diff --git a/files/dl/algo/prog_dyn/cours/prog_dyn.ipynb b/files/dl/algo/prog_dyn/cours/prog_dyn.ipynb
index e8f7f94e..ef3f9147 100644
--- a/files/dl/algo/prog_dyn/cours/prog_dyn.ipynb
+++ b/files/dl/algo/prog_dyn/cours/prog_dyn.ipynb
@@ -1 +1 @@
-{"cells": [{"cell_type": "markdown", "metadata": {}, "source": ["# Programmation dynamique\n\n", ""]}, {"cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": ["def knapsack(c, w, v):\n", " \"\"\" \n", " Renvoie la valeur maximum que l'on peut mettre\n", " dans un sac \u00e0 dos de capacit\u00e9 c.\n", " Le i\u00e8me objet a pour poids w[i] et valeur v[i].\n", " \"\"\"\n", " n = len(w) # nombre d'objets\n", " dp = [[0]*(n + 1) for i in range(c + 1)]\n", " # dp[i][j] = valeur max dans un sac de capacit\u00e9 i\n", " # o\u00f9 j est le nombre d'objets autoris\u00e9s\n", " for i in range(1, c + 1):\n", " for j in range(1, n + 1):\n", " if w[j - 1] <= i:\n", " x = v[j - 1] + dp[i - w[j - 1]][j - 1]\n", " dp[i][j] = max(dp[i][j - 1], x)\n", " else:\n", " dp[i][j] = dp[i][j - 1]\n", " return dp[c][n]"]}, {"cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [{"data": {"text/plain": ["19"]}, "execution_count": 6, "metadata": {}, "output_type": "execute_result"}], "source": ["w = [2, 2, 2, 3, 5, 6, 8]\n", "v = [1, 1, 1, 7, 10, 10, 18]\n", "knapsack(10, w, v)"]}, {"cell_type": "code", "execution_count": 38, "metadata": {}, "outputs": [{"data": {"text/plain": ["19"]}, "execution_count": 38, "metadata": {}, "output_type": "execute_result"}], "source": ["def knapsack2(c, w, v):\n", " n = len(w)\n", " dp = [0]*(c + 1)\n", " for j in range(n):\n", " dp_ = dp[:] # copie de dp\n", " for i in range(c + 1):\n", " if w[j] <= i:\n", " dp[i] = max(dp_[i], v[j] + dp_[i - w[j]])\n", " return dp[-1]\n", "\n", "knapsack2(10, w, v)"]}, {"cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [{"ename": "NameError", "evalue": "name 'w' is not defined", "output_type": "error", "traceback": ["\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", "\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)", "Cell \u001b[0;32mIn [2], line 13\u001b[0m\n\u001b[1;32m 10\u001b[0m \u001b[39mreturn\u001b[39;00m dp[(i, j)]\n\u001b[1;32m 11\u001b[0m \u001b[39mreturn\u001b[39;00m aux(c, \u001b[39mlen\u001b[39m(w))\n\u001b[0;32m---> 13\u001b[0m knapsack_memo(\u001b[39m10\u001b[39m, w, v)\n", "\u001b[0;31mNameError\u001b[0m: name 'w' is not defined"]}], "source": ["def knapsack_memo(c, w, v):\n", " dp = {}\n", " def aux(i, j):\n", " if i == 0 or j == 0:\n", " return 0\n", " if (i, j) not in dp:\n", " dp[(i, j)] = aux(i, j - 1)\n", " if w[j - 1] <= i:\n", " dp[(i, j)] = max(dp[(i, j)], v[j - 1] + aux(i - w[j - 1], j - 1))\n", " return dp[(i, j)]\n", " return aux(c, len(w))\n", "\n", "knapsack_memo(10, w, v)"]}, {"cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [{"ename": "NameError", "evalue": "name 'w' is not defined", "output_type": "error", "traceback": ["\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", "\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)", "Cell \u001b[0;32mIn [1], line 14\u001b[0m\n\u001b[1;32m 11\u001b[0m \u001b[39mreturn\u001b[39;00m v\n\u001b[1;32m 12\u001b[0m \u001b[39mreturn\u001b[39;00m aux(c, \u001b[39mlen\u001b[39m(w))\n\u001b[0;32m---> 14\u001b[0m knapsack_memo2(\u001b[39m10\u001b[39m, w, v)\n", "\u001b[0;31mNameError\u001b[0m: name 'w' is not defined"]}], "source": ["from functools import cache\n", "\n", "def knapsack_memo2(c, w, v):\n", " @cache\n", " def aux(i, j):\n", " if i == 0 or j == 0:\n", " return 0\n", " v = aux(i, j - 1)\n", " if w[j - 1] <= i:\n", " v = max(v, v[j - 1] + aux(i - w[j - 1], j - 1))\n", " return v\n", " return aux(c, len(w))\n", "\n", "knapsack_memo2(10, w, v)"]}, {"cell_type": "markdown", "metadata": {}, "source": ["## Plus courts chemins\n", "\n", "$$M = \n", "\\begin{pmatrix}\n", "0 & 3 & 8 & \\infty & -4\\\\\n", "\\infty & 0 & \\infty & 1 & 7\\\\\n", "\\infty & 4 & 0 & \\infty & \\infty\\\\\n", "2 & \\infty & -5 & 0 & \\infty\\\\\n", "\\infty & \\infty & \\infty & 6 & 0\\\\\n", "\\end{pmatrix} \n", "$$"]}, {"cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [{"data": {"text/plain": ["[[0, 3, 8, inf, -4],\n", " [inf, 0, inf, 1, 7],\n", " [inf, 4, 0, inf, inf],\n", " [2, inf, -5, 0, inf],\n", " [inf, inf, inf, 6, 0]]"]}, "execution_count": 8, "metadata": {}, "output_type": "execute_result"}], "source": ["g = [\n", " [0, 3, 8, float(\"inf\"), -4],\n", " [float(\"inf\"), 0, float(\"inf\"), 1, 7],\n", " [float(\"inf\"), 4, 0, float(\"inf\"), float(\"inf\")],\n", " [2, float(\"inf\"), -5, 0, float(\"inf\")],\n", " [float(\"inf\"), float(\"inf\"), float(\"inf\"), 6, 0]\n", "]\n", "\n", "g"]}, {"cell_type": "markdown", "metadata": {}, "source": ["\n", "### Bellman-Ford"]}, {"cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [], "source": ["def bellman(g, r):\n", " n = len(g)\n", " dist = [float(\"inf\") for i in range(n)]\n", " dist[r] = 0\n", " for k in range(n - 2):\n", " for u in range(n):\n", " for v in range(len(g[u])):\n", " if g[u][v] != 0:\n", " dist[v] = min(dist[v], dist[u] + g[u][v])\n", " return dist"]}, {"cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [{"data": {"text/plain": ["[0, 1, -3, 2, -4]"]}, "execution_count": 10, "metadata": {}, "output_type": "execute_result"}], "source": ["bellman(g, 0)"]}, {"cell_type": "markdown", "metadata": {}, "source": ["### Floyd-Warshall"]}, {"cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [{"data": {"text/plain": ["[[0, 1, -3, 2, -4],\n", " [3, 0, -4, 1, -1],\n", " [7, 4, 0, 5, 3],\n", " [2, -1, -5, 0, -2],\n", " [8, 5, 1, 6, 0]]"]}, "execution_count": 12, "metadata": {}, "output_type": "execute_result"}], "source": ["import copy\n", "\n", "def floydwarshall(g):\n", " n = len(g)\n", " dist = copy.deepcopy(g)\n", " for k in range(n):\n", " for u in range(n):\n", " for v in range(n):\n", " dist[u][v] = min(dist[u][v], dist[u][k] + dist[k][v])\n", " return dist\n", "\n", "floydwarshall(g)"]}], "metadata": {"kernelspec": {"display_name": "Python 3.10.6 64-bit", "language": "python", "name": "python3"}, "language_info": {"codemirror_mode": {"name": "ipython", "version": 3}, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.10.6"}, "vscode": {"interpreter": {"hash": "916dbcbb3f70747c44a77c7bcd40155683ae19c65e1c03b4aa3499c5328201f1"}}}, "nbformat": 4, "nbformat_minor": 2}
\ No newline at end of file
+{"cells": [{"cell_type": "markdown", "metadata": {}, "source": ["# Programmation dynamique\n\n", "\n", ""]}, {"cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": ["def knapsack(c, w, v):\n", " \"\"\" \n", " Renvoie la valeur maximum que l'on peut mettre\n", " dans un sac \u00e0 dos de capacit\u00e9 c.\n", " Le i\u00e8me objet a pour poids w[i] et valeur v[i].\n", " \"\"\"\n", " n = len(w) # nombre d'objets\n", " dp = [[0]*(n + 1) for i in range(c + 1)]\n", " # dp[i][j] = valeur max dans un sac de capacit\u00e9 i\n", " # o\u00f9 j est le nombre d'objets autoris\u00e9s\n", " for i in range(1, c + 1):\n", " for j in range(1, n + 1):\n", " if w[j - 1] <= i:\n", " x = v[j - 1] + dp[i - w[j - 1]][j - 1]\n", " dp[i][j] = max(dp[i][j - 1], x)\n", " else:\n", " dp[i][j] = dp[i][j - 1]\n", " return dp[c][n]"]}, {"cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [{"data": {"text/plain": ["19"]}, "execution_count": 6, "metadata": {}, "output_type": "execute_result"}], "source": ["w = [2, 2, 2, 3, 5, 6, 8]\n", "v = [1, 1, 1, 7, 10, 10, 18]\n", "knapsack(10, w, v)"]}, {"cell_type": "code", "execution_count": 38, "metadata": {}, "outputs": [{"data": {"text/plain": ["19"]}, "execution_count": 38, "metadata": {}, "output_type": "execute_result"}], "source": ["def knapsack2(c, w, v):\n", " n = len(w)\n", " dp = [0]*(c + 1)\n", " for j in range(n):\n", " dp_ = dp[:] # copie de dp\n", " for i in range(c + 1):\n", " if w[j] <= i:\n", " dp[i] = max(dp_[i], v[j] + dp_[i - w[j]])\n", " return dp[-1]\n", "\n", "knapsack2(10, w, v)"]}, {"cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [{"ename": "NameError", "evalue": "name 'w' is not defined", "output_type": "error", "traceback": ["\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", "\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)", "Cell \u001b[0;32mIn [2], line 13\u001b[0m\n\u001b[1;32m 10\u001b[0m \u001b[39mreturn\u001b[39;00m dp[(i, j)]\n\u001b[1;32m 11\u001b[0m \u001b[39mreturn\u001b[39;00m aux(c, \u001b[39mlen\u001b[39m(w))\n\u001b[0;32m---> 13\u001b[0m knapsack_memo(\u001b[39m10\u001b[39m, w, v)\n", "\u001b[0;31mNameError\u001b[0m: name 'w' is not defined"]}], "source": ["def knapsack_memo(c, w, v):\n", " dp = {}\n", " def aux(i, j):\n", " if i == 0 or j == 0:\n", " return 0\n", " if (i, j) not in dp:\n", " dp[(i, j)] = aux(i, j - 1)\n", " if w[j - 1] <= i:\n", " dp[(i, j)] = max(dp[(i, j)], v[j - 1] + aux(i - w[j - 1], j - 1))\n", " return dp[(i, j)]\n", " return aux(c, len(w))\n", "\n", "knapsack_memo(10, w, v)"]}, {"cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [{"ename": "NameError", "evalue": "name 'w' is not defined", "output_type": "error", "traceback": ["\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", "\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)", "Cell \u001b[0;32mIn [1], line 14\u001b[0m\n\u001b[1;32m 11\u001b[0m \u001b[39mreturn\u001b[39;00m v\n\u001b[1;32m 12\u001b[0m \u001b[39mreturn\u001b[39;00m aux(c, \u001b[39mlen\u001b[39m(w))\n\u001b[0;32m---> 14\u001b[0m knapsack_memo2(\u001b[39m10\u001b[39m, w, v)\n", "\u001b[0;31mNameError\u001b[0m: name 'w' is not defined"]}], "source": ["from functools import cache\n", "\n", "def knapsack_memo2(c, w, v):\n", " @cache\n", " def aux(i, j):\n", " if i == 0 or j == 0:\n", " return 0\n", " v = aux(i, j - 1)\n", " if w[j - 1] <= i:\n", " v = max(v, v[j - 1] + aux(i - w[j - 1], j - 1))\n", " return v\n", " return aux(c, len(w))\n", "\n", "knapsack_memo2(10, w, v)"]}, {"cell_type": "markdown", "metadata": {}, "source": ["## Plus courts chemins\n", "\n", "$$M = \n", "\\begin{pmatrix}\n", "0 & 3 & 8 & \\infty & -4\\\\\n", "\\infty & 0 & \\infty & 1 & 7\\\\\n", "\\infty & 4 & 0 & \\infty & \\infty\\\\\n", "2 & \\infty & -5 & 0 & \\infty\\\\\n", "\\infty & \\infty & \\infty & 6 & 0\\\\\n", "\\end{pmatrix} \n", "$$"]}, {"cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [{"data": {"text/plain": ["[[0, 3, 8, inf, -4],\n", " [inf, 0, inf, 1, 7],\n", " [inf, 4, 0, inf, inf],\n", " [2, inf, -5, 0, inf],\n", " [inf, inf, inf, 6, 0]]"]}, "execution_count": 8, "metadata": {}, "output_type": "execute_result"}], "source": ["g = [\n", " [0, 3, 8, float(\"inf\"), -4],\n", " [float(\"inf\"), 0, float(\"inf\"), 