[Tutorial] Similarity.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# String-to-String Similarity Problems and Algorithms"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABugAAAGqCAYAAADp+8LoAACAAElEQVR4XuzdCXhTZb7H8dd9AbdxHZo6znbnjs7OnaF1RZukLUtXyr7viooKKK5UhaYIIi1dUpBFUEbqLjjusiguiIKiooCDQMuiqAgoO+e+b5La9M1J26RJ0yTfz/P8npTk5Jz23+bEOb85J0IAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAiEv5+cfqdwEAAAAAAAAAAAAIgQSb4z6ZQxabw6hN4a5L8vJP1JcFAAAAAAAAAAAA0EQWW+HBuuWcK0dEXuVx+rIAAAAAAAAAAAAAmsiknHNFXw4AAAAAAAAAAABACOjFHAUdAAAAAAAAAAAAEEZ6MUdBBwAAAAAAAAAAAISRXsxR0AEAAAAAAAAAAABhpBdzFHQAAAAAAAAAAABAGOnFHAUdAAAAAAAAAAAAEEZ6MUdBBwAAAAAAAAAAAISRXsxR0AEAAAAAAAAAAABhpBdzFHQAAAAAAAAAAABAGOnFHAUdAAAAAAAAAAAAEEZ6MUdBBwAAAAAAAAAAAISRXsxR0AEAAAAAAAAAAABhpBdzFHQAAAAAAAAAAABAGOnFHAUdAAAAAAAAAAAAEEZ6MUdBBwAAAAAAAAAAAISRXsxR0AEAAAAAAAAAAABhpBdzFHQAAAAAAAAAAABAGOnFHAUdAAAAAAAAAAAAEEZ6MUdBBwAAAAAAAAAAAISRXsxR0AEAAAAAAAAAAABhpBdzFHQAAAAAAAAAAABAGOnFHAUdAAAAAAAAAAAAEEZ6MUdBBwAAAAAAAAAAAISRXsxR0AEAAAAAAABA82ovs9iTjLoPiTSvx36tPYbwyBG1M79ce6yn5/43ZM7UHkPgCoR7nq/LnKo9NsXz2HTtfoROuXDPeJHMMdpjc4T7sYna/QgRvZijoAMAAAAAAACA5qWKngMy6sDs09pj8z33fyV8D6AjPH4r3DNXKdUeU8Wcuv897X4EZ7ionXWW1/2tZfZ57h/vdT9Ca5yonX+y1/2JMkc991/vdT9CSC/mKOgAAAAAAAAAoPmpM1jUgdm9Mqd47jtRZpfn/gc996F5rBTuuVeL2mL0bJlDnvtHe+5D05wnc1i4ZzrH6/48z30qf/W6H6F1saids/eZcqqUU/cdkfml1/0IIb2Yo6ADAAAAAAAAgObXT9QeKM/03JfqdV+S5z40j9tE7ezbee6r+R2pM4su8tyHplOXt1Rz3SlznOe+xzz3fVGzEMLmE+E769c89y3zug8hphdzFHQAAAAAAAAA0Py8L3M523Of+nwo9e9NgstbNjd1mcuaS/ypz0lTnvH8e0XNQggJ78tcXi1zgsz3nn9zecvwGydq5/9HmV+I2jNFb/BaDiGmF3MUdAAAAAAAAAAQGTWXuVSXVVQ2ev7N5S0jo+Yyl+pWlUa7Pf8e470Qmsz7MpcOmSs9X6twecvw877M5UhRe3lRdXnLNl7LIcT0Yo6CDgAAAAAAAAAio5+oPVDO5S0jr+Yyl6o86uD5mstbhkfNJRXflsn3fM3lLZtPzWUu1VmiNWfucnnLMNOLOQo6AAAAAAAAAIgMdZnL/cJ9cFxdRlHdcnnLyPmNqL3MZc3vg8tbhscw4Z7vQZlVnq+5vGXzGSfcM1efA/il52subxlmejFHQQcAAAAAAAAAkbNQ1J45p2J2eUt12b/ZntypPYbQqrnMZU30y1seLzNCuC9PqnKdzHF1lkBjqMtc1nzuWU3+VmcJ979LZV6XmSVzed2H0QTel7lUqe/ylhnCve+Zqj+AwOjFHAUdAAAAAAAAAEROX1H3QHly3YfF6TLrRO3noS2v+zBCrOYylyrqbLpf133YVVSox76X+cnz9ZQ6S6Cxai5zqaL+xr2p14E6u07NWJ1hpy47qnK190JokjWidv7+Lm95tszXwr3MNu0xBEgv5ijoAAAAAAAAACByzhe1B8nNLm9ZKdxnGnUTFHTNQZ2tWPP70C9veYrMKzJvyJwo80fhXk5dJhCBU2cn1sx6gvaYOiuxQCbB82915qha7oGfl0BTTRa18/d3ecuZMv+V+UZQ0DWZXsxR0AEAAAAAAABA8/uFTD/hLiZqDpJPqrOEuwRaIHOrcF9+joIufFQRpH4fJaL29zGqzhK1jvXcqueo5VSxisb7p3DPeomonfWfvRfwoj6nUZWmzwn3cupyi2iaK4V7/h8K90zV/wHA7PKWajl1FmknmY2Cgq7J9GKOgg4AAAAAAAAAmt/fRW05obJH5ld1lqiLgi680kXd38cO4S6H/DlB5hnhXvZ67THUT10S1HvW6rP8/LlfuJdRn5GmLj+Kppsv6s5fnSWnU//ngE+FuxhVNgoKuibTizkKOgAAAAAAAABofhcL90FvdRD8eeH72XM6Crrwai/cv4+PZJ4U/s/oUtTnAqpLXarfh37WIxp2l3DP+j2ZUpkz6j5chyqyRwj35UZVSde17sMIQpFwz/9t4f77Pbnuwy63y/woc5Hn3xR0IaAXcxR0AAAAAAAAANDyUdC1DBfJfCKzT2Zg3YcQYld5fZ0i3H//6qxFhN/Tou5Zdt5BkPRijoIOAAAAAAAAAFqu/xXuz4u6SbgPjq/3/PtS74XQbLYK9+9BnfU4zivneC+EJlNn2ak5q8th5sm87PVvhN/Vwr2fqck3Mrs8XyNIejFHQQcAAAAAAAAALZf6fDN1AFfPw94LodkcFr6/C5U/ei+EJkuQWSbqzvhFUf/nAiJ8NgoucdlkejFHQQcAAAAAAAAALddpwn1ZRT1nex5H8/qV8P1dqJzgeRyh9Vvh/nxAi3Y/mpeaf6J+JwKjF3MUdAAAAAAAAAAAAEAY6cUcBR0AAAAAAAAAAAAQRnoxR0EHAAAAAAAAAAAAhJFezFHQAQAAAAAAAAAAAGGkF3MUdAAAAAAAAEAca5VX+rdWOSU3ndOjdMF5PUo+Pqd7ybbTukw70Cq3xCCElBi/6Fby/bk9Sqt+2bvk/XN7lLwg73ugdZfSHPnauUB/PQGAP3oxR0EHAAAAAAAAxJnWecUXn9aldPIZXUt2JPQtqe54Z9nK+2eUvf7MC+VPfvhu+dw9X5Y7jWpnKSHEWfrlauesFcvLH/33c+XPFj9a9uLQSWVv/v3aktXy9bPn9C4l77bqMm30aT0qztFfZwDgTS/mKOgAAAAAAACAOKHOljuza+nzv+hW8l2f8aXL33mz/DG9jCCENC77NznL5z1T/uS/bihd0rpL6a7WOSUzTskptuivOwBQ9GKOgg4AAAAAAACIcWflVZzROrek7KyuJd/eWlL+GmfIERLabPmkfEba2NKnW3cp+fbUnNJxon3+8frrEEB804s5CjoAAAAAAAAghp2aO+3/zsgr2ZR6W+m7Wz8re1gvFgghocsnK5yllr4lS1vllqw4ObfsN/rrEUD80os5CjoAAAAAAAAgRrXuUtL9jK7Tvp06r+w5vUgghIQrFSVd7imf3qpLSfWpOaX/1F+XAOKTXsxR0AEAAAAAAAAxqHVuybVndyvZvmQJnzNHSCQyzlk+tVVuyQ5KOgCKXsxR0AEAAAAAAAAxRp05p8q51e+Vz9VLA0JI8yW/oryoVe60rVzuEoBezFHQAQAAAAAAADFEna1zRteSnZw5R0jLSPf80pnqM+lE+/zj9dcrgPihF3MUdAAAAAAAAECMOCuv4owz8kq+eujR0mf1kqC+HKkqL332kbEvDB9x3VcdMrvts6VnHk5JSTHsHXMOd87rt/fm22774pOlkyj8whQ1/4WP3vGfu+8Yualn7577s7KyXfPP7dLl8OAhg/ZOeWDsZ+uWP/Co/jwSmqj5Pz//rhfvvuvmLb7zH7hHzv/T9e8UztOfF0gu7Ffy5qk5peP01yxcjuk1YEC3gcNGvN6lW49dnTOzDqj5Z+bk7uveq+/mAUOGO6+55ppf6U9CyBzTb8jwXkOvu3FpXreeP2RkZR9U88/O6fJT7/6D1l0/6raxeXl5x+lPQuD0Yo6CDgAAAAAAAIgRp+ZMK7XfWvq2Xg7Ulydm3vJi58ysvbas/oev6l9qXHHjq8blo1YYl4/93LjilneMK6971mjf414jJTXjaP9hI6q2rX5wpr4OEnyenDXqpa5dc38cMHjYkanT5xmL3vrIWLZms7Fq8x5j8eqNRuVLbxl3T5hsZGRlHb3j9hs2M//QpnLm6Je6duvyU//Bw47WN//MrKwjd9x23cZg5//pBxXO1rklO0/JKbbor9t4lpWV1SMzO2dn/8HD6/3775yReahrj97PtG/f/hx9HQhet549+2Zld/lOzt/n7/+NVV8as5980Rhy7Q1Gr74DfpjifCRTfz4CoxdzFHQAAAAAAABADDg1t/jvZ3Ut2bn1s7KH9XLALPs3lpaPuqHn5+kZeQeuHLHIVcjVm9EfGu17TjBSO2YdenPR/U/o6yOBRc1/9I29v+jeo8fBp15913VAvL4sX7vNuHdikZGbm3PwvZfvr9TXRwKLmv+Ykb3XdevR41Ag88/JyT6w4qX8Bfr6GpNOt5ctbJ1TMkN/7cajtm3bntCxc8aCvG499gUw/6MdO3b6zmaz/U1fHwKj5p+ZnfN01+49DzRm/mWPPGFkZecenfrw/Hv1daHx9GKOgg4AAAAAAACIAafnlTw/urjsZb0UMMvhzeVlNwzruiGz57V7rrh1jW8ZV0+uGrbAsKd3Prz02buf1NdLGhf3/LttGHnLLT+t2LDT52B4fZn7zKtGRkbnQ8uYf9BR879+ePcvR44avT/Y+X/w8r2P6+ttKFs+KXv4tC4lu07LLjlbf/3GE3W5xNTUtGdH3HTrnmDmn5beYbfVav2zvl40jpp/h46dF10/auy+QOb/+H+WGZlZOcZDFY/eqa8TjaMXcxR0AAAAAAAAQJRrnVd8sTp7bveXFU69FDDL1IKhb+d26/39Fbd96lPANSZXXvu0kdah08HP33I06XO54jVq/kOGDNy9cuP3PgfBG5PHX1hqZGZ2PsD8g4tr/kMH7w12/nOeesnI65K1d8+6aRX6uhvKFTeVvt2qy7TR+ms4nlit1jF9Bw75Juj5P/ni4bT0DluSk5NP0deNhtlstlsHDBm+K5j5q32POpNuYokzRV8vGqYXcxR0AAAAAAAAQJRrlVs6qef9pUv1MsAsm1dOnt2xQ+o+25hl+/XiLZBYB5Uf7tMze8eRqvIyfRvEf9T8O3dK3//GB+sP6we/A8m0mfMO9O+VvZ35BxY1/05y/q9/+OVRfaaBZOSoMQdmPHT9En39DeXRp8ufPbNryUr9NRwvUlNTf2m3p+6S8/eZaSAZfsPNO61W61B9/aifmn9aWvrupsy/ZNbjRq++/Xfm5+cfq68f9dOLOQo6AAAAAAAAIJrl5x97Rtdp299+s/wxvQwwS7Fj2Ft9Bo3YphdugWetkdWl5+7Kmbe8pG+D+I+a/9g7Rn+jH/QONB9u2m0M6N9zF/MPLGr+t995+/f6PAPN/EVLjKFD+nyjr7+h7N/kLD+za+nuVnmlF+gv5XiQkpJyy42j7/hKn2egUfPvnJn1nr5+1E/N/5ax43bo8wwkat8zaNgIY8Dg4dfq60f99GKOgg4AAAAAAACIYq3ySv/2y94lW/QiwF8G9M6q7jrm0W98C7fAkzVyxvfDB+V+pW+DOEv3fVlierlRNf/KRS/9oB/0DiZz5j+6g/mbx//8s6sr//PGfn2WgWbJR18ZWdlZh/T1Nybtri/9rHXutFz9tRxL0tPTT9LvU2w225uP/2dZSObfsVOn3fr64eZv/mlp6e8sePHNJp09qjL7if8Y3Xr0WaOvH/XTizkKOgAAAAAAACCKtcopucl+W+k7egngL1kZ6Xtsty4/oJdtwaT9qOWHc7I67Na3QZyldpvtSEbntL03X9ftC+fk65a+veieSlUayfnvXbx6o88B72CyZNXavczfPDY5/6yMtD03j+j2eZ35Z3b8MRTzf//Lb420tLSj+nYbk2snlb57Xs/SKfprOZakpKQcktksM0fmuquvvrqtKo3s9tTtoZp/amraYX27cPPM/6trrrlmlvf809I7fBOK+S9e9V8jMztnj75d1E8v5ijoAAAAAAAAgCh2epdp88Y/XPqaXgL4i91mPXzF2E98yragctsnhs1mPZKSkmIQvzmalmo7qL5OT7MfVPNasWGnzwHvYPLBxp1HmH/9sdlSjsq5H1Bfu+ZvtR4Nxfw//OoHw2q1+myP+EQVRbs8X++WMzscqvnL9R012R7xipy3+tv/3vNvNf+Q7H9WbvzesNtTj+jvR6ifXsxR0AEAAAAAAABR7OxupR8+/ULZU3oR5y+ZGel7rxz1rm/ZFkSuuOVdIzOjw159G8RZ6nWQ/GiH9NQDt9/S65PtHz84U53BpS7Ppx/wDiZLV288rM6I1LdNaudvtar52/ffNbrPR1+vmfJwdlann0Ix/7c+rTYysrKCOoOu8rnSpxP7ln6sv5Zjidffvyrovr3mmmsmp6am/kKdwRWq+XfOyDykbxduXvNX/+eAb+T8HWr+6R06fhuq+XfqnHFA3y7qpxdzFHQAAAAAAABAFDuzW8nXa1eWz9FLAH/p1ytrq23kMyG5xOXVI549rNanb4M4S+12m5xN5rYX/z32Oe/7+/fO3vbEK283+TOgVJ56acmPzN88av7q8+Zerhz7rPf9/fvk7nji5bd9ZhloFi5bZfTu2y+oz6Bbvqx8fps+JVX6azmWpKSk7Jd5w2azXeV9f4dOnT8M1fy79ej9g/e6Uctz5tyrMld43985M2tNqObfpWuPnd7rRsP0Yo6CDgAAAAAAAIhip+eV/Pjteud0vQTwl8n3DXo3a+Adu/WyLZhkDLhrj1qfvg3iLFVna+n3qah5jZtQ8KN+wDuYjLs//3vmb5565++YdFCfZaCZOn2ecfe4O3fq629MvlpTMfPsbqW79ddyLFFna+n3KfL+u+8pePCAPs9Ao+Y//IZb3tHXDzd/88/IyLp/XOFDR/R5Bho1/4HDRryurx/104s5CjoAAAAAAAAgirXOLT16pMq3BPCXT5eOf7RDx04Hrhiz2qdwCyTq+elyPWp9+jaI/6h5ZWVlHHx33dc+B70DiXp+ZmZn5h9g1LwyszIPNXX+A4Zeazz7+JR39PU3JoerystadymJy8/vat++/e86dc7YF4r5X3/z2OH6+lE/Nf+MzKwDTZ7/kOFHe/Uf1FNfP+qnF3MUdAAAAAAAAEAUa5VbYugFQEMZfWOftbY+4w/qpVsgsfYZf2j0yD6f6+smDWfsLf2+uH9S8WH9oHcguX/ilP233dT7M33dpOGMvWXg+vGTpwV9mdGnXn3X6N6z5+EjVeU+625s1OtWfy3Hi9yu3Z65f/I0n7k2Nmr+uXnd9shVHaOvGw3r0bvfi02df3Zu3i7B/AOmF3MUdAAAAAAAAEAUC6agq/pg8uwOHTvtu3JYpU/x1pio53Xo1Hm/Wo++btJw1NyyszP3P/rc6z4HvxsT9bzs7M77mH9wcc8/60Aw81+xYafRo1dfo3JuYZMuLRrPBV1KSsr5nTMy9wY7/+69+hy97sZRI/X1onHU/DOysn8Kdv7devY+3H/gkCH6etEwvZijoAMAAAAAAACiWDAFncrbi+6ptKd3OhRoSaeWt6WmH5nx0IjF+jpJ46Pmn5GZeTjQg+Rq+Q4dmH9Ts/yFe5/MzMo6Esj8V2783hg55g6jb5+ee1578o5n9HUGkngu6JRO2XnJnTMyDwUz/8ysnA02m+0qfZ1ovIEDh17dOTMroP2Pmv+No28/2q1nn5f09aFx9GKOgg4AAAAAAACIYsEWdCrLF95Tmd4p80D7Xg7j8jGrfMq4OpGPq+XSOmYdnDHlhiWdO6X91NSSIt6jSqLc3JyD6nJz73yxw+eAuHfU42q5nJzsA6qcY/5Nz9uL7n8qJzf3UGPm/9oHG4wh1914dOTI4V+/tOD255o6/3gv6JQ+fQZclZmduy+A+R/p3rPP2zabLUXaQUnXNENvuNmeldPlQKPnf+0Nh7rkdV/Stm3bE/R1oXH0Yo6CDgAAAAAAAIhiTSnoVKpXTZ41ZPiQr2zpmUfa97zfuHLEQuOKW95xlXLqVv1b3a8eHzxs6Ca1vHqeKieaWlIQ9/zH3nrd5sys7KPjHA8ZT77yjrF49UbXQXF1q/6t7s/Kzj5y26hhG5l/aOOa/23XV/mbf+VLbxm33+swMjIzj0x9YMxHR6rKy9Tzmjp/Cjq37gMHtuk3aNibmdk5R+qbf8dOnQ/06NX3gfz8/GPV81Q5R0nXdIMH32AZNPz69zKzcur9++/YsdP+Ll27T6iZP4KjF3MUdAAAAAAAAEAUa2pBV5PP33LMu23syE+yuvbek9oh61BKSoqhbjPzeu+R93+qHtef09SSgtRGzvfRSY4xnw0aPGBPdk6Oa/458nbQoH57JhXcsob5hzdr33zgMTn/tXL+e3+ef27OwUFD+v9QPGnUh5tXPuDzeX9NmT8FXV15eb3/MGj4iDk9evWt7pyZ9aOav7zdm9et5/ruPXtPaN++/QX6cyjpQqfXgGEXDx0x8rHucv7q8+lq5p/btfsXuXnd7jWbPwKnF3MUdAAAAAAAAEAUC1VBp0cdoNXvM0tTSgriP8w/sgn3/Cno6qfmr99nhpIuPBo7fwRGL+Yo6AAAAAAAAIAoFumCTiXYkoL4D/OPbMI9fwq6+gVSEFHShV4g80fj6cUcBR0AAAAAAAAQxVpCQacSTElB/If5Rzbhnj8FXf0CLYgo6UIr0PmjcfRijoIOAAAAAAAAiGItpaBTCbSkIP7D/CObcM+fgq5+wRRElHShE8z80TC9mKOgAwC0JBd2dJx1UZrjIn9pYy+81GJ1tK/Jr+z3/1pfxixt7OMT9W0BCL8KIc4oF+KimpQJ0Vbetq8v8jl2mX4yefpjKtOFuMp7nd6ZIsQp+vcAAAAQ81pSQacSSElB/If5Rzbhnj8FXf2CLYgo6UIj2PmjfnoxR0EHAAjG79LzT1fFl8Ve0E4VZQn2gn6J1oIB8utxrtgLHAk2x2wV+T7ztMxiTz6V921UkV8f0t+PmjFfe30fn1hqvj9r4Rs137fre5c/h/vncQx1/Yw2h91VDtoe+L36+du2rThBnw0Q7WYKcZoqu8qE+IunMMtShZlTiFvk7TiZafLr2TIvyCxWkfetl7cbPdkrY7SQ7Ha6v6cvnJ7v1en+3mfLn22C5+cZ4fn5ctTPWyrEH+XtefpcAAAAWqyWVtCpNLakIP7D/CObcM+fgq5+TSmIKOmarinzh38mByhd0ZcDAMSXX6dMOD8htfCvidbCVFVEeYq2B1wlldWx0FVe2Qs2yNtq/T2EuFLlKfzetNgKXlJzS7Q5CmRGqXkmpE642pIy4S8J6RMt+uyB5jBD/megU4i/lgtxjefstDHy34WeoupVefuhzCat3CJCbJb50DMjNSuHvB3tmeHVMn+mzAMAABHXEgs6lcaUFMR/mH9kE+75U9DVr6kFESVd0zR1/jBnckDRFX05AEBsuKh9/pl68ZZoc5RbrIXPyP3/O54zyI7q7wuRivx+ttec2WYWucwSS+3ZeCo/n5HXUOSyR/TtRSoJ7p/zM4v7Z5gnU5hoKxyRaCvIaJM64R8XpBecq/8uAX9mCHG+U4j/Kxciu0KIG+XXk2Qel/9e7ox86faNs/bMOpUPnLVnsvnLQqf7LLf5Jo/VZIOz7npros6YMyKQn2TWyrwkUyF/D3fI+feSt5eXCsEldgEAQHi11IJOpaGSgvgP849swj1/Crr6haIgoqQLXijmD1/6AcKa6MsBAKJDm875p16YWnhJgr2gk8XquD7BVvigxX1JyQ8ttsLv9f19CPODp/R61/uykBa7I1+VgAl2x+2uM8dcpWBBTs1nxqnvteZz4URe5XH6z9Nc2rSffE7N93FhSsHF3p9rV/N9u+O43X2Jy8Lp7p+x4BWLu1Rb7/n5w3mZzn0y62rma7EV3Cv/PViVrYnXPPBb/WdC7CoW4qTpQlxSJkSGU4hbZEplXna6S6oDJmVRU6PKpo0yHzvdZdgzTvfZY5PVJSHl18M9l4S0qUtCqsivf68ui6nSkj4HzutynX9Q36fnrDd19ls/+e87PT9Pifr5ZJ72/Lyfy+z0zCJUOSLzX5k3ZGarAk/edlGXEZ0txMn69w0AABCQllzQqdRXUhD/Yf6RTbjnT0FXv1AVRJR0wQnV/FGXycE/V/TlAAAtx2+shWckpjj+mWAt7Ov6bDer43GLKsZsjh36/rwJUetaXXN5RlcZZC0c6/7ctYKMms9ds1wzPkH//iBEG/v4RFX2JaYWXCFn18E1N3vhXTIPJdodj1jcpZ76rLutJrMPNofl7+VzefuCTJH827hRbbtN2sQ/RLL8RPCcQvxeJlNdflLGKb9+TWaLSdkTTHbIfOp0F1CPyDwotzFWFVXThbDK/E1dBlP/nuKdKvec7rMTUz3F3h0yUzwzVLP8ROZbk3kHGlXeqc/0mypzndzeNaVCXKB/PwAAAKZaekGn4q+kIP7D/CObcM+fgq5+oSyIKOkCF8r5o5bJwT1X9OUAAM1PfT6ZOjvKU7SUyP3z6xa7Y5u+zw4wu2Q+tdgKX3OdAWZ3jFeXUrRYHblt7IWXus5ky88/Vv9eEF6JtoltEu0T/lRzxl6i1XFnoq3AaXGXbR/LfGfyuwws7s8M/I8qCeXvfrj6fVtSp/xC/17Q/EqE+FW5EJ2c7s8ym+t0X/bxR5PSprH53uk+222RXG+ZKt7U5RXLhLhCbUvfPkLPc6ben+TsO8qMkL+LB2QWyLwrs83kd9bY/CDzvufv5BZVpj4sBK9jAABQVzQUdCpmJQXxH+Yf2YR7/hR09Qt1QURJF5hQzx9uPgfvPNGXAwCEjzoLzXVJSrtjjNwHz0+0Fn4gb/fo++ZG5oDrLCq740W5zjK1Tnnb5UK7oy1lTHQ7t31+619ZJ/7xQvtEW6LdMdB16VDX2Y2uz+GrNvlbaGy+VpfqTLQ6piRaCwaoz7y7qH0+l9gLg9lCnFwmRNsKIa5VxZnMCmfwn5GmLmf5olzXNHk7UhV8qhAqFuJ0fbtoeTyXKf0fp/uyoEPk7ST5u3zO6f7cOnX5S/333VCqZJ6X6xgvb3Nkfq9vEwAAxJFoKehU9JKC+A/zj2zCPX8KuvqFoyCipGu8cMwfFHQA0KzyKo9LSC38qypBLOoShO7LHAZzZtSPMh+6Lm1pd+Qn2Ar6XGgtvJzLTcY3S/KUU9RZeIn2gqxEm2OUTLnnM/K+MvkbaihHLO4z9+YlWAtustgKrvpdej7FTwDyhTheXSbSKcTwciFmyNvVModNipX6oi5DqT5brqRCiJtUCSdv/1etW98eYke+/P3K3/nF8nedJX/nt8qvZ8m84wz80pm75fOXy9timUGqwNW3BQAAYlQ0FXQqlESNC/OPbMI9/19m5nNQvh7hKogo6RonXPOPdyYH5FzRlwMABCg//1hVxlnsjqEJVkepxVbwvty/7tP3tw1kp8xydanDBLvjZvV5Yr+y3/9rfVNAQy7Jyz9RnX3nKu9cl890VFrcBdxBk7+7+vKpfO4ci9VxvTrTrm3bihP0bcWrMiHOkslwClEos1TmJ5PCxF9U8bJMpqRciOvl7ZVcthBmKoQ4R/6NXKYum+l1FmYgl0PdJfOSfF6+vE2T6zlD3wYAAIgB0VbQqVASNRzmH9mEe/5XWjsYlET+hbMgoqRrWDjnH89MDry5oi8HAKif+rw4i7UgJ8HmmCz3o0tl9ur7Vr+xOnbL2+UWe0Gx+mwwdTZcm/aTz9G3AYRc+/zj1Vl36jKoCdaC++Xf4dPyb/AzeXvU5+/UPD/JLEm0FUxQBfJvrIUhOdh/vn1SK/2+lmaG/DbLhejqFKJU5lOTIsRf/ivzlDozaroQ6Wo9+rqBQMm/pUs8f4/qs+5ekF9/bfK35y9r1FmeFUIM5iw7AABiRDQWdCqURPWH+Uc24Z6/OoOOksi/cBdElHT1C/f845XJgTZX9OUAAN6MY1yXqrQVjpD7zPkym/T9aD35WmaRTKEqRdqkTfyDvnYg0tp0zj/VkjrxX+pMOfm3OkPmQ0vjzrY7qj5DMcFW+KD6XMVgLot5vn3SeXI9ay7s6DhLfyyS5grRShVqFUJMcQrxiUnJYZaNMk/J3FYuRKo6y05fLxAu8m/uPPm310GVwfL2aZltJn+jZlFn2S2Uz7tZ/r3/WV8vAACIAtFa0KlQEvkP849swj1/9bqlJPKvOQoi5u9fc8w/HpkcWHNFXw4A4lp+/rGJton/Z7E7xsh95AsyP+j7TT/ZIp/zpMVaODbBWthRFQ/6qoFocVH7/JMt9oJ2FlvhSIu7mF5v8jev57B8DbxtsRXcm5hacIU6Y09fr04+Z556bqLd8Yj+WHOrEOK3TiFGlgvxqrw9YFJkeOeQ53KDD8nbbFWO6OsDIq1EiF/Jv9FuMlNl3pc5aPK3XCfqbDz5Wvi3vB2iXhP6OgEAQAsUzQWdCiWReZh/ZBPu+avXrXr9UhKZa66CiPmba675xxuTA2mu6MsBQLy5MKXg4kRbwS0Wq2OhpXGF3AGZFYn2wocSbI68Nvbxifo6gVjjOdstO9HqmGKxFbwnvz5k8tqojdWxW76uFiTYCvokpBScra+vjdVh9V5enYWnLxNuFUL8wynE/TKf60WFlr0yL8vcUy7ENeoMO31dQEs3RYhTyt2faTfK2fiz7DbK5WfK5/UqFeICfZ0AAKAFiPaCToWSyDfMP7IJ9/xrCjqFkshXcxZEzN9Xc84/nvgcOPNEXw4AYp36zDeLtbBngs0xW+4Hq/T9oh653Ha5/DOJ1oLRbWwFl6mzi/R1AvHm3Pb5rS0pjjSLrXCifJ2skDmiv3a8cthiK3hfXQ5Tfp19UYf8Cyy+Z+VVN8elLj2f3zVBFQ8mZcTPqRBipbwtkLdXVwpxor4eIBaUCfEb9Xl06qw5+fe+Q38dmGSVXHa8fB1dmi/Esfr6AABABMRCQadCSVQ3zD+yCff8vQs6hZKoruYuiJh/Xc09/3hhcrDMFX05AIhFrs+Rsxfe5b4En+Oovi/UskVmbqLdMTDxmge4xBfQCOoz6CzWgpxEm6Ncvn6+NHld1cZq/hoM16UuHxbiF04hRqpywaRwqMl2mdkVQvScJcS5+jqAeFAuxJ88r5XnZX4weZ14Z6fMPJlunFUKAEAExUpBp0JJVBvmH9mEe/56QadQEtWKREHE/GtFYv7xQD8IVhN9OQCICe3zj0+wTUjxlAWb9X1f3RR8m2B1PC6/HmxJdfxOXxWAwLVJm/gH+Zq6zWJ1LLO4zqDTX3fmCeWlLiuEaFcuxGNOIfaZlAsqm+XjRdOFuCqfs4GAOiqFOK5MiCT5GrlTvlZel9lv8hqqiXqNLZLLDqTgBgCgmcVSQadCSeQO849swj1/s4JOoSRyi1RBxPzdIjX/WKcfAKuJvhwARKvfpRefpA7uuy9dWfCtvr/zivrsrCWJVsedltSJ/xLCOEZfF4DQUWfXJdoKMuTrrtDS8Oc8NulSl/I/bI4pFyLXKcQ7JiWCykb5uEOVd/pzAfhXKkRr+frpIvOI0332nP7aqslhmTfka2woZR0AAM0g1go6FUoi5h/phHv+/go6hZIosgUR84/s/GOZyQEwV/TlACCq5Ocfm2B3XJNgK5wl92m79H2cV3ZaXJetLOz+G2vhGfpqAISfxV4wyOS16ZNgL3VZIUSeU4g1JoXBXvnYTHl7pf4cAIHLF+JY9XqSeUBmnclrriaHZF5RZ9bJ1yDvvQAAhEMsFnQq8V4SMf/IJtzzr6+gU+K9JIp0QcT8Izv/WKUf/KqJvhwARANLyoS/JFodU+R+rFrfr9Wm8L8We+GkC62Fl6siT18HgOZzvn3SeQ2c2VongVzqUp0N5zQ/Y+4T+diIYiFO158DIHTka+335ULcKm+XOt1n0OmvRRV1GczKMiEy5OvyBH0dAAAgSLFa0KnEc0nE/CObcM+/oYJOieeSqCUURMwfoaYf+KqJvhwAtFTqzLdEa+G1FlvB+/q+7OdYHZ/Lx+9tY3X8XX8+gMiRr895Pq/X+rO1oUtdzhWiVYUQ05xCHPUuAsqFWCLv76gvDyD8SoW4QBXjnrKuzmvTK9+o1+50If6mPx8AAAQolgs6lXgtiZh/ZBPu+TemoFPitSRqKQUR80comRz4ckVfDgBamjapE/7h/lw5x0/6PsyTTRZb4URKOaBlkq9Nq8nrtsHUd6nL6UJc4vS9tN57Min6sgAio1yIhAohbpK3K0xKupq8LzOcM10BAAhSrBd0KvFYEjH/yCbc829sQafEY0nUkgoi5o9Q0Q961URfDgBagrZtK05ItDl6yP3UO/p+y5M9FqtjZqK9kM+UAlo49dmP8jW7VOawyWu53phd6nK6EFanELu9DvDvlBkg/6PmGH1ZAC1DmRB/kK/T+yuE+MqkpFPZWy4Xk/mT/lwAAFCPeCjoVOKtJGL+kU245x9IQafEW0nU0goi5o9Q0A941URfDgAi6Xz7pFYJ1oKb5P5pi76/csXqWJZgL+jXpnP+qfpzAbRsbdpPPifR5ugvX8tPy+z1eX2bp86lLiuEuNwpxI9eB/XfmCZX7b0dAC1buRDt5Wt3nsxPJkWd63Utl8nOF4LPjwUAoCHxUtCpxFNJxPwjm3DPP9CCTomnkqglFkTMH01lcsDLFX05AIiEi9rnn2mxO/Llfmmnvp+yWB27LbaCkgtTCy/RnwcgOsnX/MkJ1sKO8jU+Q772t/m87r1Sc6nL2UKc6RRiS81B/HIhHssX4nht1QCihHpNez6vbrVJSafyhXydD6kU4kT9uQAAwCOeCjqVeCmJmH9kE+75B1PQKfFSErXUgoj5oyn0g1010ZcDgObkOmPOVniH3B99p++fLPaCDRar4/pz2+e31p8HILa0sRdemmgrmCBf+2t99gU296Uuy4W4z+vA/ZsVQpygrwdAdPKcHbtA5qBJUbdVPj5qihCn6M8DACDuxVtBpxIPJRHzj2zCPf9gCzolHkqillwQMX8ESz/QVRN9OQBoFnmVxyVaC6+V+6Ed+n7JYit4z2J15Kpl9KcBiH2WVMfvEq0Foy11P7euuuyYY9fXHLBXB/P15wGIfuqStfL1PV6+zr81Keq2lQtxPWfUAQDgJR4LOpVYL4mYf2QT7vk3paBTYr0kaukFEfNHMHwPgFPQAYgMi63gKovd8ZG+P5L3vX2hfaJNXx5A/FKfW3dJh/sG9+h905Ly1H+tW5b5q2cWpyVWbnzgn1dtmZLM2TRAjCoVonW5EDc7vS5r65VN8rE++nMAAIhL8VrQqcRyScT8I5twz7+pBZ0SyyVRNBREzB+B8jkQTkEHoJlZUqf8wmIveEzfD8mstKQ40vTlAcS3qqLklOqipKdl9soYJvlpa1HSC1uK2nXQnwsgNqhL2arPoXMKsdmkqHt7uhD/1J9jplIIzsoHAMSmeC7oVGK1JGL+kU245x+Kgk6J1ZIoWgoi5o9AmBwQp6AD0GwS7QVZCTbHdm0fVJVgL+gnhHGMvjyA+LXjoX/+proo6TWTQs5/itu9va243cX6ugDEBnVZS6cQN8hs10q6IzLTK4Q4Q39ODflYP6frpDwAAGJQvBd0KrFYEjH/yCbc8w9VQafEYkkUTQUR80dj6cUcBR2A5mBJnnJKgq1wlrbvOZhocxS06Zx/qr48gPhWPbWdrboo6VufAq5x2b21KDlbXyeA2KEufen5jLr9WlG3Rd7f0XtZ+e9T5f2zPY+v8X4MAICYQUHnTqyVRMw/sgn3/ENZ0CmxVhJFW0HE/NEYejFHQQcg3BKveeC3FlvhKm2/syIhtfCv+rIAUD0t6bLqoqR9JsVbIDm8tSiZS+YCMW6GEL92CvG0VtKplOcLcay8vVjmE6/7D6vCTl8PAABRj4KuNrFUEjH/yCbc8w91QafEUkkUjQUR80dD9GKOgg5AOLWxFVwm9zHfee1vDifYHPeJvEo+AwaAj83TLmtTXZT0tUnhFkx+2FLU7vf6NgDEnnIhsp1CbNNKuqUye/XybroQl+rPBwAg6lHQ1U2slETMP7IJ9/zDUdApsVISRWtBxPxRH72Yo6ADEC4J1sKOcv/yo9e+ZmtC6oSr9eUAoEZ1UdIsk6KtCUl+Vt8GgNhUJsRZTiEe1ws5k9ygPxcAgKhHQeebWCiJmH9kE+75h6ugU2KhJIrmgoj5wx+9mKOgAxAOlhRHmty3HPTaz6xpYy28UF8OAGpUT73sD9VFSUd8S7amZUtx8r/0bQGITdOFuKRciO9MSrmfIx+foz8PAICoR0FnnmgviZh/ZBPu+YezoFOivSSK9oKI+cOMXsxR0AEItUTbxP+T+5U9XvuYpb9Lzz9dXw4AvFUVJ92ll2uhyNaipCn6tgDEngoh+jlNLmlpkjX6cwEAiHoUdP4TzSUR849swj3/cBd0SjSXRLFQEDF/6PRijoIOQCglpBScLfcp1V77l3fPyZh4mr4cAOiqi5Le1cu1EGWdvi0AsaNCiFOdQsw2KeL85bB6jr4eAACiGgVd/YnWkoj5Rzbhnn9zFHRKtJZEsVIQMX9404s5CjoAoZRoc1R67Vu+vLCj4yx9GQAwU12U9I1JuRaKHDIq847TtwcgNjiFeNikhKs304W4VF8PAABRjYKu4URjScT8I5twz7+5CjolGkuiWCqImD9q6MUcBR2AULHYCjp47VcOJKY4/qkvAwBmPs2/5MTqoqSjJuVaSPJ16T8v0LcJIDZUCnGiU4hZegnXQG7Q1wMAQFSjoGtcoq0kYv6RTbjn35wFnRJtJVGsFUTMH4pezFHQAQgVi63gvZp9SqLNUaA/DgD+GPni2OqipIN6sRaqfFnR9gx9mwBii1OIkTKHTMo4n5QLMUd/PgAAUY2CrvGJppKI+Uc24Z5/cxd0SjSVRLFYEDF/6MUcBR2AUGhjL7y0dp9S8O1vrIUcDAcQkOqipE16sRai7NG3BSA2TRfC6hTiW72QM8ka/bkAAEQ1CrrAEi0lEfOPbMI9/0gUdEq0lESxWhAx//imF3MUdABCwWIvcNTsTxKsjlL9cQBoSHVxu2dMyrUQJHmZvi0AsUv+R8jvnEJ8alLKeedwhRCn6s8FACBqUdAFnmgoiZh/ZBPu+UeqoFOioSSK5YKI+ccvvZijoAMQCt6Xt2xjdVj1xwGgIVuL2vXxLdeanq3FyTfp2wIQ22YKcZpTiIUmxdzPmS7EpfrzAACIWhR0waWll0TMP7IJ9/wjWdApLb0kivWCiPnHJ72Yo6ADEApyP1L98z7lmvEJ+uMA0JBvJl52WvXUpB16wdbE7N5eful5+rYAxL58IY51ClGgF3NeuUF/DgAAUYuCLvi05JKI+Uc24Z5/pAs6pSWXRPFQEDH/+KMXcxR0AEIh0V54l8XqGKci8iqP0x8HgMaoKkoablKyBR25vjv1bQCIL+VCdHcK8aNe0Mn75+jLAgAQtSjompaWWhIx/8gm3PNvCQWd0lJLongpiJh/fNGLOQo6AMG4MKXgYoutcKTcf8yQeV1msUqCreAVi9VRoR5LtE/4k/48AKiPUZl3XHVR0ot60RZklq8vTj9J3waA+FMhxD+cQmzRSro1+nKVQhwn779SLn9HuRCPya8Xe0VdMrNY3t9/hvyfVfpzAQCIKAq6pqcllkTMP7IJ9/xbSkGntMSSKJ4KIuYfP/RijoIOQCAS7AVd5D7jQ30fUk/WJNgK+ghhHKOvCwDMfFvc7vTqonarTAq3xqc4af224ivO1dcNIH6VCnFBuRDLvQq6wxVCnKoeKxbiJPn1KJMSz1+OyrxcJsQV+nYAAIgICrrQpKWVRMw/sgn3/FtSQae0tJIo3goi5h8fTA6eU9ABaNBF7fPPtFgdC/V9RwBZ/OuUCefr6wUAM1+Xtm9dNTXpKZ/irXF5ZcuU5F/o6wSASiFOdAoxq6Zomy7EpfL2YplPTUq4xqa0QogT9G0BANCsKOhCl5ZUEjF/38ebM+Gef0sr6JSWVBLFY0HE/GOfyUFzV/TlAKBGQkrB2XI/8bG+3wgi6y9q/8AF+voBwIxhiGO2Tm3Xu6oo+UuTEs4sVVuL2g1Vl8nU1wUA3pxCjJQ5JPOAzHcmpVugWURJBwCIKAq60KallETMP7bn3xILOqWllETxWhAx/9hmcsDcFX05AHDJqzzO4v58OZ/9RjBJsDneFPn5x+qbAQB/Vla0PaF6alJGdVHSDJkvZH7wFHJ71aUstxYlzamampzH580BCESF/K8cmb0mZVuweUDfBgAAzYaCLvRpCSUR84/t+bfUgk5pCSVRPBdEzD926QfLa6IvBwBKoq1whL6/aGoS7Y6B+nYAAACaU4UQ/zYp2ZoSdUbexfp2AABoFhR04UmkSyLmH9vzb8kFnRLpkijeCyLmH5v0A+U10ZcDgEvy8k+U+4et+v4iBFmrbwsAAKC5qCJN5qhJydaklAsxQ98WAADNgoIufIlkScT8IzP/w5vLyjetKFig5v/q4ze9/9SMYRseKx20deaUvntUnBN7HS4t6GE0NZ6CrkXnrLPOMq644grXrf5YuOMpiOI6gc7/mGOOOXD8ccdtUZH/XiHzhIxDpr9MO5lGX3qIgi48TA6Uu6IvBwAJ9oJO+r4iZLEW/lnfHgAAQHNwClGol2shyreVQvA5mACA5kdBF95EoiRSYf7uNMf8VSG39KlRyxc4h252Tux1aF7JkB/V/N985r49ny4uObjxvdnG1x8vcGXPuueMn75c1OSo1+3XH21o8Vn69ItGdmaW61Z/LJxR89fvi8cEMv+qlWuND15c4sor858xyh1Tdo3oP/iry/+VtO6X512w5bhjj/vxmGOOWSzfNsbJXClzjPZ28jMKuvDwOUjuib4cAMh9w1R9XxGqJNgcw/XtAQAANAenEO+blGshSYUQ/J+QAADNj4Iu/GmOkkgP869NOOb/zZrJc19bcNOKWVP67X2kaNC+1xbcuXftsvIjuz5/xlWgqfnrpVooEy0FnUogJVGoQkFXm1DNf+M7HxvzS2bu65PTbeP5555Xffxxx20X7g/T/pv+vkJBFx76QfKa6MsBQKLd8aq+rwhVEq2OKfr2AAAAwk3+D59jnEL8pBdrIUyOvk0AAMKOgq55Eo6SqL4w/7oJ1fw/fv3uF/5dNnjbjMl9D75RefdPVSvn+ZRnKhR0dROqkqixoaCrm3DMf/mzLxuDe/bdcvppp+084fjj18i3k+7Cc0kQCrrw0A+S10RfDgDkvmGNvq8IYebq2wMAMxtntz95c/Glv93yUFL76qlJGVVFSf30uB4rTv77jqntztefDwDeKoQ4w6RUC1nk+ofq2wQAIOwo6JovoSqJGhPm75umzP/DV+58aW7xwO/mlw7bt3Zp+dG96xf6lGbeoaDzTThKIn+hoPNNOOc/f9rDB35/0W+rTzzhhK/k28pQCrrwMDlI7oq+HABYrI7P9X1FCDNf3x4AVBW3s1QXJ/fYWpQ0pboo6dXqouSt8tYIMLuri9qtkrcPV09td9224nYX69sBEL9mCXGuXqqFMhVCjNC3CQBA2FHQNW+aUhIFEuZvnkDnv+6t+56eWzxw1+Pl1+5b99b0o3pR5i8UdOYJZ0nkHQo684R7/oseqTz010v+XOUp6Kz6+w2axuQguSv6cgAg9w0f6vuKkMVeOF3fHoD4Y+SLY6uK2l1dVZQ8dWtR0gaTsi1UqZbbqKia9q9rjMo819UaAMSnCiFO1Uu1UEauv5++TQAAwo6CrvkTaEkUTJi//zRm/rvXFc18bs6IL2Y/NODg2qXORhdzNaGg859wl0QqFHT+01zzP+vMs747/rjjn5ZvM4n6+w6C43OQ3BN9OQBItDme1/cVoUqCrfAOfXsA4oe6ZGV1UfIkVZyZlGnhTXHS5qqipDu3l196nv59AYgPTiG+04u1UGW6EFfp2wMAIOwo6CKTxpRETQnzrz/1zf+Dl+98ecakPgeWPnnvwd1fPOtTjjUmFHT1J9wlEQVd/WmO+Ve9/5kxctCw70888cRd8q2mi/7eg8DpB8lroi8HAAl2x3h9XxGqXGifaNO3ByD2bS1qd/nWouRF1UVJR32Ks+bPT+pSmhR1QPxxCrFYL9ZClEMzhThN3x4AAGFHQRe51FcSNTXMv+Ho8z+4qdT5wrwbPp1bNOhA1cp5PqVYIKGgazjhLIko6BpOc83/tcefM875xdm7TjjhhAr5lnOy/h6ExtMPktdEXw4ALLaCq/R9RWhS+P3v0otP0rcHIHZtKU7+19aipDdMSrKWkN1bi9qN/TT/khP17xtAbJL/o/IOk3ItFHlF3xYAAM2Cgi6y0UuiUIX5Ny41839uzqhX5k0b9N0L82498MPnwZ015x0KusYlXCURBV3j0lzz/+/bHxn2K6/57qSTTvpMvu0k6O9DaBzfA+Xu6MsBgMjPP9ZiL9ig7y+aHKvjAX1TAGLTjqntzq8uSnq0umWcMVd/ipM/2Tb10n/qPwOA2FMhxIVOIQ6ZFGxNSpkQGfq2AABoFhR0kU84Sjrm3/g8MfOmJR3SbUefmHHnEb0ICzYUdI1POEoivSAi/tOc879r5Jj9p5x8ynb51vO/+nsRGuZzoNwTfTkAUCy2wt76/qIpSbA5tl/Y0XGWvh0Asae6qN2A6qKk73yKsJadQ+psOsMQx+g/D4DY4hSiXC/YmpjX9W0AANBsKOhaRkJd0jH/xmXt0vznZkzue/CFRycczcnqbKx4scKnDAsmFHSBJdQlkb+CiJinOec/bfykQyeddNL38u2nnf5+hPrpB8troi8HADUsdseT+j4jyBxMsE1I0dcPILZsKrv8rKqipCdNyq8oSvKz2yf9pZX+swGIHcVCnO4U4guToi3gVAhRXc5VXgAAkURB13ISypKO+Tec1a/e9eLDD/Y7+NV7s12llyrnQlXSUdAFnlCWRPUVRMQ8zTn/x8tmHT3l5FP2yLcgLkUUAJMD5q7oywH+WKyFY+XfzOKmJNHmeF5fL1ouS/KUUyx2x4sWk31HAPnJYi3I0dcNILZsmZb856qi5C99C6/oS1VR0optxVecq/+MiF1OIR6VWdzEFOjrRctVLsRFFUKsd5qUbgFks8zF+roBAGhWFHQtK6Eq6Zh//VFnzqlyrvqDR+sUX6Eq6SjogkuoSqKGCiJinuac//ySh41TTzl1l3wb+r3+vgRzJgfNXdGXA/xJtDnm6H8/QeQ7fb1o4fIqj0u0F95lUUWb7++zoaxMSC38q75KALGlqvhfHauLkvboRVc0p6ooac3WyW3P0X9WxKYQFDUqC/X1omUrE+Is+Xt7ROaoye+zoVSWC3Gevk4AAJodBV3LSyhKOubvP/99d8ITD0/ue6DmzDk9oSjpKOiCTyhKosYURMQ8zTn/qfcWHml1aiv1mXTn6+9N8GVy4NwVfTnAHwq6+PbrlAnny7+Buy12x0cmv1fv/CiXeTbRWpgphMFnOQExrqooqV91UdIhveCKjbRb9c3Ey07Tf2bEHgq6+CZ//3+WmSZ/h5tMfq/e2SHzsMz/6esAACBiQlnQbV75wOxUu+2QOjhbk1S79ZC6X1+W1J9gSjr3/K3a/G3M3ys7Pp40b8bkvvvXvVlxVC+9vBNMSVf94QIjPS3159lbrSnG92uf9VkuFInlgk4lmJJo3ZsfGGmp3vO3GlUrP/NZjjScYOefGsT8b7vupkNnnnb6p/Lt6AT9/Ql1mRxEd0VfDvCHgg41fpeef3pCSkFygr2gi0w/V6yFHS0pE/4i8vOP1ZcHEJuqpyYPqi5KOuJbbMVOthYlvWBU5h2n/+yILRR0qDFDiPPLhWgv00v+XfRTkb/bFHn/r/VlAQBoEUJZ0Kncc2uf1d27dN6pDtDm5XT8Xv1bX4Y0LsGUdGreuZkdflDz79Gt89fMvzYHN5U65xYP/PaDlx88pBdeZgmmpCt2jDFyszu7CooeXbN9Hg9VYr2gUwmmJJp8b4HRqWMn1/x7devh8zhpfIKZ/6T8CUZaappr/v169fZ53F9SLrvywLlnnz1Nf39CXSZFiSv6coA/FHQAgBpbi5N6Vcd4OVeb5Pv0nx+xhYIOAABErVAXdOpsrQ7p9gPqAG16mv0AZ281LXpJV71q8ix9Ge+oeael2g6q+XdIT2X+Xlk09/pPXpg7Zr9edtUXvaTbtrrSZxnvqLPoOnZwFxRTJ4z2eTxUiYeCTkUviTYsX+WzjHfUWVzpaemGzWYzHho/0edxElj0+W967xOfZbzjPosxzUhLSzOKHZN9HveXtUtWGGefedb+k48/+Rr9PQq1TIoSV/TlAH8o6AAASlVRckp1UdIB3yIrZnNky0NJ7fU5IHZQ0AEAgKgV6oJO5e5b+6y22axH7hnT5yP9MRJ4akq60knXvtmxQ+r+hkq3O0f3+cg1f86e+zkfvHzny48UDd7/w+eBX3KypqRbMGO8kdm5o6uE05fxzuT7bnYVdO+8UO7zWKgSLwWdSk1JNLdkhpHRubOrBNKX8c59Y+92zX/xk4t8HiOBp2b+E8eNl/PPaHD+d426Laj5P+GcY5xx2um75NvSGfr7FNxMihJX9OUAfyjoAABVJf/6n+qidt+blFgxna1FSRu2VrQ9VZ8HYgMFHQAAiFrhKOg8Z9E1WCSRxkeVc3a79bDVmnL09lt6rdEf9w7zr5tdn0+dpT53rmrlPJ+iq7FR5Vxqqt1Ik5laMNrnce+oAq9zp/Swff6cSjwVdCqqnLPb7UbH9A6uy1jqj3tHFUjqMpeN+fwz0nDWvP6OMaB3P9fsczKzjAfvq3/+XyxbGfT8+3bpfvQ3v7roMf19Cm4mRYkr+nKAPxR0ABDftk/6S6uqoqQ1enkVP0l26DNBbKCgAwAAUSscBZ3KipfyF+j3keCybfWUWTWXrVRJtdsONVS+Mf/aPDf7us+XPJm/Ty+5GpvtHz1hdEh3X7ZSRV3CsqGz6F574iGf+0KZeCrovnx7tZGeVjv/Th06NngW16J5lT73kcDz4cvLjMzOGa7LhtbMv3PHTmGbv7rU5Zmnn3HotNNO+1/9vQoUdGg6CjoAiG/VRe1m+pZW4U7yVpllMi/Lfz+9tThpsfvfSdt8lw179m0tuvxCfS6IfhR0AAAgaoWioDtSVV66cP6dL919981bevbudSArO+fIzwfTO3U60qdPj92jb+y97vm5YxYe3FRWrj+fNJzP33LMG9A3q9pqtR5VZ9HdNar3xzWPqfk//9i4l++5e3R1rz69D5rNf9QNvb+Ix/l/8eZ9z8x6aMD+3V807Wy2L96aa1w3uLch5+/6fLPigjE/P/bjhoXGsicfMCbf2tvo3yXVyO3onr1KRgebMbRPtjHhruuNN5+bZuxZ/7zPuoNJPBV0KitfWmoM6T/QNX91Jp33WVw7Vq83XqpcaEy4f6LRp+8AIzunS22Z1DnDGDRgsHHv2LuMVx5/1ti2ap3Pukn9+e87HxmlEx9yFXVq/h3SO/jO/4lFpvPPzMwyrr/2euPR0oeNLe837oy6e0ffYfz2ol9/pL9XgYIOTUdBBwDxa+tDyVkmhVU48lP11Hbzt05tl/t16T8v0L8Pb98Wtzu9elrSZXL5e6unJr9vsq5w5GH9+0D0o6ADAABRq6kF3ZOzRr3UrVuXn/oPHnZ06vR5xqK3PjKWrdlsrNq8x5Xla7cZC5etMkrnPH5owIA+u/NyO33/7n/yK/X1kMblv+9NnDOkf84mdUadOouucuZoOf+8fQ3Nv2z2/IP9+/f6Id7mP7eo/3drl5Yf1guuYLPxvceMkdf2Mzp3THedRffagslGny4djKFdrjYevvEK4537LjU+m5isPuPAlfWTkozl915mPHqr1RjRP8fo1yvPWLN4ls96A028FXQ1+fj1t40Rg4f9fBbXwkefNHr27GMMGDLcqO/v3zlngTFMLtOnRy/jvRfe8FkvaVxe/vczxoA+8u+/k2f+jz1l9Ozlf/7LPtli/HvREmPUbXcb3bt1N1a+uMRnnXqqVq41LkywHP7j//xPB/39Kt6ZFCWu6MsB/lDQAUB8+rKi7RnV4T9j7ZuqoqRb1Lb07TdWdXHy3+X/hpoj13XYZP2hyv6vHmr7S33biG4UdAAAIGoFW9Dt31haPvrG3l9079Hj4FOvvvvzAdmGMuOROd927pT2E5dgbFpUUTdiaNdN3Xr0OBTQ/Gc/vD1e5v/hK3e+9HjZtXv1cisU2bzy38btNw0w+mRbjTfuueznQq6hPOHobeRkdTLWvx385+GpxGtBV5O1S983xtw0yujRq68RyN//I3MWGNmZWcZHry33WSdpfNa/tcoYO2psQPOf+/QrRlZWtvH+fxb7rE/Pg/dMMC7+/R/W6u9X8c6kKHFFXw7wh4IOAOJTdVFSqUlRFaocrC5Kvu+biZedpm83WFsfSvqHXO+7JtsKUZLv07eJ6EZBBwAAolYwBd3hzeVlNwzruuHGm2/+ccWGnT4HYhvK3MrKb7MzO/zw9ZopD+vrJg1Hzf/G4d2/HDlq9P5g5j9vwfxt2ZnpMT//ucUDdq5703lIL7eamr3rFxr3jB5qjO1vN76aUnu2XGNTWdDbGDawp2s9+robm3gu6LavXmfcfsutxs1j7jCC+fufM3u+63KZaj36uknDcc1/1K3GTWPuDHj+jzz9stGta3fXZTP19Xpny4pPjXN+cfaRS/7nkr/q71nxzKQocUVfDvCHgg4A4s/2okv/VB2mM9Lk/7b5uOqhS/+mbzMUjPz2x1eHr1j8yjDEMfo2Eb0o6AAAQNQKpqCbWjD07SFD+v+wcuP3PgdgG5t7xo3dcdfoPh/p6yYNxzX/oYP3NGn+d4+ujuX5f/z63S88Vjr0R73YCkUeLc83buzb0dgyNfByribD++YYyxeW+qy7sYnngm7W1DLj2hE3Gk35+x8ycIjxeuXzPusmDWdmUZkxfMTIoOc/zvGgMfE+h8969Yy59kbjn3/7x3L9PSuemRQlrujLAf5Q0AFA/KmamvQfk4Kq6ZmaNPvT/EtO1LcXalVF7e722XYIsrX4X1fo20L0oqADAABRK9CCbvPKybM7dUzbt/iDdQf1A6+BZPlnVYfUeja8W/iIvg3iP675d0rf//qqDUf0mQaSdz7btL9Tx9SYnf/80kHVny0pOaAXW02N+ty57IwOxkeOxl/W0ixzb001Jo670Wf9jU28FnTr3vrAyMrMMl7/8Eufv+lAUj7zMWPCneN81k/qTyjmv/yzrUamXMdnS1b4rN87a+Xjp7VqfeTcc89trb9vxSuTosQVfTnAHwo6AIgv1dOSLtOLqRDkyNai5Nv1bYWT/N9P5SbfR1MzUd8OohcFHQAAiFqtc0uPHqnyLTj8pdgx7K2xt9+8XT/oGkzuvGvMjqkThizXt0H8R83/9jvG7NRnGUzuuP2m6lic/46PJ82bManPgaZcQtJf5jvvMybckOVTuAWaZfmXGsP75fqsvzH5ccMi47QupT6FRjxkdrHTuDt/gs/fcqB5dvEHrrPo9PWT+qPmf1d+gc88A81d4ycZM+S69PXrsV7R3uhwjX2c/r4Vr0yKElf05QB/KOgAIL6E4ey5o9VTkwfp2wm39cXpJ8ltf2by/TQlH+nbQfSioAMAAJFyrMxVMu098b7ExNle9//e6/46Ts8r+fHb9c7pesnhLwN6Z1U/uXBRSAqiJxa99L1an74N4izd92WJU79PRc2rctGre/RZBpMnFj2/PRbn/9qCm1a8vuDOXXqxFUh++OI5n/tURg7vZ7ySf7VP4RZovpiUbGR3svusvzH5Zu1C47yeZT5lRixl6wef+9yncv3Qa40FL77p87ccaN76tNp1Fpe+fuKOv/mPCNH81TquHT7CZ/16SsZPMtr+5W8b9feteGVSlLiiLwf4Q0EHAPFjy7TkP1erQq3Ip5hqQtqN0LfTXMJwNuCRb4vbna5vB9GJgg4AAESS/G8RoQ7QqYz1ur/Sc99BmX943V/Hmd1Kvv7iA+dsveTwl6yM9D1LV29o0uUta7L0o41HsjLS9ujbIM5Su812pHOntJ+uH9p1g3PydUvfXnRPpSrtsjI67F28eqPPLIPJslXrfozF+c+a0m/3lpVzj+jFViCx221Gl5wMY9xt1xpPzpxgfPzGTFdp1y03w/ikMPjPnqvJ5oeSjDS71We7jcnGVc8Zvx1Q4VNmxFLsNruRm51j3D16rPFY2Uzj3UWvu0qjvNw8IxR//+//9zsjLS3NZ7vEHbn/kX//ucY9t94elvmrdXTp0tVnu3o2LF+lLnN5tHP79ufo713xyKQocUVfDvCHgg4A4kd1cdJ0k1KqKYn4VQ3k9/CKyfcVdKqKklP0bSA6UdABAIBIUv+vr00y6iDdHhmLzJWef6vcX7uor3O7l6x+5oXyJ/WSw1/sduthdXBbP+AaTNR6bDbrkZSUFIP4j5z5IXWblmY/qOb1/+ydCXwU5fnHX/FGrVq1anYWqdrTHv+WKhutbSS7GyJyClqPqvXEW6vWeq8KOxtAYENCNkGkHnjE+6qi4NGKKOKBt1YFJRvAE2/lmv/zJlkNz0w2O7PvzLzvzPP9fL6fQDJ5j+fd2cD7y8wsfOtDUy2d+OySj9ZR/Xv3wNqajo8HDak1kom4sXSyOXCza9vUmBG36Is0e2BtbcfHg4YMgfeLhCHi9f/c0k+NeDxu6ovcUF6jIQce2PFnkfVftOQTo6amxhTIWclvczn6oBEp/LMrjFgEJR3i4wiiJyigIwiCCAf8yrB8NvYlDqTK8A7DYBvhfrymfWpljcXYyvF03AehJhTQEQRBEAThNzXgepBv1PEr557v+vNLbMPbXprY8bDGW8Zf3TQXB3E9OWJY7RePLV66Hm+4OhHaMfgVYbgPMtdY2CDnAdqQA2u+vvyCo597/6XJV48YPuRLXjdcSyc+vvidNfyKSNy3yj5++znzH77pnx/iq87sWqh/TTJhjBx+kJGb9E/j/RdvMw49eKjxSl35V9B13OLywGpTv6U469ZbjL9ecYspyAiS39W/pgbqP8KYXjfFeGfB4o4ruES8/jtucTlilKlfstNC/QfXDDZGjRgJr/9sZ/1Hi6v/CGgX92vlpWf9w6je/89P459dYcQiKOkQH0cQPUEBHUEQRDjgz4mzCKOc+sr7jVVb4z78wEhVbQLj+dhijA6tbMB9EGpCAR1BEARBEDIwEyxcNcddA/5hgyMs2Gb0tHNHXtr4LA46evLoI0a03zbn8W/whqsTb3vwiTW8PdwHmWusSSbWHn/0qLZ5t114Z/fPH3PkyOW3PTRfSEB624OPfhG0+t+SO+Hd1x5r+ByHWnYdXJM0zj39WGPRnJYNPn/6CUcYcy/ZzxS42XX+5fsax4529gy6MydfZ+jND5iCjCA5uKbG+PtpZxj/vevBDT5/8gljjVvnPGl6Ldv13v88bxx1zLGmfslOeTB63ulnm+o/VmD9jz66tPr/+7pbjZ/uvie/Ojz0WAQlHeLjCKInKKAjCIIIB/nswMfNYZQjV+frK3+H2/cTGNNsi3E6tPIu3D6hJhTQEQRBEAQhAzuwzufNFQK66Rt+2Zq+oxr37ndUwzIcdPTkpCuOeyp15RWr8IarEy8bN+4z3h7ug8w18qvl8Oe4HfUfn/4S19KJqStTHwep/mvfm96Uqzti9arX7jSFWnblV8vhz3H/Ne1SY+LJSVPgZterz9jfmHjOoab2S3Hv01qMhx9YZAoyguQ7CxabPsdtmTzNuDwzxfRatuvUluuN8VdmTO2TnfZU/2aB9U9fUVr988++bmy91VbG8YcdtjP++RU2LIKSDvFxYadfTWavaCI9LJrUT4km9HQkoc/iaonM7VCvR+HjrYXPgVdE45mT4fMj+fexMa0b4/aCBAV0/rDroHG7wetseCSRuRBem7mO12NSfwBqeUfHn+N6M/z9PK3wOiSkp98QffuKRHo//v7B30e+f5/R54CP8nON/z0a1ydHk5mL4e+HaTV1++w4rG4b3BbRC/C+3K86/Ut+fkQTmVO1RPryQr2hrq283uD9361BXL+MrwtY079qwi64uTCwdMqAXfPZ2DpzGOVI3587h2nPDjzTYpxOXYDbl5VZjG3RzNjvmxg7BDw7x9hV4Kwu7wcfBW/o+nsLHHsZHHdsC2O1DYzthtsLGhTQeU+KsT7wGvsV1O0w/nrLff96fAT+fhP/M3w9Cx9P56/DRvinKG6DkA9Yu135evH3GfhzXdeaXpvrfI+Z022d0+C54Cjwt3DsprgtojhQs778fb3rHDoHPma61feerprfwf8OX88V3tenMxYDt8ftEQQhB+eB3a+gWwr2/p9A+E/PdmMaPln8zPRrceBh5SuPj7th+PCh3z715vumTVc78u/n7fD2cB9kz/J6jRgxbLWQ+g876Jsg1f/dhelbrq8/9lMcaIn0rQU3GAcPG2y8fVV5t7k8aXSV8djsy0zt9+abz95l9DsqZyx/4X+mICMMvjj3SWPkqIM7Xr/4NW3Hv514svFA6z2m9sniiqr/sVD/ObfcbWq/Jwf+7g/GUaMPOwX/+AobFkFJh/i4MKFVTt6Sb8JGEpmrtET6aajHF7g+Nv1WS+qL4eOMSDxz1G7JK3+M+1QZCui8Yc/a+s21ZKZWS2SuhnqttKhhb36qxfV7tWT6uD1rU/xZ04TP7FSV2jqaTI/gYWokoS+xWDM7vgLtzIwmM3/hQR/uK+xEE3UVWjxzeCSuN0KtXgDXWNTQjh9Ae3dGE+m/R6rTP8X9BZG2bGysRRDlxMWLmgdIt+GanzowYTFWhw58HbcvC82M7di1aTsTXAx/XpszB0p2/AxcAF4F7Y24mrEf4j5VhgI6b5jF2HZNjP0VanU7+KVFDYsK65TPdYY9w1OMbYLbJ7yHB3KwpsfAmtwMfozXzIbfQFuLoK1JYE09Y5vjvsIO1Ofn4MlQqxsEvWe9BV7LQ7tGxnbB/REE4T0/Ab8C+UbdI10fubnuB/XEDw9tmHHaVdOfwIFHT55/1pGvXlk3+Vu86WrHKydMWX3+2X99DbdN9i7U7XWo3xpcUzteWTfp6/PPOuJV3LbKLnrwwofumXXm+zjUEm3dJacYk8dWmUK3Un3k0v2Mv45KGl++da+p7d6smznbGKvfZgowwuSVF6eMKydOM72mS/X2h58yDj/ir8bKkIac5XrFJZcbV04qs/6H26v/ESPHGIcMHXkN/tkVNiw2HTvExwUdHn5EEvqYaEK/B+b/Na6HC/5Pi+sT+sUzf4R/Wm2Ex+M2Wjzzax4IiBDm8rnF/Oy6DrdbrloyPRvP2w78yiTcphNhHGfgtu2wc3Lij6CdKzQeCJjr5tDMkbgfu8CYJuG52hXG8gJu1y0qkuOiuH9HJtNH47btEkmMr44m0rdoPLw3rY0Qv+0IY+PpUUG/grcY/Qfr/aNx/SKoxwKLGon2f5F4+kotMYH//9kx/Bc4TK85B4o4xzHLp+67d76+8mVzGGXT+lgcty0D79Xvu4dprM5dgtv3kwbGdsgxdkoTY/MtNmLd8Mlmxi7km8Z4LF4A/f8NXCLI1Tnz/OzKA6clIuVXIeF526FrM9/UrgMrcdt2gO//JQ+L4eM3OXPdHDmDsbJ/ES7XeYXRkjK9GrfrFrnOq81w/7ZtYaysOy7w8AzaOQx8PGexNoJcBV4Dr5uBuP8wAe8B+/DQMtcZpuEaifZJ8PRrGNsJj8MO0MbonMXrzq5+/WwhCL/gm0WPgXyTLg/yBzjz3/zhf18P4n9Ybwf273Ir/oltRjfuW3Fkw4q1bU3TcehhZduzk2aNHDn06xvunmfafC1F/n38+3k7uG2ydzvrP+wbp/Wffffc9SNHDP0qaPV/+OaznvnPHZeuxKGWaNufbzUOGVFr3HeB/WfRLZ1caRw5/ADj4esuN7Xbm1+8dZ/xm7EtxoP3PWMKMMLk/554zhg9ekzH+wh+bffmwrc+NA474ijj3tm3m9olS7Oz/od4Wv+LzzzXqD0grswtiNzCYpOxQ3xcUNmtZvyuWjIzUUtkPsE18NB3+e3TeICAx+cW0UTdHyzGETTn4XnbIZrM/MmiTdtGE/oluO1SiFSnd9AS6QbNhcCY3x4T92cX1a6c7BfXd7fo375x/TTcdinsNSa1Gf9eaONNU5su2hHWJPUTWVUqHFcUwDyjSf1Q8GGY/3pcD498lN9+lqVSffDweoNfjWfRnm357Thx2yLgV761T41dmM/GvrIIpXq1bWrs37hNWVhev/9OeLxlyPdPfCfXeXs4fkvAb3PmDVevfAYcW8+YZ1duQ39nWowjaF6O520HeF3806JN20I7B+C2S4FvsMP334vbE+Ay3JcTBF2FdA9u1y26rj7E/dsW5v173HYpwPft2Nx528oPcJsu+xS/qg6PJ6jw91Go81kw75csauGF/BcWbgT/gMdWCl1XVOI2bTudsd/gtgkiyIwFC1fMHdP1uT3Ar7s+t5R9f6tLfhsV/ltiheP/2vV5tsOhDc9Omz393zj06Mkn77u0ddiwg1bb3aTlxx94YO26GVNOfRS3SZZuZ/2HrnFS/9rawWuDWP/brx77v5fmTf0QB1tu+OIjVxsjD0rYCumWTa00/nFElXHSX0caCx9oNrXZm7fd02r86axZpvAijC64/xFj+IiRtkKiRUs+Mc4870Lj+GOPNx6/4wFTm2TpdtZ/lKP6X37RZab2evOaqxqM2O/3aSv8vAoreFOvID4uaPCrkvhzu2Cu3+C5++g6/ky7fkl9AB6vaCig6x3fArqq1CadQU76I9yWIJfiLp1AAV1pDBjQvKkWT5+g8SAet+Wtb0SS+iA8vqDAr4LmzwiF99B3LObul69FEvoYPNZiyB7QFei62uwhi2CqmOuXTR0o7abakllVW1iM2alLcftewjcvmxj7N97U9Fl+FdnUGfBPTzxe0eQooOsVvwI6OH5b+L4pOTFXJpqE131Zd08oQAFdacxibDv4nnG5zlvdmtrz0LmN8E8BPL6gAOv7o1zns/k+sZi7L8KYHuZX8eGxFoMCOoKwD78VSpZ1vN+yqWD33/47ruvz3CGs80q7u8FvwYfADQK6vqOmDdvt6IZlpV5Fx+Uh0ahRI77htztb8MZK04Zsd/nX+XGjRo38lodDQw8a/NXc2y68E7dJlm5n/Ud+W2r9x02atn7UyOHfBLX+10/728p3F85chYMtt3xx3gzj0BE1xuST/mS8Nan4M+le0Pc1Tjv0AOOSM44wFtw/3Rg1YqitkI5fPbfPaTOMG1ufMIUXYXXBfY8YY8Yc2vG+0tvrf+6zbxknnnKmccG5/zQeue0+Y+TwERTSlSkP6UbbqP9Jp5zRUf/2594wtdWbD990l/HT3ff8vPDzKqzgTb2C+LjAUJXaJJLUz4Y5rsJzlsw5FTXjbf0n2Q4U0PWOHwFdZ4jU8dxDUzsCvR736wQK6HonUjP+APieN0xt+CisW1P/qtQWeKwqoyX1IzT/A9BiLozE62J43FaoEtAVaK+PHZGfGltpEVCZlPnqOc77jVVb4zGX4Wu4fS+Yztj2OcYawDV4Q1Miv21ibMY0xirw+EWRo4CuV/wI6GDdBzV3PivO1I4ooY+TcL9OoICuOPAfxY3g2BNz5T1bTrRfQR1K/jeaCkxmbEuY18U5/wPQYt4Cr4V+eOxWUEBHEO4yGuQbeWeyzn8kbBDQcbY/pOGJCxsb5+Lgo5j55ydd889zjn97+IgRay/Tpxi3PbTAePSFJR2bsvwj/zv//IiRI9edf+4J7/Dj+ffxcCiIIZHXdtb/pCW8vsXqz9fn/HOOfyvI9f/X1KM/XvHCDZ/jcMtNl7/QamTO/5sxakjcqDvhT8a8S/czXq7btyOU4x8fvng/44q//dkYMSRpXJ893/jyrc7v4+GcnZAud+NNRvyc60zBRdh9a/7zxuWXXAHvLwd3vM7x67/1wSeMCy/XjeHDRxhXZ3PfPfeMh3MU0pWvrfpPnW7ruXPdffreeUa0IvJN959XYQRv6hXExwUB/jwizZtnEYmSP5dtFr8NJ55LuVBA1zteB3RdIcOn+PuFG0+fgPt2AgV0PcNvT8rPXdP3SmI0nnlWGzQugsetGpGazG9hPk/h+UnqenjtzOxfleKPhegR1QI6zrvT/7h9PhubAa63CKq+V9JnzxVoqx+omcbs1KmxJ3D7btPC2J9zjC3DG5kS+3kzY5fwDWg8l3LJUUDXK14GdCnGNsl1Xv2zDn+/C/4S9+8ECuh6Bo77JfgE/l5Z5FdR8mfh4XGrBsxjJMznPTw/SeVXSJ+fgnMdz6M7FNARhLvwK+gO7vqzZUC31Zhpv97+0MaPFz8z/VocfvTm60/o108cd+rzxx571KejRo1cXV1dza+WW83/PnHc6S/wr+PvCWJI5Jed9T/jBXP9j1w1cdxpz4Wh/ldP+usXH758y9c43PLCd56ebVwz8WzjtKNHGIcMrzF4/Q8dMdg449iDjdkNFxn5524xfU+pId2bz95l9D8mZ/z34edNwQXZ6cuPPGVMn9xgjD3plI7no/H6jxlziHHK2FONa+pzxpv/fdb0PRTSibNY/Wf1UH87vjxvgbHjD3dY2/3nVRjBm3oF8XGqE02mR2jyXzXXk6u0ZJrfwUAYFND1jncBnbFR53MQzd/rhhXJtJAHq1NAZ03HVXNJfbnp++RzqZaY8BM8fhXgz/OLxNNXwhxWW8xLdtsjyfRBeE4FVAzoCrTX77N/Pht71RRWdboYHy8by+sr/2wxbqfegdt3k2bGLssxthZvYiriW9MZ2x/PqRxyFND1ilcB3UzGtuG3wsPf55If8Cu78BicQAGdNXDMKTl/n2lZqo80MrY1Hr8KNHXezrLVYk7SC6+fRS3wzzQ8pwIU0BGEd1gGdJytDm4472fHN7yx+r3mkm91aSXfoMWfszJoIZEshrH+zXVHfPvZG3euxQGXH/L6489Z2VtI9/n/7jMS5840xuUeMoUWZM/y+uPPWUkhnTuWWv9SXbLgRWOrvn3X459XYQNv6hXEx6mMFs+cDnNai+eonukH+1dN2AXPzwkU0PWOFwFdR9CQSN+Ev8dFV+IxOIUCOoyxUSSpX6Cp9V7zXkU8U9JtiWQhOmjCHh1XAJrnoprT+HPz8PxUDug4i5oHbNpWH7s4n4193T2wasvG/o6PlY381MrTLII2R7ZnY5Nx+26Q6rwy6Vq8eamg/Mqqqa2MbYbn6IQcBXS94kVAx29jCse8gL/HRYUF4xTQbQgPu+DrN+LjJXeealfSwZirwRUWc1HJr8CxeG4cCugIwjt6DOhYKtXnB2Om/bvm/MancABix1IDIm6QQiJZDGP9p+uHrf+q6xaSfltqQMctFtKdOvE6Y+gFNxkrHN4aMKzaCYgopBOvnfqXIr81Zp8+fQIVRDkBb+oVxMepihbXL8NzU9x8RTKzL56nXSig6x23AzoezsHX5+DjXTWp34bH4RQK6L6HBy28tqZj1fCVHYfVbYPnJCNaPD1KU/dKaCsXRmrrtO5zVD2gK7AsO/An+amV87oCq7VLpwwQfqtm0bRlY7fhoM2xUweegtsXDQ+zcozdjzcuFfepJjgN8FztkqOArlfcDuhmwFs2fP0dfLybwljOwuNwCgV0G7Qdga+9jI9VQZjPTXg+MsKv/IQ6p3LqXglt5bWzGNvgmccU0BGEd/Qc0AE7jWncepvRDYsPv2L6PByClKqdgIgblJBIFsNY/8b0YaaAyy/tBHRcq5Au1TTb2OfUfxlLn3nDFFiQxbUbEFFIJ1a79S9F1vkzK9TgTb2C+DgV0RKZM/G8AuK30Xj6b3i+dqCArndcDejGtG4Mr8/b8bGum0yfgYfiFAroOtmpKrU1fH6e6TiFjCbSt3Sfk4zAOM8H1+GxK29SX94vqQ8ozDMoAV2BtqmVR+WnDrwRf142jFTVJvnswE9MQZtD+e0+cR8iSTHWJ8fYzXjTMiCuaGZsIJ6zHXIU0PWKmwHdNfCjEb72Gj7Wbacz9t17ablQQNdJC2M/hc8txcepJIxf2L993YCHWDDOW/C4A+LT/P2gMFcK6AjCfeaAS8BPQL6p90HX300nzdYjm360zZhpL428pMFRSGc3IOIGISSSxTDWX+WAjts9pLugYbbxh1NmGa89+aopqCB710lARCGdOJ3UvzdZ58+sUIM39Qri41RDi+sHa0HczP3e9ZGkfjaed6lQQNc7bgZ0kUTmGnycF1bE9d/hsTiFAjpoc4i+PXxuoekYBY0k0pa/YOk7qVSfaCKdw+MNmJ9HEuOr+XSDFtCpQn5qbBgO2cpw3Uf1A3+A+xBJjrGr8IZlwPzcKvgplRwFdL3iVkDHnzkHn38WH+eBn7YytnH3sZQDBXSM8RAEPrcSH6Og3zQy9ovu9ZKFWYxtB2s332LMgZGfSzDH/ny+FNARhPuMYx3vLSZ/3P2gAtuMbNhh69ENT/7qpIanP/pfrgUHIsV0EhBxVQ+JZDGM9Vc9oOP+596cER88xPjzCZOMNxe8ZgopyNJ0GhBRSCdGp/UvJvxIUj6IKhe8qVcQH6cSXRvxQboNWs/GM//E8y8FCuh6x62AzscrO1fxsKP7WMoh7AFdxdBUXy2pP2n6urp+uEtt+rvfcpaCjnBOyOtMBb+M1Iw/gAI6f8jXD7zTImhz6mLcvkhyjA0H1+MNywDKn2OUwPMvhRwFdL3iVkAHn2vFx3jkA93HUS5hD+gaGdszp/6z0L6Th2Dwn9uNcN385GrGfgj1XoTHGlDfgddUlAI6gpCRE5s33XpUQ277Qxrevfb2putwKNKTTgMirsohkSyGsf6qB3RzH77T2OuEq42jzptOIVGZlhMQUUhXvuXUvyfhp5HSQZQI8KZeQXycMqRSfbSAXNFSouujCf0YXIbeoICud9wI6PjzA+Fz3+JjPPL+7vMrF0HBiaIBnbER/PkO09fUdxqum38YG0WT+rUWYwyucf2zSDI92vR5B1JAVzorp+y9ez4bW2sRtDm0sgH3IYpmxnbNMfYx3qwMsJ81MWb7yu8cBXS94kZAB38+C3/dK6HvC7vPr1zCHNBNZ2x7+POb+GuqCzX6C66bX0Cdt4UxPYfHGHBfEfUeQQEdQbhA39ENB209umHZr09qmPvUf5t6vZqunICIq2pIJIthrL+qAd2rT99lHD1utvGTY2cYN932ZEcYQSFReZYbEFH9y7Pc+lsJP4bUDaIEgTf1CuLjVIFvSuK5uOTaSEJfAh9fAB/t8k3wY4tj3Xb1bskrLe9a0BMU0PWO6IBu93hmW/j7Mvx1Dz0fz7EcwhzQ8TU1fT4YruZ1wrXzA6jzBIvxBd+4/pnpcw6kgK50eKBmDtmc2za18iDchyiaGJuNNypd8gtwCfgU+Cj4eNff+VVt+Fi3fdvurQtzFND1iuiAjj//Df6+Gn/dK2Ecf8RzLIewBnRd6/gA/nxAfCvF2Ca4dl4D72ebwVgesRhfGPzU4nO2pYCOIFxix2Ezt+k7quHirQ5ueP+nxzXOnXb99GYckhQsNyDiqhgSyWIY669aQPfAA3cah6duNKJ/bTEumTbHWLLwjQ0CCQqJnCsiIKL6O1dE/bHwI0jZIEoUeFOvID5OBSLV6R20RPojPBdBvqwlMnXRZHpEdNCEPXDf3amomrRjRTIzlB8P3/eGRVtCjSb0NB5Db1BA1zuiAzr48zT8NS/lV+/hOZZDWAO6SCJ9E3xciz8fHNOuXf1TKtFE5lTzuEg7UkBXGm31A7V8NvY1DtnK8Mv25gF9cT8imM7Y/jl3bm25tomxx/gVSPDnxAzGdsZ9dwfGsTscezh8z4yc+7fAWwP9DcNj6I0cBXS9IjKgSzHWp9nfW/V9Vc/Y5niO5RDWgM7DXwLwRRmuooNx3IjHRdqTAjqCcJntxzRvu/XoxlO2OrjhWfj4wU+ObXjwiHFNM5tvbpq+8Ilcw0f/a2oRERBxVQuJZDGM9Zc1oPvgtXuM1xbeZzz80L1G/XV3GkdecVNHKDfwtOuMyTMfNd5e+LopjChIIZEzRQVEVH9niqp/d+FHj5JBlEjwpl5BfJwKuHBVy5pIInONFs/8Gvdlh341mb20