{"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":"![image.png](attachment:9ac866bb-13ac-4bf9-a839-f2b9139f82d6.png)","metadata":{"papermill":{"duration":0.04325,"end_time":"2021-07-20T23:07:40.385155","exception":false,"start_time":"2021-07-20T23:07:40.341905","status":"completed"},"tags":[]},"attachments":{"9ac866bb-13ac-4bf9-a839-f2b9139f82d6.png":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAA8UAAADeCAYAAAAUy6KpAAAgAElEQVR4Aey993cbV5Yt/P6e92bmzfS43ePudrfbloMsyZYVLCtYomTlQFGkKJIScyYIgiABECAAIhdQKBRQANUzvWZN6J6cc449Oacf3/fT/tY+hUuUaCqDlCVdrFWr8r23Tp17cXftE/4X9E9LQEtAS0BLQEtAS0BLQEtAS0BLQEtAS0BL4CWVwP96SZ9bP7aWgJaAloCWgJaAloCWgJaAloCWgJaAloCWADQo1kqgJaAloCWgJaAloCWgJaAloCWgJaAloCXw0kpAg+KX9tXrB9cS0BLQEtAS0BLQEtAS0BLQEtAS0BLQEtCgWOuAloCWgJaAloCWgJaAloCWgJaAloCWgJbASyuBLUHx//t//x/+9V//Dd//m7/FX3//b/SiZaB1QOuA1gGtA1oHtA5oHdA6oHVA64DWAa0DX1gdIHYlhiWWfdzf50CxKuT73/8+vve972F9fR13796VNbf1vpaH1gfdH9RYoMcDPR7o8UCPBzs1HjQ4B2mswzYtGMks8itrSMwuITLsw8qwD+E7M1idDCA+tYjkXAiJmSVkAlGkAivIBKPILsdQWEmimMzCypdgly04FRuOXYVTrcnSqNfRaDQ2Fj4b99UzbtZ3dV7ds373rnsv76s3UCtXUIynoMqV69fXkQ1E4Dt5FanpAKqmJfc4lo1I9xBmDn2G5IRP2sPr2ca1ST/mjp5HIRxzr63YMCJxt9z1BhqOg3I8hbpdddvLZ1hv4O76XVmL7PR8Ts9n9Xx+oy+zb23uzzu1L+NFrYqGZcExDDi5LJxUCvW1JGqxKKrLS7D9Ptiz07BnZ2BNT8GamYI9O4UK174Z2AE/qosBVGZnUJmaRH7ve8h+5UdgnDoOe2YKtXAI9UIe1WgYxrHDyHzp/yD7lR+GceQAaqtRFM+cQuZLP4D8/j1olEqo9N9C7itfQu6N11BdXkQ9n0e58zKKJ4+iGl6SscY73j2x/BoulqQM6olVmOc6UDywB9a1i6jHVtCwTNhjwyh881UUXvkBlD54B05kGfbEKIzXvwxnJSzjXdU3jdL7b6G24JP99eY4/d3vfhfEsPwpTPuo4PhzoFgY4u9//x6l4YPrRctA64DWAa0DWge0DmgdeFY6QGBXJ/jLGshHU8gur2J1IrABiiODs4iO+OQYATEXguM1fxjp4AoyS1Fkl6LIR5IwU3nYRRM1y0bVqqDWBMf1miNgVoFc73qr577nvALTBMRNMF0xSljpG0d2IYyaVXGPr6/DLpnwn76O8b0nkRj3oWa5wNiIJuDv6ETgsy6UEikXYDcaKK9lsXShB0tXbskEkEA5emsE8eGpJuCuoxRLolosbbRfnsVpgnw9j9PzWK0DXygd4IesRsVCvVSEk8/DyaThJBOoxaOoRkICei2C3/k5VOZmBCRX5n2ythbmYfv9qEYi7vnpSRinP0X2668i+9orKJ47jWowiFokjHJvN4zDHyG3excyP/qDKBzYK8dL504j/SP/G/k9u9Aom7D6bgqoLh7/GE4yLoC9TsCeSct2o+GOJVuNg496jB/qCF7dD5wN1LNZ1JYXN+rjR4LqzBSMN7+O/Gs/AvPkUTipJBp2Bc7yIooH98KJRtCo2bBuXIP5yQE4iRgaVXsDGHNMZnsIjIlpH+f3OVBM2lkzxO6kZ6e+GCll0vVpxkXpAtdaH7Q+aH1oAVDdH3R/4GSHrK6RyAqwXfOFEB3zY2VoDivDc8IUh2/PIDI0i9iYH7FxP+KTAWGTyRanAhFZCIyN2BqsbAG2WUbVslAnU9xkjBsKSHLyVicLW//cZFrpowLF9aoDI55CIZLYuJ7nCEzJCk8e+AzBS31Sp5q0sQ1zRy/Bd+wSiisJ1J26sMNkieePXUTs9oS0iRPIarGMld4RzB05ByuTlzpygRBmDpxGJWdI+1ITPoSu9MLK5mXiyeerNVlo1V41puh93Z+ULuj5xrOZb22A4mIR9UIBTi4HJ52Ck4jDia3AXg7CDgZgBxZQXVyUNdlhe3EB1YV5AbvW2DCq4WVUA35UpiZgnDyKzNe+jMxXv4zsu28i+96bwhLb05Mwu64i+8r/FRBcDS3B6utF+kv/B8aBfVhn/fEYjMMfonTyGOq5DBpODQ2n6oJNWqM8wGLmcccTsahhmXVH6lkn4HZqcEKLMDuOw7r4GezJUdTzWWlX5cJp5L/5FRjfeBXWlfNwVkIoffAe7L6bqJcMNPJZYZjVeMz2EMsS0z7O73OgmD7E3o6it1uTMi0LLQutA1oHtA5oHdA68Gx0gBMeu2QJIE4HY4jPLCE66kdoYBrL/ZNY7ptAoGsYof4prIz4ELkzg9ioH4mZIFL+CNKLK0gthJHyh5ANRmEk0qgUisLgCiAmKCZT7NQFoMoEi+bUBMaeCSHfv5p8qTXNr1NzQQQu9SE1H4KZym5cQ/PoaP8ExnafQPBKH6pllxUm650LRjBz9AJCXXdcsEtWOJ1D8FIv5k9ekfMOmWzTQnxwClP7PkVuMSJll1NZTO07gYxvUSauRigK35GzKCxHBYwb4RhK0biYX2udfTY6q+Wu5X4/HZCPb5WKyxQTlJItztKMOgknvoraShh2MIhqaNkFx4sB1JaWUIuEUAsuotR5Bdndb8M4dQzW5ChsP9ljn4Dd/IF9yLz2CjKv/giMT4/CJggeHED2x76E3K5vwl6YF/Pq3JtfR+7N1wWEOwSgsajL2lZtCFBtssPu+NdyI7nfMz3KcTVmijVNE2y7xxw0zBKcTApkqOtFA9X5WRT37kL+K/8XhR/7YTGxtsdGUNy/W0CxsxoVltgeGYRsbwLvxLSP89sSFD8u4tfX6y+O3o6g9UHrg9aH1kRA9wfdH3R/aE9/UKbTmeAqkvNhRCcWEBuZx1LfJJZ6x7HUO4Fg9xhCA1PiY0xQvDq+IGbUBMXCFC+EsTZHX+MICrE1lMkWl8r3mE+zHpmw1mnqR9/ge4Gx64dcEwBNdleB5nImj3DPKCYPnkPgQi9KyYyQDJzw2YaJSM8Ixt4/gcTonABVHidADl65hckPO+A/3YnYwDjWpvxY7hzA9KEz8H1yHstXbgkgXrrci8m9J5Ca8kv7WN/0vk8RuXFHGGYC8flPzoMMMpnv/MIyot13UIwmXL/iJrAXE8amKa0en/T4pMen9oxPlOPj9CeOK+vVKuqm6foVFwpoEBzTjDq1BiexCicchhOJuObUBMThZdTiMdSiEWGGC3vfRfprX0bug92wRodRTyZQGR8Vv+Hs66+JD3Hh0H7YwQXxM86/9Q1kv/kaKuPDqBt5VPpvorDvPRAQ180i1isWGtVq0yVDgWC1frzne1x5rEsMhOYHx4qFqn8G5tGDMN7+Bsxjh2DPTKJRNOBkMyh+uFvMpxtmURj28sXP4CwFmqRuq71tAcXeDqK3W51Fy0LLQuuA1gGtA1oHtA7svA4QQJKNNVM5ZJfiiM8EhSWODM0JKA71TWHp1gSCPWNYvjWByOCMLCuDc+JjTN/ilJ8scRgp37IwxvkV+hbnYOWLAk4d25Y6yBYLMG4yxNze8DWuN8RMOjY8g5RvCaVk2mV+62yfg0IkjvkzXRh9/yQWL/fByhsbjLGVMzBz9BKmDnyGpc4BhHqGsTYdkKBa9BWeOnBaALKxEoe5lsXqnQnMHDyNhTOdSM8FER+ZEfNp39ELSI7OIDE2g9E3DiBw9rqw3TSX9n38GVIzC+Cz5JciCJ69jqzfDZLjsuG2AGStwzuvw1rmWuZeHRB2lIHxrLLLFpMdzedRz2ZcM2rFGMfjwoJWwyFUmj7Gtfgq7EU/zBudyL23C5nXfhT53btgHD+C/N53UThyQEyps2+8htw7b8AaviMMM89nv/aqmF43SgbqmTSqwYD4NtO/uWHbaDAAWDPgoPj/brKS8T7Ddmy7cqmI73At4IcTXpaPBGSOaVJN82qaUZtHD4ic7Dt9KJ8+DicSEt9jmmGrdrUVFD/OFw82QF+vvzgqRdT6oPuDHg/0eKDHg9YkUPeHp+sP1CXxJ45nJJp0rBlgK3R7Bst9Uwj3TwtbTPNpLuGBaVkIjmlinZhewtp8SNjidCCCbDCGfCSBYiIjfr5Vs9yMRO0CY5cpbkWidn2ObQHHDJ4V7h3FNP2BO64jMeVHtVSG4zioEYyGVrFwthtj+04h0jeGQjQJI5JAZiEk5tOMOj2+91MsXuqVyNR2oSj+whMfnkJ0YFyYa/oJh7puY+pAh4Bjti8fioGAePrDU0gMT0uQrcn3j4oJdWpyHqmZAGb2n0R6LiBMsRGJIXiuC1n/suwXo3GUokmZ8Gp9fDp91PLT8nvq/zfGLKC7RsXCerncZIvzqOdyqDejUTvJJJy1ZNOkOo7K9ATyh/ejfPWyBNiy532w7gygePpTAcUEvKkv/xCM4x+jfKtbIktnv/YKytevwsnnJHK18dFeFE8dk8jUjXIZddZPdpjBqpqgWEyn6e/bNJ/eUX0nCGe9bAvbZVdQj0ZgXTkHY9fXYHz9FZgH98LqvgZ7eAClj95H5eYN8T0W/2LH2cCjbQXF3heut1uTGy0LLQutA1oHtA5oHdA6sHM6IKbGli1Rp9doOj3ul8Bay32TWO6fAtfB7lEBxIvdo2JKHeqbRLh/CtFhRqReQHJuCamFCDKLjEIdQy4ch7GakmjWNG+ulS0JuiXm0dXaBluiWGpGq66WK3CqVQG+syc7MfTOMUweOo/onSkxxZZgWRUb6fllTOw/jfH9HZg9flnMqRlwqxRPIdR5G8O7jiA6MCF10l948eJNjO05LmwwTapTswFhjuc+OY98KArbKCHaN4qJPcex+FmXBNMy4yn4jp5v+hUHkZ1fwvzRC1jpviPns75FYYrzi4x8bSExOInYrWFUS6aYhWv93Tn91bLWsv6cDkggP8cFo5YJh1GojQKcQt5ljPM5OJmMpGqSAFzZLOzFAPL73kf2m1+F0fGp+BBXl4ISqVqiT3/zNaRf+SHk97yDcl8vGGGa0aiNY4wonRAWujI+gvLALTcwFaPeC/gkQ9wEoWSKGWTLA4o5BtK8+XPPsC0RzVsp8QjU6T/NdE1mxzGUjh+EPdArKZpq/jlhi60rF1wmfW7ajUxddzba2RZQvKNfBDTDvPFFQymblr/+Aql0gWutD1oftD60JlS6P7yc/YGTMuYnLkTXsOaPIDq+ALLEDKpFQEyfYoJhAuONpWcMoVuTiAxMu2zxTBDpwIoE2WLO4mxoVcozMwXx+WVALIJe1sOcwa7JdB1mJo/CSkIAtLmWk/O8lsyv/7MeDO8+ifEPzmC5awhpmmfPBpGc9CM6MImlq/0YfvsTAcYE3gTc86c6BRSvzbiMrplIY+7oRUzsO4m0b0kA7XJnP8bfP4rwDRfgZnxBTH/Ugen9HcjOB5uBtyYxufc4li71CmguhGLwH70A38fnkJkLINzZj6XzN0CGmKA42juM8JVbqOQKEpiLZuHrd9m3WpNd3b9ezv6l/mP0+9/Z98++J2CUJtSmKfmCmbdYAk0ZBTcqdTbj+hGLOXUMZvd1ZN9+A5nXX0PxxFFYo0Mwe7qEFS4e+xi5d99C5muvonTxHCrM93tgH/LvvQlreBCNUlEAt5NNCxjfYIiFKa5ivRkN2o08TXDpAagN5j13/4tdkLz989P1et1lyZeCEnisHo+hwQ8HiVVUujtRev9N1ObnJGK31XUFztKCRN5X+twWUKwK0+vWREzLQstC64DWAa0DWge0DjwbHRBQXCojG05IkC3JTzw4J6CYAbYWb4wIGOZ64fqQMMbBm+Ogr3F4YEryFycFFEckv3E2FBemmCDbTDdzFpMpLluSv9hljiuSx5hscuDKAIKdg4gMTMGIJiEAmiA9kkDw+iBG93VgZPenmDp0Ab6TnUj5grCyBhITfgztOgLfyWvir1xO5zHxQQeG3/kE2cUImMqJgHdq/2nJT8yyae7sO3YR/lNXkV+OCru8eK4LUx+eQmxgQvyU1yZ9mNz3Kab3n4IRWUUlbyB0rQ9Tu4+KybQRiSM+OCnAmIG2yA4TFEc6+2EXDNAc2zaKbgAuDyjW+v1s9FvL/eWUO8c18eGlP2/ZRMM0UWeKpqIbfZnbjEhdmZlG6frVjSjUZIELnxxCdtcbyH/4PrJvfg2Zr78Ks6sTxbOnBRTn9+9BZXJcok7nP9gN89J5MDCV6ztcQZ0M8YaJMs2nXYZ4nR/LJNhgDY2Ga0JN32LvIu3eIV9jVZe0pVQEfYjN44dgvP26mE5XpydgXb8sbLLkV6ZMm21rKyjWX4x29ouRlreWt/ePUeuD1getD62Jku4PL3d/YBRoyzCRXY4j6QshPhWUCNP0HWZwrUATCC92DcPfOSggmebTEeYtvj0toDg+FRDzaWGIIwlJ7WTEUiitZYVprVm2+C3TjJpBsRjduWKYsApFRPomMHHwPMYOnBNwXAivCrtM8Lw2F8T86S6MfXAaQ7uOYvLgWeSWYwKoY0PTGHzzY/hPdwnzTBPqkbePCjAurq5JBGuma5o/cVmiThPcRvvHEDhzXcCyXSiJr/BCx1XEh6fF/9mIJuA/cQlT+08iOT4HpmyiafTE7k/gO3xGok/TTzkxPIVAx1UkhqZQWAwjdLkXq/2jcCwL5UQK+UDInTySAbr7cuuXfn79/p/F/60LimtYtyugfy9BMdlcYXS5zcjUhTxs3yxye96F0XEcldkZSZ3EFExkiLO7dyH71S8j/eoPo3TpHKzb/Sh88J6A5NLVS2CALqYxqkyMucDbog9xExA3QfF6rYYGF8lNXMNdBtqSbYJix2WIPcCYjLECq5TbdvYfAbhkjBOrqPpmUD7xsfgUm6ePwx4bRvniGZQ/+QjVmUnXT9q22ms+7VUMvd2alGlZaFloHdA6oHVA64DWgZ3XAaY9UqA4MReSiNL0FWaO4sWeUfiv3BYwLIG2rg8LUCZDvBGFemgO8YkAUr6Q+BMzyFY+uobCagpGPC3+wFWrIgGpCHTNVB655VUUk2kwsBavWe4ZxcThCxjZ14H5sz1YHZkTX+Kl63dkHemfwPgHp2WhCTXBNQNtERQHzveIyXJyYh5Dbx6WQFyVfFHY41DXIMLdQ5KfmObVaV9QFrLRNauCXCAswJjl0a+ZvsVLl25ibWIeTqUC+hYzSvXcwTNITfhgpXNIMlL1IR7rEIZ4pWsAgZOXkZr2o25XJV3Tau8w6sIOceLbMqHW+r3z+q1l/vLKXBhaqyw+vjShrpdKLigum6hzoa9xag3Fi+eQ/dbXkD96WEyhq83cw4Wjh2Ec/RjZ138M+X3vSWomkzmMee2HuyW4Vp2m2IYhka4Z7bph05e40mKKmyyxAGMPOFYgmf7FDH7lZYtle1t8iu/VBaaQazgOnNUV1EJB0Je4tuh3o1MHF2B+/CEqAz1w0knUQotibq1cQtrCFG8n4mfH1+XrL3LePwCtD1oftD60/gR0f9D9QfeHz/cHguKKUUYunER8ZklA8cqwTyJM00x6oXPQNZnuHm2yxBPCEq8Mzgowjo7OIz4ZwNrcMjKLKwKICXQZfZrm0aVkVsySaT7N/MWx0XlE+iYRG5pFPpoQv2IzmcXabBC+z7oxsrcDo3tPYeS9E5g5dgW55SjykThmPrmM0T0nER+dExPryC0XFId7hgV4rk0uYHjXJ4gPzwiTTJ+5zPyymEnTj5nvXlJDMVWLMDNu1G3HpiljQ8yekxM+FKNJ1+fZcZAcnsbypV6kZwKoFkpYG5uTKNRTe44h2jMkPsXx/lEsnLiIzOyCsOLJkWlErvSiXqmI/7QCxXr80eOPHn8+P/4omWxH/yAQrROolk2JQl03S+DCfYJiCbKVSqK6EoF59TKye99F/sM9Emgr9+6bKHacgDU+gvzBD5F9/SvCFpNZLl2+gMJHe8QHuV7Iu0w0zbRZl6RfsnBXGGOaTrss8XqTKVYsMcen1jYZ43vNqLdDHkrWXLN8AcViEu36ODPCdM0/A6vnOsrHD0ouYycURHXBB/vOLWGU2wqKvQ3S263OoWWhZaF1QOuA1gGtA1oHdl4HXFBsIrPMHMVLiE8tIjrmR7gZbMsbXCvYMwoxnb7TzFU8PCfX0nw6OctgWxEBxTSbZp5igmNJm9RkjIvJLILXhzB17CrGPjonDHF2KSqgmSxyfNKPmU+vYXj3Cdx+4zAmDpwTH2L6A8+duIbh904g0j8uoHilbxzD7x5DanZRQK2xkkTg/E1wLYG8ONlz6uDzPVCvGi1zReZrJojlQgBNn2SaXTNwFv2LF05eFtaYgNhMpmAmUoj2DIppNSNT54NhhK/0InrjNqpFE1Yqq6NR7wDj9cD3q+t/sP6/wPKRvkwzZjLExZLrT0zfX8uSbTvgh9nbjdpKRAAyAbBx8jgy3/waUj/6QzA+PSom0kzBlH3rdQnCZQ0OoBoJCVi2fTOQVEX0WVaAuAmK18WvuCp+zQTEdWUyLQG3HGFo3eBb7rYbkfre4Fvqg9p26bdbPuusS6oqe3QQxT1vovD6KzDeewPW1fMSkbp85gTs8WH340JTX9rCFKsH24kvAKournV9+gul1ofWZFP3B90fdH/Q/UHpwMs+Hiif4kwoLoG2ErMetnhwBsu3JhHsHZdo08u9ZImnXYZ42IfYmB+JyQASM0Gs+UICigsrSZRSOQG6DLRlxNYkzZIRT8mxcrqAYNew+BAPv38Si5f7kfKHkJgKINg1hNjQDPznezH4znGJPk1GmWX4P+vG8HvHsdI/LqbPjDA9c+wSKtmCgFgC10qhuJHuic/1KPOf9abPr0ygmxM+2RZwXMf6+l0B2cmxWcT6RyVCNf2KS6tJhC71YGbvCfg/OQcyxrHeISyfu46sf0lMr1Pjc1Dla317tPfxsvdH/fztm5802J8JRhkZuuBGnBa/YrLE4k88h8IHe1A8c9JNwRQOoXy7H7l9u5H66ivI7voGKpNjqIVDKF44i+w3X4Nx5IAE5VKMsPgsM/2SAsPVSnPb9S2WwFqSiongtwZhiMWM2pHUTDSVdnMWuybUm32K1bih1tQPL1h+Un1RY5zUV3dQC/phdXeifPwwyuc7UJ2ZQHViGOUj+2FduwQnHhWTcFVfW0GxeridXH/nO9/Bb/3Wb+Gv/uqv8K//+q/47//+b1m4zWM8x2t2sk26rtbEVMtCy0LrgNYBrQNaB3ZaB8ik2iVL8hSnA1Ek55axOhUAzaIjTd/i0MAUZOl3I05HhmYRHfUhNuGHsMTMU+wPIxNYQT6ckABbBKiMEp0NRpGYWsCabxkExmY2LybVyZlFTB+/gqH3Tm4AZEahZhTptZlFTHx0FoO7jiLUPYziagrBqwNiPh0dnBJ/YSuTR3xiXlhjykyBWpkIc/8RIz+ryeFGGZykqvLIIq+vC1PM4GAMpEVf4XIqg8j1fsweOIXA6SsSWKu4Ekf48k3Qx7iSLwh4pm/xTr9PXZ8eQ7QOuDrg9uO6MJxOLgtZjAIaRQO1WFT8h0viI/y6AGHjwllk3nxdUjDljx5C+uuvSiRq5jAmO1w8d1rMq63B227QLsUOExQrYMxjXOhX3EzHpKJPy5qAWB1nJOpGXRavX7Eak7zr7Xin3vIlQjfNygt5ONm05Ce2blxBpfua62McXoaTSW+MZ20BxQphq4fbqf3f//3fFwD8Z3/2Z/jlX/5l/ORP/qSwx9/+9rdlm8f+9E//VK7htTvdPl2f24F3Sh+0vLW8qQNa39r3RVrLU+vT89qfODFiDuFiMofUYgxkiuPTQaxOLCA2viDRpcO3p8GFwDg8OIOVEbLE83IN0zGRJc4sRsEcxQy0xXRMpbWM7Psv9sF3rheLXUPILNK8OingmMGu6CvsO30Dd94+htvf+gS+s91g5Oh8KIbZE1dx580jCFzpF5CdGPNJ9GkG2HJqDE5zr6khGdnWJK8JjJps8cP6J+/j+9sAwwTUUp4LrnmcZfA6pltipOnw1V7EByfEjNoumcj5l7B46gqWL3SjsBTBSlc/1iZmhLkmi/286gef+2Hy0+f1+PdF1O+N/mxXUM/lJKiWAONCHtXgIvKHP0K57ybM3hsoHNyP9FdfQeqHfxCF45+gGvDDOHIQ2a//GIqfdcCJrcBZjcGemkB1OYh1Ab9NhphRpyW4FiNdu1GuZV9Foq7auEsAvBFoy3EZY/oV112mWPIUN6NQc5xR441ae+WrxiueU4v3PI89aF/d3/C4jkg9hTxqSwFYnZckJVNxzy4xm67OTKE6MYJ6ts2gWDV+p9Y//dM/jb//+78XwEsg/LB6eQ3BMe/hvQ+7Xp9vKaSWhZbFy6ED7kTUDbbAgfsh/nqeQfvlkI/uB/o9P186wAlS1bJhJLJIB1fFhJqpmVanFhFrAmMG3mK06TB9iYdmZVkZnhPQTB9kAcXBqOQnzkeSyEXiAni5TXPrO+9+itvvnoDv3E2szYfEz9jKFlDOGmBk6clDFzD41lEM7+kAGWQG6PKf6xZQ7L/QCzOdQzGeQqRvHKvDM8IUq4njTusbfYzNtbTkXW7U68JUMyK170AH5vafxEpnP6LdtxE4eRHMY0wQzcjWbK8C1zvdZl3f89Un9ftq3/sSYFyrwcnn4CQTbnCtRBwVvx9mdxfy+/fBXlyAvRRE7sP3kf6xH5UcxQyoZS/4UTh+RKJNl851QHL1OgyepdIuMcq0G21afIgrNiTKdbEZjZpm02I6XRUzbvEhFp/iZqwDjgkCilsf+DY+9vE6YZE/P8eSsaT5UfDpdKVVL+uqTo7C/Hg/jF3MU7wHlb4eWFcvgD7FTjAgH/hUfW1hilVhD0LwvKYd5wlqaRr9m7/5m49dHk2pea8Cxu1oj3r2dj2fLq81aOj3oxm/newP3roIirX+af3z6oTWh+dLH/jumEfYXMsjG4ojvRhDimbUzFk8E0RieknAMYNvERATDIvp9Lhfok4nZpaQmm+lYxIf4nDJbKgAACAASURBVJUkmLOYptNrviXMftaDwfdPYmDXMYx8+BlCveMui3ypT1IwxSf8mCAwfvuYpGAqJdNY6hrE0DtH4T/Xg2I8LXmOnWpVcgcTjCpQvNP6RnmpSaljV4UhXjh+Eb7DnyE5PIViJCam1bFbw6iWTGR9iyjFkhsstLevcHun26/re776p35fT/u+1iXlkURWXo0JMK74fTDOnJIUS4WPDyD77pvIHzmA1I/+IAzu73oD+SMHYYeCqK9GUbxyAflD+8HAWus0fZYgWk2fYbXfBMrryqeYPsY1G24wrZr4DzMvsYBeAuFmEEDFEKsxRcaHJmPMa+W8h1xorz60GGmpv2ajQfPpSBhOKoXawjwqF86gOj+LumVKcDBVf1tB8eZBcTv2yfYSED9p2byXZTzp/fq+FmjVstCyeJF0QA3e3vWL9Hz6WXR/fZl0gP24VqlK/uBsOIlcKIFMMCaM8Zo/4oLjKZpTBzbA8er4guQmjk8uIjm7jNRCBJmlKPIryY1UTPlQHPnluOtnHElIIK3pTzsx+PZx3Hn3OIZ2n8TAt45guXtUzKnFjPrNoyAzTN9j5ioe239GQDFTNtEEWYFhNfY8i/ckda83UK87MEIryMwswAjHUMkVUDVNZOcCCJ7plLzGjFAduXxTfJB5H9ur2WI9vjwLvX0562Sfc1MdEcgS6NnBRVijIyheu4QCga7fB/PKRaS/+RpSX/oBFI4chNnVicy3vo7C4QNgzmKyv042LRGrFfNLv2Bhh8U/2BYALL7CBMwEyASYXDO4FscutZZUTEzH1GKLN9jhTS4h2z/ONZli9aHPMmEPDcA8dQzm4Q9h7t+N6uwUnLUEnFgEDbO4gQnbAooVwlbKuV379AumGfTTlk8fZO1j3BrAn1ae+v6n/eKn71djB9c7qU/uRK7ls1evN1CtNlC1aeLj/vHsZHt2+vl1fTurb1reOytvpiIqZwsSbKuwmha/38zSKrKhBNJkewUchyVdE3MSc6G/cWzcL5GnGT06E4y67HAojsjQDObP9SI6OIsM0zSF4gKWK4USUr4QJo9dxu23PkHfNz/G9IlrMKJrCHWPYPC9E5g6egnZ4IqkQlodnUPo5ghKCdeXTU0SZTxqRo2mruykvog5pgK4CqjXGyivZRC5dgtT7x7G3MGTSAxNIHSlB6ELXRK9WrVTgWM9Xur/c6UTO6m/qs6XRf8kF28TGDuGATuwgHL/LYkmnf9oL/KH96M8Oozc7reRZ9Tp114RkGwxCvV7b8H45BCq0RU3irUyh65VsU7/4CZjTPZ4nedk7QbSkqjXYjqtok7Tj9g1iSYIluj4snZBu5cx3qwPHDP4vtTYsXm9+fpH3ZdghOLW4baBAJhm0/nXfhjFb72Gyo0rqPrnYHVdhhOYd9NINcfbtoBipYzbuWYEaUaWvp8PMU2jCXa9y+/+7u/ix3/8xze+AKj2sQyWpaNSt4Cxko1ea5m8GDrgsheub3Dzq6r4CbvbHHwdp4FcvoJ02pJt06yiWLJhWg6schUEyC+GLLRO6/f4cuoAWdhyrghjNQ0jnpUlH02JOTVNqmlOnfCFBADHp4ICjhmIKzEVRHI25EaeDkYFGBMER0d8GKU59O6T8F/qFzNqBt+yCiWY6QJ8Z7oxd7obd946isHdn4r5Ne+bOnIJI3s7EB2cgZXOwzbLqHHiSVbFw6JQT9X+s9ZZ+gQyXzEDbI2/+ZH4Fq+NziA1PQ//4TMwQlFY2RxqxVJznNTj5bN+Z7r+l3OcY0RoOxJC6eYNFE4ehzU65JpNv/6q+BLTPLp45hQyX31VgLGwyNcuozI/hzoDaAkDXBXWl8yvywwrv+Fa87gLgiXvMEHwxnIvM6zYYRcgu6CUQFkBXtHRJmhWx9SY510/jS5LoK2m9YqUyY98piG+004+D3t6Aubxw7AnRiW6tjtPdHWnraB4O7/QKNDrFZSq75d+6Zfwb//2b/if//mfe5b//M//xK/+6q9uTGzV9SyDjDPL3Ko8dYz3slyW413++I//WO7zlsd7HmX/u9/9Ln7xF3/xie9XbXuU+mgmrtq6+Xp+LPijP/qjjef7l3/5F/zO7/zOA+Vxv+fjx4d/+Id/wM///M8/9v3f+9738Nd//df493//d/zHf/wH/u7v/g6/8Au/sCPyud/zKBnzPN8V3xmP/ezP/iz++Z//ecP8nm1X75Ln//Iv/1Keg1HPN8tb7z9a/6Cc1PIo70dd+zD5qsGXg1+l4mA5ZGB+IYeCUUFyrQzTtFEoVBAKF5DNEiTXcfduqy0PK1+ff7bvV8tfy3+r8ULSMpkWmEO4lMrDSGaRj66BgbLyKwlkl+MuWzwXwtpsyDWlHndB8drsspujeHEF2aUYMosrkn6JuYgH3j6OwXdPYOKTSwKUmYt48pPLiAxMwYilMHHwHG6/+QmC1wdhMsXS+DwmD53Hcvew7NcYuKbmNE2nVZC/ewHxVs9DPVfLdp+vVyrI+Zex2jcKguEyg3AVS8jM+BHp7EPNLMtx2zBkwrvd7dHlawZa6b4e7zeN94066mtJFC+eR3b3O8gfPQx7YR5m13Wkv/IlYY8ZjKt4+QKyb74u+Yu5L+mW+HGODDHBsGdh3mExjxazaNdn2N3nNvOc84MeGWL3HBlhmWd5AmWp/c+9LwWKN6JUux/UWvO0Tc/3iPhK6YeUc1elr3PH17u1GupGAbVgQAJv1eMxN++yPEsrhkxbQbFq0HasmXNYgQ1VPgEfwRQBK5nfzaCYx3iOYIv3q/u4Zlmbj3nPc5ugmOWzHm6rRYGkzdc/yj7B+D/+4z/e05ZHue9JrmE9rG/zvQTE3//+9wUQ/+Ef/qE8Fxl2ymqr6zffv3n/937v9wQsbgbFm6/bvE9AycBnBNT8QPEbv/EbAoop81//9V//XLs3378T+5tlyDZSD1j35ndJMP/bv/3bW1on7ERbdR2tCaP3y5870JIZrsMwbORyFvKFCvyBHFaiRSwvFRCNFlGrObKQQeY92kfOK0+9rfvX86cD7MdOxYZVMGFmDZTWcjASGRRiKYkEzaBZKvjW2nwYybmQ+BozGBejSacDK8gur4pPcY7BuvwRYZUDXUMY3nsGA28dw533PhXm+PY7xyVQl5U3sNh5B3d2HcPEofOSyqlimGI6nV2KolI04XCCpkBx0wdPxhxOFr8gbDH9nCm7mmXJmqwxwfD8odOI9Q5JvmJGo+Z1HG/1ePn89Q89pr0474yAr9zbg9z+vch/cgjZd99C5o3XYXScQP7QRzBv3UAtHkdlbhpGx3FUl4KoMwex+Ad7WGLHETDs+gs3Uys1A2i5ZtLKVNoFxQTIG2OXx+pl41jzQ57a51qxyBKFehss8qQOjknSHge14CLMowdg7P4WzEN7UVvwSY7iull0gbznY2NbQPFOfMEjeFKm06o+Aru/+Iu/ECBCILUZFBPkkcX8kz/5EwFeagDg/SyLZfKYKs97ntsKFHv9j73XEwCR2SSo4zXKVPsnfuInhIXlcZ7ndbyP4JF1sl1kcQmwuJApVUDbu89tPh8BK8shiCQ7yWMsm/cRiKn28xyBPs/xmf/pn/5JgJs6r56PwcbIgCumXJ3nPWRCWQ6vvd/z8Xqyyqznb//2b6Uu3qdAMe//8z//8402etlUr/x4De/7mZ/5GamP5yhDlqs+WHD/D/7gD+SYkqVqL+/n+/2bv/kbkQ+BtGoXjynGmc/La/kBgGV7z7HO+7WXzC8/qPCdUdf4jigjLupd8jzLVO+S7X6Y/NgevkPVHpb9cz/3cxsyUM/HtnnlpfefRB7uJNP1FXaQz1eQyVYEFJetGqpVAuE67KrrQyy+KOqLp5a/1r8H/D/o/vgk/fH+/7fbIU9OiupOHXapDCvvmjjTzJmMcSGR2QC7KX8E6UAUap1edE2mGWWaDHHKHwZBMVnm3PIq0gsRxCcXEOqbwGzHDdzedQwDbx7BbEcXGDyLvsYSdfq9T7E6Og8rZ4jJdNWqoF51JB/xBihWAWqaKUwoB3cy12Q4tvAxdid92yt/1YZqsYzUlB+BE5cws/cYQhdvIDUx66Zmiqw2AbHbVrZd/39pRlf1Za0PO9cfGo4j6ZcKBz5A/sM9kpM489orKJ7tQGV8FLk976B48RxqKyHUknHU8zlhSsVneCPPcA13ZTxSUaXdD14u8K0LQ3xXwCaBsAuGveOVgN1HGL+899xdb/kUt0tf1NjlfqirSx7nSm8XKj3XYff3wB7ohbMadaNoewAx628LKGZB272Q9d082BIsEpAQ/ND8dzMoJlD+lV/5FQFBCmSpdrIslqn2t1orUEwAtPk8QTCBJdtAcMNtxbJ6WVgCKJ4jWPu1X/s1AW8EWbyWZt+bWVbvPrcJoPmMLIfPwm0uLI+AlM9NMEogRoDpPfdf//VfG23ytp91EzArEO89p7a3ej4+K8+TOScY5L4yweYzsR1Md3W/Nqqy1ZosrCpTHdu85nnKjzJW2+ojBe9nvWwDgS5lxWNKhmwHn5H71AWvbHgfrQUe1F7Ww+sIxlnHT/3UTwk45vt92Lt8kPzUe+X743vke2O7Wf7m59f7jzO2EACrxTWdIeANRwxMTKUQWSkKOzwynsSsL4O1tfLG5NN7n5b548hcX6v15YutA5wg0VyZfryWYcLKFVFM5VEkaxxPiyl1NrgqoFiCai3FxFxaUi+J33EEwZ5RTJ28Ln7E0aFZAciMJJ0LrWLq+FUw+vRtRp8mWzw8B7LFTMc0+fFFzJ7sEpBcsypu+qWaC4qdJlMs0aedOuqcjG7BtHiPKV3jMbW9XWtVr22UhBVeG52WCNRmPIns7ALiA6OolUyU40m33TL2frF1YbtkpcvV7/2Z6wCtYrJZFC+dR/aD95Hd9U2kv/pl5D/aB2ugH9adARQOHUDh6GHUIiGJuCy5iJuRpFUUaWUyvc6PdGRxPcwut9W44N1Wx9S4pNaPIpPHufZRyuM1EoSMbRWAXkfdyKM6N4Xq9ATsiRE48Sjoh71V3W0FxZtBazv3CWC//e1v3/NHQIaRgIWgieBlK1DM4wQcZOK87dkKFHvPU7AEPgR/BFtqIZgkWGaZBKrqJRE0EdyQTSTQYttYHkGZl7H1mtzyvAJwBJSb9wmsCHoV40nGlHUoJpJ1cJ8glaBRAWS2iddwXwF1HlPPR1BHEKn21TOofYIztplspjrHuilLsrGsk+Wq6ykHnmM7vW3kecpCtVFdzzL5vGyfArje9qk6lXk1y1TnyfiybaqNClTznVAPyMLyWspVyYPbPMdreE7dy+d7WHv5npUM2X7KjgvL8b5L7vNZ+KzUm/vJj8+k2qbeKxl5yo8fPViOWrzy4jG9/7Av8Mzb5w6ETr0Bo1hFNmdhNW4KMM7lqygYtrDEDKhVrTa/gjaBtPYhfph89XnVN3V/fH7GI058GIVagHHJgpV3gXEpXRBgnI8R3MaRDyeQDccl0jT3GUBLzkUSEo16puMG7uw+hZH9Z+E7exMrt6clzdLIh2cQHZ2D73Q3+r91BFPHrqC0loVdLCMdiCDcN4FiIo16tSbtIBh2AXHd9Smm+bQyoW6yxWqyxrVaqHNqm+ut/g94XOnoVufVuUfR3/W7bn0E7TXTgrm6hsTQJEKXexA8cxWFYFgW5iv21qvqeNr69f16vFW69Cj6qvVlXZjfyuw0srvfRm73LgHFxTMnkd+/F+Wua7CmJlC6chFm93XUU2toWGU30nStirt112xa0sPV3TzC0q/pc9uogwyxAsiqv3OtFr4j93r3OjLAW70/de/m9+W99mnfNwNtSXl1R3Iwm0cPovjBu7AGbkoaJpqMuz7Qn59vtxUUb36odu4TNCjz6c3lEqgRqBC4KmDMbQJmmghvxYh6zac3l6f278cUE9CxPQooqzWP8RzBDVlAspM8R0CvgNVmIOUFxazXu+/d5jney7JUfd6yN5fL672ATj0T1wSWPOc95t3mMxDcsX51nLLg8ykQ6T3nbeeD2qjK4prAVIFr73HvtqpTgVme89blfb7N13qv4zavZZ2qfAVuH9Zebx28V93H7c0yV3US5N5Pfmynuo5yZjmb267aqNetAevRZOGyxBx0i6UqpmYy8PmzWFouYOBOHDf7YsIY06+4NZArZlmtH7dOff2jvRstJy2nZ6QDnLg5dTi2jappoVIwZaGPsUnGOJkV/+J8JCEpm5iT2A3ElRQmmefz4bgwwLNnugUUD+4+heF9Z3Dn3ROSmzjUPyFBuYb3dWDg7WNiVi0m00VTUkKxXoJimkwrUOxligk81bIBkD0TTjVeeddb6RPPb3X8SY9JffU6rFQW0e47mP3gOHwfnUT0+oCwxdHrfbAyeWFmnrQOfd8z6heeD/D6HTz/78BlR+twclkxk05/4zVk3viqMMVkh3N73oXx6VFURoclh7GTTsEpFVvBthhkqxn46h7w2wTFElxLtj1A+AFjlKRmegQda/eYRV12GWK20wGf0x7sh0Wz6aEB1HMZMf12U0d9/r23BRRvRvzbsU/zZxVo637l06+TzC7Nhh/EQPJ+b6Ct+5VHsEJg6y2LAlfm2uq4934yxQREbK8CPYop5r0EUtxXgxABkmI0ecy7r8CTYhTVvfRZVvdzzfrVOX4A4P5mhtp7PdvtZU5V+3mcIJAfErwsLs/zwwNBMa/xnmO5BMrqGbztUHWq8jfvE2DSpHnz85DFpj+vAousW93L+ilfMv8KsLJ8dS3X3Fey4ztQ96j3wfP8YML6H9ZeVYeqn/ewfdznvTyvnk/VSVDslRHPK/mxfZvbQ2aZsuU5lqvKU3Xq/a2/OG6WDwNpVWuODIhG0cZSyEA8aQozHAoVMDmVwtRMGumM1fqiuWHy15pMank/mrw3y1/vu3+wWn++ePpDM0DHrqJatkAfWQa+quQJWA2UMy44VsG3iomsBOIiSE4tRJBeWBEmObscQ3J2CbERH0K9YyBAHth1DLffPobxQxeQnFlEdHgW4wfPY+yjc4iPzaNilJoMcVVYYgHGBOhe0+mmT7Eyo+bEVG3TH5r9ajMY9k4kvfrG49593qv21T1q/1H6qwqeVckXkJ0NIH5nHLm5RRRX4kgOTiAfWAaj1LrMjDuGPk753vY9Snv09a33qeWlx9tWf/AA1XodtUQcpWuXkHv/HdDHOLvnXRSOH0HhyEGYF8+iurggfrb1Qh7rZbLFFTfytERipsVd0+qu+ZGtNf4oH2PFGq9DGGG5vgUw1fWt9rXO7cT4oD4Q8LnWi0Uw0jZ9iJ1oBOs8JibhW+tPW0Cx6pzbuSZQIwh5UB0EjwR7ZFC9QGqre1iWCjS11XkeI0ghKCbY2XwNmWACK8VeszwCJt7jBUSqDNV2mhoT1NGsmGUyUBTbzLawLJarGGcFshSYUwGyCLp4L01xGbCL+zznBfAsj3JQ9XrbT8aUYJTgXAFuBdqU7zXbwWvYTgJsPhuv572bz3nb/KA2etvAbdbJNhMA89lZj4ooTl9b1sU6WTfPsS1sE+vj/V7ASjl7gaVXdtymLAhGeR8/nlDmBPMPay/rYPtU29kWLtzf/C69dW6WkVd+3utYzua2q7r0ujWQbpaFGnTVmhGjg6ECgssFZHMVrKXKiK6WkMlasG0H1Vod5XJNGGSmZLrXh7gFiDfXo/fv/w60bLRsnhcd4DghDC19i0sWbKMM23BNqcs5Q8Bxac31M6bps0SpXk1JNGnfuV74zvdiZWhWfI2N2BoYRdp3tgcTR69g8ugVDLx1FLMnO8UEOzm9iOnj1wQ407eYDDUBuSxNE2ovY6wCbikg7DLFjPrqMatuMjKUtxrztpI9z211/EmOSa7PZn5Rts2xKqiWTFiZHHL+oPgZEyxLuz2TzCepS9+jxxKtA0+rA62+z7HDyaZh3uhE7r1dEok6t283CseOoHj+DCrTk6iGQ3DWkpKmqFGxmqDYBdfed6HGGxlbpJ83r1EmypvY4HaNQY9TjssMc2xsjY9MN1UZ6IV59iSq/jnUi4YbaVsB/03tVs/cVlC8nV8AvvOd74jpsAKhfABVnwJqZIiV+TS3CaCU/633epZBM2SWqQThPa+OkcHzgmJVH88TkBLcEmxx8V5HP1d1nOCLC8Ep7ydDTfBGRptglkCP4JrtYZu5zXLvx3ASiKmy1fUKYHvrZR18fi8o9rafTCvPK3NslkXgpsra/Hx8BvWhgecI/HkP28I2EzwqgH2/NnrrV/ImeGfZ6vkpR/pHq/Osk+dZD+sjSGb9PM9t9XybgSWBp2Kvuc0yKBMlO4J/xVA/qL1sC+/hO6EueUHx5nfpZYAfJD8FipW8VNupb3wutWwlL3WO65f9PGXAgVNFjx4eW0P/nThmfFmMTqzhVn8cfQNxLCzmUbacl15eL7u+6Of/4jG4OzaecTLHiaJdRc1ygXHVrAhjbOWLEnyrnCuCkanNdB5W1pCI0Yw67Tvbi8G9pzF26KIE2orcmUKwexgjH36G+fO9iE8sYHjfaWGMA1dvi8k1I1YXVhKwTTedEYGxsMQ0oW6aUbcYYUfMuyWwjfgWfz6qq5ogcq0Wr+y24/9gXaLCtoLWVIsmMrMBRK73I3ylF6WVVZSicRe8U746r/vGf/d2vA89fr3E49cTzPdo7lxLJmCcPIbsrm8g995byO/bLdGoraE7qM7PoRpeFvPihmm6eYqZXo19eVN93vGH57n/pPqoynrS+9W4571fWbSosVHqqNqwJ0ZhdV2BE19tWrQw5sznn8/7vG0Fxaqx27Um6FAgyFsHozgTACnAo9YEQpuDF/E+lqFYQ285T7JNc2kF0rz3E1yybi+I955/mm2ypnxmBWC9ZfEYz/Ea7/H7bbN9D7qe4Hyr52N5PM7n36rsB7Vxq+tZjmLEtzrPuh50fqt71DEFQtWzbCW3x22vKvtha1Xnw67T51sfAx5VFhz4yAozkBbZ39V4CaGIIQzxcqiA6dmM+BCTQSZL/Kjl6use/11omWmZfdF1gOMFA26Jb3G5IoCVjDFzGG8s+aKwxma2IKC4mMggMbWIhat3MHnsGkYPXsDYoQsY/uAM+t86iokjlyRdE4NpDb5/CoPvnRCGmGxzrdwExBVb8vwSkCtAvCU7zAjUTd9iYYk9ANg72fNub7fMlW8e20uTaeYpXjrXhXj/KFKTPlgp+ufdH6hvd/t0+Xrc0TqwtQ6ICbHjoLoSFiBMn2IC48KRAzCvX4U1PAh7fg61WBQ0o5ZI1E4zNWUTGG+HbDletLtcNU7JWOTU4KSSqAUDqJMJLxlo1FspNx9Wf1tAsRex82G3c5/mwpvNnh+nPprLkiH1vpTHuX+7n0+X33798TK4Wr7tl++z6D9qIjYynsKd4TVEokUkkiay+YpEla43GWTmIVapTlSffxbtVXVr/Xsx9E+/z9ZE7HnqTzRNJjB1UzRZYkpdpTl1sSzb9DXmtpHIYG0+JAsjUWeXV5GcCWJlcAaL1+5g9KNzEnCLgbb8F/uQW4ohcnsKE0cuY/LwRaR8y5Ib2bFs1MV02jWhZt0N+hNvRKB2hGlVYFhMkZvBbtQYR11T22pCp/a3ezxRUVzZLitbQC6wLGmZaD6dD4QExKv2bcUwbXf7dPl6PH2exp+d1NfWGNEQ/2GztxuFTw6icGg/jDMnYXZ3whodRm0pCCeVQt000ahV3bzjHuD6PMhXPgBI/uQG1isVlyG+fllSL8n4xHPN5WEWLW0Bxax0pxayfDSDJbh93Dp5D+/diil83LL09Tv3zp9W1mTCaQZNM/unLUvf/2zfu/oiaNt1MLBW180orndH0TsQR3dvDIOjSUTjJfEjdieQ/CrZ7i+Tm8tTdajjan0/WXmv32r7fvfp47r/aR14Gh3gmECAJ2yxVYGtGGOz7DLHpiX5jMn0Lt+awMyZbgS7R5CYCYKBtwz6GQ/PYrbjBkK3JoQ1Hnz3UwS7hlCIJjdY48JKUoJ6uf7EHp9iAmSy1Z4o1PcAYpWeycO+qsmtd+0GivGMM81AN95ruP1UsiIYby4E8YwynZqaR7RnUAJs0cdYTCmbk1Fe+zT16Xu1/LQOtFEHmmMI5z8yLlgW7IV5ScdUvPAZSpfOoXyrB/bMdJMtLrgBtxSrunF/+9r0tGPSvfrhjm/uGNUQf2EnmxY/aScSQjUwj0bZ9OQs9gDjB4xVbQXFO/VFgaCWjDF9aL3myfern9fQZJoMsRcQ3+96JfgX//y9yv7iP6/2SVG6zfXz9L7VYMo1cw+nUhbq9XXMzWexGCxI6qWh0TWMTKwhFDZgbZMPsfy5yGSTPjVNf7t11wdQvkR6JrOUsWKpve13t5UPXgsUu3mS1f7z9X6eN33S7X059Uv6L32Lm7699C+ulm0BxNVyRfyNq1YFlaKJ9OIKfOdvYfTQRUydvI6lGyMCkKeOXgVzFudDqxJ8i77Ft989gYXL/SgmM+D9taYPsfgwS6AtlzFmfQ7NqVn/PVGoGVir4Y4XzZzFjADrHTeos2r/c/rbBMDyfPeZzD7ueL9+9+7GhJKm4LnZAJbOXkfoYjfyiyHUK/Y9IPhxy9fX6/kI9VgtWh+2WR9k/KjDKRZRDS6iMjKI8sAtWGMjsJeCqGczbt7ipgn1A8ebZzh/VMG0pH3C/nJcrKOey6Iy2C9BtWgu7Z5/fP1qKyhWyr1Ta5rFMkATAS+DHhH8smNx4TaP8RyvaZcP8U49287W03I+39l6Wwqr69WyeLAOuDpKdnh0Ko25+RySyTJSaUv8iQk+eY6LAqIPLu/J5K0mnSxbbXNNc23LqsGyqrKWQdqpo1CwkMmYsO0aHJoglm1UKjU4ksbE/YLbioStALH7BXQ72q/LfLL3ruX2YshN9VmXLa6hZjUDb5n0MXZBcY0MMvMZG6aww5E70/CduymRpkf2n8XArhPiW8wUTKVkFitDM+JbPPjepxKAyyqUBFwrlpim2gocS0ooTfPk5QAAIABJREFU+hqTMfamZVIMcb0h4Fh8irm96SOb2t88/qjjXLdLVzcYmHWms7JRTqZQCEZghGPIByOoV2lq+WLohX4O/R5ffB3wjCdM+2aWUFuJSM5iezEAJxFH3TDQsG0BmWpM+aLJ5V7yoQ5Gzq4bBTiZFOyZKdSz6Xs+Hj5u+9sCip/lFx5GkKaPMaMJ0zSaAJgLt3mM5x4WZfpZtp8v7FnWL398jXXUHDUhf7btedby0PV/Md+/GqCZXqn7Vhw3emO41BnBtRtRLC4VBGjKNTSVliiorclhW/sXoy42ATEBeI2T2fV1AeOlko2ViAGfL41MpiwDc3zVwM2bYQQXswKY0ykT4VAeuawFq0wAXZV73cms2wddxrg1SWlr+5/xeKP71xezf/G9qOVF1zc+J32LCVZd5rYqgJiMca1MJrcKM5NHNhxHIbaGcrogDPCaP4TAtUGM7DuDwd2nMPHJZaz5llDOFpANxeA/1yvRqMuZQhMUu2mY6EfsBvhyATjLF8DcBMUCgDdAcTMv8T0fze79AKfejxoTFaPsjiHt0y93buCOowTxdt6Alc2hWizBsZq53vV48kznb3o8bZ++v/DjHz+wKVey5raMIZYlEaiZy7eez2Jd0jO1Am6p8eaLI59WRPxGzUZ9NYaaf64J5t35H8euJ21vW0CxqlyvWy9ip2RBU4J6Yx0Ovyp7FOFR62enqFQbKFeaSblVp3mCsh61Tn3dzuvJ8yhzNckjC1smC1txBECOT6UlsNbN/lWcPreMzp4ocvlK8+ug+rjTAsXtfXY3aX293kC5XEU2S1NuE/m8mw+Z2729YfR0hwX4EsRPTiRw9UoQBMhkk2emkxgbXUUyUcRKJI9kvCg+0GSSCbTb7wOt9a29OqDl+bzLk2OLzcmgVUWVKZrIFJcrsk3T5+JaFoGuYcydc/MTZ4IrWB33Y+pEJ6aOX8XcuZsY2tOBsYPnERudF2BsGyYqBdNlgZu+w3VaiDDqdM1NxSTgWIFkBYqbUafd/MQeNsfDEqux0Ct3HpPFwyhvPu/df9xtZabYcByY8TXEeoew2jcKm2wS5wltZKUft236ej0GaR14TB1QQFj5GIvpsWe8qdpgaqY6/XBr1Q22+AsnZxn36LJWh4DifM41/W66tcmYyA+fT4hh2gqKv3hfFLbZRv/uTpbfygumzAec+jpMq4Fkvo6C6TJW7h8lO0vrelHqJsO14XTftMWvOg2U7QYcp+kjqf5oN9+vvwjrL8I7rO/UW+oz/YPn/HmEV4qyTXDMAbFarSObt5HOWrCr9CFpR3+8F0y7/UkBbZcRJnDlcQLjRKKE3t4IurvDiMeLqNoOSsUKBvqjuNkThmnaAnjv3Inizu2ogGeyxDe7Q/DPpxEJ5zE1EcdqrICiYSObLaNW47O4/dEFyG6b9Pjajvfbmkhoeb7c8iSoE59iCa5lgT7F4g9s2ZKnODW3jJmObgztO4PRAxcwduA8br/3KXwXbgmLvNwzhtED5zC077SYV5fTeTelkqR9qgkbLSC45gbWUj7EKsjWPQwxP4Y1fYhbY869DLEaD9UklfqrruV6q/PqWq4fV99VeQT0RiSGxO1x5INhECS7jNPLrT+PK099vdaXp+mPO6I/nP/TSsVx3CB6ng9fO1K/B8jerz4Zf6wy1qsE7i32+37XK5k/6vm2gmJVuV63Jl5tlYXnS4hZaSC05qDfV8GNSQtZoy5+jferT776er4M0Vy6aNWRMRyULPdexThzsu9OxrfpOTyKf7/26uMvs+xd/aNvMEFmvmCj+9YqbvbFcWdkTQCyVal5BkR3QthenVF9wF1zElitOWIWnUqVYRi2mD4z5VMuZ2FkOI5LlxaxvJwTEJxJl3HmjB9zcylhkkPLOZw/68f4aBzxmIGhwRgmxuNIJovCGhMYkzlemE8Jc8xn5+IFxe19vpdZv/Sza11ydUDMgotl8R+WdEymJUA5sxhFuH8SsREfIgNTmL/Yh7GPL2Fg1zFJxxQdmoWZyiEfiYtJ9dTxa4iNzKLOHJkq5dKmKNNb+hBvmE03fYnVR+kt1o/6zhRAftTr73edAtwExTWzvGEyrY4TxN/vXn1cy0brwPOrA8p1ol1jSTt1oeHU3IBgChQ3x8p21dEWUPyoCFw1Wl//+F/MXOV0TR0K5Tq6pywcvWnig2slLKzS7JKTd5fBok+la/rkfmkmg+aCYppaN1CqrCOSqsEfrWI166BcXYdhNVBz3Pu0T+Pjvx+l21xr/W6P/IqlKoLLBnJ5G1OzWQyNpnD+SgQnzizDH8i7vrgi73VhWdQ7eHr5u39mG5O/JmPNfZpDLy8V0NUVQk9PBKGlvAT6oj/x6FhCgPDkeAL+hYz4Fg/0reDcZwuYnkxiciKJMx0+9NwIY3R4FVcvBTB0J4rb/REM9IXFvHrBn8LkWAKFfAWptZIAf90f26NP7dMP3R4ly+d5vFP9mz6+DI5VzhmwjBJq5QqMeBr+q7clJdPa7BJy4Tj8V24Lazz5yRUMf3gGi9eHYCTS4oNsm2VUS2XJP0ymRQJ5Me0So0x7cxILI9wQFpnRpmXhpG4LH2LK1v3fv3dy/bDxzftunub9cM7ASNmFxTCiNweR9Qfd8UjmE+5c5GnKV+182PPo83q8Ubqi9W1n5pdq3OH6i9b/BN8o9w1+mGuzBWNbQLFXYfX2vX9g7ZMH/4TWkTcbuDRq4Y3Pith9yUCfryJBslxzJvePiv7FZIJpGm07DdTE58gFvATUUyEb/XMWFmI21vIOlhJVhJKMiNsyRdBM8Xa9R13uo/QJDsaJZBm3h9cEiNJ6gR90aCZdLNkwTTLFLpPq+o60my1ugCzwaqwIsr6Fgi1+zYwaTfNp+hCPjsZx5kwAff1RxOMlYYn7bq0IML54IeD6D08lceTwNJaCWWRSprDFtwdWkM2aGBqMovNqEL7ZNQwPRjF4O4rx0VX094aFNfb7UsJIqz+oR5Gbvkb3L60Dj6YDXjbENsowMwasXBFmKo90IAKyxQS+wx98Jsud3adkzbzFi11DGHy/w2WNh2ddv2FJseQG1aJ/sALGXIu/sIqML+aJTMHUTOW2BSvMPq+W+73Ph52/332PepxzCqZiYqTp7FxAAm2pOrl+1HL0dY+mj1pOWk7PVAe2sPz4IvZzjksbyyOMk48r07aC4i/aF4Xntz2uyabb/tafY6W6juEFG693FPDOBQOnb5eRpy9xUzF4PYGtZTdA82peb9dcgEz/47K9Dt9KFZ1jZQwvVBDP1TEXqWEiaCOV90Ztc//wWnlTOVh98b4YPb/v1x38dfvdL+BKf2XwarKx2XwFC8ECLlyLCltMMCxRnz0DYvvl5wZncP8I3D7IwFnDw3GcP7+Is2cXMTISl+BaNG02SzaGhlZx6NA0Ok7NY3o6hWLRRmfnMk6d9KFUrGJ0JI7DBycRXMxhNWbg9CkfRkdWwSjUt3pC6OoMIhLJC3PceTmI6ckEJkZW4ZtJClgOBtIoGhXxTb63P+7MF2PvH0r75a0ZEC3f1kT4WepX3XZQTjHKdF7A8crwHOYv9CE6Moe1+WUs9Yxi/DDNp49j+IMzCPaMCmhe7h3D9MlOJH1LAnIdBtRyHFnW+ZHZYx694TPcZIg39pv/39QFNRa6Y1CLKVb726ovzZgM0ob1Bhg1u5I3UC2Z8mzeurn9LN+Xrl/LX+uf/v/0jknt0oe2gmJvA/V268/+yWXhglN+FSHYjaQc7Ltq4isnDOy5YmIu4kiwH+V3qAJv5UoOimVHwDFBsV1toGA24I/WcH3Cwo0pC7ORqjDGg34bvkgV1WbwIJbF+mS98eWIkSbb8Ty6jCfXhRdXdmoSRr0j+I1ESwKIY3ETF66u4GjHEq7fjAlzrK51db59MpFyPcFqVD0Ev8wxXDBszPuzuHAxiIuXgvD7s4hGDTGJ7r0ZweXLQQm4VTAq6O2J4OjRGeSzFUyMJ3Dk4ym5lgD7yqUAgsEsDKOCgb4IJscYZMtAT9cyOi8HsBLOY246ia6rixKAa3oijmzaBFljt0+6fVPrUfvevZblyy1Lt++vSwTqYjKL7NIq4lMBBDqHMXnsGvyX+rHYNSyRp/1XBsSMenBPh+Qtps8xfYtrliUgkhHkFShWbLEA42aMADXO3G+tdFGdV/s7siY4bwJz+hGnJnyI9gzCTHDsefJorjvSdh2nRDP3Wge0DrRBB9oCituF0NXgqcvjJKVlIlB11gXQHugy8eqnBXzjtIGrYxaMMsEqI9Wuw2mQHV6XgFuZYgOG2YBVZXCgBowy76+KHzJB8WSoCl/ERu+0hQFfBWmjxUazLDJS/FOm+TXX+n3oL3Kqb7r60V55sEw1CczmKpiczSEUKUoOYEaWHhpL4czFEDq7Y7BsR/TTC4qfTj9d/Vb1cy1lq6B2zRRlPM5AWytRA9evh3H8hA/nzgUwMZFEOl1Gf18EF84HkM2UMTqSwHGC4rwl+Yl7esJiBh2LFiS4lgrStRzMiUl1as1ET9eS+BYzbdOd/hXc6AxiNVrAYH8E0UgBgwORZromzRA83fvW8tPya41f7Nd3yZA2Iz9bBRPGagrR0Xks3xzH8q0JzF/sx9jhS+JPzIBbhZUk4lMLmL/QC/oXRwYm3bzDTf9hMsTKZJprYYzFhcn9P/WOd+540/ooofbVejvG2/u9f9Yp9TKqdb0B2yjCLhgu2y3ycdt5v/vZ1p1sr65Py1vrm/4/a/d41BZQrAYnvW79uT2tLJTNPCNEDy7Y2HO5iNdOFvDVk3kc7S0hlHQECHMCT//hktUQ/+B41kGqUJd9ssTMQUyGuXfGwqVhE6OBClZSDmbCNvpmLMxFbFjVZoAuAnH5WuyaX7Ms3u/+Qbus9dM+l76/fTryoshSTca4jidNdPfHMTGTkeBxkg/YciT1UjZnb+j8djw7fYVDIQP+QBY+XwYTk2sIhQog+0tATF/mhUAW4+NJjIzGcfHiIvr6VpDOlMGUSx0dfhDgRlYK6Ly6tBGFOhjIShRtplyi2TTZcAbsImiuVGrIZEz4fGsSaIsRqPt6Q2JOHQxkcOGMD1yTMWaO44pVc0F7G76IbocMdZm6fz9vOrAx/qyvC5g103nJSTx79iZ8529hdXwB0WEfpk92CTscuHYH6cUISsksCtEkUgthMNCWpF9qMsVes+mNYFpN0LlR3xb7W8mO16vj3nt5zHtOXfOka8USl1MZFJajsPNFdz7A+hVg1uPOxrt4Ujnr+/QYqXXgi6sDbQXF7UbsL2d5LYaYfsE0b951wcCrJwr4yski3rto4I7fRtFyfYAZfIjXMWBWJF0XAMz0TOUKmd515Ep1TCzZuDJq4dZsBUuJmkSeHgnYGA1UsZZ30zHxD1ExxDTVDq/VsBCrwrKbgFkzxpoxb3OUP/GRbX6EIUgslmqIxkx09a5KlOn5hYIASPlA1DTfp56qP5SnHR/cclxmWCab6+vC6EaihvgRd3aGceECzaWX0N0TQef1MG7cCCMcKUhOYvoOMw1TbNUQxvjihUUwBzH9igmEWT7NKQl83cks95sfnmSS6U5q5dkNG3aljlSyJKbSZJX7esK4eHZeolAvLWaQSppIxo2Nj1QtH2N30vy08tD3txhEvjstj5dPHhxrHLuGlD+C+fO3xGR65lQXxj++hPGPLyN4YwTB7lFMf9opQbiSM4uwCkXUKrZEmfaaTiv2WQFZ6pTa5nrz/rPQNwbvVO1kFFe2q5zKgsCYPsVso1qeRftU3Vzr+l++/qjfv+5/Sgd2qv+3FRSrxut1S5EfXxbuJJ3s71ykir1Xivjy8YIsb3xm4PywhWjacSNFNxqo1BpIGXUsJxwE4zVEM44E3yrbdRjlOuZXquiasAQUjwWrCKxWMeivCHO8uFqTQFycCLjmqIzwW0e+VMd0yJbyGL1a/YE//rM8jRz0vS+yvDcmh03rhHS2gmjcFN9dssV3RlK4cG0FvoW8+M27OtpenZA2eH3oGVymQR9iB4ZhSX5h+g2PjibQfSOCa1dDOH9hEbNzKQm2NTAQxfnzAfEXXgrm0HszjHK5Ku0lEFZ9yl3fr+1ekNwQRtkqV5HPWRgbjiHgT6NUtMWMmmwx8xiTeWqV6d2+Xx36+Ivcl/SztVG/m4woUzSFBiYxdbwTM6e7MfVpJwb3ncFMRzdi435ER3xYuDyAmY4bCPdPoJzNS/oiAmqVj9jLEPMdbYx5zTo27z+T9ygfJe9tG8EwfYrZvmfSJg8Q1/W3Ube1XLU+ax14qA60BRTvFIJXA+SLXh/zcNF0uWOgjK+cLODLJwp47aSBQzdMTCzT3Nn9EyNwZoqm8FodgZgjwJjMb6nimlNH1mqSeunKWAV9cxXMRAiIXdZ4YN5C1nA2/vjcP+g6ynZDzKuZtildcCQNjivv1h/kiy5//Xw79EVaJl0NAZHLkSJGpzJIpiz54JM3bIRXis3gWgzysh361wSk6+uSa9gFwxWUyzVheNknTLOK0HIBkYiBwGJOgmwxP3EiUcLkZBJXrgSxliojn69gcTErfr/3MrgP9slvTTxdcKv2Gdwrny2jbFbF5PriuXkxo46Ec6jX3f6vQLdbX2vypPV3h/S3OcHQ8n4x5O2aD7t5L5kuibmKF67cEWA8euii+BTThNp/sR/BrmHERn0SkCsxFUAxmUHNtFCzbDfgVs3ZyEGs5i1cq/7NtVq8x7m90/rEOutVR9rPnMpue/R4Qjk8i/ex0+9f1/dijF9aX9vTX9sCitXL0OvWH8nTyILm0H0+G986a+CV4wTEBey7UkLXlIV4zo04TTNM5hwmM8yo0ovxGhJZR4JvEdimCg5GAzYuDpdxdawsaZfoT3x1tIyuSUtyE0vE6SZTRjPsSrUu/siBmMsoK9Np94+8BUqe5tn0ve3RkRdDju7EkMG0yAh39dKXOOuaTDcaAjDp5+tOHrdDbm79FdtBcCmPO4NxXOsM4/qNCAYHVzEzm5JjA7djwhoT+NKM+uw5RpDOIRTKg2xxLmdJWy2rupE7WQHWFqN7v/a7YPje61sT5rpTl2BcE6OriMcMYYxpYl213Q9aum/eT676+IsxRuzQe2QgKQGtBK7utlN1kPKHMfdZD0YPXsDQ3tMYO3wRs2d6MH3yOmY6uoQtXvMti38xcxzbpbIAY/oXs++6/XPTMzSDernj2rP/X+Vz0+zbsav35E7W+rPpvWmW7aEsm9YZrTPPuw60FRTrL05P9sVJmS+T8SE4XYjVIJGmTxTx5RMGdp0v4sJwGbMrVfEVJiA2rQZWs3UsROn7Sz9hBwVzXdIwMegWzaSvjVk4c9vEzRlLAPKNyQqujJYxE7JRsurNL8LuBJy5jRO5OoKJGuZXbGGq680/b6Xk+v0+2fvV8nP/KLz68/+z9+bPUVzp2uDfMDMRs0XMLzcmJr6Y+WHis9tbY/fX7e77uTf3dbttYywhQCCzGoRACIEQAiG0on1XqaTas7KyVt2+nu6G7gttszRmMRhjY7AsIwuEjCwQaoSw/Ey85+SpXGqRhCVvZEVkZGZl5lne857lOe/GeL4vxlSVaxs9oGP7rm6sy21jzrZowah/n2g43/eUJuUTCkdg7/WjutqBoqJOFBZ2YHthG9atb0TO6jqsyqlDeTnFIA6irKybqVBXVvWw+85OL5MyE6g1SoipvMYJ8kHKT+VTAmG4XTLIrnh9TjU6Wt0gFWvOVwJUzz99HqS8gtcXor2s8ljjzzfNX4oko7OsjjnZKv7tSgaOSZV6X8YbqFy1BdWrt7GYxc7GTni7euDvcUHxSEyVmqTNeuD7XeZfNt4KFW9LQmpJiBdgvv0u8781X1nrB8Gf8wqK9ROWdW1cEKemhyYVIg+PLimCdSUyFi334NEMD36S7cGSfAn5FQEGVCnEEkmCybEWAd/yphDqu8Kwkdq0zMMw0TOyG87Z6ce6Ej9zzJVXJmNFocTsizt6uVo0LeRpMqTQS/QNhW2iUE01rSF4/EK1Wlt0p67DbOtqvWfRUOUBVX2Q1JP3lDlQ1+QBqVBvLexCTZ2LqVAvHK043zP+V/tAVLUl9vkU5myL7JpLSmyoqOxhKtPkcIskw80tLmzOa0LZ/m4Eg2EmIaY+xMu6UH0lyuwU62p6UFVuYwBZkYXN30LlafXVheM/i7bfF9pS3w54/bDXtqBxSwkqs/NQ+up65niLwDHdH1i/A7byejjqWuFu6WJxixWvxCSvKaXFltRRHTOtvvB96QtWOS1efRh4YF5AsUDYgmDW/Vx29PmCmiZf8vpc1RrC7zZKqtq0B8+/4cOa3QGUNgYZ6A0opDYdRVM3hVUKorI5xLxOO7xcekyxi0ly/EaJzEBxQVUAO2oUJjHOKvAxibEvIGw0ybFQFF5/FA2dIQaICRS32iMg1Wqh0jkfEi/BG3S2+GMu/PFDpBfnPwKVFJd4x247yqud8HgV2Ox+9PQGDKB4fvmFS3Spv9ERClN4pAgoXBIBdLLlDQYjzH6Y7Ih9PpIO27E0qwaFO9qYLXFtrRONja44IJ7f8iVrby7RJmkxqWg3NTjg6JUQIUmUztZa9LGFL8/Dzr9W/QWv/ZDHczFGUB1JvZicbxH4ZerUL76O3S+sQunLa1C1aiuaC0rRWVoDe1UTi2PMJMZqmCYWp1gdb37I9BI8YY0/1vggeMHi92TzucUf32X+mBdQrK+gdT3X3SRVUhzjUuK1qpSYbImfzfEgo0BC3n4eW5hUnAn0kpozAV8CxSTdJckvOdzy+aPsmpxqLd8uYd1uGcW1CkhKvGy7xGyJ6V0C37SYpkmf4hCTLTJ5ruZHEB5JbwtlSaIsnp4rT8/0Puc/cqZFHqdL9vWgqKQb9l45rm5IvLkwdOf8TOkT4CXgu6/MjsLCThQWdaJkr41JgcnL9P79duZN2uGQsSm3CRs3NaGz08eAKbcfTt835q8OWplJlbpgayOLaazZWy8UrWZqR+v5wvCoRdfvCl2pD7O5klSLYzEE/TJ6aluYoy0K1VS6eB32LV6HqpytaNy6B23FFbBXNcLdYoPscCMoy0zLQ6hSf1fqZZXD6mMWD1g8YPFAIg/MKyi2dggfbAdIANSqtjB+u4GkxB48meXBbzf4sKrIj5J6BQ4fV4+2u8nbdBiVrUEcaA+j2R5mTrUIyJJNcUVzEDm7/MjMJ1viAAPFG0tl1blWiIFgNtFHyetvDA5vGE22ENp7wmh3RNDriaixVMkLJzGMtvC32vfB2lcMPBb9BP1oQyaCzi6Z2RI3tXixr8KBji4/A8ULSy/Oz5FIlNkE19W7GBjO29KGzVtasW59E7JX1SJreQ1Wr2lAc7MbATkMkg5v2dKKunryME0eWoUNsTao6ts3GSDWP6c6zvaep8WBL4V6OlDdDYfDj1BIbF7xZ7NNb2Hpm5wec6mvVT5OQ6s9xXjx7dJDgGIWx5fAcTiCgFti6tIkIa7dsBMH1heiftMupmLdtqsC9uomeDt7EHD7eAxjNmbwmMAWf3+77WnR36K/NR/Nfv3xsPWXeQXFgnjWmQ86s6MD34kmO+H1ewNYtMLLbImfzfEyW+INe2XUdXKv0g5vBO0OsiUOoaIliGZbGPSfV+YhmLqcYRZ6aWmBhFVFEgqrA+wgSTHFHfbIQt2SS4nJy3WzjRx1BWFzcHBN0mglSBLpCESMYgGiZ1efudTdevdhpSmpTtc3eVFe7QLFKHZ5gvBK5L1Z24RZCNowXo7bEXMP15I/CGZL7FWYF+ri3V0o2WNDZmY1kw53d0vseVVVDxoaXMw7tn6zaCHKmTxN3m9DoTAaanvhcvqZCjUvi9WXktPMootFl3nmgSgHtzSWkMdmyeGBs7kT5IW6s6wWnaUH0F3ZAGdjB/NKzRxvKUpqb9SWffECaQXNc7tb7WS1k8UDP3gemBdQbO1oP/iONgcBMeY9+g+bJfwo04Mnl3nx/AYJ2UUytlcpTF2aAeLeMA50hFDeQqCYh2AiEBsMx0AAl8Iyrd/jx7LtfmwqpW9lpjpNsYkp7jGpTQsJlxKKsv8ILJMqts0VZqCZYh/7lRiqWhWmVk3AmC+qxNnaYbL4/cH5nXiJ6OeTwmhq9aJorx0dNi4hZoCVeZ02Lmbmld5q+pQXeZ2m2MTBENeOCATCKK/sQVu7F05nABs3NWPxkgrs2WuDT1LYu/Q+1YH3W94n5rV8aSXIPD+yed5T1MbiF5MjH8vm/+vzox44fnPtyfncyu/7236kUs0kyKRmHYkwkKx4/MzZlrerF1K3AwGXF0FJBoVponf0vCbGQ/1/Fj98f/nBak9rfWj13+93/50XUKwf0K1r44I+HT0ECCAgSnGJSTpMHqd/uopLideVBFDWHESvJwybM4zazjBzuLW3QWHSY3KuRZ6j6SC15+I67lBr5U4J2yo5ICYpcV1HiIVq4mpg3KGXwxPG/kYFxTXkbVpBU1eIxSimMoUjEeyuVbCrJgB33L5YA8Xp6mQ9m337P6y0Ih5zexR098gsLjGBY719LPHpQtGGHMvJchhkJ9zS5kFVtQM1tQ40t/Dr5dm1DBhL/hCTGtP962vq0dbmUcuoD7OycOVMV/9wKIJum4+FZeLg3OK5dPSynln8sZA8IOZxkQePd8y1OkT837CiMFBM91aftfhR8Ip1tnjB4oHvFg/MKyi2dkjmtkMSjfUxFeVuVwRL8v14apmX2RI/v96H7J0ytlUqzN6XAG9bbxjV7SGUNlAYpiDae8OQAjFm/0tSX7IL3lQaQEY+d6i1vSqALfsDKGtU4PJFmJdpshGmCZlCN5GXaXq+s0pGab2CjXtlbKvgXn8JOBRUyMjZIaHHrYVvstp3bu1r0Ss5vYgHyet0a7uEbrsMjzfIpLaCP4U68ELQj6TCrW1e5Be0Y8OmZmRkHUDm8gMg8Ju57ABeW1qDjKxqBpS9viBK93UjO6cWZfvtIHteTW164eMoJ9Yi3AeYAAAgAElEQVSfJg8OxJ0OGbYOLwggi/8S309OfzEJW+9b9BG8QGeLH2bHD0w6rKMXA8V0H9O+Z8C3r4/Nt2yTT3UcyP5XVTAtemv0svjP6n9Wf7D6w3dhPppXUKyvkHWt3/2ghSxfzPJJke8iO31R1LSFkVum4OerfXhsqQfP5viQsY3iC8soqeNqzwSaW+wURziEypYwGm1hJhkmz9HhSAxuKYqS+gBW7fAjq0DC5n0kJQ6gsEph4JkkyWIxHw5H0d4Twtb9fqwvkbCzmoBxAL9b78HizV7m3dothZFd6MPGEgkS2SGrdp7iLOpitbG+ja3rufADhV8qLXfAZpdBTq/0i8W5pDO3d3k+ciCMji4f6utdqKrqRVWNAzuLOlGwvR1l5Xa8llmNpctq0NDogtersPBLnV2Sakv83Wjn5gYHmuodCDJ1bj62zI0W3416WGW22sHiAYsHLB6weMDiAYsHvgs8MC+g2Nrhmd0Oj9hhpoaXg1Gm7vxqvsSA8BNZHixa7sGLuWRLHAB5jK5sCaLDEUZHbwRNNu5gi87dbvI0HUEgFIUSjqGlO8LezyyQsLLIjzW7/Vi724+yJoU54RJepAl4EMglz9XrS/zILfVjV42MDbtlPJ3lwnM5Luytl7G9IoCsfC/qOoJMEk3lpW8FKOY2jNpC3Gr/2bW/6PAWvfoQjURR3+RhkuIoOa7R8Zigj8Zz6ScL8f5M9NWnJ66pHATKG5rcaO/0Mbvh3SXdWPxaFVavrYfNRvGAvy0bYl7vZPUj22YKz0T1sGyKrf4neJ/OyfjFeq6NIRZ9rP5i9QerPwgesMYDazwQvEDneQHF+gSta22wEbRgYFgnbQ2FY7B7otjXROGTZPwsx4Onsjz49XofMgpkrNghI69cQWVLiHmaru0Io8EWQqs9jE4nSYl5vGIWt1iOYk+9wtSms7b78cZeLmUmUE0q1cGQKiUmW+FwlIHsgnIZ63ZLyC+XUVQdwPICHx7PcODpLCcy871YnOfB2mIJssIX3R4/qWpHdBI9IXlOrKuos3W2aJOOBxSFnFxxnowDVHXjJdW9SI+e07X+PfEs1ZneJfBLnpsDBCiVMItBTKrbdrsfK1bVoqi4C5IUZDbHO3a2Y0lGFUr3UaxiJb4hlCr9b/p/RQmho9WNUFCodFv89k23gZWfxXMWD1g8YPGAxQMWD/xweGBeQbG145Jix0VdxJPElkIvdTgiIKDb7Y6iqjWE17b58ZsNPnbOKpSYPfGWcgV7GoIobVBQ3hJkXqebu0mNOsKcbnn8ZBscY+rR5EwrY5uE1btkbNgjY12JH7vrFKZWTfFgSd2ZQIHXH0V5M7cfzi2VUXwgiG3lMlObfjLDiX9d6UZ5k4K8fX7YXWEQmCf7YgLjXpnfc0DCQbHV3ina27IZY6A1HX84XApc7iCLtaup4/chEo0x8EoANhbr49fqhg5NPMlUrQU4pufimvMpDdQqgGbfxuB2B1Fe6cAbuc1Yu6EZK7LrsCSzBi+8WIEXXqpA7uYW5oTLL4ewb183tuS3wNYtsXS/SxoSshyCrcvDQL6oYzp6E22s51Z/1S/eLH6w+MHiB20xb/UHqz9Y/cHqD/MKivUMZV3zBbkAkXQmyW5VWwjLdwSwtiSAfU0K8ssVBoIzCiRkFvjZQZLitSUycssCzCv1rgMKkwYTOCa74vqOMFrsIRZSiWIMF1TKWFMsszBMZIu8sTSARluIxRumBTPlTR6u2+xh7KgKMLXp/P1+bK/kEuNfvu7Gj5c6mJS41x3G+mKJxUCmEE7kxEuSuZSNea9WJd5iIW61szaIWLSYHS2IHyUpDLsjwKS2gkdJlZm0GZRghB2k5UCOseig8En0jECxAMaUjjiI9uLafKZn9H1LmxdrNzThpSVVeP7Fcrz4ShUWL6nGq69VI2dNA3Oy9cKL5ViZU4+OTgkkjaWDvLFTmjPxPH9nZhrM9r3U/KQCfaZ2PnO5Uqczc1mtby0aWTxg8YDFAxYPWDxg8cDDwAPzAoqtHbZUO2zagpVsiG3uCErqg3h1q4xXtkjI3C4hY5uMZYUBZBbIWFrg58d2GVnbZRanePVuGev2BLB+j4z1e/14o1RG7j4FBVUBFB1QmMfobZVBbCkLILdMxvoSGcW1QTi8PC6xWICTDTJ5nCaV6fzyAAPEefv9eG2rF4uWufDzlS7sqJLR0Kng2Ww3cnZKaLGFmbSZ2VSSxE3YFZMXa1UCR53Eav9U7c8HUYs+ifQJEdANaeFJiE9JK4E2b0i1msImkZozOcWiuLwMKLNvCBQLibGw9TUCYuJJSk8cdE9AmlSmKRRTe7uPhVvqsvlg75FBNsRNTW5mP5yzuh5/eLkCq9c0gpxrhZn5gciHg9F07Ul5pntOZRHlE9czvZ/4nNKg+nEtEMumOJG/BG3pnEg/632LPtoC1+IPqz9Y/cHqD4IHrPHg4R4P5gUUC2ayzmJgUcGwutB3+yOoaQ+hui2E/CoFr2yV8HKehJc2+/HSZpmFY8os8DOv04u3SPjDZok53Hpxk4SXcvn1Cxt9+N1GH17MpUPC0gIJ2Tv87CD7YZImb6sMMK/TJE32B7R4qgQ0Oh1h7K3noJhsiguZxFjGCxs8eCrTgf+e42bhm8jWmOyLn13pwu5aGZKipUOSOkqLFv4EkK32Fu1tnefKC8xRVJCH+xLglYAr8ReBYYoTzA4pzAAyAWVuCxxhIJlAdTrJsUiTzvQegVsOrrk9MYFuOjweBWvWNWJ7YQecbgW9Dhl5+a1YvKQKK1fVoqWVVJRF2KPkPK/PS39NNKF7PW3Ec/1/c7/m0nJJUhjYn0mCPff0LX62aGbxgMUDFg9YPGDxgMUDDxcPzCsotnZYxA6LCopjMfgCEeysCTKpL6lMkyOtJfkyXiVgvMWP32+S8LuNEp7fIOGXayX8bJWXeaH+8TIPnl7hxTMrPOwgz9T034+z3Phxlgs/yfbgtxvIMZfEPE+T7TEBXQrjZHOGGbggCRItwv2BKOo7Q9hZo7BQTIVVMopqAkzd+tdrSXXaiZdyPdhbH8Rzr7vw46UuLNniRXtPBKRCTQfFK27u5irbpAZOUj3u1VrrMFb7i/bnNLHokZoeXAWa8yddi/tQKMqkw14pCJ8/BI8UhNujqJJjFRgHIwgSD4YpTjcHxwR8yYs1P9SNGxWUUtoEiP3+EHp6ZbR1+JgqdV29C8UlNvzh1Uq89EoVA8f1jS44nQqqqntRUdGDnl4/A8UC3IozTZTUvgLkiv/19+K5flIVz8X3+meCX0Ra4l68o93z+nGaaWON9tziv3T0TaSnRS+LXpZGgTV+pJ6vrP5h9Q+rfzwc/WNeQbFYbDzUZ1WKSurFFB+4tjOE59/w46c5En6znqTAfryU58e/vSHh+Td8+MVqL57JJsDrxZNZXjy+1IMfZbjVw4PHM914ItPNznS9aBk/nspy46cr3czrdEGlgn2NCnbWBFDXEWJhmNjiOxZFOBpFjyeCiuYgtjNJsoyCCg6KN+6V8avVbvws24Wl+T6sLfbjv+e48ItVLmTke5k6NYVsyi/zs2tSr36jREJFkwKPFOHAWFUHfajb3KKBQRo6G14Q4JDOBO6YNDdMNsQkKY7AK4WY5NbpUuBwKgwYE0CW/Jr36KASQShIHqW5yjWdRToCMIp8lGCYqUcTCF69rgmZy2uZg63FGdV48aVy7mjrxQoWn7h0XzdTs/b5FKZyLdIQZ6qfuE52FvXXPzP/J+6ts7apZtHCooXFAxYPWDxg8YDFAxYPfFs8MC+g2NpB0e+gcHVJWhB75DDW7ZHx5DIfFmV78YvVPjy/UcKv1vvxk2wvfrzcgyeyvHhsqRc/yiAwzI/HMr14VL1mQHi5B8/mePCb9T5mi0z2yM+t8eB3G3zILpSwtyHIVKPjUmIWV5V3KrJlbu2JoKSWq06Tp+rCaoVJilcXS0xN+mfZbvx2HYVh8uPlTR4mKf7pchd+vdqNjC0evF4ksbjFL25w45VcD3ZUB+D0cSkySeas9te3v0WP2fID9REGYiPkCIsDYqYmHYjA4wkye1+bXYa9N4AeRwAEkCmEEgFj8g5N6tR0MLXoOECm0GEkQSZJMgfcNLgGAmE0NnmwJb8NGza1YGNuM3aX2JikeHNeK8orepGTU48X/lCO1zKrUby7Ey53gKVD35sB7nzdU9qzpZeYJKz3rf4meMHiH6v/WOOBNR5Y44EGIq3+YPWHr9Mf5gUU6wtgXXN1RlI5brGH8ev1En6U4cXjWR4sWu7DM9leLFrhiwNhAr/iIFD8GJMKe/BklgfPqGD4+Q0+UAxish0mJ1kU25hsj5dslbB2tx9lTUGmNl3TRqGTuGolhVMiwOqSKOxTkKlWbyInXaVcUkxOuRbnebEoy4mnMp14LscDAsxFNTIIEJMaNQFgAsnPr3Xj2RVO/HS5Exl5XhxoUxAIcRVqIZGz7Bq1QdnqAzPTQoBKAq/CXphsfAnwkkS41xlAp82Pji6JObxq65DQbZfh0IFjP9kFB/gRtzdWuLdqkjgzu2OyPSbpcyAMt0uBzeZnjrY6OyUmDa6tc2JbQTtIKtzZ6cO27W1YklHNYhRvL2yHyxUwAGJRbvM5onqoNv9P94IfzM/E/9Z5Zn6xaGTRyOIBiwcsHrB4wOIBiwcWkgfmFRRbOzR8h4ZUp71yBHkVQSYJfiSDJL8kEfbhsUySBnvxCIHhTPWcwSXGT6/w4dkcL3651ofn10v4/SY/FudJyCmSmafpAx1hVLaEWLimjHwJywv9yK8IoLSR1KeD6HJGmMq2AKiRWIw52CquVZC7T8bGveSIS3Wytc+P323w4vEMJ57McOCXq91M2kwOucgRV16Znx0EjhdlOvGTLAcyt/pQ2azAJ4sQNZoEjfK02t/aodMPVun4QYBFOpPTLZ8UZFJgkgaTVNjWLaO904+2TgkEiFvbfOjo8qPL5oet288cYlGcY7I3ZtJjKQRZ5l6rKT0Bkimer0RpexS4XNyRlt0ux8HxtsJ2LF1Wi6JdnWhr98Jul1BZ1QsCxCQ99ni5Myu++SO8XqsbT7qYyvSc6q4Hvnpa0LWgh6i7uBfvWfdW/xG8oOcX8Z/FHxZ/CF6w+EMbTwVNrP5h9Q/BC1b/sPrHg44H8wqK9Qz5MF8LW+JfrZdAgJgfHg6EVbVoBopf8+BHmR48tdyL59b48MJGCS/lScwJ12v5FK5JwsoimXmVPtAeQltPGFWtIby+S2ZS4pU7/Dw0U42C8hZySsTtfFk84RiPi0wOtgoqAlhfImF9iZ9JmosPKMjfL+P59R4synLhX3O4VLiiWUGvJ4yO3giabCFsLvXjJ8ucWJTpwK9Xu1BSG4BP5mF0aGFPUj6SyGnSYmsH62Hm+werO3eC5ZPCTD2apMEkIW7v5EC4udUHOggUt7SSgywf2ujo8DHATO/FQbJQr2YAmUByiDnXIlBMgLi9w4fSMjvytrbhjU3NWL22kdkV/9uL5XhpcRVWr61HyR4bOjq9zP6YpMQkiWYq3uTBWhcnmXhef+jBsP76wWhi9SOLbhYPWDxg8YDFAxYP/JB54OylQfRfOBrXJvsh1/X7Urd5AcUPisgFkX4433MJUq83ipW7ZDyWZZQIc4kxAWFhM+zFf8v24vmNfizeKiEjn4dlWsbiFPuRVegHxSkurguisSuEJlsYRTUKU6V+RZUg5+4LYGu5AgrDFFB4/kRXWphTrOLy5iADwgSISf2aPFRXt1JaIWxRgfHvN3qxvEBCab0ChzcCpzeGblcYq4skLFrqxI8znQwUlzUo8CsUniYG8kDd46E4xgTEeX5cQm1JjH84/Mwn5IWuD4FNUoV2eRSQDTFJiJtafGho8qC+yYPGZh8am73sv6YWL1rbvGhopsODugY36D+S8nZ1S+hxyExyTJJnAsXC5pjiE5eV9zB74uzXG7B2fRM25rYgN68FWSsOsPuXl1Rh8WtV2LylFfvK7LDb/Uzt2gyKySzBCIiNEmTR/wQ4Xmj6WelbEhIxl9LZ4geLHyx+0MCk1R+s/vBd7Q8D48DU5xcMoNji12+XX+cFFOsZ7uG71mwGCRQGglFQeKSf5vgMEmKSDAs1ajqT/fCi5V68sElCZoHMgG7GNj8DxmQ/TICY1KM37JFZqKT6Tg5w1++RWYzjlzdLeL1IxprdMt4oldFq5xItsRAnaTWpQu+pU5gdMYHizWUUzzjAbJ3JAVeLPYTsHRJ+v9GD5dt9rNwuX4TZJfvkKJMuP53lxBNLevGLlU7mtZrAst0VQk2bgqoWBXaXli+PXaynhzYxPXx8YdV9tm1OoJPsiV2eIAPFJA0+UO9GVY0LNbVu1DV6UFvPATKBZAGW6Vxb72ZHY7MHre0+dHRKIPVokgyTnTClS+rUTlcA9fUuVFT2orLKgfoGN8iuuKXNi9VrG9iz4pIudr0suxY5q+tRU+NgsZJDLOyTFhdZgGQBjPm9Fs9b9EHzebb0sN6z+o7FAxYPWDxg8YDFAz9EHjiIoxcGMPLFOCanAXw1hYnboxi6dBaHk0UyOT2A0dvjGE913BzAqWTfHTqFS9dGMX53CtP31Tw+PIWDyd5N89/h81TWCUzdn8bU3XGMjQzg3JH07TL3bw7j3MAIxiZ4WSfHxzAycC45PdKUdT76y7yCYmuHIwa7J4wVRTKzIX4008ecaFGYpaeWcRtisi2m4/EsL365jlSkCfzKyCzgUmKyFaZrOlbskJG3P4DyZgKhIew8EMTyQopr7GPgeU1xADlFfpDTLJLwknMtAua0GJeUKJq7IywM0+pdEtYWS9iyP4BdNQo6HGEooSi63RHsb1Kwagd/XtkShM0VBgFikjrvrA4w9eknljjws+VOLN/uZfbG2yv9TJW6pTsMr5/bMZP3YGJIylswpsUP3+6O1/eF/gQuGSh2B5nqNEmBq2pdKK9ysqOm1oWaejcDytV0XetGbYObSY/rGz2ob3QzoMxUrNs8TPW62+6Hwxlg4NjlJnviAOjsdMpwOGX09MjM8RYB8IysAygq7kJ9gwu1dS5UVveibL8d7e1e5uWaOewKa16tBSgWEmNxL/hfgGHzfar2EO+nem71Jz4BW/SxxhPRF+hs8YPFDxY/aODE6g/fg/5w6ByG7hASTvGbHMGFQ6b5jsTJ6X5TI7hgHg9PDGA8RTZTt/px3Px+0vtDuHBjMnnO05MYfv9QfK2vjccH039z4WD8mzi/HrqAkVTZ3B1OpIcKiuPfz/P9vIJi/QD1cFwTABQgkMcEru0MM7DLpMKZFHbJh99skPDLtWRfTJ6mSUrsZSGZXtjox2vb5DgoXpLvZ16lXyVV6m1+7mCrRkFNRwgVLUEGfsnj9G/X+5gUeXWxjNd3+ZmDLFJnZotrFZg6feSUS8H6krO48E89X9/Gh+4wc8hFNsikAl18IMBAM9kMkwSYALZHCqO2PYgX1rvx40wHnskiZ1wuZOZ7sK3cjzZ7mKlQk2ScgHRTVxAuXxjkdZvAwg+3/fsxdn8a0/FjDP3ztXN1fhgT8XSnMT0xjHPzlfaM6VzAyJSeT8YxMOM389POxLckzSX16U6bjNpGDwPDZRUO0FFR7WQguaLGxe73VzlA4JikySQpJmBM4ZaaWzxM8kv2xp1d3NaY7I27bBKXCreSlNnFwG91jQOl++zI39aOF14qx0uvcJviXcVdaGpyM0deXl+QqU+HgiQlFocxFrKQFtNZgFv9tfiPzj/cPjE/fGDR5/tLx/4v9GPiNMb6v791sfjQajuLBywemH8eOIhLoypS/WoK49cuYfgOMDXaj0vXxjH1FV9/Td8ewFHd2uvg1Qn+YHIUg1cHE4/+C4b3Y7ELGLmnruWmxjF48RROnT6Fi9cmIHDyxGfHZ1yPHB0YV9+fxuRIP86dPoVT5/sxKgDs/VFcUgG8oNXcvzmKgdtqqaYnMfLxOVbWcx+PYlLQY/TSnKXbojwPcp4XULxQiF1USEvftHCfHMbZpDscyXaMDuLyLcESxDDaol+kb96QGf+UDwzieWJ5oojG3jOBCZUZ9afpLzF15xb+8ff3UJRHUmMfi1n8Yi4Hv1nbZSwtkPHqVj/ov5fzJJCjLVKVLn/rBgQP6pMU11P3pnD78+u4cvo/VVAcRSgSY9LgXQcU5BSdSQDFl/0RKKEIsw8m9ee69iByS/3YVRMAOdvqckTQ46YjjKJqGS9u9OA3a1x4OdeDggo/6ruCaLWH0GwLMhBNjrxy90qo7wyyeLOajfFM9PsGn386w27bV9OYnprE+M1h9L9/zNAJje0/AGNK4xjoS8Zv2qRi/D6NhOO9ERhwabIdwFnz+1zzN/WtJP0jkf95HuZ+I3gz+TkZvShOcQReKYQuu4zaBi9K9zuwt6yXnfdXOlFW6cC+8l7sLevB3v297L6iyonKaidq6tyoqyfbYg6KyfkWgWJy2kWq1PYeGW1tXib93by1Des2NCM7px6ZWbXIWl6LVzOqsWJlHTIya/DK4iqsXFWPwh3tqKrpZV6rKaRTKERgmAPjeBzkuLMto00xbQqZwbAeFM+aH9SJ0Xpf37/MfJqcy+L/jg9YEsX5Gp/S8KN5DBBzJ40ZFv/q+deih8UPFj+ItcTDNT5oa8exj7nElMZNYVN88MNRFYROYOiYtn678DlfFU6NcNvjmfrPof4xPv19NYHBE8bx5ux1FU1Mj+FymvE8FjuLYfVVKp9e5brvr1cwpkKpyWundOO77psR7RtW3kOXdd+c1gD5mWEV30xhRJUii/odvDKm0mMSQ6c1eojngofm+35eQLEo3MKfExdE45/O0nNbnPhiuaSBYlHudBO7eMd8jkZnAYpFlnSeuoV/b/DiubU+5mmawC+pSpOnafI8TR6oX8z1IaNAwpZyGfXH0oNiQ9Jf9OPUwShTf67rDDPV6Tf2nMWlu/q3buMjKQInsx2OQApEIckR5nWavFu3dIcYuLU5InBJYeasq6w+gB1VMoprAuwZSZPz9vmRle/BkjwPMrZ4sHGPxGyYw9EoA8bfOWnxTKBYTyIA0xMjuHxKU/XQ2l0b2PgniXykvat15Fn9lwIUi29PDYwa7EqGPphj+rrdR5Gmdjb3rdnXy9xvTKQ03SZLl3tyJmdb9t4A6po8KC13oGRfT/wo3mtHcYkdpWX0nx2799oZQKb7snIHsz8mqTHZFre0euO2xWRLTOGYyH54+45OLMuuZx6nCRRvzmtl3qhzc1tQVe3Avn12bMptwcqcOry+th7bCtoYqA4GIyBpsTgIJAuVaQ6UhQd2AsPi0CTHAiATrfXgWKP9fLbjdz2tixjS20alsodKyatmPjWxl/l2fECbgFOm+V2n2bdYPrM927WLSelpHgP0oPjh5PNvsc0sPk/KoxYfWjz5rfLAWR0AfI+3BQPFKtiNxQ7jOEljT5/EUZ0EduA2n9Qmr5+aBV8fhBiLp0cvJb5/SFu/jl1Jww/xtagRoAv6XRBqhZPDmj3zA3wTT2diyCTtprJpc/3s6p6mPnMYE+cVFM83Yk9MTyNSfO2jE+Envi925I7GGSX+HUnCTDvogpnEO+Ofiu85sZOlPytJsUhQnMc+Q+MbJCWWQSrTS/Ildry8xc/DMm2WsKpIxp6GIDpOzh4UU/Jfjg/gL54IyhoVbC0PoLDqXfQbRM238ZEvwsDv/uYg7G7NURZJeMljdS2Ff7KHmBo1eZd2eMLocoTRaAuhoZOHavrV6y48u8KJF9a78HqRD+VNCuRgDBQOKhLlIKGvz8ikyegnOhmdF/T5HEExa6rpcQyc0OrAy6cNKrw5NZD3tct/IbmkWNBI7BgKNpoNf4pvZ6avuW8l9o9U9TP3G1G+5Ofk9CKwSOCTHGRRXOKqAy7sK3cy4EuAuLC4G4W7uhkwLt7DQTGdi3bbsLvUjtLyXlRUO0D2x7X1LiY17uryo6dXZk62bN0SGhrdzKt0yd5upm7tcARYHOKt+a0o2WtDZ5efhWIi9WubTYLTGYDfH0ZQCSMYDDNpcSgYBR0EjLnEWHPARYCYq05r4FcA4mRgOBU9RZv9MJ+b+o+qDTEzf4rxwcynybks/q8OFP8w6Wken+b5Pr7QUSmagp7mMYBAsUXvmdcPoq/Pnv/nuX11i0Wrvaz2svhxIfvXFYwJleDxQZz+ax/DJeklwAcxeIePveOfHgGFcCIHXaTROPDRuwZHVLz/avNj8vXhUQyp2thTN84bQLO+/8dVtlPNz3GptraeOyTUvFNpOF4SknD+TV/foXjdiAb6/MV4GFc3v/3NaXzNKyjWd6iFudYaPL7oATCTfrymlqD/SmtMUdZkE7t4luzMF7xmSfE9XDx4DPKb/8Cf/vMk/viPT/HBzS/1GQO4h3cDEjIKyMEWV5f+Qy5JiLmkmEDy5rIAajtCsL1jAsWTt3Dm1Cc4euoqLg3dwT21k2kZTGPkvT+hsErGmmI/A8UDwr6AvXQbH3kjONAWxKqdEgorZYRUe0gCsw5PFJXNCgv/RFJjAsRkY0xep9t7wqhpVfDqZg+zMf5FtoNJivfWUfxi7n2X6CSAgGZvrQ00yej4jfxnBsV6+4zPRjA2PonpBFoCUFX0v5Ey6hYoyfKL76qpjT2/0hhz30rsH8nKRP+Z+83kaBK7l7gtzCVm8pCYFldP9niD6OqWWSgmkgCTujRJiHcyUGzD9p02FBR2MjBcWGSLg+ISAsZldpRXORigJqkxqVMTwO3oktDeIaGtzYOWFg+aSJpMqtatHvYOSY0XL6lmYZl2FHWirs7FQDGFZCIpc0BWgTGBY4UkxlH1HEE4xEGxkBwLG2PRB8znxHp/B/rGDHw3v2VODYpnl4+ZT6cxPpyG3y6dNUz8s8vjYWuTNPVNA4otWqah2zfap6xyWLxo8cD3gQfOXtNJqL6awtR9sim+ZJAMG+uhzXfTeutPseCfHMVlneAmFhPz6xRGVL/lqbgAACAASURBVGm0Mb0YhOQZty6nnBvja03dJqghnfNCgDOJYVW1ee7faHVLtZY9d0M1KNRLpBd4bJ0XUJwM4esJOH/PNSIKnmBn0o/XqRsY89P03A3fJLGZNC/uRUMZ09N2wEkqSurT142GoLh0MIhORwTtvWFsLA3gZzmH8bcRY+5Xj0t4bZsfWQUkKZbx/AYJv1rrYyGalu2Qsb1KYaGTbCdvwKD9/M/r8G33s1BKpQ1ByEeTeJn7/AqKSyhtCUU17+JTEyi+7I/B5gyzuMQEnr0yB7K0oCfP03UdFMc4yBxtERgmaXGvJ8LCL+2tDeD5tW4synTguRwnU6HefSDAwDN9T16og2FVUsacC1HcYhqwNUdDqegpeGZBnptBsa6zx/NL4Rlw7Mp3Ywc5PuiorDQTf86Nnua+pYHiOH3Uwch8n9hvHoReXN1YIhXqngCaW32orHExdelde3pQtLsbRSV2Dop3dqOgyIb8wi7s2GXDnn092Ftmx559dpSV96L6gAsH6txoaFRtjBko9jG7YvJUvW9/D4p2dWLrtna8vrYBr7xajd+/VImXFlfipZcr8epr1cjOrsXatQ2orOqBzxdkjsDIGZgAxUogzMAyncOhGFOnZoA4TPXQNoYIFFM7iDNdm+n3cN2LSVtlYtpVNmnspKeHmU/JFulB+E1bRKbP7yFvrySg2KKXxW9ibrHGs4d8fHjo57O5tv9BnL4ygon7RjxAd1MToxg4f9i0PujHqHh3ahwjn/XjgzMfoH9Yc8xlFNyI+VUDxebxOq5xqK6Bzc/pPr7WTOmTQ5ePOv/qv0k+Pui+YYBdm8vFWjZhPBHr9lQS6wXgv3kBxXoCLOy1RkQzSwkVBHP+mrG2+Qtt0S++SVzcawsn8Y7xzEGxEXRO4dNjHEiWNQfxu41+PLbUi8b3DcgUV49xCfFLeX68soU8Skv49XpuV7yySEZJHQ+d1Hz8BgzOo+8Ow6/GJ67vDDFvz29/pupDiCpODsOT78NrW30orX8Xn+g2p4Db+DgYZbbEBIzrOoOwu/6Ekx8M4fM7U/hSJ9S+PzWF0aF+HHuTS4tJalzW8B/496siI36+feUfDDCT9+lWO20IhPGWUKVQXx37+C+GXanDZ/oxfHsy7nGPXiNHV2NXk8UmE52J3prE8JkYYkfOYfDmBKZo90wHcI3to2s/0bnU8qT85tAlbRAS7xp21Mw8mMhHVIaDRy9o5RPpTE9h8vYwLp+5GHfOph8MtF0+9YP4QGDOUyQozollYHHiiL763UXK/4shXEpqK23OIzHNVLSde7/R2iU+kNKkcOM9BP/6Lj66dguTgg+nv8Tdz6/iLcWPXSV2kHS4YKcN23Z0YUtBB5Mat0Xfx6Ub/8RdMXkQWe7fw62hfhz+6yfgzg2ncPUtD3K3tGHxazV48ZUqvPhGH0InP8PNO19qdCKneOM38cHb/4nq3GZUq862CBALUEzAWHnzLZz95CbuxAtK8QanMfXP27h++QwuCpub8QGd5oRa70NHceHqKCbu6RqHvh0fTmrHbqTRBc77X+j6DmvXwXjswMPnBzF2V7dTNz2F8eHLOKXbODS25WGc/XiYxTMUHMXqcncMg+cPG/otfZesPAMj4xoNU9XF3AfjmfGLVGO4saxmPtUmf+N7Go8l/H/kLPqHxxj9mXYIOdm7T31zhC1KEt43aUNMDpP0+TDOxdswsa/MbXwTfEFxJccwaei0fFxM2n5HzoHobnifaH93DENp4lEePHUJQ8Q/Ovbj7T2Okf5zcacq+nY2NRW71Y9d5nf1z2J6YH2PwofwmJTjOv6fvjeO4SupY2g+2HiahgcWWNqQjIes/6z2sHjA4gHOAwdx9Owl7snZoKE4jfFPZ/YMTWkc1JnbafbBYq2cel6MqySnWTfHx/NU7xwbAkcdWj5z/0abyw3zhW5sPiqwTXwtvPD8M6+gONmOg34Q+PrPNSImTNJfjWPguHkHN837X9OmWEh+fPJ7+NAgyp3CtTMUpiiKtXuCeGK5F48v9cH1ib7Ed3CiQ8birTJezpOZHfEvVnvx7Otedr1mt4zyliA6nRHUHE0ExcESGQWVAbh8qsqyftHBsrmFv2z0YFmBH5Ut7+KyGRQrUQamyUu1xx9F5wXVW52+iPrr6bsYOv0nOL0RlNUr+N0fR2CA+OND8Jb7kb/fj4rmAGTlz+hXnQOwZL4aQ/8hbUft+KfC1bs+E+2aXNIb46iJjs7fmb41aozBlmbHK85/5gV5yh2wGM4Kt3uiSPdGWGgkzr9mntIWxIK/j1/W3MmLJFKd9YNBX5+xnogPBOY8zanxMoj8Y6eFQwfze+KeD7zx99kgZM7jwW2KRZ2M6WvtL9qEnscHUla0aSRVEaJn07fxrq8bRbvtKCjswnYCxzskhM/dhh4LixomnqcweFRC9QEnivd0I29rG/5+LfEtwz//HMY7YYp1TLbFIQRVteng2x/hpqHPG74y3qigmOrM6JEmJh/78KvJBC+MRhoR+DZmEb+bHsfITUNnjz9iF3cGWb/S0z8WO66FRDC+rd5pk7RoT2N50rSZ6v0ynp+5D5ryE1444+8nlSCb+VSblEX50n1/+IOReKgHU/bx22RxHA0bP/fHMCrCSLCvjP1v7uNbDLET/RjT7WPECxO/mMLI+/r57VTcO2j8FdOFGEcN9DgziAk9GDZ9o405McQlCuZ31Hu9zZqRJwAaA+LtYZqfKJxdqt/EoLYgFN8/+Hiqp5euPCk0XkR+BnrpFmjWc4uegjfobPGDxQ9flx9oXqHNYLbpd1tMABMYPDYb/kpmk6utIVNpOIq5LN18m0xt2cDv8TFdW/+eN6k6G96n/hIH8WK+PB8PHUWOtBLe168P06zXRRsk+148m0t/nVdQrC/AwlybF0TGqZW8reldhx8XuwzG19Q7rTFFWQWziNfF4l48N58JGO9vPo+zBlHuFK6fjaK8JYznKDbxlkOo+9so9BhxaugKSvL8WJwvMXVpAsT/bZWHeaRevMXPVK4pNnGLPYw9h2+oOzJqqe4OI7RXZrGJA8EIV0uOM6go+S38eYMXa3b7sa/pXbxvWMBzSTGVnQ7yFn1Cb+cgkjCf79/E2UiY2RsvyTuL84Y0/4l3JQ8KymUcaA2iJ3oFw7p1z/Toh4iKxUVCWc0Z8XujnbjW0ZO+nWo3S+RJZ/OCPN03cScCam5fae7r9R7x+FMTHyV4OU9a4vifRh4z1XOOoDjOn1dm2ORguU9gyGCLYu5bpnrpaWm6nmu/iZfTJHWMEyXVxbX3cWA3tyneUtCJ/X8fmSUgpgTv4erfKU6xGw1NHjQ1vY0Bk4JFsmynLh9H2dYW5oWabIkV5TSuGXg/2Ve6/3SgOBbTxeTTvZJwqXMeSLQyA46E9+fwhwgFIdrggpjI0qZh5JU5ledWvzYmm/ugKc8FlxTPcuyhYrGFio7PzTxuLLqur8wyD+P4lsrEx5SLGiKQt93luNMW41vGO2M+MVy+ZXyecBcfc2bmO/3YZeYJ/TODpDghQ9Mf6gaq4M/Y1xpPF16qEC+njles/yy6Wzxg8cBseIDmFW3e07QUDeNnmrFFzEtaGtpmaXKPzdo6b+JqsugqaruJNaRh7au1aTJHXLEH+CY+Hxm0MUU+mpMxrX7i2cKd5wUUzxdCF0yUOj2tQdlUemfUpOrKF2/se7Ma7MSoaSdeW8iI/ASDiWlaMKZ4zqXDqm1sNIZuVwRZhe/i5CwW1yLNO599AlsFSYn9+EOejOfWevFMthdPr/Dgl2t9zM54R00QTbYQc7SVfygRFPeVB1Hfxb1Gk63uIfNi8851tK/3YGOpH/kVZ3HOANpv4xPV5pHoTXbRH6kYampiDDcGr2JgYBjDE0J/VZQc+OLyn9DWE2Kepm0Xjchg/PJx7KqWsW6XD5uOjWof4UuMstBBBMLfxlXVk554YXp8EO8e6UPs0Cn06+NI6xZnCRJU8bE4mwCuaC8DP5lppPsm4f34jlY8AwzEByYTDxo0Dg6i37TonL47gv6zR7gK6pGzCVI54jEt/+SgWDxPtfAUz0V9T6ux6KbvjmOUOSG6hpFxsQsp6qQNxvz7JPVS62xO33xv7jdaDuYrLtXTf2+uE31B5b516zb+aWbB6c/xdokdO3Z1Izf/P3FRv9ME4O7wZfzZraC0zIFW70kMGHjtHgbf8qC9w4vubhmdnRdxjTZupidx4+MBHD98AX8/M4xbCWQaxXs9TkgS2RVH8P9dNm04fDmJkStn8DcWs/hvODNwW42tp9Zdx2exD4T3RUGXKYx8eBSH+vpglmCOf6pNWMkkdlMT4xi/PWEwPxCp4qspTI6PY5ycx8X/VC9uXdb47digccMN0xi/ehZH+vpw8FQ/xnQieP2klKw81Gbjt8cxac7wqzFcMUh8k/C44bm+P/CJT88viZtS6Rxtcadu/PujcU+XgiTTd4Zw8cRh9P31GC5cmzDRSouNSN+n53ExlyTJYzbjm1hMiIIxthzDsHBON0yq9sC4uohh9RHaIPd5bPXBq4MYIhV2sxYBjaNx+p6L78yzrO4OMZV6Su/g0ZO4+PEwxu6OYuCvOgmQGeSn2LE392Mxd9KYpEkIdBWkclNorgSGATQJx9cdT5Pxz0z8ZT3n/UVbdFr3uv5gSYi1+WOW6wOLfzj/nBIxgnVrW5pXNImtDhQPqP3v/DAm7k9jemo0SVxhbTNVD4Dj87PqJNZA//gaRHOQZXgu+Fsfuqn/kMGEqq/vmDaX6tcT+m/UOMxs/Gfzj25uvHUlnt5BsS7XbYbGy6Nbn2jq4Qs/Ps8LKBaL8YU/mxbu4wM4KoiqzrfTty4zyUQ8SDX7nxbjpm91jrZEuc0LH/3ELt4RElYlGMWW8gCeWXUa/5gtKJ66i08//ARB1x/x+1wJJCH+6SqPCop9zKY4Z5eMsiYFzd1hlDeHsOFPw8aF691h9FXIsDm56nT04HvGhQ7JxC6fxe83evHGHhlZBWdw2lC+2+hXnQHxOp3Ex0ODOPu3GAKhKNxSGD2eEHo9f8bx60ZUMjl0nnmh7nGHsS80gBu6NQ7uXEN7vgfPrXTgj5/pHkyN4D1yNkSS6YMDBok5dB2BleVQv076MYWR82JiNi2kGXAaw/DHF3HyqAYeRBslPZv4JKVNMQ305oWggVfS8ZFZesPVYPTlSbd4TG1TzOmQ/ltBqxguXk1um5qgOXFnUJPi6WLC8dYTC30tXX099NfmfqNrfdOlpuoqvjfXaeIqt2FlTt/6rmDEALL+iY+ivdhVYsN25VMjL90bxqG95GyrB6X7e1Fe6cDZz/XZc0kxeZ2uqOrF6o5zOH7kL9jychWyV9WjutqBVavrsbHsHQwZgPEUhk764XLJkP0hXPpCnyYwcfWtuJMtNjaYeUcnKTbXldumavQ1qO3f1uLrmr8z7PCaJWlfjTNTBUHfgx+bQLzOi2N8QhJV0kt1yWZJ/61qQkDpJpZHZ3ds6MOUsDb58jKZ+rJugSDKnP5s7n+i8MnOOh6O20Cp75mk8bHYQcRtrdRXJj47Gp+8E3j8/iTGhvtx8Z2jWh/SLQpYErMc38z5Tn9x2RBqQ9Dj4CHdWHdxMLkN7glh6yXoMYHBuC25JkVgT6dGcfGIxn8iH8M5CT8bnqsLYzNPGOZOcxp3BnX1SwS+2gLv646nM9RNLXuy+lj/WbSzeMDigXnlAd1cPXnjIhsDGSimOMWHTuHyqFh46NaNuvl0aoR/w8t0EOc+Exu5uvdpTNPlM3FN8xERO3JRwwp3BpPEBda390FcFkKqqTGdh+uDOHVVmD9OqwIv8d0DfHPoMsbUNd7UF5c1865DpzA4rj64P4qL3+BYPa+gOI7w1QrM/71pQcSkMNpuCV8GkPRlAOO6HXMOlE3fGiR8vFHNCx+9zRQxIq8PVzuuaQ/h3zb58WimGRRP4/qVT/Gntz/Bm2/R8RnevWGwwCXlPJwJxvB0NkmJPVi03IOnV/jw+01+5JYFcECoTjcEkfMfJlB87xY+fP9T9PdfxdDnSSQzuIO/N3vwSq4P2Tv8+NXa03jHBIqvBHgsYdE+TI06THFXD+Otkxdx6ZOr+Pizm/jCLF0c+wR2N/dCXddxBGf0AmHcxT/sLjy7/Bwu66o7OXyGSaMJFCeATd3Cn9NXs5GgthwfEDu0yRfSovxi4Ep7nwQUp3zfLLlRF+78/SR8JPhdNxgxXpy4Zhh46Pt0i8cEibgJMMR3ADmjIzl/igGK+PUQjr5zDpcGBjEowk6p37KTIf0k9YpLmHiaqehl7jdCasgkQSQNih+jGFDd94v+lI4eBIw/Mkje76H/kAMUfqn4mAHx4v7VswwQs9BM+3uxv9KBd2/qK8tBMYVjOlDrZMB6R1EXiktsqJf+iiOnPsKZ0x/h1MXPMWzoL8BnR2qwfn0D2tveNqlOT+Czt7U4xVSnqBkA6ECxkU606SP4W6Xv+yLMAe1sjeC8Sv9UNJqRH+n7QyZpsK7NE/gp3t/U8vxV/61mY56+POYQXdpGCC9v8r4s+EH05dT3Zj7Vt7H5WgeKTSYRZGqTwM/xOIpqOrod7YS2e0/MB7r+ZtYwmdX4psWN5LlqmwgJ5UvRH0nKe+7SAAavXmMxLCd1En6aa0RoDkrPuFnMc5y6O46Rq5e4xo5u8cHyT8LPoo305UvGE/HnKdKIPzeNzUIroe9rj6em/pWCfsnqk5r/dO1tpRffNLLolWQ8sPjD4g/DeHrIuP4jB4+EU8RZnXZo/CMzUDE+GnxUqBo2kwI/YxoTgzzsoHifNngN4/HUJNcaE5hoehwDJ7T0Ne0rvSAqhr5/6CLbkAPHiXGmsaQWk0m49eaqLP8TM3yj1k0/XhjqNz2FCYMGHEWX4JvBWv3U9ckC9a95BcViclm4s2lBpKompvYwTc0n7OFM3xqkf5zIxoUPdxairwuTBMWicElhrCyW8dRyHx7JOG1Sn57Ce30yfrTUg0cy+PFohhdL/zhqdE41OYKOLA+eyPLgyWUe/Px1H5bk+7GjhscIru8Mo6BKwYo/mkCx4Mik5y9x9dhh/GylG4u3+LA4T8JPs0+ZyncLH7rCCJK0WO2w0RMXMXR7KnmcXn0+X/Qzm+L6jiBqOxQ0nTWKza7940/4ZeQG4v0Vk7h+im8iSIEIlHiANH2iqa81SU3qhbS+fdJemxZe6STFhgGFijcxpAO3afgoxeJPXy5z2gaJSjzGnEoTHYChNNJ/qy3WYofOoZ88c4tBMBWJDemnqZduYNfXRVzP1G/Ee8nOyevEeYb6m8FhG+7hk786GOAtO2eUgN795CQLx0SxjUlKXFmdBBS/5UVrmxcdnRI6OyW8+c5VjN41akMkI9XwyVYcOOCA1/sujCa4t/EJ20yiDSW1zEl4gOqhTTzJckjyH0kyVbonp5Fo75naLXXfMbdbklLo/prA0DGeZ/rypAbFvP1TlycZfyT+Z66vrogJlzpQbOr/AngZ0k/SduK5kVYa0BTP2dmUR0JxTH/w8c1cH12Z0/a7gzjXP4oJbbA1pS5uzWU9jssjQsIg3hHnaUx8ppMsUP5paKKve1qemCkNE93ibTPTd3MZE9PSUvQl66xvU+va4geLB+abByhqgS5Sgxh66UzaR1dN4y8btw7iVLIwTvcnMJLSY/9Bno95DXhvDP0J0UfEnJxk7jlxGSNmz4xfTWNiWC+1NtHoAb45fnkk0QHk/QkMf6DTQvuGxvB5AcXfFIJPWFiqoLiv7wSGTNIdwWs0wfLypV58iPIbFz4aKBbPGSiORrGvKYLn1sp4JMOH/5pxGqcMeU/hrOIHAeFH2OHBo5k+/CjjMvpFodj5Lo52ePFklgeLVnjxq3USSHV6b0MQjd0R5n16/R4ZGX2zBMXTU7j27mlk5njwy9VeZOZLTIX66WXvJIDi92xh+BW+iI/pd3aoXF9N4fb1fpw5fgyhkyYv0zc/RmGljJrWIOo7g2i1X8ANvVRiZABOfeip8at4KxKF1x9BbZsCz8cmI1AWCoXCoSQ/Jq6fYzuNM0lQRfuIATzpvWnhpQfFhvd16hyiuTSbD+r8SfhI7FgJOz/xoQqm9eknWzxqz8XgpCagglbxPNm3VGfxnNWfvBvrJPWgncSbg7h0/iSOmtU7DeknqZc6CBnSN+dnCldDJZ9Zgq1JcJLVSZ+fsU/ew8DfnKiudaHiHYMYGPeHzqGyyomqGheqD7iZl+lzhlfuYfBtL9o7fLB1+/Hmh7eMTrombuL8347C3e1HzOiuHWMf+dFl88FuP4Yhgyn9BIbepn4USwuKOV+a6YuUfM/7wxj6VfonSHQ/1eiXlh/Z90l4SuVXI23JJiF5P+TlmcCwas6QvjxJQLEhjnDq8qTtv/EJ0UzHWcYpNml/kKQ4Ib8k0mR6h/jRSCsj0Izzq3mMmdX4ZrLzZSHn9O1r6t+sPCaJAzXdxCgGL51j5iQJZTXQX03vyHFmQzyagKpJJU6XfxJgGq+vbnxI1o/j9DVL0PV29lQfE93i4+3XHk/5Yi1ZeUXZRPta99rC1qKXjv+TzHcWfSz6zMd4cfjEKQzdAaZGrxjMANPxF31z+sxpnDqhgcV071PowBP0/ulT8TwS3j+vaqipplUJz1WfE5TGqdPHmc+R2dSfNJjom9NnTujMZZLNZ4KfDuLoO3OtX7r0Hmz8nxdQrCfQwl6bFkT6ydU8edPqnMI0qdKNpIvH+EKLE8+4mNBAMa8TSXuiDEzm7Fbw1AoJ/zXDy0Cx0WZ3Cu94CRBrkmK6ftR93WgDiXvsvaeyPPjpKi9ezpOQtz+A6rYQ6rrCKDqgILvInx4U06KLYvsOf4I/+hRkbSfJsAcvbPQiI1/Cc6s9eDIzERSfrArA4YmwsEzviXiqDIdxyW4kSvWMwHXaBIpHrmBXjYzGziDae8KguMTnDEafX+KeDjSMfXQY5Y0B1LVRLOQwQu/q1EMpv7tDmg2BqS2MfJRkIZ32fW2Cj6djWnjpQXH8nSMXMXzXYMSqxUWO52fiQYPGQaINnNHD80zS3vT1NC88DbalavnMdqJGb8Pp0k9XryT0jNPDDILM/Sb9t+Y6GSXn5rQp1rAX9U0e1Lw5CIMfrakbOOFws2d1DW7U1rtwPgEUe9DWTpLiM7iqj1x0qx/Bwg40NblRV++E/6L+IXD9RDNWr6lHfn4Lzhm1tkHefUlKnE5SLPjLqCgxjbHLOhtRHT3F++KcnkYztVvqNo+HXWB9H5i8poXDEXknO6cvT2KbCfVdnlbq8iTLK/E/c32NADXxfZX/zADra9kUp8hTLC5Ues52fDPyRaLn64Q6mTe3TLbgxnlMV9ZDBzX7Zz2/nRiCnuOnbvDNSJaveV41+CHQ+nZanjCnoZ+3qRymsTkuKY593fFUK18CDfX1t67jG0QWnSyesXjgm+OB/i+mIYQ/3xrd1QnD7OPkWyvPtzwezysoTrbDoCfs139uWhAZJtdkDjve1Q32pm9nYVM8OToI8urJj6vsfOnKVRx5532UZvnwSCaXFJtB8bl/fxs72t7GzjY6n0T330cwmKDmdgt9y31YtNyL5zdIWFVEUuIQmrrDqO0MYVuVguwdAWSZJcWTN3BW4QCd6EkAlpxybdgj43cbvPjXHA+WbOWq0z9b6cHjGWZQPIYjOyQ02cKgOMXGBdQEBt+KMsdBpF79HxdMKt83P0ZFi4L23jADuW5fFNLJ69DhYLEcBKZv4lC1F1vL/KjvDMEnRyAHLybQYWq0H2dVZ1lUH24fN4TR63qbv+QL6Tnxk2nhhclRrW2HRzGeIDHhVdGH+eL5JeEjtROTDa/Z+zTl03+G7+odOduPEf3qk0lVtQF4Jon4oasGlQTmOXn0s0EMDg8wRwRUPuPidBqjH+rSP6K3EWVxZ3AhPgAlqZeQgMfrJ3b0eJqC/kYeAoz9RvQfce7HBXWjKrG8RkBNz41pk9OrAFN/bmo+jo9N5Phy7FO8HY2hpTWK8NF+xP1WsKa8h8+O+dHZJcFmuwKD4v+tAfS1k1dqP4p2+eD/wNhI1441oqioAy0tbvSZPW19NYmbqvfp2JGz+DihgTWHWccGTQWensDQxePxXdS+viM4fvoiBkZGMHBW2wE1tqlGI07/JO1maK/kfYfG5b64J0rRbacw+vFZHFUdM8Vt0q+NYvjSbMtjHlM0UJa8vMDU7WEMXh3CtSundeO1lp+YQ5J/b05fx+8G/j1lsgcHpu9cY96nY4eOJvE+PYFBXdx7Mx8S0Ofl0ef3gSkaAkkBZh7fjpr5grrm+AgGPzqPU6fJJ8AQRm5PYmL4PKePCWQKqTcvz2HNMyhrVh19zg9j8h63Hz73Dvd6TrQ9/MGozuQFEKCYpWe2Scc0Jm8OYfDqMAZ0EuVkPBqnj6m8YkMy/tw0NgtQ/PXH0+TjlZGf9O1nvS9oQ+d4+xjGE4tegkYWfaz+Injh+9xf+Ng9ieGzVntSO84rKNYzyMJcmxaABlBs9LqGBEmA6VuDhI8P9MaFj1goJjvfQpgBYg6KjY6skr2f+N+9y+/hD8s9eDbHC4pNvLVcQV1nCO2OMKraQsgtU7C80I/l/z4MQ0SlezdwPsJBMTmvUkJRlDYoWLJVwk9WePD8eg+WbpPY+cdZLjz2WiIoPrjRgz31CpN6GyXF3K7h9q3buP3PBBQPjPazcExt9jBszjAcnjB88tswrWl4ZT/9ALuKvahoVpiqdU2rgsIKP+zvmtRWE0nD/hELM85HqRf2s+azpIVMkbn4e3IEF+KeW8ViYAY+Mi8ARVopzkbJ6Az1NDueEWnqYsmZJcWkDj9Bjq4mktiML5BNsShW8rO2SKe2S7aY1repsU9O4ca7EXg8Crq6/Yi8fwszWwSLUhAoltBt96O39yyuGVTMgfu3vsD1oVHcMuFW+vrm+y40NbtRW+dETe1JfJZ0F0jkYzrrx6gE1XbTu/Fbo21PehrNuXSIIQAAIABJREFUwI9p7dSPY0B4eIznnezC6IAjfXlSg2LeromblyJHzeuw6GvJzub6GvlJzzvm64NmNV6RcZLz1I0LBoBu5kOj9Fsrp8FpSJJ0xV+G8W2WfBEfK8ySYgLRIkSXWdFF52grwbu9KIzhbPYoanZkqb2sD5ORlifMY6K+TxDgMo3NAhSz9jN/q2Wf9CpOIxXImXnAutd41aKFRQuLByweuHhtHOOjAzhrjZlszp8XUPzN7ZiZFkTq5Krlr4XVGP9UC6fBn5u+1YFi8b1x4ZN0zlX/FKCY1KTNjrbSfcef3Rv5DA35HizK9uD5NySs3h1AWVMQnY4wOhxh7GsMYW2JjKztfmT/MREUc+ketwn2+CPYWqHgV2s9eHqZG0sLJPbdsyvdeDzDhceWJILiN193I7dUhtMXQfTEVUyYjfFFFb6cxB290OzmxyiplXGgLQibIwzKWwpEcfhTM5L4Ep/83c/siJu6gthY4seSPA+yCzzYuq8Pf7lyd0Ywo19wzSRBFe0nBvik96aFl6hiqvPUF/0shielaUwvCR8ZJFIxnLg8islUNDVlSIs4Lf3koFh7fjQhzjFLTg9uKT53kv0Mnu0UJvTq4QtkU2yqounWaAOabDGt1TcZwCJV5Rg83iDs9j/i71fuzMhLou5DJ/yw98hwugL4a79B+dpQxi/Hxw1h0IbfaUfu5mZszmtGyZ4u2P7yAT6fLTBWxyjhjyB2ot8UL92QNb/RhU0i/ktvwzsTPybhKT2/HrqQxGTAVCbdpsvM5UnSZmab1hTgVAAaffvP3P+M/JT4vr5/xXD84zGDVNRUU3Y7efOKwayDymOcG4xA3Fjeg7hwPUl8aFNGxvGtD7PhC40+h3DpZspOjqkJff56+ph4wVQmup2+PYATev6IxXBsQITg0H+g0YDqn6wfU1uw9jC3twkUp7QpVr9/8PFUzd9UH2N7GfljJv6xnlv0+j7yT3z+SVjPzK49v+73Yizg/Yf6JQl2tP4pnL7q+xf9x/7v4+/p3xfX+vfFf9/H9hFlt+ozO35cSHrNCyjWF3Bhr00LQNPkyvI+NoDxieEkux6mb3WgWJTZuPDRLwDM10ZQPJs4xffu3sPnV6/jzcjb+MNyL55e4cUv1/mweKuELeXc47TDE0VrT5h5oF5Z5GfglkIyGSTFDMiQ52geVqmtN4ycIhnPrPDg56vc2F4VQEFlAM+tTg2K/32ZC1kFPrTYQ0z9+uCpyxge10kSyf367et4/9if0feRzsvv6McoawqgyRZkNsk+fxQeKQJX31WM60l0/yb+0RtiMY33N8p4LseJX+U4sarQi/JGmale//XUhxj6YhJTeskG2Ujfm8Do1XNxlVLeNqbFnB4EznZ3ayZQTE6G7k1gbGQAF9LGPp6Zj1iZj5xDP6ll353iDpXUtAcMiJUWrPoBfxb1pHh2wybvhQbv2DHEjpzDINFWAHNqz7tjGDx/2BDDDgY6zrJeSeg9+35DTKItpolO6RbT9NyYtvg2ilAoAr8/hB6HDP9/vo8r177A+OSX+PL+l7g3cQs3Bt7HoMGv2x18HAkwQOx2K/B43sTbF6/hC7336S+n8MXVDxDtCOAvAxpDk/p03pYWFO5oY2rUe0ttaGt7E6c/uo6btycxdX8aX96fwp3RG/jkk5sG0DU18h7EgiJ+Pvg2LgyMYPyeGpJBZDU9hcnbiTGm09NopnabBU/FDuPspSGMEa8KnqEyTVMYhlHON7p2T1+eVG2m5/MYDp8fNOVnVPXn/d74Df/PXF/BE8neTfHfkXMYGBnH5NQ0VZGHxLg/hYkvRjBAfURXV3GdnA9TpE8qyWcuzWF8U9MhNW7iC2oHMS6qYyLFRD5riClMXkzHoIXmAPcvwcZOvXRXT5+DOPXhEEbHOc8KtmNNnXTcFfUj76fDjF+1bzRv5ESjtDxhlvaa523T2GyQFIu2eKDxVJTfOgs+nvM5KminaqeJ9rDOSceJOdP3G6QjzT8PWj4xd3Ew+6DpcB5iacV00U+i2rW+fAIUs/PXKLs+Teta9GfrnIoX5hUU/7B3aKLwBWJYWaQ62cr0qjbF/Ez2xY8upRBNmoOtR+mdDC/oTCGanlwu4ZlsL5593Yfn1kh4eYsfBH5L6oLockXg9sfQaAtja3kAWYV+LCuUsbpYxppiGRv2yswJF6lLEyCmgSUYjqG8WcEfciU8tcyFF97w4UB7EDYXAWVSp3bjsdeceOw1UqN2Mcnx46858FSmE/+23oN9DUEoQZ4eDXZ9fTFmT0x5SHIELl+EhZ9yE/CVIqhtD6KsIYDadoUBW6c3gi5HGJ3v3jRI6+4Nn2cep3vcEeatelFGD3672onXd/rQ0RNEiMJB0SDXRzuB/JruKX89o/7w+OldDOsl7xQuLO4I7sF2yMgb4XF1sfzDo1cqGxcxucYgB8JMndrtUeD2BOFy01mBJ3YZo3qAd/9zvOcNsnd9viAkKci+leUw6HC5AqisdqKgsB2r1zRgc14LNrzRhKysGqxZW4/c3Bbkbm5BXl4Ldu/uREuLBx5PEB6vAtkfhCyHEJBDOGl0T407V99CKBhGKBhBRA3fxHj/AXfsf5D949BRnNTZuIo6Pjz8zMe97019jxxn3k+//fLO/3gqeI/O3379Uo1/3xC/xPqYZ31OE5qfv+XyWPl/rfWRnrfnxt98jcbXZ3zundv3xK9ifaet92Jqe9J8SMDXzF+UB1sfqhJj83Pr3uqPep6eL36YV1CsL+AP7TociaK8JYzfbPDjR0vJlpgO8j7Nj0dUAMxBsQjH5GWhmR5b6sXT2V78LEfCL1ZLeG6thBc2SlhaIGPTPhk17SE4fBE4vBFUtYawsTTApMTLd/ixejcHxQSUSSocZgMIH1g8UhTbqmT86+sePLPCzcIwcadWUabi/PuNHjyRyQExB8YOECh+PMOJn690Y0OJDAKufJEugAaB7ShTjSYP1U5vGJJM91HY3WG0dIdQ3aqgpkXBgTYFtR1vmrz8TmLweJipVdO3+WV+PLWkB8+tciB7uxdtdg0UU+xicsAVZjGTtcH2+807AxgTzmx0bvMPHj2LfoOUmBxbD+PUN7hT/P2mq3HDhO1Yv3cdE/8cw/Ur7+Pof/6JgVuSHv/5yAVcHTdZG49egtcbBAfEISZlJiCrBDgodjgDKCmxIW9LK7ZsbUVVtQMFBW3IyKzGa5k12LipCTuLOlFe0YP3btzD5wNX8Paf/wypyQW3OwBX5DDevvi5MRY57uKztwMMiBMIV5Qw81L9w2oHc7tY91b7zicPWOPpt8VPtC6IRGMMGIuNvG+rLFa+D96nQqEQamtrEQ6HDaB6bjQVYDaCYDAE5WscVA4GhGktGwwjogTZEVVCMBziPZ1UeW5lfnCaWfk8vLSbF1A8XwhdMOJ3Iz0CaTGQMyuaGEiteUlBAD/OlphEmINiDoy5F2odEM4kiTG//1EmqUpLLA7xr9f7mA3x73NlpjadXSRjZ42Ctp4I3FIUNlcEJfUK1uwOYGmBH9k7OSBetyfAQjQ5fWJHLcbUnjt6w1hd7MfPVnnxm/UerCkmT89BJnEme99tFTJ+kUMeqIXE2MlA8RMZTizKcuHlXB+qWoJQwjTxaaA4Eu1jYJUkxJSOV+aS40AwioASZUB6f6OCPbUB7Hdcxg2h6ke6dXev4+9MshyG2xdGcY0fP1nqwE+X9eK1PA8aOxWQhJnAMTnf6uwJMYk37Qhy1ZzEHcPvBj9og0T68pjVOzWFQ+PVNMau8JA86dP7vtNjocpP/GJSDTYSWHd3F0MnSDocYgdJhgk80xEIhKAoESY1djhklFf0onh3Fwp3dnDJ8NYW5Lxej7ytLSgrs2Pfvm78xxWTly5dTvrLL0feh6/DA6czAL8URDgcYeOJ1d7WDreY6+hs8UM6frDG04XnD21uo7WONxBDS28UlW0R7GsJo6ItgnpbGA5flGl5aeXhayTtnqez0PdKJIge2QlHwI1gJPSA/YeXXd//BFDjfTPxufj/+6bRJkkSFi9+FbIsszXtg7QP0cYr+VFaXovN2/ZgY35x/Ni0bXf8mv5Pfb8bm/JLsK+iDh6fH2FJhre8Hs7Nu+HILYIjdxccucXq/S448/dC6exBLBJhEmNOf2u8fJD2E7Sjs/V9uvnme+d9Whu89Y28MNc0KPLdMQKry4sCWJQt4ZGlJBkWkmKTtFgFwgIQk9r0E1leJh1+MdePFzdLWJLvx5J8GRnb/FhTEkB5M8XwjYCkviQJLqxWQBLiJfkSlhX6kbNLxsZSmYFXWdGrTkeZVDlzm4TfrPViWaGE/HKZ2fsSwCYVaPIS/XKulzvcUlWouTo1lxb/IseNN0pkJgGm0E5iUqAzSYvpoDwpHjFJdQkUk7o1gWWKVVzXEYT7Q6PDotv9bzGJN8Ul9khhtNqD+N1aF57J6MEvcxxYX+zF1n0+rC/2oOQAty/2ByKQFZ4+5UGSY9qM4O0qzt9k23+dvA6aQqLoYZK4nsbEZ+eSxwy1JMdz2M2exYL5q7v4/IOjTIosVKX5OQS/HIIcoCPMpLiSpOBArZNJhPO2tmJnUQdK9tiwbn0jli2vweq19Ux9uuq0wVhZNKrhfP9mPw5198Jm80GWg/FNJ7HxszBj1tfhW+tbq02+izxgjacLx5fqGodJ4vjc3mSPILcijBfzg/j5OgU/XavgZ2sV/CZXwet7gqjuCENSQ0LSOmHhypaaFzukbqxuzENuyw7YZUd8M30uZdGbbclyAHZ7D1paWlHf0ISmpmZ0dtng9fmYdJXqqSgKHA4HXC5XfJ00l/we5N1IJILubjvcHg8ikQeX8no8Xvzmt7+D3+9/4PYKhkIor2nA//3Eb/A//cuP8T/8y1MzHv/jvzyF//W//AT/16PP4f958rf4Px95Dv/bf/lv+H+f/jcU7i6HZHfC/no+qv+PRaj8nx9H5f9CxxPq8Tiq/venYFu6EWGPpM6fqXniQehrfWPRMxkPzIukWCT8w9mB4JMF7QjSgJhXrjAJ8SNLJaYu/UgGlwQ/mimBroUK9aNMQixsir14LMuHZ1b68MImCS/l+fHqtgCWbJPjoJi8Rtd3hUESYLLZbeyOIK88wEAxAeKVu2TmRCtvfwCt9jCTWAug7pOj2HUgiCVbJCzbLiF3n8wANsUsJjVsApf/P3tvApXVlaaNrtXdq4da9/73/vdWV1V333+46///vt1dVanKqEZNTGJi1BiNM4NMiogoooCoQSUiooiKoqLM3zyPVlWnUpoqTcUhBjUaS6MYgiKKKCKECCh57nrfffZ3zgcfBhO0ov2Z9eVwztlnD+9+9/Dsd9pZ5cWUNDOeJS/UbFMsJcY6PBupw4hYA2YsMbM6NAFg6kdqr/xR+21OH4deIhVqu9PPJ8V2lw9leg+2V3yAU0EBXztQ+56HHWxRuCbyTG22eZGZZ8VrCTq8MKuarzPTDEjPtWLTThe2lDixvtiBDcUOVs0mr9YEkgkUqzYsYvA+Kvx1+KzizEYrQSe/ReTM50YjztWEJcTEa9+9P/fhZF0zWinclJbW5JyouxNtzfX45Pc+eNw+tul106GOUwBgh9MDB0mI7W5YbSRBdsHh8PB1W7ER2StL2a6YAPHy5cL79LzkrUhP34FluiM4cYHCN90JAsK4cxdf3biCT9//HUoLKqHXOYR0WNGCeFT5+fGb3x+t+eQ/Ov3D8+mD4Vftek+H6Bt3+TA5w4nhiQ68PN+BKUtdiFruxIwsF8akOvFCkhOTM53YXuUVNqK9DnC/+3x+bwmOzH+jrggjVr+J8euiUGwqCYBi+X6g44X2UjqdHrlr1yE5eQFmxcYjMioGkdGxSJiThBVvr8Lu0lIGxJWVlUjPXIrsVasCAO1+y7vf9ATE5ybNx5rctXA4HX0A7UDzk6BYSooHSh9t/larDQvSs/G3P3kSf/n3T+Av//7n+KsfiSv93ff+F/hP/3UInntlKuLnZyI1czXikzMx4vUI/NNPR+HZl6di46btsGzejqInxyHv736GdT/4OV/pb/rl/+BnKHpyPJzl1UxzbX2oDeH7gY2Xb9Pf/5HpO6igWBL/8bgKcEgqQ5OX2PFsDAFiISEmdWkhDaarBMUWdrwlpcT0fEicGWPmC0A8I9OK6ek2zMigH4VNsrNKNIVgMtl9bFO8scSFRbl2VpuenW3H3FXC0dayAicqDML2lwArSXXJmdbiPDtmLrEiOcfGcYd3VHoYrFYYSfLqw/odToxLNuGZgJSYwLEAxHQlNerX5hqxcI2N82PbXrYhEocC1I+UDwFcUqOmvykNlU8gmsBrhcGDwhJSpbZzDOOi3U5sL3fyN1L9ene1G2k5FkSkGZG4zMwq1QU7nAyWoxYbMS3VgKS3TZxHpcEtJMUKkBBq1eETrcdjTD3cfpSHO/JKvEs/AscCFHtgs3tgsQoHXBT/2GhwYtduC/Lzq7AiexeyV5Vi1epSpKYWs7Q4IrIAs+dsRkrKVnbANXv2JsTGFGB2QiGS521BVtZOFBcb2OkWlatKhsV8ot4/XFqE+SdM7zAPhHlAywNyXqRrUbkXb6Q58XyiAzOWOZG3w4NqM/n8oAN7eu9BSp4bSbkuVJiETaiqzfVw6UqgeGTOmxifF4li464+YFHbxv7+pjZX63RITV2MqTMiEBc/B0vSlyJ75WpkLV+BxKRkTJsRiYWLFqOqqholJbswKzYBySmpAVDcX96D9Zwku+PffAuLl2SylFesJ/dPawmKv72k2AeTxYqkRcvxNz95KgCKJRgOfX0CT704GRs2FcOt2AZ7vF7sLq/C1FkpLEEePSkOu4pLULUoG+v//hnk/e1Pg0Dxur/7Kbb8YiyrUH/btg9WX4TzuX++e1RpNiig+HE8saFBSCBwUZ4Lz8fbhKfpaAUUzyIJMdkMC1DMIJmAMr8nb9MWPMPOtSyYvNiGmZkOTM+wYepiG6al2xG51MFxiPOLXag0UqxfP6pNXqzb7mIbYfI6TWrT5HmaHG3lFLnY5pjqRD+3148tpW6krLYjKtOK+TkOLN1AoNSN4nIP2z+TZHdNkRNjkkx4lm2KCRCTJ2qhOs0OtyL1eH6WHm/MNyNrg509TcvJR1x9IAdjDpcAxwJU+DlOLP3tcvtQWk2Ot1xYVWjHojVWrC60s+R5V5UbpTpSDfewDfHuKje2lrqwaacTuyo9WLnRhgnzdBg7W4eYDBMD5JJKJ5xuOoWmASgBRdjr5eM4vrQT5sNpn6oBQXGOWWLsIidbwraYALHBYAfZFVdXO7B1m4FtiMkDdWJiERamFiNpXhGiYwpBwDgurhDJyVuQmLgZSUmbkZZWjPXrSdVNnOjL8UN8LCTEkqfDJ9wPp7/VRTxcXlii8PDnm+8//9EcRT+b04+o5S4MSXBg1goXSvXi4JvmLnEoLbxQc1QKVp1Wo0YQXXl8UV4EflwutgHVAuag8UcaYH4P/D4n/D63kDiH8Crt8wmtNFWDTTr88qNQvxUjCBSvi2RJcWB+5Tq44fe44dOoGweVr9TX4XAgNW0J3poyHctWvI3q6uogR1RmswVr8/KxclUOq0wTKI6JTcD8lNSARh21nehHas7kzIodSPF8r443sQ6ovCDopWrkBein5EV5kO8JypNA7BtvvsVg3Wa1BsB/qPZQPvLX+z21Ras+3fv9QO5VUPwk/upHv2DJsATDfSTGP34CP/inpzEnJQtGk6i35DW6rtu4Ff/y3Hj88H8OR2TCIuiKd2P7KxHIUyTFUmK87m9/iqInxsJVVq0xpwuvnwPpL8kLWv6Sz8Lfq+MzFH0GBRRLYj8+V7EYbK/04MVEG56S3qYV4CvBL6lNq4BYdbRFgHhonAWjEgkUWzEz046ZmTZMy7CxtHjWMjvHJt5a5uFTWDqNJUnwO1tdIOdbMzOtrEJNwHh+jp3BMqksCzsYIbHdsMONBe8I9erYZTaMSzZjapoFC9fYkVfsZMnu6s0OvDbXFAjJJEExS4sj9HiOfpF6DJulx6QUI0uWaeEjSTCBYWFnrE7gXq+QEFNIJfqb7Iu3l7tBUt/iMhfWbXNgZSGpcTs5BnJJtZslxtUmUun2oMrkQWmVBwSQE5eZ8FJMFcYm6hCbYUTuZjvbLosyxWJNdaBFkRZIBsnyqlkAHh+eUxe1cJsePC20izTxssdN6tRCjZolxkYHysut2FioQ3Z2KRalbUfKgmKQCvWcxM1IW1yMNWvKkZ9fiR07jCB7ZLEBosOcB1//cBlhGod5IMwD34UH5By4ucyDUfOE7fCmUi+v/WK9VQ8SZdqgq+JzxedywqfXwVO0Ca6CdXBt3QyvXgd+TtoyCvj2ee3wOUvhM6+Dz7gSPst6+JwV8PvoIFHub3xwuH2oNPlQWOpBwS4Pa7GRxh6FrbQ4fNhQTZLiiaw+vd20i+ddn8MOT2U53FsK4SrcAE/JDvisZIsqVL1702lbcTEmTJqKtCUZMJnNIedsAVCFVFwLiiUNXC4n9HoD2yBv3lLE0mST2aSAYwpvSaH+jLBYrBrATcIEJwxGCulnFnUnzTuXm1W5yZ55S9FWlJdX8LfjJ7yFtPRM2KzCSVbvdgzk/rtLiv3sZIskxX/9k1+GlBT/xQ8VVeofPcFpyG44d/0WOJ0uZe+m8IHfh6LiXRg6ejrnQ6rUaRmroFtTiI3/ZYQiLSaJsfhJSbH2kGUgbQ6nCc+N35YHBhUUPx4nEMrk7BeT8Kgkcq5lwZMaSbC0IdaC4wBAjqKYxGYMi7Pg9fk2lhBHZNkxeYkVMzLIptiGqCw7sgqcKKkmKTH9fCjXe5C9yYW4FST9tSE6i2IYO5Ce70BxBZ0eCokTTcgUHil3m5PfkS3x5EUW9jJN3++odMNs94BUVbILHRg9x4RnIwx4lgAw2xYLD9QEhofQfYQI0zQ8Wo/IDDMopJNTUTMliRrFj+MFS3GqweqnHuHIiyTIJJGmeMbkeKtgOwFjJ5aus4HUqEt1blavpvcCGPtA6tEkPZ5DoDi2GqNiqjA5Rc+q02Y7nZAKQG53ednLNQFoAurU7t5M/njwmzp5hdtz7xO8B0sfoZEQit9dLi+D3vIKK7ZuNWDXLjN7riZ+ZL6UatKPfZztP2f/hCUED5b/w/T9j0ZfP/lM8fuwZL0HQ2c7ELvSzWBUSIfFekvacuTvRE8RJWw+vtLfZBrl8vrgczrgKlwPy+RxsIx8DsZhv4Rp5DOwTpsA96YN8NrtAvh5rPDrl8JT9Cw8638MT/7/BW/BP8KzbTi8ptXwkeTYL3yY5BZ7MC7ViefnCnXu11OdmLbUidcWuJC9xYPc8iKMyJmE8fnRLCn2WS1w5qyE9a1xMI18GsbhT8H82gtwpM6H12hQ9jAqf9OcTSrJE9+ahl27SlkqS2X31/+UnkBxbNxslhTTPdn7EhCeFROPSZOnY8KkKZgydSaSklOwa/duVhnW641Inr8Qb2evYgAs8yfAm5KahszMZQyWOa/NRSyJnvjWVLz51lTMiJiFVatzMXb8mwyKreQkK4REXbsnkvnLZ/JeguLvYlNMnqfnpa0IAsV9JMRka/yjJ/B//PchGDJ6Ghakr4TOYAoAf66Xz4f8wm34+fA38Xf/9DR+8E/PsNS4IG8TyiIXIP8//QIkISZQvO4HPwuoTxMolu3p3b7wvdhDhukzOPuTQQXFkjkf5atcEGhzTAvC+FQ7niZ74mglHjGpTIeQGKtAmRxsmTE83srhlyKz7AyCp2VQGCYbpi+xsVfplZvcfPJJZVjtXpBtMdkOx6+wI2452RXbMCfbjuzNLrY3pomY6kbS00qDDysLnVi2wYnUNXZMJ6/WGTZsKfWwDRA5xyKAuWStHS8nkHMtA55T7Iqfi9KxyvSIGANLihkss9SYnuswY7EJJVUinipJbbWAWP5NkmIC8gToqf70K6lys03x9gq6OlhqTECfwDCBY4pHTIsrAWSr3YMla614Ja4KI6IqMSFJhyW5FpTp3EwLSr+60IZFq83YuIOcFWlBcV9w/CjzW7ju6qHA94IWrI0g66Tlf/E3jQH1J0K20bjk8Rni4OZ70aaw5LrPgVq4XySPh6//UXlB7Cn87FX62QQ7FuS5YbQrc5nfz4fjq7d6MCHNiddSHHg1xY7RKQ6MTnHirXQnNu9ysmTWNOIZmJ75N1imjIc9MRaWt8bC9OzPYHr5ebgKC+Bz2eHTLYJn43+DJ/8/w1P0NLw7XoV3y8/hWfe/w1P4v+Az5fDhN9kyD0mw47nZDnbwlZjjwpiFDjwZa8cvYuxIL3Ajp5RA8ZsYlxeFYv0OOHOyYXplOINh64xJsM2eBfPYl2Ea9iTsybPhc5O0UuVzUnWOT0jEzMgYGIzGoHfadPJvopNWUux2u1C8fQfenETgNQrr8tez52oCvwRoY+Jmszdr8hwdHROHxemZMBgNyrrhR8muUsQlJDJgJkC8fcdOTJ0egekRUVibtw7bd+zA29mrMWnyNLzy6tgAKBYHtmo7ZP2+6SpB8be3Kf5mSTE73yK16h+R060n8Jc/fAJTY1JQqVO9dcs1cktxCSbMTMLwMTMx7NXpeOalyZgSlQzr7koUD5+KvP/t5yB7YgLGYUnx/ff3N/FD+P29aToooPjxOqFQJLKKFCg+24EhsTaWFAeAsWI7LIGw8DotbIwpDNOweAtemmvFW2k2TF1ix4xMB2ak2zGdfpk2UNzhDTsF2CW1aIoLuK3cg0VrnYhbZsfclQ7MW21nj9KbdrnhclMnClBMqkVbyz1YttHBjrZmr7DjzRQz3lxgxswMCxbkkG2vA7NXUKgmE4bNIkAsgDGBY3Ku9cpsI15LNGJIlA7PsaRYSoz1GBalw+wVNlZ7JvBLG37Vay7d+9mmmUA3SYlJvdnjJZtoDzaXOLFxhxObStwsKSZwTKrV28rcDIirDB6WEpO6eF6RA+MTdRgZWYk35uoRtcQQ10rZAAAgAElEQVSARTkWpOZYkJBlRGyGAQtWmZBdYMWWXS6YbYpzj142OzTAHy/+C7fnz9+fJAGWanfyEEYFw9rxQONS1Ff7fnBOLOXi9eenR7g9si/C8014fnp8xqOYs1LWuvFcggPz1riht/pAEmSa12htJ8ns66kuvDDPgReT7Bgx145nEpwYkeRAXoEOtqipMD39r7AR+HQKqbDfboM9MQ7G556ALSEa3vKN8G0dCs+6/xPeskj4PUZxiOgsg69kErzr/hM8xS+iwmDHpCVODJntROZG0nYTAF1v9TMwfzrOwaD4nbKt7Ghr3LpIbN2+DvboGTAOfxL2xSnwms3wVpTCvjAJhmd/BvPokfDu2B4EfAkUE3CNjkngcEff1J805gkUk6Mtsik2m81IWbiIQWtpWRnnLQFfbt46BsY5a3Kxs6QE0THxSE9fyqrQYg7xYffuMgUUp7LKdEZmFiZPnYFtxTuYLrS+UH6Ux5ixE7A4YylIUixB8TfVt/f7wbApJkmxVJ++t03xE+yVmtL88zOv4528TSyJl/ShK/OXRuDicLhQpTPC5/HCWalD0S/HYd0PVFActikOr78Pc/0dFFCsrfBj8XfAdtWHrI0uDI1TPE+ztFh4m+6jOh2lSJIpNnGsBaOTbJiUZsfUdCsDYQLFFJaJYhQTKKZQTGabkLJSjOJ8shFeIzxOL8h1IHWtg9WpKUayWBzEREmeqnOLnEjNtWHeKhtmLLHieZL6RokfqUQ/F0Wq0sJemMIxSSkxgWKyKx4Za8DoOUYGzMK+mFSopVq1DiNjdViQY2WgK6XFUjpGkxrVh0I+iTBNog0k/aVYwxSbeMtuJ3K32LEwx4q0XAsKdzoZJBfucGJHBQFjN3aUu5C2xoLR8VV4PqICw6Mq2cZ4TEI1xidWY1xiNSYl6zBriQELV5lQVu0OLJI8sUo7JZbc0SIuvHP7gyR99z4Reix4VXMCHm7Pg+pvCXgfVP7hfMO8G+aBMA88bB4Q8xpJg4cn2vHmYid26bzCd4mi/UJaYDqrDzqz+JGdL0mOX5xrw6a8ElheHcHq0p6yUgZy0n7Ys7MYppeGwjL2ZXg2JsGz6Z/h3fxzYU8s126fBz7japYguwv/Ddt27uBwUCSRJtthn7Km08F85kY3hiVqJcUTMG5dFIpyM2CZ8BpLhu1pKbDPiYFp1BAYn/s5LBNehWN5BnxmUxAoJltfCnU0ZVoEdHo9t5d4j0CnBJ7afQTtN6SkmLxPU0zjqFlxmJ04D14f7Tuo3wQtLVYz57sgNQ1FRdsYeC9ZQpJi5SCA8tq1WwHFC1BaWoqE2UmIi0+E20OOxwQPUJlWq5UdbZFNsRYUyzQDvQ6mpPhvOCSTkAb/xd8/gb/gcExP4K+U61+yE64n8Lf/+DRen5yAHSXl7JtG0Kd//qb28s/jgTGnAPn/+Snk/eCn2PLEWDhKqwRvhfc6Af4YaN+H0/XPc/3RZlBBce8Tqkf7XpxoEXh9IdHKKtNPk9dpjZSY1aiVmMXkjVp4nbZgWJwVryRZMZWBMEmKBSCeuVSEW8rY4OS4wxaHsMmtMnqRs8XJXqRTcuwsAU7Pt2P9TjeHa6LJgiYVr8/PNrpZBXYkr7Jj/moH2xMPjTaCfs9FGzAk2sBgl8DxkCgjg+Nh9FwBzaQ+PSJWjxfiFCBNYFhRnyYpMr0fGqXHq3MMyNvqYpUmKpsWJrJpobpIUEwSbpOdVKiFGrXT7WcHX1t2u1hqXFTqYnC8ZrOdr3lbHdhW6sTOCgLFbna4lbragonzdKxK/Up8NcbOqcar8dUYGVXJwDhpuUk44bJ6WUJNNJD0IKaWf3P9aNLUgOJHm//CEplw/4VPiLULV5gfwvwQ5gd1k/ddx4OQ2PnZjnjcIiGhXVroVgAp7TnE+kprLP0o6kX+Tg9GJjnZrKxqyy6YXh4G06ih8OiqsYdBi9greMvLYH5lONv2utfHw1P4P+Hb+ix81sIA6N6zxwefeS08hf8DzsJ/QdGObewBe8xCJ0hLjctnsOlF9mYPhs91IX2DUJ/mkEz50SjKWQzL+NEwPvUvMA15gutjS4iCqyAfPpORVafFHkFdT+l+5eocvDFxMgoKCtk+mJ6p9BRl0z09p99OjU1xRWUlIqJiAvbFMg3R0+F0coin+QtSsXVrMWbFJECCYklPCYrnpyzErt1ClXp2YpLYyyj7G8rT5XZhwsTJWJwuJMXfB5tiAsXkbOsn//IC/vsvXhWeqBUnW1oJ8ovjorB9Zxm8Xo9y6KClr+Bhld4KT8v9pd0B3eJV2PiPwxT16UqmTe/0j9z9gSOoOVYT+B07fhQH933zeN57sg6NdSex71valMs5MxS99h48ippjh7G/X43L/Th8rAZHD+7VjI/Q/Rcqf1k2XR+V94MKirUEeJT/pgmJgSDHA/bi1WTyQG0O2BVLtWl51YZnenqWBSMSrHhjIcUjtmPaEjvbEbMadQZJgkkC7ATFEiY1YpISkz1x5non2xqTt+nM9Q5Wj95c6maJrDxlIwdY5BGbPEzHr7AieqkVryaRirQJoxJMeDGergYMjxWA99lII0bGGfFighHDYxRgHKnnv1+MM2BYtI7TPx9NwJhUqcWVpMfDovSIyzKjVKeqMYlTVAGKyW5ZSopJakwnyuQQi9pEkuLsjXZs2O7Eyo12RKcbkbrKgpxCO0uNt5Y6GRBT/GOSGOcV2RG52IAxCTq8HFuJkZEVGDWrEtMW6pC51oL1xQ6WLpusQmXb4SS6eVBQbEfeVjtKq1y97I7VieZR5sNw3cP9GOaBMA+EeSDMAw+GBwTgo0PurEI3Rs514KVkB5Zv9qiH8QQK/T62L95c6sWkdCeGzXEgZa0TruoqWCa+zlJZV+5q+FjdWQBpZ/YymIY/BcuMSfBsz4Z3yy/gKfgH+HRp8PkUj8QeK3yVyfCynfEzKKnUs70ygW6Km0zlkhkLqXHHZjtBds8Zik0xg+J1Udi6OQfW6ZNgeOansMbMgGdHMcjxls9ug1dXxV6wQ3kuJq/R02ZGIjI6liW3JD3W0pj2gDqdHkVbi1FVrWNQLOMU6/QGJMxJYmdYNpuiMq5Iigk8ky1w1rLl2LmzhJ1nkdSY8pLgmbxLky1y8vwFqKysZHA9M2IWh36SdaC0O3aUYOz4id8rm+K/+ckv8eP/byQKNhdje0kp/svPXoKUDkup8d/901NIX54Lu8Op9KE4ZJBt6+/KknrlEMLrdKJ0ejK2Dpn0+EiKL7aj778edDSdYVDaH11qb/YAPa2ofQCS8tPN3Vyl1s/3BvF/oC5nW9ADoLv5dOj3D6BOgbL/THkPCih+VE4AJLHvVV8amNJmkFSHl210YcRssikmUExeqJWroi5NwFi1KRZOtl6YQ062hKr0tHQRp5ikxhSaiWIPr93mRpVJADxyPrW9wovUtXbMyrJx/OL0fCdWFFJYIw+fmAobEz9sDh/W73Bj3moHorKsGDXbhKGzSDpsxAtxAgAPm2Xke3auFSUA8ohYI0bEGgLA+PloAZJJ7XpUvAHPs92xYl8cocMQlh7r8FK8AYtz7expUqhRiwWPvE6bbVR/ISEmm2cKz0SgnQAyqYWX6zxYX+xEcbkT+ducWJJrRVa+lb1Mv7PFgR0VblbPLqkUoZyilxgZFBMYHh5ZyQ64RsdVYWaqHsvW2bB1txPby10o2uVE5lorotJ0GJNQhbGzq7BwlRlWh3Ii6Xt0TqQGwo+U5l78Gn4fpk+YP8ISVDmXhOeD8Hww0PlAHv4Tz5jtPsxfK4DxiCQnZmQ5kVngRu52D1Zu9WJOjoudbA2dY0fkCvJS7YXPbofz7UwYn/lXmEcPh3NFFtxbt8C1LBPm0SNgGvZLON/Ogt9aBW/pNHjW/xCezf8Kf1USO9bylcfAU/j/wlvwD/BWJHDUjIV5LjwT78AbaU6QF+qiCg9S8jwYNkd1tLWmnGyKJ2JcfiS2VW6FM30RTMOfhvn1UXBmpcOzZROcWUtgeXMMrFFTOVxU7/FBcYALNhZiyrSZiJ4Vh9zcPJSXU5x5E4dZKizczCrWJBFeu3YdA1RpU0wxjin9+AmT2Is1AWxSdd5NqtBzkjB56kwUF+/gUExz5s7DzMhorC/YCJPJhNLScqQuWoLXx01gUEzfrd+wkZ12LUxdjMqqalitNuwqLWVv16PHjPte2RSTpPhH/zwCy1fnY33hNvzon4fjr35McYufwF/8iMIy/YKBMr0TBw0SEMvrvcenELzsYTV2a8E2jl/sLNcJKboCkgbK37LPvzfpGRS3o14j8d1/thkES7Wg9GHWV4JifNmAg33ouxd1twSM14Lih1m/P8d6NiigWDLf43Ol01E/iis9GDPfhqdjREim3mGZVC/UFK9YxCmmcEwjZ1sxMdXKNsRkRzw9g6TGBI6tSFrlwLodblDcPZKqkjdmcrqVnCNAcdJqOxatdWDVZifH56MQRVJqTQA6u9CJqKVWTFhgxrAY4UgrKsOK2GVWjE0248UEAZTJlphUpl+MN2BknAEkGR4ZJ9Sjn4vUMUB+KcGAUaRGHakHSYfpuXC8Jf4eEqnDG8kGrClygKS0BHpJIkzSbQLoJLGVEmKXWzjmoHuRxsugOCPPitwiO94usCNyiRGJy83IWmfFlhJSo3aBpMbvbLYjLtOIN+bqMCqmEsMiyjFsZoVwwpVYjXlvm7B8vRWLc8yYvkCH1+Ir8WJUBf+i0vTI32rn+mm9VMtTWXXhF3QUUncxQQe/C0sjHp/xG+7LcF+GeSDMA2EeGCgPkF0s7S9WbHZjwmIn2++SRHh4Iv3twJDZdrw834GkNS6UGrxsJ+rzeeA16GFPmQvT8CfZ27PphWcDYZnsC+axtNZHKrT2HfCWTICHwjDl/xDe9T+GN///hmfjf4WvdAZ8zkp4SDPP6EPcSheGzXbg+UQ7RibZWYI9MtGBp+JIUuzCmtIivKDEKS42lsBbWc5epsmW2PT8k2zjTFJq8kjtXJLKDsBC0YHALYFf8kRNjq6mzYhE1KxYzIiM5nsCxCuyV4Ikw2xTHDcbKQsWseMoij+8JD2TpcIRkbPYRnjqjAj2Ir0mN49tgEl1OD9/A2ZEiPxIlZpCOJHzrSkzIkD2yS6Xi51wZS5djrcmT+e05JmaJMdJ81LY0RaVo0qk75+nCYyPf2MCyPu02PPcfx4Wqw3JaSvwN//wFAgY/9O/jcJ/e2J0UIimvyS74h/+nB1sFWzezqBY7sOktmOofpDPKK3X4YTX7oTHYkVZ1AJsG/YWnGXV37reMu/vxVWC4l4S0Po2ALdqWRJ75ko7WupPo7apHZ3dHWg65UdNfQvab9Sjhr47Vo+Wtna09/q11NcEJLl7a2rRdLMD3Xe60XGzCbU1/UiB/X4wKO7sBP3XdLwXXxxqQAd60HOnt6R4P05eakH77W50325Hy6WTGkn3GTS2taD++Ek03FDqcKMBJw8E591vHf9Yiyb+Xpt+L85dbUfrpZOBNj7I/hxUUPw4nCAI21ShVrRgrQvPxSrSYbIdjrH1sSlmFWqSIKd/DNuRSzDp9mB0khWTyPN0BnmftoGkxQSKo5fZkfKOAxtL3NCbKd6wH+RIiwAwhWCKX+FA4io7A2SyMaY0ErgR4CvTe5Ga68CUNCsijZ+hpu4WGptb0Vz7KXt1pnjG63c4sXitHa/NFXbGJA0mD9TDZulZqswOtSKEevSL8SRhFs63CAyzTXGEAMhDlPvnoylMk5m9SJN0W28Sat+kKk02xASAhQdq4YWaJcYuAs1eVBgEMCZJcdFuF+auMCM204hFOUJivG6rk9Wp395gQ0KWCVMXGAKOtwgUj4iowkuzyDt1Nd5Mqsbrs8nWuBwjI8oxOqYSselGrFhP3qmdqNC7oWOVdC8sNi8273SwLXKF3hWgodQA6AuMxQB8HPhXO1kMbnv24tQXV9BYexJ7+5wohulHdB9ceofz+/b03IuTdY24UnsqwKv30z+Hzzag8dIZHNGc6N/P93IMfvv6h8dTmN4Pf/wTza0UXrHaA3K+tSjfg7k5biTnupFV6MHmMgqrKMynpA8Pn9cLj0EPd34eHGnzYZ8bB3J45dqQB59RD59HRI1glWlXGXyGZfCWRsBbMhG+8lnwGVfC5yRHSl72Skyq3CSFfqfYgwV5HvaIvarIg7iVbgyd48CyQjd2mQxYXb4O66o3weQ0cxmkKu1evxYOcraVlAD74gVwb1wPn8kAP0cSCKanBIcUEqm8ohLrNxRgxdsrkZ6xFEuzloOA7c6du2CxWuHxekEenLds2coSYPqWpKDkwKqgsBDLVmQjY2kWVq5aja3bilnSS5Jo2m8QECV16eVvZyNr2QoOuUThnDZt3sJhnSgd/YxGEzYWbuI0mUuXgbxYk9p23rp89mLt7qXefT/jg0DxG29MAB0CyHbfz/e0XyL75tz1W/D//HQU/vonT4LUqP/2H55iUEw2xn/941/ib34i7n/4v4ZjxOsRyNtYxDbWcg97z/mQaGqyYveUJJSMi8GO0RHY9D9exK4pc+E2mLne9/z+UVh/e4Fi2R4GxW31vH+obwd67nSj+8tmXLlUh9OHFODa3YzTtB4dOo26Sw1okL+r7ehGD9rrD4r9x6dN6Py6B503GjlNU1s38HUHGo4E87/sfwbF7Y1o+BLouXE2CHR+0tQJdDahqV2AYlHfg6hv6wG629HEdWhCezfQ01aPQ7xe1qMdPeju7ER7E9VTvKd8PpHraaCOV4Lr+BFpfNVAFHtC3U8p4Lzl7MPRCBtUUEyEfpC/gKi/t2I+M0EdajRG69+2HoHJ3udHzAo7nomx4pfsddoUZFPsuUyVuAWPokYde1CxF2i8oIBiKyYvsWLyYgLFFKPYiohMG4Nact7F3hxNPhSXe5C+az+22Q+jaKcdiSsFKF5b7GKnVTShUJ0IhG4u9SBqqQ2jqy/jBhXfcxddt++i+dNDLHX2kEt7jxcFO10YO8+keKRWPVELiTCBXuFpmmIVj040giTGrycROFakxVqJcYQOL8TqEbfMgs27XTBYPKwiTUBYSopJZZocYxBwZ7Cs2BiTGnWFwcOSYXKyRRLh5ettSFll4vBLmXlWrNlsQ2aeCMM0c5EB4xJ1rDpNHqmfjyjH8AiSGFdgJHmo5l85XowsZw/V87NNyMqzsBMvAsGFOxxYV2TH8nwL4jL1/I7sluWkzLRU6EkTvbz/trzyfflO6yzhgdXpj3RqSP/aUR8Y56oThuByT4NNVdrrv+V8QBMrFfVtv3+w85BsKy1gwfT47uVynrQAShp/KtSr2i9+97xFvb9r3wxWPQYxH4VGARWvkLw60PLOoOUO9Ws3mj9Vv9l/pAY1Hx8MAtmSDwZy7dOvsn+/T9d9B3H0WA0O9zrVH0j7vk2a70rTb1Nm+BuVp3vTgtdJAic+HyhUJEW6oBBNBpuPwbIM0SjWTcUJl1xHXU54TUZ49Dp4zUb4XM7A+kr7F/6Rp2avHT63jiXDPrcePq9DvFPslosrvVhS4MG2Ci+MNh+DcALiEcvIltkOimNsd3lg99hh9zjhIQ/Wih2qz+nksr0GHXucDtSB7X2D281tVcYe/U3gmFSZCUASALbb7QFJJ+0VCLg6nS64XMEeoknSS8CZviO1ZwLLat5ij0FpOG+zmfMlgEvP6CfTch1cTpgtFs7LbhcAloAshZCi9zJt7377pnsKITV58lQu+9vmQd8ZTRZkv7MBc1KyMFvzo/uElKX8o+eJC7OQvDgbm7eVMF21Qoh+6+rzw2NzQJ/xDqpTs1G1YAWqF66AtXg3HDYbe/nu99vv0xx6r7r0AsXUnr01Dej4GuhoOMh7JV4nbjfisCYfxj3aPUHg3WEBZtvrlfRC3bmn5ZxmnVKe3TgTci8mQHE99n7eCnzdiroAhhLrINWL6hRYW3mt7SVVPtLIEuWWszTGxN6t88phtTz5zTF6/811PHi5gwoM7IEONnQAd1pwJtDu4LE82HwxKKBYnnjIyj2oewGKe5QTCPW0pKm1k43BexTmGIzyCeCRHTA5ziI1aQGMZdglCwKgOFq8f2rJUfz7Z834rW4PO+YiMDxlCf3smJbhwPR0KyKX2pGe78D2Si+qTT6UGTwsNV7wXhMDjsYaAYoX5jpQUOIGObOSIN3q8GHNVhcmLrTgneO3AXThuN3M0t9Xk8zYWUkhi0RsVUo3OlGoVrMX6mghDSYwLACxiEtMoHh6mhmL1tgRs9SCZzlmsbApJkkxOd6iKzndItC8vMAOvcXDXrDpRJeAMYFg8kpJ9FJBsR8WO4VnEqB4ca4Vs5ebUbBdqEpHp5vYsdbC1Ra2M6b3pD49dYEe4+bo8EpsNV6MJttiBRhHVoqwTTPLMXxmOV6IrMSYhEpELzZg7nID2xhn5poxd5kR01Kq8VZyNRKzDNi0k9S+yVGY6kWTwbuXBpWqTj0Y/CJ5n64PO7/T18kqRQWrD6b8vawyQ6oyqhdEMQF2Xw92wrBnTzDwuv/6BIPi+//+4Zwo8gJGdJcnoMqk/V3q2wc8hQDF3yV/vz+4b/4c/Prd6h9ifJ0WBwdy4d6zZ1+AV0mr4X7LO1LbhJam2sDGhL7X9sv95jcY3z+U+SXAaw9x/EgJiGbD823o+1DoMwjj+/s83gRwVU2LJJglr8oMyhRwS23g+z5ATexXyAu1/FaCsEB6juuu5sfrMOXvF0D87S0ejJonbJrJnnnjbiGxppBR49OcrLrN5UswTuu4EleZw0Ep99ryxHt1Ix2Sv3pFrZC0oHxEekGX/vtPglaZXuwx1O/V8v1+1as1vac8tT8qTzwXeYasr+abb3pPoHrDhgIG7P3XX61D3/yU/vSJgwFSpTZZ1J/ZaofJYhU/sxVmfmdlR1t0mCDaprazb/5iviFtAY/FBq/ZxlJjj9XOhwkVlRV8gCBp1N/33/v3DIpJEkwqyeLHx6+tdYG1hteZXsIA3t9pQKJo/16h+tzVjNPvy/n6HB/odtxQsRFJlJvae4COxhA2w4oUmstTQTDnf0EFyVQnubYyYO1qxskg/jsmpLtXSYVb7N20h/h79tAzecj8zXXcc5gEMDL9QQb+nU0nAmPkQff/oIBiyYwP+ipAsSSWOojp9EECZnFaoX13v3/T4CXA5+c4w0/NorjE2p8AxgFQLOMTR5lB9sTPx1swZj7FJrZjuhKbmFWoM2yIWW5HVoET28rdKNd72ZFW7lYX5r2rgOJjwhFXWp6T4xgT4KS60ARptHqRtcGBMfNMKP5TF0upfzPHzNLgodEGLN/ogN0lYhovyrWDVKMpFBO9oxjGKiCWDrX0GD7LwKrR5Cl6YopJ8UAt36ugmIDxyBgdZi42IbfIASMDTQoRJU6VCSDzTyMpJidc1UYPO9wq2OFE2horO9uiv5eusyFxuYlVpuevNGNethlRSwwcl1jEKdbx3+Roi6TDJDEWUmMBihkYR5UzMJ48vwqx6QZMTdHh9YRKjIoqx5tzK7E4x4QtJQ6UVru4HsLTtRN5RTYU7rDD6RYHCAIcE48IWqv398s3f770gvdVUPygx6GavwKK+3gm7Au81G8GQqdgUHx/3w4k/8FJwwuY5jBiMOqpBV+cXwCoDE6dQ4Hiwaj3nzUPhUZy4X4QdenTL5pNwUDK+67fD6SM75xm0Hnt3jz7SNDkPvv5O/fBn708uQ7KmL3qvWibvO99pb6mZ6H6XKYN9U4F0LSHKNH5ELXChRFzHeyJesxCB15McoDCNJFKN5lryXLYW3Fg3Zblh6pDqGeh6xK6/o92Wto/aqXS366NKt0l/UPno00n6TZQ+vdN53K7GRh7Q6iPhy5flvk9vDIo7kCjJiwThTrStoPnxN6gmNTuNKCY0h++SErKvSS2CiDt/rKvzXHAJrnX+OS9o1LeCUVd+oSU5irSZaqTXFtPXusGOeWSJnSi7nsZuIo0fUGxAMoStyl7xnvWcS8fQvdQ+aw63budD7ZvBxUUP2gELyRikri9JADKgt7JpxV++E81oeNOD1rra1B3sxs9X7Nf8YBIng2925XnpInc1Y6mC8ewV5nYSfoZT2EAYvZDf+IWSDWf/3W046Nf/UGVFEsv1LrLuHz7Lj5734IxyVZ2rjU9w4H5hrOoaepGN/k1B3DnqzZ88elRmHUebC334vDVu+hmdT16eRfdXXdx4/xx7KzywKtINGlSqzZRKKZanOu8i667Iq8u+vv2XZz9nQExWVaYfv0xzl5pQ4fMr7sL1z+vQ0G2Hs9SuKVIcr51BAeb76Lr87OIc15G01dKZt1dqDvxJ6RF6FSJcZSQFBMoHhqpw4uxesxIM2Hdtr2o+aIVnbJRPd3ovHkJn+z3gxxuOdkZ10nUt/fg7vULsNVcQWunUs6dblz/4jOYNtkxK92EGYsMDIinLvBg56EbaBE6usCdLlw8eATTf3cdXbfbcLC6nNWoCRA/v/wgfnvhK7ZlYEp0d+GLT05izeJyjI6tYM/UKSuNWL3xV3j3Ty1o7xT0ojwba46g+EQL7t4RTgzoRFhM8r/DsQtNaO9SOurrHnS3BzspIP6uu9mDno4mnD7ZoKbt7kBz7WHs2XMEtc0dgb7u6WpFncbJgRgfe1HTTzna8RMop6YOrYE6EZ0bcFJRcdmzpw6tdOJIvE08zKePol00WWnzE/fn0NLVg+4W4dQh8P5sMzrv9KDzerBNyb76VrZvaTlHk9BJNHX0oOdmHU/iJ692cHlc8NfKyWereKcFXvvPNqp0+robHc21ATOHQPnKRK3eq6D4cG0zOuTY6+lGey8JnpyU+9D0disaTh0IWnBIgtgnXUczao+ISVYtnxxJ9ObvBpw6IE9kRXpewAgUH6lFs6yk5JsQNjwHTjWg9XbvOacmsMBQ+VqgwPXRABW1fr3rq/RNax32avmSxuUN1axEfK8eWGjz2x+ybjQfirKONrSj504nmllFKuxUs4kAACAASURBVLj8002d6LnTjoajkj6hefzCsX1B/VHXqoyls02BPg4+YZb5qeVR+4iGku+62xpwUkqKAxoLKj0Efyjff3QBTZo5v5v7XuVrSQ859k6RBoCyjgSPsVbUacZXaNrVaDQq/EH9Kuskywu+D8175KgkOP3evvMVtaf3fFNHY7gDTZ/WoK5V5b1uGh8n1f7gNitTH3rEeO64ekrtrwMn0dDaqc5t3Z1ovXRK42DFjz1yza3bjzNN2nmwHU08Pyr92R9NB1HjgmgaTK/w/aNAD9Lk2qXzguIlz8lxI36lE+SROr/EA5tLHMQHjxcxtsP9Hebv7z1/91KfDlVfLSiW7xm4ajVqjtSjvacH7RePqPMzz3en0NwFdFwWqtjy+3uNFy0oFhLaHrRcIEltJ5pOiPlagmLOT5EgU3goNX8h/W2vp/QqKFbf07NuNJ+m9ycDdVTfq+u7rCurc99pQS2pUitS7nulH8zxP6igWDboQV2FREwFxUHl9AbFyn13dzdwpxPtzU1ovqZ4cGOmIhTRiVbFGLy1k3YExGhCF55OLZcVfohfXxFgru3yVez74At88Fk72u52oY3Bm2pT/KTxKsiJXN0HFryWbMXUdBumV36OBgKoPd1oqG3A4RONuNJB+d3FzfOHhDT40EUc/uwWu2W/dakeR07U4+Dv93N8YOl5mupCsYwT3v4I+oN1ONRIedzGqQ/r8O4HdajcbcCElGO4zODvLm42XsXv/1iH90+3c51wswFr2Y5Yj+djPkYN1b31K1y/24WG8xdx/sJVNH6ptPPkUZYYS/tiAsTa38gYFyy1JKkGutua0XjpEhqb29BNwKyrGafeE7bGVcZPcZGSfXUbX/Z0o6XxMj453YiL7aKc21+cQFymCZOS9Zic4kbFWZFn160bOPFxHT4634qW7rv44vJXrCp+zFTJsYuH532GOt4Td6Hx7EW8d6AWhy58Bfq664uTSJhTiRkLqhG92AXnBVnPG/jk48/x8flW3LpzFy03Sf28G02f0MGDAMWHL7YJFfzOVuFAoKkVgiUI9KgLrwAtnejsIYDXiIbLLejgzWQnWm91o0f5vvEG+e2jg45gWwhxwkesN9ByetB5swkNlxrR/KUAA+TUQLjPP4FzUj0GnWhhxwfCOUPQ2AicDooTOKrTucAzRYWGO1S146Dva8kd/9cyPp4KpujdwdN1aLjUAmK5nnaqXwMazkkVFyXt7Q50fN2NdqLTpSaIMQYE27yotFXrrIDi7m50k78A2X7y6MDlSRsa8e1pOr2k553taL7cgIbLEkgHnzAerKfTVaCnowWNVN/LLaKPiR4BWxqpdUK+JIi/G9DY3B7g79OBdArQQTe6u9T+DKTtCeYbWTa6O9DSa85pvaCeFmtBMdNDA4pV+vSmmaR3Oy+Wgl4NaLopzErIyUXARjmE+vSA6naI7IbIGUdv2yTFBrezSXjHDJxiqzRpuNdYIqci5LejowVNTS1o1IDu3u3dq4DfwLx9uRntnT3o6SJXI+pptvZQJpCHnPM1/NhCBxld7QKQa07ng/pAcWzSQox+RzoXOYcTyvgZEO38oUFxoG6BsajynpwfAvxEanIa3ut3HuHNh8pPft6IdaOT6CTnJjq4EwMJ9YcEL50414CGJjE+OluE+l3dabG58u87zRsZBGjXiGZlLJJ0QB6c+OUa3NnJjlh4LMq+D6jD+QPOYvqjaV+69Ob38P3jSyOhGUdacmRLzFE6HEIjjYQD4hA73P+Pb/8/xn3bCxSH6kNeezRrEaUJgGJaJ5S5OGjeDawffhwmENnTjgZ5OHrgDM/d3deCTdxk2VpQLO19eTOlCdFEdZKSYv8+AYC7m2Vs5f04Q5JsiqPM65MKimUZwZLigdaxFq0kbCHNb8XeWs3vwfLIoIDih4XgA6CYTxxUwlD5LPpHD1o+U04wlAWagJqqc0/fCB11YhwJdkT9hdE6G5u/72fV4C0f3+DNQ9ufjuMNVqGmGMVmTPBeZxAmHW1xaCYFFH/xRyveSKUwTH/AB820kbqF9ysdiFhqR/I7dqzf+SHqyEHP162oKfNgaYETs38j1KevnHBgfo4db29yocIowh6QFJOk1kWlbszMsGLUHBOqzhHYuwV/jFCNHjLLhJd2/wn1N2/jZu1BvLXIIlSmowxIOypA5VGdnmMVT0o9hlP0CF04anZj6iITqk1eVPtP4U/0vOMKtipxiilEk5AS6wOq1UPt17jtHReOoWKXk2MVkxrT75QNa2fTcVY9L9WdQh2D9G7UffAb7Kp0c8zirF3HcYYw6e2r8OZYMCPVgDer6tFCNWo8hzXzdIhaYsSMVD3GbvsCjTxCu0Cg+LW4KszV1+KzG2342GTD8BlCnXpEhA2WCwS2b+LduAq8ObcKEyu/QCttlK+dx45sI5asMWFZngXplotcFoHiiwcdqDa44f1QOJESNukam6AjwgkCbtUFNn48aaEHWiCz56MrDA4ZbO5TT9AOXyECyBMy2gxqy9Hw70fB5RA/inKAjkskgVYkLH6FR9GBBs0G+X5sivcqm2TViZDw9tdDh0d0OnhM1l9MftR2YUOsgmK1PmIC7NemWNmgq+kPo5GQVQBoCxqo7+W9Aop5PO/VtF+Chh601kr61eDCtXZ0tqlAmfM70sh90n3tpPK9UPFhNaQAPf3wf9qI9jaSsCn5aYLVy80+5ScBGdm2yPrKPuoN8vf9Sdi49tyUEvk/4nRDKzo7BLCR3/v31aKVDs3aVIdinKdWVSoEKA58ryyGARtuGr5B/CLVrLQLi9qPYpH5IKhu9Izz19RNlKfMm3dacFZLP4VegYUrBI/z95qxJG3S5Vjqe+It+V3yA91r5m32Uin7/zCu8AGlunAH6KFsLqj82psMm9F8OpifzrXw0VWQU7fefUDfa58JelD5fWnH9Hv/QlC/9v+9tn1++CXvXdcATb8f+861MOiXWlDiRB+sxkYOWQL1OVCHVmpOZ1MAtAtQrKaV/Xv4kiBaxyUNPQK8Fkz/szdU2snvVZMlVaIgQTGPMc38JMeD3FjJ+gZoquUnbXsC/B1cH/k91UWtj+QHDT3C7x9R+vgh7YMJCFMfMyBWQHG4/8PjQY79R2r8S1B8j/mO58TeoJh8xsg9AefBm+Lg/wW+OYwL1xWBjNQgJM1GzXysHT/BoFhdg7Rxk6lOcu4meu8jbTQhhxB1uNOOxpPyILYvKA62Kab5WWhU8soboo6yfmJtbkd9wGb64cz3gwKKtQz6IP+WoLiltgY1Ab38M6hrIuF8r82AssBLVYJAvRSJR5/ntIB+JjYfrReEncup69RtbfjNKivHKpaxiJ+K/hB/JMRF3qcpHBOFa9KA4jdTbZhZ2oAm2oucP862xRFLbUjLEw60LO8dwd4PDkK/y4OMfCfify1A8dVPHEhb62CHWnqLVywE5IDC6UN+sRMTFprxWpIZLhaV3sJv5goP08NijBx2KWeziyXMkxYKW+NnIw2YuOcGg9gvDhgwOdWMpfkncJZAaWsDVkbpMXGBCWU6D3uS/nU9UbEVvl5eqAkYyxjGlefvArevwZSgR8IyMzbudIKcgPl8HwrA03kVR9w+lFSewueECb9shKfIzqGX8osd7CnazhLcWzica8WcZSZsY5T+FQ6XV7FX6WkL9Zg8X4fRsZWwfkF07sIxs3CuNYI8UM8ox/MEiOlKzrdmlCPjIwH+awzlmJRUjZJP6f42PnWbkJBp4F98hh4zF1ThV5dEnud/a0LuZiv2XqQNYgcaD0lVanWDxQNTA+J40uolaZUqI8JbsljEmd8Cm0zxjJ0UcDlq/pIvQ5ajKVem28ub2WBtCTEuBmhTrPB/wHGBct/6uZjMAuMiUHc52fUGU9QGBRT3Z1OskR7K+ou6BoN6+U69KqA4xPf+feIdAVE1fV96BgBnYLHwg0MfUHgCeYqqbJq1+XA/9Olfyv+gwt+qRJR5QTlI0OZBgCE0n/Su52E0stKCKqHn7+QCSPUL9EPvb7X3St+ErLfyLuBoI1Q/avOSf/etmzhQEQePsr1nGDB1oEGRON43j4ess6yD5qrwqXZxlnXwk0rXPSXFdOpM07V6uBX49liTONDS8gkdWmr7YMCSXlnfvrTr06/98R6p42s2MKKewV7eBY2D5wDZnpqrNOmKwy1+pmyiaE2Tafi6TxzQBdEzJK+Jw7GQ403OJdJkSflegne1vNA8NxCaqHn0qn8I+oXTPi40EuuwkAxTm4T0mJ1f9Wu3/Li0PdyO8DgeBB54CJEEOHrAkf3B68r9zMvfUEfeW4Ras++njG+RdlBBsUT4kqkH+15IxHj/0+d/ZOd5VmvzF2KB5/ooz6XdlPQCx1c+ugDaL4pJ+QvaHHU1w5njxIuJVjwTS56mzXgqyoxyAocBUGzuA4oj9wig23DEyrGKo7JsyFjvxObdbpTpPWw3s3GXF+RUK0pJe/W4A0vynFi/ww2TTQApWgjMdi+yC514I8WCqWlW/O4Sgdc2/D7DhueijCBQPCTagGWl+3H4zGWcvHgLdTe62N5YEurSIRMW5NiRtf6EkBRfPs9OuN5MMWFXtYc9Sb/foAHFpDathG5iiTFJjSP18HEoKnDeZNNMNtB3uu/irrRt7W7GSbcPuyQobr6A7AIb3i6w4Z0tDqzeZMM+rv8tfLDcjLnLTRAguRX/Prsa5GhrQpIOY2dXYdSsKiw+QkrpXThursSr8VV4MaoCw1ccQPXeizh18SYa2+6ii09EqKVdIFA8dX41fvMFPbyFvfOq8HJ0Of9eia7AqMgypCt5nnJWY3amHvu53dImV/UMyDzBJ1kEOEmCrKpBfqoszuIEu16oqbfXB060eQwovNam8JPk3yCeU7wQCptFFdiG2jAy/2okvXJ8BUDxPU4gqT4ivQrujpEEVJ5e7lMkce31LBUXm2sV7IRSSRUngOopoqxPqLSyfHmwRZJqNb1YBNR7BRTfuhA04Yr36gZbTb+X1bmvXG9Fe3snuiUvEktowA5JrDrlyWRXB1qvN+DcCTXMDuXHdCc1Ydkv2it9qwFLnFZzcKHWR6qkq/1J7T9w4hzqm1rQ3kZB7VVbcJlnoHylDM5PM49p85f0pGuA3r0cYMj0AV5i/ghFPz/2H7933bi89+sYXEqAtGePqk1AUnUqT/TvPeincUwWqJeycMn6ijb14g8NHfq8l4A5YFOstlGkVQ5vAu+1/Ka0oV3EiaT0gXppxlPgmWaRlfUdCO3u9b1sT+80Mn/5Xt7fc7wrGjvt9Qr9QswXQTTRHGjtkd8q4b9EecpYlH4DtOPhjlgwJbAO/b2GPzVjMYjOA+n/wPwl+o6+l/ToTZ/wvZa/w/QK80OYH8LzxaMzXx6sOYGaM43o+LoHWsfJD2u+H1RQLCefB3WVG+pgSXE/cRVDbKK4XsrzntshPLS1iWdk10ZgR25S9pd6EJllx9B4K4dnImC8+bRQYfaQlDhIUmwBSYoj/E34EkDDERumZdgRnWVH5nrhVbpc72Eb4YKdbqS8Y+e0JKu8csyOJesc2Fji4bjDUmVIZ/ayRHlSqhVxy234PYO4NuxLszIoHhptxLAt53FFcbDV/eWXqLvUgqOHz6P6vWtcj6vH7cgudCDx7eM4RoU1nsewWToQKC6pFKD4g8sCFPujVDti8lA9Kt6AYVHCKzWD4rtduHyxFXUXW1Hf0IorTbdw82Yb2m61oe1GPT5y+1CuP4V6Elpcu4Bl+VZQ2KXlG2xIzbHg3YsCsP4hw4iINAOM5wUtf7/YgPGJOrwWX4WXYirxQnRlECh+OaYS082XcUOYJaP9xk3UXWzCoQOnsf1DAZ6PGcsxN8uAdxWp97/HlmPEzFKMmKH+Fh0SaWv0pRg9qwx2tpG+i862NrS1taFdcxV/N+KMEo5C8sQJp4fDPZFX6yqDUNUmSZTe5IbZ6oHT5YX3lIwxKw5ZPuXgvT3oVPisvc+VyhELmCxHtQVVNjeaTa4cZ4FNsmbDLt+FukrA23hIsRtWpIgsgWKQJ1SqpYMDkUdvoEH1UcCGZmPdf1pRfzmGVfVtpV1BdRf5SvAV3AalHl824AP+5rAIJk8AmOy829rR2tyI+nN1aCE267UR33vwNOqbW9HRpYJSdpymSOeY7uQQrE/fKPPFDcUvgQRP/Ug6RTs70PABtW8vTl+jwUCq493obG9H+40mNHx+mr02SlBM7ezT7/3NY0H0CtU3Kl2D8+ydduB1o/oF2ZlLzZrPpTaBPAz4jjwe1DalHfeiQx+pZ+829senlLcKiiWfBdNLlB/q2Xfq1xBtDF2G2o+yfvcc773pFGK+EPmEoEnvb7mOIh26OtB3vhJjoqWeQnHcS6uhd3/ci6Z92yvbHb6GaRPmgTAPhHng8eaBcze62cFr+6WTAbPFh9nngwKKHxaCD2yoQ9gUa4nG9dEs8EH1U9TlAmqifU6chU0pSWiFumUrPtF7kbLWiRcSraAQTU/Neh/7yF5YkRQ/HW3B0wH1aQveWGjHFFMjbpKV6+nDiM5yYM5KO7I3ubC90oMKI4Ue8GDddjfmv2NHtCIpvnLcjqUbHNha5oXdIYAUhT2qNHqRvs7BNsVJq+x4T5G07okz4tkoA0uKN31KwLILlz/+LWZn29h++LkoPca4roAUiVs+86DK5GFJ8Ql6cOU8hsfo8MZ8I4eIopBPIt9W/CrOiCGkQh2hw4RkEyLTzXghRs/A2ETqzLevoYo8UkdUY1hUNcbNMyI9z4pKg0fEKnb5UWk4JRxtsaTYgbQcCxasNiN+qQk2BqG38MEyE9sUbz4pVJ1PuUyYkqLHa/HVGBVdybGKjbWEgIVN8ciocuy5SH7K2vBegVChHhFBcYsrMP2AALonLJVYud4G8znSS/0Kh8vK8UJEeRAo1p8TeRIofjGiFAXHqfx2XPy9BzaHl71+E93pUEKGH3C5vbDaPThPzqe6ruH9rTZk55uRkWNEek4tq8