1, 7],\n", " [float(\"inf\"), 4, 0, float(\"inf\"), float(\"inf\")],\n", " [2, float(\"inf\"), -5, 0, float(\"inf\")],\n", " [float(\"inf\"), float(\"inf\"), float(\"inf\"), 6, 0]\n", "]\n", "\n", "g"]}, {"cell_type": "markdown", "metadata": {}, "source": ["\n", "### Bellman-Ford"]}, {"cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [], "source": ["def bellman(g, r):\n", " n = len(g)\n", " dist = [float(\"inf\") for i in range(n)]\n", " dist[r] = 0\n", " for k in range(n - 2):\n", " for u in range(n):\n", " for v in range(len(g[u])):\n", " if g[u][v] != 0:\n", " dist[v] = min(dist[v], dist[u] + g[u][v])\n", " return dist"]}, {"cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [{"data": {"text/plain": ["[0, 1, -3, 2, -4]"]}, "execution_count": 10, "metadata": {}, "output_type": "execute_result"}], "source": ["bellman(g, 0)"]}, {"cell_type": "markdown", "metadata": {}, "source": ["### Floyd-Warshall"]}, {"cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [{"data": {"text/plain": ["[[0, 1, -3, 2, -4],\n", " [3, 0, -4, 1, -1],\n", " [7, 4, 0, 5, 3],\n", " [2, -1, -5, 0, -2],\n", " [8, 5, 1, 6, 0]]"]}, "execution_count": 12, "metadata": {}, "output_type": "execute_result"}], "source": ["import copy\n", "\n", "def floydwarshall(g):\n", " n = len(g)\n", " dist = copy.deepcopy(g)\n", " for k in range(n):\n", " for u in range(n):\n", " for v in range(n):\n", " dist[u][v] = min(dist[u][v], dist[u][k] + dist[k][v])\n", " return dist\n", "\n", "floydwarshall(g)"]}], "metadata": {"kernelspec": {"display_name": "Python 3.10.6 64-bit", "language": "python", "name": "python3"}, "language_info": {"codemirror_mode": {"name": "ipython", "version": 3}, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.10.6"}, "vscode": {"interpreter": {"hash": "916dbcbb3f70747c44a77c7bcd40155683ae19c65e1c03b4aa3499c5328201f1"}}}, "nbformat": 4, "nbformat_minor": 2}
\ No newline at end of file
diff --git a/files/dl/apprentissage/cours/knn/knn.ipynb b/files/dl/apprentissage/cours/knn/knn.ipynb
index ed76def2..b1f000ec 100644
--- a/files/dl/apprentissage/cours/knn/knn.ipynb
+++ b/files/dl/apprentissage/cours/knn/knn.ipynb
@@ -1 +1 @@
-{"cells": [{"attachments": {}, "cell_type": "markdown", "metadata": {}, "source": ["# Cours : Algorithme des k plus proches voisins\n\n", ""]}, {"attachments": {}, "cell_type": "markdown", "metadata": {}, "source": ["Nous allons classifier deux ensembles de points (rouge et bleu) issus de deux distributions gaussiennes (de param\u00e8tres diff\u00e9rents), avec la m\u00e9thode des $k$ plus proches voisins :"]}, {"cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [{"data": {"image/png": "iVBORw0KGgoAAAANSUhEUgAAAigAAAGdCAYAAAA44ojeAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAA9hAAAPYQGoP6dpAAC7mElEQVR4nOzdd1xT1/vA8c+9CXsjCKK4AMW99967avdubX/9dtvd2mF3a6fde287ta17a917oYAiMmXLXknu/f0RQZEkBEjCOu/Xy5ea3Nz7gJj75JznPEdSVVVFEARBEAShEZEbOgBBEARBEIRLiQRFEARBEIRGRyQogiAIgiA0OiJBEQRBEASh0REJiiAIgiAIjY5IUARBEARBaHREgiIIgiAIQqMjEhRBEARBEBodbUMHYImiKKSmpuLl5YUkSQ0djiAIgiAIVlBVlYKCAkJCQpDluo2FNOoEJTU1ldDQ0IYOQxAEQRCEOkhKSqJdu3Z1em2jTlC8vLwA4xfo7e3dwNEIgiAIgmCN/Px8QkNDK+/jddGoE5SKaR1vb2+RoAiCIAhCE1Of8gxRJCsIgiAIQqNT5wRl69atzJo1i5CQECRJYtmyZVWeV1WVZ599ljZt2uDm5sbEiRM5efJkfeMVBEEQBKEFqHOCUlRURJ8+ffjoo49MPv/GG2/w/vvv8+mnn7J79248PDyYMmUKpaWldQ5WEARBEISWoc41KNOmTWPatGkmn1NVlXfffZdnnnmG2bNnA/D9998TFBTEsmXLuPbaa+t6WUEQBEEQWgC71KDEx8eTlpbGxIkTKx/z8fFhyJAh7Ny50+zrysrKyM/Pr/JLEARBEISWxy4JSlpaGgBBQUFVHg8KCqp8zpRFixbh4+NT+Uv0QBEEQRCElqlRreJ58sknycvLq/yVlJTU0CEJgiAIgtAA7JKgBAcHA5Cenl7l8fT09MrnTHFxcanseSJ6nwiCIAhCy2WXBKVTp04EBwezYcOGysfy8/PZvXs3w4YNs8clBUEQmp3yvELO/LmVU9+vJWt/bEOHIwgOVedVPIWFhZw6dary7/Hx8Rw6dAh/f3/at2/Pgw8+yMsvv0xERASdOnVi4cKFhISEMGfOHFvELQiC0GwpBgMHF35D1Lt/Yigtr3zcv184o759Av9enRswOkFwDElVVbUuL9y8eTPjxo2r9vgtt9zCt99+i6qqPPfcc3z++efk5uYycuRIPv74Y7p06WL1NfLz8/Hx8SEvL09M9wiC0GLsuOddYj77Fy55d5Y0MloPV2bt/QSfiLptwCYIjmCL+3edExRHEAmKIAgtTV5MEn91u9Xs85JWQ+frxzP62wWOC0oQaskW9+9GtYpHEAShpYv7cT2Sxvxbs6o3EP/LJvQXTf0IQnMkEhRBEIRGpCQtB2rYAVbR6dHlFTooIkFoGCJBEQRBaETcQ1pBDTPvsosTzr6eDopIEBqGSFAEQWjRFIOBhGXbWDfrKf7qPo9V4x4i9utV6EvKGiSesJsnoxoUs89LWg1hN0xE4+LswKgEwfFEkawgCC2WoaycDZc/R8qqPUga2ZgYyBIoKr49OjJt49u4Bvo6PK49D39M1Lt/Vntc0sg4eXtw2f5P8epovumlIDQ0USQrCIJQD/uf/pqUNXsBLoxaKMbPbHkxiWy5aVGDxDXorbvo//JtOHl7VHm89fAezNz5gUhOhBZBjKAIgtAi6QpL+CX4CgzFlqdy5kZ9jW+3Dg6Kqip9SRnpW4+gLy7Dt3sHfLqKDVSFpsEW9+86d5IVBEFoyrL3x9aYnCBB2ubDDZagaN1caDtlUINcWxAampjiEQShRbJu8Fiy8jhBEGxNJCiCILRIrfqFo3F1snyQqhI0sqdjAhIEoQqRoAiC0CI5+3gSMW8akmz6bVDSagga2Qv/3mEOjkwQBBAJiiAILdjAN/5H4LDuxr/I57u3SoAk4REayJhfnm6w2AShpRNFsoIgtFhOHm5M3fAWp3/ZSMznyyk8k4ZroC8Rt0wh4rapOPuIbq2C0FDEMmNBEARBEGxKNGoTBEEQBKFZEgmKIAiCIAiNjkhQBEEQBEFodESCIgiCIAhCoyMSFEEQBEEQGh2RoAiCIAiC0OiIBEUQBEEQhEZHJCiCIAiCIDQ6IkERBEEQBKHREQmKIAiCIAiNjkhQBEEQBEFodESCIgiCIAhCoyN2MxYEQbiIojeQuGwbJ79dQ1FyJh7tAom4dQrt54xE1moaOjxBaDFEgiIIgnCerrCEdTOeJP2/o0gaGdWgkBt1huSVuwka1YtJKxbh5OnW0GEKQosgpngEQRDO23X/B2TsiAJANShVfs/YEcWu+z9osNgEoaURCYogCAJQkp5D3E/rKxOSS6kGhbif1lOSnuPgyAShZRIJiiAIApC+7Riq3mDxGFVvIH3bMQdFJAgtm10TFIPBwMKFC+nUqRNubm6EhYXx0ksvoaqqPS8rCIJQa6pi3fuStccJglA/di2Sff311/nkk0/47rvv6NGjB/v27WPevHn4+Pgwf/58e15aEAShVgKHRIIkgaUPUJJkPE4QBLuza4KyY8cOZs+ezYwZMwDo2LEjv/zyC3v27LHnZQVBEGrNs30Q7S8bRtKKXaj66nUoklYmdOYwPNsHNUB0gtDy2HWKZ/jw4WzYsIHY2FgADh8+zLZt25g2bZrJ48vKysjPz6/ySxAEwVFGfPEo3hHtjCMp0vkHJUCS8OkSyojPH2nI8AShRbHrCMqCBQvIz88nMjISjUaDwWDglVde4YYbbjB5/KJFi3jhhRfsGZIgCIJZrgE+zNr9MSe/WU3sVyspTs3GPaQVXW6fTsS8qaIHiiA4kKTasWJ1yZIlPPbYY7z55pv06NGDQ4cO8eCDD7J48WJuueWWaseXlZVRVlZW+ff8/HxCQ0PJy8vD29vbXmEKgiAIgmBD+fn5+Pj41Ov+bdcEJTQ0lAULFnDvvfdWPvbyyy/z448/Eh0dXePrbfEFCoIgCILgWLa4f9t1iqe4uBhZrlrmotFoUBTTjZAEQRDsSVVVMnYeJ/dYPFp3F9pOHYxrgE9DhyUIggl2TVBmzZrFK6+8Qvv27enRowcHDx5k8eLF3Hbbbfa8rCAIQjVZB2LZetMi8k4kVj4mO2npevdlDH7zTmQnsTWZUDNVUZBk0ePUEew6xVNQUMDChQtZunQpGRkZhISEcN111/Hss8/i7Oxc4+vFFI8gCLaQF5PEP4PuxlBSVr2VvSQRduNERn+3oGGCExq9snMFHH/3T6I/X05p+jmcvN0Jv3kKPR+9Siw7N6PR16DUl0hQBEGwha03L+L0kk0WW9nPOfIlfj07OTAqoSkoyTjHihHzKTyTViW5lbQyTl7uTN/6Hn49OjZcgI2ULe7fYpxKEIRmzVBWTvyvmy0mJ5JWQ9yP6x0YldAYqaqKekmN5O4HP6IwIa3ayJuqV9DlF7P5upfF9i12IiZdBUFo1nQFJSg6fY3HlWbk2j8YoVHKOhDLsbd+I2HpNpQyHd5d29Ht3rm0nz2cM79vsbjDde6xeDJ3Haf1sB4Ojrr5EwmKIAiNXl5MEoWJ6bgG+ODfNxxJkmp+0XnOPh5o3FwwlJSZP0hV8QgNtEGkQlOT+O8ONl7xPEDlKFt+bAq7H/iQU9+uNpucVJIksg+cFAmKHYgERRCERitzTzS75n9A1p4LfZO8u7Rj0Jt30n7WcKvOITtpibhlCjFfLje5xw4YV2aE3zrFJjELTUd5fhFbrnvFmIRcPE1z/s/ZB0/WfBJVRePiZKcIWzZRgyIIQqOUufsEK8c8SPa+2CqP559MYcOchcT/vsXqc/V++npcWvkgaUy/5fV89Bq8OrWpV7xC0xP343r0JaXmd7BWMe7LZIks0XbqYJvHJogERRCERmrX/A9QdYZqRYuoKqiw8973rKotAfBoG8jMHR8QMnHAhU0AARd/Lwa9eScDX7vDhpELTUX2gZNIGo3lgywUwEoamc7XjcejnZgetAcxxSMIQqOTG51I1t4Yi8eUZeWRvGoP7S+zbqrHq1MbJq96jYIzaeSdSEDr7krgsO5onMXwfEtl7dRMxO3TOPnVKiStBlVvqPy9zYT+DP/0ITtH2XKJBEUQhEanKDGj5oMkiUJrjruEV8dgvDoG1yEqobkJnTmU6E/+Mfu8pJFpPbwHI794lG73zuHkN6spSkzHNcCXzjdMIHhMn1oVbAu1IxIUQRAaHZcAKxo7qSqugWIfHaHu2k4ZhG/3DuTFJpvsk6MaFHo9cR0ArfqG0+q9+xwdYosmalAEQWh0WvWLwLtLO4sFiloPV0JnDnVgVEJzI8kyk1a9hldnY4G0pJFBOv+7LDHk/fsJnT6kgaNsucQIiiAIjY4kSQx64042zF1o9ph+L9yKk4ebA6MSmjJDWTlIUrWaI8/Q1sw9+hWJf28nYdl29MWl+PXoRJf/m45nB7HPTkMSe/EIgtBoxf+2mZ33vU9ZVp5xNEVV0Xq40u/5W+jx8FVi/l+wSFVVTv+0nmOLfyfnUBwAgcO60/ORq+l4+agGjq55E5sFCoLQ7BnKdSSv2kNRYgaugT6EzhyGk6cYOREsU1WV3Q98yIkPl4EsgWK81UkaGdWg0Pe5W+j33M0NG2QzZov7t5jiEQShUdM4O9Fh9oiGDkNoxFRVpfxcAaqq4uLvjSRJpKzZa0xOoDI5ASpb1x964TvaTRtM4ODIBohYsIZIUARBEIQmSVVV4n5Yx9E3fyU36gwAXmEh9Hz4KpJW7qocLTFF0mqI/uRvkaA0YiJBEQRBEJqkvY99RtTi36us9io4fZad976Hxs3Z4kZ/qt5A1v5Ys88LDU8sMxYEQRCanPTtx4zJCZjc6M9QUl7jObRuLvYITbARMYIiCIIg1Jqi05N98BRKuQ7f7h1w8XfsQoboT/6pbDlvkoRxsz9zZIkOc8VKnsZMJCiCIAiC1VRF4djiPzj25hJKM/MAkJ20dL5hAoPfusthiUrOkTjzyQlUJieSLFfbcFLSyDh5u9Pl9ml2jFCoLzHFIwiCIFht1wMfsu/xzyqTEzCOpsT9sI4Vox6gPL/IIXE4ebrXeIxbSCu0Xm6V3WEljfGW5+znxZS1b+Ia6GvnKIX6ECMogiAIglVyDscR/dHfJp9TDQr5Mckcf38pfZ+50e6xdLxyNJm7T1StP7mIpJGJuGUKvRdcR9xPG0jffhRJkmgzrh+drh0n6k+aANGoTRAEQbDKrvkfEv3pPxanVtzbBXBN4q92j6XsXAF/dbuVsuz8aqt1JI2M1sOVuce+xqNdoN1jEaqzxf1bTPEIgiAIVimIP2u57gMoTsnGEZ97Xfy8mLbxbdzbBgAgOWmQnDTG5/y9mbL2TZGcNHFiikcQBEGooiTjHPknU9B6uOLfuzOSbPws6xrgbXnlDODk7e6wPZJ8u3fkypM/kPjPDs5uPIiqqASN7EnHK0ejcXF2SAyC/YgERRAEQQCgKDmT3Q99TOKybZXTJh4dguj33M1E3DqVztdP4NR3a82+XtLKhN882VHhAsYVRB2vGE3HK0Y79LqC/YkERRAEQaD4bDbLh95LSca5KjUdRQnpbLvtTUoz8+j5yFUEj+lN+rZjJus+nDzd6fnwVY4OXWimRA2KIAiCwKEXfzAmJ3rT7eH3P/UlJRm5TPznFdrPHmFshCZJldM/XuFtmbZ5MZ4dghwYtZFiMJC0cjdHXv+F4+//RUH8WYfHINieWMUjCILQwhnKyvnJbzaGUvPt4SVZZsCrt9Pr8WsByI9LJWX1HgxlOgIGdCFodG+H1Z5cLH37MTZf9zLFyZnGzQFVFVSVzteOZ8QXj6B1d3V4TIJt7t9iikcQBKGFK83Ks5icAKCRKIhPq/yrd1gI3vfOsW9gNTh3LJ41kx7DUK4DqDLtFP/rZsoLipn0zysNFZ5QT2KKRxAEoYVz9vYAuYbRDxVcWjWukewji35G0RtAqT4RoCoKyct3kbknugEiE2xBJCiCIAgtnJOXO6Ezh1W2gjdF1RvofN14B0ZlmaLTE//HFotLniWthtM/b3BgVIIt2T1BSUlJ4cYbb6RVq1a4ubnRq1cv9u3bZ+/LCoIgCLXQ99mbjPvVyCZuC7JE5+vG49ejY5WHDWXlxP28gfVznmHlmAfZfudisvbFOCReXVEpqs5y0zhUY8dZoWmyaw3KuXPnGDFiBOPGjWPVqlUEBgZy8uRJ/Pz87HlZQRAEoZYC+ndh8urX2XrTIopTsowFp4oKEkTMm8qwD+dXOb74bDarJzxCXnSScXpIUcnYeZzYL1bQ/YErGLz4brsWzTp5ueHk44Euz9LmhCpendrYLQbBvuy6imfBggVs376d//77r06vF6t4BEEwx1Cu48wfW4n7cR2lGbl4hYXQ5fbphEwa0CCrSZoLxWAgZfVeco8n4OTpRuisYdVaxquqyvKh95J98KTZZcnDPnqAyLsvs2usex/7jKh3/6jWk6WSJHHV6Z8aZOlzS2eL+7ddE5Tu3bszZcoUkpOT2bJlC23btuWee+7hjjvuMHl8WVkZZWVllX/Pz88nNDRUJCiCIFRRmpXHmkmPkXM4rvLTu6SRUQ0KHa4Yxdifn0F2EosU7SV9RxQrR843f4AEnh2CuPLUj6anjGykNDuPfwffQ1FShslEqd/zt9D32Zvtdn3BvEa/WeDp06f55JNPiIiIYM2aNdx9993Mnz+f7777zuTxixYtwsfHp/JXaGioPcMTBKGJ2nrTq5w7Fm/8y/kVHBWfohP+2sbBF75vqNBahNS1+5C0GvMHqFB4Jp2C0/ZtmObayoeZOz6g87Xjq8Tj0b41wz97mD4Lb7Lr9QX7susIirOzMwMHDmTHjh2Vj82fP5+9e/eyc+fOaseLERRBEGqSG53I0u7zLB7j5O3BtWd/R+vm4qCoWpb9z3zN0TeX1FikOvf4N/hGtndITGU5+eTFJqN1c8GvVye7jtwINWv0Iyht2rShe/fuVR7r1q0biYmJJo93cXHB29u7yi9BEISLnd1wAGqoMdHlF5F98JSDImp5AgdH1picOPt64tUp2EERgYu/N62Hdse/T5hITpoJu07SjhgxgpiYqkvOYmNj6dChgz0vKwhCM6YaFOM+MDWM/VrqjyHUXtm5AnKjziA7O9FmYn/c2wVScjbbZIGqJEtE3n0ZGhfnBohUaC7smqA89NBDDB8+nFdffZWrr76aPXv28Pnnn/P555/b87KCIDhAYUI62QdOIjtrCRrZE2cfT4dcN3BYd5OdQy+mcXXCv09nh8TT3JWdK2Dvo58S99N6lHI9AC4BPnS+dhynvl+Lvqj0QpIiGTPHoNG9Rf2HUG923yxw+fLlPPnkk5w8eZJOnTrx8MMPm13FcymxzFgQGp/is9lsv3MxySt2VY5iaFydibxrFgNeuwONs5PdY/hn8D3kHDplcpRE0sh0vWMmwz5+wO5xNHe6gmKWD7+fvOhEkyMlnW+YgFuQP6d/Wk95QTE+Ee3oetcsIuZNdcjPgdB4NfplxvUlEhRBaFzKcvL5Z9DdFCVmVL9hyRIdZo9g3B/P270PSUH8WVaOeoCStBxjMzGo/PQeMDiSqevewsnTza4xtARHFv3M/oVfWxyxmrXvEwL6d7F7LKVZecT/uoni1Gzc2rSi8zVjcQ30tft1hboRuxkLguBQJz5cRlFCBqpiojGWopKwdBvp/x0leHRvu8bh1akNsw9/Qcxnyzn57RrKsvPx7BhE1//NJPyWKWhdRe2DLUR/9q/F5ETSajj55SoCPrZfgqKqKkdf+4WDz3+HYjAgazQoBgN7Hv6Yvs/cSJ+FN4nGfM2USFAEQbBazJcrTScn50laDae+XWP3BAWMPTD6PHUDfZ66we7Xau5UVeXckdMUJWXgGuhLwKCuSLJMcXKW5dfpDRScsW+vkxMfLGX/019V/l1RjHUwqmLg4PPfofVwpecjV9s1BqFhiARFEASrlaTnWHxe1RsoTM5wUDSCLaT9d4Rd93/AuSOnKx/z7BTMoDfuxMnXg/Ic85vtSRoZ1wAfu8VmKCuvseneoZd+JPKe2aLnTTMkFosLgmC1mub8Ja0G95BWjglGqLe0/46wesKjF7rynlcYn8amq14gcHA3JI3524RqUOh8/QT7xbf5MOU17Easyy8y9sYRmh2RoAiCYLUu86ZZvmHpDYTfNNmBETVfqqpSlJJJflwqhrJyu1xj94MfGafszNSZZO45gdbL3eS/uaSRCRrVi7aTB9olNoByizsV1/44oWkRCYogCFbrPn8ubsH+pvdhkcDJ2529j33Gnkc+If9UiuMDbCbif9/C333v4LfQa/kz4iZ+CbqSvY9/hq6wxGbXOBd1hpyDpywWwZbnFND/hVvxjmgLYOzQer4gNXTWMCb++0qVrq2GsnJivljBP4Pu4ufWl/NX93kcfWMJZTWMgphTcd0ajwu37jihaRHLjAVBqJXChHT+m/c6aZsPmz1G0hhvZKO/X0Dna8c7MLqm79jbv7H3sc+MicBFb8+SRsa/bzjTNi/GyaP+S6iTV+9h3fQnLR8kSQz94H4i776M9P+OkrUvBtnZiXZTB1VLCnSFJayZ/BiZu05UjV2W8GgXyPSt7+LZPqhWMaqqyj/97+TcsXiTfViQJHy6tGPu8W/ESp5GptHvxSMIQvPj2SGIaRsXM/f4Nwx6806Tw/+qQUHVG9h60yJyj59xfJBNVGFCOnufON9p+5LPjqpBIefgKY6/+5dNruXa2rfmg1QVt9a+SJJE8Oje9Hz4KrrfN8fkiMWeRz8la29M5esqKSrFqdlsuf6VWscoSRLDP38Y2dnJ9NSiqlIQf5Yddy6mNDuv1ucXGjeRoAiCUCe+ke0pSTtn+SBJ4vgHyxwST3Nw8utVFkcCVEXhxCd/2+RarfpF4N011OLGi07e7rSbMbTGc5WdK+DUd2tMj3JgrE3K2BFF9qHab+AYOCiSmTs+oO0U07UuSrmek9+sZsWI+XWeShIaJ5GgCIJQZylr95q9KYHxxpSydq8DI2ra8mKTqGnWvSQ1G31p/YtmJUli8Ft3Aapx80UTBrxyu1XLd3MOx6GU6Wq6IBk7omofKODfJ4zhnz4MsulAVYNCQVwqR19fUqfzC42TSFAEQagztYZN+wBUQ8OXuWUdiGXfk1+y8573iHr3z0Y7HeDk6V6l6NQUyUmDxtk2LaxCZwxl/B8v4BbsXzUOHw+GfjifbvfOseo8NcVspFp5nGknv11jeXTJoBDz+XIUg9jFurkQjdoEQaiz4NG9yYtJMrlpHxj7ogSP7ePgqC7QFZWw+dqXSV6xC0mrQZIkFIOBfQs+Z+gH8+l6x4wGi82UjleOJvarlWafl7QaOl4xul43+kt1mDuS0FnDSF2/n6LEDFxb+9J26uBabRfQakAEWg9X9EWl5g9Soc34vnWOsyAuxexIT4Xy3EJ0+cW4+HnV+TpC4yFGUARBqLNu986ucYqn+/1zHRhRVVtvWkTK6j2VsSg6PSgqSrmeHXcuJvGfHQ0WmykhkwYY28ybKgiVJCRZotfj19r8urJWQ7upg+n6v5l0mDOy1nsZOXm4EXn3ZWanYCSNTNtpg/HpElrnGJ19PKkpQ5E0Mlp30VG2uRAJiiAIdebbvSMjvnjEePO8qDdKxZ+HvH8/AQPsv9OtKbnHz5C4bLv5BEqWOPjCd44NqgaSLDNp5SKCRvQ0/l2rQXYyDnQ7+3ow8d9XaNU3vCFDNKv/y7cRer6gtiLBqhjp8evZidHfL6jX+TtdM9bsSB0Yv1cd5o5C4yI2imwuxBSPIAj10uW2abTqF87x95eSsm4fqCptxvWj+/zLCRwc2WBxnflrG5JGNp+gKCo5B09RmJSBZ2hrxwZngWsrH6ZtfofMvdEk/bMTQ2k5/n060+HKMY16l2aNsxMTlr5I8uq9xH65goLTZ3Fr7Uf4zZPoeNWYeicOgUO7EzJ5IGc3HKj+byobR5d6P3ldva4hNC6iUZsgCM3Svie/JGrx78ZpHQvmRn2Nb7cOdo+nNDuPxGXbKcvOx7NjMKGXDW/UCUdjpCssYevNi0hctt04SiNLqDoDroG+jPn5aUIm9G/oEIXzbHH/FiMogiA0S77d2teYnGhcnfGw8+iJqigcfP47jr6xBEWnR5KNozrOvh4M++hBOl8nOu1ay8nTjQl/vUheTBIJy7ZhKCnHr1cn2l82vHIqTGg+xAiKIAh2o6oq2ftjKTyThksrb4JG9UY2tY+PHeiLS1kSchW6gpJqXVnBWCfR5fbpDP/0IbvGceDZbzj88o9mn5+w7CXaXzbcrjEIgqOJERRBEBqttP+OsPOe98iNOlP5mGuQHwMX/R8Rt061+/W17q6M+m4Bm658HpCq1C1IGhnPTsH0f/k2u8ZQdq6Ao29YaB4mSex/6ktCZw0Te8mYoBgMpKzaw5k/t6IrLMGnSyhd/m86Xp3aNHRoggOIBEUQBJvL2BnFmomPVWuaVZp+jm23vYmhVEfkXbPsHkeH2SOYtmkxh176kdR1+wDQerrR5bZp9Fl4I66tfGo8h76kDF2BsbdGbacREpdttzzNpKrkHk8gN+oMfj071erczUHBmTSKkzNxDfTFp2vVJcilmbmsmbqAnIMnkbQyqkFFkiWOvPYzA1+/k16PXt1AUQuOIhIUQRBsbs+jnxpHLMx0mt37+GeE3TTRJrvy1iRoZC+mrHmd8rxCdIUluAb6onF2qvF1OUfiOPzKTyT89R+qQUHr4UrEvGn0efp63IL8a3w9QGl2fmXNiSVl2flWna+5yNofy55HPiF965HKx/z7hjPojf8RMnEAqqqyfvZCzh09DYCqN37/KroS73v8Mzw7BNHpqjGOD15wGNEHRRAEm8qPSyVz53FUxfxNWV9YQuLfjm2S5uzjiUfbQKuSk/RtR1k+9D4Slm6rTC70RaVEf/I3/wy6h6KUTKuu6dW5TY3JCYBnx2CrztccZO6NZuWoB8jYfqzK4zlH4lg79QmSlu8kY0cUmbuOm+97IkscfuXHGvctEpo2kaAIgmBTxalZNR4jaWSKU2o+riEoBgNbbngFQ7mu2g1SNSiUnM1mz8OfWHWu0BlDcPE333Zd0si0Gd8Pzw5B9Yq5Kdl1/wco5frqiZuioqqw4653SPx3R5XGf9UoKueOnKYkvYbdtIUmTSQogiDY1KUbz5miGhSrjrMnfWk56duPcXbTQUqzLmwemLpuP0VJmWanp1SDQsJf/1GSUfPNUePizLBPHjJ2aL+kCFbSyGhcnRn8zj31+jqaktwTCWTtiTY/uqaqFKdmU3AyxaqiYYMNdnUWGi9RgyIIgk35RLSj1cCuZB+INXuT17q70GHOCAdHZqQYDBxZ9AtRi3+nPLcQMLZJ73ztOAa/cw/njp2x3IEWY5KSH5uMW2u/Gq/X6aoxaD1eZf+TX3DuaHzl423G9WPw4rtbVHFs4Zk0q45z8nKvsYeNs58X7iGtbBGW0EiJBEUQBJsb/NZdrJ74KCqmC2X7v3I7Tl7uDRCZcQrh5Ner4KKwVL2B079sJGtfLF3umI5qJrG6mNbD1eprhk4fQrtpg8mLSaIsKw+PDkGNqr2+o7j4W9cPo83E/iT+u4PyvCLzSa6HK4UJ6fhEtLNliEIjIqZ4BEGwueDRvZm8chFelxR/Ovt5MfTD+fR44IoGiStzbzQnv6qanFRQDQp5sUmUpJ2radNcPEID8evduVbXliQJ38j2BI3s1SKTE4CAQV3xqKHeRuvhSoe5Ixm7ZKGxqZ+Zf4uSs9msGH4/hYnpdohUaAxEgiIIgl2ETBzAFSd/YNqWdxj17RNM/PcVrk39jW73zG6wmE5+vbrG4sv4JZsIv3kyyOazlD5P34iscUxH3OZEkmUGvnaHxWP6PnszTh5utJ00kLFLFppMJsGYUJbnFXFk0S92iFRoDESCIgiC3UiSRPCo3oTfPJnQGUPrvaNtfRUmpJlfunpecUoWwz5+kI5XjAaM9SmSVlO5OV2/F+fR5Y4ZjgjXrhS9gdzoRHKjE2us97ClzteMY+RXj+HkbZzikzQySKBxdWLAK7fT86IGbJm7TlhMKFW9gVPfr0Gp4d9UaJpEDYogCC2Ga6BvjQWwzn6eaF2dGffrs+Q8FUfczxsoy87Hq1Mbwm+ZjEe7wBqvk3cymeSVu1HKdPj3CydkQn8kuXF8HlQMBo69+RtR7/1J6flluq6BPnSffwW9nrjWIXslRcybSqdrx5G4bDtFSRm4tvalw9yROPt4VjmuOC3b5D5KFzOUlFd2+hWaF5GgCILQrKiqSm7UGUqz8vBs3xqvziGVz4VdP4G4H9aZfa2kkQm/ZUrl3/37hOHfJ8zqa+sKitl6y2skLtsOsoQkGfcA8urchrG/PkvAgC51+6JsRFVVtt64iPjfNle58Zdm5nHg2W/IPhDLuN+fc0gypXVzqXEnZ7cg/2rLsy+lcXVusIJrwb4cltK/9tprSJLEgw8+6KhLCoLQwiSv2s3SHrexrPf/sXr8I/wRfhMrxzxI1v5YAEImDSB4bB/jtMIlJI2Ms68nPR6qWwGvqiisv+xpkv7daXxAUStHagoT0lk9/mHy41Lr9oXZSPLK3cT/usn0qISqkrB0m8M7/FoSfvNki1NyklZD2E2THLZDtuBYDklQ9u7dy2effUbv3r0dcTlBEFqgqHf/ZN2Mp8iLTqzyePr2Y6wc/QBZ+2KQZJmJf79Mh8tHnW+eRuUndN/uHZi+9V082tY8hWPK2Y0HSdtyxOT0kWpQ0JeUc+yt3+p0bluJ+Xy5yeSsgqSRif70HwdGZJlfj450+b/pJkdRJI2Mk5c7vZ+8vgEiExzB7lM8hYWF3HDDDXzxxRe8/PLL9r6cIAgtUPqOY+x5+GPTTyoqhjIdux74kJnbP8DJy51xvz5LwZk0UtbsxVBajn+fMILH9LGqe6k5p3/ZiKTVmP3Er+oNxP24juGfPFjna9RX3onEGhvQ5UUn2ex6ik5P4j87iP91E6U5BfiEt6XLHTNqNdU17JMHcQ3wIeq9vzCUlFU+HjCwKyO/fqzaUnah+bB7gnLvvfcyY8YMJk6cWGOCUlZWRlnZhR/A/PyWtcOnIAi1p6oqm69+0fJBikrmzuPkxSbh0yUUgNL0c6Ss2UvSvztRDQoe7VvT7d45dLt/LlrX2q82KjtXUOMKIX1RKYrB0GBLlJ39PG1yjDVKs/JYM/lxcg6dqixMTt96hJjPl9Pt/rkMefdeqxJCWaNhwKv/R68F13F2w0H0JWX49eqEf6/a9aERmh67JihLlizhwIED7N2716rjFy1axAsvvGDPkARBaGbStx2lODXbqmMLz6Tj0yWUhKXb2HT1C3C+iBWgKDGDfU9+QdKKXUxe/XqtkxSvjsEWR1AA3Nr4N2j/lM7XTSBrX6z5lTGyRNj1E2xyrU3XvMi5o6cBKr/HFd+bEx8sRV9YgpO3B5IEwWP70m7GEIvfG2dvDzrMHWmT2ISmwW41KElJSTzwwAP89NNPuLpa1xL6ySefJC8vr/JXUpLthhoFQWhY+tJyUtbs5cyfWzkXdcZm5807kVjzQee5BHhTnlfIlpteRVXU6smEopK+7WidakUibp9uuaBTlom867Jan9eWIuZNwb1tgJkiYQ1urf2MNR/1lH3wJGmbDlmcTjr5zWqOf7SU4x8tY8OchfwZcTO5JxLqfW2h+bBbgrJ//34yMjLo378/Wq0WrVbLli1beP/999FqtRgM1f8ju7i44O3tXeWXIAhNm6qqHH3rN35tcyVrpy1g01UvsKzX7fw79F5yzn/Crg9r98Rxa9uKVv0iiPtpA4aScvOjCIpK9Md/m99x1wy/Hh3p8fBVJp+TNDLeXdvR/YHLa3VOW3P28WT65sX4RLY3xnW+CR2AV1gbRv+wwCZN21LW7LNYjFtJr6DqjPeCoqQMVo17mLIcMbUvGNltimfChAkcPXq0ymPz5s0jMjKSJ554Ao1oEy0ILcKBhd9w5NWfqj2evT+WlSPnM2vPJ/h0Da3z+dtNG4zsrEUpt3xjHfz2PUiSRM6hOGNNhIXRjpK0HMpyCnAN8KlVLIPevBPP9q058vovlJzNAUB21hJ24yQGvXknzt4etTqfPXh1DmHOkS9J23KYtE2H0BWXUnD6LKlr97Fm0uMABAyOpN9zN9Nu2pA6XUPR6WvsX3Ip1aBQmpVH7Ner6XVRN1lb0xWWkLEjCqVcR6v+EbiHBNjtWkL92C1B8fLyomfPnlUe8/DwoFWrVtUeFwSheSpKyeToaz+bfK5i6e3BF75j7M/P1PkaLv7edLt/LlGL/zA7KhJ240Q6Xz0WMDb2sobGxanWsUiSRPf5lxN5z2zOHT2NoUyHT2R7XHxtU3hqK5Ik0WZsX/z7hrNixP3kxyZXmY7J2hfDuhlPMfKrx4iYN7XW5w8Y1LXGgmGTFJX4XzfZJUFRdHoOLPyG4x8uxVB8fjGGLNFhzkiGffwAbq39bH5NoX4aR+9lQRCapdM/bbD4SVrVGzjzx1Z0BcX1us7ARXfQ5Q5j7YSkkc/v7yKBJNH7qesZ9d2CymPbXzbMcq2IRiZodO96dSeVtRpa9Yug9dDujS45udiRV36qlpwAoBgTvR13v1OnKZe2kwfi2SnYummeS+jy6/ezYIqqqmy5aRFH3/z1QnICoKgk/rOdlaMeoCy30ObXFerHoa3uN2/e7MjLCYLQwIpTs5Fky3vfqHoDpZm59U4IRnz6MD0fuZrTP22gNCMXj/atCbtpYrXGa23G98O/bzjnjsWbTFRUg0LvBdfVORZ7KDtXQMKfWynJyMWjXSAdLh+Fk6dbvc6p6PTEfLHC4r+NojNw6od19Higdt11JVlm/J8vsHrcw+gKSyxeo8rrtBr8enWq1bWskb7tKGd+22zyOVWvUBB3luiP/6bPUzfY/NpC3Ym9eARBsBu3YP8ai00ljYxLLWs9zPGJaEe/52+xfD1ZZtKKV1k79QnOHY03FomeHzFAgmEfPUC7qYNtEk99qarKkUU/c+ilH1DKdUga4zLmnfe+x6C37iLyzll1PndpVh66/CKLx0hauc6N21r1DWf24S84/u6fnPppPbr8YlwDfChOyTL7GlVvIPKuun9N5pz6Zo3lJnqKQszny0WC0siIBEUQBLvpfP149j/9ldnnJY1M+zkjHV486t6mFZcd+IyU1XtJ+Os/9CVl+HbvSJfbpjaqosljb/3GgWe+rvx7xQ1WX1TKzrvfRevuSvhNk2p1zsKEdMpy8nG2ZupJBSevuo/UeLYPYvDiexi8+B7j6VSV/259vfqGjZLxWl3umEGbCf3rfD1zCpPSa6yJqShqFhoPkaAIgmA3nu2D6PnwVRx7u3pfEUkjo3FxqnHEw15kjYbQGUMJnTG0Qa5fE11RCYde+sHiMfuf+pLO14+3qvlb6vr97H/6K7L2xgAgyRKugT6UZheAmVEuVW+g45Vjah+8GZIkMeqbxwkY0IVj7/xBUUI6AF6d2tDzkavpetesem03UEFfXMrpJZtIXrkbQ2k5JennQCOBwczScow9coTGRSQogiDY1cDX78DJy42jb/6Kvqi08nHf7h0Y+fXj+PXo2HDBNWIpq/eiLyyxeExxShaZu04QNKL6ysjsgydJWbsPVa+g6PQcevF7kC/c/FVFpTQ73zi9JUnVVkBJGpk24/sRMKirbb6givPKMt3nX063++ZQfDYbSZJwa9PKJokJQG50ImsmPmrsLixLxq+v4ndzMWlkIuZNs8n1BdsRCYogCHYlyTJ9n72ZHg9fRer6/egKSvCJDCVgYFeb3ZQag6z9sST9swN9SRn+fcLocMXoOu3pU6Esp8DK46qusinNzGXTVS+QtvVI5WqmyumNS2/S52/ekiShKiqyVouqGjvshkwawNglC+32byTJcp13jjZHX1rOmkmPGUdM4MLXayk50WpwDfSl+/y5No1FqD+RoAiC4BBOnm50mGP/vVQUnZ7k1XsoOJWKs68noZcNw7WVbYpwTSk7V8Cmq1/g7IaDxs6skoSi0+M8/wPG/rKQtpMH1um8Xp3bWHlcSOWfDeU6Vk98tLJlvFWrZxQVFZWBb95JSUoWWg83Olw+klb9IuoUd0M68/sWi0W4pkaKAgdHMvqHJ0UflEZIJCiCIDQbSSt2se32NynNyDV2i1UUZCctPR++iv4v34Yk27b1k6qqrJ/9DJk7jxv/rjdQcfsrzyti/aynmbHzAwL6d6n1uduM64tH+9YUJWWabEAnyTL+/cKrTJElLN3GuaPxdflSCBreg9bDetTptY1F8qrdlTsnm6SqSBqZkV8/hqIzEDCoq9gVuRETjdoEQWgWzm46yIbZz1CamQecHz1QQSnXc+T1X9j3xBd2uOYhMrYdM31DVFQUg4Gjry0x+VpdUQmFSRnoi0tNPi/JMiM+f8Q4TSNXnWaRNDKyi5ZhHz9Q5fHTSzbWOQmzalVPI6eU61EtTOeA8eei8/UT6HLbNJGcNHIiQREEoVnY/8zXqFQfwgdAhaj3/qQ4zbZLSc/8vgUsdUtVVM78tRXloiWuucfPsOnaF/nJ9zJ+73AdP/nNZuutr1MQf7bay9tOHsjU9W8ROCiyyuPBY/ow/b/3qj1elp1f600OkSR8u3eo3ECwKQsY0MW4ZNkcWcKvd2erVj0JDU9M8QgOU5hfxtaNp4g6eBaDotKlWyBjJ0fgH9DwG6gJTYuqqpw7Fk9ZVh6eHYKQNJrKaRazr1EUzvyxle73zbFZHLrCYrNLdCspKmU5+bi19iNrXwyrxj6EoVxfOeqi6PSc/nkDSf/uZOaOD6ptnBg8ujczd35IwelUSjJycW8bgGdoa5OX8oloR+auE7XbB0dVjdNfzaBgOeK2qRx8/jsUVQ+mBlIUle7zG3ZHacF6IkERHCL2eAZvv7SBslJ95Qfc6Kh0lv8Vxd0Pj2TQ8A4NG6DQZCQt38neJz4n70Ri5WP+/cJrfJ2kkSnLyrPqGvriUk7/spGzmw6BqtJ6eA/CbppUraGcV4dg0zfCSySv3EP4LZPZestrGMp01aaEVL0BXX4RO+5azLRN75g8h1fnkCoFsaZ0uWMGJ79ZbTkYybhyRdUZ0Hq4MvTD+Q4pXrYlVVFMTmW5Bfkz+ocn2XLDK1VXL51fZtz5+glE3DrFwdEKdSUSFMHu8nNLeOvFDZSX6auMvquKikFR+fit/3hxsTehHUUVvWBZ/O9b2Hzti1w6jn/uyOkaX6vqDXh2DKrxuKx9MaydtoCy7HxjwaVqrO3Y//RXTFj2Em3G9q08NmTyAA6/+pPlE0oSucfPkLnreJWkqlp8BoW0LUfIO5mMT0S7GuM0pfXQ7nS9cyYxny2vHoZGxr9vOJ2uGYu+oATPzm3oeOVonDzqt6ePoxScSSPq7d859cNadPnFuIW0IvLOWXSfPxdnnwv1M52uHotnp2COLf6dpH92ouj0+PfuTLf75xJ+0ySbF0oL9iMSFMHutqw7VS05udS6FdHcdu8wxwUlNDmGch077373/IhF1R8ma5bTat1cauyKWpJxjjWTH0NXUFLtvLrCEtbNeIq5x77Cq5NxCbBXmOURDeNJVDSuLuRaSE4ulncisc4JChj3EvKOaMext36j5HzNjdbTja7/N53+L9+G1t21zuduKDlH4lg15iF0RaWVoyIlqdkcevF7Tv+ygen/vVdlKXngoEjG/bKwocIVbEQkKILdHd6fYjE5URSVg3uTHReQ0OgZDAqH9iZz/EgaqqoSEdma1pnJ1ZqSmXRp19Dz+7wMeffeGncAjv1yJeX5xaYbeykqSrmO6I//YdCbdwLg0TYQv95hnDsSZ/G8+pJStB7WJQbaeu5SLMkyPR++iu7zLyf3RAKq3oBP19AmmZiAsd5o0zUvmdwVWTUo5J9MYc8jnzD62wUNFKFgLyJBEezOYMWnW8XCHhlCy5KanMfbL24gK6MIjcY4lbNhVSzuThDp3xrvnAyLr2/VN4LsA7GVf/fqHMKAV26n09Vja7x2wrJtFruOqgaFhKX/VSYoAK2Hd68xQTnzx1Z6P3UDGlcnDKU6s8e5+HsRNMI2vUhkraZZLKNN33qE/BjzOyqrBoX4XzYx+O277dqQT3A8kaAIdhceGciZuBwUM2/8siwRHmnbltdC01RcVM5rC9dSkFcGgOGixLVEB4eHTGTQpmW4lhabPcfIrx/F2ceTgvizOPt64t8nzOoVKobS8hqP0V9yjFKuR5Jli8t7ixLSkVSV7g9cydE3fjFbWNv7qRvQuNS9PX5zlH3gZI3fX0WnJ/d4AsGjejswMsHeRLWQYHfjp3ZBtTDHoygqk2dGmn1eaDm2bzpNXm6pyWRWBQwaLamdLPysSFAQn4ZnhyDajO1Lq77htVo+GzCgK5LWfI8MSSMbe21cROPsVK2RmimysxP9X55H5N2zjStpNDKyk7Zyv5zeT91Aj4eutDrWlkJ21lp8/6ggErvmRyQogt21aevDrXcbt7SXNRfeyOXzb+qzrupJjz7W7TvS0Kx5oxTqbvvaaItTLMgyGSGdzD+vwuarXyTPwpSAJZH3XGaxh4hqUOh2z+wqj7WdNshy3xFZpvWIHjh5uiFrNAz7cD5XxP5An2duosvt0+n3wq1cnfALA5pJLxJbazdtMDWt5XYN9KGVFUvNhaZFTPEIDjF2cgRt2/uw+u8THDt0FlVRiegWyOSZ3egzsG1Dh2eRYjBw6ru1HH//L84di0d20hI6cxi9HruGwMFi5MdWyvMKyTqeBJ6+Fo9T3SwXkaqqyokPlzH0g/trHUPgoEj6PX8LB5//ruqeLucLb7s/cDkhl2z+1276ELy7hlIQl4KqN9XyXqHX49dWecg7LIR+z91cYzyKTk/ivzvJ2BGFJEuETOhPyKQBLWqprFfnEDpcMZrEpdvMrtbq+di1yE7idtbcSGoj/kiYn5+Pj48PeXl5eHt7N3Q4QgukGAxsuf4VY0vzi1aHSFoNqqIw5qen6XzNuAaOsnk4/uEyvv3mIBltO6OauQFLqASU5tFj7TKL5/LoEMTV8T/XOZaEv7dz7K1fydgeBUCrAV3o+fBVdLp2nMlRjoIzaayZ+CgFp89W/pxUbFY4+K276zR1k3Uglg2zF1KckoXkpAHV2MvFu2sok5a/irc1S5xtTF9SRuGZNDRuLsYOvg4a8dEVFLN+zkLSNh0y/t/TGyp/73bfHIa8d58YfWpkbHH/FgmKIFgQ8/lydtxlurMnEshaLVcnLRFbtdvA2hlPcnx3PAdHzrB43OD0KNx377V4jHtIANck/1rvmBS9AVTVqk/nhrJyzvyxlcS/t6MrKsWvZye63jED73DrRghVRSH+t82c+HAZ2YfiMJSUgom9hSSNjFubVlwe9TVOXu7VzlNwOpWod/8i/rdN6ItK8ekaSrd75xB20yRkC/U1lpTnF3Hw2W+J/Wol+iLj5oY+kaH0eeYmwq6fUKdz1paqqpzdeJDTP22gNCcfzw5BdLl9Gv69wxxyfaF2bHH/FmNigmDB8ff/Asn8BnSKwcDJr1fTe8F1jg+umVHKdPjkZBJ66ihJ4b2M3/OKT8Xnv/9BaQn07RfMyf0as3UfklZD6+G2W6pr9bFOWvx6d8YrLATv8La4Bli/5FVVFLbe/Bqnf95wSR8XE8XCBoXilEziflxP5N2XVXkufUcUa6c8bmynf/77k33oFNtuf5P4P7YwcdlLtZ4K0RUUs3L0g+RGnakyxZIXk8zWG1+lKCmD3k/Y/+dfks5PcU3ob/drCY1Dy5nIFIRaUvQGco8nmE5OKqhU6bkh1F3g4EgkjUzn4/uJPPgf7gW5lc+5FhcScXwvo91y6HbvZagGC4WsegPd7p9j/4AvcvLb1fwRfiN/97mDFcPvZ0nIVWy+/mWKz2Zb9fqYL1YYkxOwXCRcSSL+t01VHjGUlbNx7rPoS8qrJm/nz5eyZi/H3vrNqngudmzxH+QeO1O9/uP8/4v9T31FYUJ6rc8rCDURCYogmCHJknEJaA3HyC5ODoqoeevyv5mAsfFrcFIcgzb/zfDVvzB89RKGbPiTtnHH6Xn/XPx7dWbIe8YC2IuXBFf8W/V74VaH9sM4+sYStt32JoVnLtykVb2BM39sZfnQeylJz7H4elVVOfbOH7W7qKpSnldU5aGEv7ZRmplrfndlReX4B0tRzCR3udGJHHz+O3bN/5Cod/6gNDMXVVWJ/vQfiz1IJFki9quVtYtfEKwgpngEwQxJlmk7dRApq/eaXT2gGhRCpw91cGTNk1fHYEZ88Qjbbn/LmGzoDTiXl8H5gtnIey6jw+WjAOh+3xz8+3Qm6p0/SF1/AFAJGtWbHg9cQdtLVtnYU3FqFvuf+srkc6reQPHZHA6//JPFFUX64lIKYmu31YOk1eDXs+py68w90chOWhSd3uzrStJyKE7NxjO0deVjhnId2+94m7gf1iFpZCRZRjEY2LfgC/q9PI/S9HOWg1Eh/1RKreIXBGuIBEUQLOj12LUkr9xt8jlJo8G9bSs6XN60tqpvLFRVRVFUNBeNUkXcOhWfrqFEvfMnyat2o+gNBAzsQvf5V9DxytFVVmoEj+rd4J1DT3231uLzqt7AiY+WcerHdQSN6EH3+dUTKElT+8JVVW+g652zqjwmazVW9em5tK5m133vE/fTeuN5DUplMq4oevY/8UXNwciSyWJdQagvkaA0oJysItb8c4Jtm09TXFiOf4A746Z0YcL0rri5iWmDxiB4dG9GfvUY2+94Gzi/u60sg6LgHuLPlHVvig6WtRQXm8XKpVEc2JOEYlAJauPFpBmRjJvaBa1WpvWwHrQeZpsiV3sriE87X9Rq+ThdXhEpa/eTvHIPfRfeRL8Xbq18TuvqjGtrX0ozcq2+brf75hA0omeVx0ImD+DY2xZqTCQJn8hQ3IL9Kx8qSs4k9qtVluusaqDqDXQSS+0FOxAJSgNJTszl1afWUFKsq2zrnZVRxB8/HmTHltM8/eoUPDxdGjhKAYyf6kMmDiD2y5VkHzyJxs2F0BlD6XjVGLSuIjmpjb07Evj4rf8AKn/uM9IK+PGrvRzal8xDz4xHq206pXEu/l5WH1tRuHropR8IGtWLkIkDKp9rN20Ip75bU+M5PDu3oddj19D1fL3OxUIm9Me3R0fyYpJMr3BSVcJvnVJlFCrx7+1Wx2+KpJEJHNqNNuP61us8gmBK03knaEZUVeWjN7dWSU4uPAdnk/P55ev9DRSdYIpHu0D6PX8LE/9+mXFLFhJ+06Qmn5yoqkry6j1svOI5lva6nZVjHyL6s3/RFZXY5XpFhWV89s52FEWt8nOvqoAKUYfPsm55tF2ubS+drh1nuc29CZJGJur9v6o8NuiN/1kuyJYkejxyFVee/IHIO2eZbEomyTKTVryKR6iZjTcliUPPf0f6jqjKh3SFJUhW7CNkTpvx/Zj4zyuiSZpgFyJBaQAnT2SSmpRndndfRVHZsTWewoIyB0dWs7Mpefz7+1F+++EA/22Io6S4nEN7k/n7tyOsXBbF2ZS8hg5RsIKiN7DlhldYN/1JEv/ZQW7UGdL/O8rOe97l777/oygls8rxCadz+ObjXTz/6Epee2Yt61ZEU1Jc886/F9u28TR6S/vcqLBueXST2u+oVd9wOl452qrNAiuoBqWyQ20F10BfRnz+iHEJ0yWnkmSZgEFd6f/CrTUmAp7tg+gwd9SF/jFVLqxiKNOz8YrnKgtpfSLbmy0Ar0mvx69lypo3cPGzfhRJEGpDTPE0gPhT2WZ7f1Uw6BWSE3OJ7BHkuMAsKC/T8+UHO9i9LQFZlpAkMBhUvvzA+LyskVAV+PXbA/Qb3I47HxyBm3vTHmFozo6+voT4XzcDXLhBnf+BLExIZ+MVzzNz54dIksTSJYdZtuQIsixVJtUnjqXz7+9HWfDSZEJCrWtIlhifgyRJFhOQ7KwiSkt0TepnZ9T3T6L1cOPU92uNyYUVfUxMNYCLmDcV93YBHHn1J9K2HAHAJcCHbndfRq8nrkXr7lrjeQ1l5cYlv+a+x4pCafo5Ev/eTscrxxA6fQiuQX7nlyfXLjH0F5vzCXZm1xGURYsWMWjQILy8vGjdujVz5swhJibGnpdsEqydY29Mc/Gfv7edPdsTAOMIj8FQ9c1MMaiVN57D+1J4b9HmJvVJuCUxlOuIevcPszcxVW8ga080WXui2bM9gWVLjDfLS0f8CvLLeOvFDehNbZBngtZJU210wBRNHduxNxStqzOjvnmcq07/xJB37iVk0gCLX6ek1dB22mCTz7WdNJBpm97hhtx/uDbtD649+zv9XrjVquQEjMml7pL+KJeSnTRk7Ys9/2cto79fgCTLNfb8uZjWw5XQmWJ5fVNWlJzJoRe/Z8tNr7Lz3vdI23qk0b1n2/UOuGXLFu6991527drFunXr0Ol0TJ48maIiy/+Bmrue/UJqLJr39HKhY2d/ywc5SEpSLnt3JFpd6K8oKieOphMTlWHfwIQ6yTuRSFl2vsVjJI3M2Y0HWbk0yuRsARj/nbMziziwO8mq6/Yd2BbFYP6HSJYluvcOxtm5aSUoFTw7BNH9/rmM/uFJY0JhbtpHUejxwOUWz+Xs7YFbaz/kWi5Blp1rXv2nqiA7Xxg8bztpIDO2vU/IpIGViZXsdP66Zr6Efs/fgpOH5V2lq19XJXNPNAef+5Z9C74g/vctGMp1tTqHYBtH3/qN3ztez6GXfiB+ySZivljBqrEPsXr8w5TnFTZ0eJXsOsWzevXqKn//9ttvad26Nfv372f06NH2vHSjFtTGiwFDQjm4N9lsHcrU2d2MnzgbgT3bE6oM71tD1kjs+i+eyJ6NY4pKuMDaT0lleoX4U5ZbtcsaiajDZxk8okON5+szoC0h7XxIS803+bOkKCozr+hp4pVNi1trPyYtf5V1M59CX1J2YQfs8yMUo759glb9Iuxybc8OQXiFh1AQl2pqGx/AOELWbvqQKo8FDo5k8spFlOUWUp5biFtrXxL/3cmu+943JrPn9wfSerjS7/lb6PHwVbWKqzQzl41XPE/6tqNIWg2SJKHo9Li29mXc7881eD+bluT0LxvZ9/hnJp9L33aMzde/wuQVixwclWkOrUHJyzMWUPr7mx4ZKCsro6zsQmFofr7lT3lN2R0PDGfxSxuJPZFZefOXNRKKQWXMpHBmXN543qhLinXni/OsT1BURaW4qHZFlIJj+ES2x8nHw+JUgGpQCBjaA/Yfq/F8ipVFlrJG5tHnJvDGc+tJS81Hls/Xo0gSEnDL3UPo0aeNtV9GoxY8pg9Xnf6J2C9XkrJ2H4peT9CIXnS9cyZenez3NUqSRJ8nb2Db7W+afl6rIWBgVwKHdKv2nKFcx5nfNhP9yT/kx6Xg7O1B2C2T8eveEX1hCa6BvoTOGoaT54WRE31JGUnLd1GSloN7G3/azRxWbXWbYjCwZuoCzh2JA4wJUsU7SVlWPmunLuCy/Z/iG9neNt8EwSxVVTn00vdmN0BVDQopq/aQc/Q0/r06N0CEVTksQVEUhQcffJARI0bQs6fpm++iRYt44YUXHBVSg3Jzd+bJV6Zw9GAqO7fGU5BfSmBrT0ZPDKdzREBDh1dFcIg3hlpW+kuSRGCQqO5vjLSuznS7+zKOvLHEZGGkpJHxiWxPhwl9CPorgfSzBWbPpRhUwiPNLGs1oVWgB6+8P4uDe5I4sCcZXbmedh38GD0xHP9WzasbqWugL72fvJ7eT17v0OuG3zqFgvizHH75R9BIcNG0mluQH2N+erLaaiB9SRlrpy0g/b8jgPHmpS8s5cR7f+Hk5c60zYvx7x1W5TUxny9n7+OfocsvrhxhcfLxYPBbd9Hl9umVxyWv2E3OwZMmY1UVBUWn59jbvzHyi0dt900QTCqMP0tetOUpWUkjk/TPzkaRoEiqg6pi7r77blatWsW2bdto166dyWNMjaCEhoaSl5eHt7e3I8IUTCguKmf+vD/Qldeu38Mbn8whqI1IUhojQ1k562cvJHXtPiRZvrAZnCzh1tqPaVvewSeiHRtWxfD9Z3tMnkOSwM3diXe/vhIXF9OfdRSDgYJTqagGA15hIaLrroOoisLma1/izB9bjXUkKpWfmj07BTNt8ztV9uPZ+8TnRL39u8lNASWNjEf71lx58gek8/sixX69iu3/95bZ64/69gnCb54MwNabF3H6l40WlzNrPVy5qWBF3b5YwWrnos6wrNftFo+RnTT0fOxaBrx8W72ulZ+fj4+PT73u3w5ZJnLfffexfPlyNm3aZDY5AXBxccHb27vKL6HhuXs4c+tdQ2o+8CKzr+ktkpNGTOPizKTlrzL6hydpPbwHbsF++ES2p/+L85hz5Et8Ioz/T8dN6cLwMcZN6S5u6CXLEk5OGh54cpzJ5ERVFKLe+5PfO17PX91uZWnP21nS5ir2P/M1hjIx9Wdvxz9YakxO4MLM7PnPokVJmWyY/UxlLZK+tJyYz/41u2OxalAojE8jZe0+ABSdnn0LLO/Rs2/BFyjne96U5xbV2GtFX1Ta6FaQOJKhrJzc6ETyT6VY3Dm6vjw7BqFxs9yhXNEZ8OvVyeIxjmLXKR5VVbn//vtZunQpmzdvplOnxvFFC7U3cnwYnt4ufPPRLnLPme806uQkc+P/BjNmYtPvkZCTVcSaf6PZtimuWe6VJGs1hN0wkbAbJpo/Rpa444ER9BscyvqV0SQn5OLkrGHwiA5MnB5pMglVVZUd97xH7OfLqzxenlvIkdd+JmtPNJNWvIrsJNow2YNiMFjck0fVG8g5FEf6f0cJHt2b/JPJxmkaCySthsxdJ2g3dTBnNx6kLMtyQ8aStBzSth4hZHw/vLu0Q9LKqBaWo3t2btMiu9HqS8s5/NIPRH/yD+W5xtUzHh2C6P34tXS9y3TH4Ppw8nAjYt5UY0JqKmmUJVz8vOgwt3FsgGrXd4h7772Xn3/+mb///hsvLy/S0tIA8PHxwc2tdkvUhIbXd2A73v7ict57dRNHDqRWq7MKDPLk6UVT8PNv+rUEqUl5vPzUakqKxF5JsiwxeEQHq1bqAGRsP1YtOamkqKSu30/cT+uJuHWqDaMUKhScPktxcpbFYySthrMbDhA8urfJpnHVqCrS+eNKa0hOKpRm5gLQ9Y4ZRC3+3UIwEpF3XWbVOZsTQ7mOddMXkL71COpFtWBFCensvPc98mKTGfLOPTa/bv+XbyNt8yHjnk0XJSmSVoMkS4z95Rk0VixXdwS7TvF88skn5OXlMXbsWNq0aVP569dff7XnZQU70mplHnpmPA8vHE//IaG06+BLt97B/N/9w3n1/VnNIjmp3CupyPxeST+LvZLMivliReXNzCRZIvqTfxwXUCOg6A2kbztK8qrd5Mel2vVa1raur1h95d2lHe5tLRfmqwaFtlMGAuDRvrXFYyt4djC2GPDpGkrf524xPnjJiICkkQkcEknk3bPQFRSjGGpX59aUnfxmNWlbDldJTi52/L0/ydxr+72pXHw9mbH9fXovuB6XAGMXaEmrocPlI5m588Mqm1g2NLtP8QjNjyxL9BnQlj4D2jZ0KHZxMjqT5MRcs88risrOLfFcd+sAPL1bxihKbeSeSLC8gZ6ikn8qxXEBNSBVVYn59F8OvvAdpRm5lY8Hj+3D8E8ewqdrqM2v6dW5DS7+XpTlmF99peoNBI3oAYCs0dDz0avZ89DHJo+VtBoCBnUlcFAkAEEjeuLZKZjCM+mmuxHLEt4R7aosZe733M14dQrm8KKfyY8xriJx8vYg7MYJoMKvIVehKyhB4+pM2I0T6b3gOrw6h9T1W9AkxHz6LxcqmE2QJHbd/wET/noB9xDbrux09vGk/0vz6PfiregKitG6uTTKKdfG00tdEBqJM+f3SrLEYFAsJjEtmYu/d42b5zl5ezgomoZ19LVf2Hnve1WSE4D0/46yfNi9dknUNM5OdLtvrtl/A0kj49kpmLZTBlU+1n3+5XS9a5bx+YrRr/Ov9+nSjvF/Xmj/IMkywz95yFg0fek1ZAlJlhn24fxq9RPhN0/m8uPfcOXpn7j8xLfM2vMxCUu3EfP5cnQFxro2Q2k5J79dw98D7iLn6Ol6fR8au7yTyZY3ZFNVsvZE81uH6zj08o92+cAvSRLO3h6NMjkBkaAIQjUarWxVS7rGtFdSQ9EVFHPqh3UcfWMJp35Yh66whM7XjrO48ZykkS0W5jYXJek5HHjuW5PPqQYFXUEpB808X1+9n7qetpONCUjF0mAwfu+dfTyYsOylqo9LEsM/fpAZ298n/KZJtB7eg3bThjD6x6e4bP+nuAdXba7ZdvJAJq9+Hb+eVRc++PcJY8raNwiZ0N9kXJIk4dUxGJ+uoex+8CNKM3KrTUmpegP6whK23rSoWY/CO3tZNx2uGhQOPvtNi5sWBQf2QakLW6yjFoTaSj9bwON3L7N4jIenM+99cyVOjWQ7goZw/P2/2P/Ul+iLy5A0MqpBQevuQr+XbiPm038oiE+rNtVTcYOcc+RLmw9bNzZR7/zBnsc+tZysaTVcn7UUZzuMKCl6A6d/2Uj0p/+QH5uMk7c7YTdMJPLuy3Bv08om11BVldyoMxSfzcY9JAC/Hh2tel1hQjq/d76+xubUM3d9RODgyPoH2gjteeQTjr//l9U1Q65BflyT9Kt1Rc2NgC3u341zXEcQGlBQGy8GDA3l4B7zeyVNm929RScn0Z/+y+4HP6r8e8WbrL64jL2PfMLAN/5H/G9byN4XUzlloOoNeLRvzYRlLzX75ASgKCkDWaNBUfRmj1H1Bkozcu2SoMhaDeE3TSL8pkk2P3cFSZLw69mp2khKTc4di7dq54xzR+KabYLSff7lxH610tgDxookpTT9HJl7ogka3sMB0TUOIkERBBPumG9ir6Tzv4+eGM6MZrCpXV0Zyso58MxXFo859uavXJ30K1n7Yjm7fj+K3kDr4T1oO3lglamF5sw10LfmpluShEurljc6XFOzsNoeZ2uqotj959SzQxBTN7zNhrnPUpycadVr9IXme1A1RyJBEQQTqu2VlFdKYJAnoyaEE9al+X/6t+TsxoMWV4gAlGbmkb71CCETB7SoT3wX63z9BPZbSOQkjUzbaYNx8Wt5HZdbD+9R44aVspO2SiGvvRUmpHPsrd849cNadPnFuIW0IvJ/M+n+wOU4+3ja5ZoBA7pwVfxPHH3tFw4s/MbywRJ2WfXVmIkERRDMaO7LqeuqpuSkQml2892N3BqeHYLodv/lnPjgr+rTGbKMpNXQ/4VbGyK0Bqd1dabXY9dw4JmvTR8gS3T53wxcz/fpsLeco6dZNeZBdIWllXVTJanZHHrxB+J+3sCMbe/bLRZZo6H3k9dz8pvVFJxJBzP7IbWZ0L+yt0xL0TLGWgVBsBnPjsFWHefVqY2dI2n8Br99F72fuA7ZpWpnTq9OwUxd/xat+kU0UGQNr/eC6+h23xzgfBdTjVxZr9TpqrEMfvtuh8Shqiqbr3kRXUFJtaJuVVEoiEtlz8Ome8TYiiTLjPr+STTOWiRN1duypNXg7OfFsI8esGsMjZFYxSO0SPGnstm+KY683FL8WrkzanwYoR39GjqsJkFVVf6KvMXYEdVUEbEs4dM1lLnHvm6R+6uYUpZbSPLK3egKivGNbE/Q6N7ie3NebnQip75dQ1FyJm6tfQm7aZJDE7e0/46wasxDFo+RtBquTf3d7iM6OYfjOPTi9yT+vQNVUZCdtYTdMJG+z97c5EZPbHH/FgmK0KLodQY+e3c7e7YnIGskVEVFkiUUg8rYyRHccudgZI0YWKzJ2U0HWTPlCVDUKoWgkiwjaWSmrHuT4NG9GzBCQbBO1Ht/sueRTywuBweYtmkxwWP6OCQmXUExZecKcA3wQevu6pBr2pot7t/inVhoUX75dj97dyQAoBhUVNX4O8DmtSf565fDJCecIzU5r3KvEqG6NuP6MW3j2wQM7lrl8YAhkUzdtFgkJ0KToXF2smrJ86XTdPbk5OWOZ/ugJpuc2IoYQRFajIL8Uh6Y9ycGKxMP/wB3ZlzekwnTuojheAvy41IpOZuNW5tWeIc17/1TGjNVVcnYfoyT36w2Tpe08SfshomETOjfYpZ210XBmTT+CLvBYpLiEuDDNcm/NppdfpsC0ahNcIiKHiBNXdThs1YnJwA5WcX88Pke0lPzueH/HLfc0dZUVaWsVI8sSzi72P6/vHdYiEhMGpii07P15teI/3UTklaDqjcgaTXEfb+OkMkDmfDXCy3+07g5Xh2D6XjVWBL+3Gq2YVrPh68SyUkDEAmKYFJ+bgmr/znBlnWnKCwow8PTmVETwpk2uxu+/tbtIdHY6MrrNmWzdnk0w8d2plO4bdqDO4qiqGxZd5LV/5wgLcW45DesSwDTL+/BwKHtGzg6wZYOLPyG+N82A1SuRKn4/ez6A+y8731Gff14Q4XX6I388lHKsvI4u/GgyQ2GYz5fTvDYPrQe2r1B4mupxBSPUE12ZhEvLVhF3rnSKq3eZVnC09uFha9NpXVw02sulXA6h2cfXlHr18kaiTETw7n17qF2iMq2CvJLOX4kDV25gb07Ezi0N6XKG64kGwuDr7ihL5dd1atBYxVsQ1dYwi/BV2AoLjN7jKSRuTrp12qb/gkXqKrKqvGPkL7lcLXnJFlGdnFi1p6Prd5vqKUTRbKCXXz90c5qyQkYP5EX5pfxxfs7Giiy+unQ2R+tU+1/5BWDWjkC0VjpdAa++3Q3D8z7k4/f+o8v3t9hTE6gyqdB9fy/6Z8/HSIxPqcBIhVsLWNHlMXkBIx7JZ1dv99BETVNOYfjTCYnYOyHouh0HH7lRwdH1bKJBEWoIv1sAccOnTW7SZ6iqMQezyAlKdexgdlIcEgdMnkJnJwb78aAqqry8Vv/sWlNrNU1NrIssXHNSTtHJjiCUq6z6jhDuflNCwU4/fOGykZxpqh6hTN/bMVQVu7AqFo2UYMiVGHtp+qE0zm0DfW12XXz80rZtCaWXVvPUFKiI6SdN+OndaX/4FCbFuiOnhDOL9/so1YTmyocO3SWvTsSGDS8g81isZWT0Zkc2J1Uq9coikrC6ZY3gqKqKpm7T3D6pw2UZufh2SGYiHlT8OnSdPc48e8XDrJUYx+P5rorsK2UZdU8SqrqDegKStC4ODsgIkEkKEIVWq11g2pOTrYbUUhOzOW1Z9ZSWFBWmTjknSsh6nAag0d04O6HR9qsedrI8WGsXBpFfl71KSxLFMU4SvHCYm/aN7KOs9s2xiFrpMp+LtZycWm8o0L2oC8pY/M1L5G0fKfxk7KqgiRx9PVf6Pno1Qx8/X9Ncjm5R9tAOsweQeI/O0yuQpG0GgIHR+LXs1MDRNd0eHYMoqZPLloPV5x97bNxoFCdmOIRqojsGYRzDdMZWq1Mt17W7cdSE8Wg8O4rmygqLK/y3lCRPOzZkcCqv0/Y5FoAHp7OPPnKZFoHG99kNBoJjcZ4U6rx3iTBuuW2i8VW8vNKa52cAAxoYSt5dtz1DskrdwHGT8KqQalc6XLsrd84/t5fDRlenZSk53Do5R8pTEivtocLGItjXQN8GP3Dk3a5vqFcx+klG1k/dyErxz7E9jsXk7U/1i7XsrfwW6ZU1miZImlkIuZNQ7YwDSTYlhhBsbHC/DK2bY4jJTEPF1ctA4aEEtkzqMl8MnNzd2bijEhWLYsy+WFCkmDslAg8vVxscr0jB1PJTC80f4AKa/49wbTZ3Ww2ihIc4s2iD2cTdfgsJ46lgQpdurdmxV/HiD2eafZ1ikHl4J5km8RgS37+7rUaQalYjTVyXGc7R9Z4FCZlEPfjeoufkI+89jPd7p2N7NQ03hbTtx1l7fQnMRSXXdhuoGKqR5Zwa+1LxLxpdJ8/F7eg2q3eUVW1xves4rQc1kx8lNzjCSDLoChk7Igi9osV9HjwCga9fXeTed8D4+7TfZ65kcMv/VDtOUmrwS3Ij95PXd8AkbVcTeN/YhOxfdNpvv5oJwaDgiRLSMC65dGEdQngoWfG4eXdNBolXXFDX85lF7Fz65nKG1/F7/2HtOe6WwfY7FqxxzPQaCQMFm6ueedKyMossunSZlmW6NUvhF79LjQY++e3ozW+rjaN3hxl5PgwNq6u+VNrxb+hj58bjz0/ATf3ljOPnrxyNzX1My/NyCX7wEkCh3RzTFD1UHaugHUzn0JfXFq19qTiz4rKmJ+foc3YvlafU9Hpifl8OSc+WkZebDIaF2c6XD6SXo9eg3+fsCrHqqrKxsufIy/2fMJ+PkGqGJGKevdPvCPaEXn3ZXX+Gh1JVVVOfrWSuB/XVX9Slmg/axhD3rtPLNN2MJGg2MjxI2f5/L3tlX9XL7rhxp/K5p2XN7Hw9al1/kRRUlxOzPEM9HqFDp38CQyy3zyoVitz50MjmTSzG9s2xpGbU4yPnxsjx4UR1jXAtp+KrDyVuUsaDArnsouRNTJ+/m71ii08MpD4U9lma1NkWSKsS2Cdz28vnSNaMWx0R3b9d6baAIEsS7i6O9GrTxtc3Jzo2bcNA4aEorVhDVFTYCgpN/4Q1VBjYChtGis0Tn27Bl1BidmvR9JqOLb4d6sTFEO5jvWzniZ1/QFABRUMJWXE/7qZM79tYcKyF2k3bUjl8Zm7T5C567jFcx55Ywld75zZJNrs73/yS46+saT6+5EkETCoK2N+floUxjYAkaDYyD9/HEOWJZM3N0VRiYvNIiYqg8ietdsyW69X+P2Hg2xYFYOu3FD5eJ8BbZl371D87NTVVZIkwroEENYlwC7nr9CtZzAr/oyyeIx/gDutAqsmZHqdgeV/RbF+RTQF+cYeEMFtvZl1RU9GjOtcp0Rl/JQurP3XfI2JoqhMmlG7lRDxp7I5ciAFg16hY3gr+gxoi8bGuyVLksQdD4zAP9CDdcujKS+78HPSo28bbrtnKP4BHja9ZlPj36dzjatcJI2MT7fGUZdTEH+W0qw8PNoF4t6megfj1I0HLCZbqt7A2Q0Hrb5e1Dt/kLqh+jlVvQFVkth0zUtcm/o7Tp5uACT+u7NyWsecooR08uNS8YloZ3UcDSHncJwxOYHqg2yqStaeGGK/XEm3e+c4OrQWTyQoNlBaouPEkTSLx2g0Evt3JdYqQVFVlc/e2cbeHQnV3ouOHkzlpSdW8+LbM/D0tk09SEPo0acNwW29yThbYHbkYupl3assNdbrFRa/vJHjR9KqfF/SUvP54v0dpJ3N58ob+tU6luC23tx691C++XhXlWSz4s/T5nSn9wDr9pzJzyvlwze2EBOVgSxLSBIYDCp+rdy4/4mxNk/8NBqZq2/qz6wrexEblYFOb6BDJz8Cg5pex19bUwwK2X5B5AwchOFsFn5pSWgNVXuCSBqZDleMwq11w67QOrvpIPue/JKsPdHnA4O2UwYx6I07q6zCsVTMedFBVl1TVRROfLDUfAKnquiLSjj98wa63DGDo68vIeqt3ywmJ5Uv1RtqPKahxXyxonL/ItNUTnz8j0hQGoBIUGygvNy6/4RlZbVrlHTyRCZ7tieYfE5RVHKyi1m3Ipq51/Wp1Xkbg1MxmaxbHs3xo2koBgWNVka56PtYkRSMmhDGpJlVRy3+23CKqMMmEsLz76///n6MwSM61mk58NjJEYSE+rD67+McO3gWRVEI6xrIlFnd6De4nVUjMwaDwpvPryc5IRegSuKVd66U159dx0vvzCSoje2TBzc3J/oMbFvn1yuKSnFROU7OGlzssLGgox07lMpXH+4kJ6sYQnpACMh6HR1jDxN66phxRF+S8GgXyJB376v39QryS/lvQxyxxzNAgu69ghkxLgwPz5qnB5JW7mbD7GeosvuICqnr9rN8+P3M2PYe/r2NtSBBw3uSumav2URF0si0Ht7TqpjLsvMpTs22eIyk0ZC1P5bi1GwOvfi9Ved18vHAq3Mbq45tSHkxiZYTKRUK4lIcF5BQqem/AzUCnp7OeHq7UJhvvt20oqi06+Bbq/P+tzHO7LQRGD9FbV57ssklKOtXxvDD53uqrDypuO8HtPbAyUlDSKgP46d2oUefNqQk5rJ+ZQzHDp1FVVWKiy13zpRlic1rT3Lz/wbXKb4u3VrTpVvrOr0W4NDeZBLjz5l8TlFUdOUG1vx7os7x2UN5mZ5Vy46zflUM+bmlAPToE8ysK3vZbEm5o0VHpfP2ixu5dLsxRevE6e4DUWSZjrFHQFXx7dkJt6D6jZ4cO5TKe4s2oys3VI7sHdybzJ8/H+LhZ8bTtYf50VNFb2D7HW8ZE45Lp1kMCoaSMnbN/5Dpm98BIOL2aRx6+QfUcp3J2l/VoND9gcutilt2tvI2oKgcfvUnqw6VZJnIuy5rEnUbzr5eSLJ8YSWUCdrzU1uCYzX+6qUmQNbITJjW1eKna61WZsTY2i3rPJddXGMzsdxzJbU6p7WyM4tYuuQwn72zjR8+30Ps8Yxqb/R1cSYumx8+3wNQZVlsxamzM4u497HRzF8wlp59Q9ix+TTPPLiczWtPkpleSFZGEcWFlgsZFUUlNTG33rHW1e7tCRa73yqKys4t8Q6MyLLyMj2vP7uOpUsOVyYnACeOpvP6s+vYseU0YCzUzkgroLjIdoWkqqpSEH+WvJgk9DYuUP3tuwOoqmq2VCOhSx90TsYbaPKKXaSuq/teNRlpBbz7yqYqyQkAKpSV6nn7pY2cyyk2+/qUtfsoOZtjtq5ENSikbz1CflwqAO7B/oz77VlkrbZKe/aKP/d+8nrazxpuVezOPp4EDu1mXKJshqo3ILs4mWwEZ0rrkT3p++xNVh3b0DpdNcZiciJpNYRdP9GBEQkVxAiKjcyY250j+1M4czqnyrCrLEuoqsrt9w3Hw7N2tSI+fm7WLDwgJSnXpm3nV/wVxe8/HLiQcEnGUY/uvYOZ/+RY3Nyc6nzu9StiLI4KSZLEhlUx3Hr3UGKi0vj8vdpvTChJNOgS2uLCshoTy9IS6/ZPcYRVy44TF5tV7ees4mv48v0d7NmWwOH9KSiKiiQZm7xdfn2fev3cnfxuDUde/Yn8k8bhcydvd7reMZO+z9+Mk0f9PrFmphcQF5tl8RhVkskM6UhIQiySVib265W0nTywTtfbsDIGg8F0MqSqxmngzWvMj3YWxKVatcqoMP4s3mHGOqj2s4Yz+/AXnPhwGYn/7EAp19F6WA+63TeHkAn9axV/7wXXs2HOQpPPSVoNPpHtcfH3QtLKqDrLU9q9F1xH3+dubhKjJwDt54zAt0dH8mKSqk31SBoZjaszPR60bjRKsC0xgmIjLq5OPPnyJGZf3Qtvnwv9Trr3DmbBS5MZNqb2baZHjutcY3IiSbDmH9t1N92x5TS/fX8AVTXeoBRFrRzpiD6WzmfvbKvX+U8cS7N481YUlRNH09HrFd5btKVO11BVGDSi4VZjBLf1QdZYrlWx1zLx8nIDSWfOkZKUi2LFp11FUVm/Ksbiz5nBoHJoX3Llv5uqwoHdSbzw2CrOxFmuXTDn4Avfs23eG+SfujC3r8svJurdP1g9/hFjf496yM+zvLsvgKSqlDsb/6+qeoWCU6l1vt6BPUkWf65VReXAHvP7Jbn4edb8SQSqtVn3jWzPsA/nc03iEq5L+5MJS1+sdXIC0P6y4QxefA9IUmVH2orfvTq3YdKKV/FoF1hz0asE3eZf3mSSEwCNsxNT1r1JwMCugDEhk84vw3dt7cfUdW/i1dm64njBtlrkCMqZuGySE3JxcdXSo08b3D1s85/JxdWJudf2YfbVvSkuLMfJWcbFte6jDZE9g/D2cSU/z/ybtarC3h2J3HbvsDpf58K5VJYtOWL2eUUxdlKtz4iNNd1gZY3EwT1JFNUwlWPytbJE62DPBt3Ub8zEcNYtjzb7vCTB+GldbHrN8nIDy5YcZsOq2MrRGR8/N6bP7c7kmd3MTjkVF5VXmdYxx9Toiq7cwFcf7OSld2fWKta8mCQOvfDd+RNfch2DQtb+WE58/A+9Hr26Vue9mF+rmpffq5KEa2mR8S+yhGs9alD0+pqTQZ3O/DGhs4ahcXXCUGp+ZM2zUzCt+kfUKT5r9HjwCtpfNoyYL1aSe/wMWg9XOswZSfs5I9A4O9Hx6rHseuBDlDLTMUoamZDJA5tkMzP3YH9mbH+frD3RJK/cjaFcT+DgSEJnDROt7RuQQxKUjz76iDfffJO0tDT69OnDBx98wODBji8QTDpzji/e31FlF1cnJ5nJs7pxxQ19bdafoqKVeH1JkkRIqI/FBAVAV8OQq7XSzxaQfrbA4jGyLHFwT3KdE5Te/ULYmB5rsRla734h7NuZaPU5jf9uKgaDSrv2vjz0zDibbmZYW6Ed/Zg+twcrl1bv7yLJEp3C/Bk/xXYJil5n4O0XNxATVbVOKO9cCb98vZ/UpDzm3TPUZI2UUw37LlmiKCqJZ85xJi6bjmHVe3WYE/vlSsvLOhWV6I//rleC4t/Kne69g4k+lm7+Z82gJyA1ofKa4TdOqvP1OkcEkJtjfhRF1kiEdzW/tNzZx5NeT1x/IXEzYcArt9u96ZlX5xAGLvo/k8+5+HoycNEd7Hn442rPSRoZ2dnJ7GubAkmSCBzSrUl0Em4p7D7F8+uvv/Lwww/z3HPPceDAAfr06cOUKVPIyMiw96WrSD+bzytPriHpTNXVFTqdwoq/ovju090OjcdaHTr7Wyy4lCRoG+pjk2tZs1xakqxfVm3KxBldazi/xPhpXSm18EnyYt16BTFpRlemzu7Ogpcm8eI7MxpFU7Krb+7HrXcPIaD1hVhc3bRMmdWNBS9Nwvn8El6DQSH+VDaxJzIoLKh5WsKUbZtPE30s3WwR85Z1pzgZbXqPIRcXLT36BFv8GavJ2eSat6m/WN7J5BqnCgrPpFksXLTGdfMGoNXK1b+289+nsKh9aA16JI2Mf58wOlwxqs7XmjQz0vLUpUFlwjTLP/t9F95In6dvNO4FJEmVBa9aD1dGfP4wna8dX+f4bKXHg1cw/LOHq402teofwfT/3q1cBi0ItmD3EZTFixdzxx13MG/ePAA+/fRTVqxYwddff82CBQvsfflK//x+lLIyvdk3kS3rTjH1su6E2OhmbytjJ0dYrDFRVWrd3dSc1kGeODlrqnSsvZTBoBJay+XSF2vT1od7Hh3FJ2//V1nnAsaRE0mWuO+x0QS18SKkrQ+H96XUOC1/w+2DCK1DvxN7kySJMRPD8fF1ZefWM+jKDXTp3poxk8JxcXVCVVXW/hvN8j+PVY6QaTQyQ0d35Pp5A2s1ArdxVazF+kpJgq8+2MHQ0Z0YPqYTQW28qzw/68peHK+h0aAlrm61extx9vFA0sgWV4Ro3V3rPVrQvpM/Ty+awvef7alSMOtSWkyn6AMEJxtXJ4VMHsjo7xegca7HdGyPIGZf3Yu/fztqssnf1Tf3p1O45VEmSZbp/9I8us+fy+klm8jcdRyNuwsdrxhd5+Jde+h6xwwi5k0lY0cU5bmFeIWF4NejY0OHJTRDkmqLtaNmlJeX4+7uzh9//MGcOXMqH7/lllvIzc3l77//rnJ8WVkZZWUXPkXm5+cTGhpKXl4e3t5V31RrQ68zcOd1SyzOE8uyxPS5Pbjqptp3ILW3lUuj+PW7A9VvQhIMGBLKfY+NttlOv99+upsta0+aTOQkCTy9XXj3qyvRaqter7xMz/7dSWRlFOLl7crAoe0t3mQz0wv46+fDxm6wQHjXQK65pV/lzTMtNZ8n7vnb7OsBho3pyF0P1f1Trz1lZxbx5gvrOZucbyyYVY01PlonDXc9NJLYExkmE09Zlghq48XC16dZ1dxLUVTuvO6XKu3tzam4WY6f2oWb7hhU5Wdmx5bTfPXhTgx640aXYPzUL0mSxeXlLq5aPvjuqlo1dUv8dwcbZpteMQLGIsXwWyYz8otHrT5nTVKT8shIL8DD0xmfgnNk7TpurJmY0B+frqE2u87BPUms/ucEJ09kgCTRrWcQU2d3r7IppSWqqnLig6UceO5bdHlFlY97dgxixOePEDLRdht1CoI95efn4+PjU6/7t11HULKysjAYDAQFVW1QFBQURHR09SLCRYsW8cILL9g8juJiXc1FbBLk5dqnp0h9TZ/bg+AQb1YsjeLU+aH6wCBPJs+KZOK0rjZLTgCuvKEv0cfSSE+t2nq+IjnSajV8uvg/xk/tQrdewUiSxPbNp/nh8z2UFOsqm699/9keZlzeg7nX9ak2xJ6fW8Jn727n5InMyjbw+3clEheTyf0LxhDeNZDgEG+uuqkfv/9gej+RTuH+3H6f+T4PikEhOTEXnU4hpJ23Q5cd6/UKrz+7jsz0wvOxXPg+6nUGPnxji9nRDkVRSTtbwNp/T9TYgO9kdAafvbPdquSk4twAG1fH4uqm5ZpbLtzsho/pTO9+bdm2OY6UhFycXbT0HxJK/Klss/8GALOu6FnrjrPtpg+hVf8Icg7HVR9FkSVkrYaej9S9/sSUkFCfi0ZHW9N6kOXplrrqNziUfoPrnvAce/t39j3+WbXHCxMzWDv9Saaue5PgMU2rMaMg1JVdR1BSU1Np27YtO3bsYNiwC6tMHn/8cbZs2cLu3VXrPuw2gqJXuOu6XyxW0cuyxKwre3L59X3rfJ1LJcbnsG5FDMePnEWSJHr2bcPE6V1p16HuUxJlpToMBhU3dyfb7ip8kaLCclYti2Lj6liTK2kqPolPmNaF7r2D+eD1rWbPNfua3lx+0Y3WYFB4/pGVJCfmVhulkWQJZycNL707o3IkZfe2Myz/4xiJ52uHPL1dmDitKzOu6ImziQJPVVVZtyKaFX9FkZtjTDidnGRGjOvM1TcPsGpUor727kjgwzfMf0+s6W3j4+vK+99eZfb5xDPnePHxVeh1BmtWp1aj1cq8/+2V1XrznInLZsu6U6SfzcfTy4XBIztyOjbLWPArSciShKIau51WjDjW5eewNCuPjVc8T/p/R4y1FpKEqtPj2tqXcb8/R/Co3rX/opq48vwilgRfaX5HZVkiYGBXZu36yLGBCUIdNPoRlICAADQaDenp6VUeT09PJzi4evtsFxcXXFxsv/GdViszfGxn/tsQZ7YGRVFURoyrXadXS7asO8k3H+9Cki7MR29Zd4rNa09yxwMjat1VtkJ9li1by8PTmStv7Mfsa3rz2J1LyT1XUuUmWPH1bFgVy94dicYtys3cJFf8eYwpsyIrb4QH9yRXJhuXUhUVnc7Amn9OcPOdxq3dh4zsyJCRHcnLLUGvU/D1d7O42mrJN/tZfcnUiU6nsHV9HKeis3jm9an1ajRnjf27kyxvUWBFQpGXW4rBoJj9Wv/+9QgGvVKn5ASMSfuRA6kMG93pfEwqP325l3UXNdKTZYnd2xLoGObPS+/O5Mj+VHJzivH1d2PY6E71KkZ2DfBh+pZ3yNwbTfKK3RjKdLTqH0GHOSOMRaINrLzcgF5nsOsHgUslLN2GoczC0npFJWtPNPmnUvAOr/t+S4LQVNj1ncDZ2ZkBAwawYcOGyhoURVHYsGED991X/425auOyq3qxf1cSxUXlJm8ck2dGViserKvE+By++XgX6vm6gwoV1/3ive10CmvV6ApyL3V4XwrncixPe9W0BFqvVziwO5lRE4zV/Xu2n7F481YUlR1b4isTlAo+vjV3Fk1OOFctObn4vClJuaxfHs2sq3rVeK760JUZjKMM9eDiojW7sqa0RMf+XYl1Tk4qlJVe2Lxy3Ypo1q2IAS78nFb8nhh/jl++3s/jL1jX7ltVVTJ2HqcoMR3XAB+Cx/Y120sicFAkgYNsU+RtCyeOprH8r2McO3gWAD9/NybNjGTyrG52X7pekpaDpLG0q+7549LPiQRFaBHs/lHl4Ycf5pZbbmHgwIEMHjyYd999l6KiospVPY4S0NqTha9P5ZuPdhIddWGJs5u7E9Pn9mDmFdbt/GmNdStiLBYXVrRzv6kRbRZnSvSxNDQaCYOh7ndCSZYoKrwwbWcuQbzYxTfO2tiy/lSNIxcbV8fWmKCoqkp5mR4nZy05WUWsWxHD7m1nKCvV0zbUhwnTujJkZAeTtT/6iiLTeiQPsiwxYlxns5/ci4t19U5OgMoEWTEoLP+zes+WCoqiEnX4LElnztW4Yip1/X523PNula6srq19Gfj6/4i4ZUr9g7aj7ZtO8/l726skhudySvj9h4Mc2Z/Co89PtGuS4tE2oOZOrYB7W/P9VAShObF7gnLNNdeQmZnJs88+S1paGn379mX16tXVCmcdITjEmydfmUJaSj4pScZCwK7dW1f2pLCV40fO1tjOPaoeyzodxRY3QVVRq7R1Dw7xJuqw5Xb3gcF1awOfkVZYY/KTk12Mqqomb/75eaWsXBrFlnUnKS7SodXKKIpaZcO5U7FZnIzOZN+uRO59dFSVJCXxzDkWv7SRc9nmN4UDQIKgNl5kpBVW2bcJjMmJs4uWaXO6m325p5cLTk6yxZoqMF/rUrFSKCIyEIDU5Dzyath0UpYljhxItZignN10kLXTn6zWv6Q0I5dt895A1enp8n8zLF6noeTnlvDVRzsBqv0MqSrEHM9g7b/RzLi8h91iaD93JFoPV/RFpkclJY1M6+E98OrYNHeXFoTacshePPfddx8JCQmUlZWxe/duhgwZUvOL7Ci4rTcDhranV78QmycngFVz1g6a1q6XLt1bWxw9kSQrvg4JevRpU/nXsZMjLCYRkgQTptZ+hYVeZ6CkuObW+K5upmsKzuUU8/yjK1nzzwmKi4xN4vR65XyCcuG4ioRi365E1l7Uzj4/r5TXFq6t8UYvSdCrbwjPvT6NgUNDKx+r+NQeEOTJU69MpnWwl9lzODtrGDamc43N1dzcnasdI8sSWq3MHQ+MqPw+WNOm3Xic+U/3qqqy+6GPjcmJmX/fPY99hr6kbs3o7O2/jact7l2kqrB+ZbRNdvQ2x8nDjUFv3mX6SdnYuM3s84LQDDV8NVoz1LNvG7asO2WxnXvPi27ajdWAIaH4+rmRn1dq8mux6r1aheNH0+h/fulluw5+zLiiBytMTClIskRYRCvGTa1dG/j8vFLeeHYdSQm5Fo+TZYkRY01v2vjD53s5l11c4whMJRXW/HuCybOM+9xsXnuS4sJyi98Td3cnLru6N5NmRqLVytz3+Bgy0go4vD8FvU6hQ2e/yqXbNZlzTW8O7U2msMD0zskDXAsISowj2q8jCVpfDIrx6x84rD2zr+lNu/a+lccGh3jj7KKxuFxZUVQ6R5ifWsg9nsC5I6ctxqzLKyJ5xS46Xjmmxq/P0ZITztXY8yUnq5iyUj2udiyyjrxrFhpXJ/Y//RUlZy9syeHXsxPDPn6QwMGNp15HAH1xKTGfryD6s38oSkjH2deTsBsn0f2By/FoG9jQ4TV5IkGxg4nTu7J57UmLx9TU9rox0DppeHjheF5buI6SEl3l6EFFnUevvm04euisxXNIEmSd7wdS4aob+9E62Ivlfxyr7BXi5u7EuMkRzLmuj8nlw5Z8/NZWUpLyLB5jnDrRMHV29amT3JxiDuyufdFpTlYx+bkl+Pq7s/u/MzW+PiDIs9rUTetgrzp1Am4V6MGzb0zl+8/3cGT/hXoPN9lAu0N78Eo+RbHeQHv5MG0lGbceYUz/+wX8OrSudi5XNyfGTAxnwyrTeyTJskSrQI8qI2GXKjlrxa7GskRxat12P7Y3J2eNcTVaDS5tUGgPEbdOJezGSWRsP0Zpdj5enYLx7xvusNVEzVnWvhhO/bCO0oxzuIcEEH7rFPx71W1Fpa6gmFXjHyH7wEmMnRihJO0cUe/8wcmvVzFty7uiw249iQTFDtp18OOO+SP44v3tVZYZVwy33/XwSILb2mbFkL116OzPax9dxsbVsezZlkBpqY52HXyZMK0rvn5uNSYoqgoeXlWXjkuSxNhJEYyZGE5GWiEGvUJAkGetExMw1n2cOJpe43H+Ae7c9/gYk1MnKUl5da63qfg3tWbvoIpdhm0lMMiLRxZOIDuziLTUfFKWbiXppS+Qq6wJV9FgQHc8jh03vsKM/94zea4rb+zHyehMEk7nVPleyLKEi6uW+58YY3FKya2NFZsFKiruIdZvKuhI/YeEsmXdKbPPy7JEz35t0DpoE0pZqxEN2WzIUK7jv1teJ/7XTca+O4oKskTUO3/Q5X8zGPbRA8ia2v3b7lvwBTmHTlUbSlYNCuV5RWy66nnmRn0jEst6EAmKnYwY15lO4a3YsCqGqCNpSBL07NOGCdO6NpnkpIKPrxtzr+3D3GurvmGqqkrrNl5k1LADsre3q8nHJclYrFkfxw6lIslStWLTS933+Bize6HUZUdfSYI27Xzw8jF+be06+JGTZX6KSJalejXos6RVoAd+fq4cnfZH1eTkIqreQMb2Y2TtiyFgYPXRO1c3J556dQqbVseycXUsWRlFuLk7MXxsJ6bM6kZAa8uFy77dO+DfJ4yco6fN1qA4+XjQbuYwk881tN79QmjXwZfUpDwz05kqMy+33Uo/wbH2Pvop8b9vBriwUup8yVHsFytwD25Fv+dvsfp8uoJiTn6z2ux+UqpBIS86ibQth2kztm89Im/ZRIJiRyGhPg22lFhVVU7FZLJ7WwIlReUEhXgzcnwY/q3cbXYNSZK45pb+fPDaFovHfffpbt74dE69dsw1R1VUS33iLhx3yY27rExPUWE5np7OdA5vhae3C4X51hdwqirMuLxH5aej8VO7cGhvstnjK/bAsZe86ERK0nIsHiNpZFLX7TeZoICx98rU2d1NToPVRJIkBr19N2unPI4qYbJAadDr/0Pr6rgtB2pD1sg8+twE3nphA8kJuWg0UmUfI41G5v/uH07XHo5feehoqqqSsf0YaVuPgApBo3sRNLJXkx4FKM3KI+az5WYTZ1Q49s7v9Hr8GrTupj9MXSr3RKL5jr/nSRqZrD3RIkGpB5GgNEMlJTo+eG0zUYfTLrzRAn/9fIhrbx1QpxuQOR07+9d4TGZGISeOplmsYairzhEBNRa2OrtoCGln7PmRmpzHsiWH2bsjEUVR0Wplho3uxNhJESz/81iN16uov5lxeY8q3YB79w9h9MRwtq6vOk1QsdR3zKRweva1X2G0pZ2BLw5GsaLPRl2FjO/HpBWL2HnPuxScvjD15xrow8DX/kfEvKl2u7Yt+Pm789I7Mzl6MJWDe5PRlelp18GPUePDarW7dFNVcCaNjZc/S86hOKTzy+dVg4Jf785MWPoiXp0af2G/Kanr96PoLPdW0heUkL7tmNW7RkvW1CKpaqPoityUie9eM/Tp29sq6zIuXSb8yzf78fV3Y+go06tZaistNb/GYyQJzqbk2yVBiewZRJt23tU2N6y8tmysd3F1c+JMXDavPr0WXbmh8li9XmH75tO4uTsxcnxntm08XTnSI0nG719oR188PF0oK9UT2tGXcVO6VFvRIkkS8+4ZSsfO/qz6+3hl8W9gkCfT5vRg3JQIu34K9e7SDq2XG/oC88ucVb2BwKHd7BYDQNvJA7ni5A9k7IiiMCEd10Af2oztS865UjasikGvU2jfyY/InkFmvx+qqpK6dh8xny8nNyYJF38vwq6fSNiNE3HyrLmjcH3IskSfAW3pM6BldWotzy9i1diHKouYL054c48nsHLMQ8w58iUuvnXrUdSQahrpqO1xAP69OuMa6ENppvnifFVRaTvFuoRHME0kKM1McsI5Du0zP9UgSfD3r0cZMrKjTW6YrlbsDaSq4Opqnx81SZKY/8RYXn16DUWFF3WpPT/vExbRiitv7Iuqqnzx3o4qyUkFRVEpKdaRlV7Iax9dxn8b4sjKLMLb24VhYzrTOaKVVd8rWZaYML0r46d1qdwCwNvH1SHD41o3FyL/N4uod/6o1igNjMPNnh2DCZnQ3+6xSJJE0IieBI3oSVmZns8+2Fm5yqliRCk4xIu7HxlFx7CqdUGKwcDWmxYRv2QTkkY23igliYztxzj65hKmbX4Hz9DqK5GE+jn5zWqKkjJNTs2pegPFKVmc/HoVPR82v4FlY+XfJ6zmgyTw62X9hzbZSUvPR65m34IvTJ9OKxMyoT++3TtafU6hOoc0ahMc5+CeZIu1HqpqnObIvGTpb111imiFj6/leVuNRqbvwHY2uZ4pIaE+vPzuTGZc0QP/AHdcXbWEtvfl5jsH88RLk3FxdeL0ySyTOyhXUBSV6KgMJCSuvrk/9zwyihvvGExYl4BaJxiSJOHj64aPr5vFUQKdzmDTxl/9XryV1iN6GJOzi64raWScvD0Y/9cLSLLj/surqsoHr29h97aEyvtexe8ZaYUsemYt6ZcUWB99fQnxv242HlvxKf78HGVRUiYbL3/Ors3SWqrTP2+oobGRStxP6x0WT9b+WGI+X87Jb1dTnJpVr3O16hdBqwFdKqetLiVpZEImDaz1FFbPR6+myx3GzsjS+b2mKq7h3yecMT89XY+oBRAjKM1OWbneqi61ZaV6CvJL0Tpp6rW7r0YjM/e6Pnz7yW6zx0ya2dXuc/i+/u5ceUM/rryhn8nna+qTUiE1Oc+uq6zyc0tY9fcJY2O3onJcXLWMHB/GjLk9aBVY992BwTiKMmXtG5z8dg0xn/5Lflwqzj4ehN80iW73zcE9xLF7uJw8kcnRA6kmn1MUlfIyA6uWRXHr3UONj+n0RL37p9kbpao3kL0/lsxdx2k9zH4t51uisuwapmpVKM+xvFrPFvJikthy46tk74+tfEySJTrfOInhHz9gdRHrpUZ/v4AVIx9AV1CEelHnZEkj49rajxGfP1zrc0qyzPBPHyJi3lROfrWK/FMpuAR4E3b9REJnDTO7QaZgPZGgNDOhHfxq3NxPo5F44/n15OcapyEiIgOZdWUv+gysed49L7eEzWtPcmB3ErpyA53CWzF+Wheuuqkff/50CFVVkWUZVVVRFJVxU7pw9c32n1ZQVZUTR9PYuDqW5IRcXN2cGDyyA6MnhOPp5YKLlVsauNhpKgogJ6uIF59YTd65ksqRnLJSPZtWx7JrazzPLJpa7x2uNS7ORN45i8g7Z9ki5HrZsTUeWSOhmPl5VBSV7ZtOc8tdQ5AkidzoRMqyLCeSkkbm7MZDIkGxMe8uoRQmpJsttpY0Mt5d7DcKClCUksmKUQ9Qfq5qIqQqKqd/XE9p+jkmrVxUpylT324dmH3wM46+8Sunvl2DvrgUJx8Putw2jZ6PXYN7cM3F/qZIkkTrod1pPdR2Cw+EC0SC0sz0HxKKp5cLRYVlZkdsDQa1MjkB4wZ4i1/eyM3/G8yE6eY73MbFZvLm8xsoLdFXDrOfTcln26bTXH59H979+gp2bI4nO6sIL28Xho7qVO8+J9ZQVZXvP9vDxtWxVXYzPhOXzaqlx1nw8iR69g2pcYM9dw9nunS3X33Dd5/urpKcVKiogfn0nW28uLhxbqZXF4X5pTX2pykvN6DTKcYmfdZM3dTQjl6om653ziRl9R6zz6sGha52TnqjFv9B+bkCk0mSqiikrNlbr74inu2DGPbhfIZ+cD+GkjI0bi5Nevl0SyBqUJoZJycNdz8yElmWq9WimPu/WHET+eGLPWRnFpk8prREx9svbqS0VF/lBlFxs/3r58PExWYxbU53bvy/Qcy+urdDkhOAzWtPsnF1bJV4wHi/Kywo4+0XN+LqpmXyZZZXsMy6sidOduoUmp1ZxKF9KRZrYBJO5xB/qnG2gq+LgNaeNd4AKnZmBvDpGopzDatEVL2BoJGmG6YpBgMJf29n/exnWNbn/1gz+XHiflyHocz61RktVftZw+gwd6TpNwlJIvSy4bSfPdyuMZz8do3F5fKSVkPcD+vqfR1JktC6O6Z4XagfkaA0Qz37hvDsG1PpPyS0Mknx8HTGzd1yrYkkSWxZb7rd986t8RQVlpv9RCzLEquWHa9f4HWgqiorLVxXUVSyM4s4vC+FK6/vW9ksTZYlNBoJWZaQJJhxRY9q++TYUnINGxlWSIi33GytKRk9IdxijxpZlqosv9a4ONPt3jlgpshb0mjw7d7BZAt4fWk562c+zca5z5K8cjfnjsZzduNBtt78GsuHz6csp+bl8C2ZJMuMXbKQvs/ejIv/hQ8Wzn5e9F14E+N/f67WreBrQ1XValM71Y7RGyjJyLVbDELjI6Z4mqmOYa24/4kx6HQGysv0ODtr+L+rf7H4GkVVSUnMNfnciaNplUtETb5WUYk9noFiUJDNVMvbQ15uaY2t9jUaiRNH0+g/JJRb7hrC1Nnd2LE5nry8UvxbuTNibOd6F6jWROtk3ffEXiM4DSEk1Iepl3Vj9T8nqj1XsQHhpU0D+yy8kaz9saSs3nNhmTGALOEa6MP4pS+a/OS7f8EXpKzbB1xY/VOx3PrckTj+m/cGE/9+2ZZfXrMjO2np99zN9H7yOvKiE0EFn8hQNC727/4rSRJubfyr7OBc7RitRiwxb2FEgtLMOTlpcHLSGFt2a2UMevNDqLIk4WymmLTRTvtbGdjFRwW18WbudY7diC0iMhBXNyeLGwZqNBK97NhttiFcO28AfgHurPgzqrI3jCxLDB7RgRtuH4jnJRtJapydmPjPy8T/tpmYz5aTfzIZZz9Pwm6YSNf/zcQ1oHoRcXl+ETGfm29lrhoUkpbvJD8uFe+wENt/kc2MxtkJ/95W9A6xsa7/N4PDr/5kfn8bvYGI2xp3N2LBtkSC0kJIksSAIaHs25losQ5iwNBQk8916daavTsSzJ9flugc3sqhoycA3r5uBLT2ICvDdO0MGIuCu9qx+NUazi5aps3pztJfDpt8XpJg9MRwvH3t2ynV0SRJYupl3Zk4PZKE0znodAZC2vng7WN+uais1RB2/QTCrp9g1TWy9sbU3AVUhbTNh0SC0oh1f/AK4n5ab3o1kQTht0wxu4+U0DyJBKUFmT63B/t3JWJqdz1ZlggO8abfINNLCUeM68wfPx2krFRvctBCVVSmXNYNvV7h0N5k4mIykTUyvfqF0LVHa7sVpMmy8Qb445d7zT7v7etKv8GmEy9HOXowlZMnMqqsMgIql+H2HxLKDf83yKExGQzGf6vd285QVFhOUBsvxkyKoMP5/ZVKisvZvjmeA7sTKSsz0CnMn3FTu9A21LfW19JqZcK62KcPi6nOuaaPa6zDgAKAi58XM7a/z8573ydh2bbKETEnb3d6PHAFfZ69qYEjFBxNUhvxmr38/Hx8fHzIy8vD29t+zbNakoN7kvhk8TbKSvXIGmPSoJzfb+bhhRMs7nZ84mgai1/aiF6vVN5kK264U2d3Z9Dw9nzw2hZyz5WgOX9ug0GlfSc/HnxqnN3qPBRF5cv3d7B98+kqCYAkS7i5ObHgpUmVN92GsOzXIyz95XDV2M7X80R0C+S6eQPoHFH7jrX1UZBfylsvbOBMXE5lXBW/T7msG6MmhPHGs+uN0zLnE9qK56+5tT/T5zSePiSl2Xn8GnJ1jRvCzT32lWg93kQUp2aRc+Q0GhcnAod2R+vW/DdrbG5scf8WCUoLVFKiY8fm0ySczsHJSUPfQe3o0aeNxRb5FTLTC9mwKob9uxLPN2oLYOKMrgS18eLpB/6lrMxQ7ZOqLEsEBHnyynuzjP0u7EBVVQ7tTWbDqliSE3NxddMyZGRHxk3pgq9fw02bxESl8+rTay0e89I7M2jfybEJ1GsL1xITlWF2us/N3YmyUr3Z5x96Zpxdty+orW23v8mp79earF+QtBqCRvVi2oa3GyAyQWiZbHH/FlM8LZCbmxMTptVtLjcwyJNrbx3AtbcOqPL4L1/vo9xEcgLGEY6MswXs2X6GkePsU3wnSRL9Boc2+FTOpdavjLHYTVXWSGxYFcu8e4Y6LKaE0zmVu12bU1JsvphXliVWLY1qVAnK4HfuIefoaWOLdEkyTg9IABIeoYGM/n5BQ4coCEItiT4ogk3s/C/eYs8LSYI928wX2TZXJ6MzzSYnYJxeOxmd6cCIjPUw1oyWmVOxsaLewoowR3P29mD61vcY/unDtOoXgWugD749OjLozTuZfeAzPNoGNnSIgiDUkhhBEWyirMTy/L+qQnFRy+voWVGLY4lW69jPCXq9cn50oX4a2+yw1tWZrnfMoOsdzWe7AEFoycQIimATQSHeFndRlmWp3hvhNUX9BodaHK2QZIm+ZlZO2Uun8FYWR3VqIknQrr1vs2oqJwhC4yMSFMEmJkzrYrFnWsXOxi3NxGldjatzTG9xgpNWZuzkCIfG1KtvG1oFephNnCRZQtZI5vduUmHyrEg7RigIgiASFMFGRowLo2ffNtVuahV/nzq7O53CWzk+sAYW3Nab+QvGoNXKVb43kmRs3vbwwvEWl3bbg6yRmb9gDM4uWpMbSnbs7M+9j41Co6m64aR0/s8jx3Vm1IRwh8YsCELL06KXGVc0ojp6MAXFoBIeGciYieH4+jv2htFc6HQGlv95jPUrYyjMLwOgdRsvZsztwZhJ4S1699DccyVsXX+KmGPpIEH33m0YPTEML+/qHVWLCsspLCjD28cFN3f77YOSmV7Imn9PsGPzaUpLdAS09mT8tC6Mn9IFZxctqcl5rF0ezb6dCeh1Cu07+jFxRiSDhrdv0f+WgiDUTPRBqYfTJ7N48/kNxsLN842oJMk4tH33wyMZNLyDTa/XEIoKyziXXYy7p4tDP6Xr9QrZmUVotTL+Ae7iZmalxDPn+OvnQxzam4x6vjHaoGHtufyGvgSH2KcPkGJQOLg3mTNxOXh6u9Crb0iLrBUSBMG2RIJSR4UFZTx21zJKS3Qml8bKssQLi2fQvqOfza7pSJnpBfz+/UH2XrTvTkRkIFfc0JduvYIbOLqWI/dcCWmp+bi4aOnQyc/iPkVxsZksemYdhou69ILxZ9HFVcvC16fWqcW8OYpB4c9fDrFq6XEMlxTMdu8dzD2Pjqoc3VEUlRNH04g6chZVMXa/7TugrcP3XRIEoekQCUodrf77OEu+3W+2qFPWSIwY25n/u3+4za7pKBlpBbzw2CqKi8qr3OiMoxgqDzw5ttE1M2tusjOL+PHLvRzck1T5M+bkrEGSjDtGd+oSwOQZkfQb3A5JklBVlQX3/UN6aoHJpbuyLBHeNYCnF9lmJ1dVVfnwja3s25lo9ph27X14/u0Z5OYUs/jlTaQm5VXZvqBVgAcPPj3W4R1wBUFoGmxx/7bbR6AzZ85w++2306lTJ9zc3AgLC+O5556jvLzhe2Ec2pdsecWJQeXgnmTHBWRDP3+9r1pyAsabkgp8+cFO9DpDwwTXBNU2fz+XU8yLj6+qnKapoCs3UF5moLRUT8yxdN5btJmfvtyLqhobtaWl5Ju9lqKoxJ7I5GxKXn2+lEqH96VYTE4AkhPz2L45jkVPryMtJR8wJiYVoy3ncop5beE6cnOKbRKTIAjCpezWqC06OhpFUfjss88IDw/n2LFj3HHHHRQVFfHWW2/Z67JW0etq7oBpMLGnR2OXe66k2o2xCtU4vXVoXwoDh7V3aGyOVlamZ8OqGDauiiUrowg3dyeGjenEtNndCAzysvja8jI9G1fHsmFVDBlphbi4aBkyqgPT5vQgpJ3l+ox/fjtKfl6pxa66Fc+tWxFD1x5BFtvKX+xsSj5t2ta/PmTj6lirjlv99wmys4pMPqcoKsXFOjaujuXy6/vWOyZBEIRL2W0EZerUqXzzzTdMnjyZzp07c9lll/Hoo4/y119/2euSVgvrEmCxeZYsS3SOaHpLYjPTCyyODIHxa0s/m++YgBpISYmOV59aw2/fHSAzvRBVVSkuKmfT6lieeXA58aeyzb62rFTH68+uY8m3+8lIKzQ+VqZn28bTPPfwCmKizO9ho9MZ2LYxzmJycjFJlljz7wlc3ZysOt7NyuNqkpps3UhMTlaxxY6zqqKyc2u8TWISBEG4lEOr3PLy8vD3b/g563FTu1gculcUlUkzml4jKmuWpCqqavUNsan688eDJMafq5asKYpKeZmBD17fYjaJ+Pu3o8SdzDb5Wp1e4YM3tpidIisqKKO83PrpM1VRiT+ZTa9+bXCqYZdnT28XIiJts5+Mm7t1//6yLEENuZa1oz+CIAi15bAE5dSpU3zwwQfceeedZo8pKysjPz+/yi97CA7x5pa7hgCYbEQ1eVakw9uP20LbUB+C23pb/NQrSxIDhjbf6Z2yUh1b1p8ym4Aoikp2ZhFHD6ZWe06nM7BxdazJHZnBmFAU5JWxf3eSyefd3J0stvs3RZYl3NydmT6nu8XjZl/dC62NWssPHdXJquM6dPa33KZfosYpL0EQhLqqdYKyYMECJEmy+Cs6OrrKa1JSUpg6dSpXXXUVd9xxh9lzL1q0CB8fn8pfoaH2W20ybkoXnnplMn0HtsPZWYNWKxMRGch9j4/m+tsGNsneHZIkceUNfc1+6pUkmDC9K75+bg6Ny5HSUgsoL7M8iiFrJM6YmObJziyqcURAo5E4E5dj8jkXVyf6D7G8906VOGSJXv1DAJhzbR+mXNbNuNJHltBoZGNfHlni8uv62HREb8ykcLx9qjeIu1ifgW2Zc01vi9NVqgrjp7W87QsEQXCMWhfJPvLI/7d33/FRVWkfwH/3TjKT3jskQwpJ6CWQQEAgEEABFxTLWlAUURQrrCg2xFVBYGVXXnbR1QV3wYYFFERKKNEQWhJaSALpvRCSTHqZe94/hgwMmZlMwrTMPN/PJ2rmnjv3mTvXmSfnnvOc5Vi4cKHWNiEhIcr/Li0tRVxcHGJjY/HZZ59p3W/lypVYtmyZ8neZTGbQJCViiC8ihvga7PlNYWysFE8uHYft/z6NtjY5RCIOAgPAFGvhPPRElN6OVZRfg5NJ+WhqbIevnzNipwSrrYxqTCJdVgZmgI1t13a6rCrMGGCrZt9O8x4cjvMpJYpZU93cHhEEhjvnKnpOeJ7Dw0+Owcy7ByE5MQ91tc3w8HTE+MnBek8onZwleOPDGfj4r4eV42w6cZwieX/kqbEQiThMvTNc7aBajgNGje2PaDMqaJh75Sp+23UJqaeLIe8Q0F/qhulzIjExLlTnpJEQYj4MWgelpKQEcXFxiIqKwvbt2yES9ayL2tCl7i1Zc1MbTiYVoKq8AY7OYsRMGABPb0e9PHdbawc++3sSTicXgucVvWaCIIDjOcRNH4iBg3zg4++EkIFeRu+JEuQCli3+ETXXmrW2e//vcxB4SyE+xhj+8sxPuFqpfuZKp3fW3YnQcM3jQbLSK7Bl4x83Bpne8n8Yzytqn9z36EjwPI+mpjb4B7hiTGwQJBKDTazrQhAYLp4tRfrZMsjqWiANcceUGeEqY5QYY0jYdxn7dqUrz4urmx2mz4nErHuGQGQmxdpOJRXgX3/7HcCNWVIcp0gooydI8eyyiVRYjhAjMutCbSUlJZgyZQqkUim+/PJLleTEz0+3aqaUoJinLR//gRO/53dbIySgvyueWDoO4YN8jBSZwsG9mdj+79Nqt/E8h0HD/LBidXyXbQ31rVi++Ee0tHRofO7gMA+sWj+r28RLkAu4eK4MpUV1AAe0NLfj8qVKtLcLCA7zRFVlA1JPFClWDuYUNUbs7W3x5PPjET3BfHolOnWO3WGMwdPb0WwSEwCQ1Tbjlad+REeH5tIATzw3zuirRhNizfTx/W2wP9cOHjyI7OxsZGdno39/1QGnZly8lnSjoqxe56mlZSV1+Ojtg3hzzUyEDPQycGQ3xM+KwNWKBvz2cwZ4EQdBzsDzHASBQRrigef+cofa/RITstHaqjk5AQB3Dwd89o/jaKxvhbevEybFh0Ea0nVmGi/iMXx0Pwwf3a/Lts/+noS06wNtmcDQOWKmubkd/9yQCEeneAwZ4d+zF21gPM/B29fJ1GGo9fvhHK11izgOOPBLBiUohPQxBktQFi5c2O1YFdL3pJ4sUpZn7w5jir+8v/tvKl7/6wwjRKfAcRweenIMJsSF4NjBbFSW18PRSYyYiQMwQssaMqeTCrodN5J6qliZ7PA8h0O/ZiF+dgQefWqsTrezKspkSDqaq7XNj1+fM7sExZzlZasftNyJMaCkqA5yuWBWPT+EEO2Md8ObWISWlnbwPCDXsdyHYqG5Cly72ggPL/2MgdFVULAHFjwdrXP7Vi23dm7WOcah89+H9mbB28dJOeBVm1NJBcoERx3GgOzMKtRea4Kbh/FWoO7LbGz5bpPmzhmGhJC+g/6cID0S0N+1y+q3uqirbTFANPoVFOIBXtS7L7G9P6ZrHQPRqamxXacvyiYjF0Bra5PjVFIB9u26hD8O56Cp0fRrZulqxOh+WqdDK6Zz+9NMHkL6GOpBIT0yOiYQjs5iNDW0dXs75GaubqadfqyLqXeGI/lY70q3y+paUJRfg+Aw7Usk+AY4d7vOk40ND3fP3veeMMaQl12Nq5WNcHQSI3Kor9ZbG8nH8vDfz06hqbFN2