ZGZiJKGvsOijPOP6TNxfKVBA1zsiA7p+SX2A5m+o8+VeY1Kb4TmWQ1gDOmEm9cVaMj07mkiPh78fH0mmj4bz+WCw6juT+hH8mYWRuN4IxywC15jaEe+X/atS2+H6eUW/eOaPMIbVFuNyW97ncx0BbFy/DN77TyqsCazRMP5nfjth/jX4ew7+/AQc/6lFO1JIAV1ptGUrmy1CtnK8Bfchihxjc/AmZZl+Al4+jbEK3Fep8Gc6NTFWBe1cmxN/dd16aPtY3Gcp5Cig6xWRAR2s02n4817KA2Y8v3IJa0AnyG/Ap6CGM8FLwKO7HMLfL7ocxD8Hx50PXg++btGOcKHfhbh2XgJj+Dsek0d+Di6AmufAy6AOxxTWBBzTtRZj+dfg443NnYH7txbtSCEFdAThIVsOb4z2HTXtuK0Pbmzqe3DDE1sd3LAEPn7ON2jhz0LcdXjK+FP8wI6P+GuktVR/f+1e/50OazJ+dtxM409nzTZOTN9jTPvXY8YL/3nZFED0JIVE9hUZEFH97Suy/gXhx42SQZRI8KZeQXyc7PSvSm0hOAB7odxgDrNnbf3m0O7xWlx/3aI/20YT+j1sTKut2z4VoICud0QGdPBxPv68p8Yzj+D5lQsFdLb9GN6jZvGra/esTf0Aj68U+g3Rt4/G03+Dtp6yaF+YfoU7P64evzP0n8fjcdGl0bg+OZLUB1UMTTm46snYKFKT+a2W1M+Dth63aN83/VpDlViR3fdX+WxsjUXI5tj2KZUjcD8iaGbs1zmBV89Be7ddw9hOuJ9ygDZ3hLYvBT/E/TmxuYxniuUooOuVZkEBHTgKXGXxec9sYuwKPL9yaaaAzq5LwIkwjv34LRzx+EoBvrd/rjOw423h9oXpV7jTdRW0l7eBfRa8mN+21O5tgjmzGNuCjxnWZXzOowC1VP1aQ4IgusE3aPHnyiGRSPwZ2lzJP+KvEWZCWn/Tpr5fig4oKCSyJ9XfX0XXn8vPb3zChw28qVcQHyc7kXjmEDwH56YfdLZZWyJjWjfWkunjoK82c9+l2XH1RuXkLXHTpUIBXe+ICuh4OGb6nNcm9RSeX7lQQFeyL4HH818iwGMqh37JukQkob9q0Z8A00/j/rwA+r7PPBYXTOp38frBPwE2wmMoB/4sUHhv1zV/nkW6gRTQ9U57NvYIDtjKdMWi5gGb4n5E0MRYFm9QOhXa0nH7ImlmbFvo50rwS9y3Detxu3bIUUDXKwIDunkWn/PaBJ5fuVBAV7Jz+FVY/GpaPCanpBjbBNo8MSco7LfwKtyn28Ab2g+g33ctxiLar8FG/ksdeAzlAm0OzHVeLe1lyGgpBXQEIQGiAyKOIiGRFIS0/qZNfb90I6CgkKh0qf7+6kb9+fmNT/iwgTf1CuLjZEeL6zfgOTj0JVfDuW7sVJXaOpLQJ2n2b1v3Er+SBrdnhx2H1W2zwa30yjGRftBijPaM65+Z2i1TfmULnrcdhAV0Qsy8A17NN90jNeMP6D9Y768NGhfh49x10LjdeDAQiWeGwLHng3eAq7p/fyQxvhrPr1wooOvVZfy2iCyV6oPHIgp+VS6/3aJF3+W6viI5Lor7c5OO20eaxyHWpP6A6Cujreh4f01kLoU+vzCNwSMpoCtOe33sCIuArSzbsgNtPw+2VHLirii5FrftFg2M7dbM2N0WY+jNm1OMlfW+OQP+WdjtVnplCeNpsxijXZ/E7QqwP563HQQGdCJ8DsYzGT4eBfPah8+NX5HJr8Tif25k7Bfw8WD4ejrXGQjyWygWvnc1fH1rPL9yoYCuuNDfwhbGXN2/47ffhb4ex30LcAnuy22gz6stxiHSNWBjb88QFQGcG/2gr1ngOotxeCIFdAQhAW4ERBwFQiIpCGn9TZv6fulGQMGlkKg0qf7+6kb9+fmNT/iwgTf1CuLjpCaV6qOJuWJhXbRa3xs37zZdzyd72WI8Vr5bCGZkQbWgplR8D+iS+nItkanjzzDEY+uVMa0bVyTS+2nJdD201bZzcuJW+JByUW3dvQ3oMlc7vY2lE+C1erF5DOUZTeqn4H7couvWlhuEyoLN81uL4n7dhr9X86v1LMbjuhTQ9czy+v13ymdjH+CArUxX56/a25VQu4mxX+HNSYeu4Fe34fbdBsZ/SK7EK2Hg2Ied3h7PLQQFNffidv1GgoDuZRjDOY2M7YLH1hvwfX2nMzYMPt7U5MLz5ziC1j2IAd3XfN1SZYbopVLP2ObQ510W4yjXsn6Jzw4tjMUt+hcmD0v5zwncr9vAvPaG18JiPB4vpICOICTArYCII3lIJAUhrb9pU98v3QgoClJI1LtUf391o/78/MYnfNjAm3oF8XEyo9Xoe+LxOzKpP4Db9oqu59NNM41pQz+sSKZ/jr/Xb1QLakrFx4CuTYvrp/HXBB6TM8Texq+AauvuUUD3Fbxu/oL79oKuMBaPx7nxzJ24D7fgz+Yz9S/OO8q94rhc+JWUmsdX01FA1zPt2cpbLQK2smzLxq7H/YiiibFj8OakE6Gdi3DbXtHM2K65Xm6FCMcscuNKqHIRFNRQQPe9/LlYw/F4nGIIvLVidwSte9ACuvdaGPs/3LfbdIV0T1qMpxzPxf24AbyONoW+XrHoX4T8CrbLU4xtgvv1iq75TcgJfEZqKVJARxAS4GZAxJE4JJKCkNbftKnvl24EFN2lkKi4VH9/daP+/PzGJ3zYwJt6BfFxMgPjHYnH78RIPHMUbttrup6lZ7WR+4WWTA/Ex8uAakFNqfgQ0K3W4voEfutTPBYZUW3d3Q/o0h9pNXX74H69YsCA5k1hHIvM43Lsh8ylcLc7vGbQ1zqL/st1nRbP/BP35xf8StjOW9WaxumKFNBZ01YfOwaHawJctyK7r2tXLzQxNglvTjqxmbE9cNte0srYxjCGcTmLjVwehlzD2E74e2RAUFBDAR1jH0GfJ6Y8uvKqXASte5ACuuedXO0oCug7CmNYZTEup3pyTsLr6CyLvkX4KTgY9+cXMM8RMJ4vLMbpihTQEYQEuB0QcSQNiaQgpPU3ber7pRsBBZZCop6l+vurG/Xn5zc+4cMG3tQriI+TGS2R/gcevxOjgyb4unlVoCKu/07jV1F9P7bVWjJTi4+TBdWCmlLxOKBbKmsA2xOqrbu7AV36o4qa8b/HfXoNH4MmMOzqV53+Je5DNNDPPNyvAFf7dSVjMbpu5fmcxXiFSwGdmZVT9t49n419ZhGwlWd97Gbcl0ianT3HDbsct+sXMJ/DYTxfdx/bDDg98HGyICio8SQMsIPHAd3j/HlieAwyI2jdgxLQvdDA2A64T6+BOZ5mMTanfoTbFw2/Ihj6+cCi73JdnvPwFp2lwm95mXNnviYpoCMICfAiIOJIGBJJQUjrb9rU90s3AgorKSSylurvr27Un5/f+IQPG3hTryA+TmaiifR4PH4HrufP7cJt+0Wktk7T4vrrfFwyXNlXDNWCmlLxMKB71O9b8DlBtXV3MaDjVz5W4f78IhLXb7YYozOT+hG4fZG4dI6tAUfivmRBq5n8Q82DkI4Cug0xWsdsnM/G5pvCtfJd0z4l9gvcn0j4M7bw5qRdoY35uF0/gTH9Kdd5BcgqP26ZZwdBQU1oAzp47WX51ZO4f9kRtO7KB3RQh7ws4Wpz5+0Ul+AxOhVq1h/3IRLo4wLcpwD5s0Sle9xBARjfb3MlPnO0HCmgIwgJ8Cog4kgWEklBSOtv2tT3SzcCip6kkMgs1d9f3ag/P7/xCR828KZeQXyczGi9P7utFL/B7frNzsmJP4ok0n/Fn5cN1YKaUnEpPMDeIe5Zc96i2rq7FtAl02fgvvyk65aR5nE6MZmZiNsXSTSpP2zqs0z5895wP7LB39vdvt0lBXQbks/GUhbhWtm2ZSubcV+iyTH2PN6cdOA83K7fwJj+0MKY3//P7xVBQU0oAzro40LcryoIWnfVA7rVUAep7u4A8zzbYpyOhLmNwO2L4jrGtsqJD6o+k/0XGjgwzkrwK4vxC5MCOoKQAC8DIo5EIZEUhLT+pk19v3QjoCgmhUQbSvX3Vzfqz89vfMKHDbypVxAfJzPRhN6Ex+/EXWrTUj7/RHZUC2pKxf2ALv0gf24Y7lcVVFt3lwK6ecyD57TZBcb1ksVY7RvXXdtU7nwmm77e1GcZwmsyjfuRFX77UBjz53gOoqSA7nvaspXV+WxsLQ7XBPjZ+41774L7E02OsRfx5qQDX8ftEqUhKKhx7b3UKW4HdE2MXYH7VAlB6650QCfjGsK67AhjW43H6tDzcfuigHGeaNFfOa6D9TgI9yMrMN7DLOYgTAroCEICvA6IOJKERFIQ0vqbNvX90o2AojcpJPpeqr+/ulF/fn7jEz5s4E29gvg4mYkmM1Pw+B0p8XPeZEa1oKZUXA7oXtg5OXEr3KdKqLbuLgR03/I2cT8yoMX1yyzG68TXcNuiiCTT0y36K8fHZLpNcSnwW4hazEOIFNB18t60/Sry2dgKi3BNhP/A/blBE2ML8eakA9c1M7YtbpvoHUFBTdgCulm4P9UQtO4qB3RvtjK2Ge5HBmBsD1mM17ZQsxm4bVHA62cx7q8cYaw67kN2YNxNeB6ipICOICTAj4CII0FIJAUhrb9pU98v3QgoSpFCok6p/v7qRv35+Y1P+LCBN/UK4uNkRkukL8fjd2IkrjfitoneUS2oKRUXA7pPtcSEn+D+VEO1dRcf0KUbcB+yIPA2l98wF64QrBia6qvF9c8s+nPqpxXxTD/cjwpoicytFvMpWwroWNdz5wY+bhGsifC1Rc0DPLkCOsfYPLw56cRmxg7HbRO9IyioCU1Ax4OJyYxtiftTDUHrrmxAB+39BfchCzC+v+PxOtSVW/9OZyxm0Vc5Pg+vR09+3ogExtwXxv6WxXzKlgI6gpAAvwIijs8hkRSEtP6mTX2/dCOgKFUKiaj+futG/fn5jU/4sIE39Qri42QmktDH4vE7M/PJTlWprXH7RHFUC2pKxa2ADup1DO5LRVRbd8EB3Rpt0LgI7kMaqlKbaIJunxipTu+Amy+XSDxzCO6nHKNJ/RTchyr8uHr8zvxnD55TuVJA1/HcuTqLYE2E65dNiVXh/twix9j1eHPSiU2MPYbbJnpHUFATloDuW/CXuC8VEbTuqgZ0b8J/BIX/co4oYHx/sBizbWGNX8VtiwDWIov7KsO10N7vcB+qAGMfZDGnsqWAjiAkwM+AiONjSCQFIa2/aVPfL90IKOwY9pCI6u+vbtSfn9/4hA8beFOvID5OZiLxzBA8fqdG4ukrcftEcVQLakrFpYDuUdyPqqi27kIDuqR+F25fNmCc803jdiB/Vhxuu1yg3TtwP45N6otVu7UlBs6lc0zzKtOwB3RtUyvH8CDNIlwToWu3RbMix1gab06W4WDcPlEcQUFNKAI6aHMc7kdVBK27kgEdzP083L5M8Ftv5jrDYNPYbfoJbrtcUoz1gXbbLfpyZHPHjwC1gQncgedVrhTQEYQE+B0QcXwKiaQgpPU3ber7pRsBhV3DHBJR/f3Vjfrz8xuf8GEDb+oVxMfJzK6Dxu2Gx1+GqysS6f1wH0TPqBbUlIoLAd26SE3mt7gfVVFt3UUGdPwKMNy+bEQT6RwetyPjehVuuxz6V6W2gHa/MvXj1Gpd+cBhrzGpzWAuy0xzK8MwB3TLpg78TT4b+8IiWBNhfsmUqu1wn27SzNjheHOyDJfNYGxn3AfRM4KCmjAEdMv5Le1wP6oiaN1VDOjWQ1vy3iGgCxjnyxZjt22z4FtHQu32w32U4ZdBeL+GGv8c5rLOYn6OpYCOICRAhoCI40NIJAUhrb9pU98v3QgonBjWkIjq769u1J+f3/iEDxt4U68gPk52YMwf4jmU4YfRan1v3AdhjWpBTakID+iS+m24D5VRbd0FBnRrtZrJP8Ttywa83s6zGLtto4n0MNx2OfRL1iVwH2W4ELevKlo8c7rF/Bwb1oBu5dSBO+ezsaUWwZoI1y+fFqvFfbpNI2O/wJuTZfp8EDZ9vUJQUBP4gA7aOwv3oTKC1l3FgO4l3LaMwDjvshi7bRsYE3obb2jzctxHGU7B7asKzOVmi/k5lgI6gpAAWQIijschkRSEtP6mTX2/dCOgcGoYQyKqv7+6UX9+fuMTPmzgTb2C+DjZ0RKZ2/EcyvQrvsE9YECz0N+sDCKqBTWlIj6gSw/EfaiMausuMKB7EbctI/D6/YvF2B2YORK3XQ5aMjPR3Icz+Rxx+6rCn38Kc/oUz9GpYQzolk2u3DJfP/BJi2BNiG3ZymbcpxfAP8g2yjG2HG9QliNMJA+OwH0RZgQFNUEP6FYF6eo5jqB1Vy6gg3am47ZlBMbagMfuxBmM/Ri3XQ7Q5pO4D4fyZ8/1x+2rCsyn0mKOjqWAjiAkQKaAiONhSCQFIa2/aVPfL90IKMoxbCER1d9f3ag/P7/xCR828KZeQXyc7GhJ/UQ8B0G+qcUzh6v+jCM3US2oKRXBAd3LuH3VUW3dRQV00UT6Fty2jERqxh+Ax+5IeG/FbZeDlkg/berDmSv5rSFx+yoTieuNFvN0ZNgCOiPF+rRNjd2OQzVxDnx9xcTfbIX79YocY9fiDUpBPgr+CfdHfI+goCboAZ0v4bWbCFp35QI68HTctozA+lxiMXbbtjAm7Dm7PKSGNtfgPhx6H25fdaA+iy3m6UgK6AhCAmQLiDgehURSENL6mzb1/dKNgKJcwxQSUf391Y368/Mbn/BhA2/qFcTHyQ6/5RyM+xs8D2Em02+BZ+w4rG4b3HfYUS2oKRWRAV00rl+E21cd1dZdVEAH7wM6bltGKuL670xjd2A0kTkVt+2UrmetCXqfTjfg9lWnXzzzR/M8nRm2gC6fjTWaQzVhfpOvr/wd7tNLmhirwRuUgn0CHJ1ibBPcd9gRFNQEOqCD12cVbl91BK27cgEdtHMQbltGYKyn47E7VNizoVsY29eifUfCOvwFt686MK/z8TydSgEdQUiAjAERx4OQSApCWn/Tpr5fuhFQiDAsIRHV31/dqD8/v/EJHzbwpl5BfJwKRBLpm/A8XPBL8LqKuB5nVSnayGLqBTWlIjKgC9rtLTmqrbuogC6S1C/AbcuIlpjwEzx2Z2bOxG07pV9SH2Bu36lpN/7N7zPGRjC3dvNc7RumgC6fjV1mEaqJsz52Bu7Ta1oZ2zjH2BK8SemC/FaaE5sY8zWQlAlBQU2QA7ov4PUZqKuZOYLWXbmALqfIFbUwzr9ZjN2Jf8BtOwXaOtOifSd+1cjY1rh91YFzag+LuTqSAjqCkABZAyKOyyGRFIS0/qZNfb90I6AQZRhCIqq/v7pRf35+4xM+bOBNvYL4OBWIJMb/H4x9PZ6Li34YSeizIvHMIbvHM9vi8YQF1YKaUhEY0H0RxFukqrbuogI6fiUtbltG+g/W+5vG7sBoIv133LZTokn9WNy+I+P6Z0H9BQmo0bWm+TowLAFdezZ2vClQE2h7fawV9+kXzbHnSKQAAGZBSURBVIydijcpXfadJsYmtTAWr2dsczyesCAoqAlyQDcXtx0EBK27cgHddMYG4LZlBMZ6JB67E2Gdhf0CHbQ1E7fvRFjLh3HbQQHm9xaerxMpoCMICZA5IOK4GBJJQUjrb9rU90s3AgqRBj0kovr7qxv15+c3PuHDBt7UK4iPUwUtkbkVz8Uj14IL+O3vtGp9cMXQVF88tqCiWlBTKqICumg88yxuOwiotu7CArq4fhpuW0YqkuOiprE7UGhAl9DTuH2HBu7ZLAUiyfTRFvO1bRgCurb6fYbks7E1OFQT6Bsf1O0nzW2tZzG2RY6xd/BGpUd+AT4IXgBW8iv68PiCiqCgJsgBXT1uOwgIWnflAjqY9+9x2zICYz0Mj92JIgM6aO8/uH0n8nMTtx0U4HU6A8/XiRTQEYQEyB4QcVwKiaQgpPU3ber7pRsBhWiDHBJR/f3Vjfrz8xuf8GEDb+oVxMepQkU800/rvA2laU4e+42WyMzVEul/aPHMr/E4g4RqQU2piAro+K1XcdtBQLV1p4DOmSIDOlG/QAGvvUtw20GhX01mLzxfJwY9oFtWX7lPPhv7wiJUE+Vny+sH/hL36zfNjI3AG5U+uQpsBY9rhB8HeJxBQlBQE9iArokxJX4m2kXQulNA5xL8GW147E4UHNC14/adCHMbhNsOClDvE/F8nUgBHUFIgAoBEceFkEgKQlp/06a+X7oRULhhUEMiqr+/ulF/fn7jEz5s4E29gvg4lYgk9LF4Pn4LY1oSSWSu6hfP/BFedhvhMauMakFNqYgK6MBpuO0goNq6U0DnTLEBXfoZ3L4TI/HMENx2YBjTujHM8Ss8Z7sGOaBbNrlyz3w29r5FqCbK9e3ZypG4X1nIMXYD3qz022bGFjUxdtF0xn6Gx6s6goKawAZ04KG47SAgaN0poHMJ2QK6riuc1+P2nQhtbYfbDwotjO2N5+tECugIQgJUCYg4gkMiKQhp/U2b+n7pRkDhlkEMiaj+/upG/fn5jU/4sIE39Qri41Qjmkjfguckke1aMjMxmhz/KzxuFVEtqCkVUQEdv60fbjsIqLbuFNA5U2xAp7+L23fmhJ/gtoMEzPFl85ztGdSAbnn9/ju1Z2NvWYRqIk3hfmWinrEf5Bh7DW9YSuRLzYydB+6Kx64igoKawAZ00E4gf2FC0LpTQOcSsgV0/Epi3LZDV+K2gwQPHy3mbFsK6AhCAlQKiDgCQyIpCGn9TZv6fulGQOGmQQuJqP7+6kb9+fmNT/iwgTf1CuLjVGPP2vrNYR5z8Lzks+OKkuP7V6W2wHNQBdWCmlIRFdBpicyluO0goNq6U0DnTMEBXdlXhoHrBgxo3hS3HSTg3LrHYt62DGJAt2Lib7Zqy8YWWgRqwmyvj7UaBpP+KvcZcDrlGFuCNy0lcw14P3ggHr9KCApqAhvQBfV2fILWnQI6l5AtoON1w2079EncdtCAOX5kMW9bUkBHEBKgWkDEERQSSUFI62/a1PdLNwIKtw1SSET191c36s/Pb3zChw28qVcQH6ciFUNTfWEu8/DcJPX9SEK/oqJq0o54HrKjWlBTKqICuqA+L0u1daeAzpmiArqu92NT+w58H7cdNGCOMyzmbcugBXRG65iN27Ox+3GgJthFyyZXbon7lpVmxvbIyR/SFeRX/I1tZWwzPA/ZERTUBDagg3YOwG0HAUHrTgGdS0gY0CVx2w69E7cdNHICrgCngI4gJEDFgIgjICSSgpDW37Sp75duBBReGJSQiOrvr27Un5/f+IQPG3hTryA+TlmqUptoAjY7PfRTLZG+fPd4Zls8FVlRLagpFQroiqPaulNA50xRAV2kOr0DbtuJkYT+Km47aGjJtI7nbdegBXT5bOxqi0BNnPWx996btl8F7ld2mhj7ETgfb17KajNjS2G8x6YY2wTPRVYEBTUU0CmGoHWngM4lZAvooK3huG0nNnVcIB1sYJ7/xfO2KwV0BCEBqgZEnDJDIikIaf1Nm/p+6UZA4ZVBCImo/v7qRv35+Y1P+LCBN/UK4uNUh2+qa2Jur+aVH0TimaPwPGREtaCmVCigK45q604BnTNFBXTRRF0Fbtuhi3DbQYPfFtdi3rYMUkCXz8YuMwVqYv102bTKX+N+VaGesc2bGcvhDUzJfamFsX3xXGREUFBDAZ1iCFp3CuhcQraADtoZg9t2Iswri9sOGjDPOXjedqWAjiAkQOWAiFNGSCQFIa2/aVPfL90IKLxU9ZCI6u+vbtSfn9/4hA8beFOvID4uCFQk0z+HuS3Ec5XceXyjHc9FJlQLakqFArriqLbuFNA5U1RAt1vyyh/jth2Z1AP/fBaY5/mmeds0KAFdfmrlcRaBmkhX5+sHJnG/KtLCWG0zlAxvZErsuibGpk9mTOrbigoKaiigUwxB604BnUvIFtBBW0fith06AbcdNGCO91rM25YU0BGEBKgeEHEchkRSENL6mzb1/dKNgMJrVQ6JqP7+6kb9+fmNT/iwgTf1CuLjAkNVapNIPH0WzPFjPGeJ/UBLZmrxVGRBtaCmVCigK45q604BnTNFBXS7Dhq3G27boU/htoNGJKlfYDFvWwYhoFuWHXhgPhtbYxGqiXJ9WzZ2NO5XZaYztn2OsXpwNd7QlNgXYdw/w3ORBUFBDQV0iiFo3SmgcwnZAjpo53DcthNhXpNw20ED5nk/nrddKaAjCAkIQkDEcRASSUFI62/a1PdLNwIKP1Q1JKL6+6sb9efnNz7hwwbe1CuIjwsaWs3kH8I8s+C3eO6Suk7WYEC1oKZUKKArjmrrTgGdM0UFdP2rJuyC23boc7jtoKEl9ZTFvG2pekC3fOq+e+ezsS8sQjVhtk+NXYj7DQo88Moxdg/e1JTYT8E/4XnIgKCghgI6xRC07hTQuYRsAR2M52DcthNhPNNw20EDavUwnrddKaAjCAkISkDEsRkSSUFI62/a1PdLNwIKv1QxJKL6+6sb9efnNz7hwwbe1CuIjwsqFfFMPy2RbtAUeT5dNK5fhOfgN6oFNaVCAV1xVFt3CuicKSqg6zdE3x637dC3cdtBI5LIXGUxb1uqHNC9V7/vHvmpsZU4UBNrZQPuN4g0MbZPM2N35xhbjzc4JfQrfptOPAe/ERTUUECnGILWnQI6l5AwoDsIt+3Qa3HbQQPm+LTFvG1JAR1BSECQAiKOjZBICkJaf9Omvl+6EVD4qWohEdXfX92oPz+/8QkfNvCmXkF8XNCJVKd36Lqt2Lu4FrIZTejn4PH7iWpBTalQQFcc1dadAjpnigroBgxo3hTaW4/bd+CnuO2gEU3q11rM25aqBnTL6/ffKV8f+585UBNofexmI8X64L6DTAtjP+VXZuQYW4U3OiXzqybGqvD4/URQUEMBnWIIWncK6FxCwoCuCrftRGjn37jtoAHzfBvP264U0BGEBAQtIOKUGBJJQUjrb9rU90s3Agq/VSkkovr7qxv15+c3PuHDBt7UK4iPCw2pVB/+vLdIXL8Z6vA1roskro8m9MPw0P1CtaCmVCigK45q604BnTNFBXQcTdCzP/esTf0Atx0kokn9YTxnu6oY0LU3D+jbno09bQrUhFo555XUXpvhvsNCM2N9waNzjD2KNzwl8tMWxvbCY/cLQUENBXSKIWjdKaBzCdkCOv6ehdt2IoxnMW47SBiMbQTz/BLP264U0BGEBAQxIOKUEBJJQUjrb9rU90s3AgoZVCUkovr7qxv15+c3PuHDBt7UK4iPCyM7DqvbJpJMHw31uB/8BtfIV+P6Z1qNvicesx+oFtSUCgV0xVFt3Smgc6bggO4N3L4TK2rGK7GB6BQtkXkHz9muqgV0RuuYjfPZ2L3mQE2oT62Y+JutcN9hZQa81JoZOyfH2JN481MCX5rF2BZ4zH4gKKihgE4xBK07BXQuIVtAdw1jO+G2HfoFD7Fw+0FhGvzz1GLOtqWAjiAkIKgBEaeXkEgKQlp/06a+X7oRUMiiCiER1d9f3ag/P7/R+R468KZeQXxc2NmpKrV1NKkf2nXrsZW4Xn4YSej/xeP0A9WCmlKhgK44qq07BXTOFBzQPYbbd2bmSNx2UKgYmuoLc1xrnrM9VQvo8vWxFotATaSvtDXsswPul+iEb5o2M3ZijrF7cwKubhBkBo/TDwQFNRTQKYagdaeAziVkC+hSjPWB9r7F7TsR5tYftx8UYG6D8HydSAEdQUhAkAMiTpGQSApCWn/Tpr5fuhFQyKTsIRHV31/dqD8/v7ud66EEb+oVxMcRG9IvqQ/Q4pl/aonMXM3Hq+uiyfQIPDavUS2oKRUK6Iqj2rpTQOdMoQFdXJ+J23cinJtTcNtBIVKdrsTzdaJKAV2+vvJSi0BNpEvb6gdquF/CmnrGNuebqKDezNgivDHqoV9B//3w+LxGUFBDAZ1iCFp3CuhcQraAjgPtvY7bd+ho3HZQgLmdazFf21JARxASEPSAiNNDSCQFIa2/aVPfL90IKGRT5pCI6u+vbtSfn9/dz/8wgjf1CuLjiJ7puOIhmanVEumGSEJfgmvpss/h8XiNakFNqVBAVxzV1p0COmeKDej4LzWY+3DgQtx2UIgk9bMt5mtbVQK6fHbg3ywCNZG+n5+6389wv0Tp8Fu3NTN2eI6xG8CP8UapmzZ3XNznL4KCGgroFEPQulNA5xKSBnT8CmRTH3aFMU3GbQcFmNtteL5OpICOICQgDAERxyIkkoKQ1t+0qe+XbgQUMiprSET191c36s/Pb3z+hw28qVcQH0eUjlY9/jdaIn051PE1XFc35Ffz4TF4iWpBTalQQFcc1dadAjpnCg3oEvpI3L5D1/Ybom+P2w8C8Pq612K+tlUhoFs+LVabz8bWWIRqovysfdo+f8D9Es5pZWzjJsaqwOk5WEK8aeqCH8/y+Vl0goIaCugUQ9C6U0DnEjIGdDCmSbh9h76I2w4Cqc7bgH5oMV/bUkBHEBIQloCII2NIF8b6b7/99qZNfb90I6CQVRlDIqq/v7pRfzjVQ/Oe1hN4U68gPo5wRsetMJPpei2R+QTXWJTRuO7rb1qqFtSUCgV0xVFt3Smgc6bIgK4inumH23du8J5Dp1VO3hLm9rl5rvaVPaDjwVk+G/vcIlQT5Tdt0/YZhPslxMHDuhbGanPwx5ygZy9Z2cTYwbhvLxEU1FBApxiC1p0COpeQMaCD9g7D7Tt0/QzGfozbVx34ebGvxVwdSQEdQUhAmAIijmwhXRjrv//++0sTUrgRUMisbCER1d9f3ag/nOahek+zAm/qFcTHEeXBN14jicxJUNs3cK0F+Aruz0tUC2pKhQK64qi27hTQOVNkQMeJJPQVuA8nwuvPs41Pr9Di6VF4nk6VOaBbOWXv3fPZ2AqLUE2Ua9unDvQ11AkbjYztkmPscvADvJEqwGbcn5cICmoooFMMQevu2c8pCuicKTKgg/fBPXH7ToX5/QO3rzpQ68l4nk6lgI4gJCBsARFHppAujPXnV9DJElK4EVDIrkwhEdXf/HUvdaP+cIqH7j0Ngzf1CuLjCEGkUn34lR9Q46W45mW4vn/VhF1wV16hWlBTKhTQFUe1daeAzpkuBHR34z4cunq3mvG74vZVRmBtpA3o2icN2DGfjb1pEaoJs31q7ATcL+EN1zG2VRNjF+UY+xxvqDqVByW4Hy8RFNRQQKcYgtadAjqXkDGgg/88b5QTdAtH8DXeHu5DVaDOm+YE3haZAjqCkIAwBkQcWUK6kNZfmpDCjYBCBan+/hrk+vPzG5/wYQNv6hXExxFi2Tk5cSstkbkV192xyUwt7sMrVAtqSoUCuuKotu4U0DlTdECnxTOn4z6cm7kUt68qXeu1xjxHZ8oY0LU3D+ibz8aewoGaSNuzlRfgfgnvaYa33BxjL+FNVYeuh/a2xX14haCghgI6xRC07hTQuYSMAR0H2rwZ9+FUmGNgbtMMdR6D51eOFNARhASENCDqQIaQLqT179jIlyGkcCOgUEWqv78Gtf78/MYnfNjAm3oF8XGEC4xp3VhLpB/EtXdkPPNP3LxXqBbUlAoFdMVRbd0poHOm6IAuUp3+Ke6jDD/gv+yA+1AReL+ZYjE/x8oW0BmtYzbOZ2P34EBNpO3ZmK/PYyU2pImxHzUzthRvrDoR2vkjbt8rBAU1FNAphqB1p4DOJWQN6GBcx+I+yvAh3L6qQF0WWszPsRTQEYQEhDQg+g6/Q7qQ1v+7zXy/Qwo3AgqVpPr7axDrz89vfMKHDbypVxAfR7jDzsmJP9IS6Y9w/e0aTaRzuG2vUC2oKRUK6Iqj2rpTQOdM0QEdB9p9G/dThufj9lWD36IY5vGlxdwcK1tA15atbMaBmkjb6wdeaxjBuR1YUGhh7M94Y9WhR+K2vUJQUEMBnWIIWncK6FxC4oAuAu2uw/04FdrbD/ehGjCPA/G8ypUCOoKQgJAGRBvgZ0gX0vpvsKHvZ0jhRkChmlR/fw1a/fn5jU/4sIE39Qri4wj3iCYzF+P62zau+7b5o1pQUyoU0BVHtXWngM6ZbgR08NpJ437KcBX/RQfch0poiczVFvMqS5kCurb62MU4UBPsvUaqahPcLyEHOcYexJurDvQtiBcU1Pj2b7SeoICuOILWnQI6l5A1oONAu4/jfpzKrzwzFH4WXStjG+fE3e74OymgIwgJCGlAZMKvkC6k9Tdt6vsVUrgRUKgo1d9fg1R/fn7jEz5s4E29gvg4wj0q4pl+uP4OfAq36xWqBTWlQgFdcVRbdwronOlGQKfFM7/G/ZTpdbgPVdCS6YEw/nUWcypLWQK6tvrYMflsbL1FqCbK/y6bXLkl7peQhxxjh+HNVbs2MTYJt+sVgoIaCugUQ9C6U0DnEpIHdGNxP+UIcz0J96EKUN+z8HxESAEdQUhASAMiS/wI6UJaf9OmPtePkMKNgEJVqf7+GpT68/Mbn/BhA2/qFcTHEe6iJTLv4DWw6Qu4Ta9QLagpFQroiqPaulNA50w3AjoOvOc9j/sqx0gyfRDuQ3b2rK3fHMb+Cp6LCGUI6NqzlYPz2dhqi1BNlIuXTKnaDvdLyEXXLd9MG6w2bcTteoWgoIYCOsUQtO4U0LmEzAFdA2M7QNtf477K8FMYZz/cj+zAmPeAsX9hMZ+ypYCOICQgpAFRj3gd0oW0/qZN/YJehxRuBBQqS/X31yDUn5/f+IQPG3hTryA+jnAXLZGZi9fApm/gNr1CtaCmVCigK45q604BnTPdC+j043FfZfpBpLZOw/3IDH92qMU8hOh3QNeeHTggn419bhGqCbEtW/n2+41774L7JeQkV+ZmdTNjM3GbXiEoqKGATjEErTsFdC4hc0DHgbZn4b7K9EkY66a4H1mpZ2xzGO8ii3kIkQI6gpCAkAZERfEypAtp/U2b+t31MqRwI6BQXaq/v6pef35+4xM+bOBNvYL4ONXZqSq1df+qCbvgz8tCNKHfg9fApq/gNr1CtaCmVCigK45q604BnTNdC+gqJ2+pJdIf4f7K9CneLu5LRrRk+jiL8QvTz4BuxbR9fpzPxlbgUE2gy9+r33cP3G+Y4VdXyLx5m2PsI7zBatNm3KZXCApqKKBTDEHrTgGdS8ge0PE64r4E2IT7kRUY6zUW4xcmBXQEIQEhDYh6xauQLqT1N23qY70KKdwIKIIg1d9fVa4/P7/xCR828KZeQXycymiDxkVgTs+BL/avSkl5OywtnnkEr4FNF+I2vUK1oKZUKKArjmrrTgGdM90K6DhaIn057q9s45k72ZjWjXFfMqFV64NhrKtNYxeoXwFd+6QBO+azsTcsQjVBDvykrX6f3+J+w0wLY/s2MfY+OBv+8bYR/roM5Bj7Bm+w2nQKbtMrBAU1FNAphqB1p4DOJWQP6DjQ/jzcnwDPx/3IBtT0MotxC5UCOoKQgJAGRCXhRUgX0vqbNvWt9CKkcCOgCIpUf39Vtf78/MYnfNjAm3oF8XGqEkmM/z+YT1u3uc3fcVjdNvg4v4kk9CV4DezIr8DDbXqFakFNqVBAVxzV1p0COme6GdDxX5iAPj7GfZZrJK7fLGtIF41namCMX+Exi9aPgG7Z5Mot89nYAnOoJsyv2uv32R/3G2a6Nqm/u30k/H06PsZvZsA/NfHmqgN925QWFNRQQKcYgtbds3+bU0DnTDcDOhjjfrg/EcKYz8J9yQKM7UI8XjekgI4gJCCkAVHJuB3ShbT+pk39nnQ7pHAjoAiSVH9/VbH+/PzGJ3zYwJt6BfFxKhKJZ4bAXD7Hc4sm9CdkupKOPzcJxrUej9Oe6QbcrleoFtSUCgV0xVFt3Smgc6abAR0HzrOLcZ+CvG/n5MStcH9+oiX1I2Bc31qMVbheB3RG65iN8/Wxuy1CNVGugfaH4n7DTBNjF+UYW483LsGmFGN98PF+AeM51GKMtuSb8bhdrxAU1FBApxiC1p0COpdQIaDjQB8P4j5FCOMeh/vykxT8zIExTcbjdEsK6AhCAkIaENnCzZAupPU3beoX082Qwo2AImhS/f1Vtfrz8xuf8GEDb+oVxMepBt9Eh3msxfPq5ov81pf4+/yAb4BbjM+eSf1E3K5XqBbUlAoFdMVRbd0poHOm2wFdxdBUX+hnKe5XkC9oNfqeuE/PqUptAq+bCVrZv4hRul4HdO3ZWJNFqCbK9W3Z2NG4z7DSzNimOcZm4Q1LZOssxrbA3+sHMJbbLcZnyxbG9sLteoWgoIYCOsUQtO4U0LmEKgEdf++CflbjfgV560zGfL8rTANjO8BY7rcYn2tSQEcQEhDSgMg2boV0Ia2/aVO/N90KKdwIKIIo1d9fVao/P7/xCR828KZeQXycMoxp3RjGn8XzsTSpL69IZvbFTXjJnrX1m2sCNqf5rTxx216hWlBTKhTQFUe1daeAzpluB3QcqOnBuF+BrorG03/DfXpFRTL9cxjDAotxuaqXAV1bNnaRRagmTGjf9degKkxnbPscY4/gzUormxhbyG8vidvwkmbGfg5jWYPHZtNVrYz5dstaQUENBXSKIWjdKaBzCVUCOg70MwX3K0r+OuW30sR9egWMYTD4Hh6X21JARxASENKAyBFuhHQhrb9pU78U3Qgp3AgogirV319VqT8/v/EJHzbwpl5BfJwK8NuZaXH9XjyXXlwTSeoX+PW8okg8faXFmOy6El7KG+G2vUK1oKZUKKArjmrrTgGdM70I6DjQ179x3yKNJvWH+1Wnf4n7dYudqlJbd72/f4PH4oVeBXT8yrZ8NrYeh2qihPavXz4l1l956/ffCdfOLtMZ2z3H2Gt4o7IXPwSH47a8IMVYH+h7rsWY7HonbttLBAU1FNAphqB1p4DOJVQK6OoZ+wH0tQz3LdB1YBPMZUfct1s0MLYbrMFsi7F4IgV0BCEBIQ2IHCM6pAtp/U2b+qUqOqRwI6AIslR/f1Wh/vz8xid82MCbegXxcbITTdRVwLifw/Ow4cKKuP473K6bRJPpEVrx23CW6gzctpeoFtSUCgV0xVFt3Smgc6ZXAV3Xe/jHuH/BruGv293idb/A/Yti93hmW+jn/EhCX2HRv2d6EdDl6wcm89nYahyqkWbb62OtuH52aGFs3ybG3seblKXazNhNjYztgtt1E+izDo/DidDO0bhtLxEU1FBApxiC1p0COpdQKaDjQF+JnPUzQ0W6CuYz7hrGyv6FkJ6AuveHfurBry3690wK6AhCAkIaEJWFyJAupPU3berbUWRI4UZAEXSp/v4qe/35+Y1P+LCBN/UK4uNkJlKT+S2MeRmegwPXRRL6rOigCXvgPkSjJfUjNEFXVkRr0vvj9r1EtaCmVCigK45q604BnTO9Cug4WjxzOO7fRedFEum/9q9KbYfHYZcBA5o3jSTGV/OfH9DulxZ9ea7bAV37lNjv89nYZziIIq0tJ6DLMXZoTsxm6OdNjKX4bTJxHyLht6OEvq6y6N+JX/j9jCVBQQ0FdIohaN0poHMJ1QI6DvTXiPt3ya/5L2XAx8H1jG2Ox2EX/h4MbR0GNf83fFxr0Z/nUkBHEBIQ0oCobESFdCGtv2lT366iQgo3AoowSPX3V5nrz89vfMKHDbypVxAfJyuReGYIjPdzPP4y5VdZtFbE9TgTfOvISHV6By2uz7To06nP4T68RrWgplQooCuOautOAZ0zvQzoOFDfZjwGl10NPg6v5zS/qrnj6rpUqg8eV3cq4pl+WjJTqyXS/9CS+l0w5s8s2vVVNwO6FdP2+XE+G1uOQyiyZ50GdE2MXZQTf9XFp9BuFvwV7q9cGhn7BbT/X4s+HdnM2DTch9cICmoooFMMQetOAZ1LqBjQzWJsC+hvER6Dy34O3gOe38JYLbxHR/G4utP1Cxa/hHGOgBqPhz8/Cn5r0a6vUkBHED5RU1Pzw+rq6q/55mw3v+afx8cSPeM0pKP6lx/QcZ2GFO8sWGwMHjz4u9rXJJNG+7Ovm44ji0v191dZ68/Pb3zChw28qVcQHycjWlI/URNzi8hi5iNxvVFLpA/ccVid49/i1mr0PbW4PkFLZD6x6MOxkYQ+BvflNaoFNaVCAV1xVFt3Cuic6XVAt2dt/ebwfvs0HocPfgjvr0u6C59rtzhOtEJu8+lWQGe0jtk4nx34Og6gyOI6CehyjM3Cm5Iu+BJ4eTNjf4QBbobHUCr8FpzQzvXgGos+nPotv50a7strBAU1FNAphqB1p4DOJVQM6Dj8PS3X+WxQ01i8FOadh49LkB/j41zwI4vP2ZYCOoLwkerq6hsGDRr0edcm7TfxePxGfAzRO05DOviem8DVXfX/BL7/enxMgDFt6jvVaUhxwdnnfhdQ1A4ebPo6WZpO63/u6Wd9V/+DDhxi+jpZmk7rf8bYU7+r//Chw0xfL0d+fuMTPmzgTb2C+DgZiSTTB8FY1+Gxu2pSXxyJ6zfz0AX6P1qL61XRRN0f+g/W+xesGFz3M/75SEIfC98zDXzF1I4YFzDBV/g5QbWgplQooCuOautOAZ0zvQ7oOF3Po3sXjyUErufBmsXnbetWQMdpy8YewwEUWVwnAV3X1QumjUkX/Rr6XAgfW5oZO49vgINV8Odf803lgvD138LHGvh4JnztOvi4zKItEV6Fa+IHgoIaCugUQ9C6U0DnEqoGdJwWxv6cE3PbYtX8RNT7DgV0BOEj1dXV+4MfFQIi/nd8DFEaTkK6rvp/2lX/D0NWf9Omfjk6CSn+c+cDxoG1tR0BxdhjTzB9nSxdp/UfXFPTUf/TTzrF9HWydJ3U/7Hb/20kE4mO+p9z2pmmr5cjP7/xCR828KZeQXycrGhJ/Tw89pC4uiKu/w7Xww9UC2pKhQK64qi27hTQOdOPgI7Trzr9Sy2R/giPJ8jyq7Uj1emf4s870c2ALj+18nAcQJHFdRLQpRjrk2PsTrwxGRLfrWfsB7gmfiAoqKGATjEErTsFdC6hckDHgX7HQP/r8HiCLMz5cFi3Y/DnnUgBHUH4THV19TtdAdFS/DXCHjikSyaTP8LHYOD497rq/xb+WsAxbeqXKw4p3pr/vOkY7OiRozoCiqsnN5i+RtrTSf1HjRhhxONx49ppLaavkfbE9X/36ZdNx2D5lXM8pJs9fabpa+XIz298wocNvKlXEB8nM5GEPgmPP/DG9ctwHfxCtaCmVCigK45q604BnTP9Cug4WjI9EMawCo8pmKaf6V+V2kKFgO5/9bWb57OxD3AIRfask4COM4uxLXKdzwAybVAGWL5pncC18AtBQQ0FdIohaN0poHMJ1QM6DszhhFx4QrqmrjlTQEcQQaC6uvrMQYMGrY/H42fhrxH2KYR0yWTyDPj4SVVV1S74mO7AMX/n9efrgL8WcEyb+iIshBTXNcwwhh001Hjzv8+ajunujCkNHQHdgvvmmr5G2ve7+k9rKan+0ydMpfoLtFD/zKVXllT/KVdmXKk/P7/xCR828KZeQXyc7EQTehOeQ1CNJNIPsVSqD66BX6gW1JQKBXTFUW3dKaBzpp8BHSdare+tCXoum8Qu1QaNi/D5qhDQcfLZyok4hCJ71mlAx5nJ2DZNjM3HG5QB9nJcAz8RFNRQQKcYgtadAjqXCEJAx4F5HJsLfkj3ENR50675UkBHEEGgpqbmh/F4fBX/iL9GOKMrnPsWXAvm8Ne7E+L6mzb1RcnDOVgDowaccNk409e7+86CxcaQAw802p993fQ10pk8nONXZfHn+k28PG36enep/mJ9ad4C45gjj4La13ZcHXfVFf7Un5/f+IQPG3hTryA+TgW0uD4BzyNoRhL6q1rNZKl+DqsW1JQKBXTFUW3dKaBzpt8BHUerHv8bGMsyPLaA2M5DucJcVQno2hr2+Wk+G1uPgyjS2nICOs51jG3VxNjDeJMygN6aYkyaX0DiCApqKKBTDEHrTgGdSwQloOPAXA7JBfSZdPyXSxoZ27rbXCmgIwhVicfjuw8aNOii6urq/3S7xeK74KPw+fPB3fD3EKVRW1u7E9Txi66acr/CV9FR/TswbeqL8O0nX+gIJwr158+Yw1cRvfzIU0bzpHrj1BPGGoccPLrjuENHjzFOP/HkjltdvvrY06Z2ydLk9R88ePB39efhD9XfG5+b85+OUK57/Q8aMqSk+v9lzKHGP84427j3ulZjxQtvmtq2Kz+/8QkfNvCmXkF8nCpoyfQZMP41eD4Bcemug8ZJ93NXtaCmVCigK45q604BnTNlCOg40URdBYznOTw+xX2vezjHUSWg4+SnVs7DQRRpbbkBHQca2CzH2Cy8URkg5/JbeuJ5+42goIYCOsUQtO4U0LlEkAI6Tgtj+8Kc3sfjU1mYz2PdwzkOBXQEoSD8mWiDBg1qSSQS7489/ez5s+955OVHnn975fPvfW7Axw+uu2veayeddvbT8UTiQzhuelVV1Y64DaJ3amtrNajfnHg8zq+g47evbOCfL9Qf6vvBKWee89RN9z32xiPPv/NhV/1XXHfnQy+deOqZT8L3fRCC+ps29UX5ymNPG6efdIoBr/MOJ6bGd3yePxNt/EWXGaNGjDTGj5tg3PngE8bjz79jPLtklfGf59827rz/MePKy/WOWwTql1zRETbhtsne5VdxnXriWCMRT3RcSTep6yq67vUfN26icdccVP/7HjXGwbFUf+e+s2Cx0XLVtI4a8tc+D6sLV9EV6j+yq/53PjjfeKyr/o89+5Yx+5Z7jJNPPMU46dgTjBfnPWlq2478/MYnfNjAm3oF8XEq0S9Zl4A5fIDnpLhv9x+s98dzlQHVgppSoYCuOKqtOwV0zpQloONUDE31jST0WXiMapp5noeOeI4qBXRt2dghOIgirRUR0BVoYuzsHGOr8Yal4s6ZzNiWeK4yICiooYBOMQStOwV0LhG0gI4DY+kHY1qAx6iiMJeb6hnbHM+RAjqCUIwDDjhgQDwef++Mc/65YMEbK77moVBPzn+t/dvTz/nnIn58IpH4P9wWURr8yrlBgwbdDX4GJnk9z/rHxc8ueH3lalzzDer/6vIvTznzvAX8qroA19+0qS/a1/+zyDjn1DM7riJ6+Ja7jUNGjzEydVOMBW+sNNW8uwtebTf08RM7rup65oHHTO2Spfna488YZ518mjF0yEHf1V+vm9p7/V/JG5nxE6j+ZfrIbfcZJx5znDH0oO/rn56Q7bX+uX/dYowaOcqYf/dDpjZLlZ/f+IQPG3hTryA+TjU6r7bIzMXzUtSndk5O/BGeoyyoFtSUCgV0xVFt3Smgc6ZMAV0BeO0dA2P7HI9VGZPp2TxsxPPiqBTQvZLaa7N8NrYCh1GkWZEBHaeFsb1zjL2FNy0V9ZoUY5vgOcqCoKCGAjrFELTuFNC5RBADOg6MZ1OY26Scus+lWwOej+dVgAI6glCIrnDugxk33fMy3owtZstN97zJvw++/7e4TaJ0qqurE1DHT2a23vcOrnExm2ffsRi+7/2A1t+0qe+Wc1vvMYYNHWbceOfDphoX85bb7u+42mvx3PmmNsnSdVr/1lup/iJ8FF7HI4YNN2bfNddU4568+d//MUZA7Z/596Om9kqRn9/4hA8beFOvID5OTYyNIgl9rJbIfILnp4qRZHr6nrX1pt9AlAnVgppSoYCuOKqtOwV0zpQxoOPwK4ojifRDeLxSG9c/4+Einkt3RAV0/GcfbtsN8lNjGRxGkWZFB3Qc/ly6HGNX5To3RE0bmAr4TTNjrgfJ5SIoqKGATjEErTsFdC4R1ICuAMxvPxjfG3i8kvsOWInn0h1RAR208yvcNkEQAqmurt4ZXGY3nCvYfOOdb8H3L6mpqfkhbpvonUL9Z7bevxTXthSbrrvtBfj+dwJYf9Omvhv+74nnjDEHj7YdDhW85l+3GCccc6yQ53KF0UL9Zzus/6xZN1P9y7BQ/xtshHMFr71jjnHoIX/puG0mbrc3+fmNT/iwgTf1CuLjVKZ/1YRdtLg+E+a1Fs9TYtujifQwPBcZUS2oKRUK6Iqj2rpTQOdMWQO6ApF45igYZxset3TG9Xv5muDxY0QFdFoyfRxu2w1WTtl793w2tg4HUuSGuhHQFWhi7Hc5xubhDUyZbWZsEfhrPBcZERTUUECnGILWnQI6lwh6QMfhz+SEeaZgnF/gcUvmGqjjZP5LI3gOGFEBHfT3c9w2QRACqa6unsFva4k3X+0I378Y2snitone4fU/54LUi7imdhx7+t/57S6DVn/Tpr4b8mdu6XVZU03tePxxJxjzWu8xtU32Lq//+Anl1f+EY6n+Th13UQrqX2+qaalepk82MlfopnZ7k5/f+IQPG6ZNvS7xcUGgX01mr2hCb9XkDupWR5OZKf2rUtvh8cuKakFNqVBAVxzV1p0COmfKHtBx+O0i+XkG412Fx++3kYT+qpZIH4jH3BMVg+t+httwYiSZPhq37Rb5bOUcHEiRG+pmQFcgx9hg8Em8kSmZH4GnpyS+pSVGUFBDAZ1iCFp3CuhcIgwBXYFp8E8DGOs1OTmvlp7TwtheeMw9Aet2rEUbtoV12wO3TRCEIKqqqvZMJBIrF7y58iu88WrH+a/mV8cTiY/i8Xg/3AfRMx31r6n5YMGb76/FNbXjE6+2fwa1/zBg9Tdt6ov2xblPdjxL66k33zfV1I6NM2/sCDpw+2RxRdV/ekf9LzO1Txa3s/4Hl1X/+a+2G8OHjzRefWyhqf1i8vMbn/BhA2/qFcTHBQktMeEnWiLdoMm1mbsmmtSv5WPD45Ud1YKaUqGArjiqrTsFdM5UIaArsHs8sy28t/9DS+rL8Tx88DV+dR8b07oxHmcxtHjm1xZt2TeuH4zbdov2+tgoHEiRG+pFQFcgx9ifmhm7DT6uxpuaPvopeOXVjCl3px1BQQ0FdIohaN0poHOJMAV0BRoY2w3GXA9+hefhg/NaGPszHmNvwPeNtWjLto2M7YLbJghCENXV1ReefMbf5+NNVyeefMY5zw0aNOhc3AfRM7z+Z5x7wfO4lk48buxpTwSs/qZNfdG2XDXNGJeeZKqlXe969NmOq+hw+2Rxef2v1K8y1dOuvP78KjrcPllcXv8r9Mmmetr14nETjRn1OVP7xeTnNz7hw4ZpU69LfFwQ0Sonb8mvMtAS6QdhzqtxDTzyfS2uT6iIZ5T9xRbVgppSoYCuOKqtOwV0zlQpoCuw15jUZlo8PQrGf5/m7RXTqyMJ/e7O2xMbG+FxlUK0Wt/bol3b9kvWJXDbbmGkqjbJZyvbcShFfq+XAV0BvoHZzNh5TYwtxJubHvo6jOEscFs8PlUQFNRQQKcYgtadAjqXCGNAV2AWY9vBuE+G8T+D5+Oyn4Et4B/wmEqF/zywaNe2kxnbErdNEEHmj2BVl327ff5H3T6/e7fPl0V1dfV/b7h73it4w9WJN9w1b8mgQYMew30QjNXW1m6OP8fh9b/p/sfbcS2deO3tD/KA9DHch8KYNvWd2v7s66bPcU85/iTj9jnzTbW06xOv5I3hw0eY2ic7LVb/26j+rttT/U8+4STj1jlPmupp11se+K9x8thTTe0Xk5/f+IQPG3hTryA+LuhoNZN/qMUzh3eFDstwPQS7MpLQZ0WS6YMGDGjeFI9FNWA+x3fMpxzjeiNu12/4s6BM43SgKs8StItq675LbXonU/8OrIjrcdy2jPD3NDx2J0bjmRrctkrwdYf39WO0ROZ2eM1+avF+XK48AFwQjafP/XH1+J1x/3aBsSYt+rBtNFHneAPNCW1TK4/KT43NInty4Cm4Zl4yg7Ef8w1d8O4cY6vwZqdg32xibJKKG99WwDzqYE6zypFvSuN2/QbGNASP04nwQ/wXuO0gIGLdwTNxu27Rwti+Fv3bll+lhduWEf7+gsfuxOkC97T9AM6/PWEef///9u4EPKrqfuP4EbXWHVBrNQkg4lI3VNRuaiGzZCUkLNrqv9W6YNEqqOACImHLBAWqYCERqStaqbu4i+AKrai4ixvWtbhbq6AC53/OJNHwO0OSmczNnZn7/TzP+yTM3NwJv3Ay7Xm9MyaLTNbUub+P25uvTO61BXA6SjFzrvEJHiPZrJHnBXJdrUnT5qX9vMndjbetNtmr2e3tEgqF3ln87Mr/yg3XVLJo+cq15nxvy8dAfM7fmXxissDktH79+vWxpV04HH538bNvObNMJYueeePDHJu/s6mfaqLRqK6sGKDPOeNMPW/WXL10wcJ4aTFk0GCdjvk/+eanuri42Hlc0pBoJKqrBlTqc4efLeY/hPl3QCKRiB5YWaVHnznKk/mb3/168OCjnMdtKXZ9ywUfNHJTrynyuKDJK5mSb6/AyIvGJpl53JIXfx+h2P/knNoQ+zVP241ue+VNXlFtb/lYAICOYX8HF4RrhxVEYrPN7+VHze/njxL83m4pH5o8nB+tmWGfI3qVVG8nH6M97H8okuAxk459DpPnBprY9wyardTxdUpNM7nL5I261N7f6GOTR+uVmmk3cE16yMcCAHjvSqV+bH4H/9r8Pj7bfJxnPi4zv5+/TPB7e6MxX/OW+XiX+frJ5vN+M5RKeIFHqsx5Z8nHTDb2e5TnBXLdj01eNrGbdN+Y7Gli3+C6aUNz1A+Htl8oFFqzbOVnzoZrKrGb5OZ8tqSzH8nG823jxy/D4fDaJ9/4xJllKjHzt7O3RaB8PNIsRdGi+Mfy0rJ4cZGO+T/91hfa/CydxyJuiosa51/G/P2ILTLtx3TO3z6HFBUlV5CqhuezQJObek2Rx6FB/Eq7oliv/HCsr73Swb5EZvOY235n77ObwPbKDfn1AIDMYl/uuEdxrEdBUc0Rjb/bf9f0Oz0/UlNqb8svmnKY/f0vvzbd8qKx8+XzcQpZlwtXZ6PjzVRq18byru8spSrqlTquecztVfY+8/nec5XaVn49ACCzXKlU5zql9rC/uxt/fzf/nV5obzP3975Gqa3l16abeZwFsnBLIUvkeYEg+IXJWhO7UfeAySuNny81SeoNr1sTslfQLX9rrdxwTSX2SoxIJPKefAzE59y0Qb7GZFVhYeG4oqKirpFo9P10XMFiY36OX9ufp3zsLOZs6qeapvmXFBfrQZVVun7qpfrNJc+m7Qqi+EssVg50Hpc0pGn+pSUlenDVQD1n+mXMvwPzw/xL9eCBg/TfLpnVMP/B6Zt/pVlX8nFbil3fcsEHTYKNvXjkcQAAwFvm+XeOfD5OIbn0SiYAACAH1Cn1coLCLanUK3WDPC8QFFNNmjYxbdaY7LPBEWkQCoUe/fvdj3wkN1xTyfx7Hv/Knk8+BuJzXh0Oh+8yse8x2Pz2R/9x3+PfyVmmkhsWLH43x+bvbOqnGnvV1qgzztSP3nbvBrefmqb34LrzkWf0H44/0Xlc0hA7//NGnO3MP13vgRaf/3EnOI9LGmLnf/6Zo5z5/+nkP6Vt/sclOX+7vuWCD5oEG3vxyOMAAIC38iM1/5TPx0knXPuQPC8AAIBf7HvY1aX2UsobZLZSE+S5gaDorBpe4rJpI3PGhnenRygUGn3meRe+IjdcU8nIMRPetOeTjwGl7NVy8jbLzmvU2Ilvy1mmkhHnXPBCjs3f2dRPNfZqIXmbzeXTZuoJsenOLJPNJZdfqydNnOKcnzSkpfmPr/2LM89kY+c/eUKtc37SkI3Of3r65l+T5Pzt+pYLPmicjb3GyOMAAIB39h1S/SPz/Pu1fD5ONvb99eS5AQAA/DJLqV/Isi3F/J88NxAUE02aX0G3yiTt76nSt2/fXkUlpZ8tffVDZ9M1mdivLy4p/dyeTz4GNs7Oq6S07Iu0zL+45NMcm7+zqZ/uPPfgE3pg1aD4/ORMk8kfhw7Td89f4JyftJz4/Ae2f/4nnDxM3zv/duf8pOXY+Velaf733Zjc/O36lgs+aOTGXlPkcQAAwDvdwrWHy+fiVJIXqT1FnhsAAMAvdUqdm6BsSzqzldpPnhsIgt4m35rYjbp/mKxv/PzG5gelSyQSuWLspIvb9TKXF9ZM/zwajc6V50brioqLrxwX+8uXcqbJZMyEKR+EQqE58txZztnU9yKTxlTrSRfPdGba1tz8wFJ9zLG/16uWv+acm7SeSRdU64lTmb9fmXjB+PbP/5jk52/Xt1zwQSM39poijwMAAN7Jj9SMl8/FKaVoymHy3AAAAH6pU2qhLNtSyNfVSm0mzw3kuk1NnjSxm3Rvmmxhcl3jn20G/3BoXE+TASY/Vw1fm7RQKLRzUVHxqmtve3Ct3HxtS667feH64pLSj+x55LnROjs3Oz87RznbtsT83L6LRov+k4Pzdzb1vchrjz2tBw8eYv8dO7NtLf96/WP9u2P/oG+fd7NzXtK2pGP+dzL/lJOW+V93k3Pe1mLXt1zwQeNs7DVGHgcAALxjnntfkM/FKeSrPn3qN5fnBgAA8EO9UjvWpeH950wWyXMDQTDKRJZxu5j8t/G25i91ObXxtqY83+y+pPTr169PcUnpf5PdpLXHR4uKvgmHwyfLc6Lt7PxLSkv/l8L815v5r8nR+Tub+l5l6YKFurKyKqmSYtnKz/TwUaP1SSecpB++5R7nnKTtac/8T2b+7c6SBQ/pASnMf4SZ//jR45zztSV2fcsFHzQJNvfikccBAABvdIvG+sjn4ZQSrn1InhsAAMAvdUoNT1C2JZ3ZSk2Q5wZyXSeTc0zGmZwt7uvfeLtNX5MiE7uR95jJIJMHG/8cazw+afGSqKz84wkXzfhuyYpVzoZs89j7J1w8c11pWdknthwKhUKrIpHIb+Q50XZ2/mVl5Z/aubZp/hfN+LaktOyjHJ6/s6nvZWxJNGTIUfGX+2tt/g8+9boeeupwff7I8/RDNy3QVQMqKYnamSV3JTf/U049g/mnMbakGzLk6OTmf/Z5+v2nVzjnakvs+pYLPmiczb3GyOMAAIA38sOxufJ5OJXkRWPny3MDAAD4QSu1SZ1SL8uyLcUcKc8PYEN9TXo0fv5bE7uxd03TnamIRqM/qagYcENZef/V42LT1910/xK9aPnK+Kas/Wj/bG8v71+xpqKy6kZ7vP06Ww7laEnUoew8qwYO/oed74W1f1mfaP4XxqavtT+f/hWV1+f4/J1Nfa/z+uPP6AkXjNeVVQPNv/O/xOfdfP7z731Mj66O6QEDKvUVl87+/n23bDlESdT+NMx/Qtvmf8ks5p/mxOc/1s5/UKvzn9ts/qnErm+54INGbu41RR4HAADSb9dwbTfzvPuNfB5OKaHJB8jzAwAA+KFOqcEJirZU8mk17z8HJOVWE7uxd4y8IxXhcLhnaXn5xMqBg14qLev/eSgU0vZj1aDBL/cfMGCyvV9+TY6WRL6w862orKyx824+f/vzKCsrmxCQ+Tub+h2VFx5aquumzdTDhp4af38uO397ddepfzpNXzmjTr/66FPO11ASpS8bm/9pzL9DYuc/28z/T0nMP9nY9S0XfNA4m3uNkccBAID0K4jGrpbPwSklWvO6PDcAAIAf6pXavC59V89dLc8PIDHbZF9hYjf15plssuHd6WE3aOVtieRgSZQRAjp/Z1Pfr9j5y9sShZLImzB/f9PW+ScTu77lgg8aZ4OvMfI4AACQXgVFNUeY59z18jk4leSFaybK8wMAAPihTqlzExRtKWV2w9trAWjFdib3m9gNvYtUw3vYeaKtBZGVYyVRRgjo/J1Nfb+STEFBSZT+MH9/k8z82xq7vuWCDxq5wdcUeRwAAEifHSumbGueb9+Qz78pZn1eqGZP+RgAAAAdbZZSB9QptUYWbSnm3Wpe3hJo1S4mL5isNTlHNbwXnU3eD4ekTzIFkZVDJVFGCOj8nU19v5JsQUFJlN4wf3+T7PzbEru+5YIPmgSbfPHI4wAA8FtBZMohPfpW/1jennWqqzvlRWK3y+fe1FP7oHwIAACQPeqVOlzelo2uUKqr+bu8lqBoSymzzf9qko8BwFVm0rTJ2TyvNj8oXZItiKwcKYkyQkDn72zq+5VUCgpKovSF+fubVObfWuz6lgs+aNxNvobI4wAA8NWQ+Zva91kzz1Ev7Fo0+WB5dzbJj9RcJp9325OCSE2FfAwAAJAd5ii1W51S35nc+VelfirvzxbXKLW1+Ts8Jku2dmTNTKV2lY8DwGVfSmNcgpze/KB0SaUgsnKgJMoIAZ2/s6nvV1ItKCiJ0hPm729SnX9LsetbLvigkZt8TZHHAQDgp/xw7THNnqfW5kdrZvQM124vj8to1dWd0l3Ombxg/ueMJ+//DgAAvDdbqVnNSqnP65U6ozrLXtbxSqU6m+/94QQlW3tyuXwcABkg1YLIyvKSKCMEdP7Opr5faU9BQUnU/jB/f9Oe+W8sdn3LBR80CTb64pHHAQDgH72JeW56Tj5XmXycH6k5J/+X07eUX5FpevSt7lwQid2R4O/QrhREY0fLxwIAANnBXjFXp9TqBOXUK7OVOkoen4nqldrbfL8vJvg7tCffzFKqp3wsABmgPQWRlcUlUUYI6PydTX2/0t6CgpKofWH+/qa9808Uu77lgg8audHXFHkcAAB+2TVa218+T4msKojWXpAXqtlBfm0myCua3C8vEluZ4Ptub5Yprp4DACBr1Ss1JUE51TwvzFbqhPlK/Uh+rd+0UpuY7+0U8z1+meD7bm9myMcDkCHaWxBZWVoSZYSAzt/Z1Pcr6SgoKIlSD/P3N+mYv4xd33LBB02Czb545HEAAPglPxp7Qj5PbSRfF0RjV+8arf2VPIcfukcn7pYfjl2X4PtMR9Z3C9ceLh8TAABkhysbXhbyiwTlVKKssmXeX5XqJc/jh9lK/drk8QTfZzry0WVKZeR/dAVApacgsrKwJMoIAZ2/s6nvV9JVUFASpRbm72/SNf/msetbLvigSbDhF488DgAAP+SHY33lc1Qb80ZeNDYpLzL5QHlOr9nHLIjErjLfwzcJvq/0JBybKx8XAABkjzqlLkhQTrUlS+qVGmHSTZ7TS/aKOfPYpSb3J/ie0hZ7xaB8bAAZJF0FkZVlJVFGCOj8nU19v5LOgoKSKPkwf3+Tzvk3xa5vueCDxtnwa4w8DgAAP+RFau6Xz1Ep5B2TOQWR2O/yCyflycdIh4Lo5P3yorHzzeMsT/D46c479j3t5PcAAACyQ71SW81W6kNZTqWQF+yVdSZlM5TaTj5Oe1UrtdnlSv3GPM7FJu8kePx05x75PQDIMOksiKwsKokyQkDn72zq+5V0FxSURMmF+fubdM/fxq5vueCDJsGmXzzyOAAAOlpBZMoh8vkpTXkrP1x7a36k9sK8cO1R3aKxPm19/7puZbEu+aHJB5hzVBWEY2MazhP7MMFjeJW1edFYofy+AABA9qhTaniCcqq9WW/y3Gyl5pmPZ81SqsJ8vt9fldpGPn4i5rifXq7UofVKHdP43ngP1nnz/nIJYwtL87i7yO8LQIZJd0FkZUlJlBECOn9nU9+veFFQUBK1Pczf33gxf7u+5YIPmgQbf/HI4wAA6Gj5kdqb5fNTB+TtvEhspY35fEnT541/lsd2eGwpKOcEAACyR71Sm9cp9bYsqDzOdyYrG7PCfA/Lmv35gwTHd3TWmoTkrABkIC8KIisLSqKMEND5O5v6fsWLgsKGkqhtYf7+xov52/UtF3zQyI2/psjjAADoSLtGa/Y2z0fr5PNTkFMQqbnR/E+XTeSsAABA9rDvsZagoAp6hss5AchQXhVEVoaXRBkhoPN3NvX9ihcFRVMoiVoP8/c3Xszfrm+54INGbv41RR4HAEBHKojErpLPTQHPw71KZmwh5wQAALJHtVKd6pRakaCgCnKmyTkByGBeFkRWBpdEGSGg83c29f2KFwVF81AStRzm72+8mL9d33LBB02CDcB45HEAAHSUXQondTfPRd/K56bAJhp7YseKKdvKOQEAgOxSr9SQBAVVkDNbK8WrAwDZxOuCyMrQkigjBHT+zqa+X/GioJChJNp4mL+/8WL+dn3LBR80ziZgY+RxAAB0FPM8NFM+LwU4i3qVVG8nZwQAALJPnVJPJyipgppLKOeALNQRBZGVgSVRRgjo/J1Nfb/iRUGRKJREicP8/Y0X87frWy74oEmwERiPPA4AgI6SH479OZ8r6OLvOcfLWgIAkBuqldqsTqkrExRVQcs6k3PlfABkiY4qiKwMK4kyQkDn72zq+xUvCoqNhZLIDfP3N17M365vueCDRm4GNkUeBwBAR9o1Wvsr83z0hnx+CkjWFYRjY+RMAABA9qtT6o8mXyQoroKQz2crVS5nAiCLdGRBZGVQSZQRAjp/Z1Pfr3hRULQUSqINw/z9jRfzt+tbLvigSbApGI88DgCAjrZT3+pt8hte7nKtfJ7K4byTVzS5n5wFAADIHbOV6lGn1L0JCqxczpI5Su0mZwEgy3R0QWRlSEmUEQI6f2dT3694UVC0FkqiH8L8/Y0X87frWy74oEmwMRiPPA4AAL/sWjT5YPPc9LB8rsq5hGNzu5XFusi/PwAAyE2zlRpUp9TrCcqsXMrXJudWK7WZ/PsDyEJ+FERWBpREGSGg83c29f2KFwVFW0JJ1BDm72+8mL9d33LBB42zOdgYeRwAAH7LC9eWmeeoZfI5KwfydH6kxo//nwMAAHxWr9TmdUqdavJ2gnIr23Or+fvtLv/OALKYXwWR5XNJlBECOn9nU9+veFFQtDWURMzf73gxf7u+5YIPmgSbhPHI4wAAyBQF4doi81x1t8l6+fyVZXk1L1z7B1Vd3Un+HQEAQLDYom62UifUKbU8QdGVbVlkcqT8OwLIAX4WRJaPJVFGCOj8nU19v+JFQZFMgl4SMX9/48X87fqWCz5oEmwWxiOPAwAg0+SFavbMD8cuMs9b78vnsQzP0oJo7Gg1ZP6m8u8EAABQr9Ths5W6qk6prxKUX5matSa3zFLqCPn3AZBD/C6ILJ9KoowQ0Pk7m/p+xYuCItkEuSRi/v7Gi/nb9S0XfNAk2DSMRx4HAEDG6lu9WUEkFrXv4Waewz6Wz2kZEvN91VzWLRrrI799AACARK5Raus6pf7P5DaT1QlKsUzIGyYX1ivVTX7/AHJQJhRElg8lUUYI6PydTX2/4kVBkUqCWhIxf3/jxfzt+pYLPmgSbCDGI48DACArVFd3ygtP+UVBJDY2LxJ71DynfSef4zowbxREaup2DcfCffrUby6/VQAAgLaartSW9UpF65SaZvJigqKsw2K+j2Xm48RZSvEfHgFBkykFkdXBJVFGCOj8nU19v+JFQZFqglgSMX9/48X87fqWCz5oEmwmxiOPAwAgG+1YMWXb/EhNaV6kdrR5frslLxJbKZ/z0pSvTZYVRGKz86I1xxUUXrS7/F4AAADSZbZSeSaD6pWaUqfUPSYfyyItTfnU5GHzWFNNqszj7Si/FwABkkkFkdWBJVFGCOj8nU19v+JFQdGeBK0kYv7+xov52/UtF3zQJNhgjEceBwBAruhWFutir7LLi8SGFERiZ5vnvUvzw7W3mo+LGlL7jC3ymsfcvjx+XzR2j/nzlSYTCqKxU/NDsWL7XnjyMQAAADpavVK7XK7Ub2Yr9XuTMebPdebj3XVKLbIxf37JfFwpsrTx/lvNsXPMMePM5yeaz/v+VamfyscAEHCZVhBZHVQSZYSAzt/Z1PcrXhQU7U2QSiLm72+8mL9d33LBB40s5ijoAAAAAAAAADgysSCyOqAkyggBnb+zqe9XvCgo0pGglETM3994MX+7vuWCDxpZzFHQAQAAAAAAAHBkakFkeVwSZYSAzt/Z1PcrXhQU6UoQSiLm72+8mL9d33LBB40s5ijoAAAAAAAAADgyuSCyPCyJMkJA5+9s6vsVLwqKdCbXSyLm72+8mL9d33LBB40s5ijoAAAAAAAAADgyvSCyPCqJMkJA5+9s6vsVLwqKdCeXSyLm72+8mL9d33LBB40s5ijoAAAAAAAAADiyoSCyPCiJMkJA5+9s6vsVLwoKL5KrJRHz9zdezN+ub7ngg0YWcxR0AAAAAAAAABzZUhBZaS6JMkJA5+9s6vsVLwoKr5KLJRHz9zdezN+ub7ngg0YWcxR0AAAAAAAAABzZVBBZaSyJMkJA5+9s6vsVLwoKL5NrJRHz9zdezN+ub7ngg0YWcxR0AAAAAAAAABzZVhBZaSqJMkJA5+9s6vsVLwoKr5NLJRHz9zdezN+ub7ngg0YWcxR0AAAAAAAAABzZWBBZaSiJMkJA5+9s6vsVLwqKjkiulETM3994MX+7vuWCDxpZzFHQAQAAAAAAAHBka0FktbMkyggBnb+zqe9XvCgoOiq5UBIxf3/jxfzt+pYLPmhkMUdBBwAAAAAAAMCRzQWR1Y6SKCMEdP7Opr5f8aKg6Mhke0nE/P2NF/O361su+KCRxRwFHQAAAAAAAABHthdEVoolUUYI6PydTX2/4kVB0dHJ5pKI+fsbL+Zv17dc8EEjizkKOgAAAAAAAACOXCiIrBRKoowQ0Pk7m/p+xYuCwo9ka0nE/P2NF/O361su+KCRxRwFHQAAAAAAAABHrhREVpIlUUYI6PydTX2/4kVB4VeysSRi/v7Gi/nb9S0XfNDIYo6CDgAAAAAAAIAjlwoiK4mSKCMEdP7Opr5f8aKg8DPZVhIxf3/jxfzt+pYLPmhkMUdBBwAAAAAAAMCRawWR1caSKCMEdP7Opr5f8aKg8DvZVBIxf3/jxfzt+pYLPmhkMUdBBwAAAAAAAMCRiwWR1YaSKCMEdP7Opr5f8aKgyIRkS0nE/P2NF/O361su+KCRxRwFHQAAAAAAAABHrhZEVislUUYI6PydTX2/4kVBkSnJhpKI+fsbL+Zv17dY74EjizkKOgAAAAAAAACOXC6IrBZKoowQ0Pk7m/p+xYuCIpOS6SUR8/c3Xszfru9maz2QZDFHQQcAAAAAAADAkesFkbWRkigjBHT+zqa+X/GioMi0ZHJJxPz9jRfzt+u7+foPIlnMUdABAAAAAAAAcAShILISlEQZIaDzdzb1/YoXBUUmJlNLIubvb7yYv13fcv0HjSzmKOgAAAAAAAAAOIJSEFmZWNIFcf5dunRxNvX9ihcFRaYmE0si5u9vvJi/oqCjoAMAAAAAAADQuiAVRFamlXRBnP8RRxyRMSWFFwVFJifTSiLm72+8mL+ioKOgAwAAAAAAANC6oBVEViaVdEGcv72CLlNKCi8KikxPJpVEzN+9vyPjxfwVBR0FHQAAAAAAAIDWBbEgsjKlpAvo/DOmpPCioMiGMH9/k8vzt+tbLvigkcUcBR0AAAAAAAAAR0ALorhMKOkCOv/4Rn4mlBReFBTZEubvb3J1/nZ9ywUfNLKYo6ADAAAAAAAA4AhoQfQ9v0u6gM7/+818v0sKLwqKbArz9ze5OH+7vuWCDxpZzFHQAQAAAAAAAHAEtCDagJ8lXUDnv8GGvp8lhRcFRbaF+fubXJu/Xd9ywQeNLOYo6AAAAAAAAAA4AloQOfwq6QI6f2dT36+SwouCIhvD/P1NLs3frm+54INGFnMUdAAAAAAAAAAcAS2IEvKjpAvo/J1NfRs/SgovCopsDfP3N7kyf7u+5YIPGlnMUdABAAAAAAAAcAS0INqoji7pAjp/Z1O/KR1dUnhRUGRzmL+/yYX52/UtF3zQyGKOgg4AAAAAAACAI6AFUYs6sqQL6PydTf3m6ciSwouCItvD/P1Nts/frm+54INGFnMUdAAAAAAAAAAcAS2IWtVRJV1A5+9s6st0VEnhRUGRC2H+/iab52/Xt1zwQSOLOQo6AAAAAAAAAI6AFkRt0hElXUDn72zqJ0pHlBReFBS5Eubvb7J1/nZ9ywUfNLKYo6ADAAAAAAAA4AhoQdRmXpd0AZ2/s6m/sXhdUnhRUORSmL+/ycb52/UtF3zQyGKOgg4AAAAAAACAI6AFUVK8LOkCOn9nU7+leFlSeFFQ5FqYv7/Jtvnb9S0XfNDIYo6CDgAAAAAAAIAjoAVR0rwq6QI6f2dTv7V4VVJ4UVDkYpi/v8mm+dv1LRd80MhijoIOAAAAAAAAgCOgBVFKvCjpAjp/Z1O/LfGipPCioMjVMH9/ky3zt+tbLvigkcUcBR0AAAAAAAAAR0ALopSlu6QL6PydTf22Jt0lhRcFRS6H+fubbJi/Xd9ywQeNLOYo6AAAAAAAAAA4AloQtUs6S7qAzt/Z1E8m6SwpvCgocj3M399k+vzt+pYLPmhkMUdBBwAAAAAAAMAR0IKo3dJV0gV0/s6mfrJJV0nhRUERhDB/f5PJ87frWy74oJHFHAUdAAAAAAAAgLiioqKuoVBotd2cbZbV9nZ5LDYu1ZKO+be/oLNJtaR4c8mzurio+PvZR6NR/f5TrzjHkZbD/P1Nps7frm+54INGFnMUdAAAAAAAAAC+F4lErg2FQp/ZDdrCwsKvw+HwPHkMWpdqSWfmfb35ujWNm+Qfmz9fI4/JYc6mfqpJtaQYc9Y5OhwKxwuK0pIS537StqQ6/3OGn/V9QdS/rNy5n7Qtqc7/zFNP/37+lRUVzv3tiV3fcsEHjSzmKOgAAAAAAAAAfC8UCh1hiyG7QRsOhz+1f5bHoG1SKeka5x8vSM38PwzY/J1N/fYklZLikVvv0WUlpfGCYtiJQ537SduT6vxLihuu4ho+7DTnftL2pDJ/e6y9cs7Of9TpZzr3tyd2fcsFHzSymKOgAwAAAAAAALCBUCj0euNVFCvlfUiOLOmi0ehP5DGSOf4tk/XhcPg1eV+Oczb12xtZUrz++DPOMTJDBg6KFxR/u2SWcx9JLqnMf1BllS2n9bWXXe7cR5KLnP+///mCc4xMZcWAeEk3b9Zc5772xK5vueCDRhZzFHQAAAAAAAAANhAKhYYXFhauNxkh70Pymkq6aDR6Rjgc/qJv374/lcc0Z44ZYedvfw7yvhznbOqnI00lxTWXzYm/bOKrjz7lHNM8tpizBd2SBQ8695Hk0zT/q2bUt2n+9VNnMP80pmn+NReM1/3LW5//jJqpnszfrm+54INGFnMUdAAAAAAAAAA2UFRU1DUUCn1mP8r7kBpbzhUWFn5rr4wzH+fI+5sL8PzXrXr2tXVyYz8dseVcNBLREZPaCyc49zfPm0ue1WWlpfr9p15x7iOp5aqZ9fHZm3/T+uLqyc79zcP805vnFy7Rxx/7e11cVBwvSKeOr3GOaR4v5r9q+Wu6U6dOgS+iZDFHQQcAAAAAAADAXrXVs7CwcEwoFHrE5G17BYXJv00WmdvPNekuvwZtU1JSspOZ4/8aZ2qzWl5Fx/zjvnjjieVfys399sacU5cUlzTNXpufh3MV0QsPLY1fuXXayX/SRw0aHD/u6MFD9OlDh+krpl+mX1r8T+e8pG2x87flUNP8S0tKW5z/EOaftjx93yN6QP8KXdz4vn425aVlLc+/8SVefzvkaH3OGWfqO6+Zr/+z/FXn3Mlk5ZLn9NZbbbVeLvigkcUcBR0AAAAAAAAQYPY90QoLCy+PRKMf//nMc5664a7Fby1avlI/8/aX2n68/s6H3jW3P2fu/8QcN6tv3747ynOgdaFQaGeT20y+CzVcRTfb3t40/2i06JMzRp7/7N/vevjd5vOfd/tDK4edMfLpcCTycQDm/+7y+x/7j9zcT0dWPLpMj/zzCB2NROPvb3bRuEnx2+17ok0eM05XVVbpSZMu1rfe93h87k+t/Dz+8ea7H9YTx8fiLxEYGzshXjbJc5PW8/LDT+qzTjsjfhWdTdNVXM3nP9HM/5b7nthg/reY+U+YwPzbE3tFnH3Z1kFVA3VRtChe1k2b0Gz+o8fpytJSfV55RF/aO1/P2e3H+rqCTfQVu2+jp0d+oU+sGqCHHvdH/dzCJ5xztzXPm6/dsesO38kFHzSymKOgAwAAAAAAAAKqX79+fcLh8Ntnnz/u2SdWrFprS6GN5fGXP1h/1nnjXo5EIu+YHCjPhbZpLOpuMPmisLAwauc5asz4l5esWOXMfIP5v/TBdyNGjlluf145PP+XF/9jwZtycz+dee2xp/X5Z47U5WVl+oEbb9dHDR6iY1Mu0a3O/+X3dU3N1PjxT96z2DkvaVvs/M8dflb8pRab5l9z0aWtzv+Jl97XsckXx6+qY/6p5+Fb74lfJWffi87Of8iASn1Ov5/rq7ptpq/PVxvN+EP20JUlJfrx2+93ztmWLLnjAV2wa95queCDRhZzFHQAAAAAAABAADWWcx/Nnb/gTbkZ3lLm3rjgPXs1l/n63vKcaLtQKBQxc/z8ypvu+UDOuKXMueH21+3PLUfnv/SWK659UW7ue5EH59+hK/pX6OtvfcCZcUu54eZ79MDKKv3sg4875yRtT9P85932oDPjlvL3m+5m/mnIopvu0hUlJXrKfrs6ZdzGctGeXXVFJKz/dddC53yt5f7rb9V79uz1X7ngg0YWcxR0AAAAAAAAQMA0XsX1zhU3LnhdboC3JXP+fse74XD4raKioq7y3Gidnb+