{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Lab 4-1: Linear regression\n", "\n", "---" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import numpy as np\n", "import pandas as pd\n", "import scipy.stats as stats\n", "import matplotlib.pyplot as plt\n", "%matplotlib inline" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Load in a csv file with snow water equivalent (SWE) measurements from two snow pillow sites (which measure the mass of snow) in California's Sierra Nevada. \n", "\n", "(If you're interested, [read about SWE and snow pillows here](https://www.nrcs.usda.gov/wps/portal/wcc/home/dataAccessHelp/faqs/climateHydrologyFaqs))" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
yearsBLC_maxSLI_max
019836882446
119841121471
219852161143
\n", "
" ], "text/plain": [ " years BLC_max SLI_max\n", "0 1983 688 2446\n", "1 1984 112 1471\n", "2 1985 216 1143" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data = pd.read_csv('../data/pillows_example.csv')\n", "\n", "data.head(3)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This data give us the peak SWE value (mm) at the Blue Canyon (BLC), and Slide Canyon (SLI) measurement sites." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Plot the SWE data from the two sites" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "fig, ax = plt.subplots()\n", "\n", "data.plot(x='years', y='SLI_max', c='r', ax=ax, label='Slide Canyon')\n", "data.plot(x='years', y='BLC_max', c='b', ax=ax, label='Blue Canyon')\n", "\n", "ax.set_title('Timeline of Peak Snow Water Equivalent (SWE)')\n", "ax.set_xlabel('Water Year')\n", "ax.set_ylabel('Peak SWE (mm)');\n", "plt.legend(loc=\"best\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**What does the above plot show?**\n", "\n", "What you see above is a plot of the maximum value of snow water equivalent (SWE) measured at two snow pillows (these weigh the snow and convert that weight into the water content of the snow). These measurements of snow are not too far from each other geographically (both in the Sierra Nevada, California, although Slide Canyon is at a higher elevation and further south), and we might expect that more snow at one site woud correspond to more snow at the other site as well. We can check this by examining a regression between the data at the two sites.\n", "\n", "\n", "**The first step to any regression or correlation analysis is to create a scatter plot of the data.**\n" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "fig, ax = plt.subplots(figsize=(4,4))\n", "\n", "# Scatterplot\n", "data.plot.scatter(x='SLI_max', y='BLC_max', c='k', ax=ax);\n", "\n", "ax.set_xlabel('Slide Canyon max SWE (mm)')\n", "ax.set_ylabel('Blue Canyon max SWE (mm)');\n", "\n", "ax.set_xlim((0,3000))\n", "ax.set_ylim((0,1000));" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Linear regression**: Could we use SWE measurements at Slide Canyon to predict SWE at Blue Canyon?\n", "\n", "The plot above suggests that this is a borderline case for applying linear regression analysis. What rules of linear regression might we worry about here? ([*heteroscedasticity*](https://en.wikipedia.org/wiki/Heteroscedasticity))\n", "\n", "We will proceed with calculating the regression and then look at the residuals to get a better idea of whether this is the best approach." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "---\n", "\n", "### Manual calculation of linear regression\n", "\n", "Here we'll first compute it manually, solving for our y-intercept, $B_0$, and slope $B_1$:\n", "\n", "$B_1 = \\displaystyle \\frac{n(\\sum_{i=1}^{n}x_iy_i)-(\\sum_{i=1}^{n}x_i)(\\sum_{i=1}^{n}y_i)}{n(\\sum_{i=1}^{n}x_i^2)-(\\sum_{i=1}^{n}x_i)^2}$\n", "\n", "$B_0 = \\displaystyle \\frac{(\\sum_{i=1}^{n}y_i)-B_1(\\sum_{i=1}^{n}x_i)}{n} = \\bar{y} - B_1\\bar{x}$" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "B0 : 127.9143\n", "B1 : 0.1997\n" ] } ], "source": [ "n = len(data) # length of our dataset\n", "\n", "x = data.SLI_max # using x for shorthand below\n", "y = data.BLC_max # using y for shorthand below\n", "\n", "B1 = ( n*np.sum(x*y) - np.sum(x)*np.sum(y) ) / ( n*np.sum(x**2) - np.sum(x)**2 ) # B1 parameter, slope\n", "B0 = np.mean(y) - B1*np.mean(x) # B0 parameter, y-intercept\n", "\n", "print('B0 : {}'.format(np.round(B0,4)))\n", "print('B1 : {}'.format(np.round(B1,4)))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Then our linear model to predict $y$ at each $x_i$ is: $\\hat{y}_i = B_0 + B_1x_i$" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [], "source": [ "y_predicted = B0 + B1*x" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "And our residuals are: $(y_i - \\hat{y}_i)$" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [], "source": [ "residuals = (y - y_predicted)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Finally, compute our Sum of Squared Errors (from our residuals) and Total Sum of Squares to get the correlation coefficient, R, for this linear model.\n", "\n", "$SSE = \\displaystyle\\sum_{i=1}^{n} (y_i - \\hat{y}_i)^2$ \n", "\n", "$SST = \\displaystyle\\sum_{i=1}^{n} (y_i - \\bar{y}_i)^2$\n", "\n", "$R^2 = 1 - \\displaystyle \\frac{SSE}{SST}$\n", "\n", "And compute the standard error of the estimate, $\\sigma$ for this model.\n", "\n", "$\\sigma = \\sqrt{\\displaystyle\\frac{SSE}{(n-2)}}$" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [], "source": [ "sse = np.sum(residuals**2)\n", "\n", "sst = np.sum( (y - np.mean(y))**2 )\n", "\n", "r_squared = 1 - sse/sst\n", "r = np.sqrt(r_squared)\n", "\n", "sigma = np.sqrt(sse/(n-2))" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "SSE : 999651.24 cfs\n", "SST : 1219919.85 cfs\n", "R^2 : 0.181\n", "R : 0.425\n", "sigma : 204.089\n" ] } ], "source": [ "print('SSE : {} cfs'.format(np.round(sse,2)))\n", "print('SST : {} cfs'.format(np.round(sst,2)))\n", "print('R^2 : {}'.format(np.round(r_squared,3)))\n", "print('R : {}'.format(np.round(r,3)))\n", "print('sigma : {}'.format(np.round(sigma,3)))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Plot our results:" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAA+gAAAEYCAYAAADPrtzUAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Z1A+gAAAACXBIWXMAAAsTAAALEwEAmpwYAACBQklEQVR4nO3deXxcZdn/8c+VfWu6Jd3SJQVKFShQKAiCGxYroFJxo4qi5hGf5+e+oFQfBVdA3FDc0Kr4iEVQRMSFVQREoIVSWpZCoSm06ZIuSZM0e67fH+dMO2lnkpl0JjOTfN+v17xm5sw5Z66ZJCfnOvd9X7e5OyIiIiIiIiKSWXmZDkBERERERERElKCLiIiIiIiIZAUl6CIiIiIiIiJZQAm6iIiIiIiISBZQgi4iIiIiIiKSBZSgi4iIiIiIiGSBtCXoZvZLM9tuZmujlk0wszvN7LnwfnzUa0vNbL2ZrTOzRVHLTzSzNeFrPzAzS1fMIiIiIiIiIpmSzhb0XwNvPGDZJcDd7j4HuDt8jpkdBZwPHB1u82Mzyw+3+QlwETAnvB24TxEREREREZGcl7YE3d3vA3YdsPhc4Lrw8XXA4qjlN7h7p7tvANYDJ5vZVKDS3f/j7g78JmobERERERERkRGjYJjfb7K7bwFw9y1mNilcXgM8FLXepnBZd/j4wOUxmdlFBK3tlJeXn/iyl70shaGLyEjy6KOP7nD36kzHka2qqqq8trY202GISJbSMXRgOoaKyGDiHUeHO0GPJ9a4ch9geUzufi1wLcCCBQt85cqVqYlOREYcM9uY6RiyWW1tLTqGikg8OoYOTMdQERlMvOPocFdx3xZ2Wye83x4u3wTMiFpvOtAQLp8eY7mIiIiIiIjIiDLcCfqtwIXh4wuBP0ctP9/Mis1sNkExuEfC7vAtZnZKWL39fVHbiIiIiIiIiIwYaevibmbLgdcCVWa2CbgUuAK40czqgBeBdwC4+5NmdiPwFNADfMTde8Nd/Q9BRfhS4O/hTURERERERGRESVuC7u5L4rz0+jjrfwP4RozlK4FjUhiaiIiIiIiISNYZ7i7uIiIiIiIiIhKDEnQRERERERGRLJAt06yJiIiISIJuWbWZq25fR0NTO9PGlXLxorksnl+T6bBGLDObC/w+atFhwJfd/fuZiUhERiol6CIiIiI55JZVm1l68xrau4N6upub2ll68xoAJelp4u7rgOMBzCwf2Az8KZMxicjIpC7uIiIiIjnkqtvX7UvOI9q7e7nq9nUZimjUeT3wvLtvzHQgIjLyqAVdREREJIc0NLUntVxS7nxg+YELzewi4CKAmTNnDndMg6q95K9p3X/9Feekdf8io4Va0EVEcpCZjTOzP5jZM2b2tJmdamYTzOxOM3suvB8ftf5SM1tvZuvMbFEmYxeRQzNtXGlSyyV1zKwIeAtw04Gvufu17r7A3RdUV1cPf3AiMiIoQRcRyU1XA/9w95cBxwFPA5cAd7v7HODu8DlmdhRBi8/RwBuBH4djKEUkB128aC6lhf3/hEsL87l40dwMRTSqnAU85u7bMh2IiIxMStBFRHKMmVUCrwaWAbh7l7s3AecC14WrXQcsDh+fC9zg7p3uvgFYD5w8nDGLSOosnl/D5efNo7woSNLHlRZy+XnzVCBueCwhRvd2EZFUUYIu/TQ2NrJixQoaGxszHYqIxHcY0Aj8ysxWmdkvzKwcmOzuWwDC+0nh+jXAS1HbbwqX9WNmF5nZSjNbqWOASHZbPL+GRcdMAeBDrz5MyfkwMLMy4Ezg5kzHIiIjlxJ02Wf58uXMmjWLM888k1mzZrF8uS4Qi2SpAuAE4CfuPh9oI+zOHofFWOYHLdD4SZGc0tbZA8DO1q4MRzI6uPted5/o7s2ZjkVERi4l6AIELed1dXW0t7fT3NxMe3s7dXV1akkXyU6bgE3u/nD4/A8ECfs2M5sKEN5vj1p/RtT204GGYYpVRNKkrTOYam33XiXoIiIjhRJ0AaC+vp6ioqJ+ywoLC6mvr89MQCISl7tvBV4ys0hFqNcDTwG3AheGyy4E/hw+vhU438yKzWw2MAd4ZBhDFpE0aI20oLcpQRcRGSk0D7oAUFtbS1dX/3/w3d3d1NbWZiYgERnMx4Drwyl/XgA+QHDR9UYzqwNeBN4B4O5PmtmNBEl8D/ARd+/NTNgikiqRLu67laCLiIwYStAFgOrqapYtW0ZdXR2FhYV0d3ezbNkyNA5VJDu5++PAghgvvT7O+t8AvpHOmERkeEVa0HcpQRcRGTGUoMs+S5YsYeHChdTX11NbW6vkXEREJIspQRcRGXmUoEs/1dXVSsxFRESynLvT1tlDYb7R3t1Le1cvpeG86CIikrtUJE5EREQkx3R099HnMH18GQC7VMldRGREUIIuIiIikmMi3dunjy8FVChORGSkUIIuIiIikmMiFdxnTAha0DXVmojIyKAEXURERCTHRFrQZ4YJulrQRURGBiXoIiIiIjkmkqDPGK8WdBGRkUQJuoiIiEiOiXRxnzauhPw8Uwu6iMgIoQRdREREJMdEWtDHlBQyvqxQVdxFREYIJegiIiIiOaatsxeAiuICxpcVsatVCbqIyEigBF1EREQkx0S6uJcX5zO+vEgt6CIiI4QSdBEREZEcE+niXl5UwMTyInZpDLqIyIigBF1EREQkx7R29lBWlE9enjG+vEhF4kRERggl6CIiIiI5pq2zh4riAgAmlhexe28XfX2e4ahERORQKUEXERERyTGtUQn6+LIi+hya27szHJWIiBwqJegiIiIiOaats4fySAt6RRGACsWJiIwAStBFREREckxbZy/lxflA0IIOqFCciMgIoARdREREJMdEd3GfUK4EXURkpFCCLiIiIpJjWqO6uCtBFxEZOZSgi4iIiOSYNiXow87MxpnZH8zsGTN72sxOzXRMIjLyFGQ6ABERERFJTmtnD2PCBL2kMJ+yonzNhZ5+VwP/cPe3m1kRUJbpgERk5FGCLiIiIpJDenr76Ozp29eCDkGhOLWgp4+ZVQKvBt4P4O5dgL5wEUm5jHRxN7NPmdmTZrbWzJabWYmZTTCzO83sufB+fNT6S81svZmtM7NFmYhZREREJBu0dfYC9EvQJ5QXaZq19DoMaAR+ZWarzOwXZlYevYKZXWRmK81sZWNjY2aiFJGcN+wJupnVAB8HFrj7MUA+cD5wCXC3u88B7g6fY2ZHha8fDbwR+LGZ5Q933CIiIiLZoLWrB4CK4v2nQxPK1YKeZgXACcBP3H0+0EZ4rhrh7te6+wJ3X1BdXZ2JGEVkBMhUkbgCoNTMCgjG7zQA5wLXha9fBywOH58L3ODune6+AVgPnDy84YqIiIhkh9aOIEE/qAVdCXo6bQI2ufvD4fM/ECTsIiIpNewJurtvBr4NvAhsAZrd/Q5gsrtvCdfZAkwKN6kBXoraxaZwmYiIiMio09qpBH24uftW4CUzmxsuej3wVAZDEpERatAicWa2AHgVMA1oB9YCd7n7rqG8YTi2/FxgNtAE3GRmFwy0SYxlHmffFwEXAcycOXMo4YmIiIhktbYwQR9zQIK+t6uXju5eSgo1EjBNPgZcH1ZwfwH4QIbjEZERKG4Lupm938weA5YCpcA6YDtwOnCnmV1nZkPJghcCG9y90d27gZuBVwLbzGxq+N5Tw/eCoMV8RtT20wm6xB9EY39ERERkpGuL04IOsFuF4tLG3R8PzzOPdffF7r470zGJyMgzUAt6OXCau7fHetHMjgfmEHRVT8aLwClmVkbQIv96YCVBsY0LgSvC+z+H698K/M7MvkvQij8HeCTJ9xQREREZESJd3CsOmGYNYGdrF1PHlmYkLhEROXRxE3R3/9FAG7r740N5Q3d/2Mz+ADwG9ACrgGuBCuBGM6sjSOLfEa7/pJndSDDOpwf4iLv3DuW9RURERHKdWtBFREauRMagzyYYc1Mbvb67v2Wob+rulwKXHrC4k6A1Pdb63wC+MdT3ExERERkp2roi86D3n2YNUKE4EZEcN2iCDtwCLAP+AvSlNRoRERERGVBLRw+F+UZxgRJ0EZGRJpEEvcPdf5D2SESyWGNjI/X19dTW1qIChCIikkltnT39xp8DjC0tJM9gtxJ0EZGclsg86Feb2aVmdqqZnRC5pT0ykSyxfPlyZs2axZlnnsmsWbNYvnx5pkMSEZFRrK2zp9/4c4D8PGNcWRE7laCLiOS0RFrQ5wHvBc5gfxd3D5+LjGiNjY3U1dXR3t5Oe3swoUFdXR0LFy5US7qIiGREa4wWdAi6uatInIhIbkskQX8rcJi764gvo059fT1FRUX7knOAwsJC6uvrlaCLiEhGtHUd3IIOMKGsiJ2tOl0TEclliXRxXw2MS3McIlmptraWrq7+Jzvd3d3U1tZmJiCRkJnlm9kqM7stfD7BzO40s+fC+/FR6y41s/Vmts7MFmUuahFJhdbO3pgJ+vjyQrWgi4jkuEQS9MnAM2Z2u5ndGrmlOzCRbFBdXc2yZcsoLS2lsrKS0tJSli1bptZzyQafAJ6Oen4JcLe7zwHuDp9jZkcB5wNHA28Efmxm+YhIzmrt6Kai+OA/4wnlxariLiKS4xLp4n7gfOUio8qSJUtYuHChqrhL1jCz6cA5wDeAT4eLzwVeGz6+DrgX+Hy4/AZ37wQ2mNl64GTgP8MYsoikUFtnL+VFscagF7J7bzd9fU5enmUgMhEROVSDJuju/i8AM6tMZH2Rkai6ulqJuWST7wOfA8ZELZvs7lsA3H2LmU0Kl9cAD0WttylcdhAzuwi4CGDmzJkpDllEUqWts4eKklgJejG9fU5LRw9jywozEJmIiByqQbu4m9lFZrYNeAJYCTwa3ouIyDAzszcB29390UQ3ibHMY63o7te6+wJ3X6ALUiLZyd1p64pXxT1Iyne2dQ53WDnDzMrNLC98fKSZvcXMdDVDRLJGIi3iFwNHu/uOdAcjIiKDOg14i5mdDZQAlWb2W2CbmU0NW8+nAtvD9TcBM6K2nw40DGvEIpIy7d299Dmxi8SVFQGoUNzA7gNeFRbSvJug0eldwHsyGpWISCiRInHPA3vTHYiIiAzO3Ze6+3R3ryUo/naPu18A3ApcGK52IfDn8PGtwPlmVmxms4E5wCPDHLaIpEhrZw8QO0GfWF4MoKnWBmbuvhc4D/ihu78VOCrDMYmI7JNIC/pS4EEzexjY12fK3T+etqhERCRZVwA3mlkd8CLwDgB3f9LMbgSeAnqAj7h7b+bCFJFD0doRJOixqriPD7u4qwV9QGZmpxK0mNeFy1RjSUSyRiIHpJ8B9wBrgL70hiMiMnKY2QLgVcA0oB1YC9zl7rtSsX93v5egWjvuvhN4fZz1vkFQ8V1EclxbZ3B9LVYV930t6JpqbSCfIGh8+lN4AfMw4J8ZjklEZJ9EEvQed//04KuJiAiAmb0f+DiwgaCw5jqC8eKnA583s7XAl9z9xYwFKSI5KdLFPVYV99KifEoK89itBH0gk939LZEn7v6Cmd2fyYBERKIlkqD/M5x65y/07+KekhYgEZERqBw4zd3bY71oZscTjAVXgi4iSWmLJOgxxqBD0Iq+q617OEPKNUuBmxJYJiKSEYkk6O8O75dGLXPgsNSHIyKS+9z9R4O8/vgwhSIiI0xbV/wicRCMQ9+ladYOYmZnAWcDNWb2g6iXKgnqc4iIZIVBE3R3nz0cgYiIjDRh1fSPAbVEHW+ju1eKiCSjdZAW9PFlRezaqxb0GBoIplR7C8HQo4gW4FMZiUhEJIa4CbqZne7uDwzweiUw093XpiUyEclKjY2N1NfXU1tbS3V1dabDyXa3AMsIhgipyKaIHLK2AaZZA5hYXkT9zrbhDCknuPtqYLWZ/c7ddQVDRLLWQC3obzOzbwH/ILjS2EhQ5OgI4HXALOAzaY9QRLLG8uXLqauro6ioiK6uLpYtW8aSJUsyHVY263D3Hwy+mohIYiLTrJUVHjzNGsD48iJ2awz6QE42s8sIzmMLAAPc3Qcdumlm9QQt7r0ERZQXpDFOERml4ibo7v4pMxsPvJ1gPt2pBNMEPQ38bKDWdREZeRobG6mrq6O9vZ329qD2WV1dHQsXLlRLenxXm9mlwB30L7L5WOZCEpFc1trZS0VxAXl5FvP1ieVFtHb20NnTS3FB7CR+lFtG0KX9UYJEO1mvc/cdqQ1JRGS/Acegu/tu4OfhTURGsfr6eoqKivYl5wCFhYXU19crQY9vHvBe4Az2d3H38LmISNLaOnsoL46feI8vLwJgd1s3U8YqQY+h2d3/nukgRETiSaSKu4gItbW1dHX1n1u3u7ub2trazASUG94KHObumpRYRFKitasn7vhzgAllQYK+q62LKWNLhiusXPJPM7sKuJnkezY5cIeZOUFv0mujXwynJb4IYObMmamLWERGFSXoIpKQ6upqli1bRl1dHYWFhXR3d7Ns2TK1ng9sNTAO2J7hOERkhGjr7IlbwR1gQvn+BF1iekV4Hz1+PNGeTae5e4OZTQLuNLNn3P2+fTsJEvZrARYsWOCpClhERhcl6CKSsCVLlrBw4UJVcU/cZOAZM1tB/5YaTbMmIkPS1tlDeVECCfpeJeixuPvrDmHbhvB+u5n9CTgZuG/grUREkjPQNGufc/dvhY/f4e43Rb32TXf/wnAEKCLZpbq6Wol54i7NdAAiMrK0dPQwfXxZ3Nf3JeitnXHXGY3M7AJ3/62ZfTrW6+7+3UG2Lwfy3L0lfPwG4KtpCFVERrmBWtDPB74VPl4K3BT12hsBJegiIgNw938BmFkl6rEkIinQ1tVDxQBF4saVFWEGu/ZqqrUDlIf3Y4a4/WTgT2YGwfH8d+7+j1QEJiISbaATRovzONZzERE5QFgw6GsEU1T2Ec63Cww6366ISCxtnb1UlMQ/fcvPM8aVFrJbY9D7cfefhfdfGeL2LwDHpTQoEZEYBkrQPc7jWM9FRORgFwNHa85cEUmV1s6Bq7hDMNWaisTFZmbTgR8CpxGczz4AfMLdN2U0MBGR0EBH+OPMbA9Bi09p+JjwuebtEBEZ3PPA3kwHISLZ45ZVm7nq9nU0NLUzbVwpFy+ay+L5NQlt293bR1dPHxUDFImDYKo1Jehx/Qr4HfCO8PkF4bIzMxaRiEiUuEd4d48/wElERBKxFHjQzB6mfxX3j2cuJBHJlFtWbWbpzWto7+4FYHNTO0tvXgOQUJLe1tkDMGgL+oTyIjbu1LXBOKrd/VdRz39tZp/MVDAiIgfKi/eCmX3fzN5uZtOGMyARkRHkZ8A9wEPAo1E3ERmFrrp93b7kPKK9u5erbl+X0PYtHUGCPtA86BAk6JpmLa4dZnaBmeWHtwuAnZkOSkQkYqAj/HrgPODbYcXKB4F/h/er3b0v/eGJiOS0HnePOaWPiIw+DU3tSS0/UFtX4i3ou9u6cHfCczjZ74PANcD3CMagPxguExHJCgN1cb+G4ACGmU0lKKbxSuBTwCSgcjgCFEmVxsZG6uvrqa2t1TzeMlz+GVZy/wv9u7jvylxIIpIp08aVsjlGMj5tXGlC20e6uA9UxR2CBL2nz9nT0cPY0sLkAx2hzGwxcATwI3d/S4bDERGJKW4XdwALHAucG95eQ9Cy/p1hiE0kZZYvX86sWbM488wzmTVrFsuXL890SDI6vJtwHDr7u7evzGhEIpIxFy+aS2lh/xI/pYX5XLxobkLbt3YG3eMHmgcdYHxZEYCmWotiZj8maGSaCHzNzL6U4ZBERGKKewnWzO4kaCV/nGD85Dfd/elhikskZRobG6mrq6O9vZ329qDloq6ujoULF6olXdLK3WdnOgYRyR6RQnBf+NMa9nb1UlGcz9cXz0u4invCReIqggR9Z1sXtVXlhxDxiPJq4Dh37zWzMuB+4GsZjklE5CADtaC/QDA2Z054O8LMqoYlKpEUqq+vp6ioqN+ywsJC6uvrMxNQLuvqgh/9CH7600xHktXM7PRBXq80s2OGKx4RyR6L59dw6mETAXjDUVMSTs4hmAMdoDyBadZALegH6HL3XgB330swbbCISNYZaAz6hyE4kQROIRh//hEzqwbWuvuFwxOiyKGpra2lq6v/SUp3dze1tbWZCSgX9fTAb38LX/kK1NfDuefCf/93pqPKZm8zs28B/yDo1t4IlBCMfXwdMAv4TObCE5FM2tbSAUBja+cga/bXmkQVd0Bzoff3MjN7InxswOHhcwPc3Y/NXGgiIvsNfIQPdAJ7gfbw8XSgaMAtBmFm44BfAMcQtNJ/EFgH/B6oBeqBd7r77nD9pUAd0At83N1vP5T3l9GlurqaZcuWUVdXR2FhId3d3Sxbtkzd2xPR1wc33wxf+hI88wyceCL87Gdw5pmZjiyrufunzGw88HbgHcBUgmPo08DP3P2BTMYnIpm1bU+QmO9oTS6BTmYedEBTrfX