From f453b92205ba7fa09423baaed35b777c541d8127 Mon Sep 17 00:00:00 2001
From: Mirko Birbaumer <mirko.birbaumer@hslu.ch>
Date: Thu, 10 Mar 2022 09:28:56 +0000
Subject: [PATCH] changed order of chapters

---
 ...ining and Optimizing Neural Networks.ipynb | 3842 ++++++++++-------
 1 file changed, 2277 insertions(+), 1565 deletions(-)

diff --git a/notebooks/Block_3/Jupyter Notebook Block 3 - Training and Optimizing Neural Networks.ipynb b/notebooks/Block_3/Jupyter Notebook Block 3 - Training and Optimizing Neural Networks.ipynb
index cca495e..063f8dd 100644
--- a/notebooks/Block_3/Jupyter Notebook Block 3 - Training and Optimizing Neural Networks.ipynb	
+++ b/notebooks/Block_3/Jupyter Notebook Block 3 - Training and Optimizing Neural Networks.ipynb	
@@ -22,7 +22,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": 2,
    "metadata": {
     "id": "dzLKpmZICaWN"
    },
@@ -31,7 +31,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "2.1.0\n"
+      "2.7.1\n"
      ]
     }
    ],
@@ -77,7 +77,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 3,
    "metadata": {
     "id": "7MqDQO0KCaWS"
    },
@@ -87,7 +87,8 @@
      "output_type": "stream",
      "text": [
       "Downloading data from https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz\n",
-      "170500096/170498071 [==============================] - 6s 0us/step\n"
+      "170500096/170498071 [==============================] - 7s 0us/step\n",
+      "170508288/170498071 [==============================] - 7s 0us/step\n"
      ]
     }
    ],
@@ -162,7 +163,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 4,
    "metadata": {
     "id": "IjnLH5S2CaWx"
    },
@@ -185,7 +186,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 5,
    "metadata": {
     "id": "MaOTZxFzi48X"
    },
@@ -217,7 +218,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 6,
    "metadata": {
     "id": "ywVIEcXDvXW_"
    },
@@ -270,7 +271,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 63,
    "metadata": {
     "id": "ug3dTdldvXXI"
    },
@@ -294,11 +295,7 @@
     "print(train_mean)\n",
     "train_std =  X_train.std(axis=(0,1,2))\n",
     "print(train_std)\n",
-    "# Standardize\n",
-    "# X_train_zc = (X_train - train_mean)/train_std\n",
-    "# X_test_zc  = X_test  - train_mean/train_std\n",
-    "# X_train_zc = X_train/255\n",
-    "# X_test_zc  = X_test/255\n",
+    "\n",
     "X_train_zc = X_train - train_mean\n",
     "X_test_zc  = X_test  - train_mean\n",
     "\n",
@@ -311,7 +308,26 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Standardize (as e.g. in ResNet)\n",
+    "# X_train_zc = (X_train - train_mean)/train_std\n",
+    "# X_test_zc  = X_test  - train_mean/train_std\n",
+    "# Scale\n",
+    "# X_train_zc = X_train/255\n",
+    "# X_test_zc  = X_test/255\n",
+    "# print(X_train_zc[3].mean(axis=(0,1)))\n",
+    "# print(X_train_zc[3].shape)\n",
+    "# print(X_train_zc.shape)\n",
+    "# print(X_test_zc.shape)\n",
+    "# print(X_train_zc.dtype)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
    "metadata": {
     "id": "J3zk7RjDvXXQ"
    },
@@ -374,7 +390,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 9,
    "metadata": {
     "id": "HQhzU-zkvXXZ"
    },
@@ -402,14 +418,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 10,
    "metadata": {
     "id": "ULyOuowNvXXj"
    },
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 6 Axes>"
       ]
@@ -446,14 +462,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": 11,
    "metadata": {
     "id": "sgKu8QervXXr"
    },
    "outputs": [
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXAAAAD6CAYAAAC4RRw1AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/d3fzzAAAACXBIWXMAAAsTAAALEwEAmpwYAABo8ElEQVR4nO29Waxk2XWeufaZYo4bd8zMyqy5ioMskZRMy5q6Wy3BaNkwQD0YgtVogwYE8KUNSGg/iNCL2402wH6R/WDAjQIkiAYEy0RLhghDboEts2GrYdAcxKFYRRaLxaycM+8cc8QZdj/EubG+c+reypuVUbcy7t0/UKidcU+cs89e++w4699r/ctYa8XBwcHBYfngfdAdcHBwcHB4b3ALuIODg8OSwi3gDg4ODksKt4A7ODg4LCncAu7g4OCwpHALuIODg8OS4rEWcGPMrxhjvm+MedMY89lFdcrhg4Wz6/mFs+35gnmvceDGGF9E3hCRvyUit0TkqyLy69ba1xbXPYezhrPr+YWz7flD8Bjf/WkRedNa+5aIiDHmj0TkUyJy4mQIA99WKqGIiKRpOv/cZif9iOjn7/Y7Ywzb5ti272u7EoZ6jKefsxv8YfNwjEhWuHaa6H0kaPu+Dq0x6uikGe7b6rl8T4/x0OZ3+bmISIYOc3g4HrZwT7PrDUdTmUwT3hTxyHaNQt/Wq+ExfznpEuiTlA2r3+G48yjOlyRJcDzHirb35+3JeDJvx7CX8Yp9LZyLNsC5skztl6XaLr4UnTCHOWelfG39d+DrtTlH4iTWa+fnncaZJGn2boP+SLaNQt/Wcruy7xn+kWZZ+Wv5PRXBvvu4J574pEfcFGyBOcFn1Pj4XPtUtmsc67jFU22nmFOm8PxxPvIeTpgHaJ+8rpXniKK4funn42m6Y63dLB//OAv4VRG5iX/fEpG/eUyHPiMinxERiaJAfuKj10REpN/rz4+JR9N5m7fFxZEL/uzE2vQDHcww0EGuVfT2Vpq1efvZa5f0+Gpl3h5P9Ro0dL2h58myUaEb3cPDeXt/R++p3dqYt4NIr3HY7+o1pmM9vlnX69W0r5WqtqOqHiMiMhrruCWYLPzxSLHAJfFs8fqP/+X78i54ZLvWKoH8t598TkTKD+Hxa4nF5+UFgD9YlVp13o4x4eOJ3vf+zt68XcU411v63dWV9rz9g9femLcf7Krt/Jp+V0SkWW9oP2CDerM1b4/GOheGXbV9EmPenjCHDRaxEPNXRKRWVfttrGg/OIdvP9jWfiSzsfnB23o/J+ChtqVdq5VAfuGvPyciIpNE7dTH89ob6hzm60XgF23fbqg9OrgPjk+GhTfD+AQV2LWC7471GW3UdJzG8XDejkovFg/u3p+3b928N28f9vU+KngWg0Bt0WjoNdrtlXl7NMI86A3m7WSqz55IcXFPYvwNQ+VhLkShjsHr1/fflmPwOAv4qWCtfUVEXhERaTVrtlKdGTI94ZcqmahRuDCXf7FOevuJcYwxOkireOvO8KvcaETzthfo8Q/u60KbxHp8HceLiKQJf62xcKba3wYWoo1I7ymequEj9I8Lhh/o5B2M9O1RRCSzfMvUayexPmAp2lmcT1J7/FvTo4B27bRr9ujtxJ70RnXSW2lSnOQxFzn8GDfaunAaLNQdPLiVSMfAGj2Ph/Vx8/LavF1v63erteKPYzrBDyIe4lpDj9vf3tHjY8487TfnaeGtu/BmV7i0RJj3tYrOt+FYFxn+YIdh9I5zvlfQrmsrNduozPrCBTn0dT7XKjpve0Odn8Nxca7u97TvAxy3Chus4MdqkOq8HY+xIGPOV329duRzfnFAi/Or1dJna+vy+ry9iTf4eAKbYfLwRYx2HeOHnIt2+R278BLKuUAvi+Nc+mE/Do+ziXlbRJ7Gv6/lnzksN5xdzy+cbc8ZHmcB/6qIvGyMed4YE4nI3xeRLy6mWw4fIJxdzy+cbc8Z3jOFYq1NjDH/SET+XER8Efl9a+133+07WZbJoD/jClvgkDK4FiNukMAdIRclUuSoScFk2GSMmnC11/R6Sabf7Q56er0UG2K4HDdIAq9IoTRret7JAP1IQW9Yurt01eCmh9pX4+s1+qBNjFd0qbjZFU/1uCzGd6yO7VNbM5cxDG7JSXgvdo2TVG5tz/jXOlzqGvjHWqRt0ix+iSvlPSZwf/s95XersEE91HZg9LyDiXKRAaiVy1d0b8Jivy+dlig6MCLcLE5IjWGPZeJrX3kPXlPphkPw5OaEDVeRoutcq+r93d/VuUoK5egY7h8ch0e1re95spbzwb2h0gQT0BWkVqLC/lORez7sKQ1yd1v3LSwpijr2fiqgafAwVkChVEAZxSO1dz/WceYel4hIZ1X3Q9qg5cZDtd94oMafgqOvYl/k+i11XEj7EqR2RUo0MHnvwqa1trmHdxIeiwO31v6ZiPzZ45zD4cmDs+v5hbPt+YLLxHRwcHBYUrzvUShElmXzkBsPrlOr1S4cc4QJ3ESGG4kUKRXu7tbr6rrVa+Gxn3uG58Ju8mSMT+miNuQkMC6VoWfjobqMFcaa47xjUETjRN1jnEZ8RKeUt7UT0CYpaJMIbtilDd1pr+VUgl8Oe3hMZJmV4WR2LxOE0B0OdDxDn5EVCM0qh++1MIYWO/o4bwiqJEBETQpXdoLIhUO4xxurOh4B/NhaqC67iEiIfo1AYzF66tpTSsekmEcBaJ0uwsoCmPIoGktEpF5y8/lQ0uQTxC0zFDPIT7yIKBQiDHzZWl8VEZEG6I2Drs7VIcfcP54KEBEZDxmVdXxkzQDzxcv0eg3MnSooyJWmzpW93sG8zfj+/kExtDJAtFaI99cWIr9W67oeMZqmj5DJQBhFp8fQ9kl8crSXd0JcfCXEs1GaF8ee56FHODg4ODg8kXALuIODg8OS4kwpFGszGecuE1NcPbgQzZbuDBfSzofvCIuftxrY6b/61Na8naVwfRMktDC4H4EdhfR37CAncNMDUxyyChJzNjqa6Xqwv6/fj3UHf21L+7fLrC2rY0D3mAkwyVTvQQSJOSLSQLTC1npn3iadEnizcxlTHsvHQ+B7st5piojIGG7+GO71EHTRAbJw261m4VwZXNAKIhkqoMw8UGv1urqZKSiXwzEMi0icAdz/OhIzEq9I0VVCPddR5JRIMXqkiqSuKpK1GLWSWR2DlY7avoZs28lI7Sgi0gRdESJZ5ZDH9YsZwe8HrIhk+TNImoeJaYew5T7GtlJKQqlf1WeDmZzbewfzdlLIyFZ7r25ppNfVK6vzNlUDDibaDy9Vu4SlBMAxnjmDSBcfkUo+onkSUJNNzLXLlzvz9lqilAujlA4P9VoiIodd/XeBAsMzWkU/mJ19EtwbuIODg8OSwi3gDg4ODkuKs41CsSKTXMTFwB2k2hsplFZTXZPDWCkJEZE0GeE4dcPb0DqIQt2lnoJumEJESrDrW1A0o74KhHzSaVHjoQaXeizqIlWgYRGiTfqiCrGa2Kh7xsB+apmk06Lb3Gmpi7Xe0bGK8LN8RJuIiHgLpk6O4HtGVvIonwaEmJJUXdQR6JQbN9WWh6XICSbdNCBctFHXe712pTNvNxt6zM6h6tdUE7V9mIIqQUJKfwB71YpRKBlol6BCFUClViYUP0tA14Fyo8aN7zOqCjo4UXkMkBwTab/abbVxwPPmUSgFlb8FYDqdyo2bs6Sv1dWO9gPPaBtRQ4xUGZRooRQJOOtriEgCjTXoauTQFnRR2jUkaA107owSRuLotVaxHsRhMZJj2NU5MvbUZgmj34z2nfZOjc7h1TW1BbVypoiWiuPiGERV/U4YIFoF4mxMhPNOEVTk3sAdHBwclhRuAXdwcHBYUrgF3MHBwWFJcbZhhJmdC79MKOaCbCyGF250NGQohCi/iMikEG6j3FsLIYV9CFVNkWUZIDSrjgwsVkapgEhmoQcKv4uIBPh+Ai6MCZRNhMolmXJqoN8l8BCqCOGnKUKrWu1iWNFJvLdvjue952GZC6bCPc9IK9eNpsgS2xWGa6KgRlwS/KGutvGg94xO+xb7Gci6G2OPoF3Twa2GOm4TZEBm4Osnk+LeBgXSWLnFg2ATCxDwGGpKsT4Oi2tUEUZYFinrI2xxd6xzmAJpG+tqey/n1oNgsZmYWZpJfzAb0ynur9/XvQNy43Vw4Gud4vNKLj9DKG7M7EaIya22lSOOoQ3OIiYZ9qwi7PWsIfzOlMJUu9AfH4KjpnAei25xLyzJEOYIIToTs2qS3tvWVqdwber8U5Kfz/gK+n7j5sOVft0buIODg8OSwi3gDg4ODkuKM6VQxFqxOYUSMxMToWTM+DKpum1PbapwkEgxs5JCVf0+9L1Z+BWCV7W6ulX1moZEZTg+ReigRZaelxazq4ZDPa5a0YyxFQh0pQhJ66MmJrM4A2SueSgFt7Gl4kvxuHjt0IB2OaHwcoL7TnJ6wi6gpFqhH0EgVzZmpcqGCB+jUJEP9/rSulJj1i9OwQf3d+ftAKFnKy11qSfISB1BNIzvIxHuuwFd8itXlb7pQ/T7h9evF/pRQ9hiq61l2EY9pWmmDP/CNQJwYymEjpKU4Yhqg2lc1JOmCJKxx8+RIAQ9lYc5lnXFHx9mnjHIMNAEwmsj2PvyJc22vHLlSuFMAWyTIjtys6PP3+BQQwT5nGQsUQfapBIiNBFUiY8s7aAkluaBWju4qbrkISktliosFCzW805LlNu8T7ieZ4rPGUWvqN3eaOrcqSLbc2VVnxORY0tiujdwBwcHh2WFW8AdHBwclhRnSqF4xkg9D88YYjfZYvc8QjZZBDcjLOkLB3ClhqBNMtAdFi5MiIgGH99lmIBBWS1qG601IFYzKUahsBRara477YOCa48dbtAXLVQ4X0H2GCMxAlasritFIyJSwd8myBAdg2JIWHoud10XnY/pe2au690oZTQe16cQ2XdRrRgl4MUQkYI4UgOu5Qha32AxBIFJcuWyUiWQkJadPXWbHxwe6PkR+SMiAnZLhl2lddIpIh98ilmpLaugAWMc3xtodMkY879Qv09E6qBvDKJhfETA9Luqcz3I53mCsVsEfN+XdrMjIiLdgVIazFo0oKp6A7VLdUfHTERkBRRahMzkNZQ4e+17Sk/t7iqd0sr7IFKMemlhnK9uqb0PYNc3b94p9CMBndnAhDGsEeBX0EQEEyhBipcx8zYBtSslCoURKhTwC2BXRlXxOTkJ7g3cwcHBYUnhFnAHBweHJcWZUijGiPj5TjpFoHxETayv6670aoPRIsVq8ExqSBLSK6BEUMMqCtQdyRABMNhTV7QNTW3u+O/SHSxFcDRadL0RxYLjphN1DVdW9J5acMPoYlZ8VMtmVfpSkAGrl1MvewBXvSCHvOCSW0dIkkR2d2fURAO0EBM7Om3aUpMpWBleROTKRmfevgt39Adv/nDe9lH+rFZTF7wJIaBGRe29v3+g59xWezcgDlWtFpNprOi1KyyTNYW7O4LQEbXdQSukIKzCSMemYvWc01IyUw3JRhYJNLTfbg+lyPKIjXjBFIrn+9JqzWi7elX7TjplhMioGEJOUYmoS6dKrwyhkX53V+fq/QcHem0kN0WB3leExDQfSXFv/fCteXu3DzrRFt9RNzZRUi9AEthE7+Oo7KOISIIknSr0w0mh+KB3J7He56hEt04RXTRBkqGwjymi0fyHL8/uDdzBwcFhSfHQBdwY8/vGmAfGmFfx2Zox5kvGmB/k/199t3M4PHlwdj2/cLa9ODgNhfIHIvIvReRf47PPishfWGs/Z4z5bP7v337YiTzfl+b6zH2ORN1rD7rYHeh9sCpTCj0EEZHRBFEeiGIJWZILQfUWeiZTRIisIUmDkS01BPzXU3WFRuNi6S2WLKNOhoVbVK8qfdABhdKpK1VSwQ41y1G1oZ/QBzUiIpLC9R4hHMPiPjhuh3m0Tjpz8f9AFmTXJEllO6eZ9kBXrKzo/W1uaCIWq4H7peSTFei533+gf7u3o1EJDdAmoaeubMXT797OdaxFRLZBk/VipVk2L2tUz2gCl1aETJxsbOpaBxlvSaAnTXd+CCqBkSNDUGkx5pEpRaEkrD6PKKLhUG1JqenAnz1TxsxpgD+QBdn2aIb5eK6ubGjCTprqGBqUtAtKJeqoGzPEs3jjLhK3QI01QVdUfER5QBtmgnqI97b1PIepzoNnn3+20I8Ez0yKaJpKVWm9ZgM69j0d82FP1w0vxP0xqg1JQHFcKucG6taDHhPLNE5g72rt4ZTnQ9/ArbX/SUT2Sh9/SkQ+n7c/LyK/+tArOTxRcHY9v3C2vTh4r5uYl6y1d/P2PRG5dNKBxpjPiMhnRETCwFHuTzjek13rFf+kwxyeHJzKtrQri/g6PJl47CgUa60171Lm3Fr7ioi8IiLSblbsUSmiK1tKXcSQAa3V1VWzkN8cjYv0wRilxliGy/BHAi6dMeruriHSpVFImoBWBSIJqk2dyGEpWmETFeAZMQIvWiqIbqlCZzaDPCVLb9WQeULX0y9V+h5hV3sP7l0XrvYYu/9Z7j5mp8jkeRS7dpqRjfNq3DEiR3xIhfqgVqoZXcliBIYBZXTvwY5eT1CWDvRUgiiIwaHa24c8b3cAviHS8/RQUZ1SoSJFrZHxGOXScFgVyT+0DCufVyFf2usqDTRJdE41a6VEHkRgeHgeeoeUREYkTu7y+375pft4vJttadeNTtNOcjqnhuQbRhqFnrajCPREzEQ2kTHmarevdvI8PhvafhZV7LOJUmA/9qGr8zbpkNuIZvFBLEwmRbsKZGMtSqS1VjQ6xRf8cEGnudpSSmPK0nqMIkKCTgVUn4jIGOX8akhC8hEqVkP0VJIer7dCvNdX4vvGmCsiIvn/H7zH8zg8WXB2Pb9wtj2HeK8L+BdF5NN5+9Mi8qeL6Y7DBwxn1/MLZ9tziIdSKMaYfyMivygiG8aYWyLyT0TkcyLyBWPMb8hM5/DXTnMxY4yEeXB6CMlNa6kXgSo6cF1NXNzV9pAItHZJ6Zg+dBAMfp+aSCqxcFmiml77YJ/XpqaBHv+xj3240I/nrz01b4/76jbu7UE7AjKihwN1c9OMMqKMMND7zuBO94ZFt/QOoisOQJsMpkhUwueSR0qkmV2oXX3Pl2Z95i4yOihmlRO4mZT7TJKilGq/q/+eYtwC6I5srmnkQyjqlr78otoiAlUy+r7qYdzdVhpj2FcKpFIraqFkqA6zv6ffCTx1cQPMqeFY+zFBiMj4QG1Zha6GRZhLs6Qfkyb6/TDAvMXnm2uQV873II70NRZl2yRNZSenvq5t6fUK0RsVvAOiGdtiBEVm9Xkfj5TuePO1N+btH//Ic3p/6zomLz379Lz93NOX5+3eEJE/VqO7bt29q5+XqFfa2UMVny50d3xRmzVDHf8A3x3uw66glPb60KiZFNcsRht5sL8PmhSKtdLvFeWjj8NDF3Br7a+f8KdffujZHZ5YOLueXzjbXhy4sBAHBweHJcWZaqH4ni8rzZmrHWLffoTEhcFQXZ4U2g9NuJ8iInUEwq9DWyOCezfoKY1hsLPMHeT79zXSwffVpcqg5fChF16et3/ir3280I+XXrg2b1vsvN/4kSaSUOOh90N11WIkCPmoLkKJ2ymSHlKv+Hvbh4RsH3Kt0wxRLEiOOHJdF1uPR8T3A+mszJJdtnylswTSnQjEkQzyuiUpFBnB7aT2RFbVObKxpnPhyrpe7xMfe2neThBRZGp6DOU6p6Bv+sNiIk8KSssDbTKZUj9D56qBdGgNUUsGehYW56kiyaMUXCQWehiUCV5fV+qoCnf+8KgqzoIrLWVZNtcF4ViNmJCEaC1WoAlMkZKysIcHOvTDzygl8vxVtdNP/cRz8/ZPfFTbCZ5du6Nj82Mf0us99yykZXtFCuXe9n09F2zcQ8IOn8sKkgR7A71v0pkeokW6qCQ0TYphmI26zoUINO4U49Hr6zWGwyK9eBzcG7iDg4PDksIt4A4ODg5LijOWkzUS5ckPKQL7a3AtTaSuRYqknEq16I5c6eiu8woSKoYHoBwS3J7R3eQpdC8SRBsgV0EuIdHowy8+P28/c1kTCUREqkioyLCrfe3prXn7DvQeDg/V9RogsH+l1Zm3Wax4UtCQKAb2Q1FXJnAtezivBe1y+alZodnwu0obLQpZvqveQGUh5EBItaZjEyOawk6LEqhprH1/cFujR9ZXlCa7vKnX+KmPvzhvX7mMaKSBup8v00ag5TK4sYclV/sAFW9efOmvad/xnS4iinbvqmt+6yaiIEC5YMpLBdmrrZbem4hI6OvfYmhjTCEXu713gL4fadwsmhxTpdMHu9vzz4JQk15WoCHCoso+EnRERAy4squXlQK7igLXf/Nv6HP24vOayNNpq/2mkI5OMr1GExo6R0llIiLdgSbtiYiEPp53aBFNMA33DzTqaHf33rzNosaQL5EAiXeVCuZ/VFxeW6jYFSHRjMWL93Z03jFZ6yS4N3AHBweHJYVbwB0cHByWFGdKoWQ2m0tq0pWto5rM05eUojiEdKR4RfdwPEQFDVAOl1H0dHtXI0F2tw/m7Qr0BoQUCiJbOg1oWOype3zvelHfoIsg/iF0R956+/q8/WBHP7+/jR180Ac/uq7921jXa3f76s7t94uB/V3on4ygw5KCWxmCTpnmuhDZacRQHgHWWpnmrv54ov0IG6hkg+ggRtaU8nikhrmwCTrm2lZn3v74R1Ui9GW42i3YrFrVMWg09SIsJptgHO5kxY60Klf02pc0UmL9iia0DIbq7n7nm9+et7fvK2WWIeYnQYLWZKo00nBYTvig3K6656RNDvoaNZHm9MSii1V7npFqXh0oYpQU+ke6oteHVotX1CAhhVJBYtxHX9RxfvF5bV/ZgLwr5kEXcsB+hIQpFBxm5aIAEWAiIl3MkRc+8uPz9qVrGk321g2t/nT9rTfn7R/98Ma8HeHZSyiPHCktU6a0PMwx/m0bkTHNhq5NlXqx4PdxcG/gDg4ODksKt4A7ODg4LCnOlEKx1so01+P0sdPOQIQI1Xmeaunu7J3tYuREjOiMekMTHArSjEgYuXVP3ZQI4Sbra+ry/PhHNNrg2hWNaLDQ9xgc6G68iMioqy7PAcIMvvuG7l5v7zB5R787GEOy9I4eM8bnPna4+4OiW5ogKSUA/YPCHxLAsR71Zy7coimUzGYyziNk4hTyp9BhYZIUaYEwKu60N32NJviZn/7EvL2+olP1xWc1wmerA9qEFY5a0EIZqXsdg7qgns64W4xWOOwhmom0EF552nB364iSoss/2NVIlShCwW1UcUniYnQRzTOGHCqfkxBFhsN8zhtvse9jnudJPacIx9AU6YOuY6HeGKEc5UpLIWgNH5LMTRQFb9TU9n6Aos/QDQkRvROhMDSlVz2D6LVStaPNjs6diBW4EDHyoReVovMQMXX9zbe1H5C+neK+LQo2N5vF5MNqpPOij6inEOsR+94F/XkS3Bu4g4ODw5LCLeAODg4OSwq3gDs4ODgsKc6WAxcrSS5wlIGzRYKZWAggjaENHthiqNVookThjTvKS683waMhvHCTyW4VPeZjH/3QvH1pQ0MEL60rX7W5qrzZ8LDIV06mCKPaUV6LvN02tIYn0DAeD3AujMHBgXLu6winiqrFsKK1lnJs/QnEhnYO5u0AvGg155s9U1KQekykWSb9XNxnghJnly/pHsYK+OkpxMvSks57Aj4xEB2Hl15QXrIOHe6IHDpCBKMIgl6CMnZCnlaPb7WKfGUf2u67KAdn7qi9L2+BNwdvvbaqoYYJ9mpSVK6PsX8xHBW5zhSheZbbFRDG8pJ3ZpQu2q5WRNL8xgzCO4ew0Wik7QS2rIRFha4aEjOnKTNgVfxpB2GSUUX3oJooKxeGyFLGPKqhkjzHL4yKolrG0z5SKCxN1QYhyvxNsH9SRbm/AbK5DfXtUepxWtaiMrpP0u7oWtMFH37Q037sH+rYnAT3Bu7g4OCwpHALuIODg8OS4mzFrDxvLkoVI1uQrtfN2xp29dSmuhzlXxoPnwyQyRai0nStqm7V889qltezz7+g54Fe9npb3ejVln4eQFRr/XIx3CxASbYAYWK37ui5/upbek9v3EA4JGigEGXl6uj3nXt6b1tbxWuvImOz1oK4F7LNYqsu3cZ6lN/PYl3tLMukP5pRDiugsBK4/z1UqB/ATTRlyyJDrVHRfo6Q5RoXNL0xhhX9bujrMRncbj+gKJqef2+/WK5uH6GDd+5o/d/obT3vC89r1vDeHnTsWRcLlA1D1TqglEalUMqU2ZuJZnUa6IRPpzqGWR43au2iczFFjiqjWdjJoLwd7zWxeq+mlNmaIQuVIYWDgc6Lr3/ze/N2tfaxebu5omOVGYYL6vWqDYjVUV97UKQ8dxDa+vauZlxOUZrvKYiiTUZ6H+22fj7A/PCQperh2ZpOi2G/A4SL7u6pXcfQwO8jhDhLH25P9wbu4ODgsKRwC7iDg4PDkuJsxaxSK8PezCVJ4O5NehCiQfZSY6iu2pVNVOEWkdfeuD5vj0bqghzs6nk7Df19+okf68zb1y4r3VCr6fXqqA5eZXYUNtS98u56BaWcXtCM0MBqea87d1Qf+uZtzdC8h6iECqiAAJW+28jqYzSFiMjduxp9cwnRDjVkqzWZ3ZZrIS84WEH8IJhrGttU3cYuSlCFcBNjiFllSdHNrCDVcYzwpFsYw7c2dJyHA6U+nntJKY0qBMsGU42MiRBJQKZj3C3249W3buq176mgWC1QG9y4eTBvG2TgBZGeuFalaJt+t4J2WCu+R7HCfQv9rUKEeopjjqy9+LcxKzYXe6tEOp6Bj/KEoIgMbBeY4nj6qY5PGOoz99rrKjhnQTFMx3ren//5n5q3t7b0eajUdDxiq/Ngf1uv/dbNYgb3t28oddE7ONA/IHP0HvT7h6BcLO61UUdEia/3MwBV2BsUxefGoFQmoO9MoOtOpaaf21OUyHNv4A4ODg5Liocu4MaYp40xXzbGvGaM+a4x5jfzz9eMMV8yxvwg///qw87l8OTA2fV8wtn1YuE0FEoiIv/YWvsNY0xLRL5ujPmSiPxDEfkLa+3njDGfFZHPishvv9uJpnEsN+/MRKUoJNSoaDd6I6USokN1IXpddWNFRA5Q9uigq27LSlPd6xp29+8+ULfok5/86Ly9inJNzZZ+NwS1worxaVISmMFuu8EufAePxz/42z83bw8n2tev/EDdx0PsPguSchK4on6lWHrr2ac0mkZSDfpvIynFh6sXM1togXY1ovZMrNp1iPsQuMQsp1fSPJrriouIpKAieqDTvvK11+btVk3v7ydRru7pp1XDu9FQusKi4vhkpGN7eFAsqfbVb+s1rOgc2cIcOdxXF9mDnT7yYaX71tvaPz9BolHAdpGWmyLCoQlarw8XnElcXk7HhDN6b2F29YyR6vzZZLk0PKNVfXYNjGmykvs/VvsNB/q362/cnrcZwfSf/lLH/0fXlXa8dEkjUv77X1Rq5fJmZ97u7er5/5+vfLPQjQOsL52a0jHf+Kvr83YU6DGNOmihqs5hipfVQmh4Y65VwyLleYDksGms8y0DJZXBxh6ifU7CQ9/ArbV3rbXfyNs9EXldRK6KyKdE5PP5YZ8XkV996NUcnhg4u55POLteLDzSJqYx5jkR+UkR+YqIXLLWHu0s3RORSyd85zMi8hmRYjy1w5ODx7VrvRoed4jDB4zHtmvN2fVJx6kXcGNMU0T+WER+y1rbNYa7pdYaY46NOrfWviIir4iIRKGx3cNZGarNNXVFr2xpleuNFdUICLEL26wxAUMKZZ0OeuqGDSFAEFZR/gpJMzdva2LG1vpz8/YKd4DpvUAbwQ+Lbs3hoZbVstCn+OFbSo9E0H352594ed7u9/W7N/uoPp6paxdDi2E8LWojIJ9FQrh9dAFX28rldPNydh50kBdh185K3ca5y+xBryNBFEkV1cszqxSBNaVyYkGG46CLM9V+vfYtTcCoIwro1i0dzyqSgOoN7dN/9wsfn7evbOoa9q3vvVXox9s31W2PKjonPfSjXtW5sLYCKq4BXXLYIikkduC+bZFH8iCsUoG+S4Cq5j6OSXMBcZ5lEXZdW6nZOLchaROLa4dVUpA6TqODIuU5wVzY2QPdh6r0a52O9h+i6K99T5/vW3fVli+9oDRZ1Wg//u//79V5+xuvFe3aaOlaE3QQ1QNNl0qotlnBerTRhi4RIoJ8T+dgyoi1Ej3YR0nEZkOf8QnoOwstlSQtPhvH4VSvxMaYUGaT4Q+ttX+Sf3zfGHMl//sVEXlw0vcdnkw4u55POLteHJwmCsWIyO+JyOvW2t/Fn74oIp/O258WkT9dfPcc3i84u55POLteLJyGQvl5EfkHIvIdY8w3889+R0Q+JyJfMMb8hoi8LSK/9rATRVEozz89c1tb0JfcWlWX88qGujgNSEH6UtzVvnZV6RG6Z3sHGiVyHUkzodHjd3bVhRv2USW+qW5YBP8nRWRF5hUjBgKjO+f/9t/9h3n769/4vvb1ab2nzY5SRz+FSInLuIf7A233UVap6hVdqjp2vCfQf4ggtxoj6iW0s+PNzNlemF2tzWSaJ2FUI7VrFe41y6hFiKzw/OI9BSgpxVyQnV11Pw960Lup63jeeaCuaPcQyTdIlHn55afm7Yav8+77N9RNFxHZxbzw9vV6UaqPzOUNdf8roFmGQ+hkoBo7SQtKiJY1L6jzE2I8uIfURRKYn9Mj+f8WZ1dRfRU/BGUAaVnSKWMkVZUe14IuyDTRvr/4IU2+kgy2x7TY8LVifC3U+dyHjW5CY+g/fkMjWG7vqeyriEi7i1KMonN1Y1Xt11rVCJ9N2LhVgU4JEnwCH/QUqJVRSSaYUXEWtGwfq7BBRF63X9TnOQ4PXcCttX8pRXqN+OWHXsHhiYSz6/mEs+vFggsLcXBwcFhSnKkWSuD7stGZ7b42ET3QgL5IE4kBVSQ4pKXyFpO+ukYffVmlYh8cqGvyxusarfDmjzQqZGtFKYYPv6AuXKOurlOzqdc+6Kq7G6dF3/Db3/rRvP2d79yYt6mTAe9fnnkW8rDQOemgenbU0oibu7G6UeVK3xyfBna1BVEv/R4qy+zMqKO4JHP5uJglfOTJJIiaQFM8VKX3oJkRRsUpaGNEbYyRyAXKYe2yUlLtTdAYqGTeXNPP44lGp2xva9uf6Jy4s1+MmgjqGnEwfqCU2+GBzrsOJHwpMMMKPuMGQ/FAgfR0UpQlL2rQHeF4JkgamyJa4SiJatFysp7nST1/JgIkpfhoR9AM6kMiddovJkbFic7Djacwbog6SpHEZXxEStV1bNtI9pl0lYr58psaeTLEMESQixYRSVlNCFKzVpk4aSPaZDqhfgkoHjg5FjovgipgWVJInCskb8WgwFZxTxYRYi1I5J4E9wbu4ODgsKRwC7iDg4PDkuJMKRTPiNRz97mJ7D0m6aRwRxLs7k6GxR1dg0ozIVzLTk39pw+/rPTI4X2timMCpUq++arKhiZIHqhE+tu2u3swb3/7W28U+nHrtv7tJtpXX1A52Qz0wf5EXaxBpu54W9R1YsRGiOiSoF7MjKujOHMK1yuDe3cwUld9mBcczk4hU/ko8IwntdyVJj2TIkIhg3arQXRRvaFVTkRE4GnLNIO8KGRnrz2nRaZ9jG0CSdcMEQPtpkabjFG96fsHGnmSlOiHNpJmTFUjHzzMyTEkXSegNA6HSsd4PVZr0cetj0o0vik+hhVf7xtBCVKDzDB5lzSPYLLZYikUY4yEOV3SBWVZxfMagjKoIVIlKCXe0U4hChBPITMcIiGmN1DaqopBqIdqyx60bG4d6PFjjGdUsmuERLw0VRojwvM+hazt3qE+P6aj12YVselEj2GkiW/KlCd0Y0CNZUjYYUCSf4oMZ/cG7uDg4LCkcAu4g4ODw5LijKNQPFlrzXZ4N9Y688+5e97rqav2ANUtjBRdoS3s3EbQRTEoMkwNkq3Oc/rlTG/7R7c1AWDn8ED7Gqg73kBVlHKFj+tv6783L2uSSNhQVy0RRA+gYs29B3qvdyBF+6Gnn9F7YAJMWDRXgKiLEYL+Wci4haSZpJ1X5PEW/bttxebROSmidAJSQZDfZFLIsBStMO4e6FkxVj4KFgdVRGZM9XPKsvZwngqK4lKCxKBaUbtTpHIaPY2oqF3SJLBszBNom1V0Rog+GCPSwUAnhkWCfa9IaY1Ax9SQ6JQkpPiid7YXXGkpyzIZ5sWq4ykiNhJt1/BcNhCd0mwXC3BPPY08maIi0yGS6lIUap5MULQZEWFhC3QYKInqitqvsq+VqjagtSIiYmIdJN5TvaHzc/tAbR9Gen/dsa5HTL6KIZvcAV3XjIo0UohKWUPcn4XmT4Bn0ztF6Sz3Bu7g4OCwpHALuIODg8OS4uwTeXLqhAU7Dw/VjdrZVfdFEFnhldz+BAH2Vy+r+xRA9jKBK+ujiGwCzYVuV92fLtyaOFZ6o4JEoxa0WkREUiTsDKbqYrU8vXaMKjBRqG574Om1fSQujOELV+F2T4uyIZLCPY/gvo5BS0RVjbhp5wWjfX+xv9vGGAlzNz6DS80d/wC0SQj3Pyh5id4EEUkQ1Agq6pqOxmr7bAotDth7BHnerqfjMWWlpfbxtI5IMeJjt6vz8+qm6nIM+qDTMO98JI9M0NcRKEEGjNRrpYQNJHlMQb9Z6HOPMP+PPO0kKU2Qx0RmMxnl1BDd/Cruz8fnLBItfok+wLwPIQrTuaoUx+6ORu9MQUG2IlBj0AySAaOA9PMA549jFEwXkU69M2+nSCIajvV5j0Wfqwl0bSpIJswSVpVCIW5ISjejkhYKKJQEUVUBqCAfNCBln0+CewN3cHBwWFK4BdzBwcFhSeEWcAcHB4clxZly4CLKfe/uambkHsV/wO2RAw/8Ih9UW9vCYcrJhZFyUP2Rcmosz9WHTm+MEMYp+NQRhJRClFjqj7WvIiIbVzbn7Z1tvd4KdLgr4PPG4NRYcuzSpmqDM0RsjNAqSYviOCZRTo7lvSjKNUAGaz3nf08TnvRIMGbOcXvg17lvwUw+hkOuoK8iIg2InAV18OHIgGTB8wlCJmnXJsK5Wshg7SLka3Ln/rxdLYWprreRiYlhX1tHubtY5zA59GSk82WCeWTAC4/Hmr23slos00eRJR/7ISMIfQUQNjvSSFpsHmZ+wvx5rTBkDzxtyBBecMHD0oaNmWqYaxsl56oVHZPVdVWUilP9vE7N8a6KkSUIM61hnFbqyAjdKIaHrjT0Gvfv6d5IvaUc/SH0+Id4/rpj7C0hRLACQTZmkmdpUTTOMEM05HcStBFSGDx8eXZv4A4ODg5LCreAOzg4OCwpzpRCieNE7t+f1VI96KpLPYFba5C9V4VY00YpUy6sIMMwRRgPKIoUv08MYxtP1RWaQIN4CHGczFO31kOsG8WMRIoubntNyzLduaXl3C4/o5l8HrJDn3leQ9LacO14BYZbTkrCPNSkGo71vpsoqVZDBqSGhC1WzGp2zqPK6Np7ltGKJ9q/DOGWlaxIC7XgUjM8y7AcVaz972zpvFjvqPs5SpSiqCL9MgbNkmTav0ZJl7y6pterhKjInmpI4cZ6Z94+OFDXfgT6YAL9aYuMy3ZLNac31nXeiIgcQjd6ivk8AQ04ZRhbTjGkC+dQZD7JaiEoA1AopE1iamSX+sI3RZb5yxDeG5NyAJWQISywgrKHBvajcNSVLR3b3T0dv9nJ9N+bl3TuDBnuiWzbBCX0Ejw3Q+jsr7SRYYm5LZXiWmERYsjnkpSnhzGM4yIFcxzcG7iDg4PDksIt4A4ODg5LijOlUNI0lf1ct3eKCAzSJo267vKvtdXFsaVyYsOpfr+QhQX3czCGtriFMBbLNTHTEe6VR+Eh0B5+rRgNE9SO13v2Au1TBs3j1iaoEmpLY7c7Q0ZdHZXds6xIfdDFyuCzspL9SlPdySMqYOFRKCLzEAiDc7PsF2mbEOMUvkNYS+9jQs3lqVIiFbjdEul5fc5mRGxk1HmH61qDSFmWFaMmpnDbh6CCAotycKG6vgegZjhHghqpONAyVbjjQ9TcE5HEYk7VNGKDpbsyRCBp3xdrV88YqeURUaRNKLjEUnIx+m1KHAqjyCqI4BhDezvBPXmgizzox9dQZoz02wSRHNOh0hu1YoCPhMx4jvU+en21wT7pXQiI+dAMrzfV9hVkyB4OlEqzlWKGbZLp+AxA145BL5JOaZQzdI+BewN3cHBwWFI8dAE3xlSNMf/VGPMtY8x3jTH/NP/8eWPMV4wxbxpj/q0xJnrYuRyeHDi7nk84u14snIZCmYjIL1lr+8aYUET+0hjzH0TkfxGRf26t/SNjzP8pIr8hIv/q3U6UWTunTgyqsK80lVZoN3VH3sJVS0v0wQhCNiwRxiCR3X0VGyokCcClHoF+qUA7WxB076Gvnl8SPWqoyzMaaeRJDa4Qq0vXQMeglrUc9jW6gTrapBSoSy4i0mNZLbh69F5JpxyJWeU0x8LsOuul5blFpBiFwvugGxt4Zbdf/52SEsA9WWiOF2ikE0q4pRgnD1EuIUTK4hKFEqGP6+s67tAfk0pd5+po/8G83cTSmEB/GoFCMka0Qrekid5DIEMYgTJAsla1qtTYnK6YPS8Ls6tnjFTz6K0KqEnauFCpneXAStRYBNqLCSqHQ9w77ETKzeJxmIJetOhHUNHnrY6EqapfjEIJIz1uB7UHLCJGrK+01RRRKAba7FevqQjXAcqutdqdeXuS8AkXSRBZEyGyLcQ87w2U/iG1chIe+gZuZzga5TD/z4rIL4nI/5V//nkR+dWHXs3hiYGz6/mEs+vFwqk4cGOMb4z5pog8EJEvicgPReTA2nkZmVsicvWE737GGPM1Y8zXpvHi448d3jsWZdfxJD7uEIcPCAuz6zQ57hCHJwinikKxMzHgTxhjOiLy70TkI6e9gLX2FRF5RURktVWxR4ksq211QeqIPKFblMBVHk+LQe2HPXW9dvZVg6SCHWsffIrpKpXAsmYeNavh2lWgyeHBN0ymxcXqEIkdzU5n3g7hAgaMHsD3U/SjcN9w53nfZU10UirdEVxOsAE81xGdchSVsSi7bqw17VECD2kTusGMXAihl1J2tRnVkDKqAfQIx4ERLX2UNSOPlFjSKdCKHqHcGeaNiIhnYH8kAgXQqaGOThXRLWmK5BuU4RoOdUH0kDjEyCsREa+gJY9ELkRX9Lvqtod534/K2S3KrpurdVvJqSQmVTFBjslDjDwJSlRjBdTVCKXMKGxjGKmEOUKt+0NEiFSaWDcQdeSBc6k2OoV+WFBjHvRZmghXIZU6Ap2VQndne0frFkwTzOdIqS0rRS3yBOtACu16rnMRqKrsFAl3jxSFYq09EJEvi8jPikjHqDrLNRG5/Sjncnhy4Ox6PuHsev5xmiiUzfyXXIwxNRH5WyLyuswmxt/LD/u0iPzp+9RHh/cBzq7nE86uFwvGlkULygcY8zGZbXr4Mlvwv2Ct/d+MMS+IyB+JyJqI/JWI/E/W2nfdNjXGbIvIQER23u24c4oNeXLu+1kR+WVZrF3flifrHs8KT9I9O7suDk/aPT9rrd0sf/jQBXzRMMZ8zVr7yTO96BOAi3DfF+Eey7gI93wR7rGMZblnl4np4ODgsKRwC7iDg4PDkuKDWMBf+QCu+STgItz3RbjHMi7CPV+EeyxjKe75zDlwBwcHB4fFwFEoDg4ODksKt4A7ODg4LCnOdAE3xvyKMeb7uaTlZ8/y2mcFY8zTxpgvG2Ney+U8fzP/fM0Y8yVjzA/y/68+7FzLgotgV5GLZ1tn1yffrmfGgRtjfBF5Q2aZYbdE5Ksi8uvW2tfOpANnBGPMFRG5Yq39hjGmJSJfl5ny2z8UkT1r7efyh2HVWvvbH1xPF4OLYleRi2VbZ9flsOtZvoH/tIi8aa19y1o7lVlW2KfO8PpnAmvtXWvtN/J2T2ZpzFdldq+fzw87T3KeF8KuIhfOts6uS2DXs1zAr4rITfz7REnL8wJjzHMi8pMi8hURuWStvZv/6Z6IXPqg+rVgXDi7ilwI2zq7LoFd3Sbm+wRjTFNE/lhEfsta2+Xf7Iy3cvGbSwpn2/OJZbTrWS7gt0Xkafz73Epa5qWs/lhE/tBa+yf5x/dzru2Ic3tw0veXDBfGriIXyrbOrktg17NcwL8qIi+bWXHVSET+voh88QyvfyYws4KBvycir1trfxd/+qLMZDxFzpec54Wwq8iFs62z6xLY9UwzMY0xf0dE/oXMpC5/31r7z87s4mcEY8wviMh/FpHviMhRSY3fkRmn9gUReUZmEp2/Zq3d+0A6uWBcBLuKXDzbOrs++XZ1qfQODg4OSwq3ieng4OCwpHALuIODg8OS4rEW8IuSanvR4Ox6fuFse77wnjnw95JqG4WBrVfDd3x+Ug8s/1I6yPf9eTtJk2O/nySp9veE77IdT6fa5neNftvzi795nuehrefKskzbqbaL423xOT5GZ3lt3/AuRILAO/a4ONHxsDiZn/dvPIklTtLiyfQ878Gu/iPZ9d3A8TEGY8txx4njOGbf0dZjCnMFYxNP9btZqbO0qynY+PgxT1OdLxa2L1i7fJFjzvOOvuN6YYD5ZfUaSX7taZxJktpj7Zpf55Fs63nG+se84pX7e/zn5XvVv9HGfMY9Tnwcw2EzPuezdo425rXSpLg28Dk76Vnk93lLnqf/CHk9HMM1x5bGycN8znBtz2j7pGf6oDvZOa4mZlD+4BEwT7XNL3aUanvig16vhvILn3z+HZ9bOX5CFBai0g9Nu92et7f393myOfZ2dcPYxzVW1lbm7dV2a96+c+PWvH1/W8/pVaJ5u9VsFvpRrdXm7XpTzzUajeftQbc3byexGjiFsVMs8h4maRjqRGmUFsmNVb1eNdLj7uK+E6sTolGfjdk3X9f7PAbvya7/zV9/RkRKD+cJK/hJj42IyHSqNq9UqtpuNY790v279+bt0NfxCUK975UVtdn+9u68ffuWfncIu4iI1Bpq12oV7Vpd+1TV/u3v6ZhPBlormPcXT/QHw+Kh9QMuPiIV2LzdqMzblzc68/ZoPJy3dw77IiLyxs2+PASPZFvfE1nLHzPvhMWSnwd4gbEly3Ix4wtXmui8rxldjlL8uI6t2qbS0DFvNtQW7bY+01z4DvcPCv2Ix7h2dvyzaE58/nQduLSmz16A4+/t6bphPT1eRKRa0Tk8Huv6EIXap83NJj7X+fwnf/7G23IMHodCOVWqrTHmM8aYrxljvjaNj39Tdnii8B7smpb/7PBk4qG2pV1PcBgcniA8zhv4qWCtfUXy8kSdVtV6eZgl37rNCc42ftje4eLyV/za0zoH47HSIDX8gq209BfTx11X8FY76OkvpI9f3lano/2YFDvCt6dGQ7//1vaO9jXWNzLP04tndLFOoE0M3Da6VyIi1Uj7Pp6M5u0UgxVgDI7eiBfxXNKurUZkD/uzN4p6Td86QvjfBQcVr+b0PGZ/039PY31L8RK9jxrezLfW1ubt9Y6qfY6ng3m7UtfvJhN9c736tEpbGK/4KNgELj/ec2p1fevbP4RnNdF5RxToM4xCgZbzih6oj4lfCbVfdbwB7uyrJ5HklM0iIoJp1yg0Njqac+gvqRy+jRdppNKPOr5fw7xd2dA3Tg9jtbejHo0RvcZKU72hZl2ft2ZN58R4pM9bOtX2rL96jTY8M47dFFRqp6Oefqetx6809XocjynO0x8Vr12rqS2jSO+jEmifGvD2TmPQx3kDv1CpthcIzq7nF8625wyPs4BfmFTbCwZn1/MLZ9tzhvdMoVhrE2PMPxKRPxdNtf3uu37JmPmmR3EnunBiPVzgXpdcsslU3WuLvZvQqDuzvqIbG626uibDibq+U7jUT11Vl9qHSx1PcO246O4Wdsixy7xSVxerWuWmiLrBQagbVA8eKOXiB6AL4DaHJVe7Alf0oHswb5NCqQT6/XlwxLt4Zu/FrkmayU53No5+X6mcaqRjyA0gtssRDRVsGE+N2n88VkrEYHO7io2veKTHzCSsZzjcV2G5lY66wevrSrnEk+L+TDZBVA84fm5idrFBFvqIKvDgEoc67/p90lx6jF+ya0TapKZzZDDS78cJN71n8+Ck6JAjPKptfc+TVrOet0EpYuOfETojbMwxwkdECg8KN5W3tpQCSyZKOQQYzwibxY0aKJcMtAyOqYHWMbZT6EZU02eGgRBjXDsFPdtZ1TWEa1AY6rWbTaVyLiFooD7Q+SgiElWwyZvpNXyMTbOi84uRbCfhsThwa+2ficifPc45HJ48OLueXzjbni+4TEwHBweHJcX7HoVCJEkqD/ZnfAfda0ZTRIj+8OBChFV1JUVEJp66HcOhcihtuCBgU2TUVzd6FKtrM7XqyrQRqRJw1xy7zCFcNRGRGHHLMXbhn3l6Sz/HNUiPxAir9H39bh3RLHVcLyi5yG3QNL2xute9yaF+JwDFMD0Fh/IeYIyRKLfhaKyuc5xqe4jPdz2lreqVUmw7YvSjKl1hRNYk2q4i0sWDizsBNXY41vnBO2+CDolKUSgRogzSVL81ATVweUspmHas8zMENWZBq91AnoHgei244CIiBvOIkSeHfaX+YjwbUR6V8zAK5VERBIFsbqy/49w15D74iHJaBRU0LUVghKBa2k0dd4Tui60wdl/3Wj08yAb5ZxWfNtIxixMd284qojpExK8cHxkVjZloRHpLn1GuJ5wuPdhljHnONU5EpNlUW1pQLXaMnIAM79SniON0b+AODg4OSwq3gDs4ODgsKc6UQkkzK93hLDqgP9IoASYycEe+BveaAfUixR3dRg2JL4gmqBgmROi5hhN1qcfU0ujq582KusEeU3zTYjTMFAkcTAAIsNtdBf1jCveqv59bm0odhEy+KUQrFC4tIdzrtfV17Qdc/oBUVd6Nsp7L4yLwPdlYmbnFcVOvPUbq+AjtvX2leHqlNHJB1MxqoDYngRbB+a0gK2sFiRaDB4haAW01GSPSAVQMqS0RkTRUOyegyaagUCp1SCwgocggYmM4UGrr2jWVsmhAksErSUlYJLRsYN5Pb2g/sq5SREeUDaOgFgHPM9LIE2SoIZPGOs+Z8FaQGfCLdq2Ac6jhO/2xzoUEVCMTpgSp9B7eOY3Rz6exjsdwqG0TlbSLYv03U+lJETEJKSNlg+e4jwiT6YTH6/krleLyOsHaxCgUUihDUH+ngXsDd3BwcFhSuAXcwcHBYUlxphSK73vSzrUMJojAoMjVIWgMUgypKbpknZXWscetQGOg01I3jIpyvVQTDpIpdrjhJnbh5lMJj26eSDHKw2DXeAr9E36HGi4RaJosUdfc95AwEEX4vPh72xsgugJRGpe2lE7xAvQpT1agetoi4HtG2tXZOMRw/2tQBGyCbhj2dWzLXv9wiOQHUEGbiNTYuNSZt7c2lGLogx6pIHGrDR0VAWU2gqsd1opJExmSyALSAYgWmsRKjyRDStOC9kKYxUpboyaMd3KSRgBlSUNVxVWlaXojvV6YRwCVNVUeF0ZEjrSLPNBWpBGnSKiLM9BLXvF5DT0dhyloiQzPfh/JWkMkLTUjHTcfE2YgOv8LqogRkrDSYkJRwtzAgp20v4F/PKVLpcxCIhXoN2rXMCFLRGQ00v72kHgnMSNPqB4KGukEuDdwBwcHhyWFW8AdHBwclhRuAXdwcHBYUpwtB+4ZWW3NeN8kVU6U4kvUa2Ye0mhUDK+x4KA2GhqS1Wwpr9xug3eFLnAFoXU1hPsxy2uCbEivxmzGkkgPKliRHg+940MPGTJm8TmFu0hlUi97dxeVh0QkDJQjiyd63OoaMjlb0AzP+VtvwT/bvufJal4pZYSwyhh2pY0/8vy1Y48REen2EAKGEMo69hoqnl6jP1Dd6D7GoIbwsVWE7CEiTcYd3UeJTZGTLpRbAxePBLpCuS2W42N4GoXMGN7WwL5KGBWzjAXVeg4PNYOYem5Xn1LhtUxmE6+sF78IHIW9+gg9DcHtpgH4bGral7SsyYg3oeduMO4TGId7OuOh8uEBs7YRRushNDGJ9buhX6yKw/OGFI1jdilWxclEOf7eUNsZRauQLR1xr64kRpUgXHAyRCgsBNxCtE1Y7PtxcG/gDg4ODksKt4A7ODg4LCnOnEJp5RmUBdoEnsYLV7fwBWS0jYolqxJkP7XgSkUomXR4qGWnRnA/Q7htdfhLa9CHzkCz3Lx3d95OS652Ba5XDaGKE/Y3hPgS3C2WzppM1b0es6BrgszBsCjME6DvHiiYJNVr7++pCx7k7t1J1dHfK8IwkCtbGyJSrBLfh8sZF4rGot8lN3FvX4WBunsabthuIhQQrvphV0PPYjjqVYTphbj2pUtafm+AMXvr7euFfjA8tAI6LR4r3TFF+T7SKfxuJnpt36jtyTCMxkXhpy5CaSshrj3V/m5uUfRrNob+gsNDkyyT/W5OX9jj53AE4a5CqbWkOMdYRm1tVUM//aG+Q0KDSiKG50KPPWrq9SYIEaSGd7up589KjGeCbEhSQbWajrPFeTOsJ3WU5usz+5LCVJhTcanMHmnO0AfNyULp4E97/YcWqXZv4A4ODg7LCreAOzg4OCwpzpRCCQNfLq91RERkhBJGCVzcAOI9PrIQpyWX7NZ11VbGxm1BOGrnALQERGba0JleR3moKnbxX//R2/N2DOGZRrOoB54URHToLtO11A5WI35fj0+RqUY3jEXbqyUtcgo5eREyLkG7HOwqJXGU2UhhooXAWknyTFK60asoncXyUHRXTSlKgBl8Bll+RpDNmmFexKyQTnEwzUZdaei4HYCe+OGdO/N2EBbfZeoQIhqNlMqJoRsdeGpjUmmcg1mmbvRgpN/tj0Av2SL1Ua/rM0DxpgBZtWNonMf5s5GVhNYeF2lqZS8vAzdGfyM8cKuJUjlViLCthMXImuYaSwli/kGQqsHMXTyja5fUlmREbtzRNSBAfYAMNMl4WORQjEXU2VRtzCxLgzlM4TfL7GrofgvoumZD77tAjYjIeKDfj0Ad1ms6P5m13ZeH29O9gTs4ODgsKdwC7uDg4LCkOOMoFE9ajVnyCcsqZdjh9kELhKAbqC0tIpINNWJkgiSfnT2IDSXQ8kWyRBOVvtsNdVfv3Hswbw9Ah2ygcnZYlI2WGBRKa7MzbyfqccoUrlPaU0ojY/VstFmSy8+Y4FM0lwcxngQ79f2uJj6AhZBw7rYt9nd7MpnK9es3RESk0VC7roCeWl/XMayDbmBClojIFpJrAqRyvfrqq/O2H+j4NBqdeXsTCUxVzKPugY75zfua+FOpqu0bzSKV4xl1ZQPoXKdIGJlijmSoOM9IgoJgEmgyHwkw07joKtchYsTyahQzO+xpUtdgb3Z/pCIXAc8z0szL9o0hMpZAD3yISAmLebvSLs7VFPTDAalNJN34sHfIcoqgEm7d0WeUYmdtiJ1loCFWmqgqLyK+gK4AvRIhumwUU6hK58X+oY55MkJ0SguJSYjw4vMpIhL5+rc6aL0M9A+f41ajGHV2HNwbuIODg8OS4qELuDHm940xD4wxr+KzNWPMl4wxP8j/v/pu53B48uDsen7hbHtxcBoK5Q9E5F+KyL/GZ58Vkb+w1n7OGPPZ/N+//bATJWkq+wczN6SGiIom3J8mS00h8KRS0jquPqt6Gt/8trrXD7Y1eadeUxf58ibcHIR23Lxxc96+i4iNoKquvIdog3Gp5FFUZd+praGuUA36CB70HvpIzKAe8e6+uvnjqbrdQb24sx/V1XxkIibIWooC7V87pyd8f1tkgXbNskwGedIOq7ZPYyZaQLtmBbrRYVE3mroQo7GO9e6BuuphAC1sD4lRmVIPO/fvzdvbuzqeuxMdzxdeUj2RJAEFIiIW7nxntTNvm1T7G/ug9aC1U6BEkOGTwb1m25QeQ+q1cK6y4nnvQLmxwXB27VQphT+QBdg28D1Zb83c+JqvFBijUDxQVRNQhWFQfF4jJPIMcR/xlOX/dGyvXtGEqz5ox/5Av7uCMnYWOuEp9P7bSMKa/VH7GyDxy8egH/SVKhkNkAiHiJQKtHaoGc4ks2lSpH1XVpXOqWD9uw+Noz70clZWivTPcXjoG7i19j+JyF7p40+JyOfz9udF5FcfeiWHJwrOrucXzrYXB+91E/OStfYov/yeiFw66UBjzGdE5DMixaosDk8k3pNda5HbSlkCnMq2tGu1cqYxDg7vAY9tIWutNcbYd/n7KyLyiojIWqti7z/YEZFSwkdHXYUUXEADxwQlV7sGOmF792DeHo8RkC/q0vWwg1yHJkWvpy7SXle/2wmVyhkOEcBfKtHEylF97Mj7SBiImBQywA4+omdsjDJQSCqoePqj1yjphlA9tI0SYgfewbzd6nS0T7ksrjmFnuyj2LXTjKyXlw5LET0wglZIt69RDOtIXJiWIicyyAzfvretfzDQu0EUiqSgLpBwY7HjP0JiRwa7sGJ8aot6JIIolAw6rlPMrxrsQQnSAfRcQthl/+BAr40IliAsJmj50OHxMcEO+6DvMj2m1Zoluvh+kQY6Ce9m28Lz2q7ZVp5Qs9HSZJoK7Fer63OyA3uxcr1IUe9lUCg/h8icirZ7hzqGD+7f1+/i+Wl1dAyGA+ru6DFRFSX6RGQ61nGvR4iEQxnDxOr3LaiVFujdWp22137wJdUrPWdMNNs70Lk6AO2YQFclleKadxze66vTfWPMFRGR/P8PHnK8w3LA2fX8wtn2HOK9LuBfFJFP5+1Pi8ifLqY7Dh8wnF3PL5xtzyEeSqEYY/6NiPyiiGwYY26JyD8Rkc+JyBeMMb8hIm+LyK+d5mKZtfNqzpST7Q7UNYxQJaPaVPczLiU7dA9Q7QVuRwh3tBZyx1ndlMuXlP5rr+g17rym2hj93oF+N1CKxy8l8jBSYojKIVVKv6Lqz2CKhAhoXQakShoazZKhSrXnFb1eJm7EqJC+1lHXkJEq4yMNFGsXatcwCOXy5pXZNRClEyOLqAJ30oOuibVFXZb+ANQFaBODCkdXL23M24Ho9T7yskYuMJrJvKFywNfv7MzbvS5otZLGDavG7yKKxRdUT2kiGQeZW31UH09H0KLh6xLK13RgbxGR0KgLX4dE8aFRuo9JHkc0je/P7nNRtg0DX66sd2ZdB5XTQGRHHRRKExRBt6t9FRG5fv36vD1B9FUdyVQtUEd722qn/T2NLBsOddy6qFZEmpJPSffwUIgYiUNDo89MA89Js41EKkQRJfb4aJMUz3EMu5aUpwv6Kd2+zgsmC1XAi2aneL9+6AJurf31E/70yw89u8MTC2fX8wtn24sDFz7g4ODgsKQ4Yy0UX5q59shkgggAaA/41A+Aex2nRVd7hKCBJipweEh8uHJJqY+nNtXl/OhHnp+3E4vd/I1n5u0xEjNS7ERPpsWdfh8yrp6n10BShaQpitmCFqpUNJrCIoyhgp15JkpUSuF6KRJl6oUdb20/gPt/lCSQLlh21Bgjfh6RsQktlCrkQcNA7WI87LqnxXtixEiK3fk63Ov1VR3DpzY1meOTP/VhnFdd2TTQ+fHyh16atycJo2RAdYjIBDYrJnJp/3pDdc9DjH8FlBnDlEzAyAWMhynObVb3GY6UcjOgATc39L4HOTVGd38R8HzVLjJIVum0QU9hngegoVbaxQSaV1///rw9RBSKZ3VM9nY1iuX5pzrz9pXLm/P24I7aaYwkmyhUGqqFSJA2qv+IiLSbGk3TQqJMAxWfkrHSYW+/pYl+E0hajxApNoXU7hiVuMr2mEBC28N4MqkxfcRqWe4N3MHBwWFJ4RZwBwcHhyXF2VIofiCd9kxDpxYikgDx6pT7ML66+jYpuiPTGJUruupWXVpVV2prQ129j//EC/P2M9fUJeshoSKq6fF0ZLj5fFDa1bbYjn7uxY/M2ymq6nR7uot+94ZWEXn7bY16QX6IxDGpI3XJKhWlJ0REKnV1vVjg+OYdjbrYh8zpEXViH81LeyisnenciIj4AaKAUGy6iq6PEKXBaIzZv6H3AfqngSiDpy6pjT/5iRfn7Utbqs9EzYyXn0OFHEYVMBKqX0z4uL+tySMvfehj83aM7xz2NVJi564ef++uUgHDCSMU9B7akFttNIoRMBYUVwKarNFWOqCPJKR+Tv8w4WgRSOJEtndm99hAQpIP3RGq83iI3JFS0plQD6hAH7Agufb/xRc1oujSJS10fu2e3rePhaO10sDnev57d5UCERH5uZ/7OT3Xs8/O2ymS/g62VUdnACni+zv67AdIFFtBJBths2IYCgs795AcmEImegC7xqeonOXewB0cHByWFG4Bd3BwcFhSnK1ajRGRfLe+xmgFBK9XqtTSQCSAKf7WhAK9AuwCN7bUzfzQC1fm7ZdfuDxvtxkJUoUr1GYlDh0aVgy6lRU1HjxfXbfLa7rDvXpJowSGwwP9wlS//wDVYRLsXnuQoaSuw3BYdMmyE3RHhkiUCCrqnh85cKfRQnkUZDaTcb7DHtfUJR4jQmHap7QsknVKXmIDVZieuaRU13pLz/s3PqZ02MvPq3vdgixxrab32GhibHHvMcbv9u2ixk3Nf3refhbSphtP6fW6faXGXoOk8e62JghVrLr5CWw0nkA3JyxSHyx4W0XCDmVxdyCvO8ylfBn5tAgkSSLbO7N7HCJ5Z4gQsKdRhDzFMzlhKSgRWVnVZ2M8UuppY02/v1LXcfjQyxoRVgevGlSV5gxBSdRq1CBhGE8xoSgFRWGRWHPpKuRyoYXyzNNqewPpYkrOhoiQYhHqZFqcU9RMMZnOHQvaMUV0Xj0q0VDHwL2BOzg4OCwp3ALu4ODgsKQ4UwrFZlYmOYUwnkAy02cQPZI8kOBTdrXb0GD4+I+9PG9f3VJX78c/pO7P5TWlbGrQnmi01T0bIYA/TY93uye9TqEfvb72cYpqGr5okkAT7l0TBXJXO+q2xXCpggiREpBXTUq70jESC3wfBZxBm3iIfLA5FeR5i034yDIrw/HM9esNQFEgmabR0HFmceZqvbiDH0Pu8+WXnpu3n7/WmbefuaoRTBuIPqjDrmED+hIjFHmGi9qA/syopEdyQJoHtBe8ZVltYU5VWWhZ+zTaPz5ywUfC2XRcTA6b4N9jUmNIIkoEcsW12fUWTY1ZETkKiBlArtVDREnW0j4xOYmRVCIia6hq9NYP3tZzoWDxc88q5XnlMigNzO1qRdsBEttqkLgNEIVih7oGiIiMoNGy90CjTbxIx7mK54/FktdW9f76EyZx6fmbFT3eVoqUVoZFbIJ1IEPB9WSibSb4nAT3Bu7g4OCwpHALuIODg8OSwi3gDg4ODkuKM+XAM5vNSyvVkYGV9KEbXWG1Zwg5lbnSkfJJW1uoNP6Chp5RdKdW0++zqrYY5JJhNEbI0GRZq5V2MRtyAN6bwlHDVDP7nntG+8Q0yHa7o8dP9TwBwiotMhVHoyJXmuJcuwfK7fkJOEpWAJ9nYi423CzNUukOZvcbQ+zrWXCaxtcxJ586ZFamiKQTHfcqOONGSzXcyQ0GCLlL8DrCquipZUgohMLAGTdbRb6xiwzWvcODebtyX8d2bU3nFLWfN9Y11DBBqGKMLNNmrNcbT4qhqVNkX07B/0bMxEWJwaNs4HIJr8dF4AeyupKXa4MWfQvPomGpQzbTYsirl2F+U0wNYbIvv/DUvN1EGGi7qfsTK+s6BlOUbYtjnVMsXba1VRSzuntH+3ETJeC++uq35u2PfPi5efvBjmZi3sPxSayhlCsIbaxieymsFedUDLuOECKYctw6eq5p7DIxHRwcHM4t3ALu4ODgsKQ4UwolzTI5HMxc5sFAw3CuXVWKoRqoezZEduJgWNRrTqbqhlXg3llPM76oxZyh7FoCfW8/ULcmgra3AbXiwSXr9YvVyw/h/t64rS5WFqurtr2truH+rlIdXYhwZZYayRA6qqj7WC/Vc+N9jFGWroYQpclEwxOz3IU3slgKxRgjfp4VF0Q65n1k7E1QEm8MIZ+CUpiIBBCbggy3HPQ0822IMR9CBaxRRWgdQtpSZDoGITTYMQw7pVeZPVRRv3dH6/++dfNH8/YLz2mI2mEXIaS+joEf6EVqmNshKKX+oCikxQzRLNP79hKdFynDbecZmIu1q+d50szpixaE06jNXoXsd1G7vIhVhJF+7CMqQHZ5ozNvX9nUdhO2bCITc4RJESJEtnsI7fimzv+gVnxmbtzR5++H9/X5+/5rN+bt23f0Pg5BTU6nSve99BJ0xXENg7DPwBSvbRkOiSzSFPauIQR1e18p2ZPg3sAdHBwclhRuAXdwcHBYUpxtFEpmZZSXKmvV1BWaJOpCWESkDCE8k5WEenxkg1m4yHeQXXX7HjIdUSLtMjL5WoiGmU702lHIoUEkwagoUPPmTdX0/sFNFenxQQ1cv65usIHgT6ul99BZUZeqhVJrEegbit6IaIk0EZF2Q3evx9Cs7sB1razNskOj8LosEp7nSTWP8mHJuS7sF4L+iUF/lTWTK/A6LYTDbt/VcXj9+3qvg6sa8fGhmuo7B6AukomOU7Wu7nWKTNbhQZEa+9YPrs/bdx4czNstZM3dvAlteFBg4um5KsgcrIFaiVBeLWwU36OGEIIacC4gemoI2iXJ6Z7yWD4urKj+OUt9WWTyjhFZQZrMeEX6oIWs1fpzGhnSaaLkHKLOPFBMMaJWjK9jUwWVk6EkWoQx9xgaI6KppSLyrVdfm7cHfZ1ryVSf1wmoNGZoViK9Hx/jkRjQvhDjExEZgvbKQDKliNhJM70eo2xOwkPfwI0xTxtjvmyMec0Y811jzG/mn68ZY75kjPlB/v/Vh53L4cmBs+v5hLPrxcJpKJRERP6xtfbHRORnROR/Nsb8mIh8VkT+wlr7soj8Rf5vh+WBs+v5hLPrBcJDKRRr7V0RuZu3e8aY10Xkqoh8SkR+MT/s8yLy/4rIb7/buTzfk2ZeFsqAljhAOatKhKrkMfSkS8kOhUSNVF2TG7d1p7gDIZo3sS3+Uz/50Xn7pRefm7cHoG9q2PlOEbSfjovB9d/47vf0PuAxraI8mw93KUAF8k1UVKcwkg9XNGCGiFd0B0O4blWUl/IRXmGgfX7kBhtZrF3FmLnOuw+Xf4xEhATZCiNEoURh8R0igxazATWGymTyl/9FtbfXkFi1va1RBU+BJrt0SYXFbKJGGqHs2p37Gq0jIvL1V9WuxugY2paea29X55pndO589KNK66xCVCvIQKGAootLCThTiKq1EZXQQ4TDRkc/97ycGvvOXbF2vDC7pmkm3d7s2YyQXGYwJ1OrNBkpHmuLc3UCmmAA+oe0yYPdg3l7c02jryoQtWuuItGP9GKiY2ZR2m2CpDEREd/os3HzlpY3TERtPAl03kaBUjOVUNujEZ5LCJ+x3NwhksFERHYhbEaKyZKeEoionYISeyQO3BjznIj8pIh8RUQu5YuAiMg9Ebl0wnc+IyKfERGpVcPjDnH4gOHsej7xuHatO7s+8Th1FIoxpikifywiv2WtLZS5sLPc7GODUK21r1hrP2mt/WQU+scd4vABYhF2rURnuhfucAosxK6hs+uTjlNZyBgTymwy/KG19k/yj+8bY65Ya+8aY66IyIOTz6A4Kk/mYzGfoIQY9cDjyclaABmpBYF/XVfX8tuvqu7wsKvu7o0b6i6vdlQDQRAh8tGPPjdvv/ScanrcvVcs0fT9H2oCQD/WN5ZnoYdBN7hV0/uu1fRzgwSWzOq9jTk2aTECxuK4KiMc4FqyJJvJjz+61KLsaq2VOHcjA1ybSTNBoO0q3GZjivcUMCkLkUfdvs6F73z9zXl7taVjePOWuqiVEJozK3rtv/u3f2beXm9rAtl//bZSJiIib719d95uNZXq8mK1Ux0hM5trsDEigmpIDkuHeq8+5m9Weq/hWxWpMVuBhr6ArsgpuqNoiEXZNU5iufPgvogUNbb7A+17ewVJM1hOjqiXIxS1/RFVMtT5+Z3vqV0z0BIV0DfPv6Cl7ppNHefDQ71erarjbJMi9WqhqX/vng7BrR2lWl7c0sS79ZWOnrei7W5PacBDy/J4OgYH/WIUyt6hUnwW0TGMlAk80jcPp1BOE4ViROT3ROR1a+3v4k9fFJFP5+1Pi8ifPvRqDk8MnF3PJ5xdLxZO8wb+8yLyD0TkO8aYb+af/Y6IfE5EvmCM+Q0ReVtEfu196aHD+wVn1/MJZ9cLhNNEofylvFPa4Ai//MhXPPLfEVFhUVYpjCAbi91k7s6KiAS+/ttgt3YAGdj7N3dxjLqye/tKrYyH6nq1VzRRJoEL1wHV8cNb6lqLiEyQDLK7dzBv11J1r/1NvdfVtobfDuE+WiQY1CA12R/qTnZ/UNzVblTVfbXY5Tdwd4eQoDW5i22tXahdrVix2az/HvQewkJVbXX2PCRPUWdGpGBysWM91607mlzho2xVbUWjTQ7h2ncP1PYNJMr8LCIBalYTSl79kUYkiIjc3VN3d9DV80bQIwk3NSKlCrniMWjAAAld1F4Z9Pv4vKQHg+giQ/ca9FtcSPIoRBctzK5pmsl+rtdTReJdA1ROBdFa01Qpg8GwSF1Mp4jUAGUUI0okvaH0ZDLVMb9/+/68vbWhGiQRkqpu3bw5b6+tawTL3/0ffrbQD5YlbENu2hyoze7vqAaJxbLTqOu83T9EUtX0QO8HX0hKBAdL4vFPFmuNSVDyLzrJjMeexsHBwcFhmeAWcAcHB4clxdlWpbdWJvmusIeA+gpcbR+77gK9ExMUt+pD+GF2qucaD7Q9QNLNU9DMWMXOcoqkmYM9lYPtQQr11g114d68VXS1I0hXhtDT6MFFbkHeNUnVbd+DVKUHWdsAEToD9GNUcksDX13WCsaHUX0ZqsAcJQwwcWAR8IyZx4IHhaQUdYM9nxQPInHqTSGmoEGorUGXfPOaRo80W0yYwhjALpVA3dJbN1Sv5tbboEniYjRMrVBZHvouh2rXtY666qzatLt/MG/HLaVWmJdBKeGgpBtSAzVg8JxM0+Mjkrw8ecQuWibY86Saj4PxdVJZtPcQ3RWiOtIUejciImMkIfmotBR4eh8hzvv2baXMvv+q0iONutIbCRLFLKSVtyDt+93vqfyviEiGSLEOImg2UDRrcP9A7wNRLLv72qfNLZ13pEZ60N2ZlvSbMJ0L4xlgLUuhveJTT/kEuDdwBwcHhyWFW8AdHBwclhRnSqF4nieNfLeemgSBz113BPAjUcKPiu5EALnJCdy1PiQc6211X1e21N1NEfGRws2vQX9hysiWQ3Wh7+6qCy4iEqFwKeUtK3B/+gNW3tEM5v2efj6cwqUGpRTDDStLIxhE71QxhjUksdQQMRAdaTmYh+9uPwqMiHh54MN0Clc5QAUZJB3RdmUwqSHNoIuDhJ+tpzvz9qCn9i5ImCKYqYHEmulQz3l9R22ZlvRImoiUmVR1nDOMeZLqvdKWKRIf9wZKMWSgOPrQ3aG0rIhIBdK7tQjROwbzlvobuQu+6GLVYkRsPi4ZKyVREwQ6QWRNrC2OZwaKiVWKPNz7CMk+PVSYyiCjvLKlz89KW+nIMfRjUhQUf/tWMV+pESo1FiDkqYO1ooaooJqP4tO4xs6uRjmtbEDeGPdACWsRkSlkswXUZqutfRpjPRqfQh3YvYE7ODg4LCncAu7g4OCwpDhjCsVII6dFJnAv6A6yqkgCF7oG10dEJOMmNzzHTPS8jVUUsPXhfmLHOkN0hM2YLKSu0y7cuUtXrgixnWhiyLSjrlQ90P5Sh4XFmVugSgbwP6dDJHnI8doiIsXiuTzvSlW/M4Ka5lF+CJMZFoUju9GNDyDFWUVR3IDRCuNictKop/RDlmnnV9ZR7BqurAH1kWJS9IcH83a7rslTGaVJN1TjpK5DLiIijX2dR2tPqas+6kIKFXomY1bRQTuGVHKBUoJ77JciYCKPMro6ns0a6DCMLTVxFgkrImn+cI3xvDYaagtSIGRwogo4LBHx8VyTQmFCnkUC1BTysDVQDLWOzqMpaDVb1XNGvkaXxFIqLIwcpxefvTZvT24dzNvbQ84vPe9qR+dRHwWcVwzkiiGjOxkVI3EyMJfNtt5HC5LITOpKUpfI4+Dg4HBu4RZwBwcHhyXFmVIoRowE+W70kIVK4dZE0Fyo1NUNC8KiKzTuUzdBXR4P7mcEfQpqMXiZ3vagpy58kup5xqguMkUiDt1/EZFmVc/VQ8QA3evNS+qqD8fqenVCjXoJQRGNpkrZTCC9mY6LEp0tSKlOQCvUQ+0vc3b8PJJgwbEK4vm+NFuze2Eh1hA2ywoREjo2lGQVETEYd7+m7QSu7BD6Gah7LCn0M4ZIHJlU9fgedG2oVUHJUhGRKmw5RgWZrS3V4uj2Nank6rPqXvdidaN76GucKg1RoJeqxSS1wVT5nATTjUENAZKhqnmE1mkquDwKPM+TWv4MotCMTED5ROhHEDI5pTiePsY3wNiGLF4cI/JEdNxCUEc+mJkpBoTU3SHs1QiU/hIRabT0ek1EkK21dN3JJjpHbKJ9XV1F8s5A15MBaLKYpaNKD1oT0VDtJoq6I3KuB+0j8Yo01HFwb+AODg4OSwq3gDs4ODgsKdwC7uDg4LCkOFsO3KhYVaNOgSfw1tHx4VHGFkNymlVkSIED31jVckhjZK4lY23vbR/gTMrhhSjiaiGktQ0N6XEpi3Glqv+edpiV2dGDGCaJ744mytMOB9oeIVvQIsOy0SpyYp1V5dD7+CkeYzwZr+bl3HO6aBJ8pgguIkXem6GiA2QkNsA9djAPREQ6TeUfWXZvzMxd7GEMuwjfQ9jh2rr2g2Fv+yj1NQZ32QyL/PEqwrykgXJwCOX0kaU3QZZdmmq/M8s5pfdArrPaKI7BAKGVzASMQUQzszXONaTT94EDb+Thn+S0DZ9L3JOPfQqxxdBID333IeQUgjdPPL1vD/YwCPWdosSgxTiP8fz0ujrv7qf67IqIVPC8jxF2uhLpNRqXsR8Foaos1n231RWdUwn28ForyrGvrBftajz9G7OMR9gnsdzzsogBPgHuDdzBwcFhSeEWcAcHB4clxZlSKCIyF1JiWGBwgts9hitZrxbD99qgUFaoCQ2XzEOWXoBwv63O5Xl7hFTF1Oi1qa89PFAdYKmUKRQ9b+OqZmTdB01TR+mzDNXEWe5sjH5QV5m0zuZ6MRuVlbh8ZL5lcOdZrimbu4yLFbMSK/M0PANXuRjmpa6hhRucTouZmBnC6zLaD59XIGzWuqRjPulQWElDsyKEmFGQyKLqd7tSfBRqTyk9tbOn5wrwzuMHatcusm17qEZOPXcPlNDqmva7vVK0K+nCKUL2YmSRBseIwS1Y5l18Y6SRZwt7BQoFYlQIl81A3Vmv2BkPYlYMd5xMdHwSzItKhKrtWB9iiGpPIAgWj2DLVmfe7qwWaQwmN6YxaKwK7Iq1JkTI4wT1Bdp1tRnn7ADnNKUxGA6QfY5nNAxBKeq0k+wUBnVv4A4ODg5LCreAOzg4OCwpzp5COQK8A4Mq3DFcxqTgahe/nuKnJ4VIUMyNcA/iP3CFasjQtMgKnMI9ozBSs6PZXLY0YtR4nmB3vtVSt6gBPWMvAp3S1x3yFza15Nt4qn3NoICTgRYQEbm7rdSOQcYZq3VHoJrMUV+9YubfYpC78Rg32pJKRxFEuMKSDjfZnTHsYRENYCM9V1hXg1Crm9E3KWx0ki3TrBjlNKKQGsTB6hBvygxKiw3VNtU6tK+rmqEZgaZZWQUtUBIX8yKdn6nB9VDeKwPNEuf3umgKxYhIJTcI2QAP9GKCi04YTeYX51ilgjmJbNsAz34V9+RbnbchsrMzUHQVRH35MCZF8MQWNbkNKMXWitJYPiibFH3yGWHSYOYusrxx/hiZ4GlaEilj6UjQaZbr3xRrXvpw0bmHvoEbY6rGmP9qjPmWMea7xph/mn/+vDHmK8aYN40x/9YYEz3sXA5PDpxdzyecXS8WTkOhTETkl6y1HxeRT4jIrxhjfkZE/g8R+efW2pdEZF9EfuN966XD+wFn1/MJZ9cLhIdSKHamEnO0xR7m/1kR+SUR+R/zzz8vIv+riPyrdz2XaOVsVtD24KbESVz6Rt7RkkvG6ucGyTUsQWYhrDTCri/dIrpbhVJrcOXrpFxMMVmCpaYYgXEpQsVyuHcs3dX0tH/WokQWysrFCaqSJ0WXrF5ROmYEsa5hX4/rwsUNcqopTdOF2lWMVmkbT3hPoE1Ac4WIaPBKiVEZbENKoPCmgSgGgzlCIS0LCiXD1AkLCWQ6TqZUso8pU1s+ynAZpT5SUFExaJ3VQKOi0hTCaUPt3xh60rfulsr01Tfm7SrF3RrQnWZCyxG1YryF2tWISOS9M2KJSTmmIDKnx1JsS0Qkgz2GKDFYwXNdw7yI+HwXKD8mBEHQK2R0lz4Lk0mRhuB0q9LmuATvr4IIk+mEtBVKHaKvISi2wC8urxzJEfT/J3hmEiTynKZC3qk2MY0xvjHmmyLyQES+JCI/FJEDq6vOLRG5esJ3P2OM+Zox5mvjSXzcIQ4fEBZn18UXiHB471iUXSdTZ9cnHadawK21qbX2EyJyTUR+WkQ+ctoLWGtfsdZ+0lr7yWpJOtThg8Xi7PrB7YU7vBOLsmslcnZ90vFIFrLWHhhjviwiPysiHWNMkP+qXxOR26c7yex/pkCbHF9SzYe/E5RcObreLJNEgsOccDxdG5a/svDZp6Axkgza4NNiOEwAl5zlz+gy+nDDEkRKRHA/GYjgoVp2gPJTfkkTPULkRCWC7kvyzggFEZEkb5ed4se3q3IoTMQSROVE/vGRB6ZEoSTITiI94sOtpd5KDNqrqIeN+4YvmmI8+n1NumiuqWZ76eviGzVOpar0SK+QCIJkNCSndLtKF2R43KhHs7qikSoiRf2Nfhd0GL5TrVbe2TbF97HHtasRfQazgv6JHsM5HAQnLycjJOUNR5r0FOM7Aegilp+LQX+ysHtsEJVT43MIe5doiAkiO4IY44XwtRhzcIIolsN9taWPUm0WdFGMko5lGmmEZKND1CHg+sdonXINhONwmiiUTWNMJ2/XRORvicjrIvJlEfl7+WGfFpE/fejVHJ4YOLueTzi7Xiyc5g38ioh83hjjy2zB/4K19t8bY14TkT8yxvzvIvJXIvJ772M/HRYPZ9fzCWfXCwRjT7PVuaiLGbMtIgMR2XnYsecQG/Lk3Pez1trNRZ0st+vb8mTd41nhSbpnZ9fF4Um752Nte6YLuIiIMeZr1tpPnulFnwBchPu+CPdYxkW454twj2Usyz07LRQHBweHJYVbwB0cHByWFB/EAv7KB3DNJwEX4b4vwj2WcRHu+SLcYxlLcc9nzoE7ODg4OCwGjkJxcHBwWFK4BdzBwcFhSXGmC7gx5leMMd/PNYk/e5bXPisYY542xnzZGPNarsf8m/nna8aYLxljfpD/f/Vh51oWXAS7ilw82zq7Pvl2PTMOPM8Me0Nmqb23ROSrIvLr1trXzqQDZwRjzBURuWKt/YYxpiUiXxeRXxWRfygie9baz+UPw6q19rc/uJ4uBhfFriIXy7bOrsth17N8A/9pEXnTWvuWtXYqIn8kIp86w+ufCay1d62138jbPZnpUFyV2b1+Pj/s8zKbIOcBF8KuIhfOts6uS2DXs1zAr4rITfz7RE3i8wJjzHMi8pMi8hURuWStvZv/6Z6IXPqg+rVgXDi7ilwI2zq7LoFd3Sbm+wRjTFNE/lhEfsta2+Xf8qopLn5zSeFsez6xjHY9ywX8tog8jX+fXkN8yWCMCWU2Ef7QWvsn+cf3c67tiHN78EH1b8G4MHYVuVC2dXZdArue5QL+VRF52cyqY0ci8vdF5ItneP0zgZlVKPg9EXndWvu7+NMXZabDLHK+9JgvhF1FLpxtnV2XwK5nLSf7d0TkX8ishOjvW2v/2Zld/IxgjPkFEfnPIvId0bouvyMzTu0LIvKMzCQ6f81au/eBdHLBuAh2Fbl4tnV2ffLt6lLpHRwcHJYUbhPTwcHBYUnhFnAHBweHJYVbwB0cHByWFG4Bd3BwcFhSuAXcwcHBYUnhFnAHBweHJYVbwB0cHByWFP8/i6icfZ2ekbYAAAAASUVORK5CYII=\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 6 Axes>"
       ]
@@ -483,14 +499,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 12,
    "metadata": {
     "id": "ebK_OgcIvXX8"
    },
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 6 Axes>"
       ]
@@ -520,14 +536,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": 13,
    "metadata": {
     "id": "V8J5jJgjvXYD"
    },
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 6 Axes>"
       ]
@@ -592,7 +608,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 14,
    "metadata": {
     "id": "9ODch-OFCaW4"
    },
@@ -636,11 +652,20 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": 15,
    "metadata": {
     "id": "Lhan11blCaW7"
    },
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/opt/conda/lib/python3.7/site-packages/keras/optimizer_v2/gradient_descent.py:102: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead.\n",
+      "  super(SGD, self).__init__(name, **kwargs)\n"
+     ]
+    }
+   ],
    "source": [
     "model.compile(optimizer=tf.keras.optimizers.SGD(lr=0.0001),\n",
     "              loss='categorical_crossentropy',\n",
@@ -660,7 +685,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 16,
    "metadata": {
     "id": "xvwvpA64CaW_"
    },
@@ -669,207 +694,206 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Train on 4000 samples, validate on 1000 samples\n",
       "Epoch 1/100\n",
-      "4000/4000 [==============================] - 1s 357us/sample - loss: 27.3871 - accuracy: 0.1130 - val_loss: 2.4806 - val_accuracy: 0.1120\n",
+      "125/125 [==============================] - 4s 29ms/step - loss: 29.2992 - accuracy: 0.1295 - val_loss: 2.6232 - val_accuracy: 0.1160\n",
       "Epoch 2/100\n",
-      "4000/4000 [==============================] - 1s 233us/sample - loss: 2.4551 - accuracy: 0.1147 - val_loss: 2.3933 - val_accuracy: 0.1030\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 2.4425 - accuracy: 0.1135 - val_loss: 2.5073 - val_accuracy: 0.1120\n",
       "Epoch 3/100\n",
-      "4000/4000 [==============================] - 1s 223us/sample - loss: 2.3458 - accuracy: 0.1165 - val_loss: 2.3613 - val_accuracy: 0.1020\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 2.3225 - accuracy: 0.1270 - val_loss: 2.4879 - val_accuracy: 0.0980\n",
       "Epoch 4/100\n",
-      "4000/4000 [==============================] - 1s 235us/sample - loss: 2.3036 - accuracy: 0.1220 - val_loss: 2.3557 - val_accuracy: 0.1060\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 2.2730 - accuracy: 0.1310 - val_loss: 2.4898 - val_accuracy: 0.1000\n",
       "Epoch 5/100\n",
-      "4000/4000 [==============================] - 1s 202us/sample - loss: 2.2816 - accuracy: 0.1245 - val_loss: 2.3561 - val_accuracy: 0.1020\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 2.2527 - accuracy: 0.1357 - val_loss: 2.4923 - val_accuracy: 0.1020\n",
       "Epoch 6/100\n",
-      "4000/4000 [==============================] - 1s 209us/sample - loss: 2.2711 - accuracy: 0.1245 - val_loss: 2.3513 - val_accuracy: 0.1030\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 2.2337 - accuracy: 0.1395 - val_loss: 2.4835 - val_accuracy: 0.1120\n",
       "Epoch 7/100\n",
-      "4000/4000 [==============================] - 1s 220us/sample - loss: 2.2597 - accuracy: 0.1283 - val_loss: 2.3520 - val_accuracy: 0.1060\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 2.2233 - accuracy: 0.1445 - val_loss: 2.4874 - val_accuracy: 0.1010\n",
       "Epoch 8/100\n",
-      "4000/4000 [==============================] - 1s 256us/sample - loss: 2.2525 - accuracy: 0.1285 - val_loss: 2.3508 - val_accuracy: 0.1070\n",
+      "125/125 [==============================] - 3s 21ms/step - loss: 2.2184 - accuracy: 0.1452 - val_loss: 2.4832 - val_accuracy: 0.1020\n",
       "Epoch 9/100\n",
-      "4000/4000 [==============================] - 1s 151us/sample - loss: 2.2482 - accuracy: 0.1310 - val_loss: 2.3506 - val_accuracy: 0.1070\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 2.2100 - accuracy: 0.1500 - val_loss: 2.4880 - val_accuracy: 0.1040\n",
       "Epoch 10/100\n",
-      "4000/4000 [==============================] - 1s 149us/sample - loss: 2.2447 - accuracy: 0.1295 - val_loss: 2.3542 - val_accuracy: 0.1040\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 2.2022 - accuracy: 0.1530 - val_loss: 2.4859 - val_accuracy: 0.1000\n",
       "Epoch 11/100\n",
-      "4000/4000 [==============================] - 1s 206us/sample - loss: 2.2405 - accuracy: 0.1305 - val_loss: 2.3806 - val_accuracy: 0.0930\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 2.1953 - accuracy: 0.1532 - val_loss: 2.4888 - val_accuracy: 0.1010\n",
       "Epoch 12/100\n",
-      "4000/4000 [==============================] - 1s 219us/sample - loss: 2.2372 - accuracy: 0.1330 - val_loss: 2.3543 - val_accuracy: 0.1040\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 2.1894 - accuracy: 0.1538 - val_loss: 2.5027 - val_accuracy: 0.1020\n",
       "Epoch 13/100\n",
-      "4000/4000 [==============================] - 1s 197us/sample - loss: 2.2360 - accuracy: 0.1343 - val_loss: 2.3555 - val_accuracy: 0.1060\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 2.1854 - accuracy: 0.1587 - val_loss: 2.5122 - val_accuracy: 0.1040\n",
       "Epoch 14/100\n",
-      "4000/4000 [==============================] - 1s 204us/sample - loss: 2.2330 - accuracy: 0.1328 - val_loss: 2.3578 - val_accuracy: 0.1060\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 2.1750 - accuracy: 0.1632 - val_loss: 2.4948 - val_accuracy: 0.1010\n",
       "Epoch 15/100\n",
-      "4000/4000 [==============================] - 1s 215us/sample - loss: 2.2312 - accuracy: 0.1328 - val_loss: 2.3562 - val_accuracy: 0.1050\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 2.1735 - accuracy: 0.1632 - val_loss: 2.4909 - val_accuracy: 0.1230\n",
       "Epoch 16/100\n",
-      "4000/4000 [==============================] - 1s 225us/sample - loss: 2.2288 - accuracy: 0.1365 - val_loss: 2.3666 - val_accuracy: 0.1030\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 2.1700 - accuracy: 0.1667 - val_loss: 2.5037 - val_accuracy: 0.0990\n",
       "Epoch 17/100\n",
-      "4000/4000 [==============================] - 1s 227us/sample - loss: 2.2279 - accuracy: 0.1338 - val_loss: 2.3646 - val_accuracy: 0.1020\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 2.1627 - accuracy: 0.1663 - val_loss: 2.5082 - val_accuracy: 0.1140\n",
       "Epoch 18/100\n",
-      "4000/4000 [==============================] - 1s 207us/sample - loss: 2.2267 - accuracy: 0.1338 - val_loss: 2.3673 - val_accuracy: 0.1020\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 2.1568 - accuracy: 0.1710 - val_loss: 2.4986 - val_accuracy: 0.1170\n",
       "Epoch 19/100\n",
-      "4000/4000 [==============================] - 1s 197us/sample - loss: 2.2231 - accuracy: 0.1353 - val_loss: 2.3731 - val_accuracy: 0.1010\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 2.1562 - accuracy: 0.1685 - val_loss: 2.5115 - val_accuracy: 0.1250\n",
       "Epoch 20/100\n",
-      "4000/4000 [==============================] - 1s 203us/sample - loss: 2.2228 - accuracy: 0.1357 - val_loss: 2.3663 - val_accuracy: 0.1010\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 2.1547 - accuracy: 0.1725 - val_loss: 2.4900 - val_accuracy: 0.1080\n",
       "Epoch 21/100\n",
-      "4000/4000 [==============================] - 1s 199us/sample - loss: 2.2223 - accuracy: 0.1357 - val_loss: 2.3666 - val_accuracy: 0.1040\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 2.1421 - accuracy: 0.1743 - val_loss: 2.5250 - val_accuracy: 0.1330\n",
       "Epoch 22/100\n",
-      "4000/4000 [==============================] - 1s 230us/sample - loss: 2.2178 - accuracy: 0.1370 - val_loss: 2.3703 - val_accuracy: 0.1020\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 2.1404 - accuracy: 0.1800 - val_loss: 2.5074 - val_accuracy: 0.1170\n",
       "Epoch 23/100\n",
-      "4000/4000 [==============================] - 1s 188us/sample - loss: 2.2183 - accuracy: 0.1382 - val_loss: 2.3705 - val_accuracy: 0.1010\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 2.1289 - accuracy: 0.1780 - val_loss: 2.5015 - val_accuracy: 0.1080\n",
       "Epoch 24/100\n",
-      "4000/4000 [==============================] - 1s 196us/sample - loss: 2.2129 - accuracy: 0.1382 - val_loss: 2.3795 - val_accuracy: 0.1040\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 2.1291 - accuracy: 0.1807 - val_loss: 2.5162 - val_accuracy: 0.1160\n",
       "Epoch 25/100\n",
-      "4000/4000 [==============================] - 1s 234us/sample - loss: 2.2138 - accuracy: 0.1402 - val_loss: 2.3733 - val_accuracy: 0.0980\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 2.1236 - accuracy: 0.1865 - val_loss: 2.4957 - val_accuracy: 0.1230\n",
       "Epoch 26/100\n",
-      "4000/4000 [==============================] - 1s 229us/sample - loss: 2.2115 - accuracy: 0.1377 - val_loss: 2.3746 - val_accuracy: 0.1010\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 2.1152 - accuracy: 0.1957 - val_loss: 2.5331 - val_accuracy: 0.1630\n",
       "Epoch 27/100\n",
-      "4000/4000 [==============================] - 1s 213us/sample - loss: 2.2085 - accuracy: 0.1390 - val_loss: 2.3741 - val_accuracy: 0.0970\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 2.1078 - accuracy: 0.2015 - val_loss: 2.5095 - val_accuracy: 0.1440\n",
       "Epoch 28/100\n",
-      "4000/4000 [==============================] - 1s 251us/sample - loss: 2.2038 - accuracy: 0.1388 - val_loss: 2.3826 - val_accuracy: 0.0940\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 2.0992 - accuracy: 0.2072 - val_loss: 2.5087 - val_accuracy: 0.1550\n",
       "Epoch 29/100\n",
-      "4000/4000 [==============================] - 1s 224us/sample - loss: 2.2023 - accuracy: 0.1405 - val_loss: 2.4650 - val_accuracy: 0.0860\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 2.0940 - accuracy: 0.2068 - val_loss: 2.4933 - val_accuracy: 0.1600\n",
       "Epoch 30/100\n",
-      "4000/4000 [==============================] - 1s 208us/sample - loss: 2.2029 - accuracy: 0.1412 - val_loss: 2.3906 - val_accuracy: 0.0990\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 2.0920 - accuracy: 0.2140 - val_loss: 2.4880 - val_accuracy: 0.1600\n",
       "Epoch 31/100\n",
-      "4000/4000 [==============================] - 1s 267us/sample - loss: 2.2023 - accuracy: 0.1390 - val_loss: 2.4188 - val_accuracy: 0.0920\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 2.0796 - accuracy: 0.2190 - val_loss: 2.4703 - val_accuracy: 0.1580\n",
       "Epoch 32/100\n",
-      "4000/4000 [==============================] - 1s 232us/sample - loss: 2.1946 - accuracy: 0.1430 - val_loss: 2.3932 - val_accuracy: 0.0950\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 2.0632 - accuracy: 0.2257 - val_loss: 2.5121 - val_accuracy: 0.1640\n",
       "Epoch 33/100\n",
-      "4000/4000 [==============================] - 1s 210us/sample - loss: 2.2026 - accuracy: 0.1423 - val_loss: 2.3829 - val_accuracy: 0.1020\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 2.0498 - accuracy: 0.2303 - val_loss: 2.4610 - val_accuracy: 0.1760\n",
       "Epoch 34/100\n",
-      "4000/4000 [==============================] - 1s 205us/sample - loss: 2.1927 - accuracy: 0.1462 - val_loss: 2.4008 - val_accuracy: 0.1040\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 2.0442 - accuracy: 0.2342 - val_loss: 2.4584 - val_accuracy: 0.1660\n",
       "Epoch 35/100\n",
-      "4000/4000 [==============================] - 1s 196us/sample - loss: 2.1898 - accuracy: 0.1455 - val_loss: 2.3927 - val_accuracy: 0.1050\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 2.0400 - accuracy: 0.2365 - val_loss: 2.4692 - val_accuracy: 0.1550\n",
       "Epoch 36/100\n",
-      "4000/4000 [==============================] - 1s 221us/sample - loss: 2.1884 - accuracy: 0.1437 - val_loss: 2.4186 - val_accuracy: 0.0950\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 2.0286 - accuracy: 0.2430 - val_loss: 2.4842 - val_accuracy: 0.1770\n",
       "Epoch 37/100\n",
-      "4000/4000 [==============================] - 1s 217us/sample - loss: 2.1771 - accuracy: 0.1417 - val_loss: 2.4002 - val_accuracy: 0.0950\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 2.0188 - accuracy: 0.2407 - val_loss: 2.4958 - val_accuracy: 0.1810\n",
       "Epoch 38/100\n",
-      "4000/4000 [==============================] - 1s 223us/sample - loss: 2.1853 - accuracy: 0.1460 - val_loss: 2.3971 - val_accuracy: 0.0980\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 2.0040 - accuracy: 0.2455 - val_loss: 2.5405 - val_accuracy: 0.1880\n",
       "Epoch 39/100\n",
-      "4000/4000 [==============================] - 1s 220us/sample - loss: 2.1783 - accuracy: 0.1472 - val_loss: 2.4159 - val_accuracy: 0.0880\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 1.9926 - accuracy: 0.2503 - val_loss: 2.4944 - val_accuracy: 0.1750\n",
       "Epoch 40/100\n",
-      "4000/4000 [==============================] - 1s 207us/sample - loss: 2.1756 - accuracy: 0.1460 - val_loss: 2.4068 - val_accuracy: 0.0900\n",
+      "125/125 [==============================] - 3s 21ms/step - loss: 1.9908 - accuracy: 0.2473 - val_loss: 2.5124 - val_accuracy: 0.1890\n",
       "Epoch 41/100\n",
-      "4000/4000 [==============================] - 1s 170us/sample - loss: 2.1698 - accuracy: 0.1497 - val_loss: 2.4280 - val_accuracy: 0.0910\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 1.9831 - accuracy: 0.2540 - val_loss: 2.4740 - val_accuracy: 0.1790\n",
       "Epoch 42/100\n",
-      "4000/4000 [==============================] - 1s 192us/sample - loss: 2.1702 - accuracy: 0.1515 - val_loss: 2.4073 - val_accuracy: 0.0910\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 1.9727 - accuracy: 0.2525 - val_loss: 2.4678 - val_accuracy: 0.1780\n",
       "Epoch 43/100\n",
-      "4000/4000 [==============================] - 1s 199us/sample - loss: 2.1591 - accuracy: 0.1460 - val_loss: 2.3910 - val_accuracy: 0.1010\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 1.9611 - accuracy: 0.2542 - val_loss: 2.5043 - val_accuracy: 0.1940\n",
       "Epoch 44/100\n",
-      "4000/4000 [==============================] - 1s 226us/sample - loss: 2.1559 - accuracy: 0.1620 - val_loss: 2.4127 - val_accuracy: 0.1210\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 1.9567 - accuracy: 0.2603 - val_loss: 2.5053 - val_accuracy: 0.1970\n",
       "Epoch 45/100\n",
-      "4000/4000 [==============================] - 1s 224us/sample - loss: 2.1526 - accuracy: 0.1745 - val_loss: 2.4160 - val_accuracy: 0.1270\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 1.9530 - accuracy: 0.2615 - val_loss: 2.5082 - val_accuracy: 0.1740\n",
       "Epoch 46/100\n",
-      "4000/4000 [==============================] - 1s 192us/sample - loss: 2.1494 - accuracy: 0.1755 - val_loss: 2.3898 - val_accuracy: 0.1130\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 1.9427 - accuracy: 0.2595 - val_loss: 2.4811 - val_accuracy: 0.1840\n",
       "Epoch 47/100\n",
-      "4000/4000 [==============================] - 1s 199us/sample - loss: 2.1359 - accuracy: 0.1793 - val_loss: 2.4092 - val_accuracy: 0.1530\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.9305 - accuracy: 0.2610 - val_loss: 2.4980 - val_accuracy: 0.1960loss: 1 - ETA: 0s - loss: 1.9230 - accuracy\n",
       "Epoch 48/100\n",
-      "4000/4000 [==============================] - 1s 204us/sample - loss: 2.1405 - accuracy: 0.1750 - val_loss: 2.3939 - val_accuracy: 0.1380\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 1.9268 - accuracy: 0.2668 - val_loss: 2.4961 - val_accuracy: 0.1940\n",
       "Epoch 49/100\n",
-      "4000/4000 [==============================] - 1s 216us/sample - loss: 2.1220 - accuracy: 0.1795 - val_loss: 2.3886 - val_accuracy: 0.1270\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 1.9163 - accuracy: 0.2702 - val_loss: 2.5397 - val_accuracy: 0.1970\n",
       "Epoch 50/100\n",
-      "4000/4000 [==============================] - 1s 220us/sample - loss: 2.1038 - accuracy: 0.1850 - val_loss: 2.4022 - val_accuracy: 0.1480\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.9175 - accuracy: 0.2632 - val_loss: 2.4814 - val_accuracy: 0.1820\n",
       "Epoch 51/100\n",
-      "4000/4000 [==============================] - 1s 211us/sample - loss: 2.1201 - accuracy: 0.1822 - val_loss: 2.3981 - val_accuracy: 0.1230\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 1.9089 - accuracy: 0.2680 - val_loss: 2.4901 - val_accuracy: 0.1950\n",
       "Epoch 52/100\n",
-      "4000/4000 [==============================] - 1s 203us/sample - loss: 2.1005 - accuracy: 0.1898 - val_loss: 2.3754 - val_accuracy: 0.1340\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 1.8989 - accuracy: 0.2697 - val_loss: 2.4977 - val_accuracy: 0.2010\n",
       "Epoch 53/100\n",
-      "4000/4000 [==============================] - 1s 262us/sample - loss: 2.1173 - accuracy: 0.1815 - val_loss: 2.4274 - val_accuracy: 0.1130\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.8867 - accuracy: 0.2772 - val_loss: 2.4967 - val_accuracy: 0.1960\n",
       "Epoch 54/100\n",
-      "4000/4000 [==============================] - 1s 199us/sample - loss: 2.1299 - accuracy: 0.1805 - val_loss: 2.3857 - val_accuracy: 0.1510\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 1.9229 - accuracy: 0.2695 - val_loss: 2.5073 - val_accuracy: 0.1930\n",
       "Epoch 55/100\n",
-      "4000/4000 [==============================] - 1s 216us/sample - loss: 2.0866 - accuracy: 0.1982 - val_loss: 2.3986 - val_accuracy: 0.1710\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 1.8804 - accuracy: 0.2777 - val_loss: 2.4827 - val_accuracy: 0.1990\n",
       "Epoch 56/100\n",
-      "4000/4000 [==============================] - 1s 227us/sample - loss: 2.0817 - accuracy: 0.2005 - val_loss: 2.3716 - val_accuracy: 0.1780\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.8777 - accuracy: 0.2803 - val_loss: 2.4874 - val_accuracy: 0.1970\n",
       "Epoch 57/100\n",
-      "4000/4000 [==============================] - 1s 199us/sample - loss: 2.0846 - accuracy: 0.2042 - val_loss: 2.3922 - val_accuracy: 0.1610\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 1.8784 - accuracy: 0.2780 - val_loss: 2.5325 - val_accuracy: 0.2040\n",
       "Epoch 58/100\n",
-      "4000/4000 [==============================] - 1s 199us/sample - loss: 2.0813 - accuracy: 0.2065 - val_loss: 2.3834 - val_accuracy: 0.1590\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 1.8740 - accuracy: 0.2792 - val_loss: 2.5915 - val_accuracy: 0.2060\n",
       "Epoch 59/100\n",
-      "4000/4000 [==============================] - 1s 192us/sample - loss: 2.0696 - accuracy: 0.2055 - val_loss: 2.3735 - val_accuracy: 0.1700\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.8610 - accuracy: 0.2845 - val_loss: 2.5330 - val_accuracy: 0.2100\n",
       "Epoch 60/100\n",
-      "4000/4000 [==============================] - 1s 206us/sample - loss: 2.0706 - accuracy: 0.2085 - val_loss: 2.3950 - val_accuracy: 0.1730\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 1.8525 - accuracy: 0.2875 - val_loss: 2.5400 - val_accuracy: 0.2020\n",
       "Epoch 61/100\n",
-      "4000/4000 [==============================] - 1s 215us/sample - loss: 2.0615 - accuracy: 0.2087 - val_loss: 2.4830 - val_accuracy: 0.1850\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 1.8514 - accuracy: 0.2907 - val_loss: 2.4904 - val_accuracy: 0.1900\n",
       "Epoch 62/100\n",
-      "4000/4000 [==============================] - 1s 220us/sample - loss: 2.0754 - accuracy: 0.2070 - val_loss: 2.4022 - val_accuracy: 0.1690\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.8390 - accuracy: 0.2907 - val_loss: 2.5369 - val_accuracy: 0.2180\n",
       "Epoch 63/100\n",
-      "4000/4000 [==============================] - 1s 205us/sample - loss: 2.0596 - accuracy: 0.2107 - val_loss: 2.4050 - val_accuracy: 0.1520\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 1.8372 - accuracy: 0.2943 - val_loss: 2.5329 - val_accuracy: 0.2040\n",
       "Epoch 64/100\n",
-      "4000/4000 [==============================] - 1s 209us/sample - loss: 2.0550 - accuracy: 0.2150 - val_loss: 2.3864 - val_accuracy: 0.1740\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.8336 - accuracy: 0.2977 - val_loss: 2.5637 - val_accuracy: 0.2080\n",
       "Epoch 65/100\n",
-      "4000/4000 [==============================] - 1s 222us/sample - loss: 2.0557 - accuracy: 0.2118 - val_loss: 2.3918 - val_accuracy: 0.1690\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.8239 - accuracy: 0.2965 - val_loss: 2.5447 - val_accuracy: 0.2080\n",
       "Epoch 66/100\n",
-      "4000/4000 [==============================] - 1s 201us/sample - loss: 2.0439 - accuracy: 0.2130 - val_loss: 2.3940 - val_accuracy: 0.1720\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 1.8190 - accuracy: 0.3013 - val_loss: 2.5832 - val_accuracy: 0.2160\n",
       "Epoch 67/100\n",
-      "4000/4000 [==============================] - 1s 195us/sample - loss: 2.0448 - accuracy: 0.2183 - val_loss: 2.4027 - val_accuracy: 0.1700\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.8129 - accuracy: 0.3022 - val_loss: 2.5346 - val_accuracy: 0.2100\n",
       "Epoch 68/100\n",
-      "4000/4000 [==============================] - 1s 219us/sample - loss: 2.0359 - accuracy: 0.2198 - val_loss: 2.3885 - val_accuracy: 0.1780\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 1.8064 - accuracy: 0.3013 - val_loss: 2.5378 - val_accuracy: 0.2100\n",
       "Epoch 69/100\n",
-      "4000/4000 [==============================] - 1s 169us/sample - loss: 2.0426 - accuracy: 0.2150 - val_loss: 2.3964 - val_accuracy: 0.1800\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 1.8012 - accuracy: 0.3052 - val_loss: 2.5590 - val_accuracy: 0.2210\n",
       "Epoch 70/100\n",
-      "4000/4000 [==============================] - 1s 223us/sample - loss: 2.0353 - accuracy: 0.2212 - val_loss: 2.3940 - val_accuracy: 0.1790\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 1.7980 - accuracy: 0.3058 - val_loss: 2.5898 - val_accuracy: 0.2030\n",
       "Epoch 71/100\n",
-      "4000/4000 [==============================] - 1s 181us/sample - loss: 2.0280 - accuracy: 0.2210 - val_loss: 2.3928 - val_accuracy: 0.1840\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 1.7928 - accuracy: 0.3052 - val_loss: 2.5300 - val_accuracy: 0.2090\n",
       "Epoch 72/100\n",
-      "4000/4000 [==============================] - 1s 195us/sample - loss: 2.0289 - accuracy: 0.2218 - val_loss: 2.3878 - val_accuracy: 0.1840\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 1.7860 - accuracy: 0.3100 - val_loss: 2.5580 - val_accuracy: 0.2210\n",
       "Epoch 73/100\n",
-      "4000/4000 [==============================] - 1s 187us/sample - loss: 2.0237 - accuracy: 0.2240 - val_loss: 2.4028 - val_accuracy: 0.1840\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 1.7922 - accuracy: 0.3133 - val_loss: 2.5843 - val_accuracy: 0.2160\n",
       "Epoch 74/100\n",
-      "4000/4000 [==============================] - 1s 205us/sample - loss: 2.0256 - accuracy: 0.2210 - val_loss: 2.4272 - val_accuracy: 0.1880\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 1.7802 - accuracy: 0.3167 - val_loss: 2.6474 - val_accuracy: 0.2270\n",
       "Epoch 75/100\n",
-      "4000/4000 [==============================] - 1s 173us/sample - loss: 2.0258 - accuracy: 0.2243 - val_loss: 2.4007 - val_accuracy: 0.1730\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.7768 - accuracy: 0.3137 - val_loss: 2.6039 - val_accuracy: 0.2200\n",
       "Epoch 76/100\n",
-      "4000/4000 [==============================] - 1s 199us/sample - loss: 2.0127 - accuracy: 0.2300 - val_loss: 2.3932 - val_accuracy: 0.1820\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 1.7647 - accuracy: 0.3153 - val_loss: 2.6303 - val_accuracy: 0.2240\n",
       "Epoch 77/100\n",
-      "4000/4000 [==============================] - 1s 211us/sample - loss: 2.0089 - accuracy: 0.2278 - val_loss: 2.3811 - val_accuracy: 0.1750\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 1.7644 - accuracy: 0.3200 - val_loss: 2.6154 - val_accuracy: 0.2220\n",
       "Epoch 78/100\n",
-      "4000/4000 [==============================] - 1s 202us/sample - loss: 2.0142 - accuracy: 0.2307 - val_loss: 2.3694 - val_accuracy: 0.1660\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.7651 - accuracy: 0.3158 - val_loss: 2.6377 - val_accuracy: 0.2300\n",
       "Epoch 79/100\n",
-      "4000/4000 [==============================] - 1s 204us/sample - loss: 1.9999 - accuracy: 0.2350 - val_loss: 2.3983 - val_accuracy: 0.1750\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.7506 - accuracy: 0.3268 - val_loss: 2.5928 - val_accuracy: 0.2210\n",
       "Epoch 80/100\n",
-      "4000/4000 [==============================] - 1s 219us/sample - loss: 2.0165 - accuracy: 0.2265 - val_loss: 2.4120 - val_accuracy: 0.1770\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 1.7528 - accuracy: 0.3252 - val_loss: 2.5811 - val_accuracy: 0.2260\n",
       "Epoch 81/100\n",
-      "4000/4000 [==============================] - 1s 208us/sample - loss: 2.0099 - accuracy: 0.2338 - val_loss: 2.4254 - val_accuracy: 0.1880\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.7435 - accuracy: 0.3217 - val_loss: 2.6509 - val_accuracy: 0.2090\n",
       "Epoch 82/100\n",
-      "4000/4000 [==============================] - 1s 217us/sample - loss: 1.9937 - accuracy: 0.2340 - val_loss: 2.4193 - val_accuracy: 0.1770\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 1.7371 - accuracy: 0.3277 - val_loss: 2.6129 - val_accuracy: 0.2170\n",
       "Epoch 83/100\n",
-      "4000/4000 [==============================] - 1s 218us/sample - loss: 1.9915 - accuracy: 0.2385 - val_loss: 2.4072 - val_accuracy: 0.1740\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 1.7316 - accuracy: 0.3260 - val_loss: 2.6233 - val_accuracy: 0.2110\n",
       "Epoch 84/100\n",
-      "4000/4000 [==============================] - 1s 230us/sample - loss: 1.9964 - accuracy: 0.2348 - val_loss: 2.4355 - val_accuracy: 0.1770\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 1.7296 - accuracy: 0.3265 - val_loss: 2.6112 - val_accuracy: 0.2200\n",
       "Epoch 85/100\n",
-      "4000/4000 [==============================] - 1s 242us/sample - loss: 1.9956 - accuracy: 0.2340 - val_loss: 2.4133 - val_accuracy: 0.1840\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 1.7231 - accuracy: 0.3300 - val_loss: 2.6130 - val_accuracy: 0.2280\n",
       "Epoch 86/100\n",
-      "4000/4000 [==============================] - 1s 224us/sample - loss: 1.9844 - accuracy: 0.2373 - val_loss: 2.4133 - val_accuracy: 0.1850\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 1.7237 - accuracy: 0.3290 - val_loss: 2.6587 - val_accuracy: 0.2340\n",
       "Epoch 87/100\n",
-      "4000/4000 [==============================] - 1s 222us/sample - loss: 1.9883 - accuracy: 0.2390 - val_loss: 2.4464 - val_accuracy: 0.1880\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.7165 - accuracy: 0.3327 - val_loss: 2.6427 - val_accuracy: 0.2260\n",
       "Epoch 88/100\n",
-      "4000/4000 [==============================] - 1s 217us/sample - loss: 1.9847 - accuracy: 0.2405 - val_loss: 2.4194 - val_accuracy: 0.1820\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 1.7071 - accuracy: 0.3352 - val_loss: 2.6491 - val_accuracy: 0.2270\n",
       "Epoch 89/100\n",
-      "4000/4000 [==============================] - 1s 219us/sample - loss: 1.9814 - accuracy: 0.2445 - val_loss: 2.5359 - val_accuracy: 0.1970\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.7097 - accuracy: 0.3340 - val_loss: 2.6798 - val_accuracy: 0.2180\n",
       "Epoch 90/100\n",
-      "4000/4000 [==============================] - 1s 208us/sample - loss: 1.9751 - accuracy: 0.2455 - val_loss: 2.3937 - val_accuracy: 0.1920\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 1.7109 - accuracy: 0.3375 - val_loss: 2.6958 - val_accuracy: 0.2280\n",
       "Epoch 91/100\n",
-      "4000/4000 [==============================] - 1s 230us/sample - loss: 1.9663 - accuracy: 0.2450 - val_loss: 2.4166 - val_accuracy: 0.1940\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 1.7043 - accuracy: 0.3368 - val_loss: 2.6446 - val_accuracy: 0.2210\n",
       "Epoch 92/100\n",
-      "4000/4000 [==============================] - 1s 212us/sample - loss: 1.9731 - accuracy: 0.2407 - val_loss: 2.4553 - val_accuracy: 0.1970\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 1.6963 - accuracy: 0.3408 - val_loss: 2.6865 - val_accuracy: 0.2260\n",
       "Epoch 93/100\n",
-      "4000/4000 [==============================] - 1s 216us/sample - loss: 1.9593 - accuracy: 0.2530 - val_loss: 2.4430 - val_accuracy: 0.1970\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.6938 - accuracy: 0.3440 - val_loss: 2.8336 - val_accuracy: 0.2290\n",
       "Epoch 94/100\n",
-      "4000/4000 [==============================] - 1s 221us/sample - loss: 1.9580 - accuracy: 0.2465 - val_loss: 2.4463 - val_accuracy: 0.1960\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 1.6840 - accuracy: 0.3515 - val_loss: 2.7079 - val_accuracy: 0.2340\n",
       "Epoch 95/100\n",
-      "4000/4000 [==============================] - 1s 211us/sample - loss: 1.9636 - accuracy: 0.2515 - val_loss: 2.4132 - val_accuracy: 0.1940\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 1.6824 - accuracy: 0.3417 - val_loss: 2.6503 - val_accuracy: 0.2330\n",
       "Epoch 96/100\n",
-      "4000/4000 [==============================] - 1s 209us/sample - loss: 1.9650 - accuracy: 0.2495 - val_loss: 2.4247 - val_accuracy: 0.1980\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.6809 - accuracy: 0.3450 - val_loss: 2.6834 - val_accuracy: 0.2230\n",
       "Epoch 97/100\n",
-      "4000/4000 [==============================] - 1s 215us/sample - loss: 1.9525 - accuracy: 0.2542 - val_loss: 2.4405 - val_accuracy: 0.2030\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 1.6720 - accuracy: 0.3480 - val_loss: 2.7168 - val_accuracy: 0.2210\n",
       "Epoch 98/100\n",
-      "4000/4000 [==============================] - 1s 197us/sample - loss: 1.9533 - accuracy: 0.2542 - val_loss: 2.4446 - val_accuracy: 0.1980\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 1.6628 - accuracy: 0.3540 - val_loss: 2.7528 - val_accuracy: 0.2230\n",
       "Epoch 99/100\n",
-      "4000/4000 [==============================] - 1s 210us/sample - loss: 1.9489 - accuracy: 0.2558 - val_loss: 2.4594 - val_accuracy: 0.1990\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.6666 - accuracy: 0.3550 - val_loss: 2.7556 - val_accuracy: 0.2340\n",
       "Epoch 100/100\n",
-      "4000/4000 [==============================] - 1s 227us/sample - loss: 1.9511 - accuracy: 0.2495 - val_loss: 2.4144 - val_accuracy: 0.1940\n"
+      "125/125 [==============================] - 3s 23ms/step - loss: 1.6610 - accuracy: 0.3593 - val_loss: 2.7341 - val_accuracy: 0.2380\n"
      ]
     }
    ],
@@ -894,227 +918,223 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": 17,
    "metadata": {
     "id": "TDp8CC6VvXYe"
    },
    "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/opt/conda/lib/python3.7/site-packages/ipykernel_launcher.py:16: UserWarning: `Model.fit_generator` is deprecated and will be removed in a future version. Please use `Model.fit`, which supports generators.\n",
+      "  app.launch_new_instance()\n"
+     ]
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "WARNING:tensorflow:From <ipython-input-16-2b68e407f232>:16: Model.fit_generator (from tensorflow.python.keras.engine.training) is deprecated and will be removed in a future version.\n",
-      "Instructions for updating:\n",
-      "Please use Model.fit, which supports generators.\n",
-      "WARNING:tensorflow:sample_weight modes were coerced from\n",
-      "  ...\n",
-      "    to  \n",
-      "  ['...']\n",
-      "WARNING:tensorflow:sample_weight modes were coerced from\n",
-      "  ...\n",
-      "    to  \n",
-      "  ['...']\n",
-      "Train for 63 steps, validate for 16 steps\n",
       "Epoch 1/100\n",
-      "63/63 [==============================] - 4s 62ms/step - loss: 34.5204 - accuracy: 0.1650 - val_loss: 29.2886 - val_accuracy: 0.1800\n",
+      "63/63 [==============================] - 5s 77ms/step - loss: 35.2853 - accuracy: 0.1780 - val_loss: 27.6925 - val_accuracy: 0.1860\n",
       "Epoch 2/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 22.8572 - accuracy: 0.1953 - val_loss: 22.4747 - val_accuracy: 0.1920\n",
+      "63/63 [==============================] - 5s 78ms/step - loss: 22.8362 - accuracy: 0.2083 - val_loss: 21.2226 - val_accuracy: 0.2130\n",
       "Epoch 3/100\n",
-      "63/63 [==============================] - 4s 60ms/step - loss: 17.9400 - accuracy: 0.1900 - val_loss: 18.0272 - val_accuracy: 0.1950\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 18.8794 - accuracy: 0.2060 - val_loss: 17.9238 - val_accuracy: 0.2100\n",
       "Epoch 4/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 14.4337 - accuracy: 0.1925 - val_loss: 15.3413 - val_accuracy: 0.2040\n",
+      "63/63 [==============================] - 5s 78ms/step - loss: 15.3165 - accuracy: 0.2000 - val_loss: 15.5570 - val_accuracy: 0.2100\n",
       "Epoch 5/100\n",
-      "63/63 [==============================] - 4s 56ms/step - loss: 11.9522 - accuracy: 0.2100 - val_loss: 12.4391 - val_accuracy: 0.2110\n",
+      "63/63 [==============================] - 5s 76ms/step - loss: 13.6691 - accuracy: 0.1898 - val_loss: 13.4271 - val_accuracy: 0.2170\n",
       "Epoch 6/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 9.9445 - accuracy: 0.2007 - val_loss: 10.1192 - val_accuracy: 0.2150\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 11.4891 - accuracy: 0.1905 - val_loss: 11.7095 - val_accuracy: 0.1880\n",
       "Epoch 7/100\n",
-      "63/63 [==============================] - 4s 56ms/step - loss: 7.8987 - accuracy: 0.1937 - val_loss: 8.2408 - val_accuracy: 0.2030\n",
+      "63/63 [==============================] - 5s 77ms/step - loss: 10.0352 - accuracy: 0.2015 - val_loss: 10.2237 - val_accuracy: 0.1950\n",
       "Epoch 8/100\n",
-      "63/63 [==============================] - 4s 56ms/step - loss: 6.4617 - accuracy: 0.1807 - val_loss: 6.9276 - val_accuracy: 0.1890\n",
+      "63/63 [==============================] - 5s 84ms/step - loss: 8.8034 - accuracy: 0.2020 - val_loss: 8.7818 - val_accuracy: 0.1870\n",
       "Epoch 9/100\n",
-      "63/63 [==============================] - 4s 61ms/step - loss: 5.4737 - accuracy: 0.1793 - val_loss: 6.0185 - val_accuracy: 0.2010\n",
+      "63/63 [==============================] - 5s 81ms/step - loss: 7.3796 - accuracy: 0.1898 - val_loss: 7.4378 - val_accuracy: 0.1880\n",
       "Epoch 10/100\n",
-      "63/63 [==============================] - 4s 61ms/step - loss: 4.8609 - accuracy: 0.1720 - val_loss: 5.4180 - val_accuracy: 0.1920\n",
+      "63/63 [==============================] - 5s 77ms/step - loss: 5.9453 - accuracy: 0.1935 - val_loss: 6.3020 - val_accuracy: 0.1690\n",
       "Epoch 11/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 4.2886 - accuracy: 0.1653 - val_loss: 4.9717 - val_accuracy: 0.1890\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 5.1056 - accuracy: 0.1842 - val_loss: 5.5118 - val_accuracy: 0.1780\n",
       "Epoch 12/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 4.1180 - accuracy: 0.1632 - val_loss: 4.5704 - val_accuracy: 0.1890\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 4.5281 - accuracy: 0.1780 - val_loss: 4.9366 - val_accuracy: 0.1810\n",
       "Epoch 13/100\n",
-      "63/63 [==============================] - 3s 54ms/step - loss: 3.6870 - accuracy: 0.1720 - val_loss: 4.2848 - val_accuracy: 0.1860\n",
+      "63/63 [==============================] - 5s 72ms/step - loss: 4.1045 - accuracy: 0.1737 - val_loss: 4.5372 - val_accuracy: 0.1730\n",
       "Epoch 14/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 3.5776 - accuracy: 0.1670 - val_loss: 4.0614 - val_accuracy: 0.1860\n",
+      "63/63 [==============================] - 4s 70ms/step - loss: 3.8437 - accuracy: 0.1785 - val_loss: 4.2184 - val_accuracy: 0.1760\n",
       "Epoch 15/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 3.3611 - accuracy: 0.1653 - val_loss: 3.8779 - val_accuracy: 0.1870\n",
+      "63/63 [==============================] - 5s 77ms/step - loss: 3.6531 - accuracy: 0.1727 - val_loss: 3.9902 - val_accuracy: 0.1760\n",
       "Epoch 16/100\n",
-      "63/63 [==============================] - 4s 56ms/step - loss: 3.2211 - accuracy: 0.1715 - val_loss: 3.7268 - val_accuracy: 0.1850\n",
+      "63/63 [==============================] - 4s 70ms/step - loss: 3.4975 - accuracy: 0.1690 - val_loss: 3.7944 - val_accuracy: 0.1690\n",
       "Epoch 17/100\n",
-      "63/63 [==============================] - 4s 62ms/step - loss: 3.1558 - accuracy: 0.1765 - val_loss: 3.5963 - val_accuracy: 0.1820\n",
+      "63/63 [==============================] - 4s 69ms/step - loss: 3.2605 - accuracy: 0.1735 - val_loss: 3.6525 - val_accuracy: 0.1670\n",
       "Epoch 18/100\n",
-      "63/63 [==============================] - 3s 55ms/step - loss: 3.0256 - accuracy: 0.1688 - val_loss: 3.4829 - val_accuracy: 0.1830\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 3.1124 - accuracy: 0.1680 - val_loss: 3.5119 - val_accuracy: 0.1660\n",
       "Epoch 19/100\n",
-      "63/63 [==============================] - 3s 55ms/step - loss: 3.0111 - accuracy: 0.1670 - val_loss: 3.3812 - val_accuracy: 0.1830\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 3.1012 - accuracy: 0.1665 - val_loss: 3.4016 - val_accuracy: 0.1550\n",
       "Epoch 20/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 2.9170 - accuracy: 0.1630 - val_loss: 3.2882 - val_accuracy: 0.1770\n",
+      "63/63 [==============================] - 5s 79ms/step - loss: 2.9404 - accuracy: 0.1740 - val_loss: 3.3143 - val_accuracy: 0.1600\n",
       "Epoch 21/100\n",
-      "63/63 [==============================] - 3s 54ms/step - loss: 2.8375 - accuracy: 0.1708 - val_loss: 3.2022 - val_accuracy: 0.1750\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 2.8307 - accuracy: 0.1708 - val_loss: 3.2359 - val_accuracy: 0.1630\n",
       "Epoch 22/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 2.7990 - accuracy: 0.1680 - val_loss: 3.1391 - val_accuracy: 0.1800\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 2.8247 - accuracy: 0.1655 - val_loss: 3.1685 - val_accuracy: 0.1660\n",
       "Epoch 23/100\n",
-      "63/63 [==============================] - 3s 55ms/step - loss: 2.8003 - accuracy: 0.1630 - val_loss: 3.0757 - val_accuracy: 0.1790\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 2.8009 - accuracy: 0.1625 - val_loss: 3.1117 - val_accuracy: 0.1730\n",
       "Epoch 24/100\n",
-      "63/63 [==============================] - 3s 54ms/step - loss: 2.7453 - accuracy: 0.1692 - val_loss: 3.0180 - val_accuracy: 0.1770\n",
+      "63/63 [==============================] - 5s 78ms/step - loss: 2.7492 - accuracy: 0.1772 - val_loss: 3.0572 - val_accuracy: 0.1700\n",
       "Epoch 25/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 2.7021 - accuracy: 0.1690 - val_loss: 2.9662 - val_accuracy: 0.1750\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 2.7126 - accuracy: 0.1758 - val_loss: 2.9946 - val_accuracy: 0.1610\n",
       "Epoch 26/100\n",
-      "63/63 [==============================] - 4s 56ms/step - loss: 2.6588 - accuracy: 0.1717 - val_loss: 2.9270 - val_accuracy: 0.1830\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 2.6384 - accuracy: 0.1698 - val_loss: 2.9500 - val_accuracy: 0.1620\n",
       "Epoch 27/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 2.6505 - accuracy: 0.1590 - val_loss: 2.8740 - val_accuracy: 0.1800\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 2.5880 - accuracy: 0.1700 - val_loss: 2.9103 - val_accuracy: 0.1560\n",
       "Epoch 28/100\n",
-      "63/63 [==============================] - 3s 55ms/step - loss: 2.6537 - accuracy: 0.1723 - val_loss: 2.8335 - val_accuracy: 0.1780\n",
+      "63/63 [==============================] - 5s 71ms/step - loss: 2.6443 - accuracy: 0.1678 - val_loss: 2.8779 - val_accuracy: 0.1550\n",
       "Epoch 29/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 2.6114 - accuracy: 0.1663 - val_loss: 2.7946 - val_accuracy: 0.1790\n",
+      "63/63 [==============================] - 5s 76ms/step - loss: 2.6021 - accuracy: 0.1620 - val_loss: 2.8402 - val_accuracy: 0.1620\n",
       "Epoch 30/100\n",
-      "63/63 [==============================] - 4s 56ms/step - loss: 2.5604 - accuracy: 0.1673 - val_loss: 2.7685 - val_accuracy: 0.1810\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 2.5511 - accuracy: 0.1615 - val_loss: 2.8006 - val_accuracy: 0.1610\n",
       "Epoch 31/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 2.5202 - accuracy: 0.1657 - val_loss: 2.7428 - val_accuracy: 0.1780\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 2.5293 - accuracy: 0.1682 - val_loss: 2.7745 - val_accuracy: 0.1560\n",
       "Epoch 32/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 2.5303 - accuracy: 0.1622 - val_loss: 2.7212 - val_accuracy: 0.1790\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 2.5957 - accuracy: 0.1690 - val_loss: 2.7464 - val_accuracy: 0.1760\n",
       "Epoch 33/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 2.5205 - accuracy: 0.1647 - val_loss: 2.6952 - val_accuracy: 0.1790\n",
+      "63/63 [==============================] - 4s 70ms/step - loss: 2.5457 - accuracy: 0.1663 - val_loss: 2.7177 - val_accuracy: 0.1680\n",
       "Epoch 34/100\n",
-      "63/63 [==============================] - 4s 56ms/step - loss: 2.4943 - accuracy: 0.1625 - val_loss: 2.6676 - val_accuracy: 0.1810\n",
+      "63/63 [==============================] - 4s 68ms/step - loss: 2.4864 - accuracy: 0.1828 - val_loss: 2.6933 - val_accuracy: 0.1620\n",
       "Epoch 35/100\n",
-      "63/63 [==============================] - 3s 53ms/step - loss: 2.4516 - accuracy: 0.1730 - val_loss: 2.6556 - val_accuracy: 0.1830\n",
+      "63/63 [==============================] - 5s 76ms/step - loss: 2.4528 - accuracy: 0.1717 - val_loss: 2.6719 - val_accuracy: 0.1610\n",
       "Epoch 36/100\n",
-      "63/63 [==============================] - 3s 55ms/step - loss: 2.4650 - accuracy: 0.1665 - val_loss: 2.6339 - val_accuracy: 0.1790\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 2.4390 - accuracy: 0.1745 - val_loss: 2.6565 - val_accuracy: 0.1630\n",
       "Epoch 37/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 2.4293 - accuracy: 0.1700 - val_loss: 2.6185 - val_accuracy: 0.1770\n",
+      "63/63 [==============================] - 4s 70ms/step - loss: 2.4556 - accuracy: 0.1698 - val_loss: 2.6332 - val_accuracy: 0.1650\n",
       "Epoch 38/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 2.4358 - accuracy: 0.1645 - val_loss: 2.6082 - val_accuracy: 0.1740\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 2.4341 - accuracy: 0.1713 - val_loss: 2.6165 - val_accuracy: 0.1620\n",
       "Epoch 39/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 2.4303 - accuracy: 0.1700 - val_loss: 2.5962 - val_accuracy: 0.1790\n",
+      "63/63 [==============================] - 5s 77ms/step - loss: 2.4485 - accuracy: 0.1708 - val_loss: 2.6008 - val_accuracy: 0.1610\n",
       "Epoch 40/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 2.4316 - accuracy: 0.1750 - val_loss: 2.5824 - val_accuracy: 0.1770\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 2.4072 - accuracy: 0.1673 - val_loss: 2.5865 - val_accuracy: 0.1640\n",
       "Epoch 41/100\n",
-      "63/63 [==============================] - 3s 54ms/step - loss: 2.4093 - accuracy: 0.1708 - val_loss: 2.5695 - val_accuracy: 0.1800\n",
+      "63/63 [==============================] - 4s 71ms/step - loss: 2.4103 - accuracy: 0.1735 - val_loss: 2.5666 - val_accuracy: 0.1620\n",
       "Epoch 42/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 2.4075 - accuracy: 0.1640 - val_loss: 2.5555 - val_accuracy: 0.1790\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 2.4189 - accuracy: 0.1618 - val_loss: 2.5568 - val_accuracy: 0.1680\n",
       "Epoch 43/100\n",
-      "63/63 [==============================] - 4s 61ms/step - loss: 2.3989 - accuracy: 0.1643 - val_loss: 2.5467 - val_accuracy: 0.1800\n",
+      "63/63 [==============================] - 4s 69ms/step - loss: 2.4044 - accuracy: 0.1708 - val_loss: 2.5480 - val_accuracy: 0.1630\n",
       "Epoch 44/100\n",
-      "63/63 [==============================] - 4s 56ms/step - loss: 2.4006 - accuracy: 0.1737 - val_loss: 2.5337 - val_accuracy: 0.1780\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 2.4014 - accuracy: 0.1705 - val_loss: 2.5288 - val_accuracy: 0.1750\n",
       "Epoch 45/100\n",
-      "63/63 [==============================] - 4s 60ms/step - loss: 2.4023 - accuracy: 0.1678 - val_loss: 2.5247 - val_accuracy: 0.1780\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 2.3939 - accuracy: 0.1803 - val_loss: 2.5188 - val_accuracy: 0.1790\n",
       "Epoch 46/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 2.3450 - accuracy: 0.1717 - val_loss: 2.5184 - val_accuracy: 0.1770\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 2.3622 - accuracy: 0.1817 - val_loss: 2.5094 - val_accuracy: 0.1730\n",
       "Epoch 47/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 2.3739 - accuracy: 0.1690 - val_loss: 2.5140 - val_accuracy: 0.1770\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 2.3954 - accuracy: 0.1803 - val_loss: 2.4973 - val_accuracy: 0.1760\n",
       "Epoch 48/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 2.3401 - accuracy: 0.1723 - val_loss: 2.5070 - val_accuracy: 0.1740\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 2.3844 - accuracy: 0.1807 - val_loss: 2.4865 - val_accuracy: 0.1780\n",
       "Epoch 49/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 2.3361 - accuracy: 0.1743 - val_loss: 2.4968 - val_accuracy: 0.1750\n",
+      "63/63 [==============================] - 5s 78ms/step - loss: 2.3554 - accuracy: 0.1850 - val_loss: 2.4791 - val_accuracy: 0.1770\n",
       "Epoch 50/100\n",
-      "63/63 [==============================] - 3s 55ms/step - loss: 2.3305 - accuracy: 0.1698 - val_loss: 2.4896 - val_accuracy: 0.1770\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 2.3284 - accuracy: 0.1797 - val_loss: 2.4708 - val_accuracy: 0.1750\n",
       "Epoch 51/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 2.3595 - accuracy: 0.1698 - val_loss: 2.4804 - val_accuracy: 0.1770\n",
+      "63/63 [==============================] - 4s 71ms/step - loss: 2.3336 - accuracy: 0.1820 - val_loss: 2.4623 - val_accuracy: 0.1770\n",
       "Epoch 52/100\n",
-      "63/63 [==============================] - 3s 55ms/step - loss: 2.3619 - accuracy: 0.1675 - val_loss: 2.4754 - val_accuracy: 0.1760\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 2.3243 - accuracy: 0.1835 - val_loss: 2.4520 - val_accuracy: 0.1750\n",
       "Epoch 53/100\n",
-      "63/63 [==============================] - 3s 53ms/step - loss: 2.3524 - accuracy: 0.1702 - val_loss: 2.4698 - val_accuracy: 0.1730\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 2.3516 - accuracy: 0.1822 - val_loss: 2.4417 - val_accuracy: 0.1770\n",
       "Epoch 54/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 2.3205 - accuracy: 0.1700 - val_loss: 2.4641 - val_accuracy: 0.1740\n",
+      "63/63 [==============================] - 4s 70ms/step - loss: 2.2894 - accuracy: 0.1848 - val_loss: 2.4347 - val_accuracy: 0.1760\n",
       "Epoch 55/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 2.3456 - accuracy: 0.1760 - val_loss: 2.4552 - val_accuracy: 0.1740\n",
+      "63/63 [==============================] - 5s 76ms/step - loss: 2.2900 - accuracy: 0.1778 - val_loss: 2.4301 - val_accuracy: 0.1760\n",
       "Epoch 56/100\n",
-      "63/63 [==============================] - 4s 56ms/step - loss: 2.3341 - accuracy: 0.1720 - val_loss: 2.4544 - val_accuracy: 0.1860\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 2.3294 - accuracy: 0.1817 - val_loss: 2.4231 - val_accuracy: 0.1790\n",
       "Epoch 57/100\n",
-      "63/63 [==============================] - 3s 55ms/step - loss: 2.3098 - accuracy: 0.1723 - val_loss: 2.4495 - val_accuracy: 0.1750\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 2.3366 - accuracy: 0.1752 - val_loss: 2.4157 - val_accuracy: 0.1860\n",
       "Epoch 58/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 2.3151 - accuracy: 0.1758 - val_loss: 2.4446 - val_accuracy: 0.1740\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 2.3446 - accuracy: 0.1807 - val_loss: 2.4114 - val_accuracy: 0.1720\n",
       "Epoch 59/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 2.3255 - accuracy: 0.1778 - val_loss: 2.4401 - val_accuracy: 0.1760\n",
+      "63/63 [==============================] - 4s 67ms/step - loss: 2.3271 - accuracy: 0.1838 - val_loss: 2.3994 - val_accuracy: 0.1730\n",
       "Epoch 60/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 2.3037 - accuracy: 0.1737 - val_loss: 2.4370 - val_accuracy: 0.1760\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 2.3173 - accuracy: 0.1800 - val_loss: 2.3944 - val_accuracy: 0.1730\n",
       "Epoch 61/100\n",
-      "63/63 [==============================] - 4s 56ms/step - loss: 2.3017 - accuracy: 0.1735 - val_loss: 2.4321 - val_accuracy: 0.1760\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 2.3054 - accuracy: 0.1785 - val_loss: 2.3844 - val_accuracy: 0.1740\n",
       "Epoch 62/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 2.2822 - accuracy: 0.1692 - val_loss: 2.4277 - val_accuracy: 0.1750\n",
+      "63/63 [==============================] - 5s 76ms/step - loss: 2.2855 - accuracy: 0.1855 - val_loss: 2.3808 - val_accuracy: 0.1760\n",
       "Epoch 63/100\n",
-      "63/63 [==============================] - 4s 56ms/step - loss: 2.2764 - accuracy: 0.1745 - val_loss: 2.4234 - val_accuracy: 0.1740\n",
+      "63/63 [==============================] - 5s 76ms/step - loss: 2.3003 - accuracy: 0.1815 - val_loss: 2.3762 - val_accuracy: 0.1790\n",
       "Epoch 64/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 2.3121 - accuracy: 0.1752 - val_loss: 2.4161 - val_accuracy: 0.1730\n",
+      "63/63 [==============================] - 6s 90ms/step - loss: 2.2581 - accuracy: 0.1863 - val_loss: 2.3696 - val_accuracy: 0.1800\n",
       "Epoch 65/100\n",
-      "63/63 [==============================] - 4s 60ms/step - loss: 2.2873 - accuracy: 0.1737 - val_loss: 2.4112 - val_accuracy: 0.1760\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 2.2909 - accuracy: 0.1835 - val_loss: 2.3643 - val_accuracy: 0.1760\n",
       "Epoch 66/100\n",
-      "63/63 [==============================] - 4s 56ms/step - loss: 2.3008 - accuracy: 0.1725 - val_loss: 2.4070 - val_accuracy: 0.1730\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 2.2922 - accuracy: 0.1813 - val_loss: 2.3613 - val_accuracy: 0.1780\n",
       "Epoch 67/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 2.2895 - accuracy: 0.1750 - val_loss: 2.4031 - val_accuracy: 0.1770\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 2.2824 - accuracy: 0.1825 - val_loss: 2.3604 - val_accuracy: 0.1790\n",
       "Epoch 68/100\n",
-      "63/63 [==============================] - 4s 60ms/step - loss: 2.2882 - accuracy: 0.1725 - val_loss: 2.3977 - val_accuracy: 0.1760\n",
+      "63/63 [==============================] - 5s 77ms/step - loss: 2.2748 - accuracy: 0.1813 - val_loss: 2.3543 - val_accuracy: 0.1800\n",
       "Epoch 69/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 2.2867 - accuracy: 0.1727 - val_loss: 2.3933 - val_accuracy: 0.1750\n",
+      "63/63 [==============================] - 5s 72ms/step - loss: 2.2886 - accuracy: 0.1762 - val_loss: 2.3482 - val_accuracy: 0.1800\n",
       "Epoch 70/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 2.2636 - accuracy: 0.1805 - val_loss: 2.3903 - val_accuracy: 0.1740\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 2.3029 - accuracy: 0.1807 - val_loss: 2.3459 - val_accuracy: 0.1770\n",
       "Epoch 71/100\n",
-      "63/63 [==============================] - 3s 53ms/step - loss: 2.2871 - accuracy: 0.1787 - val_loss: 2.3875 - val_accuracy: 0.1740\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 2.2876 - accuracy: 0.1748 - val_loss: 2.3418 - val_accuracy: 0.1800\n",
       "Epoch 72/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 2.2614 - accuracy: 0.1810 - val_loss: 2.3840 - val_accuracy: 0.1780\n",
+      "63/63 [==============================] - 4s 68ms/step - loss: 2.2897 - accuracy: 0.1750 - val_loss: 2.3383 - val_accuracy: 0.1780\n",
       "Epoch 73/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 2.2592 - accuracy: 0.1797 - val_loss: 2.3802 - val_accuracy: 0.1760\n",
+      "63/63 [==============================] - 5s 78ms/step - loss: 2.2948 - accuracy: 0.1782 - val_loss: 2.3343 - val_accuracy: 0.1800\n",
       "Epoch 74/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 2.2569 - accuracy: 0.1825 - val_loss: 2.3746 - val_accuracy: 0.1790\n",
+      "63/63 [==============================] - 5s 77ms/step - loss: 2.2457 - accuracy: 0.1813 - val_loss: 2.3315 - val_accuracy: 0.1830\n",
       "Epoch 75/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 2.2550 - accuracy: 0.1768 - val_loss: 2.3719 - val_accuracy: 0.1800\n",
+      "63/63 [==============================] - 4s 67ms/step - loss: 2.2500 - accuracy: 0.1842 - val_loss: 2.3255 - val_accuracy: 0.1830\n",
       "Epoch 76/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 2.2430 - accuracy: 0.1733 - val_loss: 2.3720 - val_accuracy: 0.1790\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 2.2510 - accuracy: 0.1850 - val_loss: 2.3211 - val_accuracy: 0.1760\n",
       "Epoch 77/100\n",
-      "63/63 [==============================] - 4s 60ms/step - loss: 2.2587 - accuracy: 0.1828 - val_loss: 2.3696 - val_accuracy: 0.1790\n",
+      "63/63 [==============================] - 5s 80ms/step - loss: 2.2719 - accuracy: 0.1863 - val_loss: 2.3194 - val_accuracy: 0.1800\n",
