8uXLuHLlisZ9nTp1QsOGDZGRkQEga/Ne9s163GxynTp1cixl4ooZ0fpkw2NiYgI/Pz/4+fl9ctq+omxLH+2TvNE7Tko19Q9dbiYKgMYHcW45lzt37oywsDC0a9cOqampCAgIQP/+/XPM0n711Ve4efMmfvzxR1hYWODvv/9G3759ERwcXOhqfWFhYTh58iQAoG/fvvkezxjjA2TuOaVSqdY1yGKxmDJYEIOlLQuG+jiOiYnBu3fv+LSH2savpaUl0tLS+AqcW7du5QPa7MqXL8/PPs+YMUNjfAsEAkycOJGvDJj9yg5jDH/++SeAnMsuAODhw4cAaH2yIeKK01y8eFFjjbu+29JH+yRvFCiTUi236l7qxTjyyrns7OyMU6dOYe3atRCJRDh+/Dg6d+7Mb87jWFhYIDAwELdv30bHjh2RmZmJlStXYsaMGTqnZwsJCcHgwYORmpqKRo0aoXbt2vk+ZsWKFXj06BHMzMzg4uLCvzbuUnRUVBTkcjmfM5YyWBBDpW2PQV4p5Lgqa2KxGAqFAra2tpBKpahcuTJq1KiBWrVqQalUonXr1vjjjz+0PufUqVNhZWWFmzdvok+fPhr5z21tbbFp0yZs3LgRVlZW/O2MMQQEBODo0aMQCATo2rWrRpuJiYn88+myL4EQYlgoUCalWm7VvfLaEJSdQCDAmDFjcPbsWdja2uLGjRto37691spfHh4e2LNnD6ZNmwahUIizZ89iyJAhiIqK0tp2Wloa3rx5gzt37uDEiRNYvnw5xo0bh5SUFDRr1gzr1q3LN0NASEgIZs2aBSDrMrKpqSlkMhlkMhnEYjFSU1MhEokgl8v5oJkS1xNjwo1jjpWVldZy72KxGFZWVnB3d4dYLIZIJMKiRYvQrFkzpKamon///li+fHmOKz0eHh74888/YWlpiTNnzmDJkiXIzMzk7zc3N9dYY8wY44NngUCATZs2oUmTJhptbtu2DYmJiahatSrat29fVG8FIaSEUXo48lnh1ioD0Gn5QXJyMqpVq4ajR4+iV69eePz4MVq3bo3Dhw/nWH8oEAjQq1cveHl5Ydq0aYiIiMCAAQPQp08fJCUl4f3794iJiUFMTAxiY2O1Pp+/vz9mz57NLxXJzcuXLzFo0CCoVCp069YN/v7+iIuLw/v37yEUCiGRSGBpaQmlUsn/l/uCoG1GmQuwHR0ddapQRkhJ4PIo51aenit9zZFKpZBIJFi3bh3mz5+P/fv348cff8SLFy+watUqjfHVtGlTHDhwAN27d8e1a9ewYsUKTJo0KUe2GcYYv4cBANavX88v8eCkp6dj7dq1AIDRo0fTulIDlJyczBdwioyM/KSlaEXZlj7aJ3mjQJl8Vri1jqampjoFglxKNU9PT5w+fRr+/v549uwZ2rVrh7179+LLL7/M8Zg6deogODgYgYGBCA0NxdatW7W2bWZmhrJly8LZ2Rlly5ZFo0aN0KNHj3xnkpVKJfr27YvY2FjUqFEDQUFBcHJywpMnT/jZsIyMDCiVStja2vJfDORyOX8JOXs+aZlMhtTUVD5YJsSQFCRHOkcsFvMb7xQKBaZOnQpXV1esWrUKW7ZsQVRUFHbt2qWxnKJNmzbYs2cPevXqhQsXLsDc3Bzjx4/nxyJjDL/99hu/Lnnt2rVaM9IcPHgQb968gaOjIzp37lwUL50UA24pmqG1pY/2Se4oUCafFV0+aNVxQWR6ejpcXV1x/vx59OrVCzdu3EDHjh0REBCAadOmaRQbALLWAm/YsAG7du3C8+fP4ezszG8McnJygqenJ+zs7HQu7MEYw4QJE3Dv3j1IpVJs27YNdnZ2EIvFqFq1Kr+0gps9T01NhbW1Nb+Gk6O+SYq7vE1BMjFU2WeMc8Otz+f+DQBubm744YcfUK9ePYwYMQJnz55F69atcejQIZQvX55/bMeOHTFlyhQsXboUZ86cgUgkwqhRowAAO3fuxMGDBwEAY8aM0VotkzGGlStXAgCGDh1KGWYIMXIUKJNSLXsAKpFItAbJ+WWn4PItZ2RkQKVSwc3NDSEhIejfvz8OHz6MoKAgHD9+HFu2bEGTJk34SmCc7JX7cutfbrJful23bh1+//13mJiYYP369fD19eXbU88vKxAIcuRTjo+P59PNZf/iQEsuiCEpbGVIbqx++PCBv7oiFotRrlw59OvXD+XLl4e/vz/CwsLQsmVLXL16VSNYnj17NipVqoQRI0bg6NGj8PLygrm5Ofbt2wcgqzT22LFjYWZmluO5z549i/v370MikWDy5MmwtrYu1GsghBgGCpQJKaDsQaWlpSWCg4Oxf/9+TJkyBY8fP0azZs0wduxYzJ07V6fiIgV17949/Pjjjzh79iwAYMqUKWjWrFmO47hAAQC/mY+bWVOfUS7oDB0hxkB9E6tEIoGjo6NGGkhO8+bNcfPmTbRv3x5Pnz5Fx44d8ffff8PGxoY/ZsCAAVAoFJgwYQKCgoL425csWYKxY8fm2ofly