38kwHICKSiIHGoH+PoNV8DsE49AcJ5vS90N2bDvF9rwb+4e5vN7MioAz4AnC3u19hZpcAlwCfN7OjgPOBo4FpwF1mdmSkm5JIIpYsWcLChQtVxT1R7vCPf8D//i889hgcdRT88Y/w1reCpuxJSHiB8efhTUQEgO7ePna0RhL0JFvQu3ooKsijqGDAGr+UFeVTXJCnLu5R3H1jpmMQEUnEQJdgNwDXA6tSmQyHXeZfDbwfwN27gC4zOxd4bbjadcC9wOcJqsff4O6dwAYzWw+cDPwnVTHJ6FBdXa3EPBH33w9f+AI88ADMng2/+Q28+92QP3DVYBERGdyO1k7cg1buXW1d9PU5eXmJXfhs6+wZtHs7gJkxobyInUrQRURyzkCXYP8MPBdJzs3sdWZ2tZl9Omz1HqrDCMZj/srMVpnZL8ysHJjs7lsAwvtJ4fo1wEtR228Klx3EzC4ys5VmtrKxsfEQQhQZhR59FM46C179anj+efjxj4Nu7e99r5JzEZEUiXRvP3paJb19TlN7d8LbtnX2Uj7IFGsR48uK1IIuIpKDBkrQbwTKAczseOAm4EXgOODHh/CeBcAJwE/cfT7QRtCdPZ5Yl5U91orufq27L3D3BWopFUnQ00/DO94BCxbAI4/AVVfB+vXwP/8DRYdUbkJERA6wtTkoEHfUtEoguW7urZ09g1Zwj5hYoRZ0EZFcNFCCXuruDeHjC4Bfuvt3gA8QdDEfqk3AJnd/OHz+B4KEfZuZTQUI77dHrT8javvpQAMicmg2bID3vx+OOQZuvx0uvTRY9tnPQllZpqPLaWb2uajH7zjgtW8Of0Qiki22hxXcj542FoAdLYkn6Il2cYewBV1F4g5iZnPM7A9m9pSZvRC5ZTouEZGIgRL06JbrM4C7Ady971De0N23Ai+Z2dxw0euBp4BbgcjUbRcSdLEnXH6+mRWb2WyConWPHEoMIqPali3w0Y/C3Lnw+9/Dpz8NL7wAl10GlZWZjm6kOD/q8dIDXnvjcAYiItll254O8vOMuZPHALAjiVbu1s6eQSu4R0woL2JXklXiR4lfAT8BegimvfwN8H8ZjUhEJMpAR/l7zOxGYAswHrgH9rVuH+oR/2PA9eFY9hcIWuXzgBvNrI6gK/07ANz9yTCOpwgOph9RBXeRIdi5E771LfjhD6G7G/7rv4Iq7TUxSzrIobE4j2M9F5FRZNueTiaNKWbSmGIguRb01s4eZoxPrIfThPIiWjp76OrpG7Tq+yhT6u53m5mFld0vM7P7gUszHZiICAycoH8SeBfB/L2nu3ukiskU4IuH8qbu/jiwIMZLr4+z/jeAbxzKe4qMWi0t8P3vw7e/HTx+z3uC1vLDD890ZCOZx3kc67mIjCLb9nQwubKEsaWFFORZUmPQk+niHpkLvWlvF5MqS+Kud8uqzVx1+zoamtqZNq6UixfNZfH8EX3htsPM8oDnzOyjwGb2FyYWEcm4uEd5d3fghhjLV6U1IhFJjfZ2+MlP4PLLYceOYA7zr341GHMu6Xacme0haC0vDR8TPo9/piwiWS0Vyey2PR3MrionLy+YCi25BL03qS7uADvb4ifot6zazNKb19DeHXRM3NzUztKb1wCM5CT9k0AZ8HHgawTd3N+XyYBERKKpz5PISNPdDT/7GcyZA5/5DMyfH1Rnv/lmJefDxN3z3b3S3ce4e0H4OPK8MNPxiUjyIsns5qZ2nP3J7C2rNie1n217OpkcJsxVFcXsTHCcuLvT1tVDRRLTrAEDTrV21e3r9iXnEe3dvVx1+7qE3iNH1bp7q7tvcvcPuPvbgJmZDkpEJEIJugxZY2Mjd9xxB3fccQeadz4L9PbC9dfDy18O//3fMGsW/POfcMcdcNJJmY5uVDGz75vZ281sWqZjEZHUSEUy29HdS3N79/4EfUxxwi3oe7t6cSfhFvSJFftb0ONpaGpPavkIcWDhznjLREQyYtCjvJlNcvftByyb6+4j+vKqDGz58uW8//3vp6sr+MdfWFjIddddx5IlSzIc2SjkDrfeGhR8W7sWjjsObrsNzj4bTPXIMmQ9cB7wbQt+Bg8C/w7vVx/qbBgiMvxSkcxu2xNMsba/Bb2I57e3JrRta2cPkHiCvq8FfYCp1qaNK2VzjPinjStN6D1yiZmdBZwN1JjZD6JeqiQoQiwikhUSaUG/38zeGXliZp8B/pS+kCTbNTY2UldXty85B+ju7uaDH/ygWtKH2913wymnwOLF0NUVTJv22GNwzjlKzjPI3a9x93e7ey1wKnAzcDhwE9B0qPs3sxlm9k8ze9rMnjSzT4TLJ5jZnWb2XHg/PmqbpWa23szWmdmiQ41BZLSJl7Qmk8xu2xO0lk+uDCq4V1UU09jaSVD2Z2CRBD3xedCD0TQDdaG/eNFc8vP6/68oLczn4kVz42yR0xqAlUAH8GjU7VYgoWOimeWb2Sozuy1tUYrIqJdIgv5a4L1mdpOZ3QccCZyc1qgkq9XX15OXd/CvTn5+PvX19cMf0Gj0n//AGWfAwoXBvObLlsGTT8I73wkxfjYy/CxwLHBueHsNQcv6d1Kw+x7gM+7+cuAU4CNmdhRwCXC3u88B7g6fE752PnA0wTzsPzazxAayigiQmmQ2Vgt6V0/fvuR7IG1JJugF+XmMLS0csAX9zKMmU2DB5zCgZlwpl583b0QWiHP31e5+HXCEu18XdbvZ3XcnuJtPAE+nMUwRkcG7uLv7FjP7B8H4nD5gqbsn1h9LRqTa2lr6+g7uodvb20ttbe3wBzSaPPFE0JX9L3+BSZPgBz+Aiy6C4uJMRyZRzOxOgm6TjwMPAd9095Sd1Ln7FmBL+LjFzJ4GagguBLw2XO064F7g8+HyG9y9E9hgZusJLrT+J1UxiYx0i+fXcM09z7G+sQ0Iktlkq7gfnKCHc6G3djGmZOD6kcl2cYegkvuuAcag//nxBjp7nT9e9ApOnDU+7nojTK2ZXQ4cRdSsGu5+2EAbmdl04ByCaX8/ndYIRWRUS2QM+p0EJ4LHANOBX5rZfe7+2XQHJ9mpurqaZcuWHTQG/Ze//CXV1dUZjm6EevZZuPRSuOEGGDcOvvlN+NjHoKIi05FJbC8AxwFzgJ3ADjNrdPcdqX4jM6sF5gMPA5PD5D1ycTUyt28NwYWCiE3hsgP3dRFwEcDMmSpqnC1G4TzVWWv33m4ACvON+z73uoNa1AezbU8HJYV5VJYEp1/7E/ROZleVD7htW2dQoC7RFnQYOEF3d3770EZePrWSE2aOS3ifI8CvgEuB7xFMsfYBgikwB/N94HPAmHgrjPZjaO0lf810CFmt/opzMh2C5IhE+sL+yN3f5+5N7r4WeCXQnOa4JMstWbKETZs2cfvtt3P77bezefNmFYhLh5degg99CI46KigE94UvwAsvwNKlSs6zmLt/2N1PARYTtGKfCPzWzB41s+tS9T5mVgH8Efiku+8ZaNVYYR60wP1ad1/g7gt0sS07pGpqLzl0O1o72dnWxRGTKujudbaGreHJiEyxFhaP3FdpfUfL4JXc2/a1oCc+OmV8WfwE/fGXmnhqyx4uOGXmvnhGiVJ3vxswd9/o7pcBZwy0gZm9Cdju7o8OtJ6OoSKSCol0cb/lgOc9wNfSFZDkjurqat7whjdkOoyRafv2oJX8Jz8Jnn/0o0FSPnlyZuOSZHUCe4H28PF0oCgVOzazQoLk/Hp3vzlcvM3Mpoat51OByAwcm4AZUZtPJyiYJFluoKm91Io+vJ7d1gLA618+ifXbW3lp115qkqx2vnVPx77u7QDVkRb0AbqhR7QkOQYdYGJ5EU9saor52m8fepHyonzOPX7U/R51mFke8JyZfRTYDEwaZJvTgLeY2dkE3eIrzey37n5BmmMVkVFo0BZ0MzvFzFaYWauZdZlZr5mpBV2yTmNjIytWrMjtSvJNTcEY88MOg2uugfe+F557Dr7/fSXnOcTMvmdmDxMMD/oqQZfInwFz3X1eCvZvwDLgaXf/btRLtwIXho8vBP4ctfx8Mys2s9kEXe8fOdQ4JP1G6TzVWenZrWGC/rLgWPzSrr1J72P7AQn6hPIizJJtQU88QR9fXsTuvV0HVYlv2tvFbU808NYTapJK+EeITwJlwMcJeje9l/3HzZjcfam7Tw9n5jgfuEfJuYikSyJH5WsIDkY3AQuA9xGc3IlkjeXLl1NXV0dRURFdXV0sW7Yst7rct7XBD38IV14ZJOnnnw9f+QoceWSmI5Oh2QBcD6xy997BVh6C0whOKteY2ePhsi8AVwA3mlkd8CLwDgB3f9LMbgSeIqgA/5E0xSUpNprmqc52z25vZWxpIcfPGIcZvLQ7uYsk7s62PZ0sHLO/qGdBfh7jy4rY0ZpYgm4GZUWJd3GfWF5Ed6/T2tnTrwjdHx7dRGdPH+95xaykPsNI4O4rwoetBOPPRUSySkLzMbn7eiDf3Xvd/VfsrxIsknGRednb29tpbm6mvb2durq63GhJ7+wMEvPDDw+6sJ9+OqxaBcuXKznPbX8GnoskwWb2OjO72sw+bWaH3MXd3R9wd3P3Y939+PD2N3ff6e6vd/c54f2uqG2+4e6Hu/tcd//7ocYgw+PiRXMpLeyfkI3geaqz2rNbW5g7eQxFBXlMrSxhU5It6Hs6emjv7mXK2JJ+yyeWJ5agt3b2UFFUkNR48fHlweEmehy6u/O7h1/kxFnjefnUyoT3NRKY2YVm9piZtYW3lWb2vmT24e73uvub0hWjiEgiCfre8ITycTP7lpl9Chi41KjIMKqvr6eoqH/OU1hYmN1zsvf0wK9+FSThH/84vPzl8OCDwfRpxx+f6ejk0N1IeJw0s+MJeiC9SFDZ/ceZC0tyzeL5NVx+3jwKwmrhBnzzrcdo/Pkwc3fWbWvhyClBcc7pE8p4aXdyCfr2sKjcpMr+CXpVRTE7Wwcfg97W2ZNU93aACeVBq3l0gv7g8zt5YUcbF5wyuqqMh4n4J4HPANMIZrL4HPCJZJN0EZF0SiRBfy+QD3wUaCMoNPS2dAYlkoza2tp9071FdHd3Z+ec7H19cOONcMwx8MEPBuPK77wT7rkHTj0109FJ6pS6e6QI2wXAL939OwTdKU/OXFiSixbPr6G0KJ+K4gIcOGH0zFedNbbt6aSlo4cjJwczbM0YX8ZLu5Lr4r5tT9BKPjmqiztA1ZjiBLu49yZVwR1gQnnwXtEJ+vUPb2R8WSFnHTM1qX2NAP8PeKu7/9Pdm8PZie4hOKf9fxmOTURkn0ET9HAKinZ33+PuX3H3T4dd3kWyQmRe9tLSUiorKyktLWXZsmXZNSe7O/ztb3DiifCud0FBAfzpT/Dww7BwIYyuKW5Gg+gf6BnA3QDu3peZcCSXtXR009LRw2vnBse0tZsHmlFP0mFdWMF9X4I+oZRtLR109iReymFb2II++aAW9CJ2JNCC3tLZk3RBtwll/bu4b9/TwR1PbuMdC2ZQUphcsj8CVLp7/YELw2Wjq6+/iGS1RKq4v8nMVpnZLjPbY2YtZqazA8kqS5YsYePGjdx1111s3LgxuwrE/etfwdjyc86BPXvg//4PVq+GxYuVmI9c95jZjWZ2NTAeuAcgnPps8DNxkShbmoPE7nVzJ1GYb6zZrIlUhlukgnt0C7o7bE6iUNzWuAl6Ma2dPXR0D5zsD6mLe0X/BP33K16ip89598mjq3t7aKAflqZFEJGskciR/vvAecAaP3CeDpEsUl1dnV2t5itXwhe/CHfcAdOmwU9/GnRrLywcfFvJdZ8E3gVMBU539+5w+RTgi5kKSnJTpIp7bVUZc6eMYa0S9GH37LYWqiqKmRAWXZsxoQwIKrkfVl2R0D627+mgsqSA0gOqsFeFSfSO1k6mjy+Lu31bZw8TyuO/Hkt5UT5F+Xns2ttFb5+z/JEXedWcKmqrRmUpoZeb2RMxlhtw2HAHIyISTyIJ+kvAWiXnIgl68kn40peCLuwTJ8K3vw3/7/9BqaZFGi3C4+UNMZavykA4kuO2NAUtr1PHlnLMtLH8fe1W3D2pat5yaJ7d1sLcKfsT8RkTguN5MnOhb9vTeVDrOQQt6AA7WrsGTNBbO3sYk2QLupkxvryQ3W1d3PPMdhqaO/jym49Oah8jyMszHYCISCISOdJ/Dvibmf0L2FfFxN2/m7aoRHLRCy/AZZfBb38LFRXBPOaf/CRUamibiAxdQ1M7+XnGpDHFHFMzlhtWvMSm3e37WnElvfr6nGe3tXL+yTP2LZs8poSi/LykKrlv3dNx0BRrEJWgtwxcKG4oXdwhKBS3q62L6x/eyOTKYha+fFLS+xgJ3H1jpmMQEUlEIkf6bwCtQAlwyPP3iow4DQ3w9a/Dz38eFH/77Gfh858PWs9FRA5RQ3M7k8cUU5Cfx7yasQCs3dysBH2YbG5qp727d9/4c4C8PKNmfCmbkqjkvn1PB4dXVx20fGJUF/eBBFXch5KgF/Jkwx627ung42fMoSA/kQl8REQkUxI50k9w9zekPRKRXLNjB1x5JVxzTTCv+UUXBWPOp03LdGQiMoI0NLUzbVzQpXrulDEU5AWF4s6aN+qmycqIdQcUiIuYPr404Rb0vj5ne0snkyuLD3ot0oK+sy1+/cjOnl66evuoSHKatVtWbeaxjU20hwXoxperBoqISLZL5DLqXWamBF0kYs+eoPv6YYfBd74D73wnrFsHP/qRknMZkJldZ2Y/MbNjMh2L5I4tzR1MDRP0ksJ8jpw8RpXch9H+Kdb6F4ObMaEs4THou/Z20dPnMbu4lxTmM6a4gMYBuri3dQYJdjIt6Les2szSm9fsS84Brvz7Om5ZtTnhfYwkZnZ3eH9lpmMRERlIIgn6R4B/mFm7plmTUa29PSj4dthhwVjzM8+ENWvguuuCZSKDuwa4C3hvpgOR3NDX52xp6mBaVGJ3TE0lazc3o9qtw+O5bS1MG1vCmJL+rc8zxpexe283rZ09g+5jazhV3qQxByfoAFVjigfs4t4WvkcyCfpVt6/rl5wDtHf3ctXt6xLexwgz1cxeA7zFzOab2QnRt0wHJyISMeiR3t3HDLaOyIjW1QW//CV87WvBePNFi4Ix5wsWZDoyySFmlgesc/cVwB8zHY/khp1tXXT19u3r4g4wr2YsN67cxOam9gGrfktqrNvWypFTDj4Viq7k/vKpAxcD3d4SmQP94C7uEEy1NlCCHrkIkEwV94am2OPj4y0fBb4MXAJMBw4sdOzAGcMekYhIDKoUIhJPby/83//By18O//M/MHs23Hsv/OMfSs4lIWb2OzOrNLNy4ClgnZldnOm4JHdsaQ6Sqan9WtAjheJGfme2W1Zt5rQr7mH2JX/ltCvuGfbu2T29fTy/vZW5k2Mk6OHFkUS6uW/bEyTfsaZZA5hYXszO1vhj0IfSgh59USeR5SOdu//B3c8CvuXurzvgpuRcRLKGEnSRA7kHc5gfeyy8730wdiz87W9w//3wmtdkOjrJLUe5+x5gMfA3YCbq3i5JiLR2RidVL59aSX6esTaLx6GnIrGOjKHe3NSOE1RTX3rzmmFN0jfu2ktXbx9zYiXoYRX9l3YP3iK9tbkDM6geE6cFfUxiLejJJOgXL5pLaWH/onKlhflcvGhuwvsYidz9a2b2FjP7dnh7U6ZjEhGJpgRdJMId7rwTTj4ZzjsvaEG/8UZYuRLOOgvMMh2h5J5CMyskSND/7O7dBF0pRRLS0BR0jY5O0EsK85kzqSJrC8WlKrHOhjHUz4YV3GO1oI8vK6S8KD+hFvTtLR1MLC+mMM4UZ1UVxeze2013b1/M1yNF4iqSSNAXz6/h8vPmUTOuFANqxpVy+XnzWDy/JuF9jERmdjnwCYJeTU8BnwiXiYhkhYSO9GaWD0yOXt/dX0xXUJIdGhsbqa+vp7a2lurq6kyHk14PPhhMkXbvvTBrFvzqV3DBBcG85iJD9zOgHlgN3Gdms4CR3y9ZUqahqZ2SwjzGl/UvUHZMzVj++cx23B3LsouHAyXWySSH2TCGet22FszgiEkVB71mZsyYUMamBKZa27Yn9hRrEZGp1na1dcXsBt/a2Q1AeZLTrC2eXzPqE/IYzgGOd/c+CGbXAFYBSzMalYhIaNAWdDP7GLANuBP4a3i7Lc1xSYYtX76cWbNmceaZZzJr1iyWL1+e6ZDS4/HH4U1vgtNOg6efhh/+MJgy7f3vV3Iuh8zdf+DuNe5+tgc2Aq/LdFySO7Y0dzBtbOlBSfi8mrHsbOti656ODEUWX6oS62wYQ/3ctlZmTiijtCh2Yjx9fBkv7Uqsi/uUOOPPISgSB8Tt5t46hBZ0GdC4qMdjMxWEiEgsiRzpPwHMdfed6Q5GskNjYyN1dXW0t7fT3h6ceNTV1bFw4cKR05K+bh1cein8/vcwfjxccQV89KNQXp7pyGQEMLNPD7LKgRWERWLa3NQeMyGNFIpbs6mZqWOzq+jXtHGlbI6RjCebWF+8aC6f/+MTdPb07/Z99rwphxRfMtZta+HIGN3bI2ZMKOXB53cM2pNhe0sHx80YF/f1SAv6jjiF4oZSJE7iuhxYZWb/BAx4NWo9F5EsksgY9JeA7BzoJmlRX19PUVFRv2UFBQX87W9/o7GxMUNRpcjGjVBXB0cdBbfdBv/7v/DCC/D5z+dEct7Y2MiKFSty/+cw8o0Z5CaSkC3N7f0quEccNbWSPCMrC8VdvGguJYX9Ty+GUpxs8fwazj9pBhBkUVPHljB9XCnXPbiRf67bnqpw4+rs6aV+RxtHTj64e3vEjPFl7O3qZVdb/Ars3b197GjtSqiL+46W2C3obZ09FBfkxR3DLolz9+XAKcDN4e1Ud78hs1GJiOyXyKXYF4B7zeyvwL7/HO6uFqARqra2lq6u/icbLS0tfOxjH+N//ud/WLZsGUuWLMlQdEO0bRt885vw058Gzz/+cVi6FCZNymxcSVi+fDl1dXUUFRXR1dWVmz+HUcLdv5LpGCT3dfX0sb2lM2bLc2lRPkdkaaG4xfNr2NnWydduexoIiqld+uajhzQWelLYLXztVxZRXlxA094u3vOLh/nw/z3Kz9+3gNccmb5eXRt2tNHT54O0oO+v5D6xInYC3tgy8BRrAFVjIi3o8bq496h7ewq5+xbg1kzHISISSyKXYl8kGH9ehFqARoXq6mqWLVtGaWkpY8bs/1G3tLTQ3t5OXV1d7rTg7t4NX/gCHHYY/OhHcOGFsH49fO97OZWcRw87aG5uzr2fwyhlZiVm9hEz+7GZ/TJyy3RcqZTpeapHsm17OnCHaeNiJ3bH1IxlzeY9uGffxADHTNs/rPcDp80ecqGyDTvamDSmeF/X7nFlRfy27hUcXl3BRb9ZyQPP7UhJvLGsi1RwnzJwF3cYeC70SJ2AgcaglxflU1yQFzdBb+vsUfd2EZFRYtAE3d2/ErYGfRf4TtRzGcGWLFnCxo0b+eEPf9gvSQcoLCykvr4+M4ElqrU1aDGfPRsuvxzOPTcoAnfttTBjRqajS1qsYQc58XOQ/wOmAIuAfwHTgZaMRpRC2TBP9UgWaw70aPNqxrKjtZPtcbpFZ1IkKc0zqN/RNuT91O9oo7aq//Cj8eVFXP9fr2B2VTl1163g3+vTk6Q/t62V/DxjdlX84U8zxkda0OMn6NvD72LSAF3czYyqimJ2xhmD3qoEXURk1EikivsxZrYKWAs8aWaPmtnR6Q9NMq26upqzzz6bnp6efsu7u7upra3NTFCD6eyEH/wADj88mDbtNa+B1avhd7+DOXMyHd2QxRp2kNU/B4k4wt2/BLS5+3UE0/vMy3BMKZMN81SPZFuag8QuXhG4eVGF4rJNZP72eTVj2bDzEBL0nW3MnnhwgjwhTNJrJwZJ+oPPpz5JX7ethdlV5RQXxJ/arLy4gAnlRQNWct+2Z/Au7hB0c28csIt7clOsycHMLM/M1g5x2xIze8TMVpvZk2amxioRSYtEurhfC3za3We5+yzgM8DP0xuWZIvo7u6VlZWUlpaybNmy7Kvm3tMDy5YFSfgnPgHHHAP/+Q/8+c9w7LGZju6Q5czPQQ7UHd43mdkxBNP51GYunNTKhnmqR7LN+1rQYyd2R02rxIysHIe+pbmdypICjq4Zy4YhtqC3dHSzo7XroBb0iIkVxVz/oVcwY3wZdb9eyUMvpHaymWe3tTB3gPHnETPGlw44F/rWPR0U5hsTyorirgNQXVE0QBX3XrWgp0A49/lqM5s5hM07gTPc/TjgeOCNZnZKKuMTEYHEEvRyd/9n5Im73wsccrlrM8s3s1Vmdlv4fIKZ3Wlmz4X346PWXWpm681snZktOtT3luREurvfddddbNy4MbsKk/X1BVOlHX00/Nd/wbRpcNddcPfdcMrI+r+Z1T8Hiefa8Fj2JYKCRE8B38psSKmTDfNUj2RbmtsZV1ZIWVHsxKysqIDDqyuyspJ7Q1MH08aVMntiOU17u2naG7/KeTz1O4Kkd3ZVWdx1qiqK+d2HTqFmfCkf/PUKHtmwa8gxR2vv6uXFXXuZM0AF94jpE8oGHIO+bU8Hk8aUkJcXfxo2gInlxQOOQVeRuJSZStAj9G4zuzVyG2wjD7SGTwvDW/YVgBCRnJdQFXcz+xLBWEqAC4ANKXjvTwBPA5Xh80uAu939CjO7JHz+eTM7CjgfOBqYBtxlZke6e2+snUp6VFdXZ1drrTv89a/BNGmrVwct5n/+M7z5zTDAXLS5Lut+DjIgd/9F+PBfwGGZjCUdLl40l6U3r+nXzX0o02lJbA1NHUwbZI7zeTVj0zYG+1BEpoeLtH5v2NHG/JkDtyAfKNI1Pl4LekT1mGJ+96FXcP61D/GBXz3CdR88mQW1E4YWeGj99lbcSbAFvYw7ntxKb5+THyMJ376nc8Dx5xFVY4rY1dZFX58flMyrintKDblrupnlA48CRwA/cveHD3j9IuAigJkzh9JILyNZ7SV/Tev+6684J637l+GTSAv6B4Fq9s8XWQV84FDe1MymE4zF/EXU4nOB68LH1wGLo5bf4O6d7r4BWA+cfCjvLznu3nvhtNOCZLy1Fa6/PkjS3/KWEZ2cS+4xsy/HumU6rlRZPL+Gy8+bx4TyIPGqHlPM5efNG3LFbumvoak9bvf2iGNqxrK9pXNfIbJssaW5g6njSvcVWBtKN/eN4TazJgzeaW/SmBKWf+gUJlWW8P5freCxF3cn/X7R1m0LajkeOUAF94gZE0rp7nW2xfkZbN3TMWAF94iqimJ6+5ym9u6DXlMV99Rx938B9UBh+HgF8FiC2/a6+/EEBT9PDocuRb9+rbsvcPcFupguIkOVSIJe4+4fd/cTwtsn3f3Q/vPB94HPAX1RyyaH81JG5qeMzIFVA7wUtd6mcNlBzOwiM1tpZis1/dQI9MgjcOaZ8LrXwYsvws9+FlRmf/e7IS+RX2WRYdcWdesFzmIEjUGHIEn/9QdOAuCbb1VynkoNTe1xC8RFRArFrW3Inm7uHd297GrrYtrYEmZOKBtyJfcNO9uYOraE0qLEiqNNrgyS9KqKIi5c9girDiFJf25bC0X5ecyaEL97fcS+Su5xurlv29MxaIE4CBJ0OHgu9L4+p61LY9BTxcw+BPwB+Fm4qAa4JZl9uHsTcC/wxhSGJiICJJag/zSsWvk/ZjbuUN/QzN4EbHf3RxPdJMaymGN+dOVyhFq7Ft76VnjFK+Dxx+G73w3mMr/oIigszHR0InG5+3eibt8AXkucC4y5LJJ8bM2yVtxc1trZw56OnkHH8+8rFLdpzzBFNrhI9fkpY0spKshj+vgyXhhCgl6/o43aGBXcBzJlbAnLLzqFCRVFvG/ZI6x+qSnp94WgBf3wSRUU5A9+mjRjQmSqtYOLI+7t6qGloyehLu4TK4KeKDsOmDavrSuYSUVV3FPmI8BpwB4Ad3+O/Y1CcZlZdeQ82MxKgYXAM+kLU0RGq0TmQT+dYNz5TGClmf3OzM48hPc8DXiLmdUDNwBnmNlvgW1mNhUgvN8err8JiJ64ejrQcAjvL7ni+efhgguCKuz33ANf+xq88AJ86lNQMnhrhEgWKmMEjkWvqigmz8i6bta5bMsgFdwjKooLmF1VnlWV3PfFPjaIvbaqnPohTLVWv3MvtQMUiItn6thSln/oFMaVF/LeZQ8PaRq6Z7e2MDeBAnEQ/IzMYregb49MsTZm8P9Z1ZEW9Lb+BfXaOoMaD2pBT5lOd9/3JZtZAYkVe5sK/NPMniDoFn+nu9+WphhFZBRLqF+wuz8L/C/weeA1wA/M7BkzOy/ZN3T3pe4+3d1rCYq/3ePuFxBUOL4wXO1C4M/h41uB882s2MxmA3OAR5J9X8khmzfDf/83vOxlcPPN8LnPwYYNQUG4MYOPBxTJFma2xsyeCG9PAuuAqzMdV6rl5xnVY4rjjsGV5DWErdCJVMSfVzM2qyq575u/PYx99sQy6nfsxT3xgtfN7d3sautKugU9Ytq4IEmvLC3kgmUPJ/X9tHR009DcwZwECsQBFBfkM6WyhJdiTLUW6VUyZWwSXdwPaEFv7Yy0oCtBT5F/mdkXgNKwwekm4C+DbeTuT7j7fHc/1t2Pcfevpj1SERmVBk3QzexYM/seQcX1M4A3u/vLw8ffS2EsVwBnmtlzwJnhc9z9SeBGgumJ/gF8RBXcR6gdO+Czn4XDD4df/jJI0p9/Hq64AiYcWkVekQx5E/Dm8PYGYJq7X5PZkNJjcmUJW/fEniJqpLpl1WZOu+IeZl/yV0674h5uWbU5ZfuOzCU/NYHEbl7NWLbu6aCxJTu+/y3N/WOfXVVOa2cPjXGmEIslMmZ9sAruA5k+vozlHzqFiuICLlj2ME81JDYM4NltwUxaiVRwj5gxvoxNuw7u4h65aDU5gS7uY0sLKcizg8agtylBT7VLgEZgDfBh4G8EjVAiIlkhkRb0a4BVwHHu/hF3fwzA3Rs4xAOau9/r7m8KH+9099e7+5zwflfUet9w98Pdfa67//1Q3lOyUHMzXHopzJ4N3/seLFkCzz4LP/whTJ2a6ehEkmZmE8xsAtASdWsHKsPlI87kypJR1cX9llWbWXrzGjY3tePA5qZ2lt68JmVJ+pamdvKMhIqLHZNlheIamjuYUF5ESWEwZjqSZEfmNU9EpEv87ENI0CEYH778Q6dQVpjPe37xEE9vGTxJfzZSwT2JBH36hNKYLeiRLu6TEvg55uUZE8qL4ibo6uKeGu7eRzBb0NcIply7zpPp3iEikmaJjEF/tbv/xt0PujTs7v8XaxuRhOzdC1ddBYcdBl/9Kpx1VlAQ7le/gtraTEcncigeBVaG943As8Bz4eNEC2TmlMmVxaOqSNxVt6/rN/87QHt3L1fdvi4l+9/c1MGkMSUUJlCk7KhplQCsHcJY63TY0tTer+V/9r4EPfFx6Bt2tGEGMxOooj6YmRPLWH7RKRQX5POeXzzMuq0tA67/7LYWSgvzmT5+8OEFETPGl7F1TwedPf1/J7bu6aCsKJ8xCSbXVRXF7GztPwZdXdxTy8zOAZ4HfkDQCLXezM7KbFQiIvsl0sX9NDO708yeNbMXzGyDmb0wHMHJCNXVBT/+MRxxRDC+/BWvgEcfhRtvhJe/PNPRiRwyd5/t7ocBtxMMC6py94kEXd5vzmx06TGlsoSmvd10dI+OEUiRLuiJLk/WlubB50CPqCwpzKpCcVuaO/pND1czrpTCfEuqknv9jjamjS3d1wp/qGZNLOeGi06hMN94988f2tdKHsuz21o4cnIFeXmxJpGJbcaEMtyhoan/RarIFGtmie2rakzxQS3orWpBT7XvAK9z99e6+2uA15HaIZsiIockkS7uy4DvAqcDJwELwnuR5PT2wm9+A3Pnwkc+Eow1v+8++Nvf4IQTMh2dSDqc5O5/izwJh+i8JoPxpE2kC+/2UTIOvXpM7DHFUxNMqgezpbljX5G1RBxTM5YnExxjnW4NTf0vLhTk5zFjQllyLehDrOA+kNqqcpZ/6BTy84Ikff322En6uq2tCReIi5gRtrYfWMl9+55OJsX5XYmlqqKIHa0HVnGPJOiaZi1Ftrv7+qjnL7B/5iARkYxLJEFvdve/u/v2cJz4TnffmfbIZORwhz/+EebNgwsvDAq+/f3vQXL+qldlOjqRdNphZv9rZrVmNsvMvghk5PhpZm80s3Vmtt7MLkn1/qeECfq2lpHfzb27t4/C/NgtomOKCw5qAU2Wu9PQ1E5NEgn6vJpKNje1s+uAKbqGW1s4f3t0CzrAYVXlbEiyBX2oFdwHclh1Bb/70CmAseTnD/N8Y2u/13e1dbGjtTOpAnEQPRd6/wR9W0tHQhXcI6oqimls7exX8b41nGZtTHFhUjFJf2Z2Xjj70JNm9jcze7+ZXUhQwX1FhsMTEdknkQT9n2Z2lZmdamYnRG5pj0xynzvcfjucdBK8/e20t7fTvGwZrFwJb3wjJNjlTySHLQGqgT8BtwCTwmXDyszygR8BZwFHAUvM7KhUvkekmNnW5pGfoF9zz3o2N3XwgdNqqRlXigE140p454LpbNi5l7Ovvp+HXhj6dZhdbV109vQlVME94phpQaG4THdzj1RwP7B7fu3EYC70vr7Ba3Htbuuiub37kAvExXPEpApuuOgVuDtLrn2IF6KS9H0F4qYkl6BPriyhMN94KaqSu7uztbkjoUJ/EVUVRXT19O3r1g7BRY88g5LChGbGlfgiM2qUANsIejO9lqA2yPjMhSUi0l8iA5peEd4viFrmBNOsicT2wAPwxS/CfffRWl3NpwoL+eOuXXR89KMsKy1lyZJhz1FEhl04G8UnMh0HcDKw3t1fADCzG4BzCaavTIl9LegjvFDcmk3N/Oif63nr/BouffPRXPrmo/u9/v5Xzuajv3uMd//8IT595pHUjC3l23c+G3b7LuXiRXNZPL9mwPeIjGNOZA70iKMjldw3N/OaI6uT/FSpE4l9ygFJaW1VOZ09fWzd0zHo59oQVnBPRwt6xBGTxvC7D53CkmsfYsnPH+L3F51KbVV5VAX3iqT2l59n1IzrX8l9T3sPnT19SXZxD+dCb+1iTEnQYt7a2UN5cUHC49glNnf/QKZjEBFJRCJV3F8X46bkXGJbtQrOPjvouv7ss7RccQU1LS38orub3Xv20N7eTl1dHY2NjZmOVCRtzOz74f1fzOzWA28ZCKkGeCnq+aZwWT9mdpGZrTSzlcn+jVaWFlBckDeiE/SO7l4+c9PjTKwo4rIDEvOIo6ZVcuvHTudNx07j23c8y2f+sDrpqdgaIq3QYxNP0P/5zHby84yrbl+X8jnZkxHpQXFgEn5Y2BqeSDf3VMyBnogjJ4/h+g+9gu5eZ8nPH+IX97/AlX9/BoC3/+TBpL/DGRPK2BQ1Bj0y3CO5FvRIgr5/mERbZ48quKeQmc02s++a2c0ZPi6LiMSUUH8pMzvHzD5nZl+O3NIdmOSYZ56Bd74zKPb28MNw5ZXw/PM8c8YZWHH/1oPCwkLq6+szE6fI8IhMQfltgorBB96GW6ymt4P6Grv7te6+wN0XVFcn1wprZkwZW8K2EVwk7nt3Pcuz21q58m3HMrYs/njgiuICrj7/eMaVFnJgj+5EpmKLVIJPtIp7ZE723vDNUj0nezIamtuxGPO31yaZoOelaIq1wbxsSiW/rXsFzXu7+MZfn6atKxjvvbmpI+nvcPr4Ml7avb+Le+RiRTJj0CdWFAGwo2X/31GkBV1S5hagHvghmT0ui4jENOgR38x+CpQRTEPxC+DtwCNpjktyxcaN8JWvwHXXQVkZfPnL8OlPw9igu2VtbS1dXf2LFnV3d1Orec5lBHP3R8P7f0WWmdl4YIa7P5GBkDYBM6KeTwcaUv0mk8eUjNgW9Ec37uLn973AkpNn8Nq5kwZd38xobu+O+dpgU7Ftae6guCCPCeVFCcU20Jzsg3WnT7UtTR1UVRRTVND/+v+UyhKKC/ISquS+YedeasaXHrSPdDlqWiXlJYXs7e5/cSnZ73DGhFJ2tXXRFibUkb+FyWMST9CrIy3oUcX+lKCnXIe7/yDTQYiIxJPIf79Xuvv7gN3u/hXgVPqf6MlotHUrfOxjMGcO/O538MlPwgsvBMl6mJwDVFdXs2zZMkpLS6msrKS0tJRly5aRbOtcIhobG1mxYoW6z0vWMLN7zazSzCYAq4Ffmdl3MxDKCmBO2LWzCDgfSHmXzkmVxSMyQW/v6uWzNz3BtHGlfPGcxGvrxRtrPdgY7M3hePVExxyne072ZDQ0tzMtRotxXp4xO8FK7umq4D6Q6BbraMl8hzPG96/kvj3c56TKxMegTygvwqx/PG2dPYxRgp5KV5vZpSp+LCLZKpEEPfLfaa+ZTQO6gdnpC0my2q5dsHQpHHYY/OQn8IEPwPr18J3vQJyke8mSJWzcuJG77rqLjRs3pqVA3PLly5k1axZnnnkms2bNYvny5Sl/D5EhGOvue4DzgF+5+4nAwuEOwt17gI8CtwNPAze6+5Opfp8plUEX9+gpokaCK//xDBt2tHHV249LaizwxYvmUlrYf+7q0sJ8Ll40d8DttjS1J1XBfagXAtJhS3PHQVOsRdROLN9XAC4ed6d+R1vaKrjHk4rvcN9Ua2El963NHYwrK6SkMPH5ywvy8xhfVnTAGPRezYGeWvOADwFXsL97+7czGpGISJREzjRuM7NxwFXAYwTjFn+RzqAkC7W0wNVXw1VXBY/f/W647DI44oiENq+urk5LqzkELed1dXW0t7fT3h6cGNXV1bFw4cK0vadIggrMbCrwTuCLmQzE3f8G/C2d7zG5soT27l72dPQwtjR752y+ZdVmrrp93YCV1SPrbA5bUF81p4pTD5+Y1PtE9vnV255iV1sXVRVF/O85RyVUxf20I6oSfp+LF81l6c1r+nVzLy7IG/RCQKq5O1ua2nnVnNixz64u566nt9HT20dBfuz2gZ1tXbR09gx7C3qs7zCRiynRZowPkvmXwkJx2/Z0JNW9PWJief8EXV3cU+6twGHu3jXomiIiGZBIFfevuXuTu/8RmAW8zN2/lP7QJCt0dMD3vw+HHw5f+hK87nWwejX89rcJJ+fpVl9fT1FR/7GaKkQnWeKrBK3Wz7v7CjM7DHguwzGlzeSw1Xd7FndzjxRUG6iyevQ6ESvqdw2p6Nri+TX86+LXkp9nvOukGYMm5929fWxv6aAmwQJxkfe4/Lx5++ZkBzh59oRhH3++p6OHtq7euNXnZ08sp6fP+32vB4qMUR/uFvQDv8OacaVcft68pL7DCeVFlBXl82IkQW/pTKp7e0RVRTE7W/fnjm1dquKeYquBcZkOQkQknrhHfDO7ADB3j1Qjxt07zex9Ztbm7r8blgglM7q74de/hq9+FTZtgoUL4etfh1e8ItORHUSF6CRbuftNwE1Rz18A3pa5iNJrcjjf89Y9HcyZPCbD0cQWr6Da5//4BPc928j48iJuWvnSQet0dPcNuejamJJC5s8YxwPP7eDiRQOvu21PB30OU5Psnr54fs2+2P7f9Y/y0Au76O7tozBOS3U6bAmnh5sa5+JCdCX3WXFayDcM0xRrsUR/h0NhZswYX8amyBj0PR0cOSnxnhARVWOKWbOpCQh6JbR2qAU9xSYDz5jZCmBfVwV3f0vmQhIR2W+g/9yfIZiK4kA3hK/JSNTXB8uXw1FHwUUXwfTpcPfdcOedWZmcw/AWohNJhpkdaWZ3m9na8PmxZva/mY4rXSLTSWXzVGvxin519vSxYuMufr/iJfZ09CS1bSJOn1PFE5ubado7cK/aLXHmEU/GefOns6uti3+tG96CmVuagtjjjZ+fncBUa/U728jPM6aPH/7x86kwY0IpL+1qp7fP2d7SmdQc6BFVFUXsCFvQO3v66OlztaCn1qUE3dy/iaZZE5EsNFCCnu/uLQcuDJdl7+BCGRp3+MtfYP78YHx5WVnw/MEH4YwzMh3doIajEJ3IEPwcWEpQXJNwirXzMxpRGkWSkWyu5B4v8a0ZV8r9nzuDtV9ZFLMK+UDbJuJVc6pwhwef3zngevvmQE+iSNyBXjO3monlRfzxsU1D3sdQRC4uxCsSV1VRREVxwYBTrdXv2MuM8aXD2vKfSsFc6HvZ2dpJb58zeYhd3Fs7e+jo7qWtM7hYVF6kInGp4u7/inXLdFwiIhED/QcsNLOD+piZ2RggsclZJTfccw+ceiq85S3Q3h60oK9aBW96EwwwzU+2TWtWXV3NSSedpJZzySZl7v7IActiN8+OACWF+YwtLczqBP1jZxx+0LIDi4F97o0vG1L19YEcN30cY4oLuP+5HQOu1xBphT6EiwGF+Xm85fhp3P309kFb7FNpS3M7eQaTxsROSs2M2qoyNuzcG3cfG3a0ZaR7e6rMmFDG3q5ent4atG9MGmILOsCO1k7aOoOhFhUlahdJFTNrMbM94a3DzHrNbE+m4xIRiRgoQV8G/MHMaiMLwsc3hK9Jrnv44WBs+etfD5s3w89/Dk89BeefD3kDt15oWjORhOwws8MJZr/AzN4ObMlsSOk1ubKYrc3Zm6AXFQSJd1VFcdxiYKkoGHaggvw8Tjl8Ig+sH/iCZkNTO5UlBYfcpfltJ0ynq7eP254Yvl+3hqYOJleWxK3QDjC7qoINO1pjvubu1O8c/jnQUylSyf3R+l1AMPVgsqoqggscO1q7aA1b0Cs0zVrKuPsYd68MbyUEdUGuyXRcIiIRcc8A3P3bZtYK/MvMKghOMNuAK9z9J8MVoKTern/9C/vylxl/333B3OXf/z58+MNQktiJhKY1E0nYR4BrgZeZ2WZgA/CezIaUXpMrS9jWkr1j0G9d3UDNuFIe+PzrsAF6CB1qwbBYXjWnijuf2sbGnfGLpG1pbk/J/OVHT6tk7uQx/PGxTVxwyqxD3l8itjQPPn/77Ill/PWJBjp7eiku6J90NrZ0srerd9gruKdSZC70FfW7AYY4Bj1M0Fs6GVsWtJyrSFz6uPstZnZJpuMQEYkYsJnU3X/q7rMIpleb7e6zlJznsPXrqT/tNMa99rXk3XcflxUUcNOVV8InPpFwcg6a1kwkUe7+grsvBKqBlwGvBU7PaFBpNrmyhG1Z2oK+q62LB57bwZuPmzZgcp4up4dzmw/UzX1zU0dKEnQz47wTalj1YhMvNMZusU61Lc0dg3bNn11dTp/vnys8WiYruKdKJEF//KUm8mx/d/VkTIzq4h5pQVeCnjpmdl7U7e1mdgVhLycRkWyQUBUWd2+NVTBOMiPpsd8vvQQXXYS/7GVUP/ggVwCzga/09HDhRz6S9BhyTWsmMjAzqzSzpWZ2jZmdCewFLgTWA+/MbHTpNbmymMawQFa2+fvaLfT0OW85blpG3n92VXnQej9Agh60oA+9QFy0xfNryDP40xDmb0+Wu9PQ1D5ocbtI9/UNOw5O0Ot3hnOg53AX94riAsaXFdLe3UtVRfGA3f3jibSg72zrorUj0sVdCXoKvTnqtghoAc7NaEQiIlFys0zqKJbU2O/GRvj0p2HOHPj1r9n+trcxf8wYvgjsDlcZSsu3pjUTGdT/AXOBNcCHgDuAdwCL3X1EnwhOqSyht8/Z2ZZ93dxvfbyBw6vLefnUzMzRbmacfkQVDz6/I+YFjL1dPTTt7Y5bBT1ZkytLOH1ONTc/tpm+NF8w2b23m86evkFjj3Rfj1XJvX7nXgrzLWUXKDIl0oo+lO7tEBRbHFNcQGNL5/4q7krQU8bdPxB1+5C7f8Pdtw+2nZnNMLN/mtnTZvakmX1iOOIVkdFHR/wckvDY7+Zm+M534Hvfg7174f3vhy9/mbyyMjbN6j8Wcagt30uWLGHhwoXU19dTW1ur5Fykv8PcfR6Amf0C2AHMHA09kSJVq7c1dzJpTPYkWlubO3ikfheffP2RGeneHnH6nCp+v/IlntjUxPyZ4/u9FqngXpOCLu4Rbzuhhk/c8DgPb9jFqYdPTNl+D7RverhBkutxZUWMLyvkhVgJ+o42ZkwoG1KrczaZMb6MJzY1D2mKtYiqMcVhF/fgd0Et6IfOzL48wMvu7l8bZBc9wGfc/bFwRqNHzexOd38qdVGKiCTQgm5mZWb2JTP7efh8jpm9Kf2hyYEGHfu9dy9ceSXMng1f+xqcfTY8+SQsWwazZqW85VvTmonE1R154O69wIbRkJzD/qrV2TbV2m1PNOAObz5uakbjOO2IKsyI2c09kuQOVmgtGW84agoVxQXcnOY50SNzoE9JoPW/tqo8Zgv6hh1tOd29PaK9O2j1vuvp7Zx2xT3cMoQhBlUVRf2mWdM86CnRFuMGUAd8frCN3X2Luz8WPm4BngZSW0lSRITEWtB/BTwKnBo+3wTcBNyWrqAktrhjv6dNgx/9CL7+ddi6Fc45J0jQ588/aB9q+RYZFsdFzatrQGn43AhaaiozF1p6Rbr1bs2yBP0vqxs4pqaSw6orMhrHhPIijp5Wyf3rd/Cx18/p99qW5kgrdOpa0EuL8jl73hT++sQWvnLu0ZQVpacldmsk9gQuLsyuKuc/z+/st8zd2bhzL6eFhfRy1S2rNvcrAri5qZ2lN68BSGpWgInlxTzf2EpbVw8lhXk536sgG7j7dyKPwxbwTwAfIJg++DvxtoslnHZ4PvDwAcsvAi4CmDlzZlLx1V7y16TWF5GRK5Ej/uHu/i3CFiF3byc4yZRhdmALeEVJCfe8731Un346fPSjcOSR8MADcNttMZPz6P2o5Vskfdw9P2qe3THuXhD1eMQm5xC0/OUZbM+iBH3jzjZWb2rOWHG4A51+RDWrXty9b3xxRENTB2YwJYUt6BDMid7W1csdT25L6X6jNTR3UJhv+wqcDWT2xHK2NHfQ3tW7b9m2PZ20d/fmdAV3gKtuX0d3b//x/u3dvVx1+7qk9lM1pmhfFXd1b08dM5tgZl8HniBopDrB3T+fyBj0qH1UAH8EPunue6Jfc/dr3X2Buy/QOZaIDFUiCXqXmZUSTkFhZocD2Vf9Z5RYsmQJGzds4NGlS9k9Ywan/OxnMHEi3H473HsvnHZapkMUkVGsID+PqorirGpB/8vqBgDOOTY7EvRXzamiu9d5eEP/VuSGpnYmjSmmMMWtpSfVTmD6+FL+mMZu7lua2plcWUJe3uDX7yNJeKRqO0RNsTaxLD0BDpPIMIVEl8dTVVHM7r3dNO/tVoG4FDGzq4AVBFXb57n7Ze6+e5DNDtxHIUFyfr2735yGMEVEEkrQLwX+Acwws+uBu4HPpTUqic0d/v53qs8+myOWLqWgsBD++EdYsQLe8AbIYOEjEZGIKWNL2LYne67j3rq6gZNqx6e0+NqhOHHWeIoL8g6aD31Lc0fKKrhHy8szzptfw7/X72Brmuaob2juYFqCsceq5B5J1mtzfAx6vOEJyQ5biPREeHHXXsrTNCxhFPoMMA34X6DBzPaEt5aoIUlxWVBdchnwtLt/N82xisgoNmiC7u53AucB7weWAwvc/d70hiUHuf9+uk89Fc4+m94dO+A3v4EnnoDzzhvxiXnS876LSEZNGlOSNUXi1m1t4dltrVnTvR2CabROnj3hoEJxDU3tabuIcN4J0+lzuOXx9MyJvqW5nakJTo8WaUGPruRev6ONovy8lI6/z4SLF82ltLB/QbfSwnwuXjQ3qf1UVQQFYet3tlFRogQ9Fdw9z91LI0ONDhiGlMjQo9OA9wJnmNnj4e3sNIctIqNQIlXcXw0cTdAlaA9wVLhMhsOjj8JZZ8GrX82Ohx/m0yUlTNi+neUFBZA/8qu6JjXvu4hkhSlji7MmQb919Wby84yz5mW2evuBXjWniue2t+5r0XZ3GprbU1rBPVptVTknzhrPHx/dhHtq50Tv63O2JtH6X1FcQPWY4n4t6Bt2tDFzYhn5CXSRz2aL59dw+XnzqBlXihFMmXf5efOSKhAH+1vQWzo0Bj1buPsD7m7ufqy7Hx/e/pbpuERk5EnkqH9x1OMS4GSCqu5npCUi2e/CC+E3v6Fv/Hi+WFDA1T09tHcEJ3Mx5z8fYRKe911EssrkMSXs3ttNR3cvJYWZu5Do7vxl9RZeefjEhIqXDafTj6gGnuGB9Tt4+4nTw++rj6lpbEE+rLqcm1Zu4rClf2PauFIuXjQ36cQxlh1tnXT3+qBzoEebXVW+b9w5BC3Fud69PWLx/JpD/l6jf181Bl1EZHRJpIv7m6NuZwLHAOkrBSv7HXkkXHopq/74R35SXk50iZl+85+PUIPO+y4iWSky1VpjS2bHoa/e1MyLu/by5izq3h7xsiljqKoo4oHngqE7kSJiNUkkucm4ZdXmfcXynP3Tfw1lju4DbWkKLhwnM35+9sTyfePO+/qCKdZmV+V2gbhUqhqzP0GvKB75veVERGS/oVyW3USQpEu6ffGLAMxsbIw9/3ltbQaCGj6x5n3v6uoa8Z9bJNdNDrtpb9vTwYwJmUu6bn28gaL8PBYdPSVjMcSTl2e88vAqHli/M+jeHibo6SgSB8H0Xx3dff2WRab/OtTW3sj87cl0z59dXc6OlV3s6eimpaOHzp6+nJ9iLZXKi/IpLsijs6dPReJEREaZRMag/9DMfhDergHuB1anPzSJOHD+89LSUr7whS9kOqy0i3zuwsLCfcv6+vq46667MhiViAxmcmXQ+pfJqdZ6+5zbnmjgtXOrGVtaOPgGGXD6nCp2tHbyzNYWtoRj0dNVJC1V03/FEok9mQQ90p29fkfbvrHos0dIF/dUMNs/p7y6uIuIjC6JTLO2kmDM+aPAf4DPu/sFaY1KDrJkyRI2btzIxRdfjLvz7W9/O6eKpg21EvvChQspKNh/ctLV1UVdXZ0quotksSmVkRb0zHVxf2TDLra3dGZl9/aIV82pAuCB53bQ0NROUX4eE8uLBtlqaOIl/kUFeexq64r5WqK2NHdQXJDHhCRij0y1tmFH2/450NWC3k+km7uKxImIjC6JjEG/Lup2vbv/ezgCk9i++c1v0tHRQXNzM+3t7TmRrB5KJXaNQxfJPWNLCykqyEuqkvstqzZz2hX3MPuSv3LaFfcc8tjoW1c3UFaUz8KXTz6k/aTT1LGlHF5dzv3rd9DQ3MHUcSXkpamKeazpvwrzjZ7ePs6++n4efmHnkPfd0BRUn7ckpvycNbEMM6jfsZf6HW0UF+Ttu7AjgepwqjVNsyYiMrrEPeqb2RqCWjIHvQS4ux+btqgkpkiyGqloDvuT1Wytan6oldhjjUMfDePvRXKZmTGlMvG50G9ZtZmlN6+hvbsX2F/ADBjS+Oju3j7+vnYLZx41mdKi7C6w9ao51dyw4kXmTBqTtinWYP/3eNXt62hoat9Xxf2ISRV89HePseTnD/GphUdSM66U79z5bL91BvsZbEliirWIksJ8po0tZcOOVlo7e6idWJ62ixO5amK5uriLiIxGAx3135SONzSzGcBvgClAH3Ctu19tZhOA3wO1QD3wTnffHW6zFKgDeoGPu/vt6Ygt2+VisnqoFxUi49Dr6uooLCyku7ubZcuWZe0FCREJTK4s3jfH92Cuun3dvuQ8YqgFzG5ZtZmv3fYUTXu7uf+5HdyyanNKphJLl9OPqOLXD9azZnMz552Q3jjjTf9128dfxRf/tIbv3PkseQZ94aX5RC+UbGlq55TDJyYdT21VGRt27qW1o5sjJlUkvf1Idsuqzfx1zRYAvnLrk/T1eVb/HouISOrE7eLu7hsPvAFtwIvh46HqAT7j7i8HTgE+YmZHAZcAd7v7HODu8Dnha+cDRwNvBH5sZtndJJImsYrFZXuymoqLCpHx93fddRcbN25kyZIlKY5SRFJtcmUJ2xOcZi1VBcwiLfE7wzHVu9q6UjaVWLrsaN3/Hd355LaMxFpRXMD333U840oL9yXnEZELJfH09jnbWjqZNoTq87OrynmhsZWXdrVr/HmUyO9xa2cPADtz4PdYRERSJ26CbmanmNm9Znazmc03s7XAWmCbmb1xqG/o7lvc/bHwcQvwNFADnAtcF652HbA4fHwucIO7d7r7BmA9cPJQ3z/X5VqymqqLCtXV1Zx00klZfTFCRPabXFnC1uYO3GONlOovXgGzZCuaD9QSn41uWbWZr/zlqX3PWzp7MpaImRnN7d0xXxvoQsn2lg56+5ypQ5i/vXZiOS0dPXT19qmCe5Rc+z0WEZHUGqhI3DXAN4HlwD3Af7n7FODVwOWpeHMzqwXmAw8Dk919CwRJPDApXK0GeClqs03hslEr15LVbLuoMNSK8iKSuCmVJbR399IStgIO5OJFcynM7z/+OM/gM2cemdR7pnMqsXTItkRsKBdKGprC6eGG0IJ+WPX+pFwt6Pvl2u+xiIik1kAJeoG73+HuNwFb3f0hAHd/JhVvbGYVwB+BT7r7noFWjbEsZpOMmV1kZivNbKWSr+ySLRcVDqWivIgkblI4F/q2BMahL55fw8umjCE/zzBgfFnQ1XpTEglJR3cvBfmxi4yla27xQ5VtiVisSu+lhXlcvGhu3G22NAexDqUFff321n2PP3nD4+rCHUpVjxIREclNAyXofVGPDzxbGLzP4gDMrJAgOb/e3W8OF28zs6nh61OB7eHyTcCMqM2nAw2x9uvu17r7AndfkOlEULJPdEX5XJqmTiQXTU5iLnR3p6Gpg3OPn8aGK85h1ZffwFvn1/D9u55lZf2uhLb/7E2r6e51ig5I0ksL8wdMMDMp2xKxxfNruPy8edREvf//vPbwAYuTRQoBTq1MLuZbVm3mu3c+u38/ezo0zjoU+0JJ9v4ei4hIag2UoB9nZnvMrAU4NnwceT5vqG9owUSpy4Cn3f27US/dClwYPr4Q+HPU8vPNrNjMZgNzgEeG+v4yemlOdZHhM2Vfgj54C3r9zr3sbOtiwawJ+5Z99dyjmTGhjE/c8HjcsdERV9/9HLc9sYVLznoZ33r7cdSMK8WAmnGlXH7evKytfp2Nidji+TX8+5IzeOKyN1BamM+WQXpANDR1UFaUT2VpclOBXXX7Ojq6+/ot0zjrQPSFklz4PRYRkdSK+x/V3dNVKf004L3AGjN7PFz2BeAK4EYzqwNeBN4RxvGkmd0IPEVQAf4j7t570F5FBpGL09SJ5KpIC/rWBBL0FWEr+Um14/ctG1NSyA/On8/bfvIgX7h5Dde8ez7B9d3+/rK6ge/f9RxvO2E6H371YZhZziQy8eYmz4b4K0sKedOxU7n18Qa+eM5RVMSZi3tLcztTx5bE/NkMJNu692ebeFPiiYjIyJfcJe8UcPcHiD2uHOD1cbb5BvCNtAUlo4LmVBcZPqVF+VSWFLA9gQT90frdjC0t5PDq/nNhHzdjHJ95w1yu/MczvHplFe86aWa/1x9/qYnP3rSak2rH883zjkk6ScwG2ZyInX/yTG56dBN/Wd3AkpNnxlynobljSF3yp40rZXOMZFzjrEVEZLQbqIu7yIiTbRXlRUayyZUlibWgb9zFglnjycs7OMH+8KsP4/Qjqrjs1qf6FRVraGrnQ79ZyaTKYn56wYkUF6Sr09fodcLMccydPIYbHnkx7jpbmoIW9GRlY/d+ERGRbDDsLegimVZdXa1Wc5FhMGVsyaBF4na2dvJCYxtvP3F6zNfz8ozvvvM43nj1/bxv2cMAbGnuID/PyDe4/uOvYmJFccpjl2Bu9PNPnsFX/vIUTzY0c/S0sf1e7+rpo7G1k6lDmGItm7v3i4iIZJJa0EVEcoiZXWVmz5jZE2b2JzMbF/XaUjNbb2brzGxR1PITzWxN+NoPbJj6gk8aUzJokbhHN+4G4KTaCXHXmVRZwnkn1NDQ3EFDcwcO9PQ5jvFUw0CzdMqheuv8GooL8rjhkZcOem3bng7cYdoQpliD/QXpNlxxDv++5Awl5yIiIihBlzRobGxkxYoVmr5MJD3uBI5x92OBZ4GlAGZ2FHA+cDTwRuDHZhbpQ/wT4CKCWTDmhK+n3ZSxxWxv6aSvL/7MnI9u3E1Rfh7zasbGXQfg72u2HrSsq7dPVb/TbFxZEWfPm8otqzazt6un32uRCu9DaUEXERGR2JSgS0otX76cWbNmceaZZzJr1iyWL1+e6ZBERhR3v8PdI5nSQ0Ckb/i5wA3u3unuG4D1wMlmNhWodPf/uLsDvwEWD0eskytL6O1zdrTF7+a+on4X86aPpaRw4DHkqvqdOUtOnklLZw+3PbGl3/ItzcF3P9QWdBERETmYEnRJmcbGRurq6mhvb6e5uZn29nbq6urUki6SPh8E/h4+rgGi+yFvCpfVhI8PXH4QM7vIzFaa2cpU/N1GplrbHmccekd3L2s2N7Ng1viYr0eLV91bVb/T76Ta8RxeXX5QsbhIC/oUtaCLiIikjBJ0SZn6+nqKior6LSssLKS+vj4zAYnkKDO7y8zWxridG7XOF4Ee4PrIohi78gGWH7zQ/Vp3X+DuC1JRSHHfXOjNscehP7Gpme5eZ8EA488jVPU7c8yMJSfP5LEXm1i3tWXf8i1N7YwpKYg7R7rISGJmvzSz7Wa2NtOxiMjIpgRdUqa2tpaurq5+y7q7u6mtrc1MQCI5yt0XuvsxMW5/BjCzC4E3Ae8Ju61D0DI+I2o304GGcPn0GMvTbkqYoG9riZ2gr9y4C4ATE2hBXzy/hsvPm0fNuFIMqBlXyuXnzVNhsWFy3gnTKcrPY3lUK3pDcwfT1Houo8evGab6HSIyuumyt6RMdXU1y5Yto66ujsLCQrq7u1m2bJmmNBNJITN7I/B54DXuvjfqpVuB35nZd4FpBMXgHnH3XjNrMbNTgIeB9wE/HI5YqyqKMCPuVGsr63dzeHU5E8qLYr5+oMXza5SQZ8iE8iIWHTOFmx/bxCVnvYySwny2NLczVePPZZRw9/vMrDbTcYjIyKcEXVJqyZIlLFy4kPr6empra5Wci6TeNUAxcGc4W9pD7v7f7v6kmd0IPEXQ9f0j7t4bbvM/BK0/pQRj1v9+0F7ToCA/j6qKYrbF6OLe1+esrN/F2fOmDkcokgJLTp7BX1Y38Pe1W3jr/OlsaepgXs24TIclkjXM7CKCGTOYOXNmhqOR0ab2kr+mdf/1V5yT1v2nW7q/H0jdd6QEXVKuurpaiblImrj7EQO89g3gGzGWrwSOSWdc8UypLInZxX19Yyt7OnoS6t4u2eHUwyZSO7GM5Q+/xFnHTGVnWxfTxqoFXSTC3a8FrgVYsGBB/PklRUQGoDHoIiKSNpMri2MWiVtRH4w/PymBAnGSHcyM80+eySP1u/j3+h0ATFUVfRERkZRSgi4jVmNjIytWrNA0byIZNLmyhO0tB49Bf7R+N1UVRcyaWJaBqGSo3nbCdAryjO/d9SyAWtBFRERSTAm6jEjLly9n1qxZnHnmmcyaNYvly5dnOiSRUWlyZQm72rro7Ontt3zFxl0smDWBcBy95IjqMcUcPW0MazfvAeBTNz7OLas2ZzgqkfQzs+XAf4C5ZrbJzOoyHZOIjExK0GXEaWxspK6ujvb2dpqbm2lvb6eurk4t6SIZEJlqbXtUJfdtezp4aVc7C2o1/jzX3LJqM09vad33fNueTpbevEZJuox47r7E3ae6e6G7T3f3ZZmOSURGJiXoOUBdtZNTX19PUVH/aZsKCwupr6/PTEAio9ikymIgSMojVtbvBmCBxp/nnKtuX0dXb1+/Ze3dvVx1+7oMRSQiIjKyKEHPYo2NjXz9619XV+0k1dbW0tXV1W9Zd3c3tbW1mQlIZBSbEo5Rjp4LfeXGXZQU5nH0tMpMhSVD1NDUntRyERERSY4S9Cy1fPlyZs6cyZe+9CV11U5SdXU1y5Yto7S0lMrKSkpLS1m2bJmmfhPJgMljggR96wEt6MfPGEdhvv4F5Zppcaq2x1suIiIiydHZURaKjKHu6Dh4aiJ11U7MkiVL2LhxI3fddRcbN25kyZIlmQ5JZFQaV1ZIUUEe28MEva2zh6e27NH0ajnq4kVzKS3M77estDCfixfNzVBEIiIiI0tBpgOQg9XX11NQEPtHo67aiauurlaruUiGmRmTK4v3jUF//KUmevucE2epQFwuWjy/BgjGojc0tTNtXCkXL5q7b7mIiIgcGiXoWSjWGOqIL3zhC0o6RSSnTB5Tsq+L+8r63ZjBCUrQc9bi+TVKyEVERNJEXdyzUHV1NVdfffVBy0tKSvjwhz+cgYhERIZu8tiSfdOsrdy4i7mTx1BZUpjhqERERESyjxL0LPXhD3+Yn/70pxQXFzNmzBhKS0v55S9/qdZzEck5kRb0nt4+Htu4W+PPRUREROJQF/cs9uEPf5jzzjuP+vp6amtrlZyLSE6aMraYvV29rNy4m7auXhbUqnu7iIiISCxK0LOcCp2JSK6bXBlMtfbXJ7YAsEAt6CIiIiIxqYu7iIikVSRB//vaLUwdW0KN5swWERERiUkJuoiIpFUkQd/R2qXWcxEREZEBKEEXEZG0WrFh577H9z3byC2rNmcwGhEREZHspQRdRETS5pZVm7n01qf2PW9u72bpzWuUpIuIiIjEoARdRETS5qrb19He3dtvWXt3L1fdvi5DEYmIiIhkLyXoIiKSNg1N7UktFxERERnNlKCLiEjaTItTsT3echEREZHRTAm6iIikzcWL5lJamN9vWWlhPhcvmpuhiERERESyV0GmAxARkZFr8fwaIBiL3tDUzrRxpVy8aO6+5SIiIiKynxJ0ERFJq8Xza5SQi4iIiCRAXdxFREREREREsoASdBEREREREZEsoARdREREREREJAvkTIJuZm80s3Vmtt7MLsl0PCIiIiIyeuhcVESGQ04k6GaWD/wIOAs4ClhiZkdlNioRERERGQ10LioiwyUnEnTgZGC9u7/g7l3ADcC5GY5JREREREYHnYuKyLDIlWnWaoCXop5vAl5x4EpmdhFwUfi008zWDkNsqVIF7Mh0EElQvOmXazHnWrxzMx1ANnv00Ud3mNnGDLx1rv0eQe7FnGvxgmIeDsnGOytdgWSpQc9FDzgPbTWzdcMU21Dl2u/oYPR50siuPORdZNXnSYGDPs8QvqOYx9FcSdAtxjI/aIH7tcC1AGa20t0XpDuwVFG86ZVr8ULuxZyL8WY6hmzm7tWZeN9c+z2C3Is51+IFxTwcci3eDBj0XDT6PDQXjLSfuT5PdtPnSVyudHHfBMyIej4daMhQLCIiIiIyuuhcVESGRa4k6CuAOWY228yKgPOBWzMck4iIiIiMDjoXFZFhkRNd3N29x8w+CtwO5AO/dPcnB9ksZ7oYhRRveuVavJB7MSteSYVc/LnkWsy5Fi8o5uGQa/EOqyGei2a7kfYz1+fJbvo8CTL3g4Zyi4iIiIiIiMgwy5Uu7iIiIiIiIiIjmhJ0ERERERERkSww4hJ0M3ujma0zs/Vmdkmm44kws3ozW2Nmj0emdzKzCWZ2p5k9F96Pj1p/afgZ1pnZomGI75dmtj167vihxGdmJ4afc72Z/cDMYk1Lks6YLzOzzeH3/LiZnZ0tMZvZDDP7p5k9bWZPmtknwuVZ+T0PEG9WfsdmVmJmj5jZ6jDer4TLs/L7HS3i/J0eZ2b/Cb/jv5hZZbi80MyuC5c/bWZLo7a5N/w5RX7vJmVJzEVm9qtw+Woze23UNpk+Hg4l5mH5nnPteJiGmNP+PScbr5lNDNdvNbNrDtiXjok5zMyuMrNnzOwJM/uTmY2Lei3n/w9alp77DySVx5NsYmb5ZrbKzG4Ln+fs5zGzcWb2h/Bv52kzO3XYPo+7j5gbQdGO54HDgCJgNXBUpuMKY6sHqg5Y9i3gkvDxJcCV4eOjwtiLgdnhZ8pPc3yvBk4A1h5KfMAjwKkE84X+HThrmGO+DPhsjHUzHjMwFTghfDwGeDaMKyu/5wHizcrvONx3Rfi4EHgYOCVbv9/Rcovzd7oCeE34+IPA18LH7wZuCB+XERw3a8Pn9wILsjDmjwC/Ch9PAh4F8ob79yiFMQ/L9zzA8SVr/15THHPav+chxFsOnA78N3DNAfvSMTGHb8AbgILw8ZXZ9HeVgs+Wtef+g8SdsuNJNt2ATwO/A24Ln+fs5wGuA/4rfFwEjBuuzzPSWtBPBta7+wvu3gXcAJyb4ZgGci7BD5/wfnHU8hvcvdPdNwDrCT5b2rj7fcCuQ4nPzKYCle7+Hw9+W38Ttc1wxRxPxmN29y3u/lj4uAV4GqghS7/nAeKNJ9Pxuru3hk8Lw5uTpd/vaBHn73QucF/4+E7gbZHVgXIzKwBKgS5gz3DEGS3JmI8C7g632w40AQuy5HiYVMzpii2WXDsepjLmdMSWinjdvc3dHwA6ovejY2Luc/c73L0nfPoQwRzukAV/VymQa+f+QO4dTxJhZtOBc4BfRC3Oyc9jQY+zVwPLANy9y92bGKbPM9IS9Brgpajnmxg4oRhODtxhZo+a2UXhssnuvgWCP1SClgzIns+RbHw14eMDlw+3j4bduH4Z1fUkq2I2s1pgPkErb9Z/zwfEC1n6HYddqx4HtgN3untOfL+j0FrgLeHjdwAzwsd/ANqALcCLwLfdPTrp/FXYHfhLGehuGS/m1cC5ZlZgZrOBE8PXsuH3KNmYI4b1e8614yEccswRw/Y9JxhvPNnwuyyp80GCFnHIsr+rIcqWc+YhS9HxJBt8H/gc0Be1LFc/z2FAI8FxepWZ/cLMyhmmzzPSEvRY/+B82KOI7TR3PwE4C/iImb16gHWz+XNA/PiyIe6fAIcDxxOc5H8nXJ41MZtZBfBH4JPuPlDrYFbEHCPerP2O3b3X3Y8naB042cyOGWD1jMc7in2Q4Dj4KEHXvq5w+clALzCNoIvYZ8zssPC197j7POBV4e29wxty3Jh/SfCPeCXBycmDQA/Z8XuUbMwwzN9zrh0PISUxwzB+z0nEG3cXMZbpmJhlzOwuM1sb43Zu1DpfJPhbvz6yKMaucu3/YC7FepAUHU8yzszeBGx390cT3STGsqz5PEABwbCxn7j7fILGg4HqG6T08xQMdcMstYn+rQDTgYYMxdKPuzeE99vN7E8EJ6LbzGyqu28JuxNtD1fPls+RbHyb2N9tKnr5sHH3bZHHZvZz4LbwaVbEbGaFBAfi69395nBx1n7PseLN9u84jLHJzO4F3kgWf7+jlbs/QzAmEjM7kqBLHARj0P/h7t3AdjP7N0HX6xfcfXO4bYuZ/Y7gGPqbTMccdhv9VGQ9M3sQeA7YTeaPh8nGzHB+z7l2PExhzMP2PScZbzw6JuYAd1840OtmdiHwJuD1Ybd1yJK/q0OULefMSUvV8SRLnAa8xYLCwSVApZn9ltz9PJuATWFPTAh6+F3CMH2ekdaCvgKYY2azzawIOB+4NcMxYWblZjYm8pjghGktQWwXhqtdCPw5fHwrcL6ZFYfdD+cQFOsYbknFF3b1aDGzU8Lueu+L2mZYhH8sEW8l+J6zIuZw/8uAp939u1EvZeX3HC/ebP2Ozazawsq0ZlYKLASeIUu/39HMworVZpYH/C/w0/ClF4EzLFBOUOTvGQu6YleF2xQSnGSuPXjPwx+zmZWFsWJmZwI97v5UNvweJRvzcH7PuXY8TGXMw/U9DyHemLLhd1kOjZm9Efg88BZ33xv1Usb/rlIgK8/9B5Oq48lwxTsYd1/q7tPdvZbgZ3CPu19A7n6ercBLZjY3XPR64CmG6/N4FlTJS+UNOJugEuLzwBczHU8Y02EEY/5WA09G4gImEhTqeS68nxC1zRfDz7COYaicCSwn6K7cTXAVqG4o8RG0dq0NX7sGsGGO+f+ANcAT4R/L1GyJmaA6roexPR7ezs7W73mAeLPyOwaOBVaFca0FvjzUv7Ph/D0e6bc4f6efIDhOPwtcEfl+gQrgJoLj5FPAxeHycoJK40+Er11NGqu9Jhlzbfj78zRwFzArE79HqYh5OL/nAY4vWfv3mqqYh+t7HmK89QTFBlvD36Ojhvt3Wbe0/O6uJxgfG/k9+GnUazn/f5AsPPdPIOaUHQOz7Qa8lv1V3HP28xAM5VwZ/oxuAcYP1+eJ/LMWERERERERkQwaaV3cRURERERERHKSEnQRERERERGRLKAEXURERERERCQLKEEXERERERERyQJK0EVERERERESygBL0EcTMvmhmT5rZE2b2uJm9Ilx+r5ktCB//LTJX9AHbXmZmn03y/c4ys5Vm9rSZPWNm307JB8khZvZBM1sTfudrzexcMzvOzB6PWmeJme0N57vFzOaZ2RPh43vNbF3483rczP4Q530Wm9mX0xD/PDP7dar3KyLZK5xn/gEzOytq2TvN7B+ZjEtEJJqZ9YbnRmvN7C+xzl8T2McCM/tBnNfqzaxqiLElfd4cbpdnZj8IP9MaM1sRzuH+CTP7ftR6PzOzu6KefyzyOaK+l8jtkjjv9X0ze/UQPt5gn+GjZvaBVO9X9ivIdACSGmZ2KvAm4AR37wwPOEUHrufuZ6fo/Y4hmBfzHHd/xswKgItSse9cYWbTCeY8PMHdm82sAqgGNgKzzGyMu7cArwSeAeYDj4TP/x21q/e4+8pB3u5zwFtS/RncfY2ZTTezme7+Yqr3LyLZx93dzP4buMnM/gnkA98A3jiU/ZlZvrv3pjJGERGg3d2PBzCz64CPEByrEhaeXw12jjWc3gVMA451977wXLINeBB4T9R6xwN5UcfXVxLMxQ1R30s8ZjYBOMXdP5nS6AO/JDiP/VUa9i2oBX0kmQrscPdOAHff4e4NB64UfbUwbHFfF16hmxu1zuFm9g8ze9TM7jezl8V4v88B33D3Z8L363H3H4fbv9nMHjazVWZ2l5lNDpdfZma/DFuNXzCzj4fLv2Zmn4h6/2+Y2cfDVp6roq4yvit8/bXhPv4Qttxfb2YW47Pea2bfM7P7wlb+k8zsZjN7zsy+HrXeLeFnfdLMLgqXzQrXqwqvdt5vZm844C0mAS1Aa/gdtLr7BnfvA1YArwjXOxH4EcHBlfD+wRjfaUxmdiTQ6e47wue/NrOfmNk/w+/xNeH3+nR0a7iZtZrZleFnu8vMTo767qOT/b8A5ycaj4jkPndfS/C3/3ngUuC3wBfD1pxVZnYugJnVhse/x8LbK8Plrw2PQb8D1phZuZn91cxWh8fsd2Xqs4nIiPQfoAbin6ea2TvC489qM7svXPZaM7stfDzRzO4Ij3E/AyxcXmtmayNvZGafNbPLwscfCo+Lq83sj2ZWdmBg4TnrUxb0prxhkM8xFdgSnivi7pvcfTewCjjSzErNbCywF3gcmBdul9S5I/B2YF+vqPD8/5tm9h8Ler+eYGa3m9nzFlywjXxX/zKzG83sWTO7wszeY2aPhOfhh4cx7wXqzezkJOKRJChBHznuAGaEf1A/NrPXDLSymZ1IkJTNB84DTop6+VrgY+5+IvBZ4McxdnEM8Gic3T9AcNVuPnADQTIf8TJgEXAycKkF3b6XAReGceWFcV0fxnU8cBywELjKzKaG+5kPfBI4CjgMOC1OLF3u/mrgp8CfCa6+HgO838wmhut8MPysC4CPm9lEd98IXBlu9xngKXe/44B9rwa2ARvM7Fdm9uao1x4EXmlm5UAfcC/9E/ToFvTrbX83patifIbTgMcOWDYeOAP4FMFJ9veAo4F5ZnZ8uE45cG/42VqArwNnAm8Fvhq1r5XAq2K8r4iMbF8B3g2cBZQA97j7ScDrCI635cB24Ex3P4Gg5Se6q+jJwBfd/SiC1vcGdz/O3Y8h6sRQRORQmFk+8Hrg1nBRvPPULwOL3P04Yvc6vBR4IDw/vRWYmcDb3+zuJ4X7fBqoi7HOJcB8dz8W+O9B9ncj8ObwnO87ZjYfgoYugoT8JOAU4GHgIYJzyWmAuftL4T5KrX8X91gXRE/j4PP0l9z9VOB+4NcESfwp9D8nPA74BMGFgfcCR7r7ycAvgI9FradzxzRSF/cRwt1bw6T7VQQnV783s0vc/ddxNnkV8KfwKhhmdmt4X0GQQN5k+xuli5MMZ3r4/lMJutlviHrtr2Erf6eZbQcmu3u9me0MD1KTgVXuvtPMTgeWh117tpnZvwgOXHuAR9x9Uxjz40AtwYWBA0UO5muAJ919S7jNC8AMYCdBUv7WcL0ZwBxgp7v/wszeQXCwPf7AHbt7r5m9MYzp9cD3zOxEd7+MIAH/DMFBcIW7P29mR5hZNVDh7i9E7WqwLu5TgcYDlv0l7Ka6Btjm7mvCz/Vk+F08DnSx/yR5DUErfHe4TW3UvrYTdLcSkVHE3dvM7PcEvYDeSXDSGBlTWUJw8toAXBNe+OsFjozaxSPuHjm+rwG+bWZXAre5+/3D8RlEZEQrjTrHexS4c5Dz1H8DvzazG4GbY+zv1QSNP7j7X81sdwIxHGNBr8txQAVwe4x1niBobLmF/d3QY3L3TWY2l6CR5QzgbjN7h7vfHcb/SqCUoMfAc8AXCM4Bo1vPB+3iTuxzx+hz4opwGGaLmXXY/vH9K6LOlZ8naACMbPO6qH1tJ2h0kzRQgj6ChInsvcC9YRJ2IcEVsribxFiWBzQl8If/JEHX7dUxXvsh8F13v9XMXgtcFvVaZ9TjXvb/Dv4CeD8whWBsC4Rdj+KIt5946/UdsE0fUBDGtxA41d33mtm9BCemhN2YpofrVxC0Qvfj7k4wrvwRM7uTYDzOZQRXPU8CTic4yAJsIugdkEwXJYB2YGwynyt83B3G12+9cMxT9PdVEr6HiIw+feHNgLe5+7roF8NuntsIWlXygI6ol9siD9z92fAi8dnA5WZ2h7tHt8qIiCSr3d2PD7t830bQC/LXxDlPdff/tqBA8jnA41E9CvutFmNZD/17FZdEPf41sNjdV5vZ+4HXxtj+HILk/y3Al8zs6LBFPKawoervwN/NbBuwGLib4Pzww+H7/4ggwT4qvP93zJ3F137A54DEzh0PXN4ZYx3QuWNaqYv7CGFmc81sTtSi4wmKlcVzH/DWcKzLGODNAO6+h6DL9jvC/ZqZHRdj+6uAL1gwPjpSlfLT4Wtjgc3h4wsT/Ah/IugieRL7r07eB7zLzPLDludXEyTDqTQW2B0m5y8j6OoTcSVBV/svAz8/cEMzm2ZmJ0QtOp7wOw+vSr5EcNEhkqD/h6BbfrIJ+tPAEUluk4wjgbWDriUiI9ntwMcsbJKKdLskOEZGxku+l6Cg3EHCLph73f23wLeBE2KtJyKSLHdvBj5O0J29nTjnqWZ2uLs/7O5fBnYQ9IqMdh9hITYLZrEYHy7fBkyyYIx6MUHR5YgxwJZwSGZ0ETfC/eQBM9z9nwRDOscBFRbU/flNjPVPCI+XkW2PZf/5+oME56HV7r49bGRpBM5F546jihL0kaMCuM7CIhUEV9wui7eyuz8G/J6gK/QfCbpiR7wHqDOz1QQt5efG2P4JgmRzuZk9TfBHGhkffhlB16P7CQ6Qg3L3LuCfwI1R1YD/RNBtaDVwD/A5d9+ayP6S8A+ClvQngK8RtHwTjuE/CbjS3a8HuuzgKSUKCbp0PhN2wXoXwbidiH8DxVFjhv5DMF7+wINs9Bj0uzjYfcD8yIlzGrwO+Gua9i0iueFrBMe0JywolvS1cPmPgQvN7CGCE7K2ONvPI+hJ9DjB7BZfj7OeiEjS3H0Vwfng+cQ/T73KgmJmawnOnQ7s5fkV4NVm9hjwBuDFcN/dBOOwHyZoqX8mapsvhcvvPGB5RD7w27Dn6irge+7eRDBEKFYL8yTgL2GMTxC03l8TxrGbICF/Mmr9/4TbRH+WA8egXxHjff5K7Nb+VDkNiHXOKilg+3vAimROeBXxMeAd7v5cpuPJNmZ2NcG485QeDMMrxf8CTh+oO5aIiIiIJMaCor//FzZoZSqGB4A3hRcMUrnf+cCn3f29qdyv7KcEXTLOzI4iuGL5J3f/TKbjyUYWTFX3Cne/ddCVk9vvHKDG3e9N5X5FREREJHPC8fjtqb5IYGZnAs+5e30q9yv7KUEXERERERERyQIagy4iIiIiIiKSBZSgi4iIiIiIiGQBJegiIiIiIiIiWUAJuoiIiIiIiEgWUIIuIiIiIiIikgX+P8nEuMzOCr/iAAAAAElFTkSuQmCC\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "fig, [ax1, ax2, ax3] = plt.subplots(nrows=1, ncols=3, figsize=(14,4), tight_layout=True)\n", "\n", "# Scatterplot\n", "data.plot.scatter(x='SLI_max', y='BLC_max', c='k', ax=ax1);\n", "\n", "# Plot the regression line, we only need two points to define a line, use xmin and xmax\n", "ax1.plot([x.min(), x.max()], [B0 + B1*x.min(), B0 + B1*x.max()] , '-r')\n", "\n", "ax1.set_xlabel('Slide Canyon max SWE (mm)')\n", "ax1.set_ylabel('Blue Canyon max SWE (mm)');\n", "\n", "ax1.set_xlim((0,3000))\n", "ax1.set_ylim((0,1000));\n", "\n", "# Plot the residuals\n", "ax2.plot(data.years,residuals,'-o')\n", "\n", "ax2.set_xlabel('Years')\n", "ax2.set_ylabel('Residuals, SWE (mm)');\n", "\n", "# Plot a histogram of the residuals\n", "ax3.hist(residuals, bins=10)\n", "\n", "ax3.set_xlabel('Residuals, SWE (mm)')\n", "ax3.set_ylabel('Number of Data Points');" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "---\n", "\n", "### Linear regression using the scipy library\n", "\n", "Now we'll use the `scipy.stats.linregress()` function to do the same thing. Review the documentation or help text for this function before proceeding. " ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "\u001b[0;31mSignature:\u001b[0m \u001b[0mstats\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlinregress\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;31mDocstring:\u001b[0m\n", "Calculate a linear least-squares regression for two sets of measurements.\n", "\n", "Parameters\n", "----------\n", "x, y : array_like\n", " Two sets of measurements. Both arrays should have the same length. If\n", " only `x` is given (and ``y=None``), then it must be a two-dimensional\n", " array where one dimension has length 2. The two sets of measurements\n", " are then found by splitting the array along the length-2 dimension. In\n", " the case where ``y=None`` and `x` is a 2x2 array, ``linregress(x)`` is\n", " equivalent to ``linregress(x[0], x[1])``.\n", "\n", "Returns\n", "-------\n", "result : ``LinregressResult`` instance\n", " The return value is an object with the following attributes:\n", "\n", " slope : float\n", " Slope of the regression line.\n", " intercept : float\n", " Intercept of the regression line.\n", " rvalue : float\n", " Correlation coefficient.\n", " pvalue : float\n", " Two-sided p-value for a hypothesis test whose null hypothesis is\n", " that the slope is zero, using Wald Test with t-distribution of\n", " the test statistic.\n", " stderr : float\n", " Standard error of the estimated slope (gradient), under the\n", " assumption of residual normality.\n", " intercept_stderr : float\n", " Standard error of the estimated intercept, under the assumption\n", " of residual normality.\n", "\n", "See Also\n", "--------\n", "scipy.optimize.curve_fit :\n", " Use non-linear least squares to fit a function to data.\n", "scipy.optimize.leastsq :\n", " Minimize the sum of squares of a set of equations.\n", "\n", "Notes\n", "-----\n", "Missing values are considered pair-wise: if a value is missing in `x`,\n", "the corresponding value in `y` is masked.\n", "\n", "For compatibility with older versions of SciPy, the return value acts\n", "like a ``namedtuple`` of length 5, with fields ``slope``, ``intercept``,\n", "``rvalue``, ``pvalue`` and ``stderr``, so one can continue to write::\n", "\n", " slope, intercept, r, p, se = linregress(x, y)\n", "\n", "With that style, however, the standard error of the intercept is not\n", "available. To have access to all the computed values, including the\n", "standard error of the intercept, use the return value as an object\n", "with attributes, e.g.::\n", "\n", " result = linregress(x, y)\n", " print(result.intercept, result.intercept_stderr)\n", "\n", "Examples\n", "--------\n", ">>> import matplotlib.pyplot as plt\n", ">>> from scipy import stats\n", "\n", "Generate some data:\n", "\n", ">>> np.random.seed(12345678)\n", ">>> x = np.random.random(10)\n", ">>> y = 1.6*x + np.random.random(10)\n", "\n", "Perform the linear regression:\n", "\n", ">>> res = stats.linregress(x, y)\n", "\n", "Coefficient of determination (R-squared):\n", "\n", ">>> print(f\"R-squared: {res.rvalue**2:.6f}\")\n", "R-squared: 0.735498\n", "\n", "Plot the data along with the fitted line:\n", "\n", ">>> plt.plot(x, y, 'o', label='original data')\n", ">>> plt.plot(x, res.intercept + res.slope*x, 'r', label='fitted line')\n", ">>> plt.legend()\n", ">>> plt.show()\n", "\n", "Calculate 95% confidence interval on slope and intercept:\n", "\n", ">>> # Two-sided inverse Students t-distribution\n", ">>> # p - probability, df - degrees of freedom\n", ">>> from scipy.stats import t\n", ">>> tinv = lambda p, df: abs(t.ppf(p/2, df))\n", "\n", ">>> ts = tinv(0.05, len(x)-2)\n", ">>> print(f\"slope (95%): {res.slope:.6f} +/- {ts*res.stderr:.6f}\")\n", "slope (95%): 1.944864 +/- 0.950885\n", ">>> print(f\"intercept (95%): {res.intercept:.6f}\"\n", "... f\" +/- {ts*res.intercept_stderr:.6f}\")\n", "intercept (95%): 0.268578 +/- 0.488822\n", "\u001b[0;31mFile:\u001b[0m /opt/conda/lib/python3.8/site-packages/scipy/stats/_stats_mstats_common.py\n", "\u001b[0;31mType:\u001b[0m function\n" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "stats.linregress?" ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "B0 : 127.9143\n", "B1 : 0.1997\n", "R^2 : 0.181\n", "R : 0.425\n", "stderr : 0.087\n" ] } ], "source": [ "# use the linear regression function\n", "slope, intercept, rvalue, pvalue, stderr = stats.linregress(data.SLI_max, data.BLC_max)\n", "\n", "print('B0 : {}'.format(np.round(intercept,4)))\n", "print('B1 : {}'.format(np.round(slope,4)))\n", "\n", "print('R^2 : {}'.format(np.round(rvalue**2,3)))\n", "print('R : {}'.format(np.round(rvalue,3)))\n", "print('stderr : {}'.format(np.round(stderr,3)))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Do we get the same results as above?\n", "\n", "No, our \"standard error\" is different. Why is that? If you look into the documentation for the lingregress function, you'll see that it calls this output the \"standard error of the **gradient**\" meaning the standard error of the slope, $B1$.\n", "\n", "This is related to the \"standard error\", $\\sigma$ like:\n", "\n", "$SE_{B_1} = \\displaystyle \\frac{\\sigma}{\\sqrt{SST_x}} $ where $SST_x = \\displaystyle\\sum_{i=1}^{n} (x_i - \\bar{x}_i)^2$\n", "\n", "Compute the standard error from the standard error of the gradient:" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "sigma : 204.089\n" ] } ], "source": [ "# Compute the SST for x\n", "sst_x = np.sum( (x - np.mean(x))**2 )\n", "\n", "# Compute the standard error\n", "sigma = stderr * np.sqrt(sst_x)\n", "print('sigma : {}'.format(np.round(sigma,3)))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This should now match what we solved for manually above." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Finally, plot the result" ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "fig, ax = plt.subplots(figsize=(4,4))\n", "\n", "# Scatterplot\n", "data.plot.scatter(x='SLI_max', y='BLC_max', c='k', ax=ax);\n", "\n", "# Create points for the regression line\n", "x = np.linspace(data.SLI_max.min(), data.SLI_max.max(), 2) # make two x coordinates from min and max values of SLI_max\n", "y = slope * x + intercept # y coordinates using the slope and intercept from our linear regression to draw a regression line\n", "\n", "# Plot the regression line\n", "ax.plot(x, y, '-r')\n", "\n", "ax.set_xlabel('Slide Canyon max SWE (mm)')\n", "ax.set_ylabel('Blue Canyon max SWE (mm)');\n", "\n", "ax.set_xlim((0,3000))\n", "ax.set_ylim((0,1000));" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We've used the slope and intercept from the linear regression, what were the other values the function returned to us?