       "Epoch 78/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 2.2609 - accuracy: 0.1813 - val_loss: 2.3685 - val_accuracy: 0.1790\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 2.2820 - accuracy: 0.1795 - val_loss: 2.3177 - val_accuracy: 0.1790\n",
       "Epoch 79/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 2.2520 - accuracy: 0.1795 - val_loss: 2.3618 - val_accuracy: 0.1880\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 2.2678 - accuracy: 0.1852 - val_loss: 2.3174 - val_accuracy: 0.1810\n",
       "Epoch 80/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 2.2449 - accuracy: 0.1745 - val_loss: 2.3569 - val_accuracy: 0.1860\n",
+      "63/63 [==============================] - 5s 76ms/step - loss: 2.2550 - accuracy: 0.1840 - val_loss: 2.3161 - val_accuracy: 0.1780\n",
       "Epoch 81/100\n",
-      "63/63 [==============================] - 3s 53ms/step - loss: 2.2280 - accuracy: 0.1803 - val_loss: 2.3478 - val_accuracy: 0.1870\n",
+      "63/63 [==============================] - 5s 78ms/step - loss: 2.2688 - accuracy: 0.1750 - val_loss: 2.3151 - val_accuracy: 0.1770\n",
       "Epoch 82/100\n",
-      "63/63 [==============================] - 4s 56ms/step - loss: 2.2382 - accuracy: 0.1803 - val_loss: 2.3459 - val_accuracy: 0.1900\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 2.2669 - accuracy: 0.1810 - val_loss: 2.3112 - val_accuracy: 0.1780\n",
       "Epoch 83/100\n",
-      "63/63 [==============================] - 3s 55ms/step - loss: 2.2497 - accuracy: 0.1815 - val_loss: 2.3415 - val_accuracy: 0.1910\n",
+      "63/63 [==============================] - 5s 76ms/step - loss: 2.2602 - accuracy: 0.1887 - val_loss: 2.3091 - val_accuracy: 0.1780\n",
       "Epoch 84/100\n",
-      "63/63 [==============================] - 3s 54ms/step - loss: 2.2261 - accuracy: 0.1797 - val_loss: 2.3384 - val_accuracy: 0.1920\n",
+      "63/63 [==============================] - 5s 72ms/step - loss: 2.2468 - accuracy: 0.1840 - val_loss: 2.3056 - val_accuracy: 0.1780\n",
       "Epoch 85/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 2.2081 - accuracy: 0.1873 - val_loss: 2.3401 - val_accuracy: 0.1910\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 2.2505 - accuracy: 0.1785 - val_loss: 2.3042 - val_accuracy: 0.1730\n",
       "Epoch 86/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 2.2188 - accuracy: 0.1860 - val_loss: 2.3311 - val_accuracy: 0.1940\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 2.2740 - accuracy: 0.1795 - val_loss: 2.3016 - val_accuracy: 0.1780\n",
       "Epoch 87/100\n",
-      "63/63 [==============================] - 3s 53ms/step - loss: 2.2139 - accuracy: 0.1890 - val_loss: 2.3285 - val_accuracy: 0.1930\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 2.2601 - accuracy: 0.1730 - val_loss: 2.2994 - val_accuracy: 0.1770\n",
       "Epoch 88/100\n",
-      "63/63 [==============================] - 4s 60ms/step - loss: 2.2242 - accuracy: 0.1940 - val_loss: 2.3267 - val_accuracy: 0.1900\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 2.2506 - accuracy: 0.1813 - val_loss: 2.2980 - val_accuracy: 0.1740\n",
       "Epoch 89/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 2.2357 - accuracy: 0.1852 - val_loss: 2.3214 - val_accuracy: 0.1920\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 2.2436 - accuracy: 0.1832 - val_loss: 2.2950 - val_accuracy: 0.1750\n",
       "Epoch 90/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 2.2368 - accuracy: 0.1867 - val_loss: 2.3213 - val_accuracy: 0.1900\n",
+      "63/63 [==============================] - 5s 79ms/step - loss: 2.2453 - accuracy: 0.1857 - val_loss: 2.2928 - val_accuracy: 0.1790\n",
       "Epoch 91/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 2.2152 - accuracy: 0.1928 - val_loss: 2.3169 - val_accuracy: 0.1900\n",
+      "63/63 [==============================] - 5s 76ms/step - loss: 2.2618 - accuracy: 0.1820 - val_loss: 2.2920 - val_accuracy: 0.1780\n",
       "Epoch 92/100\n",
-      "63/63 [==============================] - 3s 55ms/step - loss: 2.2250 - accuracy: 0.1830 - val_loss: 2.3118 - val_accuracy: 0.1960\n",
+      "63/63 [==============================] - 5s 77ms/step - loss: 2.2383 - accuracy: 0.1817 - val_loss: 2.2895 - val_accuracy: 0.1770\n",
       "Epoch 93/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 2.2146 - accuracy: 0.1895 - val_loss: 2.3072 - val_accuracy: 0.1930\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 2.2282 - accuracy: 0.1838 - val_loss: 2.2856 - val_accuracy: 0.1770\n",
       "Epoch 94/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 2.2122 - accuracy: 0.1895 - val_loss: 2.3067 - val_accuracy: 0.1940\n",
+      "63/63 [==============================] - 5s 79ms/step - loss: 2.2556 - accuracy: 0.1828 - val_loss: 2.2826 - val_accuracy: 0.1750\n",
       "Epoch 95/100\n",
-      "63/63 [==============================] - 3s 55ms/step - loss: 2.2005 - accuracy: 0.1883 - val_loss: 2.3044 - val_accuracy: 0.1900\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 2.2430 - accuracy: 0.1825 - val_loss: 2.2804 - val_accuracy: 0.1750\n",
       "Epoch 96/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 2.2147 - accuracy: 0.1877 - val_loss: 2.3037 - val_accuracy: 0.1940\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 2.2424 - accuracy: 0.1860 - val_loss: 2.2779 - val_accuracy: 0.1770\n",
       "Epoch 97/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 2.2011 - accuracy: 0.1970 - val_loss: 2.2991 - val_accuracy: 0.1900\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 2.2644 - accuracy: 0.1782 - val_loss: 2.2741 - val_accuracy: 0.1800\n",
       "Epoch 98/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 2.2002 - accuracy: 0.1863 - val_loss: 2.2952 - val_accuracy: 0.1910\n",
+      "63/63 [==============================] - 5s 77ms/step - loss: 2.2469 - accuracy: 0.1855 - val_loss: 2.2724 - val_accuracy: 0.1750\n",
       "Epoch 99/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 2.2168 - accuracy: 0.1978 - val_loss: 2.2939 - val_accuracy: 0.1910\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 2.2493 - accuracy: 0.1883 - val_loss: 2.2691 - val_accuracy: 0.1800\n",
       "Epoch 100/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 2.2042 - accuracy: 0.1898 - val_loss: 2.2920 - val_accuracy: 0.1990\n"
+      "63/63 [==============================] - 5s 72ms/step - loss: 2.2231 - accuracy: 0.1805 - val_loss: 2.2679 - val_accuracy: 0.1780\n"
      ]
     }
    ],
@@ -1149,14 +1169,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 18,
    "metadata": {
     "id": "bvbHxalovXYk"
    },
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 576x576 with 2 Axes>"
       ]
@@ -1204,7 +1224,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 18,
+   "execution_count": 19,
    "metadata": {
     "id": "VflXLEeECaXC"
    },
@@ -1213,8 +1233,8 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "10000/10000 [==============================] - 1s 106us/sample - loss: 2.2606 - accuracy: 0.1885\n",
-      "Accuracy on test dataset: 0.1885\n"
+      "313/313 [==============================] - 4s 12ms/step - loss: 2.3050 - accuracy: 0.1786\n",
+      "Accuracy on test dataset: 0.1785999983549118\n"
      ]
     }
    ],
@@ -1245,7 +1265,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
+   "execution_count": 20,
    "metadata": {
     "id": "Ccoz4conNCpl"
    },
@@ -1256,7 +1276,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
+   "execution_count": 21,
    "metadata": {
     "id": "Gl91RPhdCaXI"
    },
@@ -1267,7 +1287,7 @@
        "(10000, 10)"
       ]
      },
-     "execution_count": 20,
+     "execution_count": 21,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1287,7 +1307,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 21,
+   "execution_count": 22,
    "metadata": {
     "id": "3DmJEUinCaXK"
    },
@@ -1295,12 +1315,12 @@
     {
      "data": {
       "text/plain": [
-       "array([0.09780582, 0.09529545, 0.10379922, 0.10234217, 0.10320072,\n",
-       "       0.10058132, 0.10370158, 0.101363  , 0.09767633, 0.09423441],\n",
-       "      dtype=float32)"
+       "array([1.1986407e-04, 1.6519021e-02, 6.6521651e-01, 3.7694808e-02,\n",
+       "       4.0283995e-03, 7.5087927e-02, 9.1714531e-02, 9.4427909e-05,\n",
+       "       1.0516093e-01, 4.3635345e-03], dtype=float32)"
       ]
      },
-     "execution_count": 21,
+     "execution_count": 22,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1320,7 +1340,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 22,
+   "execution_count": 23,
    "metadata": {
     "id": "qsqenuPnCaXO"
    },
@@ -1331,7 +1351,7 @@
        "2"
       ]
      },
-     "execution_count": 22,
+     "execution_count": 23,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1351,7 +1371,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 23,
+   "execution_count": 24,
    "metadata": {
     "id": "Sd7Pgsu6CaXP"
    },
@@ -1362,7 +1382,7 @@
        "array([3], dtype=uint8)"
       ]
      },
-     "execution_count": 23,
+     "execution_count": 24,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1382,7 +1402,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 24,
+   "execution_count": 25,
    "metadata": {
     "id": "DvYmmrpIy6Y1"
    },
@@ -1432,7 +1452,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 25,
+   "execution_count": 26,
    "metadata": {
     "id": "HV5jw-5HwSmO"
    },
@@ -1448,7 +1468,7 @@
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x216 with 2 Axes>"
       ]
@@ -1468,7 +1488,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 26,
+   "execution_count": 27,
    "metadata": {
     "id": "Ko-uzOufSCSe"
    },
@@ -1484,7 +1504,7 @@
     },
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWAAAAC6CAYAAACQs5exAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/d3fzzAAAACXBIWXMAAAsTAAALEwEAmpwYAAAao0lEQVR4nO3de4xtV10H8O/a+7zmnHnduXemM7f3Mfe2BQotEFopBHloiBGJCCEakEAp/oOKSgJIEROIijaGGB9oSEDRiATUyEMgalEoUCnSNm2plHLvLXfmPmbmzuvMecx57+Ufc4qz9ne1vae0Xb1nvp+kyezfXWe/znSdPet3fmsZay1EROSpF4U+ARGRvUodsIhIIOqARUQCUQcsIhKIOmARkUDUAYuIBJIJfQIioR04cMDOz8+HPg0ZUnfdddeatXba92/qgGXPm5+fx5133hn6NGRIGWMWHunfNAQhIhKIOmARkUDUAYuIBKIOWEQkkIGScIVsbEsF9yWJZy4fE2cfdRsAjPEdIeFQerIgw58Z1vM54tt/r9ulWH172z0D3wXBd7LeC6AzuxjW284Ts499TON5XRR5Xue7Qal7HUUxNUmSnifGxzTp+/PYh0OSJEjsRVykyJAYqAMuFTJ41fWzTqzR5nbZicPOdjQ+S21yWf4fOU4aFLM9t10vzlObJB7h/Xs6mM2NdYp95+57ne1qnc8hzuQoZi7i1vV6fI2+2ecScLue6fAOE/eDJrb8wRMZ/pApFflcY8Oda6/rnsfY2Bi1qVQqFGu1+FwzSL1Pnr+1uqnOfKve5EYiQ0xDECIigagDFhEJZKAhCGOA2Lgv6SX8J28hNeR7/Nhl1ObA1CjF6psXKHZhecnZ3u7x8XoR/wnfbfJ48g9P8fehK5tVZzvxDF10uy2KxRHvP0qNTxdGCtSmkOdYu8vjOMaz/047de09/tP/yOEDFJsY47e5Xq7xMVPbxSIP7bQm+P5srJUp1t52r2nu8jlqU0m1aZxZpTYiw0xPwCIigagDFhEJRB2wiEgg6oBFRAIZcDY0Axg3w5Yf4e/IHrncTbrNz/D3SW2Xv/MZZziZFk24Satai9s0PPUb5zc831dtVCk2d9l+9xxynu8Ze767O1LiZFS64CHvSbj5Chka27z/zc1tipVK7v4aNU/yrsuJuVq5TrHKOt+fI4dmnO2NdU6KFkt8TdOTJYo1Mm7icnqqSG3Sb5u3YERkiOkJWEQkEHXAIiKBqAMWEQlkoDFgC4NO7M4hMDPNX7CfzLvjwisP3Ettem2ec6Hb4rHKJFV4Eee5gKNYGPfsn8c4czkef80X3THNfIGLD44dP06x7UaZYulRzZUVHkNNj+MCwNSkZ3y07Sn+SG23LV9PeZXPa98+vqYJzzwPUeS+b9ksv259jQs4ZvZPUOzgZW5ByPY2v9/nlped7U7HM/+FyBDTE7CISCDqgEVEAlEHLCISiDpgEZFABkrCdW2CzYabTMnWOHG2sFV2tk11g9ocmuVZu1rVMsUqFTdWKE3yeeW5wCJOuEhh3DMrWJxzP4NKo1xYks1woURtiwtJikU3QVjIcnJqbWWNYtaTTDtyeJpi9Yp7722bE5ITY5zky/ElobLFhR7nl9z3acSTkOz2uFjioQW+pvGim1jc7vD70Updt6feRWSo6QlYRCQQdcAiIoGoAxYRCUQdsIhIIAMl4dqdHs6sbDqxjVVOgBVSy79PFz0zaO33TGFGtV5At+0malrJFrVpbnISaGSUE0hTBb7caiqpNDrKibPlJV4qZ3yc27XbbqIpXcUHADMzMxTL5fm6S0WONbbTVYE8cxvynHFrW66qy2R5/ysX3CRcq8SvSzwzltks39eVVJIyl/Wt4KzZz2Rv0xOwiEgg6oBFRAJRBywiEog6YBGRQAZKwhkTIZN1K5yaVU6KtbbdBEzWM81geYtfl425FKqbpJJ1Hc80jZYTetkuJ8CalhNUK2U3ibjQOkNtSuO+qRv5s6uRqhLMF7LUxngSTwaeSrsaV6ptlt17Fmf4erabXKFnE54Kcnqcp/A0xq2+y3gSeq02V+2tl3nqzx7c92lyjO9Fve6eq3JystfoCVhEJBB1wCIigagDFhEJZKAx4Ewmg5kpdxazyDPO2dtyxwSTbR7vrdZ43LDkKdgwsfsZkc3xKUeeMWDb5lgh4bHWHNyBx0qLx1C7VX5dq8lj0TPT7hhqq8VtNjZ4Zrh6nZf5ueKKoxQrpWYYW/csLT8+wQUoFc/y9T9cO0uxmZkpZ7vZ5Ne1uzwGHMf8OW6Me88yMb8f6WKTuMzj9iLDTE/AIiKBqAMWEQlEHbCISCDqgEVEAhkoCRfBogg3mXLsqiup3cxYydlePPUgtSkW+Fv3Ge/HgZv0MeAkULfFyZtOi4s/Ek/hQibjJhELOU4Exp6ChJkDniWVPIm5tFKpRLHJyUmKNT0FFZnUrGO+oo5ej5NdviTZSImTdelj1pt8D+Ms34vRIl9TlLgJtk6D99VOFXWka25Ehp2egEVEAlEHLCISiDpgEZFA1AGLiAQyUBJuJJvF1Ze7S+qMlIrUbvbQYWc7ynK1XGNriWL1yjofNLUEThzzKXc8M6T1cvzZsv/yOYolXXdZn7Oe6rLEM1vZdp1nGLPWbXfo0OXUpuuZpW10lO9hq8XVcSdPLDrbmQzfi1q9TjHfLHO+GdiMiVJtWKfDSdBM1lfllrr/bV4CqZmaNc9zSiJDTU/AIiKBqAMWEQlEHbCISCDqgEVEAhkoCVfI53D1sUNO7NS5ZWq3srbpvm50H+8r5mSU7fL0h5WKu2SQifiUrSc2NX+YYte//JW8/55b2fXAydO8f0912UiRE2fpBOHBgwepTbPBFW6Li3zMnCeJeN11L3C2sxmuZqt7pv5cXuKpJ7s1TtZVUstL9VClNrHnXgNc5ZbLuMm6mdkJanPZrLvU0+p9/LskMsz0BCwiEog6YBGRQNQBi4gEog5YRCSQgZJwxhjEOTdplfGs43bypDv95JEZrkA7csCTmJvjaQ17Wbddvc5JprEJLqE6du0LKbb/8NUUi1OVY8c968tlPSVhs/v5mmxqasvE87qOZ521ricxZ3JcOXb06Lyznc/lqY31lJO1PWvTVbc4wXZ64aSzfW7hBLXpbfGads0q72tx2a1qXL/AbWYPuUnKKFqlNiLDTE/AIiKBqAMWEQlEHbCISCADjQFbWPSs+6X7fWNckNDZN+5sT+T4MNUGj4V2PMsBPeN5N7htOjwL2dgoj4XumztKsVaLj1kvu0UjSYv3nx/lsenONhcymLx7/hnPTHFxga9xfn6eYpHnnqWLP7a3t6mNb7a1ifFJiuWyfG5jE6PO9tQEF3os3n8Pxcodvq+zc+4Y+aZn9riHzqw4260On7vIMNMTsIhIIOqARUQCUQcsIhKIOmARkUAGK8SARWTdhMso1wtg8rC7FE8c5ajNrffeR7Ey1wvg545e5WzPHebk2sQ4J4tanuKDsz/4HsVWzrvL/OQKnoSeJ3F2fukhisWpJNmxq5/DbbK8/3yG928yXMURZ92bnV4CCQCqNS54SBIuLsl4lnaCce/j8SufxW3SywgByHsSes+63C2yqHgKUD77pVvd89SSRLLH6AlYRCQQdcAiIoGoAxYRCUQdsIhIIAMm4YBs5CaCRkY4qQS4bc5v8AxaP7hwhmLNXpZim7Wasz0THaI2vSwnsdZXlih29jTP7pXUy852Yd9+alNf5/OvrK9QLNdxKwCN7VGbfN5zv7KcyfSt/BNnUve+wMnHgidhmHQ5uxUbvtdxzq2Ey/qmgXsGV6sZy+c/PTfjbI/PTFOb8+Wys728/k0+nsgQ0xOwiEgg6oBFRAJRBywiEog6YBGRQAZckijCSOxOzWgynFRKT5J4tu6pzvKU0DU3eXrF5eWzzvYVVxznE4v5c6TtmSIxm5pKEwCmJlJJK0/irL5Z5mO2uSJs39iss13yJCijLN9yY7lSMIq5ei2dhIsivu5ej8+/57mmkZxvqkw3MWcNJ++K45ykbHf4vVw86VYYHstz0u/IETdRl8txG5FhpidgEZFA1AGLiASiDlhEJJABx4BjZFNf1u8Y/rL+VsstnrjQ4bHdsZkJ3r9nhq7y1pqz3diuURskUxTaN8Wx3NXPplhcdfe/scbL3jeafP4ZwwUJB/a7x8zluVCiy0O7iIwn6JO61b7Z0Lo9z7I+nvfIRJ5p7EyU2vQUYniKXrJFXrJpKbWkfbNToTblnpsb6HrG7UWGmZ6ARUQCUQcsIhKIOmARkUDUAYuIBDJQEg7GAHk3CVNrcJHF6bVzznbVchIrLvChMyX+Iv6F2qp7PM+SO8aTu5me5VnTpibHKba56C4tlOstUpv8xgWKtVqeZFdqCrO2Z3Y368nC2S4nqHpZLs7IpAo2ul3POaQzdQBy3hnYPEUcqaWLeh3ef7vHxSzgU0WUcV8bNfh3oJg6XqQliWSP0ROwiEgg6oBFRAJRBywiEog6YBGRQAZPwqVm4Dq3zgmq02vnne1mjmfjSjxVXBYcO31mwdl+8NSD1ObooXmKjXoSbtkSV8eNzrhJpajAVV3xEscunDlFsUYqGVireRJWnsRZ1GtRLDPqSZKl7lmnw/sfGeHquyjDVW8J+Dy6Lfc8qhvr1GZzhZdistUyxQ6MuRWTY56kYq3ZcLZjT8WeyDDTE7CISCDqgEVEAlEHLCISiDpgEZFABkrCdbtdrJY3ndiplbPUrpJKKrU9/bztccKtUOCpDqOMO/3kN//ndmozN3eQYi+Z3HdR+zd5N1mUmeRznfAkkEzGs6/cmLNdqfDUmUmXE5ITRa5Uiz0VbfWau79Mht++2LNMETxLC/U8yzPVKm7S7cz376c2nc0Nik15lhvCpJvwND2uAExSscjoeUD2Fv3Gi4gEog5YRCQQdcAiIoGoAxYRCWSgJFyr28EP15ac2FqTE03dnJtAsolnnkFPzLcG2czBaWd7+QxXYn3j27dR7Ir5Kyk2exkn69Lr0G3XOElWbXimfSxypd12avrGTnON2sDyNbbbniRfjZNW29t1Z3v/gf3UZmKC19rzFBgiNlwdV9typ8U8f4an5pwZ4XOdKI5RLK3nScJ1U9NRehOIIkNMv/EiIoGoAxYRCUQdsIhIIAONAbd7XSyU3SWCmr49pIY5szz8BxjPC3lYEpkR9zNi9vAMtVk4x2OVPzj1PYpN7eNx2/RyPa2qZ4mlEzzz2YpnmaKk544f52IeL61WeWmejqc4w4Bv2tbWlrP93Oc9l9rs8xagFDmW4ViM1HJTtQa1yXuWJDo0y+9JJjW2nvPMyGZS48mxp7BEZJjpCVhEJBB1wCIigagDFhEJRB2wiEggg82GZhNsdNzCi47hZJFNLZ0Te2a5ijxfuvesUgSTem3OUwiADCexFs8uUOyG61/I+7fu+SdtTjJVy2WK/e9JTvLVt90EXiHLtzfxFKA0PcfM53iGtG6qXfl2XjJobnaaYlc/81o+D8P3bPbgnLM9/8xnUptzCycptulJzM2MuksjtTttapP4KkRE9hA9AYuIBKIOWEQkEHXAIiKBqAMWEQlkoCScNQlasbvckE08ZW7GLYXjdI8/6FuuJ5txl7vxFMuhkOfEXKVSp5i1/Opu4p6rZ94zTM/NUWz/Fif5uuvNVISvZ3xshGOGlzfq+mYw67n34sICzwz3X7f+O7+wzW/z3NF5ilWr7mxoVcv3dSPhCrq7T69S7Jpj7rZpcxIul/rV6fl+l0SGmJ6ARUQCUQcsIhKIOmARkUDUAYuIBDLw/H9JlKpyS889CcCklp8xPU5G5SJORuVznOApjrjt8gVOWI1Yjo16pmDkMwVaTTdx1utxGm5mmqvLphemKNaO0sszcVKpaLjCDZbbtRo8FWS6VHB6lKeeXHzoLMX+ceWLFMtOHqbY8oq7hFK73aI2ccy/Msamk4/AwqKbILz2yiPU5vjMZHpP1EZkmOkJWEQkEHXAIiKBqAMWEQlEHbCISCADJuEMkKomi3pcslU0bgVVqcQJt1KO12cbyXPiLD1tZaHASaziFCfh8tlRjmX486bWctdos56pFRs1TjKNpKrSAGCu6K6N5puC0XBeC72uZzrKDiekeqmkXpwt8esO8j1cWOKqwKXVMxSzqak/49hTd+grFUz416hywp2ac3Wd19WrPPugs73d4vsgMsz0BCwiEog6YBGRQNQBi4gEMtAYsLEG2Z473joS81jovqI7/lrM8hht0uLig3adBxij1Lhkr+lZ0qfBA6uZMZ7Jq9XcpliUWhanse0ZLz1/nmKdFh8zn3GP2Wny9VjLY7v5mO+PsTxunmRSs8x5ZkxbWeXzL3e4EKaX867/5B4PPCZrPUUjUcS/A7br/g4srFWpTfnbDzjblTqPtYsMMz0Bi4gEog5YRCQQdcAiIoGoAxYRCWSgJFxsMpiMDzixsbEx3mkqcdZqczKn2/YkXBLv4kXuOXiKA5pNToh1PEvg1KtbFCsW3cKFjVVe5ufUyQcpVihywcPkxH5nu5Tn5FQceQpJRvgettqcwGta9z5WVngpoLNlTsJtJ57PWU+NhU3P3ubJ06ULYwAg8cx2h9RST9ZwUnS95R6g60lQigwzPQGLiASiDlhEJBB1wCIigagDFhEJZMAkXIzx/IQTSzjXhXYql9JuczbHembVsolneSPjZot6XIgFY/kyJic4SdZLOBlYqbsVWueWl6nNysYGxY5PTfJ55N1zzWQ401Uq8QxmxRE+17jNF7p+oexsn1hcozZbDc+N9ST+bNvz2ZuqhIvMY78fgDefBxu597rlecOzSFf76XlA9hb9xouIBKIOWEQkEHXAIvK4zM7ujFpdzH+zs6HP9ulp4GXpReTp6eabb77otrfccsuPfbwVrll6QtruJQN3wCadT/NMT2ji1DJCea6CymY5Zj2nky4Iiy1XXY2Pcxpo/uh+isURJ5VOn3WnmrSe6TWvvub5FJv0JM4yqSo94zkeVZsBqFc5mZZYvqYLK24ycH2dE1tJzPc18twzWP7jx5h0jM+/0+Hzjz0VbNa47XoZztZmOnyuInuJnoBF9rin+sn5qT7e05k6YJGnkUE6J2D4O6jdhrHjVhJORCSQgZ6AI2NQyKXGJg3vIpdzx/ayGW6T93x7v+WpIag23S/0x56xywP7eAn6sTwv6VNe4tnDNs4uOduzY5PUZmJiH8WSJo9p9nruBXS6PPba8MzS5hsr3vbMhra0khorpjFbIPaMYUeGz8NG/J6kx4B9hRi9xFMJ44lNjrn333hmsdtaS1+jZ/q1S5SeZOViaAhCRIba03noQh2wiIjHU/FXjMaARUQCUQcsIhLIQEMQmUyMmSl3NjTf0j9RKpdiPIUA3QYvSZTxFGeMj7ozeTWbNWpTrW1SLOc5r16LY6Nwk1YjnuRdr8bL/NgOX1M2dm+nMZwQy3tmQ+t6kml3ff+7FFsru9ce53lfvrc09iRKE+9MZ+nz4KRY5PnM9kyQhmuuOuoer1amNnevu0UwWpBI9ho9AYuIBKIOWEQkEH0LQuRR6Pu88mTSE7CISCADPQGfW15du/mP/mrhyToZGR6fe/D2x/Oyo4/dRGR4DNQBW2unn6wTERHZazQEISISiDpgEZFAhqIDNgYfNAbvfhL3/yFjcMYY1FLxtxuD7xqDe4zBN43Bs/vxlxiD+4zBncbgqn5s0hj8hzGPfM+NwT8bg+Oe+FuNwUee4Gv6ijHgad5E5CljrL30pwA0Bh8EULMWH36S9v8iAAsATliL0V3xcWtR6f/8GgC/Zi1+1hj8C4DfBDAP4HXW4l3G4MMAvmgtvvYIx3gOgD+wFq/z/NtbAVxvLd7xBF7TjQAOWYsPPVH7vFQZY1ax8/6KPBmOPlL+7JL9HrAxeD+AGwFcAHAGwF39+PMBfBRAEcApAG+zFpvG4CcA/DWABMCtAF5lLa7pd3yfAJDDzl8Er7cWJ3Yfy1rc0d83UvHKrs0S/r92t9M/fhFAxxhcAeDwI3W+fW8C8Pld13cTgPcBKAO4F0CrH58H8DcADgBYBXCTtVjsH+Mf+ufxeQDvtBajxmAOwGcAjGPn/f5Va/ENAF8A8A1AHbCSyxKMtfaS+w+w1wH2u4AtAnYcsCcB++7+v90H2Jf3f/49wP5p/+f7Afvi/s+3APb+/s9/Adg39X/OAXbkUY5b88R+HbCnAHsGsFf1Y88H7B2A/SpgDwH20w//26Ps+zbAXtv/eQ6wi4Cd7p/T7YD9SP/f/hWwN/Z/fhtgP9f/+YuAfWP/57c/fK6AfRdg39//OQbs2K5jngDs/tDvp/7Tf3v1v0t1DPilAD5rLbbtzlPoFwDAGEwAmLQWt/Xb/R2AlxmDSQBj1uJb/findu3rWwB+xxi8F8BRa9EY5ESsxV9aiysAvBfA7/Zj91iLF1mLnwJwHMASAGMMPmMMPmkMLvPsag47T7QAcAOAr1mLVWvRxs4T7MNevOv8/x7AT+6K/5Pn+r4D4Kb+MM211qK6698uADg4yPWKyBPnUu2AnzDW4lMAXgOgAeDLxuCnH+euPg3gtbsDxsBgp1P+fQAfAPDbAD6GnfHhtAaAwuM89iOyFl8H8DIA5wD8rTF4y65/LvSPKyIBXKod8NcBvNYYjBiDMQA/DwDWYgvApjF4ab/dmwHcZi3KAKrG4IZ+/A0P76j/rYOHrMWfY2fs9LkXexIPf8Oh79WAO3YM4C0AvmwtNrAzHpz0/yt6dvcAgCv7P38bwMuNwX5jkAXwi7va/feu838TdsZxAeAOAK/3XN9RACvW4mMAPg7gBf24ATAL4PTFXKuIPPEuySSctbjbGHwGO8mpC9j5M/thNwL4qDEoAngIwE39+K8A+JgxSADcBmCrH/8lAG82Bh0AywD+MH08Y/DHAH4ZQNEYnAXwcWvxQQDvMAavxE7SbbN/7IdfUwTwVgA/0w/9CYAvA2j395X2JQCvAPAVa7HUHzL4FnaScPfsavcbAD5hDN6DfhKuH38ngE/2k5P/tuv6XgHgPf3rqwE/egK+DsAd1sKzFKqIPBWG4mtoF8MYjFq78z1eY3AzgDlr8VuBT+tHjMEIgK8CeIm14NneH/v1RQANa2GNwRsAvNFa/MKjtP8zAF+wFv/5uE9aRH4sl+QT8OP0amPwPuxc8wJ2nk6fNqxFwxh8AMDlABYfxy6uA/CR/tBCGcDbHqP9/ep8RcLaM0/AIiJPN5dqEk5E5JKnDlhEJBB1wCIigagDFhEJRB2wiEgg6oBFRAL5PyMQLm7+jZrgAAAAAElFTkSuQmCC\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x216 with 2 Axes>"
       ]
@@ -1513,7 +1533,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 27,
+   "execution_count": 28,
    "metadata": {
     "id": "hQlnbqaw2Qu_"
    },
@@ -1529,7 +1549,7 @@
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArIAAAI8CAYAAAANw0goAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/d3fzzAAAACXBIWXMAAAsTAAALEwEAmpwYAAEAAElEQVR4nOz9y5IkWZKmiX18LiKqamZ+ichbdU11Vc8CNL3oJ8ACRHgLvMmsgO0s8AB4B+ywnVeY2czQABiiaWqaqu6qzKyMm7uZqYqcC2PBfETUPCIjIsOzspDpykEaqm6md5Mj/B/mn/9fVJVb3OIWt7jFLW5xi1vc4s8twr/2G7jFLW5xi1vc4ha3uMUtfkrcgOwtbnGLW9ziFre4xS3+LOMGZG9xi1vc4ha3uMUtbvFnGTcge4tb3OIWt7jFLW5xiz/LuAHZW9ziFre4xS1ucYtb/FnGDcje4ha3uMUtbnGLW9zizzLST33gwzHp56+m7d8yrkW+8/5D5kt5Kff14v767d+/eP7rp1b58Af+fPryMd/1Xq7+b6/5na92fa8P7vft1/3Od/3i88jvv9O3boH66yn7dwfwm6/W36nqz7/rmX5K/OxnP9O/+7u/+2M93S3+wPgf/8f/8Y/69/z888/1b/7mb4CXx82fIn7f2v+98R1v7/e+46t1vd/p+1f4/tgf//rfF9ef78Pv9n/6n/6n27r8C4o/9rp8eP2Zfv6rv7Z/qP1PVem9+7Fk/0YhxkhMERFBRAgSQK4PY2EcivLBv/efvbytWzaBIPLB832YX2V/nCpduyUkEX+s7PeSDx9xHS+z+Q+dHVR1W1dy9YZE5PqD7Gvvag3K1fsa8Q9///d88cXv/sCT0u+P25r8143vW5M/Gch+/mri//p/+W8QXyCCHUwxBjvYYTuweu+0Vv1A7bYwgBACMUZgB2xjMTN+Kv78ovZ04gC2ZdBk993u04mxItIRgUBgW2IC4gura/PXgt6VfrUu/JOAP1YVuoKq0IGqOIi+eo98NwBvrfuJClQDqrKdxOwt6fb9qSpN7VmaKqVDU2gdSuvbe/y//z//0//+U/5evy/+7u/+jv/hf/gf/phPeYs/IETkj/r3/Ju/+Rv++//+v9+S5Aev9cd8qW/FRwHZ78Cp13hVvwPICoJ8x1rczxvj9ot3yfaDPyKQ/dWvfnVbl39B8cdel5//6q/5v/0//l+gHRy8lnXh+emRWovlyGo58v7hgbdvPyPnzDRlDvNMCIEgBkJFhBCEGOx2FCEI+23w/DeO9I72impDBKYpkqI/NrDl7hB8TQkECYgIpRQulzO1VmKMzPNMjNFzvYNthIDs62NbXp2ujb3IdA2Yr8CwRymFWipg2GAA+ZTShhNaa5ZTe6e3hmpHRMgpb4/B3/v/+f/0f/xj/glva/JfOb5vTf5kIAt7xZABTEUICt0PatmYC7IfsDJuGzDdcsNV4fLqrW87yYFvxf9vj5HtXnL1KmF7Dt0rxdq3dyxjB8wOQBWhX71fFbvdEboKKtBVaN/Kj/oiqdni2z/Ky4U73vdWa/XLABx69X6uvhLhD066t/h0YySUf2ng+uHr/eEP/I5/6st/j8aL+MrfF56D2O38Me549fvvXDSyX2/L8fcvrt/32cbPb4Yyt/gxIRjwhLCd0HsLxBhAAw2l+wIwwAohQHTAGoJcAVm2+4jn0eufxyCebZSA5zarFFmO5goQy16hDTLApuw/c7C8vQb4M+5JSTDgCS/zX+9enBq5/uq7uK4AC4KqbiB6PE8QQUIwEB/CtyqxIy+GDdjbdyuj2PUnOv/d4l8/fjKQVYTmB7PSgU5UAY0EhEDE9lABUTVQ65XTUbG1RfLiSe0Qv2Lu9q60UVkayFch6A6OheAV274tTmADrAa0G+iogHb/PaDi1dZA98+lElCJqAi9C13sPk2hjarqeC7Fns8XV/STj70v26luRVzxhd2tMu2f0C7CtiF4EeMkcmMz3+JHxouT/p8oPho0W9tiLO+Xv3pRRL3axI6EpexJS/weo0vjj/rg3W5PqKo/+N6/jy71p9os3OLPO0RgygkrpIz8k2glUUWpKL1aGSOKkiKkKKQo5OxgjgFeB+DsCBAleHUWUoAYZTvCxYtAzTs0gUAMkEJAxO67gVbYNo0DwKJCjIL2sOXrsb7GRtEquzvgHUui4V0h1asVZ9Xe7fkdbquD7LGeDJyGbwHZINYZ3TOsUS9SiKSU9i/76vVu8ZcfH1WRteW4VxVtmehOE7iCmiOuK7PXcb1D+64jcOvI69W/P3zmF9XQ692bg9lxW/r2JKrhBTmgi6DjQqCLAXar2FpVVtWffyRfDeAnlZ0+MN5T2Lk83/pc13VXryTpNUlh7Hrlw6/rFrf43hhtuX9pQPvHAnIyyiu6d1EMo45jf38dtQf4HfZSz3aukavb31uV/eHK6vd9vj/F93uLv4ywVrwdr1Z4VWIwQBeuKq5d907lqMqGKwrAoA2ANx/kg0uQvbJqCwXxUpPdZ698hlH13Cqx472yUw38PhuI/fC+7D8LYf834IDTC0cjBYq9n71ras/brzok2/0+uOyPGFnRXkw+uD/cGpifWnwEkB0AcRyoV+DQ918GAQPajRe0cdc+OOAF48A5jnvxGmMheS0XNTYOaLBqqhgEFW/xa1c0eDtjw4jqXLpx+AdfBMEuaqC1kVAN1C5UhY7SOrQmdFVaE1rVKyDb/dl02ynHAHFb7F4phhc7WBXZsa1sH3WLbWfL/j3d8uUt/pD4UwGsP1ZVUq8B7PjZ9v/v3tlubctvtRF1oyF9N71gf84f+p6+7/PdQOwtfnz4sShGbbMOohLEc8eLf++X6JfAoAGMp9tzrlxVPANWod07fNb9k9H586KL5Twr5Fhrn72bweh27lSCvYLkOV/Yf+aUON3WpBWNSims60JXJcVAimFHvdqv1q3THXi5XbUcqFslerzs9hXo6GzuwHw/Z9ziU4qPqsjuLfvm190WxlY+9aWlHW3Nkk68bmHsldNRdBFRB226dQxDjKDQHcgaeh23NxgL2I6Wvi9w/DV00BauFo8B2QgS6BqomugIS1OeVwOxtUNpiqpQa2ctoN1eMfjz5wjZ2zl24rGXyDGSYrATUlSidBBFpaOjKuw75rGzDIM6YeQmRHF+00f9pW7xCcafEsz+obGBQ71KhvpdwFO2U8nVmYKuahw8FJHwMon1fd3vlZ9vvesf/V6/C8zeQOwt/pCwTOjkNQezUTpRrPDSHbDadSf5JYqSghLC4M5avmit0bX62ohIV+OJyuDIKp1G782ArjaMiqDQxdaICEi3PDg6qTjfNQav3OpVUWWn5Rkg3sGtajeo3Bu9VXpX1uXC+XxGe+cwT8TDhDilQbxCIxKcTuRcX39OA7ajkOVAVqxINJqeDbUiWbDCVPDmZ+96dV65xacQHwlkx8GyD1J5348NhY2d41XFFq5bBS+TxNh1bdQB2WnlLyjszt95+XZ0ews7tGXv2m87zbH3MyCrDoqVYANdXSmtUdsAsrbua4W1OCj2pCs43cCftYkQx27SWz2KINpdRcEJGaP1I17B1h3MCnvb6AWx/Ra3+AuJb1Vgt83rVQK6WiQGHOXF/burnITwcq2P59o7GWMN/fRFdOPD3uJjQ0Q9F9lBug8pO7j98GeelWxD5oNasm/S9oqs5TW7Di/uY3lwTH/AWCCjkmkFoevZjOvx533rKNcVWNWteivbYzz3dqW1jmqntkatFe2dlqLn5/1yXcjiw9vy8rVl+wc7pcdzozod4zqGpNktPo346UBWnbS+VRYHTPRSv49PBQl2gI/CyhWPZeO0ylZu2XZnW/dihAw9vYgQieFIDDOgaC8uL1LR3mjafDfpIFjGYncujtrpoRGomukqLA0eF6X0TmnC0rLRCjRQiXSEKtCTWCWoN7RVBKXUztJs4SfZp0YvtZNCJwjME8x57Dr9szBaRv2qJeI7Ut/wahCiKi/ONbe4xV9YbMyfb/3CKD4DzFo7FNZSuCwLvXdSyuQp+3CKSfMIeDszbpy/W9ziXzPCdXHHq5ujSjroacgoqShh/F59UHnkQPBKh+cK6VcqBEqQ7jl3pxSId04Fgd5Mzk6MmhdiuKIF6MsCyvXleg1dAckOaGsAXC4XkxRrlbIsLJcLYzD6ME1IEh9c3mHw1o21J2bHzC9B7/47B+MbIA8vgev2/d7iU4mfDGQFCINSMBYUuvF1ggSiC1qJKATbG8qu8WFAc5MYYKvWdueebj3F7b4RQiaExOn4wGF+oGunLGdqXWlt5XIp1FJsUROsdYEgEjFekNB7HDoLFE10Aue18OX7C0vp9JDpMqES6JKMO0ugx0AN0WgGZaG2xfXsVrSuiAppoxkotAq9EILwcAqcDkbaP2Qhx+gnrm4DYcK2AEUg6DhXubyX3BbmLf6w+HMbRroqnm5paNvDqek9D13mp+czX797R6uN+TBzOh1tsrnvQPYwTRzmaZPmCVFePPctbvGnCsGoA6rdwVcnaPdyirX9g1PPgnQCjUAkaEW0ID0QxJWABEQ6EgzkRWGjHQQZOu6KSkPFKH9CM0AMaOt0DV4YGn17pbdm1DyNaL6uju5V0UHXGyVQBVtzGBB+//jIb//5t6zLQquVVtatqnp3ujMaUNj6rZjqwN6J3Z5fu+9u+1Vlefy+WxHNv8vtO3XQu92+xScTH0Ut2Cb3h3ajvmw3XLcF7CIb80D2O14/4V6ZHc+w4dx9KjGGYGLR8+ziyNUOZozvo5hclsg1hTziUJuG7UAbkaYGt2trrBWWonbXFIBIJ9IlgwidiEqy5CqdRnMeUqQ1MdB5/Q1URZsSgzJVyM2oC9mlVWwvuek8bCcH0WtqhPGUgn74Zd3iFr8/vgvEfmxr/Y8e129nz20vctBIkKMA1L1yVFtnXQu1VkIM1DYZZ70bH1+AnpI/9uPXzoff3a3Ce4s/KEbO23Zrg14AL2kGcE0v2H/nAE6GXqv/XnYKwTUl4frno7I55CJtKLobzfWq6jlMVMJ3aD1ep2tV12K/Wqhdla5KqZXLZWFZLsaVLZUgUGuz19zoAN+lxPPBOUb3nDjew/jO9MP7Mc4T++3bjvXTiY/jyAI2MJV3MLtFdAkrcYUcB2aiW2IKIewHqrLxbIzDMxb7ALNeqQxCyon7169589mvaLXy7puvOT89ImVB1mffBRrwFFckQDNooDXlsjaq82DP60LtcKnK8wK1R2KYyOmExEyQRAgTiO2IgzFgkTghcTIqQ13p5WC7yF6p3dxMJAUkZlSg0HiunehnhIZthOckZHdUCdv3pPuivFF9bvET4s+pEjtiFzL3dOXnhOG+11qn1k7rnWUpXJaVWgshRnKpxBiIvtG1Do5fv9hKA9sm8Q95X7e4xcdFMFabpzhxaS3dzvtWhWzslAOnBgxqgcLQjtVxXx1Qt6P+AiLbi3iXbygWNHxx7ddqSgbqzmK1We5qNROD2DDVViX14Sp7+wwFodY7pZuL5VqrXUq7ArpmyNDVrltno8tJ2FfiNZ99KBLo1oWx5zeX0LZx7AelQvT6eQbI/vM7B97ip8XHA1miC8h5wviARrOxVWRvD2jXbS3FbR/ZUZ82frF4vFK5TTUGYZoyn/385/zVv/13rMtK+KcJQkIuZ+T8TL8U279qsEEuDXTNoJFlrbx7rCxr5bIUvnlcWGqDkCGfICQO6cicX5PyBDGjcdAMIi1EFIilENcF7UpvK70WtDeWyzPr5Qx0khyIAZTOhTPruhBFqaocmkl1qQqajKCffOp0LNrte/pwj3CLW/yI+HMCs8Naci/7eIJundobvSulNpal0lrn+bLw9Hym1AohkOeJlCJTzuSUNhH1nSu/G6hsL/D9b8iu/kU+7S0+tRD28/0oV8TgTlyiNFEHpu1KYcBa6NJNyTxEIaqZ53Rt0B0sKmgXehCYbIbEUnFDcNCnjTEcpl29yDSm/jvaG7WsrKWgmmk504JYd2NwVLVvDmFIsg6ImGHRWgq1dy6lcl4Ky1JMQcHQpg9RO4jtSmuKBpAOQffC1V5R7aj6PIo2MybqndYrrQ9zI4ghmKOXPWijFv05nftu8fHxkRa1o8rhYhkCewPApETCNRIb1VauQOqLa7aBjnFPGb3GMSXpfNmYM3meUYSYMhIiEqMBzkEd0OAGBkLvAVVhbUYfuJTOuXSel8pSKiFHUhxadAEJiRAzGjISM0OqS0P09+knhN43+kTvEYkFlRUQNLiorHZ6X+jdlE9yh9Lt07U+1IKuHMBeWhndkukt/v8+bOnvR+p1IrkqtOx3GXJZV5y7lzBz6FzubcvWjVLQWqO2RvPqTOudrsb7G0Mx21DptVj11bX82FV1A7S3+COFXB2Do4G5L5mdAjBoANf0gu13jKHqUW21rqWB1XH886Lt/+I5/LX2guVLWoHJZwV636kAo8J53brfuD7+RE11X4++RnFe+ljHev2w7We/D3D6wLfs72F/L8Ng4eWqfHHO0T+s63KLP+/4KIvaTqYTaC15m2Hn4aSgTD6xb2xTdf7b4LjotoPaFt743SBsw1WFxnZjWmFZI19/8c/EeKSUyhdffMm790+s68Lz0jhXcQpBpValNljWhdKgts55qdSmlAKLRqoIKUxIOiApE/JMSJmQMoQEKTFUE8bUqBHxE6qdsna0BSRAnqbt84yJVO2N9RIpxT5Oa50ldFIUhATYkFgMkSRA6ITerX3jki23HeYt/tC45sn+SwKxD0Hs/trwYeVTedmqt4RmSbT25vqwSu1mblKbspZmILY2lrXY+qkNlYBEO/fUZu3TGpWmAwD7tUD4iOVzA7G3+NgQhkaqaYgrLqnl16KKtkZvjTEMJsNoSEeb3nVXMYpBcJqebNJcNjAmdC/8NFTbt1rtY3i4I9tsiWqjNRuUVu2cnwOlrNvaHIoG43ySdSLlhATopbEuC6VWnp6e+Orrr7icLxwPM/fHI+REU/V8ZiokXbsVoTRcPfcYhFNa8z7tcD8Tf0yv9G5DayJxQHp7n44nWm9ov6Y23OIvPT4KyFaZaBpNqopgB6IfQFPqHLLxgLLANGRGhKtWhaLO83lZnL3Wnd1Z8r2tdK30s/LPv/5HHt9fqA3ePy+cl0qplXfPlecVSlG+eVy5LJW1dN49Vi5rAwloSJgJAjQ1DitxZp6OhDwRpgNhMjArMRGS3V9FnAoBPURScBpEb7QaCB3m+cBhnlGUViu9VVqrrOfI42J6shfpJBpTEqIYp2nqgSknJATQ5tqYA4j0P5DVd4tPNfYipPxJTuQDxH4n2JPRSr2qvCib5J7q4L8qtTWel4Xaug1ylUrrVuUp1cFt65RS6d2uVSISbYCzVEvGKXVqV6JXiToGYscp5g8FpTcQe4s/ViQUHcoFXBkiiAHTYSagvRmlADMcUhU2Byuc9y3NZRt3owSzke2IjJy60w+utWINuAJqxvJKp2nbnLhCDbRafYYFW98yiil7oj4wEyTQW2W5nLmshXfffMPv/vl3PJ/PvHn1ipysm7kDWXHHzIYEMfofo+raHQ+oUSfUhttisO+q905vhd6qF5YEQrRujSriFeXWqm8IbjnzU4mPpBZYG7+pXXq3nZT1BCBGq8RKgOSLTXUfvcCrtxbD+OA7phJH+8TbKb1X1nUBeaZ1WJdGKZ1SrXLTVagdaoNSlbUol9K4rBVCJKTkRgXQt9ajcX0lBLv2YZHBtdtaoePdiPGb+tDCxX44fLP3hRmgu+yXcwdkcySy99gatA132N6da73d3yuyeYtb/L74juPlB07s3+tT/h2H4DWI/S5+/Af2BgxTgd2YZOjCGlBt3UXUW2etThnoptPcFVq13/duieta7r13pUmnNUt4IrKB5OvP9OIjXEvavfzFd3/+q5/fAO4t/pDYaAJby1tfXrxl/rJ6Oi6mbyNX+qhjpZqRwqhL2rV8+BrX8ybj5hWlYajmbLMZo4vj8pgSRi7aK7K9t63V38cgVmvU2ii1UErxtdqdGtR9PXbjyyqEblKbG21gvA9/He3d3bo+lNly7Vi5ojdcUSBeGDDd4pOInwxkO8K5CbUHzj3TeuR8OfPNu0dKWTkdhIc7ISfhYRY+uwvkTbQ58F0e6IP/aq5zV8LPANdLUzuX5cxSAq0Lz6tyKdBUqT2gYYbYkRSRXK3FPyWgEWIkTjMSouFt16WUKaMiNBSJQp4S05SIKZFSQiRQWqWUYguyFtpa6KpIq+RgCyeEQIzWLukVSrMKEpII+UjvjVLUErXC09IIKIccOE2RQ/KTnn44nHKLW/xwqLrUzdWJfXia73NU6pVR8aGQlyB0o2lvT3qF4/T6XiNBf/dxOoCk+htTNR3peJV6m4PY0hrnpbKWQqmN82LDIwZw7T23ptRqia/WurVBS2ssy0oIwnm68Hw+k0Lk/u5IaydSDBwPMzFNm0nKj3UY+SlV3Fvc4tuxD1xJL7Yumw929Y62SnM5uVYKWqtJQA7jA+foSY+uM96IW2VX3Lq2m7a7VltvraKtWAGls1k3v1zMZmrUB7Wgro4LHcwGIaZkxZnxUQQz6nl+JqXE5bxQ10IvFbqZIMUQ0K4spdCB949PfPH1l+ScSCGQohBC4O505O50RBDTgh8KBc26mUGEFKCLnSe0mfkRPgcjYqoLvRn9wIyKimm434o/n0z8dCCr8FyEqpFzm6ia+fLxmf/8T+94Pj/xcJf4/LPEPAV+8Wri7ngkxECUnS+0kcjZQSxAjDu3T731sRUrBVQbl/MzpVVaF84lsrRgg15phpgso+cJ6dZqkXlCQiWkRDzMxJhsCrJWS7Ax0gN20ohCnhPTPJFTYpoyQYTLpdHXYs4oZaG6sxAiZHfziklIyarT66AX9A4hEaYT2irrWihlpTRlkgoV6hx5OEzUyTjFiTFE119SL25xi+8JVaWUsoHHAViHTA0fJKnvBLMvce02xDksla+PRbkCtNePUnbKwAZkMd3kocNuUjwmp7XWxnlZOS+FtVaezos/Vrb829U6Pqq2rsq6mtKJV2AEmHJkypEYA2/XV6gqOSdCiszMm1j8Xoy9XljfDVlVvvs3N4B7ix8fivRqx2qroA3pxRQJ1Fwi67pSS6WvK7QC1c7/SjNprQjSIiGI5SCv0EbM/CeIErQhoyvqbXgrUsq+O93W+XDmbKhWalspZaE5hae1RoyRPE+EGDbe0tiAIoGYEstSKMtKqw16J0ogesv/vKyU1vj6/TtChJQiKQo5WsFnLa/o6s5ko5qsSquFXgtBhBahOSWj14rWZu+jK0iCaEBWRQ3k1tW/41tV9lOJj6YWtHHRQO3CUjqXtZOnxlIsaZXmrjyMWcxvP9OHaeF6UEWvEqheJcBam9EautC7uMBzgBCt5RIwLckgSLQFLiESQkJi3PXnrASMXg+sONcwBLOcNTK96f710SLyyU7TqxzcpRfNVPZ2jGwXS87ik9hQe6e24C0Yf6Rzk16Wxm5xi+8Pha2dPjCacP3Pq3a/+iF2dZ8Xy/A67/md5DtQnXxw7YS6fVDk+v192Op3YD2GNcaw17i97eFG1/CKjmADXWMt2tmlVnwIxkxOWuuEYO1N/fCzvfyavv0FXFexRb7HXO+2Rm/xwyGeCzYjoe+6DOlJpxeIjgvbbdO03J9HVNxVcyTJ68sohISrhe6gkWH7vOepXQarv6ANICObOYkhVMK6EltjXSrrakC2FqP+aFdaM5pB186yrDxfzqQYSUHIUYgxMOXMPE2bjFYUAZyG0BUCpg6k1zxac/3SYINw6ra/qs2lO03G7BafTvx0ICsRmV9BTbQ+USXa8Fc80GKnElhagiKsNVNaIvdAEtDQdjbMtmtyMrs4R5TgVRtPaAoFoTQTVj6XwFJsCXbJkCdCTMT5jpAPhNyZe0FDg9I49AixOrVgIsRoFVlJ23vQrjRttNpopdAMxTIl32GWSAkBCUoRtV1xb4ABZxWhFa/ydttVqo9Ot1ap1UnoCBIT0FlbMT0+Os9LY8qNFBTJppmrQXZ77lvc4gdCVVlqcwe8SBAxHrdrqqpcccmC/W7ssa7dor8brMp3VGS/530sF87LgoiQUiLGiCXJCIhz8qzyo92qMilaP6IjtDZAsPPZuzrvDloIpBBcb9lakeBazN62VKB0I6DX3u0MI0Kw3ihbiVkd0V9/Ir9S5Fu/unVHbvGHhrB7S+5DiLrpxIbeiWqjV1GVrDCpJegZO2ZTU2R1+/VulASryEIkmHpBbVZMccqb1ApA0IgZAw22nhiloSpdK1oLKVhHQzWScnZcbbSfUoz/el6LS97Z8GXvSlkL5+eFWhvv3r/nd19+zVrMqCTkRBDhMGeOh+yV10bQTgyBt29e8/bNG6ac+ezta968ekWMgRSEFG2mxuS8XJHAc7OdD0xVIcXEpAWmjPH5Fhs8uSXNTyZ+MpAViYTpNSJCq4HSoclMC0d6UKqIt/uFpWVKT9QWaKHTZd+d7tOUnlxQr26ataz24QoCiwqXDk3heVWW4oNkcyakAyFl8vGBNB8JpbG2FQ0VSY1VE5KqDXDljAQfTgtGLejNBKF765tHdAuCzJkpRnKKtBSZopHuA9bGGILRGiKIbAMq3dufww/anr/Y7lYFCQm0GYe2mhTJ09rIS2VKtiZzsIWM3tiyt/hxMYBsjNGQqUvGqYyBRauyoHqlcMBL/LZj26uQl2D298UYGNPOuiw8Pz0SQuBwOCJTRoi+cQzbumjV1kUMQo6RECIhJpPPunpz1RNq752WIjXGbe22WlDVzS1JvAVaW4fQNvAwPkOQUZeCHZl+0JH5vZ+R7VFy65jc4keEYMm2X62zgBJ6R3oneE6J2kkoWYWswqSyAdnQlKAGTGVUWwWSGkgWEWJtzhtVpBakVe+8KIhpoG+tiWY0vdYErY0owpSjnTNcO31ZC8v7J9a1cl5Wvn73xFIKz89nvvz6G9a1sC6Fy/OF1hpraVxWq8IObdkhNRbFgLfWlV6M0/6zt2/57O0bjocD/+7v/pZ/+zd/zZQz98cDx8NEjIGalNRsw9pKpZZq73twiXOihwI6WSW2OLWg34DspxIfUZEVG54Y+jaC6TpKNHOCEDwhCRIyvtSwauvvoX1upaCxzC0RD+7s0Jfc7e7UeKRiiU9isl1gMHK6BJMQGYltE0nfXsNA85A2GWEcnUaNddP1U3Wd3Gu+ob/pnY84RKVHFfk6DV+BA2F7H0bEH2LvSu2d2I16oKPvyxXguMUtvid2Z3bns11d2LRdZffcuKo8grfQr+Ddt2gD3/vaL6u51pa0Ft/eGtzbni/WhWDJ2gFoDIOT+xJSa1erMHuvU1V35RAdmpq2TmNMG+1ne56rNbVRLV4srg2u/15u7NWDv/8+t7iFh+FO6zyMLsJ229fGZrOqVw5VXTaHR6XRruTVt3XqhHPjiAraLDeqgz+L4MDX7jvaL9a2NxvYbds4cqWr9XQdiiKdUitrcWfMS2FdV9a1sCyrWUg75317TB92sp3mVrmtrPR1IQTh+XzhMJ/RrlzOF5aL8d7nZFx3Ri52u1yjC1W7XYxHKz6L0mKEIWF2G/b6pOInA9mUMm9//iviU+GxPbNqJeaZNB3ITbl7OPHzn7/iOGfeHhuHYyXFjqjQ1VoCVhIygWdCtB2jCMoE2I4wpANBJqQr5fHM02VxkJcIORJT5nj3lvl4j4REnI7ENKFUYuqE4rtRxRe22rZYdscRwHatDrLXZeXrr74ixYiWwpwS85SpTn5HhJwyOWeCNEptrHUxANx3KkR3bqxg5PZ5ivQeKGun+ommdkWL8Zgez4UgynEO3B1nuiQI1oq5rclb/JgQEcJ8sA1ajGiwDWF3SblRDTUANxKX0XzGdu4FZxZ2kuqPgGzjXgre8m/2KB0M+cHbs39PKdoAaLX2YZDdZR5cAi9GRIJXZKtx8LxLY2DWwQBmAxqiJeKcTXkkhMB0OKIh+PPuIPb6M8n4Mq54xGPr/a3v5QOKxS1u8X3Re+P9u3dWqXQgtpbC5flsXNJl8Z83yrry/PxEq4WcAkuyeQ7dqGyQYySn5Md5JCVb3yFCSLY+1lKp1c0DQrLZEWSzlAelad+ksUqr1NbddSxAh2UtPD4vPJ8vPF9Wvnm8sCyFy1JYK5QW6GTC5AC5dzR1X5+d7Dz3KEoSO8us5ycWtQ1zrZ3npzO1VH73uy/IKTJNmedXD7y6PxFj5DRPzFNGtVOXiykrdLvd6so8T/RWOJ2OaGvU9UJvldpuPNlPJX4ykI0p8/Znv0TzM797VEJdiGkiT0dqF+7v3/Lzn/8Vd6cDd+GZOb4jUQit0cuT+0lfLSpbgVgldgKdQSIp3xOne6iN+l54uhjHNM9H0jSR8szh9Ja7+1cgwZ5DIl0LMRZCaARpY0LMkrqUrWIlIfqaHnJXQllWvj4/W+tHO3fHmTrPxsFNVvVNOZPThDB2p8s22LIB2BC3inJKkRmr2NKi78YDpSlltRPJ46XYrlszb/tMF39vxH23fotbfF+EgEwHX1s+WOjV2G29bXSeQZizCunmtmfEH7zI+QF8fQn+XhyVV1SFoQPZWzOlAB1DLJ1hmRkEcoqkGBBptNoJNK8e24YxxmRueSHQamOtBmTFLaN3HoT4aSSQkgm5D1UGEefbSWDUncKH39sLOsVLMuyPg/C3uMXvj9YMyKrPSwwJueWy0HpjXVa6z1AYkH2mrCspBnIyG55aVkpZATgeZk6HAyEIOSVyNrAbQkeCgcfmmssghNg8H42cZ8PFpRZat1ppFWjgh72toaVUns4XHp8uPJ8X3j1euCyVUipLFVq3AlTIyTZ8XRHfaI6uqwApGlUO7TwR6MXAfGvK0/OZdVn54osv0d6YpsxyPrOcH0gxcjrMHKaM9k4pF6vo9kZZztSycDwcCLDNoKzLmVZ3EH+Lv/z4CI6sEFMmxuzXlZgyKWVSagb0ppk8zSSpBElGFujOfb1u4W23fMgrJEIwrdc8n0jTHRIb07wyTRXFaA0xTPb6MRmnB0El2nM7MJXr7PpdMRK3qrdZDGxKa5h25e45fZ22zUllGCbsCXNUiwwcGLbGW6ZDGzeE4Uri703s+2itUyuU2p1moG7ve01luMUtfn9oV5ZlRXAeqA83hQ3MOnj1w2+Y98RrEMogp+49+P3I/6AUec0XdQD8ElxeX7h+EXu4cxGGkcgwHxkaXTEIcRiVRCX1SBf1DWLy++7vI0Tzd3/x3lwJob6YBB+12avHXr3H7fb4XgVeMNVvG8tb/AGhqqzratXPtmshFwdftVYforL2/bKuxgWPgdasa1lr2YDsyCcxBDcZSK6yo0iwDelo8QuCdCUEdSCrPiNir1W75dQaxDi8gm08g2lS12ZGI82VBPYct3dUR59DglMkXi5BT3Pi793XM5YLU4QYo4Pvbt+DX0ApxWQ7DbyuNsvSq92uBQmBy7qS1oleK+tSNs78LT6N+OlANgTyfGQ+CfcPFdIRiZmlNE7LwtvP3vDqzVvujjNHEiftJF3pl5XW3kFT1MnftnVL0BMSEg8PP+f+1S9JaeZ4/zmH01ta6xw/+5rPv3lPrY3n84XLshJT4nR4xWG6o3dlKZVaO70CXQhEApEYook070vOdS6rLaBaKK4Lm0RJYsMnuzjJRu0DgZgSh+PBnIT8pNFaZ62VUkyV4VqnM6bI5CcPsIXbanVlg06gc17NIrDUysMpA5CjcDcHpjjdykK3+ME4X878z//L/3tLLsKVCYkIIVp1UoIwT5nTcSbFyN3xwOv7EymEKxrpld7qC16px9ZvHzXWHQArkHLicDwSQyBl33BubU0cMA7OvDDPkJItMPVhT6uwGue++f1V1Ts4rn4Am1RX18ayGj+uNq/KqG72n6BoaxjZULfPMagIOUVrZ56OzPNkG4Lw7WHLjb57i1v8iFjXlf/8X/7xis6m9NaprmNeysri+eeyXHg8PxFj2K1nMWpB7w0ROMwzx3kmBKvY5mTduyi6qY9YGHgNPj9incK0DTuXWm0gKwiaktlxikDIqASezwvvnxbOl5XLWqmuGqQhErPZsiNDlAtonc7gxV69CzWAarazwpQnBHj9cOL1w4kYAnO2zxFD3IB+7wKtsl6E1ipPj++5nM907ZRiCgrTPHEucDw+O9g1msZS6p/kb3uLf/34qIpsmg7Mh8Ddg5qDVkgspbMsC2/evObVqzecjhNzD5x6JejK2p9p52R8n62CA10jSiLKxHT/OW9/+Xfk+cjD61/x8Orn9K48vP2Gn79/z7qs/Pq3v+XLL78mxMhxfmDOR2ptLMuZVhqtKvSAS0W7SPPeWux+Rmluo1dKYbks9FbJKSDZ5Ir2yWmukrW1PA+Hg+8gu3vEN/Sy0JrzDT1pCwZcQ7SduUggxWQ78rJa1bcVLuczrdj05zenGRHhMEVynEgxftxf+hafRJwvF/4//+v/urMGYJPiMopLIE+JGIX7uzvevn7FNGV+9vY1p8OE5Dw8QYBrcMoLFKtOx3ECLLBv+sbjUs7MnmxTys51FVckgcGBFYREgMmSq17D8I0j69XZaI4+GxXJlUKqD5K11dz3Wm+s67pVtsq6sK6LDYuVSndZovElhRA4HmfmeWJyVZOYkm9mB0v2RjO4xU+LUir/+OvfwNZdkxcDkM1dI1X7C1vY6yaGDUTa9TxPHGYz+MgpkqNtVCPspkNh78SENDaRwYexowNZqwJLiMhhRtLociRUApel8HheWJbCWm2mowNIJKaxLkZWVbrUjSv7soPxEsjmnIkh8ObNG371888IIVBXow0Mi/fWGr1B7RV6o5TCV199xePjoxWh3PUv58zj0pnn2Z0zV6MY3IDsJxMfYYgg2zSwYUIDaDl7JXGanE+aCBIRItJ3BYKdTuAVzw6N0ZKIhDQT0kz0a+lKyjNTXtEOKWZTRZDo1SbT0bt2GhJGF9XbMNG4pq1fVYK3UpPJErXeSbpPbcZgoNNs+uy1ECEEJboWbAjDDGGnCojqPvk5nn98c14BGkk9hEDvJpGkaha9tXVKsapwdVWDW9zih0K7UrwiOdw1roFs7wGlE4Iw5UwpBRG3tXXNY5Uds4rqCwDr+zhQ2YbFttLsddXWj+8QTctWtjXiFdjB2eUqS3s7f8i1j6QvW7tz3HdQgBpgGrGlWRVoLcXBa/OJageyi01Xq3YHsu261+Lnhx08tD5AhmxdmBec2Stewm1l3uKHQlVZ1pWRN4fhz6CstVYpPgRmBj57EWQDtJhkXhCQ2pBQCBJcjs43hALR739NeQtd7TkJRjEINpBVquU8iRBKI2xqQQ0Vpx5c0Qs2o4LBrWdkcJfW2kwLBufeV8jgxovTImLcTBDEu0Aw1BOuvpur+7dNAcXlLdWkOKWb5bsGG5arZahB/Gn/xrf414uPkt+SmOm6cr6sPD6dQQJvP/scRHi4O3A43Zm3clsJJRN6B4mWHHx4elMKWDtPSyHlwNozYX5FPJzMdCE/QFdSrky50VskxZkUDMymmEkxgxroTNGSsuUeJQbheDyQp8xSCuvzs7d0nK8ToQClFGop5HQgTzNTTtzdP/Dms884HY/uJmaLNGV1H/jGWgry9IxI38C9Vb8yacqgUJr5aCt2gundwXXKpHygSSSshRaVRuDxrHRdOa6RHIXWPsqE7RafSChQ3JBDHdxZe9Jbj0FYF+d1qzIFmxJ+dTxuiUpQ6tAN0J1eEFxaz5Ccozu1RDYw5pY7REjTbG1MEW+Thn1z55mrX4FfGYTdjdtut4c8lgFsn8iuZmfbeudSVs7LYla3y8LlcjauYSlW5erdPexdM7o4/25QDrQTY+D+7sjxdOBwOHB3d8fd6WjAPgwYsVdjdwh/q9He4oej1MpvvvzaN0zZ8sTQUVfLI7WZmk/MiZxtwHFbu1h7XrupgKTayUvZOxXe2cgxkMb9R7lIQOLgpTq1wAclO2LSkrETS0QymyKIYsoHj49n1mJW67VdAcSrynJX2xiW2ljX4jxa4wMLMKXAIdmmNqWJLNn4sSlQy4qIsixn1uVsutNTpE6BHBP39yfujgcuy4Xz+czz5QxdKU1oGNWhngtysfegDmLbDch+MvFxFdmY6QQuy8rT+cLpdMfD6zdM08RhSkyHTIpCLBOhJ4QKEjauqajYMJTCWpSnp0LMkdISMt0Tpjsk30O6R3onpoWcCi0JKcxEMSAbQyaFBE4ajyHQQtiagSEIh3m2Tuj5zLvnJ1qr3mZJWxW3VgOlR51JOTPNM8e7O169esPd3YllWblcLps6gSL03nl8eraTRPOqkpHuSCkxTTOg6GKamr3j06Vm9hBDIkYFDUhckdBpwPNioLnWznG+0Qpu8SPDdReHscdwzNq7BrwYAptTopaJZTEzEO1umED7FpAlClH3LoxVb69k7K4queobuRitQxMcjG5dCwbv/CrbvBgI28ZEriqeewVoLYXH50dKrTxfLrx/eqa1ymVZjEPn4HVwEHvtJr83vh/fkLZWaK0SY6SWwlpWm8hezHNe/D0OuHqDrbf4KVFb54t374khkvNsXO9RyVR3lOw2oJTbxCzBc1vcFNh7020DFtzAYFRsg19PKZN9zak2X8uCRNkKRwZkx2B1BklI7KSWCGnoxjbn7lbO5wtla9Pvq3Gromqn9boB2VKKOXJWc8kEiIeZmCMxBg4pcMiB6O5drVmnZC0Ly3IhhkA5TvSWIQZOpwOfvX3N83niiy++IKboIii71XtZKq31rYiEz8Dc4tOIj+DIwua+pVy1SV62zUcFRkJANLxowwdPqt5MYHhHr+vK09MTtQkxL+TJXDqqT3WaFqZuiXmjEXjlJ+dE791bK+N3lgDHDnbwha6TpnprxBQZEjlncp7Ik1235m2f0W5kfG42A4SR9Oz5xn0GeDXpo2vb0OAcQNVuO+/oklvSUbHdpqkZ3KREbvHDEULgdDzRe7MKZO/bdPMYPBw8u+PhwGE+MOVMir6h0711J/gE8ph8doVXwKqUIlbVlEE3cHAq43m4Zu7Yca07tWa0Ho0Va+/tQ5B49RRXRAB7TB8amP5CKtfrzCehfU1GAhrsdiDQGGokzbojstOPouvWWsiL1/8QzN5S5S1+TChOaZNA6GomlrAlpyEHiYCESAzD3Cf48Sym0epHXu/7yohmZ4dZ9nheY7d33w5a9Y0sY8gyXum3R7pzwTfDITcf2t4bsCW+baF6SeeaIRSE0O0cIW5sklJgmjIpBg5T4jQlQoCcAjmbKVErieZVW/Vhrhit0jXOYTFFUopoUzDTMv+cbMD12nn6Fp9GfERFNhDSDCGZK1Vt7rrRabWjGdtRxkjUTMozISgtZaaUaD0Rg0mCqMIcOlNQuhZ+9+t/4lz+Z6bDPf/mbxZ++W/cM/78Dr08Gpm7V2Iakj2KSCdGuLs7cDjNnC9nnpcn1noxqzyX0RJRI8nHSOtKqR2X2jOuUWuklHj16pUNw7x9y9u3n3N3OvHNu29Y1pWixcwMXDKllMK6FuMTDe0+MSmRWqqBhxiZo6sWSEGLWQnm2fhILWZ6NWpCECWGhgSlS+e8FHpd/mh/9D9J/OpX8JvffPfvfvlL+PWv/7Tv5xOJw+HAv/9v/hsTBh9ANlqL7sON3zQlToeZGCOvH+5IThuwiuXC0H3Fj+cYEiGYzA8hOo+P3TkMXGDd5XnULJY3nisvFRR29oBgjlyDQ68brWDAXMB9gQY/TqnaqdohCGnKBE3klDgeDgzr2iF1FPw/7cq6FMpa6K1xvjxxuZxJKfL64TUPr+85zLMNqW3v5yUPduCCcfsWt/ihUGDtNoilKkT13OWDjGbVEUHgcDhx//DaZjCuqiHrstCb0tXlumoBYJ7MMIAQCdOReDhaa72stLraUwTbbIYQyPmApAkkEuPsVdmASqRKoEun9pXehY4S4kSSiNmtm+064DMpoITNejdEKyZZRbbSq22B376+55eff8acEw+nAw+nAzEIuHJR743HdzOPjxntjVZWvvnmS9blwPqzN4QIKUeOpwOvXj1wWSuP5UytRmNorRuVYEOvettkfkLxURzZEDMicT+QNr257hUYs4uNMZPUXLBSsuqP7TZN5qqrkgOkoNTe+PrLL/jNV43pcIfEV0ynn5FCIJRHwjh4tZmvesB086QTQzBukSsEzHPk+ewl32bVlyDKPGViipTa6A5KwYFs74QYOd3d8fDqFfevXvHw6hV3pyNrKXzz7p1pzHp1uDYD8MU1AYcc13i+2ppxgdx9pfvglnVBArkHwAbJzEPavOKDVISOSuGyLlTWP8Kf+08Yvw/E/tDvbvFRMU8T/+5v/9a0KR3IxhCYcvbKjlctML5sjnGrjEQHsr13Pxb7ZqsJEEIjBqPkhBQJ3Z34XJBWVSm9+TAlqAZG6elaT9Zso23AI/igh4igwZ0zR9HniqoAPgytDmgdzDbXu4w5EVSJ02TDZSr03ra1nSQRJdnQzXlhuay0Vv31lJQSd6c7Xt0/ME3Z5IFkVyvoY3jzW9/4Dcre4odDFYofv240SZRdkQPRjcc+HY4cT/fklM3yvHUvsiiIgdfWCsti7f4YJyaJRhHIM2G+M+CL0DYOu3ctQiLlA0wHG5ROB0KYUIXaoXdoNCrN81hEYrKqr+4bQ6vs7tXZFxrpwe7Tg9CDre2H+yOfv33NYZ5483DH21d3hGCSWq0bvSdFJQSrxH7z5Zn379/TWqHWQnDFksM8czodIRRCWEzZwCuyYy712z3RW/ylx0+nFjDag96KvPKPHmTrMYkMLwnk3j/BqNr2XEEguQDe0iu1L0hIlLJQyoqGQCgroZoIcqkrpa7EGMk6j6c0maucbMgsJVKOUCGECs3bHjEQxapFKYGETkp7GyeG6MYOedO061fWswoO3tsO3P17uV46o63ZO5vZwocizaZaoGgMxGitE0FNDxBh2NPexJ1v8ePiSg1DgoHDzbhjFy8XsONtaFVurTh1nqxdN+faWntPCdKREEh4S1NtU9a10rVzvixcymrV2FEGugaxPrwS/N/RZYMsUQ2Fg12WK8RIdCtOtY9n54tgskOgRFVicIvarW0K2iI9NkRNtWEAWfUTUWuBUmZaL6SUTHprmsg5bc8BYx3bFzQGv27Uglv8QSFjHQz6SrL1FyISBNPs6Ts953r638/9G3XNS6xDFzamRMoTMY68Zcd56pPrK9uGz9RK4rZB22Xu9tfsgyY41Al0v1y/fsAoRddDmU6q95SvxKgEkg14xeQufubkFzcjoUjoNpA65V1SLI1zAWJuZ6XQakXEnMxiaNC7VX0x+oP6xKlu3ZxbfCrxcaPwEm0nVyplLZRcqKUSQzSOqthCVQJrVaQptQkqyXaPokQaKsqchLtZqA2eny6cn0zY+PHdlzy++9ISXnlE6jO1Fr559w1PT4/kPJEOiUOYiClyvJuZj0diFh4e76h4218brZu8iIRER2hdyQd3r22Nw+FA78rheOTu7p77hwdSzpRqQySmT2nAdS2Fp+cLtVbWtfoJwAH6sAjsivriQzrKAMWjNWNTm7YzDWhvntg7kWqgoQGrGCfoFrf4wVA7VoZqh4137OBVbKo5YCAypWGDbJP8ik0ml2bDiefnM+fnZ3pXA5rBugen+xMHB7/PlwvnxTRb/+k3/8wXX37lm1ZrmBpgjZuuZbri/mW/HaNx6IyOtFeQj6cjb968MX5diuScCVE4zBnkznVn947i5mIGVslybnmQQBRbf+VYKKXSeuP+buZyuSemyOs3r3i4v7P3NWXoBp5rtdZnkECKtl79q72ly1v8qBCEGCfyNHM43ruOqp//BWpdWVcbzmrVtFC1O7DtQ2vWZLB6VyQlJteGvXt4xcOrB2JMzMcT83xEgdyKU90sX5VabG4lZFTNDr51sVkMpwe23pwut7qVbqf34rSCDtqAvs2aGEMoOOVIgE5Qc+SKKZNFiSFwfzpyOs7MOTPnZEPgwQfa4kzvnSzCaZ5Yl4V6PrM8PpJDZD0vfPPVN6BKULg7HGlV0Vq4PD1aWzbPEJOPz8kLCcFb/OXHR1ELxqBUb81dqpq1Cloy+SsXXzb7VdCqtnAwcrn4ZHRAyRGOWViDIn1lPS/U2ricHzk/vzfSd3kEB7JP5/c8PT8yzTOv+isIikSYDhPHuwNdOse7A5dyQIKQni2RIgGJyUfLhOy6eefns4HiVEx14XjgeDySUraFPfivXp2ttbGsC6UYpWBUau278YqQWkXLWqiKyL6z3kBGiNswnPZmE+UoySvWWiutip0/bnGLH4qriShxBttQbA2YxmQSWwspRJPqEaF1Ex63iWVv23flshbePT7Re/cqktGCJCfiZBWfp2dTDXh+PvMP//CP/MN/+SdbJxLpzjNNMW7V35ysGhWCMKW0DVgd5pmUIiklux0Dr16/Ik+ZrkdmRrXUXLiGOYK1/HcKw9ArsfORT3mPCrVCnau7BnUOc2JZjCf88HDP6e4EsG026UrHOzFiFa3ATXTrFn9giMlexTQxzQemPG1T+2MQmVW8EmpdEC9sboOTw3K2q1oV1tfOfDpxvHsgxmjPPc+AkPtM88JJOF+QdfU8FTz/2RZXO5uUpHUZ63bpvZkigRpDfejFBufG23ozu2gRgdBNjQglx8AhCSkGjvPMYcpMOVtl1i2pp5xs06hKEjjkxOVy4evDgZwyUQJlLTw/PvnmN3GYZi6poLVRLmckusZ7CKgPUl8Pid7iLz/+COKkfrioHdwxxD1peQWm+oBFLyu6VnozolsQSLK36iRA6Db4YUTyjrZKrcX4drVAM9ce8QGPlN16b6gARGu5WEsye6ulW0sjJbqKtz1tBntMao/WZroyPhgDKWMiclMmgA3QfqhWMBQcwJY8zidqHZvkBAzA7q0dxcwcehS0BwKd6Am5a0BiMKu+W9ziB6L1zrv3jxvFBzUd5Tw0HENkitkqlDGSs7ftpaGmyujJS30IUogpI70b1cbpNsPsRH2SunWlts6yVi6X1YCwBDpWcRqyeOK83KEcUnJybrsNtuSU0AnmbNI74lSBODQ1/WLUpo2l93KIejsnGccXhS6u2QmoNj/HuNe704qGLSawOf6BC8uHwecNtkkQn8q+Lctb/IgwKs/ILUMiygHsiyElNkoPbvVqUlPDhVKvu/0OcNXnPUBiQ0J1JYTrp/aejOyPc3r47jMyLh/w2ZWwP4cPYwYx0wZbQ762xahEBsDVXARzIAfvZMjoloyupNpAWZXtc+NFnhQTU55cc11ZlpUQInkyYDyMJMAoGK01mlQbPI2J/iHB/hZ/0fHTgex15Udtj5di5O5w4HQ6cjoctt3X01L47RffsJ4fieWZtDaCKvdJiFMwO8ygpGiAMAXjvIlUSnnm+dmEpKlntF0QgTxlpuNMzhPT8WAWfDmRppk8H5k6HE/3rLUj8cLxvHplWFlrd/6PbILsKSaOxwNgk98xZUIY7RdFrofZmi2cWu3SnIzf+y4HNCgEQ8Gg90Cr4oNfszugCUgEzNkr0mjRIHai2663RromNOWP+0vf4pOIy2Xh//u//kfj17XuG0y7iPACyJrO8USIgfmQOd7NpnCQnDsOSJo53FmKG9WeECKH40yeJ6iNxiOXtfF8qXz9zTO//d07G4TExbrEBsuCqxfEjSsbmCcDsvM88fr+gXmauL+/4zQfiDmSQ+KYJ455YkqJKZjd9GgdmgD64A/CkOZTVVop1HW11ixX8oBeDQJISQjB1lathfpoldplWc31LAjRgW5KkdPpaPSGEDZa0C1u8UMhIXA4nIwyE90u1pU1DNCNlp7JLS6XBQnBaQZ1m9MYDo+i0DWgXThfCsoTIQamxTqKIsE3nmnjeIeQbIDL+a/WIexW2FEDpzg4JZuNtfFmjYtun2Mf8DLQawA9p+wbTCVoQ1COU+J+TqQYOB0PTMnMfeiNsu75rfmgmOlYd0TheDjy5s1bp0VUfvfF16SUeHh4xekYzQa+2XRa68ranmn2ZgiT0wxucyWfTHxcRVZNZ3Ls5FIQ5snagsaDsQRQWuebxzPnpyfmvnDonYgyCV4VNV5pCN0s1AMEsTZGayvrYoYD2i7QFkKI3N3fcTgefaArI1slNrlbViNPB6a50DtM00wplRC6CU+PNr6fP0IITHmidyXnydqnwVow/eok0h2gDq6sTU3upPzri/1+GNN2ened28nMEmxXboR97SA9UIlWwcIqs0igt4jKzRThFj8cayn84z/9xo5Bb6tb78ES5jWQzW76EWPk/tWJ1/0VOUUOh8zp6IkpJPLBOhTz4cjheDIQlxMxRzqFTqBUZS2dp/PKu/fP1K5UVQYjxqUuNyqRYMNi05S21mNQoc4HcjRqUpBACiYdNqVEjtHUS0YZNBgvvYutzW240pCt0XJKudpguoJBNmcj4+6KUxSUda0Ud/16fj6zLKtVkHN2/Urr/ujWUUmuc3mLW3x/mH375B2/oYahuwNWH2rKQm+dgnUGzJzDHL8kmLOeiBsBYBqta230y0IIgdqUWk07ep5hGsNcau9B1DoSY62IXM9sGMdcgqBDzxynIqj6EOb1p7L8H5MBWaMZmGazAIcpcTxkcgzMk5sjCaAm0dkBbc001a8qqKIw5YnT6Y5aK+/ffcPT0xM5Z+b5xDzv6x0fiiutsipITCQJmD7LDch+KvFRQFZ8wjiE3SbPLvazYXKgCk2FpoHSBWlCxHT16pg0HG2Lrb/h05atspbVaAPaELWEGFNmmg/G20suA6bCslbC84XLZeGyVNa1UaqCBNPlo5GiWTDUrujmHz2UB75dYdGu1pp0cNpdeDbEaBPTA0z3DrXRtQJCDLg6A8RoVal4Neiye88HVJQWhBisziSbu1InxQF4b3GL74/eO0/PZzaZEJsWQTD9xxgiJRZLdHog5IQGLBGsK10jIp1A2wwCkmvQbsdtCG7WMQTXZVMDmaaJ47BzVgOzgmxANmzUAq9wHmZyThznmbevXnGYZ1493PPq4Z6749GmmJ3f02unavFEbu3X0XIttW1dkObWvMuysFwWA7H9GshGVwfx9SdmcFLWunVYLpcL62oV2Tzl3SBlypuRib2tG5C9xY+L7zQQupqtGBlzHI9Dw1hktNIHAHZXLrG1GEMiJe+yRDtWh4a7KRtACmwSVRKjdwqHdbUfw65woL0Te9wBbhjv329vnAQ7v8QYbWhTBNFOUGORT3koFcgmc6cMkxUz/FE3ThhMgX0Y2jV2R8W12ED1+6cnmiqPj8+s67ptAvoAtmIzO3jF+xafRnzEsJdVVFKMzDkxT9mrsZnDbJSCAKCdrsKqmUUzS4twMf5nCIHTpORo1IQYI9DRAF06ncrz+gyPXxNDYs6BKQWCRA53D7z+7Ge+mBOESO3Cl18/wTdnlrXw9dfveT4vaK9AYp4OphMbzPnreVl5Oj+zlEotxU8WA37LVo21dk43+721ml2fCPNsFAE7+QRPgAudC6rYwJaD8+iatyEEpskeZ+DAdui9CfRAVfvOqCv0hSCNeY4kmf8Yf+9b/IVHKZV//Kffegs/eM5p9GYn/eh6jCEEXr9+BVNkkgkW0MduigExMPtU8cPDA68eHjbZrJgSBHPlqq1RfQMYUiJPE2/evuZXf/VLt4g1ru3gotqQVubV/R2HeeYwz7x9+5rj4cA8ZV7f3zFnO3883N+Z3WZO5BDQ2ljbytktZ9dqlarWO5d15bxc6K1TW6UWG1obQFa103qjd7fLTJGwyX4NfizWruxm01uKmZ1IDOR5IqbE4TBTtXPXKjEG5jqR4m2DeYsfDhEDZiqB2kwTdnOzxDoL4kOYgqA6eKlumBCcmO39DAmZGK1zOM1HDoeju1pmppyQYLdTStv8iDlZArxEpiKyVTcZ8yDd1llw/nr0dsqY8dZNxcBnY9wcgV6hF0SVOQWOkxBFmFKw4ow7c/Zu9IMeAmYN6MNsrsOuIuRppitcSuebpzOq8OW7R0SEdS18/e4bai0m/9cM8ErvEIJpSPcbkP1U4iN1ZH1YIgZyii8uySuyqPsREChEegu0CkHh0mB1slsWO+BHVdb2bp1SV86XsxHlZSbFCSSQp5nj6R7EJLu6Qm3K5bKwlsZaKo9PC5elEOikEIgxI8F2e703YrGBjiGrtW/gxhCW7RS7A9nWdgMEEFLOqCqTL6LmvwvFZH62CW8Z/vbq3Lq02WCOCm2XTo2CRtCmdBqoUSFyFnK4Jcxb/HC01vjm3fsrLcZAb4XWDNCZ1rJLYk2J160QNLJWCIut5woUgRQD8zTDvctahTHgIVQfCGtbZcc0kI+nI69fP2y2yt2pNUOx4DDPfPb2LXenI6fjkV/8/Gfc3Z3M8efuyORak7MPgYGaKUPrtFJZLiaCfllXni8Xams8Xy48np9t/ZVqXu+9G5BdFueqm7QQYAOhwb8Dr14ZvWfwbPdqr4RAPkyknFjrkcPpRHCbTICW/wjzsrf4NMLLmX2TxlM3FdCtGDIUinXTjxrucvgjbFhZJCIuhZfSxDQdvEiStgHOlEeeEUJKGy0hDAvmbeDMo49B5THfYfrqh3l2WcjxkNHSb6h2bKSzeReootWs2KcgzAmCeHeS7q/RGDI82pXun7lVkxcb3NaYMlJss3xeyiYLNhw1z+fLNocyNqAKUOuNWvCJxUdTC4KTyvOVX/t+rKtTAQIpz0xTZSkTTSOtd4oGKoqNOzWCVLp0iIWYE6KRmILz2Kx1KcHaJSnPzIcjIgFj3Nou7by8M93YUt0y14Wmu42e9N7c1q6bTp5LiQTBiOgqROloW+l1oVehtYw5FFnFqnvbpdVqn6MW190btrV15yQ5H2pQMDYno6GboBDc+k96gbYiWkjSzKghmOtZHgTDW9zie0K7UpbFpohTQsN+3ArmMDfNmRgDx+NMTIEQzCnofF4RMO56EFoILMcDy/lIqxmi6TNKCC60jm3q1OgvU47c3x0pr19ZJ6PaRtH0kn2oa5p48/o1p9ORecocDhM5RSRALSvahSJCWQ34mrORJcziQLa3xlIKT5eLJbRl4en8bHSGat2V3pVSCsU7LTo4sgJBdQOySiCmIai1d1BSiBt3PueJmK19axvPaNx19IoDf4tb/P4QEXKe7PYYkhaIwW1eezXVGu1sys86eHYu54igXkGd5tkGL2PkuHHXTZbO1pNspgsigsSdXzvkHu3NbP9jTD53H0Izyb2wqSyMgoyVZDvanWuwKaR0ei3U9WJqJhGobrYSBUlj2BPL2AIqTtPRnapkheFhgOIAP0TwjXOpbbfexhz9WlWkq7U9o4H0W7r8dOKjgGwAcoqcjkfKWjgejpvYulivDu3m6nF3/0BImdZWvnn3Fa3BWSMXolVspaJS6KEjE8xH26/Oh4npYMkj5UxMEykfuXt4w5vPfm6LNc+EmHl8fOTLr5949+6J1hrLWijVd3/tAn0Az9WmIWsFt7qdknCaAz1F5ljpy3tqrBQ9UWKnxwS9MOVECEIpC5fLmVorz88Xns5n825fVi7LAr4QzbnLdsQ5mQZmFEwqXtVcTVTRuhLqo9nwSmMKq8mGiXLIxnG6xS1+KHpvPL37+kqLNZKnxOE4kVLg7u7E68/chnWeOB4SISjnpyee37+jt8ohRI7RaEOhFVKvpJTITyem08l4eSkTkhmiSC8cp8SUAvmvfs4vPnuzq2B5e3KabdAlJ6MNzPPsVR3nwrs2dKvF2oS12oBlraxr2Wxzl2WlN1u753WldavOni8Xb00O7diXFZkBXK3w5QoKITBlyH0A1uh8e9kGu0IIpCkT3PkrzzMxu8pBU1qv/yp/51v8eUUIkePdva0Jr/ybLJ4di6rVKHDajWKAcbjzNDPPR3OaC2Hr7aeUSXnyYcSJlIcSiV2s8mpyd8Ym2Cuw4sUUkCvprQH8rFtpm8FmXYsNxO4mCChme6vmXNma3X85P/H8/ht6q6QAk/NjpygcXOXjkBPzlDYJQJtdgVrNjEUxgKsS0BCRlAl5BhXW+sTzebHnmTPT6UBTZapK7SawV9TUUq4r2bf4y46P7ouN6ed5mg2oubWkKXt0aztKYJoONBUkThQSTRtFE5Xs4C5QRVCvyKZcUAmkbJZ2IdpBL8H0K6f5wPF0T0yJaT6R8mQasYglu26cVms5NHopaK8+GLKa7lwfk5sQooHZHqwiS1vpRWgp0OpkYFgbMfoOWZXqFZ9SVtZlMZDsrQ8RobeMJnX+k7jW3vCTt4lRG2DrhF6QtiB9IUgnp0YOSgqQww3I3uLHhfZOWS7QEi0IoomchOz8ueNx5vWrB+bD5NJSAqq0Vnh+fG/dihjBVQLO88T5MJNypmqjtIrESJ5m8jRtyTFH27DOOcMDXKfGGCOHw8EURlLi7nQi50ytlfP5TKkFbYWyLizLxcxGLheXt6ssl3WTu1uXQmvd3PacErQU2zw2rwzt+q/7UGX0c8jQx0RcF9YnwYNCyrINcqVpnzCPORNidDWT5NUsH0q5DZTc4kfEVpFVNovkGKyLYfkgoWpW0EIgSkLEJLvu3OzAxYwdpKbNunmnsJlT2KAOyQCd14DuxUC1vHh/w0ik94gArQ1HQN2phDKe06bH3C7E6T+VWlaWy5lWCylAC/aYHgXNZg0vx5kUQLeBZ1uvm3TleJ+4zbVEJCYIzTTpayWluLkBmn6u0hRqV6idpv1Wkf2E4qdzZMcUc4zENBHTRIimjaoOKIXdyecwzyC4xl0CqXSZqHIgilAlEUiodCQ28mwCz71VLuf3xJhJYSbMxu9ZLgvv3r2zAZR8JsbIu3ePXC7PqJppQi0rtVVfZKsB2d4p3jZpLn0CHdFOlEqgkaUyhcoUIklXtJ5pPdIbzqPr5oqXEyJQamXKadOXLVIHI94HTJTWEq0ZjaFp31QZRFeUirZC1Ao0ksAchMmlyJKoNZduK/MWPxC9d5bnJ3pOiDYTFk8PHObM8Xjk9cM9n719zeEwU+rKWhZa60SBnAKBuMldpRjcF92qKjYrbZcUhClGCPKSQrONNlsIu5rB4IbHFAhRkGYAuqwLa1lYyspSigHW1ky+TqFLwPKZECaxFmLOME2mO10qaVpNBs/dj1C2oTZ8ExlGgpc90c/TbFq6IXA4mC518EGZmEYrdm/VGqd+WHc2P3/c4hY/EA44h92sor7Ril7lDJiWjzliWkXW5inEuwni1LqN5xoGFWY/pvf5DrYKK+P66uZLcAsDko7h0N0cxU1FfGizh2Fq0F1yQGnFNJd7K5R1ZV0XWiloDIQU0CB0n/FQEVcxwge87HRhg9VXjly6MytSykzTTAjC69evmHJiyom3r++4O87UrpyXwlottz+dF9badpm+W/zFx0dVZE0IeWKeT6xHSHkGEr2bmlz0ieDT4cBnrx9Y6oHH9++IaaZUpYUH1vgKDYEuhS4rSCMcMncx0Vrl/eMTj998QUwz83RPzgckJL7+5h1N/zOA8XnUhju++fp3NLexvZwfWZYFcMCq0FRZmy2gfTF2gq7MXBCp3MWJh7RwTNC1Uc8rSqCS6GRUzVrw7jTTWkKc9FSrvY8hYK1q7dAQhDLcwoBIIdIQrbT2ROwLQiPrmYNUchTuU2SeonGRXAP0Frf4oWi18O6L35JS4jLPxBQ5zYHPXv0db96+4Re//Bl/91//Lae7I19++QX/9Ot/5NJWpmg+5z1FjilxlzIpBg5TJgcb1khjUyXKMSeOh9k1ZbPJz4mbB/gkv3pzX8TuY6oke9IttXNZn3l8erK1+2TrtXXTwhwyd6Zdl5AsTF4fOlyBZxuyrJucEZ7Ggzv9WQFpNy8YNIMBZIfBwTzPTFPeN+nBprmrG54oSimNdX3aOPZDCeEWt/i+sIrs7FKOJg+XUiBPaRuYnrIfo9ugl7iclnctc96oL1t1lmvO61UVU9k6gfYG+ADI2s2dfjM6DMPcx4pBvVXK5UKr1Tezss14jEG1VhbW5YneKk+P73l8/55aCocpoXMmhUAKM4opKHQVo+Wo2dKaa6W5WA4RTLX5aqcXHnj1+hWqyts3bxA685T5xeeveXV/Yi2Vr9898nxZOV8WfvflVzw9X/x8c4tPIT6qIjt2hmYFa2234eNsCSv49LRVZEMMTDl7RTahkmlyIIRoRgBmyorEQpYVqYHev+Fyfk/KxsEZ0/7LssK796h21nWltUKtheXyhGqht5VazpRy8e6fvd/WoXS7NvO9bq9JJ4oNnO0V2URRlxFC6DLRxT6fVWSjE+wzOZm+ZXI9PcunoyIbPPG55Io2oCK9Ql3QfibRibGSpTMhzDExBztd1N42B7Jb3OL7QntnPZ9pMUKvxBjRVjhMmfvTgVcPd3z25hWn+xPr8mwaj+w2tl0wB56cPAEFH4YUr8YqUawim6NpVU7T5Pqq+6Q04rqVsFUzx4BJH9PFotRWvDK8bhVZMx9hG5gcWpciwYZWZOjbWlLvLtvzYWxA9mpY5fq2TXlPV0B22mTxxoR2V2Vdi3vQN5bFknpXGxpt7QZkb/HDYVrK0YBjYJO2Mr1X01SeHNQa/cAeZ/KMe9cjRm/TDTQqQ4t8dENeatB+K218R5FSN+WEvikRdG0m29cbta7UUrZuzGAnRJ/qrltF1hRD1tXuH0WZYkDirg073lNXELfBFifqmiffeM+y0XZiSkzTjPi5KUXheJj4+c/f8vrhjmUthJSZny9Mz888ny+01m8V2U8oPqoia63C5BWZ6hqT0Q9GE+UQEePCxEjEzQP80hXTgqyROAtpPiDS0bpaq10iMc3ENBPCtFn31WhJ5XI+o6rOea2eaM6uImCXWoolw5hAgg2WdOPSBIEkJmWSUmaOJxINCdnaE6VyLp3HpdM6pMM909FajbXr1iYq68rlfKbUynK5UNaVrtYiGjp94gszCEhobsGrhGj84CBmC5qjLVQxBwXX6wvfPiHd4ha/J1qrnhgnpmni/v6et2/f8Pnnn/Hq4RXzfGBKE6fDidev3jDnmfVwYj2e0K4cUuLkKgN3pxP3d3fGUZ9n0uST0qc7DofTNgw19CoHJ1VRmho9IKjx3JxaSsd+LyEyH07UBiHNNKKDxk7ximwYgHXTwdz1X2MwjuCg+3xrkcg+8LF7s49ELN5SFRsSU6VfFpa1+IMNSajL89VmUmKmTtJ2XiA3asEtfjgs7/Sxs3M5xrCB0K6mXSxi9KBerQNg2rJ2nM+HowHXEAzg+oZRfIBxMFq3Y35/cTcLGJDVea3eqTSqQ0dbNSDbTX1AhzJPK/ReHTQD6iZIMXiXMUHP9Bg4zDOn04laK4cpczxMplZymMnT7NJ3NsiFWAnJ2zabg+b2pv1nIQRyMhpfTkZ3CiLUUrhcLpRSzQRBlRgCx4NVvkMI3OLTiI+qyOYpMx1mDscjrVsSIyS6CC1EqkTjxETfvbXINM8cDgeGd/T798/EGDjkB+bTK2IQWgn0KtS6kg8X8rwiEiml8fj+vR3grh+nLt48Wn9jR1lbY7k8U0ohxESOmRCSJaZaWVu34ascvW2TeDgemSO0uvC7r97RW+ObxzP//PUjpXU++9mv+MWvTMMWiYiYDNfl+czXX31FKcVVC16aK0gI1LKyTtauDYfINEdElCiWqHOEeYZjNvHrkLDtrzb0ykbwFrf4vlAfQjweDtzf33O6O/HzX/ycf/u3f8tf/ZtfcXd/4tX9A3nKvH3zmTlaFUta6nzuKUTmmDbd18M8WfJMiZCyUwUmUsqW3OI+QIU4/7t3ai2sZXUh+Ehynlz3hBvSxN2rN0zHe2rt3L9ebXK5Ni6LydnZMOfs1axR5bWkHZyLa2z88fnBqESm4NC7Jepa604/sHsav6831mri6aXaoKaqW0+77Fdbqwu1j7Zr90QuG13hFrf4vlAHqiEYB90m9sNGYemtci7LRkdb18UF/a0oJGLmJMqoUELYONyyOc0ZmA3ba1pl1zaVqt0k43qlOb+7dqfHaKe3xTZn2vGBEHO+9PNDDPtGNYiZIMUQaMnUF8Z9Ygz01phz4uD3mafE4TC5QoOdH0bBSxDoQm9scnajmKqqpBg5HGYEG8iObnVr3ZELvSmtNALKlBJvXj1wd7qzzfUtPon46GEvG96wqiy+iFRcJXVUIl0HTlSdimDtFPUJxNgCXcWktaKATqDmvR5iJgRLnqYNuW7Jtw5bSt2HLoZ5SevVWoCtOZvcrf6w6cbWQcTEmINYC3KaI1MSLs+Ny2JDKO8en/jq629Yaycf7nmzrjblHCBEq7LWWlmWC2uplLVQS9208JDRHsUTfKBPE6ZZqdaylUCIkHIgZvGqbUelG1VDgjlI3OIWPxiWJEQgT5l5njkdjzw83PPq1attoCnFyDzN3J3uac0kpEY7PYfIFGyDN+XMNGWXq0re2hef3k9bh3Nw7ZS2VXy6dlMS8IpT3zZ3ozAVvUqTSV2tKuuKBBIXam2knJkPBwPDznFX9YTtLckgNg0NnrzthtvNmssXYUXL7mA0KlH0ZlSG1ljW1Wx6HYT3ZtWpVirqQu3dq2ohysZvvMUtfkx0NXvWUYkNW0XWDX2qtfJLWVnWdaPLqOeuaZ6prTptxyTgBl1mgNhNPfUKCI7r3pWOud813+RVl83SPtz/TAKM7n2TbhrsqKJdTHJLPHc5aBVRBBuEnKaJXg/03piSy2y5O1hK2SrOrdmwF04l8HXcdEjZmv78dg+n7ImYmUsMdhZptdJr800rti5FmKeJnLmtzU8oPmLLoqANoZt2XY6o7gclQTZe5yiCDH3Gw2G2KsmyUuoKvVObUp1uppKReLAJ6vmB411xYBjcO90dvNbVuG0H47ZZeIIS49WklIhpYj6dSNMMS+VxUbQWVMU08DBAKyETUiCkiTQdUAkcTsLDa6G2zjQfqbXBstL6sllaPj0/UWozZxKXDFKFEJNbYAqJRqaTCBwCHJMSRTmGXWZrTo0p2ncbpA02kQ2G3VqYt/gREULkdP+KV28+51d/9de8fvOGtz/7hYFEFdaqPJ9t3SxrozSht7BVdUR2Fz5B0C60op4wuwsSiJsU6T44IoNnZ6up9cb5UlnXioRO6ZGYDMwO69ohkdda901qo/uaKqVZu3Vt9hwSKLWxuuTWdRs1hmjT3SJXqgVKqbap7N149MObfQezVp3qrfqQ5kqpxZP2fj8ZJvUMfGCAuLe+gelb3OKHIriSTS0FQSgoFzHA2Fql1cWKMr1v3Gtz7spb0Uh9Q7gpdGyMUt8kDm1mxSuwfes8dG1OcTCOtx3nzQm53WTAGK5hvsJUjeZmI2QbR3ZUgIFtViaEgPZMnyZ67+QU3TlvdxOzCrLzYj2nv1Cwk0E1sM805je7I1VtfcuL9MJwCOvdsYcIhLRRNm7xacRPBrLWHihIUKY5malBh1qNwK1udSnd2wR29DNNMw8PD0w58/6bRy5PC007a+lcSidrJIaZlCMSG8d7gTDTauX58cLj4xO1VZ4enzlfzkzTxC9+8XNOxyNDWL21BhKZpwxiagrHhzek6Uh8OvPVU4Wl04Hah6NIQNJMzIk8K9OxEXOB/EA8vqWpEiSxrIVlLTw9X3h6eqa2xuWy2rR16yxr4XJZQYTDwYZhokCWhVkLkwj3qfJmMi2904Tb+CkpNNOwVaVrQXU0YW9MvFv8uIgp8/Znv+Kv/qu/5v/w7/8DP/v5z/nZzz8nHe5Ye6RcGufL2RU8ug9VRWKORLGBrY5QMC3oSwWpltCg+cUF3Aes23LGALLNLWIvlFJAAjF1QoiU2ng6n1ndRnZTBGhqvMCum1C6OEeuOag8X868e//oVamdLztNk1Vtg/hQpVEF1rKyuqb05XJhuZw3kGsamJ4Qx7CLJ3oRc/YKwbh4U0g+9CauZW0Pa6XSbrnyFj8iBDt2em0s68VNPwrretkqomau0c04xGXg5pw4HI+WR3Iy4NqbG39Ughp4FbXKbqmV4hSaWq3yyiYVNwYTfaxq8Gv9EgUkmq5tDMm7hWOORKxq69SbaGVRVKzCnF0hIATrMKp2ohifVTCVn7FRZhjIqm4dFhejdB6ucE14H1J3Zoqy0GpxCsZCb4O7a+eDlDKn0z0pxxuO/VeOX/0KfvOb7/7dL38Jv/71H++1PopEYkDLWwwpoA2km2MVY3Kf60oGm3JBb83oBX5AW8KC0IdQ+QTSSHkmT0ckVMA4bLVWlnXlclnGOyGlZNUWP+BlkNGD8fmmaSbPM+e1bdqS9hl2SzwZFIhoXEBFyEE5xtlBevN2pUl9PT09u8FC36pKrVk71ZK9tT2DKFG6y24FV0Wwoa452UVEiZhvdffp0TG6KjKYBbeVeYvvDwmB+XDkdHfPq9dvefvZ55zuH5Bgw5UGGI2So9saFVDTsZTB696opKNv57f951YBGa+J0/JeAtl1bZRiQ41SxSqztfL0fOGyLJ5s3cCgK+pANsbI5Il8WEBb5+OZ9+/es5Ziw1+uqTnPM6XZuh7OXr0r67qwLIsB2fOZy+WymRiouoKIjAEwG4YRbHBmGpq3IRCT2PoLgRAUGXqXqhuX7xa3+KEYI1hWfTUKwcVVMHD79MHvjmlQB4Tox2II4ariOugxwfKED3S1blzw3q/t0hXt1el3PnA8wGuQrfoaXC96mPdE14wdyiW9C1U7venVEKWbhgWTuVPvjqgO2txLxRD7TIEhWWnykrpXZWVXKxgxKD3jMxXfBJey0qoPc7uyCQSnMN3iXzt+H4j9od/9lPiIimynrhd67QiNKIqKUqV5x3G044RrYY0YDMhqV1JMzls1asGyNno368dSrVrSWoR4IFA5nBqikXVdeXaKQXcJj+75tlZrV8YYTRYrJ6b5yOl4Yj7eUbtwOj6ajm2rlPVCaY11CSzrwdxIat+UDmIQcghEhaYL5VKprbOUxmWtLstjHtDmSmIDb0ECeTpwON6RQucUlFNQpiTcTZlTMiHsWczbHkZ3VgkK2k1v70XR6Ba3+IGYpon/6t/+W37+q7/i7uEV+XBEJfL4XLis6pPI1lbca5H4INeyyfYAL8CsF1z9OPSuy66oc7XHauDDibs8lYAUIFBr4f3zM4sD2ebmJE6AhW7KJnNuhBiNM1gr2jvP55XLuRq9J3TE2z2tCbVZQh6VKlMTKa4g0lmXQlnaBmDtA47kbRWgLcEjiAa7dFNF6J50+/gi5JqDd0Ozt/j+sGGvlbIWLmezNh9VRgQb+J1OhCBM88zxcCS4I97pdNpc6oacnFFhim+kZOOpr7VSet2oBYo9f4w+EungdVRboysAWBV2nzHZQC3sa7+zKfX07goD0sFlM/ddnVx9bnuwIsPo0zi/iG8cx4lkp0sYbccpP92KRqW4NfV6sdu9sy4XqgPZGI2LO1clzUckpA/x8C3+guOnA9neWc/vqd3UX1O0FkcU21WOZGAHtR2wipJj5HA42vTmZNPQNjHceb4UYmwEUat8COSYSDERQ+f+1cz9XeVyPvPN+0da71R36OodWleW0ljXwjQJd2nicDxxOB558/oNp9M9IWS+endGCDw/P3N+fM+yXHiOyvl5InS3CZREjCCSCCGjCktRlvrMulaeLytP54XaTCd2iFB3IhLN8Ww+3XP/8IYcOg8i3AtMEd4cMq/nRBDIoZNknBzYlBhsp22tVXq/Lcpb/Kg4HI/8+//wH3j1+g2vP/8Zx/t7WlO++vrsyce6B6oGzvpo63lrjjF84R0AG9/cQ3yYau9kvDwwZfBmR9XIN7Rjs1lr5fn5kbUsXmXxSpE6fUEhJxtSSyE6/cCGOZdl4fl5MSBuE5HWVowNiUbn6X1XJ+i10ooNr7Rq4u5gFdfhNy8xuKSKVV03vVwNhO7t06q00FCxTXnQIUIftyrTLW7xfdHVugLrsvD4/pFS1o33KkGYDgfevHlNzhPzPHM43VkxJiVzw/RNmlEETJ/8Ui+2tnp30KqU3ijafDAq+DCWWdcmH+BMMZJcfSDHsLX8h3sfsFVvR+UXt4+1rqOhWmmu3jGo8qOaKtvIGYPJMyrAsFd9LdrGW+9XQ2jL5WK2761zPj9vnZVlXajFJPouTl0KQchpIsbI8VjJhxOEvA263eIvP346tUBtalDVdORMNH2067rt/hhTmvaQQZWNIZhxwCbsbDxVk94QelCCqrc4BPUJz+iLs3kbcSwSo7zpdm1gEBebTrbbTaapOeXMlLNZULpO6xA7N+qAyaHEsR110WqzzxNaN85ebfvF1udgs+5JLsZETJkYOlkyWRI5ulZsGJSD7Zuy4bhBhr2iEqiT+G9xix+KGCMPr15xujuZ21aI9FpZ1upT0erJ0BQAhre5IpsslgHZMdbE5sHOxordN1xsvxkt+qvCzNVv+thstsLlUlk3m9dyBWS9f9NBJNKCbutzDIDVqvSmtk68IqtNIdrrm7SQt1Pr4PQp6rxZ+0xOoBBBA2xemAzoPnSfYVCEtCsaxvlmTFWHG5C9xY8LB5/Dha7W6i54xiG1yX4DsdN8YJ5nyx8xkrINMjYGwPQOZNtB7QC4lUaj+wAW2DHKpjI0XMRiGHluWORaHrI96NCYhR3M4mIGe0VWg61pCft9Gc8xPvb2nH5btjLXi3ttYPYFhcA6nmspNuTZrEhVfOBz8dtBAr0LMXZiypTaqGNY8xafRPxkINt64/H9IxIzYTpZG16UINbOsOpKY0iChGA12TEw0VonxUT2RZoHuEyJec4cDuajvpyfeH5+tunKtkCz6ePSlTgdkJRZaufx7HxZyeQ5MR0PnO7fcP/wwDS7GHMwusHpcKQ2a7OejndGxA+B5+czdbnYwvcWS9FA0Ujr8P7pma/fPVtVaWmsXWg9OK3BWqgxZULMaIisTXheKzkqhyzMMSMR1phYY0RQqpirmNIp2qg6RlBcwgylatumU29xi++LWitffvUFz+dnLstCniZaU9bVuG0m5L8PV2zQVFzmDbkCdr56R4vRuW52/73qIlfoNWycO3mBat2WHe2mQZkdpOqYnIIt8cRo+tMdS9htUHMlXNnfetuyj8/UHZw7tQCF2gzAenV4tDXs3wZmR3XaZIHMfjeI0MO+oQ0p+vnL2rPCVQXrhmNv8SNCQmCazcVSQjAgG2zIMkjgeDyS8oRE00sem06rUFZsQzVmJwZ4dTODbi15vIOZk3gVNpJcazbH4FVY60iMfZq183FqT98qvLWaLFdvzQ2HDISXspi01jRxOh1JMaEpWkEmvOS3DhBrt032UrYujYHyy3phWc+03lmHQ1jvrMvijp3NqrPrauB1sYrsALuDmlF9nZ8vF7788kuenp4opf7J/863+NeJnw5ka+OrL79iPp64fzOTc0JFSdKMu+ZVDFQJRGvDYULGKUZ6UnLOTNNECI1pnsiz+Z6/ef2K169f03rjH//LP/DNP//O+HbrE61caK2yNCUfToQYeF4b7d0TKSXu7u6Zp5nj3R0Pb3/Om9dvSTEwz9n9rGce7u9ttyvC+flsi7FeePfuHdqMWjBS/aUoj2undlhW58V2NSmgFtAuLgtUEYTDceYwHegxcm6Cnis5wSEFDmmGBJeUyMMDHp8q1cZF7XVUAioRxLynK40+PAtvcYvviVIKv/71P7ntpTtiaUA1GkCVUUWUbUjCpn5tjY4arAvteCXHW+8BMz/ANqOjbSnO37OfO9h03dnND37QjDSQMB1bW2cmM6eidBl+Q/jWbr8ooCKEZFa3vfVNQ1KbaWOatFDzSW0nzV9byPpwmxWcraqq2pEr62ihe/s1kIJtcMNsG+xtUy4GeGOAm537v378Kaejf2qEEDje3aFdOd0/+ObJjq8x0JUmG3BUhFKvtJ39OoTdgOOaZqDdJLskCFNOTIfZNV5Ns90bi/umy3OJotS2y1q1dTWOea1cnH9aa+Vyfjbt2VY3furd3R3wlnme0Z5Jwc8ZuoPXQd/dN51WjrH3Xmm98/T8yNP5kdZMTeVyuVh3ttatE2OWt8YpXi6LDXuy9U1QMdOGKsKyrjw+PgKwrGMY/BZ/6fFR8ltrKcTsnFi5JoePiUTfnWm3RDYKNFuxZk+UxheyS54mcwtrlRjTJqZeWzc7ut5QAiHlrVLSujp3LZrrUJ5I+UCe5peOJMF4R6MCnN0ut/XAWru1IumbU9h57TxfGrUra1PWuvNxOwEVNRkjJ7B3sWlqJNIRagfpStVAC5EmQiVSvaoz2pcNWIkUn6BRiUA0UXmFrje7vVv8cKh2Lpezr60hdxMRzLDEfhavAGjcf7ZN/gpgCE3cPhkxE4DgE1HJJ6lHMt6HUBJRuwHmCKIGjoNidKPRY1Rv4m8VWUHE+w5Xk9k7VejlcJrdTW0Qy9uRu/OWg9fWkKG4cEUBGBXZ8Voy+h86uiPGeu+DVKGD8zNIQFfSRfaMf/w/5C1+dPwpp6N/eljuIYyV5e3+LQcGV+IYVLu+teR3Ss9uoDDUfvYKqK2roTIgw6zIbVqDXCtsyEZRGN2MwUVvpdBadfBYKLVsXNTem1dqOznnrUrbXd9WfSj0uqO/c2X3Nd2cYtGGy2ap/pqFZV23IdDmVdbqdIztPfrm1CrMYevkDGrCqNT+Kdww/9v/9r/9vb/77/67/+5f/PVvYfGTgWxtnS/fXXglM/camEI2vpkUGJzOa6a3L8jWzIKvlMJaFtb14tzYzpQT82Hms7ev+et/80sAphx48+aedVn43e9+y9dffbEd0N25stM0b8D01avXHA9Hm/Z89RnpeP9ieCxOgXxozDJxaMLx/oLGmbpMJn1VTRP2fH6i1sa5wHM1QNp6oCLWec2RJHZKigeYOyDBfOzno4GEmNCY6BEuAd45lW8twjsCiBI8MbbejDJRTZ+zS9x4s711/gRr8hZ/AVFK4Te//setUoPiQDUjGHANMe1A1zsl4pbLQ35rjH2It9jBgKxVZHkJZD1hWvIcHvBmLCIxexUzEyRastXBysX59M4/NbS7JWlLspbUGMmtVgOvrW0yYqaHOygE3bsXzs/vDn19MAwBjQH1NmsKbE5BKZgcl/EIsUuAGEzRYFCntqEYDTcMe4sfHZ0dvFoxJ2w0HHGDgXE4jRrHqMZeXyMGWCUGH5LaRy57a9RSbVPZKm0A2aE4KXuhScEHiTt1LTw9vme9XKil8Pj4SFlXaqueo5sPihq3PQRhuTzY7ErYweQYKMXX4kb5qbZ2u1tBj8rueTlzvpyt2rqulO112suKrIPT2o1mN6hLIuL8eXPiq05FuAa8t/jLj58MZEvr/OarJ1o68jONkCZoJj+l+AHkB7S1N12CozXWspqxwLKwLOcNyB4OmdPpwC9+8Tn/9b/7t8QY+cUvPufx8W95Pp/53/7jf+R///u/94ViyzGGyHw4eHU18+BANqXE6XgyZQTtRK2INqJm5lOAXKmSubs0wnyhXI6kGGl1obx7x/n9hWUxIPu4Qu1WJVVvl+Y8kaeDEehTJsbJCfbZgALiU85CC8JjTKwkQoevLkpcLWlHPyl135221my6W6Gp+ADblWPaLW7xPbGuC//wD/+JXit1Xeit2+BhyA46MylPhFF59WvRQNhmlgeYdeA6qAXRwaxgSiIOZA3Uuu1mSsSYQCIhHwlxQiSQ4kwQfw8+JCVcVYoEOxsFYZfHMvesbajFN8E6Kq+D/+r0A64eZ1d2NrpqAeHTpuAV5DmPyW1IyTVjxUDslIwja6BWCQGi9J144XzFW9zix0THxTZicNUMq9LuPFJe3NpAq28krznq9hxWhayD54ppRFdWe6zbcA1agox/B+fJbooEyrJc+Oabb3h6/551WXn3zdfe5u+b8931UFYIwpvXr02vPQT6saMh0Lo7W3aTt6wunbd6vu/NeLGX5bJxcUsrW6W2O9936MWbUoHNxdiXaN9NcGQfQtgkutZloZTC09OTcWrrjSP7qcRHUAtgKZ3S1PRO8WERMS7bC9ilLx/XtzbDmCy0AYuUIjmbI9fhMJNSpPWTJ5nE6e6Ow/HkAxr2tCEE8nwgJ7OpzdNMmmZijBAS6o2cruJC5oIyBPOi8/gSEv3SO0iia6Cp0PBhLvCBGG+/hoQ4BzGm2Yj6o8IV/GtVN5YVoZEpZIMH2gndoEJ0vb7eldoDTQ3Ilu5DLiqbYsItbvFDoaomW1MLZbnQWyVItA0WgZiyOQK94MPa2hBGhXE3ShAf3hrUAitsmoTPBmQHF883cjEmCJGQGyHORvcJjRCya1ZGArINTFnSFcTYD3uyFrOmHJXXUW3ap5vt5zjnVcc798cH9krVJtUlwpBHsFxuXN1tqO1qiGtc7P3oVWXsqqL8L/8nBW4tzD/3MIqo6ZN7PXGj1m2yVS+IM3xrLRiQ9NtjgHFUV7eKqG3w7B8Mrp9x1/F1PJy1gCEI3btu4LFUUwlYvc3ftW1UhDGY2Vrf2vc7tUe3aupGCajVrKZruTIyKKxroatx203C8kpGTE0ZqPnz7iZD/nHUN6e6UxaGEUrz1yyl3FQLPqH4CNUC+GaNHEvi3DNHzVTtNJnREKC3rY3RJW5JckhvDKWC4/EAqrx9+4pf/uJz7u7uePVwIicbpDjOk5kS5Mwvf/VXNEks68pXX33Du/dPNIW2dqSsxLWz9sh8blYdis8GLkdFls5aG0+XYtfPz/zuqyfOy0IvC/XS6K3z1BJtegA5EjrkWYgqu9amCDFN1joNAmlC0wQiDnQNJPRt2lrokljVifdXfOLkt3tQqhQqDUWoImZ/ORgaf5y/9y3+wiOIcDpOlNXK+nXYHI+1F8R449GBnW/0tJplJiOJNH0BEuEKwIk51u0c97BxZ0NIvpmLxHRC0oxINEDrFVlkn/zfLs6RlyHW7hVT7Z3ezPFLRxWWMYrpt3YmxF59wq6jv8fkw2kQmIKYD3wUToeJw2R+8FMWslecY9CNShB9IG2ct0QCMUDOgZRu0163+OForfLNN18zTxN393fknDfaTxCnq42GAi83SOoJoGg1FQ6cWtMqdPWhrLJxZDfXLV8fIQQOp5l5tmGyPCWiJGRbQ3Zvy0tGFcgp0aeJ4XbH1WYOlMNhRoHWGpdl4d3790RXYyjFQGqrxqk1M4grR65Rhe0GcMd99ors1c+7zeJUr64arUfM/7IUgro29fm8VWGbnyNu8enET+fIqvDlkjmumceeOegEQA8VNKKhIMEnhwkbFcCGOqwNOeXM3d2RIMLPPn/LX/+bX3J3OvHm1R052sxUPE4cjzPH2igSObz5nKfnM+v/9p/48mm13drSaK0QQuFpUWJcjJfTrE0hmP2roNTeeV47tStP5zNffv3IZVnRXsB9pGvL1Ok1mpVIZCYx/IBURpsnbJxCUkbTKCe54oBuRkUANCJN3Xea4Na1wtBj710pUkwFUKD7sNho59xamLf4MRGCAdk1NNrCxoFT9UQQItMsxGxUAhFbm3VpaFstQbRKWYt3TrpXZAB1v3bnu78g9IGthWD8WJFIzHeEOINECLOJlINRdLw6JEPDUix57ok0u1a0AVl89HkkdSJ7OddvGwCVXQQ+ml61iDCT3T8e5ijcTZEUI/fHmbvjgRAgJ6MR2GsWeq/GrwdsIOzKBSkGphwMkNziFj8QrVa++vJ33N3dmYJOsspsEOOvbkVG/TaQBRt6LGVlHS35slJX45m2Wl3TXV8MSo6uZ0qJ129e0R/uXL/96KohbI+xCnHwoWszYTAQLJtpw/XGc55nwKq4vTXW5QIYqFyd/zo0c3Wr1A472n2As5RCLetWEW7NFBiKKyaodpci697B9OFTpxx0f8z5fDZZLq8E34o/n1b8dGoBULqwdigNiqM2cQ3K8Z/6DnFvjY/EJZtziQ1sZQ7ztC3ya2I7eNUn2qBKiMkUAZoNj5XaaNWqsF0LMahLeNhkpPlu2XXtyrnY9WWtLKWx1O5SPfbBugoqybl2CZG893g2OG7UhJGY8ZYREjGaxd4ospnnYPQGjK6wDdMwpKFtStqWujDeNZuC7G1Z3uKHY1Qkr/3UR0sSnP36gi7qHunV5LWsqmNDT+IWl6Pdp73aANloZX5rcyXOuTXlDtVESLYmzEHLqsLNuw6Dv2dt/oj2bok0RFTxIRLTv90VUMa1IGk/SezUgGBmKldfyFBGiV71zTGarubVbQk+2CVqm9U+6APDIuHld/yiEn2LW/xAqO5ArQ9Knb70Hv+uVvhwxts44tVkt0aFUx3IbnzxoS2Lup2sZZQBEPUDrVfR/XX343qoCA26UNwHPsO4r2yDXZ3R7udFW/9bQNaNT4arINcAt7+kJTTnxg8aotEyBBn0DL/voBR0pyIM84Zbvvy04qOoBV8tkB8rf//bb3i3NOYAd0lJAlmFibQJjBsnD/I0czyemKZKisJxTsQQ+Nnnb3n9+p7j4cA8Z0TsoCxVWWvnshb+4R//mf/9N1/w/HzhP/3DP/GPv/ndNgAyXMGGnJCqCZxvHNxhs4ewqjULS608XTq14b7qOyXAMS1dIk2ypbQBALhiyInVe1v7FqGJbfp7JFv1lNtHcxRUOirGG5amhG0nOdpNMmbmbnGLHw4FutFochRCDrRu0nWKQuh0CqIwTwfuTkdiiLRjoZ5mtFml43JZTB6nrKxbqxB62SV0ehv8vI1xxzBTkJ6IUknSiDlyvD8yHe5oqizNOiJdldKNQwedTqUjdLVkLdtuUL0Nu098j2RrA2xCygZYp5SYsvuuzxOHw0wMgbvjgXu/fT9nTjm5taUQo3+OWqha/TM1wGQDYwzEPKQBEylHck4c5olpnv91/s63+LMKowOs9LrSykJLAXq0coVP/et25t/pAYxNpHaez0+8f/+OVhtlXSjrsmm1D23YfXJyZ7oTZBui0m7lXxmtQq+0RIQpT+ihUaptKgdtRjfDkr6ZMNjwlbo7577R667jvEvnWSHGOK/tJaDvnbqslGXdK8gue9lq2ygCG0h3m9zi9+vrSnc5sLbd5wOe8S0+ifgoasFXi8D7wvybr3l4vPBwyPzy9ZFDjswCEhMR54EOuZtp5qQd7Y2744S+uiPGwM9/9hlv37xink0sXbylvpTC86Xw+Hzh7//LP/G//Mf/zPP5wj/9+p/53Rdf0Vv3g96t+zr7hP+VViRqxgMqkR5nNERPmuIKkcYXFBfE6l5N7kS6JEDGsPN2mhg2noqgXbYq0wZcQzTJof1eVuRtmBMLoxLrgyoKwYFAcFae5XjxZH+LW/xQKPRCoDMloUn0ll6nK75BrHSFNB15eHVkyhktDV1twviyLDw/m9vO5XLm+eKqGotSZLT+oOle/elbtcXehQQlUkmhMWd4+/rI/as31N55XFYu1YY8ntdOcfvY1oeNrCBtdHZGq9XWRAy7Bq4E181Mwpys4nqYEsfZOMCnu5NZ9cbIm/sTr+9OxCDcpcAxWvl58Au7Ni51oZYFXIKLgFEkQiZn6wjlHMmT6VDPh9lbrLe4xQ+EqoHYutDKQo0BjdEGfzdVH4txfNvD+qbz+vT0xDdff21VT5/S37itg6IT42ZysvFlg2xVT+2jc6ovNomBwJwzokqqCQFKyVb9HS5fXbchL3P5WrnKhs6hHx+CF1S82irFjRvW1SQutXfKZaVclm/xezfnMtgdzLTTilefu0llDue+fgVkb0Nen178ZCALJg+1ts55rYS4kgKc12w7vmgDS7iCwS4yAmPHZK255FZ6JuEzqiyKVSlrbSxr4bIUzpeF5/OF83lhWQqleEvBKQTot4EsDB6QA9lgRHFcCaD78JZLsm/V1u4gUjdaxAC88EGjcePj6Ae/G1p348e7uNF38aCuH3v9/j2xb4++xS1+f4w23GgxyotpaK9q9IZ02fjjloSAYAON0f3YxSsucWsrjoLPB6okvkl7mT5MUk/oiChTihzmTO2dFoTYkmkmByV6K7LWoR/rWfBDIBuEFIZSQiAn48tNKXHIiRiEQ04cJ+vyHLL9PMXIlCI5Or1AcFmu7hJazU8cfa9sMQDFFY83DDqBGUCM27e4xY+OK+6qqngFcvul/z9svNWXfNcr84+rCX+Jw8TEQ0b38EoVYTAKRjXzyo5W1YyA9ue9fs0P38NORbhu4Q8xEGud2NUAsua4t1MF2tVnGMNd45nEzyQvAOlWrd3Pb6gD8/G7f7E/2C3+HOInA9kOrBJ5LJ1ff/We6X3gqznx/v17Dinw5jTzq1d3lkymyJ0nkPOy8PjuHdorD6eZ+/sDc84cp2yTxcG4obUJa4XffPHIP/72S94/X/jPv/6a337xyLIWHi+N0iOqgS7BzBgUq6JslVh7r5bMnSwg7qAiwr589s3pAKx9LI5u1SzTAhIaexIfjxSGJuc4acjW8rTBE2vzhO9YbgZ22bhDLzbKChAIeUgl3eIW3x/aO5fH99biKwvaG03VhMTV1m1TJYRIQnhMmTlnooqJw6nCWgjV+HdJG8k5s82KO3QZwLh/cI0fy4rSaH2lVOHAxJtXJ/7ql59bVXM+ElKmlJX3T48sqw1tLqV4i9CGH0d512m+BqpdLSEGITsXeE6RY04EEQ7TxGGaiCGQciJP2UFoIy6PgHJphYtb2nY3UFABDRCzfciYTFIsxEieMmnKpBSZDzPTnMnZaAU3asEtfkwIkILZG2urtLoimtDgcpCwbRAtX/jmUWQ370iRacobBz5G2/ClFLehxpAiMmTxQiTEQHIVAhuc6jw9Ni5PTz40Zk5epj5w2ZQF1lpcIcRoNoOr232zd+0cKOMDOs2oO42ptEpx+a16xZcdMllWhQ3M03yd8HyzrZsFdb8CrQOAa//94FVEblS8Tyw+YthLKATq2nheF0Q7xyx8fYhMSfjlmwdyiNwdJu6ZTPIjGpB9//4d2gqn6S2n+RXHw8Rh3oFs6YHShUtp/ObLR/63v/8t758X/uHXX/HbL8xxaymN0kc1JO4tfRn1Gyd9b7tEd/vBKiy2+7NdZPc7dk/kXa+ALB1x29hOgO4DKhuX1Yenw77zlQ3ImjyPCEQf34LrNWsV4W005WqH2Z2kP0TmJdxkfm7xw9F74/L06O3IYu04hTqYbq1BKYgIWZWnEKk5M4dESKZzTK1IK4TeidpwjQHqGAITayNuQ2AveGnjutOa8ehE7njz6sS/+eVnHA9HPv/859zf3bOsK9988w2Xy9kkdC4Xt6I0jtymUMA1R9aG06KAY07mGDglA7JzzhyygVd1vdiunfPlmfNyprfK+fLM5fJsyTlG8OpqPh3I84REsfNVzoQYyLMDWXcenObJgOxhJk83IHuLHxFix6yAGQxUM0PWttlrjGkrL2qO9scAZuEFkE1BaNHyX86JnJK7SUbENZ2zm5YEMTtX7Z3aO0vZp/0Xn/avrbGUxZ0ldRtIM8ODYYoydJVtMDTI4PPCnsLMfUtdVeD5ct6cOIsDWTNDsf/mEJld+WMMbhnld5wD2Kq2NsB2dc7Rqy93VINE9nPGLT6Z+MlA1gChtUBaU9BOacJSFVVhKZWlVGvr5UppxjPdhZQtycTotpYhbI1Es7Ezp6vLUng6G6VgLZXaOq3rBgTtzXzQVmE4YelgF2ykAb2+r/3Er3lx/UPx4n5X3J6Nv3c1LT2g9Yc11Y1T663U3bzLXVhGFSqmjTN1i1v8YPgOTb1yAXacj42ZdpvGb61RSyEAMSrVeWqt1Z171jv0fSJ6DJZ8i4d2/W8Z/+5WzdHu7fuGaGcOgWNKxN5pOZH7RA2B1BulmVNPG7aXVwnL9qm2koKoVYqBOQRm15/N4lJ7vhntmCKJtmIcxWZSfa2bHNmwCNUB0IN8cAnbdbi6XK/xW9zih2I/dgGuVT9eriOjA41r3XPb6E4491U3Wotu3NLx3ON5VUdF0/nsTZxnum5KB6UWq8r2Rq02qGX7Rqc8qDrV4f/H3r8H25Jt6V3Yb8xH5lp773OqblU/1A2SGglkjBAoJCEjRNtCwSvAIAdYvGxACkNYxoHDGIQFVoAInrbANpIc4GjCViCBA0vIQkjYgB4NmMCSWkgtiYcebkuN1H27b91bt+qcvddamXPO4T/GmJm5dp2qOlWn+lbXPTlOrLPXXo9cmWvnyDHmN77xjZ4jipf6ZaUYLPvXJ/FZlaPzXFcKQVtzTIFNuGOh126S1Me0iv69ba892hNYjMaw3fbuma+PffpENgSOxxtqnZlnpdVAxfiysyhff3biy/nrHIfEm3dHaIUhRabzTGliTRRp5Hhzx/F4IA8HJCSUwPOHB776/gP3Dxd+8Id+hB/4wR/idJl59/0zpeEl0riSTnkcUGThAi1OsSSsK6fmauGm/TmWxNRWm6HPw6NP7gKXKnJ3MaK9a1emRErJy5++QqY3k/Xuz5UDaLJdYYVpPZnIKS6SY3kYbFLZbrt9jIUQuL09UOaZ00NZFn1dJgdwWo0wnU48U0ghcomJKeZlXHLz7uGpFianJ0xltlJhU1otS4fxNiBvvbA3ZUzne77+1R/hR8bMl+6e8h05caPKsVXeqAUVqEG4DJnagm0/tSXYi//c8uEMNeooTUGLJdm1VZ5XnybUqqsiNC515lJnFKiiaLDR2ekQSeOIxEA+DKQx21CGnJAUCSkSh4E8DKSciHkg5kxMyZtq9gXmbi9jm0EFjjiuYMymIgc0LYhP5zKamZ/3pdiAjxAQVWpvEmtKmYstXmenwinAKncjzrlFleZjY5cmqUVZQL0K2CuMDhJ7wXObRNbSmH0HV61p144tpuda66pvK6rWJ+MJqjVee5+KIzilFmoty9CE4u99vCBfvkPpkzqvB0oEr8TsqezrY586kQ0hcDgemedoY1WZqW3mPDekVQInMo0xReZ5Zogw5gTzjKoQJRLzyHi83SSypsv6/HThK197j/efP/Bnv/wV/vSf/RHTe22R0qKtBhcPwy8Qj07aR4FvfVg3zrA+t3BTHTkNHqCkT/NCvNvFUtKOyPTvYklkcyZ7aXPbhdlpDMt+ujqBuqasXXcsYAfn4uVsc+vHcSSmV+rL2+01sRCEm5sDlzNcLkItFmgscPmL/JyczmeeT4bILonso8rBrJW5y960Zp3H+LCRpTHqg0ksnmA2YLqc+PrXfpQxKOHpm3D3lEMayMAtkIEWhVkSTeMG/QXPwi1wt+pi541Z4eIUnHOZeZjO1NaYzieenx6M81dnTmUyjl6AEhRCIB1H4mFEkhDHyHBjE4/imImDLUIlm/ZtSJE0JNI4GBcxZ2LKhF4l2RHZ3V7a/JrfetPU+tiC0uL88K5q45JXqEIpNvvDE1nxqoWpCDj/tF1smIc2pwzMS7KL9gTXF4iyStgt49WJaPDpeH56W/hzQKiuPNhV/7VQynSVyH5AZ1r6DEGvlNSu2iM++MemnxVvOuujZu072KCwHu+t3tMRWbBR9J325New3TdfG3ul7GjbFdmTyZ4QtgZTsdGO01w4T3Zyx9aIrq0qwQYcSEz0gQI2waRwOl84XyYuc3GieKVp2PjGdW1iXe36CxwkejHpuy/fNsu4hVvT0Ve/wGyOrSOo/XlxMXmRtdS4LGPZSH+5qV9ETPJrLYd0ZxSnIcRgTSYxxmVU5+6Su72M2WIqUkr0EriTWip2Ahq0AnQ/rYApefRBBUuziZfm28KF7XJbrCjscnqvKMjGq8D5dnWemM5npuFMOT3QHu79s6wJVDxY1Ufoa09k8c+XWnw/GtU1aMs8MU8XaqtM04V5miiOxlZPprU3eQYbfxu8kSvEcEUhcIfeBHn37bA2ia6d4N3Xd9vt46wrhlj3fnWgoywTudrS4LgMS9De6LTqr/axrXXhkyq6SFWpI5rFP8voA9132MTGleIgC9pqIXFTrO8v71XNZjTCLuNVi6O6/fNV2aoewIYysL0ieC+IYEl7C23Z/nqt2YBNj4L41e+y1UtZfXGnFrxe9gocWZ+JHgIhREJwSQwfz3qpyvPTzDn4BJBSGFLgyTHzpVsr04V8SxpvSeMBDZmpQqmNr733nD/z5R/h2f2J954/Y26Vik0Gaa16MPmg9M1WnmRFY5c0ccGaRD7Y8Rhg5aH6CF17rSOyIhASBENseqK5NKCETkUQHwq0Xrh855YybFUbXysSiDkusmODd0abHqZNOAPQVimtfto/1W6vkcUUefrWU9LzxGW+mIrHXJhLW7Rel6KAmhJBQ6x83xylwW4AlUbp04G0dy1vOX7dtknt1RKRVmaev/8uqUyM9894XyPPvvouB4SjRAYJ1zw4DG1Bba5dxRh7kzbOrVBpPLTKs1qYtXGqhedlpqhy0cq5FZoAOaKDIafpeOBwc/D7jsiGQMqD0QgkLPpiEoLJAeZMSons1IIY49p4udAK9nC528dba8p0PlHniTZPhGiTs4Z7U9joE6563NDmigHNYp4qnryWhSq0SndtGy4LXaGnLzzh+iwNThug92Is1wPf3jJJT1H88zEUdZ7rMmioo7MialQ7T4gXNHRJnO21qy7sOnKW2mg1Qk/UZ6MClVLW65Wb9mtM319/XBAbUa0L5v1Z/dl2+4LYKyOy22RWWzNdVpS5Nu5LIajxd8r5Qo4Bffspb9zdQhyQfCQON8RhRENmbjDNja8/v+fLX/kqzx9OPHt4oGil+uq0uU5OSiyJ7HbV1ssSW1NYAnN34H6/0YGqLkItGzoBnsT2ABfBS4oxxoULu9WTXDos1VaspVp5xEbtefKgxkcKITJG09wNITKMJrAeRBhyIAWh1WozteueyO728RZi4O7pHQg8v3/uPgMik79CHZGRJTld0BkxYk3E9FkFoWqjev2gK2ksXPIXBAzZRJoePFsrnJ6/j5wfON7f86wEnn/t66gEimQ62X1BYXpTiQgVZRKlCZy18pxCUeVZK7xbZyZVS2pboQAlCnMSNAaGuxtytDG5483I4ekdkiJpHIjj4Kiq04VEkOgVGJffys6FTdlUC4yzniDE5Vqx224vY6qNeTpTRJiny9JDEVP0Bsu28Ek7msoCzKyNl33wzxaVXJJSUUJQA5Xo0+9kedEilxV6NXV7BnegxXVle/LarDmytUotjctl9oSbZcJu8iEhoY+/FV8Y914VxUfpVj+mugA8LTRC3SLO60jbjxps8Pi54FXQZejQnsy+VvbZEC97Sb6X1EVAVxJ3VSjNcJa5NKa5klIzBQKFpkJthsbOzSb9zKUwu4CyeZ4upbyO9mxPVZMoeVSC6EFb+kLuuvSwhFyBZUb8Uh5dy4yLssCGvtCT0vV321p/XNUndqmvazsvCl+o+naDeIkzrMmzlYgqNGuYKd5huttuH2eCLGXzuNzCph7htqW8ACprctpEXYu5n8Pb5khd3vU4VMjVFjcPduH3WtFabazkPKPi/iaGyIiu22xiDRwFuEijAmcqJyqFxrlVLq0wqTKjFDH0VmMwfmsw2bqYk30fqdMIXkQD2uzuxt8fU6eW4L9cD9ipBbu9nKlSy4wgtFCtrB4Crdq52Af7rBSedVLVkshqHwZwvemujrPGnvW07HJX/hvbQsqK1orHpQXfteYtjLZQ+yStTXLZPxPWqZ3bY+0Vmy0tQVu7eqwjyT2p/aBKAY+CPFePLTHc7/Sfqm1HZV8ze6VEdhl5tzBSAirJR7s2WrDkcm7Kw6xEga/fz+R37zmeZr7t2y7cTwpJmahcVJnmmWfniWfniYfLTFEh5gxR0RLRGhanq6WAmF5r8KAYWnDESJ1zo75nYsmwbvReN1xYXVnta+Lq6GwXmK4I1S8UTW2amOnFWsKAyILGokptdmvYaL/abDZY8GaRmBL5eMPhcDQELAR7fa08XB5oZbKrjneI77bbx5kEYTwcKHPleHuDSECbIOH0wtcviaNCFdNLDroqHBhHtpf728rt/ohtLfFmKQHaOV1VqTJRzmdmTZQQ0VARSda80iqipnd79uT0gca7NC4oD9L4ujRmlHOAhwBVQHOkjhlCIB5GhuMBiYHx9pbDzQ0hRYbDgex0AuKi6Gy+KitVSoIsSNaVDFfnxDtndivBtdtuH2e1Fp6/9zXgul6xTTivG5r88cWrdAVBWMe59vgjEpd12Xab25/dVikr1sRWrQqyenej0xOaN1iqL4djEFLKjPlAiHFJvFH1hi1XQmg98XaKRO3SXms2rV031sGbx5PF7Ov4oI8tjdb+u0VuH3DSysIr3u31sFdIZPuKiw082usWoATj3qHM2qAUAvDsVEjxzOlSeP808zArocBMJWljmmbuLzMP08zDZOXCkBLaWLh7ltvZCrEPHpAgBLUAs6zwWFFZvWqY6peSFQXtdAJdnpelHBN9dnVrxnWyS0qlNttO8ovAFhXWnsi6Q87VEGgJMEhA0kBIiTwcGA5H29tWfSxv4f7+nun8gABRm5dqdtvto00kkIeRYSwcDgdQmC7zWmL8EDOawVq9aC7hsyETfKAK8uHbesHWmyXEVQr1cqGQqBIhCQRFmjVySbNkefLbMxrvUDihPBfl3aBMAiUFpiGgIqQUSEdDXvPNgeHujpgSh+MNx+ONcV6HbAticF5w309Hh5xgtyCvG53YRRN6+/gyAGVPZHf7eNNWOT1/3+7rGjs7L3VlvAkhioMnONVl3U5P4GKIi+KAYShrxWD7uv5527i07SMxXff+3JrIivjYadUlSZUQTdNcIjkljjdHUsqUMjNPF6chWBWx+RjcXklUC56PvhSj3FV9caPY+jL9QDLboaZ+f2VQ9Ca3sieyr5G9GiL7whNlW4Lv9xUb0GrBciqNEBoP5wvvP3+gaWMcE4dDYpoLU6lMxegFta6IkKpcfUwPPL2T2MoMVupQVcKWRwSLOyheOpW+Muy1ib7//R3280OLqV7D6XJB6gT3pQS7+fwgwdW77GKQkk1dUXDeU0OrrSStC9tuyyQVedG6dLfdrk2Ea9H+DaK4vKb/tzmde4lOPaz2delSBfyU1iWTexWlqTKrctHGhWA/fUzspVVaq5xQ3ncU9hnKvVTOopxDYI6BGgTNkTDaiM94GA1tdf5rGqziYeM6H42kXuhC/Xuwxff1XHpZry8vIEys3+Vj/erdfrzar/gVv+JDn/sX/oV/4cd+BxTocnUbKNQqmurVC49PLSBhE6+0Ay9rh74NJLBHjXPqzcZhe36unvu4j+SqjG9Iz+L78Cj6OWVB8GbQHpP69aXH+F4F9SS2KyXodle215wt8vpxFxl5fHfL8VX/LBwJrq7Gsieyr4t9+kR2jU5rbQIsaLSVtGLoTqCJ6bxdqvL+aeY0FX7wh7/K4fgD3BxH3n77Kd/69huUWvna+/d8/dkD58vEuTSm0lAV49O6Gp2EYKPzxIcOxEhQNYCnmRZl8Akj9LnNvcvTKQfmCea8S9lww4/rZgjvmihvvwIwDlPRDc9nedK/E4GUE8n5sIebW4bx6Chv43Q60VplvpyoZaZMZ07vfY359EDOiSc3NwzD8Kn/VLu9RiaBlAZSmmw4Rx9TGYUYZeHQLSjkxtpyai9Dk68Wby8DcOjVfVkeKyiTwkkbX2+Fr9aZKShHTVwQLlp4rz4w1cI9ja+2mTPKKQrvpcAsguYRbo8QI+k4cnN3dBR2ZLyzMmcaB/LhYLqwIZm8n4jTnSxgWyHGFXMXioA1cpkkV9wktSuKJRtorHOO9+Xlbi9nimi5BjiaIl1ZQLaxZeXOapfPA9bVl1HbtPkCrRVqDQvXdUnvNnH5ujlMHt33hV2XlmQZmut0A3UoClKwBDalSEzW8NyHGHTt13maFyS2X2K2vrIqLjjvd8PNfRE49niCXkdjoyfVrbnsmEuSzfNlaZzb7fWwV2/26gmbbk5WRzJUZBmz1xHZuc3UUogBvvLu+wx/JnA8DMytkseBpo1nD2een85cppm5CUX7CtTnti4ntisG9CYO8FKMd23Gtq4OW1sWxGspxVeg0rk/LNeKRZeWTXK6Kd30J/uKsrI2yyxUpwXlCcSUkJh9wMGB8Xiktcb5MjH7yMDzwwPzdKZOF87PnjGf7jmMIzfDAF4W3W23jzLByo4xRpPFi8Hl3bxEqQJVrxPUTQLbw5duN7h9qsO0H2O6wiVWdVBr3JpUuW+V91ulITzRihJ50MI77cJDm3neKl9pEycaE5FTyhQJDGngeMg2aev2yPjmHSknhuPIeHe0Y82ZlAfzVZVNrr6iWcIGgQ0uBP8xo2cX5Io1Mu9o7G6fxMRlsVaUci23S/cvEXtcTVfdVpdhqTh6+KOpIOLgSROkBUdvPfl83DS1sUVlZ+HX+sKsl+dZu/8FFr58wLi5IRqIFGNYdM67Pm5z5YFWq1EBr3ypXxC6FJfaDCCur0dX39nm/Vtv63SC4NeZ0iqtGqWglnkZELHb62GfPpEVIYWA+oU/iDWMsLnYL6U7WJxXPcFVYJor96cztVbee3bP3XvPUG08nC6maFCNZ1r6kJ8Gqt6BLatzS2t4lGbx9E4U71NQJCDBEu4QdROY+3AGC/YsyedyBMtP3fy21dFcOE/brwcjxQcPkjElb/Jy3VgvYYZeyVxujgJtvsnWqQuf+o+122dln3uJ8iVt7X62gBE88LTeALnhwy3vuQo4q99uKo4fvP/YHvtNR5Cw87gAZxr3rSIivK8VWuGklWcCZ4FTFKaUKDRaToTDQIqRfBgYDpmYE8OQGJI3TKZEWoaHhGVh2v1oc4BLAtp1n0MIRj9wXeqVknGd2IYQrhpsrhj3e0K728faSifowM/6lF4tErVXOnkUa6T/jsc4WePrEh+u+/Ufn5vbiZN90M7ymsW3131dhjO0BsHR483wg9CnBjrvtm3ep8sxcXWtuU6wNwg16xL6o3zqat2N0xlqpZY+4nanFnxR7LOKp59+RK0Ix8OBMF8o9QKtIqJLKd+6gKOv7jYTgUQQiVSU9x/OTNOJFIX37+/5ytfeBeBHv/oOD2frfryUylRtvGarYqtPEWKy8a19NnxMzQPTynOTGNeQnExFAdYSh8ga+OzmygOsqfiV9JDfW1DatvJymlopJQQbaCBiI3mHcSSEQD7ckIejXTySzXNvEqgxoC0QNFBSpNWIOlVCQ6SJUGplKvOn/VPtxhcnAX1VU6C2ZpOtysQ0TzStpCEwaLLRkgVT12DNU0GcWiDL48sGOwip17+/qKq+HT+iLs+jmCKCKjyI8iOtcKlnjiTenxs3LTHReBYaUxBKyszjgRptbOzBKQTHw5Gnd3fkbGObb29uDXkeB+IwICHQxEZeKp48910VVvWBFKxpJQgxWZUEv6aEkAgxkFIm+cIzpezXm1XSLPTrxm67vaSpD7W5SuIWdQBYF4+ynLi972LzrN2XNYndTpmkT7CDpRlsec32tZvH1/N4k4T2rn+ttNl6N0ShRRt5W+bZpowB8zwxTZM9NhutQFvzASO2Xd1q4La2iau6UTf5kO/Nj11FFsqDcXc7bXBmupy4XM5GcZgnqk832+31sE8/2UuEnDNNKzEEk8EBHzXZQOKKXOi6SlQCKkYzeLicebg/Iyjn6cKzh3tEhIfziWk2eatprpxnn2ZS7dwVEXJTUtIFNVF8QILEpTFqWdUF49H2/Q6yOnoIPXntt46C+gVHOjyqa1Ksxm3qK9M+fQVYhOSDCClFxmzB8HA4MB6OWIHJpn9VsBKNI9shWBm4dU3ZYOWlql26a7fdPt76RLnaKqUWlEZMQmqBFrDSnrQ1kVWcOnqN5hgyJH6+v8wnrzjlEnp9YVh92xeB97RQFYbauAgcqCa1FaEIyBAJNwckR8JxJD+5IefEzThyd3tkiIlhGLgZR+Oz5oSkDCIUYKaXK1feby9B9ppkR29jNK5fn9Ynfk2I8XpMdEdrl9n0YWfH7vZJbasewFIxWZ9WT1wxBLQnsQs9z0xgAwq5j6ku5yX9vsh6vm4T2avk9QUJpMdwUUWrNU/RqQPVR1o7HxURai0uu1XWJNb3e7keqK4KBgsn1i4v+PXno7+51X23cVjVmqT7mGqjOMz+3b7s32W3L7q9umqBdzOGvojcPCayymb0E3bhAXUKgHNda1Om2aZgldI8kdzeVt27/tm9vNGngXSuzpXz+sWgrwIBkI1I9ManFx6SrCM67TMsMNYu6twPo5MlRF3+Q4hiJdwYAilGI8XHyJAiQ44oQmjGGUSaUQs8hTbxaSPOS3D+ksuJFf9udvugvS5o60uZqmu9+rkZbHxkjMH0Vr022dqWR4c3QwHIqhICvqLbJLPdhT7s43s9X8QWrKE3jxjvLkpAUqTFYBWHHGkhQRByiqbhOiTi4UDIkWEcOAwDKUeGnEgxEKMt/iTIMtGH5Yh1kda6imPCJqCvvr8+v+XMXisYXF1LWH+ux7xHzN0+2rbC/x94js4sWGqF7j+dIrNSdJZ6Rz83wcazSlctEK8ybpLWD/k8q8xvlGqXJrRmcnhdAcCT0yBCqw2oa8PWR3BxjYXgIpxXfSnLh9vdZWeufbbLWfbvZvs4PTFuZWn00lb76psXXAF2+ya2T53I2kCC2UndNuxAUaI2aNU4eajRCsSCy1o5cQeNCWEAlEtpzM9NtN3GahoRvHbt1uWnJaHbMbSlWJL3eBW6IChAcqc3sfO0cOjWsX1rELMmmeSJscuJKMzzjE6TN4s1esogYsgqIow5MgyZGCO3h5Gb44GYIje3R443R1SFqdqEs7lW5ouYzm6r1HliOp/RVk2e63gEbcxTYdbLp/1T7fYamYE5lUZFohIixCyMx0SqQquROvdKgi6Nj03Xcc0K1CWobZoXr5JZWWh9/uy6D0slJHoiyMIFTyJITrQQqDFRxwM1mrrA3XggxETKifHG/CYNgcMxE2NgyIHjkIghkFMgZ280CX5tEF28ssGSSPfEQLwxZdkZuFpMS+iNcV22a72GxEeoLJ2w+OPI9gXdj28rtX4AEV2oA76IAhYljTVh9cSUtXfjCmUNnRdu78X537Ld5uN+Dum0ubVJeeXPK1JmS2Rrpc0TWishVXSjElTKjHos1romtj3RRBu1WCK6Dk+6tg9LN7fJ8UKRULWKicflabpQpzOlzJT5Qq2zJ+Z7Evu62SshsqZH1xD1jkZAaARPYGVhzOkaN7b/hQDYeMpSK222oqCEXi6xIQTW5LVBjx6tbq+mgGzQk5XPZkEshoCpH6grKHdE1YNt7MEsknJeEtlaO7m9EjxhX4kI28MRUgzkZFqxQ7ZbSonDkDmOg12z5uIXIVsAWIdooxUjrIPaEIYg1gVaJ9rOkd3tpWwdL2ltx0qIQhoioYLWQA3VJ9BBq17ZgKUK0oCwNHqwUGtQr0Rs6oBXiE8Pzv3+ksgKHmuJIUCONvUvJdo40FIipoHxcEuOmZwTx5vR/TAwjIEYIUVhyGHxs+CLz6VLG64S2aUVvC9Ur5DVdX/73a4w8rghpvPuF/799Vu/ULYnu5+PdQ1lY7bIWlqEdTF1VQUIH7zPdVXhcdVg3Q4foA8s+7Hu0BpTnfJQm5fjWyPUasNJalk4sgpIKbbQ7ZXDtpnG1SkF60cs9LuP/nK4WhR+cCACC11iu/FaZuZpolZXKWj7GPfX1V4JkTVZK3XExTrwr0bGqemvXkH9HZXFErjOlrlaQ23oA1dVAk9iXyiNs0lkHws/BwGqBacYA6qNWg15UWysbYgRYkQahNCLk+u+62Yn+jUi+AhedO007Rcqy8X9tdjqVFvxharNnNfW5UKKy4XYKMB+dEHkSsbom81+wk+AH/mRFz/37d8OX/7yN3Z/vlnMSu5GJ0gpIqKgkRYFrWoLOkdjmyt4NL91SRzTXPUNLoisXFEL1hL9Bp3dJokdKfKKRU9kxxxJMTLEyPEwMKZMjplxzGQfFpL9NTF50hpt4hHBmsia30SgLsmrbCZ2cU0L4DqR/WBie50QbMLxhsb0gjKqbL6n3Xb7MJMXJJ4KnXy+XUSxOSf7P5bfuE78PpAkduRVFhTXBh5c83NxJLbpSg+orrsurVFLQTxO1Y7IqkJKBFVizvaaEK54sctwoCV267qfj3e1h1fRDxzGFTAFS9+LNhufXZ2ja/zczef1cPnNGzZ3e4G90ojaUorRCDAeHgELMArQoBUWRES2Ses6PEA6ytrxW59Y0n3A/EMWJHb59E2ppN8WTo3Iwp01jo4iPmYvxsA4DMZhTYlhPBhymzNDEEJMNBVUq++zKRLY51WC90JL6CtrFtTWmryUaOPcicIyqJc20+ZgxzQX2tyo80y5nJjOD8xlps4z6uWnEGTpjC7e+PXNZh+WxH7cc7t9uPX8MebIeMhIqLQaaVm8KVHRqkuVwxDZnsh2D93ER+nhswfRFZFcAuw2GeyThXyh51VRF1I3bus4ZFKK5BC5Gw+M0ag4h2QyWzF4k2SIxGgjaCV4EhvtY1pMTNE+tygUR5YqLOoLQVz+TuxzoycKXWdXZKMbu1U72dIk1Ljx4sopdVMeDn2owm67fYwJNmpdnMKy5Wl3mlvozkJHU9dktG9jMX30iwKuK9uT404bsHDcE1auuK11g6j2eKmtQZkdcGm02e7HPFCa2sh4hDTe0BqUafak0qqKrdQ1Ef2QRZ4s9UxdaQ9s83LLF7pKQXR9z+YasaXMTJczl/MZa26t6/fSFRx4MSf5G20vUwXZKyWvZq+AyILWCmwRREvebOnkHYVb97s6qzfOtnilbDg7/hlbRHb5bL36uX18i9ZeOWabQZuVNrWRYrQkN0b7/CDeZNWlQbYobF/xrZQCcfSp72OnUXi8XZq4+k9atRG0io+ibat4c7EhEc3J6kZdcOmwzUr+x5PA8+54P47Nk8iYIqlFNECTtAwFsTXdNbVAPZG9OsU6KiQ98G4S1+Ul101QRgu6YhYQBHIIjhJHxtES1yEmboeRISZLXlMiisnX5ZiWhi5rerQEtUbbL3Vk1paXHZVdaiYLOtwpAkEe6cBK2Ozv5sZymAtdYZ3sJcv9EMLqq19UnsEr2DfK/79prjO+qLK7H6QK9PsrHeD6nJLNf/ZMT17F7240dzZ0AfDzuK2obKt1OY/r5n4ps8dLp/k5R7aWYqoFCsRMVKXMxRHZaNvozV+dd7+5kLy44Uyvfl7F96t6iF9DAJW1IboWS2bXwQebz3vB97fbN7e9+mSvT2Cf36n14ycBfBV7HQPmbt9g+7BT7Bt16n3o5+zn/hfNvmmS0M/LPq/r/TdHuHzt7Bvpbz/efFs+LconIl8B/vRnuzu7vYT9ZFX91s9qY/vf8XO3/e/5zWH73/Gby/a/5xff9r/hN5d96N/zUyeyu+2222677bbbbrvt9nnaN18H0W677bbbbrvttttur4Xtiexuu+2222677bbbbl9I+7FJZEW+C5E/+iHP/WuI/EUvsY1fgsiv+wSf+ZMQ+Q8Q+a8Q+S8R+S5//D9B5A/57YcQ+a3++N+KyH/hz7/tj/1URP6tj/gMQeR3I/L0pfdrfe8vQOSv+JjX/AxEfv0n3vZuu30CE+FPifAtL3j8bxbhw1n8L7/9/5cI3y/CfyHCvypC9Mf/LRH+kN/+lAh/yB//+SL8YRG+T4S/wB97U4T/QOTDr1Ei/GYRfspHPP+9IvycFzz+c0T4NZ/y2H6nCF/6NO/dbbePsh9rv9xs77eJ8IH4LMI/7NLl3+K//63uw/+JCG/7Yz9VhA+NkS7C8LtF+NAY+VkfpwiDCP+xyDe2eX23Hz/2jUdkVf8+VP/LDzwuEl9xy/868KtR/W8DPxf4Uf+870b1Z6L6M4H/DPgt/vp/EPjLgP8z8Hf5Y/8M8Cs/4jP+BuD7UX3/U+zfLwA+OpFV/SPAn4vIT/oU299tt1cyVX6bKp9Fy+nfpspfCvzFwLcCv9i3/7er8jNV+ZnAv83qi/8w5lv/S+CX+WO/EvjnVF8sBinCTweiKj/wSXdOle9T5X/xSd/n9huAf+BTvne33T6xfYZ+iQh/C/D8BY//ROCvBX5w8/CnjpGqfOIY+WmPU5UJ+F3A3/5J37vbN4f9WCayCZF/wxHS34zIDQAi34vIz/H7zxH5lxD5fuDnIfJLEfnjiPw+4OcvWxL5xYj8UUS+H5H/+AOfZAhvQvU/BED1OaoPj17zFPiFwG/1RxowAjfAjMh3A19G9U98xDH9j4B/Z7PNvweRP+z79Rv8sb8Jkd+LyB9E5Hci8u2ODv8y4B9yZPi7P+KY/l3g7/iIfdhtt5cyEW5F+B2Ojv5RkasL/T8own8uwh8R4S/01/8SEX6d3//1jqZ+nwh/XIT/vj/+00X4fY6q/uGOoG5tE8QSMPBI0MdlaP824P/mD82YH94Aswg/FfiJqnzvRxze4osiRN/fP+rH8w9tXveLfX//uAjf7a//BSL8dr//q0T4DSL8ZyL8CRH+fn/8Oxzl+UO+3e/27f024O/8iP3abbePtM/LL0W4A/5XWDL62P4PwD/Kta9exUj3gS+r8lIx8ht8nL/VP3u319B+LKH4/xbwP0H1P0Xk/4KhGP/io9fcAr8X1X8Yke8A/k3gZwPvAb8H+IP+un8C+OtQ/bOIvPmCz/ppwNcR+S3Anwf8TuBXoLodvvw/AH7XBk395/11PwT8j4HfxMcnkD8f+J8CIPLTsZXpX4HqO4i85a/5fwN/uU9n+PuAf9SP718FnqP6L/r7/8iHHNP3Ab8C+N99zL7sttvH2V8P/JAqfyOACG9snntHlZ8lwj8A/CPA3/eC938XVt34qcDvEeHPxxZk/7Iq/4YIA/DCSooI/76/9/8J/OZHT3838CObgPjPYxWVE/B3Y9eJj0J9wHyxJ8I/E/hzVPmL/bPf3LwuqfJzRfgbgH8S+KtfsK2/BPjLsevRHxThd2DJ6r+vyj8rRo24AVDlXRFGEd5W5asfs4+77fYi+7z88p8G/iXgCuQR4RcBf1aV75dr6dpXi5Hf2OP8oxh6vNtraD+WiOx/g+p/6vd/I/BXvuA1FSsxAvx3gO9F9SuoTnDFw/lPgV+PyN/Pix00YcHxH8FO5p8C/JJHr/k7WQMfqP6HqP5sVP8m4BcB/x7w0xw9/p4FQb62t1B95vd/IfCbUH3Ht/c1f/zPBf59T1R/OfDTX7CdjzqmHwW+80Pes9tun8T+CPDXiPC/FeG7VXlv81wv6/8BLGC8yP7vqjRPOH8A+Asxes4/LsL/GvjJqpxe9EZV/jrgOzBE5xc+evrKF1X5Q6r85ar8VZjv/jDGSP+3RPiNInz7Cz7iO4Cv+P0fAH6KCL9WhL8ersqaL3Oc/44qJ1XewRbQPxf4/cAvFeFXAT9DlWeb1+8+utur2DfcL0X4mcBPVeX/8ejxG+Afx8CiK1PlP1TlZ6tyFSPFuOnf4+99bG9tfOUbdpxq06knEZ58yLZ2+ya2H8tE9rFA7YsEa8+PUNMP2ZL+Mgyh+YnAH6A3Z632Z4A/hOoPoFqwMsPPWp4V+RYsOP2OD2zbEtZfAvyfgH8K+HsxVPVFZYqCyMd9Z78W+HWo/gxsZXr4hMd0gBcnB7vt9klMlT+O+cEfAf4ZkatgdfGflQ+vzHzAh1X5N4G/GTtH/z2RDySp288/Y2XGX9QfE2vI+Fvggw0jTjn4lRhy9E9ipc7vgRfyWU+4b6nyLvCXAt+LITb/2uZ1n/Y4/2Pgvwv8WeDXi/D3bJ7ffXS3T22fk1/+PODniPCnsPj200T4Xgzt/POA7/fn/lzgPxfhJ/Q3esL6S3jJGCneoPk5HOcInD9kW7t9E9uPZSL7kxD5eX7/78JO/I+y3wv89xB5G5GMN4gApiag+ntR/ScwFOYnPnrv7wfeRKRPffiFwLah7H8I/HZUX3SS/3Lg16A6A0fMeRq8cLX5x2Dpkv7dwC9mVTzo1II3sOAH5vDdnsFmtfjhx/TT4IMdpbvt9klNhO8EHlT5jcCvZru4ezn7xSIE56z+FOCPiakE/IAqvwZLUv+SR595J8J3+P0E/I3Af715yV8N/Neq/JkXfN7fA/x7qnwN87/Gh/vifwX8+f453wIEVf5tLBH+pMf5i0Q4iHVm/wLg94vwkzH6w/dgifHP8s8S4CcAf+oTfsZuuwGfj1+q8q+o8p2qfBdWHf3jqvwCVf6IKt+mynf5c38G+FmqfHnz9l8O/BpVPlGM/EYep/vuO76Pu71m9mPJkf1jwP/c+bH/JfCvfOSrVX8YkV+FlQ6+DibN4/arEfkLsIHrvwv4/kfvrYj8I8DvQkSwcsX3bF7xd8ALuiFFvhP4uaj+U/7Ir8WS4q9jnNrH9juwQPcnUf0vEPlngf8IkYrxeX8J8KuA34TIu1iy++f5e/9d4Dcj8ouwbtB/6EOO6a/iRcjxbrt9cvsZwK8WoWENVf+zT/j+HwR+H/AU+GWqnEX424C/W4QZ+DLwzz16zy3w20QYsYXy7wH+1c3zfwdbio/bBvX5a/2h/z1WypxYO6a31n3xdwJ/DvB/lVWq6x/7REcJf9j381uAf1qVHxLh7wV+uR/nc1gQ2Z8N/H9UKZ/wM3bbrdvn4ZefyjwZ/bmqfPIY+Y09zj1uvsa2j6j9JGYNaf86qn/Nj9H2R+A/Av5Kp0jsttvnYiL8euC3q36gUevHhYlwxJLPn+/8uE+7nV8FPFf9QCPqh73+XwZ+myq/69N+5m67fVr7AvjldwD/uiqvFCM/6XGK8FuAX+F0ht1eM9sne30SU/1h4Hv4NAMRXs5+Eqa2sCexu+32EeZNHv8khsZ+I+2P7knsbru92FT5YeB75CMGInzW5uoFv3VPYl9f2xHZ3Xbbbbfddtttt92+kLYjsrvttttuu+222267fSFtT2R322233XbbbbfddvtC2p7I7rbbbrvttttuu+32hbQ9kd1tt91222233Xbb7Qtpn1pH9smTO3377bdBAiEkVwoXbFaz/RQJCKaerKrrz9Z48aAvQExYdf1VXBr2+vHrXz7EPuQj9EOekMcf/hImj+5Y79yHN9Atx7IcU9/Co5/9V22oFlQbAH/8T/zAO6raBz+8sn3Lt3yLftd3fddntbndPqH9gT/wBz7Tv+cwJD0cRjt/JICYz7XWUN3cR0FBm52rEoQQ3NdEIATYDF7fzmBfTnFVtCm1FFqp6wvFzu+YAiHaNeDarWTZUPcUkf75ti8xBiQ89o1rv+o+pA2a2sZaa8uxbj8u50ROyY7DD6A1ZbpMTLNpqMcYCTEs2xY7RGqty/fkFza7LgVZ/PfZu892v/wmss/aL9+4GfXbnt5QVSmt0VSptXGZC62pn1bmKSkGcooEPwcXtMkceOM3fk6GNUZuTZuyNHOLbHxYlner6nVs7u9ljVMidi1QVZqu76mt9YCHqrqfy0fGUUX9EHTd/mZ/+qu2rt59UUQIYtcFVait0twv2wua1h+mmamUTxjRP9x2n/x87aN88lMnsm+//Tb/m1/5jxGHW4bbbyWkA1EiOUREAjklxmEghEBpylybBY7pxOV0T2sVoYG0TTJ4fXILkEIiSkQEkkRi8EAT1ter/9umuxaw7US33Hp7Pq/O153XAukaxJYUU7hOOnuA27i8vWfd5iY8L4l9jJEQAiJCyokYI4L4sfVQH/sHggfyWh6Y779Cne4B+Kv++r/lT3/av9mL7Lu+67v4vu/7vs9yk7t9AhORz/TveTiO/GU/7y9CUiIdjkiMXKaJh4cHSilcLhP3Dw+UWqEqWm2BNN4MHJ8cCSlAymgezMkCa3IrAsECWj0XyqVSp5n3f/Sr3L/zLgikYyaOiTwk3vyWJ9w+ORJEiH5ThVoErWLjgVpDG+QxcnM3kIfAcMjcPBlJQ7r2OV8EqyohRmKMIMI8KfPZEvTT/Yn75889WQ8oQkqRP+c73uY7v+NtUgy0Umm1cjld+JN/8k/zgz/4Q4gIt09vOd7eEERIMRIlUmvldH/icp6QGInjgRAzEgI5JL9mCL/nN/3u3S+/ieyz9stve+OW/+Mv/Wt4fr7wo++/x2maeP/hwpfffc55KqQUGYdMCIE3bg5865MbhhRJDYbmoa42tFRLHgXUw0aIAUme7noO2VSZppl5ni0Z9vgDeJ4oNFXmWinNfOdcK6U1FKgEmgghJNJwRGJirsppLpSqXOaJ5w8nSi1oVXS2pDZLZIjJEs7l6C0++/KZ0vrnKBLCMvW9+cLYDsP+BRGGlMjRYubxcGDMA3MtPHt44HSZ/Dga1cGeHn6/97/+/36Wf8LdJz9n+yif/NSJrCIoCSXTZIRwRCRSYyZIoIWEhgENAYd+EFE76cpEqwIUg1PQBQTabh2EKo0olgCWANERnxANuVHMCZu2JWEVEVprFA9Y4o/1oNiRJ0NbLJkVsRsiG8jJk9fgK+MQ7GLgq1NzOuWDi0FHoTxxDUEYZCR6AIxxIKVkx7FZKavYUSNiF6YgIA1NI63tk/d2ewkTIQyJMAykm5GQMhoDpRSCBGr3ByDkQBptIXV8cuT2zTtCCswqzCq0ZTHnJ+jm3C/KkhC2Ws3/RAgpkMfEMGYOx4Hb2wGRQP/XqjIJVFF/j6KhEnIkHyPDmIgposKKrDbzSdVGq4b8hqi0bFWfVh1B0mAL2NpotRliGgLBF9bHcSTFQAkTtTTqDNoq86UjsgltlrjnmIghGOI8N9t7iaQ0EIexX9Ko9TMDfHb7ZjYJhOGAOAJbSyHHwLe+eUdtVjE4jpkYA2OK3OZIFJACYXZwJARIlr3GFHwhB32Ss1VZ1PNcJQUDRKAXShzyaQoO8GSEFALVyw8zllCW2qgKEoVQZqQp0hrMBVpDaiWi5qdA9SS1z5QGrtHkxdQSVEBVqHOlemzrCLKIkFMipUQQIXeEOgTfnhKAFOxxVSWEQF0CsSG6L0Kpd/vmtFccURtQIkoCSahE9OpntgDjSaKd6ommQmuCqiyrp7AuKIFt6cF+6yVBdYRUl+qFUqsupQXx8mRrUKrSql0EDA3teaoll60pda2OeDLrqOoCqirSLPENqsS+d2qlyaUKsiSzKwobtCfMAdUIko1uIRkJedkf8zp8Na2WwIZopdUYbTEQ4qv9qXZ7PUxAYrBbSkiKhBQNkanNy4T2uhCEmCIhCGlI5DETkiWGpeijILSistppBdoWhLRXRCQIIRqlIKVAyhEhEIiImm+E0mjuawA0hQAhiu+r+MMr/UGbJbIdQUaE0BQVteuIm+oaEFFbUgpCDIGUIikK2uxm1AVLfBUs2Z+spEtUml+UtHWEKBCCIcG60Bh2He7dPt6MMhMghKUSGINwHAZUhCEnjoeBGANJIIv7X7BksQc8O2eFEBMxW/hWLUsCu1IO3NeCbPZCEV8Q0qkA0pNhr5iIUNXRUbXkVVpDpNpPrZbIakO6zz/KFx3euaqQbp9c652WULda3V89CReLe33bQWS5iQVo/07Nr5taFiILNWKbO+z2OtirJbLaua5eohddTmKVhlKdAwAhmZelITAcR2qNUANUEFWiNGJfy7UV5TTuXrAVbUpINJ5bjEKMgCqxhYVD2lHQ1pqd5G3l7pljBC/zy4LeqPn9QiXoSalqvwCZR3REVrws01pzpGh9T9+IOaC/PgSGIZNTNEQ2BPr1RRx9Xh1QsXWsX4zYA+VuL28igqRESGlJKEOMxJRAlTRHUgyoCilHxjETYmAYMzF7woshLuYXG86bl02CB0KtlsiKNgKdEWPJcYwrP7cvS0UFbeZztTZUK7UVWiu0GtZg1qAVQ1RbUcpUN9UP84ekQoiKhDVoE8R5svYZUXRZLAYRol9LQghLYNTGksi2Ai35Ncc5tYgg2UNvSsScCTHSSnNU91NPx93tNTLFKm4SrDqQk51HIWUIwXmxwRaX2gjak05FW7VQ27FMWSOFc24WP+tcW1WMhrdN7jriIkJb6HUNVBBVB2r88Y3LR7+vAtlh1lYb7jy0anzZ1tR8Rzz5FnGAqtMEbNtBlRgiii4xclsBFSDnvCCyPV4KoK3RBF9Adl7u8g0/Tpt3e03sUyeyloBZEFMajYYQNlyYhkql0SBaEIkCGjKabmitIfMFpohoI9VCLjOitrXamjlsCJCTQbbDgObByw2QonOHqJsVZkeNGrXUtcnFHSVIcBQqsFnCcr2WZf2/Z7d+gQi+zGvtOpHV6zcvnNtOi8g5k7Jx/kRkKZFAo19W1mVA82/YX7N7524vayKEg1EKQjJeWUuJYcjUYBSDnBOCMh4yN3cHQyqPI3kcbKFFgdKg4TSc4H5l1ILWEdkyo6U4bciS2JQCQ46GxArU5kkuoBpotVJLpcyWwJZ6oWkhJaGVtjRuaa0IME+Fy/2FWo26IMH8e2xYUtlZD1F8Ia1WriyVENZAGKMlCzEGr3AEoyk1pRS7RtTUqFGRJKSQGceDfR+evCKCxohKYK4z9TIznabP72+92xfIhCYBiYlxPIAIwzBwvDkSU8Saeqv9LBWdJ2+oqtQ6W1UgGBCCGNumiXil0xI8wfiyMYRNldF+Nm/MUgWpa/LXvPFMFJIVRohiaHFTW9RFw5EIqtRgyW6ThtQZrZbIzsUrlFEQ8Ua1aBz14Gh0b44Up9OpQkuN6mBT9AVmj7PbnpUeL7UVqopXa5onsobGilrc7MntHjZfH3tFagFsUZKl6WpZMdrKLIhxbcACThwSoTWECq0gTcjaGHqSiLVp9ARRgyFBGgOaLECG6A6Gr+KM3EZXTmhqyKk6eb052hNCsJVej4C6AWEfeb+84P7aKd07wXVFZZcX2muj8/PEy5rR0VycTrCuJHvaukK7a2nmOrHebbePNBEkRqMWeNIXghBCRIN6oBOaKwOkHC2RTdax3ysa5kcd7dyoGThmIkvJv/WPtUDaqQXuX0tHNCxN1601Lyc2R5vq6kuwKgRgyGzxxFSCH5sIdVO1uXIU565r2y5SWY7BkNiFJeFJgK4N4V5kEomEYAh1ypmYEwpUPMAjhizviOxuL2lW4LDG3xQN8T+MAzlHW+BVq0bU1qjLO3ShIqCGXK4VEo8i26oHLIli0E5rU6s+LDvhxQt3SvEYZmCrXO2riLUgB5QqWBKLLvFKW/NYaIoGTT0xdi59h3UXxYGNH4LBOD2mJk/CtyoFjxtQtg3ats8dlV3j6VbpaLfXwz6DZq+ELqe6cVdVjGbQANROsbDUKgylVGFBSkSUeDoTn78PtaJzgalYAnh7S9BbQ2TrBZ2GhRvT5UmieOLLeqIHi5i2EkVIPUeNgThYORVV5IrjtiayupEfku6PrGJZ1cOaAcIVqb4azAPkwe5LhJSc09vdrCeqRhvoHL71YqSIBqxxJSD2hXaBht12+0gTEfI40puyqvPPJAYi1kiVckJFyWNmOAzEFJGczH+9XikiiLqfhbgksT3ArSVPR4JkI0fVQKtSpsYUCgElYY0atVbqXGiloDhlwBGh6NeGjsYEEaQJdWiUUDt3wQJhjM6Hl005RKHTCfy5hQ4ka0IeQiA613UYBg43R1Th5vaWw/FIyonbJ7ccbwyRDdm4xrU16lRozSgRtc7Uujdh7vYy1huZAjlbA+QwJFJXs0GBaKBNa2hHVaXDLLrK0knw0NT54izJXwjeBKZKa8Epfl7ZVPVglowD3pQghepgjyzVxUCKnikLViFsjahKEpNQSCiiFVGLsbU2SgNUCFJoatz4WI1S0SumKsbF7U0xIQhBjA+botGeNtBS/+acGWEUBm3NkuYNRz2GYP0rGzBrT2ZfH3sFRDbQGPC+R2yttiayTaCqJXCiwTug7QTNJjlAboUhB6Q0uNzD134YnWfaeYbzjITI8NabHNobSAwQMoglzU0STRIByMFWjarKXGeKlyVFdSlrRg222kyRcBwQL+dI3RLf7dZCpObo+w7B5U/Mh80p59TQZLSKOM2EeUIIyO0bhJDRCDUm6jDaHmh1ysDa1QkgjhyvjqvQAhIiaEQ0Elqg6e6Uu328hRAZb25ptTBPk8ncKVa+jJFcLXkNRRhvRsa7AykligpFxZuqul4jxGDBFsQ6kjshvDVaqU4BUGJc+eOiClWZz4VWmtF5gCimcDBfJuo8O7ZTASUoRCJZjA4xpOzUnESrQinVOqTFrjG93C+uLIBTnMRXnRYre9LaebF2i8EWmEPKHG+O3D29AwJPnj7l5uaWlBJ3T285Hg8QhJYMwZ7nwqU+p8yF0mbmemEup8/zz73bF8QUqNqQEDgcj6gqKQo5e79GgBDsPJ5bo4VoPZC91CFqogUprA2barit0JbYFePayW8Vj15Jsb0w1NVurTWmSailWKXEE1lbKCbEVU7meTJlEmBACaIkKkEraKW1yjRXSlU0KdYU2ZAWiRjNoFdbVplLR3uDVYMETKkgrNz63vfSvDrbtFFrodZiPSre5B0kkHO2BB6uFq27vR72itSCSNdq7Kut7nS+hlwL5NoL5LKktMabFSSAtgLTBZ0m5GKJLCEQpiOxTEgLiDSQSBcOUdegDS7L1VpDygS1O7jzYhGimm6rtGY0B/VuzE0iGzqxJjZjrQbrjA5ePQwNxOV2glbEg7CUiTBdbKVcC8G7sBuyqg08EmlvjsauJmsuK47CXn1vu+32EiZCTMmC2EL7cck5DMkMMRC0N4G5qkHD1PD6Ztj4qzdsdnW65WTUtuo+ygYdVZZGSsCVRiogrhvblv3blhp6dcIaQIyKE4PzfFmJN8aV3VRMPPAtNJ3NqlS2P1nvd9Q3pkTK2RbWeSAPAyknhmEgDxkNQnWpv4W3jycJftttt4819wmEJeEy7ulG5Ua8erkpwW8v/CK98uFnulNp+sKtn9fboTvbZG5BbWFpnoohOPqrhrCILsmmiKG41SuFwYGhKDhA5BQDTzqbqt+sgaxpozVbEPdGTtjel7Wa4/vedeJ1JRja77Lk7Su9oG9HdGnmXKpJsl7Ddvvmt8+AI8tyMou2hW9jpfFGa9ZgZd2TQlBoWCmlEqmSncidQZOhljEi42C8WEnU4nqT0wWKMXQqmSaenEZcz06ZW6GsZ7wVbFQIGoka0BTQS4IUjFZQfZ8xAr0ANQVqTrQQPJF1DlGTJaktg1IPzgq+XKgnS2SjjsRwa6LyaYBB1xWmkevwCowhXOIoL51Y4EsAWX8+up7tttuHWg8MUQJZhOacvCGNjoQUzlOGGWL25pEgK8+tI7IY/ScQVlqQIyPi3LpWTb95OxmolsJ8mRARWjYqg0ighd61bDSHFDN9nScC+XCgxUBBfZuVoMpMo8aweMY2GNa5oqErezikhAf2oIQozv9NRiWQaPSFPECKNAK3T255+qWngHA4HsgH0/KULHZ1FHU5PJAorpYijGPm6ZtPqMfDN/6PvNsX1NZlIlh1Y57rhiZgzcN14Zvaa2MMqNiQjpRsYambBaA07+r3BK7T1EKw83/5RC/8WZM25juxc2k9/izrMt+GwJCM8jA7ibwAhyFxczgQUqPJzMMsBiyF4JRCXZPtq6/ArhtzdZXqaLQFa8K24xZZP1+W3V7BphBkoUHQujwXV1XV3V4v+0wS2WsZ5A7KCtpM9L/JWiJRFb8pTTNWcBcCA6LZ9yq5JIFQJTDPDWqjPXuG3j9YmUESjWSk9WRKXiowSaNIn/BhztA0EFs0ma4glsy6VI8hstDxY4CaImVI5uhVCcW2FRqeyArlKJRbv6A8XODhDBIY2gjh1hQWhgMcjNvUV+Sq0KKRDAJCIywBfv2/631aIrvImey228eYCKQ+bjbYYnLImZvbW2JMSGic5ntkFtKYEZeEUzWeW60KzXhrqBB7IgvOU3MeXbNu41Ztep7igXmaadosiJZMTFaiDKkRQiVEYRgzKdnUvzyYVFcaEi0GLihBG3MzykITpbq/amuIa8RqU+OrSpddlhWlCRb0YgqkIZOHRIrZaBIhEmImRggp8+Zbb/L2pVqJMrr8WBTCGNCEJ8WOhkUTos+OYj85Hl4g+L7bbi8yz7p8HJeqOMdzdkrMOqK5NF2SWcSak1El50DOYUFuvT6BqBDadiplW9DWuOo8OoC7ajIHgdisgaw1RbVSO1nW43kMMOSEAEUboVWqKk0CTyUxNkXjheezoHM1JLZ1hNZltXpm6cltqxUtXq2JEXWecByMhyuOOq/IsvsfVsXVGFzn1lULgiWy2ybOntTu9nrYp09kO6q//L6unjq+uKxAteeKuqSLshQmrNihSwkzQkjY8AAWNBNVtFTaPNlFAF24uE2hRWiitNDW1Wp/nwbXpg3mxISFgK4dbV0xHysZilqyW5VWrBml2QgTAFoOaDHUijJbg5qEhTdIbcv0lB79N8wC+5xlVd0VGq5fsa5Md6fc7ZOZ+IW9ycoTjS4/FWIgtLDoI6/l8o1+cy/DA5vTkM0pe91BTD/NTV+2qaDRqxCh+cK1IVjibCNmg49rDoQ+zasjQxj6s6C9HpkWH23q3dEYBajHTLxjWrqmpV1XrDE0IARicB3qlEhDZhizNY04Oi2h1071iqKw8m4d8Y6RJHsqu9vLWU89OxfA5LUsIesydz1qaqeaPaIKbJM1D45WlQwfLKRb/Fx1CAQWH8PvEwx0Eq9kXG1E+8TNgLPxSI4apyjkJLQGKVVijMRqFR2WBsyVztDpPh0wbmqocPOqEbA0fq47tyYYokvavtmmrHSL7ffEmgTv9nrYKyGyK6FcF6h/YRbQE11dXtc7MI0zp8Rq00KCKhISjAdIlZCOpGxae8SChga1mNRPJ7LmEY2DIb4dkdUGpSCtXO2nEKgkwOS7dMi21FRs9B49gLpMdIrIkC2glYaW9RibX1zaGJCDk+7PE3KaQALhzS8hT29NveCQ0NCdsROblCiryHT08GpDzjzJ793YeIfqB64wu+32YtOmzNOE1rpM3ZrLzMPpgRAi0+WyjHcUhTYVmgh1Vhv12hbmmvlFM71YsIVdE1xyp7lsVVsqGeI8vxCNi5sPiXwwbdqUB0IypYDxMJqmspc+QxBr5vRAToNa/HrRpbSANlfqNNtn14pWI/UebgZu7o6WYObIeMy05g1uSxKwKhWkFMlDQAnGiR0HQ8fEaYJiTao9JqcYkRBNeeHYqMEqR6IstKTddvtIE/MJVTGkUqEqXOZCbYWcIqMrdSC22FyUblwDVnqT1KPtOsa7/N5jxfq6jdRjR5+0J8Zh0xxiTHRFKd5notEGOJg2rTA0MQmuKtxFYWgmdXkuymWulNnoEqpwM2SejKP1woRgjVwKcyyUyRL5FCIpxCUpbc31ojFUFu0N5HYkxoWNmPcGpyI4r96Pv2u37xHz9bFXkN/qZYqVeN2le+iryRU/oavPNS+V2wqrEloj1GZ0guMN2pRwvCMdn7gjX2jM0CoaA4zWmMHxFsaDjdQLSgsgtcL9iTC5JM5SZwjUlGghojnB7RFyMvqbWmmmoRS1RDamSBoGG2NbKjoVnyhigRxAckSyt6d0uTARwvGWcHOHxAhjXvi7qoISV54PXOHR9t34JUdcuktciXrRXthtt4+2po3pcrGM1ANgm2amqRj/WwsBIYeEtka5zI4MRVq1QCsS6FPNW5kos4v+p4CksGi/dvktvBxoJX6T/4k5Mt4OjLejJbXjSMzGP83jaNxZDDbuk4h6nG1NTYaPjgDb9aRMhfnhTKuVcpm4nM6gytO3nnC4GV3zNXG4OXhXeF6WgF2tIIXEkDPDkEAsqR4OA6U1838P9J1zH0VIPolJY+NIREeTHJrnmVp3HdndPt5ErALQVNBq53ttcJpm5jJzGAdSzp7AWRMizUcLuY+FDfq/LZ1Lh1oXHoyXJ3TlmVopcl109jeGEH07FQkFqlGESquOrtqEwJSTNXupRfKoAdHIrEJKGQ2Ry1y5XArn00Rryu048OR4JLlqSG/kupwunA1MdkWTuFSQrIHSkdeefi+JNouUHgoa2RZ9HcBdpTl3e33sM+LI9vK/LEjsNu9akdp1WEKHcPss595dbdllsEQwCKoRcLpASqiP9CNnGAyR1aCW65WApNnkvDxIem1keZ/mbDqvQ7L99UDZ58WrYojsmA0NDaVnovbDWeUSo2tagiGn3niWs0uFdeJ942rNLNelj+2FpbetXHWadmfeSbK7vaT1wCfueM3L/Sg0qpXd6X7pk+/a9jxc/+98WEUR705cuKg8ek8v8UUbihBTXMbepsHvh0DMrnXZF27YprR45UYflVb7h6iNxmylUXw6mLa2JpNiJdYQTYPzas5837cuwxWiSQQFGx4RxFRJmvPu1g7LzZQhFQgKQShbPsVuu72ESQjmQ46IKn1CpC56qB1RXQaQbErmPXaswbWffLIktWvxbhuA+1Pr0IQuLbmU7juS291xS/Hp2xRr1sYT0OQAS06RITmhvCl1DrSq5BjJMRgv3VFZBYpL9amuOuqdRqfO5unasX1f1sMO67XGH1suRcIj2sEeM18Xe+VEtiOzC/q6ILF69aqFcqAu2KVKmM6EZ/eEuaBfe4f21S+DNvTNmRaaJZ69kaw1l9+IqAakQpidMB6qjcybK+k0EXxspPhYSUt8scSzAWGC4oG+a0r3Dk8R4lyJ1UNsKaiXMyVHwuDbnMsy6u+qmFPvqZeTOVoekJwhRNp4QMeDJatXXJ4lZegZvo/bY52etAfL3V7SjEpQ3CctAS2XifPze1qp1gA1+ohm8Q5gbBjBEC0YLdrGPr2uTsUTWYU++rlU45c3XTVaU+Tm6YHD7UgaEjdfumF8YmoJKadF9zWmRAgW7ObJJhrVqXB+mKhzdfTUxssOw8jx5oYYIw+XxmlqzNPMfJmZzrNxZSvElMh56N8C6txBvPkreONXGrIjw5nUWCYsBac4LYNT8OaSmBjzgWEYKdPMaTozPZyZp5lnz58zXfYRtbt9vIkE0nBgnk3fuZTKVApFTW/d6ASRGJMvHk2uTmNAvNkrLJJzsiz4WIqOLlvVK3n0ip/5LJWlZ2OhFNiOmYsQaNlQ39oajWL7EIRZG+L708N6EMhBiAg6RLgdKE25DJFTDLTWGGJiTK6o4Dx2VaVGZY6maCTb3IH+Eda7sjwqa+XGJOgdcV6i7prIb5vmdnt97LMbUduRH1clAK5Xd70pxJs/RBvhfCZ+/T3CdKG+86Pwo3/GtkehZTXlAtwxHbFBMyBIEUIP1CbSQ5gL3F8IDxdbkXnHtMYARdEYLYElQnJaRPPJRzmTjiMSI6FVYp0QNTkhpslee0wQBnOsc4VTsQtMzoh3drZ2oTQLbiFnJGWT4nrzLUODfX90WXm3lWKgtuIVDB3q3x3X4Nduu32oqSrTPLu0mwWD0+nE++98lel85nBz4O6pif6nlKwjWQIpene0BFvcWQMyU/VpVjY7E1qwcZqlOsqrC+80DYm7N265e+uWOERu3rphvBttTvuj6VogzFNlmiplrkznC8/efcZ0mkgxcRwOxJDITyLHJyPjOFKeT5RL43KamS4z54fJrjsVUhrIQ/8suybVas0oBIg5kA4DOWdSHkgxU5uQcyZnS2QTYbl2dRnBFBOH8cg4jlwqvHeeef7+A+fTma++8zUe7h8+x7/2bl8UkyDk8UjRM1M9cZknplKoTX1wULDFlQ8zaLVX8WwYSQdGe9UTXeUcQ5IlyW0SLJnFKGvBOqq8ybk6daEPFmJJjKvEpdRfWmX2hasGmNVApEDnohqNaPSK5xASxyHRFKYpcx4StTZrItuiqcFygDIpc1SqKK3JMn1zU5yxCYDN9eAdmQ4heKeLo65bRYarhrg9kX3d7NUS2W3Nz390asGVcsBSQnTOjpc9pVakzMg8Lz+hQZmgWoOJQZNxLTX2JNk1YEER6aitURV8GLpxBANIs/KoiNh7SkOodjGo9lNCQKpPSKkNKRVaQ0qB2RJWKWLDFoJY89lcvDzjVAjBR+xOvcpiCenmc3QZFrH53nBaxoZnrM5x2iLZu+32cdY5pjYiWpegMJeZuRSyj6xdz79e2pMF3dmeabI4tCE76r7UufELPzYGQ3tzIo2JNES/WYBdg8uWNmOIb6uFWiplKpRpRhLUUJFo+rbLkATZRDq/dVxmGXIgwbjt6vqYOEC1KY+uCbUn18EbLj2hsKTBvoXtRDAQWm2UuTBPxVDhHZHd7aVsLXU3p8M0td6I3tzUX+evXjHHx1mZbl/n/2/O780Ttt1Hb99WEE29p/uOj3TGFD/EurKNAuHvDRaqNgwGWVFigBhoMVDFfLerAiG6KAWF7fvZXIt8w+o8XEtwDYEVcKAMVv6BrhSDq29vt9fNPnUiK+ATSHwCSHfFJfDZCdn9KgCoEFslt0KojXC6mC7s+UKYZuehgcwVeTiZxlzIaOj8m4C24E4wWY4bgMEUu9CKRqUNK6wprokXarPpYLUQSoUQWafzqKkZXE4QgyFNxeZI6zyj02Td2acBuQzG3Z0anK2BTfQIyTRly+lMOT1HMZ3KkDKSB8LxCeFmNqdMioY+Ts/n36rJg1V/TCUgoesJ6nIh2W23jzLrOC4+/FWX36s2qjY0CvE4ksfBO4aTJWniNB5dJ86pKCkIOUZaE+bWqJfZxkTOM7VURIQ8Zm7ujuRD4u7tW55+651prt4kQhZPHj321EadGq02zg8Tz955j4f3L8yXmfv37imXmTqO5BARUdfZrIYmRSEfRht3HX0+vSrjMPhxdAUEC7hxELKPwNShcW4XWjWaQZYMYsMicky0oKABUS+5moyJNcNoQJtQZuX++Zn3vv6cy8OZ+/fuOd3vI2p3exmza/k0F+4fHnh4eDA1j5QQbyjU5hXAZo1Z1ytKripzJgNnOu0xmcoBnpQ2x3+C2Oh2S1jDFQq75ab3eB19LDXAEJPnxWt8x9Fe8fvi4FKfxUVfIKdAa8plmrmUmdZsCIqkLi/W+puM4+6f3xV6tKlx4IupDwXXmRWv/nR+bU9wdUO7WEGfPaV9nezTI7IiayK7XV91FLF5cxQsZREBYlXyVAi1EB7O6LN72nQhzcU0GZ1/yv0DLQRaGmlx8Hcn0GhobplMaisK8S4jnpi21FzqVVG/FxSkCNJs5Rv04koF9k8tYtGyy4A0S2RpCvNMu1xQbYRjJlxGm15SQGf7HjSCHgZUoJ5PXN5/jqoSPZENw0h+ciI+mSAkpInrb9kVp69JHV9GRIzzi435q6rm7Lvt9jG2cGRRqo+NLKVQWqPSIAXSzUg+jiSNJKJP9rGGMANPXNNSlRQgp0itYojpPFGLIaelFKMVHBI3b9ww3mSevH3L02+7s3WnNJoPJ+m9JIpS60y5VC7PH3j/K1/n2VfvqaVwOU1GWbipHA4DIUJthaaFRkVSIB8OaLCZ7Nn5f4dhtKRcbNymcQ6VeAjEQzLu+9A4tQu1NQYdFg3NECI5JRuIoJa0dgQJNTQo0BPZxv2zE+997RnT+cLzrz9w3hPZ3V7CTKVAmaaZ5/f3PH/+nMPhyNOnB6O7RCuxV8X4qLoO6JGexNqWABYVABEb1BGSZ4Z1KYLixDx6Emo6tUJDFmCkD0hA1bSmfdFZ1BLsrqJje9OT4eCL01WjNrR1X2MMtKDUSbmfzeeCRku8rQ5D7xpr4lq6iMlwSkBbY66F2RPZuElg29IYvlZZTfVEXc1g252zx8zXxV6JWrB0VtJLIyvFYFkQLdwee6x3UveyYOsnYoyQsyV1MfkKMtCHJnROaSenL11SnXMnwaRag5U2tqZN6ECLyWD1vV3o8FflGydAeDmEJSFfiQBWJtHQ90NcIgTbz6U+20suPcnvpIFOwv+gqy1v0/5JvbK7O+VuL28LL90XluKTvkIfRpASUSPRE9lWm3VOdxfusyrFtBvVmy1EN4NPvCxvgw1MlSDmSExhKRHKpg7ZrwOtNupc7TYV4+B2pQXMlfsoWCvS2MhaK0t6M0eyJhihkVJYRdE7TWChO5iurVF92nJblRewa4j6IvdRadaz72VxXmuzhUHxqWZ1r5Xs9nJmk7z06txbmpNgTcJ01WX/qLL5Gn6vn1kizhKHdZGxWwkJ/bze+IGuW9gSHdoSxyx+Nlny0CVWLcgoFgOX3Vo2tCoxWGjvqgPqsw/6EJRHY4Gk5xnrQx+Kt3bKAXu8fN3sFagFQozJhqWH6KszBfVW/p799fvIMslj9mBTb4/M3/o2UgpDe4NSvg1Q5hCYg036qSHRerdmWBu/pJSFrxqOGRmMJF/qqu3YOUlBoVWxEbPY2E3nwDMHK9uLT/QRhNgatTVElTLPTNMFbWpjNA82KKGqOPIjhNsb4s2NfWYUxuFgxx6jDXrIiXZzy5yjS4uJLWrpA4RWHtJyfZB+EbHe8z1c7vYyJiKkIVNKYZ4LtVZCCByOR4Zh4PbJU56+8ZTDzZEcEqMMCMJ0vnB+ONFqZb5MTJfJNGIFxmN29KMSyZQq1ONAKwMxJ27fOHL31o0PQLBBBwuyi6zKIAh1Kjy8e+bh/RPnZ2fO9xfm80xMkbu7G2KK3Dy54e2f8BbjzYEUE3O9UE92PDFUxqgMNwOHu0wU4ebNWw6HRBgCISRSPKDSiEN0hYZISI3aLpTauExnHkiUWl34PVrwX6pLLFR8RSi1oa1wmWbOpzOn05lymSml7Yoiu72U1dp4/9k9l2lmSAO3N7cMw2CoqieUxccyR5YWr2URJUvi27fYJ+UJVrDv4I03D6v6GGmnC7VqzVsdFBJDZWtrVJ+o1QeoVG2U6kipv66H8lmrob0hkOPKd42oxywf/ACMw8DhYA1tcUikg6mKtBiZZSLUxtys18SuFz0KmgpJiEZpsKEp0YYhxLgktsE57rvt9ukTWYHoSWxHT4OXw5er+2ZG+9L7hVCCI6i3N0g0qkARYfTV3Vxmyjxb40rw7msxXpxN5FIbi+KlhZATIUWb9d4q1cYTLUEpYqhsUB9SKb18D3O0n4bG2D5GVYpfAGopzNPFZUMiLVvJpQahONITh4E4DIgKaTgwHJ/aBShYqVNjpN4eKCmu0xCCoVW5+f55yccvVwvq1afi7onsbi9jIkLKVirXyRomJAQOxxtElbsnT3jyxhscb48MceCQDgSE+2fPjYYwF6ZpZpovJqGTMsMhG+eOStJKKZbI1jqShsjd0yN3X7olDdFeG20RK83KGdYwZs7VLsr91888e+c5l4eJy/OJ+TKTYuTuyQ2HmwO3b9zy9nd8iePtkct55uHZiXKpSMEGFER4esy8dXsgp0i4PRAOCUmBNmR0lEVyK+Q+bUxpbaJo40ImtEhtjVKacWC9TmMoEUvlR4FSjN4zXWbO54nTw5nqi4TddnsZq63y7Nk9rVZyHogxkWK0RBargJRqo/XUpaoEWRoqYTktARYuqKJ+/rK8ZuGrehNlb9QM2rXQQSU4/QiKq4+0Wq3qoJ2WhHPtN59X+4jaBsmTS9lItkcbPIIIc24cDiNVlTQO5OMIAkUClwZSG2FWNBSjBHeuLDaUKIS88IdD/576cfYKzCNsdqEY7PZa2WdDLdhQDGApiHNVOPcl3VLGV4FOJ2g2/rKKo7ZqJTyjHqhN0xJMD7aXJzqh22kFSjC+jTgz1pNTcUQ4BC+lIIaKItQALdoK9QpY0d7EpdTQqDFa93QnnXsiW8P6+b0zmxiR7AcafMpKDI4my7rPXmtdmVDLl3r1lS10hh362e1lTLAmiM0wgD4IAN1M3vJxrTFG021NiZiTlQdjuKa0LGoafUSkkGJgyPaelI2HashJwIv7tGU+mAu+N2jFaAVlqrTSlv0L0XRehzGTR5PEijkSS13BJqccSOjT9+wzJfcqh/l5C4GuY7coNOh6TdJOMWgrP30Nfv06QZ+D4q8VQ69qtY7z9qgZZ7fdPso8rlm1LRA6La3zPTc0gqX8v9ACNinc0qW/6rZ3ekB/vJfseeH2PN6J9H7GzU1YC/sm9yN9qqSadmvz64H5gyFUIcgyvXJrqzKIa02HYMIEfl966ZHHruSVzq1kn1/PRLevXZUglu9gT2JfS3sFRFbIOUFKtBQtKWUNDD2ZXX86SiMgwUrsQSISB0MeW2P22e2XuXLBRmfqpsNSqy4TiPp4WRR0NkktRSnqJZfOPRLTZp1VbRytQPL9sLGwXdN1XfkaDFpdZ1apNjuWWCE5vNxEKCYxQJJICsbfkzy4NmfPvi3ASo4gm2S2R+eN321XnKvpsvLebbePMxEhHTINJZ2jISgofZZ5CokcB3KyWxoyQYTj3Q0SxLmfjfv371G18nmZz4aStEZOkRQDEu843h6IOfHWG095ene3TvPSiKIkb/YqpXJ6/sDlPHF6duL03omHZw9oU/KYSDly98YtX/q2L3H79JbxZuBwdyQfspUop5mYKlGEHEzH8ngcSXdHYgrIkJHRqh0tmtqB0pjPhVJmggjjzUi8sR5LpFGkWFCmEVwST4j9wmBcQKz0epkuzFPhfHrgfHrgcjp1iOpq4bnbbh9u6zXfYoNVFstkMpPJF4fBG65iFFf6adgUH+OfRs/k+qIKVvqZIER/nY3E9cEnlkUvQE6VSCHSRJkD1Bhs+In6loRlqEBT02JuiutH2/jZpjaiOggMMSApEkWgbShF2AhcUEJMhJRAhJgbcVQbqhJniu9/l9gzeoIpMaxJrOcA4ovqJbyKL7TXRHaf6vX62SsnshojLUUroWug9mWibG6wnHyKWCILNhrS+ypbKT5ysnIOM6c+Prav2hSqdmeDiE0MMaeuy2q3qXdeizWhhGDIalTno4oN+ZJeQxTxUbi4oyutKLW0ZY58X2nGBr2IY+ivbSiFRo5KDDCMA4w3jhpvvrCtlqXfWUpF13DwdWxUHr9gt90+1CSY8H9tzWRrxDhtEtS5bYkcMznabPc0pKUBLA8DtTROz0+EmKA05jpTJpO6u8mRQ4pIgEMcIAZijrzx9Cm3N7eG+kartlgqaexuaRPP7ivn98+cnp05vX/i9OxESpHxdiTnxO2bt7zxrW/y5M0npCEy3gzEbMoJ4zQQUyGlwDgkYhAOh4F0NxpnLsRlip8hwUJToV4qp/sLAeFJKITRCiOqzdUQ7H7no/cEQwENQqtQtTFNM5