{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"# This Python 3 environment comes with many helpful analytics libraries installed\n# It is defined by the kaggle/python Docker image: https://github.com/kaggle/docker-python\n# For example, here's several helpful packages to load\n\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n\n# Input data files are available in the read-only \"../input/\" directory\n# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory\n\nimport os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        print(os.path.join(dirname, filename))\n\n# You can write up to 20GB to the current directory (/kaggle/working/) that gets preserved as output when you create a version using \"Save & Run All\" \n# You can also write temporary files to /kaggle/temp/, but they won't be saved outside of the current session","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2021-06-11T06:18:25.223613Z","iopub.execute_input":"2021-06-11T06:18:25.224035Z","iopub.status.idle":"2021-06-11T06:18:25.240066Z","shell.execute_reply.started":"2021-06-11T06:18:25.223946Z","shell.execute_reply":"2021-06-11T06:18:25.238793Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# 1. Vấn đề bài toán: Quora Insincere Classification\n\n<div> \n      <li> Quora là nền tảng nổi tiếng nơi mọi người đặt cho nhau những câu hỏi để học hỏi thêm nhiều thứ. Tuy nhiên có những câu hỏi được đặt ra mang tính nhạy cảm cho người đọc. Ta phải phân loại ra những câu hỏi mang tính tiêu cực, không đúng đắn. </li>\n                <b> Input </b> Câu hỏi của Quora (dữ liệu text) <br>\n                <b>  Output: </b> Label Sincere (0), Insincere (1) <br>\n</div>\n","metadata":{}},{"cell_type":"code","source":"train_data = pd.read_csv(\"/kaggle/input/quora-insincere-questions-classification/train.csv\")\ntest_data = pd.read_csv(\"/kaggle/input/quora-insincere-questions-classification/test.csv\")","metadata":{"execution":{"iopub.status.busy":"2021-06-11T06:18:26.626026Z","iopub.execute_input":"2021-06-11T06:18:26.626444Z","iopub.status.idle":"2021-06-11T06:18:32.573876Z","shell.execute_reply.started":"2021-06-11T06:18:26.626406Z","shell.execute_reply":"2021-06-11T06:18:32.573105Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## 1.1. Phân tích dữ liệu\n- Dữ liệu có thuộc tính `qid`, `question_text`, `target`\n- Khi phân loại, ta dùng thuộc tính `question_text` là đầu vào X, `target` là label y","metadata":{}},{"cell_type":"code","source":"train_data","metadata":{"execution":{"iopub.status.busy":"2021-06-11T06:18:32.575579Z","iopub.execute_input":"2021-06-11T06:18:32.576007Z","iopub.status.idle":"2021-06-11T06:18:32.605428Z","shell.execute_reply.started":"2021-06-11T06:18:32.575962Z","shell.execute_reply":"2021-06-11T06:18:32.604724Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_data.info()","metadata":{"execution":{"iopub.status.busy":"2021-06-11T06:18:32.607387Z","iopub.execute_input":"2021-06-11T06:18:32.607695Z","iopub.status.idle":"2021-06-11T06:18:32.866932Z","shell.execute_reply.started":"2021-06-11T06:18:32.607665Z","shell.execute_reply":"2021-06-11T06:18:32.865954Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<h3>Bỏ đi các dữ liệu NA, không tồn tại trong bảng. </h3>","metadata":{}},{"cell_type":"code","source":"train_data.dropna(inplace = True)\ntrain_data","metadata":{"execution":{"iopub.status.busy":"2021-06-11T06:18:32.868379Z","iopub.execute_input":"2021-06-11T06:18:32.868692Z","iopub.status.idle":"2021-06-11T06:18:33.251679Z","shell.execute_reply.started":"2021-06-11T06:18:32.868655Z","shell.execute_reply":"2021-06-11T06:18:33.251033Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<h3> => Từ dữ liệu được cho, ta thấy: </h3>\n\n - 1306122 câu hỏi `question_text` chứa giá trị text data.\n - 1306122 label `target` chứa giá trị 0, 1.\n - Dữ liệu không chứa giá trị ***NA***\n \n<h4> => Giờ ta sẽ xem sự khác biệt giữa dữ liệu label 1, 0 </h4>\n","metadata":{}},{"cell_type":"code","source":"toxic_data = train_data[train_data.target == 1]\nnon_toxic_data = train_data[train_data.target == 0]","metadata":{"execution":{"iopub.status.busy":"2021-06-11T06:18:33.252996Z","iopub.execute_input":"2021-06-11T06:18:33.253591Z","iopub.status.idle":"2021-06-11T06:18:33.340854Z","shell.execute_reply.started":"2021-06-11T06:18:33.253545Z","shell.execute_reply":"2021-06-11T06:18:33.339421Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Những câu hỏi Sincere của Quora ( label `target` có giá trị 0 )","metadata":{}},{"cell_type":"code","source":"for d in non_toxic_data.question_text.sample(10):\n  print(d)","metadata":{"execution":{"iopub.status.busy":"2021-06-11T06:18:33.342407Z","iopub.execute_input":"2021-06-11T06:18:33.342713Z","iopub.status.idle":"2021-06-11T06:18:33.395532Z","shell.execute_reply.started":"2021-06-11T06:18:33.342684Z","shell.execute_reply":"2021-06-11T06:18:33.394450Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Những câu hỏi Insincere của Quora ( label `target` có giá trị 1 )","metadata":{}},{"cell_type":"code","source":"for d in toxic_data.question_text.sample(10):\n  print(d)","metadata":{"execution":{"iopub.status.busy":"2021-06-11T06:18:33.396714Z","iopub.execute_input":"2021-06-11T06:18:33.397061Z","iopub.status.idle":"2021-06-11T06:18:33.405568Z","shell.execute_reply.started":"2021-06-11T06:18:33.397033Z","shell.execute_reply":"2021-06-11T06:18:33.404235Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## 1.2. Nhận xét về dữ liệu ","metadata":{}},{"cell_type":"code","source":"import matplotlib.pyplot as plt\nimport pandas as pd\nimport seaborn as sns\n\nprint(toxic_data.shape, non_toxic_data.shape)\nprint(\"--------------------------\")\n\nsns.countplot(x='target', data=train_data)","metadata":{"execution":{"iopub.status.busy":"2021-06-11T06:18:33.408161Z","iopub.execute_input":"2021-06-11T06:18:33.408568Z","iopub.status.idle":"2021-06-11T06:18:34.465099Z","shell.execute_reply.started":"2021-06-11T06:18:33.408525Z","shell.execute_reply":"2021-06-11T06:18:34.464152Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<h3> Hình 1 - Bảng đếm số lượng dữ liệu label 0 và label 1 </h3>","metadata":{}},{"cell_type":"code","source":" print( \"Dữ liệu Insincere chiếm\", toxic_data.shape[0] / train_data.shape[0])\n print( \"Dữ liệu Sincere chiếm\", non_toxic_data.shape[0] / train_data.shape[0])","metadata":{"execution":{"iopub.status.busy":"2021-06-11T06:18:34.466331Z","iopub.execute_input":"2021-06-11T06:18:34.466617Z","iopub.status.idle":"2021-06-11T06:18:34.472612Z","shell.execute_reply.started":"2021-06-11T06:18:34.466587Z","shell.execute_reply":"2021-06-11T06:18:34.471657Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Vấn đề \nTrong dữ liệu train, có 80810 dữ liệu `insincere` <b> (label 1) </b>, 1225312 dữ liệu `sincere` <b>(label 0)</b>. \n- Ở đây ta thấy dữ liệu `insincere` chỉ chiếm tỉ lệ 0.062\n- Còn so với dữ liệu `sincere` chiếm tỉ lệ 0.94 => Dữ liệu mất cân bằng\n\n![image.png](attachment:2c6a09f6-82c2-47dd-8950-57b9bf735223.png)\n\n<h3> => Độ đo sử dụng là F1-Score </h3>\n","metadata":{},"attachments":{"2c6a09f6-82c2-47dd-8950-57b9bf735223.png":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAzMAAAHxCAYAAABZFOIgAAAgAElEQVR4AeydCbxO1frHUzIfQ6U0z5QUSYMIEU2XlIZ/g0ahmUplPOcgolKqW7e6ruaJJipuRWZpoHm8lZK5IjPnnPf3/3zX8Zy2t/c9M87hWT7Lfs/ea/yttfd+fut51rN3kAdHwBFwBBwBR8ARcAQcAUfAEXAESiECO5TCNnuTHQFHwBFwBBwBR8ARcAQcAUfAEZCTGZ8EjoAj4Ag4Ao6AI+AIOAKOgCNQKhFwMlMqh80b7Qg4Ao6AI+AIOAKOgCPgCDgCTmZ8DjgCjoAj4Ag4Ao6AI+AIOAKOQKlEwMlMqRw2b7Qj4Ag4Ao6AI+AIOAKOgCPgCDiZ8TngCDgCjoAj4Ag4Ao6AI+AIOAKlEgEnM6Vy2LzRjoAj4Ag4Ao6AI+AIOAKOgCPgZMbngCPgCDgCjoAj4Ag4Ao6AI+AIlEoEnMyUymHzRjsCjoAj4Ag4Ao6AI+AIOAKOgJMZnwOOgCPgCDgCjoAj4Ag4Ao6AI1AqEXAyUyqHzRvtCDgCjoAj4Ag4Ao6AI+AIOAJOZrbQHIjFYsrKyhJHC3bO/i6tR+tHtG+59QUc4rHILb1fcwQcAUfAEXAEHAFHwBEoOgLIavmV14pe25YpwcnMlsE5TJz4ybOtTKiC9sPSx+OxhYbCq9nGEIifRza/trFuenccAUfAEXAEHIF8IcB70BaO7bgtvxu3azJjg52vmVHIREwiwmeffaYXXnhBP/30UyA2K1as0KRJk/T2229r+fLlIQ3tKW1h3bp1mjt3rmbPnq1ly5bl9MOwTdSnX375RR999JHmzZtX2rrr7XUESh0C3IOJ7sNS1xFvsCPgCDgCjkBSBKJylz3348+ROTMzUxkZGdvUe2G7JzOb+yXPpCH8+9//1sknn6z//ve/gS0j0Pfs2VM33HCDfvzxxzCpjPgknakl6ILdKEuWLNHIkSPVtWtXffDBB6GF9Dl6A8U3+4knntBFF12kZ555Jv6S/+0IFAgB5tr69es3eSjzkN6wYcMm5wpU6DaW2O7Vbaxb3h1HwBFwBBwBKbzr7DmPHMnvtWvXhgXjjz/+OCyas5j++OOP66GHHtLdd9+te++9N8ie2wqpKZVkxgbNjrnNZktjR0uLELR48WItWLAgRxgiTWGDlR9fBhOFcM899+iggw7SK6+8EljxDz/8oCuvvFLnnXeevvnmmzD5jPgUtg2J8lm7El2zc5Ymvu12nWM0TfT30qVL9eyzz+rmm28WNw3X6Acam++++04rV67M2R9j5T///PO6+uqr9dJLL0WryPfvaP1WZr4ze8LNjoCNz+aqyMpnjn366af6z3/+o4ULF+ZU98477+jJJ5/Ub7/9lnNuW/9hmFg/7b7g/pwyZUrQnHIvEuyapfWjI+AIOAKOQOlDwJ77dsTKByugMWPH6MEHH1S37t10wQUXqGWrlmrQoEGQQffaay/ttttu2n333fXuu+8G0mP5Sx8Cf7W41JKZ/NoAMkjGVG3AOK5atUoI1QhCkBq7ZtDE/23nEx0tbaI2GZm5//77dfjhh+u1114Lwj7mZp06dQoTDaGfMqydieooyDlrT/wxURmWJlHbo+mtfdZG/ibQPwQmtEtgynnIDCTxxhtv1Oeff55DFq2833//PZim/fHHH3Yq6dHaZwns77zaa+n9uOURYIwYn80VrHzMGtF4HnXUUfrwww9zqrv22mvVqFEjffnllznntvUfhon1k3uQc5h/XnXVVRowYIDQBsens/R+dAQcAUfAESgdCPAcJ/CeRQbj2f7ee+9p+PDhuuyyy8L7r+YeNVVxl4qqcUgN7X3s3jqw5YE6rN1hOvzsw7Vzys7aqexOemvcW05mttaQM4hr1qwJAjQkZPXq1aEpNrjWLv4m/vnnn0GwZpUW0xMGnxc9ea+55ppg7vT111+Ha0Y8yMdv+9vKtIkTFdRIS3kI8osWLQrtMhMXK4f89913nw477LAcMoPwj5CBZubbb78N7cJcJlq21cuRtpiAEj0f/5s01k+u0S5iory0j4g6EjzAiHpoA+cJloZVXfpHOsqzdnI9WifnbcW8SpUqGjt2bBiDaJssvbUpEdbWL8ozPK09tJe2EqnL2mJ5/Fg4BGys7ZisFLsePZKWvxmL+fPn66uvvkr6kCRdfLCyouftXPQYrQcy/Mgjj+iQQw7R+++/n5MVjScLBxBpS5+ozryu5RQYuQ+ieaxd0XT2267Z0c5Hj3aNIyH+b0sbPW9po9e4j7gvP/nkk5x7xe4tnm19+/YNOKG9In9u90t8+fHtSnTd2uJHR8ARcAQcgc2LAM9gew6z7xjtSr9+/dS0aVPV2K2GKteqrD0b7qlDzzxUja5tpNOHn66LXr1Inad3Vvfvu6vbT92UcnCKKlSqoPHjxyd9T2/eXhR/6aVKM4PQi0nW66+/Huz+0Ha8+uqrYVM912yQeVkjfLPJHM3LXXfdFdKPGzcuCFpoQsaMGaO2bduqVatWIQ17WVjFpAyE5VmzZmn69OmbkACEgalTpwYWbAIBaj3yvfjii6I92COyqZ+9JKSxduVFZtBWTJs2LZjN2DDbhKWOyZMnB20Ggn2iQFrI0P/+978wuSFLrFY/99xzYW8Kv3E6YBhxhBR+//33AU8Y/cMPP6wJEyYE8mfpaD8CEfaWw4YNC33k9xdffBHyQyYoA1wgjpRJP/71r3+pcuXKGjhwYCBwlIsGh4CJHdjiOACsIXPkSWQGA7bsxaFs8Pz111/DDUhbEWS5kRHkaK+HwiMAfuCLEEzkdyJMbV5YumhafjNfMPHq3LlzuE+i12md1RPf0kT1kTZRPVYGZIZ5BpnhfrWAxrNevXphjlpajokC9RLzCtFy7Lf1LVFeyrS2J+tbNE2yMqPnrT7OEThyjsUFnoNnn312cMIRTUcd3GO2r8jKszKibU90zc5ZXxLli5bhvx0BR8ARcAQ2DwI8f3kWIw+xeJWamqr69eurSo0qqn5AdR3Q8gA1vaOpLnnzEt3yyy1KXZ+qO2N3alBsUDjyu9+GfkqpnaLyFctr3PhxTmY2z1AlLpUBJGJW0r17d5144olq2bKlmjVrpmOOOSacQ0A24QChGMLTuHFjHXHEESE9ZicM+tNPPx2E9jPOOCPYDGI3iC3hSSedpN69e4eVTVaWzzzzzFA+wrkREkzEqJsN7AjxhJkzZwYNy/HHH6/WrVuHPJi9QA4gHqZVSEZmzj333EDQsP2/5JJLgkCCcEJ/o/Uee+yxevnll4MmyvAwtPibCY5QP3ToUO2///667bbb1L59+9C3PfbYQ82bNw8kwLQqCDcIgNdff33AkOtgcNxxxwWTFLQe4AkpQntE/0455ZRA/sCyT58+gYxQJ/uBGA9Wwvn78ssvDyvjZcuWDUfSgyf10VZsOU8//XSNGDFCkLh//vOfatKkiSZOnJgjoJEO7EmH2RCECSKTnp4e0jK2J5xwglq0aBHyo6GLx8Xw8WPeCDDXIAes9BB5WDKnosHwRThmnFFtQ/C537jGXKeMW2+9NdjmspgAEWdsKIs05LW/rWzmGfOSa9HA3ywKWD02xqQnGpk5+OCD/0ZmuO+Zj1Y29yv1RwPXqIO68wrMRdLRR/JwfzB3+R1fLn/TVkwtmbNgafcy9XAdPEgDjpTD9Wi5hj1HMOSZxLiQ1p4plMN9zDUWbHbaaadwT5KetpLX6rH+k976Ed9n8GDBI1o+v6mTfjAWifobX47/7Qg4Ao6AI1C8CPC85z3BewXnSWhiKqVUUspeKTrkjEN05sNnqtv33dR/Q/9AXAZogNJiaUrNTFW/zH5KzUhVWmaaeq/qrZRDncwU7+jkszQGkcBLFU0Dq/q8sPm7f//+OvLII8PLnJcxL220EJh0IXyzGQphghcxGgXcIaOZYUW/Xbt2gtRAUti8jsDMyx/hiYmCQBQlM2hfOMdqsL3UEZhGjRoV6uEcwj8rw7vuumvQQFAeIS8yg3CEVufQQw8Nmh36jCDB5D3//PMDSUI4pI+GRyg4IhxhBgYelSpVCnaTkAe0Hgg6BxxwQCAlc+bMCfnBgNVzSMTo0aODgINgCtHYc889A0Ghbtq99957B00WQhCRMmgLwiR5sMeHsHAeYQmtDXsZKlSoEMpjvFhFYMxoO4QLMgKJ4W9UnRBN9tjQP/pNIN8+++wTyCNjfeedd6phw4ahPoQrsEa9ipc4NDuUZXgbNn7MGwHuGcaMsccrXZcuXYKHOoRn5gC4EhkbxtzU2sxzSDPzn/uL+YdnO+4rCDXjDGFFK8f4cS9B/tFckpZAuSxScA8y3nYOQR9yy9yiPdTz1ltvhXIYY2sL92IiMlO3bt1AZhDOmV8zZszImX/WH9pEHTgM4FxugfwskLDowIZ6CDyRvnFPWJm0i/5gXtmjRw9dd911Yf4yV40k2AuJNNyv4M69w7MJDzP0iecB/eQ+BVMcbGAW+8ADD4Sx4j7jOkQGjPEOWLVq1fAM4W80zbQDbSi/eSbyfOIZR73xJni0n5ckzjx+/vnnHHLF8xONMwsKtJW+G3nNDS+/5gg4Ao6AI1B0BHg2m1zEe4TFwooVK6p81fKqdWwttX20rW5ffPvfCExqVmogM2na+C+WpvRYuvqs7uNkpujDUnwl2AAjjLB6/3//939hpRPhGgEXgRwzJtIlCghTCPNsluLlTbAyEeJY8UdjEyUzCG2Yr0TJTLR8fiMYYr+PidVTTz2VI5gnIzO2Z4b6IQhomvA+weSlPPqw3377BVd6CIwIMNE6rd2cp08If/SdtNF0CCMImBA6hCmIBFomXPRZGaRnpRihqXr16kFoGTJkSFhlRwBCkI3iQT7qhGSgOYKwECgfcggGbEoDEwLlExEC0aqBIwGBjfZBVKjD0nEOoZSN3IwRmi+EOggaQioREkM6xpx84OAhfwgYzphuQurZZ4JQzJysVq1aIDbMPzBlPiIYI9BCuLkP0LZBhhkjiAvjwu8aNWoEIlunTh0RIansoWGuQJYw7Yxq4RCYMRXjXrTwxhtvhPmAtpA2oYVjgQJhn3Gn7cyVZGTG9swgeHfo0CEsbEBa6If1ByLCPY57dEJuc+f2228P+KCB5HmDBpP7DCKFIxEC8x7C0qtXr0DC6SdpWUigDogEAecfaDYPPPDA0C7uHTTEaD/JAw6mhbn00kuDgwNMYtG0ggG4c39BaHjWsJDA4olpQsEcokl9XKfMW265JSw8QBpZSGC/Hv0FR9pNWYMHDw73PWad/A1eaGxtrHk2kRfCYxiGDvl/joAj4Ag4AsWOAM9n3lnIUCx4sUBfpkwZlatSTo1vbaybf7xZd2beqf6x/oK8BAJj5CX+6GSm2Men0AUysBZY5Wd/CKuQCFSYa7HSitCE0IugTrAXtgkx9hJGSLviiiuCoIQQxgudSDoECV7iaAuiwjuuhOM1M9Ym8lI/Aj17ORDkWXFFKCDkh8wYGUETAgOnbMgGJAQtC22z+kKhG//jHP0iP2ZYtWrVCn3gHDcB+XAJjaBJm1ixZdUYwsZqN+lMC4TwR9py5coFcoX5HmZz/I3QhGCKkIO2i0Cd7IvhmpEZ+swKLhqiN998M5Aiyqce6xNCqpEZMAZbSBtaItrLajdaGcgLBNWEW9xbI0giAHJkjBCuEd4I5PVQMAS4lxBgGQcLzDsEZPZi2Hn2TKEFRVi38Wd+IZyjsWBsWf2HDHOfsNeKuUBkXCBDmDW2adMmaCFs3qIVolzcdRNIy71NfjQ0BMg59ytkAk0DITcyQ/2mfWAPD/MTwkV7qZd7FY0Dc9vmLfMzWYCwc19BjHipkBYyBHE47bTTQpm0m3sHMkedhhEY/uMf/wgLCORD0wJxZ0GBgJaRexHCYlhznnsAXFlgsACRYM6DI/VxX6HFwUQWAop2hT7afc8CCXhzv3MfsSjAbxYTWESwMQDn2rVrh0Uh7mkiGk+eq2h1CJTN2EOq0F4TaIMHR8ARcAQcgeJFgGczkfcne6Z57pfZsYwq7lFR54++QAPXDhSmZIHExFLjqcvf/3YyU7wDVNjSGFQCK/0IPawQIswitOAz+8ILLwwCCgIMgnJUwLA6efGaQM3LGK9HF198cVjB5FpByQzlIkiwMnvWWWcFgQmhiZVMBHmEFSYiIS8yQ/8QTFhJ3XfffcMqKedYDWUPCkKItd36Y0fScS1KZuiflUnfMLVBcEOAQzhi75EJhuSl7xwRanCOUL58+UBEEK4QiBCAWB0Ha4Q6NEAIp1EzMxMKKQuNCRhAliiDsom0CUHZyIzhzkoz584555yAA/uD8IZmmh1IDgIg+4owtSFCzDDNw6QN8uShcAgwJowZRwvMN7QOmCgyryD/aAy5z3DKwFgaQbUxpAwiplXcAwjLBCuX+cI1VpcwqeI85XDPch8bIaWM6HyxMtAcMAdx/kCAzLBwkMjMLEpmmKNomzDXZC4TuM+Ya5ATm5fhQpL/0KSg7YHQcU/TRtpNfp5FYAAe3FMQE0gIiyJmDolGCu0h9yXaJdJB4oyMDBo0KJARnieUBTaGj+FH07hP0EJDfNBQEVicwGMZZIb80Xz0l8Ud7l/GkXZjb82LkWcBgf5wv7JgAfHiGcC+QAgmCw70g3aDI3sOwZH70eoJhfh/joAj4Ag4AsWCAM9WIouDyFAsRO9ccWft0WgPdZ3dVXeuuzPsgUnND4kxWuNkpljGpkiF8HInYOaFwItgwEsXYYKV0FNPPTUIFQg3mFYguDD45COacMTksGBkBuGYPAg0lt40M5igoKlgQnHdNDMIUJyjPgQT2sSKJaYzmERBBhDkEbTzS2aoG8GGdrECjWaJlWnIA+UhYJAm2gfrC+don5EZMzOj37ST6yb0IfhDLhDOIF4ILZRLO4ms0mMGhmADvpRBu+gr9vloiDD7QUCiXQg4mB5FNTPkwbsZ2ik8yIFhtO2JyAzlsPqNYEo9CM5o3MCD9qM54G/ahlBsZmYcKZ+VaA+FR4DxQShmBQitQs+ePbXLLrsoLS0tjAekkpV67HXBn/Q2pvbb5gpmf0Zm7BpjmF8yw3wjPQGtD3MIDSx7cRCwTaORHzJD/dwbzDnmD0SIdqLhYDEELQl1kSa3wP1NerwjWj+51yEWkBnKYC6iqeH+g3hzT7AYwRHtKlpNnjXci+zJA1PyoXXimYY5GmVY3609