qj+Tyyco1YU2jBtt12VH98Tdik14tDFglGrxz+Zpsg7QY5iH9DbHLvx6aYxsme40JltP1iLce167x6TGxqeWz0oKXuCjtVkmqQonx1Y6vWR9lY95HAqWm5PI3ELTCGP1UndjU/8UxKoCUoD36vqMJKO1zl4IGkVVob4ID0VCMBlHOEmt9+nFQMUqVNLIO+XkBXTS/rJ/qP+wgdCNWfnI8iFQ2058sGjkeo5qexc1ZAkrbfub6aeaxv/WV9FHprJcwB0KUccARU0YP7Rh5sUJgFsk9V+0tTN03/SbtXUscN0ErzXvI4Hbj0X18N/ajOmu/V8nt9r/Qz8aRKP6W+is1tQGLZJ072hSD16aDvFSmz9vCkdx9Qeu0z9XsFMGpoJ+q/L+iwo//vtfRWDxwufAM9jvVSkVbr40fAvOK0Qr8Q84VIr4LiwPcBXtPOT2K8aQ+oAumD+EULirXfk4QimOfk9wGafkN7Zfpv4qfw+2B+CtMjTA8xH/Uej+F7OTbC9Hl0JMiyzx70ejAooFhW9kFfQ22o+y0zsMArm4PAJvEEB5oOckaivNt7rgmtzfU4zRs6H/7YIJyRtJ47iIRsB0YmWvF0tBlP7r6M6yz2Ed6nSZ1aC4onpdkwNf0ojhNOu90E80Y7UnIdWFvsQZn+PZy42om7d27g6DY3knPsiPmVULW+/qf9yN7kREmVB06PAGQej4/thDPyHUhcaWPJsldRDfZFmfBctBEj4gzYqkiu/1TpQswyC4Pi4TGGgKfoW597oTN7kLP5E+FQ68p5DInSs6R4/XYX8rY5YbsgJLjvJhowNEqHETF6xC61YNoiE4djejlBj6m/vcHesz/7tQtDIqr5Nzy6GhNTDuHo5eu4eOoYTFYvKo2KpPj6BeRssmP2MhPilprYiZaZJcOt+HiDjWMVv2VoEI62rpzHegrRtNiAsXOqMWO76miL1KJHRlZgD6lvdzSimEIzzSRAXI4REb/Gr1ituwufunRYs8mGTP9lkefls8hUpMQjZ5Ri2rY61XmXsQxjYssxofIiH2C0fXYIuZus2F7mQGmVCwbzu/hTYyuu1p2Cp9SOt9eZ8OE1wtqNMKZUYlpyBd5KLMeUpM/QQPxw5TPMSK5A/OIqJC/TIes94YCrtdYFp8sH3/GrbL/YeZW8NKuAmyRO5662ovni6UCA9cCGMcC3/W9yxbggm+jevN7fvbJJ7exEt9b+UPFMrHW6pY6v4I2teC421toNs3geKq2oy8DGsAI2qG7Hg9sQcBAkPRL2N86V5yrYOYG65na0NkgP2Uq+UvVWkXZzKIBejtREm06gvqUFTQEP24pUlyw4uT819ZROpQIeHFUAotKTHK/Voz2USrZGRfu+bIoVx2TaMrQOwkL3zcDrJr5XvE3fahGq1L29USsHLn1ocj883pvn+V4pl2gTZHO7F7W3gtV4tQcios4yDTm9U6Xafv9+xQFXsEZBqLHHz3odlsgDBQnGA3QfSL+GaKPgPeEsLZAXpTvSgPY7Pei4ovCuQmPtQZBIrziy03q8DwGKRdoQ/a6MGe2BMc1NdXQQGPAuquHz4/VoudGEc3KM9jcWQ3g8pzqEpqkm/xA0CqJL+H3g4ClMlzDfhHkgzANhHvjuPDCooPhBI3ghEVMdjNyzvBALtEy/90KrCM1yuwX1Z4TTrjP1SngW3nQJwOL2nlbCHHWj/shJZO8+hPX+S/i84y7a2hSb4ihhY6wFxW8ttmFGph1pH9xkSWH3res48ocj+M3vP8Gfrt4GfXn32lns3O4GSX7jVnyGi7Svu92Gc2cu4uSR37LElRic4vZROKbM9XYkr3Zg3io7zLUCvPqijXguWkiKJzJYJafOdaiu/BA5uz+C7+xt3LlDpQGtn3uxrdyDwpJPhAT32gUMjSZQbMSKAgeiM62oVrxav59mwsgYPWYtNWNqqgkjZukwPFqH0bMNGB79IT4kETi6UHf0JDbuPIBV1Sex/xJJqoHm4/tRsN2Jwp2fCEdb1z/Hqo12xGSaEL/MjNhMIyQofj/NiIjFBiQu90F/lqS1BDjbcfXaLTRe+wpdPXfRSIHX0IWPDeUYHlmOfJbq3kXj8ePYvPMPWLvrY/zuYhe6uoX09xNbJXuaTn7bi7LTSp5ftuGLizc5vFPX3btovCbyPGYow0tRZXg5ygZnPfcKWs59BnPVH7DTeAwffvEVSCO989JxrF5hwOS5lXiX3GF3XEZ5bBnenF2BCQnlmDTnLNh315XPMDmxnIFxREoF5u4RUvprJ03YXemEzfEeatk9bA86b3yBs8drUFNzBvU3ROgc8uIspZ2hNubMv5pNruTngKOhNgqNVKuJY9z/ibAAp8E2spSf8PgnnNZ9EiTB6Qt09+w5Jg6YejrRcrkB9Wdl3Ly+aYmXKX8tKA7UX9ncqvdiw86hdrrb0XjuJDvWO3tJcahHHuWlhsc+YeeKrlbUnznG6U5+3oxOirVKgDOgPq0chqEbLZ+L/I6dqkPzbRp4Yk4R5avp2q+cw0ly6HfqHBqVQOXk3VkuPGJTTyGFukFpTx0/hpoz0jOjxuu4X8mzpwNNSltEOhFyQKpPU/nafuf6aOYxlT5i4lfvFXpTCKueTjQr7TtVR3QgQNPOoXdEerVvxL2mbudPCQeG3Ibgusn+o6vw9E+DNfjQQuQnASjxeD3OsENEDY9rJPratmrzl/RV2yfae0iG1LrdjLpTNTh2/KwYO13dYp4NaCyobZR5+WVIJvQI9fW2dnTQNNBBERkFn8jyAvXS8L+QgFObGtFQf0aR+H+i8L/o12OB/u9Lu0CeGiAny5N13LNHyQ+h+KkDVwJetzU0bq4L8GjTl9zZgbCCnH+o+YLr0BcU79mjSNS7ReipCzKOsQLCoRmLJ881Cg+k5LhOxvkOaZNMfReiP/x+CIm3StMjGnoPhB/60q//+S6cX1giFOaX8PiQc214PgjPB/3NB4MKirUM9yD+1m6ovzF/zWYyVNrDtc1KfFmGYvy/no5m1B4hQCzUXS12H/J/fR6ftghgKVLexfWaYxqbYgGKpffp+g+tmE42xMvsmJX1a5R9dBNf0V5F86/rxgW87/Fiww4Bikm1epcm3d3mz+BRwg+QxLi43MMS5VlZVsQvt6GSwWsrvJEmPMPq0yY8H/M+9pz/UnjhlmV91YKDR5rY+/TNWi8KS9zsFZrxa1MthkYZ8HqSETFZFrwUb4RFAdv7l1kxNsmA9dudmJpqxIgYHV5J0GM42xpX47m0w3iXgxDLggjld6Hug8OIiNVh8gIDYpYeA2swX/+cJcUEfiOXGBGTYYJeqf97yXpMStYhM8+KjLw92LKvAfUtXSAHXl3trfj4vb0wcFrhQGs4eZpOO4j3epXdde0iig8QUu/CSUcVUleZkJJtxLgEB3J+fRFfKE7Hutpu4qD/37FdUTWvMZRidHQZJsypwIyUvfjVOQV0BZp1F9fOf4qdK/TIeMeItFUG7L0iQPHu6FK8HlOK8XFlmDj7MwUUn2VQTMB4SlI5ptovsxry5cOVyMgxoGCbDVb7QXx2rYMPRgLFoAcdzbUB9VXiv1CbaOZjzSZX5evDqLtJByX0L9g7sJqm1wmaYg/aR8rbx5Ov/C70xnbv6SZ1HA3Aw/HAxrC6YT99VQkvIFvHY1TWSVz7jOWvu9H6eS3YhDsAikniVotmEVRayU3Y+LbW1QQOI5he+06iQQHBgYQUo7WXjUugj7Q04C7oQHOtCp4pz729wxhQn189jYZeno4DeUrw9A3zmOhftW9k/NpAvbtbURfkbVtNK3ljoHWT6f0SJGlD/8j68vUwapuD+434csA8HpSXtq/34iQdjCjO0riN1D4CvHy+I2Mx9m0j1/3ASTTcUJyc3elG580GnFS8mauHJ/2MvX2n0RTgHVUrY6C069Ov/bXxwBlNOUov3ulA09neXj5D0Jj4vjcvh5wviKa9aaaMpc9bCBAU2wAAIABJREFUA/TVzg1920kEb0dDIBSHVn1a22f0dz/90Q9NA3zWH43CzwMHc2Fa9ea18H2YJ8I8EOaBb88DgwKK+0PcsmO+z+/JrTid8B89qKjV+cQJiscr7H6X5DsxZr4DE9b9Ect3HsTCbDM723oyygz5Y5viKDOGxlkwLsXKNsRRWXbEve1ASq4dSze+C+t7R/H+B0fw3h4f9GY/ynRebN7lBuWfutaJlZud2LDzXXjeP4h/twlQTnbNFI5pS6kHiSvtmLzIitfnWTAizsS2xM9EGfFMlB5Doo0YP9+MyWkWvLz299hQ9SHydu3F4iwriso8cLj8MFi92FHpRlGpC6s3uzA1zcygmFSsR8Ya8VKCAXPetiL1HStmr7BhxmITsvJtLB0m4Jy4woJXEgx4LqIaQyP1HL94SNpvkUWS4p17MYtVqavYzvj56Gq8HKfDjDQTMtdZsWaTgyXFCUtNmJlmwNSFRryZpMOYBB1en1ONNwkcz9dh5iIDJiXr8Wq8Dq/GVWLy/Grs+YLEObdxeHclq08Pn1mBEZEVmJq7Fzk73kdOgQcjZoowTSMiyvBaXAXmLjNicY4Jk5OrMDKiDCM1NsWkQr2DVc07cLikDONnV2DKvCpELapGzGId4pf8CptMH6DM+gdsLdRj8SojluWZsarAgg3brJi/XI+xseV4OXI3XptVirGxpRgXX8bX8fGlmDiHQHIFJiWWY2JCKabOLcfM+RWIXVSJrFwDikrsHNu5ePd72HfwKI7WfISD+0QIsO88XsjF/ZH9fW0uH4IEZv+Ro9wOasOgj/cDR1iCKcdo6PxF2JqaY0dx6Bvi2hHdOaTbx8LzdOj8/PArIQOOHT+C/ZqNeOj0SvkfH1TiOquTspp+Lw5+LOabwwdCvRfP1PQDvQ8GHXveP4SjJKFVwiUMLD9RN6LLEa0X/1D9eVjYnEs12/7zFzQ5dlzljcHhjwM4TO37pv5T+qz/+hF9BTjELRFT+hvrd+Awjh7cFwRK9uzZx/1KtKN+vXd5A3x/4LDg0W8az8SjJKFWaPGN9Q/Vn33mB2rP4QDPa9uz9+BRXi8lbw1KeQeOqOvvgOoXlngR3eVP2z+D0h99+CFMb0nrMH0HOH+F+TM8Pge0/n7/+GlQQLF2wnjU//axlNgPu9OL/J1uxKxwYHiCFU/OEvGIJRCWV7InJu/Tz0Sb8Xy8BWPnWzFtiR0xy+1IfseBjPUOvLPVic2lLlQafQxOdSYvKgweth0uKPGgcJcH5QYv2+KSdJqkwzImn9nuQ06RE1MI8M4mG2IDnok0sISY4hA/G6Xnv99IMeOtVBOGxZA6tQFjkszI2uBAtckLvdkL8nhNYZ7I3rewxImkVVa8EG8AOeMiYDx1kRHzsq2YlWnF9DQT0nJtmJJqwkvxekxbZMSkFJIY69n5lrAlVj1UUxzjocpvWIQOw8gBV2QVXkvQYfZyE5JXWjEv24wZiwyIWGLg65QFBoyfq8PEZD2m5HwA94V2NNZ8iOy3TRyS6aWYKoxc/Ad8QA7Nem7AmVCNsQlVeIFCMpHEOKJMiVMsrvxsBtkXl2Pi3CokrP8A3s/b8cUHv8f0mWXsaGs42RYveh9/ILvgu9dhjSjFy6w+TV6oyxncTk6qRFRqFeKWVCNlhR5Lc03I2WjF+q02FBTbkJSlw5hZ/z977/UkR5adef4/s10FFIACkFojtQitdWZkRkqkFkitIzJDaxehVQqgipyHtTUObYdjy4cZ250H7prNC8d2aVzjzlA9kENb2hjHvrVzPSIT3V3brO5Gk0CPPxy7ft1DePg91yN+8Z1zLocOXRLdhhTb7rek0WdOYcCWxrA9jSE7h2Ebh1GCZCcHFQNjAY7lDJZ284imb5EvvYcvUkO+fI/vf0dak/pLnzvy+T/+UP2nvRY/DcW/2ff+V/UQ+I+X6/rn+tw//n3/4H//M/zd3/0F/sNHf0bQdfqDP5ZSaRpF5X6z1+7Hn698HvK1kn1A9gHZB2QfkH3gn9YHPikUf/n/WNLFl6oOk7K6df0e/Z4bPLdKQPzMTO2jQkx9VmTLVMG3VoLiGrrd0rrEjne3mD+5x9bVB1zEPyDAfceqSJdvCYx/h61VLJR+B2Lpe5RqEgj/7u/+y4d1biUo/l1kyrSW8R16pyp4ba3gW6YOl/DCSJWnpUJbVIG6zV5Em6PMlmh6YcyzQluLx7fwJe4R5r9jr/Ph+98FQTatlbx6TspzheUVU6j0pvcW80c3mN6tYf3iBqXad5g9uIFyoYzp3Sp6nXk0kSpN6xfXjRRjgmGCZFaJminItJ3FK0OWhVv3OvJS2PR6iSnETCleLDJ1eGymAPV8Ecq538f/9P9IIep/+xd/jn//7/5P/Pv/+Nf4zywl+B/wJ//m91hV6tPQHdZPa2gzCgyMWw0iK7bVopPAuFVPyy/x6DSJUM3+K/xe/TX/y1/+Of7t//LH+IP/4y/xZ7ROFv4B/9f//D+ijcFyGqQet+k5dOp59FsFKNwiBu0CK8I1YOWhm8uw0Glv+BZLe0WmCncaUugxUfh1Gv1WDgMWDmNOAWNOnkHxoC2NEQft46CY4qCY5GGcFeF8K4VSx7l7pMX3iKVJOf6O/Qny5c8fWVH4+Avsn248fxiKP/X7/8e/pqUSpFwQKvL0+1+QotQoOIb/9vf427/4M/zff/qf8dd/K+Xy01r0VLzrU18v+fXk+8E/z/1A+hEp+5/sf7L/PQKVPB/k+fBj5sMngeLGerq/De3333+P776nUOPv4dq/RZO9gm/MZTw1EgyX8Q1Vn6b2Z+y5qYxvzWW0OWlJphr06zew7dxg9eIOJ+F7XKfeI5n9gCxTbj9ALH7PVNzyzfe4+0DLAElqoXQNpW0K4U7nPmBqt8YqTH9rLrCw6VfmIp4bC8xeGAsg+/bBinhhKKLJUsCW94aFTJMynMh8qK+r+wFx8QPOIndMDTavVTC1W8Xa2Q08+1XYtypMSRYLHzC5U8Wh/04Ko3YSEEvVpiUIzuK1kUKpJSAmKCYYlizz0Laas1DNF+DarkC7VIRmsQDFXB5jMzkMTuWgYtt5DLh+F5E//Ev81d/9A8sp/q9//w/4L3/15/jDf/lvYHNl0WoU8MaegXYhL7XzOXSYeTTrODQzKH5suyw8hlwixqZ+B/E//Ev8Nb3m/yvZ3/75f8K/vv99aHRp9FjSGJsS0apLo1WbZm2bLo12QxpdpjTGXDwDXwqV7jLU4dfKocuQRJc+iV5jCv3mNIaspBBzDILHXRQ+zWGEFGNLCgpXGmOONCacaajdHCyLIt7uFXAdrSFXfI+LYBVc9u5nxl/6Y+a3YT7Jn+GfYiz/N/zJX/0N/ubP/sPjn2r1JcU+3fX/I/wprUX+N3+Dv/7TP8K//uSv/5u/Tr/3v/4x/tPfSmBPcP/f/uvf42/+4k/wR3/wm3/vTzcO8rnK11L2AdkHZB+QfUD2gd+ED3wyKCaoaxidaGO7AXufe/+772iNXcky5Q/Y8t1Du17DC3OJqcNPjQTGZNSXWtpu9L8xlvDMWMILSwmdU1Uol2uwbdewdHqLk/AHnEVvcZW4Q0x8D6H4HdL5DyhUPuD2/Xf1a9VoJSD+/vvv8P7994jw72HeqOK1uYguV4ltW7dqDIQJfkkVfmkiEC5IZizgJYVXGwvQLFdYkStfnN73HmLxPdLZ9whx9ziL3OIoeIvDwC22vbdYPK5gereCxcMqA2bPfgW2zRJ8sVu431XQ7ZBCokkBJvB9qc8we2UQ0WQkCM7glV5SiAmQX+lFZk0GEYPuHHRLBdjWizCtFlhesWK2gFF3FiPuLPpdGXTbaF1iEe1mAd2WDDpMBMEipjZL2PXWoJjJgnKG24w83jhErB6XsXJUZv1WHY8mLQemGGs5tBt59Fp46OazGHEJ6LGQ6sujy0yPSaNdz2HUxePdRQkJ4RYL74roNXFo0SbQok2iTZdi4dGD9Bx9Cp2GNEadHAh4e4wptOuS6NAn0WNMot+SYiHTw7Y0xpwSAJM6TFA8XgdhCqFWuDhoZ0gt5jG5ksG7syKSwi0CsSoCsRrK1fcPc+ZLnD+f+/yWz+/Lvj/L4yePX8MH5Pvjl/f7qjF21MrjJ4+f7A/y/bzhA5/b/eCTQHHjw33ZLUHpd7i9/46pu9q1G/TO1PCMqcQNIP5pGG5AMbUExM8ptNlcRqujguG5CkwbFUzt17DlvcW7q1scBO9xEb2DN3GPQJpyjD/g7r0E4o/XTjoP6t/XoXhytwbtSgX7gVsksu/h3qvVleJ8HYbzEiTrqU+WY23fVBHbvhsGw6XaB4ilD0iI7+FP3eEicsuAedt7g9WzGpaPq5jZr+I4cAMuf89yjI2rZQbJ6sUig+JmYxavyCg8Wk8QnEGbOYtOaxbNDIxFvNZn8JrgmB2nvogWYwbDUzm4tmgJpgJTji2rBShmSSHO4o2DoFgC4i6LgF67iG6riA6zwABYOZuFajYL41IOM9tF9DtE9FpFDDhEDDhFLB6UYFvNocPEo0WbBuUOd5s5DDkFBsc9Zh59Nh4dxjTa9GlMTIk49ldQu/mA+/ffIVO4x/xOAW3aFNq1aVZEi8KiRxy07BKPqbUs3u4XoJ8T0GskIE4wOCYgHrKlMeqomz2NUTspzKQcp9j+cUcaqqk0tG6etWp3GuZ5AfPbWXjDFUTTN9g+LSDO37AIhUc/eLxpyvvkayH7gOwDsg/IPiD7gOwDsg/IPiD7wG/OBz4JFDcUVhqoxja1X0T/w4f6OX/Ah+++Y+HK/Z4yXlN+rrWCb0gdNpQkY8pwEU+Nkn1jIqW4vm0s4pmZqkEX0Wwvo8ddhnq5BsN6Fc6dGmYPa1i5uMP6xS1WL26w779FXHyP6u0HfGDn8IFB+Xff0flI+wjYhPx7nEcklblc+4C4cI/h2SKeG6SQ6ecPEJzDt8YGEOfwrSGHdnsBa+dVVswpX36PROY9guk7nIZvseurYdtbw9ppDZ69Ktzvypg9qCKRuQeXu4d1vYS9qxvsX1VhXCYozoKqSr9uqMKGDCumRUDcY8+izSKivQ7HpB4TDEvqsYAmg8Ag17hESzBloZnPY3KrCM1CHsPuDAZcIgiGSSXuJFXXSTAsPaeb9pt4BrzDkyKD4tl3RXSZefTbBXSaOWgXMlg7KaLfLqJZm0KzNo0+K4fRKRE9Zg5dJg7DTgEDdh4TbgGHV2VUb+7ZuN/cvsdlqAqqHN2uS+GNJQ3LkgjTggjzooi57RwWd/NQz/B4Y6Yw6gR6TUm8sSQZEDfgd8SeYvnD1I7aUxixJ8H22VOYcKagmExBM81BNZWCZYHD1KqAnbMCQokq1vZzuAiUcXcnjX1jDn0R8+ejCAv5fL/A+588fg/fWbL/yv7buPd+Mb9f5Pkrz9+6D8j3L/n+Jd+/HiN+f9X58Emh+OMB+RK3a7cfsOG9xXNLPWeYATFBcR2M6wDcAOGP22+MRVAI9XNzCc32ErqnyhhbKEO7WoVhrQr9WhX2nRpM61VMvqvi7ekNvIk7FCrv8b4OxQ9A/B0pmB9A51OsfECm+B7lG8oJfg9SdpsseTAYZhCcZaow9QmESSVutE2mPNy7VQRTtwyGvTEJiPeuati8qDJgXjyuYnK7DNd2CUtHVfCFexz6q+w4n7/Hgb8G3WKR5f9SfnCzKYNmIwGvCOpTXvDYTB5jnjxoLWLLapEde6UXWPj0a70AsnZzBn2OLMZmsrCs5DG1XYR6Po/xmSwGJzPotYnosgrosUlGkEzKL1N/qdK0gUO7iUOfTcAgKcA2HitHRWampSx2vRVWGKsBxT2UV+wUGBT3WDj02zioPAIOrsooV+/ZHxF39x8QTtWgmhGYStxlTEFPKu5OHq61LCbXs5h/l2OFsnqMCXTqEugzJTFoSzEbcUiq8KiDIDiFYRuFUtMxAuYk+k0J9BoS6DMkGBgTEKunONiWOLhXBSzvZuENlbF+mMeht4hSRTqvxp8kX+Icks/58aYsXwv5Wsg+IPuA7AOyD8g+IPuA7ANfhg98MiimH/INo8FvbEs/8D//PoFpoXIP/XoVTxrqsLGIJwaCXYLiIp4aCswIgBvb1DIgrrfPTEW0OEromylhYqkM7UqZQTGFP5s3KjCsVmBcr2Byt4Jd/w3SuXvc3n3A+/fv69dMam/v3jMILpTfo1AmQH4PPn8Hw2qZQfAzfQ7PDTk812fwrSGP5/osnuslKH5pzONbfRbfGrIwrpZwEqL84RtsX1awdVnB2lkVyycVrJxWMH9YxuROGZPbJRwHaziP3GBqp8yU5Ov4DdzbJSjnChjz0NrDGXRYMmi3ZEC5wlQsy7VZhH2jCPd2EUeBCi7CNQxN5fBSJ+BV3Ug1bjLwLByaoNi6modzowDDUg4KTwaDriyGJjMYmhQxNClg0CWyPOBeq4BOE60z3CiolUYzFcXSc2gzcBhzizAtZTAyKUIxLbICWS3aFFr0aQw5OQw4OLyxcuiz8tDOiTjxV1Cs3NWB+D2S4g2sbzPooLxhfRK6WR7ujSxcqzm2f3Y7B7WbR6+Jcozj6LckMeJIYdRFodWk/nJQTlG+MYVRJzBkTWDAksSAOYlBawL95jjaNVF0auMYtSWhcCUx5khC7+EwucpjaSeDk6sSNo7y2D7OQ8jePvjBoz9IkQxf2nySz/fLvh/K4yePX8MHvsTv88a5Uyuf/+f/+0seL/l+0/ABeb7K87XhC/8c9+9PAsX0A/5Lt9v7e0SFW3ROlvDEWPjI8gyMCY6fGgvMnhho36M91efxjYGMKkLn0WwvoH+mBNVSCdrlMoxrZaiWy1AulWFap2JZZdi3Slg6qeIqTmvW3oPeX7qG1BIo36NUvUeudI9i+R6V2j18iRt0u/J4xgCYIJgsgxfU6qR831ZrFi+NWbxg+zPQvC3gNFRjtn4ugfDcfhkz+yUsHZexeFTG7H4J62cV8LlbnIaqoMfF+FtsnVegf0trCecw7KbcYREdVsr9FdFlE6Gaz8G0TICbx66vArFwi1zpFqsnJRYK/VLH46WWxysdGYd2M4/BSRHquQz0SwTHOUzMZJj6OzIlMsgddYss7JlAd2RSQL+dR6s+jSZtSjJdEk2adD1MOoV2Qwqt+hRatUk0a5Jo1SfRR9WgXRz67Wlm5iURZ8EySpVb3N/fM8uXbrF2lGeVpNu0CQzZU3CsiLAviwyOPVtZWN8KeGNOokMbR7chjkFrkgHwqJPaJMadKWaj9iSG61DcZ4pjwBzHkDWJPmMcXVqC4ihMc2nYl6jgVhKqySSmVnks72VwelXE6n4G2yc5iLkaOzcaf8m+/Hn1pd8X5POXfVD2AdkHZB+QfUD2AdkHZB/47feBTwrFRPUfO83n338cYMoxPQje4IW5gCeGEgPhp6ZiHY6pLeIbkwTGDSBmkMxgWILiZ6YCXpjzaHUWMOApQr1chm6lDOM65ReXoHhbhn61xPrm9RI8B1UcBmpIirco1+5x9/497t/fS3Z/j+rtewbMFFZLCufu1Q1em0kdloCYlGDaJij+1piFbbMMx2YJLVYJlum4brmAy2gN/uQNDgM3DHhndkuY3Clh4UiC4rn9Eo4DNeRKdzgKVHGduIGYv8Xbowp0S5T3m0WXLcNCpwmKCYbn90uYPyhDtZDDxGwWy8clJMRbVGv3SGVuMbmVR4uJ1GIeBMcExS0mnoVIkyKsnM3A+DbHQLjfIbCWVGOCY7KJGVprmFRgKW+4zZBmCnCTNinBsSaJFl0KTXUQpiWU+iwcRid5aGZFphQPOjg4VjPwRSooV2/x/sN7Bp3lyh1O/SVMTHFo1cTRZ05C4eZgnBehdHPQz/FYeJdhBbMIhrv0cbwxJySV2Ekh0gTBSVDY9JiTCmpR/rBkgxYC4gSGrQl062PoN8UxbIvDOJuGe02AdSENzVQSkys8No9yOPYVML8lYPM4i1xBgvb3BO7vpVDqHzefpD9UPnwgf5agn1rq0+s0ILtxvPGaj/NTesxjX5oXn2e/8fmkc2z8yXFfj7R4/DNB+vyNz0rt4+dpXK8v7X4ln+8Pj+fn7K/Suf20/8nnK1+Pj+9Hsj/I/iD7w+P3szwf5PnwzzcfPgkUP/wwratwX0b/Dvf3d7i/k9ps8RbW7XJdASYwzuOJvt4a83hCZsjjKdtPxyQjlZiM+nSMKcW2PPqmC1C9LcK4VmKmWy1Cu1KEnmy1CN1KEVPvyti4rOI0XEOUu0W2eIfqzR3u7u5wd3+H2s0dCuU7BsRC/hYze6UHIG6AMSnFr01ZdDuzuIhUMfWuiBYzKcciXuhF6N7m4Y1WEUjWcByoYvO8DPdOEc7NIqbfFeHZK2J2r8gU4iN/FWunZZyHqjgJVOHaLEC9kEOPg/KIpYJZFP68flqCP1HFaaiCmXdUTToL7WIWU9t5bF+UcBGu4DJSYeHQL7UcyF4xS6OZQp+NHPodHEamBAy5JFN4CIQFto/2KzwC1LMChl083tgpBJpjinCbgQppkVpMJi2hRIrwyCQBMYcJNxXVSuONNQXnqojrKOUQ3z5c02rtDr5wGRoPhzZtHB36OMZcKYw4JRt2kGIswDDPgfKIO7QxBsQEumNOKfx5xC6FSo/YaF8C484ERu0J0H6CYUk1jqNLF4FyMgGdJwXVZAL2pTRsixz0nhQcS2ms7mWwd56D8y2HpXcCUkIVNw/jf4+7X2o+SX7MfKfu043t/7+W/J/8jM2Dn2ol8Pyc5vHHn4Gd990disUbZLMV1Go0vnS9fvqz0HOkz9DY3/hcP9tv7Jfbz2nM5XOR/VH2AdkHZB+QfUD2AdkH/ql84JNAMf2r8fEJfxn9xg/jOxaqHEzfoGuKcojzUog0A12C4gKemiicuh4uzeC38AjFpBRTuHUdjL+hQlemPFrseQx6ilAsFVkOMcGxYY2AWAqfpr5zu4y5wzLL8X13VYU3VkM6e8tCpSl8+ub2DuXqHVOR09kbWDeKeKYjZVjKJSYwfmUU0U/rAC8XsHVZxuB0joVPExQ/14tQLxIU1xBI1PDOR+sQl2DbKMK6XsD0jgTE+9dVhNM3mCFI3i3iJFiF512RhU4r56RcYqog3WYWYF0tsLxhWkIoxtVwEqhgYb8E8woV3BJZ5WhSgOf2iuizCwyGX5NS/ADGHF6TamxIo8fKod/Jg6pKT0wLGJ7kMejkmWo84SH1mNYqpirSAoPidmMKfTYOraQQaxNo1ibQZUphxMVj0JFGjyXF+rRPP8/DFymjWrtlaikBEv3hEExUYFsW0a6No1UbQ585wVTifmsCPaYERl0puNZEViiLgLhTF2fFs6hwFinCBMfUjtkTzCgcWjmVwkQ9X5jgmPKH+80xdOsj0M6kYF/iGRxb5lNMKXav8rDMpZhCvHGYhW0hzbYDsTKy+RvUKGrgAegkVfPHzS/pDxV67O3tHarVWxTLNygQPOaryOSqyOZryBVuGFDe3Eh/FrDohI+A8nOcvz8NxffI56u4DmThvcogl689/plwR38m3aJUqqFSucHt3V1dPW/M91/mekpfRJ/j9fhx/iCfP10nefy+xO/nxx+B8vjJ4yff7+T50PAB+X4g3w8avvCb+H7/JFD88Q/WL2ebgECyQvkW6xcVPDM9gu/X+hy+JtBl6nCOqcCkBD815PBUL9kTfQ5PqG/Igbal/VkQGFMRrCZLDu3OHMbmJGWYVGLjGinEJUzvljG9W4KbQplZv4TlkxIrVhXja8gWblGsENTcslxYLnvDKkQ/02UgGSnBGXQ6MnBtF3Dkr8C2XmAhzi90Asie6wSMenI4DlTgi1WwelrC1HYBhuUcM9pePS4hLtRwnaiyYzveEqJcDZPbBegWc0wF7rIJaDUJbOmk2d0CrmIVpMQakkKNba+fSFDc76LHcei0cGi38HipSeOlhsMrTQqvaFublra1KbymgliGFDrNafTYpKJYBLy9Vg5DLg5jUzx6WT+NkSkqrJVEhzGFXmsarzUJZu2GJPrtKXSbU2jTUU5xHN2mJFOIL0JFFEo13N7eMiNQiqQqcK0KbFmlFnUMbboY+i0JDNiS6DHF0WmIwzDHQelOsRzidl0U3YYYK5o1QI+zSMWzhqxxjDoSGHMkMOGSlGLaJht1xDFsjaNXH0GnJoJhW4wpxAZPkoEwte5VDpPLacxuUhVqDqbZFFb3ROyd5bB/nkM0WUaloW5T1MCPMII/UkwJBEulG2RzVSRSJXgDBRycZbB9KDDbO8mAjPYFIgVk8xUUCgSVjfnw497vx5zTp3xMYxypzReqOPcKWN9MYO8ghUSqyCC4Wr1BqVwDx5cQieVZWy7XUKvRHw03zA8+5TnJr/V5+oo8LvK4yD4g+4DsA7IPyD4g+8Av6wOfBIobP1jpzRvb1H7+/Vvc3N6Ay9Zg2aSK0gTCWXyty+FrXVYKoTZIkPtUT20e31Cry+IpGUExPa5uj/0Me8wLYw4vTFm0O0g1zkO5VIJ5vYCFwxK2vVThuYLVszLc74pwbBXg3C5i/qCId1dleKMVpsSmxCr4bA1CrobtyzKaLSILjSbgbbWIUC9lGfCSIkjqb5NJxEu9yKCYwHhwKovN8xL2vGUWJu3YzMPwNssKXbl3CmyJomCygtWTIt4eFZEQqgglCZDzUM/noFmgdYQFdNl4THgyWNjP4zRYZoWrdr0lzO3loVvMoMfGo8lAKnAar3QpfKtJ45WW2hRe1o36rzTJByMwpn6zLskKZnUa0wx8qVBWn03a7rYkQdakiYOU4lZdEq/VcdYftKfQb0uiRRNHkzqODkMCjhUesXQZBEjkgzc3N8zSYgVzOxl0GeJoVkXRoo6y7V5THF2GGNq1MQxYEzAtpNFDfaoarY+h1xhjVaSpkvSAJY4+E+UJxzBoibFw6VE75QzHGAhPUBi1jY5H0aUJo10dQqcmDN10AralJMxzSain4jBMJ7D8TsD8Fg+lKw7ddBw9Q0AhAAAgAElEQVRbRyJWdgVY5hLYORGRK1Rxe0dAL30Oan9+PknHqzUJBsVsDfFUCVehAo4vs9g6FDC7loRtLg6jOwrzTAxWTxQWTxS2uRjmN1I4OBPhD+chZqQwZLpejTn88+/3q85ves2fPX/qS39Y/Lj3k86rWKzi/ELA/FIY80sRbGzHEQhmwfNF5HIVJNNFXF6JODnjGRhns2WImRIEscSAmeb7j3u/L+H+9auOh3TdP934yq9HPiVfT9kfG/cW2R/k+SDfD+T7gXw/ePyN96XMh08KxR87wOe+fUPAdHuL2s0NonyN5f8+0WfxFYPirATHegl4H2C33icIJkhuwPAPtU91GQmgDVm8smTRastiYDoPzVKe5fHS8kXX8Souo1XsXpUxf1iEbaMAx2YBcwdFrJ+XsH9VxnFAAuR0RlJl+yalfOFvDSKGp7NYPS0yYI7xVSjmsszGPFkGx8+0PPonM1g4KGBurwDLah7G5Sx0S1mYV3MMgiPpKgNjguB3l0UQIG+cFTG5mYdyLgPVvAjVnIgeB49eBw/FrAjDUhaaeRGkDNP6waQAEwRLJoFwA4YbbZMujU5LmqnDH4OxtJ1gcPya8oS1SaYKkzLcokugTZ9gucOkDjdr43itiTMobtHGMTGVZiD8WhXDa1UcA/YUjq8KyBcpN1eCYWqzhSrWj7LotybRpIoya9NE0UPVofWkGEfRoYtB5ab1hRNo00RAxzt19JgoC4WmAlpvTATJUfQYougzRRkYUxEtAuRhawzjpBLbougzRtGlDaNNFUSbKoR+UwSm2QQMnjiDYIUjhtl1Hp51DqO2KHTuOAPhhS0OCmcU02tphBNFBnEEkwRyEsw9fqbG5yuWq+AzFYTiRZx4s9g85DG3noZrMc7g1zAVhdYVhsYZhm4yIpkrDNN0FM65ONxLcWwf8Li8ziLNlVGrSu8hwfgjQP5681mC4sZneHxtaT/7jPXxejz2eDNtHK9UawiEspieC8Hm8mPKE8Tb1SjOznmEo1nEk3lc+UXs7CWxs5uEP5gFx5UQiebY/jRXZH4hfZZP9dk+Pk95+9fzE/n6yddP9gHZB2QfkH1A9gHZB/55fOCTQXHjRzq1NJhfRr+GSq2GCFeFconygjP4Wp8BwfHXugxTi7/WZhj8Si1tZ+pATMcle3x8vc+UYxHf6MgyrDI0KbzdrixG53KwbxWwcV5iocdRroIwgam/hNk9guI8JncKmNsnRbnIYJYA+TJSxuZFER02EQTEpBK7tgrwJyoIpaqYP8hDvZDFWaiExcMC2q0Cnml49DoF2DfysKzmoJzPQL2QgXYpC9t6DgfXRWTyVFSrjLm9HLyRMjbOClg8zMO1mWMwPDYjQD0vYsDFoZkqQBs5dJh5dNs4dFtJGU7hWwbFBMNJZi+11CbYNqnDPbY0dPMCFvaysK2KeGNP47U2gVfqummSeKlOMHvFVGAKjyb4jTEjlfhVfZv2Ub9ZE8OQI4Um6quieKWKsZxg96aAo6s8IqkSCqUqs1N/AUp3kqnDTaoIawl6SSEmAKbc4V5zFMqpBDrqQNymjqBdE0GnLsIAmHKECYgJeMne1KF4wBJlyjCpw0PWGAZMEbwxRNGjk5TiVmUAHaoQxuwRGDwJjDtiGLNHoXXHMLWSYiqxyUPKMcf6YzZ6XBybhwJiqRJqNQptvkG1VgNBYaUi9Wk7zZfgDeZweJHF0nYazoUELLMx6N1RaFxhqOwhZhpHCGpHiMGxzhWCYSoMw1QEFgLj+Rg8K0ms7KSwd8KjWKzU5y6FHdN7SSHov958JtCWXktqKZz549eXjj/uu2F/VjWeQ1BcqVaRTBewsByGzuKDxnQJ29Q1ZhdC2N1P4upaQCCUwdEZh7XNONY2Y7j0CghHcji/5HFwlEQonGXXT7o/Sdfxy7pfNf6w+FLur/L5yv71Jf0ekP1V9lfZX7+c3+/yfJXn66efr58EimlgvkSr3TSguAL1MkExwXADiDMgEH4AX10diOttY//HLQEzg2YCZ50IUouf6US8NIks97d3KodRDwFpHp79Io4CZUS5KlN6uWwF/gSBbwlT7woMeClX2LqeZ7m+cwcFjMxkWFj0SwOPXlcG+74SiuUaA2NSfwmGQ6kKjCtZvDbyeKblQPnAYx6RGSm/BMVkjs0crqJlJIQKdi4pJ7kIMVfF5lkBK8f0nhIUD03xGJ4SGAQTFFMBLMtKBpNbWZiWM2gzpfFC0wiTTuLlR2DcakhhzM1h7TgPCl8mkBNzFWye5aDwcOg00ZJKEhi/JPClbWYEwRIIfwzDBMRSPwpSinvMCbxWR/FaVbc6OHcb4zDMp3HmL+AiUGAh0e26GJpVEckIeLVRphKTGkzgO+qUllAiGG5lFn6AYiqY1W0gOI4wACYQJih+Y4ww66NjBjoWQa8+zLap36kJgaC4VRlCry4EpSsGhSOKIVMYg6YQDDNx2BcSMM8mYFtIQOmMYNgchnk2jrnNNC78ORRLFeQLFQiZMtJCCYl0iW2nuBIOLjJY2EzBMR+HyhHBhDUEpS2EMXOIbU9YggyKSSnWuSQQ1k/RNoFxBCZ3hIVTOxdimH4bw/J2ir0HjVMDUCUw/vXmN3utGwmE6fMImRILaS5X6u9TB+ZcvsLCnMvlKm5qjfeUHpPJlfFuPwm18RIK3QXUpktYXNeYmQ9icyeGswuOqcEHxyksr0ewvBbB0UkKF14eO7txbGzFcO0XwV67fi5f4j1LPueGX8it7AuyD8g+IPuA7AOyD8g+8Ol84JNAceMHNA1MY7vxY/rz7VfZuYr5KjYvS+iepKJadSDWivi6DrXUPtGLeEItKcN0rG4EvU+04oNRn2CYKcR6AS8MIposInonRYx4shiZzbEcYPtmHo6tPJZPivBFy+CzVVAYbKlSQ5yvYM9XgmcvD3ocqbpkpAK32QQ0m3i0WHgo57NIChX2vFCyjNWTAgu1dm3n0Gbh8VzL4YWOR5ORR4uZY+HPmsUMDG9F6N+KmN/Lsfc+vi4yEI5xZUTTFawe57Gwn4NtNcNCpak6dK+DVOIU2s0cXJsi5vZoWaY8g+PX+hS+VSeZMSBWJ/CtOoEmXRLaeQHX0SLKlSqqtSroT4hqtcoslChidieD0ck02o1JvK6HRr9Ux9Awyh0mBfiVilopdFrqR9FM6q6F8osjeKUmpZjComNoUkbQpIwy+FW4U0wh7jTQMQLiKDveUodiAuIOXQSd+ggGrDG0qinXOIw2NQFxmB1ro9xgTQhdujDeMJU4ggFzBP3mCAPjPmMEZL0GAmR6TBj9xjB69JRPHEKbMvBgbwxhDJrDeKMPoUcbxKApDLUrysDYuZSEzh1jYEygTMrx0WUGvmAWh+ciNvY5rO3zWNpOYXOfw/oejxNvBqt7aegnI+hRXaNb5Uef+hpDxgDGzAEQFCvtQWidkjpMIdP6yRD0rhDM7jDM05JZPRHYZyOYX0vi+EIAJxQZPEpzV5onv858pjGvVKrIZEu48IlY30ngwicw2G/4A7X+YAbb+0kkUwX2eFLFi8UyRLGIS58Am+saw4pTjKpPMaE9h8nuw5QngOXVCPYOEzg95/FuL8GAmKB47yCJg6MU1jejWF2P4OpKYEp44570Zd2vJGW7ce6/zng0XkP+/F/S95U8/rK/yv7auHfJ9z/5fiDfD+T7wW/ifvBJoPjjH7af53YF1WoFlcpHVq2gUKrgPFyCaomWUcrgqzoMEwh/pRXwtabeaqklGBYk0/D17QYgC3iileypjsczvYDXZgHtdhGjsxkGwvqVLNSLWeiXc7BvSkZq7eZ5AcFkCbmCdG6kpCX4Mg6uipjczjGFd3iall0S0GTi0W0XMOgW4NnNIcaXEUqWcBkmlbeMWLrEjr3QSurtK30azRTubOUwMs1DvSBCsyDCvZ3BdbTEQoyXDnJYP8nhKlbC1nkO09sZBsSWFRHjMzzazWm0mVJoNiQxPMXDsS5ibjeDqU1afomKaSXxWpfAS20CLxgQxxkUN+uTMC4J4MQSu+50/dkYsLaKaqXCVNBArIDZbRHjUyl0muIMdglyX6p+2Ah+XxH0qqMYdiZY+0oVwUtVGNS+VoYfjEC4TUvKbxhNyjCaVT9vLaowOnVhdGqlY63qEFrVQVYkq0NLIdBhdBAUa0Po1ofQawij3yTBb48+zPZ3aYPo1EgmbQfQpQ2gUxNAu8qPVgXZNTpVfgbMg8YwBo0hKBxhaN1ROJfimN9MY3GbLMXU35nVJOwLcVjnYjDORFg4tH46wvKE59aT8Kwmsb6XhvttHKOmILoUV+gY96FHdYV+7RVGjH6MmwNQWPxQ2QMMhhsQbHQHYZwKMdNPBmGeDsI2G8LMchSLG3Hsn/KIpwooFOtjR/OG5k9j7Op/bPyiuS49nmC4gnKlgjRfwOp2DCanH05PEFcBEflCCaVShRltH56kMLMYwuklhzRXQCiSwfklh+3dOFwzfoypTjE0cYrB8WOMKk+hM1/C4aYQ6iA2t6PY3SdFOIq3K2EGwdTf3U9gZT2M5dUwLi45FAqlhz9mftH5y8ekP6/k6yBfB9kHZB+QfUD2AdkHZB/478EHPgkUE61/fLE+tz6d2wMQ0w/7WpX9UCd1dHq3gM5JWk5JAl8JjD9WhAWmEj8C8cd9UokFpg7T8ae0FJJRRIstgw6HiGGPBMTWjTxmdvOY2skx0HXt5GDfyMG4moVrK4u9qyLSYhnlsgTGpXIVcb6MzfM89MsZjHoy6LQJ+IaWN9LzDK5jXAnBZBn7lwVEuTJKlQquYkWMTAtoMkpLImkXRbg2s7CtZRkMU+izal5g4dEEq75IAUsHWQTiJZwFC5jfzcC+JsK8koFhScTEDI9uK0FxGh3mFAPmpf0Mdi7yMCzxeONIQzXHM/gddKbxQlUHYwqhVifQakhg9l2GqY4NSKKxkPyDPivdZKQ2zpewfZaFZYnHkCuFDiMB8mNe8SMkS8owhUx3GuPo1EfRpCGlmIA4gleqMJrUEbxWhVkYNCm/zWpSiakvGSnFTQ1AVoZAajAda1EGGRB3EEirgszaCXZ1BL9B9BpC6NFTG2E5wwTKndoQg+Y2VQBk7aog67erA+jQBNCpluC4QxUAwbDSFYFjMQHzXByWuRgDX9t8AmZPHEZPPR+YCmK5IzB7orDOReFYiME6GwULc16Ns6rRO8c8jr0CFjaTGDIE0Km4QpfChz7NFYYNAQbFYyY/yJTWALTOIAubppbUYq2LFGTJTNNhuOYjmH4bxfZ+GvsnHHaPOYTjORSKVMlbmj/0RwaN14+Z39J4l9m8E8QiFtaiUBl9UJl8eLsWQziagSAWkEjmmDJMhbJ29hOYexvG2lYM69sxuDx+aMyXmNBeYFR1iv7xYwyQjZ0wMFbqzmF1XbG8YgLflfUIllbDWHgbxvJamOUR7x8msLIWxtvlEC4u0sjnfx6Kf8zn+Zzvb/L5f97fP/L4yOMj3z8egUKeD/J8kOeDPB8aPvC53Q8+CRQ/AOfHSuxntl0ul1Gu1K1cBpcpYeOigJHZLF4YCYIFpg5/peXxtYZU4roaTK2OFOKGOtxoJWX4Gz2Pb/QCnmp5PNVSyLSAZouA3skMFPMZBr6zezlsXxawc5nHzDsCYgqJzuDNpIChaQHTO1n4IkVk8wTGklEO5VmQ1OIsxj0CWkwcnmpSeKFPQ7skQsiWEUkVcezPg8uUIeTKTPFVzQl44+LZc2Z2Mljaz8K5ngHtJ+VXv8gzqKX3o8JXuxc5pjLT6zg3CKJFmJcFBsS0PjBBMa0nPOBM48Cbw/F1DlunWSzsZnEezCNXKCGbL2H9OIMWfQLPVTF8S6aOszDoAWeSVVKu1IG/UiFQkuzj8SA1kfYThMXTRRx4s3Bv8FBMJ9FriaFdH0OzJoLX6jBeqwhqI+gwRKCeSaDPEkULHVOG8YqUYgLjhimlbYLidlKEdRGQOvxaEcJrZRBNiiCaqFWFmCpMYdMEwW3qIFpUEhAz8NUG0a0LMXuEYaouTUZ5w360Kf1o1/iZStyjC2DEGsa4LQKVK4JxRxgKZwSmuRgcS3HY5mNwLFJOcZwpwaQGOxbicK8kMLkUk0B4jsA4Asd8DBZPBFbWpxDoMIzuMINklSOEbpUPnYpLvNFcYcjgx5jZjwkLKcV+KKwBaJxBqGwEyNf1fX6MGq8wqPPhjcYLpdUPszsIszuEhbUYdvZT2D1O4fJahJDJI5MtolCgkOrG2El/3vyiec/GtlxGJlPEymYMKosXbxRn0Fmv8O4ggSu/gAsvh+OzNE4uOOwfp7CyEWGFtOaWw5j0+JkSPKI8YSD8ZvTooe0bOZTUYsUJjFYvnO4reBaCmF0IYGbOD7fnGksrQRweJ5h6PLcYxPxSAGfnKfZZGn8+/aLzl4/942MsXyP5Gsk+IPuA7AOyD8g+IPvAb4sPfFIoJvL/+MJ8bn06N/phny2UcRIsQr+SR4tVygMmKKbcYQbADILr/Y/g+OE4qcJanuUOU7j0Ux0HguPnBh6vjDw6HAKGZjLQvs3BsJzB7F4e+1d5+KIl7HqpsnSOKcDtVhFNJg6jHhFrJ3mQcp0vSlBcLJVBavD6WR6aRRGvjBy+0XL4RpNmOcNLhzmchYq4ipYY4G9f5ODezjL4VcwKUC8IsK1lYF0VoVsUoJzjWRVpUoYJZhf2MgzGL8MFBtee3QwcG1k41gUYlgToFjkoZnmMuAmMU9DO8dg8zWHfl4dzXcDORZaFfDcg3hvOY3SKw3NVHC/qRmrxK00clmXuI8VRUg8ZGDN/qQNy+VGRlKBZusnkChQWXsRFMIfNEwmU9bMpjE8msfiOoK2EcKKAmQ0OPaYomhRhtLIK0hG0aii3OFzPGY6Clk8atsfwxkI5xmEpf5jlDofRrg2iiwpq6aR8YJZHrKV+CJ3MgujW07EwU4y7dKQIk5rsZ8W0KHyawqoHTCH0GYIYMIZAlaTHbGGMWEOYsIcxSpWlZ6Iwz8aYEQSTGmz2xGBbiLEwateStI9gd8IWwrAxgAGDH73qK3Qpr9E+7kPbuI+1HRPUetE54UOvioD4GqPmawbFSpsfSlsQE2YJjof0Ulj1gM6HAa0XvcpL9Cp9ME4GsX2YYlA8bgnAMRvG3GoUKzsJnHoFpua+OyRw5ZHNlR6iGWgu/aL5TX5Bf3AQAFO4tMJ4iX7FGdRmL1Y2CLwTWNuOY3kjimWC4ZUwPIshOGf8MNp90Jm90JouwaB47BgExf2j1ErbA6OHUOnPYLZfMih2ewiGA5icvmb9hbdBBsRbOzEGy/NLQRaKLYqFB7iXohQev8x+0ef5xz6vfPwX+4N8feTrI8+vz/v3kTw+8vjQfbphsj/I/tDwhf+evr8/CRSXy/RjmfIDqS2zlralfgmluvrJAKq+/+EYe470/AZgfaqW3rf0cF50HrRMTwnBhASmQ54MnhkEfK3j8ZXmF5ukHJNKzOMJmYaTWtbn8EzHgapCvzTyeE25v04BE3MijKsSfO76qLBVAUfXeczv0/JIIl4ZODzTpNFi4WDfyOA0UGDVfynvka5PWixi15uDYo4KZyXxVJ3EE3USz7UpDLp57HnzLGeXcj8PfTm4NkT2WIWHZzBMSrFpWVKICYqnd0QE4gXE0gWsHGWwcpDFyXUObw8ymNwUYVsVYF8ToF3gMTGTxoAzhW5rEkOuNDw7IlaPMvDsCHCt87iO5qUxp+tbls517p0IqiL9jEFxDC/qqnGrIY6zQE7yi/rj2bgwv3lUx39o3NkYPvgM+RbBVonloxYf9pfYvu1TET3Geh6xRoLedm0YPYYIuphCHEKbOoQeygs2RzBgiaCXimIZIuhgCjG1FA4tAW8XQbCB8oipXw+D1gRAQEz5wo8K8TU61X706gIYowrQzhA0UwTAEUwuxzFN+cGLMZjnYnAvxzG1/JgrrJ4MY8QSRD+Br4byjq/QrvAx61L68EZ7jUHDNTqVV2ibkICYoLht3MusfcyLHpUPw8ZrjJpIIb5mOcS6ySDUjgBTgSl8etx8jX6NFz2KS3RPXKBLcQGF5QreaxHxZBZvN+PsuMZ+DcdcCJ7lKK4DIoNi91II82tRhGNZlpP7Q+Mk3QNoLEsol6TxOT5Pw+TywTJ1Dav7GmO6CygNl7BTHvBSCHNLIbjnAjA5vVDozzCiOsHAxDH6x44+skO23TdyBFKI+4YP0D9ygIGRQ0xoTqG3XMDp9mFy5hqu6SvYp2j7iinD61sRrK6F4PZcwTN/jbOLJAQxX78v/WK/++HPKD9Hvi6yD8g+IPuA7AOyD8g+IPvAb6MPfBooZnBSRKlURLFIbQmlYhHFkrSP+rSfIIb2SQAtPZb2N4CaPe8HwfpnQfuX7RNkSu+XEorY8RZAha9I0SWV96mex1dqHl+rOXyt4UAh1F+pOfyE+rSfgLl+7CmptaQOa+ixaXytToP2ERS/NnFosfHocgoYniYoFmBaFTD9LoutsyxOA3ns+bIglVf3Vsr9JdB9rktjZIbH8lEG/lgefLaEXL4IIVPEnjeLMQ+9nwTF7PGaNLpsKRz4cigWpfDltWMqkCVgbIaDZoGHZVVgqq91RYBmnozD5mkGsVQeh1c5LO6JuI7mEIznMf9OhHONh24hBc1cGuPTaQxOptBrS2DUnYJxkcNVNI9oKo+ZHR7bZxkImYI0zvXxKhSL8IazGHVLIdTPlVG8qFuTJgbDQhKZXAHFYoFBkzTWkq/QxGqMPbU/pt/wM+Y/pRKyuQKm11Po0EkFsyg0+pUiyEKkW6kKdN1I9SXFt10XRDflCBvC6DVS6HRQyimmnGCtBMKkCveZgugxUOh0kCnAvfoA+gxUWTqAFoUfzRPXaFZco0VxjW6dnwGxYToKMq07AgvlBC/GoJ4k5bgOvzo/OlQS+DLQVVyhdcKH1gkvutU+2Oaj2DvlcHolwBfIYHU3iU6FD21jXrRPeJk63D52iY6JSxY2PcLUYVKIr6G2+6GfCkHrDLBtpe2ahU0P6XwYIbVYcwnaNkwGWJ7vzkES+ydpUDusu0Sf8hwa+xWmFoKs4FU4loHFHYR9JoDrIFWMlsKof368pPlN+ynU2h8SoLV6MaE7h2MmgI13MXgWgzDYvTA5r2CdvILB5sW49hSDE6foGz1G7wjZEfpGj9BLADwqbRMQ9w4foG/koA7Fh3gzcoARxRF05jMGw1OeKwbH9klSjn0MgpdXKWzaz44vLPlx6UtDEHIM7H/+/KU/on6s/8nPl69Xwwd+zP2q8VjZv37c/V2+XvL8aviAPL9++d9HjWsn32/k+408f365+fNJoJgBb7HIfjBnchLMcZki+AwVkCqCtgnwcgXKTZQAmSYre97HIP2R8vfxpP71tyU4p3M7DxcwtZPF4IyAb01UNIvUXwJhDj/RkKXZNvUlUJaAmEGxlvJ66zDMWo4994mWw3Mdh2Yzh24HjwE3z3KFCYq1SwKc2xmsHGWxc57F9nkOc/uUUyyi1ZLGN5qUpPxO8XDv0NrDeabmErwnuAK2zrMYmOTwVJ3AU1VDLU6gyZjC8mEG2XwRvjDBqgDXBqm8AjQLHPRLHNTzHLQLHFRzHAt53rvM4u0hqcc8ts4y4IQCTq6zcG8KmH0nwLTMQeGRgHhoMoVOcxxtpgSmt3j4o3lsn2bgXOPgC2dBEEx/cDTUWhpLyj9dOxTQrIuBoPhbZRTPlRG23aqL4+Q6g3xeGgt6PP1R8auOrfT8ug8Vi9i/FNBniaBZLeUKExC/UgRYXnCHLgRSfal6NFmnjkKlQ0wlJkju0FK1aQmKqbAWbbep/OjUBlhhLQLgDqomrQ2gm8KstaQSS0DcNHENMgbFWj/GbCGoXGGMWoMYsQYxaPSjU3ONNiUpwA349aFlXLJWasd8aB2j1gsKiXYtRRFPZZHPFxCOZ2GZDUsgPH6JtjHJ2sclIO7T+DBo8GHUKEGxzhWAzuWH1umHxk5tAFqHH+OmKwzovHijuWR9gmDPShT2uRCmFkJY3orBMRtEz8Q5+tWX0DmvMb8aZfsnTF6MGS6xsRsHx9cV/5+bq41xLSHN5eD0+PFm4hRvxk5gdl1heSOM+ZUQzC4fxrRn6FecoG/8CL0EwAyCJRAmGCYwJrV4SEH5xEd4M3rIoJiBMcHx8AEGxg4xqjyGiYVP+5ga7JjywuaSoHhm7hpzC9eYmvHB7rrE0nIAvmuC4jyD9ob//Kr+Jz/v8Ue7fC3kayH7gOwDsg/IPiD7gOwDX7oPfBIopnDki1Aeu748lo9zWDzMYfWUikvlsX2Zw8Z5DlsXOex489i7yuEyUkAinQdBar5QQKFACqIESNQ2jC5uY/tXPy6pk7lCgYVNLx1moVwU0WrjWeg0hUJ/RSBMqi9BsVra/kot9b9SpeuqcVoCYPY4SSEmpfiJJoVn2jReGdPotPPodXEYmOIxPMNjbJaHdlGAZU3A3B7lDeewdChiYU+EZVVEn4vDS30Krw1JDEylYd8QsHAgYvucVOUcM887Ed2ONJ6oJCh+QmCsTOCFNgnzioDzYA7TO5QHzMH4lod6Ls3AVjmbxoSHw/hMGrS9dSoiFM/Ds83DsJTGWSDLxmxyg4dpKQ3nOu3n0O9MotOUQIcpjjZjnKnF1uU0No5FaOfTWNrnkeTzTPFtqO/SGEljGIrnoF9IMpWYllZ6pozghTKMl8oIRieTCMayyOWlMf/p5/9q40++Q6855oqgRR1iIEwwTEbFs1poeSUtqb9UNEsC4E4CWx0pwFK/Te1nkNuqoqWUKFTaD6oeTfnEHVra9qND7WdKcb+RoNgvqcMTfjSNX6F5/Aotiiu0qa7RQaau95XXaFX40DxO5kXLBIGwFy3jl6ylUOk2BfW9aKXcYKUPHQovhk3X2NhPgBPyODhPo1dDKvElU4kZFI9LKvEbjQ/dSgqf9rI84XHLNYNhvcsPMo3TzwBY5/RD7bjGhOUKbzCrx4kAACAASURBVNQXGNJdwjITYI8ZNXihtFzB5PZjey+BQc0FA+MxgxczbyNQW68woD7HhOmSwTMpwKQE//x8pMiBIsRMDjsHcQyqTtFDucATJ9BaL+GeD8Ds9GFYdcoguJvU35Ej9AwfPhgBcQ9B7+ghBieOMKo6wZCCwqklKCalmMCY7M3wPkaURzDZz0HqsAS/XlgcFwyCJ6e9mPb4MOn2wjl5ibcrfvgDHFOK6c8G8ptPc3/51Pcr+fUa93x5fD7F95/sT7I//dD9+lf7vm1cy5+//8uvJ9+v5PuVPD++/O+bTwLF07tZWDeyUC5lMDafwchcBqNz9fV5V7PQLEvHtMsZmNaymN7NYeuMx3koh1Aix3JcObHwAEuSY0kw+7GTUejtT/cfB+AH9xNs16Gb8mg3z3Mwr2fYmr8vTVQgi5dU4ToUkxpMEExg/BN1CgTGXynTbB/brkPwE3UaT9Qplt9LFaEJil8a0uiwcRic5jExx7Pc3vE5HppFHqYVAe5tEZ5dkcHxzDsR1jUqhsVjyJ3G6AwH7SIP56YIOka5uXO7Iqa2RKjmebSaUvhGnQABMTNlAs/USYx7OFyGcvC8o0JaaQbApPRqF9LQL0owPDqdgnuLRyCWw2Uoy7ZXDkVcRbJYPhCgW6DHSWHTPTZpveEWfQwthhheaWMMogmoD7wiLMtpnPqzyNXDoH/2mheKBYjZAnYvRHSYYnipIiCO4IVCal+qI7Atp5DiSa17/DOEXoeUZ6Y+f/SniLT/p8ecPa/YeG4ByXQOhrkEg9+XigBeKoJ4OSFB8WuFVEGaKkkTFDNFWBtklaVJEZYqR5NC7EebOsCKZtGSS126AMsjJkWYYJjAuFtHqrEfPXoCZz8D51bl9QMUN09c4dEaINxovRIUMyCWIJhyhKdXYlA6g+hUERB70ae9Qg8rqOWF2hnE8k4C9vkI2kghrqvErXWlmKpNEwwTFHcpLtGr8mLY4GOK8KjpCgrrNdQOP9RUcMt6DYX1ikHxgPYSvYoLjJu8oPzhIT31zzGkuYRnOQyD85r1x/ReOGYD0Np9ME1ds2N7RwkEQgJTgilcnca74QO0TbDJwqYtXgwoTzGqO4fKdAGd7RIKwzkGlCfoGT1E9zDZEWsJgiUj8CWV+JApwwPjhxhSHKJ/9JCFTUswvI/eYckIiocnDmGwnD+AsNV5ySDZ4jiH0+3F5IyXQfHkjA9rmyEEQxyD9p/1vcZnkNt/5H76M3NTvl7y9ZJ9QPYB2QdkH5B9QPaB3xYf+CRQPDBDyquA58a6mUR8Y5DCk19ZBLwwi3hhFNBs49HmEDA+L8L4NgrPnoCtiyxOgnmchXLwRfKIp/MsPzSfJ3DKo5DPP4At6zNVWYKij3/cNrappcFhz8/nkS/kkRIKLDeXoHPUI6DVyuO5nsNTXR1+NVL7E1UKZATAP1En2Xaj/5U6xUD5a4JhUotVKWZU+IqFQOuSaDKl0ONMQznPM+BVzHFQLXCwrwuYeUe5xQLcOwKDY+cmVYfmYVzmGAwT2C7si5itw7B1lcKe0+hzptDnTGNgMoUX2gSeKOMslPqZOg5a7ug8mMXaicjAdmImxXKASS0mKFbMpjDhSWL1SMRFMIP5XZ6FQhMgU1ju7DsOSg8V0kqi3UjLKEXZkkqvdTE06WIsfPrIl8HxVQYzWxze7vMsrPfja9243tLY5Nl1T6RzmFxP4aUqjOfKML6lEGpFCC8UYXQawlg95JkKShAljTONpzRWNF4SYEl92k/j2XjPxuOp5YQcHMsJdOjDTBn+djzAgPg1KcWk4ioJdANs7eFWWjdYG0A7hUKzQlp+dqyZHkNKsToAUow7CYANpCITBFOY9DUD4A71Nbq11+hi5kenmlTgKzSRCjxBrbTdNO5F84QPrB33ommMgNhXby/RPHbJ8oe1k0F4AwL2TlNQ2oPo1fowYqKiWlcsr7hf78OY5RqkBhMQt45dPBiFThMMdyou0DlxgT71JfpUl6wdMdLrXGHM5IPGcQ2t45rBL2sdVxjSedE1fo4+FRXa8mFIS4W3ztE1fgqNzYeVrSgm54NY3Y5i7zjB7OQijePzFE7OU9g/SeLgNIUzbxqxhIhcfX7SeMSTIjxLAQxrzqEwXkBv90JhOMOQ8hR9Y0