fzmdPpgfP5zPl8JoiQyAvmvNtuH2/BQ06wql9r1DKbIk7OVhUJgRSEGDog69VLbLhBF7AqTSmuCR21Q0XefOkAToyBlMSrCq5OgFBbYNKABqUEoWnwBNGoDRICIdvo5tqM32sVwZnCbNSDZnJ+QRXNiRSixbhm2ukWwwISojHpYrJrSk9ki9KkQjwv1LklwfcKUohxOabFtqFw0wSmbJrWdnvt7DOhFvRVU1OTuLLzaSloLBjt9Zs9lXu0cHqM4/aE7/Epaifu9Xu2H7jBQxcSet/nvqDrK7mOrjRdP33BQBdyryyU38f7bJtaj1u9pHKFogpeuHnE6vko39sdc7dPYVv+2GPvedS3vOkkNvREo5XtY0rE2qhzWZQPamuU3jCZTJ0gesCT0FEQ8eYUcVkv56Y3QZpVUaLY/HWbLJZJYyIP2RCkFCwhls53M3mhqI725rCoFhjfXCz2ddWQZkG2NUOCy1QJIjQfOd3Xkx+gPtHXzJ0MsSFIaXPZn82NwPVKdbfdPtwUA2K0NkMim+mUg/srq99ubUt5kR7YemVTZPOaTSzu9IMVyHS1AbU5QuL9GIpxZYP5jUsOQAymfx4j2hpBrDoptS7XC3XJPgOXLFFurrawzCPq8dBj9SbULr7V1LXSmyXqcUlMnS4gIE2XuL0qLHQahLwwhO6I7Otlr9TsFYKYLuWQIGZqU6QEmjai89WCyCphJR5MLIskuLCHIaHVb15K8VtwrbwuWbWEmbZ2DEcnnNuOreWbGDzoYTOhVxx0ZQPZlKJ1RdfDV4igEXMYpzOEDQk/AH00QyQQm/1EhfoC5wqLA+qS1ALL8fcfBsCuNAzWV+2228eaACkEqlito/ZKg88uX6dyresko/sEQrak8ebuhjff/hLTZeL9r32dy+lCrYX7OnM6V2IM3I1PuL09QI60IVADiChS3HWB3EyDNc3KpQ0oB1ISLk/eYBBLYG/evCEfMofbI+PNgZATBBviIGrXjPF2AFUbMzsm080MkRpNAi8FISTz7PN94fmzB0qp3D974PnzEzEE3jq+QXxLrPwpDaRP5Wurz0kCyR6UC00rrVnSoW1Ga9lUaRQNuk4w3G23j7DWmqldTBOX03NaKYxD5sntgZwSOWVSTK5iYGitYoldb3QSQFwVKMW4SMVJSIhEow6JEsMK7RSnCswKBWscPiNciEZ7TYI2853QlKBKTInh5khKidoqYb6Y+snlwlwrOglaC+VSraLRlFjsc5M2ChWRQMHH7yKUqrSpogLnqXKeCnOpnC6Fh4vRCEm2QAabRqja1mW4o099oRn65AZnTizSe9tEfrfXxl4NkQ3WPBJSNE3HWmkq0IInfdcrQ1t16lKeD5s+/d4fqb0cGVxXUtrmxFwxzkpzrctrUnhw9FVkLbGIGAoksOhBal8a+hg89U/G97fPZkCtzIha71aIjnKpc3R928GRpz5ZZfmOfOuiXZxar4Cg/v1sMTMj7n/4anO33T7M+rkepat0cN20tSSxnuF2f/HzOmjgcHPkyZtPrUv/4excUeUyT5RyIaZI5oa7Q0SGhGZrnBQgNoXJPjdpYEAIpXHQTGMgRWG+uWOIiXzI3Lx9R74ZSGNmOAwu7G6lf5oSY2A4evKaI3E0eS86dw9rADO/hDJX7t8/M08zz5+feP7+AylG5m+ZnWMoVkaVPvWvV3YE89CMSqNRPaE16S2bgGJjq7t/qkHPn8efebcvmKkq5/PEdD5x/+wZZbrw9O6Wp7dH0qJgYCoGfQw0AJtE1icX2Jkagw36QLAhBdEel0KQHtPaMhVyFpiAinBBONNjZlomZAVHdkPODLe3RjWqBaYHai1oENLlRKNRJjVuqxpieqlqaj+qZCxuN3EN2t5PMlcUrAdmbsylcpkr57kYX14Smuy7KroOWWgb/msIPhhButyX3a4aw3Y09rWzV0hk+0nDJon0zv3Acr+XBJoYxCi6FgQUWVAhkNWhYjQ+Df3klYW/utQlRNzZDXldnXE9oWPo5PH++HUiq1giqxgzrs+mx8UGeua7IrJYcuz72/VxbYReXEqiy1PLUrIfYb+/PrbQHOiXqM0jG0fd15i7vaxFsYYoczxY4NfNrdMFFr9ClgpAiIGUM00h5URMEW2V1oJNxwuW+JKME6sBGs2aPJpxyelSdxqIDWKnFMTEYRydy5s5Ho7kw0AcIjllYvQJfWpLy7Cp7LhqHdK8FFr7SrOh4ioCLiWEusJCsv0PMS5qDkYRaktC371xWXTrunCW7XObO72C0+fB77bbR9mKyF6Yfbxza80rmxaz+vmk1YZtoIoko9V0+sBCJ2BDKfBksXty84R34anrqh1rdcRMlMHOc43QTGVEm9MEsIQ3IjQr17jvbG99e+u2FzUEltBpzymmEFIbDVzjemYqpuPcB7Fso+Oin7ApHfWK6aLQ0BfkItd9l3sT5mtnr6YjG/v0HJujLBJpRLT56MeQFq5o81GsrXojleqa7ALERIzJhKBjIo2HTRm+23q2r/p6mwQaWboeRfBENmwSQlmco/9suia1SyIrQNAlGe1TjroELNjlYPknfSiEEFKiepObuLe5u29Q5RcEzyWpd+rDgmCvC4Xddvs4Cwi3klEisnBcvXER04qsdZ1K1VyL2c53C0jDYeTuS0KZC+eHe+6fvc88DZzPAbmYPmu+HUh3pt9cUqPpRKgwzkq4ACpEHRg0IbVxkEhLA4chcPfGG5AieUgc37w1dQJRWjR/rxQKFxrVEM/g/leFeO/Lvaow+yUgNUqaLY89z8bDDYHj3Q359oaUErdfekK+teawVoqP7YW5wVwBsc7xSECkEsRUF5BGCtbYmWTDsVUbvrIS53fb7cPtMk38//70f0OgEbUSpHF7Y2jsYRhotXI5ny2ZLYVSJlDlMGaCjDbpLph2+5LK9vjkRDfVZqX/crEYEzqAFGh5hDQQwsBx/BKHfGcFyVlp1Zqtp/sHyjQzV5PJm8RIchIjEoVQssntNZefS8niWExoGmyqphitSRCPrQYUzVNlnqxB7OvPHnjn2XPmWnn+MDH7DKXajEcc/NgCWJIKS3KqulZOW2tLQm9FGuOta+hjPL+hf+LdPkd7JWrBgpZER2gkkNRkPWKIJJ/4pdI7mJV5O6bPx9NZh2Vc+DEhJVJHOhbpgutCu2z+70klsElkV0T2Q23T7LVNZHWrXdtLiP75fcJKRHDci3VFvA5wWBbK9kE2R7p/Zk9lF+C1I7Fh+X1Zay+J7Cf72+z2elpAOEpkUoFq6E5vVgIbD9tvC182+Hnt51gaMyFnaqnc3N1yc3vDlC9oqFQphhIdE/EmIVGo2pi1EZuSiqKTggaiCtkpPJnAEBNpHDi++Qb55kgaEscnR/KYqa1yLmdqqxRmJhEaxlOtWmjaCLMQL4pUoCg6eyd2apTU7HpyqUbjEVNvkHEg5cThyZF4GIhBmFqhtGqBs4nJ6wlEX0gKQjB9P2IwnmIMq+i7iAXpLiC/224fZ/Nc+KEf/hHGHHlyyOQUKKUSYyTnzNwa52mi1kopM/M0ObVOGccEEr3vJC6xQL0y6bPSQYVaG/M8W4QxsBW8HC/DQEgj+e4J8fglWlPmS6XOjflyoZwnLjpTGlCUIq7E0wc3JJPQCrUQqktuNkVjQlOmj5SvmGJPo9l4eLW4f24zpVae3T/w9fcskZ2KJbAiYjOOvCC5FDS7SY+k9n9vMAsev1tYFdl3PdnXzz59Irtc1D3B6+ijeDKp64qpUalq/Nd5MidtzZLd4MluIFv3s5cnZFlVbfFLZZP/LUnkmvjJolDwMnyZXsTX5d6KkPbj2eLBPbHsv8l2T64+ZrOfsr7Ouk6v92e7f+IXo95xHsRw3P6d7rbby5h6F7C23m2/LsZ6+bLVVdwfFFJClhIm1jgRhTQkhuNIiIJKhdB8DPNA9iqEeKIcVEgYcpkQovamSBD1ZLk1Q4NLsWEkraItQmsGvKrVMKJaS6WN2rT7oSqxYKM2KzZjWm38dC/whwbZh6OIc4SSdP56lyDqfqdL57WoUFtFtHg3uY3fbNtyJe6j/RqXwj5yb7eXstaU81QQgcawlOtDCMQYqMF0U1WVFBO2AoScXSYvRkKEPhN2HfTBQtFTrD+jeuFStaMpLoMVjLIXYzLZOpQq1RHOngg2VCvTPFGqVSZK8J/ThXmulKbUri2L+I0lbkVxkbCKg0M2Qaz293X2jysP9OhuCgbNK0NcAUGP0dUlmWfBm5Y4rroOn9jt9bBPTy0A17Vr1pBFNbFmsekdtTam6YI2mNvEVE40tfLJ6f4ebc1E2fNoCe2TJ9ykp1Y6EUFCRKU7Sw/Cxp1b0kc/Ux8jsngy2xULeFEO6B6wHYu35K6I6eNtjtV+yto1efXMo++lB71N96Td7x2X61u7tm3/XFFZJhT1RFxFjA64224fY9oa0/nCfJ5s1Ow02RPWom/ctPOFEKNPp7KpO8Nh5BCscZLQgyzcvHHHt7ZvpZZCnS+U6YIEZbzNjDkDSikzdS7EBrcKh2ic2IMEsgpaG6EVmGeqVs7P4HJ5II8DQWZaGRCF1PWaFYRoYuwqRB+UEiYlnhtSLUkNjkTNZeZSC0rjKIFRsgt/ZBo2oCFVpU6F5s00QQISlNImprlgA3gbUSeCwBghZ9cmEefkNvUE3+hUQ8pEX3A/e+/hc/hr7/ZFsVIbP/reA28+ueHJ08w4HIjDgXEcOYyjqROojT0OrossIhzGgePNwaZntWLKGX0x6v0dnWXXVDk34VIsZoSQCBLt53DDcLgzRHY8kIbR5OnOF2gzojNCIYgNEXn/2XPm0mi1UOYL2goxmF8EUWqpXKoNTFDEpgMGk9WTmEzybp6ZWqOpcGlwLlAqzC3QSEsNtFP3LqWB2ACTIQrZ+1Esf+8pq9Cjb19awprYWzhvnhvsqezrYq+mWoCVPtY1mbrclo1zLNNsCe184jQ/p1abjvPw/H20NYY8chiMJ3Q7JOLtrUnpSCJEl+EAKp3Ebs1iqC4n6lYtc1uC36KxL0JllbW1aj0eWR5/VNd4lMzavcdbXFzLgdeeaG9nQi/75ahv7FuSdTOdUN8h7yq7S+72cqaqlMmaKcrss9Y7uiFQamWeC3EygqmgPkVHaHVYfSYKooHD7ZEohu5SZ6RMgBKCLWC1WYdxKY2ocINwDKbokUW80QukNajF0B4aOgVaGRgOxkmNBJJkokSCAi2YUHpTQgs0hTA14gVrKBMvsyIwF8qk3pwmxJzcbyLVSUCxQZsbEvz6EQLiHOG5TIDpbEatxCiM2XRtW3HPdbS4N+fEGDkcR4b0SpfQ3V4Tq0157+FCHkeqJCQPhDSQc7ZJkFrR0biyaRgYxiMSAsOQGccBEZiniWk6W6WFSm11oan1+Dg3uFRz9pQjUTJIRuJIGo72mWkkpWz0c1FQU+QQqUho1Gnm+fN7Hh4m5nnmdLqnzDOHQ+bNJzeMQ6JV05RWNfAliaASkBhJOa85gFh2MCtMFUpVaguoR77OoVW1IQ86Vx90ZI2kHZgy3+v0u0bbDCLpWLJ6TqBdd3YPmq+NveJAhPVnb6Ds5XScWrCUMEuh1EItM7XMtFaNX1MqMUTGnHz8ZSLnkZQtqGq0RirkutC/lu29ZL+kpZuMcEkf9QqZNd64LK8QWZ1AfN/XtJS1rOgHutAP6OzWR9SEfucRteED6bR/jmx3mY4SW9Rc9G13p9ztpczO0RBsgEBK0fuRVtoPyznlclyi5qfFaAZGDdKlqhFTguZFRNPHQbTQZ1eKWpNUwC4o5pPueX5diFFIKThtyJQDIoqUis6FhiWuUKk0qtZFucDpqsswhT673dBQcdmihLh+dfJBCaIC1a9PswVJQnDtV09o4YrCQ5fdW3lTa5AU+y7ykMkpMR5GR6V32+1jTLA+kN7177I4Nmq2eef+GmdCDIsSjoUmWQgAdl6GZVHWWqNpW4cTdHAnRKMRpOQUPntcW6WVmVZmtBS0ztAKAeODpyDklBgG+8TLJSLBkGJTAUn22WJUCPt9BV4IXrmUhazglIc+QGHVse5JLDjnlU6R6KpGawP2UovVx/SKnsjrEp/3ePl62SvDCV12S3xFpr3pSRvzPDHPlfPlzMPpnlJnLufnnM/PqLVQp5lyngH42lduubu5JabE7e1T7u6eEmPicHvH4ebWVBCGkdQDxzYr3DSE9ZPcR4LQ083uY1eJI905rlPMRXPTt309+u5acmzdzvr/x9rSYIY7ur13ebcExOWEWm2Uahe73Xb7WBMgR+Jh4ObuhpBsQElz1YKck517tdFCpRbrdp4uE32s5XBzZMwJETGh9qNxfNIUiRifVc+NdnZqgrq0VTOkNjSfiKBKVVMeGG8SYbTA1geGiICcztRp9uaNYPsmfX68NZQmL7NGAjlag1kIgRCSx3Qrnxq/UMjJfPTSCpdzhQAxFapckBjQHNBsxxRVbLymiE0VS65LHQNEG71tvS8KMXLz9I43S2XImTeePOEwDva9/4E/9vn8vXf7QlgMgbubG5ObywMxZRThfJ55Hs/Q1sZB8VgXHe23HFddkzUCAYlKkIg2ZS4XLsWGd8wNqjo1KB8YPabmfCCGDCqU85n5XKilMD+8Tz2f0NZIVEIORMno08ztDZxOZ2ptiATGw8Hi8Tisi2A24JWApEQcsiWy80xRm9w1tco0F0ptzLU5X1apHts6TcCYdUqSYHxZrNGyPy/0qX+WGIvHxeASXnY96Jrxezb7utirJbLyKKnrCa0nf6VUylyY54nLdKGUielyZppOtFo4Pb/n4dkztCrPh5H38kiMiTfffIvpzbdIeeCNVokhEJOJRocXICCWEG5OWrW6/koIWLsYr1LWhcP6uMtxPQb6ylG3b3P+aniUAG+2vnWiD3WoDQ/3+hVeHgnWtNNaH5+7224fYyIQbXTscBgMbW3NBM1ViTH6Oe26ka0CgToXZoEaI3HINjM94uNnB0QbuSm5KpRKrUK9mGPkaPPhEaVSfG67q5IAKkoeTMXA+r28GQ3QqaAUWlPrllYWPUpEjCs4dNRVSNESbGtciX7QwUqRKDkIY3Jfn0EnUzMIp4bGgkZBD9YF3tHkFIzXb/q4phe96GRKWIhTBAvmN3eFwzjw5I0nHA/j5/BH3u2LZiLWIDkMg03wiglFmObKZSqICVcuiVjMhnz2qqaqGh/V5SzNPaypsTFRHNmtTV0DNhDSQBoOjspmS3wV6jxTq1VG6+VMm86gSnSVgpAS3GTGaqo/z56PzKWSslEehoMv3joS2qzqqiiSjN9rXHtxTVpLWudmoExprRdzaJ7QAlRjDRgtwWMfYg1sS311U/JUtcFIoli89BBpYPEH6p+7fRPbK3Jk13trImtNIikmxnEkhgShojJR6sCchSEptdqkHS2maSkItc6oNs7nB54/T6SUbf47EFOmFOuyFgmkZDzahV/aS/1LRfDRfbiGYnvq6BuQzQH1qUedrPA4EbXfxYPn1dY2z3/wvj+wvHYhQ/SO8o2jSk82PmQfdtvthSbYpL3oqEzsQzoiqmrPha28my/lWqMWG8HaSsXGC11zXlQx/daqNvLWM8+6ke4yWa/uW3Y+h8XHfGEYZW1QaV4SbNCqLr6lXt4Pgk/TWlGplf5ji9VVUN2uPx2RiSlb87fgGtU2QroUpZ4LFWy85pI4r8NU1Eug6wJybb7sjZhW7tz9crePtxgCT44HbsaBnPoEL+V8mYgoKcKQ7Nxqy+RJ1kXnZvG54qCOgvbmTPycx4YKpZTXOClh4ci0UqnF6H5abPQysDYYx4CKJdK1NW7vbgkpcrwZOd4cOYyDT7xzfm7zyotCCpEQgyedm8SzA16PGrCtqdR8aCtvtx22FIJN6WTZpl+znH7Qv6OOyK49NLu9LvYZdCrYyRk6qkFENXBzkxniDa1BbU+Y2xs0LczTA/PlObXMvPfuu7x7fId5nnj2/ns8e+89Wq2cz/d87WtfIYbIze0b3Nw+IaWBN770Nk+evEnKmSdPnnK8uSXEwOCNGQZGhcVZpAecDySyH32Wd76NWiuoT/96lMyKWAOLdKbAomL3AW7tR397+LSz1UGR0IuvG0T2pf4Yu73mJiKkQya1TBojVaMlcO7q4zj4OEyT/umT8FoplLPJA+Wc0bs78LPQChzGMW2XCqXQzjP1PIM2iq66yNsglFMkeiIdXVtZQjB0KAS7NhRDZ7Q16lzXQO2JqqqJtTdV47eG4tsPjsgKNvTatTRDJObR9K31wHi00qM4eNtqY57OnMuFKoLkzCFlNAgtRpqP6q3eeDKXZioH/i/FQI6BGAW1lpvP4a+82xfNhpz4Sd/+LRzGzN3xwJAzWipf/ep7vBfg5jjy5pMbco7k2YYGBTG0suuu19oopaLKwqFFlBATaRhRbcQ8MKo1cN7e3nI83q4VU7VF6nx6YDqfabVSLifaPNk0v+FgWu4hccy3EA/c3t1yvLtlLoU8JG7vDqQUqbUyz/MyUKUv7OpcKNOMtmYUHUdmQzB/DVii3Lm0gUAOzowNroQk69jpKELOkSEaV7jLd0EfWug0QOcYC15R0b3Z63Wyz4Ajy3ISW/JoK7+YI0O0MXgqB5ocUBplvqFMB2otRKBOE9Plwvl0TymTSfmURpkNeX14OHF8/pycRyudlMYwjCQXko4tojFaMcVXfJ2zG7oW7SMO7KMj4BGeyjWfYJUHubIFLe0k/HUrj/3nQz9d17W1BW9Hz9BFd7OPDNx9creXMZsuF1dUNvYqiflEzNEHmYTFT0RMSL1cTJGgzWsj17ZKoc1R2NJoc0PnagmoVqq2BfmUaI1YygZZifYzxEges5VWG5TZOXK9w7lzwV3/ttbG5PzB1poNb/B9XqaSSVqRGglIzCaRtVkbqgnPGnJ0aczniRYEuU3kHFERSkdjnVpUtevw9gqPXVtiDN4Q5ovdvYq528dYCoG3ntySUmLMtpDTVrm/P4HauX5zsNHNlrz6WNdNIruU4VWX5qre1BViRH3JCtZYNgwjw2A0gFqr+3AzGb3zyTScpwvUGTQRBqtOhBhJh5GQDoyqDMcDpTVSjoxjJkShlMLlcnG5MFPxEBEu5wun1qiFRfNWetIagg0GWpBaHz9vd5dIuqKybOTu4opKS1/oAn2xq2siq0vT2B41Xxd7pUT2ak55h/sN219K+0v5w5FR63wcjDx+vOXJ0zeYLxPzPFGLy308nDmfzhZAApRi4yfP5xP54TnzPDEeRi+fJMRHwoYQGQQkrod1xX1V++/xgIWNSyzH9cG09AXp6cKxZSnDemF0Lf5syiAfMJErqgHLN4dvpUfhlwJ3d9sNZaMxuWlaXILA8vsqqL7warySobiKQajOGw0uQ+v3SRyOR0v6auM0nbnME6rKpRTq1IgxoBxAsiEsEohB0Ca0ooj2BNGqKCrBhNQVnzjWFp69TdHSZT8FPDDaATSt63SgmqitgNi1IebkwS2irSGhMqSBMU00EYpCLQWiIJoQR3Z7E3mtNsTlcj5zOZ25nM+cTyeGMaM6Lkjvbrt9lIkIB2+g7Is0fEw02oxWsxJ+HpXYey9/V8kxXqoWowQo6hQ8l5UTm5S5UAqwhqmmjRQjY87IMKCtMmulihKj9aDY4IXgU8SsbJ8IS9V1KRqKTc40BDQsdIIYTFNWpTnYZLE+VSXEQlD1Rs2A4tcA7UCQO12nFbkySU6ZYUieDzSktMU360Ir0oVq5OyC3V4je4VEVpdAuTpbWyCQbRJn3FmX/8gHUjYJn5SOPLl7k1YLb739Ft/+bd/KPF/4yo9+hXe+8hVKKZxOM6fzPeJdjM/v78k5cz6fePrsfYZhZH77bZ48fWKBS+5I7lQdoV14RpvEtSery08/Jlj5Nj2DFLFV36N+sBXtQVBdkdktRHOlbKDrp7wose0Xsm3T2HZyym67fZxpU6Zp1ZBt1ZDS5qhirdbd3EJAmsncSHPEJNvlQBWfvlet6URcVCtGwjgQVXh6e8fTEKml8s67X+Pd995jmifef/Z1vv7sGSkl3nrjKU/aDSkEDmNAckAazK1QxJAc4/FFD1gDMUZarS7RZ+M20yX6fqs3vuAyWRa6yjxzmc5eclXSEIktEceB4ebWKjM2j9aOHUusi1buW+F0OtNSIOQAMlgS680z86Xw/P1nvP/Vd7mczrz71Xe4f/6M27tb7t44kvLh8/xz7/YFsRSELx1H5lK4TLOfz9b8pKrMh+bDBUyVwPjawRdrfXFZ7b5i/SLN9I9DioQhG60oJWKyZqthtOZpBJaVWYiMtw0dMrUULg+BebogIZKGkZAykpyql02xIxDX+CNWoYwCMmRDP0WWSZytJFKOiCjj4cBd88WoBC5zBSmkXEg5GzWvrbQA9WuTxW5biOYYubm54cnNgdaUaSrMpVJq4/7hzFzmhfvrl4W9QPIa2mek5r0yujdp7PrUwqHps6KtSSvHDIejD0eIHA6J+XJhnmce7p8zTROXy8xcJuetCpepkLOR2AHG0Qjo4zhYo8qxrsl13wuxxLGX8NeRneaU14oHj+97arrxjquEduHffvA7eTyIoTeR9KxY/ee6gly/x05iX3ftMQVit91ebLVWk65ZUFl7XEWtSWP51887g2V7MFLwBkyQGKwLWNTRWZvaczzc8MbhSJlnnl/OxIcHqI3zXHj//oGcMsfDkXFotCgMGdTnvWonuUVjBQQxNYLDYSTFRKuVeQ60WgnBRseW4FJ06DKPvZcjlUqpE01hrgOlFns+CHEYTG+2NAiKtspYJloqlFa4tIKUGVFXMeiITrNFQa2V6TJxfjhxPp04Pzxwfngg5YjSkLXfc7fdPtREhEOKSGtMtdJmk5yrvjBrPlygc72XSCrWDLU0GHscaLUxlwIiDMlHvQcbBpKduhBiWhofJeiyH3EYkBioZUbKTNAGnR7QG8eiOGfVlBI6H9VUTgAx/urjoUOh04pCIOXEMIzEWMmXyTiyrfkAFuO8irYlnvbPsLAoC2Uh58w4jq5iEJBQYbZjb5vrW99GT7r3aPn62GfT7LUkbb3859iiVURsXOSS0K6JoDinTYPNlD4ebkkxc3f3lKdvvMnlPHE+V0I4L06k8wTaOJ/uSUGoZeT99zLaZoZhQNvMdLklpsh4OJJT9s/sItT98zuc6myabYlfO0a7RVnXhLarC6ziXS9ymoV08egR5wd1rpA/IfIoZ+0ftuyrXG1rt91eZJ0Hx6aEh+BDAHBZKYxrGiAka/gyLMMbJn2RuBQlnIdWfDpXI1BqnywEYx54cntHDIkhDwQfGm0BWp3XGgwdYq05LKLvTikoc7ESa+u8WXOMYRxILRlHsLR1AehJZEyFEDJNlWHIVr9ojTJNnB8ejJcbEilYaTcNI0dR5lo4tZk4na28WW37TrpbGi1Lrcy1UFtbmtV6cD0ej9/gv/BuX0QTsGEDIuQYEI2UqmizxEvosnTNObI+JGEDvGxvtVnjFwKhJoI2AtET37g2V/XFaa0L790GPHdNVh/DjPkgfRSzWveHYTk+VaQ149VuwZ6eyIZ1dHvypi5tUFOfSCK0Vn0ARHVaQDPfKp2/a5z7Tu+JIRBjWK5j1ogqNlUwBOLpbIAQukiUBacoOYn9G/TX3e3ztldIZGUpfxj6Gf0hnw6kdhMvjWvPamXt75fOudFAun3C7c3R+GpEcjpwPp+YZ+XZs3vXo50p5WQcnHrh8vAeMSXun32V8XBgGAa+9KW3ub27YxwPvPmlt7hxQejj8UhKeVOycb7ekkE6OquyBG6cdXOVgG6S4OvkdfvbBx1INz+vFApYE/4Fmb3yQblKwnfb7eOsy2ilmJBs3fVNDIMNKaAJNCmSA2kYiC6ZE5Pz61wHp0tQhdZowHme0POFLMLTlJhTJiA8vXvC7eGW5w8PvPPOu/xo+CpRItpgqg0JEQmZlA7Glw29KdOav4KYOsfDw3mpnHSJsJQjd8ejITiKlyFxzp45yTQXzpNTEabCNM1omTg9f8blciGExJOnb/LkyZuEELgZM1GeMM9W7bk8f27KC5dKCzagxeK5UItymWYezhdaKYSUORxuuLm944033+RLb731uf2dd/vimAAjGNKZMyVEpmoqq7UpgUAtNj66zIVS6lLB6yhlcapQbcbbPp3PgKDJKT8iEAIpDxvaTlq45qVUj8sCIUHAxsFL9ioM0BpBjLcbaKtyj9PzqHUp46PqQ0sMgRUREsLNeLCkMkRUhVBMaWSaZy7TxDTPzMXG158vE6fLhGI6tgGFZA2pfYJezsl7YQLDaMn66TLx7HRBzhOtNqZm31kKXc3hc/tT7/Y52CuOqJVl5XMFInopMjg663naVRIoS1ndktkQBmIYaK1xd3tmvkzkPDAeDl42MX5bmX0CkTS0TsQYqeVCfsgM40gQKGXieLxhPIzEGMg6MA6Dqxusq8cF8eylV11H4l2jrLI4xoqz9iPeUhE2Ka68hCfJVb764rf173dPZHd7CVN1Lp0aChpiX4zVtXIS1KAT7wi2pslI6s0o6kodnXvgqGxtlVKLdfg7x09CYMgDcbBBC0MaiI8RWe0lUgtQMXbFhA7WCDQL4l2JIKWOJgfyMJCSSW31C4k1tJiz5FJI2RLZBx6M36uNOk+0uRJi4nhzZ8E2RmseiUKMEzlmG9TQlODDHvpK0gAqpfTjVptwlHK2a9N44HDYObK7fbwJEFCSCC2a5qsqJG9UBh8OUNvSrNkVM67GSfvvW0S2VhtPa+FVnBoQ16Yqp/JoU9pVBHM01mOLubyiUTu3xnbNNWM7B2LZnm9pQWjFkN4+qKTURkwmnddVRgyJrY48m1/NZbbjCpACqOuzr6hyXKS8TLovUZWFOoEPDKrNhvO2TQzf7fWwV5ffYk1oBUtQRRZVxyVPVOnKqM5pEVkGAXi+uzhsTInxcEAVjocjx+ORGCPzNFmARWhaUTUkpxRQrahWnj17j1JmTqcHFOX+/jnDMHB6xSug3gABAABJREFUcs842pSTYTi4XEhEQqKLN3elhQ0z1hID2Hjt1YHbnRe1SPYh85t89wOv6ujrsklHgrdo7fVLdtvtY637pMlECQTIIYNAHJJpyMraDGlVQ5ff8gQzuAC5lkKjj0uuNFEqcJom3tN7UojcHm44DAmJiSdPn/Jt3/7tAAxDMp5cSp7YGs+odzzTk1hYhqhU2ehSYsF39uaYlfXu++ivqa0u8kISAsOY7XiaUJtRELRV5ulMDZEyKWeg1pnWlHE4EFAuKtSp0ERowcZtizaGIXO4PdBKZVK1kbrAPBemy/wN/uvu9kW1DugkEeeIBqoGSoNDHhiyIZDR9ZH7eW4/lRAjURVpgSAXX2fqUqK3MrsQky1I53lmulxotXE5nZjOF6IIbRg4JKPqNAKIobZVraEyzAU9nQjFeLyzg7IAizqPf26nyFUJyzqzNz7fny88ezgxz4Xn9/c8nE5cponz+cL5cqG2xjTPlpBjqHTzhHerrrJGY7tm9QQ9pUTK2Sq+CMUXwaUamrxLF7w+9sqTvfptSVxl/T0ugKdPzjEm0IZ8JwszW7VRfbLPkEeePHmTYTjzxptf4s033+R8PnM5n3j+zFRV28IjsmoHGNL68PCMkBIpDbzzzpcZhgPjOPLGm29Z88k48vSNNxnHkZwGDsdbUspGKk+Do7VGm1BcfHlZCa++IUEWasTqLy9QrrtiHLwo4e2cAthOUloQbta18267fZwtvhgESVYyjDkyjAlxh2zBzrkUDNWMQSjTzOVsElppGEjjYIGxVFCf/INJQ6oq7z7cc5nvGWLi29+K5OGGOBz4Cd/5ndw8fUothfPpxDRdbLRsCGg1Tm2UYWnW7Iu2IA2GQFqGj9it1sb9/QOoIzTOc7Wk2C8efceAmAI3+YgC01SZL7bvrVw43T8DTOWglAJYj/jt7RvMWrm/3FMezlSBluP/n71/D7Zt2876sF/rjzHmnGvtvc8590pXryAhwsNxIDG4UqZKYPmRlOOAXS5kbMoOFhSpECdQpmwcEhxHlZBEiY0LAyEYUUY8HeyQEGzwC2wVGCMeAoKEwNiFBUYgXd17z9l7rzXnHKM/Wv5ovY8x1j6vffa59x7uOaPtmnvNNR9jjjHX6KN//Wtf+xp4h6Nyuj1Q6hPSNfG82gReFS7nibun5y/9H3WPj0U4tcWXD8ZYxuCI0VMRxtPI6XAkRN/qOtoMIKsfehRni71aub9e6V7HqRSkLa5wQhxHtFae3d1x9+w5pRQud2emy5UYAm88eUI93TRyJYC3sWnyg4ymTJkuqOsyuGaP5zzBDzhxlE0GBVZWNldrQVuq8vxy5a27M3POfPZzX+DNt95imhP315n7iwHZVCopNx2tOlCPk01DIu063fYdiif4QIiV4XBgTBm9TtT7M6kx2F7y3tb9ExZfFGnBQk7qBtzKw58qW3fVDT3ZJyU6I2u6uWEYUFXGcWgVi9YpxBhZbTIA28R2BTenZMAvBLPuCZFDY3en47FpZT21ZOpYTHuDIkQIveVtKwpjxdw9xbpoJLTJCxY5AdujWx9ArdiNNUW7foHb19kdY6cfcrDy4DV77PHe0Vkc5200Bu+JQ8QFb56r2vqiy9oWUmslp2QpROeQIdqysxRqMdYxeEfzA2LKGb0UUojMuVjrWO9Mkx4jOSWeCU1rZ9cK2zYsdn3b09w5vGK95LVS2zgvJVNKXjR33rOwu9oqqMU7XHCreXo0oFCLUl2zxquVnGa7Dkwz0zzjRDgdmuyoZFAxy7I2p3Y9f4ie8TAamA6W4lSsK1lK+cvxJ93jKzz6AhO6E0CjJ1xAxTGE2HxcQwOuncLYOAN4b+OvFqsTWbL9lqZ31Zhc58xlIOfC9Xolp8z5bEB2iJH5eEMemxZdO2mjq7SBSiJT1JoS4azhiPpoFnWuWeS1zl7bIrRUK3POFFUulwvnswHZy+XCdZqYU2ZuOlnLmliTBwGqA3WCqltB7Ivfo/TGJCY78CHgfDZOTJVCkxi8UwZ0j49tfEhpgV3ol8Vje8xY126E0RMqnXnc0Jpv35xtp1mOeG+VziHYzTVB+RKLno+lknmBirVS8gyqTCh3z99ini5cLyM5zUvXk5ubR8SmeTscjtaT3UdCY2dVhdpkAuI8TowRkhiR1s1ks0OsxkZvH0jd8np59gVpgT0sLKOwf1/7kNzjJUOEpauWNp2q94EYB3wDst1CJzjzc9Zqvq09DVkF8GbnU0uilmRbG6I1NQAKSnI21u+mC8Ozp3hnVjmn0w05JdJkBVKtF62xoKrkPOBcWMdKG7q929i6YlRK9aTkWstLb53DDHmj1QOm+6uqUNQAaG0g2Zlez5Ie5lNp7icZrYUM3J0z58uFuRaenu94Pp2NGMoeorkziCghOrQGxtOIohxOB4bjyDCOH8nfeY+vvPDiHnAU5ldcKFRUAzYQLFXvfMB1XXjXsGpdCpKrdoeDSp0SExBTbR617blcyClRc8GLMITAECPDEJvDTyUXe35OibvzmevlbFmbAIj2Du2WqQkDhzHjfSDnwpSsYCt3dlaVVIo1RanK3fnCm3dnUso8f3bPdJmtYC1XXDsu73RZVPsG5IMP9ryaVNE7IQRr8OCCs6YswZtWfRhI2QB3bU4LudaeOvoI/sp7fBTxykDWVphWZdi7nS8AtoNYqWvqfGFh3wuYbSqWXaDWSowDwzBSSjHt0JJzf4Ce29vbPrQq6DwrRWbSJFzvnzcLIE8chuZPN3A6HQkxMA4Hbm4fE0JkHI8cT48IPiIu4L11JxmGkXE8mfhcBB8jvZVe359FTq9QaSC7/aub7+CdGiI86A+tFao0WcN7gP899tiEiBBjNKamnU9DHBgPdp5TK9Irj5s2VlHS9cr5+XNKLow5WwMF59CaqGVe2FtaWjQ7JbnKTObHnj/lfH9hHEa+4Wu/jjdee52cEpSMR8lp5v7ZM6brlRAjPo4goU3G1rkrhsDpZBpB1xwJRCCXwjTPlFqbHU+f5Pq1RpnSzCVNVK14F9DQCrZEzO9VzYO2pMk+MyVKzpRauLtcOE8TqRbenK88SzPiIBw8PhrwkACHU8RHoeoNw3HgdDxx8+iW083NR/r33uMrI5wI0YdFQaaY1VspyVjJ4oFqWtjgzP84Bqqa6w8KtWSUTEHJasVUqWSmXJjvr8Rh4I3XJ1JrHT1PM/NlwhoYmI3dMIwcjweOpyM5Z9KcjDG9Xvn8F77As6dPccETTwM+egOslys5F4Y4cjrdElsW5jIncq1c58Td5UoqxYBsMiB7Pl949vxMzoV5nrleZ7MU67ZavU6mZYXGMHAYIjF4gnNIrQgQgmc8DOZ+MkTwAScwHA4cVchVwXlKBRVFqGRd/WT3+PjHh5QW0OqZVpOqDRJ7qP/sj73UdrtcQR744b3YYGAbqltfVvscM3U3XV5tanURwV/N1D0MkTQf8cFzOBwppZgU4TijKk07GwmhpVhU8H6wVWQ1gfqLuJpe/NWI1c6urrKDF0CpLsTyA9L6AVu7Y9g9XjpaalFtHbk1Fg/egzhLJ6pSqK1osi7dtHIq+JwJuSBO0Zqt206TH/QTs2LrLNXKNc/WbrOd53EYcSJL8YpROtaoQZwzbXtnlIrd9765LCyFHL0dppC9LYr7cThxbZvNnSFLq+i2bYr6VdbkBGlAtr+mM1u1VqY0c3cxfd05zVxLQhwMzhPU5BI+BMQ7gnrCEFGEYYzEIRJi/Aj/1nt8JYVJ4/QBwLJ2zLoZW23+c86AW0O+pmhzy0xrC9XV5/haTLedGoik15CU5intpV0DHMF7fPBNptecOUphmmYu1ys+BoiCCqQ5cb1Oi8dzcJEarKnDdU7kUjlPM8/PZ1I2IHtN5kpwvr9wd3+m5ELJ1fS3Ci4I4k0ioCius9DOdLhePEtDCLXOYa75yTb/Pit4CyYtsO5l3WUEiuraDGaPT0R8OGmBNuPkrrODB0hMUMNxC3n6DsVQm8pMnFukB7VNrinNTNcL18uV69VuqFKLJ+emOmoXAUEa6DVtkXOO3iPFLzZWbYio2RRdpysuO2sdmLIJ2uMzxreeNtuvgHMREcc4njgebwkhcHN7y+PHj/HBt1RNtM4q3i1Vp71LCphcomuKe1Xn+g20r3M5il4cx7L/7w7h99hjDRHr7uOaJMYmjoCIgdhaEvl6bRICJbvWIrNkyyCIknNmulzM7saD96tnay3VJpbg8G6EoqTzzDxNzLXwhbfeYgyDAeVSGA8jzgnjcaTUTPARH6OlTj2Ewfa7Oyx0MJpzXX4PIVjjTnHLgrZk89vsOjvtRcpVFod5XS9IbWzZtaW7JtRqhSaXaSI3fZ00D8qc7RokriBaYbbxfDgeOd54DuOB06MbDntDhD1eMgQ7NUutVJTUUvq5VEKOlJJxxbXn20JRPLBxMUCXbGAcBhRhupx58/k9IUbefOuOt966a5l106yD4rHCRu9Da/08k+fEPE/M85Wck7HGIZoH+5M3OJyOpHnmLj4nzdYOVxRKyqSUuU4zqQHZu0siFWsdO7fF6VygVtO8OucYY9PwhmA6dhGmkqG1pQ0dyHZXlX7UXcvfOvW54QChMCQlqSNOM72pijZpwYJH9vhExIcAshuwSh9mS+PLtm40jaz0dLpsCdqH0MyJgPNWiZkt9Zdy4nq5cHd3x/l8z93dHXd3d83ZwBtrw2rF41paNXQw620EmO9s92Jdu3iVlJjT1NL+aztqBbS6pgIwAQUI42EFsq+9/hqf/tQbxBh59OiGx49vLAXSvCWdc8ThQIwmS3DBdIGG2k17a19jZ7LXz19yLe1bdcg+KPd4uRAhjGPTx9roDDHgvBmJ55S5f35PSYnioQRLx01pXoqcUpqZZzMxP92MDDcHA5qKaV6dYxisAUlNhad399zdPWXwgegD89mKSl5/csOjmxvSEM3/2VmL6jgc8GFoY8QcQ9ACJZmlDyDFriM+BMZhxPnuJmLjplRlytYlKGelFDteVwSKtMXz2szESCrbthPT3JVamHPi2flMAeo4QrT+8SlN1CkZIL5W1MHheORTX/spbh8/ZogDj28fcRh2jewe7x/dOqqoLpX9U0pcr5aS9wdPSjPisQJHNUrDGheM0OYvcQIuE4cD4+GIiuP+s2/xN3/kC3jvefLodW5Pjxli4CZ6Hj953SQ42aQ+zju0FqbrxdjWy5nL5UyaE955DsOB25vHfO1nvp7Hr73GfL3y7PgFI5OuE289fcY8TUwpc3+ZmHLh+TTz5v2FORcKQmnzWElKrpbJHLxnbO2tQwz4waNAmCdknkCE6EK7CUHEwHcHsWKL3fF4IJweEVJhqg51kWnOOB+aRVjTyL6LfG+Pj2d8SPutd+JXH/5cNbIs6fIHagPpT7f/25O1Wts5m6h6txPr3qFqznGqrlUw2s00dJW6VIQ2BobGiC573aoktVKqMVGlmI1JVaUUJee1B3Yv9hrHK9dpNqcDp8RobKzzhTgoMXhEKlZc6loq1JlfbS3g/PodLEUtL0gRYLEq20hvl1TLHnu8V1ihSNOFtsYEi6Uc3Xc5k+dEDUIRWsVvXZehtVJ7Z6HadeB2vlbtrKa5i3ST9bk5C1yuV+7cmeM48OSRdeTy1VKANm6saMNM291SyFmLUGpuiiRd0p5eWVKOPX1oaz0Dp9bOsz24LA5pacvlVxbXkDa5dSeF2jw4rc87bbHbxn5p2natqFTqaOzwcBgZ4kAYoqVh99jj/WK55i9JR3MJaC1brfq/SV4WOqi90bUW0s5DdUumzxoFeOuQdZ1xznO5TFwuE1orx3AwWczGVUBE2hxX1ltu8iHMDSCEwGE4cDre4HGkwwUH1NzcAKplS3u3sJytzay5l5gXrHZips1c3TbLiSzuDIoy9wLPzTxuWdMVFSwyoXZt8yFQMb9cH0pbpG+/205N7fFJiQ/tWrBoYcFYRjV5QK+YXl/VpxVhaZKuq95HmtVNrZXL+cz93TOulyvn+zvSfCWnGYcyBE+tNgFNV2ONQgh4b7pAX7RpTk30rU4XOxPfbHM6JrQVrrdCNbE2e6rN05LSdLWVmq2AK6Ureq7tAlKBRAiB8/2RZ0+PtmIcD41B8hyPtxwOJ7z3DOOJYRgR8YQ4mCtCS+X45o3Z079t5C9tfaX1w95jj/cNobkTYC2Y1Qqm6v09ANP9PdP1SkkJ8BA3+jLp9YWmXZUGFqV12JpToU5zS+1XcipmqRUcxye3SIW7dGV+c+YwDBCUVBJdOXQ4nQCHuACt8NJuztxAsErqWm1ytcnXxroBcnMOWRamzttw8YrzAdXmbNBkPI2XWS5TVeviL1mbxjAOA6ebW6pAGUZqHFAtWOGN7bcbPOKFw+mIjwN4TxWYNdvx77HHy4TYeeudR0UYhoHT0djZw3ggOG/tomma72LacCuI6kW/dfFHlXaCplQ4X6xBwt/60c/hnOMwDHzNp1/j0689sm56reWsiBBTxXtbzN6fJ66XRE2ZPBc0KXUq5MtEur+gKTGIx8eREhOHEKkhM6UCuaDJfGdJBSnFfKm7ZrX5z6IwNI2+Q/DBWWMG2oJymWcTCYHgqMWj1S/gu8uKvPPEJt0bx4GKcLweGI8jw2FocsRE2XHsJyo+JJCt9Cp9WwlJM4OT1g2r2U3JSpb0FAnSTc03aBfT4N09f86P/eiPcr1eePbWm8yXe/I84agcBrP+uL+YCN20q2rpCgfe2fTlxFr6OWkdjhbGt7FLbYXnCagYeevbcRRfEDGfuxmTOBiTNXM+34HAdH3G3fPPm+dt9AyDbynL2Cy8PLe3T7i5fUTwgZubxxyPN3gfuLm55XA82ePHW8bxaCnXOOJ9tJ2RTlX3+zvzs8f7h7QixlLN5xSFNE1M9/fUnMnTlel8Ty3FGhOMfmEvO5DNWkk5LeyNOJMlXKcrl3MDxOPEYbi2nugDt1/1GmXOvPk3P8fzzz3lECNzmbg733MYBr7q9dd5/Oh2YTprtWrkGP3SD957hzbGuNbJPGxVqLmCUwO9bbHnGiAWLzgFgqHV4J1VPHfmidXnMhe7X9QM24tWhuOBxwhFhOQ82fuluC3NxhqPpyNxHIiHSDwcDcg64Vpn5t1Gdo+XDTFv54BV64vzuBCpqgzHgRgiwVmhk7YWtVIX8XdjbLU5jWDEB8KcMs/vzo0d/W/47Oc+z2Ec+OZv/Hruv/arCcFzGCJjsLEeSXiEkhLXpxfm89Ws6aaMpkK5JOZnFyaJOIGDC7gxQCrcDyMUI5HItrDVKcE8Q6mE0RGCN615BJo9XQBikwF2uYCr4FRx1WzxCgm7cHnyMaAalyyJb7ZcsVmIOV+5uQE/DMzzxOnmyPHmQE7mRlKaxeAen4z4UBrZhWXVjR6lFZgsaoKNTkVboZWyrQxjBbNNKJpzbiL0iZKTVXSq+cnF1hGo93Lv6c5arTVuTy1UQCqYL9/D/ZCmm1WRZoViW1wBOTiXUV37wXfNTa02c6UkOGetPEtypLkZNTevS+9908EaY9ylDeaCYC4MNUTGaB3GUFBvDLItANbk0l7qtcfLRi9yrDYDtHNW7QI/z61AqlC14B6MzfXn1t9VN+O0p+FV1QozsC5F8TBaNb8qWSuX2Sx/ztPE5XpdCi+8DwurZJmG3h2vLTJdy0A4x2bgsfSZbxmTrdymG8bLkr3oj7bXqKysLGuKtf9zzvwoBajOUdt2XJM+dA/eOAz2ui5xECiN4d1jj5cJOzfVtOIqNn5MBLe4cSznbp9TNxY2SnfcWGUy2uQ1pXXIulyulFKZDwPP78/cXyZi8Mu86BBK+9yaMnMyaYCWipRqbgelUlKmzAmcIzSSxrFJ/UvfRwPYy8QJJmhaiBhpj5l0bpEJ0OR06CJ96NuzAjFFXxxay2cbQ+u9J6hloHyw7E5tbWpFt/PnHh/3+FA0n6XqyrpyrNpnwkVO0M73dkp18NvKijcqNq3FTMpzMiPnbKn9YTjw+PETSs4cD0feeOPTpJT5wltPefb8jlIr0zxZh51qlaBNeUovRBtiZDx4fBwYxpHHjx8xDIOl79ukmXNmmmZKLZzPtrotJbXGCSOrmHcV/Eo/orqazC/6PrV2gofDkRgjNze33N48MpnBEJuzgjkn5GT6JmkdVnAOcRXBUWtqK8w9V7LHS4RYdb9ooczWx3w6n7l/9ow8zYRgfpLOCX6MpqFzQhXHjI1V7xwa46Kvra3RgHdrx71SK/eXs2lGbw646PFEjo9veDQngjiyKM8uZwrKeZ44pXkp8BqaF7OZevSsTuvs5zzDeLDJrRZySaAVKYpv2l0b7wmnukgB0ErJhpMFWncuWcbkOlnqsji2HvbtGlUzuc2ewxgZxkiIgZvXHnM4HcALGgXFJAW6qBn32OO9Q0RaUaOdl1WBUpDeGauBR9VKSJlSKuQCZJAZxFHmiTJdKcU6dd3fnzlfruQ5LVrSXCo6z1RVPvu5N6lVCcFxGEfGIeJEOPhg6flS4TpDytbnfZqRZFrXGCPn8xkfHOMY8cFxvlx4ej5znScuKVFoi0tpuVmt5JKp84S8UJzpgdgWoYP3DN6hogTvGdsi2KRCXWrUMirNwSSlAi5Tc7H9RnECwQshOIYhMI4RQbk6YU+UfLLilYGstbQzdqfUaieXgtSuI7BJwl77ALLSgWwffKCtqGumpERKNgHXohwPR/zrn4LGnnjnmFPiR3/sc3z+zadcpys/+tnPMr31FlrVKidrn7CsMOwEPHKeMBy4uX3E13zd1/Po0SNjg1q65XK98uzZM+Y58eabX+CudSQZx5Gb042t/qJ1RAGYpiuXy4VabBV8vV5Mf+cUkQI4fBi4Od0wDAOPHz3i8eMni64JLL1SSjI2t9n+VM2I9zhfELFWurm0C9see7xPmI1NIGkmXyfmOXF5fsfzL7zJfL3y6Mktj568wTgOxgy1BgfFzVxbSkKcxw3RGBDnzP+1WnHJ8XCwxd7dHefz2YBefYyLgegdp9cfIyFALqT7K2/dPyfVzBvXJ5zmI2McON2cOB4OaG2Td+2FlR3IOg7DiHOO6Xrl/m4m52wsTmdaVBGteGycV7VCzVoKqS0qXXDNL7P5Pncw2xfdtI5CwVK5pdi1x3vP8eaGcTwQx4Enn36N06MbUs3cpzvmMlGxCbbuRSV7vEy0Wo7OTHbPZdcIGy3VAGkRwpypKTdLqYwyW43G9cJ0uaPkzPO75zx79pzLdWKaZ/sIIKfMnJRpSvyNkvncF9608TQODM0h5GY8MMaIRzgiRLAmKdcJSZnxciHNM8dxMNeQG+sKOOfM8+lCKoVLSmQH6h3VCYVKpkCZ0akhVidL4XLA1D9OBIaB4AdEIATHcbTFMc5kQ96ZO0P3mU45M88ZxFNzQUtZJAoaHEP0jGPkcBxBleCE9FH9nff4SOJDMbJLiq5XRLZ5Rjol2n8+eFP7KfrCw+t2oNlpOdcMx63YI8ZoDNCcuDtfuc4Z5xzDYAMOultA09/2JgiuGScH8947HE4cjzetR7tvml3HPCVEPMMwLvKAGCLDMNrnDpFhPCwVlKurggnMxRzoF4lB8MHa6LVbCFYBvq0g7RIJ2R5/T9fg1u9lZ372+CChLNXFtVUmWwFVbyzgFokNWDrQ6vVbCtStbEpVOxPtcQO+iCytYZcaa2lSg3Ggukw5T1aVXVuRVTXg183Puy53Xe7azz5+3OLDLGs1cm+CAEgHkT2tqV3+00BqS1Fu48XL0ZrmbAvsNu7s2hPsNkTCEKlFkYyxu7aUXyy99tjjvUIw6cySXm/nedezdfkMQmuS0G7OtLKAZeaa04DpYc3JR2tdKv5NQWPbS8kKJk1qZK48vjnnlFIJbZypOKQUJFvBlkvCNM8mhasFDYovnrkU5pRJ1RaLVXWpf1mcFtS69VkmpDsutGNWA7aljR+7ZMjiWqAiLRu5ioP6AreUfh2pUKvJFN3q7GPdAN0D7/hdkPfJiQ9fQaTLfwtuXRjYxR1gvd/O6BWWNf2pc2bL4UR4/OQ1nDiTLZRMLTYwQqtyzjkzPnrCa8+ec52u3Dx5jTfffBPQVrHctt18ecbxwBuvv8bxeOT29pav+szXcntzuwGyws00cbx5QkqJ0+0TwnBkmqbW0u+4SALGcUTEWaOGaabWwuVyMXZ2k/73PvBVX/XVfOpTnyKGyO3tIzOnlnV4CQ4voRWwmMn80r2kXQD2EbnHBwlVtdaU00yaZ9I8UUtu+nJHcNLaSTc20lkePkQYRocvlmmcUgUVS+2VgveOOASGMVqaPziGw4g4E/A8f9ZaQCOcjgc0FrRADQPHYcBHR6VStJKrZRlQG/deTL5ArTZkO8vaFqIuBII0f81sVdtaC5oTi2h20c6xmKI4Z78DBMtDUlVxFXItCErIhVBNmjA4ITd7n3EcGY5Ha9MphUu5kEtirjNJk4Fw3ZeXe7xsNBs8VShtpmygz+Y3RQqIOms4MCVjbVuBogJ5nklzIufEPE1crxfmaSZ64bWbI6VWxFt6XmHNlmKSvzJn1FUuQE4m/1EfGJ3DqRK1EgSyg+yE7IWklfvrBRW4psSz69U6eNXKtVjR5DknJpQkHci2Nu0iaLF20XNVnJohYG2ZEe/M3906DnZAu9px1WqA+3KZeP78jjFljjcXxvGAOqEGcxqSWohOGGOgBod3snv8fMLiQwPZFyXV20KvrQ52wW+ib38McF5wYlqZ119/gydPnmxJmnaiG9ArpfD6Vz/nfLlwnSY+/Zmv4a233lo2v8XMADFGbm9uGGLkeDzxxqc+xfF4NC++4BEnpJS4TpNt+9lzHj35FPM8E7wnxmGpzj6Mo/Wg7wNWtXUcuxgD1kX1znF784ibm1u8c4zj0WQJjY3qEgPRzeqTlv50QnUPTd33CXOPl4laK9PUOm1NE2maqDkRPLjo8F4QKtImFe9N+z1E4TB6ayxAYUq1zbmNfRE4DAOnRzeICMPpwG1K1Fq4ThNvvfWU4D2v3T7h5nREiiISYUzEEPAxULRQGohNebb+785SiUWqgdlWoW2sTgVRQhwQH6hpIhfT5tWSyZP5ZYYYGeJoQBazqjM7P5q8ScF7fMuGzKpILTiFSCXmhKCMXijN83Y8jIw3J/CQpZBLJpfEVK+kOptCXndbvD1eMgTEO6Tq4kRgEvBiTUbwzV5OyXO2cVsLGqNlAATm+co8T+ScuV4vXM9n5pQZnPDG4xOKMBwODOOBWpW7+zOXy4VSK9dpZk4ZBMseOgjOUeLAwXuCCKdmC+ackL0jeWEumefnC3PJnOfEm/dnppSttqS9fs6ZK0oRXXStqwKoVZKUapIAbBFZaiE4x2kcieO4MLPLYhRprXMr57PNrYfjzM3tYw7jwTS4g4fgoWSidxyGQJ2DMc0f7V97jy9zvDqQbYCt69WWVAG9eLGnG9fX9ye3+HTd2Lrppa/yAvEM9BnD4nC+MOYDxQR1HE83NkjRF4CsvT+EtehqGA+EMOC9tZR1bTXoA4SiOFcZxsThcMK50LSxASfOJAJNu9dTliYRcE3PtKY2nXPLZ1kFdABpxs3Nfmz7z6QExm71TKnpqJoTww5l93jJqC0F1/uNS2M9Kta0o2c2vLcFpBU8mU0dAi5VxFnbV7bSFmHxbg0aEKBWB5NZZZXGMvV0n/cOiWHptNezC73wql8Slkrmlubv1dnWAY+FbV1Mz6H5wLJUcANLVfiiv2+OJajgRJscoqd5bZA56TdLhISWnnTeZA3qlNqsj4qWvnfLd72Pyz0+aCzznz6c9xbGohV/VVdRX9Da6yda6/ZeINbm3+CNjQRhaDK2Witz8CRvvsu97Wv/EGmf0+dubYVZ3TGku/kUVeZSuObMlDPXZD9xrnXOFXKtFFXrjqctU7FCg8Y8NzsxsM5mreajqIFfx1aJ2B0abOFZSl2aIXWJlKjDhaZrUBu/3smShdkTmZ+s+FDFXjlnSBmXEp7Z2MXaVlNUqtTGyWpjF3VJyS1b6aC3n/TtV/upW+VC0/RZj/TrPDHnTK6VOIzcPHpkr1n9tJpqwRohhDjgvKeq53zJzOlq9lvYWb+1Fppnwfmbpoc1za35Ujrm1AB2kzCogsqRMMQF3GvT/iiBabbtT3NuGlra+5s2UR2u7YNISwM50CCoF2oupGumzLsv3h7vH6pKTiZ5MbmO43hz4vjkBi/C7eMbPvXp14ljpGgi65XaLLPCQci1QIQkmVIVJ5WiCWolV0spOhzeBYYYqCUT3GRMaFHOd2fKlAjO81o8cBwPRO85DiPHGM0RIWfmeiF6Tx0Gq+YuFXpBVq1kKqqttWxrYtAnflGbmAkB1BNCJAZrKqIVas0oLHZBC4CmIKJUKWRaO1xXCE3axDgyDBFiwAVHcZlKJeWZXJJpYgHp1ee9o9gee7xEGCgrJg8opl/33nxjXVtcinNIhTwnczWgWlWTiMlpmpJmHAYe3dxQSuF0KLxWSiM/jMmsohA8YRjQqjwKA+XYF6NYozARRmfNCYJznEIket9cTIRrLpznxFvnK+d55pISb14mA7LiMDeF5i7QwSvdK6jNxXaPopCbdl1TImvFiyNVsw3zzrIz0dl8LYwE75rW11xLVOHu2XNC068f9IbIgKuFQwjcHA7onIkt+7JD2U9OvDqQ1QZkXcLNiaIzTgWnRuo7KoXCAl5bYUaflFbQV5dCjhXYspC0+mInk5YeNFNzW5nG8WAdd4CNMSyoGEO7YT+rCveXBGTz16wdWHdBL6b1C7cEugTYjqFUJZeuL+zCchAXCXHTAreh75wLaW4m7CW1lpf9WA3we/Fm7i601rZq1NDgwIuZ2F8LNe2GInu8f9iFfzZdbEtnPro58tWvP+E4Djx6fMOnP/06cYhc5nueX98il8SYPG5Ss8/xhakKOevSwlKrkmphrhWPVWCPLlBcJsgZp9bM4Dydua+VQxx4/KmRw+2BwXkOw8gxBlDMv1IThEBtCzop1SyAakUrpGoaOat6du3Y2HhROlyIAMQQWiGl9XevjQUS7/BdMIsun1Ulk2S2hbYUgiiII4YB9SdqcFwHx+wKRQvXfCWpdTSTIEhrzbukTvbY432iEzilFOZ5JucCKgTnwQXEmS+qESZCmmZwLVsS2lxQC523HIbIo9sbK94EnBjreb63FrUF8NEzVpsXnQuLM45zIL6zMNY5zDvPYRiIIZBr4TLNzDlzPyfeur/wfLpySYWn14kp1+WoLGHSijLFrP9ca9SwFKAppArzpgjtmjNOZCkg8yKcgjOZg3fE4IFIVZhbe/pSKs+fPkNqNQvBICaZKoVD8Dw6jNRpJgaPd3tT909SfCggm1KiSoJpwtW2stQFMuLUCipq5zJa2qHWls6om7SlruzsknHR1gO+Vwa39KJp91oPdLB+7w2MWsqkncK6ShN6LaM23Y4BU8hFF/P4hSV1DlMcrA4D0O2BDHg7pwY6EZNCaK9CtcdUoRQl5w5kzUpkAe0tZRqkV4TTJAkYkBVF1BjZnIv55+2xx0tEr+L1rXhyiJHDYeQwjhzGgxUyDZFKYioR55VEIFYPTgnRE6Jbcv61sUBWpNn8U7W7bNiCzGHFXBVW9nTZoS4baPkZ1WaBtRqpP5QpNWmN0agtibFV3hvb0husdBcEJ4K6wjtNYX0B2ZbLD/5twWgncpbjbPXYWqsBC7o8qK+XdyC7x0vEkl7XZitVzO9ZQjt/TTZnxZPtXO162mp2jku71lYYTbSaEkfFSaVWx+zn7nWDF4iuyfK8a17lTU7kbJ+0CnQnk3azYq01C1L7cG33F6lbO/XVqTVX0SZJaJcOG0HtNfQxTWtW0KQLPRPqhFIFFUXdw+uH1kpBKc6RciLNM84ZyaNt8WsdPD3B2c17vxOyn6B4ZSCb5pkf/uG/Af6IhrdQN+JxBDFAazk+cyfvjKwqq8YHY1u1VzguMgJ9MBH2CkdYGVmwNEWtKzCsfcJrfiDGwhp8dc7YFt+aDjgfmkShZTRbSmQhZdvE2Idjj0UGQVuFLsC3pYRYuwzZYrc20L4F2103ZNtszTbbSrl57zpBomCl2pk631uF9h4ffXzN18CP/ug7P/eZz8CP/MiXd39eCOeEw80BhxCabOXJ7YnPfPo1juPA6XTk8eNbQvTEgyOMkGtimO6Qq5JKtokuWKHFfBXmSVAcwwhOEqgwXWdSscm2lsIQBkDRwVpSDiFQvXApiUrltsYFtHanK6pCqaiUjVVYpSKo+lUz1+x4nEhLGbKMKUUJPnA6HPHOMTlnCru6LiBVlblkUkkUColKxnanTEqeFKWQypXsKzV60lgoOpg/bclQrUGJ5U88Kor4ir5gI7jHHu8UVa0oeJ4T1+uVnDIxDoxjMOlbCMRhQJxjqoVUki2hpokqBZy1Pz8cDqZ5PRxwj20xWOYLZT5bU5/nBU0XpMJBAochtLHjFzmc91bZbwu0gtbWSa9UpCYDzrlaty+FIQ4cVFCfmavH+7LUgjwgnsBQbG3jExYSKgOldc2rbYGsArMCORPEMXjfHHtai1tnHtApF0pOpJQJzpGnifEw4gczDCyqVqw2jpRj5vHjR2hVwg//2Jf7z7zHRxSvDGTnlPjhH/5vyDow6w0Fs5GKLuJoep6SFmBaFmC6WoJYlX+7vwGJi1/csiq0Eg3npDkGWAeTUrZMKRgjaytQA6OhtbILHA9HYhisa9BwwPvQVpqb1rTL0a3sLHTsLO/ws3+Wo/e9NkcCv75ms83tElEX4L0+KliV9gJkvUBNSL6HugPZvy3i3UDs+z33ZQpxjvE0MvqB2+FI9IHHN8cFyA5j5Hg6WMeeHBgPjlIz8epgyMxlxg+CO9gYu54r17MVfpn2zcb0PCXyJYMKUQJDiLZIjB7XmODqHJc8oxqs97laO8ruBi/NCF7FugNqKW1h69rNFq5L605piz6kafJskgzeG5Btn6u1+zsnUrFtzzkzldn8NKk0ty/yrOSpUlHO5crkZ7R49LYzzorWbJN8yzh58cZIOUFl95Hd4/1Da+U6GZCdrldytszB8XBoqfTA2AqJS5rQbK4FGUjVZAY3t484jOZpfvCeY/CglevzN7neZeYZnkmBdIUKYzwx+KHlSxyiRr7EZlGFKkXsHC6qzLWSOwtcynLOxzAwigdXmKvDlYJWk0mgNl+X0r1hjUhapBSNyCkOk/ZIq6psqf+5WsYxOMeNxuYj65r+waFUUspM18n2uRSmc+BwHDkchiaf8oTxSBwDeqw8efQI0dZJbY9PRHwI14Kmn9NCqskmFqmIM52MloKWvEwGpa/cWvUvbSJaDMU7IwtL29slxdJQX61mT2JMi93W19BAYXc7ENSBNo1cyQUh4735UdLSHAZkZU07Kg9B6IP7LwDcNtGKWAdrs9byiJRFl/sA+CIPE5EdPTcQbzXVNkHaAYPUjJQZqfMr/6k+aPxtTjru8T6x9Dnf3LRJcapasryn95zz0FjNGAKIEkNqxVOFEoUcTbe6ZD+aTrxqMZNzHEjX31nxlWvyBgRw3bm2MbKsi7iqlVLFWjUXW+RmlIKjoPhNk/ZG5CwOBl1m1C0P5IUbNNa2uyTU1XFAGiSmtgIxKjnbxGqenmZRturZXwjdLsT32OO9o2cQ+gSwnMsspy9Ltm+dFOycLa2Cv0txqiLeJHCoufz47kQSPM47cBCCJ8bQCqAdUmXjDmKf4DCA2fOPrjl99FsvBIsKxSuxEUmV2hog9SxjZ2abb4j2T+j6A9tuz7I8pHU2BI+sR29g2GpTLLMiVncigs+FOSXmOeGCQrDCOQVrZBTjg7l6j493vDqQFSG4CBqIEnEabN0n1iNIXAN6DSB6Vu1PbbPiQ43ZKi3YAtwHtRRtcJnEbguCt6/bSAuaLVZnZc183S4UYmZBdL/J2nVwaheNFwdBJ4x7ReYCnmHzWdufDcT2/d6wvGsaRpeWe/Z7Xaja6qE6wZFwekb0y8fI/m1OOu7xXqFKThmKcp+VII5aJoQr4xA4HA/c5ltCDPhW+O9c4DAceeJeo2hu7SMduWQGlwluXrIXi2SvVEoqaDUteM0FHwLH4YZT806O3nxqo4KqMOWEQxgQvLPOYJdp5gqkkjlfJ1LJVPFkiag4hhC5ORwIzuO0EL03Pd2m85g4b1merufF4cQyNSnN1oShZGpjjaKLBB9bwWghp4lZK8/Tmacl4Q4Dh1vPcBPa5G5gQRrbW1tHpcv5YhXVe+zxPtFng+Adx8OI1mpWWR68FKiJkoQiQi0Zp+ZulYudbyrK7AJnafrP4wGv1mUS7wmnGxgGTq9PPK7GjB7HE4fhCFXJkznf1FrJc2ZOpZ3bNmcBRK8EwNVC8RBqIZRK9Y6pVq7JE4A5W6F17+6VSmHKshRyW7d6RSpQOthVmp8fHnNKsDWuadu9E8QH1HuKc0ylwjyjpXC+TlwbI1u1F44J7s3n3KdMiJHTbWY4jJRcW0t4vzOyn6B4ZSArCN4FlGCektq0OOobqWg6F6Gvsnqrya4V1YU5WpWofXW3KfBa0CvQGEsDxMWk5PowgW9G5RhDg1vf28CuZTXMBsupgtjkJlpBSxtsbb24lcjKug3QpW0edE2tLPf7heGhPOEdwHGtlDSbIbY2vWAD+NlZ+06RRJALe/foPV4mzBavmGeIGnBMWSj1jhCE0+mGmWzNQQ4jj/wR5x2jPzAMAaXixFMb+AvuivO94nq95akyXwu19ELGTHQQh2BNQLwjRPNj9aVQz1emaSYgRBcQmo1eStRSmHLm+fncrH086iOI5zSOHIMhbkGJ3ryDsji0eTeLcyZdailPy5DYYjfl1Fp75qV1ZnSR4CJZC/d6IWcltc//wnQmHAde+/QJn4+LO0lvYtKbnqRp5nx3z3S9ftR/8j2+UkIaW9gAVnBCcFaspZrJyeYeawltb5BS0WR62cSFi9o2vFZbKDqH8554OuFq4ZgKj8Tm2pvDicN4QEvl+vzCdLlSUmEqhWmyrOEggdDIl9AKnJ16aoCgjlgqeLPJmrzDK6RcyLUy1UJRZcoZ52hd+5TUipopqzyHZX41gB6aXMg3Ntg7h3hPbdKCqVbKbNeG+2nmep2brteRVZiqUp7e8fwyMYwDjwsc52zNh+LAcTyYRd8en4j4orSo7SerYCC0uwPI5jUsRRGbtEl/z5aNBeAhkF3w5JLqa+nCDfztn9k2Y5POYn/exhHGDCHmG7lWJlvhWM552SdkKy144VjppvMrIJfl9StgXYFrT+m8HcjW2aySUNMLUk0Mn51QHDgp4JMV2eyxx/uEqnULMjNxmxBF4HpN+GB+leP1Sq0V74VcB1uzuT5gTZNqk60SvDVQsCxKdyuwymcfrIOWZssgmOynUmsGHOJDc+MwBse1a0OpFSN1Cylbp69pTlznmSklsyPygCsM3jr5VWc9u/q4FGdWf0tRaKt8rn2go0sWxImjtuzM9nvqcoPaDd2btlY2RZpbMVBvjYs2S7Kyyabsscf7xCo2e/CfRU/Rt8YdrhE43Umgu7NqNWeQnDPzPFsh8+AQb5pyEbFCKSzL2OtNShu7dZn1+rXB7OS604k4k9r4NraqQnDVcpjOmbSgXStKkyhV70jVmY+8tMY+2sZfJ5AUpLWcDs4R2sKw9eFbmhnI8nVUtMrS9n3JdvZFpXPL74pQcmGek12vnH+7jG+Pj3V8CCBrRRC1Qi6OXDMOt6y0+pQhbQjWdorWzshqB6OlTSplAa82zB4Yd9AtcBbJQeUBcF3vb1L+LD4gixzVBmx3HLDPBzXWpvllLvB4kTTp8jkdbG87q3Tytn3oEgugtV/WQdpfUCuaMzSja+urXanOkaOneE8IyuEIMe7Dco/3D62Vy/1lFZgBkPgCV4TKze2J1853DMPAa08eUykMQyQGIUarag4ucHu4pdZCaAWTpVYubkbTZN30bgdcdNSiXO8z86XgpHBNT9G35rWzXRzwVRlTIpSCVyFVR1RrC31/f2aeZ67zzBeePec6z4QQGcYj3gXy7Q3H6BEdCc4zxgEnZj/kQuspXyp3lztQpTeNVRQXhPEwGmgvAZ9LKzytpHkm5cylGb1POTHNM9M0U4NrBakGh0ub0GvJTOcrOSVqLpRpwuXd33mPl40mrxMjNlSgtTwAeq7RrLLGEAzIlkAYIhUlVyt8KihP54lnT5/ivOPxk0c8fu2RZWPUUcVTa+Xp/YWa7+w6kDKaK7VUUq2mi/WecBg5tDHVwWQuGUmQSjIQq5BKxbMWWqdaGRojm0pkGGIrwFRKmzdzKeRUTAdfihFFqPk+Nw2ra6BdBIIXK/GshgVyI3iG4AjHEe89j043HA5W8HZzc2Jo4/tynbi7vyf6wHw4EEOwYrQ9PhHxoVrU0tiXnGxy8s0XzzXB9pqd3wDZWimNySw1U0tqjGhpTI624o5W+KS1AVsrLusrNK3ObEPoTRZsp2pt/nvLirfrUnUp5Oo3A84FqKvxe1+5vs1aZC0o65eclV3esMwvrAMfFoe1r64zz6q4Wqx6W9U0fH2FexgoIRBHTxkiQ9z1Pnu8f9RauZ6vyzmsqpR8YZ6fUWvi5v7EJV0ZxoFcE4fTwKEeGIfASWLzk/TchBNL22RVcs0UKnNN4CrjTSQcoslr3BVcRasyX58zne9Md3u4IQ4H0/opBBRfIWeIBaZ55unzZ1yvE5frxOfeesplmhjiwOl0QwwRR+WNmxNBQOIA0dwRgnPgrYjsWi5crmdqrUTvGWLTtnohDoOxQ6lY8VpV0nxlSpmUG4Dtt5SY5xmGxkC3EVtalialzP39HfP5gqgSVHH7+nKPDxSrtruX927oDsDS7DGaFVfQyqCRqsr99cp0PVNK4TpPXOepFXY5Djc3iAhFxTTmqtyfr1zu7w0M4ggY25qrog5wjjAMjIcjTsxz1ouQSkLJeFG82HztW2YDtfenWgldI1srQw3NmUjMXgusMCtn69SXc+vQpYTWRrcDWXEOVK0jXy0tU7JmbIN3+ODNZu904Hg4EmLk9tEt4+HANF25u7vj7u6OGDx1nhlipO5A9hMTH6JFbSXnmZwdKVXmYrY0uLykDNaVZq9EbO3zmrVWqYVaU5t0C6qN3ZBmQ0W32KnLax4CWel4ugFOXZosrAzoCmR7SMe3JuSxo2lAuncSW5s1sKQveeH+yq2+O5B9qODdvKID4lqRbkCNmiG02CQt0SOh3Xbh+h4vE8qyCOunYq1WAFZKYppm6/xTKufzhfPl2jRsA8EbK+Od2KSHLcS8N3Yo+kT0garaUog2Dfvg8FGoBao3Q3Oc6XRrG9OFNZvaOcxMpQh2c1C9UL2jOKGgOLWUaBcRvW3k6TraLM1okoMeldWxYPtu0/+aS8KcrbPQXAoKuBBwwTezevskLbYoqLmZTuvqvLCr8PZ42ehESm+L3ucuoI0nZ7K3JaPYSJcuSRBLwy8d6to4r7mQk827VvPYZDSNHBGFKs3Wbtn2QweDXozWf5pns6OILo8b9rW5PYijii3yEGtOEtRY5j7vO8WcP0RwauMZrBNfjOGhw0jT4FfphFEfZ0Jo+xKCI/h2c6vswnWyqLFOWqtJgHZxwScmXhnIlpJ59uxzXBI8PcOUzaoj+mA9zulOkGoG5xvNStfvGWhsKzAtqNoKyiakrhXoelltLWlbE4RmmwPtgrBK/JaQBzn/9thK0i6vF3rHsdJWhh1Yd8uwdxgQD4DyC1B1Rarv8PjD57QdPyLU6FAvuGFkeP0J4XQiBOFw8gxB3ra9PfZ4MRQrxnAii1Zszsrl7so8Xbnczzx/esZ5z/2zM2lOHI4HXnvthk996gkxmmXPMMRFb3c63KIoPoyM8UjRwmW+cElX68ZTA+Jt0syhUAbT01YuTDq1jls2+UgVkjXJImtlOjiyGyjRBAGuuQDMCJnK7LR5ULI07ANL8+fUrgUC4/HIksFpTFEuhWme26LUOoUVLVzTzN3lwpRmPv/sGZ998y2KQD4GjjePiaeROA54hFIq092VdJ1N/18qQ/DGyNI6Gu2xx/uEtgWmuNaoR+zcStlapVsRcqsHkcAwNmDZ26yr4pt23YmQcAZ4i3K9v/L0809x1uPcvNxFkOrQ1AkZR22ex06tCNoKrVq+VAVpGnavyiAe7xQqeBJZrVPY4D3e6rkZG7tcqjZZQbfTa81IfCEF1xjWSK0jIhDjsFxfYF2U5pKs7T0rmJWunxUDrcfDwDiYI8HohehMuzs4+26CiBFDOe849hMUrwxkay3c37/JeVKe3hWuSfHOMTjfGNmKa64CnY2186qr2ITuTqALc1KbhtXSggC6AbKL5Za27pYVtgLVdYXXfwea7qivPt/++laNvGheWVoIrobo6xK5g+PuiNDXqtbNy4QUHchuAe2L8117ZWvrqebsEDwaPfEYiY9vGR89JngYBwg7IfsVEb/qV/2q93z+O7/zO7+kn9+7WYmzdpderNPOdJ65Xq5NfW6rvjQnEDgcR+b0Gn4UxjEyDiMHrXjvOY5HxvHYJpTIEAZKzYgohdQ643loQHYWJTWZQUrG/DoEXKC0Fpm52qgpQUiDUFykhlbQkge0FPI8Q60kV42lcRsg25imnBJVFT9EhsOIOMgpGcAtBmTnebYUpfMgjooy58x5unKdZ57en/nC8zsIjsPta4yPb4iHgRBNIlWrki4Tl7uLpV+jFas4FE9d/Hj32OP9YimUFJszslYuyZoJOLFGjgKEaNX9C1Oq1g3PixCdt+IoMSCrqqTrzL3c47zncDoxHqP5uVdBszGblWoFj4BvZj6+T4VaG1Vs+QUHRHF48VSp+Kb2c1ia30Erugp0L+famyC0RiSqSvaZsthv0VQVwjAMDMPQ3ru2u07ZkxYgu44rL8a6OucYYiAGc0OJzlq8m/uDyY28iB1P0bdPunt8bONDamR7JRVLKmB1IliZyYeJPV1lA62X+Zpr7+BPXsjab7ah71CNqC1Vo0tt1/Kzg1j6/oiaWXN/fU/xsO5G18X299trbXW4ZYXWz+gPdE2tbo7pHUI223YecWqr6cOIjIEwHghDJARvVmF1LYTbY4/3i97KeeE7Nie0pdVt8ThdZ853F0op3NyOTNcJwfqWx2iXhlLr4kjnxBHF4xxEFxh8oIgje9smNHP2aClOh6C9lbMXut5I/TouXDQWKThhqBUJHs3WlIBS8SHYBKm9V31dnBBqtUWg036My7KytcFsH6Kmr1fMNmhOs5mpp0RuVeDSil98iDjvKbkwXa7UXHHamSjH8RiI0RnrO0/W9WuPPV4i1hbrlorvLK15ILfGBMIC7IxEsQKt2hZvtdbmd27tzbX5oDc0ScmZNCc736s20GzZmehaX7xGwBgD/GCiXX48fF3zfW0SAkEWSzrEMqPS5m5jlGWV9zX9K4uMQRgagyoiTanTgKu37mM25W+uEV1L2ySLPetSckYRSkq2APAeL2LFYUtB9x6fhPgQrgWC4BFtPnCYHY5o7xFinTiW9PtyUm0hpi4ArQ+cbl8lrS9zn4cFbdvuIHGzPXm4XVRpLlsLYN3A2nVfxPRGgpmzd0eEKjYB28foA0C+/Vx94ETQZfvvAl5bl6NFHyWCeCEcIm4I+CFweP0xw80RFyLD4USIo7UqvL9nTrtf5R7vHyJizQ4AodoiT+tSP1mSGYyXWsmpcL27EGIgXa+Mo3A8Hbh9fIM6JYSAiCfIgHeewXlOMVK14lEG391KPOiF0jMqg6BV8VmXNtKusUqqoN7srRyBw3FANHIscMo3xvCmTLlOaKmcwkgVYcoFp8LETBBHLsWaJxjFhcve0raqBO9QJyQx9iircp2NgZ1L5gtP3+Rzz54y58w1J4gBN0SG04mbx1b9fX76nGdf+ALRe56cbnn90S2H44FPfdUb3Dy64Xo58/nP/gjnu+cf6d97j6+MEBFCCGizggOhVKzoMGWCEzQYYMytwl9QK0hMs7VZTok5zSZRKIXR2/TtxSMVcyy5O3P3/B7VSpkSg4t457gZR45xBJTaPJVNDw/m7yqtu6zNYeYk4MwpJIQFaBfnmuuCmFc8Yk4IyRx/aiuCtE7xDlpth3M9+wkxxkUjW7Vb3SmlOnKMdEb27YRqk/zlTFWYrldKI7Y8ys3hwOA9t8cDYwyE8OHdRff4yogP95fW3vm8Fz90VzhammIFf6IP2cwXBQf2sk2LyWXLDR62iXDlWw2QKg82Sm+agC5ErX2aqnlesjYsYPMZimnojP2V1tpWF+D54qhaHn/w4Mqayub/h6p6QV37LoLHHUfCcSAeRm4/9RrHx4/aRcJM44smrmmmnM/v/nfYY48WVpzljKXs+pstI5sr8zSTciFdZ87P7vHecTh6PvXVt6Q8I0EYTgNRI4M/kH0BB34YOfrYRkXBOyWVbAVTJZPVZAAEY1ClgOspvlLQdr+6JtcRiBIIMoA6tHpQR0mZ+TJQc2FUj1Yx6yEKmQwNyOZidj6ueGotrbOgVX2r6qI5RK3D1/3ljjlnnp/veHZ3R6qFuVbUewiBOI6MxxM5zdw9fZO7529xczjw6eMNr52OnG5v+LrPfIYnr7/G82dPme/vKNP0Ef/F9/hKiM74G0FiLj5VlZSt4FC9dZ3EmXa21kIpUHImT9bUI6VETk0q02QGi7ytgmplnieu8wxqEpjozAf6NB65ORxBK2m+knOyRafY+7SnJjEGtc/AvmlPFW0d9Ty1ZWecwV1zIWoZk6VAmlYY5nvr6uaH2wB9aM4itQrFNBIU3xqb0Dyb231lZajn2cZ9KZVL61DmfWA8nBhiZBwij25OHIdhb4jwCYoPuWTpFEvXo9RN6r/LDDZ1/AtN2n7txVTomgZoKcLF4ErXTmA95bBRMtg+9MrHJVZm1GCvLun8hU+VtWXuoi/o71p1BeuWNq/pvz7ULNiLRdfPW7dgP9S1/YwBCeaDGQ8H4nEkDMFSNa1zkJnaQ5mu5Hmm7K0w93jJEPeCNKdV84t3uOAIMfQZzLrJVSXNmfP9FQTGw8H0swg5pFUTq56i5mdgbM0AeIY4M5ZC0GKG7sXGrOia9lS1ss8+QfUMhkqlUpZ0oVYx25yakZZC1eqXZW+Riki7Jlge0w5Tu4dlkyypaWRTLeRayFrJYibuVdo6u03FQZQQbAIvKVlL0GK2eKL2XBzs5r3iXMU5NUZrdxPZ4yXCtOsZxVFdMCJEdXUJ8Z4Y/FKJzzblDosUwDed99ZHUpwH71FVgvMLgPONUQ2tsr/lM1th2ZZbWQFj1Q25tPAwJiWo7T1OaNrwVf7Q27uLrnN5VxSZLGkjD2j7In1jtb14mee7jK+DWgPatdbN/E3zvrXvK/hW7OXW73CXFnxy4kM1ROh2WLUks6ixM9xOQjUngiZM7TMHvWts28IqLeiaGhGohd7Pa2ufUxuCNRArJl2Q5m+3tIXd7mEbGFv1wXLrLNXqWccD26JVSNB1Rg8YWFkvABZNI+gwDS4rEYZADQ71Dgme+OgWfzrgY+TmyS2H09EGplbqdSanzOX5HfN1gpzgegd5fvU/1Zcg3quo6Tu/8zvf9/k9vkQh4IeWWahtHI4Odwh4IuMgcHTUWpjPievzK6VUnr915of/+o9xOAzMcyGOgfEwGNAbPEED3ldCtfHm/MgpnkyOEw4cDhO5JJ5Pz7nM91RV5lTJrU1lCbXpAQuzXNGcUCDViaLJWt1ezLRdMrgZ8+VSs/sqeJLzTEHJfbyLyQlUHaU0P+mSqMVSsc+vF55PV3It3Gvh4pTkIUVHHQNUq84OYt2Nakqc33pKLYVynZFSCSLc3HievDEwHjxxSIi74P3MED2H4fBl+bO+zHj6SnvN13wN/OiPvutL+Mxn4Nu//cu3v1/KqLVwf3eHuICEaBNFLYzBXAbGGDgdD1a05J21L1dbFFqaX6A5Ai3kS0OjLkZcjFZI5gTvbSHrxbzdvXOM3rEY33R13WYOE1gKj4Ulr4qK2ethu4ItR21erq2BUMmFmlLrFuisWYk4nAdz7usdx+xnt9BqsyvqAZRczLNWl5m7g2tn0gOgiFAAxDFGRwhC8IHj4cgwDMTgOYTI4P3bEqZfivhyjZWXKSL+Yu3Le43Lz3wGfuRHvrzXkZeJD1XstV2NaTUfPHXNakvr4hdr4hvotKhJhLSl8Vcw2DHmlhbtzRDWns0sQBYamKX1D1s28GAn1403cPwA7HZHAt28VlfBwlbmsOzni5/RitdE+8rTjrntuQHpxohJCPjTgfj4EWEIHF97wvHmhOZMvr+nThPlOnF5+pTL3T2ihVBmvO7mznu8f4iA+LbQ6yRHEPMkLp4YBdfYUapyvTOm43KZePNzz4lD4Hgzcv7qR1QtjMeBKY9UCcw1kHTA4xndgTEebaj6SIwzqaY2Xq3xgMkPLMOQWyMUKdaSVqpVZheyOSnUTJ6v1FRwxRFTwFdnTK1aarM4JSHWrtYFfIjQUrXarkc5F1KaKdVsti7ZCrqutTCLkh0Ubw4hqBJ8bI4GQi2Z+VxMh5cyrlp92jh6TreBODhCLIjMOFcaC7Tr8F413gvEvszzX0lRqzJNE84XvGIpd1Wid+DhMESO40Boj9NIoqVlOs2pwK/Nhjrx40PEt8YfddMh04vDt/cF52gCAdPPP0jaNOazb1d7IseuIc51/9kmCVKrJ9FSqEWpOVNLbo2IhBCNwfUevG8Als3+dncBWR+3j+7VKg8LxEtzNpKNXMgJRO/xOEIwScEQI9F7ou/b/1L+RT++8V7j7m/XMfmhrsLaUugLgFxWefZ882ReomNJaXdUZGEvt/KANTUvi1ShVzwu98Xcai3tYStAe/NmeDaJgmw+04m1AOyWW7W6VeLgdEPI6ma/2l5tUj3rMNseuD1aGwVcvaDS7MgOA34YcDEwHsyn0nmPlkq+zmjOpMuVfDEpQZ0zFLP3cfXtSuA99nj3aIxJsfaQvcNOThkJ4KMxmmGIxKEbkztKUVyuXK+Ju+cXci4Mw8DhdCCXyCAD1xAJLiAuIs4uH7U5HXhxxBA51COlVkQLSUrTyxacWDFYyJ5ag8lwxHwya3XIYDIDQdG5kmuFCq44vGZiiOAEj4eqSDbbIOccfrIxV3IipQnVynVOJDUGSYInuhFXK4dSSA340my5lvHqsIl6CKhmbh8diePQUi1mAF9qsU6ApXcp3GOP9w+bq9ZWsA5ZbBVjaADMO3Mn6O+hkykPFG9bXYDd1dpkA9LYTmta0K2rzItVFmlBl9nVRYJnrWVrlaUwVNr8WTq41K3mvpE3Yt64zrUuZU42U6Is+/52+d/6nUgDytKyuX0ubTlZ24+W2RUxt4SVN3Z2bHTc2giyJR26xychXh3I2nnU9HftOm/ZPlrhI66DVCuHpJ9qixGyUZXtMZbntyxnWxfaa1zvcOIY4mBsirAA2s7cdqa3e9RCn4dsdeibqbTVwzysjuwgdhWat9+xJhAptQYGi20YVhjWLg6W/mgyiTFCHHDBc3r8mPH2Fh8C480Nw/GE1kq6XDg/P1PmmetbT0n3Z7MWmSYkZ/OrlKY12mOP9w0FnamlMk2JkiuX+wt3z89MlyvHRyOPn5wI0SM4aoI8G4szT4U8Vz7/Y88oVIZDMIcDMUCbHieSZrwP3AyZU8mmlQ3RrG+8w41POMUbqlZzCUjZGijkibkkcsk4rJWs4Ah4HI6aCtkf0JyZ7ifuLs9IaYYk3F3voQjj4cgNj/E+UObMPCVrdlBXr+l+XwENDg0OnBBPRx7fHKx5wuPHPJ5mkyzUstgfmXjQGKg4REL0DAfP7WtHiIHqPCkXlMR1SlynzHXatet7vH84JwxDxHvPMHizemvaWBtDgSFGxDlyykwNL3YGs0v01hlym2lURAsOGILgfQSat6pYQVZQm+y1ykImdU1sv11LJTepHcXY1wV4Lw4DxT67Wc9K05ozBJM2eBtv1gLXQGqf23tdyWLj1fS9XWduZNIK2LuNV214QgAfAuZrIPSWS9556/SFuaPUnCmlyxr3+CTEh8+LtZNZ2iRgJ6suC8YuAVjBbHuTroD2oXPBerezr/3BLhb3zjPEgSEOyGZwAy0d02y0liYLQi9gdM4Rgmcxcl56Oj9EiouP30bWkNRa/dWu+6WtGNugQoz9yYhZAgWPHCIuBMLtieOTR3gfOBxPDMNITon5+T3z3Zk8z1yf3TPfWW9sp60fmijO6bL/n7T4cmrbPmod3RcnGhtbMyXZwmueEtN1ZrrOjKdAGDzDIZJTZThmnC+kKVlFsFbu766oK8QhcDgN3L52ZDwk/BAIh0gIwVw1xOOdx3mHkwgC3vmWlah4JqJL5JrN79KBd0IpEQCPI0o0P0xXyNVZmjIZM5tKomSlTBXNkARkPOBR5nnicn82fV4plJKXiaszQOE4Ek4jzgXGIXK4vQER/HhgTJmqlZQTuRgY7WySD4HT7YnheMAHGE4VdVa1nWuFUsz1IRdy3v2d93i5sHS7JwZjXr13jDG2iv4G6ETQsrptAKvGFB7MUw+Eea2pQXCtVbyYq4FvTRBcMb9lZS3AKhsJX9VKqoVUrfizZvOP9k4YghWMLTLClra0gmkxH/RmKbZ45W6L0ZYULevzLYzNdcv342Srj6XZY1rHsZ599aHXn5i0sLeplbZfWq04bCtb3OPjHR8KyAqWhjCRWztpal0qorUnSLqf7IaVXbfRQ9ezV3rl4oNkyvrKtjKsvSisyQ66dqczqAsjq0rtNq9VKYVWJNIyJct+yGZXmkZXWRmeTdf3rYqn62C1pzmdUajhMOCPIz4EfPQNv1fSdKWmZDZD5zP5eqGkjOYEujb0fXD8+5jc4yVCVQ3ATonr+UKaC/NlaoCvNmLHWWce7whDAMwL0rXJqhZlvmZqUc53E3dvnUmHzBAHDscDsRSChNZtxwNW+CliE7U43wpSHN4Fm1Sr2XY5hOKH1qnIM7jBJl5fQSKUSlQhP7owhsB8SZzLRJkrGpRZTbs61Zlrna2tNIq6lmx0HucDzgnxdGC4OeBD4HA6cjgc2mSYcM5Ta7UGDtXUg70YxXlPHEZiHHAenBS6IwsEVD0ihRAHwrBr1/d4iWjnlTUS8A/sqLqDTlE7x3ItzE26olpbi2VdCreks5tuna+6ID5jRWIioK5aYwDAVcW37WS1wmnLYUjr9rfJUFYrvLLPldYtq8sJWBimngk1lYEuDPL21smnPrW++PwWmDfCFpZ9WwGza58XnW/worsOteYM7TVOWDKye3xy4kMVe7kqSBE768qa0m/tN5Z0X/eXbdAXWe4/3GB3JwA21ZTvcEK2/so9qbeu8lbtji02VwVQf0xE8HUzANunPACym7t9kEJjeJtiaEXdLQUixsL6MeDGiATP4ckjxse3iPPEGHHeesTfPX9OOk/UnJmf35EvV7RUdJ5xtfRDXKRG+7Dc42Wjlsr1/sL57sIXPvuU62Vinmams4HZUtXSfSEQD8rhkbEv4syzsmQhp8I8zThnBSN1zsbg5oJERxwCU7pyf3iOF8/oDwx+wLvA8XDLOBwtXeoiIYxULTgn5BrJJRPxpJwILnAMB6KLeISxFaecz895bRy5Xi88fXbHj/jPcblMFArP9UzNjilduMzPKaUQ/cAQR5zzHI4HxtMN3ntONydOtydC8Nw+esSjR48A4TJNTNeZqpVZM5m8LmbFwPB4OBKHwSROLoEWwEGNgMO5yOFYQONH+efe4yskun9qbBICH3xzGLBW0rVCKhlV5TzP5rZRKjllUrKWr6MPjCGaDjYGkweJIMX0rBXlmmamPINADIHgW1taBdcZ2NKcekTI0oGsTeGlWovrOSVyLgsT2y28jP10bdFqi1RVpVSaFoKlnsV7hwu+uf60zKesXcG2gLYD1tWGs5rkB2vNG4LRsJHQsqS0JkZtTm9aYJNZWCHrLi345MSHkxbUJmWtjU2tNO+3futsRQelbRXHOxUvNXp0w0A+tLdawZxWQUuhLKu9rr1dP2sLilffyvZ+3XrMbVeF24H0wt7pRnMrW30sK+IUzB92iLjoGY4jh5tjK0yT5TuZLxcuz55Tc6Hcn6nTbHKCunY5e5G57o4re7w9Ph6SgC9OqEKaEtNl5vL8zPneWtCWlBvbArjGyAZPHAM1VNJkEgGtSkqV+WrLxPt4sfTiIXL7+g1PrhdKDajLFDfjxJFdYpamWfcB5yPeeWIYCT6i6kAzoUIWh8baKqkDp3gkurh0DYviOASPlonrNaJUvvAsMGsiF2WaE7nCVK6cy9UK2rzg/Gi2Q2Mg3BwIITDeHjne3hCC53R7w+l0A7AUilZVPJlEebBUdM4xjiMhDFixi5jXLQJY8wYnEONIPfRrzR57vHc432Q4PjRtrGlne6au1EKt1mRkKtnkK8lkQd3CjtpcBLovdK/96GxrzkzzjIjJ7EpoGlkalaS0AmtLy9fGBlfcQrrWWlvDEdOZZhFzQpFuddlEDm4lhB76DHT9rDTQ2x+T5ecDEAsLHStNEqAbVlVai1yco7s2dwa7f6L18OtPvnMmd4+Pb+zeMa8aHwRUfggA+nHGrjsA3WMbX/Zz/e0JmFfbzL7C3ON9Yj9DPnxsv8Mdpu6xDXlVNkFEfgz4a1/c3dnjJeIbVfWrvlgb2/+OH3nsf8+PR+x/x49X7H/Pr/zY/4Yfr3jXv+crA9k99thjjz322GOPPfb4KOMTauq0xx577LHHHnvsscdXeuxAdo899thjjz322GOPr8h4NSAr8kOIfPodHv9HEHn3Cp6X2/YJkT+IyF9G5C8i8p0vPP/zEfnB9tzvaY/9ZES+D5G/gMjPbI8FRP4wIqf3+Kxfh8jPfp/9+T8g8g9+wGP4dkR+4wd6z6uGyL+KyN//ZfmsPb4iQ4RfLsJfEuF3f5k+738lwn/VXOk+vXlcRPj17bm/IMJPb4//ZBG+rz32M9tjQYQ/LMK7jl8Rfp0IP7vd/1ki/EUR/rwIxy/Rcf2rIuxjbY8PHfuY/KId1z4m9+i2Uh/wBj+k8OlXeu/7b/uk8Pe1+4PCH1P4H7fff6LCn1N4vf3+1e3nv6bwLQrfoPD72mO/TOHb3+NzPqXwvR9iP/17PPftCr/xS/L9vP2zvlHhP/qyfNZ++4q8gf5l0G94h8fDl+jz/i7QbwL9IdBPbx7/h0H//ebe/veA/sn2+L8G+i2g3wD6+9pjvwz029/jMz4F+r2b338z6D/9pTxG0G8E3cfafvvQt31MftGOax+T++19GFmRm8aO/v8Q+QFE/onNs78MkT+LyPcj8lPa61cmUuS7EfnNiPwZRP4KIj+nPf53IvKnEPnzjUH9iS8g6zOq/2m7PwN/FviG9uz/DPi/o/pme/6z7fEEnNotIfIa8HOB3/EeR/fzgP9gc6z/MiJ/uh3nb9mY3n03It/W7v8QIv9XRP4s8I8j8j2I/OvtWH4Akf/BO3yHPxeRP4nIn2sM8Wfa49+ByL/ZtvFXEfnlm/f805vv6N9AxLfbd7fP+X5EfkX7Dv4a8ClEvuY9jnWPT2iI8JuBbwb+fRF+hQjfIcLvFOGPA79ThG8S4T9pzMsfEeHHtff9BBG+V4TvF+HXiHDXHv9aEf5oY1l+QISf9eJnqvLnVPmhd9idfxT4He3a873AayJ8LS+MXxFe4wOMXxF+CfDzgf+jCL9bhG8V4Y+J8AeAHxThIMJva8fy50T4+9r7TiL82yL8oAj/HxH+pAh/t1gzo+9ux/f9IvyKdlx/DfiUCPtY2+OVYx+T+5jc44sc74l04ecpfNfm9yft5w8p/LJ2/59V+K3t/spEwncr/AetochPVPgbCgeF36DwT7XXDArH9/j81xT+qsI3t99/v8L/TeGPK3yvwj/UHv9xCt+j8CcUfprCr1X41vc5tt+u8HM3v7+xuf87l+fsOL5tc9z/4uZ137N8P/CzFX7gHb6H17W5Qyj8EoVf2+5/h8J/rjAqfFrh8wpR4e9Q+HcVYnvdb1L4hQo/Q+E/fvDdrPe/S+HnfdSrov32t+dty8KAfgfo94Ee2+//Lug/0+7/YtDf3+7/e6C/oN3/paB37f4/D/qr230P+uhlPnezzW/Z/P5HQP9u0B8H+j2gfwL0p4H+WtBvfZ9j+u2gP3fz+3eDflu7/62g96A/frPP/2a7/1NA/zroAfRfAP032uP/XdDc9udngP7Hm22/trn/XaD7WNtvH+q2j8l9TO63L97t/TSy3w/8DxsL+bNQfbp57v/dfn4f8E3v8v5/G9WK6n8J/FXgpwB/AvjfIvK/Br4R1cs7vlMkAP8W8OtR/avt0QD8ROBbgV8AfBcir6H611H9VlR/JnDGGNy/hMjvROT3IvKT3uETvhb4sc3vf19jTr8f+PuBv/Ndjun3vvD7vwWA6h8FHjc2eBvfAPyHbbu/8oXt/kFUJ1Q/B3wW+AzwDwA/A/jTiPz59vs3Y9/fNyPyGxD5h4Bnm+18Fvi6d9nfPfZ4Mf6AKn3c/Uzg97T7vxP4ls3j/067/3s27/3TwC8S4TuAn6rK8w+7M6r8dVW+VZUH47exVL9XhJcZvy/Gn1Llv273vwX4Xe2z/jLmBfmT2uP/z/b4DwB/ob3+rwLfLMJvEGEfa3t8OWIfk/uY3OMV472BrOpfAX46Bmh/DSL/8ubZqf0svHuHMH3b76q/B/hHgAvwh96jUOm3AP8lqr9u89jfAP4AqgnV/xr4Kxiw3cb/CfiXgF8O/FbgXwT+9++w/QtwAEDkAPwm4NtQ/anAdy3PvT3u3/cYH8ZvAH5j2+7//IXtTpv7/XsU4Lej+t9vt5+M6ndgcor/HvA9wC9tx9bj0I5njz1eJl48h186VPmjwM8Gfhj4bhF+4Qd4+w8D/63N79/QHtvGBx+/7xwf5hj3sbbHlzv2MfkesY/JPd4r3k8j+3XAGdXfBfwrGKj9IPGPI+IQ+QkYq/hfIGLsouqvB/6/wE97h8/9NcAT4J974Znfj7GxYK4JPwlbqfX3/b3A32wM8Amo7fZOVZZ/Cfhvt/t98H0OkVvg2z7AMf4T7bO/BXj6AmtNO45+YfhnXmJ7fwT4NkS+um33DUS+sR2vQ/X3YReV7d/iJwE/8AH2eY89evznwD/Z7v9TwB9r978X07yxeR4RvhH4UVW+C5tMPsg14Q8Av1CsUvrvAZ6q8rc22/57gb+pygcdv+8Xfww7NhqT9OOA/wL445iODxH+O8BPbfc/DThV9rG2x0cR+5hkH5N7vHy8G5Pa46cC/woiFRN//y8+4Pb/OvCngMfAL0X1isjPB/6niCTgR4D/84N3iHwD8KuBvwz8Wazm6jei+luB/xD4HyHygxiD+StR/Xx7n2AneC9I+y3A727H+E77/QcxhvS3ovoWIt+FDYYfwVI1LxtXRP4cEIFf/A7Pfwfw7yDyJvCfAD/+Pbem+oOI/EvAf4SIw773/yW24vxt7TGA/w0AIhG7ePyZD7DPe+zR45cBv02EX4mlBX9Re/yfA36XCL8aK+DoC7RvBX6lCAm4g7ezPyL8coy1+RrgL4jwh1T5JcAfAv5h4L/C0pW/aPOeVx+/7x+/Cfh/iPD9QAa+XZVJhN8E/HYRfhC73vzFdpxf376TB2NNhH2s7fHliH1M7mNyjw8QX7oWtSLfDfx7qP6/vjQf8EUIkf8M+DmovvWK7/8e4F9A9aMbRCL/GPDTUf3ffWT7sMfHLsS8Ii+qqAj/JPALVPlHP+r92oYI/xnwc1R56xXf74GoylWEnwD8YeAnqzK/y+v/MeCnq7KPtT2+7LGPyXd8/T4m93hfRvbjHv88ltJ46yPejw8TAfi1H/VO7PGxi58B/MbGyrzFO2cbPur4sOP3BPynjdUR4J99twmzxT7W9vgoYx+Tb499TO7xJWRk99hjjz322GOPPfbY40sYr9aido899thjjz322GOPPT7i2IHsHnvssccee+yxxx5fkbED2T322GOPPfbYY489viJjB7J77LHHHnvssccee3xFxiu7Fjzyop+ODhHwYuWF2/vQfyoiDTGLvcbJ9nlAtL1y83h7steiabu//K5vb6H1wlaX560f72YbvNN7HTgPODR4iBGcQ1OizjPUak7QYu+tKtT2IU4rohWAokKp7XPF2QH3L0fszUWViiIKgiKqttdq92mf8+KBCPD5op9T1a96x0N/hfj0pz+t3/RN3/TF2tweHzC+7/u+74v69xwPo55ub9YHXijmfDjS+mjZ/m/3zJb54asejBrRB+NXxAaWog/GqN0Re+fy9n5lWP+zsbkdmfrirr89tL9z+/vm180G3u1+/5CH+7c9Yt28bL0YPfhcIKe8j8uPUXyxx6VzTr3zAGgfW+36D4p3juAdToRxiJyOI945nIBzD8ffcuZtxpVuxg1om3KE7Sh+r3iZkm9VqNW2X6qSS0VVqbWSq/W8F+dwztP8321qF/DeEbzHibT7hh22k/qL+/riPr3nsWzHd/v5I59/ylt355f9Ct439jH50cZ7jclXBrKfjo7v+HEnRg+3gzJ4GB3cBkd04FACFSdKcDAGcA6GAIfB7gvVei2I0v8hII4F+dYKVZWqMCclFagKOUOpoCpouxz0CbifuQVrQVJq5TJXUmnbKSxgE0AFCAfq+AbqR/T116lf+7XoYST96GeZ/toPUa9XrgL3DgrCuTrOxVNVOaQrh3RGq/JscjydHFUddRjQOKDOGTCOkarKfUpcSsHVylgSsdp9P8+4UlCU2akBZRSnC9bnd3wh/bVX/Zu9U3zTN30Tf+bP7F7SH1WIyBf173m6veEf+J/8gxsA9hDA9YkHbGJwgCDtvo0f5wLeRwS3rCoF0NbUR0QRV+0mig8F521SyzlTSkUrlCJotVsprs1ZDq0ekGVflErVSiozqnb+11rQtjjsC9H1OLDJW9fnat0e33qctdb3vK92gbHvafMhC3RQpZRC7dsu5cH32L/fH/tbP7aPy49RfLHHpXeeN177FIpQxUYdWpGSEJQnp5Gveu3EcYj8hB/3dfxdf8dP4PHtDccBbg+Cc1CpVEo73z21elQhF6WUBmAlARknMAYhuhU0958rwbPu39uArOvEi9rYAnKuXK+ZXCr3l4nPP71nmhN308ybdxdSqcTDkfH0CO+DgWkVnAivP7nl068/YoiBN54ceePJDcEJ5BnyjABGI1nU7T514ktkuVYtB9CuAdTNtaK98xf/X7771f9g7xD7mPxo473G5CsD2eDgjQMMHm4iRGe3Y1CC0FhGmxy8WxZoQAOfFTtBOwOJAdbOXkobgKUa+OzgNed2v0Cu9v7atomosb3tw0o7pXO1W2m33MDwlv+pKFUy6oR8d2b+wlvUGCnPnlHnmZoTSaB20FwdlGK/lYy2DYp4gg8UHMUPzD6iIswKKWeqKnMppFLxWomqC0jtzHVtP1U6a/s+q9E99nghBEHRhVlVtfvvZLen6DI9PAC+7fGHOY5qo0oLqqVNVgXR9nhnZx1ImwDtJNa2YtS2ZdeWrn1FWVvCQto532G2AVVZRqq0+VXb/ukCaJdJ7QHSrZtUTE+V6Jr9aKBU+kJ6/RgejLr+ns335mTdqz32eKVoQ6KqknMhiZCzgcVSK+DxPuAdFMpyjisOtAFiUar0894gYAVcGwaCGLOLbCYTG2fbc365L9KZJnuxtzeJKE4LvlR8dfih4Am4LKgkKoVcQKeEc4XgPNF7xDm8c4zDwDhEhjgQ2jGhAcHGqBfwbaQv47F/R53h3VyPbM5tY7dhAVsgNFprnzQ/MfHKQHbw8OOfGKA9esE7A5HB20RWayXnukyg/QQUsLEmNiWidrqWWim1n4DSs/CUaoutCkyZhZFNDZxquwj0CcU5lvmnbrYxJXt9rTA3QKuygtmSC3m+UslcLolnT+/JTvDpSpjukWqsTN8XrWKUL7Tp2AaRk0iMR5w4LnHgEgcSyptp4tl8tQFbK74qA8qIcsDmeQ8G+sEuTD1FxMuniPbYY2EwVJbJoAPaVTKwpilXeNnXlhXV2l7bUyQdxBZAqZqgZJssWlbFGCFT6HRmlLpZpbWfKtIyKdvQhRVWwHWgq4pQ6WvcPhdXUaR2UK2NVeVBqpJ2HLoA2sbzaF1ftwG40kDt+iVuwXB98Pl2gdpH5R4fJtbRl0vhOs3UUrhOiTllUspwCAzDQPCOXDO5JpvviqeoAxUjeqqNiaqC4pCWMSht/h2CJ3jjO21+XbV7faGoPTMhgnpv414c+AjikKr4oSIForsypIDGhC8BdYmiM/OslOkCCKcxcns84ALEGLm9veUwDpxOkXEcjHRyDim2Lx7Bd+me9DnVCJ3lmtb/183+Nka2j13teGBHsp+YeGUg6wUej/bz4AyAIeC8TWi5KKqVWtdFoNDOUnqqcGVTtEIt6/b7OVgrlAZWS5MTVIVS7PE2h1HbgK1KY4BkSU90JrYUe75sgOzyGpRUC1VhSoXzdSYBkcxIwmlDsP0N/T52cNqZZHE4CVTnUR/JPpC0clHleSmgyqgwNHZns4lFRtuxeNcVr9KCfWDu8QGiI9R3irc9vkGK7/rixqZqZ1INSEpjN7dycMAkCE07K2gjZLf5hYefJZt7D7hPffs7pD+hm316d/3Bqr/VvkDsjI828P4C46orm933WPvzy6KgM8d77PE+sR0YsC62sMdNIlBxArmUJn+xM9Q5j/cOlUrF2blZ+wJTGvBrc2H7JzRJGrbNKoKKs3O637b70dlctf63Kt5YIeeREO1nUZwqOMVFxcUZVwUJMyqeiqfUwpxN/jB4h9aKqmljhxgZYiSEgHfegKwvGIS1/9damjZGRd+2XlzGq+i6+JTtQlTe/VK2x8cyXhnIOuAgahpYFFdbpqPaCdYZmaKKqLQnWeZEB8vko9pS/4bzHoz3XKVpZIW5wlxscM+dkcXAaR+PtojsEnqbN0uFuQilKkVXVrdugaxCrtpUSOvyT6kUsWlbtKVrWIFmx7RJbPK9OuHiHRnhHgOvc63c5cI5lTbpm4bYt/2TNiCdKl5tJepbmhYVNgvSPfZ46Xivrn0LH9vYC9fZ1/a4Nu1P15sCIMVu1PaztM+xidf0ta5tC7w39scm5MaStFR/n3DendncgGd9++PGxnSdawOhfbt9f2tZV8edVdU+4dUGUHXJpvY0kM3z6/fhgOrcC/vQJt19ZO7xAaOvL7fYVlFKtTqOOReuc+I6J3IZbLnYgarqUjNSekYSh4prw8ggoaIUnYFs53yR5TwPweHELZ9r4BUDsKjpY90KZHEmDwAhtPFa1XGbYRwzuQq391e8j1ymiZTOiw7fOWfFXTEQhkgcB3zwSC/2IizL1oXsAkQqssiNDNBCJ3TsdSo9Dyrt2Puc2RfP+9j8pMSrA1lRHvnCVmtWWXWptSipWGqD6pYJJjvTqAptMmopwVzU9DXwYOIqCqXKAl5TcwtIFXI7Z8tmrnNOFluEiqUlahWmurK5ucrynixtfyuteAMy0LU31VVmb3A3qBB1ZWm8tO2JkHFUcdwHz330JByfz5XPz5VUC1+4Jp5eU0uY2gTqRFBvQv4OYkOTKejCwjYw+6p/qD0+cfFAL9ofayOkA9OeCekpcxFn2pb2Wm2pxlISKSVAcV5xvk1QviDOZAalZlv+iRAlIt4jCM7bcq8WBcmQa2OaDAxb9bZVdgqWzVhAbTsG6ua+rCyqakFLboUeusxfy6pWrZBLa1l+X9jYJldYFtUNAHQJFALeubY/ioawfJcL27y8fo89Xj4MxC4w1s4xDJzOOVOqcJlm7i5XfPA8modWzyGtcNlea/OlUhGyeooMlmfwgLeFWk5nUr4iohSMCHIORu+JEugZFl4Ageoc+LAAWvEBnMcHh5eIiCMMR4bxllIqPh6YsnK+XHnr6VPu789oMTrIB0+IgWEcOZ6OHA4jMYDzTXPrxECtdmlRD3MC0j7mZR2vxsjSmOx2rWv3l8zMeyzi9/j4xSsDWQFiU7F0ZqTJ5NpjnZVlZU0wCrTbThmQtedzMYD7INOBNFlB08WqgdeK/cx1ZVNrZ1IQY4Abi2qMrVqRV5UGZJt9liilAeMms2GpetQ1/aiNhu3srWwGVNfyZqwQLDlh9o5ZhQnlWiupVKZSmUvFo2Sao4KznbZ/9nmtTsYA74OxuE+Ze7x8LK4Fm1VhT6/bj4dgV7cnNSwDsdZKKZkOb3vaT7WPrsaM1tpsgirgGqvplu05pzjX04CN2e1ZGlGWwhXW3VjlAJv7C5Dtt8r61Obi0UA6/drUv41FWrB+UIejbmFiwYnDObcw1tsd6wC2uzzssccHDZH1rOoMba8RybWSciHlTKl1Of+WYbuZE6tqY2O9ndWdyNFCFU/Xy+YmpTPyRej5iGU/eqq+T2rOtYIv+ynOIeLxLuCcX+arWpXDYeIwjtSqxBAaLrbt9nHkvMOHgA8B59QypyjidGVklwPtEp5+1HZN2SRK7XFx6xzdF6G0cb8Py09UvDKQVVXmZJNIaVqeKlC80fypCFmlJR9toNn7KqXYMKqt/kJVyGWVFtiAbdqfDeOaVBYg20Ftf80CZJvdh+lf23611Wgv1MraBrNCaYysNglD9+bb2oCUNsAMTNuAKdqncajiqOLJIkwKl1pIKqRc0GReX6EqowoeYRRlECUiBIQgrdCrgVirOO0V5ntl9B5f3FicDB6ARJZMP6pULW2MT1ynC6gyHjwhepvbPPiwTCl094GqJiUQEWNwxeOc4oNt3/kGJp1NOLUo2nSzVrxlK8P12sALq9sObvvoa0BcN+CUdUIW7zbv7c/216//wCZdaV+CE1nTrw/ArC7gdyvG2GOPDxqy+d9YWaAqKVvh1xAD05zIuZC9s59NP1v6fIkDL3jXfM+DswrsppVVJ7aYK4lcC1odWR0ebwsyZ+d6l9kIGBD2HpwxxmIaIZw4vEhzBloLkr1AdEL0whA9xzHiHZyOB063Jw6HkcPxSBxHwjDgndly2kCX5Svozj8mTVxG9obFZpHZLUWm7TlZHEQUdbKM7z0+GfHKQLZWuFwLdQGXgjqheoeKPZaWlZ/NfIpQU6Km1NhY84lFWYEsa7OBfr+vybIKuU1fSXUDZFew5zeOB7Xpa6rae2tncxszW0Ua2O0TOo2BscHpMKa11LatKhQTE7Ga3QoVT2pA9qLC85JICtNcqNeCViXmykkdAThK5QB2ExjblgIGaC3D0w5i5Wr3WXOPlw7pejd9++OL/HPN1pkUoFUzV23eqbVyvV64u3sGKOKPjMcjzgkhQIg2YeailNr9LZVMRpxj8I2FFRik2egVQBSXrRC0pNUJpI8nI1OFWtr0uhSILYN0Rbod1OqasO3Wmd4L2gpJepGLMT2rhMCJbwDWVK+dbe0yA10YbOis9CIz2FeYe7xCNGO8dt/+r2rneVXlOheen68gwvkyMaWMd0JKmTSba0FOSs6K4nHeE+KhGbVHiN7Aa/BojtRamC9nUr7gxeHV4zXgxBFCIHhvJOwGpLrgbIrbzDsOh+8yIGcFnwUlijIGoQbHzRh5dHMk5ciTJ7e89vprHA4jt08eMZ6OjEPEScFh9SJmYylNFlQXvTt9nm20kmi1z5WN/MDZa3sWdZEU9Kzqni35xMSrM7L0FL0aSGyPqTRA20DsAmSbAK/iLM3RbrV5ZHWfV22guDQQtxoFGIjt4DU3AE17vqqs+jVtjKzq4vu6SBSW+3Y56dvuB7Ws/Np/fSh1TFAbI4taqgWkpXYcVYyBTi2VU014a7YlarJ236ZW39I8HQ53Y/r+2SYx6JzTbsC1xxcvms850BdLPFgkbbVmpRZykxbUDhwbGHauESh1ZUpVrTDTqU2yCzvqWsUzTWbg7X51tTkeyGbO7OAVVPu1Q9/htuWz1vvb+asXnvVCtH5fGtvqxVsnImyifuBUsOjutt9L7b+8bZGwxx7vFctcwjvn2fp5Vmsll0LKhVyarlxt/JmMR63hTiustnPVIa0wC2f6VyWhDOZzLo6KNEs+m5dtMecRtwJZEYylddKA7LrvSxZCV2cdaePbieCd4J0jBhuvMQaGIRKHgRAiLnhc8A86Wtpi27X7jsXreZvvUJbxuozOJYtky81F17jQXvuM+UmKDwFkhQuBIkLC2amzDAqhdNDZgGmt1tmnZCHn3vFnTR+WsvGFRZpbJQsQVRo7qisj2yUHPcko7b0rI7uC1d7lqyoLq6u6Vjuqdr5lc7npY6UNCXURwoiKw/kBF0xgX7WSarV9qtW6NqgSauWghQqYOYkSRDk54cYLB4HozfmhJTfXz+3IG9nu0R57vHfoyhi+a1tWOmPLhuhcnQpMd9o7a/U0fZ98awO42rQ91pErlxnQldFxQq2Z7P3yuzkYVJQMUpAF0DYgW20xCEJxtphdx0WbSk37YNvrScc+sJWFbZW2Et3+XEFqB82CEzP9ecDUsp0G105hYAvvzvzo5vE99niZeKeFV39G27mfq3KZEt45LtPMNCe8wxol5GxuQBlKBsSKhJ0zIOt8QOJon+ABDZRinrQ5G0hMGapmfHMUid7m7OANwDonxMEsv5Zxs7lOoNZ5Lxfr4ldrK/EW8MExHAZ8DRxPJ463N4yHkXgcccFD22bHyFIdVG/jqVS0tMVyI4FArf27rnOyrojf3ifYNasvpoG3adv3+FjHKwPZgvCcSMExi6cieOeJLuKcM6urYtDQunDZhJBnR57X1pU9vWkdtxp7CmTaCcrGDxbZsKpijgibta0gC7uJSDNVNpa4sHrt5a0Z+zI+usxg1QoJ9nqnDkUocUAPt6gLuPGIO9za+64XpvOFVK2pgs4TUitDhqFtc0ZJogQnPA7Co+AYBEYpDciKWZyIawVxzQ0CFkCxxx7vF5Y1WCjE93ztdiI1z2cDfN1ZoLaJxLk+mVlDBFXrPlRqRrWSy5Vcpv7pSwGJ925hd0IMeN8N66zIAzGbOeeUWh2CQ6tNdDmJZWsERK15ZZ9UOxvsfWia3kUuZzKBBcyaBlAeHPEK8qVdMdwClNtrWgqmT4h1Y/VVGpDVWim0VOgee7xEvBdDaNZXBu/mXLk/T2it3J+vnK9XhEopiVJMlpeyklMFiQyqeOfMbzYO+PHQFnARUQOcaa6kbOftnGbqNOFEmOfcmFQhBk9wDh8cMKDRmNYQZNHOaymNkCrknCi5UkqhN0UJQ+B0e6KqcvPkltvXHjMeDhxujsgQkBBwrrYFZMu5tuxGLQXtFdzdOk9ZACtt3C0LyV50qjS22hbWtU30+p7f+B4fp/hw0gJxFFxjZB3VkuZ4hIJ5yKraz95Vq3T/1yYpMP2ZmLSgkZBFjDVFt16vssgCOmFZNqnRF6XdPcGwFmQ9ZHe3zKtsXtPjYRqos0HebEl8hDhAHEEETZnq5tXPtvVvd9r1UOBFqQIBJYgQnSNiWlzpt8ZadYPnNQe8D8g9Xj7e3hjgJd+HbthZpbduNfCo62TWJgytpQHgsgC9nvswxrci1bwkndMmxWEBut3wvFdLO9fe42QZE2ymI5MHNI7WCd631+kqJ7Aq6ZW9dQ/8XzffT9e5NtXfCmA3RZbaN9wmR2xir+3xPVOyx8tGX1gt+uoHT67XeVuI0qQFnlwKpRRKNYlBzwKYtZzDDNx1mS7s/Pd0ElXESBjnAyIBlYwlDddMoDqh9vPZJi2z1fLOzESaZnUpvGzA0TqHVQOR7TjEOfOJBfOOjZEQg7GxrQj7BaHrYjMpNlGyOpSwjtNNponuSy3mZqCoCeMr9NbX+5T5yYpXL/ZCOBOMkSVQEVwVpmTqm5orJZVWOGKeslWVmgvVbCnbzSYQa0jQALI2P9o2vjvArA3Mom26XGYR2fzP8pq6bv4BiN2yL0uKQ5omtU+MbWCJOARr1+cOJ9xrr6NhoMQD53ikAOeqXHPTEmoharELTc5UsVVlVMGpEJ1wGjy3MRBQRoHYPDi79lCcUrcWDjshu8dLx8pSQGMeH86aS7q928ytzUua80Cty02cMMSwaNBSnltDoQRipnNVc/OGNWpUenNJlSZnK1Yxnd2yHbPtYmFawCOueVRGz3iIxOIRiTgOQDBg2sZJL6xegKz2o1tlAdtWnEu6UVl0htB0sXR3AnvhwmrrFjSYPrE2zaK5r9g432OPlw1ZBeo8GJhLBaYVTF7nhKCcp4nzdUJEjSpavJT7gtDOz5oztKLF4HzLhIRme5eIcSTGkZwdk16ZU0FQ0ty6gAmExsyG6EnpQBwi3tv4d96x7Ww5p8zlOlFKZcoVdQEJQgwBxhEVGE831gBhiFRgyuacEDwE32pD2uJUlLW/vNr3YTKD/q10e8+1m5eW2q51xoBJl/rUYrUp77CI3ePjGR8KyN7J0BhZb9X8tSLJAFhJhTxntFRybRWW2iDlssoycNcZVgOyShFdGiv04kX7zS06oo1TB1sgW+lsqi4AeEtqatcGKmsKsr25F0c7ZS1MEWfWJuLRm0fUT30VOo7cSeROAknhTuE+F2o29e1IhVqY5ytq3DJeHIM4BhFux8DjGPEoB4TY4bQ4EKHkSso0xmv5wvfY4/1DaaDrITO7FEKKFVL1+XQ5vRo4g4epdOdgGKNtWDIpNwmBJFRsRSpimtclbyJ9cNIWlIpmK1Bu+RT6J7v+UwLeafspBOdBA05GvDshhLaobGWRoos7QW/UKdCqnuty0B3I9pafJhVwCwCV1hjTSOame4Ulhalqi0pjnZWSywpkm15xjz1eNoyQXAbjC/yEPZ5r5TLNlFK4P195fn+h1sIYhePg1gwJHcgWSk62wAOiN+DpgnXV8j4xjBfiNKM4ShGmOaO1UPPcmhcYUBYgDoFpOjEMER884xjx3m9IFSHl3PbRnAvUB8QHovfE1rnr9OgR8XgkxkAV4TxNpr/1whBMKhS9JzQrMCcOCS0TWx3qWkmaW8dxh6baSbFq+9zToaqKlKbv34HsJyY+nLQAs6PqP6ltIqlKLUpunXxKtS4kS6oSllufTIuu01t5kU3dAlFdP7+nYuxeT+Jvn98QmUuqvn2ubB+W5WlaOpOeVeyTYe87HSMaBrPcwpOqkkQoItRWtemdN9htuR5rP+sEL0Jw660p/1oDh7Y3i86A1X2ENR27xx7vFy8WeW1JIOhgdkM/LszjhuHpTMeS4gcDuRUTpBaTAXSaZplYbRG5DqM1RVhrZ2P7exbyBZMj1AUI+9a/2bUx00FsdwqRNpBtyDZ1oRo1u+LYFbAvY4o+lh7mHnuB6Nu+x/Y9PWzAsILctcp6jz3eO3qh4vZEk8293mJWlTZvCrlaNsMaI/h1Plp+0rapyyDvNnpOWiMD55HmztG9kWtbtJVSFwKmOwkgkFJGnMlonGsOCX1Cbh7pOTfJg4h1AxNBvMfHiPPemh9412o/LMuh1fS8tdlp9a5ldvytymUDEKzJg7N5dPtdap+j+6BevwNTAz24cO3xMY9XL/aqylvnZEC2usXMmcbI1mLSgm5fU+sKM7u6zGGMLGA6Wt2wsDycfBeGZ1t53Gaybf/lPrGpCBUrLtkCUhWWtKIs77HtdAZ35X0A70wP6wNziFxVKBXenGd+bC7MpXK+P3N/vqK1clQ4xgHRitNCtE7YHEQYRYgiHJ0QRXFacdVusAIQV6qVpGpnufYKzD0+QGj3VQSWFH57amFzWIArgJZKSRmtlZwTOSdUK+IV5yriwA8VPxjQVCmoaxOgFKT1PXcN0Fp2o8satDGgujxmc7CBVCfmG2BO063rl8ug4F0ghmJV2YBIANwih+jZFW3AeAHfbBapCubLbJILbXUk9lV1XxNMNtAmQ2Nbu4Sg3zf2ay042TU/e7x8bBd6y4nbV4q6asFVTYonYq3YJ7Ws3RgH4s0R7x0uCT7ZEi6MoXXT0VUbu86Q5lHuPTEEtBbTrYZAKZAma8Ag0uwgxeb2ac7mGiKZy3U2tlcE70zik3NhmmdKVbPVioOxrWHg5vYRIUZONzecDjf4YP0sVVObrB29k58Q7aZC93E2gKqoa5UvvcsYwGacC2Udg9oxfbf5011a8AmKVwayuSqfu0tobZ29FmsBXdObdTW06gBUYWEfnS7t3R8wslvCpHfy6G/r+r4N1FzA6wOmtz1hvQtkKSCx2Xa7Ku4vap396EC2sT7eo+MIIZJC5E4dc4XPXyb+1rMzcy5M1zPT5WKvHwLjeMCheC14Ch7lscCtmH/sjVh7X0HxVZHSVpKdza4VqQn7RmSZmPfY4/3CAGwDXX3i3KwKO5uzzKPtVnMlXxO1VqbpwvVypmolRPADiIPROdzowBmQrc78ZZ3UZsDHg1a3PYugKNSCatOCe1tUeiA6h3fOJjLtSvaMONMIeh8YhrTYd5ntliMXJSVdgGddjq9lOVjHcZfNLexzbQXR2r2mu064LoC1ltykPWY11O9rzYt0Yxev7/FBwgkbJl+XeQfYSOS0+cja41OpXKsQVHgUDwyPHhODt4YIc7UFmnsByEqfv6TVeQjBe4ZhQFWJMRKiyYWqKinnNi6ddcUqCnPGZfOuzW1R571nHAac85RSmFOmaiUOIwcXQBxDHHny+DWGceRwOnA83eCckMuVlC0TIwuQdaAR0UNbcG75aelKO+ss5prjibRsjiriDMguycz2DUpf9O5A9hMTH6JFLaRS26TQWJ+Nn+NikaGdfWl32sTSo2dGuo58+8wG+m4+mHXwN2Aqy6t79eZ6f8PdrqmdFeq2i8lGWtDe0RkjbbICdY4qkFVJtTKXypwSUy7m0VesU0mNftl+L04JQJB2Y6vp26SEjLZu399qN7L9onaL5z1eJt7dP1YfnlILmmWjEbWCptJ0ZsUBxc7ZjdoAXczqdPnXPuHhtjtLSr8W2Pjri9XGwyyZlT7mu/ygs7320y/aQOluB2yvEC8a7mwXrLr+vrl1aVMHtS8eUS+IW7ex/awdxO7xciECIXiT5mh54czpdE8bG2JuG5aW94j3iG/WVa2xgNSCNPlNr/vYjsMHn93ArGuLRucc3ntq8VijgZ6VFFol56Kxr6U2CYHVa3hX8N5cDUrdSmxoTKoQfCD4gHcB3wrPpLrtDA3NJs+JMyuuPgdvEOlyXRDX9lOX72hBrwvR0+dsc0nZFnru8fGPVy/2UuVyze0X6VnD5ed2QNnQaJXBrIB1CzfXqUKXk7q/pt/vHpFsTtrlp7ZJsQneVaAuFZ6rrkilp10NWHfNnWu6VXMtqEj1iCp5HEhjpPrIvSpvXi5M4nnr+TOePn1qq9I0UeYJD1Q94DQTUY4lcajmlXnSyrGJHZwYDBBVNBdqqbb/tVdkrpYqD5D7Hnu8Xyjo23Sb+o73bfHZFpzdqaDoonPvVfqaK65Cyg5XPIJSfUK1NUFQXYu2YBnpfcEG0Hv8Wc+FxspqZ2GatKAxKWbZNwOZXCo6g0jA+wPeV0QCtTrodl5OF/bXoQuIrrlS2mK75ErJzbWgyqoj2mDb7fezdhqT1q8etLURRdr93hxhjz3eJ4YY+fqv/2qmaeLu+T0pWfYj59J0qKs+9NGjI2+88YTDOPC13/C1fPXXfT03Nwee3I6Mtye8E6pMpHo1IkkVpYC6xRYLsLEtDtVqjQqGASdwe3uLEyHNM4J14AJjjBfs1xd1VcmlkrK1o3euWtdOreuysa/1KlbAjbNWuOIaUBVy25ZScT4QfcQ7z2E8cRwfNUlBXT4XJ0uv6V6EXWshJfOvtbFZF0zru45WxBpDyNYbeo+Pe7w6kK3K/ZRtAtK1+t/16/qyaGoT1WZ1tLI3wpZDWbIrm/dbeqSD287eyJI+hE3fH/n/s/cvO7ZtSZoe9tkYY851cfd9OycuGZGRmUWxWCSgBsEW2RAbgtiQXkCA2KmWBBEEBQGCUG+gaqhBUA3qASRAb6AO1SGgJgGJFxFgFatYmVGZEXEu++Lua605x8XUMBtzTt/nRJwdsSOjKs5edrCOL1973X2OaTZ+++3/hSSBiFvU9j2qD10RTBioejFru1TXtwtGVMcXlHQ+3W5HHkdKTF7IXrgovL5/x5s3XxspvmakFAYB1ULUwiDCrVSeufPK2AqD9zebX1BFqonrCiYJGLTTMbzNqt/Ema5xjV8f+rSQfX/Si21PwgvWZshLq/ZTq8lN4fxQyEiAUCKherEpmSbW/g9qGzBb/xtJO9+sWd1qklkqwXusvpq9mBURv4+YmUqbjS7QTAQeCcR4ZBhAwgAkEOM8mNSQfdYunYcqpRRyMQH5VqAVDMGqETTyjS9l+c5YWrQLAd+J++rXtSfcax17jQ+IcUz87Gc/4uH+gV9J5XyGnLPxRmu1QSw3D3l2d+AnP/kRx5sjP/nZn/CjP/0px8Oe/Sjsd4ZMZgXNs63ZZsoaqsGMTKrL31XrgKgqKUbCbiSlwF29YxgGK2TFqAbg/G+c2pCzbQKp5NKY54IqhJCI266OeA5vGKdVba1HictFxNWJmnVZIsEK2Zg47G64PT7DXP8qrRZbfzEg0bbHtVUaRnGY5jO5OKUp2roPWG6XYMhtiMm1dK9581OJj1It6Mozy95HN/jhtwCJoiyTmU9uf/KrvPfPuhyQKz1gc/0J1aAjr2EZaMEnl+kJFKUt7UnjziK2CDQ4lUDDquMl4oNj4sYOldywdkuttFoJtSGt2iBNqzbAJUIUJQUrZGNTQuvC0StBfTs9vu16srRw/VvQ66K8xgfGdk2ofqPWWik0LC0NazcGK0yjXbdB5drZr1i2kmV6f9vMfKLd2l9HNx0TX5vLBSPAKbIe5Ns2zNIsNYUEoylVmhYfEBUWO69eeS5tIXu88XU7j9XRnSdfxqYnJGtxL31NbmXElu/O3rOAIbVX0YJrfECEEDge97RWOBz2gBKjFW+1iheyiRAC+/3I4bDncNibves4ksaBaDRUluTgoQsK610VbYQmaFgHPju/XNVoBSklVBspDQzD4Fxws6cWEWqt1iHcrNlv5m158vPJGlvymiX9Dl31994d/+xnH/JqLB7XS/fVflUXh+8GLGDPYU//zaHuaxH7acXvXMga7tGxFy9lRXkf0Bc2iCmWG1b9xxVZ3bb3HFyx6xuEp9MG8J/L2vICVCXQxgFiWpEi7ICXFJBoiGyg0TucPdEuKgdikiNGS4AmiazC3ISpVM554lyVPGefIGlEaaQAoyh7bdyUwi7AkcrRVQtCq4g6g9/JgJ2HtP3sazG7KNxei9hr/FbxdCO5KSwdZVw4ZIFFt1HGEW6OoMp0OTNd9tRWOE/3nKeCilMD1HWgF+qLWVMnsZU8hEgMcek1PkVkbVNpEnU2TNloFN/49WFLBSS6QYm3Ta2knSj1AVoiyICE2fhzGqBtE5+/s1AJqVrH0uth6XQif7GALINitTsGeRGrm9W4RG8BA2hY6+ZrXOM3xG4/8q/963/B+XTmRz9+xXSZOJ8vvHnzlpyz2ctGK2R/8MMf8rM//zOOxwMvXj7neDQt1ijVKARtw12tjelSmabZ0NLhljjcEGNivx+JcfB9nrXcAcbdjhACwzAgqhz2O1qrzPNErYVaKkECJRREAqVUUjLr+ZSiu+V1lz1Z3rd47sw5Gzo6RFLNBA00a4cApk0dk5j6QsRakQgqzS59vTVzEJvLTC6ZUjOXyz3ny6N1X6vxiFNMpNQlxjANXZf8usanER9RyG6pAXZit9JrVQ/o9zKtVMsviya6dD7dWsgCyyDnQlUQa7cvSXjZca2IUovBrO1iRA8jOg6gkJoSmslwpGSk8+Zt++rvPvhP4/3YDq/2IhOoEskamBUuuXE+zV7Izp7UK1GUMSijwJ7KTc3sGtxQOaqJTat2ooMJwzdHkJpjTv1LWL8L6V/fNa7xu4Wvs20nI4ZIjHFZS9GRjyFGxjQgwHw5c7nsqKXw+l1lrg+mK7vRauwtBNN5jaQ4ECQwxoEhuoGC9illiKFvQvtGVUArWmaKmmNRCUrsRWj084JijmO+fkqxrWiIA1EyQiCoUQVs4MM4eg7qENKqUiBOng+ii+yfWX6awF3EZQI7j745Ktu9p0M/L8nynV5z5TU+JPb7HX/3X/tz5mleOLL39/f86ldfMk0TIcSlWPz8Bz/gpz/7Gfv9gd1uZH/YGTWvAc04ta0ptVZKrlwujcfHikhiHB8ZxxMpDaRBGEh2vIYVqdxhclzaGmOKlHKk1sL59EjOMyUXVHXRoW1NGQabh9misL14jTF6W19ozYagESGOkVIGYgo0NdAHUczpPRCiIBGzmhVB1ZRNFuUVDH2d85lpniglc57uuUwnO2e1RIwRHQb27MwZ0BFek+u7rs5PJT6qkF1x1n5Sf4IjLojNes+NIuomyS732rYhl1t1RViQJwtpEYZ22RBiMAmRuPo3g670gfdaIL2H/xQN9QlQ6ZOgwYpc6ZOchhKJ9OGUaE5gjk/3wqB3Orfxa2kD/l2pPEVge2cTvvFU17jGr43etWCbdMSKuxgjyROP6UJGgghjSuwGK2S1FWqxrkbs6K26HJ3HkxThVIHu8x5C9JUuhlpKp6j3DWhHXJy/vlCOfC084T6wiLR3dNZWQ0C1AHEZuLKzz0ZyZ+1SugQfiwylbO/DOiC2dEpkfU99By5+0rLPsBUAvMY1fnNIEHa7ARGopZBSpNTC8XggRiHEZJvBGIxSMI6M42AIaKfAOcqz5I5lX2kFn0hY7aW1uUPfmm+33XZxOasQA1GtgxJjpFYrAsOSxzyX8fTnNv/2WPoXrmSwvSydUVg+Twg94T/NbpYLraNjRW2ltULzQrdp9U6K+N46spqUGJXoyiz4tOKjClmRrlbH5iebVLJii4IsBVxfiFa0+nM9TUGspewqjRV8ClKCkIZESBEE6hCpyQpa9jsYkiXJqULt05ebBda5NiildaQ0oOKS7BIoIaIiZBKTBuYmNKyFMQbzNJKYTEdPM6MrFYwSrI0KNBWmZsNq2sriHW10n17uu4LBN/aP223ANa7xYSEiDC5OHhx5jSHaepFATJFxGP16YEiJEIQxDex3IwD3b1/zhsKcA8PJLGNxjp0ZHvQNqeIzygSxy5B27MZxs7YtqaRohbMhnYawtlooNENLXQC9o6FbSp70uSw1KS5L6NUpB4KQaGrIk0pExVGoFEkhmEqB4tR0Mb+FuppZ48iryobGoxFV5951BLp/x/28ptf1eY0PiyDCfj+SUjRnq9rY7UfGMZFzsTWaBiQEbm5uOd7sScmsYUMMy+YJMdfI7s7X9Y9rra7XmillRlFyGZizFcLBByIVTOZRLQPHEAnJitIhD66Yg7ly1Qa1kktlyjMxBHuPEnz20Y7+htFyBFMJqbVRqs2Q2LCooa/7YTQ96sOO3X4ghkSIQvOuDWG1nWbJi5WqM7meKaVQ2oxqpiJQG01td3qZRxQY0mDfm8Tr4vyE4qM4sq4hsEFVO0rxzR1Wv0/Xy/Mc9wSJ7UpxT1HctRAWCUvbIA0DaRzQKMgYkWT0grYbYYhQmsuCVe9y6lLMNt+9NW0UNTcfFZfZkUALiSIDTQIzgbkFJqBJIDn6KzGRdjugkdpM0pmkyqjrFHdtylxBtKHFWpWi2++jo774ztmLWdnCQf07uWKy1/juCCKMw7Bw4IzXlthteHG7cedrKLHb2X3GYWC/H53TWrlcHpELDIPZTGqTVRFHenclGDVIIkFsUnhII+O4t3OBG/JY+9F4sU3NbKC1Ri0z2jJKs0JWih3ljtIIssig2PnClDxsqQSb+Ea8iB2wvkgCqUAghIGQRjvPKEgTtLGgwtLVFgKsArm9et5uLL+hrdIXLe8rQlzjGt8WHZEdhsTow1WHmz03Nwdqayu1QISYBoZht7TrJfRsYVP5jdV2emvkAVY45pJRoORMzrZRjR0B5WnXM7jZgAgUHwDThk3+R1tHpVTmOZO6jq0jsR2MMvqPIg1qM33Z4FrUtVoHZRhg3I3EKOz2A+Nu8M4Ny4Y0hN6NcWhJG7UXsuXiDnszDXP901ptTdMY5gmw8nfHHsTpTdf4JOIjEVneK8z89uX/3zyQvnWTtBlM+eY/bfsh9qJ9sEtD8CGtPjzshaHLgTUvfsXpBd3X2f1+ECBKb3sMEHcgkRYTcRhRCVyqEqtxbWMNhvAoaLPWpmpDaoXiCKsITWxpmq+XI0itoa6mYFSE3ipdvyXxxy8Fvq7f4VWC6xofEiJiKGuMTwrZYRgWZHa5fUjOy7MOR4oGfQYx6OapvLqval/sXW2gWz43UYJsxNG9/bhMEQeXyVJFW/B2/6r92B0SdLG1FZ7OOG7eSVcjWM4ZbWkpgk1m27BWRBf5EfeCF3t859iHIOao1zePS4foybfKtydF/cY9r3GNXxedl67Bzv0xRGJKUNtK+ZHgJgJrEbtQ7fwY7d3NpW0vqwLAMj+yOSz7fqvfFzZ0tr5/2xzeXWd96V6+9zl0s8l7SiDcgjL+/lozEy/dDp1u1Us2j/f3vaiCsJ5vvrXQ6H0hVZrW1ZmvNbQPVl/jk4jfuZAVf3Bw0KQLoW95Y9sCrf/8xqL4lvtvr6+8VUN+xBd5CZEmkQZcaiO3ClEJIdnx3oRCXPg+MZpIsqRI2I/EFAlDIh32hBRJw4Hd8TkhjjAM6G6PSuCXDw/kr99wygXNSp2UXJXLfOFyOdNqIT8q+XKhqnIeIo9DIgnMDAxUM1g4TxAmQlOGUkmu2VeXU8VTHLubSyjdxOEa1/juSDHx2avPiDEyjF68psjoKGxMkcGHSmIMxMHoBylFUjKOnQpM88w0zZRSvd3evdYFAlS65qww50phtoI5FoZUPbEmG96S1alI1Vr4NBvmkBht+hFYNm1LEQx4MsVJgbJIbnWugIB68hIBKo3qiHAjdM0/SeA+8UEjUZ2UoEJUH1ZVWTL6onXtBf1S2DvNwDazf4i/6DW+D2HH42BT+T5nEYKQols8u2qBdS+SKwF0Lqk9h9ZGy95oVKU248HGGDkcTL5rvx8Zd52S0Oc51jxqlWsvFF07VhvFkdycM3nOzPPMNE3kXOzde95lKYTX6AVrVy2wwlIp1ZDc2AIhBYYWFvtYUyywGRMRG9REGiLNZ0M6L18IyS4qIFF5ImQLqFRyne0zSmPKIwTjCV/j04iPKmTfL163Mlvrveznr8ct+lCI/7bdSfrFpLJcXkvcVi8EWghUlEtVJm1IgxghiiWlKNEml2MkjSMhJuJuZPf8lrgbGQ97Di+ek3Y79vtbbp//gDTskXEkHA5oiPDFF3z1Vz8nXC60SyWfK6U20sMD4T5RS6ZezpSqSFOmMXAeB6IIQ1SSgLRGcOH30BraZpPkXKwRAk+X5joZvbj+XoGfa3xAxBh58fw5KSVz8tkgst2ask8ZhyCmDiDiCCkmrA7MpTDNmVKrixR0ak8EUapWT8hmV6m1kJJy2FXKrlqRnMBglrAmQtRak8H1OmLAfHCh69T2rot1KCzZLql4oxO7ptPg3D6hmuIzSHAXvQ0FKghoIAQlJqP5DAQG3CFsU8i2jnjBk6GZTktSuqHhdYt5jQ8IMa1YbQEN1QtQoTU7ppchSRGT0erKIlGI7npetbjajR2mrVknIsaB/T4RQmLcDQxD16T1crXnVoc4t6mkaaM5l7WUajzUUsh5Zs6ZUostydBtYtd4v5gVWIpYVaXWRi6FpkIqyZzBvJAO0exy7XmdBN99q3FVEbWVG6IVsjaXAhL0yYsrjVKzyegFZS4XQnzPGOYa3+v43akF8j61YOV4diL6Nx/z61t07xdqndC+vBhr6ur0NMNfBI0JUCQl0v5A2htHLzqHLwQvZFO0AvaZFbLDfk+6fWZi0/tbwv5IGPYYoWdvi363ZzgcGETYR+V2MMs+k0NplHlmHgYenYdbWuNSKjEIFUjeTh3EkDCaWX7awAtuc+mfWd77AjZw9dUK8xofEp1aEFNaitZ+6fzyrRf5++LhKu8vxN7GBJbtlXuk+wasPekm6JP/96drvtFUPNHRi8DeB+32r+trGM1mQw1YaALqn9Vf8WlGpakNnqiaLqXdfSUgdZME8ef4tu32wgDU919HWE9L193lNT48DIxhoQJYTRcWbujynx+Ti2LA5hhdn2yFO0zX1QrhGHydx/fX99N1DiybNLOm7gVtvxjHtdvdiktxLeYFPCH7LEoGy5vrr6G6INDa2mLrvGEgPf1wm7W20tX7JnoDk8n6DlYqRPPBt0Kp8ZozP6H4SGqBfgOVXfk80DuC4oNU7+OyvSBdrq/A7HqIyuoAVFSXyX9taoY/KcLxyLAbGfYH7n70JxxevCDGyG53sCnLGAnjiKRIHBLD8UgYEiElwm50J6MdpFtaSNQglGDcW+4qtz9qDDnzKu2Q4YAC7776mtdffMF0PvPP8pl3f/3Pybnw7rHwOF8QgXFIDCmRJPDZsOfF/hnSKhIf4Hy2ts6saKnLd6HCYvdpyV3ZSERf4xq/MUIM5qXuSGxHXlft2E0y6pWcx3oMBkQSEhKKUKu7+LVAYPDkKgSxZFFUqdoWEXLtxacYCqo0prkYlaAPr4j7tQdTF6CJc9ssQUdxGIpmULFr0prklqFYi3oAq9V1U5uaVjFbzOhFcOia0uoW1mq322utvaRNmbzZOPvqE7Xhkg13sF1X5jU+IHpujAFEkm+Q3AjAC7uN3tsiFyfCQi0wFk2zoWHpCgKREHaEMBq14LBz3VnTWg5xTfHf2KO2Rs6ZMs/kPHM6n5guF+Y5c76cuVwmz9+BlIxTP447YoqrLW4HsJYCPS6Fc9dvhkbJME1WHOdcqNXXtbBoR/cBMlTozREVU19Jgxk7hBj6/tKLB+uQ0LJp1UrjfIFcLosD2DW+//ERw142SiHSKSuyGBws91h2krIcpOvQ0jd5oU/Wsv+O+G0CFZuO1KamINJM/mfYH0m3d4y3t9z95M+4/cEPSePI8e4Zw35vvs3jCDEaH3a/J8RoM1gde9EIOtAI1NaYWjHk6EY5ijC2yvHmGXfPXiASeP2rX3F32HN+fOTNP/85oYHmymMpzFMFgXHcMQwju5S42d/B7TNCdUmwqlAymie6IPsTtFlX7GnbRL3GNX5TBAkcDgcWofJl2MpXohezduf31pz29blp+WmgNS9k1aS2+vO14JzXWtBgbXx7Gm/F+7pVVXfmKat0XnQ6gLf7AbR13ZKA0nlzAdFqW2atS0FrydNXhsjq1UCzcwQYZcclAvvAS9CAagQiihXs4UmCX7fbrQPGdLkxb0F1niz2ete4xoeEATxhNf3wDVRrltNqXTsNIrY0OuXHQpf/BJPOEgIxDaS0I4TEbmf6s4agpqVtv+q5bjKwKrUUZi9kp8vExQvZaZqY59k2wml0WlIiDSYhVmulYPSGp92dFdpSFcvVCKVU8mxrvJRKbQ1pas5ejnx1KyVdveyhd1TT4Hxfr+51mzEbVQuiAS0NmZRYwoImX+P7Hx/JkV1730+6A9Lb5cr2H1fTBBa9xoUVI0+fZGXLeLtTxKxnQ7LCdLeH/Y4wjqTDDel4Q9wf0WGkxUSVyKTG36NhUj2h2sIsDQn27qs2T3IBrQlUyK1yaYXWlKllLmWioRwOjSENbv+353A8msXf4cB+v0dbo9QztZozitSKSkEQ5lrJrS3TqiTbYeqUaBKWhblMffu30H+9YrLX+KAQ3kss23a5rzmfXl6OLS86Fd8c+sBJSmlBWETX9r84XzYG34AlR5tiIHUKQ1xpDAaqytLZV20LBxXWlivizy/GxxMMiVVHYG3oyqvvhSu7QbF6F6gXsroRhIcNjWCzPVTTrpXlOVaqj+dLlrORI2e9lDaHpd/XH+4a3/cQb4evqXEtXO34V99M9la53cWontvb+hqNzq9dqUKrKQD+QNscNs89sOwvbVjzW9rv26I0hM5ljcQUSanbwcqGdrC6fG2pB+vH3OY0nl7f/rJU2LJ8KV3xpHOIg59fTHbs6fpeCUv6rZ/rGt/f+Cj5Ld/rLQfuQinoXcsnzb91gGlbs3Y6wdZJ5wk6KYEq0egBx1vi4UgYBoaXLxme3SHDSHr5knhr19vNcx7DHm1Q7i80JmqrTDlTqk83e3umtkou2dqRRZnnRmuQa+VSjDy+Px64eX7LMA6M8cDxp0f2hz0JZTdEptOZr//5n/LFn/4pjw/3/PXXv+LN67O9ZmnAxJgGboYdKY3sRPhs2LEbd2jOtFzJ2ZEmLaB1s0lY25fXZXmND4tvL2Clt0caNK8oFWiyXWuGsISQOB5viSHxbveGIIMllhbRaqbOw5gYdpZI2bsVbRD2B5uaDjGYbuY4UFu1QRWs1de0orVYQnZElQXJwdulZtrQ3bxsg1eWFdFlA9YzTz952BS20u1mq6FWoqQhIBrMoKRmlEijUKsNkAZJvD/QAr0wdmSpVVqzzlAuphV9jWt8UEh92nIU9cEl43RLa0vh1z0LRPHNkqGnWivalJQSh8PBi7lkHQwRs3SdJ1tDsSHBgJpaimu6+krxgrduEFWj2EWaKsMworjj2LAjxMRuHLm5vWEYBuZ5BgmU2s0cbLhst7Nupw12Cq03XTzBKxFVu715Z7W5trMtY5POE42uWR0Y0s60bSWyG8+uJ1vJeVpQ7FVZxNQKVmfPa3wK8dGqBT2MS8OaQDfF7VO9OKMgqOjyuG/QCTbXW28vhkjY74m3d8Tdjv1nn7N/+dIK2RcvCDe3tJjI45EsA7lVHi8TUynknHl8eCTPM7VUyuVCK5VSMvN8obbKnCvnc7ZJy1q5ZCtkX33+ij/56Z9wPB758Y9+yn4YuTkcSQLjEJjOF1794HM++/xzxnHgq9M9pSm5mK1ebUpOhXeXC/vdxD4l7g5H2O9hnmjDAzWcrYCtLvi+NEbXIva6JK/xIbGst82aW6KjPW3FLurm+GpLIWv8ciEwDNay1GZoaKviXL/BjRcgxkaIptU8DJE4RDdfcPSmQSyB2AKN5rJBdqwHdDlvtGWyJS5tUWnNbu+yWFqh29Sqt/rZftb2BGmW1pY6OUTjzUron9yK6oYlvhDtG/y2GS67r9IUqkJtUOq1kL3Gh0anwfSCy27rKhymbdw3mHacCQLVjAbAOKkdhYzBJPW0KaUKtdnaadoopSDSCGrHvapSsqkR2AvoUsx2hxCT+eoIrBWmTc3ha3DFn3G/Y384MAzmQJaLGZgsCiliRkUim4Gw7kukvWIQpxysBmMrTLMt8lelkZgGBi8w0jAylJFaC3Wxre1faO/EuMTZ39Jf8hr/8sXHGSL4zw2d5SnH4L0jSTYC/50Lswoub9sRfhCKIGmAcbf8ZLejDSNZBK0NCRWZZkQuaIjkGWoq5FJ5dz4x5cw8Zx7v75nmGS2FOs1oqYYU1ZnWGnMunM8zpTZybUyl0IA839r5R6R3Ir3VYzIpISZ2uwPHuzuqKrvDkSENqEJ2hLVr6k0lI8ClNS7N2kQTQnEeU8D4gEEX8GzTBL0uy2t8dyybHrVN0CKjilN1VJeWYDdn3jTanUcnpBhpMTKkxJgGLAUKtVTQQM2NVgxuCU7ok02yWrgK/ptZ5gZomySKLt0bF+HbqCZ4q7RXodKRWzM8WDd86xb5m2tk1X0FkO15yhO8qisqiGlg9tNQ0/VSVanNbDhL06UQr80UTK5xje+M3i0HFsmdzbGJrtSBJ3rGauL+qkrNhZaLO0SqrVGxDVY/DFttKNkKwWoufapqslq1Ore2F5qb7KL9Je2NxpgYVKyQHUb76frTW959iO/RiMQL8aa+bDdenuJSeMtC7NyJ9TvoAur9XCBqQ6UhNGJIpDiQ0oiIkMsMhIVTrGqfLcXuGvYtO9JrfC/joyxqu/zOQglYSWh+j3XNhiXdrCy1AvRm4WqkYHpxDUNh080d6dlzZBjh+Uva7TNaCDxKIp8mmsyUh4kqNtjR0o4WElOe+erdO06XC/M08e7tGy7nC6KNUAvSlBiEIVkinnLm4TSZR7RCVqM13Bz3ntgHhEDOjZwrqpGYDoy7xMsf/gk/+1f/dR7eveP1ZeIXX37FebpwOj1Q8olWG/fnMw1hNwyEkChxQHKjSkLTDqmFQSpRKwFlR8MBomV45RrX+JAwlHDrBbee0hcZHF1R2SeFLEqUwHG/Z4iRu5s7nj97QZ4npnzmfH9aNnXShBgFboZVZaBFpPnQh01TIioMMZkkXSvolE1AHUNMwVCn4O86uNSO8dhlQV5t49sw3mzr6c6L4P5/FnKrITXVHosV3QKQoDUbEOudE9FGk0hsEUXM4bpZkTAXK1qrb3iLI2Nztg7ONa7xXSGYHqqh+r7+pB+91ubv6xKB0KwzkOeZ6XJ2M4RKywVUGMY9h/0BEB4vmdwyrSllnij1gqr45stevyO5IQQO+wPjOLIiM3a/pmJWIm6sAJDSwP54dPUfISZz4wspkMYEUUynelwtZ0v1ba8PqgURdmpIbx9A9R0ktIpo2ewyex0QfA40IEmIkkhhoB0bKY3kPFMr1CJee8vCGz7s9uZY6E6F1/j+x+8FkX3y23azRdfEk1Xr2O/UPdPbAtzoUuxawzGARMLuQLp9BsNIu72j3d5Rgcc58zDbQNYln5iLtUg0DmiInKeJL16/5v50YpouvP36ay7nExFlUHuFcUgc9yNDilzmmfvT2QpZjJeLRKbLGRBiSOCKBrUaOhTiQCJwc/eCz370J+yOtzz7+V+x398Y2jpdAGvXXuZM5cyuVA6HTCo2tWkT4gNBhcrEQCBqIyGE94qMa1zju6LnhwXlYL0ugC6+7GsRu72uautwHMzU47DbczwcmUMkzxP5YvJXQ4oMIdGGwDACO5PQEvUCVsUltSwxp+ijoVWJItbYV0OWeq9V/E3KgsK62oLiz2NrsiO+uqA5XqjS57lX5Gu7BZT+AkEgWmelitlbBkC0Ld9DaVCdy5c7EluVqTbX2GxeBF8L2Wt8WCw4jytfrEVsH2RsvS2yIJktF/L5QiuFVhqaG4Iwph27YURFmHIDMUgol8w0meJOzo1ctpnDCr2UEmlI3hmxS9NVXVlCYIwDQSLDOHI8HEnDgBFyGo1mKGyKxADRZS1DCLTavGvDwqQIAVTHZWhLerdmQaTdCKX/ulCjHG0NwVHZyK5VR38T6XQmhAyw6OYOaWDcHQw9/ha++zW+n/FRHFlBvpVWsPBjthOFT8HaBUNZppddJsfUCSLEEVKipYESE4TAXCr5cqEonOaZSy40tUTTGiCB5szd1prvfNuTNmZvJeI74w74EASJhgmlODCkHRITh9sbbu5uOd7dMu53GA9Jl6FpbbZrPRxvaE25ub3j7tkzYoxc5onH0+OapL2tm/PMebog1dFhbQQfXFGEJGad2dslV2LBNT48bIhxPWB6kdjbJj1x2r89SXN+UIcgOPmV3W7k5nhkSIlSJvI8oTRSHEAFrZCnCmEmBCg1kMbg9piRpD4QJqbxWGsx/cnahe9WPdi+8XWZWaMW0BOZ0sR0YJfTivatXt3skdfnfOIYqD4EAut5RkBpiJpmtLTm1pZQ1HiHzU1OSm3u5W7SQa0Xvdfp6Gt8SPQNmhqio86VhZ4vHenx3LIMaDal5UotFc2Gygp2e3CHS1MYKYgEhgaqwZVJbNgL1v5MiubyNwyD0+RclzYEH8KywbPcKkEaIrbe+nuU/n6FZa1tL5a+TVmkyzN3q92lS8LCErKuzNOz0LoJ96/FTCJcczZE6mL8EAnSHdBcLzsNC+XvW8nu1/hexkchsoQVQ4X1gAvON1sP0J42/d6eoGqD4gMUUayVLyEg4xE53iJDotzcUnZ7qsDb04n7dw/U1jhNM5e5ICKkYUdMAxoCumu0IVHqTOtkcNoy7EFtFC1QK0EFjcBgns+hDcQa2d/ccPP8FcO45yd/9jN++nf+nBcvX/Hs1SsIkVLNpSg0K8KP+zvCD4W7Z2f+9Fdf8PXrNzw83FMVHu8fqa0SCFCVqoW3D/dc5pmgkFojtkbSxgEYQyApNAJ5wablGya217jGt0VTZZozawW3bjqX0/oGoV03n64DK0C0LoG2xMuXLxhjIOfMzWHHflgHLawobTzMJ8qbggQY9pFhDMQUuLkd2B8SiC4T1E0rtcxos/ZjDD2huQi82IYyRHcSgoW7JDVAUHfFq6bFrI7E9p+b70K6XBFKrZU5zzbNrauwetVKaxXRwKCR1GzzWFqgNaE2ZZ4LpVjxmvv0t6oZmlzX5TU+NCT6dP7GMAQvYXsXQzHTDrU2PXOjPE6UOVPnQp2sYL09PmeIIyFG8k4oBJoq4846B6ow50ouzWtnOwvEGDkej+zGnc1uzGZOYIiskGtDq9oarY2bm8rtzQ0RG8bUbtJQccWFZjhtK/TOZRpGuiuZBCtuUwxWNNeKaCOKGSFEUQLdGKJ3cMTWtZ+gYogECTRptLRz2lJgHA7MqdqQ6WhKKSlFxp3Ror5NgeQa38/4uEJ2K7vFylPp//QUgV3vuGkoGA8NXJvSC+E0EvYHGAbquKOlRFblYb7w5vFEbY3LeWaeMyFGDgdbwMRISxENUFs1xLW3HTczKIbItsVZyFqNAUm2qxv2e26fP2O3P/Ds1UtefPaKFy9fcTjegPiudTM8PQ47hmeJcbfnxavPePXZDxjGHb/65S+JcWChUjRrY16miTlnRIRdiIbAepnfJDAAAwIEIkYz4Ir8XONDQpXS6tNile5hta7XNYFu0ZF1DQciRLg5HkkhUHOmlkyZZkopnM8nzheTorrkzDmfkQDjHEljYBgCQXagCQkQUyMEW4uW9NZEZQWsO/zIOkzC8l5C7zgiRKMQNYXWE1XrH/0b0TshJjVUPZ+nBVEqmPKAoNCKd3SEUqNxBquSa10Q2dJtO3065jobfY0PC1913amyJ0RhQWkN4HSdZnV1n6q0qVDnTJ0K+ZJNR7U0okRiHEhJGZwekJJ3HRvEVEmle8/ZkWqOl3vGcaQ19VZpJRTrbtZm2ut5mmml2uBy02WjqyLWGQnOG/CLSeQZFSC5aon4GhNHZReNW99ch37O2a6hTSep610LRi2QYLMqADU2YhiIcXAXw8EoDjHasFe8Dnt9SvFRw162IFfkh+XAXNsQ72OyKgEN0e4eI3EUTx2m8UiI6LCjDiOaIpemzNNMUWWqlYoXvkFIKbrrR2RIkRaDz5es3KPm4tAhBGKyBBaaobdpHEnjyLDbEffC6J/p2YtX/OBPfsL+cMNnn3/O4Xhk3Jk134bG4wJ5uMSI0vy2kAIxRcbdjuPxSMmFVme0GZ+nD9wQbKitSTArvmHPIMJAI7XMoA2pBZkvtHpNmNf47jDqmQuV69O5fkSfILOhc9E80cS4KXR1pRmkaB2Lw2HPs2d3lFKISQgRcinMOnPJ61roFyEQQ0JEu/COF3+BLgHU1aUDxs2TEJ4gsorYOcM/nYboHZ1GC2412zapsNcLdJ6sAzxO67FauS00Ci+tEaC0Yj0Qb7G2FnxIxnVyRYnCwtu1d31dl9f4wJDuMPct1J6lPbLdgJpcXBBz1KtqGrJmydrXdXe+Wg06TJdVSX1IUllofCGERaLLjvFuiiAMw2AFbqhIVaqYyUlrlZozOEU9Blkuqj7LUStNGqTk/xa8wHVKX2uUoqgG5nlmmi7UmoBhoR6oTYa9V0uof0/q6gyF4jqyIiyKCWbW0Ato6QD0NT6R+J0L2ScGBn3eoVNRO9Li69JyjUuZx0BzGoCMO/a7vSWqrisngXJ7S7m5pYrwdcm8ubxzCRybHAaIY2IcEiFGjscdu92OIsJJIOMtTDU+mwIxDYwi0AZkSKCNw/HI8fkLdvsdh+MNz16+Ytjt+PwHP+Jnf/GvcLy55cVnn/ODH/2QcbdHfNCkYZyi7h3YaqWIUqPCEAi7gVR2PHv+nB/+4EeUeeby8Jbp9EDTxtwKWc1drMRAC5EwDIzPnnF7ODBo41gyu1Zp05n8+gvqufyuf6prfFKhNoUPhmgut69lYy9grXhdHXlS6lax1S5qJgJBBlQj4/CKly+fUWvhzZvXvHv3lmmaKF9kHs8nS9DFDBeUQGJgFw+YK1cGLxSlKU2FELqZpSXYFM21L6SB4NqVDTEePaDBHqOtUQWjFzShUc29Fts+iwRz62rqYtfivNaMSGBMAyGY6kGVShUzhS9VkVq8xRlBo/MIZdHMTjEs32bn8l7jGh8U1h6gqQ9cdn3F/sOP1e6kJwhRIoOMSMDoPHMzya0K3TJ6GCI6Gp2vFhtGVpTUcyostrBdU3aeZnvdTmcIgcPxyDCM1FKY0ok6Z4YhUcvE+VxJo5kixBjRFtgPkRLUNNnzjDZlPwzshsiQzAilNHsvtRZqLoQAQ1AihZQSt3c3BMEQ1Bi947Kp61HUJSxbq8x5YpomaitIFMbdQIyBcTcujmOdT3uNTyc+ilrQE8emH9CPvfUmWWkEVtAKNUYjqe92xOMthECtoMWq4LY7UMYdGeVxmnhzmWiqxGD82yDCEAJJus6dXYx+tA5hLBecE0c0e9torf5ht2PY7Rn3e453z/jshz/icDzywx//hD/7i7/g5vaO3fGGm9tbS6oNatl8vGWozac+g6JRbJpzSOz2e25vbsnDBPOZdglmkLDh19UuwxID8XBgvLtjUGXMM0NrVBFyTN/4nq9xjV8fbeXdbeg/PTks/DXBJ4DFvdTD8uim5gIUo7X2IbDfDQQRaq2WXFqxbsibgYDrsDZoBTQZMSaFAdxtx5ISqBpZwAwROqXBtShjXOwwxeV8HMhdsRkRe/1iBSsimzXJAmzJZn3auaAiKC22Lc7jhTfG262mk2vVc/NWr71LESG6UHsQK6qv+fIav01YxwSjxnwjZAGBjGpgvYpIRImIivHCaUvHw7oeYVUF8aFKVZ8JwZ3BpCC1UmslNzMJ6mhuP7bHcSSlgTJnKIUiQgwO1NCQqAiDM/GEFAzNKoIhss1aMcnfD9Jo6pJjrVFzRkSZLpFzEoYhsdsN1HoAIEhgqUGXQbKuTmLnkForpealu9JR2JTsnLF+0deV+SnFx3Fkt8VVzyLeFnjy7yJoMIWDGiMlJWsRjjvksIcQaRVaNRvInBIXhALoODJgbRNtNpihCCV0wXTIogRRiioF03xEhN1uh4rZ7u3G0X3gYYxWEB+OR16+esluv+Pm7hmf/+hH7PYHXr76nBfPX7A/HEnjjhS6haW1VlCWAazWGufTA+/uX3O5nHn9xa948+UXzOcL9XxmRIgEZoTklKKg2NRpEHbjjuF45OZw4NnzFzx//oLUGvt5YiiFGZjTSFlLkWtc4zsjOKetI7Cxo7ASrBjzSd/kiKyEteBVrZRSjFdasrUVgd04MI6DcV6HwG4/gijP7m6Zppc+yW9t+GFM7Mc9QxzphbVq8AK5+XT0uhXGC9Xm0ng29NElwbwgff9DbpdE7yV68S49AbKepVozJ6Ut5ciqUn8amyGxNSqCqCX50ExSTKRzjf2hV+mta/wW0bnfyyaqKeqDWVoVZ7DYZtG7AagNUGlU44mqHaytKjkXo+K5rNVic+Ibt0Uqki2FgCebPZOPWymC4nSCcRiI2IYx58ns2xkZ9gEYaNpIQQiS0FKtO1LtfDFPF8vVrEOYrRRDbVV5qJnpfE+MkXmemPNMSon98cj+eERCIKaBmJKbIhg9oVZz4yw52+aza1Avm/K+cd0iadf4FOIjC1l8iyn0SQtdTvSyXDQEoxSEQB53TPs9LSZ2t3fIy1dISqbVWE38/6Eqb4sx19rxjuOzSGuN0+me6fxoKUqEEmzqWaVRqFSUqZkzFyHw7PlznoXI7e0tP/3pT3n+7Bm7ceD53S3jOLI/7Hn+/DnjaNZ7t89eMDhvdr8/EmKkVLVJTlUrxnuxrs3br4Wvv/wb/vv//r/j9PDAP/vv/jv+2T/6R7RcOMrAcxmoEmhqU9C5CVOD0JQkkWd3z7h99Yrbm1t++ud/hx99/gNCqaTTI2HOPH71Ky5f/JJJ3n30n+oa3/8QbE0EkaVIDRJIPvwQQ7CNGeJuW5FFFcg7GaUULpczrVUulxPT5QQIz5/fMezuIML+MEK45VD2xCHy8sVzam3M80QumRgDx5uR3ThgU9oZxZQOpknJrXNmC10/0wpNF4Zv1RHVxcj2SYdlRV9lLWLxKpRofOAO4+KbZOflVq0ENd97DYoMsraMsGJ1lESS0YraGhaASEwvcEGHWqt/wL/uNf5YwzaOEfHjRcEMNuaM1oZWoAjSQEUN8UfQJoxpT2LgEmeEaLa0uXE+T5RmklopJQJCzs2t0RvzlJlz11k15NIGqJwr24xmUGtbBqaSd0vHEKA1LqcHvnr7FZfzI4ebPYTMbj8amnrYWRcnZx5Lps6Z+RR4dIvbkJKZJgB5mjg/PlJr4cvTPafHdyDw8uULXr56wTDu+MGPf8xnP/iByVneHNgdDo6N2X+lFqbLifP5BM6PtUHRQIwwJDtH1NrZx9di9lOJj6MWLNnkCQzLmj86kiKGwAahxUiJiZYScRwZ9nvEfZ27DeR8mbmUmYZ7Kx/2tFqRfKFenOLeddcDFDHtuwZUbYbqBGHc7RjGkefPn/PjH/+Yzz//nMN+x2cvX3I47Nntdjx79pxxHBnGHYebW2Ia1vcMTHNGz5PtXJfEiisQKELlfHrk9Ze/4uH+ntdffsHbL7+E2tgfn7M7DlQCo5o2rKoYIuuozziOHI833NzccvfsOc9evEJKMZ7gPFMvJ2Iar1Ii1/jgWBAK574Gsfbjej2uPFmnE9gAom9Gm+m9llKY55nz5YII3JQ9SjNu7RAZGUgpQlPGNFBr5XK5ME0TEoTdEL3F2Kc3IgWhhEST4uito6jak49V1cv76Tz8TWpaR2UWbJSuHLugsdZnXRDYJa1tKEf95ST4cznt3TYDgSTBp8dl7XTSzGZXrRDRei1kr/FbxEJ38XZ7rdRico7UrlZgvO0OmsSQkCDunheARqtKKRWJlaiG1qrThnyvRamV7Oo4aaMo1KkET/i69H2hINHPD9qYp0DOM5fLmRCVPO3MBMGpBSlGYhDwzWctmTxPtFoZVInJ1AtaLZScKWXm/t09r7/+ElBqsc3tbr/ncLzh9u4ZbWgMu4Gxb3BpXqAWQ2RL9i6S2QaBbvjqsgxyXvuXn058pLNXlx1f23hmbiDOi3U6gQRKiNQgXBAemlJr4zzPPD6eIcVler8Bk4K6oHFz/k2thWnOXOYJVB11sp1lVmUoldqUKRdyaewPR17s99w+e87LV6/44Y9+xA9/8APGceTu9oZxHBiGkXG3t91jTMZ1rc05rz6cVqq5rDhiJa7UcHp85PH+HfPlwptf/II3f/NLTg/3zG/eEi6TcezCRMXkSyKBcdwTtXG7H4nS2B0OvHr5GS9fvuLmeMPxeMM47ggxEWpDYmLcG71BvMC+xjV+Y4hJ7ATpg1xGJQgbfvlWait0NFNAQkRR9vu90Qqq6TDP2QrZNA4Mu9GKvHFA255W2+Ld3moz/npMXnhWWrVp/zQkYkymMV13fsYoFHzaeVOwwiqRh4qZM3REdvM5t4hscEH3jiyv7knr3Raqgidw8SrWaBWCRPHBLqMRRRF/qgbVCvyWzdCh1so8z9R6HcK8xofFIinnnHSwdnmt1Y6vLIscXpZIJBDUTABCEgNbjjeoNuIwGKJaK6k1qtqmUEJgGAY3BDJLWkQYUiKmRGuNEGZyLsTYXIa52LR/n/jXlUHeVQc6PWGeJqDZ0GU9GG9WN1x3tXWiYhzWIUZ7TzHZplZNYWhI0TbMOXN6eCTPmTdfvyaNO4ZhYM7ZJSrXJy61cjmfmafJ3y+IJPo0aPCzivtIXOMTio+mFjzVapOOsRhTzlvxOSYucaCEwFsiX9VGbtAezrT8NepT+3EcLbEMO2S3B4SCormQS+bhfObN/YO1R3zcOYiQLhdStEU6XwolF159Jvyrz1/w0z/9M374wx/wr/8b/wY/+clPHM0Njli5Pp2jrLVApS47ZStoDYEJuhAl0Fp5/Oprfv4//BNODw/85X/9/+Ov/qv/hsvjI/O7e8LbB4KCzkI5G5I7SOTu5jktCDeHgbof2B8P/OTP/oJXP/oRu92Oz15+zu3tHVIbcdghpVAuF8bjHWl387F/qmt8AhFE2A02lBVDXIrV6GLsgWB8WefOdmecGM0hCMG6BDdHajWTgzlfAGV/PHC8PZJiYtwNjEOilsrrL79mSG/NnvJ4pJZKzoX7+3eczo8mUn7Ysz8M1DqTgjKXgdJmTnNxXdm1EajqSKeCSLSziiNIT9HYsBTggcEep4ZImfSYDYnY5rrR7fgajaoFs8AMhBRMwsulWCKBqIHUX7G6q1KpTOczZZ4ptTLN0+Irf41r/MYQzD3SHSQDAZ0h58w8Zy9kFRrUUNFgx+FIIg0DUYXDnRVuTZW4O5BLpSLEYnqxEgMp2YZRVRl3O2pthsgmox/UWjmfrWtSa0O4LJJV/SfebcSLYbyYLSXz+PjA5SIIjed3N0iKBLUZlYht+rQWGkqUPYfRTBs07yjTSBHYDwOHYaTUyny+cH48EWJknipv3z4wjCOvPn/Js5fPXVqrOwUqc5mptRJTNCALk+2SXsg62T2IbhDoa3zf43cvZN/b8SwTwstFHJE1R5ASIiXY0NO5wSzmKpJ1QiWQmpKwtsYu7RiDCZ+3WpbCMpfCnLMVl74FlBBIqoRgGnvzZaZmS6TDuOP29pa7u2e8fPmSV69eLST41nVJWm/zqIud95bPtuWysn6D2iLPlwuPb97y8O4d91+/5uGrr5lPJ/QyEeZMQECySf6EQNiNDGmAFOF4gOPI/njk2d0znt89YxhG9vs9yaWBIiApMex2RjNIw5Xxc43vDjH3HuPAhicDXyJdtbXrx26luIzXJmIWk0qk1sowjqa/jJpW4zDYYMZ+z+GwMz7t4yPTaaDVRguJlpQYZh4fAq0qGpQYIkO0iec6jFZYFpOgMwHmp4qsqh1RbWaHsKG8PUFldUW68DZj35iuq3b9YY93aoGotyHFigzxny5GL755rU0XV6KaM3nO1FbI83wtZK/xW8XW7APwgSxz09LiyGKrlFhRgeQbNUFIw8BufzAVgBR9ALotduxBZVH/AGwArOlSyMYYqbVRSvWp/0qMkRjbk/emsnZGfOLMefSNkmdaFUrOru1l6zRgHVLj2bdFss6e3yhGMQRT5/HbtCm5zFymCZFAGu5RCaaeMCZT/4ldkcAR7GaymqqJVio62EZVNrrUXTv3Gp9OfPywl4f2zOGtvRYCGn2icn9gd/eMlBJDM31HUWhESjMzAHX5rdAaIVWkWIKYpgvzPFFyZr5c7OD1wSu6BE6MpDDSaAwJhMow7hiGkWEYCSFRijJNRoLPpdiEdVNaXQdNavFFsflcW6mizY3ElBh3O/b7AzfHG57dPmNKA224oMNIIHA83HE83Jo+3vGIHA62g73dI4cd437Hze0zxmFnE5pqhPbghPwYBI0BHQZ0uFILrvHdYeWc+447F3YZ9pLg0jjJhzuSUVl80GMYR29FWoFXW3W/dHPF2u/35FK9NT8uLf04JIadUWisBe/e5ym4QsBT3ry4zFbQaJPGneuwoSfp5vN0utKSXp0rYDy54FPKEcE2pKIR0S4h5uWswdHL+UlpK0fWb9FmOrg0JVeBaja49VzQ2agWec7G63Od3Wtc40Nj2TwGk9QKS1HnOqnBctKyPsQMfwgRIRD94FeFmoSSAoSwDEKJNitYg29eg639bSEbQmO/3xFjpJRiHYg4byTqbP3WZnq1cbA8V2pGW6HWmVoaec5M00wIwZ4HL6IlPCmmtVlBrtVFbZsyxsTN/kAdeiFtZUiUSM2FrDCdLpzHEyEGxjGR3IyoNsvddUikGF27ujKPu6XrZO6AXGe9PqH4uEK24cMT0EGTboinIrRhhBhJz54z/OjHtHHH/TQTLxPURpsq86XSqhJRqlYkKBoyGhLaGvfv3vJw/45aC5fTiTIVe81uSJACcTcwpD2qSmRgSI39/sh+f2S3uyGmkctUuX+4GM9mnharyVrqgs72zzGkyDgMi7bmMBg67J1Js+IcR25u7ggEXr58xQ9/8CPmy4UyXSiXiSDC8dkLjrfPCcPA7sULxrtnhBQZjwfSwQqIcWcToIidoOZ5JvoQThiiFbG7Hbrff9Sf6hqfSphcj2lLdq5sYjdYkToOA/vdnhgDu92B480tMcZlU2bHvBBTQLXy/Pkzbp/dkfPM6XzP6fLAXIXDzcEtnWHcDxyPdnwGEkEip9OJL74aqKoEH8DsA1gxJSQ0mhRCjoS2ceRzpHShGdBLUS+Dl9lSmwI36gC2sVXtszC2IW2sep3dPQjx1zFN2dCtchVartSpIRV0mpln0+1s50Kb7TwxZ2tt6nvF8DWu8Z3hG6+YEjQxlYAhGYhCXRBZSYEwJoIkJA5IsE3jMA6M+z2gzLjCh6+by9wpAnvftIpz5X0D6xx1gGEYF3WSIZ2ZppnWGtkVDFotZJRahWE3cri9RaIwT2ce3p0p88TpNPL4cKLVtmjSxhhdHcFeE7VhLlqllWx0odrYjyPj8xe01rhME5dpNqOg3MjnmRoKD/EeLZUQTeZvHAdj3bdKbZWUEi1X5sNE3u8YY0KqEmI0Hn+Mv/FPcY3vV3xcIdsRib6bo7clwCFFiJEwjsTjEd3tjbdXQUJFnSbQnMJWq7UIalWaF5rznLlcLqYhl12qpL+2qOk+uhezjXiCSCPG5BdbVKa7V42XMxeyUxZytp/iE6IdnYnRitfAZlu3QY1CCOZeMozsdnv2hwNBhBICRYw4v7u5Yby7JQ4Dh+fP2T9/QYiR3fHAsB+for3q0uzeCtLoSLWpT6PXhXmNDwijEYQFHbHCNBKTF7duQxljdHrAkZTSegyHSEqBYTAry/N05nR5ZMoTc52op4a4yDlOUYghEAeTEUkhEUMi52GR+1nNDHoN6hu1FljbmZth0Q2FYMFhlc1Iqf2DzWLJMrls9w8LzUkWpJcVxV2eu+uSeHQ1A2/z1gwtW5eozsU4sq1LbrVlqOTKw7vGh8XaeejHvITenQg0NwNajtM+fBUixI5yQk8DJiHXVeNWg52mzVztnCNqxa0Pey6Ug2B7vhDIo82C2HENUAH1+xgyHFNiGEdKMfeuWt2YoJixQq0+aObnnYU+gQ9IIqYn61SEGCJpFLfbBdQoFqVM1nENah3YaSbGTo8CHHm2bmojz2ZvGyVQ5kIZC1FtsJRwVfn5lOL3VMg6ny1EakpoEMZnzzh8/kPibk/47HPST/4U3e14/foNsX6FzDN6UUq9mFJAR1xCo5ZCKaYduwgg17KKs4sYbyiANPXWqbcnhhGA3eFoslUxoQTmUpnmbEMaufhEtlJcIzZ4QhZfkE0xSZGiq6+z2KJBlbTb8eyzz9jf3nI6PTJNM3meuJzPTOczIsLu7o7dzR0hJYabW9Lxxk4o44AMfkbqjkf9a8TcvqY8UTKcpwuXPDNduXjX+IAIMXB7e0dyE5AYIsOCwlqn4bA/2PVxx/5ws7T3Yhy8+PUEq8J+f+D58xdMeeJ8OfE2JlQb0zxz/3gPajy3pqZYEKIPloyRkHyQKsCcM+GiSFBiqsuwpvQBmM0m0a6tk9O2abXbZdvOF5tTRhq04Dw+BYkLv9ZcudoT/l/vJDVP/N3ituZCmasjsorMQF2LW0u6fchsmf68xjU+OBZ+um8wh2H0rVhFqtBiYxhG0jgQQyKEZPQZQk+zi2HHIJEmJj+ZtSHOOZWL56pYiWFARKhtIMbS34HPn1g3IjqC25o6rx5qiVYsp8TucHDXP2W+nBeufC4FmeyzpGRdRUOZR+O+AznPCDBPM3nOtFaJjnGhXSLQjEeGNFAGMzYY48gQzX52CIkkdt4xIwSjMOVLto5JbgwxMV9mhjFxvL1h3I20qzTeJxMfMexlE8E+2m+obIqUcaQNiZvPf8Bnf/fvsbt7Rnz1OelPfkobR77+q58zXCrhdEJPhVwbNZuEiMbmIu1mI9taI08T83QxyZt5ouXZ25MjEkGiyf8MaTBry3EkpMTx9jnD7kCIIyqBy5yJZ0N2pzw7x824sQoMfeIzRAjdSUzJxaw4UTU7zpQQEcabG35wvKO1ynA4cHz+kpxnHh8feTw9okDc70n7PQShBXMzszTocl5qnrcLj2gZPDNP6VYK70733M8XHsr8e/qTX+P7HCkmXr38jHEYOR6ODCmZCsHhYAWmF7WG1CZitM6ALWdbCzE4OBQCt7fPGPcj8zzxeH7ki9dfUErmdDlzyWcEJWkj0hAxZHbYJYYykMbktpHW+pwLpEE43gTGQQhNrNhtAe3uX8CTglZ18/t63YCrDst2PVwrNoP2ErhhVrogQQnBXb1EXdnAuL+0ghYol0w+FygQzopMPnxTBGmd0+BKCTYmfXXCvMaHhwiEgKjLaqXE/rCnpkSJlaiRVhu7cc+421shKwlL09K5ewCkZBQgBU5lRstEazBNlTJNxouNu2V9xzRshkC99e9F9TAOSxFrVAEhl5milTSO3Nzd0eqBYRzQVpgvO4LAPNvg435/4ObuQEoDMSYvZAWthcv5jLbG5XRiupxprTGmSEi942lyfS0otRnZJ4TAftyzH+w8NcbEEJK7k7mjV1OmcuEikGJkPl98GG7H81fP2R8O1HItZD+V+HhEdoUSrRXubfCw2zPePWP//AXx2XPis2foMDIcjkgakJhAwqLZ2hqoNJtR7lPCrdFaNZpBrVbwdWOC1ujOAn0Se0nObm8nrnyg2G6zlGpToj4pCtC29pcSlqmu5tvfWgu1zN4iTT7gLMRhzzjsoSnHu2fcvryQ5xmGER0Hk0jZ7wjjiIpQFSrG45PWie+Y7I/Igm5be8j4S7VkcimUWilXB6FrfECIBMZxx24c2e8PSyHbKQTDMDCOu4VyE4IlyVobRcuilwz4IEhiDDsQXNXA1kepFa3Fhyusk6c+ULWirM5LVee2qTnuqQ4sU1bi9184Sb8uNgNgCyrr+gSdSuDcWecc0J3BpLcm1/+xkAs6paDZoJhW25i3qkh1SkMzY4ROmZLNW5Xf/KavcQ2LTkPxY1RUFqMSYqJFc99a1Ea2XNO+Y1pyhK+ZZLrPUtduXmumGmBHfEQ1LOvAXL3MkKfLU4WYnIIAGkzGqoawrN8QApIGNAZqyQzj6Hm40nKmy9uFEBc1guAIb63F+OSuA1+r5W9zyIy+jtf1GYObtYTw5HqQ3in19axmf1SrDclpNM3akguqyv5yMMmv6zDmJxMf5+zFWnipQhhH9q8+h+ORu5/+jFd/53/E8cUr8m7PdLiliCDjnmF/YChtQU/7lH6X1DKuqBWxtVRKLrRmC0Jd+LloJTSBEjidT7QYiWnggDBIMMqAgMRoiGhvI2qXK9G1zePJrZaKSmO6nHi4f0spmcvlzOnxntoq4ziw31kR8OzuOS+fvyKGyJwz6XBEdnv2KaG70b4TF6NWLPlK97vu8iZitrVuYEJtlVYL8zzx7s1rLo+P3L/+ivuHd5zOp4/8U1/jU4iUEp9/5jaP+73bVxrNYE02ie4U1weXTqcTDw8P5qAXFEmOfEbbMJYy83i5Zy4Xcs3UNlGbtQ1bjLb+JIAGci6czmcKEwxmbKCtoBQqgYISNVLV6DJ9HQaCU216eWgKIp0/u+3ke3qnF7NeytIQh0m3KO7m6jKZ7QVxUyhGNowECOZUptKMsoAXDW01kljezJaze41rfGesk/0t4Osx0Z26QghUB4QadpDGbjXn2I3ZDOHobJ/XsAFlRaGq3XE5Ps0hrNTstsygOlm+lsAwmmLOAkhh+8AUbdBYSIiMiCjjkIg0yjwZje7xgVqLOWimYaEXGF8X57may1eeM6VYxV2jd0Q6ZzhEgsIwRCQad3cc7Tl7kds3mrWYfNhSd2CzJSVXW9PMPN6f3Hr3Cv58KvHRhWxTdZ1FJe2OHP/kT4kvXvDqX/17/Ph//G9y++pz3kwzvzpfKLnA4ZbheMuuCcP+gTgOvtScJytre712W7t58rZCAfdIbxXQRtFGe7znVArDsKOGyEESU6k0THyaEKiqlLZFZKsNtnQSvGIcXIWvv/qKv/wf/imnxwfevnvDl1/9ipwz+/2Ow/HAMAz8yY9/yp/97M/Z7fYcb++4efaMUQLx5pYxZx8km21ATZVQC6XajrFVoYoR4aniTiRGZK955vz4yK9++Qvefv0Vp3dv+fr11zw8vPv9/MWv8b2OYRj5yU9+ZoNdafBhD5aBiXUgA4ofk61V3r17wy9++TfkeYKgaGhIgGEXGXeR2grvHl9zKY/kkpnmE/N8sQHHNFKTtTAv80QMkWmemNuZsKu0Vm2Qo2aQQNZKaImiGTB+uhKMnoANOfaK0YBaHyaBpWDthey2uO3XG9ZpkaXK9LK4F7GqTk03DrwWRaoQ1cwiNCg1FJqjVn2ALDiq1ZZn1e9Aka9xjR5exAIqJqelEdKghNDQCnMs1pmQPkzZbLMWxLuXlWoEb9CGVNvwBYHdaMVolYL7i/QSD5qZhKBYXvKiMsbI4XjDMO4WAxWT62vsBnPi6vJXIQXqPHN3c6SVzOnhgTdffcE8Tex2O9N+TYnaqhWa2pjnifP5TKuVMk8u92UUPaP72Ofq7oJpiOxCIgRx6czk79mK0lYrJVfybANpS70goLVSghW0tTViilY4X+OTiI8rZHW9NAz9jPs96XBDOt4y3Nwx3N4R5ITO2cjlIZgFa0rWguhtyA2KYi2S5lONHZ1trFlDl0vTRqkVSkGCCT7XDfLZW40rcrxeWP/Z2p/VTBUul4n7+3c83N/z5u0bvv7qa3KerZC9HBmGgePxlofHB0prDPuD7Sqj7SyTBGuhqOnUNjV93G+bcF6MAFUX68tcMtPlwul04nI5k69WmNf4wAhB2O/31raMkSCLeCqwFoL9t76+cp45n0/M84SGamhkgF1LVBKtVXKevUVol1IzglDElToQtDVKKOQyUykQmhWi0mjSkZRI095d2aKt8vT9LdCTGG9A+796SxKnEyyP0+XU8H7Lf20zrsWnrqcRf8yqoNB6t2ZlLqxtYf/lSo+9xm8fm6Pc0VmjHfQN5mbTRcf81/WwpMmew7yjGHtB1wcanbqzZNSFPtMo1dQGmiaGWojNkNS4tEUNLe0D0DEZ190kpSstJkrOVriW4vq0YUVPNzm757S1C7oyEvtq7nzdsNASjFoQJCw9D22uR938s2zOG/13W+KNPJte/CK9d43vffzOhaxiHTkl0GK00/p4IBzvSLfPmCTyq7fveNuEt+czXz88MhUTUe5C7MNg5PAaTUqk8/NMqWCilsI8XVxhoNG10xFrAUpMhM6JHUaiP19wf2dxGSzoKO+6wKyYbUute7mcePPV10zTxF///K/4J//oH3F//5bT6ZF392+ptZiO386cjvI0c348sd8f+OnP/pza1F2QRlIa+hu1k5ZzXufZpruLFwS0PsCWmeeJt199yenhnseHd/zyb/6GN19/SbmYIcQ1rvEhsZXcWTdOm1a7I4wIJi3VCrlmHk7v+OKrv+F8OVFbXtDSYR8Z9hHVxuv7r3jz7mvrljRztwoC51ao8+yv0oBGqZnT/I5JH1FtVJlpUikEpqy0Fq04nq2tamIA4ujsBmcVaCu06hteR2PVSwJt9MFTbXUVTW/F+PVqnNdWbdetFagYXaBBbLaRjiEShoAGJYxQ1RRKukkC2gfJQLfc+mtc40Oib5r8+A7oMrEfvBhswUq0UosTZQItNAK2UQzRhid7/kIViV2CDkIaGGIy/rdYd6M1JU+zrdnSyPPFpK1KIg3RBiElQI1EiXSnS2MoKK2awo46GNPNVLrudIpmRqJqEpnT+WRD1dPFnDh9viOmwYCjNIDPyFiOjgtf12yybW6k+nBXLoVSrMvZu6pbh7R182ufdbrMTsVoXOPTiN+9kFUozdokGgc0BHR/Q7x7SXr2iotEfv7la+TdmVOeuD+fybVyvkykYWSveOEXSTUaIqK2m8uXmelyopbCNJ0p1YetOg9PhBASIRrHNo07hp05eVkx60T5sCbzrQZkv8RgO+AocD498vO/+mfcv3vHX/4P/5T/+r/6//Lu7ZslaSvdccWI8F/88pf8/K/+iv1uz8Pjo2nFHm549vwFz569NBmjhahverXns01tWpGebZgtX6xV8/jI3/z1z/n6yy84PT7wi7/6S969eU3QxtiyubpcIaBrfGfIInz+9IDZsEuDD3+g5Gro6dv71/z1L/+Sh4d7pnzmMp9QlDiKaT6KMpcLcz0DSkxmFiJAqYFLM83Y0iZqyzStTPVEbpOjoQ0NhnS2OTFpsKGqUqGpceNSwOSgZXm3yjfRVfCBFe38OcyFS5VWC6XOSyFbddMZcRc/Chi3RwkVQjV+7pgGhjCgVQlNKJLRphSK6Xz6OaqDw9Kuo17X+MBYkP+16BQxhZAgSomN4KhnK1BzRlUIRMZoA1Ld3ASgFqME+agFEacQeUGMCC1AE6F2xZ8yU0tlvpy4nC/ElAiDoNIMjZVk3UQE1b6hTJQcbI5FvZB12bDDzQ3jbnD7ZlPfyfOF8+MDpWTmYvMeTdU4tKPRj8IwItHAHomuFCRmRRui2WG3aprvTStzzuQ8+xB0pTabqUkpLFz/fn6rpTHnyShT9VrIfirxUdSC5q22xWUrmCVtC4Falel8QUtjyrO5aXkx2XX0ukizBIHW2xIsjltWeNYnVIBFo6BPdm40MGNMhBD98p4wOjx5HnuufsVQ4PP5xOPjA4+PD5weHzg9PgINcxpSe5+hi8Bb23beTTw+PHA+nxEJHI833jINSxukdTS4+gCbt1u0VbO79IV6cTrB+XRiulyYLxeSwBB1bW9e4xrfFfKk2c7TK+sNzbnorVVKLUzzxDRfuMwXLtOZpo1YhejSU0UnSpu9xWnJRxC0Nlo1mkKuNgTWtFHINtCl/Q0oTYO5+jnCSbWWZwBDlpb3uR7wT4wQvMOxTnHL2mbdKBB0Piz6TQrB8n30whT7GVz6R1WJfh5T16Cl03Y7ZWlpj14X5jU+LPTJsd2HjcOqMtBRRtqimqPa2D5sm9NkOSg7f1wWExRcQcSWh9MM+vP5ILU0G8wynfRAC3UxOjHdWjEtdb+Aru6WTjsgRJuTscSNNrNyLqXQXKu90ySedEll83l792jpIglQF7pA2+TQX7dxNL66cYtrtSHx6yjmpxMfRS1orjYgKUFMTK3x1du3CMKjfs3r9lfMCHFIxHEwYecYUZfxWZyHgk1cau0IS6WUmebDUasTitEJYkrc3D3ncHtHGkZunr1gf7wlxsT+eMM47nj+/Dlj947naRHb2yNoY54uZIF3b17zN//853z91Ze8/uoLWi1uTGb8IJMRaosFbyuFh/t7LpcLv/zl3/Dsn/4TDscb5mlGCIQQmaaJecrUWng4PXC6nHyhz2gt5Dxz//VXnB7ecXp85Bd/89e8/uIL8jwxXc6g1blLwQWpr3GN747OrXtazHaumaGLinK6nHl4fGSeJy6XiVYt6YlEQhwQbQTzMzBuaxVytqHMkouZGqBIMetJpZmagVoHo1JoVEvYavqZ2ho6AcWcfVouaG2kMbA/JuIgZtFJg2glrLrMXk+sYGhoaJ70uiyfKlWBPpwlQsIGtIIGQotLUm2oSSA1Qaq5fAWBKAGNyjgORAk+ZKIL+rUUxeIb26vEzzU+KDZgzOaQMStZ6/TFaJuoWqwY1AY1WvcwEJAYSd6ejyFQg7X8zVTAS9mwopSd3y0KESGJAU4xhM2Qs7uCiW8KgyybQFWoFcgBaWbpLNoQzHrWhq2Epo1cMrUULtOF8/lEzuWJy2d0nfYQ4yKrhXNjn66gToN6ukHc5u7o9IotTdBAL/s5z5N/f9e1+anERyCySm2NgJCGARlGK2S/fk25zPzi3QP/5IvXPM6ZZ8+f89kPP2e323Hz/DnPPv98QWSjS/c0TD6jgiFE3dO81WWRBi9k07Dj7vlLXnz+A8Zxx7OXn3G8fWbGCONIjIlnz56x2+2WIrhzZIHltlaLF8yV119/yc//8p/yy7/5hQk3l5nUvekPO5NGaZXSbJd5uUw8Xh4QEf7650eCBA7HI602xmFHjIl5zsyzWeCeLicuk7VltRVoJhb9y1/8DV9/8SvOp0d+8fO/4u3XX2ITqTO0iki0z3W1qL3GB8YKOnZ0xfl0mNxWt2V+PD3y9t0983ThdL64BmVAiMQwojRCsAJPtdGKkKdG00rTQtOMSQdNaJ2s2GVG8cS4IJmRqCOBgZaV/KjUWUxfcp5ppTLuIwqM+2ie6YAkS2oaXNHEKlADRZubFDgdqRe4xrWNIBBFnfMHoVVitGHOJj4Mqlgx3AvZ5O5+CnEMMFjBXuZCyf5aTZdKRDrB/hrX+IDom6ht5bbktl7MqpLFdJpbbYypGIIq1hlILklFVDT5Bu/JMblFau12UXlSyA4hUBaN2rWQVel6tSz2sYrSZnU6klnIiz8GDMgyK/mJkl2u8vRIzpm4pRPEaLnZKX9mKQ+9cNX1I6zRO6mbr6wDU+b6WRbK4DzPSwfXbHPrlSP7CcVHIbLLRKGY8kBTJZfCNM+czxceHh55mGbSMHB7uQCwcwkO64j4YEoIZs/nq3xVFugHomxaEUYOj2kwK79hZBjGBX3tA2QppcV67321gO3vhrhkcs5Ml8k4uTn7/Xgi7C7aJ5lt1XWaxDxPnE6m83q5nJnniRSNF2sqCm7q0NaBlE4rmKaZ8/nMxYe6cs4IjaSezIVv/QzXuMa3xzrRq5v11Ck6pVTmPFvyyZl5npmzDYGoytLulMWu2RKX5ZvuAOb0n1Z9QjnTmhe1FFRcYWOZZJalha8qaBO0Cq0KpSitKKGYXXQo1qmhubyQdPzU6QR+SlDVpZB9kum8uFxXi9nJrkNkvRHrxABlMVhYRszECwyVhfokrnwiy/DZdT1e47cL3VSxJjiwpc+w5Jl+hzUP2sNW+oEVoBJ6r6W3/mFR3gBHUJ8e2319h83a/DXvlpW208AHH1nkv3QtPN8bpO6XwKa4lk0e/QbauqR3usvgskHc5r3Nut7OvnSQqnpnprWnFMJrfP/jI3VkTalO1KYL5zLzcHpgKpnLdEa1GprTDAWSaJOHq8+0uQ6hyiVPvquyorJ5ksTlPyQE4rgnDDvG3X65pDSgYDvAmBY9u2EY2O12HI/HpR3Rd3PBV83FebHz5czj4wOX6cw0TYgqKUZwp5WcM2R8gKQui8TsaiFPF9589SXnxwMvX7zi65evGIYdIQ6EONBlUmKM1FqYp4l5OnF6fOTdm9e8ef01eZqoeXIUSRhDIgWzzh2HgeFKLbjGB0RDmatJ3c0bPePL+UKphXmaeTzZVPH59MjD/TtKzry7fyBIYkgHgg7ENuCQJRLMSGQImaQzpRUul4nzNKFaaVxoXIAGoaJidAJzJopEb31KSDZUNYyoRMo8ky+FUirtUmlvzsTB0NDhmIhDMJmA2BaOqk1vQ5BIIi18eRE7lUkAiT3BV7rzkDZFqllbSgHNatdrMGqBOMdPqtMSohezdp5q42DfZcteTF+L2Wv8trHqGHeqzLZeSzEgKDEKZsbhlqzdyKcjuqyFof3a6Tcsm86uPGDHvRJUSBIJAQ7DaCpBMRDGgTDYvEcSl+tbNp2Wt7qxD0bf7Z9k4duWYucV000vRmoSowwOw0CI0WgFm7mVbys0RYx+0IdDUSvYoz9+pWasPxd6Qd/Jivh94xX8+YTiIwvZhrFGrZid8myDUtPEZcqoNoKY1mvOGQkmqdEt+IZkxaaocjnBNE82+FTygvaICHEYEImkcUfaHRh2e8b9gXHXregg5wK+gHohezgcOB6Py/vtOnfGz4HpbGoFjw8PPD7cc7mYyHuKiV0aTOzdnbb60FbXfAUYB/v65suFr+cvGceRly9e8uLlK8bdnuPNMw43d/7a9rqtCdM08fDuntPjPW/evOb1V1+hNVPniYiSgrAfE2MMC71hCNdC9hrfHapWyOY5czqdKKVwPp148/Ytc545nU68e/uOnDN5npnOZxtAzAVhYEiJhFEHcDQUMcmbQWaiTrQWyGfl8X4yikE8oeHivNGKROPnxjAQxI7nYbBiM0gijkdCGrhw4UHP5DyjpXKaLqhUhl1knwfSaARdGdqyflI00fYhDEjqAu7BOL0ifebUknkttFYQNTqEVEytICvMzRDe0ty1ywvZYIM33W4TbJ1bwq4m6r7orF+L2Wv8tuEFoA8NdgQVgZhceSAYsVU9tzZthIW3KmhwVNU5pqtls26KZEVqddc644gPEmlB0GFHCuZ4yRhhMLWE5NtEe2u6IqOue7m4UvoAVh8CK8UGRedpIhd361vAKlMRSjGZyokE14R9yhcOYZX1MsS3utyXFbKtuQTYewPbW54sOIc2fXsn9hrf3/g9WNTii0zd5astbczg1AFz7lgni+FpK2XRem1W9G4P1O2kY3SduRg30lrvKRL0Qa6tSPOT6FOfPgRSuw90q8u0M6gtLAk2DOI8w96QXJOXN3CatVeLGHo7zxMigXqo/p5gO3FqTkeZUoqpFtSCVrPv6+2V7Xe3tFM+5o91jU8iVNXoAvPMNF1M9m26cL6cmV0Z43y5kHOmZOt+aO3DUr2t6Qes01LX9r1JB4lrWq58OrfUlK4A6+uQ5WmWtmZge2x3vduAUmkKrSmhdR1Lf86mhrRuJqA3ZwhPWt4qXfgB6q3I3o5duYOyfCbd0BL6+amhS1PU37cn2RBWBYPl+35vVOUa1/i26KmvI7KquhyPPed0ynU/xNbU9dQEQHvXv5NkOt0NK4A3JHnLaV5wttr8ddeuJI62OpEIYJWXg28knU5ZWClLa8dj+RxbKsFyWZ/Mlt7TQvb9AtXWNQtq3edc1u9Tt3d9ur6XIvaaMT+V+CiO7IyYL3mxdsIUkxWiwDAkbsOOAgy7/SoJsrRDLEGkmGipLi4ei1tJjIgG4mAGAxIju8Mtw+Fm4caqZ8k+sZhiZLfbcXt7y/F4JLk6wrYFsSwsYJ6Nn3o6nZinGcCL4EQaBpuuTJE4DKiaY1hZ9CqN49qL4VLNxeTx4R1vX3/Nfn9g2O24ub3DZIrsW9PWuJzPPLy753x6YL6YjqxoIwmEFElB7CKrteg3mfDXuMY3Y5om/vF//4+Z55nHhwffWBmdoNRCnjOXy8UGKWujFdvA0SpafAK4ZWo15YHWCk0Lqo08V6RFIso+Hbndm8byuQpTdQRJMkpd1hExEUgkSYwxIUQiEGi0BoebAQl7aotMRakaSUNYTU2C4jKThCAL770boSxobPAhGFHUeb0SPGE3RRKEAbQafcDkgnCjAzvv1FLIbVNk+2bXziOBmM05ELECoTsBXuMaHxJbV8lemIrTVFTN/c66Gj54qIJE58BKXdBZVKD20tcevykR6XMYeZo9vxQu7x6ZTyejuA1mORswXfaQUodZnbC6Fqt2MYtoXzCWp1ul5uxSW3lFRUVMn1bVfvoMjCJL/q21UWvzl/Ktr5pCSCmFPkC28IM7ag0G/mwGuTrlIEhAkxf7K8vgGp9IfFQhm7EWQy2mODClcWkZpJS43R/QENEYbfBj82jFi8YUSaXLW/VdmSxac8NonNiQEvubO3aHW2JKpDRiaKhSm1n16DiyG8dfW8jWaj3Bvij6UNrpdGKaJ9tt9vc0DKSYSJudYi5u7dcaWWeqmsRHLVYgtFq8kP2K6XDk7sULWHQo1XeYjely4f7+HdPpxDzZiSagxCAMIRCDkIL5voflpHJdldf47rhMl6WQfbh/MCHx1tySch2MUHX5qa6RU5sVeaqUMrsAeaNWL2pVEWkEIqLCPh2RvVFv6qUyzQWotCY0zWaQMiYEoxMkSYwpIYTlmFYVDseBGCHXCJNSaiYMLie0oQpIgBhsult6skyJINGK2DhgQmGVnmxjECLGs21ViQM0qVSZya0n7oY275RooQiEEBedaLBzWUomIVRcdqhWLyquhew1fovYFrHLIBU4AKQgzY5195wNwW7rReyqL6s0rXhvkbVXaPJYrVbKNJPPF8qceXz7jtO7e2KMHO9u2B33LFJcyagzjYJWL7CX9qOrgSA+C6M+3GUuXqVkqqugWB278lx7Ifu+BKYVrN6txDsyqm59bTm6Px/0LmUwPr6fy4Bl9gVYfvZ6wDaY15z5qcRHOXs1NaWBuqEWtIXMvoodI1utuC5B0sWBNihtZx7Iph0Yw8b0IHkRm54Mb/Xoi6ajNvDNacq1XdEL5w3HZtOiWFuV62JoTalhLYafTn1ae6VVX+BDXuRBEKMTdN/pWo1SUKppaKLb3W+/ul7vC/qaMq/xXaFNTQFjmpimix+D1TsJbWlt4oWsqk3umztP12TcWDn7z7X1KZ4AIykOAKQwkOJA0+AtTiU69hqIVmw+oeM4yiOuUxusoA5BkBacHrC2JLcTzU8vYUF8RGzNbznsTnJYqA4sVKaN7mbbrF9Yzgm94DeqEsvrddnAZRDmmiuv8YGxmvqsdAL6LRukthdvLJc1PyzH98IveC93bNr+tVaTj8uFPM9mSxsju8PeCtbYlQ5YKXpLDnq/ja9P1kanAW75qU/b/d8kxPW6YItKu50nfTitI63b525t+/pPX+t9ysHydV6z5ScVv3Mh24CTCq1BrVbAXqqS/aKiaGLlzwR8J9eopSAh0GpdeEPNd1JVm3F20oBIYNgd2B1viGngeHfH8eY5MSWOhyP7/eEJyns8HtmNo3k/iw1VPT4+LMlZW/MkZv/eC03bIUvvX0K3t3VkJro6gYTZCvfWSNXsNRtGMbDzjaOt794w55mH+7c8PL4lhOh+0ZXL+cz9/T33796RpwvTfDGKgigViJ4gizaoxickNJscvcY1viNKyXz5q1+6XmxeilBtaxI0GTh842WoYqcZsCSrLrPTMKxDiNJ1HJUoe3ZpoLZKTIH97khrlTmfyflicyRJSFFIMRihoBrtqLg8V20NYiUOdn5IzddeghB1oROEIEi0n8HXdkoj425PDMnVQXaAUFum1OlJsqXTDIaABBgPO0IzmkE7m1yPOD8WtWK4ZLOnDSGQRuuQhCgMuwEJQs7FBkylfvsf4hrX+JZYQJxeHFIXhLO27NJRNjAZESTYfaxx4pvRBfjpMxcNVdN1RRtCo+bCw7t77r96TZ5m3vzqSx5ev2EYR3qhnOrIcLMDxifrXv18QFN/CXEwxnJ3UyXPE3meKaX4Jhnb2DUW+av+8/2O6DqgBV3eT7Uxz/P6PbVOH+jW8qu9fI+tAsIWrCqlbDi71/gU4iMLWUM3azWKwVyVXBuluhxI02XTuLYkjGcqpdgOzHeZ66BYMxTF7WbTuGPcH0nDyPHmjpu7Z6QYORyO7Mfdk0L2cDgs8lsCTM57tde2n53OEEKguoVep0OwQV1E7D7RZbxsICVQakNKRWOjxUIjUMJsgy2qzJPRBuaceXh4x+nxHgnREp8Xso8P98ZfnCfyNFNrsRMFunQqq/VfjasXFQ1XQ4RrfHeUUvjqyy+fDGpseeFP3HBK9bagorUuhayhkHb/GIToiGQU424DThOwY3YYdhx2mdoKl/MjUzgjoqRQiaERgpBw1Ne1ZhuVpk4dGKyFmpoPn0RFIhDMPSvEsBm46usyMY5mPBLiSEhWyOYitLl6q9ZczNBGCkJIwSym9wMJoZVGLjPZzxMm1SVQhYptAmKMxCHZWoxhkRNCBLkE41dd4xofGp1WgNHMmhegtZmta7+t4xYSFJVq7FSpXsz6U2FSWU0rrRaWQlYr2Tnyb796zXyZ+OpXX/Duq9fs9jv2xz27/Y6GcqjVhig3VAcbDKuLNm0vB2trFN8cz26r3juM62faFLDvFbL+iZ6aFSzAsixGBnb7upluPkAO/FqTg277vtCoWn/8NT6F+CjVgubsHMdu1o740vrgG7uiXtR2Tby1XRcJ0UjiBNsxhhgYdsaRTcPIbrdnv9+bFe1+z363X0juAuz2e4ZhJKboxapPYW7oAVtKwPKf0x+e0Bz8WUVsaKVLgyRHiluti8XfagtoJ4JOgC++2LeFrGntGUm+VU+43goxuoYVEk27F7c3Yq+Un2t8QNggYpeiw4/rDXrRbGixo0D9hG+i52ursT9m5WdvaC997kNwioJPQqtNP0eCJ1QxOUiUVgAaKuas1fx4Z0sfCEZzWI5zYX3dJTb/+B7VoD/X07ery8+G6cWa+5dRDAj2ul0fUzYJedvuXdq5neKwGKVcOyXX+LBYUqHnwOXIXI6xvhZ7sWuFqfbFph32sUd2VZD3XmVFe9uWIqSuXrD+XIa7PGfrssk1Pq5uikZlRUlba6Z0ohvVgvfa/v0z9C6QLPQIntx3S4l4SlNoLBrQuhalW0pBj29XSegngmt8CvERiKwwBYEgNB/MaAriiCXBF4ILLJtPVaQ1KLniGYRht0dC5HB7x+3zl9RaiWMiDdbSf/7iFXfPnzMMIy9efs6z5y9JMXF7PHLc7x21tEM2DYnbuzv2hz0xBsZxXEjgnWva5blAGLorWBpJcUCCMfsgoGL8ujQMHG9uSCmx2x843NyaOsG7d7xTNbK7VkrNViTUyuPpkVwyb998zeHLG0IIzNmoBZ16cHq8pxbf1TZL8KWrMIhCCDQx6j7B0LBrXOM7Q6HlsiSQJwUZjor67+uks62fkPxn6NKR7i7nG0WaUr1loEsRrOTiqIwqYa6k2gvqTGkZRJlTc6RVkBGIXsCGgCSTuouer1vo6JNiUlgCLollOdc2nSbf5ZaX0TU1vTjtibH48FfTSmwVFESamyYoIQXSEL2QlbUA9hYrTjOwZ7TzXIiGEqdxWLs517jGB4QdUV6wsvLPbVPZLVcLtc62VqXRxAxAStuhWlERguhSzIW+9VSjCpmUnSGtQU39wMYuI4lAVCH4BrTVtoAq2Ye3tDU7h9T3ENlazWSlVht8ni13ZZfx26KiS5Hbese1D2+xcHcBQkgL+lxcBciogPY6dh87V3QwaTsb07us28HuzmHvQNY1vv/xuw97CWRHfPqlgVtLmve5NkNXe6pUL3ZLaYhUEDEv5hjZHW843j2ntcq43zMc9qSUePnqM56/fMk4jLx69TnPn79kSIm72xtuDgc7wP1tGJ8tmW91CCT/iVMgnrx5YPACNqXBuHbi/tNiyVOx6ei9UxYWHk5tiCrT5ULIkVIzKQ9OmyiUy0wpmfv7dxxef70pZM3V6/HhHefzo0t4ZZdUAZeSpudiDZ2dsR2Wu8Y1fkOouhHARiqutwp7MevFbYy2RrqRQIxeIAaInZagnT/q6NBCSygLF626L7yqIq2SnCdXLoWcJ0Nho5kNhCGQNBFGcxmKnQeLFdLqwy3VxeAtCQbCBonyd7YMey2T0SJIlQVStmfoskWF2gqi9tlCdKQ6BeIQTc6oOlrbO0beKSklGxDtVCMJpmCQUqS1j2pqXeOTiw1iummfWyevI56FVi0vNMG1lwOtZStQDbbxAnA7TIXxWtUc7dC+ETVHvGSjl6YZ2+9bG62YY6WpEFgBXTeFbH8Nk++bF9WBPmPSzwX9/b/PhV0sc2tFwQv1gqqrM/jzm6W7UZ1Kthwq0lVD4hN9ePu86/mgD3irmoGCqj5Bba/x/Y6POwvLmjTUWweiinjrMiwJxRDclTD7dCL6fYu53g4FlgVegy20PM9oq8xTWuSpQm/3iZBqdA1KTEInyILGbhEXEM7nM3mebTdZypKk8eKxL4yub7edrKzLe/cdKJ0yYUVDFTHnpMsFEfFhr8I8zxvdvU0rhZVCIP59icpSxOp1UV7jA0JxfdNt2285rn1L2X9+20XUklx/NnMpsOu1oW52UvqacVRn5bw1Lzr7UvbjtnPmqtIaSPMBywVmtQFmr2NXdsB2YfRWpXQkqSIYd51gG+PaerHdFgpTR6RFe2Eubq6wUgnWL9ApD2xarv5dBvAukyfmpbtzjWt8YCyI/9rW792TfrB3tNZu67xynjhLLrjMlgy/6bEoLIVfSolxGNntd4y78T3VHxuybstAVbUN3WawqteLbYO4vn/paOwTKuH2s66/bpmH71ES2pJfnxTBqk+ub4vUnve3SGzrm+9rzvxk4ncvZAVIbnLQiek0QslezI5WRMZIjYESrNiVVmnzBWpiniemabIdXZ6XXWSrhTIHWm08vLunlEqMkfPjiTdff0UMkcNuxzgOPoQSFs5aHKINh9AHViwl2sa3J3A7wM+Pj7z5+kumy4V3b96Q59kQZXSZls5z5u2bNzbgsVl8D/fveDw9ULLZ85VqMkc5Z+Y8EUrm9ddfGkcW09ssbqLweH9Pmc9+MmtLFdHnn9tCMRCbXJVwLWSv8UGhrXE5nQzJdA6c7TfFaTi+Lvxn8ha/tSqtD2Fr0JuguaDFpqVbLsuGr7gz2DZExCxuUyQIaIhIHMzIunbuncJk2q0xNaIYV09UCdVcvKpjsUtRK0ZRatIoFEtedULqow9jjaT9DCKUNjOXGVWbXi7Fub/NNtjmRKu9Ll5arIZkN6gdifWiQJVKsS5MCDQgNEu40Ye+rnGND491Y6nNKAVWwNUniGa/LmLrQYiUOpNLQaMQpBnP2+k0gWCdPXF+qcC433Hz/I6WC2OIPLu7JQ2J55+/4ObZLUQzKbhcTrRaXa5vtg1nWZ0uq3dXa1uVUGptZB/YnuaJ8/lMqcWK52C68M1NhLYFLKzyfhZ1+Tdbr6bcUL2rJAIG8BrKaoYJ8g2aQUdhO/Bk1Knr2vxU4uMQ2ShPT+TakFqsbamNIZgObI6CBmsbilZaztAqZZ6Yp7MjnrMTvA1tKVKQ1tDTiTlnQghcHk+8GwaCCOMwMHSXn+7EE9xnOXbW0Kbt4otxq/xa8szl8YGSM4/3D9ZOWdBkk74qJfPwsAowBxdvPp1OnM8XarV2jC2eugx6Ady/e8d0vgBsbHCb8Wrdkzr0oRynMpjEtb13UywzflO4LsprfEC0pkzTxKYNsbbksGM6+rDSqkjQmytrm1Od81pzps7Gmytzpszzcr0Lk0ff9IUQGMKeiHH4WgygkarCXAStvhfMThNooKETY82a1mpKJTgg2nVpCd5+VbX7lxlKcK5qITWT9KtaKJS1k1N7IduQpv2p7Kvx85Q0K6DVtPX8mxTn95vKAi24SUvoQNqiYXuNa3xwrETvTdegLV2NpZj1ArfvuIRq5iO1WBci+jrog03BNJGteWHrPu0GDjdHtFb2aaDe3hJj4Pj8ht1xR6Vx0cw