5cDyJpNtre3L+qXSIiG8PDwfI+RSqVwd3cvgd6UThQoE1JA2oJKsViMIUOGoG7dupg7dy4OHz6MtWvX4tixY1i3bh3atm1bJM/97NkzzJ49m5/RMjU1Rbdu3dCmTRuN47i0cNwx3GYlhUIBoVCYZw7l7OuVCTE2MpkMaWlpAABPT09+LHBfcl++fAmpVMpfOdmzZw86duyI+/fvw9/fHydOnNDIbjFy5EgolUoEBgYCAObPn4/x48fn+vyhoaE4f/48hEIhJkyYUIyvlHzuuJSI2TeSaiMWixEeHk7BciFRoExIISkUCr7ogYuLCzZu3Ii+ffsiICAAL1++RKdOnTBo0CAsXbq00OsUo6OjsXDhQmzdupWfIe7YsSOGDRsGU1NT2NjY8CV8ASA2NhYikQhWVlb8emNzc3MoFAr+v1w1wpiYGP53JyenHOuVCTE22dfYq5/TXBAtl8shFosRFRUFW1tb7Nq1C127dsX58+cxfPhw7NixQ6PNCRMmoEKFCgCALl265Pn8XAXAFi1aaE1fR0hRcXd3R3h4eL6b/MLDwzFgwADI5XIKlAuJAmVCCik5ORkfP34EAP4yb8+ePeHn54dp06Zh586d2LlzJ86dO4cjR46gVq1aBWo3KSkJx48fxx9//IEzZ84gNTUVANCkSRP8/PPPqFixIl/wxN7eHkqlEnK5HImJiXwxFWdnZwDgizBwa5LT0tI01m7GxcXB3t6eX7tdmI2OhBiK7Gvss5/TcrkcUqkUCoUCIpGIL+pz4MABdOrUCcHBwXB1dcWcOXM02s0vQFapVFixYgVfdn7EiBH8l1BimExMTFC/fn3+34bSli7tu7u7U/BbAihQJkSNLssPJBIJvw5ZPbh0cnLCokWL0LVrV0yePBnPnz9H69atMX/+fJiZmeHjx4/8Dxdscz9JSUkIDQ3lg2MAqFevHqZPnw5fX1+IxWJ+1lgikcDd3Z2vumdmZgahUMgHAly/spcCVigU/NpNbjkG93rpg52UJtw5zY0H7gsjkFW0hPtC6evri6CgIAQEBGDJkiVwdHTMcx2yupiYGHz77bc4c+YMgKwguUWLFjSWDJylpSVu3bplcG3po32SNwqUCVGj6/IDLhBVP1ahUCAuLg6Ojo4IDg7GlClTcPXq1TzXNmZXoUIF9O7dG/7+/qhUqRJSU1OhUCiQmZnJFw3hngv4LwDgggJumUX22WH1jBhc32lNMintuHEdHx/P5xTnZp5lMhkyMjLQrFkzfP/991i7di2mTJkCR0dH9OrVK892uXLz0dHRsLCwwLx58zBy5Ei6KkNIKUKBMiFqdFl+kFtQnZycDJVKBZVKBWtra/z+++9Ys2YN7t27B1NTU5ibm/MBqomJCaysrCCVSsEYQ0ZGBpydndGhQweNS2pWVlZ8blgujZVYLIZcLkdycjJ/OS4zMxNCoRBpaWlIS0vT+lrUU2fZ2dnRmmRS6nHjWv0LJaC5+dXS0hIjRoxAbGws9uzZg+HDh0MqlaJVq1Y52svIyMCCBQuwZMkSMMbg4uKCdevWoWHDhlAqlRQoE1KKUKBMiBpdlh/kFlSrB8GWlpawt7fHkiVLAGRttouLi4OlpSWio6ORlJQEiUQCV1dXpKSk8MGr+iXi3HAz1yqVCmlpaTA3N4dSqUTVqlUhl8uRlJTE9yf7a+SWX2i7n5DSJrdxzV19MTU15ZcszZs3Dx8+fMDp06fRu3dvnD17ViMP+uvXrzFkyBBcvXoVANCsWTNMnDgRXl5eSE1NLdDYJfqnUChQvXp1AMCjR48+abKgKNvSR/skbxQoE1JIuX34isVijeIFQFa5aYVCAaVSCRsbGyQkJMDa2hopKSlwcHAAAFhbW8Pe3h4ODg5QKBR49eoVAMDBwQEpKSl49uwZbGxsYGJiAltbW/4ycmpqKiQSCVQqFcqUKcOXqVYqlUhLS4NUKtUIhnMrugJQijjyeeGyXyiVSri5ucHR0RFisRhBQUGIiYnB3bt30bVrV4SEhKBSpUo4ceIERo4cibi4OFhZWWHkyJHo1q0bLCwsIBKJaJ2/EWGM4eXLl/y/DaUtfbRP8kaBMiEoeAWwgv6Ryl6xKz4+HowxxMbGQqlUwsHBAc7OzqhYsSI/w8sFqKampnj9+jXu378PCwsL1K5dmw96X758CR8fH40ZMPUPZy71GwAIhULY2tpCLpfzx+U3e5x9OUlBX29hK6gRUty4NI7qqRC58SKRSMAYg0Ag4NPKcRlstm7dir59++Lx48fo0qULOnTogF9//RUAUK1aNYwePRpVq1YFYwzJycl49+4dKlSoQF8yCSllij6PCSEkBy4ABsBnyuBmr4CsAFf9d6VSiczMTKSkpPD3c8slss9aZWZm8rk0uZzKXNleoVAIIGtNpVwuh0wm4wPp7LhgnZuhJqQ0SE5ORlJSEmJiYpCUlJTj/Hd0dIRIJNJIKycWi+Hi4oIDBw7A1dUVz58/54Pkfv36Yc+ePfD19YWbmxtffe/Dhw+4e/cunjx5AuC/scitgSaEGCeaUSakBKgHxOrrg3PDbeSztLTki4FUq1YNqampfBtcAZL4+Hg+G4atrS2io6MRHx8PKysrODg48DPPXA5lbdkwgKyAQiQSwdTUlGbESKkhkUhgZWWlkQpRXfbcy8B/4zQ+Ph7z5s3DTz/9hJSUFPz8889o0aIF3r59C3Nzc4jFYr5C2uvXrwFkBczAf5tmFQpFsZSzJ4SUDAqUCSkhycnJsLOz0/rBnH1tMHcMdzvwX95X9Q9xsViMlJQUfPz4EWXKlEF0dDQyMzM1ln6oFxPJqwhCYQqOqPebZqGJIRKLxfD09NS4rSBr8eVyOYRCIRwdHbFv3z54eHjAxcUFjx8/Rnp6Ol/IJyoqCnFxcXyxHzs7O36cUdERQowfBcqEFKG8PoDzCkTzSjWXkZGhsSaZu5zLzR5z1QEtLCz4TX7cEg31HfjqOZS5NcXqfS3MRiTucraVlRVf5pcQQ6d+3np6emqMWwAa49DGxgYikQiJiYmwtbWFVCqFUqmEpaUlxGIxnjx5gnfv3qFs2bIoW7YszM3N+XLBFCQTYvw+60CZyznLbeYg5FPlVbAkv0D0w4cPOWaaueCae5z6kgtTU1NYWlrCzs4O6enpsLS0hFAoRNWqVTUCavXAnHt8TEyMRuES+kAnn7Pk5GQkJiZCJpNBLBbzGSzc3d2hUCgQFRUFkUjE7wXggmQga/2/paUlJBIJLC0t+f0BxLAJBAI+5dqnfv4XZVv6aJ/k7bMNlENDQ/HTTz9h3759FCSQIqPL8oXss1h2dnY5juGCay77BHc519raWqM8r3o+WIlEgqioKCQkJCAuLk5jXSZ3fGpqKkQiEVJTU1G2bNlCp4VzcnKiZRfE6GQ/byUSCWQyGf/FUX2dvnrArFAoIJPJEB0djbJly6JMmTL83gHuGC6YlsvllC7OgInFYjx8+NDg2tJH+yRvn2Wg/M8//6BJkyYYP368xh8xmlkmn0qXD0b12efCrA9Wfz7uQ5x7bqlUirdv38LMzAxyuVwjIFBfs1y2bFmIxWK+jC+Xq5PLvZxf8Mw9P40bYkyyj1NuHTN3vltaWmo9XqFQIDY2FnZ2dkhLSwOQlVmGW2ahUCjw8eNHhIaGonz58pBKpXB3d6fNfIQYsc8uUA4LC8OXX36J77//HosXL+Zv5yqb6SI1NRWpqan874mJiUXWT1L6ZQ9ECxJgq++kz/5Bz90PZO3kr1y5Ml+dL7vsz6e+2U8oFPKzYrktIylKNI6IIVAfE9nzh6t/EeWWNqlfxVGffeZmpl+/fg1XV1fKekGIkfus8ii/e/cO7du3R9OmTbF06VJkZmYiICAAnTp1Qu3atbFq1So8fvy4wO0tWrQINjY2/I+bm1sx9p4YM+6SrXpO1ey5kwtCLBbnmr5NPYgGsmaFnZ2dIZVKkZycrDWHMtcvICu4dnd3h0gk4meUuaUcxYnGETEU2sYpd7v6F1QuJVx8fLzGcWKxGBUqVICFhQWqVq0KKysrWnphoBQKBXx8fODj4/PJua6Lsi19tE/y9tnNKDdu3BivXr3CkSNHsGHDBqSnp6NOnTrw9PTEmjVr8ODBA8yaNYvPY5uX6dOnY9KkSfzviYmJ9CFfyhW2gl9um/x0XbKQ13pg9RlqbsaLm63mllZk39yXvV/ZU9eVxNpjGkfkUxXV0h9uPGjLNc5ttlV/Lltb2xzPL5VKYWtrC1NT03zzpRP9YYzh0aNH/L8NpS19tE/y9lkFymXLlsW6desQGBiIvn37omnTpti3bx8cHBwAALt378bYsWPRo0ePAgXKIpGI3/xBSF4KuwZZF+qXjmUyGb+L39PTM9fnL4l+5YfGETEU2sYDN8uc/Rw15DFFCCk6n1WgDADlypXDokWL4OrqijZt2sDBwYHfxNevXz/Mnj0bFy5cwFdffaXvrpJSpKR3v6vv4k9OTs51iQftyifkP9wVG/VZO65iZfbS7rmNHRpThJQun12gDAAuLi4IDAyEhYUFgKzLZowxxMXFwdHREXXq1NFvBwnRUfb0btl38evyWELIf9QzxuS2NyB7sRIaS4SUHp9loAwA1tbWGr8LBAKsWbMGcrkcX375pZ56RUjhaFsDnd/MFvcBz82YFXd2C0KMUX7jSH3sJScna1T8I4QYv882UFa3d+9eXLhwAQcOHMD58+fh4eGh7y4RopPCrIvkPuABlEh2C0JKI/Wxlz2rDCHE+FGgDKB69eoIDg7G5cuX4ePjo+/uEKKz7LNe2ZdTcL9z1IuO2NjY5JgxU388BdCE5I5LvQhkVfzjcLfRGDJMAoGAnxQrihLWRdWWPtoneaNAGUCtWrVw6NAhnQuOEGJI1IPb7EsxuN8/fPgAOzu7PDf4ATmrBhJCsmT/EiqTyZCWlgaZTAZHR0dIJBJkZGRAJpPxY43GkOERi8WIjIw0uLb00T7JGwXK/4+CZGLs8iqJzf3O5XXN74ObUlwRop223ONckAzoPtYIKQnh4eH5HsOVXCeaKFAmpJTIqyS2rimruOPpMh8hmrJ/icxepIfSwxFDwlWRHDBgQL7HisVihIeHU7CcDQXKhBSDog4wC5LCTSwWF3j2iqo7EZK7vMaveho4ABQUGymlUonmzZsDAP7++29YWloaRFtF3b67uzvCw8Mhl8vzPC48PBwDBgyAXC6nQDkbCpQJMQK5lcAmhJQ8Go8lKyoqqkCBni5UKhVu377N//tTFGVbxdG+u7s7Bb+fgAJlQowArRkmxHDQeCw5UVFR8Pb2hkKhyPdYsVgMqVRaAr0inxMKlAkxAsWx7pEq8hFSONpKXQM0poqDXC6HQqFAcHAwvL298zyWNqOR4kCBMiGfqeyXj2UyGb97X31zEiGkYLgxFRMTwwfT6gEzV72PcivrztvbG76+vvruBvkMmei7A4QQ/ZBIJBoV+WQyGVJTU/lCCYQQ3XBjCgD/JVSd+pdTQohxoECZkM8UlwOWm/FydHREZmYmABRoPSAhRBM3ppycnJCamork5GSNsZT9yykhxPDR0gtCCADwyy1oNz8hn4ZL1Zi9uiUtuTAsRbnxr7g3EdImRf2hQJkQwqPd/IQUDRpLhk0ikRTZMrOibEsf7ZO8UaBMCOFRVTFCigaNJUJKBwqUCTECBa30l1/FPfX0VTTTRcin0TYuKbMFIaULbeYj5DNCu+4JKV40xoyDUqlEixYt0KJFCyiVSoNpSx/tk7zRjDIhnxFaN0lI8aIxZhxUKhUuXbrE/9tQ2tJH+yRvFCgT8hnh1k0WdCkHIUQ3tOSCkNKFAmVCCCGEEILw8PB8j/ncSoVToEwIIYQQ8hmTSqUQi8UYMGBAvseKxWKEh4d/NsEyBcqEEEIIIZ8xd3d3hIeHQy6X53lceHg4BgwYALlcToEyIYQQQkhxiYqKKlBgRkqGu7v7ZxP86oICZUIIIYSUqKioKHh7e0OhUOR7rFgsLpYSzkVZEKa4i8tQ8Rr9oUCZEEIIISVKLpdDoVAgODgY3t7eeR5bHJvHuDR+htaWPtoneaNAmZBSRF9p3/KrCFjQYwghnxdvb2/4+vrquxuE5IoCZUIIIYQQUmCfUxo5CpQJIYQQ8llJSUlBjx49AAB//PEHLCwsDKItfbSvi88xjRwFyoQQQgj5rGRmZuLkyZP8vw2lLX20r4vPMY0cBcqEEEIIIaRAPrc0cib67gAhhBBCCCGGiGaUCSGEEFJkqJAI4ZSGTX8UKBNCCCGkSBhCIRGif7pu+jt06BAcHR3zbVMfATUFyoQQQgjJV0FnivVZSIQYhoJu+pPJZOjevTs6dOiQb5sFDagL4uPHjwU+lgLlIsQVVEhMTNRzTwgpWQUpJvL+/fsCHUvjiJDcvXv3DkDBx9HVq1chkUg++XnlcjkGDBgApVKZ77GWlpaoU6cO3Nzc8j1WX+NcvdJdYmLiJ2WTKMq29NF+cbG1tYWtrW2ex1SuXBk3b95EbGxsnsdx519BAmpdFOSzS8CoXFaRef36dYH+MBDyOXv16hXKly+f6/00jgjJH40jQj5dfuMIoEC5SKlUKrx9+xZWVlZ8KeHExES4ubnh1atXsLa21nMPdUN9L3nG2m8g/74zxpCUlAQXFxeYmOSecEfbOCppxvb/gfpbvAypv8Y0jvJjSO9rfoylr9TPginoOAJo6UWRMjExyfWbibW1tUGftHmhvpc8Y+03kHffbWxs8n18XuOopBnb/wfqb/EylP4a2zjKj6G8rwVhLH2lfuavIOMIoDzKhBBCCCGEaEWBMiGEEEIIIVpQoFzMRCIRZs+eDZFIpO+u6Iz6XvKMtd+Acfc9O2N7LdTf4mVs/TUWxvS+GktfqZ9FjzbzEUIIIYQQogXNKBNCCCGEEKIFBcqEEEIIIYRoQYEyIYQQQgghWlCgTAghhBBCiBYUKBNCCCGEEKIFBcolRCaT6bsLRcaYEqWUpvcdMK73Pjtj7HtMTAwiIiL03Q1iIFJTU/l/G+P5bCgyMzMBGP57qFAokJaWpu9uFMjr169x7949fXejVKJAuQTcu3cPzs7OuHLlir67orPIyEgEBwdj/fr1uHbtGgBAIBBApVLpuWf5M+b3XV1KSgoUCgWArPceMPwPGE5YWBhmzZoF4L++G4uwsDA0a9YMZ86cKXVfuAxBQkIC4uLi8PLlS313pUAePXqEHj164Pz58wCyzmdjGYeGJDQ0FF27doVCoTDovwkPHjxAr169cP36dY0vSIbo4cOHaNKkCYKDgwHAYD+fX79+jf379+PQoUO4f/++vrtTcIwUq9DQUGZlZcUmTZqk767oLCwsjNnb2zM/Pz9mZ2fHatasybp06cJUKhVjjLHMzEw99zB3xvy+q7t//z5r2bIlq1evHmvYsCFbt24de/PmDWPMsN9/xrL+H1haWrIff/xR43bu/DFk//77L3NwcGATJkxgSUlJOe43tPf+6dOnbN68eaxPnz5s27Zt7N9//9V3l