Z37vJlnNNqZ936wpzjpU5OP8bx48avURu7qc79iou+z5n825LrpxrytyrbtQnH39Cu9+XK6j5Yf7JlXNN+Rvzb1fs/AdXVOjafX/qlHCtZfLPfqIHFRfFXzZTnrelzJ16mT6090Er5YIPGlnMUdABAAAAAAAAARMKheacee4Fy+TGdzIZPmrMy+Y8l8pzo3V2/vZlLeVMk8lpI0Y9lYPzr+4fLXlYbu6nO/Y9z2qmXOrMNJmcdOJQvXD+Hc65Seux85980QxnpsnkpBNPZv4pZtL5F+iRv+rtlG9tzel99tITTzvNOW9LOf/PZ+nwkX0fkgs+aGQxR0EHAAAAAAAABEjfvn17RaLRD5euWPWN3PROJo+/9P46c55Pw+FwN/kY2Dg7/2i06OOlr364Xs40mTz20nurzew/zrH5/7ZXj57/lJv76cxzDz6hB1YN1Gb+zkyTyV/n3qAnjal2zk9aTsP8B6Vh/tcz/xRi519ZUqyvbuU951rKld031+WF/fRLi//pnH9jOap/lR4QLc21/6AgabKYo6ADAAAAAAAAAiQUCo0eNvzsJXLDO5WcOnzks4WFhSPlY2Dj7PyHjxr9rJxlKjl52BlLcmz+vbf88Zavy839dObyaTP1pJppziyTzW2LnopfxSXPT1qOnf/E2HRnnsnGzv/kE5h/srHzP/vIQ5zSLdmc1mcvPeuCsc75N5aD9++tB5X2/51c8EEjizkKOgAAAAAAAMBfdtPyuMZ0bnb77s1uP6jZ7e0SCoUenXfHwlfkhncque6Oh94tLCxcLB8DSpWUlGwhb7Ps/G+4+5FVcpap5Opb7ns+x+a/mclnLz+87Cu5wZ9s3n/qFec2m1NPOkXfdN/jziyTzWMvvqcHDKh0zk8a0vL8n3DmmWyYf8vZ2Pz/9Ifj9LQ9uziFW7K5eM+u+sQB/Z3zJ8q///m83nqrrdbvu+++28gFHzSymKOgAwAAAAAAAPw12sRu0Nlc1uz2xxpv+8KkoNnt7RIKhd5Z/OzK/8oN71SyePnKdfZ88jEQn/N3Jl8UFhYuNB+H9+vXr48t7cLh8LuLn33LmWUqeeiZNz8x535bPnaWu+PiCyYul5v8ySYaier+5eV6xLDT9TWXzdFLFyyMlxZDBg3W6Zj/k29+qouLi53HJQ2JRiK6ory/PvPU0/W1f20+/yHMvwMSMfMf0L9Cjzx9hL5u1hXfz39Q//56To8tnMIt2dhzVEbCzuMmyk31V+uf7bHnR3KhB5Es5ijoAAAAAAAAAH/Zq4aWqYYybq1Jb9VwVV1TaXfSD4e2XygUWrNs5WfOhncqsZvk5nxrTexHsvHYItN+/DIcDq998o1PnFmmkmVvfro+1FAEyscjzWLLCvuxvLQs/nk65v/0W19o87N0Hou4+X7+Zczfj0Sj0fjH+PzNzK7ttqlTuCWb6wo66ag5pyzjEmXEScN05Mi+d8vnoiCSxRwFHQAAAAAAAOC/A0y+UQ2F3OMm9qoo+/n9zQ9Kh1D8Crq3vpMb3qnEXgljrwiTj4H4nJs2yL8yea+wsPDcoqKirpFI5L10XEFks3j5v+25c+0Kxn222GKL9+Qmf7Kxsw+HwrqspFQPGThI/+2SWfrNJc+m7Qqu+EssVg50Hpc0xM7flnG2FD1q0GB95aV1zL8D01TM9S8r10cPHqKvmtEw/0Hl5Wm5gu5vPX6k+4fbVtDZ958rDxf9QS70IJLFHAUdAAAAAAAAkBkuVD9cNWfzX5PuGxyRBqFQ6NG/3/XIB3LDO5XceO9j/7Pnk4+B+JxXm9wSDocPF7c/+o97H1sjZ5lKbrhr8Ts5Ov8Xbr782vflRn8yKS4q1qPPGqUfve3eDW6374H2jzS8B9qdjzyj/3D8ic7jkoYUFxXpMWef48x/2MlpnP9xJziPSxpi5z925HnO/Icec2xa3oNuxu7b6mNLipzHlVl292LdpXPn71TDleKBJ4s5CjoAAAAAAAAgM3RVP1xFZ/OXDe9Oj1AoNPrMc8e+KDe8U8nI0dWv2/PJx4BS9mo5eZtl5zVqzPiVcpapZMSoMc/l6PxH/rLPoc/Izf5kYq8WkrfZXD5tpp4Qm+7MMtlccvm1evLEKc75SUNamv/42jTNf0Ktc37SkI3Nv/7iS/SIw/Z1Crdkc/7+3fT4U4Y655c597QR+vBDf2HfTxWKgg4AAAAAAADIVDPVhlfQfW6Sv8ERadC3b99excUlny599UNn0zuZ2K8vKin9zJ5PPgY2Lj7/ktLP0zL/ouJPcnT+O3fq1Onzt5Y+v15u+Lc3zz34hB5YNSg+PznTZPLHocP0PfPvdM5PWk58/gPbP/8TzPzvnX+Hc37Scuz8B0TD+upumzmlWzL53eE/1/dec6Nzfpndu++m+/36yFK5wINKFnMUdAAAAAAAAID/7MsgrlMNxVydydrGz+9pflC6hEKhOWMmTnlfbnonk7GTL/44EolcIc+N1kWj0blja6Z+KmeaTM4fX2tf3nKOPHcOuW3siHNekxv+6cikMdV60sUznZm2NTc/sFQfc+zv9arlrznnJq3Hzn9ie+d/DPNPNdUjztJnHryHU7q1NdP37KKPLitpdf53XHmDzt9l1y/lwg4yWcxR0AEAAAAAAAD+2sLkFdVQyD1vsqnJ7MY/25zww6HqRyblJseZ2KsS7NcmLRQK7RyNFv3n2tse+EZufrcl19724NqiouJV9jzy3GidnVtRccmHZo7r5Gzbkmtue3C1+fl9kOPz/+VWW271n/eeesXZ+G9vXnvsaT140BB93e0Lndm2ln+9/rH+3bF/0HfOu9k5L2lb4vMfzPz9ip3/wOIiHdtrR6d8ay3XdNtUD+53pL7zb/Oc88qEj+irC391+FS5sINMFnMUdAAAAAAAAIC/atQPZVy08Tb7/mUfN97W9FKXPU1ea3asjf3zLo1fk5R+/fr1KSou+cKWbXITvKVcd/vC9cUlpf+1Xy/Pibaz87NztPOUM24pjeXo5wGZ/32TRo15W278pyNLFyzUlZVVSZVEy1Z+poePGq2rx1Q75yPJJdX5jzDzHz96nHM+klyW3PmA7h8OJVXSXVfQSZ90+KF63IiznPPJLLzxDr1j1x3se6ra/6gEjWQxR0EHAAAAAAAA+GcTk2NVwxVxg8V9v2y83eYgE/teYy+qhuN3M7lDNZR05zd9QbLiJVFxyarxUy5ds2TFKmdDvHns/ea4b0pKyz4KSDnkOTtHO88JF834ti3zr55y6eoi8/MK0Px/tfVWW63y4io6G1sSDRl8lJ44dWZ8vnLmzfPgU6/rU049Q58/8jz9/tMrnHOR5BOf/5Cj4i932eb5n30u809TbElnr6QbcVAvfVUr70lXv9uW+rjCI/R5p53RpvlXREv0b375q3q5oINOFnMUdAAAAAAAAED2sC9/2eQi1VDQDW12W9Ki0ehPikvLrykrL/9q7OSp39x0/xK9aPnK+Ka4/Wj/PLZm2rdl5f2/LiuvuM4eL8+B1Nl5lvcfMM/O18454fwnT11jfz4lJaVXB23+m2yyyZ3DTxz2niwA0pXXH39GT7ygWldVDdTVsenxeTef//x7H9Ojq2N6wIBKPfeSWa2+7xZJLg3zH68rW5j/mOoa5u9R7PzH/fl0XREO6TP67KWn7dlFz9ntx/FSzn6culdXffqRh+qKkmI99y+XtWn+d109X++0w45rzPLdSq7noJPFHAUdAAAAAAAAkF3GmLxtss7kEpWmlxALh8M9I5GiCysGVD5XUlr6SSgU0qVlZZ8OqKx6PhotHmfvl1+D9LHztXMeUDXwBTt3O3/7c7A/j1AkMjbA8++++WabffavBQvXySIgnXnhoaW6ftoMPWzosPj7o9n526vrTjvlVH3ljDr96qNPOV9D0hc7/zrm71vs/GdPvkgPPfooPbC0JD7/weVl+tTjjk9q/h88s0Lvu9fP1v/6578cKRcyKOgAAAAAAACAbDfM5GGT1SYrVMNLXwK57Ox99tjrHVkGEEIyK5PPHav36tnrLbmA0UAWcxR0AAAAAAAAQPb4cbPP7XvR2Y29uma3Ablo08032/zZmvMu/EIWAoSQzMiTdy3SXTp3XtvngAMOlgsYDWQxR0EHAAAAAAAAZIejVMNLWx7W+Oda1VDQXfH9EUDu2mPzzTf/7P4bbvX0pS4JIcnn3WUv6wP33X/9YQf1mSYXLn4gizkKOgAAAAAAACA7DFANL2tpN/Oa8onJfs0PAnJYVefttv9sxSPLnIKAEOJfhh57vN5nj72flwsWG5LFHAUdAAAAAAAAkD16mIw0Gaca3otuxw3uBXJcp06dLj54/96f2it2ZElACOn41NX+Re+800++2n777bvI9YoNyWKOgg4AAAAAAAAAkC023XKLLW4LHf6bLz54ZoVTFhBCOi431V9t33fu23333PNQuVDhksUcBR0AAAAAAAAAIJv8aMstt1xUVVz+5arlrzmlASHE+zxww222nFt74H4HlskFisRkMUdBBwAAAAAAAADINltuteVWTx5VXrX6/ae5ko6Qjowt53bsusPaQ3ofdIJcmNg4WcxR0AEAAAAAAAAAstHW226z7UOHH/bLr1cuec4pEQgh6U/jy1quPeSAg06UCxItk8UcBR0AAAAAAAAAIFttut22212zV89eXz+/8AmnTCCEpC/1Uy7RXbbv/O2BP9u/VC5EtE4WcxR0AAAAAAAAAICstt02213YZfvOa66/7AqnVCCEtC/vLntZDz32eL3TDjv+b6+ePQ+R6w9tI4s5CjoAAAAAAAAAQNbbbLPNDt9um20/HXrs8etsoSBLBkJI8ll292Lde5/91u/Va4/lZpltL9cd2k4WcxR0AAAAAAAAAIBc0XnnnXZ6qFePnt/ecsV1TtlACGlbPnhmha4570L7kpZrD96/9xS50JA8WcxR0AEAAAAAAAAAckrXrl2P7rJ9588qoiXrnr73Ead8IIRsPHddPV/vs+fe63bvvtsb+fn5+8v1hdTIYo6CDgAAAAAAAACQi7bont9txrZbb/PtKf/3R738/secIoIQ8kMemn+nroiW6B26dF190H69T5cLCu0jizkKOgAAAAAAAABALvvp3rvvcf22W2/zzW8HDNKP3nKPU0wQEuTcceUNOnxEX71Dly6rD96393SzZraSiwjtJ4s5CjoAAAAAAAAAQBB03qfXnjO7dO7y9f5777Nuwsgx+vmFTzhlBSFByFP3LNbnnjZC79at+7pdf7Lz5wfvd0C1WSM/kosG6SOLOQo6AAAAAAAAAECQbNI9L69s/732eWSbrbf+9leHHKZH/ekMfevcefqdf73oFBmE5EL+/c/n9U31V+sRJw3TfQ44UG+/7Xbf9Nm/9/0FuxQcIRcIvCGLOQo6AAAAAAAAAEBQbdGze8+Bh/Y++ObduvX4YKstt1xny4ujKwbqC4aP1H+bdpm+d97N8auNbN598iWn+CAkE2L/bTb9O7X/Zu2/3TFnjNRDyiv1wfv11vbfdq8eu73384MO/nv3/O6l5t/+ZnIxwFuymKOgAwAAAAAAAACgwdbd8/IKf3HgIdU/P/Dg2/f/2b4v9eze48MdunT9cuedfrL6R5tvvs4cYzfUCcmo2H+b9t+o/be6e4/dPtxv731ePOzAg2479KA+Y/N/mt9X8b5yvpPFHAUdAAAAAAAAAAAA4CFZzFHQAQAAAAAAAAAAAB6SxRwFHQAAAAAAAAAAAOAhWcxR0AEAAAAAAAAAAAAeksUcBR0AAAAAAAAAAADgIVnMUdABAAAAAAAAAAAAHpLFHAUdAAAAAAAAAAAA4CFZzFHQAQAAAAAAAAAAAB7Kj9R+K8s5e5s8DgAAAAAAAAAAAEAa5EViE0y+ayrn4p9HY9XyOAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAALTP/wPeRyAy49ViKQAAAABJRU5ErkJggg==",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from IPython.display import Image\n",
    "Image(filename='fables/string2string-logo.png')\n",
    "# Image(url='https://raw.githubusercontent.com/stanfordnlp/string2string/main/fables/string2string-logo.png?token=GHSAT0AAAAAAB5KHQETSF3UOOWXYPJHECZAZAXW5HA')"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This tutorial provides an introduction to the **similarity** module in the **string2string** package, and presents a few examples to showcase its usage. While it is not an exhaustive overview of the module, it offers insights into its key functionalities. For a more comprehensive understanding of the module, please consult the official documentation."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "String similarity refers to the problem of measuring the degree to which two given strings are similar to each other based on a similarity function---which can be defined on various criteria, such as character matching, longest common substring or subsequence comparison, or structural alignment. There is a natural duality between string similarity measures and string distance measures, which means that it possible to convert one into the other with ease; hence, it is often the case that one uses string similarity and distance measures interchangeably.\n",
    "\n",
    "The string2string package currently offers the following similarity algorithms:\n",
    "\n",
    "* Cosine similarity\n",
    "* BERTScore\n",
    "* BARTScore\n",
    "* Jaro similarity\n",
    "* LCSubsequence similarity"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Before we begin, please make sure that you have installed the string2string packag via *pip*."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# !pip install string2string"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Cosine Similarity"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For two (non-zero) vectors, $x_1 \\in \\mathbb{R}^n$ and $x_2\\in \\mathbb{R}^n$ the **cosine similarity** between them is defined as follows:\n",
    "\n",
    "$$\n",
    "\\begin{align*}\n",
    "    \\texttt{cosine-similarity}(x_1, x_2) & = ||x_1|| \\ ||x_2|| \\cos(\\theta) \\\\\n",
    "    & = \\frac{x_1 \\cdot x_2}{||x_1|| \\ ||x_2||} \\\\\n",
    "    & = \\frac{\\sum_{i=1}^n x_{1i} x_{2i}}{\\sqrt{\\sum_{i=1}^n x_{1i}^2} \\sqrt{\\sum_{i=1}^n x_{2i}^2}}\n",
    "\\end{align*}\n",
    "$$\n",
    "\n",
    "where $\\theta$ denotes the angle between the vectors, $\\cdot$ the dot product, and $|\\cdot|$ the norm operator.\n",
    "\n",
    "In practice, thought, we need to ensure that the denominator is nonzero; so, we often compute the cosine similarity as follows:\n",
    "\n",
    "$$\n",
    "\\begin{align*}\n",
    "    \\texttt{cosine-similarity}(x_1, x_2) & = \\frac{x_1 \\cdot x_2}{\\max(||x_1|| \\ ||x_2||, \\epsilon)}\n",
    "\\end{align*}\n",
    "  $$              \n",
    "where $\\epsilon$ is a small value to avoid division by zero.\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's now learn how to use the CosineSimilarity class in our library, which can automatically detect and handle both tensors and numpy arrays.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/machine/opt/anaconda3/envs/string2string/lib/python3.9/site-packages/tqdm/auto.py:22: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
      "  from .autonotebook import tqdm as notebook_tqdm\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tensor result: 0.9746317863464355 (type: <class 'torch.Tensor'>)\n",
      "Numpy array result: 0.9746318461970762 (type: <class 'numpy.float64'>)\n"
     ]
    }
   ],
   "source": [
    "# Import relevant modules\n",
    "import numpy as np\n",
    "import torch\n",
    "\n",
    "# Import the CosineSimilarity class from the similarity module\n",
    "from string2string.similarity import CosineSimilarity\n",
    "\n",
    "# Create an instance of the CosineSimilarity class\n",
    "cosine_similarity = CosineSimilarity()\n",
    "\n",
    "# Compute the cosine similarity between two tensors\n",
    "result_tensor = cosine_similarity.compute(torch.tensor([1, 2, 3]), torch.tensor([4, 5, 6]), dim=0)\n",
    "\n",
    "# Compute the cosine similarity between two numpy arrays\n",
    "result_np_array = cosine_similarity.compute(np.array([1, 2, 3]), np.array([4, 5, 6]), dim=0)\n",
    "\n",
    "# Print the results, along with their types\n",
    "print(f'Tensor result: {result_tensor} (type: {type(result_tensor)})')\n",
    "print(f'Numpy array result: {result_np_array} (type: {type(result_np_array)})')"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the context of natural language processing, the cosine similarity can be used a quasi-metric to measure semantic similarity between the vector representations of two pieces of text.\n",
    "\n",
    "Using the string2string library, we can do compute the cosine similarity between words using GloVe or fastText embeddings or between longer pieces of text, such as sentences, paragraphs, or even documents, via neural language models such as BERT.\n",
    "\n",
    "In what follows, we will demonstrate how to compute the cosine similarity between difference pieces of text at the word and sentence levels.\n",
    "\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Cosine Similarity Between the Glove Embeddings of Pairs of Words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The cosine similarity between the GloVe embeddings of \"cat\" and \"dog\" is 0.92\n",
      "The cosine similarity between the GloVe embeddings of \"cat\" and \"phone\" is 0.32\n",
      "The cosine similarity between the GloVe embeddings of \"cat\" and \"computer\" is 0.35\n",
      "The cosine similarity between the GloVe embeddings of \"dog\" and \"phone\" is 0.36\n",
      "The cosine similarity between the GloVe embeddings of \"dog\" and \"computer\" is 0.33\n",
      "The cosine similarity between the GloVe embeddings of \"phone\" and \"computer\" is 0.72\n"
     ]
    }
   ],
   "source": [
    "# Import the GloVeEmbeddings class from the misc module\n",
    "from string2string.misc import GloVeEmbeddings\n",
    "\n",
    "# Create an instance of the GloVeEmbeddings class\n",
    "glove = GloVeEmbeddings(\n",
    "    model='glove.6B.200d',\n",
    "    dim=50,\n",
    "    force_download=False,\n",
    "    dir='/Users/machine/.cache/torch/hub/glove.6B.200d'\n",
    ")\n",
    "\n",
    "# Let's define a set of words\n",
    "words = ['cat', 'dog', 'phone', 'computer']\n",
    "\n",
    "\n",
    "embeds = []\n",
    "# Compute the embeddings for each word\n",
    "for word in words:\n",
    "    embedding = glove.get_embedding(word)\n",
    "    embeds.append(embedding)\n",
    "\n",
    "# Create a matrix to store the cosine similarities\n",
    "similarity_matrix = np.zeros((len(words), len(words)))\n",
    "# Compute the cosine similarity between each pair of embeddings\n",
    "for i in range(len(embeds)):\n",
    "    similarity_matrix[i, i] = 1\n",
    "    for j in range(i + 1, len(embeds)):\n",
    "        result = cosine_similarity.compute(embeds[i], embeds[j], dim=1).item()\n",
    "        similarity_matrix[i, j] = result\n",
    "        similarity_matrix[j, i] = result\n",
    "        print(f'The cosine similarity between the GloVe embeddings of \"{words[i]}\" and \"{words[j]}\" is {result:.2f}')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# We can also visualize the cosine similarity scores using a heatmap\n",
    "\n",
    "from string2string.misc.plotting_functions import plot_heatmap\n",
    "\n",
    "plot_heatmap(\n",
    "    similarity_matrix, \n",
    "    title='Similarity Between GloVe Embeddings',\n",
    "    x_ticks = words,\n",
    "    y_ticks = words,\n",
    "    x_label = 'Words',\n",
    "    y_label = 'Words',\n",
    "    valfmt = '{x:.2f}',\n",
    "    cmap=\"Blues\",\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Note about their similarity. / can also do this using fastText embeddings."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Cosine Similarity Between the Language Model Embeddings of Pairs of Sentences"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import the ModelEmbeddings class from the misc module\n",
    "from string2string.misc import ModelEmbeddings\n",
    "\n",
    "# Create an instance of the ModelEmbeddings class (if device is not specified, it will be automatically detected)\n",
    "bart_model = ModelEmbeddings(\n",
    "    model_name_or_path='facebook/bart-large'\n",
    ")\n",
    "\n",
    "# Let's define a set of sentences\n",
    "sentences = [\n",
    "    'Tea is my favorite drink to have in the morning. As Arthur Wing Pinero said, where there is tea there is hope.',\n",
    "    'Richard Feynman, the famous physicist, made significant contributions to the field of quantum electrodynamics.',\n",
    "    'Coffee is one of the most celebrated beverages in the world. It is a brewed drink prepared from roasted coffee beans.',\n",
    "    'Albert Einstein published four groundbreaking papers in 1905, which laid the foundation for his theory of relativity.',\n",
    "]\n",
    "\n",
    "# Compute the sentence embeddings for each sentence\n",
    "embeds = []\n",
    "for sentence in sentences:\n",
    "    embedding = bart_model.get_embeddings(sentence, embedding_type='mean_pooling')\n",
    "    embeds.append(embedding)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sentence 1: Tea is my favorite drink to have in the morning. As Arthur Wing Pinero said, where there is tea there is hope.\n",
      "Sentence 2: Richard Feynman, the famous physicist, made significant contributions to the field of quantum electrodynamics.\n",
      "Sentence 3: Coffee is one of the most celebrated beverages in the world. It is a brewed drink prepared from roasted coffee beans.\n",
      "Sentence 4: Albert Einstein published four groundbreaking papers in 1905, which laid the foundation for his theory of relativity.\n"
     ]
    }
   ],
   "source": [
    "# Let's print each sentence\n",
    "for i, sentence in enumerate(sentences):\n",
    "    print(f'Sentence {i + 1}: {sentence}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The cosine similarity between the BART embeddings of Sentence 1 and Sentence 2 is 0.52\n",
      "The cosine similarity between the BART embeddings of Sentence 1 and Sentence 3 is 0.73\n",
      "The cosine similarity between the BART embeddings of Sentence 1 and Sentence 4 is 0.55\n",
      "The cosine similarity between the BART embeddings of Sentence 2 and Sentence 3 is 0.53\n",
      "The cosine similarity between the BART embeddings of Sentence 2 and Sentence 4 is 0.80\n",
      "The cosine similarity between the BART embeddings of Sentence 3 and Sentence 4 is 0.53\n"
     ]
    }
   ],
   "source": [
    "# Let's create a matrix to store the cosine similarity scores\n",
    "similarity_matrix = np.zeros((len(sentences), len(sentences)))\n",
    "\n",
    "# Compute the cosine similarity between each pair of embeddings\n",
    "for i in range(len(embeds)):\n",
    "    similarity_matrix[i, i] = 1\n",
    "    for j in range(i + 1, len(embeds)):\n",
    "        result = cosine_similarity.compute(embeds[i], embeds[j], dim=1).item()\n",
    "        similarity_matrix[i, j] = result\n",
    "        similarity_matrix[j, i] = result\n",
    "        print(f'The cosine similarity between the BART embeddings of Sentence {i + 1} and Sentence {j + 1} is {result:.2f}')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# We can also visualize the cosine similarity scores using a heatmap\n",
    "\n",
    "x_ticks = [f'Sent # {i + 1}' for i in range(len(sentences))]\n",
    "\n",
    "from string2string.misc.plotting_functions import plot_heatmap\n",
    "\n",
    "plot_heatmap(\n",
    "    similarity_matrix, \n",
    "    title='Similarity Between BART Embeddings',\n",
    "    x_ticks = x_ticks,\n",
    "    y_ticks = x_ticks,\n",
    "    x_label = '',\n",
    "    y_label = '',\n",
    "    valfmt = '{x:.2f}',\n",
    "    cmap=\"Blues\",\n",