8GS1gnidqyQvlAEdCc1lR2IVO0Jq5WTYl52x22KUwDAMyCkHDUliaDJ4++b2jsKvxAZ47e+ezd1Wg1s2msg9ZY/l4mYFZvlYflvYi+BqfRnxEIbtyY3sbvm0oAThpXUqgtkZtioZAmSemywkJgct04XK2QnY6n5gvZ5oqIWVCTEgQSh5cx1Uow8CQEkECc4qksBay4oVsSCuHZoPHWpLaILKCFbLz+Uytmcv5kZwn3xGuWnvAk55I9dZq5xhCtIW2LNSVLtBqpYZi6Gqry2P7EMnT71IXFKi3VTsbAq7Ugmv89tFNQYwj1zdIHen09dC6qL8NiYChMV1+p+ZMzX0AJNM6BadWqJ4sJDgNZsPY62oEnmCD/+ybQfHNmTSQjppW61D092Y2s76CpU9uuzoIlSoFrV6JR1MQWCSKUKjqeutG00FXbmBrZg/aHKmlbb4T50Us7cmGr09dtXdFOlz8h/yTXuOPNvRpjtigsv1kv223L1QcfE2IeAHoYgP4Yfqtr2XzKBICIQVb30Oyrke0Sc6uKNB0pRQ057p3+p/6Oukg0DcoBbWabGanG2xy31bjtdMO32fxfOP72Vxf1Q6++Qm/7baVDfVUQeEan0Z8FLVAQkQkEoN5kVeJFIFGo+aJy/1bNEZKjOSYUBEuj5F3b80R6Hw583g62W7wMjFdLnbwBrO0tddwXVds9xX79eBmAeLar313Gr2o7e3UXhbqupAEv6lVmk+HdvOEWgopKPLsligJOlcWKLkwzxlVS+A3RxOBf9DC5eJk91rRZnqcRZXmyOtiQegnsuWs1QtVcIkS5x0H40Fplwu6xjU+JAQkshaKTidI0llAimhZCrpavHhVm5pGbdCjo7D9Os0ME9S1ZmmN0CzRJBlsaKwPk/jgZ4jBhsYUkkZU2toSTOYCFqa6IqY+MQ2N0EX9RCEqGnoZbj2LLI0SM4gQdol4sCEsiUJIfUOoayejBS9mvVAvVujWqcFUDBGelVBsDUawwU+wIlrElEVaRkP1IiHBFZG9xoeE5xtthVo2CgWdWlBbd1dGay8OK6rBHPOkkWulOP+sbtyRtec5sSNe8dw5upylKmmfoO7s/hFmMWBlykbFabWS58k0zbeocXvq9FccnS21Ms2zPcc8kUum1GprO8Z1+MpNSZxVYCFsuOVb2GZzeb8Y/VZ5reWrXTjG1d9be5+ve43vdXwkIhtWToxP+/fWeC2Z+XKiilBCpKSIiiz8t6bK+XLm4fFk3NLJhMlV1XaM7x2o39B/fVLchfXAXjRdWZDiFYllLSD9mcUFqdURZFXl5rjHhlY64GLPnVujuHd9jIH9bodq43J+8B1sXS6Gwiq1c4A2O84+YCLyZG3bJzXgyPjEnjyvrIJrfGjYwBYrhQCIYq5c76OxvXhVGq0WapeQmzNlsuO8LYWsrookWHteOlUhDU4n2Oi5YpvQIJZkUgookRiFwTektEabi6O7FXJdOOouPGdIqFWVlrDEErdSqH3J50io1sGJQyKN1sGxza5vaDUsybkVpeVmQy25obkiTYkVtLrQV0dct18sJlyvvrlWCVfKzzU+MDrK2RxA2Q4KN1p1BLaxoJ+dm9q0IRK9oFSa2PzlBpux9aF9s2dSdTIkG7hWJagVtcYzdeUBl40seXZr9WIa1AvGIvZ632JdXUpZpCNztiK2c1O7Ax/yLfaxvR/63rLp/Nn+4t+Kqi4cqPWyNkvtsd1yfsvNvcb3P37nQtYSVVwR094uFGd49mlDoLVKKeq6qIZUKmryQPj0sDtu9cWpmyIPb/H1Qnb7HtguioUK0Jbbl1bHZmEstxmjCOg6d5akay3M8+TcwmSe8s6ZjTG6fmVfNHXxhX6ivrDEty+nXnf3wmPdVG92nUG8KJc/uhbmP/gH/+DX/ts//If/8A9+H378Y/jlL7/9Tj/6EfziF7/2Of7YIohsqAQ8QTesldkcbSlWyOpayOJdBOssOFfO26BL6x1wnSrfcC7Q0FIs93JzxYvW4csu6bUoInR4qXNVNwW3OrVgWQLSyULWwVCA0iA3NApgOpuGoIq3UWVDG1DIDRyRpfb30b+b9Z1Db3KyfiLn7wkslrt/TPEv07r8TUsSvl/L0riextlehrl0a0jSq0dZH6BdMcNuaz5k2MLGhMDuzJM8I+/li96UhKVl3zuErdMHdC0CO1i0KCS4msBWZagPfBU/T9jLbhKZv62uNrBk602Os4/56/Lj0zxo9MGnSOz63N3xa1UVap1/8UcSv4+18vf//qebKz+CWiDEYTQh85iWQlbFmoJZzdKuokwKp9Y8bzRyMz255gWnhMgwRsbRWh+9PQCr5Ai4XeuWh/vr3hqw0An61d6qVNZE7i1VNgtWVblcTnz99ZcMw8jxcMPd3XNiNMmSm+MRFKbpzOVyotRMno1bW2tBW7VTwFJfb//f398G7emKD9LF3J1HOERCcCpFtNv/UPEhB/t3Het//+//Lbyxj4nfdBbwf/t9JOff9O/b5/nbChEYgrXtae5j3rUdXRqudc5rK9SSnbO3UmJarcuwh26ud0TTlpEnL1cDseLUfpcnzpBWlKYgptMMhKYGfZaKToWWDQWSqstjW+fKig2KLkhvpxx5ual2Z3uPApoimqo/zlBTexu6vvdqSgw0RS8V5manhAKt2nM3WTfI/TkaSmlOmQ3BeLV/IGrB721D9y9R/KYl2f/9D7rZ/VsMVaOvlZK5XC421LTJPwCdV2650Q09SmPODTSwSzN5zGgLlFioQ0PDItDI6sznG7nQN3KWd1EDXYorJNSupOOt+FLMplrEVBDwoalSDOGstZLn4nSCmfPlQnFFE9uobjVe7TVr7+A4KNNnWvqQ1vt81wVkCkIUA41SSmYNHcKTwrVr165oLBvVh28DlX7/8Yc69j5krfze4g+UKz/0Ph8SH1XIimvRhRB98QSTwgJTHKjrdONUKkUbpTVyLSisXJogpBhJfnDnUsCnF+VJIfsUkX0avWhtT3diCj7x4c+jqy6mNuBpIQswzxMPD/fEaBzZw+HGDRYSKY32Hn0wrF9M5Ll6cf5+4fptv8jK4+0b1m5RFmRxK+qI979sHcwPONav8S8gBIjLweLuepvitZZCmSe0mRRcLbOjs9WqONhYV/rlPXSj78E6woPi2q+uQat2wxadtbUbLOn2NVobWuyCKmGhxVpBu6XbIGrrwjgSTw0StKHOadKodlazSTPU+fK6FOFqNIbqhi2zo7P+tpra3rIt3Rp/sP/ohqHLd3LlyF7jA0IVb8dn5tlBj+UYYzHZWJKBsrTzczbL6Vys8BSpGwmr4DScvkA7auldUeP20M1tG2oFpm610Fejn47IEny1Nl3uV2u1wrdZ0ZvnQqnFN5gbapF94mVYzD5/sLGX91DVXxf9fr3o7Zzb3ukFFuOFLarbkeKuB3+NTyM+Sn5LxZZQg2XAo/ntGqygNVRRGaJpOAZtBOevxBQXRYIYgvHmFHSeacwuZt6W4jA4J3cN65mI89dWtHWBdTYt1bb8rtJ3cWHh5qijP6ihLE2tzVlKMa3b1ryQNf7rZbowublBcccSA1q3bZ1NS2X7lrfX3yexe+Ha3YqkI0Py/oOvcY1vhiobZYHi9IBKyzbU2IpxXs2QxPmpzRKqrNMjHmsb/8lrdNMEtbW+tEGbbwZbr0LXlqfx1Q01oa2T2/2x4sXwet0b/EIvf93IwJ6tedG8oLJebC50oYVTqwu1QFhfX7rXfFvq7g3a3NuqLJ+j0wo22+GVqHiNa3xHqDZyyV6MlkV+sbt/RKzoFEdlV7ON6goHuN56Bg3Usa5Fp2tAf5NX2ovG4APLlt9q6xrLZtAzT2YyVHKm1YINUysiwWlz1bWX17Z9VyjoygKLclBfd0vHsxey6/taaHXvcWj783Rzg37b+3SCTjV8P7YUpnXtXuNTiI8a9qrSW3eGkzbpt1khGwcb8NqFCCHRnLem3iZPw8DgpgYddGmt8fj4yMPjI6vvtLUK+84MWBULAHGISLQZquSClVtHo540u3yOLSDjKaGdzO4T2RIozQY7Hk4nalvbJiFEL2RPnM8P7js9bVoiYYMay1oUbCkRW/pSWIvY4Oh0DJHgVAbxAvaaLq/xIaG1Mj08orVSluJ1LWS1VlrJ3u1oS4EXBJ/UX1uTAEqg8c02XWMtQJs/r6iipaDezjRYVZdmAwS0VXI2NJjaoDRC83zeVkRXmo2sqAghwroh7K1FQVxnUosu2pkhNiT0+1Xc18HUGjbvP/jHaVW3TiSb4qCTIlyeTHBuIcu30aTxhEVxjWv8mmit8Xg6GbXgfKbUSghittAitK5/7uBGTGabjPSOn3KJFx7lkZQKQ9hx2N1AUEJoSOhSXR3Q8aLQjQQIq0b7PM+cH0+UnHl4947L6Wwby+LUOKcL2uPXpyy1kkvxQrgusppDDAzDaColMTk/tT0prJ92Uq3YBEE39AMRISUrSd7nxK4atG0BnbYA0ILYBiGFtDzfNT6N+J0LWcdwlpN9b3A0EZoENEYkjd6SH2AYaT4cRrKDbhgHhnFkURZw3/asQsq246QUNNjuNQ7DkxbDsntFlws1O1Ky2uACNrDiiGyr/fZI07KS7UNYdr/d2WieM7WdCL1gd77cPJ+Zp7MXwq5bueT/tT30DSR1e/X9wa6+eDsaGzYz0ddK9hofEKpKnWdqKeRpWlHYPEOri80saj7svUbsTl/BOwALF21Twm5/LkCK9p78Zp3VutAB+uEvLvdq9FgrrruEV3/C4N2Rjs5azwTQvmmVJ2+kdbc+N1MAkGqFbH9rXaPISQJeVK/PZV0iR1uXYa8+DLO2ac0YYU3qTZXKH9+w1zX+xURTZc6zOz9mF/YXlLhoLLe+BqTLU9n66ZzPkmemMNEKlEM2vjusnY7elVzCc0prtiF02l8tlXmeyfPM5XzhfLZCFgdyRAIpNQNtwJWAxHTR22ra0E0OEKMJJu+uqqobl7CcKN5HZNcCdS1StwYHvSPZH9uL2P7aYK+50A+Mt+Bd4M6jv8anEh+lI6vBijsVK2A1BIiDKxAMDLsdMURaGkm7PXR9WLcTitGoBQgurlwhgDgiqVKN7xcMmYwpmXMXsiCZ0KmljhDFiGhdd2/do72t7c8u9m6WmWVpc0ooLndib1WVRbVAlkLWCtTYEqFERMXNV+x15MkCEhYR2CcFrS63LQNsYsMwVkz0Nk1YnI2uWrLX+JBQVebLZEknZ9+0VacQrC5XeOKQTofxJLgybNryfNsktEVmm+oir7XcTy359t+bdAWCsKyd9WJJT3xtNqcUrM+zJkMxOuBGPHOlN9B/KL6TlOU9vF9p6nK3TolYb+wU2u19N4/YtEzXO/whBkqu8ccfquaAlZ0nW2olRjFtZTEqi7hDXm2rk1YHY7qGa5YZjULOmZrLcgyLZ/L+GMt11UGaRsk22JnnmWmamabpyXXt3ZTWtZ4Hm31xKS2RsAyGVS9iRTrvNSxUHtTOCysA8831oU/u8y0yW3THL31yfYvIbruYYQP+bNh81/iE4qOGvVpKnjK8yEsj7PaEmBh2e+LtHZoGwm5PONwgIaGBZSK4NSOeqyp5mmk6oVRDcoedtT17e8HbLSH6KMvCHZXFKdeQnGoUAzVe0sJ721AMzC/eJkb71LZJhPkEt4NMCsSQFlWGZaU6VKTO320tLzZ6snZll+LXfwP6zlSXgiFIWEjycRme84lN5z+h32ztXuMa3xa1VO7fvFmRUV256YCrC/hwVRCimNZrN0gA06lUWJoK6tQd27BiRWdni6qbfbRq99tw1NWnpEWCmxSI8VybXUxtQG3YC0NVl+HOntzsjeG7PrqbVpfwEn/b3Z7T3Lu2SgVtubpuIHsK9Y3mco5YgeWF39fRo/5dKKwc2uuavMaHhVELzpScOV9O1FKIMZJGO+enOZLz7O38CmIAS86FnAu1NC75Qg33xDCxHw/cHm4Z0kAbAtoCiNLKqudaq8notVaZzmfydGGeZ968ecvDu7fkOXP/9g2X85nWqnHnayXEyDCMlo9iZNztCDFidrVlURuKKRHBwSjxteEKIqw1pSDrcnbEtvl9uvTX+9ElypYNsUttmTb1KoWZBqMmJs+dTzbV1/hk4qOoBSqdeeYmBDES0oikhOz2xMMNMgzE/ZHheIekaANhnhRKtd2p2dw1cBkeOiKruiwmkY397Bb2FBd+95uCtgWdZXtAt/UAj7kPu1RCTItFHyHaT3riM25sVy9YM5kS60DMA9IqSEOJdETW3lanITydau6tlI42dUS2T2f2orZTGK5xjd8mVBvz5WIVmeudBrA60O6xoKbSvEnyHuK6BSr7E6ijRn0f15HJ5XrfAG5gzcVTXRpo3DzpmtwW6+iOttiH2Hwgx0KFlR/gTyLv3VW8UO7tC8OZN4VqR2rXt/EeQty/Q12FGhQj1Gpb0N71S/jt/z7X+DRDVZl9OLirFsRWURIhBs87zWWqFIldF3VVFChkmjO9WXgAAQAASURBVM7EoOR59sEvH6j2lnytzWdAGrXUZcZkng15naeJy+XC+Xwh55nL5cLlcnFnLzNGCCEwjtUKVXfkjJ6Pi9usi+errq6zdCw2G7zl3GNfwPpdwMJpb/rNRdTnVYorF7UNChvdTU9gHYpehs22YNE1PqX4KNWC3x7C3z7gD3OwbScj35+S/DWP+M5bf/0z/Lov5A//ua/xCccf84l8y6X74Adde4nXuMbvO64F4TX+WEJ+14NVRL4A/tnv9+1c4wPiz1X1B7+vJ7v+Hf+Fx/Xv+f2I69/x+xXXv+cff1z/ht+v+LV/z9+5kL3GNa5xjWtc4xrXuMY1/kXGlYh5jWtc4xrXuMY1rnGNP8q4FrLXuMY1rnGNa1zjGtf4o4y/nUJW5D9C5L9F5P/+t/L833y9/xCRf2zjnvL55nZB5D/xf/svEfm3/Pa/h8h/4bf9O35bQuQ/Q+T4G17nP0bk3/1b/Bz/D0T+7t/a81/jkwwR/iMR/lsR/iDrUYT/UIR/7EZ+n29uFxH+E/+3/1KEf8tv/3si/Bd+27/jtyUR/jMRfu16FOE/FuHf9ev/ExH+GxH+PyIc/pY+1/9ZhP/p38ZzX+MTjmu+fP9xf4HI/+o77jMi8p8j8nED69f4XsTfFiL7HwD/Hqr//pNb//YOuv838D/jm0Ts/znwd/3yvwb+U7/9fwP874D/BfB/8Nv+t8D/DdXTt76CyGfAv43qf/5R7/Q3fwf/KfB//Kjnv8Y1vhn/AfDvqfJkPYp8pGrJr4/f23pU5VvXowifAf+2Kn09/vvA/0mVf1OV8+Z+v8/P+H8B/sHv8fmucQ245sv34y+A31zIqs7A/wv4X/4Oz3+N71n8/gtZkf8r8K8A/09E/vf8/9n711jbti2vD/u1/hhjzLnW2nufc++tF5WqwhgcOwFLUEQmgbgsJYojmaAIFIGsGJnkA4oUpBJIQXESlxQixY74EtsfIkiMgeA4D0IwiSwRCEoZUzwLqArkIRdVRVXde+495+y911pzzjFGf7R8aH2MOdd+nLPP2efWvXX3+B/Ns+aea87xmGv00Vtv7d/+f5EfQ+RPIfJXgT/VVlt/ua3u/hIiP9A+92sQ+QlEfgqRP4LIfXv9e9vK6+8i8tOI/LaX9qn6k6j+7CuO5ncAf7IJXf4E8ASR7wUSsG+PhMgT4LcDf/ITzux3Av/RxXn+ZkT+E0T+HiJ/A5Gbdm4/jsjfaY//Ynvvj7TX/zzwDxC5QuT/2j7704gsg/HHgf/Ktsrc8EVBhHU8ivCjIvyYCH9KhL8K/CkRfkiEv9yyoX9JhB9on/s1IvyECD8lwh8R4b69/r0i/D9b5vOnRXhpPKryk6r87CsO53cAf7LJOf8E8ESEl8ajCE/4DONRhP8u8N8C/qci/G9F+BERflyEPw/8AxEGEf7ddi4/KcK/0D63F+F/L8I/EOH/LMJfF+GHmzT1n2jn91Mi/Gg7r58DviTC93y2v8KGDa/BNl++PF/C/xz4be0cfhSR/1z7zN9t38NStfxz8HBxvuEdxaUTxhf2gJ9V+HJ7/mMKf1th1/79Hyr83vb89yn8ufb8Lyj8nvb89yvct+d/UOFfa8+9ws0b7fe8zd968e+/pPDDCj+g8FcU/prCb1D4owo/8inn9O8p/Pb2vFP4GYXf3P79SCEo7BWG9tqvVfhb7fmPKBwUfnX79+9U+GMX23588fwvKvymb8rfZXu8kw/QnwX9cnv+Y6B/G3TX/v0fgv7e9vz3gf659vwvgP6e9vz3g963538Q9F9rzz3oa8fj5X4vtvlbL/79l0B/GPQHQP8K6F8D/Q2gfxT0Rz7lnP490N9+8e8/Afq72vMfAT2A/uqLY/7ftOf/WdCfBx1A/xDo/6q9/p8Hze14fhPoX7zY9pOL538M9Hd+q/+m2+M76LHNl6+aL//Cxbb+LYV/+WJbu4vz+8a3/O+3Pb7lj1+uZq8/j+pS7vstwJ9pz/8U8FsvXv8/tOd/5uKzfxP4VxH5MeDXo3r31kej+vOo/giqvwU4At8P/MO2Ev4PEPl1r/jU9wLfaM//KeCrqP7Ntr1bVDMQgT+GyE+1c/lnLj7/N1D9R+35TwH/VUT+DUR+G6rPL973deD73vocN2x4Pf68nsvvn2s8ivBjwK9X5a3Hoyo/r8qPqPJgPLbM8X8gwqeNx1fhb6iyjLffCvzptq//N1ZS/XXt9f9de/2ngb/f3v8zwD8hwr8lwr8I3F5sdxufG77Z2ObLh/hrwP8Qkf8B8IPrd6NagBmRm7c+xw2/ovHLFcgePvcnjWPzXwZ+EfgTiPwrn+HTvwj8Zy7+/f3ttUv8z4D/EfAHgD+OcVT/9Vds6wQMn7K/HwU+AP5Z4IeB7uJ35+9A9f8L/EYsoP0jiPxPLt43tH1t2PDNwucej2qc1HU8ivDtOh7f5hyfYmP4rwC/vx3Hgm18bvhmY5svL6H6Z4D/Rtvm/w2Ry4bLHhg/ZT8bvsPxrZDf+k+A392e/8sYLxTgJzBeDRe/B5EfBD5A9Y9hA+c3foZ9/XngX8G6Mf854DmqX73Y9j8P/BKq/z+M/1Pb41WdmP8Q+Cfb8/8P8L2I/Oa2nZvGa32MrTwr8N8G/CuPSuT7gCOqfxr4X7xwTr8O+OnPcI4bNrwNPtN4FOEHgQ9U+dzjsakX/HPAc1XW8SjCPw/8kiqfdTx+Gn6cxqVr2d0fwMbwX8W4tYjwzwC/vj3/MuBU+T9hk/Y2Pjd8q/Auzpd3wDnLKvJPAD+D6v8S+L8Av6G9/iXgQ1TTZzjHDd+J+KZwFl7m/Pyhi9/9oMJfVvj7jYPzA+31X6vw19vr/6bCL7bXf6/CTyv8pMKPrzzTh/v7Awq/oJAVfknhj7fXReHfUfhPFX5K4YcvPiONj/p++/c/rfB32v7/S6/Yx29T+NMX//7NCj+h8Pfaz+t2Dn+/vfZvXPCWXuT8/Nfa+/6uwt9cjwu+W+FvfKv5JtvjO+vxCo7sH7r43Q+C/mXQv984qz/QXv+1oH+9vf5vgv5ie/33gv406E+C/vjCQ31hf38A9Bca5/SXQP94e11A/x3Q/xT0p0B/+OIzAvoXQd9v//6nQf9O2/9L4xH0t4H+6Yt//4kXOLJ/4eJ3A+i/2/b5k6D/Qnv9CvT/CPoPQP8s6N9t5/3Ptn3/3fb4r7f3R9B/CBq+1X/T7fEd9Njmyxfny9jO+e8p/KjCH1b4f7X58j+6OIbfpfBHv+V/v+3xLX98+1jUmh7dyY5Kfjfwe1D9Hd/qw3oAkf8Y+JdQffZN2v6PAreo/q+/KdvfsOEN0fRbT6qoCL8b+D2qfFuNRxH+Y+BfUuXZ5/y8B6Iqowi/Bvi/A/+UKvNr3v/fBH6jKv/jz3vMGzZ8IdjmSxD5s8Afxqh6G95hfDvJPP0m4N9GRIBnwO/71h7OK/EHsbLks2/S9p9hhP4NG77V+E3Avy3Cd/J43AP/DxEiIMB/73VBbEMA/ujn3NeGDV8k3u35UqQD/twWxG4Avo0yshs2bNiwYcOGDRs2fAZ8K5q9NmzYsGHDhg0bNmx4a2yB7IYNGzZs2LBhw4ZfkdgC2Q0bNmzYsGHDhg2/IrEFshs2bNiwYcOGDRt+ReJzqxa89957+qt+1fcBivWLWdOYoqDaftorIg7nPCDL2+y9TQPMtlGptZ630TZp72nP6/J+25SIoChV9bxZZ+3HSns/D5vZnAjOe0SkPXd2fCJ451sTaNvB+SAujsuei7iL9168tZ23otRSKFpBlap1PXb7LAiCOFmPt5RC1bp+fu3Dq+dd/Ow/+vkPVfUrn/4XejN8+ctf1h/6oR/6oja34TPib//tv/2F/j2H/bVePfnS5/68vPYXcvH7VzeIyif86/UbXvBJTafy0rMXN6ev/dcrtvvgJXn9rx68rq9+Q/v4h7/0j7dx+R2EL3pc2t/zB9d/P7iMVF8zPuT8zs/Tk/3i/PSKrb8tLg/rbbb34veRUyLnAii1VqoqThyxi8QYbX+fcn4/+7M/x4cffvhFnCawjclvNT5pTH7uQPb7v//7+LN/9t+nVgvYVJVKpZKwi69QSkZVib6nD3uc+KZ77ixwyzMpzWitHE/3HI/3qFabNETbe6Bm0AplKpTZojrxFgRWUZJUslTECS56nHfUWsk5U2qxoDV4nAgxBvZXO2IMxBjZD3tCCHQxcjXs8d4jCE4cqKBaqDVZYFoLtWYAvAv4EFnC0MvguVZ77/3hnsPpQKmFlCdSnhEEHzzeO5xz9F0khEDKiWd3dxxPJg1YSkW1IlVwxSHVxuPv+p3/6s993r/Zq/BDP/RD/K2/9be+yE1u+AwQkS/073n15Ev8i/+dP9z+tSycXn0vl6at1f7x6vddvMfJ2XrHtWv+4m0P9iXIg5lNZHnP60POV87Wsv4Ph+DVykg2Ru13FShtjazY/UNRHIqjnhfUFwot53MSRFxbGHvEWZFKRVCgqpLapKoKpdrC+cFJA3/8X//vb+PyOwhf9Lj8oR/6Qf7G3/yJB0mNUgrzNFNrRbw7J1icwwe/zi1Uu97SNDGPkyVuakuMiNAPA93QN0OuiwXnK8azPPzfiyf9mmBUlyH0ZvgMAfRLAX218fbBL/wSH379G+SUOBwPnMaR3dWeX/1r/0m+5/u+t31HAb+M18ux3fb/X/jNv+UND/jNsI3Jby0+aUx+fh1ZARcdThVVd86sSrBppFQLxqri2tSjqjjn8S4gCEkrNWeKVLRU8pyotdhs6W1QhhBxXbSgMoKmdvG3QFec4HqPix5xFiQ6b/tb4Lwj9MF+Bk/XRwsmnSOGgHMWVAZ/zsguP7U6arVjryVTsjQNamlTKOsNxCEgHpyjqjL0ClhQXWqiasaOvoLYNroWyHrn6cNM8fX8l1Hw4okS8fJqt9sNG16EBWPK5bT0adkL+YRAFmw0iVwGqheB7Isfa0HgOkvJUtF44e0v7e6C6dQ+IxevOsCLtPOSiwn5IqOlbs2eqiqKB7GgVqS27VWcKCIQg+C9LVy7rieEDgRqO8OcC3eHIye9CGBr+wbO0fmGDZ8IWxS16lwLQsfTyPNnz5jnROwiw27Ae0/sOga3Q5xYVbFatfL2+S1Pv/GhJWhKoeSM854vf9dX+NJ3fRfee3CCOEsUqS7VRFkrgOeR+HBMLr99YchevPthFPv6q14eBJUPxoeex/QrN1RtsVlz5nR/z/MPP2KeZ54/f8794cD1oxu++3u/l1IrKoLUer5nbTKi7zTeyhDBeUFV1oveLvY2iKQC5VwdafGZCPgWaIq7GGBqmd1aq73JKYIjeEcIoWVHbfJSPQ8scULXd/gurCtZowvYylawbGwc4vq72Aecaytf51oAynn0Xgw+dSDFAnXB2+RYl1r/w9uATfYOnEeqEnwk+II6xatDl89rRrFMcfAB7z21Kt55vLOA1WHn4MXR+57gvp28KzZ8O0MuMi4vZk1fdb9/bZB7+brIRdb1YsvycAw8/Pz56RLYykN20cuflOWGIWuw/PBxGXBfBrMXO1huEiKtams3IMEhokYpQnEOvBOiF7xz9J2niwFE1kB2dsJpdHgntv5sC1Bt35uyBbMb3gxLYLlQ6krJjOPEPE3UWvHeo0HxPjwYqMucM88zh8M9aU6UnMk5473n5tEjaq023+nDQb6Eq6IXC76L+dOSQg/efCY0XLIa5OGo/bS4UV8TYMolDfCCOSHtvaqg1ZJa82lkmiZOhyPHu3uC9+SUqHqurti8/IrvecM7hc8dHYkI3hlX5XzdnDmpKhUnbUCqIC2L4X0geDPSqbWQU2oDxVFVqBW0tGyrVBwFoQCCFkctbt2/tKxTKYqUZaUrSFGcc8TgjIJg4SyC0QVqVtQBTvEWI19SYUEqKm3Vqlbit0FWV+qDUQgsoPXOssFLMOzsVC37q127cTmqeqCSKyzxulIbx7cQvKOL9t04seP14gguriWUDRs+L157f1d9dTB2+bpqC9ouX36RGLAEoMtEeJ4uL+e1y0SqbVLX/O6LZUdZny2LZG3bkAeTsl5mffUicFbWYNPCU6NIBAHnhD56ht4qIsPQ0fc9VWFOlVSq1ZJUbcBq2/Drvq8NGz4FlznJWippnpnHEQFyjFCVGMP6rpwz0zRScuZ4f8f98+ekVrmspRJCYBpPpHlGq1qlMfhWPmkBpVr/BW2+Os9VyyJMEDmP3vO4Ox/xi2exjO/zr5eI1NnjNWf/IOOry5OLsS6sldXYBbRWovcEJ3gRtBZKTqCK9+5cqdyC13cabxHIOrq4e7C6gyWUXZZyF8NWbWh4F/Atu6goKWc0CSKeWoVcodRKyS3zUQtaMyCUFKjFwlIfLJviFHK2QBAgWdqE4D1+bwPAqcNpwBPQquTZMjTqzWzdeyvh5GIBKq6gLoO0jFAbbFULSuM31UJOtk/12rLLgg+KOMv6dNFoA7AE2QXVypyP5DpZgxuF0jjGMQRk8C2ADWtD2bmcumHDNwvKwxnsMvI8p2z0/O7XbuccTJ6bM1+6enWZS2W5NdjzJUhtQae2xeSy3bpsTRbSgaBii0xEUZXznKbOqjjLGbWg2QtEb+P+ehe43vd4H9jv9/T9nlIqt4cTnApVKk4r1HJRkbXgQLchueEz45yGLKUwHo4cDwdqLgTnKDFaINtWfHOauLt7Tppnnn74IR997WukaYJWmQgx8vjJE8bDkdh1xL5DiTYfeY94h9ZKSYkyza1fxBn9zlm1UsRdBLFtvH5aXNgyyufFXTs773FhqZ9c3ifOmVPLML9wJ5HGV3cWyMYusBsGnMAQA5PzBIGaM/M4EmIkhvCWNeUN3yn4/IEsgnfhQRr/QRC7DIgH5QNZObIAznsrhzi3lgG12mRVWhanuEp1FdRRqlJKK++p5VirglOllgpIy5ha4Nfmv5aZEWiT3EK2d3IehNoywbW28g8VpeLcmaO3TMyLUsKqsiDnjBAXpQ7njB2sgGvnUGuhqKfijMNfcztmGk9X1u/WyUIz+ITy74YNL0BexyH4JKhdx+cF00VUexHM2tMXGXOv2eSD8uhycK96o1xQDuyGIcu+2gC1o7nY70Vqx5pe9MH+UFk/9YBegWVkTaVEiN7RRY/3Rivou0DOxRbAcjkdX57Aw5zVhg1vghfHjFYLBkteHgUnjlrqSkOopVjWdpqYpxPT8cg8TTZvigWpORnVYKHWafUoDufP+63VOLUiAs63JI1D1a052NfXRZZx//D1lWa3VCmwKseDTejFNtb/P1Q1Wse7uLXSYufi8NkbDVBsHtTWRO7cC1xcaDSDi1zylqR9Z/AWzV6C9xG4kIlaJ5qX8zbLBeakBa5L41eIlkUhknIgZ0hayNpWddWh1YLSec6k2bilwxDpujZSp5YZbQ0ZqlaeSRn6vsP5hI9GjLeDsAC06wIl7wjRU0slJ2tOk1BwXUZcBamIazIgRSm5Pgx6EdRdjpglOwTn/mpaJre2ALm07Gyh1EQpCXB43zUJMMvILtSMupQ1N2x4Y1ysID/lXecnnxycKawB50u8tE/4zOu2/LrQcAlol6lVL56/tAW1sVXV1FLU6qltC34dgV4geJPc6zvHEB0hCH3f08WIc8Zfn+eZlAvjOHE8nZiTcRHrwt/bAtgNbwVlYcXYT/uv1troAZXpdOR03+G95/ajj/nwa19jHkeef/gR4/09OSVCjIQYqVk4He559vFHlqUcemJvn93dXDNcXVFy5v72lsOzW5wThv3O+kpCYLje48JSIb3ABd1uSfysDaEipJQYjydyybDQ7BSG/Z6rEFnC48vF5HLOOTVKYa3WtFasl6Yfero+mppDmhmnkXkcmU5HpuM9IsrdRx8x7Hv6YUfwYZXiOn+7ehFwb3Pmu4LPHcg6cXSxZ+XKNDzIz+p5tfVA2kNsNRViJHY9Ih6lZ04d8yyk6kjFeLElelJw1KocDonTKeG85/pa2O8jqkqaTR6naiWXmaqFGAI315m+7y/vGjgnhGgljN2uZ56VvouUouRU0KqEodJfZZxXVBLIBFRKhmLzJaIewWgA1V1MvaqmTqDSGkNaTnalJWRUM1VTWyWbLJd3ga7rjF6Ax7keJ6HxqJIN9g0bPgXLxGgZyk957+fI8i8Bp77w2msOZv3lq99zLkGyBKCvCKj14j16+ZlWftWaqDqhVETd+h04idbYJRBa9tU7YT90XO87gndc7Tt2QwdAypWURqY5c384cHt3JJfKnJJRgy5DbnnTpcKGDbBmIV9I7izBoZbKPI5kJ6AFpxXnhG/84i/yCz/zM4zHE6fbW47PnlNrZdjv8Ps9VZW7p0+hSUzGviN0HbHr+K7v+x66PpKnmY+/8XW+8YtfxYfAk/fe4+r6iq7vTSFh2MFag7ygFYglh0rOJhHWmqNFhHmaeP78mcmHlUrJGRSevP8++6sbfPSrOgM0/fY275eUOR2OlFKYpolpmhAnPHr8CHHXzDkzjicOh3vm04nD7TOOz5+SpxMf/uJATieubh4xDDuG3X4Nrlm/T90qmO8Y3ophIuLWeWfl1sBaMtAHHc3woG1aaMoCHnE2OZTy8IEIuaWAarUGjGkuOKf0vZKyDbRpVlJaAtlMqYkclBAzphSwrCqNFxd7KyuKOKYhA1bOybMNPHUV31e8GPdWXQYKtQqlgOoSwvqXyUQtc3NuhrnMTrc1qiq020ZdGsi0mljDwrV1zrRs4fXSSBs2vAg5c01fDLM+SU/2s+5j3frSPfwp731VTvZFEsNyyK9+/SKYveQrNarPwj+3PFCT35NFe8CoCwudwHtPDAHvXVMNcZZt1UzOliFKrSs819oqLxfnePF9bcNywxvjFSvLJb2jWiklW59ISszTiBNhPFrH/ng8Mh9P5Hm23oocmyxXIc8T4+mI855cMiEnSsmkeYb2nnmaOB2P1hy229F1nVUh6sJFv0jBvjBWFxqdQ1c1glIKKSXmeW7VTOPL5mw/L5eoF2u/87k2CbGUEvM0Ic5Z5aMaTbCUQs6JlBM5J3JKOCdMxyPj/T3eB1MwaBJcctEMLfIy7WDDdza+EKr0qxlz8uDZItljdBoL9nIuTNPIPCUOhwO3t8+Z5sScYM529cfYEUK0jOxx4nQacc5TamScPFVry8hmG9RlopRECIVpdMRoE1yt9tN5IXaC9zAMA8dDoeu6pnubUFX6HeyOEAKErhCHjDhFC2gBVAgCwQWcGP3BEZqJgmniKsvK0GgMq0y7sppFmMGCmSyICLVlap0oSDFNQFV8cBslb8Nng5HO+LSc4efKyuqSPH1VJeaz4MxTeCCQwDlXW1umthGTWDO3gLHpKznP5PmIajHTBJHWjLondh7nPLs+cjV0eO+42vfshh7npAWxpnwyzpnjOJNSYZwLc1FKhaJnXdkH1IJtTG74PLiQjqq1oKVyPE2MpxOlFvoY2A89IvD8ow/JpwM6T2ga0WwGCtPJUbTivLeAcJpx3hO6Dt9F+t3AkyePmb/0hDzP1DSj2RSCtGR71LJq1EILWC8SUgIm+XV7yzzPeOdM1905TqcTeUqmA58LZbYAtqSMFtuu1gqltEWiUMW1ILaYFm7O5DmRZ5v/ptPIKQTmceTu2TOef/QRaRqp44lIxZdMPtxzdCClcPfxh3S7AR8CXb/Dx4iTJsG5qfy8U/im9vwt088D3UlYuxbTnDgeT0yjlSk++uhDxjExJsc0O8AR4o4QdlStnE5HxvGEc577o6fvZV0FlmIBYc4nSplxztPFgvejGRKUiVoLzrMGsl3Xc319Mp6NZlRn0Mqw91w/DoTouLqBm/eEEBrPR+28OucQr9Y5Ej1eOpwIpSRqzWeuTjttsQ6T9eZlgXemlESpGQRKnSlqwTASEC/NTcnh6maIsOENIVhz46fQCz5rEPtgU6qfs2v/4j6w/GvltJ5zsPLg3ZeWCAvrtdg40UqeJ6bjLbUmvBgf1jtH7yD6nhgcN/uOxzfX+OC5Gnp2Q2cForaYzKVwOCWe351IuXAYE6dkVZyiZ13ZZYqXRa92C2Y3fBashQQrM2o2U6D757d87WtfYxxHohP6YPd+nUd0OqKlUKeROtt8dsoz9WgGCKfbe4bYW+m/aaoPV3ve/9ITpu9+nzwnyjzBPNu4zYmaEyUHai7UbOFrRqmKJWDa4U7jxNOPP+Z0uCd4o79550kpM4/TasyQZpPEKrMFt9roCLUYd7aKUF1dqYDzNNvnxpl0mhARTs5BNo7w0w++zjd+6RfQnOlSoqcgRZmfPyUf7sinAx+9/x7VO7q+5+bJ+/T7KzMWkr7R/L5Vf+QNv9z4ZRKvuLjbtwq7KRRYV6aVETJzmplTYp58C2Q9pRZKLdSqTHNlTsYdclMBSuPw2KrSbGmtLOEEail47x4Ejs5DzrL+FCZirKAZmIFKxeNCJXYOFxzD3rI21jlpk1htQYINFmk0i9bYtfIC7Q1rqdfS0Wu3p15q0774EMs2iTjr0GRbYW54M5hQvzVCvQmd4PL560tySxaJh7X/N8rFvuYYzlt9QIc4v36hSftaioKNoVpyq2zY2rKqAy14UePIerOnDt4ToicEWxgWrdRGXSqlknIh5UKuNqmv6kKLasNl0XSl9G4z5oZPgbaxpbRsJTZnqc1bKWVOxxOn04nooLhmCV0ToXm0i4msY2oGSlJrfE5VcEVN8aAWSrGsaZrnVdGg5rxmSbUUs5Vvc28pBcUystaZcq6OLDSH6XSihAC14p0nl0JdgtVcqLnNxSWv26ylUkuTrXMCzQH0oVKDZXRFbF9p8k2hwR7kTESbNXZFsyV+8tQzjyem0xFVZUiJkDPioVbFuYd85A3f2fjlU2FbJwJttns2I3rnG09NCEHx3gLR49Hc0rs+0FVrKqs1IDKAOJRA1csyo/FkfABxESceH4wHJLVaIFgLgppObQWtpivrfWEpHoIyToVxqqYzeXSk5IidcLWP3Dzq2jGbYLN3NikuPFYRb9xZmqc2FwFrURuEybh4tRRTYsgJ5yuiI6UoPkTENR1ZPN5FxMVXfq0bNrwWr5Hhel0Qu/z7zflln/V9S1gq66tn3twFt+ACq27sCw/R2jwEwTnogqDVXhfNOHVEV+mj0EVh6AO7Ia78WJMuUuPdT4k5ZY5j5jhlclVSEXJbPL5uStySsRveFCkXPvz4GSUbX7WWyv3TZ9zd3hrv83DPdHfLfDoiwZsijxO8U3oHgjNN2GAukFQLZlG1gFQzVQR1po2e/MSzb3xE/3P/mJoLz7/xEafnd2aWoMp4OBK6nuNxZLi6tvkVVhvmRTv9dLjno69/jfF4sGxntCB5TUZhiaSUss1zaabOM13XG22htBHsHRI8CpwOBw73B2opTMeR6WSGELGLdH0gTRPT7VNiSQiV3juGRaJTbDHpS+L08cco0A078jQzXN/Q73a896Wv4PZX34K/8oZvFb6QQPahS/PLeHHCqtUeghB8pPhKDI4QFZ8qKc8c7meqeoYyUGrB3EIiuNgyTpFal4nGQcu8ONdBUwtwznirVSsqESn1IjtbSQjjyYTUhboOXucyzhdElOsbx+He0XWO7/oe4fpqh4+BECJ9b+YOITY9XBGjAEhowWtqrl1KqTNVsxkpTDN5LuSaGU+JaZ5x3lELzCmtvGDxjuA6uuiJrvsi/lQb3hEsWrIvjspPCmIvX//kYPYtMh1rFvcczF7cHLgkGLSjoUrjxiIsNAO3bkXxDnx0aHVozmhOOITOF/ad0HWOqyFwte+bvJ0HHJXCPBfuDzNzytwdE3enTFXICoXAWfTr5ca5F490w4bXIeXEL339G8zjzP3dnSnRHI/Mz55R54nj7TPG508Zj0dcH9H9YPf/LjAM1ovhosfVQKkV5kIpRn0hFzLaEjzWTIUqH/3S15imES2V6fk96e6IOOF0f8B3EecD4atfx3ddy8he9rtYYmcaTzx/+iHzNOJFiN5b06T3xNjhnKMUq5Sqwu1HH/HR1z7AewuYpfFvXQy4LqBI08U1rm86jaTTCGqW995bJXW6/ZhYE15gFxz7zgOyymtqSRy+8QH3z54RdjsO9we66xtuHj9ht9vTD8PW8PUO4YvLyH6Ga2bxmgagNWaIc2Yg0ASVS63Nstbceoxvap38IoIT36StoMWxtjnXEjtipgIiDmptpYaCisNMwKycovV8XAuntVZBmvHCPCnzqI2LtzgHiR23k6YywLLT9b+L+uhKJzg3d5lerRZbVddiOZ9SKq4UnC9rc5quXdfbhLnhs+PzUgte/pye4883phS8iJd4CRdP5ZXv1IVff+nQtX7axs25a9lRxRorLeRVMzRx5gLo2z1mOXxVKFXJpZJLbZUaoxRYvldaP9oitH5uhLk80o1ZsOHToFUZx4lpnDgeT8xTQqexNWFZiV2zNWFRHFKt3O7UbJKdgBdT3qBJWTmWioXppwtGU3BVTYt1mjgdjlAreZwoKVmfhgi5VMRnfAU3W5Pz2T8PUKteLrSCJZCt3jWqToRSrPG6KCnXNZsLYomdxgMGrBGrjyCQk1UhtSjzeCKNI6glosQBtVDSjGuvWYKqjbpKq8RWSpotmAfm0xH1gX4YVqfMDe8OvhjVgk+7aJZVlNpFPI22GsupYHqrnr674vrmSzg/MTy/xeg4jr4f2F3tcM4TQsCHgHOOvot00TKftSxGCLTO4sUZJOK8R7WSc24SJ80JpVYrz+S6BtbarMBEKk4sIxu7Qi4JneB4ctzdZtKsBFd4dKVIABzNFQnwgkiwEmf20FbNORWTKqmFnCo1KbVATUqZleoqItlcXaowDwnvZwgODUZ92LDh2wOvCEY/FQ+iYB6Egyux/FVBt1wErSBtAeqoLC1YwXvisEM0kh1kik3+DqjJMrVqY0gRUlJSsTF5d5x4fpjMBCEpWf0LVALBv9R8ZsezmSNseFPMKfHzv/QB8zhxf3dPmhMhz/TzCV8zmmeuo2PQyC569t5cHoO0XhI1N8vg7MrvnJK9o1ZI1UyEUKHiKFVwpaKHe+bcmrDGRJ2T0e9yxoWAeE8oFRcyGAlulYY8c3EzghKDmYv4JbCkmKh6zdSi5FQpanS6XItRd0pFs+mfh11HzAPiXEsAOSqVlGbub+/Q2vjsbQaPZEIL2Gs1ec0L+z9TOZlmRBO1VE7dc6aUCc6Tpqlxc7dg9l3BLwtHdg0Um1LBdBpbWcSmJHGBfrjm0aPvIsaZjz52hFCoKvS7PVdXV+ZC0vfmSOKFXd/R9RbIpnQus+Qq1Co474lDjw+hBbCLDJeu1IKaK3m24LFWy4hWBSeK9zZN1Xoi5XtSzhwOwvNnibEv7IYMDnufV3CNh+sEp3ZcohUpVubJc2E8TtSmoVcr5hQ2KXlSk/fSRPJCLTAMUyvPeGqXIWyB7IbPhsuw8dsD+lL8+uKzy98+SPzK8o9lqVoQKk4q0Qf2YY+jMgvMNRnlQEDrbOIGakoiaGXKhdNYmFPh+WHk6d2JUpVjVjKLOojty7JhFsxeHq0CRbdgdsObYZoTP/Pzv0iaZg73B3JK7KXyvhQ6qbg0cdM5nI9E7+mDVScFsWSICKIQxaNOqd7oeUWUTCHVZnOuYsaVCtNtQrhrDWY2Bztx+BgtyeMcMRV8tIysOU6aEghLg1mTqPMx4rTitbRgsyCtSaxkJc1WzUhzQscZwKS55gQK3dWePmd88PRDz7AbEIF5mrl9/oxaMkErXiveCTf7jqEpi9SqTMka27w4M1aoFVI2uuCcmFUoxxPRm3xXyfnb7N634ZuJLzCQfYmNZ/9vfLtlIGkLGEupD94pYk1NzilOlkYnWZupfLDmqr4LOG9NYEMXLPB0rslvgRSxFam3z/jgUW1Eca1IbaZ5VREpVqpw1cr7okgrZyyBLDlTpVn4VSFn+13JS9NYxS8OJotXvbz8PaheUiouaAiyaB3Y/UNl4RBb8F1dWXlBGza8DV5HM3iza6t99uXo8g0+uvB+LsUHlLPs1gvHs279Qqn2wZNlrNnD6APeNGQbfcDUQgAulEDUMrK1VHIp7bHQCpSqi8SWrp3bwuJK9PCIqrZTeG0mecOGM2qtjOPEPM2Mk6kJBKfkYMY7TivBCU7NtMO5hUbAqpjBCw5WTgR1IFUu+iBbyhTW+fYMa5ZCWltzo+sZvY9G4XNnqo9Kq2xaFdRR8Wra5pc6Ok4rzkMVMJ7fMkrkwT6XhzjX7OLtd4tMV9WCaEGausFyzLXdKkSXMbmosjTqQi2mWpLSWZ3hLCm04R3AFxDIPizEXUKcXci1KimbHMc0zYyjZWS9XwJNRdVRqxkfKH7lAfWD5+ZRR4yBm+udNVt5x9BHumjZE7OFVXKu3B4Kp8myuaWZJWh7j3Fiz1kU5z1+sIFZaiWls6e0c3ZzGCQS/B4RJYSJKR3JtfLxs5kYP6SLnkc3j3l0Dd57grMGMNSs/XITsHbe0zUCeqkZVZMVU18JfTN2KLNJjeXKeD9RZ6X0sAun5hq/YcMbQr/Z2djPkO9dZqHLj0JbOMr5tTWwlYs3teYVaBnYggBBKr0XHI7oFS8VYVnARmj+8POcqArTNHEaJxDH3XHm9jCTcuU4zcwrNcnuOdD4tZhs19XQ0bd7zdJAl3LlMM7UvFlHb/h0lFK5e3bLnJKZCeSMD0LqxCgEpbbSvfVeFOcozkGISNdZBFcK0mSupL0XVboQ2aui4vBdh++6lr1sTciI9YZgQWSIHS4EvA/srsyqVmj8VLBrvBoVzzUnPOeWALap+yxBpCqpmoFRXah9bUM1V2qy8eH7iB86xAkxeroYKDlzuL3l/mpPSQlfZqSk83fQbgNVPBXrjYnOEYxIi3Mmt2e+L9V0eUuxgDalLfnzDuGtA9nXXivC2oxlIsimLzeO5s5VSyF2Hb3rUJVm/xqpRVENSCvTD7vAzaOOvo986cmex4+u8U4YukCMvgWdlm2a5kL46MTz+0RKlftjZp7rcjAtK0tr1rIbQexcMzKo+Gh6tdb0Zdve9YHrK3MBGsdnHA9K1YRwIqcjXRDSVyrRRWIM1NDTBUAxi8tsA0q8o/MD5u6VqVjTl4vQ5UjOmftDpUyFkgvj/cgsibpXroZrgtsMETZ8u+GzZCLPWdmHm5DGOJCXY+M1+6uNkVpxNO66QOctexXESqYCTRovGo1IISVbSI7TxGkcUYTb+5FndyO5KIexMDXRdloWTADTNVA677jZ9VztetbskCrjnJiSac5u2PBpqKVw++yWlBOnaSSXQug8iUj0jq650gURqnNk51DvkBhgGMA5JCWcyxZAtkSPAL1zNtc6R+h3xN64qLHrCcGCWnwA540jGyMuREII3Dx6xLDbGX0meLxzrTE5o3VppFwqhwu1h1U/dul9sWoGF++wpBGlJZGcUC0Ji4Wl1qx1//QZu6s9Jc3ILDC3pi/fGjdFmMWTJCACvfNE5xApFtQ6bJHcXMRogWxp5g8b3g28VSD7SdfJ5Xy1ZkRfeLy4Lcusnre7FOCd8MAl67zNc+nPe0f29UJFoO2zUQZYLOvaaFuqJ4sP+6JAsO55SRQ5h2s6t84FVDxaC6VCmhv/NZkAtHNCkIq6JotyIX/knGsaeO3c9fy6WzRoOZcqTcmgmtlDE5jesOGN8E2/f3/WUnqrCZ57oh9u5aKJ4zXkB0Bxcu5i9s4yWd7ZxHhO+F64bqk2L/bm3V4yqq49N0pBrfXh/UZYM2NOhOAdIZj2rGIqB7W2wFm2uXLDm0HVejNyyqSUjNYiSslutS9/cDEtA2GR3XDeejGMS7cutpS2eHMenCN2HaHrEOcJsSfEaEGuD6bs4RwumvSWj5HQ3u9ECDFcBLKhTbCyzk/tTBAa/S2UVuG0wzr/tqFigSxQRTG1TG30Acv4eu+tF6T4lcuz7Eo5/6wLlQJTXbio16zBgFxkZUvOW0b2HcI3t9lruZDWTshzqUJEcN41r2jIuXI8TpxOs4krt03kXJmmQimJj+uJ46G21aPDO0eMjsc3Pft9xzQXxikxz4l5rkxjYpqMJO6Cb1SHdjgKxUsbXCCqBGeco1IhN4/oUr1Nhs6a0rruCtUOoZKmEU3KdFKm00xNBe1AizaSesU5b6vQ4HDeoVRymillptTMOB+ZR8vc5rFQpmKyJQgOJUlmPE442TKyG35l4syXe9jgtby+JGIvJ6fzb9ts2LKjuz4SnLDrPTdDwDvI04E0TtSS14wtmIRfmSviKu5woLrnKI7DsZDm0kxRmjKtgBfjK3rnuN71VvUJgSePrtnvBrOuPY2Umqk0ma5trtzwBqilcLy9Y5pn7g53zCnhdx23eUeNHufhUaTNi0uvhUN8RPo9EjxOWZsOhcbWEaH4SHbBgtRgQaqIza3ivSVrWjYWWDmri3KBzW0O74MFlSzBpCV5Fo10g+3/zL9dMrJ68ds21itGHVJzKqtiAXtNEyWNOKDvIkMXKVS0OBNwxhanXisVacke06+VdiwiDqcgVXEUXJ4twXS65/D0G8TOU3L65v5RN3zb4AsIZB9e4JfQy0C2RY/SgjopxmcT5xE1LcfT6SKQVRsEJVemsZA8jKcTZiNLK3dA3wfK99yAwJwK05RIcyLNlXlKTFO27kxVnF+oDmcuXnVL0xV4bzcSTbXpvSqlBIpWG5Q+EPs9aEHnRBrvKVKYTmqKBNHbnFtbiTLYzcQ5oes6YtehtTJzJCFQhJogHTO5ZPKYKXMLZNWm8CSZ6Tiu3L0NG37lYmmm+rSc7sNAVjBNyyE4boZIDJ6rXeTxVY93wvEucTtap/Wiu6kouVZSUZCC3h+Zi0dxTNkxF9d0qt3acBmkEkWIwfF433NztSOGwPXNNf0wGL9/ztQ5r0HsFsdueBPUWjne3XMaTzx//pxpmghXPbd6jfbWxFz3HXiHq03TkRaY9jskWpAZWkLG+0jwFryq76i+8WhXIfWHo2hxnkQrpWRqLRbItiDWeW+PYCZCPphOu4jRd9Y06eXIXTKn2izXH/4Wp4KYqTtVlCKgVObTgfmkZJQudgyxM6voWVBZjriZKThpTZyh9bEtgWy1xuxamzSZmSDV0z2Hpx8ZpTjnb+afdMO3Eb5YQ4SlHP8JQddCCVBYJa9yNpu7UuvZrWR9/7nOUOqi9crapemckFIh50rOtSkiaDNS+AQ6w0oxkNbgxVref/Hwl0YxEJzzTV3AoYvmZDWThKXrWVp9xImzbmon7aeJsTvnV0MHk7W2Ae/E49vrjoBTf+FEtAWyG36l4uVrd7EYeIkU2/rCLCHU5K8avSg4RwzenO6CJwTXOrzP/L1l+l7pTNWki0oxRz9w1BrQelkqNZqAd6bdGX3bT7DAYeEiAiafV2u7v2yB7IY3w0otyFb2LsV6IVLOzA6Sd5YDkZaB9AEJwQLZEBAfraq42KG311pmCPGBhVOnjaqjLVv6sFz/wlhcgt61Qm96spR6ZuOpPJjTjXGw0ADkwdx6+byqazbSFsjWRq2rtbbH8t4XR9H5tXZGsM6sTVHkJdqAcei1VNI8maNZ3SQr3xW8XSD74CK0LCIPeGPni3tptqoVpqSUXCinSqkzOVeePj9yPCXGKZNzbatBk9JZyiK1QtHaGibPVrfPbydAyLlyOFgWNjV5rGX/tZ4DYLdy6Gyl3GotBCe2oMU0+1AjrJtntAWb3nW2anQ7krvC3MJ6lA7EEeKOYb9vsiWmysCixec9WiuhgkgACXTDyJBM0SF4cyVx4oje7DRDF9ld74h9fKs/1YYN31q81OXFMq2udMAWhIqY8LqgRC/sYod3cH2140tPrulioAvCLgoiSvJiZcgm37N0MNdSGk1JUEZydSCeKh0qHYrRk7w4gnc82nVcDWa08t6jK26uBkAoVKZpZBxnjscT9/dH5mzyXVsgu+FNUErh7tZUC+Z5puTMaYaP7+EYPc5d88Rdk7uO/fUjrr/0ZULX466u8TePER/OMlZAEWHV8lBBSzk3WSmAKXHUpnCw8Fxtcbe4Ri4Omgq1Gnc3WxNyaq5ZC3VgpRBUe805T4iWvRVnMpm0LGia03lebZq21TV5LgEpCVdnSkqcjkemeabMs5knFBtRwSneWROm00RQ4wVLFduuFtPTDe5iDCpzGvnoww95djiS0vzL8Jfd8O2At8zIvriaWtKySwB7kc1sjVWlwpxM6mqcMqdjJuXC89uR05iYp0wuCjhELBvpvGUpVSq5reJyahnYqtzezSYnV5VxKsy5rkHuclz2vFqDhlsaNZRaXOtWbo1jS0Z1TfAotdTGzzGtW4DqdqjsqVKoElGJIJ4QB4bhCucdPoZWorn4hpydm3MRCPTdidSZC0nf2cD13tN3AyGYZq7vbDW+YcOvbLw6mLVni7SPBbHehiV9dFzvopX7r3d86fE1fRdxVIJYw8jJi3Fja0ZqtaYPtUA252wsW51wxVwEJQDeI07xEvHBMrCPrgaeXA3EGHhys+d635NK5e5k3vDTZPaih+OJXJVStjB2w5uh1sr9/T255FVJ4zRXPj4WOu/ohp4vO4/GnrC/Ibz/Ffphhxv2+KtrxHlKreSmkVprsURNa5BeEjV6keVcMp+L5qys2deyNnItgayZBdk2Ukrc3d01J8qlIfKsba5aiTHS98ParBWjGSxM08jpeKKWYs6VpVUuHI3GB32AwYOWzOl0skA2JSQXaNbwThWvjT5AwrcqjavtPJqQs3pnsXL7LuZ55v6jD8kukOaNI/uu4PMHsgvZW2CdjF5BKVhfaUFtKYU0Z+ZUmKfMPGfSSgvQRh+wDywd/loVdW1/K02gCZxXIbfMS21NWqXUFtgu7mHSBmrrfGzp1oVWUKuizlHq+QawlEiW8ofJci3H1s6sNWAp7qz+UWmcncteNzlrWcOaaXbeE2JH7Pozj1gV7zxd3xMah8nFsBL1N2z4VJzjwy9sc7ztJlu1Bj1vbxEiERY1Amtw9CJEb//ugm8ZWEeMnuA9wZuPu6w8pTMH/0G7yYslz0Yz8HoWdPdOGqXg4uGaAULbblnME3JptIJ6QSvYKD8bPh1LIHg5h4HNBeK8NWiESPUB9cHkshoHdlnm2XzVaHlLkkhfSB9djP3FFMD+cxfHwoNjsEpppTSt93meOBwOjOO4zn3LvLgEsl3XoyqEEIgxGOVOlZQy0zi2zK6aQRiKijVWiwgSHbFzlhkGm+OcX78HhzVceu9QcThxaAuC7R5xee+wSk5d6Adq5gplcfLb8E7gcweyqkrNJkjMBRl84YfCBddUtfGBCne3R7761Y8Yx0TOkBKUotzdjfZasqC0tLLdnBKncWzuXYsGs6ItasyaOd4X0uRRhbzwY4HSuKt2SC3X6kzxwDKsrM0epUBK0nSn1bK6Cj475jnjfSUEAN/UgjziB0Qqc1bujkqMldBlYkz4UOh7pVPFOWfSJktHqDOengsdj8Wzv3pEIzUBTU6suakg9v6XiLsbNnwShEaZ+fSb+Sdx2h+8Tz/f1PCqPOwSwC7+6t4JnXd4B9F79n0keMe+jzy5HuhCYDdErneR4D0lF/KUqCWjJZmAezVawUJPWLK8qqAlUXXJyPZEp3gHQxfoh94ayIbIvrdxpyUzTZUpFe7vj9wdJ8Y5M06JVOrZBWwblxveELkUai2WPayV4Ht2148Z+o7h0ZdwN+8j+z26f0QOO8R3eIJJMdYzp3bJyNZaQcAHT7DJyTKostDvoGIlfuv/XzKyLZPLeTE2z4kPv/Eh9/f3HI8HPvjaB9zf3a1qBiLSeloyVZXr6xu+8uUvMwwDu92eR48cIQRun9/z9Q8+YJ4mSjWTopW53ji1X3p8DY+vLGxwnqubazRnpPPIGHEoMTpCMMnKWYS0BLJKUyuoeEk4NfEya/qqmMlKMgWETX7rncFbZGRBc2krJcfibPeC3CuLCUEpFswe7k98+PVnHI8TSqCqp1bheD8zT5b1yLldlBVySkzTbEHdMv3VxvOpBa1wKolxPK9WtR2ItMapZUqzZi3LxTgnSBU7/qZxuVz4FVldSnLOpDlTvWscWSvVWCDbI1RSzRzHmZArfZfpu0QIi5OK4L2at7W3r1ucx6lChK4f1ljjwffmvvDE2oZ3DG+ic/qmQeyyQUtUfv6r0jIoBtNrtZ/RwRCNp953gUdXPTEGrvqO92+u6EKgi46hD3gnTBVyLdTSgtklQGj0obVJZM3YFmqdERdwWggowcGu8+x3HTF49l1k6BY76mKqJ3PmeDxyfxiZUmVOiVxaM5nbgtgNb4YlI2tZ2SZF6T3D/pr9fkd//Rh/9cQC2eGGHHrERWsqLhYK5mzVTNNbLyttwHszMhARCnbd16rn65/FzdI1qt9F70pbjM0589FHH/Hhhx9yd3vLz//cz/Ps2TO8D3R9j/eBWmsLxpX3338fcFztr3j0SOn7HarC4f7Ih9/4iNPpSC61LfpaVZSmm14zV10geKtMDvs9UgvOC+KdOeoFa7xUMQtcaVrRrqrZy3M2LVGtiBZK4wQbtkD2XcLnz8higan4xYfcglla1ubBBLmUONqF5ZxJeiiCLCRuWUolbbXZShgpzczTeG76ahFzLXXtStSleWvZFbAGrWu35jkQlpUiJMZAX+gRSxlyZesJJTtyNiH1SxWCush+ACVX5tnoDPNsEmCleGL01BoeJG2W4HR9aen+fDFvJQ9+bNjwxli9yHmzQPVN3nMZvC5NIw/3+cL7Xzome1Hak0tKwcKJNZMDm8CCd8T276X0v5qdrPtr22md3t47aNUYarXXLrqygdUn3guruYJrwfTlxF9axutsnqArpeB8AFtGdsObYzXCafd77z1d39H3PbEz21hnGpAtIfOQOrBSBlrj1drU1dR+rJrYsr6X1AFYttAkuFjvD7VxY3POjNPI8XhkmmYUVrmvLnb4YIGs5EytSghmnuDaw7eg1IdA7CKldLhqmdJ1LYkFsjFGxHvTkQ8Rp8UC2VLWIFWCQ7wzGqCWxntXyBWkrjyK9Tta6RJNWWQjFrxT+PyBbKmMdwdc9IShbxp0HmITT25BnkLr+i9oKXTR8/hmx9AFVD1VA6XC8VipdWyrzpHxdGdkbmZO4zOc8/T9FV23W2WrpIl7nAd1OzZtVAJ/mZG1nmgRkMSahV2kPVYOnWrT1jOZkzwH0myBdNd19F3fgmoWtRPyOHGUE95BGhPjcSRGz1e+8sgGdoR+sfvj85doN2x4U7xJNv8zZWM/42c/iU7gWOgEMATzT49BuB4C0Tv6LnDdm8zWrrdHF2wBubDSV/qAQAyBYdgRgm8c+bRmv0wzU8nFJnkR6JzSByV4pZdKh2VoqZ6crYQ6zYk5F6ZUOIwzpzlZgxcCzuTwVm3ODRveBEsm0QcEZX91xZe+8hUePXrE9fUV+92Oruvw3lujIoCqqexIW5wKZjIArU8EpnFinmeWoNaatposnZx10pdkiSydzYgFrQrPnz/na1/9Gr/wC7+Ac46r/Z6b62v6vufm5hFd35NyZppmSilcX13x5L3H9H3P9c0VV1d7YteRy3sUzaR5XukE7TQs8SPC9X5gvzdbXMqAlGQZ6pRxKbdjdKgzRWjRjFOzn+V0QufJHuMRyRayLlSLXGFMlWOxJu0N7wY+dyBba2U+nvBdBOfxsa24gn84iT7Islaid1xf9fTRgtiqnpyVLp7WzsmcZ+b5SNVCyieOR+PpXF2/x25XTPojdM1TXY0LtHZWwZIpcdWapIwkv2RkYRWq1Np4dbo2oqmqNWEF4wXl5EnJNGDT3JH71FaggRADAsxlRvOESKWkRJpm+t6z2/fcPNqvGazlhrJhwzcf3z7EFLl4OMydSFCCCJ1vj+jZd54uOPro2XeOGBxDdPTBdGNXrdilUQYLqH3w9H1nEnfZuPdLlikn6/Y2+0pr2IxO6Zz110SnRDETBWqmZOvkHmczP5hyMW5sKs0Eoa1g5aUayoYNr4daAgNpTllO6IeBx0+e8N777zH0Pf0wELzHi1BLWYPYl6g8LT27ZF1LXSzMz43KIs2EJ4Q1A2yxqzwYkCnN1FI43N/z0ccf8cEHH3B9dcX3fd/38ejmEcNux3vvvc8wDKSUOJ5O5JwZhoHr62tiDOz3A8NuoOs6lEeIN0qea2YLy/y3zNGuNViCorVf52CjDVgQLuJAjAohNeM0ozmj8hzlANraNdfG7LrO4fNk43XTkX138FYc2VIKUnzLuJqNq3kvn/VbAdOMTdn8j2uxUp6TFk8at9V5WV07wDhA2kSTpRirri76dyoXVAVZ5UXWDs0lpl318lrT10Uga30wRqJftmUk+DboRZsQem1zp1EHSlaqCBAQF22LdQbz6mpdqIq0zudLtYJPDCy2WXHDNwUvlxdffPpaLtnl29/y+lyWkUYlsPEavNCFQB9dUymwjK3WTJqhZsELlGFoHEB7D+u4kpVW4HzArxOaVVOCD3QxWCCLImL61KZRa7w6LZmaE3hPKVYirdU0qEs1FZXFweucYboICjZseCPYtelWbfSzSY5cuHHZOz8d1nhl85sZLWT0Irh1jTsbWoNxK1O2xWCrjVQllUzC+lByaoo/Ct4FYuxwzqh10zQxzTOHw5GcE7UqXdcBNEv4hC7PkzlVmsbsy4GszdfLeZS1+cyp4mo7OmdNz4qSVSkq1rwqDpxHXZPdWmiLzSxJzx1sG94hvIVqQWUerTOR4PGdcVadCOr9hVSHcrw/cn97T5oTeZqIruCCSXIonuyVvvPmj1w9IkqtiVozNZv2rK+RklOT7DhTF0QcLgw4H9vgnikltwE+UWtu1IPz1a3r/8+cONcmRLOqjPQh4j0gBWQCVcpcOE5W+ohdoOsj4oTglegrzguxv2Z31dP3jn5wdJ0jBGnSXfW1McOGDV8s2vX9ie84+2udP/NJeAO+7atea8GrcWEtyxq90EfPk+ueofemAVtGRBPpNHN4eqSWzOPHT+i8Q3d7QvAm9dMCWFMBUWLXs9vvras7jeRZqFoJ3rHrA1WVlCspLROpR4rdS7I4ppJx3qN1T6kDpcJpKpySNaskFQp+5eK3lreV77thw6dBscSP80ZRCzHQ931rAjaqymLg4z5JarElbHIzE6ilME2jOVldNFV577mpipfGLQ+h8cUXOowjp8zd3R3j6cTTp08tSE0FFPphx9X1DSlnvvHhx6ScOR6OfPzRR4zTxHvvvcf3f//3c7XfM46JVCHGyHEcuTvck3Np+1lGy5mXi57n4BdNG3xLZnU+EH2zpTXaME4rUTy+6yklM6kwZ7N0l1KQWpmrLTjd8qEN7wTeSn4rzcl0GVO21Z5C8cGMdYpJhdRamU4j4+HUnEMy3hlHTp01aokzf/MQPD64Fshmas0WsNrSqwWlTT/2opklhA4fdra6axnZUhOlJEqezsT3xk09M1QXdT5Q51fvaSeeGBTvQbVQdUa1kPNISgsFIqB0ZpHZewimgxe6Hf0gdJ1p5XkveC/t/mNlEzvu1yfCNmz45cNnuQgfdDq9hNdNG0Jr5hKzfx2aLuzQB26uBvZDpKSJdDxRcibPJ+6efcQ0TQjKe++93+SFIjFe8FOdw6nRgGI34EO2aghWzYneoV2we1XTqq4VUqrklAGhJEFrxrmAOI+Kp6i5D05ZjRerNEvqc3nFwtmtdLnhDdGqBQKEEFswG9es7KLlCjzIXNpHz6mX5WfOmXkcrUnrdOR0PNr7WnY3hEDfdQx9D96bEY9r2VhtV2+pHO4OPH/+nLu7O6ZxbhbxQux6ht2efH/P7d099/f33N3d8cHXPuB0PDGeJq6vbii5mRR5a/I6nkZuDwdyKTw46Mvj17NZUcUqHohlgb03S+hd7OhDwTmhi9YA6lE8nuAjuEACxlpNiisXXBurZqa0BbHvEt4ikLUBV6saMT3Z6sunDFWbg0mmlmqZ2JTIc2q8tWRZUhEzIihKLRkhIxRC8AzDQK0F06ESvA/EuMP7zga/2IS28tWWzmUXrQ0acC5SJWNLwHIxDUvjDTWKA7Ttx9ZV2RPjgA8mDVZqQVWa5qxivnvVtqny4FFLbvwgYW6ez7mYi5BqaeVQt+raOt8En3l489oyPRs+Lyz7fzntwQPO7HqPXzjbZw9z1p+sY8S2KcsV+vL+LsbTuov1g5Zp6b0nOJuQ+ujbI9A1/qurQpG2yGxct5zbfSPNpDSbCol2sDS9sAx/17JY5jKkIYJWUN/GKHhfycHuV06KBbtqMn65VMQVqnhcUSrCXD1VXdOtvvg+Won204L6DRtexGLVvtiV+5Z51UrTpFuSHK9pJHxhStCFt9Yam2ncWJE2r7SyviwD5YIcu7hXLpVTq56ej0Hr2QAhzTPzPLd53Oa3UsrqpgUPx/uC2uzdtZ37ck+qVU1eU6G2+4pY2aadjkMwbWnnTJIvemcNotUhatSCrDAVxSmEppBQWrxwpilueBfwFvJbpoRRU6EeJpxPhJDQlHHOM8+J4zhSSuVwf8/98ztySqSUmMfZVp2uQ3xntrXjPdQD3hVurnf0/fe1fGlHJeLEMwx7un63Trg2n/nGO1p4qz2oGK2gAOrNzk9HVEszQnBrabKLXfONPlvudX3P/mqP846Sj6R0i9bEKNls96p1UtZcwQnVF2oIFBzTeORwcExTIPhMyUd88AxDR993iDi62BNCxHvPbrej6/r15rNhw9vAiobV7MhZJhqL+i7rEMA6wYmYJmNwK3sO17KNC+WmkUIvwtzzNpyYwcjDrbPKZXnn2MVIF/xqctBHUyV4dN3Td57ZZeromDNQM9N04nQ8cBgGbu+ek0vh+vqKYYg4F+0YnZ1tCJ4YIrVaoEwXQU0VwZsFfDNTsUD2cH/idJzIOfP8+T339wdKhUk9SR3iI/H6PcLumoo1irrWeNLc5795f8AN37FQNXOcYdix29t9f+k1WXRgV0mrZv260AyWQNA01NtiytkF7rpIbG6VXpy51XlP18Umi9XUeySgIrhFbK4q85wYTxPzZMkl1xQ5Us7M08zpOHJ7e8uz5885HY6cDgemaSJNU2umWpQYWkXVe5tfa6XUwjzZXF9KoeTckl+1NadhykKtKWzoOqBDvCO4yD5a0D90kb5rgXpxUArqRu6K8PFU8bUy1EqoFfUCXSD6uAWy7xA+f7MXzSavVFQTIkINxQjbzjG2QZBy5nQ8MZ1GckrMUzLru6qIT+DNcSenEzDhpDIMHbv9FeAp7Kg6IGLOISH4lrGZqSXbanRdwXq8DEBEJOLDiMsFyC0zW1v2JqyUhK7bWTa26xh2V3YDGHqG/Q7vPTk5nJup1VFKYHKwMn5qbT7XSi0gOFKaGccTIXju7yvC3G5efbuxeIZ+Txd7YoztnCwTvPLwN2x4S6yycoB5ua4tStgK8GybLC1rGpyueqquKSmbVeT6qba9JbPDOvl6J+f9rvPs0tRlTl19MBH0/WBBbYyefR/ookdK4OSELIBWu1e0isbpdGxd2L41hxg3VlwT4XNuLUlapSM0WS5TPhAsntdmW+3UIQrT5NCaGY8HUqncTZXDXAndwCPXs4sD2uT7HK3JUx/ktjds+ExwzhG7SN8syIHWqLSo2ri1IWzJqK7VlVWQVc/j0DlcCPg2xqMzi2XvPD74i6ysid7JkpVtmzJeeWpa6boaD9VSybkwzzOn44nj4ch4OjFPE2meW+O2mRRIy4Ce9yWNKqGkXFp1JZNSutBmtl4TH4Lp57bjjsFRxeNF6bw1hA7R6EiqShGo4lAXGItwl5TQguNea4sBfDMf2ibTdwVvFciunZBrGUQa96WVJNJsdrM5XYg0F7SYc5dNQxlVwUmlC5ZFQQSV5ttRHLXaIGxppjYp2cUrKuScWtuKaxNzaRnZtDqA1SbPUauCmCNKyYXs83qjWOz4jEpQWsfmgTSP1DoznVIzPtC1g1rcmcfknTbnr0oIgm8lTO+VkjOpdzinpHmiC5UQM851oLaS7Xcd0b3dn2TDhnOkJWsQ+6DufxGKLWoCXi7sYkVbZz/EaNlOEWlBXasaLCVNWB16bBd13cOiYxmcsIuRGDzeCdFJc/SqaClUMWqRts+uh75MhmlmnsM64YbgV4khacGyc24tUV5SDqQdI9DcBwUXPCFYc2iMkRgjuIrPGe+bPFL7voxKYCF8s3B5wFXcsOFNIY1W0EXjyMYY10rGsqBccVGKXzOLF+wAL+aG51GC0zafKl6qNXiJEqltQfowoFvyJUtVckmoeO8fUBrOBkU2n9eySF1dZIi1JXSaBNj6aAHyOE5rIJtbz4y9J1tTWqm4UtZA1glo8eTUm2sfnlo8tVhFpLSqStHziDRaX4sHqunVi8tb5eQdwlvJb1l24iwjArQmCjgdT9zf3pqlY3vUWigpUdJkAaUr4OyC7l3iybVSVah4Ch21OuoxMk22uvJi8jeqSsmVlAtgAbNVTx3O9YhE62qcD+Q8Nx5capzbAq2jMjlrSFv07vzBr1xbFyyrU8tELUfbRh7NrlYrZ6FpJYRKiAXnlP0usxsSPhRu98J+Z2WXYah0nUmPeXcwb/ku8pXvSjx6fE3fd3z5K4+JcQtkN7wFpHHneF2gdVYzcEBomdkolV4qTqD30Adz39vtOnb7vXHSnUdcaLxXf6boXGZk9bz9hYjgxEwPvGthoJpFpyjkKVFmZZ4nKz1W4845Z1mVUot1Vk8jzguPH98gomv5VPAQ1By9tDYOe1gzwrLw5Zd4vip9P6BVCDFydW1ORilXkkxkybgYmzygUTREy0X4utB/lpz3lvXZ8OmQ1oA1DAOPHj/i0aPH7PY7fPQPzHXs5yIjaQkTXRdQ7YoT6L0SAlDBS8bLzNmPEhAPdKAdKlC0srQay7JFJwy7HdfX1wB0LUu8BLRgDcopJdI02RxezvSGNbuaEvNoc/o0TkZVSJm723uePn1OapzaRcVooRnYYXqkufdN+5790DN0gUdeSUFR7wkUnHaoCrlCqcKcK7kKVT1FlVwrrljwWqYjmv3aPLfhOx9fSNS0ZD7Q2roezVp2Gkem2bg0y0W82M/WWqFWcEYG966yHyzbmtWR8eTicaOn1ubnUzClg6XJrNamIJBRtcYw5xJOmi90k99aVotn+Stbk4rYcT2QO7mclwS0ZrROTaMuU8pDoTpb1SreK84pOVXmueCdkufMPDU/6QH6rrkRkXFU+j4SYgc49vvC4yfXS5J7w4a3gi4NSQ+i2XOACRaSOVlsYitBbKHVeccuWFnvaghc73uTunIR8aEt9nzj3j0MZJca/pnHrisH1xyszTSkVOy+kFN7zYTZV95dW1DWWpmmE7kkpmlPyjOxBEQCUexYtDXPrDZCcnb9sgXupRGJtq5x20ff9wzDgEuFmJSQwfnQ+LdLY2cLI3QpzJ6D2C3ns+FNYZavkWHYsb/a0/Xdyo1tE8N5/tGFGrTIV50rK2YmAtGbwUevlc7ntkBczWdJZAqlJYcuOfJteyImAzYMzCkRmkWuXFQkVK2Zew1Em6jykgGtalzYnE25KCdr8k5zZhpNd3ZOZ6e9lwNZB84CWdECNaMpMl8P5LlHQqAET2mW9lk9pQW0Vc+OnZalrdSiJIFavTWUbXgn8PkD2VZa986aRJwXSla0JJPemiaOxwPjNF1cxFBTXvlAPgZ8a3QqKJVKVeEwOsapKRxkoeSls9GC0aqFaTowL1JYS7ZVBCcdIh6zrbMAV5t0V1U9z+3aypLuhe7Gde5vNwTN1JqgBeFabdVnYs/eJjV1a0NIcooT83l3YhJi3gml9OQUbVJvJdxSHIf7RAzGwZ3GTE4FcZYNE3fxZW/Y8AZQjENmF7IsfV4WULLMl3Ju6Mq5jbxCkowTpafDxR6HtAnT4bxD/LlRcvFXB2ndxW7dz8qfbUckWCBr9Fdz2XIFqrPFZK1CLf4BL/DyjKqa48+S/UnJzEeiN26dTa7NCEX0oQV0pfm1t4Cg6UzSbKi7fmB3dYVPhV11zBrAedOVfXAveCHL/eDJNj43fAraNe2cEMLSoGhKOSGGNpa0UeDqet3ZtbdcY+25Kq5mfJ7N9Wo+4uY7aLKUooq6iIQdxD04MRlJjIrn2vHUWo0DexqZpomcy9khs9HmlsXgpQSYYtruqWQLgHMm5YyKNYktygYpmzFCaYmrWuz4lgXrSnHAZLhKKdScKU7a3G8NYW6e1/1mCRQ8qRQqzU5eFyteQ6lK1kWFYcO7gM8dyAoQA4Qg9NG6g6ecmcYDaZ65f/oRX//gaxyOJyqQ20W1Cx1XXUdwnt1+z83jR20VaBdjLsovfHDLh89umROcjolxTKgq83xCZKTWxOn4lHG8R6tRC0pJjTPnW4ZYCMFfZFuXUknj93AmqRuhvpHul1VokxapbUW6rE6XGSyEgI/xJaJ7mgPjKSBOuA+B0CbbYbii63at+SUQxNF1lZLvuXteePR45vHjG/ZXA8E7+l0gti6bbUBueHMIxUVETVJqydIsuUMv0HnjldZ5RMc7tGbmPJHyhEMZ3nuMH94jBqEPsOtaN7IPuJaRNfkg3xaPF1JBC+dvCWbbBOPUFm+1Vkp2VpXRSimB2tz+xilQSn4QzFatzUmoMk0Tx9PJxu8wEFyT22o8+CWQXTzp24EAS4KrjXME8ZEggZvHnmH/iDlnZH/E358oFcYCc62gYs0lsnFiN3x+LItIo5n17PcD/TBwdb0nxLAmaExhx5oZgXU+suu6moSjFnw+EcfnSJ6Jh48Ihw8RrabLiqKhRyRYIOsr1UWKhtZJAqiQ8sTt7R0ffvgx43hinOYmjWXZWuctW9VmxSY66agoqRQOpxEXA8WB7ztiiRyPI4fjiTlljqcTp5M1fGut1hhOowFVXYNliqkOJQeTFKQWxnHkNI445zjN1jCN89QwoCEyzxMZkOAR8bgiOIWsypQyo17Y1m/4jsfnD2QF46k5WpckJCo1zZR5YhpP3N/dcXc8UjGFWBBkp+xjh4gQY2C3t4HsWnkh5UL46J6UZ+a5kpMjZ2eNIOUAHCglcTw+ZTw9p1bTuSsl2XE1qSDvPX3XN77PWYLECONLQ4qsXZpWQjlb1S5UCFUoRV8691ojEaMl1KVbWoRaPTkvWas2yJwnJaHrLOjtghJcYJ4h+JmS7fPjKZHmDNHT6caV3fDZoS3XqmJOOII0nqf9fmm+cgIF431rnqlppE4nqxRc9ThqaySBGKxJxS2Z2da08qAs2ra9pHAWDvkS0rqW09RaLaB1jlpLG0tCKeFB1/O6KTWlAQVysQaS4C2bVatx1Zeu70sqxZKTvWj0Zs1St3KSiNDvAv0AcyocizMJrlJJxxltDSmW0b0Ijjds+BxYKhkhmDRW10W63hqtcsmU1Ax/WBtQGs4NVs1CACkJl0ZcmvDjHf7wHDSv+6pxgJuTlerFsRgJIUKLISmlME4Th+OReZ6bckEbQg84uxfjiSUjq8w5MaWZkDrmnFCBOZtF7bxQDJpaAU1Tds3KtGNY7OZRoRShZMjCqnIgzkHO9mbvobMVZSr5IiNrlr9SxTTsayXVcxZ5w3c+3srZq+QCtTKpWbxO40hKVhLIuTDnQkqFilAARM5c1YWvJ+dBlXNhTplpLqRUSVkpNaN1buX82gwMHEM/4LBtlX5RRVhMGprMSYytZHNx3DVT8nTWsVvPZ8nIsnZhGtne28p2LXfaSDdR6BFgbRYTzpJDIo4QB2Locc3M4VyF1NaUctkVanInKZV1cl4yxcuNbMOGN4WotuYPC0a7phjQe2HfebyDLD0pR2pWiKCdXcNXu6Fpvtp1mOdE9RXnFedb8Og96o1aI960nLVd92sUKu48bB4c3NmZy+NwArkFxr5xb70PqyblIrGntclyOUfwnnmeCSGgpaC5XEy1BuPvNbek2gTa4SIzawnrqpBy5XAcOY0TuSqp1GV+5cGM/kpsg3PDm8H0wpdr3BO8OdMpFcmXcd6Fqcnl3HO5LVpgqwWtCbSsE6pQrfGyPeQigLyc9LTpvZZaLvpOlh4SW7iac16HasL7vM6vy1KVZQ5blAtaHwwo3jtUfTsOD3pWSkFopg7nZrhFjqsgzezgosJSBVwGvNEXSqE0Zy/UjBGEtmjf6LHvFD5/IFuVeZxAK6dqAWWaM6fTSMmF4ylxOCXux7xKXyFGLci14vVc6nQC05Q4jhPTXLi7G7k/FOZUmWYoNSMCwWViAMGzD4/Rq5tlvgQnlFI5nWbmuXCW3zF+UKkTWgtTzozHO+b52MqcZZX9OZchtSkcQN/tGXaP8BLW4opqZTweGceD6VK2bLJxiawj2znP/up99rvHhDDQ+T0aL3bi2uq4Zkp25JQYTzOnw0Qtlf1V9+I3/nn/VBveKSjQZN+04LSwD5En1zu64Nl1jkdDJDhIh8oUTtQcW8OXTSjDMLAbzCiEmjkdDmYH6y3AFDFt2OCNQmMW0QGWysdCOVjVTGS97AUzUFAHDjEBc6DWQoyRXDJdjPRdbxkiWCeyWivH49H0KEttxxEs01PqSxmYOWVSTtSq5FLJreJi2s9WMi21ceqK8vw4cRhNyq+6iDZxeHMuawtifcj/3dK0G94UFhianF3XdXRdR9/3xC7CpExpXBupFlkrnHvILVsLIOYsKVrQMlPLCWpB2kKw1mwuILnRihTjkl4cT9XFgdMcu0qxZubakijOO3yI9Lsdu6s9ziXmURES0QeCGE1BqykbVFXmlJjTRMpmFx8b/9eJtIZQaw4N7b6Qa6U0x0vLUgeC9yQC97NaX8wy90mBIhAKZRoZp5kpJTvvqgQFX8FVwVVdq1AbvvPxdoYIpVJLNjmtUsi5tgFRrQyYK3MqbXVpF2upZeXcrN3FWEZ2mhLTnJnTRUa2nANNESW0kqAPsTVcCT56nBdyrggjTs4lFpsAE+TKolJZ8kSaT03rNrVOaZoMijQlBAtkvXOgeyubakUpQCXnkdPp1vh9SyALqOb2mse5SPBdoyeUpS7D+ck5I1tLbdp7ZdXJtHPeSiQb3hyWkbAMjFMr4wcn7DrP0EX2nefRLhK8kGrHOFpG1rfMixPBh4AP3gZEVXJK0Hix1ds4JgRo174sVAK5kPgRy6QuEkIPIlk5c2jNalqbi1Fz6WuuRr6Ng7Loy9Zq3dFACmaZWX1LqdZXBLJzYmrOQnO2+5FRi5qZi0Ku9u9cKscxcZqLBeSdb9WYS6vQh0HrFsxu+CwQWTSP2/XdrvHFwetBmLn0cZx5MbQ2RdvWOoc0q/SaEXPlATXpuMuMbJtu2++Xaeis/mOPszbsshejQlhGtiQbp9XVlQbkLHvTVIhKy+5adnbJyDq3KJu03hVnmWgAVwq5WhIoxGDzujP127mcz1vVDFBwlrau2TKyy7GLsvJ/7f73Tfszbvg2xFuoFog1O7nmPOIq3lW8BKRWQj8Qu56YLPBDKyLQxbgO4lor4zQhknj6/Bnf+PgZ05x5fjsZZ6eAEAk+4hx0Uek7HjSbiHN0Q0+IsXU5enyY18nV6Awz86zk7IAd0/7aVocXg9Y5wQcriy7lFlXl+voJjx9/mRA6Wm8ltVZub58y9JZBWviCYJmlJZC9vvoy+/37xNjz6NENNzd7nHN0nSdE6/re7Tr6LhK7QNXKPM34IE3GTC+C3w0b3gSK00J0yj56gjge7SNPrgd2XaRzSucaTxX7abSdpp8sDsSzGJAorAYg0rKoCwfWtfG3jOeFLmB0AHlAnynLRIo2N8C6bvNysbYYHPiW8b2cYJcu61KKceBTMlvKasYKqK4UnVorp3GyZtNGFUhtYvShw4WOVoA16pMK6gIuNCpDO7clZDh/u9uQ3PDZ4bzj6mpg2HWEYGNjWR+t15Ks67z2i7pWAqRRSVcOeHO3qj7iYo90e6gF9QFxHg0D6kNrUry8ah8uKG3cy1mhgDOX12xuOx4/foRzwjhMBInM08yjR9cMg7lTijjSnCBnxtPIeDo1imGmpNTiA2vWEhHUVWrL2F66fFEKNbXFcS7kmJYvArBAVqJJhGlOUApeLEg2q/qKE8WrSWK+mte04TsRnzuQdU7odjtKruA7SrEVoFQrWQyHmeHqmqQO13odncB+2NF3HTEGci48v72nauXnfuEf8zM///NMc2ZMA+M8gHj6/pqh7/AernaOfV8R59bVmw+R/fVjht2elDK73YHTaWSxonXOkdLE4d4zzyemIRCkMk8jwXv6Pq4r467vcM5RasvUAo8fv8dXvvK95ot9Ech++OHX+eDrXyWnhPeycnGX7KqIZxjeo+8fE3zHo0df4urqsa1wo8M3wfmhD8To6TtPKZn7+wNKIc1XlNKzaGBuQ3LDm0BUiXViHwPvX0WG4HnyaM/3f+URV0NHmUfK6b41MmVmS2VYuS9ExDlUfGv1omUvCzbreZPRauogvilyhBDbhCYmayVLdQKrYrQgdJXJatlT52zcmOOWrnJ4Pniz8aydianPVn5MOVOORxbbzugDIQST7Gmd0eM4cTyeyKXw/Pk9z27vLLg19gHiPdeP3uPq0WMLVoNp4yoO9QHvlxk+NCcz7HcXFMMNGz4rYvB85bue8OTJNX3v8R7ELWoA9nNNvoAFsWrpUyeWiV21YBWqC5TQmwLBcAN5NAMPZxav1XfU0K3VT133wXkyacoEthAtOG+SjyYR5okxcH19xa/6Vd/HPM+Mx5Hn790xjzNdH7na7wjBM+XM8Xgkl8rd/ZHnT2+ZU26ZUdOULi0LvWSd26lR8kxJCbDEmPd2H7lfs9QLPd0W274lgALKDZnBezqEoBGH4FXpukpVNdWFDe8E3oJaYOV9xfgpIka2tnII+K7Dh44QZitvYhmY2JxDvHMrpyaXwv39PR8//Zhpzqg8RvGIU+gqwYP3QghKjLZ6DJ1NeCEGhl3Hbr8jJtNhhSVrG1fFgJJ7jDtYSf0eLzZQ9/uBGIIJQ+9M5SBn08erqrz//pf48pe/TN/3bWVrGVfTsj2RUrKM7Mr10xbIOrruMV18hPeR/X7Hft9bEB7sBuKc+cfHYMFtbVyjnMN54m+8RdlWlxveAILiqQRRhuDZ9YGrPnKz69gPHYnEOFkmZKEeKJyzqS0be2n9WKrRevSinVnErdnXB/qvq00RLNlYLgJZVaMBgI0Vo5a7BwHiYrJw3m7bmuqZWtDsagErMzbe7DiOHI8nUs7c3d3z7NktpVSKQlExs4O4I+yucR6cBFybLVUcTjwqpvxgarvyUhD7cjC7hbcbPhniHLtdT9/bYsktlQguqhEsjOyzCoc1bbL+/py9dajzVBdQH5HYnzO44sB34F6WjXuxoiAtI+sWZ761T1PWhunr6ytKGehihxaY+9msdjuz2J1LMSOEnJnn2R4p49FVPxpn5X4RoTbahKpSU6KkGTBTg7pkUtfmF9bvSgRisiRQ50CjwweHRxupwBulyrU7yjZnvjN4C/ktwUXj1EmxAp0uXM+q1lFYCrkUfCsxuubOYTI+nlxbp35OHE4n7g8HppTXTKtTK+dbZlWATKkZtJBPR1ROOB9bx/GJWirTOJNSbryj5Vou680ixo73338PVIld4Go/EGJYS6TihFKN46sKj24esdvviTE2tQUTc765eUT68ncZtcAtMl4tI1sLIo6+e0QXrxHn6buB2NmEXUoiz40znO0mEoNDtSPNAQWmMZFzK7OI38bkhjeDmFtXcBCjp4sWqOU0k6RyOh65u70lzxOnw4G72wMlZ/q9sHMDziviHeI7K7Er+DYBuRDBh1YmjKiLqLMJFefXgNigrfGxTZtL9zRY0Iiu1J922OtEfpEyWuXylm7qxXYypcQ0m/rIPI6MxyOlFA6HE/f3B6v2PL/n/u7eGr0qpIo1r1w/Zl/Umtuwc1kCWZt6jY14NkE4s2HPQcHGjd3w5nACQ+/poyN6CKIEaTz2lc9Ks1rmLBNp+nTth9FsUEh4iu8RHLV/ZJqqTf9VxWgyqbuiuI7qm8nHEhi2Cp9rNs+hSdl1XWc/e2tEizGuvNZaClSYdpPREbwjNCm+fetTKW0b3gkpJZxWvBZEFe8t24o4VFwzbbG5sGSjEPgQEbfoVDdOLfZd+UbFCK2S6Z0weEfnHOSZcnrOcR5JosxA3mQL3il8/kDWOeLQQ8rkWtFcKEnJqazBacrmwlMcBAGvQhW7YH0MjKcT98cT0zzx9PlzPnz6MXPK7Haw2zli7IH38L7De0Gp5GqKAqfxyDyPgMPFjxDf4cQRg1EFzDFFCU5BMqBohd1uz5e/9ITd0Ft55GogRE+palJhVVenS4C+H9jtr/DOo2qZWFUl9h2PHz9eOzzP3tSlBbJC3+3pugFojidNluz2buRwOrUyqx1X8I7DoafvAtOU+O7veZ+bR3tC8K386l7xV9iw4SEEiA664Nh3kauhIzohTSdIwvPnT/nwg68xnkZOx5HbuwOlVG4eC4/9FSF6ul2kD1cW0CIEWRosPS7YYo/YoaEFsj6iLqyZJGHhd9eLAHZxKDpXGEQuA9/l6B8+X2T+Us7rIlFV8eKIPuCd4+72lmdPn5JT5nA4cn93JOfM/f3I7f2JUitTVqaihK6jv3mf6/eXxtGICz20wLUujTALN3bttdH1eBboFsdueEN4L9xcRa53gSEKvVezhdaCVMG1zntZrJtLXRdOvm1jpehopUqkhGvwFR92+N17iLYg1VKgZh7gB1TMXto1ObzlPc7ZPNn3Hc4JOe/x3nO137Pf7djtBrQqfbOm7UJnyaJ+PjebCVyFQNd1iHMc7u/5ypM9OSWkzLg8WpXI9/gwIOKM8uB7VISiuVGXWBVPnBOGGOiDtwUAmSjVcq7maEtFGDWS8IzHez7+auGYEwVlVtOtr9tC853BW1nUindIbVeWLGLL5r9sunLGVaFCdUuzx9nmUhUrRyQr5Y/TbJZ3caaviVqto3kpWy4TjQWdM+N4smzJnFExysJ+t6OLXVMtKdBUBhZzg+ADV/srbm6uLCN7bYFsSoVxShSro9pkjWVwY2ydlNpEl1s2KYbYJrYLJ6Jmi2s+7gN9N1BVOY0jTO3mpJmcm4tKNn/o4G3Q1lwY9z05FUqx7tCHma4NGz4ZIialdTYtgJIL4pQ0TZxOI+PpyPE0cTxNlKqEXWaXFXXg1VFdwIm/0EW2cbuUOWxy9G1WbK8hCIUHFueXWpZLFrMR9V53ScsLH6967qwuxZiCi11tdY5pmjkdT8zzzPFw4ng4knOx8xwnSq2MWZlyJRZlTpmi4LRlXcUKoItu7EMqwfL/8y+0Zb6WzvINGz4NIkIXjVYWXAvIpOX9LxZ8QFMVOCsI6MV1t5r0iCO7YLQd5yl0lmVFVwtaJKLim5X6he5r28llU1etlRA8tYam59r0bastGlXtHmKVSau2mJqQ0nWRq6udqTFQkTKRc0KywyU7rxCWQNZTY29Btkgzx14CWePXe+fYd4EhBrwoA5megoiucp0Zx22NnKo3a/sQmcXoilmUitsKJu8Q3tI+yobG0vF8tmutKwGoVuuOzsmavU7TxDjNlOoZ55lpnpnnRK2CDz1BPc7ZYakqKU2cTvfWVOJnvE+UmjmdCtNogWFlRsk4J5Q5ExrndZ5muq43Hcm5SXWomoC7D5QKd3cjCMzzzOFwIueyamaa+1hH1w848VQ1rT1zASvU5vxzWQ6tanaZzjtubkCvAqrKeJw5jSPznHj+/I5nT5+hyiIYYtxhMQGROVVSLpRc7bvc5Lc2vDGEimMucDjNlFoJVE4u41Du7k7cnRLTVJmykIhUgVk9xywEgTxVkk+IqxYQt47msJQTnaMvEIvindlKlxoQwLfGTtoELbos3s6ZzXP1gkYVuBRif1jCV7WgtTSB9UXKS9WkuESEeZ44nU6r3FbKiZIruWlimhynNZDFrsfHDudbk5fzlDZ+K3rOyF62knNhSKKL3z3bRLnhjSECPjpC9IToidHjgzX6OucpxSpviz1sbfQCM+lpnHK1/gtQSlULWcV+a3NJ8wxYL86m4axqGd+SVqWCharggykT+CaH1XUdV9dX7HZ7hmGwY1erT8ToEVFTDqmFnBOqlWHYcXN9TQie/dBxvetsbhwPcLqDWnGxx8cBxJFCTwoDFZjSzJSmVslZuLHS+L2OKkqpQmrnVxqDP6uSilKqQ1MiBs8w9FSU6Kq9b6tiftvje74HPvjg1b/77u+Gr33tzbbzBfigSltFNc04LziVldRdW4PGPE+gMOx67k8nYvYcTieOp5FpnqkqxHCFSCE4W11SK9N4RHhqQbJv+tC1MI6ZebIyY66WVRJZSg8mDWQD1CwA9/trutBZt7QP+NgzjRO3t/fMKXE6nXj+/JZ5TmsTmRNH7Ab6fm83m5zJxQZvLcVEp5fsjElDt2C5Erznu79bcGL7vLs7cTgcmMaRb3ztQ77x4TcAoYsdIUTLIuOBwNTObZ4tqK5bHLvhDWHtiJ4xK0/vT8STIDUjZYRaGI9H7u7npvcsJKzE52oHs+ArBC2EMuPEEZzxbQXWpkYnjr7PbXw5robCnMxUoXeV4Cwr5EXxC6FAxR5Lh4uwLgiXoHQJZvUc9VKbdmxKqY3rxQraJlSA4/HI3f098zSRZtOjro0qZCYI4GKkjz2xH4j9DtcNSIjW/d2a2+oDDc1GLrhIlKELteAhW3bDhk+DiBCHQDcE+j7SN2varg84F5qk3DmQLWpFzkJF6nks1FpZ1lE257Sfctkwtu7VnitISXYfaMkmG0OFECK7/Z5aK33fU2vlyZPHPHp0w/X1NU6EEMx9L6XE1dWw2s6eRtNiv9rvefz4kSmX1ILkgmgl3z2jPP8QzRnpelzfgziOvuPkOrLC7f09ei8PmkFdU0hRbzrSWQtaZVn5IrWYRW6yYFbzzBAD7vqKKlCCoi0e2fDtjdcFsZ/2uxfx+QPZixv8kiWx5t/FZ729rWVlczYt2dz4biIYZ7SUppkq1tSlrSO67aSWQk6TZUmrGSCYMHqllCVQPgswF9eGuHOtWSrR9x1dN5gL0JJDFUetMM+ZcZw5nWaOx5F5nvE+ELvYtiHUsqyabQCrVrTmRiNYvoCFWmAllxAC85xJyVbQKRXSnJnnzDRNjKdxpVfUaqvR5bsodRGq1pXcv3zHGzZ8MqzjvirMuVALUBOaEtTMNCfmbC5XVYWCdTlnHKmKWbkWpSTTZCzOHk4EXyu+OkSqTTRS8LUSvTn1eAdezVZyaTpbeKS6PuF8GTfagF5Kcy29YVyUUhfR8xcMCuqiiFBqy9ouTWEvFDGkuYwtFpguNNUFM32o693mTCGQ5fmruLFL+qtRpTZs+FQIrcn54uFcaxRuWq4XXBu9oK0s4v7LfHA2GbnY+FI5eKF5UrVRePTic42brnpu+DINaQsoY4yEZhHtnFj2uFVlSunx2QyISs2UYuoFfd+aw1Tx1ZrXcp5I44D6hHQd0neoOIqLZBctU+zPkgQrnUJYqyOoUlTRomsgS2t6K7lQi6KlNMmwYI6B0ShSGx3v3cHbZWTrwn07O5aEEEDAh7DK85Rixge1FG7v7tk17dRpNl5szhlVIcYe5yre+7XsmOZj4/c5QuzxIbJMLj4EpOoq52GwgFbVgXrAU4swjbb/EDwfffQx85xIqTJOhVLMD9r7vpV6LiY5FVJuXaOlkJuYuy5BJqw3BsPCl3WM48zz5/eAcjicGE8z05QpVdpkem56WTJUS7PYKgRft+zPhs+GipCrMmUl2wtm7aieTAfxCucLLGNEBEJPxtm1mRWP8by9VOsYpk16bcKdihJTNp3mOXHqPMEJ171nHy2o3TXNxzZAWmDYmLJtfJdsTnilqYSAUKoyzZnTlBjnzJQqc1Z67+m6nZVDQ6CPJv/jxcwTTJKrMierjNwdEnfHiVpBYo/EnhA7rm4eEfse16gFl3gQ/168ZlllPfNi9fIT24S54ZMhzhNurnD7AY2B6hxJlToloHAazW415YLWZIoF2gwLxD0MylbO+jJ32Pha3rH+lDMrdvm4PPzfKp9XSiG3huTj8cTHT5+SSyFGz37XEYI3PeaUKW3xuJbuayLfPwMn3I8zx/uRkjPz8cDp9rlVXVxTBRE4quNYhazK/eHI/eG49tVorQhiGtHeIyihPUxasOnqKkgLmCUIcefp9zvwgnagXnDh4dje8J2Lzx3IqoI2r0dpGU7nPaGLSG2GBY1nmkvhcDiScsLs9GazqGsRnAWEjqHft+ys2e7VWkjzHaU8x4mn393QDVdnknrscLVak4nkxqcD1WrcUw2ggZKt/GhZ4Ixzgd3uOc51hLDHuYBqJMYrvK8PupFrFebZmsaqFhuUtCxRSx858dbvgqzORIpwf5hI6SmqyjzNpNRUHAo43zUq0EJKX4LY3B4mZbbIDW3Y8CZQoKizzudsHFVQa8pUAemh71ox3eEWHp2PJPVoFZPxKVY1Ma3Zto2VwiZ4P+NaV/EuOPpgEnJfuh6o+65lae3neeJ0axC48P1SytSSKLU0Co1QinIaE/eHidOUOM6VlCrd4Bn21/Rdz27ouNnvCd7x5NETvvLlL9vCN0Mq1hD69PbI0+cnclXUBdQHnA9cPfkS/bADcRRxtvTVh0HsBSX2XH16MdO7YcMbQoKjf/8RobtCu47sAzUpZZqoVTlNM6dxMmv3OTGPozVgOU9sjlg+WCOW0ex8c6EzKazVTW/5T9p8dJGhtScX5AORZoTim4Z5toSTKl/96gc8e/acvu94/GhP35t0lkiwfYq2RjFB8kQ6PiWXzIcf3/GLX/2IcZo5zpnb0YLy3OY+rZVTyozJrGyneWac53MDWTWTa2nHaU6Y0XpInNCHQAyOKPDIKzun9Nc7vvz4K1y9f4MEQQaPRCHELZB9V/AWGdlzB+WCxUtaWyZ2yTgqmP9yqaScmaaZ7F3znA7rNpzzqEorn2Blw5opOVHFE8pgQu5uCRg5r07FAWZEsHYVt+Yp1aXb2Ry7xnFsGWAF+jUDbNaUrF2jyzlWXcqVS2B5lkGxAzczCEQRlfU+UXJmap9NzayhlLbSbt3kIufvadnfUgZaLGqtkikXd6MNGz4B2oolrWIgil2XgIhnSUJaJ7NbDQAqcu7Ob448tTVvnYvvNrSs679Z22pFK9TqSdmqFk7k4t6w1jNX+kAbXWt1YzlWa+SCUtSayEqj2CgoDucjIUZi7On7vql9WKdzrU0vNltGeq6eMTuj6oiJx4uzxbZrFZfyCYHpi93i59cv3rSVLze8CUSQGJHg1+pERcm1UnNtlLLaLJYT8zRZMqNxRZ3Y4hS1ANI0zYNR7ZpSgSVSaGOvzY8PDqGNbWRV5zhf20pplZGcE9M0ATbn7YaAiDZpLJp6ieDDss1KTTPkmfl04HB3x/E0cciV57NR5dI8rec0TYlxNtWeOSfmnFcK0TqntnuRc46+74mdSYTtYiAGT++gj5he9mDfQ4geCQ7XeyRYLLLh3cBbUQusHHgx0SzBIKxarl3XMfQ9+92eFMzONbWOYu8hqKycNt9KFVpZRdBLSeQ8gTjcbLw8EbPF9D5gE6RvJUxvmV4WjbwB7wK15qZDWUhp5Pb2Y+PBxoFhOBFCZ/tdGkzUsqGKnVup56B2mcyt/F8Qga7r6bveCPQrs045lkRtPtLG77Nji13kJjziLPy8OBn5tcxaiikklCZBtmHDZ4ElQZfGqqWP2ZZn55jxnCXVtuBcS+UXc8DKb5VVQKstOJfpRqnF1AHmouQKTo3isDoNsfxUTCzz4fFq26YiFBWyOrI6CgF1PfiAj3uG3TW73cDVfmcSesFbtmdnnc+pCHOxQDa7gVkGcqmkquT2XfgQLs71zEEUveDCvjL1+qLc1rlxbcOGT4asnfhn5zzWuTOlzHg6MafE/fNbnn7jQ9KcCM4RWrLDrNQtaLOMrAXFPhr3W8R4osE/fG6GO+f3mPGPY55mnj99yv39PSllDvcHpnlm6Ae6EEyXFmWeuhYUJ0RHAKMQDp1xXOeEn2ckz5TTkfnuOdNpZCqOuTqyilEScmnNaorHIU7pQiQ0MxW7N1x8X63CGUM0DWsRYjAKk5MmNdZoT9pkP0XBaasybXhn8BbUAlMjUIx/fVaJbCoGPrQgtiPvdlxf35DmRCkz0zwBlRB0LSeqWle0iJDzEsgWUp6Y08nmZMmkcmrE7s7UB1yg76/xYdfI8xHvQpPb6fE+kNJIORwpJTNNiePxFtVKjAO73SNCMPmR2AZ6KdaQplXXVfKS5V2s/GotlGZ8cH11TZBr05qteQ1yD4c7jsd7RITd7oZh2ON9YLd/RL9QJMQ1x5UKOlNyNreTksglUap/kA3asOHTIC880zYm9UH4xvq783v14uflBhf5c1l/oypQ7N2lFOZmVjLmylwtVq1I46DKOZjlwvn9kv6tQl0fjlw9uQYr+zsFKqG/Yn/9hKv9jkdXO957ckMMnlpykwJS5uqYirNMbjyQwh2pFKZkjW4K4I1nr/owiD1/KZfZV70458tfbRPlhs8AI5mDO1c+qpo6Qa2mbXw8HBmnkQ8/+Dq/8I9+lvF0wiH4xnQ198lGIfAeCU0isouElrHsup4u2nzW91a18N6z3+8YhmG1nfXecxpHPv7G13n67BklF6ZxIqXMvN8zdJ0pz2pl2nW2CK4FnWeohTgMwCOjOkwzOo6QJvL9PeOzjzgdR050nGQg49BFrrJlgL3YeXUxQOP6LlXche+7PpemPY8poTgWty+WJYFVXWvGqUfVb6PzHcNbZWSXyWhZXbb8xtoxuejLWoDpUV+tCYvzpFCbbNY5CNaVoH7eS23vzUgVVBfy+7Iaa57ty6rVeePrtlWoql8DUNVCznPTgxVinLC9B7yniT+XpkpgZgVL5tnctRZerw3MpbSy8AeN7mDHXEsizaZO0MWeUrrmYS0mVSLGUrQO0kLJpo+56gde0AvW7M+GDZ8CuWx+lAtSgLYg9aW7/Ktu+/rCJXcRxD4IaM/hcmmNl5WFOnDO5K4/HpBP1z2dny/3k/XhwNlU7pxxXL0P+GDyRSEEamtAs+qQo4iAAx9nfIxUcbgK4ut6PC9mZM/EhlcMM32YidUH39c2ZW54QyzjSV68fpoVcynmYpcS02jKNg4z7rBAVtaeCvHeHk4IcyR0s6ns9JnSWROmlooW04r17qz3rrXiQ2CeJqZpZBpHW4xOMzkXYgjklMgpUYJrUpMFckZLglrQEtBSUOfPTVpVoRZqztRGB6zStN5foOcs3F2jCbpzsOoWquD53iEXP893M2mcqeU+8dBAQvXF6smG72R8/owstBLJuVFCVVZx5lKUlBLzPFNrwXtAHSEM7PoOVEklt8ynyW+JE1x1rezfmUpAiPh8ls0yGZ5KSpYRDqEyDHV1KDEZEJMCub55xDAMHE/31HoAZqapMM8j8zQhe/ByRRdhv4+89+QxXdcxzzOn04lcinVBz4Wq2nypY5MhyU2QHd578pj333uv/R6CN6/pn/v5xPH41IJwnSllwnvoe8/NjYlN52w0hpIrpcykeWSahmZp2yTGTLlsw4Y3hix815du55evvSaAfWEp+fC3L0zCaxTYVnKcHw+k+NrH6nnawRapjUOvRktIWckqSOjxXSWGigbL5PS7PbEpD3gfzJfd+RZoVipKypXTXMjZGmjGOVkDTSltMsV4D23fdb13veKr+lRsQeyGN8NC4nGiOKmIKN4LfReoXgm+NWiWAiKErqOrBYoiTcHApCbNBYuqSKkgMI4TiyRcFzu6rtm1d3F1pez7ni5GC3yDJXrmeebjj41aoItFe6kcDwdqrex2ex49umIXKrHu8SiDFjxqurSp8XhTWpWDFGs+897jsWqj4nAKKmejo/Mt6Oyiucb3a9Br96oKy0p8bV7FCzk4cvB4L0xqpg+uKFNSnLqtUfodwltkZM1ByIp+2gZak3lTmgxOIqWZWsoqpN6Fjr4zTurt/T3T7XNyKdaEIh51EINxX6vP5DKR89TI7VbSB6hqk1unBVXbfoyeYdfR91ZG+fJ3vc/19TV3tx3j6WNyOZHzxDyfOB7vCUH4/7P359GXZFteH/bZ55yIuPf+hsysyqpX9V6/7tcDgywQ0AiE3WKQZFkYARqghQck95JlW8MSyFaLBWYQRpjBYBaTZQm0mASSsDDCLYFlMfVqwELMCBAgWND9TPebasjM3+93742Ic872H/uciLi/zKzKqqx6Va8yvu/dujfvEDfi/uLE2Wfv7/5+vU90rXJ50fLp1++x2+04HA88uroixkjfm/lCzlpcvjqcOMZxYBwHRIRX7t/j1VdfoQmB3a5lt2s59kdu9m/xxS/ZxSfngRg9IQjdxnPnzg7NaoYQfSJrLi5mB/p+wzCUQDZmUjL1hBUrPhi827n0OPHz8RhvpiRoaS7RGsxOJUFKw0UNjK2Wn7WszBaTX1IYotKPmTE7XNMROsGp4ssqbrM9o2lLIBtaxAXjHWYli42hISb2x4ExJg5HcxGMxS47VXJw1hLMLjPA7473FeuuWFEwcTrLzQchFKWQEJzR1WJEgLazOTLHRB4jFKm6WKqDJAsAsyrHw5HD8QAKoZmVDWp1ErH+E2dSBlOzdE6Jw/HIOIwl2WLjoWla9jc3dF3H/Zfu8PLOs5VI8MK2cXReSNkxDkeyq6ojZdlcv9c7PA4/LWw9iJ8CUl3c3653TInb2hhaFYLUKqq2k45x2xKDx3nHsVAfJCYYRkiQ1uzPC4PnpBZIPS8nlYEpQ1tOwFScrrS49ohYI5iVCt3MgRFXFAcoagb2Ha6UGyiSQlB4a6UEX9UDls+BFmkrsQDaz6WKSTpr6pAs+nROaNpA2zWkPNI21qmZM8SoJZANtqqdFAZyoQkE2ibQNE1xbWlRMj6U5pby3pxNwovyvUaZSAtebZ7uqyLELPuzBrIr3h2PnyXyTi++A25TfOZsrD7zhk4pBbL8ZFUEwM5tLRmnVKkJUigFWpx+sOuCFD/2KWCupVksmM5qnN2Y5mvPbfvb0wcssj9P+yWe7ThXrHgn1CI4mlEyooUmICcnI1VSy3kHWU2D1Rir05xgAgQCZQ6MY5w3kc0GOrk0OeEZ59Ret0DWdJwrncDK/7aXiEyuecMwMJZbE8ToAupI6hhlJHtFcmRMGZeVtKAYPk7Rme+muIF6KagvzgvM+tQ0fvXk0+Ac4h3qXDFQKO/Jlp1eW0teHDxHs5eQcmPZkGTZlaxKTKmQ1xN98T4/HvZcXz8i58zlxQVt2xRVg8Bue2Z0Adcg0liZcXLqiWQdGOORnE2wOSVTCpDC99Oc2R+ujUDuA8d+T9t0bDYbsg7c3Jyzv7ni0dVD9vs9fdHJq6M6l2DbnjIHL+cafGjKZSeDNLYIbDd03daa2frjVBIJTVO0/IoQ/TDSF8OFmCguSgNjcSD7yle+VH6nzPFobmI5KXEcyTERx4jF7EZcz8VxacWKrx7ea3BWV1ynW5gm0Zp7LRrRlAAzKySEUYXjqByGbEYO6orr2FxiVKnNYNZIlsWh4kkkYhZShn60jOwQE8MYC293QX8qRMUa1p7wfZ/vB1ux4h2gQCSlnuNwTYpHyA6JATLEdCR46NpA1zZ0XYcgJBcZkUlfVXFlrvOTpW2Mib4fCu1u7uOAmlRSci59HsxrQOPl1gzo6WfGmFAij673/P++8AYPr/Z0TrgIjsYJOTTEbod6k8I6D+BRHjzaczUqfRaOOdOngVQOvxJ6NM+c2fqc/X+W3zqhH6gZIDjv2G46ural2wTO755zcblBGgfBMWTLLKujJNU+HiP65//8n//U137Vr/pVX8U9+eTiuagFKbWW3SzdiCmbt7n5oGeOxbnr6uaGtx+8RYwjzsHl5QXOmzzI2e7MgkjXgmupPNgqvTWMe47HK1KWYtFXg04tGZjEYX9F3x+sI/NmQ/ANm82GmI6cX59xPB559PAB+/2eYejLgC9NW2rKBBlFxOGcNZOEYALQzkMINtm17Za221HNH+owCU2LOCPep5Q5DplDPzKMyfQwowIDysAwDvAV4er6ClWl73v7XcTRNhuCa0wqTCcfE1IWUnruv/WKFR8OpklqQX0rHNmq5SilJCgmOGsd0Drb5I5ZOMTMfkgcozKqkPCIFIVbUcxOFpJK6fx2ZHFkHFE5DWTHSD+O0/Uiz6F0yQJZgPzxmOpWPBdee+3pxuyf+hR88Ytf3f15Aiwba+fjcUgMODMpiQFUGOOB4IWu8XMgK47RjVaxyNlK82Ka502w6mDNqu79AZF5XqNUKAC0WMOnZK6Xc8hqeyalYap0Khc3y0TK8PDqgOobZkogsBEIAhoa2GxQ79l1DXfOOhrvON4c2I9KysIxZvqxtyyt6kTtSYt5vI5Fk7osTWVTGdL20KS2oG0bzs86ttuOza7l/N4ll/fOSJI4+IExjWZvK6BSMrQrXgg8V7NXzXTkUuIwZYHZE7oGnjWbWm8pRZIr3kL13gecb7AAuejNkQu1wCGTUsGtvVDjzdpA8AgDmq3k3/dHQnD0fV8ctep2KYNXS8CciDGZ9MgwMsY46diqFr4fUqouRSqkdIDW0WbbsUFKnEs21Ru7dlWKWFeqd5YZHseeGCPeeYJrwZ2uRq3sWsXl1xLminfGB3vpftbzTU/unvh6pR5NY6FcO5ipSDlDKhOeFu7tcsK1yuJMb6hTcp4+a9eeKi6fc834PPl45kzsOx9nVUR4+uGvE+ZHiqcFse/22lcZHjHjAq0Ug7ke4AS8CMk5gnemxpEVTZnobQHnvOKSce4sI+vnbv9lpz+cnK81Kzs/txgRi8C3mqPU57QkeoYxWt+LKCrgARobo+I8DmXTOHLwJFVcCCAOyRHVhYxlmQtnbfaSoZ32caYALUeklmpvLk1vMw2/2vS6xzK+6vigL4YrPsZ4/4FsVvo+FY6OTiuucSwZ2TETx0yMmXFMjNEav67317zx5ps0TcNmc852e4n3gc32gu3uEhAOhz37wx7EGqxCsNWp0Q2stBJTngZnziNgJ3aKIyKOYTiQUs+jh11p2rKAMecMovjgyZq5vr6m74eiDiBFa0+QomnrnIlQO3GFg2NyYia9ZSOl7488eGSDbL+/4ebGMr9vvPGWWXDmNHdt50zfH0gpguoUXGto0NayTognZUeMjhiFYYBitLJixTPio72K14VbLllYzdG0JHV28SprPvoEfTIN2sOQzWJWBRVPZuaxZzV/9ljoAlksIztm2A/RRN2HyGEwNZQxzdQBPclUzZnZ94J3DWg/RHwNJB1XvAOCOO6154hKkdMSnAY8DeAIozC2I0EFdmeQhRjN2t3Jtc0TSUmdJWJCkZfMKdFcN8AcvFrCRE7ul4mQamFbs7FQJDKbxiS9xOGDud+JwDhaoseh9FK0ioYBjgPiHGnoaMi0TeB8t+VTr7yC944vfOlNbr7/i4xxYBgGjv2wME+yiogUnquIsNl0tN3W5HbF4ZyQU+Z42BeJMGEYew7HAC5zPPZ0xxaVDD5TWmmQDBMracULgfcdyGaFvi9NS0VRICVlGM3lZxhNCsduloUc48j1zY19cQi8dM9zfn6Ppu04P7/kzt37iAgPHphvMyihaQlNawT4OOLcOK3MJp7NibWdQXDsb66m1aVIzagaLcAXf+mrmxucHNjvjzy6uiH4hu1uy+WdC0LT0LYNm02H99aIZplhD+TidmL0gGNvWnxf+fIbvPGVNxnHkeubR+YvrYorqg05J4Y+MfSHaU8rJq1aceQspCTEKIzjGsiueG/4OFzD57GpaIqQYpnELIjMwJiVIVsgexgzx9GMFWzaLAFvscjNhT6QcqUXeLJ4hgT73jixh37kMIxWGi1ygFNTmLjFvsGt9jPsmfd0hO/5E+8XXyNJxxVPgRfPvebMMoqlTOBdILgOJx43KMf2iM9CkEAXNqScaXxDirnMIzK58oVijpBiJDThsWzslNlcLL6m4LW2cS4Wd048IbSEplkEspY8MjpeKuMxTbx1KTnmPI60Xti0DS/du8dnvv6zbDYdQ4a/9wNfIMeB4Xjg5vrGzIUoVZDiPuabYNKV51vOz3d47wjBXDrNEWzgeDCqwzD01kTtM8d+oDsOlvtpTEseAS0eLB+Li+CKrwqeS7Ugl5pg5bnMVAMAEzAP3m4+NGTN5nHOIrgs3ZjOySQVMg/KyrMz+1ZXlA2kCCHPmFei8zMJTTrRAqpCgrng2hmvqib2TMaJDRrNsNFuakZrW1MhmCx3y0BnQUy/3RkdUyKmuFBUKKWTas2pOq2WZ63NWbmhHr/RNObbOjI/BvgaSI19UE0O77aVp4ZxE8WtqD/X64TqRJOxey3ndp7NFJZNGmLvlcVm63tSKVdKEmLKZcwVjn4db7aReVvLfy+Orwaz89w38XqecuC6CIC/esHsiq9diEDrjN9ao1EvC/vZQjtwmIJOCB7JjrCkD1ANdExOyxe3K+eWCzSdz1u53bq/IOosDAeqUkFVD6qa7DZXG40BbE61vu5y9tvB2GI1zXPg0olrrprmiWY4LSBlNgCComjkaizgipmSTI1pVokt1LzBMfYjYz/igtAGTyhmEWV6Pw0RVnyi8VzUgmGIC7K2PVc1T9t2x0svvcpms2O73dG0DeM4FJJ6i3OOXaEVuMJxMWkqKZzVWPRlA027xaVYmsryVI7POU4c3cLYedKe2iptEexas5ftr7mS2CDauh0+OO7cveQbv/lznF+cs912nJ+fmYi0BLwzHu84mNlDHYQKjONIjCNXV484HuF4VGK09ziV6YIzaWuKGSx4byYObbsptroNKcE4JLxPDEMmhK+ebMHaZfkOWFNjJ7gd6lWZuZQhJ9PFRC2NanqQQlJPUmFMmcMQ2Q/JmiNjZkwzmaCkjaamlZiVfow26e4PSFn8Prq+5q1HxS/+OFjjl7hyRVgGsKdtLjBzbaUG3U87ME4XymsAu+K9wKgFZydzDlAaTRIhjbjU49JA4xuabUdG2O/DpPzRdi3dxqzYQ+HSjuPIm290uAWfvG66DJ8SdM77MnlkiWVixRW5rxI4hiZwdr6jbbti0mCGB/1w5Or6EWMcbD6TKSVEjonkHIf9kbfefki3P/Lo6pp+GKw/JadpJ6pFtWWmHaIJUSEINN4CWacKKZk7mcmUEFPk0fCIRwJd15LHyPXbV1xcnPENn/sML53fNVpCsIRV65sP/e+64uOB55DfMomtKdsIMGUfhbbdcu/eK+x252w2W0LT2EpKyokq5gvtXSgZTqYMZ6pSWzEhztO2W2sQiyM5J5wzb3WXxiLdBZKmPpDHJiFdTGW13IkUj/jBtmElFfBBuLxzwTd87rPce+keZ2db7ty9IIQAWdBsWaLD/sDNzWFSQBARhnHg4cMHfOlLXwTJuEeQkvGCvArqy1LReXMkQgmhoW03hNDRtBuaZoP3LTkL45jwoQayazZ2xXvAc54u7xamPTmkWzZ4SuG0J6RkbSj8uITdYobjENkfZzpAnGbhRcZIjWaQsjIUZyMtaiMiwtX1nrcf3RBTZIyZmKXsU9GpZs66Vn7snH6dm8cm1qzOR3jaJMPTs7QrVrwDvHjutTvTfI1msJNTZtSIpkxIEZcGJPeExhM2pprTtcFkHkVo25bzi3Pjx3pH4z3jOLDdzIHsSUYWuD2KZUpVlvxuaaY22UnLfjbBc7bbsd1ty6dsXr++uaIfDsQ04kQISOH8Yu6UY+KwP/LgwSOapuHq+mZqns5pzrwCUyW3COkhZLzTKZDVXLj0ORWVE+sn2d/sGfqBtm3Ix8jN+TXplZf5Id/wTbx6dr8YIzm8lzWQfYHwfBa1y7J5ebbOP2YX2yECMY5sNmeEMIDqRMJ27vbXLyePuYRRyx3Oe5wLKExCz06KLe7JJFQnpxrA2qylT5idF0ZEJ52QtbxiJRZX1BXKtrOW95Q6Rv3cSedoNTwoZggqkCtnyZXfoNAJnC/HU2kFYqXZxe/wUTWZrPgaxFf9VLmV5ZzoM2UM6rxL5SUyTNJ3sSgMZD3d9duZJEvSZKII4hJujDiRSWUkLbZRg9U6HqUGsHXDKovdnkuwp0nXBVXpscf1GnNrJ1eseApuZ/trUmWmpRX75ippiSOmaI5e5RzXnMnOlVNTF3PvU75zyZF90tL01txXDYjmdaSUpi/js7pKaVDmps2cJ+vcfhi42R9ompHjcSjV0+X8dYu8OlGOlFSUg5yYbrqq2dDnNFcjfWlya0IgNIEmmA1vE+xmsp4WyD6ucrTik4r3z5HVTEqlYanYUzpXOT0OOd/RtK+Rc+Lm5prz8zuM40jfHzke9pO8Vm3lMD8fowpoNuMDzbmU31s0B3SjBB/MKKFozpl2bSQXodV5iJzygUwiz1ahIkUyzxthXdXRNmYf6wMIVk4Zx56+Fw57Uy7IKZOTDbqhjwylkauO0WHoGceBlEa7xZEYi25tKpadziHtzhzHnCeEjrbdEXwzd4q6opc5SZnZ/YoVH2coMi/cslhwmdLU3AJCUmFIypgzxzGx70duDj1jzKfe6MuIstRIY7HUHGoJtFB1jmPkONqEmTG1A/vc7WC1PlUztovk68THXfBjF/z/pTPYGryueK9Q7JxPKTGmaAuyGM3IICX6OJDJqMAwHOn3AzErb7/xJg/efJNhtMytE2dVvCYwtoE4RsZhtIxn1tNhU+kGNaB77JTVk/c2TaBpG4L3RTJzJIRA121pmkDOid1mC6qkMTIeeusxiZFxMLv268OBNx88xDnH9c0NVzcHUjEyWionlG812bwxoyny4O1H9McBgUndxHixEcg0jefi/B5tYzq7r7zyMpcXF9y9d4eXXr3P2Z1Loxv56l7mWPFi4DkyskrORyq72thooWQwhabdcn55AQhn+0s220viOHJ1dcXDByZLldKRcTyUjE3GCo6KaiorVTMpCL5BvSIOmtAQ48AwHIhjj8TRTvycnhrEooI6KaXG2d3ES/GfRgmNx5eVHKKkFInjyOAdx2PAexuMKZqQe0pKTFU7NxfOsHFkcx7Nf7oEszZBG1fXgtcNNRtr/NgN3jd435ixgjjb50UQuzp7rfh4Y54sl7qPKeUFO1XMxSsrQzLzguMwcujHqdnrsYarEnIKlr099sM0umvQGTOMWhtIAHETP3C5f3PvS3k8bf5WWmtRAZ1KTjypMvJuBIwVK2YkVWLOjCmRUmaMkePQG8Utjla9EGUYBh5d9Qxj4uGDt3n08IFlKp2jbTtC05DalpxaYoxF4rHauOpifCzvBRVd8O9O4ZwQSoOzOYblEsj6otyzYRxHNt3GAu+sHAsFcFDlkG3Q5EdXVBWj5SCcDUjKeC4DfdKYJZPGa26uTNUolxjAOUfbtTSNNY3fu3eHy8sLNpsNn3rtVe7cueTi4pw7L91le35mwbsvC9Y1kH1h8FyqBfWa7h4rz592DFrguCy9L0roCyw9p6twenmhxIG1/D6X4U+2Ow2W+rGZ1F53qJZPqkZeLT80ocE769TUrAz9wPHYo5nCJfJT5yWqk22swkTgTzFNDiopJ25Lgp0e62LfnjAZ1rm0Tp4rtWDFVxPvPTzTk/tTldZC7UHKwq+qDhRObVUrqEHmoilk2noZ31mtsxvmcr9VdGqWdcl5vb0Lj2do52D0dJvLQ6p0iOnpdSiueI+wpmQm8475vC/9XsyqAdV0x4niBRrvIDscSk6RJDAKgJX0Y4xU3Wb7rlMd2bIH8/lemqNP2/pnSh0i5JSIKDH6Eixbf8pEP1io61SzIBCcKIpfbNbeUxV86v5NS9bFWMpgCifMxgmIFlqDqQbtdtspkD0727HbbdlsNqascFKBWfEi4TmavSBGwXuHbywL671ZujqxFVUcUmkK60lxmNQGVAQVZxTvaTIxHTgbYNlWZGp0gTpgrOvfTzay3geWvtPTOIXpokC5QDjnLWccgq1qg32+6zq887Rtx3ZzRgiBoc984Qfe5O03b4yP03YTT9aXjGkIjVnTSrXUTYzDwPXVDfvra4790bT/oMSrbjoGXztEixB0DbTrhJ5vTfA1+7tixVcDz8Itq6FfDSGLDxCV51ffUycyLf3NiUw/mlLBcRgZoslm5ZopqgHncoarkyE20eW6beZtn37rzF9d1Gbmxex0jShft8y+Tp+sjxaBALcNEdZZc8WzQYEhQoxmAJISxCREdSRVVAI+dDR4ugi7JjIK3NltGO6eM44RnBAPV4wIN9mMQVKMXD18WEr3t927hLSYN5YmBMsxYnNUnRtba6o67Ikx0XUHUkq0nWV/xQndZkPO2eh2peHSl8A2NLYNcc6C4TLnD4OZIpjUXkbTsoJa9qa4aCrzwlYwo4SLizMuLi745m/5Jr7+6z9L0wbOz8/ouo6maei2nakeAZLq8X54f88VHy88V0Y2ZTGhf+8K59NhCzpFY/F3zkoch4nXaqsym2xslVqmIAFxteyfUU1kzWihLZRagclruISIL52Wpk3rvWeSAVMtK0Y/ZWunQNaH4hZmAe352TlN0xYSuQWscci8+ZUHJg3mPN43OJEikdXhvGO72bLdnZnLV06kbDyhw/7A8Xjk2B+M21PLK4WvVLPJzrmii7uYeJUiUTSvSPMic7Xi4493ki6Dj7982XtpkJjfOdUPHitrMj1rt6wwpEQ/RvoxWpk1V0EeN3/4Cae7Yja0NddaA+WakbVJkIkLq1Ml9ZaL160A9rGvm2d3lt0068T4tYuPWlJQFcYMMQtjqskJiCpkdeiUoBEan+gajwfONy3xfEscI/t+5Lrfk5Ky7wf2pZnq5ma/UAWYv8/mQb113tqgEDkNZEUE7wMhBFLO9Mej8dGHFkVpGjNICCEUtYQR562fI3hPE0wmbLPdcHZ2hi/v6XvrEXHOVEyg8F9P94hpAar1Pk0Bdtu2lom9c8HXfd2n+UE/+Jttvi40CDveUsHNK3v9RcRzBLJS+JxGFZgCU0ubWOkw5bkcTy7ZVuPj2M22VLVd48Q/zYvV5XzSz2X2msSsQepML1DVyY7SFR9mwdQNRNxJOWXyfU6pvN9KHzEl4mgSKd4pmmvJJ+F9xrRuzcGMkjUlS+G1WtlICy12/rXm/Z1RjyejxRy6FlWptId6wVln0RUfEp4WvJ6eqadZWF28R8SynNPjUhqdKT2mBaslkE3ZhM1TXkryzOVJpolunmyX+zPbEdziL508fDJd57F/FN7AFMwu6ASnj+ej52QiXrOyK54N1dFuCtoW9LJqIuCy4rwn+IAgk7Nk9IEsjogpd1QTnpSE1DVobBf0nFuVBWV+TUAKiVSKDazJYc5ToxPBh0DTWsJnOQellEBkmqMBnPd0GzMR2m63nJ3t8N4zDCO+NI7Va0xKmX6wJmotDV05z3P6tMMyGxk1TaDrWrrWEk6uXl9qjFDjAji1pV3nzBcG7zuQFRHarsM7AfEodnIPY55kM8ZSSohpJOUR1Yz3ma4zzkvOUlaMyvHY8/CBfXa/P1gzWM4l+DQdyZyTdUkWdxDvAuqV4Bua0E48oWqOUBunlq5ZzrnFgBzZHw945/E+0DQJ5zxNSMRRccWVrGk6nDgLaGnwHjMqCAnvLPvrxINmNDtSMgJ7HaAoRavPTe5k2NPWHZqM6N9oC2Sk8IJcGbQW8K7dXiveC57tIv5+JWpqG4UURpwXCCKT45CrWskuGA1JlVSsaPf9wM3xyDDWsXy64lNk4sQ++Yhm1QF7fq5q3JYYOg1e9dYLpRq04CQtA+jTz5f0ri6/eQ1iVzw7pnNLStOx1MqhBYNNsGDVGpEFzXnKRqaUOfQ9+8ORlDP7Q3mcMsdDz/Fomc+cspkPqDlOmkW0mQlUe9g8Wd3avqgITesny/UQPHfu3Jkyp7UokXJi7I/klBn6o6kvaGaz3fDqa6+x3WwmDqsFsj3H49Fs2YeRvh9IMfHg4UPefvsBMSaOhyOHQ1/oBDWYLfKW4mi6hss7F7zyyn0uLy/o2oDmaMeQ4mIkzsvXyYJ3DWRfGLz/QNZJMREog0FNLmosk1OMiXGIRhHIsbhwZcRlQmOUBD/MK71hGNFCPej70fi0qsUe3U7I2ebOgjrnHF7dpC1n3dLVJs/0WS2IXfJl3czBiQnVofBWEzkbBcHcyRzORQuQy/NOGkLIaBZiUJMNqXIf3iPqUZWJElDHUWEmWRDr3LzyVCVrIqVYmtWqJJkWLb9SLtU1I7vi44HbIZzVY4wU4MSyOZVPLq4uQiFpJiqmWBAjx2EwF6+lHMdio7cDSpkaOZfNYJT7E7LByaPbw6ZmbqYXazZ2STNQ5gzsSfB6+37FiveGWi6/fXNS+i9K9TCUhEvbtmy3xkk99j27Y2cc1v2R/cYkIftdR9+Pk9lCjJFqGJILd9YUdeJEcTA6j0zNZsE7C2Q1W+9I2+J8MMm7vi80wUTfD/Yd41Ayw0rTtdx96R7nZ2ecne24c+eC4D1D33M8HorUWHHrjAkfAuMYGQYLbA+Hvvw2Nh4RKRRCRwie3W7LnTuXnJ/vaBpvcQK1Qmu/q5sUiExDyZV5c8WLgefiyFqm0CYplCJKnq28XlZxWeeyohYOTLXIsyDTtpRTZshxyubmYraA6sx308zkzTydpHO2FVEk236J1O27qdwJwpKTOh/DosSfZdKoBUg5knKYH8cRfCbGgXE0gWjNAdU8XUBmVYQnZHZ0/sc8GOeOzhoALJUZLBu9ZmRXPAN0ScbhJJJ7KoXgmbOypwGcKw/nEV3Lj5kotqgdfDRebLTmlDFGYvVlz3oSGtatnX7b47ORvs8g8nFTgyc95hkePylju2LFO2OKW7EFn4rgXRk9tVm60ESzeFCdSv5auv41J1Iur2EUviZ42uDJZe6sjV9Tc3VWhnERyCb7nKrZPudsiZjgQDRTHTqzPbSbUty1Zov4Om5Sygx9T18qiMEbb3YcR4ahLza3uehK56IPX+l+M02gBgNO1NzLgqNrAk3wNMHhnaApEceh/qIzzdBZ83id31VOryorPtl4DkOEUurPSoo1A6nEVAeBErMrgWEml2DPAtzSnlFWolmyGSUcD+Sc6Ic4DTrnHQ5npZJs2UudBlHh+fhAaFpU50Fmq1y/CGRnBYPTBiuLLE0+K6KuUhPsMzmbV7xzflJScM4zxiPDcDAOT1EwqA1tXbcBgcPxpv5YgLl7iVbR+HphiLg42AXLgQ9mvmCqDC0iliEex/Rcf+gPGq+9Bl/60pNf+9Sn4Du+46NtrljBB1paux26eVdk9wCP4DHt1mGM7I893jk0JQ59ICvEMnkex5HDsWcYIklnX/iJN3j7EBYZ2OfBbVODuVxSVpZTFvb0OJdZn5LAvfWOFSveHTZODNkVTrkT8B51gprC1hQ1aqHcVC1mVImxZTzbnGZesxLjOD2uCjqqLGQg1ah60aqcYzQ99JyVYYzEmBlT5upoxgyoEgmoE1JOpGhjN8bE0A+MYz85kqHKcX/gK1/+Co/aliY0bLrWsqNL3q2zCqmqMg4DXkC9wzu7lixrLN4J59uO3a4zqsLZlrNNSxscsd9z9WB23qw8X/XBqkG1KZxa4VzxIuC5LWpTyozjbDFZ5bRMK89KF6qOnF2ZCKr0R82QGod1HCP7/X4qV6Qyw4lbzDW5GiWkKZAFKcoCJgWSdcGXcZUfW6kF3MrI1vJDsQWU+cSPjEXuSyfSuWV6KUoFo3F0qppB6qZMbgiNyZMsBJmnzGtRwax0Ac2JlEZ88IUr5cotGMdQ3Mnv8XHB04LYd3ttxVcJH2JdrWaWaiBblohQePL9MBp3XpP5rKsSS5VmGKNNnqkuat20zQ8LJ0HsrUzsqWNXfc+7/3wfr9FoeJbO/I+6e/9FhmViS0ArNZB1LIR8jH6HgtrEVxuh7T5Mj6dAEkuGpBRLcsleUyAvAtla1tesjONY9M4zx+PAOEYOfWSMN4xDIouYi6Vq6fewpqwUSwA9xhNaztAPXD18xCEEcx4r9vEmXRnKHNnQtA1gAXYNOr0r2rOLLi3vhE0bON9t2G07tl1D13iCd6Q4cNxTmsAaoxSK4EJCvC9ZWfdEjv2Kry6+mtea58rImipBXQVWYrjMJfP63tKRKQoZjxAwU4RgkSrOMrjFajbrnIOptASbZmowWKad8rhmT50Dp54qfO4W8ltTRrbsz5SdxZWLipt4qVPWVmrRdKYf5Gml7EhpnFaEMcok+jwrKZTVYl4KSNeV6rxfvqwum6al6za0jXWAVoqBEffXQbniWXBqnvHBxYcza1Xmf9ZvtGqLMxvZMUIqvINcmk1q9nWMs/d6pQioPEPO9T1Guk/ilL8rtUBPeXXPREV4T3u14kWGK5bsrkh8KNYDoiKmeiMyBbC1ATHnbHOUKlMHhWqhH9hjMVaoJXucorn0gTjBq0OzlqDRo5qLi1exj0YmW/mzzUjOEHEcsxIxa1inCc0JRyYIqDMSu5QKhi/cWk0JXCar8VUzShKbD0UzogkEchwQte01Xth2je1XmZODd+w2LduuZdM2NE4Q1bKNcg9FPTqX45/Iizh5v+SjFV+reN+BbFbo+zSVHPKUIQXLdM4TqpX/G0CRnHG5Iavi/ADSoKKMKRkxPEdE/CztNY9eaxrTWESVR1KOtk3n8K6x75OMOlvinujILpxI6j6WnQMowWRTmlW8ZURL2qmS0FOOSLQgVLM1sYkIMXnGGEqGeiR4h6pp8gXfkF2eJEPmRhjT3w2NyZxsug0Xl3e4e+8+5+eXtN3OLGul8Iw/XsyCTyQ+MdmqZXT1AV7R60Rhk4mC2OIyZSVpnrh8/VgmpCLtMzFd1Tqph5hIWgk3tu68HRB+EAHiswSip1/45OB3mc09EZ3/MNPIKz5REJSGkSyKc7nYxSrqKQsox9RgmItKjWqR2dKTLOzU1JwtkZOTkPKkZXnSzFhpNFUKU1VJJRubkxkVxBjph0jTNNwcBo5j4s2rnn0/ErGKISnT6MguCAkPNWuMSVS6OIBYWImY2152Qu7tWhC9Y/Q1WE+4EoxebBxnzQ4RTCHImyb8xfmO3dmWtmnYtR6fIy4rPkd8dsajJREw6bAgGX+S5V2H54eJj9tc+XzOXqlwePKs+Xq7dQPm7CNYllQFyBlxDUhAMc3WMZpxgveNlStFcKVZjEVTlGqa7uftF6KNZlTmQLZmXl2hGTwJs22tm+z3lhq0FqOXzHOhE6QyDZuzl1ENTA0hTlJbZnlrmegpbr6V9a3Z2NA0bDZbdrtzNpudKUI4N/+aL2jq5+M2YF5k1JBUyrk/aTgWrhyUDFFisYCU+bNiC9xUqzd1w2V81Oe+WkHs49SC+vzTq5KnFra6zpYrngmCUrwsEbHS/+l0VLq8VMFpocqqyUxJfiwLq2qBoqqSBbyeus7dbuCshgmaaxOYoikzBk9Mia4ZSckax/bHgZvDkXEYQRNeR1LOBE3Gm/WVMpcnegGlObouToV5kWoauZBcmWt9UfoRoW0cvnBq26albRq8d5zttmy3G0LwtF4sg5tBckY041SKWkrGAV7KY3F40ceOf8UnG8+lWmAVkGo0sOCSLqKvqfRfPpNzJqnZ0TpvTiGaG5z3hWYgUwlfREhp5q6lPHdLVp1VoBC8S7lDjKZgE6mfKAW1Q7KuUIFS5bcVpHcWTLqiNetrRpaZEuCLvSwUa97iS+1csf1TRVzAuYDzSmg6us1uWjlDkRRpWrO4DQ3dZst2e8Zuu2Oz2dB1HW3blG7Wqfi6lkpWPDO0Ti5lXqx0mqdNdE+TdjuZDLTGm8v3npbY6xk7jTPErGdlSUeQiY86B64zfah884eycHt3asFpAPs0mkF56+NveoHw1VpgfpIWslIyGs4tKSxljC2y/FqalWX+4GP0G2u0ss9JjRpvf9+tx5qNiyqAipJFQJuSvBF2uzzNyUPKbLqWMSbOjqM1iKXMOI5TkxdT8orHV39aefQlgeVqEocSyNrxee/x5fubpqEJAe8c2+2GTdfig2e36dh0Dd57ujbQNgHnHU2R3TSqhLfeEqkJJFnXmC8QniuQzYVf6gPU4C4EjyuBrRUYmMoiVRLEBgO0bcN2u7VVV9PhxJOwhikttIGYZkqALgaL8w4fupLpYfEey+AaJzXMTV5TPsnKMqBTJhYRmhDoiv2sd54QmpJBKhlhmLK19fuq/BCL/fMZXBaCa9idZ1woag25yo44QtsRfEvbdbz08stcXt5luznj5Zde5s7du7StdWg6YuH6roHsivcGKyKUbKNIoQI871lUQ1CYNHkWS1Vzl/Rkme0SajQtzJNeDXLnILbMg3X3dNYq+CAUCx47isfoAsujowS0S7UCPblfseL9wIENh8nScvFiHQTYODAVScVpCTixz9VzsPZj1GawnN20mDyZLBb0guk8Xjhnpta2k7Ky3W2JKRNT4pVX7jKOkZgSx6M1h1W1hFwtpYskZJ3foATLleNbE0VQdGGlzGeFciQ1IeSnxJQpEZijWRNM3rJtG0Jjj7u2mx43TYsPwTTtvQW3Im5y/3KfoEj2a936/MPG82VksVWPOCN3WyDb4I39XVagxYYuVaWBTLFcJng7GcFWZnXiMzu9aG8qJYspxQQlu+rxvgaV9RXBekItk+pcM2VkoV4njM+3DGSrh3TbdWWFWANZU1qoLiiVwlDHR82W1m2rKuKS3XA07cZ4hKrkFMtgF0Iwsem27dhuzzg/v2Cz2bHb7dhuNovfMJdA/ZMZyH5Q2ZZPUtbmg8EcMFomVC2YfdI79b2V4aYAWedKyYyaOVooy1r6B60powWXvgayy40vc7wf5Dl/m+c6P553aw5o5yNbg9cVHxTqOJuG27RwYxonNlZd4Z+Xsnw9ZxcJEydSNNp1mgNPN1o3fOscrtsv2/R+fk/bWp9J0sz52cbku4pGbO2DqQEtMFVHawV1epxOx0zVip2t5Gfqkfd+ygjLpKFuigc2v0vJurpJqaB+JjTB3C+rFFel9JVg+KuRkl3nsI8HntMQ4eOAZz9ZRb62K4GfoAXmik8abp+b67n6NYt1cv4QILf/+XjAefrow/nqd93+guHztc4z/dre+/eOD0p+72tx3Mr7zTiIyFeA7/tgd2fFM+AbVPWVD2pj69/xI8f69/xkYP07frKw/j2/9rH+DT9ZeOrf830HsitWrFixYsWKFStWfJRw7/6WFStWrFixYsWKFSs+flgD2RUrVqxYsWLFihVfk/jqBbIi34vI/Sc8/9MReWdtiXff9g6RP4jI30DkryHyOCNZ5GdY+7/8g+Xf34bIf4vIn0PkB5Xn7iLyX02aXU/+rt+HyDc91/6+E0T+CCL3PrTtr/jk4qMaYyLfgMgfLePpuxH5uvL8D0Hkz5fn/4fluVDO8d07fNevR+QnvMv+/DJE/sfv8Ri+A5Hf/J4+834h8msR+Ue/Kt+1YsWKFS8wPvqMrOp3ofpBtML9WlR/KPCjgG9D5H86vSJyAfxc4L9ZvP/fBH4K8G8A/3J57hcBvwLVJ8hLAyJ/P+BR/TvPtaci76QW8R8C/+pzbX/FiiU+/DH2a4Hfheo/APwy4FeW5/932Lj7KcB3luf+FeB3o7p/4jeIvAz8OFS/5x33RPWXoPpHnvB5/14O6EPEbwKeb/Gw4mseItwV+eCu5yL8JBH+i+f4vBfhLy63IcLvEeFvivBXRfhtIjTl+Z8hwl8T4U+I8HJ57ptF+L3vsH0R4Y+JcPlRHrsIr4jwX35Q373i440PPpAVOSuZm7+MyF9F5GctXv3XEfkLiPwVRH5oef+cJRH5HYj8eyVL+t8j8lPL838/In8Gkb9Usjs/6OQ7Vfeo/vHyeAD+AvB1i3f8O8CvBo6L50ZgV24jIt8MfBbV736Ho/tfAv+vxbH+5HI8fxmRP1qe+7GI/NeI/EVE/r+I/JDFcX4XIn8M+KOIvI7I95Rj+quI/Piy1e8C/ufvsA8rXnR8/MbY/wD4Y+XxHwf+qfL49hi7C/w04He9w9H9DFhMQCK/BJE/W47ztzCLcf4ORH5mefy9iPxqRP4C8O0lK/wbFmPrxz7hN/xpiPw3ZZz+EUQ+VZ7/pYj8trKNv4PIz1l85mcvfqN/H7MO9GVf/mr5zf/35Tf6PuBlRF57h2Nd8cnHXZ6SmBD5SOQvfy7w128993uAHwr8cGAL/Evl+X8d+DHAvw/8L8pzvxxL+DwNPwX4y6o84iM8dlW+AnxBhG/7ML9nxccDH0ZG9icDP4Dqj0D1h8HJqugNVL8V+L8zZ2hu43PAjwX+SeDfQ2SDZUx/A6o/EvgHgb/31G+fJ8saWH4rFqD+wVvv/JXYhPoLgN8M/J955wEK8G3Any/bfQX4rcDPQPVHAN9e3vM3gB+P6o8CfgnwKxaf/1bgZ6L6E7ELw/+nHNOPAP4SAKpvA13JTK1Y8SR8vMYY/GXgny2P/xngopy//zfg/wj8Tmwc/GLeqeJhmMeY4Tej+mPKcW6Bn/qUz72J6rei+p+Uf+/KsfyrwG97wvv/JJb5/VHAfwL8vMVrPxT4J7Df6N9GpEHk7wN+FvBtZbsJW9j+SOAzqP4wVH848NsX2/kL5XhWvLj4VcA3i/CXRPg1Jav4J0T4LuC/E+FzIvzV+mYRvlOEX1oef4sIf0SEvyzCXxDhm5cbFuHHlOzqN4vwE8t3/KXy3MXtHRHh67Ax/x8sn1flDxUvAwX+DPMCNQMdZSEqwo8HvqjK33qH410mez7qY/8DZX9WfMLxYayK/grwf0XkVwP/Bap/YvHa7y/3f5554ruN/0eZ6P4WIn8Hm1T+a+AXYty734/qkweSlez/Y+A3ovp3MK7rrwO+47H3qv4l4MeVz/0E4AuYfdfvxTJJ/yaqX7r1qdeBr5THPw74HlT/btneW+X5O8DvLBktBSvTFPzhxfv+LPDbEGmAP1D2p+LLwKeBN594nCtedHx8xpjhO4HfjMh3AN8DfD+QUP088JPK574FmyD/OiL/IdACvxjV//7WNyzHGMA/gsjPwybTl4C/BvznT9iz2+XO/xgA1e9B5LIE30t8HfB7EXm97MvfXbz2B1HtgR6RLwOfAv4x4EcDfxZLCm+xcfqfA9+EyG8C/iDwXy22U8fxihcXPx/4Yar8SLDyOJbQ+GGq/F0RPvcOn/09wK9S5T8TYYMlnj5btvM/wugr/5Qqnxfh1wP/mip/SoRzTquPFb8eW7A9FuSWbTbAP49lbcGSPX8E+AHgZwP/KfA/e5fj/TaMUvRxOPY/h2WQV3zC8cFnZG1i+lZssv3liPySxat9uU88PYi+LWyrqP5HwE8HDsAf4ulNFL8F+Fuo/vry7wvghwHfjcj3YsHnd1EbvqDal/wijH7wb2MD/bcCc0lxxgHYPOW7K/4d4I+XDNJPu/X+m/mo9HuAn4BN+r8DkX9h8b5N+a4VKx7Hx2uMgeoPoPrPluzmLyzPPbj1uVrx+DlYRujnYePtNuYxZpnifxerYvxwbFw+bfzd3Pr348d4it+EZXt/ODbxLrfbLx7X31GA34nqjyy3H4LqLy0VlB8BfDeW1V5mu9ZxvOJJ+DOqJwunx1Cyip9R5T8DUOWoSuWV/33YOPxpqny+PPengF8nws8B7qoSb23vpwJfVj2pdtzGvwt8jyp/onznH1blR6vy0zC60B8CfrAIv0+E3yrCkxo2X1Ll6mNy7OtC8gXBh8GR/TSwR/V3A78Gm3DfC74dEYdxVr8J+JuYSsDfQfU3YmWLf+AJ3/vLsWzovzE9p/oQ1fuofg7VzwF/GvjpqP65xSf/BeAPlUzpDiun5PL4Nv468C3l8Z8GfgIi31i+/6Xy/B0sOIUnZYLn/f0G4Euo/lZs8vvW8rwArwHf+9TPrnix8XEaY/b8fWalj1/A7VK+yE/EqBB/i/c2xmpw+QYi58DPfA/H+LPKd//DwENUH956fTlO/1fPsL0/CvxMRF4t230JU2u4DzhU/59YoL78W/xgmEunK1YULBddkdN5+N0SJWDVwyPWdAmAKr8K47ZugT8lwg+99ZlvA366CN+LUWn+URF+d31RhH8beAX4P9z+shKwfgdGFfo/YePlT/Lksn0Uece44qt57OtC8gXBh0Et+OHAr0EkYyX6f+U9fv7zGE/nEviXUT0i8s8B/zwiI/BFTnmnlHLoL8T4qX+hlP5+M6onXKDHYBJA3wH8T8ozvw5bdQ7M5PYl/iBWKv0jqH4Fkf8t8PvLJP5l4B8H/i8YteAXlfc/DT8J+LfKMV1jATVY+fJPoxqf9sEVLzw+bmPsJwG/EhHFqAX/2uJzteJRG9J+C1Y2DE/Z7z+IZUj/A1QfIPJbsWDwixgd51lxROQvYtSef/EJr/9S4D9F5G2sUe0b33Frqv9dGdNVnm/EjvMA/PZbgTyFMvQtWHlzxYuLK55Syi/4EvBqUQW4xjjg/6UqVyL8PRH+aVX+gAgdUBU5HgD/a+APi3CjyneL8M2q/BXgr4jwYzC60N+oX6LKL6Ccm6XE/52q/Ozy738J44T/Y6o8ib/+bwG/UZVRhC1W3XjaQvRvYovjv/0xOPZ1IfmC4ONlUSvyOzDO3+/7qHfliRDZYl3Z34Zq+pC+4zcA34XqH33X965Y8V7xcR9jACJ/EvipT6AnPOvnvxv4zluVl68uRP4Z4FtR/cUf2T6s+FhAhP8Iq3D8v7GF2neqzk2LpST+c7EKwd8BvleVXyrCD8IUA+5jC6dvB76+fl6Ery/b/BcxDus/ggWYfw34DtUTisxyf37Sch9EiMD3wUQJ+P2q/LLy2qeB36rKP1n+/e3YIvAB8E8XdYDltn8x8AVVo9h8lMcuwncCvSq/6Sl/mhWfEKyB7HuFyD8B/PXSyPJhbP9/U+gGK1Z88PjaGGP/EHBA9b99n5//bj76QPbbsebOBx/ZPqxY8VWGCK8Dv0uVf/xjsC/fgzWEvf1R78uKDxcfr0B2xYoVK1asWPE1CxH+OYwi8Ogj3IdXgG9T5Q98VPuw4quHNZBdsWLFihUrVqxY8TWJj96idsWKFStWrFixYsWK94E1kF2xYsWKFStWrFjxNYk1kF2xYsWKFStWrFjxNYn3rSO7abyebQIKTDRbEcR5QEAcOIcg9SVmc51yf8LPFRDQ8n4z0pHpX6DknFFVclbGOBJjBIX8GM9X3uHfj3OC9Ymv6a2Ht7dpL0g5NpPLfDpEZHqPVmNrtW3U3ZcnfsfpXqWc31DVV97xy94D7t+/r5/73Oc+qM2teI/483/+z3+gf0/vvTZNQETw3iEiqCopp+m8y9mkIp1zeO8RIGclp4zeGh9SxuXtM1MVe68ymbSzPK/FPmunfDn360Z0+s9ivM/bm79Pytgq23KCL9cUcYIrL2rKpJzsWpAzqnlxaVlef+bxd/Kdi/snQuddvnXFmpBU13H5CcIHPS7vXrb6+itbNCs52VgUJzjn6sm5OM/KgzLv6HTegmp5LAGkTN/ii7+JENqWEEIZb84+r8o4jqSYcM7RdhuatrWtPm3e0joupieecmRP+LwuRpPOI70+sYwZ3DQvyjPNo0/e1yfv3ec//7288cYb72ODT8Y6Jj9avNOYfN+B7Nkm8JN/9GtkhTHZiSlhg9/cQXwLYYN054j3OFG81zIkE6IRNNeRaSe2+HKTeZCK4LG0sapyPBzo+yN9P/ADX/wSb775Filn+iESUwIE57wF1Ej55OPncZ3Ib09oSi73irKQiVUBlSm0LpcShIwINEFoGruQyPId2SZ2EaFtW5qmQVXp+55xHMlZSSmRs83eDo/g5guM2D5mzdNAfetq/33v92/2JHzuc5/jz/25VbP9o4KIfKB/z9AEvu6zr9N1LXfunNM0DX1/5PrminEcGceBQ38g58zZbsfdyzsE79nfHLm+2pPTqTyyBbvusQkmxkRKmZyVGCMxZnLOjKMtMEWEEALee5xzNE1jkzZ1/GlRVbeJPOfMmGyh6gSCczgneCe0jcc7x6ZrudhtaUKgbRu2mw3OOQ43e26urkkx0R97DodDGdsy3aRM9gA55WlRHInkMu4RnfZrum4opHJ9s/FqQbKU36Zu8+3jYR2XnyB80OPy9Vd2/PZf8W2Mfc/h6oo4jjRdw+Z8hw8ezZCSolkhKjpkmyIJZA2oOsboGEZHVoc298jtSyANrjnHhXOapuVTn/k091/7FN4HxHfgWsZh5Is/8GXefPMtdrszvukH/VBe+8xn58WuWwbKdvbX8QHYvLicK1VPki7La4OWxFJd0OYyB6IZ1VTGXGZUm4O70NH6FhGhCYEQPEs8Sy/6cr+X+/kP//h/6L38id4V65j8aPFOY/K5nL1EZFo91ot7nTtCcLSbBgkNwVuwJwLkEU0DaCbFSIrRTniBLDWoVVSWAwdySgzDwGF/pB8GhmGwyTSXSSnbRJSzIOJQ9GSALadhV5+vGWOpWaWakVK0bAMbg7aNOtcJBO9oQkActMGVQBbIFsDaftuFyQJZT9N4m6gJNL5klkcLCgAk2w/pnOAbCwBUlZhPLyQrVjwNmjN93wPKfm8BZEwRELz3qAZabdGc6bqOrusIPqAZcrRAzQLTsUxGmZROJzGgBLG5TFb1/Cwh4K21Yw1q27Yl50xK0T4DUzZFRHDCZCuUNaMJUCEKZMl45xjGES0LUe89ToSUcqkGCT4E2q4rk+cTslnlQjXv7ZwAw75u8WPOk6RNyNPTi4vdihXvDgFCgJwE39h56YKbKg5ZlRRLIJvU5hyd0yqQEVG8yzh1ZI6QroCA5p403EBouHmgNKHH+QB+h/gNMWaOx2s0J1KKHIeBm8PRFmJOHluk2phOpZKTGWMklbFeqzrzIrEenz3O5TOqkLIuAtmE6giqxJwYc0ZEON9uOd/s8N6x227xvrjUPiXLens/58f6hDlynTNfFDxXIOtwJfTL5BLIigPnodsELu/sCG1H23i2m2CTztgT+z05J/rjkePBskMRSJpRBJVMlpKtzUpOEGPk5nrPwwcPGMaR/dWe46EvGUvIU00yATaxOXTKmNSA1YnDeY8rZR3vPCJC1myTp+oiYrWBmPM0oyElqt1tG87OWrx3tI3QtfY9aYjEMZ7UIkWEpm0IIYAqYwcpe3JKHA8wDmX1GhVNCd8EdruWtmvIWRliJpbJ+wfe/Mik+VZ8DSDnzM3NDX1/ZBiO+OAIIbDZWEXAe0doPKhyfn7OxcUFTQh0bUfXbEgpcXNzw9VVIqWadU3Ttpe0mPmek0DvNo1GRNhut+x2O1JK9ENPjJGcFUmJpCU3qyXCzNloDprJIuQUC0Uil+qOp0uRrBbc5hLIOu9pN56mbW1/F8HrOEbG0SZncrlmlffU4HkZ0NZIt07mOetJ8AsOladXZlesWEIcNBtAhJQ8PmScdzhv85IlNTIpKqJq0xggkhFJVp10ivN2HqackOEGVWFMMEQQ8ejw/dw8uIu4Bg2XaLgAPCm3ZG0YR8/1zR7/4Gqq+tn3yERPUM3ElMg5M4yRq5sDwxjJqsSkhconE82hbAGAuKisxGQ3VUV0AO0tkC0JLOeEV+/d5dV7d2nbhldevkfXNjiR6bpya/MTbgexy/v5hef7m6342sH7DmRnztnp0qmuML13tG2g7Rq6NrDbtjgnxEEZibbiS5E4CAk7cUvB3wJO1Cr6NeObMzGO9P3AMI6MYyTFtKASyQnHDqVkdeedm4NZC2IrR9A5Z5minJispstnHUqy5TG2h1LoBJ5NZ4FB1wptY5SAEcVV2gT2YzgRQnCE4GybzpOzlSyzpZvQrKRsK9rglK51bLpATAqS8Hnty1vx7tAyUahmkIyPjq7r2GzaUgqHUuGnaRrawqlTFStvxkTfDyeB6MxNzwtaDtP96fxxWgmpgaz3fqIXpMIRRDKiGZexjKoIDqvOaFlUZp0XgzEKY7Sqhhch+pHs3JQcFSn8WW/ZorpvFowqEuvYPr0q2F7X22m6VikB8eKYT6LXNZJd8YxwTvAevBfQmg211xTIyZImkrFgdooyCy1PFHG5TLk9Lo92bg6RPCRUHEcXiakHadBmIIcBcQ0+XOKCI+fEMI4ch3H63lq9nJM6WhaaieMw8uhmz7EfC40wT1Q4VSljYh4DMStDTFPQO06BbI/o0SqxYySOo82doeFs05FSYhyj0fHcfF05WVg+BWu1csVzZGSlTD4O8YLLEJrGSpVNy27TsusCbRssiBMLCkFRZxNU6yEFR3LgM7ishXlaJzMYNTOmRE5xumlOBG+cuRrE1iuCZXQtaK3cvtosQglgQwhlUi/NIyLkLMQkqDryNHmXDHPd8wyacimqFM6BQk6QxILXOI7EcQSFEDzBl4k2J+JYuIE6B8nBO2g8mpWololugqdtPG0TcC7bZJ7yE/8KK1Ys4b3n/OIc1UTONasaGcdxOu+UjCCkmKxhEkFzRsThPWw2HZeXl1PQKmVBOY4jwzCQc6bvB/p+oE6F9tZlZkSmILbSCYwXnqeypVBoPkY4pZGSIU2W2ck5nyRialMI5RoRc8Zr4a2LIGgpl1pAq3WIKoTQoNkqL4hDXLD9dYCry9RsmdqsjENtJlXEKTnpKZ2gUBlWdsGKZ4WqQwsnvJ4/4o1nHQK0GzvPNCkaS0WyVCVtgVWrCVb9qEMnNHYeZ4TsPUMOZAmMYyCmgDhPq0Ijyjgk0oM9j+KDxeLNxpBzDlcC2Zyt4tEPlpHtp4ws0/5MjWfKlESKWaeMbMpKLJxyzQOoUQptwowE7zheRmLKVvWo1VBkOkbgHWk8axC7Ap6TWgAeceAB8dB1HRfnO9puw/n5jrtnG9q2taYoEqA4STjJZJdxQXBdsEaPXFdvEHGkUq5MGsnjSBpHoyXEHk1K23q831KVEqSkmepU7UQIjeAKd7wGtJaFDVPjCeVTKQkxWvk0jpl+sNWxE2t2ASFHJZYJ26FIzkjJ2NqFJzMcB4ajZbSa3YYuNIAyDCPjMIKADx7vjYO7CQ71RiGILpFipm0Du01gu20YYyFNxNMmnBUrnoQQAq++8jKH44GHDx8yjlaWd+5Yqg+Cc4I4GMaRoR8tSFNKB7Xj/PyC8/OLqUmx66xUf319w83NDTFG3nrrLWJ8UILUDJzSDuqCTdUC174fJs53znOjpxPLoHoAMRWUFB3kbO2W00Q1VzisOUwZYyJi9IITNQMfSom0UgswPrwPqEKTbJ/FCU3X0HR2GUw5kbIF91dXVxz2B1LOSCrVmJI1npo63dwos2LFO0ERUm5sIeU84jMuOHxwiLNx6UOpzI2J8ZjKY4iDlJxJLhXDEn5Kxgl0Qemc1Qsf5cCNdsTccDO0HFKD84HznbBFyeNIv3+LkaMljLDFoyV1pJT1S7OzKmNKHPrB5qGyyFSVsmSVkuCpvFibM2PhxWZVa2BDydkW1qjSuEQrmbbx3LvsGcZICMF+G7BgVis3GEpmasLUTL0GsSsK3ncgO2VCp6xlKR+GQNsEyygGRxOsTqB5PjmdqBHcHTROyLJY2c1LRJgmxITmhOZsN3TiuCIO5zzi3ZKWinMQ2lLGoc5/Ula/AefqRGd0AefKSjA7y06VXXCU7mQAZw0pqjUjWybtrOQy8C0DlqaGsnqvJStlQYVtuE68giNLBq+I2j4H7/BeyDpfYFaseDc4J2y2G1KOUybVgk1bCNkk5HDI9Lw1Wjl71tk4rhI+2+2G7XZ7kmUdhpEQmikwffqpOX8m55r9La+USciCwmnnLQAtjSATRaHSdOwThUZkC91aHXFzsnaqFCEgat/v1JcFKSQxrqGIsNl0tJsWUGKKpJwYx5HD4YA4Z0Lbi4rPvF9rNnbFe0OuzYc1lTotjEC8TFUEFJIvVDuZ6TEKUxO0oIVyYMka57GF3yjE7Bmzp0+OQ3T4LDQJQqn6XQ0D+9LGkbQQ5kTwUueZqtpjGdV+HEtmVcip0gkKDQemxewUvBa+fNa52cuoe7aA7Xwm+VL5iJmUFzx0rdne+h0FC6rFO0rlrXgh8VzUAhXLSrShATEuXhs8jXNIigw3V+Sjt0A0WZkuxYE0HqfALo6jnfAqiNoE4cTjscYoyRHNI5ojkHDOyPBU6ZvS5OG8KzI/LT6EwidKSFU/KNeOmXdXJsuSUfHOE8rE7MS4gilmnAjelc8WrhIoOSbGYSwSQRC8fY8XTxOa8jmT/LHJ3gJuxWRWck4IFrCaJqajbRzq1ZrCgJQSMSaOfc9xiKxY8W4Q59hsNuSc2O12Cy54ldAyHcuc4HjscVxb4OoDjW8QcbRtA9gCru/7iSN7OOzp+74oGuQpMFxyUXOmNKiA9wFfxmXl7i7hnCM0Db5oaYqr498avGKUcp0oGdwSUKpCUiUl217bNKaHW6ozPtSMbCXDCt2mIYTWxlnb0rQd3ju2Z1s2W/u9rq6v2JdjTNmuSSklnAuMk4rDsjta1il1xTNBM4y9JTniUHpEtMxd4kpTcy3JZ3CFfhagacsCVC2xUTnj8wIQEEVVOETh4VEYs3A9KvuU8Q5csIXtmBxv3SQeHX2h09g5bIEstiBcBLKmWpDm4LQGshPNpwSsTwhkK72gbieWBNGuFWitX+Tqes9bbz3gsOnYtg2Nd4TgLZZomxLo29ieM7Snv+370p5d8YnCc1ELkgihadhsz/AhsO02bBs7GWUcOD7Yl2xsQlOZCNJIigNUHmq2hi1cQFyw4NQ3tspUxRW5Ls0jIhHvrdTug10ERAQfPM4HmtBweeeyZJAyMfXkXCdP47jFcWR/OBLHOAvCl0yQK4MzeCGNI9Ep1dJBAI2QMF3XPCp9TjiH0RxaPxHmfWtcI5tQPSKKcwHntFAXTN7IOcd2006amU0bbFIvF5dxjPT9yM1+z/44PM+fasULAu8cFxfnhOCIMU5BrTVYWQZ2GEayZsYxcrg5IiJ0bcd2s8U7z3a7mWg4w9BPwdvxeLQgr3Q0h+AXFB0pNAKbwGxRGaZFWc6ZYbBzuGY1m6aha63hTMqiWIBh8KCJcXSkmBjU5LpMVq+oUadEHMcyuQlN09pYcx4fGqMaFY6sc47Ly7tcXN6haRpeeull7t69RxMC55fn7M7OGIeBH/jC9/OVN7/CYX8oXEZPjJFDe2As3OBhHKZmtZTSE8xYVqx4HDlnjtcDOQ/GFdeBJmdcI3hxRlsbc9EeB5xVOryYdCUlG5piUeERowgBpJwts5qE697x5UeeMTtuonKII8G7wv32HAf4gTcSbzwqCkHOoaXJ0i/6WCDCLWUPzdaqoZWXXug8WedKac6zgkmekkYQNZvkFsLdsxbOWmL0vPn2Qzoim64hxZHj8UjXtbzyyj1eunfXFrvOglmwXVtWc1asgOekFtSVnPO+ZF9M19GJoCmRxr7QASwjq5rRFMnJMjpaBgAI4kGCWLlEPLhsmdcipIzm0rk59YZYptRZU5f3pgzQtQ2bTYdqYhgzs757Xc0Zr6hWC52T6X5qAPOWwVpSDFiQ66fMbmln1uymhFF1G3KlXFtXrRO3rvD7UirUhNrDVRvRvLdGlrL9lE0KZYxrRnbFu8M0W208hhAK1aUGmTIFpTlZa1NUqwwIjsY3qNfJ7GBJS6jNXtbNbGN3KouW81p1wSEVShViNkGoE1DN0NbHlcZQb5YFtc/mMqGfKCGw1Kuc5YDKDzCVbEHAgfMmybXdbmnblsvLS+7du0fTNFxcXnB2fkbf91zdXHFz2GMZ3A1t1+G8n2gZKRmHtjbBwcrTW/GMUDPiSMmk5eaye16U4fNElwHKnFCypMYrQN1MlatryIxMGnIxQx9hSHCMSh8zKcMQhTHCOCrHfuTmEKdAFnElkM2PBbLLrK3qbA6yHNuVCz9JYeb63GyOEFWLdiwM0RFTwDuhH0YOhyM5RW5uDlzv9sQUuRzOiTmZIZJT6nL5SbSCx557cuJ2xScYzxHIKkNKkBIh2qBwwF6tE1/jAMMRcsI7ITjBiZLiiMbBJtaUzBABcKHBhc5Khz6h3vQlNfaIRoSEd0oTbFB5L7hQMjuhyvtYQOtcKd8na9yq2SUzcPA4acrFQRn7RK2UTIFmymy6Bm3McUXLwCR7crTmtOAqLYCiMGD8Ay3fmxWGIZbSCiZ2rW5qQMmFW3vsR2LMOOcYm0zjvXWHqpJUGcbIGNfMz4pnQ8qJ6+srDodjoQJYJjHG2UnO9FQzk6ocgB7RlEsWdmAYjtOiqwahKUViLO48cZwm3qpEMImfT7BJUfU0EM15fk/TNGy6rlB/rPqRYjRKQYzFNrdsTWdXPhFH03YI4ENjjoAsAlyETdex2WwJTcNLL7/M/fuv0LYt9156ics7dwghsN3t6LotiGOzPWN3dk4ILZ/5TOby8o7pV9/cMPRH9vs9X/jiF+hLdjbmZBq2K1a8C0QgeDM1sMCzZFTLeiurZWSnsVE69cVU9ObMUXnZ2DhlTDmMQpeVMUX2w5E+Bm4G5TBGGu84axt2TaAfMvtDz83NaAs+b6oGDsVjOs2WPCqJJxaBLIJi7ptVj12qwkCex/rUy8nS4askrUQmx8ucM8d+4OF1ogmOrJnrm2s2XUs/9OwPB9q24d7du1yU5lO/aLZcZ8QVFe87kM2qHGMki8MPo2mgDiPpsMerQhxhOCA5selazncbnBNyHNChJ+fEOA6Mw4Ci+NDhmxHEoX4g+8bKGmNEcsSRaYLSthYwusotdULTWvdnE0w+yBe5nDgmhj7ixETgvTMheC8t6h1xjPTHY3E+gnql6NqG3a7DOyHFXDhNxs3VZOWfpvF0jUecEKo2IMqoyTo1Ven7SD6OhZBv5gu2SjW5r6Ra5IwUL46uaQjBAtmxkOOTKn1KxDWQXfEMSDHx4MHbHI89V1fXRiNYNmOUwJMy+WiRxxll4MDBJtzgaRq7NFSum2HOBtV7y9KkKWt7mgeZA1sTW6+qBXNWs+s6dmc7q76UTs0xDuX6MJ58v4msZ5wqITS0RZfWl/JoleQaY8R76LZbXr7/Cl234VOfeo1Pfep12qbh4vIO5+fnOGeZ2iY0uNBwfnGHYz+QNXPv5fuA0YCurq84Hg+8+eYbPLx+xFtvmzV2jHHK1q5Y8U4QUZoQkVxMPrLHBQtkFUt+jEMkxam0ZyMpZ7RI5DkxGoII+NL0ITXgVXtvnwYeHfb0o+f6OHA4BprgOWs27ELL4Zi4vtrz8GEP4mjaxlR8aiCLFl76aL0tyKSjbITdBpxRinwIxu+t6Voql5fyeB7/qcxnNVGUy1i+3h+Ix4QTeOPNN/Fi8+9bDx/w+ltvsdvt+MZv/HpC1+CdtznS++k7a/56JRm82Hj/HFmdT86UMy5nomYkZ5JmJI7IMEDOtKVMj7hZKDJntOjDmqakx4kH504dd3JGMOF0EaMTKDJ1KVduqy880xPajJbJ2hUKQCnP2IquSGrlohE5TbqKNpZVDsEjRHJyQMY7+56MmuTPJGdk5Q/rSE3T75NzJiYbvEFmxQMW+nuxZnBFy0RPCWSNYJ9REnmtk6x4JiwpAEYRSFMmtgaF0y3rnB2BEkwyZVgLiWaamKrzT+XnWWlRT7epUEhH0/e9E5etjl3Q6fudyMSFm4JYlhke216lH0hpFishMyBUW9yu2xRDiC3b7XY2gWgayyqJmz4nzrJMXsy8IQRvXHbv6LqWY3+kaZopkn+yLeaKFU9GLZFLmQplaqxanNtZZ5tkKXzTKl/gnGVMWSZnSxNU4b3lwoOP2WTsYrJKaJW/qnz5mKLRA5KbGqmklGgmOlHpX8lVbstbM+W0oM2KSqlIzOtQlutZuxzMDoDT8ZZ/p5wZc0RUGfKI5MjQNlxdXXN2tiFl5XCsEl1qtCld8PJLKLuOwhcbz5WRvTkMDEMmjhC8x2s2ArtqeRxN9D8EdsmcSWx1V6JQlTJQFZFEkhEQskSSG6aT0wOo0komeVOltQxMCWJ9KHJAnjgmDvlYVn+uNIFUztx8sUBNwqvr2sLRm3mzbWu+8MaTBXHZXGudR7w1b4mlfm3lLFqsepn4sVnKIB3LMeVMDqVEq6XBrUiwqLNjGZISU1pc2Cr1aR2qK54NOmX5M13X0DQ2Jvp+NImblOcmpUUgJovTa7arpPBeH3++KiHkrIUfO3cqWylRp2ylBb51obYITtUoD8PQ451VJLx3NMHsrIWyoM1G1VERs69VG5MuJbxC67yplXjPnTuXvHT3Hk3bcvfuS9y79zJN27I7PyO0Lc57hjESH12RUubho0dcX1+TYiqKBUe6TcdnPvNpXn75JXJKXFxcMAw9oQm89qnXuLm5pj/2PHj4gOPhCFhj5ooVT4Uo4os7V9Yi3qpTKtE5MWqcK4YHZT7JCnGxsEtWsyeOgqTKA7cmxyE70IZWggW9XUsXOjMPKpnX4KEJppAj4oyWF0wxyBdqjytJI1VvGdnSYCneQWgqrwFgohVULWoLUpd9IUzvq86YWvpepmyq8xN5AYSsjgeP9iR9g+3mCvGe6/3BmsBefok7F+eWne02NKGhKhJV1N9ulel6cfC+A9mUMo9uDnhxtO6IF4fXTMgZp9B6Yeu9cUlDYiwC5EalqZmcwinNGXSk2AMZPzRnm7gab93RCp2ABkdWYVTBQkSxMmPboVp4qWmYjA+6rpn4ODMp3W5OTKoISobJV/tZR9tahjdncINJpYjPiPemg2teg6UElGbxdsHcWrCA4VjUBkLjCUVBwSblwjlyAjiz/4uZnJSqNTslsm4N1BUrnoacM4fDoWQjTWKq70eTxYqJIY+k4qRTcpfTZ21dJbduSypBlfAyBzGT9KqT0zLbC6ClMSSX83mpcFBpCco4DPSHI13X4DcburaxQLZqN6PT/joFlZKVEoe4RHJK07a0XUfbttx/5VW+4eu/nm6zYVc5rz5wfn5B120AOByOHPuew+HI3/7bf5vPf9/nAdhuLXt7795dvuWbv4Wv+8xnrRegPzDGgd1ux+c//30cjnuur28s8z0FsGsgu+LpEMAFRVJt/c+LC7xR09qNg1RUcYr7XVRHjjJxTmM23mrKQlLLjroQkKZlSB5o2LhAUFMfQXY4L5xtzCI9NKays2m9UQvCrK9cVJMx1ZFQnLZszKmIpZGdL9WPudnSaDal4gM2uS2qNiWFNIeXmuyWHYTC0wX7PZIjobz51hVvvPmApg08vLnh733hS5yf7fiWb/oGPv2pV2m7lnt3reqiFAWykqHNawj7wuG5VAuskSkTsxHYsxq/RlSR7GlL+0bK8wR3W4Nx+mf5LGBKB4VojgfR2lddKAWUDNLJ2Wp1+7nL2uLM277vJ5+YmsCsacs0L7GAtvjSz6Ls5TuW0gWyIJzPh3Uy+c/rwlP9yROiYZnUTa6Euv4tQcZcMF2x4t0xL9rqROKWgSlz6a+el4Vmd4KlIcHUU+JkCmZrWb++t3xq2oc61kt/ByJLvmx927JRzLIypwF03d9T+kJGcNh4sWFoygchmEX2drdjs9mw2WxLZSXgQ4PzlaOujGNkGAb2N3sePnw4ZZVFhLFI87WtGSU4pzQpsN1u2WzMIGIcx0ItWjoErljxTpgpbFkVV+bEep2XUp0rVlvlE67cpNDtbM5J6om5BLIa8BrI6hE8XlxxsvOIDzhvfRxVScc7P+kuV4UephnHxgEiljkWIVcehEgJZIsawXIMUoPYMuOpNTejhSpRu6bLUdm1IKP4xQXBleuE2dLHcSTmzPXNAR88WTP7/YFj39tvEBNZrVq6rFzavq01zBcJz2WIIC6Yq1Xl32VTIhBVxGeCKskJQzvSD6ZUgKbS1phLyXDmplI7NlUnTqyo4nImIwR1BBWSglb+X8wc5MA4nnKHFKYmrlwaM+oEL2JBa101AmWyL1+flTiagPQwjvTDaAYJpUNZih1gTMk+o2bWANa01TQB9cpmY88p1kDjgi/7U7KympExMcZUaBcZRHHOtAODc6ULO80B/4oV7wBbHFrH/9D3OOfL+bXglNYAdfk5nQp/9ppYRcImOpv0Kt+00nSAwsGVSc2gKhlwe9vT2DuV2RqGvnDHlX4zYOyBRPCetm3RYQTiVE3JRcJHMO3YEAKXd+7w2uufZrvd8KnXXuPll+/Tdi3eNZMs4G635fz8AtXM8XicG9RQXJmcRapjWObY91zf3BRx9obNdsPl5SUvv/wyDx68gveBL3/py3jvP+w/6YpPAHJW9jeJYUwcjqZE026EnYBvhBgbhjGQsyMmk89ShRgbxtSiKiQcSR2qQqYGro6GhlZaYml8vtyZusBme0Hb7UAgkUiaabXh8k7D6K2MPy9IF2oAJdBWe8ukCOIWYXXWPGVkx5g5DsbFN6rQUMb7SB7L2M3GuUVAu4QmIePJTom5QbBGNucdoopXUDzi4GY/EtMVh8PI2faLDMPA2W5nNKPSYxNjmprJXKmUplWy8oXB+w5kLSvTgEZSGiElJGdkjBalBYfPieQcx9bT9w2aE060SHwY32cKZHXmC6HVQUtwOUM2TmxUoUFMPSAmxj6Dc4wJ8EYnCG1rBgll0tTSpV3lgqZGFZkHrx1PmcRVTko7Qz8UN6NkDWLlc2a4UAPvCGqBb2ha2iZMy8EQfKExmfsYlAavasvHYA7aWfEoOMU7oW1NhSHnzDiK6c6uWPFuWASNuWg015tJ07nThsjF53RK08rEgbXmKHPOuri44OLiYgpCzXkuMo5pkvkyXmyprBSNzEoPAEoWyDKZRr3pycl0ovttZxJ9KRZqRGucwEPJMiNocQ2zsmhL07bcu/sSn/3sZzk7O+OVV+7zyquv0jQNKdn3Ouc4Ozvn8vKSnDMPHjwipmR8dGUyREFkakA5HI9cXV+z2XScne+4uDjn2Pe88sqr3OxvcM7z+d2OEJoP/2+64mseKcHNVaIfIjc3I2NMbM8doYVGhT523Aw7Yg70yXOIwZzl8oaYt2jJwEYNxkGVAGL6yztp2EqDOmg2PXfOe5w47t694OJiR8rK2/uBq8MIznHv3oZw3gKnWUuRSgOAeUmr5FL+9Jpo84DT+fqiqvRj5qY3567jYc+NDsSYGMcj/XBjFdaU0GwNZroBHQOqgegoZkFljiwVDo9DxWQCr/c9D65u6Frb1sMHD7i8ODfjpXJt2ZeGMO8dTdfhvS3gV7wYeC5nLycWYE4RaI1Ha8OH2sqtZlJyzpb5cMvhoyePbhcpizt1KbkXfnzhJJi2ay5ajhZ4upxx6kpHtvFsq1jz7As/Z6Xqt552XM8loCX3z8o/5f1lX5C55FnLoDV96gpdwR5bd7UqOGcT/CkXsQq/zxxZ54w4f0pVWLHiHVDpMGX8zaOqOF3V8v7iA/N5/4TNyXyemlZzsY10MmUwa5Z3eYPa9DFXSk6/t2aPdeLZzQ5kOh2HLK8vtfhZS6TF9KFpW7puw2azoW072qY2cNp1oTr4VYOIWoVZHvJMXcjFfc8yS6FUUYwT7GmKE1nTWHPp6i604llRFXJysaPNUUkRxAtjFProicnT50AfG5I6Um5I2qLqiDkwqgd1xQWzwTkhakOiKbW/hJOId8aJbRqHZC2ulZRqpKdhIa9XrhGz7e1MmbNeDmOdBoUmOXyhwGUx6kBWoc2Cy448eobgjPwTYSSDRlQjpGjzcR6hxAUUisKyQUzrjk5yfUqMGe+SVUj7gb41h7QULbvd9z3HfsQHTwJ8CI9RCVd8cvG+A1mzV91CimY9UswRcCa55VGcZsgmcD6OI1IkrNRbVijrzAUSJ1OnMhQpnkrey4pT8GqDSbOVHjyLrmaML0MUkubSqJWsyaRMwtUuk/IduZDUq1NRXWG2rWWD/GICFCwrGssqz3nLcIkYNzjGolodj4zHwUKHRaYna4YE1X86LfQ8pwl7os3qFLx6p2jhN61Y8W5w4thtzsz4oJT4cjUXUCXGWMTLb0Omu1yykgpITCARr0rSshD1ghezdlaxjGZtsPChITRSyn1VZ3VuDFFKBrRIXYk3i74scOgHomb63rhxlffmXMB7wfuG0HY45zk/u8PLL73CdrPl/r1XuHfnZXa7HWfbC5rQEXwgeEUb8M5ztjvjbLcz296uo2sCY/DGxxsL7YnMMPRkzXz5K1+h23Scn59zcXHO+fk5KHTthvOzCx7trmlCg8jKkV3xbBDset5448fGpLzxwHixV73nrX1DnxqiBvrcmqqrOrKOKELSkZStKrjptmw6c8cCSw5pVq7HzNWQ7XuGTIgmAXndH7k67El4o+kV2aySslns4a15pgS6iuI1ItpD4aU6BBXomkBoOhRH7IQ7OyGnyPWV8PDtA+OYubk+crx+ACh545B4Zi5ivoUs1jRd2IXWsGVPq1gzmFfFh0DTtIVS1OAAzYmh7/nyG2/z4NE1PoTJBGVYlUReGLz/QFaEXQlkxQuSEzpGstSgNkLsC2/WM44D4Aneg/g5e1kbO4pGK4UXO60JC63WBpIFs5opgawNupjNMECyyV65lBBxJJ/LyjTQNA1NYwoG1Ss+Z52841OyYDvnzNnZlq5ri/WupymBbM6zs1GDn2S7sjI9P8YE0bI+m03HdtsB1hiXi+xWKt+9bH6RWxq4UigY6iCEWWR6xYp3gnOOs905Q9+ThkhOiRwTQ1+MEfLcVDlBFg+kGAsU4wHjyEFQ04fGAU6sYYSi4+iLY51IaboK09imVERSzXgaWdeCWOeQ4JFgtsz7ocfHgVgm3wymUuAbvPpC29nifeDi/C6vvPwaZ7sd91/6lAWy2y277Y42bPDOFxMSW4yeLwPZTUfXNvSNB82Mo9Ebxmj0pJhGvvSlLyEC9+7d47XXPsXLL0dUodtsOb+4ZPfwEaFpi67mihXvhuKLJUrrM4nMTVTeuIE+C2/vHV941HIcW5IEkrSmFkACRiY5RlWcOC4vPHcuNgQPOJP2ylm56jMP+0Twjm5MtKNZnFsge42Kx7VCE6qeq5asKEbhK3s7NYDqHOhKjkjuQQvNjoBJeAWadoNzATRAbtGceLuJuOEB/XFkvNqTrr9s8+75BsZ7FjY3G/uObBbXqUx0SeyGs0A2iCOUgLlpu1IZAk2Jvj/ypa+8yRe+/HZx6zujaVqGYfxq/oFXfIR4/9QCKRlU53DBI7nkUnO2DEsEs7njJNs430udN0/ec6oSUDmolXWuC71LG2WPta3oKUVh3t3qB/94RHhbUeGk1HqrXFopB7fqktP9REPIi8flGJc0hZnKUPZNS+FUpp/m9hE89syKFe+KWyXzx0//J5xX5VzNujhvoTSBZNC5BFk57tMEWJQSTDXBGkmyLsZ1+cbbtAGbqDMpF446dVhJaTyD4ANt0+JDoGs7y6x2G6MS+NLY5czQwBy/TLrPe3P1m22qZ8tMLY2nqsW5b0EjYvm4XKzsmPxEK1ipBSueBXbOz5W2Kk4wRDgmpR+VYbRmsOyELAmVqrf6eCBblXlgHqtZLZOpVFpfvS+SXpqxPGpGJJcZ1P4n2EKUuUg67fjcAGoWu8LpGJFCubPki0ntoSZj2QRPDh5vUWfpdLZbpRbU4TUdz+LrJyLULerSUt1E1eRAx2jxRhgj6JPn+hWfTLz/Zi+URpSmCWy3rbleoTSAU9NePF5fkdPIpgtst03JjljmdWquYuauzrFi5c1osdEs8l6TCw9UCSCw7JAW4VUXTBzae8em7SybWsp/lX83Z0KlZFXtvm0bVM0QoZZGdeqEdIuJd17NLoNRXHkfTIOsisLHpMRc+YITgWLqCjcLXIrbEJCZ9ANTWgflimeDqnI89qRxLK54ZZxpnXSYuHBaubPzpwGKUkZGVGiaQGis838cI1dX18iCYzrGyLHvTee1yGB1mw2ghCaYBmyK3Oz3do5PskC+OHLVIHnmwadcupCTNXh1mw2qwuXFHe7ff5Wu23D//it8+tOfYbvdcvfuS3RtR+NNT3qz2U42u00TcM5bhaXMeof9DW++8QbX19ccj3tUM+Kg61ratuX84pxX7r/M659+ncuLC87PzmxboZluwYcTBYgVK94RAqFJIIrPJjkVB8+jY+ZqSByGG3I+YioEDicBEIaUGIbSmZ89SZ1RbS4DZ9u7NEFoGiWTUDLiFB8wyoFXsisSWR58MJnJplFCY2X3U4MSmeYmu5uYsvbPnCEJop6YhGE0nfUcR7JcmWRdgKa1no/truHOnUuGTcfN9QO6tiXGETB1A8ThmojLGVNLmL+/xrfTXmide51VdJ2naTu67Zb2OFpMURbD4zCSiyX3ihcDzxHI2oc3wXN5cUbXBNrgOWsbvDgON1c8ejswjgMhCG0wDdhJ7F8rO6dkQ6iZj5IdrdaZpZlLEbLzqNj6sVrTUvivWjMkZYIMIVjzx4JOcNsXfdlBXQPj2nilWh2QzMjB4abyf83qLAeKBeYm/eEX2pI1eB6TlUvBbP6qTq13NtHmbHxih6ekjCZHo9LUvWLFuyJnpT/21iWc8qJBcs4sOpk0QTAxHabX6virUjaIK655jjFGrq6v7d3l3I+xKA/kbNzZEOg2RqdpqjzPODIMo/HkxZ1ay06LuluBbDL+uvemDetc4N69e3zd132Ws7Nz7t17idc+9Tpdt+Fsu6VrjBfbNR3bzWZSPei6FnFC2zRlsoP9zQ1vvfkmN/trjocDqqZs0HUNu92Oy4tz7t9/mU+//prxbs/PCjUpFO5tOLlurFjxbhBRfDDeuUsBFTM7eHRUHhwSMR3I+YgjERBaKfYEMTEcksWQuWFILT40+HzJ2UZpG1NjNbKCToGs84BnDmSD4BubN9uQaZoiDTlVVEoQWZut0CmYreGlKmgUyEIehTSIWeHqSNLeMrKuwfkW74XtrkHjJWM/8vDtM9q2gZK5TeMIOHxM+JSNb6vuNB3LIiML9rq4cg0JhLal225o90fTgAfImTgOxOgep1Ct+MTiOQJZMdcu7wjeuK9NIWMH54ijSeMgxcnVqwWyVOUAJg5dHSh1EXh7HVUqjCVQdTiVybgAqeYFsw7e3H3JTCcogeft0sSsZiCTXmaVDwItMl2lKWvay7qCnUOA6XeZFrSnK8taop3qtpUusfg9l6L19lvIZPywri5XPBt0wb+u59lyOrDkhrDkw8mCyiMW+c7V9LmErrbtuSZhrzVNw3Zr3NWzszNrjGI+Z4dhsOoGFG3aGgQat9ZsbqfLwYS6uAyhKS59HdvtdjImaNuuKAiEyWmsBsn1Zt81j3WTCIuM48A4jKRcF7e6+A0e+0lvUYKmwbxixTNjMvQo5l7VqTaX885JRiXhnRB8xgHepUIDUAQ3PZ41nqtCz1JZpyaNFuMUFvPeTBfQErDKNODLzk7jcPH+LFNS9JR/AHWOdE5pGkfwjtwExqYB1eICOM/xWRWpOvKcXKbmEbgYYlOmuE6f0yvy2L7U32EdoS8O3ncgG5zj5fMdTdOwK7aSm67l7OyMEAI+WNkwjiPkEbJ1BmuO5DTOJ5lM/2ESWtdsRiBAdqC+ZIY2HW23I6mQskOSIylIzEjRmbSypwWDOSXGIshcSw0hmC2t937K1GoZaKFYY5qG7Iiqcjj0xFhE3qUaKLjJHcj2N83HokviQGng0sV1ASbKhEguk3jpCvWFd6yUlk1bbaeoT+k0X7HiFDUDKmpjgkp/qdf74uxjTVjmo24fnBdodsLWRZ+bA08387ir0YD3ntde+wxnu3PatuP+/fvcu3fvZLF4PPZ86Utf5MGDhxyPR958801ubq6LasjAOCYbe6Wa4nygaWxS3G3PuHv3Zdq247XXXuczn/kM5+fnnJ9dcPfuXZqmZdO2bDed+a9vOrv+FEOFzXZD1b21zHDP9fU1b7/9FvvDnuPhUAxdhGHoEWc0owcP3uYrX/kK5+fnXF5esN1uOB4OHA579vv9ZKqwYsWzIGflcIgMA9zsYRxhvxczNXCBJmQ2frSMaQPbtizsAoxJGaMQtCXoDhc6tmdn7HbnhKZhfzgyDkerigBNUxZvMMl94RwuBJtf/ILSw+xDVBV27K66fTEFoIqSvE3M4uwaAYrzStOaxfvlxYbXX71L1wauHx15FG7ojz1f3rbWvKylsToOZBFCjhYXGDFg+r3mnhKjPFA4v2OCISrDqAxjpi/yW2qdofZbqzIno1a8CHjfgaz3jntnO3zwNG2Lc45NZwOsaRp8sPpGjCNpPBL7azRF4gi5nrw1iJ0mzfpvKfp1oEWWQ7wjbFqa3Y4sjpw9LnvGlEnHgdRbh+K0FlNI0TTsYM4OiTi64su+FHUOwdO2TQlkIzEOpUEG9jfH8tmS/S3E+ZQiWlbSThZKC5zE59OKsYYKS03balUIlrkWhw3I5KxMojXQWAPZFe8OVYgxYt4+M2WnnoFOZOGXzpThkMUJq2UxOVvR1gxqnfEgBFMB2Ww2fO4bvoHPfOazbLdbXnvtde7fvz9lQ7337Pd7Pv/5z/Pmm2/y4MED/vpf/xvknBmGgb4/0vcDbUtxDXOWkQpYIHu246WXX2K33fGpT73K66+/xvn5BZvNlvPduQWsTUPXNIUe0BK8L9cl2z+A4/FA35s27M3NNQ8ePOB4PHA8Hsg5ISrmSFSa1B48eMDZ2RnjOPDqq69weXnJse85Ho9lW/0ayK54ZuQMh0Oi74WrK+OX7gcha8BJoGkS243JzHWtst2Yw2NW5XB0uBECDZ4t3m/Y7s7YnO3wvuEwjJOMJGJNVq7Q4GZHPdNdnow/FtOKVTwXSSAphkGlwunqdcFZ81ku6iW1+ct5CK0QgnBxseHVV+6w61oeNTe0KhwOgd22tfkvZ1KOjHEki6Op9D113K6GLINZgKxCTMo4KkPMDDExjokx5kJLktKgqk8trqz4ZOI55LegbQrXc/H81AAlxgV1XtFkXb6KImnu9pVSZ1BlDmoXBU+Eoi8L4oORuzcbsjhSDqg6fMr0GaLaiq4WTEwWpJkMCepgaNuWtu3ounbKyKJVF7ZSC2oDlhhlommgktEzxZUrmqafWmYVkjVrqeLy7F+/9HC37OxcCjmlQMxcJAHEizkiACE8PshXrHgyqjmHonVxVccRtiCUsmJKRZZgpsPM2+AWBccCWsoE6dgUA4LNZlO0Vi+mx7vdbtJfrsHs3bt3J2e9l156iXEc6fse54Tj8UDTNuzOdjRNICedqiBV/3W73bHZbGiaZpL4MkUCu6/VFO+8TchSbaeXBg2zekPOaZrkKybN3ZK9PR6PtG1bgtcj/fHIOA62OK/GDesCc8WzopT6cw246pPLkn+9uaVJjs2BKp4gwTKrhZqzVOHIWXETRa3QdrTqlS8zrsWKdsGLn3axNjTf5vncPs0XdBv7nN2cE5pCMwwhTLTDqndbNzUzCRb0pyWV4Nb36a3/1ubQvNiPeSzfpj2s+KTjOTKynpfvXDKmxHE0rtk4jhyOPUM0WRvxjTUzCQQxWoETgZzIOZJTJimlbC6IB1RMP680mjRNi2safNtx79XXuHz5UyCOQ/YM6hli5OzRNdf740JFwLitoWkmL/Q6mXVdx+XlJW3bFqF4m5DMI7onZdOA9U5QMk48bbspbiyUyR9iGqes7Tj0DMPRup+jGUNMDWY5GX9JRhhrznZuErHkmPGevKv3jjYE4xklZTs2k8UnX+nf759sxYuAemGXOQtr1IAyDnzAhQ7EMYyRYz/MHNWyCZOWYuKnGkfVs91u2G4tmLz/yiu8/NJLbDZbvv7rv57XX/80bdty7949Li4uLTvadoQmEGPk5Zdf5ng8cnNzwzd+4zfx4MHb7Pd7vvjFL3B1dTVN2EiZpEqyc7c746V792nbjsuLO5yd7YwysOnYne1MbaRp2LRdCbaLBbVz+GDyW7ZtJedISqc3M0LQ6RqQckIE3nzzDVQz19eP2Gw6+uOB/X7PG2+8waNHj7i5uTba1BrIrngGiEDTwJjMACFJUeGRUG5jaWROU4NvFstCivOIF9pmg7QX+LCh22xMIccJKWeGYSRnpWta2tBAqfb1wzBR2bwYT9Wp4ErJfg5mZeoBuU1/zYWAVF06J9v3lIhR8UFQXKnUBDbdht1mw7hJHDb2/U1oy3WlWLxzyhGutHy9FczaUryMUXQKzM26PdEPiWHMxe7dpAF1DWRfODwXR/buxRmHvmeMkZgjKdrE6GK2TExoitasgiRMXDaT4kCKgrhxJrsriM7sUspq03cdzW5Hu9ly95VXuP/ap8F5evWM2SbjZvuA7upmmsQp5cEayNYVqojQtu0cyKZMjJGcM4fDnkePHjIMAyDGy8WI62dnNviceBN9RhjjYELqObHfX7O/uTbx+TGi40hWK52O40DKJgifSx1nWkmXo4VSoim3EJTtNtC1wRoDklEcDG+93z/ZihcGlS9g//LVgEAEFxpcswHxIOZPrppmLlzhgbvSOGVZVcuu7HZn3LlzSdd1fPbrPstnPvN1bDYbPv3pT/Pqq6/SNA1nZ+dst9vJ+a9t23m3BPq+5xu/8RvZ7/dcXT3i+77v+3j77bdJRdg8pVTKglaN2HRbzs8vC5WhY9NtS+NXy2Zrwuhd07LtNsXNzyZbESlmKLNhQa4L6OmWTgLZnLIxkVR5+PABOScOhz3bzaYsVgcePHibm5s9x8OelApHfsWKd4EI+AbcaIGslkAW8eBCyX5mEFM2yLlaQEvpzXCEtqPdneHDlqYs3ERMYWQYzbGvC60lbwRiHompUuvmCoVjordSc7M1mC17e5qNRef/lWRRzkoqSjyNumnO9t4XjecNfRfpup6cMmFywbO5tCqUTNnZRXGjkqGUhezkxOnVIg9oVZthSIwxnTgHrnjx8HyGCE6KFqSd+BnTXpVcOaCWIUHVBJGdCZMH3yCY5aTzDYhZypZFXXH2soDPNx1Nt6VpN4R2Q2g7cJ6UHZqtWaVpzP+cRUexc9bsMQWyziEITfFJDyGQXbHpy5mxZJ2C92V1aa855/HOtGiXgWyTPDGabJaTInCdMmkcycOI5oz3R5z35JyoVIn54mFQnXl2s94nVI/pGvj61UBoxbOgBKNVRs6VzOo8DozvWjlxtZRYM5kiQpj0V2c+eXXHq9zYtjEzgk23KcoBzaLcP99mm2bbvcqrrRWLi4uLqZzf990k9F6915vQlubMgPfNtM26r6c3oz9Yhnk+volwNFVs6m91u35ZJ2lbhB4OBwR49OgRIQTGcWR/c0N/PNIPQ2kSW6fOFc+CGobpRC+wFIYWU5HTgK5WFtHluW2NkL5SC8qW6yJ0ohAVfWZyaXyaqG1F1afOQDUN+qQMZiHO6wkR4DRQvG3sU7E0OqmbWuzpfDvlGJxu/1SaAKMLlsqRWxzvIh80qwhpyTaveFHw/gNZQLzHeY9rPE4zY0r0Vw9JWenalrONceU6LzShwTnwncO7tuhOblG/LeX3VAI+JRTfc+c95/fus7t7j7bbcH7vNXaX9822MmZ8yvgYucgZ1wRAcWWl532wjFBoK3kHxMqsoWlw3hvxfIzkbKoFeRwZwmAbKIOl6zZst2dTedZ7b6TynMxyNitDydbknInDQBxGckpc31yx398QU2R/fcXhcGPadnFEizhsihFNtqI89r0R1zMMjKTBMtvtZmPNcytWvAtMDqvFCzTeFlh2zltDpoq31BBCjMkso4G2bei6Fu8cZ6VT33tP123oug3euSJ9taHbdFzeucvLL71Mt9lwfn5B13VFu9V4s865KbitmRywfdntjB6w3W4nDqpqnhpWcsqMZVzmknmpC2MpAzM0Zg+tag0uzldFEU8o1J0QjC8rCkjNKNUFqhQXorIIxRaiOWXGFHnrjYG3xQwh3nrjDbbbbckU22R7c3PDYX9TsrIrVrwzLJM4kvBEbRmBUZUxJ2J2+FyUekTMca5QyYRQGqcDTbtls73ENxuapityllYdbYJHFbpCu8maGVLPGEcEaHxD8LWyuDREyRNHdSknKTBzVyWX8r4Uyp+Nl5TN6Cfl06BTi3lBzpjddKwVRY8QQC0JVW1pTxUK8hzQT/s0b9x7aBqhaYRQxrwrfSg5lYUoqdAg1lD2RcH715GtWU5fypbBspE3+xvGIRI3G4JC4wNt1xKaDcE5fBMIvjOuizREaa25opTqLZuqeAEfAtuLlzi/+yptt2F3+RLd2V1TNBhGiBGfjNbgvNT40zi5vuF8d07XGB8QV3m3i47NnEmNBaOiyrDpCc4Xf2d779nZOXfu3DlpXJk0YssASylOXNuhHxiHgRgTjx69zdXVI8Y4cP2o4+aqRXNE+yM6mIh8GgbSODLGROyFMRkHaNRIdJmmhW5rXMUVK94NgslHeSdTIBtCmEqRiiOL6Rn4shD1WCC73Zoj1r17d7l//37hxwaroIhMpgCbzYbzs3MuL+8Wbded6UeHUJopWyvtB4/zp656rgTEVTHk/Px8arrSklWJMXI8DKRkZgtXV9eFAjRr5Abvp/qoSNXONImwWkKtGrJTyVLzlGE60WwWij1tRnMipsSxPzIMAyLCl/2XJprFbmtBeF8UDFbR9RXPCuO/Cgk1CUk1WkDMiSYrWQWvWECbqtmsGeYggdB2NN2O0GwJvp3UBEx3ttrCGmc8lfMypmjXBB9sfCwoAxas1n2r/5W5simFnyqFfiMK4iZFoaQ66eHq6VZRlUKLU1JSo0pgDWqoMy5r3VDJzs7mDDWIPf39pCxYQyP4UIPYUuiszreFdlAXnSteDHwgaT5rjnLlJmQn+MLFmdwsVWfv9sL9caGh6RSfMxJt0lOMhhAcuBDoNlvablMm4rAov+jklb5ctU1d1rJk/cyU9gklqHXiUKc4b5aWqFrXcyHSz2LrpUzKolxZxokTB0HQbPxWzSASabuObtzgoyf2B9LYk1MgZdPScyKoc6hzFryX70FmyoYixkWKq9TPindHtVsOTmgahy8BZds0iHNkhKSOrBbgdm1LytnkrM5NA3q73dJ1XdFaNg3WmuGso8qqE3VshOnx3H29LC3qyf1yX504cPOYRs2NyHtH1Xe2/TAeea1FOu+mgJVCBxAowaycfD+U5pQYJ078Scd12Zd6rYAiMJ/zlCFDjH8bvF0yx9GqLnkNZFc8I1IqLq+pOGJNydAaeDlUbYxljBqEmFMWNBO1xhdnyCXsdJ/HXv13nfik1uHLsBBY0Ojn2dEqFjMN7mTSrNlTljGiRd6aLeuaUyLGRIzRKq26zKpK2VeZ6HhLtQZgsX0tv4su9rEWVwXnWaio2Gdzns0VbiWJV3zC8b4DWTvFEs5B2wSbPHIibTpG59m0DZvgLQuLlRiSgyxCKiSXZnfJvTuNre7SSEqmBdsERwiC94HdnZfYXd41kfRuY+U/tRL+OJiWYx4HNMbSzGLf6QQr2TOAOMQrhTCIhDlYlGCDfbPpkHxpk3YNZEVou7ZYZLopi2Tco7l04b23TlHMcjYEo0744Nmd7UhxZLdpuDnbkMaB/QPHUZScEi5nRs2IOHY7wbeWrR6TNYglhOtDD8fxuf/YKz75cM5xdnZG2wTONi1N8KWDPyDOMcbEcUikrISmZbs7A+Du3bu88srLE9+860x/9dHDKx5ePQKFi4sLQmhRFUJo2Gy2ph6w3bHbneG90QmkdEdP8nbMGVm4FdjKrFSZ68Ql1ZXLKEJV3/lU99l4vCIOJTMMR5w44+uGzsxFHFDoBMNw5Pr6iuvrK479kZjipLvpjFg77ZNlhfL0fXGMpJTw3hPHSFOUGI6HI2Ncx+WKd0fO0A/C4ei4OXpujp5jEnLKRZ5AUO1QbczkpwR42uwI/gylod3eod3sCKEzukFZxLnisIlaY+fc3OyAsvh0Dgn2Hqm8WZaOmmpUuSnQXfBcFUuqqKn21MynYPrpmhNxTAjCYb/n+uoKHSOHw4GURttuiZqtUiK0wbSe21AqR6UHpGZTc1bTq6UG1zYufXA0nadtPW3naFuhCWXfomWJo1be8RrKvih4roysTToQiv2cNoFN0xCALgRa7yZ/85jVjAScJxWuzqbbsj2/wHlz00ppRMT0aZvGyp6bszt0ZxeISJEjsW7JFEfiMJByIsdoej2uKAsIxXc52VCVXGw5fRmefmLIV3vOpmlxO/NnFlfdT0p2q7HSaozxZDKt2ZhQMlJ2AfF4b683jSflTcm+Kk0QxqGH4Yj2R1KM5DCgySOidOKRoCYlMkZyTGRVjn2cSkUrVrwTxJlz3aZruDjblUWmm5qujkMk0xNTpnXG+xbneOWV+7z++uuTvrKqcUYfPbrisD+YtE+3KfGe4FwoFrGdVR7KYi8EN2eDmOkCj2s9LjNCp40ZVvUwA4bquLe0mDVqQSi2lwJZieM482KnTE/dBy26sGaAMI7GZ88Lviw1A1a7T2rJNVf1kRHvHJoTYwiknIvF7TouV7w7VM3NaxiFfhCOg2PMUhp6c+GKNnaO40g1EA1neC5AGkK3o2k7gm/x3pqOK03GOwtSJ+1kygJRyxgpZHABTPNy2rM6Um+V8+dG0DKblspH4aLmmdZjvR4ZhzD2PYf9AclGs0tFRWSyyhUK7ckC2uAF7439h+giq6ozxWFBfPDeKk2hcTTB0QTbjqplvJNarDF/44oXAc8RyBYyD1rEmo103jaFcxcCTWmQilnNyUMhOUeSCM7jusRGBBVvzRpFNiQ0nqYtzVWhupGUskVOpJQY+p6xPxYt2JEcowWfOSOlFJliNI6Pc7YKdcm2p2bVR1n5TmUUX3i0JZCtQvBLUfh6vywp3p6EpxKPczisXGSZrg4HtG3H0HY450nDiCa7CKgKASFlRYaRMWXGGBnzgTGtE+aKd4fU5i5vfNWubSY+O5gu5ZAUlzI+NLaAKw1Ym43JWaVSHgROuWsnM8PcpDFPeqclR3PIPc101sfL7dTS4FweXbqKuSkbq5qn76tTr12GrEkLoE0lEOd0wTkMA/u92csOg/FvqzFC1cXMydQTcrKm0ypFFoIHNTqDNbB5JCaSW6kFK54RArXdPqsnq6njuFKPMEqMQzxoCKTQWvCpG3LuEAlGOyi181zoZqrKGBPDaOM1LmgzVW91ak7OllSqqkA2AdbdK6YLYJS7haJJbZTWnOuoAyxodjX7W+LNGBPH4xFRpe9HxqE3Ccqi2XzKg2W+rui8sJ30YivPoDynWUkxEcc40YRiMk57XZjWTGxeLEZXfPLxfKoFJDzQiBIEmjawkTM0YzJbLiAIb+/3vPHoEX2MRBUGFXCeT/nA5t7LtN64em1XbCZbT9uWALaIKKecOB72HPY3xBjZ35gKAIqR0FE0BNR5C2STchgO5JjtAtE04BztZsN5a0YNNnyteUudoi5QXbjwczBadfmQeUKuHCDDaabJldUxzjLWTjxnZxd0XUsceiQmPI40jgTf0rd7xHn8ZoNrWpIqxzESU+bq+obj93+R6/7qef5UK14QSCmvb3c77t69y267mScOVfoxETYjKSu73RkXl3domobz8zPu3r3Ee8/NzZ6rq+tpMqkNVha4njrcae1Jqdzx6f2ndILTfazEu3myFIqRXZnUvFdETIqrbS3ATGnBb0WLygmMw8hwtEZR74XtrgNRUnLE6BjHgQcP3uL7v//v8ejRIx48eJu+P1qFpTgi5Ww6tv3xWCZUpQnWlNJ4X/i6nm7T0YSGYRh4qFdrs9eKZ4SA96hrSLTE3KA4gnicCE3TEbY7fAjEbsewuURdQAZFehAVlA1SbNd7mAK4B1d7HlzvEYSm2bDZJHLKDONAP/Q2B4WMc0a/82L88kqFdQClOZLi3OeDnyoVVT8BF+mnKoQQnPW1BCk261k57Hu+/OUv03pnigZjNqOkwzUxDkY1yImkGTRbkitZszVZwWWoFIZppWwXmXEU059+mCFGrl+65nzXsj8eOI4DQ4qWLKvUgjWSfWHwnM1exrPxAipC453pwqpNqK50R+v+hqvjgf0wMCToE4h4zu6+TFJBXcC1HZvdGc47ujaUQJYygSXIiXEYONxcE+PI/uoRh/31xIt13iNZYZOQbPzTsT8yDiM4h4sN4r2VMCYB+NJsUke0lFXhxE2woFT8zBVSP69268Q9vzhnYwEE4+qaUPSWTlti2zLu96RhJI4jOdmK2YXA9uKCZrshlYAj5ow0Df4rb7JOlyueBQJFlLzl7OyMs9220HEsqPRNBh/Jqlxe3uHl++aatdl0nJ1tERHGsUpK6SK7UzOydo5rKcjUBs4laiY2LZqhJj3Zup8nzVjMNB8soPVF49l7PaEWVH3cJbUnxpG+N2e/7bgt2ScpdCVrPLm5ueGtt97i6uoRNzc3xDiWQFYnLm8cTXGkKi74hQ2ocfsC2+2GEGzBfXOzZ7h1GCtWPBFlpabiyQQyAXBTUBl8i2/OcG2LdpeM2/tk1+AY8GOPy0rWprgHZaKODEW2cX/suTkMOBH60WhoWW3hF6PR9VKEmLJlfr1xy5G5kiJOkMbUfUw6LxSVk3mMx6jAOPeGiJsUhoSi/NNHHj58iLmqW6k2xsgwHMkpmsymzk1gdaFcyLvUxre6eJ7StWipxA4c90rjhOOx59j39MPAmCIx57L4XpOxLxrev/xWvZXBoK6UJ3CFYiZlZWS6jP//9v482JZsW+vDfmPOmZmr2fucqltVt3ktfRMOYwyWQlggnrBwIAzGMrYsjABDyLLABpsQ2MiSxcOBZWQshQWyMQIkBAh1bjCtQghCAcLCSHTvgQA/wo/3Hu/eW7eqTrP3Xmtl5myG/xhz5lr7VHdunbp1762T34l11tq5mszVzJxjjvGN75tSZIozc4EpC+JK/fHV0oCyyGPV1kRrJyuFlK2MEOPMPE3kFMlxpuRUM561LFmcdRLnTMnmNJZTRJxDnTmYNBHz+8ffjtnGEk4s4MUGuBO30A98zc4ULQS1DG6z3muKBpWQ2xhMy/8YgYC+3zBstnjnmfuJHBOuC3R9T9/1FEzqswBzTOz3ew6n8SN/ySteIlxQYBa0MqNahaDvOxRh2JipgRkahPOkdlnWbwu9tl0uPN5hoR5YplRqFfLMZ021CRPeHczWA74YM/f33yogzjUJr+ZSdJbPejaKviyp5qpNbcFsZJ5npmk2K9oaZJvySXvOOXCXUgNrxwXFQZb3S+0PaMe2YsUHwQJBSEmIOTCX3hREMFcsjQGZHK4IUSE7RV3BRXDqcJjFrBePF4dZ7Bgtoe869tstTuy2E7N5F/E48XUcOXKu1DnasNHKdjAVHslnll0V6licN1WrvFWBXIRcr7V6/aRitrfRQyzFEi+lQE6klJemTzsWk+YSS8raeKuTb+Pp6mWAW5+YC0xj5ODMivbx4xu8czx5euR0moxqATWG+OS+2xVff7xQRta3VV2VpWpDC4U5ZcY4k0rhZjrx6HjDYRxJxUju4gN345HTNIMPDKVUyRFPcUYRUFXmnJnGkThP3N0+5ebpO5SUSNORPE9W+h8GBHP2yuNI1JqlOR6I82TGDSUtEltaEmYL62rDSO3sLJf8VpYuS/HVyM+fJ7oud8QQ7QRQuU7nU0tL57IExK7SDEqKaCx0vqvZH1+97AO76z39drO8JxcCu6srbsaRfmdd5H/hL//Ai3xlKz7lEGHRPG5BY8658kIL/WbLw4cP8KGzjOxrr9J1ParGMTNHOl+tZStvvaJpqXZVAaFlZlNMTNNc+aQ2phovdZ7nxSHsMlBeGlLaQlLAVsNtWwtiLVjNNSC1l2i6sdzLKoHZ0MZqEV3lo5nnicPhwNOnT7i9veVwODBNpnjSiL4mcReXCVdRithn0VU3M1cn/JIzWsrimLZixYehZDjcFe6Ontt5x828s0C2WNbTl46Qvc01gyLHhHjwYs3EQYReBvqwwTsh1YWaquMzV9c83F9XakFnigYp0/mePhjf/xenAABp5ElEQVS1qGTPPNlkZPzX1qhd5ybncEVxXimuZUBbAGvBZZxhjo4YlRyVKSolKa44ZrXzTRJFfcE7hfmETkdyjEzTDFilVgvEaJnTPtcoWU2+S+u4T6mQmylEXThPufD2O3c8fRQZ+o6bp0eu91umOfP24yPHca7W9OHeeWvFpx8vyJFthO+zvuRSk6+NWXPOjGnmOI/cTSNZHUkdPmemFJlzIqRkshtiK0a79iiFXBvF5jgzTSPj8WATSRwh2Q8XX0XlsEAxiSOnRIozKc5I8XgnqAZKitYQthy/WyZlpEpMC8tKtVldPouWXWral7WnZQlklwDYtUYxO8TiA5vtjBQldhPTOJJTxgfPsKn6nV3HsN8ResvOvvrqK8yrg9CK54Jx3VpzVyvftazIIMJms6UfBvb7Hbv9nq7rztUOpAZoAe/LvYyjyWJVpQNaIGtVk5wS6gQnRgm61G1tLl/3Mqn2gufgtg062sTFwoX1wUN+d/bzHBiftzUaBUItrTpyTovl7Ol0Yp5n05NtMzXV1atmchtUzsfbZI2s9NmcxpriwYoVHwxVmGaYojCmgbFsKEUsq4kg6nHFNJBDhD4XnC9o7/BbTxDBS8C7rs5lEa3W55thIPQ9wNnu1ulir160Zk8XbS1dCp+o4p0Fs9QAVqoCgKOqFNSsaU6mgZuzkDKkLJRsfSZZqgRnNl14j6JJYYrkOBFr86jNmRbYm0oB56Rra+hc9qv1OfZfViWeZkhHvHeMp5nN0JELzNl0eb03fu+KlwsvQC0QvJhuRtOsa8p0qpAR5lKYcyZpoVRJOy2K5kIhk7LRBWIISyOH19rxsVQuLzTh1ByqXSuJeCv5BWlqefW1Xe08LqZALc6E4W2CtourntTiaqCq56wyFwyHSy/3SzSOrNB4sgpcdHHWTLWrzgzOVR5SMfMF5wPiMuCs1JIhzebC4osSeusmR5UQPMOwOnuteB5YIBdj5FStX3Ou3bxiGdZh6Bk2G7q+OwuKL+oEptJ+WbpvAV5K0cZr7MglWQb3gkdr/RoFp3Iu71/oyLbspbsMCkUXbqxIFWO3oWQLwCKLcUgLbpdObLFrwZQFStFlvEGbDLPJ9V0Equ12thl6KWMuDSZgMnxG0mee53ZISwOMZXNXasGK54MqpNkaluYkzMmTVUjZUVRwWfCFZXHUOUfw3iQsxVWtWE8IlpQRUZzYPBk6b42JIkvgKcB2GEibRCrKmJQp2Xkg5ULOJp3VBQjObrsAzjvEFXwniFe0OpBZRlaZZyUlc+xa5mTx+C4gztP1wrARvFdSnol4Cjbf9UN1/Ou6ZUHsvFvm8YXwpyDu3HOibULWFl84KEJMBSFVGqPY5+jawnNdYL5MeKFANjj7YSJGHi+AFe1hZuIuRcY4c9JMDqC9o8RCyhkphWk+cTjcoiWz3+8pJVGKle+a+1drxCglgxacFkQKfTB/acSZywfVqGCeyanYa8UIOeG8s8cPHZu+o+8CfbCB57yrmVgH1VVFRG0gUTmyrmWKzoPjnjamnDskvQt4VzVlvauvz3IRHKHbUAalqAPp6iq3kNIRJ9BteuP1omhJbLc9D8vVR/2qVrxE0KJM81ylaPJCBWhOXf1mwyuvvsJuZ5nYzvsa9xVKTov0lK9ufcq55H4aj9webigUxmkkl0QupiiSq2azazy6GkxP07S44gH3KA8igtdLPltdTAqLHzzeUYLH5RrQZqWkYukitSDYOWEYBgAzZKglEqNUKPMciTGSogm3x5iY51Qb4EpVQKi+9y2jVV2JRIR5jrb4FAsmzsGrLE5fK1Z8EEqB051yGOHu1HE796TiGHNnAZgWgla9cefYf2ZgO/R0QdgEC8yGbmC7GQjeUUpCs1U/Qh8IXTB2XLLMZClKh+dq2DGlzJuPT9yNMzEVnt7OHE4ZQQku46UsAbQTEGeW8+KsWbvUSimt+ghIVlyOQMGHgX63w3c9+33Hw4cbgoM7hOn2lqxKN2x5+ODa1H78luI6xHeEzi7ifH35lhgSkLPcnqmuCFkDRTtcVsoxEyRZpSjYa5iJir/H+V/x6ccLnYWd1EnJNWpB6y+0gHYqhSlnohbUCeqxiUgzIkrKkXke8c5VWY4mrbO8XM341JKfFsxkUwle6MW3ukMt+Wlt9KouJTkvWVzvHaFevLMA874Diq36mpWfLOWX+xnZ92xckfPxOufMQrAGstIC2YriwPmADx3OJ8BW5FqUkiNoNueyOJNTQLXQd57tpl/lRFZ8KBQsqIwFpeCrQULoe5wIPgQ22w27/Y6mkwogqovl8/2MLDWrmUg1MO26jpTjhW5jFT1Xy5aUKoK+KI7Avdul8kvP1QxLRbnaQAUWyLZrL1LF3OW8eKz2sdSMrikbmLPR8lmUYr72OVeNWMvCNr3YnPMi07NkZGGhSzRprVzLos45+hAWbmxTL1ix4sOgBeYJ4izM0TFFTyyeUw7k4vAlEbIFlTkKvfNsQiB4zGa6ZmS7YEYgRqy1qSV0gdDbHJxlKUQiG0fne7o58faNaSbHBMeT4+bOlHs8BSd1jlzmvIIEhzij+xWHjU8HXTDdeKdKV+dj54XQd/h+oN9s2Ox2BO8Yb59SJFDweN+x2fTk5JnpmPE10VMVh5zDLY1ltdnaotcWP5srKE2Ht6AlkjTbuBRfNdtZzmtrGPvy4KNnZKV18zu0BoKKmq2qKkkLUTOzZoqA6zxeAkWFkBTrhFKbIHPzQM9VoLwsEbE1mHRQMsNmQ97tEc0MZDrqJFkELVXvrnJsjAdbyQ6CnQS6bmmEacd+7z0t188fMC5dlhfPp3J67kW4yw4aAbdSMpruLUqeMzlOIMrpcESlmNLDeCLF6av6fla8nChaGKcJJzBP56ap/f5qCU4vF2El5zpf6BIESnXjY1lEWsZ1mkYOhzsU5ebmhsdPHrPdbOgCbDYdqq7yaq3C8GxGtlnLtqys1gbI1hxpYWQrJ17aU9ZesGKvOU9zYwAhCH3fMQw9vsoGdV2oZdayyG/N0VQL5mkmpnSmFyyB+JlCcSna3nYuC3fYE2ow23Xd2uy14vlRGxQ7lxlcwqFW0XBKCI7BWSPXdrdh6ANdcKQ0M46TJW8ksukzwUtV7rE5Ydj09BurSOSs5FSbs2ImpULMyhCUVx5sOI2Z20NBjkaNyerMgAejA4ponRdd5dYZ3xYRk8QrGecKvuotINa0Vdqi2DXFkxpIFlt0ChCcN8MHMTMI8Y7Om0MXzswepAaz1N6UpqCgqsbBz4FCRtThi+JUaoxQNW29I/hq2btmZF8avEBGVnB+sC5DLOuYS2Yuxok9lcQxJ445Er0Stj1DcbiQbKKsPJhpNlmpeTZlAhEYcl8zqVYq3Gw3lM7j0itsPVAyPo74bFaT0xiJMddlW65lwYJoxpEJHjZDz263ZdhsCF2PD6GxbS/e0fsEsM+Mhya/dWlzCSzyYU2j78yo5xzcOrVB6wP4Dtd1+K6naGGeJsbjLePoSXmiu+2IJXFIJ+a8erqv+HDknHn69OnCaVUtvPH6G7zy6mcqlcYvwVhOyTr8a/DWhYBXs3NtVj1Fz1z2pzdPuT0c2Gw3vP5Dr7O/2rHf73HuO9jtNjVgtZ95jLEqBTzFe888z/R9X9UQLgLnrjPOOuesq9TMiwWvVAqBKZGcDubOlbMdF8BnXn2VB9dXdH1XA1nbx+l0sjF1OnE8HLi5ueHu7o7TaSRGE3ZvY7hlYUupC+EayNra08a0855+GOj7vmr1Diu1YMXzQQRcwDth7xLqJ2bxBFWyeobtjv2Dh4Su47WHA9f7gaFzvP3OU778pTdJKTEetmjcE4IZAxwPRwCurvbsr64QMZMEc9EyJY6SM873XF1/ltdee5XbY+LmqDy6Mw3aaS6kallrQ742WztLUNm0ZeYkQTKDn/GS6bwiXUYdeJTiHMV7CB7feUI1FCopU1LGI2y6HvWZWToLap1ntwnsBjMrSgipzsKpVkiaaoKZr2SKL5RBEC344nGabXHujSrY94Gh8/ShariveCnw0c/CIoizbITQvDeEjGVjU1GiGq3AMrIOX7yV0JPxWRHrLM451mavmpGtDRfWXGGSP0VgGDb4tIOSEK9IrLa10byeG/ncUkwFk9gyndkQbPIOIdQO5Noe9mzsKpcb3j0SFgMEtWzshczdvee1f+2zWm5eZGSblIHU2T/lRJxncnZIUGLyZAqxTGQ17uKKFR8ELbpIS03Vwvnq6tqamC4yss4JGS5sLCtdBqp2LMDZeCDnTEyZohMpp+qQ9YScE+M4VmksWSadplrwbEYWMJ5cpRkELTZkpTZxcNnkwb3xqcWC7KY6MM8m6ZOur/He0dXxHWrXsvVq5apBHRc5sHQvI3s/kG1Na/dH/gWnt+3DB+MYh7UJc8XzQatBTucyvZiUXJZMBobOMrHdMLDZmSFQ56GkxN2dyUjuhsTxqhC8cHt7x+3tHcDicCcipBxJOVb6jVU3u37Dg1ff4JWrAREL9px3Vj1FiJaNMvlJteMsRZocetVpNzk6VzIqGadK8UoRFg34Zu/u3IXFbZX0axlZ80iwQFkusqfi/TJjVklZfG1AzbkFsqDqURdAC74oonWmdUZPCLWZO/h3z90rPr14sXSCVNMCNQHkWBJjnJhLZsozUTOZTBE9Z1ucyVIhUKQw5xlJQszmtON9MN/0Guh6H5Bhg+aEy3uCKJoz6h3qPOSMiyBFcLngNOEwa0uKlVrl3vEKpYlQaw1kq+rAUtqs2VOVWmYtzkouLW6tGdnzJHvm0V66gi19LHVyRFj4g6XyCo0/cZG8FTWOUikmyq520nDvirhXrHg3tGZU7Xdm2848bwsm59l0HXM6W76Kc7W5yzqXmxnCWfOVJcuTUuLu7o5Hjx6ZVevTG+4OBxNjd01BwHRkx3G0DGu2xrNhGOg6C/5C8EhV5XBO8M7XJq9zVeNsjUu1ujxWZy4LkgEePnxAypmuclq9M6rOPM/c3Nxwe3tbr+84HA7M03y/WbONT969VJS6CG1UjPZ4sID/kpO7YsX7IWV4cqccEkwRUrGLKWUYL92LSWGh1UikWKJHS6KoJXlKzmTEqil1IVdStmhSlDTPjPO4LCRzSgxD4jOvjQQyvctcbYSHe88UIWbHnGojprOCoUjBuakmqBxJA6XK7RXMjCFRSMUaJX2GnAuu8tBjKYgTkipWE3WkIsyxLhJ7O0e44Ol7z9D7Os9e0PS09ZZJ1ZaVuuANFuGiuCJIybUfxUyJQjhng1dmwcuDF8rIEoIRyKtV3inN3JzuGHPkdjoylomJSJKM+MrNKw6XLTObyBzjiaSZ02T6jk6cNVfkqoPXbXCbrXUhbwZKvEZTYr59Sjzd4WIiaiDJEckZzwnTeYTaqVJLIy0T6inqTIi6JnBBkcoFbO+tuLPZg/PeuIYX/s/lHCXgqr6mYAMKJxcOKlXO54KHl+vnlbWgVZbMLgVxVdarRCQamT5oMS/qFSs+BKUUDodT7UJuXNjq8uM8OReOx9O56lEbu7z31j2M0IWO4D3Jm4tQzUdSsmU2tRTeeusrqCrX11e8+soDXnl4zWa7wXvHMPTknDkcDjx58sSeXQPi7XZbn3dN33eUfbJSfeO5iZy1aqvjXqucjOPI48ePefr0xly6Rgtkr6+viDEuJg4+mJnK8Xjky1/+Mk+fPuVLX/oSb775JqfTiRhnSj5zYRfJrWVM15rKcr0QchcXMFVdePcrVnwY5gQ/8CYkgdlbU1Yqar0h6tCSCQ56B6KFeZrJUqqRyUzJ1vwbY4cWYRpHToejUW52e1zV3BpPJ57e3ZBTYjyNjNPEfrfnC5/9PAMTJShvXAsx9tyNmdNcOEzgROlF6UQJLrF3JzoXmUrPk3TFVEyhKLtAEUU143LCl4KLSpoz4hJzTEwpURCmoszqiRo4JuEwWnb2qhOuNqZWsN/1XO87EMcpJiQmWj22yWE210zzxAY0WJRbUpX/atrZpmzivFuaRVe8HHixjKwTUDERci2kkplytaLNiayFTEGr3h0qZ91WseaOVBKSpUr5VBmfUn+wzrIrvvOYJBBo8EZ0n2tGiYiEERe6qpPpcLla4NUuTCPktvYVW92VVk5pGRYj4wBnYwZ7wjljc7bifCaovFA2kHtLwXMW95z5KUundPvXJkmpJRwASs3MalmzsSueGy0j6y9UOeAcSJY6ec5zxDu36C0atcCfKQbNhlZk6VtUNZ3mpJnT6cTNzQ2qhcPhwGk8VY5eWsZJa/bSi0BRVRnHkb7vUVU2/WC0g0prUNfoPo5FVxasMSVbFraZGrRAdp7mhSpAPf62/8PhwN3dgePxuBgiPJuNbcd2DxesoGc+YbSauVvD6zphrvhw5AK3R4vBygB4WWSySq3MiagZE2gzAyjVRc4uRW1uLLXCkVMExNR5AFTJKTFPEyml+nsf8eIoKeLJdAKbXthvPUWxpqhqne5dwQt0Ttn4mUEsKxukMNemacWhohSUXOz8sLyPC7qOqLMmMISCkIowJRtnO8TK/8EufTCqXcyCr9NyowSCqTY4EdCzRKYuQe05kHX+fK7gXvfLik87XrBTQchamKtD12me7BInpjyTKJRasL+cJpqIesyJnE9Enzgcj9wd7sg5s99dWSOHgFdzEbLfcaC1YPh+Q0gJ8R19yiDm4KOAeG9yXlMxHbCqlmAOJ5CyIlXlWauRtKiV89Fixg1tEodzEHwRyL7LbnNp7tLl3Z67n+0EU6oebqyWu3GeKPMEKSIlmQ1hzUr13hOco6hDSiZfNJWtWPHhkOrOZTSBximdponj8Ugphb7r2AxD5bR5QlUTaJxWt5gQsDQ/qQiC1tL+SNcFjkcr96sq42lcXLQeP37Mm2++ec8zfbvdAnA4HNhsBuP+bbcE79kM1kTlfaDrhnNAjasB8MThYPtaaAsIp/FEnGdi7CsX1prcYoyVD5so5Syndzku2yITWinzYgIUWbZZH2khJWtD886bc9mqWrDiuSBkAt73bLY7XH+FccpMQnJ3tWM3OEKwZMfdMaGaOcVUEzRnbrv3ju12QB8+QES4vtqx7TuKKqHKWFU2n5knAClHxnlijHA6zZxOiXEqpFTq2KjGQ6JL5bG0bRfj1/nKgzXdAQRTYghaCJoJJRNSIiD0omwHjy8deepI245SHH3vCcHscYcOtps6ddaqTFskLkYp9aJqY9AczbQ6kdWEVa1ktkbsOqI/6S95xdcJL6RaYKuxwmEaOc0TN+MdT463nOLELIVZElnUwtk2SOqzFWWaJ+Ix4cTzaPeIB/sHbDc7ttsdrzx4BVTNfajp1QYBDRSf6bfmKJJzQkJHt9uTU8T1PfN4NB1WjMyO78jqiBk0KSEW1DUj6UbGyUipPZOl6lhecAORs1uRqtYJ1y+8WMsynzPNwCIlVkohTiNpnsk5Mx4PzONIipF8OqDTiKRI7wRf9T4H7wniyFqISdZAdsVzQbHsSAgmD9V1HSJuCWCd8wzDwOk0cn11xdD3i3xN3/fmKtR1hC7gk194taXqy7Z9NH6qqvLk6RMevfOIaZy4vrpit9vy9OlTfuiHfoi/9bf+1jJRAmw2Wx4/ecz19TX73Y4vfP7zXF9d0fc91/s9fd8RQsd2u8P7UCfvAAq3t7c8fvyYd955VDOyIwjc3DzleDqax3p172uZ32mamKaJUrIpD9RsbWm8+IusLBfT35kz26oqdj1X/jHAPE9rRnbFc0FViGWgC3uuH77KsHtI3we2u96anaqiCCiHY+TxzUhMmXiKqKt87CD44Aid55WH17x6fYUTx9Vuz/VuR86ZJ8HXnoqCB4ITnNhce3u4Y4xwc5t5elM4Tso4mUQXYoFr5iyf6TSTqJlgMuIdvjMNdq+KLxY0B4FeM70mhjTTR0+nws4VHu57YlCCbnB5S8mZ3TYwdNB1sN/CwysL0vcJYnKXg8+uSqnNXoUYMzmZWkkTB7LxetaojsWoe2sg+/LghbVjzPIuEWtW1rKzkeRqV2PLetTMjl6oAuSSmWPEkZnmuepfepPGqZaWWjlylgGtTl4Izne4kMA547ZWFQXfT/iSrW2rSnKonJu7Si2D5Oa71wTg2wVz9WpZ10VPUs86k/bnWTT+bKLQ6Au6XBs3NlNyMkpESkaLmGdyjJSUTMFaM76S1p0IwVlGVopQXC2jrFjxnDCO7Hmx1br3U7JGLxBSnRAaD61xy6zRq5mFwGWAB/bbz6UQmxrAZJneEMKSBbWy/pGnT2+qnbM9e5om+sqhTTFytd/jREgxWgWiZLqu4H0gBLUGDm9VmJZRnqbJMrKTNZLNc6zvLVUVlHzmotfb2qgAFxq6eu//c8MXwjLmLf9cH1P5sRkW29umvrBixQdDUDy4QNcPDJsNmyFwfTXQdX6Zk5rByBgz85wouWUdzwkTJ0LoAp3zeBE2Q08XPILWErwubLpGC8ol2xwdhTkW5rkQo1Jypc0J9xZvtf/sHv1N6nE4jzVaVcUAIyeoBbda8CXjsxDANGKLp+vsUhyEKo3lnNEFOw/O21jzTbagHYuqGTzUQFtzATE+sKufR+O5W7Ks+ovqGsi+TPjIgayqMkUrVxzGkcN84jRPi45sUsiqVdHj3HksGI+1CYxL5cSdxhPvPH7EcThyff2AVx68wjAMuC7Qbwa0uKpR2QildlIQBAn98noh7YxvFzrr/PSB0G9xYQAXQDwpFzSle9QCh+LF2ELN0xnFAlEuy4/3B4cs/50D2PpASo7kGpRPxwPj8UBJmfl4JI4jpSRcivSYJW8JnkKPA4JzNEGS7Mo6Jlc8F5rLVdf1DMNA35uv+TwnREacCwzDiZwy281mCfBa9aFRC7quo8+Zvh8Yhg05F6Zo/DuFJRAV4O7ujpubG4oWpmk8mxDUwLPREFJKDENPTIndbsduu2M8nbja7xmGgVcePGAYerquZ7+7qsYDRjUA4Yd+6If4yle+wuPHTxYLXOeEp0+f8vjx4xq4piVY7fue119/ne12y3d8x7fz5MkTTqcTT58+5eb2lpwTp3FknKclYFUqO8j+q1SDM7UoZ6uyOBHGcVzcv1as+CA4J+x3PdttR9c7QhC6Ttj0Qt8JMSlTOTcEj7EwxYIv0AWPq8Ho6XQkeWeJDrGkR+Oi55wZ58nU65zgqvuciOd4Srz95MgYHbdH4TRBKsKDnbDdBLzAEMSCyuIJUXFloGigzz1FPS4YxeGy+RFqFnSakVKYfWEeQIOQ4wwlI2Q0zaT5RE4JLTNxOpqZSI6Mh1vECTkVUrrMxtZqyCUVqJwpQQvkfG2Br1Vcck6seDnwkQPZooXjeOL2dOTJ8Za78cixzJxSld0ScwqpogCVZwMiDu9tBeh9skA2Kzd3N8yjdR4Pmw273Y7tdkvoWpnROHzOO1SdBbGhXyYVDcai753HpR1hnsEF+nFEfIcfduB7Ct4cT8pcVQts6RmcYwiu8n8a45wqY2K2sS2j0+AWfizLxAfnwDelmTiP5Bg53Dzh8OSJZWZPJ/LUiPRKT+Xkdj2EAKjRf7TajWphHZIrngfiLIDbDBt22x3DMOC9ZxyNu1qqGkjf92y3u8pfdUt3vogFwsNmQDFO626/Q4HbgzVM5VJMKst7Ykw8evyIt95+m2meuPuWLyx6reM4cjweF7WBw+GOEDquvvIWwzAwDD0/9PAhm2Fgs9nw6sOHbDYDfTew3++rBew5kP3iF7/E93//93N3d7e8X+c8X/nKV/jSl77E3d0d0/Ta8h42mw3f/u3fXju/LSg/Ho/8wA/8AD/8wz/MNE88evyYOc5nFRLOc+j9hhFzG0ux0oZyxgHztKoWrPhweO945ZUdw35gu/F0g7AZHFcbx9A5TnMhJ1skpZw5jInTnNkH2AyBINYQfXt7Mpku8fhKuTPJKbMzP4wnc+JC8GpBrPiOJ3eRQ7lhTp63bzuenjx953nj4Yb9rrMsb/BmF58y87QhpYTLwmkSk1twULM9NvepmBNXzkzHkeIdvQZGn8hBiPNodD3NlDQyn25JMXK6qGy+/aarDWeQUiZVO2i5/L9mi8QJXd8Tuq5WhawprvGGW7bae3tsiquJ0MuCF8rIppxJORudIEWSJoqav7nZETxTlNRzMNsksUQEFaMnnNLJVp3jycqGzhFTqr7qjqaX3gTTzVSA2skIIhkJ3RKIhm4wDqwLiAvQtPBqR2jLyKoqDqyxSu9VNmxlV00amqh7K03aPi85cpeTYXX/quXNFGfiNKI5k+cJjbOVfrw/e2c7O76Wvpa6GnXicPeMGlaseG8IpvRxaacK51J4jIk4R8v0p/xMhcHGlVQJLO89PvjaNObr69jvWZ0ipRBCqNSCkXnuSTEtFJz22JRSDWpPeG8artM00fcdmjNDDWQ1ZzabDX3fM03zuzKyT58+4XAwBYLGd21B+vF4IoSwBOyNx77ZbAghcH19zauvvkrf97zzzjtsNhsAfPBnaQLVyyjW/r9QILEFqlVHMmKT7josVzwHRKDrrEu/qfZUPxzTbwWaQk5phkJZKYHKE9eaHImVy5pxWCCbksN7M0GIqcpX1eYpZ2LoxFyIYyLmwhwdKYvZ4nbC9bYGxD4gzpGyOXVKDMSkhFzwqNEPXKPayFKOtCxotgpsEnJMODU1BRsv1eY6JetdyRaw2rmnVPoddv8SyDbZkBYvgPOezXZLN/T1s6iNak4sa+1d1bG267Va8vLgIweyWQt304nDPDKmyFQSsxay3A8EgXdtEKkZ2c4zbAbjsEUlx0zMkdvjLW+98xV22x3XV9c8fPAKXRcYdADLX54nmUWNwFZnKatRBwoU58GbRl2pHRxaMqQZpJkm1AmqOKBYl2cRXHGLnFCMdpIIIVhDibglQKjv6N710mWZCylGYm1MOR2PaMn4lHCacTiCEzrfsrzN2YhqOK1LQ92KFc8DEUfXDXRdX7VYq+TWXH/DPjBNVo0YK980+EAXfDUkcXXR1rqkLy7PyMw1DvnxeOTxo8eUXLi9veV4PDJNEyLCMPRG8anNkoBlhkuxpg2F7jSy3ViFYjOOeO+5vbnD+cYdLGhRnj59ypMnj5mm2VQWggXYb731Ft/3fd/Hbrfltdde45133rbmsetrrq+vKcWOaxxHUopsd1ve+OwbjOPIGGeOp5PRf+ZITGmhC51D2Hq7RQg0KaJ3FTlXrHhvCKgUTjFx92REvWfTzzx5Gum8Y86ZcU7kojw+FJKaSYALha7LdD7jteDKDJo5nEZOx6mW2rHEhwj90NP1HeLEKpidR/FMBGISYnbE4sjqERfYbXteeTDgnTAEITghZTiNjpSUm0mZKnc3AxGT02pT+jlo1nopuJDwXpimkfF4IMaJ26c3PH70mDjP+Oq+BVZxNLqOMs0z82y1x+CrlnTNtnrvCc7Tbzbsrq4saJ/Noc85R9fZoltLpuRYlQ1WvCz46IFsKTw53nE7HTnEiWOaSaJkV3mxF+Txe23AUH/4Qtd3eAloUU6HkTmeKDnz+Ok7CMp2u+P6+gGfeeUzpjuJrTLtJbW2LZqkVsyFkgtzMkkRLZiVXWeBblELLFumdOHA1uvkHTn7SqgHmYzg3hpXQK1kuzVP+S539zM2epG8adzWlJmnmThNHO/uuL15iqiy9cLgBPEQvDB0z3wNjROk4FymPJMFXrHi/eCcY7PZmsmAD4h4SomM40QuJlPXdSdSypyOJw6HE2BlxXlrWdBSyjkje3FxtRGxSC3jOdNzvL295ctfFsbxxOPHj5Zg1jkzQMg5n9UCKg/QOTMSOZ1OOHGLFNdmM9QeTPv9z9PM4XCqzVy5SmspIXTGoXeOH/zBxOPHj+i6wKuvvsobb7zOMAx867d+K9/2bd+Gc45Hjx5xPB7IOZtiwn7P6TRyGkdu726tSS0Xe31MtojWeMJZlMuC3DNn9l36sytWvCcU9ZnDPPPWzZFTUrwIvVjDFk5bwZBEIGoH3uE7GDZC5zOSChJHSi7c3Bz48pfftgpLvfjg+dzn3uCNz75m9KC+Y7MZyOo5jh2H2ZOKYy6eTEB8x4PrDZ97bUfnlH2n9N70Yacpk3Lh7VvhdrTq66yOWEDVDH+0BbStOOoU7zM+KN5DTkcON0+YxolHb73Nm1/6MnGe2V/tuL7eL42aU4yUXCw4P02IQD8M9J2pqPR9T+gUCR2b3Z5XXnsNzco0zqRoi+9+6AjBE+eJu9uni3vfipcDL0QtiDmZFqyWxabWfthy+UDO1ILzZmvOsmBOxcoD9pBCTJHTeAIR5nkipmgTYStVcPaBtuc07TvLzOYauV6qeBStfs2VWlDaUnIJti0b6oqA2OryHMjOC7WgdSkXva9esOxo0b+rigeXZdaUEFWKmPqCNZnJkum6/GxbIFvqfi+5uStWfBB8tWu8tFUtxRZ6pXbxX5b+L7v7y2X2X5pd7Nmm1sqVei9Tm3OuagLGjW3KBe1YGi2hTSwWzIrJ8pXGPTdVAhGWLGwpph17e3NHjKm+xv336b1jHE+UkpYsj/eOzWbDgwcPGMcR5xzzPJuLUil47xdDhq7vCCGQc6lSY/baSxaWFsu+t8D6OlmueC7U5I0CUyycYsapMFVam3jwXZ3UvIfFZrWNM+4toGJKnMaJOKfKSY9G85ljzVJaZsV7jxZfK3uQi5kBaZ2jnBOCt0sfCoO3xBDFKA99ULxTbF2nC19VlwFyYQdbe0VEFMTUF1oDZkqm7xznyCYlk8Nri9vc5sdsFRERvM9kl1FcNRAyOO/pup7iS1UeMk5sCFZVKiXdU1pZ8XLghZq97uaRU4pEjFJQXKWvVriLHzvtaumSrpsCaBH6oUPLZpHQOIxHshZu7m64uX3CZtgQgmcYTBez8Yya+0muk+fpcGCaRkrOxHki53gvw2OD0NECzVa2vG9soDRjg9Z9rapcXe158PAhXc2g2jEFW0y7yivCRNxtge0RvLmcZCWmjGghYhp8qtYaZkH8had9/fwUwDuKgAsvrJS24iXAYtcoxuFsTl6laTHWSUGp8nfzRAiOOQ4LBaB15mspVvpLJmnVdR1XV1cA7K+u2O/3gJgXfJ2sbm/veOeddzidLItq5f+wLMTapGengPN5oHHufUr1mMtZJeA9gkXjyCVKMcmunG2x67wHgc0w8OqrrzJWqsLd4cDjJ08opVQ1h565etW3ao8/XCwWn4lkn2USt2NYqQUrngd91/Ot3/JtXI89bK85TJ3R36qcuZi5FYhJ2+V8NJ53zuQ5IS6TxxNxPJFTZB4jZJOMdAhBpDYfztzeHBiGnv12S98NuOIIk8dUau05VspPvPnOEdVI52AXlM7VALRYv8vTk3J7KowREs16XfDi6L3DK3TOTBGcVB5vzEhRvBeurnb0feDhwwe88uorxHmmH8Iy14Vglu6lmHKPC7bo7ULABztvdH2g64zWZwYPihMleJAgxjUWM0kQXVrMV7xEeCFqwdPxyFQiM4VUg9jLTKktjGTJVN6bHGDRxVOFgb5yXArTaeb2eMsUJ57cPObR47fZbnZshp6r/da82MXhxIMWtFrWzvPM4eaWu8OBlCLH8chcs6ntIjjzjxfLJJlWnzmQLXyjpWhClTYZUYVXXn2FN8aJYejxznN9dQ3YArqtTL0THAEz5wsIAdSTM8TZ9O1CKUhwleag4C7kyFpmthoseFWk8/czZStWvA9axz5QKTGtQatUI4DzY1PO5o4lsN1slqxn0bNmsmVJIjll+r7nlVce4n3gs5/7PG+88QY5Z95880s8evst5jny5MkTvvzlLy8OW13X01WjgvbTvtRfPgeFxbqkK5f2Mgi/DGTb+LD7zKKzsfVEYJ4jp9OJoe957bXXOR5N5ufp06e89dbbaCkWhO92NWssbDabugCop8PL2FQv9GXbMbRqy0otWPGc6IeBH/UjfyRPD45+HzicrAFrjIVcWmbfFG/m8Y7xcKCUiE+FPJp26ul4x/HujpwS42mGXCxg1NoQpTAdR57IUzabDa+/+hk2/UAqjs4roVrLUquT45T4wS/f8vYTm24GpwQ5B9UiwpSFJ6NjSg71UILN2V6EHkcn0EkVMxCgWF8IRQnB8+DhtcmCnQ4cDrdM80RJsdrrQug6+srL7/rAJg5cBgoijn4wpYK+78ypUOpcGOw4REBcxmw8L1m8K14WfHRqAZCqdWppHY3LfboEsffKce/D8xSoHchNd7ZOvs6MFmKMhDAv6gFGLzAqQFtdtjK+dUZGYopM88w0T+cJp9IAvFgg3Dh3OdfsTxVOt0nR0jCXfvGbzca0I1Hmqg9rfMKLILMF7VT6hJx969vwXAIF1aVsssjjLhp9bgn6nXP3Mt0rVnwQ7tmw6tlW+Vlo0TO1oJh9cinOuo9rIFtqZla11DEaFmmr/dUVKUW6YBJUrau6acdqlaw7GyucA9H7yh9Y1nPZr9ZLqWP73e8PlNK0/Yw8BGA0pMkyq5cUh5zykt1tn4cpgsiZA+zuV0RaJvZy6MnFMn2lra94Xjhx7Lc7UhGutnZuj0kRn8nlrFaAKpIdxRcKhSAFR7VOv6imtN9tMyQothMEscRMs1Jf/tkYqctAoFAKzNH27QXiEshWvVgnzFmIyRndYBnDLPtthpbAxRhWiti+fPCm2NAHUxtAiVSNV23SWZaFvaQQ2Nivya57DabtMGzh6pyNWLm4tLhgpRe8PHiBQFaZSqUVqMltXVrQPovLkLY9Rpp8gYB4wWFOVubnrCiFeZ64O95RtFSuah1Rev7Roub4UapIe6wB7M3dDcfxuPAB28QanDWt5HzO/JRGM6jBbpvUWrOXqiK3SsEyU13neXB9xWbYsN3t2G+ty1K9LhI9Xdex3e1MDPv6mutXXqHkBHFizhEUZpRZC04cvTvr1Da36IIut1es+DBYMJkrP60s48U5myycmNSbFq3uW4cajAa2m4Gu60wep1jjkzVuneoCThbqwn6/49VXXyXGyDtvv4UPVi48nUaePH6CYuOp7zv6rqML3ZIpvgxsl+OGZZxaENvGZak0n0ZJOAeS90eFjVfj646UUrg73PH06U1tFgl89o03UCz7vNkMxJQ4TifGeaomEGEJku+FsLUM6sTRd8HOH07o+4Cvnyt3p4/1e1zx6ULXdXzujc/x4AE8eABTtAblUyzW11Fl6kopzCfP6aCUNENJkCZQJV11xHhlBgSnkflkjV/NMbKoEoGo5vwlznF3OJKKMJ0iZcpocYQ8M2iPqIMyULSr9DthlhbyAgJZhblArr1oXmuehWIXzTYSs1nQ55iZfSI7K/133pqnu75nf31F1/cc7yDOMypK1/dLRWRTqkxXLtaIeRqXZJirQazDzItUFOfsOEWwzCyQXeuZe29O+4pPJ16AI6ucSiRTSNa2AdTMZ50LXMsy3isiVmjjyNn2VlYvpSAecAWVwjifuL29oeRsNIFLBqliQWwpplmXMmmOzOPIaTrx5OYJN4dbs+eLM7mkxf7ViSxNYtZUVUy2S023tfEMW4OMOZmduL27IYSOrve88uABu90OFDbdFoLgFlNe4955J3Rd4MErD5mmIynOHG6ecDqaEsJUCmMpBDFfbDp3bhJrWbWVibfiOaGqxKrlmrNdhxDoe5Pjcs4vWclpmrm9u2OaAt45+t7ThY6YMzEm5hh5+vQpd3d31kzSdXRdjw+Bq6srXnvtNWKMfPlLe7quQ4Dj8cSjR49wzkqCfd+bJFANZrWO22UxWxeiqDnyZM7NXvdK94IFlK3hZfmv3l1PL8ZpN9ev25tbHj9+zGazYbPZ8PnPf+4iq2xW2KeqO+t9oOt74ylepmFbgQTBB2/82s5KnK3UCcBX3vmafacrvvnRdx3f+vkvUApL5S/lwpgskI0xMk4TORfmsWc8eHKOnA4n7m6P1ozoPc5b9UNywdWFapon0jyTcubR4cST01gVbxxPbw/kUpiOE3mcKXi87kzKUjtUPaUMqApZPQVZVICy6nnMCQRZhBWwLGhGSFAKqglFSVqYyHgH9I7gPeJg2AxcP3xAnGdyThyOR6Qo/bDh6urqHoc+pkTJjzkdRxoXSkSWYFZqtrcdjMMoBl4gO8Fbi9gn/h2v+PrhhTKylVRg/+4J9j9bkHvXk98FAZYuadrEVMWhL/zT277vvZDq8tfCcq2TdS723JSNCuBEwJezalZ9CXtsnTydWwJtVSvxgJk2oFBKZp4mpprJiZVmICIUQnVWaSyDWhJxHucDrmTUOTthKCSFpAVUlhBYqXJhLYjVtTt6xfPjklJwphVcBIEXZfycsmVxKqfVOqIzc7RKRPtt55LxGmoAqUtmtlRprQbjt6ZaLpRasj8rHKg2nt45p3rOst4/h5xHQytnVjxb/19OGPacRt1J2ahJIQUrb3bdYqLgnKeo8fh8CPhgi9t7uNhhG8fNZMJ5VxvZVjWRFR8OQRYKTqu3pVzwyeQV51lxYvSXjoDXjpygxMip9pF453C18uG94usCMHkhe0cqhVFhRiwJUs4LweCFLlTbc7WJX51QgkODo6gDdZalFTVXzlKzmnKex87VCqs83rtQUFcVhGhzVh2/lUJQauOnYMpDTs561Y060OT53vNzrM+BRmGUZds5c3vucVnxcuCFWuGz5CWUXWaW5ffXhJrPs85lk4TB0X6vLfQ13kudbMRZCXROzF0iZ0XNSBpVkxEpKouLGE7oh4Htfk/xsD1tmHQyNxSXIVX5kAtXMeddHTyOrjZ22Aq0bi9S5RiUEgspjpQiPHn6iC/+8N9lO2yJU0KK+dMPmx2bzQ4EUozkHElx5hQzUR1RPbMGJg2kLDw5RcYCffAkhf2SncoYCVCrRdo6MFd8OCwjG20BdsEFXSRyGj8Nc9dyzuSqjOVjbljzHE1GKyUeP3nM6XQiplgXe9YdfDweONzdWOZ2mmzSW4jeljnabHbs9ztyhmGzJXTV/KTyZy1ArJw275DgzII6l+U3b9J87eev92gGS2a37lYrc8/V14spcbg7kFNmu9kQnBkoSJ04i1O6fmC72ZJzqfQIzueAiwBWpGZktxt7rRDYbjd0XQtOvv8T/JZXfLNhaeylSVhZxTKILgFgcooH/KajC3tyTtzeHnny5CnjOOF8QFyHiNA5sbI9VTGgNk6/8dnX+PywuQg8bXF5miamOaLqSNqRCSiexJZMhyLEKs2VFWI+Z2aPUyYlo8CdcqwLxJlDHHEl0klmcBnf5tXB4YPgXKEKMeBQvHMUX6Uk60CzaqjNeUu2tyaRFtqO2ntAlb7r2G43KBBzJhdr7GrNZr4UQi90ufa5rHgp8NEzsqJkyTVjePmDqaslqfyV5Qn1quhiHScOwC0ctPqXEcCdqRKUorX72ZqyQFA9O4sUtQms1BfsNxu2QPGwOW3Z6EhMjkREXa6lQgsQnRe63ibPe+8NlsKEKIilRxkPI9N4Qgs8efI2Lit9P1CS0slAP2zYXydSsZJMnGdiNP3K45yYcCQ8owZOGpAC8ykRpsTQBcQ7ilSiejZdFtHalbqOyRXPAQtk5+V2u2780qYOUGoDVy4JV62gpzjjnDBNc3XBSjx+/Jjj6bgoH6RsQejhcMvd7Y01TE4zpTrsqFpzl/ehGpo8JGfYbHZ03VBf51w7acclVTrLVTUPTXWBLArO7JovSAVLyd9uy1m7ui5AxVml5PbujpQSrzx8Be/8YsOJOLyHoduw2e5IuVxweNsO7CI1W+RDYLPdsN9f0fUd+/2efhg+ia91xacAWcvC5wQQUYIrVpp3hVybnQkBt92TS+ZLP/wmjx8/4e7ObJkRa/Aaes/Qh6qZbBbP/dDz2c+9wee/5QvWwFh7LlQLc62uKEIqQlFnc2tUUrZ5NBnVlaJCzJ6iwnHMvPVk4jhlbk4TX3w8E6fah3I8QZ4ZfKGEhHdKFzqc782Ny9n8JSgeJTiHOtO4bizWgi6VVpOhrHruYo3SDS1m6LrAbrsx5/lkJiZwbjoLJdENRglcmzFfHrxQRlblgux23lqvL8qFreV+mYNkuW4yPO0pKiaPZR2ZF3JUevmaC62ururOJUqp2VwbyA7vPMXlarl5Jr6ZuLM8s/1iV9qyxFr5tC1Ta3fmbPQCCozjxOk0kgu40COhN+mSKgmWU+I0RaY5kWJmToWYrWxTskmXoTDGRB+97bOUqhEIrUN0xYrnwWUA265bma9RCsDWSpbt1KWpUUSWSc/ctNJCUShazl3/i6pBrqXFS5UEMxxpAW0r5Xvvl2bKSr87dyIvY9FRXDmPNb04j9znFFzcaueP6sonLevDIriul7pjS6bVLRSBEOx8cVZVaNz+Zz9blsW7XnzGK1Z8EEwTOdX5zi2LSuqCsqkSoOVelcQqhhal2ZxXEIRczLpcqbqzamofzjvjcIdACI5Qpa18cHTB1+DR0YyB5tmcMC8DWVUhFgtkvRMOJ5PAG2NV6rhUQ8mmUFBUcRcZZ1fL/cHdfx9tvF9+LksgWys0pakHyf3HnBuxK/XvXqR6JhPUguoayL5EeMFA1q7vZV0b902pEhjg6r9aCFnMDCxD4s8Zmcqh8+oIBARh22/oQmcCya17DKpYe+FZR6K+6/HOoRR2/ZYpjAQVkpuXMqblcMGLMz4SrcPDcMmXvXgKPR4NA8UppMLh9sjJzQhvMh4hhJ5+f82we4CCebmPEyUXTuNp6aaepok4W7OXk4KTwtA5DseR621HFzxXQ8/QeYJz7LtAVyVKVqz4ILQJs40TEcg5Lc5WtsizgDWEQJcC4mQJSM02crLGk2oG0u5DmwrCBW9VLnVnldPpRFFlV8dlCNZEtdvtuLq6qpSFuATNzXe96zxDVRfwzldpMFdVRUzB4F5UebE+tqytnVN635tCQnXvkrrYNUUT+1z6PtTMkedqf41z5uO+3+7Y9L1p7l5SoIpSRInzzM3NDdM0EkLgeDxeUAtWrHh/pJR4+62vLOog3jsoGU0zqKnnxJjQonSbnjBsIAQePHzA57/181wfR+Y5MY6pcld16ak4zYWYR6ahcDyOjNNIXwLBbwg+YAvYmiBxnn6zo+sHtBRSTNbMLKA1wLXA0c4lTw8zXjI3B4ssv/ROgZzQXMjF16qOUoq1fItA5x1D77nebXjl2lR7nL9hnKyasyjzFLOZ1aJL4OucqQmdZTOhpAQFYuoRcXRdTy4F5rmqEUHjx5sdttAvCiQrXga8uF3UOUlpK6DWZahlqc+LWEZRRGqW1IJUy4QE+wF7jw8eFIJ6OjpA2IQNfejofDBNVrCJqU40LZDVWrbp+w7oKFrYdhvGsMEXIcoEi1ByDbDF0YvHLYGsBeBZMy61iUzO/FTxSOjRosxj4XA8oiqMR+XROyPiAn64xg1XKMLhcOR4GtGitWEt1Y9MFo6QkIBMHxx3d4GrrWcz9Hz21Qdc7TZsAvTBEWQNZFd8ONqYgHODhhl/zEs2o20PXaDP3WIzaw2LMMeZaZrseTHe49pelELONRfVZeF3Gk/M1YihqFYh857tbsv+ao8fPYfDYTm+EDwhmPRVPwx0nSkolJJJqVnGyrIfe14drUvWRhe9yW7ouNruFxtawRa9JRdzHMLBIISux6NcAcNgnu773Y6h60nOgt60aHeanF6MkZubm4XDv9kc6FbHvRXPgZQi77z9FTbDQLne03edlUSSSUrm+vssqnTBuv2dt0D2c9/6ebNqvh15enO0oHeOxClVWcqEjjN9zBxPJ6ZxBO3Ybjq60Ajk3tywuo4HD6/Y7a9tYZpnyAmWfhFXs8Nm3vPoxpPSxLbPTNHRObMj06Lk4iglkKW6BtZzQh8cQ+e5vtrx+uufwTtPTIXHN7eVytSyy8bTn6cJxEgF7aQS62fR3DdLKmyq82DX9ZASqlSpQWsKLxb1WvNY59eM7EuEj+ksLEuAeVm8bw0ZHlnctLx4vPiLoLbyz2pWRBWTx/GWPQmLi9dF2Y9zea9Nqrbq5N5jav4XX58fxNUQ1h7vcfURbbDbdqeVh6ttArX35DDejrbGmUqtyNl4vDhFmHE6UxRO48w4RhtkjZguVvpsHaBn4ebCNGe8U8AxzokuZEQhpkLnVzmRFR8NrTTnqrf5MkackLNbuoRTqoFvNQ9oFIJnX+uyuC+cKQJgWRZ45vmqONe6/P27xuiZYlS3tbK/07OSQA1a33duqlwF249xYUM1ObAu6TNcox6gS8d0O0cYN+98PG3XNEpBUYpmBMgpnU8OK1Z8AEoptfEwEZyQug40IzkhLZBNpmYQ5pk4R1wwSkwXPLkL9H2grzatrYnKpBodWkwOzigy9TfM5fxYaIYBZkIgsMxxl43PlcRWG5y9F7zDLmJUt9aU5Ro3/SJiPJ8PagJLnDWiLRTCc6WIxo9VO8eUc37nvvRku/2MSQt6oSVPM0SgOmWezzMrPv14wUBWaP0Y1owhJu+BWbV6sSDRu47gOtNndcEaLuqk451fOoKbw0cojk7t0Hq/oXMDw9DTh0BotrZJbWWoGS3VG161uqQUcoy4AkEdENj5DV3VaF0cVETwxS0nBCudgiuCzywDifq/U0WcuZhJb6R9LeZ8MmXQVIhxIh6NIxhjIqa8zHXWCGOf29JQogLqmIuid5m7MTN0hTkK+83EbgjMc+Jq27/YV7XipcBlt/KlLFbLuF4+LqVIjPOSGe06v2Rw04ULVgjhgttadZsv3MKsXLoxKa8YmWNkCoHT8cjxcGCaJrrQcbW/wiH0nZX+vW8TriwUAIouj1FvFAMvk5UtYQlma570nJotdv8QAg+uruhCx7bfsKvUpH7o7T2GQN93bIeBgtEwUpyxzmeh8x5RKJLPtRs1zWlptCkEzYU4zWS3stdXfDiOhxN/+S9+D5tNz4PrK4YuELzQB4d3siyWALb7PVePH+JDIOZMR8H3nv56YD/4upjSyvsW02juAl3wvP6Zh3zmlT3OCZ3PlPnWXrRUFRzXo3lEc29jOc9oTuBsESeVf0qlEuWSzXAhRrRkgoPeO3wXcKVQsqOTVDXjCyZ9ZWO0FNN+d9mRcrqn2uCcXeY4M1fnTC4WxHKRuDKqn6mxHA9Hnj65sbk6TvRqDec4U1NxAbphh2M4qx6s+NTjhQNZ6+oHY9g4ArKYDjQ6QHAdnR+q0UALZN05O4lZ2YUWyC7UAvC+x7uOTdfT1dWmCGTKEsS2QadVF7PkTI4RyUpQh8ODH+g01DJOPpcpa8LJJui6vchZ/oezs5bDsliq4PtA1wVKEfIxLVa3hzRxTLGqKdhnRH3e4mR2CbVtkmGeE45MFxzTmNn0fglgxzm92Fe14qXCZUD7rC3rZYme2nzhvdQGk/tlfFVdFpjPvlabeIMPbIaB6Kps1zQxh8B4PHI8Hkkx0nnP1W6PlmLNKEs1xp3zNDUoFYS+am7mVBZK0TmfA3buqdnhYhknxALgq/2eoe8ZQs8QBjofjH9baQx91zEMNpHP48n2qRbIBufBKVlkUS45k5HkrB5SDVRWrHgenE4nvuev/hdsNh0Pr/f0XWAzdFztB0LVMHa1WrHZ7dg/eULoAvv9nusH14TgkX7AXQ+18uBN5cM59vsrdrsrM+nwjj4IaGGeT6R4WhIygi34yBMlGzc354jmDOpR152buYr96lO1fc8pQs4EEXrvyCiiAS0OX2ZcsTYyoQWzWgPZiHNuoQAo1OYvwSVIaeZ0uquB7Dk73HUBX2k7OVsDaYqR4/HI7c0t5h2U6aQqELmMSMF7x7bf0Xm3xBMrPv14gUD2HJSJgKjU0nsdlOLt4s60gHv/9KIruFIQlpqlnvexyO60AfYepbx7HdM181Sy8XjODWgWYKs6ipR37csyQpdNX+fbVRDootwoNC3bJv1VFLPqXfg69sEsNL77hcqLm+e/s1qBxBWIWXGxMIXMOMfKdVqx4sNxGcQ+G5he3q65RavK3yv58a7Hvt/frjZsBe8Xrcdc1Q5itYvOKSMIwVvJvymKiJwzsVqqs5c0gfMmy9cKkmejFKPi67tGVDs2d+9cUysqpZiNtS9Lo1upC9cUIzGaccoiIr+cd6Senp4dxS0vvGLFh2NRBpmFeWpW60oIQpf9YhggNV3pQsDPZirSFD8si9loAGbkYfcFvO8ss+sdxZsiQpxPpDjaAdTpOqgQ40yIs427lCwZ5BRcqqOsqSicKy9N9aRZxIpgNvKieDH92yad6S7OP6ZnXakCjeogVdXAtQW0LBWfhb8fhBBsbDsnlAJdEJuLNdfXMP1YcTVR5IRQpTu99+8+ma341OKFMrJOAw4xfQEROucYfPeujKwgOK2ro9x4dLoEtzUuxB6ixJhJdfKTYHwXC1ATOcdqlFAbt7RQciJFE2o+HU+m3zpH4jzVgLbUsoknU5Ds6omEe0GzE185NsU+GtWaMG0TlrTqJ3NUpmRNLmNSpqykokQKpZWKMA6RxbRlCRaW6a8uAOwacGGZRA9TYYqRKWVySWz7tYS54vnRXKgaHQDenZltpTwBUslobFwzloitSf8INlE0kxIvbml02m93PHzwgNPpxJNHj7i7uyOnxKN33qlqI9ZpfH21By3sNxtOm03lvEVSLGjJiBZi5bhuNkPN2trkVKpMnvNVVitXS2mMxpNz0182SSNRs602y01lGk8cDrcMacPrb7zO9fUVMUW+9KUv8tZXvsLt7Q23T28YT6clIA/Ozgfi3HnsXvD2Si7vCuxXrHhvCE6CNQdPkbm656US8d6dudwC7u4Oefy4BoPeJsa2wHPUjGVH19ui8Gp/xdV+j/eOXd+zGUJd/CUQc5zsh47QBULXcxoLu6ujVUJrQOh8oNuM+M6yssHZdk21GSxnXMn0GtkScT7TdQmRag8fFVFh3ztTBek6BCHGCJXGVLIFzd4Vhl4r79az7XvAdN19FYT13uF8W8JamLwZeq52heAmgnNses8QTAXJh0pXELdw41eO7MuDjx7IqgWnntoogRBcoA/9ot/aOVMk0KJQOacUqZavLRtrpUR1BVcfk1O2LE7N7KozCkHLtqo0ojdQs7Apme7lOJ4Yx5EcLbgtOS8ZWWlBpWmN3CvzC26xzhMaX0Ix574WgDbrWLMOnHMiZWXKhakUsr1NSos55ZyPvZ9wbiR8qaVVayCjeESFooUcE6KZ05wYpxOdWyfMFc+PS4WC+xqvZy3YRi1QjM9WcrpPO8B4toshQPVNb/zyZiCw3Wy43u+thlEKp8OBkhJPHz+m857NZsPrr7/OfrulpMR2szGZq5QYx5mSEpoFrTbPw9CzGQbL9rp2sXJkoz8kLmSCsNVlu14uGD+9eCVOE6fjcWls21/tqwzezKN33uHu7o7D3R3TOFkQe9GYFmoJ99mFQCrNrWnFig+GVQQdWmCaEwjMyYwHnG8mQBasplKIVb95PGUOh0TJ1p/RmqiGoWPYmPHA9X7P1X5P8I6r3Zb9doN3QtdZVtN7z26/ZbMdCF1PjMrpNOGcMARHCFWFI0dC39txdB7xDk2zBaql1EA2MRDpfWbXJ4JXcirE6r636W2BG0IAoRqoWGOk1p4WL0rfWQPZEDy67RBR070NjYrXGsYc3nU419GFnt1GCTLTec/QObaDLQJC8NUuugWw76UCveIbDZ//PLz55nvf97nPwZe//Hyv85EDWVMEMEUBX+kE3hmdoAVnC6tNW/mfiybfSgpXQUVxpTlQn6W1rIej3LtoyajzS6kDmuh5tpJmLWeW6g9fqoD0IryuNbNUj68pE1gjWi3tlItA1nqxKgtAF5mRonbCSUXNDcXutsdJEyKBc/h6iVa3PGfKLHyuwe3y6VTKQnm/11mx4t24DFpbAPZe1ILzSf/+81rnsNRt0igw9Qd+meF1tVGs73viPC9ZTNQaqWKczWN9kQQzsXfL5lZb2+X1zvs8j/ezCPpCURJZnH/EGW2gUQ60FEpKZGfOXeqtelP03O3czhclN43asqg5NF5d13XVJtscvfzF59guzeABgCo5tmLFe0IgBE/rjBbRmjU803oswcHSTFmKkksmJbNnp3JPbfFp86QlXyxADt5RUibN0Zq9OiF0tgCdUmIYZ7quI2XHaTSZqk3vCb5WQaZI6I2/PvSB4B03NxOHw4HTcWI8jaT5RJ4ncikUyaYhWxvGVMu9xmu9mMcv5y9776B1Lhb1iChd5wmh9c6cKVLe9TWQNT35rho9eH9eZDe3sPuXFR8Fv+E3/Ib3ve+3/Jbf8rHu6/2C2A+771l85EDWiWMjOyOYh65ayta0vljwWpplejYHEK18uIto9h6Xz7kqapwatQDLTlaZm3kamaepirrb/tBMnCdOxwPzPHPz9AmHw13djZ0YnPN0XY9zHkqh866KShvBRioPKXhTVii1bIiaEHo14kVnC5JTKRxz4XZKpFQYY6k+1ZCX8fNBgedZmmyhGahahlhLJeZbSaVJ/riyDswVH47GxbsMztr2cyb2YszVCFCzcbzRJmPH0khVyBecU7N/bIvGEAL73Y7dsOV2GLi+2jP0Hc4J4/HEEx6z20+88uCavN+CKrvNhvTwmnEcmccjsS5KpXLUS0pMp5E0OzNxEIcE85gXV0dGcHQ+oCgxJcRcecnzzN3NDV3o2G22uM2W4AMlWsNKioFxPHF3d8s8N71cO79sNxsePnxopdi+pwsB5yzobpy7RsWwLPaZWvDFv/E3PuFvesU3E7x3XD/cYjOiXYwXe3b5KqXakotlaKkVk4ySVckxk5L1dxwkghsRIPhbQn2dvvP0VUPVByvPOxG6vjPddu/Z7fYMw1AbozoLIL1nu93Q953Z3g49nQ8cp8Q7TydOU+buOPP2O3ecxkjvYR4sqyo1MBcRpt4zjzvTke8DznV2ftGCVPOfEGDooHghVPMS54Sh7xj6nmaQJE020/d43xF84Gr3gM2wtfdUNacr/4cmJablnAha8XLghTKyvQxnOkEtw/mFS1YWmatSrAS4pCzvleNaKVNwxcK6Jv8DteSZEk6EOE+kOIMGnHTgpbqTzMzTyDiOHA633N7eWnalM/ceD3QOJDicmge0U60DJQCC94EuDFVrU41/p1rVEdpFiHMiqnFij3Mi5ULMQmpv7Xk+uzbG2ozYwlktGFm4URKaXJhjcZdYseIDcN8kRN/FEztTB2Rx0gET6rhcZFowW6sEuWaBxPqSzXXLKiDeObabDUN11NttNpZtrQvPkk1YfZ7GuiAtDEPPVTEqwo234NWyuEY5KikTmUjSfNdtQr58L80JEGyk5GS0iJIi492BFAJBhU0wrl7OaeHZz9PE6XQ0hYVavQGlH3qur65q6Xag6+09NctPd8E7hrPWJQBrILviA+C8sL8a6jySMF3Xlkmsc56aCUCzaz7nOJSshZgK05SqwU5Z5qhSkikPCFV72carc4JUnql3Fhx6V4PU0BG8Yzv0i3TXbjPQd2bXvK3jOBXlGItR6ObE7e2JGDNzEFLnarOWt0yu98SpJ06TNYQ5JRRfi7JnRQPvlC5YNXQYAn1vDVrb7YbtsLGKrgSkash3vid4e/3dsKfvB5s6l9OBUIqcQ4tFiWgNZl8WvJBqga8/tEXouAZzls2RZV5c5sfLy3ugtCyl1edppUepnZCGi8ySE0oppJyJtVM6pURMyaRJ0KpvZ22S4sRUeqogwblL0jzXnW8e7Sb8bEEkVbfPmkZiynbJpm2Xq698k2NeBpMsR9tu1E+t8nfqZyTc/ziUc2m39oHVEbs2e614PrTg6jL7ennfWbVAKcUtv7f3eh3jsb+bZxun2exoS2HbD/S98Vq3uy1X+x25TrJay/hNxaCUYrauauoGXQjVj12WRpbmnOWaOPtF49liblAnar04VlVl6PolgxqqpJ+v3PfL4zFJofQuWbKiBUejEdSO61KQYvJeUvU2V6rPiq8GArWcfrletN8Y1MRPpROUapBwuVA6T2OC1oytV6l0HHfWaL3YZ9G6CK0UOilKcQWRmqyppiApZ+ZKwZk7y9qmVAjBkwvW1KxqzWmxqpCoEPGULJROwcmyiD4bGNjcibN5zollcIN3i6Z7UydwzuZfWc4FDhG/NHi62hCHs0j1Mh+manruLW64pC+ueDnwAtQCofeDBYLqIdemEbCu4mIOfG2FVJOzzwSysqQnRc8ajapmkyfi6Lyn7z390OG8q5neTMxKKplxnjgcj9ze3TFNE3fHE4fTia7rCEOPBG8DxFokwZnhgVbKQQjdIm/ShWrXWRTJ1R4vJtJUSArHKfH4dmSOkcMxMsbKEVJHqYHm5Srw8rQil/8tZHTujzXTIuKsXtuck/xZv3LFig9BC2Dfmxd7vm0ZWTMSWJoi9b6k1MJvEyFrRp0yTxPvvP0OPzD8AFf7PQ9+/BWfefVVtpsN3/nt306aZ8bTyFtvvcXTp0+Zx5G7m1t6H+j6jofX12yGDU+fPuFw85Q8G79U6o88hLCUPpfFJuBDoO/6JRPbjtRMGExUfQg9m94es9/u2W/3OOcR78kpEZ0wjSPH45EYZ6Z5WhbCc4xM1coXEQrGAU45LxNt3/X3dXU/9m9vxacRItB3pYr7W4Wv5Fr5qwmSabaFXs4WXJYCaY6gGUEJHtxgDRtFa/C6cGrrrFH54KpKLJlUbB7J+RwQx1QWiaxwdNXdTpZFX6PTXBqqQE3s1G5n5xynSiPs+o5hM+CDZ7PbkFIyjfVcyNEys06VoROCc3g6hrAFLbigS7Nb6LpqLWvGSc4ZncjMkmryx8VqHY1lYYucb7e513HRc7LiEp8k//WTxAtRCzpnouUtQF1iVKHSCbTqMkJzCLB58aK2vrhdcdHsYatMwTourfQRagamOnNlBSnEFJnmidM4Mk4Tp2linGeKwA61FZ53VmJx93VrvXeEPtQMUKjORs76vLKR6aWYp1AqMMbM3XFimiPjnJlT04utx7/cuCCbnyPYe++ZZ4NcwDQPKv9OauMN1pi2SomseB682/RA7v1977HYAlOERc6u/ZzPj6/LKQXEKhAC3NzcEIJnevCQ8mMKV1dX9H3PG2+8wel45O7ujtubGx7lTIqR8Xjk0HXs93v2n93x6quv4oD9bsfh9vbyiOi6ju1mYyLwF1nYvu/ZbLZLk1Wp2ayUIinOqCqdDyb75xzbYct2szGVA210CGcB7DQSYySlaO5FtakmVj9372Ndcwou56UsC0K417iyjssVHw5rclREjB9bKvfcmpL1omKhpGR2tXbbDH+MC4ulNRVM3byiZj4VSCmTitmhp6g1qG3juZ0b0sVxnX+/riVYGi1BxNzunFTdV1n4rIjDuQAi9ClRMAfAZgxUqi50yYoUy9F03l4viNER0VL1aysvOJhSgohDvDfTBxFcsPlboPL1qxpJdqZ2olByDWQFe6x7nzLTNyk+KACFTz4I/UYLiF9IR/aC4rkEsMDyh2AZUClS6yrnx59fQZ65ZWV9Gkm9C/Rdt7gB2UNMv7WoVpmS5hhi+nrDsKnNGr0JsLuAE187j6k01LNun9TUaCnZtPO0BQSW3Uq1+SwlWy3n6vanelHGaOnVJu21fEBy8bnI+UNr1+faaP3YzuLTKpUzyyXd4psDz/ND/7ge81wSHs/xoG+0wflRccmNfb+sLNQAti0qL+/Ty2u597Nr815OjWt64unTp7z99tukGDnc3VUThGQNJBsbi31vTR1dZy5bm2Fgu91yfXXFdDqRcmKeJ3LOz/jEW5AN1IkxV/rROfPUOMFWZrnQp4ZaunU102Wvczodub29IcbEaRzPtKTKzRcRYkrLOaVlaL2zLJhPdtp81xr1mwCf5Lj80Md80JgE+Nzn+Dxf/ljkeb4R0GhuRcS0W8UCRJVWHbFJwVfdZBHwpfJbGxHtXtKo0vfEsq6WoC00swVov1F73cZ7v8TS+ElLKMkyfi7lqxa1kEXvtlIBREz1wJv6QTtWbSVYLZWPb7qxRpITHAEoduzV9ejs4KfWLKb5gqLYJs0Ljfbat2LGRPUxF+/v0xPGrvgwvKBFbcVlNNey+yKI94AiziMaatmyhqwqiyZr+4m2NWZwdvFeuL4auNoPdF1gsxksK6vKOI2M08w4Tswpo+Lw3cD1w1fZXT8keM9+t2Xo+kVC536ppMn4gNUmEqmk+nZkkaKM08zxeGKaM8fjzPFYmGIx3XX1y7FTtXHflX195u+F29NKPXo+XSgtgFWKFJRSJ229d1L5WuPjmsg+KTyXhMfHpPPxYe/7G2Xl/EGOXAuHVsHat4xS8F7B7EXt5PxrrlqxOSYOtwf++vd+L4/fegtUOR6PTONILoX9bkv/hS/QdYHr6ys2my0PHlzz2mde5Y03XmO7GShp5o3XX+Pu9o4vfemLHO7ujBublaL53vHmZIGsE2dZ1Nzk9SpDXcycxYUO7+w4Sw1I55iYarY1/nDirUePyCXz6PETbquCwWE8cZpGFBjn6bwYsIOgOSo5+eTGYsM325h8LnzYuHvzTT7oEe3p3wyfjYhlJLPY3KcWuaFezJ2qzoC5QPaKD5Zldc7O/aW0ILbONlq9P1SJqiStLnWalttIWXjkqu6Cb1szly1oXYqH5+C1BcBOrHLp3bk6an/bbeccfd+z3Q5GCeq8BaAlgjqcepw6OldwnR2/agfsgFITUrm+tUYGUFBrzhSBgodiNdpSTZa02GfSKAVFm4umKRDVD3XFS4IXDmTlIhBtK6JlYMh5aPhlEJ5D1kUjVTlPRpgV3RBMmmS32bLf2SDpaudwLsWyKaeReZ7JxTq4nBe2/bCUAbddZ7qWF+VJy67cfw+t9JqLdS8361lVK1vO08w0ZaYpMc+FGJUioPW1VJcQtQYDz6x6L//WJUZ4Jj19Ts+qaF1lL+2XL/QdrXi5can7erHxPGnUH2Vb08FlnaQGsNLGs/09T+aeN51O/F2U28ePcc4tEj/OOTabDddXV/gq7dN1HbvtluurPQ+ur+m8J82f5Wq/49GjRzx5/IjxeKzyfQq10SrXjKw21y4n5JSJKWIGBxB8FZzvOrPKBMz1L9eSayROMwU4TCP5yWNKKRyniXGaiZUfO8W4VETaZ1bKJVXj8oP9OL+lFZ92uHMOhyLW+FRaBhMLxlyR6lDVEj1Cya5yU1kqfqkW/+zHbYu+okrWTCEvD22NzVyEqJdWsIu+erv/8jet4OVcuXTOGV+1cmiHqlTQ9x2bobeKi60gq56sufWJKl4KzrdAOtRFoi5VEFMIasdugSy000A2l6EqSabO1/nZWQCrUJbYQlD5xqLifVKLqK/3Yu3riRezqG0dlEtCtnJTlhJFC1vtMfbTdQvnLpdluqBWF3BQBZpN8Lit/pw4KLX5KmfyHMkxVgcs6JzpPPpQJbfEbHJDtQhxzwSTrbTfTBia5izUyavGpDlnYrRmEOvEvijatLKM6FKWsZKHXcu9x1zuv9EIWJq4hKadV6rwu7l9SXHNm4F15lzxsUEEUT1Pbg487iJQO/9um9ZsawSx59qjvDdFgJRSNS2w0qnFolUlAIi1kWUaTSbvdDoRUzS9192O8XRiMwxshsFGZA1en6VILIvjelxKLcvKWWO2Pb7kQtSIKkzTxGmcbLIXKGLhg5Zihg1q18579KIhx8b7OehvPP6LD/Jr8vWs+HRBEHOokkqNcYqTgtRMqc8Z56SaIBSyt9+el4STaut+QWdzqUA2VY3kxZqh1ObXRk8TkbOu+VIKrHN0m5/duZkStQrh5TE7J3TBxnXofA1eayA7mLNY3wWG3hRCuiAWtJIRTWiZa9CZjI9HJcy2+VJbgsmOrSmVNBvoJbhWQJ3RMmrQv8T2z1AOREz9YB2aLw8+erNX5cbAObwy/oxfShKtlO/lnOk3Gau6kMyKq8Fh48U4MW25bXUW2fQ9fW3eKHNkrJzV0+0dp7sDuSidwtWwQZyj7wdC6KqByjOuRNQJrk5IRdXs91RrxaY+pmD2k0UZp8jheOI0RaYpsrhv1WDTSjSKua5gq0FtskHnlWEr1aCKymVHec2GOeg6V52FitniSqEkiJmlsWXFig/DewV/79XshZwnsc57uuCXbmaRswyOd+Zn3ncdXd/ba7csZW2IHMfRMjdaOe7qyDGSsFE1jicUUxh48PAhJWe6ruNqv+fVV18lOM87b72FVN77OI6kVJtSYrTO7Gq6skhw1XFtTSgXwW0xjp5ZSFtQ+vT2jid3d9bV7RzqjCawu75is9vhU6DfHOnG3hq/ppmY9SJzXT+yZz/CC3LRihXvByeOTb+zhV3IdZGkiwtWKYVU0rIt1+2myDHXAFerEpByHCPHMVrzpXOIC0YtyEKqNASTpWrplQvOq5gmtGDzzpK11WVZi3Ae/32wCkvXeXa7Dh88fRfYbvuq9uPZ1qbpYePp3YwnISWS58mql5rRYmcDuRzDEvA1qHZecL6NbKtIlqKkagRRBHIJqLNkWClV15pzQOvF41yH9wH5BCQrX+Ys6DcSXjAjy7nWVicRX7VYpTZHiFT9uCWQhdSkuKRUa9e2pnI4ga4OFFv52UpPsNJiyuZDHceJeZzqcXh6b9qRQzcsOpWazC4PtUDQiPCWii0osqgq3KunogVKNu08k0WZmaZIrEH30iF2j1/Urs/3XWa1zomuxkUQLsuVIhCcWyZkG4Ni9reNzP8iX9aKlw7PBrDy7G+ORhuQe5I7y0QjVezce8QJm2FDP5jzTsnNRjMznU7MY1wmvVLMWShX2SpVZY7z0sh1d3tLX4PY1159levra+ZxZL/fczoemeb5njvZorqgXDSdOHxdMDpH9amXhZZgdAKT1EopczgeuL29tQDBO3AO33UM+x2hM1943wVc8Gg22lB5j+B/DVtXfCSIEFwPrlhQtyQmqhW6FnIOy29dK881Js8cm166krIuWrOmUFAoxZHr88TpuYJXaQg1LLWqpJxlMkVAfJvCpWZ8oTlqCTb++9C4sIHttquW1IHdbiAEoxMMncc7MV1Yl2u2OaM51mRwle6ixg2LXqyvQa3QOb8kx0oLZCkUjZBTpSO6muRpQew5kK0ftPXkyFkib8WnHy+UkTW/dKAGWZYtabR1Y8JaM9M5I3rm6tQA19dSvFaxWQTVTC42/FJypFnOk3DtUiYXfMuH1AySE4evGSJK9WCvgarUhqlSj8lVboF3LXMlC4U3ZtP0i9l4sTGaAUIpeg5el2C1BQPLJ8NSsF3oBtTM7VnAvTWPtM/PThQdm6E2yGkCMpFCEnMYWyfQFR+GFui9p9zWvYav81qrTQSlZTirMxA1YHSV8+qDJwTjt6mr2pXFI6XgsYrMpbVrMw4wD/ZqjFCVDYIPlJwXJ63T8WjNHZwD7EYZaG5ardHzzFvFjry6+iDNRdA6pYu2SV+XhbVVM92SkcoxMo2jBQVzrMF5uThf3ft0aee6xitcseJ54MTRD9XZq7RA9pxLVFW8z0tJvWVsXZW7aiYJqRolFKyKmHOh6zN9zNacnAoxN045NO57m5NBaHu25sWakQXQlrWVGsg2fVlrsOp6R997QnD0vSxmBt4r3hfj9jpB5DzXq9Yx975lDT1PpZdhwuURi8dV/qJzRhtojzsXNi9e1FyMPuI39fXBmtl9MXzkQNY74Wo3mOtVaeL95wKGtJI7CqUOXqiZW5tIgjdlA8UyoGZja45Z85zITnAlIXPVq61ZIK3Z1o6W+Q2LSLnkgpTZJuoqNt2yTkuBpXHrnBBqmcUWsMY/OowHnt6eOE0zT29PHE4TU8zE4sBVHT3ccgZ4VyC7/CH3gtX2WC5v1cxY13leud7z4GpTuQ0TWhKnMZJOR/KcP+pXteIlhIhlcYBzRvMZXGYyMkoqirhCJ2Y9KU7wfaDvzbq532wYhqHKY1XpIFU0btCcEIxH66uFkYJNKqWQYyTGyLEUvvzFL/L4nUdcX11RYuThw4ccDwemySos5yC2Um2GwSorImZZXWzR20TgzZRFKicwE7Fu59j49MUoS33fUdSqQLnSmcbD0eTCqhJDnKYqNF8W+957n9wztpfnYHYdnyveH8579g9esfkrXQayZ/3mhYmtZoiDQs4TOY2WsS2JXCJFlet5YJzttzrHqmleYI7KFBtH3HjiaJOzstcv2FwnNZHUHPV8DV4veeiLwoIKoXNstoEQrIel34ipGTgl+FQlxSpdoc7+RV0dQL5easTa5sNzPmg5ZvvAamVIFO8dqmZ1a5QnowWWpoPJeRxKUcgJIwyvNcyXBS+ckdX6g1qaM+C+C1UrldRgt5UUoOrGOVMwyLQskg3iXMziLqlxRIGFKtBePtSiSRCHr0FlKaWOVEXO3WRWf2xZUa3ZJoRqK3JxSgEtwjhGjuPEaZzriaLmRKWtFGtgfFGibWWbdvv+0vMyoG2UhPPK1XvHZtNztdtAyeisaHaQjHLwKdN3XvEJ4T25sZf3U9kuKBnFaf1V1tWZ87VT2Vs2NtRsq6/ZUgEkeFMTuMz4lrLIY7UsZ8mZuRRub245uAMpzlxfX4HCPI/kyom9XBA7Z4qT6qxz2yww76t5WNOWHXPJZsMJZ8WDUoPg4L1x9GtlRxVSnIlxNvvpaaZk6/5uGasmGNg+q3NUu+ZjVzw/RBz9sLHEj2uNT7XIf+74XdAys6V4SnKoZopGcvGoFrpe6JNpqc6xEGsgO87UQBZSzEYBUoWSlixlCyPFyRLIOmlN1ZWic6FfnmuTmQ/CZvD44PBB6DrBe6pGbG5t3PfGxeUWWZpJ331OkloZKqUtTBftkRrMyvmBVN0jLe9+KYueTeFkxUuDj0dHFnj25/uxvMzHjQuC24vsRu49+cNf6fLxX00w+uwryzpzrvhU4uu0Qnuu8fQc43sNaVd8GN51Mr+8ef7ja9EJ8WIz81fx2/6wh36iw2TN+ny98UnSJeTDMjbv+0SRt4Af+FiPZsXz4DtV9Y2P68XW7/HrjvX7/HRg/R4/XVi/z29+rN/hpwvv+31+5EB2xYoVK1asWLFixYqvJ772QmsrVqxYsWLFihUrVnwNsAayK1asWLFixYoVK74p8fEHsiK/BpG/gci/9bG/9nvv73+GyN/GfGJfv9guiPy2et/3IPJT6vYfj8hfrNt+Wt0WEPmPENl9wH7+T4j8A++x/bsQ+aMf83v6dxD5sR/ra654efAyjEGRv3NvX18PiPwfEflZX9djWPENDRFeEeFXfYyv910ifKT5RoR/XYSviPDXntn+k0X48yL8FRH+cxH+3rr9F4rw10X4syK8Vrf9aBH+3Q/Yh4jwp0V48PV87yK8IcJ/8HHte8U3Nr4WGdlfBfxsVH/xva0iH6NCwj38OeAf4t0k7H8Y+LH18k8Cv6Nu/58A/3Pg5wK/rm77lcAfQPX4nnsQeQ34+1D9Mx/rkb8/fgfwv/yE9rXi04d1DH4QPr7P4bcD79+au2IFvALvHcyJfJyqQc+F3wv8nPfY/n8AfpMqPxn45+vfAL8a+HuA3wn8D+u23wz8cx+wj58L/FVVbvg6vndV3gK+JMLf/7Xcz4pvDHy8gazI/xX4UcCfQOTXIvLdiPx+RP4c8PsR+RGI/OmaiflTiHxHfd6PRuTPI/K9iPxmRO7q9i8g8mcQ+SuI/DVEfsa79qn6l1H9O+9xNL8A+H1VDf7PA68g8gUgArt6iYi8Avx84Pd9wDv7hXCxuhP5OYj8TUT+EvDfvdj+GUT+UH1/fx6Rn1S3v4HIn0TkryPyuxH5AUReR2SPyB9D5K/W9/c/qK/0Z4F/6GsYeKz4tOJlGYOGX43IX6rH/BPq8b7fGHz2c/gvIfIX6vv6nqUCIvKPX2z/nYj4evm99f1/LyK/tr7vHwBeQ+TzH3DcK15u/BbgR9ds52+tWcU/K8IfBv4LEX7EZYZUhF8nwnfX2z9GhP9IhL8qwl8S4UdfvrAIf48If7lmSX9m3cdfqduunz0QVf4M8Og9jlGBB/X2Q+CL9XYBBuo4FeFnAF9W5fs+4P3+YuD/9Q3y3v9QPZ4Vn3acvZ0/pgv8HYXX6+3vVviLCtv69x9R+GX19q9Q+EP19h9V+EX19j+lcFdv/9MK/2y97RWun2u/59f86Rd//ymF/5rCdyj8xwr/qcJPUviXFL7rQ97Tv6nw8+vtjcIPKfxYNSnrf0/hj9b7frvCb6y3f5bCX6m3/1WFf6be/jnVc+R1hV+o8Lsu9vPw4vafVPipH/v3s14+/ZdP+xg87+tX19u/SuF319vvNwaf/Rx+u8Ivrrd7ha3CT6yfT1e3/18UfqnCT1X4kxf7fuXi9u9S+IVf9+98vXxDXkB/BOhfu/j7u0APoD/yfe7/daDfXW//f0D/kXp7A7qrz/+joP910L8I+h31/j8C+vfX21eg4XmOp277iaA/CPpDoD8M+p11+8+u+/gjoA9B/0PQz3zI+/0B0OtvhPcO+q2g3/v1/g2sl6/95ZNo9vrDqJ7q7Z8G/MF6+/cDP/1i+79fb//Bi+f+Z8AvR+S7gf8yqrcvfDSqP4jqd6H604Aj8G3A36jZmn8XkR/3Hs/6AvBWvf0TgO9H9fvq6PsDF4/76fV9geqfxrI1D+r2f6du/w+Ax/Xx3wv8bET+RUR+BqpPL17rK8C3vPD7XbHi0zcGG/4f9fovAj+i3n6/MQj3P4f/FPhfI/K/Ar6zbv9vAD8V+M8Q+Sv17x8F/P+AH4XIb0fk5wA3F8ewjtMVXy3+girf/0EPqFnFb1Xl/wmgyqhKo938ROBfA36+Kj9Yt/054F8W4dcAr6iSvorj+ZXAr1Xl24FfC/yeus8/qcpPVeXnY9WVPw78OBH+byL8LhHei8/+GVU+6BzxSb73dWy+JPgkAtnDR36m8eH+AeCHgd+LyC/9Kp79w8C3X/z9bXXbJf53GN/n1wC/G+Ol/sb3eK0TsPkq9v18UP3/Aj8FC2h/MyL//MW9m7rfFSteFJ/WMTjV68zzuRSePwfVPwj8t+vr/nGsaUuAfxPVn1wvPx7V70b1MfBfAf5j4J+qx9mwjtMVXy0ux2Pi/jz8PPPMl4AR+K+2Dar8FuCfALbAnxPhJ3wVx/PLOC8K/32wZq+GGrD+j4D/M/Cb6uP/E967bJ9EPjCu+CTf+zo2XxJ80vJb/2/gH6u3fzHGBQX48xgHjov7QeQ7gTdR/V3Y5PFTvop9/WHgl2Kd038f8BTVL1289s8Evojq92EcoFIv77XK/BvAj6m3/ybwIxBpnJ1fdPG4P0sb3CLfBbyN6g22YvxH6/b/JvBqvf0twBHVPwD81mfe34+D+92lK1Z8DPg0jMEPwvuNwfsQsUyr6m/DOH0/CfhTwH8Pkc/Wx3wGke/E1BEcqv93LOhex+mK58UtvJuveoE3gc+K8JoIA/DzAGpW8++K8N8BEGG4yIA+Af5bwP9ehO+q9/9oVb5XlX8Rq6J8NYHsF4GfWW//LHgXB/bXA79NlYgFi8r7j9O/hVUx4Ov/3tex+ZLgk24m+tXAv4HIr8fKhL+8bv9fAH8AkX8Wa+hoJfbvAn49IhG4A96dDRL5NVgW5/PA9yDyx1H9J7AyyM8F/jZWvvzlF88RbEJqzVX/GvBvYZ/Hr3yP4/5jWKf170Z1ROSfBP4YIkds4myD9buBfx2R76n7/GV1+28C/m1EfglW0vwyNsi/C/itiBSsAeZX1uP7HHBC9cvv9SGuWPEC+OYfgx+M7+a9x+Cz+EeBX1Lf15eBfwHVR4j8c8B/iIjDxuT/FMvq/Bt1G8A/U99DhwXX//mHHNOKlxSqvCPCn6tNTX8C+x1f3h9F+N8CfwGrVvzNi7t/CfA76/0R+O9fPO9NEX4e8CdE+BXAPy7CP4gFmH+97useRPi3sfH8ugh/F/iNqvwe4H8M/CtVSWDEFEbac74F+HtV+U1102/HgsUnYIHmM/hjdR9/+xvgvf+Dz+5zxacT3xgWtaYdeUJVEfnHgF+E6i/4eh/WPYj8J8DPQ/XJR3juAGRUE6ab+TtQ/ckf8PhfC9yg+ns+2sGuWPFV4tM+Br8WEPlHgJ+C6v/m630oK1Z8I0CELwC/T5Wf/Q1wLH8G+AWqS0/Kik8pvlHknX4q8K/WLM0T4Fd8fQ/nPfFPA9+BHd9Xi+8A/r2a0ZmxFfAH4QmtYWXFik8Gn/Yx+LVAAP6lr/dBrFjxjQJVvlQbwR6o8m5KzycEEd4A/uU1iH058I2RkV2xYsWKFStWrFix4qvEJ93stWLFihUrVqxYsWLFx4I1kF2xYsWKFStWrFjxTYk1kF2xYsWKFStWrFjxTYk1kF2xYsWKFStWrFjxTYk1kF2xYsWKFStWrFjxTYk1kF2xYsWKFStWrFjxTYn/Pwc9hoKKr+cWAAAAAElFTkSuQmCC\n",
       "text/plain": [
        "<Figure size 864x720 with 30 Axes>"
       ]
@@ -1563,7 +1583,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 28,
+   "execution_count": 29,
    "metadata": {
     "id": "yRJ7JU7JCaXT"
    },
@@ -1594,7 +1614,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 29,
+   "execution_count": 30,
    "metadata": {
     "id": "lDFh5yF_CaXW"
    },
@@ -1625,7 +1645,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 30,
+   "execution_count": 31,
    "metadata": {
     "id": "o_rzNSdrCaXY"
    },
@@ -1634,8 +1654,8 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "[[0.09914809 0.09525293 0.10366846 0.10192148 0.10296609 0.10004853\n",
-      "  0.10344286 0.10120125 0.09786693 0.09448335]]\n"
+      "[[0.11279975 0.10304905 0.10780791 0.09079217 0.10205003 0.09319472\n",
+      "  0.07174336 0.09929511 0.11256534 0.10670257]]\n"
      ]
     }
    ],
@@ -1647,7 +1667,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 31,
+   "execution_count": 32,
    "metadata": {
     "id": "6Ai-cpLjO-3A"
    },
@@ -1661,7 +1681,7 @@
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -1688,7 +1708,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 32,
+   "execution_count": 33,
    "metadata": {
     "id": "2tRmdq_8CaXb"
    },
@@ -1696,10 +1716,10 @@
     {
      "data": {
       "text/plain": [
-       "2"
+       "0"
       ]
      },
-     "execution_count": 32,
+     "execution_count": 33,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1728,7 +1748,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 33,
+   "execution_count": 34,
    "metadata": {
     "id": "0XHhKCJOOGMM"
    },
@@ -1737,207 +1757,206 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Train on 4000 samples\n",
       "Epoch 1/100\n",
-      "4000/4000 [==============================] - 1s 259us/sample - loss: 60.3141 - accuracy: 0.1430\n",
+      "125/125 [==============================] - 3s 21ms/step - loss: 60.2776 - accuracy: 0.1605 - lr: 1.0000e-06802 - acc\n",
       "Epoch 2/100\n",
-      "4000/4000 [==============================] - 1s 186us/sample - loss: 26.2780 - accuracy: 0.2125\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 29.3304 - accuracy: 0.2040 - lr: 1.0798e-06\n",
       "Epoch 3/100\n",
-      "4000/4000 [==============================] - 1s 191us/sample - loss: 20.6708 - accuracy: 0.2225\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 22.7748 - accuracy: 0.2260 - lr: 1.1659e-06\n",
       "Epoch 4/100\n",
-      "4000/4000 [==============================] - 1s 189us/sample - loss: 17.4517 - accuracy: 0.2362\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 20.0149 - accuracy: 0.2488 - lr: 1.2589e-06\n",
       "Epoch 5/100\n",
-      "4000/4000 [==============================] - 1s 202us/sample - loss: 15.0166 - accuracy: 0.2533\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 17.4151 - accuracy: 0.2610 - lr: 1.3594e-06\n",
       "Epoch 6/100\n",
-      "4000/4000 [==============================] - 1s 203us/sample - loss: 13.0606 - accuracy: 0.2618\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 15.5394 - accuracy: 0.2690 - lr: 1.4678e-06\n",
       "Epoch 7/100\n",
-      "4000/4000 [==============================] - 1s 198us/sample - loss: 11.8550 - accuracy: 0.2707\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 14.0803 - accuracy: 0.2887 - lr: 1.5849e-06\n",
       "Epoch 8/100\n",
-      "4000/4000 [==============================] - 1s 200us/sample - loss: 10.4342 - accuracy: 0.2747\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 12.6668 - accuracy: 0.2945 - lr: 1.7113e-06\n",
       "Epoch 9/100\n",
-      "4000/4000 [==============================] - 1s 195us/sample - loss: 9.5936 - accuracy: 0.2842\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 11.8588 - accuracy: 0.3090 - lr: 1.8478e-06\n",
       "Epoch 10/100\n",
-      "4000/4000 [==============================] - 1s 196us/sample - loss: 8.9411 - accuracy: 0.3030\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 10.7524 - accuracy: 0.3147 - lr: 1.9953e-06370 - ac\n",
       "Epoch 11/100\n",
-      "4000/4000 [==============================] - 1s 190us/sample - loss: 8.3025 - accuracy: 0.3045\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 10.0094 - accuracy: 0.3198 - lr: 2.1544e-06\n",
       "Epoch 12/100\n",
-      "4000/4000 [==============================] - 1s 190us/sample - loss: 7.5267 - accuracy: 0.3185\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 9.4757 - accuracy: 0.3110 - lr: 2.3263e-06\n",
       "Epoch 13/100\n",
-      "4000/4000 [==============================] - 1s 194us/sample - loss: 6.8614 - accuracy: 0.3320\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 8.6934 - accuracy: 0.3310 - lr: 2.5119e-06\n",
       "Epoch 14/100\n",
-      "4000/4000 [==============================] - 1s 205us/sample - loss: 6.4472 - accuracy: 0.3385\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 7.5893 - accuracy: 0.3485 - lr: 2.7123e-06\n",
       "Epoch 15/100\n",
-      "4000/4000 [==============================] - 1s 197us/sample - loss: 6.0191 - accuracy: 0.3520\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 7.1017 - accuracy: 0.3668 - lr: 2.9286e-06\n",
       "Epoch 16/100\n",
-      "4000/4000 [==============================] - 1s 202us/sample - loss: 5.4516 - accuracy: 0.3623\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 6.4868 - accuracy: 0.3663 - lr: 3.1623e-06\n",
       "Epoch 17/100\n",
-      "4000/4000 [==============================] - 1s 189us/sample - loss: 4.9921 - accuracy: 0.3750\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 6.0443 - accuracy: 0.3803 - lr: 3.4145e-06\n",
       "Epoch 18/100\n",
-      "4000/4000 [==============================] - 1s 206us/sample - loss: 4.8239 - accuracy: 0.3868\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 5.7332 - accuracy: 0.3820 - lr: 3.6869e-06\n",
       "Epoch 19/100\n",
-      "4000/4000 [==============================] - 1s 190us/sample - loss: 4.7828 - accuracy: 0.3832\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 5.5397 - accuracy: 0.3898 - lr: 3.9811e-06\n",
       "Epoch 20/100\n",
-      "4000/4000 [==============================] - 1s 201us/sample - loss: 4.4205 - accuracy: 0.3950\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 5.1825 - accuracy: 0.3970 - lr: 4.2987e-06\n",
       "Epoch 21/100\n",
-      "4000/4000 [==============================] - 1s 188us/sample - loss: 4.1003 - accuracy: 0.4103\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 4.9526 - accuracy: 0.4100 - lr: 4.6416e-06\n",
       "Epoch 22/100\n",
-      "4000/4000 [==============================] - 1s 191us/sample - loss: 3.7402 - accuracy: 0.4153\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 4.4758 - accuracy: 0.4252 - lr: 5.0119e-06\n",
       "Epoch 23/100\n",
-      "4000/4000 [==============================] - 1s 182us/sample - loss: 3.5342 - accuracy: 0.4195\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 4.3288 - accuracy: 0.4280 - lr: 5.4117e-06\n",
       "Epoch 24/100\n",
-      "4000/4000 [==============================] - 1s 219us/sample - loss: 3.6627 - accuracy: 0.4005\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 4.0896 - accuracy: 0.4322 - lr: 5.8434e-06\n",
       "Epoch 25/100\n",
-      "4000/4000 [==============================] - 1s 194us/sample - loss: 3.2692 - accuracy: 0.4160\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 3.7686 - accuracy: 0.4435 - lr: 6.3096e-06\n",
       "Epoch 26/100\n",
-      "4000/4000 [==============================] - 1s 196us/sample - loss: 2.8876 - accuracy: 0.4392\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 3.7501 - accuracy: 0.4363 - lr: 6.8129e-06\n",
       "Epoch 27/100\n",
-      "4000/4000 [==============================] - 1s 184us/sample - loss: 2.6240 - accuracy: 0.4495\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 3.2985 - accuracy: 0.4633 - lr: 7.3564e-06\n",
       "Epoch 28/100\n",
-      "4000/4000 [==============================] - 1s 199us/sample - loss: 2.4698 - accuracy: 0.4582\n",
+      "125/125 [==============================] - 3s 21ms/step - loss: 3.1863 - accuracy: 0.4652 - lr: 7.9433e-06\n",
       "Epoch 29/100\n",
-      "4000/4000 [==============================] - 1s 176us/sample - loss: 2.3354 - accuracy: 0.4638\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 2.8219 - accuracy: 0.4888 - lr: 8.5770e-069 - accuracy: 0.50 - ETA: \n",
       "Epoch 30/100\n",
-      "4000/4000 [==============================] - 1s 179us/sample - loss: 2.1678 - accuracy: 0.4647\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 2.7807 - accuracy: 0.4875 - lr: 9.2612e-067 - accuracy: 0.48\n",
       "Epoch 31/100\n",
-      "4000/4000 [==============================] - 1s 178us/sample - loss: 2.0470 - accuracy: 0.4720\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 2.4144 - accuracy: 0.5095 - lr: 1.0000e-05\n",
       "Epoch 32/100\n",
-      "4000/4000 [==============================] - 1s 170us/sample - loss: 1.8744 - accuracy: 0.4880\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 2.5267 - accuracy: 0.4922 - lr: 1.0798e-05\n",
       "Epoch 33/100\n",
-      "4000/4000 [==============================] - 1s 187us/sample - loss: 1.8277 - accuracy: 0.4850\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 2.2568 - accuracy: 0.5225 - lr: 1.1659e-05\n",
       "Epoch 34/100\n",
-      "4000/4000 [==============================] - 1s 185us/sample - loss: 1.7947 - accuracy: 0.4830\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 2.2646 - accuracy: 0.4990 - lr: 1.2589e-05\n",
       "Epoch 35/100\n",
-      "4000/4000 [==============================] - 1s 173us/sample - loss: 1.6551 - accuracy: 0.5025\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 2.1021 - accuracy: 0.5095 - lr: 1.3594e-05\n",
       "Epoch 36/100\n",
-      "4000/4000 [==============================] - 1s 186us/sample - loss: 1.6081 - accuracy: 0.5095\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 1.8296 - accuracy: 0.5493 - lr: 1.4678e-05\n",
       "Epoch 37/100\n",
-      "4000/4000 [==============================] - 1s 198us/sample - loss: 1.5493 - accuracy: 0.5195\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 1.7501 - accuracy: 0.5435 - lr: 1.5849e-05\n",
       "Epoch 38/100\n",
-      "4000/4000 [==============================] - 1s 200us/sample - loss: 1.5226 - accuracy: 0.5128\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 1.8584 - accuracy: 0.5130 - lr: 1.7113e-05\n",
       "Epoch 39/100\n",
-      "4000/4000 [==============================] - 1s 191us/sample - loss: 1.4990 - accuracy: 0.5197\n",
+      "125/125 [==============================] - 3s 21ms/step - loss: 1.7223 - accuracy: 0.5355 - lr: 1.8478e-05\n",
       "Epoch 40/100\n",
-      "4000/4000 [==============================] - 1s 194us/sample - loss: 1.4510 - accuracy: 0.5188\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 1.6808 - accuracy: 0.5400 - lr: 1.9953e-05\n",
       "Epoch 41/100\n",
-      "4000/4000 [==============================] - 1s 211us/sample - loss: 1.3729 - accuracy: 0.5480\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 1.4447 - accuracy: 0.5723 - lr: 2.1544e-05\n",
       "Epoch 42/100\n",
-      "4000/4000 [==============================] - 1s 198us/sample - loss: 1.3944 - accuracy: 0.5318\n",
+      "125/125 [==============================] - 3s 21ms/step - loss: 1.5951 - accuracy: 0.5450 - lr: 2.3263e-05\n",
       "Epoch 43/100\n",
-      "4000/4000 [==============================] - 1s 199us/sample - loss: 1.3661 - accuracy: 0.5387\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 1.4580 - accuracy: 0.5548 - lr: 2.5119e-05\n",
       "Epoch 44/100\n",
-      "4000/4000 [==============================] - 1s 198us/sample - loss: 1.3787 - accuracy: 0.5232\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 1.3944 - accuracy: 0.5660 - lr: 2.7123e-05\n",
       "Epoch 45/100\n",
-      "4000/4000 [==============================] - 1s 198us/sample - loss: 1.4931 - accuracy: 0.4900\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 1.3392 - accuracy: 0.5695 - lr: 2.9286e-05\n",
       "Epoch 46/100\n",
-      "4000/4000 [==============================] - 1s 189us/sample - loss: 1.4215 - accuracy: 0.5140\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 1.3687 - accuracy: 0.5518 - lr: 3.1623e-05\n",
       "Epoch 47/100\n",
-      "4000/4000 [==============================] - 1s 186us/sample - loss: 1.3758 - accuracy: 0.5148\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 1.3353 - accuracy: 0.5667 - lr: 3.4145e-05\n",
       "Epoch 48/100\n",
-      "4000/4000 [==============================] - 1s 198us/sample - loss: 1.3440 - accuracy: 0.5240\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 1.3154 - accuracy: 0.5665 - lr: 3.6869e-05\n",
       "Epoch 49/100\n",
-      "4000/4000 [==============================] - 1s 197us/sample - loss: 1.3474 - accuracy: 0.5272\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 1.3372 - accuracy: 0.5598 - lr: 3.9811e-054 - accuracy: \n",
       "Epoch 50/100\n",
-      "4000/4000 [==============================] - 1s 194us/sample - loss: 1.4087 - accuracy: 0.5163\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 1.3050 - accuracy: 0.5627 - lr: 4.2987e-05\n",
       "Epoch 51/100\n",
-      "4000/4000 [==============================] - 1s 195us/sample - loss: 1.4345 - accuracy: 0.4920\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 1.3123 - accuracy: 0.5535 - lr: 4.6416e-054 - accu\n",
       "Epoch 52/100\n",
-      "4000/4000 [==============================] - 1s 191us/sample - loss: 1.4276 - accuracy: 0.4972\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 1.3686 - accuracy: 0.5293 - lr: 5.0119e-05\n",
       "Epoch 53/100\n",
-      "4000/4000 [==============================] - 1s 190us/sample - loss: 1.3955 - accuracy: 0.5105\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 1.4103 - accuracy: 0.5225 - lr: 5.4117e-05\n",
       "Epoch 54/100\n",
-      "4000/4000 [==============================] - 1s 186us/sample - loss: 1.4635 - accuracy: 0.4875\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 1.3974 - accuracy: 0.5320 - lr: 5.8434e-05\n",
       "Epoch 55/100\n",
-      "4000/4000 [==============================] - 1s 193us/sample - loss: 1.3851 - accuracy: 0.5157\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 1.3539 - accuracy: 0.5360 - lr: 6.3096e-05\n",
       "Epoch 56/100\n",
-      "4000/4000 [==============================] - 1s 192us/sample - loss: 1.4518 - accuracy: 0.4790\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 1.3960 - accuracy: 0.5285 - lr: 6.8129e-05\n",
       "Epoch 57/100\n",
-      "4000/4000 [==============================] - 1s 169us/sample - loss: 1.5017 - accuracy: 0.4767\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 1.4210 - accuracy: 0.5080 - lr: 7.3564e-05\n",
       "Epoch 58/100\n",
-      "4000/4000 [==============================] - 1s 189us/sample - loss: 1.4920 - accuracy: 0.4720\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 1.3646 - accuracy: 0.5250 - lr: 7.9433e-05\n",
       "Epoch 59/100\n",
-      "4000/4000 [==============================] - 1s 194us/sample - loss: 1.4705 - accuracy: 0.4762\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 1.4185 - accuracy: 0.5102 - lr: 8.5770e-05\n",
       "Epoch 60/100\n",
-      "4000/4000 [==============================] - 1s 199us/sample - loss: 1.6413 - accuracy: 0.4128\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 1.4707 - accuracy: 0.4803 - lr: 9.2612e-05\n",
       "Epoch 61/100\n",
-      "4000/4000 [==============================] - 1s 196us/sample - loss: 1.6659 - accuracy: 0.4145\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 1.4065 - accuracy: 0.5042 - lr: 1.0000e-04\n",
       "Epoch 62/100\n",
-      "4000/4000 [==============================] - 1s 192us/sample - loss: 1.6273 - accuracy: 0.4168\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 1.4203 - accuracy: 0.4963 - lr: 1.0798e-04\n",
       "Epoch 63/100\n",
-      "4000/4000 [==============================] - 1s 191us/sample - loss: 1.7323 - accuracy: 0.3758\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 1.5497 - accuracy: 0.4678 - lr: 1.1659e-04\n",
       "Epoch 64/100\n",
-      "4000/4000 [==============================] - 1s 200us/sample - loss: 1.6940 - accuracy: 0.3923\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 1.6588 - accuracy: 0.4310 - lr: 1.2589e-04\n",
       "Epoch 65/100\n",
-      "4000/4000 [==============================] - 1s 192us/sample - loss: 1.6742 - accuracy: 0.3938\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 1.5678 - accuracy: 0.4390 - lr: 1.3594e-04\n",
       "Epoch 66/100\n",
-      "4000/4000 [==============================] - 1s 191us/sample - loss: 1.7220 - accuracy: 0.3640\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 1.6629 - accuracy: 0.4123 - lr: 1.4678e-04\n",
       "Epoch 67/100\n",
-      "4000/4000 [==============================] - 1s 183us/sample - loss: 1.8148 - accuracy: 0.3462\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 1.6737 - accuracy: 0.4055 - lr: 1.5849e-04\n",
       "Epoch 68/100\n",
-      "4000/4000 [==============================] - 1s 212us/sample - loss: 1.8972 - accuracy: 0.3072\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 1.6150 - accuracy: 0.4123 - lr: 1.7113e-04\n",
       "Epoch 69/100\n",
-      "4000/4000 [==============================] - 1s 192us/sample - loss: 1.9677 - accuracy: 0.2940\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 1.6676 - accuracy: 0.4150 - lr: 1.8478e-04\n",
       "Epoch 70/100\n",
-      "4000/4000 [==============================] - 1s 197us/sample - loss: 2.0168 - accuracy: 0.2435\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 1.7472 - accuracy: 0.3823 - lr: 1.9953e-04\n",
       "Epoch 71/100\n",
-      "4000/4000 [==============================] - 1s 175us/sample - loss: 2.0817 - accuracy: 0.2237\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 1.8559 - accuracy: 0.3315 - lr: 2.1544e-04\n",
       "Epoch 72/100\n",
-      "4000/4000 [==============================] - 1s 181us/sample - loss: 2.1317 - accuracy: 0.2240\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 1.9034 - accuracy: 0.3115 - lr: 2.3263e-04\n",
       "Epoch 73/100\n",
-      "4000/4000 [==============================] - 1s 167us/sample - loss: 2.2026 - accuracy: 0.1500\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 1.9967 - accuracy: 0.2560 - lr: 2.5119e-04\n",
       "Epoch 74/100\n",
-      "4000/4000 [==============================] - 1s 193us/sample - loss: 2.2767 - accuracy: 0.1210\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 2.1371 - accuracy: 0.1885 - lr: 2.7123e-048 - \n",
       "Epoch 75/100\n",
-      "4000/4000 [==============================] - 1s 190us/sample - loss: 2.2664 - accuracy: 0.1303\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 2.2103 - accuracy: 0.1462 - lr: 2.9286e-04\n",
       "Epoch 76/100\n",
-      "4000/4000 [==============================] - 1s 191us/sample - loss: 2.2357 - accuracy: 0.1530\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 2.2582 - accuracy: 0.1395 - lr: 3.1623e-040 - \n",
       "Epoch 77/100\n",
-      "4000/4000 [==============================] - 1s 181us/sample - loss: 2.2636 - accuracy: 0.1258\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 2.2285 - accuracy: 0.1430 - lr: 3.4145e-04\n",
       "Epoch 78/100\n",
-      "4000/4000 [==============================] - 1s 179us/sample - loss: 2.2405 - accuracy: 0.1445\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 2.2112 - accuracy: 0.1485 - lr: 3.6869e-04\n",
       "Epoch 79/100\n",
-      "4000/4000 [==============================] - 1s 190us/sample - loss: 2.2885 - accuracy: 0.1130\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 2.2830 - accuracy: 0.1195 - lr: 3.9811e-04\n",
       "Epoch 80/100\n",
-      "4000/4000 [==============================] - 1s 207us/sample - loss: 2.3153 - accuracy: 0.1160\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 2.2711 - accuracy: 0.1268 - lr: 4.2987e-04\n",
       "Epoch 81/100\n",
-      "4000/4000 [==============================] - 1s 204us/sample - loss: 2.3092 - accuracy: 0.1100\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 2.2832 - accuracy: 0.1163 - lr: 4.6416e-040 - accuracy\n",
       "Epoch 82/100\n",
-      "4000/4000 [==============================] - 1s 188us/sample - loss: 2.2971 - accuracy: 0.1072\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 2.2898 - accuracy: 0.1140 - lr: 5.0119e-04\n",
       "Epoch 83/100\n",
-      "4000/4000 [==============================] - 1s 194us/sample - loss: 2.2986 - accuracy: 0.1072\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 2.2965 - accuracy: 0.1098 - lr: 5.4117e-04\n",
       "Epoch 84/100\n",
-      "4000/4000 [==============================] - 1s 188us/sample - loss: 2.2969 - accuracy: 0.1070\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 2.2946 - accuracy: 0.1088 - lr: 5.8434e-04\n",
       "Epoch 85/100\n",
-      "4000/4000 [==============================] - 1s 195us/sample - loss: 2.3012 - accuracy: 0.1077\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 2.2930 - accuracy: 0.1085 - lr: 6.3096e-04\n",
       "Epoch 86/100\n",
-      "4000/4000 [==============================] - 1s 191us/sample - loss: 2.3004 - accuracy: 0.1072\n",
+      "125/125 [==============================] - 4s 29ms/step - loss: 2.2923 - accuracy: 0.1095 - lr: 6.8129e-04\n",
       "Epoch 87/100\n",
-      "4000/4000 [==============================] - 1s 182us/sample - loss: 2.2980 - accuracy: 0.1067\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 2.2958 - accuracy: 0.1055 - lr: 7.3564e-04\n",
       "Epoch 88/100\n",
-      "4000/4000 [==============================] - 1s 183us/sample - loss: 2.2979 - accuracy: 0.1072\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 2.2978 - accuracy: 0.1095 - lr: 7.9433e-04\n",
       "Epoch 89/100\n",
-      "4000/4000 [==============================] - 1s 172us/sample - loss: 2.2982 - accuracy: 0.1065\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 2.3043 - accuracy: 0.1047 - lr: 8.5770e-046 \n",
       "Epoch 90/100\n",
-      "4000/4000 [==============================] - 1s 183us/sample - loss: 2.2990 - accuracy: 0.1063\n",
+      "125/125 [==============================] - 3s 21ms/step - loss: 2.3001 - accuracy: 0.1032 - lr: 9.2612e-04\n",
       "Epoch 91/100\n",
-      "4000/4000 [==============================] - 1s 187us/sample - loss: 2.2985 - accuracy: 0.1063\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 2.2994 - accuracy: 0.0997 - lr: 0.0010\n",
       "Epoch 92/100\n",
-      "4000/4000 [==============================] - 1s 182us/sample - loss: 2.2985 - accuracy: 0.1063\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 2.3023 - accuracy: 0.0980 - lr: 0.0011\n",
       "Epoch 93/100\n",
-      "4000/4000 [==============================] - 1s 189us/sample - loss: 2.2985 - accuracy: 0.0950\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 2.3040 - accuracy: 0.1037 - lr: 0.0012\n",
       "Epoch 94/100\n",
-      "4000/4000 [==============================] - 1s 195us/sample - loss: 2.2985 - accuracy: 0.1018\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 2.3002 - accuracy: 0.0948 - lr: 0.0013\n",
       "Epoch 95/100\n",
-      "4000/4000 [==============================] - 1s 170us/sample - loss: 2.2985 - accuracy: 0.0997\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 2.3002 - accuracy: 0.0988 - lr: 0.0014\n",
       "Epoch 96/100\n",
-      "4000/4000 [==============================] - 1s 185us/sample - loss: 2.2985 - accuracy: 0.1015\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 2.3002 - accuracy: 0.1025 - lr: 0.0015.3000 \n",
       "Epoch 97/100\n",
-      "4000/4000 [==============================] - 1s 197us/sample - loss: 2.2985 - accuracy: 0.1018\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 2.3002 - accuracy: 0.0988 - lr: 0.0016\n",
       "Epoch 98/100\n",
-      "4000/4000 [==============================] - 1s 187us/sample - loss: 2.2986 - accuracy: 0.1023\n",
+      "125/125 [==============================] - 3s 22ms/step - loss: 2.3002 - accuracy: 0.1018 - lr: 0.0017\n",
       "Epoch 99/100\n",
-      "4000/4000 [==============================] - 1s 188us/sample - loss: 2.2985 - accuracy: 0.0983\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 2.3002 - accuracy: 0.1020 - lr: 0.0018\n",
       "Epoch 100/100\n",
-      "4000/4000 [==============================] - 1s 180us/sample - loss: 2.2985 - accuracy: 0.0975\n"
+      "125/125 [==============================] - 3s 23ms/step - loss: 2.3002 - accuracy: 0.1042 - lr: 0.0020\n"
      ]
     }
    ],
@@ -1966,7 +1985,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 34,
+   "execution_count": 35,
    "metadata": {
     "id": "u64JiZ--RfNn"
    },
@@ -1977,13 +1996,13 @@
        "(1e-06, 0.001, 0.0, 20.0)"
       ]
      },
-     "execution_count": 34,
+     "execution_count": 35,
      "metadata": {},
      "output_type": "execute_result"
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -2027,224 +2046,224 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 35,
+   "execution_count": 36,
    "metadata": {
     "id": "SMqahW9svXZ2"
    },
    "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/opt/conda/lib/python3.7/site-packages/keras/optimizer_v2/adam.py:105: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead.\n",
+      "  super(Adam, self).__init__(name, **kwargs)\n",
+      "/opt/conda/lib/python3.7/site-packages/ipykernel_launcher.py:36: UserWarning: `Model.fit_generator` is deprecated and will be removed in a future version. Please use `Model.fit`, which supports generators.\n"
+     ]
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "WARNING:tensorflow:sample_weight modes were coerced from\n",
-      "  ...\n",
-      "    to  \n",
-      "  ['...']\n",
-      "WARNING:tensorflow:sample_weight modes were coerced from\n",
-      "  ...\n",
-      "    to  \n",
-      "  ['...']\n",
-      "Train for 63 steps, validate for 16 steps\n",
       "Epoch 1/100\n",
-      "63/63 [==============================] - 4s 62ms/step - loss: 86.7751 - accuracy: 0.1115 - val_loss: 62.2731 - val_accuracy: 0.1220\n",
+      "63/63 [==============================] - 6s 81ms/step - loss: 87.3796 - accuracy: 0.1150 - val_loss: 63.4311 - val_accuracy: 0.1440\n",
       "Epoch 2/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 50.4389 - accuracy: 0.1525 - val_loss: 50.5928 - val_accuracy: 0.1440\n",
+      "63/63 [==============================] - 4s 71ms/step - loss: 51.3846 - accuracy: 0.1505 - val_loss: 48.1949 - val_accuracy: 0.1500\n",
       "Epoch 3/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 42.7557 - accuracy: 0.1612 - val_loss: 43.8770 - val_accuracy: 0.1670\n",
+      "63/63 [==============================] - 5s 76ms/step - loss: 41.1413 - accuracy: 0.1573 - val_loss: 40.6570 - val_accuracy: 0.1600\n",
       "Epoch 4/100\n",
-      "63/63 [==============================] - 3s 54ms/step - loss: 37.5669 - accuracy: 0.1735 - val_loss: 39.2737 - val_accuracy: 0.1700\n",
+      "63/63 [==============================] - 5s 72ms/step - loss: 35.6244 - accuracy: 0.1663 - val_loss: 36.0196 - val_accuracy: 0.1730\n",
       "Epoch 5/100\n",
-      "63/63 [==============================] - 4s 61ms/step - loss: 33.2105 - accuracy: 0.1793 - val_loss: 35.9195 - val_accuracy: 0.1750\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 32.3886 - accuracy: 0.1657 - val_loss: 33.1440 - val_accuracy: 0.1710\n",
       "Epoch 6/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 31.5966 - accuracy: 0.1740 - val_loss: 33.1607 - val_accuracy: 0.1970\n",
+      "63/63 [==============================] - 5s 79ms/step - loss: 28.9966 - accuracy: 0.1650 - val_loss: 30.4703 - val_accuracy: 0.1570\n",
       "Epoch 7/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 29.3372 - accuracy: 0.1883 - val_loss: 30.7882 - val_accuracy: 0.1870\n",
+      "63/63 [==============================] - 4s 69ms/step - loss: 27.2292 - accuracy: 0.1698 - val_loss: 28.9967 - val_accuracy: 0.1640\n",
       "Epoch 8/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 26.4772 - accuracy: 0.1852 - val_loss: 28.4816 - val_accuracy: 0.2040\n",
+      "63/63 [==============================] - 4s 69ms/step - loss: 26.0068 - accuracy: 0.1717 - val_loss: 26.4203 - val_accuracy: 0.1790\n",
       "Epoch 9/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 25.4928 - accuracy: 0.1830 - val_loss: 27.0500 - val_accuracy: 0.2070\n",
+      "63/63 [==============================] - 4s 69ms/step - loss: 24.0249 - accuracy: 0.1737 - val_loss: 25.5074 - val_accuracy: 0.1690\n",
       "Epoch 10/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 23.7661 - accuracy: 0.1947 - val_loss: 26.5256 - val_accuracy: 0.1910\n",
+      "63/63 [==============================] - 4s 68ms/step - loss: 22.6697 - accuracy: 0.1743 - val_loss: 24.8954 - val_accuracy: 0.1720\n",
       "Epoch 11/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 23.9458 - accuracy: 0.1937 - val_loss: 24.4593 - val_accuracy: 0.2050\n",
+      "63/63 [==============================] - 4s 65ms/step - loss: 21.3804 - accuracy: 0.1850 - val_loss: 23.4098 - val_accuracy: 0.1780\n",
       "Epoch 12/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 21.9132 - accuracy: 0.1988 - val_loss: 23.4824 - val_accuracy: 0.1920\n",
+      "63/63 [==============================] - 4s 68ms/step - loss: 20.5605 - accuracy: 0.1772 - val_loss: 22.9010 - val_accuracy: 0.1800\n",
       "Epoch 13/100\n",
-      "63/63 [==============================] - 3s 54ms/step - loss: 21.4851 - accuracy: 0.1857 - val_loss: 22.9383 - val_accuracy: 0.2090\n",
+      "63/63 [==============================] - 4s 71ms/step - loss: 19.8422 - accuracy: 0.1840 - val_loss: 21.5172 - val_accuracy: 0.1800\n",
       "Epoch 14/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 20.2159 - accuracy: 0.1965 - val_loss: 22.2125 - val_accuracy: 0.2220\n",
+      "63/63 [==============================] - 4s 68ms/step - loss: 19.0382 - accuracy: 0.1950 - val_loss: 20.8096 - val_accuracy: 0.1810\n",
       "Epoch 15/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 19.6982 - accuracy: 0.1880 - val_loss: 21.6698 - val_accuracy: 0.2130\n",
+      "63/63 [==============================] - 4s 69ms/step - loss: 18.4138 - accuracy: 0.1822 - val_loss: 20.0728 - val_accuracy: 0.1740\n",
       "Epoch 16/100\n",
-      "63/63 [==============================] - 4s 61ms/step - loss: 18.6534 - accuracy: 0.1992 - val_loss: 20.7887 - val_accuracy: 0.2180\n",
+      "63/63 [==============================] - 4s 68ms/step - loss: 17.7271 - accuracy: 0.1885 - val_loss: 19.4107 - val_accuracy: 0.1810\n",
       "Epoch 17/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 18.1630 - accuracy: 0.2087 - val_loss: 20.9183 - val_accuracy: 0.2240\n",
+      "63/63 [==============================] - 4s 65ms/step - loss: 17.6251 - accuracy: 0.1760 - val_loss: 18.9971 - val_accuracy: 0.1980\n",
       "Epoch 18/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 17.4065 - accuracy: 0.2035 - val_loss: 19.7803 - val_accuracy: 0.2090\n",
+      "63/63 [==============================] - 5s 78ms/step - loss: 16.2954 - accuracy: 0.1970 - val_loss: 18.3570 - val_accuracy: 0.1950\n",
       "Epoch 19/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 16.9620 - accuracy: 0.1920 - val_loss: 19.0655 - val_accuracy: 0.2150\n",
+      "63/63 [==============================] - 4s 69ms/step - loss: 16.0583 - accuracy: 0.1935 - val_loss: 17.7447 - val_accuracy: 0.1880\n",
       "Epoch 20/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 17.0973 - accuracy: 0.1998 - val_loss: 18.5932 - val_accuracy: 0.2090\n",
+      "63/63 [==============================] - 4s 68ms/step - loss: 15.8284 - accuracy: 0.1805 - val_loss: 17.1217 - val_accuracy: 0.1910\n",
       "Epoch 21/100\n",
-      "63/63 [==============================] - 4s 56ms/step - loss: 16.0869 - accuracy: 0.2072 - val_loss: 18.2922 - val_accuracy: 0.2230\n",
+      "63/63 [==============================] - 4s 69ms/step - loss: 14.8200 - accuracy: 0.1975 - val_loss: 16.7020 - val_accuracy: 0.2060\n",
       "Epoch 22/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 16.0888 - accuracy: 0.2070 - val_loss: 17.7776 - val_accuracy: 0.1920\n",
+      "63/63 [==============================] - 4s 67ms/step - loss: 14.8035 - accuracy: 0.1877 - val_loss: 16.4884 - val_accuracy: 0.1930\n",
       "Epoch 23/100\n",
-      "63/63 [==============================] - 4s 61ms/step - loss: 15.3753 - accuracy: 0.2023 - val_loss: 17.2345 - val_accuracy: 0.2150\n",
+      "63/63 [==============================] - 4s 68ms/step - loss: 14.0777 - accuracy: 0.2013 - val_loss: 15.8489 - val_accuracy: 0.1890\n",
       "Epoch 24/100\n",
-      "63/63 [==============================] - 4s 56ms/step - loss: 14.9182 - accuracy: 0.2083 - val_loss: 17.0521 - val_accuracy: 0.2000\n",
+      "63/63 [==============================] - 4s 69ms/step - loss: 14.2543 - accuracy: 0.1830 - val_loss: 15.5276 - val_accuracy: 0.2040\n",
       "Epoch 25/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 14.8265 - accuracy: 0.1965 - val_loss: 16.3193 - val_accuracy: 0.2230\n",
+      "63/63 [==============================] - 4s 71ms/step - loss: 13.4636 - accuracy: 0.1875 - val_loss: 14.6891 - val_accuracy: 0.1890\n",
       "Epoch 26/100\n",
-      "63/63 [==============================] - 4s 61ms/step - loss: 14.4575 - accuracy: 0.1968 - val_loss: 16.4406 - val_accuracy: 0.2290\n",
+      "63/63 [==============================] - 4s 69ms/step - loss: 12.8913 - accuracy: 0.2037 - val_loss: 14.7856 - val_accuracy: 0.2010\n",
       "Epoch 27/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 14.2733 - accuracy: 0.2058 - val_loss: 15.5790 - val_accuracy: 0.2180\n",
+      "63/63 [==============================] - 4s 69ms/step - loss: 12.9247 - accuracy: 0.1965 - val_loss: 14.4304 - val_accuracy: 0.2000\n",
       "Epoch 28/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 13.9852 - accuracy: 0.2050 - val_loss: 15.5298 - val_accuracy: 0.2290\n",
+      "63/63 [==============================] - 4s 69ms/step - loss: 12.4494 - accuracy: 0.1972 - val_loss: 14.0287 - val_accuracy: 0.2110\n",
       "Epoch 29/100\n",
-      "63/63 [==============================] - 4s 62ms/step - loss: 13.9091 - accuracy: 0.2052 - val_loss: 15.2373 - val_accuracy: 0.2180\n",
+      "63/63 [==============================] - 4s 66ms/step - loss: 12.5309 - accuracy: 0.2058 - val_loss: 13.7324 - val_accuracy: 0.2090\n",
       "Epoch 30/100\n",
-      "63/63 [==============================] - 4s 56ms/step - loss: 13.2436 - accuracy: 0.1980 - val_loss: 14.8881 - val_accuracy: 0.2240\n",
+      "63/63 [==============================] - 4s 69ms/step - loss: 11.9367 - accuracy: 0.2002 - val_loss: 13.4195 - val_accuracy: 0.2140\n",
       "Epoch 31/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 13.2120 - accuracy: 0.2048 - val_loss: 14.3383 - val_accuracy: 0.2120\n",
+      "63/63 [==============================] - 4s 71ms/step - loss: 11.7203 - accuracy: 0.2035 - val_loss: 13.0633 - val_accuracy: 0.2100\n",
       "Epoch 32/100\n",
-      "63/63 [==============================] - 4s 56ms/step - loss: 12.6683 - accuracy: 0.2037 - val_loss: 14.2273 - val_accuracy: 0.2150\n",
+      "63/63 [==============================] - 4s 69ms/step - loss: 11.4939 - accuracy: 0.2030 - val_loss: 12.9967 - val_accuracy: 0.2120\n",
       "Epoch 33/100\n",
-      "63/63 [==============================] - 4s 60ms/step - loss: 12.5617 - accuracy: 0.2002 - val_loss: 14.5992 - val_accuracy: 0.2130\n",
+      "63/63 [==============================] - 4s 70ms/step - loss: 11.5324 - accuracy: 0.1910 - val_loss: 12.5197 - val_accuracy: 0.2280\n",
       "Epoch 34/100\n",
-      "63/63 [==============================] - 3s 55ms/step - loss: 12.2824 - accuracy: 0.2110 - val_loss: 14.0668 - val_accuracy: 0.2120\n",
+      "63/63 [==============================] - 4s 70ms/step - loss: 11.2674 - accuracy: 0.2002 - val_loss: 12.1535 - val_accuracy: 0.2180\n",
       "Epoch 35/100\n",
-      "63/63 [==============================] - 3s 54ms/step - loss: 11.9499 - accuracy: 0.2072 - val_loss: 13.5388 - val_accuracy: 0.2160\n",
+      "63/63 [==============================] - 4s 70ms/step - loss: 10.8380 - accuracy: 0.2093 - val_loss: 12.2492 - val_accuracy: 0.2010\n",
       "Epoch 36/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 12.3029 - accuracy: 0.1982 - val_loss: 13.3450 - val_accuracy: 0.2160\n",
+      "63/63 [==============================] - 4s 70ms/step - loss: 10.7348 - accuracy: 0.2025 - val_loss: 11.8892 - val_accuracy: 0.2110\n",
       "Epoch 37/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 11.6939 - accuracy: 0.2097 - val_loss: 13.2882 - val_accuracy: 0.2100\n",
+      "63/63 [==============================] - 4s 68ms/step - loss: 10.5163 - accuracy: 0.1928 - val_loss: 11.7170 - val_accuracy: 0.2050\n",
       "Epoch 38/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 11.4354 - accuracy: 0.2115 - val_loss: 12.7657 - val_accuracy: 0.2150\n",
+      "63/63 [==============================] - 4s 70ms/step - loss: 10.1681 - accuracy: 0.2023 - val_loss: 11.5104 - val_accuracy: 0.2070\n",
       "Epoch 39/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 11.3911 - accuracy: 0.2118 - val_loss: 12.6634 - val_accuracy: 0.2040\n",
+      "63/63 [==============================] - 4s 67ms/step - loss: 10.2094 - accuracy: 0.2007 - val_loss: 11.7209 - val_accuracy: 0.2120\n",
       "Epoch 40/100\n",
-      "63/63 [==============================] - 4s 61ms/step - loss: 10.9773 - accuracy: 0.2122 - val_loss: 12.4000 - val_accuracy: 0.2070\n",
+      "63/63 [==============================] - 5s 71ms/step - loss: 10.2874 - accuracy: 0.2030 - val_loss: 11.1544 - val_accuracy: 0.2140\n",
       "Epoch 41/100\n",
-      "63/63 [==============================] - 3s 53ms/step - loss: 11.0405 - accuracy: 0.2065 - val_loss: 12.3065 - val_accuracy: 0.2110\n",
+      "63/63 [==============================] - 4s 69ms/step - loss: 9.8548 - accuracy: 0.2000 - val_loss: 10.9128 - val_accuracy: 0.2240\n",
       "Epoch 42/100\n",
-      "63/63 [==============================] - 4s 60ms/step - loss: 10.6174 - accuracy: 0.2247 - val_loss: 12.1353 - val_accuracy: 0.2170\n",
+      "63/63 [==============================] - 4s 68ms/step - loss: 9.5805 - accuracy: 0.2013 - val_loss: 10.6923 - val_accuracy: 0.2160\n",
       "Epoch 43/100\n",
-      "63/63 [==============================] - 4s 56ms/step - loss: 10.8657 - accuracy: 0.2042 - val_loss: 12.2361 - val_accuracy: 0.2080\n",
+      "63/63 [==============================] - 4s 68ms/step - loss: 9.8389 - accuracy: 0.1980 - val_loss: 10.8398 - val_accuracy: 0.2050\n",
       "Epoch 44/100\n",
-      "63/63 [==============================] - 3s 55ms/step - loss: 10.7345 - accuracy: 0.2050 - val_loss: 11.7414 - val_accuracy: 0.2160\n",
+      "63/63 [==============================] - 4s 70ms/step - loss: 9.4249 - accuracy: 0.2060 - val_loss: 10.5591 - val_accuracy: 0.2190\n",
       "Epoch 45/100\n",
-      "63/63 [==============================] - 4s 60ms/step - loss: 10.7250 - accuracy: 0.2002 - val_loss: 11.7004 - val_accuracy: 0.2150\n",
+      "63/63 [==============================] - 4s 70ms/step - loss: 9.3991 - accuracy: 0.2105 - val_loss: 10.4659 - val_accuracy: 0.2060\n",
       "Epoch 46/100\n",
-      "63/63 [==============================] - 4s 60ms/step - loss: 10.2594 - accuracy: 0.2042 - val_loss: 11.4515 - val_accuracy: 0.2210\n",
+      "63/63 [==============================] - 5s 79ms/step - loss: 9.1690 - accuracy: 0.1988 - val_loss: 10.2464 - val_accuracy: 0.2140\n",
       "Epoch 47/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 10.2420 - accuracy: 0.2167 - val_loss: 11.4376 - val_accuracy: 0.2130\n",
+      "63/63 [==============================] - 4s 67ms/step - loss: 9.2827 - accuracy: 0.2035 - val_loss: 10.1138 - val_accuracy: 0.2190\n",
       "Epoch 48/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 10.1477 - accuracy: 0.2070 - val_loss: 11.4256 - val_accuracy: 0.2210\n",
+      "63/63 [==============================] - 5s 71ms/step - loss: 8.9197 - accuracy: 0.2155 - val_loss: 9.9359 - val_accuracy: 0.2200\n",
       "Epoch 49/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 10.1751 - accuracy: 0.2140 - val_loss: 11.1165 - val_accuracy: 0.2190\n",
+      "63/63 [==============================] - 4s 71ms/step - loss: 8.6668 - accuracy: 0.2205 - val_loss: 10.3965 - val_accuracy: 0.2140\n",
       "Epoch 50/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 10.0919 - accuracy: 0.2155 - val_loss: 11.3012 - val_accuracy: 0.1970\n",
+      "63/63 [==============================] - 4s 71ms/step - loss: 9.1509 - accuracy: 0.1980 - val_loss: 9.7671 - val_accuracy: 0.2210\n",
       "Epoch 51/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 10.2196 - accuracy: 0.2035 - val_loss: 11.1318 - val_accuracy: 0.2180\n",
+      "63/63 [==============================] - 4s 69ms/step - loss: 8.8055 - accuracy: 0.2120 - val_loss: 9.7981 - val_accuracy: 0.2210\n",
       "Epoch 52/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 9.8787 - accuracy: 0.1988 - val_loss: 10.9818 - val_accuracy: 0.2170\n",
+      "63/63 [==============================] - 4s 70ms/step - loss: 8.6455 - accuracy: 0.2058 - val_loss: 9.5791 - val_accuracy: 0.2190\n",
       "Epoch 53/100\n",
-      "63/63 [==============================] - 4s 58ms/step - loss: 9.8847 - accuracy: 0.2052 - val_loss: 11.2438 - val_accuracy: 0.2240\n",
+      "63/63 [==============================] - 4s 67ms/step - loss: 8.4916 - accuracy: 0.2097 - val_loss: 9.4260 - val_accuracy: 0.2280\n",
       "Epoch 54/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 9.4437 - accuracy: 0.2135 - val_loss: 10.9776 - val_accuracy: 0.2130\n",
+      "63/63 [==============================] - 4s 69ms/step - loss: 8.2758 - accuracy: 0.2107 - val_loss: 9.3284 - val_accuracy: 0.2210\n",
       "Epoch 55/100\n",
-      "63/63 [==============================] - 3s 55ms/step - loss: 9.5829 - accuracy: 0.2118 - val_loss: 10.7154 - val_accuracy: 0.2160\n",
+      "63/63 [==============================] - 4s 71ms/step - loss: 8.1894 - accuracy: 0.2090 - val_loss: 9.3506 - val_accuracy: 0.1980\n",
       "Epoch 56/100\n",
-      "63/63 [==============================] - 3s 55ms/step - loss: 9.2288 - accuracy: 0.2157 - val_loss: 10.5924 - val_accuracy: 0.2280\n",
+      "63/63 [==============================] - 4s 63ms/step - loss: 8.2820 - accuracy: 0.2128 - val_loss: 9.7846 - val_accuracy: 0.2130\n",
       "Epoch 57/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 9.2192 - accuracy: 0.2167 - val_loss: 10.7033 - val_accuracy: 0.2250\n",
+      "63/63 [==============================] - 4s 70ms/step - loss: 8.3598 - accuracy: 0.2090 - val_loss: 9.3707 - val_accuracy: 0.2040\n",
       "Epoch 58/100\n",
-      "63/63 [==============================] - 4s 60ms/step - loss: 9.2124 - accuracy: 0.2198 - val_loss: 10.4734 - val_accuracy: 0.2210\n",
+      "63/63 [==============================] - 4s 70ms/step - loss: 8.0148 - accuracy: 0.2160 - val_loss: 9.1976 - val_accuracy: 0.2070\n",
       "Epoch 59/100\n",
-      "63/63 [==============================] - 4s 61ms/step - loss: 9.0680 - accuracy: 0.2165 - val_loss: 11.0154 - val_accuracy: 0.2320\n",
+      "63/63 [==============================] - 4s 69ms/step - loss: 8.0729 - accuracy: 0.1995 - val_loss: 9.1469 - val_accuracy: 0.1970\n",
       "Epoch 60/100\n",
-      "63/63 [==============================] - 4s 57ms/step - loss: 9.2504 - accuracy: 0.2097 - val_loss: 10.4972 - val_accuracy: 0.2230\n",
+      "63/63 [==============================] - 4s 69ms/step - loss: 8.0123 - accuracy: 0.2175 - val_loss: 8.9841 - val_accuracy: 0.2120\n",
       "Epoch 61/100\n",
-      "63/63 [==============================] - 3s 55ms/step - loss: 8.9054 - accuracy: 0.2140 - val_loss: 10.3327 - val_accuracy: 0.2250\n",
+      "63/63 [==============================] - 4s 70ms/step - loss: 8.0255 - accuracy: 0.2075 - val_loss: 9.0406 - val_accuracy: 0.1970\n",
       "Epoch 62/100\n",
-      "63/63 [==============================] - 4s 59ms/step - loss: 8.7671 - accuracy: 0.2103 - val_loss: 10.5284 - val_accuracy: 0.2210\n",
+      "63/63 [==============================] - 5s 71ms/step - loss: 7.7979 - accuracy: 0.2138 - val_loss: 8.7216 - val_accuracy: 0.2030\n",
       "Epoch 63/100\n",
-      "63/63 [==============================] - 3s 53ms/step - loss: 9.0359 - accuracy: 0.2007 - val_loss: 10.1828 - val_accuracy: 0.2270\n",
+      "63/63 [==============================] - 5s 76ms/step - loss: 7.6717 - accuracy: 0.2048 - val_loss: 8.7191 - val_accuracy: 0.2020\n",
       "Epoch 64/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 8.7825 - accuracy: 0.2140 - val_loss: 10.3114 - val_accuracy: 0.2070\n",
+      "63/63 [==============================] - 5s 72ms/step - loss: 7.9337 - accuracy: 0.2080 - val_loss: 8.6152 - val_accuracy: 0.2200\n",
       "Epoch 65/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 8.8007 - accuracy: 0.2138 - val_loss: 9.9774 - val_accuracy: 0.2200\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 7.5857 - accuracy: 0.2095 - val_loss: 8.5533 - val_accuracy: 0.2170\n",
       "Epoch 66/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 8.4969 - accuracy: 0.2120 - val_loss: 9.8941 - val_accuracy: 0.2270\n",
+      "63/63 [==============================] - 4s 70ms/step - loss: 7.4746 - accuracy: 0.2150 - val_loss: 8.4395 - val_accuracy: 0.2160\n",
       "Epoch 67/100\n",
-      "63/63 [==============================] - 3s 48ms/step - loss: 8.5745 - accuracy: 0.2142 - val_loss: 9.7652 - val_accuracy: 0.2270\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 7.5529 - accuracy: 0.2085 - val_loss: 8.3611 - val_accuracy: 0.2250\n",
       "Epoch 68/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 8.2715 - accuracy: 0.2153 - val_loss: 9.7300 - val_accuracy: 0.2270\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 7.4024 - accuracy: 0.2135 - val_loss: 8.4891 - val_accuracy: 0.2080\n",
       "Epoch 69/100\n",
-      "63/63 [==============================] - 3s 48ms/step - loss: 8.5079 - accuracy: 0.2100 - val_loss: 9.6696 - val_accuracy: 0.2100\n",
+      "63/63 [==============================] - 5s 76ms/step - loss: 7.2022 - accuracy: 0.2135 - val_loss: 8.5370 - val_accuracy: 0.2220\n",
       "Epoch 70/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 8.1360 - accuracy: 0.2128 - val_loss: 9.5810 - val_accuracy: 0.2080\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 7.2330 - accuracy: 0.2157 - val_loss: 8.3368 - val_accuracy: 0.2150\n",
       "Epoch 71/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 8.2250 - accuracy: 0.2068 - val_loss: 9.4582 - val_accuracy: 0.2260\n",
+      "63/63 [==============================] - 5s 71ms/step - loss: 7.1719 - accuracy: 0.2163 - val_loss: 8.0086 - val_accuracy: 0.2210\n",
       "Epoch 72/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 7.9919 - accuracy: 0.2145 - val_loss: 9.4491 - val_accuracy: 0.2210\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 7.1011 - accuracy: 0.2190 - val_loss: 8.0636 - val_accuracy: 0.2130\n",
       "Epoch 73/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 8.2732 - accuracy: 0.1998 - val_loss: 9.4745 - val_accuracy: 0.2230\n",
+      "63/63 [==============================] - 5s 84ms/step - loss: 6.9326 - accuracy: 0.2268 - val_loss: 8.1927 - val_accuracy: 0.2320\n",
       "Epoch 74/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 8.2165 - accuracy: 0.2040 - val_loss: 9.2288 - val_accuracy: 0.2180\n",
+      "63/63 [==============================] - 5s 77ms/step - loss: 6.9288 - accuracy: 0.2260 - val_loss: 8.1182 - val_accuracy: 0.2160\n",
       "Epoch 75/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 8.1323 - accuracy: 0.2030 - val_loss: 9.1234 - val_accuracy: 0.2230\n",
+      "63/63 [==============================] - 4s 71ms/step - loss: 6.8918 - accuracy: 0.2338 - val_loss: 7.9932 - val_accuracy: 0.2310 0s - loss: 6.8640 - accuracy\n",
       "Epoch 76/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 7.8084 - accuracy: 0.2097 - val_loss: 9.2111 - val_accuracy: 0.2250\n",
+      "63/63 [==============================] - 5s 76ms/step - loss: 7.2133 - accuracy: 0.2115 - val_loss: 8.0421 - val_accuracy: 0.2040\n",
       "Epoch 77/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 7.7830 - accuracy: 0.2077 - val_loss: 9.4077 - val_accuracy: 0.2180\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 6.9548 - accuracy: 0.2142 - val_loss: 7.9215 - val_accuracy: 0.2200\n",
       "Epoch 78/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 7.8126 - accuracy: 0.2125 - val_loss: 8.8898 - val_accuracy: 0.2180\n",
+      "63/63 [==============================] - 5s 79ms/step - loss: 6.9369 - accuracy: 0.2185 - val_loss: 7.8404 - val_accuracy: 0.2120\n",
       "Epoch 79/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 7.6488 - accuracy: 0.2072 - val_loss: 8.9457 - val_accuracy: 0.2290\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 7.0997 - accuracy: 0.2120 - val_loss: 7.7585 - val_accuracy: 0.2230\n",
       "Epoch 80/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 7.5386 - accuracy: 0.2135 - val_loss: 8.8794 - val_accuracy: 0.2160\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 6.9580 - accuracy: 0.2150 - val_loss: 7.8094 - val_accuracy: 0.2220\n",
       "Epoch 81/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 7.7827 - accuracy: 0.2103 - val_loss: 8.7974 - val_accuracy: 0.2080\n",
+      "63/63 [==============================] - 5s 76ms/step - loss: 6.7121 - accuracy: 0.2202 - val_loss: 7.7070 - val_accuracy: 0.2130\n",
       "Epoch 82/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 7.4790 - accuracy: 0.2292 - val_loss: 8.6569 - val_accuracy: 0.2220\n",
+      "63/63 [==============================] - 5s 80ms/step - loss: 6.8451 - accuracy: 0.2180 - val_loss: 7.8513 - val_accuracy: 0.2430\n",
       "Epoch 83/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 7.3493 - accuracy: 0.2173 - val_loss: 8.8846 - val_accuracy: 0.1990\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 6.6499 - accuracy: 0.2310 - val_loss: 7.6191 - val_accuracy: 0.2230\n",
       "Epoch 84/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 7.5967 - accuracy: 0.2095 - val_loss: 8.4244 - val_accuracy: 0.2280\n",
+      "63/63 [==============================] - 5s 77ms/step - loss: 6.7305 - accuracy: 0.2198 - val_loss: 7.6156 - val_accuracy: 0.2000\n",
       "Epoch 85/100\n",
-      "63/63 [==============================] - 3s 48ms/step - loss: 7.3262 - accuracy: 0.2087 - val_loss: 8.5192 - val_accuracy: 0.2180\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 6.5236 - accuracy: 0.2210 - val_loss: 7.4502 - val_accuracy: 0.2270\n",
       "Epoch 86/100\n",
-      "63/63 [==============================] - 3s 46ms/step - loss: 7.1841 - accuracy: 0.2198 - val_loss: 8.4598 - val_accuracy: 0.2220\n",
+      "63/63 [==============================] - 5s 77ms/step - loss: 6.4832 - accuracy: 0.2313 - val_loss: 7.4636 - val_accuracy: 0.2400\n",
       "Epoch 87/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 7.3674 - accuracy: 0.2030 - val_loss: 8.4037 - val_accuracy: 0.2270\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 6.3933 - accuracy: 0.2150 - val_loss: 7.4960 - val_accuracy: 0.2260\n",
       "Epoch 88/100\n",
-      "63/63 [==============================] - 3s 48ms/step - loss: 7.3697 - accuracy: 0.2100 - val_loss: 8.3432 - val_accuracy: 0.2270\n",
+      "63/63 [==============================] - 5s 77ms/step - loss: 6.6315 - accuracy: 0.2118 - val_loss: 7.4906 - val_accuracy: 0.2370\n",
       "Epoch 89/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 7.2712 - accuracy: 0.2175 - val_loss: 8.4862 - val_accuracy: 0.2210\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 6.3671 - accuracy: 0.2307 - val_loss: 7.6169 - val_accuracy: 0.2300\n",
       "Epoch 90/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 6.8406 - accuracy: 0.2300 - val_loss: 8.3376 - val_accuracy: 0.2250\n",
+      "63/63 [==============================] - 5s 76ms/step - loss: 6.6556 - accuracy: 0.2170 - val_loss: 7.4098 - val_accuracy: 0.2310\n",
       "Epoch 91/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 7.0440 - accuracy: 0.2222 - val_loss: 8.0393 - val_accuracy: 0.2350\n",
+      "63/63 [==============================] - 5s 79ms/step - loss: 6.4345 - accuracy: 0.2222 - val_loss: 7.4402 - val_accuracy: 0.2240\n",
       "Epoch 92/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 6.9428 - accuracy: 0.2188 - val_loss: 8.2111 - val_accuracy: 0.2260\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 6.4228 - accuracy: 0.2205 - val_loss: 7.2602 - val_accuracy: 0.2320\n",
       "Epoch 93/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 6.9066 - accuracy: 0.2185 - val_loss: 8.1855 - val_accuracy: 0.2200\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 6.1637 - accuracy: 0.2202 - val_loss: 7.0793 - val_accuracy: 0.2170\n",
       "Epoch 94/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 6.9384 - accuracy: 0.2118 - val_loss: 7.9436 - val_accuracy: 0.2200\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 6.3256 - accuracy: 0.2062 - val_loss: 7.1821 - val_accuracy: 0.2160\n",
       "Epoch 95/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 6.8164 - accuracy: 0.2107 - val_loss: 8.0330 - val_accuracy: 0.2260\n",
+      "63/63 [==============================] - 5s 79ms/step - loss: 6.1190 - accuracy: 0.2250 - val_loss: 7.0112 - val_accuracy: 0.2360\n",
       "Epoch 96/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 6.9694 - accuracy: 0.2160 - val_loss: 8.3398 - val_accuracy: 0.2150\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 6.1672 - accuracy: 0.2188 - val_loss: 7.0114 - val_accuracy: 0.2250\n",
       "Epoch 97/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 6.8458 - accuracy: 0.2227 - val_loss: 7.9509 - val_accuracy: 0.2330\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 6.0777 - accuracy: 0.2145 - val_loss: 7.5616 - val_accuracy: 0.2170\n",
       "Epoch 98/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 6.7733 - accuracy: 0.2198 - val_loss: 8.0392 - val_accuracy: 0.2200\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 6.0525 - accuracy: 0.2177 - val_loss: 7.2203 - val_accuracy: 0.2060\n",
       "Epoch 99/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 6.9146 - accuracy: 0.2132 - val_loss: 7.9781 - val_accuracy: 0.2170\n",
+      "63/63 [==============================] - 6s 90ms/step - loss: 6.0467 - accuracy: 0.2227 - val_loss: 7.1099 - val_accuracy: 0.2290\n",
       "Epoch 100/100\n",
-      "63/63 [==============================] - 3s 48ms/step - loss: 6.7163 - accuracy: 0.2202 - val_loss: 8.2036 - val_accuracy: 0.2310\n"
+      "63/63 [==============================] - 4s 71ms/step - loss: 6.2256 - accuracy: 0.2235 - val_loss: 7.1370 - val_accuracy: 0.2340\n"
      ]
     }
    ],
@@ -2289,14 +2308,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 36,
+   "execution_count": 37,
    "metadata": {
     "id": "Q6_cz3hgvXZ7"
    },
    "outputs": [
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAewAAAHiCAYAAAAqIP8QAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/d3fzzAAAACXBIWXMAAAsTAAALEwEAmpwYAACqG0lEQVR4nOydd3xkdbn/38/0TCY92wu7LEsvCyygoHQRG1gQ4aLCtWD3qter3qs/5Vqu9drutWEvKKJeBRVERIqKhaWzsMCyu0C2JtnUmUz//v74njNzZjJJJruTnUzyvF+vvGbmzDlnvjNJ5nOeLsYYFEVRFEWZ3fjqvQBFURRFUaZGBVtRFEVRGgAVbEVRFEVpAFSwFUVRFKUBUMFWFEVRlAZABVtRFEVRGoB5JdgicpOIXF7rfeuJiGwTkXNn4Ly3i8gbnPuXicjvq9l3H15npYiMioh/X9eqKNWi3wHTOq9+B8wyZr1gO79I9ycvImOex5dN51zGmBcYY75f631nIyLyARG5s8L2bhFJi8jR1Z7LGHONMea8Gq2r5MvFGPO0MSZmjMnV4vwVXk9EZIuIPDIT51dmHv0O2Df0OwBExIjIIbU+b72Y9YLt/CJjxpgY8DTwEs+2a9z9RCRQv1XOSn4EnCoiq8u2XwI8ZIx5uA5rqgenAwuBg0XkpAP5wvo3WRv0O2Cf0e+AOcasF+yJEJEzRaRHRN4vIruA74pIh4j8RkR6RWTAub/cc4zXxXOFiPxZRD7n7LtVRF6wj/uuFpE7RWRERP4gIl8RkR9NsO5q1vgxEfmLc77fi0i35/nXiMhTItIvIh+c6PMxxvQAfwReU/bUa4EfTLWOsjVfISJ/9jx+nohsEpEhEflfQDzPrRGRPzrr6xORa0Sk3Xnuh8BK4NeOdfQ+EVnlXAUHnH2WisgNIrJXRDaLyBs9575KRK4TkR84n81GEVk/0WfgcDlwPXCjc9/7vo4SkVuc19otIv/hbPeLyH+IyJPO69wjIivK1+rsW/538hcR+YKI9ANXTfZ5OMesEJH/c34P/SLyvyISctZ0jGe/hSKSEJEFU7zfeYN+B+h3QJXfAZXeT5tzjl7ns/yQiPic5w4RkTuc99YnIj91tovzv71HRIZF5CGZhpeiFjSsYDssBjqBg4Arse/nu87jlcAY8L+THH8K8BjQDXwG+LaIyD7s+2PgH0AXcBXj/0G8VLPGfwL+GWsZhoD3AojIkcDXnPMvdV6v4j+Yw/e9axGRw4B1znqn+1m55+gG/g/4EPazeBI4zbsL8ElnfUcAK7CfCcaY11BqIX2mwktcC/Q4x18E/JeInO15/gJnn3bghsnWLCJR5xzXOD+XiEjIea4F+APwO+e1DgFudQ59D3Ap8EKgFXgdkJjsc/FwCrAFWAR8gkk+D7Exu98ATwGrgGXAtcaYtPMeX+0576XArcaY3irXMV/Q7wD9DphyzRX4H6ANOBg4A3sR88/Ocx8Dfg90YD/b/3G2n4f12B3qHHsx0L8Pr73vGGMa5gfYBpzr3D8TSAORSfZfBwx4Ht8OvMG5fwWw2fNcFDDA4unsi/1DzwJRz/M/An5U5XuqtMYPeR6/Ffidc//D2C9097lm5zM4d4JzR4Fh4FTn8SeA6/fxs/qzc/+1wN88+wn2n+sNE5z3pcB9lX6HzuNVzmcZwP5j54AWz/OfBL7n3L8K+IPnuSOBsUk+21cDvc65I8AQ8DLnuUu96yo77jHgwgrbC2ud5HN6eorfd+HzAJ7trq/Cfqdgv9jEebwBuHim/8dm+w/6HaDfAdP7DjDAIWXb/M5ndqRn25uA2537PwCuBpaXHXc28DjwLMBXj7//Rrewe40xSfeBiERF5BuOi2MYuBNol4mzD3e5d4wxrgUVm+a+S4G9nm0Az0y04CrXuMtzP+FZ01LvuY0xcSa5wnPW9DPgtY4lcBn2j3FfPiuX8jUY72MRWSQi14rIdue8P8JehVeD+1mOeLY9hbU8Xco/m4hMHLu8HLjOGJN1/k5+QdEtvgJrGVRisuemouR3P8XnsQJ4yhiTLT+JMebv2Pd3pogcjvUA3LCPa5rL6HeAfgdM9h1QiW4g6Jy30mu8D3sR8g/H5f46AGPMH7HW/FeAPSJytYi0TuN195tGF+zyUWP/ChwGnGKMacW6L8ATX5kBdgKdjvvVZcUk++/PGnd6z+28ZtcUx3wf67p5HtAC/Ho/11G+BqH0/f4X9vdyjHPeV5edc7LxcDuwn2WLZ9tKYPsUaxqH2Fjc2cCrRWSX2BjnRcALHZfeM1h3WCWeAdZU2B53br2/68Vl+5S/v8k+j2eAlZN82Xzf2f81wM+9wqQU0O8A/Q6YLn1ABhsKGPcaxphdxpg3GmOWYi3vr4qTaW6M+bIx5kSsZX8o8G81XNeUNLpgl9OCjcMMikgn8JGZfkFjzFNYd+VVYpOFng28ZIbW+HPgxSLyHCcW+1Gm/h3+CRjEunjc+Oj+rOO3wFEi8nJHaN5JqWi1AKPAkIgsY/wf9G4mEEpjzDPAXcAnRSQiIscCr8deoU+X12DdV27Mbh32H6wH6w7/DbBERN4lImERaRGRU5xjvwV8TETWOokmx4pIl7Hx4+3YiwC/c+VdSdi9TPZ5/AP75fcpEWl23rM3Fvgj4GXYL7wf7MNnMB/R74DxzNfvAJeQc66IiEScbdcBn3D+7w/C5q38CEBEXinF5LsB7AVGXkROEpFTRCSIvXhPAvn9WNe0mWuC/UWgCXsF9TdsQtGB4DJsPLIf+DjwUyA1wb5fZB/XaIzZCLwNmzCyE/vH1DPFMQb7ZX8QpV/6+7QOY0wf8ErgU9j3uxb4i2eX/wROwMaLf4tNTvHySeBDIjIoIu+t8BKXYmNaO4BfAh8xxvyhmrWVcTnwVedqufADfB243HG5PQ/7xboLeAI4yzn289h/6N9j43/fxn5WAG/EfgH1A0dhv1wmY8LPw9i605dg3d1PY3+Xr/I8/wxwL/YL40/T/wjmJV9EvwPKj5mv3wEuG7EXJu7PPwPvwIruFuDP2M/zO87+JwF/F5FRbBjqX4wxW7AJqN/EfuZPYd/7Z/djXdPGTWhRaojYMoBNxpgZv7pX5jYi8h1ghzHmQ/Vei1I9+h2gzARzzcKuC46rZI2I+ETkfOBC4Fd1XpbS4IjIKuDlWAtfmcXod4ByINDOQLVhMdbt04V1T73FGHNffZekNDIi8jHg3cAnjTFb670eZUr0O0CZcdQlriiKoigNgLrEFUVRFKUBUMFWFEVRlAagqhi2k0TxJWxLt28ZYz5V9vx7gDdg2/P1Aq9zahPd51uBR4BfGWPePtlrdXd3m1WrVk3nPSjKvOSee+7pM8bM6mEg+v+sKNVRzf/zlILttKn7CrZmtQe4W0RuMMZ45wvfB6w3xiRE5C3Yxviv8jz/MWzbuylZtWoVGzZsqGZXRZnXiMhTU++1z+f+F2zNuQDfNMZ80Wmu8VNsjew2bG/zgcnOo//PilId1fw/V+MSPxnb9H6LKU4RutC7gzHmNk8f3b/hmR4jIidiJxf9vtqFK4pSP8SODHwj9n//OGxnrUOAD2Anhq3FTjb7QP1WqSjzj2oEexmljex7KG3EXs7rgZsAxM4X/W+c0XCKojQERwB/N8YknMEkd2DrwS/E9qXGuX1pfZanKPOTmiadicirgfUU27W9FbjR2EHqkx13pYhsEJENvb067ldR6szDwHNFpMsZLvFC7HCHRcaYnc4+u7CeM0VRDhDVJJ1tp3QSy3IqTE4RkXOBDwJnGGPcHrrPxv7jvxU7Hi4kIqPGmBJXmjHmamxjetavX6+F4YpSR4wxj4rIp7FhrDhwP3ZGsXcfIyIV/1dF5ErgSoCVK1fO7GLnMZlMhp6eHpJJHeLWSEQiEZYvX04wGJz2sdUI9t3AWhFZjRXqS4B/8u4gIscD3wDON8bscbcbYy7z7HMFNjFN416KMssxxnwbpyWqiPwXNhS2W0SWGGN2isgSYM8Ex+oF+AGgp6eHlpYWVq1ahZ1wqcx2jDH09/fT09PD6tWrp338lC5xJ4b1duBm4FHgOmPMRhH5qIhc4Oz2WawF/TMRuV9Ebpj2ShRFmTWIyELndiU2fv1j7OSiy51dLgeur8/qFIBkMklXV5eKdQMhInR1de2zV6SqOmxjzI3AjWXbPuy5f24V5/ge8L3pLU9RlDrxCxHpAjLA24wxgyLyKeA6EXk9drzgxXVdoaJi3YDsz+9MO50pijIOY8xzjTFHGmOOM8bc6mzrN8acY4xZa4w51xizt97rVOpHf38/69atY926dSxevJhly5YVHqfT6UmP3bBhA+985zunfI1TTz21Jmu9/fbbefGLX1yTc9UTndalKIqiTJuuri7uv/9+AK666ipisRjvfW+xgjebzRIIVJaY9evXs379+ilf46677qrJWucKamEriqIoNeGKK67gzW9+M6eccgrve9/7+Mc//sGzn/1sjj/+eE499VQee+wxoNTiveqqq3jd617HmWeeycEHH8yXv/zlwvlisVhh/zPPPJOLLrqIww8/nMsuuwx30uSNN97I4Ycfzoknnsg73/nOaVnSP/nJTzjmmGM4+uijef/73w9ALpfjiiuu4Oijj+aYY47hC1/4AgBf/vKXOfLIIzn22GO55JJL9v/D2gfUwlYURWlw/vPXG3lkx3BNz3nk0lY+8pKjpn1cT08Pd911F36/n+HhYf70pz8RCAT4wx/+wH/8x3/wi1/8YtwxmzZt4rbbbmNkZITDDjuMt7zlLePKnu677z42btzI0qVLOe200/jLX/7C+vXredOb3sSdd97J6tWrufTSS6te544dO3j/+9/PPffcQ0dHB+eddx6/+tWvWLFiBdu3b+fhhx8GYHBwEIBPfepTbN26lXA4XNh2oFELW1EURakZr3zlK/H7/QAMDQ3xyle+kqOPPpp3v/vdbNy4seIxL3rRiwiHw3R3d7Nw4UJ27949bp+TTz6Z5cuX4/P5WLduHdu2bWPTpk0cfPDBhRKp6Qj23XffzZlnnsmCBQsIBAJcdtll3HnnnRx88MFs2bKFd7zjHfzud7+jtbUVgGOPPZbLLruMH/3oRxO6+mcatbAVRVEanH2xhGeK5ubmwv3/9//+H2eddRa//OUv2bZtG2eeeWbFY8LhcOG+3+8nm83u0z61oKOjgwceeICbb76Zr3/961x33XV85zvf4be//S133nknv/71r/nEJz7BQw89dMCFWy1sRVEUZUYYGhpi2TI7euJ73/tezc9/2GGHsWXLFrZt2wbAT3/606qPPfnkk7njjjvo6+sjl8vxk5/8hDPOOIO+vj7y+TyveMUr+PjHP869995LPp/nmWee4ayzzuLTn/40Q0NDjI6O1vz9TIVa2IqiKMqM8L73vY/LL7+cj3/847zoRS+q+fmbmpr46le/yvnnn09zczMnnXTShPveeuutLF9eGCTJz372Mz71qU9x1llnYYzhRS96ERdeeCEPPPAA//zP/0w+nwfgk5/8JLlcjle/+tUMDQ1hjOGd73wn7e3tNX8/UyFupt1sYf369Ubn5yrK1IjIPcaYqWtj6oj+P88cjz76KEcccUS9l1F3RkdHicViGGN429vextq1a3n3u99d72VNSqXfXTX/z+oSVxSlbiQzOYbGMvVehtLAfPOb32TdunUcddRRDA0N8aY3vaneS5oxVLAVpZZsuQM+swaStS2xmat87DePcM5/317vZSgNzLvf/W7uv/9+HnnkEa655hqi0Wi9lzRjqGArSi3ZvRESfTA8bgKtUgGfCLn87ArLKcpsRQVbUWpJyrGsxwbruoxGwe9TwVaUalHBVpRakhqxt8mh+q6jQfCJMMvyXhVl1qKCrSi1xBXq5GBdl9Eo+ARyqtiKUhUq2IpSS1yXuFrYVaEu8cblrLPO4uabby7Z9sUvfpG3vOUtEx5z5pln4pb5vfCFL6zYk/uqq67ic5/73KSv/atf/YpHHnmk8PjDH/4wf/jDH6ax+srM9jGcKtiKUkuS04hhZ9Nwy4fntbj7fOoSb1QuvfRSrr322pJt1157bdX9vG+88cZ9bj5SLtgf/ehHOffcc/fpXI2ECrai1JLpWNg7H4C/fAm23jmza5rFqEu8cbnooov47W9/SzqdBmDbtm3s2LGD5z73ubzlLW9h/fr1HHXUUXzkIx+pePyqVavo6+sD4BOf+ASHHnooz3nOcwojOMHWWJ900kkcd9xxvOIVryCRSHDXXXdxww038G//9m+sW7eOJ598kiuuuIKf//zngO1odvzxx3PMMcfwute9jlQqVXi9j3zkI5xwwgkcc8wxbNq0qer3OlvGcGprUkWpJa6FXU0MO5u0t+nEjC1ntuPXsq7acNMHYNdDtT3n4mPgBZ+a8OnOzk5OPvlkbrrpJi688EKuvfZaLr74YkSET3ziE3R2dpLL5TjnnHN48MEHOfbYYyue55577uHaa6/l/vvvJ5vNcsIJJ3DiiScC8PKXv5w3vvGNAHzoQx/i29/+Nu94xzu44IILePGLX8xFF11Ucq5kMskVV1zBrbfeyqGHHsprX/tavva1r/Gud70LgO7ubu69916++tWv8rnPfY5vfetbU34Ms2kMp1rYysyz+VaI99V7FQeG6ZR1FQT7wA8RmC34fALAbGuRrFSH1y3udYdfd911nHDCCRx//PFs3LixxH1dzp/+9Cde9rKXEY1GaW1t5YILLig89/DDD/Pc5z6XY445hmuuuWbC8Zwujz32GKtXr+bQQw8F4PLLL+fOO4serJe//OUAnHjiiYWBIVMxm8ZwqoWtzCzZNFzzSjjj/XDm++u9mpknOQ2XuCvYmflrYfvECnYubwj4pc6raWAmsYRnkgsvvJB3v/vd3HvvvSQSCU488US2bt3K5z73Oe6++246Ojq44oorSCaT+3T+K664gl/96lccd9xxfO973+P222/fr/W6IzprMZ6zHmM41cJWZpbUCJic7f4118llIDtm71flErexNdLxGVvSbMfvWNgax25MYrEYZ511Fq973esK1vXw8DDNzc20tbWxe/dubrrppknPcfrpp/OrX/2KsbExRkZG+PWvf114bmRkhCVLlpDJZLjmmmsK21taWhgZGRl3rsMOO4xt27axefNmAH74wx9yxhln7Nd7nE1jONXCVmaW9DxqJJLyfIFU834zjriXC/aeTfb4lafUbm2zFMfA1kzxBubSSy/lZS97WcE1ftxxx3H88cdz+OGHs2LFCk477bRJjz/hhBN41atexXHHHcfChQtLRmR+7GMf45RTTmHBggWccsopBZG+5JJLeOMb38iXv/zlQrIZQCQS4bvf/S6vfOUryWaznHTSSbz5zW+e1vuZzWM4dbymMrPsegi+/hxY+3y47Lp6r2Zm2bsVvrwOIu2Qz8F/9Ey+/z++CTe+F9a/Hl78+eL2X70NttwG75k47gdzY7zmN+54kk/etImN//l8msNqP0wHHa/ZuOh4TWV2knLcQfPCwnbi1+0rrWchN0WMzHWJl8ewM3EIzt2JQ15cl3h+lhkOijIbUcFWZpZCb+3Bui7jgJD0CDYUBXwiClniZS7xdAKCTbVd2yxFHJ+441VUFGUSVLCVmWWexLCf6o/zzu/fYR+0rbC3U12kTCTYmQSEmmu6vtmKmxiuSWeKMjUq2MrM4lrYc3zc5JbeOAH34qTdEeyp3vNEZV2Z+WNhq0t8/5htOUjK1OzP70wFW5lZ3Bh2dqwYs52l5PKGz968ib3x9LSPTaRztIiT9e26xKe0sN2yrrKyj8zYvIlhF13iKjzTJRKJ0N/fr6LdQBhj6O/vJxKJ7NPxmpZ5IHnyNrjvh/CKb9t6lj99HgIRePZb672ymaO81Cm2sH5r8XLX/9hM7ue8q7Dpyd5RvnLbkxzcHeMVJy6f+NgKxNNZWnAs5YJLfIowwEStSdPx+eMS1zrsfWb58uX09PTQ29tb76Uo0yASiZSUjU0HFewDyeM3w8O/gAu/Yl2eG39pv5jnsmB7rcfZJNiPXA8mXyLYqYzNfIqnp98BKZHKEpMxMhIm6L7HqVzimUli2PPFJe5a2KrX0yYYDLJ69ep6L0M5gKhgH0hGd9nb1Ij9Qk6NWNGYy3gzpWdTHDs1vktSKpsDIJ7KTft08XSODhIk/c0EI212Y7UW9rgY9hgE54eF7TZOUZe4okyNxrAPJCO77a0rFqmRqUt/Goln/gEbvlO6LVVmYc8WUiPjXNHprL14SrgW9uDTcMdnq2rDlUhnaZEEcYna+LMvOL0Ytvsaxjgu8fkRwy64xFWwFWVKVLAPJKNlgp0eLRW0Ruee78GN77MDP1xSIxCK2fuzqRY7NTou2SvlCPZoyhHsTb+F2z4OI7umPF08laOFMUZM1JqNkbYqLGwnSc3ki+KdTQJm/rjENUtcUapGBftA4hXsXMZ+OVdwzTYsqRHIZ6DXMxg+PQptToLF2EB91lWOMbY+vMwV7brEE65LPOdceFThBXEt7GHjCG1TexVlXZ6seTeO7fYXnzcucRVsRakWjWEfKLwWXXq0KNT5jP3iDoTrt7Za4YrOzgdgiTOsPjViBbt3U2WLc3gn7N0Cq8oGBDx+M/Q+VnwcaobjXwOB0P6vM5NwrNqkzRT3+e1SXQvbdYnnMsX34OWpu6BnA4gPjn45tC4lns7RQoKnc112n0hb9Y1TwLYjpav4Gc4Xl3hhvGadF6IoDYAK9oHCta7BiV2PlD6ea4LNa+z91DAsOtqWr1USsL98ER74CXzg6eK2fB5++uqihevSvRZWn77/6/R+9uk4ROzgeTdLPOG6xPPObfmFxq/eCgNb7f3RXXDex0mksrTIGHtzEfJ5gy/SXn0MG4rxdNfqnyd12H7Hx6cWtqJMjbrEDxTeOGhqpDR+Olfc4u572vVgcVtqFMIxO8GqkoU91DO+rCk5aMX63Kvg37fDP99UPFct8J7H4xYflyVeySWez9k1P/vt0H6Q9RBAwcIeMU2MJLPWwp6yrGvMfi7gcYnPL8EW0aQzRakWFewDxWiZYJdb2HOBgmA/ZIUN7HsLt0wsYKN7rCWb95RSJfrtbesyK/ZNHfax14Xsnnt0H5pGeAXYc7HgusTj5S7xpGf/0T02jNG5GlqWFDwnyVSKmCQZoYmBRNrGsCtdoIzsKlrT2RREHRd6xlmH+9w8c4mrha0oU6OCfaAY3VO8X54dPlcEOzVqM8IzCeh/0gpSPmO3TSRg7oWM1z0c77O3rpgFnDZ+5YJ9y4fhRy+b/jq93o0Kgp1IOxcPrkvcK/BDzozrthXQsqgg2Mb5fY6YKHsTacejMDh+DNU3z4E//Xfx/bjvcZ5a2MUs8TovRFEaABXsA8XILlubG24dX39d3ku6UUnHYcUp9v7OB4oXIuHWyklYxhRr03NewXas5uYF9tYVbDeD2mV0D/Q+Pv3ZjOUxbHdzeVlXJQt72BHs1mUQW1xYvy9t9xkmymAibTu65bOlmfGZMXv88A77OJuCaGfpOuaZYLuNU9QlrihTo4J9oBjdDbFF1j1cXgM8FyzsfM7WFS87wQrsLq9gTxDDTg4WhdprYSccC7u5294GJ7Cws0l7vLv/RJS7W0ti2F7Bdsu63KQzN0u8koW93FrYqSHIjBFIFy3sgXjG/q6hNBTiJh6mR+yasmPjLWzXJV7nOmwRebeIbBSRh0XkJyISEZHVIvJ3EdksIj8Vkf1O2dc6bEWpHhXsA8XILvsFH26xAnAAY9gjycw+TaCaFu4FSKQNFh4JOx/0CLYnhr39XvjkCti7tWhdQ5lL3IlhF1zijniVC7bbi3vomYnXte3P8OlVpSGJiWLYhV7iOdsqM+dmiZcJdsh5P44om5FdBLP2vRZi2C2L7f7eZENvpzs3oc21sF3L2r2AqOPwDxFZBrwTWG+MORrwA5cAnwa+YIw5BBgAXr+/r+XTaV2KUjUq2AeK0d3WhRpuOeAx7P/89SO8+Uf3zOhrFOuHY7DgMOjfXBRxN4adGoaHfm5vt99Tan1my1zi4dZiqZt7m6lgYQMMbZ94XQ/9zFry/U961uqNYXuzxIuu9bFMbmILu2259eXGrCinB3ewzNgLgl0ssIJdsLA9FyXeXvLu2gsWtrOmQuOUurvEA0CTiASAKLATOBv4ufP894GX7u+LuIKt07oUZWpUsA8UroUdihVj2G5sdoZj2HtGUvSNzvAs6pRHnDvXwPD2oli5MWyTtxPKwDZL8Vq9uTKXuOsOByuOgUixlaeLK/Kum7ocY+CJW+z9uCebvCSGXfzsXZc4QDyVrVzWNfRMsXNbixXl9OAuVvt2khc/o5GlDCQyU1jYo8W1h1vAFyheOKTrH8M2xmwHPgc8jRXqIeAeYNAY444y6wGW7e9ruS5x1WtFmRoV7GoZ2QU/uLDorp0O2TSM7R0fww63FgU8n4PrXgtP/63mS09lcgV3b01IDsEPXwYDTxW3ucIXjkHXGnt/54PFbW698YiTcNX/ZKmYlWeJRz2CDVawx1nYjoBPJNi7N9oLByiNc6dGbQIglNVhFz+jeDpXOenMtbChYGFnh3aySnYRjy4n1txkk85CzdZ1PpGF7VrSgYjd15t05g+Bv349jUSkA7gQWA0sBZqB86dx/JUiskFENkw1q9mnSWeKUjUq2NXy9F9hy+2w++HpHxt3LMmCYI8U65PdxyO77IzmrXfWdNlghShdy96Pux6GJ/8IPXcXt6U9sdeCYN9vb90Ytkv7SusyH50ohl1mYYNNwprQwp4ghv3EzZ5zei60UiPFudyeGHbaK9ip7PiyrsyYrRFvcwzLaBf4AuSHd7FadjEWW0VHNGSTzqCk7AsoWthpj4UdiNi+4RmPYNffHX4usNUY02uMyQD/B5wGtDsucoDlQMVYhDHmamPMemPM+gULFkz6Qj6fusQVpVpUsKvFdd+WlxZVg/tF3bK4TLBjRQt71PNlXmNS2XyJGO03rrVaybUcikHnwfb+jvuL25ra7f3ONbDmHNhbZmFP5hIHxyVe5tYvxLAnsLCfuAWWHAfhtjKX+LD1bgSjFcu6wHWJl1nYbqy8bYW99fmgeSGMWgs71baK9mjIxrChpOwLKB3+ki23sD0u8foL9tPAs0QkKrYV2TnAI8BtwEXOPpcD1+/vC7mNU4wKtqJMiQp2tbjiUjbhqcDGX1oLvBKuKzTmxLDTrmC3FgW8INjxyufYD1KZXG0F2xW/Sg1IQjH7nmKLbBjA3ea6xA99vrXAxwbsQJBCUxRHjPN5a8WWu8SDTeMvltxjXLf3P75ph3KAPf8zf4e150FzV6lLPD1q1xiMlrrEMzlCAfsvkUhXSDpzLXnXJQ4QW0i4byPNkiLXcTAd0aBHsBeWlXW5902xm1sgYruaeV3ide5yZoz5Oza57F7gIez3xNXA+4H3iMhmoAv49v6+lk+HfyhK1ahgV4srqBNZ2Ld/Cv729crPuS05mzqsUJi8tdhDMWtlp0eLFwQzMB87lc2XJFTtN657uVJpmluO1HWI87jFWqJda2DN2XbiVqfjMt+90brHoSi+yUHrih5nYYcrlHWNAWJ/N0Pb4cZ/s6IN1m1v8nDQqbYBS9wbw3bCEd7YMfZz6mq2pcWjqWyxrCuXtvFzbw22S8timgfsOFHTuYbO5hADiYy1GFvKLOyR3SB+5zO06/nEzVtISsRT1jUrLGyMMR8xxhxujDnaGPMaY0zKGLPFGHOyMeYQY8wrjTH7ncnoc76BNIatKFOjgl0trqCWx1Fd0omiRVaOm20cCFuBBhjZ6cSwWw+ASzxH3kC2VmZMRZe4I3zu+3Pd4u7jUDO85pew6MiimGPsAA0oCrZreY5LOmsqTTrLOxawK/j3/sCez7W2XXFtP8ieK+GNYTsDSSoIdqcj2Il0tvT3mRp2zinQsrS4PbYIwYpNYMFa2qMh0tm8LQuLLbKx6dSIFf94L3Q479cR7Lu3j9GXCnhGr8ZnhWAfKIpZ4irYijIVKtiV2H6PbezhZSoLOxMvxjzLcQXbH7YCDdaSKiSdjc6sYDsZ4jVLPItPEsMOuha2Y0WHW8Yf37HKzpKGouC6MWz33OOSziKlFrYr8K743/t9e+u6rV3Bbl1qXeLlZV0VLewc58gGQmTsxC7veM+kI9gti0tncjvlWykTINy5omCh946kPKVdu53XN8X1OutJEaQ/E/SM1xyru0v8QKJ12IpSPSrYlfi/K+G2T5RuK8SwJxLssdKJU14Kgh20bnCXQtLZsCeDeAZi2E78umZxbFf8ygU70FQsR3Ld3t736xIIFRO3yl3ihT7iFSzsEsF27rsXBiN2zCXDO2wcfLjHWtbBJusST/QXi33To9ZVXxbDXpju4T39V3Gu7x4n6Sxra6TBtiAdfKrUHQ6FBilPm0VEm8KsXmAvWJ7sHS1tnuLGr7vWlrzPFEF2JwOlMex5ZGH7dLymolSNCnYlxgZLXai5TNENXCnpLJ+326dyiftDpRana2GnRz01urW1sPN5U7CsU7USbPezKU8687bTnMzC9j5fLtju5zzOJR4uvVhyBdu9MABY9Vz7Wcd7S+ulo902Lp4ctKI9gYUdztr30+pL2zrsfAaanNahqRHb7MX7elCworeZxTQF/Ry60L7fx3ePFi3s0V3FKgP3fTuehKQJsXvMhymM15yvLvE6L0RRGgAV7EpkEqWzm73u1EoWtiseE7nEs17B9licoRb7OJ+FgW12W40tbK9IT2hhu68NVjgna/UJlV3iblzYpRDDnkiwHdewG9N1P0M3oa1iHXYFC7up3VrQvoBNaAMr1l7Bds8V73M+X1Mxhu3P2Yux5mC+WNbltg4d3WPj411lgu1a2LIUv09oiwZZ2BLmid0eC3tkd9FDU8ElPmrCmJRrYc83l7i9VQtbUaamKsEWkfNF5DFnSs8HKjz/HhF5REQeFJFbReQgZ/s6EfmrM/XnQRF5Va3fQM1xrWXvKEhvvXAlC9vdNpmF7QvYlFg3hg3FpDOobLXWAG92eEULe+eD8KXj4BmnCcrfvwFfOXni/uZu2RWUegPS8VL3d7AJug+zMeRKLD7GXsB0rLaPXS9EeR9xl8AEMexA2L7O6tNtQhvYOPZQT9Ht7opuvK90IElZHbY/Z88f8+eIp53GKe6xOx+wt+WC3b6SHH62+lcXNq1dFOOJPSO2KsAfcizs3aXHO56EUKSJhAnjy6ftBUImUcwDmAdoDFtRqmfK/oci4ge+AjwP2z/4bhG5wRjziGe3+7CTfRIi8hbgM8CrgATwWmPMEyKyFLhHRG42xgzW+o3UDFd8vaMgvd2qKlnY7pe+WwZUTi5tE85gfAxbPNdMgaYZEOy8536FGHv/Zns7+BSsOAkGn7Zr2HI7HPGS8fsnB8E45ymJYY+MnzB1xW8mdu+uuwxWn2GnVfk9JVuJvqJIegmWZYl7W3te/AN7MeR+6e95xL4HtyOZO1fb25Al1GJ/F87v2xhDMDcGfmgOeCzsZmct2++1t+Uu8dhCPrnmR9y5o5iItnZhC9dteIa8AV9skbWwQ6NWwL0XD8ChyxaQ2+58Rum44xKv72jNA4lmiStK9VRjYZ8MbHZqMNPAtdg+wwWMMbcZY1zT82/YtoUYYx43xjzh3N8B7AEm71VYb1zxHRssCoBrYUfaKwu2uy0/mWA7vasrxbBdOg+2+2bT1ApvD/GKLnE3tjo24Nw6zU6e+H3lE7ru8KYOK9KFk8fHJ5jFFpa6yb34/EV3eCBcfM+V2pJCcfiH+zvxtvZs7rLraeqw1unTf7XPjXOJ9xaboIRbig1LjI3zN2HXEPXlio1T3Bj2RBY28LRZSDTsEexFMRLpHNsHx6xbfHi7TYaLLbbr9dkyrhw+OmJROtra7YFjewHDHdviXPyNv1b+3OYY2jhFUaqnGsFeBnibNU81pef1wE3lG0XkZCAEPDnuiNmEm/xjckXxdi3sjoMmcIk7+03kEs+mii7eYFPRqnZj2C6uGNTQyvZa1ZUF27kYSewtvX3ilsqZQG48v2O1tbALmdfxcRZ270jKWqpTEQgXy7oqdTkDK3QmX8wTcC1yt1Ma2KlebcuL3c7GucT7i258N4aNgcwYqWyeJrFriPrzxSxxT6/vfPNCBnNlrnpsV7RoyF94fOgiexG2ec+oDQls+xM8fhO0LrFrdC7SUiZIV0uE7k57UZAdspnuu8d89M/0dLVZQqFxilrYijIlNU06E5FXA+uBz5ZtXwL8EPhnY8w41ZjOdJ8Zx5v05caxR3bZL/1I2wQucUfEJ3SJZ2wsE0q+sMdZ2AXBrl3iWUnSWSUzxi0nK1jYA/aCYmQn7Hpo/P5uFnfHKutRcC3d1Oi4BLNXfeOvfPy3jzAl/rDnPCMQaR2/T9BtYeoItTeG7aVtefGiyrWwA2HbTzxRHsN2LjDScdLZPFHsuZt82WIdtj9QWM/W/CLees2945YWT2dpDhejS2sX2ouwx3ePwNn/D573seIP2As1bMJZVyxE6/IjABjc/HcA9mYCdMXGXxjMRbSXuKJUTzWCvR1Y4XlccUqPiJwLfBC4wNuyUERagd8CHzTGVJwdOZ3pPjNO2mNBu3Hs0T2OO7NpAgvbdYlPlHSWKgo2FL6wrZXndYnX3sJOZjxJZ5VGbLoWtusKH9trY8tQOu3KxXWJdzpJVq4ApkdLLOz+0RRb+uLc9/Tg1IsMeAR7orKmQLlgj5Vud3FF2he0gzlc3OYphRaqseJ6M3FrYTsu8aZC0lnGnsdJDHyKJewZGW/5JlKlFnZ7NMSCljBP7BmFBYfCae+0P4uPtjs4FzZJQnQ3h2lafgxZ40Oesf8e/elAoQHLXEfrsBWleqoR7LuBtSKyWkRCwCXADd4dROR44BtYsd7j2R4Cfgn8wBjz89otewbxiqVb2jW6y8ZjKw2ggKJLfLJOZ17BrmRhh1qKZUD1tLATA7DgcFh6PDzxh/H7xz0WNtg4tjHjBHvjDhsrfrJ3dOqGLd4+4ZnE+OQ1KCZiuZ+/K/DBCQS7dWnR3wrWzR7v88ztbi2WT6XjpDK5gks84ssRT2asB8EfLFjYT7Gk5ALIJZ7O0hwqzd88dFGMJ3ZPkGnvhEFSxlrYXW2tPGGWE9ttXfl9KX+hRepcpzBeUwVbUaZkSsE2xmSBtwM3A48C1xljNorIR0XkAme3zwIx4Gcicr+IuIJ+MXA6cIWz/X4RWVfzd1FLMhUs7JHdtglGMDpF0tkEnc6y6dJ2loX+2rHi/ZZFRaGqVww7l7ECHO20oj28Y/z+iT4bGvA2FMmlrbh5ks4e3mE/u0zOsKVvivcTCDvnyE8s2OMs7AoxbCgKdtuK0u3NTj/xQtJZrLjedIJUNk8UR7AlSyrj6U7nXFRtyS0iWcFLkUjniIb9JdsOW9TKxh3DXHXDRpt85iXsdYmH6Y6FeDi/inDKlsv1Jv3zxiXu1mGrR1xRpmbKsi4AY8yNwI1l2z7suX/uBMf9CPjR/izwgFPiEh+03ySju631mxqp7BJPT5F0VsnCDjbbTOlgMyDO6E1HqGrY7WzSLPFsulhTPTZQtLLd+uFchcSneJ8tk3IvNFKjpaM1HTbuGCbk95HO5dm0c4TDF1eIS7u4ZV3uZ1uNYGemEuyyFqLN3bZHfGrUZmkHIkXXeyZOijxNjmCHJUc+m7H/HR6X+OPZhaRyFSzs1HgL+y1nrmEgkeaavz/Ffc8Mcv3bTis+6XxOKYJ0NYfoiIZ4lNXAnQAkTHjeuMTdsi5NOlOUqdFOZ+V4rdvkkLU88xnHwi6rBXZxhWZSl7jHYvJa1j6ffRxbVHSPz5BLfFwdtpvx7Q/b2LWbId7UURpXBju28qm77DHR7uJaUyOeRC6PYG8f4vRDFxD0C4/uGp58kYGQvXhw33elGLbr+s6UW9gVks68ty7Rbrv2R39tP2+RCV3iITIEcT4rj0v80fQCkmWfYTaXt9Z5mWAvaAnzhVet49KTV7Ktr+z36Ylhd8VC+HzC0+G1hafHCM8fl7jGsBWlaqqysOcVXgt6bBBGHLdwy2IraJmEtbqdLxp7jCfprPw5sILtrVE+7IXFGmSA4y6Bg57tcYlPEPvcBybtdOa6wxccaudHF/p4dzoWtucC5LZP2KS7QBgWHVVMlkuNeCxsu/7hZIZt/QleuX4F2wfH2LRzivcTiNhpWJnxlnpxHyeGnS2LYY+zsFfCYS+CteeVbj/4DCvWJldsCON1ifuKSWchyRHAyfj3BWHteeTyhuG/BQFDLm8KlmHCiWk3l7nEXbqawwyNZUhn84QCzvWxI9hZCRWEfm/LoeQHBR+GBPPHwvZplriiVI0KdjkFK6/ZusTdvtptK5yRm8aKhTfZyWsR53PFiVUu2VRp9651l5Y+/6LP2VvXFV5DC9sbcx2XdOYmnC04wpZwuT3FK7nEs6miy3z16R5vwEjRK+GI+KNOwtmRS1t5cs8odz3pGaRSCX/IXtS44YhKvbQDFSxsX9CGFUrOFYBLfzz++IPPhHdsKN3mWvLpUdKBPM1UsrADcMRLGF75fPjbLYC9CHKFNpGy+zWFJhDsmBXegUSaRa3Oe3A+O+PxujS3tLNzZCnLctutS3yexLALLnFtnKIoU6Iu8XLScSsO0S7rEnfnK7ct98Q8y+LY3seV4ti5TLHT2WQEiy7aWpHK5lglOzlKto0v63It7IW2Dph+p6dNU6e1pPNZmwgGVrDFEaWSGLZXsK2F/bAj2EcvbeOwxS3sGk4yEJ+ke5vbJ7zMUi+hUIftWtjJ8db1dCmUdSWcxin2YiBIlqDXwgZGPQ1gvBdBg2P2fbU3VbaIux3B7vM2QnEte8/6F8TCPGJsqVxyXrnE7W1eLWxFmRIV7HLcjl2RNusSH+op1vSWlxa5eB9Xak9aHsOeCDeeXeOksw8Gfsxng9+YwMIWmxEOPP6o037TdYmDtbLzOetKPuE19nNYdFQxWa4k6cwR7O1DLGwJs6AlzOFLbPx3065J3OJua9KMx7sxbh9XsB3hyybHl3RNl0IIIk4qmyu4xINkCYgnhk25YBfDDHudC5GO5soXZN2Opdw/6rlgcSxsCRXX390S5pbMsQyFlxInQke0igu8OYCIIKKCrSjVoC7xctxpSU3tjoXdY4dI+HweC7tMsL0WcaXEs/Is8ckINde4rCvPEumnS4bHZ4mPOh3cYrbBSLZ3MyYQQEKxYjJXNlWsuelYBf/6WLG+ORRzks6K7T57BhLc+NBOXnTsEgCOWGzFadOuYZ69psJQDyi2Ji0If6Wkswp12PtrYfuD9veSjpeUdQXJEHBd4j77LzKRYA/E7e97IovYdW33xz0WtiPYfs+Qj+5YiKszzyG9+mLaHu8l4J8/19J+EU06U5QqUMEuJz1qBSPSBnu3AKZY01sQjXKXeBUWdqBGgp3YC3setfeXHDvxvGmHVDbHIhkkJomiYA9tt2VOI065WlMHAKtkF9lwO0ERj4WdAZwvU3+4tBlJuKUshh3jk9dvQgTee95hgM2WDgd87ByqkF3v4rYmLcSwKyWdVajDLs8Q3xecEZveLPGA8caw7ecwkUt8b8Jazp3RiQTbbvda2PlgDB8QCHsF276Xx3aPzht3uIvPJ6heK8rUqGCXk3Yad0TarYWdGoGDnBraCV3iHgGvZGFnU9OwsGOTx7B/+eZiy9ATLocLvjzp6dKZDJ0yhB9DNu2s++ozbPvR0d22YYsj2FFJEQ8uJwilLnG3/Xu5QIYdC9tJRvv7jgy/fWgn7z73UJa2289KROiOhUtjuOW4JWSu8E/WmtT97DM1iGGD83mPOq1J7Rr9JktIHIF2XeJJj2BnvRa2E8OeQLBbwgFCfh99HsGOE6EFCIaL79MV7Cf3jLJuRft+v61GwqcucUWpChXsctxe1k3tTvevdLGmd6Kks5Is8YmSzqYh2JPFsHdvhDXn2IEkblb3JATG+vA7FrI/PWwFL94LD//cWrZHvwIi7RgEwZAMttMMBXH+48ZniIQjnOrZViDcYte6+2HoWMVP7+ulsznEm844uGS37lioNIY7bpHlLvEqY9i1EOymDhgbJJ0pxrAlm6I9LNax4CSdxSeJYbeEA8WSrTJEhK5YqGT61p50mBagKTpesNO5/LyzsNUlrijVMX8CZdWSceY6R9qc+cs5j2CXWXmFY7wu8QrtScuHf0xGODaxSzwzBsM9sOIU6Dq4OPZzEiKpvsL9QGa4WJrlrqtlEfh8jPmtaz3hdzqSOev90s0b+fnfNzvbKgn2iJ0VveQ4tvTFOXxxC5FgaYlTVyxcGsMtx/1s3OloZYL9VH+cy77zD1sGla1hDBvshdnYXnLpMXziiEYuQ0fESV/2j49he7PtBxJpOqYQ2K5YiH5PlvyelD1nc7T4PrtbQiX7zyesS1wFW1GmQgW7nHTciWG3F7cVYtgeC3vDd+AbpzuP40Uxy2WssH7hGNh8qy2Lymerj7eGmu0aeu6BT68utaL3brW3XWvs9LARTx/wzx0GT48fhtaULAp2KD1c7GZ25EvtbctSAOKOUMd9TkzcWW82nSKbdruKlQlJKAbD2+0alxzHtv44q7rHW8ddzVNZ2JHi+whExtVW/+evH+Evm/vJ+sKeOuyx2sSwo50wNkC+JHEwRYd76oplXR6XeCIztWA3h0ss7GfG7MljrR2FbZ3RUKHfznxpmuLiEyGvFraiTIkKdjmFGHZbcVvBwvbEsHs2WMsyHbeP3RnO+YwVnqGnbXJYzjNEohrcpLMn/2jbhT76m+Jz/Y6l27XGWsbJQStgvZtsxvfujeNO15wuzhcPZIrxZta/Dl75fTj2YgBGfHb9I1JqYYfIFAV7nIXdagUbGOk4ksFEhoMrCXYsTP9oeuJuVq7wju0dF7++7bE9/HGTHQCXkVBFC/vRncM82buPmfVNnZDYi8l4Et5yadrct1pFDHuqEqyuWKgkhv30WJjXZ/6N5pP+qbAt4PcVEtfmnUvcJ9pLXFGqQAW7HG9Zl0vbMnvrLetyrdvR3VbkXYHPZYoinRnzCHa1FnaLvQjYeb997J1JvddpbNK5pjiKc3R3sRtbBVd6c2Zv4X44O1ycex3thKNeWnifw9jM7EEbwWbHqBWlqD9LLuO2AS0TEk/v8K1BO8t7Vdd4we6OhUjn8oykKmTQQ1GwE3tLMsSzuTwf+80jHNzdTCToI0WoYh32B37xIJ/47aOVzz0VTR32IiblWNiRdsimaQs75m6Fsi6vS3xvPD1hhriLm3TnXrDsGBrj0diz8Uc7xu0HzJsuZy4+0SxxRakGFWwv3rnOrgBH2oqlU66FnR0rxo9HdlmRD7sWdtYj2AmPYE+zrGvng/bxU3+1fbbBdiJrXmCt+dhiu210T7EbW2p8c5KWbD9559cczo56Bnx0luw3hH2PA8YK5h832/GYzzqohVxmIgvb+VxalrI5bi9mKrrEPaVNO4fGeO13/lFoOOI97zM7thM3oYJ79P5nBtnSG+dfzl3L0rYmkgQr1mH3jqQYSU4weGUqop1gckRSjieiqR1yaVpdo9nTOKXdsaRLXeJTx7C7YyFS2TzxtD1u11CSxW3j4+9uHHv+ucRRl7iiVIEKtpesU8LkjWF75yoHPC5x18IeeAowlS3sbHL6LvFwzK5h6Gk7wCKfga132Of2boGuQ+z9FtfC3mWbu0BFwW7L9tMbsOIeyY6UjtD0MGCs0Pbl7e3WASuAy2I+TMHCLhVs41rDS45lW18cn8DKzvElWV3N9ri+0RR/2dzPnY/38o+tnv7iznkDqUGeGDT856+ta/+uJ/sRgdPXLmBxW4R4PugZr1mMYQ8kMoxlKiT7VcGoEwqIpZwLsEg7mBytQXu+ZN7G00dT2YIFnHTq2ZOZHIl0bkoXtvv+3Tj2rqEkS9qbxu3nnr9zniWd+X2aJa4o1aCC7cU719l1iXvHNPp8xclS7mSrAScRLOKxsLMeCztbWewmxNs05KQ3QrgNHnfc4v2brTscihb2iFewx7vE2/N7GQguJi0hmnIj1iUeiIzrJtbvCHVv1t72O7rY5M8TMI71WuYl2Dxk3cZ7YoextT/B8o5oxfKmooWd4ul++xk/2etJ8nI+my4Zpam5lV/cu51kJsddT/Zx5JJWOppDLGlrYjQ33sJOZnKMZXKMpfdNsD9zhxXq1vQe5w23A9AesO95yPlVxlPZgjC7FvaA0zSlYwqXeFehn7iN4+8YGmNJawUL23WJN6tLXFGU8ahgeyn0so4WLebyucrBJhh8qvh47xZ7G/YknZXEsCuL3YR4S5qWnQhrzoInbrFNXEZ323IusJ3KxOfEsF3BHj93uiO/l9FgF2P+FqL5EUgMjHOHA/Tl7IXCroy1/PocXWz25wjhvIeyi46+jH1Pd4wsY2vfaEV3OBSFqG80zVN7bXJXSZKYc94Qabo6OxhNZbnlkd3c+9Qgpx3SDcCStgijOT+mrA57MGHXliwfbFIFxhgeG7Gej7aMx8IGWnz2dYZSVklGklnamoIE/VJ4Ldet3zlBH/Hy998/mmJoLEMyk6/oEj92eRvLO5rmTR9xF59PG6coSjWoYHvxNu4INsHRF8FhLyjdJxgtlldBUbALLvFscSxlJlG8P53GKQCty6G5C45+uXV73/pRu921sH1+O4jDa2GXJ50ZQ5cZIB7qIulvIZqPW5d4mTsc4K7c4dyZO4bH0raveG/CfoE2+bKE3MlVZe9hS/hI/pw7im8/vYRtfQlWd1XoUEYx67l/NM22flewPRa2Jzbe2d5OezTIp3+3iXQuX+g/vrgtwpgJkU0luGPTLnthFIgUrNx9cYkPjWXoy9mLjPYyCzvmCPag42mIp7PEwgEiAX/Rwnb6iFdrYffH0+xwTri0gkv8wnXL+PP7z55XfcRBG6coSrXMr2+GqSj0snYsxYu+DYecW7pPsKlYG+0LFsXbjXnnM0WruiRLfJoW9pLj7O0RF8CKZ8Hd37KP3Rg22KEd/U9CyiaIuTHsbC7PM3sTkBwkRJZkeAGpQAvNZtS6xKOlFrYxhgfTS3ht5t/pS/lIZXPsTVl3d5MvR0hcC7vUKnxaFvPqzAfZNORnNJVl9QQWdtDvoz0apD9edIlv2TNaLPPyWO6+cIzzjlxEz8AYAZ9w0iq71qXtEVKEyKUTXL/BZsunJVhoDVqtS/yrt2/m0qttvXrvSIpBJ8muNe1Y2I5gu4NABhwLezRpBTsc9JPKlrrEp4phFy9YUuwatq6LShb2fMW6xFWwFWUqVLC9TNbL2iXYVHSdLzyiWCblxrBzmWLc2usSr3b4h5t57Qq2CLzg04BTZtTpafvZsrhY/iU+SI0yNJbhtd/5B2d+7naG9ljLOxmxgh0zcZsl7i1Zw7qTjYGQ38dIKkv/aJqM07U2Ih4Lu8wlPhjP0BIJEPTbtU3kEgeb+by1L85AIsPStggjqSy9oxXi+8FmXnSsbeZy3Ip2YmG7jsWtTSQJYTJJHttuM7pHsgEGHJf4WCY3cZ23h39s3cvd2/aSzxt6R1MMOuVsi3CS4JwLryasJTyQtO7veCpHczhAJOgruMQLMewpBDsc8NMSCdA3mi4MQVmigl1AO50pSnWoYHvJlFnYlfCK+eJji/cLMexcaVlXdpou8faD7LnWnFXctnQdnPIm+3reZLHYosKaTcdqkvFBXvG1u7jryX5yecPe3U8DkIosIB1spZVRxyVeamHH01aQF7aGMQae3psgbUeAEPFlizHssvcwOJZmaVtTIc48kYUNtrb4/mcGATjzcOt2f3KPvfAx3vOGopy6posVnU284OjFhc1L2iKkTJB8Zoy+AetJGMr4C6IJ1cWxewbGyOYNfaMpekdS5PAzbJpoESdo7wh2OG+FdW/STjxL5/K0RAJEgkWXuBvDbm+aOubcHQvTO5pi52ASv09Y2KKC7aIucUWpDhVsL5MNn3Bxa7GjXaUJaW4Me8Kksyozf1sWwb8/AytOLt1+/qfgyjvK9i0K2l0jC0nFhxgey/D+8w8HYLjXWtiZ6EIywVZaiWMquMQTKStAi53M5af646QdCzs0mYWdyNAWDfLG5x7M2YcvZHnHxJ6J7liIEadb2NmHOYLtJJ6l8Qp2M0G/jzv/7Sze8NyiN6E9GiTjC0EmSUTs5zuU8THoEeyp4tjGGLYPWGHeOZSkd8ReTLnd3TISKjRj8Wft30J/0hS6nDWH/I6F7caw07RGAlXFnI9e1sbtm/bwQM8gC1vC+H0y5THzBRE0S1xRqkAF20tVgu2IUmxxsRYaSl3iJY1TXAt7/zJ/xzJ5+hJlzUGcbmd58fPgWDctkuTP7zuLC9dZl/LYXtsBLRtdSDbUSrvEkXx2XNKZa2G7cdWtfQmy+DEIYbKEJY1BCl2/XIbGMnREg5x2SDffueKkSUXIW6p0ysGdNAX9bHESz8bynt7hQfvZi5SeS0QIhpsJmTRhx+Lfm/YVXOL2M5pcsL312juHkvSOpgj5feSci62ML1L0Ijj5DP3xHHHngiYWCRIJ+Ek5ddh7E5mq24i+97xDyeQMf3qiT+PXZfh92ktcUapBBdtL2lPWNRFu4lXLomItNJRa2CUxbEe893NQxZdufYKXfuUvpRsdwd7r68JE2vGRJ2RSLGixr5Uf2kHChPGFW8iFPL3Ry1ziCVewPRY2CATChMgQIkvOF4IyER1MZGhvqk6w3EzpruYQLZEgBy9oLljYCeO5mJnkYikUiRKUHEd02j/bvSlfIekMpk486xkojkXdOTRG74j9rMT5PDK+Jo9gx8niZ2Asy0jKXhTEwv4Sl/hAfOouZy4HdTXz+ueuBmBp2/gM8fmMX2PYilIVKtheMp7GKRMxoYXtLevyZIlnp5klPgE7BsfoGRgrESjXJb4108HBy5yLh9QIQb+P7liI6OjTPGUWEQ4FyIU9gl3mEnctyKKFHbeJZP4QQbKEyZCT8R6CwbF0oV3nVLj9sQ9ySr/WLIgVBHss57GwQxNfLEWa7HPH2ZA5fUkpi2FPLtiuOxxst7HekRTdLWFCrfaEWb/Xwh4lJwEGE+mihR0OliSdVdNH3MvbzzqEVV1Rjl3eNvXO8wgRIad6rShTooLtJZ2wbt/JMrrdGHa5he1md4/rJT7NpLMJcIdPbOnz1Fo7FvZ208URq5Y678E+v7AlQmfqGbaaxYQDPoybFAfjXOKuhb2oYGEn6IiGEH8Ifz5DWLJkfaXrT2ZyJDN52qoU7G7HEj3IGQ6yZkGM7YNjJDM5RrNewZ74Yqkpap87rN0KaO+YMJDIFFzxU7nEtw9awe6IBgsx7AWxMLF2G1PP+puKnpBMgrzYLPRRx8JuDvsJB/2FaV3V9BH30hwO8Md/PZM3nbGm6mPmA37tJa4oVaGC7SUdnzx+DUXBji2ygzjAJpS5SWXepDOTL7YLrZFgu5nV7hry+BgKLWHlEsfad7qdLWkJsCS/m21mMeGAH+Od712eJe5YkG6p0VgmZy3iQBjJpYn6c2QpFeahMSti1brEux03vdtrfHGbzUjvG02RyELGOKI9STiipcVedKxpsq7tPQkYTKQLrvxKLnFjTMEV3jMwRiwcYO2iFnYOjdE3al3i0TZrYbe2thVzDdJxjAQYSKQZdT6flkiAcMBHKpPHGMPeKkZrluPTZLNxqEtcUapDBdtLJl5IepqQgkt8kbXEo13Wjet+0ec8FjbYlqJQfR32BLiZyk96LOy8P8wbsu9nz5H/jLgWvnOBcGh4kKDkrIUd9JXO9y7PEi+zsMGZGOUPQS5F1JcjI6UJZ25L0Gpd4svam/AJHL7YrrM1Yo8bSWYZy2QLWemTXTAdc/KZdp19dwGwM2ETybwXGuXc8Xgvp3/mNjbtGqZnYIzlHU0sbYvQMzBGfzxtY9hR202tKRorXnilExhfgJFkliHH7W7rsP2FoR+pbH5OjsIUkcNE5H7Pz7CIvEtEOkXkFhF5wrkd3zJv315Py7oUpQpUsL1Mx8J2S6pii63IuxnU3qQzgOSgvZ0BCzuezvLH7DG0LVhadMk73c7WBuw0sa156xIXb7MUr7UNJBzLtLM5VGiC0ukKdjZFxJct1GW7uOVU1dQgg23F+Yf3nMH5Tm11q3Pc8FjGip97/kk+/+jyY6F1GbLFlrf1JoWhsUyhzWelGPZT/QnyBm7ZuJvtg2Msa29iSXsTO4eSGAMLYqFiiMB74ZUeRQL2/jV/t/Xs3takxT7ic2+yljHmMWPMOmPMOuBEIAH8EvgAcKsxZi1wq/N4v/FrpzNFqQoVbC/pxKRJT0CphQ02lh1sshnUvoBT1uUpv3It7GrrsCegEMP2DM1wXdnN4QCESgV7hdkJwDazhHDAj88RpVwwNs7aj6dziEBT0E+LY/l2xUJ2v1yGJskWBdVh0HGJVxvDBjh4QaxQrtUSsRc4I8ksiVSueEEw2QWTCKx9nr0oAlJOdvmSdmthJyq4xN2Rlrc9toeegQTLO5pKuowtaAkXPQ7BaDGGbXI0RZo467AFbNplP9PmkNPpLJun3xHseTC7+hzgSWPMU8CFwPed7d8HXlqLF7Au8VqcSVHmNirYXjKJCWOoW3pHWf/xW9i+6Ew4/d+gw5bo8Oy328dge4vnM8VEM4CxQUDssI79wHWJP7U3QdqpA3ZFPBYOFC3stBWXRdntjJgm+mglEvThb2ohZ4RMaHyGciKVJRr04/MJrY6QWpd4GHIpIpItiKPLUMElvm+C5brEh5MZEmnP+ScrqQNY+/zCXfciYpljYVeKYfc5wnrfM4OMJLMs62gqxLzBEWzXwg5GSzwh4g/yhVetY3lHEy3hAD6fEAn6yeUNu4dtJ7S5aGGXcQnwE+f+ImOcK0HYBSyqdICIXCkiG0RkQ29v75QvIIK6xBWlCgJT7zKPyCYnzFJ+fPcofaNpHhxpYdnZHyo+ccg5xfu+QGlrUmBgby/tgfC4RiDTwW2NuWZBM0/2xnl6b4JDFsaIp9wOXAEIO+t2LOyOpM0QByEc8JMJGoZpxh9qp7xtRzydI+r07HYt7M5mm3RGNk1Ysozly2LYY9NziZdT4hLPeC3sSUrqAFaf7sTW0ySdDmlLnLrmSjHs/tEUoYCvcJGzrD1aMilrQSwCuC7x5tLQhS9IezTEj9/wrELuQCRor3F3OBnn3XMwhu0iIiHgAuDfy58zxhgRqaiyxpirgasB1q9fP6USa9KZolSHWtheMslxE6lcRpLWotzhDG+oiN9xiWeLgr1rzx5ysn/XRa7r+7gV7UCxpWdBsMMBax06A0AAmka2sc3YeHE44CPk9zFsoqSDFSzsdJbmkPUAuK5qG8MOQi5FSLIkzfiks6BfiIb2zXPgdYmPOTFsM1VJHdgLk1XPASiI/OLWCCKVY9j9o2lOWNlOt9O4ZXlHU0mnse6WkMcl3lQq2H67xpVdUc5y2qlGgvb9uoI9xy3sFwD3GmOcUWbsFpElAM7tnlq8iMawFaU6VLC9ZJOFXtLlDDsu6V1DYxWfBzwu8aJgt0rcdgnbD1x3+HHL24GiYJe4xEVsHDs1Atk0/uFnioId9BEO+vh1/tlsX3zOuPPHUzmaQq6FbW+7Y45LPJsiRKa0fSg2ht3WFNpnz0HQ76Mp6Gc4mSGeypGVIDJVhr7LyW8if+TLMGL/fNujQZqC/ooucZsJHuGMQ63gLutooqs5RMjvIxYOEA0FINwGR14Iq55b2pHON957EAk4gj2UJBzw7fMFS4NwKUV3OMANwOXO/cuB62vxIjZLvBZnUpS5jbrEvWSTEKjcNnLYSbLaOYGFPZTI0OIL4HPLuhzxbiVBTlr2a1lua8zFbREWtYYLmeJuD/DmsCMa4RbbOGVgG2Ly9IVXQNaOd0z583wu+yoWrjyW48rO77WwWwsu8WLSWYj0eJd4ovouZxPR2hRgeCyLzwc5CU2doe9y2Pn4DjufrsdvoW80TWdzyAp2BQu7bzRFV3OIy05ZyYpOK9YiwuK2CIWSaJ8PLv5B8SDx2Rr6Cv3fwx6XeHds/0IdsxkRaQaeB7zJs/lTwHUi8nrgKeDiWryW36eNUxSlGuaXhW0MPHZTicu6hMzYhBa2O2lqIsF+4w82sH04y4ate0inkoW65xYZIyu1sbBj4QCrupqdXt8UGnq4M6MJt9jGKXufBCDefJDdHPARDthftRvL9VIpht3VHC4knQVNlkTeX/KlavuI759gt0SCjKRsWVfWF5o6Q7+MBS0RQn5r5UYqCHYqm2MkmaU7FmLtohbede6hBYFds6C50HVtHG5Gv2/89azrEt8+MDan3eHGmLgxpssYM+TZ1m+MOccYs9YYc64xZm8tXktj2IpSHfNLsLffCz+5BJ74feXnJ7OwnRj2rgkE++m9CYzPz869I2zvGwRP3XN2Px0ZXtd3VyxU6J9dEsMGG99NjULvYwBk2u14ynDAR8gR7FQFwU6kihb24UtaOKgrSmtToJB0FjBp0iZIwiOIg4nM/lvYEWthJ9I5tgdXwMIjp3X8wpYw7dEgIkJTyD8uhu3WSldqbvL5i9fxhVetq3xiN45doXbevfDpHU3NacE+kNhe4irYijIV88slvuNee+u07xxHNjnhVC3XJb5rOEkub0pGSRpj2JtI09QSoSMrZDMpaCsmd2UqDM6YDgXBjgRoawoW4unxVBYRinHUcAskh2HnA9C+kljHAkKBHkRspjhUtrAT6ZyN5QIXr1/BxetX2CecpLOAyZAmQCKVLVjzQ2MZjlzaOu5c06G1KcjeeBqD4Udtb+blrzptWse/4OjFHLrIZpVXimH3j05cKz1pD/BACFJUdIm7FrYx86IG+4DgFx2vqSjVML8Ee+cD9jZTIXEsl7WDO4KVLWzXJZ7LG/pGUyVtPBPpHOlsHvEHaRbIp9IQKZaoZvbzY3ZfuyUcoDUSLPTxjqdyNIcCxThqKAbDO2DXg7D4WC49aSWHLLCCFprUJZ4txsG9+MOQS+PPp0kTZDSVZaHz1GAiXROX+FP9Cfw+saVp0+SSk1cW7jeFxrvE+5ymKdNuH+pa1pO4xO15VbBrgTZOUZTqmF8u8ckEO+tsm6CsaziZKbTtLI9ju65Xnz9ILGAgm8YEIoXGHumykqin+xNs2FZ9+C/usbBbm4Kks3mSmRzxVJnQhlthZCf0b4Yl6zhmeRuve45t8OL3CX6fkM6NT8zyWtglBEKQTePLOxa2Y8Gms3ni6VyNXOIZxtI5mvYz27qShd3nWNjd0xXWgku8koVd/JfpbJ67NdgHEm2coijVMX8EO5uGPY869ysIdsYR4Qks7OFkhoO7rbW6c7D0eDem7AuEiAby+E2GlAkwZuwXf6pMsL946+NccvXfqhbt0VQWn9M61NtwZDSdLcavwcaw3VaoS44dd56Q306a8pLJ5Uln84UYdgn+MGST+EyWlAkVLhyGCm1J98/CbG0KMpLMWgu/FoKdKXeJ76+FPXFZF6hLvFZoHbaiVMf8EezeRws9qAvi7CXrbJsghj2SzHKoM2lqIgvbHwjQ5MsTIsPeJIxhz5U2/nH7Z/OGt/34Xn55Xw+v/97d3PTQTiZiJGljxyJCmyPYQ2MZ4p6YMlBsTwqwpLx4y5YkpXN2NKSLazVHwxUsbH8IsPumCRTKyIacLmdt++0SD5DO5RmIZwp14PtKJZd4fzxNJOib/sVAYDILW13itcangq0oVTF/BNt1h0PRJf7YTfA/J1rruyDY4y1sYwzDYxkO6owSDvjYWdY8xR01GQiECPsNQcnROwZJpz92eZewobEMB3VFGRrL8O6fPsCtm/bwu427Cs/3DCRKRHXUI8yuSNqGI9nSxh1uW8/mhcVpYh5Cfh87Bsc483O388O/bgOKozUripqn61iaYKHjWt8kyVzTwa35Hi1/H/tAJOhnLF3qPbA12PtQKz1pDNvrElfBrgU+nzZOUZRqmEeC/aDtBNbUWXSJ737YxnuTg0URr1CHHU/nyBvb6GNJW2RCCzsQChF2LOzd8TxJx8JOlnUJGx7LcPTSNn74+lP41mvXc8yytsI5nuqP85xP38YP//ZUYf/RZJaY04HMHc4xNJZhNJWrbGFXsK7BJp794dE9PNWfYMNTA/a9pSazsIvehjSBgkvcTeba3z7arR4LfX8Fuyk4vqyrfzQ9/fg1FN/3VBa2xrBrgt9HyQWqoiiVmUeC/YCN64aaiy7xdMK5HZ3UwnZLulojQRa3RcbVYg8k0vjEWth+kyVElp3xPGPOcIryLmFDY1lam4KctKqTc49cxIKWcEGwn95r1/S5mx8rbKtoYY9lnaSz6Qk2WFf09gF7gVKthZ0iSNxxn/eNuIK9fxam2wYVqJz0Ng2aQj7GMrmSL/7+eGr68WsoCvUkddgAneoSrwk+rcNWlKqYH4Kdz1lrevGxNgs84wi1e5tOTBrDdsuqWpuCLG1rqmhhd0RDiD+I5LN2ulXOX7Cwx/KlH/NwMmMbkzh0REMMOOLs1g4PJ7P89+9tA5TRVJaY4z5uLYthVxbs8QlnAOccvpB3nH0I5x25mB5HsHsd8a04JtMjWGlj67DBusT9PqFjf5POIrW1sHN5QzKT58ofbOCuzX30j6b3zW3v/g1UcInbmnbbOW5/E+UUi09Es8QVpQrmRx12etSKc9symwXuinNBsOOTZom7Xc5aIgEWt0XYXdY8ZSCRto04fHZaV4AsKQJk/REwkMj5McYgIiQztmbbm7DVFQvRH09jjCm4m19+wjJ+8o+necuZaxhNZQsznwtzpMcyJZY3AAedBs96G6wZP+AD4IMvsp3EPn/L4+weSZLO5tnaZ9ucru6u0KbT4xI3/hCj6aJLvLM5hM+3f3202zwXLftb1uW6qh/bPcLvH9nNo7uGrWDvk4U9cdIZWCvbTQJU9h+/T1ADW1GmZn5Y2K4YByJWkN14tesSz8QnrcP2usSXtEXI5g398VTheWthB+0XfDaJD0PGBCBge2OnTKDQErRQEuUR7I5oiFQ2z1gmR388TdAvXHHqKvIGHuoZsjFsR5hDATvlqj+eJpXNlzYcibTC+f9VnI09Acs7mjDGtlnd1h+nrSlo11+OxyXuD0YKnoa+0XRN5kC31NDCdl3qj+2yXeye2TtGOpffxxj2xGVdYC8O1B1eO3xah60oVTE/BNsV42CT4xJ3Hntd4pNY2F6XuCsy7kAOgIF4xrqHfUFrrWOTtCTU5NwPFsqnvOLv4rpt+0fT9DuZzWsXtiACT+wZZbTM9d3WFCxkqlfsUDYFyx1rvWcwwda+OKu6mytbix4LO9YcLdSf942m9jt+DaWfwb50OvPSFLJ/ypt2jQBw0qoOYB9LrwoWduU1RYJ+TTirIT6fxrAVpRrmiWA71nAgAsFoUcC9LvGsxwovw+sSd4XTza4G6xLvbA7ZL3jnnBkC+MPWwk7jH990xGthO4K9N5523LghmkJ+lnc08diuESeGXS7Ydr2xStndU7C8w66rZ2CMbX0JVndNMCXLY2G3tcQKcW8r2PsvWJGgr9A9rhadzgAe2zVCU9DP5y9ex0mrOjh+Rcf0TxaY3MJ+wTGLOe+oRRWfU6aPX0SzxBWlCuZHDDvjsbCDkfFZ4pk45JymKpO4xK1gW2FwB3IYY4ox7GywIPxpggQd13TaBAv7u+LvLWly63n3JtL0xYtx10MXtnD/M4P2tT3C3NoUKMSem/dBsBe3RRCBLb1xdgyNsap7eeUdPUlnHS0tbO8ZK8TZa2FhiwgtETsApBZ12GAt7JWdUVZ0RvnZm0/dt5NNEcP+9xccsW/nVSqiSWeKUh3zxML2ZIAHPDHsQiw7Pmkd9kgySyToIxzwFyxa12IeTWXJ5Ayd0VBJVnFeAkRjVrAzBArlU5UsbNclvtdxiXc7jw9ZFGO744Yut7Dd5iX7YmGHAj4WtUS468k+jJkg4QxKXOJd7S0k0jl6BsZIZvI1sbChWFdeiyxxsF6KgybyGFRLYR72/nVyU6rD5wz/UCtbUSZnfgi2K8aBJidL3BVsa6WWlnVVzhJ3Y9cFl7gjwANxK8AdzaESi+w9LziWw1fY2VZpAoym3Bi2Ew/3CLDrEh9IFF3iYC1sF68wl8R+90GwwSaePbTd9h2fULA9LvHudjtK84GeQfu4VoLtXLjsfx12UfD3X7CDpbfKjOJ38idUrxVlcuaHYLsx7KCbJT5B4xR/CHzjP5LhsWxBYF3hdF3ce53BH53NwRILe0F7DH/ICmHG0yXMtbC9LvHWSICAT+gZGGMskyu6xBd5BDvidYl7BXvfLNNlTqY4wKoqLOwF7XYt9z89CEB3S20Eu6XGFjbAyq4J3k+1TOESV2qLWx2oiWeKMjnzRLA9FnYgYh8bU5otnklOOlqzaAlaYXAF2G140h4ttbDxhwsZ5ymCJYIdDfkJ+osfvYjQ0RziiT02w9l1ka9ZWBSeEgvbI9j74hIHa2G7r+W12EvwxLCXdLYDXgu7NmVN7mvXqg4b4KDO/bSwp0g6U2qLW8+vcWxFmZyqBFtEzheRx0Rks4h8oMLz7xGRR0TkQRG5VUQO8jx3uYg84fxcXsvFV03GE8MONoHJQy7tcYk7ddgTCna26BIPuRa2dXG77UM73bIuF3/IZqQDGVO0sIfHMhWnXHU1h3h89yhQdDdHQwFWdFphjZWVdbnsq0t8Wbtd24TucChxibfEosTCgYIbfUHNYthB/D4h5N+/a8fausQnL+tSaovbgEgNbEWZnCm/JUXED3wFeAFwJHCpiBxZttt9wHpjzLHAz4HPOMd2Ah8BTgFOBj4iIvtQZ7OfeOuw3Trr1AjknVpqt9NZhYQzgJGxTMEl7vMJ0VCxTMudhW07nXmsRH+w8Fp2NKUV+KGxTEWLtiMaKrQJ9dYOr3Xi2BMJ9r5a2MscC3tCdziUuMQlEGZZexPJTB6R2k2qOmxxC2sXxva7a5jr+fD7hKXtlWeaV40mnR1Q1CWuKNVRjVlzMrDZGLPFGJMGrgUu9O5gjLnNGOMEhPkb4NYJPR+4xRiz1xgzANwCnF+bpU8Dbx22a0Un+ovPu3XYFRLOwFrYrWVWrVew/T6xgu51iQeKLvG8FF3iw8nKFra3c5Y3oWvtIptpXpp0Zu/7fVIyjGI6uC7xai1s/OHCMR3REIH9tIhdXvec1fzuXafv93kiASvYy9qbSsIN+4QmnR1QfKIucUWphmrMs2XAM57HPViLeSJeD9w0ybHLprPAmpCpYGF7BTuTsF/OFQZ/gJslXvyoYuFAMenM6XImImUu8SCEbWZ1Phj1xLCLfcG9eIdUeK3Xi9evIBLw0+5pHeoKfnPIv8+W6equZt5y5houOG7pxDt5p1X5gwWrvFbx61ricy5e9tsdDp7hHyrYB4KiS1wFW1Emo6ZBOhF5NbAeOGOax10JXAmwcuXKWi7J4pZseRLBSi3sUSfmPF5I3WEdpaVU/pKks0If7vKksyVHwsu/xabfttOSKrYmPWJJMfvbxZ181RIOlCRQrVkQ493PO7RkX9fa31d3OFiBe//5h0++U2EudBhEChcas7UtZ3cszGGLxn+200azxA8oamErSnVU4zvcDqzwPF7ubCtBRM4FPghcYIxJTedYY8zVxpj1xpj1CxYsqHbt1ZMZs6Lj8xXd3q5g+8PFOuwKSWfePuIuzaFAoTXpXrfLGZSOY3RLxI59JeFwqNA4ZcKkM8dqrab3dcHC3g/Brgp/AMRXsDgLFnaNSrpqzc/e/OxxFzf7hAr2AaWQJa4WtqJMSjWCfTewVkRWi0gIuAS4wbuDiBwPfAMr1ns8T90MnCciHU6y2XnOtgNLNlVMKAuWxbBjC4sx7IqDP5y2pOHKLvHBhMfC9gq2J/7b7OyfyxtGUtkJk86AqsZBth4owQZ7QeMImNuDfDa6xAGWtjfV5jOZYlqXUlvcpDPVa0WZnCkF2xiTBd6OFdpHgeuMMRtF5KMicoGz22eBGPAzEblfRG5wjt0LfAwr+ncDH3W2HViyY0XL2im1IuEso7mbdHKUnt6BijFsV5i97ufmcKDY6SyRKYhtqUu8KGoxJ0nNFf+Jyrq8t5PRHPLj98l+ucSrJhAqeB5WOBb2otbK2fRzhlWnwbPfDkuOrfdK5gV+dYkrSlVU9Y1vjLkRuLFs24c998+d5NjvAN/Z1wXWBG/JVnmWePMCAtkHwDRVzBJ3Bbu5XLBTWYwxDCbStmkKlLnEi+IfDfnpG01V7HLm4rrVq7GwRYS2puA+dzmbFv5QwVvQFQtz9WtO5KRVnTP/uvUk0gbP/0S9VzFv0MYpilId86MzhLcpSnnSWfMCfORpJY4JRCjPuXZj1bESl7if0VSWeDpnB380V0o6K62VHk1lC33EJ7Owq3U3H764paR16YzhD5dcfJx31OKZf01lXuHTXuKKUhXzRLBTkwh2NwCtkiDjD1MupfGChV20ZpvDAZKZPH1Oo5OihV1Wh+3ZP57KFi3syPiPvSsW5pzDF3LaId1VvaUfv/FZVe233wRCpfXYilJj3LJ5TTpTlMmZH4KdGSsKdVmWeCbSVRDpjIwX7BE3hl1Whw0URl92VHKJe8Q7GvYTT+eKozWj4y1sv0/49hUnTfedzTxlFrai1BrXws6rYCvKpMzd4R+9j0PfZnvfW7JVyBK3SWeD0l44JM14SzI+QdIZQM+Abe5WrMN29vEFS6Z+xUIB0tk8e+PWIp9w2MZsZJKGMoqy32y/l2U7bgEgrzFsRZmUuSvYv30P3Phee98r2K6FnRoG8dOXLSaapaWyYPukdHyjK9jP7HUs7OYyl7i/9Dzu/ndvGwAqx7BnLd2HQtch9V6FMle570ccc/9VgLrEFWUq5q5LPDUCxiaMlWSJ+3zWxZtLQaiZ3Uk/RziHJB0L+/O3PM6aBc1cuG4Zo6kszaFASQvQmBPPLlrYZWVdgXLBtvvf8MAOjlvRvt+znw8oF3273itQ5jKBCP689Tzl83Vei6LMcuauYOfStiEKjB/sEYxYwQ5G2TVWFM+UI9g/2/AMRy9rs4KdzJbEr6E4YrNnYAwRj8U8gYV95mELefWzVvL8oxbznEO693sylaLMGYIRfDk78U5j2IoyOXNbsFPD9n5mrHR0ZjAKySEINrE9XowKjBln1nWymNEdT2fHdc8qxrDHCvOcgeJ4zbIkrUWtET7+0mNq9c4UZe4QiOAzWfzktA5bUaZg7saws2lIDtvizmyq1MJ249mhZp6JF63dpAmRzxtG01mGElawR1O5CQV790iymHAGOpZRUaaLk9AYJqMWtqJMwdwV7FzKxrAzCadxisfqDRbblG4dKW5O5IMkMjmMoWhhp7KFmLWLG5M2xpNwBkWXuGZVK0p1OBfSYdIq2IoyBXNXsLPOwLDEXshnSwZ7GMfCNsEmtg4VD4nng4w607kGx2xcbTSZHdez2/u4kHAGxbIutbAVpTqci9sIGXKadKYokzJ3BTtnLWTizvAwz+jM4ZwV1F1jPoYzQl6s0MbzAUZT9rhkJk8yk7NZ4mWC3RT0FyYMtXtd4oWkM7WwFaUqnP/LsKiFrShTMYcF27GwRx3B9ljYY8YK6307U4CQc9xyI7kAo07vcLCzq+Pp8Ra2iBQyxUssbLfTmV9beSqNjYi0i8jPRWSTiDwqIs8WkU4RuUVEnnBuO/b7hbwxbE06U5RJmZuCnc9bNzgUBdsTV06LvT+cc4TVGbk5mgsUXOIAg2MZ4hUsbCgmnnV6Y9gT1GErSgPyJeB3xpjDgeOwo3U/ANxqjFkL3Oo83j+Cbgw7o41TFGUK5qZgu9Y1eFziRQvbrbdua2sDQMIxAIazwYJLHGDPcIpMzlScO+0mnpW6xNXCVhofEWkDTge+DWCMSRtjBoELge87u30feOl+v1ghhp1GDWxFmZy5KdhZj2AXXOLFGHbSWEE946iD+J9LjycQbgZgJOtjxGNh73CGe1QSbHdbadJZ5cYpitJgrAZ6ge+KyH0i8i0RaQYWGWN2OvvsAhbt9ysVYtjqEleUqZibgp0rWsmM7ra3Hgt7zLGwo82tvOS4pRCyFvZQxs6tdulxBHsyl3jlpDMVbKWhCQAnAF8zxhwPxClzfxtjDFBRYUXkShHZICIbent7p3glR7DJaOMURZmCOSrYXgvb+cLwWNgJx8ImFC25HcqUxrC3D7gW9vje382VLGzXJa512Epj0wP0GGP+7jz+OVbAd4vIEgDndk+lg40xVxtj1htj1i9YsGDyV3IEO6J12IoyJXNTsEtc4q6FXRTseN6xhN3M8VAzWfyMZg2jqSwhvw8R2D5oh3vEwuPrqmOVks58PhCf1mErDY0xZhfwjIgc5mw6B3gEuAG43Nl2OXD9fr+YdjpTlKqZm73EnWECgCdLvJJgNxduM74wiVSOkVSW1qYAmZxhx2ASKCaYeamYdAbWLa512Erj8w7gGhEJAVuAf8Ze4F8nIq8HngIu3u9X8cSwtXGKokzO3BfstNN71FOHPeI0Tim4xBccSn9oBYl0rtDZzDB50tnahS0c3N1MOFAm5l2HQNeaWr0TRakLxpj7gfUVnjqnpi8UdGPY6hJXlKmYm4KdTY/f5rGwR3PO23ZF/Dnv5lt7zyN+b0+hs5nfJzzVb13ilZLOLj91FZefumr867z1rv1dvaLMHwoxbHWJK8pUzE3BdpPOAhE7C9u97zCSdQW7ubAtGvIz5rGwQ4FieL98HraiKDXCCR9pa1JFmZq5nXTW7MlQ9WSJ78i2jHu+ORwgmzfsTaRpiQRoayrGpt02pIqi1BifD+MPO2Vd9V6Mosxu5qZgu3XY0a7iNk8d9u2ZI/j6UT+G7kMK25qCNha9ZzhJLBwoJJM1Bf34fcWZ2Yqi1BbjD9uyLq3DVpRJmaOCXWZh+0O25ArI5w2JdJ6x9rUlh7hZ38PJLDGPhV0pfq0oSu0wgbCWdSlKFcxNwXaTzlzB9ljXiYydxlVeqtXkcXvHwkHam2x9dYvGrxVlZglEbFmXCraiTMrcFOyChd1tbz2dx+JO69Fyy7k5VBTwlhILe3wNtqIotcMEIjpeU1GqYI4KtmthO4LtSThze4WX11Y3eQQ7Fg7Q5sSwNeFMUWaYQNipw673QhRldjM3BXsyl3jKusSjoXILu/i4ORyg3bGwKzVNURSlhjgWtg7/UJTJmZuCXZ50VsHCLnd1RyewsLUGW1FmGCeGrUlnijI5c1Ows6Uu8bQUB3TEJ3CJRz2PWyKBQtKZZokrygwTiOi0LkWpgrkp2Lk0INDUCcC2wWJHhni6ctJZNFhmYatLXFEOCFJwidd7JYoyu5mjgp2ymeGRVgD6U8W3GXdi2OXJZCVJZ5EAkaCP849azLMP7kJRlBkkGNbhH4pSBXPTfMymwR/ChFoQYG/aR9wZ6hGfIIYdDvjw+4Rc3tASDiAifP01J9Zh8Yoyv5Bgk41ha9KZokzK3LWw/SFSeWHUREgS5NGdw0DRJV6eJS4iBbe4JpopyoFDnBi2Nk5RlMmZm4KdTUMgTDKT4wmznK35JWzc4Qh2Kjthf/Bo2I9Pin3FFUWZeSSojVMUpRrmpimZsy7xRDrHy9L/CQiv3D4EwGgqN2HmdzQUoDmcQ0SHfSjKAcPtdKZ6rSiTMjctbMclPpbJAVZ8H/ZY2LEJ2o1GQ35aNCtcUQ4sgQgByWPcKXuKolRkbqpTNg2BEGNpmxF+yMIYT+weIZXNkUhnx8WvXaIhP5nc3PxIFGXW4vT69+WSdV6Iosxu5rCFHXYsbDhpVQfZvOHxXaOMprIT1lavWRDjkIWxA7lSRVECthOhz+1QqChKReamOZnLQCBMIu0Kdic/+cczPLR9iHgqR3csVPGwT73iWIxmqirKgcVpHSzu0B5FUSoyNy3srBPDdgT70EUtdDWH2LBtL/F0tqQNaTmacKYoBxjHwvarS1xRJmWOWtgp8Hcwlim2IX3Wmi7uerIfgyGmIzMVZfbgxrDz6hJXlMmYoxa2m3RmmxM3Bf2cuqaLXcNJdg+ndKCHoswmnPG3GsNWlMmZm4KdS5cknTUF/Zy2prvwdHlbUkVR6ohjYftVsBVlUuauYAfCjDltSJtCfg7qirK0zcbK1MJWlFlEIYatgq0okzE3BTubAn+QsUwOv08I+gUR4dmOla2CrSizCI1hK0pVzE3BduqwE+kc0aC/kPl96ho7KrM5pC5xRZk1BG0M26+CrSiTMkcFOwOBEMlMjohHnM86fCEnHtTBscvb67c2RVFKcSzsgLrEFWVS5qZvOOuxsD2C3dkc4hdvObWOC1MUZRxupzOjjVMUZTLmnoWdz4HJFRqn6KhMRZnlOIIdyKtgK8pkzD3BzjputYCd1tWk8WpFmd24gq0ucUWZlLkn2G4/Yn9YLWxFaQTcOmxNOlOUSZm7gu1Y2FG1sBVldiNCmiABjWEryqRUJdgicr6IPCYim0XkAxWeP11E7hWRrIhcVPbcZ0Rko4g8KiJflpmeruG6xB0LO6IWtqLMetISIqgWtqJMypSCLSJ+4CvAC4AjgUtF5Miy3Z4GrgB+XHbsqcBpwLHA0cBJwBn7verJKLjEnRi2CraizHrShDTpTFGmoJqyrpOBzcaYLQAici1wIfCIu4MxZpvzXL7sWANEgBAgQBDYvd+rngxP0ll5WZeiKLOTjITUJa4oU1CNS3wZ8IzncY+zbUqMMX8FbgN2Oj83G2Mene4ip4U36ayscYqiKLOTtIQIGHWJK8pkzGjSmYgcAhwBLMeK/Nki8twK+10pIhtEZENvb+/+vagj2DlfiHQ2TzQ4N3vDKMpcIiMhguoSV5RJqUawtwMrPI+XO9uq4WXA34wxo8aYUeAm4NnlOxljrjbGrDfGrF+wYEGVp67M5p39AKSxlnVTaO4lwivKXCMjIYLqEleUSalGze4G1orIahEJAZcAN1R5/qeBM0QkICJBbMLZjLrEf3f/UwAk89aybgqpha0o00VEtonIQyJyv4hscLZ1isgtIvKEc9tRq9fLSoigusQVZVKmFGxjTBZ4O3AzVmyvM8ZsFJGPisgFACJykoj0AK8EviEiG53Dfw48CTwEPAA8YIz59Qy8jwIjiYS9zdq3plniirLPnGWMWWeMWe88/gBwqzFmLXCr87gmpHxNhE2yVqdTlDlJVeanMeZG4MaybR/23L8b6yovPy4HvGk/1zgtEo5gPzOcA9AscUWpHRcCZzr3vw/cDry/FidOB5ppSsdrcSpFmbPMqQCvMYbE2BgATw1awVYLW1H2CQP8XkTuEZErnW2LjDE7nfu7gEW1erFssIWmfKJWp1OUOcmcCvAOjWXw5dPgh60DNoFFh38oyj7xHGPMdhFZCNwiIpu8TxpjjIiYSgc6An8lwMqVK6t6sVwwRjMJMAZmuBmiojQqc8rC3j2cIkQWgC0DGUAtbEXZF4wx253bPcAvsQ2UdovIEgDnds8Ex0676sOEWgiSg6zGsRVlIuaYYCcJYYV68157qzFsRZkeItIsIi3ufeA84GFsdcjlzm6XA9fX6jVNuBWATGKoVqdUlDnHnHKJW8G2FvbuhPXW6fAPRZk2i4BfOnN6AsCPjTG/E5G7getE5PXAU8DFtXpBiVjBTo4MEGxbXKvTKsqcYk4J9p6RVMHCzuDWYatgK8p0cOYGHFdhez9wzky8ps8R7LH4IC0z8QKKMgeYUy7xPcNJYsE8eXzknE5n6hJXlNlPIGoFOz06WN+FKMosZk4J9u7hFO0hQ84XLGyLBFSwFWW2E4i2AZCOD9Z3IYoyi5lTLvHdI0naQwZjwgBEgj58Pi0RUZTZTijaDkB2bLi+C1GUWcycsrD3DKdoCebBby1sLelSlMYgHGsHIDemWeKKMhFzRrDzecOekSQtgRwSsBZ2VAd/KEpDEHEEO59UC1tRJmLOCPZAIk0mZ+jID+JrWQhYl7iiKLOf5mgTSRMEFWxFmZA5o2i7h+1ovrb0bvztK2hrCqqFrSgNQnMowAhRJD1S76Uoyqxl7gj2SBIwRMd2QtsKlrRFNIatKA1CU9DPqGnCl1LBVpSJmDMm6J7hJK3E8Wfj0LqMN5+xhoBfM8QVpRHw+YS4RAlnRuu9FEWZtcwZwe4dSbFM+u2DtuW89Khl9V2QoijTYszXTCyrgq0oEzFnXOIjqSwrA3vtg7YV9V2MoijTJuWLElLBVpQJmTOCnUjlOCgwYB+0La/vYhRFmTbJQIxQLl7vZSjKrGXOCHY8nWWFrx/8IWiubgavoiizh4y/maa8CraiTMScEexEKsdyXz+0LgXfnHlbijJvyAZjNJkEGFPvpSjKrGTOKFs8nWUxfRq/VpQGJRdqxYeBtMaxFaUSc0awE+kcC/K9Gr9WlAYlH4zZO1qLrSgVmTOCPZZM0ZnrV8FWlEYlYmdiq2ArSmXmjGA3pfbgI6+CrSgNioStYOd1YpeiVGTOCHZbeo+906qCrSiNiDgWdiqugq0olZgzgt2R3W3vqIWtKA1JIGoFOx0fqPNKFGV2MicEO5c3LMo5FnabtiRVlEYkEG0DIK0WtqJUZE4IdiKd5TDfM4yGF0O4pd7LURRlHwhF2wHIJlSwFaUSc0Kw46kcR8tWBtqOqPdSFEXZR8LN1sLOadKZolRkTgh2YnSI1bKLeOeR9V6Koij7SHNTmFETIZ8crvdSFGVWMicE2+x8CJ8Ykt3H1HspiqLsI7Gwn1GaMCkVbEWpxJwQbN+ehwDILlLBVpRGJRoKMGSa8SUH670URZmVzAnBDvc+RJ9pJagZ4orSsDSHA+w1rQSTe+u9FEWZlcwJwW7eu5GN+VU0RwL1XoqiKPtIc8hPPy2E0yrYilKJxhfsbIqW4c08bFYRDalgK0qjEvD7GJI2IunBei9FUWYljS/Yex7BZ7LWwlbBVpSGZtTfRlNuGHKZei9FUWYdjS/YfZsBeNwsJxr213kxiqLsD4lAh3NH3eKKUk7jC3Y2aW/8TQT9jf92FGU+kw532juJvvouRFFmIY2vcLkUAIFguM4LURRlf8k1OYIdV8FWlHLmgGDbWFcgFKnzQhRF2W+au+2tWtiKMo7GF+ystbBDYRVsRWl0fM0L7J14f30XoiizkMYXbMfCDqqFrSgNT6i1i7wRTLy33ktRlFnHHBBsa2E3hTWGrSiNTltzlEGaSQ+rYCtKOXNAsNOkCRINaw22ojQ67U1B9ppWsqMq2IpSTuMLdjZNhgDNKtiK0vC0R4P004oZ1aQzRSmn8QXbtbBD2jRFURqd9miQvaYF0SxxRRnHHBDsFGnjVwtbUWqMiPhF5D4R+Y3zeLWI/F1ENovIT0UkVOvXbI+G2GtaCejELkUZR8MLdj6bJmUCamErSu35F+BRz+NPA18wxhwCDACvr/ULtjcF6aeFYHoQ8vlan15RGpqGF+xcJkWGADG1sBWlZojIcuBFwLecxwKcDfzc2eX7wEtr/bptTtKZjzyMDdT69IrS0MwJwbYxbBVsRakhXwTeB7hmbhcwaIzJOo97gGW1ftGA30ci2G4faBxbUUpoeMHOZ9KkCdCsk7oUpSaIyIuBPcaYe/bx+CtFZIOIbOjtnX55VmEAiPYTV5QSGl+ws0nSBNTCVpTacRpwgYhsA67FusK/BLSLiPuPthzYXulgY8zVxpj1xpj1CxYsmPaL5yJd9o5a2IpSwhwQ7DQZE6BZk84UpSYYY/7dGLPcGLMKuAT4ozHmMuA24CJnt8uB62dkAe4AELWwFaWEhhdssintdKYoB4b3A+8Rkc3YmPa3Z+JF/DEVbEWpRMOrnMllyNCsFraizADGmNuB2537W4CTZ/o1W6JR9tJK58iOmX4pRWkoqrKwReR8EXnMaZjwgQrPny4i94pIVkQuKntupYj8XkQeFZFHRGRVjdZuyaZIEaC1KVjT0yqKUh86okG25zsxQxVD5Ioyb5lSsEXED3wFeAFwJHCpiBxZttvTwBXAjyuc4gfAZ40xR2Cvzvfsz4LHkbO9xNtUsBVlTtAWDbHTdJEf7Kn3UhRlVlGNS/xkYLPjDkNErgUuBB5xdzDGbHOeK2lN5Ah7wBhzi7PfaG2W7XmNXJqsBIkE1SWuKHOB9qYgO0wXMvxEvZeiKLOKalziy4BnPI+n0zDhUGBQRP7P6Un8Wcdirxm+fAYJ1LylsaIodaKjOcgu04kvPQypkXovR1FmDTOdJR4Angu8FzgJOBjrOi9hfxot+E0GCYRrsFRFUWYDbU0hdhinFlvj2IpSoBrB3g6s8DyesGFCBXqA+40xW5yWhr8CTijfaX8aLfjzGfwq2IoyZ2iPBtnpCvawxrEVxaUawb4bWOuM1gthGyncUOX578Z2R3JV+Gw8se/9xhgCZPAHVbAVZa7QEQ2xE1ewtbRLUVymFGzHMn47cDN21N51xpiNIvJREbkAQEROEpEe4JXAN0Rko3NsDusOv1VEHgIE+GbNVp/P4cOoYCvKHKI1EmCX6cAg6hJXFA9VNU4xxtwI3Fi27cOe+3djXeWVjr0FOHY/1jgxuRQAwVBkRk6vKMqBJ+D30RJtYsTXSau6xBWlQEO3Js1l0gAEwmphK8pcoisWZq+/Wy1sRfHQ0II9Grdl3aFwU51XoihKLelqDrGbbhhWwVYUl4YW7JFEAoBwWF3iijKX6G4J05PvtBa2MfVejqLMChpasEfjjmBH1MJWlLlEd3OIrZl2yMQhOVjv5SjKrKChBTvuWNhNEbWwFWUu0R0LszXdYR9oHFtRgDkj2NE6r0RRlFrSFQuz03TaBxrHVhSgwQU7MTYGQFOTusQVZS7RHQux3XTbB4NP13cxijJLaGjBHksmAWiOqoWtKHOJrliYPbST84dhYFu9l6Mos4KGFuxk0rrEQ1qHrShzigWxMAYfo03LYe/Wei9HUWYFDS3YKcfCFr8KtqLMJbpidmTuQHgZDKhgKwo0uGCnUzaGjc7DVpQ5RTTkJxL0sTuwxLrEtRZbURpbsFMp20scvwq2oswlRITuWJjtLIZMAkZ313tJilJ3GlqwMynrElfBVpS5R1cszJM5ZzKvxrEVpbEFO5tRwVaUucqCWIjH045gaxxbURpcsNOOSzygSWeKMtfoag6zMdEG4lMLW1FoYMHO5Q35rBvDDtZ3MYqi1JzulhB7EgbTthz2bqn3chSl7jSsYI8kM4RM1j7Qsi5FmXN0NYfJ5g3Z1lXqElcUGliwh8YyBHEFW2PYijLX6G6xF+KJ2Ap1iSsKDSzYw2NZQpIhLwHwNezbUBRlArqb7YX4YGQ5jO2F5FCdV6Qo9aVhlS6dyxEkh/Fp/FpR5iJdMWth7wkstRvUylbmOQ0r2NmcIUSGvLrDFWVOclBXlJDfx/0JZ2pX/+b6LkhR6kzDCnYubwiRVQtbUeYokaCf41a08btdMVva1ftYvZekKHWlcQXbGEKSxaiFrShzllNWd3H/jgT59lXQp4KtzG8aVrCzeUMQFWxFmcucvLqTXN4wEF0FvY/XezmKUlcaVrBzTgxbS7oUZe5y4kEd+H3Ck2aZjWHnsvVekqLUjcYVbONY2D4VbEWZqzSHAxy9rI27R7shn7GjNhVlntK4gu0kneksbEWZ2zxrdSe37+20DzSOrcxjGluwJat9xBVljnPSqk42ZZfYB30ax1bmL40t2GQw2kdcUeY0RyxtZYQoifACTTxT5jUNK9hulrho0pmizGmWtkVoDvnZFVqpLnFlXtOwgp3PG4LkdBa2osxxRIRDFrWwOb/UWtjG1HtJilIXGlaws3m3rEtj2Ioy1zl0YYz7xxZCegRGdtZ7OYpSFxpWsHP5PCHJImphK0pNEZGIiPxDRB4QkY0i8p/O9tUi8ncR2SwiPxWRAxaPWrsoxn3JxfaBtihV5ikNLNi2rEtj2IpSc1LA2caY44B1wPki8izg08AXjDGHAAPA6w/UgtYubGFzfpl9oIKtzFMaVrDdpDONYStKbTGWUedh0PkxwNnAz53t3wdeeqDWtHZRjF7aSAVboXfTgXpZRZlVNKxg542NYfu0cYqi1BwR8YvI/cAe4BbgSWDQGOP2Bu0Blk1w7JUiskFENvT29tZkPUvbmoiGAuwOrVILW5m3NKxgF8q61MJWlJpjjMkZY9YBy4GTgcOncezVxpj1xpj1CxYsqMl6fD5h7cIYm1kGvY9qprgyL2lYwc5lc4Qkh6iFrSgzhjFmELgNeDbQLiIB56nlwPYDuZZDFrZw/9hiGBuAeN+BfGlFmRU0rGCTzwCoha0oNUZEFohIu3O/CXge8ChWuC9ydrscuP5AruvQRTHuSy6yDzSOrcxDGlewsykAjWErSu1ZAtwmIg8CdwO3GGN+A7wfeI+IbAa6gG8fyEWt6m7miUKmuAq2Mv8ITL3L7MTkrIWt87AVpbYYYx4Ejq+wfQs2nl0XVnZG2UUnmUAzQU08U+YhDWthS85a2CrYijI/WNkZBYT+ptVqYSvzkoYVbNTCVpR5RXM4QHcsRE9gJezRTHFl/tGwgl2wsDXpTFHmDSs7ozyQXwOJPuh/st7LUZQDSsMKdtHC1uEfijJfWNkZ5eaxI+yDLbfVdzGKcoBpWMGWXNre8auFrSjzhZWdUTaMtGHaVsDWO+q9HEU5oMwBwVYLW1HmCyu7mskbYXTpabD1Tsjn6r0kRTlgNK5g5x3B1hi2oswbbKY4PNNxCiSHYOf99V2QohxAGlawfeoSV5R5hyvYG8Pr7IYt6hZX5g+NK9gFC1vLuhRlvrCwJUw44OPx0QgsPAoeu0nLu5R5Q+MLttZhK8q8wecTVnRGeXpvAk68Anr+AXd/q97LUpQDQsMKtuS1cYqizEcO6ozyVH8CTn4jHPI8+P2HYPcj9V6Wosw4DSvYfk06U5R5ycquKNv64ySzeXjpVyHUDLd9ot7LUpQZp2EFu+gSV8FWlPnEWYctJJnJc9umPRBbCAefBTsfqPeyFGXGaXzB1qQzRZlXnLqmiwUtYX5533a7YfHRMPQMjA3Ud2GKMsNUJdgicr6IPCYim0XkAxWeP11E7hWRrIhcVOH5VhHpEZH/rcWiAXz5rL2jMWxFmVcE/D4uOG4ptz22h8FEGhYdY5/YvbG+C1OUGWZKwRYRP/AV4AXAkcClInJk2W5PA1cAP57gNB8D7tz3ZY7Hry5xRZm3vOz4ZWRyht88uBMWHWU37nq4votSlBmmGgv7ZGCzMWaLMSYNXAtc6N3BGLPNGXqfLz9YRE4EFgG/r8F6CwTyGbL4wdewXn1FUfaRo5a2snZhjBse2AEtiyHaBbsfqveyFGVGqUbtlgHPeB73ONumRER8wH8D753+0ibHb9JkRfuIK8p8REQ4+/CF3P/0oM0WX3S0WtjKnGemzdO3AjcaY3om20lErhSRDSKyobe3t6oT+0xGBVtR5jEnHNRBOpfn4e1DsPgY2PMo5LL1XpaizBjVCPZ2YIXn8XJnWzU8G3i7iGwDPge8VkQ+Vb6TMeZqY8x6Y8z6BQsWVHXiQF4FW1HmMyce1AHAhqcGrIWdS8HeJ+u8KkWZOaoR7LuBtSKyWkRCwCXADdWc3BhzmTFmpTFmFdYt/gNjzLgs830hoBa2osxrumNhVnc3s2HbgC3tAtilcWxl7jKlYBtjssDbgZuBR4HrjDEbReSjInIBgIicJCI9wCuBb4jIjNdXBEyGnGhJl6LMZ05Y2cG9Tw9gug8FXxC231PvJSnKjFFVDNsYc6Mx5lBjzBpjzCecbR82xtzg3L/bGLPcGNNsjOkyxhxV4RzfM8a8vVYLD2jSmaLMe9av6mBvPM3WgQysfR489HPIZeq9LEWZERq2JipgsuR9KtiKMp9Z78Sxr9vQwzXp0yG+B56oaQWposwaGlew0Ri2osx31iyI0RoJ8PU7nuTDjy5jyN8J9/6w3stSlBkhUO8F7CtBkyHna6r3MhRFqSM+n/ChFx9J70iKv2zu46bes7jkiV/ByC7bUEVR5hANa2FbwVYLW1HmOxevX8HbzjqEEw/q4Jujp4HJwR2fqfeyFKXmNKxgB8hoDFtRlAJHLGnlyfxieo9+PWz4Njzxh3ovSVFqSsMKdpAseS3rUhTF4cglrQDcvvwtsOBwuP5tkNhb51UpSu1oXME2GfI+FWxFUSwrO6M0h/w8vDsFL78aRnfDn79Q72UpSs1oWMEOkSHnV5e4oigWn084fEkrj+4cgSXHwbGvgn980yagKcocoGEFO0gGoxa2oigejljSwqM7hzHGwJkfgHwG7vxcvZelKDWhgQU7p0lniqKUcOSSNkZSWXoGxqBzNZzwWrjne7BnU72Xpij7TUMKdj5vCJEh71cLW1GUIkcsaQHggZ5Bu+HMf4dIK/zySsim67cwRakBDSnY2VyesGQxvnC9l6IoyiziiCWtLG2L8KmbNjGYSENsIbzky7DzAbhj3GRfRWkoGlKw89kUAEYtbEVRPESCfr5y2QnsHk7ynuseIJ83cMSLYd2r4c9fhMFn6r1ERdlnGlKwsxlXsDWGrShKKcev7OCDLzyCP27awx837bEbz3w/YGxDFUVpUBpSsPOuYKtLXFGUCvzTKQfREg5wyyO77Yb2lXD4i2wCWmasrmtTlH2lIYd/5FzBDqhLXFGU8YQCPk4/bAG3btpDPm/w+QROeTM8+mvbTEV8tlb7sBfUe6mKUjUNKtj2Cllj2IqiTMS5Ryzktw/u5KHtQxy3oh0OOg0WHgV3fNru0LoM1j4ffA3paFTmIQ35l2qyGQBEBVtRao6IrBCR20TkERHZKCL/4mzvFJFbROQJ57aj3mudjDMPXYhP4NZHHbe4CLz8G/CSL8HzPwnD22H7hvouUlGmQUMKdj6TBNBOZ4oyM2SBfzXGHAk8C3ibiBwJfAC41RizFrjVeTxr6WgOceJBHfzh0T3FjYuPgROvgOMvA38INv6qXstTlGnToIJtY9gENOlMUWqNMWanMeZe5/4I8CiwDLgQ+L6z2/eBl9ZlgdPgnCMW8cjOYR54ZrD0iUgbrDkHHrke8vm6rE1RpktDCrZx6rBFk84UZUYRkVXA8cDfgUXGmJ3OU7uARfVaV7VcctIKlrU38bYf38tQIlP65FEvheEedYsrDUNDCnaxcYpa2IoyU4hIDPgF8C5jzLD3OWOMAcwEx10pIhtEZENvb+8BWOnEtEdD/O8/Hc/u4ST/+rMHSp887AXWLf7AtfVZnKJMk4YUbOP0BFYLW1FmBhEJYsX6GmPM/zmbd4vIEuf5JcCeSscaY642xqw3xqxfsGDBgVnwJBy/soN3nXsof3h0N4/s8Fx3RNrguEvgvh/CwFP1W6CiVElDCrYbwxaNYStKzRERAb4NPGqM+bznqRuAy537lwPXH+i17Sv/dPJKgn7hF/f2lD5xxgcAgdu1z7gy+2lIwSbrJp2pha0oM8BpwGuAs0XkfufnhcCngOeJyBPAuc7jhqCjOcTZhy/k+vu3k8l5kszalsHJb4QHr4Xdj9RvgYpSBQ0p2CanFraizBTGmD8bY8QYc6wxZp3zc6Mxpt8Yc44xZq0x5lxjzN56r3U6vOKE5fSNprnz8bK4+nP/1brHf/EGSMfrszhFqYIGFWyNYSuKMj3OPGwhnc0hfrahzC0e7YRXfBt6H4Xr3wamYi6dotSdhhRsMq5gq4WtKEp1hAI+LjlpBb/buIvbHyvLlzvkHDjnI7Dxl/DLN0E6UZ9FKsokNKZgOy5xXyBS54UoitJIvPOctRy2qIX3/uxB+kZTpU+e9i9w1ofgwevgO8+HREN5/JV5QGMLdlBd4oqiVE8k6OfLlx7PcDLDh69/uPRJETjj3+Cffgp7HoUb3qHucWVW0ZiCnXUtbHWJK4oyPQ5b3MJbz1zDjQ/t4qGeofE7HPp8OPcq2PQbuPtbB3x9ijIRjSnYuQxp4yfg99d7JYqiNCCvf85q2qNBPn/LY5V3eNZbYe15cPMHtdxLmTU0pGBLLkWaID6Rei9FUZQGpCUS5E2nr+G2x3q556kKsWqfD176NYi0wv9dWez9oCh1pCEFm1yGNAECfhVsRVH2jctPPYgFLWH+9boH6C9PQANo7oYL/gd2PwQ3/4eKtlJ3GlKwXQvb71PBVhRl34iGAnz91SewcyjJG36wga/f8ST/et0D7BpKFnc67AVw0httLPtL63R+tlJXGlKwyWXImAB+dYkrirIfnHhQJ1+65Hjuf2aQT920iV/c28N379pautMLPwuv+aW1uH/1Voj31WexyrynIQXbl0uRJqAWtqIo+835Ry/mz+8/m7s/eC7nHrGI/7t3O1lvv3ERWHM2XPQdyI7Bn79Qv8Uq85qGFGzJp9UlrihKzVjW3sSCljCvXL+c3pEUd5T3GwfoXgvHXWrd40/cAnd8BvqeOPCLVeYtjSnYuTQpAgRUsBVFqSFnH76Q7liFfuMuZ7wP8lm45iK47RO27EtRDhCBei9gX/DlM2TUJa4oSo0J+n28dN0yvnvXNl7yP3/m4AXNfOaiYwkHnJ4PHavg4h9Achh2Pwx//V9rZXevreu6lflBQ1rYvlyatFGXuKIoteefn7Oa845cRFtTkOvv38GP/vZ06Q6HvwjWXWp7j/tD8Pev12ehyryjMQU7n9akM0VRZoRl7U187dUn8qM3nMJzDunmf//4BMPJzPgdYwvhmIvh/h/DU3/VYSHKjNPAgh0k4GvI5SuK0iB84AWHM5DI8PXbn6y8w7PfamPa3z0fPrvGJqLl85X3VZT9pCEVz41hq14rijKTHL2sjZeuW8q3/rSVB54ZHL/DoqPgHffCpT+Fo19hE9F+cgmkRg74WpW5T0NKni+fcbLEG3L5iqI0EFddcBQLWsK89Zp7GYinx+/QvgIOOx9e/k140X/D5j/A9y+AeP+BX6wyp2lIxfPnbdKZhrAVRZlp2qMhvvbqE+gdSXHO5+/gwq/8hR///enxO4rASW+AS66BPY/A918M6cSBX7AyZ2lMwTZpMhJEtDWpoigHgGOXt/ON157IWYctJJfP8x+/fIir75wgrn3YC4qi/YerituNgYd/AXu3Vj5OUaagIeuw/fkMWQnWexmKoswjzjpsIWcdtpBMLs+7rr2f/7pxE5v3jPKucw9laXtT6c6HnAunvAX+/jU7V/uQc+CWD8NdX4autfCmOyDUXJ83ojQsDWphZ8g25rWGoigNTtDv40uXrOONz13Nr+7bwZmfu53v37UNY0zpjud+BLoPhWteAV88xor12udD/2btkKbsE40n2Pk8fpNVC1tRlLoR8Pv44IuO5I/vPYPnHtLNR27YyLt+ej93PdnHiFuzHWyCK34L533cCveZ/w7/9FM47Z1wz3dh02/r+yaUhqPxzNSczdLM+kJ1XoiiKPOd5R1Rvvna9fzPHzfzpVsf5/r7d9AaCfCbdzyXlV1R21zl1HfYH5ezPgRbbofr3w5LT4DWJXVbv9JYNJ6FnUsBkEUtbEVR6o/PJ/zLuWvZ8KHn8Z0r1pPNGz7220cmPiAQgld8G7JJ+OWbIJc9cItVGpoGFGzrbspJ4zkHFEWZu3Q2hzj78EW84+y13PLI7sojOl2618ILPg1b74CvnAz3/Qh6H4NshTpvRXFoPMGOtPPZNT/g1sBz670SRVGUcbzuOatY3d3MW390Dy/40p/44h8er7zjCa+FS35iY93Xv80K95fXwciuA7pepXFoPMH2B9gZOohRf1u9V6IoijKOcMDPVy87gRcfu5RY2M8X//AENz60s/LOh78Q3vQnuPJ2uPCrMDZg3eTaj1ypQFWCLSLni8hjIrJZRD5Q4fnTReReEcmKyEWe7etE5K8islFEHhSRV9Vi0TljCGibM0VRZilHLGnl0xcdy4/f+CyOXd7GB3/5EHtGkpV39vlg6fFw/GVw/qdsQtot/w8GK3RTU+Y1Uwq2iPiBrwAvAI4ELhWRI8t2exq4Avhx2fYE8FpjzFHA+cAXRaR9P9dMNm/wqWArijLLCfp9fP7i40ikc5z3hTt524/v5d6nByY+4ITXwrGvgr/+r63d/v5LVLiVAtVY2CcDm40xW4wxaeBa4ELvDsaYbcaYB4F82fbHjTFPOPd3AHuABfu76HxeLWxFURqDQxa28IPXncw5hy/ib0/2c/HX/8p3/7K10GgllzdsHxyzO4vAy74Bb/0bnHsVbL8PvnYa3PIRO1REk9LmNdWkWi8DnvE87gFOme4LicjJQAiYoAFv9WTzBr9O6lIUpUE45eAuTjm4i6GxDP963QP8568f4ZEdw1x1wVG857r7+f0ju7nmDadw6ppuK9oLj7A/R70Mfvuv1uL+yxehZQmsu8xa3QNb7YSwztX1fnvKAeKAqJ6ILAF+CPyzMWZcNoWIXCkiG0RkQ2/vJKUQDrm8wa96rShKg9HWFOTq15zIO89Zy8/u6eHUT/2Rmzfupq0pyId+9TCpbK70gI5V8OpfwAeehkuvhQWHwZ8+B1tugz2b4GeXQ2aC2Lgy56hG9rYDKzyPlzvbqkJEWoHfAh80xvyt0j7GmKuNMeuNMesXLJjaY55TC1tRlAbF5xPe87xD+cwrjkUEPnPRsXzxVevY0hvn6ju2VD4o1GyngL32enjfVnjvE/Dyq2HnA/Crt8Bjv4OhngP7RpQDTjUu8buBtSKyGivUlwD/VM3JRSQE/BL4gTHm5/u8yjJyGsNWFKXBufikFbxy/fLCmOAXHbOEL/zhcZLZHP9yzqGEAhMYJdFOe3v4C+H0f4M7Pwsb/w/8ITjj/XDav4BfO0HORaYUbGNMVkTeDtwM+IHvGGM2ishHgQ3GmBtE5CSsMHcALxGR/3Qywy8GTge6ROQK55RXGGPu359FZ/N5/DoLW1GUBkc832OfvuhYoiE/X7ntSX70t6dZ2t7ES9ct5U1nrJn4BGd/CNa/3lrXf/1f+OPH4PZPQqAJghEIRuGkN9he5vkc9D9hB5H4/Afg3Sm1pqr+nsaYG4Eby7Z92HP/bqyrvPy4HwE/2s81jiOfB79a2IqizCFi4QCffeVxvPCYJfz+kV08vnuUT960CWBy0W5dYn9WfB8evxme/quNa2eTsPdJW9P91F9s69OBrdB5MDzrrbZ8LDkIN7wT0qNw4VdsjFyZtTRkQ+5sPk8o0JBLVxRFmZSzDl/IWYcvJJc3/Mu19/HJmzbx0w3PsDee5u1nHcIbnnswxhj6RtN0x0IlVjqHPt/+uBgDf/pv+OPHYcmx8Pz/god/ATe+F275MIgfMNad/o3T4Xkfg5PfaDPVK/Hw/8HGX9rSs1B0Rj8HZTwNqXo26UwtbEVR5i5+n/D5i9fRHQuzc2iMwUSGj//2UQDufKKPOx/v5fDFLbzg6CU0h/0cubTVloV5EYHT3wsnXgHRLvv4WW+F7ffCvd+D1Iit9w5EbD/zm/4NHr/JCveioyA5BL2bYNl6GNhm98kk7Lle8sUD+4EoDSrY2ppUUWYMEfkO8GJgjzHmaGdbJ/BTYBWwDbjYGDNJyy6lFoQCPq664CgAUtkcr/ve3Xz8t4/SFPRz5ekH8/ct/XzBGS7iE/j2FSdx1mELx5+o2SPkIrD8RPvj5bKfw4bvwO8/BF8/DVqX2UEkJgeLjrbH+UNw9Mvhnu/CqufAMRfZ8aCbfgOLj4GuSVz3yn7TkIKdzWlrUkWZQb4H/C/wA8+2DwC3GmM+5cwT+ADw/jqsbd4SDvj5+qtP5Lt/2cZLjlvK6u5mAMbSORLpLK/59j9454/v44dvOIWjlrYSnG6zChE46fVwxEvg0V/bWu/uw6DjILj90zDcAxd91z6/ZxP84vVW4Ed3Q/9miLTDZT+zMfKdD8BBp9nEN6VmiNseb7awfv16s2HDhkn3Oe8Ld7BmQYyvvfrESfdrRDKZDD09PSST2gxBsUQiEZYvX04wWFqqIyL3GGPWz8Rrisgq4DceC/sx4ExjzE6nEdLtxpgpM5Sq+X9WasP2wTEu+J8/0x9PIwKru5tZt6IdQRgay3Ds8jaOWNLKzRt3sXnPKJ975bEcsrClupOn41akXas8OQz3fh/u/jaEYvCsN9tY+VCPzUY3OVh5KlxyTbEMTZmUav6fG9PCnsMx7J6eHlpaWli1alVpMokyLzHG0N/fT09PD6tX17UF5SJjjDsjchewqJ6LUcazrL2J699+Gn/Z3Mf2wSSP7BjiT0/0EfAJ0ZCfWzftxhibjR70C6/42l/5xMuOpino59BFLazonCSJLNRc6kKPtNpSsVPfUdy29jz4w39CbCG0LoWb/wO+dQ4c/QpY+SxYdToEQjP3AcwDGlKw53LSWTKZVLFWCogIXV1dVNOy90BhjDEiMqFrTkSuBK4EWLly5QFblwLLO6K86qTKn3nfaIpNO0c44aB2+kbSXP7df/D2H98HQEskwC/femr1FnclYgvhpV8pPl54pBXtP33eWtyRdlh9uhXzthW2HnzpOnucUhUq2LMQFWvFyyz5e9gtIks8LvE9E+1ojLkauPr/t3fvcVFXeQPHP4fhJgMKiJqKiSSCsjgCKnhHqc3bgppalK6XXTNz17Rn6zG72G71etrNpy2fLdvSMvdxdbNNs/LyBF43NbxkCV4WDVS8kEIiiBdgzvPHbxhBLqIhM6Pf9+s1L5jfzPzm+zvM4cw5v/M7XzCGxBsrQFG3IF8v+oZ5AXB3c3dW/6YP+3ILQcGMZXv51Ye7mDbgHtZlnub+yLtI6fkTv2yF9IGpm43h9OytxmpsubvgyEa4UnT1eYGh0PE+iBhmJDdxMxmz0929odW1mZzvbK7bYDvHP7HbTn5+PomJiQCcPn0ak8lExfru6enpeHrWPqS1a9culixZwvz58+t8j969e7Nt27YGi3nmzJmsWLGC48eP4yZrzN8qq4EJwKu2n586NhzxU/l5e9C7ozF7/K/jY0l5bwezP9mHn7c7mw6d4XThJab0D8Xsaar2pVFrjdbUb/KvpxnCBxu3CiUFxuViubuMRV32fAjpf63+2q4PweD/unoe3FoO+z6GbfMhZgLEPXqzh++SXLLBLrNq3E3SYN8KzZs3Z+/evQC8+OKL+Pr68rvf/c7+eFlZGe61LFrTvXt3une//hyohmysrVYrK1eupF27dmzevJmBAwc22L4rq+u4bzdKqWVAAhCklMoF5mI01B8ppX4FHMVYdljcJmLbB7Biai/KrFa6BvvzzCf7eDMtizfTsnBTxgx1fx8P7o+8i7b+Tfh7+jEUxmVkFbPVAc5fKsXPy/36o0I+gdC+t3HrMwMuF8PRbXD5PFjLwKspnNhtpBQ9sNq4hMzLD45uh6KTxuPrn4G2MRB8S+ZdOiWX7I5Yb/MhcWczceJEHnvsMeLi4nj66adJT0+nV69eREdH07t3bw4dOgTApk2bGD58OGA09pMnTyYhIYHQ0NAqvW5fX1/78xMSEhg9ejQRERE88sgjVFy1sGbNGiIiIoiNjWXGjBn2/V5r06ZNREZGMm3aNJYtW2bfnpeXx8iRI7FYLFgsFvuXhCVLltC1a1csFgvjx4+3H9/HH1/NTVM5vn79+pGUlESXLsbQ3IgRI4iNjSUyMpJ3333X/pp169YRExODxWIhMTERq9VKWFiY/dyz1WqlY8eOTnUuujZa6xStdWuttYfWOlhrvUhrna+1TtRah2mt79VaFzg6TtGwLO38iW0fiIfJjddGd2XBIzE8MySCxxM6Mi7+bizB/vw9/RivrDmAv48H5y6W8sCCbWw4mEe5VbPoX9nEvvQl4xelc7rwBq9y8fKFTj83ruu2PGQkNkl8HqZugehxkH8Eju0wJq+N+RCe+Bb82sDHkyDtJVg1HTb/CQ5+YVw7XlnefiN/+G3AJbsMZXfIkPjvP8tk/8nzDbrPLm2aMvcXkTf8utzcXLZt24bJZOL8+fNs3boVd3d3UlNTmTNnDv/85z+rvebgwYNs3LiRoqIiwsPDmTZtWrVLk7755hsyMzNp06YNffr04auvvqJ79+5MnTqVLVu20KFDB1JSUmqNa9myZaSkpJCcnMycOXMoLS3Fw8ODGTNmMGDAAFauXEl5eTnFxcVkZmby8ssvs23bNoKCgigouH6bs2fPHjIyMuwztN9//30CAwO5ePEiPXr04IEHHsBqtTJlyhR7vAUFBbi5uTFu3DiWLl3KzJkzSU1NxWKxUJ/0sUI4mlKKIVGtq20vvFhKfvFlQlv4knP2ApMW72Ty4l34ebtTdKmMnh0C2X30R+59fTMt/bzw8TIxZ2jn6iuw1VerSBj6Ws2PjV4EHwyFf/3ZWBimOO/qY4GhEDUWik7BniXg2woe3WhMeHNhLtlgWyUfdqMbM2YMJpOR4aewsJAJEyaQlZWFUorS0tIaXzNs2DC8vLzw8vKiZcuW5OXlERxcNUdMz5497du6detGTk4Ovr6+hIaG2hvJlJSUKr3ZCleuXGHNmjW8/vrr+Pn5ERcXx/r16xk+fDgbNmxgyRJj3Q+TyUSzZs1YsmQJY8aMISjI+OcRGHj960N79uxZ5XKq+fPns3LlSgCOHz9OVlYWZ86coX///vbnVex38uTJJCcnM3PmTN5//30mTZp03fcTwpk1a+JBsybGl+6QIDNrn+hH2oEfWJtxitj2AUzsHUL22Qu8vekIl0rLyThRyLiFXzMqJpjiS2U09/XkmaGd8fEwsfKbE3Rs6Yulnf/NBdOuJzyVBR5m43Kxy8WQlwkndhlJUDb/EZSbsSzrvhWw/GFImGMMtTcJgKAw49Y0GNzcjHXXT+2Fi+cgpB+YnK95dL6I6uFOOYd9Mz3hW8Vsvnqe6vnnn2fgwIGsXLmSnJwcEhISanyNl5eX/XeTyURZWdlNPac269ev59y5c0RFRQFQUlJCkyZNah0+r427uztWqxUwhq6vXLlif6zycW/atInU1FS2b9+Oj48PCQkJdS5w065dO1q1asWGDRtIT09n6dKlNxSXEM7O28PEsK6tGdb1am88tIUv88ZYALhwuYwXPs1k9bcnCfZvwtEDJaRnFxBg9iQ9u4AmHiYWTexO73uCuHC5DLOXu/11u4/+SAs/L0Kam2niWUs60CYBV3/38oW744xbr+lQeALQ0CzYuEZ8+cPw9zHV9+HuDc07GufOzxjZ0fBrY+QVj5tqrABntRo/axrZPfCZ8UWh/1M1py0tPmOMADTAqLBLNtjlVo3bHTAk7qwKCwtp27YtAIsXL27w/YeHh/P999+Tk5NDSEgI//jHP2p83rJly1i4cKF9yPzChQt06NCBkpISEhMTWbBgATNnzrQPiQ8aNIiRI0fy5JNP0rx5cwoKCggMDCQkJITdu3czduxYVq9eXeuIQWFhIQEBAfj4+HDw4EF27NgBQHx8PI8//jjZ2dn2IfGKXvavf/1rxo0bx/jx4+0jFELcKcxe7vz3WAvzxnRFKcVXh8/ym7/v4XThJX6fFMnSr48y6YOdBJo9OVV4iSE/u4tH4trzwqcZfH/2AgDuboquwc0YGRPMuLi763+ZY7O2V3+PGArjjZEx2vWEKyVw9pBxbvzsv+FslpHUJG4q+ATBzoWw7j/h2DajB75nCZRdBHMLiHsMev0GtNXIPf6v1439njsOSf8DxaeNnn2TQGPm+6ZX4RdvGufnfyLXbLAl+YdDPf3000yYMIGXX36ZYcOGNfj+mzRpwttvv83gwYMxm8306NGj2nNKSkpYt24d77zzjn2b2Wymb9++fPbZZ7z55ps8+uijLFq0CJPJxIIFC+jVqxfPPvssAwYMwGQyER0dzeLFi5kyZQrJyclYLBb7e9Zk8ODBvPPOO3Tu3Jnw8HDi4+MBaNGiBe+++y6jRo3CarXSsmVLvvzySwCSkpKYNGmSDIeLO1pFI9unYxBp/5EAQKDZk+FdW/P8pxm4KUVLP2/+9+ujrM04TUs/L956OAaNZv/J82zNOsvzqzLYl3uOl0dE4enuhtaakivl9l75dd1T6QoSTzP4tjBmn9ek8y9g+1+MFKQoiBwJ/ncba6SnzoX096DkrJFzPHai0ZBveQ0OrYGLtrkxbh5gLTV6922ib6rcruVya4lrrenwzBpmJIbx5H2dGjGyxnHgwAE6d+7s6DAcrri4GF9fX7TWTJ8+nbCwMGbNmuXosG7Yrl27mDVrFlu3bv1J+6npc3Er1xJvKLKWuLgR358pZtXek4yPb08Lv6uny7TW/Dk1i/lpWZjcFAE+nly4XMbF0nIswc14JL4990fehclNMT8ti/0nz/PSiJ9VueTspuTtN5ZhbRZcEYhxmdnuD6FFONwzCMLuM7Z//Q7k7oTgHkYPu+B7CB1Y9frzOtyWa4mXW40vGNLDvr299957fPjhh1y5coXo6GimTp3q6JBu2KuvvsqCBQvk3LUQ9RTawrfGjphSiifv60TM3f7szCmg4MIVzJ7umL3c+WLfKZ7++DueXbkPs5c750pKMXuaGD5/K88O68KomLZ8c+wcL67OJLKtcZVMxcS567p2pTWloEuycbt2e/w0YNrNHXg9uVwP+3JZOeHPreOp+8OZPrBjI0bWOKSHLWoiPWwhaqa1Zs+xc/xf5mmO5pfw6IBQ7mrqzYxl37Dr6I/4erlTfLmM1s28+aHoMq38vHgs4R66tw9k9bcnWZ95msul5Xh7mojr0Jzkbm2ID23e6MdxW/ewZeEUIYQQSili2wcQ2z6gyvaPpvZiR3Y+K/ecoFVTb6YP7MihvCLmfLKPFz7NBMBNwYBOLQjy9eLHkit8/u1JPtp1nEUTupMQ3pKME4VoDZFtmtZvGdZbzGUbbBkSF0IIURs3N0Xve4KqLNrSrZ0/a57oR1ZeEek5BfQPa1ElrWjx5TLGvrOd6Uv3EBfanA0HjRw3AT4edGnTlGB/H/IvXOH8xVISIloQ3sqPFbtyOVV4kZ9H3kVCuLG/zBPnST2QR+fWTRnRrQ3upoZZN8RlG2zpYQshhLgZYa38CGtVPZWor5c7H0zqwci3vmLH9/k8dX84bf2b8K/DZ8n6oZi0g3kEmj3xdHfjT+uMJZkDfDxo39zMa+sP8dr6Q/Z9mdwU5VbNXzZk8dKIn9Ev7KevcuhyDbZC0a2dPy39vB0dihBCiNtMq6berHmiH1pDgNnITjgium215x3Nv8DhH4rp0zEIbw8TxwtK+C63kGMFJbTx9+bezq346vBZ3kjNwtujYdZgcLn1PZv5eLBqep8qK+uIhjNw4EDWr19fZdsbb7zBtGm1z35MSEigYmLR0KFDOXfuXLXnvPjii8ybN6/O9161ahX79++333/hhRdITU29gejrNnPmTNq2bWtf1UwIIWri7+Npb6xr0765mcTOreyNcbtAH4Z1bc20hHtI7tYWs5c7P4+8iy9m9KVHyPWXQa4Pl2uwxa2VkpLC8uXLq2xbvnx5nQk4KluzZg3+/v439d7XNth/+MMfuPfee29qX9e6Ng3nrXIjS6sKIW5/9V6ZrR6kwRZVjB49mi+++MK+nnZOTg4nT56kX79+TJs2je7duxMZGcncuXNrfH1ISAhnz54F4JVXXqFTp0707dvXnoITjGuse/TogcVi4YEHHqCkpIRt27axevVqnnrqKbp168aRI0eqpL1MS0sjOjqaqKgoJk+ezOXLl+3vN3fuXGJiYoiKiuLgwYM1xiVpOIUQrs7lzmHfUdbOhtP7Gnafd0XBkFdrfTgwMJCePXuydu1akpOTWb58OWPHjkUpxSuvvEJgYCDl5eUkJiby3Xff0bVr1xr3s3v3bpYvX87evXspKysjJiaG2NhYAEaNGsWUKVMAeO6551i0aBG//e1vSUpKYvjw4YweXXXN3UuXLjFx4kTS0tLo1KkTv/zlL+3rhAMEBQWxZ88e3n77bebNm8fChQurxSNpOIUQrk562KKaysPilYfDP/roI2JiYoiOjiYzM7PK8PW1tm7dysiRI/Hx8aFp06YkJSXZH8vIyKBfv35ERUWxdOlSMjMz64zn0KFDdOjQgU6djBWQJkyYwJYtW+yPjxo1CoDY2FhycnKqvb4iDeeIESNo2rSpPQ0nwIYNG+zn5yvScG7YsKFB0nBaLBbi4+PtaTh37NhRaxrOilSgkoZTCFEb6WE7szp6wrdScnIys2bNYs+ePZSUlBAbG0t2djbz5s1j586dBAQEMHHixDpTS9Zl4sSJrFq1CovFwuLFi9m0adNPirciRWdt6TklDacQ4nYgPWxRja+vLwMHDmTy5Mn23vX58+cxm800a9aMvLw81q5dW+c++vfvz6pVq7h48SJFRUV89tln9seKiopo3bo1paWlVRonPz8/ioqKqu0rPDycnJwcDh8+DMDf/vY3BgwYUO/jqUjDmZOTQ05ODtnZ2Xz55ZdV0nAClJeXU1hYyKBBg1ixYgX5+fkA9iHxijScwE2n4dyyZQvZ2dlV9gtX03COGTNG0nAKIWokDbaoUUpKCt9++629wbZYLERHRxMREcHDDz9Mnz596nx9TEwMDz74IBaLhSFDhlRJkfnSSy8RFxdHnz59iIiIsG9/6KGHeO2114iOjubIkSP27d7e3nzwwQeMGTOGqKgo3NzceOyxx+p1HBVpOCunAb02DefGjRuJiooiNjaW/fv3ExkZaU/DabFYePLJJwGYMmUKmzdvxmKxsH379jrTcJaVldG5c2dmz55dYxpOi8XCgw8+aH9NUlISxcXFMhwuhKiVyyX/uN1J8o870/XScEryDyFub7dl8g8hbjeShlMIUR8yJC6Eg82ePZujR4/St29fR4cihHBi0mALIYQQLkAabCfkbPMKhGPJ50EIAdJgOx1vb2/y8/Pln7QAjMY6Pz8fb2/JTifEnU4mnTmZ4OBgcnNzZS1pYeft7U1wcLCjwxBCOJg02E7Gw8OjyhKXQgghBMiQuBBCCOESpMEWQgghXIA02EIIIYQLcLqlSZVSZ4Cj9XhqEHD2FodzoySm+nHGmMA546orpvZaa6dOnF3P+uxq5e5IzhiXxFQ/14vpuvXZ6Rrs+lJK7XK2dZQlpvpxxpjAOeNyxpgamjMeozPGBM4Zl8RUPw0RkwyJCyGEEC5AGmwhhBDCBbhyg/2uowOogcRUP84YEzhnXM4YU0NzxmN0xpjAOeOSmOrnJ8fksuewhRBCiDuJK/ewhRBCiDuGyzXYSqnBSqlDSqnDSqnZDoqhnVJqo1Jqv1IqUyn1hG17oFLqS6VUlu1ngANiMymlvlFKfW6730Ep9bWtvP6hlPJ0QEz+SqmPlVIHlVIHlFK9HF1WSqlZtr9dhlJqmVLK2xFlpZR6Xyn1g1Iqo9K2GstGGebb4vtOKRVzq+O71aQ+Xzc2p6rPzliXbXE5vD43Rl12qQZbKWUC3gKGAF2AFKVUFweEUgb8h9a6CxAPTLfFMRtI01qHAWm2+43tCeBApft/BP6ste4I/Aj8ygExvQms01pHABZbfA4rK6VUW2AG0F1r/TPABDyEY8pqMTD4mm21lc0QIMx2exRY0Ajx3TJSn+vF2eqzU9VlcKr6vJhbXZe11i5zA3oB6yvdfwZ4xgni+hS4DzgEtLZtaw0cauQ4gm0fikHA54DCuFDfvabya6SYmgHZ2OZLVNrusLIC2gLHgUCMBDifA/c7qqyAECDjemUD/BVIqel5rniT+nzdOJyqPjtjXba9p9PU51tdl12qh83VP0yFXNs2h1FKhQDRwNdAK631KdtDp4FWjRzOG8DTgNV2vzlwTmtdZrvviPLqAJwBPrAN7S1USplxYFlprU8A84BjwCmgENiN48uqQm1l43Sf/5/I6Y5H6nOdnK4ug9PX5waty67WYDsVpZQv8E9gptb6fOXHtPG1qdGm4CulhgM/aK13N9Z71pM7EAMs0FpHAxe4ZsjMAWUVACRj/ANqA5ipPpTlFBq7bO5kUp+vy+nqMrhOfW6IsnG1BvsE0K7S/WDbtkanlPLAqNxLtdaf2DbnKaVa2x5vDfzQiCH1AZKUUjnAcoxhtDcBf6VURd5zR5RXLpCrtf7adv9jjErvyLK6F8jWWp/RWpcCn2CUn6PLqkJtZeM0n/8G4jTHI/W5XpyxLoNz1+cGrcuu1mDvBMJss/88MSYWrG7sIJRSClgEHNBav17podXABNvvEzDOhTUKrfUzWutgrXUIRrls0Fo/AmwERjsiJltcp4HjSqlw26ZEYD8OLCuMobN4pZSP7W9ZEZNDy6qS2spmNfBL2wzTeKCw0nCbK5L6XAtnrM9OWpfBuetzw9blxpwc0EAn9YcC/waOAM86KIa+GEMb3wF7bbehGOeY0oAsIBUIdFB8CcDntt9DgXTgMLAC8HJAPN2AXbbyWgUEOLqsgN8DB4EM4G+AlyPKCliGcd6tFKMH86vaygZj0tFbts/+PoxZsY3++Wrg45f6fP34nKY+O2NdtsXl8PrcGHVZVjoTQgghXICrDYkLIYQQdyRpsIUQQggXIA22EEII4QKkwRZCCCFcgDTYQgghhAuQBlsIIYRwAdJgCyGEEC5AGmwhhBDCBfw/ie2BjjFbpGwAAAAASUVORK5CYII=\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 576x576 with 2 Axes>"
       ]
@@ -2334,7 +2353,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 37,
+   "execution_count": 38,
    "metadata": {
     "id": "f2H8GEeevXZ_"
    },
@@ -2343,8 +2362,8 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "10000/10000 [==============================] - 1s 86us/sample - loss: 7.9333 - accuracy: 0.2262\n",
-      "Accuracy on test dataset: 0.2262\n"
+      "313/313 [==============================] - 4s 12ms/step - loss: 7.1838 - accuracy: 0.2212 0s - loss: 7.159\n",
+      "Accuracy on test dataset: 0.22120000422000885\n"
      ]
     }
    ],
@@ -2374,7 +2393,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 38,
+   "execution_count": 39,
    "metadata": {
     "id": "Mwgg1e0VvXaF"
    },
@@ -2383,207 +2402,206 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Train on 4000 samples\n",
       "Epoch 1/100\n",
-      "4000/4000 [==============================] - 2s 389us/sample - loss: 2.7780 - accuracy: 0.0833\n",
+      "125/125 [==============================] - 4s 29ms/step - loss: 2.7341 - accuracy: 0.0967 - lr: 1.0000e-06\n",
       "Epoch 2/100\n",
-      "4000/4000 [==============================] - 1s 237us/sample - loss: 2.7679 - accuracy: 0.0855\n",
+      "125/125 [==============================] - 3s 28ms/step - loss: 2.7231 - accuracy: 0.1023 - lr: 1.0798e-06\n",
       "Epoch 3/100\n",
-      "4000/4000 [==============================] - 1s 233us/sample - loss: 2.7571 - accuracy: 0.0862\n",
+      "125/125 [==============================] - 3s 28ms/step - loss: 2.7117 - accuracy: 0.0960 - lr: 1.1659e-06\n",
       "Epoch 4/100\n",
-      "4000/4000 [==============================] - 1s 237us/sample - loss: 2.7408 - accuracy: 0.0900\n",
+      "125/125 [==============================] - 3s 28ms/step - loss: 2.7018 - accuracy: 0.0997 - lr: 1.2589e-06\n",
       "Epoch 5/100\n",
-      "4000/4000 [==============================] - 1s 224us/sample - loss: 2.7360 - accuracy: 0.0880\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 2.6871 - accuracy: 0.1042 - lr: 1.3594e-06\n",
       "Epoch 6/100\n",
-      "4000/4000 [==============================] - 1s 222us/sample - loss: 2.7119 - accuracy: 0.0887\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 2.6773 - accuracy: 0.1007 - lr: 1.4678e-06\n",
       "Epoch 7/100\n",
-      "4000/4000 [==============================] - 1s 240us/sample - loss: 2.6955 - accuracy: 0.0993\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 2.6678 - accuracy: 0.1047 - lr: 1.5849e-06\n",
       "Epoch 8/100\n",
-      "4000/4000 [==============================] - 1s 241us/sample - loss: 2.6770 - accuracy: 0.0978\n",
+      "125/125 [==============================] - 3s 28ms/step - loss: 2.6484 - accuracy: 0.1067 - lr: 1.7113e-06\n",
       "Epoch 9/100\n",
-      "4000/4000 [==============================] - 1s 238us/sample - loss: 2.6724 - accuracy: 0.0988\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 2.6282 - accuracy: 0.1135 - lr: 1.8478e-06\n",
       "Epoch 10/100\n",
-      "4000/4000 [==============================] - 1s 223us/sample - loss: 2.6479 - accuracy: 0.1018\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 2.6128 - accuracy: 0.1070 - lr: 1.9953e-06\n",
       "Epoch 11/100\n",
-      "4000/4000 [==============================] - 1s 225us/sample - loss: 2.6333 - accuracy: 0.1047\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 2.5955 - accuracy: 0.1168 - lr: 2.1544e-06\n",
       "Epoch 12/100\n",
-      "4000/4000 [==============================] - 1s 229us/sample - loss: 2.6210 - accuracy: 0.1042\n",
+      "125/125 [==============================] - 4s 28ms/step - loss: 2.5668 - accuracy: 0.1185 - lr: 2.3263e-06\n",
       "Epoch 13/100\n",
-      "4000/4000 [==============================] - 1s 237us/sample - loss: 2.5956 - accuracy: 0.1130\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 2.5601 - accuracy: 0.1175 - lr: 2.5119e-06\n",
       "Epoch 14/100\n",
-      "4000/4000 [==============================] - 1s 237us/sample - loss: 2.5834 - accuracy: 0.1192\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 2.5297 - accuracy: 0.1222 - lr: 2.7123e-06\n",
       "Epoch 15/100\n",
-      "4000/4000 [==============================] - 1s 240us/sample - loss: 2.5585 - accuracy: 0.1203\n",
+      "125/125 [==============================] - 3s 28ms/step - loss: 2.5098 - accuracy: 0.1287 - lr: 2.9286e-063 - accuracy: 0.\n",
       "Epoch 16/100\n",
-      "4000/4000 [==============================] - 1s 242us/sample - loss: 2.5298 - accuracy: 0.1295\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 2.4952 - accuracy: 0.1320 - lr: 3.1623e-06\n",
       "Epoch 17/100\n",
-      "4000/4000 [==============================] - 1s 244us/sample - loss: 2.5137 - accuracy: 0.1277\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 2.4701 - accuracy: 0.1320 - lr: 3.4145e-06\n",
       "Epoch 18/100\n",
-      "4000/4000 [==============================] - 1s 232us/sample - loss: 2.4886 - accuracy: 0.1410\n",
+      "125/125 [==============================] - 3s 28ms/step - loss: 2.4431 - accuracy: 0.1440 - lr: 3.6869e-06\n",
       "Epoch 19/100\n",
-      "4000/4000 [==============================] - 1s 229us/sample - loss: 2.4633 - accuracy: 0.1458\n",
+      "125/125 [==============================] - 3s 28ms/step - loss: 2.4264 - accuracy: 0.1515 - lr: 3.9811e-06\n",
       "Epoch 20/100\n",
-      "4000/4000 [==============================] - 1s 228us/sample - loss: 2.4527 - accuracy: 0.1480\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 2.4027 - accuracy: 0.1480 - lr: 4.2987e-06\n",
       "Epoch 21/100\n",
-      "4000/4000 [==============================] - 1s 228us/sample - loss: 2.4356 - accuracy: 0.1545\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 2.3835 - accuracy: 0.1520 - lr: 4.6416e-06\n",
       "Epoch 22/100\n",
-      "4000/4000 [==============================] - 1s 218us/sample - loss: 2.4110 - accuracy: 0.1660\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 2.3644 - accuracy: 0.1593 - lr: 5.0119e-06\n",
       "Epoch 23/100\n",
-      "4000/4000 [==============================] - 1s 217us/sample - loss: 2.3872 - accuracy: 0.1665\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 2.3339 - accuracy: 0.1752 - lr: 5.4117e-06\n",
       "Epoch 24/100\n",
-      "4000/4000 [==============================] - 1s 221us/sample - loss: 2.3640 - accuracy: 0.1850\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 2.3023 - accuracy: 0.1822 - lr: 5.8434e-06\n",
       "Epoch 25/100\n",
-      "4000/4000 [==============================] - 1s 213us/sample - loss: 2.3401 - accuracy: 0.1905\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 2.2856 - accuracy: 0.1918 - lr: 6.3096e-06\n",
       "Epoch 26/100\n",
-      "4000/4000 [==============================] - 1s 223us/sample - loss: 2.3202 - accuracy: 0.1887\n",
+      "125/125 [==============================] - 3s 28ms/step - loss: 2.2573 - accuracy: 0.1980 - lr: 6.8129e-06\n",
       "Epoch 27/100\n",
-      "4000/4000 [==============================] - 1s 234us/sample - loss: 2.2908 - accuracy: 0.2075\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 2.2410 - accuracy: 0.2005 - lr: 7.3564e-06\n",
       "Epoch 28/100\n",
-      "4000/4000 [==============================] - 1s 230us/sample - loss: 2.2663 - accuracy: 0.2135\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 2.2158 - accuracy: 0.2177 - lr: 7.9433e-06\n",
       "Epoch 29/100\n",
-      "4000/4000 [==============================] - 1s 227us/sample - loss: 2.2405 - accuracy: 0.2177\n",
+      "125/125 [==============================] - 3s 28ms/step - loss: 2.1895 - accuracy: 0.2243 - lr: 8.5770e-06\n",
       "Epoch 30/100\n",
-      "4000/4000 [==============================] - 1s 231us/sample - loss: 2.2248 - accuracy: 0.2195\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 2.1788 - accuracy: 0.2303 - lr: 9.2612e-06\n",
       "Epoch 31/100\n",
-      "4000/4000 [==============================] - 1s 238us/sample - loss: 2.2063 - accuracy: 0.2265\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 2.1541 - accuracy: 0.2377 - lr: 1.0000e-05\n",
       "Epoch 32/100\n",
-      "4000/4000 [==============================] - 1s 238us/sample - loss: 2.1786 - accuracy: 0.2415\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 2.1251 - accuracy: 0.2520 - lr: 1.0798e-05\n",
       "Epoch 33/100\n",
-      "4000/4000 [==============================] - 1s 222us/sample - loss: 2.1478 - accuracy: 0.2545\n",
+      "125/125 [==============================] - 4s 29ms/step - loss: 2.1084 - accuracy: 0.2595 - lr: 1.1659e-05\n",
       "Epoch 34/100\n",
-      "4000/4000 [==============================] - 1s 235us/sample - loss: 2.1260 - accuracy: 0.2565\n",
+      "125/125 [==============================] - 4s 32ms/step - loss: 2.0844 - accuracy: 0.2607 - lr: 1.2589e-05\n",
       "Epoch 35/100\n",
-      "4000/4000 [==============================] - 1s 236us/sample - loss: 2.1038 - accuracy: 0.2632\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 2.0649 - accuracy: 0.2693 - lr: 1.3594e-05\n",
       "Epoch 36/100\n",
-      "4000/4000 [==============================] - 1s 239us/sample - loss: 2.0768 - accuracy: 0.2702\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 2.0467 - accuracy: 0.2842 - lr: 1.4678e-05\n",
       "Epoch 37/100\n",
-      "4000/4000 [==============================] - 1s 221us/sample - loss: 2.0551 - accuracy: 0.2755\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 2.0186 - accuracy: 0.2882 - lr: 1.5849e-05\n",
       "Epoch 38/100\n",
-      "4000/4000 [==============================] - 1s 225us/sample - loss: 2.0291 - accuracy: 0.2870\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 1.9956 - accuracy: 0.3060 - lr: 1.7113e-05\n",
       "Epoch 39/100\n",
-      "4000/4000 [==============================] - 1s 235us/sample - loss: 2.0061 - accuracy: 0.2980\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 1.9865 - accuracy: 0.2980 - lr: 1.8478e-05\n",
       "Epoch 40/100\n",
-      "4000/4000 [==============================] - 1s 228us/sample - loss: 1.9852 - accuracy: 0.3083\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 1.9583 - accuracy: 0.3108 - lr: 1.9953e-05\n",
       "Epoch 41/100\n",
-      "4000/4000 [==============================] - 1s 231us/sample - loss: 1.9538 - accuracy: 0.3130\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.9443 - accuracy: 0.3175 - lr: 2.1544e-05\n",
       "Epoch 42/100\n",
-      "4000/4000 [==============================] - 1s 237us/sample - loss: 1.9299 - accuracy: 0.3235\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 1.9211 - accuracy: 0.3310 - lr: 2.3263e-05\n",
       "Epoch 43/100\n",
-      "4000/4000 [==============================] - 1s 227us/sample - loss: 1.9079 - accuracy: 0.3392\n",
+      "125/125 [==============================] - 3s 28ms/step - loss: 1.8980 - accuracy: 0.3383 - lr: 2.5119e-05\n",
       "Epoch 44/100\n",
-      "4000/4000 [==============================] - 1s 231us/sample - loss: 1.8937 - accuracy: 0.3320\n",
+      "125/125 [==============================] - 3s 28ms/step - loss: 1.8754 - accuracy: 0.3440 - lr: 2.7123e-05\n",
       "Epoch 45/100\n",
-      "4000/4000 [==============================] - 1s 236us/sample - loss: 1.8620 - accuracy: 0.3553\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 1.8504 - accuracy: 0.3557 - lr: 2.9286e-052 - accu\n",
       "Epoch 46/100\n",
-      "4000/4000 [==============================] - 1s 233us/sample - loss: 1.8346 - accuracy: 0.3650\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 1.8324 - accuracy: 0.3557 - lr: 3.1623e-05\n",
       "Epoch 47/100\n",
-      "4000/4000 [==============================] - 1s 241us/sample - loss: 1.8134 - accuracy: 0.3702\n",
+      "125/125 [==============================] - 3s 28ms/step - loss: 1.8225 - accuracy: 0.3627 - lr: 3.4145e-05\n",
       "Epoch 48/100\n",
-      "4000/4000 [==============================] - 1s 228us/sample - loss: 1.7957 - accuracy: 0.3905\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.7926 - accuracy: 0.3828 - lr: 3.6869e-05\n",
       "Epoch 49/100\n",
-      "4000/4000 [==============================] - 1s 242us/sample - loss: 1.7616 - accuracy: 0.4008\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 1.7666 - accuracy: 0.3910 - lr: 3.9811e-053 - \n",
       "Epoch 50/100\n",
-      "4000/4000 [==============================] - 1s 239us/sample - loss: 1.7528 - accuracy: 0.3975\n",
+      "125/125 [==============================] - 3s 28ms/step - loss: 1.7480 - accuracy: 0.3988 - lr: 4.2987e-05\n",
       "Epoch 51/100\n",
-      "4000/4000 [==============================] - 1s 238us/sample - loss: 1.7231 - accuracy: 0.4087\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 1.7273 - accuracy: 0.4132 - lr: 4.6416e-05\n",
       "Epoch 52/100\n",
-      "4000/4000 [==============================] - 1s 233us/sample - loss: 1.7001 - accuracy: 0.4193\n",
+      "125/125 [==============================] - 3s 23ms/step - loss: 1.7036 - accuracy: 0.4223 - lr: 5.0119e-05\n",
       "Epoch 53/100\n",
-      "4000/4000 [==============================] - 1s 236us/sample - loss: 1.6721 - accuracy: 0.4375\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 1.6816 - accuracy: 0.4363 - lr: 5.4117e-05\n",
       "Epoch 54/100\n",
-      "4000/4000 [==============================] - 1s 219us/sample - loss: 1.6483 - accuracy: 0.4433\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 1.6528 - accuracy: 0.4335 - lr: 5.8434e-05\n",
       "Epoch 55/100\n",
-      "4000/4000 [==============================] - 1s 232us/sample - loss: 1.6249 - accuracy: 0.4532\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 1.6421 - accuracy: 0.4423 - lr: 6.3096e-05\n",
       "Epoch 56/100\n",
-      "4000/4000 [==============================] - 1s 240us/sample - loss: 1.5993 - accuracy: 0.4575\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.6144 - accuracy: 0.4530 - lr: 6.8129e-05\n",
       "Epoch 57/100\n",
-      "4000/4000 [==============================] - 1s 230us/sample - loss: 1.5796 - accuracy: 0.4803\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.5863 - accuracy: 0.4622 - lr: 7.3564e-05\n",
       "Epoch 58/100\n",
-      "4000/4000 [==============================] - 1s 226us/sample - loss: 1.5496 - accuracy: 0.4855\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.5684 - accuracy: 0.4782 - lr: 7.9433e-05\n",
       "Epoch 59/100\n",
-      "4000/4000 [==============================] - 1s 228us/sample - loss: 1.5305 - accuracy: 0.5035\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.5419 - accuracy: 0.4947 - lr: 8.5770e-05\n",
       "Epoch 60/100\n",
-      "4000/4000 [==============================] - 1s 242us/sample - loss: 1.5062 - accuracy: 0.5045\n",
+      "125/125 [==============================] - 3s 28ms/step - loss: 1.5241 - accuracy: 0.4955 - lr: 9.2612e-05\n",
       "Epoch 61/100\n",
-      "4000/4000 [==============================] - 1s 220us/sample - loss: 1.4774 - accuracy: 0.5157\n",
+      "125/125 [==============================] - 4s 28ms/step - loss: 1.4922 - accuracy: 0.5145 - lr: 1.0000e-04\n",
       "Epoch 62/100\n",
-      "4000/4000 [==============================] - 1s 221us/sample - loss: 1.4558 - accuracy: 0.5293\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.4605 - accuracy: 0.5250 - lr: 1.0798e-040 - ac\n",
       "Epoch 63/100\n",
-      "4000/4000 [==============================] - 1s 230us/sample - loss: 1.4167 - accuracy: 0.5470\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.4500 - accuracy: 0.5310 - lr: 1.1659e-04\n",
       "Epoch 64/100\n",
-      "4000/4000 [==============================] - 1s 235us/sample - loss: 1.4026 - accuracy: 0.5497\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 1.4160 - accuracy: 0.5418 - lr: 1.2589e-04\n",
       "Epoch 65/100\n",
-      "4000/4000 [==============================] - 1s 215us/sample - loss: 1.3733 - accuracy: 0.5580\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 1.3833 - accuracy: 0.5598 - lr: 1.3594e-04\n",
       "Epoch 66/100\n",
-      "4000/4000 [==============================] - 1s 223us/sample - loss: 1.3473 - accuracy: 0.5780\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 1.3639 - accuracy: 0.5645 - lr: 1.4678e-04\n",
       "Epoch 67/100\n",
-      "4000/4000 [==============================] - 1s 225us/sample - loss: 1.3170 - accuracy: 0.5863\n",
+      "125/125 [==============================] - 3s 28ms/step - loss: 1.3266 - accuracy: 0.5820 - lr: 1.5849e-04\n",
       "Epoch 68/100\n",
-      "4000/4000 [==============================] - 1s 222us/sample - loss: 1.2806 - accuracy: 0.6033\n",
+      "125/125 [==============================] - 3s 28ms/step - loss: 1.3012 - accuracy: 0.5905 - lr: 1.7113e-04\n",
       "Epoch 69/100\n",
-      "4000/4000 [==============================] - 1s 230us/sample - loss: 1.2494 - accuracy: 0.6145\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.2574 - accuracy: 0.6165 - lr: 1.8478e-04\n",
       "Epoch 70/100\n",
-      "4000/4000 [==============================] - 1s 238us/sample - loss: 1.2187 - accuracy: 0.6313\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 1.2467 - accuracy: 0.6192 - lr: 1.9953e-04\n",
       "Epoch 71/100\n",
-      "4000/4000 [==============================] - 1s 223us/sample - loss: 1.1796 - accuracy: 0.6465\n",
+      "125/125 [==============================] - 4s 31ms/step - loss: 1.2082 - accuracy: 0.6360 - lr: 2.1544e-04\n",
       "Epoch 72/100\n",
-      "4000/4000 [==============================] - 1s 239us/sample - loss: 1.1587 - accuracy: 0.6575\n",
+      "125/125 [==============================] - 3s 28ms/step - loss: 1.1601 - accuracy: 0.6610 - lr: 2.3263e-04\n",
       "Epoch 73/100\n",
-      "4000/4000 [==============================] - 1s 235us/sample - loss: 1.1041 - accuracy: 0.6845\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 1.1304 - accuracy: 0.6618 - lr: 2.5119e-04\n",
       "Epoch 74/100\n",
-      "4000/4000 [==============================] - 1s 234us/sample - loss: 1.0765 - accuracy: 0.6920\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 1.0996 - accuracy: 0.6795 - lr: 2.7123e-04\n",
       "Epoch 75/100\n",
-      "4000/4000 [==============================] - 1s 229us/sample - loss: 1.0372 - accuracy: 0.7143\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 1.0548 - accuracy: 0.7020 - lr: 2.9286e-04\n",
       "Epoch 76/100\n",
-      "4000/4000 [==============================] - 1s 234us/sample - loss: 0.9984 - accuracy: 0.7220\n",
+      "125/125 [==============================] - 3s 24ms/step - loss: 1.0141 - accuracy: 0.7207 - lr: 3.1623e-04\n",
       "Epoch 77/100\n",
-      "4000/4000 [==============================] - 1s 247us/sample - loss: 0.9581 - accuracy: 0.7467\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 0.9744 - accuracy: 0.7345 - lr: 3.4145e-049 - \n",
       "Epoch 78/100\n",
-      "4000/4000 [==============================] - 1s 226us/sample - loss: 0.9233 - accuracy: 0.7630\n",
+      "125/125 [==============================] - 3s 28ms/step - loss: 0.9272 - accuracy: 0.7520 - lr: 3.6869e-04\n",
       "Epoch 79/100\n",
-      "4000/4000 [==============================] - 1s 235us/sample - loss: 0.8757 - accuracy: 0.7755\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 0.8962 - accuracy: 0.7655 - lr: 3.9811e-04\n",
       "Epoch 80/100\n",
-      "4000/4000 [==============================] - 1s 240us/sample - loss: 0.8245 - accuracy: 0.7990\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 0.8426 - accuracy: 0.7918 - lr: 4.2987e-04\n",
       "Epoch 81/100\n",
-      "4000/4000 [==============================] - 1s 240us/sample - loss: 0.7824 - accuracy: 0.8165\n",
+      "125/125 [==============================] - 3s 28ms/step - loss: 0.8049 - accuracy: 0.7987 - lr: 4.6416e-04\n",
       "Epoch 82/100\n",
-      "4000/4000 [==============================] - 1s 239us/sample - loss: 0.7358 - accuracy: 0.8325\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 0.7616 - accuracy: 0.8205 - lr: 5.0119e-04\n",
       "Epoch 83/100\n",
-      "4000/4000 [==============================] - 1s 232us/sample - loss: 0.6827 - accuracy: 0.8570\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 0.7175 - accuracy: 0.8325 - lr: 5.4117e-04\n",
       "Epoch 84/100\n",
-      "4000/4000 [==============================] - 1s 226us/sample - loss: 0.6456 - accuracy: 0.8577\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 0.6672 - accuracy: 0.8510 - lr: 5.8434e-04\n",
       "Epoch 85/100\n",
-      "4000/4000 [==============================] - 1s 231us/sample - loss: 0.5995 - accuracy: 0.8777\n",
+      "125/125 [==============================] - 4s 28ms/step - loss: 0.6195 - accuracy: 0.8595 - lr: 6.3096e-04\n",
       "Epoch 86/100\n",
-      "4000/4000 [==============================] - 1s 244us/sample - loss: 0.5666 - accuracy: 0.8857\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 0.5772 - accuracy: 0.8842 - lr: 6.8129e-04\n",
       "Epoch 87/100\n",
-      "4000/4000 [==============================] - 1s 236us/sample - loss: 0.5197 - accuracy: 0.8997\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 0.5274 - accuracy: 0.8953 - lr: 7.3564e-04\n",
       "Epoch 88/100\n",
-      "4000/4000 [==============================] - 1s 225us/sample - loss: 0.4602 - accuracy: 0.9218\n",
+      "125/125 [==============================] - 4s 29ms/step - loss: 0.4971 - accuracy: 0.9005 - lr: 7.9433e-04\n",
       "Epoch 89/100\n",
-      "4000/4000 [==============================] - 1s 232us/sample - loss: 0.4264 - accuracy: 0.9262\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 0.4486 - accuracy: 0.9190 - lr: 8.5770e-04\n",
       "Epoch 90/100\n",
-      "4000/4000 [==============================] - 1s 239us/sample - loss: 0.3980 - accuracy: 0.9283\n",
+      "125/125 [==============================] - 3s 25ms/step - loss: 0.4084 - accuracy: 0.9287 - lr: 9.2612e-04\n",
       "Epoch 91/100\n",
-      "4000/4000 [==============================] - 1s 252us/sample - loss: 0.3819 - accuracy: 0.9317\n",
+      "125/125 [==============================] - 4s 28ms/step - loss: 0.3708 - accuracy: 0.9352 - lr: 0.0010\n",
       "Epoch 92/100\n",
-      "4000/4000 [==============================] - 1s 227us/sample - loss: 0.3279 - accuracy: 0.9470\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 0.3318 - accuracy: 0.9517 - lr: 0.0011\n",
       "Epoch 93/100\n",
-      "4000/4000 [==============================] - 1s 234us/sample - loss: 0.2987 - accuracy: 0.9563\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 0.3165 - accuracy: 0.9457 - lr: 0.0012\n",
       "Epoch 94/100\n",
-      "4000/4000 [==============================] - 1s 239us/sample - loss: 0.2846 - accuracy: 0.9550\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 0.3029 - accuracy: 0.9435 - lr: 0.0013\n",
       "Epoch 95/100\n",
-      "4000/4000 [==============================] - 1s 235us/sample - loss: 0.2763 - accuracy: 0.9457\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 0.2939 - accuracy: 0.9485 - lr: 0.0014.2888 - accuracy: 0.\n",
       "Epoch 96/100\n",
-      "4000/4000 [==============================] - 1s 251us/sample - loss: 0.2681 - accuracy: 0.9500\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 0.2818 - accuracy: 0.9448 - lr: 0.0015\n",
       "Epoch 97/100\n",
-      "4000/4000 [==============================] - 1s 235us/sample - loss: 0.2705 - accuracy: 0.9460\n",
+      "125/125 [==============================] - 3s 26ms/step - loss: 0.2832 - accuracy: 0.9400 - lr: 0.0016\n",
       "Epoch 98/100\n",
-      "4000/4000 [==============================] - 1s 230us/sample - loss: 0.2598 - accuracy: 0.9457\n",
+      "125/125 [==============================] - 3s 28ms/step - loss: 0.2942 - accuracy: 0.9337 - lr: 0.0017\n",
       "Epoch 99/100\n",
-      "4000/4000 [==============================] - 1s 231us/sample - loss: 0.2946 - accuracy: 0.9273\n",
+      "125/125 [==============================] - 3s 27ms/step - loss: 0.2673 - accuracy: 0.9380 - lr: 0.0018\n",
       "Epoch 100/100\n",
-      "4000/4000 [==============================] - 1s 223us/sample - loss: 0.3040 - accuracy: 0.9255\n"
+      "125/125 [==============================] - 3s 25ms/step - loss: 0.2838 - accuracy: 0.9280 - lr: 0.0020\n"
      ]
     }
    ],
@@ -2620,7 +2638,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 39,
+   "execution_count": 40,
    "metadata": {
     "id": "TI33T-Xefdgt"
    },
@@ -2631,13 +2649,13 @@
        "(1e-06, 0.001, 0.0, 20.0)"
       ]
      },
-     "execution_count": 39,
+     "execution_count": 40,
      "metadata": {},
      "output_type": "execute_result"
     },
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYYAAAEACAYAAAC3adEgAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/d3fzzAAAACXBIWXMAAAsTAAALEwEAmpwYAAAaEklEQVR4nO3de4yld33f8c/n3Oa+M7O7s+u9goMdU0pjJ4xMI0hkbsZYCCcEEbsRNQ3SBgRSkVo1tJFKRRSFqgpRiqO4W7CAKDFJAQc3OJhtVGqQCHjWMrDGGBvb4L14Z3bu13PmzHz7x/PM7DyzMzsz55y57O77JR2dc57n9zzPb/bRzGd/t3McEQIAYEFuuysAANhZCAYAQAbBAADIIBgAABkEAwAgg2AAAGSsGQy2j9j+v7Z/ZPsp2/823b7b9gnbz6bP3ascf29a5lnb9zb6BwAANJbXWsdg+4CkAxHxhO0OSScl/Yak90saiohP2v6YpO6I+P1lx+6W1CepV1Kkx74uIoYb/YMAABpjzRZDRJyLiCfS1+OSnpZ0SNJdkj6fFvu8krBY7u2STkTEUBoGJyTd0YB6AwA2yYbGGGy/UtIvS/qupP0RcS7d9bKk/SscckjSS0ven063AQB2qMJ6C9pul/RlSR+NiDHbi/siImzX9dkato9JOiZJbW1tr3v1q19dz+kA4Jpy8uTJCxHR04hzrSsYbBeVhMJfRcRX0s3nbR+IiHPpOET/CoeekXTbkveHJX1zpWtExHFJxyWpt7c3+vr61vUDAAAk2z9r1LnWMyvJkj4r6emI+NSSXQ9LWphldK+kr65w+KOSbrfdnc5auj3dBgDYodYzxvAGSe+T9GbbT6aPOyV9UtLbbD8r6a3pe9nutf0ZSYqIIUl/KOnx9PGJdBsAYIdac7rqdqArCQA2xvbJiOhtxLlY+QwAyCAYAAAZBAMAIINgAABkEAwAgAyCAQCQQTAAADIIBgBABsEAAMggGAAAGQQDACCDYAAAZBAMAIAMggEAkEEwAAAyCAYAQAbBAADIIBgAABmFtQrYfkDSOyX1R8Rr021/I+mmtEiXpJGIuGWFY1+UNC5pTlK1UV87BwDYPGsGg6TPSbpP0hcWNkTEby+8tv0nkkYvc/ybIuJCrRUEAGytNYMhIh6z/cqV9tm2pPdKenOD6wUA2Cb1jjH8mqTzEfHsKvtD0jdsn7R9rM5rAQC2wHq6ki7nHkkPXmb/GyPijO19kk7Y/nFEPLZSwTQ4jknS0aNH66wWAKBWNbcYbBckvVvS36xWJiLOpM/9kh6SdOtlyh6PiN6I6O3p6am1WgCAOtXTlfRWST+OiNMr7bTdZrtj4bWk2yWdquN6AIAtsGYw2H5Q0nck3WT7tO0PpLvu1rJuJNsHbT+Svt0v6du2vy/pe5K+FhFfb1zVAQCbYT2zku5ZZfv7V9h2VtKd6evnJd1cZ/0AAFuMlc8AgAyCAQCQQTAAADIIBgBABsEAAMggGAAAGQQDACCDYAAAZBAMAIAMggEAkEEwAAAyCAYAQAbBAADIIBgAABkEAwAgg2AAAGQQDACADIIBAJBBMAAAMtYMBtsP2O63fWrJtv9i+4ztJ9PHnasce4ftZ2w/Z/tjjaw4AGBzrKfF8DlJd6yw/U8j4pb08cjynbbzkv5c0jskvUbSPbZfU09lAQCbb81giIjHJA3VcO5bJT0XEc9HREXSFyXdVcN5AABbqJ4xho/Y/kHa1dS9wv5Dkl5a8v50um1Fto/Z7rPdNzAwUEe1AAD1qDUY/kLSqyTdIumcpD+ptyIRcTwieiOit6enp97TAQBqVFMwRMT5iJiLiHlJ/1NJt9FyZyQdWfL+cLoNALCD1RQMtg8sefubkk6tUOxxSTfavt52SdLdkh6u5XoAgK1TWKuA7Qcl3SZpr+3Tkj4u6Tbbt0gKSS9K+r207EFJn4mIOyOiavsjkh6VlJf0QEQ8tRk/BACgcRwR212HS/T29kZfX992VwMArhi2T0ZEbyPOxcpnAEAGwQAAyCAYAAAZBAMAIINgAABkEAwAgAyCAQCQQTAAADIIBgBABsEAAMggGAAAGQQDACCDYAAAZBAMAIAMggEAkEEwAAAyCAYAQAbBAADIWDMYbD9gu9/2qSXb/pvtH9v+ge2HbHetcuyLtn9o+0nbfFcnAFwB1tNi+JykO5ZtOyHptRHxS5J+Iuk/Xub4N0XELY36LlIAwOZaMxgi4jFJQ8u2fSMiqunbf5J0eBPqBgDYBo0YY/hdSf+wyr6Q9A3bJ20fu9xJbB+z3We7b2BgoAHVAgDUoq5gsP0HkqqS/mqVIm+MiF+R9A5JH7b966udKyKOR0RvRPT29PTUUy0AQB1qDgbb75f0Tkm/ExGxUpmIOJM+90t6SNKttV4PALA1agoG23dI+g+S3hURU6uUabPdsfBa0u2STq1UFgCwc6xnuuqDkr4j6Sbbp21/QNJ9kjoknUinot6flj1o+5H00P2Svm37+5K+J+lrEfH1TfkpAAANU1irQETcs8Lmz65S9qykO9PXz0u6ua7aAQC2HCufAQAZBAMAIINgAABkEAwAgAyCAQCQQTAAADIIBgBABsEAAMggGAAAGQQDACCDYAAAZBAMAIAMggEAkEEwAAAyCAYAQAbBAADIIBgAABkEAwAgY13BYPsB2/22Ty3Zttv2CdvPps/dqxx7b1rmWdv3NqriAIDNsd4Ww+ck3bFs28ck/WNE3CjpH9P3GbZ3S/q4pNdLulXSx1cLEADAzrCuYIiIxyQNLdt8l6TPp68/L+k3Vjj07ZJORMRQRAxLOqFLAwYAsIPUM8awPyLOpa9flrR/hTKHJL205P3pdNslbB+z3We7b2BgoI5qAQDq0ZDB54gISVHnOY5HRG9E9Pb09DSiWgCAGtQTDOdtH5Ck9Ll/hTJnJB1Z8v5wug0AsEPVEwwPS1qYZXSvpK+uUOZRSbfb7k4HnW9PtwEAdqj1Tld9UNJ3JN1k+7TtD0j6pKS32X5W0lvT97Lda/szkhQRQ5L+UNLj6eMT6TYAwA7lZHhgZ+nt7Y2+vr7trgYAXDFsn4yI3kaci5XPAIAMggEAkEEwAAAyCAYAQAbBAADIIBgAABkEAwAgg2AAAGQQDACADIIBAJBBMAAAMggGAEAGwQAAyCAYAAAZBAMAIINgAABkEAwAgAyCAQCQUXMw2L7J9pNLHmO2P7qszG22R5eU+c911xgAsKkKtR4YEc9IukWSbOclnZH00ApFvxUR76z1OgCArdWorqS3SPppRPysQecDAGyTRgXD3ZIeXGXfr9r+vu1/sP3PVzuB7WO2+2z3DQwMNKhaAICNqjsYbJckvUvS/1ph9xOSXhERN0v6tKS/W+08EXE8Inojorenp6feagEAatSIFsM7JD0REeeX74iIsYiYSF8/Iqloe28DrgkA2CSNCIZ7tEo3ku3rbDt9fWt6vcEGXBMAsElqnpUkSbbbJL1N0u8t2fZBSYqI+yW9R9KHbFclTUu6OyKinmsCADZXXcEQEZOS9izbdv+S1/dJuq+eawAAthYrnwEAGQQDACCDYAAAZBAMAIAMggEAkEEwAAAyCAYAQAbBAADIIBgAABkEAwAgg2AAAGQQDACADIIBAJBBMAAAMggGAEAGwQAAyCAYAAAZBAMAIKPuYLD9ou0f2n7Sdt8K+237v9t+zvYPbP9KvdcEAGyeur7zeYk3RcSFVfa9Q9KN6eP1kv4ifQYA7EBb0ZV0l6QvROKfJHXZPrAF1wUA1KARwRCSvmH7pO1jK+w/JOmlJe9Pp9sAADtQI7qS3hgRZ2zvk3TC9o8j4rGNniQNlWOSdPTo0QZUCwBQi7pbDBFxJn3ul/SQpFuXFTkj6ciS94fTbcvPczwieiOit6enp95qAQBqVFcw2G6z3bHwWtLtkk4tK/awpH+dzk76l5JGI+JcPdcFAGyeeruS9kt6yPbCuf46Ir5u+4OSFBH3S3pE0p2SnpM0Jenf1HlNAMAmqisYIuJ5STevsP3+Ja9D0ofruQ4AYOuw8hkAkEEwAAAyCAYAQAbBAADIIBgAABkEAwAgg2AAAGQQDACADIIBAJBBMAAAMggGAEAGwQAAyCAYAAAZBAMAIINgAABkEAwAgAyCAQCQQTAAADJq/mpP20ckfUHJ9z6HpOMR8WfLytwm6auSXkg3fSUiPrHWuX90dky3/tH/UamQU1Mhp6ZCXs3F5LmllFdrKa+2UkGtTXmV8jnlclYhZxVyOXW2FNTVWlJna1GdLcWkXCmvtqaC2pryairka/2RAeCaUM93Plcl/buIeMJ2h6STtk9ExI+WlftWRLxzIyfuai3qza/ep0p1XuXFx5xmZuc0MF7VZKWqyXJVU+U5VebmNR+h6nwoYu1zl/I5tTcnIVHM5ZTPWfmcVSrk1NlSVFdrSV0tRXU0J4HSkgZLSzENp2JercW8OluL6m4tqau1SNgAuKrUHAwRcU7SufT1uO2nJR2StDwYNuxgV4s++Vu/tOHjqnPzGpupamSqopHpWY1Oz2qqPHcxSCpzGp+paqI8q8nynGYXQmUuVJmb18jUrE4PT2tkqqKxmarm5teRNJKaCjk1Fy+GR3tzQd2tJXW2JOGxu62kPe3J8+7WUtp6ScKpvamg9qaCbG/45wWAzVBPi2GR7VdK+mVJ311h96/a/r6ks5L+fUQ81YhrrqSQzyV/fNtKDTlfpTqv6cqcpmarmq7MaWZ2XjPVOU1X5jQyNavhqcpiiMzMJi2a6dl5TczMajgNmaHJikanZy9f75zV1Zq0VrqXPTcX82l3WhI+7QuhUsprV0vSXdbdVlJbKU+4AGiIuoPBdrukL0v6aESMLdv9hKRXRMSE7Tsl/Z2kG1c5zzFJxyTp6NGj9VarIUqFXNLFpGJd56nOzWtoqqKhyeQxXZnTZGVOU+WqxmZm05CZ1chURcNTFb00NKUfnK5oZGpW5er8uq6xEC6daVh0pS2Wi++LS/aX1NlSUEdz0mXWUiRUAFzkWE/H/GoH20VJfy/p0Yj41DrKvyipNyIuXK5cb29v9PX11Vyvq0lE0s1Vrs5rpjKnibRLbKJc1dj0rEamFwIl6TobTZ+Hp5KWyuj0rMZnqpe9RiHntLurSXvTLq+uhfGWJWMp3a0ldbeWkjApJS0ZAgXYGWyfjIjeRpyrnllJlvRZSU+vFgq2r5N0PiLC9q1KpscO1nrNa5FtNRWS2VS7movaV8M5FsZeRtMQGZma1dhMEhjjM0mrZXiyogsTFQ1OlvWzwanFLrLLyees1tLFcZL25oJ2NRcXu/N2t5W0t72kve1NyaOjSV0tRbXS7QXsaPV0Jb1B0vsk/dD2k+m2/yTpqCRFxP2S3iPpQ7arkqYl3R31NFFQk+zYS9u6j5ubj8UwGZ6qaHgyaYlMlqtJV1ilqsly0nqZLFc1Ua5qeKqinw5MaGiyoqnK3IrnzeesXc2FxTGSXc1pF1drUbtbL4ZKZ0sxLZOU7WopqVRg6Q2w2eqZlfRtSZf9b19E3Cfpvlqvge2VT7uYah3Mn67MaXCyrAsTFQ2MlzU4UdbodNJaSbq5ku6wsZlZnR2d1mg6oH+5yWDtTYXFbq2FMFkIl13puMmu5oL2tjeppyNpqXS1FJXL0UIB1qshs5KAlbSU8jpcatXh7tZ1HzM/HxqbmdXgZEVj6RjJ2ExVo2kX2FD6vDCGcnZ0erHc7NzKiZLPORmAXxIkS8Okramg1uLC4smC9rSX1JMGS2dLkW4vXHMIBuwouZzTQe+NtVIiQuXq/GILZKGVcmEieYwsDM5Pz2pwoqIXLkwmoTM9e9kWSiFndbeVtKctGXjvblsIlKI60nGVhTGWjuaiejqatH8XgYIrG8GAq4JtNRfzai7mtW9Xs25Y5yj9QqBMVeY0PTunyXI1DZOLwTI8WdFgOtX4J+cnFsNnZnb1qcSlQk5dLcXFKc+lfE67Wora217SnrYm7WkvLWvBJOGzu72kDhY8YpsRDLimLQ2UBb+4v2Ndx5arc5osJ2GyMLtrYLys/vGy+sdmNDI1q9m5eVXm5lWpzmt0elbPvDyuwclBjUytvuixlM8trjvpaiklA+9pgCyEycIiyIWxll3NRbU3F5RnLAUNQDAANVqYRlzL4Pzs3Pzi2Egy82t2cQHk4GRlcVrxyHRFp4en9KOzyZqV1WZ6LWgt5dXdenGa8J72knY1F9WRBkdHc+GSNSrdrUUV8sz2wkUEA7ANivmc9rQ3aU9704aOW2h5LCxqHJ6qaHRqVuPlqsbTtSnDkxVdmKzo3OiMTp0d1fhM9bKBYktdLcn6k67WUmYq8f5dzTrY1ayDnS3q6WhSaylZ3NhSzDN1+CpGMABXkFIhp56OZMbURlTn5jVZmcu0UpK1KUkLZTBd3Dg6nQzc/3RgUsNTlcuumm8p5rWnPV0x31ZSdzq1uStdj7Kwkn5h6vDS7jrsbAQDcA0o5HPqbEk+Wv7IBo6bLFd1bnRGZ0emNThZTgbp08fodDKt+MJEWedGZ/T0uTENTlZW/XyvrtairtvVrP27mrWvI1kJ35OuiN+bThHe296krlZmdG03ggHAqtqaCrphX7tu2Ne+7mMWFjYutEIujFfUPz6jl8dm9PJoWefHZvTjl8c0OFFRdYW5wqV8Ttd1NutAZ7MOdiVdWB1NSffWrpaCjnS36oZ97Rue0oz1IxgANNR6FzbOpx+5MjBR1oXxsi5MVnQhndV1bnRaZ0em9b0XhpJPJJ69dIxkb3tJ1+9t0972psUV+j0dTTrQ2bIYKt20PmpCMADYFrl08WB3W2nNKcKzc/MaT7+E62eDU3quf0LP9U/ohcFJPduffDbX8FTlkm9xbC7mdLCzRQfSAfSDXS061JU8H+hq1nW7mtXWxJ/B5fgXAbDjFZd8EOQv9LTrTa++dAXj3HxocLKscyMzOjc6rTMjMzo3Mp2MkYxO67FnB9Q/Xr4kPNqbCtq3q0kHO1v0ij2t6aNN1+9t0yv2tF6TX91LMAC4KuRz1r6OZu3raNbNR7pWLDM7N6+X08H0s6PT6h8r6/xYMu5xemRaX/vhucziw5ylQ90tun5vuw53t+hwd9LiONydBMiettJV2VVFMAC4ZhTzOR3Z3aoju1cf/xidmtWLg5N6cXBSzw9M6oULyetTZ0Y1NFnJlG0r5XV0T5uu39uqV/W061U97fqFnjYd6mrR7is4NAgGAFiis7Wom1u7Vmx1TJarOjsyrZ8PTennQ1P62WDy/PS5cT361HnNLZllVcrntL+zSQd2tejI7tbFbqpDXS3a19Gsno4mtZR2ZjcVwQAA69TWVNCN+zt04wqD5ZXqvH4+NKmfDkwmYxtjM4vdVt9+bkBffqJ8yTHtTQUd7m7R0d0XxzZ+cX+Hbtrfoc7W+r5rvh4EAwA0QKmQ0w37OnTDvpVnWM3MzumloSmdHZ1R/9iMBibK6h8r6/TwlF64MKn/95OBzOLA63Y161X72rS/o1k9u5oWWxl70kH4PW0ldTQX1Vxs/HevEwwAsAWai/lVWxtSsq7j5bEZPXN+XM+8PK6fvDyu5y9M6rsXhjQwXlZlbuUV5bbUVmrsn3KCAQB2gFzOOpiusXjTTdnpuBGhkanZxa/KXfgU3slyVVPlqibKc/p4A+tSVzDYvkPSn0nKS/pMRHxy2f4mSV+Q9DpJg5J+OyJerOeaAHCtsS8uBlztS6gaGQw1f26u7bykP5f0DkmvkXSP7dcsK/YBScMRcYOkP5X0X2u9HgBga9Tzgeq3SnouIp6PiIqkL0q6a1mZuyR9Pn39JUlv8ZU6sRcArhH1dCUdkvTSkvenJb1+tTIRUbU9KmmPpAvLT2b7mKRj6duy7VN11K0ROiWN7oDzbeS4tcrWun8j2/dqhfu7DRp5/3bCvVurTC37dur9uxp/99Yqs9F9K227aY3rr19E1PSQ9B4l4woL798n6b5lZU5JOrzk/U8l7V3HuftqrVejHpKO74TzbeS4tcrWun8j23fCvWv0/dsJ926tMrXs26n372r83Wv0/dvse1dPV9IZKfOdH4fTbSuWsV1QknKDdVxzK/3vHXK+jRy3Vtla9290+07QyLrthHu3Vpla9u3U+3c1/u6tVWaj+zb13jlNmo0fmPyh/4mktygJgMcl/auIeGpJmQ9L+hcR8UHbd0t6d0S8dx3n7ouI3poqhm3Fvbuycf+uXI28dzWPMUQyZvARSY8qma76QEQ8ZfsTSpo0D0v6rKS/tP2cpCFJd6/z9MdrrRe2Hffuysb9u3I17N7V3GIAAFyd6hljAABchQgGAEAGwQAAyLiigsF2zvYf2f607Xu3uz7YGNu32f6W7ftt37bd9cHG2G6z3Wf7ndtdF2yM7X+W/t59yfaH1iq/ZcFg+wHb/ctXNNu+w/Yztp+z/bE1TnOXkvUSs0pWWmOLNOj+haQJSc3i/m2ZBt07Sfp9SX+7ObXEahpx/yLi6Yj4oKT3SnrDmtfcqllJtn9dyR+FL0TEa9NteSVrId6m5A/F45LuUTL99Y+XneJ308dwRPwP21+KiPdsSeXRqPt3ISLmbe+X9KmI+J2tqv+1rEH37mYlH2fTrOQ+/v3W1B6NuH8R0W/7XZI+JOkvI+KvL3fNLfs+hoh4zPYrl21e/CA+SbL9RUl3RcQfS7qkuWr7tKSFb+Oe28TqYplG3L8lhiU1bUpFcYkG/e7dJqlNyScpT9t+JCJW/uYYNFSjfvfStWUP2/6apJ0RDKtYzwfxLfUVSZ+2/WuSHtvMimFdNnT/bL9b0tsldUm6b1NrhrVs6N5FxB9Iku33K235bWrtsJaN/u7dJundSv5D9shaJ9/uYNiQiJhS8h0PuAJFxFeUhDuuUBHxue2uAzYuIr4p6ZvrLb/ds5LW80F82Lm4f1cu7t2VbVPv33YHw+OSbrR9ve2Sks9Senib64T14/5dubh3V7ZNvX9bOV31QUnfkXST7dO2PxARVUkLH8T3tKS/XfrprNg5uH9XLu7dlW077h8fogcAyNjuriQAwA5DMAAAMggGAEAGwQAAyCAYAAAZBAMAIINgAABkEAwAgAyCAQCQ8f8BRvGSTmB95lEAAAAASUVORK5CYII=\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -2655,224 +2673,222 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 40,
+   "execution_count": 41,
    "metadata": {
     "id": "56buB1kGceaB"
    },
    "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/opt/conda/lib/python3.7/site-packages/ipykernel_launcher.py:18: UserWarning: `Model.fit_generator` is deprecated and will be removed in a future version. Please use `Model.fit`, which supports generators.\n"
+     ]
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "WARNING:tensorflow:sample_weight modes were coerced from\n",
-      "  ...\n",
-      "    to  \n",
-      "  ['...']\n",
-      "WARNING:tensorflow:sample_weight modes were coerced from\n",
-      "  ...\n",
-      "    to  \n",
-      "  ['...']\n",
-      "Train for 63 steps, validate for 16 steps\n",
       "Epoch 1/100\n",
-      "63/63 [==============================] - 4s 61ms/step - loss: 2.2243 - accuracy: 0.2410 - val_loss: 3.7818 - val_accuracy: 0.1730\n",
+      "63/63 [==============================] - 6s 83ms/step - loss: 2.2542 - accuracy: 0.2598 - val_loss: 4.9940 - val_accuracy: 0.1770\n",
       "Epoch 2/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 1.9376 - accuracy: 0.3025 - val_loss: 2.2632 - val_accuracy: 0.2520\n",
+      "63/63 [==============================] - 5s 78ms/step - loss: 1.9629 - accuracy: 0.3025 - val_loss: 2.8735 - val_accuracy: 0.1970\n",
       "Epoch 3/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.9164 - accuracy: 0.3167 - val_loss: 2.2741 - val_accuracy: 0.2600\n",
+      "63/63 [==============================] - 5s 78ms/step - loss: 1.9255 - accuracy: 0.3198 - val_loss: 2.1162 - val_accuracy: 0.2780\n",
       "Epoch 4/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.8825 - accuracy: 0.3282 - val_loss: 2.1843 - val_accuracy: 0.2990\n",
+      "63/63 [==============================] - 5s 79ms/step - loss: 1.8948 - accuracy: 0.3192 - val_loss: 1.9912 - val_accuracy: 0.2830\n",
       "Epoch 5/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.8613 - accuracy: 0.3310 - val_loss: 1.8329 - val_accuracy: 0.3400\n",
+      "63/63 [==============================] - 6s 92ms/step - loss: 1.8634 - accuracy: 0.3410 - val_loss: 2.1961 - val_accuracy: 0.2630\n",
       "Epoch 6/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.8214 - accuracy: 0.3397 - val_loss: 1.9534 - val_accuracy: 0.3080\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 1.8725 - accuracy: 0.3250 - val_loss: 1.9157 - val_accuracy: 0.3180\n",
       "Epoch 7/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.8233 - accuracy: 0.3467 - val_loss: 2.0647 - val_accuracy: 0.2740\n",
+      "63/63 [==============================] - 5s 80ms/step - loss: 1.8330 - accuracy: 0.3363 - val_loss: 2.0854 - val_accuracy: 0.2960\n",
       "Epoch 8/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.8114 - accuracy: 0.3397 - val_loss: 1.8218 - val_accuracy: 0.3380\n",
+      "63/63 [==============================] - 5s 80ms/step - loss: 1.8108 - accuracy: 0.3473 - val_loss: 1.9207 - val_accuracy: 0.3270\n",
       "Epoch 9/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.7918 - accuracy: 0.3575 - val_loss: 1.9308 - val_accuracy: 0.2840\n",
+      "63/63 [==============================] - 5s 80ms/step - loss: 1.7873 - accuracy: 0.3625 - val_loss: 1.8341 - val_accuracy: 0.3500\n",
       "Epoch 10/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 1.7776 - accuracy: 0.3490 - val_loss: 2.2103 - val_accuracy: 0.2730\n",