/L04MED1rx5c1azZk1WuXJl9ttvv+m7S3lSqVRs8ODBzMbGhnXq1ImdO3dO4z5SMNzfhGnTpmncbmjv4YMHD5itrS0bNWoUi4qK0nd38hQaGsrEYjGrUKECK1u2LHv//r2+u6RVWFgY8/DwYPXr12fOzs6sc+fO7NmzZ/ruVoFQoFyM7t+/z8RiMZs5cyZjLOuPwbNnz9jly5fZu3fv9Ny7vH38+JE1adKEjR49mmVkZDCZTMZ2797Nqlatynx9fVlKSgpjzPACBsaM+31XFxERwezs7NiIESPYzp07Wb9+/Zivry/r1KkTe/r0KWPMMN9/xrL+eEskEjZ58uRcjzG0D0d1kydPZn379mWMZfVzz549bM2aNWznzp38MYby3t+/f585Ozuzbt26sVatWrFKlSqxiRMnspSUFIN8jx89esTs7e3ZtGnTWHBwMAsICGANGjRg8fHx+u5ansaMGcMaNWrEunXrxtq0acPOnj2r7y4ZlX/++YdJJBI2ZcoUjdtTU1P11CPtPn78yNq1a8dGjx7N3xYeHs7u3bvHXr58qcee5cR98ZgxYwaTyWTMx8eHzZ8/n6lUKoMa+5GRkczV1ZUFBgayjx8/spMnT7KyZcuyGzdu6LtrBUKBcjFJSUlhnTp1YiYmJvxtHTt2ZHXr1mUCgYA1aNDAoGc74+LiWM2aNdmRI0f429LT09mtW7dYtWrVWOPGjfnbDWlAGvv7rm7t2rWsXbt2GrcFBwezVq1asZYtW7Lnz58zxgzr/WeMsaioKCaRSNjw4cMZY1kfhAsWLGDDhg1jffr0YadPnzb4oMjf35+tXr2aMcbYF198wZo1a8YqVarEKlWqxBo1asQHyfp+71+9esWqV6/OAgMD+du2b9/O7Ozs2IsXL/TXsVykpaWxoUOHsqFDh/K3nT17lnXq1Im9f/+ePXnyRI+9y9vu3bvZ4sWL2Y0bN1j79u1Zu3bt2L1799iSJUsMLoAyNNHR0axs2bKsffv2jDHGMjIy2MSJE1nHjh1ZtWrV2MqVK1l4eLiee5klJSWFNW3alN29e5dlZGSw9u3bswYNGjArKyv2xRdfsC1btui7i4yxrC8eIpGIzZgxgzGW9cXd39+fNWjQgD9G33+fOBs3bmQtWrTQ6M/XX3/NNm7cyHbs2MFCQkL02Lv80RrlYmJmZoYZM2agcuXK+PLLL9GuXTsIBAIsXboUoaGh6NChA86fP48FCxbou6taWVtbQ6VSISQkhL/N1NQU9erVw6ZNmxAbG4sZM2YAMKy1p8b+vqtLSkrCkydPkJSUxN/Wv39/jBkzBgCwePFiJCYmGtT7DwD//PMPKleuDLlcjqioKHzzzTc4ceIE4uPj8fz5c0ycOBG//vorkpOT9d3VXGVkZCA0NBQbNmyAtbU1/vzzT9y4cQO7du1CYmIiunbtCkC/5z5jDCEhIfD29saoUaP4dYl9+/ZFuXLlEBkZqbe+5eXff/+FtbU1//v//vc/XL9+HX5+fmjWrBl++OEHPfYud1ZWVjh69CgaNmyIKVOmQCKRoFOnTggMDIRIJAJgPHsH9KFx48aIjY3FkSNH0KlTJ9y/fx/VqlVD69atsWbNGgQFBSEqKkrf3UR8fDyePHkCuVyOKVOmAAC2bNmC/fv3o1mzZpg5cyYOHjyo515mbSydOnUqFixYAJVKBRMTE8yfPx///vsv1q9fD8BwPpsZY4iKikJoaCgAYMGCBTh16hQOHDiAtWvXok+fPti+fbte+5gnPQfqpd7t27dZ9erVma+vL3v9+jV/+8ePH1m/fv1Yu3btWFpamh57mBP3rW/OnDmsSZMm7OTJkxr3p6ens4kTJ7K2bdsaXN85xvi+Z3f06FHm4+PDzp07l2NmICgoiFWoUMFg13j9+eefzM/Pj5mbm7OvvvpKY93cxIkTmYeHBz8jbki4meIdO3awNm3asLZt27JZs2ZpHLN3715WvXp1g+j/mTNn2MqVKzVuUyqVzNPTk+3Zs0c/ncrHjBkzmIeHB5s3bx4bN24cE4vF7MCBA+z69evs8OHDzNTUlG3dulXf3czhyZMnrFGjRvzvbdq0YWKxmH3xxRfs8uXLeuyZcXj79i0bNGgQs7S0ZG3btmVyuZy/b9euXczW1jbHZ40+qFQq1qdPH/b999+zTp06sdOnT/P3vXr1ig0YMIB99913LCMjw2BmbBnL6nd8fDzr2rUr69Wrl0H17/nz56xJkyascuXKrEePHkwgELDDhw8zlUrF3r9/z8aPH89atGjB5HK5wfRZHQXKxUylUrF79+6xU6dOsfT0dMZY1mUnxhibO3cuq1evHr/e19BERESwL774gn399dfswoULGvft2LGDeXl5sdjYWP10Lh/G/L6ra9KkCatTp47WoMzBwYGtWrVKD73Knfofuf3797MRI0awq1evMsaYxnIFc3NztnnzZr30sSBevnzJ/Pz8mEAgYAMHDtS479KlS6xq1aosMjJST73TjnvvVSoVq1u3Ljt48CB/3+7du9mtW7f01TUNYWFhbMaMGWzgwIGscePGOQL9li1bsjFjxuinc3nIzMxkzZs3Z1FRUWzgwIHMxcWF/frrr6xr166sQYMG7NKlS/ruosF78+YNmz59Ojt//jxjTPPvReXKlXOsX9aXW7duMYlEwgQCATt69KjGfZMnT2bNmzc3yICOMcb++OMPJhAI2JUrV/TdFQ3Pnz9n+/btY7Nnz2b+/v4a9y1evJjVrl2bKZVKPfUub7T0opgJBALUqlUL7dq1g6mpKQBAKBQCyEq9Vrt2bf53Q8IYQ8WKFbFp0yZERUVh6dKl2LFjB4Csy9L37t2Di4sLLCws9NxT7Yz1fedweUZPnjwJpVKJvn374uHDh/z9CoUCVapUQdmyZfXVRa3UU2b17NkTEyZMQP369QEAJiYmUKlUiIiIQLVq1VCtWjV9djVXjDG4u7tj06ZNqFOnDk6cOIFFixYByLrcef78eTg4OGgsHzAE3GVWgUAAiUQCS0tLAMD06dMxevRo2NnZ6bN7vJo1a2LBggXYvHkz0tLSYG5uzt/HGIOZmRnKlSunxx7mxBhDRkYGGGNo3LgxLl68iBMnTmD06NEYMWIEKlasCE9PT3130+C5uLggMDAQTZs2BfDf34vY2Fg4OjqiTp06+u3g/6tfvz5OnToFANi0aZPG39709HR4eXkhIyNDX93LU6dOndC2bVusX78eSqVS393hVahQAb169UL58uWhVCo18lO/f/8enp6e/OeewdFjkF5q5fdNMy4ujs2YMYM5Ojqyhw8fllCvtMvMzMyxe5/7nfvvw4cPWbdu3ZiXlxdzd3dnrVq1Yra2tuzevXsl3V0NefVdG0N63wuCey2vXr1iPj4+zNvbmy1cuJAdPnyYTZkyhdnb27OIiAg991K7vMbArFmzWK1atfg0d/qSmZnJX2VQv039v0+ePGH+/v7Mzc2NlStXjjVv3pzZ29vr/dzPS1paGvP29mZ//vknmzdvHrO0tDSY2eTsRo8ezQYMGMDu3r3L4uLi2MyZM5mLiwuf1UUf8jp3g4ODWaNGjdjt27c1bv/48WNxd6tUmzVrFqtSpYrBXaW5dOkSc3FxYQ0bNmTDhg1jAwcOZDY2Nuz+/fv67lqeFi1axKytrVl0dLS+u5LDw4cPmY2NDVu6dCnbuXMnmzp1KrO1tWVhYWH67lquBIzR7oNP8fr1a1y7dg2mpqaoUqUKatasmefxp0+fxt69e3Hu3DkcO3YMdevWLaGe5vTo0SMsXrwYr1+/RuXKldGqVSv06dMHQNaMplAo5DcJyOVyREZG4uTJkyhfvjyaNWuGKlWqGHTf1RnS+54fxliOTRiZmZkYPXo0Hjx4gPfv38PJyQm//vqrwb0ObX3nnDhxAufPn8e2bdtw8eJFvc4ePXr0CAsXLsS7d+9QpUoVdOrUCR07dgSQ89yPjY3F69evcerUKbi7u6NRo0aoVKlSifZX2zmdm4yMDPj5+eHDhw+IjIzE33//zc/ql5SC9jc4OBgbNmzA/fv3UbVqVchkMhw6dKjEz+vk5GSoVCowxvK8UpCeno7k5GTY2toCyPt8J/nbu3cvLly4gAMHDuD8+fMG9/cMAJ48eYLg4GBcv34dVapUwZgxY1CjRg19d0sr7nz88OED2rZti4MHDxrklY4LFy5gxIgRMDExgaurK1avXo1atWrpu1u5okD5E9y/fx+dO3eGo6MjXr16hYYNG2LlypUaH6Lchy3n9evXOH36NFq1aoWKFSvqo9sAgMePH6NJkybo3r07qlevjrNnzyIiIgIdOnTAL7/8AgA5LosaisL03VDed3X//vsvtm7dipiYGNSpUwdff/01/+WDO29Y1j4C/hxKSEiAUqmEWCzW66X/gvZdPYgIDAzElStXsH79+ny/UBanJ0+eoFGjRvjqq6/g6emJU6dOwczMDE2bNsXKlSsBGNa5/++//+LYsWPo169frksS1N9rpVKJNm3a4N9//8X58+dL/AOoIP1V/7t469YtPH36FGZmZvjiiy/g5uZWkt3Fo0ePEBAQAJlMhvfv32Pp0qXo37+/xnuaPfDP/nedFE5YWBhmzJiBJUuWwMfHR9/dyROXVcYY/r8zxqBQKCCRSPTdlVzFxcUhPT0dIpGI/+JpsPQwi10q6JpAe9u2bfxlJX0XKkhJSWH9+/dn48eP529TKpV8rmGu0AJn27ZtBlOdqDB9N5T3XR13+alDhw6sR48ezMbGhrVp00Zjgxu3CZExZlDVlj6l7/re/KlSqdiMGTNYr169+NsSExPZ/PnzWZ06ddiIESM0jj9y5AiLiYkp6W7ynj59yuzt7ZlAIGDTp09nMpksxzHalgrs2LFDLzmJC9tffXn48CFzcHBgAQEBbNeuXWzSpEnMzMws16U1+j4fSiNDKzhCSHYUKBeSLgm0//77b1alShXWv39/lp6ebhAfFK1bt2Zz5sxhjDF+p+nUqVNZjx49