{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"## <center>Recognition of Handwritten Digit using Convolutional Neural Network (CNN) with Extended Data (QMNIST)</center>\n<center> <b>A Deep Learning Analysis with Real World Data</b> </center> \n\n#### <center>06th July 2022 </center>\n<center><b>Mohammad Shahid</b></center>","metadata":{"papermill":{"duration":0.022732,"end_time":"2022-07-03T08:26:59.921864","exception":false,"start_time":"2022-07-03T08:26:59.899132","status":"completed"},"tags":[]}},{"cell_type":"markdown","source":"![digits_recognition_cnn.png](attachment:88e93e1d-0c93-439d-8ef1-9161fb4f92ee.png)","metadata":{"papermill":{"duration":0.018608,"end_time":"2022-07-03T08:26:59.959986","exception":false,"start_time":"2022-07-03T08:26:59.941378","status":"completed"},"tags":[]},"attachments":{"88e93e1d-0c93-439d-8ef1-9161fb4f92ee.png":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAZAAAAGQCAYAAACAvzbMAAAABGdBTUEAALGPC/xhBQAAACBjSFJNAAB6JgAAgIQAAPoAAACA6AAAdTAAAOpgAAA6mAAAF3CculE8AAAAhGVYSWZNTQAqAAAACAAGAQYAAwAAAAEAAgAAARIAAwAAAAEAAQAAARoABQAAAAEAAABWARsABQAAAAEAAABeASgAAwAAAAEAAgAAh2kABAAAAAEAAABmAAAAAAAAAEgAAAABAAAASAAAAAEAAqACAAQAAAABAAABkKADAAQAAAABAAABkAAAAACIES7eAAAACXBIWXMAAAsTAAALEwEAmpwYAAAC5GlUWHRYTUw6Y29tLmFkb2JlLnhtcAAAAAAAPHg6eG1wbWV0YSB4bWxuczp4PSJhZG9iZTpuczptZXRhLyIgeDp4bXB0az0iWE1QIENvcmUgNS40LjAiPgogICA8cmRmOlJERiB4bWxuczpyZGY9Imh0dHA6Ly93d3cudzMub3JnLzE5OTkvMDIvMjItcmRmLXN5bnRheC1ucyMiPgogICAgICA8cmRmOkRlc2NyaXB0aW9uIHJkZjphYm91dD0iIgogICAgICAgICAgICB4bWxuczp0aWZmPSJodHRwOi8vbnMuYWRvYmUuY29tL3RpZmYvMS4wLyIKICAgICAgICAgICAgeG1sbnM6ZXhpZj0iaHR0cDovL25zLmFkb2JlLmNvbS9leGlmLzEuMC8iPgogICAgICAgICA8dGlmZjpSZXNvbHV0aW9uVW5pdD4yPC90aWZmOlJlc29sdXRpb25Vbml0PgogICAgICAgICA8dGlmZjpPcmllbnRhdGlvbj4xPC90aWZmOk9yaWVudGF0aW9uPgogICAgICAgICA8dGlmZjpDb21wcmVzc2lvbj4xPC90aWZmOkNvbXByZXNzaW9uPgogICAgICAgICA8dGlmZjpQaG90b21ldHJpY0ludGVycHJldGF0aW9uPjI8L3RpZmY6UGhvdG9tZXRyaWNJbnRlcnByZXRhdGlvbj4KICAgICAgICAgPGV4aWY6UGl4ZWxZRGltZW5zaW9uPjM4OTwvZXhpZjpQaXhlbFlEaW1lbnNpb24+CiAgICAgICAgIDxleGlmOkNvbG9yU3BhY2U+MTwvZXhpZjpDb2xvclNwYWNlPgogICAgICAgICA8ZXhpZjpQaXhlbFhEaW1lbnNpb24+MzkwPC9leGlmOlBpeGVsWERpbWVuc2lvbj4KICAgICAgPC9yZGY6RGVzY3JpcHRpb24+CiAgIDwvcmRmOlJERj4KPC94OnhtcG1ldGE+CkroQPQAAEAASURBVHgBfN0FsGVHtcbxHRjc3eEO7u56Gdw1uAwkSEEChBAgVAEBioJCggYnBAnu7gzu7j64u2ve/nXyn+p3a95bVT3dvXrJt1b37t17n3PP7PPEJz7xmOc973nL3/72t+UEJzjBss8++yz//ve/l23bti0nOtGJlmOOOWb517/+tUT4yn/+85/l73//+xg/4QlPuBzveMdb/vnPfw4+PYWdv/zlL4vxE5/4xMvxj3/85b///e/yj3/8Y8ixg9ihT8Y4f/oK4kvJBnlyxmcb/JERg3HYUX7TwWdPDSd8dCJxGKMnJ+yS18bTJ883OThgooeXf/1szzbzRU+bbXrlHw48JVl+FIQHx0lPetIhwzfZ8mFcrHCWBzU+G40PY8fZo5sM2+TybwyFYdY3hq+ebZhDdvDYKXfNHR08xXi+8OFIt7pY8l1c2eVPjiNycsq/dvbJ88GeWrEGTnKSk4xrgB287LJJH7FHDyY1oms9i4svY7CVC37x2AtHfWPlKb3kyeIhvqwjsnxps8ePAmNx0UdioA+HNhtk84MnZn02w9E6ZAO/nOLDRK7csKlsxUqWbVj+9Kc/DXmYyeLxTUfukDHy+TOOioUsXXx49cWSvfKOr5Azbi9g+69//euwFW4xGOeTrhjZRWzSg40eTNkky0bXnb2t3IeJb7jJIr6MyYkx9rTh4pM/hV06dMNC3xgf5MnABjtsarL0tNkPP1n95o8tffLG8k0P0YOzONpTsk/GGBt0+BxxpQhgBgTIiX7gKOHrZ4yMvsJwNvADmQ1jCjImsZLFromkj8gbV+Mp+YBV33g+47EtWOMFzTY7TV6TYxzRNcY/kmw89tlB9C0mfJhbMGEwzo9FBVOLx3g41PTJWQDkLCSy5E52spMNnnF+s5ldsZE3qWyxQYbNiCy5cBTLKU5xihHjH/7wh+Gvi5ZeOYRDYZP94pebsPBFng++4IG1PJE1lh025pibY/bYyYac5gdfYQMOfPbZCZs+2/S1yRXLHBs+PPTCyHY24S8mesnxTUYhL9fk2AsD30gNv/njS86zQ55f9ooLr9yGiZ9ss6lPhh0Y1Uqx8FfsxQM/3oyTjTCwabxNIZv5gJEN9uDCDzM7xhGefCD26Oino+aTfOPx2EH6jZMll741giePCn+wNE/aybInLn2UbNir+Sp3yZKHna9K65lNRLYckOEPDrj0kfFTn/rUAwe+uaFH1ph+/rtmYRFzmOjxSb75o08XafOXHnv67GkXp1qfrfLIRvGwTy8dfOMK28ZgYUfhN4z5xBcf2eJc7RwLPiGKiEDJokBZ4Io+uXTJ4wOJr+2JJmds//GPfxw20lWTQ/TIdMGEwbgA+eYLkZsDyY5xbZj5ZZO+NlLrS4oEm6wmzLg2X9mnzyZeGzf7JRsGYwo5C5JdcWcDn1yLk74+29p0+cWLb0wMbNDXV/gli59PfTGlLw7++WPfItPGQ+yRZw+VG/349MpVMvrp8k+eHJ1i1HYiQ8bMoZsZjOKRwzYbsnKFwkY+u/zRQ2Llswtemz1EJ1m1PMx18yZ3ZOnCqIY7XidUNrvQYSm28olHp3kjz5bx5iA88BeP2visl33YjLHBFv/yh6fGI0sXL3x8NyflY6sem+ybB7j04RIDwmO7nOWnvOnzqVYi84c/62krbBuHd84xm+Y8LOWcjhjtD6jrqLkxxg45+NkWJzza6YuLnGKs3JEJPxnyfCjWnvHTnva0y+9///vRly9+4BVjPujpw8VO6/bkJz/5Hl/4iL98sadPH8EFo7jkKKzhJMe/mi8HQLLx2IADZZsuWwr7fGc/Hlv8IuPNsTmhrxQvWfmBIf1yyG6xsLPaPHbjZaB2Gxhn+JQZd9KajZI3zkmALQyyBceJvnF2ATNGx5i+8eT4IouH+JMUfUVweOS7cNIn34TQQeQQfvHhSWDJUIuD3fwUF5wKv2yQIa+dr2IgJzakXe70Z74xdkyemlwxwqYt1+zq//nPfx41PjuKWJAcFBfssCnNDRn8NnMLUZ/Puc4/fnkIl7o8GhM3XTrJqsnUn9vx5NCF96tf/Wo5zWlOM05veOFVy40aGZMD/sXMTuP882GMvD5c4laHp7WsJqOQcWGS+c1vfjNqfsSD+JAnfbLsk2XD+s6Hcb4QGVjKDT9shlmN0jWefrjZaGMSO7JGyFkDSJutfLGH5MKYGwzs5V8sv/vd7/bEbrz8sBNe9ir4CgzG2WKHr3SMkS9P4kHkw2wNy5ebNmzyR56eWJHaXMiVazJfZNjUR+zyj49HT788DqH1H/iMVfPJd/PbDVo+ybDn6VxNTpzx9eH1Gg6d/vSnX9wwEB9K6wEOOSkPZNhkT0020haDIq4o//GsBTLNg3Ft+tnEk298saLGyRSLNmzywH7rTF8e+SkGthB78CfDLhlFm+2x+hinBAxqYgiWgIwEXM2pWgFKTZ4dNrQB3dzcXC560YsOQGScNt785jeP5Ftk7ASU/+wBGS68cGmXMH75Y0NA2mr2onSTo2MhleguAuNi5sfCOOUpT7lc6UpXWs573vMuP/nJT5a3ve1to7bgT3WqU+1Z1Owg+gp7Cj8oXHDAia80GY3TxVfDgMgg8WgbZ4N9F5xCVj7wFX4sKpuufguBDfrk+TSWPzGwz7cxZEw7nbCwHxbt4s+WtWIcX2FTLZ9XucpVxo3jF7/4xfKlL31pz8ULI1zkmnPtfLJdHOExruhXyiE+Coc42CWHF279xuC2Nntaxg8D3+UrTDaT9PnFnwseYiNd8uZFbT7Mo7VYzswnPnkEM6JD1phrSi1f+ApfdOiLrXEy9OCyJsiUMzx6ZPCV8MOnjbTpkNMmJx/yU+GTLVjo8d+cmnc6M0Y4ET7bcsk+Gfp4ivjnuSKD8PiOwlrNv0JGflE6fLbGZqxuJG4s1772tZerXvWqA0c5h8se8PWvf33ckO1bbJiL4oSNv3IhV9oKn8aVxtlE5CLjMLETnx88NhREDl4x2ovUckqOHh3YycttNunya72VYzz+FES2mtycq+Izrr2NMAHGkODmwAlxmCJ5RdKST4ecBSQYd20BXvayl11uc5vbjEkRDFkT4VT0sY99bCQCn66CYGnyJQo+mOKTx4cpPNr1+UD6iC45/MbwYdWXaERGW5H0zfXGd5e73GU5//nPv/z4xz8ei+sTn/jEmDCPvE5WbCN4tN14YFfg7oKCHZ5iIasfX0z6+Hjs6MsJnQiPX1jPcIYzDCwWDj1yLhYXQRjYJasWazjZm3OhzS/ZiE3EpzK3w8luFAY+bFbWCCxk1Be72MWWO9zhDsuFL3zh5Yc//OFyxBFHjJuIpyNPJb/97W+HH/qROOBiA4YZJ3/yU22MT32kr12fPlt0zLFxOZNr8+lgcKYznWnMtzzKM9w9vbFJvrjcaOQeXrbQVix8VcgUh3WB8NhrjrIFpzacckDPNSWv5zjHOYYfY2zDDydii67aU57XM3Da6IqfPbbpllM4yhVfiliTYZOsgsojPGTYb+0Ykxc8OW1Nyif5c53rXHtumGTxy4Oa30g/32q4wspf+Mt7vHDr80lPTuRQ7Hjyhy8fPsuwEcvlne985+V617vewFjMML31rW9dvve97y27d+8ea0RssIiTTb5ar/CHCZZwq8VBlg6ac4qXbDLtqfCSZc/GfsYznnH4aA7YhUe+FbG4pqwNbXri4ENbkRMF8YfEwE/+8ejDVUxh30aQU0ooI3iIgjZwZIzXZ8SdOBBAk7dgv/Wtby33ve99x4Zx8YtffICla/yc5zzncv/7338597nPvRx++OFj4iRJUoAsOH7yF04+8QqQfG3JSV7NDtx8pi8mOvElXwxi408sFsa+++673O52t1vOdrazjaSf9axnXR74wAeORUfmKU95yrgBskcfmWB2+W6SYOKbT/ZhQeQQHt/4cqCvwA4bImuBsyO3bDvJW+jbt28fFyD5Ymbvfe9738DIP9vllL18qtk0Trec6Iej8WT0yYkHD2Y8pJ+uBYcPq/Yvf/nLoadvns9+9rMv97nPfYYdF/UrX/nK5cMf/vBY7OWKPhwRjK2R8OCh8IZhb7EYK098IDnG/+pXvzrW441vfOPlJje5yZhzNhAcsx82HIBe+MIXLu985zv32CTHrpyTCZOavkLGOmnO8cJiTD7paqsV+OTNZnC1q11tOfjgg8fTHDtsv+Y1r1mOPPLIcR2x9f3vf3/gduNzYyyO5NlW6JJXt04aU0fa/KvhhglO2M2HOttuHOzxSd7mZdxh8uc///lymctcZjnkkENGrvHJiE9sL3/5y5dXvOIV41VR8+JaY4s/GOAtd/zQDzs5vPIGP1k3KGRcn4422+Xq8Y9//GLujZ3lLGcZmMjo8zvHLCZxuhbFnW11mNTNWzzY8dhkD58PfDGkr5ab/JbbeMZ//etfL/e6172WG9zgBuOAQxaxRY4f9ac//emxTj/4wQ+Ow4R9jH+xo3DCg08/3jwej11EfsRHgWLJyhDnSsLaAiGXEUaVEuIG4m6s2Gxtwpe85CXHIqOXTYtuY2NjLCIbi3LmM595Od3pTjfswaCYbPZNFh8D8JooNT7s5CSPbTISo+BbhOo5NvHQtSDhNWZBWOA/+9nPlstd7nLLbW972+WmN73pwMgHHOw4OfEDj9Mdv008GbbYVdNT40flDb9Yyol+4+KItMkgfm1cN7zhDcfN7YpXvOLI7SxPVnxi137pS1+6fOc731ksHP18k1H4pK+GfSYxIHGSKQ9k01fP+Q8vHjmHhc3NzfG0JFfmGA74zne+8w275t9mhy/PHUT4t+bYah7JsGuO+Zo3D7HhhZucPtImy04Yk/PK4qCDDhp+rdcLXOACe/SG8l7+gcvTqXXgZGoNsc82X/mBSYFbXHIsj+EnFy4yCnn22fIk7An4yle+8jhEOEh0ICs2+he5yEXGnNAVl2vsox/96HjtCptNMWzG+IXD/OHzi+o353Ag8qh5L1bXDV02kTVKprg8WZrr/ffffxwafI4gFq+15Im8nLBBD+/tb3/7OHTIFyo/xsUHrzadGXdrBXbjSu1i95RBzppzncNiTO1VNfrkJz+5fO5znxv6+tas8fOc5zzLPe95zzHvX/nKV5Y3vOEN482EwxDiD1YULr7EyIc4wyPO4qBnzvRR66c5ME/xjJO1V8mltWqN7I3MgTHFk783P+9///vH536uN3bh5IdNa5IfBSZjMLFDXu7xEXn9cduKmXLBBwqfQg4LlAE3CwlS9CXpFre4xQB8qUtd6n+BYONHP/rROBU52XsU93qLXRujMTYEDLSAYOOfT4EWHDk8WBVy7CjRPBavWNWwuwAs5ktf+tIjSZe//OWXW9/61gMbffHMRI9vY+WLHbwuqnwYR+Esb8bFIkZ65PgpNuOKWNT42mSuc53rjJxd85rXHHw5DUu+1HJrg5NDnzc55VvM/CE6Wyn/cCp88wlnWOjo0zeeHW22xaXYNGBz073+9a8/biTFwjZf5Nki58Nsa8DTHP3G5psJXri02SMbD5bwsK/NtnFy5duc03dT9a7bRuLGQc4cZkdfyabY9eniWStthJ6eXNT88oXIyVu+8YwriA3+5Cq76vKqbfP3Ws1TkZtVOsa0FTeYjfVAhofU7LrGEH9O+J6y2POEDdOcPzrllI5xPNe1vLGhz59aXvHDiq9dfMatve3rE7LXQW4g8NNxsyHHX3asMwci2Gzy1qybj/mXH/blUuGfHbyIHfbKPSyITNjokOPDq/U73elOY97xPR254bLx7ne/e9m1a9eQ5c/h1jp243YgIu/w43qyZsXp8zxPOtZDPtX2Bj7hCtucR/7IyYdaX5EPfbbLl7nggx1vIMyvNt2o+dDnB3b5h9lNEU6vj+WYn+To6ZMLhzr7xuSCL9jgUtbxYxe7Acqo5KvxMs6ghZmMMfoM+VBUOeyww8YG527OZsmiKwGf+tSnls985jPLHe94x7FpexUEnM9D7ne/+40FAyif9NV0FX1Ent2SrB2RgRE2fJPHRoRXMmwkLiyv2my2XqkZ65RMD2aLQJxtbmyR68bBZ6c6vLCpFdjZ0kbFwwZ8+skaL850YfAqwo3ZayufJRRfsmwgfvD03YjJOyE961nPGovPGJl88g87HjyN64t5zqXxYuCr+dduztRyYE2Q10c2Av1wZkeff5sEe6j3rXwbI6PABm92urnwgW89hJ8dOq3vsOK5Wcnp7W9/++Ue97jHOFkaLy/pln82o5mH7/Mcp1I4Pe3BpszxpaOGU9EWnxhgFLO8saPAY4OyYfBj0yObXbzaZBvDMyZOhyLXody6wT3/+c9fvMrgt2uHXnOvRuVCHGzlC25teOWcL/rwtqGwHYlpv/32W25+85uP1yd0kZovtZOturnyJOVG41p75CMfOd7zs883Hb75JY/w9NVwK2TY5F8MeK0RdhzC7DUOEPLg2nLT+NCHPjT8OKnLuzXiSYUOf+WNX18KevCDHzz8fPGLX1ye/exnL2984xvHE4FxduEoz9pstDa0YSx2NZzqYpJT7TlW2OTTa0BrDyYy2SGrZINfdvFaVw6TDhJkkPwpbLGjTacxuWstGCunbI8P0QMvWAIlLDBqwgXPMB45xcbm3aZJd6fevp46yGcnQIC4eH0gzRZ/JhHfKQtoenPwsBlXxy8ZBUuHzFY+eWPswqLm1wdhHumue93rjlcrHgUvdKELjYsU1hLnBPSe97xnvCeF71rXutaex1wLzCSI2QUaFhi0s1HCYeHfONKHJb7FgoyXB33jeHLmhiBfbRbZUs/t7LAvNiUcbGvD2OIjjyeH5YtcdrKfj/hskOOnWNyUXZCRUzq7iI9saSvIqc2NTnyvetWrBr9HbHaRmFuffJZjdbbw2Z/9JCePipPboYceOnz6EoK1ym4Y0x9O13/06WUn/MnJlznxYTXi21oLt/Fkm0v9sIbfZgWDOUHWF1xOyZ4+XPzZKW/6SL/48PIjn3KLnO5hcSJ1M+ELBrLRbFcMZOBTwkxWmy3ybKjLn7brwvV017vedeTb+gtrvvSV4o/PrqfnW93qVmM9vOlNb1q+8Y1vjBsNX6jc8qWg1rJ4jCswGbd2tMXkSUg+NzY2hu+f/vSny+te97rlHe94xzidwyRvrmk24XGNs6HPHhk8Bdn77n3ve4+9xE3amiJbTuAWJ0zFDQseuZnHXrL0rAF9N0Ovog888MDx0YBDJJzFT0+7+aSjhNc4PA4p1pbDibHGyYqVz8bgCid9Y2yIg+woHDJiMCogAtrGySl4JdKd7Ac/+MG4G3ul4v0seUGryaKAuggsaI99n/3sZ4e8ZBsHtATT0w88W3yrBQqrNr18xJ/9GcdXtHs6cPL0Abn3mi2CkpM98nD6cNc7WZ+JuOP3nhQ2MuVPH4kd34ZgjF1j8JaXFjO+cT6NR9p0jbux0POqzyO0hR1lX+3DMo+lXmfwj8SNOonNeOdFoF3+yPBfztT6cMIBezJqGPlXfCtkx44d4307n+JyomxzLUZ6SF/burjEJS4x5Pn4+Mc/vriw6fOH1OVKHY/+fHMpJ+Va/pJ3ILDZX/3qVx+Y+CfHhraSLDvx5cAN3AeunZD5Nx7RE0dzLafGxaCUI7b01WTKk77rCs8FvLFucHDe8pa3HKdaOgjGKP94jTemzidMbpbWu03oIx/5yPBhA0IdXkZn/WeOXVzlBEY+zYW8xacPg+vLmLn0RH+jG91obHLs0SVDv742Xvy59orI2wH7hLcTnsD4FIuC5Cu7+sULA9vGFOSzAl/Jd917DYXvxrRrfVX14he/eFzr9q9smAN4Osi4vnxwjZ9N/q0Hny+xjW/+fIHFB+zWJSouthG5/KjNnVzKo355VRsTjxu/Pcibmwte8IKDX/745CP7fNChG58t8g54/MDeem1fNM6W/GqHiz0UtmzibeMIBTpQg7n+U1IYRuTwcmqRuylItM2NEzZMvoQD6vQumTYFj4Qew9z12fHEgoA2ZqEAzk4JsXDgJG8MljlB+vkdxtZ/YAi7STeu9oHp3e52t/H+uieHWUebrgJT+WELD7ElLjE1SeTw2SSnXc7oZFMblcf42TFmgvguXq8inNCdnvlkt/lyE/fd9Be84AXLFa5whXEDYYNddhCs7LNXHPDyEQ41HXwyiL6CLx7EDio+ego5F5Jvsmxubo4bKJ5Fak6LJ9tsFDseWTdory+cYL2nxbeuxJs+HW3ye8OAD4+YtcmSU9zIbJpyqLBFBoXB05N1Ys1G1qVvW3nF8YAHPGDcqPNf3mZb9PjH41dO9fnAU7TFVb6t8WRhttnv3Llz5JQuoheR5RvZrFxr2ZRz+M0VOYV94xX45QAlUyx4fNFvvuFE9I01H80NOWt/+3ogdMr3msiBBpHPh375yLax5qGaHCLrJksmYq84wmMcfvqtD4dD4/R7IrJGjX/ta19bXv/61y/vete7xr4DKzwOfih71q714ANza9G+h2zC5F2XDs+eslyf1i9br371q4cfsnJfbuWLHvtzXvDNc9j1ySiuB9e2p5w+g6EbiXnOSTlonC3+xeKG5w2MuWrN8UkGqeWMjeanMT7w2Ar/sbvBcZ4oUaZQaeGXAIocAuGk4WTswgeGDjknPXdhJ0knHx+imjgfjLkITYgJdGHS4ZddVBufnxKjVsiFrYAEZeL5bpL5gJ28BeQE6avD3h/6cN/iYr/kiVsMKF+wOJXYeLT5jfjilx4fYbPAjCnGkHZx6ZM3lk0Y4s12XTw71hP9zW52s7GA3KjJlTMxeh3h3avat0SM56MFUI3Pb/r8whDPuDj4VYdRO9niSoZtixEfWaQ9cQzG+g9M9JGaf7bp8s++Iv+eYjoZi48sOfa1Ffb01XDg6bfR8mE9sMe+D7ed2nptaY3QTW8AW/+xKXzgAx8Y69KN0CbsIPTa17525NgHkL46CSu7Cio+9soTPkzh1EdkERtIrHA6CISJHevOqbNNRQ7yS06RQ5h9a8hTBTzitratc5sFCie79GBijy4e6uDTvPJlLF1xpYunXwzJ+mBajm12NtHwkleS5xt2/XT146kR/23otY3Brw+rNhtzTtifcYvTmrQH4fsygTcLu9anD5/bkmdPTEr4ipcfh7Rs6sNFh6y2G6ZXb3IuB+bvuc997ogDxmJiQ8GLmodiYVdp/fJHv7ViDDa8bOkjtXF84xVx+dDda0Xr2gHe9WW8eVCjGRt7Ch6bCK7mbZuOQUKMIe0mZb6BMCAYpzSP1r6D7K67sbGx5+I1IW4e3mW7Ybjzs2fy8Gzk7tZd7Ow5Cdn8fBfbZw7f/e539wQWFkmBr8TMk15wBUjOKYwMrNoPetCDxmc1XgP12opvH4B985vfHBNu8n2QjuCzGbLtsc8EZB8muorxJgEfzmKb+bNubbJIny6Sd31jnticOLzvdAGQSddN2h81Oun4/ryLthMlGSVbaosU3i4OGOfx4uOXn3IdrvLJBurJQF9+fIvFHBeTmv38WWdu5DCTdYF5srKYySmoC5g/myocfDXHcJNJnl0Y6xd3MjZGr1J8fnWNa1xj2VjXKlk6zRO/XlfahK1R+GwKNgSHnBe96EXjG4K+jCDHzUE1ffjEyi/M4ldaH9p80kmv2pg8KnLobz1sxuxsJTpsKvIDs19IcEL2JOK0isy3r9H7sFcuEV2+xFre9MXEF+xK+ZQnmIohPPja7BUz+94mWIeub3rGmpd0ydXmh7/WiLFyoi3fPq8gb93A0WGFLuyofLA1Y5IfN2E3U3MvP15b+cDcaytxW1vWQTcEuWKD/eZEnwxZWG28fKFd603IoZGsg7Ic2PO87nT92mv4ke/WXbjzw1Y5ZhO/MTr2IfZs+t6geI1Onh04yeqL1yGHX1jtdV7ZJuumDpv91Q1xY70WxJMNduSaXX5n++EyTh5mMtswNBSk5lBNEBUwWcCc4N1lvffzhGEDcaFJpABe9rKXjT8k9D6eDY//gFrcTphoq7/t66OvBW8z//KXvzweBelEBVey1HMg+nD6gChZbRuVE7xF5BRqMhGsbh4m5UlPetKY/AMOOGBMDmxOnp6SutjkpA2HXwtGzZeC5EebvkIH4ZFF8ZPDV8TaZLJN19OHxdKJnC69cvnUpz51XFhkxN48kYElIs8eP2xbzPouFjrsovTywUZ26IifLBtqm1QbrZuc+XbBIr7YcdE6VJDzmsXNztclfW3WiYg8W3DAZGGSZRu+eMZtHnC4gDs9wcRXcfAZTnOM/G2PL0zM2PjsxseWi8o3ab797W+PC9Z/ccA/PA4h/OrzRReVm9FZ/8FXymO4jGvP80BGCaO2DcC1tbm5OV6xih0ZQ/zJC6xqG4Wbm83A5mCd8GF+nH7lxFfSHc6MG7OhyP0XvvCFEZvrgX32xNrGwBf+7Lt4m6sBav0n/WLELxfp4JHDJ6dtTF6tEU+JcgEnjHKtdrKXl7e85S3jEEU+W6NxnF325nyRY88aM/fy6gs8nia9BbEvRGTFFP7WktzIJT4ZPuSncbHY/+TSWsL3lVk3UGvN5xXGrGe5l+P8sFVhG3b2yMlLfvjXlyPXjTjMJ3m2EB173ec///kRnxukg8gTnvCEEX+5UbM1zw3f1j8SZ3OiDS8dfvC14VHwXGfbCqiFUxILSJ+iYjN1atxv/Wqei9IkuLi8qvL1QG1PGP7IxusWi0Dyeky0wNlxIwHcuOBLHB2gYCKHjwKvnokMnIox8iVc28LzFUsnA5sbn/gl21/wemUBg1OnBWbRwWey3AjF4/Sfj/DoR+Vu5oXVGJzRjJEMHbzyLCcWi5OMG6rTqFNMOWHHhHsC8SgOm42VfrbylV/xyWvjMCHjs139dNThos8+v2p8bRuYb/e4UJx2+HBqT48/c++05+Yh7w4YDhvwt3DJk43yIW5YrSHjMBlDMKFkw8sWUrsAyFlX3k3LdxeFcac6rzLkWyywWdN8wcovHXHRsy6sr72RPDbX2uyXS2121MYax0uGT3PvtOmQ1UGHr+KGxwHLurRhWd8+3DUuRv7ZETMf5OTdu/OeRJzEk3HdyjFc8giPsfCygfAbU8sFX3SyBQO98kMOD2Z2rFHjUT7YsvH5LMLnXr0GkwNxmA83lfzQ55dtBc35ZBffXgMjG/TlU3v3+jMkPoPtq/p0rUN6xue4HJbtB8abD30y+cE3b26AL3nJS4a8J15vM3zWetRRR40nRHLFDHPYrC0kpop++aMjR54cfDPNvmCs3JO1Ln3RwJd9PKn5ezrrWZy+jCD+cqWWSxSPPZTN8lCM5Mkab37FM57IKAvCAAGCFLUzmEOgbGg+87Bx2XS9e/eXuBYAORNnMXdXtXDYLxHZ5kfhP55Exedbko0r+uTCok1WXzEeufglzgR6BNZmA0m2C8fXA92pJUs8bHgisRjIOKW4uYiTDOITkY0HYzjCp4+MhYs8vgVbXPKiFJ+abRedBQ63kwRih74xG52nOheZBe01kHf25oYcGxFMLgB+6St4+nsj4yi5YobZGtHnA1avMfn1GrMnifTkkazXVnLtUAEnzMq8oYSZPL4LxcbmdaYbuLXELgxkyhM81ky5wZdPfWOeOJ0Inbi3xiVHbhxuIL5J6LTrJsOX3DQ39Mqfutzik2tewwSfYowsvpJ/fPgQOTLicjF6WnBahmUm+mT8DZU16+c+bFhe8VgnSB4UsnJv0yDja6XeFvAvJrZ3rE+28mwerHOY6JBR4CunbBeHNoIbHnw3Mde8OfOULw5kjC2fz5DZXJ+qyM3kOnUt2j/MtSf+Nn7+xVK+3Ei6VvBmrGyGiQ5dubAG4DRvyFw1v/rayHjXs/XKNl902UDss9ucwVDe2KEvVq+cfbHI2vPNL2t+XqPygtjOnn4xqfMBR7hc9w641mzXmjGHMTcP15ibB53t69scfvqZHX8sa20jOg4T3l7IuadXsvImZvowwKadjnGYkTY7ZMZnIA0ywJiBjFAYgquCn9CwSZFDbiAWtDuv1yj4FmLGAZBcfJNkcvjig32AJAuPLLls6+MbN4lqvOTV4YSli5tvm453sd4XmkR+3ZXJuKh8+8LTh7t5Nt003ATdxcm5CW5sbAxdCwBeBeW7xPMZX0xwhz37dKL08MgXq3G+5dVC0Y7IsgmLhfKoRz1qLFK6bjI2BacUchW6xhFeNf/6YW1MPmd5bbzmKaww2HSdbK0HG1FE3unYHFj0TsxuBnKtth7YlbN8hUt81o+btg3QWiADL7v80sOH2XzQwdNXK/Lmycca8JrV05Fxha71BH8Xj9ccNii65ZgvcuTx+W/eipU9BIdxBU+M+ZtjZFsOyWfXOD9qp1YXu6eQiE1PEX5bzus/n3XYlMqPfNHPRrmhbw27UYjTjcMciMUrFpucdWYTgVVsFViKhZ1iU9NHanHaMGH2mtiXauiyZ/3uXk/7noJscuRtfGIno+9AKiYbIJxovpHpl0e+5Y4efbk0Nrf18eWCvHUwzxlZ9q1D49knY2y2Z8z6La9udvyTbX7zQ5Zf8dlHzNXGunfw4aQutuzngx3t8hF282cMWYP4/PjsRq77xYQhsP7jVbsbhT3NvIoPFj7hIM8OYgt264wf4w7J7BpDYuSPDTjKO/wKXtjlZPwhoSAQJaU+o9omgiGneSekThIMAeyCRxJugiUNEH32yADOTglRV4by+g9ZvCZXn52SzLZxxLeAEPn6EmOBuAu3sXmUc9L0p/z03Uyc3Nhmo81C3zheiyMs+PnmM9/8wkUOTjLshKdcVpMtx2T5Qfhk2LGZKWxEZOXTqx+bI2qzkFubTjiMwafYzL1aZEuBjY9iKQ46eI2lX58uPwhOrwTbyOIVj2/o+ct3N2uYFXMCL91yW3x8NdfmxonUK1EnV2uHPFky5qr88metlQd99tkzv/Cp6RWHNciHjdj7cLouti5y9snyp2aPDBvyyzfiK8Irr+TLlTZKh414+OVY25Mkm3KMz3fy8vnwhz983PTcFNgnw67DEaIXv+sMbuWZz3zm2NB37ty551q1XtyE+CTf3LKFx355Kw/FLA+IjD+i87cVnpZbf+S8HnzGM54x3lDAtf7X2WMN0GMXmd/WcpucOv9ktOUMv82RXzaNyYEa4cWHRUzGGqfXXkSvdUg3GfpyRp9fOgrCKyf15YYd+jZw8bgpun77rMKTg3VItnllhy80+yk2Y/hsw2rv9Uelrrsw2MAdLuSGPwSHm5jY91s/avB18MbYFoPr0Q1e/jvUw9W6IYPI88U+LF0HbDe259d4XYhAG1C0gVH0JcH7bac1hnKg7YbCkUCBKPklpgSz2cQ1xna28BRyyZZoNRJoxI8+jL2vtHB9V9/nHi42JKHu4L4VZlOR0DYMukgcxZ1vsfAbhhKr3+KkG6b0xVh8kk+WDj4bxUwXXx8/PZ8peI/qqa78q52anTa8wjLGLx04xYjYSsemZCNWLLb45OjymX9jc/7DKBfaavIWnM3CHzV5X42HkmfXwnTDJmvNdFrOR++Rh+L6T7jMnZuHb0J1k4ePbzT74NecZZOcdca2tWBT8/rKqTCSK6cuNw5fnoDTh6DZgL31R6d+FxD/ZBF/xlFY4OmCxyNDl542at3AopArLrazCQcdY2zuWr/t40lTLvUVxEY3QDybQ2OtYzFb/8aaT7gU/tzcbaza7MHBL1kxFUfY1HTZcy1trKdtcnQRebm17thi2x8ENu/mR3zdxOQg221iw9D6T7mzPylkycCmoDlv+vzjtZ6zbQzGOQ/mo1zwpSC69Phgb76G+YcDHnLFrU/HXPn8sleR4TNGXl++2MSjzwfSJsMW3+bGa2LXW58z0ofTT0L5xqsDsj243InPDcLNZvt6kJ6JjIO0J0P1xjp37IVRjeCBTR8OdTmBG0+9DeCUKBYggII0JsGELUzjxjz+2shc6IwhC8WYjYNejujxQ44td2OP0P2mE10nZQnxuCt5SBDsKQWBz57CX3x2XWDew5o8j2lNijqbEi0xiviQxQDvXOiwjwe7xIsd4dUWozZ7YSUTduOIDh5ZRX/mk+PHJuodpdcv3luSFaNarn0dme7GOvEwybUL0zgKgzz63MFG7j13Fy1bZKLkYSuGsJPRLl6yFqabgs893EiM0UNsy6Vi4buB2yTkGT44YXZKcwMMs5ubm6NfCrW5w06PvHmTG/PLD3/8KNpk8l1+9X3Zwwf8Nld+jMnv7vW1ymMe85iRj07Nxq0n2IohbPTgNw6D8ag8kolmDMYb0w57eaajrciZ9dZ6IW9u4fXZgGvGeDmgSwc1BzAb51PJn03e0zjs4TEmH4qbEr9yqc9GechHtXFrwHXkD+i8tsJDMMOgduP2c/2+PeXDcfjxkTjkkw9tlA38meiQgy/MZLTNS7jw+MYXo3Y3inKK52ncax9PC4h89oyXn2zNumQRGZjaP8gi49r8ouK1BskjuY2yZyxdOmTab+D0FGE9I/bJK76t6r8U0HZ9h8vc4PmChevITaQ8sy1eRWz8hEM8zf8cE0zl2bjrwHVJb/wlegYArF2gKVrMjDLGqXenT3/608eEtCmXfA5QiQgUvk3RazC/n+XDazo2DB9AOXm7awvEJBS0YNmCpQTTCyuer196H+9nSmzAEsgvOYmCMXv0jKUP69zOF902MU8uJimiz65Eyk35UsPDvyIWcYtB3xhdxD7iW27F55WAjbe8l3NydF2YbrbafLswfZgqPsS2C2TXegp67GMfO+YKbvLw8hVGbfLGyi1MCh4Kmz49RUximSl5usZggI0+Er8x7+F9VmNRw9+Y3yLyeZpcsk+WrxYy++ziGUsP/uLoQi2/Q+i4f8JXLScwsS8v4tena33zAQd5snzwHWmXg3JEjzzd5ptdPuZrggw+ff6Mq9ksX2x4qvSZnFci5n22Xw7oWKPyJg5rnG98hR/4jfHL/lzDABteOjMGPNiMy4e2V9Y2NR/6q2ERA5nidchwkGPbXuF1JlzmKHm5zL9DhP3AwYKNyLhiP4CFvnG2+GOr60sfduN01Ma8PfFERJZ9rwR9o7TDh5iaazLlSFtp/rRbm+aCHD94ZMq7m6qbsrlgOz9ks0++G008+M0THbbzW1zhEpObuH3AXLBP1rgc9brYj8Q+9KEPHTdy84Hkzzdo2ZQTn51srIfR4iDDllIexYXESS4yPv4OpAACQIABThSgZhKgCXeSBIhhuhX2WnjanArSJufVjG9sWCyS5Z1+ry2MSw75ksg/LCgsxoHPj5q8DcsrNuMFDVOT3bds2KKD4CxOfXw6cBjTd8OD2YkVz8TzDz8/5EcyVx6s5cEYfQUmZAzxiRqzWXjkdFGaYIuPr8oQXv+xYMSK+Nyxfnju8bZXNRaVLwjIqfmxeTTxfDWvdLXZL8dqfFjj5QePPlvikh9ETonwO2zgkXejwLeIxeePNcWnzxbbvl5Kxk3EWoFDroqfTNjYNcav+MjIiXjU3dTZsT5nCis+m2TosG1MjOyZW21+YER08LZSeSyXzXHY1WQQ/Qo5vo03R+Ej63rwpA6nzzHIlRdySriNhVm7+dF2rbqhd9BhL6LfHJRT4zDSVePjlSd/m2EefdOo/HeNkSNvDlyL3gi4Lo8++ujxxQqYYWNLm549AAavnq0DB6iwyJsxBVYlTPTx2dCOT1exxrze8VTrmt3c3Bw/3e+JzN/P0KWjwK2wDx+/bDZv5ofNcMuftpzTg8HTjcORzyscBBF+xA+SV3bZm0kfJuN8w6LND1KLyc3Yr0/48gHin5yaXrUxN0o2EZ9smBtPpG40xsLBBiqX7JRTeuTKl7UEy/gQ3YCklTD1DETbpM7AGOr1RBspJ2T1JU6fnMmzmLyT9m0N3wxAXsn4Gp/3cV6HdcqiY8Lo8wkognPGVcDk8SVIQtT65P3Rla8/8tXJPj0TxD6dsPIDr43YDcPF4vMUr0PY9KrF131dkMjFiW9SYNZmSz75b2OQE36bMHKIf+TVlddCTnROSWyQNy4+pVyYPCd52NyQYUPknSq8NvD46v2+GPkqL0PwuH/KGz0y7MOsn8/aXUDhN76VbMZem3n37YnNvIkDXvn0ylKMDg+R9/NOhDZL6yn//CF5LAewRORgFh8sXRzJGFMi8tnGY7NxMdHDc5PAZw8PDjm1MfhWn7iKnZyToHfJfqQwTPTIzBsA2eIIBx4ZPsjT4xcP2Zx9zkTOf9DmxkvOtcWWNYXo6KsV8nQRWblvkx7M9R/zoohXzS49VK7UiG1kLpysvYu3To3jwUHGN+98s8q1vn19wjTuJkPGU4gDjTZ/xqwPv2Yhp2J2s7GuI9eW1zTWlPWBym3Xkz5dRQzFoo1s6j779HrbQdAf4tFVXCfeXLAND10kD/TxFLFlmwzc+U/GuFzD71tpvtlkrxCvQ585bn7Y06dr7dHlU5v9ZO0D3tQ4WCKxIjcQ355yndkz4WEr3K5D8+9bkvYIc4bsWeK1v+1eX40qbu7wNPfa+Qlf/XKbH/GMlUopACaiRFJUBKhokyOvkHMXk0xtCWBcIWthSCRZm4a/G3AhIu9FfRPGqcOJe2NjY5wGJaNEs4Fgyjf/sKjJSZS2wr+aP2TyfHjIj43KGDtNGJxk8cLuKcVXf9l1wfZbX3RdID589ZMrbkYuPsTOHD+eOBA9xA85mNX4s38yLSBtmOg0HybfJmBxwOfGAZubXPLs8esChlX8bLTA9PkQf7bpwB4mfmE0biHr0ytnfHWxDsfTP56iLOznPOc548NVc6/QV3ryoMIum27GD3vYw8YGbNPhM5Lf8kUedn3+4dXGE58beLHY1MlYlzMZV8wxvXJhLrTV4ibjIuMTHq8knCp9mGkzMG4M+YqspwMfZhpD8skOIltbvKiY2FZsPHxXhtBx/yTrbzlc/G4GEbt05Va8avLWrnyE32bvHbqTdxjU5VHs2go89NhSsweXdvZqZ4O+9enr5f5PFE/r8iWH5qa1yD5bbNOB0Tp2sODXASKylvxMi79j8TcO1nz5gQkG/unJcWN4Ch7Stkny53Nbe5IN2QmcDU+8bnrmTB82NhU68YtZjWdtiUf+9cmLB8EXGYOlPMZXm7/42q0TY/huer6G7noWH39Im3+fj/ALE14x+2aWuH3V3xOdPRp5GvOrG/ZD+uzKDX3EPxt02RN/cePxpUbtfQMRpkI5ECkCHfCAMsCYCbehSRxAHApcW+J8EOyu78KyeN1RjXsa8OuxFoZNw0KbwfEDh4C0nWC152TBADO+YBH/ZPnAh9vTjh8Q80rHX27TKQ4YWzBk6Tld2Sh8FuNC9FU8ck5Pvh3kj7jc0U2KuItZHtjKL6xIbOWWfViLlV0L0Dhf+OW/nLPh4oQ/3y5QF14nExjSy2Zx6Zcf86KtkDeGB3NxaMODylM6aic6/JmyZz30fzS7kVj8XsGQF7vxXrXpyw2cNnSf41iUNhw5MQ4bLLDxISfGw15ejbGD8MSFyok2GSQ+J05j2griT2n9kGfL2rV5O1W62FBj5G3o/LFJvtzkOz90FDp4M058seIbz76aPblxmvSk7kblydLpXP7otJ7U4qEnZ2TkzP+46ClaLOzRUStyTV7hu3zAV5tMenzqs99c2LB8mOt9ulx4XeV07Ppg15pxEzHGpr2APTkTC1wwsQ0D2zZ2v2Dg4OfgQb7YzDV5/pE+vdZL8ZDvOvLLEvz7nNU+xK8DrdpNz55ENjtsu6b1FTblhF/41OTxEXzaanjI0Nf3GYzNvnySJ8sGXj7KOX18+9URRxwxMFt/5JF5dq24bmBwYKIjfgdbduAXm31lpmy4kZKh3/wWC324wycWeBTy+GEeX+PlgFKJIUhIwVdbDNpIAL7l5JHT10qdjI171Hfa8dgNuInywXZ3KwvEE4HN0CIDRmJNbIunyTJWkvkUEApbuMLMl9dAHpW9KvMoyYaLzw/pwQAbfb6c5rxDxPOhO798WKzu2hY3go3se9/73nGB+MBfbPyno91k4CGTgq+wKxa21E1UE2G8eIszOfrw+k0dNn091c3DjReJnyxbHpnh3L0+muJZUF1cxhXER4tBHz9/YYWHbUVsZLTZyw5demKG0xpwWrdwfb7h6UhuZyKPbG7y7xUmns3ZugqbmoycofIIV4SnD5OLtQ1UbWwmfX7EYLMwz+Rgx5MPMvzxrXgytvE6BFkPeGTEr+01oY3SyVbs+Eq2WgvZw1ei8igGc+urri5s109/w0RWW05dX6478m7Q5K1rfvI147Nx29y9thGvMYTnG4++uILCLR/ilw+yYoJX3sSADyeCAV+Rf0+SfpLEmwb51bf+rF2bmmvGU4V59lrGdW9dGUPk4PE0AJ9fuPDq2WENLn7ECJcSzhlf80OODp/G4fOk6PoQq/3Bf3QmJw419LzOsmE3P9mHrXXCr5iQcWuuHKjtG9aK+bJnWBt8OowaR+Ft7ss9e3KK4CEnFz7DsIeKKR327AFwmUc3EPNiD/TkJ8fmzg2G3/TU1oucK/SR3Jc7fsUZhnLNDoxwNPcDswHCagbVCqLMqXEbJwVtE+N0aXIBtXGTk1Dv3PzFejcGzujYDHwe4RTvFZB3o/SNs8tnCdRX+A8sGwhGvgSCpwhaQvzVs18m9QdMniJsYsZgcTImS9+CsljFYSKckouNPzoWsRsjX254/vtKr99c4GKR9HIlBy4GemybTDWMePyaOLZdhGjOsXH87BlvErXZ97ThUb9TEx2FnLx7PPXLvI94xCMGTv5gN47YFuPMw4eRTHLJkicLs5yJI+xbbegXD32P3jaJbBmf2zYSC//JT37y2HQ8SsufeMQShbc50VfKqTbc8uPC1mcjvMWUf30Xlf/syGYBh7liT+HHvHq69OTqywkKe2jOtw3JU7T3607QMLBFJuJPXvbGI2NcKbc+2PW+v3fXNoLywobPYNzQrD/zC5fPXlyb/JDNpycjcbgWYTNuzIZkY/NqyCsi10hzzJ71II/a8qaP9Oc4mm8yriObHNueNPizyVkvNmi6iC8HNk90DnbskUWe/HyN+9GPfvToO3h4Wp03t6251Ocffm1YZ1zNOz8wIk8bYXQ9uzH7Az161iGs5sCeJg9smp+IL2tNbNaqNxZiNR/2EvHJvQ+5Dz744GVjY2PMQ9c3W3yV83KqxleTUXv9a2+CE6/4+PP62nzLX9cM/PjemiA25EANk48NHEKsC2tBfAoip8CgaBuLz0ZrNV31+BaWQQFKHmWEpyBjtQfjuH/cwX1t1uTTE6C7rwsKceAiBcSEPO1pTxunNidTvth192RbIkwyWxJiDJkstskIRl2f3RHEOqltBJLtBuUG4HEVRsnv5MCmyXbK21gn14bCboucbRekr0/6/9o9bvPj7m4MriaNXhcA/8b5QhZIyYeRfXaSkSslMkaeneLSJ+9Ga0Nmx6aCp7i4xOXCFbObYt9uY8O4mm+46CB2y6M+PttklMaKAU43RGP8a1uI+Hh0/y/iC/FB3tx6PSG3cFsv/LBZDpJl11jY1GyQUxB75sgTgnitN5jk0Zj8hMEF5wnJ1xuPOuqo8QSUf/bose/pdefOneObNPzLR1jgdDr2a7e71yc9mxy8fNAvd9p42WWHXLjNicIf29aRC9+J1a8Cy42nTRswXUQOidVJ00bitWY3DrGG1/q3xv2f79YEXTyvmbwKczhyKBGPPM0UVtdisavlV66LTXz4riEblzF4YIdFvNau2gbmUGcdb6zXHTxiN19sqNmTe/EZd60lR5ZftsjLnbE5n42p8d1oyYnH3sKGvi/sPO5xjxs/tOoLKOx4KvGaSNu3GMWqTRcOPrUR+7A4KInbN65syvY9T19HHnnkeBPgZmJ/Eztd9sJWvtvnxJ5tvsj1ROHzH7pKMuQ98RgrBzA44JIzXpEHb4oc2My3+aIjH3yVV30UL0zq2WbjeOMV1jyoDbyAXYQowznCA8Ci337cqxQ8RDd5dzwbsQUlkR5JTYynBXLaEqjNnuDyDTSgbOHBhcggC8GFA5Mx+i0qtT+gsyh9/tEElAj6FqnCNpsly+sIpweLSNv7S4uPj/DCXy74kqfiYAvu8NNTYFEjMZHRJwe7jchjrwucLR96yi974nR6Q8aywzfi34neBmFRF1N+yJQnYwr/8+Za/LBEZPjgU9uN14brNaRHfmP9gdMcD312+JnJheT3xuTWKbCND07rLVzsll+4jBe32BW5MabIEX/a5goWF4z4Njc3x0UVPhuaU2MXfnbYb4NwgfVHiMXBtjz7exWf3fl2UJtC2OCHG16yza+af2ScTSV98vXxbEKuGzyvXz2tp58PT9Uw4pt3+MoNPDYNp3hPUmz6phubsLvxyRM+PfJqfT4jPPlB4oGzmJKJZ/3ahNk1z+YQNpsam25SvoHnWoxg9MbAiZ+MQ6abouvMGH3+1TCqFSQPfKvrq2Fu/bBpbaBi1caXA08jXqPy4UlNvsh5i2Ij7oZMni9j+ZUPN3c3CU8KxsVhbVvj5FoX1nm4+Q+3Gp+svPLDhzUiDp99elLuCZ1u14W2G7YyE5vsIDYRHPZfrzxhNSfZIa/wp9AxhvSNIbhgZUte9uAlgElJoIIxgQRsGPoSbGOSIKcZJyVGjM3EocXsxmHiBH/ooYeOpwo6fNADJMB8arPVotFvsvADG0bjkqSe8dNBJtYJ2R3XhuFkYWFbnEoJEqeL1esfuNizMfgdJ6cvE8d/eMuPfOHxp03GokXaLhZ8uMmE01i+q+nQlWvvPP2Gv6cm77phJkcPqbMHuxuOOD0xiZNfeYcN0YUZD5505UwhB6M2jOSVGS+fZPAQG26sR60neHoI1taKvsXqdWILmW84nK59Q0u+ndbgEgcc9PnRt974pccGTHyRLwY500bZKQa113nWQa8dybHJF3ub641F+b+oHObDQcITntdW/jKcbfbgRXzqs43gZoM+Piq34bAJGVfIK+JysLGZ2ozE1oneuvAEyC4bZL2a3RsZh0neuiF58jB3sLOryLs+OfboKa4Hc9acqK1HGNkzvxvrkwQddsjadP19yN4oPNVikk/fCBKnm7bNrblmV1turINiLs/62nB17confmPGEXzswdgYvs9sPA3DZLN2E+HPXuXp7f8je5V9zpdrxELf569e0beW3Ujh41fe+JFnPH4QXAjP+LxGrDPrwHXtC0lsuAE7PMxx0jPGBn0F4SP8/PArX/kvt/pkwiF3fMhdRAZ+vrSVfdZXCcd4H1pgnDKKXMgMmRiGd693ba+F/NewJpuDQJJjw1f5/AW0TcJNhzMXGTnjanajJlnQLQAyTTS/ShebINmBEV8bSYo+XQUJkF0bmonwdUuLpLu2i8lG43F2Y70Y3DRMvgvYpDUp7MLMt4SSw7Mw4CAXLv704YdJH0b2xKqPYCwfcPIpRjh9/c4JWY7pkFWz60lG7QTv/zoxJxY8TPIMD3l+YFbDDA+MdBUyyZExLqZyCJs2bGzqazdO3g0MTu/sfeAMP19eERx44IFDx9OaGxxsbhr02YKFf1iQ+IxVyPChNid8s588fQSbtrngI5zm3Aead7/73cfFxy5iky8FVY/O9A8+e+YNeeLyakIsqFySYRs+JCYYEBvaeMblWT+M9cm1voyJs7hs2g4JTvhevTkdw2S9KIh9NMciTsU6ceOwSbOTDh9wt/nyq89GOWIzPLCS5UvxIbTcqpH8I/LG2QpXbWuUX7mDx2c+8il2PptbuUoXHvwZn75xmOa8xidrjK5xtb6Tt+vH5q/vxmlMPptPm/5DHvKQ8VXk4iUjBnbhhNeTi8OQ/c6Nkz3FuGvA/GnHLwfZkCs849kvfrzWOlxi8MrRTfawww4bT3rmsdjkgq/s0O1aYMsNzluVgw46aLy2xGt9hSsbcLGlr7CplH/yxq3LgS2ABZwAJwWrRm4ITug+y7AxCJgeHckm53QiWOSu3AL5M4VfAABAAElEQVQNCFDkFDyTwU7jtcmhAuSnvjp5C5etLgh6TRxdRSywO+307RA6JtopwlOJfjGzpc9O/i0y8lv5sJAzhoyjeXJhghOPrFjESbb8i4eMm4GfroDB1y99XkSHrNcQ/g5Fjl0I8kvHKahY+RaHPtt8z+18koGDX+MoP8bMC93mgy2y9ON5QnMStbH6MNkGwYanUPjJ44nbTYAevPzJF5v62mHhO8z0kYuhi9kYGTi0kTZ/6bHvicwrU3jMnQ8jmxM66VbjIX1+6fkNMR84uxl5ehIXX2KBnYwaHkV8CA8ZZ6uiH25tvvIvB9r42tnyTSo6nk6PXr+5uGvXrmHfH+SKCWWjfOHB45t7XuX6zMamKS/ybKwY2Q5X+saRMcR+Y9rmwwbKliczr66SJU82TPrFCrunDjcyX0KwbsNBnr2Z0sMzzgf5sCRbvsRnDH6lvcfaIDMTX+JAxiv0veazrs0jPlvWKXl9fNh9gcGaKrf8V/DhnXGLh31j2uWZXW1jcLKByOFbCzZs5Gu9rnt7gzVOJ+JLcYDzbUxrRg5gdtizt8DeXLVOyzs+XPHZ5t94+Ng3rq89/kdCAAuOkQLIIEOKu56Ny6OajYFhxpooNgQt2S4yOngcaSPjiE8lvsSRDQc5emFoYowDr58u2eTYxqebLcmHafd6WvcaDm6+xeMUQl+fDDJOnx82tBUYa/OnPZNc4FeSZY9ttvjBh5E8Pv82ufD6S1M3Cv5984qO4ltsNgSfQTjNu4k7VeVPTsxFdrpoylNxGQ+/MXx1OVbnk2yLBR/R1TbH1oOvlDrliIMsTHLLJgx4zYG+Bc0G+XLEHl+w4CsIDrKIvWKrT59N8mxo59OF7dAgzy46+WpDYGv2zZ4xBVb23BRdiDB106CnsMkPn+LTLi942WYXH+5k2evUS9eYOPlON3xqc4rcEHxzyhM0fOzYjMsPOX7pwGtefMbgBqjPT4UOG2rYtfOJBw/Z8PEV8UNn13pD4A9u15GY8k2vNS1X4vL5ER1P/taHtU8//zDo880OgkGbvlofsY/mcf34xcCeEh41v8YVcWXboUls3gI4fCCY6IQBD174ybdW8Oe1yiZs4TMWDrbClX39YmNrjkObP+TvYrw10e8LLexlR0z9h3kOmu0vatelAotrQ+z80oE3THBos1vhuzzB46Y3xg4//PBjjlrfZzNIcZ7EJn42RoYBtdJJFZiClmAg6Sv0yeJxOvMC1QJlQ0DkED1tuOYgjYUr++pk6MTHszDisZ8sO/qw8U22gqcgumExXizly5i49LXzkR5/5c0YKj/k81OcfMRX4ysofW12TSZ9mwS9NiLjiN/80zU/+snXhydZ4yhc2mwjeFD90Vn/KVa1fJODhU/Y2TeGxz4ZYzM2tvAUix6xA5dS3o2j2Scf+GRgY9e4thoZ14aFXL7xyhtcfOPxTVaOXSN82DQU/eLLF13t/PDZ2p7nj239fNDJBryw8Z0OG/js4ynmzSZtjK4xWOklB3f540tpvsOa3zCQJ8Pf1jnAI2ccwZGemM2R3LDRnmLTclqWK3piq+SbDmzFoo3Y0WZXmx4MSjkOgzHt7KSLZ5/Kp/Gw8kcOyV1yxcdPWNhBdOjTw6Onj8jzIy6khr0csUWGHjl6+kqULD0F4fFLDka+zS0eGRjMlTmFXe7JGguvm7w8mwcHD1iKTU2ufvlp/vkyxg+CfcQPmIbBHJYAY/HJ6KdMBrVAjQOP0qtmoyDIJSsRglEQcPkbjPUffXYUPhvPDszG9I2ZGDV/YdRvomZ9MogN/GzgadMvkU2kMbIKIkNfMXHkw6CtzLaNhZ1+ONmXD7JsKuJCbLNDllz9cOij4kmPbHp45BQY0jGOsssmHvnmk93kYY/yDyt9xThdOmoy/JFh14LU166wF74ugDlHs2+6bNPFV8JEV58tlJ4ajnyQhwdPzR5d64+seVSc8vizxvWT1+efLipP2eSHD/IKu3KDT0a7tc4u2Rm3PtmKMQUl6w0ADNZM68Z4uXFNkVXyRz482SqPYVMjcZNNJ0z4My7jxsSj6Nu8yOQXdk+oeOa/vMwxw61ffvI3wKz/6IeFn/p0tNX8ySfSLgZ+e+LDkxs8NSyITTr8N7/65NgmB6OCZwyFQ63EJ88mYpMOImNMjVf+6BXLEFz/EYt8xadTjs2/eW+DZ0vevRHAc9Mmz/fs042DLn/lsZjIWcvJ5xdGBHdxaMvlNoYw1QmUTAEwAhCjDGWUTPJ4ij7SzoY++8b4aFLZapGnS04wgk4fj28lXTZR2I3lA58vNosJz7jSZBpT4FDYmOX4Mp5tY2yygXolwB7KX/bKBX12RrJXm+Iii6+QR2FrDD8bxstfsvpk2G6RyR19PtgLOxlEHi8Z2C0mtsRjHC/f7GlHbDYH4Sdv0eFr0zGmP/Nh4EeZcRUH2zM+GBF++NMLPz4eG4g+v82NMWScHXqK9twnh589Y3KI2BQTG05v7Ou7NsjRwQsHPp7Y5SMZtsiQRWygcGiHjxy/9dXmmGyx8qENp7p50ycLRz7Y01cjMmzqq+k3xi4eX6haPLXFQEdBfIcRjy88BAt5PDI25+LIN1n8cskGGX21Ej444C+ebNDRNsYe+fDhywWecXZRuuxro/zoN1fmgk5FP9vk6euTp4dHBg58pViM5QOfPP18G8Of14/+1jlz86DLrrHio88ePgzI9cCGMXlDZOAjxw5ZdvAVbTr44tKvTb+1ta2JYCylDDNee1YqQDxGyXHCsT5AqH42yNHNLt98lghjgBlPf7Y1mOs/fEgGOT7oscOPm1J8Y4o+GTpk1PoVdvlJJh0x0YWTXnzybOjPccOAB1+28dhIfs5PONgjb4wfWGa7+WXHOCpvdBTybZxyqI9gZ5uNmYqHHfLJ8aEg+JRyg8cOHqo2jr9nUa2+57jhg8Fmok2ODgzZKF442MoPPp0wJJ9PcsXKbv1sqMXDP90KPspuPsov+TDC5IIVk0KXfLbZ0S7PtfHzN/uPX67Jp6NGxvIRL9/WuDbqc4bm08nTWP1izz6binE1IiOH6uTKM9zJkG+d4DempqfIGVvsk+3GwzYyR9rss0c2n2p8NdJmu7jpskkP33rijz084+HOX33yePHZYRcf8QlzdvFmXW1YxcNXTynkXXfG2CQHU/pkI7KVYoZHey502ChuNsnlQ5zGrQMkD2Elxwcs5BDZXiu2LshEYVLzDQsKI7/0FOP88aO/5z+UAsAgRxalNkVGgJmNliTjiOP0ySmMF4y+8QBq4xkXXH1thZwxMmHIhrrCB12yeyNybKXLdsmY5cOJRxaVSH16fGmTVRrPprFktI1bbMUcTw2zcRS+2uo2K4uUff7YNg/lRt8YwheXBRcufOP8oXLcnJEjT4ZNZEy/OPQbU7OBigEuMvyjsOGTtY60EZ2KvnG+wmgMHnz+2cXTZlfR54tOFw35xpJXk8mPGi87+tpyEC8c7MEcFn3zaJy8i1IhQ9c4aiw8ZIyLo1zDDL8x9tJvPB/42U9OHpA4EFtse2XhKRK/zbqnSuPssG9cnc/sqPEU4+GPzxce3Hja1iV5OUqOHxjJZU+NVy5gpmPjw89eueYLjw32xK7wmSyMCh3ETnGRJaekbxxfblD6sx/tfGkj+oh87eaOTJjIaIsx3OKe7dQmG5Ghxy8fcLZf4Dc+5641382LXXLymS5bkTF2FTEo5NjE40eZY6ST73JCni8lPPnYlmOMguGIsH41o/U5zEkAjCH8ajylhBijK9H8NjF8aBvP7jCy/hNgC2YeY8cY+/HV+dauzz4M4WGrRcsGgiec+uzTQ+y4GEt6PPbwjLMnLjr66iaOLp4+u+mTSZ5++dAmDxtMZGo77fBrc2Y3W+RtHOTFgiwsffbyxUe42IGnvBRj48PI+g/bCjkU5vp8IH1y+uzyCwusdOSHj74JAgt+OMMS3mLWR8bJl088/vD4Jgc7P/nORxjVdOAimz11/th3sdb3WYiLtnf7fBlnB3aFvn7YYeU7fHzh8U+GjbDwRz8s5BQY6dNNXszaYlTLZXPKbsQ2GTW7+c6OOr/VZPhDdMhE7CC+wkdWX0H8GesApM0OXTLq7JPPPhl8+MuT2NONRyc+XhjpKuJAxoq7OI2XDzLlFsZiC496tslP/sgbL2a2rBPyzau+Mtsxxk++6Bc3LGStKWQ95R8fkQ+rtsImG+SN6bOhz78+O3y2X8gRf/TJGJtxaudHW9xslHdY9KPxH0oR5IzRkk+IIXWB6aN5wQkAAYJPvoCMsYmXPbJN3gyebfIFQy9e9vGMk8tvtuvDT68k4UuUukVAB1b2yEXJ5INcWOkitsUXFrLsR/HpsldOyiu5MKsROXYUOemEx7cx9vk1R+F1kRYrPbbCQRbhI33j2WNTEQc+ufw0J+LAQ+qwlgd8/tj01ApzeRGzEtZ8qNlyIZcn/sgVi7qFz4c2HTGoyeKRU8pv4+xar2r+8qnmK2IrDGqY2COnj8iIUVwz8YmMa+cbLpR8/DCwL9dhg8eYkm/6bBrD44Mdtulp46vJiLWbGz0HDDw2+UpWe8atHy620CwjbrqIX2PlTzseG8XROB5f5qo1Qscmxtec+3yIj56inT99bQWp8fhiJ7nskplzZrw+PbjD6xqCSR/efLGB8MOnT4at5MLRdSRePDT7pDfPQzbKoXgdTsQAb7HgZzMsbJFpjG/yxvHyHe54xrPBf3JqNo2xU2z4ePkyZi6LOblxAzGISi4HlBEAOSZHpkTODpOnU2KaPDw2A46PyKF0BW2MfQRw/sJonLwAyKnJqLNPVx9t1UvGhlFykzMWJn4ah4tPY2TgUhuf/WuTo0tWnTydbKr5R3QUcrA28TYBC9yYRY6M2yyMWVj6MPBFT63AIDfG8BVtfpGaDCIfRn1jzS8ZvvS1YYaHTljpIjybF2KDHB1yyYSnGIqbnEIOjw7Kl7YxdvlB5I0r8MXjQ95QsasR3WTDxaY2e8az6WmOHl4bhNN+esbCy5/5Nca+nNXnN9xhz09xF4N6xlpO8Iylz6b55z+ccupvg/gPrza97MBBJzvsFrsxNtUoLPpySmfGYRyxjfhS2MdrY8sf2zDyh9glr1bw5QyPLzw1Hl/6aI6lvrqYjJNvvYeZfjlLJlthVqM5NrjYzn4yxZEvtsSM+GoNkIchfvb0tfNFrvhh12cTsUVOkR/2+HI9Fgt8+mzgwUBemy2xI7oKXjJzLHj5J6ev7trnn3x+txnkSG1QYVzBR5TjUUQMA60fD59OAPQ5S7dJpYsfOG22CjhMYQmb8cAbS59/PvQVRDa/2uTJiYWM2h8MsR35i3Q/F4DCPevYKNjJdnJ48IetZIs3nkXAL0xszvLaczx05oXAT/Lpk4GjmPHnjTO/8zi/+ii+Nl0LjD35UIzXp4fYTI/OnIv4ZJOjj9SKmMjRI6fPTjyy+vkT82zDmJJ98og8OfLGXVzhSf5YyWPjzmb2jM0YxM8HnOatkzMeohdGevLOps8j8quvnQ81e+qupzCrs6lGeAo/1hFd9mAQp3Y4/SW4n/nH80e+/vsBegp77MzERvGxXSxk0jPOHl8oefnQ7gZBv3Ht4sPTZ0O8XRMOGXzMtsUx42QfZjJkjbM1YyEvJ/jk0ydfvPj8ksNrvc046bFrnLzCr4Kn1GbbuK8kK5G/cfFHfeTCFVbY6eUHn3+4wkcmXzP+4pZrMShhVJN13fr7Du0Omt0s8MSMZvt0YfDmAC7rlixsYoiM4TdGhx1rgF34xtHNgI7Sp/UEKDKCSo4+ACUdPyfkGleTETQikx8/CSGBCCB/WUs+neyR14aLn3zSwU8neTLh4dc4UidT4v1KqZ+i7328k5y/9PWXnhY5ORhLKht4Wycy++Uvn/zBEoWNnDGx4bEpHjVd/uTeYiDDn3F6SBtWv9njL6tNZmPsskGmhcNeeSsH5Yg98mGe63TYUdg0zoZ2+YU7u+Gz4P0EB/L5gRyXi/SyWV7YVNhiE6nJlfP8k0Hk0d767Ja7YiGbX2Otr3LPfjiyawxPjh04kqHvR+3cPJysm4PZFlkl/9pITR4fBvatNfzyMfuPR06bD/JiQX5k0f/Pw5bfZXII2rVr18ArN3QQ/XKFxx9b2VeTQemQ4QdWBW461cbxy/VQnv5JNnmvaqxfa3c+kGnzyRcdFB4bYnliRxE/OXwYxDXnJ8xzHGFgmyy98PMhhq47eKxjcuJjzxjc/k8Wv47NHpmfrb8c4Rcu7Gv8wQWPUm7Kw+w3Hh3XsTF+9MPNB2KLvCJ2stp+ZQBucWjD6CfxXXN02JEjbbbo8VVM9OgjbTbJaivk57nNRvrjQ/SEKQugfsqM41MGyLg+RwzlSJsMO9pK4Mmy5zFbkfh0Ba20oIyxgwoof/yzg1+S1SUR37iC1HzHh9UPzO2///7Lvvvuu+d3sJwo/JKmE9z29SfqLahiZYcNG4UbLGIPJvGJgx91mPInDj7n3NGPXw7bQMTBNx32iot8cdiUnTrw6PFJpzb7xc1O+YGJDbzGkxXL7BMfJacWK2IHNVassCoWrx//Iw8T28XLRxjYqG2cvfJnDbCrrxS7WuGHfDJylT11+UqODpnssKkPT/EMA+s/+MbhLkYbs40aLjaQ2maIV0755ZM+nhJp45fHsOSPHH/8GsMnq5+PcPPjmmnjUyO/1sqPm53DECyIPTqonLDbmrGpoPDCQa6+MXliB88YfGyShSv7xacut65tPnrNpqbnd/PkUFthX02XH32krbBnvBj0+RUHnjzA1dzBxQYe/9r5UJMzxmZY9fkq12w6VJLly490+il9/sj4nyn9KKR8++8OisNYbfZqw5ovcbIDC/l8Fmd68cPF1sbGxvgLfzdluNiyR1krCrxh0MYjIw/sNPf884enkGVfHpA2+2SaF3zj4/8DKaGcaSNtRAExyjhjykzJGmc0HTIlir5F5HHbT1A3SU7S68+pjB9Z6+QPA1vsZltdkNozDvxwqyUEhYc8fgWfL08+ThTJwofIiQFGuNPjs8nHC5M2mwqiC0M8dfJ02ETaSkRGLDZgi5Ycn2FxsVmgfhHZ/xHhP2Xyo29igdXCoG+y2cWDQ2ELH7FpfMZUPtUz8Y1XjhvPFh9tXmSUzc3N8WumHuv9bpf/L9uPVqabTXnJNj22kPGwGVf0K2Sy1fhcw1bsybGP5MI4W3jZlI9kssUGjC5K/yWA/7mOPnLa9AunfmyRDfrGskG3eLLHlnE1efywJB+PDzy5lY/WHT2HGDcP43zSQezqO1zQKQ/G+FXCTxaRYzNs2uJhk2xydK2hcKcHF3m+XD8wyRc9bXJtZv6DLKd3cuxbE/7/Gz+wSIZ8mPLNp8IeHTjU7CJ8/uWJvjGULTVb+I0VU3b0+RALefa0w8AezGJ3vcm/vcP1hlyvcOmTKydt1GyjMJRLvthU8GrLJUz8k5FPNtgubutvx44d478VD7OaD7/fdvT6w5v+6pwNufW7WOxsXw/H7IeHz0hb7HzQQ2SV5kCe+Qjf+DuQJoVCQWpT4jQnAohywJBxNTv4+tolQl9SvdZwkvO/wqXj5O8dIl98q1HJ0BYU+8bzyza+fr7VFWONs4GMkffqIZ348IoVWSD8w4JPFl7+9ZFcZEMdwUUWlcuwGDM5xW7cGJv4LbhsZMcCIOc1hf/VbXPdoG0g8imvPa6KjR288scGPh9qY2FT58vCEEdYtcNPjj1j5LX50SfDB/twIjcO/w+HGs87VrWNTY7Th4UuG3iITW25CCsZ+MxHOngKgon9bOGxoV98PTmKK1/kyMhLpK3QQ/y5+NykxSRueP044671NZHx8qAub9psh0mNJwY1O/zo05E/fNiKFZ9cc1OfDgww2YQdwhAe8oRKj006FeOKPuKfz2LVj8gYg7u2Wh8ZQ/Bmj238cJgTY+yWP9e+eMjA6b+xFQ+7bGWPLsKnT16bfXHRicQaLvxwZ5McHOKkywdKDq91mR0+WztqT3R+ffqQQw4Zv2oLD5tq158nKjGyWaHHTvtmmIuRPn/qGROb9NjRtna16afLl3XpRx9dK8bZUWAlR8bTndgOOOCAgcOPcfolZHsu3OSN81NO+MafbcJJRkzWPBm88S0swgb/h7W7e7W2qvo4vp+H+1/owJPYt0SCVFQWlb1tPagQMs3eX8nKUqkDiyASI70r8UDJMoOCdoZZZhhlr5TsIu8isIiMTrdH1l/Rc31m+7sbLG7o5Bkw95xzzDF+4zfHnNfLutbaa0VOv8lKbhOxUE2EDSkQ/5LRIjYuqOey3nfwRnWxjMOU4AjRRVQ9uRnDRxxccdEmavYmx4awodNPJ56N21hzLRa/TshO0gS2WBMD3/p82cCii7N2uQtfzVdOjIVPT/DiZ1HZwTIn+H4xzc9Y+j1pPzbjpav5KPz5zTzyFUOssOGQ5pQ+fnwIv3IE13g28dZnUwx21p4+Xi4afjPCj/e4awpf/HzFswfgyHkY4fOBTZor/+ZgfnzgOTmpCR+FPx28mYvGisMmfuKwxd/Xpvta/XLL3h5pHfXjmH/xjYVvfsZh4w4/Xrjoq82BDaHj0/zi2EHvd9L9NC0bcbxK9XXecg2HiJW0p3Bhb0wtLh18RVsxnk21WHjxgydOHOGwo8eZnROcvotC83Fys+Yw7G9PAsQ1DoMvaQ64EPba7NjDJXzKgfjau3r9qV8G258w4RGYYYiFp69C98rf7567ATWOtzx7le1E3Z6Z6wtP33zKjTh0+s1LzHjwIc3THiTzzW7581skfokUP/mQXxcG2PaHvuNJHL9sycdjNt/S7IJHxFDEJnEyPzpcs4E7bY2v78LS4Ci5DDhpAyE5IZmOvTa7FoUtX3gL/GQh3K356UePAFxEWnx+xvQlMA5NQlwbKH4WLN3kwZ6YYDh8shGHnz5986jf4tnoxGKzI21gJ2tifuIpbJRyUG2MHXxxCW7Gw81PHQ+2uLA1VzyNZ4uLTQCbrTv8aaMtBjyiVujLEay4xEcNb/egDWdygFvf+G7b3ZLN2ytLbzZ+4AMfWHl84IEH1gk5juahTXBs/jOucaU1Kf/NW9981ETNnz08/Eh3tNrZarPRZ1csuYchZ97L8QNDvla/+Yrt9yDM03rALg68+MOfPnMO2vCtNxuxy2WcjLFzcjJOL5YTinp/f3/vlltuWTcWxXFC8VMAfv8BN/7GitEccdOmt+/bH3QEF/EUPORIO/+OCzqihgGvun0mX14t936HE54CCy6e4RYHDszwZ36146TmKx9qvOiIvrVUw6U3Hq6+OMS+ZYc7HzFwLPc33njjupHwBKX5edPab634cSy/1cNHHPlsf8K2BvHT5y+W2OynxAEWrmp8+MBx4wLf3lO6yfVK9Oz2eOr2229fb+r7uV4Xiz7oUc5nbvjCx0Ncoh83/dbBfKynUs6Mn/4mOnLE4ATVV4BKggl3wFqIFoiPvolG0hhfn1LwCsQEnfTYEJvcj7f49UJ3ICYEX2GjiAtPbRLG9MWiYyOGPn54sCFzEfkYM08Lb4wP4T/FxaKYYvDLX5/wiSecYobJZvpkW06Mk/IWRnPW124T2ch8y7dxc0joza35x9M4fVjx4M+mml7fvPJVh1cc9qSxOJirje3VnR/a8fvbV1xxxTrx7m8nOhsZx+Lgw95esm4KbDitp365F1ObFBuG0qbWbs3zi2+++bMVh77Cln+YTnpOLPj7QbLy4sD0YQufvAm/Mfgzh2HPcfhTbwzOtIkbnYPfRcMdaHehXnn4EMiLXvSilfPm5cL2pz/9ae3xTpbGxJsx9PPZ5aIvbrz44aMQa6SEx45P62C/yaP4XcTcqZt3Ovb82bU/wzVmnyjFMJbERc0/fsbjEj5O2bATn/ANW58f4Ucac5J20vYBit03yNnJt/dxvAo0V37mGQ+1vniw496cYdCxU8x5cjDWqwhjxHwUdvYnLL8A6r1lvztv357dzrXiejXqN4TMw9ztpfmpVxgK3uVcX5s/fX1xcHTc5nMmxzYEwnQkQG1gCuJsAWXHh65xY/pOxILa7H43u5di9J7T+T1id3eeLTvxdLWHV0xYCdLGlCYwJyg+MRa/bOmag4tZMcKWFJ9gufrqq9dvuUu4DcFHHiS/jWhuxTce1uRi3NxnjnBS4qdOYChtDnq2+qQLRz7ievaNZwcJu/CLwY6YQzVMduakJnHXxpvA4FfMpdz+sGXTPqDHHQ/x/GqlVyBefbhzZ6+IC4sN/w5mY7CyU8dLnV85ai7G8qFTYNPhkp3+bIc9c1TMxuTVQeilvxNIc1T7pTcXSI8uHJCNmUO5pQtLrYhnzrjox9W86PjO3OPExkmiXDmJuTA7WfidbPMsDvwu4vZGWMZhk/bu9CnOMjj5AwuvfPETi232xuER43zEzE/beKW7ZzjE3JozHXt99vrG5xzEUHBio2ZDJyZf3Oir6St0bPLBAWdivbWtp7o5umB3AqcTiw393/72t/VLlS4e1ie+4sHAQ82HxM+4+dUvFhv6/OlJ+YNXjoyxM+aVnePMo2I34s6zLhJeFfnXBOcOuGzjo80fDsxyaZxeITNOuWPPn8+ZgA1y5pBTQAxJAZs4Pdv81IBtYjVxEPr0hd9yduLm4wrpZV8fmXX3yccJ20KysSAtqn4LJ4aFolPiZBwGoWMzebEtURJsE8TRmIPUAvBzR+HKLfHmVxy1QtiVB7Hpy1vJZZOILdfmFQ5+hD093nJhXFsJlz88fcL+BS94wTpZ07MtN/ho46c950mfLRzjeMc9TsUNI/305U/4xk3OzIHwTbzydBPhxAZbXLbmpVg72OVUuxI/dWssHgzrCK8cFDf+9LtzgxumeGyLJVfWyYn6mmuuWc+8n/WsZ53uJfzddfK3BkRuyiOchC6hz05dTOP6YrIpj/CbQ3mS18suu2w9i3dHXHx2SjiOI0VeCVx4ybTHkV950CZqfmzjJDfs2SrGCf3ELB6duK1BsZbT9sc4XXtfn/BX8KeLU/HmXOSN4MNOf/Kb+PBwkU8x8eYjBmy2ChHD+Ye9nLPP1g2wN6+dJ5588snT8xYfOPFVK/Q4qYsldnlbAbc/xvAvX/TwxDbGvzWd/n5WmJQLMWF439Gvah4fH6/jhk/nV7Yw2SlEXyHyMPmnNxa/xQHBHEsgkgYjzIZkR29cAAJwLiA7j4le/epXr4PQZ6Y9OyR8XdG9AeQRln96MSl6m37GLBFOGsUyviviG2/S2egba8JhuDt2MGaHrzk6UXgJ6A0mH9H00s9dExy2/Nlp8yH6LVx44tGXI/0OBn76RBtmvPIXz6Yxbza9kjOusPdhhJtuumk9AvQYMB7GZh7klLRx4LW2HWjxYccfFqEvJl1jjcdXDQtnY/rZwDEPb0SzOTw8XL+jjZc1gMlefvAm+vRJfMsj7PDNZ9ffWLhOFoQOfn7h1y8WLLm9/vrr1/71yZvEvvnFL36xPrrrJqgYMMwt3jgR8bSV5qZtHuo4aGdvrnFiJ6cuWh7zEp/CO7vdlHnFzk9Mhd8U2MUSO67FLaYYxu1hcwhHmw0ubKawMa4kbGBY18bmScrP2IaZD5zi0cVTu3XbPXlm04kUP/MnYuPck4zmxkYcdnTazQkeXmzwnT7OT944v+GGG9Z5yriPxvaTsd6M7mIkvnF5DkO/XBTDfEj7XVycHSNsSDo4+TVHY8VoPnwah2XcOdaewZU41so1DAI/P/32iJja2RljR5deHP5nAEtCxDJQS66JMtQ3ee36EkSAEQQVJyu/3y35ntG6eMBBwkfPTMpJ0cussIxlA69E4OAuE0+4eIobX5PlW/LxgEnabLXVbM3X82v/AORVhwsHPfGs3p29cY8qfPqGxIddOYqLVy/iK8VmM0Xf2MRhjyOd+alJ+Go6xQYzb/bERe7FL37xehntDgRv0iYRi601i1vx2F2IZ/HV5qYQc2aPTznVJsWZ+uzDw8GnWLzs98rTyYQNfHzbRwvw5E+48SgXHXjiwZg5Zxt3uDDmHEC3d9jFobhsHXhybd928WArPryjo6P16NWc2KfHgxSvdRCjXJUP/amDqxA47JyovFL2CM0HT9yQ4fWc5zzn9D0PJ0rvd/jkjU/oweTvOLHH4VSKiR9dvOjplNYQD+3mkh5H9ubGH4628dahuWSHC34+/iqnMMs7P6Kf0OnD0WYvDnx6OthkxrKu9OzUxWFXLooz90a6cNWORRfoz3zmMyv/boDbo8b+8pe/rBtMNvZBvPhq4yU+Hvr4G1PgGLePCV3CtnMxn2Kak0JnD4TrePLIykd0rb95GXN82Rf+YdoeFqtXSm6CYFgXeGLy0VaImk7BNVy22go+cnH6ZYo5l3h9yVE6uWkDTYCzK1nqJuvE6zFAd8AShbSJnDt3bh0cDhA6xJTiFIOPmBJZAuMpdtIE+Rk3FgZcvGCVAI/Sfvazn607YZvAM+WJ13zg8yFwi228UjKLk31c9MVmn4gVx3jlV2zjLbK2OGybV/FxcofqIl1MeYQndzZksYpPFy+YSjYwlGKxi7u2dSD4JG10G5XvxGKDIwzrzVZ8HBsTi/CNF59il/ds1GI0Hlcx6JVihq2mCwt/duZTLuF59eFgdNJL8HIC90kXFxg3NHzNRaGDw46Um7gUO77isVVPTny94vUxbTdX1tQrDifeTiywj7dHErCdIL72ta+tf3L0eBCWNWgdWpfyCp9MXuUubmo46csPP2Ow4eKRjdpJCUc3huKxEcdFzs2DN/xxNOdilSd9/nDiBgMP8ekV/cT41DV39uEXS9/4zEP8zYewEdvaOsm6KfQ+k9w7joybi/du3VjCiiP/4hcLVuc2PMSLD7/WxhwI3s1PzZ+N8TjyKz9s7Ek3FHLrphdP4ubXP+96DG8+9A9u/1govvXpPb240jcXdbmBVV7w2J2HsXUBMWiykyAQZBFl2BhQ9vWNASZsvaFjUj6y6+7JSYOwF8PJzr/9m4SkEbGUTnxsYdFpL6JbWyJw0m+jdEVugvR89AlutWGWZJsBV5s+O7hE3aLNPiyLGid98cwLV9h8xRFTja/2xKaPF7/siwWf0Gerbs7Z0XUihp+fNnzjTm7mEj+6Oa5Nx5dPcdV8Gm+usMoxP1I+2MZh4i2j7U/2sIzXz69YE7NYrSHf/GHQTyxjco6nMUWc/MK2jsVjw0cufVPC2972tvV+HVu+9ohHrp4nO4HAgh9/8eWKqOHBgknEIexr80k39R7p+qoMJ1sH/9ntcZVjiI3y9NNP791///0L2ytPJwsnuYTNhfJCT4o/c6htHF91/S5a/MK1Htryxw4eP21zN65N5EhxjLooOimXExiOQTHYuJFjNy/G4bAlYinyS+izoa9Nz0Ye4JcPOmM4OFbpce68AxMXvOC5iFsDAluuvYL2BKWnJ8VV7+YBrpidM2CUP/qZC1wSbcV5he8urjnxx9MNuty2R2C4SHifzgWmGyHnXZg48IELB29iLG5yIk/lN72Y2mz56a+P8VLocGBgMCM67aRF0i8AX8LP3byvK/CmuU/hJK7q58+fX59c2d/fX0mxWSykAoN/mCWXrotbcXBQ8IqzOgz6eIapnuICJumSleTDHycHbwsIX0mKC7f81MYDljnM5McLdnh82Buj008nljEHFh+Y8Aibapu8WOaDc2N8wqR3gOjTFy9e5a/cF5MdewU+u/Br68PHj3/zW4Y7f8TjV/z2HDO6avFIeHzSaePS3PjBVVpTY/HQJvk3Fg9r7eJx5fb1EE7eiYOxN0x/8pOfrMedPWuH4SQhBtxi4QuXaBdLv3ja6dXWhX86bY8oPIogYbur/MpXvrLsnZDbo8to+4OHmHIqD9YQpnZzZ0vHTtzZZ8PWmJhErU+MyTF8c4ETfuuqzlc7PnTw6XzU20X5eHs11b40DotkG//JB4biBEgfD+cJAh+OWoGvsOOXXWtBJ46YLtAu4G9605vW4yG6xCsPFw+P4T32bt8WT0yYcsSvOPrG6Cp405V/HIzFWa1fDHsOR8KPPQwfj5554GNcTPsDHwKvdlzgxUMcOOxaY3b1s5PHeOKwXoGsCNsfRk2eY0EBKYSOsCUR1vb/Hl55OAi7eCDmTsn33Xz84x9fV0YflxXciXH6R5YOD9JVUp+PJMTLOF640MUpW7FbIH786Yg2H4WISerDcFcE0xh9sVoIeuOdwLvjMF7O1GHHi47QN6YvBl4w8bZY+oRvHPTp87dR5NI4HX+F4AZHTJhtNvZ0xVTXplfkSk3MqYOQXTbGxNJvfsabg3F9ghu7eONT3zi7cpvePOESfoROYdN8+Cm7sdik41t+HNxhO1F7Vuy/jL165oO/+bqA+M4xH4f0vlN7trVhV57EMSc8HJz6jYlFby5x0J9zoffGrY8H4wQbhpOHx1XEicsFDn/cxJvzczfqPRN8jRsrB7XFEVetGE/HRlwSV+NK89I2L7HV2U1f+UkP2xgJx8Xjy1/+8jpneI7PxloStvokntriJ2GKXxsf9q1RengVnBR2hA37dP7lwFfje3Qo7+zw8tTEecyNp+Nt98YBlhjlCi7f4qSv35hct84w6PkSvIhY8TeuzeelL33p+qBHey0/Y3JhTmzlLf98YfIzll829dXxra3fGtKt/wNBkjMRlCDBkJHSQjkhFdQ4MYaQT6a4q+8ZmzEHgk8rfPGLX1wvuXpjEg4yXoZ1EMBrgk0aD3YVfXyIOp/4qyc/Nvlq82cjphJWerZE392I59AtBF1zzo99/OHqs1G0u8p7tQWnTSFGnLXh4d3FSF740zlZeDkKEx7R5q/IuxOHMVxseHjiaZdLY/zoYbOZBW4xrSl7Ioa583NigFF8/vGgy5/99A+HPRu8zAsm/8aNEZzLNxvCpxO3tnjwOlnBaT750sGkJ9p4ZWee9quLQy/3GxOjT1u1jtYBJr9OfM0TtoI7gYO7mg+9cbwbx53EiX0+5hCeR1vhyRuMcgSb0Ll4EDZO0rAanzmbXI3DJq1bPsXX1zauJPR4iAMfTmsUBl34+cFg5yTNt5zQa+cbZ37iGyvO5Ma+uPiwK6a2/KrhG3cc9qjKMaftvVGfuPJ+Al6JpyePPPLIet+090c6qccdPlzS+hhTiNhJbZxxxIld/sbnfMw3TGNi+QaP973vfev/1rzXnPCDpYbb+UKuFHpFXHs4O/32Ip04+DSurfBhK3/qM4EapOBIp03oCT1BTnIlUBHURcJ/6vrKCldFV+jEwrij8tLbZ9dbGDiRiSwsyUGalNj6dPFkp6xJnCQsHPwn3+ZDB0ucEgCT0M84fCxaHPGlY7OLjwcdG/zEwR23Lhgzj8Uxzk+Bwc+Cq8U2rrDHoxj4siFiahs3L7bFUrtwqeU9HH58+BKxCfwpbOLKRgz8ph1+0y5/2PgQNmRyM67gZhx2vOFp15dLvvDEZmvv4SK/E5ctga0NCz4sJ4r0bPjx916dx67tW3pxfNrGV1XIoYuFA1mBxaa5x1WscHGUM2Owsm0cj3jxy5ettXJRK+fmQU/44MDGfJqfMW3P5v1TmVcq9hCc4scbtzDNg7AJqzYbPPM33rzpWo/mGI51wVN8doqYU9i42XSBjo/xcg+zHNHDI7DiIb42noR9c4SjsDEuxpwfe5jG5Nl7BM5h3k/yicZs2Xkl1z8++6dBTybE4ZeIRcevucZfjPIgZoUvH+PxnW1Y6cUyF3uUDl+vln0rtz3MNhFL7OYchvFw+Ihlb6vxgE9214OuufBvDcxj/R4IBRBGiESAo+ABc9BnU2FjIgcHB3vXXnvtmhCdcXdrXpJ7tGVR2vDGHARqgrDYktNJm24RPOGEn4IbDoSPdpOjg8mOsIVRHEmiU1eyhTGFX/+xziaOM3bYDoYpxRCXn7q5Fw9OczEOizgxstEPBzcLlw87OjlU2/h8+PKxXvzNMb42DJFftmIWnw17enwTehiN0eNB6PKDVT5hEhzKC0x3TH3pXJxmDQ8OwYPAoucPH165Mm86QpdeLbZx/nxh8JUDOn5yd3Z738M3HPf7DvT83fR49em7u+699971Ue5eVfMPw8EHXzxz0cZdvGy0lfixVejY4kHwTeibgzZeCh/SnLT5tQ5ieibuZs2dcjdr9OKzsw9h2it0BDaJY3gzjrE4GYcJhx4WjPzotRXCdlfo4PFnX56sU3M11jqXW5jxrc2/mHBg89OGkS7cmXN+jgkXXV9U6mLRG/pwXOCOt/dpzNGjdzzipR0WLsUyL7jG6Zt/43KBU2PlrvGwcClPMNmZl3V1g4AbXnLGzjw86vQf6GJbawKPOHbxas+kL0/mQlehJ81ZDAVPtosPQIaKQeDIMhCQM8PIA/SKA4hxk/SpFV8B4oCER2CZyLntI7u+ObZHQfAUn3OPULEjDhe+A96YNlvCV8IUej7G2LVpcFDyKU74bAmM8NXs2GhbGJ9kcOdp7uIWi29ttmKxoWMHBzdj8hg2uzizJWyNtwnEZWcNbGS1E5pxOXOCcJepT+D0pYphwwtfDROGdgXf7NPxE1vfOAmrtTBO2BgjbNuUxsPJ1nr6jYLbbrttbeo+MZKtHMwDRJ8Uo7o1YFsMY+xhzXjaeHXRgCcPBI6DzVfoXHfddesGiI7AkmOPXH319dltT9PJnzlqix2X5l7smRPc2NHxI3TtDbVxa9nJvmMKrsK+3OYrln1lPi5sM6b3Gr/+9a+ffgLH8RmOeHh0nIiVwIRjTMGpeHy0CZt480kftnG4/K2TcwUbmFPw77FsMfnhytf61OcHH44iRjngO/Vs45uNvn3ghAozvT6O/L33YS84McMUz/HmH0d9BNYrELbtPTcP8g+bzjrCLjbMuMy5GxeffTlrrfXzMUcFTmsB3z7k7xOk4vJNvE/zne98ZxU29hQM0reg+x8RvOW3vDUH8fnFXR1fevHo2k90/wsspxwANUHB9Sv6JQmgk4GDzJvmnWSMs0fSqw/4+glsG0g8rzo6yNl3woUtcZEtJp8SB5P95Go8MV4C6eqrmzPcxugJDLG9ArH44rVRW+zpkx9dc6+tz8fi46kNu7j44RLPcmWcvY0KHxf/Q/Pwww+vO0z4xPyd8Ng7IHDnR9Rww4GtzaccZMueLX8czbc5s1Xo5YINLFJ+tNkUUz9h687InRwx/xlXG2Y5gVk+wjcOP8FFoZ+86djZ13JuTBGj/cXGex5euXlvoccCzcuJwydtiJziZY+yixvbOMGe68aPTiH8xWzd0+uXT5zhNSd1Y83BXrB28Iqtb57mS5xY+l+F4uDGR2Hf2rGHXV7jWc7Nt/njw1YsGLDjyp+PMiXe8l6MxosL30WwucJlC1vRpmMPv5hw4mM87s1TzTcO1fQkTHkk/lfFeaw19ojoxz/+8fqVUjfA8suWPywSlpzEmY1x9cyJnMeh9Wg/0MNir4633CjG8ZWjzpUz/iKz/XGxdi7w/0rxiJtvQ/eo1nnaeogDGx6RV+38xMQDV7yy1W4ea34cgUmQugkITBiREm6cDgmPpa7cPnHlDVyBCTwbzH+i+2ppB6o+MkqJRZatpMBmA1sdQTG0TYJfie0qnE6t8GdP9OE2P2NwzItOu1jG4hcnG+otb3nLmoOFYSNHMHGPo9pGUBsTV1uOmpO2eDAqcY43PWmO7J0w8mPny9tsbM9hfT+POLi5q5BncejMwTz56jsocCRsYBmbHNjly854OWFnbsb542qsudCzJ8azXYrtD38nie7uwo5LsWHyTSZPsfRJsdhrG4Ol3wHDTt8eMmafyQF7XLxXd+mll57GM87ehcM3FLh780ovDDmFoxRD3OYeB3EJrgp/pT2nXR6N86PzSkF/xtGXj+LZT2y92d+YY8EjK6887Al70/y8evaIQ9vdNQz/x+DRVvmGAU9pf8iPfScuLvJCpx8/9njHPT17ceBnXy741FZPDP5i0Km9MoFNX375wGRDR8TTZmdd6rPLtnVnJ8dEHvnMvBrTh6E4EXvVbC94egLPOH582cBOBxcHEj92SnOrnX4Zb3/MCQ5R87cecnAhLMf6zTffvC542cC09m7oXST488XbfD2l8ETDfuBjHF/ciONEm32x8WqecYcFl55sGP+5onBucpITeQGBkQAlGClvPvpWUOTZOMn62ceHHnpo76677lp3eAjBNa6GLS7SCr0+OxyKZUJisydq/l3s2LNlw1by8WND6OHSsStZ9OYGJx0bOn1zcYLRNxd3dC6WsPjCwtM4P/oWfvLBgz2pzd48xCk2nE4OjWfPlx0fJw45d3KDy9bF252Fl68uJrjAU/jBcQFha5xOG0/FeLH0+ZWf/GESY/nQaZs3bjDZ52tMSYy5O3YRYcu/uPnKaRhw+BtTmo82u/INH06+zYeeD6HjR5xwn/3sZ68bH+992EPsnDRdiD0C8MiCnvAzv/YpPW7iacPmryb0Chu+9ObBZvIRj3SXaT72I1vxtPk7Gdh77I3Btf5w2dg3jkVf7OcTjh65elbPz7i199F6/j//+c/3nnjiidPHmZO39YAtP7hoK80L19aZHy4KjuwIvyQdfzaJvvyQ8uoVH5FPPMwpiQMf8eKjTfRnsX/Y0vGFKQfs7f/WRmw2culGwnEVT7nykV3+XvnDJLg1dzitJz18vIvL13hc2Bgj4mSPh/MNO7ppxx+Ok742/o5z3zHou9r6d4iw3DC4EOKRb2vmI9N08uEChAM89vRkco2/Gr6xOKaz3qcf4wVggABsIpwYpkMOIJ2NrUgAoXeX861vfWt9064DtWAISyAcC8JWv6SK10TY8tNn12IXIyz9/LKlM85HHHxLojo7Y3zZkl0+rtZOzubGFh8F3xIJC4ZcKMb045RODLZJbfba5VjfAit8LS4d3gSPYuvj5re66fxzmROGAwKm+eArdiej6U+vsJu5gBUv/mKbDz079vD5GtemJzhr05Urem28nNyciIkTZ/b6fEhzDRcXBS4ctRJHPvniyaY5GcPTuBw0hosTAw7FOd4er331q19d73uwxc1YJyDtTvrFdLIl8AheJG54EBzoSGNxaW1gGJMf9oofs/L/CF5Z2AvloOOtEwWu3VzYt73HZM3MsZ+Q5sfHnTUsuVbE0m/OuOEyubKTAzZwteXbHJsnPzp4arbhwiL8jIcth9q40cPQ5oerQqfwVepPHjD0xbRm8qmNG3t1nJsvO/9T4z3ccsTGRdajYq/49cuzGM3X/GZfu3nNXPEn2WqbH7054mZMCZNe0Wdn35UL+4DeR3ddWMwZhlegvlRVHQ5MXNT2lQIXJhv5KV+40zcH7XImNsmnnC4cAwjkrK9NJxiCQPUlXN8dgxOXOzgHInsk2btC+soHbWNI8wmXrUIHzwQRykYcot9kYCsONrYmoib8xTKeb2NqsdTGtdmy83jCPI6OjtYdis9+O8FNH8nl5yAswfGnx6OYxmdc+rDM03gLiDccws549nD5mb9abGO40BlXCIw2ljyw0+fHXswWXZ/99GXPhtRWy5ECRz+f8g6L0BsPN199hT/BzRc/fvCDH1zfZCvv5mVcCaccwucT93jAlAPcjCkzdnHjDo+OxM0BaB8pJB+PTTzzpvcqSR6tV/OHhVN55CcO7uUjLHXxjHdyZB9/4/jri6VN581OHyVVu1P0+Gl/f39x/W9/xCX9r9W0d0dtbu0Xe07Bj8StNaEz1lzMv7zSK+nMX785qc2LLzwnQL5Enr2X5x8dtWHw7RjBj06Bwy8b/PXls/2uz04cwpbAbH80ZzaNi+d9Ad8+4Duv9AkbjwT97xps+zQ9X4XenLWLK5ZiTElww4VOrZSvuIRRvtTF4t+rQhxdGPUdQ3IsJvHo9Wg7l6mdd+PZ3vIqjG/7TU3KuzhyKjZ+cHHVVxM13PrLroUwUIJLUIkAZpx42eNrgj/72c/uXXXVVYssO+PskHR19LiigPwQQZqOnVgIiM/XeMWYkr86nYkq+kSCtEtIXPXp4WenFkvSFCcTn7bh4yW/C0gJE4Otws5dHbvJBbYYfGxqYk5i0xnj3+YRk96BoOYPTyknfGDQwVSXX5uZn77aCc6nRD7xiU+sOyn4xTZuDnC14cDO1xg9fuo4amfPhp4frrDgw1KvDbTp6BV80+PCn/D1SNDe8IakvSGfxYYvJoEpJv84F4cdXfzENCamtnhsCGxiDJZxJzJ39b7o00WCwMLHe3Z0Pm5sfcTJzwEbLjxxFJj05bT17ETHplzlX0xxcVT4W0uveH0VkFcd3udiA1NM/t7XwE/ccPLPlj6Oy2j74071/PY1Qp4M+FoW84Qbd/Z4ljN6fWIMtrk0rq/gTQ+L0PHDTw6I9xJdENvLbjAPDw/XEwr55VseYcHoIqJfXLEIezbWvXH9YrPhz15MNvqdJ8zBnFycfd2Sbx/QVyYPr+TMgT3sOXeYxTNf7XKm3Vj82OBTvsSiU7RJa8HXHBUx2cCRqwodO3p72qNXxXzxNgYvXHNny4+YpzabbNVhhp8/O1yaj5qoz5iUxWgCauTpiWDERAXI1qZsMY0bYyPpSnYmDaukRYSenbEIa09y2nCNx2eR3jjR4xpv+hYZHz4tAJv8iqXG30nZBcJJ4kLCrrkbh8MPJvzmyc6c1HO+FpUNvuWSr75iceHAnLG0CV9tOOzEJ3wJLMLGCdqGmhtfm7DnjwMd+2KYn2KcDqZarOLo16ZX6BSY9bVxVhJ4xm1uMXAUj606iSMbbTyN8y0/teOqhi+eMSXu6sbaP07Q73znO9f/KxnDxfsHPvoqnhyWM7zEZycOvtpw42P9ihd/+4AffRjx4s9OjbM7Q1guIH5awHcw+Xhx3NkqePpvaFy9ehaXn33BN2FrrnHExeMtb577RKSLB5/i41UO+fGPK0xjdK2THBSjGo8wwmXP1sXQh1GcsInx8lVO6edaw5Ubdvjjw499fNnQNdfJk59iLD7afPkl5V5um1fjaj5yBQtOe6HjnI1iLfBl63zCT98YP/7xqy9nRD97/eJpwzXOFj+YjiE3GuKYj1cc3/zmN9dPDJgHWz6k+U4ecW5Oy/DkD3t6Yq7FjTMOxq2VHDhWziDFwAQNClBgY3TGCD0nfSfd6esuzss+z+GcjAVrk8PQ5wdDwhCkjxz82mp2kolPC0RHxIXFbvI1NhdDAsQgfLXZs4Gpvb+/v37V77HHHlt4/rPTGHvjijnD0o4bTDgVi6ddrow3Pz6K/hwXQx9um5NNseDFw0lCvzyyI+HJyXyFYoHhwG2u7PnBEBN2AicubIxP23KtJjCM6+MlThj0FbawyORRjvkVi0/zoYPd/NuLkwcb/uVxd4xeab44nN0+qum5twMRB8VvnPdNCfTdVIgZPl9Y9iSf1oae4EJH2JlLNuWqMTV7cy1/ah/W+Pvf/74+XuwxhD3129/+du1PPr/+9a/Xe4zWlsx5lW+xHeAdI/h2sPPpxCN+xbi2PcZPH05zj6c+Ect4/ubJhq/SOjpuvNpT5Izw4WtvKmzp+IdZ3I6D4vKHT8ydHq42/3D0cWIL0xhd+8MYP3nSlhOCS7kS22MiOdHmazzOsPnDJ2p8sqMTl6SzFuIpbOEl4ZgDPR+FbdjieQPdTYZX8MR52I2BPezJT5gzT+WcfTlhRw9fnOYVv7jzwRtfpVyq2Z7RQDCiHIC3QAWhk+gmpc/POPGM1dc++BSLg9RdEkxB2bAXS5vOAvJHnNA3vhTbH7EkrYViy44vWwWuQs++ApsYS/AxDgcGHjaM/5Z3sPqPeicXduHCkcDJzbg+Md4YDHlT+NPP+fMj9LXZabMrH2rzoZ/82dgA5pC0UVy0849v/bgUR98YkYdyH6ZxMdizq+jHiQ8xl3JuLD8njw7McOHAVki4zX3OmY2+XBLtihj05d14uNVstK21NSZyZY0cdC4U4ht3svaeA74wFXbxg6NtfuyNw4oPbDp25kJ2efBnT89WX01XjElJqgAAQABJREFU3n7zm9+sj2l7TOxNXTdlftfBTz93clDzxYMfHuG078LEw102G/ngh18c42wPJDAUGHwIP7r8jZFsm7daDtnxDUN8OkIfLns24eNlTuWHvT6bdLDwLWdsCMxyUZ9fvrg2d8eHvHShhE1cLHxbgrt6ePTi4AeHVMdLTFLOxOBnbng2NzbG2Fm75lEO4xqW+HRsYTkHwjs4OFi/7smOiOVY66Pd5ZkvDDyJOCQ+MPEhYsgFezH4suMLP77yA0fujOO0vo1Xx4SAMtAHxlGdBKrPzliFb4+1IkunIFLCi8WvRBejhGWrNhZePjaAdhONu1jwq41Lkjpbvsb1xRdD0pww3HE4EcfZgdyiwyXiGq9dvuIMl8Bmq9RXy2k2xYZnE8AyFwtFxIRD+MSLH16K8Z5pN1d25gRLPAtu3uZDb7OwMV6+wqKPHz7a5Tq9Ggfj7El4avby6U1SMdjDD6d54QGjnGs3NzYd4DiTuMJR6vMRQ59oN05nHIZcZWMcJzxnHlxczIEPe+348+VDpxRLnQ5uXMynOHQELp0xom4/q4+3T4O5aDgRi+V9Dx8xlQsHLBztBCd4s+Ct30Wcj7hs5QGuuOXXfqM3h4nXHNnTE37wjIkBtz3Nxl7Tt6YdU/LbHPnhB4e9YhxOY3CJOMb1tYsbL2NikdawuauNN2YO/MVxnpJf/Kw3Yev9ufvuu2+9CpQPnPmVO7WCDw7ylw5G68KnPNOJW3xx+M5STuEqMOlqNw9zDc/8CP6Oa7atw8wPvdgKnPz1SZjG4gg7jGW0/eFnn5hLe2XxnBMBJqCNSs+AaNNzVIgJKoCJOn99vnwEREjRn5hI67OpPfmwh0maMNsOriZMl56/zVSsYotTmx+7NogNxd7n5D1CiI/PVXvTUjw6giffDmYHTG1jCim+vjyRMHBR2BA2xvBqnuHkwxZf/9Tko9Pyou9NYR+XxsG8rQ/buTZi2Wiw5QB/upkDPpNPucKPPs58xI5zuU5vHxjzitTjTONvfOMbF+f2yLyDMT/4/LXFTafmrybxh1+utBUCJ958ygW+7GFpO7GK5z92vbF8dPTvT+LxwZHtzAVbYlweCDv68rKU2x95ZycW2eXHfvJ1UMKhm7nrJoLO/mxe9LjBL2/FYQOr3HQcq8U1d/zY6bNLlw99eDjFTTubuBrDoznp22fmL4bjxi9/svFRWe+DdIEpR+JrK+W03DVnfJprc+TXXI3xDSd7NcGbrePF/Nnbn37fxUd15ZSNN/h9Okx8+HQJbKKOl3mHayyO2vQwCDttebMnxYNBryZ4zbiwjNGx07dvvXJmC+9o27eOMY8+5ZYN+9ZNTuhaN3j40xEY+nzwpWfDXwySL538saNjx3/9oFQECwBQuz5gpQ3p5b7/MgfWnbMfW/EjMT5CKDhfgqDgCoFBV0KbjDF+uJDISngY6vybiJoffG1xTRAuvTj6CXw48cuOzV//+tf1Az5iSr5Ssuj4wMsHDj1MsRJ6Or54NS86vsYJvomcG2Pf3PlnKza9uyb/HWsTGnPXROLAHyc1/nDZEtz5GC9vYhSXDg5hx7fcslHYqIm2wlaxF3A0Luaf//znhSGeN/58Nxr+TizmI5YYfPngbIzwoSsWnfHmVW7ghAXHOMGLnl192J4V//SnP10HIn7f/e53VyyPsfjnx4c/Kbfq2uyKxSZffGvLHTu6cBtTK8bYFQumfvu2k0Br1riYxdJmzy9ca6HdGmYvDi7ZlR/+9OZnjIiliNM66RM6FwP2Xm3AhYEnW318/COuO2Tia2NctH2tjRiwqpfB9ofOOsEv38b0Ezybw9wHxvmUS21+eLAn8cfRty0r+NOz5ds5Dbb5qIup3frAMF4cY+Wh44CufKrhiKVuvWAUP+zOtda/tTXW/3kYt4/Pb5+uE8tNhico5Q0eXmJWm79+a8PWcRqnuLNvznxa+8kZPvv/2f5b/F9eMgNGRC2hagvZpgAKgJ6zK56rIRKKgDYKO+OSQi+ImqiNq2Els81XTEIvUYQ+f0ltjK128cxBH8/dmo49TqRE6Vscc1DEmr7h07NlEwcLoM9fEQOuQuhwsggKCUcbHh9j4oRjrvTGZ5uPdbFh2Lvj80pJm559uRAHP1xw0G8zaMdXHH1x1PGHJV/65mg8m3gZ4y8+20544sD3Zh9eBBcXELZ8iqNdnnC35vrsiht+PPnS4aOdhM1f4Y8LXr1BDgN/B5zcyQ1bsYmxmROcjZcbtX7zbh7lp7xlYz7G9OMbHiyFhMsGbyKn+riVM3PRDottMWHQ65eXLo72ilwY548DKZf65ZwOBjz8xdfu+Knd3HGKq3WGz799p1aM+ToefmxIc2FPquPZePZikzBwpsOBL65ygk+2rS1bNi4SbBRxcIO/u3/kS2EPQw7lqbkWC6ZY/BUCL876bPjhLSb7jgfxCT78jbMXq3k5v7poewLBxjcQ2MN0+nLLnzT35hc3sWE6Js3LfPgaV8pDWNPPGE7sibmfKZCOEpi2ACXGhAQmkiKAC41JdcD3vxKSwr/gfOZEYJqIuoRpK+LQKWzwUfRbuCYAP3+6FowdX2OEnb7x2hNbzMYtluTylVz2cZKrYmQ/ebNjnxgj8mEMR9zU/AmbMJuX2HLPJwy6uMm39xfEsqguJmLQxxF2/jBgW5N48xWXvjnyMR43fT4KW9gKgZ1dWHQEZrFxstHxw9ObfebSXely2P7gKA57uPzhFsM4P9h0xui0FbkpLkx882VL5NS+ta7wrbM9yw6/cOCyVYpTrozR6cPAiZTH8gOLP/205VvBmZg3oSdi8OPPl4gFU2l8Nz4u8TQWf3Ombx2NhQu7dnMRgw0+alxw1ddW6OGps1HzhSMP9jobOnsA77iLS2+cDnb84w2DvjlpE+PTBh9Yuzp+zQ0XbTZs8xHb3OQIvpOxPejErN9c1faLOgx8YWUXT3V50W68/KibKzwiZhxxYoOTNtv2Cjv/h2e8vNm7u3H4sKEPTxx9OTdePxt9Y62Z/McvDHMhsM194/yfO2dtySHaBJg2MIEibuwEYCW9ACZNHJhsXWSQUNhk1yI6AdZmg6AkJXRim3DjcYSFn5qNmg08NvhadKJtrMSyFUutNM7GHNjR42cO2Rkn8PmXF/76xi02DtPO2LQxLj/x7wBl15i2uPraFbg9ooBZsU7i67dB8oFhjI02G33j+opYOBljFz9982SrnS8bbQKLjZqfvKmtJZ0D0xyLGX7+8sneuIJP/XCbk3jabOD0aoy9OfA3nj09W33jnSSyNY6HvQKPbbz1ywtcbVzZyEc6/fDEoeebaItD+LFV+KUz3vFWXmCxMcaPGJNfHOEqdGxxm3nmY/+q5ZHAw4/wae+J3RhuMBW+Yqgbp2++/ODQGSdsCRttY2KaR3kwbiybuMDQJuKSYoWPc5xgEz7lz1zpuzAUh5348uQcRQ8TPl329lTnMljGwogTrOZlrLnB4oOLcUWMYrU36PPrnBcPtTUm7UU3X2LDVcTTh10s2K3tct7+GE+vzU9cvvI4dcbkUXxj+s2dT9j0crQ+hUXJgWgzVACbJMBsSoiaT2T4moRgiKkJEoR9dVjsxDHWJLUln75NVCx1XIyLz5bA1MYHD+NOCPr88FCLM2Pmzyc+ruji6BtXG9/Nh5h05lr+aotBxC1PMMtXOVDjRU/0ic0jNhx+CsEjDH32+sr0NXd32viR+GiXAzp+9fEg9NrwzL14tZtjfePmyN4YruJrG2sPlCt+Np/xDlhzdaDQiQ+PPVv9yQVHfXp+uIrZPPjwV7Mrd9r144YfOxzVxFi2xstFOnZiF4+eDdva4tMpbEn2xtiKQ7LTxpWejYLX1IVFB4PQsW1MjYe60j7kh4cS5wVygsPeGLuJwTaf+BqH2wWZ3r5VJ2xInJpLe5t/Uv751GZP+Dff8h9WueInNju65hmWPmEHr9w6ptir7Ud71+NOOn14iv3ZPIoJ05i4xsLXL068jONiLHu+4rLRJuyUckRvnLB1Q0CHQ3r2+nDV+vYRfmxal/rTHhY9bnzdnIYLr/j4sJN3OhjrHwkZUQYKiCEjY0R/OWxjbBULQO+EmwhM10T04SgkQsggS68mxYTbBMIq8eyaiGTipJTsFkPCFBsA12KJDytOxZRsOrH5ic+PvsRq8xdL0Sd8+IqtZk+04Sj8xMIjPxhs6NmYRzV/8cn0Z0PE4udEDEffXHGxAQhspTh0/MtlfOnwoheznPCLrxpuY2Gx4RNW/OnKpQNPHz578cpl82nMwWE/wbEOMOSOOLDxiEM5Ncdw6PT50fXyni8/XNiEg4c484KHC125wEUhMPjDb176fIi20hrnq8YnHDaw8tGGofBlx4YPbtnjxUa+ibmytTad7IpjDK6c89GOsz47eShWOtitj3FCx7710IdnvNh09iN9ucM7jmwV42Ri6+PBVhyc9WHirG/u2dDBwZmOiFVdLsTgR2DzU+OsNk+1QsyPuPkyBwUfa8AGH3X7GId4iMUfH364kXjNdeADB3Z5NZ6ePz949AR38c2ND2ETR33+hD4bNS7GcMSPjh+dNr2+0hyMsY2XYwlnOsWxui4gOgAMNllA2i1Q4K7MApiE54QkEu5E+LTgcI3RlSwLR18yYEWqycGkI3zZ1zZZmOITMXFkH0d6bTbFV7MTT1uBq8w5ilfC4IinFINutvX5NMfyqFYseJLdnA9dHNixT4eHsTatOdrwYXdAsfePmx0c5k2Hp0LKMb1iTmoihnGFXr8x4w4eMeVUfDnUJ/Iorr615Q+HPxyFjrBhP8enjg178+YjrvU1f3tNfCco2HRsm6N2+mLQKfp4s8W9uYkjhhon+I1p03d3bRxXc82Obfa4iyMeKbZxcc1HyX72a4tnXBzx6MUiMGDSm4O+95H40MuH+PrGzZevsTCMh5VffNSw2RQbj06+/Ojn/tMXlxQbZ/uPHd/yrc8eH7j56M+50RuHFzdc4dI1H3Oi02cHX82Xzhjhq5QfOjZs6dTGcUtgGYOT0JUbuuzjMGPCnGKsmHD4lJtulozHXez4FwcmX0Vbacwexa3j3nwci/mww4GeHeEjhv3ceaP47Rd9NnCq4VgvAmu9iV6gDBlwUIAgTYy3KTnrS4S6yYQhEElfYBuKuHqxbRGLxY6PEg/24RaLPdtqfBR+9I0Zlzh+RJ+NedGZmzj69GGow04Pk62aaE8MJyNjc/PBKJYabvG05RNGfItBJy7ucmyR4wRTHsXrYNVXxKarH584LOLbH/gVeES/es4dD33xCSy4+gqOdOxg1Jbb7MzBpmbfhoVlvPUxxt+8ig8jvbb56c9NLp/F5ic+YSeucTmhhw2DnbjG8IEdd2OtC5uwGlfD5ht38RP+CjuF1GaXLX/zN1bbWJyMwZnzN87eXNT6uDc/8y2eml1z0Y6XMW2+5RIWG3GzFZvQkeKLzT+OLuzG5JbkryblNx+6sMwXnlIc4wldc5BvmOHjZ4xNutr6joPyI17HKB2urS+f8oljEi9xFHbG4wA/XzzYmA9c8RR6NX06+Nrl3/rpw2UHR79Y9ii+xswBL75iN4fmqzZertX6BF98+Norcc9nGZ38iW9x4wYLxurPjjbSyAqq6NMjqTaWj6uYN3YE6Lmc2BHmS9gjbuLhz4mzhw+bXl8CkdRukmoCA/kSJxn6+WqLSdjAgIUn3ApObFssCWXrzo6uBGqXRJhslBZfXJzYa8Plg0M6fWI83VJsf/gq/OSUX/nUVvg5SNk0Fy+zG2OPPxvj4vFRi8cvvuWnsdYWFuFfPvFiZww+LOM2fDrY8tPdlLhKeWef3+QgltgKPPbsxCNquIQNLmFX07FpzB7DvQNu2mVTbUy7fIkvDn9tY3jBZ0voxSD82ITHT5EX8yT65tGc4BjPx3zp2NNV89Vnu4vhn10Trzw9FdjlDtNc1HBaO3ORm6QY4sTLerZeHZP68My5ecddvzH88cUHnvwl8qbYJ/YryRYPRZ/AJjMfdOZhTmLKqRjNlS99HOSFvXjswi7W9Idh3NzLGR825ciciD472PLJxlj4+bOFgXf7Ej85gSG3an784RiPi7689slFH3+OCx9t9uZHwslG7PBhyQve7NoHjZsLrObBF4984DcvY2zJehPdBAugBsKYMG5ywCTCSc7kgfSRSKQUNkjB5NdkYGkbR954PtqNSZiYFRhEHze8quMJx7gxJ1B9MRQcjZE2nwVMLzY/i4ADX1LcsOuzYQ8Tvr5FiT+9ObjbFmPaG4NHpxBYOJCJqV0pVziI6QLnv08Tnwv3z52kk4PYM0+wEmsgLjx8tMuHfi9/+cRXW2y+RL888MVR7sXUlwNz7AREz4cOpkJwoIM1Y+nzUYh+AodfeQk3Hxw7SOlw8VX99Er7AE/CRmyY2oo9Xq6LZ53582MPJ47NgW2YxsI1nr4Y2aqV9o0YuKgdX/75LvHrk1/4whdO/xcAlt8x8W3CbH2tjf0w88qGiFFs64KfOckjfvpqcfNR08UfDh0fRVs+YIvpk00dF/IDS78Y4tKJxX7K1MHTh63mz15f3Vgc9HExpm7t6HE3z3w6XunZWuu5X2YsPjNv8Y1f8xPX3GFbB/Gtfb7NQ8zmII4bAH3xm5++whe+i2265mOv4K6m4wuPHQ6OYec0/njQEX12zuO46BcHDkx9Y2q+dMaIml5cNqefwoow44JpK8Y6ERhDoKD02iYQeEHpBdHPR2IVuIvACRZb2PREu6TxJ8aKRaetZhuPJi4x+ExO2ngobTB+xP9VFNu4f74qyeIo/Nk0V3XcjItdjHiEKQYd+ziJY1xpPjZSfbV52EBiiYH3C1/4wr13v/vdC0+OvD/gC/fUYvCzgZzQ9fnwDReX+NLVZisO3dwP2jaMMQcHPnzwpzMf/LWJtpjmYszJm8CVU3OC6cCla37L6OQPLk5A4igJXFLuyjcsbesJH4e5fs0r7uLXhm8P8fGPhbDdvToAXaz7KnK48lBuxONrDs0d7/S14w6fLWHfHODlb0wMuE4ur3zlK9fX1+CBoy/79DsW1pYUX47h+c9qX8cBT8GFiEHEh8+WDsfmboy9Qt+aywNbPmHAZj/ngIuSyBeBAzN8a5g/e7HU8kM/ddr80sPUDlcbN3q28SnXYsEnavbZymH47Jt38eGKjb82PaGDBZuesIFBD59+cmaLm9K+Fo9dhd4+5t95CZ49yZZv+5yOnVcn5oHTjKnNB2ex1fGvzSc/9tpxFpeoa/ObfXNeFxCBSEnXBwSQE3DGDkh9m9lk28TGTVhNzxdpJxA6G9TkCQyb3ZiNyVYiSg4dkvpq/niV9E6wMHEhajZiOinwLXnGS5QxeuPiOskqsCyEeLDMlb45Fcc40efPDvfyVxz4uLBnZ1zR5ksfB3j88GfDj+DEnp7gIu/Pe97z9t7xjnes30GHo/jWVl8j4yLilQhfG03u5EssOlzLNcxiNp+4ihVHNtqto7WTm7V5TuZpM/u9CXe/xvl7uW2t+YtfLmGVC/HwCZseF0XbmhNt/OG0/nzDoo8nnXzSsYGtFAceXr5TDC7e4pVv34eUmKf58G096fjbpy6mYvGFpcRfLT5ehB0dfqT84RZn/uKw0/bVH34t77Wvfe3Sw5ATFxbYid8R8Rsi5uzr390MHR8fL4z44ViscsO/tTZOcCm+NlsljvBwa8z8cElnfWCyY0M6wcGFNfOgrZQn9rAUuuzFgJmvtpJ9emulbT7tn9a+9ZBj+MblxDrysSbGOmboCP/mwo9eDY9/PNmZuxpv49oKG344d27pXAfbmFrBnS8efNM3V/5y6oZRHMIW9/JoTjAIjPJJBw9PvuUwW3bp4rRAtj/0zSNO6zfROSWAJIUzYUgKpg9IjXCB2cNB1rj/PHVim+Lk99a3vnXv+c9//nrJ7buRfNOoDQ8LpthwxEMWFmy4pARpi20sHsb4z4k62NsY4vjOLvKRj3xkfcmf7+7iI7ZFKEF+6e/cuXPrh+vZixMX/OJBz1ffwrfg9PiV2/JULJiEb7ZqJ15f4JjcdNNNe9dcc836Om8XEDj7+/urn407ZL+seMcddywVzN/97nd7d95558JysnRHjT8+xWyudMZgK3HUbp6+fO7pp5/eu+2229ZvErhA4UvC0y4GX21fj/25z31uPXLzfVh++e+iiy5avwBpf9jIDmTrZM1xiaP1gB0WPGsZp3jLeZyzh4WfvaE27kL3rne9a+/Nb37zuti5oYDV3PGHVTxjfPzevN8LoZ942cHGRRzx4cU1HmziCDdO8LIVP725w/dK2COpKWyImq+TSTdzHnE5wXik9fjjj6/jCwd7Ezf5zs9c40oXrhwr7WWx2DWujxu8eBqTA3HoOs7YuhDC40NaX5h0rePEl6PyRK9N4suvmOr2Rflkj58xPnJg7mIpzj3/XyLWxRdfvDiW0/KDQ3PHo72izU+frTauiXG5dMyZ0xQ3FB/96Ef3/Ay37+/zg1LOV7viV1YdVwQPMYpZXuiJeNXGyh0dH+Otkz7OyvpFQkYGDZAmD4he8hWTJQ54ehulibJ18nPhsOGvvfbadbccYYl1N+duybfHukL+6le/WgdmuE2quHzFFYuNfhzr42OyJQB3/iWmE507Mr8v7be5Pat0EcOlu5Rw1LD54fztb397bTw+JbE54RVXPC24+BY8rjDYzJzGMZ5w3E040X/4wx9emxGGOdgkfg7WCSIsseLL1wb1s6EJPycexUnSm65ynW38zdOc8KGD05xwMwbLid5mtGn9jPGll1665sqGVBc/X30XaBju9OXHvNTuou0fv4Hx+9//fs3NvOLChy1O9PpOACReuOItHlt6fRjNg4+vfvCjRnJ79dVXrz3ggmr+bBP+lXQu2opfAvR7N7iIV2z+1oWITcqHGm8ycePG17g58qUnYjSvsMRgTw+LVK/O9oetV9Ive9nLVj7k+Pz2ZXtywC8u8eUvNpEL/mEWn57QN59s4MDEl84+8e3QfibWozb2ji/fePujH/1ofYmic4Njq7hslPJZDPHhih9fPLKTC+1yFkd7Kzy+eDl25cDNjL4T8M0337xuqhx3sPAxFz4wyhX8xrXts7ixgef3hLz6d04Ti50xmArMcPBkJ4aTu2OAsIeNJ1s2zlEeVbOVi+zl8FWvetX6glI3j/h5pYoLX7bOw44rvNxQ+B0ZXMolTPbNs3WIiz5hQ+Aq5l6uxF2vQBibJEUScDWgFsamQACgl2GS4ATnZTTdJZdcsr6+2Uktf7iCO3ARODg4WAemq6cfcjJJeklsY+rzwS3iOIRJr28Md/3iZNdFzkn4uuuu27v88svXIrP1wzH4Szh/uHg4CBwA7lZxefTRR1eMklniYWhXcC+H4iv6xuOHqxK/bMR1gfPzn35HWlx+9N1dyseUsNlNgQnDIyXr0Tei+uZeJ0PxSfm6EGc6ONbaHeTb3/72dWJwp2XtSXHFkLtdfmzk0i+oNR+Y/OScj1cl+p7bO5gbU7OFiQtbNc7ybIzAIexnzcZeswdf8YpXrHx4Jdfvc7O9EF8HnsdY8Bx0LsJXXnnl6fp7BdX7ZcVsLovA9gdXY+UZ1+adzlyUDmTzYiOX8OjZsiHFMk6nz6dvOHZzJpfm5CSGs9rekVs3d9YgLDGsrZiOgeJO3nT6bNVszS0MfZj4uEC7cOHsFZ4bNePGvGdHPGbF2atZMQmbsPW1FXGMlYPyk40aNm7G4sWe37STEydaxwR75wI/ayzH7UtY/OKzALY/9ePY2mYvnguIXB4dHS08Ngob+uajJvJqf5J0Lvxu0vBzswXX+cBxFwYdX/OxrubtYuI9sfDwZWe9HXt8XZQ81XCBh0+Hm9hsCT+c829cX0yFLx+65r8uIJQKI6LdSUKC0xeAziQQcYI+3u7uX//61++9973vXRcRJz2PKdwBEf5dafXhW1ClPp2JIgdbcvTFJAiXwCYHV4mrtkkaV/RdJLzSuPHGG9cveeFE/+STT67fmJZoiyE+ju6uvXryXsJzn/vcvQ996EPrjukb3/jGqs2ZfxsEJzHp8DVO2vC4lTcxzI9oFxOHT33qU+v3o/2kbvNZhtsf/rjBFy9f4+Kaq5NFi63GyeNB4kT6yU9+cs8c/Cwqe+PEWhIbUBx6MWB49cL2hhtuWK889vf319icj3G2U6dP1MZb56Xc+eNNYndWfsnS77GYoxN3HOAq5iwv6vqgiq8tXmsj/x6ZunjJaXvS+ri7a24z12Lj4M4NrsdBftPbyfk1r3nNmsf999+/9k6x1XESW3tKnOQZpvjZsyu+2NmaX/Z8SOvFJn8nZF9H71NaLuwukL0XYj+8/OUvP43n56bjBz8ctQLfcactZjZzXJteYWONFBeP97///XuXXXbZ2veOHSfERP5ciMV3E/PAAw+sVyXizXzFD6YYchAv60naF/XxiHf8slPTya2fgnUylm+lPcYmYTv5pP9vtfOHV1Xm7IYYhhjmI7a+cZwVY7WL6SbNDau96gaFjRtAT2uymTxgK/LTeXaOO+b8Do9XW84v9vXtt9++9gfsjnvc5BRWNW7tN/ps4BuzNtn++yyyDViINg4HAgT5yNL1TM1V0sZ5z3ves062NomXXCZM+BF3Gp7TKYg7qN0FGrfwRBspJzGx6XERlyDbhtLGS79xbQulFkMNR3HltRAuDBLtDtzFwyOJe++9d+Hb3O5+JedjH/vYOsEYEMddi4OzC01ccMZRDHV9sRVCFw/YzQlv2MbM2YYRQyz881sg2x+c//CHP6yLoXE4CjFvuXeys4Hj0iaAZfM6sbhbiacaT3USrs1F70QL3zq7OzJ3uIQt/oRe0eenbg3KBz2fchNPJ7qXvOQly8dmt688KvKqyV6TKz7s8RIffvHhsqFX94rSDxi5qfEI0KNBwvZ4u9nxmwruguXShSVxkfaoRSE9juEnt3KoFodf3GYe4mlMm+BmDtZcnuJsTJtMDDzMg0/jy+jEDrZPW/k9CPvYseUEJI4Lnjt+eNb97Pbz0jjD64QNE7Z84igPRBu2PYknvtrs2JN84Bt3YnbD5ZW9k/KFxEXTiZGIbV3tLfu2fdO5QHxSnsQtlng4yg87en2+cNsD9OUOjrabKeencJsTfzoCg++FpDF1og3bfOTZuc++csNmzsblHb4c44wj0Wfj+FKsVT+pjdvMBx88zZ/EgY6Us/iUL69ICTzx2HuUjaO2tZ2CZ7kzLzaw6bWLayy7MwYyDgBhk+LMKWN2gXlzx8HpWZzHPXAUZLOxSfxozyOPPLL3pS99aW1yV0EXkCbZxPT5iSmOBEpYCYdrrDgmPifE3rhiYWA1boFhuet0cvrBD36wd3h4uB61ias4yZizN02dvFxsXFjEVWwEdnNhxSJixR8/sfTFx1+fjXkRem21MRcP/9fhwuZuphyyheORijtN78fYFLi4qNh0TrhXXHHFyosTsVd++fEl5uTOz6sxnHGqLp9slTgux+2P9Wk+dNqJ2B7pwHLX6SJTTLhEv3WYfbpszMOrRPsC1+PtJG/fyDUbuVPb8GLxlSPj5VVNb9yF0t23vYn/P/7xj/U+iHEn21tuuWXFkUvxzFneYVh3ebSn/S65uM1JPFztJzocxLPfjNHB4qM2lsBunI4NPh3EbNngS0/cWHhu725SLHp+1u/73//+3j333LNujByHxvy6oscUngR4jGwu8MxFLrwXBhOGWISfthrn9gYbfK03bvRxbsy4vJm/vWXd4NAXwzHlYmuPO7G6mLnouAj0SEuMyQdGMcSFZx50+qR9mB1dfuU+LuouytYPnvUi1t8rOHvBzYwc4cyGwOcvrvOBC7X1T/B2c8nPj5WZH670rT8+1hkW0Ra/3IrnmMaRsK/gIb9T4jZ1tWGLnQ0O4trLzsdunnD13iSe5iV+kn0YcMxfIXy06ZUzDE2mpOubmA2rMKJLBATialnSbYykYMh69uqxiTdK3c1ZPJiEnckq4p0S2tpTZiIlwgRLUrz4WnzcLIL5GNNXi+lTIS4MHkE4idjQxkoGbBvDnR2B6Q4WN5uCfVzo2oByUb74wVTwhN2CwMenedLDg2ODeMTgkz70LnT0cM1V30XEiQC2i4e4+LpgePV0uF0Qzd3LVnrx4Ksd2D6VA5+IQWDhI5ZansqHMevlzs1BJj6ezYctPC/ZnZTchXqDnY24/N1hzg0npjhEnFk7ySj8XADMwTytK34w5UPBQYHVXMxTLHsRjlzIC7/z2xvJLsC4OdjhuVtUOznDgGu+8u1O2puXHsngQ+hdMMWBY39bT0LHTix5UcdNDVvdftBmIy5fRZtOGy9FPK+GfKLOuANdXox5Y9pFGy8ncXMnP/zhD9d6fPrTn175twaekYvtzvOZZ55Z/MWJD7844UDipm3M/JT8ms9TTz21cPUV8+crl3iK51yBj7XxNMD7fObiwu4JQbj8YBDcypHx1t18STyag3riZAMDJ/5yxC+xho73hx9+eO8Nb3jDOp85X7k5Zj/j4usRIf7WAG5zdvF0k+yTUOZo7sZcFOCICYuPWmmf4D054SZvYbN1fOHKTjEmVwkMxfzcNODK3nGAKwyCG1/7wPmYPX548bef8aJnF18xjbGJvzab0/9EzyAjg9pACBAHDp3i0yxOsA52QUkBHaTuph977LH1MruPkQoeLnsxTVIyYBpjo9+k1PQVdjN5eEkCKSHGcYLl8Yuf0LTAEuf5dico2JINE46+k5a7Xy/JicX0ptutt966NppXCvLQBuMnTjnQJ3AJTGNwlObWXMU27o04PxNMLLoDBQY7fRcxJwr4NqiYeNj4XoGwtyngEX5EfAepu6PuPo3BUfDhk5+ajzH44jzxxBPrRKCNU/P1qsarORh08muTWlcnLq8qXPT0SZzE0J4Fhr7aCU98v1mNe/zwkvfy2J4VmzTuwPURVhc+mPi7Y4fpMaG7L3EIHXy14sIjzx65zPdurM/R9iapN62tV7EmH5j06vDUOCTGZ47NBX92rTnMjjWv8Nw1upA2zsd+kON0/Pnh77GWj6nLD52bADd71sbNiP2EQ/nDDwe1Ur7jTAe/PJkDoXcRcEdL4OKNk7jwnVz1y7dj1IVPYW8v8yFwxamt1qeHocaz9YavpJeX+IuvLR/xYmftmrt94TfFfUjGnnHjYI2bzyJy8ufzn//8evyOf1Jc+8w5w8XUo3L82gPs2cWTb+O4mJ9+uPzY0uPvfBQn87A38beOxecjPy5ubuLcLDv/en/RK9GEHf8e6RYnjnEph/zSaYunNI7H6W+iA29AmyNJF1CJsBm9pEPIJ1ecWF31kNF38Ls79VjCwrobNMn84bZxxJNEMbSR1FbaCMhGnl5/+tkoRPziiKscb3fgf/zjH9cBJHn5L4ftz4xpQ9tI5hcHrz7guJPto4D4KvjxNx+ijRsx1lzo4Olrs+fPnsjFvCugV9iwlTcbjpif3F+5fdLGM28bBW4Svr6DwSsvF2pr5UAh2bOFF0/xtMWTSzE9N3URdqBM8erHBoUhhjVP5Mwbq+J1MTUP2OblYHAydzGfIv/u/K2ng9LBgANe4vCf7dYozmzYe1R5uL0q03ey8pjHnHCIh7j5w7R/vUfn/STrATN7r1rvu+++NRd7gZ+x8iVma9l8rH9rbgwX9vkan3sELrt4srVunWzxEUcRG448Wj+vVnB2UXGBdGftwuGRi5zCFJ+9mPzpiLjxoJ/z5sNuzk3c5iCmePHRj6OTnI/0elVs75UDviQeYoTPvxywYYuPORJ7El86duLCkSvSOrQu8bJ/7QnzF89aOz/5x0t5stfwFQdnnyjj67zlXODDGPY6nbjw2bpAOifIkRO4MXtejHjh5BjQJ2zMC4Yx4kLhhowdXPNk71zqJ8e9EmXvgz1sxd0VN8ty46bjwQcfXB9WKNds8Ts4OFhccRZbDLnEF77868eLXt+8YdeHx37NyGADasJJQUBho0/YeCTglYZXIh5veP5q0yKJlEQ4iVgAAuPsyRt62jAUyY44n+JakDa2tgnhwDfhF1Y28MSnh0ckBRcLV2Jg6TcnWG1M9haQlA/2sM3HQd0YO3H4w1a0FT4Vcdo4LRqdcaIWCxY7fbV5zTnyddJ1kvOpEo8H2IqXwLUJnXwPt5OoOywnEcLOOFx+SvjiGwurOfCzruLmT8evWs7ctRuXA9xteo8vxSimHLoQOUB98szFmq3Y7IgDxKsZBzQsvsatD0knvnj82TQXerZyYyzO9Ppy0/z50VlHB6d/dPUKyEkFnvXAx4Esh/EsFn94MIyJYY2MK4TOWL50OCl4dAw4udHpu4DCzJfdzIMxJU5xcKIW1/sjcK6//vplE18x+BF41fzFUorDB2f2ky8dmbVxfWsLSxz571sTPFoJj41Xw+zNi701dVyJTxdHuPGKq3kZ52+MDWxrBScM49bCPvrlL3+5Su8nOWfZI/Yff5jiOkf5uP/BdqKlhwnDfoArDhHfRcmx5VNlYripMw4nTviYHy4ERvkUv+Pd/yp5zGyvWXvr6lUnXH0f4IAJu73twkbnYigO/d13373ieYXq1aF54m4ObkR8NFgM/x/iwo9Xa18+cSQwm0d7xfyU+usCwrEJcgTAYHcjWeAWJBKecZpAn+O3cWxiviZUwnwu3JXd46AISB6BKX6JR9qk2MGQ8Jl0PvGFb6IEDr9w2ODi5BB+fuo5R/7i0rlIWDQ2OGQL2zgdTLHZ8XMCLa7Ft/Bsy+WszamNJS7BHY55so2LPoElllcp7uhcuN0tN9a8+OHpbqbnsrjgm+CfH3vx0oWjX97V+MGYerb8WyNj2rgq9ohNzJ8vW+thHi5ualJMPJxYfNTY409vssLkp8CExV4cnPjoZ1dsenvRmHVR6PDg54QhT3DZOED9b4A3eHtVRG8vfO9731t3qtaMrxgEHsFJGweiViYXfAlb+sZhhidemGpj+BHt8FtLWHLIr3nh57hyF+qkxI+w7diAA5etcbnQL1/wihvXfGA1r+La+/Etnr3ZidjNR2vtlZK19X6BmwQx+cCEZ6/iKl45pS8WW/PIxhysbSLP/Ng3R/b2ohguXsYdo/E27mbYK2L/LGsPeHVdvsMWU275e5TpSyw9GnWS9spPvDiL1fktnM4Pzac9aQ7OocfbkxJ5crFV//Of/1wcXEh214YPgWXOirZ54Gk/eOXiEbDzhQslnYuIC4d8kJkrfb7xMwdx5l5go59usWCoTGEAyCTVOaiBupIb86zZhcUkJcyFwiMIC2Sxw9nf31+fpe/gpEeelFgTpONXu/jZstfG1xhps9Fp46tdItjZJPwshDFcJYuE4+TGRnLnBQAmP3N2ZccdBungs4BtFPpia7PNR0y2+mp98RX+5UVbkQt3EC687o7k1SdqHKBwwxNPG28vx7034S7UKy85h9VBXBz2FfOem7K1M296scSIKz9cSToY8Q5PbLk3V1j4k5mvXSwfKPBsujsnceHOnC6Q7Q9excwOHv2UclV+xcfLScSnrjw79n6SvWw+zQ2Wj8t61OHxZ/syLuVlcpA7/WJpE7aKfnzo5UQfH7j88KOLi3yWZ7U1sYetqbziRYev4u5UKTZO9rX9ap/Ts1foxOVHLxYRRx92XPIzxr486Lu58TjVucAJ+HWve906XpxgPUrF1UnyoYceWidguTZXGNZBu+PS/PGYcenYse945oMHO23z0TaOK+HnBB9X+VL4mbuTrouGV55ucuUND/4KHrDd9HhUC9v7Uofbq3t5dV6ALU582MCni1Px8YNPby302XoFQ+wH84QNBwYe+gQfWF0U4VgjNnythQu1m3oXjoODg1UXR03UtfX5wy2f4rQGdOZG8NA2tj6FpcEZmDYQNaHPUbIBVYxL3gLaJuFz1l4iKZLBLjwT/D/K7uxFt6Pq4/iB9/wfPudKBEFQwRFsDwoOCCKIglNLjBPOIAaVcJDXgGIcElHjEI8SxeFOES+8yYVGMBgvBAVBaEHw33j3Z+V820Vz9NUF1VW1ag2/tap27dr7ebrbwkICtKgEeTpuLMnw08KUFCSBxhW0bcIJM9uILLzZM+7OrkYmpUnmC58uHuIzP8ZgwXNq8RmP+PWbWDLsiAEOj7TG8PUV42zCDVvjYRnHx4/8kuGXnEVmoftNar/d2xNHi46fsPPltOHD3i9/+cuz4bnhFCd7xZMvOqicqRV2xck2nOTklg3t9OiKi3y6eHLK11U5PE8kNg5rxWK/m66btFMS7HJAj/0wsZs/mGBAxhXjdIyJlR2YyBVDebEm3ZC7eZARq9MfO9YovDZHRA/lQ40HT7nAQ+VZrRg3BqM+HbqKOI3jG0fFYhzfBlxei4c9eFF+5ZSsmh5yI3e4w7eey6k5Dnf+2KOrH49teFE48Vxf6HRcA16rWqfmj6x1Kn8+f/T5kd9lciiwN7gZw1Lc1pk4xZWf5pA/PJjIaYuLPPz6ZPHIKNkRCzlERjx4bhr45tZNz9OnXxZln1x2R/H44RWmzwIfeeSRuXm4tsh285cL1yoMYgorH8bYCxu8O5+tU5hh0ycLR7GwWx+P3WJhCw/RlS9jqEOYNhtK+QgDv/SbV7L5KLd808PXZof+5Z8yySFmgBJKgeEAGpMUehaC952+ynl23O327wPQ2QRAHwz98pe/nMXEZsnRRuwWIOAFgidQNaovCYoxPhQYLRC6JlVhvwTS50ccJs4Ngj2nFToIrqeeemo2Za8ETIixku0G6rMIPI/mTlp8sKOEXZuvYsWHr9zCQAZGJyAXn7ZfrvO1R091Yon4l3c13+w++eST8xvdTh4Wt3wUq/gQn9XG5Km+mhx7XajdgGEjayybbGlnBwaYyZR/No3n3+Zx8/jw3+sNH/AieUFi4a+8pdM4fnJ84yv50k6GnNzCguS3U2frwJgTmm8oyXdk0/N1b68pyPS1Tj7LQ7JiYxt2/lCY8Mjry2d8eBW6bMpbsZozOM0tXk8S2mywRy8/2jAa558uP/GL3+c7Tti9IpSD5GBo8+O3vNFV+MQjj/DCQN414G8x+UKHOe2g6WnYb0A/9thjc13YbPl1cmZD3OLVZkdO+BCnvvyZH/nt1ZZxePmHK19w0aejbv3yoeCxQxfJK3KQOT8/n8/kHCb4ISNeOvw4mHki9u02ByC5lC/jER+wIBjosSM2chX9eHTyAxvM8oHE31ybWzrFoc22XJCRU3a63vmP7BnFjEe+cfY2TnwljOzrbxzp8ofmjykKbgsboIQKNhBqTl3oToke/Zw4nCjOjpuHk1pJIVviAsomvoXmQ1dfpRM4nkRtoosPS5OvDRN72i0ociW0SYCDXLotQnrhMlHs4eXLTcJip0fOwrLpIXxUIk2Q06vJt6h8cOwVkslhDwb2YVPCzR8ZdvDUCnk3CwubTV+D9XVYWMgj9pygfBDmNEmfnsdqr39cHC5q/ugo5YwPMdHJVvmAF+mHWx+/fMGUXTLslTdtRJYMCrPau3DvY70uEh/iC8FnoxaTrxyz5cKA03zqky0ObeuFH5hQ+LWTq8ajAxvc2jDY+Bx+io+cDc669K07G5jXMeWfDzazKy446cdv3eCXn/IQBvLadHZstY27rqw7n8+IlQ3jYqYHiyJHxsXVjcD16Vs55UZMfsnQDQTRI4+6XlwLSjbLSX7I5lfNr2Kt+Zaba5p9r/sujldV8uBDZgcwB4c2bfb4gakP1+VKyZfPntg0P56ebM7wGEfy4K2Az8uKu3ySk6vwFYdxPlxT3pKIwe8DaSPjcuvruH7PBkbXu3XpRugzOWtYjpGch0kO9engwacg+MIjJ/ji4L823+TMl7F02C0uew07xvDC27rgK3m5RfY8cpvq0xMvf3LVujUePnrNubhaK8bHv0HEmHZ9PAoCwishwHPqwvL7B7614olD4jxZ2DyzYXH40IZTthROLTinagvKb1d7T29SkuOrIOjwCXCEh9T5ajLxu8DwXEgSg/T5tIAsBD6Kix1yEumG0SsUer6KTAeFg55N3HtRTybydON4rHUDZcfEsUWOTrb54Rf2HYcYXTROv141+KUlNzIy4mGHTf5cTE7H99133yxuuNh1kpZH9vmGiV26LRRy5lqBa5ftS67MBxv4zQd9fXbJ0Gdf4ZMO2XzgiV1ubIaeUp3gyLOFtH0Zw83XabUPV9lG/CnkkFhQa1YbHr7psJttNT4c1oLcedrwOtCNzDtvG1S2bbKeIptP+dTuQuwiJp8vbfa3T3iN51t/z3v6ZLYuvk3YOvJKyDrwW8TNo3F+6ERsZD8+OXmq75Wcjdymaz6Q/NFD7MsNnnF8hR1j2mEVh+td37z60ymumexYo94ueHUlfpswO/T01TZuNxxj4rGWyChIX9xeiVnXcPEX8e8mz4/rxhwhMcPazRtf4RM5tLopeWUlN+WnWP1WuqeNRx99dF5J07GHia/X8vw1B3CxbV3ATM6H8fzLt1w2D/CT5QtPX5sttZjkmg45/Mb44YOeHPFlTB91g9WWL3sqMj9yHrEtHn/fjk0fJbhZouZdTuARgxrxYzyCD9b5UyaAAEVAu8QS2kFLoMkga2J9NxpAMhLvq73+KJ6TE3rnO985NwoXKCAVPlyYNvEWn4QALKgWrAudjo1aDR/fJZffiE7YJVPbDQxeemKyCbgxeOVkA2GLXTWiwyb5Fkv2jaEdw8VxyvKbq07MEuzm5ImAf3220stH9uAxzh5ZsbphuLAsQn04wmdx+Zr0gw8+eI1f+aMnHj7YKz98wCBPCrm90PXZNVauYAmv8eaAXjGzq88fXfLiwuOfXnwHjBauvDhsWDPw5rtNyPz6Tr5TnjVAjwz7YlLr84HCxhf/ivUTNrIwqpExNvGsASdzT0JeuVgj2ZALJ2ZfSXezkWM2bOhsi48NeLq42NUvp2yFEx/RUeKzCTsig6/G58f14gNom5yno2RG4b/8wabipu3vg8Fhg9w4mYSh+cpFsVaz03XWPHuKcyO2Dxizdu0L9gA2bVDyuEmefd7g5k2erT2n4oVvH0Bhk/P8s+3ahgm1FviClxyeYg2ZHxu7g6u1aE1GZOTdN9du3749n9XYrIuJH3ZhspbgU/jAFwN81onDn7mzfhxGHC757lqGl27rSRziQnAblx+++EV8KnCSJ6cUn/XCv7795+zsbOZE/tyg7WXlCV4Y3bw9pbjJeUXnGmQTNnbEFsEvH2zAbZzs3NA2IycGgWfMRS5ARiTe47Rk+FbQ6XQaY546fJ3Nn1HwLQWPnt63+q1MAbArIZE+ECboU5/61CTHJuzrkoivFkCTj6+N6EsYGXbDq23MKd6Nyaslr9aMS7xYPf0Yw093jB4/xLupTcJnCk46MJcTi8ajrVOQTQk2fpIxUbCQVyNx5bOFABM9v/nuYvT5i4VOlh57Fqbfp/DtFb+tjse+OYIDsSEn5UBuWpjktWHULh/lhC9tOvDAXC7I1m5869Pd+rA4OXq1KRY6ajdsixY+GCxY75RtMC4O39MXp42HfTxUPDCUM2PsNq42zq5a3hT5SCZ7xt2oT8fadYHRIeuEbg17InbRy6ubOCwu6AiebOYvG9a6HPOFJy/lDHZ9pL3zjRd+p0MbkOL6wM8nLM0PzPqbsqE2VqEvFteu02nrVX7gULMXLvJ09dXGxcO3Ojm1p0W/lOYwKGd0XVuuPWvU5kSfHURfsUdY72wg/pVN+mTZtBn7RUCf87ElRjzXM/uwyTVZ14V4rfkw+9tonn6tyw6H2WfL60r7l2+V0jOX7Nn/zFs+OvTYfOFwEBGLWK0TByRvLKwBvn2V3isxsmJls5j4p8c/vnFyYigX2vlXs8m2tmuIP58ndjCD3Q2ifVo+s8c2sv4V5KDHl69Ve/Jml3088mqx44tPH043j7mBCAZTEAYU/YgxdzgT4MTm/XybpQve7xv4PvQ3v/nNAeobDRaPyfLah75vBvWbk9kGymnbiUDy3HgkBLFvkXvnaJEAKgk7EQVXQHCXbBidbmDwKgCfvGIB2PjdFCI5sGA89nkdRx6ZDEm1cXuk7akAny0XCp5FwL8cSnKLtpwmr28MaSNjHondkOXXBso/e/Li1Y5XAV7t9GdY6MlrE60vZya5jcLCQGTYU+B0QvHI6gLgQ6EHT2sh/2LEQ+VPnzx/iGzrwZw6NPSVWFjEgI+ae686vbt20jfvYmbTjYZ9vuAhr69tU9CWP/w9RpcPvvDLrTZbiK7iFYaNqwsIfocXTz++9kzHZsgW+eY1f2Pszo+dM/JswYn0851ueOgZq6/Nl7lyTfjihE3A9SAnMJBVrC+/Fd/GxlZ+xWZzRNnWDgc569s7f4e+3pOTjeDY/fT52fFmExbfPrROPXmwT84J3+bKnv6mdPNF1ytZMfGNH9E1t/YPX6d2LcqxOXezdQPR77pSy5c1qTgkWJP2JHuCMXkwP/6Wm/2FnK+99wTsgOkmQEa+jZsHep4sYLI/wWWPaU8sb2oy8mx/dPh0gMIzx8UnD/C0dtk3ppAVo9jokIWjcfuG/cprzj5GKGdkYVDEoGYLaW//DhQOK56+XJP69PMPEx151caHuXLYfdpwSgZMpARxZPO+eby79hc+T6fTKAqInovfJuCPvVnw5G2C3tv2Ds63FrwW8L7e47hkKS5gN4/eRbpwyAhYoiwqr4csThsR4OwbF4yJFRBZZAxfYs/PzwezxSWWFhXMFqrN2C/Z2GTxjJtg//vDBYD4Ievfmfr6ng9SyfInofyTkR8++S8vZPDItziaOHnGszDIkPUnEpzeyG8y7rXV/x7/WldbjviQDwua/03mzgLwJ/Z7b06PnPwpTnBOjRYMLHCLBSY3GHiKDWbELlnEhnZY4bERkXUS85rKxcoOYhcZN19ulg4cXsXh2azV8oGsO/66cMq5MXEozTs9BcGkwMc3PlnxmSNkLTiBeoXpdN+cmV+/TdxBIOzFSjeb4kXwwcZPvtmXC3qoMTkgIy7UnODxadwaVIqHnEMNGTjFzLdrzt9l8jkDXb7Yten5L3tuIHDQYzefcqHv79dZazDYkHsKE8vOO5908MoDGXZRsZHzxCZ/Diferbu26cJCB5FLV7t4YfJ3mxzQrHV9hYyY6YRN33WKxz4M1lPjcKZXXtxUz4/9wAFNLHSMid+rPHsXnHLr+uLDL+nam/jBV8OEjNNXs+c6bL2QQzDwYx4cAuhag/HJKXLQPhaPTPbV1gA/bCnsmnd813mfleZ7ABw/9Is12/qb9BU4PAyYPzdlJKfWuDhhkgdtfPHCPbnHxCAcQAHXBtJJwusHfHdUr5p8GCc5XqlItlOnDdumbBNDFvvDDz88H47aKPgy8YpHW3acYPA94ZwfE63t4nDS6K/HuuvDx3+TIHC4JcdikGS1EwUMLhKJ8a0epwuyEi9JHrmdkNiymVpcHgNdhIgdp343MO8I82uMX77Yo99maww1ceKQ9OSNaRvHh60nD++mnXj4IRNlXx5seHxZ7HLtdxe6iODJpjkQh3xtMp9eE3lUlV+LExY4UW18OJp/fZjiW0DGYNFW+PbaEF43N7kRH4INsSHXt2/fHhzsZdMYneLJn3G2YYQPrvyP0Ts/yNNFdOAoLjxYFHKepj2VkhMHvz5Upe9gYX3ATN+YNcNeBQ6kDi9MZNmHl2z92qN0/GB769G9G2byxsgahx++20f+rHHkZmvcCdI/I9oXP3zw8I/0+VY8OZO1+W2ZETx+FL8cGNdnp7Y+vwq7cgSf16zyR9ap399Ccx1G9BF9r6N87d3GZT06rInTuupAQl4ekTHXrDUQkcXnHw79CHbXF4xukuy0BrRdH54e/Ekd+uVYzr05cRPE+08IzvJMXt/N0O+MeMXtxmStkWEfacNovcAev3VTHmBu3dPD93mWpw4HNftqtthRIm25RmTKf+NqMfJZrOzzp9Dhnx1Y2dKWU3wxTWJZ+REAAEAASURBVDQGFRRgmzZhG5ubRw4YBpqcRdgrC5uwiUAWgFdE/ried8ruli5cAJQS7uTiCcXG4kN4/iwSG6b30YiuRQYs+zZ3WNgpKerwwW0RF7h3256STJDN10nA4rHRWdwWkEdvj4L6bmxuHr4M4IJIFmaJU7PNHrzdvPJvQZCTT7iUFod4skNGzG6+MHjyIYvYRXIt9/fff/8sHH0xwOq0/++IfXbgUtcXPx6ccqgNH1xyt+do6+Prk1PDWpsNTzzWgHlGLThj8uWm7R2zp1E59trKnMofW+xbyLDARics2ehCL8f4sKCwscW3MfrllG16DgSn02k+c7JJiMGTsZMcWY/xXluy51WXE6yvRluniN1y6SbOVhegMWvfKzGla4jdcsaGtjhbF2JlU4E5vjjgS8eG70nfHLou2CHjA2mfodWHJx359crSBi1+ObTuLi4u5hqwYfCL2ELiQHwr8CeTbX32isN1Yl7lT9xePxtzsyKXHTw2XFsK/+wrXZt8s0+HXW05gq/1Um7YE7fx8GqLgV/y+khuIxi8elLuRnwq7CP42OSrPn/G3dDtdz7PE4N13eemZF235YANOvCzlQ18PmDk15h+sYTH618HRE/SrjUyiD7KZvyuAQd/ry3te/TtI9aoBwF7pPVhruQ7DDCzq8AZ5nJP9ojl6c3DaSbnQEt8gVMowBvHSdirnozQaZIBMTEuNo/HXlOcjgvMoiXPpnY6AsajL8AHHnhgJgCfHf7Jwqg2MQLRrpZYtpOha1xB/LkJql10EuvJSXLcyT1huQD5kwMbjNdWPvjq9MR28fMLM3lFn201m+zjk6eH8EwKGbnVh88G5uQubyZPvGQUMvTdKJyI2CsufDL6ybKP0tduHE/sNjwYPRlaiIhdRR7J7XmFQZ+dMJMJh1iMy9s999wzJ842ZTkgh8jwacPwhKQg+ohvhwgbjwMEghNZ6Gw5FMiVHLGnlAdyzQWb/HbhGBM7PfZ9jsSfU6nPGsQlJoch//bYXDqwsOcdtps1vy628iFeF6ITtpsmWWN82hA98ZLnhxz75Q+21gZfiL0dD9loyzsxe0VM1jUTFWs1f7VtEG7c3sPDJmdsyhN8+sXFXrltXeA158bMi3gVxJZxlE+bqFfPTuBs88VPcyPu2r2KhY1t+TLuGuOPLbUbMx1++YFPWx7w5RAfmWt6ZIxZW+aQTX3j1ikig8TAj76YkD7ciG37D1vyxgY5417B+UOI9g5rW8x8WXfskS/X9fmjy0f5x6OXf+P88kUf39sVazX/+CjcbCGyinGfH/l855Of/OTsJx/60IfmQ3frwkHZA4HYXBPyBoO5oA+DEjZ4+Y53+T/RAeW8QBkKXIYYAbSJGqR3fngM9UrKac0CcFd2qtbmLFsmMBvssdUHx8YE0gLVJ6MfaIHFx2sxsg+nuzL79MRwdnY2SeOHHR/o+8qoDdvXjN2N4fO42wepbSBhFmI22bEwEBwSCjM+GXi0dxxhbrHA1eJ/4vj2B3/33nvvnCrokzPOnpgq4/TOD/Ht0liydDfB6ltP3vu6wedHjC0WF6Oc4sHPvljY1EbhIpN/G7McsrnzRN6aMubm5aRfnsRoodJxM3MBeh/NL3+IrI0DLnbZKtd8w12c7OHlrzzg0eFLsVGxqY3o021D9jrRgQL5Zou1+ZGPfGR8tR7kEu4bx2HKWggvHfnDg6N88iUH4sFnh0++Ffrljw2YEXlthZw+v3cjmLY9Nt0sbx+vvPwJkdaqjZYc/zZRa9H8ySUM5FoPxQUzPoJdHxbySJ8sXhh2ruQcH3Zy7PNPnm9tYx0s5Ku4i8u6w+eLX7VibumLw82HbfOLR8dnqf6MijlhU/ye1nxeRCcKfzHzUXzJuFH84Pgz6fY2Bx2+4BODpw8x2kfEYh3AUNxskIUBGePLuFwp+YyvJl+c9PyxRwcJB6tyZDwdPvS7ZhzMfd7rcz7r2V5j/zNuH2CHX32Y5BN+uZFvPHE2n3ziiY/P+T0QQQRGW2FA0jl2epQwm4CvmJL1js8HsiXMJqDv9VMLxQIpOQVWkgRpTI0XBkkCzOQbB9YYUrPNruDY1O5CNeZE4wRI7+y4efg8ps9k2HC3FbyNwe8DuLi8siLvEe/ieLRnQ7xwtMBhbOG3icoD/2JQsxsuY/TFACeCn6wx/hQT5nMk+eYL5i4k42SLf4wcP9hVkNpF4X1yr1lgxacHl7558t7ZgjIGKz4fiKx28wQnXgsFP5vFpq8tX7XZKr7ywo+Tk5v23UiObDI9KZpPr7zcXD2diYF98xtOduLLORvmhZy2uTKu7FhgcyLzeZA59lv+8EUwXsVpnfwrgs9rG7m3gfHl9UX50IdL/uDCb17DJaYweHIxl3S8XiGPxPHviDwiZx35Oio71nY3DX7aSGEpT3u+4DDnzTf/cJLBV1ozxsgrbBUPPl/IGJ/m1Bpn19wgNvHp4sNJ1zWBtM0tYo9sxC5/9Fq31hCbkba3DeYmTG2GrjE2XBdyRpaP4jBmvYVJ3N5KeMrwNEyWX/bg8rbCZoxnDDY+w8ce28bCLrbm1fzhN49iQfyyhYw7hHkCQTDj8aHNH3023SjMvc887ct8iYeMjw7Ckl9zw0a5Z79ckaEHC7nWAJvzxxQN7gQy3sXqzxKM4AFU4nxww7DTuvfZJoAD+vRclAIQNJ6ElgByyQZU8smQJafUFwRskoQHh4nC4wsvOyaGbcn0uN7E3jhOiWToGhe8OzEbThS+py0O37Lim334+STTBGq32Nne2OhZaDAhWNiApcVEpgsAbvbgUZPzYT99PJjlw0LpUXsMHz/4dtrpFRScnv58jVpt3ppgflDxhIdtPL7gik/WGEzG1cnqk2UTTm088bmBOOl57+11mziNl0N20y3mbLNPzreDfBhMjowbHTtuejYB8mJD5lMfyUe+wqgPVz7U+jYyF4A162DEt1h8qG59kKMbBrFmhxz7xQ6jPFtDDiwOWeLAk0+4ktdnS984OXYUGPgwxr9XJL6lhsiJGTbj4TXGDpzs0lP4cTCyQXjSdO2ybU2TI88mntqahbM1ww8biAx/dNjmD4b6xvDlgS18JR364mPfeGu/mMmVE7LmWOEH/6p8uOFjozyyU27cOGESD1w2YfKNG+PLAcIXEozLgXFjxVdOrRf461t3/JUbPhS6MNE31p5kTJ8NtO3j0wufeNmGB7GH6LBjXA6aH/yIrfITzxp34/BlJAci68L8hp9NbbEp+nKKV97Y6poLD8xk+Wdvjl6UDCjAZJyAtgSYHButx2LyNjHBWjQFxYlA2OEgcPr4eOzjqwGWGI+cSNtYNukIQB8GfvhWK8bhy48xGNj1JPTQQw/NxcO2MVScahPtYpPc/OKjYpvO8YM/WNiBBSbxaJezbBtDbMIID3v80SkXsJKh7/2j12v+iB99p15fnXZSKnaxypWbnhN0p0k888E/nGyaOzW89JoPWPgnx0/+jcND3riajNLitSjFEX7+yokN2bivZXtSpddcy0VzrR3xUR7zzaYF7KnRZ230rDsxaiN65VqfH/jjhZ/NiJ/G1fD522E+IxCDPERbjy03HfrmQV6N8+mJw0nPhiQv5OQqPPkMD/t0w4FPRoHHPMFCxudFnsKMibd1zjYf+LAg480vOevMDb0x/iIxIPLsILGTsZ7gIGMc5QPWfMBAHk7zwmfrSh7gEQcenOy7OeDRpYfHF3k8sm50+bcGEFvGs6OtwMgOXPHgIx824/Tw2VeT5cOTY7nAK4flQA7NrTH2yhE75lmfvtibRzr54yv8dMiQNx7hsY1gReySYRfhK3RdX8ZRsU7n+IFPThzmVYwKW+VYLHwiY9qw8UXXXGaDLB/02S537Stw07+eAoOUGAOCU8ocMIDn3Z/EZ7xgGMJjlI5gySM1vkKGP770+SKPZ4xeZJychWQMjnCli7ftCo49CWTLhkyHH3LFwYd+CdKHga6C+CSvFN9VfGTZMc4WvHjZssHgsWEcFjViiyxdtSJWp3jfZpFrel5FsMGmeC1Sp16P0+Gn2wWbb/jxwhd2fPJsNwZXerAZR9mHqxzGhxuxy5YPKb1ygs9TAx6C3RPK2fFq7uo3XmBR6HjU7ivd8LTuPInw7aKFl13jiK52eceDHW6yxtTFTA6xgQdz766zk21y5UceYXBNWGNyRd864yt75PkIDyx4ZNilJ1/ZVSv5zi7bNlzY+CsXZOHgl0+HuuKFF9ElR0abnPyyzQ8sCN8cqvnQZpscHiLPFp1iKCZ87WxqV8hq86smy651lJ2revo2SDX/5OCvxmfLmDYqPvjJdYNKxjhZdUQOlmyJOTvFRKbc4O1Cr2s3O3gojPphtBbYY0OdvnF48bYMPt8wKcbFI0ZzagwPsYn0vVrz+bObgJug9eN60i4f4dSHnb5aPx571huKV47ghAnG4rmOqRA2WPDAtgAKgjNtyuQDtPnj+fgBBLlkam9fLSi+yBUUWSD1FUlTI8GwTX4H1HhJV8OIjNVmWwwKYoevxndywtxFr8+/GqZ0tx289OCji/ijK1Z4yCH+kL5i3Oc0LiY3D34QOW21xWRDRekVDx45GPhRK+VDzhE/dGBU2CWTHzbKW+P08OWKTfFoh9vi9UG4z3TYDqvXcMasJ9jpsJM9N0vv610EfDZGDibzzRd7xUUO8Y3Pn7FqekpyZNlji4y2uFD6eGwo2sl1ek4uLPT5zyZ/eBV2IroKmTA23hyWT/raZNneuuLRlyMbkHzSJ4+XTXxtvthDZPDKmbUpzjCRY1+fDMpeGMsRX7CRz3454xPffMOoGOPP+itf7OMjfozTbdOrz3d+jYsfr/hcC/rFXIzlhr98VLfG6CH2K/pyIF+IjD6/SGzhwRM/nwodY3jJ81VbTQ5PLGTVCF+bLz7IaKsdwrS9KvW5LT35YY9PNw/ffLVWXdv423Z9uNihg8cPn8WYb1iMG0vWGJ6C2D+wPR0IBQQUCjwwxiwGzrS3cW3OjaE9IWzocwTEVT1ABGSMnD6Q5MkCH45AWwxk9uLIZ7GQhQcfj63KgDx+8JFPY2yq8eLjKRG7xvDCo27hJpucXOLlm0/x4qvxi4+c8eT55EveyckFHbaVeOp08NmTI5gQPbmokHVBkwknv1H54kshj7QRf3Cxp43INFf82iTMKbt8eKKywC1+8bCVH7rkPdmSZ5Ntcvqw4mnzy0/rhW9j+U6XPp4++2o4lI07jMabhy5K+uka41vBK2/V+NrFq11++dMPC8woHe3GtZG+whcc7MK6+/IcZnxtOmTDiKeQzR88CJ5k2daPtm962z97ZMkYy64aVvzWTjEnz7686IeDPURPO+z5za7amDpiQ2Fz62nDXBzkyfHLhzYZa8t81ydHB+HxRSc98eHTwWuO+UJh5ENOywV5usabx+ymS5bt5OAjE7ZsuY58gan9mF+yPjvzhMyewxpdthS8rt94dIox7GHiK/8wp0sO0YuuUyLMSZNrwRWMcWPbqHaTTU4/5wwHWI2Ma/ODyBaY5AJpPH4+ydCt0BWMcYSff7Lp4TeZ2tnVzg8s6YulRch+uuHPNvn0yLCbjRaMeMjQUdBVWeMK+/T5V9tI8ehlgy6eWsEv/rvFbkwMxsiyS19fLU7268MXTrLbH33YyCpInAod8vHZRuwnbxwedryO8V7egaQFyVcXo1Onsfy7uNnZ81E8+azP746TT4UtsvXJ6+OLq4sZdn7CW81u+NX44ibPBv1s4+MpyZLnMx4+uXDkR13+jJGBR00Hj0w++Q2zMf3yJIflJ5xkIuPImNIY2+FtnB15IlMMtc0b+fYMtooNFnLFQRfp80/PWt98PCTe9gQ24tPVLrZk1cVRXrKrz0aY97zQgxNl27h46ecHT0H80FGTi7KrD2OYa7MHQzrk+RRntsiGkx0y+mr62fKZnVfz8HW92K99eclcKOThVPgpJnblhK3mTxtln279cMMBJ5mNUX+yYJAwowBzgvDxNghtRL6iz3C1tgDpam97ASUbqOzwTY9OCSSHp+xkbpxspsuWhJW0YmsxJbv9a0s2XfLhx2c3KnZYULGxbaz84bPDprYxBY+uSdY3Xj7Y4L84jSPjbCDjSGz4EYxs8t/Nnx2P9mS1+aajJlebLpnyV9s4f437PIcuKibtYiKH4BYLHPzo9xRhnF2yamN9ay9/eNp80OtUVd7yEway6RYjPxsr2WS0EXuoMeNIHo3py4UciANfia+mq8AMF5/NUTj3WD7IKWTS59t4+vrGkmvj1UfWTvi0Yc5eayFdGMiyR4Y8P/hkimUMHz9aA+QQuXToF0drli22YdA2Ts44yocxJ2d8GPGbh/TwWz/siBvPOAzsaiM+88UOnMVonGxPGfr8tfaLQaza1ms8cuyEnx1yKH94Oy58OtmDVRt2tsmmAyf5/OAj8mLDR/jJDuP4YZxNhTwZ12a4YCBTLvARWTwEm3ztGPc8kCOP6Cts8kVHPz8jleGCbDDnOxA8fQ7ViPEo4/rssE2uQhZfn+yeJGMlLwz0bYQmQeAoW3S3zlXb7IeXDlvs8qHQ3322yeCni6dNzhiCg689iRYmXj7I8aHPtzEFOX3hm6T8q8nhKdrlgv+IjXKfjJr/5Nly4SG6fCnbjjG2xEAeaWeTrDYZ43Ki8CF+MRjPfjZamGHIXro2EGMtUH7Z4IdtdTbJmHu+2G3+6aAwqreuMXYQ//yRCZtcFBeeNl/xyeJlwxieok2HbDGRM1b++dXmWyn3+Y+vZi9iIwzx9NkyhthorWWHfzGqXSfmnl7y9G2kyYRXXwxRfDXbxshoF0NYsmdOwhHGbG69bJJngwybZLKpzQZZfo2rjbfx6xcfHnvJ0VXYQdp0iyue8Uo884BHXs1PWFwXET7Z4mms68/8RGQaNy+N4aPyFEY8sbBfvPT5pIOviBm25trnI66RdOBHZItDv5iLzzgMbIctHj24mi99euWn+vIVFkHEoMFKAPSBRtqcZrQJxFPIAaJoKwVFB5GrNs5vcmQKiEx+djt82cjvnpQtzzabGyPcirF8aIcjm+zgFTc+eX21MbbFqI0n9mJIl534bKB8kClH7G4d9ir4+ceL6CD24xcvuy748Ifb4hR//OZxY2HTwlTYZYs8GQSLnNPVrhh3odOx0MOdb3Jsqa09bTLG84FPX+FDPG5AxsOdX7qtN+MIL7/6ZOsXC16+jRWbWmFTQfyX0+03f3hITLXrq/NTvviDA6npIXyy9fFgyWbj8gMbe2TpkNt5S9f8keHnqs9wkc0GWX7MYfGHBx+p2VInDwsbCjIe7mTqG9dWymt5zzYZPONhD292yFbIkwvD3lTjiwcWhe18OazInfWaHTXfEVl+XTvFYwwvm+wXf3Ft7HySzQZ/YeKfLfKIXPayb9y80GOrm0fx4aH6MCP6ePQRvuuJfVj4RvTDSKZc0temrzY2X+PdCcrJBshoRgKTAWNAxddG23G8+AWYDEC7TY59WApEv4Tj5T9bBVWQfIqLXLbyb0zJhjb/1WyUB7wwkMk3PjKJZE0CSjc5fQQfHTEgfBPWAmsS+bDo0cZNX0neOBt4iG39MFYXf+PhTgdOsrvQYZc9BanhNQarEj424KJjAzGWfrE0xg9SwxKfPju9ciiv+tsnuQo7YYCPPf2oWIvDmLL5/OvzXS6KmR1tfHKKHOCJMTnj+mIynv10ssHexqtNF5Ehj9gxhvCzm53ya+Moj+zwGw588niuZUQm+XzJJV8KWePaxrXTy775Jce2OcqeOqzZw2te6aBqMvnuptM4vX3DCytdbVgUsbYntCauypqPZMNVjY/UePAjfhT28VrDjYdXn9908GFnC7FrHKbi1jbeWH7Z0q7wHRkrJ8ZdE15bid23NfsFSv5R/vjMT/wROH6wwyYikw4ef3RhV/OTf3Kb5m9hMRDwArk6OZRyok1HIUe3IAMWCDqB2mNs4AuaDWCTU5M11uQ1AXyyTYecvpqfcLAd1vxbSCg7ZBXEHv/62cVLRl08YeUXvvyzo11N3jh/2YKlOMkVJ5kKHh/p0VXYxjOWz/Cr6Sn5IovII36LkTy+0oa3bexc4rPFdsRHONnYtvUtOHo+v2A/e+zIiY3HODuNlRc8hF/+9NktvvAUK3/lhCwiwwadsJPhd8ekbbw6+eY2H+yFQUz4eGxuGb71IxiSy69x63pjbGzr0y0GfhREj438woynoDBpFzv7xSjXZItVvcf1+UXZMq7dmH5zH478pxfeMXT8yAf/iB4d64Ese4gPbfLaSD/M9POJX47I4ivh147o0+UrW8boKd206CrJq2sn2xrQDzu8EX6+2dImp5CzvvC0rV9EJ1/kkHH5wQ8XPv/JVKcfDvx8dC3hJbdr9ujxgbTLOZ7ckE9GW7nusQ04jtSoYAkEWjtADCO8AmvcmMJGC4VT43hbr6CMIfYV+nhKgOkKosCyBR8/ye7aGB/h4INetqvxlBIlmTCIjS4bCDb+EXlj6bGFdr6K21i66mTZt3jyX3zlZePeOvywky0+2aIHe7JiCAMfxYXHF53mik39TeTw2Mu+cbaKh4wCAzmYtbc/ffLFRz6b5OjoKx6p5aRTM564yNFPT80O/OznI1vwGucXL1l2FMTGpvh0FX2+s6POvzE+8cjiN5ZN/S7+5NQKXePs0NcPp3FtsckFuWI3JiZkLFtspA8znfr5oUM++/r0UDb14+HTJa+utI42PnGQQ9rsKWyJDW/bih+WbNMhzzb8+nTx6PCNr63QI9PaIheOroXk1MbMCfv11crGTU7pM7tympw45Z9vMsVmvFg2Dm2y5Y4uwgszPZQ+vtIa8hczxMmfHCTfGtFnTy0WmMWJ+MdTInLwonLKHxtKY/zTq5+++jrnLlaPw4QYKAkZLdn6EgDMvrACXcIKLDv6V3klhwy7SFspWHa16cJGjl+6JUIb5bvA67eImuxiMhEXFxej2w9fJXVqRv4OlX64w2BMPvgvX8mE1Tg/+PwXk37YtZMvlvJAxhh+cYhrU7blo802fTp8JpMfOWCbrRZyeNgmZ6y48LSzSxaFXdv4xmls2whL+YALjq2rjdcmyQ95hb5SW7zFlY3wkEP64aKH+FVgI2d8U/JkUDb1tenF1zfH+QljMvW3Xrnjh01zpk0nn/p09dXZ1+cPaZs7smzmi53mVC616Svph08/X2zos5POxrpxwGDMPGUrHnvNEZ1tX7u++eMPxQsD/WzsNh9hVIev/MKiGCPLv754kFyThRvRN06OvDFYkL9O4HcqrpJ/CtW6Laf8ZYc8exX9bNamF4kPqeEpL+HGK042wywmhW8FiSE9PjY+MvrGlcazqd88sNUaoIefPHyIjebI2HwGYkCn4CnXxqeQkwzRIYMYFYQEuzv6zWI29OkVtHd35PACYXI9BZVA9shLWEGQR8kM8MNOGNXJtCBgwifLlrabg99FgMP/2fC/HvwNqrAky78/KeL71vSMX/UJD3xsG8tXso1XG4cN1abn7g6XugUOH7vZIm+MvrYxMnKnjYpfjsqfHCjNJx0+UTx68dhGeMajfOAliwdPmMg2Zg30ZzayFT5yvrrr5pwPGOkWNxxiD1vj3vuKuVyFl02y+tnUznf4+YCXDJtkksPvQ9TG2KXTutt8PDiMs68uRva18x8m9iL+6DSWn2wZZx8+bbaMtYlZD43R1c8fPsqWcbR9DeP4kV3xoKuYjCuIXeNq8oqcmCvUWDjMp3HY/DIpPP7elfWBlzw5c2tcrHDmI59kw0GOXevcuL71FA8WNvCRWp9PMtYQ8vmBPxCLJ9f+YvQrX/nK+ZcQ7NKBzV/z9bf19l8koFMMzaU88FUMxaEmX7z65YhsOGEyll6x6yN58/e53OT6DASfXw8A5qM8FjM/Ct94aliQPh/lFS9f2ki/eIoBP5vXgVIYJlji1IhSihKfQzWdFo++Iogbx1+TtUnoA8CZ345EAsQzYXSNmURt9uMViHq32aSTP2Nw6CN1STJW0oz5A33+Vzvf/u+xvwDrz2wUAz3yFpZ/tuK3Pk+n0/DIGJMLbbIIr1JSySD+5QMmMmjrs1O+yBiTA7bzx4a2QqYiBgspe8ZtLuwhc6ltnD12Ijb0i1cddmN0EZnWhIuVLTyxWDP65PXD7KIi4y/cwgNnNlyMFrq/JGw9hLcL2yO6+NktDriab5iSzSZfCv/0YGFDnSw9hEcv+3TIIDo2IbjYK8/lSg2HPMNGBqnFkR1y7KZPJvzxksFPlh1FXykG/tJnq00T1vTpwURPrYixfNBRyMmBGpFngxxe/NpqY2GCRRt+PmA0v2yz6wZs3Bzrs22cvH8D4eaBwmOcHXuCtptMB4/mij3yxtlB6uaxGGElSw6ZEzbCb1zOsoVvvfmzQTCw5y9Jv//9759D5Rg5fuA7EPuDnv4Cub0BDwb54CM/+VaXJ3b4woMhoq/Q3XnWb/7YMNaNGHZf2bVnsenwaw7sA2y5JuGzFsW6cRmHG48d+njsyxUef3wj/fDhi1U/bMbpXn4LS4AlhgETT7HgOWGgJBlDgTYG/Lvf/e452ZuYQJLbi48u52o3j2984xvzv4Of+cxnjg49Y1cTkP89HmY1ogcjvBLPrwWtf/PmzfkPcm5u/sSxv61v0V8li9i42HZe2A4XLHLUItJHMNORYG0YED242bQge1Xm3+n6f+z+nHy2yLMxE3TEpa2wgaftX1B+4hOfmBPJs571rPlTIBaXIhcK2XIIH//ZClO4jSn6SnHSh4sefvNm0ZpveSWbHYvXX+R9wxvecPmXhOkYZ8u4/5/x6KOPTv5cABalvzpMRk7DCaM4woJvLtMpFnrk9MnkL36x4pOBg09zxF8xaRvnL5/qLhb2FISnzU68/KRrTDubaqSOB2/zJTdurvD0z9jkmH2brLWz4zcvcPBjHvT5YzO/bLHPNp9i10fJJRsmNT2y1q8223zDUuE7G+xlVx6NWSNO7Q5q/me7Ez475Ss88JP92c9+Nv9+mi3rmDyb8Cn5NQ4bwjMmPjjZQvTIGAtn68b17Rr0j8L8P5js8ulfGLOzqX/HLRf+Ejas9o18k9UOkxhR/psXdvNlXO7KNZzakbFs0hOfPcM/wXvhC184umJV5Bse/6rYwdef8XdzsSbEXs7Dx2555Sdc5PgxRgbhdY3IH2ou2LuOSYEiBkXt+k0ymRYLGacF/0DF//S+cTxxAAGs/yPt33z2J9IDwqmyAWibTCBtoP4Rz06iNptq9tNnUxuegiWDr8ZHan2yxsQk3mc/+9nzmMq2i8IkNNn0+PK/3v3XMn9Pn24bszE+N6b84PNFhk01XZNr8vnw5OPzFTpk3cT8f2MXCx75KNybZ4wcPXlzA2Zf2/++lkN6ZEw8XTjCAreC8Mjqb994xvDkjG3xiM+82+wt5NPpNHZapNmWX68HbXrbLiz69D0NNrfsWvD+vLr/5eLPo7uBO7GaL3jIyF83LJjYy2dxFrtxvpoHcciVP9/PthOaue/JmB39/v+Kjdy6tmmwrfCvsF1OyhNffCtskYdZXQ5qF49x8STv2vFqVd7MpQ0VDmvPgQx+7+flqrkTH7tyKRb/60ZsCD/b2oiegh/e+MkYY49t8ZKHFS55zGY5JkM3e2561vPb3/722Qc8jbqJmM90w8a2tpsjH/7NtJqc18hPHH/h2U2omPHzky5e+cdDeCjssMIuJk9DnjT8e2BzHHk9lD88uNiVe4c0hz2HT3852uvtsJDlhw/YykdY8PZa37lin5xCLpvVeK45uT8/P5/DphwZT09bXOTsuzCKA58vFIbpHD/oGoc5ggVOa5Keoq+4DtgwT/Zse8udPfHpkyolE0eYUUlQUHVB4lkgFoV/r3i6s5FwYDOzUOhseWAAJsNPZIPwj4TI+uuz+gUALNIvULpsCVYpUIsD8Yunr9Blk2//ZMfpwenOog5TNsbA8YO8f3XrQrRw/a8L70H547+Fkl6TITbUhUcuHWPwyJcnLYRnYZh0tuj9p3Q6cu7fr8oRXb76J0LmgF9j4meXDN6eE2181JzAVBxsypucKvLiNHZ2djZPGP5nCZtkKmyJyeJiO7v4SP85z3nOzAE/9OCwQM0JnputeXIjsR6RWhzk5DG9sPIFI74FTt4YXnGad4cb/WzSE5dciaULhR0nOpu2cVSsdMSIXEgRfoVfbXbKO7/5pqPNXyf7syOvnkZht7G5ebBBjgyb5K1LMubH/MIAv9z6o5Qw09MnB4O2go+X//Dpsx9uMnDnl64+fnJ08CJj7DvcvPa1r531Cat568ZsTshcrcn4V8L+K6V4jXvK9uG1E/Xf/va3iYNumMOKRwcWpTbMrX81fIoDjLXLzsXFxcBng4+HH354/uinawjPpmwvuHXr1txE2JRj/06BTcQnvGo6FTxYFDz+kL72xi/+yJhSjuTAXuupvv+OueeAbbKuTQc3fddIh0l22cOHn6x+ZB0ZYxM2lDyeXOCLz7o3vw7B1uK8wqKsMBTpIzVFhXMTAMCb3vSmuYM7bXaBpZs8OYUNtYUOSCDJa3uS8QTiglEj8oLgs8DChF+A+UqWvDjos02XHp4ThkXiP/e5i1sAxrc9eoq7uNOgtg3N/xPP5wA8fuibePaLUY2Pkm/hql1QbrDkEPtOjmry6nKUnWwbQ/ridBOP5MFFKj4TzAY5RYyIf7T5bOqTD0P+k6Njg8L3xOmG71Ffjv4VwcOe0hzUt16a563/ohe96PKdtD8J/9Of/nQ2cfrmTz7kW2FLPPywZy7JwKmQaXO2Eb385S+fNfvc5z53Li5+2ROTXGqzhdim7592/eD4H9jWrT5/5MTtQjJvLtRyt3Xx2MtHseOLRx9ettViYNcmQEY85hc2sj3B4sPCNjzmSFtN1snzRz/60WBr3uEyRo8/mLUVMjCypcBRCRs8yZCnFy628y8OT3cONk73TvpR8bMDK6o2pu2wpkRO/ubQgcBfcubbZsZGucWjmy1jeIiMUh+PnD/s6abkSSIe7A6J/NAxt9aRQwSb3ZTdfBz45BA2uaG78ZBHxpC+thLOGTh+4MmleWXDHMgHvr41IZ+vfvWrr51Op4kFruxUs6dtXb7kJS+5zJXDr5yZU3loHsiyz4+x8OMrMKmzK0bY7C1vfetb5xDM1jwKEJYQQgwiQTNAqITgkzEBbhz9+1LOC5r+njA6iF6JIkM+MuakU6KN8S0wtfFsklGS0Wa3gg8PeROsFhs7nXBdZE7IThZkti1yCrsWkSJxqAvOOB+b8NhRwpgtsSGLwSsa/2muHMBn8dDjh11x8wU3W2TymQ+by/Of//y5+eLR89rtK1/5yvxfAAufTraNJ4fPJsJDxZM/4zDKPRkbmAvZaxanROPpqot1jB0/jLOpaBtnW00+PHzEc5DwhID4cqHybxxVa2cDjw25cjCwJl/xildcviY0Lv9urt4dn46L8P+j4nHhWR+PPPLIfN7j5m+d5c8GW57MHaytEbHiZas48ZE5NpaMtld4cL7gBS+Yjdhm/N+SDQ4m/xvCX27lt7lga/szBzAjeapfjPh4NiVPB894xjPm86uL49RORixqMbOrb43bG9w8xISaN1gUZIx884enHw9mG6XXYdYeebL4yHzjxTcGC7zbZn7YNYYcsrztqA+Tcfl2XZlzY+afbwfPZOVLmw+24VEjdXLx9ZOng/aaLl61HFpT5KwPPLl1ze18wnc34p8Na8A16gnEv1s2f+Utf/S1iyGc4Q4rvvyQU1yj9p3eolxvQiXfpAiAYYZymhMyiuD+/ve/X/vLX/4ySS4hgdt6eIgMvk1cgE2AMW1A+dGmAxdfqEXPb5NNFpHp7s0H4kfi98Imp+/Dsy9+8YtzY8Cz4Vgo7GWTDW0b0gMPPDAn0RvHKwX6CD7jGyv84S7+jZ8uef+73COwNhsm10Yp77Ahi4WuG7VXOfJlY+HDpkDvPe95z3yjzIQi8i5wj9y/+c1vxj4eP/LSPOg35+yVMzbIGFcXjxpPTNaHG5NTmkVMl602D/7owkceZrl1wXrqMvd0yBnfOdSusMkn6oIyx3KYP3by04XnQOA9tSckTziwk1fz6QaA9OW2D6jxwiLXilitDW1Y/YvZv/71rxOfjc0NDlmT5UwfTj6Rmh36itzo44uFLr5YjPmA1pqDz6aBTx62jT2s/GoXn6cXrzhggNU88YHYkic64eVTgQcOtYJHBuXf3FmXbnI21eTEIbdwqOXm4rjBWCdu3Hxa4zZmn2WYz9YFHUSPvfp43kb4z5Y//OEP599ot2mSIUtnE54SkSsmsnIkFiQ/nqDlRMmmWPSLH1/cvtrrlY14xAWbGyVKNl/ipdfcsBcusshYeLVhkxe88k6OTT7l02eGZPAQHyh569ReUF6u+tKHJd98ZoNdfbbp45NVtPHpGXMz9dmUdSXW6xQT1EYUlAI3HgA8C9Vi920aOhJvUgRLlnFydADIh4Xjcex1r3vdpe1xePwA8HScDtlhA9FnJxzxyMLHbjgtagvEhcYGv/Dg6SvkJctjssd8G8hb3vKWOfUa54fvYuKn2LSLY+PQvkpwd2FpK3RhsfA64bSwjcHtJgUDPpxs+FCUb09C9DxllGN8xL4bzR/+8IeJy4JDxtlpEcgV4o8OX8YQWX21wgf5CkywO9H86le/Gn3zDJPc2zjgLU4b+Bvf+MZrp2NO3TR97mFzp8MWH/mFob42zOZGzR5efHLJs4PMM7ke9btRyTc+HW1F2+uIn/zkJ/NfELsBy4dYvJp9/etffxm3z028f7b52TzMgzjlCrFHV42vhq8xGPHlkVzt1ri+YsxBwrv1Bx98cJ6Qw2xtyr2+i5asPCK+vHpx6vetIh/25pMePPToNJdhpCsOfHkig0dfDSNsaps5bBfH2sIjj9hujeHLj73ht7/97eTMYckrlQ9/+MPz2gj/3nvvnRMsP3yi7dta8oTwta99bQ5bNsf2ltm0Dj0xwKVmJxv6rWsYxWFeizGePrzk+U6HfYce+bWurQVrwpp68sknrz322GPzL6XlVtziZ4sdVDt7+qj4Nj584+WyPpnmRtx8enPBX3iTwUO+FHDz+JapmzUqr3DCWK7DpW5uyYarXOYrf8m7Br70pS/N0xrc8y9tC4CyQtjkSKY2I02cxALkJGxBASEJFo6CbIYWAWAmQ3HnchF4/CHnFZIkRex2UZT0fJcs9sJa8GwoJazx+HQRPtzk4BCPG6Bxk+S1gRtLdvCdmt71rnfNDdMFRAexDZuC6MBjPCwtDGNyKScmt7ySpV+s2TTh2uaBTTmBxQbm9Y5FYtNzUbkhIxuir+/5EN2/hvX5Dn9izq4a8QlTfT5qFxO9SjHCqTgh2wTMLwzwmk+5ZcvCEyMMbMDqg8mXvexl49/TgfwbYy98M3j8kBenXHNj85EzMl0ofKD05caYTcKpspOimymc8NNpjviUS9/r969BjdFlh67XX9rk6MitJyh1+eFfmxyZ8gkToqsdRnMpBvLm07g45an8siHWri21cbltXbGD50bDFkzIjdo7cp8Z+HaQzca1GUZ6sIQXHw+FHT6242sbkxtz7InmH8cv4Vpz5ZM9dskhcvTly+9MRPz5RqPfwRI7/Orw5Zu8tng9IXrqM6dk2TWm8FdhQztbbMBFnh341LDJvbWSjfyRF5Mbhid45EnOt8he9apXzcHud7/73Xz13LcyPem7vthB4dKGA/GL9I2HSSzwNqf6Wy75YlLL55/+9KfRaR06xPtGoevrYx/72Bw0y1G+ipkvY4p1BY9csI3S098FNrashWJ03f/5z3++Jh9szZ8y4YiTSBCUC1JyER5nnLgpWPCKcYZNQEDpctACu3Gcrr2DNjGB4cfm7a5mgZIFlj1j4eA7LNotbuOw0NOmJ2AEq7jCQJ9ePsgYtyEq8KPssGtybChO9zZnN702ZvbFwS5ZfaQtB4gvBBNsJo0fixhPjQdHMdFhI5tqur6v7gRvQcO1x92cb926NY/WTqDlgU3UApGL8mIMfrJq9mDSVssDLMXADh6sp9NpZIoZnvT4ECNZi91C/+hHPzp2vKKTLxR+vpDaxu7rq34vyId/7MIeDv7Y5Qt+xXgxeWXjpmYT8BXYz3/+82P7bj9sZm5ydM2DE+aLX/zi2bCSh8mrKp+B+Dei8IS3nIqj3NErXzDCSj6ZxvC3vjjIwWI+3GDVCl34FLlnw6ZOB2afTfhquHVqLTm5f/vb3545aJOjxw6dcMEKB8o/ORhgCWPxus59G6hYio086jowzofDji9FsGkNiMXh0bt5sWdXrdBzKBWnPcHNyg3RnmKMP7aQPmKTL4UN9dVY4ApbmPMtl9r45dXThr0NZr/P5GZhf7C2PFW7EWr7YkaY+KUvF9r5y09zmS/jZMVRLMayw64chUtbLox7Fez6gtGhzIEMTnPDBnvZhokNBa/1RIZNxKYSVmPw1SdTnsNtH1TIHHb++TjIUUFwaEwfX9vmwIEFbNzketXgRCEgZAE0iS44jsh6P+0ril5ldOIj74T1ve9979qPf/zjSxt8IAAV+nBk10IKl3FttUBhpY8n4CibLjJ4yXo95Gu1LkIbTzbTEUsnQLG6sOmzZXIq9PDgDK8+LLDCIU8uDDp44RST3CqILF4x4XkXL3e+SSSnqJzA53cDfADIBv980Ndnq4VT/sJNtnKVxwd9ONlqPFz8G1MXI1twdKO1uM/Ozi7zRU6ew8QHe4r8esqz+TlMuGDIhRkWRV6ViK71ZMwrj+9+97sz7ibivTGMxuQAVnGYVzW++YTJzePNb37zvAKig9w8vO/mo4MD39Y6XeOuiXzgsWsOxZo/MaDmppjU7JHjAy628LTJ84tHVk0O2dB9S7BXsOKAR7HJ7Byxo083rHDaiIxV5KgYwmbM3NBD+ccPd3bpkHV9Wat+edBaoCvH6uy4FhS48OD2FO3wQAcOduBkly855RflP5xyTlbJppzQYVs+9dnlz7zR1Re3sec973nzxqH9zBMp8mbCEx49h0lPVw5t2nCUW7U+e2pY5Ka1wF948cMSbjGIE58+f2ziI33XF/vmzpOnQ6V1jtihh+iJEUZ7RvPGdnLa/MlNpM8PnHQ2TjjwyZTL+QzEgNLGzjkjgTHRiMNKyaJnc8RnPB2gnED8kpy7uddBbiIco8AJ0mO3O7r3+xs82+yYAPb5zG828pkeWQlmn+3kyUm6DwJf85rXzOO0GwJMsCGJTZ6+d8velf/85z+f31sg03iTLplIX/7gIJMtY2TEIo/FAg8fZHeM5OWwsZe+9KWzSGxwLYQwiO+J4xtl3/rWt6hdbhzG6csZW+TkXV7Czb9xfZgRjAhGeuzk62ofn1767Nn0fV7gq7Lk3fi6CMfw8SP/8CF2XIy+UukQYR3AidjUDjcdvHzyrxSnJxiP9fpiMB/s07ep6lsDiA3z0eYAt7XAHh3kFdj3v//9eaKxYZPf4+Wg+VUrYmdfm7y+Gq81HQ9WcVWyKU6HAzVdtoyxYYOzLjyN2vTc0Lzu85Ts1Z/5psMmPT74pauNipGcvJDFS04bxuSq6ZKlxxZf1i8/8mzz9zRkYzP/5KJixPM1bU+ZDj/02Pf3pmz2bkLmi312o9YnO61nY/zjkYVZzhT67Grjo13DYZxd8+/zDk9I2dh2nfDZs1nDy78bifWKwkrXODxyg68PAz9qpOYfZhjo6St0mwe6rRu59Xse8krea3fXV3kpfva1+bY3W+ONsc2e64BtduCEIV/GER55mMixp2294Yvh8ltYGSOQAwoFliFyDLvoBWRjswHjIRcuWRebC9I3GDo157SkOd15l2YR2XxKsHGFvAIPXPAAnS/88KnJsqEmX0L0jTuVwuR05MKLjCM6EXn6Sq/p+JVsY9rhgBXBw79xtmA1gWyEgZzckKMnpvLNj8kxuWqvAXzhwGbhZteiCyebdE+n0ywqG4EYH3/88flGF3vlDG6yCtx0Fe2wwKmvRvioudjtsJQLm7ebs9cUNpAoOX7JssUHvr6b9K9//etr3/nOd+bbHT2h8C1O+NXhpc9WxA57ZNpUuvDJ4OevfMmTm9yN47UqHXj1UTmhx7ZXGF51ZUftCck7aW1ULHSRuGDEh7d8ZltNBrGhbJ108dgwTkfbCdQByEbitabYvGKxmcmhb0e63ugqzZ086rce2LPG9JOxLvNXbHTyr43IlFN1fHm3Tu0J1i4fO9b6bHhV6c2D9/vIDcMrIvGxTza/2oiv/JGJVx2OYm3Nm2MxomyTQeXauvWKna+L44sCXqsXJzl+3Tz8HoQvLMgdm26CYshuGFqj5RYW+dFXkrPOkxWv9o6dHPzkFDc4ryzDb3/xbVhz7gsfrTX4vM4+Pz+fD/13DvngK8xiK0ZyfCr45LTDqI1gEdPTWTwYOzDtAqHASEUS8JyCvPrxSyU2DvL4bULk3EzcZBAwnGbbzcP72re97W1zkUoAeRRYsiVKW3BKgcNkMzDWZKsFJuke/xEsiHxxDeP4UfKMIbYQPxa0D9Lg+sIXvjCyndZKNH026an5bwGzw+7OIV746ciX8TCy7wTmptwmgYfKxXSOH/hnZ2ezabMBpw8f2bKo+fa5g1OdzYF+8xMmtSIeeOhoo7Br49WHXwm7uOX8KsHBNrv8a9tg9PPnw17f2nHj6yaZLXLWDFm++IEhjPpsqlEy+toKgrX8yq0bgtOmV2z4LraIPXni00HDRohXYdMNz7fRfFDvIJU/eJEaRrb5pqtvrWizob/zSI8dfslkg5wiDzYw771dc16/0vfk4YPdnkL7kJs9/thzIBGT9ZEP7eIOpzG+4KPHpz4Ka3nAM84HrMb1zZ259PTU6zey7CmRA6MTtP2Bf2PslKt0kjeeL5jomEtxiM0Y/AiPLXjouU7odHAjz48xOmS96lHg90vDnjwRP/4iAl8f//jH52/9OVTY82zk5H2jsydbtssxfePhkld+w1vO4IVPDMaTx4PNuFgc0mAON3m/6/HVr351DkBeGVoX9FxLnqq8/XGQ9KbCPsCWeWcD8WWvhEWb//hqmPlW0hNr8v9zXCS3nnrqqTHIeAaAUBBhfH13PEC8L1Z8X93pAUgLxmRxJAA1PaXkSIjXDF//+tfnxMQmHj2BSDhZJEhjTbggYEGCpVsiyGnzxZbAybCHLzaBO+H4WqGL0eIgo7CrhDNbfF8cJxIbhnhKKH4LIhzlKJlswJYM7PqI33TKUXhchJ7MxOLDUvjDGE528OVa/sk6wbvxedJyUnLCdypF5OVQTvhD6vCIPczG+CGPyCjG8dUofbpuek7yTmlybYxccZJnA79izdj0rCvrAjbrALHZuuCTTXjY3PnXD1/5Y7+1pdZvXqwBr1f7EJ0++7uwQ15u5djcw6oNr/VjQ/Jhb7LswIy08dFev2HF167WVvYcwI3g8hrCb3f7wNSTk3ickh966KF5+pCb/LGjLXdw05e3Nl25gik54/KujsICj8IeHj36uy9PiD+4vNbxBQpfVDDGbuuhXFgr1rV1Ttact9aKW62EFY49BgdMkfHmnV9tccFqrLXVtYSP2PDBuL++7VWqbwHCBTPsXVcOZT7od/OzhmCh59Vha4RP9mATE7/yjlc8/LKLRx41Np3jh5j5tebI2nPF5FBj/tnwy6I+83OQ8frPDdCNrdd/5GFVXF9++9564Ct8avbLkTEEV21YkFjIK9rG5xUWR5IlwYJCDJhQgpLDCEfJeeTzFbuSRJ7u1h9Dd35wioyzY5Iujo35dDqN7fyp4UHaSgnXhkMNE2LPBOHzYawa30XOHp6LUCIfP17xOFWIt6SQgcvmK7Y+wDPOBj9k+IGHPTxtJJlk84WnDc8k+tCTu+2vuNgUD3swIdg9oeE75TpJ4GXDBuaR242DnoLg8piLzB8f5shXVsPNnwILmRaKfsSvUo7oFgsfdPLLlpOkG57TudOaXCr0LFo3NOtFP13+vMP1Ssaih7OvCDePfJGDI/wwssE+DGRgCE9zQJ6McaQtX3B65dPXhOUg+2JO1mtOmyBMnpDdkNnS9pqmfPnFTafD/MEBD1k4xaLgkVG04azPFlk8pE+n2OXXu+++xUiGLnzWrCd6G4QbWv7gKH48ceIh8fInXnkvd+SQGn769BBZ+FoT9MRRHrSRzzRtePrWaYdOTxxyiG89ezXrSUWe1daoV4M2TthgYhsGbTwxIzLINYEHG7tk01GLoxzTpycG15ExPDoOL33+Vpz05EBfbY79eR1969VNkL78WOPyAQ97zT2M5UUM+dTG54O8whZsxiL29v7jSYKsOfD7WL4dyC9sfulS298h649ywiXnXhnKMf8ojPphwq+vhgM+BJc2zOI3zsZ8jVewBiRCu0H9bVCALkCPSRLuAtQ2GRaJoOhwzI4JZNdi6VUWMGzaoJsYQdK16NIlUzLZR2xFxptgcvzyp/Roy75CD3Z3c6cd7175op8femLyy1geUeFrIumbGD7U4YCZfn32JJUef5tvDD/MxuiSV/CThwVuNzybl28nZdNislj8TSryTvtqp2J1mOVJXryqcQLBbxHhR+ZJHxZ5tmCz0YI2hmBUkDFySjmwMXud4lTEJtvwmn+vKt/xjncMFnyUXbi9riPr8wUXcmujNUW++YQBX97YCD/sSnMEI1n5NHeKNUDH6c0XJGAj4yZtnL6NwUWH7yTqF6dOx0HnAx/4wGxA1qt8e5J1UfrNXE95bIuFPz7E2Zywhcr9xk6OXzKtG/pya8ymZ9z1Zu5dg8iNw+8pyYvTMV/+TA4ceJ4+OygZKy9sZZsfPivhhQ+Pf7iaa3lH4TOGF3ZjfFsL8iZX/LneXFswiMd+4BqzPvkQm/Xnl/WM8y2P6nzSxSNnrbCtmDcE08ZfbMWaTfLG+EX64ZK7YlXLA3/4MJs/3xTzobonQlhdp/jkyar5khOxG1OM8cVv/tXk8YyFBV+bjjFPwmL/xS9+Ma/XtM2/eMnJOTuf/exnB6uDRetEDApSw0Lf2oWRL3b4wWcH1SaD8PlKR3++hYWhI4mM6UsAgxmr3SR+7nOfu3ZxPEH43rRTj03Z4pZoSRM4gAKxyG0QbCtsk3HTEZBaP1/AFhB+AaZfHW7YkODYiw+rxFpsbiom24LjRxKNFSd9E1Kc2QuvPj2LWU2PvnF9fmGWdDx2GhNDGKvDqB9mNuls0ueHTD7cELx29DgtDqfh9773vXMzD9u24XWNxe/3GXzoh8jBG3Z+2AoDvnzmHz89Y/iwa4sZkSnnxuQamV/fsPJhr9+Y9Xd06GWz9SY+/GIVNxtwybuSX/bhR/ybY3Vzop1Mi54/vqzN4qXXJsAW3075bng+P/rMZz4zm7ON2GsCWHx2It9y2uvbcsmvgvCKRRseY3zDUi1mcbYmyg2c2nz6bMt7edda//So+MXnlYrrzDXHj/L48aTtFyZ9yMqO3GSbbuuKPoKnudCvzb8xNhEbKB5beOTF1xrQLtee2ny24BddPS35fQpfQbb5InPghmJt05EfdvGR6zc8cPAXbm28/DfP9MSYLXa78ciXddmhlVxxaLNlveFlW62PwqPNDrL3GS+PrWd9GBQ8RI4P1Hi++EXygWDm2w1BrHJGtjiTx0PyTi5i35h5NMavNl5rjK2IjFLc9LXVck6Wvvp64NUKxwxrI4a0t1GABeVkjO/G4T2cWnAM2zQVJzu/iq++eee3qPlgw+nPAjEZeN2ctOmqAedDH2jygpYEOI0bg7HA3K3J5sekFbhJ5oeeQqb4vENkq0nGN05Xuwnknw0+xEEHFrIV8uUvP5PwO7kVEyJnfOdcW3z8NqZW8BC/NjV9/tn2tc6+SbL9G7MpOjUXG542Ob7gEZ9+9rST2W26qBjooxaVdnnQtjmy6RWFi9PTE5xsksseWTG2DsRoTNFG+YSLX2O7nS1ye4xdfT71ICyNAABAAElEQVSNKfHYlUfYfCPL47/cejVlwyZrk7CJK17FsYO8c/YKThy///3vB2dxlTtjiH9+jPPFrryUA3i0FWNqMebLE7L39Gw0V3LuWvR5gtN8v1CWP7VTq6dPtvi1vsKmL7fGFH7xyh0McCkoXDCRaV7MWePZHsbxA0Y2Pf3D7vrxi578+RPl3mKIx5dy3AB9DuGw4Vrnh0/ELuzswaXPBhxk1OHqetInZ1yNjMECOx4Z9pB2uaiGHRWrdnL0YW5Tx4eDbusrfOnBovBJXtEmR4dNhQ2kbn3CYgyJg1667CBzwVZkP/RlkPvuu29+JcE6gpdc9uBB215zzrc2+2p+1XTmFVZKElSwQFNoUwCaXME7UdhwvUtmDJjumPRMvkLeyal3yGqA8Nmmq6bDt7Hw6OMjbXy4BJ2N6nyR1UYurPy06Lqb85v9JsQ7Zl+F6+RMpwtcbuiKky4c5UNC6XRjwydT7abFByz4YlLwFPEUU4vUeDbEwq/CbyW/Xhc43VkonkYQXwoM4kRssqHmR2xilC824TemyCuCQUHNAZsoWe18GBMLki8yfJYf68UrIu/DySpkkHgUtpo34+EyRhbW8pkunTDgJVsOsku3cTzjbMmd+fdEDZ/XbV5N+YymeaBn/bqJ+Pq5L5DYuN0QXQueUIpHfuVQ7HK8MYdFzPgwwA5HccSjy2+5cvF7120c38br5ubLE3Lc30pzI3Fzs1HA69UWX+yzBZu+ucdr3Zm7/MGJ+KKD1GELA4z41hOix37xtIbgIyvXnoo8QXsN2JO/XMo3ueyUHzbDYAxGuCNtMmFtDcIEBzsIFgdWPDL06CjJFIfxfZ3ky2c2cmtu2XEw67qCMVzZpYfPPtz11XgwqcOiT9e6ly+FH2sMJn3jfKJ9rcqlvPosiZw9We0zSL59ZuKbW9YFm3gV/vEUuQwr/vZRnozP/wMp6BIiaYyUiEAGuslhVLLpmXR1yZcENjjzQZqFiodKkA2G/AC5s+gKIn0+tMMCQ4S3sZd0Gyp/vU4TD2zG4VHw9CUKjjYNv91JD8FFToGDXhMoNjw1fMWghkmdfTUeXyi9xssH++niITIKjMYsBpsG8lTHD0xykQ45OgifXY/qcl0MxrccO/T3RUhf3oyJk7yCyCWbb3LGxce+vMEbGbd5hB8WPAWlCyc98+AiQnwkN4zjBxk8vhrXLg/G8q8dTn7I6LMvtzayD37wg3NT+OMf/3iJ0xi82ZcPm7FXW/fff//8omy2yw3bCj6CqZLfcgQfHl35QMbolnu+s60th2pyCjmboq9zwutm4q9Iq91c8HaO+LNmuj5du/ptrDCEiQyCjU92zDudYjBOHt/cyak2XOKja/117Rin71UmGXa9vfABta/EWtdi3Bj5KFf8Ij7gU+SLH3mCkYw+PHzxUSzGbazhYgfPeqXfTRUv/M2xV2zve9/75oNqPvDxEF3+FHz6eNYxm/yInR8kPuNwldvGyOIVM7384SH2mzM8fXPub4j5ht6nP/3p+YICO8nzKedK+bU+yMBSnHKjRPC0JvhRyM4NJOeECyRFRhXG1Dlq4jJmjBMkCcYBM3lOaH4z1ekCcSypdJpodpUNOpt4/AhaEHTYlwC2yE0wd5LuQ1ivFfz5AW2/YGWS6SO2UIk04Xy7K1u4bCMnN7+j4Cu83pnSawwmPukhmIzjw1hO8cjAKC/aeHTDTie+Md/f9ljvlYhXFnCRNca/RZNf/Wyxg/SbC31fNfQZhN/+JROOcLUw9Y3DasHnMx5f+WNXnHT4yrc+vS4Off748OUE68C3RPDLA1tRuvLXxaatIDra+YRHv3UBOyz8xaeDT6YL2dpM1/zD6+Ij5xTn73f5kN034Yzzp3hd4RWXJ5W+wMBfeShWPsmzV3vnMx49pE52GMcPffbKSTEaN6YvRtjFe3F8JukJ5J577pmvKLvReTJ94vjmTvLlgj+4la4LPL4U2OWH3XTx+UyXLePmqU2STBsobPpyjccXW07Srs39ewswnJ2dTbz+GKRXWOT5CCdffCvwIW2YENnGWn98o+Jhy5Mj/77F1DWBz062+IKZPZjdiD01+2zM61f7hIOqMdcjPfGSD1P44Wq9kTeOjGvLEWJDH19B9MLOPv0+t5Gz7LFhvny+5DM68dlvsz3Gjh/N6cbAFzvGavMFt4K2PDywWptzA5GsiCBgFPFdMDZf7yd9aPiP452whSlpnJEDkmxgLShjLkgfnvs7Tr4aiwClIwlAmmhJwgcMKASgfnbJhSnwMFi44eUTBl8P9Ke9fSYAqw9DLRTvPZukcLDpEdSi8BRSPDCI3Z8x8c0Qm0pJVvMTJonHQzAg4y2GfImF/TCIjwx9ixXdPD4ncmG5abiIfJPGYhUneXJKuWDbCdlF4UQXhQcONvyJC7Wc01d3s4SL7YhOeWiOvRJxU5Nr44pvzvgwn66+ufs/vu6cxdbi2+P4udAvZJ9MRNFAjJuTGJgIJprIcTqKgjghTggKehREHHHk2HqcEAQNTDTZgdzU1MCg38m9z6fsr3ex6f9dUF1Vq9bwW6vqqWfYT+9dbOKCwYnXRgu//4x1MWFh8wkDKmcuNtihxw4+O4gPmFof+sWoZs8YfTpq8uEvz3OcXf1s1/eYymch5oUN82zt82HNuVuxpuXd1bL14a6FfAQ/XXVxGpv82uUNFrLwNM/pFKtxcog/8ZkT60V+/VY97E6U5ny/fZDucYa55qd4sqtuc5t2Z97LdfOilgf2XBR49OdlAo/XXPSU9z5ncLXvv+ZhNEbXNywcbycM6wKZezZaA2S0xQYLvXIgb8h84OuHXR12Y/KlTo4NL/b47IU/NjzWoWc+1Xy3J5kH+5j/b7J2HZ/isH95FGT9d3e1QG1/zIm5YUvhG/GtjVcsan06tYtPX+4RTOTso/Liwlh+xVrO7FdOat4g9A0gHglG3hK0BzpmrW97tNxG4VKzWc714WgcNgUPHUkw8KgFBjQDALta8M9LXsGzGThYrl+/vq5q5mZSUhhmB1CPg7xp4XlhwAR5ul0p1Q8IDC2ImVC4YCmpbCMYyTm41eyZSBPv642duCxcB7uDx12IN0AshrCaZMVihtNGp8+XBDtZsmuhdRDBqbCJRx4+2PHLW5iMk4vkmwy7qPjdwlvUbo8tEu/Su5pAYm6ytdm2QYhP/D4gc9VpA0FskkHyYS7Cyb82/wos8NWmU67xyMNmk/Bf+U5Sxi1om5PXbh148sEXfOzLh4Xtu9BctV3c/vkJicNBCZ985YuutUZPGy4xGtdXyiO98sYmnPqt5XLOBnvlgx475NkKg9p8y1PP49k93jY4V51yHE662m18vtq6Hz2iY6ycwsFnc6EPi3FkLBl9+MkmY3yS3JFHYZJLV8cu8hyj8EbnbcB8iIFtWPljEw+u8ig/rTWxkmte2JfXPri3LhwjruitB3fu9K0Ja8d6fu6559Ya4DcfYkD6ntt72caH6/aIcMLIVvNIB5bs0Ie9nGkjMoqcNa49Y3S82dO8Ueo1aLnkTyzySdZr3D4X9V/oPvsQk7z4fydvnrq7YxdGlK+Jw5h4EJxw4bGvkG1tqJsXsSpiNQdsXLlyZR0X165dW77Jsln8To7w84HkzeMsr63D6wRjnF+42XYsIzr5ziY+v/DikYcx+ytqAw5WiSkBBG0M3it2JS+hgnGl6zVCJxUECFnEeQlhh44DUptDC8PzOQvFZu9MXnIdwCYQZY8tPunDNoPGQ7CzXVDuJixCfScG/iXVj9g4oZFHJZ6uReRzGmP4sOy3zbGEWzTswUoeDmRy4DAWnsbwUAuncfphIMOX3PsfA3g7CZATO2qC21AdlD4sF4+N2Ubhv6rJZzu9fjjJgQCD3KrJKdrFpK+thJcdd3BsuwpXjMHi5ODVYDaN05MTfbbkwgm7DxzzJceInLYrKTj3W867mrMWsiOH2WaXjtzhNSdsK+fFAIcxlDzbsMJtfk+3ixoHpatMd7B84Dtw2UT8Ibbyb25au21QYSFbDDCw05g2Hpvi02+N09M3VoGVfHXHDfuOKzjgJU8OtQbnusBjg5wY1Y2HLxv45oG8+bbO+GLDOvVN1q50HcdsydvVq1fX77+I02vFrnbZsReIU47YLz44vXzgdWOPC9m379DvZM+2mMSqFjsiE+ZiwdMmmx4ZmBUYWk9subN48cUX17HuAlOsXtF2F5cNeBxnaseRV+E9GoRbbOyxxTcd/mAMG56xsCVbLFNObhSkpmtunbx8hY2vsnHXI0+w8m39udj1lMCbVrA77pAXQXy2JLew89WFhfWm5FO+FDjx4ONf3iZWPH20HmERBgYxAJS+4tGOK/OCcfZyNQpIVJsMqt8E4gnY63k+T/A6p4Uk+cBKtgXaCYR8BCx7ijYf5AUAHx1j7EiMsd9++21dHXkOiCxwJz4y5xGb7KlN1n7byHyY527LolEs5hJXkiU2my3Kks0WXGTx1EoxGNd29cinW3wYuyqwGVgIFgUy0Xzgw7Lb7dZnJeaDH/hhKRaPExwQnie7KpRffuiSl6ewsR+2cslWMfENa2SMnEd+XsX+T0QGnnJRu8WntvC9meVRoVdnxWNdyBmMagUVG/9hUJdb/hT6ZOGm21ozFtFBZKwbeXFVSU/uYLO5zMeC6Rr3ii8djwbMoRONHCnWpZoNfvidMZQH9siRKdYwkhEH0qbPH1l2yZlPj9A8svDI1h0pO2y6ULNpuKsKh5otOVOzASub+igZNozr26AQPHguYMy9bzywPtmg72Thjp88GY+9HTvGyMCWH/bw/KOsfaEfEOtKmx9F3GHJDl3tcLce5lowTp8PtbFqj7XtQ74WxN2F/5Pi5+J2kqDnMwR7XOsof+ba3uJq3ne4sUlGjthW6CvxwhEfZkS2vIkvOW26yN6mbd5d5MqntaoNo1w7RtlyEebYcUIXUzasD4/aPPIyP61Dvtnnt3UmHuPpmi9txwh52FB8/a39f5sxYWByou/9cYvEc+vGOUL6BV57DZz9EbiDzUJ3G+WKnj2BAqWgeVCwIyA+AISn2lhtfiUvXX3Epk3WQe3xgoVsEcNvI/n/yCMU/9viDskzbVfbfEpexD9sfMtdCS8ncODDreC3aNjQNl4csOF5rdGi3W0nBn2LxHv95FB+6Rar9rRNDh6PA2we4hCTkwc5BzZM04Y46IhRrMZgxuOHLoyN8YGMk8NXEIzhZLeFuQbHH7fRPt8xH/Lt7sP64AsOfAUWixxu/sIQRnMNY7kXW7nRLnf0yhP7xvTxk9dnz4bmMwM2HaT+2xixRUZMLoD8doU1Rs7FUcSe2NVhoKeg6dM4fbbJKzNOduAtjk4g4ueTvBPcAw88sNYKbORdoboAchJhHz88fE4MMImbTRiNk6WnraAwwKS4K3Vh4uLScUKOjjGbmLb5jNi3Fn0OZ0zfxYOXVOQTbjw1/0oEXzpkGlPzC5uYGsOHA7mjnX2y9gV7El+Xts80nDjcPblYRnTJOY7kWV9uXGC4mhe39QkXO8b5L0flt5zokxGDdjFqt4lrw4/gJUsfjy/5tie5yPS5k9fMjU+iR5YtZO/zaE4M5gIGMnDTJZcf8XVM09WvkKWHyCN92Lbyz8JRC4wARWc65OB2oPhg11UHReORvsIJYDM53v7xA0HOjhLtdsrJw4bkdoosnWywyY7CByzKnJgOsHTCz04JcFV8uj2S8FmNpFkgbrc9SmEbkc0PX2L0xhO8Njd2HXjsdgXGFiwlkR0ThhzcLQJ26bDLDmJHXxF3vt0dwSLPbHgs5ZEbP/wp5xF9hS2LkDxfNmdvXPk/BYsO1nJGRt7ghIP+jMUYKufVDkI24EN0jdGP6OKzja8PV5sePh4M8uzuyBpzVWeujMmZeMXBX2spu3QjMmyiZMnR5x8PJQcvH0jbuDFzbG3Ciedgg5stJ2Frlh++1XRsfA5Mc4qMsWmcf/3muFzgs4nIFgt7+gg+Mo2zF2bybMHmKt3m5nGGRxbuBDr52uRcXbt7Jo/oZovtQz/ksm+MDHn82mrzg8e/E77HTvB447FHJtmmb92gcLjTcMHQseDq2Mla/h0H5LTNgzwo7JWjsBWDmo5xxThduY+SsdbIkMdzYetRt89rLl++vPYkY2wo1qSThSt3c6eI2Xpo3rOtrtBFavYQXfmYuBqTd+P6xlsj+vBan8Z7m9Rnte5Euijgd/o0H3hOHr4bzZ29GJG1DQd7/JhPBYbwk+NbHtktjvBW47N1xBCDFADOkEHFYwX/yWrcK5j9NgJHkywGV2/O7gEUtNs9Y/P22sICBAgTy7ZAYAm8jbkJj19AcEWCV8LvBGUifADNr7are8lzJ8Um+RYB24rEe6bpLO95rQOSzeTyRxaO+GEsX8bEQA5PW075lBdtmLSNk0f9449x/uXLBuHNp/9EbPpSSldwDmq2nJgdpE6CMFho4gjTtFWu6YWbzYmdjBOvHJpL9vq8Zdo6bIsRDvjMpXjlir63QTy2wveZlTnrJMV3Jx2yYWl+4Sn38idfzYGx5gEeMZXjZPGsOXhg5IuN5PlzgYMnh+KGA2VDHWnzC4M2+9njwxg/jWuLBRkjIw/yjp9tdmBBxqwHrw7bPKxt69UFnTUCn7XuMZC8Ookgdl0YyRc/5bKaXVjzqc0vPTbphdEYebIdG95eIgNLr40ux9sfG7T1Yl7FTt9acGIjz0fx0XH8k+EXBr7Y5g92deuYPBl2jeGHDR+lay6Ns42HxKTvJIxHh9/803Hx5ULHfgZrvhxnZM1jOaV36L+5LHbj5CIY8Pgmi7TZbIwuHjl4zK+vkCeDj5o7bXvcfnv87gUcbY/ZejuQb/mGnV02kTZ/yPjs04HNeHOlrRTz+jZegxiSyTABADlTXG2dnJwsJ54XSihZci0GAXo+uN8C8AhJYBaPYC10ffLsN2lNAJAwOLAtpGzjKYKCjz4yefhsauOHFW79xvFhcya2MNjijz2ybCjabNm89cnA0kI59KvPBjlxoHJHHw8ONb5CNhlj2kjbRupKAU45dbJzNeaxVvpiEo8+XGpXdR6neOSlD7NYkD4s9KY//GSqjR/mBF7jNgwHm8cNyAd65oqdOS9rcPtD3ty7woPNIw55h8PJAsY+jKcDoyI2/sSmpsNH2IznT42MkaUf8UNXHtkqLnLIGIzZ0mZHQfTI6osTNZ9yUj7ZIVOOYaWLYMUnX02WXQW//LI3+ezgqcOvD4OXQrxhSNe4TZyMR0M+2PWIuDwZn/iKCa/YYdNPrjZdJUpOny484veUobs3/x+jbX/A98Wa1rT8wtTabBMuD453JzoyzYWYYAlzWPSNqRU09wt9WJVyBqd88T9zzp+7SY/72KfDP3nErhM3f8i4OVDDxr5CV06SUYdXm0/j8JKnm3z7A5nWmHHEBnk4YJA38+tuD4bmgV05d4fkYt8FJT3jbGVvGd3+zH7j7LFTv9yqW+fsaZOFF77/eu+99/7nZDs5GDT5FEoKwYAa8zmCZ55zoxdcJxQnGo9Q2GCcM5OE2Nc2RifA+GSNsevMaeM0jiepZOhkR59Petr8ODkInhweP0ig+PrqiUsbj74JosdfpE8GDxabg778lEBj9NnJfuN8GlOzTwfpV+ggOmIOk8XiNckeDeCLGR8OOWJbzj0Gkju22MFH/LX448kP32T5g59t/KnPTpiLgZwNTJEPfTmix2Z5h1HbJuJDXCcLsk4qDlAFNj6bZ3izwy97/LJTXvAQ3dqNr4GzP3jiUstVc0SnGGEwbl2zj7LJfnMIN53ypE1emXGcuV45aw7Kjz69iVVbMabUz1/2YOKbjLtRdxweJ9vYzL3X0z2uEIcrT3LkrY+OBbbKZfbZNYfiaJ7zyUY5KV/6MEbsid+4jdmFl3Uhd8bsA33+qd/nJOaDfXKwGFMr+Z19vPowhwGfb3bUinE4lfYHfG1+5KO1JQ7yiA1jbLa3aTv2yNuTyhd7KEzh0CejaNODg186cNcnU/xyiDqeySr5czEuZ+aXLxeK9oX84PGh9vTBSYQsXmsq/2pz7iIOFnetbOeTDqKHF15teIuBTPGut7AMKgwoBtVAWYwF42rCIy2LQeCcu3JAAiVrI5/A6c4CDOIDkTVpwHaFYML1px2y8ebCx+O3g4UOmsloMsmxjWCiC0fJqG+MbDFms/H6M7HkjRerNjm44I3kNN9NCqxk4msrNl+PUOBvLsi2oMRCrnzlg188+FqYYhEnnyh85Yb94lLrK3TIOimw527OxpC++GCCvViNKXj8skOXnNoYfPwoydKHjx7SNobwYGFDrlGy+uwaV6Pa6esjfX7KKRvG2GDbeHhhCzceDOTDzQYyFlY2jOsj9vRr86VfHMWKxxdb2QgXG3xZC07C5t9nBh6x9ZPQdJpzx1G5xOeD7XzAoq/woZDjOxltc4fYQ3DQqYYpPY+r4XNis+GSaU/gn6w6+/CxBTMbxuFAePT18emgdMkr9dfg9qdcsiv/xRdectrpw0DH3LcexS23/NqI2XCMRcaVcqzd2skv3PywjQdn2MwdnQjO/OORLa7i558OXXdM9mAxlMPiJGMvbR2IwRjZsMA19Rrji74+vXRhMUbfWDGqVy41lJKkTdigwLQ7GIBT8DkLaAkr+JzhI33UpOrntzGyJoUNtpMp2MAbj8g0UeQQOTwlWfaSIwNHY2pjJU/fRLHDPlzFYSy87NDBy6ekh1utsI20i2sxtj9hkgu6+hU8+fVCgLEwqztp5MN8wJwNc0S/2NicfXxY6Mlb2MiEsZhhEB+d9NhnkyzCJ+8kk+y0axzZkGBUxBD+sOnnhyx8+mIuL2qUjnbj9IsBn24+4EpHjdTskYM/29rlgZzxZItZDHOdkCtn2YKfXvzsh0nfuIKXbXztxvmCQc7cbXhUhWe8Y9I4knf+2BOzOh/FwQ6evE0im09tRI4ePnt09I2LP5vmHrVeGqMDqwtN+WAHqemGVc0Xymd5Y0M7vOnIkUKevWyXL3rGySvGyWobkweyymGewqYmq/CfL/LpkNFWk9EWK994YSt3YiSD+I7gmvLZj69mV47JzXmca40fsvzmR138LirpFxecyJwpERkF8ZUtdsK0PkQ3wAjHAmJYorQDAyC5irOzMXp4QNHhiD4Hxo0F3HhEjxwfnTiMkQc6P+yUDPxwGecLjw18hU92C5xNvhTj+PDoayvpauc3m3xH2sbTZyNsZLKnTQblV5tuvvDl1AGv1CcjBjyy6YST3QqbSD+5Yiz35airyDCzl0864sALtzabZLTlo1w0v9lQR/EschjYZgMO+hVj2uwqiN05d1M2XDDNWMtR/vWVMJFFajbwtfmBC5HvwDEnYSOn0IEl/Q5WuOklly3jKAz0wh8vWf3mCh5Y9BtnG8GbHTzroxoOY2r60xfc5PDU2W5eySP2jOGTpeduR42yyYaC+DMuXkW/fSA564BNdyViJdcYf/yybUxhb/oiD6Ox9PKPr22sONT0ZxyTx2c5YDsM5OWY/3mssF2ujZHHU/hhO8x42SkG40ryfCB185WNcDXOlsI+DOqoXLBbPozxG0bxGQ+fcf0w6SPj+aDbHIUbVj7Sq0/2KIMBbFFMBaA4wCNPRq1fEhlH+Aqig3KYLD6e5CBA6OsXvD4ZxA/C01bYSB+esOBPOT7DvIxsf+glX2x0tMPFTjFKqILCWix4ybGbHTztam0U7vrFsAa3P2JpwdIlD6sNWI3wEd3wGGeTrpO7voMX2eTYaUHRZwsP5SP57KvZFytS4yWvzTa/5SebZPNRrMbEx0/5Y4OcmpwranF1UVHu6ZKZc10+wtdYOTWund4KYvuDx57YyKjJ4LXm2HIXpI8PGx5Z+vnAozspPHjs06WHtOuXI33tfJVTuuYRP5vlQQ6NtWEXk5oeOXX+6YeZP4RXDub8iUefHDv5JpsdY2wjbbbphb270cbk0nhzlG0xKObbGOIDj3065V+/GCafTvbI0FfTmzbJRWJRsi2fdOBwDJVHdXZm7HT5xFPgQa1XfXbIKIj9SrbwyZKha1yb/fjdjYsFFiQ/xQAjHTW7bBiLinGuI7Ko+coXu9nLV5jIoPyydyRIjAKTSP2EJygGMxrQ5OghgSiIzKy12SNrbNoqIDgAg2FOfgHmNz9qY/Tyx64+YotMtuaBkS81fyi76ePRxS8Xajwl+3zik5v+6YfPmAmaOaZT7sJBrljyGZ5s8VG7MViQPhvhwzfP+nTYJFPbuHZEV0G1y/vk5SMscisWfPnMj4NSO918kaVDtvwbY0+/uOjVNoY62MiT5dNYfsjPYlw/TPTCSU8OjMtRREbcYZk5IKvvhGdcjGo8dtUKyg+d5ohtRAam9PDJ6GvToT+x8FPekiFPNnlx2HgUFxT4ybCH4uWP3fwYwyerDaM8GUf4MOCHT5u/uXHaoJInl2x2ipttxfHBL58KwqenXz1t4mcvzGJJxhgZvthSw2rO4pNRGk8mn8vY9qc+HGRmDjq+skUHnuKRF/ZR+mEjwx775Y+s/LXWYeZDTU4pXvaya1xBbOYrveTYl6diSr6+OiLLDn909NdcNWEBK1iOUyqw+owKvDrw5CYPAAW/yck+W3waKwl45CTM5nI4wcv49ocMPfItFGPhC1v+S0Q4S266E0OLAA8Otsg1CbC28XX1x7dxssYnzdjw9fknH/6JL2xskZlj+ihf6q7YLTRz2f/ByJ2FKI5s0J348Oc4PEjNP3tkxE+uYkzRn/aW8tCnC2PzQl7u6LApVryZc3zycm+MH7Hkpxzwxb5CRk2mfjlqLGxqPESWjwoePSXfxZ5tOslVkzWeXfrkKuTItBGwmR2xpp9v8umSxWe/nJBH5Y1fPCU9PCU98uyUc236/M95pk/HOCKfLWN81A8HObaMkzfHZNjIT7GTMUYONd9sWct8a+MfUn7JZIeMNuJLPPr5xucLf9rE4ycc1hgdPHU4soevHfZyRDYfan3+FTriJksPT50t8vHhoKsYT4eMYxuPPr5avskWGzvZIiuXqJizXzzG6E97xvCsB7bCwh99tvhlX197/SKhDpApMcAwCljJY4xhpI1fm0GUA22y5NgMYCcG43gtHBtyNslLPv+oBMw+PxNLmAXKzixskVXYoKsmS08/3/lQkzdGBlYFn2xJVeuL3xtq3kTzzDfbxsTMDv0WhLgm1jDg01H4pxMvn/rlJFz0xclXvrVhhwl1dbM625/mjC1yYtNG5WF1tj/sI/ywqm0YjZUfMmzD0xtcfPOh5Jc8rKj5oSP2bBmHCZ8ufvPCb77ZoMd3eZt5ZIMuojPjID9l+cmHNnmY86+GS0zlLp38h4tdMuwZy0/xdbzhI34UffLZn3jC7tgpl3hInhT2w2xM31yx12eYMPJT/MvA9kd/EhmyxWCMPYU9lA13IdrG6MkLnny1z+CxR0YdLnZgTl+NyBSffnbLQzGwh6cPF738dEGF35pLj02Y4AiTmhzs8gm7vjYqh9p8NI9k6PLDvnjwiktMMDYeZvrGJvb6xeTNNrbslcg437XZYodtOo2T4X8eg2Qi40r5UDdHZLTZ41vb+LLlT+AMagMAiDEAkORRMsYRPlkUEPr1ybChcIjSMRavza5kGuMXZR8P0T8sjRkPBz18eKP09OEMi77E0g2D2njxaLNncbEpF96h9s61fPi9cYsqOTazDwt7DvSSnr/GwkknXTU+nhyiJo49PJjSMeZDT7hgSr48t6GIqZyFMf9qpTzAR6YDRruSDX7SM0aX/+7OHHz6cnb4oSy7fORTG7GtXdxsovzgi0NsSrJhqyaP9NlkJx/08IxZb2Sbw2LgI9vTpzZ9tcKOEj565hjhk+VHDtTNTz4b58t4GzAbeHCRzSZevozxwTYssOOxmb51G6Ywq/FhIa8vD+TQYc75C59xPuuHnw08ffLZmu30ksuXuOmj7GQ/33QV47XJa8PLhiKO8qDPDhm48NmTG2PayBjCKx7402dDSYaeQoZN9hG7/Osb16anHY9NpMZTxMRWcuVCH57wJp+O2hyyRad1QE9fbX6LI7/6cMNWnOTS41Mxznd4+KOb/lGDBjJgkFHBA8cRAzlqnLNAlCRO47GNj0dfnz0LNxv89o86/EfkyRhXZj9dsnzBEd4pex6W5NNVw8gmH4gNhI/0m4j8+A0OX6/s9UT/DeyfePgz7ru/xKrQE68xbYU9/oqvuFtsZA/xwK3gh4mefnr+kcv3aPlPVAtJrm1G4nCinnHCwY8CDxuwn5cfsnKjhIOt8iJO/Gxpt5Zgpcc2rGp6chIG8uwVD5liVJNDZCK2lDkWL3646fCBH+7mGg+FXR8+RbvYyDRP2vJLhh459tIpVnLIuFKc2mjmiT4+zOwgvrOpNmYe1e4o8Zp7sn3WsZTP/sAHj7gVJ3EkFmMov9pswwUPeWsIsY/CrF0u6aDw1KYz80q+Mf6VsJ0353yxaUwpB2zApyD+2S6O4iqGmW+y+Hyr2UxOG2YyCpnk+deGSc75yAb7+Grrgg19RE+78frGyilbxvWLMRzG2C2mjlN2ED6fLhy1YUqGLXblRozw2w8Qm0qxkzNOtlzST7dc4Wkja2P9oFRMAGoXTM/SAsDpTCZDeAiAQOmzl7w+kG2mrlD50HcwCNziBhgv4PTYYFuBj10YCsQ4vppeicDjs/HwkWFjyvGjX60NH11JZ8dE6fsuKD+S1Fc3fPDBB+t7c+hI6vRnMiN44eYbhtqNq8NcW81ecTQPfGWHjm/o9H1ETmzm7L+3H7qBvxzxxWf26KDsrs72hy+k5oNObX1El1328bJFDp8OHtvix1ebV3zjdGFKjkx2yGpbE/jkK/LLrvHsquEoFhiMhzfd5MRQPsjJI11yEQz6xvMnr/jZtS60e0WVLN/8TP9ssqEYVyKy/KjFC4t1ln99xF72y6O+PCqw+W9weNiixwZsTiz6ijEysMgBLNqNq40XO3mEV61dDGGHCQ7YG2v+4Oz7zlxw0XfM80UnX+zzR15u1dkPt7o8ik8bj1zU+Fxf/GSveOnDgE8WLrbIwt6jV/ljf22Ym3xzwx8duk7sclDMjbFfO5z8I30lnOyHQY3UfLPNFj/8yzHcxv3zpq9qh1tpHzXGNn3t4tYvBm3UGH3EvrZxGJGaHTXfyr9fpsigAQqcMkBYP+dqY2qJMpYcnnby+hKC6CA6CrB8WdwWvS9pJGsjBkqy6MAUSV7+2tAlk52IbQQHfjUsfNYnwz4Z/tRKcur6xvXp+k98P6Lkq5R9pYSJI+erBfh2UoQ9HW1+6cPOJ8zidpdC3/eEscE+SpYNushYk4mP6LSQfG+W34M4Pj5eP4wll/4BsW9LZRNVl8vmhS34wsFvfugZw1OSCyee2Nkqp8bYKhdsZLPvwrKRyIO5bCMWI918wdCc4eUThnwUg5jFpd84fUWfHTrZzI8adXHADll2xFUfj0wxsoOqjWsrYi+n+sVDBhbUOHvFYry2ubXedrvd4s3c0KeHyGtbS77ipG9ydkLx87J+D8ZX3ZgL+Zn4wgBTOOPBjMIkF8aKX5tvJZvs+E4uG66vXbEGyytbZBFddtXiog+f9XC6fQuuNUIfXjLNvbYSZY8tNhRERjtseMk23rFDruPBGMIr/o47fDbwyYvRNwHYrPnz2acnAOUpeXW4m3t1bfbgR3RR8RSrvlIe5AoWXxXke+l8nf/F7RvH/Ye6H7vyZaV0+SgH6eKz5atmfL2+vcvx1zokryCxk6dbPluH+uwcEaBQIILIqYPcoxnJ8eENJWQ8IEDi62tLbhPPgTEHIR19E8cXnuD9doBfPHSw+JoMi04wM6CJEV8QUbhXMGcHCBz6auNNaljDwm4xwYWvkEPGFAeBM/w999yzfiPi0vYbAq5OfNWI3wboR3vEBB8dNvTlAo8NOfRtmhYafEjMvgXYgYOKtXyqERtiYY9uhR9Y4LMQ+HQXYtPwlen8kqEnNjaQNtvsFLex/MVTo/Bolx/y5U/NljzypeBlxxyUG/NH1obnQPS1006sZOlU+InHlzYb8fH4gY2OunaY8x+WdPHJNg4PGTZrh0MOHWx0fc2FWIzBougbU+Tf2Nyg+Cg3+SyPxvhEbJGzQdx6661rU7AZ+OoYczvj5YstfLnj169U0mNDXvywkCt9X0vvyzDTMT798q3Pv7mhW0xk0yNTm0x46Lkgsom5kJEjX2via1ZcVMmHvQResSH5YUtBxhxX7qSN+U0TBX6x8Q0LCjv/5YwP/WS02cyHttK8iVN7knG2xaaYd7llgz01f2JxgvMTEU7cYsL3zbfWMlkn8PLFB9vVYaIjNr6QPkzGUfpqOTEuPvPq90voyrX90y+q+m4034dmvo21ZsqLmOHyQ2k33HDD8uXbhn2LennjqxLO1vvMFxk66x8JCWJwRBhQG5LFqO0k4hmbxVBglCWF7jK0JYEuO8gYwPoBYEsijAnc7zf7XWJnQWdPG2Eg01PzQVdbyS5Zhb30yGqLRc13GOkmJ8FwG6NP3kSSmb70ycHoJ2fdeZCjI/kvvPDC+nptk9VCIs9PONlTXFn5kSKPvoyz4dfCXLX5ls1O0nyWQ22leOrTRXJG35fqOYhdjYiJHJlimjxt/GJvjA+lvGgjtqprsy0Ghb6iLU7t1oIc6CM6rtjkydw7gVjwNsn4ZBQ6+aDbnPARBnwYySrhOdRPHl9sdLQVZLw1qs1Ocwy/H22CWX5d6OBZP/RtMtr0YMQLB1/64csX2el7gTj7U9x+48N6cxz6WnRX50622abPDpziqR0fLi93kIfZCUSOizOf+uUQfsQG7GwaE68xvsgnoybnmHZR5Bc0/Z6Ni0C/2OjrxJ142ci2ttKGaP3C6jue7rjjjvXLnMbcmfvaFvzyB5d4qmvXD2+Y4pNrbtgiZ87w9cWnrcBmLMzlgU5tX2joh8buvPPOFSt9F2zmyFeuh8OaYaeTEL6CqmGUU332yWvDBQ/CK1/mUG6sDXsSH3KIb57t03TpKGzwgceG2PySoZOO+XSy8RMAbLBFLt/pqWFD9jnj+sr6EJ3hAjFood52220X7rvvvnXGcivsatZiKNhlbfuTE0F30GU8IPqSHGkr7mycybUlWfA9G2WP7exLBDv5kRyLQmni1frkjMNDXh+W+tlVFzu/ZOkjbUQGSZwDwwTQMW6By4mFYwL4EAsdcvnNlkl24rQA4EMe4SG3wzaqZMNRTtlkT4wIH+nLmzsZJ3oyKN/Gi4FO2LMT3hYbPl5zR1c7G+HLP5zlPBk2+OdLTZaejYItV5oeA95yyy3r6jq89PIFDxzlAV+Z86pPh23t/FhL/JqDsLCl3fpNd/pmhx6cHUxO+n6Hw1Wf+fZjXV6a4AMWdshrK/Bmu/Fi0Jej+uGFodjZImOd+N1xOKzNjz/+mNiKsdyGV2woXf7JGGdHzNYeH61LOvzAABcizx98bNEzjthUku24YUdu/Ja8iyN3EHzYZNUzXu1wsdNVOhs2P3tBv3xqg+PfzwH8+eef/+4LxU4/bNqww4f4URC+cX3janrpGg/TbJOTL/kwru84l08XvU4eTtDsuPh1t+Ri0HoRj/zQtS+g7sTCgheG6R+fL4VsfTX8zWPffhx/v9+vR9Z+c8U+VSzNk1gQfSd788Sek5Dfl8FD+Tam0MPTRq3dcrJWB2bCQFs8FoAfr2lBuTWjZFFEQJYEetotPjIWgAIAm6ikOEkJTrIRXUmGgw2+yKrxtPkrgPp0jcNlnC86Aj+0Fc843xasuwG6fvzIFSa7JZs+Ig+vW1eE7xbdj0A5sdhM2EP8I1hKvhyIzTPLGQ85vwNiQ2XfrSQsijiLJ1t08dllE8mrr5129efRFWzkLW6b9O+//76u8slmpwUxsRiji2dOwk4P0Sk2+KwLMYudvMeRfhLUXdTp9jjOlY0xdhV4yfPhB6r6+VB5Saa5JWM94MOjbr6NRdoKXGQaO+Szi8cmO9mcMYrBBkHWuJyKye9m28zME/zmyN2ivNNnTzv7eOGoHb7yqC6esBSr484VoosJ+n4Pxrp89tln//0hLzgRnWLTb37oWZcIZoWcOeAvfMY7/rTLITtsILLaeJUp507JyxseRe92u+WHHp/ZDqca4Yvf5kqOXSeTdKwtn+XY7PwAlTsw65k+PetPHIo5pa+tNl6BXRG7GPDJIO3w4ZFR84lgYxuRY6ere4+A/KxF49a7Xy+0R3o0BAt7Chm+FfkXQ+uFbX1+w0JGfPwVEzl2koUxG3hse5TuM096Tgr4xYRHhj0nO7rafBZvfXqKPp/wTDt0k5HDIwsNYAdyZAI5NKHaJRCPUQEzoq+NCjA+x8nSB6iJdVt+cXte6kocpaPNZsQ2O3gl2JiA6KwAzhJLjg+yMGdTjZItUQ5UG7erJvYsVDno7M1OiXOisTBcfdgUfXX1W2+9te7KnLnlLx9wlJPilgf+ba4eNznj88OHKwHPUtmm57ZfnQ69SCzG4mmbI7n06MBmw6YYXd3D1tdstyjZKCdq8sXaQmMfr5j4qZ0+Hnm5vummm9ZVmU1EzA6kfo+7NUBejOJ1lWkhuqL3YR5fzRnsfLCDr80PTM0xHsIPMx5dlJ42v3QreChZfGTDkD8nChuFR2vuwv2EKNxO8KfbRuEuj0/5zGYYw86ngozBQx5PXM0tO5G1xoeTq8cTnnGTM79yhcjQZ0cu+NfmQ5ufKP94ZB3n7ODjqfPPLtLPXjJiKk5jzYG7IvPuBRg/aevKG5F1V+7RE7v80tHOtrhgFgv7+vYEvEnyAbN9gh36xWkMXroTd/kutmzSRzN/YuRfkWcnMT+Y547KxvrXX3+tk3dy1oaLNGs1n9aKeD1NcZKDz1zzq7DLp7YxFKbWEBmx8YPYly8XCnTkD69ckS0+cvy7uLWnOKbkLF980CPvosgFs1w0H9nhgw7Z8MHC1yRyxvGVI84oWQACAKRkZogSIJwFRkD4koNf4awFAqRChyw+eYl2G+jzBD5t3DZmgfFZIkuUGs/YHOeTTbbFAbckKHT4ra+mGx4LxYEqoe4mGueHXXW6t99++4W77757XQH6wPuTTz5Zt6s2FrL8K/yxL0b6YcaXL+W11167cP/991+4995714EBp4NHLsidnJysD+N8ODfjMwY/XrHh6csbef7w1HAYU8uRDTJs7CBzGq9+PuiKny1t8cHPd/Ml506sHl/YRNhzJfbVV1+t+SCPF14f9JF1Ve/E8emnn65fVHSw8ktOgRdpw1AfjvjGkDG64ijW5PHhZ0Ne6MBjo+DTmmvuzLETuas3V2nvv//+su2kb115pn/9+vV1YuGHbXbD3fyEC78xOGGHpXx03BRfenJmMyqGYoSDDCwwG1fYCQ/7qDWgTUcxxgZ/Nt82YLnAY4NcMcGrn5/yp5Y7J1ufA7jzhRcZs3/4unmP+vhk3zoRN9z5MZZNPI9w+S4G9mBQ2LfG2VbaJ4x1zIcTZvbYRviI3XzTg4k/BJu1YO4dm07iPjOyjl3s2RvlyxuYHmd6tMiuMevCm25epMmesXzAU17NlYLIlGt9bTjghsXagVmh75gj48LXcYSMOVHLt18mJSPH5FH5VXvE7bjzuaM7a3LlwJpT2JNPNXwwiANpV7PP5pLTIKSjAOmWWbIjMgLnNMd0GOWIUzIIEDaAY8M4nQC5gnOmtHFLhLOs53b77RmeK312S0Cg4cJTWzzshpndcPODjIUJHvySAZcztlvQfl1xProobjbT2+1264UCcXnTwU+1OhhdhYk7HyYw/OzAasyCwCfr+bnFCUdY+XJ146UFvvwWcxsXGeN0+RevNnt8iEufD1Tu4PTzuO5Amg9yydDFR+yjcmmMrPiTaS7w+TAmx+bM3Y6FbeP1iMf/oNx4440rfrbZs36ccC1eB6rHfz5bg0/sbCryVbwwiZNvOLSNNZe1yYWvmq3i0OafDXOiLo9OrA6sNkP88Ng0bJbmzM/5WqPmXOzlQQ7kPpv5nNj4J0fGJhHGati0jYtRP8KHqTH6+igdPrOVHjsIDsR/umGFS5uMtoKqk6PHHj/hkyMXD45laxkmJwE/a/35559f+Pvvv9fPMrNHR87IsE0eXoVteZ5EHqawk4fF2iArFvPIHkxsGo/MjbH8kc8Hf+yySZdMflyhu/NTO/7UTlhqPyPsUZ0LT75t3PttPTjGXFTyWUzwmMN4sBnDV8sFn3BF+uRggq09pfjYo+dC052+Ywk5iTmO/POwuWBDTGo6fLBtz5v7Ll1+yKo7aZVL+qgYyMGisEdOvb4Lq0lUS5zHOv63IAKCIVcTkqetmGgJaRLwOCgI/RzF44MM0MgtoCtRB6hNx0SXLABRk8EGf/ps8NtiY7fxuWDph6naJLgtph8+Y+LMHr54PYv22EUs8CgmzzjCL048k6yIo3Z2+XPSsih9DQo7NrCIvIPSIm7jZ9sJjq744MOL5IJffETOuA/13n777XVnIg4+4Wge0tcXOx25FV/2jcGE7+RoM9W3BixwxC8MyJjHUz7go8sWbPyWXzbwjLvzUsjxgdgji8hlA6baxrOhppMsvcbZ5NuYvMDHjrWneMTn1Vd3l3KO3BnZ/KxFj2jE/eabb658wmq+2DUn5VM+ULkoFrzw4Sn8l199NpB8FgedSFs8Hk1YL+Igp+CrxVKcbOJFybVO5B21nsizg+RHQfQU/uWdHht8+Udax8WM26vGPqt45513lo7ja64XduQyzGo2FRhQNX/IlT9fLkxc1IrRWibX+mMTiaf1oSaD19okkx6fjk8Y2BSn484jy+amfDlRuLv3qM5dKh0nSp8terHB0wsXRCjc7Gkr7PErr/jhko/kjOmThbmaTTKwyAV87uDdCcm9NWgfkQMnB3tgvvhmV12enSTMH1J7cwx+xwH/5cc4n+lOrNowIrk7ckAowNjMJePy5cvrNUuOkcly5UWZEn4LjSMGgQdK0c85UAifLLAWF3nElrYr2IJX58fYtNUixIOnwjbCN8n0YMTPt7sfG4Pn2q4onCxLKhwWiolpUdlAfLbglVPj5YhdMoh/i9HCps+fvhj4ZZ+8cbjg86qfTcrjLB8myxc7nr36PxFyr7/++sqJvIvZOHvG+LGZ8cEXzMYPyYYDN5/sIPgsUDx+jdNlC19c2sb5s4CRGPDVdMi5qilXZMjaNNwuw9hBTs986NPVR2o6+TivH/7mF2ZUDHDKiRi02Zht+ohvBV45kz9rwP/22DzEKs/ffvvteq3YG4hOLjZux4f88UmODzj05cNY+saUcKvJwSTO2vh0w69vE/C4Rq4mkXGX5yLAekLJuBOg55hiD/ERsUvW+pPD9PHhTHbOj3w2J3TEBgNdV+Pu1vy/B7/InH/55ZfrVXJ5TVfM8qRPN5/6/Bk3F82t9WQOHJfo4tn/lcj/119/vXhyTV5hBy4+WhdslmtzjcQpJgSrfnnXdty5iPAasXVLz14oN7vdbj1yZ58sfRe77rKsDfmHI5tiRDDQ58cYm/HIkzNfCj4SE1kkD+UL3wnb4zUvLdCB44cffliPCuXNCabjgC9+Edzse9rjwq5j8nT7PO+LL7648P33368LEzrs8m++UTbC17h+vPUZiEkEVhE0sE4kAkV4gojwGag2jjjkRD8eGaD0jfUOsitVJEATaOHxIQl4sLCnT7/AG8cjg8JiI6XDpySQ5Zc9+uJ0AIhPMh2QPjy3IJDFQ94Ctnn4jMQGbwG7MnXVofDtgERsi4tf9umLwwFCrhyR1YcVNh80W6QwkqUnHgelfPDnYDGGjNM17pGLN0H0Tbzc2UAmwWwzEqPFCKMc0OETLv5gmnNmvNxnT2zihIUen2JQ8I0juq2TZI13UjauoHTERQYfFsVYc8uesekn3HjZgouuutjYwSMPDzvGXKx4DOCqshcPPFb0Pvxnn312YbdtGu465dCVr3VRDPDwW47kgo/iih9PDUc5hzde7XRdqPgsycYZ8eWZ/JNPPrmedXtc0QnLpmDjs6ZhdEyxDWvrxdxaH9a7K/lyAzfbCv/wqcmzoWRD7QQgJ/A5gVij5dqYNwxd0cJhnRkrTu3WHlt88aM4ZmDYbTn/5ptv1vqxvh2D4uhRMz15VLTDzwb77PBXTtVkmns+8m2stSG3Ptfwpl0xeSPToylX5+46eo2bDTZt2F52gVnOFPsBX9qtAe1whwfWGf/Enz4/iC4bcif3LnSsSfPoP869deVljy4U80GHXWvVsQ+bz5ydIK0DZM68udU/qZJHdJo3fbb0W7+w4ckfXEdd1TiDGVT6DISiheG1Rc/ajFFEDOo7MG0k7Gi3sQIo2U22E5LXgl1he4SEvPvtUYszuc3OxLQI2JZQttnId3w8GFBBqhEZE8V/E6Z24vAMPoz8+rzBragxeiYITncevWrKLl+e7VtcFjYbbK4kbnriJQN/GMMCh4mUH+N82Li85bHbFqGDhB1k3FXNlStX1vvvbpcdTK5AbRjyZLHbbA5JThAbcqfgtSjKEz6Zcige86qkx47xbKWLp23BwWw8srF5RgyzA4ws7PxbH4o2IuvgdFfocR078oZgKOdqJNfwyTG7zXHjxthQN+dhp28OFBuN3F/a3sCzFtiCyWdGjz/++Fqb1qf5ZVuhZx2q+Q2HvnE2YOYX6ccjS46Pclt8+Mbjm1MHeTEUj7t2372m79Gy+bLenGh8oacTSFSe2JRb/6Tp5Ain+YKLHTU/iI6Cx64Ymws8Y+6+5c3ac6wUj2PWRZhxm7E1zq7CD3v0rZeOD2sCsQ2ntUDmu+++W8eGRzX2I+S4YRu56kZsN//6Ysu2MTb5Rvp8s69Njrwc2pDdhcqtfnPpM1lX5taxty3NCYz827DdYfsMd8bHNn3xiqsc4pNTlxvjeM0/nHQRPrxiYEufL8e/GJAP7F3I2j/ksgs5euzkjx8XD71N6KLTWMenOfQZJh7bMKePN9vG2A977fUar8GuhrWBV5w0TKordA4YbKIko4ky2RaxBWoiOJccJHECdEXvcYE3sBzEyMLzNpKTiwmiCyAfijaffGtHAi9oGMLVlTY5/HyzZdP2zz+uoJzNkXHYbczidtXv8x/PGZ088smfRcOPs72+mPiQB2QMwYXolkt+YKAn8WLVfv7551eOH3vssYWJDTng252GKwz/T2EDkTsbG59ssEee7fzp86l20MEsd2KnJz/G6DefzZVxpM8mOYWcMfHAbJwP1MnBhYCTg9x4kwn2q1evLj/ya75dmeKLHWl7TATbRx99tDY7bT6tHfHlK2zG4sHYusDXbyy8+uyQ0yZn3tQRviLvcuzq18WDAw/JlZO2zVr8yMnFWmfLxZZ109zm89CHnM28wqiviNc6glO7YyjMfLLnbsMaRvyRg83c0FXIsUXXCw1ya6ORWySe7JKlC8Phmgg/vjZf4u3YDbe3kPyPij4s6ja+5kFM1p6+vMkFHjnt1jJ8s61vnF2fhXjsCy9Sk+UPsct+MeFZn3LBT2uXvDjU7kDlVIziEiMbu+2izjHo8w95g1XxGY/90AYOU+vKWPabQzx2m1cY2A5v88Wv+TCGJw5Fvsh65O4ioWPfScyFAR145UFbrbABC//uoJ544okLDz300HpzjJzj1IXwSy+9tPQds+Vc7uQ7W/zTiS+n7JMRlzweEVIwBMFYSoQDpa0wRo4RfYHalF1NAhORYwepnem8/hZxLkgBOVjJSDi7dCWB/WmDbvzkyGi30PWRvricGF3h+9zD5uBkYUxcNjevISKvn/ofDVjcASC2FP8J64N+NTwWFRxsWBhyxybZFnJ9C4QcEpcCb3HMq/JyZoOCuRMZfw4g+SWTreyKE59/+NTGmls8+cbn21gLvIVLxhg52LSza03QnQcsPScDdw+epVrYd91118qNg+/VV19da6M8sNtjluKk78IjXzChFjRMqJiLR58OOTqwKPrmg08EUbjWygAAQABJREFUM6JHHrGZPX1txT+uPfXUU+sKv5MHPoye9zdPbJljBy9/3tD69ddf1+MAPtgnEzY28Lujkdvp11y0lpq3xsOqFp8NUZnEPiKD9NmBQW60mwN+tJXyRV4samu1XJEVQzHBSafc8sW2CxV3b+6MYGDDXNMjXzGGV27Yc9xkn5xN+ZDk2tuJ5kCePbHgg2/67Gmzr7ALN3tq4/gdo3Kiz6557oJGXHJA3vr1qEcOdtvJxJpyxe91WZsyWT4VbTF0ccY2G2w1Bq++MvWMw5N8scspWZ93OHn43xN7ADtOYi4su6vkmx0Yxcs/3cjx1TGGJyZPXszZbosNhZm+/DUn7JCHD59c+bVmHM/r23gJUAKcERsrYYsVeAAZQmQV43gM+azAB84mbZKrMweYM6bvjVKcURXg+GLbwcWehadu0ZGRcIQHPB7/8ZK3QAqOTAvZ7bbFB6Nn38YQPXG6rXMHhO8Rgtu9ToQ2DXdJPlj1rrdFZ4yutskXgwJfccABD76JlSOYiyU9J1Ub8H6/X5jcGckHOTrlyiB89FATnE14tCtknHBcXflCO29qwAIX++TUCJ+/cq4uR/yQZZ+MNhu18cXnCtfnB9aP/Oy2hXl8fLzsH/6h67GoR1x8eSTo5MPWxMIX2fgwwRy+MMA4CT4042mcjjlyBenKmV8+b7755nUCsT5Qfvm2wfTINTuzdrHhS/S8lgwLv+WNbdScsmscxjYOfX4cS9ZUG5p4FWPV9LXZlR+6+ki7nJFDxvJpnA++rU2+jIeFHEzG8LNv3BhdfCUSpzX69NNPr/l3xQ5/8Tie2XQH7bNPJwj2Ir7I4sFH1j5is0