pjnDRPSOO+4IpnGQGYgLIUpmzLyU/lCOkRn2HJmGh3sYIgee3DfkR5uG2Rv3HvnwkkObMWtDC0o/OPJc5dmECS3p8sLN+uBHR8ARcAQcgfwhwLOV9yBWMCzgVahaQQe0PkA3fH5D2OCflpG26X4YIyy5HZ3M5A/8zZnKXpi8cBFIsPtGgEWogLViEsZqIS9lXvwIHAhiCO72Umdy8DJHmMD8gxVT7MYRRtgzgNDD5KFMyAyCG6uXURKBYMcLntVKCAEmbWh3qBuhG+JAuxDy2YyL0GJkhlVQBAE2EdMmhDQ0TAjtUXMPykV4w5wK23mEHtJaPxLhbEIF/cPkCwLESqsJkvQH8oEwxuZt0kMGWUFG62OEgf6jXUEQhaAhWLEij9AELgTSYMpGXgQaSAh1Yu/PajSBtKx8Q+iMiHGOSN0QQMbs0UcfzRGIKBftCgIVWh+OaLNoA/0AV9qKxo0N5BYoEwGOcSVQvof8IwBe4AfxRmPCfgwIPhqHlJSUHDJj+ycQpiERNrfIz9zkbxvjeDJjczdKZkyQ5xp7ZphzppmhLMaTfR60B7Mmxt40sQXRzFjbMA3DZAqNJBoRNBncEwjvhLzmDWSGlwpEjzKZr0ZmOE/ALAwTUbS9LHxgYsm9TOQ5xb3FPMZkEgIE3txDzHe0sTwjeF4YXjxP0ISgDWNPE4sIaIAh/JAZ8CREyQzPAPCjffQpEZnhfkbDxPMNYoP5GPWjvaEPBPYIQcIwzbP2o92lH2jVeGYQ8sItJPL/HAFHwBFwBPKFgL1bkYfYc1m5emXt32p/XTP7GvXP6B+8k+GlrMD/nMzkC//NmoiXOwGBi5VZCAACMSYwvNSxUUf4QsgiLfs82NzPpmJWQ0mLJgLBCHMOyAGEhv0Y2IlDSFiJRkhHOEcgQDti3oHwYsbKLgIIpk+sDENy0ITw0qcMBDjahykJG27RSphwQtsRCiBCmGdYG6kXAQsNjAmFXINo8c0ICBll0FYmeDLBwfLSHgQjSAT5EVggBNjMsx8FHOgf6RGEKJs+ISChFcHuHhLHKiykhzrpK8IkK/KUhxAIcUHAwpYeQQjcEHwwPyLQB8yUMHuBfIAJK89oriBr1MXKNaZi1nbyMC4QQwQr2gtZREDmGm3h5macEQIZT/AG/44dO25CjEIj/L9cEQB3MOWeQTsHUWCs0YBBLvAsxhgwtowxc5R5AsFmXMhLJFAW0cgMiw6mmbHrHI3M2J4ZS0990X1PaAuYX8xhNCi0CS0hwjbzjHlDyI9mxvrJvGMe8uxg/kGOmDu0wfoRCk3yX15khnpoN4sP7I+BjFFnNEJUSAPRgZywmMJ8Z/7SJ547kBDaRHmYg4EV9xzkDzLEIgT3I/nZ90NgAYf7gfsNjMlLJMSTGc5TB8QEEgYmEBSea9y/1E1AE3b00UeH72fxrIv2A5LFPenBEXAEHAFHoPgQ4PnM+4j3HTJT1RpVtW+TfXX5xMuzNTJZaSqQaVmU8jiZKb6BKmxJ9nJGQObli7CAKRkCNloBSAlCu60qIjAwGRDsIQuWHlMLzGhY7eelzUucVWgEbty/oslBMODFzeZeiBKCA0TF9mjwm1VqBBOEAsyp2AeAqRtCCSvXCP2QIco34YDVWIgVghmTFcEFzQSEAmGRPiIgECEvfKeGDdgm0OcmcFleMzMjH1oh8EEApR+Uh1MBE5TIA2li5ZcVcfoJUejWrVvQfIABaRCGEKbAiTSUh2cmVmtZZWZVGFM/iBBkhEAfEHggiBA7BEhwYXUczNBOkZ6N3NRB34i2as3KNuPKeHLdIkIV40d70KgxthBCiCbjbekKO8+2p3yGOXMAPNEEMtds/BC4MZdiHCDAYA+xgfyyiGBCM2PEuDKWNn+5HyFCaC64F9F+MLYI38wdCDPaNcaL+YN2Am0i9xGB+xPtA/cQpIB03C/kZd9XvJkZRAgNgwXmM2QKDQyBvlIGcxbtBoQIsmRaSq7nFRKRGfCiHDChDPqP2RpYoj0yssG9wH3Pfhe0ImiOuY/QMPOMoB2mfbVyaC91QvoxL+O+AUvIDNppFmasfJ4XEDRMQ80FNBpVxoz7F0JkZmaUzzMAbTTPMrCAoKKJJT19oG4cMPBc5JmGO23q5zzzBa2v7YfKD3Z5YevXHQFHwBHY3hHg+UrEUoJFd7T8terXUoenOyhtTbrSikJkIDVOZkrOFGOg0bqwSokwhHDAS5tzJgzQWtLxUkYIQ/jF3IM8vKARLLhOYGWXMiAnZm5hL3OENEylLC91MMkQAhBIrAzSIdTQHggQq8yQCo6kt3T8ZnUbskSADGGqQbsRUggmyCDgIchg146gYm0KiRL8Z/01MsMGfdqL5gRhiVV2hE/KiQ+Uz4ovwijtN89u1m6EGNqIAIMwCFb0F+GUgGBEveSDABEsL/kQ1MCevpOHawimCEPgaOk5T//Bh/aATXx7SQNuCGyYurC6jKCFgBWfNhTs/yVFIIo3ZIL9Ggj67H9iwQBiy2o9fxvZxozwpptuCiaGEA0IKgIxAj17maxMyqtZs2YQkiHwmGdCLLj3eEizOZ88EHmI0aWXXhrIDCtRBOYc5BltEefYvwEx52/2pKCRI3D/QhrQwDIPLNA20kFwCbSLyLxDg4EJI+3iPrNrljfZ0cyuICvMNdoIWWNRhb0kVg73OeZxkHg0VLQd0zYI/eOPPx4ICOSEPBAa+gl5hzTQJsiK3ScQSrQnaGF4HkA6TXPJOe5pAvcdCwo77rhjwBIyiRYLIsgzAHO6KJmh/RBGHJfwrAA/CKI9fzhyX1M/2jGeQ+QHO7RJzAHIGX32+y7ZjPHzjoAj4AjkHwF7/vIs57lf/YDqOjn9ZPVe0lvpmemF18iYdsbJTP4HY0uk5AXKoMdHzkdDsnS5pYmWkSi/nStKuvi89jdHBAOECIgBm2zx6IXwznlLF22//ba8UTKDAEo+IliRJlEZnMsLy/jrVl60fs7Fh/h8Vr/VaX9H89m1+DpIwzWL8WVHy/Df+UMALJkfkEM0MAjKCOBoYNBSIhBDhiGY4M3chNCgeWS/FyQEMoMJJWTCxgYNKho+BHUEX8zY2LdFGQjBaOXQBkAQ8EiHORWEhw/iEiiH8hDKqQMhHxNJNHB4NIPIEtAkoBmlvRBnC5SHVja6H4a66SvaUDQVmFFRD+fzE9Am0k8WOKwsFksoh3ZZoEyIOppcNE0I/xxJQ99ZUKFtaEQgj2gWIV8QGvahoXVBi8siCfcz5YAB5UCMIIBoYdC82oIC/QJfxg7SBOaPPfZYWMxhsQfyhxYIAmNjRB9Y8KBMiBJjZtfsyEIHizmMF/Uz5hBYzG7NXJW0HhwBR8ARcAQKjwDPYyKyH8/ZajWrqd6F9XT9l9drQFa2++UC75ExEmNHJzOFH6DNkZOXpw189JjopRq9br+jbYovK1pG/DXy27mipEuWF4GRPSEIKQgPmMZhRoKgEs0Tbb/95jrpomTGtBWct7Zb+uiRvIaNHaPX+W3no8dom6yM+Hx23vLZdTsfLSP+GnmSBctv5eaWNlkZfj4bAbBkjqAlQ9hGWEewRfMAIY5qBm28mKtoWtD4oZ0gDdcsUjJCO2XxcDZNA+cpF8Gba5hAUQfn0JqY2RTpKAthG1KDGRtaCIRxNB8cCYw7ZVMGpN8CGhuEbcolUBZpIT+YW0Iu0C7aecuX25E+00bKsDIpH9w4b8EwIB0YodFEQ0R+rkFUMPtkTx8aSDS9tIXfrMihNeIaRIw2Uz7X0ByDB32nLPCweU+5jCGkdMaMGTlaUK6DC5iRx9JbW22JDzwAACAASURBVGk/18CPMuID58jDs4T6iWhMDdf49P63I+AIOAKOQMER4DnLcxXnUnwCoFbDWurwbAelZ2SblxWZyLiZWcEHxXMUDAETGBBOWJm21VpWRO1aXiVaOgQWtDms0JpgyDUPjkBuCDBHbA7xULU5Y+ejee2cpY/+HZ/OyiJNNCTKa+XkJ118mkTl2zmOEB4IF+QLLQ8aFoR8SxMtL9lv0iZKn+h8tN/2myMRDQn767hHIXq0Cy0MpAWygMdDTOsgM9GFCKsneoy21c5bffxtwa7Z33a089G0do1j9Hr0dzSN/3YEHAFHwBEoPAL2zEajj/l1ym4panxzY/WY30P9Y/2Lbl7mmpnCD47nLDgCsHJWrPH4xUosIZmQkax0hB+EI1Za+e3BEcgvAsw15gzRBFd7yMaXwXWukTZZGs5bjJ/HifJandG6LD9Hu25HS8ffdj16ztJxZN8ce13QyBDRgFo+y5PXMVpeNK2dj57jt5VvGFkb0a5gJsYGfky2WInDVA2NEeYF9iFbNDvR/ttvq49jfCCN1Re9bnni00fbmeianSN/tH4770dHwBFwBByBoiFgz232VLKYtffxe+vCVy/UgFgxmZc5mSnaAHnugiNgQkdUECl4KQUnQYWpw/M4AiUdAbuf2OiOK2b2pqAZMScgRb3PCtN/6mTRAo9m7Jdhcz1eAtmvhLc2iBcOQbZG2wrTH8/jCDgCjoAjUDgE7B2FJQ37SFNqpOiE7ifo1nm3Fq9Wxs3MCjdAnqtwCDCxbWW1cCX8ZRpS2PyezxHYVhAwQgA5YD+Z7VvZ2v2zdmHqhhYV5wVRs1C7vrXb6fU7Ao6AI+AIbD4EkPd43uPUBhf/1Q+urnOePkfpWenBlXKx7JVxzczmG0Av2RFwBByB7RUBXl6YFsSHZOfj0/nfjoAj4Ag4AqUfAd4DPPf5ZMF++++nehfU03UfXaeBWQOVmpVqNKR4ju7NrPRPGO+BI+AIbL8I8LKwWNJQ8L0oJW1EvD2OgCPgCGx+BHgnEfhOGBv/K1eprJMHnKyev/csfhMzNzPb/APqNTgCjoAj4Ag4Ao6AI+AIOALbCwKmnf/q66/CdwUr1qqos585W2lZadkmZrFiNTILZabH0tVndR+lHJqi8hXLa9z4cVt1D2lxjvUOxVmYl+UIOAKOgCPgCDgCjoAj4Ag4AskR4CPUaGf4vhhezPZvsb86TepUfB/JjDdOczOz5IPhVxwBR8ARcAQcAUfAEXAEHAFHIP8IGJkZdt8w7b///mpweQPd9PlNTmbyD+EmKV0zswkc/ocj4Ag4Ao6AI+AIOAKOgCOw+RAwM7NbbrlFNWvWVNPbm+q2n29zMlNIyJ3MFBI4z+YIOAKOgCPgCDgCjoAj4AgUBAHb/M/xkksuUUpKiloPba1ev/faPJv/3QFAQYbH0zoCjoAj4Ag4Ao6AI+AIOAKOQDIEjMysW79O7c5qp0qVK+nMf56pPiv7iE36qbFidsvsZCbZUPh5R8ARcAQcAUfAEXAEHAFHwBEoCAJGZlasXKE2bdqoQpUKavfvduq3pp+TmYIAGUnrZmYRMPynI+AIOAKOgCPgCDgCjoAjsLkQMDKzfPlytWzZUuWrlNdZI89Sv3VOZgqLuZOZwiLn+RwBR8ARcAQcAUfAEXAEHIECIGBkZsWKFWrdunW2ZmZEO/Vb62SmADBuktTJzCZw+B+OgCPgCDgCjoAj4Ag4Ao7A5kHAyMy6devUtm1bVaxSUf945B/qu6qvm5kVEnInM4UEzrM5Ao6AI+AIOAKOgCPgCDgCBUEAMkPEPfNFF10UvJmdeu+p6vWHezMrCI7RtDlkxsD1Y/YkcxwcB58DPgd8DpS+ORB9wflvR8ARcARKIgL2nZkbbrxBu+22m5r1bqbb5t2m/ln9lZrl3swKOmaBzGRmZmrDhg0eHQOfAz4HfA74HCjVcwAC6sERcAQcgZKMAHI34a4hd2nfffdVw04N1e3Lbv7RzEIO2g6QmB9++EGTJ0/W5MmTPDoGPgc20xyYNIn7i/vMo2Pgc6C45wD317Rp07Rs2bJgwlHId6JncwQcAUdgsyMAmWHh5fnnn9dhhx2mg9ocpM5TOzuZKSTyOyxZskS9evVSlSpVQqxWraqqVk3x6Bj4HCjGOVC5cmURU6pUUdWq3GMeHQOfA8U1B6pVrRrur1q1amnq1KlBs+QamkJKBZ7NEXAENjsCppn5ePbHOv7441XtgGo678XzlJaVprRYGp+5LN5/sbTgXKDP6j5KOTRF5SuW17jx47R27dpAqkr783KHxYsX6+abb1aZMmVUvXpl1T18P9Wru7/q1eXo0THwOVDoOcC9dMT+qlNnH1WuXCHcY7V2311HHHqo6tWu7dEx8DlQTHOg7qGHqmzZsmFBDg0NFgel/eW82aUpr8ARcAS2GgL2fMI98znnnKMKFSuo9ZDW6rOiz+bxaLatk5klSxarR48eqlatirp0bqfVKyZo/Zr3tG7lBK1fNdGjY+BzoJBzYC330ropWjBvjE5pdayqVqmihwcP1tq5c7X+5589OgY+B4o4B9ZxL/3yi1b/+KPqHHSQdqlRI5hxOpnZajKaV+wIOAL5RMCcAPTu3Vt77rmnjr7yaN34xY2bx9Rs+yAzt6patUq6tms7Za6bLGVNkzKmSpnTPDoGPgcKOwc2TJVi7+u3xW+p9SnHBhOzx4YOVWzBAmnRImnhwuwjvz06Bj4HCjYHuH823kNZ8+cHMlPDyUw+xShP5gg4AlsbAds38+abb6pRo0bard5uumDUBeof66/UWDF7NNueyMw1XdopY80kxTKmKmv9ZMXWT/HoGPgcKOQcyFo3WbGsGVq68C21bnWsUipX1qNDhijz118DoYnNny+PjoHPgcLPAUgMiwMZ8+Y5mdnakpnX7wg4AgVCwEzN2O5x/vnnq0LlCjqp10nq+VvP4nfRvN2RmbWTglYGIiNWlj06Bj4HCjUHstZNUSw2U0sXbdTMbCQzWREyI7Q0Hh0DnwOFmgMQQbQzTmYKJEN5YkfAESghCGBqhoZm2H3DdPDBB+uAkw/QZeMv08DYwOL93sx2RWa6tlNGIDPTwmq8Exkncz4HCj8HIDNKRGYQwBYsCFoZJzJO5nwOFH4OOJkpIRKZN8MRcAQKhQBkBg3Np59+qrZt26rSLpXUvE9z9V7WW+lZ6cVHaJzMFF6Yc0HYsdue54CTmcILqS7gO3b5mQNOZgolP3kmR8ARKCEIQGQgNGvWrNGQIUN0wAEHaL+T9tPFYy/WgNgApWYW094ZJzMukG8vAjmmhezzsOimhkWb+05mXCDPj0DuaQo/T5zMlBCJzJvhCDgChUbAtDNz5szRBRdcoJSaKWrUtZFu/unm4vNs5mSmaALd9kIESms/Y+unZpsUZkxTLPN9ZWV8oKyMD8MxljlDOdd930yB9804mSm8kOoCvmOXnzngZKbQ8pNndAQcgRKCgGln+IDlyJEjddRRR2mX2rvo1PtOVeqq1Gxzs6J6N3My42SmtBKVvNqd7a1ummIZM7V+zWT9Mvc1zf7oSc16f4Q+/+xZ/bZknDI3zFAsI5vU5FWeX9/0XnEy4wJ5fgRyT1P4eeJkpoRIY94MR8ARKBICEBri3Llz1bNnT+2+5+7a58R9dNHYi7LNzbKKaG7mZGZTAc0F1m0Hj9iG6fpz2UR9/NETGvF4T93e4ypddUVHXdaxo7pc3VED0rto9Kg7NXfu61q/BqcQ207ft8Q8djJTeCHVBXzHLj9zwMlMkeQnz+wIOAIlCAHMzYgzZ87MNjfbLUV1zq6jrh93LbqrZiczLsBuCcF3S9YR2zBVEJnFC8dr1Et36bJLz9Vxx7ZXo0Y3qnnzoWrV6gE1bZquo4++TE2bnKaed1ypDz94SqtWTHJCUwBzOyczLpDnRyD3NIWfJ05mSpAk5k1xBByBIiEQNTcbM2aMmjRposq7V1b9K+vrxi9vzCY0mJvF0lTgf05mnMxsSaKxuesKm/w3TNMfv72jJ0b2U8uTW+vYY29Up06zlJa2XoMHx3TXXTENGhRT797L1aHDSzrkkDN05eXn6INZT2rtKj6k6nMiP+PkZCZbSEXgLEx0Ib/wQv72gp2TmSLJTp7ZEXAEShgCRmj++OMPPfnkkzriiCNUaddKatilobp/013psfRsMlNQQuNkxgXX/AiupSVN1vop2rBupsa8NlSntjlFzZr1VWrqskBe0tJiSk3Nyonp6dLQoTHddNNXOvjg03Vzt4v13bejlcU+Gv+oap4OAZzMLAhfZ89auFCFibGFCwv1IcXtRZD3fm78VpN/NLOEiWPenG0BAdvDEX+kb5zzsPkQMEKzZMkSPfroo8Fdc4WqFXTUpUfppm9vUtr6NKVlphVMQ+NkxslMaSEqebUzbPjPnKkF88aoc6dz1KjRtbrttl91551Z6tcvQ5CZtDTlxNTUmPr2XR8IzcWXvKvDD2+ip57sq7WrJyuWMS1PYT6v9mzr17d3MpM1f76WfvedfvnyS/1c0PjFF/r9+++V8euvTmgWuIYmGXFzzczmE6i85NKJQDz5yO/f+e0tezr4Yr0TmvwiVrh0Rmh+++03Pf7444HQlKtYTge2OVCdZ3RW6vJUpW3YSGjyo6VxMuNkZlsRujPXTlYs82O99sogtWl9jtq3f0aDB0FkMnMITJTMZP9GW5OpO++MqV69Lup+U0f97/vRysp4X0FYL8Aekm0Fx/z2Y3slM0HAXLxYq374QXdcd51Ob9FCbZo1y44nnaQ2ecTTWrRQqyZNNLhnT83/+mtlLVoUzNSSCbR+fvslO05mCicoea5tEwEE4JUrVwoBOD7+/vvvIsaft7/Xr18fCApEZcWKFVq6dGmIaAeikTL4wKORpG0TyZLRKzC28XjllVdUu3ZtlStfTlX2raL2I9qr14JeSt+QHrQ0qXg7y43UOJlxMpNf4bWkpwuamaxZGnLX1Tq5RRd16TJbgwZlBrLydxLzl4YGsoO5WZtTH9fZ7Tto0nuPKGvDB+HjmiW9z1uzfds7mfnjm2/UrFEjXXvUURp5+ul69h//0LNnnqnn/vGPhJHrT3P9vPPUYe+9dd5pp+mr2bOVuXix0PI4adl+SUuysXcyUzKELm/F1kUAoZcAkenXr5+OO+44NWjQQEcffXQ4NmzYUI0bN1bTpk03uWbXuTZx4kStW7dO//vf/9S9e3cdeeSRqlu3bhCg69SpE44I061atdLDDz+sP//8M3jesrq3LgLbbu1GaDZs2KBPPvlELVu2VLly5VSmTBnVPb+uOk3vpPRV6RqYNTCb1GSmKi0rgQmakxknM1tTIC7euqdJGdOU1u8ytWl9u7p3/0UDB6J52dS8LJ7YQGZwCnD+BePVpnUHjR1zj7IyZrlmJg+tlJOZb9S+ZUuNvfhirbvzTmUOHarMu+5S5pAhyePgwcr85z/1z1NOUadzz9XXn3ziZMbNzJISWScz264Q5z3LPwJGKFatWqV7771Xbdu21emnn54T8YpVs2ZNVahQQYceeqjOOOMMnXbaaTnHs88+W9OnTxfamS+++EIdOnTQrrvuqgsvvFAjRozQv//9bz322GMaOHBgyEc5J554or755hshZNsX7PPfYk9ZEASM0IAzWrO+ffuG8dmp7E4qX7O8julyjDrN6KT0NekarMEaqIFBS9Mvs5+IqZmpQnOD84A+q/so5dAUla9YXuPGjxMf6qR8m0MFaVdJSrvDkiWL1aPHrapWrZKu6dpOGWsnBYHXN3hvoyQuc5rSUi/TKaf0ULduczeSmaxczMwUzNCGDMnSuee9odNOPVdvvjHMyUweRAYS6mTmG7Vr0UIvduiglX36KDMtTZmpqbnG9X37KvPuu3Vvs2a6okMHJzNOZJISGbQ1TmZKkjjhbdnaCCCQop0xkzI7QlTOO+887b333ho2bNjfrpMOrQz5v/7665D2kEMO0QMPPBBMylavXi0iZc+fP1//+c9/VL58+UCaFixY4GRmCww8Y2OkEUIzadIkXfB/F6h6jeraudLOqrJPFdW9oK46PN1B3X/srv6Z/QOxGaRBgdz0V3/1j/V3MlO8GoJtlCjkQ8DdmjgiXMcyP9Cwe6/VySdfpas6va9Bg/BeltueGSMzMbVs+YAuOO88TZv6mJuZ5WOsncx8o3bNm+vFc8/V6j59FEtPVyw19S/7xTgVYCwtTRn9+il2zz26t3lzJzNOZHIlMsVJZrBLz08srauXUUGoqHKXlYVg5aHkIcC42FzOyMgImhPMkyAze+65p4YPHx6uo4WxdHZkbCEz5557rg466CA99NBDIQ3lWBo0MeynadOmjapVqxaEaiNCpfX+KHmjmLxFdv+BOcRy1KhRuujii7THnnuoXJVyStkzRbs32F1HXHiETht2mi5/93Ld9N1N6rmsp9Iz09V3XV9Vq13NNTOJhHG0OFnrJosN5oWNiTRBCIPx5SVKl6hNds7axbGgea2MbeWY7QDgI/133DCdflp7nXnm48GTWTaZSWxqhglaejpkZ62OOKKjet1xpX6Z+7qyMma6mVkehMbJTDaZeQky07dvIDO5qgA3khndc4+GOZnJU5BPto/EhPysX39V5rx5myUGjUgc2Qr1UefGyN+J0iVrN2lz8uZzj1RxaWYQEAoak4sbfsUR2LoI2Fw2YvPpp58GgmJkhuuQEkvH0Vb948mMlWFHSBCmSVdffXXQzrAxHfM2y791e7591B4dN7RlP/zwg/jIZs+ePdWsWTPtstsuKpdSTlX3qardDttNex69p/Zttq/qnFtH9S6rp3JVy6lCxQoaP368m5ltIuBnTFcsa5ZimR8WIc78m6vfWNbMBOXNkDLyr9nJaVfWB1LG9L/VsUk/8hBOS3tayFwsY4aWLBqn7t0u1DHHXKbrrvssaGfYF8M3ZrLdM2fvo4HkENn837btszquUQu9PGqw1q2eotg2jlVxjLWTGSczyQT3zXoeIoAHuD/+UGzZsuKPf/whLVkiRQnHwoV/1Wf1/v67wreCouniCFAODqRZsuSvMpYulfLxnaGikBkTCDDZePbZZzVy5MhgPsP+gGh84okn9Oqrr+rDDz8MnqAQlyxvaRCdEFoxI/rll1+0ePHiIMQWtt2Uhfcrylq0aFEQYAtblufb/AhAMIifffbZJpoZakbbEg1GRqJk5p///Ocmc515j4aGDzriZKBq1ao5e21K0z0R7Xdp/Q3eBBtjzAC5L2d9MEsvvPCC0tLSdOFFF+r4448Pbp2rVq8aCA5EpsyOZbTjjjvqrbfecjKDsMeX4GMbpum3xeM0c/rjGvv6vXpz7DC9MXaY3hx7X57xrTdIM0z/HfeAvvr8uRzNSdb6qdqwdpo+m/OUHn+sl0aO6KsR/+4Tjv/75iVtWJs/YTpz3WS9MeYejfh3bz37TLoWzhsjyi4OQbW0lgGhydwwQxMnPKgO57TTccfdpG7dvtbgwVlh/wwfyiT276/wIc3Bg2O65JLxOvjgVkpP7aqf576u2IbpOWNVWnHYEu12MuNkJkdYTybEF/P5INwvWqSV//ufPp4wQZPHjk0c33hDk4ljx2rSmDE5MaS3axuvR8uYMnasPnjnHS347PNswjR/fiAsa378Uc8/9phGPPigRj70kP49fLjeffll/fbdd+F6aFeSvnIttnSpvn//fT31yCN6/P779f748Vo5d26+8kJ6MubNU52DDlKNGjU0efLknFXn3AQVhAEEgZ9//jl4fOJL23hvOvzww3Mif3P+mGOO0cknn6yrrrpKzzzzTBDkS7rwZsIp/cO06JxzzlHv3r01d+7cHCEoN3yi16JYDR06NGwQv/nmm8OKPOlMsIrm8d9bHwETdItKZigHEoMHs9mzZ+vWW29VlSpVgnYGUyeu+RzYOuNtzyG7Rxmr1WtW69f5vwqN3HvvvafXX389LNTcc+896j+gv3r16hUiDhwgtdvC2BXJAQDCGuZGH7z/hK7t2l4nNq6nJiceGWLTE49UnrFJdprmzRrorju7BpMyVvwz107VqpUz9OjDt+iIugeqXr0DVe+I7OOQQV017+c38vUV+vWr39PZZzXVkfUOUsOGtfX+9EeVgXZiO9YqBO3Mhmn6c9lEjR51l9q3P0uNGnVShw4vqVu3H9SnzxqlpmaoZ8/luuaaOWrT5gE1qP8PdbvxYn3+2XPasDabxG4JMlDa63Ay42RmS5MZTLv0++/6Zto0denYUaecdJJaEZs2zYmci8boNX5vci2S18537NBBr4wcGbQomIXFFi3Sb198oeMaNNARderoyMMO0xG1a6t1s2aB0GyArCQhMuCTyfU//9ToESN0wjHHqG7t2hrcq5d+/eorZS1enKupWlE1MwhhuKKtXr26dtppp+B2lhc9Xpsspqam6oYbbtApp5yiAw88ULi5hRT8+OOPm6xabx1RJnmt9A3BBoGmY8eOoX8nnXSS5syZE9rN9fwGE2YRivF8tfPOOwfCxwo9YVsQhvKLRWlKx7gRC0Jmvvrqq0BWWRjA4xmmS3fccYduu+02de3aVe3atVO9evV0+eWXB2Jj+2VKEy7bYlujpCZKbOw8poHLli0L2tlff/1V8+bNy9HKbAt4FJHMTA4fT3z3nYfVonkD7bDDDtpppx2VklJBe+21m/bacxfttVfyuPdeu4brBx5YS3f0uDiHzGSsnaIVK2Zo0MBO2nHHMipTZodQNuXXP+rgoMlZv2Zq0LLkRkzWrZqoBvUPUtmyO6lSpfKa+M792WSmAGZqpV2gTtR+IzR//P6uxo+7Xzdcf5Fan3Kemje7Ti1bDlDr1kN1coveat78CrX9x1m6a9A1+vbrF7U+nxqxRHVuj+eczDiZ2dJkhj0yWrZMH7/zjhofc4wqVaz4V6xQQZUrVlTVKlVUvWrVEPldZoe/nq9cr16tWojVUlJUuVIlVapQIZTBNeJRhx2mR++9V7Hly4NGhI+aLvrkE6XElVW2bFn17d5d8z7/PCkpgeRk/Pqrslau1Ij77lONatXCs77b1Vfrp88/z9Mtd3GQGezN2UsAmenTp0/QXGB6hi06EYH9p59+0oQJEwKJQVOzzz77hLRcN2GhpAkECLG0DdJl7npvv/32YF9PW7me32DCETgMGDAgCLRdunQJ+OS3DE+35REoDJn58ssvA5mpVKmSGjVqpEsvvTSQ4UsuuUQXX3yxrrvuOmF+hgvnbWVVf8uPzOat0Z5J9gygNjtn93JB7v/N29riKb1YyMzECf9Sq5bHhJdQ9WpVdOYZjTVs2M26d+i1GnbP9cnjvdnX/vlgd02e+JCy0Jqsn6qMtVO1csUMDR3cZSOZKaM9a+2icuXKquxOO6pPr8v0y9wxeWpnIDPHH3dYIDM1alTWpAkPOJnZqJWC0GRtmKYN66bpx/+N1nPPpiqt3+W67pr/U9fOF6rbDRfq7qFd9M7bw7Vi2buKsS8KE731U7ZrM72CkDInM05mtjSZCZqZpUs1f84nuj89XT2uvfaveM01uuO66wLB6N+jh/rfeqtuuOKKQFhYKCJe2L69+t52m9J79FDqzTfrjuuvV49rrgll3Hbttbr92mt1d9++mjVuXDANQzMTyMynn2qXatXCh9wCUUpJEWTm+KOP1hvPP68NCxcqC1fKcRqaHDKzYoWeGD48h8zc0qWL5m5BMrPHHnsEG/L09PSwcmlCQFQA4Dc26f3799cuu+wiPiTIhwZNOCieV3LxlUK7CNjSf//995oxY4Y+//zz8HdhaqE8ysI0hbLwkBW/76Iw5XqezYdAYcgMmhm8mfFdmptuuinsFWO/2EcffRTGnBV95oLFzdd6L7k4EbDx4mjzojjL39plFROZeUStTm4YXob77FNTQ+66TorNyf5CfOY0Ka+Ykf0xRwTFsKdj3dSgmRkyyMjMDrrw/FY69JB9tPPOZXVMw9oa8/q9Wr9mWq57YCAzxzWqHchMtWoV9d67w53MREzssjU0UxXLnKlYxixlrJmkJQvHav68V7Xst/GKZc7IdsAQiIyTmIIQGdI6mckmM/GumXHBnDCmpmqDuWb278wU3psZxGH+fK2dO1erf/rpr/jjj+Hchl9+CddJ88V772mP3XYLz27IzKsjR2rVRtMv9qEkK4NrEDXISA6ZqVo1kJn6hx+u01q00O677qry5coFQjTvs8+CdoY6owQvGZm5uUuXLaqZMTKDSRkb2xHSbR9A9OWPEPDBBx+EDwZihoPpDSFqsmVCg73c7W872nk72nmOBPvbric6WproMbe8tNvMzvJTXrQsa5fly6usRHnj22llRY+J0kTP2e9oHv+dOwImtBbEzMwcAOy11166++67cwRfyoqOAX97cARKCgKbh8wMvlax2OwgHBdUAExEZnjJPnBft7Avp9Yeu2inncrotlsvCtoZvHORJ1E9Tmby5+wA/ILgHbzSva9YFnFm0MQEd85J8E2EuZ/7C3MnM9/orBYtNOq887S6Xz9lDRigLL41079/wsi1jLQ0ZQ0bpvtatNCV/tHMTQT/KAnI8zeewBYtShzxPLaRiHw5adImZOb1J57QGrxUbUyTWxl/IzMbNTNNjz1W/77nnqCV2bFMmXB89emnlUG5pYDM4PHLiEz0RW1C/Lfffqsr0GhVrhzMcUhjgh3CHr+JJvjFn4uWaXlJYyGa387FH63MaD47Z22xPJynP3giS2QWZNfj8yVqh6W1sqyO6JE0RPJzJNjfds7Ox+eL1hnNY+ctfzSf/06OgOFWGDKD6eWQIUPC3GH+WPQxSI63X9l6CJQaMvPYI7fpnfH36/jjDg+mZ8c2qqNXX7lbG9ZND+ZpiYRoJzN/CdaJ8Ik/F7zTBVM/zP0SE8T4PP53coydzHyjts2a6dlzztGSXr20vF8/Le/TR3/27Zswcm15795aMWSIhjRtqsvPOUdfzZmT576JseYoAgAAIABJREFUPAX7ONOm7SV98BLG5vpc4pdxmpnXnnhCqxcuVCYal1zyhb0q8ZqZjWTm6COO0McTJ6rLpZeGfTnldt5Zt3TtqgVffKHY4sWbEJqSqJnJi8ywcs0+AsgM5jgEBDwCZljsUcEUZ82aNTnn+BtHA7g1JiCom7BOfZh/vf/++5o1a1b4aCH7dBIRD8sLmWATL5v7yTdz5kx9/PHHwZyMTb4InhZoG3uA0DhxzdpqZfE3X3FH4KUsTMjQPrEnAk9V9MnIB+WyT8jKsjqiR8pjszHtszLpF3sxlixZEtpm5UXz4SmL/UtgRZ2UsXDhwoAH+Hz33XfhY42J8kbLye03eYsr5lZPSbnGWBALQmbMzAwyg+c6QlEw35pYFNdYl9b+b03st3TdpYbM3HfvDVq6eIKu7tRO1apVCtqZW7r/nxb8OjZ8O4X9NvGCtZOZ5IJ2PFb+d/Fjtb2TmWXffKMzTjpJ3Y8/Xk+dfbZeOPdcvdChQ67xubPP1suXXKKOtWvr0rPP1jcbyYwJz9sLEdnc/UymmYHMrLH9LXFalERtopwcM7ONZKZenTr67MMP9daLLwatTJkyZYIzArQzmUuXKuzr2UgwSyqZgUggBJoQwxFBHgEb1898Y4N9M3h5IkAuCAjd119/fdgkz94ShPE33nhDPXr0CG5s+Vq3lUUeBHQExlatWumQQw4J+3Dat2+vESNGBKcDpIm2gTZBTCAdeJlq0qSJ+FI7Xtbq168fNmjzjRw26lubIAY4L8A724svvhj6QFspl37STr5Jgetp9gFRFu6pTz311OC+ddq0aSEP6SEcuHodPHhwKCt0OvIfaSBMkyZNCnlbtGihQw89VIcddpjOOOMM3XfffUJYxgMWfSHYkTw4FcDBAMSMetnDhPcs2vZ///d/4Yv07FuiHmJBAumpq7hiQesvSFuLK6311cgMpmO46SYw9tFAWvq0rZCZ7XG8o+O5vf0uNWTmniHX6bff3te4Nx/QccceFmyz0dK8MnqIMtcn/hq9k5niF9Cd9OQf0+2azCxapNU//KBeN9wQ9k/kuP9t0kSt8oinNm8e3AgP6dVLC3DPi8lTPgTrRMK2n1uQ0FRtc5KZow4/XJ9Nm6ZVc+eqa8eOSqlcWWhnbrzqKv32zTebaGdKIpmBgEAEEPYgMCbk8eVzBH9cN+PpCYEfLQaBawhPb7/9tmrXrq02bdpo9OjR4Xsc++67r6pVqxY8Qz3yyCMhHaQIzUfbtm3DV9RZBT/xxBOD2+e9995bu+66a843YaJCGUSBD+IdddRRoQ0Ip7jJxV30/vvvr913310VK1YMBAmNDwFtCKSLj+RBKPiAJoH+QQz4YjjfDKGdLVu2DCQGL1aUB2Hr3r170IiQh/T0nzooi0D7CODEl+AhYrSJDypCtCB+tK9WrVphUznYTJ8+PdRPHhOqH3/88VBn8+bNQ50QGHChXfQLzMEF98D2xflQcT7/MzKKxqw4orU7n9VvlWTgS2TeQrJPOOEEPfnkk6Et4BENpGMs0SDisYy0jz32WEhiYxxNX9J/Mz7cZ8Ux1pRTGsa7pI/J5mxfqSEzw+6+Xr//NklrVk1R56vbKSWlYnADfeP1HcJHOxPtnXEyk3/B20lK8WO1vZIZJxCJCURJwmVzkpn6devq80mTpOXLgyezJo0aBQcDuIp+7ZlnlPXHHwoupDeaqQXXzHHezLaWAwA0AQjsaDOiERKBKVfnzp1zBH+8mhGiAjkaED60iSDYtGlTQTYQ5tE2TJkyJZhnkR5zqgsvvDAQGQgEmha0FQhNECI0Lrh/RstiRInrY8aMCWQJ5wPkxxwMTQ0CG6ZifBwP07ennnoquJSmfUuXLg3ECDJw/vnn55zHXOzBBx8MC4NohjBZM20O7cAsDPJFmfyNQIvpGE4SICoXXHBB6D/njfRBtCB5u+22W9BGUQbth3xMnTo1xzwPTBCayWv9e+KJJ3TwwQeHD59CWq699tqgoaFf48aNC1+wx0MeGqx33nkn1B0vkIeTCf5DEOXjoWPHjtUrr7yiV199tUiRccDckPYTS3pgXDFdhOAyFrkFsMpv2tzK2ZrX6ANaz+Ia77feekvslSOUhvHemthvrbqLn8zsXTN8lyQj4yOtWzkxfDuGTeTJYta6yZvsz0jmAOC+e27Q77+9K+ljjXvzPjVpXC+8IE84/nC98vLQ4HWLsqJC+eYgM6F9aycH5wYZaycpc+2kTdofrd9/Fz9BKE2YOpkp+UJ9SSIYW7Itm53MTJ4cNDCrfvxR119xhSqUKxc8m/Ehz5U//JCjnSkpmhnc0GIOxzdkzjrrrEAIOnToEARotCcNGjQIAjzOaPbbb7+g+TABHuHGhGrTzFAWEUcBmJ5FBSBMtSAbCOaYdZnZFAIYkbSvvfaajjnmGF111VXB7AcBAVOhSy+9NJApSAlkK1GgDIRX2kRZkBlM0sqVKxf6haBKYP8O7aMdDz/8cCBFECbIBYQrGvib8iAzffv23WS/kPWd/TWNGzcOGpQ777wzELdoGfxGQ8C3SlJSUkLfOGdk5j//+Y8OOOCA8F7v1KlTIFPR/Ji3ofFBW3PXXXeFS/Q1t0D/icuXLw/EEC0PGiLKKGxEc4YpHjgYLrm1If4aeWyst9TR5gJtAY/c6o2m3RptjW+bza94HBP9bePN4gOe2NAuFmW8WYwgskDBnCaASbJA/TYnaLfH4sUgGe6cL3Yys9++u+vee26S9LUUe1+KfZR71Cwpa7q0IXvPS+5k5j1lbpiulcsnqPtNF6hq1UoqX66srunSXst/fzvsnYluXC9OMgNRCsQra4YU+1iKfSrFPpNisyW8f5lHsIjr49IkeHtbi594OZnJJjMmOBf0uCWF++2tLhuLeG9mxbFnJmhmpkxR+AbN779r3EsvqXnjxtmLTw0basyzz4YPbqKdKSlkBk0A5APBHhOq+Ii5FaZXuGK2PSS8QBFeCCZwGZmB9LDXg7RmsoaQRjpWePkAIQQKzUci4QgBH6Jx0kknBW0MedF6IIyj+UBbQj4jAlEhytrEkZiMzEBqbrzxxtBvTMZYyaYeC5TP3xyJtD0RmTFtzqOPPhrah5kYWhgjVJRn7UOD9OabbwazO4REsKBsApoZyAwmeRAbyBp10gYi+2jAFK1PIpfY1u5ER+qnbjQTaHqKGjFFTEYmE9Xv57YsAow3GigcVRR1rMnPvGH+eCi5CBQ7mdlj9xq66cbzNWfOi/p41gjN+WhkwvjJRyNF/OG7l7Rq+bvShmmBEORFZjasnaLMDbP09vgHdXKLo8ML8thjD9PoUYMVy/ow+9s2GwlFcZGZbBKTTcyWLh6naZMf0LNP3a6nRvbQhLfvDt9lUewDxTI/yHZx7IRmEw3Z9kqUnMy4ZqakkqTNTmamTs3+SOaiRVr+/ffhWzOVK1YMH+i86sIL9WdEO1MSzMzYkwGZQViGkMyePVtz5swJHwnk45CYFOFlCwKA9gJhKRqMzGD+xJ4ZTLr4SjoaASMD5CFiUsaeF0y10LSwwt+7d2/16dMnJ2LKdfzxx4fN8wj2aFFYGUa7wqZ60+YYEYi2xdpm9SUjM/SD9u68886BvGH6dcMNNwhSAhlhb42VQf+SkRlrA9qU6tWrhzL4SCfp7Rrtoyz+RlN17LHHhv5DYAw7IzM4DIDwkNbykIZxOPvss8O+GRwqECxvtP+5/bb+FMcxt3rir1l9mPaZM4ZBgwYFpww4ZvCYGAOcY2DiZSQbHAsSDPfiOOZWr7WLe2ru3Lk5zw3mrMeiY4AJLI4pWLxJFoqdzJQtu5Nq1qyuww47QHVq76fD6uQer+7UVh/OekKxDIhAtslZZtxHM1nlyjYze0+Z66Ypc/00Lfv9bd1x2yVBO1OpUgVdcfmZ+m3x+E20M8VBZrKJzEea9/MYDbnrarVpfYIaNDhJ9eufqfr126lBg5PVovlxuuOO/9Oc2U8FQhXVDm2vgrz3e/v9aGZJFeC9XX+Ryy1BZmK4eP711+DBbNyoUTr5xBMDYWhYr55eGTkyaGf4+GZJIjOXXXZZEEYQnhCULUYF66iAbi9WE6rffffdQGYwQ8KLmaU1YYp0kCW0EjvttFPQMkCk0NJEIx/xRBuEFgYhH3fP3bp1CyQpamJmQpS1I3q0OpORGdoC2UJgxOSqQoUKYb8KZlQQCojd888/H0gNacEkkWaGOrmG9zPKwDMa6QwzaxPt4RzaGNLi2po9SqbZYWM6mhm8soGjtZ88RITCc845p9BkhjLoh5VXHEfrW15H6zukFLKIc4NTTjklOFxgz1RhImUki4Upr7B5krUhej6+7Og1+x1Nwx4u4plnnhk84NkcAcf8hi053tRF2yDxN910U/Dehwc/j8WHAQ5SWMRJFoqdzEA8dtppR1WqVE4VK6KyZ9Xn77Fy5fKqVLGczm5/kmZOHxG0GvaBxjzJzNrJytrwvsaPG66WLRoG7czRDQ7VSy/cma2dWZu9d6aoZCYGucr6ULNmjVD79s11/HHnq127R3TVVZN0/fWzdf31c9T56unq0OFJndj4cp1xRjO98jJt+CDpt29cyC9+c66SiqlrZv4Snp1IlCwsthSZwQ1zbNEiLf36a/Xt3l2VKlQI3s0u6dAhaGdw67xh3jxlbWUHAKaZgcywUdyEXhOo44/xL1TSExDC2QeD5oUV+GigDNKhDYE8YMuP5y9MxtiQzj4Zi5wj/ve//w1ulnEYgEkYGh823pvJC2UmC9bmZGTGhD1MaNjQjscz9gsdeeSRYV8OZOPoo48OwjcaBdInIzMImwij5cuXV79+/YIWywQ8ax/t4RxkBo9mlI9GygRVIzPUCXbWfvIQi0pmrLziPFrf8nOkXvqKhg/HDR9++GE48tvjphiADRHTQjQdjH9BQ3GOs5WVrA20jzTcpzfffHPYe8f+O4/FgwF75SC33DvJQrGTmV1qpOiC81vpmacH6Mn/9NLTT/RJGJ95sm84P/Ht4Vqy8E0pY0a+zMzQzGStnazYhhlaumic+va+PHg2S0mppI6XnKY/lr6trA3TQ1lFITOxdVMUy5ip/333ss44o7FOPPF6XXXVBPXqtVTp6Rnq3z9LA/pnKT09U336LNc118xS69Z91KrlCXr37fsTOiQoqUK3t2vzECwnMyVLgHdC9dd4bCkyg0ttPr6ZuXix0M40P+GEoJ05sk4djXnyScWWLdO6n38usWQm2Ysz/nw8mcFhwMSJE4OAY2kRdhB68IoG2cGRAB+TRMDFfIJjNHIOsxW0Htj+Y3qGmVnr1q3D31aelW9HE6zsmIzMWH7ajrcyhEbM62g3DgEgHGiHcIRAOwmJyAz1ENEY4eIZ18kQFsrlvAWrD0cG9B8nAA888ECOCZGRGbArLjJDnUT6h3kfWiMz6eNY0IgpIGaBaJTwsGZ9sj7m50ieZMHam1cacGVu0C+O0WhzkTqsfdFyc/tNXhu33NLZNeaotSHZHGZOUyZ5aDfpIOPRNvObeU6aRO1Ohley85TPPC6O8WbM+a6SedCzNiaqm31UuFzHWQWRNngsOgZgyXOTuZYsFDuZ2Wfvmho0sKvWrf1Aa1ZO0Lo1k5LGtWsmaT1alvV/CZN57ZkJZGbdlLA3JXP9DI1/a7haNG8QtDNH1jtIr45m70y2yVpRyAxtWr9uprrd2E5163bQddfNUnr6aqWnZyktLabU1OzI70BsBqxXt25f6aSTblb7dk30+9J3FMuAVP3VNycN2xcWTmb+Ep6dSJQsLLYUmWHcITSxxYu18MsvlXbrrapYoYKqVqkitDMr5s5VBh/e/PNPPTF8uGpUqxae5VvaNXNUM4MpgwlfyV6c8edJTzDNDKZSCDFRwceEOQT9jh07BnMp88pF/vhoeTnyEh85cqQwP0NzMn78+FA2giICokXSxsfcyAxpLS9lIVByZG/Q008/HfbtoCl57rnnQv8SkRnyENgDgraJvT58WNT6Q/n2Gw3PSy+9FPbW0Be+N8M1wuYkMwiZCETXXHNN+HgpLravvvrqAkfy8WFP3EbzAVLDL3Qgj/8sLRigCWPfD9ownBkQb7/99kCUIK18yyd6nmvgi0MJ8MJzHEQMLQDmh5g2cUR7xzk+Tgq2eM6z8Y2fF4n+Jm2i88nOYTJHO6mfb9OAi0X+JqKpA3vKYB8W7sBpI5F8RNpN+yENmGEyZ6PtzgPanMvUQWC8wZi2MM5FGW/y8q0ltKcEqyP8keC/ZFj5+b8/mwqKSQK4c04VP5nZp6aG3HWdYrE5ysRMK3Na7jFjWtj8b4J+fsgMaULcMEOLFryltH5XqUqVCqpatbIuufhULf99QtDOrF05Ucc1qi328VSrVlHvvTtcGeTNyF2ohshkrJ+pLz97Rvvtu68uvPAtpaevUlpaZiAxaWlSNEJsuJaevk6dO0/XQQedoJdfTJVic9whwHbsDMHJTMkS4J1Q/TUeW5LMGKHJWLhQ/x09Wnx3hs32dQ46SOOee06xP/5Q1rJl2w2ZwW0s+2DwyoWHNDQG8aQEQY79LHgYw3MWL320I5iBoS1hsz3EAsHWBAITRDF1wQUygiwhGZmBhPz000+hHH4b4bAyEUDRAkHMcCVNSERmLD1mQWzsp30I45RN20woZfWdNGhwMDFr3759aCN9J2wOMhMK3rinBxz5yClCfmEj+YmsErOPybC3enI7GhaMK5ovHDmAL3tGMNHDcx17hpgXmNVwnWu2fwTzwpdffjkQTlb/IeBgffnllwenC5AChHe+P0R5zBXIBt648ttOiDzzhTHNLVAegTrY68UcoU72ARHpF5o9jji5wHySgAkn/cFRBPMAhxMQGT7oiYkn6dknwz4uSH9+2x0Kj/zHfKYvjFNhx5p8Nt7ce7Q9P8Ha7Meik5d4DHPDv/jJDN+ZGXytYlkfh2+x5BAPIyB/O25KLPJLZiA/kI7M9dP13/EPqGmTI8OKXp3a++qNMXeHvS5r/ny3wGTG6l+zZpb+9c/rVatWC/XuvTiQFbQwURIT/Q2hweSsV6/5Ov64Hrry8pOl2CdOZpzMaOmit9T6lGPDXoFHhwwJZjcm4LmA/ZeA7VhsOSy2BplBOzP/yy814LbbVKF8+RztzKpffglmZiPvv3+raWbQEuy4445BoCoOzUwiMzNexLycERQhKbhnZg8MHrr4kCPnWHFnE/GkSZM0ZMgQ3XLLLUHbAyFg8zjaGTYV80FNVvYhHKTnA5QQBbQeCIj33HNPOEedRmbYz8IHNe07M5AdNEMIkqNGjQr5KQshnT0caA0OOuigsFkfTQsBMsNKO+Zk5513XjhHn2gf2iNW1hHIaSOaA7RTECuEb1a2IWFobxo2bBg8lpEXsyOOkBkcJ+S1Z8Y+yEnleQndoYEb/7N2Grkq6pHyChrIA6nDOxOYM+6jR48OJAVNGH3HlBBtAmNp1yAxeHhjfBDUwZN0uLFmDoAxwj8en9DePPTQQ4FU8PFWPrzK2OTVXvD417/+FRxX8CFWQrI8dh5HDrgxh5DQH9pIxAOZRdpnxBZTRr5FxFx85plnQntpN+1n/oIBZo141oPQLFmyJLTB6isI3uQp6hhH8xemDQVpr6ctGgLFT2bQzEBmYrMDmTGNS36PRiZWrJihIYO6aMcdywSSEvVmRhrK45i1YYYW/PqmBqZfrcqVymdrZy5qo5V/Ttaale+p0TEF08yE+tdO1YoV76tzp1NUv34P9e+/PJ9kJkupqX/q7LOfV+PGtbP37mzHwnx+x3xbTeeamS0nnDsRKhjWW5rMGHnHpGzCq6/quAYNgnZm/7331oRRo5S1fr3+M2zYViMzCMiQGcy/zAFAQYQXE6qxq486AEhUBuew6WfV11a2Ed4wX8JkB1Oo0047LXhFw6MYpmsIVdQB0ULIw9SsRo0agWiQnhV5yAXn8UTGXgFICcHIDMIvxMnIDCY/5OMbOwjRrJRTFmSIsiAlkBkICk4CaAMbcO2jmXgWI9AfrnEEO0gM7YCYsDqPYI72ACEWbUKzZs3C93UQ6smHcE6AqJEnmTczXGWjzWGsIHkEwz38kcd/1k7qLI5IeYUNidqCxub0008PWivceifS1pGPPkNaIBFoRSAK1h+7zvwCT7BknCBBVmd8mzlPoFxMqu699948SYTlQYtCO9AAMZ+sHdEjaa1uyEzTpk2D1zsIO0Q2Pi1mZmh50OCYVzvSFDRYndHyi/Lb+lzQdnj6LYNAqSYz2YQG7cw0TXz3YZ3Y+IhAfPbfv5YmvD1ca9fMUMOjDymQmZmRmT+Xz9D5556gE08cogEDVigtLSPslYlqYzb9jWaGfTRrdPHF/1WD+gcrY/VEKSP7+znbqsDu/dpUsxjFw8lMwQRsJyRbDq+tRWbwXrbg6681uFevoJ2pXKmSOp57rtavXKmtoZlBuEEb0rZt27CSzbdcTHAviPBiwtZHH30UtDtXXnll8MSU6DVuQhbCPAI6GhaENwRPvB+htUCQQ2MyduzYoJGxPBxZrcYLGivY7E8hD2ZJ/EYY7t+/f9jIj0BLeoTkxx9/PJjwsJnZVt1ZrWfVHALFRzPJDwmxNkBE2KfBN3boH8IudfPdG9JDagjUYUd+Q3jQBmDuhGaAMukbQiy4oKGhbstjhIQN9RdddJGuuuoqgSNlWaR+BHLaD6FB2CcY7uGPUvQf/aLfRDPxwwwOEy3IwbBhw8JHHy2NHW0cIDO4wYaIMj4QH0sDQaBMNB2QUnA3UkCa+EBbCFyD3N5///35JjPMA9oxfPjwcN9QN22hfouUS7uJkJkmTZoEzQye+pgH1nbDAWcXOJFAu4emiPaV1nGOx9r/3nwIlH4ygzOAsHdmvIbeda0qViynKlUqquPFp2rZH5OKoJmZqauvbKkGDW7Ll2YGEzScA6SmrtA5HV5U4xNqK2v9NMU2apCiQq7/Tk4AtiVsnMxsOeHciVDBsN4aZCZoZxYsCBv+Z44fr4ZHHhm0MzV33VVfTJmi5x9+WLtUr77FHADYaxUBDIGQjzkiiCNcmYBnafI6Wno2d2M2g+DNF8hzCwho5KM+vIhhVsTeFPYX4PGLsghWtpVleSBhrGKTB/McfrNKT3+i+RASERAxTTKtk12nLCJmagiXzz77bE4bwIM2WBry0FbI3pdffpnwmw+WliNaJDRVOA+gb+xBQEvENas/emQvEeZGybBD8KX9CPII/tG84Y9S+p8J6uDKnhHIDBv4IaPxfQQ7yEGUzDDfOGfY85uIaSCaE/beTJkyJVznfHyw8eAamrnCkBnIJWNrhIQ+WbR28XeUzDA3mKvUSxpLz1xknw0e79hbZtfi2+1/OwJRBEo9mUH4xWMY2pkZ0x7XsY1qh5fh7rvX0OT3HlKzk47SzjuXzbcDgKCZWTdV7Jl54L4u2nPPVurde1HYD5O/PTML1LhxL11ycVNJvmdmWyInBe2Lk5mCCdhOSLYcXluNzOCqedEiLfr+ew3p0yf76/MVKuiWzp311PDhW4XM8EI0gSv6cizs74KURdpo/ZY3/hhtC9cQ/PLKF81jaePP5acca2M0b6Jzdt3abnXa3/FHSx89WproufjfpNmWgo1BUchMVLsBiYZIDhgwIGg3MPMzJxKGb/RI/fwNEYHMQKRMO2nXounBnr8JaALRzKDBQ5sGsYpGTB0hqaSnrHgyA0GFzFA3v2nnI488ErSEmD2ahs4w2pbG3ftSvAhsI2RmirLWT9fiRW/rnqHXBfJSqWJ5dby4TXAMUBAyg8AavJll4M3sKe2+e01dfvkUpaWtVVoampdETgA4xzdn1unaaz/Qfvsdr+ee7inFPhUfAi2oEOzptw3NjZOZLSecOxEqGNZbi8yYdiZzwQJ9PHGiDj/kkLBXZc/dd9fNnTsrpUqVLa6Z4ZWKQEUsqtBkQpsJgfl5XVsea4O1wwTGZGVQRzSP5UuUnrKSXU9Uv6VN1AZLnxtWlia+fYnKs/ZanmTY5XXdyiltR8OxoGQGDQ5ewdDkIfTzgUm8d2HGBynBXJA9V2hAwM6wjx8TiAQRLQnmXWhm0OTZ+Wh6a6uVxd4uyAxmhJicUR8RU0D2g7EH54033gh1U46RmZ133jmQFkzhaDtHNIOYEWJmydfe0RJC0pLNh9I2zt7ezYtACSczXXN1ABAv9ENCPv7oSR1xxP7BfAGTsxrVK2vHMmVUtWr+XTNTbub6KcrI+EidrjxZBx7YVj16fK3+/TcoNRX3zNnfmsnW1EBwMsO1O+74UU2b9lLTpodp5YpJG78zk+2sIL6t/ve2QVhyG0cnMwUTsJ2QbDm8NiEzNWsGArHDDjvotSee0JqFC5W18fsweY0J5QRNy6efBq0KLpfr162rz6dOVQxToAV/7xPfnclcuFBLf/xRQ3v3DmSGuveoWVNld9xxq5CZzfua9dIdgeQIGEEoCJnBzK9q1arh3sET39577x0inuJwkHDIIYeEPVcQnCiRoS6+wYInu0QRJxR8yBSvdYmuk5cQT2aOOeaYQF7Yx4XHPPbqENn3hVcz0huZgajwnMAZBF75aPtee+0V2s2RvVZ4wYPIROsKf/h/jkASBIqJzPxLrVoeE15C++xTU0PDd2aK5s1sJd7MBkfIzL036vff3hMfzcQULF6I5BzamaVLJ+reodeFm4UXpMWCfGeGshFE2YuzeNF41T9qH9Wte4VuueU73XVXpu68kw9lSunp0sCBsXDujjvmqlmzVNU9fH/NmPqIFPvYtTLbuSc3JzN/F2QTCbd+bsvjFCUzaEXsOfn6k08WmszsVqNGeO42OOKIXMkM4039eDb7avp07VOrVk79CDm05ZYuXTT388+ViRet+fMTkqJQDtcWLlTGvHnhuzV4+GKvAIKQCVxJ3n1+2hEoEQgUlszgHrts2bJhkzwfUWXPFBv9caudbG8SDhxw242TBczPcJVtkXOHH3548DqHV7/odX6j7WFvDBpglUdvAAAgAElEQVQbu7fMAQBOIXA0kSyQnn6imcEhAZoZXHxDdGg72iNcSrNnzfZ8UZbVk6xcP587AoY7RDKRpi163TRwpLPfpQn/YiAzs/TO2w+r+Un1w0tor71208D+ncNHMzPWTPob6YgnIfF/254VXDMPTO+Uo5m5e8h1+i0XMkM5bLbPXD9VX3z2nA444K8XNC/HsmV31Lv/vS9fH820NmXx0c/YB/ru29Fq07qBdtmlvk488S5dffVnuv32P9Wr1yrdcMMPOu20Edpnn1N03LFH6J23hysz8yMnMts5kWEOOZnZ8kK6E6P8YW5k5vOJE8P3j4zMvDJihFYvWlRwzcycOapUoUJ4B9Q5+GB9OnlyUs2MkRBMzZbNmxc8m1n9drypUyf9CJlZssTJTO7yil8t5QgUlszw4VHcbbNHJSqAUh5CqAmqBg9/4z4ZT3i33npr8GKHJz1+2xEX3XjSwz03Xu44b0c+hIqjCQRdE3KNzKDNQbMEEWEhISo4R9sCmeHjoLQbT304OTDB2drN0X5b2/2YPwRszMEUDIsjUKbNLxv34ii3uMsoEpmBeMQ2TNPiBW9q8oQHNer5/nrjtSH65ovnpIwZCTUoRhRyO7KhP2PtNH3zxbMa9cKAUO53Xz6v9aunhM3+uebdMFVrVk7SrGmPadQL/TX6xez8r42+U0vmj80zf3zZGWveU2bGDP322wQ992x/XXLxGTru2KY6rE4L7bdvY+2zT101bdJQgwd10bdfj9K6tTP8Q5lOZAKJdzKTP8HaCcjWw+nPb7/VmCef1KjHHw9x/uzZwQQMspPvcVm4UGt/+kljnngilPH2Cy9o+fffB41JbmVQB4RmyWefafTG+q0dn77zjlbPnRsIEWZpycoJ11wzU9xygZe3BREwobOgZmZGZvCEZ17BTIg1AhHfDerKy8zswQcfDO7K483McOkd72HNyAx5+Kgr5Vt/onWbkB0lM3i4oy2J2h7N67/zRsDwjZINfuON8MuvvtS48eP0r0f/pbT0NHXv3j3sZeKIRoxvUmFeyIdPcbf+4IMP6bXXXtfs2XOCi+742iE2icY4Pt2W/rtIZMYEf7Qha9dM1aqV07R69TRtWF/0/RBoWTasnxbKpFx+c87qzO0IGVq/bqpWrZqmVas2tmvVtKC1yS1f/LWgmcn6UMqcqe++elYvPtdLfXqeqysua66unU/TQ8Nv1LQpj+nHH8fq998nKGOj84D4cvzv/I3btoaTk5kCCMQFEZ49bVLhPpnQn+w8e2NWL1igVfPnhwi5SJY2t/MQEyvH9tzklj56jTassjZsPK7faIoWTZfot5OZLS0yeH3FjYAJhpAZvu1TqVKlsAk/njhQLwIqhIU9M5iZlS9fPnyr6P/ZOwu4Ks/2j78WIQJiOxsDscUuEDuxFQxMdGInSpzgEOb2rud6Lv66ehc6F24TAXVT5zp1tmJ3wInv/3Pfx0fZBgqSwg2f+/M8nPPkdc7hXN/7d4VwMMVz6Z3ZzK5TU0K0pTbrLqBCKwAgrkU8rj2nbatdq3YeATOiEMGTTz55G2a059KfXzwm9hUwI5qmiut+++23ZQWz7Fx7+mOqdfv7QdhVs7moCCdKqos+TzNDZ+Lv70/DRg2pVbsWIp9K9CUSRSNEvpXIrRL9jNq2bSv/Fo9XqFARkYMlcplE3pXoEzVx4iTZR0gUlxC9sbRzifNq74fC8FrkCswIyBA9VbSRVei4l/Oa/rjZPab1H9ek9Xy51zm154UjarXsZu/uVwh9eCCdO3WkXdshdO48jY6dJtGmTX+6+3YiYvl4vtv3GjbrN1iFUpVBPo92TLUsXlBTUDBjPXaM+xnCMbzbLHhGzqR67P6c/8JkN+uthH+Z9H+fMCPuR+wvRyaJ/5ndswCh2/veWs+qMqRgpjC4EeoacmIBzSEUACGap4oQLOFk3gtmBBCUKlXqbzCTlesQzmj6oTnDAiqyW5q5b9++sppZemVGc3bTX4t4TJxHwIxominuUSgzwjkWoJTRPun3V+t/t4BmT81uov+SqP4WMj1EAkidOnXwqOCBs6szLrVcqNW1Fs2CmtFpQSe6RXTDqaKTfH7t2rWyqW25ctVp2XICXbosoWXLidSt25Py5Rvi6OiGu7u7hBtRsS4wMFCCq+gFpEGo9v75+xXm/1+5AjPCSRdqiAw7y2Vn/n6Pmf56snsMURXt5vUd/O/dOLp1bU23bjMZPfotQkJ2MHfuD3LMmLGLkSPX06XLREaO7M/mj1Zhs+yUIWbi3ApclA0KBGZSUrCdPXvfg5QUuEtYT2YOqXr8wYUaDWJzCrI5OU76fbNzHXJbFWaW/56DOmOuWUCDGZGcL8oaV65cWSbai1l28aM5rNq6cP5Fk1cxg16hQgXZYFVzLO/norTji+MKmMlO08zhw4fLqmTr1q2T+TiZObaa8y363wgAEqqAKCGtYCb7r5j2fhE2PXDgAML24nXwbuKNKIDiVt0Nz16edF3SlSEvDCH4s2BCvwllwc8LCDsYxqLDi3Cr70bFShUlNDdr3pxKlZoRFPR/LFt2mIULf2PWrL1MnvwFw4e/hr+/AW/vEbi51ZFqoKenp2zuGh8fLxvxpn+fau+l7N9VzvfINZgpCs67gB4BMjeuJZLw1dO0a+tNjx7RzJ37HTrddaKjrZhMYlgwGq1ERFwkJCQZP7+FDBvSh8SEp7FZvkGEpxUFe6h7yBmQ5TfMiFK4qYcPc2TfPv7au5cD2Rhi+6Pffce1v/6SuQ5ZnRlXEPPgQkxReO0UzOTcCVBHKFgLaA6gcApFRTIxw/7LL79IxSKjKxPbi8phb775pkzIP3/+/N+AJ6N97vaYdn4BM6GhoVmCGe14onra66+/LuEqfQiS9nz6pTiPaOgpqpeJfYRKo1SZ9Ba697oGiyIXRihbQilp3LgxZV3LUq1VNTrM7sDwV4YzY8cMlhxcQuSlSAwWA9HWaKIt0ZgsJqKuR+HWyI0KlSqw9pFHaNasGZUrt2TSpE+IiTFjNIoh+iZa0OmuERZ2jNmz9zJ69Lv4+kZQp053ypRxok6dugwaNEjClKhCJ65Nu75730nub6FgJl0ejoAZkf9z+NCHjBrpR4sW04iIOI7JlIbRaJMNM0WPGa3PjHjBxXMzZ+6la5eHCZk2iHNnPlf9ZdLZtDgDUX7CjIAPS0oKfyYnM27YMKYOH8607IwRI5gaFMS2d9/Fev78XStIFQUnWN1D0YAwBTO57xSoIxacBYTCojWKvNtVCDAQ24k8Fw1G7rb93Z7T9hdgIRLBH3300dulnbXnMts/q9er7S+Ol919tH2L81LYTYCCeM1FvpSoLNe+fXuZr/RQu4fotLATQe8EsfDPheiu64i2RUuAMdqMGGwG9DY9eoseg9VAxNUIXBu6/g1mKlVqQXDwJxJg7L0URSN4m/R7xSS+GOLxsLDjTJnyBf7+egk1zs7usjmrqH63bds2RJGIggIaBTO3HG+7KpPEtSsJfPTBKqpU8WLWrJ+JjRVNMi3yhTUYIP3Q68ULLhpo3iAo6H+0b9+DLZtXge1bVZpZAU2+lmYWOQepp0+T9OGHVHJywujry+qePVnVsyere/TIdKzq0UNuE9m9O21q1OCZ1auxXL6MWeTP5CB/QsFC0YCFwv46Kpgpzi5e0bz3ewGEdtdiu6xuq+1zt6U41ssvv8ynn34qq2CJbbNy/Kxsk/682d0+/b7FcV3YSwyham3ZsoWxY8fK8EL3uu60GN+CMRvHsPjwYgypBgkxAmAkvFj1EmQM3Pq1GRDPRV6LzBBmJk78VEKLvRm85uuKSXxtAh/5vACb8PAziO07dpxLpUreVKxYTao069evl5XwCgJoFMykgxlLajInT35OZHgwdesGsWqleCEFyGgv7L+X4oUWYWfz5v1Mly6zWb5stIQZFWqWsxCtoqDo5Kcyo8HM9vffp5GrK2fDwrAYDPah12PJZJj1etIMBg4tW0agtzePr1yJ5coVBTMK5O6rqll+w4+CmeLo3hXte86qs685ublpDVGE4MKFC3J2PSvHvZ9ruJ99snItRXEbzVbiNdm4caMsEFGyVElqtK9BnzV9mPvLXIw3jBJiNAVGLDP8vS+Y+bvPq4FNdLQNk8nG4sWHCAh4jgYNBlCuXBXatGkj83D++uuvfFdoFMykh5mbyRw4sInA0X3o1Hk1a9YImDHfA2ZsklaXLUthwID/Mma0LzbLjiyXkS4KTru6h4zBraBgxsvVlfPLloHRiM1olEuxntGwCeCJjuZoeDhBTZrwmIKZB8KJz29oKKznUzBTFF04dU8FZQHhPKufwmEB8VoIhePMmTO8+uqrtGvXjjLOZWjQryFBbwfJcDKTzSRDx4QSkyHApH80F2BGm9gXUUn2iXwRnZQqiwW0bj0ZN7eH8PLywmQyISqeiesXIz9+FMz8A2b+/OMDhgT407Pn86xZwz1hRkhyglLDw88zdOiLBAzuZA8vMiepMs3FPNSsoGBGKDPnwsIkyAhYuduw6vWYjUaOLF9OkLe3ghmlyDxQMKdgJj/cBHWO4mIBTQkoLvdbWO9TAxnRiFSATPMWzXF0d6TJsCbM3BNKjCVWhozpRShZVn9zEWbuQI1Is7CrNHPn/kiXLgspX7429erVQ6fTIUpGi1ys/AAaBTN/g5kd/HVgE2MD+9Kx48psKTNhYSfp338tgWP8sJl3ylLVSrHIWLEoLnYpSJg5HxZmV2a0/zqZLG0i/Mxo5KiCmQfKiS+sSkl+X5eCmcLqjqnrUhZQFrgfC2ggI0LLNmzYgI+PD07lnWgyugnzfp1HnC3OnsyfWThZZnCTBzAj3ApNpYmJsbJo0X78/CJwd6+JKOEcFxeHKDmeH1XrFMykh5nUZFJOfo4+chJ16oxkxQoRYpZx8r/mG2o5M3Pm/EDnztOJigwC23eqAEAxV2UEsCmYUUnw+e3cF7fzKZi5H3dJ7aMsoCxQWC0gVAzRMHXTpk34+vri6OqI91Bv5v8xX4KMzqzLDFfu/ngewYzmC+t0Ftm6ZMmSgxJoXFwq06hRI5599ll5P3mtziiYSQcztrQkblzfztZPH6VypVrMmvUbMTECZjIDmjvVzALHvEsbn85s++K/2KzfYlG9Zop9rx0FMwpmihtc5Pf9KpgprC6Zui5lAWWB7FpAOPxCxdizZ4+sWlbasTQN+jQgdF8osbZYdJb7BBmBOXkMM3aVxl4Qa+HCP+jUaQGOjuVo0aIFCQkJeZ4/o2AmnYKglWdOObmV8UHdqVplIDEx52/X2LZXctBK1YmSzWZiY9MIDd2FT+vxTJvaB/NNlfxfXMLI7nWfCmYUzOS3c1/czqdgJrvuktpeWUBZoDBaQMtXSklJISwsDJdyLtTsVJPxm8cTb4vnvhUZTa/JB5hJDzSi0aa390icncvSs2dPRP6PgDVxn3nxo2AmHcwI59SWmog5NZn9f75LvTrueHoGExFxltWrbcTHQ0yMfYj1Vatg5sw9eHmNoXevdhz44x2wfo31RkKxVyXu5egXh+cVzCiYKW5wkd/3q2AmL9wCdUxlAWWB/LaA5ui/8soreHt7U96zPH0f7YvJarq/HBkNYrRlPsGMKAggJv4NBrOscubh0ZhKlSoRHR0tTSqUp7z4UTDzD5ixA812zOYd/PbzBnr4t8DDoxVdu8YzZcpuFiw4wbx5xxg/fjs+PsupUaMNwRP68+MPb2Ix78KaqkCmOIBKVu5RwYyCmfx27ovb+RTM5IVboI6pLKAskJ8W0EDmxx9/ZMyYMZRxKkOrSa1YmrJU9pDJVtUyDV7+ucw3mLEXBTAarUREnGPgwKcoXbq0LAjw008/5Vm4mYKZDGBGOKrW1O1YzMkcP76Zdc8sYeiQ7jRr1hrPeu2oV9eHli1aM25sf15/Tc/xo5tIu5msSjFnYsusOP5FcRsFMwpmihtc5Pf9KpjJT5dLnUtZQFkgLyyg5cqsWrWKmjVrUtu3NuM3jSfGEmNXZf4JJvfzdz7CjBZuJtSZefN+wstrGC4uLoSETM+zUs0KZu7igIuQM8xJXDi/jQMHPmDvNy+R+NVTcuz79lUOHfqISxe3YU0TPWWKdxnioggjOb2ngoKZhq6unFm6FIvoISNKL+t0mQ6zTkeqXs+hZcsIVH1mikV5Zuvx41iOHcv2sB47hhgSIApJPx4FM3nhWqljKgsoC+SXBTRV5rvvvmPIkCGUrVSWbpHdCL8QjtFqlIn7We4lczfIyWeYEeFmBoMVne4qgYHvU6ZMWR566CG+//57zGZzrveeUTBzF5ghNRGbVGmSsKYlY0lLJC11O+bU7RJgrKnJWFOT5DY5dXzV/kUPBgsKZpp6eHBer8e2ciW2FSvuOawrV3LUaGRi8+Y8vmoVlitXMB8/jq2QOKz5Pdtf1M9nO3MG28WL9zfOnYOUFDh+vFCAn4KZ/HK51HmUBZQF8sICWuL/o48+Sv369anXsx4TP51oz5XJTlPMu4GMeC7fYQbZUFNUA168+C+8vIbi4ODAkiVLuXr1aq73nlEwczeYufWcqHImhygQYE6SQ8CH9rgCkaIHIll5TYUap70HxFKEJgqAsdxIkMN8fZsMVTx1YhO9erbF1cWFZ1eswHz0qH1W/OhROdMtZ7wFPPxjZMeptp44Qerp02x//308nJzQd+nCiu7dWeHnd+/h789yX1+aVavGEytWKJgpghAnwNSWksK1Q4f4essW/rtiBStjY1mRxSG2XRMXx//Wryflxx/h1KlCodAomMkL90odU1lAWSA/LKCBzMmTJwkMCqRs+bJ0WdaFsFNhMldGAEiu/RYYzFjR668wcuRGSpYsSaNGXhw6dEjBTFacTLVN8YSL3Hrd/wYoNwWgJNwat6BWwK0AXQG2lmRslh3YLDuxWXZhs3yN1ZxuWL/l0vmt9O3TAfdy5Xj+kUewihKF585hO3sWOUsuZsrFOH0a26lT9pnvkyflzLd0Qo8fvwM9t0J9RKiQdORuOd5iO0tKCr8nJjI6IICHhw0jdPjwrI0RI3h45EiCR41i68aNWC9cwFJIZt6zA3Rq28xzlCQonz5Nyq+/slavp0uTJoT6+RHq65u14e/PYB8fJo8YwdeffYbt/Hn5nixomyuYyQ+XS51DWUBZIC8sICp7CaDZtHkTbdu2pXLzyozeOBqjORfDyzQcKhCYuaPOLFh4gIoVm0h1ZuPGt7h27VqulmpWykwWlJnccpLVcQofZGngchtYUreDOREsAlR2YDPvxCqq1N0CFAEtWJIlyJhvJHDl4lYunPuMM6e2cPjg+/z+69t8//0bfP/dG+zZvZ49e94gYdvTdOrYDBdnZ/RLlvBDYhI/JiXzy46d/PHNbvbv2cvR73/g1M+/cPHP/Vw7eJC0I0ewnTwp4UZAjgQgAUEaAAnwEeE+YptbsCMA5Or+/Xzz5Zckbd1KohhffJGlkfTFF+xJSODsb7/J46oQs8zBoKAd+Ps5v4Bf6+nTHPnxR4wLFjDUy4vEGTNInDYta2PmTHT+/ozv1YvPP/xQhqiJvJv7uZbc3EfBTF64WOqYygLKAvlhAZEvI0aULkrmk7QMbsmc7+fkfohZgYWZ3YEZne4inTotoUSJkkyfPoPTwq/Jxb4zCmYUzBSbnjgyFEwqLXaFBVG4QSgr5h1Y03ZhTRPgsgNL6nZuXvuK0ykfc+Twh/z4wwaSkl5iy5bH2bAhhpdfjuC5dUtZu3oWUeETWbp4LAvnB/LwjKFMmtifoMBeBAb2YuSI7owc6c/gwV2oWtWD0qVK0b5Va8YOG8644SMIHjmKKYFBTBs7jtBJk5g/LYSwWbMxLFzI6qhInlm5gleefJKNL7zIpg0bSfh4C98lJvHXt/s49etvXPnrIGbhUKakSEfVeuYMnD0LFy7A+fNYzp3DfPasHJazZ+XfVjGjntE4d86uFgllqAiGWRX3e9Jg5vBPPxG/eDFz27fHumoV1ri4ew5LbCyW1avZEBhI6MCBCmbyw8tR51AWUBYo0hbQmkdeuHCBESNGUNatLD3jexJ+Pjz3Q8wKAcxER99kwoTPKFmyNC1btmT//v0KZpTCUfgUjsL4mmjwItUXs8h3SsaaJsBFKC7JmG9u5/KlLzh+bDM//fwWSUkv8tFHa3jphWU89shcIsMnsmhBoASUgICudOvWgiZN6lCnTlWqP1QBd/eylC5dkhKlS1DKpRQO7g44VXLCubITrjVdca3lSoVGFajoVQGHcg785z8lKFeuOpUqeVOhQkPKl/fEza0O5Vxq4FK2Oo6OFShVqiwOpR1lfk2VihWpV6s2TRt50alNWwb07MX4ESOZM2UqkfPnsyoyihfWruW9l1/miw8/4rukZA5+/z0Xfv+D1MOHMR8/gfnkSdJOnsQswtBEFaujRzEfOSKHWBdDq1JV2CpVFXcAyc37Tw8zcYsWMat1aywmk6x4Z9XrudsQFe/MMTG8MWIEM/r3VzBTpF0sdXPKAsoC+WEBrXnkvn376NixI2513Ri1YRQGiz1ZP9dyZQo8zMyuzoi+M4sWHaVs2aq4urqxbds2bt68KcPsNLDLid2VMqOUmSKjzGghY3Z4uVWBLlXAS7KsQHfp4hccPrSJffve5Ksvn+atjSaeeGwBushJTJ06iH79OtC8RT0qVCyHQ+lSlChTAkcPR8pVd8GjgQfVWlalRoca1PWrQ4O+9Wk4oCGNBnvRJLAprUNb035Bezot60SX5Z3pEdNDzrIMenIQA54aQJVmVShZ0pGWLYMZMuQ5Bg16kn79HqV3r1X4dzfh52ukY8cltGo1gyZNgmjceCiNGg2kQYO+1KvnT61anahWrSUVKjSgnEt1HBzcKVPaCQ83NxrWq4dvx06MGz6SxaGhrNXpee3JJ/jkrbf5+osv+fPbfZz5809uCnA5eRKrUHJSUm6Hk8l8in/k4OSmI62OVfAha/+Emdk+PhATg000BLjHEKBji43l/0aO5GEFMzn5vlX7KgsoCygLSAuI8sTCid+wcQONGzfGs7cnIYkheRNiVqDKjPYVIwoBXKdu3T6ULOnAc889z8WLF3NNnVEwo2DmgYWZv8GLKKGdloQlVUBMEqk3t3P2zFb++msT3377Op98/AgvPBfG8mXBMvTLx6chlSu7U6pMSUq7lMalSlk8PMtTtUVVagpg6V6XBoMa4DPTh24R3Rjw+ADGbBzNxE+DmfHNDOb9MY9Fhxey5MQSll9ajujQa0z3a7KZpFQcb4sn8koUDfo1pHTp8gQEvEBsrJW4OBsxMfZhMkF0NBiNonOuhYiIKyxbdpbFi48yf96fhM78limTv2DMmLcYOPAJfLtF0qrVdBo2HErdut2pWbMDVao0p3z5epQtW4VSpZxxLeeCl2d9+vr5M3vKNB4xRvO/l15i5yef8NuevZz84w+uHD6C5cQJCTiieICWJ6PBjYKQgoeQ3HoNFMwoD0pZQFlAWaDwWEBL/l+5ciW1atXCZ5oP83+eX0RhxoZebyY29iadOy+lTBlnwsPDSUlJybWqZgpmFMw8cDAjIUaUxk5LxHIzEWtqIpbURC5f2sbx45/y8y8bJbw8/tgCZs0aQffurahWtTxlHEtRplwZylVzwaO+h1RLanSqgdcoL7os6cygpwcS/MkEQveFsvTMUiKuRMimVTG2GOJscbdHrC0W8Zj4NWEimmiJMellYb1Nj86iw2gxsiRlKZ69PCldyp0BA55Fp0uT0CLARa8XsxXigy4aTNmBRkCNgBsBOQJ4YmNtEn7i4+3LmBixnZXIyOssWXKS0NDvGT/+YwYOfJKOHRfToMEwHnqoLVWqNMXDwxMXlyqULl0Wdzd3WjVtxtjhI4gJW87Gdc+x65NP+Ov7Hzh78BA3b1Uwswj1RpTyFUMpNkUih0jBTOFxYtSVKAsoCygLyOR3bMybN49KlSrRLbwbS48ulZOgwn9I70/kynqBVTOzh5kJfycuLo1Bg9bh4FCOcePG52qJZgUzCmYeGJixNzDdjkX0cklL5ObNRC5c2Maxo5v5bu/LvPGajoULx9C5azOcnMpQyqEkDm4OuFRxwb2WO1VaVcZ7VGP8Td0JfC+Q0O9msvTUEvRmPbHEIlQUAS1iXQCK+BV13oXqorfY4UQAiliXj4nHxRD/eDKoBy+eE8cIO7OM+r3rS5gZOOA52RVXAIu9Q64mwWpLO9jYIUeAjhgWdLo7Qzwm9hXHEMATGwsCdMQQ62L7pUtPMnPmt1LN8fOLokGDACpVaoabW03Klq0s/5k4OjpTr3YtxgwdxqMmE1+++y6Hvv2W0wcPcVXk2NyCGpFro0GNpt7klmKgjpM/6o+CGeU8KQsoCygLFA4LiPAyMaw2K0FBQbi6utJ3bV/Cz0cUaZiJjbUwfsKnODm54+/vzx9//KHCzApjwrm6ptwvTCBUGNmIUlQhS0siLXUnV64ky6T97V89ycoVDzN4cCeqV/egRMkSlHYqjaObI2UrlqV6x6r4PNyawesGM23HNJaeWkp0WjTxt34FtAhlRQMWnVmHGLdhJYezI/cHMxrUZGWpgY8AHbMEHgE6GuTExcGKFQJ0wGhIY/Hi40yevI0BAx6nRYtJVK7cCmfnCjg4uFKqlCNlyjjg06wFc6ZNZ8Nzz/PHrl2cP3KEG6KEosizEQUERJ+bW6qNApH8AZHcsLOCmcLhxKirUBZQFlAW0BLeb9y4waBBg3B2cWbgM4OIuhqF0Wa0T5DmtjZTwMqMwWAhJsZCSMg3ODmVp02bNvz6668S6nLjHaGUGaXMFEplRvR9sVzfhk30e7HuxmLZw4XzX8ieLRGRk2jXzgtnZwcJMCVLlaSUUynK1S5H0/FNCXghgFm/zpJgIsBlJSuJI86utFgNMvxLKCxSZbFmrKrkhqyb9zCTGfAIyNEUHTvkCCVHqDga4AgFJzz8CiEhu+nf/3EaNRqKg0NlWc0QK40AACAASURBVDaxRIlSlChRgqqVKxM0fASvPf0MB3fv4eap01gvXpQNPiXYHD36t8adueF0q2PkDSApmMmNr0t1DGUBZQFlgZxbQIMZkQDfq1cvnFydGPLyEHQ3dEUcZqyEhv6Es7MHTZs25eeff865MW8dQcGMgplCAzNChbHc2IbtZiLwDbCP61e+IjHhaSKjBMA0kuWQy5QpTcmSJWWlsdo9auEf68+kbZOk8mK6aSLOHEeMNUbOawigkGpLHkJLZuBTcDCTEeSInJw7gGNXcEQBAiuxsWaMxussWnSc8eM+pmvX5VSu3JrSpRxlbxxHBwdq16ghe+O89tTTHNyzF5voZ3PpkmzqqZV4FqFoCkbyBkZyalcFM7n2nakOpCygLKAskCMLaDAjesyIcCvHco5FHGaET2LFZLIyd+4BGRGiYEbBR6GBj9wIq7OlJsowMmtqAlh3gG0vFstOfvvl/1izZi5duzWnevWKuLqWpUyZUrKPS/1+nvRd04eQnSEsP7Uc4xUj0anRMllf1GgXOTAGax7Vas+G9Fu4YCZjwNEUHPGPRhQVEI2tjMZLLF2awsSJW+neXU+NGp1xcnTH2ckJD3d3mnp5MT14Ih+9/gZnf/8dLl6UTTqFw205ckSpNffRdFSAYPp+P9ldvxdIZgQz1qz2mdHrsag+MzlyXtTOygLKAsoCmgU0mLly5Qp9+vTBycWJgBcDirAyI/wPEWZmZebMH6Qy06xZM6XM5IYTrY6R+/kt2bGpPRcmAauAU9s3WNK+5uTxj3hrYxzBE/vToH5NKlf2wMGxDM5VnGg0pCF91/ZhcsJkFh9cjO68DkOqwQ4tosmUBi8iET+DZPzM1JO8fLzww4wGOEK10YZF/tOxFxi4QVTUWRYuPMD48Vsk2NSq5SdnVdxcXaVa49upE7ERUez5fKtszCnUGk6dwnqrGWdOFYVis/+pU3YgPH8++8tz5yAl5a6qWHqYiV+8mLnt2tkTqkRpvHsMm8mEbeVKNo4ZQ+iAAapppuaRqKWygLKAssB9WECDmdS0VIYMGWLPmXl6oGzjUNRzZqZN2yVzZtq1a3c7Z0azx32Y8vYuKsxMKT35qvRoEGMzJ2OzfsPVy9vZ/fXLxMQ8TM+ePjRoUBN3dxf+U+I/1Oz0EL6RvozfMp4Fvy4g4kwE+lS76iJKHhdGgEkPRw8OzGhQI5Ya1IhqaWIIuBE5ONcJDz/JnDk/MHr0Rjq0n03lyk1xcHCgetVq+DRvQfDoMWx44UVO//Ir1nPnZCiadKKPHVNqTSZqjagOl3byJN9v3cqckBBmT59O6LRpWR6zQ0KIWLiQAzt3grB5JqF+6WHGtGABE7y9ObxoEYfmz7/nODhvHocXL+bxvn2Z3LMnn3/wAbaLF2UxiIKGTXm/J09iPnoUL09PPDw8SEhIIC0tLdcSS29/W6oVZQFlAWWBXLKAcODFGD9+PG5ubvRe1Yfwc+FFvJqZmXHjNuHo6E7v3r35888/VZ+Z7KgAatuCVWGE/UVCvwAZm3knNuvXnD39KZ9+8hiLFo6lU6dm1KpVBScXB9zru9NyYksGPzuYh3fOIOxoGPqbetmvxWg1yuQ4qbwUIgUmPcCkX38wYcYONnfKQWtV0swYDKI/ThqRkZelWjNx4mf06hVPo0ZDKFu2KkKtaerVmMG9e/P06rX8tnMXN0+exCacbLFUUPM39USGlp04wbWUFN567jnqVahA3IABrOzXT45V/fqR2VjZt6/cZpm/P+0bNmTr22/LHCZRbS4jwNBg5sQvvxCzZAkNK1ZkSJMmDPH2vucIaNyYoc2a0aFWLUb27s2OLVuwnT+vYCaXnBp1GGUBZYHiZwHRZ0b8LFy4kMqVK9M1rCtLDy8t0k0zRZ+Z/v2fkK0hJk2azGHRuNtiyZWJJ6XMKGUmT5WZOxCzS4LMoYPv838bYpg5czgdOzajShUPHN0dqe1bG79IX8Z+MJb5P80j8lKkzIGJtkY/UABTFGBGlHYWzTrjY23Ex90asVbiYi3Ex1pZEQ8r4yE+Lg2DPoWF83czNmgD3X3DqV2riyzx3Kh+A/r37El8lI6dn37K5YOHZBU0W0qKgppbKo0GM1dSUlj/xBO0cnPjtyVLOLR4cdbGkiVsmzGDHp6efPjGG3DlSqaAIRugpqRw/dAh9n7yCS+uWcNLa9fykljeY9zedu1atrz+Oqd++kmGEmamAmUEU3n1mFJmip8TqO5YWaAoWEBz4tc+spY6derQalIr5v04rwjDjMiZuUn79vMoXdoZnU7PqVOnFMwotaXg1Za7vQaiN8xtJcaczP7977F+vZ6QaQH4tPHCvXw5ylVzodGgRvSM7cnkzyex9NhSRAK/yWq6DTCZNaRMDw2Fdf1BVGYEyAhVZvbsn5g6fReTp+1kUkZjqnh8B1NCkpkyPZmJU5MZGfgR3f3jaeI9itq1u1KmjAt1atakX48eGJaGkfDhR1w4cEBBTQYw89pTT9GlYkWuxMRgMZkQyfn3GiIp//fFixno5cWHb755V5jRYEJAjfnYMW4cPnxfI/XIEdlvSDteQS8VzBQFt07dg7JA8bOABjPvvfceTZo0oU73Okz9aipiAlf4Drnu1xR4nxkbOt1Vatb0o1QpB15++WUuXbqkmmbezZFWzxUc6EiA0cLJLDv468D7vPmGkRkzhtK8RX1c3JzxaOBB01FN6Le6LyGJISw/u1zGiUbbom83sCwsSfw5+YfyoMJMePhVWreehm/LEfRrE3hrjKFfmwyGz2j6+oymX9tABnacSMemI2jbdgETJryDn284DRsOkCFo1apUo59/D2IjIknYvJlz+/djO31alnYWDmlhmOXPb8dc3LNoQCqVmSefpKOHB1cMBix6PVa9HttdhnjeYjDw+/z59GvYMMswI+9RhPzd5+DkyQzD2PLbdtr5FMwUPydQ3bGyQFGwgBZm9tPPP9G1a1dca7oy4vURGMx5VMSogGFGVEudP/8vnJwq4eFRgaSkZFJTU2WImSoAoELE8jRELDtQeAdikrGkJnHwrw/43/9WMnfOKJo398TZxZGKXhVpFtiUgU8MJHRvKFGXozDZTLcT3qQKk/vzEQV2xAcNZvR6G9HRNsLCLuLuVpel7VvxXEB/XhoygJfEMrMxuD8vBvRj3dBBjKxXk1r1hmA0nCU8/BwhIQn06BFNw4YDKVeuBtWqVGdgr96sMhjZ+elnnD9wAOupU9K5FuWINSe1OCwzgplrRiNWgyF9RYYM120GA1ajkT8WLMg+zJw4IeFRnD+7o7C9LgpmioJbp+5BWaD4WUArACDKMwcFBeHi7oK/yf/25G6u+0MFBjP2HnfR0TcIDHyfkiVL0b59ew4ePJhrqox496icGQVEOQIi0SdGgkxaEmk3kjhx4mM+/vi/LJgfKCGmjGNpKjSsgNdQLwY9OYi5v8zFcMMgIUaWICyAZpY5UVuys++DCDOmWzBTwaMJXwZP5EZMLNa4eGwxcdhiMxkxcVhj47i6YhVr/TtTvW4AUVFnMEVbZV35qKirTJ++A39/I56evXEt9xBVKlVhSL/+PB4Xz7cJ27kmwpcE1NxytAub05wX11OQMJMX91MQx1QwU/ycQHXHygJFxQKaOhMXF0ft2rVpGthMTvSKUPtcDzUrYJjR6S7Qtm0oJUqUZN68+Zw9ezbX8mXE+0HBjIKZ+4YZuxqThPlmEufPfU5y8kuYoqfj49OIEiX+Q/m67tTvU58B/+3P3N/mYrxplBAjQsjkB7WQ9IPJDqBkZ9sHGWbKl/dic1AQV3U6zHoDFp1Ohj6J8KZ/DvFcml7PlWgT8b7tqV4nAJ3uLAaDWebfiBLPoqCAXn9NQo0IPxM5NeXKVaeChweBQ4fxf+ueY//332MRoVeiSEAxgBoFMydyrMQpmCkqbp26D2WB4mcBLW/miy++oFOnTjJ6ZeRrIzGm3Wo9kZtxJQUIM0ajmXnzfsfdvT5OTk589NEmrl+/rpSZ7IQ/qW1zP39GQIw1NVGOq5e+4qef3ua5dcvo1astJUuXwKWqCzU71JQ5MQv2L8CUZrJDjPUWxOTmB7QQH+tBh5ktY4O4LvI2jEZEWBOZDPGcyN+4FhPLCt8Ot2FGxMjae9QIkLHKvjV2qBFKzS66dVtG1aotKFu2IpUrVGTOtGls/+gjUv78UzbgtIq8jkz6phSECpDb51Qwo2Cm+Llv6o6VBZQFNAtooWZCpQgODsbF1YWOCzqy5PiS2+H32ZlAveu2BQIzoledFZ3uEkOGvEzJkiVp0aIFJ06cyFWQEfZUyoxSZrKuzIiQMtEzJjWRm9cTOX78Y97/3ypGj+5JOTcnHN0cqNq8Kt3CuzH/z/nEWeMwYZdLc10yLcQQo/1DUTBjh5n0DCSgxmAQ4WcijvYKU6Yk4OMzjfLl6+Dg4IJnrTrER0by044dXD5yxA41RbRAgIIZBTOaU6OWygLKAsXTAiLUTEDNs88+S+PGjanZuSbj3h+HyZLLoWYFBDMiQmPhwj+oV6+XVGVMJhPXrl3L1RAz8c5RMKNgJsswI8otW9OSuHolka93vcrcuaN5qGZFSjmWkiFlLSe1YMbOGcQSSwwx6C363I/7fAAgRsGMPcxMU2bSw4x93a7U2KEGdLrLjBv3AZ6e/XB2rkCJEqXo1KYdrz/9NEd/+400MYtTBIFGwYyCmeLpvqm7VhZQFtAsoMHM/v37GT16NA4uDnRY0IFlZ5fJFhW5NhGc7zAjVBkbUVGXGD78DUqXLi376Rw6dEiCjJYvpNkhp0sFMwpm7gkzMqzsZgLmtF2cPfM5zz+3HJ/WjShRsgQO7g549qnHuA/GyjhPATLiw5drH8AHCF40iNGWSpn5tzLzd7C5AzVxceKf3lWGDHme8uUbyaZapUqWYnJgEHu/+IIbJ09iOXkSUfGsqISe3Q1m7hbWJ8L9clrNLLdD5grqePK9cPIk5qNH8fL0xMPDg4SEBNLS0nKlq3ROv2DV/soCygLKAveygBZu9sQTT1C/fn2qt63O6A2jibHEoLPqcscLymeYEdVRDQbRs+47atbshqurK4sWLZLhZSJXKLd/FMwomLkrzFiubwPrDmAPO3e+yIiRfri5l+U///kPlZtXYcDTA4k4F0mcLU72iNFZcumDlzsf3wI9ioKZe8GMloYjmmlZEA07V6wQpaFT6Nx5Cc7OFSlRogSNPOvzaEwMx3//Hc6fl8UBikIZ53/CTAcPDy4bDLKYgug1c7dh1usxGwz8ej99Zm417SwoAMnN8yqYyW2XQB1PWUBZIL8toKkzQrWYNGmSVDGajWnGgkMLiLHlEtDkI8zY2zxYCQ9PoUePFZQpUwZvb2+OiwgLm00CTW7bWMGMgpkMYUaoMZYb28C2h7OnPiPGNB0vr9o4OJShbFVn2sxsw8zdM4m+EY3RapQhZUWh0aWmquTGUsFMVmHmDtTo9WaZMBgdfY3g4E+oW7cPjo5uuJQtS0C/fnz29tukCoXm9GksR4/muBpWbjrW2T1Weph57amn8K1cmWsrVkB8fNbGihX8uWwZAd7e2WuaqWAmt79H1fGUBZQFlAVyZAEBNEKxeOedd2QfFpfqLvhG+WJMNWKwGMhx35l8gxl70r9eL/rKvIeLy0NUq1aNp59+Os9UGWF4BTMKZv4FM9abCWBOBtteErY9S7/+7alc2Z3SpUtRt0ddRv3fKJadWIbhZvEosXy/YKNgJrswI6BGhJ5p1c9usGjRX/Tps4qKFb1wdnKiaSMvjEvDOPzDD9guXLjd9DG7IFEYtk8PM68+/jj1ypThy8mTSQoOJjkLI2niRDaMGEHbatX48I034MoVLMWw8SgqzCxHTpTaWVlAWaDgLaApFufOnUP0nalcpTJVW1Vl+PrhMvJFZ85h1Es+wYwo8hMdbWX69CTq1u2Ju3t52RT0woULuZ70n/5VUzCjYOY2zEg15uZ2bNavuXblKx59ZD4tWjTAuawjbnVd8Y30JXR3KFGXo+RMgcF6v25+8dhPwcz9wMwdlUYUCBAxtxERZ5g06ROaNx+Pa7lqVK1UmdFDhvLl++/DmTPYHlCVRoOZ6ydT+Pj1N2hSrx4D2raln49PlkZ/Hx96t2qFb5s27Ni0GS5elIUSCgOo5dc1qDCz9F/nal1ZQFngQbaABjS//PILM2bMwNHFkbp+dQn+NDjnQJMPMCNAxmSyMmvWt7RsOREXl/L4+vqyb9++PAsv015vBTMKZkCUXE7dji01iZs3drJ3z6tMnjyAunWrUapUSer1qceojaNYcmSJVGMExORY8izQbJb8gSkFMzmBGbtKI5QaURVNr7/K3Lm/0KfPaqpWbYWbqyt+nTrzRPwKzvzxB7bz52XYmXRuH7AwKtFP59yvv7JzyxZ2fvIJO7I59mzdyqU//4RTp2Q+UX6BRGE4j4IZ7atcLZUFlAWKggVEuNmNGzf46quvGDx4MA7lHGjUvxHTd0yX7S70Zr3MT862F5PHMKOBzJw5P9C+/WxcXCrLnjIbN24kNTU11/vK/PO1VjBTzGHGDjHbwbKD82e3sv5VI35+rXF3L4uDRxk6zOlAyPYQIi5HIJUYocbYsv0xKgbo8m+bPOgw8+nYsdwQ5ceio5HZ+SJDP5MhGmtei41jlV/HfzXN/HsFM015yfrSXhVFgJGVsLCTBAa+RePGwyhbtoIsDjBj4kR2f/mlBBrbiRMPZrUz0SBUGykp2LIzTp5EhFoVBrjI72tQMPPPr3T1t7KAssCDbAFNnbl48SIffvghPXv2xMnVicaDGzMtYRqx1ljpi2V7QjmPYMb+/WzDZLJJRaZDh1m4uj5EkyZNZJ7M+fPn8xxkxOutYKYYw4wAGdEA05yWzP4/3yEm5mFatWxIyTIlqda2Gv0f68fC3xdiSLUDTLY/PMUSYe5AzYMIM9HRNpYtu4i7ewMmNPFG59uNaD9fjL7d7jqifX3R+fvTo0ZV6tQbgi7qHEYZJibKM2YdXDLfVhzHrtJERV2U8bhduiylYsXGlHd3Z0j//nzw+htYBQScOiVLOOe3Y63Ol/O+Mdm1oYKZB9ltU9euLKAskJEFNKAR+TOiIEC3bt1k/5mG/RsS/HEw0ZZoom3R9hYYWZ1cznWY0XJbrRiNZqZNS5YNsN3c7CCzdu1aTogecbeagmZ0n7n5mIKZYgozdpBJ5tqVBHbseIFZs0ZQp05VKWk2HdOUsR+MJfJSpPzQCCVGgcwdSMnq2oMGMwIkhPASEXEDPz8T/p3m06frIvp2XSSXYj2z0bfrYvp2W0KXDnMZOOhFdLprUknJHE7uF3DsQGM0prFo0UECAtZRt64/jg4OdG3XnmfXrOHM779jEUBTBBttZtfZL+rbK5jJTXdAHUtZQFmgsFggPdC8/fbb9OrVi1IOpajtV5uA5wNYlrIMk9kkI2Wkf3YvqMk1mNH6wwk1xkpExHkCA9/B23skrq7VaN26NY899hhHjhzJN5ARr5mCmUIIMxI0RCK+zGOxL623/ian13srP8aalszpU5/x3nurGDmyOx6VXHGt7UqH2R1kyWWTzZS73WeLoUrzIMKMgA8R+xoWdoqFi44yf9ER5i/M+hD7LFt+Po9ARgMgu0ojVKTIyItMmPAR3t7DcXBwo0GduugWLeHnXbu4IWaFCmnYmXDC73cUdUDJzv0pmCksrpe6DmUBZYHctoAGNJcuXWLz5s0EBARQrlw5KnhVwDfclxnJM4i6GiVVGm3SOdOJ5xzCjFZlVERICIgRxXnmzPmePn1WUrNmJzw8qtK9e3eef/75fFVkNJsrmMkpHOTi/nZ4ScRm3oE1bRc2885b6ztv/Z2MLQfnu3P8JA4f+pBnn11Ot24tcXR1oFqbavSM68niQ4uJscZI+TLTD0UxBJOsqjHpt3tQYUYAjQCFmJj7GCahnORWaJkGLxkvxT9XURzAaLxBaOhe2raZiatrLdzLuRMyIZhtH3/M5cOHCx/QiNyY7OTE/GPb4pofkxHkKJjRvsrVUllAWaAoWkADGlEU4Ouvv+bhhx+mXr16OLk70XhYY4a+NJS5P82VUGO0GeUk9N/ARig2t4Z4PvJaJK4NXalQqQJrH3mEZs2aUalSCyZO/FSWVBawYo+o0NokaCHetlvPW1m8+C/GjHmHVq0mU65cNWrUqMm4ceNkjo+WIyPCy/LzR8FMDuAgxypJunMLSLGZkzDfTOJ0yid8t+9Ntn31NJ9//hjbE9bx048buXB+q3xeVB0TFciyc34JMmmJWFOT+PP3t4mLfZgmTeriVN6J+n3rM+zVYbLkslBkdNYc1jNXsCMt8CDDjDYLY1/e+ad2779F5THtn2HGEJK7oWd2ydtkshAWJroNx1ClSgtKlXJgSL9+fLRhAxf277cn1x8/DmIUcLWztGPHuHLoEJcOHszWuHzwINcOH8ZczHrJ3O31UjCTn+6COpeygLJAQVhAAxrRVPPUqVOsWbMGPz8/KlSogEd9D1pPbc2I10cw+7vZhJ8Nx5BmkM3MRUNzATAazIgiThFXIzKEGdGkOjraIiMzNJgRk4WiZ4zRaCEy8hILF/7GuHGb6Nx5ERUreuPuXhEfHx/Cw8P5/vvvSUtLk6Fl+Q0y4jVRMJMOKLIDB7m5rQAZa1oiFy9sJSnxedaumkvwhBH06TWA7n4D6Nd3ACHTAln37HK+3fs6ly9+KcsoZxVoBMiIRP+0m0n89OObLFwQSM2alXGp4kLzoOZM2jqJaGu6hDIFI7ligQcZZnIXOPIeauyNuuwzR8OHr6d27a6ULu2Mb4cO/N/zz8s8GlECWTq/BQU0ornjsWP8kpDAa48/ziuPPJLl8eqjj8ptP3zlFQ7u3i2LHMh7KWAwuxto5MdzCmYKwrVS51QWUBYoCAukT6bfvn07M2fOlDDh4eFB+drlaTKqCQMeG8DkLyaz4LcFhJ8KR3dNJ/sCSrCx2JUZt0Zu/1BmWspebiaT+VaIuJiUvEFExFkWL97P9Ok7GDLkBdq0mY67uyeurm54eXlJNeb9999HFCoQPwK6xCiIHwUzBQwztlShliRyKmULb7xmoLtfN+rW7U3nziZGjPiQceMSCAjYgI/PUurX92f0qIFs+uhRLp7fii3t3gqNABnLzURuXE+U/WOCJ/SlYkVXylVzocOs9sz9YS4xtpg7SWS54sanD7YqvusKZvIeYtJDlwAaIZHHxdmYPPlL6tfvjYODKy28m8jCAMd//hmzyKMpAIVGOt0pKVw7eJAnY2KoW6ECvjVr0rNOHXrUrk3PuwzxfG9PT1pXrUqPtm3Z8NJL2EQ3ZaXQ2OFUQOLRo3h5eiK+1BMSEuQMYUF9qRbEF7k6p7KAskDxsIAGNOL/25UrV9i0aRMhISEy8b5KlSo4l3PGo5EHzcc3p/eq3ox5bwzTdkxjzs9zWLx/MYsOLMLN042KlSryiAgza96cSpWaERT0OkuX7mfOnF+YPv0bqcAMGPAEbdrOoHr1tpQp4ySVoIYNGzJq1Chee+01Tp8+fVuJKej/twpmChBmpGKSlszZM5/y3LPLqO/Zgs6dY4iKusDq1TZWroQVK2DVKrFuY8GCP2nZ8mH8fLvx7tvxXL9qV1wyU4kEKFlubuf6tR0kbFtH7z5tcXF1xLVWOfz0fiw9vNSeH2PRq2pleQBxCmbyF2Y0aVyns0igmT37e1lhxcmxArVr1MK0bDkHv/uOtJMn8x1oNJi5cvAgT8XEMLZ5c47Mm8fF8HAuhIVxcdmyTMf5sDAuREXx3ujRzOzRg5fXrcN28aKCGa24g4KZ4uHFqbtUFlAWuG2B9FBz/fp1kpKT0Bv0+Pv7U6dOHSpVqoSLiwsOzg44VXWiUutKNAhoQLOxzXBwdaBCxQqI8sktWrTA2bki9er1xMtrKNWrd6BcuVo4OJTD2dlZThDVrFmTNm3aMG/ePLZs2YI4nwYv2nXcvrACWlEwU5Awk5aIOTWZDz9YSbu2HfH1fYQ1a+zJ1zqdGeGU2YdYNxMbK3IRLtOu3WyGD+1NcuJzWM1fY72ZkGH+jPlGAjdvfsPHm/9Ls2Z1Ke1YCvcG7gx8ZiCR5yJlaJnKj8k75UjBTEHAjP2c4nMjPi9Llhyljc9MypatiodbeWZPmcpv33xD2unT+QoD6WHm6dhYprZpw9WICIiNtTclNZkgsyHqZa9cyeeTJjG/b18FM+lC61SYWQF5Duq0ygLKAgVuAQEUIo9GAwtxQWfPnpXq9Oo1qwkMDJSKTY0aNSSUuLm54erqKiuiifXVq1fL50WFNPG4eKx8+fJUq1ZdFgYYPHgwUVE6Nm3azHER0ZDup7BAjHZJCmYKCGYEgFjNuziw/13mzQ3E23scpmizrM6k11sybDQogCYuzsrs2b/QulUgURETuH51GzZzsizjnF6hMV9PwGr9lvfeXcFDD1WiRMkSVG5VmaD3gzBcN0iQEc523rny6sgKZgoOZoRKI4BGVGQT5Zu7d4/G1bUODmUcGDNkKL/s+hrr+fNYjhzJl4IA/4SZSa1bczEsDJvRiFWnw6bXZzrMOh2W2Fg+nTCBuX36KJhRMKN9f6ulsoCygLKAhBkBF2Kk/xGQc+3aNf46+BeJiYlsfGsjL7z4giwgIFQZUR1t3bp1EmrEcsOGjXzxxZf8+utvXLhw4W+QJI4rjv9PeEp/voJcVzBToDDzNZs+WkvfPkMJCHhThpIJByx9HsC/162Iyk3d/VcyZsxwvvv2FWyW3TKcTMCMDC27sR2z+Vv+7/VoKlcuT4kSJajdozbTkqZhTDUiKloU5bLLAiJ0Fp0cEtju1Uwqj5BOwUzBwoz47NgLA4hkxmsMHrSOShWb4FhGVDrrz4/JO+y5J0eP5jnQZAQzl5Ytk6qMAJm7fegtej3WuDg+Cw5WMJMOZESBAaXMFKT7oM6tLKAsUNgsoKk1Ajo0uNEeM5vNiCGqjmmVx7S/tee0dNViagAAIABJREFU/cQ+2n7aY4XtXtNfj4KZAoAZWYZZVjBL4qUXl9Gx42hmzPhRhsXYk5gz922EahMfb2XkyLcZ2H8s770bi82yF8uNBKnOiGT/a1d38cK65VSqZAcZz/6ehO4NRZ+qv1OiL48c+ALTY2wG2RvHgJFYYlnBSlawijjiiSba3jcnn5UoBTOZv4//Del5ua29Tn5U1FWGD3+NalV9cHJ0oq9/D/Z8+SW2S5ew5DHQKJg5kSfAqGAm/de5WlcWUBZQFrhjAQ1INEVFAxZtqcGK9rcGLdp+Yvmg/CiYKQiYSd0O5iTSbmzj6ScX0bHjJObPPyZDYrIKMxMmfEnfPpN5+aXl2Cx7MF/fhiU1iQvnt/Hk44upXq0iJUr8h0bDGjFr3yz0N4o2yIha6nHEYTAbmb0/lPFfjibosxFM3zeFiCvhxNriMGGyw1xmIHcLiEQekQCRnKpXCmbyElCye2w70Iha+aNHb6DGQx0o6+xMz26+JG7ebE+oz0OgUTCjYOZBcQrUdSoLKAsUPQukBxRtXdyltq4tH9Q7LzQwI0sI30iQCoNQGcSw3tz+r1yQ9HkhD/S6OQlL6naee3YJHTqMZfbs/dlSZgKDNtO/XzD/96Yei3kPade3cyrlcx5ZM4/atatSsnRJmgQ2Yfb3s9HfLNogI5SXyAuRjHpnOE0meFG7szf1O/jSsL0/dTu0psGQhvRa2535++dhtEZjlN1w7+QLCWjRwENAT5xtBbE2oejY4Uc8dz+Kk3bMsDPLqN+7PqVLuTNwwHOyjrvI6b7TaTe7jrna/v6UHQE0ViIizhMU9C61a3eX1Vr8u3SxA82lS7LEb170R1Ewo2DmQXUS1HUrCygLKAsUdgsUOMwIiJFhV9Zd2Kx7sVn23RrfYrPuBnNiptW6HmSYkWWZzbt4a2MM3bsPIyjoM1assMmk5bs5akK5EVWa+vVfx/Bho9i+7UmZ6H/i2BYee3QBnvUeorRzaVoEt2Duj3NlJ9jb3V/vyyW/Hzc+n/ax2bvczvtzHp2XdcGzfUe694kkMOh/TJuaSMi0ZCaM/5QBg/+Ld/vBNA1sTfDW8RLuhJIjlRebQSo2MdZYwi+HM+vXh5mcPJ6Hv59G2NmlmCwxxNhiZZ6RtGM2bKhgpjBClwY0Fxg37gPq1euNk5MT/l27kvzJJ1jyqH+LghkFM4XdGVDXpyygLKAs8KBaoEBhRoKMeSepN5I5fPA9Pt2ylvWvhPPKy8t59x0TP3y3Xia22yxfy8aSDzK8/PPaheokyirv/uZlJowfSceOEbJSmciJ0euFw/VvR1CAjMlkZenSk3ToMIfZoYEcP7qZEye2ytCyxl61Zf1wCTI/zCXabM8Vya4Tnk8okg0syOCKbPaKbPP/mEfnpb40aNmPMWPeYemSw+j1aURH27vBi5n48OVnJNy06xhKsxEdmPD5eBmOJjviphmlYjPg6b60muZDk8EdaDbAn2aDu9BivA/dY7sxY/d09Nfs4GSwZV2lUTDz7/dwRu/r/H7M/vmyEhl5geDgj6hXtyeODg709vNj77ZtsmyzaKwpAeQfCef3q9oomFEw86A6Ceq6lQWUBZQFCrsFCgxmpBpjTmb/H++w7tkwpkwewcABo+jXdzL9+oYwoP84Ro8cii5iPF/veokb15KKFNDYFalkLpz9nKeeXELLFn0ZN+5z4uPtsf06najA9PcRHW0lJiaVgQPX4d99IK+/ZuLo4c955ullNG/miWN5R9kQada3szBZTfelJmSADTmDjrzYWyoy0YSfC2fgM4Oo38aPwNEfEW1KJdpkwWC029AexiWgxoYpxsrsWT/QymcS7WZ2Ze5PczBcMhL86QR8Jnekud9g/HpFEjD4JcaMeodhQ9+gV584WvkG0mxYO0asH86ylGUIAMoq0CiYKZwwY4cnmyyDHhV1ifHjPqRWra6ULlWawX36si8hgZspKVi1alm5ADQKZhTMFHZnQF2fsoCygLLAg2qBAoEZUT7YmpbMnm9eZPHiyfj5jcLfP4phw95gwoTPCQ7+ijFj3qdv38fo2GECI4b34cMP1nDpwld2oElNzLBJ5D/Vj8L+tww1S0vm5x/fZM7ssTRrNpKJEz/HaEglNlaAi+32EH9HRp5j+PA3add2ENGGUPbu2cDzz+nxad1Igoz3SG9CdoRIkBGOdJFUZLCXlTbZYgjZMZUWgV3w6xqNKeYmRqNQtaz/UrXETLxQvGLjbIwdtZmmvgMY+EQ/gj8Ipvmwzvi0m07w+C2EhZ3CYBTqlw1jNESEX2RGyA78/CJp3LMjw14ZxvLTyyXQZKU4gIKZwgwz4trsIWdRUZcJDHyXatV8KFmiJGOHj2Df9kRuHD+Ra0CTEcxcCAvDajRi0emwivLLmYw0nQ5zbCyfTJjAnN69VZ+ZdHAp7XrypMx18vL0lI3hEhISZNlRkdCqfpQFlAWUBZQFir4F8h1mtNCyn354g1mhQbRtO5aRo94lPPyMzAURTSHFEM67cE5DQ7+lffs59Ondja2fP8XVy9tkLxVxnMIOK/e6PmmL1ERSr29n9zcvETJtBB06jGbQwKeZMvlL5s77jQULDjF79k9MmPAJvXvH07nzYMKXh5Cc+CqvvmKifXtvHN0d8RrmxeSvJt9WZLLibD8QKkxGyo5I4E8zMOSlABp36EfIlN0SVO7Wo0cAjVBoloWdo4v/Uur39ab5sI60aTODuXN+JDbOSrTJruhoYX4CaMTjERGn6O5npGn/zkzYPB5DqiFLoKhgprDDjOhDYwca0Ydm5Mg3qFy5OWXKODJ17Fh+TEoiVYSb5YJCkx5mnomNZVrbtlyLjMQaH48lJgZrbGymw2wyYVm9ms8nT2ZB3768/Nxz9uprx47lSbnj+w2lK4j9FMwUfSdF3aGygLKAssC9LJCvMGNXIpK4fPEroo1Tad9+JGPHbpHgIkKo7GFVYnZdm2G3SbCJjDxD06bjmDwxgF9+fkuWIC4KMCNgRwJNWhJpqUn8/ttGaZcB/fri330CvXotoW9fA716LqRnjyCGDe3P448t5pef3+Pdt1fTrVtLHN0caDi4IRM+nYDJZg8tK8ogI+4t2hZN+MXl9IzvS6s2UwhfdlmCigYhmeVg6A12oBkc8CzlK9XHs2FfQqbtsOcq6cR7TgtPszvh4m/ZRT5W5CkdpUXziXRZ7M+iQwukre9lZwUzhR9mxHtFA5ro6DQCAl6kUqUmODuWZdHDM/lj927MGswcP37f8JAeZp40mQjw8mLbxIkkzphBYkjI3ce0aSTOmsV/+/dnkq+vUmaUMnOv73X1vLKAsoCyQLGyQL7CjGzsaN3NzuRn6Ne3NwMHPoXBcF3GrmcUHiQcjagos6zyNWPGd9St2543XzNy5dKX2NKSikzZZjvQJIJ1J9evJcrwu6efms/SpWNYuGAkuqgJvPZqOL/9/H9cvrSDL7Y+Q79+HXAoV4Z6vesxbtO4YgEyQkkSACGgbempJXTX9aV9+4USgrNS6li8x0To3ojhr1K9Wke6dVuOMToNg0EDmYydb53OTPwKK0P6v0LzQT2YsCUIUf1MhvJlpBzdekzBTMb2zAw2C/JxDWhEgY1+/R7Hw6MBFd0rEhcewdEffsB68mSOCgJoMHP90CFefvRRujRtSnD79kzt1IkpHTvedUzu2JGpXboQ0KoVY/r04f033sB2/jwWpczYXxMVZlasnBZ1s8oCygLKAv+0QAHAzF7WrnkYX99JTJ2aJFUZocTczZERz4uyxV6Np7Bk0UT+OvA+1rSdRQZmtHA0CTWpSbIktc26D6zfgG3nrb/3YrHs4bt9rzNqlD8ubk5Ub1+dEW+MwGgxymT/eykFD2xYWTpgsCszJpadDcM/ui9t2s0iMlLkywjHWSgrmTvQGswMGvQk9esPZuTI9bcryN19P5FvY+XhkH007zacQc8PINoafc9QMwUzmb8Wd7N3QT0n3h9GoygMYKZbtyhcXWtSp2Ytnl61itO//55joEE43ceO8fP27bzw+OM8umqVHI+sWsW9xn9Xr+bx1at59+WXObh3L5w6lSO4KoiQsLw4pwoz++dXuvpbWUBZQFmg+FkgX2FGJP7bLDuYO2cYvXvrWLjwMDEx9vCyuzkwYmZ89Wob/v5PMXrkEPbseQ1r2i7ZVFMDgaK0TN80VNyX5cY20m4mcvDA/5g6ZRAeFV3x8PJg4DMD0V3TcbtnSjqnvyiAS0b3IGBG3K/upo4BTw2gadvhzJvzl1RcMlP37O8t4aSKQgCpdO26HE/PAUya9HkWYcaKKcbGwvnHaNVpIr0f6Y3OrMOIqGyW0VXaH1Mw82DBjHifiCqCIrcqIuIirVtPx9m5Ei0ae/PGunVcPnQo50AjQqTOnMF2+TK2K1eyPy5ehNOnIQchb3kBFQV1TAUzxc9pUXesLKAsoCzwTwvkG8wIkMGcJBtghkwdTP/+q1m69LSsHGUP8cjc8REws2qVjYCAtwgYHMDOHS9JmCkqeTN3AzFxj5bURJlntGjhGKpW9cC5sjO9V/Yi4nyEzB+5V7hT5u72g/mMuF8R5jV+cxDevbswdOB64ldYEO+TzNQZoe4JIJkz9zeaNgvE07MXU6Z8kUVl0B6etmDeEVp1DqbPo33QK5i5qwp2t8mJwv6cABoRjrh48VG8vUdRpkw5urVvzwdvvMmNEycQPWhy4rwLB9x89CjmI0eyPSxHjypFRuXM/PN7XP2tLKAsoCxQrC2QbzBjd9iTsFmSmRU6hD59olm06Fi2YKZX7+cZPnQw33z9apFWZtLDjfmGKDP6DWtXz6J2rSqUdiyNb6QvYcfC7EnoogRzMfuViofVxJJjS+gW5UuDRv1lI9GYWKsEmr/Dsb0ss9EoQohS6dEjDvcadXiogQ8jRryWBUXHPlsfF29l6qQdNO8+mKGvDlYFAO4SzlfYYSUr1ycKP8TG2pg9+wcaNuxHyZKlGdS7N9s//hjL2bMqXyUdUOQE7HK6r1JmirX/om5eWUBZQFlAWiBfYcZ6MwGb9VtiYybTvfsMZszYc2tm/N+9QdI7HMKxWLUSmjabzdzZ4/jj9/8VC5gxX98Gtn1s/nA19T1r8J///If2c9uz6M9Fd3rJFDOQ0cBNKwTw8NczaDCoCQ3rBBKpO8uqNWAyaSqfqF4G8SsgboWNnl0fpWbL5nRf2YV2oZ1p13YeUVE3MBjuHuooCwCstNHLdy3Nh3VlasJEYq1xqgBAEQcaoeaJMvFTpmyjTp3ulChRguDRo9m/Zy+2s2exHj2aI4Ump4682v+EKgCgHBllAWUBZQFlAQoAZvby1db/4u/Xi4CAVxDVg4TT8PfZdM0ZFbPiYoYUFi46wUPV27L+pchb1cySi1wBgPSKjAA/bN/w+y8badq0LqVKlaTRsEaE7g7FmGa8pyOtOf1FdilyVawGDKlGpu+aToPB3lSp1IGhA18nbOkpCTQxMRAZcYVJ4xNo7DmBqs08GfJyAPqLBoa9HIBnu86MGrqRlatECWbzrYabd4oIiBycqKg0WclszuxfaNCwPz1jehF+IRyjLVpWVrubfVXOzJ3PcfrJiQdpXYYnmlIZMWI9Vau2poJ7eZbOms2lv/7CppLwCxzmlDKjvBhlAWUBZQFlgXyFGeGsi5LKovHlwgWjaO0TyLSQZFmpTDiOdqixz5KLv4WDKWbWV62Cho1mMGRwD77/bgOW1CIOMqnbsaYlc+3yNgYM6IizswMVW1Rk4taJ6K7r7E78XRLP7+ZgF7nnbAb0N/XM/3U+vVb14qEO9ahSswX1ag2kfp1hVK/RkWrNG9B2TnumJE4h4kok0VYTi0WI2nJ/6tbryaTgL1m52oYpxu58a4UEok1I0Fm06ACNG4+g1cR2hO4NJdpiuifICDsrmHnwYUbkYIkQxcjIi/TuvQo3tzp4edbn8dh4LKdPY9V60BSSsKviptYomFFOjLKAsoCygLJAvsOMNXU7NstOvt+3nsAxg2jSNJApU5KIj7dJqBGz6WLEx8PKlQJsrtC6dRhNmzRly+ZHuXppG7IqWup20isZRWXdJqqXpSZy40YyYUuCcHd3wbGCA8PWDyPiQgRG690raBU5WMlKGJ0AuzQDEWcimP3jbII2j2XQc4MZ8MwgRr41iul7phN2bKmEHgkZNj0Gs5E5P82ly6Lu1KzfiZ7d1jBv7n7Zc0a8/wREL11ykhFD3sTTszctxrZl2vZp6K7r71rBLL39FcwUBZgRTTVFhTMrCxcepGOH+ZR1Kk+Xdu14f/161e+lgCFOwYxyYpQFlAWUBZQF8h1m7OpMIjevJ5Kc9DwzpgfSrFk/OnbUMW7cZ8yZc4AFC44wLWQ3/fqvo2HDYXT368b/3l3BpfNbsYryzkUVZETlspvbuXolifWvRFGtWgVKliqJf0x3wk6EYbAYsuxIp3eqi8W6zSBBT8Ce7oZOVnqLOBtO1JUoqWT9CwJvKTr/z955gEVxdX38i13sxt4RLKjYwIZItVJVUEBFwAKKBQFBpGynqemJMSam+MYUk5iiMUYTezeJmqjRNHvvHdid+X/PuevohtBZYBfu+lxndufOzJ0zs+z53dOi/4yG95s+6OEzEJ1tXdGz+1TY941Gn14zYWUzEtbOdnBTD0fkT5FIuS8rlvw5zFQOmNG7xVECiRxERBxAr56BqGdhAe+RI3F423YGNDx+5lKFuJxxmOFKDJcAlwCXAJdAhcAMcnaCLBCPH+7Asd/W4M03FmHqlAlwdfHFwIETMMB+IoY6jIWvjy9SkmZg+7Y38eDuVmaRqSwWmNzXQdYmAjWCvP1730GvXp1ZwL9tsC1i/4iFIpuDTGFQxoqGEtSISlbUkgpb0jrbj6w3uV3z6L1OicQbixGxdybGrRkP97QxcEkZDTfVaHi/44OQH6Yi/kJcieTPYaYywQzFUpHr6wNMmfINOnZ0RtPGTRARHILb//wDgcfPcJjh+gSXAJcAlwCXQIVIoIJgZhdA7mYUQ6PdjVs3NuPnQx/g6y8z8cGqRKxauQiffazEtq1v4vy59foYmZzdbJ/cEFBZ3hPM6LJ245+/v8TkySOYRaZFvxaYdXAWlI+V+jiZorhc8T56aJHgJTfA5JaPAfxQIc5FV+IRezYG8RfjkPwgGUqditXy+Q8I5T5OHu85zFQmmCF3M338TGLiDfj6vo0mTazRqX0HLM/IhHjzJnQXL0KsYLcrHjPTBDt2UDr7HIiiWCE/qvykXAJcAlwCXALlK4GKgxkCGQIachvL2aNPtZyzF9qsHch+tJ0lChC0ByBo9z3pVzljZAjGSA5C9m5cvfI9Xlw2D42b1EOd5nUw7sNxUGZxkCnMIlPa7WTRoUZWHLWghlrQsEYZyxjEFAZEeYAMjYnDTOWCGb27mR5oYmL+gYPDIljUaYTB/e2wd+NGPdBcuFAhFoqqBjHS9XI3s/JVGPjZuAS4BLgETFECFQozklWFKfNZBDY7Ae1u1shqQ+mJhazKCzGGIHPvzjZ8/dUy2PbqjFr1a8F+rj3kD+Q8TiYfUCgtwOS5P8XRCPJnjRIFlOIfh5nKCTPkbqZQZGPWrIOwtvZk8TOBvr64fOoUdFeu6GufcAtNuUAdhxlTVCv4mLgEuAS4BMpXAiYBMxLUSMq9PlvZrkqZrczwWvXXuws5j3fjp4Or4efnjBp1a8DS3RJRJ6OgETRMsS6NQs33rTgJcJipjDBD16S3zshkdzFp0pdo2rQbGtSvj2VyOR5fvgzdxSfFHDnQlDnQcJgpX4WBn41LgEuAS8AUJWByMJNb2a/M75lFKmcPzp9dD4V8Giwa1EbzXs0Q8PlEpIlpkOtKZxmoODWen5kkwGGmssKMYfzMVbi7Z6JOncbo2LYddm/YgKzLlyFcvFjmirzkalWVlxxmTFGt4GPiEuAS4BIoXwlwmMmpIAsQy16mLyD6ySep6NGjI+q1rIehiUOhzFHqQaaEsRocJUxDAhxmKi/MSPEzCoUWCxeeRdeu41CtWjX4e3ji7LHjyOHFNMsF5jjMlK/CwM/GJcAlwCVgihLgMFNBMEOxQLrsvfjp0AcICnRHjTo10MXDGjGnY6ARuXuZaeBI6UbBYaYqwAzVn8nC9Ol7UL9+azz3XDW8lbkEd86cgY4DTZkDDYcZU1Qr+Ji4BLgEuATKVwIcZioAZiT3shvXNkGjmoEGjeqgRZ8W8P/En7mXyXSyUoSdl04B53sbTwIcZio7zND1Uf0ZHZTKB3Bx0aBGDQt0tbTCoS1bkHXxIgQCGh47U2ZQw2GmfBUGfjYuAS4BLgFTlACHmQqAGbLKaLP34vPPMjHAvjvqNK2NgfMHQP5Yn72MFX/kOGP2EuAwUxVgRh8/Q0CTlHQHrVoNRLVqNREZPBWnjx2H9vJlnt2sDGGOw4wpqhV8TFwCXAJcAuUrAQ4z5QwzBDKibj/+OLkWIcGj8X/P/R+sx1hjzm9zkCqmQiZwq4zxbCMVeyTzhhmyOJSkVQ2A0cfMPLtWuVyHtDQBU6duhYVFc1SvXgMfv/km7p89C4Gna+aWmfL9Xedn4xLgEuASqFIS4DBTjjDD3Muyd0GbtQvpabPQsmVjNO/RDGPf9oVGp+HZy8zeFvNveDJnmFEqgJK03Ep+VXqvBxod7O3nonbtRnCwH4BDW7dymOGWmSqlVPCL5RLgEjANCYiiCEEQoNPpoNVqWaN1avR5ZXpxmClHmBEeb4eoO4gd21dg5MgBqFmvBuwj7bHo6iKoBBWrQv9vdZi/M2cJmDPMyOUi5E8tM88sEPnBCfXV9y+8b37HMPfPSWZqtYjo6NNo2aIvateqi/TEJFw+eRLC1asQL1woMwtFVU3PzN3MKpM6wq+FS4BLoDQSMISXygYrhcmFw0w5wYwU9H/vzlYsiJqAxo3roaNLR0zdOBVqnZoXx6xkVhmCMHODGblcgEajhYfHSnTp4gdrq7Ho3NkHlkVo1K+zlS/69AlHYtIDKJSSi1rVghuyzqSm6uDh8RoaNmyPnl274duPP4GWJwIoE5DjMFPYTzzfziXAJVDZJSBBDC3pJb2/d+8efj/5O37c+iPee/89ZC7JRFJSEqKiorBgwQKcOHECOTk5rL+5y4jDTDnBjJC1A6JwCF98noEBA7rDokVduGnckHQ7CUpBCQWvKVPpcMa8YEYEKeLpadnobjsTAd27Y4XveLxLzWcc3vPNv73rOw6rfMdDPswZjRtYIn7RLSiLYdUxd4vMv8cvQqUSER9/CV26eMOibiPMnz4Dfxz6CcK1a9w6Y2SXMw4z5q6C8PFzCXAJlFQCErTQktrjx49x4vcT+HDNh4iLj4OPjw/s7O3Qo0cPWFpaom3btmjZsiWef/551jZv3sz2kfYv6ThMYT+TgRlmucjaAab0Z++CSEUlqRbL4x2gbSgn6CiL87Cgf+0+XL6wAVMmj0L9+nXR3a87Ig6E693LBHmlU+TN2T3MWGM3P5jRIiMtG5Y2oUh1GoJzSYm4nZyM20mJuJOUlE9LxO2kJNxITsLukGA0rt8BcfE3qzDMUHYzsnAJmOD/CZo1s0Gn9u3x/muv4dH58xB4djOjWmg4zJiCGsHHwCXAJVCeEsgNMRcuXMBnn3+GmJgYjBw5EjY9bBi0WDSyQP129dFmUBt0H98dA2YNwOCowajdtDZq16mN7777jsOMsZR+AhXWtHsh6g6xmBII+0BN//4QoNsDwYyBRkdWGd0BrFgejy7W7dDYsjF83vJByr0UbpWpxBhnljCTng3LHtOw1NUJt5RK6FRqCEolRJUqzyYoVRBUKmSr1Ph5xnQ0adAJcfFV2TKjT9WsVApYvPg67OzCYWHRDIFjx+KnH7dCvHYNAo+dMRrQcJgpTxWKn4tLgEugoiUgxcIQ0Jw9exarV69GSEgI+vfvj+bNm6Nhq4bo6NwRA+cMwqgXRiHgswBM3zodc3+ei9iTsVjwxwI06NwAdSzqYNOmTRxmjAEzDGJy9kDI2YOrl7/FDz+8ihUrFiIzIxxLlkRg1aoE7Nm9Ag/u/QhBuw9C9i6zs9AwS5NuP/756wt4ejjAol5t9A/vj/kn5vOg/0oMMuYXM0NuZlpkMJgJQ6aLI27J5dAqlBDkcogKRd5NLoegUCBbocCBaWFo0qBjlbfMkNsZuexpNCKCg79H27ZD0KZlKyxTqnDnn394djMjuppxmKlo1Yqfn0uAS6C8JEAgQ+3BgwfYsGEDZs6ciT59+qB+w/po0asF+oX1g+erngjbMg0xf8SwMAaFTgGVqGKx2RSfLXssQ8OuDVG7bm18t4lbZkoNFQQyBDG3bvyALd+/gkXxoRjr64/hw0Pg4jwLLi4RGDkyBH7jx0OpmI6fDr2PRw92QsjeDZgR1Ohh5iBeeSkKHdq3QLMezRD4eSBkj2RQCAoeK1OJgcZ8LTN6mLktl0NHVhm5PN8ofoIcgpkchRIHp3OYeRY/I4KsM8nJd+HouAj16rWE94iR2LPxO4jXr3PrjJGAhsNMealR/DxcAlwCFSUBKaaFgvVPnTqF9PR0ODk5oV69emjZuyX6z+gP/9X+iDoRhZQHKXp4EdRQivp4bCrELtfJmc6Z9CAJDbo04DBjLIsMgcyNa5ux5kMVxo8bi4EDp8PDYwVCQncgMvIYZs/+DVOn/oDhI16GnV0ggqd4Y+uPy3H/znaI2bvNIo5GpAKZ2n04e/pLuLn1Rx2LWhgSOxixf8dwq0wlhhgp5obDDGU0y5eDKv02feyMiOnTd6Cz5XC0adkaaUnJuHfmDLfOcJipKL2In5dLgEvAjCQggcyjR4+wbds2hIeHo1mzZmjYriFsxtpg/LvjEf13NBRZCqhFPcAweKFYbMPEUqKCwU3yw2QOM8YBmV3MXSzr4W58tjYNo0d7wXFoAqLmn0Bqqoi0NJEt9esCUlMFzIk8jN69gxEwwQMH9r2HrIcUZ2P6LmeUvADiIby7gAKMAAAgAElEQVT6ygK0atUUTbs2QfCGKZBnPXnIDB+0KqDcS0p+VVlymKnaMKNQ6K0zKSn34eIiQz2LFhjj6ordG76FeOMGt84YAWi4ZcaMtDI+VC4BLoFiSUACmbt372L9+vUYM2YMnqv2HFr2aQkXhQvmHJ0D5WMls8SQpw/pHP8CGEO9ksOMcaFBn9nrAI79ugaBAT4YPDgGcXGnkZYqsCxA5GtOM5r6Rus6ZKQLiI39C507j0FK0gycPfMNhJy9Jm2dYW502r24dGE9hjr2Qh2Lmhia4IC4s3HcKmP4BavE6xxmqjrMSJnN9NYZa+sxaN2iJeQL4/Dg7FmIVHvm4kWjBcNXxcKZHGaKpRvxzlwCXAJmIgECGYqPuXXrFtauXQsHBwfUqFMDHYZ2xIQ1E1ioAllinkJMYboUhxnjw4wu5wBefnEuHIcGITh4C9LTdZDJdPm6naSk5GDpUhE+Pp9gyCBnfL/pFWiz95i0dUZ4vBOC7ie8+koUmjRpgEadGyJsWxjk2Xq/xXzpubAHkm83GwlwmOEwI5dLsTP34eamRt26zTDccRj2UezMzZvcOlNK6wyHGTPRzPgwuQS4BIosAQlkbt++jU8//RR9+vZBrfq1YD3cGuH7wpGmS2MWmAItMbk1JQ4zxoMZslZAtxf3bm/GjGk+cHNTYNGi61Cr9ZaY/PzryUpDfRISbqNrFy8sfy0Od29vgUjpnE0wZTOrkZOzGzeubkJvW0vUrF2DWWUSLiSwVMzkz1hVXK2q8nVymOEwo89sRnVnRISH70G3br5o3aIVkqJjkHXpEoRSKvNV0RpjeM0cZoqsH/GOXAJcAmYgAQlk7t+/j6+++goDBg5ArQa10MWjC+Ydm4d0MV0fzF9cLZLDjPFghmX2Eg7ij98/wvhx3vDyehsKhZCvRcYQbmiGMzUV6NVrFhIXh+PM6a/16ZqzdpQ6s5oxYoGkYxBc6R7vRHbWfqx+PxF169aGReu6mLl/JuQ53CpTleDG3GHGMDVzfl9SynQmpWbm2czyTnagt86ISEl5gBEjlqJWrYZwHjwER7Zug3jrFnS87kyJXe04zJiBdsaHyCXAJVBkCZBrWVZWFjZv3gw3NzfUrFsT1qOsEXUyioFMijaluBij789hxtgw8xMO/7QKXp5e8PF5HyoVzd4WPoNL1hlKDtC7bzQWzA/FqZOfQ8jZDwIkCSRMYSnm6BMcUMrpIUNsUL16NQyKHoSkK0ncKlOyr6DZ7mXOMJPh4ojrMhmy5QrkyGTQUs2ZvJpMhhy5HI/kcux/WmemahfNNJyEkdbp75feOrMT1tYeaN+mLVKTkiFwmCkxyJCFhsNMkXUk3pFLgEvAxCUg1ZH59ddfERoaiuo1qsPSyRKRP0eWDmRIi+IwY2yYOYi///wE/uO9WCpmhUIHpZJmNAsGGprd1GiAnj3DIUuehXNnv4GgJZjZaTIww6wy2bvw+PEebPp2KZ577jk8V/05hO2extLm8boyVckuo88sooIKi64nwGqEFWpUbwRPj5XMGlmUZ15ShMtnaVA0s+c0vDZqOO5kLgWWLAMylwBLaD3/pluyDCfnzMXzjawQF89hJvc9o79fNHGTmHgb7u5pqFatOpwHD8alo0d5ZrNSuNpxmDFx7YwPj0uAS6BIEpAyl928eRMymYzVkGlt1xpBXwYhXUiHTCsr3cQuhxnjwQyLb9HuweMHWxE5ayxcXROxcOGlIsfMLFx4FV26jMKqtxfj4f2tELWUBMC0YIaKel679gMCAtzwf8/9H2wmdkfS1SRQFVYe9M9hxtRhJj0tG5bdJiO4a3t86OeNdf7eWOfniS/982l+nljn543P/X2w1HkQ6tZqjPj4m1CyyYmCJyhyK/yV/T1ZZ5RKHcLCtqBdWwd07tgRK194Ebh3D7pz50ploTCMI6lK6xxmiqQn8U5cAlwCJi4BssoQ0KxZswZ9+vRBo06NMDxjOJRZSuPojxxmjAcz5AZGbmGC9ie8/VYcHIeOQ0DAl0+ymWnztc5QNrNly0S4D38Tzs7O2L7tTeiyTSv4n6CKXMyyH+/C7h1von79uiwf+OSNk1gecA4zVQtk6GrNzc2MrKQatRYjRrwM606j0M1qDLp1HoVunUcU0kayfl06jUB3myAkJt6DUkkgw2HGENAIZlQqAfHxF+DkJEOd2rUx2sUV9/7+G7h6ladpLoGFhsOMiWtofHhcAlwChUpAApkTJ04gODgYderVQe/g3og9F8tqyJAuUWoNisOMcWGGKf3afTj7z5eYFjoWffuEYd68X5GRQYqPwFI0y2S01K+TArB0iYjZs35G27YOeGFpFK5e3mRydWbousgqc/nSZsTGBKBajWpoN7QtEq8nGoeqS/8o8yOUswTMC2b0rp4EIYsWXUN09B9YEH0KCxZI7SQWLMiv6ftER59CbOxpViPKUInn64aJAejvXA6Cg7/F8027wapTJ3z1/gfA3XvQnT/PrTPFBBoOM4XqSbwDlwCXgAlLQHIv0+l0eOmll2BpaYl2Du0QtC4Iaq26ZJnL8tJ1OMwYF2bIOkOpi6lOzLYf34D/eG/06R2KadP3QKPRITNDRHq6iIx0EZmZFCcjYvLkjbC0HIHI2QE4fnwttI/1xzCFgH9pDGSV0WbtwqED76KzVWvUrFcT3qu8oMlRF1yVNa+Hjn9WKSRgfjAjKd1UG0WEqgRNb5GRjsOXuUFOHztDRYD/xqBB89Ggfn1M8fOH7uo1CLyAZrFhjsOMCWtpfGhcAlwChUpAssocP34c/v7+qPd8PQxdNBQJ1/WlPIwWnsBhxvgww4AmZzfu392OHza/jvCZQbCzG4vBg5MwduynmBq8E8HBO+DtvRp2djHo18cNC2On4vDhNXj0YIfJFcvUW2X24OaNH7D89RjUrFMDz9s0RdylOKgEFavQWmoTYaVQ76uWFMwXZghCJDexkiw5xOSGmGfv9aAokz1EwMTPUbfO8+hubY3jO3fxIprFtMrwbGaF6km8A5cAl4CJS0CCmddffx1du3ZFB6cOmLJhCtTCk4lwY+l+HGbKCGbILStnNx7e346jR/6H5W/EYeaMyfDynAg314lwdw2Ar89ERM6egtXvJeHvv9Yh65HpWWT0liaKl9mDE8fXwtfXkRU4Ghg1EGlCmt5EKFYtJZ5frV4C5g0zHEieAYixZUFAo2PutVZWHmjauAmWyRT6RAC85kyxrDPcMmPimhofHpcAl0C+EpBczK5evYrJUyajfpP6cIhz0E+EiyoYtcA6h5mygZlnELAbom4v7t35Ab8e/RAb1i/DR2tk+HiNHJu+exknT65l7luilgL+y24skrtYcZfkXibm7Maj+9vx1boMtGjZGA3aNMDUbSFIwxOYMRZZ8+OYlQQ4zBgbAirH8fRFNAUkJV7H6FEvo1atWnAcMBD3zpyBcPkyxBJYKKpSBjPDa+Uwk6+exDdwCXAJmLgEKE6GgGbT95swcOBAPG/zPPw/8odSpzR+9lsOM2ULECwhAMsEtgeC9gAE7UGIuv2ssXV6n7PbpFIwGwKPwCxMe3H2zHosXjQZNerUgOVwSyTdSWJZKIzm72hWajy3zZAEOMxUDvgoGwuNCIU8BzNm7EGjRp3QpFEjbPv8C+hu3IDu4kUONEUEOg4zJq6t8eFxCXAJ5CsBycVMqVKiXbt2sJ1sizmH5xjfxYz0Rw4zZQszEhiQ1YUKYFLqZglwWBrnrJ0maZF5Nu6d0Gbvxo7tb2HwIBvUb10fo14chTSRW2WqOtJwmOEwkz8IkauZgEWLzqN/v3CWpnlOSBgeXbnKYaaIIMNjZvLVkfgGLgEuAROXAFlk6HXv3j1MnDgR9RrVg3uaOxJvJZbNRDiHmfKBGQkOzGnJoEu7G3dubcGbb8TCon5ttLJrhTnH5kAjavRZzLhFpcpKgMMMh5nCYEYuf4DAietQo4YFLNt3wJmffoaOu5oVOW6GW2ZMXGPjw+MS4BLIUwJklaHX0V+PwsHBAQ3aN4DfR35lV8qDwwyHmfwAi1mScvaxdNHTp3miduPa6DOtD5RaJQeZKoswz+xRHGY4zOQPMyQbyhSnQ3T0X2jWrC+qVauGT5avwKMrV6C7dIkX0SyChYbDTJ56Ev+QS4BLwMQloNVqWbzM2rVrYWNjg45uHTFj+wyWAZd0h2eahJHWOMxwmMkPZsg1Tpe9B99++xL69LFC0y5N4fehH9LFdMi0MuM/jPyIZiUBDjMcZooCM3L5LTg7yfHcc89hqv9E3DhzhsNMEUCGu5mZuLbGh8clwCWQrwSk4P8lS5egQ4cO6De9H6KOR5VNvAxpThxmOMzkBTN6F7M9uH/nB7z0wlzUrlMD7R3bI+qvKGjAXcyMNJdgVvCS+5o5zHCYKRxmBCiVWQgN3Y5q1WqibcuWOLV3nx5miqjQG2b3qmrr3DKTr67EN3AJcAmYsAQkN7MFCxagWbNmcFzsiLhz+tqE3DJT/Bv3f9euXUVc3EI0amSB2bN8oH28HdCabvawvMCiIj5jyQm0+3H8+KeYPt0LtRrVQt+ZfcompZ5Zq/S5Vfyq857DDIeZgmFGcjWjRADX0ahRdzz3XDWsXbESDy5c0KdpvnixyPEjVQ1kuGWm+D/4fA8uAS6BipeAVF+GlpMmTUKDBg0w8oWRT4P/jVpfRtIfuWWGW2bygiUWL6Pdh6++Wop+/azRzKYZxn8wTu9ipuMuZlUHWfK/Ug4zHGaKBjM6qNUP4eiwCNWr18LMycG49Oef+kQAHGYKhDlumal4xYyPgEuAS6B4EpAymWVlZ8Hb2xt1LerCc7knku+nQCkqjVssk8PMTuSlxPPPdulr3uj24OG9H5GmCUetWtVh6WaJ+afmIxWpPPhf+vJU8SWHGQ4zRYcZLcLCdqJmTQu0a90GR7dug0BJALirGYeZ4ulJvDeXAJeAiUtAghlKyzxixAjUaVAHvu/6QvZYxmGmhPeOu5nlFN/yJLmYnTj2KUJDPFC7cS3Yz7GHPFvOHsT85+r5lqokAQ4zHGYKh5lnrmbJyffRtKktatSohY+WL8e9s2e5q1khMMctMyX85ee7cQlwCVSYBCSYuX37NlxdXVG7fm34vsdhpjQ3hMNMSWDm8Q6I2gNYv/5FDHW0RfOezTHu3bFIFVMh15VBSr0qbuEwVwDiMMNhpqgwI5frkJYmYPCgaNSqVR8xEbNw/veT0F25wlM0FwA0HGZK8/PP9+US4BKoCAlIMCNZZghmfFb5cMtMKW4Gh5liwgylYxZzdkGbtRMvvjAPLVs2QecRnTHrUASDGVl+8TKigvlBEuxQH1qSsmuuijofd+ES4DDDYaZoMAMQzKSniwgM/BIWFs/DwX4gft21GyIV0ORxM/m6mnGYKcWvP9+VS4BLoEIkIMHM05iZenXhsdwDyfeTuZtZCe8Ih5liwoyQvROibh8uX9yAuZHjUcuiBnqH9kbCzQSooGK5vHOruQxaRAXUogZpYjrSxQy2pPeU+5tDTW6JVY73HGY4zBQVZqiAploNxESfRcOGndCwQQN8++GHyDp/ngMNt8yU8Oed78YlwCVgqhIgoBFEAZMnT2bZzEYsHYHEm4lQiSqeAKAEN43DTDFhRpe1A6LuIPbueQdeng6o37Y+3NPcodT+NwMFpdejphFToRRUiL0QjRkHwhDy4yTM2BeKmPMLIM/SQ46irDJYcNtPhUmAwwyHmaLDDMlKhFKZAysrT9SsWRcvyOW49uefEK5e5daZfICGW2ZK8KvPd+ES4BKocAlIdWaio6PRvHlzOCY4Iv5cPFSCqmwmuHlq5uIHyFfmjGeUklnUHcD778vQq5cl2g1ph8lfTfpPvIyUJ1wNNWb/HIERL7rDboYD+o53Q1+f0eg7Xv9+eKYbZuyZBlkWJQ8oIyKvMHXeTC0sT1wCyWpWmn8cZjjMFAdm9K5mgLt7OurVa4Fgvwn4/dAhDjP5gAyvM1Ph+hgfAJcAl0AJJaDT6UDWmWUvLEOHjh3Qb1o/RB2LglpQc5gpgUy5ZaYYlhkWL6PdjUf3tyFh0RQ0alwPPSb0wLzf50Mjap49gKQECwoodSpMXOcP26BB6OPsDxc3BXw9V2HCuE/g6/0eXN1V6OsciL5BThj7wVgk3U4qO3/JUqnlpVHpzWRfA4ChPO9SK40bIIcZDjMlgZmw0O1o0tgSPbt2w7avv9bXmylAoa+KxTKla+aWmRL86vNduAS4BCpcAhLMfPHFF+jRowc6uXXCjB0zuGWmhHeGw0wxYIZZZYT9OP33FwiY6Iaa9WvCId4BSff0EPJ0Jp/MedlKTNoYiA6De2DwkFjMmL4LSUn3oFYL0KSKUGtEJCffQ/iM/RjmlIRuLo7weccbisd6EHp6LA4hZS+BJ+ZXmhEhf1VZtgzJD5NZkgb6jFpJ7geHGQ4zxYMZfdxMfPw1tGkzCBZ16+ODV1/Fw7NnIfKsZnkmAeAwU8Jffr4blwCXQIVKQHIz+/W3X+Hg4ICG7RvCf40/FFpFifSNQqeNuZsZdzOT3OZ0lJJZdwhbty6Hq2t/NOnSGN5vef87YIseGK0SC/5YgDaOrTGwXzwWJ16FJlWAUiVCLqcmsKVSKSI1VUR8/DnmWtJpaG+E7Q5hMTakCBf6cPIepZcA3S9BybKIzD0SiUnrA+Hzvi88V3hh/MfjEbolGNF/LoD8STGr4twTDjMcZooDM9SX/jbQhIetbQhq1WoExcI4XDzxO4Rr1yDwrGb/ARoOMxWqj/GTcwlwCZRQAlJGM0rPPHHiRFg0sICbxg2Lby5mOmVJJlAL1E84zHCYkWCGimWKwkG8804SevbshE6uHRGyKQQaQaOvL/NEMZbflcPtBWc0b2qPxKRrUKl0LPUqgcy/lRsCGx3UGh2io0+i/8AI2M6wgSpLzdzUCnwwS6/G8yOQa1mWHDH/xGDs6vGwCxuGro6O6NrHA91sx6Kr3WjYjHTEkGhnTFk/GQnXEop1XzjMcJj59/e9cHnIZJSiGRg9+iU0bNiexc38um+fHmYuXPiPMi+5W1XVJYeZEmpSfDcuAS6BCpeAZJ1RqVRo164dbCfbYs7huWUTN8NhhsOMBDNUX0aXsxuLE6agWbOG6D3ZFvOOzXsWL0MPi07JMlK0cmgFT/f3kJaeDblcyzIVUbYivXJDS4N1pQCZ/DGCgjagVW9LzPtrDjRILZv0fBxh9BKguKZsBeYdm49hi93QrvsgOAxIROiUrYhZcA5xsdcwf+5fmDjuC9j2CIPl4H7wWu6D+ItxzJJTlFkTDjOFK+/FVfYre3+CmbQ0IGTqZjRvboNBffth69ff8CQA+cQMcZipcH2MD4BLgEughBKQ4ma+3/w9Bg0ahGY9mmHCmgn67LjG9s7hMMNhhmBGzN4JCPtw89pGBExwRY1a1eAQ54DFVF9GykJGCrJWgYjD4WjYqg3mzzkNtTobCoUOCoXAGrmWSesS0MgpJatKwJw5R9CphwvGf+GFTCxlxTW5daYMJCAqWJBd9F/RGBrngk4d3RESvAXpmSIrXKjWgNX80KQCGZkilOpHGOX2Btr1sIPH655IupVUJKDhMFOeMCNNEJRkWZ7jLPhc5Gam0QDR0afRtu0gNGvSFB8tXw7thQsQL13iKZpzQQ2HmRJqUXw3LgEugQqXALmaUbt+/TomTZqEuvXqwmGhAxZdWfRMrzTWBDSHGQ4zBDMsXkY4hMO/rIa7mx3qtKiDMa+NhkL3RNl+8qBQAH/w1mA0bd4TKcnZUKnvITHxKqKjz2D+/L8QG3seycnXoVQ+hFJJFhuKnxGYK1pM9En07BeEkcudsATLINPKjPUY8+NIEnhyn1IeyjDmdQ906D4UwZM3I3OpCJlMC5oZl2KaaEmfEYCmZQBujkth7TYAkzcEQa0zyF4nHTvXksNMwYq7saws+jg0LRRkAS1GI4spuXn+1/2zfMad9/VTrRlAJsuGdZexqFa9DjISk3Dzjz8h8noz/3Gz4zBT4foYHwCXAJdAKSRArmYENK+++iqsra3R0bkjpn47FRrSMXRGjJ3mMMNh5hnM/IzPPktDv35d0HpAa1ZfRi0+yQkuwUyWHCE7pqJp8+6YE3kSrm5paNfODY0b26Bhwy5o2tQWXbqMh4/Pu4iLOwu1OocpVBRXs2DBCXTv44fRb7tWPMxQPIkgZ18m+kJJdXPKwE6SCwHK9gx0TRohFbMOzUTvIAc4DExBaroeZCRLWW4lk6BGrRYRH3cNffqGwjFhGBbf1gfpFSQXDjNlDwUEmiqVFhrNY6SlFr+R5VSpJKtp2Y+1qOcgwKLEIE5OKahfvw0iQ0Jx6tBPEK5fh8DjZv4FNBxmSqFF8V25BLgEKlwCEsycOnUK48ePR816+iy5iXcSmQdIQTpGsbQlDjMcZp7BzCGkps5C+/Yt0MPfBpEHZ+uD/yXfRnIzExVYcHoBLNrVQ/VqFmjT1hGjRr2E6dP3IDLyNwQHb4GjYxKaN++LTp1GIDh4I8tepFTqEBFxCG1t7DH5x4lIR0bFuJmJiqfnTUUaMpDJxqKBhl0bmymg6zTTf/SHgWoC+a3xQS/X0QiZtBXp6WSBIVfA/BVaAprUNBGj3F9B30BXzNw/DalCWoEzJxxm8pdnQbIu2jayYIhISbmPKZM3o2fvCFjaTEcnm2noZBNWhDYN1j1nwnGYClFRfzO3QrrHRTt3WV4XWWW0zN1x4oSP0bx5D4x2ccPO7zZxmMnlYkZJDzjMVLguxgfAJcAlUEoJSO5mL7/8MiwtLdFmUBsEfh6oL8Yu6Zel1bk4zHCYoXgZCv7PebwT4TO90bChBQbNH4SFZ2IhWWZISaYUv7K7Mni/54XadRvBx+dtyGTXodE8YhYYsr6QJUajeYCYmJMYNCga7du7IyDgI6hUd+DtswqtBrZBwq0EqKAqd2sIXQOdNx2ZUIhKzP0zEsE7AhC6ZxKiz0dBma1CGtL1RZ1EI5o/S/slLer+TyCMgGzkKyPRx2EyouadhkajT5ddkCJLs+VpGQKC/L9GPy8vBKzz4zBTAPwVJEtjbCPXMPo+JSZdgbNLGqzrN0K6mwvUw4ZA41R4S3V2wLTetujQsi+mTN3OMogVBrTGGHdRjqFPAiAiIvwA2rUbjF7duuPL1f+DePUaj5nJBTQcZkqpRfHduQS4BCpcApJ15s8//0RwcDBq1qmJPsF9EHv+mY5Z6glkDjMcZljwv7gXt69vgsfowahWvRpclS5IvJ3IKsUzM6BAaX4VmLZtGpp17ggfzw+QmnYPCgXFxTxLAKAP/qc4jBzEx5+Hk1MyunbzweiRb8C6hwdcMocgXUhnLl6lfniLo+SLCgZmi28sxsjXXNF+RGs0794JbbsOQJuu/dHStiOs/Swxdo03Fl9dDLWot9SU2xiLei0F9aMvM5RIyU6G+9KR6O8wAwtjrjAXssJm5RnMpIuYOmkL+nmOhd8nPkgV0wq8T9wyU3YWDLpfBDOLE6/AzTUd46074nZ6Jm7J5LgtVxTelEp8GxqKAR37Iyh4GzLSySJSsHWuKCBijD50bWo1sDD2Ajp3dkfTxo3w1pIl+iQAly+zRABVNRVz7uvmMFPhehgfAJcAl4ARJEBAk5OTg48//hh2dnasiKarxpVlXaXYbKZnFqTfFLaNwwyHGaovA/EgTvy2Bo5De6NW01rwetsLcq2cKcdMaRVUiDu9EAPnD4aNVTCUqjtQKHNypWQm5U6fbYmUY7Vai9mRB9Gv/ww0a2+FviEDsfBCLLN8FCX1r1FA4olrHKUpjvgpAt38u6JTd2d4eL2KsLDtmDPnV8yO/BVTp34PV3clLPsMwcA5AxF5eDaDH4VQRtVqC/tilma7qMDo5aPRe2gg5kX+AU1qES0z6SImjPsMfX3GYNL6CUgj6CwgQI/DTHnBTBomWLeHLmMJshUK6BRK6JSFNJUKW8NCYd+hL4KCt5sUzNDfCEoCkJL8CDY2gahZwwKquDjc+esvngSAW2aMoDbxQ3AJcAmYmgTI1YyA5sqVK1AoFGjWohmLzZ7wyQTmblbqhFAcZjjMsExm4s/4ftMr6Ne3C57v+TyCvgzSV2l9osyTi9mMPTPQYWAvBPhtQHqmPhNW/jO1pLAIzOffw+NNtLGzwuRNQSw4nZRgo4BKUY5CMKNVIPxgOLr42KJP/2mInH0IyUnXIZfnQEHB0UqqTJ6FxQkXERLyHWwHTYD9rEGIPEJAU3hWr3K7liJcL8mWYmYmfu6HXsNHINDva6RnUAKAgmblCUCpuKkAV6cM9JvqjNlHIgr1Z+UwU34w42/VDmJ6BrRyOcSiNKUS20NDYN++jwnCjH7SQ60W4OAQj7p1W2DetOk4c/gIRJ4EgCcAMDUtjI+HS4BLwCgSIKCh2jNHjhxBWFgY6jSoA6uRVgjbFoZUIbXAydNC9SwOMxxmtI+2QxR+wVsrEmBt3Q6dR3bGzJ0z9VVadXLmapbyOAUT1wWgnfVQ5rqk0pACXFhAsQCVWsCUSd+im5MLgn8MRLpYfoH/LEZGUIHqrTgsdEbXbuMwf/4xqDVU90YHBauJQ9cBtk5uPUrVI4SG/IDeDuPhonDG4pv6rF7lZkkqArAU9KUmwCAAm3t8DgZEDMOgAVGQKx6wOj8yWV73S5/pLDVNwNzI4+hpFwAXtRuSKNOIqGJJEfI7H4eZ8oUZZGRAJ5cX7aQmDjPMrTFNhJfnG2jatCsCfMfi5+07INy4wTOaGVhnuJuZUXQofhAuAS4BE5EAAc2jR4+wefNmjBo1CnUa1UGP8T0w6+CsZ0BTkiRMHGYKhhmKJxGyqO142liMSU7B+1GGMHNpzM1MOIiEhClo1aop+ob1RdRvUQxmZDoZU2qT7ydh7Id+sLT2QHISpXvVz67mb5kha4fAZvunh+2A7TBf+K31Roa4pPzqy1D65Ww5gr4MhNWAIcyilJZBLlf6Wiu5x07jpaFf/UIAACAASURBVKxrCuVDeHm8CVtvF0zZHFRo7Eh+yn5Ffa4UlZA9kGH8/8ahy5Ch8PH4AJpULQNLfc0Sfe0ful5y+UlNFZCYeBmOQxPQZ8JQhG0PhUpQFwgydG0cZorGFbmfs6K8p3vzLGYmDWSZqUwwIyUBmDJ5Pdq2GQDnwQ7Y9PkXHGYMQIZnMzMR7YsPg0uAS8BoEpDczW7duoXPPvsMjo6OqNu4LnpN7IWIvREsgy659xc7hobDTN7QwTJ8UZYv7R6I2gMQtIeetAMQtXshZu9CZYAaug6CLiFnNyZPGo4GDepiWNIwxF+I/7dl5kEyxn/sj45Ww5G4+AFLG6tXhvNX6EhxJtelsJAf0dNpDCZ+PRYZYubT1Mhlqewzq4yowqIrcXBPG4UeNpOQlHSXub4VVESQQEeTKmB2+C+wcwmBa7oTVLrCFfuyvJZiH1tUsLik2DMxcE8bCet+7vDxXIX4uAssGQBdH1liKJaG4G3+3N/g4ixDjxEO8P/QD8m3k1jmusKsURxm8n/2iwIsBfWp7DBD3zOqNRM5+2eWBKC7lTU+fHMFxGs8o5lhEgBumTGaDsUPVIESIAW2JM1wyAXtL2XLMuzP101XAhLQXL9+Hf/73/8wcOBA1G5Qm1lopv84HUqdUp98qjghCRxm/gszDGRyCGL24t7tLTh18lPs2/sO9uxeiWO/rcGNa99B1O2FwKBmp9lYYPKyFDGYEfbi7q3v4ebSDzVqVMeoF0ch6W7Ss4fpiYVjyndT0NamH2bP/A0ZGZKbGVk6aP2Ju9bTdLbSZznw9/8EXd0HIPzIdFBtF1KCi62gF3MPOodG0GDesTkYFOkOx0EKkFUmb1crQ6WUUuKKSFh0A87DU2A/ZyAS7yeylM6FKfdlfU3FOj59sQUlFvy9AO5pI9DdxRkOzjEY57saIZO/x/SQ7ZgSuB6eHm/CznEaeo0dAr/V45FwJQFKoWhpsznMGD43xl2v/DCjL9S6MPYcunfzQfPnm+IVjQYC1VbJZZ0wVO6r2jqHGdNVyPjIii6BgkCkoG2GZyionwQz1Ie/zEMCdK/ovklAM3ToUNSoXQNWo6xASQEWX18MlVbv6s6sNIW5nnGY+TfMMItLzl5cu/I9tm19Ha+8FI2oeaEInjwJk4ICETlrKtJT52DDhiW4dGkjhJx9Zm2hIRc6iAdw+u/PYde/G6pbVIfPKh/Issi9TMlMffQgqQU1oo5GofvYvnAakgaF8jqSkvQWGgrkpQJ/hkBDyhh9vjDmLIa5JcI2xBaJd5JYUoFimw+LCTKk9OthJhWRP0XAPtQdI51fRcaSwotHPsu09BgjRi9Dv+mDEXd94dN6O8UCihKM26jHZ19uFRKuL8aET/zQP2IAenk5o+/Isejn7o8+Iz3Ra6wDhixyQOi2UKTcSSlWRV4OM8YFGENLTWWHGfqeqVRAYuJd9Oo1GXVr14UiJgaPzpwBrlzh9WaeAB2HGfNQzPgo85fA48ePsXXrVnzzzTf4+uuvn7b169fj22+/ZY3WDbdR3++//57FVpDSe/LkSXz33XdYt24dc01au3YtpEb7/vzzz7h9+zaz/uQ/Er7F1CQgAc2dO3fwxRdfwN3dHRb1LNDctjmGpw/HnJ/mIOVRCtNFSR+lCeV89UcOM89gRrLInP7nK7zxRjx8fbxh138iBg9OgJvbS3B3fwVDh8rRv38wXFxGQaMOx6mT66DN2mO2QKNjaZkP4fDP78HW1gr12tdD4LoAPM37TTSsUyDpZhKmfz8NPYN7onH9nhjt8TJ8fN7DlCkbERX1G2SyxwZAoy/4J5ffgf/4/8HG2Q2+//Ms19gTCWbmHpmNgTPd4OqQgfRMETJ5QVm9SDnVW2YSF9+F6wgN7CIGIeHOIqigzv9LVNHQUtD5nwCNWqfB4tsJmLY3BOPWeMJz5Qj4f+6L2UfD2R8LShpAlpx8/1DkcQ4OMxxmDAGseOv6yQ9yN7O3j0KNGg0QNWMmbpw8aTKuZgQSJW24ePFfWclKalHiMGNq6hcfT1ElIFlJrl27hqlTp8LJyQnDhg172gYPHoz+/fuzmiO0briN1seOHYurV6+y2fvXXnsNDg4OoH4eHh7w9fWFj48PPD09MXr0aHb8t956C1SUkc4rnbuoY+X9Kk4CEtBQUoBdu3Zh8uTJaNeuHWo3ro1ek3ph4mcTEfNnDGSP9BPshlDD9BXJYsNhRg8zBDJCzh5cvbwJatUMDBgwHO7u