    "    )"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### BERTScore and BARTScore for Semantic Similarity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertModel: ['cls.predictions.transform.dense.bias', 'cls.predictions.transform.LayerNorm.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.dense.weight', 'cls.predictions.decoder.weight', 'cls.predictions.bias']\n",
      "- This IS expected if you are initializing BertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
      "- This IS NOT expected if you are initializing BertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n"
     ]
    }
   ],
   "source": [
    "# Import the BERTScore and BARTScore classes from the misc module\n",
    "from string2string.similarity import BERTScore, BARTScore\n",
    "\n",
    "# Create an instance of the BERTScore class\n",
    "bert_score = BERTScore(\n",
    "    model_name_or_path='bert-base-uncased',\n",
    ")\n",
    "\n",
    "# Create an instance of the BARTScore class (if device is not specified, it will be automatically detected)\n",
    "bart_score = BARTScore(\n",
    "    model_name_or_path='facebook/bart-large-cnn',\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Let's define a set of sentences -- source sentences and target sentences -- in the following format\n",
    "# source_sentences = [s1, s1, s1, s1, s2, s2, s2, s2, s3, s3, s3, s3, s4, s4, s4, s4]\n",
    "# target_sentences = [s1, s2, s3, s4, s1, s2, s3, s4, s1, s2, s3, s4, s1, s2, s3, s4]\n",
    "source_sentences = []\n",
    "target_sentences = []\n",
    "for i in range(len(sentences)):\n",
    "    for j in range(len(sentences)):\n",
    "        source_sentences.append(sentences[i])\n",
    "        target_sentences.append(sentences[j])"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### BERTScore"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "BERTScore similarity scores (precision):\n",
      "[[1.   0.34 0.44 0.35]\n",
      " [0.33 1.   0.37 0.55]\n",
      " [0.52 0.39 1.   0.4 ]\n",
      " [0.37 0.59 0.41 1.  ]]\n"
     ]
    }
   ],
   "source": [
    "# Compute the BERTScore similarity scores\n",
    "bertscore_similarity_scores = bert_score.compute(\n",
    "    source_sentences,\n",
    "    target_sentences,\n",
    ")\n",
    "\n",
    "# Get the precision scores\n",
    "bertscore_precision_scores = bertscore_similarity_scores['precision'].reshape(len(sentences), len(sentences))\n",
    "\n",
    "# Print the BERTScore similarity scores (rounded to 2 decimal places)\n",
    "print(f'BERTScore similarity scores (precision):')\n",
    "print(np.round(bertscore_precision_scores, 2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# We can also visualize the BERTScore similarity scores using a heatmap\n",
    "plot_heatmap(\n",
    "    bertscore_precision_scores,\n",
    "    title='Semantic Similarity Between Sentences\\nas Measured by BERTScore',\n",
    "    x_ticks = x_ticks,\n",
    "    y_ticks = x_ticks,\n",
    "    x_label = '',\n",
    "    y_label = '',\n",
    "    valfmt = '{x:.2f}',\n",
    "    cmap=\"Blues\",\n",
    ")"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### BARTScore"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "BARTScore similarity scores:\n",
      "[[-0.24 -3.28 -2.87 -3.51]\n",
      " [-4.11 -0.27 -3.09 -2.94]\n",
      " [-3.76 -3.32 -0.27 -3.49]\n",
      " [-3.94 -2.27 -3.12 -0.35]]\n"
     ]
    }
   ],
   "source": [
    "# Compute the BARTScore similarity scores\n",
    "bartscore_similarity_scores = bart_score.compute(\n",
    "    source_sentences,\n",
    "    target_sentences,\n",
    ")\n",
    "\n",
    "# Get the BARTScore scores\n",
    "bartscore_scores = bartscore_similarity_scores['score'].reshape(len(sentences), len(sentences))\n",
    "\n",
    "# Print the BERTScore similarity scores (rounded to 2 decimal places)\n",
    "print(f'BARTScore similarity scores:')\n",
    "print(np.round(bartscore_scores, 2))"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can also visualize the BERTScore similarity scores using a heatmap"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAcwAAAHbCAYAAABGCKqlAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy88F64QAAAACXBIWXMAAA9hAAAPYQGoP6dpAAB6jElEQVR4nO3dd1QUV8MG8GfpZelKUVHAGlFEsAQVK6ho7LFEo6BRY401r1FjwBS7iRp7L7FrxB5rUCzYsVfEjoCg9M79/uBzZGGBQUHRPL9z9hx25t6ZO4V9Zu7M7CqEEAJERESUL40P3QAiIqKPAQOTiIhIBgYmERGRDAxMIiIiGRiYREREMjAwiYiIZGBgEhERycDAJCIikoGBSUREJAMDk0qc1atXQ6FQSK+S5H22zc/PT5qPnZ2dyjg7OztpnJ+fX7G2Q057PiZpaWnw9fVF1apVoaurKy3TyJEjP3TTqIRjYBahTZs2oVWrVrCysoK2tjZMTExgb2+Ppk2bYsSIEThw4MCHbuIHV9LCkNusaHxMYerr64uff/4Zd+7cQWpqaqHrHzhwAJ06dULZsmWho6MDIyMjVKhQAQ0aNMDgwYOxefPmYmi1fD4+PtK2aNq06Qdty6dG60M34FPRp08frFu3TmVYbGwsYmNj8eDBAxw7dgwPHz5Eq1atPlALPx5169bFzJkzi30+b7PN3lfbAKBly5ZQKpUAABMTk/cyz/yUtPa8rY0bN0p/16hRAz179oS2tjZcXV0LrPvTTz/hl19+URmWlpaG+Ph4PHr0CKdPn8bp06fRvXv3Im83fXgMzCLwzz//qHzwurq6olWrVlAqlYiMjMTFixdx+vTpD9jCj4ujoyMcHR2LdR5vu83eR9tea9CgARo0aPBe5pWf2NhYGBsbl5j2vKuHDx9Kf48cORLffPONrHo3btzAr7/+Kr2vWrUqOnToADMzM0RHR+Py5cs4ceJEkbeXShBB72zUqFECgAAgKlWqJNLT03OViYmJESdOnFBb//jx46J79+7C1tZW6OjoCCMjI/H555+L+fPni9TU1FzlX88LgFi1apVYu3atqFWrltDT0xMVK1YUv//+uxBCiLS0NPHLL78IOzs7oaOjI6pVqyaWLl2aa3r//vuv6Nevn6hdu7awtrYWOjo6Ql9fX1SsWFH4+PiIK1eu5Krj7e0ttaFJkybi2bNnYsCAAVL9nPMKDQ1Vabe6l6+vrxBCiFWrVqkMzyktLU2sWLFCeHp6CktLS6GtrS1KlSol6tevL/z8/NRvpBzedpvl17YmTZpIw729vcWZM2dEixYthKGhobC0tBRDhgwRcXFxQgghNm/eLFxcXISenp4oU6aMGD16tEhOTlaZnq+vrzS9ChUqqIyrUKFCrvX2et38+OOPwsvLSzg4OAgTExOhpaUlzM3NRaNGjcS8efNy7VM5t82///4rli9fLmrXri309PRErVq18mzPv//+W+B2XbVqlejTp4/03s3NLde63rNnjzReU1NTPH36NM9tl92TJ0/E2LFjRY0aNYShoaHQ1dUVFSpUEL169RJnzpzJc/uoe/3777/5zmvu3LlSWUNDQxEfH5+rTGJiojh69Kja+sHBwaJv377CwcFB6OnpCUNDQ+Hs7Cx+++03tdPKuY3Pnz8v2rZtK0xMTIS+vr5o1KiRCAwMlMrn3DcLWsaMjAyxdu1a4enpKUqXLi39H7Vp00bs3bs3V3tybuuQkBCxYMECUbNmTaGrqytKly4tvvnmGxEdHa12+c+ePSt8fHxExYoVhb6+vjA0NBSVK1cWPj4+4t69eyplk5OTxZ9//inc3d2FmZmZ0NbWFtbW1uLLL78Up06dUjv9VatWiSZNmggLCwuhpaUlTE1NRZUqVUS3bt3EggUL1NYpLAZmERg+fLi0E5UqVSrXxs/PhAkT8t3B3d3dc/0zZR/v6uqqtt6kSZNEhw4d1I5bsWKFyvTGjBmTbxt0dHTEoUOHVOpkD0wHBwdhY2OT77yKKjCjoqJE3bp185yGiYlJsW4zuYHp6OgodHV1c7WvadOmYtasWWrb3rt3b5XpvU1gxsXFFbiePTw8VA4Qcm4bd3d3lfdFEZjnzp1TGXb9+nWV5ckeqG3atJG1LY4dOybMzMzynK+GhoaYPXu22u1TUJioM3v2bKmstra2OHfunKx2CiHEwoULhZaWVp7zrl69uggLC1Opk30b16tXT2hra+eqp6urK27cuCGEKFxgJiYmCg8Pj3zLjh49WqU9Obd1o0aN1NZr3LhxruWfPHmyUCgUec5rx44dUtmIiAjh7Oyc73adM2eOyvSz75vqXlZWVrK3VX7YJVsEXFxcpL9fvHiBKlWqwNnZGXXr1oWrqyuaNWuGSpUq5aq3adMmTJkyRXrfqlUrNGzYEOHh4VizZg3i4+MRGBiIUaNGYenSpWrnfeHCBbi5ucHT0xObN2/G7du3AUC6ztKkSRM0btwYy5Ytw/PnzwEAM2bMQL9+/aRpGBoaokmTJqhZsybMzc2hr6+PqKgo7N27Fzdv3kRqaiq+++473LhxQ20b7t+/Dz09PQwePBj6+vpYtGgRkpKSVOZlbm6OmTNn4vz58yo3RWS/Hiinu6937944d+6c9P6zzz5DmzZtoKuri0uXLuHMmTMFTgN4+20m1/Xr11GhQgX06tULZ8+exeHDhwEAAQEBCAgIQKVKldC9e3ccOHAA58+fBwCsX78e06ZNQ5kyZd56vgqFAg4ODvj8889RtmxZmJmZIS0tDbdu3cLWrVuRnp6Ow4cPY/v27ejWrZvaaQQGBqJChQro0qULDAwMEBERkef8KlasiJkzZ+LgwYM4dOgQAMDMzAwTJkyQytStWxeOjo74/PPPERQUBABYvnw5fv/9dwBAamoqdu7cKZXv27dvgcv56tUrdO7cGS9fvgQA6Ovro2/fvjA2NsbGjRvx8OFDZGZmYuzYsXB1dUWTJk0wePBgfPHFF/j++++l6XTv3h116tSRliU/2feZtLQ01K1bF9WrV0e9evWkedSsWTNXvVOnTmHYsGHIzMwEAHz++edo3bo14uLisGbNGrx48QI3btxAnz59cPDgQbXzPnv2LMqVK4devXrh8ePH2LBhAwAgJSUFc+fOxeLFi6Xr65s3b5b2KQcHBwwePFiazutlHDVqlLRP6ujooEePHqhcuTKuXr2KrVu3QgiB33//Ha6urujZs6faNp04cQItWrRAgwYN4O/vj6tXrwIAjh8/jqCgIHz++ecAgK1bt8LX11eqZ2BggB49eqBChQoIDQ3F7t27Vabbu3dvBAcHAwCMjIzQs2dPlCtXDidPnsQ///yDzMxMjBo1CnXq1EHDhg0BAIsWLZLqe3h4oGnTpkhISMDjx49x4sQJ6fPonRVJ7P7HpaWliTp16uR7hNOoUSMRHBysUq927drS+D59+qiM27JlizROS0tLREVFSeOyT7d69epSF9uBAwdynRm8PpNYvHixyrjY2FiV+WVkZIgzZ86I1atXizlz5oiZM2eK0aNHq9R59OiRVD77GSYA4e/vL42bM2dOnvMqqLs1vzJXrlxRGd6mTZtc3YshISF5b6hs3nabyT3D1NbWFqGhoUIIIRISElTOLnR0dKQux1u3bqlMb9euXdL03uYM87Xw8HCxc+dOsXDhQjFr1iwxc+ZMUaNGDalOv379pLI5zzDt7e3Fy5cvc00zv/bkN+619evXS2VKlSolUlJShBBC7N69WxpuYWEhDc/PH3/8odLmffv2qSy7UqmUxnXo0EGlbvZ6q1atKnBe2XXs2DHffcbJySlXl2ynTp2k8U2bNhUZGRnSuLNnz6rUv3z5sjQu+zY2NDRU6abO3g4XFxeV+eW8XJJTVFSUyv64cuVKlfFDhgyRxtWuXVsanvMMs1OnTiIzM1OapqampjRu3rx5Uj0XFxeV5bh9+7bK/OLj40V4eLgQQojLly+rzCPnumzTpo3K/F8zNjaWhuc8UxdC/udCQRiYRSQ2NlaMHz9eWFlZ5fnPVLp0aRERESGEyPoQza+LIudr//790ryyD89+ze727dsq4yZPniyNO3TokMq4hw8fSuMOHjwoypcvX2Absl87yP5PWaZMGZV1sX///jzn9S6BuXDhQpXhOa9RFVZht1lB7c8emE2bNlUZl73LulmzZtLwtLQ0lemtWbNGGvc2gZmYmCh8fHyEhoZGvtuyZcuWUp2cgTlr1iy16+tdAzM1NVVYW1tL5TZt2iSEUO2O/e6779TWzalbt24q2yinrl27SuMtLS1Vxr1LYKamporp06cLOzu7PNetnp6euHnzplTH0tJS9v/5okWLpHrZt3HPnj1V2jFu3DhpnL29vcq4ggJz3759stujUChEQkKCECJ3YB48eFBlutn/j15/9uT8nBs8eHC+6zfn/3h+r+zdrG3btpWGW1hYiDZt2ogRI0aIpUuXirt37+Y7z8Lgc5hFxMjICFOmTEFYWBiuXbuGFStWwNvbG0ZGRlKZyMhI6c7Mly9fQgghe/qRkZFqh2fvvtPR0clznJaWau/76+6hZ8+eoWPHjnj06FGBbUhJSVE7POdzd7q6umrn9a6io6NV3tvb27/T9Aq7zQojZ7dq9m0jZ7u8rfHjx2P16tUFTievbQkA1apVe6c25EVbWxuDBg2S3i9fvjxXd2z2SwX5yb4vWFlZ5RqffdjrbtuioK2tjf/9738IDQ3F3bt3sW7dOnz77bcoXbq0VCY5ORkLFy5U29aC5PV/nt//WGH3mcK0RwiBqKiot25Tzs+5gv5n33ZdLVq0SOoCjoqKwr59+zB37lwMHDgQlStXRvfu3Yvkc4jXMIuYQqGQHj3o168f/Pz8ULFiRWlj3b17FwBgamqqUq99+/Zwd3fPc7rZr59kp62tnWednB/G6uzevRuJiYnS+9mzZ+Obb76BiYkJbty4IesRipxtKK4vJDA3N1d5HxoaqvJB9bbkbrPCeNft8rayXx+uWbMmNm7ciKpVq0JLSwvdunXD1q1bC5yGoaFhsbXv22+/xW+//Ya0tDQcOXIES5YsQUxMDACgdu3aqFWrlqzpZN8XwsPDc43PPszMzOwdW61epUqVUKlSJXz99deYNm0aKlWqJIVL9n3G3Nxcug7cqFEjdOjQIc9p5nUdvyj/x3L+H40aNSrf6+Z5PXMrp01mZmZQKBRSaIaGhhaqbT///DP09fXzrQMAtra2OH36NO7du4ezZ8/i7t27uHr1Knbu3In09HRs2bIFrVu3lnV9PD8MzCKwZs0aJCcn46uvvoKxsbHKOENDQ2hoaEgfvq+D0tDQEM7OztLF7aioKIwYMSLXThgTE4P9+/cX27N/OY8e+/btK/2DbNmypcjnl3P5EhMTYWBgIKtuo0aNVN7/8ssv2LFjh0oAPXz4EBUqVChwWm+zzT4G2bdns2bNpP0mMjISAQEBxTbf7Ns1+wFYTtbW1ujatSs2bNgAIQT+97//SePknl0CWcHyev+MjIzE/v374eXlBQCIiIjA/v37VcoWhX379uHatWvo27dvrgM1PT09lf0w+z7z+qYYAHj+/DkGDhyYa59LSkrC1q1bi6StBW2L+vXrQ1NTExkZGVL5sWPH5ir34MED3L59O1dbC8PAwAC1a9fGxYsXAQDr1q3D6NGjVW6oS0pKQlxcHCwtLXMtf6lSpVRuWnrt+vXrKj0Hly9fRs2aNaWDmNc6dOiAXbt2AQAuXrzIwCwJQkNDMXnyZIwcORKNGjWCs7MzzM3NERUVhW3btiE9PV0q27p1a+nv77//Hr169QIAnDx5Ek5OTmjXrh3MzMwQFRWFS5cu4cSJE7CxsUGPHj2Kpe1Vq1ZVed+2bVt4eXnhypUr2LZtW5HPr2zZsirve/bsiQYNGkBDQwO9e/dW2732Ws2aNdGmTRvs27cPALBnzx7UqlULbdq0gZ6eHq5fv47jx4/jxYsXBbbjbbdZSVe1alVcu3YNALBs2TJoaGjAwMAA69aty7O7ryhk366RkZHo27cvqlevDoVCgaFDh6qcJQwfPly6yzM5ORlAVndeXndjquPt7Y1ffvlFOkDo0qUL+vXrB2NjY2zYsAHx8fEAUKTfERsREYFx48Zh4sSJcHNzg6urKywtLREbG4vdu3ernNVm32fGjBmDnTt3QgiBe/fuoUaNGujcuTOsrKwQExODq1ev4tixY0hISECfPn3euZ3Zt8WFCxcwYsQI2NraQkdHB9999x3Mzc3Rr18/LFu2DEDWneznz59HgwYNoKenh6dPnyIoKAiXLl2Ct7f3O3872Q8//CDdkR0fHw9nZ2fpLtnHjx9jz549WLhwITp27IhatWrB09NTuuN62LBh2L9/P1xdXaGhoYGHDx/i1KlTuHnzJnx9faWD6O7duyMmJgbNmjVD2bJlYW5ujpCQEOmzAiiiA98iuxr6H1bQM0CvXwMGDMhVd/z48QXWy3kTRfZx2W9ayHnzRvZxOS/Yv76DMzU1VdSsWVPtfHPeCZv9ObX8bizIa15CZD2QnNczm6+fa8vvxpoXL14UyXOYb7vNCvPFBdllv4Ej57i8ttnb3PSzceNGtcthY2MjPD091W4zdV9cUNA6y9mesLAwYWBgoHbekZGRuaaV8w7lrl27qp1nfo4dOyZMTU3z3HYaGhpqb2DKa30XRM5zjgCEl5dXri/CWLBgQb7PYarbn/K7Ezq/bXHp0iW1N30ZGhpKZRISEgp8DjPnvprf/3VB7fXz85P9HGZ4eHi+z2Gqm0fVqlXzLWtubi4ePHiQ7/aVgzf9FIGRI0di27ZtGDJkCOrVq4fy5ctDX18fOjo6KFu2LNq3b4/t27erfZZyypQpOHnyJL7++mvY29tDV1cX2traKFu2LFq2bIkpU6bgyJEjxdZ2bW1tHD16FD4+PrCwsICuri5q1KiBpUuXFsuvYOjq6mLfvn1o2bLlW3X1WFhY4OTJk1i+fDk8PDxQunRpaGlpwczMDK6urrLPJt5lm5VkPXr0wJYtW1CrVi1oa2vDwsIC3bt3R1BQ0Ds931kQa2tr7N69Gw0bNpR1DfS7775TeV+Y7tjXGjdujGvXrmHMmDFwdHSEgYEBdHR0UL58efTq1QunTp3CmDFjCj3dvHTr1g179+7F6NGj0aBBA9jb28PQ0BDa2tqwsrKCp6cnVq5ciT179kBTU1Ol7pAhQ3Dp0iUMHDgQVapUgYGBAbS0tGBlZYUmTZpg0qRJuHz5cpG009nZGRs3boSLiwv09PTUljEwMMCBAwewYcMGtGnTBlZWVtDS0oK+vj4qVqyIL7/8EkuXLpWelX1Xvr6+CAoKgre3NxwcHKCnpwcDAwM4ODigd+/eqFGjhlTW0tISZ86cwaJFi9C8eXOUKlUKmpqaMDQ0RLVq1fD1119j/fr1Ks/TTp06FYMGDYKrqyusra2hra0NAwMDVKtWDUOGDMGFCxdkXaopiEKIQtyqSURUBIKCguDm5gYgqwvx4cOHuUKGqKThNUwiei+Sk5MRFBSEly9f4rfffpOGDx48mGFJHwWeYRLRe/HgwYNcz+E5ODjg8uXL0s+GEZVkvIZJRO9d6dKl0b17dxw9epRhSR8NnmESERHJwDNMIiIiGRiYREREMjAwiT5Bfn5+UCgUUCgUub4kOy8PHjyQ6igUimL9Kj2ijxEDk96ZnZ2dygetjo6O9GPV2aWnp8PW1lalbHF9UTt9vHx8fHLtIwqFArq6urC2tkazZs0wZ86cAn8UeNasWbmmsWfPnjzLN23aVO189fT0UL58eXTo0AE7duyQygcEBKgtn98r58HLpk2b0KpVK1hZWUFbWxsmJiawt7dH06ZNMWLECBw4cOCd1iUVLQYmFbm0tDQsXrw41/C///4bT548+QAtok9BamoqwsPDERAQgFGjRsHDw0PlO39zWr16taxhBUlJScHjx4+xa9cudO7cGRMnTiz0NNTp06cPvvrqKxw8eBARERFIT09HbGwsHjx4gGPHjmHevHlYtGhRkcyLiga/uICKxZIlSzBhwgSV34GcN2/eB2xRyZOamgohRK7fDyVVM2fOBJD1xd0bNmyQfjrr1KlT2LNnDzp27Jirzrlz53D9+vVcw3fv3o3o6OhcPyOVk5mZGSZMmID09HTcuXMH69evR2pqKgBg+vTpGDNmDCpWrCi17bWDBw9KXxwOABMmTFD5ebHXvwT0zz//qPzOqqurK1q1agWlUonIyEhcvHgRp0+fzreNH0psbOw7/YLJR+2dv42WPrioqCjx/fffi+bNm4sKFSoIpVIptLW1haWlpfDw8BBr164VmZmZuert3LlTtGrVSlhaWgotLS1hZGQkHBwcRIcOHcSUKVNERkaGrPln/9Ll7F/6vG7dOqnMhQsXpOGampp5fuG0EEJkZGSItWvXCk9PT1G6dGmhra0tSpUqJdq0aSP27t2bq3xaWpr48ccfhZeXl3BwcBAmJiZCS0tLmJubi0aNGol58+aJ1NTUXPWOHz8uOnbsKMqUKSO0tbWFoaGhqFChgmjdurXw9fUVr169ksrm98XqhflC9qtXr4oOHToIc3NzAUBcunRJKhsSEiKGDx8uqlWrJgwMDISenp747LPPxLhx49R+gbkQQly5ckW0bdtWGBkZCSMjI9GqVStx4cKFfL+cOy/qvoR906ZNwtXVVejr64vSpUuLvn37iufPn0t1Vq5cKZXX19dXWWdCCPHy5Uuhra0tldm0aVOB7cj5pf/ZnT9/XmXc1KlT1U5jyJAhUpny5csLPT096f2ff/6ptk72bZVznY0bN05lvqdPn1Y7jZxf6p/zC8pfGzVqlFSmUqVKub6sXQghYmJixIkTJ9TWP3v2rPDx8REVK1YU+vr6wtDQUFSuXFn4+PiIe/fuqZRNTEwUv//+u2jQoIEwNTWVPhu8vLzE5s2bc00755es3717V8ycOVNUq1ZN6OjoiA4dOkhlC/u/+rFjYH4Crl69WuA3+/ft21eljpxfXkhKSpI1/+yB6eHhIZRKpQAg6tWrJ5Xp06ePVKZjx455fiAmJiYW+CsKo0ePVqkTFxdX4LJ4eHiofCgdPnw4V3DnfN28eVMqXxSBWbt2bWFoaKhS9nVg+vv75/lrHwBE2bJlxY0bN1Smfe7cOWldZ3/p6emJFi1avHNgtm3bVm1bHBwcREREhBBCiKSkJGFhYSGNW7Bggco0sweqmZmZSE5OLrAdeQVmfHy8+Omnn1TGqfu1keTkZGFmZiaVmTBhgujUqZP03sXFRe188wvMefPm5QoRdeQG5vDhw6UypUqVyhVy+Zk8ebLsX/4ICwsTjo6O+e7nXbp0EWlpaVKdnIHp7u6u8v51YL7N/+rHjl2ynwANDQ189tlnqFevHqytrWFqaork5GRcunQJu3fvhhACq1atwqBBg1CvXj0AULk2UrduXXzxxRdIT0/H48ePcebMGdy8efOt2mJiYgJvb28sWLAAZ8+eRVBQEBwcHLB582YAQJMmTVCrVi3pB3VzGjVqFA4fPgwA0NHRQY8ePVC5cmVcvXoVW7duhRACv//+O1xdXaXfT1QoFHBwcMDnn3+OsmXLwszMDGlpabh16xa2bt2K9PR0HD58GNu3b5d+l2/p0qXSD+hWq1YNXbt2hZaWFh49eoTg4GDpB2+L0qVLl6ClpYXevXujcuXKuHXrFvT09BAaGoqvvvpKuonF0dERnTp1QmZmJtavX4+HDx/i6dOn6NKlC65evQpNTU0IIdCvXz+V333s2bMn7OzssH379iL5hZu9e/eiWbNmcHd3x8mTJ6Vp3r9/H+PGjcPKlSuhp6eHAQMGYNq0aQCA5cuXY8iQIdI0tm7dKv3ds2fPt+p+zuvGsNe/qpHTzp07VX5cuEePHrh586Z0w87Fixdx9epV1KxZs8B5Z2Rk4M6dO1i5cqU0zMXFReVHit+Gi4uL9PeLFy9QpUoVODs7o27dunB1dUWzZs3UzmPr1q3w9fWV3hsYGEi/LRkaGordu3erlO/Vq5dK1/SXX36J6tWr49ChQ1KX7/bt2zFlyhT89NNPatsaGBgIR0dHtGvXDkII6Xt/3+Z/9aP3QeOaitTDhw/Ftm3bxPz588WsWbPEzJkzRdmyZaWjvZ9//lkq6+TkJA1X170UGhr6Vl2yXbp0Ebdu3ZKOgL/66isxefJkafz27dtzHYW/FhUVpfKbgStXrlSZT/Zuttq1a+dqR3h4uNi5c6dYuHChtPw1atSQ6vTr108q2759e2n4xo0bc00rLCxMJCQkSO+L4gwTgPD39881r+zdc1WqVFE5s3/27JnKmfDOnTuFEEKcPn1aZbo//vijVCcmJkaUKlXqnc8wW7ZsKXXlZ2ZmipYtW0rjdHR0pPXz8OFDlTZeuHBBCCFEdHS0Snfs6+EFyXmGqe5lYWEhrly5ora+l5eXVM7R0VEIkXU2lP1sXN2ZT85tpe5Vt27dfH9XUe4ZZlpaWq7fBM35atSokQgODlap5+LiIo03NDQUt2/fVhkfHx8vwsPDhRBZv4uZfXr/+9//pHLp6enCzc1NGmdubi79v+c8w/z8889z9Ta96//qx4qB+Ql48eJFnt1n2V8DBw6U6gwdOlQarlQqhaenpxgyZIiYP39+nh9EeckZmEK8+dDS1tYWpUuXlj6409PT8wzMffv2FbgMr18KhUL6wE5MTBQ+Pj5qfzQ3ZwC8NnPmTGm4rq6uaNq0qRg4cKCYPXu2CAoKynXNtygCs0aNGmrXX7169WQv97hx44QQQvz5558qw7N3HwshRN++fd85MNeuXasyfs2aNSrjg4KCpHGdO3eWhg8aNEgIodod6+TkJKsNQuQOzJkzZ4qZM2cKX19flR8PNzU1FZcvX1apm/MA45dffpHG9ezZUxpuaWmp0g0pRMGBaWlpKf7555982y43MIUQIjY2VowfP15YWVnlOc/SpUtL3d8JCQkqXbGDBw/Oty0LFy5Umdb169dVxi9YsEBl/Osu/5yBuW3btlzTftv/1Y8dHyv5BHzzzTfYu3dvgeVSUlKkv6dMmQIvLy8AWXcfHjp0CAsXLsSwYcPg5OSEpk2bIiEh4a3b9PoHgtPS0hAZGQkAGDp0aL4/4xQdHS17+kIIREVFAQDGjx+P1atXIzMzM9862Zd/5MiR6N27NzQ1NZGSkoKAgAAsXboUY8aMweeffw4nJyeEhYXlOe+8ppufatWqqR1emOV+vS5fvXqlMtzS0lLlvZWVlexp5qWgaWZvQ/YfhN64cSMSExOxZcsWadjb/ED0a2PHjsXYsWPh5+eHkydPonLlytL8f/jhB5Wya9eulbragazu2Ne++uor6e+IiAjs27cvz3mamZlh5syZ+P7776XljoiIQNu2bXH06NG3XpbsjIyMMGXKFISFheHatWtYsWIFvL29YWRkJJWJjIyU7qZ9+fKlyr6X85dfcsq5X+XcfjnfZ+/Gzk7dfvu2/6sfO17D/MglJCSoPIzdokULLF26FBUqVICmpibq1auHc+fO5apnbGyMffv24cmTJwgKCsKdO3dw48YN7NixA4mJiTh27BhmzJiByZMnv1W7WrVqhapVq+L27dsAsq619O/fP986OW/1HzVqFMqUKZNn+de36L++PgoANWvWxMaNG1G1alVoaWmhW7duKtfRXtPS0sLatWsxe/ZsnDp1Crdv38bt27exY8cOvHz5EteuXcMPP/yANWvWAMi6TvxazgfmXz/mUBBDQ0O1w7Mvt6OjI3x8fPKcxutfpjc1NVUZHhERoTKd8PBwWW3KT0REhMr7nNPM3oYmTZqgZs2auHr1KmJiYrBkyRLpmqeOjg569er1zu0BAG1tbTg7O6s8WpLd6+312utwVWf16tVo37692nHGxsYYO3YsAGDgwIFwdnZGQkICMjIyMGTIEFy7dg1aWkXz8alQKODo6AhHR0f069cPfn5+qFixonQA+HpZzczMoFAopNAMDQ3Nd7o5/5/Cw8NhYWGh8j677I+/ZKduv33b/9WP3gc9v6V39vTpU5Xuj99//10ad+vWLaGrq6u2K/Hq1atqH7X47rvvpPJffPGFrDao65IVQrXLJ3t3cH7XMLN3p2W/5pJdaGioStdY9utk3333nTQ8IiJC6g4GIJo0aaKybtR1E/3+++9qu1Cz32VZqVIlkZKSIoQQ4smTJyrXC3P+S+XXlfvayJEjpTLm5ubiyZMnucqkpaWJv//+W0RHRwshStY1zNeWLl0qjc/+GEf2fUKO/B4rSUtLE5UrV5bGKZVKaVxQUJDsbkIg63JB9sd18rtLNuc+u3r1arVtl9slu3r1arF48WIRExOTa1xERITK9cEffvhBGpf9GqZSqcx1t25iYqJ0DTM4OFilLW97DVPdMrzt/+rHjmeYHzlLS0uYmppK3WO//vqr9K0hK1euzLO7cOzYsTh79ixatGgBW1tblC5dGs+ePcOqVaukMjnPYgrLx8dHOuqsX79+geXNzc3Rr18/LFu2DAAwY8YMnD9/Hg0aNICenh6ePn2KoKAgXLp0Cd7e3mjVqhUAoGrVqrh27RoAYNmyZdDQ0ICBgQHWrVsndWHm9Mcff2DdunVo0aIF7O3tYWVlhejoaKxdu1Yqk33569atK91lee/ePbi4uOCzzz7Dv//++87dTcOHD8fixYuRnJyM6OhoODs7o2vXrrC1tUV8fDxu3LiBgIAAvHr1CqGhoTAzM0P9+vXh6Ogo3QH522+/4cGDB7Czs8O2bdvw4sWLd2oTkPUQfosWLdC4cWOcOHFC5c7bnj17wsDAQKV8r169MG7cOLx8+RLJycnS8L59+75TO2bNmgUgqzdl//79Kmf0DRs2lP7Ovu8qFAp07do11x228fHx0uWLtLQ0rF+/HiNGjCiwDSNGjMDs2bOlu5KnTZuG3r17q/Q8FEZoaCgmT56MkSNHolGjRnB2doa5uTmioqKwbds2lW8wat26tfT3Dz/8IN3pHR8fD2dnZ+ku2cePH2PPnj1YuHAhOnbsiFq1aqFFixbSdpsxYwbu378PR0dHHDx4UOWLEUaMGFGoZXnb/9WP3odObHp306ZNU3sEXaNGDeHq6qr2DKdVq1b5Hn3r6emJs2fPypp/XmeYecnrDFOIrBsbCnq2K+eybNy4UW0ZGxsb4enpqfYM89tvv813+hoaGirPs4WHh6s8b5i9XM51mZ2cM0whhNixY0euZzTVvbIf7Z85c0ZtHW1tbdGgQYN3PsNs2rSp2jbY2dlJZzE5jR07Ntc2UPdQfn7k3CULqN70k5SUJExNTaVxHh4eaqedmZmpsr86OztL4/I7w1S3bOoe+pd7hpmzXF6vAQMG5Krr5+dXqOcwq1evnu88CnoOM69leJv/1Y8dA/MTsWDBAlGlShWhra0trK2txYABA0RUVFSeH9iHDx8WI0aMEJ9//rkoW7as0NHREbq6usLBwUF4e3sX6k7ZogxMIbK+PWTDhg2iTZs2wsrKSmhpaQl9fX1RsWJF8eWXX4qlS5eKqKgolTpbtmwRtWrVEtra2sLCwkJ0795dPHz4UOXDN3tgnj17VowbN040btxY2NraCj09PaGjoyNsbW1F165dRWBgYK52Xbt2TXh5eQmlUikMDQ1F8+bNxfHjxwv1TT/5CQ0NFaNHjxY1a9YUSqVSaGpqCgsLC+Hm5ia+//57cfLkyVx1goODpTYplUrRokULcfLkySL7pp81a9aI2rVrCz09PWFhYSG8vb3Fs2fP8p1G9ruVX9/VWxh5BaZCoRBGRkaiVq1aYvTo0eLRo0dSnZwHTRs2bMhz+jm//OD1oxsFBWZYWJjKJY5atWrluptabmC+fPlSbNu2TQwZMkTUq1dPlC9fXujr6wsdHR1RtmxZ0b59e7F9+/Y8l+HMmTPC29tbODg4CD09PWFgYCAcHBxE7969c3XTJiQkiNmzZws3NzfpW7BKly4tWrdurfabl+QGphBv97/6MVMIkeOWPyKit5ScnAxra2vExMQAAG7duoWqVat+4FYRFQ1ewySidxYUFIRXr15h7dq1Ulh6eHgwLOmTwjNMInpndnZ2ePjwofReR0cHQUFBqF279gdsFVHR4hcXEFGRMTIyQuPGjXH48GGGJX1yeIZJREQkA88wiYiIZGBgEhERycDAJCIikoGBSUREJAMDk4iISAYGJhERkQwMTCIiIhkYmERERDIwMImIiGRgYBIREcnAwCQiIpKBgUlERCQDA5OIiEgGBiYREZEMDEwiIiIZGJhEREQyMDCJiIhkYGASERHJwMAkIiKSgYFJREQkAwOzBIiMjMTgwYNRvnx56OrqwtraGq1atcLJkyeLdD5NmzbFyJEjC1Xn3LlzKFOmDADg2bNn0NfXR2pqar51jh8/jnbt2qFMmTJQKBTw9/d/yxYXn09tnU+dOhV169aFkZERLC0t0bFjR9y+ffttm10sPrV1vmjRIjg5OcHY2BjGxsZwc3PD/v3737bZRe5TW9/ZTZs2DQqFotDzfVda73VupFaXLl2QmpqKNWvWwMHBAeHh4Thy5AiioqI+dNNw+vRpNGzYEAAQGBiIOnXqQEdHJ986CQkJqFWrFvr164fOnTu/j2YW2qe2zo8dO4ahQ4eibt26SE9Px4QJE9CyZUvcuHEDhoaG76PZBfrU1nm5cuUwbdo0VK5cGUIIrFmzBh06dMClS5fg6Oj4Ppqdr09tfb927tw5LFmyBE5OTsXZRPUEfVAvX74UAERAQECB5b755htRqlQpYWRkJJo1ayaCg4Ol8b6+vqJWrVpi7dq1okKFCsLY2Fh0795dxMbGCiGE8Pb2FgBUXqGhoQW2r3v37uKPP/4QQggxbNgwMW7cuEItHwCxY8eOQtUpbp/6OhdCiIiICAFAHDt2rNB1i8N/YZ0LIYSZmZlYvnz5W9UtSp/q+o6LixOVK1cWhw4dEk2aNBEjRoyQVa+oMDA/sLS0NKFUKsXIkSNFcnJynuU8PDxEu3btxLlz58SdO3fEmDFjhIWFhYiKihJCZO3YSqVSdO7cWVy9elUcP35cWFtbiwkTJgghhHj16pVwc3MTAwYMEGFhYSIsLEykp6ernVdgYKAwMTERJiYmQlNTUxgYGAgTExOhpaUl9PX1hYmJiZg6daqs5SuJgfmpr3MhhLh7964AIK5evVqINVN8PvV1np6eLjZu3Ch0dHTE9evXC7l2it6nur779OkjRo4cKYQQDMz/qm3btgkzMzOhp6cnGjRoIMaPHy8uX74sjQ8MDBTGxsa5dvyKFSuKJUuWCCGydmwDAwPpyE8IIb7//ntRv3596b3cHSwpKUmEhoaK/fv3CzMzM3H//n1x/vx5oaOjI27evClCQ0PFy5cvZS1bSQxMIT7tdZ6RkSHatm0rGjZsKKv8+/IprvMrV64IQ0NDoampKUxMTMTevXtlrIn341Nb3xs3bhQ1atQQSUlJhZpvUeJNPyVAly5d8OzZM+zatQutW7dGQEAAXFxcsHr1agDA5cuXER8fDwsLCyiVSukVGhqKkJAQaTp2dnYwMjKS3tvY2CAiIqLQ7dHT04OdnR2uXLkCLy8v2Nvb49atW3B3d0e1atVgZ2cHU1PTd13sD+pTXudDhw7FtWvXsGnTpkK3ozh9iuu8atWqCA4OxpkzZzB48GB4e3vjxo0bhW5LcfiU1vfjx48xYsQIrF+/Hnp6eoWed1HhTT8lhJ6eHjw9PeHp6YlJkyahf//+8PX1hY+PD+Lj42FjY4OAgIBc9bLvYNra2irjFAoFMjMzC90WpVIJAEhJSYGGhgZ27tyJ1NRUCCGgVCrh7u5eou4GfFuf4jofNmwY9uzZg+PHj6NcuXKFbkdx+9TWuY6ODipVqgQAcHV1xblz5zB37lwsWbKk0O0pDp/K+r5w4QIiIiLg4uIiDcvIyMDx48cxf/58pKSkQFNTs9BtKiwGZglVvXp16XEMFxcXPH/+HFpaWrCzs3vraero6CAjI6PAcsHBwUhPT4ezszMOHz4Ma2truLu7Y+HChahZsyb09fXfug0l2ce8zoUQGD58OHbs2IGAgADY29u/dZvfp495nauTmZmJlJSUt2n2e/Gxru8WLVrg6tWrKsP69u2LatWqYdy4ce8lLAEG5gcXFRWFrl27ol+/fnBycoKRkRHOnz+PGTNmoEOHDgAADw8PuLm5oWPHjpgxYwaqVKmCZ8+eYe/evejUqRPq1Kkja152dnY4c+YMHjx4AKVSCXNzc2ho5O6Vr1SpEoKCgmBlZYVGjRrh0aNHiIuLQ7t27aClVfAuEx8fj3v37knvQ0NDERwcDHNzc5QvX17mmik+n+I6Hzp0KDZs2ICdO3fCyMgIz58/BwCYmJiUiAOcT3Gdjx8/Hl5eXihfvjzi4uKwYcMGBAQE4MCBA4VbOcXgU1vfRkZGqFGjhsowQ0NDWFhY5BpenBiYH5hSqUT9+vXxxx9/ICQkBGlpabC1tcWAAQMwYcIEAFldIPv27cPEiRPRt29fREZGwtraGo0bN4aVlZXseY0dOxbe3t6oXr06kpKSEBoamueRZUBAABo3bgwg6xk/Nzc3WR8iAHD+/Hk0a9ZMej969GgAgLe3t3T95EP6FNf5okWLAGQ9RJ7dqlWr4OPjI7u9xeVTXOcRERHo06cPwsLCYGJiAicnJxw4cACenp6y21pcPsX1XRIohBDiQzeCiIiopONdskRERDIwMImIiGRgYBIREcnAwCQiIpKBgUlERCQDA5OIiEgGBiYREZEMDMz/qJSUFPj5+ZXor/H6lHB9v39c5+/fp77O+cUF/1GxsbEwMTFBTEwMjI2NP3RzPnlc3+8f1/n796mvc55hEhERycDAJCIikoGBSUREJAOvYRIR0X/e7eeJqGptkG8ZnmESERHJ8PH8ENlHoFanXxAdk/Chm/FJMjcxxOUdk1SGXX8Sj4xMdpAUF00NBRzLKVWGHbkVgdQMrvPioKOpQItqlirD7kckgrt48dFQAA6W+Z9VZsfALELRMQl48TL+QzfjPyMjUyCdnybvVWqGYGC+R5kC4OouOdglS0REJAMDk4iISAYGJhERkQwMTCIiIhkYmERERDIwMImIiGRgYBIREcnAwCQiIpKBgUlERCQDA5OIiEgGBiYREZEMDEwiIiIZGJhEREQyMDCJiIhkYGASERHJwMAkIiKSgYFJREQkAwOTiIhIBgYmERGRDAxMIiIiGRiYREREMjAwiYiIZGBgEhERycDAJCIikoGBSUREJAMDk4iISAYGJhERkQwMTCIiIhkYmERERDIwMImIiGRgYBIREcnAwCQiIpKBgUlERCQDA5OIiEgGBiYREZEMDEwiIiIZGJifmEmD2+L+wd8Qffp37F08DBXLl863/Nh+LXHir+8RcWIWHh6Zii2/D0DlCpZ5lvefPxhJl+ajXVOnom76R2fJnKno3KIOGla3QdNa5TH46/a4eul8vnVWLpyN3h2awr1GWXjUqYjRA3viQchdlTIvIsMxadRAtKxbGQ2r26DnF+44sn9ncS7KRyE9LQ1r/vgV33Vphu71HdDXwxlzJg5HdMTzfOtlZGRg/fzpGOhVD93q2ePbtp9j85LfIYSQynSsZaP2tWP1wuJerBLtz1m/watRbdR2KI161cqib7e2uHzxXIF1qtkYqry8GtVWKbN53Ur07twarpWtUc3GELExr4pxKYqO1oduABWdMT4eGPJVEwz4aR0ePI3CT0O+wO4FQ1G7y69ISU1XW8fdpRIWbz6OC9cfQktLE5OHtcOeRcNQu/OvSExOVSk7vFczZPuM+c8rb18J4ybPRNnydkhJTsb6FQsw1LsTdv57CWYWpdTWuXjmJLr2HgBHJxdkpKdj/qyfMbRPJ2w7dAb6BoYAgJ9Gf4v42Bj8vmwTTM3N8c/ObfhhmA/W7QpANcda73MRS5SU5CTcv3UV3QaOgn3V6oiPjcHy6ZPw2whvzN54IM96f6+aj3+2rsGIX+bBtmJVhNy4jHk/jYSh0hhf9OoPAFh15LJKnYsnjmK+32i4ebQt1mUq6ewcKmHSlNmwrWCP5OQkrFk6H9/0aI+Dp67AvFTeB+OVq36GlVv2SO+1NFWjJjkpEe7NPODezAO/T/EttvYXtUKdYUZGRmLw4MEoX748dHV1YW1tjVatWuHkyZNF2qimTZti5MiRhapz7tw5lClTBgDw7Nkz6OvrIzU1Nd86x48fR7t27VCmTBkoFAr4+/u/ZYtLhqE9m2H6sgPYE3AV1+4+Q/9Ja2FT2gTtm+X9Idth2EL8tfsMbt5/jqt3nmKg718ob2OO2tVtVco5VSmLEb2bY5DfX8W9GB8Nrw5dUb9RM5Qrb4+KVT7D6B+nICEuFndvXcuzzvw1f6P9l71QscpnqFK9JibPXITnzx7j5tVgqcyVi2fR3ftb1HB2Rbny9ug//HsYGZuolPkvMjQyxuQlm9GoVXuUtauEqk6uGDh+CkJuXEFk2JM8690OPo96TVujTmMPWJW1RQPPL+Ds1gR3r12SypiVslR5nQn4BzXqNoR1uQrvY9FKrHadu6NB4+awrWCPylWr4we/aYiPi8Xtm3nv4wCgqaWF0pbW0ivnAaT3wGEYOHwsarnWK87mF7lCBWaXLl1w6dIlrFmzBnfu3MGuXbvQtGlTREVFFVf7ZDt9+jQaNmwIAAgMDESdOnWgo6OTb52EhATUqlULCxYseB9NLFZ2ZS1gU9oER8/ckobFxifj3LUHqO9kJ3s6xko9AMDLmERpmL6eNlZP9cHIaVsQHhVXZG3+lKSlpuLvjauhNDJB5c9qyq4XHxcDADA2NZOGObnUw8G9fyPmVTQyMzNxYPc2pKSkoM7njYq83R+7xPhYKBQKGBqZ5FmmqnMdXDkbiKcPQgAAobev4+als3Bp1Fxt+VdRkbgQeAQenb4qljZ/rFJTU7H5r5UwMjZBter57+MP74fA3bkiPOo7YuyQvnj25PF7amXxkt0l++rVKwQGBiIgIABNmjQBAFSoUAH16tXLVW7s2LHYuXNn1j95nTr4448/UKtW1lmOn58f/P39MWbMGEyaNAkvX76El5cXli1bBiMjI/j4+ODYsWM4duwY5s6dCwAIDQ2FnZ1dvu07deqUFJgnTpyQ/s6Pl5cXvLy85K6CEs26lDEAICJaNdAiouJgZWEsaxoKhQIzx36JU5dCcCMkTBo+Y0wXBF0OxZ6Aq0XX4E/E8SP/YMJ3/ZCclIhSltZYuG4HzMwtZNXNzMzErF/Go1adz1GpanVp+PQFq/HDsL5oXtsemlpa0NM3wKzFf8HWrmJxLcZHKTUlGWvm/Ap3r44wUBrlWa5Lv+FIio/HsI7u0NDURGZGBnoN/wFN2nZRW/7ori3QN1DCrUWb4mr6R+XfQ/sxZpA3kpISUdrKGis3787zkgMA1KpdB1PnLoF9xcqICH+OBb9PxdcdPbEr4ByU+Wynj4HsM0ylUgmlUgl/f3+kpKTkWa5r166IiIjA/v37ceHCBbi4uKBFixaIjo6WyoSEhMDf3x979uzBnj17cOzYMUybNg0AMHfuXLi5uWHAgAEICwtDWFgYbG1t1c7rxIkTMDU1hampKbZt24aJEyfC1NQUixcvxrx582BqaipNt6ikpKQgNjZW5ZXf+iguPbzqIPLkbOmlpaX5ztOcM74bHCvZoM8Pq6RhbZvURNN6VfD9zG3vPP2P2T7/LWjkWEZ6XTp7CgBQ180dG/cGYtX2Q2jQpAV+GOaD6BeRsqY57acxCLl9E1PnrVQZvmj2b4iLjcGiv3bir50B+PqbofhhWF/cvXW9yJerJDu2dzt6fF5Rel2/GCSNS09Lw8zvvwWEwKCJ0/OdzskDu3Bs398YPXUhZm86iO9+mYudaxbj6K4tassf8d+Ixm06Q0dXr0iXp6TbvX0TXCpaSq/zQVmX2uo3bIwdh09j4+6jcG/miZEDeyPqRUSe02ncohVat+uMqtVrwr2ZJ5b+9TdiY2Pwz66/39eiFBvZZ5haWlpYvXo1BgwYgMWLF8PFxQVNmjRBjx494OSUdcfkiRMncPbsWUREREBXVxcAMGvWLPj7+2Pbtm0YOHAggKwj69WrV8PIKOtoo3fv3jhy5Ah+++03mJiYQEdHBwYGBrC2ts63TXXq1EFwcDBu3bqFnj174sKFC4iOjkaDBg1w8eJF6OnpwdTU9G3WS56mTp2KyZMnqwzz9fWFn59fkc6nIHuOXcXZaw+k97raWZvS0twIz1/ESsMtLYxw5Xbe13de+2NcV7RxrwGPb+bgacQraXjTulXgUK4Unh+fqVJ+46z+OHkpBK0GzH23BflINPHwQk1nV+l9aeus6+X6BoawtasIW7uKqFm7Ljo2qw3/LWvRb8iYfKc3/aexOHH0AJZt3gcrm7LS8McP72Pz2qXYciAIFat8BgCoUr0mLp07ha3rlmHCb3OKfuFKqHpNW6FKTRfpvbll1udBVlgORGTYE/y8bGu+Z5cAsPqPX9Cl3zC4e3UEANhV/gyRYU+wfcU8NG/fTaXs9YtBePogBGNnLCnahfkINGvVFk4udaX3Vv+/jxsYGKKCfUVUsK8IZ9d6aNXACds2rMG3330va7rGJqawc6iEh6EhxdLu96lQd8l26dIFbdu2RWBgIIKCgrB//37MmDEDy5cvh4+PDy5fvoz4+HhYWKh2SSUlJSEk5M3KsrOzk8ISAGxsbBARkfcRS1709PRgZ2eHLVu2wMvLC/b29jh16hTc3d1RrVq1Qk9PjvHjx2P06NEqw14fHLxP8YkpiE9UPbMNi4xBs/pVceXOUwCAkaEe6taww7KtJ/Kd1h/juqJ981poOWAuHj5TvR49a9VBrNpxSmXYhW0T8b/Z27H3WP4X/j8lhkojGMroTsrMzERaPjebCSEww/d7/HtwD5Zu3IuytnYq45OTkgAAGhqqnT8aGprIzMwsfMM/YvqGSugbKlWGvQ7LsEeh+GX5Nhibmhc4ndTkJChyrk9NTYjM3Ld8H96xERWrO8G+quO7Nf4jpFQayeoyzczMLPCGyuwSEuLx+GEo2n/58V8TLvRjJXp6evD09ISnpycmTZqE/v37w9fXFz4+PoiPj4eNjQ0CAgJy1ct+pqetra0yTqFQvNWHgVKZ9c+UkpICDQ0N7Ny5E6mpqRBCQKlUwt3dHfv37y/0dPOjq6v7QQJSjgUb/sW4/q1x71EkHjyNgu+QtgiLjMGuf9/cMr9v8XDs+vcyFm8+DiCrG7a7Vx10HbUU8QnJsLLI+oeJiU9GckoawqPi1N7o8zjsZa5w/S9JSkzAigWz0MSjDUqVtsKrl1HYsm45Ip+HwaNNR6ncoF7t0KxlO3T3zupdmfbTGPyzcxt+X7oBBkolXkSGAwCURsbQ09OHXcUqsLVzwG8TRmLkhF9hYmaGgIN7cebEv5izQn0X4n9FeloaZowdgJCbV/Hjn2uRmZmJl//fNag0MYW2dtZNfpMGdMXnzb3Q9qt+AIA6TTyxbdlclLYuC9uKVRF66yp2rVuCFh1UP8AT4+Nw6uBu9B3z8TzmUJwSExOweM4MNG/VFqUtrfEyOgobVi9B+PNnaN2uk1TOp2sbeHi1x9f9BgEApk8ej2aebVDGtjwinodh/qxfoaGhiS86dpXqREY8x4uIcDwKvQ8AuHPzOgyVStiUtYWpWcEHQR/KOz+HWb16delxDBcXFzx//hxaWloF3qSTHx0dHWRkZBRYLjg4GOnp6XB2dsbhw4dhbW0Nd3d3LFy4EDVr1oS+vv5bt+FjNHv1YRjo62L+j1/B1Egfp4JD0H7oQpVnMB1sS8HC9M1R+7fdGgMADi0fqTKtAT+tw1+7z7yXdn+MNDQ18SDkDvZs34hXL6NgYmoORycXLN+yX+pKBYAnDx/g1cs3Bxbb/loBABj4lerzfb4zF6L9l72gra2NeSu34c8ZvhjVvzsSExNgW8EBk2ctRqNmLd/PwpVQURHPcTYg63nLUd08VMb9snw7atZtAAB4/uQBYl+9uWdi4A+/Yf2C6Vgy5QfEREfBrLQVWn3ZG92+Ve0pCvzHHwIC7l6dQICmhiZC793Bd1vX42V0FEzNzFHT2RXr/Q+hcrab1B49CMXL6Df7eHjYM4wZ4oNXL6NhblEKrvUaYPPef1We29y0dgUWzJ4ivf+6U9a+PWXOYnTu3vs9LN3bUQgh71H0qKgodO3aFf369YOTkxOMjIxw/vx5DB8+HG3btsWKFSsghEDjxo0RFxeHGTNmoEqVKnj27Bn27t2LTp06oU6dOtJdssHBwdK058yZgzlz5uDBgwcAgIEDByI4OBhbtmyBUqmEubl5ri6q14KCgtC9e3c8fPgQjx49QvXq1fHq1StoaRV8LBAfH4979+4BAGrXro3ff/8dzZo1g7m5OcqXLy9ntaiwbf4DXryML3Q9KlgpMyUeH1W9gevKozikq+lWo6KhpaGAU3nVLrr918ORmsF1Xhx0NBXwcrRSGXYvPBFc3cVHUwFUsjIAANx+noiq1gb5lpd9hqlUKlG/fn388ccfCAkJQVpaGmxtbTFgwABMmDABQFbX6r59+zBx4kT07dsXkZGRsLa2RuPGjWFlZVXAHN4YO3YsvL29Ub16dSQlJeX7WElAQAAaN846Szp27Bjc3NxkhSUAnD9/Hs2aNZPev7426e3tjdWrV8tuLxERffpkn2FSwXiGWXx4hvn+8Qzz/eIZ5vtX2DNMfvk6ERGRDAxMIiIiGRiYREREMjAwiYiIZGBgEhERycDAJCIikoGBSUREJAMDk4iISAYGJhERkQwMTCIiIhkYmERERDIwMImIiGRgYBIREcnAwCQiIpKBgUlERCQDA5OIiEgGBiYREZEMDEwiIiIZGJhEREQyMDCJiIhkYGASERHJwMAkIiKSgYFJREQkAwOTiIhIBgYmERGRDAxMIiIiGRiYREREMjAwiYiIZGBgEhERycDAJCIikoGBSUREJAMDk4iISAYGJhERkQwMTCIiIhkYmERERDIwMImIiGTQ+tAN+JSYmxh+6CZ8stStW00NxQdoyX+HuvWro8l1XlzUrVvu4sWrsOtXIYQQxdMUIiKij8Pt54moam2Qbxl2yRIREcnAwCQiov+8kBfxBZZhYBIREcnAm36KUHL6h27Bp00vx95aq9OviI5J+DCN+Q8wNzHE5R0/qgw7F/oK6Zm87aE4aGkoUNfeVGXYsTsvkJbB9V1ctDUVaFKllOzyDEz6aEXHJODFq4K7UajopGcKfoC/R2kZAqlc3yUGu2SJiIhkYGASERHJwMAkIiKSgYFJREQkAwOTiIhIBgYmERGRDAxMIiIiGRiYREREMjAwiYiIZGBgEhERycDAJCIikoGBSUREJAMDk4iISAYGJhERkQwMTCIiIhkYmERERDIwMImIiGRgYBIREcnAwCQiIpKBgUlERCQDA5OIiEgGBiYREZEMDEwiIiIZGJhEREQyMDCJiIhkYGASERHJwMAkIiKSgYFJREQkAwOTiIhIBgYmERGRDAxMIiIiGRiYREREMjAwiYiIZGBgEhERycDAJCIikoGBSUREJAMDk4iISAYG5idq+JBB0NdW4M+5c/ItdyLwOLp0bAf78mWgr63Arp3+ucr47/gbX3i1RFkrC+hrK3A5OLhY2vwxmjSoDe4f+BXRp2Zj76JhqGhbOt/yY/t64sS6sYgInImHh6dgy+wBqFzBUhpf3sYcSRf/VPvq7OFczEtT8q36czp6t66PVs62aFvXAaN9OuHG5fMF1tuxfjm6N3eGZ80yGNTVEzevXFAZ//RRKCYO7Y32n1eBl0sF+I7oh+gXEcW1GB+F9LQ0rPz9Fwzp1ASd69mhd3MnzJ4wDFERz/Otl5gQj6XTf4RPS1d0qlMBY75uizvXLuVZfv7P36NtTSv4r1tS1ItQ5AoVmJGRkRg8eDDKly8PXV1dWFtbo1WrVjh58mSRNqpp06YYOXJkoeqcO3cOZcqUAQA8e/YM+vr6SE1NzbfO1KlTUbduXRgZGcHS0hIdO3bE7du337bZJcZO/x04eyYINv+/PvKTkJCAmk61MGfegjzLJCYkoEHDRvh1yvSibOZHb4y3B4Z81QTfTdmMxt6zkZCUgt0LhkBXRyvPOu6ulbB4SyCaeM/GF4MXQEtLE3sWDoWBng4A4En4S9h5TlB5/bxoL+ISknHg5I33tWglVjm7ihjx03Ss2h2I+Rv2wbpseYzt9yVeRb/Is87RfTuwYOokeA/9Hst2HEXFajUw9puueBkVCQBISkzA2H5fQqFQ4I81/pi/cT/S01IxflBPZGZmvq9FK3FSkpMQcvMKvvp2NOZtPoyJf6zEkwf38PPwPvnWm+c7CpdOH8fYKfOx4O8AuDRoiokDuuJFeFiusqeO7MOtKxdgYWldXItRpPL+z1ajS5cuSE1NxZo1a+Dg4IDw8HAcOXIEUVFRxdU+2U6fPo2GDRsCAAIDA1GnTh3o6OjkW+fYsWMYOnQo6tati/T0dEyYMAEtW7bEjRs3YGho+D6aXeSePn2K0SOHY/feA+jUoW2B5Vu19kKr1l75lun5dW8AwMMHD4qiiZ+MoT2bYvryA9hz7CoAoP9P6/Dw0BS0b+qErQcvqq3TYdgilfcDff/C46NTUbu6LU5eDEFmpkB4VJxKmfbNnLD90CUkJOV/APhf4NnuS5X3Q8f/gr3b/kLI7etwdWuits6WVQvxRbfeaNOlFwBgzOTZCAo4iH3b16PXwJG4dvEsnj99hOX+/8JQaQwAGD99Ib6o64CLQcdRp0HTYl2mksrQyBi/LduqMmzwhKkY9VVrRIQ9gaVNuVx1UpKTcPLwXkyatwY16rgBAHoN+R5nAg5i3+bV6PPdeKnsi/AwLJ4yAb8s2QS/oV8X78IUEdlnmK9evUJgYCCmT5+OZs2aoUKFCqhXrx7Gjx+P9u3bq5Tr378/SpcuDWNjYzRv3hyXL1+Wxvv5+cHZ2Rnr1q2DnZ0dTExM0KNHD8TFZX1I+Pj44NixY5g7dy4UCgUUCgUeyPigPnXqlBSYJ06ckP7Ozz///AMfHx84OjqiVq1aWL16NR49eoQLFy4UWLckyszMxDc+vTFq9Peo7uj4oZvzSbMrawGb0iY4euZNj0RsfDLOXXuA+k72sqdjbKQHAHgZk6h2fO3PbOFczRZr/E+/W4M/QWmpqdi9eS2URsaoWLVGnmXuXL8M1wZvwlRDQwOuDZrg+qVzAIDU1BQoFApo6+hKZXR0daGhoYGrF84U70J8ZBLiYqFQKKA0MlE7PiMjA5kZGdDJti4BQFdPDzcunZXeZ2ZmYvaEoejSdwgqVKpWrG0uSrIDU6lUQqlUwt/fHykpKXmW69q1KyIiIrB//35cuHABLi4uaNGiBaKjo6UyISEh8Pf3x549e7Bnzx4cO3YM06ZNAwDMnTsXbm5uGDBgAMLCwhAWFgZbW1u18zpx4gRMTU1hamqKbdu2YeLEiTA1NcXixYsxb948mJqaStOVIyYmBgBgbm4uu05JMnvmdGhpaWHo8O8+dFM+edYWWWciEdGqZ4MRUXGwKmUsaxoKhQIzx3bBqUshuBGSu7sKALw7uOHm/TAEXQl9twZ/Qk79ewCta5eHp1MZbF29CLNWboepuYXasjEvo5CRkQEzC0uV4WYWltI1SkfnOtDTN8CSmZORnJSIpMQELJz+EzIyMhAVGV7sy/OxSE1Jxqo/fkUTr04wUBqpLWNgqES1WnWwackfiIp4joyMDBzdvQ23Lp9H9Is363Lbyj+hqamF9r0GvK/mFwnZgamlpYXVq1djzZo1MDU1RcOGDTFhwgRcuXJFKnPixAmcPXsWW7duRZ06dVC5cmXMmjVLCrTXMjMzsXr1atSoUQPu7u7o3bs3jhw5AgAwMTGBjo4ODAwMYG1tDWtra2hqaqptU506dRAcHIxNmzbB2NgY165dw5EjR6ChoYGLFy8iODgYgwYNkrV8mZmZGDlyJBo2bIgaNdQfrQJASkoKYmNjVV75HUAUl40b1qOUqVJ6BR4/hgV/zsXSFauhUCjee3s+dT286iDyxCzppaWlfp8sjDk/dIVjRRv0Gb9a7Xg9XW1093LFGv+gd57Xx+jQrq1oXbu89Lp8Pussu3b9RljuH4AFm/ajnnsL+I38Rroe+TZMzUth8txVUhC3rWOP+NgYVHGs9Z/6X/p3zzZ0qWcvva5deLPfpaelYerYAQAEhk6ake90xk5dACEE+rSohY6utti9YRkae3WCQpEVN3evX8bOv5Zh1K/zPrr1W+hrmG3btkVgYCCCgoKwf/9+zJgxA8uXL4ePjw8uX76M+Ph4WFioHu0lJSUhJCREem9nZwcjozdHKDY2NoiIKPwdaXp6erCzs8OWLVvg5eUFe3t7nDp1Cu7u7qhWrXCn+UOHDsW1a9dw4sSJfMtNnToVkydPVhnm6+sLPz+/wjb/nXzRrj3q1asvvf97+1ZERESgikN5aVhGRgZ++N8YzP9zDm7fe/Be2/ep2XPsKs5eeyC919XO+texNDfC8xex0nBLCyNcuf20wOn9Ma4r2rjXgEf/uXga8UptmU4ezjDQ08H6PWfVjv/UNWzeGp/VcpXel7ayAQDoGxiiXAUHlKvgAEfnuujZsi72bvsLX387Ktc0TMwsoKmpiZdRqp8vL6MiYF7qzVln3UbNsPHwBbyKjoKmlhaMjE3QqeFnKNOmUzEtXclTv1lrVHV6s75f34iTnpaGaWMHIPLZE0xZsT3Ps8vXbGztMH21P5ITE5CYEA/z0laYNnYArMtVAABcvxiEmOgX8GnpItXJzMjAill+2PnXMqw6UPBdzx9KoQITyAopT09PeHp6YtKkSejfvz98fX3h4+OD+Ph42NjYICAgIFc9U1NT6W9tbW2VcQqF4q3uRlMqlQCyzvo0NDSwc+dOpKamQggBpVIJd3d37N+/v8DpDBs2DHv27MHx48dRrlzuC9nZjR8/HqNHj1YZpqurm0fp4mNkZKRy0NGv/0C0adtOpUy7tq3Qs1dv9PHu+76b98mJT0xBfKJqT0JYZAya1auKK3eyAtLIUA91a9hh2db8D7r+GNcV7Zs5oeWAeXj4LO8b5nw6uGHvsat48Sr+3RfgI2SgNCrwwxkARGYm0vK4I15bRwdVHGvhwunjcPfIugkuMzMTF08fR6ev++cq/7pr9+Lp43gZFYmGzVu/wxJ8XAwMlTAwVKoMex2Wzx7dx9QVf8PYVP7lKj0DQ+gZGCIu5hUungpA31GTAADN23WF8+eNVcr+NKgHmn3xJTw7fvXuC1KMCh2YOVWvXh3+/v4AABcXFzx//hxaWlqws7N762nq6OggIyOjwHLBwcFIT0+Hs7MzDh8+DGtra7i7u2PhwoWoWbMm9PX1860vhMDw4cOxY8cOBAQEwN6+4Js1dHV1P0hAFsTCwiLXmb22tjasrKxRpWpVaZhXyxZo36ETBg8dBgCIj49HyL170vgHoaG4HBwMM3NzlC+fdbYaHR2Nx48eISzsGQDgzp2sG12s/r/L/L9qwYYAjOvfCvceReDBsyj4Dv4CYZEx2BXw5jLFvsXDsOvfK1i8+TgAYM4P3dDdyxVdRy1DfGIyrCyyAiEmPhnJKWlSPQfbUmjkUhEdv1v8fheqBEtKTMC6xb+jYfPWsChtjZiXUdixfgVehIehaesOUrlR3h3h7tkWnb/Ouj7Wre8QTB03FNVqOKOakwu2rVmCpKREeHXuKdXZt309KlSsAlPzUrh+6Rz+nDIBXX0Go7xD5fe+nCVFeloapoz+BiE3r8J3wV/IyMyUrvsamZhCWzvrKYQJ/bvArXkbtOv5DQDgwsl/IYRAObuKCHv0ACt+n4xy9pWkMDQ2Nc8VvJpa2jArZYly9pXe4xIWnuzAjIqKQteuXdGvXz84OTnByMgI58+fx4wZM9ChQ9bO6uHhATc3N3Ts2BEzZsxAlSpV8OzZM+zduxedOnVCnTp1ZM3Lzs4OZ86cwYMHD6BUKmFubg4NjdyXWytVqoSgoCBYWVmhUaNGePToEeLi4tCuXTtoaRW8aEOHDsWGDRuwc+dOGBkZ4fnzrAdyTUxMCgzbj9X9+yGIinrzzNrFC+fRyqOZ9H7c91lnz1/39saylasBAHt378LA/m/OUvv06gEAmDjJFz/+5Ff8jS6hZq85DAN9Hcz/8SuYGunjVPB9tB+2ECmp6VIZh3KlYGH65hGlb7u5AwAOLR+hMq0Bvn/hr91v7sj07uCGp+GvcPj0rWJeio+HhqYmHt2/iwM7NiHmZTSMTc1QrWZtzFu/B/aV31yCefb4AWJevrnJsHmbTngV/QIr501DdGQEKn1WAzOXb1Hpkn0ceg/Lfv8VsTEvYV22PL4eNBrdfAa/1+UraaIiwnAm4AAAYPiXzVXGTV35N5zqZj2JEPb4IWJfvVnfiXGxWD33N7wID4ORiSkaenyBPt+Nh1aOnsWPkUIIIeQUTElJgZ+fHw4ePIiQkBCkpaXB1tYWXbt2xYQJE6SAiYuLw8SJE7F9+3ZERkbC2toajRs3xtSpU2Fraws/Pz/4+/sjONu3xcyZMwdz5syRHh+5c+cOvL29cfnyZSQlJSE0NDTPM9Zp06bh+vXrWLduHdatW4e1a9fi0KFD8hY+jwvOq1atgo+Pj6xpZJecXnAZent6OY6BbJuP/892V74PpUyVeHx0qsqw0yEvkZYh6yODCklbUwG3imYqww7fjEQq13ex0dFUwOOzrG/n2nctAm1qWOZbXnZgUsEYmMWLgfl+MTDfLwbm+1fYwOR3yRIREcnAwCQiIpKBgUlERCQDA5OIiEgGBiYREZEMDEwiIiIZGJhEREQyMDCJiIhkYGASERHJwMAkIiKSgYFJREQkAwOTiIhIBgYmERGRDAxMIiIiGRiYREREMjAwiYiIZGBgEhERycDAJCIikoGBSUREJAMDk4iISAYGJhERkQwMTCIiIhkYmERERDIwMImIiGRgYBIREcnAwCQiIpKBgUlERCQDA5OIiEgGBiYREZEMDEwiIiIZGJhEREQyMDCJiIhkYGASERHJwMAkIiKSgYFJREQkAwOTiIhIBgYmERGRDFofugFEb8vcxPBDN+GTpm79amkoPkBL/hvUrVttTa7v4lTY9asQQohiagsREdFHYd+1CLSpYZlvGXbJEhHRf56DRcE9VgxMIiIiGRiYREREMvCmnyIUHpsKXhEuHgoFYGWsozLsxtN4ZGRyhRcXTQ0FqpdVqgyr1elXRMckfKAWfdrMTQxxecePKsPuRyQgg7t4sdFUAA6W8m8eZGAWISEAfn4XD3VdIRmZAulc4e9VdEwCXryK/9DN+M/IEEBG5oduxSeskH2s7JIlIiKSgYFJREQkAwOTiIhIBgYmERGRDAxMIiIiGRiYREREMjAwiYiIZGBgEhERycDAJCIikoGBSUREJAMDk4iISAYGJhERkQwMTCIiIhkYmERERDIwMImIiGRgYBIREcnAwCQiIpKBgUlERCQDA5OIiEgGBiYREZEMDEwiIiIZGJhEREQyMDCJiIhkYGASERHJwMAkIiKSgYFJREQkAwOTiIhIBgYmERGRDAxMIiIiGRiYREREMjAwiYiIZGBgEhERycDAJCIikoGBSUREJAMDk4iISAYGJhERkQwMTCIiIhkYmJ+Q2dN+QeN6NVGprBmq21mhe8fWuHj+bL516jtVQVkz3VyvCWO/Uyl3/mwQurZvhUplzVC1fCl0btMCSUlJxbk4Jd7iP6aic/M6aPCZDZo4lcegXu1x9dL5fOtsXbcc3Vo3gHuNcnCvUQ7enTxw8t9D0viYV9GY7vs9OjV3hVtVK7Rp4IgZfv9DXGxMcS/OR2PSoDa4f+BXRJ+ajb2LhqGibel8y4/t64kT68YiInAmHh6egi2zB6ByBUtpfHkbcyRd/FPtq7OHczEvTck2f9ZvaONeGy4VLVH/s3Lo2+0LXL54Tnb9ZX/OxmdllJjy0/9Uhj96cB/D+vVAgxoVUKeKDUZ92xsvIsOLuvlFrlCBGRkZicGDB6N8+fLQ1dWFtbU1WrVqhZMnTxZpo5o2bYqRI0cWqs65c+dQpkwZAMCzZ8+gr6+P1NTUfOssWrQITk5OMDY2hrGxMdzc3LB///63bfYH51CxMn6dMQdHTl7Ajv3/wra8HXp2bouoF5F51tl39CQu3XoovTbu2AcA+KJjF6nM+bNB+PrLdmjSzAN7D5/E3iMn4TNgMDQ0/tvHWxUcKmHczzOx5cAprNx2AGXKlcfQPp3wMupFnnUsbcriu3F+WL/7GP7aFYC6DRpj1MCvEHLnJgAgMvw5IsPDMHLCr9hy8DT8Zi3EqWOH8fO4Ye9rsUq0Md4eGPJVE3w3ZTMae89GQlIKdi8YAl0drTzruLtWwuItgWjiPRtfDF4ALS1N7Fk4FAZ6OgCAJ+EvYec5QeX186K9iEtIxoGTN97XopVIdg6V8eNvv2Pn0TP4y/8gytpWQP+vOiA6Ku/PlNeuBl/A5r9Womr1GirDExMT0P+rDlAoFFi9dS827DyEtNQ0DPHuhszMzOJalCKhEEIIuYUbN26M1NRUTJ06FQ4ODggPD8eRI0fg6OiI9u3bF1mjmjZtCmdnZ8yZM0d2nXnz5iEwMBBbt27F5s2bMX/+fAQGBuZbZ/fu3dDU1ETlypUhhMCaNWswc+ZMXLp0CY6OjoVu9/OYVGTKXpvFLy42FtUqlMYm//1wb9JcVp2fxo/BkQP7cOLCDSgUCgDAF57uaNy0Bf430a8YW5s/DQVgbaKjMuzq4zikl6AVHh8Xi8Y1bbFo/U7Ub9hUdr2mtSpg5IRf0LF7H7XjD+3dgR9HDcTJG2HQ0so7GIqaloYCNW2NVIbZNh+PF6/i31sbcrp/4FfM++so5qw7CgAwVurh4aEpGOj7F7YevChrGqVMlXh8dCo8+s/ByYshasuc3vA/BN96gsE/byiytsttV3Z3wxOQUYIyJD4uFnWrlsHKzbvh5t4sz3IJCfHo0qoRfpryBxbPnY5qjk6Y8PMMAMDJgCMY+HUnnLn5BEojYwBAXGwM6n9WDss37kKDxnlPt6hpagCVrQwBALfCElDNxjDf8rJPEV69eoXAwEBMnz4dzZo1Q4UKFVCvXj2MHz9eJSxfvXqF/v37o3Tp0jA2Nkbz5s1x+fJlabyfnx+cnZ2xbt062NnZwcTEBD169EBcXBwAwMfHB8eOHcPcuXOhUCigUCjw4MGDAtt36tQpNGzYEABw4sQJ6e/8tGvXDm3atEHlypVRpUoV/Pbbb1AqlQgKCpK7Wkqs1NRUrF+zHMbGJnCs4SS7zt9bNqJ7Lx8pLF9ERuDS+bMoVbo02rdsglpVbNGlrQfOni7aXoWPXVpqKv7euBpKIxNU+aymrDoZGRk4sGsbkpIS4eRSL89y8XGxMFQavdewLInsylrAprQJjp65LQ2LjU/GuWsPUN/JXvZ0jI30AAAvYxLVjq/9mS2cq9lijf/pd2vwJyY1NRVb/loFI2MTVKue/z7+y4TRaNKildrwS01NgUKhgI6OrjRMV1cPGhoauHj2VJG3uyjJDkylUgmlUgl/f3+kpKTkWa5r166IiIjA/v37ceHCBbi4uKBFixaIjo6WyoSEhMDf3x979uzBnj17cOzYMUybNg0AMHfuXLi5uWHAgAEICwtDWFgYbG1t1c7rxIkTMDU1hampKbZt24aJEyfC1NQUixcvxrx582BqaipNtyAZGRnYtGkTEhIS4Obmlme5lJQUxMbGqrzyWx/v26F/9qJyOXM4WBtj2aI/sXHHPphblJJV95+9uxAb8wrdevaWhj18EAoAmD3tV/Ty7of123ajRi1ndO/YGvdD7hbLMnxMjh/5Bw2rl8HnVS2xfsVCLPprB8zMLfKtc/fW9aw6VUrjt4mjMXvJejhUrqa27MvoKCz7cyY6f+VTDK3/uFhbZJ2NRETHqQyPiIqDVSljWdNQKBSYObYLTl0KwY2QMLVlvDu44eb9MARdCX23Bn8i/j20H66VrOBsb4E1y+ZjxaZdMMvnM2Wv/1bcuBqM0eMnqx1fy7Uu9A0MMeu3SUhKTERiYgJm/DwBGRkZiIx4XlyLUSRkB6aWlhZWr16NNWvWwNTUFA0bNsSECRNw5coVqcyJEydw9uxZbN26FXXq1EHlypUxa9YsKdBey8zMxOrVq1GjRg24u7ujd+/eOHLkCADAxMQEOjo6MDAwgLW1NaytraGpqam2TXXq1EFwcDA2bdoEY2NjXLt2DUeOHMk6Url4EcHBwRg0aFC+y3X16lUolUro6upi0KBB2LFjB6pXr55n+alTp8