Th8NmEC1Rrxv872NzhQ3IRbjEhaxI2dq1J+Pnhn83d2fp0QYuPyBebk+F8C9V6IQebtULO8cmPE4r1zEeYOrbIkYGRfsW4vVXMnvLg46nFYo06efjcwiPDCEZ3YS5k7W8w0MPnm5/yYW9z0Wx+yEVy3hor/2Fngy0FFvj5Mff8KGT5MLdOwEcEKQi0iQTSJDlzeffZYx3OOGCUETUjzmZ0bYR4SGKQ96NfeeWVdcCyZ8L8I4tnj4heSVfDwjYc8dnSLjH1w82OAAtcn74+PVeVl7bn3SbFCc0zWz5sdArM4nSrhy9OPrSdQHxG4lGAs60xfDE3KXxJMvsWhSsReSkGSc8e3fIHm0m2+bpa17aBkY/IK0i8/E9iV8mmNmLbAry4PXZ5+eWX1yKXjw4wuUyO/Wzg8d9Go802eTYV8mKz0NUwwe5Cwa0xnBY9HSdItrTpyZF14Ln8G2+8sWTJs9WGRra55ZsPugr+HNeHqfj1zQsewudffHhsiclB67Hsjz/+uB6lWZNicECQS9/da+tl+mCbLzwbCDmbU3r4+Sxf4eafroJgIo/vUZjn03KkHyVb3Tw3Tl9BZBoXk5cr2LVG8dt0yNGxVhV9c6CGFa5yIYc24rDyIy5ryvN5jyP1fXbQyzds0JNrnzU988wz60N+c82Hwn750caDMfz6MPLtA197jJN+x0m6cNAhixefTTjKjWMTuVDzWaz8nG5vI5ETQyeubISRP4812SGL+HRB7IN/e4S1A4Mc0UPizze99ozwkXFMkKErN8hxw5/PZ9hAaidoc+UEwA878iE/8u5Ey499y3p8+OGH10WxeaWveMrihNhcw4qvFhMs7CJt/HhySYYPubR/H2G2iARhk/W7HO4anAFtEJTIMUaZQQvcla3bKf/E0oFhLCcWT5NmQ+uRDBsSwB/fAsDTZ8eYwhb//LLDLhmETwcuATtAOujcnrlqfOSRR9YH4iZJkl3tnpycrMB99uJtEicRvkwuOxU+LAp3L/x49ln8ZEx2/uFEna1hgwtWOmIko81XsWob6+BkNyKbXfJInvDkofHslpf08eUjXX0lOb7YMK6u4GdTjPoTczbUxsQJF5nICdfbHU66xrLjCu7y5cvLl89N5NdChimcarbZC4d42eGrNQBvMmFOV7842eDfnPAVmStjDqZffvllzYPPQFxMtKa8KedLRJ0gEJvZZs+8OU7UDlgHFJxkovrlGh/O5hA2Y+S8+cOX48SVY/Flq/jq54ct7dacPnI3LWb/Fe4OkX6++BOPOLTD13EsFvYQvXyRQ80RG05Sjz766LrjdRwZQ/A7buXUSRHxF7EJK5/pTPva8uPCxGuzLjz4as/gO9ywFgd7xtg0Pn3A5KT/7rvvrnGbLfkHH3xw/X8FbOJF8BX3xG3cBZMnE+6IzD27MPCp8F8ujWUnu/JLjl1jsEf49ix+6JJxMe5VZm/hiUlOnTxczHory9p1oQ8LW3ScKM0/G2w5AbPhcZxHtuVOfvhE5VEbDzbjZMslGe11AjEZioCRA8tVsQNHcoFhJAMl1+Zgk/Ahr2Q6KxorUdqBYh/g/guZH8m1eSO2gUL0SjzfgZUI/DYjcum4csCXUDKXtjsOZ1+vakraH3/88e+rdzYFscLLvhPNbrdbWNlUIpPhzRNXhR9++OHa8EwsH5JKX4wmi019pRyoW1jG6SA+9CdPLOlpN07WQehRmgNIHhykxc6eKySf73RHhSenNjZ4+IUxHe2wqYvZuCImftT6zT3ZcGlnr3jI89ednr65h9eVqs91tMmwzxY7FfLyhfAQ7OzDgFdtTJsOSh4Pqdk3V4id/MSni2dD8cjC3YYD00lhvz1W9LzZydC6Cgd5bevAmnO8zIPdGGK73Ml3JJY5Bgtiy1pVy5lcdUIyLh526LOryKN8+idIj1b4pss+eSdob++IYeabD/p45LRR88GOvPHVRufqttdH4XCCi8h6A6tNKT7bkbzpT16xJ1MNfzHAIv82ancM1jhb1kkbIz1tfDbhRmpYjclVMq7CHdPG8cLkCt8TF3ubNeCCkyy57MoVHY+tfN9bH7Q379niHx5xNOY4Zqdjy1wp7LNLN6x0s6GGxz7rUakciMtjQU91+Dg+Pl5rgO0oe+wrcuAxIOwXt6cTfMIDH1l2kLo2PeMTI1mF/lrtFgBAlAwQdhJxh9FiY9gYIt+mbZPqFtY4QJLABrkmzsnIQkB4DhZff2zR5Rsg/thRSjR7DhT4SrSFzQd/ZPmC5XS7JbXZS662qy+v5Bkn71aPvNttbym5XYQBJlejFgybiC8TbON2MLKh4M/FIcEwsouvwIu0+VWHXT95OhaEWFto8sGvOymbmkcQP/3004X9tqnBQbf42fWhnxNdBxe/cmVzg8MGJz48NZ9KhK/giV0biRWPP8RXY/rx4Y3o6KvNJXLV6LGWR0X8w8WuNtnmkH1UfsMYjjW4/SHHNj1jzRds2YMtrPKtjVcM8i3H8oivnbx1JN8+28BzUkHwIvaULprgNYZX/tjlC77p0xyTLfbWtXhgMtfuFvDJwAWfXLDVWiOPT8fGbsxJUPz0WkuOObKOVzgVMRmHK2K/PJFhRz8ZfReM7sRd9Xpxxrpy8eIzDtR8aTvePQpXu/vw2NrxBSu7+YINFvjlTK6drMSvb5/w6uxu98+H2c1R2OALO79yAj+bxaktJ+UR3zzJSTgca/YjF59I3vbb8eYCGTZyav5aN2xaA2zxAQfbCgz8aavJlB81O/kOT3PaWPLwOIZcKNiv3KXS8XQET67YSj6fePzyL/fyaMx+jZcOnkJfqa+mryYPF4ytjbM8/5NoQgSQtsmTIM/19TvAcioIBvRzIIn6E5x2G4Yx5CDxA0j+I7tJBJQ/BUhJMWH6DhKBwRC1SPBhpZ99AToB9sYCG8bhKwnwlzA2LXS3dv6702SJyRUD/M7+DhbybOEhMuwpsMLEt2IMLjr6arx80mFHn22y/LGDHKye0bt9Rx75yAM5ha9igYltxJ4xJLdOOK7cPE5MZuKQk0jbGP2wsSd3ijH8Yk4PT+6TIQdjfsg1N3TFC1vrJbzkySn8sqvIX3lhSzvbzamajhKVo+yRKdetZ/LsWW/yi8y3Qsb6Rx3c2sWvpt8jDDHjFUPx4EV4ipjgEZ8+KgbHizllu/XGhvVBD056YfJCAF2vRbsTIJtNbXJOitrkOvHxI0d44SrPxvgwp/ToI33HrZOrXB5vV75e1jCf4kcweuRysj0udqXvLsgjIjh9V5a8Zld8/Cv2Bf+D5X+gzIeThwtAL7DAIq9IXuBVK/Cpw04Gj2219U+XDX60yYodZnh94OxzkebbfqC4m4aFfXKKfLYXiSN/7MLFp6JPL8oOHHwr5jJ+mNMzjsi7yJUXfvlHTtzisS7I5rfjix1kr3MitOda1+3pYkm/tVU8dNmDH09cjZVnPtdrvAYAk3DtSsFzVNAAaRtTGEZ0Sh5QqEBdGXNKBgHrWZ5NfrddWQDLJtLmD9HHVwSj5oPffBSYYNj3qMQzbXyLGJ//Jp0dC9WtqoVPHrEpuRZ4vz3Q4soXG+yJZfLKF5/x5VIs+vjaUbEmi+8W3VWQxUTeVZ7HVh6vsE9WXvhnixw7/Kiz2aSTJ2PxzJOHOJsHOvXZDSf7xpo/Mqg4+aivLifk6ChsODhgppeOPlyIP3z1LOHDY9tmRk/e8YyzYcy6hRXPOHv4sCj6dB1U5IzJjblllx2vcvvcwYnAYwt3H20ccPKRnoOPvg21OTYWfjzjCp5cwCEf+jCgcsI2PnkybClw45GTR3iN46GZC+vVG4QevaBsa/ONmku5KOczV3QmZnHgkW9O5Ur81pQ7CWPy4K4dRjr80XMB5G0hd9DkjVnfHqGImQweEhc9H5T7b/r45sBdjCtndzrJleupm99kio1MsdOTW2tBkVNyYnHn0yv+TnS+MNUJzFt2bLNBnzzs9PVbU2r5MBY+WPTLqzaCoT0kWfps65OXT7WCnAQ78S/G2R/Htr2juwsxeSLRI21ifNlznfg9ehQrf3hdKPADbxi0y6X4YVaLkT1t87he4xWYSbNADDIukNqU9SmRzThnZAWLxyA7wJQQNgSur0QS4qqF3RYU+0idP3a1UUHmHy4YGiswb0U04WLilw+1k4l/arRhuP2Dw6Msj6mc0HzwRGfa5N/EFPsa3P7ok1Unr25hSrZY4FTg1oe5GIvP803PJifJpdxN+XIoVrqIf7lEdPKrdkV6cXve6SB3Wy4W9tTlFbbyJcf47JMLNx98K+XfONJPtjlir+JkaLGy5VsNnBinfzpIThREls/mtDE1f3DwqS9Off7KiTpZ7Xywmy0bhitejwJcSLDntWebB7/WhnzSYT8s7Jl3NZ/NN9xskxcfnXhqfNR64a8Y2GYPqcmy3VziiUFJR5s/m8Kl7dGL4xCxj1+bHP38mgs2+KSjbUwRL3KioAMDkouOC7HZQMnb4E+3R8X8kVcQGXbpOL7c3RsrHvL8I/gi45PYQcmr5TX76njp8RsO7WzAS9YYnnkTv38odJzYTJ30zL8LyfLPn5Kf7BgPi7GJCRZ9hUw4yLUOxM+Ggi+n4nec+tzKPJkP+h5deUztJOyi0n4mt30FjzfC4PIZibc6nUTsdS6KxWR/cceSfzXim56LJyQH8MHbvNSnAwvfcmh8/aStRBICSEKBxiNcoaAwjMcpIotfP2f6HJDv9jfQkmScriTik2VXQEgy2dXPNz69AmmCws5XZ1S2+OEDsS8mz4m9geXkIVa6JuXatWtrYlxxFCM/xvmnj9jlFxkjE45i7qBcQmd/yhlbdIoBP9vwsSUn6uaA3Xj4bKjh1JZf30Pk850OQONwuMty8Lt9hRfJifH86hczPLCxyy85ZFwha5xuc2G8vPAPqz5/2k6CHqW5ovx2e3PFCb7cT9t8RvkuP3wWc77gIUcv/PTDpk3WgjcOCzvyqu9/nPyjlv8VYtvVnA9F5dMB1VqiIy61xxrs1w+DGo5i57d48OAs78UhP+QObemTbQMRB33yyvRPH9FBcBiHNcLjMz/5dYzBxhc7+Ch9evj6ZNVy0wbSHMmTNlLPHISPrrHWdPiMa8OHypl2mGyosMAZ5Y+MNr3slF+yxhS6zY9xNsXhnxT9H4uXFtyJeGzsAqcc2X/44J8dfPraYsLnl208VFvfOD+RMSW+mj17L3vanpC4C/YhvT7s3qiD2SNtLwWxaT5dnLk4nDjzpXZXxxYdd6hioQNDcWiXa/HB1Lg2nr551Yen+Vy7IkMYBJusAKklwgGPMtiE0BUMXYls8XAoQAvGbSwyjsg6WC1G9vgmm40ltP1hV2DG0ZxIeuw1Scb08RuDLRl4bLJu952ZXbGTKykS7IDd7XZrMoqPb3LZZ09BtY0jdcktL8mwU5sc3Ep5xjOOR1efTzkpDjEo5Oipuwrxn/Twe5WPnsKnDzA9FsiWPLMXXjJ8slWs5/knT06+1OnTCxfMMBiHG3n+7Z12LzVYuP6T2HPxXmaga42Qhw3BiuDITnHz0Ri8rTc6sCjljg4sxhT2sgmnNQAHG67q/G+ID1J7YYIOWzZPdq1HfTbMc/kwhmcMXnqI/2oyYSGjz2/5TzebxhR2s8cWv2Tpsx/GMJBXotm2ZtjsAocvRVzs4psDbfkpv/zwSV+bDH8uDNizZ4SLX5gUcZLng614MLFH1xi5ibN1EB5+kitv2VDjGWdTLPp0ww0b3/rGxOEOy5ckuvvwAX9xeBRkb3IBARddPowXG33x5FddfrT5yRc+3+TZ0RafPjt84CE8ha5HgKdnd3bWHzknCvjw2UXilh96HhUi9uBA8uFlKMWH58XSuBqvNcAefGp8fsg0V9kuliMCnEuoQQd54AgxlDGABGdyBC+5iAMBdpUCDCLLFkCBwTc5XrN1W+bWyhh77KTT5Onzo1bIGitwPHgQ/AK1qZLJBtvkfE+T590SGXls5XbVSSW9Fo246dFnW1+uxASHMe3wwERO35hS3HMB5ludDXLlmj47fCnabBuHLX9syq0TcY/t4PGITo7ZRHhk+TI34imObLGvTYdsvvKdPhtkw9HBoM8ufbbp68u1V0w9w7ZJkzdfbMCVfb4nTm0y+RMHGRTWxtkgnywMfMNARm1ujTtZuRLzVljfL+TzOK+tKy526OSHXQWxCT87eM1PcZOZGMOHPzHSTQ42pE8G2bzp4s2av+6K2DCGBw9ZdXNe7bi2tt2FOrbptLnyJR/mDl+BQTzss4esGWMI3vjk6Brnjww9hZwT9LSZvPGIP9jF4ZGiO0D/D+M5vWPTyyM9syfLLvni09dm27g2W+psmzMlvWLz2NLJQzxeCtjv9+vkYbyY2Cgex4B4jCn1xYJfXtRhMla+y4WcGVfgUsOGLx77KL44+FB7i0puzFfxwdVxSdfFojXtYs18I1gU6wCO5pZupE0fvggmfAWm8i4f5CrrP9GBUAqMIwYIa6sljJLgABIE4gCfvqDJGsOTlADgAYEsaK8BsuOg9WyOLB0yU4c8Pmzx8VDBsG1ccmAoqeyzK+luT30lti+j48+B6LmhNzzctiJncvLsVfguxiYaj299bUQPdn1+tWGBK+xkED65JkmbfLjx+SfHt7jMQ+P64TOOzI9Xk12FiNcimv+8BRNZuBF9PusXxxrc/oSZjEIekVfYg89YscAthzbr7i7LHV067MgD++WDfrkiX2z8sDnzFE4y2tmXG20lzHySSY5/eXLX4UPGS9vnBvo+XPRztfxZ3wg2fuFCxvhoDoyJAbEPDx45GBAc5T3ZYgtXsZEnm111dsOPxz8fbKuNwaimr0bGrQNXrD6D4Ne8ZEtNX5kxWVtiZ5M9pIafTbHBkX/yiA02yZHRLh59WJp3WNpjkocDz5W3NyHpW8tOeuFjj5w50w4jHjuKGMlH5BCZCvzFAAtcXp7w+jSccGQPDrJioK8WO5pjeGyFsRwkQzeZcgynGNhPThsGZJxNtg7xki8v8MuVizUX5h7L9ZloMcuLuNiSP3y2EX32YDEe9uIk66IGhnySp7+VfwACQJCTkqBWGOKEAcbxMiQZJUbSyZQU7QKlzz6ib6NRAENwsEOGLFJP28b02dQuUHJ847MNE9/aFX1JdPJiwxWHTcMzRc8cJZwMm7Cg7IQ7f2oUlvRgoMsnHlzq5PBhbIwN7Ui7GPLJnjZb2m1weOTVCh8eJTiJwG0+/XequaBrUeI3b2EJazmFRXwV49oWHZlin/Lh5os8f+R7FZZNfDjlaNqGBy72xKdmr5xpKyj/+uyRYas+mSlPBvEJE3ytN1do+i4kPFLzrNhzcPllo4PY8RA+bWPsqZsTfLGJGeWXTPhgR3gIdnL6dLWb6ykDI10Y0tO3lpF4xDCJPhlvPP3888/rxYCwqmHKNxzmAA8OumHT1w6DfvHQgRfht/bJapdz9sjilxd6+mIyHhZ8OfS4hT67xmZJHi5tNmASLz94SnHRJcNeuupk4afrCp9fV+p08OiwXZzstI7k37hCHrFLvvb0p43KnzUDg35zoy9+a09+6LCPTwYfD56pxxY5up482Fcj8mJpjrTJq42FVwxyhvhCxsgZwyOvlHs+jzAMSo4BwpxF2gJQKCiIbM7ZaIw+WYQvEa74vR3gap+O20Z8BQF5WODB40fNZgGxi/CMFfBMLBn9au9RO1mwK4EOLv8nYlKcVMRZDvjRZ7cY2coevjZs+RaLA1n8iF+6+jAi8vgKPTVefDyy+PzDRj9Z9vhVK/BOu+XCm0ThF6s3n5xQ6OKHMZl84sOiNqbAkk9tY+ri1k5HjW8t2KA9X1bk12cLvk3VFaWxsPOdf7GIK1x8GWNXHZ8cv0pEJv9sNM4GgomN4+Pj9RU9PpT0TNn78R6VuHpjnxw79IqFXjbYzeb0g09OCTcdNpXw4LGPyiGdePjJqvkID7vWWTbpaZtvXynku5PEIQavosq3Hx7yWM6GR9YYO9ph1YbFBmTMfIeB//xoG0PaYWOTPh005TtGxWA8OTW/eHyq82uMvH62yPJHzlj7Qf7kBZ8MHXWUXn3+ENzV7NIrP+GiyzcdbYUc/2Rg1S8GvGSKmb42Ip8tvuimUz6mPbbokFWyyZYxushjWfuFz0Ld/fvWDzmBk10Xly6Osp0d+tp0i0WfDvvNn31EPxIDmXUCyajBhAosowE13kIjExnXNy5QpK8IxCuqrkhtHPQVV33zrSe66agFUrKzB7QxB4SaPTyyaOLXViRBcVD5KhN68Fqgzth0yfFlTMzFk391cpKJSjJd8sbVCmILGSeb/cbVYdHODjkkj/HDAZu2MsfIzzz4Hxg5Z58d8haQhcIPUheTdnkyxjYdpJYrMtkzXnzG2DEXZJxIPSayiD2T1Xdn5IsdffVCCxneTiTlbq4d4wo/Cj8ID1Y62rAgtYKHyMOebRuplyh8sysZG6zHJR7zWAeww8O2tlznS18pbjatazw51Z85gpduOWQLkZ9xkKGL6JBX8iUf+GTgykd8d85e07RpkPE6quPMmzf77Zk+O/T4KZ4Zg/GOIXz9bPOL8JXZZk8pTr7p8qFd3rWzj1ce0lfjs2OMDUU/PhviJkvGmD5M8chW4vOLh8LBtuOXvvljq7VEz75kDBUzGW01HcT2XKt4yfOFyEdw8o2yh8eXmPilBy+54mSTH7zyoE1Xn5yTA1zavoXDesRzMeEzZheP7krYnnMbPnV+J/bwGIczHMWkXpkiCFQJAC6DS2hTJoMfcPzk1eQ5EkQyTQS+SfMs1l2Id5eR12d7lKDf5LGF2M1HfTjYxZ8B8asUdPLVxiSWD/YlnB3Y2IE5u+WCXP7xsk+HvHH+wpkeHplJ5I2Hz6JNhl1j9emVg+mHDMILFx3t5MXAB5s2FbHh2WjyQadCFpFX8Mmzp+SzGOc4PePhr862vpOHjduHoUj++0Cvx6ZshLt54Zs+v+JlEx1iIqMYL+/qiZtt+ebXBYu+VyS92uyDSRcjxuWKPxjV2ckW34iv8hFGMvHLaZiSLa/65NWo+GBAjeWv+NIjk818nG5v5vgsT3xel3ZStN7ZTFabzdZUvtTlyFj+ksUTk6IdbnKwkqNv3Bg+Hmza8orsAfh4qDzRIdtxqE+uPBjLPj3jZPG1EVl2w4Bvk083ueyyZ206RuRJYWP6pMMmG/Si4iFrTEHklXhzXN6yQyZ70ycZJb1s6aOpbyzddNj0LRqetPjMV74dd9aDOxTjckI3/TkX7OFHM5fkjLMRTjk84gQBl2ED2niIYpONZ5yhJnwJbX84Sc94OsY7c1voHh+x4cBtUfGxAG18pJ8vtclmnwyagRrXL0Dtw0IHJhvpxC+hiA3jiB0lG8b4Nk5X3MaLT11/2pFoNoqrhUeeXMV4sunTQ3wq+o3BwhY9+RNDubdJsF8sxoujWI3xx152w8RnvrXJhG1iEa9+sdt0kb65dnJwx2feHJw+c/JPmsbIwp/Oob9wLYPbH+PiQ/mjT05tTBumTkB4zQkM3vDxz6O+roINH5r6Zy2fe8gjntzB1AeG1qexciBe7QofytyYYVXI0EXpyWNEhk9ycKqRNkqfHD1FHumYU3GKXWFf7f8XbIj8ujDDp1dsaoTHLspfWOgk13jx6Ncmp9RX8ysf1hl84SYnHn21MX7lTSzpsGFMrZSX2mETqzkNA79k2SELP7vlp9yGiawrdN8ibq79b4/HmPAZY3fGqh2PTTLlK6xqVO7YCq9aH2U/G3jwk2GDfW0FbjGpjWkjeVPCgteYNeuk4a4Uz5MAcSM2q9mEiQ2YlfySMV5hhxw7xZfM+k90wRBAGcLLMQMCYJATbeP6AinwZIwBZxyVDG3yXsv7X8Lu5uX2qvzj+A2efyPZZ9JAbWAEYmjdnEkkBWEEVmR3D2BEpRGKg/KBMAeiUEGZmvmQRhgRjYwgTqHSKMKBTkwOTfo3fuu1PO/D9du/Y78L1r3Wuh4/17XW92F/9773DggZXU28fOfraj7wND5g04wVj505Od9hN544xGPTpjeH2xw/3fCQ4yF9cdq85uyQ2PRr+Fo1o0uWLn8aeTpbuP4Uy/q00fAQn/zwp01e+W/mZV2+6VkDdnTMixluc2SuZSf+1IFj5mVuDek5EffpJuvi5bSX0ewd/Hziw+KEc+zXvJzCKBa+OOHHI9eysebloMdvb3nfwxumvqbCXRmZ2rKHp9qa82MOIyqOMR6ZJl9kTEfMZMYanPzBnSweObspw0PlVE+HzAkSuaig1tQn7/jt4kzGN/tsyWHIZ3kVk42xRk/t9OZskbnxxGzMV/uEHp1spv/yJaPfcWpM1hqKo8GK9Oadr4rFpnzowWwOk2acrd5Fw0fffYTfnvDKDU0/xvwXnx1qv+SzGPTmXknOhi+55isenfj8GIehMb7a0dXnyxzRi9qzbtbSZ+MYKx85tRZsp59itlevVl/6+OpwzfrK84f6agnGEqwg5ppgHCLyWVRyDpPnAw/ggBprCuxADYBY+WXTxsDLpqKwKV4Y+SuWg8oY0UPmLVoytnzz14EoLqzplkf4zMUKk54/Pug0LgY54g/ha+HNb/XJLl09GT/pdFC0BmRww5A9/+UHc7bVLv/0862+5hpbNtUhHTK8OafHNvtqNOvkouE9DzIt/8bVnv3Md0/Wn2LBflxHscmrRfj5jci1PpXiH0l9ZY1XRj4uKn6Y6JVHuZLll7w1U3Nxa3TImtfnU4/HnzH9cM59R4dMbnp1JGdDPzmcrXO5qpFmj2jwaGxr7Iyr56yheGTxjPHoihUufKRPBy52ekS3vr1AHsEx46QbT8x8ZFNc/GTVw9x4rhcfbIobtslj50JbfnTot07hmhjo1MLCp2YubvblEwZxUPvImKy4xmExVif+IvuVbvHwxSpX8ZH93p7nh04UBr75MqeD2Oe7/YNHnsw4Hy5K+welGAuSMicUbV4kEHmByCM8jS29kklOF1/jD/ASIgPUhjcGqFjs8eijfIhFLyx8hS3f5hWAD/MKUM+eDhs6/Odz1oFOucGRXTy2csbX8o/HdvomCxtfZHga0s88s62m5hof9GAvZhjwymnqqb25ZpydHi/qpF4scVAY6ZOZ68WauDspe5WBwm4cLjwx+cIz5iOf+OKmgz/t8e0ZedCzoY1bK/p8aMZieKTmPTcXDneb/tucH1iKyxe/eGHNZweanOmR42npwshXPu1TuJon18PGF+KvnOOxQcXnV6OH+BUXT83ZaXhsJvGfff7p8WEu5zAkN0fm/NXjkRWDLeIL0YNRbTVkrsFArjXeJ6HLa0CHjB35xBSfjJ764IUDnz6e5sJAD8Fonfi0D2D1KJstG3K6xvw2pq8Vw5jvaQM/Hjv8YuIZs8E3b5/wB295wohHPz59ONlq/E8sZHRhZ5vPMMDl1Sm+3PNLrjb4bM2NxaunL14kroboaez2CwGgKQNkTBDY6QCPXBDgEFABMDdG9LQSFNwcMH2F4ZMOQPh6utNv84DTsxAoTHuy/sDFFonBNt90kXk9Hfmw0ehH+NUiXhj4ogsLojcP3GPM9Mnphb981SQ+O+PmfONpYrfA+GGhO/NubejQd6Cwz3b6sYFgwyuX6tRmDw8fiJ5x+tWbXjjJxGvO/6xvNmHKNt/FFI8PelFY8ekhdsUwT19t8R1I8vLo1IVDbjCUIxmbasGGbzy+w1MOYne8ZJMem/SS8Ycf5ub0nPzZ5s8YbmvjQJ9Y5MaGjB6faOp34gwDuTGCBwa2cjeePONk22D9qQbmE0u50OcfhnxWLzrxjMu7uM35YMNPuNnFJzPW8NnN+DNGONWTr06S9DUkT5RP/iI2xRCXDR5d4+pBhthqdFC4jfGypU8vn9MvvrmePXzpykNt+TFGdOFB7RtjfK1HxHjhMkbtw9banA5cPQ7lI7zvWb1XKzwtjPv8McEbo3hAGmtznNNknJIf6xQQSEEdLMBJWo/odNJTqPyQ84+M8Vu8ihKfLOJ72vEfpa/no1h08PR4xhFe+YuPps/GsxZ4dPHCnF4bLIz80cGXP0rHuA3CPhzk5uEsVjqzjuzR3DD0slEHY774T7+axqeDioufDj5brU0oHp3isIOL/3wlNycvdmP+imecT/kf+5BHa0oXmSP6YmVPDgs5GSIPG7k5ufh0yNibI738y49+RLd49rx5ezw9fY2f/FZTGNhF+NOWnI0GQzJzuM2rEd30y5lsnwBWTvIjzy4M4YMhf/RQteCn+GHkh56LGTxqLYbGT7Xnh03xwpNPfL4Qf+mGNbx8prMH64+4MGr89Yo4bK0RPX7LT6/BzpY+nfjlYp8Xn7y6GBeD33IhZ8snfjUvp/w3l0fY8lOMdMz5R8Wcc3zxxM1Gzy8Kg7qj/OvVhy+4ypNtYzK+9++BZMgJJUYtdo7NGXEAAONjymlB6bFTbOOAO5iM2zDA0jOXrHk4YCOrx6cnRtjCDxvimxye4habPf349MuLDXmx5UgPXhgQHn0+jIujhx0G/PTxjfVswmoeJuNy4rv4LTwbOqjYbPHo5qfFxc9GLeExZ8snfXby0iM25LDyo7Eh1+Mb8wVrPsKlb0xfM6efj2Kx5V98Mc319LMVD599WJKZ53ODX3/wssejSwdPM+7iZsxvNzRiZ2NMFo+tGpYvvyifZGKzSca/xgc7jb65Rh+Pfq8Y2OMlg5Xceh1T2NozYvXJseJVD7r8mKPmEz9e+PjKR3HN0+cLVkS3Pdpeoles4pp3TLPNBs86IzHow1JN2qt44QrntMNjLw49LT09ykev+LIPCx28sJvzidjiN+cT1uIUk07rmJytMdKHtXzx88M2vHylP3XSJYeXfjZk/JqzrZnLm8w69QolXRitRXmKxzd/eHCJV2xzYzr7u7BKUkCKAnJurMfXT6KTwwkkXgCyoS9OfHN+6SO9JlE6HZh8m8NiscNKt7js8cPON+LfmJ0xHXb84bV58BBdrXj528LLcn5gq6j5zV4MMgSfhqfnz5hNsvDpW8RwzJxhZKeh+sb8wR3BUC50ycz55IscFjHJJ2bzGl2+yfNPhsJ5tXG6apxvemLO2vNVTejxmb9k1ar6TD2+6JHRK8ftZP1pbfXdyDjxomTGbOHg2x40lgMd/vH12YhjjvCKbU5mTs6HMXt9e4qsvOXARg7G9FA2YcieHT+oOpNp/OSXHC8c5Z2fY1v64eOnuPjmsPGdjBxGxGcUL3txvPfgo7N8aNbChcL/LZRzNWGHzDX+xFQfBGMyc7FrbLMXN9z0zItvLhfEtjHf9MSkO+O0x2D3vxVkTsj+v8jHptkWOzz884NPlw++ydOt5njFTwa/NvXNJ/FXbD1c5UCvOuDzq+cj3XKnm7xeXJjNy8OYPbv9bbw5wCREnDMuqDkHdEvAmL4DrrshtuQaeyQBm0UDhr5m7J/MxNBcGW0yduz5FxeJU6HMw0GXjsWJxEF8sNPowyEOm4mNfGI2psdPMrHxENvm/ObPmA09coTHR9iN6RdPHy76ybLlS2OH6FsHeuxaJ3Jz8axFi80mn8Vk05i8+PhavujArcfLLv1whA+fLgxkyMGWH/Z0rdWsB306iE546ZLxW+yw0EHm6Zl3kmSHn199jZ69Z7/h2Yv84bWP8cVEyfdk/aHHtxz19iw7rbjlwKa1m7nhFUMtypM+vXj4Wrpiis++WHoYvalZLdQ9jHTVXB3Zlxd5scIoDuz5N6dnbhzF59dY7nRgi9i5cMjl1ltv3f+TACtscLD1Dc2+Pr0Y8PQGd/uMP5i0csZjo5UPfzDQQ+EpR/40ePNnPqkc8ymeRj/cfqKgj0vTh9/Hwr0KzJ4N3/zkC1b8qDGb8j+WqR05H/Ip59bRnDwf6g9rPHZ0YBPPnE++zPX5Fzt+fu0r4/Z8+tYbLV/vXWn9A1KgSoxCQQQOGCfInHMtmTGfqMXpo5x0HGz4Fhs4/+DVSTcfengkxpe+mOZimMPUxq1Q+OSIH+M2Vv7wZj5bef3Bp6MO+hk3fxNj+tnAYoxPr9rlX97wkYkfZnrGbI315sbZ5AOfvdrRM1YTfvHkGuHRQfrqIL82hDF7sfiih6exD9fEkc/i0NEcYOWmFmJo+Sh3vTXnR63E4h8GYz7oIGP8sJCzM8+GDw3RrdELDzt8eNiSFR+PPN/m/KkLe2O1I6/uYhujfNGN2DQPf/Hwa2E1dwHgNzzqlm01rpb4dKuzeB1HcNGL6CYXQx50xWzvkKPqbl6jj9SAL77ZkssJ8WNMh2/El3h+uMu3xHq1wab83Dy+/PLLV75eiG8Ef2sbdjHhEJdffoxROmzCBg+d9ry5Vwlia8f++IKfb7p82s+w+OAFmVcdvtfPL1j6ETJ6mu+ak4efBehnFNjzWe1g6VxFxm7WrrqXT/Z0wkVmnF85sJOj3MXIrzEfyLi85GPORs8XGwQrG3mTWTsyY3HVrTrhoSV/rwgmlOopcsaoAI0LRJ+sRRIQASAhBfcPLb5C/cYbb9wyPvynpK9d+OAHP3jyiU98Yi9myYjrv0L9rrl/9ikR/thWtApZYjAg+nSmP8ULK//metj1FZ+NnOkqtObKLR88xDc7NvmBi507JxdLB1A++UfkGvuaeGyR2onhrji96i4OXRtQ7YsrXpuHTCy25Pya4/Pp+5H4gK9aha3cyDWEp/GF8MVmg1c+xuokJz25MTLORi5s+CTXuzvFk4eaqbOD3Jxfc/bZ8JEf+dGpRuJN/Obk2RYT31hc+jC2lvhsNHxy9q0l23Igh0HOPt0FazWnxy5fcuCLn2T1YpLR0ayVuThkSC8ef2LI2Rzh9d1ijhdyN1/qaU9Zf/btyWrYMQtT+ZHBTVbu5jW8sMWDAVYtXM3l4wfOfKHmhfXV+WFmg3y1kdqhvqcNTnriyI3PalRMfFQcY3p8kLHlo/qZq0vHDz/mYrGDk69s6fu4t7X48pe/vL/6nw59e9OrKT/SFrG1b32rMwzi8CsOHBp7/tVa7En0woBf/vX5gktDfFQH9nQ0seeakml0+QsDP+ZqRN6+w0P4Gv1qV/xwke1HWASCKwShcY7aXJRRTgS10WxWP8xyww03XPkm0F5l0PWZexeJm266adv74z+BHXgf+MAHTj71qU9diUXm4sKvgrJHARazOXsEF9xTl565XhGypy8fNuR6BbfoiiQuebkXT58v+sgJw4Yy58dY3vJlj8emBWWTP7Fh1sRroVq0+GzxzNnKGVaLLb6LAT454gufDTxOIH4b2V2TdYILj8xdk5fd1YKsmvJR/fjXyGfuYuKRiYv4xuNTM9b4QsZ8q7X/THdnClux+GHnIC0H/u0XNxVdcOCnyx85om/Ml7FW3eghssZk1lvO8iLTUD5hQfYiHux6c8cIezbyRsZk7PjPHyxhDTceOb3kyfLDxnjapg+L/SYenm889sWV9l+4YOWbjjztm37RLmxs09Pj62HJVm7WDOHJnU889siapEPGh7r6cbP1z8pXalJt1Kx9XE788sEXW5jlTq6Ro2oCSzWjp5lPf61TfHGrGT025nyKq/d7Gp/+9Kc3ZjV1EWRHv9hwqLOLoK9m8hjLzTLMYsJGXz7G4sidfXUuF9jyGx467DX+OvboITYa4l8O9NjxUXw8NnDAi4/Sod9ewoNfw0cw4yN9Y3Lxr7wHUkCCClWwbb3+4Eukr6UASNH90t/HPvaxfTcRULp8AeMuAxBzYNGFdUciaScFNsZesfz6178+eeGFF/aCsWVHB9FBfKMO3FlcxSpp8ZA8wmXMDx0bWA5OWAqNR0+R8BUIr9q0+dzp9T8FYrBn59WKl+UObBjZsa9u+TEXw1xPp8Ul45MMtWHowSOGnp66yIW+mOrga+vhtC5OKtbm+uuv3/75o8Pet9C6WNMTH08scdQCiaG5aPFJx/rpxaTHHztzJA/+vLHIN7zlCKsLBhs3D1/84hf3d1GJkQ7f9KoVvy50vnLCd1c5YNMna7138PVHbC08+vTYhZV/OWUvF3JEBgdMxl208fim2960h9jRS/8YQ3x64RIHH0/Nih2/vWZd1VHM9O01j0q8gkduDnznkZ9odRIrRzgQW/h8wZ4fcHvnnXe2TzmUR2tavdhUG9iMkT7/zeHC04spFr/WEbHHo0Outy/gLq464tsb2cpbbfiFB9HJT1jN6XXsisVveRc7fPTYhI+t+GrtRtgx48KHXJDhEYs+O78fJCf4fTnsc889t/Ohi8jEEtdYHuzx2OuRWsjXXHz6iC7s7PIDszksePLji3+kL386/DpW6Wj0UTmTiePxnF8LFVtMzReMemLR/ikm/4g/Y/L9MV4DAUrAnHOFpQywpmA27e23376DBtjjqcPhsJ2/3x8JAu2rJP70pz/tEwywNlLFdfIlEwdoeCpAfunCChsi57tCk8Fvnowc35w9ubH4xhZGr7Wg5GzoawqvF9eJ2R3KddddtzGwYSs/XxHu9xmyo2+M6Myx/MyLHUYXJPmzhb3Y6q1e5mxtOicaJzg18zXlXtGRsyNzl+9il2/x2HoTsPqrFX4b3bj6zzqHk73Gvjrp4ROHrTFbOuqD7wDzU7K+wNAF5uabb979Ls5/+ePk2H+O+7lTB3o1YdZaG1fP+DDLC14Ep6ZmfKihWpVHdvxo+B2IPZ7kT0wncTUWg3+9nPmmyzdbtRBDoxMO/vlpH8YXky1dNnSqt9j0/faHRyn8q636qKe9g7IVg+2ly19i6pUnG3w6fB+T+PHpsreG5SAe4kMN+WFDD08zb/9O/2R8uRlwU+AVkb1Av9zyKU55G6sjPVRMteDTXDNG6e/J+gM7W77RtMcjO3/+/H5k5ebWe7MRW7moiePMd2h5v0NtvDJ2I+YRVvWBEw7HQecGeMQQlx9yuWl48ohPlxwu9vr2Et32Lr1wlTcffJHRJTfnJ17rpLdnXCx7muHY+uMf/7hvMpw3+At39eBXrmLuj/EaYAqCGGj4giByV6bPfOYz+5mmN8QmkbOJSqg5PzbIP/7xj5Mf/ehHsf9P7w7ACYMuHxbhuCCMHLgK0mKFnU0bojz04mttJOPyy3+2im7B+La5/QCWi4UY7H0Jn+e6ToTH5OsyfPOsjZZ/NnyaI3MxzWEz1qthmygb+vHaUHSdMDw69LvychHPY0LvNx2Tl9e+nVNOTjB0YVRjvpAYHaDVj1/Y2Gnyt3nbVGFmB7t8ykUM9taJnL1Xrn6TvvfD2KgLvYlj4seH2W+LuHP2A0leebGpnvp80DeHUesgg72TvdrSI9PShx1WcjVAxlp5sXMRx/N4w7NwPsSSD7494y7VScYJkow/fT71MPMrvrXoYM+XGhrToevY8HsffHks5OZAbHIEm+ZHu5zYrHuxnfS8sneCuPbaa6+si5iOMQ0GBE++2JtXIz1Sp3yby7s9ni3d43USQ038XslvfvObfRF28kXpVgtz61Zd+cUzb93YwYEPg0ZPbHp86SN2dOnYk9bv9PR067ip8m3NbgzKj28+uvA6B/qVR9iR35dxY8Sf2Oyybc2T6flrzenzXZ3ChQ+z3Dv/6dXXRQvByI5Puu1Xc/uTX2tN37nCXvWqn1xDsPjp69OVv1jIKw+/IWL/OseLgcSAC037/YNSmClOYQAFUngnD0l4KWzcyYRe9ul6NeFRTzIHFVl3SB/60Ic2EIkixUPiAFqiii2ORaHbhsdTnDDOjSFZ9mQKO3XEwBMvG7g0/vQOMAsknt7vZztgFRk2mLwKkZ+DtcLy2RwPDv7Er0+XrFzZ8ck/3PjypUuGRz8dtXT36aW2RxZksNO9tO406SNzufqyzJ/85Cf78Rqfvn0Ufnq1WSt2fCK4zdUdJhvZ3Lg43b2oV7bsZ+0Ph8N+lObiS6ZVF3HUzcZPxrd8Wztzb1LqXQjEMSZPhy0eMoadD2NEz1wNnJwjcjJ700mldaFHJpaTnLtS9nDS9yrURzqLVQw1csLxSNaFmi07fpDaIT7Y1shnzejgWW+v/K33LbfcsvWtH3x8uKCKZY3k4DGfR5S///3vt66aw2QvwyIeuwg2dWNPhurtSXWsFvCZV6N8wEkHkWv2hWPEhcxJjK0TsB898k3NsPBfzWGHy16Gh364xBNDK4axONWxPWnObzcv/M1jq9zoOal+5Stf2ecYdeg8pRbee6PjePntb3+7f+2PH3puKsWWM/z4Gh7/xp2j+GLDF3068qJXDvKLyCO2dOTC1pgftnwUr/o46dsb+OrhmITjc5/73MmHP/zh7ZYtHxqZuRzpunA4Nton4sFOp/1hHo79HgghElBgCnMzUEaK7WTkynbXXXdtQBIDhA3iy0s6j6L8ZGyFPV1XOS+VAEaKztbGLn6bki8LKSHJwEJHwWwyPlsAutnBji8PmOBmZ4xPtwOUTXZhpOMuz128Vxdi83FYJz93KnDwhfj15pmP7zl45UXW8+WpRx+2eOZiiY9nzB8d7criLLzqQwcWNXMy8BjIycsmzo+6eDzo4JQju2xd8B0M6qlVM33xYWI3awiH+eTbRHyoRVjDWw5qYW3pOdnajF4dwWyDW59JcvOpPI/+yNSFbznxWX3k4WQJc/tTbGQepnJqzhe/8qDfuhtbe2SsqYlWbtXMheKzn/3s1rcOyCspr6om8c2PE6YD2UlULeQQZvrlhQdb8fR46ToOvPryitcjSusf8cnWb5u4WDguxXPC9GuUfCE14wcuccnhrB7qb2y/yLf81EYtELvqmi1+2MuxOrdnfJpS/LOzs/3YtHryxQYmNnrY5S2+Xkx8edA3t/8QPgqXMZ3Zk0XJmsuLXG+d7FFx1BS56L344osnb7/99sZz6fKNmdzVKftqzD9bfdhgpa+OcpWT40G+6dLH14xr/Iqh8cEXHpzG/KhxFzxze85v3nhVRMYuvG7YvVIyry5i8eni4SeRvUrlz6sQ5z/7SSz58Od4LKdw7kdYBaMIsOQkhChKuASdLD3n93JPUem7OwJQkbzh+eqrr+6rtZfweIAA5fGOK5yX4fxKRAIa0uOJX4Fhw4ctXPzBg4+HLFJ2U1/CMLLXxMUrNlv52RTu7lw8fCDApkLiyMlmcuDZ5HTF8jOtFy9e3JuNb4tjAWARR4/wy1evhRvWeMZwlRvceOw1r4xcPJxMbBIHo0cVLhz8vfbaa7v2sKtbd38u/DYXHzag1oEMd+tLHsEEB4KJHix46ofwkLogGMgP64Lrp2udxHx1uguKk5SLqzvzchYXTq8sPBZwARFLjflWZ/5gQezUtD1qLqZGPz4fqDqTi2NeLlNWXl4Fyc3Ngpf83lx0YMrP3jhdN0F0I76qEb49bk949S0na2M/iKW2dGA0Z1f92o/mdPR44qqRDx70ftGsB7wX1/576aWX9nPrTmyOOTXUxEPlzr8xah87YRcX3xjR06ovnpw1OnwbI3p48FUXdtYUz8XXOcMNhPfp2HpU4li2NtYVDn7gVy81IKPLF157VI3wNPHMtepqHH7jiZFPMdj++9//3r9M6bhyHkvXnvWK0znMcaWWLt7VBg6x6LGBVTPmn8wxJgZeWKsZP9WWr7DCyZY+HTJ++eLXftIjryj8P4ra2SsuEN6fVuOrkYuitxD8IizfasDWq0IfrvCq0LnCOqGZa3Wp1mHfPyhVwpiMgJGADYZXschchby54q7HXaM72yeffHJfFAAC0sawuc+vN6WQg8DdupfzFsLJzIGJ+FZUAOkhIMn1s2AtkmLCYvFQxW6R8OBGswj8mdMzRnKHyQXQyzzP6GFyAvBKymL++c9/PnniiSc2TrnD62QhD3N1svnbQOVSPDHi8S1PGCwePG0YeIzzw2+bRz28gvNPTC4k6msjeJb8s5/9bGNQK4+0+BVHTJu+9eXbGslXffjWiwOThkefHvxsqhW9DgBY2SbzXgxc8vnIRz5ycrpOtl6B2iPWyU2HxwCwuYCxdYETj1+YfIIM5tZHLPp0u+BVU3r48IlJ11ze2dPFo9teEa89IF9j+9RJ2slMHDc4Pv/vYGqt+HVBRPyLW5xwvvXWWzvfv/zlL3vfihtOWNRUw2t/s5U/f4jetHPD5ZhyTKoRkpPY9uivfvWrfVzBqlbWTk7y1JAYETv27cnk8q521YwfeOgiY61x+sWj197jjx+5eXQmJ7EdL44zr0rc3cuNPZkc2dvHfFsPPfzhJi++momhtR70YOQzu2LzVW588OU3Yh5//PG9Hp///Of3HqbvZOxTbW7GnPPc+FgzPopfrc3tEzL58iu2+uPDA58GFzLGFwvPuL1pDqdzCrkLm/3pJsxNuJt4/Avrzf5PfvKTu17yRh5fvvnmm9s+v/hw+a9/n3J1g28d+AgvXU805CDPsMoFFvxyoavh7d8DsdgMUMIWQpAWggxZ6JI350hj44r4pS99afvxeMfdnEKTidOJVjwL44ACTHH4FEtMcjb02ZGh5GQlRmaM6vHyV7LisNfo4YvlSu6ln0cEbGxcF8hnnnlmY3eRdAJSbLnLFW4Y4LbweAhu/uUsHvxsi0tPXHKxxStXY7gsGhldPGMnXQdWJ0KYHnvssf2znDY7Eo9vMWFD4hqT8Q2rkxK+OGKko85wyRPRN2YLB994dMzZwq4W3pNxZwmnE521hRuJTx9PTBjM+dH45Ud+sKhpezL5PPHymW/YwgebWPzpG5vTMQ8zLLD6wIa7Nq8625tkLohiw8OvTwx5FcoXskfaC+SaudqqmTiotSWHWd7tFzK4xGEDHx901cla84OPh+rzo0+XP3WL6GrsyVD5i4cXX17hw8uu+OTx+IPZnK45ys5aqaXc+KyG6fFlj6g/OT9sqlnHuxhs8MPBl9acrTk98fg2ro78IzXHJ2cLK7vWyitgN2Xf/e539x14dfVpN+vgEaEPovArN37hE1PDx2MHP1xhEx9fa0xGX4OLDCb48DpO1Ml7X6frhszF9pVXXtkfJHGTTo+cLjuvJNyweftAbfmko7byZ6957Co+G9jp0XexVAO8MMBLrxitK8w7P8Y1RhWZoXFzTuhJsg1vASTg+RlnnnG7E3In6cLRm8zA0bUQFZwPPM2CooqKZyxpByrw9MUOD6zhnr5dPemz57cxu3wqghOGT7G4Qzf36oK0jtwAAEAASURBVMhFg08F94rJFdsGY4c36wMPXSe7MNG1ecyROT15tGH5wINLLdo0bPKj1tXAnZoF9xLbYxQnWTYwu0CbO9mx0YotjvxhhF+sCB4Y6IQtez0b+nzRQ3TFJeM3rOY2pZfN/t9EnZCNSIb4lC8ZLLDzl24YrB1dByP/4qgvvQ509mTw8aGFSc3Y4xnDS5+cPp6xNXdi8N4C33o1jPjXPI564403do08JnRnx499LM68I4UVRv4cE+Za1BrIFcGCYMSrVuIi/vH4c/frlcZtt922Pw2oPnB5VecYs0b88dO6iYfXvFpVP3HEKB65scau+PBpYsTTa/gwVm+25nryak/uomzOxtzNGrw+li0/x436sTWuBuYzFzU159+eEo9PmLX08dUJT7xs2IWRLjls7uq9Yubr7OxsP4I1JvPKX/7OGW7cPNLyXnB1p8evhtTc2NqJFWY4xOycRNecPV8anNmwg+9wOJycX6+SNedSzTFkn8lRDb0y8egULq+E+YIhv2rNxt4VHxZNDPHVnC6ZedjkAlM25UoXtv0prIpOiaOA6zVAKnqBBcDXnGw55scmB9JFxOMWcnOJ2igSESdA5mwtlLE4ga7gO5v1hx4/fFYg+hIUO2zGdCRYgeiYW1QvCb2P4JNVFuPvf//7ydnaND5e6BEQHf6daKoNbGrDB+zhJEewiSUOWbjSNUcwwiZHNuJkwxc9fbX3mMx/8n/hC1/Yb/Dzwc5L0O985zv7RC2Gk121ZcuHJpaYbWr2KKzVKLzZsmOv3vDxIS6e3lxcdy1ewfXcldzFw7NjG9n7CIh/+MTjj38El7ryhcTjg/9s0slGz0dEl02NTDNH/NTEssZqemE9AsimvP/1r3/txxr2o7tSDSb47GN+kMcy1s5FpFhhgi+ssOGzq27JJj64EF1+22/m3oPzD2uHdSLxcXL16BGykwh/1c1JuFhyyudxjehbC0S//GHiT1zNmLw6hR0G+WjwkhsjvthGHgV7g9anGe1but6ktXfcpNkriB+x+KZT/DDkkwzp6eMfY+RLTemwb9/Ra+zmlK11VTdz7yfR95jNDVGyLiJO0j2Ch50/tXXOCI9eDcTnS12S0Q87XvujXu4w8dea2AvV2nuL9q2aeUXk5s2rYzcUbiY1x2J7Nr/8yREfLvyw8S1uuuKrUXHVkh29dNji7zfRKQNbkxhqYfQMJe9AaqE5RO7eBPPIBzhvmPWyi5y95/X+/8MBas6WTQUTE7++BCWN6CHx6bRwsKMK5oAq8fy1aO4y3Pl84xvf2F/xwbbCyKnnnfzBV5xdqKXLD56mgDDS40eviRk2foxhrn7mcqErN80YH/HdZoTXozXPYj2Xp9em9EbXAw88sHN1gKqrZ+JiTTLnG96w6xE+vBp5jbzcsm092Mm3fGxoFzcHWJv+0vpwxY9//OP9PNbF2oHZWssBiWGsVUt9MnHFqD70w2IMc/uwA61c6KUr//Y13+ZuGpwYEJt6WDwf9oraBcSrP0RXzu1BcauZcbkZ819ubPDKmU3YiinH9PHUgD5d9TS25uwm0aPDng4M8jK3p5H4dK7W+JOPGDCqUdjgaY3j84GKSU5PXMQXnsYfGRxq53ulHn300f1q0n6GE7HV2MinE5sazprxBQed4sKKj/Tm5WmM+OkYYw8jTGokb2Pxy5uu84eLtfdGvv/9758c1kUbyUMcNw5y0D/11FP7JslJG4WHfz5bR7jFJyeTAyzhJSejbxwOPuFkQ15e5OJ75dGrkG52Zw3557NaslcTFE88WJA4YSIPL5460YXHmBz/mvUmzEOepwnMESe1Ctu8heaYrqJwgvAEcNfpxOcVyPn1kkuimjsPb7KK87e//W1fKfktoRLHQ8cLYC6+4olpTrcCtxHI6cHmJZvN7GRA3/Npdz5eeRzWxnA34dMHTz/99H4UQK8DXoH4sXHEYV+u5Y6vVWy2dDS10KsLLDarHr9awU4H9mR8OOHaGHfeeee+GPvEFT2fYvrpT3+6T3IeXXjTy+Mt/jz39AYaoov4Fw8+daanITpyIg9TOWejh0eOalGt6amBDxK4QHj/w119deDbJ1ji+69je6Lahk/OqPWCzRiFTWxEpk7hh60c6LZW8jTnm66xni4b+8wdmsdWh7UHyLVyth88GrJebOyn1pgtnjksasAuWz0c4pUbXhjkUay5V+iHkb0mBlt8sdzcuPnxCsTcIzWf8oKPT7U3Tj8sxQvHXCN45IPINQTbzJsNHspfdRBPruRw06VT/o4pPOt/ad1Y2KNuhpwT7B+Pg5wznIT5rFUfvjTE/+TTDbMxEpc+XTJz66anAy88mjGeWovvwsbW8efmwYXPJ5PU2rnMGiC18ajeY3rnkH/+859X8ifnz7Hi3AEvn+2dcNErL/7UCYXP/mOD75/9vLfMF+LX8eWRmr3sY7g+4i4n+Th3iMN/5yp2jg2x1IXerJ0x3PhqpSUPp14MMmP+rlnvWzzk5VAFFgAxBhgQIBiat2E4EKwDlg5dL+m95FYAm8QbvPj0vGHpztnY3bIDwcK1mHNB+Ub8iqXBSAely3eYpiyb8qg4LmLeS7D44iu854ZeCopp4dkiODX87qL5QS2W+XGx2cMiNltjeSD1xZuyas+O3Gb1Ks6buy7EcHqPxuMU73u44Kur/1nxctoG8nl7r0Ss0TzRVxt4iwlztYKlRo7ggaXcYYcBwScGuUdVLhxObDY0e34diC7UsFtvfI8DzGF0YXdgwMH3rKUYYQtPNcTHE1+z9nBExccjY2eMLxd1MRYPRjc16pcu/72nZb3pwWisJmI6OPnsAiO2OBNPteY3KgZb+nAgflFyGPgS1wlE3fGst/+l8YoPz2NXH8msJnLjg3+NDwSLGPTEpcN3RJ6sdS8+mYbYsDXnH9GHRS7iN6fLR/H06tYnyrz/1P/P2AvsnIjlC4sYbMTh37wYfNszYdY3bm35E796pA9zMfCsIV3jaoInnrhehXhM5aO+fFkDewY5DpzP3Pnz6XyG5h7jW+MPiQGjC4AebqQn01e3+Omrj/Oqfat2h3Xz48bSuYI/x6FjEBYXa77E1fOhFQNffcngw9cidSdno5e7MV0yuvJE16xnwQ+5wu7JElBEFLUAJBecI4lqzekWwEXESc6m6QRiYcwl6WrqI3zuoMRjy5fkS7oC8JmOMbm5JMyLuUGvPzZDvAqjCOk7YfQVIE5i8Cg+PDYNjLDyU07G8PCDYCWjj8icoNKv+PgaaiFsvA7E7OUTz8nKG+aeF8MKizfIfA2FRyv84bnzd+K2qd0t+YisVyU2k8YfPbWoJnCwhxPBNMd4rad6yQnRKQ81kwPs/LsDspEdWLCLq/GtuUHwD3cecblrc8fpYs0PH3yLxWc4w9Q650uPYDRmF6+8WvvqQB5+m15O7nwvrbth8bxHo1dDa39YB6b94P8XvGHKnk+x5AUnXDCYG0cw8IVP35w8fnO2GkoPRrrhNecrnhOVD314BaLu6ugZ+Kw1n+FUW7iLP2MVGx5xqiFb+uw6Fs3DyI5Mz5YdeT7SU6O5nnSau6n0PqmToPVwQeSzj/TyAbvcqxfexN/eSw6HcXO6eOw0+Gpk3TzQx7funRBhQWTikPEBH0w+TmtPy8l5wKspFxIXECd4+4seP3o+6PKvx5sUdvrWVSx7F1Vv+9B+dNPgVZBzqmOdP/V0s+amWG9vu9mUV+vEp3H+GuvJYNZXI1j4trfi0UXW0bHbuf+adaLaj7AUjGE9Q8np8RpzPB3YaGQdsOT84LlD7Z/JjJ1AJGwB3ZHqffIBKPoSaSFLFk+DQUKAl5yELAhd9mTp4nclrhD0Fd6dhZjwODBtAI+DXPi82UcmBuJXPghPq6gwKTwevWoVRnyxNWP6dBB9c/mqCZv8uQu2GW0aJ4qLFy/ul9FsbSB4DutE56TsAuLu5He/+92+w+NX7no+Ub7515CeP3qzkZmjamttHUj5cdcIo31A1x2ax1Q+hOCErPEvZ/WRI1v19TFEFxAHoQOGvfpcDQP7agqPOT3rXR1bWzxyZFye9OShZnrNQW0/epziDlONHaQOwC7aHl2oq3jIvuKjmk1c5GKSw6CZa9mHqf0S/jDzr0awzZN3e8ex0gXEHvYY2KM260BfbT1ygb9X9WLwLyZqzfWN+U+uby3IYc9eH4+vZHKWW3ng2y94sFUTvuVm//rfBRcQPM1e9xSg9/DKedZY7Kj1pNe5QzxNfHyNvfj0jeFRYydbH9ftvVqv9L3P6OJsL8Npv2fPB1sye8YNhj0Ou7yd0L265tvjLOthz8NGR88ffLDAKB9+NTpqVk3laT3J5OE48erNfrRP3Zz7Tji1PH/5FYj9Y+3hELu3JfhFrWM15Re1pvgIBnHp89m6ksGZXB7m+010Qo44ZVwQDjQFLTl6FYEdsBXanEyhFM2G8MkrSX/729/e/02r8ArtUZIFc+fnBA5oSbKHxVw88ZOLgddm4SvssMJpkQ/rBOvrwr3Bb+H/+te/7h7W/tHGolxYn2hQdCcPj7U8YnFCL1/xjJE4xlpjRUwmZ1jFyL6i08evPvJRa5ujXNnQcVLzfkcnaCdsfD7YefbpfRy1RDanC2FxbRpx+Ud7oVc9Wzdx1A9VV76rKV54G8PJH6z0NLWHzU2CDe/ViPzdAZlbD6/wvAKBT/v617++bxrY5Btu/szDAKN41REeY00+ckkuj+pMDmNyPhG+OObqA5sTsX3h5OCu2MfQ1XTud7btMWMxycVDrUm54NFpLYyReOnEMw9XNlt5/YFf3Pj8aYgv7y+5qBi3f9TOJyI9krYu9iN85SMWf7CLHZWDeXuEnL/yYIvHlt+JR3w4iuWVKD3nBjx+xLWHHF9i4EetOR/swklOr7wbw+HYEJesGuHPWvDDH55XOt5LsPZevbv5QunYs85V7uDtBzZkYYEN34Ubz17xBMDFmk+vRHoFYA4Tyo81wCte+ZezGlQvdSJ3DNmjLhRk+M5lbhScN/EurVfS1sMnIeXlaYBPkLlA+zCAvNTFXkLGMPCvNa6G5sga0eW7faDW5mpP34XyHAajHJcwBU5aXHKtBRO8zUSHM0WKJO/E5mrdnb5iIbZ8RQoBhwXiS2w6yDwSG1a9xofCwEx/+u2jry4OHtF5DOQu2Zx/hYXPRQyPD/mgaiCnfLIRk0xMuSRTJ7YdGHCZl4ueLh5benwZk7mzMCanp6H8Zc/GBvEJN8+R1czdvGfhbG1mPYzwWgO++Mbjpw2CH48uYtsakzdnN0lceja01onMBdmm9RHTYrnYqa+DwIXZ3Z6Tm5Oc+PxUD2M8sdmLr87yrv5kmoO0+pGHW3586L060xyE7nLx+eG33PiwR93QqB+ih+CSKx17WxwEEz4c8ufTWMzG+YcrYk+v3lhD7ORMn+2UZZ8vx8vp6en+uhWydOH2CtoNlEcurUG4yTVzWItDT/x8kRWLb/J8tfbtzfDiI3VSE+vDv8a3vegGwolOndnBoRdPa03Z8589PfEdj0j9329flGu+5SueH1b76le/un1WB7icc5wb+gc9ewWRlVv1gMfF0ReTkjlhzz0DE99qRq7B0f5mL8dqShdPbnQitXQzRtf7m/w4ZmCqrmrklbPjSG3vu+++vY9hsOfPr1cmfIhl/+o1pK/mfMJQjnr6cGcjfmTcxQym/XsgHFj4nFJqQ5SYZDnFlwRjMroCAt5FRNL83XPPPfuloTt7vr3UQ8bA5VsS7LWS0yssmdjGeOIZ86FvAWAz5+OHP/zh/koSr2xgs3F788uJBN17771Xftt4M9Yf/hHf7MIpZnHpaGTVJBvxtYqPX43yDR85Wy853UnYwDC6oPHrhCy+Gtoc5Px861vf2l+x4WtXrIM18AjQ5u+Vk/htNnHKofh6NSNTW/p48MiRvlYO2cfXI/rwsrXu8Gouhh4dksNsLA4ix9c7EF1QzPmAY9aYfjj0YaZr7i5ILcXOP8w1PK983HXaj3ffffd+fg2zWEh9PdP2FRb+L8iBx85JpRqKRV/fPoDfWH5kYsajFx7jGp3iii0PPPlXK7r50ZcrO/lHZHJ3ErEPYEH8+SoeefhSTRcTdSIngzc8/Gl4YSs/fGN5lHtY6KubeXrwyMFa4LlZEMfeLiePN33s3FMBN5RhhpuOOE7e/CJ54SO+w28eNrphJtfw4HGOgsc68jX5fJjDemndwf/85z/fj9DsSXYznpMlf9WOL3G8Umk9+SK3HuXreGanBviIPt8averL3pxfsvPr5O+RmpsvH89/9tln93fKkas9v/KWn73qnOY4I0dujv3nvBvL1r89SUezf8TiT+MPBljIw8ifOhWTXEz5WudzTlAS0yjORBhzTBkAYw7oBVZwibj7lETfJ2UTu3A42XXQ8h1JwGIgC8dvc77Fow94dnTEQzBIUg87fThsQu+vuKMkQx5LeN7pLiT87h58iskG8ajNj6h4eXo4HLYNPfnyq4mlGcNHLi6Cz1xPxk5sLf1qV25k3iz3vVE2owuBr8pwR2NhqvnputP85je/uR/FuZP38VNxXDDYwN2bazY/qkb0qh9/c53FJ49mPjCSl0eY6RjrbSIHh7XtYOFLPeHg+z/rqxXoWZcoH+qC1BQufDHNUXpk1Q6vlg0/8hWPjK5cvc+F570tByKZj2t7TNEdlDhuauwPe5Uvd3X+Mct+th/5y3+1FFsMfJjx+Vcb+uKWH3my2dMji9jwg4/4QtXHK03vyzhuvIfgDXT/W2XsQozYys2HRPyTqRORj3ezJSu+uNWvOPzSgWPiksfkyZkfOmz56QSpJnh0+NKcXzxG9vFzr5phkidyvuiHpewV2MVCbPnmrz2BL64GA/msMxtz+4083M4J7tS9T+iYs+bFcZPpUZRPY/o48Ywlttz4FNNedkFy8+kRlnMcurQuQr7/zAWbP/r02PIhL7WBq9qY04MZFnFhrob2o5sbPM1jbeRGWP1gcd70iPgHP/jBvlESB4ntIuJDSnx00eNHnGpL17gY5moKDz6CHw+Jayw2G3Xc34VVkpQylByDFpuOcQEVVLKS8PzP7wjbhL4P6cJ6X8FJUJDs6fPhZO35fkkqeCccckQXUH3JJMsfPv/4WkXQ25ROah73wGSh3YnySc4WX2/xnYh9yZg7fXf/+PLn11iTq+KL3wkRvvzRR+nrJ9HVkI3Fn4sBXDYVPORiWHAnZ3Wx4Z0AXVjwvLHugHCC9Mkr+OUKRxu2zdgGFVPu8hFX4zv81RQf7tZ45mDMhxOCeB5FOPGKabO6y+cTDvm487EnvArwSiviW6sefLYu8MGi1+jQJYfNGA4xyOVnHi/fYcifPWZf4tsbsJE5ELzMd5BFcvGRaHnJRy3pIvZiWH+8/JPFs3Ya7HDRr+bG1bz8y0WPZiw61QcejwbtAWQfOI7sGzmI6abI/nUicWPipOgbChxz+OKHW18+rT+/eDXzbOSiFvCUV1itg+MZX7M/k7kQu6Hzu+IuHu1Neuyshy8rlZ91KaZ84LCXwpwNDPyg4phXU3w6sGv8uHGwd62FTzjCgrxX4B8DxfZK3vp3koeJLb9qBK/j0WNYF+nwetXtAuL9tJ50iEUfri6M8CM+jfltvAXrD9zZ4VlfPyXgWz2sIbm9xk5trbN/5HUOVgsXDW/kW49uLPLJBh5NjmqBB4c60YOLH2N6EZ56TD4f5ypMyZlzyMCYgcXkTGIzCJC9Qe6XCp1Q2DjZ8YH4oSdpB6wi+zgqUgAFV2CJVBg24vEhSUResubwmpPTR2wU4qX1dQQW0MtAJw+JogqVrTszd5v+o9TdENxk/ORbPvzjZ9cCNw8bvRbCGF+9Inbx6TkZ4DnBOsB8OqQ42bUeDkrPtb2ctWEtpgsemnce/Nt07I35h0M8ZC3xNbHww4lXvnjm6qsefJJ5Se0GwZvhLhBOYr7N1AcTyFsvG9urUQcbWySeGw550xUDPnwknrUqd+uVDE5zvozpqoG5nNhNfTW179SKP69Iz87OTu64447/VQ/+rTsduFwI2Ygh71kP/uUHNzs65vhauMjZJa8m5uUj39bIuD1nTE/jZ+5dftxJwwmj/vnnn98XOhcRr1TlguBRF37CJl64q7u4aqlWZM1n7LCxMeYvX+ZhFZed49hFC59vfRdjcsSX4985QVwn42JXI3GQGuA151N8cznyhVfNjfHFst/o4XmM5oe2vOL3ha/OV2ROtG4w3M37j3k3lGHfAC7/+d73vrf3vZM18kk+evIwtj6tKb9ksGnm8zxn32rqXk3oa+bWl52bNd8M7TGrGpDJRU+urs6h/HsF5dGVCzK/ZHTp0VdfPmA0R3q2qNgTO118tS03c37lvb/O3YSwzcGB5ASzeAUBygkk8v8cDz744H7u6gAV5JjYeCzgY3oK7ArpZCNeBRBXDCd9vQYkHWQOIz08uMhRY33NScDLfTY2BwzZy42excR3hwaTuzf4nZQmVfziVY/mdPk2r8jhFQeZt7ngdydhg/sqAv48hrAJOvi30frDrzsiuTgQ5OUuFEYbSzz4+BRfbnjVobriV1M8fjVElw9z49Y+Xrnovap85JFH9l2YDx+ooUdE3sBzQYEhO5vXI7eZk4u0u3v7wQlDPDZ8G8PQ+uKpGX56evUSl506yAdmRB9fs+7uOj3i49fjS/jfj1yM6Xo05OLDl1zEYK9FagwfrPjHNcNjj2Cx1ub4jis8vvHkFM+cX3I9HWM6Yslb70Mh1p/M/rUvTk9Ptz+8sOrNNXXjr3rxOak9Eo9NeeYzbHT4Nu/C4AbHSVQuX/va1/Z3jZWPtehVqLga3Pa1T42xEUvPJ98uQuzlK/7ct+zpaVE8tuz0/HVzIHc8r8pcIPh10+schuh5Ve390/vvv3/74F/sfLtI29Ow+hShR1bq6iZa7nzQZcN2jsU2F5dMPvC1v/CsAR0fgvARbX58GKjalevVevV0PvHxXk8nrEv49Rrf+uoqnhY2svaI2AhejS3d1sGc3f4Yrw1uwShUaAqcmDvwOXeVduV24qXvJOLjpN4U4wwwxVUUBZaIjXVx/R/DL37xi33ScJXvQKBfDGDF0AIX6IpeQvFhM86GL3PP5tydeaUDq8V1Am6j0vPS1YLjw8ROnPIXK6KPKnQnMLpqhvQTyxwnF9+GC4f3XFw0+e37xGBAFkodndC86jCG2VqpsXjswlQtYJ31MudLXGO4tPcjtsnpi5dPa+7jkA46NRBb7xFFuFsLMauNWJfWozab229JG3d3VA50+GAjP3uDr+TGeEiu/NNDZFp7WH3J3QXbA/y6W4KVj2pQbfSeMf/yl7/c+9UFPp/F5I8dPNXIPAztGzz+zFE14K9c8OhF+M2NrVVz9S9ma+yRLDmf9gS89pHHKMhFHQby9PJpTob45RMebeKgI+fySh9fo4v4UyPHuT3s49B+V9wjHvFRNnQjH0d1Itf3uDuM9NlWZziQmPjm8BqL3b4pH5j5KJ759OW85Jhi75UyuZodDoeNm5+Zn9jiVQ/nNtjtZceli7lzjJjFYWMshobY0wlPuKuhOR219CpCPfDcbMFjf6eT33o3ZT4B6bxnn4vRcVAu5cUmLPyJ3/rDaQ4nooeKYz/imaN9AXEQcG4xBHZwO/FScvVzovVyz7NCG4MD+vQ62DjjR9JeSnmeaENLRnFdZAJJB5HhOYkjoAPWXJySyU6CFYM9uXm2kjTuhAK7QsmPXjgsFD24ySK6+PzOAs7C0tHCQZ9fOrV80MOr8Gps7A1oj6W8XLVJ6OPTL1cY5UMOTzjFzS/c5tnyH9lECDb+wwQPCj/f1UXscqDDr7g2p1rlkw2adTIX35ui5ekgs7ndRMBsP8EhBr96PsJtLAZdLYzwqUW6fEysM3966kt+cd3AeNWLqquYYti/xh5r2PPGdIprjsTWyOAxFh+ZIzzYInL5kZevMd/02qfmiL145vTNI3P1YWcv0IFFLflxjPlhMTm4S1aL/PFh3RD/4eWLX/Nwkxcr/WzFJwsvuQuXOjoB+zVPjwjdwLVH6ET2co8HPT6GzzmEv1mjMJKXgzrSCx99DQ/2bGBkoyZkzgFqgVoL+mR+jM2Jl41zG/zz/bBw691Eq7ebEhcPd/tqjuSC+G2NxKyO1dYcH251o6tOer7hC7uxx5UusPYoHRjgVjP5ssHn1zGmd06l03GKZ65nU/3Mo+qnb63Fp4uyhxt+Lcz7TXRAWpCcUPC4ykdx/QcsYC4CV3s51YmaH/8B6c1dd5uSl6xNrge6DQFcQMQGXGxxASYPExlbScHBLh0J8slOq4B06ZAhNhURHuT5KxILdr3GBhY8C2Os8YH4RXQb861NYlN8ucCE4LSB6LP3PoI5svHpslVr9uIgvVqQ84Ho8UNPTzZzwBM3P+k0p6/ho8b8Ivjw+LCR9VprR4cvhE//0nqF4eC0L+Sjhm+sr5t28B0Oh62Px46+XMRv7iTowIKBLEzkxWJnfZA1p0/WgcQWRrVw0vLPV7DQhbOLhTn/ehccmKttsfjA0/NZTcQOm3jVTC7GZBoKizlf5cKXBpO1FYOOHiU3dgyRyVuvwZuuO+P+UY8+OXtYOlHh48HIjqzcyPBr7MnNj/GyxVdvJ13vdfUVJfKQz/TH3vufPloqvvcjHH90kVhqJJ786BcbPv6uVrdyoIvaN/ji8Ftd+cVz/PPvAqBe6cDjsZZ8wmFfuLnwuM2n3jwu7LGh91fV3w0w7O0NdREfFmRen0zd6NhzeBqcbDQ4nRO8MiYTB8FjH6sFvO2rbPERH+zUTX5iaYhvDZEZVz86+eCTnA98NYMJT1PPfQFhkILeweU9AW8a+Vy5f7T6b+SE4SSokC4gPi4nMR9FQwLzG9ASAECS7qjomAMGj4TY6Ong0zFmb06mUMbNKyiZRlfx2ZpPHLCRI8WK+KSrcMVKFgY4ioGXf/x05FEuFp5+GyG/8HoMwc6mkgeMNgkcLZ74xSwWX+nzkw695uxbX341cnZ6rZrkKx0yBCt/MPmEB756wSunbg7sG/HsBx/b9CyXHv2+/sHYOoermPjVEY88GT6ig9843/DSJYcTJrz0zD2HtwZ8hbuY5Ph8yFE+DhR1Q/jVQEy+4chGLy699jA9fuvTnWvGp8Yfe/3EXU4zFkxs+OO7mB6jPPzww/uRsn2DZuz8y1kMfjR+kB5evvlE7Kt5Mcm0MKupuZOs9zM8chWffrH0yPtfnkzYJx53qbVjXxz6ET487GDMHk8TGw9eDYXJmMw6sJU3G/7FmfHwe4LiwuaiRu/85U+LwSk/FxmvoD0O9bjNXnIREjMcxmLJXVx++K8W6ZJp5UDPuHrCT84ve7HoqEmNjj0hHnywOPd6hUdXTcRjb8xXVCwxyMxRczx2qNzM+ctvmB0n+010CikBBoxCuQDYGIAFJKcC49H1xu5L65NP9Nh5CesgRQILKFn6qOIY42kSEJtuCZfssV5zBeXXQWmh5MCWHzI8Yw1f49/cZqI/aRbdGPGDKqBxRTcup/T55xefHr64sITHvLz19MMDm1o4APhC7PgKP33zKQ9fOuT55SN8YaNXU0Mtyi8eP2HEdyBpam691ccrJXrmsCP7wKNDtnLpJOHEbCwnPloL9hGM7QcHCB+wIny58OFkYGyvicFHmPkl06oXWzlo5cQvXTyY6PIRBvJ8NGZbXcjw2evTaQwzfXriG2vhEEfcfNJnSxeJk085GiP6as+PmuKrt6/TcGC7eHvDmKx47KpRPuPho/xXh+Z6seSB+FQnzaOVvgY//3SdgOXGtguyCwdea8FfNbQOiJx9scWCRzOmryc3rkbZkKHW0ZxMb9+wgQ0GPvB8iEV9n3zyyf2ehhqSt7fo+b8xPtnDOh/p2svFEDss1YiNho/EROoCP2x6GFBrO/dBcekhvtmpFztjOOuN6dCHN1k88+oKF//VmI054g/hGWv05LA/xsshZxQ0gPokE2Cc0+GQHjkHyMtAL517VjcLQs5WEbJlL7CrpqLzLzmtIigePbZasYzxKwCfGjkbYxuDL0mWDzlbfGSuwQEDO3I42YQHvxNYG4ScbzJ6fOCxRy04O7Lpmzw9NuWR73zyK08NzhaNPH96VBwbnR9+yeAQnz0ixycPb3mYVxM64hcz/Xzwgxxgau6i4RERMpcTskfYIr7kYm5MB27z1gVWMrFhgIdO/vCqiR7hzRZmtvaSGsBkbtzaiCsejGqGHyZjezgM8IVBTHJxyo0eHZjokRnr0w83TB07YWHDFz/ZTns+YEXkWjUqLp/W4Y31mNAHQnyowSNkF3rP5/lng4zVo34z1x9ycRE5gil88IareqpdORvX+oSgOroBZZeeXpNTMdkh9eHbnA2MGhzh1YcFP4z0Z1On6qZOiC5M5UeHzI0IMnf+8KiVThcR8TxqIg8XO80rBJjab2FiT1cucMjZHF9jo/HBltw6GtMzR8b49NTL8ce+fY3vHEzGH4zhrJ75oUvGNx/y0vDMxUL8GDfPBn/myXb5em/TK6wxYAJ5JOW/WCNJJGth2Ci+q7dPhuCz1QBCCipwcwXFoyNpIPh1sjBGMwG6SBJIDHK2fOCbo5KuOMUuHr0WU670xUyPL2N8/ZyLhcceX3xzfvglZ2eO6Jjjk3cCb6GLAavGl9rYwK1DOvyRa4g/bcYmo4+slfgo3GTZxSPPBm/Oi9e6kbU+DizrhWBgy4+YYqiRJmc5ITrJ+Tbn29rLuXnx6BrzF0a9JiYfngGnp0fl2Dxbc7jVHyZzPpB6wWvOvj1BBwY+yBpXB7bJ2CF6xYwnN2RurNHhL5xh0TdOpmdb3nzxEU5+5OUr/13Mb7755v2jYz4C3vtobLMrPp98hxmmYpUXWXkkz8bJxD7o+KVHHx7NSVddkfNEmPXm6u4cUj3DWP56uvmjB8PERhYecWAgn5g7VukZlwds5tVDfI+T4bJHzNsXbLOjP+tiTIavpQuDJgaehuiEWwwNL/9s1BaVu568Y4XMHJUTP9VMXytuOPNVPzEb51NPh10+tnD9SW9/CusYMGVgvBzmoDlHDOlbTL2DEklY0SXFBrXg9PD5tDj4rpr4dC2SGHzxoSCouMYlSd+4RTGehWJTYclQmMQnD39+6MDAN2wITrriVEB+NHM+HDh6c0Rmzo5vvvgU11gz1vITnmKzT6fYePzl3zx9Gwrx02afGNhoiF92+Qk3Phv4xKGD6JW7uRhkcJVbscjoti7qBw89fHoaoqO17uJn64TU2J4SL9ti8Z0PtuZk8coXr/3Eb2sMA1xdrGFnKy5/aOaeX/7I1aic2BafHb44iA9jPVt68GjyMmff2m6j9acYZOyS843Y5lOvsfHNC/73xX93e7bvUfK0zx8fjck1vvnBR3jm6oRg0MRG9OFH1ik/jmGPtenO48Ocb/by57sY+uIWzzwdYy2ZXMWfeODgB007euzCy7Y5LPy60QlLe4QeWf7Y45HT1fMj1+KRlyc7empk/+BXL/uOPz7YqBNf7Xe46Oj5YEs3LOKVO4x0xEBswkNmjOiQIbpTBxb+YCgmHTaILh3+yGHGMz/HMDAMOCEoOGVEhxNEphj0tficGwc6MHj80sXjiw+FoxvYCpl82vPRnA1cJQJTGIqhFzNfFkDM8OFbsAhfK//4bMQtJ36N9WguujkZ33KCr7hkbNSNz+pBvxNZcjx+i8OfMX/hpxs2vhvT4VtD4meTX7r4/GrZGiPziA2+eLP+xSeXq7w0Ok4QYooBz7SnQ7/8is8fvmYcFv41mDRjZMw3Chuf8qZDJhZe66qHr33n7pcO3fA6mWSv18QSg2/j6pA/9vRgJ9PMydmxyQddfvR4dORKP//JsiHT6PGdD7m1v/ihLzcncXtK7Naf77Dknz92fJNVy+KJxWd6W2H9YZMuO03uYnk0jbw6pKe+Yqt1tWnOd03M/OLxWXx8GMJtTp883GLT4TviB9HNJ1tY4xeTL3pw69Ux7HTEwTfmi7547Vfx0fHFkQ1ddpo5Mq6uej7xxMm/eXx29DS4rDG/8rUP8qVGYYmnr578sDHXyFDzPVl/wozPX3J4jGFD5xTARGBGAuAZUxasnkFjvRYAQSq4zcInH/j0BOVXgVELhOegRTAETs8OwdTcGLEvfsnCQh52NuZasvxUAJgReXZhIsNjQ14eYVCbuam3o/UnORlii8LDDs8cjublb86H+PTgaRH5Iau+HgUYd8dCj1zjv5zMO5nywYauZqwVn11Y6aL86WctyNjCzFf1Kgfy1skY2fxww8yXxi9bvozZ8EeWr+rRHWP1pZ89fXma06eDF0Z58YvEU1vUK5RqJD4fiB92CJ8dTJq5xn9yYw1f7Ig/fuzd6kuPP+TY4JOsxgdiSzd/jeWSLj/m5Vfd8OmElT8yvrQpEycM6edfTHJNXc3J1Efd4Mcnx+cHFn7M8Ytn7dn1Pw7m2eGzCR8/SB50rJlerHyT53/qlYMewchvOMuBjdbeyRddZE632MZ84tfY8zt16KkBGX5zPPP8iyvP/IqZrjj0xEF8ITw+sgtH689O3fkhKxaMfODRST7t7NFqwYcW8UO281r/Sf6Qj2YCUQDjghsHuAQ4ElTjmG5gCsQGX9+GcHX3b/90AUxHXL5QOPCyx29Mr42TLlk+woWH8qsPP4yS3wW4XEi6MPFdzk50dBUr+2oEewtBFp+famLMNmzmYdDjTzu88tTzgyeWZl5c/Dakg6L/QA2T3IrLpjrwyxY5aZJVR/4jfszZhYOdeS1dfbpkYcbTUFjMYeOfntaBRAaPlh1/bLPTs0kuH/jpsUNh2JP1h071nLLWnx45CpNxNngIZjbFT5defmEw1vJpHMZ4+ZBH+ZZbvsSLimVOjvhSm3ylkz053/h41awxeTblRqd6579YycyrXZjLr/3OL57e8a+JwTc+3+adqHxTgIsIf/YlfXr88VEcY3Z6vlrz4suNb82+yra4sLPJji/EP1sUT2yYfEBBr4WH3+KIzb95/InXmE/+YTJWP2OUHR/sHc/hZjsbXXM4erXceQoOMvbeImid9eJpiI78w7+Z64951BiO/BhX53Ds6gFOKSI0Z1ChjTXBc25MrwBtADYVTJISkuThcNifEHGy89lqfJuFP75bcJtHgYBFxaQzF6q49eQRXsnOHMTIh4LS0copXzCIGw7y7MTR2BezePS0Nge5uXrwh+hmhxe/jUXOHi44wsmGDnIH7yOR6mqDk3kDVaMvJl6YzRGf+Pwi+dEJQz0ZG/N84CH15APNHu4anPOuvjrju5FAfbIlf8VhZxzWrbz+4CE+4IK9eHQRzHKDq1z41/KXPR59+1C9WzM+yfLTgconWb7pwNBcL4a4/NJF5sUylkc8NuLmg17jcqKLxx9MfJNpfE19uvh05MlfeuztjYkFj5wPVBx+yOKba+nI2zohY/HChSfGjBPG/Fh7Yyc6e9gvLLqYOC+4yfTeApIDv+zhROzw5K52yfHFROUhBz6QMXkyvTm+NTZ2XOn5ZudDCH1ZodrT8wnVPl0mdnWCj208eBA+Hhle6y2+Zv+JZW3ytQ3XH/Py4kNjA1/z9kS2PgjQucH5AF6+NXUVL+Jn4sTnP99k/B7XmZzsHGVUkQkkGhjgjI832TZeSevpp2czCVaRXFRcME5PT/dXBfiaBQn53iH/ZdnH4MRF8FRoY40vmwtJGI9+mM3hQGzx4WUXznLC17qbYFNs2Ctocbq6k/Elz/Kb8YzhEIctn5rahZUcsce3oGTZGcMmjnz1eIhfYxsYFp+08RsnvuBNLnz6cjf/AW7s4GaPzJGYqLrxhRdG/sVhJwf8MIdjO1h/6GjkenbGzfXZk4lpjmxgcvnDlj2ZcXpi0mMbn8ycTMPX6KmdWMiYnG5NTloHE1/09DNvvtJTx+b0ih8uscRB4SALm3EyfbGKZ84O0TXuWFInfjQ4yxt+dvTwW990+dHoaMZskLjm8hOLTX6z0SPy9JuLiWCrxvTtV77Ewzfv2BHLY23+4ODDWPPKwy9s+ioR/zvka+r9w96l9c+o1Z5/OJC9LwbM/PBdjvzBQN847Ob4MBm3//iDhx59cucjWB1T4vuG7NN17hLL3P/XvPrqq/tbvOVZDL74RjBpyeiFaStc/iMm7DUx6LGL6GgoH3q48dXDDaR6sHPxhdf3FvoQhX/s9CWhLsxw2Cv5KYb4UfHUq/UsD7JqTd/8yqewJMwAkwHFFscYQDLB0+Uk4DmUGH2fpXaS8zUoFsXXA3z84x/fn1OXpP9IdfdBxge/kuM7XxUy4Py2EcIXZnZ85MvYgk/ihz7d4phr9DV+9b7zyw/NWFTfSeOrDPhLX08X0UfViG91QHTw6eNnAwsqrr78beTWoAPGwYgHw0033bS/wdcP9LiTy4+v66im4vGDiitmvttIsKFqzA5lAy+Z2DDCozfHp59NB5Q5Oz5sbrXw1e/qiaeeFy9evGLLHxv+qsMGsf7wI1Yx6KBjGzp08WcvXgewnMmS43eSbq+