foJiAeITtC1/AhuhgQE+RKYNwA4sEJWo+YCmwRIB+w/OGeoT00jBTj/tF9jCgOoTedwuq8YGBstJzBYKGK1BewT17AMXkBi/0MzqlLbG6HEY7yyGSoWNjHfifdLxr+9ejPj/eYj/1PPi5fL9kfHu/H8nyQ54M8H+T5IH9/yr8XGj7wpd8PPgkUP9VLObn/QpPG/6Dh8C8+soe+msKUKQc3jZdmHt02Pybm0zCsipjeFbFwmMH8QQZrJ1SFOYtoKsdUnVwuh1w+h/yDEXg1+hKE0Q/yHzJ6HoVzngUymNwWMDzDocWSxnO9FPbMQJeBMAFwwyQw/okqUYdiaT899itV8sGkxyfwlUqyJ5oEnuuSaLOkMDaThmIuDfVCminA1jUejg0eU1s8XJs8prZ5TG7xmN3lsbjPY+dcxNF1BvuXIlYOBUxt8jC8TWNoKoXhqRQ8Ozxmtnm80MTxRBHDE2UMT1UxdFni7LnrJzw0CykGxmPTSYy5U6C235mAfS2Fy6DI1FvnOoedcwGJdJa9l3YhhT5bHG0GAuIIXmoiGJmMw/w2Bc1sAkpPAvaVFKzLKbaPXieTyeFxTD6+7o/7M9kce89hZwwvFCE8V4TxfCLItr9VhjBoj+DcLz6MbzaXY69LY9V4/ccxprF+fB/23rkcOD6Due0U2rQUJu3Hywk/Xo5Tey1tT1zjNRXCUvmZEtxExbCYIiyFQ9N6wpQX3Kq6ZuDbSSHTGiqWdc3AmFRhKqrVrpaOExRTbnAHmfoa7VQlWuVj4dEEwA82RiB8ySBYamlbMgJi2iZVePswiTSfRTKVgWctymCYllOikOcJ6zVGTD4MGXzo01AO8QVaRs/ROnqOtrFztI2eoX38DJ3j52gfP0e34hy9qgv0KC7Y4wd1lxgxehn0qm0+1ios1PdiUHuBjrEzdI6dsVDpAc05xoyX0DuvMLcSAqnB6+9i2D9J4OwyhTNfCkdnSWzvx7C4HoZ1+hoa2yVU5ktMzvlxeZUGz2fgD6axtB7EhPEMI9ozqIxnGNefom/iiIFw18jBAwgTDEu2j+7hfXQPSdYzvM9U4f6xA5ANjFNxLYJmAuLdn7K+4T2mFBttpyBl2GQjGD6H2XbG+jbnOcgsjjO43JfYeRdGtA7FH/vT/8feewbZVZ39nl+mbt0P975TE2reqpmpqTt3qnwtCcVW55xPn5xD56RuSd3KanXuk3PO53S3JKKJNmCCCQYMCAwmCAMChJGRESCTRBKoAbn+U89a+3S3CDa45LcctqoerbX23mefc9Zee/X57f/zPEusr9xbYl+IfSGOAXEMiGNAHAPiGBDHwL/qGLgsULxWPw8yBsD6PKuvbq/R8f1rDQsMmK8w5rFBE0B9fx4yirvdexgdBw/DcuAIusYIko/ggPMIbOGjCCaPIDN/FARa/CKtADE9mVh94QptAifafvSqq5ibMCmwmtEF1PbOY6slj41GrvoyN+lVUMwU4kJbn+WKMYNjAt8cSCXmsEzAnMF6piZzKN6gy2KLMcfU4paBPGRD81DtnGdw3L43j66D89g2sciAmDI475pdwKTnCIPhcOooEtkrEU5eCWvgCPbaFlmML8UCa3bksHOaoDePzXoBijUciqs6Muw8u2bmod6ehWQgi+beDCT9lDU6y2KAye2ZIHjMSeechy1wGJ7wEWwbz0O/M4fm3jSq2pMoMybR2p/GyBTFHy8yaKZyx+Q8dk7lMTqdRzBxZBlkr7ySu0Cv7n+q04MIMgK+fXPzqDTGBZWYQ3GRJopyQwz9Byi++AjcoUXsn8tjx3gOw4dy2DWVgzOwgNw8JZri5yu8B11furbzC0fZa+otlFQrzOCXoJjiiAmKSSWmslwrQLEujHICZG2EgTGBMMUN0zJLPKEWJcMKM7fp1g5ylSbFmACZEmvxRFkExOQ2TTBcT/Bs5JBcTXHAugDbVqen+GJSiQUQZkrxKigm92mlD6QSRxNcZaUHAsHoArp2RKHsoszREbQPR6DrC0PTG4a0neKGSSX2CEoxh+NKhRu1Ki+D4jq1Gw0aL5q0gum8aDF4oerkQKxo9zEYlhi9zDWagLhR7WaQPDgawe6xBFOGp+0ZTNvSGJtKYc+hOIPkgZEw2gcC0Hb6oDB50ayhWGEnGhQO5iLdORjA7oMx9AwHYejyQtPuhqHLA6XRhWaVA/VyUoS5kULcILMyIyhukFOdgzEDYgWHYALhFsUcU4upTkZQXIBjKgmKFVo72rvd6CGlmAC434POXjd6Byiu2IPufg/au10YHPZh1ppAJnsYR5fnkW+O38L8sXq8FepUivu/fb4t9JHYP2L/FMaCeL+I84U4H4jzgTgfrMC1eD/8Y9wPlwWK1+jyWKvLLdsaXQ6XWh50DMExL3P4sdyHzcYMKnrm0TK0AMXIApSjC1COLEI9ugj9nkV0HFxkCu9e+yKDx0SW4PgoriQjBfnKozi6XL9SqNM+vj2Z5a7Fpr3zqO3NochMrs9c7V2ny6JgBdWXQHcds/Qy/K7TpXGFlizzLZbGeu2KUYboEnMGVZ1Z1PZkUd+bRetAFvLtWZj25NF9II++sXmMztDySItwhg4jlDiM7PxRLCweRTp/lIEngeuYYwEjM/MYHM9DtzOLMnMaRYYUSoykFCexSZtCmTmJgbEcevZzuK3tTKOlL42W3jQkA2kGs8H4Ycz6FjF4KIcDtnm4w4vYPTMPw0gG6uE0JH0p1HYmUWFOomtflmVxdgYXEUkdRjC2CLt/EVbfIsad85hyzSORJoWX9+9R6mfhOrA+Z23adhRHjhxBJHkY5pGUoBJHUKSOYAvF+2qjaO1JoHN3GqrBBOrbY6g2xVBljKHWHINiIIGhsTRcwXlkckeYWy59lnj6MHLzRzDlzKO5M4ZiQQ0uUQVRrAoxK1EHUcqM4DiECi1Xiiu0VCf1OIgKWvOXklzpyG06zGCXlGGCXkknQTDfT2WjMYxGU4hZrSGAKq2fQXCjMYh6fZDVm4xBNJmoHUCl2o8ypRdlSp9QelGm8KFC4UW5wotGgx9T9ixTumnsUl8tLh6B1Z2FcSAMVVcQpkGC4zBrk2IsMflQT2CscAvmYmW10oVqJcGxC006UotJNXajXu1Gk9YDguA2I2WT9qCF9mvdrFR3+tA1FGLQe2AqgbHpBPZPJLBjbxQ928Mw9vihNHvQpnexDNIEwbUyO+oVdradAJiSZzUq7WjROKE0uaAwuqAyu6G2uKAwONGkJBC2ok6weqkVZATBBMD1slnUS8nmWNlIoEuu0gpKpDXLYocZGCvn0CIXlOK2GTQKRtuUOjvMXVwpJhgmAO7scaFngFRiN9q7nDB3ODC0wwe7I4l0mrtPF7wNxFLw7mD3sFgXx4M4BsQxII4BcQyIY0AcA/+6Y+CyQPFaAl7tChSv1RMAr4DxJW19DtT+HzIP1mrT2GDMobRrHjV9eTQNUjnPQLmyO4+6vjyaB/OQ75xH+/4FpqCSckxgfPTIERw9yo0GMAHGkaNHGIwRNBM8zfoX0TM2j6bBPLaYsrhCz8GW3KDXaTPLRlB8absAwCmsI+gltZiBMQfklXYKV2hTWK8jME5hoy6FraY0StuzKG/PoKI9jdquDBp6s5AOZaEfycGyJ8uU32nvAvyxw0hkjmDhMIfIhcO0jvARRFJHmPtxIE5uzwuQbcugxJSEfCiN5r4Ud6HWJFFiSMKyOwPV9gyq2ykGOAVJfwoNPSnm+jzhnMc+ax4d9J5TOYQTh+EMLsCyO4vWvhQau5Oo70yiroPOFYdiWwrbJ3IMgF2hBbiCC7D5FxgUTzjz2Debg9W7wD4zuS8T+JKxvqe6cD0K7YXFI7D559HSk2BAvFUdRpEqgmJSi/URVBqioORYxeoItqpoX5iVJZowqowRqAbiGNiXwtDBNDpHk+jencTIRBrK/rjgJk2u0ty2qoIgK1EHGBSXkGKrDTKrpIRY2gDKKdaXMkBrg6jSEuAGmXs0uUQTAFPZYKJt3DWaXKmbLSE0ETAbAqjTB1BDia0MATRbgmixBFkscIuZ4DnAMklXKL0oJyMIvqT0oFLlxba9caSyi7zfhD6jBwrZ3GHsnUhCSepubxjtQ2FYhmiZpQBaGRSTQuxCpdzF3KepTlBMbtSNOg+a9V406tyoU7sYGDdq3GjUutGgdoHqzVo3ZCYPTH1+DO+JYu9EHLsPxTC0O4yObUFou7yQGtxoVrtQr3SiRmZHjcyGWqkdtXI76mQ2aDu9ODgVx9RcCu19PpY0q0llh0zvhNzoYnAs0drZ9jrpHLjNcjgW2gTFLUobGuUchgmMuQs1QTFXiAmMSQluUcxCwmKKuVLcKJ0GWZN0Bq2KGWhNlGjLiY5eN4ztDhgsDtYmQO7occHU7mBQvHM0AIcziVR6HocPH2YP1NgDCZpHhIdqYlt40CX2hzg+hL+nhXui8PdVbIvzhThfckGmcC+I/SH2hzg/cv4q3BP/6P1xeaBYAOJLQTiHDYYctrbnUdWTR1lXDptNOVxhICjOMSheo02x+kYTP668K4+Sjjw2mXKgbXT8FnMOxe151PTmoRrJY3SOq5gcxlbAuADIVB4+fATeyBEMT9HSSHlstVBcsKACf03xXQ3Dq+tcGRZU4lVq8QZdGhv0XB0mUCYYZmAsQHGRMY1SSwYllgzK2tMos5ClUNmRQWNfBtqRHPrG8ph0cyhO5/jnpQFFSYAIjHP5I8jkj3Cw9y6gpT+Nqo4UlNszqOlMYYM6iY2aJLbqkwx+yQW60pJETUcS9d0pNPaksN+ahyu0CNNoBqrhDOa8tNbwYeyYzKGlN4kqSxLlJoojjqPUEMdmSoSli6OuIwHpQBL6HSkYRtJo351Gx+4MevZl0Lsvg4EDGQyOZbF3NodgfBGLh1fgmF0TAZSpTuDsjy5APZRkUExK8YqFBSAOY+syFIdQpAphq2DF6jAq9RE0dkTR3BFFvSXCrIbWCTaQqzQpxATD9JogtqoDDIoLYFyuCaKMgXAQ5ZqAUL8UjilmmOCYlGOC4XpjEHWGABrNHIjbOkklJpdpDsWkEEvag5B1haDoDkHaHkKzKYg6PcUVcxCuosRYah8HY6YQe1hd3x9GKJpn47PwMIHGa+FhDsXvmgZDbD1h40AI5sEQFJ2UYdqLahWpxByKGRArCIi5kULcqCXjKnGdyoV6tQtU1ih42WpwMyAeGA1jeE8EvTtC0Hf7ITW60aAqQDCB8GqzoUZqY3Bcr7RjcDSEKWsSgztDkGi5a3QzZYnWOiDVO9CisYNih+ukZAUo5iVTipnr9BxaKFZYQcoxV4sbKIkWU4rn0KrianGznEMxJdQixZgn2SIg5tammoWZgLiHK8Raow0agw2mDgfau50wdTpgMNvQ3uXA6O4AXO4UMpkF1veFyVssV0BY7AuxL8QxII4BcQyIY0AcA+IY+FceA5cFitfpclinz2KDMYvi9izq+vOQ78izzMo7ZhZwyHUYe208uZRl3zzUu+bR2BNEVXcaZZ05lHfnUdGdA6nDxe0ciNcbsuyca1kCrByoXd6Zg3okj93WBcTIjfcwJX4iIzDjddoWzxzBXitlgM6hvDOLDXpShdPLxgB5dZupxsJ+DXeVXqdJY52GlOIUrhCUYALijfo0NhnS2EAKMbMkNuhSrL1Jn0KJOY3qrgxKaUkjSmBlSmGLIYUt+iSKTUk096Vh2pXD6EyOqbbJLLkjr4AlfX5qLyweRjy9yJReUpfb92aZSrxBk2Du0xs1CZYcSzGUQl0XLaGURLk5gUpLAu17MgjGFjDnnWcu0n0H6b0oe3QOkn4Ow8X6GIq0MQbDm7UxbFJHmRVpo6jt4NmiKf63tj2Gtr4kK+vaKXt0DJWmOOraY9g1nUU2z1XPxcXDyM8fRiZ3GNHkImx+Upaz6NidRI05hi2qMDOmBhMcqwiASR0OgSvIIZRqQmhoj0AxEIe0J4o6cwTl2hBqTBE0d3Cj5ZHqTBG0dEaZKzTBMIFxkTKAraoAipUEvxyGCYjJKjQhVrK2im8ro/hebYCrwoIyXKsPoMEYZGoxKcFkko4gWswhBsKtliBknSGoesMMiFU9IUg7gmg0BNnaw2UKgl8ParV+VKs8KFO4wbYpvWgy+jHrzGB+gffXytilOFd+vWedaQbBdRoP5O0+6Hq563Sd2oNKuXPZagSX6XoNuU47GRCT2zRThjWUgMuFWoUT1XInqkjlVbkgM3tg6vczMNZ0+dCscaJGbkcVg147qqU2ZgTB5CrdoLCjTk5KsQ11MjsaVXZY+nww9XjRqnWgllyhad1glQ0tajuaFOQqPYdacoeWWVlZK51FHYGvEDfcIJ9Dg5y7SRMEMyhum2Exw80KgmACYF4SFDfLSRGehURJyvEMmtqmmEmUs1DpbejqI9dpNyydTqj1VkE5djAQ1pms0Bqt6Oh2YO/+ELy+NDJZrhTThM/njK95Ogj3obhf7B9xfKz8TRLvB/F+EO8H8X4ojAFxPhDng8JYoPKfbTxcFiiu788xCLbsy2P79AIm3IvwRhdBwEewRMBK6i0ph7QtmDgMmz8Lypq82zqP7TPz6D00D/mOeQbJ6405Fu+7VpvFaiNX68ruHLSjeRyw8/NzuFhkgEz13DyPoe0+OI+GviwoKzS5aa+G4m+vcwC+dJ8AxQIAE/wyKCYw1nMQJjjerE+BFGJyl27qS6N1gFym06jsTKPERACdZLZJm0SxMQnJYAbd+7mbcjixiPlFAuFVcH/4MIPiQGwRE655OIILIHdmxVAa69UJEBiTFWkTqOtMosyUYMmyqtsTaOhKMvglt+XOvWmYRtOw+ebhDi0w1bfCxJdJ2iIA8SYNh+FNqhi2aKKQ9CYwOJZBc0+Ctes6YpAPJFCii2KzOorNqgg2q6Io1kbRsy+NQGwBgeg8Zj15HJjLYWQyi87dSUj746gyRVGkiQhAHMEWJXeRXg3IDIyVIRQpydU5BM1gDKOTaYzbs9g1mWbtSsr8bA5D0h1FW3cUze0RVja103rEQQbCW5UciIsVfpDrdCmDX1KIgygXILngPl1Gyyhp/CAIrtUHUWckpZgDMblRU1ZpcqMmCG5t527S8q4QlD0ciFW9pBITIBMwk9u0D2UKih32oEIpGANigmIOxvKuAOyeLJLpBXZt+cMPfs2pHo7No3tHmCnCFXIXCIzbaA1hPQfiCgGKSfkla6BljoQ4YYJhihcmIziuVXIgrpY7UCWzs7bE4Eab0Y16lQPVcjsqpRyIq2QchquFkqC4Tm6HkpZSMjpZHDFto3hicpXmoGxlCbQobpiMllkiZZiAuLbtUoWYoJiMXKSblHNoVvGYYVKHmes0JdASVGECY6qTcSieRatyloExAXJBJVZo52DscKBngDJNu2Cw2KExWllp6XLA2M7bBMXdfU4cHAsjEEwjK0Dx6gldrK/82BP7QuwLcQyIY0AcA+IYEMeAOAb+VcfAZYFiFh8bpTVsOcyxH/xMxSXXYK7oLi5ycKWS10kxI2heRH5hEaHkAvrGeUKsjcYs1uoyIJV4jTaDNRoC2wxbO3izKYPGgSwse3OY8S4iO08uvAvsPIuHDzN33R3TeZbcaquZoDrNjam+aazVpJhxFZi3qU7byQ2a709irSbJVOK12iTWaZK4QpPEeoJiXVKAYFKA09hiSKLElEJNVwrSbRlodmagG+FxvgS/Fe08MRYDWUqQpUswF2j9SAY7p/NMLU5neZ/whwfUT0KfxBfhCS+wGFSC3KZecp2OM9tIyxzpEijSxVFCCao6kmjpTaFzTxqTrhxMoyk09yaxZyaLeIpU4zw021MoNZCbNCnEUaYSb1RFQLZJFWH72nenMOHIYXAsja1aDrVbdQTClCSrAMURbNVEoB5KoG9/CsYdSbR0x1BjjqJURxAcwmZVCFvUpA6HsFkZWlaJt1BdMAJiXg9iizLIEmdVG8MCGKcw48ph53iKLZVE8NvYHoa8LwrdYAzK/iiM2xNo66as0gTEfgbHvCQw9jN1uFTlR8EqNBRP7Eelxs8TZpFLNMUSMzgOoNlMyy0RLPvRZAlC2hliYCztCoJUYQUpw51BtHUE0GL2o9FEMcYUP+xGqdyNCpWXQTHVyUg1LpW7UCZ3oVbrgb4/hOH9Ucw40vAGc8jl+XWncu9kAi0mH8plTlSQyXm8cI2SXKZJ8XWyhFo8ZpgSa7nQZvJCaibzsIRaEiOtP+xCg5pUYIJfB6qkdtQoHKhTOlDL2jZUSq1MIa6WkUJsRXWblblJk0JcK7OhWeOAvscDfTdlm3agRkb759g+VpIazFRicpdeDcNW1LbNLhu5Ude2zYAU43qKA9bYINFYGfRSm8C4oY3KGTSSEiyfZfUGwUWauVDLZ9Ai5y7TlFyrVT4DlX4O5k4negdpKSYHSBWmbRqDFaYOG/RmK1S6OeY+PTjkxuRUBKFQBrlcYe4peJmszE80T9EfgcJ8JbbF/hDHg3g/iPMBnyvF+VCcD8X5UJwP/xXmw8sCxQuLpH4tYHFBMKoL8Ptny4VF9jp6bX5+AftteUiGsthkJBBOf8PWkkuzLo2S9jTahjMYHM8xpXJ+fhELC5QAagEH7DnoRzMo66AM0gS9BRBOYq2awy4DXo3Qpm1qDr6ry+W6JoF1q2yTPsldoY0Ew0mUWZKo7kyipT8F9fY09CNpWPZkYNqVgXZnhrlLFxtJ2Y3jCk0c69VxlJgSkA6m0L0/g3FnDpHEPPv+9B2YLS4gm19AIEYK7zycwTx69qdRpIthg5obuT/XWOKgssQQQ3V7HJrhJNzBPGx+Wn84N4EzSAAAIABJREFUwdpWbw6z3hxTiWvaY0zxJRfpTWoOwxtVYZAVacKQ9cdg2Z3A/rksOvckGdRuUoaxSRn6mvHjS3RhFGtDKKIEWmoqOdxuVvKSgJfqK22K+w2iXB9CsYZcnum4ADfFSkmu0JQluqkjzKC3TBNAkcKPYnUADZYgLDtiGB5Lomd3At27Y2huD6FE5cNWhQ/FZEofSpQ+lKp8bHuZygdmah+D4Rq9HzU6H2p0ftQZSS32o87gR7M5gHqjj6m/rR0BBr/SjgCUPUEouwOQdPjRTEmv9D7UaLyoVLlRxgCYgy+HYxdKZW6UyAiMXSiT8X2sZHUnmg1e6PoCGLcmEYnn4PKloe8PMPhlQMygmKu9NZT0ihJqKV2opSRYGhdaKCu03gWpyQ0drRnc4wW5RKs7vVB2EBjTa0kNJgD+ulkZFHMwtjIormJQbGUqcIvGAYXJhTaDA20sTpjcqQmcZ1HTNseN4oVl3Ah4awQQrpHMMCCmslBnUNw2gzpShJXkOs1V4AbZDOoJiNum0SjjVt82hTrJFOolU2hom0Jj2xSaZVMMilsYHM9AqqYEW1ZYOh3o7nPB1GmDUjcHmZpgeRYGixUawxyUmhmY2q3YMeLFzGwU4QhB8Ty7v/7snPR95i3xmO83v4v9JPaTOAbEMSCOAXEMiGNAHAP/IGPgskAx/chcBuNFSmbDFZnCj096ulCoU8naBNDUSatKV2ge+t2UdIu7PP9YkwYZqcRrNKll22BIo6Yng64DpBbPs9hbgkiKm+0dy6G+j+KICYAFVZjU4lUQTHHCy9CrTjKFmNRgbgmuEAuwTK8jBZnAeL2GlF6uFJMbNAFxZUcC9T0pSAY4FJt3p9F3MMuWNzKOZqAcpgzUSWzU0usTWK+OYashiaaeJPSjaeyZzcEXWUAmy4F4UXigkM1xKB6zUzxxhsUKF4B4szaOmo4EWvuSKCK3Zm0cxboY+g+mEYrNY781B2k/LWuUwe7pDBTbkssKMQExg2IBhgmIN2siqG2PYsqVxbQ7i4GxNMoMEQGEw9isIjBeMVKNCYSLNQTDBMh0bBibmfFY4RUFmMcMl2rCqLOEoR6MoX9/EtqhGEgVppjgIlUQWwiKBSNYLvpG28/AuFTth6Ivgt1TSewcT0I7GEG9Kcjgl2CYjJRhguICGJOrdIXaxxTigipcpfWhjuBW72Ou0qT6UvywpCOAZpMfLe0cisntmRTitg4/Go0+1Om8qNX5UEkxw6QGK91CPDFXh0tlLpQTKMsIjjkUU5vDsROlMifbXyZzotnoQf+uMCzDIVB8cEElrlTQkktOFgvcoCFXab7sUrPOzRRiSpolM7khs3gYEJv6aT1iLzTdXgbJLToXc5suADG5UFe2FeDYCnKZrmyzgmCYKcUsntjKVGFNh5tBcZ3cyhRiKmvlAhQLLtL1ChsaleRCbeVALJ1jELwCwrMrbQJlKcHyDOqYEQDPoq5tmlm9AMUEx/WSaQbFVGdQLJ1mUExxxW2qGVByLYVulmWYJjdpUonJbZqgWK6eZYqxzmiFQjvLoLi904Zde3ywO+OIxTLI5+fZnPOt89GqCVvc/y3ztdg/y3/DxPEhjo9v/J4R7w/x/hDGgDg/iPODOD+sMOA/2v1wWaB4foEUmNUmKJ4F5fPPlPPz81iYnwedI56mdXmzqOjkMcA/1qTwYzW3AhQTpK7XJ1HdnYZhVwZ7rXk4gnl4I3nsms1BvSODYjMlxyKFt6AMJ7BW/d22Tp0AM0ERXquOg2ydYIX6FZoEU3tJ9d2ij4MU4FJzAvXdSab8KodTaN+TRu+BDLr2klqcgnoHZYNOYLM+ztTijZoYivQx1HbFodmexNB4BjMerhZnc/Og/iCjH/DheB6jUxnIBhMoN8ewUR3BBnJ3VkdRoo+ixEAu0GQRVFviTOE9aMtCMZjAtrEU3KEcevclUW6keOAIO26TOoyNyoKFsFEZwlZNGPodcSTSefgjeVQYI9ii4W7PpBLTMRuVQWwSVF9Sfmk7uUUzUwWxmWy1KsyyQQdRoQ+hqTMC884oDtnSiCbyyOXnEUvmMXwwAUlXmC2XROrwFoVfsNX1wjY/tsh92KLwoUzth2k4goG9MbR0BFGq9HOVWOlDOWV9VgtATGsFq32o0JBC7EO1TjCtD7V6HxpNBMQ+NJn9aDJx6G2xBNBi8TNVmIBYYvGh1cwV4lqdF9UaihkWAJiWRdJ4UK/3oFzpQomMjINvmcKFWorv1RAgO0EQzM2BMhlZAZAdKJc5UC4VTGZHhcyOKoUD1Qoniwdu0DhBRjHBEqMbLXonKylOmJRhY58Pxj4vK039PsjMbtQq7Ax+q6QryjCpw9zmUCmdQ2XbHKrIpNzqFFZIdA40qgiW55g6XC2dBTNSisndWTEHidYGiY5iiQmGCwA8jRrJ16yt0J5CbdsUaiVkHIZ5nSvDdQIMk0pcJ5lEvWSSQXEDKcaSSbQqpplCTLCrM1vR0UNZpx1MJVYbZqDS0ZrFs9AZ56DUTqNNOcXa3b127D8QgNudQCKRZffUsjfGn5mTxGO+//wt9pXYV+IYEMeAOAbEMSCOAXEM/DOMgcsCxYWOoKcjhTqV37ddAMH5+QUcctLaxBQ/TCpvGmvUKwoxiw/WpLDJmEJjfxqm3VmMTGex355jyaW6D+QgGUxjs4FgmKCYXptgdYJiBskCHK+hkqnHK7DM9xeAOIErtATLK3C8nqm9BMQchsstSQbFpARLBlPQ7CDXaQ7F5BpNCq9hNAXFUBK1nQkUGeIo0sdRbo6joScB9fYU+g+kQGqwJ5xHIr2ATG4e2fw8CJDDifllKK4wR7FVT67TEVByrGJ9FFu0UQbKVB8eT7PXbx9PQ7c9AWcgxxJg6XdQQq4Ic5cmeGbu0gLoUhzxRlUI5cYIOnYl4I/mYfVkUdcegbw/hvqOGMr13EWaYoS5YrziSk0q8VY1QTW5Q4dQog0zCK41h9HSHYF+OI7hsQSc/iwDYVLB+QOQBfYQhFzm7b4s2nfGUGem7NOkGK9AcJFypU7bOThzMCZFuN4UAinHPI7YhzINxRFzKCbXaVKIKX641uBHgynAXKXJTbrB6Odu0gIASzv9IDdpUonJRVreRYqxn7lP12rJTZrWGPagUr0SP0zu0bStTudBtVoAX7kLpXInyuRO1Gk92L4/huG9ETTo3KhS8Tjh1XBMyjADYoJimQOUTKtcZmdWIXcwl+k6tRMFa9TSGsROrhbrXVB1eKHr9kFHCnG3F4Y+P/TkTt3thdRI6xM7ljNMM3dpUogJhgVjwLwMxgS/VtQqrGhQWFErm0ONjMCYYolneVZpuRUyvR36LhcURlqSyYpqySyq28hdmmB3FrXMdXp6BX6XwZhvK8AwlUwtZiDMYZirxeQ6TVA8ydynG6VTkCinIdPMQt9uhaXbjs5eBzp7HMyNmkBZoSGbgVI7C5maQ7HePIfhHR5MTIUQDNJyTPlL5qYfMj8V5rTvO5+Jx/MfB2J//XV/D8XxI44fcX76/r8fxftFvF/E+0W8Xy7X743LAsUrULuidP6wbXnk5/PI5/MMDg270thiTGKNJikoxVQmsYZif3VJlFiSUG5PQTuSZnG5g+NpdOzPQL8rjRpyVdaTCzTBbQIEvwyAmfq7AsBcOY5jrYofd6mSHMNa9YqtU8fALY71lLVZF0eRgSA3gXJLAs19SQa+2p1JWPak0LmXK8aW3WnodiYh35aEfCiJ1r4EmnoTTPnV70iyfV17UxidTmPWk0UgmmdKaiSRRzCWw5SL1gdOsuPLjATBpBJTDDBPjEXLKFG9rS+GUDyHZDqPnZNpDB9KwhnIstcS8BL4XmLL6m8IFBds2BHH0FgSB+ZSMO6MY9tYAsl0Dg5/BqOTSbSPxlkc8CZFCJsUwWWj+OBqUwgNHZQZOgLLSAx7p5OwetMIx3PI5HLsmubzVNLYyGN+VUnXm8ZJKp3DhD0J/bYIKnUEwj6uCsu9vL66LfNiC22Xe1Ak86JE5WUgXKz0okTpRbGCl7S9UutFrcGHZlKChXjhBiMpwz7UG3yoM/jQYvExAGZg3OljIKzuDaLZ7EOD0YsqjZspwxQvTK7SJXKuBpeTiqt1o0otKMRSB0pkDqYKV6qcMA8GkEzlkM7kcGA6jvbhIBr1LlQqBKVYakfZKitfVaft1K5cVowdqKElktR2NKgdDIplFheU7W5oOj3QdJEbtQeGPi/MAz6WJEticKBeZUO1jNThFRAmdbiybZZZlYSU4llmlRIq51DFIJdAeI4pw1VtMwx6KSa4RW2F2uKAqccFjcWBJkXBbVpQhFunUVMwUo1bp1AjIZtETSu3AhQvt1snULvK6lonUC+Z4GAsmUSTdBJtqmkodDMwdlhh6bKjvdsOY4cNasMcpKppKLUzUBtmoTFwKFbpptHVa8ee/T7YHVHEYpR5mo/FHzYv/bXzmfg6sZ/FMSCOAXEMiGNAHAPiGBDHwD/SGLhsUEyAUzDqgEK9AD7f3c4xIM7lciCLJnMYHKdEWZQAK4E1qoQAxwS6CWw2JFDdlYJ8OAXtzhQIoNUjaai2p6AbSbOkV0VGUnnjWKOOY42K4FYoCYBVpPwmWEn7yL7eXn38GlVUAOIY1qmiuEIVxQZNjNkWXRxV7XFIBxNQDXMzjCTQsScFw0gSpl1JUFsxlEBLXwJt/ZT8KoGufUn0j6XQuy+Fnn0p7JxKY9yRgTuYhTeUZSrvlCuNkakULLsSkPbHUGYgGA5hg4IAl9yow9yUIXTvTSIcy2J0KoltBxNwBbMMbCtNEeb6vImgWBnERkWQl0o6TxCbVEG09kTgCmRg82XQ0EEKcZQBcS6fQzaXZ9ckGM2hpYtihgPYqAhgEyv9aO0K4ZA1yQA4ksgik80iL4AwvZ5dUwbE5DLN2wTIBMcclAvb+Pt4g2kYhiIoVhXA2IvNci+KGBR7GAgTDBegmOrFSg+LHd6q8KCYTO5BmZIyTntRqvSgXOVh7tHSLj/aOv0MgkkprlR7UKf3slhhcpOWdXEXaUW3HwTINVo3qrUeVKjcKFW4UCLn0FssdaJC6UK9zo0aco2Wczfo0gIUyx2Qtnvh8CTZd6TvPZ/PI5PNweZOonPYj2aDm8Ox1IYyqY0pw2VtvL7cltpQIbOhQm5DlcKOaoUN1Qo7apQ21KsdaNY5oOxwQdPpZm7T5n4PLAPkSu2ButOFJq0AxMw9ugDGAgwTFEtmuLXNcChum0ElmYS36+QzqJHO8JhhigGWz6JNZ4XCbGNgLDfY0KKaRZ1smh1XLZlCdSs35kZdAGKCYQJjAYqpJDCuaZ1gVts6DgLh2hYqeZ2guL51HPWtE2iWTjGVWGchldiGjm4bzJ12aI2zkKomIVNPQWOcBblRM1dq7TRzo+7td+DgoQA8vjiSSUqyVXhAwx/EfPd8JO7/YfO32F9if/2Qv/fieBHHizhexL8/fy0viPOHOH/8beePywLFq2/wH1ovgFPhR2s2l8OumSwqO1NM4f2xOgEyUns36BIotSTQOsiVYtlQCortKVBJUEygXN+XQJmFH8vhVoBjUooFKC6Ua77RjgqwHEUBhtcSFCujDIjXqnm5Ts3BuNgYQ3VHHC19cabmKocT0O1MwLwrCfOuBIyjKQbL0gEC4jgU2+IMcndOpLB3Jo0dEykMHExh+0QSY/Y0U3dJ4Z10pLF7OoXBMYLrBJp6aBmlFRBerwxjgyKMDcoQi/8dmUxCvz0G1bYofJEsIvEszCNxlOkpdpgvj0SuzhQHTDDM4FgRRJk+hMGDSaZmOnxptHSHMedJc5gliKAHFfkcooks5P201BK9NoBNigA2ygMsVpg+6+pryOrCAw56LTsHAxKCEjpfHnmC7TwH4cLxqXQW2w/FUW2g2GFSiLkaTHUC4y2kEMs4GBcpqHQL5sFWOZmbQ7HCgzJaGomyQ2u9LJEWKcKUbZrUYVKJG41eNLFYYh9aLT5IOwiEfWhtp8zTPjQZfahgibS4OzSBL8UKExCXKZxo0LtRp6XM09xVupSAmaBY6mCwvOtQdLkP+fcTHgoIDwb84RT6RoOo1ThRKoAxwfDXrYLAWMgeXSW3ok5pQ43SzqC4Ve+EqoOyT7thGaR4Yq4SExA360hRtjGlmCXJolhiQSGukMyispWgmAC5AMKrIJlBMQHxLFtuqV4+i0ZaK1hrhcxgg9RgRat2Fi3qWZZNuk5G4FxQigl2uREg8zpXiKtXQfEKIE+gpmWcATFB8QoYc0BubJuERDXFYobNnTZ09TnQ3m2DwTLLoFipo3hiDsM6MynFM1DrZ2DpsmFoBy3FFEIgmEAqlRWgeAWMf+g8JR6/8iNG7AuxL8QxII4BcQyIY0AcA+IY+GcbA5cFigtgQ51TqFP559uU+IYfTyCczXGVMZfLYs6bhmwoiQ16UopJyY2jqjsHw96rMWK7AXvc12O383qMOG7AiIPX93uuxz7PDdhuvR6Dc9ejc+I6mA5eB8P+66AXzHDgJ7AcuhEd4zehfewGdOz/Cdr3XYf2fddy20/162DZfy0kw4sotqSxVhHFWiVZBATDV6gjWE9uy5oItuhiKDXGGLQS8JJarN9JMByHbmccyuEk5INxKLfFId8Wh3Z7Ar37k9g1ncKYjdYDzmDfbIoB8IQjzZRihz/N3JgHx5JsWSTDzjgkfbTsEk+ydYUiiPXMQgyU988lEYpm0NAZwfChBMLxDIvjLSdlmSCYkmKpAkwVZm1FABsEqzSFsH0sjnAsg5GJOHr3xpnaS9ciy1TfPHLZLBKpLDRDERRpgtgg9zMgprJCH8TO8TjSGeF4Bn0EINwYBBcAuTAeshxQ6Jh8nr+O3mvClkBrZxCb5R4GwVSSQkxAvFlGJijEMoJhF4rkbmxl4OxGUQGKlV4Gx2VKN6oIiHWUUMuLWr0HVUz59TL1t1rnQYPRw0C4rcMLSbsXih5aksmHWor/VZObNAEwd5UuZkBsZ6owwTDFCJP7NCnGlUpSigmI7ahSOmDZ5kecKZMr45vuBdafTH3n33nfeARNOidK22yCWZk7dWmbFWTlBMO0rJLMiiq5DTVMKbahWm5Fg9rOlGBSinXdpA67YejzwNjrhtLiRLPWhgaVFU0aGxrUVlTLZlEumUVF2xwIiiuYSjwNco2ukEwzq5TwdpVkmrlLU5xwo4oDcaNijgExuU63aefQqJhBg3wa9aQmk0LMjCvCBL/VkklUtxAMC+pxywSqSRlm2ydQ3SKoxBKuEBMYk5FiXMdcpyeYakxQ3CKfgkI7zUC3o8cOgmOtidykKZnWNMhVmlRjtWEaCo1wbKcV23e6MTMbRpBBcUbo/x8yP/2l+UvcX5i/C3O+2BbHV2Es/OW//+L9I94v4v0i3i+CR2Hh9+HXfy+K7WWmEueL/5j54rJC8eob/C/VCRIKx7B6lsNCMk1QnIF+NInyjjg26ePYYkxgh+Mm/Ozep3Hrvb/GT+8+xuyWu47hlrseBSvvfhS33H0Mt1B51zHceOejuOGOR3H9KrvhzmM4/tIfcOqN93DqxdN44eeP44WfPYrnV9kLtz6K397+OO66/Ul0jV/P3KXXqiJYNiVBcQRb9FEUG2KoMMfQ0hdjKrBqOA7jSByGnQlotseh3k4KchzSgRiDYuMILZuUZG7RBLPkIj1uT2G/NYVpdxqeYAZ2XwZ7ZpLo2xeHZZTOFYe0L8YU4Q2qEEglvoJie1VhSHqjiCWyLPa3vpMvU0TxwJUmUpIJnlcAmOrM5LysMoXQvSeGwYNxDByIQzEYgS/M4YGuB8VgMrjN5pDJZNG5K4YSLSnEfmxiYExQHGAwzaBYAOHCNf2ukgE3m+hWgNjhTUHZR8swkSpMUFwwDsOsLXMzMGZwLPNgM8ExA2UqOSgXK9yoZksmeVGl9YDgl0Ex1bVUJxj2osXiZS7S0g4fZJ0+yLv9DIxrtC5UqlwoVTjYGsOkDm9tc2Kr1IFimYMly6KEWUwZltmXlWIC4lKZHbIOL+yeJAMw/mBgZcKnNu/XLFKpDINncoteDcUEw2UExDIyrhIzF2qZlUFxpWwOtUobGjV25jpNqjDBsGXAC3KfNvS6ITXa0ai2cSBWWVGnmEOVjEB4ta2GYV4nKCaj9Yhrmfo7g2bVHFq1c2jVzDFlWKIhIJ5FLSnDzApAXIBfDsNMJWbu1BOolnAYLrhLExCvGIfhglpMQMxcpyXjqJeMo6FtHBLFFIPezl47ClBMCjGBsUI3zYBZZ5yBxjgNuZqOnUF3nw0jo27MzoURiSSRyfBx/V1jUty+epyKdXE8iGNAHAPiGBDHgDgGxDHwrzcGLhsUF37wM8AVlLHCNhpYhfrq/fzHKsX7ZZFOZ5hiOetJs0zKupEkWgdoDeAEtKMLuPkXT6Bztwf/9j+k+Lcftf1F+68/asN/+ZEE/+VHbVipSzAyHcOrp97AR8+dxK9Le/Cr/1OCh/6P5kvt31vw3IAT9/7scZS2p5lKvEYZwRplmNk6dRhXqCLYqo+iyhJlSjGpucptMZhH4+jYk0THbipJJY5BMUQqcQzmXXH0H0hgeDyJoUNx7J5OMFWYVOMZVwquYBpWTxq7pkkljsEyGoVxZwzS/ihKdDwu+AqKDaa1hdVB9O2PIRhJQzMUw7aDcQQiGZbsqlgbwnq5H+sJgAlgFbRkUgBbVBySi9RBGLZH4A+nMelIoq03it1TcXYNCgBBZSabEYAihx2HKNmWHxvkPmxU8LJY48fA/ihTkgvXcvX1LVzzb7/+dO4sfKE09ENhFhtMoMtgl6BYAF/epu0uAYRdKFJwpXiLzMVcqHnbhRKlGxUaN3ObLlO5UamhuGA3anRe1OvdaDC40Wz2QNUbgLY/AHm3D8puHwNi2k/Kb7nSyVRiguBiqQNb23gscZncAUqgVXCVLm6zoVjKjaC4QefEyFgIqXSGeT0UvvtKfxQeOGTg9ichMZJKzFVhcqHm9TmUtc2hXMrhmNXJ7ZmgWGllQFyrtKJZb4ey3cUg2NTvBpmhxw2Z0Y465RzqCIaVPHs0d5EmpXiGW+s0KtqovqIQs2NoOyXJks6gQUGu05RYa47FEUt1VuYqXS+fRq10mkExU4NJ+aUkWm0ExRyIq5gqzF2mq1rGQVbL9o2juoVboc0V4kNMFa5tOcRAmMGwZByNbRNokU9App5k6rCli2KKrdCaZxgIK7STkGkmGQTrSDlWTUCmmoTBPIPeATv27PPA4aQkWylh7in0P38Y983rs/LQ4tvHq7i/0Gdi/3z73zOxf/g9Io4PcXwU7gUqxfEgjgdxPIi/Hwpj4O99PrgsUMwAiiDqB1kamUwaqXQa8QQBMcXUUsIpirFNwbInAdX2OORDcYxYr8F1tz2EIskQ/qf/p+Ev2n/6b40Mnnv3eXHrLx7B/Y88jcThW1CtGcWPqjsRzF6PP772Bk4Hr8EjBMX/teYb9pvKARy7/kFIdxxZhuG1BMWKMNYqwrhCGcZmbRglhjBIpZUNRBkA63dwMLbsovjhGINa02gMtJ2U4u69MXTvifNyL4fZXVMJjNuSsPtSsPuS2DsTR9eeGEw7ozDsjEIxEEG1mVyhA8w2awLMndnqTcA8EoF2iAOuy5+CpIdihwmGfQIY+1GkofjfENp6w6gxB1GuDzAodvpSOGSNw7wzjHAsxa4FPZwgJZOuS1owuk67J2OoNPixQeZdti1KHzpHwwhHU0hn6Hp+nzHArzudOxpPo2d3GBVaUoQ5EG+WurBplREMU5u2b5Y6eSkjKBa2yZzYLHOydonChVIlKb0ulKtcDIor1VS6UE3LGdE6vx0eyDq9UPV4Ie/0MkgmWK43uFChIiB2YKvUjq1SG7a22bGVEmCRGzMBMQNl2mZfBuKtFK+rtKNrhw/hSIr3GeuHS/uDvi/1I5WjB4KoU9lQIplDadscSqks1KndNotSySwrCYwrKHs0ZYNWzqHVYIfc7ISm08XcpU39Hqg6HGgz2NGgmkMdwbN8FtWymWX4LW+dRsEIfCtap1DROolyVlJ9irlSV7dNgxJsqcw2yAwExHOQ6kkpJkie5m7RBMIEuV83QQFmINxM8DuBqpZDzKqbDy0D8XKdtjEbQ03zIRAU17WuWKN0HDLNFPSWWQbDBMSmjjmo9DOQKCcgkY9DzqCYsk9PoU05DqVmEpbOWQxss+PAmA9uTxTxeJL1+fcbm99n/IrH/D335bWPn8If33wB93yvuUi8ln/P11L8bOL4FMeAOAbEMSCOgf/IMXBZoZieBKx8eFoGpdDmQMCVxxXYIhgORNLwhtOw+9PMhfiANYUdk0n07E/AvDsJw2gCY97rcNXN92N9Y99fBGKC5v/5xzKo+qbw0qun8dnnF/Dll1/hgw8/wbHfPI9qzS5UqHbixp/dj49O/gHPtIzgkX9v+QYUP1nSjUevuhfKXVctQ/E6ZQQExmuVIVaSGzWBcaXlMMYzv0A49ROmCKuHYzDsjME0Eod5lCA5BqPvNmSuvRNuB28T8BIk0z5yXyYwnnalYPOkcNCaQPyhV/DYXbfDsCMCxUAYlSZSiLla3NQVht2bRDyZxvZDMUzaEwhFU0ydLtZydfgKuU+IHfajzhKEfnsYo9E7EMzcDBOBsSGA1p4Q1INhDB2MIZFKI53JMLU4laKHFfxGLIDxlCOBhvYgNsoJin0MjDcrfFANBGH3JhBLpNm19IVTCEVSSCQ5FLLXEzAzKKRz0vvQg5A0hg9GUKOnhFqrFeIVCObbC0DsYm7SBMdMIWYgvIjd4VsxFb4V0+FbMRu5DYemkihWuFChdqFWR2sLE+xS3Y1mEyXY8qDV4gHFElObXKbr9S5Ua5wsTpjihxkMSwl8yY3ajmo1JdSysyWXSpg6zGGZlGL1xI0IRw/D5o6zsU/fjfqRvufK+BcmtWwG8USKuTxXyASVWJLFvuAt2L97DmVSK7psN2POllnm38GgAAAgAElEQVQG4rK2WVRIZ1GjsLIEWtpuJ/Q9Luh7nNB2uaBqtzOXaYofrlcRENPySxQ/TFA8K8DwFANgco+ukEyhYtdVsAVuxMFdHJAryRVaOsUyTPfMXY+f3Pcsbrn6Ouw2WpeBmNRgcpsmGDYeuhYLtz2FqzNXoZ/BMQEwxQ1P8pKAmMC4dZVR3HArKccFGD7E2gTEBauTUFwxKcaH0CQbZ4qwudMKc+cczF1zIEVYpScAnoRUNc7cpQmImWqsmoCxfRY9/TaM7HJjaiYIfyCBRII/qKB56evXQ2zfjDt/eT/uvJmPz3/8/ngIf/gMeO/5a9m9WPg+195+D558/iU88+hduGkVLBf2F/5mZbO8P+6/+6ZLXr+yv/D37Ov9dS1uv/d+3HXLd+3/+vFi+6+5H6995AWcevERXEvX8Lrbcc8D9+D263hf/jXn++b1/we6fjffiV/e93PeF8KYvtzf5+a7f4l7buf30qX9fSeeO/kqnrz78vTXT5/5A9478xx+vvx7Ubw//uXGs/j3Wfx98ndy/18WKC6A07eX9KM0hXSKlMgUUqkUYvEUfKEErN4kZt1JTDppjdwkdk4lsWMigaHxBPoPxtG5Nw7L7jgm/Nfi6lu+HxT/5/+3Gf9fuQVHbrgbn57/HF//F124Gf+93Iwy+XY8++xLeOOau/Hr9WY89G+1l4DxChRfiR8rgszWKIJYowgt2xW03q8mhBLDr/DsZwD+9A7umIxA2k/KcQTqIXJ/jsI8GsVNpy4CWMKJnxEwR6HfzvdphsLQ74iiZ2+MuTCP2+I4MHcPjn8E4MPfIbEzAuVgGLUWnjBrqzaA/n1RRONJzLnjTFWedsZh2RlBmZ6WTPLjCplv2UgxrjYFoB8O49qTS8Bnf0C+M4h6SwC15gAq9H6oBkIIRvm1SaZSSCYFIzhO8esXCCeh7A9is8KDDTJum+Qe1Jn96BgJwbQ9BElXAK2dfsh6/ejdE4bLnwCdr3Dd6VxkiWQKe6aiqDNRRmk3V4YFRZirxE5skjqxScZLUoipzZViKh3YJHVgi+xZ/P7rF5ja515HYrsTlQS6KidTgAl6q7W0trALjQYXGgxOVKgdKFc6UK4gF2mu/ha12VEk4UYAXK6wo0xmZ4m0SmQ2lEitKG6zYquEl7kXLwDvPs8UycL343228vCncF/Qdrs7hia9nanEJZJZlEiewesAXr+Xq8O3vwHgjWeYUlxGawfL5tCosTHwJUWYYNjQ64Smy4k2ox1yC0ExAfHct6jDHIZJEV6xRfz89a9Yr71+7xQqKfZXOo1GZRiHf/sp2/7lha/wBQ3X82/jVjvFDU+gioC3NYDUrz/BFwC++Fw45txrSPWOo1JwlSYYLqjFVO5ZeBz3X38Lqkk1bh67xKqbx1DdfBDVTWOobRlDXSu3xrZDkKomoDPPMCA2dczC2D7DFGGZmuKMx9EqO8RcpuXqSdaWKcdh7pjB4JAde/d7YLWGEArRckyCer/K86FwPf7VytsefwGvPvPgsgdIOv0SztHt8vI3x+p3982DePqVF/DYz3/Ia/6Djn3uPeDLs3h6+Vpfg8feoIkZuPjlRVz8E03B7+GFu77989x24hxo2OPcS6v66NuPvbR/HsObSz+0H7/Peb/HMQ88jVdfeAy3LX/n7/Gaf4hjb8NjL7yKpx9Y+T7H37kIXHwPx+nzH3sTS1jCm8dW9l96Tf7Jt798Dlh6E4/9Da/lS+wtHuP3wur+/uUp0F312an7fuB98l3X5Amc/fIizj75XfvF7f9SY/tvOKbFfhTvpb80Bi4LFNNTrdVvdGk7xd2kUynEkykGX+QmPe2MYcyaYEmldk4mMTCWYOpwz744uvYl0L4nAdOuOHQjcRzwfH+l+H9bp0DPHg9OnX4LX37FfuKwH0WF/+564NfMhfp/WSOHbnAWb772Bl7oteHh/73p26F49ChThldgOCi0g1hHUKwOYav+V3jqPH+Htx+/Ga29UcgHI8w0w1FYHM/hD/SDDEt46TauDhMIa4YjaO0NMYA27Ihi28EY9s3EcciWZOXwWAwdu6LQDkfQ2BnEFk0A9R0h3m/TMRh3hOEOJFnm5goDLZdEccTcbZrAeL2MA/IWtR8VBj/yJzgUL/SEIO0JQtobRGNHAG09IaY8r0BdhoPxMhSn2bUj0C3VerGBqcUeEBQXq7wo1/pYuUnhwSZKfqWkbV6o+gM4MBODN5jEnCuOWVccgXAK49YYmjt8LIM0c40ugDFzmyZ3aFKLBTAmGF5uO7BF6mRKMQHyFtlxBpS/v9+JIsoQrXCieN9DePpj4IuXHmIwXKV2sTWHqzROVKqdTBGu1XFlmFyiy1hSLYodtqOozcaAmLtN21GmIHWYgNjGwJjU4mKCYnKtllhRpbLjhleXgA9OMNgv9N+l458r5oV92/cFUKOyoZgB8XdDMSnEBMQtehu03VwZNvSSy7QTum4XFBY7WnRWtGjnWFKtasWKuzRlly5rnUaZAMOUYboAxdqfvoUvvriATz8nECcgnkKDcgaDd72NL3EBz92QhUQ7i2bV1bjzzFfAH1/EmGQSla0TqMy9hk9xAS9eleTg23snnvkI+OLlB2EoQDHFDbdy12mKIU499znwxlOC+zQH4wIgcygeQ00LuU8TEB9CE7lNq6dYNmljxwyMHbPMbZpUYKV+CnLNBFrltH7xGINhAmSJfIIpyO1dcxje4cChcT+crgii0SS7Lt92Pb57vlrl4SD8cf5nef1jjNxe4h4N7LsVoPjPzd9f7w96zaUw8vfSP8ffA/De8ZXv9+RZXKTP+vi1wt+nO/EqPWx877lv/r269jjeu3gRSxcuAh9eCsV/+ftdCsV/+fgf0t9f7/+vtb8Fjv5D31/wALr899NKn37r91kFad+6f9UP63/K/auu+9/q+xWgmJ3/b9zfbG567wVc8zcbTysw8Lfqr8I9IJ7/Ms5v4nhY+Xv2T/Z75O/1frksUFz4wf9tJSmF8UQS4WgS3hApwwnsn0tg12QIOyco8RRPSGUYiUO7kywGzY4Y1NtjUAxHIR2MYof1Ghy58fspxf++UY1m8wFE5m/CJ59ylaAAxFT60z/B/7VFh//035rwv65TwBo8ghdvvBdPVQ1+E4qvvBfK0aNYIw9gjSKAHysCvE5teQDraM1fdRCbNQ/hN+eB9z88D3x4Cm5jEDXtITR2hSDpDcP6m4+Apa/wJYPiCLr2RGEeCWPsxpfx8psf4fzSEt79wxt48KeHsXc6ztyh73v5A/zxxOPo3ReBelsIjZ1XIXHsLZz9aAkXPv4ALz/+S7jGoqxvd44/hgf+8CZ+euRJPHL6PD7542twy7yCWkwlB9nI8xeA86cR1/hQSWv2dgTQ2hWAoj+IA7NRJJNX4d7jp/Hux0u4eOFjvHv6OO69OolkktvY7GE473wNL585j0/On8fvT7yCtNONzQo3Nsjc2CBdhOeBN3HqvQv45L33cOynN+Dwb97HyUfugKrfD/+Db+LM756Ht93LkmVtlDqRevw9/PYXt66C4AV4HziD3wvn+O0Dd8Pc5sCmNgdTiDdL7atKrhT//j4Hiijel9YTljuQpO/5/suYUDpAMFw3eitueOYsXn//Aj59/z288OsHsZ8UYAF6lf5H8MCLH+KTzy/g/TfO4BepGAPfUpkNlFBL5X8Mj716ju1/69VXcY0vhlKpFfp+N+49TRLRCfz8mdM49+kSlj49h7OvPIwbBVWc3RPX3IdnT7+L9z9ZwvkP3sMLj/0Se1tnUdw6g+LWpwWleAalkhlwpfhpVMpm0aS1QtXhhJFguI+AmJJrOSEzU4ZpUpDnUCOfQZ1yFlVSSp41xUC4rGUKKzaJshbBeh/EC59/hdfvfhBPnwdO3z+FVu0MlOY53PI7AuDn0aOhRFtTaFBMombqRbyNT/C4fwKVLRO48qWvgLeOY3fzIVSStYxDf9tbwMW3cD21m4TtzQS/hzB/7B28TRDy+Sc4/fpp3OEfQ1UTqcUxRO4/jdPvfo5P330HL97/Mwy0jKFeMoYWxTiUIzfi9uNv480PLuD8B+/i5FP3w2Wehlo/CVnsKbx0+gX8JPAwnjzzOb6ge+HBWzGy9+e4/6V38fHnF3Du9Ance5SSbCXYw53UTQ/jxJvn8NnSRSx98i5OP3sHg2Ua1982Z/11267Gfc+dxrvnPsPS0mc49/YpPHEn94z45vmewKkPzuLErx7GyXeE408fxx2pO/DEqbP4+MISPvvgNI7fe/Uln+/qe4/j9DsfY+nLJXz8zqX77ztxFudOPYE7nl0Zh6efuw9XC+PwiVPn8PEFAF9+hnP03vfTZzvBleJXHsbJt/l52ft+1+e+/wTOfvAZU1OXPjrH3q/w3f7cZyscs7q85Pi3T+LhB+jcp/DE8n0j9OcnK3PRfdd8V3/S9odx5gLw8e9uW+mzJ07h3NsncN/yOVNIsV/6Z3Bs9bbU1Xjqjxdx8YMTeIo9OLj0Idfqz83q19yHp9h1ovF0Fid/dQxnaBp4qfD5vnss3ELX548n8ODq97/mKZymcf4If/3V9z6FU4XrcY7Of+k4KHweds0/vQj8aQkff3AOp5742vt/775L4cZHTuJsYezSWPxG392Ih185+91z3Orvk0qBxhuNx8JnTaXuw4m3Vz4j7T974kF2zo8vCP34yI3C8XR/fIylPwEXPz23fA3Z9y1cz0fPMKX4zKOF77y6/FvdX08tf953Xyhcq1X35A8ex8LnfOw4Tn/wGZY+OY2Hv9aPhf678ZETOEPH0NxC1+cb4/gv3C/X3Ifjp99duX40V1zyXjfi4RNnlvfTHHmCKcXH+DW5pL/pc3/faylcl5tojqE5eAmf0ZhevtbC/qfoAdY5nPiz9/jqayzWC2NDLMWxII6ByzcGLgsUF6CJLkwiQQBFP0YTSCSTCEcS8ATjsHoSmHSQMhzHtkMxdO32MzjUEgAPR5kphmIMgmXbOAwTELcNRDA0ew0Wb7jve8UU/+f/3oz/e6sejuhVOPfhJ6t5mNUfevw4A2Zn7GrMBg9j3JXDK8++jLeuuhNPbO5YdqNm7tMExSNHsEbuXwFjeQA/lvnxY7kfLBO0OoBN6ofw5Hng1LGX8Ca+wnPX+bFJHcBWXRA1lnvw1EfAhy/+Hm8JUDx4IIIDD72DL/EVPnztd7jnoeN46o0l4KsP8Jurqa/iePZ9AO+/iG0HI9Bsuwn3vvkV8PkHeO3EKzj+29dZ7NxX757AoiOG1p7n8Xt8haXzF/Dmb1/BXXc+gAGpB+sJjKUeZutlHoQFKA5K3dhIKq/Gh2qjD23dAQzuC+NXr38GXPwMZ393Eidffh3nlgB8dBI/DccxZY9i4XmSwy/g1FPHcf3Pj+PJt78CPn4dEYJiaQrh314AvvgUv330Rdz9wO/w+/Of4pPzwCcv3I9yjRsx9v6vIyJzgoB4k8yFn50BPvnt/dhI4Cu9Bree/go4/x4efeAF3P3oGbz5BfDJi49A1sZheJNQEhxvkRag2I4imR2lCidKh36KX70DfHHiQeYaXa25A7/+kFyB38Nj9z6Fm+/5Pd6icz57L1tGqWjgMbxC7dOv4qajD+C2Fz/FF1+8h9ttc1wVnn8V71/8Cu+/+BJ+8cDzOPb6BeDiOTwYcODgdBiP0K9h+kO/9AFOv3ISJ393Fp9dBJbOHMPV7GHCL0HdiqUP8PJTx/Gze4X3f+YekPv0ChRPo2wVFNer56DqFNylKX642wFdtwOqDjuatXOolk+jRjGDSuk0auTToLhgyihd2jIpGHeZLmuZADc/Es9dYFC7v+UuPHMe+MMvZ6DpmIO+04rbTgN46xkGxPXyKdRIJ1E7TVAMvH43AfD1ePwc8Okzt6Gi+RAzco+u7DmOt/E5ns6OsXjhyuYxkBEU+284jl+/8RVw7jTuu/8xpCbGUNmUwx3kvv3pO/j1/c/i/odP4+0vgE9f/CU6ZeNoU92CZ+h6ffYunn3wSdzzwCm88yVw/vm7oTVNQ3qY1OrPcf6dD/Dcg8fxwLMf4Au6jz78BG89/zye+s3v8C7B3/sv4GdsLnoAp2lYf3oWJ59+Ak+dOoelPy3hzLErlx/2rJ6/CvUCMH/f9gO/ZxeZAfnzvz2BMx9dBC6cwTHhgdKl5yMYvYilzz7Gu3Sv/e5d0K1G7Y/ffg0nX3kNZwl2lla9/tHT+OxPF/EZ238Sr71NcLqEM4/yB1bHhHH48cdn8dorhf0X8e7xq9j3fODpkzh97iLw2VmcfOV5HLuNHghwKL54kX5on8TJV07x9/3yLH4tPDC45PvffgzPv3IW9M7s+Kcf5H24/NlO4eTye9NnozCMlYdq9PeBte88AfooSx+cXj5+6cIS/1EsHH+MXTRhLir0x3f2J70H/y7vHv+W9xPOmUpdjeeZmvws+xyFz3PVY2ew9KeP8f+z957RUVzbvu+nN+5479037njjfrrjfXjv3r1tkyQUQQmBpM5BORBEzkGpg7pboXNuSUiw99nBGxOO4zYGjAMYjI3BNrYxmGiDMSbZYGPAYIKxEfv/xpyrqrslwGEfzjjn7NMaY6pqVa1Ka81VXb/6z7Xq+Mu/A5fjlaND1svXIPJvxlE++SuJ9n77Nm7/KKCY8j7QF+j6t53GTdzE6dfFCxnKv/H4NQ77phcCK9ftx0Uqm0uncWjPezhK9Tx4BUdfTuSXz2fTO4dw/MJN4EdxLnt3iPJ9YNnJ5R8vD5Gfrx/kW6L+Tl+6jZvf031N9r+NOHqJTkq65mOn+beBXiJsvF/9rlzJQEUvC+VzXblSfnEg6oeAa/D7m7h5SfgAHRNSuaxcuQN76Rg/krsex/HDb/Nx3qYXFvI5xSHtfvX9L29ffD7sE2L/7Bc/DuLmzYtc73up/mQ/lq7h5PmbSPZj8pehdSHKjX5TN3P7Euc5ePs2rp37LH6dcpkN8c+frJ+VP3kcKk9qTlx/Rw7h0JFzuEbVefZtqX7WCr+Xf/u5fgnAxe8Yn8+Q8hZtjV4C0bp4XUrtOV6Xkj+u3DS8vV/D4N3buLB3XZJ/iLK48MH96lMsk8tjePmk0qnyIR9I+Yf0+zrs/p5qH7+ufTxEKJZAeAUNbrMC/QMrEKF+w8EVsHsHGIYX2/sxyzSAqc39qFsURPWiXmjnLodydh9Us/tQNqsPill9nFbMWo6ymX0omd6H6bZ1+PPTvwyKaaAtCqHWTLPg62+u4G9/+9sQMP7+9g+sIH977TouXbmKk2e+xLXvbuD2+W9wukeMRr3zv01CMhSPIBCW1WJDFKMkG1MeRUYVmQTF2/+Ip08COLkPKvpsUXkEuc98iR9wFW/1HQb1gvp4Ux+DbteftmHL6zvQYenD9OY+zGp9D6d+AC5+NIBgdAAHLgsoXtTeh/rnz+AmbuD45n4EIgPcX3fyHz7FZQzikw0xZJQfAB32m3efY1VYBmGajmY4DqJkSgSrPrnNSnGPgUA5AALlzPIgh0Srzftw7u4dnH9/DdffipUrMLD2IM7fHcSJV3qgbgxjenAnHv/nZ5GuIwgOIL3zE3yB7/H+HwMY23mEXwgcevb3UE0LwTgrgvzgMXwJAb302aQ+guYbp9BnSIRGb0yC4rF9J3AJ3+J1t1CGSR3O6PuUl73aQcs8bFmsFtO81Kf4hzv4gfrBSn1hrx/fC+dCD/LKPSio+DNMa3fhjz4P9w+mgbOW7v4WuPY5IloXsv70Ob7DFWyZ40K2hmwNoqtexOK5TuRo+/H8SQoPfgtGTTey1U5kq5djw+fAjyd3IdozgN0EI7iGz15O/Jiv/YDeel/D8U3UFp7B5nc/wDubB1A/x488XTcW77oCXDuJqLpzKBRru/DSOQDn9nFodO1ML+pmelE93YvKRi/Kp3qgrnUyDFtca/HndduxyPwnFGoFACeAWIwsTWrxeJWdrfpxgskreNNpR7HuVXx0Azj7Zjeqp7lQNdWFuVvOM+xvj0QxUWdHsTaK31N9cX9n6jP8Mj68TiHX7ShQCSgWcPwBTuEODj9pRYEExAzFKiuK1Vb8nsOnPwCFShcqLSiMfYLLuIyd3dSXmMyMCbGPedkbXTS41koEn92J5/q6OXS6dkoX3HuuAN+dxOP1HShfTddxBx8/2w61oR1KbRC7LgA4/TbmznejzRyAe98l4Mfz+GAF3ZPoYes2zr1D9SPuUa+/tQebn6X5xDKxLlGHtG7z5s147733sH79+qSHt6F55B+cZ1/eg73vbU7k2/QZrtFx37pffvFQee3Exnh+fui+eRqvyZDxFilh1/DZJtp+rYCMbw5hrbx+5VrsvTDIL60ITMRD+0Xsj0d2yNsIAKTzTAY+cd7iPG6evM95S7AtX19iKpVnfL18nAefW2JbURZ7zg8C1z5jMBDrpGshpYiub504r4sHBdBznnV7ceFH4NrxRJkN3e/w87q33Dd/TK8iBJDGt10nYO3aCVEG95bRsP28J9r2ve2doFjk/Wlf2IjPriXBxsqNOH4VGDy/R/gCw8c1HI/X40bsfGcHNsbTw85HVgxlv/jVZScd/8LeJN/azOc4FECHldvL5N+DuPDesPORzoNA6eegGNeTfUDUw83PX5PaxFCIpvriuvkJKI7XqfSC5Ne0L3pZMjikDNbyCxS5Xu5tXyvx8358CBfvDuLCB8l+LC3jcrtP+5PrMT79BfWz7ueO8yw2v7sXe5J+ozaeuMYv7XbScTYdF3WZfJ5cv0l+el8oFnXPdf0TdbmDiHxIe18JXnb1eNJLlXvrO1Gf9/ex1PpUuaR8IOUDD9sHHhIU08P/ANvAwAD6+lcwvDm8/Wjt6sdi+wBmmZZjWstyHlSqdsly1CwIQDe3ByUzejFhWh+KG3vZiqb1oLixDwVTepFb34PM2h7Utq3FH58SUPy//b9l+N//lxL/52/V+C//U3HPaNT/5f9T4P/JFkrxpSvX7oHiIYScnLj7N9z+4mt8pG/hzzS9P34mdq/ZCuPSVRhljGKUMYLR1Ge3PIKx9L3fCqEGZ1VHkFX1Jt4jpXh7GCOfJD34Kl7qCmO0IYy1NBLU6f1YOP+QgOKNvZhn6cMyex8sf9yEDW8dwd5DZ3H2wlXQsGDfHOxHtHcFDkpQvMTWi9D+q8CNM9joXI7ZbT2YtqwXnd4tOPIdcOn9F5FmEFB8cltAgl1ZIQ5gtNaP/JoQ5rb14NVTt4FbZ7C8gqDYjzFakZ8Gzsp89gv8cPcqvjx6DMc/OY6Pjx7D/o8+57DAL995Cuk6P9J0fmhs67Fy8xG8svcLnDx7HbcJih/3I/3pc7g9eAFPVvhRPjuCqYujqJ33DN67RCrgDqRrfUJJvnEKK6sCyDFSn2AvBBRvw1itB/N3fcuK7g5Siclep+k5XMIdHHzajQyNG5laNxiKtTQvoPjSscPY+sZhvPbGYWzbdwGXfgAuffQaasrdyDO6Md44gGXhHXj+jUN45+hFnLpCavTn6NO4kDlHKMW4cREf7Hof/xRYgwoeSKsb43Rv4dAt4MsjB7HldWGv7TiIjy7cAa6ewAsDEhTfPIWt0ssgeiG0cuV+XLwrXnBQe1ixYg3Wrn8bG988iLePSMe/fhL9qiQo3tbFsPvqFwC+2AddvQvVjR5UNXoYiI1T3NA1uFBS3oUifSdC/Ruw4eX30Rl4mvsLE/yOU9oxXmnndIHGgXwKm1aSUvwKRytc2rMeE7R2qKu34OBN4NxOJ8ond0Nf1w1l5Vq8Qir94B1cPvs1zl+jh9qvcZ5B2IYCpQTFW0XYNAFwgYLsfZwipfjPFhQoLCgk0FXTp5gsmKBKQHGR0gKy5rcus0q8c/t+bCOlmO00LhPoPmdHZZ0D9dN64V2xHVt3H8C+Yxdx7luqr8/wRLWdofgGvsdHj1uhMlihMVix40vg5sdbsWiJB1ZbCP7tZ3Abl3FkgO5LrwuVZPAmLp4+gg92vohnVqwA1YuoGxqMS75/0VTAMk137NiBw4cPY8uWLeJFkVTHdBOm9bLJ6TUbXsGeA8dx/PQFXL4iKbm7EvtL5D8Kat6XP04cj1+uXDmadHzKcxvneHsBhJePJPLzvj66CNy9iP0rV4qXM7fPYXfS+cv7lM9veHrFCnEeBHPyudEyfomw697rE3mGr/+AYVU+N8rDx5PPTU7Hy2sDA9fNz7fGj8n5+UXSZRyl/Enz8f2tWIH9FwFcFJCfOF+5fKXyeltODz3/NW/TS4ZBXP5YwK+8/W6K0Ll5Gq9L5yeXkbw++fg0v4Fg4ubpYe1dqp+k+kz2hStJvkD7YGVYqqsVGwlIbuLUdul81wqlmCJLzp04iD2vbmBYfdD5rPj4MqunVO+ch8vuCpejvA2Vr1x2tEz2B7Fe1N/Fj4aWF1+ndI4MULfOYad0DLH9Vo6AoXocuj+xfxmKE8fbzb8l5POU/971K3jZ7XO7pf0NzU/74bqRoHjFLil8+iG2r2tfUoQURUyI6SmKCLj2GYObfOxE+0r48ZDrTy5/mr97DRek/fG+KQKLXu6c2ID7t79fXz/cXug4dO7x8z+Fy9JxxPmtwYZX9+DgJ8dx+vwVXBbhTOJ+wed5EfuH1K+4vnh9DCnvofev5LqU65uW0bYrVkj7ufSZODf5/DjC4YL04pLuP/fWt/DP4eWRSg/xt3vac6p8UuWT/Hue8odf6w8PBYoHVgyAbWAAPX0D8IUHYHf3Y6mjH3PMy9HYQp8fWo6KhTT4VC/08/ugmeHDxGkx5E3uwbj6HhRN7cXExl7kT+lhGM6q7UF6VQ9GVfSgomkN/vCkgOL/4zcqjC6ZidK6VlD/4eHfLSaVeJ4liu9v//jLgViC47s//IgbH3+OA5VmfDhxAUNxxbJVSGMAjiCnOobxdVEUTaY+w8JoPqdaVorDGKl/FW9dAb7Z+ypGzt+Hjwfv4PCzERjmSFC8qRfzzL34/V7CAODOtcu4eP4sjuw7hvPUFfZQP5/ggwAAACAASURBVGK9AzgkQXFrZx/+SoM5XfoETYt7YJgTgc21HIHwE3j7Iqmwr2OMbigUkwos1GJSgwPQzhDbbKf+r7fOoK+c1hMU+xmiaT5j6zfcN+3qN5fxzcXLOH/+Ms6cu4TPT3+DA1s2Ik33O/TsF6MTf/fVNzh55AReefGUUIr/7Efalm+AW+fwuN6HcRV+TKgNQNv4B+z+Grh+dAfSNQko/l21H8W1fkys8+PlL4Drh7Yj2+BDaP/3wK3rfEw67uenJDv9DV5fJ6A4Q+NhMCY4ztbvF/1xt4v+wfkVHhRUelDz7Dn8gG/xmtODgoXbsJdUi8Hv8eXpb3Bo7yE8t+9KHIqzCIybXsZze89yP2YedfnK5xiYR1C8j0e3/u7Li/j81EWcOnURZ899g6+/vozLXx7B1hUSFF89hjXyQylP5R/4AQys3Q0eAPcuHf8iDn1wAH+l418/iQEaAEv7IShy+fTr3dDUubCNpPUv90M/2Q0C4XIJhlU13Sir6uY+xAWkDGuE5amlcGkJiKn/MAExGYdNK+0I7PkOuHoGa2LPIrD8OcR+9zY+uQFcePevCAV+hxKjAxMpXFoXQeuqd/Ha6x/hpWeex2zVLhwfvIUPf29DgepP2PkVcP3Ay/HwaYZi22ERPv07GYopjNqKCSoLJmpkKH6fgZiguO/DW1If469w+lTCzpz+Gu8+60BN6xYcpn7Ig9/j63MXcfyjA3j5ANXXZ/hLpQ2GJz7DDdzCgb9YoS23wlhlwxsExZ+8hqYWHzo7wwjvEFBMgMUPVmu34oNPznFf+Ts04N2dyzj4Qq8ExZRHvNCTH8J+/XQNdkujHN++Rr5xCscPnBNKcRJcJvY79KGSlt8LYxLk8fYJ4EvsYwVWMOgJkEx+aJfzDN/n8LT8UC7Dudgu+bgJUJL3KbYRsD4k/9vD8iadW2JbyrMT527JUJC0TXL+I8NAT6rHe88/afsVEpxLYJp8zDVbj+Py4CCundg5tJ1uP4WboDDOrdi6Tdj+C7eBq59h67bN/PIkeT80v5Mg+tpn2CD71pD2TufzC3xhLZXxIC7sWYENFDp981Qcyvl4z+7EQXqxcv22GDH75jnsXpt8rUnzw6H4V5eddC4fJO1zxQqs+YTlbIYmvuarx4eW3Yo1rHAL8Bm6LV3DcFBKhp77r09AMZdBEiSJdAKK+QVAEqTJ6xNTuqbESydafq/vJPu5yH/76mVcvjTMvjzKL0B4e/lFBtf5L/Rj6u89fJ+XLuPCEXopdO95Jq5BLlPKQ2qznBbT5PpZQXX+U8dZuxvnKHxaynPh9HEcPJuoX97+RwGoieMPq98h5T30vO+t6+S6FOU0eH1YuXKZfIY98XYk/2YOvc7E+aSWp8oi5QMpH/jX94GHAsX9A/1Y3i9gzunvh9W5HEvsfZhpWo4pTfTJoV7o5/VCMYvAtwf5U2LIqXIity6KnPoYcmpjKJjSg0nT6Zu/PcitiyGjOoa0qh6MLI/BuGx1HIr/R2Y11j6/FR8eOs7fIh4OxaMmzcD2t/Yma8C/fP5vfwOB8ZWd+3Bklhtvr9mKqqZVyK4JI7+eRn6OoaQxCu2sGPRzyGggrSgmThEDbZ3cHsEIXQizdn4L3PoCq9+6Clz5DIHaMFQzZCjuwXzzKzj8HfDd8W3o9i2HJ0j2PvefvXSYyrE/rhQ7PH1wv/c1f+5pw5wopiyJweVfjtkte0BddE9u8WO0DMXbAxit87M6TApxOn0yaUoIjctiaLL14hmC6xunEdNSHh8rv2O0PpBl/ukUbuAq9v+pB3ULIsirDiC3IsB9f9M4zx58Mgh8/OJKXkbh0FldUvj0n71Ii53AN7iOXX1epGu9SNN6MXbuXhwbBL47tB3pGq+kBH+OfqMXk+p8KJ/5BwFaR7Yjr8KLLBmsNW6M1bg5VHqsxgWyDIJXLS13IYPCnrUu5BoSUDze6EZBhRcFFW6Mn38IX+J7fPgXDwpevEBxcXh+ngv0SaVsjRMLKHz5uqQUq7thmN2HTHU3r8tuehvHbpEC/zxytC/jg2vAF7ueQLa6G6U1HjRZwoj2LAf5PL0I4j7FFKrLirCkQG6UQmd3DqCfJJq7X+PFZS6M03YhR9WBxfTW5PpJrNB2YYLhQ5yhrzC94eLwaAHF+zhMWlvvgqbWidLKLkwydmKCnvoO02BaDjYCY1KGxysdGK+0MQTnq2UgFmlavpk+8/Sgv3N7MUFrg3fLGZzevwttWhsK1SI8uubJM/jh1mmsJlVYacVKCoW+/AlCskqstKLpjcvA4Bd4WkFKMPUZFopwkdKMEl27NPr0+yiiMGlSi1/5Crh1CmuVJkxUmVGsNGGS2gy1sR1V9R2of5XCuM/jNXMn6qbS4Fp2dLxzmaH4cWM79ymWodhYbUN1vR1vngduHnsNbSY/nK4IYm+cZaWYoJgV4bWrsXpgAMuX98PZvxn0TgLn3kUkIsBYqMb9DNCygixU5ISi/NPpfSIy4MCauAI9IPvALqFID93+SPyhXT6e/NAupwcGKA/BJ22fUOUS6weEannrHN6UIxbooV1SwCmf2OeR+DkNT8cfyo/Kqjldb+K49AOcfDyRHr4+cW7J+V/49Bq/JKNzk/chr9/Hiu9Bhix53evc6ZHU/QEMvEawehOntiUf/wURdnzmzXv2J/axOg5p8j75eJuP4DL1nzy3e8jxKA+rrA9qFxAvG+R9cX4qj8MEIBex7572LiBsYED2hdUJf4r7gnw9q3HwIo1Pt5+v6dqnL/A1yeWzeu3qRPqZfbhIfWslZf2e8zkqvUCQz0cuu9foRY98PKnszu6Ml138emTfOv16Un5xfhQ+zf7E9zChJCaOL72EOHxv/VKeDyi0/8rRpONJ0CP52hEpvDqxv4G4UiyWDc1Py+JgSj4Sh7T7Hf/Xtq834y9qks8nUX7Dji35tFDfD/K9hbaj/Ml+vELuArExuX0ln68El0PaX/J6qj9xbsPrn8tXrp+dUleLjXJ9DzueVH8H1ybWc/scsv3w9ibql9oNl8mQ8h5avqIujyT5zwBombztkUv0E/zBkPVyeSXKW7Sbyw/wp3vzJ/v3sOuV6ie5/lLbp8or5Q+J9p9qDw9uDw8Finv7liMcXY5u/3K0dfVhobUXM1p7UbekF5ULeqCeHUPR1BjG1ceQVRtFWmUEIzVdSK8MIbuWlkd5fdmMGEpnJPKmV1Ff3giMS5/AH558jQfa+u9jytHiXIk//fNmjNMtvEcpVjSYcOQYffn17/+7cfRznOj4PUNxXcsqFNSHUTotAtXMKEOwfnYU5XOjMMwRppn5Jn+S6eT2EEbqgxg570N8LH0N6otdT6OwPgzF9IOgn66PN8Ywu/XPDIM/nt2L3i4KhV6Flz++ysrxpUPLGbrk8OluXx+cv38PJ4lnj++CzxzDtKVr8IfDNKjVBayeR4D7EfcpPrktAcQExYV1ASy2xtDl6cVCcwz/RN/UvXEKMY2PoXi01gthPqTrnsYbNEDVF0fgmO9FptGLKU+ewneD3+P9P3mRptnGg4l9uWcjpk8NoH7B89jwGfU5/R7v/9mDNM06vEADKF27gBdWvwBb7A1s//J73KZBrA5sQ7rGg/R1VALf45MNL8A4fx3+svdb/ubt9cPbkK13I2PuLhy8Rfl3oF7tQrq6H/bXL+L24EVsbHUiQ+PEWI0TmVonsnUuZEtK7qntbow3uFBQ/keYl+/AVvqU0Pfn8EcC5b+IPsNvRJYjS9ONTNfbOEihwddPok/dhZmvXQR+uIiNri5kqbpgDB/El4PA56/1IkvdhSYC2B8u4rVoH2YtCSCw5m2cvQXcPrML/f39AoppiJhz7+Kp/n70r9mCYzyg0efYtnw5om+cxU18izfCPchROZDr3I1D0vFX6joxqfxD8KeJdzphnCIpxV/sg5qV4U5MNIqBtIr0HSjUEgyLPsIEuwTE4xS2uDEYqwmGaR0tb8d4RTvyVTYU6+yYZLBjot6OIs3LPNDW6ddouU30If6zGMDq0w1Po0ZlRU3wLVZsL+35K/IVFhRQf+BuUoXv4PzbGzCzzIKawH7uB3/9w80oKDOzFRL8qiwoVptRrLHgdx/d4u84u2eYUKgwoXjmdhy5RYrzFsxTmzFJHUBgx9f4YfBrvN7lQMOT9A3MK3hvZRRV9XaUh3bhYyqvGyfwuNEC7V9O8EBbHz1uYmCun+oQUHz8NZitAXh9UfTuFFB8hOrjLZq/jbO7n2AIXhrdgaM3aOCul2Cx+rF8+fK4UX3+fbYLQkDcgid4H5uwjxRHOu5b99un9FB5NLGOX65cOZJ0fAGf8vabCDIHL+PYlic4zxNbjjHs3Ty5PeGHt89iV9I1DN8np6+fwJY18nHvPY/+/qHHvbc8aP0gLu5/Kn6uP3du9+yD62QQ1069jze3bsGbB87j5iDdMC+D66x/EyhKefDKMelcn8CWT+io9OAun/u9U37Qv3kKW+QyeOZd0Va/lNqmvPwnpsPL7J5zX7NPDIQ1vL3T2XN9/kJfIFAZHOSxB45tSFzLcwS5g5dxZLNYRvV87S7tW9T7PedD+X+8iH3PyPv49WX3wmEq29u4ePhdbNm6Be9+dg23qT5kf1rzPs7/SJAjl+NTeJfGUhi8iH1xX5KPL6YvHLsG3L2GE28+hf5ntuMAhcvGy6ifoQlD/F0su31O3Ff7+0U53vxMblPSvVY+J7ld/yu1r37Zd5Lv8/KxZf/51X7cj6fePcsD/Yl2fb/2N7Qcqb63n6SRAhNtn/Zxk7+5Lbf34XU+7Dh8ntdwQrp39G/eh/PUVyt+Pds5FD7R3qh+RX3F62NIeQ89bxmKk31TQLGoyyfeP8/+dfZd6Z4h/0ZeOoLn5LLcKr0I23rv9SfvNzWfKp+UD6R84F/TBx4KFAej/ej0Loepqw8LLH2Y1tKHmkUEw6T+Uoi0pPxWEhBHMaYihhHqTowuDyGjOsrqsGJmD4zze1C1sAe1S/p4Pn+y+PSRZuFf8PhzbyBbs4D7Ef/fowwcOv1fH9HcF4rf2XuYB9O6e5c/DvyzdPy3O4P48dJV3Pz0LG5+egbn//kVHJ7uxNtP7UD1sj9DMT0KzcwIDHNjqF0UQ93iHkxeEkPVghiMc6PQzU5A8WhDEKMNIQwcpf6ZX2FDa5hBWjVDQPHRF6KY1hTDovViVF3cvYMf7wJ3vj7D4dPfMBTLSvFhdHj60NrRB/v6j/G1GPdIXM+tb7D997/DKFJ9ZaVYhmJdEOl6P6YsjsAX7EOsdznmm6JYQSB9nz8a/XmM1ot0zx7s/ZqCuil8lewOvnhvG7RaGinai7ZXLuAGFeldkef6yTOgcXP3rQ0gy+BF+twX8MyR6/ju1h3cvvU9Tu54Fa+cpRBvCYo1T+OZTxPn8MOZ/dj5hVhPynC6xg3tykP4XPrmsziH7/HJi0+yUpyhFmCcrXMj1+DGOIMIbx5+SdcvncP6nn7uT5xn+Gc8Lx/zB7qm6/iAwnElKM5Ur8Nz0vofaD0NDPbp+2hSdyFbTaC8Dk8evIIfaYXkToPXTmGX9DAoHqI/wxEanVV2t9sXceDF5ejpXQ6L43lsOCFdMx//O3x4kMKBT+IPlV1Q1+0Dj6210w3DFCe2Sn2KFVVdKDZ0oFDnQJGOQqVJAbbHoXgoDJNabEO+hoDZxgNr0XqGZKUNRVoHf2JJUe5gKJ6gFaNPn95mY1AuMThQrI2g9/WvQYMey3/Xj78DE/UTVlHfYQuPKm3e+MWwPG/DTNBcZgYBMYVIT9JaUaIlMLagsOs9nJbK9fh6E4pVZkz93X6cHlLHt3Dy1dWom9qB+mlrsfWkVF5U6IPf4SCpczdOYFV1OzQyFP/JhNoGByZPc2DneeDWp9vg6AghFOrBcn6AkwHrCez+nIYEorBp4bc/fHUUf5zZiukzOuH3xwTcLadPOJH1szKSfNOlt6o/l9504CJ/QgZ3RQHePHOWw6fp4ffe7Yc+VNL6ZBgT+SmPgGqR3oR956TrkPzs5pf7sEk6X95eesiVjyfvU073S6opFcXF/fTDKp9H8vUljitfc3x7foB9ArvPiAdmCvslAB0YePGB5/ag8nzq3RO4fGsQgz8O4va1c9j3YeJFBh9v8z6cpSFy6Y/vOTdx/sNNP10/G47g8l3qnyuuh69f7GHYf9k3xMNF8vXJZSZf+/3Of/XuUzx6r9zeb545xi9FBBT348V7fEGE0ssvOMTxBGgSGFIUQ+J4m3Dwa3qhwszM09tfH+R6lvMkn2//ml0iNJZezJ3cIsrnJ8ruftczMLAaWw6e50+WcX1cOoY3PyEFWkAXHe+JN47hsjgtUZY/XsaJNwSoDzkfuf2s2YVTcv3RuSWVEeVPBil5exmk5PSmowTrVP8XsU9+AXn7LKvX8Zdd/4rtS77P0/ncr31RWQ7147ND/Jjra/M+nJPLgf14ENc+3y29PLtf+7vf/eJFHPhKKnzaB700+Wxo/RDoDm0vg7h2chcfh9pnfHsq0Ls3cfYchU8n6rd/8wFclOuXPoV16QhOsNq7S9y/kqBYRJPIL4HufcEh168M1ORf5E/0iS25zeDWRRyQXvxQfvmF1nbZf2RYTqXv8/uRfL+4n7+k1sv3yvvf71LlkyqfxIsF+fdGLpOHAsU2dy+aHL2Ya+5Bw9IYDHOjmDQtyuHRY6siGFNBnzAK4zFjGGPKRXq0rhNjK0PIb4hAMTPKMNzY3MMqs83dBxpgqmR6DCONEeRO7sfaje9genMQ/9djOvzXRzUPtP+ZNxl1C13Y8c4+/qzEXQqJ/hn74bsbOLfmJeyvNGN/uQm7R9Tg4MIgtr7wDkpnLIdqBvUJjqBhcQwzW2KY3hTDlKUxVM2PQDcrAkVjGLnVIYzWBzBSH8AofQBjy4OYMDkM45wIahdGUb0ggrqFEUxbGkXtwgjK54ZhmPMHdP9hMwYG/oRlth7YXb1wB3oRCPeJ8OmvPkS7swcLLTG0dvQgEO5FZM3L6Ar8RVJ6fRit8WGkxodRWh9GabwYRfMaH3Iq/JjbGkGHuwetjhiq54VQWOtHhoG28d7HPBij8bKpbX9Fe+xp1Gk9DLsF1T7oZwQwbXEYnZ51eGHLq9j4VB/6+vr4nJraI6iYFUDBkhUw6AluSTmm/a3F9q+A7/a9gjS1gN50UoA1bqSxEuzCWLUwWi7Pj1U70eBaD7t/NXRqoQ5T+DQpxRmk9mqcGG90Ib+c1GInxhlcyDNSeDTNO5FncCK/nMzF8znabmQ2rUNHdB0mq7qQoepEBk9pXlhm01p0Rp/GgtldyFSRdSKLrQOl1U5Yep/Fpldfwcan+9C3vC+uLiYrjavWv4QtLz7F5dLT24cudxT6yU5WiLOXrUFXZA2maxwo1HdgUmUnNPXd0DY4oWtwQt/ghLaevk3chYnGDhTqHQzEpA4TEOepCXZtGEembEeuQhipwawIq9qRp2rHOKWV0+MUVoynTyepbSjW21BW4YCikqDYhiJNO09LDDZM1NLAWe0opqnaijxFP0zhtVg2w8IKcX6ZPDUn0mUBLPOuwbIZJuSXCSsoM6FQacYENSnFZkwkpVhNsNwWtwlKgmITJqopdNqEBa4nEQ7/EfOq21HVYGcoJjCm+Sr7avSueALNlVboK60wVrfDUGWFSm9CmaYNugoLJk/rwJTpHZgzrxutJh9c7jCisV4u/+R6ofm+J9bjr5teQo+rA5PKmjGxpAkKdSvmzOtCMBiTtrl/vQ7f14PTT2HT1i3Y9HRCfX5w3r8zz+r1eGnrS1i/+u/cPkkZ/7c8t1WrVw1tQ+9RDMJlHBl2ftymXl6PVcOWP+jcd525jdtffYinfmH+B+3n55evwvqXt+Cl9cOuI37cn/OFd1h9/fqjB2z/9CZs+RfW8y8vu1VYNcyf1nP4+1m8Eb8e4W9PvbiF73E/Xz7/Dv1z2LXc9xq4ff3yNvyr/Hjrpn+ZX7JP/PQ+uM4fdJxfsP2/bv2KNnFvm1mPT+iDEMfWD70n/JL6SuVJlVnKB1I+8BB94KFA8RJbD2a1xlC/NAb1rAgmTA0jty6MdAqTNoYxyhjGo4YQgzGl0yrCyKF+ktPCMMyLom5xFHNMPbC7+xCKEhTSYFQxTGqMYXRFCI/pw2gwr8O6TXvw15ffxTObdz/Qnnv5Hby4fS/2fHQCn39xGWfOX0myb5PmablInz73DT798FMceeV9HHnlPRx8eQ82b9yDGebVHPqsnB5C5fwIZjTHsMgaxVxTFFOXRlExTwBxYX0IGeUBjNYHMUoXYMusCDAs1y8KY3ZrlK1xWRSNyyKoXhCGflYQyulhTJwcgmF2GDOaogy+3d7Hse7Vj/DV98D1E6/C7qT+wDHYnD2wdvfAMItGlhYAPJqnHgmGKRRaLB9r8EI5NYCGhWE0LAxBNyOA2c1hLLVGoW0MIENP23hE6DRN1QSwHqTrqC+wB2N1XmQZPChp8KF6XhBNNlKce+AP9XL99Pb2goygmKY9vb3wBV/j0bCvn9gLd4sbabPXwrXlPG7Td5vXUig09Q12I03ljANxuoaWO+NG/YXT1d1xo1BpSo9VdzMMZ+kodLob2VoncvVOFFV6ME7fjVwyXXcciPOMThRUOEHTbAJitQS+tB9lpzBVp7ScALmTIThHI4CZgDhTScu6kK/vwoLmAELhHoZhvm7p2umhispANjnd09MLpzeKutke5CjtyFbZkU1hzgS3WgeKjR0Mv2LwrE6UVnZCVdPFNqm8A0V6O6vCRXr6tJIEwwqCXjGfq7CCTMAxQTABsIBgXlcmlhWo2zk8usRoR4mR4JhCp9sxQduOCRqxboLGCrJinRXFWuo7bEGBikBYgmCalgkjtTiP5yUQ5nU034YipQmleitKdCJ0muCYlOP8slbkl7aiSGHCJI0FJRrqQyxMabBAQ4NlVdtQM9mO+mkdqJnsQGUdfZqpHYbqdmjLLdAYzSivtTIUl2nboNKRSmxD48wuTJ/VgUVLXbDafKz69vT0oK9PBuNE3VC9BQJRLFrcDZ2hDcUlTSguXQaFuhnNLW7eZvnyRH6qU7k+h9dvKi3K6e8tn2c++hp37lzFiV3rsaqvD09ueg9nKXLg20/w/H3a0z9WeT+PV7e/indOXOXPhr3z7+B6t528Adw6jwPb6IXeKjy//TB/G/zOl+/E722p9nDv/eDZuB8/z3781IvvD/Hjv7d9/GP5e+Ke+qDyWLXrLG7/eB7vr77/7+l/tvJIXe+/7PclVX6p8iMfeND95uf846FAMamn9UsojDiMgoYwMqrDSKsMMwQ/og/jMX0IjxrCGG0MY2xlGNm1YRQ3OFG1IIwZLT0gqHYHekAwQQ+vnd5ehmTFjChv84gujIyaXpTO/SOqm59AVfOquFUPm69u/guqW55AXdsaTLaswxTrPycZpddhCi9PmjevxZS21WhoWiXZE1DP+T3y6ntQUB+CsjGM6vlRzGiOsmo7ty2KukURGOeEoWgMYVx1EGMMAoZH6gIgy6gIQtkYwqzmGFocMVZrF1iimNcWw5QlEehmhlDcEMLEhiD0s0KYujiCJdYYHO7X8enVO/j+4sd49Z960OGOocUeZdWX8mWVy6qwrAyTOuzFSFaJRR9hdWMQdlcMgVAPFpgimN0Shtvfw2Xb6oiiuM6HUWoBxgTHBMK5lV4opvgZhPUzA5i8MIROTwwxrhOxLTkTQyE/yIm6onRvn5j//Ysf4cwNKfyawt6uX8eBl9ZDo3YhTe0UQKwS88kwPHReQDGBcLKRQkyqMFmOvhvj9U7kGgQQ5+hoXTcDMoEwWT4DcRcyJCAW6rAExElgLC9nEFZ1MCBnqTokSO5E/VwvfIEY+nrFtSeuP/FjLxoZgZh4QeDxxzBtvhe5ageyCYoVduTQ94IJinUOlFR2QVUrVGFShksqqG9xJyaWd2CCwYECnV2Ylj6vJPURloCYVOHcMskIjhmAqe+wAOHcMgvGlVk5XRhXhO2YqJP6D+ttKCWFWCfDMAGyFUVqK4olQGYoVhIUCyMQZis1I6+UQFdSiEvFtEhFg2pZoCxvR6mOPslkRoGijYG4SEFKcStKtGaUas2YpBFKcanGBE2FFRW1NlQ32FE7xYHaKXbuR1xRK1RhgmGNwQx9hQUVNVZojBaUqlthqLKgodGOxlmdmLugGy1tXnR1hxAKx5L8k+pH1Inst9FoD9zuMGbN6USpohkTJi1FSVkT5s7rlLYTMC3fNFPT4T7+sNJP4p1TV0U4JYfH3sHtyyfwjhR98o9d7u/h/K07uHPrMj7ZsWoIdP6bXfcTr3J4LY/MzvVxGze+OIBXn3hY9f2Pup//zH78j1qnqev6N7sPJb0gTJ1Dyg//rXzgoUBx7eIoVDMjyKsP8eBZo8oJhIN4VBfCo3qh9I4yhpBbE8L4uhCP4Fw934O5bRE43D2IxAiICbx6EI31oN3Vg8bmKMqmRzDGGOL9PKILQjba52+T0o/oQrzuUR0dM4jH6LhaMU/TeFobwKPaAEZQHm0gbvdN6wLIrgpi4pQg9LPDqF8cwSxJ8aUQaP3sEFTTgyiZEkRuVRBj9H6M1PkxQuvn8OmsCj+vn9UUgakrhvbuHpg7o1hmE6oxKbBl0wJsFXNCmLY0hCXtUYZgb7AHshHMWrtiqFsYRHG9HxlGL9I5BNqDkRqPgFutFyMJctUeHiCryRZFKNqDQCiGJZYIA3Kspwc9vT1o64hgYj1BsTtuWUYP6hcE4A/GQEBHIaiUv1fahsGX6ydZIRb1RXlofShCxwqjoIpCp90Yo3JhjNrJlqYR86QWk1KczgowKcbdbHGFWNWFdJWA4XSepzSBrZNV3WxdF7K0XQzA2TSSs45U4E6eUuh0rr4L2RqxPJfyqiUVmMG4AxlKYQTANJ+p6gABMOeT1vFyNa1zQFXnRJc7wko4l8EwhVxeRo2X/DcWo/KLYvYSH8ap7cimwa7U/0qJQQAAIABJREFUDuSqO4RKrOlAkcHBKvGkig5WjKnf8AQ2h6QQ23hArSKdHYVaG/LU7fG+wgKICX5tDMMCiClNkEwwLAGxUoRSF7ECTKHRFDJtxQQCYZ0Vk/S0zIpClYVDpmkdzRMY8zKVmOaVmZCnMIGnZSaMLzUhTwLh/FKC3jYUKk0oM1ihr7ZBV21Dqd6CIhXlEwpxYVkbJqhMKNWZoDBYUKYVKrHKaIGxhsKmbaif5kDdVAeqJIW4vKYdxmr63BKpwxZWiHUVZih1Jqh0baifYseURhtmzu7EkqUutNv88HgjIOglH6f6EHUj+WhvL6LRGFskEoPdHkBVjQUTSgiKl2HBwm5x/5F8ObF9sr8nXgKl1svlmyof8rOUP6T8Qf4tSPlDqj2k7gep+0HqfvAf93npoUCxemaEYTeNQ50FjBLAMqAaghhpDIHWFVEf23kRTFkawUJLEJ4AKZEExDF+KCWwCEZ6YHH28GBUxVMjGGkQwPtbbRBkMhjfM02CYALfn7NkKB4+/6jWh8d0AeTVCbW3cl4Is1vpnKOY0xrB5CVhAcUzgpjQEEQmffeXoFjrwwgNDXzlR1ZlAKrGIGY2hWHqjMHhirLyau2OYZElgunLSH0Ocl/fugUhzG6JMPwSBPtDBLQ98Ad74PKR0hzlMObi+gBD8RgdqcMibJpgWAbikWo3Cmq8sDmj8AZiWGAKYVl7hOGYXjhQ+Ta1h1FYTQDtxiiNAGMKpyYophcSlIdNyi+nRSMXAJxo8AI6aLsWWwQTaz0Yo3ZJQJyA4jEUMp1s6i4OmZahmKYEw+m8nNZJYKzuwlgCWskyNV3I0hAUi2VZmk4eIZrCnscbBDgTJBMwU74MCYpJDR6r7MBYVoiT4FjdgRwtAbJDQLLSgUzJCgxdaLOFEIkOv+ZEY0+UQy+XmdcfxazFXlaEs5Q2ZFP/Xwp/1hAY25GvIyP47UCBjvoWO1CgJVXYhsK42VHISrENBRoaLIv6CcvqsFCGhVJMMGzhPsS5pRaQERTn0aBY6nYUUYi0TjIpXJrgl+FYCpcWEEzLBBxT6DSpwwTBBZJSnAzE48vaML60DXkMxCZQP2LqN6wytsNYY4eqoh2FSqEMU8g09ScuVLSiVGdGmd7CKjLBsFJvhq7CykBMYdM1k20MxdUNNlTUtLMqbKi2Qmu0QG0wMSCTYkz9iatq2jF5mp3Dphcs7kabyQuni1TiKNdBcp3I87FYDwiGY1FxnwkGo1ja5ILW2IpSZROaW938Qo7aiLxNanp/P0+VS6pcUj6Q8oGUD6R8IOUDKR/4R/SBhwLFFGKcXhHASKMErQyoAYwwhHgwrYnTwtDPCaOxKcb9ce2uKMMGAVeMgDgWY5WNANntj6HZHkXDkggmTg1jlCGA32rI/HiEoFgbiBtBN63jZRoC4SAeobTGj0e1fgZbmlJaLKP1fjwqGcGwPE9TOf0Yr/chpzqIkqkB1C0IY6E5wudF4ch1CwUUq2cEUVjnx1ijH6MZiv0YQQNd6bzIMPqgbAxi+rIQWuxh2J2kAvcwHJNiTCp5w6IwquYFML0phCZbGN1eEfIcjsRAFgj3oNsTQ5M1groFAZRMCbASTANlMQhzyLQMxW6k6TzQNPpg7Y5gdnMQVXMDPE8KJr14oId+c0cYJQ1ejFK5MJqgWOVCpsGFxiUBEDz0xKJcJ6wU9wr1k7aneqIGIEOy/CKD1tm7I9A1ermv8BhVN8YonWAQVjsxWtWNtLgRHFNaUoQ1lBbzrAyTciwrxBIQExSPVQk4zuQQ6i6G3rHKLmRqqF9wB0gVJoWYgDhDVn6lUGgBwwTEDl4n0gTBDmSpCZDFfIaCgLgDGQrqA+zAzMU+rgP2T+lFwdDrFwDFPhzrgS8QwdylPg6PzqKQaaWNQThHRcDrwHiNDROMHcjX2jFeLcEyfV5JTevtrBJP4MG1aIRpGlBLwHBOqeg/nEPgq7CCpjllwmhQrdwyM3JKzcgtNXO/4kJJES7WCTCmUGgyUoAnaK1sRWoLJlD/XkM7h1HTp5MoTaNM5ytMyFcSFJt4uwKVgGSGYVaLJSguFQowqb8V9XZUNtgZegmG80vaUFDaChpYq4ih2AQ1wTCFP+tM0FVaUVnXLsKlG2wMxwTEZFX1NlTU0uBaZtGfuNzM82q9iacNU22YMbsDcxd0YVmzGza7H15vmFVgUT/C12V/J/+kthQKCWim5bTM5w9L/Ytb4XD4Jf9OeikkqcYJf0/Ut7zsQf6QWi/KMVU+yffLlP+k/CHlD/K9kaYpf0j5Q8ofEs8cqfbw76M9PBQoTjMG8JiOoFXA6Ah9AKONAYytDKKwIYiqBSHMbCbFNAJ/MIpoNCogWIJhSkdjUYQjUbh9USy1RVC7KIQJk4P83V8ZdAloGY4lyJXTtIyMwJaWPaLxSZaYf1TjgzCxLJGWl3vxqNqLRzViOkLrRXaVD5OmBFE+O4TGpSHMbQmjYXEIullBaGcGoZkRwMTJfmRX+jBG78VInQ8jtV6k6b0YX+3j9TOXhdBsC8PWFUGnW1h7VwSLLbTPIGYuC6K5PYwOFw1mJcogEqWXBlFWv1zeCEP15EVBKKb6QKHOssqbrnOjsMaDLKMbozUuFNd6sNQSgj8YwfSlQSxsC8EXoLKWLBqFtTMMxWQvRqqcccs2uDCvJcj1QvXA9SGdg1w3os5EPcnraeryhlE3z8ffDyYAHq3qitsYVReSLU3ZhTHKLqSpOhl+CYBpfgynO5Gm7ES6KmFjeTCsTmQz/HYindRelbB0eUogS/Px0GiRprzpSgfGKhwYq7RjrGKoZSjtyFDa4us4rbAjU2GHbnI3vH4Kx6XrjUjlJ724kX02RmHmoqyovOc3ebkfcI7ChixFO7JI3VXbkK1sZ/Adr26HgF4bK8P5mnbka8U8KcIFWvqkUjvyKVxa3Y7xKmscfrOTQDiHIJhNguNSgmITL8tVmEBQXMjKrwX5MhDTKNAqM4poRGgtqcIiXaAyidGiqQ+w0sRAXKhqE2BMcKwQADy+pBXjS1s5JDqP5ktaQdMCRSsUegu0lTTAFinHLcgraWbLL21BkbIFhYoWlGja2CZpWkEDZdGgWdWTKXS6neG4WpoaqylU2gyNsQ0qgwmGagt0FW1Q6lqh0reiZrJVhE3P6WCgNVs8cDr9CAYJiofeU/hFTiyKSCSKYDCCUEiqT6pTybf9gQg6u/wIh0UdU32K7YbWdWpZqjxSPpDygZQPpHwg5QMpH0j5wD+2DzwUKBZ9dkmp9WOk3o+cmhAU00OoWRBmuKW+uKQOh8PigZTeDiU7FqXpQTUQisLpi3K4cM2iMPLqqT8wwbZQiwXwijTNyxBModLJsEzpeF7Kx4qwD4+qhclK8JC0tI6gmPoFjzH4kW70o7DBzxBMoc71C4OomR+EkfsTB1A2NYDCOh8yK3wYTSHNNHqzwYvsSj9DsarRj2lLgmixR9DeHUanS0Cxwx2FqSOCRWYaEVqoyPQyIBgmGBaqLJWHeElAUBwBQbG60Yfscg/GaGm0aDcm1Hmhm+5DUa0PY7QupOs8KKz1omZ+AOWz/bB0hnkfVNaiT6WAYuVUXxyICY6zjW5MXxKELxhBS3sIs5qCmLrIj7nNAXS6w3xODIASBPL+YlFWRxsX+ZBjoJBpAmJhNKgWK8ZxKBawTFBMEEyh0gKMCY47OWya00oBxqwOqztBUEzwm8Gh0B2sBFP/XwqPJpWYYJiUYlKIuX+wRgA05SFlOF1hR7pCVoiFIkyAnKEiWE5AMqnFlM5S2TGhvBNddM2RKGLSy5r7+iuVRSyKQDCCRS0+FBsdDMOZinZWiQmKc0jtJaXYYEexwYEJRhEqna8hABZWoGkX4dIaGlCL8guIzlVakaMQ4JsrTYVCbBaKsaQQZ5eaMU5By8gsGK+wiBGmJTgmGI4PnKU0o1AC44laCybqJNMSJIuQ6bwyAcUEwWwEwBQ2TVMJjGkQLQLiCao2TNJSf2EzCpUEygTFLRw2TQrxJIZh0Z94opoG22rj/sE0qBYBMcFwzeR21E21s0pcXkMjUQsw1lWYQKY2tEGlb4Oh0ozaye1onNGB+QudaDG54ej0w+sLxX1bvqck1xe1IbrvUH1Sm6J2IOfjdCSKnph8PxJQnLw95U2l5fIRZZcqj1R5yG0o1T5S94fU/SB1P0jdDxLPFan28B+7PTwUKCaFmOA1vTyA/PoAKKy4dmGIw4IXWUNweiMIhwk0BGyIh1P5IZUUnAhC4Qi8pNy4I1hkoW/4hpBVRaHRfvxWLZRgWRFOnsbhV+3DIz9pXjxCSrA6MWVlOL7Mg0fUHjwq2RiDD5nlXlaC1dP9MM4OoGZBAIbZAVBaMc3P6ybUkyrsBX0GaYzOgzSDBzmVHpRN9cEwy4+pS4JYag3B1EHwG0aHO8zXSPNtjjBaHZQOg1QrKoNIJGGUJiW2xU6jUwcYinMrSRV2Y7TahYJqN/QzvChtIEh2YpTKidE0yrPWhZxyF2YuC8AjKZ7yfh3OMHTThVJM+QmKCWKLql0oa3Aj2+BEhs6JsdpuZOi6UVzrwiJTAL4A1d3Qc1vQGkBBlUsow8oujL7HOjFa2YkxSSbDLy0jZXiMsoMtTdmBZEtWhWX4zdKIcOexKgG6pASzMeBKYdAEvKwMJ8CX4FiGYKEQU9oWtwyFDbR8nMaOhS0+Vg4j0QjIyDeFyf4qprQuEAxjYbMXE8spXNqKLIUVmWU0tfA8hUbnSYrwOJUVeZIiXMT9hwl+rciTjJVhgl+lsOwyC7LLzCDoFWaKT0kpzi6ltIlV4nFlZowrMwlTmFBA3wrWUri0MOozTCoxGYVGF6pMKNaaQaNG83eF1WbkKdpAfYZlACYIHlfSMsTySlt4ROliNQ2uZYbCQH2KBSCTMpw3SVaJmzFB1YJSbRsmqVtRrGgBQbG63ByH4erJVlTWW1HdQNbO8+U1Fg6R1pWbYKgysWJcqm6BUteCylozpky3YfosO5Ysc8La7kG3M4BAIMR+OfSeItVRJMoKcSg01HcFIEvKcUSuX3k6tJ7vt9/UslQZpXwg5QMpH0j5QMoHUj6Q8oF/LB94KFA8Su9HXm0A6plBVM0PYerSEOabwqwOU2hpWIIpVhsjpIaKB1J6u0QORbAVCBEUR2Htor7HIUyaFkRauVCAGYoJeAmQpRDp33I6AJryvKwcJ4ExhUgTCMuWnCYgfkxaH4djDeUlMPZihMaL0XovCmr9KJvqh25mABVzg1BO82PSZD8mNPhQ3OBHUb0P46q83NeX+hFnGr2s1pJKXD47wCHSAopD3PeW1OIOlwBkS1eEw5md3jBfvwibFjBGZSIrxW1xKA4gp5z6ELsxkga00rmRU+7GuHI3RmtdrB6PVLm4n/AotQuGmX4+FkMAPfzHonB6wqiZ48NoCaBlkB6rpVGhadtujFJ2YZRSqL5j1N3Iq3Rh+hI/Wu0hLLME0d4ZQpM1CNUUD9LUCRhmtTgJjDl0ejgQU7i0BMTUj5jAWMBwJw+MRcpwmqqDFeKxagqZJsgVfYIpdFqAsAiXJpU4TWmX8jjA0KySQZkguUNSiwUUi/7DMiALEBZwbOe+xIVGB2ydwYS/kp/GoUmoiPINMBiKYJnZj7LqDg6RppBpCpUmMCbLUQoIprBpmicoJnWYlOACrTAKlyYY5r7B1D+Y5y2sEGeREqywJEExhU1TWsCwAGIROj1OYYIwM8YrTJhA3wrWiz7EQikWCjGpxBwyrTFjgsaMQimkmiCZYJn7E0twLA+oJYPxeFKB6XvDyjbuG2yobYeu2ooJKlKIJRhmpbgZBWXCaJCtYmULisqaodCbUFFH6jB9k9iG2ini28S1k6kfsRUExKQU6yuFQqyrFCqxQtfKoFw72YoZsx1YsKgLJrMHnZ1++H1hfoEh14l8P5HT1H4otJrCo6k9yfcaef3w/FHpfvSg9cPzp9Li/p0qL/FQkPKHlD/IbYGmKX9I+UPKHxLAlGoPqfbwH6E9PBQozq32QzszgPpFQTFoVDupw2EEQxS+O1ylkdVGeXkYoXCYlUgCtmZbCJXzgsipprBnAbwCfL0JAFb78Bt1Is1QzDBMUEsgLFTf+05VQg1+ROXB/exRXu7Goyo3HlO7MdbgQW6VF8UNPiim+TBpsg/F9T6QQkxWWOdFbpUHGUYPssrdyK30oKiOlGIvtNN9aFjox2JLAC32IKydIXS4Quhwhlg1NneE0GYPwt4dgtsvlVeYHuJFmZAS2e0Jo6k9iIYFfiimepBb4WJ1d4TKiREEsGonW3If4XSt+BSSYboHDmeI90f7pLrw+UOYudSPTL0TI5UCgKkfMMEtwfBItm6eymmC3fxKJybVunhaXONEUZUTGVqxzSilUITFVIZkWiZU4DEKaap0CGWY0ooOpCkcSFMKk1VfOS1gl4A3yRR23oaWsRqstDH0kspLn1DK0YjRozlsusyO9DJbwpKUYQLh9LL2uFI8lubL2pGptEFd14luVzBeZiK6gXyW/FUY+XWT2QdVrQDiTAmECYazJctVEQwLIKZlBMf5MgxrCZATKjFBMcHxeBUBsVCGs0pNIEuGYFkZ5mlJG3JKTMgpaWMQHqdoQ26piecJfgvVEugS7CrbkKekfG3IV7WhSG1CESm8NFK0ij6r1IY8hRQuXdaK8WWSQjypBeMkG09KcGkLJqhboam0oLyuHdpKC4qUrRg/qRnjJzVh/MQmVovzS5uRV9qMgtImFCmaGYxVRhOq660MwJV10rRWguFaCwzVQhmmcGmNsRVqQyuU+lZojW2oqrNgaqMNc+Z1YFmzE+02D5wuP4IB4dvJEQzyPN1TSEUOBpP8X2pbcp7UVL4Xp6YpX0j5QMoHUj6Q8oGUD6R84D+3DzwUKDbOCWLKkgAWmIKwdgXhDRDg0WdSEg+tDGVJD6VhXkfrQxyG2u0JwdIZxJwWCr8OIM1IQEzg6+VBs36r9iTSai9+o/KwJdZ78VuVh436Bcvzv1W5efAsBmQZeGmfKnfcKGT6EZUrbpxWu/CopMim6QX0Zld4uE9vVjkpwwKCM41uBuJ0vRtk2eVuVooJYNWNHpTP9mJOcwCLzQGYO4IMxF3uEDpdIdi6gzA7BCx3uoLw+sVDPD/QB0OcdjiDWGoJoG6+F6ppbuRXubj/8AhlN2QjtXdEHGi7kEbfAdZ2Qz/dA1tXQEA21UcoBI8vhDlNfmTruzFS0YWRik4R/qwWYDuSAbcLoxSdcSNopsGwaBAsVnkpLYMwbU/KsqIDoyXjdTzv4GUiTfMOjGEQJih2CCimfr+kDktwnEbgS/MKm2R2ULh0cppBWQp5TlfakKkiKLZxn2A5LJqgN00yhmAFQXB7HIZpvTArxjI8W5FeZuV+wK1WGrxJ+KYI+6eBmkT5BYNBtFh8UNc5WAHOLBWh0iJk2oJsDn8mOLawUTj0ODVBsQWFOgsKCYjVpA5bkKs0sxEQU59gBmIKjS4zMxAzGJe0IbuM4LgN2TRf2sYh0zTPUFzWKoVOt4HAmJRiWREmGKbBsmgQLQ6NVrSBBtISwCyvE3CcR/spacY4AtqyVuROamajZXkMyQJ0iymc2UifWGrDBGULGIAJiic2YdzEZTylZfmlTSgoa0Jh2TIo9G2oqLOgut6CyjoLg3FVPU1JHTajvJZGmqa+wwKGGYr1LVDpW1BRY0bdFCumz7Rh0RJSid08OJbXG4gDL/2IyfcXqieKTAkEQwgEgvE8dJ+R6zA5v1y/8vaptHhZmSoP8XIy5Q8pf0jdLxL311R7SLWHVHtItYd/5OeDhwLF05cFsKw9iG5PEP6gDMMhhCWQEACceMhIFKh4UCUYtHWHGKqr5wVQUOfDCB2BrWQSHMuQLE8JjBPw++B5ERJ9P4WYwFhWjN14RO2KGwEx2WP02SKtG6N1wkZqXBipJRPLKYQ5TUehzC4OYU7Xu5BT4UZBLanFBMZeNCzwYU6zH8vaA7B2SmDsCoHgmFRjgmOCX3oxQGXhD9Co0SG4vEG0dwWx0ORH3Twv1NM8HMo8RkNKbjdG0ueP1NK8soshmZVeglZ1NybWubDU4uf9CSAOYl6zDxNqSEkmVbiToZjAOK72SjBMKvDEGidyDN0MxATJo2VQVgpgpv7CZATEMhSPUggQlgFZTGUgJnAmtViCY0kpJkCmZQTDPM/ATKowGcGxmCcAFkBM6jGpxGL06Cy1HVlq0UdYwLCsEAswpmXJsJwMxOmlBMcCiMeWWnnU6GaLLwmKxYsdKj8CZVO7D/rJHchRWZFRakFmmQVZZWLK8xIMc/gzg6+FgXicygIyUoPJBBTTZ5bEIFkyEMsKcXxa0oYsBmITskvIkuFYADIpxLnUH5iVX9FnmMKhKU1gLCvCpAqTESiPL2vlKc+XtoCU4HHylOC4REDx+JJm5JcRGIuQaBpReqKGPsdE2zSxQjxuEgGxDMXLkFcigJiguFjZDG2FCTVT2hmGK2oFFNOUQLmilr5ZLAGxsU2CY9GPWFfehjoecbqdVeKmFidsDi9cLj9DL/UVTtxLpAgL6Z4jVOJgHIRlIB6eP5W+twxTZZIqk5QPpHwg5QMpH0j5QMoH/rP5wEOBYlJA3d4gq8OkpAVDwkgpDgUTD6ZUuPLDKeWhNOV3eoJotVP/2wDKpnmRbiRVWNhvVO4hqrBQiGmZWyjHNJVMVoOHpgXcMuCyIpykEEtqMYVKk1JMeWTF+FGVAF8GYI0bBMMjNBRW7cJjsqndGEH51C6MIJPyESBnGFworPWgpMENw0wPZiz1YaEpgBZbAO2dAXS6AnB6Quh2B9HhCsDhDKDDGUSXOwiXJwi3R8y32f2Y3eRD9RwPVFM9GFchVGDq+0uqsYBlN3KMXRil6sIIBVknhz9n6rtRPdsDR3cAHl8Q81t8KKruZpAdSSBL6rKiAyMUHTwdpSJI7kCmthMzFntgojpZ5EFpXTdyDGKkaAHADhD8knpM0EvzDMPKxPzw9OgyO4dMj1bQVJiA4OQ0zdvYGIaVdowpsyGNlFwaLEtll+Yp3S4NqGVDhsqGDFaCbUgrbWcTIJxQgwl8WTEutSKdzSKFTLdjQrkdOep2ZLBSbEW7w8cqowBh4deBYBBWhx/l0zqQq7Igo9SMTIUZmWUmZJaZpTTBsRlZFAJNsEufVZIU4CxSe2m+zIx8GthKbUGeiqDYhBwyWl9GIdNtyCwRRvNZJa1xI5VYTmeXtCKHVGP6RBKHTrciTwJhgmAOmaZPKila2fIVrawS05TCo8n4k0s0z0DcLKYSDOcS6BIQlzajUCnCpkkdLlaJTy1Rv2GCX1KHyXInLsX4SaQUL0VeyVIOnS5WNUNlbIOxxoSqBgt4cK06M6vD1HeYlusr21CmbUaJuplDpjVGAuJmtvJqE+omWzB9lg0LF3fCYiWV2AuhEov7h3w/oam4nwiFmFRisY6mYj75/iPnH759Kp1Q1FPllfi9SvlL4uWg3EZS/pHyD9kXUu0j1T5S94PU/eAf4X7wUKDY55cgmB8+JSgmOGYglh9OEw9bouDEcoINgsFFZhrd2YeCOi9G6ERotBwinQzCApJlUJaV3gQYx4FYBmUKn5ZCoQl+BQDLYCxCpgUwC7AlCH5MBlytW8CuBL0ExI+onHhUskeUYv4xStO8lKb+vjTQVZbRhfxqN5RTPahb4MVcUostfrTZAxzWTOowhU2T0TzBMhlBLEEzTZusfjQu9qJilgeKyW7kljsxRtvF4dFTF3lh6wxgTrMPueWkGHcyEAswFkqwdpobZocfi00+TKpzYrRKziNAOBmKaZ6MBriqm+vGwlYfFrX5eERmAmOCYFovVOEECI9RiT7DpALLgDx8KodOMxCzIpyAYRmSGYjLbAKECY6leVKE77V2HjGagZhVYylcOgmKWTUutSKNIJjUYFKFk6A4U9kORY0Ds5e4kKejfsVWFOht6HZTaK7wX3lq6/ChanonjxhNQBy3MgLjpDTBMsEth0FTCLUAZIJiMgLgPIZiM8YpKWRawDADbxIQZ5a0QrYsUotLWkEgzFbaimwyCYzHlVF/YgqjJuClEGoCZKEGi2X0WSXZBBBTSDTBsKwQU9i0rA4TEOdQSHRJM39rmL4vrCo3QVtpQomOvk8sAfEkGYjFdPykpcibtBT5pctQVNaEUm0L9xWm8Oiqehp52oyKOjN4EC3uN9wCha4Fk1RNKFE3MQhTyLRC2wyC46o6EyY3WjFnvgPNrU7YHR643X7phYV8X5Gn4v5C9SXCpmm5bMPvPal08o9Xaj7lDykfSPlAygdSPpDygZQP/Gf3gYcCxVSIAh4ETCTSMhjL6+9NE1CT0jyzyQf9LC9yqjx4TENQLNTg3ygF8CbSBKYSBCtdrBJT+jc0L9kjalovpZMUYAG1pAgLuJUBl0KjM4wutkyjC5nlbg6HJsV3hNrJYEwgTBD8iLKboZimsjEkU1ohllH6MRrESuNEptGJkskeVM3xYNYyL8Npi41GcvYzrFo7/LB3Ewz70Wrzo8nqA6nDtNzS4ccSsw8Ev8aZbpRNdiGLwpk1XRhX3o1mq4/za6a5MEZNsCupvgzHHRit6mC4JRDXN7qQLuUZqaR8DowoE5acHlnm4AGyMnUdyK/oREldNypnuTDe2IHRSgdG0XoC4zI7WPVVOpCt7+T1mVpaR8sdIGV4lMLO+SgkmpViVotFn2EC4NEyAEuKsAzBpBKPKWuPG4VKc/9gVofbuf+w6ENM/YipTzGtF/ArAJjUYkpb2AQci/m0EgsoTHpsmQXFFTYsbnHB6vAhn6C41AJFjR3kk4EAfe5HWKfTj7qZHcjT0LYmDpseW2qW5iWluMSEDDIaIEthZgDmvsCkDisJjgUQ55Kay+HTFDYtA7EpDsAEwkItJihuYaOQ6axJpBi3sOWUtjIEZ5e0IJf6+1IfYDn8mRRglVCIx9E6Al/qL0zQW9osFGIbl18nAAAcBUlEQVQlhU7T55UIisXy3JImEAznUhg0q8VNHB5N4dLUh9hQa4axzoxSHW3XhHESEI+buFSaXyqpxMu4L3GxqglqYytDMYVJk1JcXivUYeo/rNQ3o1TTjInKZZiobGIwLlU3gYzA2FjVitrJZjTObMeipZ0wWZzo7PTA66X+3uI+Ik+T7zd+P9WZqL/7radlyflT6VR5pPzhwb/PqfaRah+p9pFqH/Jvaep+kLof/KPfDx4KFMvw8Ounfri9fiyz0ijNXpRO9SDNSKDrYsgl0P1fSicbzSfMKc074yAs1tFyWiYD7NCprPA+pnbiMXU3A+8YvZO/6Ztf40JBDYU8i2luJUGyE2N03XhM3h9DbxceUUjGYNyFR5RJy6R1jyq78Bh9nkhDYc5OGGe40bjYg4VtXobZ5nYfT5eavVjy/7d3nsttY1sWfpVxtoKVsyWSAJhEKgdLlm3JtixZzFFiQGCQ+zluzXvdeZY1tfYBSNrXHVyjH9P3HletOgGgWr25AfHj2ueg1gfbm0ofV6U+vpb7yNd4Xl/OP73uIvXew8ahi7ktB6t7ClSz1T6+lPhYpDZeWjZeWGyHIijvXXg4uuSu0dwVuoUxgeGmADGhdSqhXF4FyJwfasxqgpqMKwAOxqptYNxqYCbRxMp2G6H9NlZ3uIFWA+OmOsbjFOF3qDtMWD/XEIRvMWndYsK8xSRlKRF8p6NKs/FbzMaU6AITgCd9BTD8yqxBqY5XRu07LSdvcfbFgdfpo1DtDkqiDz7YuL//hm/flByvj7NLB6ubdUwbVUwZFUwZVelPGxWMaoZlz4RiKZWuCAgLDFtVEIYXYxVxhhc5JhCzHNr0HeFIGbOiEmYjSnOREgLNR0oQGapdMEsQKCb0RpX7y5LolSjl7yBtlbBoFrFIyDUIuUUFwzF13hrPtdSxRSOPhfBQi5E8Vsw8XkeLiLIE+qiGrcMqwokiliI5KZdeDGchCmWxFM6CTjFd4tfRPKxUEen9CrYPKtg9qorSe2Ukd8pIsEQ6rYB4w8ohZOUQjrLNIhLLIbXD5xJX8easjovLBnIFG7d3HXQ6fdx/o1M8/MJitM/37L7/zX//fn7O6Pm6r2Okc0DngM4BnQM6B3QO6BzQOcAceFAo5rdIo4n1x2M6Pr/B635DpnaP/cs+Vo56eLFJh1gBMEuhCbkEY0KvGg/hOIDngUMsr/NBmG7tiCPMEmeu+aXzO77pYSLliTO8vN9B6E0X4ZOObIxlnnYROeFcBzw2veXKo4+GDvEQgJ+LO/w7Y0JxnK91MLPlIvq2gzdXPXwp9ZGrK+ClC/yp0MO7jALmD9m+gPNFtofLQk9Kly+LPZxcdZE487B+qMD4A9f7NvqoNb/hY66L8XgbL6I2XtIhDqCYjzxKOFjetbG842B60xY3WUExy5zbWNtzcPDZQ/SEu0crYCbwEozZypzfV2M1H0AxXeEAignGi2m1S3QAxTw+hOI7AeZRIFZu8RCQ6RQThCcIxALCdIuHQMx1xEoBHHMDrSEQK7fYd4TNOqaiLJseOsRTdI6NGuZidaRPW2g6fXjde4FjOr+c/5zxBLy+/fYNne43fLh28DrNTbWq4DkE4gCKWUI9CsWEYULxLDfGihKAuYu0AuEl7gAdpztcGWje4trhIRSzTFrGPhSzPHoAxUZJ1g8ThAPRMV4wFBQThFdjav1wsGaYsCxAPOISE5rpEsvGWYRji5DL0um8gPNCOCdwTCd4LVqQDbUs7gq9W4G1VRHo5fphSsGxAmMC8etoAXSIzVQRqd2ybKK1e6yAeOugIpBMGCYUm8mCuMQE4pAAsQJjM5kXl/jgpIqz8ztcfW2hUnNhOz3c3/8mX1YE95jR+wsdYir4QuO335SjHJzLdvR8Pdbx0PnwK3+vdb7ofNH5ov+eDMFJXw/6evh3vB4eBIoDV+2vt/e4v7/H/bd72N49rsp9pC96mKJLLBBM8FWu7+OoByoYq9bFk6ir5qRV46cxF0/9EmYpbY67eBZnubOLl0lXYPhV2hPQnd/xsLzvIfTGg/m2A/O0g+iZao0TD6Fj7iLtYiKpHF+6w099F/hp1Mbv6VnUxqjoFk+lHKwduEi974CuL93dTKWHy2IXb6872P/UkRLlg08dcXVPrjp4f9PFRa6L82wXh587iL11ETp0sfXBQ8vpo+3eo9zowzpx8MJq4bnVkpZ9ivA7znJnusN+S2c4cIIX0m28verA9vrIVbuIHLUxFW/hpdnEC7Mp7UuzgVGNmQ38qAnuAB1XCtzgcfNOziMcj5l3mBDn+BbjArxDCOZ5nJN5HjPrPxUdYAFeq+47xXXMROsKeK06Jo0aJoyatAN32KhhihBLtzg4RqA1qwjv3CFT8tC/v0fL6SF10hTQnY/VUKx2JDe9bh+fMgTiOugAE37pEo9C8LRRxnSkDGmNMmbMYX/WZBm00kJUPS5pJV7BUowqY8Hyy6MjJcz4ChziYVvEbKSIuR80b6g5tnSCF+gGE3zp+gZ9v13kDtEjJdLqkUssoy5ghWXQftk0IZhO8XxIQfGyoZziUKIIg+t+k9x9Wm2otRDKglr0tRSiS5yVxy9Z6SKSuyVsH/IxTFQV2wcsoy5ja7+M+FYRsXRBQPi1mcWGlcW6mcVrI4NwLIfEVhG7B2Ucv63i/NMtsvk27hodeB6h+H7g4P94r+n379HrqfuKOvb75/74Wj0eVkboWOhY6BzQOaBzQOeAzgGdA/+JOfAgUMxvC0aD9+djljiqD7AN+x6fS/ewzroYT/kw7Du+4hILEA8hmXNP44RkgrBquTZYjV3QvX2e8AYwzDLpsU26w66UQy/seViRUmkPG8cejBPCcBcWwdiHYzrFS/tcE+yAZdD8mQTipzGCsS3l0qNQzPLp0TFfQzB+zucAxx0s7HhY3XdB2N4+7+BDtodPuZ4C3ssOti86Uh69de5hR8qduxAwznTwIUNQ7iJ66mL9wMHuRw939j2+FLtInrkYl/XB7REgbvpri/1S6WgTsmZY1g83MRZtYj7dxu6Fizub5aj36PW/oVDrYuu9i5kkd5VmCTYBWkEx1xkHcEwo5nriUTgOHGFxiM0GCMrj1p2I5xGWh/BL95djBcRSIs1y6hEgVqXTqhya87ImWB6bRCiuS9n0TEz1xQm26gLFQYl04AgPxwTjqmghUcfJJweEXuZgvdlDeLchULycrKPZ7kk8rgsuNra5yzRBmOXSQyAW11jGCoJ5zgwdX7+ds1RZNKGY8Lscr6pnCMfpEpfFKebGWYRfAjHdYQHjsBrLsXARs2Effk2WUQ/hmOuHF4yiSK0pZkl1wZ8ryPpiHldriFXJ9LLlg3C0gDVxiBUUs2xagXFQOp2TtcSr0YK4woRiOsUsm2Y5deASB1DMsukVI49VM4eNWB6JnTJ2Dqs+EFewc6hKqNP7ZWxJ+bTaSEtA2PSB2G9jqTx2Dss4PKnh7EMdX66bKFUcWUvc7Y4Cb/DHiuBL/YZer4++QLEqr/61+9Gv3r/0+Tq+QQ7qfPvzv/f6etHXi75eghzQ14u+HwS5wFbnw/+vfHgQKA4A91fafr8vH2SLdz2c3XSxfsxdn108jgZy8DhKBePvW3GKo47AMKH0SeDkxtUa3xdxB2MJB+NJF5ObLqbSjji/a4cuQscewm88RE48mKcUy5s9WG8JyZ64ulMpG89jdITbeGr5LR1ii+M/1jOrDeo5N6RKtLG6byN87MA6oVvs4fDSw9lXusaewGnynYvEmYvNdy7S7wMw5jke3t94OL70kHrvwnzjYPOdg0y1g9Q7F1OJFl4l6Aw38dyiw9sQl5dOL+fUuOGvC25hLtXC2m4L+x8d1JrctIhgyEfcqPei3uph54ODyGFboDcA4ZfmHSi6voF+HAfzhF0pfY7+6AIrR3jcGDrD4wZh9vdFh5hQzJbwSxGKKQXDdIerAsQTAr1qzLnvFKliMsLy5yriRw1Ubjvy/9zt9ZGreKBDPGVWEN69hdvpIV/xENm9xbRZwVSkPCKOh3PTRgmBZvxNsTiesxQMz5v+RllWBfNWSQCZm2XNGsoBngkXQU2LCpgJDzUbLoCaixQwZ/itP543CuIOE4TlOOfDecyFCwLHnOca4iVxiQviCtMZXosVpCRani9s5rFEyI3QGVaaZ0m0kcN6rIBQooD1WB6hRB7GZhGvozksRugQZ7Cw4SuUwXKYDnEOoXgesa2ClEjvHFWwc1QG2639CtJ7JSS2ufEWS6eLMBN5bESzCEUJxRmsRW4QjmaxuVvA/nEFb3yXOJNroX7rwnW66PfUlxg/u8cQiLvdntxPfnZcz/mVOf4XkToeOh46B3QO6BzQOaBzQOeAzoHvc+BBoJhQpQDrXtq/Mu71euLWZSpdvLnqYPmADi/BV8Hwo6iNJ7EAjB08thQAB8c5VuXSysF9FrfxfADENsaSNma3XCldnk47WNpzQCDeOHIReePCOHVhEoLZnjow3zrSXz9yMcdn/ibaeCZQ3MITgeCWuL9PrRYCPRNg9scmjxOWW3jmi1A8Fm8JFEdPHexcuNg+d7D/ycXpFw8nXzzsXThInjmInToCvun3jsydXrniEr//6sm5b689nGcJyC4KtQ7CRzZS722s7rUEfp+z1JlwbNzhBZ/rm2hiZrOJ2c0m1vZaSL2zBbAzZQ+NdheMf1+AmDCh1On2cdvq4uiT48MvneI7vDRuRQTfoM82GI+xL7B7hzGjjnGCrGyURQeYa4KDsuhbEISVagK840YNgQjA45GaryomTb8smmt3o3Uod7iGaYsbaKnj45EqqIlIRZzgyUgFgegOB322q6k6zq9tdASgerLJ1sVXG6/8EujUSQO1Ow/W/i2mjLKUS/PYq7ASgfhVuCSaihCICcwl0TRdXymnLg52jaZTPMNSZ4FjtiyZLmImQhAuiDvMNlAwL8dChGG6wwqOB4DMEmeDAJwXAGY7S4XUeD7CVgEvy6OXzbxoxcrjdbwItlw/vGzQEaY7nMN8KCviGuHXsTxifFzSrlJkk+uK81gzuYY4i/kAiDcysrnWSiSD9WgORrKA9AGfSUwYLktLdzi1V8LmbhHxrYKsM45u5mHEczATOUTiObw2brAW+QozmfNd4gpO39dweXWHcsVBs+Wh01H5yZt3cG9hznIs9xGvi06nOzgWnDN6/q/en/T56kuIIJY6Hjoe+noa3n/09aCvB3096OtB/31U7PfvdD98UCgeTZA/7qsPubbbw1WZjyvqYHbXlfLkAHof+XBM+BUFY7aWLc4w1/nSESYMv6ArvEln2MGrTRsTmzZmtx3M7zhY3h8CceiYEOwi/s5F7EwpyvW6xy5W9h1Mp22MxbkuWLnBCojbPhgr6A2g+I/aZ4Rkru2NtTCbtmGeOOIKH3xy8OaLi7fXLk6vPBx+ojvsIPXOkXb73JXjF1kPn/MdfMx28OHGw2Xew9eSh4usizeXDjYO28hV1aOSXlh3IBTTGX5u3mE60cTuuSPnvf/qoljvwPF8+O37rQ/CARD3+z2BxWylg7lNwjBBWDnDoyD8s34AxYOWcEsJKAcQzLbmzysQHhsB4gCMB1Asx5QL/MqqYTpWw1y8htl4DTPRGiZNwnDFB2LfGY5UMGEMoZigHGjaqiJ92sBts4Neny55D7bTxennlkAvATd+dIedtyylLgsovxIILmMqTPBlGwCyAuFRQBZIJigTeAWQCckKiOkYz5kKlnl8FH6nQ0MoDuA4aOkcK8eY4Mt+AL8KihUY5zEX4fGcwPAiQZcl0VwvbPIxS9xAi2Dsu8RWHnSEKbrEyilWULwQzopDvLlfRvqwIopu8VnFeawaWXDtcOASL4pLnMGqmUUolkM0TSguY/uoLKXThGO6wwTirYPSAI6NRA5WMgfCMd1hQnE4lsHmdgG7h2UcnZTx/qKGTK4pLrHjdNQXOPKeDW/AvL/QIe54XbhuV5ziP77nfP/a4Fzmf9DX7c9jpOOi46JzQOeAzgGdAzoHdA78u+fAg0FxAFfBh8w/G3d7XXEsP+c9bH/0pLz5cbSNRz7wEnyV2gLAwVgdV+XML+I2xpM2JlIKiGfStgDxTJo7PttY3nPw+tAB3d+1QwehYweRNw7ME1c2qCKocrxx5GBp18Z0uo3nsRaex9p4GqVDrMQS6idWcyDC8BOzOVAwDlrCMI8/Yxkzn7mbbGF5t43kOxtHl45fFu3iHV3gKxeHn+kgK+d359yR+cuCh6uCh885D59yCoaPLx1ET/j84BY+3DioNTqYTTbEGaZDTCCmzKM2qnceul3CQlcgkPHu9dSYQCh9jjnf78l59WYH5hHLrm/xwneHXxp1H5DrkL5Rx5gA84/j2uA4gZjnEnpFJvs+FPtgPGZUIaLTK+dVMRa4vrJxFoFYQe6UxZ2ha1hM1rCYUP1Jg0BMlaVU+pUPwxN0dsUhLmOSxwiyRgXrWzVc5WwVj14P3V4HjZaHvXcNBcDhkjwuiRtkKXdYOcKTAsTsFwci9E7JuCBzMo4UMSWur4JfgjC1FKNYTs0xgZgOcRHTofxAnPtuLGXUvgPsl0XPhHKY9UUnWIA4nMcyN78yWDatYJhQLI9XigzBV8qkCbUGy6SH7UI4I85vAMYrRhbr8Tw298rYOiyDQBxK0D2mS5wBQXhh40baxdANVo0MNuIKiOkIE4S5qRbb5HYB0VQeCXGIi1JCzbGRyIJrh424Kp1eN28QS+extV/E3lEJJ+8quLy+Q7HcRrNJl1g5wD+7n3heF7bjwev4ee5/0cOb9s/OD+Z+dlyuk//D64OfzfZnP18fH34Zp+Pz6/mp80fnT5AD+vrR10+QC2x1Puh80PnwcH8fHgSKR9+Qv9rnh9Bagzssc6MrtakVwfdR4AwPoHgIx4+sNh5bBFQFxVxLTEf4VUqJUEynd2HHFteX63hZLr1+5CB0pOA3/MaBceKPD22sH9qY37YxtdnGy3hLYFhBcAuPCb4+GA+heFg+PQrG7A+cY/Z9KOZa35fRJmZTLZgnNrbOHRx/cQVqLzIuqLNrFydfHJxeKX3MuchUOqBre10kGLv4UvDw7quL+VQDqXdttJ0uSrcdvEr4MEwo9sF478JGs90ZQHEAx+qDfwAQXfR8aOZ82+ng5LMtAPzcuMWLSB0vCMQj+nE8euyloaCYEDzaZ6n0GEuifVdY9RUgB2AclECPtlwjTDdYnF+ziplYDcupOpaSNcxG1aZZBGIRy5hNlk/TJSYMq3YiXAY1a5VxeN5Ay+6oLwHk/7uD2p0La78Ogq/Ar7QEaQXEygkmDPsl0wRkUQDIhOKCADIhWa0NVk4wS58X+fzgWBlL0SIWrKKsDab7KwA8aIdw/D0YK1dY3GHfIR726QrnMEe311Tl0mwJvwt+WTRheclkibQSwXc+nBVxbbCsDw5nBZIJyqtWDmaqgOReSXaPjm+XsGbxucPcWTqDZSMj7VKY64hVn1AcSeaQ3i8NSqZZOs1xcqeAuKwjzmNzJy+QbG3mYW3mRKFoRrnE0RsktvLYPiji6LSCDx/rvkvswHHVFzs/u6cQlm2ng3ab4Ew3WS0J+Nm5vzKnrpHhDfZXXqvP1XHTOaBzQOeAzgGdAzoHdA78PXPgQaA4gC4mQdAPPlz+3tj1OijUu+Karh1wPTBLognACnwFfv3xI5OAqtb2Pjbb0ieA8jUE4qlUW7S4a2NlT4HuxpEta4cJveFjG9apDestnWE1Nt7YAscre23MpFuYSLZk/S8fbRTArkAu/7sCx00FzD7wPjEbUh7NNtBTuslmA08NJTrFz1jWzDW+yaaUPCfetnH02caHGxuf8+5An7KOlEhf5l1cF13ZAIplz5my6nOtK0uhw4dtFGoePK+DYt3DZPwWhNhnRl0cXpY9n1y20bIVUBAYAvG96HQURHQ7Cpp5zHE7uC44mEneCQgLEEfoENfxMlIbiG6vgLA/J2PpV/EyQudXOcPBaxQIs8xZQTEfmzQWVo7wGF1eIyiBVq4vQZjub1DyTCimI0xNWxUsJLieWMGvOMT8GeGyaCJSGkDxRLgEioDMNrJbQ7ZkCxB3uioeruchW3KwEOM5RUzS+Y0QjouYDPlj3x2eDBcwGVLwqxzjAl4NxpxXJdBBWTTXDfMxSSuJkq8i5k21Rng6nMdUKCeSjbVCOUwHY3GP1ZhzM2ECcw7KJVbuMMuklWNMwFVuMFsC8gLd4gg3y8pi2cxiLcay6SwWjSwWBiCcwXxIiU7xIh3gcBbrsZwA8eZeSdxhQjCBeNXKYMXg+uEbLIZVy/G6lYG5mUdyt4DdY+44TTBWSu0VkdzhTtR0ihUQh+MZgeFYKgcjkQGhOBLPIrGVw9ZeAXvHJZydcy3xLQqlJhpNR1zif71/8P3rwrY9NBqutMzhP7vf/NXjwc/6q+f/6++nvnTSr1dfUuj46HwIcuBXPh/o60dfPzpf/vrnaX296OtFXy8Pc708CBQH0PUrreN4yFXV2tq1Qz7mqI1H5oiskb7MKzAmHFPPooTYNma321jYaWN5z8bSLts2CNkr+9z1uY2NozbCx0MRltcP2lg/VMfnd1oCxeIS+1BLCH5sNvDYaOCJ0RQ9ZksAlrk/b59GCMZ3eMp1vnwEUayBxe0mYidt7J7bOLt2xCW+zDu4KrjgI4AIw19LDm7KLrIVD7mKi0zZkTZbdnD4sY3zr7YAL6GufOtiKk6HuC5QTDB+ad3i/VcbbccDv3hotj3cNT1pCb/fv0cKrst1Fxt7d3geqYleSFsd9DlWqvqtGiv4VUBMKB6qJqXQUg5tVEAAHopQPDoe6RNwffeXZdAiusDcvTlexUKCzxmuCOgKDIdKGAuXFBhHFAgTjgnC46EiJkIlTJtlvPnYQNtWXwbIlwKeh2bLxdvPDUxFeJ6SAmICsA/GIQXDBOKh8n4/j1chX4RXbohlKC3FiljbLGI5xh2n8+ISz0SUIzy1oYCYYBzAsMxtZDG1kcX0RlYgeGZDwbCM2ef8RhazoaEWjCzmI1nMhTOYC2cxF1ItoXjJVCC8EMmA8DsXusHsxlBzGxnMsRSaLrCZRSSVw+Z+EYndAtasrEAwS6aXjRsshL9iMaQUALGRzMlu0duHRYFhOr0sgU7tFgSI41s5UKm9vGyiFYqyTDorQLxufsVr4xpmMoPkdh7b+wUcnZZw/qmGTLaBWs0ewO73+aryl0Bcv7VRq7XB+8jPztFzP17reqxzQueAzgGdAzoHdA7oHNA58Hs58CBQDP1PR0BHQEdAR0BHQEdAR0BHQEdAR0BHQEdAR+BvGAENxX/DN03/yjoCOgI6AjoCOgI6AjoCOgI6AjoCOgI6Ag8TAQ3FDxNH/VN0BHQEdAR0BHQEdAR0BHQEdAR0BHQEdAT+hhHQUPw3fNP0r6wjoCOgI6AjoCOgI6AjoCOgI6AjoCOgI/AwEdBQ/DBx1D9FR0BHQEdAR0BHQEdAR0BHQEdAR0BHQEfgbxgBDcV/wzdN/8o6AjoCOgI6AjoCOgI6AjoC/1kR+Oc//wf/+Md/o1AsI5FMY2p6Do+fPMd/PXqqpWPwl3OAOcPcYQ4xl5hTzK3/9H+/CsX/C9QVwei+nNIoAAAAAElFTkSuQmCC"}}},{"cell_type":"markdown","source":"thanks https://www.kaggle.com/mlconsult/1-35-lightgbm-ann credit to @KenMiller","metadata":{"papermill":{"duration":0.040695,"end_time":"2021-07-20T23:07:40.467127","exception":false,"start_time":"2021-07-20T23:07:40.426432","status":"completed"},"tags":[]}},{"cell_type":"markdown","source":"> 📌プレーヤーがサヨナラホームランを打つ。投手はノーヒットノーランを投げます。チームはポストシーズンに入ると真っ赤になります。私たちは野球ファンの関心を高めるいくつかの触媒を知っています。現在、メジャーリーグベースボール（MLB）とGoogle Cloudは、Kaggleコミュニティの支援により、サポーターの関与を刺激し、プレーヤーとファンの間に深い関係を築く他の多くの要因を特定することを望んでいます。\n> \n> このスポーツには、数字主導の長い歴史があります。少なくとも4月から10月までのほぼ毎日、野球ファンは選手に関する情報を見て、読んで、検索しています。どの個人を探すかは、プレーヤーのパフォーマンス、チームの順位、人気など、現在不明な要因によって異なります。これは、データサイエンスのおかげでよりよく理解できます。\n> \n> 少なくとも1990年代初頭以来、MLBはデータの使用においてスポーツ界をリードし、データと人間のパフォーマンスを組み合わせたときに何が可能かをファン、プレーヤー、コーチ、メディアに示してきました。 MLBは、テクノロジーを使用してリーダーシップを継続し、ファンを引き付け、新しいファンにアメリカのお気に入りの娯楽を体験する革新的な方法を提供します。\n> \n> \n> \n> MLBはGoogleCloudと提携して、データを通じてファンの体験を変革しました。 Google Cloudは、Vertex AIの立ち上げを祝うこのKaggleコンテストを誇らしげにサポートしています。これは、MLワークフローを統合するGoogleCloudの新しいプラットフォームです。\n> \n> このコンテストでは、ファンがMLBプレーヤーのデジタルコンテンツを将来の日付範囲で毎日どのように利用するかを予測します。プレーヤーのパフォーマンスデータ、ソーシャルメディアデータ、市場規模などのチーム要因にアクセスできます。成功したモデルは、どのシグナルがエンゲージメントと最も強く相関し、影響を与えるかについての新しい洞察を提供します。\n> \n> MLBオールスターラウンドをシーズンを通して予測できるかどうか、またはチームの25人のプレーヤーのそれぞれが脚光を浴びているときを想像してみてください。これらの洞察は、アメリカの娯楽のファンダムを深く掘り下げるときに可能になります。この種の最初の方法の一部として、プレーヤーレベルでのデジタルエンゲージメントをこのきめ細かい日常的な方法で理解しようとします。同時に、Google Cloudのデータ分析、Vertex AI、MLOpsツールを使用して、MLBがイノベーションをより簡単に構築できるように支援します。 MLBファンとプレーヤーのエンゲージメントの未来を形作る上で役割を果たすことができます。","metadata":{"papermill":{"duration":0.040553,"end_time":"2021-07-20T23:07:40.548856","exception":false,"start_time":"2021-07-20T23:07:40.508303","status":"completed"},"tags":[]}},{"cell_type":"markdown","source":"> 📌あなたは、2021年シーズンに活躍するMLBプレーヤーのサブセットについて、4つの異なるエンゲージメントの測定値（target1-target4）を予測する必要があります。データには、時間の経過とともに変化しない一連の静的ファイル（players.csv、teams.csv、seasons.csv、awards.csv）と、日ごとにグループ化された日次データ（train.csv）が含まれています。特定の日付を予測する場合、次の日のターゲット変数を予測します（つまり、日付dの場合、日d + 1のエンゲージメントを予測します）。\n> \n> これは、時系列モジュールに依存してモデルが時間的に前向きにならないようにするコード競争です。時系列モジュールは、テストデータを提供し、送信ファイルを自動的に書き込みます。テストデータは、ターゲット値が含まれていないことを除いて、train.csvと同じ形式のデータフレームで到着します。提出するには、評価ページの指示に従ってください。ノートブックを送信すると、表示されていないテストセットで再実行されます。\n> \n> コンテストのトレーニングフェーズでは、この目に見えないテストセットは、2021年5月の月のデータと今年のアクティブなプレーヤーのセットで構成されます。\n> 評価フェーズでは、テストセットは約1か月の将来のシーズン範囲になります。\n> コードは堅牢で、モジュールによって要求されたdate_playerIdの組み合わせを予測する必要があります。各チームが選択したノートブック（チームごとに最大2つ、最終提出期限までに選択）は、評価フェーズ中に再実行されます。\n> \n> 詳細に飛び込む前に、データに関するいくつかの高レベルの資格：\n> \n> 一部の自明のフィールドには説明がありません（例：季節）\n> バイナリ列には、ゼロだけでなくNULL値も含まれます。プレイヤーが何かをする機会があったが、しなかった場合、ゼロが発生します。プレーヤーが何かをする機会がなかった場合、ヌルが発生します（たとえば、特定の日にピッチングしないプレーヤーは完封をピッチングできない可能性があるため、ヌル値が予想されます）\n> ほとんどのゲーム状態関連フィールド（ボール、ストライク、アウトなど）は、問題のイベント後のゲーム状態を表します。ただし、ホームスコアとアウェイスコアは、イベント前のスコアを表します。","metadata":{"papermill":{"duration":0.041956,"end_time":"2021-07-20T23:07:40.631837","exception":false,"start_time":"2021-07-20T23:07:40.589881","status":"completed"},"tags":[]}},{"cell_type":"code","source":"# --- CSS STYLE ---\nfrom IPython.core.display import HTML\ndef css_styling():\n    styles = open(\"../input/competiongoal/archive/alerts.css\", \"r\").read()\n    return HTML(\"<style>\"+styles+\"</style>\")\ncss_styling()","metadata":{"papermill":{"duration":0.070386,"end_time":"2021-07-20T23:07:40.744547","exception":false,"start_time":"2021-07-20T23:07:40.674161","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:35:06.519863Z","iopub.execute_input":"2021-07-24T08:35:06.520701Z","iopub.status.idle":"2021-07-24T08:35:06.553727Z","shell.execute_reply.started":"2021-07-24T08:35:06.520401Z","shell.execute_reply":"2021-07-24T08:35:06.552327Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<div class=\"alert simple-alert\"><font color=\"black\">\n⚾ <b>Competition Goal</b>:あなたは、2021年シーズンに活躍するMLBプレーヤーのサブセットについて、4つの異なるエンゲージメントの測定値（target1-target4）を予測する必要があります\n</div>","metadata":{"papermill":{"duration":0.041112,"end_time":"2021-07-20T23:07:40.827757","exception":false,"start_time":"2021-07-20T23:07:40.786645","status":"completed"},"tags":[]}},{"cell_type":"code","source":"\n\"\"\"\n!pip install pandarallel \n\nimport gc\n\nimport numpy as np\nimport pandas a\nfrom pathlib import Path\n\nfrom pandarallel import pandarallel\npandarallel.initialize()\n\nBASE_DIR = Path('../input/mlb-player-digital-engagement-forecasting')\ntrain = pd.read_csv(BASE_DIR / 'train.csv')\n\nnull = np.nan\ntrue = True\nfalse = False\n\nfor col in train.columns\n    if col == 'date': continue\n\n    _index = train[col].notnull()\n    train.loc[_index, col] = train.loc[_index, col].parallel_apply(lambda x: eval(x))\n\n    outputs = []\n    for index, date, record in train.loc[_index, ['date', col]].itertuples():\n        _df = pd.DataFrame(record)\n        _df['index'] = index\n        _df['date'] = date\n        outputs.append(_df)\n\n    outputs = pd.concat(outputs).reset_index(drop=True)\n\n    outputs.to_csv(f'{col}_train.csv', index=False)\n    outputs.to_pickle(f'{col}_train.pkl')\n\n    del outputs\n    del train[col]\n    gc.collect()\n\"\"\"","metadata":{"_cell_guid":"733fb600-c8da-4658-aa35-320c6817a8c5","_uuid":"62aca0d3-6af3-4760-9db0-0a397fdc5191","collapsed":false,"jupyter":{"outputs_hidden":false},"papermill":{"duration":0.051479,"end_time":"2021-07-20T23:07:40.920494","exception":false,"start_time":"2021-07-20T23:07:40.869015","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:35:06.561372Z","iopub.execute_input":"2021-07-24T08:35:06.561758Z","iopub.status.idle":"2021-07-24T08:35:06.571028Z","shell.execute_reply.started":"2021-07-24T08:35:06.56172Z","shell.execute_reply":"2021-07-24T08:35:06.569337Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Training","metadata":{"_cell_guid":"8d708767-e50d-4684-82a1-790feb5c0c1f","_uuid":"ea4d85e1-21d7-4e7a-a8e7-36e5f53f4612","execution":{"iopub.execute_input":"2021-06-16T09:14:33.869905Z","iopub.status.busy":"2021-06-16T09:14:33.869464Z","iopub.status.idle":"2021-06-16T09:14:33.874766Z","shell.execute_reply":"2021-06-16T09:14:33.873097Z","shell.execute_reply.started":"2021-06-16T09:14:33.869879Z"},"papermill":{"duration":0.041432,"end_time":"2021-07-20T23:07:41.003571","exception":false,"start_time":"2021-07-20T23:07:40.962139","status":"completed"},"tags":[]}},{"cell_type":"code","source":"import numpy as np\nimport pandas as pd\nfrom pathlib import Path\nfrom sklearn.metrics import mean_absolute_error\nfrom datetime import timedelta\nfrom functools import reduce\nfrom tqdm import tqdm\nimport lightgbm as lgbm\nimport mlb\nimport matplotlib.pyplot as plt\nimport seaborn as sns","metadata":{"_cell_guid":"3181642b-6cb0-424d-b5be-a0e6a5cb0457","_uuid":"2139878b-da24-41e3-bb59-76b60a1e16ef","collapsed":false,"jupyter":{"outputs_hidden":false},"papermill":{"duration":2.313042,"end_time":"2021-07-20T23:07:43.358143","exception":false,"start_time":"2021-07-20T23:07:41.045101","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:35:06.607703Z","iopub.execute_input":"2021-07-24T08:35:06.608265Z","iopub.status.idle":"2021-07-24T08:35:09.51901Z","shell.execute_reply.started":"2021-07-24T08:35:06.608223Z","shell.execute_reply":"2021-07-24T08:35:09.51763Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"BASE_DIR = Path('../input/mlb-player-digital-engagement-forecasting')\nTRAIN_DIR = Path('../input/mlbdata') #../input/mlb-pdef-train-dataset')","metadata":{"_cell_guid":"5f9dc680-5158-4bf2-856f-d43b6aa620de","_uuid":"3540a2ce-95e1-416f-9892-bcfa92cf6047","collapsed":false,"jupyter":{"outputs_hidden":false},"papermill":{"duration":0.050191,"end_time":"2021-07-20T23:07:43.450973","exception":false,"start_time":"2021-07-20T23:07:43.400782","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:35:15.041615Z","iopub.execute_input":"2021-07-24T08:35:15.041991Z","iopub.status.idle":"2021-07-24T08:35:15.047847Z","shell.execute_reply.started":"2021-07-24T08:35:15.041952Z","shell.execute_reply":"2021-07-24T08:35:15.046461Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"players = pd.read_csv(BASE_DIR / 'players.csv')\n\nrosters = pd.read_pickle(TRAIN_DIR / 'rosters_train.pkl')\ntargets = pd.read_pickle(TRAIN_DIR / 'nextDayPlayerEngagement_train.pkl')\nscores = pd.read_pickle(TRAIN_DIR / 'playerBoxScores_train.pkl')\nscores = scores.groupby(['playerId', 'date']).sum().reset_index()\n#scores = pd.read_csv(TRAIN_DIR/ 'playerBoxScores_train.csv')","metadata":{"_cell_guid":"e67cda79-ae71-474a-861b-2b437b94a0a8","_uuid":"38e056cf-cf8f-45f0-9ad4-cda6d11917a2","collapsed":false,"jupyter":{"outputs_hidden":false},"papermill":{"duration":4.400832,"end_time":"2021-07-20T23:07:47.895718","exception":false,"start_time":"2021-07-20T23:07:43.494886","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:35:15.050332Z","iopub.execute_input":"2021-07-24T08:35:15.050775Z","iopub.status.idle":"2021-07-24T08:35:19.801626Z","shell.execute_reply.started":"2021-07-24T08:35:15.050737Z","shell.execute_reply":"2021-07-24T08:35:19.80042Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"targets_cols = ['playerId', 'target1', 'target2', 'target3', 'target4', 'date']\nplayers_cols = ['playerId', 'primaryPositionName']\nrosters_cols = ['playerId', 'teamId', 'status', 'date']\nscores_cols = ['playerId', 'battingOrder', 'gamesPlayedBatting', 'flyOuts',\n       'groundOuts', 'runsScored', 'doubles', 'triples', 'homeRuns',\n       'strikeOuts', 'baseOnBalls', 'intentionalWalks', 'hits', 'hitByPitch',\n       'atBats', 'caughtStealing', 'stolenBases', 'groundIntoDoublePlay',\n       'groundIntoTriplePlay', 'plateAppearances', 'totalBases', 'rbi',\n       'leftOnBase', 'sacBunts', 'sacFlies', 'catchersInterference',\n       'pickoffs', 'gamesPlayedPitching', 'gamesStartedPitching',\n       'completeGamesPitching', 'shutoutsPitching', 'winsPitching',\n       'lossesPitching', 'flyOutsPitching', 'airOutsPitching', ###\n       'groundOutsPitching', 'runsPitching', 'doublesPitching',\n       'triplesPitching', 'homeRunsPitching', 'strikeOutsPitching',\n       'baseOnBallsPitching', 'intentionalWalksPitching', 'hitsPitching',\n       'hitByPitchPitching', 'atBatsPitching', 'caughtStealingPitching',\n       'stolenBasesPitching', 'inningsPitched', 'saveOpportunities',\n       'earnedRuns', 'battersFaced', 'outsPitching', 'pitchesThrown', 'balls',\n       'strikes', 'hitBatsmen', 'balks', 'wildPitches', 'pickoffsPitching',\n       'rbiPitching', 'gamesFinishedPitching', 'inheritedRunners',\n       'inheritedRunnersScored', 'catchersInterferencePitching',\n       'sacBuntsPitching', 'sacFliesPitching', 'saves','holds', 'blownSaves',\n       'assists', 'putOuts', 'errors', 'chances', 'date']\n\nfeature_cols = ['label_playerId', 'label_primaryPositionName', 'label_teamId',\n       'label_status', 'battingOrder', 'gamesPlayedBatting', 'flyOuts',\n       'groundOuts', 'runsScored', 'doubles', 'triples', 'homeRuns',\n       'strikeOuts', 'baseOnBalls', 'intentionalWalks', 'hits', 'hitByPitch',\n       'atBats', 'caughtStealing', 'stolenBases', 'groundIntoDoublePlay',\n       'groundIntoTriplePlay', 'plateAppearances', 'totalBases', 'rbi',\n       'leftOnBase', 'sacBunts', 'sacFlies', 'catchersInterference',\n       'pickoffs','gamesPlayedPitching', 'gamesStartedPitching', \n       'completeGamesPitching', 'shutoutsPitching', 'winsPitching',\n       'lossesPitching', 'flyOutsPitching', 'airOutsPitching',###\n       'groundOutsPitching', 'runsPitching', 'doublesPitching',\n       'triplesPitching', 'homeRunsPitching', 'strikeOutsPitching',\n       'baseOnBallsPitching', 'intentionalWalksPitching', 'hitsPitching',\n       'hitByPitchPitching', 'atBatsPitching', 'caughtStealingPitching',\n       'stolenBasesPitching', 'inningsPitched', 'saveOpportunities',\n       'earnedRuns', 'battersFaced', 'outsPitching', 'pitchesThrown', 'balls',\n       'strikes', 'hitBatsmen',  'wildPitches', 'pickoffsPitching','balks',#@\n       'rbiPitching', 'gamesFinishedPitching', 'inheritedRunners',\n       'inheritedRunnersScored', 'catchersInterferencePitching',\n       'sacBuntsPitching', 'sacFliesPitching', 'saves', 'holds', 'blownSaves',\n       'assists', 'putOuts', 'errors', 'chances','target1_mean',\n 'target1_median',\n 'target1_std',\n 'target1_min',\n 'target1_max',\n 'target1_prob',\n 'target2_mean',\n 'target2_median',\n 'target2_std',\n 'target2_min',\n 'target2_max',\n 'target2_prob',\n 'target3_mean',\n 'target3_median',\n 'target3_std',\n 'target3_min',\n 'target3_max',\n 'target3_prob',\n 'target4_mean',\n 'target4_median',\n 'target4_std',\n 'target4_min',\n 'target4_max',\n 'target4_prob']\n\nfeature_cols2 = ['label_playerId', 'label_primaryPositionName', 'label_teamId',\n       'label_status', 'battingOrder', 'gamesPlayedBatting', 'flyOuts',\n       'groundOuts', 'runsScored', 'doubles', 'triples', 'homeRuns',\n       'strikeOuts', 'baseOnBalls', 'intentionalWalks', 'hits', 'hitByPitch',\n       'atBats', 'caughtStealing', 'stolenBases', 'groundIntoDoublePlay',\n       'groundIntoTriplePlay', 'plateAppearances', 'totalBases', 'rbi',\n       'leftOnBase', 'sacBunts', 'sacFlies', 'catchersInterference',\n        'pickoffs', 'gamesPlayedPitching', 'gamesStartedPitching',\n        'completeGamesPitching', 'shutoutsPitching', 'winsPitching',\n       'lossesPitching', 'flyOutsPitching', 'airOutsPitching',###\n       'groundOutsPitching', 'runsPitching', 'doublesPitching',\n       'triplesPitching', 'homeRunsPitching', 'strikeOutsPitching',\n       'baseOnBallsPitching', 'intentionalWalksPitching', 'hitsPitching',\n       'hitByPitchPitching', 'atBatsPitching', 'caughtStealingPitching',\n       'stolenBasesPitching', 'inningsPitched', 'saveOpportunities',\n       'earnedRuns', 'battersFaced', 'outsPitching', 'pitchesThrown', 'balls',\n       'strikes', 'hitBatsmen', 'wildPitches', 'pickoffsPitching','balks', #@\n       'rbiPitching', 'gamesFinishedPitching', 'inheritedRunners',\n       'inheritedRunnersScored', 'catchersInterferencePitching',\n       'sacBuntsPitching', 'sacFliesPitching', 'saves', 'holds', 'blownSaves',\n       'assists', 'putOuts', 'errors', 'chances','target1_mean',\n 'target1_median',\n 'target1_std',\n 'target1_min',\n 'target1_max',\n 'target1_prob',\n 'target2_mean',\n 'target2_median',\n 'target2_std',\n 'target2_min',\n 'target2_max',\n 'target2_prob',\n 'target3_mean',\n 'target3_median',\n 'target3_std',\n 'target3_min',\n 'target3_max',\n 'target3_prob',\n 'target4_mean',\n 'target4_median',\n 'target4_std',\n 'target4_min',\n 'target4_max',\n 'target4_prob',\n    'target1']","metadata":{"_cell_guid":"39e17ba0-96e5-438c-94bd-6c184905c35f","_uuid":"046564c8-2d25-4540-9a94-9e629e263a22","collapsed":false,"jupyter":{"outputs_hidden":false},"papermill":{"duration":0.06259,"end_time":"2021-07-20T23:07:48.001452","exception":false,"start_time":"2021-07-20T23:07:47.938862","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:35:19.803652Z","iopub.execute_input":"2021-07-24T08:35:19.804186Z","iopub.status.idle":"2021-07-24T08:35:19.825494Z","shell.execute_reply.started":"2021-07-24T08:35:19.804131Z","shell.execute_reply":"2021-07-24T08:35:19.824121Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"player_target_stats = pd.read_csv(\"../input/mlbdata/player_target_stats.csv\")\n#player_target_stats = player_target_stats.drop(['Unnamed: 0'],axis=1)\ndata_names=player_target_stats.columns.values.tolist()\n#data_names","metadata":{"_cell_guid":"5ee50f59-b249-4698-ae86-e35289c6df01","_uuid":"09b65c0a-4ac9-45cb-86ee-d81e871960a7","collapsed":false,"jupyter":{"outputs_hidden":false},"papermill":{"duration":0.079213,"end_time":"2021-07-20T23:07:48.12281","exception":false,"start_time":"2021-07-20T23:07:48.043597","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:35:19.827228Z","iopub.execute_input":"2021-07-24T08:35:19.827576Z","iopub.status.idle":"2021-07-24T08:35:19.877333Z","shell.execute_reply.started":"2021-07-24T08:35:19.827543Z","shell.execute_reply":"2021-07-24T08:35:19.875992Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# creat dataset\ntrain = targets[targets_cols].merge(players[players_cols], on=['playerId'], how='left')\ntrain = train.merge(rosters[rosters_cols], on=['playerId', 'date'], how='left')\ntrain = train.merge(scores[scores_cols], on=['playerId', 'date'], how='left')\ntrain = train.merge(player_target_stats, how='inner', left_on=[\"playerId\"],right_on=[\"playerId\"])\n\n\n# label encoding\nplayer2num = {c: i for i, c in enumerate(train['playerId'].unique())}\nposition2num = {c: i for i, c in enumerate(train['primaryPositionName'].unique())}\nteamid2num = {c: i for i, c in enumerate(train['teamId'].unique())}\nstatus2num = {c: i for i, c in enumerate(train['status'].unique())}\ntrain['label_playerId'] = train['playerId'].map(player2num)\ntrain['label_primaryPositionName'] = train['primaryPositionName'].map(position2num)\ntrain['label_teamId'] = train['teamId'].map(teamid2num)\ntrain['label_status'] = train['status'].map(status2num)","metadata":{"_cell_guid":"69624101-e45d-442a-bbb4-fb5514840e21","_uuid":"6cb3c4b2-a69b-4ce5-9d70-6633354c11ea","collapsed":false,"jupyter":{"outputs_hidden":false},"papermill":{"duration":7.909372,"end_time":"2021-07-20T23:07:56.074869","exception":false,"start_time":"2021-07-20T23:07:48.165497","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:35:19.878993Z","iopub.execute_input":"2021-07-24T08:35:19.879359Z","iopub.status.idle":"2021-07-24T08:35:30.202548Z","shell.execute_reply.started":"2021-07-24T08:35:19.879319Z","shell.execute_reply":"2021-07-24T08:35:30.20135Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### 追加説明関数","metadata":{}},{"cell_type":"code","source":"#train['add1'] =train['target1_mean'].shift(1) - train['target1_mean'].shift(-1)\n#train['add1'] =train['target1_mean'].rolling(5, min_periods=1).mean()\n#train['add1'] = train['homeRuns'] * train['hits']\n#feature_cols.append('add1')","metadata":{"execution":{"iopub.status.busy":"2021-07-24T08:35:30.205591Z","iopub.execute_input":"2021-07-24T08:35:30.205957Z","iopub.status.idle":"2021-07-24T08:35:30.210647Z","shell.execute_reply.started":"2021-07-24T08:35:30.205923Z","shell.execute_reply":"2021-07-24T08:35:30.209196Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### 日付いじるとスコアかわりますよ。","metadata":{}},{"cell_type":"code","source":"train_X = train[feature_cols]\ntrain_y = train[['target1', 'target2', 'target3', 'target4']]\n\n_index = (train['date'] < 20210426) \nx_train1 = train_X.loc[_index].reset_index(drop=True)\ny_train1 = train_y.loc[_index].reset_index(drop=True)\nx_valid1 = train_X.loc[~_index].reset_index(drop=True)\ny_valid1 = train_y.loc[~_index].reset_index(drop=True)","metadata":{"_cell_guid":"ffc2b454-9e48-44f8-b4fa-d2d87ea5e89d","_uuid":"ddc79e19-ca6c-40d2-bb71-cf5df58549ce","collapsed":false,"jupyter":{"outputs_hidden":false},"papermill":{"duration":3.851396,"end_time":"2021-07-20T23:08:01.476623","exception":false,"start_time":"2021-07-20T23:07:57.625227","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:35:30.212941Z","iopub.execute_input":"2021-07-24T08:35:30.21327Z","iopub.status.idle":"2021-07-24T08:35:35.949635Z","shell.execute_reply.started":"2021-07-24T08:35:30.213237Z","shell.execute_reply":"2021-07-24T08:35:35.948611Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_X = train[feature_cols2]\ntrain_y = train[['target1', 'target2', 'target3', 'target4']]\n\n_index = (train['date'] < 20210426) \nx_train2 = train_X.loc[_index].reset_index(drop=True)\ny_train2 = train_y.loc[_index].reset_index(drop=True)\nx_valid2 = train_X.loc[~_index].reset_index(drop=True)\ny_valid2 = train_y.loc[~_index].reset_index(drop=True)","metadata":{"papermill":{"duration":5.640031,"end_time":"2021-07-20T23:08:07.170097","exception":false,"start_time":"2021-07-20T23:08:01.530066","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:35:35.951488Z","iopub.execute_input":"2021-07-24T08:35:35.951807Z","iopub.status.idle":"2021-07-24T08:35:38.856028Z","shell.execute_reply.started":"2021-07-24T08:35:35.951777Z","shell.execute_reply":"2021-07-24T08:35:38.854793Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"![image.png](attachment:ba8790f2-2fa8-471d-85fb-478e501ca0e0.png)","metadata":{"papermill":{"duration":0.052619,"end_time":"2021-07-20T23:08:09.64932","exception":false,"start_time":"2021-07-20T23:08:09.596701","status":"completed"},"tags":[]},"attachments":{"ba8790f2-2fa8-471d-85fb-478e501ca0e0.png":{"image/png":"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"}}},{"cell_type":"code","source":"def fit_lgbm(x_train, y_train, x_valid, y_valid, params: dict=None, verbose=100):\n    oof_pred = np.zeros(len(y_valid), dtype=np.float32)\n    model = lgbm.LGBMRegressor(**params)\n    model.fit(x_train, y_train, \n        eval_set=[(x_valid, y_valid)],  \n        early_stopping_rounds=verbose, \n        verbose=verbose)\n    oof_pred = model.predict(x_valid)\n    score = mean_absolute_error(oof_pred, y_valid)\n    print('mae:', score)\n    return oof_pred, model, score\n\n","metadata":{"papermill":{"duration":0.055942,"end_time":"2021-07-20T23:08:09.752555","exception":false,"start_time":"2021-07-20T23:08:09.696613","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:35:38.860244Z","iopub.execute_input":"2021-07-24T08:35:38.860582Z","iopub.status.idle":"2021-07-24T08:35:38.867514Z","shell.execute_reply.started":"2021-07-24T08:35:38.860548Z","shell.execute_reply":"2021-07-24T08:35:38.866489Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# 残る唯一のチューニングポイント　@","metadata":{"papermill":{"duration":0.047815,"end_time":"2021-07-20T23:08:09.848022","exception":false,"start_time":"2021-07-20T23:08:09.800207","status":"completed"},"tags":[]}},{"cell_type":"code","source":"params1 = {'objective':'mae','reg_alpha': 0.14947461820098767, 'random_state':77,'reg_lambda': 0.08, 'n_estimators': 3633, 'learning_rate': 0.1, 'num_leaves': 660, 'feature_fraction': 