\n", "\n", "This function gives us our R value, we can report how well our linear regression fits our data with this or R-squared (you can see in this case linear regression did a poor job)" ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "r-value = 0.4249234045616491\n", "r-squared = 0.18055989974426292\n" ] } ], "source": [ "print('r-value = {}'.format(rvalue))\n", "\n", "print('r-squared = {}'.format(rvalue**2))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This function also performed a two-sided \"Wald Test\" (t-distribution) to test if the slope of the linear regression is different from zero (null hypothesis is that the slope is not different from a slope of zero). Be careful using this default statistical test though, is this the test that you really need to use on your data set?" ] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "p-value = 0.03047392304371895\n" ] } ], "source": [ "print('p-value = {}'.format(pvalue))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "And finally it gave us the standard error of the gradient" ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "standard error = 0.0868316797949923\n" ] } ], "source": [ "print('standard error = {}'.format(stderr))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now use this linear model to predict a $y$ (Blue Canyon) value for each $x$ (Slide Canyon) value:" ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [], "source": [ "y_predicted = slope * data.SLI_max + intercept" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Plot residuals**\n", "\n", "We should make a plot of the residuals (actual - predicted values)" ] }, { "cell_type": "code", "execution_count": 19, "metadata": {}, "outputs": [], "source": [ "residuals = data.BLC_max - y_predicted" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "For a good linear fit, we hope that our residuals are small, don't have any trends or patterns themselves, want them to be normally distributed:" ] }, { "cell_type": "code", "execution_count": 20, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "f, (ax1, ax2) = plt.subplots(1,2,figsize=(9,4))\n", "\n", "ax1.plot(data.years,residuals)\n", "ax1.set_xlabel('years')\n", "ax1.set_ylabel('residuals (mm SWE)')\n", "\n", "ax2.hist(residuals)\n", "ax2.set_xlabel('residuals (mm SWE)')\n", "ax2.set_ylabel('count')\n", "\n", "f.tight_layout()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "That distribution doesn't look quite normal, and there seems to maybe be a negative bias (our predictions here might be biased higher then the observations).\n", "\n", "There does seem to be a trend in the residuals over time, so this model may not be the best." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "---\n", "\n", "### Making predictions with our linear model\n", "\n", "Let's plot what the predictions of Blue Canyon SWE would look like if we were to use this linear model:" ] }, { "cell_type": "code", "execution_count": 21, "metadata": {}, "outputs": [], "source": [ "# Use our linear model to make predictions:\n", "BLC_pred = slope * data.SLI_max + intercept" ] }, { "cell_type": "code", "execution_count": 22, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 22, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmoAAAFNCAYAAACwk0NsAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Z1A+gAAAACXBIWXMAAAsTAAALEwEAmpwYAACcdElEQVR4nOydd1hTdxfHvz9AEQfuVRduAQUF98Tdum21aq1Wrdo6X6u11tq66l61dqhtHbVaReseddS6NwrurbiruFBUZOS8f5wkBggQQhZwPs+TJ8kdv3vuzc29556piAiCIAiCIAiC4+FkbwEEQRAEQRAE44iiJgiCIAiC4KCIoiYIgiAIguCgiKImCIIgCILgoIiiJgiCIAiC4KCIoiYIgiAIguCgiKImCImglDqrlAqwwrg9lFL7Db5HKKVKWXo7yciglFKLlFJPlFJHbbjd3Uqp3rbanpB6lFJdlVLbLTBOnPPe2iil8iulLiqlsthqmwbbPqqU8rb1doX0iShqQoZFqyDpXhql1CuD712JyJuIdltbDiLKTkTXrL2deNQF0BRAUSKqHn+m9qYaqz0Wz5RSIUqpVrYUUClVVyl1UCkVrpR6rJQ6oJSqZksZDGTpopQ6F2/ajkSmfZnMWB5KKVJKuVhQPt2YEfFenVI7NhEtI6JmlpDTUmj3tUwyi30JYBERRWrX8VZKbdc+nDxVSh1XSrXQzruolHrfYPw62m3EnxahlHKJ9/8wfL2lXXwGgPGW3WshoyKKmpBh0SpI2YkoO4CbAFobTFtmb/msTAkAoUT0IollDmmPTS4ACwCsVErlsYVwSil3AJsA/AAgD4AiAMYBeG2L7RthDwBPpVR+rXwuAHwBZI03rRaAvdYUJBkFL5fheU1EgdaUxVFRSrkC+AjAUoPJGwHsAFAQQAEAgwE8087bC6CBwbL1AVwwMu0gEcVovx+Kd6yzE9Fd7bwNABoqpQpbcr+EjIkoaoKQCEqpUKVUE+3nsUqpVUqppUqp50qp00qpckqpkUqpB0qpW0qpZgbr5lRKLVBK3VNK3VFKTVBKOSeyHb11QCm1WCn1k1Jqs3Y7R5RSpQ2WraC12jyObwUwMu5bSqkN2mWvKKX6aKd/DOA3ALW0VoBxSR0HItIAWAjADUAppZSrUmqGUuqmUuq+UmqeUspNO3ZupdQmpVSY1nKxSSlVNBH5CiulTimlPjcyu5x228uJKJaIXhHRdiI6pV23h1Jqv1aOJ0qp60qpd0zY9yyKLaf5tN+/VkrFaBVDaH+n2UaOwV0A18A3awDwA3AWrMAZTnMCEKSUaqmUClZsjbyllBprMJxOkXuqPf61tNvupZQ6r92fbUqpEgb7Q0qpAUqpywAuGzueSaGUyqs9Hs8Uu+W+VVo3pDJi4VMGLmpl4LLU/tYz4o29Xik1VPv5S6XUVe25e04p1T4JmRI9l5P6HyildMfvpErcalgDwFMiuq1dJx+AkgB+JaIo7esAEelcsXvx5ncEgHoAphqZZpISrrXiHQfgUJZIIW0iipogmE5rAH8AyA0gGMA28H+oCNjNMd9g2d8BxAAoA6AK+IJtamxWF7D1KDeAKwAmAoBSKhvYIvAn2CLQBcDPKvFYmOUAbgN4C0AHAJOUUo2JaAGAT/HGIjAmKWG0N/DeACLASsJUsCJVWbt/RQCM1i7uBGAR2GJXHMArAD8aGdMDrOT8SEQz4s8HcAlArFLqd6XUO0qp3EaWqQHgIoB8AKYBWKCUUsnseySAY3hjKakP4AaAOgbf9yRyKAxv5vUB7AOwP960w0QUBeAFgO5ga2RLAP2UUu0MlgPeWL8Oaed9BeBdAPm1Yy+Pt/122n32SkS+pPgJQCSAwgB6aV/m8CeATrrjrP1dmgFYoZ1/FazQ5ASfw0uVEauSieey0f8BEemOn28SVsNK4HNDxyPtGEuVUu2UUgXjLb8HgLdSKo9SyglAVQCBAHIZTKuNlFlLz4OtroKQKkRREwTT2UdE27Suj1XgG+oUIooG36g8lFK5tDeBdwAMIaIXRPQAwHcAOpu4nTVEdFS7nWVghQgAWoHdlYuIKIaITgBYDVZE4qCUKgaOQxtBRJFEFAK2onVLwf7WVEo9BfAf+KbZHuwq6gPgMyJ6TETPAUzS7RsRPSKi1UT0UjtvIuK6jwBWNHYDGENEvxjbMBE908pPAH4FEKa1CBneYG8Q0a9EFAtWjAsDKGjCvu8B0ECrgPoAmKP9ngVANbCSZAxD61k97XL74k3bo5V/NxGdJiKN1gq43MhxMOQTAJOJ6Lz2d58EoLKhVU07/zERvUpinIeK4690L0/Fltz3AIzWno9ntMfLHPaBf5N62u8dwAr/XQAgolVEdFe734FgxT5BDCRMO5cT+x+YQi4Az3VfiJtaNwQQCmAmgHtKqb1KqbLa+TfB4Q/1wMrVZe1xPmAwLQuAIwbbqBnvWF+NJ8NzrRyCkCpEURME07lv8PkVgIdaJUH3HQCyg61JmcA3g6daZWc+2HJgCv8ZfH6pHRPacWsY3hwAdAVQyMgYbwHQKVI6boCtX6ZymIhyEVE+IqpJRP+AldOsAI4byLBVOx1KqaxKqflKqRtKqWdgC0QuFdft2xXAHQB/JbVxrdLSg4iKAqio3afZBov8Z7DsS+3H7Cbs+x4AAWBX5WmwZacBgJoArhDRw0RE2gvAR2tFqglWUC4AKKydVle7DJRSNZRSuxS7gMPBFsx8SexuCQDfGxzTxwAU4v5et5JYX0c+7W+me50H/zYu8da/YcJYCdAqPCvAijsAfABWogAASqnuihNPdPtREcb325RzObH/gSk8AZAjnuy3iWggEZXWbv8FgCUGi+gspjprKfDGYlofwBEiMoyRPBzvWJdGXHIAeJoCmQXBKKKoCYLluQUOeje8aboTUWrT9W8B2BPv5pCdiPoZWfYugDxKKcObVXGwgpQaHoKVUm8DGXJqkw4AYBiA8gBqEJE73liblMEYY7Xj/KkSiduLj1YhWgy+8SdHcvt+UCtje/DxPKed3xKJuz1BnJl7F0BfADeJKEI765B2WnYAh7XT/gQHlBcjopwA5uHNMSAjw98C8Em839aNiA4aipD8rhslDOyGL2YwrbjBZ11CSVaDacaUfx3LAXTQWvtqgC1h0H7/FcBAAHmJKBeAM4j72+tIyblsDqegjXM0BhHdAruDDc8nnaKms5YCbyymJsenGeAJ4GQK1xGEBIiiJggWhojuAdgOYKZSyl0p5aSUKq2USsr1ZQqbAJRTSnVTSmXSvqoppTyNyHALrJBMVhxA7wPgYxhYP8yBOLHgVwDfKaUKAIBSqohSqrl2kRxgRe6p4gxRY/Fv0QA6AsgG4A9t/E8ctIHmw5Q2EUHrzuyCN4pQUjImue9a69txAAPwRjE7CHY/JqqoadkHYCjiukf3a6cFGbglc4CtepFKqepgy5OOMAAaAIa18+YBGKmL0VKcjNIxuX01Ba3Vdw2AsVqLpxc4I1I3PwysxH6olHJWSvUCEN86ZDhesHYffgOwjYieamdlAyuTYdp96InEFWuTz+VEuI+4xy8+R8GW3CJaWXIrpcYppcpo/4/5wHF6hufTXnA8aQOwyxNgi2tJsNvUZEVNcdapP9haKwipQhQ1QbAO3QFkBnAO7Ib5CxxDZTZaV14zcDzYXbBraCoA10RW6QLAQ7vsWnBMmCVuHCPAgdmHte7Nf8AWKoBdk25gi9lhsFs0AcQB9++C3cELjShrz8HWmiNKqRfasc6ALXamkNy+7wG7p48afM+B5G/Ge7QyGxZu3aedZrhufwDjlVLPwYkWK3UztIriRAAHtG6/mkS0FvxbrtAe0zPgOMeUossk1b2GaqcPBFv8/gNbJhfFW68PgOHgoHtvsOKaFMsBNAFbDnX7dQ4c/3UIrEhVwhuFJw5mnMvxGQvgd+3xS5D5rD2/FgP4UDspCnw+/AOOszwDtnr3MFjnEoAHAO7plE/tg8lRAO5IeExqqYR11HR1/toA2E1vynUIgtkoDjkQBEEQMgpKqR4AehNRXXvLYi0U17fbB6BKMgkY1tj2EQAfaxM3BCFVWKwytiAIgiA4ClqXbgU7bbuGPbYrpE/E9SkIgiAIguCgiOtTEARBEATBQRGLmiAIgiAIgoMiipogCIIgCIKDkm6TCfLly0ceHh72FkMQBEEQBCFZjh8//pCI8sefnm4VNQ8PDwQFBdlbDEEQBEEQhGRRShlt7SauT0EQBEEQBAfFaoqaUqqYtinxeaXUWaXU/7TTxyql7mgb94YopVoYrDNSKXVFKXXRoCUNlFL+SqnT2nlzlFLGescJgiAIgiCkK6zp+owBMIyITmibIx9XSulauHxHRDMMF9b2n+sMbl/yFoB/lFLltH3q5oKbHh8GsAXA2wD+tqLsgiAIgiAIdsdqipq2MfU97efnSqnzAIoksUpbACuI6DWA60qpKwCqK6VCAbgT0SEAUEotAdAOoqgJguCAREdH4/bt24iMjLS3KIIgOCBZsmRB0aJFkSlTJpOWt0kygVLKA0AVAEcA1AEwUCnVHUAQ2Or2BKzEHTZY7bZ2WrT2c/zpgiAIDsft27eRI0cOeHh4QKI0BEEwhIjw6NEj3L59GyVLljRpHasnEyilsgNYDWAIET0DuzFLA6gMtrjN1C1qZHVKYrqxbfVVSgUppYLCwsJSK7ogCEKKiYyMRN68eUVJEwQhAUop5M2bN0UWd6sqakqpTGAlbRkRrQEAIrpPRLFEpAHwK4Dq2sVvAyhmsHpRAHe104samZ4AIvqFiKoSUdX8+ROUIhEEQbAJoqQJgpAYKb0+WDPrUwFYAOA8Ec0ymF7YYLH2AM5oP28A0Fkp5aqUKgmgLICj2li350qpmtoxuwNYby25BUEQ0joTJ06Et7c3fHx8ULlyZRw5cgQAEBAQoK8v2aJFCzx9+jTBumPHjsWMGTMSTE+Kv//+G1WrVoWnpycqVKiAzz//PNX7YEmICBMmTEDZsmVRrlw5NGzYEGfPntXPz549ux2lM06PHj3w119/2VsMwQGwZoxaHQDdAJxWSoVop30FoItSqjLYfRkK4BMAIKKzSqmVAM6BM0YHaDM+AaAfgMUA3MBJBJJIIAiCYIRDhw5h06ZNOHHiBFxdXfHw4UNERUUlWG7Lli0W2d6ZM2cwcOBAbN68GRUqVEBMTAx++eUXi4xtKX766SccPHgQJ0+eRNasWbF9+3a0adMGZ8+eRZYsWSy+vZiYGLi4pNt68oKNsZpFjYj2E5EiIh8iqqx9bSGibkRUSTu9jdZipltnIhGVJqLyRPS3wfQgIqqonTeQiIzGqNmUf/8FVq2ytxSCIAhxuHfvHvLlywdXV1cAQL58+fDWW28lWM7DwwMPHz4EwBa48uXLo0mTJrh48aJ+matXr+Ltt9+Gv78/6tWrhwsXLiQYZ9q0aRg1ahQqVKgAAHBxcUH//v0BABs3bkSNGjVQpUoVNGnSBPfv3wfAVrtevXohICAApUqVwpw5cwAA33zzDb7//nv92KNGjcKcOXNARBg+fDgqVqyISpUqITAwEACwe/duBAQEoEOHDqhQoQK6du0KY7eHqVOn4ocffkDWrFkBAM2aNUPt2rWxbNky/TLDhg2Dn58fGjduDF2M85w5c+Dl5QUfHx907twZAPDixQv06tUL1apVQ5UqVbB+PTt4Fi9ejI4dO6J169Zo1qwZOnXqFEcZ7tGjB1avXo3Y2FgMHz4c1apVg4+PD+bPnw+ArX4DBw6El5cXWrZsiQcPHhj9fYUMCBGly5e/vz9ZlXffJSpb1rrbEAQhzXHu3Dm7bv/58+fk6+tLZcuWpX79+tHu3bv18xo0aEDHjh0jIqISJUpQWFgYBQUFUcWKFenFixcUHh5OpUuXpunTpxMRUaNGjejSpUtERHT48GFq2LBhgu1VqVKFQkJCjMry+PFj0mg0RET066+/0tChQ4mIaMyYMVSrVi2KjIyksLAwypMnD0VFRdH169epSpUqREQUGxtLpUqVoocPH9Jff/1FTZo0oZiYGPrvv/+oWLFidPfuXdq1axe5u7vTrVu3KDY2lmrWrEn79u2LI0N4eDjlzp07gWyzZ8+mzz77jIiIANDSpUuJiGjcuHE0YMAAIiIqXLgwRUZGEhHRkydPiIho5MiR9Mcff+inlS1bliIiImjRokVUpEgRevToERERrVmzhrp3705ERK9fv6aiRYvSy5cvaf78+fTtt98SEVFkZCT5+/vTtWvXaPXq1fp9vHPnDuXMmZNWrVpl9LgKaR9j1wkAQWREnxHbrLn4+gJr1wIvXgDZstlbGkEQHJEhQ4CQEMuOWbkyMHt2orOzZ8+O48ePY9++fdi1axc6deqEKVOmoEePHkaX37dvH9q3b6+3NrVp0wYAEBERgYMHD6Jjx476ZV+/fp0iUW/fvo1OnTrh3r17iIqKilOOoGXLlnB1dYWrqysKFCiA+/fvw8PDA3nz5kVwcDDu37+PKlWqIG/evNi/fz+6dOkCZ2dnFCxYEA0aNMCxY8fg7u6O6tWro2jRotpDUxmhoaGoW7dusrIRkT6o28nJCZ06dQIAfPjhh3j33XcBAD4+PujatSvatWuHdu3aAQC2b9+ODRs26OP4IiMjcfPmTQBA06ZNkSdPHgDAO++8g8GDB+P169fYunUr6tevDzc3N2zfvh2nTp3Sx5+Fh4fj8uXL2Lt3r34f33rrLTRq1ChFx1pIv0ivT3Px9QWIgNOn7S2JIAhCHJydnREQEIBx48bhxx9/xOrVq5Nc3lgWmkajQa5cuRASEqJ/nT9/PsFy3t7eOH78uNFxBw0ahIEDB+L06dOYP39+nJIEOtesTt6YmBgAQO/evbF48WIsWrQIvXr1AgCj7szkxtHh7u6ObNmy4dq1a3GmnzhxAl5eXkbH1B2PzZs3Y8CAATh+/Dj8/f0RExMDIsLq1av1x+TmzZvw9PQEAGQzeGjPkiULAgICsG3bNgQGBupdp0SEH374Qb/+9evX0axZszjbFQRDRFEzF19ffj950r5yCILguMyeDezebdlXEtY0ALh48SIuX76s/x4SEoISJUokunz9+vWxdu1avHr1Cs+fP8fGjRsBsIJTsmRJrNLG4hIRThq53g0fPhyTJk3CpUuXALCCN2sWJ/qHh4ejSBGuT/77778nKbeO9u3bY+vWrTh27BiaN2+ulzEwMBCxsbEICwvD3r17Ub169WRGiivj4MGD8erVKwDAP//8g/379+ODDz7Qy6yzcP3555+oW7cuNBoNbt26hYYNG2LatGl4+vQpIiIi0Lx5c/zwww965TE4ODjR7Xbu3BmLFi3Cvn379PvSvHlzzJ07F9HR0QCAS5cu4cWLF6hfvz5WrFiB2NhY3Lt3D7t27TJ5/4T0jbg+zaVECSBnTlHUBEFwKCIiIjBo0CA8ffoULi4uKFOmTJJZmH5+fujUqRMqV66MEiVKoF69evp5y5YtQ79+/TBhwgRER0ejc+fO8NU9pGrx8fHB7Nmz0aVLF7x8+RJKKbRs2RIAJw107NgRRYoUQc2aNXH9+vVk5c+cOTMaNmyIXLlywdnZGQArb4cOHYKvry+UUpg2bRoKFSpkNLnBGIMGDcKTJ09QqVIlODs7o1ChQli/fj3c3NwAsCXs7Nmz8Pf3R86cOfVK4Ycffojw8HAQET777DPkypUL33zzDYYMGQIfHx8QETw8PLBp0yaj223WrBm6d++ONm3aIHPmzADYYhgaGgo/Pz8QEfLnz49169ahffv2+Pfff1GpUiWUK1cODRo0MGnfhPSPSsqknJapWrUq6eoFWY369YHYWODAAetuRxCENMP58+f1rjAh5Wg0Gvj5+WHVqlUoW7asvcURBKtg7DqhlDpORFXjLyuuz9Tg6wucOgVoNPaWRBAEIc1z7tw5lClTBo0bNxYlTRC0iOszNfj6AhERwPXrQOnS9pZGEAQhTePl5ZUg6F8QMjpiUUsNklAgCIIgCIIVEUUtNVSsCDg5iaImCIIgCIJVEEUtNbi5AeXKiaImCIIgCIJVEEUttfj6iqImCIIgCIJVEEUttfj6AqGhQHi4vSURBEEAwBX6K1euDF9fX/j5+eHgwYMAgNDQUFSsWNFi25kxYwYqVKiAihUrwtfXF0uWLLHY2JYgPDwc3bt3R+nSpVG6dGl0794d4dpr9e7du9GqVSs7S5gQDw8PPHz40N5iCA6EKGqppXJlfj91yq5iCIIg6HBzc0NISAhOnjyJyZMnY+TIkRbfxrx587Bjxw4cPXoUZ86cwd69e5Ns9WQPPv74Y5QqVQpXr17F1atXUbJkSfTu3dtq24vfvkoQLIEoaqlFMj8FQXBgnj17hty5cyeYvnjxYgwcOFD/vVWrVti9ezcAbjxeq1Yt+Pn5oWPHjoiIiEiw/qRJk/Dzzz/D3d0dAJAzZ0589NFHAIDx48ejWrVqqFixIvr27atX4AICAjBixAhUr14d5cqVw759+wAA9erVQ4hB8/o6derg1KlTePz4Mdq1awcfHx/UrFkTp7QPxGPHjkWvXr0QEBCAUqVKYc6cOQnku3LlCo4fP45vvvlGP2306NEICgrC1atX9cemffv28PLywqeffgqNRoPY2Fj06NEDFStWRKVKlfDdd98BAK5evYq3334b/v7+qFevnr4rQo8ePTB06FA0bNgQw4cPh4eHB54+farfZpkyZXD//n2EhYXhvffeQ7Vq1VCtWjUc0BZKf/ToEZo1a4YqVargk08+cThlV7A/oqillsKFgXz5AIOLjCAIgj159eoVKleujAoVKqB3795xlJXkePjwISZMmIB//vkHJ06cQNWqVfW9O3U8f/4cz58/R+lE6kcOHDgQx44dw5kzZ/Dq1as4LZZiYmJw9OhRzJ49G+PGjQPwphE7wL0vX79+DR8fH4wZMwZVqlTBqVOnMGnSJHTv3l0/zoULF7Bt2zYcPXoU48aN0/fO1HHu3DlUrlxZ34YKeOMSPnv2LADg6NGjmDlzJk6fPo2rV69izZo1CAkJwZ07d3DmzBmcPn0aPXv2BAD07dsXP/zwA44fP44ZM2agf//++nEvXbqEf/75B9999x3atm2LtWvXAgCOHDkCDw8PFCxYEP/73//w2Wef4dixY1i9erXesjdu3DjUrVsXwcHBaNOmDW7evGnybyVkDKTgbWpRShIKBEEwypAhln+Gq1w52b7setcnABw6dAjdu3fHmTNnTBr/8OHDOHfuHOrUqQMAiIqKQq1ateIsQ0RQSiU6xq5duzBt2jS8fPkSjx8/hre3N1q3bg0AePfddwEA/v7+CA0NBQB07NgR3377LaZPn46FCxeiR48eAID9+/dj9erVAIBGjRrh0aNH+hizli1bwtXVFa6urihQoADu37+PokWLJiuj4fTq1aujVKlSAIAuXbpg//79aNy4Ma5du4ZBgwahZcuWaNasGSIiInDw4EF07NhRP87r16/1nzt27KhXCDt16oTx48ejZ8+eWLFiBTp16gSAG8GfO3dOv86zZ8/w/Plz7N27F2vWrNHvkzHrp5CxEUXNEvj6Aj//DMTEAC5ySAVBcBxq1aqFhw8fIiwsLM50FxcXaAza30VGRgJgRaZp06ZYvnx5omO6u7sjW7ZsuHbtml7RMRynf//+CAoKQrFixTB27Fj92ADg6uoKgK1bupiurFmzomnTpli/fj1WrlwJXZ9mY25AnZKlGyf+WDq8vb0RHBwMjUYDJyd2Hmk0Gpw8eRKenp64fft2AkVOKYXcuXPj5MmT2LZtG3766SesXLkSs2fPRq5cueK4Zw3Jli2b/nOtWrVw5coVhIWFYd26dfj666/12z506JC+EbyxfRIEY4jr0xL4+gKRkcDly/aWRBAEB2L2bGD3bsu+krOmxefChQuIjY1F3rx540z38PBASEgINBoNbt26haNHjwIAatasiQMHDuDKlSsAgJcvX+LSpUsJxh05ciQGDBiAZ8+eAWAL0S+//KJXyvLly4eIiAj89ddfJsnZu3dvDB48GNWqVUOePHkAAPXr18eyZcsAcJZmvnz59DFxyVGmTBlUqVIFEyZM0E+bMGEC/Pz8UKZMGQDs+rx+/To0Gg0CAwNRt25dPHz4EBqNBu+99x6+/fZbnDhxAu7u7ihZsiRWrVoFgBXIk4l4UZRSaN++PYYOHQpPT0/9cW/WrBl+/PFH/XI6pc9wH//++288efLEpP0TMg5i/rEEhgkFnp72lUUQhAyPLkYNYKXi999/jxOrBXDAfsmSJVGpUiVUrFgRfn5+AID8+fNj8eLF6NKli969N2HCBJQrVy7O+v369UNERASqVauGTJkyIVOmTBg2bBhy5cqFPn36oFKlSvDw8EC1atVMktnf3x/u7u76mDCAkwZ69uwJHx8fZM2aFb///nuKjsOCBQswaNAglClTBkSEWrVqYcGCBfr5tWrVwpdffonTp0+jfv36aN++vT4uTWdtnDx5MgBg2bJl6NevHyZMmIDo6Gh07twZvrprfzw6deqEatWq6ePuAGDOnDkYMGAAfHx8EBMTg/r162PevHkYM2YMunTpAj8/PzRo0ADFixdP0T4K6R+VXjNMqlatSjrzudWJigKyZweGDQO0f2pBEDIm58+fh6c8sKWYu3fvIiAgABcuXNC7KgUhvWLsOqGUOk5EVeMvK/8GS5A5M1vSJKFAEAQhxSxZsgQ1atTAxIkTRUkThHiI69NS+PoCO3faWwpBEIQ0R/fu3eOU3hAE4Q3y6GIpfH2Bu3cBaf0hCIIgCIKFEEXNUkiHAkEQBEEQLIwoapZCFDVBEARBECyMKGqWIn9+biclipogCIIgCBZCFDVLIq2kBEFwAHQ9LX19feHn54eDBw8CAEJDQ1GxYkWLbWfGjBmoUKECKlasCF9fXyxZssRiY1sCDw8PVKpUCb6+vmjWrBn+++8/s8caO3YsZsyYAYCbu//zzz+JLhsSEoItW7akeBsBAQEwp6yUoTyzZ8/Gy5cv9fOyZ8+e7PqLFy/GwIEDE0xv0aJFnAbz1iY0NBRubm6oXLkyvLy80L179wQ9XO1J796947QBsxWiqFkSX1/g3DmuqyYIgmAndL0+T548icmTJ2PkyJEW38a8efOwY8cOHD16FGfOnMHevXuNtnyyN7t27cLJkydRtWpVTJo0Kc48IorTRstUxo8fjyZNmiQ631xFzVwM5YmvqKWGLVu2IFeuXBYZyxjx234BQOnSpRESEoLTp0/j9u3bWLlypVW2Yw6//fYbvLy8LDJWShBFzZL4+gLR0cCFC/aWRBAEAQC3djLW6Du+FaVVq1bYvXs3AGD79u2oVasW/Pz80LFjR0RERCRYf9KkSfj555/1LZ1y5syJjz76CAArDtWqVUPFihXRt29fvQIXEBCAESNGoHr16ihXrhz27dsHAKhXr16cPpp16tTBqVOn8PjxY7Rr1w4+Pj6oWbMmTp06BYCtW7169UJAQABKlSqFOXPmJHsc6tevjytXriA0NBSenp7o378//Pz8cOvWLUyfPh3VqlWDj48PxowZo19n4sSJKF++PJo0aYKLFy/qp/fo0UPfGuvYsWOoXbs2fH19Ub16dYSHh2P06NEIDAxE5cqVERgYiBcvXqBXr16oVq0aqlSpgvXr1wPgDhKdO3eGj48POnXqhFevXiWQ++jRo/pG9uvXr4ebmxuioqIQGRmp77Oqk2fOnDm4e/cuGjZsiIYNG+rHGDVqFHx9fVGzZk3cv38/2WOlw8PDAw8fPtQfsz59+sDb2xvNmjXTy3r16lW8/fbb8Pf3R7169XBBe//buHEjatSogSpVqqBJkyb67Y4dOxZ9+/ZFs2bNkizJ4uzsjOrVq+POnTsAgOPHj6NBgwbw9/dH8+bNce/ePf3x9/HxQa1atTB8+HC9xXjx4sXo2LEjWrdujWbNmiX6G5w9exbVq1dH5cqV4ePjg8uXL+PFixdo2bIlfH19UbFiRQQGBgKIa/Fcvny5vqvHiBEj9HJnz57d7OOdKESULl/+/v5kc86eJQKIliyx/bYFQXAIzp07Z28RyMnJiXx9fal8+fLk7u5OQUFBRER0/fp18vb2JiKiRYsW0YABA/TrtGzZknbt2kVhYWFUr149ioiIICKiKVOm0Lhx4+KM/+zZM8qVK1ei23/06JH+84cffkgbNmwgIqIGDRrQ0KFDiYho8+bN1LhxYyIiWrx4Mf3vf/8jIqKLFy+S7vo9cOBAGjt2LBER7dy5k3x9fYmIaMyYMVSrVi2KjIyksLAwypMnD0VFRSWQo0SJEhQWFkZERAMGDKAvvviCrl+/TkopOnToEBERbdu2jfr06UMajYZiY2OpZcuWtGfPHgoKCqKKFSvSixcvKDw8nEqXLk3Tp08nIqKPPvqIVq1aRa9fv6aSJUvS0aNHiYgoPDycoqOjExzbkSNH0h9//EFERE+ePKGyZctSREQEzZw5k3r27ElERCdPniRnZ2c6duxYnH2Ijo4mDw8PIiIaNmwYVa1alfbv30+7d++mzp07x5En/j4TEQHQH//hw4fTt99+m+A4xZc3/vG7fv06OTs7U3BwMBERdezYUb8/jRo1okuXLhER0eHDh6lhw4ZERPT48WPSaDRERPTrr7/qf/cxY8aQn58fvXz5MsH2DM/PV69eUUBAAJ08eZKioqKoVq1a9ODBAyIiWrFihf64eXt704EDB4iIaMSIEXHO7yJFiujPxcR+g4EDB9LSpUuJiOj169f08uVL+uuvv6h37956uZ4+fUpEfP4eO3aM7ty5Q8WKFaMHDx5QdHQ0NWzYkNauXWvy8SYyfp0AEERG9BkpeGtJypUDXF05Tq1bN3tLIwiCAxAQEJBg2vvvv4/+/fvj5cuXaNGiRYL5PXr0QI8ePfDw4UN06NAhzjyd1SspdK5PADh06BC6d++OM2fOmCTv4cOHce7cOdSpUwcAEBUVhVq1asVZhoiglEp0jF27dmHatGl4+fIlHj9+DG9vb7Ru3RoA9NYhf39/hIaGAgA6duyIb7/9FtOnT8fChQvRo0cPAMD+/fuxevVqAECjRo3w6NEjhIeHAwBatmwJV1dXuLq6okCBArh//z6KFi2aQJaGDRvC2dkZPj4+mDBhAp4+fYoSJUqgZs2aANh6uH37dlSpUgUAEBERgcuXL+P58+do3749smbNCgBo06ZNgrEvXryIwoUL6/uZJtYwfvv27diwYYM+xi0yMhI3b97E3r17MXjwYACAj48PfHx8Eqzr4uKCMmXK4Pz58zh69CiGDh2KvXv3IjY2FvXq1Uv0N9CROXNmtGrVCgAf8x07diS7jjFKliyp7x+r++0iIiJw8OBBdOzYUb+crj/s7du30alTJ9y7dw9RUVEoWbKkfpk2bdrAzc3N6HauXr2KypUr4/Lly+jQoQN8fHxw5swZnDlzBk2bNgUAxMbGonDhwnj69CmeP3+O2rVrAwA++OADbNq0ST9W06ZNkSdPHgCJ/wa1atXCxIkTcfv2bbz77rsoW7YsKlWqhM8//xwjRoxAq1atEhznY8eOISAgAPnz5wcAdO3aFXv37kW7du0sdrwNEUXNkri4ABUrSkKBIAgOQ61atfDw4UOEhYXFme7i4hInPisyMhIAK2FNmzbF8uXLEx3T3d0d2bJlw7Vr1/TuN8Nx+vfvj6CgIBQrVgxjx47Vjw0Arq6uANi1pYsdypo1K5o2bYr169dj5cqVevcSGYl50ymIunHijxWfXbt2IV++fPrvT58+RbZs2fTfiQgjR47EJ598Eme92bNnJ6mM6tZNbhndcqtXr0b58uUT3Z+kqFevHv7++29kypQJTZo0QY8ePRAbG6tXOpIiU6ZM+m0kdZySI/7xfvXqFTQaDXLlyhXHba1j0KBBGDp0KNq0aYPdu3dj7Nix+nmGxz8+uhi1e/fuISAgABs2bEDJkiXh7e2NQ4cOxVn2yZMnScoc/3c29ht4enqiRo0a2Lx5M5o3b47ffvsNjRo1wvHjx7FlyxaMHDkSzZo1w+jRo+OMlRiWOt6GSIyapdFlfjpgUK0gCLZn9+7dCV79+/cHwAqKsfk6i1K+fPkSzEspFy5cQGxsLPLmzRtnuoeHB0JCQqDRaHDr1i0cPXoUAFCzZk0cOHAAV65cAQC8fPkSly5dSjDuyJEjMWDAADx79gwAx8L98ssveqUsX758iIiI0MdyJUfv3r0xePBgVKtWTW8FqV+/PpYtWwaAj2O+fPkStVqZS/PmzbFw4UJ9HN6dO3fw4MED1K9fH2vXrsWrV6/w/PlzbNy4McG6FSpUwN27d3Hs2DEAwPPnzxETE4McOXLg+fPncbbxww8/6G/wwcHBCfbvzJkz+hi8+NSvXx+zZ89GrVq1kD9/fjx69AgXLlyAt7d3gmXjb9uauLu7o2TJkli1ahUAVmBOag0V4eHhKFKkCADg999/T/HYhQsXxpQpUzB58mSUL18eYWFhekUtOjoaZ8+eRe7cuZEjRw4cPnwYALBixYpEx0vsN9A9bAwePBht2rTBqVOncPfuXWTNmhUffvghPv/8c5w4cSLOWDVq1MCePXvw8OFDxMbGYvny5WjQoEGK99FUxKJmaXx9gYULgf/+47pqgiAINubVq1d6NxUR4ffff4ezs3OcZerUqYOSJUvqA6L9/PwAAPnz58fixYvRpUsXvRtrwoQJKFeuXJz1+/Xrh4iICFSrVg2ZMmVCpkyZMGzYMOTKlQt9+vRBpUqV4OHhoXcLJoe/vz/c3d3Rs2dP/bSxY8eiZ8+e8PHxQdasWc264SdHs2bNcP78eb17N3v27Fi6dCn8/PzQqVMnVK5cGSVKlDDqZsycOTMCAwMxaNAgvHr1Cm5ubvjnn3/QsGFDTJkyBZUrV8bIkSPxzTffYMiQIfDx8QERwcPDA5s2bUK/fv30+1e5cmVUr17dqIw1atTA/fv3Ub9+fQDsJi1QoIBRa1zfvn3xzjvvoHDhwti1a5fJx2Hx4sVYt26d/rtO+UmOZcuWoV+/fpgwYQKio6PRuXNn+Pr6YuzYsejYsSOKFCmCmjVr4vr16ybLoqNdu3YYO3Ysjhw5gr/++guDBw9GeHg4YmJiMGTIEHh7e2PBggXo06cPsmXLhoCAAOTMmdPoWIn9BoGBgVi6dCkyZcqEQoUKYfTo0Th27BiGDx8OJycnZMqUCXPnzo0zVuHChTF58mQ0bNgQRIQWLVqgbdu2Kd4/U1FJmfDSMlWrViVz6tGkmj17gIAA4O+/gbfftv32BUGwK+fPn4enp6e9xUhz3L17FwEBAbhw4QKcnMTZI5hGRESEvlbclClTcO/ePXz//fd2lip5jF0nlFLHiahq/GXl32BpdMGgEqcmCIJgEkuWLEGNGjUwceJEUdKEFLF582ZUrlwZFStWxL59+/D111/bWySLI65PS5M7N1C8uChqgiAIJtK9e/cka2oJQmJ06tQJnTp1srcYVkUeXayBtJISBOvz6BHQrh1w9669JREEQbAaoqhZA19f4OJFwCAlXRAEC7N/P7B+PbB2rb0lSUB6jf0VBCH1pPT6IIqaNfD1BWJjgbNn7S2JIKRfbtzg9/377StHPLJkyYJHjx6JsiYIQgKICI8ePUKWLFlMXkdi1KyBry+/nzwJ+PvbVxZBSK9oq9pj3z6uW2hC4VBbULRoUdy+fTtBgVlBEASAH+aMddFIDFHUrEHp0kC2bICRas2CIFgInUXtzh3g5k2gRAn7yqMlU6ZMcdrlCIIgpAaruT6VUsWUUruUUueVUmeVUv/TTs+jlNqhlLqsfc9tsM5IpdQVpdRFpVRzg+n+SqnT2nlzlCk9N+yJkxOX6ZCEAkGwHjduAG+9xZ8dzP0pCIJgKawZoxYDYBgReQKoCWCAUsoLwJcAdhJRWQA7td+hndcZgDeAtwH8rJTSldKeC6AvgLLal+NXkpVWUoJgXUJDgRYtgBw5RFETBCHdYjVFjYjuEdEJ7efnAM4DKAKgLQBdH5DfAbTTfm4LYAURvSai6wCuAKiulCoMwJ2IDhFH5y4xWMdx8fUFwsPZJSMIgmV58YLLc5QqBdSuDRw4YG+JBEEQrIJNsj6VUh4AqgA4AqAgEd0DWJkDUEC7WBEAtwxWu62dVkT7Of50x8YwoUAQBMuii0/z8ADq1gXOnAGePLGrSIIgCNbA6oqaUio7gNUAhhDRs6QWNTKNkphubFt9lVJBSqkgu2dcVarEWWiiqAmC5dFlfJYowYoaEXDokF1FEgRBsAZWVdSUUpnAStoyIlqjnXxf686E9v2BdvptAMUMVi8K4K52elEj0xNARL8QUVUiqpo/f37L7Yg5ZM/O2Z+iqAmC5TG0qFWvDri4SJyaIAjpEmtmfSoACwCcJ6JZBrM2APhI+/kjAOsNpndWSrkqpUqCkwaOat2jz5VSNbVjdjdYx7GRVlKCYB1CQ4HMmYFChYCsWbleoShqgiCkQ6xpUasDoBuARkqpEO2rBYApAJoqpS4DaKr9DiI6C2AlgHMAtgIYQESx2rH6AfgNnGBwFcDfVpTbcvj6AlevAhER9pZEENIXN24AxYpxKRwAqFMHOHoUeP3avnIJgiBYGKsVvCWi/TAeXwYAjRNZZyKAiUamBwGoaDnpbISvL8fOnD4N1Kplb2kEIf1w4wa7PXXUrQvMmgUcP85ZoIIgCOkE6fVpTSTzUxCsQ2ho3E4Ederwu7g/BUFIZ4iiZk2KFwdy5RJFTRAsSWQk8N9/cRW1AgWAcuVEURMEId0hipo1UUpaSQmCpbmlLbdo6PoE2P154ACg0dhcJEEQBGshipq18fUFTp2Sm4cgWArDGmqG1K0LPH4MXLhgc5EEQRCshShq1sbXl9vdXLtmb0kEIX1gWEPNkLp1+V3cn4IgpCNEUbM2klAgCJblxg3A2RkoEq+TXJkyHKsmipogCOkIUdSsjbc313oSRU0QLENoKCtpLvGqCynFVjVR1ARBSEeIomZt3NyA8uVFURMESxG/hpohdesC168Dd+7YVCRBEARrIYqaLZBWUoJgOeLXUDNEF6d24IDNxBEEQbAmoqjZAl9ftgI8fWpvSQQhbRMdzdayxBS1ypW596e4PwVBSCeIomYLdAkFp07ZVw5BSOvcucOlbhJzfWbKBNSsKRY1QRDSDaKo2QLJ/BQEy5BYDTVD6tYFQkKA589tIZEgCIJVEUXNFhQuDOTLJ4qaIKQWXQ215BQ1jQY4fNg2MgmCIFgRUdRsgVKSUCAIlkCnqBUvnvgyNWtySRyJUxMEIR0gipqt8PUFzpwBYmLsLYkgpF1CQ9lC7eqa+DI5cnBSgShqgiCkA0RRsxW+vkBkJHD5sr0lEYS0S1I11AypU4ddn9HRVhdJEATBmoiiZiskoUAQUk9SNdQMqVsXePmSkwoEQRDSMKKo2QpPTy4dIDcOQTAPjQa4dcs0Ra1OHX4X96cgCGkcUdRsRebMrKyJRU0QzOPePXZlmuL6LFIEKFlSFDVBENI8oqjZEsn8FATzMaWGmiG6Bu1EVhNJEATB2oiiZkt8fdkqEBZmb0kEIe1hSg01Q+rWBR48AK5csZ5MgiAIVkYUNVsiCQWCYD7mKGqAuD8FQUjTiKJmS0RREwTzCQ3lDh/Zspm2fIUKQJ48oqgJgpCmEUXNluTPD7z1lihqgmAON26Ybk0DuDtBnTqiqAmCkKYRRc3WSEKBIJiHqcVuDalbF7h0iWPVBEEQ0iCiqNkaX1/g/HkgKsrekghC2oEo5RY14E2c2oEDlpdJEATBBoiiZmt8fbkW1Pnz9pZEENIOYWHAq1cpt6j5+3NfUHF/CsIbwsOBc+fsLYVgIqKo2RpJKBCElJPSGmo6XF2B6tXFoiYIOmJjgZYt2dqs0dhbGsEERFGzNWXLAlmyiKImCCkhpaU5DKlbFzh+nHt/CkJGZ9YsfnB58gS4fdve0ggmIIqarXFxASpWFEVNEFJCahW1mBjg6FHLyiQIaY0zZ4Cvv2aDASDuzzSCKGr2QJf5Ka1tBME0QkOBnDmBXLlSvm6tWoBSEqcmZGyiooDu3fl/tH49TxNFLU0gipo98PUFHj7kdlKCICSPORmfOnLnBry9RVETMjYTJgDBwcAvvwCenlzXUxS1NIEoavZAEgoEIWWYU0PNkLp1gYMHOZBaEDIax44BkyaxRa1dO57m5SWKWhpBFDV74OPD76KoCULyELHr01yLGsCK2vPnwOnTFhNLENIEr16xgla4MPD992+m6xQ1CcFxeFxMWUgpVQBAHQBvAXgF4AyAICKS3F5zyJWLbzqiqAlC8jx9ykpWahU1gN2flStbQipBSBt89RVw4QKwY0fcGE8vL66ndu8etzYUHJYkLWpKqYZKqW0ANgN4B0BhAF4AvgZwWik1Tinlbn0x0yHSSkoQTEOX8Zka12fx4kDRohKnJmQsdu8GZs8GBgwAmjSJO8/Li9+l+LrDk5xFrQWAPkR0M/4MpZQLgFYAmgJYbQXZ0je+vsCmTWyWdnOztzSC4LiYW+zWEKXYqrZ3L7t6lLKIaILgsDx7BvToAZQpA0ydmnC+TlE7dw5o3NimogkpI0mLGhENN6akaefFENE6IhIlzRx8fbkq9Nmz9pZEEBwbS1jUAFbU7t59M54gpGeGDgVu3QKWLAGyZUs4v2BBzoiWhAKHx9QYtVwAugPwMFyHiAZbRaqMgGHmZ9Wq9pVFEByZ0FAga1Ygb97UjWMYp5ZapU8QHJlNm4AFC4Avv+Q6gsZQSjI/0wimZn1uAStppwEcN3gJ5lKqFJA9u8SpCUJy6GqopdZdWbEi4O4ucWpC+ubhQ6B3b64uMHZs0st6eoqilgYwyaIGIAsRDbWqJBkNJyegUiVR1AQhOVJbQ02HszNQu7YoakL6hQjo3x94/BjYtg1wdU16eS8v4LffgLAwLoArOCSmWtT+UEr1UUoVVkrl0b2sKllGQFpJCULypLaGmiF163Jc6OPHlhlPEByJFSuAVauAcePehNckhWFCgeCwmKqoRQGYDuAQ3rg9g6wlVIbB15fr2Nw0mq8hCEJEBCtVllTUAO5SYC2++gpo1kxufoJtuXuXy3DUrAkMH27aOqKopQlMVdSGAihDRB5EVFL7KmVNwTIEuieekBC7iiEIDoulMj51VKsGZMoEHDhgmfHi88cfwOTJXL+qShXg22+5GbYgWBMi4OOPgchIzvJ0MTGqqWhRjpWWWmoOjamK2lkAL1MysFJqoVLqgVLqjMG0sUqpO0qpEO2rhcG8kUqpK0qpi0qp5gbT/ZVSp7Xz5iiVjgogVarEAdISpyY4OgsWAGvX2n67lqihZkjWrIC/v3Xi1E6fBj75BKhfn+V+911g9GjO6j52zPLbEwQdv/4KbN0KTJsGlC1r+nqS+ZkmMFVRiwUQopSar1WW5iil5iSzzmIAbxuZ/h0RVda+tgCAUsoLQGcA3tp1flZKOWuXnwugL4Cy2pexMdMm2bMDpUuLoiY4NkTszps40fbb1lnULKWoAez+PHqUrQ+W4tkz4L33gJw5gcBAbsmzfDmwfj3w6BG7oz7/HHiZouddQUiea9e4ZlrjxpxIkFJEUXN4TFXU1gGYCOAgTCzPQUR7AZgasdsWwAoiek1E1wFcAVBdKVUYgDsRHSIiArAEQDsTx0wbSCspwdH57z/gwQO2GNnajXfjBpA5M1CokOXGrFuX9+O4hSoMEQE9e/INMzAwrqxt2vBNsE8fYOZMtqLv2mWZ7QpCbCx3H3B2BhYt4moCKcXLi/t9PnlicfEEy2DSr0pEvxt7mbnNgUqpU1rXaG7ttCIAbhksc1s7rYj2c/zp6QdfX+DqVW46LQiOSHAwv0dF2b6TRmgo9+k05waUGLVr87ul3J+zZgFr1nCbnvr1E87PmROYN48VNCcnoFEjoG9fbjYvCKlh9mxg3z5gzhygWDHzxpCenw6PSVc/pVQrpVSwUuqxUuqZUuq5UuqZGdubC6A0gMoA7gGYqduEkWUpiemJydlXKRWklAoKCwszQzw7oEsoOH3avnIIQmLoFDUAOHHCttu2VA01Q/LnB8qXt4yitm8fMGIEx6MNTabUZEAAW8+HD+eYP29vdo0KgjmcPQuMGgW0bQt0727+OJ6e/C7uT4fF1MfU2QA+ApCXiNyJKAcRuad0Y0R0n4hiiUgD4FcA1bWzbgMwfBwoCuCudnpRI9MTG/8XIqpKRFXzp5XifYatpATBEQkO5k4a7u6WcxeaiiVrqBlSty5nfmo05o/x33/A++/zsVm0yLTOCVmzcsD3kSNAvnxAu3ZAp07A/fvmyyFkPKKjWTnLkQP45ZfUde0oUQJwcxNFzYExVVG7BeCMNk7MbLQxZzraA9BlhG4A0Fkp5aqUKglOGjhKRPcAPFdK1dRme3YHkL4eQYsXB/LkAf79196SCIJxgoMBPz8uN2FLi1pkJCsw1lLUnjwx390TE8MKVng4sHo1K7EpoWpVICgImDABWLeO3U9Llkjxa8E0Jk7k/+L8+UCBAqkby9kZqFBBFDUHxlRF7QsAW7QlNIbqXkmtoJRaDi6QW14pdVsp9TGAadpSG6cANATwGQAQ0VkAKwGcA7AVwAAiitUO1Q/Ab+AEg6sA/k7ZLjo4SgGffgr89ZftrRWCkBzh4RwkX7kyl7U4eZKVFFugKwRtjQbqhg3azWHUKGDvXrZmVKpk3hiZMvE4ISF8o/zoI+Cdd95kugqCMXQK/ocfssvdEnh5SYyaA2OqojYRXEctC4AcBq9EIaIuRFSYiDIRUVEiWkBE3YioEhH5EFEbrcVMt/xEIipNROWJ6G+D6UFEVFE7b2BqrXoOyRdfsBvkiy/kiVpwLHQu+SpV2KoWGWm7C7qla6gZUro0ULCgeYra2rXsvvz0U75ZphZPT451++EHlsfbG/jxx9S5ZYX0yatX7PIsVIjPF0vh5cUPRpLU5pCYqqjlIaJ3iWgMEY3TvawqWUYiZ07gm2/Y/bltm72lEYQ36BIJqlRhixpgO8uvNWqo6VCKrWopVdQuX+ZyCFWrcsadpXByAgYO5ADxunWBQYOAevXEyiHE5euv+ZxYuBDIlcty4+oyPy9csNyYgsUwVVH7RynVzKqSZHQ+/ZSDkr/4gmvjCIIjEBzMlqfChbniebZstotTu3GD42eKWKkiT926bLW7fTvZRQFwsdr33uP2PH/9Bbi6Wl6mEiWAv//meLULF9jlPHo09zwVMjaHDwPffQf068e9ZC2J9Px0aExV1AYA2KqUepXK8hxCYmTOzD0CT5/mfoGC4AgEB7M1DWClqUoV21nUQkO5F6GpfQtTii5OzZS+n0R8gzxzBli2zDpWPh1KAd268U3zvfe4X2i5clzSQx7iMi6TJgF587Lb3dKUKsX3IFHUHBJTC97mICInInJLTXkOIRk6duSm0V9/zbEIgmBPXr/mC3flym+m+flx8LstFAZr1FAzpHJlthCa4v789Ve2co0eDbxtoy52BQsCf/4JHDzIx6F3bz7+O3bYZvuC43DxIrBxI7eIyp7d8uO7uHBtQVHUHJIkFTWllEcy85VSqmhSywgpQClg+nTgzh3g++/tLY2Q0Tl7ljM8dRY1gOPUXr7kG4e1uXHDupYrFxfuwZmcohYUxDFjzZtzLKmtqVWLrX4rV3Kwd7NmQIsWtu8SIdiP775jV7s5vTxNxdNTFDUHJTmL2nSl1GqlVHellLdSqoBSqrhSqpFS6lsABwB42kDOjEODBkCrVuwGffjQ3tIIGRnDRAIdfn78bu04tehofmCxpqIGsPvz1CkuQ2KMx4+BDh3YurV0Kbt/7YFSbHE/fx6YMYOtbD4+HNsqxXLTN2FhwO+/c4ZxwYLW246XF3D9unhzHJAkFTUi6gjgGwDlAfwEYB+44GxvABcBNCIiscNbmilTOHh44kR7SyJkZIKDufJ56dJvplWowFXMrR2ndvs2l6ewpusTYEVNo+FA7fhoNHxzvHuXkwfy5bOuLKbg6goMG8b9gQcO5Li1MmU4fklusOmTuXO5LE5yLcpSi5cXx2LawloupIhkY9SI6BwRjSKiAG2NsypE9AERLSWiSFsImeHw9gZ69QJ++omLjQqCPQgO5hZnhg3RXVx4mrUtatasoWZIjRpsJTOWUDBxImdgfv89UL16wvn2JG9eluvsWaBJEy6cW64cJyJJ/bX0Q2Qk3wfeeedNZqa1kMxPh8XUrE/B1owbxzfFUaPsLYmQEdFouNitYSKBDn9/VuKsqRBYs4aaITly8D7Gj1PbsQMYMwbo2pXdi45KuXJcgHf3bnaLde/OSuWePfaWTLAES5cCDx6wFdXalC3LDy2iqDkcoqg5Km+9xX/OFSuAY8fsLY2Q0bhyBXjxIm58mg4/Pw5qv3LFetu/cYPjsooVs942dNSty67P6Gj+fvMm0KULWxjmz09dw2tb0aABcPQoW9QePAACArjh+6VL9pZMMBeNBpg1iy3YjRpZf3uZM7OyJoqawyGKmiMzfDjHxQwfLq2lBNtiLJFAhy06FISGcpFdaxSVjU/duhzfFRwMREUB77/P76tXc/mOtIKTE8fUXbzIMWv//sthFIMHS2JSWmTrVk4eGTbMdg8LXl6iqDkgyZXn+NDgc5148wZaSyhBi7s7u1/27AG2bLG3NEJGIjiYm4Z7eyec5+XFCpQ149SsXZrDkDraS9v+/XxTPHKEW/SUL2+b7VsaNzdg5Ehud9W7N8c4lSnD2aI6q6Hg+MycyZ6VTp1st00vL7aUv35tu20KyZKcRc0wzSR+B9heFpZFMEbfvnyRHTFCqpILtiM4mJW0zJkTzsuUiUtDWNOiZu1it4YULsyV2WfN4mboQ4dySY60TsGCnDF4+jQro8OHSyZ5WiEkhC2igwcb/w9aC09Pvs9cvmy7bQrJkpyiphL5bOy7YA10raXOnuVaOoJgbYhYUTOWSKDD358tatZwycfGcpyYrSxqALs/79zh9ylTbLddW+DlBWzeDLRvz43kE6sZJzgOs2ax271vX9tuVzI/HZLkFDVK5LOx74K1eO89LiPwzTdcFV4QrMm9e1xk01h8mg4/P77hW6N8zL173BHBVhY1gN1Lfn5AYCBbDNMj33zDv9kP8Z0jgkNx5w6wfDnw8cdA7ty23Xb58hwPJ4qaQ5GcolZBKXVKKXXa4LPuexoN4EiD6FpL3b3LT8SCYE2SSiTQoUsosEacmq1KcxjSogW7ct96y3bbtDVVqgCtW3M7oufP7S2NkBg//MAZn0OG2H7bbm4cBnD+vO23LSRKcoqaJ4DWAFoZfNZ9t3L1PSEO9eoBbdqwWyYszN7SCOkZnaLm65v4Mt7ebHmyRpyarYrdZkS++YbbYv38s70lEYwREcElYd59FyhZ0j4ySOanw5GcotYeQAEAd4joRvyXDeQTDJkyhWtbffutvSUR0jPBwZzA4u6e+DKurkClSunHopZRqFYNePttzgB98cLe0gjxWbgQePrUNgVuE8PLi0u8xMTYTwYhDskpakUBfA/ggVJqt1JqklKqpVIqjw1kE+Lj6cnp9nPnWrfYqJCxCQlJOpFAh58fW9QsnVAQGgrkzw9kzWrZcQVm9GiuqzZvnr0lEQyJjeXQltq1gZo17SeHlxeXcbl61X4yCHFIrin750RUG0AhAF8BeAwuy3FGKSW2UXswdixngkprKcEa6BIEkopP0+Hvz260mzctK4Mta6hlRGrV4v6g06dLI3dHYu1a4Pp1+1rTAMn8dEBM7UzgBsAdQE7t6y6AI9YSSkiCwoWBzz8HVq7kwpyCYElCQvjdFEXNz4/fLR2nZssaahmVb74B7t8Hfv3V3pIIOmbOBEqXBtq2ta8cFSrwuyhqDkNynQl+UUodABAIoBaAgwA6ElFVIuppCwEFI3z+OVCgAPDFF9JaSrAspmR86vDx4SbOloxTIxKLmi2oX5/7g06dCkRG2lsa4eBB7jc7ZAj/p+xJ9uxA8eKiqDkQyVnUigNwBfAfgDsAbgN4amWZhOTIkYNbS+3dC2zaZG9phPRESAhQqBC/kiNLFs7+tKRF7cEDVhxEUbM+33zDJX8WLbK3JMLMmVwzraeD2D8k89OhSC5G7W0A1QDM0E4aBuCYUmq7UmqctYUTkqBPH6BsWW4tJdk5gqVIriNBfPz9LZtQoMv4FNen9WnUiAPXJ0/mJvSCfbh6lePTPvmEuxE4Al5ewIUL0rbQQUg2Ro2YMwC2APgbwAEApQH8z8qyCUmRKROX6zh/Xp6IBcvw+jU/RZvi9tTh58d1/e7csYwMUkPNdijFGaC3bkl7Onsyezbg4gIMGmRvSd7g5cWW7RtShcsRSC5GbbBSaoVS6haAveBCtxcBvAtASnTYm/btOYNrzBipiSSknjNn2DqbEkXN0h0KpIaabWnWjGurTZ7MJRkE2/L4MddO69LFsbpiSOanQ5GcRc0DwF8AqhNRKSLqRkQ/E9FJItJYXzwhSXStpe7d47YwgpAaUpJIoMPXF3ByslycWmgokCsXkDOnZcYTkkZnVbt+HVi2zN7SZDzmz+f+zUOH2luSuHh68rsoag5Bcora1wDWE9E9AFBKlVdKfaaUetf6ogkmUacO0K4dZ289eGBvaYS0TEgIJ6qUKmX6Olmz8kXdkhY1sabZlpYtWTmfNEniXW1JVBT39WzSJOl2bfYgVy628Imi5hAkp6htBVvVoJQqA+AQgFIABiilJltXNMFkJk/mwpXjx9tbEiEtExz8xkKWEnQdCiyB1FCzPUpxBujly0BgoL2lyTisWMHeEHsXuE0Myfx0GJK7Iucmosvazx8BWE5EgwC8A45XExyBChU4C3T+fODSJXtLI6RFYmOBkydT5vbU4e/PN5x791InAxG7PsWiZnvatuXerRMmSKafLSDikhze3kDz5vaWxjg6RU1qddqd5BQ1w1+oEYAdAEBEUQAkRs2RGDOGG2V/9ZW9JRHSIleucEKKOYqarkNBat2fT54AERGiqNkDJye2ql24AKxebW9p7M/Tp9Z1A+/cCZw6xbFpSllvO6nB05OvCbdu2VuSDE9yitoppdQMpdRnAMoA2A4ASqlc1hZMSCGFCgHDh/NF9quv+CIgT0KCqZiTSKCjcmW+2aTW/Sk11OzLe+/xzfnbbwFNBnwOv38f+Okn7tqQJw9bGA8etM62Zs4EChYEuna1zviWQJf5ef68feUQklXU+gB4CI5Ta0ZEL7XTvfCmCK7gKAwbBrzzDicW+PoC5cpxQdyjR0VpE5ImJIRr8+kuzikhRw4+11JrUZMaavbFyQn4+msu07Junb2lsQ2PHnG/0yZNOHh+4ECe9sUXnI1Zty4weDBbei3F2bPA1q28LVdXy41raaREh8OQXGeCV0Q0hYj+R0QnDaYfJKI/rC+ekCKyZwe2bOG2MPPnc/berFlAjRp88xsyBNi3T2JQhIQEB3O8TObM5q2v61CQGqSGmv3p1Ik7nnz7bfp9uHv6FFi8mB9