4qKqheGzdgXawkvHfetIRFx+u9j7/UTkVszwmTjr4eJo5fbzjkyQ5Kja5Ex6s8kNs5WbPkNPlc8rowGs/0Jn45Fd8Y774R+VuDAtZpBatQ7xyEaM8+XIHj3xNua9AubB+IgAe+8t36NFhIycx8gOXGHIK84xlXG2M+Sn3aheOsOKXF/2ai9v59V/pjjNYEJ6vePIvDvmpVvljj/htbF6t6IknNz6sAR/m4iRjg6+xQXzO9WTPDt8YvsPhsC989i5d5y7nYx9wMNfywy+MbOOJYz55MIWFPh9onz0oEsYsUYo5lxi5g6UCpO8g1AQpUT+W5Htl/M5FiXh56CWU75TxqsTFw4ErtiLoJVFhbSj+xLGx4OwESBdOMnrs8sOGrh6fnhY+OZHJKTk/Goz+4xR2PzxDV0y/I+A/ffngW8wI3nAex6VDhsJRLDzxzcPKHk895RPBoXa+fwcuXwEPl38UU0u6LswuKOpabvkSI3/1YaaD8OdmhIVdmMjh0OOzq+HLT/3Uxlyzvi4efsHNNzfzf1gb3B2eA1GefIihRWLgFx/fWIzGdLRs2wfi0pUfmZj2LTInR+qXHj+wqK8bHXx2au7gYwMPKmf9jE/HvPU2TteYTw3RhQUfta/M5chOPvj2Pn1jRE6PvRzoWW9fP+RxCzmZb531ap9tVLwwkrGHi10Yw8FPOaRjnh6/5umVB19qRw82+yB7aw+nH0o6Ozvb3/BM5sLtJogdn+XBd/nSM6/Rk0v5mGv069mQT3/TZ/UhhwGx+c/6QlAnZHiLEQ55qhs/nZvySecYTzHw6dlb9MpHLz4egj3Kxrxc6Gl8wZLdhXUh9g0ixRMHfjXtZkgc8jDOcbjKTcxwiRVefBiv/CY6Zxg5L4E2l3nObVpNYAcd/na2EjH2/K07i74PS0CL8frrr5888cQT+8TXJzDya6NZDL7gwdeArnDhlAxsqLEihhsfNvlki8feZqY3C8NW/I9+9KP7i+FcAJ0M3d3pybLhg0+EZwxnC0neApIZ0zvGx75FacHohqvaqwn/sPT5fhcK9fS1z/6BTGzvLXgMID9+OmmSwcRPNwDwoGNc8GhRc/Hh5wsvu/TkOR9Z0GPjq9/9sJdfXnRhQ+7o2D/33HP7Wwz4ZQ8jgt2YTljoqAsiM9foIPao2pVzPHJryCb/6kQPD1bPjl3o4BOXngu0bxF2sbMe9r346qixtT/gsN/YiKPxwS8Zm3LJd/jLB9ZyFss4rMWVR7myd3KQs/8+93vjh3VxzsbjIN8c6ytNUMdWtYGNPTx44aJrXPzwiyuWPrl+knky+NKtNub2sF/2U2uv9OXqlarv0PvDH/6w38ezv9U3ez7VCcEAE5lmjOrx2gfpwILwUbr2QDK6xUlPLeni03Midpxp7uiti9rRC1f67a3851svt4id3OyZsOejtWRjveClg1/MZP4PyO/yOIc51vhojzovOD75gAc2jQ4/+mrFXy1+OOjAmt05BcDkFBPVt2AZSbSDiH4XDzYC4rnbdBV0t+niUaB+JMUvZtnYHmnxzwf/SK+IKH+A45sbOwimfrpTR5GyT559ufGF2PEpL887Xfg8IvIDQ74Owibxm+O9imqzVFA58GGeT3XAV1s1499cDEQuT7zwHG8+OvyF20bw+/K+ctr7SF5teBn98ssv795JzDNPMemKJS6/8OHzicTkF9GZRPf/21j06fERbjbyMQ+zk5M7eneaNrRY7Iw92rQX6FSncPDBXwQvnkYXyaXaGWt8l1s2sGhkKBtx7RM+8fzKnD3rB3zcyUe+58kFWk3Fbk3gs3fZd1CKY135o0cHLnK8cIRJDLjCzZcaNufDOB5/1rV4fKNe7V9Yx50TReSCCLOvLHcB9NhIHvxWR7rihiOM+PTEwENyCE+y8tJ3kYBTw9NXI3PHvacSXkH7ueN8w/Xuu+/u79TyBEDOYsuXDh9i8oHMw0oXP6xiIvuNTljIa3x1PNLJF3m++OikLm/4PA7yqVVf18K/tcgPTGz5Kg5s5q1hOvyh8qFfzvmgA6eeD3z2SFx8ucHhnOWi3BMfFzo/4udxoBtN5zExwtx5lj9++NZgQHRRdWlMn0zO+nP+aBQ5bc6gwhsjgWw8xjYLGycufID9tsX/sHanrZYd1R/Hr9CvQX2gclpUoogDRsUptlHECeOUqDHDNRJNTIhDgjFGxAERJzSORMW0GAkGRcUJB7QlkCBENKghUcR+IPgy/O9Ppb+XxaH10X9B3apatYbfWlW7du19zr33iiuuWB/gWCiCJOMU558n+e+F7NsAXcD0+CwIY2xLGl3BwMBGGOlok5kJh08CqumkW02HL8U4X3QsBp/X+F8GJsKfcTm5nY4//elPr/9zIPlk6ZRgE1G+smes9v4Ewd04jGwpqLHixocZH49v/7zLazXviz3++8unN9988zqt9b/f04dBmx1t+uZL3mDI735O8RW6jenTgYUtRRs+uUtOPvSNKQhmG3P+8Kf95gU/29YXfzDgpUs27HhKWJLPP99hqG3MJoXYsr74cKLzl4JdfD7Itemyi9iwrsVIHrEhn/TU7LLDpoL04TOmoFk3DjfbiLz+JDrZIVeeYDEGg5uHf+BkQzaebTH4L6H4XsvY/MwRH+TYI6vgI/xwhkO/uMpBMmr2nGyzw25xwVCbjL/T5T9DelLyryHoi883hfwGvX3i+PaZQ3tDa4FNcnBqw5Nd/AiPTViQMW31lCPT9amtILaL0b4mr7vdbvki40/D+FcVnug8RcEpxnTok5v5rR0GeMi0dpLnXxvRKYZiYptOfe3i85HADTfcsPA66Dr0+PDc4fIb3/jGugatlXCWV9j5mfPOD2qMP+OIPxjV5oDMMZPCCMHphCA+oRTJ6APCqCQriGN/D8ujJ4DG8Xz677H0Qx/60NHX4ehLQK888sEOPPnW5xuxp82mQqZg8EuEhQGnvnGyK9CzXOBk4PXKws3DCd/G/NGPfnTdDD1ee7ROH45s5lONFz4y8WobiyS+GNQwwAwHOzYom72bmhuXR2WL4oILLlgLxAnTqyxycg9fvtlR5JdtfDmWj24g8m6Mr/DVJoeSpU8e0RGXPnm4UbHKEXnxWUdO8W52TvU25fDQJWuR8ydGvOzw3boQh7HGw8I3efqN1Q5T4/owKewpEUzWLJz+nwWcSBzGyLqBsMU3vtJGpl3O1Aqd5MTJDn7x6YcNdrHgwVd8xvGQmoxCPtty5F+wHh4errp5sS7IkrMW2PIZmff59K2zbJLhV50vY/luDH6xROV84ueLHbkhj4ybT9jgPe+889Y3B93YyBbX3dsfgfStTBh9q9ANDwY+6bPDZnbLZXjKDX5+1fCzY07o6iNtmNinG9+YNp9ucFddddX64of1jKegsKnFi6YNsfGhhr1YycXjF179CK9xPPbzhe9a4YdNfHjo9H+a7LvIl1S++tWvrieQ/rkYfTpq+wbf9g62wjFr7XzlT0x8yJucnsndQ0mNCQAFg4ghYxyjgsJXJNCGZsO78MIL16sJixR5DfTtb397/fcuPICAYA+5g7MRrwTiRSWwPhwWKyr4ZNjJnrZxgZYAOsYlwYXmlY/Nw8neKd6j6Sc+8Yn1HxXJIgu/5IsVdrYtQHi1S3S+yfBZzrTJGMfrYofbGLtuFMYRm8bkx8W02+3Wo7+bBzmPpf4jGjtiCF/5Y1NB+Q1rPoyXR3jxFXx2EQz0jPPBlpqcMSU9fLnGc+OVU9/ttymziU+GLf8d8Pbbb1/rg/6co3y5aC10/l0ocsRGm6I2Ms7GxMUfnrEID4mNP/LWgFw7GXfzyD95rwB8Ldo6RnT4p6/Q5QflMz/hKv/4eGzQa471xWIcj0yx4tNXjCt41q/N2GdLXgXSs1bkzBr2OaNXndY0W+Wvi7/czVyKGx8e/uDIJ/3ilT9yxlHxyYM2uXThKR44vJXofwuxDbM/A++zD5ueaw0mGGBF9pbmBF+bLe18lpd4+tpKbbbkCbGBj+BrThdj++GGiy9/DhiekHx1F1ZPSXKQDTYrePTgCgub+/nMT2tHvxjkD59NecYvx+wbUyO+fN7sK7teB3bd2iN8gYJeBzR25K14w8wHe/DiIXr6ZFHY1IhOtD5EL1gOGAkoQ/oUc5CyuoUlCI/KTsn+AKML0+Oed/RuIIC4gbS4AsEuH/jsSY4ar2SQ1WfDuIWlhlUNFx2LoEAnbmPiUfNDhm2nIE8eF1100cFu26BPnTp18K1vfWthtnDYLR/lx4bGDj6SmzCQUfBgVfjDg4ceXrbCBAuenCF9FxJZCwE2H4r1eRI5F5tHU/+f3I2bbXOhhocun9ouUjFrR+VJf+Kvn25y7Gor4ddOHmZ9c9M3ruTWZx2wiJWMzcDN484771wXZGsLBuPq1h4/xWNuFfJKuSMDK5Jr/GTiwcUuHI3hsXd8e13iFZD1G7EJp1cqNrb+InV++NBmIyx0YS034SNXHPy3HvDoIznTLsdskoUZX2FHfti1IfhCghu0167p0XHN+VzJP26i5zO9fjGPH7LkjMFinbGLZyyCL1x8kicHmzFULrXp6zcHavKueWNe9/gczI1aXzw+w/PZohz7Z07h448PRQ7chPJJF5Ftreg3H9pklOYcr+sWX5zGmhu4xcemcdeR68pnCY15Wr733nvXoc2BDSbxsUEHwa0g/PLR/OSTL2NKeJbS9iMdMsWsLo+ufWP2H9e1L074zM7BXW7xXF/eovALo5of2GCAnQ3UHOPzo2hH5BR28PllB6/9dn0GQlHQyOQj/ZwKrEQAw0DOLGgBeF/ohIHc/b6z/YMpG7LAOHdRIrr6CmDsmDT++GgSjCH++DdO13gJoUvOeHbxtAWbDzJ86JN1wrj44osPLrnkkrWJeF/44Q9/eC0S7+zlYMZMn0/6iH1Y8x8uYxILq3iLQQ0PneKgo7DJTgu4CeLPI6ZXQF6v7bYbiTGLpO/Q0803vsXBTvlio/ls4WQfDmQ8O/ra9BDc5Mh0g4MVrziMtZg8SvvlRt+uYac1wxacNrPPf/7za+PoQ3UyMJer8hKGbPAhrua/uNlG/CWjTw5PQXQRPaRvI/aZ3W7LrTjwjPvarm8Kzv+uFx52yx1s+OWE3drklOaWbbJ0+NIWM8x4XpUh82Rjt17IITU7rjWf0Xm14tVEvmH2IfmXvvSl9Zd42fDk5EZi3Zu7/LFDXt8a5RuO8qQPNxm8fKQXJjkwRs78hZN+82kMFqfja665Zm3M8iA23x60P8g1GXbpZbdXs3JBpyd0+QoXX8YUbXjo6/Ot1lfYV8QqPhi0jWkXiwOZNWxvsEb58hbA/LgRyUM5MqbwhWBjUxzkFHb545uecbWSHt38G599bXL8ILbE5bW2fRdWN2j6Prv1ZO8XB+nwn19t+MRL1rpAcCDjiH00sbGhn01L3yOgAABAAElEQVQYyeuvf2lbkJhKiUmIEzzGgXBHlFTfsDk8PFx3azcPdpwmbrvttvVb0RJp4ltQ6sDni2029Y2p8QItWSgM+OHDJ49gkxzyZGfA+ILmB9nItOFhi00XmcXDnuQ2xu5MKnn2lDDxhYxZjOTZSYY9sZEzBosxfovTRYTvRtDClzuLuJO8k4UbsycQX49F/IWXPnszJ+IsH+Fln0x8uLRhYk+breS1ycAmT7CKtVj4gIGci4x98dLPF/3yow4Xf/kUTz7J8wOnmNirpq9kUxuR1c4GHhn88suXONiC1RcSpj06NgsnZLHCKQbFmmguySEbHB+K/MkJ/6j46RqXn9aBPiKPF0ayxck3m3D0OZjXVp7yERv0bcZ+T8n6aB6cmsVV/vIDAx4/ahj54wePPrswJVN+yRnDbz6MsclOdvOh7zBhkzvnnHOWD3n1QbSbm5seW/tPQvTlGmY+yyOM5lHfWPHLBV/h1q4Y6/DKJhmY+TX/4mDLuvD5iwOFU73XxWx4HejDaE8g9jjYzIeCYIervJgzNtlD5RUvX3ja4pi51Z86bPKXPeN4vqXmUOn3wewNcFoDcHrtSl58ChIfGX025BTJB57rgQ7b6RqHTcFTsxFmPHT013h1MqBGOdVmgDNjAPgFJQvZo2lPHvF948rmJ7kWJKctCnYKqA2CH4Ak38SUfP4CzUaTwQ8bJaTg2KCrTza7+MVkw/D04QNzPI/9p7bXV7BaNPTJs81GCYcN4SlwwsHXvNDp8otnHIk9LHw2CdnEC3exkJE7k2tMvL798dOf/nQ9mbDZfNCnR45ePsvPYmw/yBini8jTVcJXTDNn5BX6cBSjvrYL1KbLnj4qRnbZdDpyMrJuLFz8/MLJrrxqszPjIZvf4jOOjCF+jZ2tHw8OxTz7rXiv25qDZWT78cADD6w/UyHvbTjwsME+vraxctU86CPjFX06xSpGWOFnr7b8itG6IqvPBpv4DjfWrGuOHpn8+ozG0xI/eHTpsd16nb7YzT8+yh+dsPNLji/tYiCPT448vzZj1xB71q1XPV6heUXslRB9rwX9uR03D/9/HDZ+2UJs6aNia02xK6Yw0FHw8bTV9WGLz1525be8qNnkw/q1FmzKnqQ9AfVnVXx1l4z9Sp0PMWnDqqbfeipfMDQH2vnEowenfJV/WMmh8mIcTvJwWQu+zeYXmx0UHNp9UcnNjox5EI892P7sVRcdN+/77rtvfVmBHCx8aSN9+GETZzEUIzmY0qOz/ie6RouGQBPFuFKwxhi1YbiAvDs2RsYYG/g+ePKEom9hAaBoK8b1+RFoQCURX589fbbJsYWmL77Twacbfm0TDg+ehWPzOr6997700kvXBPx7+9bH7373u/VaRYIlnm++sidpfPAVTuNsKi0aSS0X5Mkq2vAbR+zBVlvfOFtkyVkwZHwW44bHh3G544O8uOhOu/TLF742UrNZTMboi2PK6csBXvzwsVN8FhcsbFgLbNMzr3KO4KWL4L3rrrsOPv7xj6+TNL55FyPM4mMDnz5d/S7ackiejPFi0w6vdnOHR7YYi4ddH/L7yq7Padg0TzD6XSU3ORejuOCKxOu0Jwdw+WAVwUFXXX75bQwmPuCgq61EfDRv+MVGXi6KM4zp8UHe07RNhR24+ED06CBj8Ovjk2G/3JU3OcJH5MLFrrHkyODpwwgLH+ny49X14eHhwW63W7qw2uB8RuPfX6dnDclfPtiavtiFhT67xrT5ImuscTxjZGbscE774sKzFlp3nvBgNsfGHNb8uSWY9dmGlVwYy6e1Yg74Zo88mbDkIz22mgNYtWFKB9aIDXw2+Scnd/SQw4PPFH/5y18uOTKK153WqM9P7Xc+8/UFi5MnTy5bdO3RsMOD1HyhcsofDKix5gBv/S2shKcBk4ooC4iyycR3Z3vPe96zPvdwE8EDWpBsGddWJM+dzwfsfknPaYQ84Ke2k/8Xv/jFZZt9J5n0gAQesalvrCDis2+c//CRg9sEa5sgY169sakf+TzEjaWFQ87NRsxks9WNTh9+Ncp/+UmH73JCBh4Tj1rc8g07W9pwafvFHycHv1V6/vbLYRYtn8Yt1OIqJzDz28SKk12yKD8wsAErG2xq86lPji39MNEhZyzSnovcTc6HuoobdFRsathsYm7UMJQTfhVEBrHPf/kwPvEYh1ttrDzgIbJyj4qZbP1w6ZtrfrxO8Xs/nkhhyx/7ZLzWuPLKK9eF6XWBd80+i+pilg8FFqU888W3+VEbyz8Z+YaRrote34XtGnKN0FNg0I/YOH369Pqs5ve///3S7yZRbvXhYx8PwcCHNaJtXRsrn/CJ2Ri98ku3nITfePH0tWxYxUIvIi9Wvmxq4mid8W0cWWf4SmvTPIq9NVgOw6IOAz3+9dXwKWJTk3XQMcZe+YTtbW972/qjjuYgIgMzn36P5l3vetfaw+wV7LEDuy8DnNw2Zk/Z9Mnzxa66fIWjePlhJ3m5gJ1OsvpIDsjZE+ylPj5AxhV54s8TxnXXXbee/OD0+YgnFeQ14vXXX7+eWjwBeu3phu7ztHDyo51fbXa7DmCHQ+xrzA9gkbaAECF8vBKgzXAneXdsCxG5mLwn/PnPf74Wv43ZVwy9KiDjDuhDSxsIewBJtqT5Sqq/gUMu4MnAIIAWAFwlPT7/eHSVGQM/bCELRoIscDH4Mw8mX6LpkZ12yCF+JMzi4z9+Nwh5iUdfH248/S4WtvhAMBknOyeEvIvMaxZfebQ56/ttYqcMN2y5trhhhoF9dvVRsWjDDL+48flD/MCgsBWO8sVm8082Pqw2JWNIDObQKws3j9aD8fLOr03RQcKcy3dzxzZilw9zBIvNT2xsK+TFAiu59Pngiz5ij74yYyw+WBxU2EH06MuPjdhN15MUOWsYXh+oeo/vVQC/dNjzauPBBx9cstkxDp++Njv5gac+XhiLCQa6xWdc2+ddftHx3HPPXbbSFbtfIvMkDSc7csVHvluD+aWTD3VrsHE+K/yIw9ovl8bYbm6NJ8eevr8l5ks1MDWG7zRsDvkUK7zWS/bMg1eIfi/EPPBLVuGP7+mPnj478FWmbLyuQXNvjfFFn21r0+cfXl2x5xBhb3CokHNfXCBvL/Ma0dosfpisZ1h9zdfrpGKWj/yTmzrybTx+ff75Qsbow1kufMkHTjI+77B3WoNigsMXLPx1Y69nuxaXse2H69aeTV/NF0zWPXtyxB+/+trGySmRMdcm2rA9tBgEYIAyYYApMzKV9ZFJyzgdm5rXQW4ggJgUv4nsmwI2FvbZyR77kuGOCoPke4XQYsBjlx4d2PBQGFZn+0EO0S0OdRjVApY4X9uUaI/RPuz3jTGbNRveL/s8B3YLiw2TrK+IS8zswaPfxNI3ptCrDZc+ObjxUZjLBxnELnzZw3Ma9XVSr1dMvAvTTSUZeJTya4wdtvnFL5/hMhaGahi0FW026NuU5EEbGcueMfOmLja2Ub7Iyqv3tha1xW69ROT4JCfvrbf04eC/HLFPTq2Qz3exqOMbkx/23ZDdhOUYse0w4SK0uZG1Vtw8nN687rJGd7vdkjWu7XM0cd9///1HGwGfFXLh1YZfbRwOfUUbTqUncJj0ERknR68hYNZnw7eXfO4hDw5n1gXM8gG/WDzJksPjh548sGEO1Pr8hYNPPDqtHbj1xaOt0EG19W1QXrtedtll6ybCH1LTtRm7/hAfKFzs8OcQypebM3zlsLnUDyudsJYzuvjWIzxy4SnCuI2fvJuXPlzm2HVvXTQPPvvwWSN5n+F4+pRP4/RQsfFlbbvBmwu4rfHWUnmDGenPWOwh5YeMcQWxXXFA81V+dsnD4WZlX7AOkW+7vfvd717x8IHsW6e3J1Xz7SZp7cuLtxpe4ZLzjb1uSuUbFm1zgPT5FSMqjqN3OYQZI6SYTMoFox0oBgQ2ib6NzsbsxMQGaqJKIDsWur4JMnFvfetbV2Af+MAHlo+JY9rhowDybTOF0cIwHt7q4rFQDg8P1y/dSJ6bnV8Kgk8xMX4hx8nJ4msBsuu76rfccsv6ZTOLwwSIwwWvyAU/fIqrRdoFiE8+ChM+38iiFnc6fDTGJvwWgVcFMOjDyJec4KnbmMWobwMhy14EL5swwcBneTYmj80vmXIOD3k+jSvZwE+n+Ngxzrdv4/jFQmTB+pos/LDxr8CUDfnoFGoTaG7zI1fwhMk4v/itn2zRKUfm16bgorcOyfpWkEd5Mi5omNlz+PGtnC4u2MsTWXMvFzZ2Pqz/ciWW8JFREB5iBz5+YBCjPt8wsGcM37ix8sqWk6ffsbJBiMc1REfhQ879LbeT26sV/3dH3xqHmQwfSJ7T0+d/kjExwcCuNlzswWTcmNoBwQnYDW9S2MmhGde+nFe3XrPstpu0bxxa8+KmoyDxZ8uYdrito/KXX7jFTZ+8fQfPfLmp3XTTTeuQi8efz5WsDX96qTcofNBnB+k3H10//M1ivHWgXQxylY3V2H7wTQbRcW2wpe068PT7vve9b30DC6944PAU+pGPfGQ9KXtrETb+HDJ8O8vTlFjc3H2MkB+2+JE/uLSRtjGxtdblFQ/O1vT6h1IYJUggjHQhGNM3LignDBu9xWIijJFhmAML0iSYFL9caCOL2PTY5fHQRSegEydOrM3Pqy0bund7YSgYQUiGoODA1w6bGoVFDcsM3sbqXaAYPCX5oDS8fiGrx1QXHF8S6mTisdtd2ylVvOJUYBSPdnhaxDCSlWQ44SCjj1x8MJPni7wbQDbZLTby+Gx4snDDbXEZk2tPcr7W5yTqtMUXnfLifad/0uNEKmbj/CbHfwQvwiPDRrFYtHDAX/71yeoXP319/GxbB5708N1MYIbRqzmboZiKk19yNk95KXdyjS8/eKj881NccOgbg7/FTtc6s8a6wRqTF5usUx7b6r7Ouds2svK4HJ75Aau8KNZ8eIqXb/j5NJYNbbrdMMTaXLDlQGQO2bTuXvKSl6wTcK+D6LLR9afvSd/pNN/s4fsdIvPjQ1bvxvHgEXOyzSWbxuAsb/rkYEX6zTf5/JCxDtn3G/1qY4gcm8WPZ/xsRIb949sbCzd6dj11n95O0NatcbrwmWNFPyzktRV8eMl0vcUnJ782ZuS6cZDxitXnAm689OTfTZc88rf+PGG48TS/vmnWt/nkyb42Y1+K2w85L394Yml96BtT+KJv3HrQh0v89i88r/vtobA4UCJr2qvXcgu/LwD4IsA999yzDpauXzh8gcTem+wysP2Qn2KFQRtPW4GPDgzsrP2PEOacFG3ClIwJBnUXs6hNcnybszudD5F8IONDHqc3p4lkbBDetTntWRT+FIO/l9QiIKcNcIHwrw0LUuvD1ILqRIAPtzEkruKgZ4MQMKx+WxMGSfcthquvvnpdqN59+qDU3Z5vX+nT9mh4eHi4voboxJcvSSQnoXiKOGDgUymGMJEpZm3korHI9S1sBem7aVgIXpU4CaM2ULXF78nJ7wfYHPmMartp+/MnsMcjczZs+PImDuPFA4t4FYSvIPjJKoisglqQ+uJGDiHWD3KRupAtchcmW/yzpdanq61uXo1njx1j8OSvdYvPhjGxF79+uG0S7Bsj71HfKwlrOXsuYr9nAetuu6l4ZRQ+OIofD7FlzaOwyF0x4JOBQzH/CirPNmMHmz77CB971mW/CyQOOtavGyPc/NA/f/sShmvEZuX6dXMiyyes+SevL16xkOFHMVaM8IXbGJI/p1qYYEGNaZtXr1vUrtt8lAvrGNmIXZPG7Q0Ob7455ObniRE+vhUEe3V5Z7O5IFcMZMVgvPh9RmsfouvNiS9H9Kdr/MFHN+3i9xVkr9n9SRNPJzBbE/7mlLVgHxGjOWxdhpMNObIeYAsLv8bIh1Xs8OGZNzcFnyn5/LgbuYO2bzXyJ5/IusyfJyj7MawOaOZcLD4r+9znPrdeI7oGxexXLuwr8oK6qfEvRlhhhIsMnrbC9/pjik0gA0AIrInQL2jjTpFdkGRsyO6EvpXiJPnJT35y3Tx220UmEZItcb7F8PWvf30B9irDRu0UxRcZm5tTAH/sC4BfQPH4qsCB30LRpsPW5JMXpFMsYqeaT3d0Pt3FfRYiwUjCkVMVORfxe9/73oPPfvaz63vsEtrihZMfpC43YScHE+JfoaNmB2bEVzGbD+NuJB5P+6UrpxvY2BOXDeJwu7FZ7HJsLuScvgVj8eUfNpgifFhRdRueMf7LKVwKv/gK3LX51C8WNo1NCnM8scPkMwbf6PNkZf3Ig4uQvvhbvPrFMnFMP/jh0k6OnpLP5ot9F6ObtPi6ccNIl7xc4nvX7KmVvA/UHZLw+YeXXPNPv00gzNN/8xBWNviDQfzmQT/c7JFhHw8Znzccv1x6xx13rBuypyc3EvYdKrzislb85rdNw2bGTtfPXKvl09xri5Esfwqeoi0/Njk3XN+wdJDBR/LhuoLBTcCfxPcUzy57bNC3jtmwrn0LztNfNsiSicRfIcOOPh9I3sJaTU585kONr5Yfn2N50uHHgbLf3H//+9+/vlFofyDv5uHJxCGsa8F1Kadi90TgCZ8+LG308JCPx5aNnn8kbrxigCO8+PThdhj31E5OntwgmpfWgL2hXDls2o/lPR/y7QbuRmz+7XUwK11vfEYwly9tvsVCRo3wjxFKoI2AUEVAFgM5f6VWwm1MSOJcWD4jAO4rX/nK+q30TsqSDKCLzmsCJ2mPT07LThiA26A9YntstIkgiQASaIEjOCw2fWMliwysYmhM3eKDW0Hp1GdTQl188NlI3JkRfY/OvRv3msDFSBduPvkmx1+LRZsfY2STK59qfuFFYglvC89ikE+y3ifbEL72ta+tGzU98+S1mxuHE5RF4m9juQmzbROC17eHYIVH4YdNNpTyWw0PPhnElrF58cHIltjYU4uBbHrlgA362aeHyJHBN6e77bDh3ayF/Z3tvXef85BlO3mxyLd4YWIv28UDD54aJqTPjpsqOWTcB+c2Dk+k1p4xvuWefeSChck34JxA/SkWG6U5Yt+1IRYlnPyVD3y+0cQDe6e9bkTkJv6uO7rlVpvNcgmnU6R16glfDtlwnXkFCws/DlH8W6dyzp5cihl/4qWTH76SYZffmUM89uVPDQ99B0bXvly6cXh6c30VEwzmU99B0oHSOsj2ArD94A9OdmE2zv4sMCAxGC++rkm8clvbNSJXvY6B0R7wsY99bL1x6ObhG2FuDE789io+bNBeXTlU+ovIXm2d3vYKH2Z7GoEHXrmDRQzWLAxiaS7VxsSiTU8bhbNrwdp1fXv97knJQQDB6WDgKUU8SM49bdJ14JcHr7G9OXrHO96xPov09OTzUXu0J1lYYVCXT+3w4WmLX2zaxo/+I+HqbIlmhADCExAAyK/Qezy2UJAbg0XrUclXx7yy2m2LgBM6JsWHz2o3DH8y27v6HsndgFyYFprHb5MZMDgUNBdLfNhaMGRg1jeuru3itFiSkQiTg5I1+X5xqEWdjLs0fQn2zp6PsNDPR8llT9z4bKnJw4qMt4AaM47Kde0wWjQWgwlHnpic2p2A/cE3GD12+0DVU5SF5tWbhQWvR9lT2+/bOPnwVczTP/zlU108XXR0umkai8+evpizASPb2VfLoU3ahm3tyKevQxaznLhxi0nbSc+akUMbTOuRHX6UqPzhaVs/8CbDP5sIr/jV7LfmjOu7EG0OLjzEp4vWKwonZL/r4oAEk5yYJzcZpM2uuBDfzSPf8jbngJxCjr3kxWDDyYaa3Wkbz0YhVzYUNxFkPXhS9srj+PaakE1UXvTlAy685jAcYTJGhry84JtnPEUf5mJqHtSKzauNzmmfvJsuG3QQ/PYAh1Lv5N2ExIisFTG5NuWZzRm/fBYTHOzrqxW4yYRRzDZYvl3L8qNmlzw+Wa/hvEZzvTjweir2eQOs/JPzrSxfWvBq22t7v11vM/Y0QoYdBDcc7FvHdOHA15bH/NPTJqdW7LU+x3K9GEf2yVPb9cyutcmeV30+PJdz15lxeHraYdNfZ7b3WtseAOwrHajC0BrNF9zaSm24rQd+2d1ieGgztckC1eKgoOgzAJzAKUYZ13d3zkl8sm4qgHpcVCTZ3VDwTv5O1ibH4sonkGzAo0bGBMhmCzp/5BA9bTjpCVChFxWjvsRlX58cm9kzXlLZtlHAbROfF582TGTowMiuNvlwyiVZRJZtJX8Wnr7JdWJkw0KwgXmMpu8zDzdif4beJutDMjl0evJI7cnELzx5J82u1y4WuA2Jb7z8FsMCtP0oXjXfSnjxlGI0Bg8SY/HoG5skLidST6huIDYMF2pY+IAJ5t1ut274Tn3Tl/Fw4dNR5LaS37DSMQanObHOYMXn203LqxeblCdpN2YXo6/MWkOInlOlC+/yyy9fm5xcuqk7QLFTIZ9PvPrwIrxigAsW/QhOMVizdPTLcXLZ9brSBuvViutJjsuLzdEazZ+1IUZPAPLvpmkztZnLCczWLx9w6cORrxmTnGeXjEKPDAonLPaEdK1pGMmzYU9wzSOfg8k3uzBZK572rF03xHKQLoyIvPnhUy4RGX6MKemqxS63DrPXXnvtykFyDlz2KnJ4sLuB2LA9ybHvEGczf+c737lO8bA6QHtCtTeUw2yyJX4+kbj11XDLUbHEhx/pe7rhq73AGJ3io2tOvapy7bth+Bq0Q6M1u9uuJXl3Q7zxxhvXTVoOxOYGbU8XN3xsKXJZTpsvfsVkTExdG+TXnzIxoGREAILEo6jtouFY0jhAxqaxgjdGxmsUH9pIvtc/ig91fMearBOzIAQvYRazBCG+2SCnCCaMTTIsK4gtKGPTfwEXA5viUJLTJqevjcjDgqdIrtriwNeHy0XIv7zAHG52jMMLG/vTB1v4+eKPDntIDsgjk23B+j0Evr2mcHIXv03BP+FxcyBvQ3Bj8YjqgqRrwXtfKs+9wuCfrZkvPCUe+4o4YBNbiyxfxvHMbTHL0T6J1biNmY5v58GFTz7f1eUGRn5hQLCJsVySKw7jyeHlk03t5kLNJhkEj402XbmXJ7kzhoy7IZsfJ2n+fSHDU7MPMc1/uWNXW7wKyh97dJFY4CpmPO1kjUfk0NSB01O/w4PNlozipuZ1p1crTqR82mDIOkl7yrZxo3Kh7RqGl381W/zBq01Wnz25wkP6xTVjM0YGztPbjctTBhuTbIxOxHLqaU+ekVddDg++Zu+V0MTGZjnmFyZ9vvXDJQY8xRpjg39zRYZs88SnmHyu5Ym+1+hO6V4B+/zBOD8K2w7B1okNW3GQsD74Y59tbfi0+ce3X+jjy/W8dpItLrhgZ4cOCru8yK3Nn29/zcOa9SrQHuCpBYXHmM+W2HFD8aUaX+319GRMXMg4HX7hkbPyK248RRxqskd/C4sSgAZQC6mkuYv5sNxEAyrRNhAL1Nd1PV04HZNvIbphWLBs2vAsZCcLpyb+2HQKaXPiFyiFTsmGqzFtpUXCl0ALnJyJQWyYWEnyhEXOSUESERtnSxY9Mm5sXs15ZBWL2OE1zkdxWmCS2kQYK/EtPjWMySwAZzBoi4kMPBaRnJCXPydzNwXkfas8+maKvy0Fi9OxR2q/GOSxHNlMnKotMJhbAHDKw5xvNlBzrj8XbX24YIJV7MXiYpObboLGI3b4ciJyA/AlC2uHLXJsq/WdirxrVocl3GrUog4zHn2l9YJHXj7JKfLJB4zhFAMyv06lcu0UGi72rB1jk6xlF6HNu81BLvgJZ37p4ckbik8eX8FDeIo8yRkKI31ynkgdwLz69VRK1sEOJr8n4PMZN4/06VgL3uWLRexy4XrIrzjZL3/V/MvZvD7hkx+41fpk5AGlay/wX0hdR7CQIWv+xCTPTtjNBV3fxHOjE5t9pUNPPpvj6Zut8siGmMjjw8IvPSQO+ZJDn9s61JhbsXstWVxk4XJ4Y6888cMm8hmE14cOEW6C+3Jsslde4SEDj7qbmXGEj4zxge86cFCx37qu5djnGHLLnljImUs3COu3wzgb8CI44LRnyIGbtNeL9jRyzSHf4cFvDdE3lj02kOtrfQsr0AQSjqdv4RH2LSqbkYlHArBYXYA2LxtDEwcIECXGxv3Nb35zfQDlycQ428CkUzBswwKD4FC28OjoK+yQFaw+W+yq+YbRonVSoBdmE2IS0+eHPvs2WR+ueyfvFYcN0qOqC9VGyTaZ/MLHF7/sNCH6Cpso+2p8VH5gocemzdYpw+Nzv33qcyOnITcnH3zZvNzcLHSvYg4PD9ciYtMJ2SbjcxG+LTy2y/Oa+DMLGY8Mv7DIB9ligRUZJ6fQKa7Whk3KAvWhvwtffsqtmh0XmlNe/XzrK06qXlv4vCHizxh/5lHBYy89eMtdN41s61sbdJAnstPbqdgrtPi+FeibduxZH9kIQ3Okb+NwYcMh9/zwnf3m1Xi5M4Yvh8Wirz1tN0Zu4uc3+/JqozXHfFjT8PoFPk8fHeLEZk15J95nY+zk17zRDz/fSpjV4lIrcMLAPxt8IjlwbfsTJPYCBx3y1q337tMGebrs2OzY8PoNTidqc++ASYdddsiXC32kH2WvMbblJIz6yYuPT9ePQ+zx7TMi6zUbxskjNZt4kT3E/LsGvTZyE1LbiNmduuXVXLITBm3+8Ok0xo+SHFuuJ3nxWYgbnTn1m/w+izZOPnuucUXfXPFhnbsR27dObZ+LeOrgF375t9bZ4RMf0Qu7fteInM655Aet/wcSiJgZVeOVBAoS6LTj8ckEeGUCiJvHtKMNiMUFlJOw95leWdmQbTLsG8+fhAienlKC+GVPH6ktfrJh0yZTkOmyY9Hjk7XIYZds/4BHgiUJRjbIugi8B/eBKZ5viEm+2PXDADccdPlWGuOLT/61yZCvbUwfP56JdeHYaH2+4ZWVvpuzk6UnjbPFaCGKx0KRT1/ndAPxnlyejVsASJ8NccImnrCFtRjFBttcRGzgIfi7cDydOgnJmZNdFwdf2ScvHnOjbawCt1dbNhMXavp0pw1t+GAVEyzNtTGkZh9+Ywo7au+IrVe6TsE2UiUKV37x5ZUt8fnGoNew9GHMB7l8hAMP6fONyCC4xVH+yeCFWz74sz7NFYxkbMxeA8LhN6UdaG644YaDw+0A0e9KyI2NB85etfHDtmI+rRfrjR9+2W5NkkH65cG4Nl5tMcHl2vZtNrbMv5sYH8p/I7Zs5L4AopbPPsuBBz61HPCJ4Mp/tunhw1KMbGvDWVx45ksu5UaOrDdrTQ7ZIDtJn7w1iawdH6g7hIjZQUf85QuG1iX/ij4yNvEbw0N8ZENfm57Y7bX8aVu3rh/lbERPbAg+n506bMiRvctaJVNetOW49S9HMKnpaJMlg6ddHNp4azdk2MRIMKOIEh4h4/o239PbXc0pgQGPeBZ0E0yvpGh753pqu/NZqE7U7HvUIi9pbObPmL7CnwSyhY/4o0de8smhkrI62w86MCM1P2QQm2HwKMiWU4THucgvF/mw1NOHRWwj9uTkgymPuW0+5MsPbGzxDRe/ajx17cZbVMbDJiZUPBat042v6YrB5uAmfTaywbkQXcBse8XlMb3NGk58PmBVYMSHTU74nVi0LSJj6ZGli9I1Jg7z6901PbZ9kCt/iA55pZzga7MPt5u0994u6C4Q4xNnvquzoR9PnPTgKt9s4Nl0XIxOunxal57s4DQOHzszTheuNeBCdnF6xeJJySFIrPArcsUeO/SVYtYuT819PuobJ89m+OXUB+Xy4Zt1LnTrz6nZ06dTKd8+5LU+xKnA6bMxa4I+e07J9NnnUyErR/JCpjzJgT4yrp8sPh6Clx259LTsdRU5X/CATRuR186Gvlex5twmByN8nrydjK19OMkp4VFPMiZ+OJDxSnrpNhf4cs+HHGn7HYvWavbz5XWP69D8eiJwUHLj8arInMkd2TAUM7uwoeY4m2Eqx+WUrnZyXQdelZk7a6CDd7pk6fHnwOhVpZuHPUFeHULJ7Ha7daOEV27NmX0l/OJD+ogOm4g8Cldt8a1dS/AABT5wEkS5pDDu4nFyx3Px4dHnTEmWDe9ov/zlL6/TdL+A6HTNsQDIkDd57PDlIhQM3yY7eyuC7Yd+wfEXVnwFFnX69Ng5vj0tweN1j8dAfacli/b87bd1+ZdQj4v6fNgQfceaXqc7tskZp8M2DBakuOrz2wQUpzFEzpiFJVY2EZvZV3td6JTj2yA+uBMbHXaKnY5N8Qtf+MI6wcEPlw3OmEJP/vNHX5GjME8MYTFOH8Z01Ij/bJNj32ZrQ+DHPJprpzuvi8gok9iw2L0e9NTkLyNbCy4WY0iu2RMTf9oKSk5+xYjCHE5jSiReG6ybLNsuNDdmF6bc2aBhchrmx43j1ltvXRejk7V4yNJFcPIFm1oeldr4+tZ2cvA0DjcZvGTVNg96Ng8YyHt1bOPCd3hzoCMrN/Jt89Xny83YZwmtAzjFTtd8wq/NLvlyRF8pNu3GilU/3Go25EXeTp48ua4FT8vyx0d+a5sjm3c3Zjcbdtglg8TDN9vk84eXXXgUelN3Gdh+NJZfc4voWwOemuxnct1m3Rrll5w8OmR6RevNhWvSOil/bMIEL4z0+cXT1+a/HOuzbRyfjP6MgQ3jeK4phxZPomStVXV5oqsPh1eVDkc+R4TVPHtVzxab+ZAD+vr5CUd2Zxx4/ChyQkcxrw/bvtv+nyadAKbkaDOSQQ44zLkNw+ZrkzBmQnLEOEcWvlOUfhc7+4qTpqRLaO9D8dxYyEsIDLBIPrmSXyLikecbjgh2FxYib4xfC9vXSJ3iwujOTZ4dcTkNuXt/6lOfWhhhLza26LHdqcVEZV9NFja1GOmIhf19zC0sY2SMk8fXl99uuvzEF0t28b0G0Ffk0wXBpgWIXODachwG+MRNhw3EJwqPtpj0K1OGLjJGTuFLHmH0msgvtbkBm0u545MNuk5MfmvW601kHB/GNkY4Wz9sKMmUL3qITzx9WBBZBRkPvxuB3JrHfpnNBep07MN+rxHNn6e5DjzG6WebL7HzB6eSv8bMWbkmi8p7uMm2ifEl3m5UMDj4+MVRT85ym51lbPvBh+sNLrjd9DxZu47kkTzc4eQf6Zt7mMmEnQ4ZdlsbMCp8KOzRwSOrtsb4tDewrWSLrPyjXoV3DZU7NrXJsqMOsza/+tr8qcnja9eHR8EXEwzya+3AiB8ub1GyCQ+7dMjaG8xH+5I54cM4Mk+IzWIVQ3POB2ILDyY5CGdyZOjlGz5j8HQwcF3JSXOUTrGYJ9eRa7zY+CoWtb58NafNYbljCxVfONVkEYzi0F+ryCAFTIOcM8QRvnG8JoqMxWqhSigwEm1hS6hHWTyvq7ooAOdQUiSaHH8Sq882P3iowOrzGR5jdGDE1w+nPr7CV2PG3eQ8gsLnFJFtNXwwkReXE5JHVd8UM+YJQ0zsi0kNL14LR62P+NNXk40XNrp88Y0nNlSMxvFhOL29NpSnYuGjhSkmm7NFxhcdVH7kALkA2CCDqslr8xdfP8zhbJy8Yrwc2OBga07lCm6nJhuFU77TsoUtPvmkq+3pw5/PdrJ2cSD2y4M+WbwwVItRCePUTV8sxYpH1kWphsupzdrl3+dz8mgz8OrC7ylZw05z5OVdjdjkW9xdL/Xzbaw2PTpqfDEh84PXxiVOpDZniIynMzravrTiuisu+RSDk6oDmzVhzZRjuOBQyPKHh/DgwUsmfGo+8FF92IqfD3w8cnJh/n1LTW3cGiefDl/Nd7bxEJmITfx42mzRCT/7iEw2wkMGr+vV+u96JcMOnldU1iqyh+EbN6bvQMmG3OPJobUOnxj5KA41LLPPbnOh3Rje5PPJVvOlXXzWgjcNSAzWKCzJaDsMWavwsgEjfWMwwYuP135eX0xwKXAgYzP3YTGO77pf/5GQYUwCBoBC6owZywBAALRp0WkToxM4oIwJWPIFpM2fC4BtgAEhZ3HwATiZ/LHHbpOvnfwCegZrbTbpK2QVthWJ9+jKJ7tuJghWyWfXY61Tic0jW3TFYBzFV5fQEl7c/GuLg9x+wTcexuJihz/4tPGdPsgiGOjiy4k+X2TLqU0Ev764+U823Mvg9oOtSeQQvhL2ZIyfzRZZOKwNBT6/6IiXPXMPH/KFBQsffjwxi5O8OSpmfTHzWRxhXIa2HxOjuMnTgyndMNMx5pWE3PZ5g3n3DTdr1eN//um1DvmBNxxsGWcvv3jhwVOi2YZLH0a5UhBdc8autSCXNuU+Z7CxGUNuPj5v9PemfHAuHq+u6CHXa3kICz94ZMSKLyYFsZ3exAsXzIg+uXLRNS5n/PPBtr6NljwSK7908Yxr4yP++GFPOx/a+SYbLv71m2NtvGymbz2xm+9ucK1V9tI11+mXq+yEkY8253CSgYMdhb9yJk5tMZPXJ6Ot0DOm8A8nXvujtYonL9pspafNt/HpA/byCPfUoVue8LPBpzH2lIlRn46CHrb9X4b/+DtKOUlpDW7BFWSOGCbL2XRE3hgKCFuBbKEAEyVHJnvG+MKTDHr80AtDAYV1+jVGxxg9FE582Mng5V8flRRtNhV2XARqiyW+Osy1s5Ne9tkzpq9o5yuM+uKlq61uscCN9MnTz554yFtwKB/45i4s2qhNgZ0ukvKIl3+89NlXZjxklfj0Jq7l7MwP+I2xeTYSV2PFTUfhAzVv2mRhSVZd3Ma1+WMXPsQOPfywurjIilMhr4+fLlk45M06YIcMu+wZR+zmi62JkYx+hawS1We3Yoz8vk19GOjAo25OjcEKO76arNyxWz8c5PFaB3TJGqcXBv0ou/phVafDnn4xk8ODMX/5Z4tcOvh4Sja1EWzGydJJt5jhVfTTyQ99PGNo2mB3+hI/O11PfHUDyS95lJ/a/M142Caj8BnJBZ51wnbj/PJPrxsvzOWFX7IozMVCb5/IZF88dCtwznFY6sPBbr7YTQ+2ck5m/T+QElPy9RHhABpTltKmKCgOFcYb51y/ZOY4PtuKJNLtYpt2jQsc0SeHly06qAvHGJoyizF4TQJdWFATnL0WinG2xITUMKRnjL3J127RtQjJ8VEuqsWB2MRDeOSzjQ8Pu/zqaxtXIzX7ePIXLz/so3KozpZaIcNGcmzhVfC1w8cn+3gRHTk0Zk7YzScZeQk3vfDhTTLGhjo5tdicshuDV3sSW/mVfz5goEseT78LiV19pN0rgfB7DcBmsbLDZ3bpNY4vfrbZyjZeecCLtKcdMgqekl1y2saKjQ0n0tYGeW2vWJyku6k0Tp4MmjbyiYf4Egd/tcuVnCA6KHvhxGOneOeaJCv3M3dkmyPj5NmW4/zjKWTzQ4Z8OPktR+kZmzpkkBjgQ1NGn41sw1CbL8W4YoxtuNggN2VhMB5GMuloTxva4VA3TpeOWo7KE+zFWAx8I3MdRtjIFmvx0CWfXP7Cxw6filyRs5bosZ8+Oe3WxrEmV+AcMLzvqGAo5oQhOhwBja+vIDbIlxxABZU+ucDjFRhdffqIjSj/BY/PJtnsspPO5JEll2z+9MOfvJoNhS85IpeN4s5eeCQajzxc5UIdJjbynT86xtV4xhFcyBg+P42rs6uNyLPR5JLPdhjlvPft6SXDRr6N5U9bCTc84cXThoUdsePRxecrfjb5wdNXk6+mk5945Iopf+LQFnNri7w+G5G28eZwxqENt3FkE3bjYBefreyxM/NBhh6fxtTZJidufGuCDePJkFPS5btxuuzSpYeyTwYur6jEs0/ZI2/9wZivsHW98DPjCU/+0mMTnmyFkw88BJN+PvDYJjvnLZvq9Gd+8bPPtgJvdsMup81ZfuiRi1/uYMtuONml5xUWwqdHTjHWnMFprLqYxYXIkUfk4CgmGzB7xZJfMmTZMG7DTocfRBZPDV+8bPHJBpmpS5Z9cqg+u8WlXaza5I3hkVemfGNs8tk4+fUEAoABjBJAKcO1czadL5RngBpH6gLLphoAfPYshnwC5IKwMLTTYStfamMwTT4bCtvzpJQeee0Smr66xRnWMCXDrjZcbIgrrGFYYLYfxqYPNrNDRpsNfDbY1sYzNmWzpY4/bePJIRswl7t8zsVdGwa+6OHRdZLtZMaXsUg/+eZEPxz8wqGP36kFLz4b2YUt4lspDvjZQC46Y4jtZMzB9E0erxwag1+BF7GrDQM7dMjXJ6uNyBXLYpzh0SHDVv710zMeRnrZzBd/8cKVfTKTjOMp+Sg3xoqBPVjCE/Z08MkqCL++Mfmlk+2wqs0rIqdPFo/POYfGlOzyoU8m/3jlrXHy1oo+u3SKx3rEM9a4mr1quMmwE4bq9PhU2qDpa0+scCrio6cNB1soWePaKJyrs/dj4tFmU1388ajxkTxsChIXsq7zxb8i7vhk6NNjiy+UT3z6dPIjPljIF1Ny+Ig8YocMvpL8zA858seAJsBwBggqDFUCg88xYBmkq43II+NA6/ORTEHrR+kuQBsWJFl0ySVLLnvqAidvjM/9Qs4Yv8g4Kh58hRySC5TPMLGhrRhjh061GOGB22LNZ3aN4aU77bARvuTDMPHRQdlhi115Nga7MfZqZ48sGWMw0lFQfG1ydMkZ11eyOeWNw2eMH3KIfv7JwGisPJOHU8600ys+vNrqdOnv9/lVwppeOVKHEbawxtNH4SUfsYuffzVeGMjh0VHCwqYcIzx9cgpKPwz42smSoUMuGXWxTVly9CK5rq/OZvbIyZW+m3X9/OArMPFXm1xt+si4/OSHbzRtuFmwZa6bo3zhIbki01rGZ3s/r1OPPHvdGMjG04YBrnSaR2PlkW7tfJGLRzdim0zjsOZn5oAc33SN64eDLT6NlYtkZszNGR+wsJ/ctMWPcaXchZdOOVDTj9Krr25OtcmyCRN/xQ470i+uzc9D38LoJFrglDlSUHXjDEfGMhpQABA+ypZ6bmD7ftgvKWSzI0HGSm6JZL/Tb8Hyx+6k5LPJrja7qHH4tRF/7JdIPDqIDFl9OUTFakzbpPCjreBn22lLu8fcJpDNMLCLr4/Iw1SOwl9eyJMJz1LafqSfDX05o59N+JJjz1i8/LErHjqRdkWe5FO/uLXFiN888YPPfhjU6bCtr9Reje0HXYUuYie89FG65LQbb7PRpw+vuj5dbTZRudcPb/4mdrJ8RXMsW8bZDiPbsE17cy0an1jYhp9M/GzzZ27ww2Es+9M3XjK9XqTX2qHHntzgwZFPGLIbhuKhg+ghPvgyrq2m49pnY8rhk53zkd2w5pcf8iiMeOThVLSLXS0GhS2y2uyVs/DoZwtPG1726vOrjRcO7XT5T4Yc/wpbajrhwNOni/iqTzd9cl2v1gD8iF+F3fDUlhu28otfvHSNKfnAgz2/MGrTVxAfiC1jcCzkGIhSgPBSxM94gMgzgpIDqOTQZ0tSSmpBGSsYIMjRQ2xqJ6sdPuMlrwSws49D3zhcFTx+pqw2P9knm56xYjbeGAzZwdNWm2BY4UkeP/z51dc21rga5XtNzGaHLSXCJxtfm69k9BU3J34aJ8NnJXxhU9duLrK1XxcveW3UnMDBR3krTv6U5jRc1dnUDyP7+tbP2fjJFQsb2rDkj4ycTkqeTWPkix2+7BZX8uyLTz+bxYk354A9F3pxGWeXT6RvDBVn+NlWyBpDcLGH8Okr/Osr9PkgG4989puj/Bqjn7/45Oizlf/8qffjZIcOOw4K9NiN0smmOMgictaqvpsZ23xODPTZVBtrnKzCbgcTNpPVDhddhR/yarrG2YPJuBywRUYsU8+Yftj02WALn44xJWxsixEmvLDDVr6zgccOezDg00NsatOpDgf5xpMvJ/ps8qsuhmX0zA+yCll2yg35fCQDF7415rpk9xgFQRoMYID4kEjBpDxB0mUQT5ujePXZKFmcomzSxQsoO10A2uwhQPXzFR5j+eZPO19hUSejVsRLHrGJ8NlFxhrH02a3/JChJw5j2WNj2iMf9tpNJrvkEYzG6bKnxivn+hV8NsKXThuWsQrbxUQu/PKtrbCXT7knJx6UPB4iS1cfdn7w9MOVPXIo23AobJNnG7Um8qHObjr6yJhS7HzFxwsv+XJLnh19JXzaYWtzxmv92djYM9YaZUdfDMXBPz1+2K4Om/HWt3b8dGClE06Y2IZDMQ6H8XBokyvHbKFyms3iy5c+qr8624/ywx6Z+uwoybMPU3aM0VGLS03GOB0FX5+cMbaLQ1tJjyyd5mNiJZNfdsiFt7mQF8ROtrI98dCnS16brHY62WFLe9rQVsg2ni7eLPhKvumJQQ7xyOrj68fXR8bpkyl3+uT0xV1OyOIZn36mnezhkZNnPIV/+vIyx/lQwseffUmf/vqf6J1YAWOAgkFlBqCNACQXIMYCkHNjMzHx2VD4zLYxvksgkPRbXNMPm4ox/H2CCxbjaCZYXMb5lSx22Kjok1e02ckH+XwWuzHtEqyfTbr8oGrj2vknk78pTw4fj1/t7PDHBypf9dMzRg/xxQ7sxcUGSkabXNjZIT/x4aF8sKUdRjbNabrG8cLPvrHG9W0kfO77ikcGiROvdUneItbXVsoReZjCTi+MfPOpZpsOfPTxEJvadPD1ySl9ayfcnVLJ8YOvrdBH9Iyp4+srMBSjXJFBdBsL8zxMZJu9sNLTVuiyx4fcTeLDOBv51GdLTV5Oksl/ffkIO7thbp7ZsW/IRdQ6YCv/5FF2jcUrBn7I69cu9nj4k4qLLW32FXbSDQOc2SeTX7rNOx4SF1kkHsQePqJfTYcMHhkY8chOu+WfjBIWdmBQyNA1po1njvDCTH760RZbPOOoHNKDhQ08NhH5yTfGZ3xy5aB7wyb/0GI3iNkCCnRJUGdQnUPGA6RWpi57ZADhq3H+FLYCxWYJw6dnrHr65UOJsptvsvyVeDbS0eYb4fFJXkmfPXJ4ZPXJqdnVthiKgy3t8PMbhhZqdrJNhzxZPL7DwXcl3MbIyWk2XKjpwkM24i+dbISBTyV+NtTiQxNbsunTI4v4QMXRAtcvV2T0yynbyDg7+saQvjZe+LLNfzbY1DemoOTZJVeuyOrjy5N+a7Mcmic+ywFb+SWbH3Zm/J3Ikgl/8tmjpw0z0maHz/JanmBMlt3yQ655DkP2woWffRjww8Y+ImMsG9mf88U/olubfjrq2mQQX2TKI36ftcizMTLJqflE5Stc+Pzqq/XJx2NL7Hzgw4LCpTaenjZe85Ufus09fXxYycMejz++8dVs4SE8hS1krBwkGx462sVFngyiY8zhhK3sa5Nv7S7h7Uf+1OmKhT475OmVy/xMDMVqjJ1kxMMGwqNjnH1Eb/0eSIoAcERR24nS5CgFrGaEHGAlQZ0M42yykwwe0p8g0sHnjw5d/pE+/8CHkz75kqvNDrnajZFFxtmqGJeA7LAtMenra6cPj3Z4w9fiNFaupm3jdPK/wGw/2CcXTn2kn998xdfnp0IWjw/6iviyi99ks0EPboUeUpcTflGxZQ8v+dr5C4M+f3LAXrndjyU5fG0LnD/YtOnXzkeyza8NoE0Anmyq2QoTHGxN7Ppykm1rjt/ypuZP0aarlEc8PtThhEFf/Hyz19MKzPhs4OeLDqIHC3/hVNOJassp0qfDlpoPY80bm3jGiwPWdNmnF0+7NcOGPhsoGXYQfu3sl0/j7OQzObWSX3X+8hUvH/X5V8gjbWP5kAslPkzGiw+frL7YyMqN9SNn+L2Kzke6E0Nx4SF2tNnkI/vZwENkyEb4MCI6xloDdOOps5WsPMPc2tdX2Cu25o9ffDbKAZ7YUbjxJuGHecYgh5E2u+r1GYgAMIBzQRUEZ4wRZEzf2FLcAg+4u7QxfePJsDkBahsjq/gDYXS6y/OdbYnhM50mWfKQxLNFJjnBlxh2FHp8TSxs1p/67PplMry5schLtoxp84P0yx88/BU/vxG8YmUr7GpUTmAik81ir69GyWjTlT92ycOG2E4ePqX8ND7l+YaXjraC9NkMc3HzS59deuWk/JMvnvz7G1jksk1Wny3y7CliwfPHFec4PWMwtR7Ih40syp41hKwZPAVWeNgwx/Tlz1j8pbT9gA8fsc0nPXKwNM6Gcf34eGQVVMyta322YS8P5SndOR94ZBU26PJJR61PBuFZI2qyfNGDX/7SI18bTm1YtLsm5S6sasW4mn1y+mzxCTNfbOHrk0Xhbv0aKy61ggcnv7C2BvrbaWyyjYqZfTb1Ff5rJ1efXePsmv8wpM8WXteUPp+ww9dc5UcOFDYV4/mkS4esWlxktMWmjvLPN1/Z0DeWnfh8sqc2DqOY+FfI82887PKWbfJRMU6eNjvli5147Ihj5QBAHUYwUAAKHBhK+Go6DNJB1QVqLKpdsvZ98B3wkjuTxQ6fZNTGJJ8dPDp8NGYcnuySNU6ezJw4+uTVEitZiokhzwa+WmzskFXTKx/G2SZjEsnoG2cn32IxVv7wteFVyh+f4Zl+6fNNln1tRZvdcOCJg388fsRNjj/xsWucnsKmmKLG6BuP2GrD7Q80woqPqmHAZxfxq++iDXN41OWTP3L1+SbPbnHyry1GNaz5EBseYmPag0E/Xdjw2I/4YaM/tOe3vm1e2aSvOPzglbva5Y0N9mEkw4+aXLb0kdrBBdEvXljKlbHsFEM5UWebHB3zjbKRDN34cLQu5JxeayB+uTWGwq7mc2KiQ37GXM6Nsc2OnKqtH+tBnuCDjTwb1meYwrAAbD/I0VHYrGYTnq6j5pZteOk1zqac44WBHeNsGtdG+/4dhBA/ZPJPT+GXTwXFX53tR37gTa4aHv7ki239CrviKD4y2eADnx16FT7JRNr8s9Mcsktf3zhd/XDyj49HNtvm5+iv8XJukOAMPLD4HJAp8UAJlmHEIBn6nKr1FTL66h7v6QAjEca0gVbStxjTK/DGsm08HTbw4VSH398JEgNfMJMPE9wuePwWE1k2+FqJ2uomCO5yxV+54U9s7LIVwYfYIm+MbH9vyQYFT3KwKPyVGzb5sXlq02dLQRa1XLFVfsghsuHho5ywlwy+drb1YRAnXf1yms38G+MD5Vs8Ch+Iz3JLb+qQUfCQunF2YSpn5Nid8WtnIxzpZIuO/Chk9I3Boq0UPwza2ZQHeq1bY+WiHKSfbvr4sLNFB7GV78b4YF+/HJI1/2whMvyVRzx28cjJcbjY59uNzvpVonJHtzWpjc9XhV3j+ddHbHcTDz/7xv3JfuP6bJYXY7XZEGt54Lc42BOHfnJ49BW24Qk3mfDRSze5+ReN6bBV/PKJ4FL4pAcXOTGwTd54OI2RU2BCeIgNcgp/fMAEh7lAxvDZZzc75PSRMfoIT771YeMjbK0PB1c22cZTk4MfNrph4wdfv5zi0SnP5ZR/ctYVGbYUvvDWnzIhXOLUGaac44LkxCalrwgEaQtSsBKl76RBP6d0jeMJGEls/yeCLXrJAawPDz18uoq+MQXVJxt+cUkIu24g+rDg5SPbbfz67CMyLlY1HXjUXaQlmW98trUVNuQDH3WB6/PBprjZKM7ikT825EjJT9j0FT6q/W8LMYSdHyXd+PrhlBc2+UcwwbAWxsBPnkwxwWbeUAuJLzGSMS5vbCN84+YAJdPFTY5feuGDpbXFf3Mw5boIYSNTvGyky7fCR38O20XYIUg7vwvcmR9sIX79aXLx+B84UZu9P6tO1pzBAVM5V4cpHPDHb05gd02pjSMylXyyZZwfY+Vt2jR34oG7/BcrPbL4fJFLlw85Y1es4c4nPtJH9LT5Yae11x5Ahh1yrTM8GPzfFRgUOVDD2DyYJ/kyN/TT01Zgh0esSD8c+K0L/GLnAw7+kw+/w5c2e+TIIHjgK6f4/JBVYBEje2wnp69dPrNHl46xfBvLFt+oGNTGYaBrjciJ68jhE754fJJnn6y+oo9Pr7zwh8yVcdchTPTYhKO4yRnLNj4Z9sivG0hBUcTkoADxCpChJhaYgFRnlFP/t8C/Mn3Oc56z/g0jvQJi06Sp/fOm7373iIrfZwAAQABJREFUuwd/+9vf1r9t5YOdZE0CArpkSKqAELzVxsmryajhpOsf7sCZHjk+ED4ZYzDRQ/rsGCNvTKGnNE6eDclV8kOWTIusCWKLXLbpFy8di8L/V3nxi1+8/hGTfGRTHumygfy723vuuedoQvMpj2KfPunCwx85NOM2pp8vC4scbOzMvLOh4JHRRsVEB58eLM961rPW/0kno++f4/zsZz9b/3DKQqaHxBUONX11pM0unOWFb3x9fOPsiEPu5fNJT3rSyqf/8WEDMw4Hv8XLF5JfZAzPTcI/dPIvYtnDo8NGFx/5cMKTLXyy+pMXZuMwI5sd4oMt9hHZYs43jHgwtnmTl3fznH2+ySjG9OWbL366idM1xqa8oPDiIzibV/a12TVOlg248ZMNU7jJk6EjBvuEf9X78pe/fB2WyMHln4z5d7LJ5Z/dcJlvtuqTgZ9/tp/97Gevf7vdfMKswEB35kjM4Q8rO/xnk2/+XFf+Z4z/Wmod29DllB7biB4f+vSNd/1lTx2GfNFLxzjSx+eXTfONYEFwyxm+OIoRHmOIDXwFXw1PMZFpnA/81kFy+jAg/thZNxDCSKKb3MATNE4RTyAlmwHOIu3dbrcK3nnnnbf+77gT3H8jJzeL+Ac/+MG6sNnkQ8LVEisw/o3BIiAkOSVEnzwZPO1iUUsgeTEaR02ANrvwF3c8MuLNXnx9GI3DE27jxtiTO23+2ICBvJg8WTzucY9b/7TKOFkkTm1jb3rTm9b/vl4D24901WKxYCxe/4tabTH7D4DsNVflQ81vi7HckW0MT0HkikmcdPkj7+YPv9jELm/G1fLrH3Ed3/7nvJMkrGK3Qbzyla9ctvH8m9j77rtv/c93WPlTK8YVPP5m7vD5gg1ZP/DRm3xrii55WF74whcevP71r183kqW4/Si++v+rFpP42fXPp/wDJ3PMhhzwj/T5lRu+1WElo1892+acXGuKLXLllk0ybMpx+Sbnvz36H+lkYSlusmyo5TvMcLOh8KemIz525Vtf/snCqQ8fe0q5LgbrVjv/bCI8uNhS8/eUpzxl/fM2427m/sPihRdeuNYLHhzy7Qn99PafFeUciQMOtmFA8OHVJ8OPG6R/U3zFFVcsuf/vH/7tMd+uuXKhr8Ai3nIhNwhOpXyS1VaXW/jFp2ZDbRzPvLg+rYPWC7ts+Mdsj370o9e1ya8cwtG+CSMb/P97+7e3f//735dtuuTw+UHlFw8GughGtumQOVM/ZJRgTIYAn4RXid/iExQQ/o3py172sqP/gQ2AjSYCQOEYsadtkh/zmMcc3HjjjcuORAkcBuD1YdOeiWA/m+yRL+HkjIUR3ys1WC1aG0+JY98GbJwOu8byO3GYGJNHBsFFTizhM4aHJDyb7OCzIe5LLrlk/Y92mOQfj1/yCCY6Sv7yxa62i++5z33u0v3e9753cMstt6w46LThsNlFvAxvP9iDlxy/ajKwaIunedUnb07w4U+WjDYb8MqH/9t92WWXrU3CxU/GTUdc4oQHNmvDTabYWqz8hIm/cozHBh4qBjw6+vLCTjz/w/yiiy46uOCCC9ZTnZjJKWQU7UhbvOT4rf+85z1vnZTJexLx3/88PZc3mMSl31ymD3c+tMmh5OrPtYXHZrpkm3O5ZM+4uF/zmtccvPGNbzzaVIyj9F0D/pXzbbfddnD//fcvnezZaMmz3Xoxxj7f1pi2uGEKD55CD99TGn/WAll5wGfbf5wk59/G+o+P119//cGTn/zkozm2PsRBl03rwj5iPd1+++3rpC+X4WGbLxQmPIU+0nYD8tqRDB9ixZ8kVnG3DsLOF8pmdo3T2e12B1ddddWKbfunfOsmZ/8QR9eVNn3y+S1/+towiQUGpG2+4GUHyYvS/LCryBPeH//4x4XFTdg8JCdufuiqyVtj1sLNN9+8bJBBxRt2OLTFq2SLvlwkf/QKa1nZfjRYQAVsAQiKYRfO1VdffbT54ZlgDt0F3Q0lJgI+mRzjAaXgOXF4nH34wx+++uSbLLKRBY1MCn/dIOCbwc5E0DdmM2OXrPj4tshswhasxZ0M++J1AdDXZ8P/8LaBwOomRAb+4mkBkYVVoctvciaUb/8X2tNZMRXjfm1O2G8Rlzdy3oUq4pGLf/7znwdPeMITlrxxFyGSD3MEC1uV5oVNY43jKxYdCr8YtGEhKxYX6ete97r1iggWhwGvisi6QUxqI8hO/uQLBqSdfzjzbYyMnCejJoPYFKd8umE84xnPWDfT49sTiDVpnCzsbNJFME0yzo9aQXw21y960YvWOrWObMguyNPbSZk9PvhvPbJtTbAnV/gwstsc0KGLFyZjdF1Hxr2GQ55ML7744rV2yLL7+Mc/fm0c9IuxOOngPf/5z1+4fvSjH601DAd56wYuGM2d9e+a4Fsc7MDLF172tbOdHTIKeXbYdHN7wQtesGT5sck+9alPXdfV1GejnNuk/GtebyUeeOCBxefXOMqPfuuka0gfNvHddddd61DLtgPAK17xiqP5XIa2H/599Z133rme4s2vfNDVdj1Zn/lm2xh8Xi+ff/7568mJnNfwrj3zBVex8NM847ElD3j0YGMXP+xi0a4Pi5zybd/xOtbbCXstGbn2VGcPJeuaZze/MOhH1u8HP/jBgx//+McHnqTMkziTkV/tbNDjB781oKZ3DCiEIUBOKRIuEcYZ8EGtoJ3o3vCGN6yFADCdiI5J+Otf/3rwj3/8Y9llywUgMU984hNXCaxa8tQWrhsIexKG38Jglx2FrD6CKxvx6MGFyCN9+AXtnavHfn7cBDzqesViM09+KZ35UU5aPL/5zW+OcsV/OPmHmzw7+sb4JGeS2Eje5xdiFqNFFZ8s/WKg00JugebHRm1ToXvOOeesG/sf/vCHdUHwi4/KDbsKe/zKnf7MafHSnfzkWjwWsPnSt2HLITLXv/jFL9Y6MAarG5Eb5mMf+9h1aitH4g4DWT7LFcxywC/Cpwe7MfEpeBHffLm4X/rSly7dxmfN1/+i8kYGvkmPeMQj1sUKgyecP//5z2vzgFOs+OzXp99chF9crW282vnj37xbs9pkbLyvfvWr16u41gZ54xPjbBuXLzhPnDixXhmeOnVq8eSJXZjZmHbCKQZ8JP+1XeMIDv4cImysDlznnnvuwaMe9ag1/15b+iwiTGr+yj8/UTLsuDF7YnLjl09yxtVhwkd4SutDro2x8Ze//GW1vXbktzyzBbvPXn/yk5+sGw1M7MiXcXsd0kfss73bnj7Er28fcXO69957D1x34meHfoWuNtvWO72w67NlHDZ8844XDnryaf61Pb25gbgZT2q9scEeP2p9sdOFzQHP9QGvzyHtG4gMIlMM+uwi+tkO6/pFQo4sDAY4nAuLMh6DPoj2qHzttdcuEOQFyRh9soJ347jjjjsWOI9U7EmUR1mnJ+9rC4qOTdR7ORt4WACuzX+TCAc+PT6NaQsueXVJcHe2UGx2FqYNz6PnM5/5TC6WnloMLlafJ2TbAoKZfpOpTY4/43LAF316bYjio4NnDKZsyIe+0+Cvf/3rJaNfHrXFpfDTvBhX8PmxmN/85jevEyh/vrRgYTlVeZ2I5DY78tUCwEPNmxiM8ZdPY9pqvuCHDR55dbpULGR4vGfHd2r8zGc+s3hyRIdtJ6crr7xybWb5L38Th7Hmj64SVnjCwKYLVo75VZsftpwGrSnznS81GcXrFKc5xA5/+OyXB/2wa+PbiNtY6BqXF/7ZSJ4svvgqsClyZdxJkbx1qY/I8uXJTV6tNXPoKerw8PDgVa961dEckbMejPfNpuaOHTb6/FHs8Lg+jZXj5XT7AZNNiz6bbJOPxGYMv3WhT09MDgdsy/db3vKWdfMWFyLDJhv8kndwc7P31COf7LInD8atN9eJmj824NFGZOVOv9jExZY+e0ibnA3evkSGDjxsw9yXKsjLGV/mE45sG2PH4eGaa65ZNw1rL4KNPNvlia54xI3oT//49MiTVfNNzr4jFmsUFvvVddddtw4uxo2l15yIadLMlTa9ZLVbc/Gyx4Z2pN0Y/AodWNc/lCKow6AiKA4ICtKY9k033bQ+CPWoxCDZEpZDpzHvhz0+ukgtEracQryj6309n5Jm8jz+2UjdZV0MbJlYIAtSXZLZg0nCmuTskYG1Pv9OvYfbxee0AO9uO0GYAHaKg99f/epXBz/84Q/XYyg7XgW99rWvXRcvWbkgLz+w8ZON2nCzbdwYIo8sWHYRHC5YFxkd/BYUeTbIuND0u5j4x6djAbOB+HOi8DmIjZw82XCHXS2X8tdYOQtbemq8Ch98I7jEYxNkz8V56623LjxyDgOMZOi5EGxm6YtJMZbdcCwH2w/4yCDzzA9ec86WvPMjD/rl7fvf//7iObHbfNlhX/w2ZWvON6vkEF+sSJudcqCvyJfN5e1vf/taF7Ag2Muj9UgPzzhfKFlY2dYn0xy0doqHDd9a81oQzxqBs9eCy+j2g31x/fa3v11PfHAiPPNzYnvi8Lagk6oc8Yn4VBB5PumrI33+xYfgnnlsfrz6dbB0jdOHEzaY6eORVfTZ8Q2mu+++++DSSy9d8tN3sp62tRG9avYq8khX4bNcGucHr3nxgbe1Yoy8cXa12bGu1PlvjeMjfYcSenOz5lNe5Jdt+nhs8U8ehcUc8MtueMiQN+YbmJdffvnB0572tCObbtCPfOQjl84yNn5kv/kUT6RdrPHUfCm12VDgR2LY52UXbmOwHtMowRSBCAi+cY+nilclNmF8Mm2IHhX/9Kc/rWAffPDB9eQhQTYMSXUidhE//elPP/r2BaAl1IWsSFwbgHEBFhDw+vymi6ffmMD4hatYnMQtcI99bgjRv/71r3WTc3HabN1AvHf1SOuVlsc8kwYj4sMpxsLXxodHEumHTVvOutDImoz/Y+teXi47qj6O90ueP+S0IyE4FQUhxw6ORJwZQZFO2guJ5obReAlJqyRR4hVyjwmdhMREQYggOlGfUUAFB44dPBm9f8b71qfMt1kcsqC6buvyW6tq165dez+n4ZLyRxlNX9S1i6/ED3U2+KaNHUmcXGDakDY+/fGPf9x4LOxki41ysdNGN73Zh5cOfoWxvD529NNFvwWADvSf//xnf3ZptwSTGMDIdzFxrGExdxF4OnCjhpdf2QkrGW0SPUhZOx5U7gKEmw1t+LSZk3Dz2a7dU2+Ez9ga+3//+9/7hkJHuNlpXvEVefq2IfIlV/1iiE8dTokO+iWykrK4ydlBZMQqPjnscDkS+cxnPrPfHWhD+qVJ8Dt+8U6uT7ndvL2jtFkzhxsvC59jRZ/HGpfG2fjBQrc2xKcIXn7pk/A2t7XbGXt/6D0iexNnsaMLr82l9YFNizlfxXDacNR0vo7YvGs0x4oXebrZFxN1+tWl+vGXtBf/2rpe6NAGi+uITxI9CJ/+dOAXW3MWHyLLJ3H19aOnEzEKJ13pMSeLMZ5s019cXQtuGo79HL9aD9nlg6SMlE/JU7+X6eZEGOA0Jt4FeqLJN7JsuSFG4Wjs6UBwItiLpTp++vZLdA0qmDFK6gF2/OMzSMdR2hnHgy7Wy0MB/P3vf78DC8BWvABwpsdrZUGMyGvjtJsMnQZIboKbWC0OeOktkOr44JtpDhJfyMDswvdOJmzs/utf/7r0+uuvb9tsCabJ4d2Or8KcC9JhAvDBRHnrrbf2IlkMBFE/eW0WBDmCS3+xUm+CkqmOBzU4cv3IBE0GFv6JTX3174b1Dx5UnGDJVu1iwHaTjP3Zpj2bZBpnOf3ZIOcpUxz1NbbdTNnOR5PXizuLGnnzxYtLTwKw4pPTqR8GbfRKtePJfjletvCR4QuyKBhPNyryx+Px5rGQm5gXqi74Bx98cPfbpdNJD53NL/q02Th5EW0+qU/7cJlTiOzEqYwflcNZXblk/nvy8NTrSZ2N9957by+2xpGfxUL9Yl17b7zxxs7NQfPPrv/atWv7mNaRofGwKHu69lLaAs43ZJzpC6MyG8UfLm3xyvXBJS7KbiCOUS32YgCTPB1k+M3Gn/70p339GBfHXK4zR4L1u6k4ifD06HpLBwzFKH3awka3hMqTkXddul7oMTfYVOeH9Sdd/Odfc0q7On7z2I3SxlgMzV+fosNr/TCHik8+GRc22RFv7caOTnGCl4w+Oo2/Lxi14UXi5fpqQ0DOvJgbEE+hxthalQ82Dd5N2vjDjmBxTC++5oX449cePnW41POdrHa42Me/sWuUulCUu/Nj1G6SeKTypj/nOeePabyEsQuy6xFASuko8JycxpWBcqTxz3/+c78o83264wEy7BmMLhhA8cvprc6OhI9TJoOJwOmck7uAPFX4HLBHQJgcsQhkMgYknN3A6BIDTx3eV7jZdZNrwYQBwYDo1tbgh51v+VC5On5lMhJd2uiCT1nSzm5joEwuWf1ipL84ZiM+eNKjLdKWXu1dZPQguLTnj7LYGKuw5Ree7CTnQsqusfJEq6/xoyN/2OOHNjrTpT0bytmJtzZ5dj2B/vnPf94vU/nkScgOz9k77I5Q+QJbY5fvcCK4PvWpT+2F+fJ6ki5W2l2wxYie7G7B9Y82fjR31ZXxhT8/6dPGR2Xz06bF0715zK7+yAJmEUBweLJzQ3TMZq6bx65Lc9eTqbIbDN3Gjl08xZRufeGFMZza8i8ftZFpjjpCdTTYNZLv+PG6Dh119rdKbbzgNx/+8Ic/7Bs+nkhsyeZ780G/drqRdhQ2ZQsyHjq0S8otlOp4quOFOR71riVxbnPpJMMYiqE5I2/c5PTCmz5tynK66SSvLIWLbX7QiegwR/sbOe+P3UR8GGBMjSdZMtbQ5oIxlMTR+LIVGRvr7quvvro3Jl5HwIUaXzl5OazZUEd8gZvcmUqOpoBATnMCUI+Uvohw98XnOKcdBaAGggywJgagbjx2bV6s9eUKAHYayXIGAUrOYAYODvrkSC7p5yB+2GuHNSfJKWvzBOIR38WjbmGAV92F6eWzHV8vSC0uyEXpE003SP7SZ0DYz45ymJT5wYZBhU3Op3CKUX3k4MxHOsOMnx44JYuaNosGO97reMx13EEnos+N0c3YDbr23XnyT59a85U+8YRDOTlY1MMFtzLKT7za2NZW3F0E8MOMpk7txYgcvckXJzJhCQceeuTFLNlw6UP44nWEerF2xnRbqCzMNjv+mEqsYDVGMCVHHxtijw+/p1LjHw5f49n1mc+ND7niRFe45OTYQOr6kRw2TzfGz6bMztbNzV9j2y3Cka/0wAuT8XOt8fHRRx/dTy/eQRgLPtPlc02LNv3mPnv8oFOMux7U6dVfgg82sZBPDMpwepJ3HTk2FA/61OGkn8/smZP89gnpcT0RwohsRD15uM4sjmTzt1jipU8dNvXkm6PhxCdF6VInw8fmjzq5dMFO1tpCb3oszmLtNMacsWDbWFoXWofwI/UwuK7Yg4FNdfi14WnO8cuN2JGqeKoj/OaZG4nYeU/suPh8HfPBRN4NTJ85yDYZdfPB8a2jwgi/+eJm7ViX3+TDiw8+8njFBsm1zViT3UdYGmPARGEOUGLyOaISBOdpgmxi29WYFEAzasFtkXbReQno6MKujV7yiG4OOI/tSEAbng1q2RRgbYIOgxzByV74BB0JwOSBpwHtRqFN8BA9bNih2l36q/nD4bD9yOduHs4X8dMvp4e9OekMmj6yqEArJ0temgOhzJf06afLjcI4ZEM7PjHWDq9v0S1q9IuDdz0PPPDAnmx4kHNUT5R0FiMLpkXJjT/bcOeXcvHnR3KwwIu04VFnv7bK8NJDHjZzBtFhHqjL819OhjwcjS8Z7Shb5eTjlWuXyGafHJst8Oalb9+18SE5OLWRVZabZxYNX5r5hJI/ZMwZ8bMz9B5Im77s80GZbu30KosX0p9MvjU+bkaemNjXBoPNF1m4kAVWn7qxNO533nnnPiJ0rbFnkaHHomyBdg3QKe7kxE5OTzGAS8IHfxjZzB9lFG43DbbUjXP+0q8NPsSOxdec9eLcQqjf0bJ3ODaUjmo6xYAJ4ZHQxKlMZ7HGwzbs+mAv185vPihLfMeLj45siLO5goqXMXCsaA6IPbJ4u4FI+OlE2aePffrg1C8mcnFi3zjhS8Y1a27aRCjDDBt+2H/3u9/dHEeb4vyx8U2ndUOcHbV1fAU/otMNykmKJ1QYkBjAoQ4zjOzK2YAvPjzZ3Tycq6EyAcIUK0t2Qy36lHDSYkYJ6iLhSINh0lq8OMCGJCguQDtBDjmbw8eWevZgmcQO+TkpyNTWgKnjLRjaDbAEm750wSLBZ2LIDbgnLrs2OwEDZfeUHTaLTRMizPCGLzz4lcngoyfclZMjyxaCA7/dDuz6tFlMyLvx6qMb0e8jBbsKvGTx8Vlc2YrPuImvYzsTjt8SGXqQuKhPErf8ow8e/NrZ0kevuthok8S2Y1ELpLNZcnjChc+YyZF2PPmnDg/d8uTUUfHWR4d6+sMVn35x0Z6ebNFXmxi4OB0JOZvGg9yEfO1mAyS22Qkj/cYRP59gKm761FELCJv4xBMuFB8/6JW0qVtkLcwdA1qU3UD6ys2i7ojDBki86cfPH3rUYZg4tBcPWNiRwpLMblj/hImMTRns02c68pkeG03vPHyE4Oah343Nouj8vgVRu7gVM/OS39mTT9InwYfYpKNYqcOG8nFX1j9wZSd5PNro0GY+w+M9j/fArkVj7mYndQ2kOzxssI3ooQ8Zc/L4pgwb2vAZV3qTcf3g9RQi1nQgbXjEXRusjjH7vL6jNrzmlRuHd49ueuYQfHylB9FRmV7l/FHGq85eWPfPuecgIY5QLueExYaw+sV6JPbI2uLFMaDJ0aHdRPHFyg9/+MP91Y02QPUDwAk7Nw65C2pXpl+ZzQIvzwGyOYsPPzLBEBzJKcNcoOlskps02k1YA2LH7k5toUAWNy/F3Czt0F2U/MsHWNmnh50wwYnk+uX1s0eenvrVyaYLPscQyHfmFgU312Tyz+AhC7IbCD10dhPYne//w29+WPDCJG4WEzcP5Xam7HQTYQt/dbGUwsIe4he++rSJi4RHrPzhqE9JHRHadFh4fe0Gu/4ZA/KNm3a+5Z86+2zSX50MDPmXDnrw8FEiy564mbfGvmMBWMS1mwGbbJgjNhN2+OZHfoufp3I5ObrZ0h/m4gCXvnSqKyM8lV1z4g4THhjDBDe/i7M6W/iMLb6uJTtM89eHCjZ9fEP0mg9kLIL0s80m/NrplBpX/WzCKZfCO8vi202UPn0zFp6K7OCdRjjDp8/7EsdWcNpQGhcybnJ0IBjp1QY7YgvW8GibtrKtPazKZNQ/yKfmkj7YzAN6xEkMLco2buaAOPp4xBOTo1BzA2mHW+xQdoxNfjVHilVzE3/jjhcf+/hQc75NWP1iwg5Z64djNWTdcGxFP4LLZsf7KdjNn+YeO3TQ2bgXa33SnBvKqHy/A5kKGFVnlCKGOKXdgFpULDoUUK5dDgA5dz2fwTp7824BmaAGwstAE+bNN9/cCzPA5KYdvKiBYBuxgU9uUMKcI/Xhl2rHJ2DkDAh/LAwmspepPpf8yEc+snG4GH214y7tJoiHXonOfGRfuZ3EtAUrXn50YcJAB9tSPoodPfjdMFxo/Lbb9f7IBIabjDzdu7D+YZcuC5ldpxhPPpPKgu1c2uSjG3Y52+SV6YCvsVAWK/30hRHOsNSmf9qMPz+9//Jk5LEaRpPYjUS7cRGb5NmjV1vyp3bUJy5luMjoS1d1/RFf8earMsKD30JFh+Si9C7B11o2QUicndFb/Dy5kaezXalrRht9YkqnfpiU9ZXo04bqV+Y30qfdOBkz+tLFNlvseGluYVZ3Hu/68vQhtq5TlI5io04XmfrZ43c4tVcmVxzTwQ/9dDUn1OGMV5+jHryOXPs7LLrdgJ3t28WHhX26WjiVw0S3+rTFDjzGc2IsxmSV9cnhQXJtsNLHvrJ25eYlvday4/F4cwNhXji+dDxo3CXYyNHFjjrZcGiHNRv8a66RQ3jhhas6Gf3akTrd+ZV+17kPnBwNus69ZjBnkDXB+2sbSacqjmTNC1jgYM88CDMb6ZWr45GKH/vkyez/E50zGGNwocQgQG4aFBDEJyDKOdJO1a7Wz1n43t+NBC+i16L961//ek9yDggwAHMA1Qtkg8lGOpTDgY8snNrU6yOrT73A5Bu7guyTUjgtbG4Ebh4uvp/97Gc7+O7iAkw/XZK48Kl2eIoHveyFpz5286t4NDB0GnyPrC6wK+vnBexEDT6MyIS1CzZGETv6LcLp8MebPtFkv5iZGOTtRmCQktVHZ9hg0hZuuRTWysUZFuXyWWZDPJBYNR785xs7kotDW3a3wPqHLL/olNTx0YXyQ5mtfIMxwi/pS55t+ow32/SJgf7GTY7P1ym+erHBcAMhB6edp4XPhQkj3+jSh/BFyvChYlC85WRqhwl+mBEcxq5duEUNL8zK8r4McyTkpnbjxo39dMeueYTwFgv14qq98W8cxAGuxlq5GPJDwovwsgOT9upilw7tnuBtJMUQL5+tBZ62XU+NKV7YyFtP8NGpnR12i6fYwJVtdTaRdnqSU9eP8Ev0aMOT3XJ8zQtPHr688gRt4c1fWCQ2xZEsO2HUrk3OB/FULq4wVA+vXD9c6clesaBLPz48yXjy8NuEjgetW2Jo3tgEe9q3IfYRkbiyS58YqdM3Y0GnhAfBgMhJ+QqDdOYfHTPohAQxIQoKCOXV5TlJ3nmrCeNxj4N0xy/QFmk7EhNeOxucNZHazdMT6OS1TYwcJh+fcvjJwCpNnHhMCD88d++99+4vmMLh88I31vf0PgzwyA87e91IydJn4FGLOb7ixC6Sk8VLTh1OvB8UQ/29S2Jb/EwUySDa8bqxORqkjy7+e9fhWMBNhF58Xvb3mE0vbBZCsRFfuPAaC4ku8ccrzbgqI7IRWXyNhT44tccnzwa8En6kDBMqRl1MZPSrs50MfXDqyz55k1+dfbwlvJJ4kwmfNpQOevGwhwdVt/nxdwqeBB0NkUFi6GbuCQqvi5QNC70+xA7+iTfb2dHHPtv0hF1bcsVg6sNn3PBZ1JzL29U7tvJU73gCj8UOBol+ds1ZuunDQ7+yNmNaWV887CjTgwfvxNtYpUu9xd/GyNPFY489thfhNkRiZ/fu4wO4zHnznF1lOmCV2A4LDNXDRUaCC8VTWbsUfj6Eny569LGvXZndNnE+DnDjcw25TujiK9xwkq/O7+zSXWwmNnL6tOWDNpT9YtGY8I9teXMs2fwn72MZc1UMkU2pI0IfP3ntkAy8yny1vqVDHS74EPzq+vVpVy6GdCjvIywAYyCcEnmTo35CnIyPUmeBnlT8TLPv7Ht0ZoS8nbEvLdTthukScAAFkA5loLNdnxyxi/THY7DJ0EW39hZ0GNlW1weD3aRF18+Ne/LA46sHu8r+ENKEgSn59Korw+MiRhYQwYWNbX5kT//sIyfBhK9Bgd0i0OfM7NPHFhwel1966aV9Tkynye2G4/e87EC7gdBnEhYLZVjgmjFWdjNCLupww9Ykpit8+JQRXtjgh0/ZGCg34fDhxyPhST5+tshoLy54ERt0JastPyzU4td8DQsZia5J2hBbqHHTHq7kqrNBj2NNFyXCY/NhgXZeL8b5rT/ZfMkeH8jyB25JjLNJjz649JEXk+JCfs5x42Yx87Mhxt4fCiJzxPVFt/lBl3GmTxuii159ytLsg0m/HH55OItz40JHRA8/5PqNlYXO8Z+NmiNAXwrS5Yur8/PzvSN2WgErGXaLIb2VySD9+cMWv5rT+tnGG251MvXhry0f1fHTw55Y4EPw24j5DNZRMoz0O3azo3fDRuTpIaef7uKrrF9SltLP5pTXh/I3PNrxZp8uPPQo24zj+cEPfrC/EBN75D2uOPtAwbGVT/67CcFHRt440i8Vi+zQxT9JfMhJCD8c+7ewdGJKoTInUqQd8JSZ1IRTbFI7gvHZrj/Ft7AiA+HcmxOc9EK1nYh+F04XCEDsTcpROVt4ENvaYAqX9vDinY6y6csEO7bj8bgXXTcEL/TdOPzqp6cmQc7X9LPJBirXl/98YC/s+cF3MWwyhJWctolPn91CN1m69Et0i6dJkIwdZhc/XPRFJgF9yfNTG510sU8WD18l8iYfHmVtiI5wk0X8w6Mu1a8vu/rZYJeucLv4pbDpR/iT7UY99dcPI/t8yC/ycGsPo5xudvThR8YEmQ/ajV3Y4KXfLs4TSJsgdiwcPof1lKoeVnNXmW2kj2326JXri19dGeEjL9fGR0l8yJDXD2M64PK3PzZB3s/hsUu2q++60gYHXcriT55vyuxpR3iQWGiTwq6cDjxhDL82/fS6bmAVc21udN57Oc6GmS5yfn3al402nDA1P8LlqaWnOv3hY4sO2GsjC2sxlfMDD8oP/MZWXd9MycJGP8JrYUZ+osVGwpgga5knPe+a2miSg1UeNnqV2TIW6qix1a4/WWUJDrqQuPJRmvjyQ8zc5NygzQdx1meuesLzB6jw9t5JbNINE9v46UblsGUPnzK5+tNxE6+OBpJwzpYXGIDxCvBsM3GcFXuf4PemPIno125ndGOdy3qJazeXDnaUpZwAKPvK2oGla6Zw4YVl4tGG6CWjzwtHfzTnL+l9hueM0C7TY50gOx4ysGT5R84FIeACKC9Y6ddmgGERO0SOPNv4809bcvgl9XxKpolCL3m2xbALNP34w7MNr3+q68OHyOYTm9rxpUdf2MMjDsnTgd8YkJ91bfq0h7+Y1w53tvAg7w2KGz52S/HIayMvhbNdlNjgMQ744aAPhuYE/90w1OHFp5xMfnaRaDd3PZ36Kq9jIE97fsXAPHGmnA5+INjCAoe6RD+8cjZgiz88MCK6iqkcFvx8QnKLBVyeos1n5MgSNgsa2+zIpeRhQ3CIgVxC+vAW8/q0Nz7a8KD6lfVrx9v40OPmYQfshuY4li/6HZc4VjEm9OClQ06PpC4mp31s0JM9voUFb3ONvHakHaWbDoSnsZg8+skaX5uH8LeRsGa48bnJzXGlHzZ6+Wn86KCbHW141E/tkWF39pMNa/OAHjwRnTYOP/rRj/bG2LE8u74a9bK8n6yBnRx94trcYoMO9tOr3Lgkwx5eOLpO+BCv/ps/506hTsKEKCvQAEgAWGTwqZO5WJ/2ujl4ZC3YJosjGbu2l19+eU8kFyQKOHl26GQTqICxq0/Ch/A0OZTpUdcvl+DiLJ10KBt0A+6dgZuHF3i+SvBHkBJ5ux4Bxs8/dpW12QGyR3fxCDe8SB8iUxxhQPqSledHuTZy9bkY1Mm3sNBjAGes8jkbsGmjl7wy38IOq3pxKW76Ub6p60MwZacyPWHFV1lOhzbxUVfmg7LdkPkg/sZDnLIBL/9Q/GQimOBgGx/Z4oNHH9uSdrJk1MnI2TA/YbMYFE/tbMKq3eLc0RDd2i3Q3pNZEOmmT14ZjzJdyvrDFD544jevzDN2UbEgo5wsfnI2Zp7u/eqAo1c+2ZxZKBxTqLv58UlsyWkLl5xO7dGsZ3P2i2NyUya95CUk5zucnpBcZxZgPiJj7unDosefdLM3E1760423eCqzEVZ1fPi1T1l92pHxDqc6fYhM+qe8sXM8aEPcmuWmAb95gDx9WBcm8WParS5nX46KaZj40/zTJjZyqXZyYRVLx9b+uwRPSI4H4Xes5gSjz/YdD+Y7n7KjDT/93czpVoexVEwmnnCli9xe5RQSTBmjBkGdMoY5j09ZUKX+MCiwdhk+1/Wo52io/+yIHF2nAwgUWwVJDnSTTF9OZ5+e+jmjjtiIlBssA1FAPJE4y3766ad38NlH5WToTFZfAVdGdMPVEQgs2vCJWTi1Fb/81McffXi1Jzvb2KE322TUkyFHF8o38oi/+Pgh11+f8qzDN3XmezEmxxaZJjTftYUdNpSvyfiPu7yItqiZ2P42yI2cHxbRxoUeOvgz8dIXVZaX9JElN/1Sb0zCqZ+cvvCTZ1efRc9xhQU6v7wwd+ODSX/Y9LdY23wUP/rYoLN41zZzsnDgRfTB0JwSL30W4G5qve+yOfP07PNs15ejLLbw0RMW+thQp0s8kLp2MrWxi4cMHcryxpFcupVr10YGv2vBRhJOx4BswO9rK9cbrF6qk8WPyE8s1fXjCysedqZP6dHXGnCKv3GmF5HBXxk/kuuzoBpzX40dj8eNX7tNqCdQN2wbjPDro1tSpodNeJTFFy8/kDpe4y/X11xRj7SHSZ6PYgybp1B/JnF5ffavzfs5R/H9DpkbHFswsSWfWNTRac4uW2HDwx8YWlOSo2/7mIAKBZKLu8FjpAtdmUIyHu3dAR1fWSjc8bR7UfbUU0/tXf7pzYNuOiT2ACXDlrpg0o/0GYjKeBqIsOZoA6OfPLw5bNeJ37mmO7eXoLDbTbKH0kufsotRTo4eZKDg14YMDHxsR3xBcrrwKldvEIsDXnb016cuiVGkTEZ7McEffrmFLDvkiiMM5OFB5PDRh8qV8dLPDn5yeFH4tMfDbrjwqouXC9HXIP4rUz+pgEdMzRE26NemTC994dSnTY7kMz4TBx+l9OCHT8KH9CvjYcfcRuJVXOze/V8WPtl0g0unhdr/cWIBwW/M+UcPG3Rqp7NFgW48/KOnMSmW8nbm3UTjJ0N3seG7G5f3c3bDjodsznxq7olIH/5000MGtsYGLhjyFT76kTZUzMIqL/58xC9pT69+iQ4JPjcOixcceF1zPiG1mbRzNg4Se+Ghgw05mdor65sYw0NHuMgoJ4uffHJh1689nfqVtcGl3w3a+FTH05i4sZvbrY9k4SCf/fCRq085rOYEX9Xxiie7bKD6aqNbm35/tOg9bjdpOGxw3Di894AdRtjpbf5lmy6Jbon9CA+8k/KvtmTop1vafwcyAwCoOiZ5QdDusVRuF88Rd0I3D+3IImuSm9wuSk4ESp4uvMDkjAmXU+w2mfBxojby6ninw/QgbXhNAHywWshMZF9a+VbeDcTnxCY7fGQlcurkDEx45fDIw1l/O8V4Noj1T/joEmQ4wqYPhV8dH6JHO3tSevQpTz+T10a/hYzf+CwaydONBz9SZgfJJTL8x4tPTife4oKf3/xB2smRzxbZbDuzt5v3gYX5Ilbh6wItLuTomHhgYEtbPtBtjCT84cOTLrzqZPNFHW+pdn6Yp45d7Oi8+M0u/S5YLyh9BOL4hU46+A2DHB/Szg7SHmmjU8IriV0+10ZGjMwxCwF/bHK89/Aymn5fAPkrc9/9NxfJ46UfJhirs0Enoh8WbcWD7CTtJe31k5XYKKdH2YJqjP1dlRe6cKGLdbTthxy9Z/SSX5zNUb7bTPCHPvbkkX7Edvbwqmebn7WRV8eLh/9IO136ELz60tEY4KNLv3npBkiPtigbfMCH9NNNn7K8pE6mGCkjdfFprhobOPS7ttKDt9jAbMPreNB73OPxuOeldkdq/vDZzcP7OeswG/TRJeVnscl/OSx49LGXb/qKFSz640m3XNqjVfBdoBopoEx7BrSrm9wmh7929MmuoCKyLjKLs51+fxVNHpEPFD0RO+zpp0O9tvj0K5fjpVfShrRNzGyRESS4/J2El3vI+5p2jukQaPySckEz2Egb0k4mfnZnWxjl+uQmZjy1yWGkK9/yRT1cfFLGP9sdB3jP5IboAqXLDj/c8MGuHamLRXVt9NEr4ZXwtNDbEOiDQR4OF064tCHYlY0hmmW8kp18T374XTT42UwfWRhh0xbGiRsW8vokffVr158OdaQtncrihOBpLqvHp99xgc/OnX/TA4/5LxcDvBZQ9ezwm23y7In79AMfHu104BNnddeSnF5fLNppeqnvCxt2fKnnKMjPwhS7FiF1usKijGCI9NHNPpKrS3DNhDfST4+c3skvRp4ynUZ4mUun37iyqPnykk43GHLKjYEYFYN0kw2DPhi0hTMdYTttn7ywn+rVT1Yy19lCteMXz/yTu868b/J1kzGwbpBvnMiH+VQfPm18pUudja5DbSi/woRPImfz5d2Xp2NzwcbMJt2a5snD0bB55Npni056+ISKo/bTeGgL+2Ze/6iThQk/UsYrhwkP0n8mIBmWN6kxRgkAjkwWj6yUSgLrixAv9HwV0qROXjBQTgVMXqD0BVJbPHC4wJB2weEIgstiZ0AsRAUuPY6v8JATYHV98SvzFw9ZiT38SF7Q4CmQ2g2MesHODp14tUt41bMhr0+OH9VeXa4/m8rplfuPnLw44xN/jEGY5PTRQV6ZX/i08RMu7dnVX/zEhD194qqMv5wO/OHOP/z4zKk+iw53fXKLHRl4WgjV6ZXIlONvTPKPT9mWn9azVezIwYXK8wsv2yj7ePJJ2e5uLh5kyYWRnfRqr67MF/gl/LXR0bWnrzjJ8Xga8um5pyD9NgkWE7tMO9Jk4GabbgS38RNXbbDQl4/0o9r0T134aqMXH9JOtjmi7Dp3AuEpzR/lipENzYsvvriPrTzlw0APfv3kwwojrPphQGzrD6e25LWFLfzGlk7xzPdk8cQvV8ePyNCbbRsDcuHYTOsfpxbeN2k/HA776c96ZO6ShTW86qiYTV3ayMklttjkP4Jr+rIb1z8f//jHL33xi1/cNw82jb0nUb/q0R8Oh0HOL3HIBgzZhI9dfNrV2VRvTmsTK6SPrH5EN7LmIu37v7SNAUAGCoQyBSl1A/F3Hn5zxW+vuABMCv8drbNOOzW7KHKS4OQIGw1iDsm1Aaac44DXznaYwhFvuOX6kscvQIKiHT8eOd34CkAT2GDinxNj8m2B9U/6wqR94mFnDqI+vIgsatLAqE8iox8+9bDSl39kYWpgTUiLCn1ukOTELZ30pYsOY5U+uvCGSU43HGSUYZDIoBmPZLVJZJqsdvR2TOZIdvVbZMQX3xxztsNKLx528yUbMNCDaqMfX7l+skibpE2uXayU2bOAiSEsKD59iF7zYupw8bAdXnMm+2TixaOcD/jZTzeb+vBpx9ui792Gp3g3YfrtNv1UjfNubeLDD9iMKT3qrsfmVPMoG2GRw5APbJOJL7/jUcdTXMlJcCFHxJ4w6PUZqQ1kX4bRO2+SZLTBSn866Ef0wsGWhKe5vhnWP/lBRlInVzs5+rUp00ePfjlSbhHEk394jQtZhN/RkT/a1e7IyLGcl9Zhk5PjCzl4s0N3+PBI6njkxlEqLr0icMRnLhyPx33zgIFuY+1PIn7zm9/szWPY4adTygYMbISrOj35O3FqU5fqpx8lo0yn/vIzQDEyjtEFpTNA+oCS127nYZDcDfuc0MtGL83nMUoGCx5w5KTa0imX8Mg5zmYOmXja5XDCZ1Dhxcup2vHRHy8c+snkPB7lMNYPmzKd9CvTSx8syvoMJl6UnmIk1xdePGyT14fStSvrH7yIHfpMcDzFKTn2w8MGvXj0w8tnk1J7femWzzIZdXzp4huKDwb9MMWfHnZR2Op3ROiYxXmyo0w/H23x6wIhkw6+pLuyvuzyKZ+zoz9b/CXPZ+34yWpDcvLaJuWz9vrCEZ+Fwm8Imed0y/GKlVQZluzhYw8pa5eyU06m+OW3uexLJk/4nkDUHb363PwnP/nJpcPhsOcFH+Gnnx46p//seTJ1bXcDxMMOm+SaM8nDix9fvulD2vQhdtUlsTdfus4c8/zyl7/cL/fdANmBlT5rBkqnHE6YwlMc5PTrx6dffcqf9qlH+JPTFmbl+mBqroudjS8/PEHRlU1zWXIT98QffrrEkH94k6msPwz6wkdGKn7a0wlneq6s38XzZOcPtF07nkCNga+tHF0XT3bySS5FdOV7fXLxReYXUsenjwxsyl1L8E2c8Fdfcv+dEAQmY0zaEaXOgx2d+HTXY6pjLHdkL3K8DymYFj/gmnQTFBtSeuWV2Qg8GYOcbTJN1DDjbbKZDOmd9vSrIzlb9ODVp4zShackBjDI8ePBL8UvpzMc9dGpTBbPxDbL6eETH+hikz6ySF929UswtosiF+9sNx7q+JURvZK2ZMKAN9v6jZ82/DDjV5dgwiPp00YPnC5CH1LYLcu//OUv7z8ic66PFx/+4gAX3VNPtlzc7Kc/m3TgkUuRMj3wScp0SOZkccAnfvx1Y3CB4oWfDfbeWH957nt7X2a5YH22jhpXsvTip1sOK7uwsTXx6Q8zWTbU2VW/vD7LtHD4mw+Llp/fefLJJ/ev68LQImfzguCXyNOlXTm/2cs+/fok7SjbZCVENp/4SY5PLbZhluMTOzcOTxrecdlEipP+TijMI2X8zUl2UPjCxhZ807ayFC/dyuHHL9Gpr7gXY330N4+U8aWHHF7JE5Xf9PP5MbxkvNN55ZVX9lOgv+w2DmEJN13K4kRffsDIfrbJ0SuJkxu8PnVx8vRpDvi61X9gZb39xS9+sZ/sfJSi3+aALbrYUiabbe35lp8wILwoufTgK07kUXq082MSzGT/51e/+tX/+WO/LgCMGSWQ4waQUcoAtcNwUTkfnp9o6pc4RGYbeX+wAtGkNNA5JIAA000+e00CsgXIRQOj4JODC3486vro1p6jfKE3e9rrw6usXzn/yZv47GpH+aNcoMnCyT7c9SnTifAgdSkbdEj0soPYVaePHN5w44VFXRz4y7bFQwzwmuBk2dGPXx1p45O88aEz3Gyr4y8WyemjX1946tNGDo8nDccankbNDTcU2ODFxzb5dFSmW1853+lD8ciLJcxsollmg5xUuzic4iYrFuYyvOw6piWj/WJ9LOKI0KICFx/0SXjJS/mUXW14wgqvsn59cMUjxnS7lryM9jWTG60n+8cff3wfC+GFnw68KBl95k19xh62iTV57eyFnw59UrEpbs0ZYwa/ev7Qk6+eMm0sxcgNxEtnvIjeFlvXKrvFQL8yXRJeeCbG5je+qZMsgkciS27GddaT54cyO80HerTRYczp86LakyD7MJu//MLXHM4uXcVRv7GgyzVZHywSqg2W5kjzlIwnDl9beVXgd/qeeOKJ/QTCfzJ46CJLV8S2lC0Y8MtL+q3LiB7tYQifPnzpxoP008fv+sVwX52UpAizDgYCmXJKt9C6AARUv8Ui3sDTwbB64LUhdQFrAICJ8GtHeAAu4PgK3nQuG/rISuSaTPoRmXRr0y9v0Z79W+B9GXaLA9/plrrjhyU7BVz7aRu99MGhL94uznyecngaG/3hrg1vmPKbjRZsutVhh0m9SRQebfSIIUq3MhkJjmxpR41H+tjAZ1Nh8VXGY3GxQKqLN34Xa3X69fNNn8SWOMGGlPW7gJWLoT5+I22ITW3qdMvpYU9duVha2PxthV2nOCF++DrLEzZ7YklWO/ywwYtfrPRJ2Yw37HRqQ+xHM+bKsNjtOjrzy9D+svhwOOxY0d0ibFHmI13FSjk89ZFRhqMjV3bi5Y84FDcY9fNJIh9/PGTJaGfPH975TTnrgIUXH7nmkDJSDyt5RAdsZPrApZgll111+NjXRpZvMMTDhnYJr3b9ylK26Eo+e3htdoyrpw/vc9PffOjmDDv+ZOmT8gvGMOBF+PHIzSey+OnG242HD/7swNdV3i37uzoy+PQh/Op08E9dok/dnM3f5i6/4JL04SNPNwq//nTio/NUP53pOQM8AUYwowacYoqQPgbxu7i0ZzAwGcRPNuNk8OqX2K1fmT36shUfvZJ+AyrNiUgn+XAlp80Fl0442JMQPIi8FJ88XOyaVGQNmPbwkVcniw/BhvJVXwNUO10RXepyMnwkw4dw4q0/P/VPHNnLd3VlRGf8yvQWz9rx5RseBJekHR8/6pv9tcdHxs3BTYMcWxYv2NX15ysZdb40wekujvpR8e3CoC8d+rMNn/ghMnjobWcJg35JH+xswyrBgtgJKxkJhVefRD495gdddMslpJ9esvBJ6hJ5xF/tjoIdDSN/8ezYih/s088mUicjKbse6OcrPS5whL+YwcWmNtjwhTF88ISfnDr7dGuXlMnxSx0GRytijBobfBZc+LRN22GgQ2qM2KITr3zaUYdHG93xsKldHckr74b1T/3a2SavTU5veMQRVu2X13Hi1IVPzJByCU8xhj98cn3JVRaPbGtjO3xiYdPgU23kuMyRlXZjip9eRIbusIdHf2sVvNmlQ5lvUnOOz/jIh53OZLUphxEfebrQLeuPqK7bgU1wmBgkiAKNR9IPZCDwMUoxYkxC00E66SJfH5n0p4cN8vhLeNiT49MOAzJw2vSR1ceGpC1ss2/aTZ4uNtod00lG0s6eCdZd3cTByw68JodycaAPaaMDFuV4yGinh/5ipi3/yYefPNI/c2U+52v4m9D6lU0WttmaOsmFbfpMrjiFhzxbEb3w0IfCRk8y2ulXJx8p4xdX/HMh1E4mrOFjm55JeNMN78QIV/OErDHCr53N8MjJdiPDm168kjYJ8VuZXXxiHg89cCcvR9PWlNVHj3ngSMjRnzL+cMmbJ2TZSK9Njvj4bBaGbiAwNlfhIROFjV27/24+YhLpYwtv9rTh0camsjZ15fjpc1PWBgfsdOQT2Qg2pI8uxA+kDa8UHm1R/fhhqM5W9vBWxhNWOsObzpmzR682ftBRXRs92sLW3OI7Pn53zeEhIyk3r+lNl5w+T3JuHMWFHRR2uKT0aE83PmVY9JuXdCIy7MGm3XVBFk6EH5HHq08uRuX0In1o/5/oAaAoIJQJBCJsAnShU1jg9SdTTh+QiKx2TpCnU86WPgtytujUHrh49LMvGRByXRDKBQSfC4aTTVj9Egz6w0gmjPImghw/XIINK4KtSaR/YqU3gl2/NvhR8S0OMz7pYceOTT19TUh1/tLLP7xwtuCmd8aE7fCnrwmpzk446aLjtA67Ngmxiw/J+WE86WIrPu141flAv/FgXx02Mo2byTyf9PRJxQle+ht79vWV9MGTf8p0k1PuSbT+FlUY0yXHCzPblWGFjyx7+tmjWxs78Z7ipj/sxUc9XOTSJUbkixVb+iN8U4d2fiRDZ3NfjM0l9s1jlCy9bFjc+aafT8Uy38Oujs/4IXW8/C5WeCU28Oaf8ULix5fipJ8OpAyvWBYPOf3ZiBd/WOVk8UTq8KH8oUu7pA0+/sOqzfxlX18+kMeDtIdHvXK25LDQp6945Q8ZPEgfvOphIMcGogce/caSDpjyAXY61LueskcHnfrkyZGhly4UP5/ZUcePR0JyfRK8MNEDK17E3u6nQIfBBopAAQ2MHA8QhNQ5CKQ2yuiRS3gzEpjqAQMkUNmhix719JIPLB36YNRm8tenXaKfbuX66KA3jPQoT38KDOz4yesPB39rk6vPASUvftkJK1596deeflgRGTHHK4kr/vyUay+uycVPnk7tfPbprPGEHUZt+YpPX7EJL1n65Omvj0yxwBP+cm0RP+AvFuzDL2mnm/10J8s3Zbj0qeOVN0+05wtedTzFVhtM4U8f29rI4oXDRcGnYgo/7ORb8NQRO3Tg1R9pY0M7HmU62cCnvThnmwyqTzmcMCrjIS9pK2+OsKHc2PYe0k2Y3saKbmW26Q2XuhjIXUPmcte+uGiX8pueKasvfPrUER7Yy31lVEzg4ksx004HWW1y/c0bbeowkC2+8bFHHiWPT0wQPjaQvLFVpgvG2sSIDm2oOauMRzs5iV51efFPln38/Kaz6wBOsgiPOplIma+ui2KiL//0pYN89uUSf8JcH3zpio9OtopB1yH99MqTxxNGuos/HXRLO27+IaghRwJMKUF9KcAvAccYGf3KBh+ZlIwCEX+6fPaHvwlMTh99LdJ0sgeHlCx92dIm0Y9XGS+qzWDmU4HBQwcio4w/eTiQNvbsXpONX10/XcpNRD6R4Udxm4OorB0p54+8eOlTL7eTdFGzgQde+uXswZANMtqdSZOhB2YLi7iThxd/vJW1F/f65NrpxAdDfdr4IuUH3urtbGHU3kLFH0cmMJET33j5wje29EnhCwuMje8G8/4/8ODXH47GSyzgcGGLAW5xL5IAAEAASURBVAwS3S6i7Ju/eNMPj0QvvBODMqz6mwswsC9O+RcefMp4mi90qCM20IxzcZgx4IcYa+Mn2zCHUxu/+RoedbbxsS3m2vCSk9MbBrnEDh362Utf/bXTJYbZyKdwTvz5LmebXfx4zQs6jBGdeOTk2cxH/eHWh4eOeCqrh0VsySC4JWNPVrsc4U+3HAZ68CP2wh5W9eYN+XRln2yxJJ9evBI+pF0ZL6IHRvL6EJtzfuprPSAHSzbwhVu7o1FrA75+/ohO/Mah9ZtdOOR49bNLV7rzcd8GdUqEDA6nAs2wMjA56uLDZyICr19qEuUgfqAYZTDn1QWmAAo+AlabfikKX87Ux2Y69KmzKUfxy+Ggp8lZADzKm7h+hl6/QPIJqcNJpgDTL8FgUPlKlzo5tkpwSPjkePSR1yblrzw+OT562ZbItwDkczLa+UfGbtSnoF7AGjs8/fy3r6PCgxeeiYVdxB7iP1LHi5oH2sjSJy/RK7ErZvmdLfFE+rThmzzZZqcz9G7ixYcc31D201XMtNdnXuLHwy58PteFuTHXrx0+5fzZRtY/4STPBh761fWpIzbowAOvujIePqHqbGsrluqNPftkkFwfOYRHaszDTgaefMebXPrUERnvW9g3b91wyNILP/6JRbl4ipN6uPDCVjzop1N7xA6e5oanPPx0iJEyYtt44wtztvBNndqRtg/qo0PiF/t41JPLX+3NcW1iRCe8EwPZ2ugQf9Scqc36iFeM6eajRLecr3zUVwzoVZ/2lOkgw1a6tNNvzMhr148PFm184FN++RTZD4b6fTW/jPzMM8/smJBhg04JhZEsO2JHpzKSs6n/LDAYKjNOSTszCpB+glJ1xid4vBZkwcoBwaEPHzucRH58zc9eaPMjjH7jJQwGgwxbANMRRrL6tdFFL2KviaKOHz6kLxx46GxX7sfq/OaMF1f0diPAl11B91fJPl+mV4z0I7rIhUkb7NrzQVs8fER0K+OhS15bvLDA3rGKMYknHLA0if09w8c+9rH9Hfk2sv4h6/8z4AMbdMMmtsow5KdcQvRPUk+WHiTXThc9xdaNzP8n7W8axDmd+PxxlM8Tu/HRkT58eNKZX8Upm/LZJkZh0T5jr9x44emiZEMdZv3aHf/Rlf7GRT/fPVnItcMKH/10SWT1uwa010cffn1shlE9Cg//ER7l+LOpz8/E+FsBmMMLIx3NFX9HYPNAhz7Ym9tiTy9MdGSLLnOsuLDJBzpgjS+/XOf60sMO2XC7NvPRomle+FsHX5mxi/DA5nen/OU9H2we5OJLNxxIGYaI7mzRp0wOX6Qc3rCJAz56tYkHPhjDLA7J1a/uBpF+68bhcNg4/dGsn3PBI8EZVjjZ08527TDGC3t8yZbjKbba8BorOJT57u9xbr/99h1bbY2jMr/89wrWW3H2l/XW33xxM6KL72yJSVhgDAcf2NIHz1kFxpQbqC4ETNoRYe0uMpOUMQQgHv3KgKgDkl51ZSSAAPo/GB544IHN7y/a/YctgoDIInJ0FkBt7HAomznLLnwFmhzChx+R1R9WARRYv93kh+saXDolJCa+DSfn5wzo80dT3YDowkOXiSbA9DS4DXQ4yeOHzwUiD5++7Cp3gbrAPIL6S9/8S0ZOh3HRx546HOoRnNrwa5fwITm84ly/XCIHRzHDG0a29PEn//Srw+rnGNyYySL6/TSH30wyh0xcf0vATrbxtTDTJQ760p8ufI05eTZhUYZLjuiQGg/yFnht5iQcbLDp97rMQe36ET3xqBsL3+rzy+fsSM6mDQaf6CPvGhHnYtt48Eeb8UDx78r6Rz/Cz298fDPHfN7pr9XtKNmAjzxe89I8wPOPf/xj9/vRvRZkvJI/kJSU6Rc7cWGXrtroZBtWSR1fc4x8OLW5JooVfv7T5Xowfy1g99133z5KEW99fILZnPD3OMrp3IX1D135KieX7bDA3TzUjw9WfOzrw6NNWb/UnJH7ks1Y8tM44w+LsnlhfI0z/MbAH4Bq97cbzz///PaBfokeMZ0pX4oNrIj98MIFN3l66lfnrz5/q+SGTM4JCuxXr17d/1Eee3TgT7+1iU/6/FSOseJDdvAlB4eU/8UznfTuMgaJkjqBxQAYZwGmgEJtTagMJMdRAwUkfuD0CRRg9CC2yLJBhn0TvPpmWv/gIYc/R+hDeNnQzg6c6qiyPvolfezDXu5/c7u6Au7/W/CzG/lcLNjKrh3TN77xjX3RmSzPPffc/o0nuuGk310e5Xc+0hNueTjpxgMvGWWDjGCEx0KH3MCOx+O+0cENY4NNJx3+y1i7e+10G6vws0m/un646RAPfTNpx6uNXnW64Kkef/rl+OBWNqH9Pxra/Jqo+MGpz0XnZxro8GTk/7gQU7zFCR8M5o62YksGfvGSwyQhfezTk3xtdOAvBvyW8ImFPguf/8rU/3ODPzzK8cDzl7/85dJLL7106XOf+9zeTeuDwULtbzn8VlHy7LNJlxyvGErqkjL+2vJHm7IcPr75XxO/9rWv7ScQiweiG7HlicNm7Ktf/ep+AiT/6quv7vjeeeedlw5rt2wT1G8qkQtfGM3BYqVfub7qbEliiOhQ54O5kp9uUvV96Utf2mNvo6bNQk0G+XT1jjvu2OuAHwu8uLjYdosjnvxUnoSHnuyU54N+mOITE+MOK8LnpgDX97///b0Qs9VcpltZXNyg+yM/8ecnXfTrp0ebMh3FDSZznE1tZOgNk7p2WLTTIZfCHl5YHMGKlyMpR5HWXH596EMfummbTUQvO5L563fq+OnvTfzBpBsPe9knl19ktKfLeNOhDZ79H0oVqAwBzSHEoZSoUwQooky/i9uujTNe0giUdnxkydBXALRr839Py9lz8crxAy8vwRUmefUw4NPOpoSyWTmH8cLSLu2wLqgWN/jCI5ciZbz0iFcXL12SfrZ78kg23OoSG3SENbwmiYFB9JNDcIqNnXz/L7b4IHoQXmPhR9f+9re/7cVVOyzn5+d7wXN8xTZeMcgG7OpIWT9MxUsOmz7yFm621OGFhR04u3E6QvN/GIiRnwnBQ48njXfeeWfv3v3C6JX1mz/e17D517/+9dLFWjTYwltiRzlsMISPXMQHSZs416dMthsOHvj102Xumq8e5/3HUm4enpyKSfrL6aLTrvm4buh2+tmjU7zFwu7QPIedPbbUw6W9MnmUb9X1x6Pf4iTmdo+efuwg3SxsLshI/vtYfxnuidr15Snf+Ojze3V4HRfL6S6uYso+vtrUsw9vc1UZnxgpN74wKvPfQvb5z39+P3WwT4+N2uX1B3r0I7GH3wJovljAP/3pT+95Y/PRjVjcUPEMR/bzg1544I7grD8f9WlHxh4Oc9HP58Nh8ScjnZKxNAauHwtwc55Ni7gbTE+l7MGYv8rhobd+c558/nQN4IGhdvyefG0qbQb8XtZhrUnF5xQvOW3sh4FOcXb86WnbWLmBROKXPj7yF2by7OvXLqd730B0ZKQBKMCcUeZgwgzgpwhIF9Ntt92274he3mYoHYGT0xEBgNeEtnNy0bFDP72SMj5yYSSfQ8p4yCHt6umg3yDXjoeuuVhrQ3hclB1DsF9s8Ls52imZcJLgkgmfnM/V2aEDaaNLXZkcUlc2aWGGVT+iX5szYz+h750CmnrV7fLsfmG3YHQMaEL7j30szsWPfsSui51t4yiXwteYq2cPHx/IwaWOz2LhZkG3Pv8zncltXmRXboz8b3qelCxybi52Q26MYv7ee+/djDe9Lm5JWaIDzbq2Un380Abr7IMN8QkWY/nhD39437Tc9OyQzUFER9S4qfPZTaOFsLH63/WflXW2TD8+1w4yjtqKmb5wwcEf2JrD7CFteJH5x6abAvJS1M3ijfWDj37MEF5YfF3DDp1IfmUtjnR5KjQX/P8W5gw+BEvyynPMi2H68Cvj1yfnp7KxIm9BchbvyVOM6SPjpuUn0cnAo+7YGm7z39OVkwBHQhbid999d88hNt2E8quY01us6NcvhuEvxuTJFGM8+e7kg92vfOUr+zrDSw6Pn5Wx0ydLP1/4Zs661uAztihMytrEJDzq+tMTD+ww4dPHppx9MmxWxusa8772k5/85KW77rprb3zxI+PuGkoPWRshsXbN8UObeaTNmmtDZ61ojpHVh9jTrk1Sh0VsUdhuvgPBUMLAKQoSJJBQhuSAf+9739u/0Gsh0YYPMWhRkbQJFgeU02ui3LhxY//2j34keGQNwgSvjwPaEB0chlVZbiJG9NEVX3ar6w9rdgT72Wef3RclrHDoc7d2QXhJzaY2F4CbShOWPvbFzWQgmw71/AsrHXThgxMPWW10Kvuvg+3S7ZD1u5CmPJ89XTg/tsDcf//9N/9/hiaLm11PeDDQT5eU33L2YAmnOKrXjqcxUfZVl4XXZHZzM87a7cDwqcOnjT8uQEcrcPmJd+3Tnhjgc+NBxiYd4qofFu34xEGZP8WRvvxSJqMPjnx3Y7XYudm6afQ1kjN67WySiVy4sCL2mj94tHdc51jIotP7FfbxSOTU6c+H9PNFH3z6ELzkwu9m68jCL2GLrfdIbh7eGeCBCRZjLc5irB7uyvT3NZ62sIgve3yf8YMlH/DrgxWG5om2fGDbYuxJQlzxwia5efn/LNgg64nKpgEG/8/7I488snfG7Og3j/hDt4RPXmzoxtscCIO6hF+/RAYpk8MLh534d7/73f3rx+nFY0Mmvl6MFxc3RRsj1xLinzhH5PXRTQc5JD76wqDOtjqc6s0p2CTtyNgqa8Pn+Nd64CmCfrbw+N0sT23mNr/F7uGHH951fvigotcEZNwAbXrM+WzKwwg/u3iR+uzTBvOZgRIECilAmFGgATfBBIdSAN3RLKYWBOfGfn46A1t4/WMx9n8iu2hMOHIuWjuiwAHhZ6vt3tyAChbbXRj0kdenDXWBGAi8yemDo0FRL9CCoZxtctNXvPTZJdkh2w3ZqR2Px0uf+MQnLh0OByw7Dp4I/L8NcDiz9cKSTSk7fFNmgy0JNm0S0geTvCRedhp28o5VlJEvJ/zf7jAZN354OhFTY2BCu6FFFmJjhjebE0sY6EHy+vMFXpNR3s3LOJL1DoDd+dRDjxvr+pXnPd6w+684HbnAYXGFkS462Gz8xNKCIeFtzMWnC5McKlZiLKlHynTTi9igjw8uMHW/eOpCtPtUj/y3sXb2SDvfHcWY66ixSuZiHbs5lrOgO0J0nbCDxAnBQw6eYq4tHXjqw0cen7kON//Ewk3EjtHC65rxu0niK27i4uZlvvgPkA5rrrIhwUGnG6GEH9GJ2JLYrg4bWXJy+ORki3c+0a9sMyGmTiPMRXzi6XRBv2vcdWUMwkA/f7w78nIdDnJ0ysMVnvDVHgbtyhI5eIujnLx4ipOY4kNw2qmbl2Qk4+8pWW5zRtbYOt4i7xr0xeg999xz87/1hidfipE2qZsz3bAhPDDo146UXXfsSXDi00bO2uSpx2bW6wJzwZOoG7O5IL5w9lHQ5bWh9L7LzdC7Du9O6ekaKgbZZ3O2FUftsBl/c8bY5Nfy+b/BlRNu4RBsbQILOKFuBBTZ+Vg4LHKOrRAgF+uCAlrZHc7EMAgGyiedzmERHYC5A7ow5ILFToCV6emi5BA5bXAWcO0cyxf9JYOgzB/+NSDK8cODB9EpsW3xMFDOkz02wqGPnMdY55HOaWFhn25YxIsP2ujV1sDATweqTR0/eTJ2tCaHC9KRhX74PXI+/fTTm9dCAAcM7H30ox/diwN7bNhl2uV7Qmzx5RNd8CDy6nCQmaQNHriU8SVLj0WeTZPVHLFZ8FgMi/zGeqpE/stTbX4eW84P2BxbWfycwZrcLgQYpGJGnm2JTSQ+CJ+45696vsCsXb15Qh6Pi8hN107e8RVMPh0VJ7x2yOfn5zftwX1cG4iePOmlnx+9c3KOf7HmvR0emwhmcVYPWwsC2fDRJeFlX9m4uMjDnJ90Iv3eg4ihG4mneGPgPdnVq1e3f40d23SKsw8s7ETNB4s4ghkWtoqV9mKpjPAgGPRJycnZtwG0yNosNF4WYWsAm/STt7ExjspiYQF3jZnzsOLlo+vAEw0+8ZDzh4xcHZ/YFWttSD/Sr5xvrmEEs3bjL8FEFp8bnneJbmyw4DusG7INnIXaU1O+mh/WN5tOm0hrWbrFiD/Zk9PFjrmgH25lxHb9yrCzDwed+NixLhl/8XL9u85tYLR7Qvn617++TwTgdG0Za08sYg6PnE62wgGLOmJLHZlH8BbzMIsj2TMONNg6U6SNogZau0FkRJsFjJMG2dOLGwAjvoN+/fXX94QlL7ieVizEjgvcRADSZ+BcePJ2qHTSg4cdVIA5L6jx0JHTDQp+8njix6eNHnx8kZNVRupk8HryQI5lHM+46dElVggPfj+77b/v9JirDTaJP+wJcvjY0UdHdtTpRfrFnF5t9NUXVvwmqBjrk/zsNzqsCe4mTc6Y2Pm9/fbbO/7k3fzTmR+w6ZOQdm30KsuNQf6o658+kKXXOD711FO73+Q0abVbnC1e9DQG5sIXvvCF3e+CtMDwif8mbPGhmy3tMIQ/DNne4Nc/2hvbGWv9dOjz3b4np8trd1a7L8EsGPyzMWIfVjpclNlhnw3EPzF24fIPr37UfKVHIk93/eHkmzZJWcpnbWRgVua/G4VrTYyurKd45/E///nP95dsNht28L5uM+/Iul5de8ju1BFHX+nBCxuiD4kRbDBkl22kXbk4qouRnL9yNuPLDwuzuQc/Yhflr2vI0dDdd9+93y2x62Zus8QWm9rwh7d4apfYRPEq4wk7HrrgNA+7OcMy+cjRxQ4eOdzkxcjNEK5r167d/LsxMjZPnkI9CeCjgxz5sBYX/NokFDZY8Ri72swxZXr4b80VYycQ6mTYswmwKXMc9+1vf3uvW+akrwV94aiMYOrmzXf26IOFDbaU5VLrJTva5eKoTFaMzgqSCgEKgUtAP9IHrHZAgPbSxtm73aidh7vlrbfeuhdd525vvvnm3pk8+OCD+9G2x236gLG4+JLBnb2FCjBJvxzBABfiWMETBGWEl3PqteWwPvL649FX2grWP3jsUL/1rW/tT3bd4S00fKdTPkkcHBvwn6x6k1Kc2FLPDuzqiE941MnCqJ8ON12U//QgO7XHHnts/+S3l48mg086HRvAilysjg0tzLAZs3nBFBu2K2cnbNotDnArsw8bnGE2mZF+vrhpeGGP7Nbt0NjVbqdurmjzdz92m3TDScYiwm+6EZ1hYhcWuV2u9rA1LvGTxyfRjxdmc0tunGx0bBDI8guPXRwcjmHNYf6IneNZGwi7zPSV0+cCt6A0xo134wUHmr4owymlCw8ftOnXng+1s9VNzi7TeLdgOErx8t/11QJEp6cNNzkLiEXOOJhbfHT9iqMx4i97xY9NGKJwhYnvxZzvMKenOnlyNo3i57NnT5nJiZFNAzIW5gxcnqhsKOyqjXfXCb4wVi5O6jAUO+3k1GGAtzEyz5B2beLFvjo5/RZo42otQG7CsHri879r2lh6MqIfiSN+c8a7ytqKgWuwG5frE+lD7OYjPM0d/fxVVyYvV2fXtS8XW3PU0yd51xifENyeWHxlmv/k87vrqjjlD1m42Dde+NWThSVst6y7/3VGgCvQDTLHEQdzQlnigElj9+Y4xzGWi9PCYkDcFJwTejFmx+fGAgiQgLg7umvbFWnT5yLhFLDxsg8sncgAT6zachAu2AWwSa6OyNODBACvi8i7DQszgsMkNii+tPHYxx942LTb99foJrlzRcdXjmsKvBwf3bVtxesf7ezDCot+dTjTHwY2YUTKFjx1SVkyUezcvIMwoY2VR1Yvcj0VwddYsJHv4lEs4WQbFVN1PHAaY3nzIl79xsnxpUdpuOi8vG62V9bOWNz6LwLgctGI8ze/+c19dGQX7eZhd+S403hZLGCgtzGbmIqdvPgWY22VPyhvkSBHJ6xiaL6ypV+5/7zHOxofTPis0x++dbHSLTmatTB7cempQBvdcMzxZEvMJDwz1xdufepiLacL0ZWca8bOXKw9wZmn+j2ROhP3NEK/Nnpd066tvniD0/VqPKRs8p9uOdvir06PhKZe9WTlMOt3XeKHDybXFpLD5gYhbm5iFlvvnsTYHHI8bM6IP8zeMdhwiDtqPrDTfCx+4YYFyfkvFU/+1IYfRgkvnfpcK27K6pK6d3kX68naTdc7M5s1x7ZOS8jCALPNhyc8cmKbLfalMIWD7fCHAR9sUvz01S629Rt7PGIr1pI1SwzNZZitVd6XGHP8UrrIugYab3GjX1tjHgY8ytmHSdl4az9T0YgRNXkZk0yoAjIXb0HgENKfPKWcMBjXr1/fXy1Y7NjAZzAsLq+99tqlv//973uSmCgWGQFmMyJT0OBSR3jgpk+O5gVccPDTCZOkDmd5MuRnoNLJd7tjC55dyW9/+9uNvYltd22AkqejOGijpzrMyjDzSVm/cjGHC07HJnS7wYo5GTdodk0aT3u9S7LwwumCs6j5D4k8zjaRs0+HcvFlmy3txQfmYlIMtXWxwQqPXD+b/utV4w2bnZB+N1UbCAuehdkN2ibCRwcWMjtM42/D4EINWzha4NjRJy74jCV8sEvKePRH6ii/9ZHTrs05tbry8XjcO/duwHbq+i12nursJmGiIxsWvz6FteOkF0/xjA9ufdrDo948aMzD2bgUfzrxI2V8dDqe9OTuKNhcIMemfvGULNJuznbx5q3FA8FiLMmIgblLJwqHMj7t2vLLHMtX9ujQhwc+c8BRrpMHvP7mwxqgzwL32c9+dmMxP2D0YQIePpjD2m04XWNwaTPHlCW2JHSKtZizhWYOMzzw0le8yPDRpsv/d+4aJ+dm7N1u73Fg9WTnyNX7PLF0I3R92uGfr/dl/rDQU6x5Cxvd4qHMfmMdLnaLJZ/gwysZH/1h5k992ul1zOoPmT11Hg6HLYOHLjaMv5MhNxL8/JbmmKvDKTXnYBIfNLHioV8bDPQg7besP5y67lxUBwABUZcoRRyzSFDCIGFOmtAeqT1hGPR47eDt3gwIostEsIN3bGX3xhaZgkx3jm2h9U/gCzgZiT6OGDQytc9Jlg/a8o0evBYLO6PjWkQcXaCCtivrH9/X2yl7WnLTs6s3URwHWBzZFxfxUGYDXjbYVm5AlFGY1MPKrpQP+ug0QS3Sjl5cjGJZjIyF2OH1eO19x/mazCY3vezCAdfMtdNNTzGBie3w6aNXO1n88dCn7EJycblRWHTJmAMWBFjtMNudO5v3tMQffxXt7xDc5MQuDNnRRocxZdfGgm7txUe9Msx80kYHOu3HCzPsEhzGUbuYuhAdScAvfuaGJzsLH10l7W5+dpwWcfM5W3M84UiGf5I2GFDxDzdZWMJfO14y6ZaLhwXE2PfkjI89doyFOeu4C15xRI1hNvA3N/TTjejIPp7K5PMLb6lxwKdfXBt3awCc9LhOxdpHF/5Y003bjtk8JutG5x2DTZA2443oZIMOJM/2aVldvPiAhxy79GsXC33K2oqNozXJTcFvjNnAkTMfzANfEbqJ2Azxz/Gwd1+eOqyd1gO26Y7oR/BnD486gi/fiqG2fCIff2MiLsrWXacg4ugPcvGxkawbiPfQbo7msD4y2ZZLbEj6UXrSBVcyynjFLFxk9n9pqzMjDZwJgziFCJlwFEj6gUKc0o+00+EYoDbtJrNHUxef5IJ1d2SX3gm6MnkDiehlR5/dClLXnqOwss1pOrMfj4XYgqxuR2Fy2NnTicg1iG4WFgvvGiwWHhFNLEFsghcr9tlis0Gms3blGcdszPZst3OxOOGzWIgXny18dkcW2OIBt3K7IDt1cYGF/W4yxdiY0auPfWMpL0ZhlodTHg97xkTbxdrhiE0vPckc1w3ZwuAG4onEwuCYC6/jNcmmw8XIphjyuSdQcYSRPTFRl4sH/eHVbyz0KcNDFx6+IrLqCI8y7Oz6nNR8wOtvFsxXu3qfwHrKEDcklsh88OThiIUvdMOSDforsyN1wZHXz25xJgszwhu+3bD+IXuq0y4ZuXHASkeUPdgdX+GdN2J26TOP4BRzqXkRBvFkm9+1FVtykjo+snileLV79yK3wXBE1Y3Ohs2CFtawe8p3w/M0a/NBP93pxEdfOLru+YO0K/MxghMupJ9M8dSHNz3mjaMomyLvONhyHTs5Qflt3H2l5+bhxmEuGQOJLnZgJi9pKzb0NPZw6KM3v/Tnrz6ETz8+fV0XrbvGUju7eOTIxv2JJ57Yc9XNhH+un6594w6XGGcftvTQoSzBEIW5fjbPmlAUasAU6ATUGQfWhWWR0uYCRMr6InoiIE1qO6MXXnhhP6bamQg+voICKB3a6AO+iyxcdCoLCAofWYkP5Jp8eKvDeliPe4ILv8fnK+u8vu/Vk8XnCMZCAS9bdk70WOTYNAByiV112F20bIeXTv35gx+fOsKHJn712c+uHZ3jPjc7+BHdiA4T39EgfBZmbezMpI2uqVsZjz5+woGSS4Zck58P6Zez50KzG7NJEAM7NjHzyM8Gebu08/WE5IbCFj366EBs42O7eaXcIqkfBoQP6Ue1N6e00dv4ZKd+/DA4gvKhhzGD+fJ6h+MrJouzfuTC84TnBu14xk2cHrpRNpTZkWDNr+KmP//gnv4oIzJS/drod424Tu2CvSx1/EOvp3n83ue4ppTpMjfgFzttKNvqYdSu3lzCQ2944AhL/qaHDmXyteUrXjtkNxLxs1EzR2CKn206POXbIHlq8s4BT7rZpjOM2TFe2vHpkxD+ynyXxA0eY+b64JubAx366NE/b6Z0kKWPTbx283A6YrtYNxLzXj/ZcNGtrQQf/fqRfvz0K9eev/rIIrn2SNk8QI7XxNOmx9iRi/C59ryXdqRoE+xJij/JZ1cdfz7Qoa8YyrsOJh7t7PLhrIsh5yhULoCYKd3M7y9aAcGL8AZKXhDIWlDs3G7cuLHP5nwRABR9UnIFQcCbaIHGMycM3nAmn025RDf76YDRVxRe2HmScBMxEHTFy4agO2bxcQCZ4pNe/pIJj3aTD4780sZ29hvksJiQeKcedtjT56LTJzlS8bh/bX06aJEzcfEYA3YR227sYqI/bC4Yj+Vs4WFfLvFZnS742FaXospsxaev8ddPF3kXpQXZUZp27zzEFxY8bibf+c53Lv34xz/eTy38LbZuGHTCSZcLmw7+kw2HfkQn/PHzO5/J01vs8Te++lDzVt3mxuO+OHlishjTm003DOfJjlbwiDnddBRH+pWRvnI68MnZkmojc0qnsnglvMbVH6+6gWhzrOqmxq5dfl8M0alN7NjOfnazqZ1ecTf2SFs+KRd/fFIycPKlOJGJX3z0iZVx9EGH+ezJzrsleuiFx6LsXY1dvXk+5yAb2ZFPXHSQ14Yqs4vk8bNlPvCTXDJ0mtMIj3q5MqIXmQPWAzcPmG3Y9NHXE343J/zZ5z8b/AobGXV5MSOjnj044FSvDSY6+OGJTjz7QVKbMrroNSaSp09rHH5rsDZ66cBrTFwz2sKjLrHNXvbDkA0yyjDfsl5sXne+P50gMB1QTik+oOwu3Ol8Vup826MpHtQAyAX/fO08vfdw1myhQYFTxpf+QKoLfuA5Fq6ck3MmHYJTH15lTxMenR966KF9VGEXZ2GDQ3/JIuax1CSx02TbjQYpCxZe+JCyOOhD9cNZcMMOI/4miXoTCq8+9eJHzgA7vvIlkJ+GcOPjBx3+OMjRkQvSwkLekREd3knAgI9ubRKc+MICczHPdrjxIe34+eli0E4GvnyTsyd+Ngfeh11ZT3Ymr6/UHA040jL28PrjPdSjNZ18LQ7q7MHPNqJfuwsX6VcvwZQPtYULf35o04+fPZgtLo6wfNkEv3q+4XXm7QbjCUofWQSbhDe9lfV3wSrXTp8yCm/Y6+O3NgkvX20gfMEmrtrMUccojqrE0ULi3QJyLXuXIKZ0GrfmQTbk7NAtIXrxFvdw5icZCcnDqF/d/MoOHnr5aPy1u8m57hBZX9850nYTtPEQW4QXWdSyAYv25oF2pB1eSXni0hY/3cp4Gl9liV5rhJ8M8pWV96HGjq4SW9r4ZMMhtmHQR3f66UTs0x3f5NkM65/Jq1wc86M5Ep+641+bBZ/Cm6+Oj23QrQme5qzDbh7sek3gScVNj5yxgiv9ckS/8WsMtemDP1z0aVOX8veMwijFDBGeRAFBwdPvs0xfVviy5vJ69A9MebLkmhwWAEFgHOFln15Uro1cNjmGmqzKJhiMBYRepK6PDY/ELixPHn7i4XA4bJ75T7bsILzY95s87vAmDPts05VtNiX2JDzqeCqHO2z85Js6PemiG29xSIeJyge/fdQvhLqZ2Wk4u/cFk52edwtInz+QM1EsLmQlPsAIH2I3e2zApF8Oi6TceClP2fyge+oi58IyD9wg3My8JLdLdrRFDj6T22JokbMo08N+MYGxmLGtHNEhwd8FHk996vGQTRc72vmiDZ/x9rRh/vqrdC8j6UZ4I7Lia+6Sbezk6vQoR2TV06WdDoQf6dcGDyyVw147XtebG7DrzQJs4/DTn/50y1gg7IC9Q3DEKfZuJm7i52vTZnGhvzFnVx0ZM3aaw2yLK8oHeKWuBe2SNnryR06PeWNeSXTL3ZgdYcGZbrnNmp8AcjxrwYOtOamMsrMr6x9+6IMhigee5KZf/HRjRGGnB7/rDWYfeVjLHPv0VB+/3GbTE7RNhGNDL9z5y7b3PIhOSRubCB5jy05jigfBUrz0NUfYJ6cNacdrHniS8JLfRqJr1NG1X0JwPfm7INeejSY9yjZAPul1k+ZH1xucbMATPvbYDqM622TES47IVv5vy2o0eBoJ5ywhCdUnOO50/m8ME3WevToDl/Ca7HbHU75gaGOjAa8ObOUCCGz25fphlBqsBkNOTs6WF3J+9tjRj4sQCZanInpM3AJBVxPN4JDHq78BpVdZrPDiKV7h7SJki06pvtrIoHxgpzZlNwoXnV2R785NBhPBjtPx2rvr+247EHhQ/nahw6+t+ODRNxcNbcVSO7vqjQk86Vfma31kUfzs+cjAxO2pzZGEXY9PSfXzyXwRczuny+tm4+KzAPOPLXZgUWZLnQ225Xjl2TU2Yht27Y1XOhoDdbojT0Peg4mxDQM58uYGPvMXBu10GG94xFEbm2zrU0f5INdfHzm82uPRp01f7cWYfb7iESuxFSOLhU0CbI6I0umJw3XniNBNRJxdoxfrrL6F25xlh49h+KB6WOKDoTHRBi9/2ZbERZzYcZOFkS1j7vrzsYrNmyfQ7IlVvmeH3q6n+MRwEjv6aleHTyIvKaPGBK9Y6mNLPzll65Pf7fLkYVcvxm66nkz5YsODl6y88YFBG90SW+lnO/xsTP+UUe1igMwpSZ0N8sriGC/7h7UBvnr16j6NEE83FTfg4mHT1ryBhx56Eb/phhW/MS0W2rM/28njCSedEh3FcP8ar0ZMMVBCIfBNPIrqx2tiW5Rz1nGLO51Fzs7IxdkOuUDTxTCSkwWGHTaR8gSsjo/tnCvIOaGvm5W7tMXLTzv45Ux3axMBBrrslOF0xOLpBFbEhoTCBK+yBJMY4GGfroh9fWGdfTA2wMr/z9i9tFpaXH8c70C/kN1TZyIEMSiHHgWCEB1JItiiQZCIF+LASxwkJI0GYrxF0QQbNAgJQRQC8QYNQgYSh2YQEM8oeRn+61Oe72Fl4+C/oE5VrVqX31pVTz2X/Zy9yc08pFd8NjCfHyj+18OBaXPwDN4bYf4xzAbCpgWE+FPq28TZtWnjm0uYy4GxSLv8wkBOrGT5htmiZNu4YqwNtb6v4HZSs4G5UvNowmNOJ4z5iNTJ3P+x2Oy8y+5RhvmBARa2kX5FXxx8ouZHm39kbcSHdcbUeiKn7bbeCdrVsUcWzSf7/hfByVo8Nu7yakwe4YPTJsMHnzDwr98c89N4cRnDM8YuopusMXb4UDvG3IU6lsh7VGH+katJPHI2Dh/0W/s2bselNcA2mTDr86EuV9Yi/whOfBQvvMboIbww6rdm5Ig/Nlz4uLPrzpN8cdKBswuHYpYfxH446Sjsyj++NjyKfnnlA2UHXj4mLjbYI2O9usP30gQ7nlj4TMZJWg7tH9Ym4lPJl/akyQ+fmi8ESzrx1XLV+oNBLso1m/qKfPrHS/l050zXYytrwknEB+pyKtbywq82npeFHM8u2hSycsE2aj1M3/DoiyH8ckwWf+sBiUoAAeAoqxVjapuvzz08k/dGkM2uf7TzeqwPG30Rma8uAfCYLFb8fBnPD1DKTLgxBa/FAUcBpc+eACWDnA3AASTpFoArYZuZjdAigdNVkVdiJ7GD2Clx/CN+8WDkCxnDU1sE2p10jNNRyKcTv4mjy68Cn8VgQbv9dJJD+K4yfSDZVyXwpaBs8GHzlANtmOA1PueVDlxILrXLYZjVxuiq87WVzv4Y4+90Xen6LiP5duUGq6/4dxA6sdN3YWFexG3jdmJn01si1gx+fuBGMPNBTo2vhpU839bTjFXbHNAVO9lil09vpPgyRY+EnDyybxPu8aCTg1icQJBchAFGByT/4TGm8M2eUhufvr42/WKA3Ri8xhW2yXik1mcz7pDccfj/g2vrZRQnaZuB+MnS67gKL16xh604whMONsjoG1PkjA051M9usmrEByIvb/CfnJzsf8SUYxcQ1rTPauTz0rrzZLe8yKG2fHaC0Gc3fPr8ken4i5dMGNThVjt2EHviYKdjVI4VPC/PWLf+EdfnNh5ZOXF3AmGDvbDTgQGxXZ7iwTXzSU/feHb0UbHhi5GMEx+7yTuW7FdOHoiu9eqRtWNObHityS10JmcMXy32bIaDT4U+nlgUPASfsXzQN7bldEx+AvqII7wSQ9kjCm+B+C9dJxDPY23Gfj9D3/gzzzzzP2+F0HfgsmWBSAoCik19bQEatwAnHjLJN1H6+Aq75B1gHYg2CVeZbCLPLL0d5JVNGJ3sfKg3KT94MOiX4MbUszThdDqAm6AmUn8n+mxC6JfTOTHyLk82V4tFLvDcRThJi8+ickCYPL75RDYSt94eH9hY6IQdDnmjxx8+DNr42vyoIzlV2CWvbZEicvTg04bBgWZhk20u+TWmsEPfPPzud7/bV8reIqHXyd3mQ19u1CgcaoTPd3kjG98YOZjSJxsVn7Uhx/CaZ/ly8nBX6vMvz+ZdXGSjGMUbnvzqm9988jcx4BtH+NGUxzOfsLirkC+5cPfuUZT1aj69Fovv0QX/dNhWPCry2ZLN2l3I6Tqhu7grJ+TNsbzxnU+YYMZrTZVbMvj6dNlSR/oK22yYb1e5jjt3Hk7S1rM1aWP2SrzHhb16Kl7Y6Yo5CmM5gjFf4cHjl6xjH58dcsfy/ETwk/fUQX7d4SG6XtF18mBLzl2AshXlMwz4/FkXYscPS+10yJZDvpR4YpB7OvlTy40YFcePfuQOwknZkwnHlVquPY6zFuAhb127m4KLDXbhaG9pPstdecZvHvgMr3FtY2jHRzhD6s5mAGTQ5qX4pkeL+tK6ikBAWajIM2W3WA5OzxMDa9EjH/Dkq4SxL1iyguRTWzLJltQCwDOejvHGBBXmPvdwsCFjFpKEkhFLsW2B9UffBo3Y1eePLp8KX2p8ZIwsYpe+CQ9ni9w42Q7SFly6xi0KH4B2m2ozYbOTtKsiCx92OYWFfeTxoQ/WXeWbP7nMl3EHKHn8YsIXCxtkm5M2SniLT1xk42nLj5OSCwcXFDYMfGsCPo/Zio9vhQ9jbW42HHPEPnt8k+MHabMJM11Ym5cwqFFxqactMRejtk3WVaUTByzWRP+H4HMDB2DzyC5b+uYsYk8RH6zmKTz8K8VORx+ZB3JwpCtGttnhS8GzgTnJ8QPXH//4x/1BKJ7xLgjkz+M4F3aHw2E/LvQ/Fe642couv+UzPkzh5kc7XGS0YdVGrdt47CniwnN37DujfNArZ058/kHQxZuTSK8gs8Umfyhs2uUqX50A4GscLvqtl/IJS3bNCTk4mg82rU93nleuXNlr1xicjiF22FSsk/Cxw56ckw/LxG19Js8OKm+7s/4YTyZeMsWtX97JwiIH2vjodF0g+D81bzmKDy77hv9Md5dH1knFW26+Dskaty/AzUa+1LDmJ998iNe4MXradJV48n3+XViYTQADgOeIMSBscBYJuaigbBhOEj5UR/iuOK+vt0E8ohAoOxa0km1y7BlzQIdjAs2Xmt7UhdPCkGSbq4UAh1tTYxG7yIbn+bYrNptIVFLqw4tnYfCnPYsYUAs9XGLRFg/5qEWuH/5qOu4gvKoLuzsQ8dPx2NDVG3LXZKN2sraR2IDJOfm4eibvCtBCUPhXT9/lm705j2ThacPRrohFLjtwOqgdjDD4zRJXlq5+LWrPZW1sFrY5DQ9/FrLNsUXZiUEejcPHnz5M8OirOwDIIHHRIYfHD9KP8GwQTgzueDyawBObteJD/pdeemm/LeYzJ/+o2Q8isZEseW0Yylv4+IZPLDDhk420lXDTFx9b9IqNrrb8ilmh5yreGrDxWbOwyL1jzlW99WzTRl5JtWk7Fqxh9uApn+GDwRjM0cxbWPHIwqHNN4qnzYY7SXjsEebXWz/u5v7whz/sMU8mHHP0kFqBRyxqPpBa4SscZPVhwFOS06YfNjYax9emb+1ZC92BOPFZA/7Hw2dI5kPOXQR7O08c9FzwiMWbbtrmgE1x54fPsNFprcpx/ospXjinbjGYL3Js+azGhTv8yJ7gBOLjhPvW/4ex645K7vlFciomJ5DD4bDvUNiTDzjpwDWx8IWHGiMbH681aT8wdn4C2VrrDwEESIqEtW0GFrerY+T5tg1P33N7AXrObTGx46rZh6UmzYmFDcDZtrkIhm0TWcLxULLqJkpSBVGSC1YgxjyecDVsISfDFnz+/8M/iqkdcGRLtpg8JnJmF4MDTynmkibZqIXAvzbM/NPRJtfEtPGSnTFO7MnDXPx4bNv4/ECMvgPQ/9y4xS6PTtJuU50wteFgp5yKIft0+J0LaM43PbJqspViFl8xJ2tuFGSeXW16FOAulO3mxknOvPgQ0FqhX8602a3gw60PjzZe+WOTjgKjPpnik8PmDP+wDqDL6zMXr2par8ZstK7QfF6HPM7gw0nEZw/WtrXs/2rExbdxhR++y1X41WFRh788Fh9M7FkbDnR29OHGc5KzTumZW1gvrbt+ci7kzK2TiY3Oh8AuGsyBlxfayMWEl299/uEKI3tk+IYJkTcOC/mw4pGlg5ccGfpegTWvZOTBZ41ya85dHTsurQnHmjuq/64XXciWU77rs80Gko+Ju1yHkQx5GMyr8dY+PTwEMx3z72SbHVfqviXDXuWkK6de+z85OTnf51wYeVXWo3r7GEzzOGWffxQWvsXKD7/65a/cFYMxeVDjZQN2uGHy8okx5AISzx4GZ3tWGMThMT3c85EWPHx43MUuPPmDk1/j4SEjTjLa5QxPH+3/A8FgXBANAC9RgbZwkH48B50F7GTQ4wjfZ/Paa6/tBcKGK9E+iJoJnElkt+DZplcy9QGHkQx+gdKD25hFYwG7yvShY5sdfbd1bvNd4cOpL9b0LZ5r68NJH1I6UI11cPGH6ucLFrhKKjljTYrcRTOW4kmumNwy47nCUePD4QNTnxcgubR4jDuZyzvMrvK02aADB/8Kvnzyq8AS7rDz1Zhxa0BfLddqNm12TsYWoMXJtvxrs8WfBWsuzLk+2/CS9WKAt4ocxNmz2OVW4ZM8PX222WUPDm18GFs/8qJfXPrsIDmC17x7Lu/u0xi83hL7+c9/vu84xEdW7uRQfE4acuuE6O7QeoINpnDmszzyyZYycRqXA/ziMCfIGB5bbLubs4G5C5IjV5p9dkcGzzpwMrTW3fmx4QLIY0z/p8A3vGpYs63mSxzGUHlUw4f4QHTTJy8GNvDCjX9YJ2h4XNw48SI+bNROLK6gbbx8uJvyZZrukqx1OBVzbA74YDs/4QuvMYV9RF575jas5pNdeXUB49jxduPJ2nTx+XC378LSJuxzJv/tb33qW5t8OcnIiTuSfMPFNypHYea/YrwYkmWDDv4xwSwW/mbM+vKEXDy483SsWQtIjNa1XHgK4MP1f6zX/Y2zyZ+Cwg4HX/jaFXGwU46Nl1M8fXiUi4QZzJB+RYIyTqaAjVN2oPVWEGAACMJCJiPhbrMcxOQdNIAIiK4+HWQcX9GO0tHvoAsrbHhwsWNBuvKBwUaADwf/EyceeQvW4zWbhAOVvM0Lid2kFDN5/sKnLZnk4DFu08Fng5y2cfHwyRYdNR25MKbtrsLVpisd+u4o2OhxFUxsiM0m4fVX4zY5r/PZeMQpHwo8Sv755afchl+Nz7air8BIP7423fT05diVjq+iMJ+uiJyonfQsdrL04VGLxdW1RwCulL0ye3094nRiEssxhSMsfMKqRsbFx7b45BOPrSl3WBuciwp5FYNxWF3oyBn95tFFiDsPuPzPjUdcNmnx0eWjeQsfX+UZNvYU/DDiayPxsOGkivDJW28er9ggbA7Wp3l96KGHdnzJ0feozaYMk8/JvM3mv9M9kpFn9pC6fMFDF5kT9uTCODzpxNdX+FDj01creGIw17C4MEvnsHLuzTzHnTskNvgx1+6oPF7hu3xml762Qj7s8ouM0+G/eNgOk5oOHhvWhVj15dbJwYUvO+TMq8c/bLqDd4csHic6L1b4sNrdErzyiooln/HyzW4xbIWzP+LFN8+tjRlnxyhx8eqrzS2c3mqzhhV3qDCLjZz9wGM2PG9n+VyPD+s74l/c8KvFCSs85TTcxsvRzGU8tvndP2nbRFDmxCQjCaGA5yByhnYgdoVPxjiykGzeEm2STICzvATY9NgGMpq+yJR0Po3BRL5NHA8BjZ8+OQHieUf+T+trSK5evXr+tSUwoHBq8+Vk02ubTng2PbbzH1b2S1oTkP9shZeuwg4eOb7wkoWVHQVfnwwdC/VXv/rV5sHn9t8CMZ6MKykfktrc0GEdqK6s4TRvcsm3Yk74CL+2wle4GiMvV/gw4c/Y+cpHuhanRyp+ytam6xf8PIc9fj2aPPs2RGvInNro/Ee1E7hHGzDxCZ8Y5C4e3/IFV/jDnQ77xuLrO9DU8sC3q0rkxHf8Ft4eOPvjm3d9m6n5uLTuSNmFGWnXzz4+PwoSr1yFB3Zj+OKCEanJGcNnz/Ntm4HNykbnjbBJ4vBYzQlYztnw/0HWhJOO2BxvbRzlng++JobJy0d5h1lb3OGkPwnfHMHiMZATiHyR0/bcHrmw8pmIebVH6Nu86JrTLrxgswbxtRUEZ7zNWH9gk9/WhX7xwBVpk5NbefE/SXLjhFc8ZPp9HXruOMXjxYX+78ZJ0NU8TI5BsbCL8LTzrz8x4DcebrxkxIbI4CVjvcqPuZVjbXsCmeZSW2xwOp6sdbl38c5u9mDiU81++vHzL590KnwiPujYY9ghR/c7zz777NdexeXYgAnsBJITfE5dOfhJUv8H4oq5qxgTcro+P3DbJFjAOZQQMsBoA6NGnKOSBVB+yPNN3rjFhscWPjmk1m/TJINsUr6ewK1pV2nsGKfDjrO1DyYtFMnOJnsmjjy5JlFOwkMGXkROzoq1CWnBh41sYzNWfLjCp+bHVbqrDsSGAxkWug44G0kY4CkP6iabTvnHRzBoq/lCLWDtcsj2nCuy5OSpOMjDqjgheC7rJOIqrjWUDrvetXdytwF6xOXk07pjU2mdkBdHJMfywBeZiU1bTHTI6MMPA/8weXTmIujbKD15Rd5geuqpp/ZnCtZetuWSbTblQy706SvhJ28sPpnk8k8WkUXFz64Th0ctHruxU+7ZcIHmCynlUoywiJO+41BbgRvJBxzk6JMrh+HHC6M2Kgay/BtHdBR8hV3rkj+fLfoJW1f6/MJH1gnxxRdf3Hk1j9al8Xnc8BFOOvnULo9qVD+MZFF6+trGFb7ieYQtt3BaX8bIILHo+38re6KLYDHyJ7fGyweeXJFXh4VfpVyHjXyY6GRHDuSJHB5ii83sGGPP+iOrDXs6ZPXpG6OvpGfcGB5KLjxqPPJiRPCyZa7wOhbDZZzOfoQloBxjGqSkppCwKxy3U54NanMIlKseG4hnboKUlIIrCDZQSeEH8TtLcsYDW00OVkmchF9iydoo/Oc2vsVsk2qi8+8DR3cgsNusjcNuHAbxw1785QMPhYNfOdDnz0KzufM5MdExrrCRfBOTr/zLpytMNuCyKfDDPrwOWLLhky/jcJbbbPGZfzUdZJy8cQuFnj6f2jC1wMnr0+GnWNiygbig8PmXX3fzaIq+Qg5WMqfrIsPjNzbEHV65QM29ujYs2nKgbh7IhyFZPPJkEPvIowjP5m1wHqmGyxhdBTZr20nExtH8iZkfcSq16RrjQzxiUPOP+CCPRy7Z8q2eY9p4dGDxuylqfetJDvtg3+va/LEpj0oxy7Ox1gss4iPLh7YYits4MsYXanz246Wn39rlz57gpQS5gxPP2iHjEZvvmcMzhse2HGuzJW/GrcOIPlzlMgx0yx8dsdFPlpw2Mobgpue48vlXT0WsK3zFhSQ8Pp/xaNbJht05x62d1hZfs/AXHrrGxIRHB24FbhibN/jCDC88sMmBubcGFPJsmV+2jSO4zX15xGMj7PSKk58wk4MlWTjYT6b44+sbo4O+s257vvbWDAPAGCgYziUMGQfSuDsRFFhgPIawOASrL8AmJ3vk2dMng9TkgOKvgOmTSyd5ONLHC3dy9CWLn55fbkfjj03Ya7Cu8OkrSM0OG9oS2QRrI1gRvErxpCt+2PGNo7Cra5NvXDz62YCBP/3kyRQ3PTj0yarzBzvCYyMdc8uWg5ZdY+aTPn6ydOuri58OCrPaeH7Z0XZX4dEPHGxOskYclPwah8XBlc3szxzj8V2s9SceOIozvPlW29Q8i/dBrw/yzU/54B92m9wLL7ywsTtBe4SEj8LHT3jx8ikW7WRhk4vGq8NGjg57ZGHki1z5lAPrt3HHnePLI4x5Ndwc0INbX17ZqvBXO5/888k+Hmza4lMXHz3jajxtBdEpz3gwmHuPeOSXLWTO3VXlQyzk1eWYHEx8s2OMP+PpGeNPwWNj+idfXuFNtn2LXbiMOZHYkGHz2Ud0aT2CsxGzI89whCV7xQ+PUo7ik9OGEdFHeDAiPD7sF8WVPnuwonKiTVcxrhSvdj74boy9bLKDbzw7/OYHjyyeGnby+dUmK5fGtXec6/8HvvZhoo7JyniG8RElyUfANvEtkpIlIdoOAHIclyw65NnOF3t45OjRR22Q/CoIFjKCLNACZxOPXUEiG4U2Ht2CJoff1ZqrJ3bSNU4XkTHmoJzUxIaTTASjeBREln99bfHwoeDr01e0ycGqZqvJP9ZrjA22ULmtj5ccfXGFCz+f5Gc732yLcY7BI3902Jv2J0Y2wkafnhyTp6vkR5ssYkM7XTL06KfTXLKVPbr8kCGrbe2pw2vjsN5sxuySNcf6ZBUEg0K3ONhsXvJDhn9kjAyedjjikWFLQeS01fCh5ke8qM3MHBiTExc+9LoYoMuGuLJVTujBQc8YokuHD/EagxGfLKoWZ336zR89MsazW67YaWPs2C532bKXyFH62uWADkyNTbtsFws+YjuZ9qjyCFu+6cqLGmnLpRNJ81XcdMSq3xrPB93k2SjXeOWCD/L0tdkyBjtecuopp403x5tDY5F2sfDLbj74o1MOySafvvjgCDMdbXLmwZiib8x8q+U3Xbb2I6z/YayEUKJsIhFgeByQtThyWKAlhl4JzWkgCopMwdEnT5acfnpkFOP5h6cAyU4+nHDRMabkKxsthjCy15i42LCgIrbYgCl/4pCHCgzaTbQ2vUi7ySGL8GDVV7T5aUybnWLId+NqOpGckBEL23TZyG/2ipXuLHSUFmH6xc02W6j4wtvcpUuOLTo2QLUDEqctAABAAElEQVQ+MsYvvD06lG88+ZMLdsizCy8bKD+7s/6QhYEuudrlM3thcKHAb1fwdBC7xuAs52yyRyZsZGHTnzFpI7WxdPXply8ysEXkwiovbDtAW0d0jSt4yYQBbja68yhn+BOTNhvx6LCliLG85c84H+QbEwM+UocLP9vssUHH3sGGXDvmFPNQfLCSU4dPmw34zRPbKNmJHz89NVk1m2Eun+6I2OUbdvaM0YGTPJz6Xl4pHifo2vTYV9jCj/AQHhvs18cLvzZfqDyJFd/8wtc43syzvoJqJ1tM+OJC8gdnuMl2PGjDzB8c5YBe8cHMHlk2UH60jYv1f17jxUABOzYQn1PGAkKOwZzqs5VDQQVMbdyYoi1YZIwNtWJ8gzzzx28Ys42H9LXZMymwkZ24tuD6k5zENYFholdccMPGBvtIXQzsmySER15fYUOfvL42yk9942TVCgpfOTCennF+jSnGIm1jYbYx04uvjdKrr1bEmWwyanjYDTOZFh1+hb5Y2UoHlrkh0iVDJ18w0WW/cTbI0GcLDrw5Tq/8GQ8ruXKJp80+O96kMZ49deuFbQV1sDVePPpswQavPqLHbkRenxx5+cKjEzb1zBdbneySN66N8mkzlpfmmR+lu3f+WtfhoauUC3a1ZwzGy295oE8OdsQuPPknV/7I4bMpf3xkx2al8IHwxQs3vbCVA/x0i49cuSfPDyqW8ONl91i+u7Zk9fmEmw6+fvPHJ4KFLRTeMIqfTrk0HnY8tuQDX5ud4jdndNkyVi7xyJNVG1eQGo9NOjCHrVyIx5h+Ntis0DeWn+zPPuxwKuEME3n8/XXuQOUEExEMgKCM63MQWH0UKEYdAPgt+iYGWAXhZSc/MynaTbCaPbjoNEkwhFlwTTwZpOaHXknEx2MDT1xs67PlwIRHyR4dfTIIP3z4/KjZMqa0WMiTJZNcOjPPsMSnDxM9No1pq9kwjmrTiycGcvSRWJW50MjzTZYNtdjosGOsnPFLP19q9lE+8chlM//0ikEcER7i11V/FAaybZ7mFS59/rTpiSEqVn7ZDpe2MfbaJOiEH27jcCPyxcJfB4zckYHPOFLDoMCDyOiTi59cPsKIP3HoKxODdnbFAKucsgWTcbFasz74n/NEj72w6WvjsQMf3XJqnN1wkgkfGXx97TCyAZcxvOziyZ084JPLlrxmi0949NXktPmgq43YgF3Jj7H8k6mfDlnj2VS7w7A3kZEzBGs2XWzBoY+vTR5mBV8pj8WER1YxhvgTh9pcGUNqsaGOSXmjN/NnnD7Cp5dNfuHPV3k2/3SMkxUD0kZsKGS60yFjPEzlUUyILfL6xrSz11q8mOGU9ROmoDDEGb5x4BltvGD0ybURk+HQBARWQqZNQPMXuOzxFWlnLz49PMWi5UOZJDnsigGV8PT0YTJeadLYzx6fYWCHrD59cvpNTMmlKxYyKPxk8eUhm8VOjr1ski1n5BvXxifHBrls1DbOZ/k2ngz/9auTZxOPLiKLp1bwxSTefJGjQ07JZxsw2XJFjh26anzyZLMnJtTaKYd4/BsPE13j+TceVnL4YlPPA5ctfD5tIOxZR7Ambw6zMe2XGzwUJn2+Wzd78GycfQQvOTra5NHUCxff2ooYyxlb9MuBGm48trPPbjpshIHdZI2j7NE1HmmToWtto/KbDNvkxC2X4lLo0K2df3ra7ORLrQ+HwiYs5JR84KHwGqOrb4wN81i82XAhYSy7bMAmr/F9FkbOOrGPKWTkNjvkxdOYfljIwAJTxB95JD9zTfFFp3F2UDbgkvP6dOEpF8aNsVGbfj6SM47Cxx+Z+mzwzYZiHK9CVkxk5ML4PpYYwOQo8NqEO/MWAOAZ4YSukvFuv8koxhC7TZJ+4/nMZolgr2QkG078Nhr6yDPO5MNEr80i3MXFPkz8WQSSoW/RTTvak8ISjx7b/HRgFQt7/JSDiYEeLMbIyw3ijxwig8KpTZYO3PTJhj+bU8f8ucovDvFlh9347KHwZst48aiNGyMfFnjCyobFpYiJTKQtT/GaA7p44mC3A1Vs8CfPjjbcijbfxSAfYcObeSTf/Ggb41+tz46iXSywwlMcxYlHTx8G/daVPt8KvhIlw69xeJpDfn2OQb+1TI9vhb8Kfn7ZgZe+gvjEh99YpXFjsCO1Pkz0yOijjnntxooZFhjKOZn8ZTNb+uU0HfL86BtjX5tdetnHr8gD0qdDXtEPB1/FsoXP/tiX8J3YyISNfvOi5oM9fPJqsnKhrSA8hbx1a53SR+Hjp9jFg5/f2lvhTAePPF9ySUeceOxniw475OVQHW79dMixZQx/5qU8w8y2Prl0ySv80EPaxtXywK/+svHNiQMjYNqMMk6I8UAyGFC6ZBnEEyQbav0STo5+RAcPkVVMBF4le1OH73wkBydZSedbCTM/4Q0/WeOIDfoKWQtN0sVNT9FXtMmxrz9jpKvQE7ONmjw/5BF5+uEmr4+fvpqOmj6sNlQ64mObb2RMzvDENv0Yp4PYIaOvpkc2nfJJ1pgSjxy9+HSyCzt78BYbG+FzZUXfWy5wIvJkkRjZ0hebvljVMy8z9+WGjMK+wq444UPwklW0i5s/PhBMbCC1MbbCmr5xY+zX9jiE3Q70cBgPV/bIiVExxm79bIiRDX00ZZPnH78cauPJ1ZyXdIsTRjb0yetPf+yhcJKFhR1FHGr88pUPNeK/OSOXTeMKf+zzr2ZzyrWW6LFFhxwc+a02Fj6+6YRL21j2w4RHpvEwwYUct8YVJ5lyZiy8xuip2VUjuKr5OaZiw6dDvnzAw2a4jIefvLHGk82GPiov2sbkGJlvtvhHzSN7YqKfDXpIXe6SmTEZp8+m8S3vzxRmoEEbKgONF5C6QMmzIekmgqzxeTDrA0uOnlofX5tsgecP3zhiE6WnRuG0YUukvpoeOy1GfXyBKyYQbjIKLLB36802H2TYpt9EkM+HzUbbZBVfcetnR00OTVvJxJ94yMOABwfZcpsv84P02Y3oiAexQ1+hrzaebbpk8BA7+SFjHNEjVw6N6SvZSD4b8oKXLBvWSTjomnfjxsqNHOo7CRkTJx2Ff/6MK3jshF8f0SOX/frkyLNpXDHGp3a5gQXWydcnQz/f7GcDX4GRPgrX7qw/8fXJOsHC7HMMffJhwGefjrF80m1M27iCB4t2Jzk8eKbtZOger1c8lC1teOgj/OZH3vhCbMLn2NJOjq42Mo70ta0POYWPTT6yN33KgWJcYZ9vNool+amvTa/P2uiGK5z80w0n+frZTobPxppjdvDDVHx8h1GdDP/tM2zBpBYLm8bZ0g8vm3jhIW9fCpfx4lbTnfNKL4IDiRO1dxU3Hhvhqo8HG990EJ2LNkEDKQDFiQWIKOKpSxrwDOkzpm3DAtQYYpydJooPsvyQo0smeXLanYTI8YnUfJQIdvEUciVFP6xk+UuXL2OSi/Sj4mWLf2P8TazaCEY2kqE7J5t+uMRYzsKors0eWWTCYaavraQ/bbbAbIDiI2McBpg7qbFpDI+sdrHzr9+C4Lfcku+gEGMy7MkfPdQCNY7Y1iZjrHWRfDGziceHml+5xZeLsBrDJwcTOTbUYSKPp+DNPn19uogM2whGZDwij/hT4Jl3I8bKEf3kytfMTXbVCt/kFH1Y2ebTfz9bU/4z2jySkTv5hEHpbo4s8pirHLORDlmbpT4fxviWR22+4S5fan3P/Y17hbWY+W9O01MrqFxo86Xwzzc9sfT6LD55vjyqI2NcDB1PeOHFZ49OuTOmT05+tMs5GQQbPl281hXd/OCzbxPP/tRhnx024KUn/9OGthiThYMNdssZHjswFJu+gsgjfXrkEL948M1jiIw9mV0xwMUu+Wp6dNStJXr5pyemiC7CJ4PgUtgQi7Z4yeDRwSv+/WWKCRIo6YBqZ5ADfQA6yBkreZzoA6zkMDAFQYY/yRAkYlMxhiSBL0mEib/G9FuoZAsEDwb++EbaKGx08Y4PEjyFT6WY2C4nbJIxVkLFgJecseTFq8BFN31xkJEjvuiHF9by3JgasU2XLH/xwlRM2cbXzl4x0Zdr/OIgVzEe5u1k/UlO3zhMCh1+FMSmkg8HHr/wws4uWVfdMJjf5Okko2afHnl98vyxJ3fG4DIejmq6yfOJ6IWbDn52m0cy1mTrUl9e8wcrv+WIHX1+K9vZ+kMGhUmN8Okh9mDQx9fXFldUjmwYMMNmPL9iomOMDXHnIz/wh0+bH7FlL1t8aodJm332FO1yns9yStcYXbWvSffNy+Ya4ftvf984XdzJsxGGfLMfftiNw6BNpnzSLbZqdiO+lCmXneIzx3yxKz/GwyNPqBzKWfbZNS9K+WSDfnkuT9lONrtwIXLsqsOqPcfExy6CwzgMqBjx+FInq21P1dcWL3kxJs9G/skZ02eLbHLa5XnLOXgETJACJiELVSD1OSBrM2dAAnKo38LFUySK3c6k7Ggj4Ngi0607/66eACWr5p8tvtjSLxA+JYUtyaSPyCBykbGJsT4eu/nAT19tjG1y7MlLfG1FDC1AfXjIy58+Yh+/Pl548fihg7QbCw+fxvWNo+z7ri9fb+FbXI07cPmCyRUgPf6/LR947OUTFoUdOtrmAuGxBUtyYWJHG5+tcoYHS/iNWSfp4/OjNpf5MU6Orhjo+cqJ4lBnqwOJLD6fSB0m9sNFJoJbXKgDzBVyPpMzx/5/RC3PreMw5FM/wmMbrxwYa606jvgs18mRKZ948t9x54SM0iOnrTbf5POpDYMaZSvb8eQGJuvJGLnwpj9tGSOvaCcvHjw4YPBbIL7Q1Jdm6ovD79b46QR3OuYXJW/O2eMTFjqw0dOuD6M2WfNRn542Yptu9sg37/DCSUZe6eiTQezQRcb0FbZgJKewY7w4zE3HJDm4y9uxvr5x+tnhj4/WsHllB8GjbSweDOywYbxctEbwioMPJ3L25Z4dbdjZEb9CLr7aWP7ZChs9fsV7EZMymkoFyAgZCjnjCE8tIPySou2LyByEnOL73ikB+k4iX14IgB9p4TefZG0SAsSXwBJABh4nEjwFHhiQNj+oMdgiuiUer8Q1LhGujNjwmEBN30bBtm/07YBnx6LvgNVnH5b02AtPPsiTE4sxeeJDmfjTw2ebPLyoqxZ8OdL3Iz6+aXYSG74Qzle+z3lKhh4ZthVULstNMhNbOPiGny4s+YA5/GrzKWZfRU2eTW1rw8FbfDbt8suGg4AP/tTWDnvymm9942qFLboKzEptOPmL4G2OyHVC4MePH/mhJrrk2FZbV74zyVf/+0pwfGuUXJjYQmHhlxwZdZjkRL8ci4MtflpzxW09yp1HXNliXw59Nbq8sde47/ASqzy7IPP9Y+JC9MVBH2nDyBdiB8GptEY7puKTERNiq3Vh3vR9+7EfPPKlptkky7/jG675nV5iFjtd8tkOb7pq4/hKuOmFTVtMCiKjsGlMLsy93Mu52lfNkzcOi3jxkd+NkeN0yglZNtnG871lvibHd36JzThMiG14xUgW8YFHBg41e8Webnrs0WkcX5GPjg97zH/PvosMZnx61kM5odO6zYexfIcL1nzATKbjUcx02T7/KhMKMQNGER+VkIzFwycPvLsJP7vq5yudKIAKLDmT45tRnWD8UplJ8utkFlqTCSjKN9uC0i9IPKQu0JLBj0DxydMzlh16xWnMgvYjQvfcc8++WoKRHhmY4PGDTX7n24aoL4Emjr42n+Fgnwx/Ciz5Ewcin4/k6JhofLmUG206LdLw2jz8oIwfb5Jrv8/im1Cd6PwWC3lfSW7R8y8O+BQ2+AzHbqw/HZh0yMGc/+TDyr62cbV44FS08wEnWbGYb7+yh2ezIwuX5+82Oz/o5RuS5djiT48t+jYdOnLU1V6+YUXwRMboGqttTGxsw8k/HXPuStnvV/jCRb+vkm1ySN8XRZK3Xv3QlPVuHF7zZ97IkcFX8x8O/Zmn5iMf+shckGWLD8QGwmPze9/73i76fNPVFlN65fEf61uyfVOyuOmSZx/BjE8Hz1g5rx8WfH7g5QcZy7fjyE+vyuPhcNjrz/Fuvm+77bYLJycn+yLSj6bJubXGZvg7AbErBx1j7MPNF/mJB254yis5PDW9GZt+/qwn/vx2xpNPPrm/XZweeTJ8w+c48ujNGqSP4DXOh4sKvxXiosM3PTupi9laliObOLzyXAz6MLMJNwpXudBXEFz8Wft42dJnxzHlgtyx7hcV/SBZucMz5vfc/VKhfZnPcsk+WT7YLkbY6YantcMf+eZg/yMhoRLeRAScYEGqSwiHMxGcu9qxqfkRnH58hQxSW0jO0IKx8UmEyTEGoAllF5aSx6dEB9wYWVQggmaDrHGlhJALqzgR+eTcHfU7yLAZ44uOwpfHQu5QLHp9fsjwCWe2yCP82rDrO5jJWpQIlvyo2UwneyaRrjwZc6XukRWeBXvlypXN+/TTTy+89dZb+0QoJ4d18JJvMZBHsGtbGOWInAIbHPja5MJHV79Y8Rs3hpLNrnjcNlsvcuzK3o84yQfKLzkbM34bizH5qRaHvHVRQpYeX3CQVeuLjZ6+XIRL7NpiMF6+8c3797///f07N67gySFXczYI/nxdfXJs/Otf/9oXH+TYgxGfXfpqGCY/HpyVqVss5JTWChnyyAnXNwv/+Mc/3pu1sWLcAmd/wuKYc1Kmf3p6utcwnfJLnK/wwKsNi4LYLxay9fnQlh82/SibebYpRzZTx45HWn5lszl0AvZzBd+G3bwhWNhVwsQ/LPr8dxw1Hrb8N26e2VG7Q3DR5STigvfBBx88v3ijR654XeD69nH2HY/FDFs2nTT9dosTid97t+79LpKNWyxtxM0J+/QV+NhW51sM4iTPB79OFmT0ydOVd3xr037rm6bd/cFBnx+2XITYmz0BcvGOV461i5fd4mObP4UMMobqb/yYFDElLWMGtQskw/EBYFzfgeZX6PyEpSsNmxybLSx2FRPnQCXrQOBzfRvwhfWjVuc/7MRPE5P9ALPHLlt4xrUnjz7ceAq5klUMkuCqV6L93rjNGGYTTQaJO/Kb2ibID1D5DRE2YSHDtj7MePTxYED45JwEovCp4VeTK7fabCL28G0ETnQeDVgMfNicXbn/9a9/3Sdii9dJ7r777tux0HdSRvzkSz3n1Th/5Uy/2PDga1zbGBvIHER4zZGrsbvuumv/Bod4HATG4FbSpyvv1o4N2q/AeRzALl9OvGq6ckgXFuNhk6PmGt94PPb5Mq6G3TibcFmT995774U77rhjr0l+itFjQL/05grPye/+++/f+mwhuJFxOuyaq/wZgyOf+m1CdFCy4iqXbDh50LMu2bCJ4fuxrscee2xfLNg88Iot3PQia8QjThu331Z3J0JHmXLlu9iMxSOrzVZ5w4MdFTsZ80W3uG1eThRkjbtL/ulPf7q/Lt/mbO8gK3dttOURlmKir00WPx08/anDPz68cmSezSuCwUmjH77i028bsSHn/NBvjYrHnHXic9IhKy4XF77J3B2XiyS2Xf37wSp6r7766l6z7LX+YQkzebjyV2xiicSQb8e2vPNv/3KCcifqx8/cNcPcUxJyiA8XRe6srSX7rZO6u8XyyS9qjujwgaq14VSMN/8XOcVkxIA65xwoDEt0i6MrIwe0Beng8whIMEA6ODz28at/bNI/HA4Xbr311v2owJU0YJIqcM95bRp0S7DEwRIuOCWbLWMlcmIMa8GS5UeNjLNhQduM4XbCcwaPLGgbB1zf/e53N9vJzqbNjgVkQhX2bNAwm3T9FoN2SWYEDrE0CWFSV4zVpssuPXmxCdx+++3795vlwlW7DySdPD7//PO9gP2crHhcAbrqg9tCN1/mj33xsxne2uGBdWIkZ6xFnVxrRJ88O2y7hfe5jBOd33CWN7J+UOhP6+eGrSFXo8ayzYYY3bnC7IedrCvj7MoD2/Knz5780sPTzpZxJVzG6mvDSscaML82EnjNK7/yylYHiR+hkm8HnHV9/fr1/WNlHn2iNp2OHb7oh51PpI/Cbj7wxIaKRU0nOceYx5UPPPDAnkObgSt5xwAM77333r6yZE+O5NZxCA9sfMipNe54s16tXXlAfJFRt97oGo/PLjwonOQVuZQfG5crbzykZgNO9thCajHYtMsZfbLZLHa84/ksP2xpo/S044lFmz/29F0sOOatM3mSs8i8+nllmMjLl4sGuRObkwVbMLmSdxJiy2PZTh5sOd7Y8TkZYo/vcgireBW5i0+WbfJkygGM9PXJ8u0RqpOwY0YuYTGnkac6nkqQ9RmOfZesi4+nn356vx3n5N0ea52ECQZ54ws+BA8igzqe9PcJBJMR4DFLOP4WWnzJUyxoZz7GgXa2tblZtBaFRe1X52xunsuTcXJwoArGs2Y2gcJ3K8sePl4HVpOvnzw8CD7jimAQbPXVDhY1vOQlpoRYGG61bWLuLGwmfgebrBMI7D5LYNPjAjHwY5MhYyFmW58c+wqCWT7FAztdGPgnQ14xjqeefDyFjgPe7bbfbIbJHJ2uRxGuIq5du7Z/4tQix7fRuMNja+YlW2p8NRnY+EAwGFOHTV+ceG1GLfD4dMlZcOzZcOX35ORkH4TunBxQf/vb3/bnXQ5Ga0HuESyIDXcg9B0AfgYVNv4cQAocCuwTRzZgyhaZ2sWLRwZOd8FOyDYAFzAeSfk1Rbf4rlodXO5MfvjDH25/1pN14rMwnykgdpXWrLyxLxa+9I2HHU9MxUCWDHmEr62mR9aJSz7hgFvcsHjGbpN455139m94OzHD7hHGzTffvP2Lwbz4iWdz4OTY8cSfOYWJzdZBOMJprDhgQjDiIZjYZIc/BTUX5MjUFxtS88GWMQVWfO349NNV6xvjT509Y/kxJkfsOenD5ziWF5u+R21k/bSy4x0OFw/2IrpOApfWD0uxebqONWs4X2S9iecizRo+rAtjOgjfr3G6qPOYHj9MxcWmwh4Sz8wreX01HXL69hnryJ39j370o31HYe/C8xKAPZQsPRfv9jBtj9/zBU/5Y1ss5ldB8XZn/QlreI/nAq6LACicMFAhnEOCnJkQi87iRT/4wQ/27467InL1YYOVwDfffHM/6sFjh32TqCA+LGYHo1f73I61SZMvEMBh4B8PHSdD8Hjq9GClazE3zoarCFed7pZcqXeSwffoxAaBHIja7PhQSnzlAx5FTG4p86NG/MiT8SaIPGxk2FHrx0tPHc/ki8vCkRtXEPIHq6uHd999d2PEJ2fTsxGKSR8GJx9tPuUh39owwGq8Me38kw2jWl8R0+zjsWNdONF5Nu8khs+Hx2pvvPHGXg/k6CvG8se/HOE54OUOya81ZHNMng3+6aCwGI/4RuzzpZY7uZFPjy6tAVeiZLPhLRprWO4RXXfL+VLD1sk0/NnITrowict48ZFRxIGynS6bZPlw9+ZkCycc8NjMXHh5dKnYtMwzvrs/vtjgW3GsWiuffPLJXi+OifJLlv+JrzzCww5Z4/KHZ12FG19B8gaXYxlWJEb6ClvHxHc5Nc4+mrz0JpZiI89HOOjhKXLYMaTtuP/JT36yN376//nPfy78+c9/3idWPp0o4LT++LIpe+POepFjtQs1axGJUWGbPyciOWbTxQhMxUdGYZ8dOSwGtXVpHF88ytTlU/5dnLsT9TmzC27kQsvJ74svvti4xQw7G7/4xS/2XYcLXgSji3p7NJ8wNq/WGywzfzDgWXcKYr/jQ3//J7qEFWTtEqVfUK5kXLmtn8HdV4s2Lc80HeQ2XI9+3FK7MsKjZ/N74okn9tWesyGy0Pzms2AEbiJaiIDO5MIFQ7za29D6I0hUoNqSUELmmCsLuNz2OZDYQibOAnJVb8wkOyBd0bOD4BCLxMNoMeCVUD7xxFyyxZQP4/pwxjcGv75xfTbkXrExuLtTPJ766quvdt4s0i+//HLfOsNg4+6RERt8wBVWMnwoiE8Y2wzCSC5dMsVDD59csvrxYNe2GN1hONDkgaz8+XDcnCNxyS8qbrrIwfDRRx+d332wR6Y8wwO3gheRaY0WA98z59rFbN06AOEsB8adsPLl9p8/fAXG5ppt/vTliY6Y8LXJaysoGTaUYtA2Bj+ePmLDJuEu2V2740MeP/jgg50bvj/77LN9cqBv85YrF2MeUyA2PUZx5/f3v/99n1y6oOLLOB0UzrDDr802X3CRJ6ddoWscNmMeWVubPlN00cWO/Dru8lGMMEweX/AZz395qa82bl3TlXN2YCVrHJ7iMz/4auvSnJO159h0fb7lYpE9fPmwNviA2+ed1msnaU9bXFDaB2zKbCIXSfLMnpzbY8TDDptsswdbpB3u5gHOYpVTuOB34rjhhhv2RY/9Vxzit1+ZW3fEveTjBO7iyEtK1oILzEi+3HU5wbiTgYk/PsJCtnkxhsRhHB9u+1LyFzUCbpAxtQASapLcSt955537mZoEklWc2a6vZ4hum5w83FoBJyjkik8C2EPsSbgDwiZNFgEKdIDJAW0y8AtMPWWNkYUbFQ++gt9YmxA5B96///3v/Uxb4vlxa2qSPDrwgdjhcNhX/e6WLCQxdMBow8E3akLYMVbCjeHpV+QtfGo0+3C6IpYf+XPF5KTrat6YHLMFi0UPr2ejTnxePfTIgo+wlEt+YNPnz7hanzwqR+WYn9pkyekXCx4cbMk9ImPcCfmXv/zlxq/Pj+fP+VLbuJ08XFFbF/oOHnZhkVe2lbDQa07JKQhPfGQRPl7x4rnY8YjBHZv8kndQ+sc3d898OgCtA37Eai23aeOVI7rhKEdqfDRx0dMvb+TYCSuMYqXrAL/pppt2Hsy9z7lcEXs84pm7jYG+dUvWxuazHI8BIx+8O3nDblMLC3/afJfPdNgUj/FimHiN69Ml0zzgO5ZsLk4eCj8evXkJxWcKZBTEBllX/NavfrGzqS8f/KA2Yvr8KuzPnGZbjR82cVhPrSn2XCDwhxwz7XdOGPRttp6S+PzAxRmeR8g2ZRce+uXHBu+xkQtiewY/SAxyWax44uAXLxJjtsTsGGqOxAmH+fOWlePJnDu5eRJhv0X4ctl6cIx18mD/dF0g+2jBydwJkJ9yTJ8fcuLiG55ktOuLhwz5LVtSGcPMmISaAFcF2ibb+9IWKQCcILfN3bo5+3aFSc9m7A6FfAkyWa6kJVHQ/BoDBmAAWxTwKC3osIWTHln4Cp5MQRZokwWTsfo2KhuJOyQ4XeV7XOCVPBuKE6NHRjaO119/fW/oTS6s4dDOV3WLQp/PSfTKt7YYKvpyTcfG0DPMjz/++MJzzz23F68rUnJyaGHZUNSInoXlNtW4HE1/ZFoY2uYRFrLk9MtR+S0mGFFY6SnkFYvTenBA2ZzpWwOu3Ngv5jYzff7E+/777++7V3PZowFtemSsKznNZ7ow8w1TbX6RPqKvrVjvDjIbsbjMu3Wh9viNrhPwfetNNnd2ZJCra5+ROBDZKRfNb3L5zB++Er7WDUx4rVt9Y2ImL7505PSVV17ZV4+d8OTMJieXV65c2Sc8Oee3+OG2YVmz7IUpu+WNP+3G6ZGhE2U33dY8XTmVQ3jgkiNX5Wz6XNTFjoshfbJOco79w7o4s1booeZJmz/yxXKMhV/y5Yo8TIgPhb5izAkDJvNsXh955JF9F2HdkWGHvvUmRnGQdSHhwtmatAblu7khZ9Nm19qX/4mHX/NJnu1igVGbLOIf6dMhb960ET9iLV5PH3z25TMbeM0vXXLI3gVr1D7nMb1csw0XeQW2csCH8fgwKNkPG9t0LmJoJJQhoCxuBUgLwC2bOw885PmquwjPWF2hRza+hx9+eL8dwHGPrpw83G6ZNAvHlQe/glXDwrZ2eOgrgjKuGDMB2rDF4z/5kmkMT2IsKnx9xK+3sZxEyCnig9cVsdtDyfVG0Ok6KOizI0fJswcL3HjG4SOD8seOsRZRNvTJk1PwnbRcQdx99937sxp225gdjOQsCovAq7LkyjG+W1TyTtzyVq7YYZ8Mwi9fs2afLMzWAXwIfmP4xrXpaZtLd2iuwvDlVS75c8VGt7yTT59deKyNfFnkE8/0y56CbFrZIoPqk9FGsLEtX+ZQTqxdjy8c+PDJhY3N40J3ppfWh6hks2EDogMrv3JijB+2xWwNyE15xSNX/shqF482ecQGotMBTF4fHseZO1GY4PChqRcrvFXjTqkYwssWm2zJJUyITzLxtJtPMvyVuykX9uyrxUGHj/zYeD1acZfn2MQnx4fcWZPulrw5BJ//X3KxgxyP/JsnOI0rfLXnwF0MdMq1Nj59PG0+1bC7g5M7r7O6W/f5kg+jrVP4lPy09vCc6FygwZFN9vStHXcA9jPx2vfmGsgmvXTDyUc5TA52pTUhDm3+0lPz4a7E4yhr8bBOxI55dxyeUFy+fHmfzLbS+kMfVv9I6A7J8cBnBAdfCv/68JYH7Xh0tPndc2tQo2SXfA60p6IFkJw3ZVwVA2xCXB07A5NxtvZWFn5ABGyhONlcW28PeSuiRccHf/QdLBYhXgktkIIsoWQsNuNkq7Uj/sOgNsYO4k9BxvLp5Oak8utf/3qPSbrJ4dfkkbWAvs0PGfbFM9uw5Sec+mT0w0jXZPPnsZ8TgxOKOBGfNlt8z8k9byaHPEKQY1dFDkZzxR4dsTV3W3j9mfjnfJODC6ZyNfEbQ8YUuTBu43fShU/b1ae5ZNtVn1twukr583jGRYX1YT3waeNBDvjyyB6qz4a5KHfiO6Zyih9WOkh+XDl++OGHe5NwcWQjtgmgclXc7HdCgRNGc0IOJnKKnKrxskO3dWMMFnJigFGdP3cY7vJtAvwZKx82CW/+2PicMGxsZN0p//Of/9wfFMs5m0iOw6aN8hPOfOtXwkuePrxsTt3sZtNxX/yeRLjDIx/Ba13YE1zVe9zqbSgnFMe84xAWvtWtK/r6/ONNSh4P9nLORnbU8ueu0mcf5swx4+TruFb+P8S/NcinDdhdrHXrsf319fje8WhTNc/WB6zNeXnNjzyxB1s5Y1dbDPJmDOHTZ884kjt39R4NGuNXjNa09eDkOInuzB3b9Ngur+FpzqyXCp5Y0mMbFv39GUgOCBVQADhCFrMTA6MMOvgsWgvegnbCsBgKWDDToUXuQz4bnKu7wLMveH75YN+mI6BK4CW3QDf4s4OULXphNUaOzYoY4LMIPBsOq764XH1aGEg+LHhkgRWzfj60Ed/FSQ7WFjIMxhGb2gobSvHByAbdeBYFO/B5q8NtK3JykD//vORDVrfZDgrPuZ2c/bOjg5HN8taiqG4j3AbHHzoIFu256PTDX25nHPTw3T57ROmRC3nz7vm3t0fcVbHjAEPWkIsQn+1oIzE3Lp/WRidPfPrhY5+8gse/dnMEv7YxsWurrS925NVJFzl50IcV2RCsC3IIFv7DwnYbBpvJ5ZtObf7Jh4cfvGTEYQMmz6ccOhHoG7NWfZagDac7Tr4dY54MyJ9N+3A4bD0Y3TFZE9ZLOModvzAoeMaViYuMsUo24K4ND8LjU+6tLTrulso5ebGbSyc9x58Ttny6kLQu3FWZF3JyzQa7bNBD/IVH3zh5Neza8WGpr26T/ctf/rIvVsj1OKrjDk6+EZ/lx77gwqI41Nar480H54eVd7hgpa9os8u3sbDAJQ724a4Y1yZLBpErx3LV/mSv8ro7+1OeDesn3GxYV/YOTyTsFfMYgnHmjS8YxGdMDQsZpTF8+JX9XViB5JBQQTJSAgAFpg3XQu4ZsefcHq0IyDidkpBzC0wQFr0E6wMgoADB0WTCwmfY2DkOmJ4yiTw9cSDjsOhbRD3/9oESmw60/inIBPFfnPTZIldcbODBgtgtR+r4ZPgOT3mgk0262ii5xppoj1m82eagc6dh8fiMxquyrkTIu0P50/onPR+YshkGdvWRA5sPBU54kH4yM/ewdzCRmUQfZatFhgcPfi8cWNBs8ZFP827DsPH50Ny6QDCimRP9cIjLWP7ZYztqDM9YcanFRi8fNis4rEPFejDv7vxs3v4HpM9qzH36YoPD/ISrMTiM57uY+YVNaS3yDWexkDXuhODK1gbrZMaPq3YvIrDrZGIz48fJ2YbojaIujthwB43ncaK33/BgLC908WbuwiwGY2JrnLw27MZglYN4MKIZU+sXn5w45ZOelwHk2tWyuxE2PdP3GMj88MMnzHyFI1vk86+tIDEo+tVs8c0WMu8uYt3ld7ynZ5ysYs5h4MdJz7cQ2O+MWS+OSycR4wqfYdVvTZBnXwxqNsNGhy21fQWfjUmtMfz+/05urdOo/Sp9PqPe3BSzNcKX8XAUI37zWf6NoermlKwY0fn/gehMw8AgwCXaRuDZptpttI1MiWzM3gwAGNBuEY1zBpzFwR7b2oAUNBnJRGQbU7dwyZCHc46zha80UfwoZBVtiXAb7fGUxWqRIAeqxY8cwP0XfY/g2EVqfiv547O4jPED0/SLRy5eNtkVFz2LRU0mX/LtHfYbb7xxH4AWmhOJq4lidFDYMJxIPLOVr3LOPmLveGHgkUPFoA1D8UwsZKLymZ+pU/zWTY9k3KEWp9fBPTf2Yb/NzsYCGzwOPLa0zQl+B5ExfFSetPEq+EpU3zhcajEh8fToRA5dUVoTvtvIZ0t0HZzX1yMKV/jabOA3j9rixC8+mOVn5pZf+Mth+RWbQoe8Dd+3KLsD8nzeQetkokSOE5/LefzmJKzt7SA+EJyORScidzMwyiXM/OsXA3l68CjGYUFkUJucsfTJInEr+IgsfWX6ybdN0Gbm9WTxidPdij3D3ZJcwMqmNj15K7f8wJu/8kguvOW6ONgjZ9xxCaM7XycBJEf0HT/F5ThrfyDjoo0+jD5PsUd49OZiY8rxDbcL5XJDHxZ9vvkiwx5dsRhD1drG4bBGPeJ1t8OOE5p/gOXbhZonQU5m1qG1K6fp28+85OQlEHd7qBjLYTj1y3mYyjvMeDAhtf4+gTCAIXilZDcBFjFhjyUk3bikF6xJOV0fMjsA33777X3yuHr16vnzRXYkwlnf4na3wiZiVxLZIqcPA+JHG3h14AWFp0+PDiKTrnFkLL3NWH8cjM7gFoN4YMuHZ/LPP//8/u9dC9vGgiSWr2KefrQnVvJ4+eZfWyEHv/FiwIvIeIZpY/V4xQnBM1v+5c7tKJnIArJZONGwB28n1OTo4iELmP+wNMcwlKfG8NgsN8WsJqtok5MXNd0WHV/WicXuq2Hwzbu4vBFiwZuL8lFO+GuzLd9w4lsr5NjhLyzhEGN2YCkmPNQBrG28Wo4urUeDrjJdGcPVRZH/e3LSw4OBTb7h1oetHJSXavaNITybWbhgCmvzAIeTrgsdr2Jbm+WCDfLm2GdN7j5sVCcnJ/t1U49d+ZBXLzC0McKJ1Owr07c+H3CWE222yJmDsIs3Yk8xJ9YXrPp06dFBdBSxyz9ZuCOy/CrasKjJ0eNfn93WA93wGVNQuQ0/eQU2cTqGyNiEnbTopaOffbZhtUk7/sy9OXLS8PbTCy+8cOFwOGy7+WKLXvuleWKvtVtummu1GMkUg1rM2arvjt1JS+1NUTL2J5/tuBDzgbrH2idrLYiHL7Ie07LvaYX4xSS/fDc/ajGgWfPNDtls8ksmzPszEIyK4BPiEL/FBbAN7MUXX9wyDJcsG55bao4QBwgIJw4bHXkbogXOJlltsp2JyQBtTGDk9GHSZ0+BS/K1LUw1HTVZduBH7GsnTxeVOOPx8GFpgyJjnN38q5Uw0tWng/QVuFHyYkHpFVP2s2Mhy6X/SZBXj6tchXim7arThuzuz12UjcSXtrHRZgOHeLOr7QQDT1jEg2BAYcIvh2qYUNjIaysT/7SNL1Zz62Khg48dm7KrKTx5DqOFDQPs5g5OPszttC2WmVcyxUInzGwpKJnWBBljxaJW2Han5H+d4Pbmk9e3nfDkT375EBtcE6N2a9d49mBV6hdvMeLDJ34UZncV7kZseEprF3a+XVnSoe8O1R2TO2k8OYZBPOHU5psvPtgJi74ihvwbV8jDyg959vTJqY3TIxuvNgzFFX525CM9MefHWrdmyCC6Sv6tl2wbn3jZU+jyRafc6zthGWMDH8FbjsSFrL/wuZL3iM0JxEUd+9auC0/rA/GD2ComtVyxE0Z9hRzSDitMybGXLbGSkxd8a9DjS4+zX3rppX1y8wj++rpDfvTRR/ddUk9T3JF49O3RoIsR+wd/rQc+sl9Om0f4ERn50SdTgQnP+ML1zaMDwRLWjwjhc0rY5CJnNbd7yGZHziS5qvdoxVk7O87YbqU9sjA55NFMGmBN+jFItpWCgweW+DBbqE2IMW21Me0mkg9JpEtHXNp45Pkmb3Nzq02f364k9PONr7BNjw0xs5k/PKSP0iGnzVY8NQqDRWPjgsdikFu2Xcm7mrC5yKn/TXCl6vbUBkfewcI2v7Bb8PrNYzHkKwzwaheTdjLq5jSccohmHMXsatgHfa7m4aHvhOcE6NVHcgpdebUe5lzx3XibEP94xsjmNzxw45HhTzG/anxUTGzh59eYR2mX1l2I/MqVde5xi1g6edBH2c9HJz7+wsHnLPRgI4NvPSK+4mXfnbG3huCjIwc2EOPa1gOf1qmLCXdO5sPdqM+VXJkieaIjx+zAxld+jJcffH1yU7cYyGUvHrlyKQ7EdkUfrjl3dNmKyMIlVjEa12cPFrp44dKGVYnYY6cYtOnDa89hw4m4GGDiy9rU5o9OJxBrwcsKHrN5kYZPd6E2a2/DuRAuF2o26CvacLAJc3jVyNiMH08fRmNskMWjzx4fsMGl75GUY9sJQ9/8twdYN9aAx9r2ECdC+pVwqvlD/OV31jCUG2tInx1tZZ8tKANK0WAJ0FcEpsYn21WlCTJGl17JwosEaYPztceeeyKAEB13J9lgmw/javYQ0KjEkyPjQAp7NRntAucDL7sWDH9w8e0NMl+yhu/uyj8N5ptudjsA8YyzUfLlhj/5KDbYlfCUy2IRD1tRfsiTYd/LCafr0aBFi1xleNbtJG1h+Cp6di2cmXs8ttVww+WqlF19FE4yiLxxfRjUYVIX69TLB6zImIPUBuw2W/HZDX0HnefOL7/88r61Fhs9uMPARzlR08vHdnD2B98ckNdmRx3G7NFVxOXgI6Ot4CM6TtaH9TgCXnfZTs7ybhN3gJIXGx2+skGfLzEYIxMG+BB5OTeGBwMdculmW82WtUjOxudEwQb55liO+fOIyrN5x5WLBCc8v0OhOOHZIOnyy7Y2n3SLIZv6YSfTeDkLrz59pK2Qj5Kjb8yGh7KtLT59Msh68exenjypKJfGzBv75MtzfDW+2PgSn7Z46eij7BlD4eU/efrs4zmR2XS9JuvkId/2Bpuyz5xcFHenxH/5dEGgT589bTV/1RvA+lPe+S8P4cEjX+GLLfudWlzWpba92P+22AOy41jz1f0w2ytQumTYVVeMh6/ciEkblhkHWXxzJYb9j4QZTJgypYIkHDht1AGSrqBMtoXsgGQLGZ9XWgHczpcOu3zFZxePDt0WgzYZfrTp6xccnjbdFg4MYkEtRJupZOr7D1PvU19e79ybdFeb/gHSf6Qbt3D4YWfaZC+8/KJw6ytwsAmTAwaVV+2w47HNDyou47DqW8Se2drcbAxOfB5xeY/eLXX46LDXYmGPvAUISxjx4SsGfdQ8NBZGdXkgE95yoiZjbViwNgOvncLNlo3YgeebDLpDbV7KST7LnZjYzLc6jM0z2fjacGSPTHKws2+MPL58IPMsj14MsRkf1onEycNjWlfz5OVQ7lx0hE+fP7ZgRezng0+yajbih4m8ebLBWh/sKeSP1yp9fH7YZAt+m4hXuX3FhicC3m66vi42PAEgI+bmhl/2869m15yRKQb2o451fbLswcF29tnJpnGUTxgdx/hiZUOZbfJeVpEDj1o8dkH8IJhR/vMFWzkxlkyxiENb7ThmZ8YWDscYm2TZgzUe7HgeE7nb9z1f9jdYZ/z5ZpOP8hMuNvHKFT67anzt+mr+I/PCH11x0IEVudCxBjxmswbgtffik+kxG11EtzZMrQd8hMfflCuP7NHNP11l8b95Y0IHUWDIBBHWTnELrD/xjWlXJKPJSjZ9fbbZ5YsOYpteNkpmgNtE6Srk86tPXp9NfQefvkSo8di2CWhbABaqA89XO3t+bIIsEslHNgo42Sh2eLJHhi248YpFjVc8dGd8+C2UbNDny+SnayzC89zTc3mPhMh5DGSzgMEGxAasZIubH30HK+ywkTfOn5ibB2MKHTL42dQmz4axZLNpfvgg70rSB3n+wfHSehQkdm/luft0AnES5IM9Y2wjPPpsqxW+yGgjtXjw44VZTIp+smryeGEvXjb4NM8OMnn1zdJOJIicxxUOxB6phClfYWHLRsmHuPRRMc3NgAz9dPlJF07FmDUqdlTcu7P+0FGsc8+1Pb6yJsTiEaG13cbBVjlke+YN1umTXP18kS9udTJi04aDDizi5k9fW+2YEy+yTrXhnGuSXfitHR/8u7qnx4d8sUmmOdTGk7dyjIfkWlzswadfTGRap1t4/ck2O4gMeX2PLcWHBzdcctudhzH87PNHlq6iXx7IKfGy2Vphw5gCv3FtfBjL8+TBKx6P290tOZ59Lub75Jzs5Fke+GUvX9pIHV/NR3ngB5UL/PThUdg1Jxc5sQBMSJQwIcocpMhRBwpHGcJXfAGcK9BuneiSQWq22WSDTUnQ1wYYKHbwCyQb+gr5ScbxCpgfcpOvzYcztqtjG53/ZZF4V8j+zd9jocPhcG6fDmoi9eFXV9jEK2dhwEf4LWSYGm/MOOzG4C6/8qBvE/Mf3P7Zym2qOw6vIVsoTizd3pYXc6kdscE//swRHAqCtTjoaoeTHiz6+OVEGx9+bSdfV5JwOnngsWUz9G0F9K0JWIzpWwNqGPFQ9uhqo2zpk0di0Q8vnjEljMnDWY7zTU6uxefE4TEQPfbwem7ObhtPc4LnuGEj3NrHePHYU+Y4LGJH8sNv8chP68U4uWywb2PA89jViQ/Rt+nagGFnE1Z8+NQVfhR44rGPwqsdjyy5MLBLDk0fsMHFn832sI4jnx84vujC5+0lPyyGzHk1XTouPhV2spdNPmcxdyg8/MaDd/bZQMazYRwu+UL69ODymNjFpceH5GHhx8VEJyeySF2ZOeMze/ndCutPvtlkm14yzQ2eccV6YIusmqziosfjKzn2FpbPFv1nvP9psjcgvuihbJd7tlHrW6woX8bpI3jKBX3tjc9Ag4QNcISnn1N9pJ+c5OMzKCD/EOZHekyAK4qoCQRMoaemC4RJ0WdHYX+ONxnkC4LtfGuHEw4Jo1PiG/OBo03YCeSWW26htq/aJNxbDd5ucPVW4tjS5hOmEsd+GNkOh7h2Us8Sn7xaPOWOX+14/HR7is9OMduMfXBuc2DHVaZNpLfZYIvKMzxssl++YFaKhUwluWSLSSwWp75c8i/uaYsOOxas3M5n3jYFj9q8lupESI8tdpQ24eaJLXw1TPlhH8kL//rsROyg4tHOVzHpN5fF4XMYuXW1yZ/iTtTrjzAhPtlwkPGtTw7xN+1PvnYY1Igs0p+Y4SFfTOzq84enTTebdP2flXVs3JWxN/RcBFoD4comGesh+9nKbrjo8YmPksOb/fTKRXOWfTn15pK7ZrllV+7dMdF1HGbTWHNDBn41W9rGzEVrJGzGyIR9A1x/2I0HH/vwIbrsZRcvzGq6PoPxo1zuoq1lJ+PT9UjTOjbOpjyzW37KRz7UxpFY5F6/fE35LbT+wJU9PH3zTJ8/vvHYcUy6GPPln97IlBtPJfrCV8ciPX4qdCcWbXlgz5j4G+dLX4EJfjk0LgbyzflFm1F3Hy3q6XQ6KHA87QxyqA+4q2NXG8YscgDIl5RuTwVtjE+BaDtIgcabC4btggFcG/Gr0FUQX64U1GyQ17eQ6fHvql0SPDf2z5G//e1vz/8fJD01mcp5wlZcJRCvSSAPJzy1N6D1hzx8xsNKFxlDxrRb9E4obk99kOdglFtf8OjL55zwnOiylw3xmmw+wtui5QMu+FBYs7GZ6w9dMbFjPtg2H/HZ0JdPtsiQ9wG0zxBsHmzaiP1n/O9///stZ9GTh8c80CMHpxpPPQtM/BqDo7a1BYdS7OVd33j22Dc2dfSRD/h9VbfPvOROzv1vhTfbtK0VxJZCj31+2cVDbRBTHr9Y6Cjw063WZkNfPhCc7NHVVkdk5d5nHDZj6xgOX2TqAsjdaTL04GUbqWFA2SeLF/7G1eGCsbg7TtnAUxDbjjE+ybuAtGZdweubc7nxOrp1AzMf5MXDTpisE36QOSFHXiGvIDb5pYfkqvyHOTv62aEjto4LdtkwbszbTC7O4NB31+RbHjydgNXJj052tvP1h00kFv4iGMplGPIVrsbTEwv7+vZm/vDIqa1Xdx+X12e37vx9YN5jK/NAN+JLYav90LhizbGdjBpeNSIv32TlNrvG8Z079j0MBmPASayEZjhDDBcYnvESLzA2jLt6czb0Pj3QHLXQCiBZ4yUpOWOAqukpxuBCfCH+6WYLn2yLhi/FQUbflYTHazY6C9trjv6z07v+bv9c4Us+P2RLmhhNIhv55lM7DGT1ySglesYBK6I75WAmJ3fGyLlyh9EjAFdx7ua8HWahyC15Exoe9lC5agy+sLCr5Lu80YunZsMYPX05ROHOJ748OZGdnJzsR4IuGOhYAzZin3ucrqu3Xus2Fh72EIza5o1vBclHmPPNp3Y21PQnZrrJNz945pF/JXIH4pGbE7U3VvpZAq9N9wiL3iRrQQ7gYwvG8JDjuzzrd4xYV2EtNrjZooOnrYQVv3HHJ3kXEu48yjVZeN2BIFfN/JOtho8t/rXh1lYXR7GwMf2yQwZPidgpTjUcxeNu8/r6MN98yZerZRdC1rTf1xCLTbpj1edk1rXHWx2D7DV/7PNdnmFF+ecXyZUx/eJMRl1+5zg9Oi4YrAcv1dig+aPjH4v7uWs8uvLReoWTHOKzHJdX8uEqR+xU+KbD3sxpeuwgunTsq04eHr/b22DzP0NyKHfljD5c064+G3jqTiD64Wiuwwq7tYvPpjnVLpcXDTrbcswgw5RNPD5BZ5oZEF5JY1DgTRh7ZG0cbvsE5azuWZ1v8GXXmZycSbNZIjYFAhybAtKHhS82kbYCr40SXn3ybNA3Rk+/4h+BPNe8vM7adJw4fIMwOa9D0oMJNvGyiZd/bbxyRJZt4xG7LQJjMCnx1frwGY/wEfvyYzH0WZIf5vFo0Mnu2rVrFy6tR1p8Z5ut8p+fbLOHGsdvHsXJhj65cJEvNvYaK37+YBS3TaCTnM8Q8GxkflXwZz/72d7sfMDHhrmUV/rsI77LV3Hwg+Bhz3g4tcmxBZs22/hIjV8s1pGY2cRDdB14PvewLhE/Norf/OY3e36th/KintjYU/ixFvjUz2fY4Cq+MJFVwpwOfHLSuBoZV9ix1j0iPFkna3dNLnjks//Jckei37ETRn3+w80Wf/rG5nrmUy74nBhbD3SzN2Oi545T7PYM38HlzlPekM9q3Hl4YUXt8RD/xSkG/03tEbKLDb7ZV2qzAxfsCj7sany48ItVDa+xbMFHvhiNhdle5ATnBOJzUthdVPhgGla+1Mmz2XHA5pw77XyTD++UC5caX2ETafNPl09t+MTo0buTsTGv5p+ufdYe6nimh58+XaSGgR3+1MUSTvxioI/gIacYF6+aL/Z2ocQgJYOSm4FphCMG1Ax2cukkwBnwCnu+FZYtE2IBuaKzqLzdYpxf4+wBgvDYgQMG7QIkY1wfGYuSp5MNQSZvc/NPQT54tHm47aPjyt5jLLYUsShwiSMf2mIWf7kJX77J8qcPo4OuMbrIeDVetmrr00Hw2zSQjcECdvDBHxZ5JA+fujj4ISNneM0vnmIcf/qjX67JpIfPJyzG23S8UeW3M1wNOTmT8zKCD8ydmH2gzm/zfrzxEwAACZ9JREFUlR47ZNu4ygle+OFKz7i2MaSNR4b9iD7+tGcMXzyIDjteh/YWnn9ylGfzDY85g7O8slUeslGu2Z2Y4QonWb7op1efTnIb1PpTnswX+WzTqW1TcxHmjrTPElxoeN1Y3fExMWmj8qIfjvKpr00m3a00/iSDNW3q82svsEabGxeNHmW6Q5VfG93jjz++LyrEIIfyTd4dU2vBHoEcZ/Dwi+Djl/z/VXVHR4rEMBCGXzYQNgUiICti4IEcLhNSuUz25nPtX9WnKp9tWWq1ZeMZBtiTn7izIWq6LfTxVjd3a60v58Vojay9GGztZY8y3fR63RnDiy0//vFr3cRcruyUuDeujg97Ns0tHzmikxucvQa9Q5LrYvgizZ/rptKNeucx7N2/OPJvjuLC1Mdbv7Ic6HApT/gQurD0//tjihkh6IUuuI3BiY6jWiA1EQgRG6W7KH79KQZJ91VDz0W13Z34QZlY/CxOoo/wTkRM/XTi4kNKDD8xcbHQxisOCR/su3iI5YLhUUX/XwHOpDhwcFO7mJTAEtoBTi++2CUUx/JhjMCHhZc6XmyJ+dATOdfvA3M+BIb82iTNFU5xtWHg3TzwgKUm9Pow6RT2+eJTP07lXG0fEG0XZDn1TNvFw4b0VtpNg+fxfv+B616MxSnXMLTZ4C2XYtLjVT5w5kevNnd27ZFyb1whxpqznGkbg2XtcJVn7zLsB4ecz8H8yYdehPYpDvDLE77lSl1hhxMRR1/hx0dceyEO/Egc+S6WMX5iE3lnay599hG+udrPHhmaGzt44mVjTQld+WRjXF1O2dAp+OAbZ37hZc8mDDhhiyOmXKa3DzzW8g0h7zLKh4sHG3vHnb75lJ+44JWwpS8nbMWKBw5ygPfmNz/4iXH2dO0JX0Z4vV7nx8XeoTo7nFsugM4sf7TQ2vAR02vJXOyneBgLW9z44CoeySYu6WGyU/iS+vj4jNEj18bw8qUaucYLF9jNX7u546HA63XBns6a0muXy/z04STs9Mv1l2AAKALMQFKME4AFUUt6YxJoAk0MEXd5ruI++HVH4g7KNwZ8COVvN7lD6cAUFxYONkdtcdmYhHhqnOCzUZekFkGfHqZfZNoQnrtKsscUDjgXD39DyuMVcyd84g/LHMXbMePi2jhsFH7ZstdWiH4Y8cuGXyUdLhbX4eaRgD9XIJ5DQun5Nvt81Xg1D4df+Ytv81lebOA0D3zjWE6tKXx9uHLvG1U2sh9hGvPsWu0xhHX9vh6x+UxBLJyMwdXmD4dOiacaV3ysW/bp9XEg5Reewm/nxYZ9ORHHesFScJBLX0QQz48y/T81DgjvSOUfXnyagz7c+KdvnnjHRUzj8ZJHsenDbj7p8KLLDje29Pa+MTEIG3u5v5og1u7JuLFdLjDDxVVhSwdbzb789xrXjxdMdvzMizRGj2dz9JqDoXjX5JGQv0qhb17Gd71h8o2TNkyF6OOXrjnQk+agbYwtbvCcI+0DsfM1rk3nQPYHY90g3e/383mp88rXY2EQ9uLEky+BnbCVM3byAVs/3uVYTWDGozmXR3358jmSJzne+duzPsN1vtqvsNnjoNZX4POvHW/rVk7jhgcd//zUzVvbXOFlY+yLsgmoA6IXGKjAjDkD2itYhOk7ZNj6cNUvef9ez+jcgbhjNe5OhMAWA3YLq5YsehIeW9IE8SDswoAjGaRa26aQcM8JHXbP5/NsfAsCR8EXjnjm44JnwxnbjVFcc8VFHL7yVlLZ4ALPGMzmakzhXw1Hu9riiu9zD3f0cIzjAY/oK8bETeduOhy6XuD4iLn+cYQR1+ahxpm9OGpxXNw9f308HsfH50i+SuhXsA6DvoFlvfnJj6KNv7Jzh2ccZ+uuFps9u3jYbzjIJT0f/LRhsi8HxWXffHA3X7V9Ka6/BuzC4d2eveGizZdNvvzFINpK7fLHFndjeGUTVuuhr+C9tXFYfFtrdXuR3jriZX3FcrfuHZO/fC2+verumY29w94aKIReTLjiKdrm1hg7OkJHcCXmtHb6zUcbTrYwtOvzi6M5ETnn3xgfxQ0SvTXVN17+Wofixl0cOpKOT5K9vBF7Kc7yHnd7QN/+kGcc1P0CvRu4XWtrg6NCzDMO6cQvnrH44Mi+tTBfY/zY8aHTVlzIrDt+/msBP85V54dXeWOf3pxgyhO9eHHDOW64iL+5M07Ss9W2v+Cf1977/f7x3BoBhwAx0MIIxkmyEAAiiPF0+mwifUB+ceg9wvIORNudk8PRpocBj6gr+iZEYBePPT756Iupjxt8ddz4uRt2OIhnAdwFGVfY0osFQ7IVmJLUfNWET7HYwCfGK/Ts8Iwbm10gevZi4WwMFp2cxo0ODj07pcWzXvQkLrjFkS9bPl3E2Opnwy9expo/G4JLfNqc39c7DJtY39vnfn9g74hHcC43YhHcCGzjYZuDvmJMHy9+YeBIX+61Wx+c9cMut3TKxtfmB8fhIJ71V/j17pcNW9jmSbpp0uavLDaOMOhJec22nFbHWZ0PvOZaDuVAruTXa8heNuabVw41Y/pwYOPQOnTYGWNnrJyyFdf8jHWw0sXNPHAynp8xuPyNLffi8yPFY8e/mOHJr7aizQ4+TG32RF+BR1rz8ONirFw0RqcNrzXBPz58t61vf9vn/Oxx55WY/OOCM5E//mzjESZ9Qsefjg+eMPDiq9CJo034sJETF93b7XY+S/ZbFd8ctV+tG5sw+cEUR4ERvj7+hD4xrm9cLH11usXB0bhHpsfvuiv/+Xw+xxgZBoJ4UbWYAtEhRgKpT1ci+LTAdEiwb9N1YPMhJQwZRCOfr7gOQGN86bWzZS95cCTSIuVL56DAqYVgzwYnGB0W4uiXNBi74LhKGkzxjJHlrx+v2vAIPEVcNTttOIr4sLJNj3tcYeGvsIWjLbfahB4WW+tJ3zsCOrh8CA7wm/dybYw9/XIoV62HfMDceenD5kvPn8g3PnC1+Yq1nForcfiy02YDjw9sJUy1GOzS6xdHjHjXZudgducrjrs8NtZZX5Fb8XCQW74dtvHmg1dzaM5is2lNDtnrHzqFiJFdeNVw8sXJfHCJm9oc5LPXlT4/mNq9duSUnsCHlS2cYtGZn7mIZW6wFWP85ES79YBJr8+On4JD2OyJPoFL17sOr1M6hR+c7GDLA+Evjto8HKLE/PRhWivz0Vfzz695quOf3wG6/inncPAoPl58kuZAJy4/cfjVV4ulJlvvmsCAJ1Zrx54vias+vmy1YYgpdjbFbt782RM+7MTQ5sfOOKxyp8/OOhrvBtE60emz/wddG1SdFVJmMgAAAABJRU5ErkJggg=="}}},{"cell_type":"markdown","source":"<a class=\"anchor\" id=\"bcImp\"></a>\n# Table of Contents\n* [<b>Abstract</b>](#abstract)\n* [<b>Import dependencies and load the data</b>](#importData)\n* [<b>Data overview</b>](#dOverview)\n     * [Diamension of train and test data](#dTrainTest)\n     * [Visualizing the data using TSNE](#visualizeTSNE)\n     * [Diamension of training data](#dTrainVal)\n     * [Converting training, and testing data into array](#reTrainTestVal)\n     * [Diamension of training, and testing data after reshape](#dreshape)\n     * [<b>In summary</b>](#summaryData)\n* [<b>Explore the data</b>](#exploreData)\n     * [Visualise how the digits were written](#visualizeData)\n     * [Reshaping train, test, and validation data](#reshapeData)\n     * [Normalize train, test, and validation data](#normalizeData)\n     * [<b>In summary</b>](#summaryExploreData)\n* [<b>Build the CNN model to Classify Handwritten Digits</b>](#modeling)\n     * [Summary of the training model](#modelSummary)\n     * [Visualization of the model using graphviz](#modelplot)\n     * [Compile the model using keras.optimizers.Adam](#compileModel)\n     * [Train the model](#trainModel)\n     * [<b>In summary</b>](#modelBuildSummary)\n* [<b>Model evaluation</b>](#modelEvaluation)\n<a class=\"anchor\" id=\"bcImp1\"></a>\n     * [Loss plot curve for training and validation dataset](#lossPlot)\n     * [Accuracy plot curve for training and validation dataset](#accuracyPlot)\n     * [<b>Evaluation of the model accuracy</b>](#accuracyEvaluation)\n         * [Performance of training dataset](#perfTrain)\n         * [Load the model](#saveModel)\n     * [<b>In summary</b>](#modelEvaluationSummary)\n* [<b>Model prediction on unseen dataset (test data)</b>](#modelprediction)\n     * [Visualise test predicted data how the digits were written](#visualizePredict)\n* [<b>Submission</b>](#submission)","metadata":{"papermill":{"duration":0.018981,"end_time":"2022-07-03T08:26:59.998541","exception":false,"start_time":"2022-07-03T08:26:59.97956","status":"completed"},"tags":[]}},{"cell_type":"markdown","source":"## <b>1. Abstract</b><a class=\"anchor\" id=\"abstract\"></a> \n### <b>Introduction</b>\nIn this project, we will build a <b>[Convolutional Neural Network (CNN)](https://en.wikipedia.org/wiki/Convolutional_neural_network)</b> model using <b>Tensorflow framework</b> to <b>Recognition of Handwritten Digit</b>.\n\nA convolutional neural network (CNN, or ConvNet) is a Deep Learning algorithm that can take in an input image, assign learnable weights and biases to various objects in the image and be able to distinguish one from the other. \n\nIt is used to analyse visual imagery. Object detection, face recognition, robotics, video analysis, segmentation, pattern\nrecognition, natural language processing, spam detection, topic categorization, regression analysis, speech recognition, image classification are some of the examples that can be done using Convolutional Neural Networking.\n### <b>Approach</b>\nWe have used Sequential Keras model which have two pairs of Convolution2D and MaxPooling2D layers. The MaxPooling layer acts as a sort of downsampling using max values in a region instead of averaging. After that we will use Flatten layer to convert multidimensional parameters to vector.\n\nThe last layer have a Dense layer with 10 Softmax outputs. The output represents the network guess. The 0-th output represents a probability that the input digit is 0, the 1-st output represents a probability that the input digit is 1 and so on...\n#### <b>Result</b>\nCNN performed well, providing validation accuaracy and loss score of 0.988 and 0.046 respectively.\n\n#### <b>Conclusion</b>\nConvolutional neural network (CNN, or ConvNet) can be used to predict Handwritten Digits reasonably.\n\n<b>Keywords</b> – convolutional neural network, mnist, deep learning, handwritten digits recognition.","metadata":{"papermill":{"duration":0.019499,"end_time":"2022-07-03T08:27:00.03763","exception":false,"start_time":"2022-07-03T08:27:00.018131","status":"completed"},"tags":[]}},{"cell_type":"markdown","source":"# So, let's get started 🧑👈🙏💪","metadata":{"papermill":{"duration":0.01917,"end_time":"2022-07-03T08:27:00.076521","exception":false,"start_time":"2022-07-03T08:27:00.057351","status":"completed"},"tags":[]}},{"cell_type":"markdown","source":"## 2. Import dependencies and load the data <a class=\"anchor\" id=\"importData\"></a>\n[Back to Table of Contents](#bcImp)","metadata":{"papermill":{"duration":0.019559,"end_time":"2022-07-03T08:27:00.115443","exception":false,"start_time":"2022-07-03T08:27:00.095884","status":"completed"},"tags":[]}},{"cell_type":"code","source":"# This Python 3 environment comes with many helpful analytics libraries installed\n# It is defined by the kaggle/python Docker image: https://github.com/kaggle/docker-python\n# For example, here's several helpful packages to load\n\nimport tensorflow as tf\nimport matplotlib.pyplot as plt\nimport seaborn as sn\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\nimport math\nimport datetime\nimport platform\n\n# Input data files are available in the read-only \"../input/\" directory\n# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory\n\nimport os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        print(os.path.join(dirname, filename))\n\n# You can write up to 20GB to the current directory (/kaggle/working/) that gets preserved as output when you create a version using \"Save & Run All\" \n# You can also write temporary files to /kaggle/temp/, but they won't be saved outside of the current session","metadata":{"papermill":{"duration":9.257769,"end_time":"2022-07-03T08:27:09.393086","exception":false,"start_time":"2022-07-03T08:27:00.135317","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:26:47.989627Z","iopub.execute_input":"2022-07-30T04:26:47.991219Z","iopub.status.idle":"2022-07-30T04:26:57.965383Z","shell.execute_reply.started":"2022-07-30T04:26:47.991088Z","shell.execute_reply":"2022-07-30T04:26:57.964292Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print('Python version:', platform.python_version())\nprint('Tensorflow version:', tf.__version__)\nprint('Keras version:', tf.keras.__version__)","metadata":{"papermill":{"duration":1.133502,"end_time":"2022-07-03T08:27:10.546922","exception":false,"start_time":"2022-07-03T08:27:09.41342","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:26:57.967111Z","iopub.execute_input":"2022-07-30T04:26:57.967927Z","iopub.status.idle":"2022-07-30T04:26:59.167971Z","shell.execute_reply.started":"2022-07-30T04:26:57.967890Z","shell.execute_reply":"2022-07-30T04:26:59.166727Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Load the TensorBoard notebook extension.\n# %reload_ext tensorboard\n%load_ext tensorboard\n\n# Clear any logs from previous runs.\n!rm -rf ./.logs/","metadata":{"papermill":{"duration":0.035963,"end_time":"2022-07-03T08:27:10.602664","exception":false,"start_time":"2022-07-03T08:27:10.566701","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:26:59.169666Z","iopub.execute_input":"2022-07-30T04:26:59.170694Z","iopub.status.idle":"2022-07-30T04:26:59.959989Z","shell.execute_reply.started":"2022-07-30T04:26:59.170648Z","shell.execute_reply":"2022-07-30T04:26:59.958149Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Load MNIST data","metadata":{}},{"cell_type":"code","source":"train_mnist = pd.read_csv('/kaggle/input/digit-recognizer/train.csv')\ntest_mnist = pd.read_csv('/kaggle/input/digit-recognizer/test.csv')","metadata":{"papermill":{"duration":4.753376,"end_time":"2022-07-03T08:27:16.226866","exception":false,"start_time":"2022-07-03T08:27:11.47349","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:26:59.963767Z","iopub.execute_input":"2022-07-30T04:26:59.964653Z","iopub.status.idle":"2022-07-30T04:27:05.834152Z","shell.execute_reply.started":"2022-07-30T04:26:59.964599Z","shell.execute_reply":"2022-07-30T04:27:05.832893Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"test_mnist.shape","metadata":{"execution":{"iopub.status.busy":"2022-07-30T04:27:05.835709Z","iopub.execute_input":"2022-07-30T04:27:05.836203Z","iopub.status.idle":"2022-07-30T04:27:05.851801Z","shell.execute_reply.started":"2022-07-30T04:27:05.836159Z","shell.execute_reply":"2022-07-30T04:27:05.850639Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_mnist.shape","metadata":{"execution":{"iopub.status.busy":"2022-07-30T04:27:05.854743Z","iopub.execute_input":"2022-07-30T04:27:05.855650Z","iopub.status.idle":"2022-07-30T04:27:05.930360Z","shell.execute_reply.started":"2022-07-30T04:27:05.855597Z","shell.execute_reply":"2022-07-30T04:27:05.929367Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from tensorflow.keras.datasets import mnist\n\n# Load MNIST data\n(X_train_mnist, y_train_mnist), (X_test_mnist, y_test_mnist) = mnist.load_data()","metadata":{"execution":{"iopub.status.busy":"2022-07-30T04:27:05.932161Z","iopub.execute_input":"2022-07-30T04:27:05.932752Z","iopub.status.idle":"2022-07-30T04:27:06.624457Z","shell.execute_reply.started":"2022-07-30T04:27:05.932708Z","shell.execute_reply":"2022-07-30T04:27:06.623318Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_test_mnist.shape\nX_test_mnist_re = X_test_mnist.reshape(10000,784)\nX_test_mnist_re.shape","metadata":{"execution":{"iopub.status.busy":"2022-07-30T04:27:06.625863Z","iopub.execute_input":"2022-07-30T04:27:06.626200Z","iopub.status.idle":"2022-07-30T04:27:06.633879Z","shell.execute_reply.started":"2022-07-30T04:27:06.626167Z","shell.execute_reply":"2022-07-30T04:27:06.632712Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_train_mnist.shape\nX_train_mnist_re = X_train_mnist.reshape(60000,784)\nX_train_mnist_re.shape\npd.DataFrame(X_train_mnist_re).head()","metadata":{"execution":{"iopub.status.busy":"2022-07-30T04:27:06.634823Z","iopub.execute_input":"2022-07-30T04:27:06.635142Z","iopub.status.idle":"2022-07-30T04:27:06.664980Z","shell.execute_reply.started":"2022-07-30T04:27:06.635112Z","shell.execute_reply":"2022-07-30T04:27:06.663833Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_test_mnist.sum(axis=1)","metadata":{"execution":{"iopub.status.busy":"2022-07-30T04:27:06.670129Z","iopub.execute_input":"2022-07-30T04:27:06.670570Z","iopub.status.idle":"2022-07-30T04:27:06.692001Z","shell.execute_reply.started":"2022-07-30T04:27:06.670538Z","shell.execute_reply":"2022-07-30T04:27:06.691100Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_train_mnist = train_mnist.iloc[:, 1:785]\ny_train_mnist = train_mnist.iloc[:, 0]\n\nX_test_mnist = test_mnist.iloc[:, 0:784]","metadata":{"execution":{"iopub.status.busy":"2022-07-30T04:27:06.693364Z","iopub.execute_input":"2022-07-30T04:27:06.693961Z","iopub.status.idle":"2022-07-30T04:27:06.700201Z","shell.execute_reply.started":"2022-07-30T04:27:06.693924Z","shell.execute_reply":"2022-07-30T04:27:06.699395Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_train_mnist.shape, y_train_mnist.shape,type(X_train_mnist),type(y_train_mnist)","metadata":{"execution":{"iopub.status.busy":"2022-07-30T04:27:06.701323Z","iopub.execute_input":"2022-07-30T04:27:06.702386Z","iopub.status.idle":"2022-07-30T04:27:06.718242Z","shell.execute_reply.started":"2022-07-30T04:27:06.702347Z","shell.execute_reply":"2022-07-30T04:27:06.717381Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_train_mnist.head()","metadata":{"execution":{"iopub.status.busy":"2022-07-30T04:27:06.719222Z","iopub.execute_input":"2022-07-30T04:27:06.719891Z","iopub.status.idle":"2022-07-30T04:27:06.744234Z","shell.execute_reply.started":"2022-07-30T04:27:06.719859Z","shell.execute_reply":"2022-07-30T04:27:06.743437Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### QMNIST - The Extended MNIST Dataset (120k images)\nUsing to improve the performance of the digit recognition model with the expanded version of MNIST data\n### Load QMNIST data","metadata":{}},{"cell_type":"code","source":"def unpickle(file):\n    import pickle\n    with open(file, 'rb') as fo:\n        dict = pickle.load(fo, encoding='bytes')\n    return dict","metadata":{"execution":{"iopub.status.busy":"2022-07-30T04:27:06.745786Z","iopub.execute_input":"2022-07-30T04:27:06.746132Z","iopub.status.idle":"2022-07-30T04:27:06.756292Z","shell.execute_reply.started":"2022-07-30T04:27:06.746100Z","shell.execute_reply":"2022-07-30T04:27:06.755181Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Read qmnist data\nqmnist = unpickle(\"/kaggle/input/qmnist-the-extended-mnist-dataset-120k-images/MNIST-120k\")\nX_train_qmnist = qmnist['data']\n\ny_train_qmnist = pd.Series(qmnist['labels'].reshape(120000))","metadata":{"execution":{"iopub.status.busy":"2022-07-30T04:27:06.757663Z","iopub.execute_input":"2022-07-30T04:27:06.758386Z","iopub.status.idle":"2022-07-30T04:27:07.628177Z","shell.execute_reply.started":"2022-07-30T04:27:06.758353Z","shell.execute_reply":"2022-07-30T04:27:07.626964Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_train_qmnist.shape, y_train_qmnist.shape,type(X_train_qmnist),type(y_train_qmnist)","metadata":{"execution":{"iopub.status.busy":"2022-07-30T04:27:07.629765Z","iopub.execute_input":"2022-07-30T04:27:07.630105Z","iopub.status.idle":"2022-07-30T04:27:07.638079Z","shell.execute_reply.started":"2022-07-30T04:27:07.630074Z","shell.execute_reply":"2022-07-30T04:27:07.636955Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_train_qmnist_re = X_train_qmnist.reshape(120000,784)","metadata":{"execution":{"iopub.status.busy":"2022-07-30T04:27:07.639272Z","iopub.execute_input":"2022-07-30T04:27:07.639646Z","iopub.status.idle":"2022-07-30T04:27:07.650454Z","shell.execute_reply.started":"2022-07-30T04:27:07.639615Z","shell.execute_reply":"2022-07-30T04:27:07.649211Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_train_qmnist_re = pd.DataFrame(X_train_qmnist_re)","metadata":{"execution":{"iopub.status.busy":"2022-07-30T04:27:07.651980Z","iopub.execute_input":"2022-07-30T04:27:07.652745Z","iopub.status.idle":"2022-07-30T04:27:07.663021Z","shell.execute_reply.started":"2022-07-30T04:27:07.652694Z","shell.execute_reply":"2022-07-30T04:27:07.662029Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_train_qmnist_re.shape, y_train_qmnist.shape,type(X_train_qmnist_re),type(y_train_qmnist)","metadata":{"execution":{"iopub.status.busy":"2022-07-30T04:27:07.664335Z","iopub.execute_input":"2022-07-30T04:27:07.665363Z","iopub.status.idle":"2022-07-30T04:27:07.677416Z","shell.execute_reply.started":"2022-07-30T04:27:07.665327Z","shell.execute_reply":"2022-07-30T04:27:07.676630Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#X_train_mnist.columns\nX_train_qmnist_re.columns = X_train_mnist.columns\nX_train_qmnist_re.head()","metadata":{"execution":{"iopub.status.busy":"2022-07-30T04:27:07.678358Z","iopub.execute_input":"2022-07-30T04:27:07.678696Z","iopub.status.idle":"2022-07-30T04:27:07.703043Z","shell.execute_reply.started":"2022-07-30T04:27:07.678666Z","shell.execute_reply":"2022-07-30T04:27:07.702171Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Combine MNIST and QMNIST","metadata":{}},{"cell_type":"code","source":"X_train_mnist.shape,X_train_qmnist_re.shape","metadata":{"execution":{"iopub.status.busy":"2022-07-30T04:27:07.704434Z","iopub.execute_input":"2022-07-30T04:27:07.705676Z","iopub.status.idle":"2022-07-30T04:27:07.717724Z","shell.execute_reply.started":"2022-07-30T04:27:07.705626Z","shell.execute_reply":"2022-07-30T04:27:07.716711Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_train = pd.concat([X_train_mnist,X_train_qmnist_re],axis=0)","metadata":{"execution":{"iopub.status.busy":"2022-07-30T04:27:07.718853Z","iopub.execute_input":"2022-07-30T04:27:07.719825Z","iopub.status.idle":"2022-07-30T04:27:08.419980Z","shell.execute_reply.started":"2022-07-30T04:27:07.719788Z","shell.execute_reply":"2022-07-30T04:27:08.419157Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"y_train_mnist.shape,y_train_qmnist.shape","metadata":{"execution":{"iopub.status.busy":"2022-07-30T04:27:08.421335Z","iopub.execute_input":"2022-07-30T04:27:08.421885Z","iopub.status.idle":"2022-07-30T04:27:08.428506Z","shell.execute_reply.started":"2022-07-30T04:27:08.421851Z","shell.execute_reply":"2022-07-30T04:27:08.426992Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"y_train = pd.concat([y_train_mnist,y_train_qmnist],axis=0)","metadata":{"execution":{"iopub.status.busy":"2022-07-30T04:27:08.430332Z","iopub.execute_input":"2022-07-30T04:27:08.430895Z","iopub.status.idle":"2022-07-30T04:27:08.439876Z","shell.execute_reply.started":"2022-07-30T04:27:08.430845Z","shell.execute_reply":"2022-07-30T04:27:08.438923Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_train.shape, y_train.shape","metadata":{"execution":{"iopub.status.busy":"2022-07-30T04:27:08.441622Z","iopub.execute_input":"2022-07-30T04:27:08.442322Z","iopub.status.idle":"2022-07-30T04:27:08.452499Z","shell.execute_reply.started":"2022-07-30T04:27:08.442276Z","shell.execute_reply":"2022-07-30T04:27:08.451335Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## 3. Data Overview <a class=\"anchor\" id=\"dOverview\"></a>\n[Back to Table of Contents](#bcImp)\n### 3.1 Diamension of train and test data <a class=\"anchor\" id=\"dTrainTest\"></a>\n[Back to Table of Contents](#bcImp)","metadata":{"papermill":{"duration":0.020332,"end_time":"2022-07-03T08:27:16.267203","exception":false,"start_time":"2022-07-03T08:27:16.246871","status":"completed"},"tags":[]}},{"cell_type":"code","source":"X_train.head()","metadata":{"papermill":{"duration":0.052282,"end_time":"2022-07-03T08:27:16.339354","exception":false,"start_time":"2022-07-03T08:27:16.287072","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:27:08.454215Z","iopub.execute_input":"2022-07-30T04:27:08.454575Z","iopub.status.idle":"2022-07-30T04:27:08.479659Z","shell.execute_reply.started":"2022-07-30T04:27:08.454543Z","shell.execute_reply":"2022-07-30T04:27:08.478802Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_train.info(), X_train.shape","metadata":{"papermill":{"duration":0.075758,"end_time":"2022-07-03T08:27:16.435268","exception":false,"start_time":"2022-07-03T08:27:16.35951","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:27:08.480822Z","iopub.execute_input":"2022-07-30T04:27:08.481682Z","iopub.status.idle":"2022-07-30T04:27:08.545290Z","shell.execute_reply.started":"2022-07-30T04:27:08.481646Z","shell.execute_reply":"2022-07-30T04:27:08.543954Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_test_mnist.info(),  X_test_mnist.shape","metadata":{"papermill":{"duration":0.066692,"end_time":"2022-07-03T08:27:16.521931","exception":false,"start_time":"2022-07-03T08:27:16.455239","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:27:08.552297Z","iopub.execute_input":"2022-07-30T04:27:08.552643Z","iopub.status.idle":"2022-07-30T04:27:08.601671Z","shell.execute_reply.started":"2022-07-30T04:27:08.552613Z","shell.execute_reply":"2022-07-30T04:27:08.600848Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### 3.2 Visualizing the data using TSNE <a class=\"anchor\" id=\"visualizeTSNE\"></a>\n[Back to Table of Contents](#bcImp)\n\n<b>TSNE</b> - t-Distributed Stochastic Neighbor embedding. This is a dimensionality reduction algorithm that is designed to keep local structure in the high dimensional data set, but cares less about global structure. Here, we use it to go from the 784 pixel-dimension of the images to two dimensions. This makes plotting easier. The color scale is the original MNIST label and one can see that the separation of the labels is apparent.","metadata":{"papermill":{"duration":0.020162,"end_time":"2022-07-03T08:27:16.612689","exception":false,"start_time":"2022-07-03T08:27:16.592527","status":"completed"},"tags":[]}},{"cell_type":"code","source":"# WARNING: running t-SNE on the full data set takes a while.