0.9101240539122566, 'bagging_fraction': 0.9884451442950513, 'bagging_freq': 8, 'min_child_samples': 51}","metadata":{"papermill":{"duration":0.056213,"end_time":"2021-07-20T23:08:09.952978","exception":false,"start_time":"2021-07-20T23:08:09.896765","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:35:38.868716Z","iopub.execute_input":"2021-07-24T08:35:38.86907Z","iopub.status.idle":"2021-07-24T08:35:38.886394Z","shell.execute_reply.started":"2021-07-24T08:35:38.869024Z","shell.execute_reply":"2021-07-24T08:35:38.885485Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<pre>params1 = {'objective':'mae','reg_alpha': 0.14947461820098767, 'random_state':77,'reg_lambda': 0.10185644384043743, 'n_estimators': 3633, 'learning_rate': 0.08046301304430488, 'num_leaves': 674, 'feature_fraction': 0.9101240539122566, 'bagging_fraction': 0.9884451442950513, 'bagging_freq': 8, 'min_child_samples': 51} \n0.5770630841609712\n\nnum_leaves': 660  0.578658\nlearning_rate': 0.1 0.578658\nreg_alpha': 0.14947461820098767 #\nreg_lambda': 0.085  0.578275\nfeature_fraction: 0.9101240539122566 #\nbagging_fraction': 0.9884451442950513 #\nbagging_freq': 8 #\nmin_child_samples': 51 #\n","metadata":{"papermill":{"duration":0.047118,"end_time":"2021-07-20T23:08:10.047657","exception":false,"start_time":"2021-07-20T23:08:10.000539","status":"completed"},"tags":[]}},{"cell_type":"code","source":"oof1, model1, score1 = fit_lgbm(\n    x_train1, y_train1['target1'],\n    x_valid1, y_valid1['target1'],\n    params1\n )\nscore1","metadata":{"papermill":{"duration":292.151784,"end_time":"2021-07-20T23:13:02.248708","exception":false,"start_time":"2021-07-20T23:08:10.096924","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:35:38.887537Z","iopub.execute_input":"2021-07-24T08:35:38.887806Z","iopub.status.idle":"2021-07-24T08:39:44.950682Z","shell.execute_reply.started":"2021-07-24T08:35:38.887777Z","shell.execute_reply":"2021-07-24T08:39:44.949671Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"mae: 0.5765479264820539\n0.5765479264820539","metadata":{}},{"cell_type":"markdown","source":"## XGBRegressor","metadata":{}},{"cell_type":"markdown","source":"from xgboost import XGBRegressor\nfrom sklearn.model_selection import train_test_split, KFold\nxgb_params = {\n    'booster':'gbtree',\n    'n_estimators':200,\n    'max_depth':40, #6\n    'eta':0.01,\n    'gamma':2.8,\n    'objective':'reg:squarederror',\n    'verbosity':1,\n    'subsample':0.85,\n    'colsample_bytree':0.45,\n    'lambda':5,\n    'scale_pos_weight':1,\n    'objective':'reg:squarederror',\n    'eval_metric':'rmse'\n}","metadata":{}},{"cell_type":"markdown","source":"model11 = XGBRegressor(**xgb_params)\nmodel11.fit(x_train1, y_train1['target1'], eval_set = [(x_valid1, y_valid1['target1'])], early_stopping_rounds = 50)\n\ny_pred = model11.predict(x_valid1)\n\nscore11 = mean_absolute_error(y_valid1['target1'], y_pred)\nscore11","metadata":{}},{"cell_type":"markdown","source":"1.0764579770208662","metadata":{}},{"cell_type":"code","source":"params2 = {\n 'objective':'mae',\n 'reg_alpha': 0.1,\n 'reg_lambda': 0.1, \n 'n_estimators': 1000,#80\n 'learning_rate': 0.05,\n 'random_state': 77,#42\n \"num_leaves\": 155\n}","metadata":{"papermill":{"duration":0.062934,"end_time":"2021-07-20T23:13:04.314298","exception":false,"start_time":"2021-07-20T23:13:04.251364","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:39:44.954698Z","iopub.execute_input":"2021-07-24T08:39:44.956766Z","iopub.status.idle":"2021-07-24T08:39:44.96301Z","shell.execute_reply.started":"2021-07-24T08:39:44.956709Z","shell.execute_reply":"2021-07-24T08:39:44.962073Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"oof2, model2, score2 = fit_lgbm(\n    x_train2, y_train2['target2'],\n    x_valid2, y_valid2['target2'],\n    params2\n)\nscore2","metadata":{"papermill":{"duration":46.603164,"end_time":"2021-07-20T23:13:50.974383","exception":false,"start_time":"2021-07-20T23:13:04.371219","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:39:44.964229Z","iopub.execute_input":"2021-07-24T08:39:44.964659Z","iopub.status.idle":"2021-07-24T08:40:36.136746Z","shell.execute_reply.started":"2021-07-24T08:39:44.964626Z","shell.execute_reply":"2021-07-24T08:40:36.13557Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"1.1568237368972796","metadata":{}},{"cell_type":"markdown","source":"<pre>params2 = { 'objective':'mae', 'reg_alpha': 0.1, 'reg_lambda': 0.1,  'n_estimators': 160,#80 'learning_rate': 0.05,#0.1 'random_state': 77,#42 \"num_leaves\": 22 1.180154592803244\nn_estimators': 1000 1.1645554527856938\nnum_leaves:155 :1.1568237368972796\nreg_alpha': 0.1 #\nreg_lambda': 0.1 ","metadata":{"papermill":{"duration":0.061538,"end_time":"2021-07-20T23:13:52.581073","exception":false,"start_time":"2021-07-20T23:13:52.519535","status":"completed"},"tags":[]}},{"cell_type":"code","source":"params3 = {\n 'objective':'mae',\n 'reg_alpha': 0.1,\n 'reg_lambda': 0.09, \n 'n_estimators': 10000,\n 'learning_rate': 0.05,\n 'random_state': 77,#42\n \"num_leaves\": 800\n}\n","metadata":{"papermill":{"duration":0.070393,"end_time":"2021-07-20T23:13:52.713906","exception":false,"start_time":"2021-07-20T23:13:52.643513","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:40:36.140862Z","iopub.execute_input":"2021-07-24T08:40:36.143028Z","iopub.status.idle":"2021-07-24T08:40:36.150139Z","shell.execute_reply.started":"2021-07-24T08:40:36.142956Z","shell.execute_reply":"2021-07-24T08:40:36.148735Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"oof3, model3, score3 = fit_lgbm(\n    x_train2, y_train2['target3'],\n    x_valid2, y_valid2['target3'],\n   params3\n)\nscore3","metadata":{"papermill":{"duration":61.878519,"end_time":"2021-07-20T23:14:54.653731","exception":false,"start_time":"2021-07-20T23:13:52.775212","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:40:36.151717Z","iopub.execute_input":"2021-07-24T08:40:36.152079Z","iopub.status.idle":"2021-07-24T08:41:46.402393Z","shell.execute_reply.started":"2021-07-24T08:40:36.152044Z","shell.execute_reply":"2021-07-24T08:41:46.401449Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<pre>params = { 'objective':'mae', 'reg_alpha': 0.1, 'reg_lambda': 0.1,  'n_estimators': 10000, 'learning_rate': 0.1, 'random_state': 77,#42 \"num_leaves\": 100} 0.4678867373414343\nnum_leaves:800 0.4620618661241774\nlearning_rate 0.05 0.4615044085912404\nreg_alpha': 0.1 #\nreg_lambda': 0.09 0.4615044085912404","metadata":{"papermill":{"duration":0.063063,"end_time":"2021-07-20T23:14:54.781548","exception":false,"start_time":"2021-07-20T23:14:54.718485","status":"completed"},"tags":[]}},{"cell_type":"markdown","source":"0.46137044709334585","metadata":{}},{"cell_type":"code","source":"params4 = {'objective':'mae','reg_alpha': 0.016468100279441976, 'random_state':77,'reg_lambda': 0.09128335764019105, 'n_estimators': 9868, 'learning_rate': 0.10528150510326864, 'num_leaves': 1250, 'feature_fraction': 0.4, 'bagging_fraction': 0.3, 'bagging_freq': 19, 'min_child_samples': 71}","metadata":{"papermill":{"duration":0.076168,"end_time":"2021-07-20T23:14:56.998605","exception":false,"start_time":"2021-07-20T23:14:56.922437","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:41:46.4063Z","iopub.execute_input":"2021-07-24T08:41:46.40828Z","iopub.status.idle":"2021-07-24T08:41:46.41577Z","shell.execute_reply.started":"2021-07-24T08:41:46.408223Z","shell.execute_reply":"2021-07-24T08:41:46.414502Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"oof4, model4, score4 = fit_lgbm(\n    x_train2, y_train2['target4'],\n    x_valid2, y_valid2['target4'],\n    params4\n)\nscore4","metadata":{"papermill":{"duration":133.125384,"end_time":"2021-07-20T23:17:10.191496","exception":false,"start_time":"2021-07-20T23:14:57.066112","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:41:46.41744Z","iopub.execute_input":"2021-07-24T08:41:46.417798Z","iopub.status.idle":"2021-07-24T08:46:18.870873Z","shell.execute_reply.started":"2021-07-24T08:41:46.417764Z","shell.execute_reply":"2021-07-24T08:46:18.869987Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<pre>params4 = {'objective':'mae','reg_alpha': 0.016468100279441976, 'random_state':77,'reg_lambda': 0.09128335764019105, 'n_estimators': 9868, 'learning_rate': 0.10528150510326864, 'num_leaves': 157, 'feature_fraction': 0.5419185713426886, 'bagging_fraction': 0.2637405128936662, 'bagging_freq': 19, 'min_child_samples': 71} 0.9897448589181667\n'num_leaves': 1250 0.9278848020973461\n]reg_alpha': 0.016468100279441976 #\nreg_lambda': 0.09128335764019105 #\nfeature_fraction': 0.4 0.9269202636946671\nbagging_fraction': 0.3 #\nbagging_freq': 19 #\nmin_child_samples': 71#\nlearning_rate': 0.10528150510326864 #","metadata":{"papermill":{"duration":0.069652,"end_time":"2021-07-20T23:17:10.33178","exception":false,"start_time":"2021-07-20T23:17:10.262128","status":"completed"},"tags":[]}},{"cell_type":"markdown","source":"0.9282364499246017","metadata":{}},{"cell_type":"code","source":"score =(score1+score2+score3+score4) / 4\nprint(f'score: {score}')","metadata":{"papermill":{"duration":0.084441,"end_time":"2021-07-20T23:17:12.631135","exception":false,"start_time":"2021-07-20T23:17:12.546694","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:46:18.874628Z","iopub.execute_input":"2021-07-24T08:46:18.876591Z","iopub.status.idle":"2021-07-24T08:46:18.884034Z","shell.execute_reply.started":"2021-07-24T08:46:18.876544Z","shell.execute_reply":"2021-07-24T08:46:18.882582Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<pre>\nscore: 0.781819183401393","metadata":{"papermill":{"duration":0.074081,"end_time":"2021-07-20T23:17:12.779414","exception":false,"start_time":"2021-07-20T23:17:12.705333","status":"completed"},"tags":[]}},{"cell_type":"code","source":"score =(score2+score3+score4) / 3\nprint(f'score: {score}')","metadata":{"execution":{"iopub.status.busy":"2021-07-24T08:46:18.885647Z","iopub.execute_input":"2021-07-24T08:46:18.886013Z","iopub.status.idle":"2021-07-24T08:46:18.899546Z","shell.execute_reply.started":"2021-07-24T08:46:18.885981Z","shell.execute_reply":"2021-07-24T08:46:18.898833Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"score: 0.8488102113050756","metadata":{}},{"cell_type":"markdown","source":"model12 = XGBRegressor(**xgb_params)\nmodel12.fit(x_train2, y_train2['target2'], eval_set = [(x_valid2, y_valid2['target2'])], early_stopping_rounds = 50)\n\ny_pred = model12.predict(x_valid2)\n\nscore12 = mean_absolute_error(y_valid2['target2'], y_pred)\nscore12","metadata":{}},{"cell_type":"markdown","source":"1.5158150124891132","metadata":{}},{"cell_type":"code","source":"import sys\nprint(\"{}{:>25}{}{:>10}{}\".format('|','Variable Name','|','memory','|'))\nfor var_name in dir():\n    if not var_name.startswith(\"_\") and sys.getsizeof(eval(var_name)) > 10000:\n        print(\"{}{:>25}{}{:>10}{}\".format('|',var_name,'|',sys.getsizeof(eval(var_name)),'|'))","metadata":{"execution":{"iopub.status.busy":"2021-07-24T08:46:18.900825Z","iopub.execute_input":"2021-07-24T08:46:18.90138Z","iopub.status.idle":"2021-07-24T08:46:24.554677Z","shell.execute_reply.started":"2021-07-24T08:46:18.901344Z","shell.execute_reply":"2021-07-24T08:46:24.553657Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"model13 = XGBRegressor(**xgb_params)\nmodel13.fit(x_train2, y_train2['target3'], eval_set = [(x_valid2, y_valid2['target3'])], early_stopping_rounds = 50)\n\ny_pred = model13.predict(x_valid2)\n\nscore13 = mean_absolute_error(y_valid2['target3'], y_pred)\nscore13","metadata":{}},{"cell_type":"markdown","source":"0.8326967548738722","metadata":{}},{"cell_type":"markdown","source":"model14 = XGBRegressor(**xgb_params)\nmodel14.fit(x_train2, y_train2['target4'], eval_set = [(x_valid2, y_valid2['target4'])], early_stopping_rounds = 50)\n\ny_pred = model14.predict(x_valid2)\n\nscore14 = mean_absolute_error(y_valid2['target4'], y_pred)\nscore14","metadata":{}},{"cell_type":"markdown","source":"1.2992890471976417","metadata":{}},{"cell_type":"markdown","source":"train_X = train[feature_cols]\ntrain_y = train[['target1', 'target2', 'target3', 'target4']]\n\n_index = (train['date'] < 20210426)#0426 \nx_train1 = train_X.loc[_index].reset_index(drop=True)\ny_train1 = train_y.loc[_index].reset_index(drop=True)\nx_valid1 = train_X.loc[~_index].reset_index(drop=True)\ny_valid1 = train_y.loc[~_index].reset_index(drop=True)","metadata":{}},{"cell_type":"code","source":"","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Inference","metadata":{"_cell_guid":"25cf48a6-84c3-4d5d-a312-1a47c050722e","_uuid":"372a2718-6f2d-4128-97cb-fefd653e1f40","papermill":{"duration":0.074396,"end_time":"2021-07-20T23:17:12.928352","exception":false,"start_time":"2021-07-20T23:17:12.853956","status":"completed"},"tags":[]}},{"cell_type":"code","source":"players_cols = ['playerId', 'primaryPositionName']\nrosters_cols = ['playerId', 'teamId', 'status']\nscores_cols = ['playerId', 'battingOrder', 'gamesPlayedBatting', 'flyOuts',\n       'groundOuts', 'runsScored', 'doubles', 'triples', 'homeRuns',\n       'strikeOuts', 'baseOnBalls', 'intentionalWalks', 'hits', 'hitByPitch',\n       'atBats', 'caughtStealing', 'stolenBases', 'groundIntoDoublePlay',\n       'groundIntoTriplePlay', 'plateAppearances', 'totalBases', 'rbi',\n       'leftOnBase', 'sacBunts', 'sacFlies', 'catchersInterference',\n       'pickoffs', 'gamesPlayedPitching', 'gamesStartedPitching',\n       'completeGamesPitching', 'shutoutsPitching', 'winsPitching',\n       'lossesPitching', 'flyOutsPitching', 'airOutsPitching',###\n       'groundOutsPitching', 'runsPitching', 'doublesPitching',\n       'triplesPitching', 'homeRunsPitching', 'strikeOutsPitching',\n       'baseOnBallsPitching', 'intentionalWalksPitching', 'hitsPitching',\n       'hitByPitchPitching', 'atBatsPitching', 'caughtStealingPitching',\n       'stolenBasesPitching', 'inningsPitched', 'saveOpportunities',\n       'earnedRuns', 'battersFaced', 'outsPitching', 'pitchesThrown', 'balls',\n       'strikes', 'hitBatsmen', 'balks', 'wildPitches', 'pickoffsPitching',\n       'rbiPitching', 'gamesFinishedPitching', 'inheritedRunners',\n       'inheritedRunnersScored', 'catchersInterferencePitching',\n       'sacBuntsPitching', 'sacFliesPitching', 'saves', 'holds', 'blownSaves',\n       'assists', 'putOuts', 'errors', 'chances']\n\nnull = np.nan\ntrue = True\nfalse = False","metadata":{"_cell_guid":"08cf9a6e-1f2c-4acd-b714-703d641d996a","_uuid":"7828a113-8dc7-4691-b2a6-b232ca0525dc","collapsed":false,"jupyter":{"outputs_hidden":false},"papermill":{"duration":0.084947,"end_time":"2021-07-20T23:17:13.087985","exception":false,"start_time":"2021-07-20T23:17:13.003038","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:46:24.557706Z","iopub.execute_input":"2021-07-24T08:46:24.558048Z","iopub.status.idle":"2021-07-24T08:46:24.567134Z","shell.execute_reply.started":"2021-07-24T08:46:24.558011Z","shell.execute_reply":"2021-07-24T08:46:24.565862Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"del train\ndel train_X\ndel train_y\ndel x_train1","metadata":{"papermill":{"duration":0.098665,"end_time":"2021-07-20T23:17:13.262366","exception":false,"start_time":"2021-07-20T23:17:13.163701","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:46:24.568807Z","iopub.execute_input":"2021-07-24T08:46:24.569153Z","iopub.status.idle":"2021-07-24T08:46:24.807775Z","shell.execute_reply.started":"2021-07-24T08:46:24.569121Z","shell.execute_reply":"2021-07-24T08:46:24.80665Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#メモリ使用状況\nimport sys\nprint(\"{}{:>25}{}{:>10}{}\".format('|','Variable Name','|','memory','|'))\nfor var_name in dir():\n    if not var_name.startswith(\"_\") and sys.getsizeof(eval(var_name)) > 10000:\n        print(\"{}{:>25}{}{:>10}{}\".format('|',var_name,'|',sys.getsizeof(eval(var_name)),'|'))","metadata":{"papermill":{"duration":2.846472,"end_time":"2021-07-20T23:17:16.183708","exception":false,"start_time":"2021-07-20T23:17:13.337236","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:46:24.809422Z","iopub.execute_input":"2021-07-24T08:46:24.809898Z","iopub.status.idle":"2021-07-24T08:46:28.481438Z","shell.execute_reply.started":"2021-07-24T08:46:24.80983Z","shell.execute_reply":"2021-07-24T08:46:28.480148Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import pandas as pd\nimport numpy as np\nfrom datetime import timedelta\nfrom tqdm import tqdm\nimport gc\nfrom functools import reduce\nfrom sklearn.model_selection import StratifiedKFold\n\nROOT_DIR = \"../input/mlb-player-digital-engagement-forecasting\"\n\n#=======================#\ndef flatten(df, col):\n    du = (df.pivot(index=\"playerId\", columns=\"EvalDate\", \n               values=col).add_prefix(f\"{col}_\").\n      rename_axis(None, axis=1).reset_index())\n    return du\n#============================#\ndef reducer(left, right):\n    return left.merge(right, on=\"playerId\")\n#========================\n\nTGTCOLS = [\"target1\",\"target2\",\"target3\",\"target4\"]\ndef train_lag(df, lag=1):\n    dp = df[[\"playerId\",\"EvalDate\"]+TGTCOLS].copy()\n    dp[\"EvalDate\"]  =dp[\"EvalDate\"] + timedelta(days=lag) \n    df = df.merge(dp, on=[\"playerId\", \"EvalDate\"], suffixes=[\"\",f\"_{lag}\"], how=\"left\")\n    return df\n#=================================\ndef test_lag(sub):\n    sub[\"playerId\"] = sub[\"date_playerId\"].apply(lambda s: int(  s.split(\"_\")[1]  ) )\n    assert sub.date.nunique() == 1\n    dte = sub[\"date\"].unique()[0]\n    \n    eval_dt = pd.to_datetime(dte, format=\"%Y%m%d\")\n    dtes = [eval_dt + timedelta(days=-k) for k in LAGS]\n    mp_dtes = {eval_dt + timedelta(days=-k):k for k in LAGS}\n    \n    sl = LAST.loc[LAST.EvalDate.between(dtes[-1], dtes[0]), [\"EvalDate\",\"playerId\"]+TGTCOLS].copy()\n    sl[\"EvalDate\"] = sl[\"EvalDate\"].map(mp_dtes)\n    du = [flatten(sl, col) for col in TGTCOLS]\n    du = reduce(reducer, du)\n    return du, eval_dt\n    #\n#===============\n\ntr = pd.read_csv(\"../input/mlbdata/target.csv\")\n#tr = tr.drop(['Unnamed: 0'],axis=1)\nprint(tr.shape)\ngc.collect()\n\ntr[\"EvalDate\"] = pd.to_datetime(tr[\"EvalDate\"])\ntr[\"EvalDate\"] = tr[\"EvalDate\"] + timedelta(days=-1)\ntr[\"EvalYear\"] = tr[\"EvalDate\"].dt.year\n\nMED_DF = tr.groupby([\"playerId\",\"EvalYear\"])[TGTCOLS].median().reset_index()\nMEDCOLS = [\"tgt1_med\",\"tgt2_med\", \"tgt3_med\", \"tgt4_med\"]\nMED_DF.columns = [\"playerId\",\"EvalYear\"] + MEDCOLS\n\nLAGS = list(range(1,21))\nFECOLS = [f\"{col}_{lag}\" for lag in reversed(LAGS) for col in TGTCOLS]\n\nfor lag in tqdm(LAGS):\n    tr = train_lag(tr, lag=lag)\n    gc.collect()\n#===========\ntr = tr.sort_values(by=[\"playerId\", \"EvalDate\"])\nprint(tr.shape)\ntr = tr.dropna()\nprint(tr.shape)\ntr = tr.merge(MED_DF, on=[\"playerId\",\"EvalYear\"])\ngc.collect()\n\nX = tr[FECOLS+MEDCOLS].values\ny = tr[TGTCOLS].values\ncl = tr[\"playerId\"].values\n\nNFOLDS = 8\nskf = StratifiedKFold(n_splits=NFOLDS)\nfolds = skf.split(X, cl)\nfolds = list(folds)\n\nimport tensorflow as tf\nimport tensorflow.keras.layers as L\nimport tensorflow.keras.models as M\nfrom sklearn.metrics import mean_absolute_error, mean_squared_error\nfrom tensorflow.keras.callbacks import ModelCheckpoint, ReduceLROnPlateau, EarlyStopping\n\ntf.random.set_seed(777)\n\ndef make_model(n_in):\n    inp = L.Input(name=\"inputs\", shape=(n_in,))\n    x = L.Dense(50, activation=\"relu\", name=\"d1\")(inp)\n    x = L.Dense(50, activation=\"relu\", name=\"d2\")(x)\n    preds = L.Dense(4, activation=\"linear\", name=\"preds\")(x)\n    \n    model = M.Model(inp, preds, name=\"ANN\")\n    model.compile(loss=\"mean_absolute_error\", optimizer=\"adam\")\n    return model\n\nnet = make_model(X.shape[1])\nprint(net.summary())\n\noof = np.zeros(y.shape)\nnets = []\nfor idx in range(NFOLDS):\n    print(\"FOLD:\", idx)\n    tr_idx, val_idx = folds[idx]\n    ckpt = ModelCheckpoint(f\"w{idx}.h5\", monitor='val_loss', verbose=1, save_best_only=True,mode='min')\n    reduce_lr = ReduceLROnPlateau(monitor='val_loss', factor=0.2,patience=3, min_lr=0.0005)\n    es = EarlyStopping(monitor='val_loss', patience=6)\n    reg = make_model(X.shape[1])\n    reg.fit(X[tr_idx], y[tr_idx], epochs=12, batch_size=35_000, validation_data=(X[val_idx], y[val_idx]),\n            verbose=1, callbacks=[ckpt, reduce_lr, es])\n    reg.load_weights(f\"w{idx}.h5\")\n    oof[val_idx] = reg.predict(X[val_idx], batch_size=50_000, verbose=1)\n    nets.append(reg)\n    gc.collect()\n    #\n#\n\nmae = mean_absolute_error(y, oof)\nmse = mean_squared_error(y, oof, squared=False)\nprint(\"mae:\", mae)\nprint(\"mse:\", mse)\n\n# Historical information to use in prediction time\nbound_dt = pd.to_datetime(\"2021-01-01\")\nLAST = tr.loc[tr.EvalDate>bound_dt].copy()\n\nLAST_MED_DF = MED_DF.loc[MED_DF.EvalYear==2021].copy()\nLAST_MED_DF.drop(\"EvalYear\", axis=1, inplace=True)\ndel tr\n\n#\"\"\"\nimport mlb\nFE = []; SUB = [];","metadata":{"papermill":{"duration":351.445201,"end_time":"2021-07-20T23:23:07.704085","exception":false,"start_time":"2021-07-20T23:17:16.258884","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:46:28.483007Z","iopub.execute_input":"2021-07-24T08:46:28.483316Z","iopub.status.idle":"2021-07-24T08:53:12.185252Z","shell.execute_reply.started":"2021-07-24T08:46:28.483284Z","shell.execute_reply":"2021-07-24T08:53:12.184039Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import copy\n\nenv = mlb.make_env() # initialize the environment\niter_test = env.iter_test() # iterator which loops over each date in test set\n\nfor (test_df, sample_prediction_df) in iter_test: # make predictions here\n    \n    sub = copy.deepcopy(sample_prediction_df.reset_index())\n    sample_prediction_df = copy.deepcopy(sample_prediction_df.reset_index(drop=True))\n    \n    # LGBM summit\n    # creat dataset\n    sample_prediction_df['playerId'] = sample_prediction_df['date_playerId']\\\n                                        .map(lambda x: int(x.split('_')[1]))\n    # Dealing with missing values\n    if test_df['rosters'].iloc[0] == test_df['rosters'].iloc[0]:\n        test_rosters = pd.DataFrame(eval(test_df['rosters'].iloc[0]))\n    else:\n        test_rosters = pd.DataFrame({'playerId': sample_prediction_df['playerId']})\n        for col in rosters.columns:\n            if col == 'playerId': continue\n            test_rosters[col] = np.nan\n            \n    if test_df['playerBoxScores'].iloc[0] == test_df['playerBoxScores'].iloc[0]:\n        test_scores = pd.DataFrame(eval(test_df['playerBoxScores'].iloc[0]))\n    else:\n        test_scores = pd.DataFrame({'playerId': sample_prediction_df['playerId']})\n        for col in scores.columns:\n            if col == 'playerId': continue\n            test_scores[col] = np.nan\n            \n    test_scores = test_scores.groupby('playerId').sum().reset_index()\n    test = sample_prediction_df[['playerId']].copy()\n    test = test.merge(players[players_cols], on='playerId', how='left')\n    test = test.merge(test_rosters[rosters_cols], on='playerId', how='left')\n    test = test.merge(test_scores[scores_cols], on='playerId', how='left')\n    test = test.merge(player_target_stats, how='inner', left_on=[\"playerId\"],right_on=[\"playerId\"])\n    \n\n    test['label_playerId'] = test['playerId'].map(player2num)\n    test['label_primaryPositionName'] = test['primaryPositionName'].map(position2num)\n    test['label_teamId'] = test['teamId'].map(teamid2num)\n    test['label_status'] = test['status'].map(status2num)\n    \n    test_X = test[feature_cols]\n    \n    # predict アンセンブルチューニング\n    #x = 0.9\n    #y = 0.1\n    pred1 = model1.predict(test_X)\n    #pred11 = model11.predict(test_X)\n    #pred1 = pred1*x + pred11*y\n    test['target1'] = np.clip(pred1,0,100)\n    \n    test_X = test[feature_cols2]\n\n    pred2 = model2.predict(test_X)\n    pred3 = model3.predict(test_X)\n    pred4 = model4.predict(test_X)\n    #pred12 = model12.predict(test_X)\n    #pred13 = model13.predict(test_X)\n    #pred14 = model14.predict(test_X)\n    \n    #pred2 = pred2*x + pred12*y\n    #pred3 = pred3*x + pred13*y\n    #pred4 = pred4*x + pred14*y\n    \n    # merge submission\n    sample_prediction_df['target1'] = np.clip(pred1, 0, 100)\n    sample_prediction_df['target2'] = np.clip(pred2, 0, 100)\n    sample_prediction_df['target3'] = np.clip(pred3, 0, 100)\n    sample_prediction_df['target4'] = np.clip(pred4, 0, 100)\n    sample_prediction_df = sample_prediction_df.fillna(0.)\n    del sample_prediction_df['playerId']\n    # TF summit\n    # Features computation at Evaluation Date\n    sub_fe, eval_dt = test_lag(sub)\n    sub_fe = sub_fe.merge(LAST_MED_DF, on=\"playerId\", how=\"left\")\n    sub_fe = sub_fe.fillna(0.)\n    \n    _preds = 0.\n    for reg in nets:\n        _preds += reg.predict(sub_fe[FECOLS + MEDCOLS]) / NFOLDS\n    sub_fe[TGTCOLS] = np.clip(_preds, 0, 100)\n    sub.drop([\"date\"]+TGTCOLS, axis=1, inplace=True)\n    sub = sub.merge(sub_fe[[\"playerId\"]+TGTCOLS], on=\"playerId\", how=\"left\")\n    sub.drop(\"playerId\", axis=1, inplace=True)\n    sub = sub.fillna(0.)\n    \n    # Blending\n    blend = pd.concat(\n        [sub[['date_playerId']],\n        #(0.36*sub.drop('date_playerId', axis=1) + 0.64*sample_prediction_df.drop('date_playerId', axis=1))], @\n        (0.361*sub.drop('date_playerId', axis=1) + 0.639*sample_prediction_df.drop('date_playerId', axis=1))], \n        axis=1\n    )\n    env.predict(blend)\n    \n    # Update Available information\n    sub_fe[\"EvalDate\"] = eval_dt\n    #sub_fe.drop(MEDCOLS, axis=1, inplace=True)\n    LAST = LAST.append(sub_fe)\n    LAST = LAST.drop_duplicates(subset=[\"EvalDate\",\"playerId\"], keep=\"last\")","metadata":{"_cell_guid":"c3148c3f-68b2-46f6-b8d6-3284e68b507a","_uuid":"bc76106c-5572-4d75-ae68-28381f7406b7","collapsed":false,"jupyter":{"outputs_hidden":false},"papermill":{"duration":11.752062,"end_time":"2021-07-20T23:23:20.541664","exception":false,"start_time":"2021-07-20T23:23:08.789602","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:53:12.186918Z","iopub.execute_input":"2021-07-24T08:53:12.187381Z","iopub.status.idle":"2021-07-24T08:53:24.173457Z","shell.execute_reply.started":"2021-07-24T08:53:12.187343Z","shell.execute_reply":"2021-07-24T08:53:24.172211Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### これが提出物","metadata":{"papermill":{"duration":1.084606,"end_time":"2021-07-20T23:23:22.742916","exception":false,"start_time":"2021-07-20T23:23:21.65831","status":"completed"},"tags":[]}},{"cell_type":"code","source":"blend","metadata":{"_cell_guid":"cd1f93f4-37e0-47d0-b05a-0df03d07ce15","_uuid":"9267b421-3cbb-4b28-a16e-e62ba8da58e4","collapsed":false,"jupyter":{"outputs_hidden":false},"papermill":{"duration":1.091515,"end_time":"2021-07-20T23:23:24.894204","exception":false,"start_time":"2021-07-20T23:23:23.802689","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:53:24.175179Z","iopub.execute_input":"2021-07-24T08:53:24.175616Z","iopub.status.idle":"2021-07-24T08:53:24.200702Z","shell.execute_reply.started":"2021-07-24T08:53:24.175566Z","shell.execute_reply":"2021-07-24T08:53:24.199808Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### お詫び。スコア1.33333にするには、このコードを少しいじらないとできないようになっています。\n### 過去の私のコードを参照すれば解けるはずです。","metadata":{}},{"cell_type":"code","source":"sample_prediction_df","metadata":{"_cell_guid":"e94e45d6-2620-412b-bd5c-cb773241c8fe","_uuid":"99a89e72-692e-4aff-874b-a9e1d44ddaee","collapsed":false,"jupyter":{"outputs_hidden":false},"papermill":{"duration":1.196691,"end_time":"2021-07-20T23:23:27.18974","exception":false,"start_time":"2021-07-20T23:23:25.993049","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-07-24T08:53:24.201975Z","iopub.execute_input":"2021-07-24T08:53:24.202271Z","iopub.status.idle":"2021-07-24T08:53:24.221063Z","shell.execute_reply.started":"2021-07-24T08:53:24.202243Z","shell.execute_reply":"2021-07-24T08:53:24.220032Z"},"trusted":true},"execution_count":null,"outputs":[]}]}