mK+vL1u2bBljjLHLly+zKlWqsAEDBuRYz6kvuvbdkN53xrI+GAYMGKARlD19+pT17t2bffHFF3yhC87s2bPZ0KFDDWI9cmH7bkgllYcMGcKaN2+ucVtiYiILCgpi9evXZ4sWLWKMMXb8+HFWvnx59uOPP+rlS9bHjx/Z0KFD2ZAhQ9i6deuYQCBgU6ZM0Rp8MsbY0qVL+XGhD4Xp79y5c0u4l/+JjY1l7dq10/jSzRhjLVq0YOPGjWOMaQb1x44d0+v5QAjRD8O/hmCgmA4JtJs1a4apU6di3rx5MDU11XuRAoVCgbS0NERERCAjIwMWFhZ48+YN9u3bh44dO6J69eo4efIkAKBp06aYOnUqfv75Z71niShs3w3hfVdnbm6O9+/f8/1hjKFy5cpYunQpqlWrhoMHD+L48eP88WKxGFevXjWIy2iF7bshZIhg/7/KzNfXF5mZmXjy5Al/n5WVFYYOHYq6devi2LFjSEtLQ8eOHTF06FAMHTpUL5dbTUxMUK9ePXTo0AFjxozB3r17ERQUhKVLl0Iul2scGxcXhzt37uDUqVOIjY0t8b4Wtr8nT55EXFycXvqbnp6O+Ph4+Pv7A/jv0nqFChX4Pqn/zejUqRO+/fZbvZ0PhBA90WeUbswKmkDbkC6Zq7ty5QozMTFhzZs3ZwMHDtQoOXz//n1mZWVlsJkhjLnvGRkZLC0tjX377bfM39+fpaSkMJVKxc9QRUREsMaNG7PevXtrPC4uLk4f3dVgzH1X9+zZMyaVStnQoUNZUlISY+y/mcOoqCgmEAjYsWPH9NlFXvZsCnv37mUCgYD98MMPfMGGjIwM9uHDBxYbG8vevn2rj27yjK2/6lc6uAJEM2fOzJE7+8OHDyXZLUKIAaFA+RMYcwJtxhi7efMmGzBgABs+fDhbt24df/uRI0eYt7c3i4+P12Pv8mZsfc++bOXixYtMKBRqLFXgjrl48SIzMTFhDx480CjSoS/G3PfchISEMJFIxMaOHauxNCA6OprVrl2bXbt2TY+9y0m9ytaePXv4ZQ1v3rxhEydOZF27djWoAjrG1l/1pRQ//vgja9++Pf/7woUL2fLlyzXW3RNCPh+m+p7RNmYVKlRAhQoVsGXLFty+fVtjp7zBJ9AG0KBBA+zcuTPHkoTLly/D2dnZYJYqaGNMfdeWBcDPzw9LlixBQEAAxGIxhg8fzi9tsbKyQtWqVSGRSPhLvPp6Pcbc97y0bNkSBw4cQM+ePREdHY1evXqhVq1a2LlzJ2JiYko880J+hEIhGGNQqVTo06cPBAIBBg4ciKNHjyIiIgI3b96ESCTSdzd5xtbf7FlauHN31qxZmD9/Pu7du8cXLiKEfGb0HKiXCsaYQFubsLAwNmbMGGZtbc1CQ0P13R2dGGrf88oCkJyczObOncsEAgGbOXMmu3v3LouNjWWBgYGscuXKet9db8x9L6g7d+4wPz8/5uHhwSpVqsS8vLzY3bt39d2tXKlUKn6mtlWrVsze3t6g/84YU3+5WeXZs2ezkSNHsmXLljGRSMTu3Lmj554RQvSJAuUiEhISwipVqsSqVKnCWrRowf755x99d0knKSkp7NChQ6xPnz7U9yKSWxYA9SAyMzOT7dixg5UtW5a5urqyatWqMRcXF71/OBtz33WVkJDAXrx4wcLCwnLN0GBIMjIyWEBAABMIBAZ1vufG2Po7f/58JhAImI2NjcFWNCSElBwqOFKEjCqBthapqanIyMgwiOwKujLEviuVSvz2229wcHBA7969sX//fvTp0wc//PADpkyZAkdHR/7YyMhIREVFQaFQoGbNmnB1ddVjz42776VdZmYmtm/fjnr16um1umFBGVt/b9++jYYNG+LBgweoXr26vrtDCNEzCpQJKUbJyckawfu+ffvQt29fTJ48GdOmTYNUKkVGRgbevn0Ld3d3PfY0J2Pue2nHjKx0srH1N/u5Twj5fNHuBEKKEfdhm5mZCRMTE/Tu3RuMMfTr1w8CgQATJ05EUFAQXr58iZ07d0IsFhtMQGHMfS/tjO19Nrb+UpBMCOHQjDIhJYRl7QmAiYkJ9u3bh4EDB6JixYqIiIjArVu3DPqytDH3nRBCCCksCpQJKUHccBMIBGjdujVCQ0Nx8eJF1KxZU889y58x950QQggpDFp6QUgJEggEyMzMxJQpU3DhwgWEhoYaTaBpzH0nhBBCCoMK1hOiBz4+Prh79y5q1aql767ozJj7TgghhOiCll4QogfGlgVAnTH3nRBCCNEFBcqEEEIIIYRoQUsvCCGEEEII0YICZUIIIYQQQrSgQJkQQgghhBAtKFAmhBBCCCFECwqUCSGEEEII0YICZUIIIYQQQrSgQJkQQgghhBAtKFAmhBBCCCFECwqUCSGEEEII0YICZUIIIYQQQrSgQJkQQgghhBAtKFAmhBBCCCFECwqUCSGEEEII0YICZUIIIYQQQrSgQJkQQgghhBAtKFAmhBBCCCFECwqUCSGEEEII0YICZUIIIYQQQrSgQJkQQgghhBAtKFAmpcqQIUPQtWvXPI/x9PTEqlWrCtTe9u3bYWtr+8n9IoRkmTNnDurUqaPvbhBi1GgclRxTfXeAkMKIjIxEhQoVcO/ePZ3/WNy6dQsSiaR4OkYIIYSQUoNmlMlnx9HREWKxWN/dIMRgpKen67sLhBg9GkelEwXKxChVqFABAFC3bl0IBAK0aNFC4/6goCCUK1cODg4OGDt2rMYfsOxLL+Lj4zFq1Cg4OzvDwsICNWrUwPHjx7U+r0wmQ/369dGtWzekpqYW+esipLhFRkZCIBBg37598PPzg4WFBXbt2gWVSoWff/4Z5cuXh0gkQp06dXD69Gn+cRcvXoRAIEB8fDx/W2hoKAQCASIjI/nbNm/eDDc3N4jFYnTr1g0rVqzQunzp999/h6enJ2xsbNCnTx8kJSUV46smpGjROPp8UKBMjNLNmzcBAOfOnUN0dDQOHTrE33fhwgVERETgwoUL2LFjB7Zv347t27drbUelUuGrr77C1atXERwcjEePHmHx4sUQCoU5jn316hWaNWuGGjVq4ODBgxCJRMXy2ggpCYGBgZgwYQLCw8PRvn17rF69GsuXL0dQUBDCwsLQvn17dOnSBU+fPi1wm1evXsV3332HCRMmIDQ0FG3btsWCBQtyHBcREYHDhw/j+PHjOH78OC5duoTFixcX5csjpETQOPoMMEKM0IsXLxgAdu/ePY3bBw8ezDw8PFhGRgZ/W8+ePVnv3r353z08PNjKlSsZY4ydOXOGmZiYsCdPnmh9nt9++43Z2Niwx48fMzc3NzZ+/HimUqmK/PUQUlK4sbNq1SqN211cXNiCBQs0bmvQoAEbM2YMY4yxCxcuMADsw4cP/P337t1jANiLFy8YY4z17t2bdezYUaON/v37MxsbG/732bNnM7FYzBITE/nbpkyZwho1alQEr46QkkHj6PNBM8rE4JQpU4b/+e6773R+vI+Pj8aMcLly5RATE6P12NDQUJQvXx5eXl65tqdUKtGsWTN0794dq1evhkAg0LlPhJQEXcZO/fr1+X8nJibi7du3+PLLLzWO+fLLLxEeHl7g53/y5AkaNmyocVv234Gs5U9WVlb873mNUUJKGo0joo6yXhCDExoayv/b2tpa58ebmZlp/C4QCKBSqbQea2lpmW97IpEIbdq0wfHjxzFlyhS4urrq3CdCSoIuY0fXzC8mJlnzKowx/rbCbl7SZYwSUtJoHBF1NKNMDE7lypX5HycnJ63HmJubAwAyMzM/6blq1aqF169f499//831GBMTE/z++++oV68eWrZsibdv337ScxJSXAoydrSxtraGi4sLrl69qnH71atXUb16dQBZ2WIAIDo6mr9fPaAAgKpVq+LWrVsat2X/nRBDR+OIqKNAmRglJycnWFpa4vTp03j//j0SEhIK1Y6fnx+aN2+OHj164K+//sKLFy9w6tQpjV3KACAUCrFr1y7Url0brVq1wrt374riZRBiMKZMmYIlS5Zg3759ePLkCQIDAxEaGooJEyYAyAoe3NzcMGfOHDx9+hQnTpzA8uXLNdoYN24cTp48iRUrVuDp06fYuHEjTp06RcuVyGeDxlHpQ4EyMUqmpqZYs2YNNm7cCBcXF3zzzTeFbuuPP/5AgwYN0LdvX1SvXh1Tp07VOlNtamqKPXv2wMfHB61ataK1YKRUGT9+PCZNmoTJkyejZs2aOH36NI4ePYoqVaoAyLrMu2fPHjx+/Bi1atXCkiVLMH/+fI02vvzyS2zYsAErVqxA7dq1cfr0aQQEBMDCwkIfL4mQEkfjqPQRMPWFMoQQQkgRGjFiBB4/fozLly/ruyuEGC0aR/pDm/kIIYQUmaCgILRt2xYSiQSnTp3Cjh078Ouvv+q7W4QYFRpHhoNmlAkhhBSZXr164eLFi0hKSkLFihUxbty4QqV5JORzRuPIcFCgTAghhBBCiBa0mY8QQgghhBAtKFAmhBBCCCFECwqUCSGEEEII0YICZUIIIYQQQrSgQJkQQgghhBAtKFAmhBBCCCFECwqUCSGEEEII0YICZUIIIYQQQrSgQJkQQgghhBAt/g/wtyb0eThligAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 760x760 with 9 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "s[1].rough.value = 2.\n",
    "fitter.initialise('covar')\n",
    "fitter.sample(20, nthin=10, pool=1, random_state=1234);\n",
    "objective.corner();"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.13.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}