qCxUC+vYFbt5kT8Tp06xITZnCCuvAgcCPPwIVKwLbtllm+7NmAW5uQL9+lhnPWuTLB+TPL4qaA5DC9C4hTVCwIF98tm1jc/7ixeyemjePzfpFivBF4p9/pMilwDfk4GDz3J46/PyA27dTVyLmxg1uoZM3r/ljCKnD2RkYNYotrOmpj/Dz58CffwJt2vD1sWdPdukNHcoPGBcvsnJaseKbdXLkAObM4YdbNzfg7beBjz5ii5u5/PcfsHQp0KNH2jjPJfPTIRBFLb2TJw9fXDZs4FY/y5cD9eoBS5YATZvyE2WvXnxRfv3a3tIK9uDuXT43UqOoWaJDgS7j01GDqzMKH3zA1vjx49O2Ve3lS2DVKqBDB6BAAY4HO3ECGDAAOHyYi/xOncoPGUmdc3Xq8IPMqFGs7Hl58bjmHJuffuKH488+M3+/bIlkfjoEJilqSikPI9OqWVwawbrkyAF07swXmbAwYM0aoEULfm/dms3cH3zACQnSkirjkJpEAh26dVOjqEkNNccgUyZg5EggKMhy7j5b8fo1sH49X8cKFADefx/Yvx/4+GNg7152cerCQVLyQJAlC5cuCQoCihXjcd99lx9yTOXlS2DuXLbqlS2b8n2zB15e7Cr+7z97S5KhMdWitkYpVUT3RSnVAMBC64gk2ISsWblX6B9/sLtqyxa++Gzfzk+g+fMD3bpxIV0hfRMSwu8+PuaPkTMnUKZM6uLUpIaa49C9O1C8ODBunONbU6Ki+Pr10UesnLVrxwpm165cBuPOHY4zq1cv5cWc4+Pry9a4adM4IcDLC/jtN9OO0e+/s9vUUQvcGkMSChwCU8/aTwCsU0oVUkq1APA9gBbWE0uwKZkzc2Dtb7/xk9POndwkeOlS4K+/7C2dYG2Cg1nJcndP3Th+fuZb1J4/5zpqoqg5Bpkzs1Xt8GG+HjgaMTEcY9unD4dvtGzJlrR33wX+/puvY/PnA40acdydJXFx4VJIp05x7G+fPpw1evVq4utoNNyPuVo1ziRNK4ii5hCYpKgR0TEAg8F11MYCaEpEUgUvPeLiwhe3X3/lJ+rly+0tkWBtUptIoMPfn61i5gRbSw01x6NnT048+vZbe0vCaDTAnj1A//5cSqNpU27D1KIFx+Devw8sWsRB/5kyWV+esmWBf/9lhTAoiOuuzZplPKt+40Zu0TVsWNqKwSxYkPt+Si01u5KkoqaU2qiU2qCU2gBgJICsAF4DWKCdJqRXnJw4nm3HDuDhQ3tLI1iLp085qNoSipquQ4Eu5i0lSA01x8PVlesw7t3LCpI9IAIOHeLSQsWKAQEBnMUeEMDW/gcP2PLfurV9apI5OXGG/dmzbFUbNgyoXZtLexgycyY/+L73nu1lTA1KSeanA5CcRW0GgJkGr48BfG3wXUjPdOnCLgZxf6ZfdPFpllTUzIlTkxpqjknv3uxatKVVjYgtVMOHs4W1dm0uLVS9Olv4HzwAVq5kpcfNzXZyJUXRoux6XbGCH3z8/Lit3+vXwLFjXOLjf/9jj0VaQxQ1u5PkWUNEdnqMEhwCX19u+L5iBfDpp/aWRrAGOkWtcuXUj5UnD99YzYlTu3GDLSIFC6ZeDsFyuLmxwjRsGHDgAJeqsAZEXGw2MJBfV6+yUtOsGSuJbdtywoojoxQXDG7cmMtvjB/PD7n583P8Z+/e9pbQPHQJE2FhvC+CzTG1PEdNpdQxpVSEUipKKRWrlHqWzDoLlVIPlFJnDKblUUrtUEpd1r7nNpg3Uil1RSl1USnV3GC6v1LqtHbeHKXSkoM/jaMUuz/37uXMKSH9ERzMFpNChSwznrkdCkJD2TWU2qw8wfJ88gnfoK1hVTt/Hhg7lpUBX1/uCFCqFCsG9+8DmzdzBqqjK2mG5MvH2fRbtnCSzJ49nHCQ2mQde5HRe35qNGwVtSOmXhV/BNAFwGUAbgB6a6clxWIAb8eb9iWAnURUFsBO7XcopbwAdAbgrV3nZ6WULlVnLoC+AMpqX/HHFKxJly78tBsYaG9JLEt0NKfrz5tnb0nsi6USCXT4+bE15OnTlK0nNdQcl2zZ2KK2bRv3DU4tV68CkyaxYublxZanggWBn38G7t3jEkEff8wW2rTMO+9w7NrcucA339hbGvPJqJmfsbHsTfLxAWrWTDqr18qY/PhKRFcAOBNRLBEtAhCQzPJ7ATyON7ktgN+1n38H0M5g+goiek1E1wFcAVBdKVUYgDsRHSIiArDEYB3BFpQrxzff9Jb9uWIFF8IcPTrj1oqLjOSLryUVNV2HgpQmFNy4IfFpjkz//qw4mWtVu3EDmD4dqFqVS8GMGsW9ib//nluP7d7Nbe0KFLCo2HYnRw4OG0lLFsH4FC3Kv1VGUdRiYjhBxdubDRUaDX+344OkqYraS6VUZgAhSqlpSqnPAGQzY3sFiegeAGjfdf/KIgAMy33c1k4rov0cf7pgS7p04eDey5ftLYllIOKClfnycdzFokX2lsg+nDnDT42WtqgBKYtTe/WK3VyiqDkuOXJwX8xNm0z/be/eZUWsdm2+yX3xBYdTTJ/OituBA8DgwVxqQ3BcMkrmZ3Q0ZxR7enKx98yZOWnlzBm+B1q6Hl8KMFVR66ZddiCAFwCKAbBknrGxuDNKYrrxQZTqq5QKUkoFhYWFWUy4DE+nTvy+YoV95bAUW7bwn2/GDKBWLb5xxMTYWyrbY8lEAh3583MZhZTEqd28ye/i+nRsBg7kmloTJiS+zIMH7MJs0IAtMUOGcOukSZOAK1c41ufzzzkeUUg7eHml3xi1qCiuG1quHNcOzJEDWLuWr48dOzpE3KypBW9vgJWmwkQ0joiGal2hKeW+1p0J7fsD7fTbYOVPR1EAd7XTixqZnpicvxBRVSKqml+yUyxHsWIcz7V8ueO3kzGFadP4JtKlC1dfDw1NfzF4phAczBelUqUsO25KOxRIDbW0Qc6cXGJi7Vquyq/j8WMO/m/aFChcmJueh4VxeYrz5/mGN3IkULq03UQXUomnJ1tIUxp76si8fs3xg2XKcC28AgXYYnz8OLchcwAFTYepWZ+tAYQA2Kr9XtnMgrcbAHyk/fwRgPUG0zsrpVyVUiXBSQNHte7R59qsUwWgu8E6gi3p0oUvuqdP21uS1HH4MGexDh3Kpu2WLTkWYcqU9KGEpoTgYLamWfqC5O8PXLrEGW+mIDXU0g7/+x8r96NHA0uWcFeAggU5qzE0lBWyU6c4iH7MGC7vI6R90lPm56tXwJw5/ODQvz8bIrZu5XtDy5YO2TnC1Cv0WADVATwFACIKAeCR1ApKqeUADgEor5S6rZT6GMAUAE2VUpcBNNV+BxGdBbASwDmwMjiAiHR9OPoB+A2cYHAVwN8myixYkg4d2Eef1pMKpk4FcufmGwvASsqIEewK3bzZvrLZkthY4ORJy8an6fDzY6VX51pNjhs3uGaWxCo5PrlzA4MGcXHXjz7iuKXPPmMrxKVL7BatVMkhb3ZCKkgPmZ8vXnCLr5Il+YGjTBnuY7t/P9C8uUOfs6aWSY4hovCUlDAjoi6JzGqcyPITAUw0Mj0IQEWTNyxYh/z53/TWmzTJoU/qRLlwgW8wuowzHZ07c/r8lClAq1b2k8+WXL7MsUPWUNR0mZ/Hj7PLPDlCQ9kVnRartmdEvvySY9Xq1QNq1Eib1wIhZZQowcWP06Ki9vw5x03OnMku+caNOdSlQQN7S2YyplrUziilPgDgrJQqq5T6AcBBK8olOCKdO/NN9fBhe0tiHtOnc/X7QYPiTs+UiQOcDxzgVi8ZAWskEugoVIhjlUyNU5PSHGmLHDm4W0HNmqKkZRScndmNnZYUtfBwYOJETlL68kt+gDxwAPjnnzSlpAGmK2qDwMVoXwP4E0A4gCFWkklwVNq3Z0UnLbo/797lauG9ehmv1dSrF1sNp0yxvWz2IDiYFVSdS8PSpKRDgRS7FQTHx5FLdBBxj9VVq1gpa9qUrfRff81tz44cAf7+m0vFpEGSVNSUUlmUUkMATANwE0AtIqpGRF8TUaQtBBQcCHd3DrZcuTLtlbOYPZvjsoYNMz4/a1aOW9iyJW5GW3olOBioWJETKqyBnx+7ml+8SHq5qChuTyYWNUFwbLy8uJSOqUlCKSQ2lushL1yYzIJE7Nn56y9OXmnWjGtilioFvP8+x6E9eQJ8+CE/LG7YAFSvbhWZbUVyFrXfAVQFcBrAOwBmWF0iwbHp0oWLk+7ebW9JTOfpU24V1bFj0qUo+vfn2LX0blUjsnzrqPj4+3NF75Mnk17u9m2WRyxqguDY6KzvFy5YZfgLF1ivmjnTIAGfiC3uq1cDX33FQf/583NCQMeOXAvz0SPgvff4Gh8UxIpkUBCX3tAV4E7jJBe960VElQBAKbUAgAUavQlpmpYtOUZl+XKgSRN7S2Ma8+bxn3fEiKSXy52b273MmsXZa5auL+Yo3L0LPHxoXUXNsENBUu4GKc0hCGkDwxId1apZfHhd6PO5c0Bw75/gd2cjK1yPHvEMFxf2ArRvzw+CVatyhrGrq8VlcTSSs6hF6z4QURrzdQlWwc2NiwGuWcMFAx2dyEh2ezZtappi8tlnfEGYkY6Nx7o+nNZIJNBRpAjHAiYXpybFbgUhbVCqFIdKWClO7ci/L5ADz5EZr7FkUQzw339A27acsXnkCD9sBwdzF4FPP2VFLQMoaUDyipqvUuqZ9vUcgI/us1LqmS0EFByQLl3Ynbhtm70lSZ4//mBXbXLWNB1vvcX1oRYu5AtFeiQ4mLP1fH2ttw2lTOtQcOMGL1usWNLLCYJgX1xcuM2SlRS1w1seobY6hDb1w/FnvsGIPhYCLFgA9OvHMWZZslhlu2mBJBU1InImInftKwcRuRh8dreVkIKD0aQJkDev42d/xsZySQ5/f6BRI9PXGz6cG/R+/731ZLMnwcFc7DFHDutux9+fK9S/epX4MqGhrBxbK6lBEATLYaXMz4it+3H2aRHUqJcZHw0vgLAwha1bLb6ZNIvjNLMS0g6ZMnEg54YNyWf12ZN167iw6xdfpKzeU9my3Inh55+5Fk96w9qJBDr8/FhZTqrtmNRQE4S0g5cXcO1a0g9fKSUmBkH9F0IDZ9T4X019vsCSJZbbRFpHFDXBPDp35sr2G8xp+WoDiLhdVOnSnBGUUr78Enj2jDOH0hNPn7IVyxaKmmGHgsSQGmqCkHbw8uJr68WLlhtz7lwcvs61LWs0yIJMmYAPPuBby5MnlttMWkYUNcE86tXjgHFHdX/u3g0cO8YdB5ydU75+lSqcCj57tmWfHu2NNTsSxKd4cSBPnsTj1GJjgVu3xKImCGkFS/f8fPAA+OYbHCnQGmXKEPLm5cndu3OJxZUrLbOZtI4oaoJ5ODkBnToBW7cCjx/bW5qETJvGWYcffWT+GF9+yYkIixdbTCy7o8v4tIVFTamkOxTcvcuFk0VRE4S0Qdmy/OBrKUVt5EjQi5c4TDVQs+ab8JQqVQBvb3F/6hBFTTCfLl046H7tWntLEpeTJ1mB/N//uJyIuTRowE2np0+3XScGojd1g6xBcDD34SxY0HrbMMTPDzhzxngpF10NNXF9CkLaIHNmVtbOn0/9WEePAgsX4lbvcfgvzAU1aryZpRRb1Q4eBK5cSf2m0jqiqAnm4+/P2YOO5v6cNo07DPTrl7pxlOIWJboectZGowH69GEl6p9/rLMNWyUS6PD3Z2X+zJmE86SGmiCkPSyR+anRAAMGAIUL43DNIQCAmjXjLtK1KztuxKomipqQGpRiq9quXY5Tcyw0FAgMBPr25U4DqaV1a8DTk9tK6fuaWAEivnAtWMBlMz780PLHNDKSn4RtqagZdiiIj3QlEIS0h6cnZ9NHRZk/xsKF3HVgxgwcOeUGV1fAxyfuIkWKcCWoP/5gvS4jI4qakDq6dOF/kaNEfc6cyY9hn31mmfGcnLhY7qlTwN9/W2bM+BABgwdzq6uRI4F9+zjj9IMPOODeUpw5w+PZIpFAR6lSQM6cxuPUbtzgOMLUuKcFQbAtXl58Hbl82bz1Hz/m+N969YAuXXD4MBvejZVS7N6dn73370+VxGkeUdSE1OHpyRXuHcH9+fAhW6S6dgWKFrXcuB98wJXzrdGsnQgYNgz48UfOUJ04kfvZ/fQTWyq//dZy27JlIoGOpDoUhIaKNU0Q0hqpzfwcPZrrbvz4I6KiFU6cQJz4NEPateMolozu/hRFTUg9XbpwR93r1+0rx48/cimNL76w7LiZMrEStW8fcOCA5cYlYgvad99x4sO0aW8K8/bowY+T48cDO3daZnvBwYC7O1CypGXGMxV/f7ZIRkfHnS411AQh7VG+PF+nzFHUQkK4NuWAAYCPD06d4oiM+PFpOrJl49rjK1emrypJKUUUNSH1dOrE7ytW2E+GFy+AH34A2rRhK5+l+fhjbptlSava6NFclLdfP1bWDLsnKMWdESpUYAuhJeLVgoPZ7elk47+9nx9nfRpe2DUa6UogCGkRNzcOaUipokYEDBzI19Hx4wFwr3UgcYsawM+rz58D69ebKW86QBQ1IfV4eAC1atnX/blgAcc+WNqapiNbNrZ6bdqUdEskUxk/HpgwAejdmy2BxlpcZcvGj5LPnrGylpp4tdhYtmrZMj5Nh7EOBQ8esPImipog2Jy9e/kZ0WzMyfxctow9EpMnA7lyAWBFrWBBro2dGA0a8Pzffzdf3LSOKGqCZejShRWYs2dtv+3oaE4iqFOHX9ZiwABWnlJ1hQNfqMaMYffm/PlJW7h08Wr//suKnblcvswtv2wZn6ZD1wDeME5NaqgJgl0gAoYM4Xj+Bw/MHMTLC7h0yfT6ks+eAcOHA9WqAT176icfPsxuz6RaMTs5Ad26Adu3A/fumSlvGkcUNcEyvP8+/6Ps4f5cuRK4eZOzM61JnjzAJ5/wPpobjzdzJvDVV2wh++0309yQuni1ceNYYTMHeyQS6HBy4u0aWtSkhpog2IWjR99cDswOf/Xy4vIc166Ztvy333KXl59+0l/zHj3i58ek3J46unXjaIk//zRT3jSOKGqCZShYEGjUiN2f1qw3Fh8iDsL38gJatrT+9oYO5QvNjBkpX/f77zkp4f33uS2VqT1IDePVPvjAvHi14GDOf9dlbNkaPz/uGKF7ApcaaoJgF+bO5UzKXLnYSmUWKcn8PH+eeyZ//DFb1LQcPcrviSUSGFK+PCt0GTX7UxQ1wXJ06QJcvcqFDG3F1q0ce/XFF7YJki9ShK1bCxfyE6Kp/Pwz+xvefRdYuhRwcUnZdlMbrxYczG7UTJlStp6l8PfntK0LF/j7jRtckNjd3T7yCEIG5PFjrgferRsXk92xw8zn6goV+D05RU1XIzJ7dmDSpDizjhzhZ9CqVU3bZPfufKk/edIMedM4oqgJluPdd9lqY8ukgqlTuWZaly622+bw4RwIP2eOacv/+ivHt7VuzcfGXGWpYkVOPEhpvBoRp8XbI5FAR/wOBVJDTRBszuLFXA6jXz+gaVPgzp03z04pInt2jvBPTlFbs4bb4U2YAOTPH2fW4cN8ScuRw7RNdurEl86MaFUTRU2wHLlyAe+8w49slqyonxhHjgB79nAXAmNlra1F+fLAe+9xvMWzZ0kvu3gxx7W98w73C02tnD178uNwSuLV7tzhYsD2iE/TUb48kDXrmzg1qaEmCDZFo2G3Z506QKVKrKgBbFUzi+QyP1++5Guzry9fA+PJcvSoafFpOvLmBVq14uRRU3MY0guiqJkJEd/7hHh07gzcvcvFYa3NtGmsHPbpY/1txefLL4HwcG77lBhLlwK9erGPYc0awNU19dvVxauVL296vJo9Ewl0ODuzRe/ECf7zSA01QbApO3cCV66wNQ3gutdlyqQyTu38+cQfyidPBm7d4vqW8UI9Ll/m5gSmxKcZ8tFHHHFitsxpFFHUzKRNG/ZkCfFo3ZotJ9bO/rx4EVi7ll2KptrOLYm/Pz+Sfvcd+xLiExjIV5WAAGDdOiBLFsttO3t2ts49e8bN25OzXgYHs4Ln62s5GczB359lefgQiIgQRU0QbMjcuUC+fFzpX0fTpsDu3Wb2V/fy4mufLjHIkCtX+EG6a1fu6RkPUwrdGuOdd9iyltHcn6KomUmNGuxjz6h1XRIlWzagbVvgr78StgyyJDNmsIVq8GDrbSM5vvySLVrxKzGuXs0XqDp1gI0bWXG1NLp4tZ07uT9oUgQHA2XLsoJnT/z8uIOEztcirk9BsAl37gAbNrCB39Cw36wZ/yUPHzZjUF3m5/nzCefpwlGmTTO66uHD/Hyd0iYymTNzOPK6dcDTpylbNy0jipqZtGvH7xs22FUMx6RLFy6SY3bwQzLcu8ePVD17AgUKWGcbptCwIaebT5v2JmhiwwZ2/1avDmzezIqrtdDFq40dm3S8mr0TCXToOhSsWcPvYlETBJvw668cFxYvVAwNG3JUglmuRJ2WFT9ObfNm7uAyZgzw1ltGVz1yhC+dplYoMqR7d87l+uuvlK+bVhFFzUy8vbndWUbuP5YozZtz6QVrZX/Ons2K0bBh1hnfVJTipurXrvFVY8sW9iv4+QF//219l6xhvFrXrsbLhTx5whmW9oxP0+HpyS7gv//m76KoCYLViY5mRe3tt/meZUjOnPxMadYzda5crIgZKmqRkdxqr0KFRL0dL19ymY2Uuj11VK3Kw2ck96coamaiFFvVdu7khrGCAZkzc1bkunX8r7QkugD+Dh2A0qUtO7Y5tG3LitKIEVyepFIlYNs2vgLaAl28Wni48Xi1kBB+dwRFzcWF4+RevmS58+Sxt0SCkO7ZsIHzu3RJBPFp2pRLXz55Ysbg8TM/Z83iWppz5iSa4X7iBD9npzSRQIdSbFXbt8/0xghpHVHUUkG7dhyEuXWrvSVxQLp04YDxzZstN2ZkJMdjPXtm/XZRpuLkxLLcvMmPeTt26BsO2wxdvNo//yQoKukQGZ+G6OqplSiRdIM/QRAswty5XPKsRQvj85s1Y7eoWd3pPD1ZUSPia+CECfzAqqv9YQRzEwkM+fBDvnz88Yf5Y6QlRFFLBbVrcxbNunX2lsQBadAAKFw49e5PIuDQIeDTT3m86dNZQ9bd8B2Bbt24U8E//9jPSmQYr7Zr15vpwcHsnrBnLJ8hujg1cXsKgtW5dIm9Pn37Jh4PVr06R2mYFafm5cUP5Ldvc3s8IraqJcHhw5xHVLCgGdvTUqwYdyxcssS2HQvthShqqSA8/DFat2ajkTUTHNMkzs7c03LLFnbLpZTQUG7kW64ca8RLlnAvz+3bHS+K1MWFFaV8+ewngy5erVw5rq+mi1ezUCLBoEHA+PGpHuaNgi0Zn4JgdebN42r+H3+c+DKZMnFSgVlxarrMzx9+4BCMr75K9iHsyJHUWdN0dO/Ors+DB1M/lqMjipqZBAUFoWTJksif/y+Eh3OBfCEenTtzeo6pJsdnz9gyFRDA1RhHj+ZHp0WLWPFYupRN6uakCmUEsmfnfqBPn7Jv4MULTp1PpdvzwQPWASdO5FiXVOHtzcpk3bqpHEgQhKR4+ZIbo7z7LlCoUNLLNm0KXL/O4WUpQqeoTZ/OmQrDhye5+N27XAPX3Pg0Q959lysfZYSkAlHUzMTLywuVKlXCd999gMyZd4j70xg1arDClZT7MzaWg++7duWryccf8795wgS2qv37L9Cjh32K2qZFKlV6E6/WrRsf31QqamvWcAxLVBTX900VmTNzsWJb9mYVhAxIYCAnCCSWRGBIs2b8nmKrWr58b3p4zp6dbGFvS8Sn6cienXPWAgON1xxPT4iiZiZZs2bFxo0bUaFCBcTGtseqVUczhK88RSjFVrV//gHCwuLOO3MG+OILtpi9/TaXbOjRg+PRLl4ERo2SOCZz6dWLLWpr1/L3VCpqgYFvOlbNm2dmdpggCDZl7lw2eNWvn/yyZctywoFZcWrNmwMdO3IjzmQ4fJhdrZbKberenSNrNm60zHiOiihqqSB37tzYtm0b8uQpgAcPWmDjxuv2Fsnx6NKFrTqrVrEP7fvvOaC8UiU2z1SrxjFn9+6xf61mTckGTC1K8VW6fHmuZ1eypNlD/fcfu/U7deJGDBERbLATBMFxOX4cOHaMc7BMuZwqxe7Pf/81o+H5H3/w05wJGzpyhENmLdVRr2FDoEiRhM1h0huiqKWSwoULY9Om7QA+wpEjxewtjuNRqRLHJY0axf+oIUP4D/399+ziXL+e7deWaFguvCF7dr7qbtuWKsX3r784q+r99/mnbNmSf7oXLywoqyAIFmXuXI7f6t7d9HWaNWPrVFCQGRs04RoTE8PKoyXi03Q4O3OEx9atxut9pxdEUbMA1auXQYMGM7Fxowvu3buHsPhuvozOkCHcSXfoUHZ5BgVx1WpdbINgHd56iy2WqWDlStazvb35+8iR3B1swQILyCcIgsV5+hT4808O+01J3e3GjVnfslbnv7NnOcHBEvFphuhCca3VCAfgnDh7IoqahWjbFjh9OhYNGjRFixYt8FzaFbyhd2/gyhVg6tQ3d3zB4blzB9i/n61pOurUAerVA2bM4OQCQRAS59kzDhlIcTZlKliyBHj1yrQkAkPy5uXqOWbFqZmArvG7pRU1Ly9uK2WN7M/oaK4hXr488Pix5cc3FVHULETbtgDgjHr1JiM4OBjt27fHa3ur4YKQClatYrdnp05xp3/5JafYW/MJVhDSOjEx/JAzdSrQurVtWg0SsduzRg3zAvabNmWFyhqyHjnCyqA1Ov917861vU+fttyYR45wOPWoUeyYSHHsngURRc1ClCrFMTxXrrTGwoULsXPnTnTt2hWx8XsvCkIaYeVKbs1Zvnzc6e+8w9OnTuWyHfaAiNCzZ08s12qLQWYF1giCdSACBg7kENH+/blDQM+e1q+iv3s3cOFCyq1pOpo1Y4Vk925LSsXoCt1aI1esc2euO26JllLPn3Nf+Vq12Iq2fj0/tNqzuYtdFDWlVKhS6rRSKkQpFaSdlkcptUMpdVn7nttg+ZFKqStKqYtKqeb2kNkU2rVjV1GLFt0xa9YsrF69GhMnTrS3WIKQYm7e5Eophm5PHUqxVe38eb6I2YN169Zh8eLFuHv3LlatWoVq1aph/PjxIKmRIzgAM2cC8+fz/+Snn4Bp04DVq/nhxprMnctd7Iz9b02hdm1OQrB0nFp4OF8vLJlIYEj+/NzLdOlSjlczl82bOTrnhx9YwT53DmjTxnJymg0R2fwFIBRAvnjTpgH4Uvv5SwBTtZ+9AJwE4AqgJICrAJyT24a/vz/ZmuPHiQCiRYv4+7x58+jhw4c2l0MQUsuMGXwuX75sfH50NFGpUkTVqxNpNLaV7fnz51S0aFHy8fGhqKgoioyMpO7duxMA+vDDDykyMtK2AlmJmBiiZs2IvvrK3pIIKeGvv/i/8/77RLGxPE2jIercmcjJiWjbNuts9+5dIhcXomHDUjfO228TlS9vGZl07NjBx8Ra+05EtHo1b2Pr1pSv+99/RJ068fpeXkQHDlhePlMAEETGdCZjE639SkRRuwigsPZzYQAXtZ9HAhhpsNw2ALWS24Y9FDWNhqhYMaK2beNOj4yMpI0bN9pcHkEwl2rViPz8kl5m3jy+guzcaRuZdHz++ecEgA4YXE01Gg1NmDCBAFDdunUpLCzMtkJZgQUL+PgCRAcP2lsawRQOHybKkoWoVi2ily/jzouIIPLxIcqdm+jaNctv+9tv+Vy5dCl148yaxePcvGkZuYjeyPbkieXGjE9kJB/bDz4wfR2Nhv9nuXMTZc5MNH480evX1pMxORxNUbsO4ASA4wD6aqc9jbfME+37jwA+NJi+AECHRMbtCyAIQFDx4sWtcBiTZ+BAIjc3ohcv3kybOHEiAaDff//dLjIJQkq4do2vDFOnJr3cq1dEhQoRNWliG7mIiK5fv07Ozs7Uu3dvo/NXrFhBbm5uaf7B6MULorfeIqpalahoUaJKlYiiouwtlZAU164R5c/PluYHD4wvc+UKUa5cRL6+ce8RqSU6ms+Tpk1TP9bp0/z/X7Ag9WPpaNWKqEIFy42XGP368f03PDz5ZS9dImrYkPe1Xj2i8+etL19yOJqi9pb2vYDWrVk/CUXtJyOK2nvJbcMeFjUion/+4aO6bt2baZGRkdS4cWNydnamDRs22EUuQTCVKVP4HDblqX/qVF722DHry6Vj/fr1SYYU/Pfff/rP9+/ft4VIFmfCBD6u+/YRrVnDn2fMsLdUQmI8fsyKSO7cRBcuJL3sli1EShF17Wq5sIF16/gcWbMm9WNpNESFC7Mr0BJoNET58hH16GGZ8ZLi0CE+DgsXJr5MVBTRxIlErq5EOXMSzZ//xkVtbxxKUYsjADAWwOfpwfVJxCdBrlwJT8pnz55R1apVKUuWLLR37167yCYIpuDnx7FnphAezuf7e+9ZVyYiolevXqVo+T179lDWrFlpYVJXbQfk/n2iHDmI2rXj7xoNWySyZSO6ccO+sgkJef2aLTOZMhHt2mXaOjpX4OzZlpGheXOiIkXYsmYJunVj5coSCszVq7yvc+emfqzk0GiIypYlCggwPv/wYbZOA0QdOnBcnyPhMIoagGwAchh8PgjgbQDT4yUTTNN+9o6XTHDNUZMJdHTtSpQ3b8I/zYMHD6h8+fLk4eFBr+3pCBeERLh8OeXWm1Gj2EJgTdfBw4cPqUiRIikKH3jy5Ak1adKEANDIkSMp1lEem5NhwAAiZ+e4lpnr19mlo1PeBMdAo+GHcoAoJZEtsbEcy+zsTLR7d+pkuHKFtz9unGnL3717l4YNG0ZDhw6lmJgYo8v88QePefx46mQjIlq2jMcKDk79WKagU4JDQ99Me/aMaPBgvk4VKUK0fr1tZEkpjqSoldIqXicBnAUwSjs9L4CdAC5r3/MYrDNKm+15EcA7pmzHnoraypV8ZPfsSTjvxo0bdPLkSdsLJQgmMHEin7spsdzcv88B1L16WU+ujz/+mFxcXOj06dMpWi8qKor69u1LAKhDhw70Mn6Et4Nx8SJn7vXrl3CeziXtqDeZjIhOKRgzJuXrhodzdmWBAkS3bpkvw/DhrPDduWN8/t27d2nBggW0ZcsWIuJwAFdXVwJAX3zxhdF17t3j/ZoyxXy5dAweTJQ1q+Wsfclx/TrL/u23/H3TJk7yU4ofgkyJX7MXDqOo2eplT0Xt2TPOIPnss8SX0Wg09NNPP9FdR7O9ChkaHx/OWEspAwey6yc1N5zE2L9/PwGg4cOHm7W+RqOhGTNmkFKK5s+fb2HpLEv79kTZs3O5gPhERRF5exMVL84ZhIJ90VmKPvzQ/Fizc+f4965enbMWU8qrV+y9iR96cOTIEfrmm2/Iz8+PABAA+sAgHfLly5fUr18/AkB//PGH0bErVSJq1CjlMsWnenUO1rclAQFEpUs7RsmNlCCKmo1p0YKzfxL7A9+4cYOyZctGPj4+9MSaOcuCYCLnz5PZcTOhofxUP2SIZWWKioqiSpUqUbFixej58+epGuvIkSOk0f4ho6yQQvnq1SvatWsXffPNN/Tpp5/SlClTaOXKlRRhola1fz8f//HjE19m3z5exkydVbAQe/fyw3j9+uYpWIbokkUSSWROkiVLSJu89oR2GQTI1axZk5ycnKh27do0ceJECgkJ0Z/7OqKioiggIIBcXV3p1KlTCcYeOpT3MTXZqZGRPIatz9eFC/m4OELJjZQgipqNmT+fj66R81/P9u3bKVOmTFS3bl16YclcbUEwg3Hj2D1w+7bx+REREXEyKuPTrRu7OCxZ4/nff/8lJycnWrt2bYJ5kZFEgYEJ61Ulx+XLl6lUqVK01ZzKmAZER0fT4cOH6fHjx0RENGfOHAJATk5OlCdPHr0l44G2VsPUqVOpUqVK1LZtWxoyZAh9//33tGHDBoqJiSGNhi2ZhQsnby37+GNWipO6tliThw/ZJfbLL6yw3L9v+6LH9uTSJaI8eYjKlSN69MgyY371Fd8vfvnFtOU1Gg2dOXOGihefSm5u9cnZ2ZkyZ86sf5g5c+aMScXWw8LCaPTo0UYfXLZuJbMLyOo4fJjH+Osv88cwh1evOHPaEUpupARR1GzMvXt800vq6ZiIaOXKlaSUoiJFitCyZctsI5wgGMHbO6GLIjQ0lPr160eRkZHUqlUrqlChAt27d8/o+mfOkNnxOklx+fLlBNaA16+JWrfm7TVtyhdmU7l58yb5+PiQs7MzzU1BKlpsbCyFhITQrFmzqFWrVpQjRw4CQEuXLiUiolu3btGGDRvo6dOnRMSZ3idPntTLvnTpUmrdujV5e3tT1qxZCQBlyZKFNBoNrVpFBAyjkiWrU6dOnWjkyJH0yy+/xLGS6Hj4kN1dtWvbvqzAixdENWuSvhCv7pU7N0/v0YNo8mSitWvZrZdWLBmmEhZGVKYMZ0ReuWK5cWNiOHMzc2ZWbowRFRWlD/6fPXu2/kHgrbd86auvvqIDBw4kmhxgCg8ePIiTWf3iBcuTmk4Hs2fz