+      "63/63 [==============================] - 5s 80ms/step - loss: 1.7856 - accuracy: 0.3535 - val_loss: 1.7954 - val_accuracy: 0.3380\n",
       "Epoch 11/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 1.7668 - accuracy: 0.3742 - val_loss: 1.8801 - val_accuracy: 0.3190\n",
+      "63/63 [==============================] - 5s 78ms/step - loss: 1.7792 - accuracy: 0.3602 - val_loss: 2.3248 - val_accuracy: 0.2330\n",
       "Epoch 12/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 1.7696 - accuracy: 0.3665 - val_loss: 1.9997 - val_accuracy: 0.3260\n",
+      "63/63 [==============================] - 5s 81ms/step - loss: 1.7624 - accuracy: 0.3758 - val_loss: 1.8582 - val_accuracy: 0.3520\n",
       "Epoch 13/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.7461 - accuracy: 0.3805 - val_loss: 1.7899 - val_accuracy: 0.3650\n",
+      "63/63 [==============================] - 5s 78ms/step - loss: 1.7585 - accuracy: 0.3735 - val_loss: 1.8053 - val_accuracy: 0.3600\n",
       "Epoch 14/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 1.7239 - accuracy: 0.3855 - val_loss: 2.0417 - val_accuracy: 0.3270\n",
+      "63/63 [==============================] - 5s 78ms/step - loss: 1.7528 - accuracy: 0.3710 - val_loss: 1.9468 - val_accuracy: 0.3100\n",
       "Epoch 15/100\n",
-      "63/63 [==============================] - 3s 47ms/step - loss: 1.7299 - accuracy: 0.3810 - val_loss: 1.8531 - val_accuracy: 0.3460\n",
+      "63/63 [==============================] - 5s 79ms/step - loss: 1.7250 - accuracy: 0.3755 - val_loss: 1.9362 - val_accuracy: 0.3190\n",
       "Epoch 16/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 1.7251 - accuracy: 0.3810 - val_loss: 1.9979 - val_accuracy: 0.3010\n",
+      "63/63 [==============================] - 5s 81ms/step - loss: 1.7323 - accuracy: 0.3677 - val_loss: 2.1602 - val_accuracy: 0.2850\n",
       "Epoch 17/100\n",
-      "63/63 [==============================] - 3s 48ms/step - loss: 1.7179 - accuracy: 0.3875 - val_loss: 1.8043 - val_accuracy: 0.3570\n",
+      "63/63 [==============================] - 5s 77ms/step - loss: 1.7348 - accuracy: 0.3780 - val_loss: 1.8932 - val_accuracy: 0.3490\n",
       "Epoch 18/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 1.6945 - accuracy: 0.3850 - val_loss: 1.8396 - val_accuracy: 0.3390\n",
+      "63/63 [==============================] - 5s 75ms/step - loss: 1.7111 - accuracy: 0.3913 - val_loss: 1.8995 - val_accuracy: 0.3230\n",
       "Epoch 19/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 1.6928 - accuracy: 0.3845 - val_loss: 1.8086 - val_accuracy: 0.3530\n",
+      "63/63 [==============================] - 5s 81ms/step - loss: 1.7216 - accuracy: 0.3857 - val_loss: 1.7782 - val_accuracy: 0.3610\n",
       "Epoch 20/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 1.6857 - accuracy: 0.3997 - val_loss: 1.9544 - val_accuracy: 0.3040\n",
+      "63/63 [==============================] - 5s 78ms/step - loss: 1.7007 - accuracy: 0.3842 - val_loss: 1.8050 - val_accuracy: 0.3580\n",
       "Epoch 21/100\n",
-      "63/63 [==============================] - 3s 48ms/step - loss: 1.6905 - accuracy: 0.3988 - val_loss: 1.8178 - val_accuracy: 0.3680\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 1.6877 - accuracy: 0.3963 - val_loss: 1.8479 - val_accuracy: 0.3460\n",
       "Epoch 22/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.6867 - accuracy: 0.3970 - val_loss: 1.7080 - val_accuracy: 0.3940\n",
+      "63/63 [==============================] - 5s 79ms/step - loss: 1.6840 - accuracy: 0.3988 - val_loss: 1.8476 - val_accuracy: 0.3380\n",
       "Epoch 23/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.6909 - accuracy: 0.3918 - val_loss: 2.0128 - val_accuracy: 0.2830\n",
+      "63/63 [==============================] - 5s 79ms/step - loss: 1.6892 - accuracy: 0.3950 - val_loss: 1.8558 - val_accuracy: 0.3540\n",
       "Epoch 24/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.6617 - accuracy: 0.4062 - val_loss: 1.6912 - val_accuracy: 0.3890\n",
+      "63/63 [==============================] - 5s 78ms/step - loss: 1.6756 - accuracy: 0.4095 - val_loss: 1.9144 - val_accuracy: 0.3250\n",
       "Epoch 25/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.6710 - accuracy: 0.4008 - val_loss: 1.7549 - val_accuracy: 0.3740\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 1.6849 - accuracy: 0.3960 - val_loss: 1.7634 - val_accuracy: 0.3690\n",
       "Epoch 26/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.6568 - accuracy: 0.4055 - val_loss: 1.8568 - val_accuracy: 0.3350\n",
+      "63/63 [==============================] - 5s 78ms/step - loss: 1.6741 - accuracy: 0.4072 - val_loss: 1.8039 - val_accuracy: 0.3450\n",
       "Epoch 27/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.6650 - accuracy: 0.3915 - val_loss: 1.7527 - val_accuracy: 0.3910\n",
+      "63/63 [==============================] - 5s 79ms/step - loss: 1.6633 - accuracy: 0.4050 - val_loss: 1.7362 - val_accuracy: 0.3850\n",
       "Epoch 28/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.6418 - accuracy: 0.4130 - val_loss: 1.7032 - val_accuracy: 0.3940\n",
+      "63/63 [==============================] - 5s 87ms/step - loss: 1.6675 - accuracy: 0.3915 - val_loss: 1.7573 - val_accuracy: 0.3850\n",
       "Epoch 29/100\n",
-      "63/63 [==============================] - 3s 48ms/step - loss: 1.6448 - accuracy: 0.4092 - val_loss: 1.7557 - val_accuracy: 0.3640\n",
+      "63/63 [==============================] - 4s 71ms/step - loss: 1.6411 - accuracy: 0.4047 - val_loss: 1.8315 - val_accuracy: 0.3630\n",
       "Epoch 30/100\n",
-      "63/63 [==============================] - 3s 48ms/step - loss: 1.6413 - accuracy: 0.4108 - val_loss: 1.7976 - val_accuracy: 0.3500\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 1.6543 - accuracy: 0.4065 - val_loss: 1.6689 - val_accuracy: 0.4140\n",
       "Epoch 31/100\n",
-      "63/63 [==============================] - 3s 48ms/step - loss: 1.6073 - accuracy: 0.4235 - val_loss: 1.9522 - val_accuracy: 0.3130\n",
+      "63/63 [==============================] - 5s 72ms/step - loss: 1.6376 - accuracy: 0.4195 - val_loss: 1.8304 - val_accuracy: 0.3600\n",
       "Epoch 32/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.6250 - accuracy: 0.4165 - val_loss: 1.9847 - val_accuracy: 0.3210\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 1.6322 - accuracy: 0.4135 - val_loss: 1.6550 - val_accuracy: 0.4090\n",
       "Epoch 33/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.6156 - accuracy: 0.4238 - val_loss: 1.9079 - val_accuracy: 0.3200\n",
+      "63/63 [==============================] - 5s 71ms/step - loss: 1.6259 - accuracy: 0.4165 - val_loss: 1.8499 - val_accuracy: 0.3470\n",
       "Epoch 34/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.6048 - accuracy: 0.4232 - val_loss: 1.7669 - val_accuracy: 0.3470\n",
+      "63/63 [==============================] - 5s 72ms/step - loss: 1.6312 - accuracy: 0.4227 - val_loss: 1.7315 - val_accuracy: 0.3900\n",
       "Epoch 35/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 1.5915 - accuracy: 0.4265 - val_loss: 1.7371 - val_accuracy: 0.3660\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 1.6230 - accuracy: 0.4165 - val_loss: 1.6394 - val_accuracy: 0.4170\n",
       "Epoch 36/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.6022 - accuracy: 0.4272 - val_loss: 1.8722 - val_accuracy: 0.3480\n",
+      "63/63 [==============================] - 4s 68ms/step - loss: 1.6310 - accuracy: 0.4123 - val_loss: 1.6955 - val_accuracy: 0.4020 - accuracy: 0.\n",
       "Epoch 37/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.5977 - accuracy: 0.4230 - val_loss: 1.8964 - val_accuracy: 0.3140\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 1.6011 - accuracy: 0.4227 - val_loss: 1.6961 - val_accuracy: 0.3850\n",
       "Epoch 38/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 1.6072 - accuracy: 0.4195 - val_loss: 1.7229 - val_accuracy: 0.3740\n",
+      "63/63 [==============================] - 4s 70ms/step - loss: 1.6102 - accuracy: 0.4198 - val_loss: 1.7158 - val_accuracy: 0.3780\n",
       "Epoch 39/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 1.5835 - accuracy: 0.4387 - val_loss: 1.7082 - val_accuracy: 0.3940\n",
+      "63/63 [==============================] - 5s 72ms/step - loss: 1.6313 - accuracy: 0.4103 - val_loss: 1.6456 - val_accuracy: 0.3980\n",
       "Epoch 40/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 1.5870 - accuracy: 0.4232 - val_loss: 1.8603 - val_accuracy: 0.3360\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 1.5908 - accuracy: 0.4227 - val_loss: 1.8407 - val_accuracy: 0.3590\n",
       "Epoch 41/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.5815 - accuracy: 0.4315 - val_loss: 1.6768 - val_accuracy: 0.3880\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 1.6066 - accuracy: 0.4300 - val_loss: 1.9521 - val_accuracy: 0.3190\n",
       "Epoch 42/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.5769 - accuracy: 0.4320 - val_loss: 1.7751 - val_accuracy: 0.3660\n",
+      "63/63 [==============================] - 5s 72ms/step - loss: 1.5782 - accuracy: 0.4408 - val_loss: 1.8885 - val_accuracy: 0.3470\n",
       "Epoch 43/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 1.5595 - accuracy: 0.4387 - val_loss: 1.7197 - val_accuracy: 0.3860\n",
+      "63/63 [==============================] - 5s 72ms/step - loss: 1.5978 - accuracy: 0.4280 - val_loss: 1.7635 - val_accuracy: 0.3650\n",
       "Epoch 44/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.5831 - accuracy: 0.4297 - val_loss: 1.9539 - val_accuracy: 0.3250\n",
+      "63/63 [==============================] - 5s 72ms/step - loss: 1.5855 - accuracy: 0.4290 - val_loss: 1.8491 - val_accuracy: 0.3240\n",
       "Epoch 45/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.5462 - accuracy: 0.4390 - val_loss: 1.7155 - val_accuracy: 0.3700\n",
+      "63/63 [==============================] - 5s 71ms/step - loss: 1.5950 - accuracy: 0.4240 - val_loss: 1.8440 - val_accuracy: 0.3330\n",
       "Epoch 46/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.5422 - accuracy: 0.4482 - val_loss: 1.6116 - val_accuracy: 0.4320\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 1.5726 - accuracy: 0.4310 - val_loss: 1.7390 - val_accuracy: 0.3790\n",
       "Epoch 47/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.5561 - accuracy: 0.4412 - val_loss: 1.9093 - val_accuracy: 0.3040\n",
+      "63/63 [==============================] - 4s 68ms/step - loss: 1.5843 - accuracy: 0.4415 - val_loss: 1.7202 - val_accuracy: 0.3730\n",
       "Epoch 48/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.5544 - accuracy: 0.4440 - val_loss: 1.6708 - val_accuracy: 0.3860\n",
+      "63/63 [==============================] - 5s 72ms/step - loss: 1.5732 - accuracy: 0.4227 - val_loss: 1.6525 - val_accuracy: 0.4120\n",
       "Epoch 49/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.5639 - accuracy: 0.4420 - val_loss: 1.8629 - val_accuracy: 0.3360\n",
+      "63/63 [==============================] - 5s 72ms/step - loss: 1.5691 - accuracy: 0.4360 - val_loss: 1.6301 - val_accuracy: 0.4300\n",
       "Epoch 50/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 1.5440 - accuracy: 0.4490 - val_loss: 1.8777 - val_accuracy: 0.3270\n",
+      "63/63 [==============================] - 5s 72ms/step - loss: 1.5723 - accuracy: 0.4365 - val_loss: 1.6796 - val_accuracy: 0.3970\n",
       "Epoch 51/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 1.5253 - accuracy: 0.4670 - val_loss: 1.7382 - val_accuracy: 0.3730\n",
+      "63/63 [==============================] - 4s 68ms/step - loss: 1.5554 - accuracy: 0.4450 - val_loss: 1.8034 - val_accuracy: 0.3440\n",
       "Epoch 52/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 1.5316 - accuracy: 0.4493 - val_loss: 1.7009 - val_accuracy: 0.3870\n",
+      "63/63 [==============================] - 4s 67ms/step - loss: 1.5702 - accuracy: 0.4317 - val_loss: 1.7770 - val_accuracy: 0.3750\n",
       "Epoch 53/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.5182 - accuracy: 0.4580 - val_loss: 1.9415 - val_accuracy: 0.3410\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 1.5511 - accuracy: 0.4395 - val_loss: 1.7265 - val_accuracy: 0.3730\n",
       "Epoch 54/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.5225 - accuracy: 0.4490 - val_loss: 1.7658 - val_accuracy: 0.3640\n",
+      "63/63 [==============================] - 5s 76ms/step - loss: 1.5601 - accuracy: 0.4295 - val_loss: 1.8740 - val_accuracy: 0.3330\n",
       "Epoch 55/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.5257 - accuracy: 0.4535 - val_loss: 1.6135 - val_accuracy: 0.4080\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 1.5442 - accuracy: 0.4465 - val_loss: 1.7168 - val_accuracy: 0.3680\n",
       "Epoch 56/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.5263 - accuracy: 0.4535 - val_loss: 1.7795 - val_accuracy: 0.3550\n",
+      "63/63 [==============================] - 4s 69ms/step - loss: 1.5656 - accuracy: 0.4442 - val_loss: 1.7108 - val_accuracy: 0.4040\n",
       "Epoch 57/100\n",
-      "63/63 [==============================] - 3s 48ms/step - loss: 1.5272 - accuracy: 0.4540 - val_loss: 1.8484 - val_accuracy: 0.3570\n",
+      "63/63 [==============================] - 5s 71ms/step - loss: 1.5292 - accuracy: 0.4618 - val_loss: 1.8155 - val_accuracy: 0.3510\n",
       "Epoch 58/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.5254 - accuracy: 0.4580 - val_loss: 1.7163 - val_accuracy: 0.3780\n",
+      "63/63 [==============================] - 4s 71ms/step - loss: 1.5433 - accuracy: 0.4512 - val_loss: 1.6529 - val_accuracy: 0.4000\n",
       "Epoch 59/100\n",
-      "63/63 [==============================] - 3s 48ms/step - loss: 1.5059 - accuracy: 0.4595 - val_loss: 1.7073 - val_accuracy: 0.3760\n",
+      "63/63 [==============================] - 4s 71ms/step - loss: 1.5463 - accuracy: 0.4372 - val_loss: 1.7713 - val_accuracy: 0.3760\n",
       "Epoch 60/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.5167 - accuracy: 0.4535 - val_loss: 1.7138 - val_accuracy: 0.3900\n",
+      "63/63 [==============================] - 5s 72ms/step - loss: 1.5269 - accuracy: 0.4532 - val_loss: 1.8304 - val_accuracy: 0.3500\n",
       "Epoch 61/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 1.4861 - accuracy: 0.4635 - val_loss: 1.7030 - val_accuracy: 0.3870\n",
+      "63/63 [==============================] - 5s 72ms/step - loss: 1.5375 - accuracy: 0.4467 - val_loss: 1.6323 - val_accuracy: 0.4320\n",
       "Epoch 62/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.4896 - accuracy: 0.4658 - val_loss: 1.7081 - val_accuracy: 0.3800\n",
+      "63/63 [==============================] - 5s 72ms/step - loss: 1.5391 - accuracy: 0.4470 - val_loss: 1.9411 - val_accuracy: 0.3150\n",
       "Epoch 63/100\n",
-      "63/63 [==============================] - 3s 48ms/step - loss: 1.4994 - accuracy: 0.4663 - val_loss: 1.6950 - val_accuracy: 0.3950\n",
+      "63/63 [==============================] - 5s 72ms/step - loss: 1.5302 - accuracy: 0.4532 - val_loss: 1.9294 - val_accuracy: 0.3500\n",
       "Epoch 64/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 1.4979 - accuracy: 0.4550 - val_loss: 1.6721 - val_accuracy: 0.3800\n",
+      "63/63 [==============================] - 5s 72ms/step - loss: 1.5218 - accuracy: 0.4552 - val_loss: 1.6804 - val_accuracy: 0.3910\n",
       "Epoch 65/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 1.4900 - accuracy: 0.4690 - val_loss: 1.9822 - val_accuracy: 0.3500\n",
+      "63/63 [==============================] - 5s 72ms/step - loss: 1.5208 - accuracy: 0.4543 - val_loss: 1.7296 - val_accuracy: 0.4080\n",
       "Epoch 66/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.4855 - accuracy: 0.4680 - val_loss: 1.6436 - val_accuracy: 0.4230\n",
+      "63/63 [==============================] - 4s 70ms/step - loss: 1.5332 - accuracy: 0.4535 - val_loss: 1.5908 - val_accuracy: 0.4440\n",
       "Epoch 67/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.4831 - accuracy: 0.4720 - val_loss: 1.8347 - val_accuracy: 0.3720\n",
+      "63/63 [==============================] - 5s 72ms/step - loss: 1.5037 - accuracy: 0.4540 - val_loss: 1.6887 - val_accuracy: 0.4250\n",
       "Epoch 68/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.4686 - accuracy: 0.4710 - val_loss: 1.8893 - val_accuracy: 0.3540\n",
+      "63/63 [==============================] - 5s 72ms/step - loss: 1.5166 - accuracy: 0.4502 - val_loss: 1.6828 - val_accuracy: 0.4060\n",
       "Epoch 69/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.5038 - accuracy: 0.4500 - val_loss: 1.9516 - val_accuracy: 0.3290\n",
+      "63/63 [==============================] - 5s 72ms/step - loss: 1.5191 - accuracy: 0.4535 - val_loss: 1.7525 - val_accuracy: 0.3870\n",
       "Epoch 70/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 1.4667 - accuracy: 0.4725 - val_loss: 1.6148 - val_accuracy: 0.4240\n",
+      "63/63 [==============================] - 4s 70ms/step - loss: 1.5154 - accuracy: 0.4535 - val_loss: 1.6676 - val_accuracy: 0.4090\n",
       "Epoch 71/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 1.4787 - accuracy: 0.4700 - val_loss: 1.6446 - val_accuracy: 0.4110\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 1.5056 - accuracy: 0.4615 - val_loss: 1.8401 - val_accuracy: 0.3630\n",
       "Epoch 72/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.4632 - accuracy: 0.4715 - val_loss: 1.7416 - val_accuracy: 0.3920\n",
+      "63/63 [==============================] - 5s 72ms/step - loss: 1.4962 - accuracy: 0.4563 - val_loss: 1.7432 - val_accuracy: 0.3920\n",
       "Epoch 73/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.4696 - accuracy: 0.4697 - val_loss: 1.5885 - val_accuracy: 0.4440\n",
+      "63/63 [==============================] - 4s 70ms/step - loss: 1.4923 - accuracy: 0.4620 - val_loss: 1.7786 - val_accuracy: 0.3890\n",
       "Epoch 74/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 1.4577 - accuracy: 0.4730 - val_loss: 1.7661 - val_accuracy: 0.3710\n",
+      "63/63 [==============================] - 4s 69ms/step - loss: 1.5093 - accuracy: 0.4557 - val_loss: 1.6356 - val_accuracy: 0.4230\n",
       "Epoch 75/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.4493 - accuracy: 0.4798 - val_loss: 2.0005 - val_accuracy: 0.3450\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 1.4964 - accuracy: 0.4563 - val_loss: 1.7209 - val_accuracy: 0.3780\n",
       "Epoch 76/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 1.4472 - accuracy: 0.4918 - val_loss: 1.6410 - val_accuracy: 0.4130\n",
+      "63/63 [==============================] - 5s 71ms/step - loss: 1.4922 - accuracy: 0.4667 - val_loss: 1.8775 - val_accuracy: 0.3690\n",
       "Epoch 77/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.4645 - accuracy: 0.4787 - val_loss: 1.7967 - val_accuracy: 0.3600\n",
+      "63/63 [==============================] - 5s 71ms/step - loss: 1.5045 - accuracy: 0.4487 - val_loss: 1.8763 - val_accuracy: 0.3370\n",
       "Epoch 78/100\n",
-      "63/63 [==============================] - 3s 48ms/step - loss: 1.4509 - accuracy: 0.4773 - val_loss: 1.6453 - val_accuracy: 0.4270\n",
+      "63/63 [==============================] - 4s 68ms/step - loss: 1.4811 - accuracy: 0.4782 - val_loss: 1.6409 - val_accuracy: 0.4320\n",
       "Epoch 79/100\n",
-      "63/63 [==============================] - 3s 48ms/step - loss: 1.4448 - accuracy: 0.4865 - val_loss: 1.7087 - val_accuracy: 0.3920\n",
+      "63/63 [==============================] - 4s 68ms/step - loss: 1.4890 - accuracy: 0.4667 - val_loss: 1.7575 - val_accuracy: 0.3660\n",
       "Epoch 80/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.4415 - accuracy: 0.4767 - val_loss: 2.1826 - val_accuracy: 0.3420\n",
+      "63/63 [==============================] - 4s 71ms/step - loss: 1.4737 - accuracy: 0.4602 - val_loss: 1.5995 - val_accuracy: 0.4320\n",
       "Epoch 81/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 1.4639 - accuracy: 0.4775 - val_loss: 1.7095 - val_accuracy: 0.3950\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 1.4876 - accuracy: 0.4615 - val_loss: 1.6411 - val_accuracy: 0.4120\n",
       "Epoch 82/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.4370 - accuracy: 0.4952 - val_loss: 1.7657 - val_accuracy: 0.3620\n",
+      "63/63 [==============================] - 5s 80ms/step - loss: 1.4904 - accuracy: 0.4645 - val_loss: 1.6765 - val_accuracy: 0.3890\n",
       "Epoch 83/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.4280 - accuracy: 0.4850 - val_loss: 1.6603 - val_accuracy: 0.4030\n",
+      "63/63 [==============================] - 5s 72ms/step - loss: 1.4744 - accuracy: 0.4712 - val_loss: 1.6622 - val_accuracy: 0.4140\n",
       "Epoch 84/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.4452 - accuracy: 0.4820 - val_loss: 1.6725 - val_accuracy: 0.4110\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 1.4704 - accuracy: 0.4645 - val_loss: 1.6428 - val_accuracy: 0.4290\n",
       "Epoch 85/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.4210 - accuracy: 0.4940 - val_loss: 1.7191 - val_accuracy: 0.4080\n",
+      "63/63 [==============================] - 4s 68ms/step - loss: 1.4688 - accuracy: 0.4737 - val_loss: 1.7446 - val_accuracy: 0.4040\n",
       "Epoch 86/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.4304 - accuracy: 0.4880 - val_loss: 1.6111 - val_accuracy: 0.4360\n",
+      "63/63 [==============================] - 4s 71ms/step - loss: 1.4813 - accuracy: 0.4798 - val_loss: 1.6792 - val_accuracy: 0.3770\n",
       "Epoch 87/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.4132 - accuracy: 0.4942 - val_loss: 1.7015 - val_accuracy: 0.3890\n",
+      "63/63 [==============================] - 5s 73ms/step - loss: 1.4582 - accuracy: 0.4723 - val_loss: 1.6408 - val_accuracy: 0.4060\n",
       "Epoch 88/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.4101 - accuracy: 0.4922 - val_loss: 1.7223 - val_accuracy: 0.3840\n",
+      "63/63 [==============================] - 4s 70ms/step - loss: 1.4584 - accuracy: 0.4795 - val_loss: 1.8675 - val_accuracy: 0.3500\n",
       "Epoch 89/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.4092 - accuracy: 0.4935 - val_loss: 1.6518 - val_accuracy: 0.4090\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 1.4595 - accuracy: 0.4692 - val_loss: 1.7522 - val_accuracy: 0.3830\n",
       "Epoch 90/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.4193 - accuracy: 0.4972 - val_loss: 1.6131 - val_accuracy: 0.4110\n",
+      "63/63 [==============================] - 6s 92ms/step - loss: 1.4523 - accuracy: 0.4735 - val_loss: 1.7044 - val_accuracy: 0.3960\n",
       "Epoch 91/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.4100 - accuracy: 0.4913 - val_loss: 1.6936 - val_accuracy: 0.4140\n",
+      "63/63 [==============================] - 5s 77ms/step - loss: 1.4731 - accuracy: 0.4730 - val_loss: 1.7716 - val_accuracy: 0.3750\n",
       "Epoch 92/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.4079 - accuracy: 0.5030 - val_loss: 1.6407 - val_accuracy: 0.4250\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 1.4691 - accuracy: 0.4733 - val_loss: 1.8304 - val_accuracy: 0.3660\n",
       "Epoch 93/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.3967 - accuracy: 0.5005 - val_loss: 1.5743 - val_accuracy: 0.4480\n",
+      "63/63 [==============================] - 5s 78ms/step - loss: 1.4652 - accuracy: 0.4720 - val_loss: 1.6646 - val_accuracy: 0.4140\n",
       "Epoch 94/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 1.3968 - accuracy: 0.4995 - val_loss: 1.5930 - val_accuracy: 0.4330\n",
+      "63/63 [==============================] - 5s 78ms/step - loss: 1.4573 - accuracy: 0.4770 - val_loss: 1.6571 - val_accuracy: 0.4120\n",
       "Epoch 95/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.3911 - accuracy: 0.4995 - val_loss: 1.5624 - val_accuracy: 0.4380\n",
+      "63/63 [==============================] - 5s 78ms/step - loss: 1.4467 - accuracy: 0.4770 - val_loss: 1.8122 - val_accuracy: 0.3780\n",
       "Epoch 96/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.3881 - accuracy: 0.4995 - val_loss: 1.6052 - val_accuracy: 0.4300\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 1.4353 - accuracy: 0.4835 - val_loss: 1.6656 - val_accuracy: 0.4270\n",
       "Epoch 97/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.3690 - accuracy: 0.5153 - val_loss: 1.7387 - val_accuracy: 0.3960\n",
+      "63/63 [==============================] - 5s 74ms/step - loss: 1.4453 - accuracy: 0.4755 - val_loss: 1.6765 - val_accuracy: 0.4040\n",
       "Epoch 98/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 1.4074 - accuracy: 0.4933 - val_loss: 1.6132 - val_accuracy: 0.4280\n",
+      "63/63 [==============================] - 5s 80ms/step - loss: 1.4494 - accuracy: 0.4818 - val_loss: 1.6135 - val_accuracy: 0.4310\n",
       "Epoch 99/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.3635 - accuracy: 0.5153 - val_loss: 1.6164 - val_accuracy: 0.4230\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 1.4442 - accuracy: 0.4832 - val_loss: 1.7504 - val_accuracy: 0.4080\n",
       "Epoch 100/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.3786 - accuracy: 0.5145 - val_loss: 1.5810 - val_accuracy: 0.4650\n"
+      "63/63 [==============================] - 5s 77ms/step - loss: 1.4388 - accuracy: 0.4793 - val_loss: 1.6360 - val_accuracy: 0.4080\n"
      ]
     }
    ],
@@ -2899,14 +2915,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 41,
+   "execution_count": 42,
    "metadata": {
     "id": "3RZxKhYnvXaI"
    },
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 576x576 with 2 Axes>"
       ]
@@ -2944,7 +2960,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 42,
+   "execution_count": 43,
    "metadata": {
     "id": "EWXjK2gTvXaM"
    },
@@ -2953,8 +2969,8 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "10000/10000 [==============================] - 1s 114us/sample - loss: 1.6241 - accuracy: 0.4324\n",
-      "Accuracy on test dataset: 0.4324\n"
+      "313/313 [==============================] - 4s 13ms/step - loss: 1.6613 - accuracy: 0.4116\n",
+      "Accuracy on test dataset: 0.4115999937057495\n"
      ]
     }
    ],
@@ -2985,7 +3001,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 43,
+   "execution_count": 44,
    "metadata": {
     "id": "H-WaGarpvXaR"
    },
@@ -2994,8 +3010,8 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "10000/10000 [==============================] - 1s 113us/sample - loss: 2.3026 - accuracy: 0.1000\n",
-      "Loss on test dataset: 2.3025851249694824\n"
+      "313/313 [==============================] - 4s 11ms/step - loss: 2.3026 - accuracy: 0.1000\n",
+      "Loss on test dataset: 2.30259108543396\n"
      ]
     }
    ],
@@ -3058,7 +3074,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 44,
+   "execution_count": 45,
    "metadata": {
     "id": "t0SeB0p8YVK9"
    },
@@ -3067,207 +3083,206 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Train on 20 samples\n",
       "Epoch 1/100\n",
-      "20/20 [==============================] - 1s 38ms/sample - loss: 2.3026 - accuracy: 0.0500\n",
+      "1/1 [==============================] - 1s 856ms/step - loss: 2.3026 - accuracy: 0.0500\n",
       "Epoch 2/100\n",
-      "20/20 [==============================] - 0s 389us/sample - loss: 2.3025 - accuracy: 0.1500\n",
+      "1/1 [==============================] - 0s 46ms/step - loss: 2.3025 - accuracy: 0.1500\n",
       "Epoch 3/100\n",
-      "20/20 [==============================] - 0s 345us/sample - loss: 2.3025 - accuracy: 0.1500\n",
+      "1/1 [==============================] - 0s 14ms/step - loss: 2.3025 - accuracy: 0.1500\n",
       "Epoch 4/100\n",
-      "20/20 [==============================] - 0s 341us/sample - loss: 2.3024 - accuracy: 0.1500\n",
+      "1/1 [==============================] - 0s 71ms/step - loss: 2.3024 - accuracy: 0.1500\n",
       "Epoch 5/100\n",
-      "20/20 [==============================] - 0s 373us/sample - loss: 2.3023 - accuracy: 0.1500\n",
+      "1/1 [==============================] - 0s 14ms/step - loss: 2.3023 - accuracy: 0.1500\n",
       "Epoch 6/100\n",
-      "20/20 [==============================] - 0s 386us/sample - loss: 2.3023 - accuracy: 0.1500\n",
+      "1/1 [==============================] - 0s 14ms/step - loss: 2.3023 - accuracy: 0.1500\n",
       "Epoch 7/100\n",
-      "20/20 [==============================] - 0s 362us/sample - loss: 2.3022 - accuracy: 0.1500\n",
+      "1/1 [==============================] - 0s 62ms/step - loss: 2.3022 - accuracy: 0.1500\n",
       "Epoch 8/100\n",
-      "20/20 [==============================] - 0s 347us/sample - loss: 2.3022 - accuracy: 0.1500\n",
+      "1/1 [==============================] - 0s 16ms/step - loss: 2.3022 - accuracy: 0.1500\n",
       "Epoch 9/100\n",
-      "20/20 [==============================] - 0s 353us/sample - loss: 2.3021 - accuracy: 0.1500\n",
+      "1/1 [==============================] - 0s 66ms/step - loss: 2.3021 - accuracy: 0.1500\n",
       "Epoch 10/100\n",
-      "20/20 [==============================] - 0s 340us/sample - loss: 2.3020 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 16ms/step - loss: 2.3020 - accuracy: 0.2000\n",
       "Epoch 11/100\n",
-      "20/20 [==============================] - 0s 374us/sample - loss: 2.3020 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 14ms/step - loss: 2.3020 - accuracy: 0.2000\n",
       "Epoch 12/100\n",
-      "20/20 [==============================] - 0s 408us/sample - loss: 2.3019 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 50ms/step - loss: 2.3019 - accuracy: 0.2000\n",
       "Epoch 13/100\n",
-      "20/20 [==============================] - 0s 438us/sample - loss: 2.3019 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 16ms/step - loss: 2.3019 - accuracy: 0.2000\n",
       "Epoch 14/100\n",
-      "20/20 [==============================] - 0s 463us/sample - loss: 2.3018 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 12ms/step - loss: 2.3018 - accuracy: 0.2000\n",
       "Epoch 15/100\n",
-      "20/20 [==============================] - 0s 411us/sample - loss: 2.3017 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 53ms/step - loss: 2.3017 - accuracy: 0.2000\n",
       "Epoch 16/100\n",
-      "20/20 [==============================] - 0s 396us/sample - loss: 2.3017 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 15ms/step - loss: 2.3017 - accuracy: 0.2000\n",
       "Epoch 17/100\n",
-      "20/20 [==============================] - 0s 402us/sample - loss: 2.3016 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 80ms/step - loss: 2.3016 - accuracy: 0.2000\n",
       "Epoch 18/100\n",
-      "20/20 [==============================] - 0s 422us/sample - loss: 2.3016 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 17ms/step - loss: 2.3016 - accuracy: 0.2000\n",
       "Epoch 19/100\n",
-      "20/20 [==============================] - 0s 417us/sample - loss: 2.3015 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 68ms/step - loss: 2.3015 - accuracy: 0.2000\n",
       "Epoch 20/100\n",
-      "20/20 [==============================] - 0s 487us/sample - loss: 2.3014 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 18ms/step - loss: 2.3014 - accuracy: 0.2000\n",
       "Epoch 21/100\n",
-      "20/20 [==============================] - 0s 498us/sample - loss: 2.3014 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 17ms/step - loss: 2.3014 - accuracy: 0.2000\n",
       "Epoch 22/100\n",
-      "20/20 [==============================] - 0s 410us/sample - loss: 2.3013 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 44ms/step - loss: 2.3013 - accuracy: 0.2000\n",
       "Epoch 23/100\n",
-      "20/20 [==============================] - 0s 449us/sample - loss: 2.3013 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 16ms/step - loss: 2.3013 - accuracy: 0.2000\n",
       "Epoch 24/100\n",
-      "20/20 [==============================] - 0s 451us/sample - loss: 2.3012 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 18ms/step - loss: 2.3012 - accuracy: 0.2000\n",
       "Epoch 25/100\n",
-      "20/20 [==============================] - 0s 410us/sample - loss: 2.3011 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 49ms/step - loss: 2.3011 - accuracy: 0.2000\n",
       "Epoch 26/100\n",
-      "20/20 [==============================] - 0s 442us/sample - loss: 2.3011 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 17ms/step - loss: 2.3011 - accuracy: 0.2000\n",
       "Epoch 27/100\n",
-      "20/20 [==============================] - 0s 457us/sample - loss: 2.3010 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 19ms/step - loss: 2.3010 - accuracy: 0.2000\n",
       "Epoch 28/100\n",
-      "20/20 [==============================] - 0s 426us/sample - loss: 2.3010 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 48ms/step - loss: 2.3010 - accuracy: 0.2000\n",
       "Epoch 29/100\n",
-      "20/20 [==============================] - 0s 475us/sample - loss: 2.3009 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 15ms/step - loss: 2.3009 - accuracy: 0.2000\n",
       "Epoch 30/100\n",
-      "20/20 [==============================] - 0s 430us/sample - loss: 2.3008 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 20ms/step - loss: 2.3008 - accuracy: 0.2000\n",
       "Epoch 31/100\n",
-      "20/20 [==============================] - 0s 466us/sample - loss: 2.3008 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 14ms/step - loss: 2.3008 - accuracy: 0.2000\n",
       "Epoch 32/100\n",
-      "20/20 [==============================] - 0s 473us/sample - loss: 2.3007 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 14ms/step - loss: 2.3007 - accuracy: 0.2000\n",
       "Epoch 33/100\n",
-      "20/20 [==============================] - 0s 446us/sample - loss: 2.3007 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 22ms/step - loss: 2.3007 - accuracy: 0.2000\n",
       "Epoch 34/100\n",
-      "20/20 [==============================] - 0s 376us/sample - loss: 2.3006 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 42ms/step - loss: 2.3006 - accuracy: 0.2000\n",
       "Epoch 35/100\n",
-      "20/20 [==============================] - 0s 397us/sample - loss: 2.3006 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 19ms/step - loss: 2.3006 - accuracy: 0.2000\n",
       "Epoch 36/100\n",
-      "20/20 [==============================] - 0s 422us/sample - loss: 2.3005 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 54ms/step - loss: 2.3005 - accuracy: 0.2000\n",
       "Epoch 37/100\n",
-      "20/20 [==============================] - 0s 397us/sample - loss: 2.3004 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 18ms/step - loss: 2.3004 - accuracy: 0.2000\n",
       "Epoch 38/100\n",
-      "20/20 [==============================] - 0s 382us/sample - loss: 2.3004 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 13ms/step - loss: 2.3004 - accuracy: 0.2000\n",
       "Epoch 39/100\n",
-      "20/20 [==============================] - 0s 384us/sample - loss: 2.3003 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 52ms/step - loss: 2.3003 - accuracy: 0.2000\n",
       "Epoch 40/100\n",
-      "20/20 [==============================] - 0s 361us/sample - loss: 2.3003 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 15ms/step - loss: 2.3003 - accuracy: 0.2000\n",
       "Epoch 41/100\n",
-      "20/20 [==============================] - 0s 407us/sample - loss: 2.3002 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 16ms/step - loss: 2.3002 - accuracy: 0.2000\n",
       "Epoch 42/100\n",
-      "20/20 [==============================] - 0s 418us/sample - loss: 2.3001 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 56ms/step - loss: 2.3001 - accuracy: 0.2000\n",
       "Epoch 43/100\n",
-      "20/20 [==============================] - 0s 439us/sample - loss: 2.3001 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 14ms/step - loss: 2.3001 - accuracy: 0.2000\n",
       "Epoch 44/100\n",
-      "20/20 [==============================] - 0s 405us/sample - loss: 2.3000 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 18ms/step - loss: 2.3000 - accuracy: 0.2000\n",
       "Epoch 45/100\n",
-      "20/20 [==============================] - 0s 486us/sample - loss: 2.3000 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 49ms/step - loss: 2.3000 - accuracy: 0.2000\n",
       "Epoch 46/100\n",
-      "20/20 [==============================] - 0s 529us/sample - loss: 2.2999 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 17ms/step - loss: 2.2999 - accuracy: 0.2000\n",
       "Epoch 47/100\n",
-      "20/20 [==============================] - 0s 435us/sample - loss: 2.2998 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 16ms/step - loss: 2.2998 - accuracy: 0.2000\n",
       "Epoch 48/100\n",
-      "20/20 [==============================] - 0s 403us/sample - loss: 2.2998 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 15ms/step - loss: 2.2998 - accuracy: 0.2000\n",
       "Epoch 49/100\n",
-      "20/20 [==============================] - 0s 448us/sample - loss: 2.2997 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 25ms/step - loss: 2.2997 - accuracy: 0.2000\n",
       "Epoch 50/100\n",
-      "20/20 [==============================] - 0s 414us/sample - loss: 2.2997 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 17ms/step - loss: 2.2997 - accuracy: 0.2000\n",
       "Epoch 51/100\n",
-      "20/20 [==============================] - 0s 436us/sample - loss: 2.2996 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 12ms/step - loss: 2.2996 - accuracy: 0.2000\n",
       "Epoch 52/100\n",
-      "20/20 [==============================] - 0s 389us/sample - loss: 2.2995 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 41ms/step - loss: 2.2995 - accuracy: 0.2000\n",
       "Epoch 53/100\n",
-      "20/20 [==============================] - 0s 479us/sample - loss: 2.2995 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 14ms/step - loss: 2.2995 - accuracy: 0.2000\n",
       "Epoch 54/100\n",
-      "20/20 [==============================] - 0s 499us/sample - loss: 2.2994 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 15ms/step - loss: 2.2994 - accuracy: 0.2000\n",
       "Epoch 55/100\n",
-      "20/20 [==============================] - 0s 419us/sample - loss: 2.2994 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 49ms/step - loss: 2.2994 - accuracy: 0.2000\n",
       "Epoch 56/100\n",
-      "20/20 [==============================] - 0s 375us/sample - loss: 2.2993 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 14ms/step - loss: 2.2993 - accuracy: 0.2000\n",
       "Epoch 57/100\n",
-      "20/20 [==============================] - 0s 397us/sample - loss: 2.2992 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 19ms/step - loss: 2.2992 - accuracy: 0.2000\n",
       "Epoch 58/100\n",
-      "20/20 [==============================] - 0s 340us/sample - loss: 2.2992 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 49ms/step - loss: 2.2992 - accuracy: 0.2000\n",
       "Epoch 59/100\n",
-      "20/20 [==============================] - 0s 407us/sample - loss: 2.2991 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 16ms/step - loss: 2.2991 - accuracy: 0.2000\n",
       "Epoch 60/100\n",
-      "20/20 [==============================] - 0s 485us/sample - loss: 2.2991 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 16ms/step - loss: 2.2991 - accuracy: 0.2000\n",
       "Epoch 61/100\n",
-      "20/20 [==============================] - 0s 502us/sample - loss: 2.2990 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 51ms/step - loss: 2.2990 - accuracy: 0.2000\n",
       "Epoch 62/100\n",
-      "20/20 [==============================] - 0s 458us/sample - loss: 2.2989 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 14ms/step - loss: 2.2989 - accuracy: 0.2000\n",
       "Epoch 63/100\n",
-      "20/20 [==============================] - 0s 506us/sample - loss: 2.2989 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 13ms/step - loss: 2.2989 - accuracy: 0.2000\n",
       "Epoch 64/100\n",
-      "20/20 [==============================] - 0s 431us/sample - loss: 2.2988 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 50ms/step - loss: 2.2988 - accuracy: 0.2000\n",
       "Epoch 65/100\n",
-      "20/20 [==============================] - 0s 389us/sample - loss: 2.2988 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 15ms/step - loss: 2.2988 - accuracy: 0.2000\n",
       "Epoch 66/100\n",
-      "20/20 [==============================] - 0s 636us/sample - loss: 2.2987 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 15ms/step - loss: 2.2987 - accuracy: 0.2000\n",
       "Epoch 67/100\n",
-      "20/20 [==============================] - 0s 414us/sample - loss: 2.2986 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 51ms/step - loss: 2.2986 - accuracy: 0.2000\n",
       "Epoch 68/100\n",
-      "20/20 [==============================] - 0s 436us/sample - loss: 2.2986 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 19ms/step - loss: 2.2986 - accuracy: 0.2000\n",
       "Epoch 69/100\n",
-      "20/20 [==============================] - 0s 560us/sample - loss: 2.2985 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 14ms/step - loss: 2.2985 - accuracy: 0.2000\n",
       "Epoch 70/100\n",
-      "20/20 [==============================] - 0s 438us/sample - loss: 2.2985 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 51ms/step - loss: 2.2985 - accuracy: 0.2000\n",
       "Epoch 71/100\n",
-      "20/20 [==============================] - 0s 438us/sample - loss: 2.2984 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 15ms/step - loss: 2.2984 - accuracy: 0.2000\n",
       "Epoch 72/100\n",
-      "20/20 [==============================] - 0s 419us/sample - loss: 2.2984 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 14ms/step - loss: 2.2984 - accuracy: 0.2000\n",
       "Epoch 73/100\n",
-      "20/20 [==============================] - 0s 353us/sample - loss: 2.2983 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 52ms/step - loss: 2.2983 - accuracy: 0.2000\n",
       "Epoch 74/100\n",
-      "20/20 [==============================] - 0s 452us/sample - loss: 2.2982 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 14ms/step - loss: 2.2982 - accuracy: 0.2000\n",
       "Epoch 75/100\n",
-      "20/20 [==============================] - 0s 440us/sample - loss: 2.2982 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 13ms/step - loss: 2.2982 - accuracy: 0.2000\n",
       "Epoch 76/100\n",
-      "20/20 [==============================] - 0s 452us/sample - loss: 2.2981 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 13ms/step - loss: 2.2981 - accuracy: 0.2000\n",
       "Epoch 77/100\n",
-      "20/20 [==============================] - 0s 446us/sample - loss: 2.2981 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 18ms/step - loss: 2.2981 - accuracy: 0.2000\n",
       "Epoch 78/100\n",
-      "20/20 [==============================] - 0s 374us/sample - loss: 2.2980 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 13ms/step - loss: 2.2980 - accuracy: 0.2000\n",
       "Epoch 79/100\n",
-      "20/20 [==============================] - 0s 391us/sample - loss: 2.2979 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 16ms/step - loss: 2.2979 - accuracy: 0.2000\n",
       "Epoch 80/100\n",
-      "20/20 [==============================] - 0s 435us/sample - loss: 2.2979 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 18ms/step - loss: 2.2979 - accuracy: 0.2000\n",
       "Epoch 81/100\n",
-      "20/20 [==============================] - 0s 440us/sample - loss: 2.2978 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 20ms/step - loss: 2.2978 - accuracy: 0.2000\n",
       "Epoch 82/100\n",
-      "20/20 [==============================] - 0s 490us/sample - loss: 2.2978 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 17ms/step - loss: 2.2978 - accuracy: 0.2000\n",
       "Epoch 83/100\n",
-      "20/20 [==============================] - 0s 419us/sample - loss: 2.2977 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 39ms/step - loss: 2.2977 - accuracy: 0.2000\n",
       "Epoch 84/100\n",
-      "20/20 [==============================] - 0s 437us/sample - loss: 2.2976 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 16ms/step - loss: 2.2976 - accuracy: 0.2000\n",
       "Epoch 85/100\n",
-      "20/20 [==============================] - 0s 365us/sample - loss: 2.2976 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 15ms/step - loss: 2.2976 - accuracy: 0.2000\n",
       "Epoch 86/100\n",
-      "20/20 [==============================] - 0s 393us/sample - loss: 2.2975 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 50ms/step - loss: 2.2975 - accuracy: 0.2000\n",
       "Epoch 87/100\n",
-      "20/20 [==============================] - 0s 361us/sample - loss: 2.2975 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 16ms/step - loss: 2.2975 - accuracy: 0.2000\n",
       "Epoch 88/100\n",
-      "20/20 [==============================] - 0s 438us/sample - loss: 2.2974 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 15ms/step - loss: 2.2974 - accuracy: 0.2000\n",
       "Epoch 89/100\n",
-      "20/20 [==============================] - 0s 403us/sample - loss: 2.2974 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 47ms/step - loss: 2.2974 - accuracy: 0.2000\n",
       "Epoch 90/100\n",
-      "20/20 [==============================] - 0s 403us/sample - loss: 2.2973 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 14ms/step - loss: 2.2973 - accuracy: 0.2000\n",
       "Epoch 91/100\n",
-      "20/20 [==============================] - 0s 444us/sample - loss: 2.2972 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 17ms/step - loss: 2.2972 - accuracy: 0.2000\n",
       "Epoch 92/100\n",
-      "20/20 [==============================] - 0s 407us/sample - loss: 2.2972 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 48ms/step - loss: 2.2972 - accuracy: 0.2000\n",
       "Epoch 93/100\n",
-      "20/20 [==============================] - 0s 425us/sample - loss: 2.2971 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 15ms/step - loss: 2.2971 - accuracy: 0.2000\n",
       "Epoch 94/100\n",
-      "20/20 [==============================] - 0s 394us/sample - loss: 2.2971 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 15ms/step - loss: 2.2971 - accuracy: 0.2000\n",
       "Epoch 95/100\n",
-      "20/20 [==============================] - 0s 426us/sample - loss: 2.2970 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 56ms/step - loss: 2.2970 - accuracy: 0.2000\n",
       "Epoch 96/100\n",
-      "20/20 [==============================] - 0s 451us/sample - loss: 2.2969 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 15ms/step - loss: 2.2969 - accuracy: 0.2000\n",
       "Epoch 97/100\n",
-      "20/20 [==============================] - 0s 413us/sample - loss: 2.2969 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 17ms/step - loss: 2.2969 - accuracy: 0.2000\n",
       "Epoch 98/100\n",
-      "20/20 [==============================] - 0s 382us/sample - loss: 2.2968 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 50ms/step - loss: 2.2968 - accuracy: 0.2000\n",
       "Epoch 99/100\n",
-      "20/20 [==============================] - 0s 421us/sample - loss: 2.2968 - accuracy: 0.2000\n",
+      "1/1 [==============================] - 0s 14ms/step - loss: 2.2968 - accuracy: 0.2000\n",
       "Epoch 100/100\n",
-      "20/20 [==============================] - 0s 408us/sample - loss: 2.2967 - accuracy: 0.2000\n"
+      "1/1 [==============================] - 0s 71ms/step - loss: 2.2967 - accuracy: 0.2000\n"
      ]
     }
    ],
@@ -3299,224 +3314,222 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 45,
+   "execution_count": 46,
    "metadata": {
     "id": "vhDjE0kQvXaX"
    },
    "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/opt/conda/lib/python3.7/site-packages/ipykernel_launcher.py:36: UserWarning: `Model.fit_generator` is deprecated and will be removed in a future version. Please use `Model.fit`, which supports generators.\n"
+     ]
+    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "WARNING:tensorflow:sample_weight modes were coerced from\n",
-      "  ...\n",
-      "    to  \n",
-      "  ['...']\n",
-      "WARNING:tensorflow:sample_weight modes were coerced from\n",
-      "  ...\n",
-      "    to  \n",
-      "  ['...']\n",
-      "Train for 63 steps, validate for 16 steps\n",
       "Epoch 1/100\n",
-      "63/63 [==============================] - 4s 68ms/step - loss: 12.9742 - accuracy: 0.2227 - val_loss: 13.4224 - val_accuracy: 0.1760\n",
+      "63/63 [==============================] - 7s 96ms/step - loss: 13.0181 - accuracy: 0.2160 - val_loss: 13.2011 - val_accuracy: 0.1950\n",
       "Epoch 2/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 12.0929 - accuracy: 0.2780 - val_loss: 11.8874 - val_accuracy: 0.2680\n",
+      "63/63 [==============================] - 6s 94ms/step - loss: 12.1518 - accuracy: 0.2840 - val_loss: 12.0171 - val_accuracy: 0.2470\n",
       "Epoch 3/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 11.3412 - accuracy: 0.3070 - val_loss: 11.0116 - val_accuracy: 0.2890\n",
+      "63/63 [==============================] - 6s 97ms/step - loss: 11.4058 - accuracy: 0.3178 - val_loss: 11.2302 - val_accuracy: 0.2820\n",
       "Epoch 4/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 10.6504 - accuracy: 0.3250 - val_loss: 10.3778 - val_accuracy: 0.3000\n",
+      "63/63 [==============================] - 6s 99ms/step - loss: 10.7203 - accuracy: 0.3207 - val_loss: 10.8606 - val_accuracy: 0.2100\n",
       "Epoch 5/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 10.0058 - accuracy: 0.3237 - val_loss: 9.7936 - val_accuracy: 0.2950\n",
+      "63/63 [==============================] - 6s 88ms/step - loss: 10.0897 - accuracy: 0.3298 - val_loss: 9.8507 - val_accuracy: 0.3080\n",
       "Epoch 6/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 9.3829 - accuracy: 0.3505 - val_loss: 9.2074 - val_accuracy: 0.3050\n",
+      "63/63 [==============================] - 6s 92ms/step - loss: 9.4855 - accuracy: 0.3495 - val_loss: 9.2743 - val_accuracy: 0.3090\n",
       "Epoch 7/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 8.8137 - accuracy: 0.3557 - val_loss: 8.5589 - val_accuracy: 0.3340\n",
+      "63/63 [==============================] - 6s 90ms/step - loss: 8.9464 - accuracy: 0.3347 - val_loss: 8.7243 - val_accuracy: 0.3240\n",
       "Epoch 8/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 8.3020 - accuracy: 0.3543 - val_loss: 8.1501 - val_accuracy: 0.3180\n",
+      "63/63 [==============================] - 5s 86ms/step - loss: 8.4246 - accuracy: 0.3483 - val_loss: 8.3198 - val_accuracy: 0.3170\n",
       "Epoch 9/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 7.8181 - accuracy: 0.3645 - val_loss: 7.6620 - val_accuracy: 0.3200\n",
+      "63/63 [==============================] - 6s 101ms/step - loss: 7.9513 - accuracy: 0.3485 - val_loss: 7.8701 - val_accuracy: 0.3100\n",
       "Epoch 10/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 7.3544 - accuracy: 0.3680 - val_loss: 7.4278 - val_accuracy: 0.2870\n",
+      "63/63 [==============================] - 5s 87ms/step - loss: 7.4930 - accuracy: 0.3705 - val_loss: 7.3292 - val_accuracy: 0.3430\n",
       "Epoch 11/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 6.9695 - accuracy: 0.3625 - val_loss: 6.8906 - val_accuracy: 0.3030\n",
+      "63/63 [==============================] - 6s 95ms/step - loss: 7.0981 - accuracy: 0.3647 - val_loss: 7.0211 - val_accuracy: 0.3320\n",
       "Epoch 12/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 6.5806 - accuracy: 0.3697 - val_loss: 6.5827 - val_accuracy: 0.3140\n",
+      "63/63 [==============================] - 6s 88ms/step - loss: 6.7173 - accuracy: 0.3765 - val_loss: 6.7382 - val_accuracy: 0.3330\n",
       "Epoch 13/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 6.2234 - accuracy: 0.3690 - val_loss: 6.2195 - val_accuracy: 0.3120\n",
+      "63/63 [==============================] - 6s 89ms/step - loss: 6.3762 - accuracy: 0.3663 - val_loss: 6.3951 - val_accuracy: 0.2830\n",
       "Epoch 14/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 5.9127 - accuracy: 0.3695 - val_loss: 5.8937 - val_accuracy: 0.3440\n",
+      "63/63 [==============================] - 6s 89ms/step - loss: 6.0660 - accuracy: 0.3645 - val_loss: 6.0316 - val_accuracy: 0.3210\n",
       "Epoch 15/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 5.6340 - accuracy: 0.3728 - val_loss: 5.6059 - val_accuracy: 0.3340\n",
+      "63/63 [==============================] - 5s 87ms/step - loss: 5.7584 - accuracy: 0.3750 - val_loss: 5.9498 - val_accuracy: 0.2510\n",
       "Epoch 16/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 5.3555 - accuracy: 0.3808 - val_loss: 5.5691 - val_accuracy: 0.2800\n",
+      "63/63 [==============================] - 6s 94ms/step - loss: 5.4748 - accuracy: 0.3808 - val_loss: 5.4712 - val_accuracy: 0.3390\n",
       "Epoch 17/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 5.1117 - accuracy: 0.3745 - val_loss: 5.2532 - val_accuracy: 0.2930\n",
+      "63/63 [==============================] - 6s 90ms/step - loss: 5.2459 - accuracy: 0.3745 - val_loss: 5.2130 - val_accuracy: 0.3630\n",
       "Epoch 18/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 4.8789 - accuracy: 0.3887 - val_loss: 4.9243 - val_accuracy: 0.3160\n",
+      "63/63 [==============================] - 6s 90ms/step - loss: 5.0009 - accuracy: 0.3790 - val_loss: 5.0301 - val_accuracy: 0.3470\n",
       "Epoch 19/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 4.6629 - accuracy: 0.3817 - val_loss: 4.9759 - val_accuracy: 0.2380\n",
+      "63/63 [==============================] - 6s 93ms/step - loss: 4.7974 - accuracy: 0.3778 - val_loss: 5.0356 - val_accuracy: 0.3010\n",
       "Epoch 20/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 4.4783 - accuracy: 0.3860 - val_loss: 4.5509 - val_accuracy: 0.3150\n",
+      "63/63 [==============================] - 6s 87ms/step - loss: 4.5918 - accuracy: 0.3850 - val_loss: 4.8138 - val_accuracy: 0.2510\n",
       "Epoch 21/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 4.2929 - accuracy: 0.3975 - val_loss: 4.4422 - val_accuracy: 0.3180\n",
+      "63/63 [==============================] - 6s 93ms/step - loss: 4.4131 - accuracy: 0.3860 - val_loss: 4.5613 - val_accuracy: 0.3380\n",
       "Epoch 22/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 4.1548 - accuracy: 0.3795 - val_loss: 4.2272 - val_accuracy: 0.3280\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 4.2511 - accuracy: 0.3877 - val_loss: 4.2133 - val_accuracy: 0.3590\n",
       "Epoch 23/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 3.9952 - accuracy: 0.3860 - val_loss: 4.0448 - val_accuracy: 0.3470\n",
+      "63/63 [==============================] - 6s 93ms/step - loss: 4.0962 - accuracy: 0.3840 - val_loss: 4.3263 - val_accuracy: 0.3050\n",
       "Epoch 24/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 3.8525 - accuracy: 0.3993 - val_loss: 3.8363 - val_accuracy: 0.3770\n",
+      "63/63 [==============================] - 6s 89ms/step - loss: 3.9479 - accuracy: 0.3837 - val_loss: 4.1604 - val_accuracy: 0.3210\n",
       "Epoch 25/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 3.7381 - accuracy: 0.3870 - val_loss: 3.7707 - val_accuracy: 0.3670\n",
+      "63/63 [==============================] - 6s 101ms/step - loss: 3.8269 - accuracy: 0.3913 - val_loss: 3.8741 - val_accuracy: 0.3390\n",
       "Epoch 26/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 3.6182 - accuracy: 0.3915 - val_loss: 3.6817 - val_accuracy: 0.3400\n",
+      "63/63 [==============================] - 6s 94ms/step - loss: 3.6984 - accuracy: 0.3842 - val_loss: 4.0194 - val_accuracy: 0.2720\n",
       "Epoch 27/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 3.5141 - accuracy: 0.3930 - val_loss: 3.5192 - val_accuracy: 0.3780\n",
+      "63/63 [==============================] - 6s 88ms/step - loss: 3.6033 - accuracy: 0.3762 - val_loss: 3.7121 - val_accuracy: 0.3350\n",
       "Epoch 28/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 3.4040 - accuracy: 0.3955 - val_loss: 3.7013 - val_accuracy: 0.2940\n",
+      "63/63 [==============================] - 6s 92ms/step - loss: 3.5039 - accuracy: 0.3873 - val_loss: 3.6310 - val_accuracy: 0.3140\n",
       "Epoch 29/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 3.3146 - accuracy: 0.3915 - val_loss: 3.5867 - val_accuracy: 0.3090\n",
+      "63/63 [==============================] - 6s 94ms/step - loss: 3.3895 - accuracy: 0.3873 - val_loss: 3.6758 - val_accuracy: 0.2780\n",
       "Epoch 30/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 3.2321 - accuracy: 0.3925 - val_loss: 3.3006 - val_accuracy: 0.3440\n",
+      "63/63 [==============================] - 6s 88ms/step - loss: 3.3008 - accuracy: 0.3965 - val_loss: 3.4139 - val_accuracy: 0.3260\n",
       "Epoch 31/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 3.1470 - accuracy: 0.3983 - val_loss: 3.3153 - val_accuracy: 0.3330\n",
+      "63/63 [==============================] - 6s 94ms/step - loss: 3.2209 - accuracy: 0.3960 - val_loss: 3.6432 - val_accuracy: 0.2830\n",
       "Epoch 32/100\n",
-      "63/63 [==============================] - 3s 53ms/step - loss: 3.0724 - accuracy: 0.3943 - val_loss: 3.0938 - val_accuracy: 0.3720\n",
+      "63/63 [==============================] - 6s 90ms/step - loss: 3.1354 - accuracy: 0.4013 - val_loss: 3.2672 - val_accuracy: 0.3400\n",
       "Epoch 33/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 2.9984 - accuracy: 0.4033 - val_loss: 3.3114 - val_accuracy: 0.2810\n",
+      "63/63 [==============================] - 6s 95ms/step - loss: 3.0679 - accuracy: 0.3873 - val_loss: 3.4932 - val_accuracy: 0.2670\n",
       "Epoch 34/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 2.9421 - accuracy: 0.3997 - val_loss: 3.0574 - val_accuracy: 0.3520\n",
+      "63/63 [==============================] - 6s 88ms/step - loss: 3.0092 - accuracy: 0.3842 - val_loss: 3.4149 - val_accuracy: 0.2750\n",
       "Epoch 35/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 2.8831 - accuracy: 0.4017 - val_loss: 2.9282 - val_accuracy: 0.3620\n",
+      "63/63 [==============================] - 6s 94ms/step - loss: 2.9387 - accuracy: 0.3963 - val_loss: 3.0612 - val_accuracy: 0.3430\n",
       "Epoch 36/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 2.8355 - accuracy: 0.3997 - val_loss: 3.0449 - val_accuracy: 0.3300\n",
+      "63/63 [==============================] - 6s 88ms/step - loss: 2.8899 - accuracy: 0.4025 - val_loss: 2.9881 - val_accuracy: 0.3310\n",
       "Epoch 37/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 2.7722 - accuracy: 0.4078 - val_loss: 2.9248 - val_accuracy: 0.3350\n",
+      "63/63 [==============================] - 5s 87ms/step - loss: 2.8424 - accuracy: 0.3947 - val_loss: 2.9952 - val_accuracy: 0.3140\n",
       "Epoch 38/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 2.7329 - accuracy: 0.3995 - val_loss: 2.8284 - val_accuracy: 0.3350\n",
+      "63/63 [==============================] - 6s 91ms/step - loss: 2.7795 - accuracy: 0.3913 - val_loss: 3.1798 - val_accuracy: 0.2870\n",
       "Epoch 39/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 2.7045 - accuracy: 0.3945 - val_loss: 2.9529 - val_accuracy: 0.3260\n",
+      "63/63 [==============================] - 6s 90ms/step - loss: 2.7378 - accuracy: 0.3915 - val_loss: 2.8324 - val_accuracy: 0.3590\n",
       "Epoch 40/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 2.6383 - accuracy: 0.4030 - val_loss: 2.8162 - val_accuracy: 0.3480\n",
+      "63/63 [==============================] - 6s 96ms/step - loss: 2.7003 - accuracy: 0.3923 - val_loss: 2.7913 - val_accuracy: 0.3480\n",
       "Epoch 41/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 2.5990 - accuracy: 0.4090 - val_loss: 2.7714 - val_accuracy: 0.3250\n",
+      "63/63 [==============================] - 6s 90ms/step - loss: 2.6558 - accuracy: 0.3938 - val_loss: 2.9557 - val_accuracy: 0.2860\n",
       "Epoch 42/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 2.5558 - accuracy: 0.4070 - val_loss: 2.8227 - val_accuracy: 0.3200\n",
+      "63/63 [==============================] - 6s 88ms/step - loss: 2.6065 - accuracy: 0.3930 - val_loss: 2.8128 - val_accuracy: 0.3180\n",
       "Epoch 43/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 2.5364 - accuracy: 0.4035 - val_loss: 2.7441 - val_accuracy: 0.3170\n",
+      "63/63 [==============================] - 6s 92ms/step - loss: 2.5631 - accuracy: 0.4095 - val_loss: 2.7912 - val_accuracy: 0.3540\n",
       "Epoch 44/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 2.5107 - accuracy: 0.4015 - val_loss: 2.6493 - val_accuracy: 0.3480\n",
+      "63/63 [==============================] - 9s 138ms/step - loss: 2.5446 - accuracy: 0.3968 - val_loss: 2.7181 - val_accuracy: 0.3350\n",
       "Epoch 45/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 2.4749 - accuracy: 0.4040 - val_loss: 2.5891 - val_accuracy: 0.3630\n",
+      "63/63 [==============================] - 11s 175ms/step - loss: 2.5097 - accuracy: 0.4035 - val_loss: 2.5991 - val_accuracy: 0.3520\n",
       "Epoch 46/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 2.4396 - accuracy: 0.4153 - val_loss: 2.5481 - val_accuracy: 0.3770\n",
+      "63/63 [==============================] - 6s 93ms/step - loss: 2.4774 - accuracy: 0.3980 - val_loss: 2.6282 - val_accuracy: 0.3430\n",
       "Epoch 47/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 2.3978 - accuracy: 0.4178 - val_loss: 2.6178 - val_accuracy: 0.3200\n",
+      "63/63 [==============================] - 6s 91ms/step - loss: 2.4469 - accuracy: 0.3997 - val_loss: 2.8327 - val_accuracy: 0.2810\n",
       "Epoch 48/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 2.3826 - accuracy: 0.4135 - val_loss: 2.6158 - val_accuracy: 0.3330\n",
+      "63/63 [==============================] - 5s 87ms/step - loss: 2.4187 - accuracy: 0.4000 - val_loss: 2.8206 - val_accuracy: 0.2760\n",
       "Epoch 49/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 2.3750 - accuracy: 0.4117 - val_loss: 2.6044 - val_accuracy: 0.3170\n",
+      "63/63 [==============================] - 6s 91ms/step - loss: 2.4080 - accuracy: 0.4035 - val_loss: 2.6159 - val_accuracy: 0.3110\n",
       "Epoch 50/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 2.3392 - accuracy: 0.4085 - val_loss: 2.7147 - val_accuracy: 0.2760\n",
+      "63/63 [==============================] - 6s 89ms/step - loss: 2.3828 - accuracy: 0.4055 - val_loss: 2.6431 - val_accuracy: 0.2970\n",
       "Epoch 51/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 2.3171 - accuracy: 0.4128 - val_loss: 2.4306 - val_accuracy: 0.3690\n",
+      "63/63 [==============================] - 6s 94ms/step - loss: 2.3392 - accuracy: 0.4065 - val_loss: 2.4633 - val_accuracy: 0.3610\n",
       "Epoch 52/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 2.3016 - accuracy: 0.4123 - val_loss: 2.4043 - val_accuracy: 0.3490\n",
+      "63/63 [==============================] - 6s 87ms/step - loss: 2.3308 - accuracy: 0.4117 - val_loss: 2.6086 - val_accuracy: 0.3030\n",
       "Epoch 53/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 2.2820 - accuracy: 0.4075 - val_loss: 2.4418 - val_accuracy: 0.3470\n",
+      "63/63 [==============================] - 6s 92ms/step - loss: 2.3193 - accuracy: 0.4030 - val_loss: 2.5730 - val_accuracy: 0.3050\n",
       "Epoch 54/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 2.2535 - accuracy: 0.4243 - val_loss: 2.6577 - val_accuracy: 0.2940\n",
+      "63/63 [==============================] - 6s 90ms/step - loss: 2.2961 - accuracy: 0.4095 - val_loss: 2.5999 - val_accuracy: 0.3080\n",
       "Epoch 55/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 2.2439 - accuracy: 0.4160 - val_loss: 2.4222 - val_accuracy: 0.3430\n",
+      "63/63 [==============================] - 6s 89ms/step - loss: 2.2812 - accuracy: 0.4033 - val_loss: 2.4191 - val_accuracy: 0.3650\n",
       "Epoch 56/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 2.2260 - accuracy: 0.4123 - val_loss: 2.4573 - val_accuracy: 0.2970\n",
+      "63/63 [==============================] - 6s 94ms/step - loss: 2.2537 - accuracy: 0.4072 - val_loss: 2.5827 - val_accuracy: 0.2960\n",
       "Epoch 57/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 2.2228 - accuracy: 0.4078 - val_loss: 2.3692 - val_accuracy: 0.3560\n",
+      "63/63 [==============================] - 6s 88ms/step - loss: 2.2567 - accuracy: 0.4017 - val_loss: 2.8786 - val_accuracy: 0.2270\n",
       "Epoch 58/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 2.1989 - accuracy: 0.4120 - val_loss: 2.3348 - val_accuracy: 0.3610\n",
+      "63/63 [==============================] - 6s 91ms/step - loss: 2.2384 - accuracy: 0.4042 - val_loss: 2.3259 - val_accuracy: 0.3660\n",
       "Epoch 59/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 2.1825 - accuracy: 0.4115 - val_loss: 2.3080 - val_accuracy: 0.3750\n",
+      "63/63 [==============================] - 6s 88ms/step - loss: 2.2352 - accuracy: 0.3960 - val_loss: 2.4744 - val_accuracy: 0.3210\n",
       "Epoch 60/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 2.1790 - accuracy: 0.4160 - val_loss: 2.3991 - val_accuracy: 0.3390\n",
+      "63/63 [==============================] - 6s 89ms/step - loss: 2.2107 - accuracy: 0.4008 - val_loss: 2.5989 - val_accuracy: 0.3060\n",
       "Epoch 61/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 2.1608 - accuracy: 0.4165 - val_loss: 2.3595 - val_accuracy: 0.3490\n",
+      "63/63 [==============================] - 6s 92ms/step - loss: 2.1988 - accuracy: 0.4002 - val_loss: 2.8597 - val_accuracy: 0.2500\n",
       "Epoch 62/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 2.1461 - accuracy: 0.4215 - val_loss: 2.2999 - val_accuracy: 0.3640\n",
+      "63/63 [==============================] - 6s 91ms/step - loss: 2.1779 - accuracy: 0.4100 - val_loss: 2.3780 - val_accuracy: 0.3270\n",
       "Epoch 63/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 2.1236 - accuracy: 0.4182 - val_loss: 2.2934 - val_accuracy: 0.3340\n",
+      "63/63 [==============================] - 6s 95ms/step - loss: 2.1578 - accuracy: 0.4025 - val_loss: 2.5887 - val_accuracy: 0.2920\n",
       "Epoch 64/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 2.1171 - accuracy: 0.4195 - val_loss: 2.2164 - val_accuracy: 0.3920\n",
+      "63/63 [==============================] - 6s 98ms/step - loss: 2.1611 - accuracy: 0.4022 - val_loss: 2.2919 - val_accuracy: 0.3620\n",
       "Epoch 65/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 2.1182 - accuracy: 0.4225 - val_loss: 2.3244 - val_accuracy: 0.3470\n",
+      "63/63 [==============================] - 6s 93ms/step - loss: 2.1421 - accuracy: 0.4065 - val_loss: 2.1934 - val_accuracy: 0.3840\n",
       "Epoch 66/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 2.0955 - accuracy: 0.4212 - val_loss: 2.2796 - val_accuracy: 0.3820\n",
+      "63/63 [==============================] - 6s 98ms/step - loss: 2.1225 - accuracy: 0.4137 - val_loss: 2.2906 - val_accuracy: 0.3780\n",
       "Epoch 67/100\n",
-      "63/63 [==============================] - 3s 48ms/step - loss: 2.0840 - accuracy: 0.4210 - val_loss: 2.2425 - val_accuracy: 0.3500\n",
+      "63/63 [==============================] - 6s 94ms/step - loss: 2.1114 - accuracy: 0.4103 - val_loss: 2.5732 - val_accuracy: 0.2620\n",
       "Epoch 68/100\n",
-      "63/63 [==============================] - 3s 53ms/step - loss: 2.0637 - accuracy: 0.4198 - val_loss: 2.2143 - val_accuracy: 0.3870\n",
+      "63/63 [==============================] - 6s 91ms/step - loss: 2.1030 - accuracy: 0.4078 - val_loss: 2.3567 - val_accuracy: 0.3350\n",
       "Epoch 69/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 2.0766 - accuracy: 0.4220 - val_loss: 2.2212 - val_accuracy: 0.3630\n",
+      "63/63 [==============================] - 6s 96ms/step - loss: 2.0863 - accuracy: 0.4157 - val_loss: 2.1938 - val_accuracy: 0.3840\n",
       "Epoch 70/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 2.0500 - accuracy: 0.4285 - val_loss: 2.3182 - val_accuracy: 0.3390\n",
+      "63/63 [==============================] - 6s 89ms/step - loss: 2.0790 - accuracy: 0.4108 - val_loss: 2.5503 - val_accuracy: 0.3050\n",
       "Epoch 71/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 2.0597 - accuracy: 0.4265 - val_loss: 2.2217 - val_accuracy: 0.3680\n",
+      "63/63 [==============================] - 6s 94ms/step - loss: 2.0817 - accuracy: 0.4100 - val_loss: 2.4150 - val_accuracy: 0.3000\n",
       "Epoch 72/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 2.0411 - accuracy: 0.4320 - val_loss: 2.4726 - val_accuracy: 0.2700\n",
+      "63/63 [==============================] - 6s 90ms/step - loss: 2.0706 - accuracy: 0.4108 - val_loss: 2.2823 - val_accuracy: 0.3300\n",
       "Epoch 73/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 2.0423 - accuracy: 0.4207 - val_loss: 2.3842 - val_accuracy: 0.3170\n",
+      "63/63 [==============================] - 6s 87ms/step - loss: 2.0592 - accuracy: 0.4148 - val_loss: 2.2665 - val_accuracy: 0.3340\n",
       "Epoch 74/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 2.0162 - accuracy: 0.4283 - val_loss: 2.2431 - val_accuracy: 0.3300\n",
+      "63/63 [==============================] - 6s 89ms/step - loss: 2.0342 - accuracy: 0.4223 - val_loss: 2.1628 - val_accuracy: 0.3720\n",
       "Epoch 75/100\n",
-      "63/63 [==============================] - 3s 53ms/step - loss: 2.0308 - accuracy: 0.4195 - val_loss: 2.1713 - val_accuracy: 0.3750\n",
+      "63/63 [==============================] - 6s 91ms/step - loss: 2.0361 - accuracy: 0.4162 - val_loss: 2.1333 - val_accuracy: 0.3660\n",
       "Epoch 76/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 2.0196 - accuracy: 0.4243 - val_loss: 2.1299 - val_accuracy: 0.3650\n",
+      "63/63 [==============================] - 6s 91ms/step - loss: 2.0346 - accuracy: 0.4135 - val_loss: 2.1840 - val_accuracy: 0.3620\n",
       "Epoch 77/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 2.0190 - accuracy: 0.4130 - val_loss: 2.1529 - val_accuracy: 0.3580\n",
+      "63/63 [==============================] - 6s 91ms/step - loss: 2.0472 - accuracy: 0.4125 - val_loss: 2.2439 - val_accuracy: 0.3370\n",
       "Epoch 78/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.9974 - accuracy: 0.4392 - val_loss: 2.1087 - val_accuracy: 0.3870\n",
+      "63/63 [==============================] - 6s 91ms/step - loss: 2.0369 - accuracy: 0.4078 - val_loss: 2.1270 - val_accuracy: 0.3920\n",
       "Epoch 79/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 2.0033 - accuracy: 0.4182 - val_loss: 2.1338 - val_accuracy: 0.3810\n",
+      "63/63 [==============================] - 6s 93ms/step - loss: 2.0165 - accuracy: 0.4182 - val_loss: 2.2412 - val_accuracy: 0.3310\n",
       "Epoch 80/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.9884 - accuracy: 0.4193 - val_loss: 2.2034 - val_accuracy: 0.3450\n",
+      "63/63 [==============================] - 6s 89ms/step - loss: 2.0148 - accuracy: 0.4078 - val_loss: 2.3163 - val_accuracy: 0.3190\n",
       "Epoch 81/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 1.9885 - accuracy: 0.4243 - val_loss: 2.2258 - val_accuracy: 0.3190\n",
+      "63/63 [==============================] - 6s 89ms/step - loss: 2.0013 - accuracy: 0.4105 - val_loss: 2.2562 - val_accuracy: 0.3210\n",
       "Epoch 82/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 1.9741 - accuracy: 0.4272 - val_loss: 2.2169 - val_accuracy: 0.3320\n",
+      "63/63 [==============================] - 6s 93ms/step - loss: 2.0057 - accuracy: 0.4103 - val_loss: 2.1052 - val_accuracy: 0.3600\n",
       "Epoch 83/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.9772 - accuracy: 0.4248 - val_loss: 2.3005 - val_accuracy: 0.3160\n",
+      "63/63 [==============================] - 6s 92ms/step - loss: 1.9984 - accuracy: 0.4065 - val_loss: 2.3521 - val_accuracy: 0.3040\n",
       "Epoch 84/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 1.9743 - accuracy: 0.4218 - val_loss: 2.2183 - val_accuracy: 0.3270\n",
+      "63/63 [==============================] - 6s 103ms/step - loss: 2.0071 - accuracy: 0.4095 - val_loss: 2.8214 - val_accuracy: 0.2440\n",
       "Epoch 85/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.9604 - accuracy: 0.4205 - val_loss: 2.1792 - val_accuracy: 0.3190\n",
+      "63/63 [==============================] - 5s 86ms/step - loss: 1.9809 - accuracy: 0.4112 - val_loss: 2.1729 - val_accuracy: 0.3370\n",
       "Epoch 86/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.9573 - accuracy: 0.4288 - val_loss: 2.1856 - val_accuracy: 0.3370\n",
+      "63/63 [==============================] - 6s 96ms/step - loss: 1.9611 - accuracy: 0.4227 - val_loss: 2.1808 - val_accuracy: 0.3400\n",
       "Epoch 87/100\n",
-      "63/63 [==============================] - 3s 54ms/step - loss: 1.9427 - accuracy: 0.4300 - val_loss: 2.3086 - val_accuracy: 0.2950\n",
+      "63/63 [==============================] - 6s 88ms/step - loss: 1.9549 - accuracy: 0.4238 - val_loss: 2.1579 - val_accuracy: 0.3220\n",
       "Epoch 88/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 1.9463 - accuracy: 0.4255 - val_loss: 2.4385 - val_accuracy: 0.2730\n",
+      "63/63 [==============================] - 6s 89ms/step - loss: 1.9649 - accuracy: 0.4145 - val_loss: 2.1404 - val_accuracy: 0.3790\n",
       "Epoch 89/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 1.9661 - accuracy: 0.4132 - val_loss: 2.0925 - val_accuracy: 0.3700\n",
+      "63/63 [==============================] - 6s 94ms/step - loss: 1.9651 - accuracy: 0.4190 - val_loss: 2.0501 - val_accuracy: 0.3910\n",
       "Epoch 90/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.9347 - accuracy: 0.4408 - val_loss: 2.0926 - val_accuracy: 0.3600\n",
+      "63/63 [==============================] - 6s 91ms/step - loss: 1.9717 - accuracy: 0.4140 - val_loss: 2.3215 - val_accuracy: 0.2900\n",
       "Epoch 91/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.9430 - accuracy: 0.4225 - val_loss: 2.2100 - val_accuracy: 0.3250\n",
+      "63/63 [==============================] - 6s 94ms/step - loss: 1.9553 - accuracy: 0.4182 - val_loss: 2.1125 - val_accuracy: 0.3660\n",
       "Epoch 92/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.9178 - accuracy: 0.4255 - val_loss: 2.3889 - val_accuracy: 0.2730\n",
+      "63/63 [==============================] - 6s 87ms/step - loss: 1.9436 - accuracy: 0.4155 - val_loss: 2.1085 - val_accuracy: 0.3620\n",
       "Epoch 93/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.9214 - accuracy: 0.4260 - val_loss: 2.0543 - val_accuracy: 0.3960\n",
+      "63/63 [==============================] - 6s 95ms/step - loss: 1.9441 - accuracy: 0.4112 - val_loss: 2.1820 - val_accuracy: 0.3260\n",
       "Epoch 94/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.9320 - accuracy: 0.4200 - val_loss: 2.1397 - val_accuracy: 0.3460\n",
+      "63/63 [==============================] - 6s 88ms/step - loss: 1.9377 - accuracy: 0.4255 - val_loss: 2.1216 - val_accuracy: 0.3670\n",
       "Epoch 95/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.9199 - accuracy: 0.4335 - val_loss: 2.0488 - val_accuracy: 0.3780\n",
+      "63/63 [==============================] - 6s 88ms/step - loss: 1.9356 - accuracy: 0.4285 - val_loss: 2.1735 - val_accuracy: 0.3260\n",
       "Epoch 96/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 1.8967 - accuracy: 0.4288 - val_loss: 2.4588 - val_accuracy: 0.2750\n",
+      "63/63 [==============================] - 6s 96ms/step - loss: 1.9493 - accuracy: 0.4078 - val_loss: 2.5470 - val_accuracy: 0.2460\n",
       "Epoch 97/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 1.9234 - accuracy: 0.4238 - val_loss: 2.0139 - val_accuracy: 0.3970\n",
+      "63/63 [==============================] - 6s 90ms/step - loss: 1.9324 - accuracy: 0.4090 - val_loss: 2.0640 - val_accuracy: 0.3690\n",
       "Epoch 98/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 1.9150 - accuracy: 0.4300 - val_loss: 2.0700 - val_accuracy: 0.3640\n",
+      "63/63 [==============================] - 6s 90ms/step - loss: 1.9279 - accuracy: 0.4257 - val_loss: 2.2967 - val_accuracy: 0.2970\n",
       "Epoch 99/100\n",
-      "63/63 [==============================] - 3s 53ms/step - loss: 1.8897 - accuracy: 0.4355 - val_loss: 2.1719 - val_accuracy: 0.3320\n",
+      "63/63 [==============================] - 5s 85ms/step - loss: 1.9366 - accuracy: 0.4130 - val_loss: 2.2642 - val_accuracy: 0.3260\n",
       "Epoch 100/100\n",
-      "63/63 [==============================] - 3s 54ms/step - loss: 1.8948 - accuracy: 0.4408 - val_loss: 1.9570 - val_accuracy: 0.4010\n"
+      "63/63 [==============================] - 5s 84ms/step - loss: 1.9155 - accuracy: 0.4162 - val_loss: 2.1090 - val_accuracy: 0.3570\n"
      ]
     }
    ],
@@ -3561,14 +3574,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 46,
+   "execution_count": 47,
    "metadata": {
     "id": "PSMyg4Q_vXac"
    },
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 576x576 with 2 Axes>"
       ]
@@ -3606,7 +3619,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 47,
+   "execution_count": 48,
    "metadata": {
     "id": "UI8zctRKvXaf"
    },
@@ -3615,8 +3628,8 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "10000/10000 [==============================] - 1s 111us/sample - loss: 1.9597 - accuracy: 0.4027\n",
-      "Accuracy on test dataset: 0.4027\n"
+      "313/313 [==============================] - 5s 14ms/step - loss: 2.0938 - accuracy: 0.3669\n",
+      "Accuracy on test dataset: 0.3668999969959259\n"
      ]
     }
    ],
@@ -3629,655 +3642,56 @@
   {
    "cell_type": "markdown",
    "metadata": {
-    "id": "XBZuaco_vXaj"
+    "id": "5zgbDpH0sH7u"
    },
    "source": [
-    "## 7.2 L1 Regularization"
+    "# 7. TensorBoard\n",
+    "\n",
+    "TensorBoard is a great tool to observe variables during training (especially useful for models training a long time, like above model).\n"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": null,
    "metadata": {
-    "id": "mIGlhfVQvXak"
+    "colab": {
+     "base_uri": "https://localhost:8080/",
+     "height": 51
+    },
+    "id": "FP5b7DGle7to",
+    "outputId": "9be8f4da-e6a9-45ab-aa2c-0c5a4e994055"
    },
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "WARNING:tensorflow:sample_weight modes were coerced from\n",
-      "  ...\n",
-      "    to  \n",
-      "  ['...']\n",
-      "WARNING:tensorflow:sample_weight modes were coerced from\n",
-      "  ...\n",
-      "    to  \n",
-      "  ['...']\n",
-      "Train for 63 steps, validate for 16 steps\n",
-      "Epoch 1/100\n",
-      "63/63 [==============================] - 4s 66ms/step - loss: 327.1665 - accuracy: 0.0873 - val_loss: 327.0736 - val_accuracy: 0.1030\n",
-      "Epoch 2/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 326.1466 - accuracy: 0.0920 - val_loss: 325.6699 - val_accuracy: 0.1070\n",
-      "Epoch 3/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 325.1237 - accuracy: 0.1010 - val_loss: 324.5874 - val_accuracy: 0.1270\n",
-      "Epoch 4/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 324.0984 - accuracy: 0.1065 - val_loss: 323.5617 - val_accuracy: 0.1310\n",
-      "Epoch 5/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 323.0910 - accuracy: 0.1195 - val_loss: 322.5514 - val_accuracy: 0.1360\n",
-      "Epoch 6/100\n",
-      "63/63 [==============================] - 3s 54ms/step - loss: 322.0914 - accuracy: 0.1268 - val_loss: 321.5405 - val_accuracy: 0.1370\n",
-      "Epoch 7/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 321.0686 - accuracy: 0.1332 - val_loss: 320.5320 - val_accuracy: 0.1410\n",
-      "Epoch 8/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 320.0708 - accuracy: 0.1510 - val_loss: 319.5321 - val_accuracy: 0.1470\n",
-      "Epoch 9/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 319.0765 - accuracy: 0.1435 - val_loss: 318.5339 - val_accuracy: 0.1520\n",
-      "Epoch 10/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 318.0603 - accuracy: 0.1610 - val_loss: 317.5360 - val_accuracy: 0.1550\n",
-      "Epoch 11/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 317.0946 - accuracy: 0.1575 - val_loss: 316.5469 - val_accuracy: 0.1600\n",
-      "Epoch 12/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 316.1055 - accuracy: 0.1620 - val_loss: 315.5569 - val_accuracy: 0.1650\n",
-      "Epoch 13/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 315.1094 - accuracy: 0.1650 - val_loss: 314.5706 - val_accuracy: 0.1700\n",
-      "Epoch 14/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 314.1318 - accuracy: 0.1725 - val_loss: 313.5870 - val_accuracy: 0.1770\n",
-      "Epoch 15/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 313.1319 - accuracy: 0.1838 - val_loss: 312.6048 - val_accuracy: 0.1820\n",
-      "Epoch 16/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 312.1464 - accuracy: 0.1870 - val_loss: 311.6259 - val_accuracy: 0.1890\n",
-      "Epoch 17/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 311.1884 - accuracy: 0.1825 - val_loss: 310.6478 - val_accuracy: 0.1940\n",
-      "Epoch 18/100\n",
-      "63/63 [==============================] - 3s 48ms/step - loss: 310.2142 - accuracy: 0.1780 - val_loss: 309.6715 - val_accuracy: 0.2020\n",
-      "Epoch 19/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 309.2176 - accuracy: 0.1887 - val_loss: 308.6996 - val_accuracy: 0.2070\n",
-      "Epoch 20/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 308.2632 - accuracy: 0.1910 - val_loss: 307.7300 - val_accuracy: 0.2040\n",
-      "Epoch 21/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 307.2966 - accuracy: 0.2050 - val_loss: 306.7606 - val_accuracy: 0.2140\n",
-      "Epoch 22/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 306.3224 - accuracy: 0.1985 - val_loss: 305.7940 - val_accuracy: 0.2160\n",
-      "Epoch 23/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 305.3623 - accuracy: 0.2045 - val_loss: 304.8271 - val_accuracy: 0.2210\n",
-      "Epoch 24/100\n",
-      "63/63 [==============================] - 3s 53ms/step - loss: 304.3907 - accuracy: 0.1998 - val_loss: 303.8641 - val_accuracy: 0.2250\n",
-      "Epoch 25/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 303.4518 - accuracy: 0.2030 - val_loss: 302.9034 - val_accuracy: 0.2260\n",
-      "Epoch 26/100\n",
-      "63/63 [==============================] - 3s 53ms/step - loss: 302.4777 - accuracy: 0.2035 - val_loss: 301.9412 - val_accuracy: 0.2290\n",
-      "Epoch 27/100\n",
-      "63/63 [==============================] - 3s 53ms/step - loss: 301.5256 - accuracy: 0.2105 - val_loss: 300.9859 - val_accuracy: 0.2330\n",
-      "Epoch 28/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 300.5605 - accuracy: 0.2085 - val_loss: 300.0311 - val_accuracy: 0.2350\n",
-      "Epoch 29/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 299.6143 - accuracy: 0.2120 - val_loss: 299.0780 - val_accuracy: 0.2430\n",
-      "Epoch 30/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 298.6453 - accuracy: 0.2175 - val_loss: 298.1245 - val_accuracy: 0.2430\n",
-      "Epoch 31/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 297.7067 - accuracy: 0.2212 - val_loss: 297.1722 - val_accuracy: 0.2460\n",
-      "Epoch 32/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 296.7560 - accuracy: 0.2218 - val_loss: 296.2218 - val_accuracy: 0.2460\n",
-      "Epoch 33/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 295.8042 - accuracy: 0.2188 - val_loss: 295.2757 - val_accuracy: 0.2480\n",
-      "Epoch 34/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 294.8590 - accuracy: 0.2278 - val_loss: 294.3298 - val_accuracy: 0.2520\n",
-      "Epoch 35/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 293.9075 - accuracy: 0.2393 - val_loss: 293.3864 - val_accuracy: 0.2500\n",
-      "Epoch 36/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 292.9801 - accuracy: 0.2260 - val_loss: 292.4443 - val_accuracy: 0.2590\n",
-      "Epoch 37/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 292.0424 - accuracy: 0.2360 - val_loss: 291.5052 - val_accuracy: 0.2580\n",
-      "Epoch 38/100\n",
-      "63/63 [==============================] - 3s 54ms/step - loss: 291.1082 - accuracy: 0.2292 - val_loss: 290.5645 - val_accuracy: 0.2630\n",
-      "Epoch 39/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 290.1467 - accuracy: 0.2368 - val_loss: 289.6267 - val_accuracy: 0.2670\n",
-      "Epoch 40/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 289.2212 - accuracy: 0.2387 - val_loss: 288.6919 - val_accuracy: 0.2670\n",
-      "Epoch 41/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 288.2890 - accuracy: 0.2340 - val_loss: 287.7593 - val_accuracy: 0.2660\n",
-      "Epoch 42/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 287.3457 - accuracy: 0.2432 - val_loss: 286.8265 - val_accuracy: 0.2690\n",
-      "Epoch 43/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 286.4145 - accuracy: 0.2345 - val_loss: 285.8943 - val_accuracy: 0.2730\n",
-      "Epoch 44/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 285.4957 - accuracy: 0.2265 - val_loss: 284.9667 - val_accuracy: 0.2770\n",
-      "Epoch 45/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 284.5951 - accuracy: 0.2257 - val_loss: 284.0397 - val_accuracy: 0.2760\n",
-      "Epoch 46/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 283.6504 - accuracy: 0.2365 - val_loss: 283.1138 - val_accuracy: 0.2810\n",
-      "Epoch 47/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 282.7088 - accuracy: 0.2377 - val_loss: 282.1896 - val_accuracy: 0.2770\n",
-      "Epoch 48/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 281.7958 - accuracy: 0.2520 - val_loss: 281.2671 - val_accuracy: 0.2820\n",
-      "Epoch 49/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 280.8683 - accuracy: 0.2438 - val_loss: 280.3452 - val_accuracy: 0.2860\n",
-      "Epoch 50/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 279.9677 - accuracy: 0.2442 - val_loss: 279.4262 - val_accuracy: 0.2850\n",
-      "Epoch 51/100\n",
-      "63/63 [==============================] - 3s 53ms/step - loss: 279.0381 - accuracy: 0.2345 - val_loss: 278.5090 - val_accuracy: 0.2860\n",
-      "Epoch 52/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 278.1271 - accuracy: 0.2465 - val_loss: 277.5904 - val_accuracy: 0.2850\n",
-      "Epoch 53/100\n",
-      "63/63 [==============================] - 3s 54ms/step - loss: 277.2080 - accuracy: 0.2525 - val_loss: 276.6772 - val_accuracy: 0.2870\n",
-      "Epoch 54/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 276.2863 - accuracy: 0.2477 - val_loss: 275.7659 - val_accuracy: 0.2900\n",
-      "Epoch 55/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 275.3843 - accuracy: 0.2535 - val_loss: 274.8553 - val_accuracy: 0.2900\n",
-      "Epoch 56/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 274.4735 - accuracy: 0.2505 - val_loss: 273.9461 - val_accuracy: 0.2920\n",
-      "Epoch 57/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 273.5583 - accuracy: 0.2512 - val_loss: 273.0384 - val_accuracy: 0.2940\n",
-      "Epoch 58/100\n",
-      "63/63 [==============================] - 3s 54ms/step - loss: 272.6652 - accuracy: 0.2503 - val_loss: 272.1306 - val_accuracy: 0.2930\n",
-      "Epoch 59/100\n",
-      "63/63 [==============================] - 3s 53ms/step - loss: 271.7381 - accuracy: 0.2520 - val_loss: 271.2254 - val_accuracy: 0.2930\n",
-      "Epoch 60/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 270.8439 - accuracy: 0.2550 - val_loss: 270.3220 - val_accuracy: 0.2980\n",
-      "Epoch 61/100\n",
-      "63/63 [==============================] - 3s 53ms/step - loss: 269.9411 - accuracy: 0.2558 - val_loss: 269.4202 - val_accuracy: 0.3000\n",
-      "Epoch 62/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 269.0396 - accuracy: 0.2640 - val_loss: 268.5195 - val_accuracy: 0.3050\n",
-      "Epoch 63/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 268.1388 - accuracy: 0.2585 - val_loss: 267.6217 - val_accuracy: 0.3000\n",
-      "Epoch 64/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 267.2439 - accuracy: 0.2558 - val_loss: 266.7258 - val_accuracy: 0.3030\n",
-      "Epoch 65/100\n",
-      "63/63 [==============================] - 3s 49ms/step - loss: 266.3549 - accuracy: 0.2643 - val_loss: 265.8312 - val_accuracy: 0.3080\n",
-      "Epoch 66/100\n",
-      "63/63 [==============================] - 3s 53ms/step - loss: 265.4531 - accuracy: 0.2537 - val_loss: 264.9359 - val_accuracy: 0.3050\n",
-      "Epoch 67/100\n",
-      "63/63 [==============================] - 3s 53ms/step - loss: 264.5504 - accuracy: 0.2775 - val_loss: 264.0421 - val_accuracy: 0.3050\n",
-      "Epoch 68/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 263.6703 - accuracy: 0.2578 - val_loss: 263.1519 - val_accuracy: 0.3050\n",
-      "Epoch 69/100\n",
-      "63/63 [==============================] - 3s 50ms/step - loss: 262.7795 - accuracy: 0.2610 - val_loss: 262.2625 - val_accuracy: 0.3070\n",
-      "Epoch 70/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 261.8926 - accuracy: 0.2637 - val_loss: 261.3748 - val_accuracy: 0.3130\n",
-      "Epoch 71/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 260.9946 - accuracy: 0.2670 - val_loss: 260.4905 - val_accuracy: 0.3090\n",
-      "Epoch 72/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 260.1207 - accuracy: 0.2663 - val_loss: 259.6058 - val_accuracy: 0.3120\n",
-      "Epoch 73/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 259.2258 - accuracy: 0.2682 - val_loss: 258.7226 - val_accuracy: 0.3120\n",
-      "Epoch 74/100\n",
-      "63/63 [==============================] - 3s 53ms/step - loss: 258.3605 - accuracy: 0.2697 - val_loss: 257.8402 - val_accuracy: 0.3120\n",
-      "Epoch 75/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 257.4714 - accuracy: 0.2697 - val_loss: 256.9610 - val_accuracy: 0.3120\n",
-      "Epoch 76/100\n",
-      "63/63 [==============================] - 3s 52ms/step - loss: 256.5937 - accuracy: 0.2740 - val_loss: 256.0830 - val_accuracy: 0.3150\n",
-      "Epoch 77/100\n",
-      "63/63 [==============================] - 3s 51ms/step - loss: 255.7213 - accuracy: 0.2733 - val_loss: 255.2066 - val_accuracy: 0.3160\n",
-      "Epoch 78/100\n",
-      "50/63 [======================>.......] - ETA: 0s - loss: 254.9224 - accuracy: 0.2781"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
-    "# Define Network\n",
-    "model = tf.keras.Sequential([\n",
-    "        tf.keras.layers.Flatten(input_shape=(32, 32, 3)),\n",
-    "        tf.keras.layers.Dense(512, \n",
-    "                              kernel_regularizer=tf.keras.regularizers.l1(0.01)),\n",
-    "        tf.keras.layers.BatchNormalization(),\n",
-    "        tf.keras.layers.Activation(tf.nn.relu),\n",
-    "        tf.keras.layers.Dense(128, \n",
-    "                              kernel_regularizer=tf.keras.regularizers.l1(0.001)),\n",
-    "        tf.keras.layers.BatchNormalization(),\n",
-    "        tf.keras.layers.Activation(tf.nn.relu),\n",
-    "        tf.keras.layers.Dense(10,  \n",
-    "                              activation=tf.nn.softmax,  \n",
-    "                              kernel_regularizer=tf.keras.regularizers.l1(0.001))\n",
-    "])\n",
-    "\n",
-    "# Compile Network\n",
-    "model.compile(optimizer=tf.keras.optimizers.SGD(lr=0.0001),\n",
-    "              loss='categorical_crossentropy',\n",
-    "              metrics=['accuracy'])\n",
-    "\n",
-    "# Fit Network\n",
-    "num_validation_examples = 1000\n",
-    "num_train_examples = 5000\n",
-    "\n",
-    "\n",
-    "train_iterator = train_datagen.flow(X_train_zc[num_validation_examples:num_train_examples], \n",
-    "                                    y_train_cat[num_validation_examples:num_train_examples], \n",
-    "                                    batch_size=64)\n",
-    "validation_iterator = validation_datagen.flow(X_train_zc[:num_validation_examples:], \n",
-    "                                              y_train_cat[:num_validation_examples:], \n",
-    "                                              batch_size=64)\n",
-    "history = model.fit_generator(generator= train_iterator,  \n",
-    "                              validation_data = validation_iterator, \n",
-    "                              epochs=100, \n",
-    "                              steps_per_epoch=len(train_iterator))"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {
-    "id": "1sfknL8-vXao"
-   },
-   "outputs": [],
-   "source": [
-    "# Plot training and validation accuracy\n",
-    "acc = history.history['accuracy']\n",
-    "val_acc = history.history['val_accuracy']\n",
-    "\n",
-    "loss = history.history['loss']\n",
-    "val_loss = history.history['val_loss']\n",
-    "\n",
-    "epochs_range = range(100)\n",
-    "\n",