LExETlNXXq1DzLF5e/t2xE5XLm0uvMqRMAgIbuTXHw+FnsPHAMTVu0xKC+PfEiMkLWNDf9tQrNPFrB2qaMNOz1NYWvffqjey9v1HByxuQps1CxUhVs/mtN0S9YCbXPfwsaVi8jvV4fCdd1c8fGfYFYtf0QGjRpgXFDfRCdzzVjIOu60MZ9gVjjfwRdv+6Hn8YMwv27t3KVi4+LxYi+XeFQqSq+HTm+WJarJOvhVQeRJ2ZJLy0t9Z8DhTHnh65wrGiDPuNXqx2vp6uN7l6uWOP/8fcyFdbuvzfDtZKV9Dp/JqsXqX7Dxvj70Cls2HUEjZp6YtS3fRD1Qv1nStjTJ5j60/8wc/5K6OrpqS1jblEac5asQ8Ch/XCtbIV6VcsgNjYG1Ws6Q1HC74soVB9Ply5d0LZtWwQGBiIoKAj79+/HjBkzsHz5cvj4+ODy5cuIj4+HhYXqEXZSUhJCQt5cK7Czs4OR0ZtrIzY2NoiIkPehnp2enh7s7OywZcsWeHl5wd7eHqdOnYK7uzuqVVN/xJ5T1apVERwcjJiYGGzbtg3e3t44duxYnqE5fvx4jB49WmWYrq6u2rLFqaXXF6hdp6703tqmLADAwNAQ9g6VYO9QCa5166Oha3VsXLcaw0f/L69JAQCePHqIwICjWL5us8pwK2trAECVqp+pDK9UtRqePnlcFIvyUWji4YUazq7Se0vrrIMKfQNDlLeriPJ2FeHkUhcdmtaG/+a16Dd0TJ7T0tbRQXm7igCA6jVr4/qVi9iwchF+nDpXKpMQH4dh3l1goFRi9pL10NbWLqYlK7n2HLuKs9ceSO91tbM+rizNjfD8Raw03NLCCFduPy1wen+M64o27jXg0X8unka8Ulumk4czDPR0sH5P/neXf4qat2wDp9p1pPdW/7+PGxgYooJ9RVSwrwhn13po1bAWtm9ci4HDx+aaxvUrlxD1IhJdWr25JJaRkYHzQSexYdUSXH4QDU1NTTRs2gIHT1/Fy6gX0NTSgrGJKdxrOcC2/JfFv6DvoNAXRfT09ODp6QlPT09MmjQJ/fv3h6+vL3x8fBAfHw8bGxsEBATkqmdqair9nfOfX6FQvNXdUUqlEkBWN6mGhgZ27tyJ1NRUCCGgVCrh7u5e4F2vOjo6qFSpEgDA1dUV586dw9y5c7FkyRK15XV1dT9IQOakNDKC0siowHIiMxOpqQV3GW/esBalSluiRcs2KsNty9vB2qYMQu7dURl+/95dNPNoVbhGf8QMlUYwVMpd3/nfnZ1TZmYm0rLViY+LxdA+naGjo4s/lm/K80j9UxefmIL4RNV9NywyBs3qVcWVO1kBaWSoh7o17LBs64l8p/XHuK5o38wJLQfMw8NnUXmW8+nghr3Hrn7QG5s+lELt43lchnJzb4qdR8+oDJs4ajDsK1VB/6GjcvUWvu7aDToRgKgXkWie4/OnpHnnuwiqV68Of39/AICLiwueP38OLS0t2NnZvfU0dXR0kJGRUWC54OBgpKenw9nZGYcPH4a1tTXc3d2xcOFC1KxZE/r6+oWed2ZmZom6JilXYkIC5s6ehpZeX8DKyhrR0VFYvXwxnoc9wxcd3jwi0q1DK3i17YC+A4dIwzIzM7F5/Vp07fF1rhtLFAoFBg0fhdlTf0H1Gk5wrOmErRv/Qsjd21i6ZuN7W76SJikxAcvnz0ITjzYoZWmFVy+jsGXtckQ8D4Nn245SuW97tkOzVu3Qw3sgAODP6X5o0NQTNmXKISEhHv/s3IoLQSewYO3fALLCckjvTkhOTsKvc5YiIS4OCf9/Q5yZRak8L0/8VyzYEIBx/Vvh3qMIPHgWBd/BXyAsMga7At5cGtq3eBh2/XsFizcfBwDM+aEbunu5ouuoZYhPTIaVRVYoxMQnIzklTarnYFsKjVwqouN3i9/vQpVQiYkJWDJ3Jpq1bIPSVtZ4FR2FDauWIvz5M7Rq10kq17dbW3i0bode/QbBUGmEKtVUnzDQNzCAqZm5yvC/N62DQ+WqMLcoheALZzHlp//Be+Aw2Feq8t6W723IDsyoqCh07doV/fr1g5OTE4yMjHD+/HnMmDEDHTp0AAB4eHjAzc0NHTt2xIwZM1ClShU8e/YMe/fuRadOnVCnTp0C5pLFzs4OZ86cwYMHD6BUKmFubq72mb9KlSohKCgIVlZWaNSoER49eoS4uDi0a9dO1h2F48ePh5eXF8qXL4+4uDhs2LABAQEBOHDggNzVUmJoaGoi5O5tDNz0F6KjXsDM3AK1arvi731HUfWzN93LD0NDER2teoQdGHAET588QvevvdVOe8Dg75CSnAK/Cd/j1atoVHd0wsa/98HOvmKxLlNJpqGhiQchd7Bn+0a8ehkFE1NzODq5YMXW/ahY5U339ZOHD/Aq2/qOjorET6MH4UXkcyiNjFG5miMWrP0bn7tnPfZz69plXAvO+vKDDk1qq8xzT+AVlLGt8B6WruSaveYwDPR1MP/Hr2BqpI9TwffRfthCpKSmS2UcypWChembxwO+7eYOADi0fITKtAb4/oW/dr85G/Lu4Ian4a9w+HTu68n/RZoamrh/7zb8t67Hy+gomJqZo2YtV/y14yAqV33zmfLoQSheRud91q5OaMhd/DHVFzGvXqKMbQUM+u57eA8s+c8ay34OMyUlBX5+fjh48CBCQkKQlpYGW1tbdO3aFRMmTJDO5uLi4jBx4kRs374dkZGRsLa2RuPGjTF16lTY2trCz88P/v7+CA4OlqY9Z84czJkzR3p85M6dO/D29sbly5eRlJSE0NDQPM9Yp02bhuvXr2PdunVYt24d1q5di0OHDqktm9M333yDI0eOICwsDCYmJnBycsK4cePg6ekpq35OJe05zE/Jx/Ac5qemJD6H+Sn7GJ7D/NQU9jnMQn1xAeWPgVl8GJjvHwPz/WJgvn/F9sUFRERE/2UMTCIiIhkYmERERDIwMImIiGRgYBIREcnAwCQiIpKBgUlERCQDA5OIiEgGBiYREZEMDEwiIiIZGJhEREQyMDCJiIhkYGASERHJwMAkIiKSgYFJREQkAwOTiIhIBgYmERGRDAxMIiIiGRiYREREMjAwiYiIZGBgEhERycDAJCIikoGBSUREJAMDk4iISAYGJhERkQwMTCIiIhkYmERERDIwMImIiGRgYBIREcnAwCQiIpKBgUlERCQDA5OIiEgGBiYREZEMDEwiIiIZGJhEREQyMDCJiIhk0PrQDfiUKBQ8AikuCkXuYZoaagZSkVG3fs1NDD9AS/4b1K1bTQX4oVKMNAv5EaIQQojiaQoREdHH4VZYAqrZ5H9AyGMXIiIiGRiYRET0n1e70+QCyzAwiYiIZOBNP0UoJjEDvCBcPBQATAw0VYbN/Pc+EtMyP0yD/gMMtDXwfTMHlWEXHsQgPZN7eXHQ0lDA1c5EZVit7jMRHZP4gVr06TM3McDlzd/LLs/ALELi/1/0fiSmZSIhNeNDN+M/JT1TID2De/n7Eh2TiBevEj50M+j/sUuWiIhIBgYmERGRDAxMIiIiGRiYREREMjAwiYiIZGBgEhERycDAJCIikoGBSUREJAMDk4iISAYGJhERkQwMTCIiIhkYmERERDIwMImIiGRgYBIREcnAwCQiIpKBgUlERCQDA5OIiEgGBiYREZEMDEwiIiIZGJhEREQyMDCJiIhkYGASERHJwMAkIiKSgYFJREQkAwOTiIhIBgYmERGRDAxMIiIiGRiYREREMjAwiYiIZGBgEhERycDAJCIikoGBSUREJAMDk4iISAYGJhERkQwMTCIiIhkYmERERDIwMImIiGTQ+tANoKIz7bfJ2LFtC54+eQxtHR04O7tgot8vqFO3fp514uLiMOVnX+zd7Y8XkRGoWcsZU2f+ARfXumrLj/5uCFavWIrfps/G4GEjimtRSryM9DSc+Gsu7p8/hpjnT6BjqESFWg3QxHs0lBZWedYL2roEd08dQtTT+9DW0UOZarXRxGcMzMs5AABiwp9gaX8PtXXbj5uDqo1aF8vyfCxWzJuGI3t3IOL5U2hpa6OqozMGjp4Ix1p18qwTfO4UNiz/E7evX0ZUxHNMWbAOjT3bSuPT09KwdM5vCDp2CM8eP4ShkTHquDXB4LE/oZSVzftYrBJt0oCW6NuhHkyV+jh99QG+m7EDIY9f5Fl+QOfPMaCzGyrYmAEAbt4Px5SVh3Hw9G2pzIGF36KxS0WVesv+DsJ3M/4unoUoIgzMT0ilSlUwffZc2Nk7ICkpCYvmz0WX9l64cOU2SpUurbbOiKEDcfPGdSxevhrWNmWwZdN6dPqiFU5fuIoyZcqqlN2zyx/nz56BjU2Z97E4JVp6SjLCQ27ArfsQWNpXRXJ8LI4um4K/fx2CPn9sz7Pe42vnULttT1hXronMzAwErv0DW3/qj74L90BHzwBGpWwweG2gSp0r/2zB2R0rYO/qXtyLVeLZ2lfCqJ+mo4ytHVJSkrBl1SKM7tsFmw5fgJl5KbV1khITUKlaDbTt0gsTh/XJNT45OQl3rl+G95CxqFytBmJjX2Hur+MxbnAvrPj7aHEvUok2pndTDOnWEAN+3owHYdH4aWAr7J7zDWp/NRspqelq6zyNiMGkBftx78kLKAB83dYVW2d44/M+c3EzNFwqt8L/DH5ZekB6n5icVtyL884K1SUbGRmJwYMHo3z58tDV1YW1tTVatWqFkydPFmmjmjZtipEjRxaqzrlz51CmTNYH+bNnz6Cvr4/U1FTZ9adNmwaFQlHo+ZYkX3b/Ck2be8DO3gGfVXfEr9NmIS42FtevXVFbPikpCbv9/8bkX6eiQaPGcKhYCT9M9IWDQyWsWrZYpeyzZ08xbswILFm5Flra2u9jcUo0XUMjdPtlJaq5e8G8nAPKVHNGi28nIfzedcRGPMuzXtfJy1HDozNKVagMS/tq8Bo5FbGRzxB+7zoAQENTE0qz0iqvu0GHUa2RF3T0Dd/X4pVYLdt9iboNm6JseTs4VP4Mwyf8ioT4OITcup5nHbcmnhg4aiKatPxC7XilkTHmrN6BFm06obxDZdRwrovRP83A7WvBeP7sSXEtykdhaPdGmL7qCPYE3sC1e8/Rf/Jm2JQyRvvGjnnW2XfiJg6cvoWQxy9w7/EL+C0+gPjEVNSrUV6lXFJyKsKj46VXXGJKcS/OOyvUGWaXLl2QmpqKNWvWwMHBAeHh4Thy5AiioqKKq32ynT59Gg0bNgQABAYGok6dOtDR0ZFV99y5c1iyZAmcnJyKs4nvVWpqKtasXAZjExPUqFlLbZn09HRkZGRAV1dPZbievh6CTr85CMrMzMTgb7wxfOQYfFY973+U/7qUxDhAoYCu0lh+nYQ4AICekYna8c/vXUPE/ZvwGDSpSNr4KUlLTcXOzWugNDJGpWo1inTa8XGxUCgUMDKWvy0/NXZlzGFTyhhHz92VhsUmJOPc9ceoX7MCth6+XOA0NDQU6NLcCYb6Ojhz9aHKuO6taqNHaxeER8Vh34kbmLryCJJSSvZZpuwzzFevXiEwMBDTp09Hs2bNUKFCBdSrVw/jx49H+/btVcr1798fpUuXhrGxMZo3b47Ll9+sWD8/Pzg7O2PdunWws7ODiYkJevTogbi4rA8OHx8fHDt2DHPnzoVCoYBCocCDBw8KbN+pU6ekwDxx4oT0d0Hi4+PRq1cvLFu2DGZmZnJXR4l1YP8e2FqawMbcEIvnz8Xfu/+BRSn1XVVGRkaoW/9zzJr+G8LCniEjIwNbNq7HuTNBCH/+XCo3d/YMaGpp4dshw9/XYnx00lNTcHz1LHzWuC10DZSy6ojMTBxdNgVlP3NB6QpV1Ja5enA7LGwrouxnLkXZ3I/ayX8PwNPZFs1r2mDLqsX4Y9XfMDW3KLLpp6QkY9GsyfD4ogsMC3Hw86mxtjACAEREx6sMj4iOg9X/j8uLY0VrRB79BTHHp2DeuM7oPm4tbj2IkMZvPhCMfn6b0HroEsxa+y96erlg1eQeRb8QRUx2YCqVSiiVSvj7+yMlJe9T565duyIiIgL79+/HhQsX4OLighYtWiA6OloqExISAn9/f+zZswd79uzBsWPHMG3aNADA3Llz4ebmhgEDBiAsLAxhYWGwtbVVO68TJ07A1NQUpqam2LZtGyZOnAhTU1MsXrwY8+bNg6mpqTTdvAwdOhRt27aFh4f6Gy1ySklJQWxsrMorv/VRXLZu2gBbSxPpdfpk1nWvRo2b4djpC/jnaCCae7ZCv95fITIiIs/pLF6+BkIIOFYqD2szAyxd9Ce6dO0BhUbWrhF86QKWLPwTC5auhEKheC/LVhLdCNiNOV1dpNeT6+elcRnpadg1fSSEADyH+Mme5qHFP+PFo7to97/f1Y5PS0nGzeN7UNOzy7s2/6N0cNdWeDrbSq/L504DAFzqN8KqncewaPM/qN+4OX4a2Q8voyKLZJ7paWn4aUQ/QAiMnTyrSKb5sejRqjYij/4ivbS0NN96WnceRqJ+nzlo/M18LPv7NJb91A3V7Cyl8St3nsHhM3dwPeQ5Nh24hG8mb0aHpjVhX9a8KBal2MjuktXS0sLq1asxYMAALF68GC4uLmjSpAl69OghdWWeOHECZ8+eRUREBHR1dQEAs2bNgr+/P7Zt24aBAwcCyOriW716NYyMso5SevfujSNHjuC3336DiYkJdHR0YGBgAGtr63zbVKdOHQQHB+PWrVvo2bMnLly4gOjoaDRo0AAXL16Enp4eTE1N86y/adMmXLx4EefOnZO7GjB16lRMnjxZZZivry/8/PxkT6MotG7bDq5160nvbf7/Bh1DQ0M4VKwEh4qVULfe56jjVA1/rVmJUd//oHY69g4VsefAv0hISEBcbCysbWzQr89XsLOzBwCcPnkCkZERcKpqL9XJyMjApPHfY/GCebh8M6QYl7LkqFSvGWyqvOmyf30nbFZYjkJsxDN0/2217LPLw4t/xv1zAegx9S8YlVK/n985eQBpKclwbN7xndv/MWrUvDWq13KV3pf+/ztW9Q0MUa6CA8pVcEAN57ro4VkHe7b+hd6DRr3T/NLT0jBpRD88f/oY89bu/M+dXe4JvIGz1x9J73W1s+LB0lyJ51Fx0nBLcyNcuZv3dXoASEvPwP0nWZfqLt1+CtfqthjavRGGT1d/F+y5/59vxXKlEPo0Wm2ZkqDQ1zDbtm2LwMBABAUFYf/+/ZgxYwaWL18OHx8fXL58GfHx8bCwUO0eSUpKQkjImw9WOzs7KSwBwMbGBhH5nAXlRU9PD3Z2dtiyZQu8vLxgb2+PU6dOwd3dHdWqVcu37uPHjzFixAgcOnQIenp6+ZbNbvz48Rg9erTKsNcHB++TkZGRyjrMS2ZmJlJSCz4DNjQ0hKGhIV69fImjhw/C79esM/PuX32NJs1aqJTt2qENun3VCz17+7xV2z9GOgZK6OQIw9dh+erZQ3Sfsgb6xgV36QshcGTJL7h7+jB6TF0LU+tyeZa9emgbKtVrBgOTkn3UXVwMlEYwUMrbx1Nl7OP5eR2WTx6GYN66XTAx+++t8/jEFMTnuPEm7EUsmtWtjCt3wwAARga6qOtoi2V/ny7UtDUUCujq5B03tapk3bD5PCq2kK1+vwr9WImenh48PT3h6emJSZMmoX///vD19YWPjw/i4+NhY2ODgICAXPWyn+lp57jLUqFQIDMzs9CNVyqzPsBSUlKgoaGBnTt3IjU1FUIIKJVKuLu7Y//+/WrrXrhwAREREXBxeXNtKCMjA8ePH8f8+fORkpICTc3cXRK6urofJCALkpCQgN9nTEHrtu1gbW2DqKgXWL5kEcKePUWHTl9K5Tq28UTb9h0xYNBQAMCRQwcghEDlKlVxP+QefCf+gMpVqqLX/4ehuYUFzHMcAGlpa8PSyhqVq1R9b8tX0mSkp2HXtBEID7mBzj8tRmZmBuJfZnUL6itNoKmddcPZ5ok+qOzmAZcvvgYAHF70M24e34NOExdAW99QqqNrYATtbDdfvXz2EI+vn8eXvkvf85KVXEmJCVi76Hc0bNEapUpb49XLKPy9fjlehIehmVcHqdyIPh3R2LMtuvQeAABITIjH04eh0viwJw9x98ZVGJmawbpMOaSnpeHH73xw5/plTF+yCZkZGYiKzHr8wdjEDNoybx78FC3YfALjfJrj3uMXePAsGr4DWyLsRSx2HX9zV/K+Pwdg17HrWLztFADg58GtceD0bTwOfwUjA110b+mMxi4OaDdyBQDAvqw5uresjQOnbiEqNhE1K9lgxoh2CLx4H9fuPVfbjpLinZ/DrF69Ovz9/QEALi4ueP78ObS0tGBnZ/fW09TR0UFGRkaB5YKDg5Geng5nZ2ccPnwY1tbWcHd3x8KFC1GzZk3o6+vnWbdFixa4evWqyrC+ffuiWrVqGDdunNqwLMk0NTVx985tbFq/DlFRL2BuboHarnWw91CAyp2toaH3ERX15qHj2NhY/OI7Ec+ePoGZmTnadeyMH31/yXVQQ6rio8Jx70zWM3prvuuoMq77lDUoXzPryyJePX+EpNiX0rjg/RsBAJsmqD4P6DViCmp4dJbeXz28HUYW1rCrLe/mtf8CDU1NPLx/F/t3bELMyygYm5njs5q1sWDDXjhU/kwq9/RxKF69fHPn/q1rwfiu95sbE/+c+iMAwKvTV5g4fQEiw8Nw4kjWgXXfDo1V5jlv3S641G9UnItVos1eFwADPR3M/6ELTJV6OHXlAdqPXKHyDKZDOQtYmL555Km0mRIrfLvD2sIYMfHJuBYShnYjV+Do2ay7bdPSMtC8bmUM69EIhno6eBIRA/+Aq5i28sj7XrxCUwghhJyCUVFR6Nq1K/r16wcnJycYGRnh/PnzGD58ONq2bYsVK1ZACIHGjRsjLi4OM2bMQJUqVfDs2TPs3bsXnTp1Qp06deDn5wd/f38EBwdL054zZw7mzJkj3Q07cOBABAcHY8uWLVAqlTA3N4eGhvr7k4KCgtC9e3c8fPgQjx49QvXq1fHq1StoaRX+WKBp06ZwdnbGnDlzCl0XAF4lZkDWyqRCUwAwNVA9iJl88B4SUgs+sKK3Y6ijCd+WlVSGnbn/CukZ3MuLg5amAvUdTFWG2baejBevEj5Mg/4DSpka4vE/vgAA/c//h6SgGfmWl50qSqUS9evXxx9//IGQkBCkpaXB1tYWAwYMwIQJEwBkda3u27cPEydORN++fREZGQlra2s0btwYVlZ5f11YTmPHjoW3tzeqV6+OpKQkhIaG5nnGGhAQgMaNs44Kjx07Bjc3t7cKSyIiovzIPsOkgvEMs/jwDPP94xnm+8UzzPevsGeY/LUSIiIiGRiYREREMjAwiYiIZGBgEhERycDAJCIikoGBSUREJAMDk4iISAYGJhERkQwMTCIiIhkYmERERDIwMImIiGRgYBIREcnAwCQiIpKBgUlERCQDA5OIiEgGBiYREZEMDEwiIiIZGJhEREQyMDCJiIhkYGASERHJwMAkIiKSgYFJREQkAwOTiIhIBgYmERGRDAxMIiIiGRiYREREMjAwiYiIZGBgEhERycDAJCIikoGBSUREJAMDk4iISAYGJhERkQwMTCIiIhkYmERERDIwMImIiGRgYBIREcmg9aEb8ClRfOgGfMLUrVsDbR7vFSd161dLg3t5cVG3bs1NDD5AS/47Crt+FUIIUUxtISIi+ijof/4/JAXNyLcMD9GJiIhk4BkmERGRDDzDJCIikoGBSUREJAMDk4iISAYGJhERkQwMTCIiIhkYmERERDIwMImIiGRgYBIREcnAwCQiIpLh/wDVuL4wFnCHAgAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_heatmap(\n",
    "    bartscore_scores,\n",
    "    title='Semantic Similarity of Sentences\\nas Measured by BARTScore',\n",
    "    x_ticks = x_ticks,\n",
    "    y_ticks = x_ticks,\n",
    "    x_label = '',\n",
    "    y_label = '',\n",
    "    valfmt = '{x:.2f}',\n",
    "    cmap=\"Blues\",\n",
    ")"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Jaro Similarity"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The Jaro similarity is a string similarity metric that measures the similarity between two strings. \n",
    "\n",
    "It is defined as follows:\n",
    "\n",
    "$$\n",
    "\\begin{align}\n",
    "\\texttt{Jaro-Similarity}(s_1, s_2) &= \n",
    "    \\begin{cases}\n",
    "        0 & \\text{ if } \\texttt{matches} = 0 \\\\\n",
    "        \\frac{1}{3} \\left( \\frac{m}{|s_1|} + \\frac{\\texttt{matches}}{|s_2|} + \\frac{\\texttt{matches} - \\texttt{transpositions}}{\\texttt{matches}} \\right) & \\text{otherwise}\n",
    "    \\end{cases}\n",
    "\\end{align}\n",
    "$$\n",
    "\n",
    "where $s_1$ and $s_2$ are two strings, $\\texttt{matches}$ denotes the number of \"Jaro-matching\" characters between $s_1$ and $s_2$ and $\\texttt{transpositions}$ the number of \"Jaro-transpositions\" between $s_1$ and $s_2$.\n",
    "\n",
    "* A character from $s_1$ is a \"Jaro-matching\" character with a character from $s_2$ if the characters are the same and the distance between the two characters is less than or equal to the floor of $\\frac{1}{2} \\max(|s_1|, |s_2|)$.\n",
    "* A \"Jaro-transposition\" occurs when two characters from $s_1$ and $s_2$ are \"Jaro-matching\" but are not in the right order in their respective strings (i.e., the characters are transposed). \n",
    "\n",
    "For more details, please refer to the Wikipedia page of [Jaro-Winkler distance](https://en.wikipedia.org/wiki/Jaro%E2%80%93Winkler_distance).\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Jaro similarity score between FARMVILLE and FAREMVIEL is 0.88\n"
     ]
    }
   ],
   "source": [
    "# Let's import the Jaro similarity class from the similarity module\n",
    "from string2string.similarity import JaroSimilarity\n",
    "\n",
    "# Create an instance of the JaroSimilarity class\n",
    "jaro_similarity = JaroSimilarity()\n",
    "\n",
    "# Compute the Jaro similarity scores of the following pairs of words\n",
    "score = jaro_similarity.compute('FARMVILLE', 'FAREMVIEL')\n",
    "print(f'The Jaro similarity score between FARMVILLE and FAREMVIEL is {score:.2f}')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Jaro similarity scores between the two versions of Hamlet at the word level is 0.66\n",
      "The Jaro similarity scores between the two versions of Hamlet at the character level is 0.82\n"
     ]
    }
   ],
   "source": [
    "# https://en.wikipedia.org/wiki/To_be,_or_not_to_be#Second_Quarto_(1604)\n",
    "second_quarto = r\"\"\"To be, or not to be, that is the question,\n",
    "Whether 'tis nobler in the mind to suffer\n",
    "The slings and arrows of outragious fortune,\n",
    "Or to take Arms against a sea of troubles,\n",
    "And by opposing, end them, to die to sleep\n",
    "No more, and by a sleep, to say we end\n",
    "The heart-ache, and the thousand natural shocks\n",
    "That flesh is heir too; tis a consumation\n",
    "Devoutly to be wish'd to die to sleep,\n",
    "To sleep, perhance to dream, ay, there's the rub,\n",
    "For in that sleep of death what dreams may come\n",
    "When we haue shuffled off this mortal coil\n",
    "Muſt giue vs pauſe, there's the reſpect\n",
    "That makes calamitie of ſo long life:\n",
    "For who would beare the whips and ſcorns of time,\n",
    "Th'oppreſſors wrong, the proude mans contumly,\n",
    "The pangs of deſpiz'd loue, the lawes delay,\n",
    "The inſolence of office, and the ſpurnes\n",
    "That patient merrit of the'vnworthy takes,\n",
    "When he himſelfe might his quietas make\n",
    "With a bare bodkin; who would fardels beare,\n",
    "To grunt and ſweat vnder a wearie life,\n",
    "But that the dread of ſomething after death,\n",
    "The vndiſcouer'd country, from whose borne\n",
    "No trauiler returnes, puzzels the will,\n",
    "And makes vs rather beare thoſe ills we haue,\n",
    "Then flie to others we know not of.\n",
    "Thus conſcience dooes make cowards,\n",
    "And thus the natiue hiew of reſolution\n",
    "Is ſickled ore with the pale caſt of thought,\n",
    "And enterpriſes of great pitch and moment,\n",
    "With this regard theyr currents turne awry,\n",
    "And loose the name of action. Soft you now,\n",
    "The faire Ophelia, Nimph in thy orizons\n",
    "Be all my ſinnes remembred.\n",
    "\"\"\"\n",
    "\n",
    "# https://en.wikipedia.org/wiki/To_be,_or_not_to_be#First_Folio_(1623)\n",
    "first_folio = r\"\"\"To be, or not to be, that is the Question:\n",
    "Whether ’tis Nobler in the mind to suffer\n",
    "The Slings and Arrows of outragious Fortune,\n",
    "Or to take Armes against a Sea of troubles,\n",
    "And by opposing end them: to dye, to sleep\n",
    "No more; and by a sleep, to say we end\n",
    "The Heart-ake, and the thouſand Naturall ſhockes\n",
    "That Flesh is heyre too? 'Tis a consummation\n",
    "Deuoutly to be wiſh'd. To dye to sleepe,\n",
    "To sleep, perchance to Dream; I, there's the rub,\n",
    "For in that sleep of death, what dreams may come,\n",
    "When we haue ſhufflel’d off this mortall coile,\n",
    "Muſt giue us pause. There's the respect\n",
    "That makes Calamity of long life:\n",
    "For who would beare the Whips and Scornes of time,\n",
    "The Oppreſſors wrong, the poore mans Contumely,\n",
    "The pangs of diſpriz’d Loue, the Lawes delay,\n",
    "The inſolence of Office, and the Spurnes\n",
    "That patient merit of the unworthy takes,\n",
    "When he himſelfe might his Quietus make\n",
    "With a bare Bodkin? Who would theſe Fardles beare\n",
    "To grunt and ſweat vnder a weary life,\n",
    "But that the dread of ſomething after death,\n",
    "The vndiſcouered Countrey, from whoſe Borne\n",
    "No Traueller returnes, Puzels the will,\n",
    "And makes vs rather beare those illes we haue,\n",
    "Then flye to others that we know not of.\n",
    "Thus Conſcience does make Cowards of vs all,\n",
    "And thus the Natiue hew of Resolution\n",
    "Is ſicklied o’re, with the pale caſt of Thought,\n",
    "And enterprizes of great pith and moment,\n",
    "With this regard their Currants turne away,\n",
    "And looſe the name of Action. Soft you now,\n",
    "The faire Ophelia? Nimph, in thy Orizons\n",
    "Be all my ſinnes remembred.\n",
    "\"\"\"\n",
    "\n",
    "# Let's first tokenize the texts\n",
    "from string2string.misc import Tokenizer\n",
    "tokenizer = Tokenizer(word_delimiter=' ')\n",
    "\n",
    "# Tokenize the texts\n",
    "second_quarto_tokens = tokenizer.tokenize(second_quarto)\n",
    "first_folio_tokens = tokenizer.tokenize(first_folio)\n",
    "\n",
    "# Compute the Jaro similarity scores of the two versions of Hamlet at the word level\n",
    "jaro_similarity_hamlet_word_level = jaro_similarity.compute(second_quarto_tokens, first_folio_tokens)\n",
    "\n",
    "# Compute the Jaro similarity scores of the two versions of Hamlet at the character level\n",
    "jaro_similarity_hamlet_character_level = jaro_similarity.compute(second_quarto, first_folio)\n",
    "\n",
    "# Print the Jaro similarity scores (rounded to 2 decimal places)\n",
    "print(f'The Jaro similarity scores between the two versions of Hamlet at the word level is {jaro_similarity_hamlet_word_level:.2f}')\n",
    "print(f'The Jaro similarity scores between the two versions of Hamlet at the character level is {jaro_similarity_hamlet_character_level:.2f}')\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    ">  With this example, we have reached the end of the alignment tutorial. We hope that you have found it useful. If you have any questions or suggestions, please feel free to contact us."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### $\\square$ Q.E.D."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "string2string",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.12"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}