btja8tj7gPDMee+oWbf8zAmNgge/3NA7bVmtYmKIDDMmjfEBJERQm4H9aCQHrG1FeGZlybq3J7nOQwa7mf02zu9ET7IB3ddr+jam+9kuQ49KDETB0As0Tzdnuc5jJs0oNfr7mSlV+CNF3/B6lf3YfP6RKxbdQgvL/oBiQnZatufOVGILRv/UPaYdH7Bt7cL+H7HWez9Mb1XcRiTra0Ir62eDtfr9WI63zueVxS6e3LpOAwaptvMQUKI+aAeFNJjI8YoZqecSynpti0v4vDAglFGiEo/Hnx8NAKl7vh1VzqKC2oBAB6eDrhW3aR1PxsbHv46lH23t7fFwiXj8Nk/kpR1OjrxPAcXNzvMf6R35ys7swofvXMIgqCaKMlqW/DFpmS0t8kx7a4I5eOMMXz3ZaraWi2ddn97HtNmRehtoUJD8evngnX/nItTSQU4e6YYbW1ySIM9MGXGQL3V/yGEGBclKKRXcq9c7baNWKL4y9aYU4xvF8dxmBAXgglxIWhsaANjDA6Otnh96c+oqmhQewuB5znETgmBo1P3izUCwIS4EDg6i/HjV+dQkKuYgSIS8YiZKMX9j42GRy9v73y99QwELRVsv/0yFRPjQiCxUyQbhXk1qCir1/qcbW1ynD1VjPGTdVu/x5TEEhtMnBqKiVNDTR0KIUQPKEEhvdLdOAoACJS6I0xPK/Caws0Jx/MrJmHNWwfQ0tyhkqRwHODp44gHF47u0XOPHNMfI8f0x9XKBjQ3tcPDy1HnBEedirJ6ZGdpTxpbWzqQcrIIsZMVS1E01Ld2+7wcp1s7QgjRNxqDQnolJNxL66wInucwsA8nJ7cKCvbAXzfOwdS7wlXW4mEMqCpvwAcr9yNPzQKE3fHycULgAPfbSk4AoPaa9ltQgOI9qb1pCQBdbn0wpkjACCHE2ChBIb0yY/agbla6ZYi707JWuvXycQLHcWhv79p7VF4iw4dv7EdxQY0JIoNygKg2gsDg6najnV+AC8IivbUmms4uEgwf3Q9trR0oLqhBeYlM6/tOCCH6QgkK6ZXhUQGYNe/GSrydOv974XPj4BdgWesnVVXU4+CeTLXbBIGho0PAT9+cN3JUCn4BLgge6Kl19pFYIsLocaqrgy9YPBYiG75LksJxip+HF43B99vT8MLC7/HmS3vw2tLd+MszPyFhXxYtWUEIMSgag0J6heM4PPD4aAwc7IMDv2QgO+uqoiDWqADcOXcQBkYad4FAY0hOzNdaBVYQGFJPFqG5ud2gs17kcgGpJ4uQmJCN6qomuHvYY+LUUDzw6CisX50AQP1A2fsfHdUlrgGhnnh77Z3Y8flpZF2qVD7eL9AN9z4yAr/+dAk5l6+C3fSaq6sa8d9PT6GyrB4PPTnGUC+TEGLlKEEhvcZxHEZHB2J0dGD3jS2ArLa52/oogsDQIGs1WILS2tqBje8fRsaFCmWyVFpUi4tnyxAy0BMvvDYZX/3nDKoqGpT7ODqJMf+RkSpTjG8mDfHAGx/ORFVFPaqrmuDkIkG/QFcc2puFnKwqjbOCfvs5A7FTQtQulkgIIbeLEhRi9spK6pCTdRUcz2HQUF+TFX1z93RAd8MvRCIOzi4Sg8Xw1RdnkHlR0dPR2ZPTmUDk51xDcmIe1m+Zh8uXKnG1shFOzhIMHuEHW1vNxeE6efs6w9vXWfl7wr4sre15nsOxg1fw2DMxvXw1hBCiGSUoxGzVXGvCv/+RhPRz5crHOE5RjXXhc+OMXjxs/OQQ7PxfmsbtPM8heuIA2BkorgZZK/44nKNx7IcgMJw+XoBrV6MQMcQXEbdZ3b2yokHrcgaCwFBWIru9gxBCiAaUoBCz1NjQhg9W7kd1VaPK44wpVgSurmrEyg9maF1jRt88PB1w933D8PPOC1228TwHO3sb3PPn4QY7fvblqm4XJGQMyLpUoax10h3GGDIvViAxIRtXKxvh6maHCVNCMCKqH+ztbbXWQOF47ranRxNCiCaUoBCzdPTAFVytVP8XvCAwXMmsQtqpYowZH2TUuO59eAQcncT4eecFNDbcWPE3LMILC58bB19/w81cYjpO72Xd19ADAHR0CNjy8e84fbxQOZ6F5zmcPl6I8ME+GDtBimMHrmgcFMwEhpiJA3SMnhBCeoYSFGKWEg9la729wPMcfj+cY/QEheM43Dl3MKbNisDlS5VoaW5HQKAr/Pt1vw7PzTp7LooLamErEWFkVD+4eWgvcR880EvrLKJOulbv/WHHWZxJLgRwYzxL57+zM6tg72ADsUSEtlZ5l2PyPId+ga4Y1c0A6caGNhw/loui/BqIxSKMig7E4OF+Zr9CMiHE9ChBsUKdZeqNeXukp2R1zVq3CwLTqXqqodjaijBkhH+v9s3LrsaWj39HeWm9csFAnudwx7RQLHg6WuOAVjd3e0RPlOLUHwUa1wQaOtIfvv7OavZW1dLcjkO/ZmpMAgWB4XxKKV5+Iw5ffnoS1642QSTiwJhiW1ikN15YMQk2NpqvodPHC/Dp35PQ0S5X1lk5uDcL0hAPLH9nqkrROEIIuRUlKFaCMYZTSQXYt/sS8q4oSrIPjPTGXfMGI2qccXshdOHu4YCmxjqN23meg5ePkxEj0o+ighp88MZ+dLTLAdyYgSMIDImHstHU2IbnV0zWuP9jT8egtKgOhXk1KqsQcxzg4+eMp16M1SmOnMtX0dYq19qGMUAma8HfPr0H51NLkZtdDZGIw/DR/RAc5ql13+zMKvxzw+/KREouv5EJFeXXYMPqBKz+22ytVWwJIdaNEhQrwBjDN9tS8dvuSyp1PLKzqvDJ2mOY9+Bw3PPQCNMFqMaUGQPx1X/OaP0L/45pfWfVWrlcwJ7vL+LnnRc0DnRlDDh9vBB52dUaEwBHJzHeXnsnko7m4sj+K6ipboKrmx0mTQ/DHdPCdJ7ZpGu5ekHOwIt4jBzbHyPH9tdpHwD45YeLWo9dmFeDC2mlGBHVT+fnJIRYF0pQrMCl8+X4bfclAFD5wu/8713fnsfwqACEhpvP4n6T4sNw7GA2SovrunyZchwwdGQAho/uG19ugsDwr7/9jjPJhVrH1QAAL+KQnJintYdCLLFB3MxwxM3s/VpHQcHuOo1nCRmovadEnY4OAefOlGgthc+LOKScKKQEhRCikfkOQiB6c+jXLO0rD4s4HPr1shEj6p6dvS1WfjADY8YHqcRuY8NjQlwIHl4U1WcWrTufUoLTx7tPTgAADKiXaZ7aqy+ubvaIniDVeF3wPIewSG8EBfe8Sqy8Q97tOj2MAW1t2m8xEUKsG/WgWIH87GqtX+aCnCEvu9qIEenGyVmCpa9OQu21JuRmVyPjfDnSThfjj8O5+ONwLpxcJIifFYE584fqVCnVVI7sv6xTb0UnL2/jVMpd8HQ0CvNrUFZcp5I8cTwHFzc7LHllYq+eVyyxgbunA2qqtQxiZgz9Ans284kQYl2oB8UK2Iq7//IW69DGVNw8HJCdWYUDezJV1phpkLVi97fnsfH9w90WMDOlshKZ7mM+BIaJU40ztsbJWYJ31t2FBx4fDV9/Z4jFInh4OWDu/cPw141z4O3bu0HIHMchflaE1nWLOI7DpPiwXkZOCLEG1INiBaLGBeG33Zc0fklyHBA1znwX/CvIvYa9P6ar3cYYkH6uHH8czsGUGQONHJluelJtdfb8ITpNE9YXe3tbzJo3BLPm3WZd/FvMuHsQzp0pxpVM1cUGO3uSFj4bQ9OMLZy8tQ1Cuxw2jnZU94b0CiUoVmDaXeE49Gsm2tvkXcZBcDwHOzsbs/1yBxRVZXkRB0GuOcFK2Jd126+huqoRRw5cxsXUMsgFAZFDfDH1rvAeF2G7VezkEORlV2sdgyKxs8F9j4zE9DmRt3UscyEWi/Dq6unY91M6Du3Lgqy2BQAQPsgbc+4bhmGjAkwcITGUkv2ncf6jr1F+9BwAwDnEH4NfvBeRz80Fb2O+PbXE/HCsu9FsJiSTyeDq6oq6ujq4uBiuhLg1yEyvwN8/OILmpnZF1zvHgQkMjs5iLH97GkLDvUwdokZr3tyPzPRKrW3EYhH+/d3DvT7GhbRS/OPDI5DLmbKniec5MMaw6PnY25rS3Nzcjjdf/AU11U1dZyTxHJycxPjgk7sttkdBEBgaG1phYysy+gKPxLgy/rkbJ57/BJyIB7teEBKc4h+Bc8Zh6g+rKUmxEvr4/qYeFCsROcQXGz+/F8eP5iEroxIcgEHD/DB+0gBI7Mz7S8PeQaysuKqJxL73l/K16ib848Mj6OgQVI7RmUx88X/HETjADQNCez7lFlDcRnnjgxn4ZO0xFOReU86cEQSGfv1d8eLKKRabnACKRM/Zxc7UYRADq88vx4kXNwHAjeQEuF5MkKHol2Rc2fobIhbPNkl8pO+hBMWK2DuIMW1WBKbNijB1KD0ydoIUaaeLNW7neQ7j7wju9fMfPXAFHXKmMQHiOA4H9mTi6Zcm9PoYXj5OWP23WbiSWYXMixVgjCFiiC8iBvvQ/XliES7/ey84jgODxv+RkPF/P1GCQnRGCQoxe9ETpNj97XlUVTSovUViKxbd1tiNi2mlWlcKFgSGC2mlvX7+ThzHIXyQD8IH+dz2cxFibq6dz1XtObkVY6hNLwBjjJJyohOaZkzMnq2tCK//dTr6S90AACIRB5FI8QHn4iLBitXx8PHr/cwXXYZhaUtgCCGAjYME4LV/pfASW0pOiM6M0oPS2tqKmJgYnDt3DmlpaRg5cqQxDkssiIeXI977eDay0itxPq0EgpwhNNwLo6IDta6oq4uIIb7Iz7mmcRo2L+IQOdT3to6hTw2yVlzJrARjQPBAT7h7OJg6JKMQ5ALOppTg3JkSdLTLIQ3xwIS4EDg6SUwdGgEQ9KdY5O88pnE7ZyOC9N7eFf8j1skoCcqKFSsQEBCAc+fOGeNwxEJxnCJR0HeyEDczHPt/ztC4XZAzzJgzSK/H7I3W1g7s+Pw0/jicC/n1rnSOA8bGSvH4khg4OVvuF3V1VSPWrz6EsmIZeBEHMCDpaC6++28aliybiDHjzW9Fbmsz4L5JSH1nKxqLqsA6blnG4HqnydBl9xs/MNJnGfwWz759+3DgwAFs2LDB0IcipFd8/Z3x9EsTwHFQWZum878fXDga4YNNO25EkAvY+P5hJCbkKJMTQDGz6UxyIda8eQCtLe0mjNBwOjoErFt1EBWl9QAUCaMgKAY1d3TIsXl9InKvXDVxlEQkEePOg+vhJFX8AcHZiMCJeIDjILITI27nKniOMt96S8T8GLQHpaKiAosXL8auXbvg4NB9N3RraytaW28slCaTyQwZHiFK4ycHo5/UDQf3ZOJCagk6OgS4ezjAyUWCwrwaJB3JxdgJUpMtCZB6uhgZFyrUbhMEhuLCWvx+OBfxfWyGli7SThWh/HpycivGFL1Iv/6UjudXTDZyZORWziEBuDdjG4p+SUbR3hMQ2jvgOXogwh6fCYlb75ZOINbLYAkKYwwLFy7EkiVLMGbMGOTn53e7z5o1a7B69WpDhUSIVkED3LHo+fFIO1WEzRt+R1FBDQDFraXkY3n47n+pWLE6Hv0C3YweW+LBbO0LDnLAsYNXLDJBSTlZpPW1CwJD6skimh1iJngbEaT3TIT0HhpvQm5PjxOU119/HR999JHWNhkZGThw4ADq6+uxcuVKnZ975cqVWLZsmfJ3mUyGwEDzXSOG9H2XMyqx/+cMXLg+1bi/1E0xYJYxdJZz6JzlI6ttwUfvHMS6f86DnZErol672qh9wUGmKDhnidpb5Yr3Qwu5nIEJDJyIEhRCLEWPE5Tly5dj4cKFWtuEhITg8OHDSE5OhkSiOnBvzJgxeOSRR/Dll1922U8ikXRpT4ihHNl/Gdv+dVLlr/PcK9Ua2wsCQ11NC5IT8xA3M9xYYQIA3D0dUFxUp3W6s5u7ZVaj7S91Q+qpIq1Jil+AM3gRVU0gxJL0OEHx9vaGt7d3t+0++eQTvP/++8rfS0tLMXPmTHz77beIiYnp6WEJ0auykjp8ueUkAGjvmbgFxwGpp4qMnqBMnBqK86mai8VxHDA5PsyIERnP5Olh+HnnBc0NOCB+tmUsskgIucFgY1CCglSn/Tk5KQZIhYaGon///oY6LCE6OfzbZUVZ7h6ulcmY4paDsUWNC0JYhBdyr1R3Sah4noOPnzPusNAExcPLEY89E92ltwtQJGZDRwYYPWEkhBgelbonVik762qPek468TwHaS8XDbwdNjY8/rJqGrZtOYmTv+errBs0bFQAnnphvEWvFBw3MxxePk7Y++NF5WwmTy9HTJ8TielzIm+7WB8hxPwYLUEZMGBAj/9aJcRQRL0cTCkIDHEzTVPLwd5BjGeX3YEHH49CVrpiwcHQcG/4+ve+zH9fMmxUAIaNCkBbmxzyDjns7KlsOiGWjHpQiFUaOaY/sjOrNK5gfKvOWwuPLh4LvwAXwwbXDQ9PB4yf1PvVm/s6sVgEmKgeDSHEeKhflFilSfFhkNjZQNMf4BwHZc8ExwGDhvni1XenYToNxiSEEKOgHhRilVxc7bD8nWn4+L3DaGlpV/akcDwHnufw7LKJGBsrRUeHAJ4DTWElhBAj45gZDwyRyWRwdXVFXV0dXFxM261uSTIulOPg3kzkXr4KkY0Io6P7I352BHz9re8cN9S34o/DObh4tgyCIGDgIB9Mnj4QHp7mtUKwXC7g0vly1F5rhqu7HQYP96eBoYQQs6WP729KUKzMDzvO4uedF1Sma/I8B17E4eU34jBsVICJIyS3OpVUgO3/PoW62hblY84uEjy8aAxiJ4eYMDJCCFFPH9/f9CeYFUk9VaQseHXzFFtBYJB3CPhkzVHUy1o07U5M4PTxAmxen6iSnABAvawVn25MQvKxPBNFRgghhkUJihXZv/sSeF79qFDGgPZ2ORIP5Rg5KqKJIBfw1X/OaG3z1dYzkMsFI0VECCHGQwmKFbmSqb04GWPA5UsVRoyIaHM5swrXrmpfAFBW24KMC+VGiogQQoyHEhQroktNKyp8ZT5ktbrdbqurodtyhBDLQwmKFYkc5qvxFg8AgAMGD/czXkBEK3dP3VYn1rUdIYT0JZSgWJG75g7WeIuH4wB7O1tMiAs1clREk9Bwb3j7Omnt+XL3dEDkEF/jBUUIIUZCCYoVGToyAA8uHA0AKj0pHMdBLLHBsrenwtFJbKrwyC14nsOCxdEANN+eW7B4LBWRI4RYJKqDYoUKcq/h0K9ZyMmqgo2tCKOi+yNuxkC4eZhXcTJ9aW3twLkzJaiva4GHlwOGje7Xp4qcnUspwfbPT6OyrF75mLePEx5eNAajYwJNGBkhhKhHhdoI0YIxhoN7MvH9jrNobelQPu7sIsFjz8QgeoLUhNH1DGMMOZevKivJhoZ7ax9PRAghJqSP729ai4dYrIN7MrHji651ROplrdi8PhE2NlP6TA8Ex3EIi/A2dRiEEGI0faefm5AeaG1px/c7zmpuwAHfbEuBGXcgEkKIVaMeFGJUzU1tSNh3GUcPXEFNdROcnCW4Y1oops+JhKub/qbLnkspVbmt0wUDKsrqkZ9zDcFhnno7LiGEEP2gBIUYTb2sBR+s3I/yUhk6Oy5qa5qx98d0JB7Kxptr7oSvv7N+jlWnW/EyWnuIEELME93iIUbzv89OoaKsHrfeVREEhnpZK7Z8/LvejuXhpduMJE9vR70dkxBCiP5QgkKMoq62GaePF2osFCcIDLlXqlGQe00vxxs2uh+cXSQat3Mch+AwT/QLdNPL8QghhOgXJSgEjDGtiwjqQ2FejU7HyMuu1svxbGx4PPZMjOKXW2bjchwHkYjDI0+N0cuxCCGE6B+NQbFiednV2PdTOlJOFqGjQ4BfgDPiZ0Ui7s5wvRcyE4l0q9kh0mNV1OgJUtjYTME321JQcVORM2mIOx5dPBYDI330dixCiP4JcjkgMPC29FVljehdt1Kpp4qwae0xAFD2bFSU1WP7F6dx9kwxXnkzDja2Ir0dLzTcCxI7G60zazgOGDJCv4sVjo4JxKjo/sjPuYZ6WQs8vBzRP8hNr8cghOhX6aEUXFj3LUoPpwICg9uQARjy0r0Y+ORd4Hjq+LcW9E5boabGNvxrw+8QBNVbO4wBYED6uTIc2JOp12NK7GwxfXakxjVleJ5DzMQB8PDS/6DVzvEmw0f3o+SEEDOX8c/d2D9jBcqOpAHXP59qLxUg6emPkfjYWjBBMHGExFgoQbFCSUdz0d4u17idMeDAnky9FzG79+ERyvLynWXaO/8dPtgHTzw3Tq/HI4T0LbKcUpx4cRMAgMlvSkSufxblfpWA3K8STBEaMQG6xWOFCnOvgeM5MLnmBKSmuglNjW1wdNI8E6anRCIezy6/A/GzIpGYkI2rlQ1wdbNH7JQQDBsVQGvLEGLlsj7bA47jwKD+s4njeVz6v10IfXS6kSMjpkAJihWysRXdOrFFfTsDrPjLcRzCB/sgfDANUCWEqKpOu6Lac3ILJgioOZ9jxIiIKdEtHis0Ykw/yLX0nvA8h4jBPpDY2RoxKkKItbOxE0PjQLXreDF9LlkLSlCs0PBRAQgIdNV4S0UQGGbPH2rkqAgh1i7w7lh0KTV9E85GhKB5E40YETElSlCsEC/i8Zd3psHHT7Huzc0DVjkOeOzpaIyI6mfKEAkhVijk4amw9/cEp64e0vW/p4a+cp9xgyImwzEzXm9eJpPB1dUVdXV1cHFxMXU4FqejQ0DaqSKknCxCW2sH+ge5YfL0gbQ+DSHEZGozC7F/xgo0FVcpEhUGMDCIxLaY/PVbkM6dYOoQiQ708f1t0ARl7969eO+993D+/HnY2dlh8uTJ2LVrl877U4JCCCHWR97ahvzvE1Gy/zSE9g54jYlA2MKZsPN0NXVoREf6+P422CyeH374AYsXL8aHH36IqVOnoqOjAxcvXjTU4QghhFgIkUSM0EfiEfpIvKlDISZkkASlo6MDL730EtavX49FixYpHx88eLAhDkcIIYQQC2OQQbKpqakoKSkBz/MYNWoU/P39cdddd3Xbg9La2gqZTKbyQwghhBDrY5AEJTc3FwDw7rvv4q233sKePXvg7u6OKVOm4Nq1axr3W7NmDVxdXZU/gYGBhgiPEEIIIWauRwnK66+/Do7jtP5kZmZCuL6Y05tvvon58+cjKioKW7duBcdx2Llzp8bnX7lyJerq6pQ/RUVFt/fqCCGEENIn9WgMyvLly7Fw4UKtbUJCQlBWVgZAdcyJRCJBSEgICgsLNe4rkUggkehv7RdCCCGE9E09SlC8vb3h7e3dbbuoqChIJBJkZWVh4kRF1b/29nbk5+dDKpX2LlJCCCGEWA2DzOJxcXHBkiVLsGrVKgQGBkIqlWL9+vUAgPvvv98QhySE3KbK8npkZ1WB4zhEDPGFh6eDqUMihFgxg9VBWb9+PWxsbLBgwQI0NzcjJiYGhw8fhru7u6EOSQjpBVltMz7flIxzKSXKxzgOiJ4wAE88FwN7B7EJoyOEWCsqdU+IFWtpbse7f/kVFWX1EATVjwKe5zAgzANvfngnbGxo2S5CiO708f1NnzqEWLHEhGyUl8q6JCeAYlXr3MvVSDmheWA7IYQYCiUohFixxEM52la3B8dzSEzINl5AhBByHSUohFixuppmrduZwFBbrb0NIYQYAiUohFgxd097gNO8nec5eHjTbB5CiPFRgkKIFZscPxDQcotHEJiiDSGEGBklKIRYsYnTQhE4wB0837UbheM5RA71xajo/iaIjBBi7ShBIcSKSSQ2WPn+dIyNDVJJUkQiHpOmhWLZ21MhEtHHBCHE+AxWqI0Q0jc4Oknw3F8mofZaE3KvVIPjgIGRPnByoXWxCCGmQwkKIQQA4ObhgNExNCCWEGIeqO+WEEIIIWaHEhRCCCGEmB1KUAghhBBidihBIYQQQojZoQSFEEIIIWaHEhRCCCGEmB1KUAghhBBidihBIYQQQojZoQSFEEIIIWbHrCvJMqZYZlUmk5k4EkIIIYToqvN7u/N7vDfMOkGpr68HAAQGBpo4EkIIIYT0VH19PVxdXXu1L8duJ70xMEEQUFpaCmdnZ3Bc1+XguyOTyRAYGIiioiK4uLgYIMK+g87FDXQuVNH5uIHOhSo6HzfQubhBl3PBGEN9fT0CAgLA870bTWLWPSg8z6N///63/TwuLi5Wf0F1onNxA50LVXQ+bqBzoYrOxw10Lm7o7lz0tuekEw2SJYQQQojZoQSFEEIIIWbHohMUiUSCVatWQSKRmDoUk6NzcQOdC1V0Pm6gc6GKzscNdC5uMNa5MOtBsoQQQgixThbdg0IIIYSQvokSFEIIIYSYHUpQCCGEEGJ2KEEhhBBCiNmxmATl6NGj4DhO7c/p06c17jdlypQu7ZcsWWLEyA1nwIABXV7b2rVrte7T0tKCpUuXwtPTE05OTpg/fz4qKiqMFLFh5OfnY9GiRQgODoa9vT1CQ0OxatUqtLW1ad3Pkq6NzZs3Y8CAAbCzs0NMTAxOnTqltf3OnTsRGRkJOzs7DBs2DL/++quRIjWcNWvWYOzYsXB2doaPjw/mzZuHrKwsrfts27atyzVgZ2dnpIgN69133+3y2iIjI7XuY4nXBaD+s5LjOCxdulRte0u7LhITE3H33XcjICAAHMdh165dKtsZY3jnnXfg7+8Pe3t7xMfH48qVK90+b08/d25lMQlKbGwsysrKVH6eeuopBAcHY8yYMVr3Xbx4scp+69atM1LUhvfee++pvLYXXnhBa/tXXnkFv/zyC3bu3Iljx46htLQU9957r5GiNYzMzEwIgoBPP/0U6enp2LhxI7Zs2YI33nij230t4dr49ttvsWzZMqxatQqpqakYMWIEZs6cicrKSrXtjx8/joceegiLFi1CWloa5s2bh3nz5uHixYtGjly/jh07hqVLl+LEiRM4ePAg2tvbMWPGDDQ2Nmrdz8XFReUaKCgoMFLEhjdkyBCV1/bHH39obGup1wUAnD59WuU8HDx4EABw//33a9zHkq6LxsZGjBgxAps3b1a7fd26dfjkk0+wZcsWnDx5Eo6Ojpg5cyZaWlo0PmdPP3fUYhaqra2NeXt7s/fee09ru8mTJ7OXXnrJOEEZmVQqZRs3btS5fW1tLbO1tWU7d+5UPpaRkcEAsOTkZANEaDrr1q1jwcHBWttYyrURHR3Nli5dqvxdLpezgIAAtmbNGrXtH3jgATZ79myVx2JiYtgzzzxj0DiNrbKykgFgx44d09hm69atzNXV1XhBGdGqVavYiBEjdG5vLdcFY4y99NJLLDQ0lAmCoHa7JV8XANhPP/2k/F0QBObn58fWr1+vfKy2tpZJJBL29ddfa3yenn7uqGMxPSi3+vnnn1FdXY0nnnii27Y7duyAl5cXhg4dipUrV6KpqckIERrH2rVr4enpiVGjRmH9+vXo6OjQ2DYlJQXt7e2Ij49XPhYZGYmgoCAkJycbI1yjqaurg4eHR7ft+vq10dbWhpSUFJX3lOd5xMfHa3xPk5OTVdoDwMyZMy3yGgDQ7XXQ0NAAqVSKwMBAzJ07F+np6cYIzyiuXLmCgIAAhISE4JFHHkFhYaHGttZyXbS1tWH79u148skntS5Sa8nXxc3y8vJQXl6u8t67uroiJiZG43vfm88ddcx6scDb8cUXX2DmzJndLjb48MMPQyqVIiAgAOfPn8drr72GrKws/Pjjj0aK1HBefPFFjB49Gh4eHjh+/DhWrlyJsrIyfPzxx2rbl5eXQywWw83NTeVxX19flJeXGyFi48jOzsamTZuwYcMGre0s4dq4evUq5HI5fH19VR739fVFZmam2n3Ky8vVtreka0AQBLz88suYMGEChg4dqrFdREQE/vOf/2D48OGoq6vDhg0bEBsbi/T0dL0sZGpKMTEx2LZtGyIiIlBWVobVq1fjjjvuwMWLF+Hs7NylvTVcFwCwa9cu1NbWYuHChRrbWPJ1cavO97cn731vPnfU6lnnj/G99tprDIDWn4yMDJV9ioqKGM/z7Pvvv+/x8RISEhgAlp2dra+XoFe9OR+dvvjiC2ZjY8NaWlrUbt+xYwcTi8VdHh87dixbsWKFXl+HPvTmXBQXF7PQ0FC2aNGiHh/P3K8NdUpKShgAdvz4cZXHX331VRYdHa12H1tbW/bVV1+pPLZ582bm4+NjsDiNbcmSJUwqlbKioqIe7dfW1sZCQ0PZW2+9ZaDITKempoa5uLiwzz//XO12a7guGGNsxowZbM6cOT3ax5KuC9xyiycpKYkBYKWlpSrt7r//fvbAAw+ofY7efO6oY/Y9KMuXL9eayQJASEiIyu9bt26Fp6cn/vSnP/X4eDExMQAUf2WHhob2eH9D68356BQTE4OOjg7k5+cjIiKiy3Y/Pz+0tbWhtrZWpReloqICfn5+txO2QfT0XJSWliIuLg6xsbH47LPPenw8c7821PHy8oJIJOoyE0vbe+rn59ej9n3N888/jz179iAxMbHHf+3a2tpi1KhRyM7ONlB0puPm5obw8HCNr83SrwsAKCgowKFDh3rcS2rJ10Xn+1tRUQF/f3/l4xUVFRg5cqTafXrzuaOO2Sco3t7e8Pb21rk9Ywxbt27FY489Bltb2x4f7+zZswCg8kaYk56ej5udPXsWPM/Dx8dH7faoqCjY2toiISEB8+fPBwBkZWWhsLAQ48eP73XMhtKTc1FSUoK4uDhERUVh69at4PmeD78y92tDHbFYjKioKCQkJGDevHkAFLc3EhIS8Pzzz6vdZ/z48UhISMDLL7+sfOzgwYNmeQ30BGMML7zwAn766SccPXoUwcHBPX4OuVyOCxcuYNasWQaI0LQaGhqQk5ODBQsWqN1uqdfFzbZu3QofHx/Mnj27R/tZ8nURHBwMPz8/JCQkKBMSmUyGkydP4tlnn1W7T28+d9TSua+ljzh06JDG2xzFxcUsIiKCnTx5kjHGWHZ2NnvvvffYmTNnWF5eHtu9ezcLCQlhkyZNMnbYenf8+HG2ceNGdvbsWZaTk8O2b9/OvL292WOPPaZsc+v5YEzR9R0UFMQOHz7Mzpw5w8aPH8/Gjx9vipegN8XFxSwsLIxNmzaNFRcXs7KyMuXPzW0s9dr45ptvmEQiYdu2bWOXLl1iTz/9NHNzc2Pl5eWMMcYWLFjAXn/9dWX7pKQkZmNjwzZs2MAyMjLYqlWrmK2tLbtw4YKpXoJePPvss8zV1ZUdPXpU5RpoampStrn1XKxevZrt37+f5eTksJSUFPbnP/+Z2dnZsfT0dFO8BL1avnw5O3r0KMvLy2NJSUksPj6eeXl5scrKSsaY9VwXneRyOQsKCmKvvfZal22Wfl3U19eztLQ0lpaWxgCwjz/+mKWlpbGCggLGGGNr165lbm5ubPfu3ez8+fNs7ty5LDg4mDU3NyufY+rUqWzTpk3K37v73NGFxSUoDz30EIuNjVW7LS8vjwFgR44cYYwxVlhYyCZNmsQ8PDyYRCJhYWFh7NVXX2V1dXVGjNgwUlJSWExMDHN1dWV2dnZs0KBB7MMPP1QZf3Lr+WCMsebmZvbcc88xd3d35uDgwO655x6VL/K+aOvWrRrHqHSy9Gtj06ZNLCgoiInFYhYdHc1OnDih3DZ58mT2+OOPq7T/7rvvWHh4OBOLxWzIkCFs7969Ro5Y/zRdA1u3blW2ufVcvPzyy8rz5uvry2bNmsVSU1ONH7wBPPjgg8zf35+JxWLWr18/9uCDD6qMr7KW66LT/v37GQCWlZXVZZulXxdHjhxR+/9G52sWBIG9/fbbzNfXl0kkEjZt2rQu50kqlbJVq1apPKbtc0cXHGOM6d7fQgghhBBieBZbB4UQQgghfRclKIQQQggxO5SgEEIIIcTsUIJCCCGEELNDCQohhBBCzA4lKIQQQggxO5SgEEIIIcTsUIJCCCGEELNDCQohhBBCzA4lKIQQQggxO5SgEEIIIcTsUIJCCCGEELPz/6kPgronMRE/AAAAAElFTkSuQmCC", "text/plain": [""]}, "metadata": {}, "output_type": "display_data"}], "source": ["import numpy as np\n", "import matplotlib.pyplot as plt\n", "\n", "def points(n):\n", " tX1 = np.random.multivariate_normal([4, 3], 5*np.eye(2), n) # points de classe 0\n", " tX2 = np.random.multivariate_normal([-2, -1], 5*np.eye(2), n) # points de classe 1\n", " X = np.vstack((tX1, tX2))\n", " y = np.hstack((np.zeros(n), np.ones(n)))\n", " return X, y\n", "\n", "X, y = points(100)\n", "plt.scatter(X[:,0], X[:,1], c=y, cmap=plt.cm.Spectral);"]}, {"attachments": {}, "cell_type": "markdown", "metadata": {}, "source": ["## Algorithme des $k$ plus proches voisins"]}, {"cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": ["def d(x, y):\n", " s = 0\n", " for i in range(len(x)):\n", " s += (x[i] - y[i])**2\n", " return s**.5"]}, {"cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": ["def voisins(x, X, k): \n", "\t# renvoie les k plus proches voisins de x dans X\n", "\tindices = sorted(range(len(X)), key=lambda i: d(x, X[i]))\n", "\treturn indices[:k]"]}, {"cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": ["def maj(L):\n", "\tcompte = {} # compte[e] = nombre d'occurrences de e dans L\n", "\tfor e in L:\n", "\t\tcompte[e] = compte.get(e, 0) + 1\n", "\treturn max(compte, key=compte.get)"]}, {"cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [], "source": ["def knn(x, X, Y, k):\n", "\tV = voisins(x, X, k)\n", "\treturn maj([Y[i] for i in V])"]}, {"cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [], "source": ["def separer(X, Y, p):\n", "\tX_train, X_test, Y_train, Y_test = [], [], [], []\n", "\tfor i in range(len(X)):\n", "\t\tif i <= p * len(X):\n", "\t\t\tX_train.append(X[i])\n", "\t\t\tY_train.append(Y[i])\n", "\t\telse:\n", "\t\t\tX_test.append(X[i])\n", "\t\t\tY_test.append(Y[i])\n", "\treturn X_train, X_test, Y_train, Y_test\n", "\n", "X_train, X_test, Y_train, Y_test = separer(X, y, 0.8)\n", "\n", "def predict(x, k):\n", "\treturn knn(x, X_train, Y_train, k)"]}, {"cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [{"data": {"text/plain": ["0.8974358974358975"]}, "execution_count": 7, "metadata": {}, "output_type": "execute_result"}], "source": ["def precision(k):\n", "\tn = len(X_test)\n", "\tp = 0\n", "\tfor i in range(n):\n", "\t\tif predict(X_test[i], k) == Y_test[i]:\n", "\t\t\tp += 1\n", "\treturn p/n\n", "\n", "precision(5)"]}, {"attachments": {}, "cell_type": "markdown", "metadata": {}, "source": ["## En utilisant sklearn\n", "\n", "`sklearn` donne le m\u00eame r\u00e9sultat, mais est beaucoup plus rapide (tester avec $n = 10000$ points, par exemple) :"]}, {"cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [{"data": {"text/plain": ["0.8974358974358975"]}, "execution_count": 8, "metadata": {}, "output_type": "execute_result"}], "source": ["from sklearn.neighbors import KNeighborsClassifier\n", "neigh = KNeighborsClassifier(n_neighbors=5)\n", "neigh.fit(X_train, Y_train)\n", "neigh.score(X_test, Y_test)"]}], "metadata": {"celltoolbar": "Tags", "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": "3.9.2 (default, Feb 28 2021, 17:03:44) \n[GCC 10.2.1 20210110]"}, "vscode": {"interpreter": {"hash": "31f2aee4e71d21fbe5cf8b01ff0e069b9275f58929596ceb00d14d90e3e16cd6"}}}, "nbformat": 4, "nbformat_minor": 4}
\ No newline at end of file
+{"cells": [{"attachments": {}, "cell_type": "markdown", "metadata": {}, "source": ["# Cours : Algorithme des k plus proches voisins\n\n", "\n", ""]}, {"attachments": {}, "cell_type": "markdown", "metadata": {}, "source": ["Nous allons classifier deux ensembles de points (rouge et bleu) issus de deux distributions gaussiennes (de param\u00e8tres diff\u00e9rents), avec la m\u00e9thode des $k$ plus proches voisins :"]}, {"cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [{"data": {"image/png": "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", "text/plain": [""]}, "metadata": {}, "output_type": "display_data"}], "source": ["import numpy as np\n", "import matplotlib.pyplot as plt\n", "\n", "def points(n):\n", " tX1 = np.random.multivariate_normal([4, 3], 5*np.eye(2), n) # points de classe 0\n", " tX2 = np.random.multivariate_normal([-2, -1], 5*np.eye(2), n) # points de classe 1\n", " X = np.vstack((tX1, tX2))\n", " y = np.hstack((np.zeros(n), np.ones(n)))\n", " return X, y\n", "\n", "X, y = points(100)\n", "plt.scatter(X[:,0], X[:,1], c=y, cmap=plt.cm.Spectral);"]}, {"attachments": {}, "cell_type": "markdown", "metadata": {}, "source": ["## Algorithme des $k$ plus proches voisins"]}, {"cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": ["def d(x, y):\n", " s = 0\n", " for i in range(len(x)):\n", " s += (x[i] - y[i])**2\n", " return s**.5"]}, {"cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": ["def voisins(x, X, k): \n", "\t# renvoie les k plus proches voisins de x dans X\n", "\tindices = sorted(range(len(X)), key=lambda i: d(x, X[i]))\n", "\treturn indices[:k]"]}, {"cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": ["def maj(L):\n", "\tcompte = {} # compte[e] = nombre d'occurrences de e dans L\n", "\tfor e in L:\n", "\t\tcompte[e] = compte.get(e, 0) + 1\n", "\treturn max(compte, key=compte.get)"]}, {"cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [], "source": ["def knn(x, X, Y, k):\n", "\tV = voisins(x, X, k)\n", "\treturn maj([Y[i] for i in V])"]}, {"cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [], "source": ["def separer(X, Y, p):\n", "\tX_train, X_test, Y_train, Y_test = [], [], [], []\n", "\tfor i in range(len(X)):\n", "\t\tif i <= p * len(X):\n", "\t\t\tX_train.append(X[i])\n", "\t\t\tY_train.append(Y[i])\n", "\t\telse:\n", "\t\t\tX_test.append(X[i])\n", "\t\t\tY_test.append(Y[i])\n", "\treturn X_train, X_test, Y_train, Y_test\n", "\n", "X_train, X_test, Y_train, Y_test = separer(X, y, 0.8)\n", "\n", "def predict(x, k):\n", "\treturn knn(x, X_train, Y_train, k)"]}, {"cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [{"data": {"text/plain": ["0.8974358974358975"]}, "execution_count": 7, "metadata": {}, "output_type": "execute_result"}], "source": ["def precision(k):\n", "\tn = len(X_test)\n", "\tp = 0\n", "\tfor i in range(n):\n", "\t\tif predict(X_test[i], k) == Y_test[i]:\n", "\t\t\tp += 1\n", "\treturn p/n\n", "\n", "precision(5)"]}, {"attachments": {}, "cell_type": "markdown", "metadata": {}, "source": ["## En utilisant sklearn\n", "\n", "`sklearn` donne le m\u00eame r\u00e9sultat, mais est beaucoup plus rapide (tester avec $n = 10000$ points, par exemple) :"]}, {"cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [{"data": {"text/plain": ["0.8974358974358975"]}, "execution_count": 8, "metadata": {}, "output_type": "execute_result"}], "source": ["from sklearn.neighbors import KNeighborsClassifier\n", "neigh = KNeighborsClassifier(n_neighbors=5)\n", "neigh.fit(X_train, Y_train)\n", "neigh.score(X_test, Y_test)"]}], "metadata": {"celltoolbar": "Tags", "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": "3.9.2 (default, Feb 28 2021, 17:03:44) \n[GCC 10.2.1 20210110]"}, "vscode": {"interpreter": {"hash": "31f2aee4e71d21fbe5cf8b01ff0e069b9275f58929596ceb00d14d90e3e16cd6"}}}, "nbformat": 4, "nbformat_minor": 4}
\ No newline at end of file