\n# X_tsn = X_train/255\n\n# from sklearn.manifold import TSNE\n# tsne = TSNE()\n\n# tsne_res = tsne.fit_transform(X_tsn)","metadata":{"papermill":{"duration":355.495505,"end_time":"2022-07-03T08:33:12.12856","exception":false,"start_time":"2022-07-03T08:27:16.633055","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:27:08.602831Z","iopub.execute_input":"2022-07-30T04:27:08.603330Z","iopub.status.idle":"2022-07-30T04:27:08.607828Z","shell.execute_reply.started":"2022-07-30T04:27:08.603298Z","shell.execute_reply":"2022-07-30T04:27:08.606423Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# plt.figure(figsize=(14, 12))\n# plt.scatter(tsne_res[:,0], tsne_res[:,1], c=y, s=2)\n# plt.xticks([])\n# plt.yticks([])\n# plt.colorbar()","metadata":{"papermill":{"duration":0.788494,"end_time":"2022-07-03T08:33:12.938332","exception":false,"start_time":"2022-07-03T08:33:12.149838","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:27:08.609601Z","iopub.execute_input":"2022-07-30T04:27:08.609913Z","iopub.status.idle":"2022-07-30T04:27:08.619987Z","shell.execute_reply.started":"2022-07-30T04:27:08.609884Z","shell.execute_reply":"2022-07-30T04:27:08.618990Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## 3.3 Diamension of training data<a class=\"anchor\" id=\"dTrainVal\"></a>\n[Back to Table of Contents](#bcImp)","metadata":{"papermill":{"duration":0.032095,"end_time":"2022-07-03T08:33:13.364895","exception":false,"start_time":"2022-07-03T08:33:13.3328","status":"completed"},"tags":[]}},{"cell_type":"code","source":"print('X_train:', X_train.shape)\nprint('y_train:', y_train.shape)\n# print('X_validation:', X_validation.shape)\n# print('y_validation:', y_validation.shape)","metadata":{"papermill":{"duration":0.042175,"end_time":"2022-07-03T08:33:13.439169","exception":false,"start_time":"2022-07-03T08:33:13.396994","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:27:08.621291Z","iopub.execute_input":"2022-07-30T04:27:08.621918Z","iopub.status.idle":"2022-07-30T04:27:08.631383Z","shell.execute_reply.started":"2022-07-30T04:27:08.621873Z","shell.execute_reply":"2022-07-30T04:27:08.630545Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### 3.4. Converting training, and testing data into array<a class=\"anchor\" id=\"reTrainTestVal\"></a>\n[Back to Table of Contents](#bcImp)","metadata":{"papermill":{"duration":0.031213,"end_time":"2022-07-03T08:33:13.502016","exception":false,"start_time":"2022-07-03T08:33:13.470803","status":"completed"},"tags":[]}},{"cell_type":"code","source":"pd.DataFrame(y_train).head()","metadata":{"execution":{"iopub.status.busy":"2022-07-30T04:27:08.632998Z","iopub.execute_input":"2022-07-30T04:27:08.633511Z","iopub.status.idle":"2022-07-30T04:27:08.645660Z","shell.execute_reply.started":"2022-07-30T04:27:08.633456Z","shell.execute_reply":"2022-07-30T04:27:08.644636Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"x_train_re = X_train.to_numpy().reshape(162000, 28, 28)\ny_train_re = y_train.values\n#x_validation_re = X_validation.to_numpy().reshape(32400, 28, 28)\n#y_validation_re = y_validation.values\nx_test_re = X_test_mnist.to_numpy().reshape(28000, 28, 28)","metadata":{"papermill":{"duration":0.040004,"end_time":"2022-07-03T08:33:13.575064","exception":false,"start_time":"2022-07-03T08:33:13.53506","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:27:08.647000Z","iopub.execute_input":"2022-07-30T04:27:08.647521Z","iopub.status.idle":"2022-07-30T04:27:08.656915Z","shell.execute_reply.started":"2022-07-30T04:27:08.647461Z","shell.execute_reply":"2022-07-30T04:27:08.656020Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### 3.5. Diamension of training, and testing data after reshape<a class=\"anchor\" id=\"dreshape\">\n[Back to Table of Contents](#bcImp)","metadata":{"papermill":{"duration":0.031868,"end_time":"2022-07-03T08:33:13.638693","exception":false,"start_time":"2022-07-03T08:33:13.606825","status":"completed"},"tags":[]}},{"cell_type":"code","source":"print('x_train:', x_train_re.shape)\nprint('y_train:', y_train_re.shape)\n#print('x_validation:', x_validation_re.shape)\n#print('y_validation:', y_validation_re.shape)\nprint('x_test:', x_test_re.shape)","metadata":{"papermill":{"duration":0.041854,"end_time":"2022-07-03T08:33:13.712401","exception":false,"start_time":"2022-07-03T08:33:13.670547","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:27:08.657971Z","iopub.execute_input":"2022-07-30T04:27:08.658748Z","iopub.status.idle":"2022-07-30T04:27:08.675962Z","shell.execute_reply.started":"2022-07-30T04:27:08.658714Z","shell.execute_reply":"2022-07-30T04:27:08.674501Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Save image parameters to the constants that we will use later for data re-shaping and for model traning.\n(_, IMAGE_WIDTH, IMAGE_HEIGHT) = x_train_re.shape\nIMAGE_CHANNELS = 1\n\nprint('IMAGE_WIDTH:', IMAGE_WIDTH);\nprint('IMAGE_HEIGHT:', IMAGE_HEIGHT);\nprint('IMAGE_CHANNELS:', IMAGE_CHANNELS);","metadata":{"papermill":{"duration":0.042011,"end_time":"2022-07-03T08:33:13.787149","exception":false,"start_time":"2022-07-03T08:33:13.745138","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:27:08.677899Z","iopub.execute_input":"2022-07-30T04:27:08.678550Z","iopub.status.idle":"2022-07-30T04:27:08.688666Z","shell.execute_reply.started":"2022-07-30T04:27:08.678504Z","shell.execute_reply":"2022-07-30T04:27:08.687220Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":" ### In summary <a class =\"anchor\" id =\"summaryData\"></a>\n [Back to Table of Contents](#bcImp)","metadata":{"papermill":{"duration":0.032546,"end_time":"2022-07-03T08:33:13.851637","exception":false,"start_time":"2022-07-03T08:33:13.819091","status":"completed"},"tags":[]}},{"cell_type":"markdown","source":"## 4. Explore the data <a class=\"anchor\" id=\"exploreData\">\n[Back to Table of Contents](#bcImp)\n    \nHere is how each image in the dataset looks like. It is a 28x28 matrix of integers (from 0 to 255) and each integer represents a color of a pixel.","metadata":{"papermill":{"duration":0.031188,"end_time":"2022-07-03T08:33:13.914574","exception":false,"start_time":"2022-07-03T08:33:13.883386","status":"completed"},"tags":[]}},{"cell_type":"code","source":"pd.DataFrame(x_train_re[0])","metadata":{"papermill":{"duration":0.055576,"end_time":"2022-07-03T08:33:14.001776","exception":false,"start_time":"2022-07-03T08:33:13.9462","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:27:08.690170Z","iopub.execute_input":"2022-07-30T04:27:08.691249Z","iopub.status.idle":"2022-07-30T04:27:08.720815Z","shell.execute_reply.started":"2022-07-30T04:27:08.691203Z","shell.execute_reply":"2022-07-30T04:27:08.719528Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### 4.1 Visualise how the digits were written <a class =\"anchor\" id=\"visualizeData\"></a>\n[Back to Table of Contents](#bcImp)\n\nThis number matrix may be drawn as follows:","metadata":{"papermill":{"duration":0.032478,"end_time":"2022-07-03T08:33:14.067314","exception":false,"start_time":"2022-07-03T08:33:14.034836","status":"completed"},"tags":[]}},{"cell_type":"code","source":"plt.imshow(x_train_re[0], cmap=plt.cm.binary)\nplt.show()","metadata":{"papermill":{"duration":0.200878,"end_time":"2022-07-03T08:33:14.300872","exception":false,"start_time":"2022-07-03T08:33:14.099994","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:27:08.722347Z","iopub.execute_input":"2022-07-30T04:27:08.723741Z","iopub.status.idle":"2022-07-30T04:27:08.936443Z","shell.execute_reply.started":"2022-07-30T04:27:08.723691Z","shell.execute_reply":"2022-07-30T04:27:08.935177Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Our first record of training data represents 1.","metadata":{"papermill":{"duration":0.032639,"end_time":"2022-07-03T08:33:14.36674","exception":false,"start_time":"2022-07-03T08:33:14.334101","status":"completed"},"tags":[]}},{"cell_type":"markdown","source":"Let's print some more training examples to get the feeling of how the digits were written.","metadata":{"papermill":{"duration":0.032556,"end_time":"2022-07-03T08:33:14.43225","exception":false,"start_time":"2022-07-03T08:33:14.399694","status":"completed"},"tags":[]}},{"cell_type":"code","source":"numbers_to_display = 12\nnum_cells = math.ceil(math.sqrt(numbers_to_display))\nplt.figure(figsize=(10,10))\nfor i in range(numbers_to_display):\n    plt.subplot(num_cells, num_cells, i+1)\n    plt.xticks([])\n    plt.yticks([])\n    plt.grid(False)\n    plt.imshow(x_train_re[i], cmap=plt.cm.binary)\n    plt.xlabel(y_train_re[i])\nplt.show()","metadata":{"papermill":{"duration":0.520651,"end_time":"2022-07-03T08:33:14.985777","exception":false,"start_time":"2022-07-03T08:33:14.465126","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:27:08.938091Z","iopub.execute_input":"2022-07-30T04:27:08.938440Z","iopub.status.idle":"2022-07-30T04:27:09.467374Z","shell.execute_reply.started":"2022-07-30T04:27:08.938407Z","shell.execute_reply":"2022-07-30T04:27:09.466127Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### 4.2 Reshaping train, test, and validation data <a class =\"anchor\" id = \"reshapeData\"></a>\n[Back to Table of Contents](#bcImp)\n\nIn order to use convolution layers we need to reshape our data and add a color channel to it. As you've noticed currently every digit has a shape of (28, 28) which means that it is a 28x28 matrix of color values form 0 to 255. We need to reshape it to (28, 28, 1) shape so that each pixel potentially may have multiple channels (like Red, Green and Blue).","metadata":{"papermill":{"duration":0.033167,"end_time":"2022-07-03T08:33:15.052259","exception":false,"start_time":"2022-07-03T08:33:15.019092","status":"completed"},"tags":[]}},{"cell_type":"code","source":"x_train_with_chanels = x_train_re.reshape(\n    x_train_re.shape[0],\n    IMAGE_WIDTH,\n    IMAGE_HEIGHT,\n    IMAGE_CHANNELS\n)\n\n# x_validation_with_chanels = x_validation_re.reshape(\n#     x_validation_re.shape[0],\n#     IMAGE_WIDTH,\n#     IMAGE_HEIGHT,\n#     IMAGE_CHANNELS\n# )\n\nx_test_with_chanels = x_test_re.reshape(\n    x_test_re.shape[0],\n    IMAGE_WIDTH,\n    IMAGE_HEIGHT,\n    IMAGE_CHANNELS\n)","metadata":{"papermill":{"duration":0.042322,"end_time":"2022-07-03T08:33:15.12869","exception":false,"start_time":"2022-07-03T08:33:15.086368","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:27:09.469420Z","iopub.execute_input":"2022-07-30T04:27:09.469789Z","iopub.status.idle":"2022-07-30T04:27:09.475344Z","shell.execute_reply.started":"2022-07-30T04:27:09.469757Z","shell.execute_reply":"2022-07-30T04:27:09.474525Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print('x_train_with_chanels:', x_train_with_chanels.shape)\n#print('x_validation_with_chanels:', x_validation_with_chanels.shape)\nprint('x_test_with_chanels:', x_test_with_chanels.shape)","metadata":{"papermill":{"duration":0.042714,"end_time":"2022-07-03T08:33:15.205239","exception":false,"start_time":"2022-07-03T08:33:15.162525","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:27:09.477292Z","iopub.execute_input":"2022-07-30T04:27:09.477875Z","iopub.status.idle":"2022-07-30T04:27:09.488646Z","shell.execute_reply.started":"2022-07-30T04:27:09.477835Z","shell.execute_reply":"2022-07-30T04:27:09.487582Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### 4.3 Normalize train, test, and validation data <a class =\"anchor\" id =\"normalizeData\"></a>\n[Back to Table of Contents](#bcImp)\n\nHere we're just trying to normalize from values range of [0...255] to [0...1].","metadata":{"papermill":{"duration":0.034217,"end_time":"2022-07-03T08:33:15.273318","exception":false,"start_time":"2022-07-03T08:33:15.239101","status":"completed"},"tags":[]}},{"cell_type":"code","source":"x_train_normalized = x_train_with_chanels / 255\n#x_validation_normalized = x_validation_with_chanels / 255\nx_test_normalized = x_test_with_chanels / 255","metadata":{"papermill":{"duration":0.219883,"end_time":"2022-07-03T08:33:15.527086","exception":false,"start_time":"2022-07-03T08:33:15.307203","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:27:09.489788Z","iopub.execute_input":"2022-07-30T04:27:09.490586Z","iopub.status.idle":"2022-07-30T04:27:10.053605Z","shell.execute_reply.started":"2022-07-30T04:27:09.490551Z","shell.execute_reply":"2022-07-30T04:27:10.051227Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Let's check just one row from the 0th image to see color chanel values after normalization.\nx_train_normalized[0][10]","metadata":{"papermill":{"duration":0.044654,"end_time":"2022-07-03T08:33:15.605992","exception":false,"start_time":"2022-07-03T08:33:15.561338","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:27:10.057531Z","iopub.execute_input":"2022-07-30T04:27:10.058467Z","iopub.status.idle":"2022-07-30T04:27:10.065395Z","shell.execute_reply.started":"2022-07-30T04:27:10.058430Z","shell.execute_reply":"2022-07-30T04:27:10.064526Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### In summary<a class =\"anchor\" id =\"summaryExploreData\"></a>\n[Back to Table of Contents](#bcImp)","metadata":{"papermill":{"duration":0.033343,"end_time":"2022-07-03T08:33:15.673138","exception":false,"start_time":"2022-07-03T08:33:15.639795","status":"completed"},"tags":[]}},{"cell_type":"markdown","source":"### 5. Build the CNN model to Classify Handwritten Digits <a class = \"anchor\" id = \"modeling\"></a>\n[Back to Table of Contents](#bcImp)\n\nWe are using Sequential Keras model which have two pairs of Convolution2D and MaxPooling2D layers. The MaxPooling layer acts as a sort of downsampling using max values in a region instead of averaging.\n\nAfter that we will use Flatten layer to convert multidimensional parameters to vector.\n\nThe last layer will be a Dense layer with 10 Softmax outputs. The output represents the network guess. The 0-th output represents a probability that the input digit is 0, the 1-st output represents a probability that the input digit is 1 and so on...","metadata":{"papermill":{"duration":0.033087,"end_time":"2022-07-03T08:33:15.739624","exception":false,"start_time":"2022-07-03T08:33:15.706537","status":"completed"},"tags":[]}},{"cell_type":"code","source":"model = tf.keras.models.Sequential()\n\nmodel.add(tf.keras.layers.Convolution2D(\n    input_shape=(IMAGE_WIDTH, IMAGE_HEIGHT, IMAGE_CHANNELS),\n    kernel_size=5,\n    filters=8,\n    strides=1,\n    activation=tf.keras.activations.relu,\n    kernel_initializer=tf.keras.initializers.VarianceScaling()\n))\n\nmodel.add(tf.keras.layers.MaxPooling2D(\n    pool_size=(2, 2),\n    strides=(2, 2)\n))\n\nmodel.add(tf.keras.layers.Convolution2D(\n    kernel_size=5,\n    filters=16,\n    strides=1,\n    activation=tf.keras.activations.relu,\n    kernel_initializer=tf.keras.initializers.VarianceScaling()\n))\n\nmodel.add(tf.keras.layers.MaxPooling2D(\n    pool_size=(2, 2),\n    strides=(2, 2)\n))\n\nmodel.add(tf.keras.layers.Flatten())\n\nmodel.add(tf.keras.layers.Dense(\n    units=128,\n    activation=tf.keras.activations.relu\n));\n\nmodel.add(tf.keras.layers.Dropout(0.2))\n\nmodel.add(tf.keras.layers.Dense(\n    units=10,\n    activation=tf.keras.activations.softmax,\n    kernel_initializer=tf.keras.initializers.VarianceScaling()\n))","metadata":{"papermill":{"duration":0.220378,"end_time":"2022-07-03T08:33:15.993677","exception":false,"start_time":"2022-07-03T08:33:15.773299","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:27:10.067023Z","iopub.execute_input":"2022-07-30T04:27:10.068013Z","iopub.status.idle":"2022-07-30T04:27:10.236882Z","shell.execute_reply.started":"2022-07-30T04:27:10.067979Z","shell.execute_reply":"2022-07-30T04:27:10.236043Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### 5.1 Summary of the training model <a class =\"anchor\" id = \"modelSummary\"></a>\n[Back to Table of Contents](#bcImp)\n\nHere is our model summary so far.","metadata":{"papermill":{"duration":0.033167,"end_time":"2022-07-03T08:33:16.060467","exception":false,"start_time":"2022-07-03T08:33:16.0273","status":"completed"},"tags":[]}},{"cell_type":"code","source":"model.summary()","metadata":{"papermill":{"duration":0.044739,"end_time":"2022-07-03T08:33:16.138914","exception":false,"start_time":"2022-07-03T08:33:16.094175","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:27:10.237894Z","iopub.execute_input":"2022-07-30T04:27:10.238911Z","iopub.status.idle":"2022-07-30T04:27:10.244809Z","shell.execute_reply.started":"2022-07-30T04:27:10.238875Z","shell.execute_reply":"2022-07-30T04:27:10.243704Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### 5.2 Visualization of the model using graphviz <a class =\"anchor\" id = \"modelplot\"></a>\n[Back to Table of Contents](#bcImp)","metadata":{"papermill":{"duration":0.033099,"end_time":"2022-07-03T08:33:16.205623","exception":false,"start_time":"2022-07-03T08:33:16.172524","status":"completed"},"tags":[]}},{"cell_type":"markdown","source":"In order to plot the model the graphviz should be installed.","metadata":{"papermill":{"duration":0.03349,"end_time":"2022-07-03T08:33:16.272436","exception":false,"start_time":"2022-07-03T08:33:16.238946","status":"completed"},"tags":[]}},{"cell_type":"code","source":"tf.keras.utils.plot_model(\n    model,\n    show_shapes=True,\n    show_layer_names=True,\n)","metadata":{"papermill":{"duration":1.220929,"end_time":"2022-07-03T08:33:17.527042","exception":false,"start_time":"2022-07-03T08:33:16.306113","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:27:10.246279Z","iopub.execute_input":"2022-07-30T04:27:10.246662Z","iopub.status.idle":"2022-07-30T04:27:11.449772Z","shell.execute_reply.started":"2022-07-30T04:27:10.246630Z","shell.execute_reply":"2022-07-30T04:27:11.448632Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### 5.3 Compile the model using keras.optimizers.Adam <a class =\"anchor\" id = \"compileModel\"></a>\n[Back to Table of Contents](#bcImp)","metadata":{"papermill":{"duration":0.035298,"end_time":"2022-07-03T08:33:17.598137","exception":false,"start_time":"2022-07-03T08:33:17.562839","status":"completed"},"tags":[]}},{"cell_type":"code","source":"adam_optimizer = tf.keras.optimizers.Adam(learning_rate=0.001)\n\nmodel.compile(\n    optimizer=adam_optimizer,\n    loss=tf.keras.losses.sparse_categorical_crossentropy,\n    metrics=['accuracy']\n)","metadata":{"papermill":{"duration":0.307189,"end_time":"2022-07-03T08:33:17.940807","exception":false,"start_time":"2022-07-03T08:33:17.633618","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:27:11.451018Z","iopub.execute_input":"2022-07-30T04:27:11.451362Z","iopub.status.idle":"2022-07-30T04:27:11.474506Z","shell.execute_reply.started":"2022-07-30T04:27:11.451329Z","shell.execute_reply":"2022-07-30T04:27:11.473580Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### 5.4 Train the model <a class =\"anchor\" id =\"trainModel\"></a>\n[Back to Table of Contents](#bcImp)\n\n","metadata":{"papermill":{"duration":0.035869,"end_time":"2022-07-03T08:33:18.013217","exception":false,"start_time":"2022-07-03T08:33:17.977348","status":"completed"},"tags":[]}},{"cell_type":"markdown","source":"__Data Augmentation__\nWe will use Data Augmentation to provide more data during the training process.","metadata":{}},{"cell_type":"code","source":"datagen = tf.keras.preprocessing.image.ImageDataGenerator(\n    rotation_range=20,\n    width_shift_range=0.20,\n    shear_range=15,\n    zoom_range=0.10,\n    validation_split=0.25,\n    horizontal_flip=False\n)\n\ntrain_generator = datagen.flow(\n    x_train_normalized,\n    y_train_re, \n    batch_size=256,\n    subset='training',\n)\n\nvalidation_generator = datagen.flow(\n    x_train_normalized,\n    y_train_re, \n    batch_size=64,\n    subset='validation',\n)","metadata":{"execution":{"iopub.status.busy":"2022-07-30T04:27:11.475680Z","iopub.execute_input":"2022-07-30T04:27:11.476457Z","iopub.status.idle":"2022-07-30T04:27:11.744368Z","shell.execute_reply.started":"2022-07-30T04:27:11.476421Z","shell.execute_reply":"2022-07-30T04:27:11.743213Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"reduce_lr = tf.keras.callbacks.ReduceLROnPlateau(monitor='val_loss',\n                                                 factor=0.1,\n                                                 patience=5,\n                                                 min_lr=0.000001,\n                                                 verbose=1)\n\ncheckpoint = tf.keras.callbacks.ModelCheckpoint(filepath='model.hdf5',\n                                                monitor='val_loss',\n                                                save_best_only=True,\n                                                save_weights_only=True,\n                                                verbose=1)","metadata":{"execution":{"iopub.status.busy":"2022-07-30T04:27:11.745872Z","iopub.execute_input":"2022-07-30T04:27:11.746233Z","iopub.status.idle":"2022-07-30T04:27:11.752303Z","shell.execute_reply.started":"2022-07-30T04:27:11.746202Z","shell.execute_reply":"2022-07-30T04:27:11.751516Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"training_history = model.fit_generator(train_generator, \n                              epochs=10, \n                              validation_data=validation_generator, \n                              callbacks=[reduce_lr,checkpoint], \n                              verbose=1)","metadata":{"execution":{"iopub.status.busy":"2022-07-30T04:27:11.753720Z","iopub.execute_input":"2022-07-30T04:27:11.754035Z","iopub.status.idle":"2022-07-30T04:37:54.928507Z","shell.execute_reply.started":"2022-07-30T04:27:11.754005Z","shell.execute_reply":"2022-07-30T04:37:54.927204Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### In summary<a class =\"anchor\" id =\"modelBuildSummary\"></a>\n[Back to Table of Contents](#bcImp)","metadata":{"papermill":{"duration":0.08198,"end_time":"2022-07-03T08:34:12.361298","exception":false,"start_time":"2022-07-03T08:34:12.279318","status":"completed"},"tags":[]}},{"cell_type":"markdown","source":"### 6. Model evaluation<a class =\"anchor\" id =\"modelEvaluation\"></a>\n[Back to Table of Contents](#bcImp1)","metadata":{"papermill":{"duration":0.081123,"end_time":"2022-07-03T08:34:12.523842","exception":false,"start_time":"2022-07-03T08:34:12.442719","status":"completed"},"tags":[]}},{"cell_type":"markdown","source":"### 6.1 Loss plot curve for training and validation<a class=\"anchor\" id =\"lossPlot\"><a/>\n[Back to Table of Contents](#bcImp1)\n    \nLet's see how the loss function was changing during the training. We expect it to get smaller and smaller on every next epoch.","metadata":{"papermill":{"duration":0.081442,"end_time":"2022-07-03T08:34:12.686427","exception":false,"start_time":"2022-07-03T08:34:12.604985","status":"completed"},"tags":[]}},{"cell_type":"code","source":"plt.xlabel('Epoch Number')\nplt.ylabel('Accuracy')\nplt.plot(training_history.history['loss'], label='training set')\nplt.plot(training_history.history['val_loss'], label='validation set')\nplt.legend()","metadata":{"papermill":{"duration":0.270657,"end_time":"2022-07-03T08:34:13.03887","exception":false,"start_time":"2022-07-03T08:34:12.768213","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:37:54.930499Z","iopub.execute_input":"2022-07-30T04:37:54.931259Z","iopub.status.idle":"2022-07-30T04:37:55.144396Z","shell.execute_reply.started":"2022-07-30T04:37:54.931207Z","shell.execute_reply":"2022-07-30T04:37:55.143559Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### 6.2. Accuracy plot curve for training and validation<a class=\"anchor\" id =\"accuracyPlot\"><a/>\n[Back to Table of Contents](#bcImp1)","metadata":{"papermill":{"duration":0.082599,"end_time":"2022-07-03T08:34:13.203783","exception":false,"start_time":"2022-07-03T08:34:13.121184","status":"completed"},"tags":[]}},{"cell_type":"code","source":"plt.xlabel('Epoch Number')\nplt.ylabel('Accuracy')\nplt.plot(training_history.history['accuracy'], label='training set')\nplt.plot(training_history.history['val_accuracy'], label='validation set')\nplt.legend()","metadata":{"papermill":{"duration":0.270982,"end_time":"2022-07-03T08:34:13.557302","exception":false,"start_time":"2022-07-03T08:34:13.28632","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:37:55.146025Z","iopub.execute_input":"2022-07-30T04:37:55.146946Z","iopub.status.idle":"2022-07-30T04:37:55.366316Z","shell.execute_reply.started":"2022-07-30T04:37:55.146896Z","shell.execute_reply":"2022-07-30T04:37:55.365050Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### 6.3. Evaluation of the model accuracy<a class =\"anchor\" id =\"accuracyEvaluation\"></a>\n[Back to Table of Contents](#bcImp1)\n\nWe need to compare the accuracy of our model on training set and on valiation set. We expect our model to perform similarly on both sets. If the performance on a validation set will be poor comparing to a training set it would be an indicator for us that the model is overfitted and we have a \"high variance\" issue.","metadata":{"papermill":{"duration":0.145343,"end_time":"2022-07-03T08:34:13.784852","exception":false,"start_time":"2022-07-03T08:34:13.639509","status":"completed"},"tags":[]}},{"cell_type":"markdown","source":"### 6.3.1 Performance of training dataset <a class =\"anchor\" id =\"perfTrain\"></a>\n[Back to Table of Contents](#bcImp1)","metadata":{"papermill":{"duration":0.083883,"end_time":"2022-07-03T08:34:13.950235","exception":false,"start_time":"2022-07-03T08:34:13.866352","status":"completed"},"tags":[]}},{"cell_type":"code","source":"%%capture\ntrain_loss, train_accuracy = model.evaluate(x_train_normalized, y_train_re)","metadata":{"papermill":{"duration":2.799582,"end_time":"2022-07-03T08:34:16.831834","exception":false,"start_time":"2022-07-03T08:34:14.032252","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:37:55.367400Z","iopub.execute_input":"2022-07-30T04:37:55.367749Z","iopub.status.idle":"2022-07-30T04:38:12.446179Z","shell.execute_reply.started":"2022-07-30T04:37:55.367718Z","shell.execute_reply":"2022-07-30T04:38:12.444928Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print('Train loss: ', train_loss)\nprint('Train accuracy: ', train_accuracy)","metadata":{"papermill":{"duration":0.091801,"end_time":"2022-07-03T08:34:17.006383","exception":false,"start_time":"2022-07-03T08:34:16.914582","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:38:12.448058Z","iopub.execute_input":"2022-07-30T04:38:12.448577Z","iopub.status.idle":"2022-07-30T04:38:12.456978Z","shell.execute_reply.started":"2022-07-30T04:38:12.448528Z","shell.execute_reply":"2022-07-30T04:38:12.455632Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### 6.3.2. Save and load the model<a class =\"anchor\" id =\"saveModel\"></a>\n[Back to Table of Contents](#bcImp1)\n\nWe will save the entire model to a HDF5 file. The .h5 extension of the file indicates that the model shuold be saved in Keras format as HDF5 file.","metadata":{"papermill":{"duration":0.081283,"end_time":"2022-07-03T08:34:18.432473","exception":false,"start_time":"2022-07-03T08:34:18.35119","status":"completed"},"tags":[]}},{"cell_type":"code","source":"model_name = 'digits_recognition_cnn.h5'\nmodel.save(model_name, save_format='h5')","metadata":{"papermill":{"duration":0.121588,"end_time":"2022-07-03T08:34:18.635926","exception":false,"start_time":"2022-07-03T08:34:18.514338","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:46:41.918040Z","iopub.execute_input":"2022-07-30T04:46:41.918528Z","iopub.status.idle":"2022-07-30T04:46:41.953930Z","shell.execute_reply.started":"2022-07-30T04:46:41.918491Z","shell.execute_reply":"2022-07-30T04:46:41.952700Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"loaded_model = tf.keras.models.load_model(model_name)","metadata":{"papermill":{"duration":0.178221,"end_time":"2022-07-03T08:34:18.896278","exception":false,"start_time":"2022-07-03T08:34:18.718057","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:46:15.901266Z","iopub.execute_input":"2022-07-30T04:46:15.901905Z","iopub.status.idle":"2022-07-30T04:46:15.999240Z","shell.execute_reply.started":"2022-07-30T04:46:15.901858Z","shell.execute_reply":"2022-07-30T04:46:15.997946Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### 7. Model prediction on unseen dataset (test data)<a class=\"anchor\" id =\"modelprediction\"></a>\n[Back to Table of Contents](#bcImp1)","metadata":{"papermill":{"duration":0.087233,"end_time":"2022-07-03T08:34:29.887905","exception":false,"start_time":"2022-07-03T08:34:29.800672","status":"completed"},"tags":[]}},{"cell_type":"code","source":"predictions_one_hot = loaded_model.predict([x_test_normalized])\nprint('predictions_one_hot:', predictions_one_hot.shape)","metadata":{"papermill":{"duration":2.445265,"end_time":"2022-07-03T08:34:32.474146","exception":false,"start_time":"2022-07-03T08:34:30.028881","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:46:12.274306Z","iopub.execute_input":"2022-07-30T04:46:12.275396Z","iopub.status.idle":"2022-07-30T04:46:14.787931Z","shell.execute_reply.started":"2022-07-30T04:46:12.275339Z","shell.execute_reply":"2022-07-30T04:46:14.786674Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Each prediction has ten probabilities (one for each number from 0 to 9). We need to choose the digit with the highest probability.","metadata":{"papermill":{"duration":0.088836,"end_time":"2022-07-03T08:34:32.652401","exception":false,"start_time":"2022-07-03T08:34:32.563565","status":"completed"},"tags":[]}},{"cell_type":"markdown","source":"#### Predicted probabilities of all digits.","metadata":{"papermill":{"duration":0.088922,"end_time":"2022-07-03T08:34:32.830662","exception":false,"start_time":"2022-07-03T08:34:32.74174","status":"completed"},"tags":[]}},{"cell_type":"code","source":"pd.DataFrame(predictions_one_hot)","metadata":{"papermill":{"duration":0.115571,"end_time":"2022-07-03T08:34:33.034968","exception":false,"start_time":"2022-07-03T08:34:32.919397","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:46:24.318979Z","iopub.execute_input":"2022-07-30T04:46:24.319464Z","iopub.status.idle":"2022-07-30T04:46:24.351774Z","shell.execute_reply.started":"2022-07-30T04:46:24.319422Z","shell.execute_reply":"2022-07-30T04:46:24.350464Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### 7.1 Visualise test predicted data how the digits were written<a class=\"anchor\" id=\"visualizePredict\"></a>\n[Back to Table of Contents](#bcImp1)\n\nPredicted digits with highest probabilites","metadata":{"papermill":{"duration":0.087654,"end_time":"2022-07-03T08:34:33.2101","exception":false,"start_time":"2022-07-03T08:34:33.122446","status":"completed"},"tags":[]}},{"cell_type":"markdown","source":"#### Actual first digit from the test data","metadata":{"papermill":{"duration":0.087314,"end_time":"2022-07-03T08:34:33.384917","exception":false,"start_time":"2022-07-03T08:34:33.297603","status":"completed"},"tags":[]}},{"cell_type":"code","source":"plt.imshow(x_test_normalized[0].reshape((IMAGE_WIDTH, IMAGE_HEIGHT)), cmap=plt.cm.binary)\nplt.show()","metadata":{"papermill":{"duration":0.256898,"end_time":"2022-07-03T08:34:33.729006","exception":false,"start_time":"2022-07-03T08:34:33.472108","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:46:30.656124Z","iopub.execute_input":"2022-07-30T04:46:30.656567Z","iopub.status.idle":"2022-07-30T04:46:30.856500Z","shell.execute_reply.started":"2022-07-30T04:46:30.656528Z","shell.execute_reply":"2022-07-30T04:46:30.854576Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"So our model is predicting that the first example from the test data is 2.","metadata":{"papermill":{"duration":0.087946,"end_time":"2022-07-03T08:34:33.905136","exception":false,"start_time":"2022-07-03T08:34:33.81719","status":"completed"},"tags":[]}},{"cell_type":"markdown","source":"### 8. Submission <a class =\"anchor\" id =\"submission\"></a>\n[Back to Table of Contents](#bcImp1)","metadata":{"papermill":{"duration":0.089025,"end_time":"2022-07-03T08:34:34.082497","exception":false,"start_time":"2022-07-03T08:34:33.993472","status":"completed"},"tags":[]}},{"cell_type":"code","source":"# test_pred = pd.DataFrame( model.predict([x_test_normalized]))\n# test_pred = pd.DataFrame(test_pred.idxmax(axis = 1))\n# test_pred.index.name = 'ImageId'\n# test_pred = test_pred.rename(columns = {0: 'Label'}).reset_index()\n# test_pred['ImageId'] = test_pred['ImageId'] + 1\n\n# test_pred.head()","metadata":{"papermill":{"duration":1.907704,"end_time":"2022-07-03T08:34:36.079151","exception":false,"start_time":"2022-07-03T08:34:34.171447","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:38:14.824234Z","iopub.execute_input":"2022-07-30T04:38:14.824623Z","iopub.status.idle":"2022-07-30T04:38:14.831634Z","shell.execute_reply.started":"2022-07-30T04:38:14.824590Z","shell.execute_reply":"2022-07-30T04:38:14.830166Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# test_pred.to_csv('submission.csv', index = False)","metadata":{"papermill":{"duration":0.12483,"end_time":"2022-07-03T08:34:36.291311","exception":false,"start_time":"2022-07-03T08:34:36.166481","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2022-07-30T04:38:14.833728Z","iopub.execute_input":"2022-07-30T04:38:14.834187Z","iopub.status.idle":"2022-07-30T04:38:14.844988Z","shell.execute_reply.started":"2022-07-30T04:38:14.834141Z","shell.execute_reply":"2022-07-30T04:38:14.843867Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"__Many thanks for reading my kernal!🙏__\n\n__Please leave a comment if you have any suggestions for improving the analysis!🏋🥇__\n\n__If you liked my kernel, give 👍 UPVOTE!__\n\n__If you have a moment, I encourage you to see at my other [kernels](https://www.kaggle.com/itsmohammadshahid/code?scroll=true).__","metadata":{}}]}