+ZHYw58QF1HU7ECtWkSmiLFt2zZq3bo1bdq0iYiIrl69St99912S1gtBsCQ6JevHH99Mu3v3LpUuXZpy5cpF58+fp71791K2bNnIy8sr0fpkbdvyTTuVXkqtTGeMTo+O5tR6gKuQK8Vtb1Lignr27Bm1bNmSANBnn31m9Aan0Wjo4sWLdPbsWSJipVV3cyxXrhx9+umntHLlSnpkhllFo9HQ/fv3KSQkhF6/5pt/wYIzqXHjJlS6dGlycXEhAFSmTBn9Oh06dCBfX19q1aoVNWz4KQETacCAN5kF0VaO1o6O5jIhShGtWsV19v7+m+i774g+/ZTjggoXjqvAOTtzuYTWrdn99dtv7OJNi40jXr0iqlOH629ZI97p0SMOlylSJG6M4pMnT2jatGlUtGhRvWX58uXLVK/efMqS5RZpDbqp4vnz5+Th4UHdunWL81DUqBHHqplLly68P4JpiKJmB3RZWiltxfHjjz8SAHJ2dqYWLVrQihUrHD5bTUjbjB7NfQh1xrIHDx6Ql5cXZcuWjQ4dOqRfbvfu3eTm5kbe3t56l54huoKTM2emTp5NmzYRgAQFomNiuPyN4TZ+/ZW/t2yZMmUtJiaGBg8eTABo+vTpRMTWtsWLF1O3bt2oSJEiBIA6GVT+DAwMpJuW7K1DRHPmsPybN8eVLTQ0lEJCQvTTxo0bR61atSJfX1/KmzcvASAXl4b6KvgVKlSgfPnyUeXKlal169bUr1+/OFb6mzdvJrBMmopGwzFUANHPPye97NOnREePcvzUqFEc6F6xIltnDJU4nVWwVy8unLx+PZckccQODLGxrHQARCtWWG87ISFchoUr5V+hwYMHU7Zs2QgANWrUiPbv309EnLmYLRtRz56W2/b48eMJAM0wqM2ju4clYkRPlpIlbVNjMb0gipoduHCBElgpTOXcuXM0cuRIKlq0KAGg/Pnzp5tm04JjodFwmYCGDfn7kydPqEqVKpQlSxajrredO3dSlixZqGki/WoCArj9kbmn64sXL8jDw4M8PT3j1BuMjeX4LIDLiBii6zvatm3K3W3Lly/Xx4j6+PgQAMqXLx+9//77NG/ePLpiSR9XPJ4+ZYWlUaOUx3kFBb0gF5f/6KOP+PuMGTPo008/pZYtW5KPjw/lyZOHunXrRkRswcuaNSt5enrS999/r4+rM5XRo/n4jhqVMhkNiYnh4qebN7OS3bcvB+MXLBhXgXNx4fOxbVuiL77gwPADBywXD2YOX3/Nsk2aZP1tcTaphnLmLEOZMmWi7t27U3C8ImQ//sjyHD1que1qNBrq0KEDOTk56Ut5HD/O20kkMTRJ7t/ndadNs5yM6R1R1OxE+fKp64UYGxtLO3fupB9++EE/rVOnTvT111/TxYsXLSChkNE5eZLiVA4PCQmhggUL6i/Wxti5cyedO3fO6Lxt23i8334zT56vvvqKANBug0qgGg1R//487jffGF9Pd/N6913zrTK7d++mkydP2qw47siRLHNQkHnr64LQE6uIr3PpxsTE0E8//UTVq1cnAOTm5kY9e/Y06RqiU4J79rRe0sCTJxybtXgxH5P27bmkQqZMcZW4/PmJ6tZl69706UQbN3JwvzW9vroMwo8/tt7+R0VF0fLly+ntt9+mly9f0pAhRMABmjMnYXE0jYbjSa1xi4uIiCBfX19yd3en8+fPU2wsx+N1757ysTZs4OMmjXhMRxQ1OzFiBD8hWqoCx6tXr6h58+bk5OREAKhGjRr0008/mZThIwjG+OorndvzTZyWqSUlNBoNzZkzJ06clkZDVKUKxyalNLb53LlzeiuC4XiffUb6shRJ3Sx1wcsdO9quwKa53LzJhYK7djV/jBcv2L1UvrzpFszjx49T375947i1//vvP6PlT9at43OjRQv7uCSjo7lbxsaN3C2jd292C+bPH1eBy5SJyNOTFbwvv2SF79Ch1F93d+7k63fjxtbZ/ydPntD06dOpWLFi+tjHc+fOUVQUUYMGfH6cOBF3nb17eZ8t2TTdkNDQUOrYsSOFaQMJO3Xi2MOUKqmjRnGMohQ0MB1R1OyELmbH0gmdd+/epRkzZlClSpUIgD57LSIigs6ePUt37tyhly9fmh2TImQMNBouDNmo0Wtq0aIFTZgwIUXrnz9/njJnzkz+/v5x3Gm67hyrVqVMntWrV1Px4sX1yQoazRur06BBpt0sZs7k5Tt3dmxl7aOPOG7LsNWNOWzezPubwp+Onj17pr8+9OnTh3LkyEH9+/fX19Q6cIAVherVHbPA7qNHRAcPssVrxAh2lVaowIqVoRJXsCC7WPv25XNj82bO2DR8iIiOjqYHDx7QxYsX6fDhw3T9+nU6e5bI3V1Dnp4aiz1oG3Lt2jXKnj07AaCGDRvSxo0b41hy798nKlqUyMMjbsmbLl24v64tFKDXr1/TL79EE0CUwqYg1LgxP7AJpiOKmp2IjSUqVIif8K1FSEiIviTA3Llz9ZlpAChz5sxUsGBBunbtGhERrVmzhrp06UL9+vWjr776iqZNm0a//vqrPlnh3r17dPXqVXr8+HGqUr2FtAHHoESTv38HAmBW5f7NmzdTpkyZqFq1avrizTExbFHz80v5k7hhTafx4/kq1bdvysaZOpXX+/DDlFv1bEFICGdPfv65Zcbr0IGVKnPD6Q4dOkTdunXTtxaqUqU2Zcu2ksqWJTKSM+KQxMbG0pMnT+jixWu0Z88tWr+eaOpUDdWuPY9KlJhMWbJ8QUBvAt4jYD5lzkxUocJzcnHJEeeayZnAX5OHB1H+/A8oSxY38vT0pBYtWtCAAQNoxowZ+ozklDwIazQa2rdvH/2mjQnQaDT09ddf04n4JjMDjhxhZb5pUz6P799n6+H//peqQ2USr169onr16lHfvsMIIJo1y/R1Y2KIcuTgbGDBdERRsyN9+3KbEFvkApw8eZJWrFhB8+bNo8mTJ9MXX3xBffv21d9A586dS2XKlKF8+fKRs7Oz/sIUrm2A9vnnn+unZc2alUaPHm2yG0xIewwfHktKdScANCslV+J4bNiwgTJlykQ1atTQn0u6bMxt25Jf//Hjx7Rs2bI4Nz6dsvXRR+bVC9P1LTV3fWvSrBmXMbFEaQUirlOVPTuXKUmNEf3hw4c0evRMcnEpS25u/enqVVYorl69ahlBk0Gj0dCLFy/ihHJs27aNFixYQDNmzKBRo0ZRv3794pyr1atXpzx58ujDQeK3S9JlTbq6ulKBAoXIw8OL3ntvMg0fTtS6tYZy5RpCSo0jYA4BfxCwmTJnvkpubkRbt96noUOHUvv27aly5crk7u5OAPTKVlBQEOXOnZv8/Pzovffeo88//5x++uknumHQLFcXf1atWjUCQMWKFUtRZ4zffuPzeMQIrlEH2K6Q68CBAwkAFSr0O73zjunr6cr9LFpkNdHSJaKo2ZEtW/hIJxGbbRc0Gg09e/YsTtr+iRMnaNGiRfTdd99Rhw5sZfH390+XLtTvv/+e5syZo1csMhoaDVGOHP0JAI1PrjKzCaxbt47c3Nxo+/btRMQPJm+99SabNCn69etHTk5OdF57B/r+e9K7L1NjERs3jvSB4I6irOmSLVKhFxtFF5+3cqX5Yzx5wnWzsmXT0IED/IC2d+9eAkCNGzemVatWmaRkGNZ0O336NG3atImWLFlC33//PY0dO5amTp2qn9+7d2/y9PSkggULUubMmQkA1axZUz9fF96h6/qQN29e+vDDD/XzhwwZQv3796evv/6aZs6cSQsXLqSDBw/q59+7dy/Z8kavX7Pys3Ytl6To25fo338TLqfRaOjRo0f07NkzIiK6ePEi9evXj95++20qX7683iK5Tft0smHDBr2iWLZsWfr555/NevD95BP+bd3dTfs/WYqoqChq2LAhOTllJlfXQyYbGxYssK1CaQkiIiLo8uXLtHfvXvr777/10ydOnEhHLZlemwSiqNmRyEh+2u3b196SpJx9+/bR+vVcVDMqKoqOHz9uZ4nMJyYmhtasWaMPmtZV986ePTsNGDBAryRkFI4cIQIWUsuWIy2miBsWadZoNPp4scSqrbMcR0gpRf/T+nN0WYbt21smgPubb0jvPrW3shYTQ+TrywkAlrawR0dzTFDhwlz2I6VERnJpFRcXIq2uTURE9+/fpwkTJlDx4sW11pVCNNygeeOMGTOoQYMG5OPjQ0WLFqVs2bJRwYIF9fPbtGmTwLVYsWJF/fwRI0ZQhw4dqE+fPjRixAh9j1Qdly5douvXr1N4eLjDPzDGxsbq44OJiA4fPkz9+vWjDRs2pCqTODLyTUeI1Cji5vDw4UMqWLAkAYUoMNC0FgN9+xLlzGn//xsRF/M9d+4c7dy5k5YuXUrTp0+nkSNH6ucPGTJE32VE98qbN69+fv/+/RPUc7QWoqjZmY4dOVbNEU5cc9HFv3Xv3p3u3EmYNu6oPHr0iKZNm0YlSpSI47YgIjp27Bh1795d/yT/aQYJqrh+/ToNG8bxLpZyvxny119/UePGjenu3eeUOzcHehsjJiaG/Pz86K233qLw8HBavJivSi1bWq79kGFCQv/+9u1Lqdu/5cutM/7Roxz7NmhQytaLjSV6/32WTdsRKwExMTG0adMmatWqFTk7O+tjWMePH0/16tWjNm3aUI8ePeizzz6jyZMn69c7deoUHT58mC5dukRhYWFW76CQXvnvP+4Fao/Dd/jwaQJ8qU8f4yV54uPjw+79lPL06VO6fPkynTx5Uq/sXrlyhVasWEELFiygH374gaZOnUpjxozRJy+tX7+eOnbsSC1btqSGDRtS9erVqWLFivpwn2HDhiV4UMiaNau+XdbixYtp8ODBNHnyZFq8eDFt27bNaJN6WyCKmp1ZupSPtkGR9zRHeHg4jRgxgjJnzkzZsmWjiRMnxukN52i8fv2aPvnkE3JzcyMAFBAQQGvWrDF6o3jw4AFNnDiR/tBWdoyIiKCZM2ea1R7I0Zk+fTq5urpSwYInqFUr62wjMDCQnJycqEGDBvTllxEEEGk7McVB18g8MDCQli/nUhBNmqSsd6cpaDRc2gMgGjzYPsray5ecxVetmnUf2AYM4ON47Jhpy2s0fEwArk1mCpJolPGoUyeW/P3ZUp6UZfP5cz7/Eqt3GJ/Hjx/TL7/8Qk2aNIkTN63z3sybNy+BogWALly4QEREv/76K1WoUIH8/Pyobt261Lx5c2rfvr3+2n3s2DFatmwZ/fvvv3T+/HmHtsyKomZnHj9ml8KIEfaWJPVcuXKF2rdvTwCodevW9hYnDjExMXGqeDdq1Ig+/vhjOnnypH7a69dEgYFE77zDFeG7duWb+Hff8fS9e4nmzl1LusKgvXv3jtPKx9rMmTOH9uzZE6cqv6X46aeftPFGnQiIoSVLLL4JPX/++Sc5OTlR3boNyc3thdGimZs2baJevXrR6tUacnbmMgrWKjtgWI9t6FDbK2u6QHCDOr5W4elTtt77+5sW36dL2hgyxL7WRsGx4QzsaOrevW+cGMP47NrF55NhS7T4PHr0iO7evUtEHF6j62s7cuRI+uOPP2j16tV6i9nDhw/p7NmzFBoaSg8ePKCIiAibFaS2NaKoOQBNmnCdn/TCzp076YC2O/GTJ08StDmxJU+ePKFZs2ZRqVKlyNXVVd+H0vAPfesWP+UVKsRnvocH9xr08OBGy4a1l/h1kjJn7kNKsUWuQIF69L//hdPs2Rwnsn8/t8QxjFPWaNga9OgRb+/iRaLgYK5JtWMH9zP8808NTZlylXr2DKQGDYaTh0cAlSjRhAYM4OrvhQvX0pvn33nnHZo5c2aiXQBSwuLFi/XK9cCBUeTqyj0Drckff/xBSikqVqwxOTtHGq0ZtmkTu2Br1SLSxmhbDY2G3YIAtyeylWLy4AEHgtvquWb5ct5Hg4YmRlmyhJfr1Clth2UI1odrgmqoVq1OpJSiTZs2GV1O1x9UWy9Xz5MnT2jx4sXUokULcnFxoUFa/3xsbCydOHHCYa1ctkQUNQfghx/4iGsttumKkSNHklKK+vTpoy9Wagtu3rxJ/fv312dW1a1bl1auXKnPTIuN5cDodu3YHK8UUatW/LRnaG3QaFi5On2as/IWL+a+fgMHErVu/YhKlJhOWbN2NGgsvZGA+3qlzt2dmyQrFV/Z0xBwk4BNBtM+NDDhZyagOjk7f0Z58xIVK0ak1BPy8VlH/foNpPLlyxMA6tevHxGxxXDx4sUpbgx++PBhcnJyoqZNm9KLF6/orbcSjxuzNL///jt98snn5OKi0cdObd++ncaOHUubNkWSqytR1armBcCbg0ZD1K8f/xZffWUbZW3QID7/LKBvm4RGw7W3cuQgSiycdNs2tvI3bGib0kFC2iY6mhMEevR4QX5+fpQjRw6jD5Dt2hGVKRN3mmEccIkSJWj48OF2fbB3VERRcwBu3uQjnoTVOM3y+PFjGjJkCLm4uJC7uztNnz7dKq47In4C08UfXLp0Sd+30LBw5KNHXIW8bFk+5vnycWsZbd1fs9FoiK5ff0ZZsrhRpkyZqVatD+nTT4/Q4MFEw4axxW7w4JP09ttjqFKllpQzZ0G9UrZhwwM6fpxo/vwtNHnyfPrnn+P05MnrBJaM339nmd97j5XJmzdvUqjWFBUUFKQfr3z58jRgwABau3ZtsiVGYmNjadasWRQREaFvQfPnn6k7FimlZ08iV9drdOXKUypTpgwVLVqWsmR5RT4+tm+4HRvLmWkA0Zgx1t3W5cusENk66/vSJbYUv/9+wnlBQfxg4eNjOwVZSPu0b09UvDjRjRs3qUCBAlSmTJk4HUk0GqICBZ5R7drLqG/fvnor2VdffUVDhw6lw4cPi+UsCURRcxD8/dnFk145f/48tWjRggDQJ598YtYYr1690rsuidjFOn/+fJo0aRINHTqUypQpQ+3atdPPN1RSjh5lhSBLFj6769Th9l2WthicP3+eBg4cqG8BU6FCBX15jwULFpCTkxN5e3vTRx99RD/++CMdPnw4RUUuZ81i+Xv3jmvx0Wg0dOrUKZo1axa1aNFCb0ncunUrERFdvXqVdu/eTZHaHd69e7deydMxcCAfH2u7GeMTFPSMgMKUPXthAkBZsuwgLy/7Vb6PjSXq1YuPswXKyCVKhw6sFN27Z71tJIaujpxBWSi6coWoQAGiEiUSt7YJgjHmzuXz6eJFov3791OePHlo79699Pz5c1qxYgW9/fa7BGQhAFSkSBF9HJpgGqKoOQjjx7N7zB4XbVvy999/06VLl4iIm/xu376dNm7cSIsXL6aZM2fGSd8fNWoU+fr6UrFixShr1qwEgDw8PPTzmzZtqrciubm5UZ06deLUWXrxgvv9Va3KZ3S2bFwg0hbx/+Hh4TRnzhyqV68e7dZGiUdERBhtcJ1SRo3i/fnyy8SXef36Ne3Zs4deaCPwR48erY9va968OWXNmpVatGihXz4mhmP03nsv1eKZReXKnMHl4tKZypQhsvd1PDaWOxcARKNHWz5m7+BBHnvsWMuOayqRkUTlyhGVKsWxlPfvs1sqT570GYIhWJcrV/h8/vFH/q4r/Ltq1SoCQLlzFyZgEP366/50G/BvTURRcxBOneKj/ssv9pbEdnTt2tVoHRudCXzixInUunVr6tGjBw0dOpQmTpxIiwx6j9y4cYNu3bqVoLr4xYucxZcrFx9TLy++gKSXRgMazZuK5KaWTXj69CmtX7+eBg7k+LaaNWvSPYOnAl1GVmCgdWROjqNHiYCTVLz4K0phmJ3ViInhnqAAJzU0bswWTe1zhtloNGzRLVSISxbYi507ed8++4xLg7i5sQIpCOZQsmTC+NYXL17Qnj17aMiQGHJ1tVwNxIxGYoqa4nnpj6pVq1JQUJC9xUgAEVC6NODpCWzebG9pbMOjR49w/Phx5MqVC3nz5kXevHnh7u4OJyenFI8VEwNs3AjMnQvs2AG4uADvvgv07w/Urw8oZYUdsCOxscAHHwArVwILFwI9e6ZuvH79gCVLgAcPgGzZLCNjStmyBahcGXjrLfts3xhEwP79/J/ctAk4e5anlysHtGoFtGwJ1K0LZM5s+phr1/K5OX8+0LevdeQ2lW7dgKVLAScnYN06oHVr+8ojpF0++QRYvhx49AjIlCnuvLp1AY0GOHjQPrKldZRSx4moaoLpoqjZnqFDgZ9/BsLCgBw5LDfu06d8Mfb2Bho04ItyeuLUKaBrV+DMGaBoUb5g9O4NFCpkb8msS1QU31j/+QdYvRpo1868cWJiWDlq2BAIDLSoiOmO69dZadu8Gfj3X/4N3N2B5s1ZcXvnHSB//sTXj47m/6GLC5+3Li62k90Y9+/zOdS/P9Cjh31lEdI2q1cDHTrwg02dOm+mR0fzf+TTT4HvvrOffGmZxBS1dHYrTxu0awe8fg1s22aZ8WJjgV9+AcqWBQYNAho1AkqWBL7+Grh40TLbsCcaDf/xq1UDHj5k69L167x/6V1JA9iKs3o1UL060LkzsGuXeePs3s0PB506WVS8dEnJksDAgcDff7PlYN06Pm779wMffQQULAjUqgVMnAiEhLBFzpBffwUuXwamTrW/kgawvEePipImpJ5GjdgIsGNH3OmnTgGRkUDNmvaRKz0jipodqF0byJuXL/6pZe9eoGpVti55egKHDrFZ2ssLmDwZqFCB/zhz5wKPH6d+e7bm7l22YgwdylaMU6eAjh0d4+ZnS7JnZ+tOmTJA27bA8eMpHyMwkMd55x3Ly5eeyZ6dj/kvvwB37vCxHzuWHyC+/hqoUgUoXpwtCZs2sfVq7Fi2ardqZW/pBcGy5M7N95z4itqRI/xeo4btZUrviKJmB1xc2A2xeTObi83h5k1+wm/QgJ/4AwOBPXtYKevcmS0Bt28DM2YAL1+yy6NwYeC994ANG9iV4+isXQtUqsTxDvPn8/ek3E3pnTx52AqbNy/w9tsps5ZGRwNr1gBt2gBubtaTMb2jFODnB4wezTeme/c4drB6dWDZMv5fFy7Mlsvp09NfzKQgAECzZnz+h4e/mXb4MFtuS5Swn1zpFVHU7ES7dhxTtndvytZ7+RIYN44tZRs2AGPGABcuAO+/n/CmULgwMGwYcPIkEBwMDBjArpu2bYEiRYDBg9k64GhhihERQJ8+HIhdsiRw4gQHY8tNj3+37dvZ9dC0KXDrlmnr7dzJFlVxe1qWQoU4wWP1anbL79jB/6upU9lVLwjpkaZNOeTGMAzjyBG2psl12vKIomYnmjZly4ap7k8ijs3y9GS3SuvWbFEZOxbImjXpdZXiLLtZs9jKtmkTxxn88gubsCtW5BvL7dup2ydLcOwYu5IWLABGjmRrWvny9pbKsShbFti6lZ9mmzVjBSE5Vq58EwwvWAdXV6BJE2D2bOCLL+wtjSBYj5o1OWtc5/58/Bi4dEni06yFKGp2ImtWvsmuX5+8RSskBAgIYGtI7tzs4gwM5LiYlJIpE5caCAxkt838+Tzml1/yeM2aceboixfm7JX5xMZyYHbt2pxosWsXMGlSysohZCSqVOEyJaGhQIsWwPPniS8bFcVu43btWJkQBEFIDZkz8z1Jp6gdPcrvEp9mHURRsyNt27LrKjjY+PywMA5Q9vfnuk7z5rGrsn59y2w/d252Ke7fzxlq33zD7926sUunRw9WBl69ssz2EuPGDS4Z8fXXnPZ96hTH3glJU78+W8pOnADat2cF1xg7drCb/f33bSqeIAjpmGbN+H4RGsrxaUqJu99aiKJmR1q14lij9evjTo+OBr7/nott/vYbl9y4fJkzO52drSNLmTIc+3b1KlvsOnViK0ybNkC+fJyE8Mcfls8c/fNPwMeHrYZ//MHfc+Wy7DbSM61bA4sWcQzaBx+wZTI+gYF8TJs2tbl4giCkU3TXkx07OD7N29uydUGFN4iiZkfy5+dKzoZxatu3A76+wJAhnEl26hTHvOTObRuZnJzYUvPbb2zR27aN60YdPgx07w4UKAA0bgz88ANnnppLeDgXr+3alTM7T54EPvxQAlHNoVs3PkfWrGFl3tCVHhnJDwLt24sbWRAEy1GhwpvkpiNHJD7NmoiiZmfatmVlbMcO/ty8OccUrV/PAeNeXvaTLXNmNm///DO7aI8c4SDpe/c4s61ECXbLfvstcPq06dmj+/axMhoYCIwfz4VYS5a06q6ke/73P3Yd65IwdGzbBjx7JtmegiBYFqXexFk/eSLxadZEFDU707Ytvzdrxq1qpkzheLQ2bRzLuuTkxBa+SZOAc+e4JMjUqRycPno0uy/LlgU+/5xj3oy54KKjWZkICOBacgcOcFxcRiteay3Gj+denlOncg0vgGPY8ublLF9BEARL0rTpm1qgoqhZD+n16QB0787ZmBMmcO2ztMa9e5x0sG4dx0pFRbFbt00bzjRs3JhLf3TtyuU3evViV53EM1ie2Fg+zoGBwI8/cjZvly5cikUQBMGSPHjARW6zZ+eEJWvFUGcUpCm7YBOePWOX7bp13Hnh2TOut0PE1rdff+XEBMF6REWxkqzrJbtjB9f3EgRBsDQ1a3LXlC1b7C1J2kcUNcHmREVx/NnatdxRYdIkDj4VrM+LF+yWuHsXuHJF3MuCIFiHhw/ZkmarhLf0jChqgpDBiI3ldlw5c9pbEkEQBCE5ElPUJJlAENIpzs6ipAmCIKR1RFETBEEQBEFwUERREwRBEARBcFBEURMEQRAEQXBQ0oyippR6Wyl1USl1RSn1pb3lEQRBEARBsDZpQlFTSjkD+AnAOwC8AHRRStmxuZIgCIIgCIL1SROKGoDqAK4Q0TUiigKwAkBbO8skCIIgCIJgVdKKolYEwC2D77e10wRBEARBENItaUVRM9aePEGlXqVUX6VUkFIqKCwszAZiCYIgCIIgWI+0oqjdBlDM4HtRAHfjL0REvxBRVSKqmj9/fpsJJwiCIAiCYA3SiqJ2DEBZpVRJpVRmAJ0BbLCzTIIgCIIgCFYlzfT6VEq1ADAbgDOAhUQ0MZnlwwDcsIFojkg+AA/tLUQGRI67fZDjbnvkmNsHOe72wVbHvQQRJXAHphlFTTAdpVSQscaugnWR424f5LjbHjnm9kGOu32w93FPK65PQRAEQRCEDIcoaoIgCIIgCA6KKGrpk1/sLUAGRY67fZDjbnvkmNsHOe72wa7HXWLUBEEQBEEQHBSxqAmCIAiCIDgooqilAZRSC5VSD5RSZwym+SqlDimlTiulNiql3LXTMymlftdOP6+UGmmwzm6l1EWlVIj2VcAe+5NWSOFxz6yUWqSdflIpFWCwjr92+hWl1ByllLFOG4IWCx53Od9NRClVTCm1S3vNOKuU+p92eh6l1A6l1GXte26DdUZqz+mLSqnmBtPlfDcRCx93Od9NJKXHXSmVV7t8hFLqx3hjWf98JyJ5OfgLQH0AfgDOGEw7BqCB9nMvAN9qP38AYIX2c1YAoQA8tN93A6hq7/1JK68UHvcBABZpPxcAcByAk/b7UQC1wK3Q/gbwjr33zZFfFjzucr6bfswLA/DTfs4B4BIALwDTAHypnf4lgKnaz14ATgJwBVASwFUAztp5cr7b57jL+W69454NQF0AnwL4Md5YVj/fxaKWBiCivQAex5tcHsBe7ecdAN7TLQ4gm1LKBYAbgCgAz2whZ3ojhcfdC8BO7XoPADwFUFUpVRiAOxEdIv5XLwHQzrqSp20scdytL2X6gojuEdEJ7efnAM4DKAKgLYDftYv9jjfnblvwA+FrIroO4AqA6nK+pwxLHXebCp0OSOlxJ6IXRLQfQKThOLY630VRS7ucAdBG+7kj3vRC/QvACwD3ANwEMIOIDG96i7Rm8W/EJWEWiR33kwDaKqVclFIlAfhr5xUB96rVcVs7TUgZKT3uOuR8TyFKKQ8AVQAcAVCQiO4BfHMDWy0BPodvGaymO6/lfDeTVB53HXK+pxATj3ti2OR8F0Ut7dILwACl1HGw6TZKO706gFgAb4FN48OUUqW087oSUSUA9bSvbrYVOV2Q2HFfCP6TBoFbnR0EEAM2h8dHUq1TTkqPOyDne4pRSmUHsBrAECJKyhKf2Hkt57sZWOC4A3K+p5gUHPdEhzAyzeLnuyhqaRQiukBEzYjIH8BycKwCwDFqW4koWusK+n979xJaRxXHcfz7U6N9JFpqXahB3ajQplghQhMFH+hGIhYMWCjaQFGXPqoiqOBCwU3rq0hBLdlYK5WAKS6KLkSoYgUfKammVYjWtlRCUaMVauvfxTnXjqU3zcQ0zr35fWDI3JnMf04Oh5s/58yZs4M8FBQR+/PPcWAz7jIvrV69R8SxiHg4IpZFxJ3AAmAvKYloL4RoBw7McLEb3hTq3e29JEktpH9ab0bEQD58KA/v1IZ5fsrHf+TfPZe1du32XtI01bvbe0kl672eGWnvTtQaVG1Gj6SzgKeAjfnUD8AtSuYDy4Fv8tDQonxNC9BDGk6yEurVu6R5ub6RdBtwLCJ25+7zcUnL81DEvcC7/0/pG1fZend7Lye3zTeAryNifeHUILA676/mRNsdBFZKOi8POV8J7HR7L2e66t3tvZwp1PspzVh7P9OzK7xNywyVt0jPnP1JyuDXAA+SZqrsAZ7nxMuLW4GtwDCwG3gsH59PmhE3lM+9RJ4t5G1a6v0KYIT0UOoHwOWFOJ2kL83vgA21a7yduXp3ey9d5zeQhmyGgC/zdjtwIWmyxt78c2Hhmidzmx6hMNPN7X3m693tfUbqfZQ0yem3/L20OB8/4+3dKxOYmZmZVZSHPs3MzMwqyomamZmZWUU5UTMzMzOrKCdqZmZmZhXlRM3MzMysopyomVnDk/SCpIcKn7dLer3weZ2kRya4vk/SJf/h/ksk7ZE0t3DsPUkrpxrTzAycqJlZc/gY6IZ/Xoq7CFhSON9NWqWjnj7SsmuTJumc2n5EDAMDpHdcIWkF0BIRW8rErBffzGYvv0fNzBpe7g3bGRHtkpYCjwIXA3cDR4BDpAWWnwDuAOaSkrsHgLuAfmA/8AfQBSwG1pNeID0G9EXEQUkf5uuuBwYjYl2hDPOAL/I9t+T7nF8nzn3A/cC5wLfAPRFxRFI/6aWa1wKfR8Ta6a4rM2ss7lEzs4YXEQeAY5IuI/WefQJ8Skq6OoGhiDgKbIiI6yKig5Ss9UTEO6RF3VdFxDLSou6vAL2R1hbdBDxXuN2CiLixmKTlMhwhJYgfkRK10QniDORyXENaVWFNIdRVwK1O0swMwF3rZtYsdpCStG5SL9alef8XUi8YwM2SHgfmAQtJy+1sOynO1UAH8H5avo+zSUta1bxdrwARsU3Sz8Crp4nTIelZ0iLyrcD2QpitEXF8kn+zmTU5J2pm1ixqz6ktJa29tw9YC/wKbJI0h5RAdUbEPknPAHNOEUfAcER01bnP76cpx195myhOP7AiIr6S1AfcVCK+mc0iHvo0s2axA+gBDkfE8Yg4TOqx6iINhdaSsjFJrUBv4dpxoC3vjwAXSeoCkNQiqTgxYbImitMGHJTUAqyaQmwzmyXco2ZmzWIXabbn5pOOtUbEGICk1/KxUeCzwu/1Axsl1SYT9AIvS7qA9D35ImmYdNIi4qikenGeJj1D930uT1u9OGY2u3nWp5mZmVlFeejTzMzMrKKcqJmZmZlVlBM1MzMzs4pyomZmZmZWUU7UzMzMzCrKiZqZmZlZRTlRMzMzM6soJ2pmZmZmFfU3yiNg75nMCRcAAAAASUVORK5CYII=\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "fig, ax = plt.subplots(figsize=(10,5))\n", "\n", "data.plot(x='years', y='SLI_max', c='r', ax=ax, label='Slide Canyon Observed')\n", "data.plot(x='years', y='BLC_max', c='b', ax=ax, label='Blue Canyon Observed')\n", "\n", "# Plot the predicted SWE at Blue Canyon\n", "ax.plot(data.years, BLC_pred, c='k', linestyle='--', label='Blue Canyon Predicted with Linear Regression')\n", "\n", "ax.set_title('Timeline of Peak Snow Water Equivalent (SWE)')\n", "ax.set_xlabel('Water Year')\n", "ax.set_ylabel('Peak SWE (mm)');\n", "plt.legend(loc=\"best\")" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "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": "3.8.8" } }, "nbformat": 4, "nbformat_minor": 4 }