-    "plt.figure(figsize=(8, 8))\n",
-    "plt.subplot(1, 2, 1)\n",
-    "plt.plot(epochs_range, acc, label='Training Accuracy')\n",
-    "plt.plot(epochs_range, val_acc, label='Validation Accuracy')\n",
-    "plt.legend(loc='lower right')\n",
-    "plt.title('Training and Validation Accuracy')\n",
-    "\n",
-    "plt.subplot(1, 2, 2)\n",
-    "plt.plot(epochs_range, loss, label='Training Loss')\n",
-    "plt.plot(epochs_range, val_loss, label='Validation Loss')\n",
-    "plt.legend(loc='upper right')\n",
-    "plt.title('Training and Validation Loss')\n",
-    "plt.show()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {
-    "id": "uW4-afX7vXas"
-   },
-   "outputs": [],
-   "source": [
-    "# Evaluate test accuracy\n",
-    "test_loss, test_accuracy = model.evaluate(X_test_zc, y_test_cat)\n",
-    "print('Accuracy on test dataset:', test_accuracy)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "id": "aEsp4N5JvXay"
-   },
-   "source": [
-    "## 7.3 Dropout\n",
-    "\n",
-    "It is most common to use a single, global $ L2 $ regularization strength that is cross-validated. It is also common to combine this with dropout applied after all layers. The value of $ p = 0.5$ is a reasonable default, but this can be \n",
-    "tuned on validation data."
+    "# Load the TensorBoard notebook extension\n",
+    "%load_ext tensorboard"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": null,
    "metadata": {
-    "id": "cecEKPOYvXaz"
+    "colab": {
+     "base_uri": "https://localhost:8080/",
+     "height": 1000
+    },
+    "id": "EuDr-lyAe9YL",
+    "outputId": "7be105fe-ee50-4101-9640-e2220e16aa96"
    },
    "outputs": [],
    "source": [
-    "# Define Network\n",
-    "model = tf.keras.Sequential([\n",
-    "        tf.keras.layers.Flatten(input_shape=(32, 32, 3)),\n",
-    "        tf.keras.layers.Dense(512, \n",
-    "                              kernel_regularizer=tf.keras.regularizers.l2(0.005)),\n",
-    "        tf.keras.layers.BatchNormalization(),\n",
-    "        tf.keras.layers.Activation(tf.nn.relu),\n",
-    "        tf.keras.layers.Dense(128, \n",
-    "                              kernel_regularizer=tf.keras.regularizers.l2(0.005)),\n",
-    "        tf.keras.layers.BatchNormalization(),\n",
-    "        tf.keras.layers.Activation(tf.nn.relu),\n",
-    "        tf. keras.layers.Dropout(0.3),\n",
-    "        tf.keras.layers.Dense(10,  \n",
-    "                          activation=tf.nn.softmax, \n",
-    "                             kernel_regularizer=tf.keras.regularizers.l2(0.005))\n",
-    "])\n",
-    "\n",
-    "# Compile Network \n",
-    "model.compile(optimizer=tf.keras.optimizers.SGD(lr=0.0001),\n",
-    "              loss='categorical_crossentropy',\n",
-    "              metrics=['accuracy'])\n",
-    "\n",
-    "num_validation_examples = 1000\n",
-    "num_train_examples = 5000\n",
-    "\n",
+    "import tensorflow as tf\n",
+    "import datetime, os\n",
     "\n",
-    "train_iterator = train_datagen.flow(X_train_zc[num_validation_examples:num_train_examples], \n",
-    "                                    y_train_cat[num_validation_examples:num_train_examples], \n",
-    "                                    batch_size=128)\n",
-    "validation_iterator = validation_datagen.flow(X_train_zc[:num_validation_examples:], \n",
-    "                                              y_train_cat[:num_validation_examples:], \n",
-    "                                              batch_size=128)\n",
+    "# Then let's create a new model and train it again.\n",
+    "# Check out TensorBoard during training (click on the \"refresh\" button to see\n",
+    "# new data).\n",
     "\n",
     "# Fit Network\n",
-    "history = model.fit_generator(generator= train_iterator,  \n",
-    "                              validation_data = validation_iterator, \n",
-    "                              epochs=200, \n",
-    "                              steps_per_epoch=len(train_iterator))"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {
-    "id": "jrCdZzd3vXa2"
-   },
-   "outputs": [],
-   "source": [
-    "# Plot training and validation accuracy\n",
-    "acc = history.history['accuracy']\n",
-    "val_acc = history.history['val_accuracy']\n",
-    "\n",
-    "loss = history.history['loss']\n",
-    "val_loss = history.history['val_loss']\n",
-    "\n",
-    "epochs_range = range(200)\n",
-    "\n",
-    "plt.figure(figsize=(8, 8))\n",
-    "plt.subplot(1, 2, 1)\n",
-    "plt.plot(epochs_range, acc, label='Training Accuracy')\n",
-    "plt.plot(epochs_range, val_acc, label='Validation Accuracy')\n",
-    "plt.legend(loc='lower right')\n",
-    "plt.title('Training and Validation Accuracy')\n",
-    "\n",
-    "plt.subplot(1, 2, 2)\n",
-    "plt.plot(epochs_range, loss, label='Training Loss')\n",
-    "plt.plot(epochs_range, val_loss, label='Validation Loss')\n",
-    "plt.legend(loc='upper right')\n",
-    "plt.title('Training and Validation Loss')\n",
-    "plt.show()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {
-    "id": "VOJYqDdXvXa5"
-   },
-   "outputs": [],
-   "source": [
-    "# Evaluate test accuracy\n",
-    "test_loss, test_accuracy = model.evaluate(X_test_zc, y_test_cat)\n",
-    "print('Accuracy on test dataset:', test_accuracy)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "id": "vc8w1Q3_vXa9"
-   },
-   "source": [
-    "# 8. Parameter Update\n",
-    "\n",
-    "In practice `Adam` is currently recommended as the default algorithm to use, and often works slightly better than RMSProp. However, it is often also worth trying SGD+Nesterov Momentum as an alternative. In addition, it is recommended to use learning rate decay. The stochastic gradient descent optimization algorithm implementation in the SGD class has an argument called `decay`. This argument is used in the time-based learning rate decay schedule equation as follows:\n",
-    "\n",
-    "LearningRate = LearningRate * 1/(1 + decay * epoch)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {
-    "id": "ZxI6XZxJvXa-"
-   },
-   "outputs": [],
-   "source": [
-    "# Define Network\n",
-    "model = tf.keras.Sequential([\n",
-    "        tf.keras.layers.Flatten(input_shape=(32, 32, 3)),\n",
-    "        tf.keras.layers.Dense(512, \n",
-    "                              kernel_regularizer=tf.keras.regularizers.l2(0.005)),\n",
-    "        tf.keras.layers.BatchNormalization(),\n",
-    "        tf.keras.layers.Activation(tf.nn.relu),\n",
-    "        tf.keras.layers.Dense(128, \n",
-    "                              kernel_regularizer=tf.keras.regularizers.l2(0.005)),\n",
-    "        tf.keras.layers.BatchNormalization(),\n",
-    "        tf.keras.layers.Activation(tf.nn.relu),\n",
-    "        tf. keras.layers.Dropout(0.3),\n",
-    "        tf.keras.layers.Dense(10,  \n",
-    "                          activation=tf.nn.softmax, \n",
-    "                             kernel_regularizer=tf.keras.regularizers.l2(0.005))\n",
-    "])\n",
-    "\n",
-    "# Compile Network \n",
-    "model.compile(optimizer=tf.keras.optimizers.Adam(lr=0.0001, \n",
-    "              decay=0.0001/10),\n",
-    "              loss='categorical_crossentropy',\n",
-    "              metrics=['accuracy'])\n",
-    "\n",
-    "num_validation_examples = 1000\n",
-    "num_train_examples = 5000\n",
-    "\n",
-    "\n",
-    "train_iterator = train_datagen.flow(X_train_zc[num_validation_examples:num_train_examples], \n",
-    "                                    y_train_cat[num_validation_examples:num_train_examples], \n",
-    "                                    batch_size=128)\n",
-    "validation_iterator = validation_datagen.flow(X_train_zc[:num_validation_examples:], \n",
-    "                                              y_train_cat[:num_validation_examples:], \n",
-    "                                              batch_size=128)\n",
+    "logdir = os.path.join(\"logs\", datetime.datetime.now().strftime(\"%Y%m%d-%H%M%S\"))\n",
+    "tensorboard_callback = tf.keras.callbacks.TensorBoard(logdir, histogram_freq=1)\n",
     "\n",
-    "# Fit Network\n",
-    "history = model.fit_generator(generator= train_iterator,  \n",
-    "                              validation_data = validation_iterator, \n",
-    "                              epochs=100, \n",
-    "                              steps_per_epoch=len(train_iterator))"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {
-    "id": "cixRpMMOvXbB"
-   },
-   "outputs": [],
-   "source": [
-    "# Plot training and validation accuracy\n",
-    "acc = history.history['accuracy']\n",
-    "val_acc = history.history['val_accuracy']\n",
-    "\n",
-    "loss = history.history['loss']\n",
-    "val_loss = history.history['val_loss']\n",
-    "\n",
-    "epochs_range = range(100)\n",
-    "\n",
-    "plt.figure(figsize=(8, 8))\n",
-    "plt.subplot(1, 2, 1)\n",
-    "plt.plot(epochs_range, acc, label='Training Accuracy')\n",
-    "plt.plot(epochs_range, val_acc, label='Validation Accuracy')\n",
-    "plt.legend(loc='lower right')\n",
-    "plt.title('Training and Validation Accuracy')\n",
-    "\n",
-    "plt.subplot(1, 2, 2)\n",
-    "plt.plot(epochs_range, loss, label='Training Loss')\n",
-    "plt.plot(epochs_range, val_loss, label='Validation Loss')\n",
-    "plt.legend(loc='upper right')\n",
-    "plt.title('Training and Validation Loss')\n",
-    "plt.show()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {
-    "id": "eCuOiek7vXbQ"
-   },
-   "outputs": [],
-   "source": [
-    "# Evaluate test accuracy\n",
-    "test_loss, test_accuracy = model.evaluate(X_test_zc, y_test_cat)\n",
-    "print('Accuracy on test dataset:', test_accuracy)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "id": "U4S4eZcArngs"
-   },
-   "source": [
-    "# 9. Save Best Model on Google Drive and Reload it"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {
-    "id": "lZ_eBernxdYc"
-   },
-   "outputs": [],
-   "source": [
-    "# Install the PyDrive wrapper & import libraries.\n",
-    "# This only needs to be done once in a notebook.\n",
-    "!pip install -U -q PyDrive\n",
-    "from pydrive.auth import GoogleAuth\n",
-    "from pydrive.drive import GoogleDrive\n",
-    "from google.colab import auth\n",
-    "from oauth2client.client import GoogleCredentials\n",
-    "\n",
-    "# Authenticate and create the PyDrive client.\n",
-    "# This only needs to be done once in a notebook.\n",
-    "auth.authenticate_user()\n",
-    "gauth = GoogleAuth()\n",
-    "gauth.credentials = GoogleCredentials.get_application_default()\n",
-    "drive = GoogleDrive(gauth)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {
-    "id": "a4wwPUFhxpjw"
-   },
-   "outputs": [],
-   "source": [
-    "!curl https://raw.githubusercontent.com/dexterfichuk/GoogleDriveCheckpoint/master/google_drive_checkpoint.py -O"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {
-    "id": "7pT6GX1lxyg2"
-   },
-   "outputs": [],
-   "source": [
-    "from google_drive_checkpoint import GoogleDriveCheckpoint\n",
-    "\n",
-    "checkpoint = GoogleDriveCheckpoint('best_model.h5', drive, save_best_only=True,  verbose=1)\n",
-    "# Fit Network\n",
-    "history = model.fit_generator(generator= train_iterator,  \n",
-    "                              validation_data = validation_iterator, \n",
-    "                              epochs=100, \n",
-    "                              steps_per_epoch=len(train_iterator),callbacks=[checkpoint])"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {
-    "id": "XW4eFU5xGv53"
-   },
-   "outputs": [],
-   "source": [
-    "from google.colab import drive\n",
-    "from tensorflow.keras.models import load_model\n",
-    "\n",
-    "drive.mount('/content/gdrive')\n",
-    "\n",
-    "# load model : go on your google drive account and get the best model that was saved\n",
-    "model = load_model('/content/gdrive/My Drive/best_model-1.939315915107727.h5')\n",
-    "model.summary()\n",
-    "      \n",
-    "\n",
-    "\n"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {
-    "id": "Y_QKt2hcFDY8"
-   },
-   "outputs": [],
-   "source": [
-    "# Plot training and validation accuracy\n",
-    "acc = history.history['accuracy']\n",
-    "val_acc = history.history['val_accuracy']\n",
-    "\n",
-    "loss = history.history['loss']\n",
-    "val_loss = history.history['val_loss']\n",
-    "\n",
-    "epochs_range = range(100)\n",
-    "\n",
-    "plt.figure(figsize=(8, 8))\n",
-    "plt.subplot(1, 2, 1)\n",
-    "plt.plot(epochs_range, acc, label='Training Accuracy')\n",
-    "plt.plot(epochs_range, val_acc, label='Validation Accuracy')\n",
-    "plt.legend(loc='lower right')\n",
-    "plt.title('Training and Validation Accuracy')\n",
-    "\n",
-    "plt.subplot(1, 2, 2)\n",
-    "plt.plot(epochs_range, loss, label='Training Loss')\n",
-    "plt.plot(epochs_range, val_loss, label='Validation Loss')\n",
-    "plt.legend(loc='upper right')\n",
-    "plt.title('Training and Validation Loss')\n",
-    "plt.show()\n"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "id": "5zgbDpH0sH7u"
-   },
-   "source": [
-    "# 10. TensorBoard\n",
-    "\n",
-    "TensorBoard is a great tool to observe variables during training (especially useful for models training a long time, like above model).\n"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {
-    "colab": {
-     "base_uri": "https://localhost:8080/",
-     "height": 51
-    },
-    "id": "FP5b7DGle7to",
-    "outputId": "9be8f4da-e6a9-45ab-aa2c-0c5a4e994055"
-   },
-   "outputs": [],
-   "source": [
-    "# Load the TensorBoard notebook extension\n",
-    "%load_ext tensorboard"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {
-    "colab": {
-     "base_uri": "https://localhost:8080/",
-     "height": 1000
-    },
-    "id": "EuDr-lyAe9YL",
-    "outputId": "7be105fe-ee50-4101-9640-e2220e16aa96"
-   },
-   "outputs": [],
-   "source": [
-    "import tensorflow as tf\n",
-    "import datetime, os\n",
-    "\n",
-    "# Then let's create a new model and train it again.\n",
-    "# Check out TensorBoard during training (click on the \"refresh\" button to see\n",
-    "# new data).\n",
-    "\n",
-    "# Fit Network\n",
-    "logdir = os.path.join(\"logs\", datetime.datetime.now().strftime(\"%Y%m%d-%H%M%S\"))\n",
-    "tensorboard_callback = tf.keras.callbacks.TensorBoard(logdir, histogram_freq=1)\n",
-    "\n",
-    "history = model.fit_generator(generator= train_iterator,  validation_data = validation_iterator, epochs=100, steps_per_epoch=len(train_iterator), callbacks=[tensorboard_callback])\n"
+    "history = model.fit_generator(generator= train_iterator,  validation_data = validation_iterator, epochs=100, steps_per_epoch=len(train_iterator), callbacks=[tensorboard_callback])\n"
    ]
   },
   {
@@ -4742,14 +4156,1312 @@
     "%tensorboard --logdir logs"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "id": "XBZuaco_vXaj"
+   },
+   "source": [
+    "# 8. Regularization\n",
+    "\n",
+    "## 8.1 L1 Regularization"
+   ]
+  },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 49,
    "metadata": {
-    "id": "izJpFarYgZOP"
+    "id": "mIGlhfVQvXak"
    },
-   "outputs": [],
-   "source": []
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/opt/conda/lib/python3.7/site-packages/ipykernel_launcher.py:36: UserWarning: `Model.fit_generator` is deprecated and will be removed in a future version. Please use `Model.fit`, which supports generators.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 1/100\n",
+      "63/63 [==============================] - 6s 85ms/step - loss: 327.1515 - accuracy: 0.0960 - val_loss: 327.0746 - val_accuracy: 0.1080\n",
+      "Epoch 2/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 326.1475 - accuracy: 0.1002 - val_loss: 325.6567 - val_accuracy: 0.1100\n",
+      "Epoch 3/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 325.1220 - accuracy: 0.1063 - val_loss: 324.5860 - val_accuracy: 0.1250\n",
+      "Epoch 4/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 324.0782 - accuracy: 0.1168 - val_loss: 323.5659 - val_accuracy: 0.1260\n",
+      "Epoch 5/100\n",
+      "63/63 [==============================] - 5s 85ms/step - loss: 323.1131 - accuracy: 0.1072 - val_loss: 322.5623 - val_accuracy: 0.1300\n",
+      "Epoch 6/100\n",
+      "63/63 [==============================] - 5s 85ms/step - loss: 322.0774 - accuracy: 0.1260 - val_loss: 321.5606 - val_accuracy: 0.1380\n",
+      "Epoch 7/100\n",
+      "63/63 [==============================] - 6s 93ms/step - loss: 321.0746 - accuracy: 0.1195 - val_loss: 320.5585 - val_accuracy: 0.1440\n",
+      "Epoch 8/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 320.0763 - accuracy: 0.1325 - val_loss: 319.5595 - val_accuracy: 0.1530\n",
+      "Epoch 9/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 319.1026 - accuracy: 0.1322 - val_loss: 318.5633 - val_accuracy: 0.1550\n",
+      "Epoch 10/100\n",
+      "63/63 [==============================] - 5s 84ms/step - loss: 318.0901 - accuracy: 0.1482 - val_loss: 317.5669 - val_accuracy: 0.1590\n",
+      "Epoch 11/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 317.0896 - accuracy: 0.1462 - val_loss: 316.5726 - val_accuracy: 0.1630\n",
+      "Epoch 12/100\n",
+      "63/63 [==============================] - 5s 80ms/step - loss: 316.0988 - accuracy: 0.1583 - val_loss: 315.5802 - val_accuracy: 0.1620\n",
+      "Epoch 13/100\n",
+      "63/63 [==============================] - 5s 84ms/step - loss: 315.1066 - accuracy: 0.1567 - val_loss: 314.5958 - val_accuracy: 0.1670\n",
+      "Epoch 14/100\n",
+      "63/63 [==============================] - 5s 81ms/step - loss: 314.1442 - accuracy: 0.1462 - val_loss: 313.6080 - val_accuracy: 0.1690\n",
+      "Epoch 15/100\n",
+      "63/63 [==============================] - 5s 81ms/step - loss: 313.1559 - accuracy: 0.1695 - val_loss: 312.6242 - val_accuracy: 0.1690\n",
+      "Epoch 16/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 312.1642 - accuracy: 0.1758 - val_loss: 311.6449 - val_accuracy: 0.1730\n",
+      "Epoch 17/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 311.1918 - accuracy: 0.1663 - val_loss: 310.6667 - val_accuracy: 0.1800\n",
+      "Epoch 18/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 310.1902 - accuracy: 0.1840 - val_loss: 309.6874 - val_accuracy: 0.1810\n",
+      "Epoch 19/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 309.2360 - accuracy: 0.1688 - val_loss: 308.7123 - val_accuracy: 0.1860\n",
+      "Epoch 20/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 308.2580 - accuracy: 0.1877 - val_loss: 307.7401 - val_accuracy: 0.1950\n",
+      "Epoch 21/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 307.2866 - accuracy: 0.1785 - val_loss: 306.7669 - val_accuracy: 0.1980\n",
+      "Epoch 22/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 306.3173 - accuracy: 0.1887 - val_loss: 305.7998 - val_accuracy: 0.1990\n",
+      "Epoch 23/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 305.3628 - accuracy: 0.1883 - val_loss: 304.8331 - val_accuracy: 0.1990\n",
+      "Epoch 24/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 304.4003 - accuracy: 0.1898 - val_loss: 303.8682 - val_accuracy: 0.2000\n",
+      "Epoch 25/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 303.4365 - accuracy: 0.1935 - val_loss: 302.9048 - val_accuracy: 0.2000\n",
+      "Epoch 26/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 302.4757 - accuracy: 0.2015 - val_loss: 301.9421 - val_accuracy: 0.2070\n",
+      "Epoch 27/100\n",
+      "63/63 [==============================] - 5s 81ms/step - loss: 301.4982 - accuracy: 0.2060 - val_loss: 300.9833 - val_accuracy: 0.2120\n",
+      "Epoch 28/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 300.5380 - accuracy: 0.2023 - val_loss: 300.0252 - val_accuracy: 0.2160\n",
+      "Epoch 29/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 299.5988 - accuracy: 0.2048 - val_loss: 299.0702 - val_accuracy: 0.2160\n",
+      "Epoch 30/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 298.6487 - accuracy: 0.1965 - val_loss: 298.1189 - val_accuracy: 0.2200\n",
+      "Epoch 31/100\n",
+      "63/63 [==============================] - 5s 84ms/step - loss: 297.6784 - accuracy: 0.2107 - val_loss: 297.1659 - val_accuracy: 0.2220\n",
+      "Epoch 32/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 296.7291 - accuracy: 0.2093 - val_loss: 296.2149 - val_accuracy: 0.2280\n",
+      "Epoch 33/100\n",
+      "63/63 [==============================] - 6s 94ms/step - loss: 295.7881 - accuracy: 0.2128 - val_loss: 295.2669 - val_accuracy: 0.2320\n",
+      "Epoch 34/100\n",
+      "63/63 [==============================] - 5s 85ms/step - loss: 294.8321 - accuracy: 0.2130 - val_loss: 294.3218 - val_accuracy: 0.2350\n",
+      "Epoch 35/100\n",
+      "63/63 [==============================] - 5s 84ms/step - loss: 293.8979 - accuracy: 0.2233 - val_loss: 293.3750 - val_accuracy: 0.2420\n",
+      "Epoch 36/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 292.9487 - accuracy: 0.2188 - val_loss: 292.4310 - val_accuracy: 0.2410\n",
+      "Epoch 37/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 292.0022 - accuracy: 0.2165 - val_loss: 291.4884 - val_accuracy: 0.2420\n",
+      "Epoch 38/100\n",
+      "63/63 [==============================] - 5s 81ms/step - loss: 291.0607 - accuracy: 0.2265 - val_loss: 290.5476 - val_accuracy: 0.2450\n",
+      "Epoch 39/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 290.1371 - accuracy: 0.2265 - val_loss: 289.6096 - val_accuracy: 0.2460\n",
+      "Epoch 40/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 289.1928 - accuracy: 0.2247 - val_loss: 288.6717 - val_accuracy: 0.2480\n",
+      "Epoch 41/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 288.2564 - accuracy: 0.2237 - val_loss: 287.7351 - val_accuracy: 0.2490\n",
+      "Epoch 42/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 287.3262 - accuracy: 0.2233 - val_loss: 286.8005 - val_accuracy: 0.2520\n",
+      "Epoch 43/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 286.4016 - accuracy: 0.2202 - val_loss: 285.8700 - val_accuracy: 0.2520\n",
+      "Epoch 44/100\n",
+      "63/63 [==============================] - 5s 81ms/step - loss: 285.4873 - accuracy: 0.2153 - val_loss: 284.9394 - val_accuracy: 0.2540\n",
+      "Epoch 45/100\n",
+      "63/63 [==============================] - 5s 81ms/step - loss: 284.5251 - accuracy: 0.2323 - val_loss: 284.0109 - val_accuracy: 0.2590\n",
+      "Epoch 46/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 283.6101 - accuracy: 0.2225 - val_loss: 283.0848 - val_accuracy: 0.2600\n",
+      "Epoch 47/100\n",
+      "63/63 [==============================] - 5s 80ms/step - loss: 282.6856 - accuracy: 0.2325 - val_loss: 282.1608 - val_accuracy: 0.2640\n",
+      "Epoch 48/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 281.7592 - accuracy: 0.2380 - val_loss: 281.2372 - val_accuracy: 0.2650\n",
+      "Epoch 49/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 280.8422 - accuracy: 0.2350 - val_loss: 280.3164 - val_accuracy: 0.2640\n",
+      "Epoch 50/100\n",
+      "63/63 [==============================] - 5s 84ms/step - loss: 279.9146 - accuracy: 0.2407 - val_loss: 279.3953 - val_accuracy: 0.2650\n",
+      "Epoch 51/100\n",
+      "63/63 [==============================] - 5s 81ms/step - loss: 278.9896 - accuracy: 0.2327 - val_loss: 278.4753 - val_accuracy: 0.2670\n",
+      "Epoch 52/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 278.0770 - accuracy: 0.2395 - val_loss: 277.5576 - val_accuracy: 0.2660\n",
+      "Epoch 53/100\n",
+      "63/63 [==============================] - 5s 81ms/step - loss: 277.1502 - accuracy: 0.2457 - val_loss: 276.6424 - val_accuracy: 0.2720\n",
+      "Epoch 54/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 276.2343 - accuracy: 0.2425 - val_loss: 275.7294 - val_accuracy: 0.2710\n",
+      "Epoch 55/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 275.3232 - accuracy: 0.2545 - val_loss: 274.8169 - val_accuracy: 0.2720\n",
+      "Epoch 56/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 274.4250 - accuracy: 0.2455 - val_loss: 273.9061 - val_accuracy: 0.2730\n",
+      "Epoch 57/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 273.5044 - accuracy: 0.2512 - val_loss: 272.9958 - val_accuracy: 0.2710\n",
+      "Epoch 58/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 272.6038 - accuracy: 0.2463 - val_loss: 272.0880 - val_accuracy: 0.2700\n",
+      "Epoch 59/100\n",
+      "63/63 [==============================] - 5s 80ms/step - loss: 271.6734 - accuracy: 0.2515 - val_loss: 271.1808 - val_accuracy: 0.2740\n",
+      "Epoch 60/100\n",
+      "63/63 [==============================] - 5s 81ms/step - loss: 270.7823 - accuracy: 0.2498 - val_loss: 270.2771 - val_accuracy: 0.2740\n",
+      "Epoch 61/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 269.8884 - accuracy: 0.2477 - val_loss: 269.3727 - val_accuracy: 0.2760\n",
+      "Epoch 62/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 269.0027 - accuracy: 0.2435 - val_loss: 268.4718 - val_accuracy: 0.2790\n",
+      "Epoch 63/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 268.0920 - accuracy: 0.2503 - val_loss: 267.5706 - val_accuracy: 0.2780\n",
+      "Epoch 64/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 267.1882 - accuracy: 0.2528 - val_loss: 266.6737 - val_accuracy: 0.2790\n",
+      "Epoch 65/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 266.2816 - accuracy: 0.2537 - val_loss: 265.7779 - val_accuracy: 0.2780\n",
+      "Epoch 66/100\n",
+      "63/63 [==============================] - 5s 81ms/step - loss: 265.3962 - accuracy: 0.2492 - val_loss: 264.8836 - val_accuracy: 0.2810\n",
+      "Epoch 67/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 264.5002 - accuracy: 0.2562 - val_loss: 263.9904 - val_accuracy: 0.2810\n",
+      "Epoch 68/100\n",
+      "63/63 [==============================] - 5s 84ms/step - loss: 263.6219 - accuracy: 0.2465 - val_loss: 263.0982 - val_accuracy: 0.2810\n",
+      "Epoch 69/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 262.7169 - accuracy: 0.2623 - val_loss: 262.2080 - val_accuracy: 0.2800\n",
+      "Epoch 70/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 261.8205 - accuracy: 0.2620 - val_loss: 261.3180 - val_accuracy: 0.2870\n",
+      "Epoch 71/100\n",
+      "63/63 [==============================] - 5s 81ms/step - loss: 260.9346 - accuracy: 0.2650 - val_loss: 260.4309 - val_accuracy: 0.2830\n",
+      "Epoch 72/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 260.0492 - accuracy: 0.2580 - val_loss: 259.5452 - val_accuracy: 0.2820\n",
+      "Epoch 73/100\n",
+      "63/63 [==============================] - 5s 80ms/step - loss: 259.1770 - accuracy: 0.2675 - val_loss: 258.6613 - val_accuracy: 0.2830\n",
+      "Epoch 74/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 258.3004 - accuracy: 0.2542 - val_loss: 257.7794 - val_accuracy: 0.2840\n",
+      "Epoch 75/100\n",
+      "63/63 [==============================] - 5s 80ms/step - loss: 257.4110 - accuracy: 0.2545 - val_loss: 256.8986 - val_accuracy: 0.2860\n",
+      "Epoch 76/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 256.5139 - accuracy: 0.2603 - val_loss: 256.0189 - val_accuracy: 0.2850\n",
+      "Epoch 77/100\n",
+      "63/63 [==============================] - 5s 79ms/step - loss: 255.6440 - accuracy: 0.2567 - val_loss: 255.1406 - val_accuracy: 0.2870\n",
+      "Epoch 78/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 254.7783 - accuracy: 0.2663 - val_loss: 254.2640 - val_accuracy: 0.2860\n",
+      "Epoch 79/100\n",
+      "63/63 [==============================] - 5s 81ms/step - loss: 253.9017 - accuracy: 0.2610 - val_loss: 253.3907 - val_accuracy: 0.2860\n",
+      "Epoch 80/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 253.0186 - accuracy: 0.2618 - val_loss: 252.5173 - val_accuracy: 0.2880\n",
+      "Epoch 81/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 252.1360 - accuracy: 0.2767 - val_loss: 251.6447 - val_accuracy: 0.2870\n",
+      "Epoch 82/100\n",
+      "63/63 [==============================] - 5s 81ms/step - loss: 251.2754 - accuracy: 0.2555 - val_loss: 250.7763 - val_accuracy: 0.2900\n",
+      "Epoch 83/100\n",
+      "63/63 [==============================] - 5s 84ms/step - loss: 250.4181 - accuracy: 0.2623 - val_loss: 249.9071 - val_accuracy: 0.2910\n",
+      "Epoch 84/100\n",
+      "63/63 [==============================] - 6s 93ms/step - loss: 249.5337 - accuracy: 0.2705 - val_loss: 249.0408 - val_accuracy: 0.2970\n",
+      "Epoch 85/100\n",
+      "63/63 [==============================] - 5s 81ms/step - loss: 248.6617 - accuracy: 0.2780 - val_loss: 248.1743 - val_accuracy: 0.2940\n",
+      "Epoch 86/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 247.8124 - accuracy: 0.2727 - val_loss: 247.3114 - val_accuracy: 0.2960\n",
+      "Epoch 87/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 246.9591 - accuracy: 0.2640 - val_loss: 246.4467 - val_accuracy: 0.2950\n",
+      "Epoch 88/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 246.0869 - accuracy: 0.2763 - val_loss: 245.5876 - val_accuracy: 0.2940\n",
+      "Epoch 89/100\n",
+      "63/63 [==============================] - 5s 81ms/step - loss: 245.2222 - accuracy: 0.2780 - val_loss: 244.7277 - val_accuracy: 0.2990\n",
+      "Epoch 90/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 244.3685 - accuracy: 0.2760 - val_loss: 243.8696 - val_accuracy: 0.2990\n",
+      "Epoch 91/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 243.5253 - accuracy: 0.2772 - val_loss: 243.0142 - val_accuracy: 0.2990\n",
+      "Epoch 92/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 242.6632 - accuracy: 0.2697 - val_loss: 242.1590 - val_accuracy: 0.3050\n",
+      "Epoch 93/100\n",
+      "63/63 [==============================] - 5s 79ms/step - loss: 241.8077 - accuracy: 0.2780 - val_loss: 241.3056 - val_accuracy: 0.3050\n",
+      "Epoch 94/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 240.9498 - accuracy: 0.2770 - val_loss: 240.4554 - val_accuracy: 0.3020\n",
+      "Epoch 95/100\n",
+      "63/63 [==============================] - 5s 80ms/step - loss: 240.1183 - accuracy: 0.2685 - val_loss: 239.6053 - val_accuracy: 0.3010\n",
+      "Epoch 96/100\n",
+      "63/63 [==============================] - 5s 82ms/step - loss: 239.2501 - accuracy: 0.2785 - val_loss: 238.7564 - val_accuracy: 0.3030\n",
+      "Epoch 97/100\n",
+      "63/63 [==============================] - 5s 81ms/step - loss: 238.4097 - accuracy: 0.2735 - val_loss: 237.9107 - val_accuracy: 0.3000\n",
+      "Epoch 98/100\n",
+      "63/63 [==============================] - 5s 81ms/step - loss: 237.5537 - accuracy: 0.2718 - val_loss: 237.0653 - val_accuracy: 0.3050\n",
+      "Epoch 99/100\n",
+      "63/63 [==============================] - 5s 83ms/step - loss: 236.7005 - accuracy: 0.2865 - val_loss: 236.2210 - val_accuracy: 0.3030\n",
+      "Epoch 100/100\n",
+      "63/63 [==============================] - 5s 81ms/step - loss: 235.8721 - accuracy: 0.2788 - val_loss: 235.3800 - val_accuracy: 0.3030\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Define Network\n",
+    "model = tf.keras.Sequential([\n",
+    "        tf.keras.layers.Flatten(input_shape=(32, 32, 3)),\n",
+    "        tf.keras.layers.Dense(512, \n",
+    "                              kernel_regularizer=tf.keras.regularizers.l1(0.01)),\n",
+    "        tf.keras.layers.BatchNormalization(),\n",
+    "        tf.keras.layers.Activation(tf.nn.relu),\n",
+    "        tf.keras.layers.Dense(128, \n",
+    "                              kernel_regularizer=tf.keras.regularizers.l1(0.001)),\n",
+    "        tf.keras.layers.BatchNormalization(),\n",
+    "        tf.keras.layers.Activation(tf.nn.relu),\n",
+    "        tf.keras.layers.Dense(10,  \n",
+    "                              activation=tf.nn.softmax,  \n",
+    "                              kernel_regularizer=tf.keras.regularizers.l1(0.001))\n",
+    "])\n",
+    "\n",
+    "# Compile Network\n",
+    "model.compile(optimizer=tf.keras.optimizers.SGD(lr=0.0001),\n",
+    "              loss='categorical_crossentropy',\n",
+    "              metrics=['accuracy'])\n",
+    "\n",
+    "# Fit Network\n",
+    "num_validation_examples = 1000\n",
+    "num_train_examples = 5000\n",
+    "\n",
+    "\n",
+    "train_iterator = train_datagen.flow(X_train_zc[num_validation_examples:num_train_examples], \n",
+    "                                    y_train_cat[num_validation_examples:num_train_examples], \n",
+    "                                    batch_size=64)\n",
+    "validation_iterator = validation_datagen.flow(X_train_zc[:num_validation_examples:], \n",
+    "                                              y_train_cat[:num_validation_examples:], \n",
+    "                                              batch_size=64)\n",
+    "history = model.fit_generator(generator= train_iterator,  \n",
+    "                              validation_data = validation_iterator, \n",
+    "                              epochs=100, \n",
+    "                              steps_per_epoch=len(train_iterator))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 50,
+   "metadata": {
+    "id": "1sfknL8-vXao"
+   },
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 576x576 with 2 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Plot training and validation accuracy\n",
+    "acc = history.history['accuracy']\n",
+    "val_acc = history.history['val_accuracy']\n",
+    "\n",
+    "loss = history.history['loss']\n",
+    "val_loss = history.history['val_loss']\n",
+    "\n",
+    "epochs_range = range(100)\n",
+    "\n",
+    "plt.figure(figsize=(8, 8))\n",
+    "plt.subplot(1, 2, 1)\n",
+    "plt.plot(epochs_range, acc, label='Training Accuracy')\n",
+    "plt.plot(epochs_range, val_acc, label='Validation Accuracy')\n",
+    "plt.legend(loc='lower right')\n",
+    "plt.title('Training and Validation Accuracy')\n",
+    "\n",
+    "plt.subplot(1, 2, 2)\n",
+    "plt.plot(epochs_range, loss, label='Training Loss')\n",
+    "plt.plot(epochs_range, val_loss, label='Validation Loss')\n",
+    "plt.legend(loc='upper right')\n",
+    "plt.title('Training and Validation Loss')\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 51,
+   "metadata": {
+    "id": "uW4-afX7vXas"
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "313/313 [==============================] - 5s 14ms/step - loss: 235.4020 - accuracy: 0.2974\n",
+      "Accuracy on test dataset: 0.29739999771118164\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Evaluate test accuracy\n",
+    "test_loss, test_accuracy = model.evaluate(X_test_zc, y_test_cat)\n",
+    "print('Accuracy on test dataset:', test_accuracy)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "id": "aEsp4N5JvXay"
+   },
+   "source": [
+    "## 8.2 Dropout\n",
+    "\n",
+    "It is most common to use a single, global $ L2 $ regularization strength that is cross-validated. It is also common to combine this with dropout applied after all layers. The value of $ p = 0.5$ is a reasonable default, but this can be \n",
+    "tuned on validation data."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "id": "cecEKPOYvXaz"
+   },
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/opt/conda/lib/python3.7/site-packages/ipykernel_launcher.py:38: UserWarning: `Model.fit_generator` is deprecated and will be removed in a future version. Please use `Model.fit`, which supports generators.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 1/200\n",
+      "32/32 [==============================] - 5s 126ms/step - loss: 8.2717 - accuracy: 0.1015 - val_loss: 8.9044 - val_accuracy: 0.0990\n",
+      "Epoch 2/200\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 8.2294 - accuracy: 0.1115 - val_loss: 8.3683 - val_accuracy: 0.0990\n",
+      "Epoch 3/200\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 8.2034 - accuracy: 0.1115 - val_loss: 8.1730 - val_accuracy: 0.1110\n",
+      "Epoch 4/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 8.2173 - accuracy: 0.1050 - val_loss: 8.0694 - val_accuracy: 0.1150\n",
+      "Epoch 5/200\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 8.1576 - accuracy: 0.1238 - val_loss: 8.0118 - val_accuracy: 0.1300\n",
+      "Epoch 6/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 8.1675 - accuracy: 0.1182 - val_loss: 7.9765 - val_accuracy: 0.1280\n",
+      "Epoch 7/200\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 8.1499 - accuracy: 0.1203 - val_loss: 7.9521 - val_accuracy: 0.1300\n",
+      "Epoch 8/200\n",
+      "32/32 [==============================] - 4s 119ms/step - loss: 8.1334 - accuracy: 0.1165 - val_loss: 7.9339 - val_accuracy: 0.1300\n",
+      "Epoch 9/200\n",
+      "32/32 [==============================] - 4s 119ms/step - loss: 8.1110 - accuracy: 0.1343 - val_loss: 7.9190 - val_accuracy: 0.1360\n",
+      "Epoch 10/200\n",
+      "32/32 [==============================] - 4s 118ms/step - loss: 8.1160 - accuracy: 0.1268 - val_loss: 7.9076 - val_accuracy: 0.1400\n",
+      "Epoch 11/200\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 8.1144 - accuracy: 0.1273 - val_loss: 7.8954 - val_accuracy: 0.1480\n",
+      "Epoch 12/200\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 8.0868 - accuracy: 0.1258 - val_loss: 7.8821 - val_accuracy: 0.1470\n",
+      "Epoch 13/200\n",
+      "32/32 [==============================] - 4s 143ms/step - loss: 8.0794 - accuracy: 0.1350 - val_loss: 7.8711 - val_accuracy: 0.1510\n",
+      "Epoch 14/200\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 8.0743 - accuracy: 0.1357 - val_loss: 7.8607 - val_accuracy: 0.1520\n",
+      "Epoch 15/200\n",
+      "32/32 [==============================] - 4s 116ms/step - loss: 8.0526 - accuracy: 0.1402 - val_loss: 7.8506 - val_accuracy: 0.1550\n",
+      "Epoch 16/200\n",
+      "32/32 [==============================] - 4s 117ms/step - loss: 8.0373 - accuracy: 0.1380 - val_loss: 7.8432 - val_accuracy: 0.1580\n",
+      "Epoch 17/200\n",
+      "32/32 [==============================] - 4s 113ms/step - loss: 8.0570 - accuracy: 0.1437 - val_loss: 7.8320 - val_accuracy: 0.1640\n",
+      "Epoch 18/200\n",
+      "32/32 [==============================] - 4s 117ms/step - loss: 8.0218 - accuracy: 0.1465 - val_loss: 7.8245 - val_accuracy: 0.1690\n",
+      "Epoch 19/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 8.0468 - accuracy: 0.1435 - val_loss: 7.8172 - val_accuracy: 0.1720\n",
+      "Epoch 20/200\n",
+      "32/32 [==============================] - 4s 118ms/step - loss: 8.0108 - accuracy: 0.1545 - val_loss: 7.8070 - val_accuracy: 0.1800\n",
+      "Epoch 21/200\n",
+      "32/32 [==============================] - 4s 117ms/step - loss: 7.9958 - accuracy: 0.1507 - val_loss: 7.7972 - val_accuracy: 0.1790\n",
+      "Epoch 22/200\n",
+      "32/32 [==============================] - 4s 114ms/step - loss: 7.9712 - accuracy: 0.1612 - val_loss: 7.7887 - val_accuracy: 0.1850\n",
+      "Epoch 23/200\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 8.0070 - accuracy: 0.1497 - val_loss: 7.7790 - val_accuracy: 0.1930\n",
+      "Epoch 24/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 7.9934 - accuracy: 0.1550 - val_loss: 7.7713 - val_accuracy: 0.1910\n",
+      "Epoch 25/200\n",
+      "32/32 [==============================] - 4s 119ms/step - loss: 7.9845 - accuracy: 0.1538 - val_loss: 7.7629 - val_accuracy: 0.1950\n",
+      "Epoch 26/200\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 7.9644 - accuracy: 0.1585 - val_loss: 7.7543 - val_accuracy: 0.1960\n",
+      "Epoch 27/200\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 7.9780 - accuracy: 0.1612 - val_loss: 7.7470 - val_accuracy: 0.1970\n",
+      "Epoch 28/200\n",
+      "32/32 [==============================] - 4s 117ms/step - loss: 7.9485 - accuracy: 0.1673 - val_loss: 7.7399 - val_accuracy: 0.1980\n",
+      "Epoch 29/200\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 7.9730 - accuracy: 0.1545 - val_loss: 7.7329 - val_accuracy: 0.1980\n",
+      "Epoch 30/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 7.9604 - accuracy: 0.1690 - val_loss: 7.7283 - val_accuracy: 0.1980\n",
+      "Epoch 31/200\n",
+      "32/32 [==============================] - 4s 119ms/step - loss: 7.9444 - accuracy: 0.1620 - val_loss: 7.7224 - val_accuracy: 0.2010\n",
+      "Epoch 32/200\n",
+      "32/32 [==============================] - 4s 117ms/step - loss: 7.9592 - accuracy: 0.1632 - val_loss: 7.7158 - val_accuracy: 0.1990\n",
+      "Epoch 33/200\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 7.9515 - accuracy: 0.1723 - val_loss: 7.7093 - val_accuracy: 0.2000\n",
+      "Epoch 34/200\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 7.9370 - accuracy: 0.1745 - val_loss: 7.7043 - val_accuracy: 0.2020\n",
+      "Epoch 35/200\n",
+      "32/32 [==============================] - 4s 117ms/step - loss: 7.9305 - accuracy: 0.1653 - val_loss: 7.6985 - val_accuracy: 0.2070\n",
+      "Epoch 36/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 7.9209 - accuracy: 0.1835 - val_loss: 7.6933 - val_accuracy: 0.2060\n",
+      "Epoch 37/200\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 7.9326 - accuracy: 0.1765 - val_loss: 7.6889 - val_accuracy: 0.2090\n",
+      "Epoch 38/200\n",
+      "32/32 [==============================] - 4s 119ms/step - loss: 7.9351 - accuracy: 0.1700 - val_loss: 7.6836 - val_accuracy: 0.2090\n",
+      "Epoch 39/200\n",
+      "32/32 [==============================] - 4s 125ms/step - loss: 7.9061 - accuracy: 0.1780 - val_loss: 7.6788 - val_accuracy: 0.2090\n",
+      "Epoch 40/200\n",
+      "32/32 [==============================] - 4s 115ms/step - loss: 7.9031 - accuracy: 0.1782 - val_loss: 7.6739 - val_accuracy: 0.2100\n",
+      "Epoch 41/200\n",
+      "32/32 [==============================] - 4s 117ms/step - loss: 7.8808 - accuracy: 0.1793 - val_loss: 7.6692 - val_accuracy: 0.2130\n",
+      "Epoch 42/200\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 7.8905 - accuracy: 0.1805 - val_loss: 7.6640 - val_accuracy: 0.2140\n",
+      "Epoch 43/200\n",
+      "32/32 [==============================] - 4s 117ms/step - loss: 7.8922 - accuracy: 0.1795 - val_loss: 7.6598 - val_accuracy: 0.2110\n",
+      "Epoch 44/200\n",
+      "32/32 [==============================] - 4s 115ms/step - loss: 7.9096 - accuracy: 0.1765 - val_loss: 7.6540 - val_accuracy: 0.2140\n",
+      "Epoch 45/200\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 7.8835 - accuracy: 0.1817 - val_loss: 7.6498 - val_accuracy: 0.2140\n",
+      "Epoch 46/200\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 7.8720 - accuracy: 0.1835 - val_loss: 7.6454 - val_accuracy: 0.2140\n",
+      "Epoch 47/200\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 7.8861 - accuracy: 0.1822 - val_loss: 7.6411 - val_accuracy: 0.2180\n",
+      "Epoch 48/200\n",
+      "32/32 [==============================] - 5s 141ms/step - loss: 7.8637 - accuracy: 0.1933 - val_loss: 7.6370 - val_accuracy: 0.2170\n",
+      "Epoch 49/200\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 7.8734 - accuracy: 0.1855 - val_loss: 7.6322 - val_accuracy: 0.2220\n",
+      "Epoch 50/200\n",
+      "32/32 [==============================] - 4s 115ms/step - loss: 7.8481 - accuracy: 0.1832 - val_loss: 7.6287 - val_accuracy: 0.2240\n",
+      "Epoch 51/200\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 7.8732 - accuracy: 0.1813 - val_loss: 7.6235 - val_accuracy: 0.2240\n",
+      "Epoch 52/200\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 7.8571 - accuracy: 0.1873 - val_loss: 7.6204 - val_accuracy: 0.2240\n",
+      "Epoch 53/200\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 7.8460 - accuracy: 0.1915 - val_loss: 7.6169 - val_accuracy: 0.2240\n",
+      "Epoch 54/200\n",
+      "32/32 [==============================] - 4s 113ms/step - loss: 7.8232 - accuracy: 0.1945 - val_loss: 7.6135 - val_accuracy: 0.2280\n",
+      "Epoch 55/200\n",
+      "32/32 [==============================] - 4s 119ms/step - loss: 7.8128 - accuracy: 0.2017 - val_loss: 7.6095 - val_accuracy: 0.2290\n",
+      "Epoch 56/200\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 7.8377 - accuracy: 0.1957 - val_loss: 7.6054 - val_accuracy: 0.2300\n",
+      "Epoch 57/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 7.8164 - accuracy: 0.2048 - val_loss: 7.6014 - val_accuracy: 0.2300\n",
+      "Epoch 58/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 7.8366 - accuracy: 0.1980 - val_loss: 7.5975 - val_accuracy: 0.2290\n",
+      "Epoch 59/200\n",
+      "32/32 [==============================] - 4s 115ms/step - loss: 7.8566 - accuracy: 0.1857 - val_loss: 7.5936 - val_accuracy: 0.2340\n",
+      "Epoch 60/200\n",
+      "32/32 [==============================] - 4s 118ms/step - loss: 7.8271 - accuracy: 0.1940 - val_loss: 7.5899 - val_accuracy: 0.2370\n",
+      "Epoch 61/200\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 7.8468 - accuracy: 0.2000 - val_loss: 7.5866 - val_accuracy: 0.2370\n",
+      "Epoch 62/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 7.8057 - accuracy: 0.2040 - val_loss: 7.5828 - val_accuracy: 0.2390\n",
+      "Epoch 63/200\n",
+      "32/32 [==============================] - 4s 117ms/step - loss: 7.8074 - accuracy: 0.2037 - val_loss: 7.5803 - val_accuracy: 0.2390\n",
+      "Epoch 64/200\n",
+      "32/32 [==============================] - 4s 119ms/step - loss: 7.8049 - accuracy: 0.2080 - val_loss: 7.5774 - val_accuracy: 0.2390\n",
+      "Epoch 65/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 7.8306 - accuracy: 0.1950 - val_loss: 7.5748 - val_accuracy: 0.2410\n",
+      "Epoch 66/200\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 7.7938 - accuracy: 0.2050 - val_loss: 7.5715 - val_accuracy: 0.2400\n",
+      "Epoch 67/200\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 7.8254 - accuracy: 0.1912 - val_loss: 7.5689 - val_accuracy: 0.2420\n",
+      "Epoch 68/200\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 7.8172 - accuracy: 0.1980 - val_loss: 7.5662 - val_accuracy: 0.2380\n",
+      "Epoch 69/200\n",
+      "32/32 [==============================] - 4s 117ms/step - loss: 7.8092 - accuracy: 0.2020 - val_loss: 7.5634 - val_accuracy: 0.2380\n",
+      "Epoch 70/200\n",
+      "32/32 [==============================] - 4s 116ms/step - loss: 7.8169 - accuracy: 0.1947 - val_loss: 7.5605 - val_accuracy: 0.2400\n",
+      "Epoch 71/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 7.8063 - accuracy: 0.2083 - val_loss: 7.5579 - val_accuracy: 0.2430\n",
+      "Epoch 72/200\n",
+      "32/32 [==============================] - 4s 119ms/step - loss: 7.7876 - accuracy: 0.1998 - val_loss: 7.5548 - val_accuracy: 0.2430\n",
+      "Epoch 73/200\n",
+      "32/32 [==============================] - 4s 115ms/step - loss: 7.7891 - accuracy: 0.2000 - val_loss: 7.5528 - val_accuracy: 0.2440\n",
+      "Epoch 74/200\n",
+      "32/32 [==============================] - 4s 118ms/step - loss: 7.7635 - accuracy: 0.2188 - val_loss: 7.5502 - val_accuracy: 0.2440\n",
+      "Epoch 75/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 7.7661 - accuracy: 0.2190 - val_loss: 7.5471 - val_accuracy: 0.2470\n",
+      "Epoch 76/200\n",
+      "32/32 [==============================] - 4s 118ms/step - loss: 7.7862 - accuracy: 0.2050 - val_loss: 7.5440 - val_accuracy: 0.2480\n",
+      "Epoch 77/200\n",
+      "32/32 [==============================] - 4s 117ms/step - loss: 7.7641 - accuracy: 0.2138 - val_loss: 7.5414 - val_accuracy: 0.2480\n",
+      "Epoch 78/200\n",
+      "32/32 [==============================] - 4s 126ms/step - loss: 7.7802 - accuracy: 0.2030 - val_loss: 7.5387 - val_accuracy: 0.2470\n",
+      "Epoch 79/200\n",
+      "32/32 [==============================] - 5s 143ms/step - loss: 7.7838 - accuracy: 0.1990 - val_loss: 7.5362 - val_accuracy: 0.2480\n",
+      "Epoch 80/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 7.7977 - accuracy: 0.2058 - val_loss: 7.5336 - val_accuracy: 0.2470\n",
+      "Epoch 81/200\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 7.7726 - accuracy: 0.2040 - val_loss: 7.5306 - val_accuracy: 0.2490\n",
+      "Epoch 82/200\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 7.7691 - accuracy: 0.2048 - val_loss: 7.5279 - val_accuracy: 0.2470\n",
+      "Epoch 83/200\n",
+      "32/32 [==============================] - 4s 114ms/step - loss: 7.7585 - accuracy: 0.2080 - val_loss: 7.5251 - val_accuracy: 0.2500\n",
+      "Epoch 84/200\n",
+      "32/32 [==============================] - 4s 115ms/step - loss: 7.7603 - accuracy: 0.2192 - val_loss: 7.5229 - val_accuracy: 0.2520\n",
+      "Epoch 85/200\n",
+      "32/32 [==============================] - 4s 119ms/step - loss: 7.7456 - accuracy: 0.2110 - val_loss: 7.5198 - val_accuracy: 0.2530\n",
+      "Epoch 86/200\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 7.7311 - accuracy: 0.2253 - val_loss: 7.5171 - val_accuracy: 0.2540\n",
+      "Epoch 87/200\n",
+      "32/32 [==============================] - 4s 118ms/step - loss: 7.7700 - accuracy: 0.2180 - val_loss: 7.5153 - val_accuracy: 0.2550\n",
+      "Epoch 88/200\n",
+      "32/32 [==============================] - 4s 125ms/step - loss: 7.7330 - accuracy: 0.2282 - val_loss: 7.5128 - val_accuracy: 0.2530\n",
+      "Epoch 89/200\n",
+      "32/32 [==============================] - 4s 118ms/step - loss: 7.7535 - accuracy: 0.2145 - val_loss: 7.5103 - val_accuracy: 0.2550\n",
+      "Epoch 90/200\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 7.7406 - accuracy: 0.2085 - val_loss: 7.5084 - val_accuracy: 0.2550\n",
+      "Epoch 91/200\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 7.7497 - accuracy: 0.2177 - val_loss: 7.5056 - val_accuracy: 0.2580\n",
+      "Epoch 92/200\n",
+      "32/32 [==============================] - 4s 119ms/step - loss: 7.7347 - accuracy: 0.2190 - val_loss: 7.5035 - val_accuracy: 0.2600\n",
+      "Epoch 93/200\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 7.7343 - accuracy: 0.2118 - val_loss: 7.5011 - val_accuracy: 0.2620\n",
+      "Epoch 94/200\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 7.7330 - accuracy: 0.2245 - val_loss: 7.4994 - val_accuracy: 0.2610\n",
+      "Epoch 95/200\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 7.7581 - accuracy: 0.2128 - val_loss: 7.4974 - val_accuracy: 0.2620\n",
+      "Epoch 96/200\n",
+      "32/32 [==============================] - 4s 116ms/step - loss: 7.7483 - accuracy: 0.2050 - val_loss: 7.4945 - val_accuracy: 0.2660\n",
+      "Epoch 97/200\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 7.7282 - accuracy: 0.2233 - val_loss: 7.4925 - val_accuracy: 0.2710\n",
+      "Epoch 98/200\n",
+      "32/32 [==============================] - 4s 115ms/step - loss: 7.7208 - accuracy: 0.2210 - val_loss: 7.4906 - val_accuracy: 0.2700\n",
+      "Epoch 99/200\n",
+      "32/32 [==============================] - 4s 116ms/step - loss: 7.7301 - accuracy: 0.2280 - val_loss: 7.4884 - val_accuracy: 0.2700\n",
+      "Epoch 100/200\n",
+      "32/32 [==============================] - 4s 119ms/step - loss: 7.7288 - accuracy: 0.2205 - val_loss: 7.4868 - val_accuracy: 0.2710\n",
+      "Epoch 101/200\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 7.7153 - accuracy: 0.2132 - val_loss: 7.4843 - val_accuracy: 0.2710\n",
+      "Epoch 102/200\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 7.7427 - accuracy: 0.2180 - val_loss: 7.4820 - val_accuracy: 0.2720\n",
+      "Epoch 103/200\n",
+      "32/32 [==============================] - 4s 118ms/step - loss: 7.7508 - accuracy: 0.2132 - val_loss: 7.4799 - val_accuracy: 0.2730\n",
+      "Epoch 104/200\n",
+      "32/32 [==============================] - 4s 118ms/step - loss: 7.7374 - accuracy: 0.2195 - val_loss: 7.4784 - val_accuracy: 0.2710\n",
+      "Epoch 105/200\n",
+      "32/32 [==============================] - 4s 118ms/step - loss: 7.7109 - accuracy: 0.2257 - val_loss: 7.4767 - val_accuracy: 0.2710\n",
+      "Epoch 106/200\n",
+      "32/32 [==============================] - 4s 117ms/step - loss: 7.7218 - accuracy: 0.2167 - val_loss: 7.4740 - val_accuracy: 0.2720\n",
+      "Epoch 107/200\n",
+      "32/32 [==============================] - 4s 119ms/step - loss: 7.6929 - accuracy: 0.2285 - val_loss: 7.4725 - val_accuracy: 0.2690\n",
+      "Epoch 108/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 7.6828 - accuracy: 0.2210 - val_loss: 7.4705 - val_accuracy: 0.2710\n",
+      "Epoch 109/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 7.7151 - accuracy: 0.2243 - val_loss: 7.4681 - val_accuracy: 0.2700\n",
+      "Epoch 110/200\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 7.7110 - accuracy: 0.2240 - val_loss: 7.4665 - val_accuracy: 0.2720\n",
+      "Epoch 111/200\n",
+      "32/32 [==============================] - 4s 135ms/step - loss: 7.7127 - accuracy: 0.2218 - val_loss: 7.4645 - val_accuracy: 0.2730\n",
+      "Epoch 112/200\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 7.7181 - accuracy: 0.2135 - val_loss: 7.4618 - val_accuracy: 0.2750\n",
+      "Epoch 113/200\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 7.7345 - accuracy: 0.2160 - val_loss: 7.4600 - val_accuracy: 0.2740\n",
+      "Epoch 114/200\n",
+      "32/32 [==============================] - 4s 118ms/step - loss: 7.7043 - accuracy: 0.2270 - val_loss: 7.4582 - val_accuracy: 0.2730\n",
+      "Epoch 115/200\n",
+      "32/32 [==============================] - 4s 119ms/step - loss: 7.6915 - accuracy: 0.2320 - val_loss: 7.4563 - val_accuracy: 0.2760\n",
+      "Epoch 116/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 7.7198 - accuracy: 0.2120 - val_loss: 7.4554 - val_accuracy: 0.2760\n",
+      "Epoch 117/200\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 7.7043 - accuracy: 0.2272 - val_loss: 7.4536 - val_accuracy: 0.2770\n",
+      "Epoch 118/200\n",
+      "32/32 [==============================] - 4s 119ms/step - loss: 7.6905 - accuracy: 0.2310 - val_loss: 7.4519 - val_accuracy: 0.2760\n",
+      "Epoch 119/200\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 7.6905 - accuracy: 0.2298 - val_loss: 7.4501 - val_accuracy: 0.2820\n",
+      "Epoch 120/200\n",
+      "32/32 [==============================] - 4s 118ms/step - loss: 7.6868 - accuracy: 0.2265 - val_loss: 7.4491 - val_accuracy: 0.2810\n",
+      "Epoch 121/200\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 7.7044 - accuracy: 0.2200 - val_loss: 7.4479 - val_accuracy: 0.2820\n",
+      "Epoch 122/200\n",
+      "32/32 [==============================] - 4s 118ms/step - loss: 7.6905 - accuracy: 0.2288 - val_loss: 7.4459 - val_accuracy: 0.2790\n",
+      "Epoch 123/200\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 7.6706 - accuracy: 0.2280 - val_loss: 7.4444 - val_accuracy: 0.2800\n",
+      "Epoch 124/200\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 7.6836 - accuracy: 0.2333 - val_loss: 7.4419 - val_accuracy: 0.2810\n",
+      "Epoch 125/200\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 7.6944 - accuracy: 0.2303 - val_loss: 7.4395 - val_accuracy: 0.2800\n",
+      "Epoch 126/200\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 7.6784 - accuracy: 0.2342 - val_loss: 7.4382 - val_accuracy: 0.2830\n",
+      "Epoch 127/200\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 7.6786 - accuracy: 0.2257 - val_loss: 7.4366 - val_accuracy: 0.2810\n",
+      "Epoch 128/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 7.6820 - accuracy: 0.2235 - val_loss: 7.4350 - val_accuracy: 0.2840\n",
+      "Epoch 129/200\n",
+      "32/32 [==============================] - 4s 119ms/step - loss: 7.6736 - accuracy: 0.2240 - val_loss: 7.4330 - val_accuracy: 0.2860\n",
+      "Epoch 130/200\n",
+      "32/32 [==============================] - 4s 117ms/step - loss: 7.6846 - accuracy: 0.2253 - val_loss: 7.4314 - val_accuracy: 0.2830\n",
+      "Epoch 131/200\n",
+      "32/32 [==============================] - 4s 116ms/step - loss: 7.6891 - accuracy: 0.2255 - val_loss: 7.4297 - val_accuracy: 0.2870\n",
+      "Epoch 132/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 7.6736 - accuracy: 0.2292 - val_loss: 7.4282 - val_accuracy: 0.2870\n",
+      "Epoch 133/200\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 7.6661 - accuracy: 0.2235 - val_loss: 7.4273 - val_accuracy: 0.2830\n",
+      "Epoch 134/200\n",
+      "32/32 [==============================] - 4s 119ms/step - loss: 7.6550 - accuracy: 0.2390 - val_loss: 7.4257 - val_accuracy: 0.2850\n",
+      "Epoch 135/200\n",
+      "32/32 [==============================] - 4s 114ms/step - loss: 7.6708 - accuracy: 0.2310 - val_loss: 7.4243 - val_accuracy: 0.2860\n",
+      "Epoch 136/200\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 7.6671 - accuracy: 0.2393 - val_loss: 7.4221 - val_accuracy: 0.2850\n",
+      "Epoch 137/200\n",
+      "32/32 [==============================] - 4s 117ms/step - loss: 7.6570 - accuracy: 0.2410 - val_loss: 7.4202 - val_accuracy: 0.2850\n",
+      "Epoch 138/200\n",
+      "32/32 [==============================] - 4s 114ms/step - loss: 7.6506 - accuracy: 0.2415 - val_loss: 7.4185 - val_accuracy: 0.2880\n",
+      "Epoch 139/200\n",
+      "32/32 [==============================] - 4s 119ms/step - loss: 7.6627 - accuracy: 0.2260 - val_loss: 7.4173 - val_accuracy: 0.2870\n",
+      "Epoch 140/200\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 7.6471 - accuracy: 0.2385 - val_loss: 7.4159 - val_accuracy: 0.2850\n",
+      "Epoch 141/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 7.6434 - accuracy: 0.2440 - val_loss: 7.4143 - val_accuracy: 0.2870\n",
+      "Epoch 142/200\n",
+      "32/32 [==============================] - 4s 137ms/step - loss: 7.6576 - accuracy: 0.2233 - val_loss: 7.4128 - val_accuracy: 0.2870\n",
+      "Epoch 143/200\n",
+      "32/32 [==============================] - 4s 127ms/step - loss: 7.6684 - accuracy: 0.2295 - val_loss: 7.4106 - val_accuracy: 0.2910\n",
+      "Epoch 144/200\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 7.6458 - accuracy: 0.2373 - val_loss: 7.4092 - val_accuracy: 0.2890\n",
+      "Epoch 145/200\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 7.6329 - accuracy: 0.2517 - val_loss: 7.4079 - val_accuracy: 0.2900\n",
+      "Epoch 146/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 7.6422 - accuracy: 0.2315 - val_loss: 7.4073 - val_accuracy: 0.2930\n",
+      "Epoch 147/200\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 7.6533 - accuracy: 0.2358 - val_loss: 7.4056 - val_accuracy: 0.2920\n",
+      "Epoch 148/200\n",
+      "32/32 [==============================] - 4s 116ms/step - loss: 7.6501 - accuracy: 0.2430 - val_loss: 7.4039 - val_accuracy: 0.2940\n",
+      "Epoch 149/200\n",
+      "32/32 [==============================] - 4s 116ms/step - loss: 7.6542 - accuracy: 0.2362 - val_loss: 7.4019 - val_accuracy: 0.2960\n",
+      "Epoch 150/200\n",
+      "32/32 [==============================] - 4s 117ms/step - loss: 7.6436 - accuracy: 0.2365 - val_loss: 7.4001 - val_accuracy: 0.2950\n",
+      "Epoch 151/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 7.6354 - accuracy: 0.2335 - val_loss: 7.3989 - val_accuracy: 0.2940\n",
+      "Epoch 152/200\n",
+      "32/32 [==============================] - 4s 119ms/step - loss: 7.6617 - accuracy: 0.2380 - val_loss: 7.3975 - val_accuracy: 0.2940\n",
+      "Epoch 153/200\n",
+      "32/32 [==============================] - 4s 119ms/step - loss: 7.6192 - accuracy: 0.2410 - val_loss: 7.3968 - val_accuracy: 0.2930\n",
+      "Epoch 154/200\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 7.6340 - accuracy: 0.2335 - val_loss: 7.3950 - val_accuracy: 0.2950\n",
+      "Epoch 155/200\n",
+      "32/32 [==============================] - 4s 118ms/step - loss: 7.6250 - accuracy: 0.2387 - val_loss: 7.3938 - val_accuracy: 0.2970\n",
+      "Epoch 156/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 7.6438 - accuracy: 0.2370 - val_loss: 7.3924 - val_accuracy: 0.2960\n",
+      "Epoch 157/200\n",
+      "32/32 [==============================] - 4s 115ms/step - loss: 7.6198 - accuracy: 0.2453 - val_loss: 7.3913 - val_accuracy: 0.3000\n",
+      "Epoch 158/200\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 7.6583 - accuracy: 0.2342 - val_loss: 7.3901 - val_accuracy: 0.3030\n",
+      "Epoch 159/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 7.6203 - accuracy: 0.2432 - val_loss: 7.3882 - val_accuracy: 0.3010\n",
+      "Epoch 160/200\n",
+      "32/32 [==============================] - 4s 118ms/step - loss: 7.6554 - accuracy: 0.2285 - val_loss: 7.3874 - val_accuracy: 0.3020\n",
+      "Epoch 161/200\n",
+      "32/32 [==============================] - 4s 119ms/step - loss: 7.6492 - accuracy: 0.2365 - val_loss: 7.3856 - val_accuracy: 0.2970\n",
+      "Epoch 162/200\n",
+      "32/32 [==============================] - 4s 125ms/step - loss: 7.6338 - accuracy: 0.2438 - val_loss: 7.3842 - val_accuracy: 0.3000\n",
+      "Epoch 163/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 7.6259 - accuracy: 0.2377 - val_loss: 7.3827 - val_accuracy: 0.3030\n",
+      "Epoch 164/200\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 7.6097 - accuracy: 0.2420 - val_loss: 7.3813 - val_accuracy: 0.3020\n",
+      "Epoch 165/200\n",
+      "32/32 [==============================] - 4s 114ms/step - loss: 7.6206 - accuracy: 0.2370 - val_loss: 7.3804 - val_accuracy: 0.3030\n",
+      "Epoch 166/200\n",
+      "32/32 [==============================] - 4s 119ms/step - loss: 7.6210 - accuracy: 0.2393 - val_loss: 7.3790 - val_accuracy: 0.3030\n",
+      "Epoch 167/200\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 7.6314 - accuracy: 0.2428 - val_loss: 7.3775 - val_accuracy: 0.3020\n",
+      "Epoch 168/200\n",
+      "32/32 [==============================] - 4s 115ms/step - loss: 7.6029 - accuracy: 0.2465 - val_loss: 7.3761 - val_accuracy: 0.3070\n",
+      "Epoch 169/200\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 7.5835 - accuracy: 0.2467 - val_loss: 7.3750 - val_accuracy: 0.3060\n",
+      "Epoch 170/200\n",
+      "32/32 [==============================] - 4s 117ms/step - loss: 7.6193 - accuracy: 0.2320 - val_loss: 7.3742 - val_accuracy: 0.3040\n",
+      "Epoch 171/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 7.6305 - accuracy: 0.2362 - val_loss: 7.3725 - val_accuracy: 0.3050\n",
+      "Epoch 172/200\n",
+      "32/32 [==============================] - 4s 118ms/step - loss: 7.6199 - accuracy: 0.2432 - val_loss: 7.3713 - val_accuracy: 0.3060\n",
+      "Epoch 173/200\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 7.6162 - accuracy: 0.2553 - val_loss: 7.3701 - val_accuracy: 0.3030\n",
+      "Epoch 174/200\n",
+      "32/32 [==============================] - 5s 142ms/step - loss: 7.6056 - accuracy: 0.2400 - val_loss: 7.3687 - val_accuracy: 0.3060\n",
+      "Epoch 175/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 7.5935 - accuracy: 0.2498 - val_loss: 7.3676 - val_accuracy: 0.3080\n",
+      "Epoch 176/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 7.6068 - accuracy: 0.2490 - val_loss: 7.3660 - val_accuracy: 0.3090\n",
+      "Epoch 177/200\n",
+      "32/32 [==============================] - 4s 116ms/step - loss: 7.6315 - accuracy: 0.2330 - val_loss: 7.3653 - val_accuracy: 0.3070\n",
+      "Epoch 178/200\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 7.5982 - accuracy: 0.2445 - val_loss: 7.3636 - val_accuracy: 0.3080\n",
+      "Epoch 179/200\n",
+      "32/32 [==============================] - 4s 117ms/step - loss: 7.5611 - accuracy: 0.2500 - val_loss: 7.3620 - val_accuracy: 0.3080\n",
+      "Epoch 180/200\n",
+      "32/32 [==============================] - 4s 118ms/step - loss: 7.5995 - accuracy: 0.2453 - val_loss: 7.3610 - val_accuracy: 0.3090\n",
+      "Epoch 181/200\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 7.6195 - accuracy: 0.2380 - val_loss: 7.3603 - val_accuracy: 0.3090\n",
+      "Epoch 182/200\n",
+      "32/32 [==============================] - 4s 115ms/step - loss: 7.6183 - accuracy: 0.2453 - val_loss: 7.3591 - val_accuracy: 0.3080\n",
+      "Epoch 183/200\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 7.5992 - accuracy: 0.2420 - val_loss: 7.3584 - val_accuracy: 0.3120\n",
+      "Epoch 184/200\n",
+      "32/32 [==============================] - 4s 116ms/step - loss: 7.5978 - accuracy: 0.2368 - val_loss: 7.3572 - val_accuracy: 0.3130\n",
+      "Epoch 185/200\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 7.5972 - accuracy: 0.2453 - val_loss: 7.3563 - val_accuracy: 0.3110\n",
+      "Epoch 186/200\n",
+      " 8/32 [======>.......................] - ETA: 2s - loss: 7.6009 - accuracy: 0.2414"
+     ]
+    }
+   ],
+   "source": [
+    "# Define Network\n",
+    "model = tf.keras.Sequential([\n",
+    "        tf.keras.layers.Flatten(input_shape=(32, 32, 3)),\n",
+    "        tf.keras.layers.Dense(512, \n",
+    "                              kernel_regularizer=tf.keras.regularizers.l2(0.005)),\n",
+    "        tf.keras.layers.BatchNormalization(),\n",
+    "        tf.keras.layers.Activation(tf.nn.relu),\n",
+    "        tf.keras.layers.Dense(128, \n",
+    "                              kernel_regularizer=tf.keras.regularizers.l2(0.005)),\n",
+    "        tf.keras.layers.BatchNormalization(),\n",
+    "        tf.keras.layers.Activation(tf.nn.relu),\n",
+    "        tf. keras.layers.Dropout(0.3),\n",
+    "        tf.keras.layers.Dense(10,  \n",
+    "                          activation=tf.nn.softmax, \n",
+    "                             kernel_regularizer=tf.keras.regularizers.l2(0.005))\n",
+    "])\n",
+    "\n",
+    "# Compile Network \n",
+    "model.compile(optimizer=tf.keras.optimizers.SGD(lr=0.0001),\n",
+    "              loss='categorical_crossentropy',\n",
+    "              metrics=['accuracy'])\n",
+    "\n",
+    "num_validation_examples = 1000\n",
+    "num_train_examples = 5000\n",
+    "\n",
+    "\n",
+    "train_iterator = train_datagen.flow(X_train_zc[num_validation_examples:num_train_examples], \n",
+    "                                    y_train_cat[num_validation_examples:num_train_examples], \n",
+    "                                    batch_size=128)\n",
+    "validation_iterator = validation_datagen.flow(X_train_zc[:num_validation_examples:], \n",
+    "                                              y_train_cat[:num_validation_examples:], \n",
+    "                                              batch_size=128)\n",
+    "\n",
+    "# Fit Network\n",
+    "history = model.fit_generator(generator= train_iterator,  \n",
+    "                              validation_data = validation_iterator, \n",
+    "                              epochs=200, \n",
+    "                              steps_per_epoch=len(train_iterator))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "id": "jrCdZzd3vXa2"
+   },
+   "outputs": [],
+   "source": [
+    "# Plot training and validation accuracy\n",
+    "acc = history.history['accuracy']\n",
+    "val_acc = history.history['val_accuracy']\n",
+    "\n",
+    "loss = history.history['loss']\n",
+    "val_loss = history.history['val_loss']\n",
+    "\n",
+    "epochs_range = range(200)\n",
+    "\n",
+    "plt.figure(figsize=(8, 8))\n",
+    "plt.subplot(1, 2, 1)\n",
+    "plt.plot(epochs_range, acc, label='Training Accuracy')\n",
+    "plt.plot(epochs_range, val_acc, label='Validation Accuracy')\n",
+    "plt.legend(loc='lower right')\n",
+    "plt.title('Training and Validation Accuracy')\n",
+    "\n",
+    "plt.subplot(1, 2, 2)\n",
+    "plt.plot(epochs_range, loss, label='Training Loss')\n",
+    "plt.plot(epochs_range, val_loss, label='Validation Loss')\n",
+    "plt.legend(loc='upper right')\n",
+    "plt.title('Training and Validation Loss')\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "id": "VOJYqDdXvXa5"
+   },
+   "outputs": [],
+   "source": [
+    "# Evaluate test accuracy\n",
+    "test_loss, test_accuracy = model.evaluate(X_test_zc, y_test_cat)\n",
+    "print('Accuracy on test dataset:', test_accuracy)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "id": "vc8w1Q3_vXa9"
+   },
+   "source": [
+    "# 9. Parameter Update\n",
+    "\n",
+    "In practice `Adam` is currently recommended as the default algorithm to use, and often works slightly better than RMSProp. However, it is often also worth trying SGD+Nesterov Momentum as an alternative. In addition, it is recommended to use learning rate decay. The stochastic gradient descent optimization algorithm implementation in the SGD class has an argument called `decay`. This argument is used in the time-based learning rate decay schedule equation as follows:\n",
+    "\n",
+    "LearningRate = LearningRate * 1/(1 + decay * epoch)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 59,
+   "metadata": {
+    "id": "ZxI6XZxJvXa-"
+   },
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/opt/conda/lib/python3.7/site-packages/ipykernel_launcher.py:39: UserWarning: `Model.fit_generator` is deprecated and will be removed in a future version. Please use `Model.fit`, which supports generators.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 1/100\n",
+      "32/32 [==============================] - 8s 194ms/step - loss: 7.9278 - accuracy: 0.1795 - val_loss: 8.7435 - val_accuracy: 0.1630\n",
+      "Epoch 2/100\n",
+      "32/32 [==============================] - 5s 146ms/step - loss: 7.5847 - accuracy: 0.2355 - val_loss: 7.9850 - val_accuracy: 0.1890\n",
+      "Epoch 3/100\n",
+      "32/32 [==============================] - 5s 140ms/step - loss: 7.4110 - accuracy: 0.2598 - val_loss: 7.3038 - val_accuracy: 0.2660\n",
+      "Epoch 4/100\n",
+      "32/32 [==============================] - 4s 128ms/step - loss: 7.2711 - accuracy: 0.2718 - val_loss: 7.1753 - val_accuracy: 0.2720\n",
+      "Epoch 5/100\n",
+      "32/32 [==============================] - 4s 129ms/step - loss: 7.1305 - accuracy: 0.2885 - val_loss: 6.9784 - val_accuracy: 0.3390\n",
+      "Epoch 6/100\n",
+      "32/32 [==============================] - 4s 126ms/step - loss: 7.0077 - accuracy: 0.2955 - val_loss: 6.9774 - val_accuracy: 0.2720\n",
+      "Epoch 7/100\n",
+      "32/32 [==============================] - 4s 127ms/step - loss: 6.8876 - accuracy: 0.3085 - val_loss: 6.9673 - val_accuracy: 0.2690\n",
+      "Epoch 8/100\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 6.7867 - accuracy: 0.2993 - val_loss: 6.7843 - val_accuracy: 0.2750\n",
+      "Epoch 9/100\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 6.6536 - accuracy: 0.3027 - val_loss: 6.5008 - val_accuracy: 0.3240\n",
+      "Epoch 10/100\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 6.5523 - accuracy: 0.3063 - val_loss: 6.6367 - val_accuracy: 0.2980\n",
+      "Epoch 11/100\n",
+      "32/32 [==============================] - 4s 126ms/step - loss: 6.4511 - accuracy: 0.3135 - val_loss: 6.2525 - val_accuracy: 0.3490\n",
+      "Epoch 12/100\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 6.3515 - accuracy: 0.3243 - val_loss: 6.2500 - val_accuracy: 0.3330\n",
+      "Epoch 13/100\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 6.2211 - accuracy: 0.3330 - val_loss: 6.1633 - val_accuracy: 0.3460\n",
+      "Epoch 14/100\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 6.1450 - accuracy: 0.3255 - val_loss: 6.0062 - val_accuracy: 0.3610\n",
+      "Epoch 15/100\n",
+      "32/32 [==============================] - 4s 125ms/step - loss: 6.0337 - accuracy: 0.3313 - val_loss: 6.0169 - val_accuracy: 0.2890\n",
+      "Epoch 16/100\n",
+      "32/32 [==============================] - 4s 125ms/step - loss: 5.9399 - accuracy: 0.3295 - val_loss: 5.9055 - val_accuracy: 0.3250\n",
+      "Epoch 17/100\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 5.8530 - accuracy: 0.3270 - val_loss: 5.8749 - val_accuracy: 0.2960\n",
+      "Epoch 18/100\n",
+      "32/32 [==============================] - 4s 126ms/step - loss: 5.7460 - accuracy: 0.3252 - val_loss: 5.7107 - val_accuracy: 0.3370\n",
+      "Epoch 19/100\n",
+      "32/32 [==============================] - 4s 127ms/step - loss: 5.6489 - accuracy: 0.3355 - val_loss: 5.5232 - val_accuracy: 0.3780\n",
+      "Epoch 20/100\n",
+      "32/32 [==============================] - 4s 129ms/step - loss: 5.5519 - accuracy: 0.3492 - val_loss: 5.5483 - val_accuracy: 0.3510\n",
+      "Epoch 21/100\n",
+      "32/32 [==============================] - 4s 131ms/step - loss: 5.4864 - accuracy: 0.3453 - val_loss: 5.6983 - val_accuracy: 0.2720\n",
+      "Epoch 22/100\n",
+      "32/32 [==============================] - 5s 145ms/step - loss: 5.3888 - accuracy: 0.3455 - val_loss: 5.2684 - val_accuracy: 0.3730\n",
+      "Epoch 23/100\n",
+      "32/32 [==============================] - 4s 125ms/step - loss: 5.3491 - accuracy: 0.3140 - val_loss: 5.4412 - val_accuracy: 0.2830\n",
+      "Epoch 24/100\n",
+      "32/32 [==============================] - 4s 117ms/step - loss: 5.1893 - accuracy: 0.3535 - val_loss: 5.1698 - val_accuracy: 0.3400\n",
+      "Epoch 25/100\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 5.1523 - accuracy: 0.3388 - val_loss: 5.2489 - val_accuracy: 0.2900\n",
+      "Epoch 26/100\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 5.0692 - accuracy: 0.3523 - val_loss: 5.0228 - val_accuracy: 0.3450\n",
+      "Epoch 27/100\n",
+      "32/32 [==============================] - 4s 128ms/step - loss: 4.9970 - accuracy: 0.3453 - val_loss: 4.9607 - val_accuracy: 0.3490\n",
+      "Epoch 28/100\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 4.9232 - accuracy: 0.3455 - val_loss: 4.9283 - val_accuracy: 0.3390\n",
+      "Epoch 29/100\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 4.8450 - accuracy: 0.3595 - val_loss: 4.7769 - val_accuracy: 0.3600\n",
+      "Epoch 30/100\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 4.7822 - accuracy: 0.3525 - val_loss: 4.9336 - val_accuracy: 0.3220\n",
+      "Epoch 31/100\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 4.7048 - accuracy: 0.3602 - val_loss: 4.6390 - val_accuracy: 0.3450\n",
+      "Epoch 32/100\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 4.6338 - accuracy: 0.3537 - val_loss: 4.8122 - val_accuracy: 0.3060\n",
+      "Epoch 33/100\n",
+      "32/32 [==============================] - 4s 130ms/step - loss: 4.5592 - accuracy: 0.3627 - val_loss: 4.4749 - val_accuracy: 0.3730\n",
+      "Epoch 34/100\n",
+      "32/32 [==============================] - 4s 130ms/step - loss: 4.4877 - accuracy: 0.3735 - val_loss: 4.6357 - val_accuracy: 0.3300\n",
+      "Epoch 35/100\n",
+      "32/32 [==============================] - 4s 125ms/step - loss: 4.4264 - accuracy: 0.3638 - val_loss: 4.3905 - val_accuracy: 0.3810\n",
+      "Epoch 36/100\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 4.3562 - accuracy: 0.3700 - val_loss: 4.2849 - val_accuracy: 0.3750\n",
+      "Epoch 37/100\n",
+      "32/32 [==============================] - 4s 125ms/step - loss: 4.3279 - accuracy: 0.3550 - val_loss: 4.3070 - val_accuracy: 0.3680\n",
+      "Epoch 38/100\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 4.2570 - accuracy: 0.3580 - val_loss: 4.1936 - val_accuracy: 0.3840\n",
+      "Epoch 39/100\n",
+      "32/32 [==============================] - 4s 126ms/step - loss: 4.2110 - accuracy: 0.3495 - val_loss: 4.2390 - val_accuracy: 0.3220\n",
+      "Epoch 40/100\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 4.1555 - accuracy: 0.3573 - val_loss: 4.1628 - val_accuracy: 0.3270\n",
+      "Epoch 41/100\n",
+      "32/32 [==============================] - 4s 125ms/step - loss: 4.1026 - accuracy: 0.3575 - val_loss: 4.1093 - val_accuracy: 0.3500\n",
+      "Epoch 42/100\n",
+      "32/32 [==============================] - 4s 125ms/step - loss: 4.0492 - accuracy: 0.3683 - val_loss: 4.0883 - val_accuracy: 0.3610\n",
+      "Epoch 43/100\n",
+      "32/32 [==============================] - 4s 125ms/step - loss: 3.9741 - accuracy: 0.3742 - val_loss: 3.9863 - val_accuracy: 0.3710\n",
+      "Epoch 44/100\n",
+      "32/32 [==============================] - 4s 126ms/step - loss: 3.9070 - accuracy: 0.3710 - val_loss: 3.8807 - val_accuracy: 0.3910\n",
+      "Epoch 45/100\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 3.9007 - accuracy: 0.3660 - val_loss: 4.0333 - val_accuracy: 0.3180\n",
+      "Epoch 46/100\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 3.8386 - accuracy: 0.3683 - val_loss: 3.8179 - val_accuracy: 0.3710\n",
+      "Epoch 47/100\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 3.8148 - accuracy: 0.3700 - val_loss: 3.7847 - val_accuracy: 0.3720\n",
+      "Epoch 48/100\n",
+      "32/32 [==============================] - 4s 125ms/step - loss: 3.7699 - accuracy: 0.3645 - val_loss: 3.9120 - val_accuracy: 0.2960\n",
+      "Epoch 49/100\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 3.7052 - accuracy: 0.3738 - val_loss: 3.7466 - val_accuracy: 0.3860\n",
+      "Epoch 50/100\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 3.6604 - accuracy: 0.3770 - val_loss: 3.6506 - val_accuracy: 0.3770\n",
+      "Epoch 51/100\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 3.6110 - accuracy: 0.3708 - val_loss: 3.7059 - val_accuracy: 0.3350\n",
+      "Epoch 52/100\n",
+      "32/32 [==============================] - 4s 131ms/step - loss: 3.5865 - accuracy: 0.3697 - val_loss: 3.6172 - val_accuracy: 0.3720\n",
+      "Epoch 53/100\n",
+      "32/32 [==============================] - 5s 145ms/step - loss: 3.5357 - accuracy: 0.3840 - val_loss: 3.5688 - val_accuracy: 0.3510\n",
+      "Epoch 54/100\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 3.4730 - accuracy: 0.3935 - val_loss: 3.5173 - val_accuracy: 0.3750\n",
+      "Epoch 55/100\n",
+      "32/32 [==============================] - 4s 129ms/step - loss: 3.4665 - accuracy: 0.3742 - val_loss: 3.5868 - val_accuracy: 0.3360\n",
+      "Epoch 56/100\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 3.4167 - accuracy: 0.3710 - val_loss: 3.3837 - val_accuracy: 0.3870\n",
+      "Epoch 57/100\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 3.3701 - accuracy: 0.3915 - val_loss: 3.4149 - val_accuracy: 0.3630\n",
+      "Epoch 58/100\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 3.3672 - accuracy: 0.3820 - val_loss: 3.3169 - val_accuracy: 0.4010\n",
+      "Epoch 59/100\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 3.3082 - accuracy: 0.3887 - val_loss: 3.3791 - val_accuracy: 0.3570\n",
+      "Epoch 60/100\n",
+      "32/32 [==============================] - 4s 125ms/step - loss: 3.3059 - accuracy: 0.3787 - val_loss: 3.4320 - val_accuracy: 0.3280\n",
+      "Epoch 61/100\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 3.2544 - accuracy: 0.3875 - val_loss: 3.3982 - val_accuracy: 0.3290\n",
+      "Epoch 62/100\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 3.2020 - accuracy: 0.3875 - val_loss: 3.4636 - val_accuracy: 0.3390\n",
+      "Epoch 63/100\n",
+      "32/32 [==============================] - 4s 127ms/step - loss: 3.1931 - accuracy: 0.3832 - val_loss: 3.2788 - val_accuracy: 0.3360\n",
+      "Epoch 64/100\n",
+      "32/32 [==============================] - 4s 127ms/step - loss: 3.1928 - accuracy: 0.3740 - val_loss: 3.1737 - val_accuracy: 0.3700\n",
+      "Epoch 65/100\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 3.1177 - accuracy: 0.3988 - val_loss: 3.1195 - val_accuracy: 0.3910\n",
+      "Epoch 66/100\n",
+      "32/32 [==============================] - 4s 125ms/step - loss: 3.0972 - accuracy: 0.3835 - val_loss: 3.3795 - val_accuracy: 0.2900\n",
+      "Epoch 67/100\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 3.0761 - accuracy: 0.3890 - val_loss: 3.1332 - val_accuracy: 0.3630\n",
+      "Epoch 68/100\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 3.0429 - accuracy: 0.3845 - val_loss: 3.0678 - val_accuracy: 0.3840\n",
+      "Epoch 69/100\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 3.0335 - accuracy: 0.3842 - val_loss: 3.2026 - val_accuracy: 0.3430\n",
+      "Epoch 70/100\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 3.0042 - accuracy: 0.3902 - val_loss: 3.0992 - val_accuracy: 0.3310\n",
+      "Epoch 71/100\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 2.9729 - accuracy: 0.3790 - val_loss: 2.9847 - val_accuracy: 0.3770\n",
+      "Epoch 72/100\n",
+      "32/32 [==============================] - 4s 127ms/step - loss: 2.9472 - accuracy: 0.3890 - val_loss: 3.0121 - val_accuracy: 0.3560\n",
+      "Epoch 73/100\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 2.9414 - accuracy: 0.3855 - val_loss: 2.8951 - val_accuracy: 0.4120\n",
+      "Epoch 74/100\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 2.9079 - accuracy: 0.3855 - val_loss: 2.9885 - val_accuracy: 0.3530\n",
+      "Epoch 75/100\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 2.8793 - accuracy: 0.3930 - val_loss: 2.9061 - val_accuracy: 0.3740\n",
+      "Epoch 76/100\n",
+      "32/32 [==============================] - 4s 126ms/step - loss: 2.8585 - accuracy: 0.3792 - val_loss: 2.8393 - val_accuracy: 0.3960\n",
+      "Epoch 77/100\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 2.8236 - accuracy: 0.3950 - val_loss: 2.8561 - val_accuracy: 0.3730\n",
+      "Epoch 78/100\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 2.8158 - accuracy: 0.3907 - val_loss: 2.8607 - val_accuracy: 0.3800\n",
+      "Epoch 79/100\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 2.7802 - accuracy: 0.4015 - val_loss: 2.8434 - val_accuracy: 0.3750\n",
+      "Epoch 80/100\n",
+      "32/32 [==============================] - 4s 122ms/step - loss: 2.7607 - accuracy: 0.3887 - val_loss: 2.8587 - val_accuracy: 0.3650\n",
+      "Epoch 81/100\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 2.7576 - accuracy: 0.3885 - val_loss: 2.7834 - val_accuracy: 0.3720\n",
+      "Epoch 82/100\n",
+      "32/32 [==============================] - 4s 128ms/step - loss: 2.7446 - accuracy: 0.3817 - val_loss: 2.8279 - val_accuracy: 0.3640\n",
+      "Epoch 83/100\n",
+      "32/32 [==============================] - 4s 121ms/step - loss: 2.7224 - accuracy: 0.3995 - val_loss: 2.8202 - val_accuracy: 0.3530\n",
+      "Epoch 84/100\n",
+      "32/32 [==============================] - 4s 128ms/step - loss: 2.7081 - accuracy: 0.3913 - val_loss: 2.8054 - val_accuracy: 0.3780\n",
+      "Epoch 85/100\n",
+      "32/32 [==============================] - 5s 144ms/step - loss: 2.6631 - accuracy: 0.4033 - val_loss: 3.2552 - val_accuracy: 0.2540\n",
+      "Epoch 86/100\n",
+      "32/32 [==============================] - 4s 127ms/step - loss: 2.6763 - accuracy: 0.4027 - val_loss: 2.8086 - val_accuracy: 0.3350\n",
+      "Epoch 87/100\n",
+      "32/32 [==============================] - 4s 126ms/step - loss: 2.6311 - accuracy: 0.4062 - val_loss: 2.7136 - val_accuracy: 0.3700\n",
+      "Epoch 88/100\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 2.6369 - accuracy: 0.3907 - val_loss: 2.6391 - val_accuracy: 0.3980\n",
+      "Epoch 89/100\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 2.6020 - accuracy: 0.4027 - val_loss: 2.6999 - val_accuracy: 0.3630\n",
+      "Epoch 90/100\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 2.5928 - accuracy: 0.4098 - val_loss: 2.5866 - val_accuracy: 0.4060\n",
+      "Epoch 91/100\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 2.5793 - accuracy: 0.4013 - val_loss: 2.8585 - val_accuracy: 0.3090\n",
+      "Epoch 92/100\n",
+      "32/32 [==============================] - 4s 127ms/step - loss: 2.5699 - accuracy: 0.3983 - val_loss: 2.8960 - val_accuracy: 0.3100\n",
+      "Epoch 93/100\n",
+      "32/32 [==============================] - 4s 131ms/step - loss: 2.5651 - accuracy: 0.3895 - val_loss: 2.9424 - val_accuracy: 0.2690\n",
+      "Epoch 94/100\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 2.5424 - accuracy: 0.3905 - val_loss: 2.6070 - val_accuracy: 0.3860\n",
+      "Epoch 95/100\n",
+      "32/32 [==============================] - 4s 123ms/step - loss: 2.5166 - accuracy: 0.4025 - val_loss: 2.7224 - val_accuracy: 0.3390\n",
+      "Epoch 96/100\n",
+      "32/32 [==============================] - 4s 127ms/step - loss: 2.5092 - accuracy: 0.4020 - val_loss: 2.6135 - val_accuracy: 0.3560\n",
+      "Epoch 97/100\n",
+      "32/32 [==============================] - 4s 127ms/step - loss: 2.4977 - accuracy: 0.4075 - val_loss: 2.5204 - val_accuracy: 0.3970\n",
+      "Epoch 98/100\n",
+      "32/32 [==============================] - 4s 120ms/step - loss: 2.4865 - accuracy: 0.4058 - val_loss: 2.5593 - val_accuracy: 0.3860\n",
+      "Epoch 99/100\n",
+      "32/32 [==============================] - 4s 126ms/step - loss: 2.4784 - accuracy: 0.3925 - val_loss: 2.4718 - val_accuracy: 0.3940\n",
+      "Epoch 100/100\n",
+      "32/32 [==============================] - 4s 124ms/step - loss: 2.4630 - accuracy: 0.3988 - val_loss: 2.5533 - val_accuracy: 0.3490\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Define Network\n",
+    "model = tf.keras.Sequential([\n",
+    "        tf.keras.layers.Flatten(input_shape=(32, 32, 3)),\n",
+    "        tf.keras.layers.Dense(512, \n",
+    "                              kernel_regularizer=tf.keras.regularizers.l2(0.005)),\n",
+    "        tf.keras.layers.BatchNormalization(),\n",
+    "        tf.keras.layers.Activation(tf.nn.relu),\n",
+    "        tf.keras.layers.Dense(128, \n",
+    "                              kernel_regularizer=tf.keras.regularizers.l2(0.005)),\n",
+    "        tf.keras.layers.BatchNormalization(),\n",
+    "        tf.keras.layers.Activation(tf.nn.relu),\n",
+    "        tf. keras.layers.Dropout(0.3),\n",
+    "        tf.keras.layers.Dense(10,  \n",
+    "                          activation=tf.nn.softmax, \n",
+    "                             kernel_regularizer=tf.keras.regularizers.l2(0.005))\n",
+    "])\n",
+    "\n",
+    "# Compile Network \n",
+    "model.compile(optimizer=tf.keras.optimizers.Adam(lr=0.0001, \n",
+    "              decay=0.0001/10),\n",
+    "              loss='categorical_crossentropy',\n",
+    "              metrics=['accuracy'])\n",
+    "\n",
+    "num_validation_examples = 1000\n",
+    "num_train_examples = 5000\n",
+    "\n",
+    "\n",
+    "train_iterator = train_datagen.flow(X_train_zc[num_validation_examples:num_train_examples], \n",
+    "                                    y_train_cat[num_validation_examples:num_train_examples], \n",
+    "                                    batch_size=128)\n",
+    "validation_iterator = validation_datagen.flow(X_train_zc[:num_validation_examples:], \n",
+    "                                              y_train_cat[:num_validation_examples:], \n",
+    "                                              batch_size=128)\n",
+    "\n",
+    "# Fit Network\n",
+    "history = model.fit_generator(generator= train_iterator,  \n",
+    "                              validation_data = validation_iterator, \n",
+    "                              epochs=100, \n",
+    "                              steps_per_epoch=len(train_iterator))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 60,
+   "metadata": {
+    "id": "cixRpMMOvXbB"
+   },
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 576x576 with 2 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Plot training and validation accuracy\n",
+    "acc = history.history['accuracy']\n",
+    "val_acc = history.history['val_accuracy']\n",
+    "\n",
+    "loss = history.history['loss']\n",
+    "val_loss = history.history['val_loss']\n",
+    "\n",
+    "epochs_range = range(100)\n",
+    "\n",
+    "plt.figure(figsize=(8, 8))\n",
+    "plt.subplot(1, 2, 1)\n",
+    "plt.plot(epochs_range, acc, label='Training Accuracy')\n",
+    "plt.plot(epochs_range, val_acc, label='Validation Accuracy')\n",
+    "plt.legend(loc='lower right')\n",
+    "plt.title('Training and Validation Accuracy')\n",
+    "\n",
+    "plt.subplot(1, 2, 2)\n",
+    "plt.plot(epochs_range, loss, label='Training Loss')\n",
+    "plt.plot(epochs_range, val_loss, label='Validation Loss')\n",
+    "plt.legend(loc='upper right')\n",
+    "plt.title('Training and Validation Loss')\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 61,
+   "metadata": {
+    "id": "eCuOiek7vXbQ"
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "313/313 [==============================] - 5s 16ms/step - loss: 2.5623 - accuracy: 0.3593 0s - loss: 2.5660 \n",
+      "Accuracy on test dataset: 0.35929998755455017\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Evaluate test accuracy\n",
+    "test_loss, test_accuracy = model.evaluate(X_test_zc, y_test_cat)\n",
+    "print('Accuracy on test dataset:', test_accuracy)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "id": "U4S4eZcArngs"
+   },
+   "source": [
+    "# 10. Save Best Model on Google Drive and Reload it"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 62,
+   "metadata": {
+    "id": "lZ_eBernxdYc"
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[33mWARNING: You are using pip version 20.2.4; however, version 22.0.4 is available.\n",
+      "You should consider upgrading via the '/opt/conda/bin/python3 -m pip install --upgrade pip' command.\u001b[0m\n"
+     ]
+    },
+    {
+     "ename": "ModuleNotFoundError",
+     "evalue": "No module named 'google.colab'",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mModuleNotFoundError\u001b[0m                       Traceback (most recent call last)",
+      "\u001b[0;32m<ipython-input-62-61720729c72a>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0mpydrive\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mauth\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mGoogleAuth\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0mpydrive\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdrive\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mGoogleDrive\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 6\u001b[0;31m \u001b[0;32mfrom\u001b[0m \u001b[0mgoogle\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcolab\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mauth\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      7\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0moauth2client\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mclient\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mGoogleCredentials\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      8\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;31mModuleNotFoundError\u001b[0m: No module named 'google.colab'"
+     ]
+    }
+   ],
+   "source": [
+    "# Install the PyDrive wrapper & import libraries.\n",
+    "# This only needs to be done once in a notebook.\n",
+    "!pip install -U -q PyDrive\n",
+    "from pydrive.auth import GoogleAuth\n",
+    "from pydrive.drive import GoogleDrive\n",
+    "from google.colab import auth\n",
+    "from oauth2client.client import GoogleCredentials\n",
+    "\n",
+    "# Authenticate and create the PyDrive client.\n",
+    "# This only needs to be done once in a notebook.\n",
+    "auth.authenticate_user()\n",
+    "gauth = GoogleAuth()\n",
+    "gauth.credentials = GoogleCredentials.get_application_default()\n",
+    "drive = GoogleDrive(gauth)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "id": "a4wwPUFhxpjw"
+   },
+   "outputs": [],
+   "source": [
+    "!curl https://raw.githubusercontent.com/dexterfichuk/GoogleDriveCheckpoint/master/google_drive_checkpoint.py -O"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "id": "7pT6GX1lxyg2"
+   },
+   "outputs": [],
+   "source": [
+    "from google_drive_checkpoint import GoogleDriveCheckpoint\n",
+    "\n",
+    "checkpoint = GoogleDriveCheckpoint('best_model.h5', drive, save_best_only=True,  verbose=1)\n",
+    "# Fit Network\n",
+    "history = model.fit_generator(generator= train_iterator,  \n",
+    "                              validation_data = validation_iterator, \n",
+    "                              epochs=100, \n",
+    "                              steps_per_epoch=len(train_iterator),callbacks=[checkpoint])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "id": "XW4eFU5xGv53"
+   },
+   "outputs": [],
+   "source": [
+    "from google.colab import drive\n",
+    "from tensorflow.keras.models import load_model\n",
+    "\n",
+    "drive.mount('/content/gdrive')\n",
+    "\n",
+    "# load model : go on your google drive account and get the best model that was saved\n",
+    "model = load_model('/content/gdrive/My Drive/best_model-1.939315915107727.h5')\n",
+    "model.summary()\n",
+    "      \n",
+    "\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "id": "Y_QKt2hcFDY8"
+   },
+   "outputs": [],
+   "source": [
+    "# Plot training and validation accuracy\n",
+    "acc = history.history['accuracy']\n",
+    "val_acc = history.history['val_accuracy']\n",
+    "\n",
+    "loss = history.history['loss']\n",
+    "val_loss = history.history['val_loss']\n",
+    "\n",
+    "epochs_range = range(100)\n",
+    "\n",
+    "plt.figure(figsize=(8, 8))\n",
+    "plt.subplot(1, 2, 1)\n",
+    "plt.plot(epochs_range, acc, label='Training Accuracy')\n",
+    "plt.plot(epochs_range, val_acc, label='Validation Accuracy')\n",
+    "plt.legend(loc='lower right')\n",
+    "plt.title('Training and Validation Accuracy')\n",
+    "\n",
+    "plt.subplot(1, 2, 2)\n",
+    "plt.plot(epochs_range, loss, label='Training Loss')\n",
+    "plt.plot(epochs_range, val_loss, label='Validation Loss')\n",
+    "plt.legend(loc='upper right')\n",
+    "plt.title('Training and Validation Loss')\n",
+    "plt.show()\n"
+   ]
   }
  ],
  "metadata": {
-- 
GitLab