6YiK+gNqtY65VZFrVWqqDvHxV5gi37mzI2ZH+ODsmQ3IyaIYmmdglJdLlyl+RmmZIf6M7za+gG5d26NFn+YI2xKqV2wpCOuRHFG/R8FrhS9s/Yegcz9HWFmNRteu/ujSZRysrHxgZzcXQUFfIyHhKhSKHBaHkZx8HZMnfYXeg/0xdKEDFl9PKDerjGSZIWtS7NkYOCeNQD/bCCiUWU8ysOV2iXumlNKMuEYjIGruKQxwmYWhyUOg0Or9N83KzcwQPmg2Q5AzV7k0MR3pYuaTlsEAkwpt0vbigIwkYxVUWHQ9AVYjrFCjeiN4eqxkMi5KgojiKb/P7lFV2C9Py0x65clmprfi6q0zLi5q1KvXBtODJuGfn3+BaAIZzcTLl6G7cgU6Wha1Uf8rVyBQ4c/LlznMVJx+xM9sAhKQgIJm3TMzMxETE4Po6GgsWLCALQlEWrRogdatWzM4MdxG63K5HDdv3mRgQuvUj/ZZsWIFvvrqK3z55Zf46KOPoNFoMGbMGPTs2RMLFy5kAMSBxgQegGIMge4XpWymdunSJSiVStjb26NRo0Zo2rUpBs4diKCvgxB9KhrJt5NZuQ2afGXticWG6WeiAkkPktCgSwPUrlsb3236DmQZrAzPw/9du3YVcXEL0aiRBWbP8oH28XZAu/s/lhRyt8rJPoDlr0ejf79hmDBhDdTqByxIlX54KfsONb37FJCeno3IyF/QpElXpGmm4/at7dCZYVIASssM8TDeXZWITp1awXK4JWbtjWAB5PLHcpYub+BsR3Ts4gT3YUsxO/woEhPvISnpIeLjryM0dAv69ZuJ1q2HYsyY1xC14CjmzTsCT8/X0bW/GwbNH4QFp6JY7ZPiKsulcrkiWqesXo9kGL96HDr3dELEzKNITdNnM9MrU7kVZP02TWoOAvzWoteIUfD72BcEAKTsl2o8JrA3yV8myFgCBspSR41dlzSzUcwxSoDEYSb3c1T697lhZoJVe4gZmSw1s6hQoNCmVOnrzHQw1TozYPFrGo0Ib+/laNq0GyZ6++DIjp0Qbt6E7sIFo8BASSwiFLOTdf48rv7zDy79/TcuFrFRX2rXT5/Go/PnjTJ+U7PMSEpBfsti6C9m05WutaBXYdsL2rcqbpOeHXIrozgJssKQS5n0uaFMSLmlV1paGputj42Nxd9///20r7TPiRMnEBISwvqsXLmS7SPta3g8vm7aEqAYGun1448/Motb9+7d0aBBAzTu2Bi9p/TG2FVjmV666OwipNxKgSJLHx9MMEOJA5IfJKNB1yoKMywrmXAAZ/76HIMH9cWYMa8jOfkmS41KbhC5Z4JJ0ZDLcpCRnoOAgE1o0bwFfj74HnIem591Rg8zvyAjYzbatGmG3sG2iPptPiuYOPf4XNjPHIJuXcciYuZ+ZCwRkZoG5utO/u7U0tP1zcfnHbRsOQidrUajfZfBaOfYCaNeG45FlxIqrPAkKe+UYnju4UjYTrJH/15zoVTdhlKRV2pmPcioNVpER/8N+8EzYD/bHgvPx1bY+E0dnjjMlB5acv9tkd7nhhl/6/bQZi6BVqGEVlmEplJja1gY7Nv3NdGimQQz+oxmAQGfoXXrARjp7IKt36yHcOtWxcEMWWEuXsSJ7duhiY+HLDqaxfLIaVlAY9tjYkDLF+Ry7Nu4ETCChcnUYIaUDWkGNfeyMgZhk7JM12moZEnKFm2jz6Xrlj7ny/9KwFBWWq2WyZRiJCiLFbmOUdwLyTknJ+epTEmu1JdeZH1p06YNIiMjWfxMdnY260vbqdF+GzZsQJMmTZj1hvahc/KX+UnA8Pt0//59FicVGBiILl26MEtNLYtaaNq9KfpO6wuPNzwQ8mMISFeNPReLxMuJSLicgIbWDVHHog42bdpUtSwzeoX+KN58Yx769PbF7NkHmGsZKRSScpHXkkAnM1NE06auWJoegZs3tkGXbV7xM3o3s4OImOWDJk3qY+CcAVj4RywUD5RwUgyDte0YTA/bhYxMASkpWjabKmUvo6XeWqXDkiUiRo98FY1bWGGYYhjiLsUhFal617KKsmqQxYFc5bIUmLJxCpp1s4SzYxpUmrsgaNHfU32tFUpgoEnNxqKEc7C3nw/rUb0QsmUKNGJqpbDKlAUYcZgpB5hJugJXl1S4t34ev0bNx5FZM/FrUdrsCLzr7YlerXthUvB2ZKTr4SGvv2MV9Rn97UhLEzF92lZYWrphcH87rPvfhxAJZoxk2SiuZYbcw8gq8+mKFWhepw4mde6MWba2CO/RAxE9emBWPi3cxob187K0RN/27aFMSoJ4/36pr8OUYIaUQ0l5zG9JCin1k5RXc1coDa8597XQe7pekgUpYPxVNAlIyioF8xPMDBkyhIGIJE/Do5Bs6SXBzNy5c/HHH388BR5pH4KZLVu2oFmzZnB1dX16iNz37OkGvmLyEpCeExooxdPs2LED8fHxGDBgAJo2bYp69eqhVq1aLANa7Za10dy+Obr5dUOf0D6sEKeFhQWz+FUpNzMpbmRa2HC4uSkQF3e2CDBDM/laLFsmYtDgDEzwc8WZMxvNDmZYwUzxIIKDR6JhQwu4p7pDdlmO8N3TYOM9GMNdXmB1V/QWqrxjTQj6lEoBycmP0KPXJIxYOgaLby3Wu6oJsmI6LxlZ7X4SFCZ/KEfQ+kmob9kYndv6Y3rYXqTIslgMlEajQ+LiW5g4/ku0a+OGjm6dMXXzlGcZzErohmXkK6lYOeZxdg4zZQkz+hpASUk3MHr0ctSt0xJNGnRAo3pti9jaoYFFW7Rr64SZ4YeQmmqaMEO1ZuZE/oJuXb3Rs1s3vL98eYXGzEgw8+Xbb2NUly64n5wMvPQSsHQpsGxZ/o22v/IKfo6JQYyLC1QyWaWDGVIKDh8+jH379uHAgQM4ePAgDh069LT98ssvTNGkrFIS1JAiYo6KPinB1CjeY//+/aBrI4WKXpKC/ODBA2YloExaFynhA38VSQKSkmoMmKET0v24cuUKli5dyiwzUVFRbBz0DPKXeUuA7q3h3xK6mhs3boBc0NRqNbx9vGFtbY369eujevXqrFWrVg1So2x5VQpmmJuZuB9+4x3h6bUcixdfh1qln7EvaNaSFPwlSwAfn3Vwcx2IP/9cx5IImEsiALpuUbsL2uxd8PIagrp1a2PEshGQX1fA+83R6O06HmHB21mBxbzc7QxlQ9spW5j7sBfRN8gZ4fumQyOYhlWDxeoICiizlMwc2TeyF+p3aIomjW3RoaUH2jYfgUZNuqJpj+ZwlDsg6sR8qHJUUJQgKL6yA4zh9XGYKTuYkb5bNFGQmHgfMTEXS9Ti4q4iJSW7QAuzdK7yXlLNJ4qZmT//d9jY+KFzh45YvmQpUIGFMyWY+WLlSrh06oTzUVHQqlTIlsmglcvzbbQ9W6PB3ogIzHZwgLISwYwEI3/99Rd69OgBmvHMq9FMKSkVHTp0QFBQEHMPkWbWpWOYg2pFChSNl1yZKKUwBSFbWlri+PHj7HNSrmg7QcyECRPYNlKk+atoEiDZkYxLAjPz589nGcukM5FFhiw1lCCArDKDBg3C77//zo5vTs+cdD18mbcEpO+k9OzQd5C+nw8fPsS169dw7NgxZoVZ/b/VeP2N15GamsrirOhvFv0Nov3N/VWkBAB6mDmICf6O8PR4DYsXX2M1UvIOEpcUGL1lZulSEWPGfITh7oPx119fPYEZ8yikqb/uPbh/ezMGDezOCmaOeXUMkm8lw0XlhAFO4Zg/909WKb4wlztyGUnPFBAw7mv0Gu2OyesDTQZmmAL+xOVMqVVCfkuOeX/NQ8A3E+GxchQ8V3lg8ubJiP47GvK7Mihz9PnMDRV3vv5fCXCYkf4WlO2SXCDJ8qkqZqN99PWfynZ8JYUg+puiVouIjT0HW9vJaNOyJV5QKIGrV40SPF9cFzPqnxtmLi5YAFGthiCXs6QL+VGhTi6HLjUV+2fNQuTQoZUSZijtbbdu3disJ9WDSE5OZlmqKFNVRkYGZDIZZsyYwVyHKGCXYhwIak6fPs2Uf3NRLiXFiZQlmgEmVxYCNAlmJLeyn376iWXhojSyS2hWk7+KJAFJIS0OzKhUKpbNrG3btiyVM7mmEbj07t0b7du3Z/eH4irIalhZlNciCbMKdqLvJ8GMNKlA7+meZ2Vl4eGjhyCL6b1791ijzyvLq0gwo3czO4L5c73g5LQQsbF/MZgpWIGneBG9m1nv3gmYFjIG5899b1ZuZgxmsA83r21Er56WeK7ac/B915dlg3BMccAgp0jERJ2FWkPglreLmaTIMJjJEDF54ib0dHdHwLoJ+sB50cSygD2BGoWghOKhAvI7csjvyqF4TPV0lCzGxmxTMOfhCvZfBDHeJxxmyhMS9HVZpJTGxVuW5ziLfi4JZhISrqNfvzA0btgAqoULQdnEQOmNnxSOLAmUlHSf/8BMdDQ57DOYyQ9k6HOCHaEKwAxlFiIXjsWLF4MySFHqXKlRJW9yt6LPV69ezTJVkVVj5MiROH/+/H/cRUxVycgNMzVr1mQKc34wQwo2h5mi382SwAy5FFFq5r59+2L27NkMpFNSUqBQKPDKK6+wWXl6xkihpfvHX5VbAnSPDRs9UxLgGC4r07NQLJhZ+3Ey+vR2w/TpO59kMss/AYAedLRQKh+jcZO+WPXWQty5sx1C9n/TPptifRkak5C9E8B+XLu8Ad27dUA1i2qY+MVEqHLUcM90hr1LKObMOvYkZiZ/WRDQEMxkZIoY7/kxenu7I+T7KaZlmTFU9AloKJZGylNOYEPtyefGU/cr95E4zORW3An6BfZdIBeq4rbCJgykiYPKsqTrpYyISUl3YWcXgTq1a2PhrAg8PnNWDzNGKjxZHLDhMPNfJUiyqJBlRoIZssqcPXv2aUYpQwWCZkMp1oTS7ZIlh9zPyO2DZkslRfa/ZzGdT0gBonFKlpmiwAxZp/iraBKQnoHiWGbo+SFLHxVUpEDwCxcusEY1SSiGorLERRRNgrxXbgkYgo3heu5+5vy+SDCjjx3ZhyuXNsDbayicnJMRF3eaFU+kWBC9kqGfGaV1UljIhSM9/TGcnF5G395dcOLYJ9BS4cwc8ymcqcsimDmIUyc+hrVVW9RoWgOBGwKRqkuD/ydjYTvSA4F+X7KYGYKVgpQokpMmTYTDgCQMiHDB3GORUAsa084ERsUkxSfFIglk+L9iSYDDzL9hhhRzigGhDF3pxWwUCE/7F/Qdq3zbyH0OSEl5hIEDo1Gjeh3MnhqMO1TF++pVsExe5Wyd4TDz35/7vGAmMTGRwQyBC4GMoQJB/elzqvxOMQ7VqlVj1dtJ8ZQUWcOzSPAgAVFefQz7S+vUj5q0Hy0L2jf3efLrL/XjMCNJ2rhL6R6VBGZCQ0Px66+/sudLevak49Eo6d7xF5dAZZRAkWCGWSmydkLQ7ce6zzUYOtQd7u7piIs/wxQTUlDoR5eUDfLxJmVFobiPsWM/QPNmXfH+qsV4cG8HBFY004xg5vEOAD9j7+4V6NSpNeq1t8CULVMYhMw/NRd2oUPhMDgOiYnXn7jdEdD8292M4I5AJy1dQGTEUVj1HIUxr3ki6U4is3xwl63Ki0gcZp7BB/19SEi4gdlzfkXYjL0ImV68Fh7xE+Ljr/3n+1X5AOaZzKS/JTQR4uiYgpo1G2FaYCCuHT8B8do1CNwyowe6y5ehPX8e3Tp3ZtmaaGaaAp/LS3EjZZFehpaZpKQknDt3jsGDtF1SIGhcND7Kavbiiy8ymOncuTPOnDnzFHoM+1L//JrUz3CZX1/Dz3OPSVJ4DfvkXpfOQZ9Tfw4zkkSMu5TuBcEMxb0UJTWzZJmZOnXqU5ghGDW8h8YdJT8al4BpSaDIMMOsMzm7cef2Vqx4cxFGjPCCg0M0AgO/QXT0aSQlPUBy8kPELryIkJCtcHZWwt5uOJSKabhxfQt0zL3MfECGAE73eDuAX7D5+5fRoUNLNOneBGE7wxiEyB8rMO79sbBxdYWX10oolPeZpYoUELJM6Zse8lLTBCQknMcAuznoO8UBEQcjoNSpTNsqUywbROUFktJcGYcZvWJOQE+THMFTt2BQ/wgMsvGGc+9xGNbLp9DmZOsLhx5e6N19PAIDvwYp9pUbYAxhhtb1Fik3twzUqdMCU/0n4MLRX42SnpksO8VuT+rMSNnMLvKYGabY08+6IczkZ5khRVWylFy9epXFN0iWmcuXLz+1nJASKr1u3brFUj5TgbvNmzezmBtKgywpqlI/Whp+Rsen9NC038aNG7Fr1y5WHV5yOcq9H7m5UdwLBfVTgUXKVEbZjnKfi85RXjAjXU9pl4bXaurrpYGZkJAQ/Pbbb+z+0HHM9VXa+y3tb67Xz8ddfAkUGWZIude7m+3Bzeub8eknKsyYHogRw4Pg6hoFV1cFXF1VcHWNg7tbMAImjMVbb8bi5o0tELTmVSiTrpUaKxaKX7Bx4wto374FWg1oifD94U9iSVRYeC4WI5aMgo3jcHh6vo7oBX9AodBCrRFYHA0lBpDJHmFWxAEMdYhDjzEDMeWbyUi5m8KtMlUAljjMSDAjgNzEPH1WYVgXByx2GIRV43zxtudovO01psD2jq8XXhruAod2Vhgz+jXIZDnMCixZLSo/2OhBcNSol1G/fntMGjceZ37+xTgwQ9nJituuXGFFM794+22WmpnDzLM6MYYwQ8HXhpYZSbkiBZMagQNVeKeq3TVq1GBpUqmat6TIUn8K1qYaLlQQkVIcu7i4wM3NDTT7/tprr7EsaNJx6adfWqeEAwQjixYtwrhx41iRRGdnZ3h4eLBsarQvKbxSf7KwfPfddyBr0qRJkzBixAgMGzaMJSaYOXMmS1ZA1yL1pyWNszwsM4bnLM168VWjittDegZK4mZGzwaHmX9bMivuTvIzl6cEigUzhkCTk7ULx39bg9UfJEGWEorIWRMRMXMCEuKD8ebyGBzY944+c5luL4MgCRDMaamHmcPYsGEZg5mOrh0w56fIJ8Uu5VCJKkT/FY2Ry0bBZtRQDHaZCx+fdzA5aD1Cp2xC4MR1GD36JfQdNgm9/OwRtC4ISbeSOchUAZAhiw6HmWcwk5YqYozPu5g0wAN7w6dDyFwKQamCqNbk2wSVGmJaOq7GL0RI36Hw8ngTKVUMZiSrlofH62jUyBIBPr74Y//BUsMMQQy5qekuXmRLWi9Su3wZj86dg6FlRixCNrOqlpp54cKFzN2H4mKkRpYXAp7du3dj+fLloPTNderUwZgxY1hxSYpxIEWWGrmhkWVkypQpDHjs7OwYmPj4+LCMVVZWVixjFVltpH1I2SfXtbfeegtOTk4sHW///v3h6ekJ2o8qv9vY2LC2jAqcPinYSftQRrWOHTuyDGt+fn4IDg5m8EP1YyhD1quvvgrKxiaBDJ2zvGCGzlUakJH2LU/FqjTnksZLqa2VSiUo7fKRI0eeyt7w2GTlo9f69etBxTBXrlzJAv8lmRn2NZd16fqNsTSXa+bjLL0Eig0zz4BmF0TdAdayHm1n6YuvX9mAR/e3QhT2Q9QeZNYNlt7YjIL+DWFLgpn165cymOk8yhLzDs+FSnIRExUMaBKvL0bgNwGwj7JHD99BsB05Gr3dPdFrlDt6+PWHk2IoZv0SAcUjxbOsYFVEoS+Nm5a578thJhfMeK9CkP0Y7J4RCl1qOrRUOJHqk+TTSAEW1GpcionGlD6D4TlmeRWEGX2tGW/vFWjc2Bp+nl44vmcvxJs3IVy4UOJ6M1eOHsU377+PT1eswNq33mJLWi+oraW+K1fi83feQWp8PIZbW+MC1Zl5AjOiQsFqzeS1rCqpmbt27YrnnnsOZAkhoCFFVFJIqc7MvHnznkICgcLEiROZ+xeBDClvkhJK7l4EMgQY/v7++OSTT1jhu8OHDzOFlaw0lL3q888/Z/tJwd6ffvopqy1CdV+mT5/O9jt06BBThsl9jNL0krWFlvSicxIQkeWHMrBRRXAqqkgppMlFbc6cOWjZsiWDHHJVo/5SYoCyhBk6NhXdpOuja6frKk0jZf/UqVNPr7k4qpMxlOqSHIPcAaW03lJKZen5kI4nvafaIQTNd+/efRovJvWpqGVxZUz9KQMgZfkrzb2W9v3ss8+wd+9ekMWTXiSH4r4k2RV3P96//CVQIpiRFH4haweErJ0QtbsBYR+DGOj0LmVUm8acQYauUYKZLz5PRdu2zdF5dGfMOzzvGcwQkIhyKEUlNKIGyY+SEflbBAI3+GP8J16Y+kMgFpyZD5VWxbYrBH3KY3NX0vn4iyYB04IZfS0kmukvSSuNOxfFjzHLzBOY2TMjFEJaOghWCgqAIQWYaphcjpVg5g2kpGSzTIlSTFpRrsWcXdLoOineaKzvKjz/fHf4jh6NIzt2lhhmJLeybZ99hoE9esCzVy8EDhyIif37I6CAxrbb28Ord2849eoFBzs7ePXsiXNRUUWCGQamVaDODLmNEcw0adKEgQgF9lMjcKEK7NWqVUPDhg0Z7CxdupRVZyeFiZRSetGSrDLp6ekg64uXlxe2b9/+dDv1IcX1q6++QqdOnZjVhfoTYBCAjBo1Co0bN2axOFTPRlLGpCVBAsXSUOpe6UWfkaJPS6kfjYMaJVMg97TmzZsjISGBjY0+p/NRfwIkY6ZmpvPTi9zwXn/9dXh7e2P06NHMekXL4jYaO1m+KJaE3ProJZ2DvSnCf9Rfkkd5Lg3Hxi4LEQAAIABJREFUWdgYDC+jsL7ldQ2G4zccX17rNCbqT9ZIqpND94xace+39KzQvmSRJBdNyhJIr+KMRxoj7cNb2chAkrGxlqWCGQlqaMniabJ3mj3AGF6TBDMr31qEVq2awsrDCvOO5IIZciei9MWCHEookSqmIV3MRIaYiXQxQw8xT1IcF00F5r0qiwRMC2YoXboOqmI22kepoBpK/87SVxy4MQbMBPcZAi+P5ZDLs6BW0XVoi3Qtxhh/ca7V2H1JdpQtcvz499G8WU94Dh+BQ1t+gEjuRSWwzEgw8+3q1RjcoQM2BgTg75gY/Dl3Lv6eNy/fRtv/XLAA302ZgtB+/ZiC7ta1K77088PemTOxJywMe6dNy7ex7eHheMfXF4F2dtCQNe7ePejOny+xdYnq47D01CaUzUyyzJAyRYoUuWdRI0tIWFgYWrRowWrLvPPOO0+VJFLk6CUpWwQltD8BxEsvvcRm3KkujWE7duwYs+oQIFF/2vejjz5Cu3btmFsaJQog4JAanYPWpXPlViBof6kP9SNAIlihWJnIyEgGLBRPQ5YC6kuWoLKAGWlcNAZKPkBJCwioStp27tzJ9j1w4ACTk6GcpXMVtKTrJGsHASBZPmjJW/4ykGRErov0fBT1JT375MpIFrmS3m/D/ejek5WRrFr0ks5RnDFJzzl9H3gzjgzouSBZFvd+FHbfjAYzhhBQWdYlmHnt1Wi0bNEkX5gxVL5JgZXpZE8bvTfczterjgRMCWZksvtITr6GpKTLSE6+jKQiNNaP9b8BuTy7ICNKgdtKCzOXyDLTezBGj3wBixdfQnJK8cafknIHcrm2wDEaG0KMdTwJZgImfoxWrfrDi2Dmhx9LDTPfrVmDETY2ODl3LrB0KbOuIDUV+TVyJRMzM3E8OhoLnZzQp08fDLOzw9Bu3eDasydcevZkS1rPq9F2d1tbDO3ZEyMcHfHuK69AvH270sGMVDSTXMskyKAfbWoUAxEQEMCsNkFBQUw5JmVJ+lEnBZ7WaXaaYmTat2+P2NhYFoD//vvv44MPPgAt33vvPRZzExgYyKww5EpD+5JLGFmE6HOCHfosN7zQ8ekz6ZySgkB9pfPTmCiDGVlHqGr8ggULmEVp7NixzKJD+0pKnrEtM9J4aEnnMWYzPHZh69J5Sbkm9zqKMaI02i+88AJvBciAZESyWrt2LXMZk+RYmLwNt0v7GGtpeOzirJN7GsEQARJBNcERb6WXAcmTXFjJjdKYLw4zBcTzSDDzxusxaNmyaDBTdVR1fqWFScA0YEafHjwoaAOGDElAf7u5sLOfj/528wpt1K9f/zlwcUnDvHmnoCyhdaa0MHM+Jhp+Nj3Rrt0w9OsfCTv7wsdO12dnH4X+/ebAx/t9xMVdhEqld7EzFmiUx3GewkzAJ8aFmQ8/BFlWjkZEQEhNZfFLLK6F4pTyaBTfpFWrcWTOHMQMG4ZZ4eE4fewYju/aheO7dxfaTuzeDWon9+zBmUOHcJfiFy5fLnXhT1OzzEgwQ6mZT58+zWanaSZSAgRS8sh60rZtW+ZGJUEE/ahLYLN69WrQcWrVqoVGjRrh+eefz7dRbMyWLVvYvuSSVrduXRarQ+fOC2byUh4kiwzBy5UrV1jiAoqdeffdd1mWNYr/Idc5X19fBjekZJY1zNA5aPwkO2M0mgnODXZ5yUL6jPpS++OPP0D3kuC9X79+bEnrRWmUOCG/Ju2f3/by+Fwag7Q0xjkp4QQdLzw8HGQNo/tYXLnTvTLGPZe+dzSG4ryke3/06FFmTa1fvz77HtKSt9LJoEGDBqhXrx5sbW1Zmni6N8W9P/ndSw4z/8/eeYBFca19PLYEUbH3ig17lyKiAoKKDSui2DsqVYqU7YA15abem56Y5ObeJDfJF9NMs7ckpvdq7C3FRtmd//f8zzK6QXZpiy5y1uc4szNnzpx5Z4D3N28rBczcJ2FGWpfKIQFXgBnGlBiNwKAhMRjaohVWD+gGQ8AQ6P0GwDBsoN2m8xsAY4APFvdoB89m3TAn6nXo9eVzNasozPwUF4dpXl4I69gWOl/H81avScx/lC9Gt2yMAV7TsDL6EExGa/2nGwEhzjpHZcPMZytWANnZAmAcma4E4JhM+HT1aiQyuD0xEaD7xtmzwPnzZWvnzgGnTgFOKPrpqjBTtGimqtT99NNPAjYIKlT6CB3qvqIw4+3tjfXr14OBzIQgLtWmfmccCLOkcQzGCxBm1q1bJ9zDSgMzPI79OC+ei+dkTA+zrLm7u4vGmJgbDTNUWDgvZ7n2ULZlUapVhYny4bF0VeKb5LI0HsP583gu+V1t6jjquLZ92b+yGhX8onOwnQv3sY/tvG2PUbcXnR+3q8dxnfeOsivrh/dIlRfHqWgrzzxUmGGKa1o7+eKBVlIm3JCt/DKgHPkip3Xr1qIQrPq7rzzPSXHPlYSZUsBMWdzMSnpbL/dXHwm4Asww1sVkBAb7pCBxeCh+TEmGectdMK/fCPOGTfYb9999Lz5btQKBPYMRNW87DDcJZn6IjcW83n3wr0mTYc7ZAPPGzfbnrV7T+g0w/+N+/GviWIweuhjLoz9ClqnqwsxMZ7uZFVpmygUzI0YgITYW5vPnUcAq90ePlq0xxsMJIOOKMTOqZaYozPCPr6qo0WWld+/eAhqoLPFDpYtKHP+wM96FVgDWlKHLmapc2Vuqf9iZTpkgwmxlDOjnmDymuI86FvvwDXqvXr2EJYjzio+Px9atW0WNG7rGcTwVZpg4QFXwqbxWlpsZx96zZw8efvhh4VL34IMPiiVTWpel8Ti2Z555RtRfoSycpTwVJ1e5rXwSUO8JoZqArt63stzron0feugh8bNEiyM/6jlKO0P+PDJOjS8LaLGUreIyYLwZ5Uj3Tf7uceZHwkwpYObRR9ahVaumpYqZqT6qurzSkiTgWjCThFU+gfgqZjXyjUbk021Iq7Xb8rg/ez32LZqHgO4jEDXv7ZsKM3N79cJ9Y8fiSmYmCnQ6u/NWrymX/XI24B+hozBq8IIqDTPMZjZlymNo1qwXxo8ejYNOSADwhhNgRvnjjwrHvBBIKtKqimVGVaYIEWfPnhUuZiyWybeU9MO3hRmmpyXIMLifChnjV9S33uobcfVNOMdVgYXJAlq1aiXibQgCHJONShybCjCqUsfvVNaGDx8uQIYWHdbBYbY0vqHnuZi2OSkp6YbBjDo3zosZ3YYNG4ahQ4cKixGXZW0+Pj7iWLrg0X3PVl7ii4P/VLlR+WIBS51OJ1Jtc8nGuCjOkU1NwW3bhzVfmGKbcmRKbe5LSEjAihUrhBvW8uXLBSiyuOkrr7wiagRxOup5bdfVbUWX6vSLbi/6nfeaMVxU+BmHRWBdvXq1SBXOdOFsMTExYtuGDRuuFlXl80CwYJpxxsMQNji2+lypS75lZ6wM0yJT+ef52K+0H7U/45MY80ULIe9dWe+3+qyox/PZVQu+8hxl/fAaeJxszpGBKs+yPBulvWcSZkoBM6+8vAHt2rVA57FF6syUw/WoJAVY7r91JOB6MDMK38SugcVoEm5FxdUDUbcxla6SvR4HXAhm7h87FoQsdY6OlqKGzfqNuPcWgZnJhamZp4wLw2c7d1U4NbOEmdL+iSy5n6okUfFzZJnhSPwjTuWWAfp0C6O7GQtWUnnndiqHXGZmZgqXDMapsMgmjyva2I+KNt88cx+Dlfv16yd80qk4M3i/uA/h6JdffhHnYh9mTbvjjjuwb9++q+5FHJtj0hLDbGZMKX0jEwBQpoQ+gh3nWtHG66SMy/JRFVimuGZBSroaMR02l2x0m2E8U5MmTQSUqtvVPkwLzdgjguhzzzGBRytRyJR1g+bMmSMa6/tQ8WY9IWawo1Xu0qVLV+91SfPlNRFsea/sfdTnhjEghGQmiaC7DzPvde3aVWS/Y0pxLy8vsc77TIsgP1wy3Tdjt5hinMDLD59Tdcnx+Yz6+/sLSKN7lio70akM/xGkCV0Vvd/q8cyuRvmU96PKrrRLyqWqNd6r0l6fM/uV957YO07CTClg5rX/22gtmhnqiTUf2xTNlDAjJeBAAq4HM4H4NnbN1bogjmIkRA2YnMqBGbVoplnDgpk6u41xGsyiRTczWmYIM7QoQa8XQONw/oSeWwhmJkx4EI0adRFFM7+qQNFMNTWzhBl7fxLLvp3KAD9MJUx3LcaYsAAl3wgXVeqoDHAbMyX973//E0oig2LpTkUllhDB/YSdKVOmiFTOLKz55ptvihTBPJ79uJ/pnkePHo1vvvlGHEOlmW/PVaWageu0CBB2qCRSweM5WceDrmNUumh5oWsZ58w387QkcBzu4/z5lr5nz55i/7Rp067LZkZll/E1VOB5LOeuXgNT7PIamKSANXXK8nGm0mQ7VlnmoPalvHlv9+/fL1zyuGSjFYGptik7Wjyo0Nv2oVLPVNZUpln4k+AzZMgQIacff/xRjEkAZj/ef96HgIAAvPPOO6UCFMqaboo8L+NeeJ3FfdTr/+yzzxAYGCgsfrwffDZ4j7i0bXyemFqZ43MuLNDKuTNom+6PHIdjcj+fE344Bwb/L1q0SFwP97GV9aPO1dnLss6jvP2dPe8bNV55r9eVjpMwUwqY2bZts4CZDiPbI/rDaBgsBlFb5taxIcgrqQwJSJix1sS8lgDgMcwZGoZ9y5dA2bAZit4IGE12m2IwATkb8HN8Iub37lNNYYYJHBSEhd0HDw9PREyajO8PHIRy9myF6sxImHHen2FVaWPmKxa6pBWDBSZpVShOqaOCQoWfVg81HoVvxqkI072LCiKXfEvPt+SEE1pcWAQwKipKbKMLDo+hcqyeh8exQKBWqxVv+hms7OvrKwpvsvgkFWUqzJwjgYZzoBKclZUlAv1btmwp3q4Tntj4Fr9Pnz7C8sBropVIddnhsYQeZlKjdYkucVRyeb0qzDD96qRJk8BxWXOnKn+KKpWqrCnjWrVq4f333xey5PXb9uU1Ux50v6KMeN9oLeDxqpy4TqikqxrvKeOt1HuqPltFZcdz8Di64THGhHCsnru4vuzP+8N7SgBjIgl+OEZxH47FRphhNjtCDOfFe0mXNI7H/erxhBn2YS2lilhmipuLq2+jLPjhzwN/Pgj1tOZ9/bVrN86RL0Vo3a2I9cpV7o+EmRJh5mO8/voWATMtBrXA0n1LoVf00CmVof7KMW8lCUiYuQYzWVkKJkx6FAM7DkVUby+kDfdDqs9gpPoOsd98hiBtmA9W9O+DIc2aVkuYYQIHZqMbM+ZO1K/fDnOmTsWvHx+WMFMYa+MKMTOqMkM4oTJH9zGmNWawa3EKpqoIElgYYD916lThakYFU1VKeRzXWUOGSm5ISIhQdD09PYX1h64/dEWjsqm+lee4bIxf4PnnzZsnFEy6MPE4Kpusb8MinnQ5UpVRWmwYy8HUvLSi0P2ISvKCBQuE9Uej0QiQIiQxcJfnoBJLZfzjjz8W+1g/h+dVx2QfJiHgMYyBoJtVVf+o8uWS10lwZMwTQY8WKt5Pbrftx2tWYYbWDbqUqTEllCH7cj/XCUSsL0R50WLCsdiK+/A4HkNYtX1uuL3oR52PapmhWyHhiserVjius6nzV8/NBA+EGVoAOT8CNd3ROD+Oq7q4EWb4/FRnmKF7HH8mg4ODRRs9Ohiu20aLORJuGdtl73dV0WfJlb9LmHEAM+YrHwD4GO+8fQ86dGiJRt0bYuHOhdBDwsytBB2VdS0SZlSYUUSNl8VL9iFwpB7Dh65EoM9qjPJeiUDvaLuN+4P9VsFnwDx0ad4D948Lq3ZuZkzxzJTYgYHZcHNrjnkzZuAY3TzOnHGKZeaT8tSZKcxmJhMA/P1PO2NRCCdU7AgIVG4dfagMUpmkBYMZzmjZ4XdV+eSSii4hiUkCXnjhBTz77LPiTTz7Mw5E7aueh9+piDKWgsDCmA26OFF55Trhg7Eo6kc9nu5O3M9+TPtMFyoeTzcpWgkIVaoLGo9Vj+N5eL28hqKxHoQxvv1ltjR78TvqPKrKUr1uypjKqz2YsZWRLcwwQF2FGW7nOCpEMAkAXbnmz59/1crF/cV9OA8e5+fnJ54LFYK5vehHnbMKM7TM0N3Q9txcZ1OPV78TZuhmRngmrLJwK6GMSQwIMhJmrAk4KDe6ItJ6WatWTbRs0Qi9enZ0ydazR0f06uWJ5s0aonbtWmACCv6Mq89h0eenqnyXMOMQZnYA+BAH9z0Mz06tUbdNXURtj7JaZaRlxkG0SGXhQdUaV8KMFWbUmivp6X8hIeEXrIn9BqtjSm6rYr5GbMI3mL94Pwb1nIIHxo2vZjBjreuj1RbA3z8dtWs3wOLISJz56msop0/DUo70xldjZp55BiE9e+Kb1ath2bQJFpNJFM9kAc3impn7meUoLg4pgYFIjIuDhJnr/8yrSqCqFF7f4+9b2M/eMdxn22yPVBXP4s5je4ztfq6rx9mOxfWi2237quvsU/TjaB/7qvu5vFU+6jUVZ5nhPvVa1XVbmKElhfEo6hhqH1qxaEljkD3jnuiKRnmr90VdV5dUPAkShBlaZhgXpcKR2kcdW10SZggmTFhw7733CtCkmxEhVW1UyAlGqmKrwgwtM4RqAjfBhgkE6BZJ+OanOltm1HukwkzjxvWxcOFEPP/vHDz/7yyXa/9+zoTnn78TkbNC0dCjnoQZOICAW2WfJXcngIP46bv/oFvXdqjdqDYit0VCZ9FKNzOJMiVKQMLM32GGFgbGf5hMpWxGC9bnWJCUcgbDBswvhJnMapQAQGGuA2RkXMbQobGoXcsNqxbMx1/f/yCKTpanVosKM9uefhreHTrg5WnT8ElMDD5euRKHHbSPV6zA4ZgYvDBrFhYNHoyEmBgJM8Vo57aKZDG7r9ukKrWqQnRdBxsriO0+R/3ZT1VguVQ/6rlst9nus91u21dd5zmLfhztY191P5e3yke9pvLADF34WPOGVjY2Ws2Y+njJkiUinmnx4sUiGJ8wwfOoTX2ubJeEGboD0spiCyBqH/VYdUlLG13GmLCB7kV0CaMViI3rdCtk4gpCD8/PpsIMXadoXSM0sQZRo0aNxDGM9eH4EmYUUdGelpm2bZrivnvXQjEfgFKwz6WapWAfzHl7oFi+QLZpJZo3b3Q1vk8F2Kr6cyotMw6gzJJHmNmHMye3oadXR9R0q4kZ/50BbZ5Gxs2UqMpXLStKZczWFWHmampmZgVTTSbFLJ2Zmvna8AqYDKC0TaMxw2CwIC7hGHz7R12FGaUU2cxuhdTMWi3d84C0tD8wePAy1HVzQ0r0SuT9egRg1fdy1GhRYea9//4X/n37YpmfH9aFhiIpOBjJDhr3p4aGYsXw4Qj38UEOM80x49HRoxWqE1ORGjOuUjSzqv7xl/MunwSovBMYSgszhA41AQDTIjNlM7O8MbkDEzPQWsK0zIy3IHCoroacHV0VaaWhVaRoY7A53dYef/xxUR+IsGHbh25wf/7551Xrji3MEE6YeYwAQ5AhRPE746MYwF8czHA8XjfPwfnyWpikgjFbdIPs37+/GKu6JQCgTPhMMEMdYaZN66a49554WPJ3gy/EXa3lXXwfFsvHMOqWonkzCTO4Vawvjq5DIcwoe3Hu9Ovo27czatasgUkPT0TmXxkSZiTMlCgBV4OZ1b6B1jozpqwS68yItMg5G3Bg8XwEeFWsaOY1mHHIT9exFaGHSZDiE0/Ar/9cPBhmdTNTYcZRnRkBYxs24t4xgRg1ZGGVLJrJ66clKyXlFAYMWIAmjRrClJJSLoixhQYCzS+HDuHBzZuRo9MhW6eDSastsWXpdKL/g1u2YP+bbwKnT6M81iHbuVR03RUSAJRPJZZHVVUJVARmmJnOaDTi7rvvBoucMr02YeTQoUMi5ogy4fhqY5wKrTdbtmwR7mdcsrFAJRtTYhNImEJb3ab2oSsZLSZqXIsaM8OsarSuMF03LS206nCdjXFf6ht6KumqZYZuZoQ3ftjn5ZdfhoeHB2bPni22E2aqawKA4mDmvn/EA+Y9LqknF1xmLPhhmPTLJMw4AoBbaZ+AGcseXPxzO/yH9UGdOrUw7p6xSDufBoNikK5mJarzlWHvqDpjuhrMrPIehS9jViHfaBSxJ/laLey1PK0G+dk52LdoLgK6uwLM0DIThiuZGcjX6ezOW72ePE0m8tdvwD1VuGimCjMJCT+jT59ItG3VCnfRVHPqVMWtISdPisKbCt1E6MNf2sb+584JkEE5YnYqCi9Fj5cwU1WRoOrOu7www8B5Zitj5ih1DBVauOTH9jvXGctCSGEqbRYv5dK2sRAnrSTLli27ul3tFx8fDyYVoHWHY6kww2xmdHFTzydWivxHBb04mOE4jM2hxWfChAnC3YwZ9QgzrDNTHbOZSZgp8vDcpK/SzcyBmxlhRsnfBXP+boSHB8Dd3Q2hW0KRdiZN1JqR6ZmrDljcjJm6EswM9U3GtF5DsHVqOD5YtBjvz5+HD+bPt9vemz8fO5YswwPjRmOQpy/mznsbBr01IL28lpayHmdV5oGEtScwtO9sxHv74N25c/HBggV2561e03vz5mHnsuWIGdwX/gPnYeWqj2EyWV3cyjqPm9Vfo6FlSkHMmi/Rs+dUdOnYCQ9t3iIymVXUIkLrDF3Eytsqev6iUFLe7xJmbpLmUI1PS4WeCmxZ3cwIM3TFYuY4Wj9oMVHdubjkuGX58Bg1ZoZWFXvHczubLczQ7Y3bVKuN2odLfhzBDI8h0NBqw7ozdFF76aWXBMzQSiTdzJpCWmbK8iQ7p6+EGQcwQyuTJXcHoBzE/Pnj4OFRD0HGQKQeT7UWzmQiAPlPSsCOBFwBZtS4i9DQe+DZfgQ6teqHLq0HonOr/iW2rm0GoUOLPvDqFo4lS/aLFME3UrFX556Scha+PvFo36Jv4ZwHlDh3Xl+3NoPRpnkv+PkmIy7uGxiNVQ1mzGB9nugVh9C923j07dETTz30kLCM3OxYlfLCh7OPkzDjHEVAjlJ6CVDhV2GGRTNt68yoUMDR1HUq/2rMDF2xfvrpJ3EyjlGajwoYtn25jTBT3tTMTL/ND6GkuPEdwQzPq+5n7A+BJi4uTtRBIswQmtT9tnO+Vdd5rZShbcyMhJkbf7clzJQAM+YrhJlDWLVqKpo0aYDBKwYj4ZsEGC1GUFmVMCMlYE8CrgAzKnykp19Gcsp5JCSdQcJaa4tfewb2GvtwX2LSGaSk/gGNJv+6mBZ17Mpe0kKzLu0C1iadFXO3N2fb7er81yafFccyvXFlz9PZ4zMBQna2goUL30anTqPgP8Qbrz77HCznz8P8228VdzUrRwIBZ8NIRceTMHPjlYbqfkYqrlRgaZlp27YtateufbVoJvepcKCuqzBD9y66YjEOhh+OUd4Pxy4PzDCLGV3TygIzPCY0NFRcrzpvnpsfupexKCuLrdavX1/E70iYkZaZ8j7XFTlOwkwJMCOCpZSPceedsWjbtjn6zu6D2M9iJcxIjCtRAq4EM0zxy0rypnI0o0FkQ76pMMBQkXLN3QiREczZoHEjxiPM0DIzc8ZzaN1qMMYFBWPnttetMHOTs4hVFEKcdbyEmYr8+ZfHlkcCBAmCCLN6MQi+Vq1awuVKjU0pDmYYcF+vXj0MHDhQvMHneW80zBw+fFhYclg0k+mh+aFlprgP58bGbGXDhw9HSEiIKBLKvuq81eXcuXPRsGFDYaGKiorCJ598Ivqo+4sb/1baxuvkPZeWmZt7VyXMlApmDuOpJzPh6dkaHUZ2wNJdS61uZoq0zNizSsjtOmG5M8CAlDOp6BLSBbVrNcT4sH9Bp7OI+iE63Y2MQeG5mBq57O3GztNexrOyz1u9VteYv73rsr9djZkZH3YvGjfqilmTw/H5rt2wnDsHs4QZYZmSMHNzFYjqenYqr8zqxUB6gsGJEyeEpUQFGVu5UNn95ZdfRHHLbdu2iaxh3F9cX9vjHK3zWFpH/P398eKLL4qMZKpSXdxx7M9indu3bxdzZsV3fuzNgdvZeF08hkH+vF7bY1RYoRLPWjcEtr17914tCmpvbDHILfSfKncJMzf3pkqYKQFmrG5mH+Gd7fegR4+OaNa7Kea/NV9mMyvRLiFxxpUsMzfCkiDPYR9Myi4bK3zSIhUQoEPdui2xbE4Ufj38CZSzZ2FxgUxizrKuVGQcCTM3V4Gozmensk4F/9KlS1etFfbkQfBgX9ZkcYaSzzE4Zr9+/fDcc8+VCDOcF5VuWo84D9VNzN581e3qMarVSd1uu+RYvC7KgTVyVMix7XMrr1cFmGGtG+qy9DTKvfAezOaPZJ2ZWyn1cmmuxSwSABzC558+jf79usK9TV3MfGEmdGadTM0sgcahBCTMOFO5r25jKcJ6x1ilQYOiUauWOxKWLcP5b7+D4gL1XSoCIM48VsLMrawmuv61UZEtDRio8MG+zoIZnnvdunU4ePDg1fTLJUlMnW9p58B+PIbN3sf22tivtGPbG6+qbVev2ZUsMyq8MIEVM/JC2Q9FOQTF8hEUyyEoyrdYnx2NFs0bIzU1FSzAymfT0X129fsiLTMlWGb4UEA5gCM/v4Qhg7xQ8/aamPivici8klm1CmcqVrcnjVmDymhU3IvaYrSKFqJZtMLlin3sNY1FA63Zur/oOFX1u4SZ8gMIg/4ZM1LeRhez660hdFWzZjQreWk9N/vdDDc1zp9WmXXrfkef3rNQr647jElJyKV7yIkTN71YpTOBpCJjSZhxdRXj1p4fFffSKu9l6VtaqVEJZVrm0iqh5ZlDaY5R+3BZ3T6uBDMCXljs3bwXiuWgaHm5O3Hst1fx0YeP49VXN+Nf/0zB5s0JCA4aLMqNJCQkiKQUEmZKgIHSWD9cuY+St0tUcr305zsIDRmCOnVqI2RTCNJ+txbOpLJeFZRtPfQwKSZkK9mV0oyK8To5MF6E20tqnFeWkiWaGKcQvKqCXB3NUcJM+WCGyQpYX4XB7+X8Wi2LAAAgAElEQVRtBIGiMMNtRqNSYrM9N/szxql4OLr+HEXPWd7vhCjOIyHhF3h5TUTr5s1xf042LMePgzViKgIAt9KxEmaqm+oor9dWAqoibbtNrt9YCaj34GZZZlgPUUBMwW4olgOwFOzDmZOv44MPHsS99yZi5copCB09FMOG9UH//l3RrVs7dOrUGo0aNRBJG9asWSOSF0iYucVhhqBFM51SsAcLFoyDR0N3+KcMQ9Kva6tErRnCFkEmMzcTi3cvxuz/zUbkS5EO26wXZuFqe3GWw76R/4vE3G1zsfqL1eI8qiWGgLJo5yKE3hkq4C9kcwiKa9wfdn8Ypm2dhoXvL0TikUSoEEQYcAQLrr5PwkxZlX2ra1V6+gVERLyAKVOeEG3q1CcxdepTpWrTpj0N9l+16gtotbmFVhVaZMxYuHAXxo69GyEhGxESsgkhIZuLtE0IDd2CsLD7MHXq01i48D0kJh6BTpcLg+FaAoXyAkpZjuN8CXLLl+9Hp06B6OPVA889/LB0MSsCchJmbqziJs/mWhKojpYQ17oD1lgk3oebATMCYsy7oZgP4NKFD/DRR0/hgfuTsHjRBIwI6C/ApVHj+qjToA6a9mqCToGd0Htmb3gv90Y737aoXbc2Vq9ZLWHGlS0qzpwbHxhYDkGjXYQ2bZqhX1RfrPlkTZVIz0y4MMKIpNNJGLxyMNr6tEUbnzYOW2vv1rBtjvpzPM8QTwEqJpisbmQFWqy3rMco0yg09moMj44e9lsnDzTu3hgtBrRAp+BO6D+/P8IeCEPc93EwFBigs7g6stifX1WGmZLdsBy7a5VFcVf70vrB9NEJCUfh6RmC5s37itaiRT+UvvVH8+b9EBb2T2RmnodebxaWFY0mD0FBWWjYsDMaNOgADw+2jsW0TmjcuBtatOiPTp2C0L//fEyYcD9iYr6GTsdaO9asauqcK2up1piJjHwRrVsPwuiAEXjn5VdgYfC/zGR21TIlYcbVVDs5HymB6iWBm2GZoTVGtIL9uHTxA+zd+yiys1YgfHIAundvjwZN66GJVxP0nNoTw9YOw9h7xmLWi7Ow+N3FiP4wGinfpMBvrR/qNq0rYcaZsODqYzEDhGL5GI8/ngkvr/boFNQRi99fXKVgZu3JtfCa6gX31u6o26qutbW0Luu1q4cGHRuIVr9tfdSoUwO33XabaHUa1UGD9tZ97FOvTb1rx7eqC/dW7mjSswlGaEbgKszka7HBsgG+Sb64vf7tYpyGnRuiR3gP9J/bH/2i+qHf3H7oG9UXvSN6o9v4bgKu3Fu6o2btmmDfAQsGYPEHi6G9rK2yiRaqMswYjRZkmcreTCYLDAbGmZTVKgMRz2IyAbGxP8HDo13hM1gD9eq1RpMmXmjatCeaNHHcmjbtJfqOG/cgMjIIMyyWaUFmZh78/JJQs6b1efTwaI+ePaehf/+56NcvCv36cTkXvXvPQvfuE9G2rQ/c3JqhZs07BNwMGrQM8+e/g8zMi9DrrVaa8lxjaY8hTNIyM3bsXWjUqAvmTJ2GT5mWWcLMVZChu5yEmeqlOMqrlRJwNQncaJixQgytMXvw1Zf/xT/uScCUKSPQpk1TNGhVH55BnvBL8MOUx6dg5f4VSD6eDE2uRmTgZbF3viTeZNmEoKwguDd3lzDj6gDizPkxpZ2ifIj337sPgwd5CWtDxAsRVSIBgOpmln4xHdOen4bgTcEI2hhkbRuCELwxWLiCjfvHOLAF64NRv1P9qzDTdUpXjNtg3Ue6p6vY1eMLxxl7/1gs3LVQuIdRgdcWWGFm2LphuL2BVXnsPrU7Fr21CPHfxCPmmxjEfhsrlqs/W41le5Yh8sVIBBoC0TmkM2q61RTn7zq+K5buXlo2oFGsWebEPDiXwiQE9u0nlbenqsIMXZxiY7/C8pUfYtnKg9a24iCWOWqi3yGsiP4QCQk/Q6ejRaRsQGONEyHM/AzChhWoa2LgwKWYNOmfCA9/FJMnsz1mt4WHP45Jkx7BypWfFbqZWQP4CTPDhq1DzZruYtzOnUOwZMkeJCR8i5iYbxAb+61oq1Z9jmXL9iIy8iWMHKkTFiI3t8aoVaseunWbjHnz3hJuZwSkyk0MwAQAZnh7x8HNrSnily3Db599LtzMpGXmWsyQhBlXU+3kfKQEqpcEbiTMiPiY/L3CGvPWW3djxfIpaN+hBTzaWCFmVOZILHh7PlJOpEBv1sNkMYmYZR3jkKkLWbTIzMtEtiUbowyjJMyUBxSumsWEeWyX1URWuF6e8W7kMdaMZgfx/TfPCz/EOh51hCuUJl9TJYDGkbrOeBo1RoXB96m/pqKVf6urMBOyJQSG8wZhhTIoBhEX42g88UNTCDN+qX5XYabfwn5Y+81a8UPEtwOMqVEbg/+5rrmkEdYYr+leqFHLah3qHdkbCd8lQFdQQirsQojRK3oY+PbBXNgsxmv3iH1u4L+qCDMMvk9PvwRv71gEDJqHEO8FCPFeiJChCxAyxE7jPu8FCPVZDO8+sxA0ehM0mguFFozSA40tzDRoYLXM1KhRE1OmPAWd7jJo9SlNAD/78DqsMGVdt8JMKmrWrCue7R49piA5+SRycujaZg22Z8C9bdNqL2PhwnfRu3cE3NyaoGZNN/TtOw+rV38ujuF8ywpspetvzWSWln4BXl7Tccft7shetw4XfvwJysmTMpOZTdyMhJnqpTjKq5UScDUJ3CiYseTtQkHuHhw7+gYeezQdw/37wt3DDa0GtMKwpGFY9N4ipP+eLuCFuprQxdQMsjaJqjQFGuQgB4GmQAkzZQEJATH5u2DJ22Nt+Xug5HN9r2hK3m6IjGEunEiA1wDLXvx5/m1MnjQcderUwkjNSKw7t06Y7ki8N1JJroxzUfFm7ZykX5LQyu8azNByozmnETDB1MklnptvAIqBmd5zeyPuyzjo8/XgDxPPd7UVpmQmiBjzjFjzxRo079tcAE2N2jUw5ekpyPgjA3qL3r7LmaKD3mKANl+LdadSsfbHBCR+H4+Uo8nIvJQp3lJwfP6A36h/VQ1mrOmAFaSk/AGPBu0RM7A3/jE2EPePDcb9YhmE+8cW19gnCP8YH4pJ7ZujTYex0GSehV5fNuuFPZihNSYj44Kw9tBqxH4lNzU1c/Ew0737JCQk/CpiajSagiLj8Ry0LCnIzlawaNEOdO48BjVr1oGHRyeEhm62sc6o0OS8pSqHuPifRPB/y2ZN8ehdd8F8TGYyK5qJTcKMq6l2cj5SAtVLAjcCZggyeZd34fvvXoZeuwTNm3ugXgt3dBnXBTOem4F1Z9cJiKGOI/QqBzqphJlywIYwieXthiV/N86ceVfciE8/fRaHD28Vvn5Hf3sDly6wqM9ua2EfpkEux3kq+xgrkO1EQd5urF41FY0b18eQlUMQ/0O8eID48NwoBbnC57GnzAsY0CP5SPJ1MKM9bwUdu9BmO6YdmOkzrw/iv46HwWwQP2zFXQfHp7UrKzcLY+8bizrudcRb9L4L+iLxh0QBM8XOQdFBc0Uj+kS9HYXQLeMwLHEk/OJGINAUghkvzMTqT1YJILqRxU6rIswYDVaYadK4F3YsWIjcnPVQ1m+EwmXOBjttPSw5G3Bp0524OygArTuFQ6M550SYeawCsSqOYOaoSDhAeCjOYkK4I9QwiUBgoAnu7oT8GujVOwIJib8JK5G9Y4sbr7Tb1OD/JUt2oW1bbwzs3RevPvOczGRmY5FRoUbCTPVSHOXVSgm4mgQqG2YIMlcu7sTXX72ExPhI1KhxGxq0bYAhqwYj5usYGPOt3iclQYyqc0mYKSNk0Npizt2Fv/7aic8OP42774rDvDnjMWqkP4b5+WDihECkpszHttc24/ix15B3hdYb14QZwpKIm7EcwpbNMfD0bI3uE7tj2e5lwiexSsGMPeyqCMzYjlkBmBFWE7MOxgIjVny6And43CFgpuXQllj58cpr7mLq+QhRig4ZFzKw8IOF6L9gCFp27oUunaZgaL8k+A5MR69u89HW0wfdxvVF+JNTkXw02WplsgUwdTwnL6syzDRq5IVtkZG4kKlBvlaLAo0GZjuN+3I1GvxlMCInwButO06CRuNMy8zNgRnCB2GGgfgREf9B69ZDxPPo6TlaxNtwO8GjtJBS2n4ck+5vTE3dtKkXJoeOxb7t78By5ozMZFYEaCTMuJpqJ+cjJVC9JFCZMEM9OvfSLnz99UtYvWo6atSsIUBm9PpgZP6VaY1TpsdMGfQZCTNlgBlaMgqu7MTvv+/EU09koGfPrmjdOgg+PjpMnPQfTJ36KoKC7kfXrlHw9ByAlSun4vBHW5GXuxciPqUM56psq4w6vhVmPhIVVIcM6YEW/Vpg1n8iqkStGZXIHS5dAWYIExadiM9Z88MauDV2E8pj496NsfTQ0r+jhgAZPTIvZGLWSxFoNaALenRegJUrDsNoAtZvADZsBLJzgKS1vyFo+Ca0aN8fIzJHYe2va60uaQ5MsQ5l9feZ2P1W1WHmjdmRuKzVQtHroTjQxLnPrNPhkikL60f43IIwY8Hcua+hffsA8TxyOW/edgE5lQEzVjczBX5+a+Hu3gJxS5bhx48PS5gpAjIym1n1Uhrl1UoJuKIEKgtmhB6duxs//vAKktbOvgoy4x8cj6yCLKunSjm8giTMlBIweANokfnjj524564Y1K5dD2FhD0Ovv4hNm4CNG4ENGyDW16+nP/pudOs2FaEhfvj4wyehmA+KaqYqRLjKUhQnshzEl188izGh3ri98e1gcLxIAgC9XYXWGUrxDRnDFWCGgELLTL4RSw4twR0NrJaZVr6tEH04+u+WGc7XbMDcN+egrbcX/AZpkbPBCi9anfWNORVNKoasXbJhEzB75ito13kYArOCkPGnNQanWLc1J91NCTPOipm5Zpmh65cDripmX/ndzHge1TIzffozaNlyoIAZZkJbunRfJcGMdb4ZGZfh6TlOpJLeotHijx9/hHLqlAz+LwI00jLjiuqdnJOUQPWRQGXADPVoS/4enDz+BtZnrRDxw/Vb18e4+8dhAzZY447L+TJWwkwpYEbcgLzduHxxN97cthl16zbE3LnvYePGvKuBtlQwrc0adJuTY0Fs3OcYMHAJpkz2x+9nt0Mp2CuynbkKyHAevDYmAbjw+9uIihyN2rVrwifWBymnUqxxM+V8sG4IqJRGOXcBmFFjZrKvZGOkcaSoTsv0vINWDELST0lX420ICUaLCXHfxmJwtB969ZgHvYHPGCGGdUX+HoxNoNFoC5CzwYxxo+9D9yAfRL4yA1lKtt34HWfcFwkzzoGZiZMeQVr6n9Bq86HRsBU4aEXdvsoPM9Z4GI5nFmmd69RpJGCmT585SEo6VSkxM1arDBAT8y1atx6Mls2a478PPwwza6qodVWKKPRq/Eh1XEqYqT5Ko7xSKQFXlICzYcYKMtSjP8DWpzLRtFl91G9dD4HZgcg2ZwuQKYtbWVFdRsJMaWEmfw9++vEVTAjzw4ABydi82aqAFF+TwRpgazTmYt681zFoYBj++WAcFMsnMF/5wOWSASiEGvNe6HSL0KpVE/Se2RurDq8SaYVLlemrNFBxs/pUMswwm1nsl7HQ5euQWZAJjVlzXWO2sqz8LCw7tAxuzdxwW43bhGk14pVZyLyksfqFFuZLNyhGRLwwHV4jRmL2zNeRvd5x/ILVQmNBfNwvGOi9EAHpw5FxKUMkpq7IL4aivyhsv0uYcQbM1ML06c/BZMrD+vVATo7jlp39d5Dl7x2maS6ampmFMePjrdnMMjMJR+pLFi6tsMRjeb7Zs19B27Z+Ivi/ceNuGD/+fhiNVqtfUXCu6Heem+ecM/tVNGvqhRE+vtj5+huwnD4trTLFQJyEGVdU7+ScpASqjwScDTMWYZXZi+1v3wt//z6o28QN/Rf1h+aiRmRlrag3iYSZ0sBM/m7kXdmJ97b/A+3a9kJ09HfIymIa1aJvS68pHFQyWRwuKek3hISYEBkxEvlXdkIpoDXEtRICiLgZ8yFs3apH//5d0d6/PaJejbImAShN2uKbBSqlOW8lw0y/Rf2Q/E0y1ivrwboy2Tb/mPM8W8m2xr+8MAtN+zbFbTVvE2/Bh64ZipTfUoT7GX+I2Zh+kEVAQ7aEou/gWUhKPFn4lrwkFyQFOn0Bgketx6B5AVjzTTRMSlalWWckzFQcZm67rSZ8fRMQGfk/zJnzOmbPfh1z5rxhty1cuEOAyzWoKB5mevacitTU09iwQRHuYoSga40QQzi+iGnTnkP79v6oXdsNNWrcjkGDliMx8VcYDHwRw+u79rvMGeuEKsLM6OBsNGjQBktmz8FXBw8JmJHFMq8Vy1StUBJmqo/SKK9USsAVJeBMmBHhDAX78f23LyJmzQzUrX8HOgV3wpqv1jgtc66EmRJgxmoa24tzZ9/FPx9MRtu2oTAa80tVJdvqHpSLmTNewMiAYfj1x/8AygGXSwZgzt0BxXIIe3Y/jNCQofDo2ADj7h3nuP5JaUDCFfpUMsx0DuuMyP9GYvn+5Vi6fymWHlgqlot3LcbsV2dj7N1j0Xt2bzTu3BisLUP3sj5z+iD2q1ho865l6yDMsLhnypkkjNKNwVDvWPEm3Vok0THMCBeeLAsmj38cA6eHYtGuechWclBZVjUJM86AmdtQr14LNG7cBU2adC1s3dCkSfGtefO+SEk5W/h7h89D8TDTseMIzJmzDcuXH8SSJfuwdOl+LF16QGQpmzPnVYwbdze8vGagcePuqF27bmHBzLlYseIg9Pp8lD12p7TQw/la4OU1E3Xq1Ef2ujSc/OZbWE6dguXYMahKvFxawUbCjCuqd3JOUgLVRwLOghnq0Ky7eOXiB3jowSR06NQCLfq2wLSt00QcsbP0FAkzpYAZc95enDjxNnKyouHpGQUG+Jf27SWBZt68d+E/LBiffvwIoHwo0iG7WtyMYtmHY0dexZzZoajtVgs+Md5Y9/stUDyzkmHGrYkbWvRpgTaD26D14NZoM6SNdX1QazTv0xwNOzXE7R63C5Bp0LkBRmSOQOwXfwcZunARZuhilnw6CaO0hJm4wlogVBZLhhmm0w0f/yQGThuDRTslzNhaE6igq3VmmJrZFbKZEWpvv70x3N3boX79Dg5aR9Sr1wGtWvkhJYU1bqwgcz3MWDPkubk1QsuW/dCmzRARm9KmzWCRerl160Fo0aIPGjbshNq164OWoaZNe2LECA2iow9Dq70Mnc75FhneByF/I5CYeFQkG/Co1wD/fvAhXDlyBMqJE9LNTLqZVR8NUV6plEAVkYCzYIZWGUvBARw68ARmzghE/Zbu8FnjjXVn1tmvsVeOF+ESZkoBM5a8vTh1cjs2bViNDh2mC3eJ0sIMFYSoqLfgP2wkvvzsCUA55JJZzZSC3ci7vBNazSI0a+aBHuE9sOoTuiuZKs1dyTYOo9LWKxlmarjVQP229eHR2QOEFS5F6+KBRj0aobVPa3hN9UJA5ghEvhaJ5N+ShX9o0XiWq25mF9IQsmkM+g6KQnLS2VK7mekNFowO2oKBUSOx6osVMFViEoBbxTJjKUzNzBTMxTWLTocCnQ4XKyE1M4tUenvHYubMFxEZ+Vph24bIyOLaa5g/376bmZ9fqrCwEJBq1aqLevXao2HDzmjQoDM8PNi6iEaQa93aBz16TMOIEZmYM+cVrF17RCQgsAKzY2i2BcSyrKsuZlFR29CwYUcM6dsfuxkvU0VBhlYTusaVpwmLSzHwUtQiJS0zVUTjk9OUErhFJeAsmGFM9pVLO3Dn5hi0a9ccHUZ0wPzt82BSjE7VLSXMlAJmlPw9+OvP9/HvrTq0auWLjIwLpagCrsBgsCA9/S+Ehz+O0BBfnD35GmDZ53IZzWglEj6N5gP493MmDBrUHa2HtEbE8xEiDsRZZsBKAxZHFF/JMMPCl2FbwjDrxVmY+eJMRLwYcbVFvhyJ+W/Nx8qDK5HEopYWnajfUxRkrsqFcy0wYPqzU9HDPxhzI99BznqLwwKGhGqjSUFiwlEM9l2BYckBwqKmVwxlKjh1dQ6OZFm4r6rDzJuzZ+OyVgdFb7DWmaEvXzHNWmdGL+rMbBhZOXVmMjIuCsubffcu1RJT1J2rqJuZ1TLTqtUgjB//AGbPflGAUkTEi1DbrFkvY/78txAdfQjJyceg1ebaWHqKju+874wtZLxMQEAm6tZtimVRc/Ed68tUURczhUkLzp6Fcu5c2duZM8CJEyW61UmYuUU1RHlZUgJVRALOgBmrXrkfn3/6DCJmBqF+i3rwT/FH2u9pIkaYL3HLons46ithpgSYEemL83ehIHcXDu1/DH1698X06duwcaOaLrf4t5n8A24ymbFmzRcIGL4GcTFTYM7bD2Z0cCUXM3UuKsx89OFTmDplBNxbuGOkdqTI0mVX+XbaY+joEa3gvkqGmW6TuyH201hREJNyYhC/2gyKAWxGi1Esud/RDy8hgamZV326EkOXBcB7SAwyNedhEBmmmGVKVWytS6t1kDBjxuSJT6JXSBCmPz8NBrNz33gUvQNVHWZej5yFixqNKIhp1mphsdPMWg3ytVpcMJqwfoQ3WnecCI3mbCleZPwdBHifTCYgNvZnNGjQTsRN1ahRE5MnX6szc80y8vd7fG3738e052bWrVsY4uJ+QHa2NUaFbmm2jS9YjEYmJ7E9T9Gxnfnden6t9go6dAgSCQfuz87BuR9/hOXkySrpYvbxu+/iv089haceeQRPsj38cImNfZ9+5BG8+/LLOP/dd8K9rqg1xva7hJkqovHJaUoJ3KIScArMMINZwX48+kgaevfqhA4BHRD1WpTTgv5tdRMJM6WBGZFSbjdOnHgTyWsj4dlpItLTj8FkYiIAZjVTmzWWRrwtN1qQkXEGEyY8iIDhwdj+1p1QLB+6pIuZADYGaZn34tzpN5AQFwG3erejz+w+SDqRVCkPnu1DWKnrlQwzXSd1xZpPYqypBS3WrGQElr+1wu0lQmHhXDMvZGLm8zPQfYQvwsY9gMzMczBlmYUCqsZM6A0KjCaLqEMzL+od9BwSjhGaIKw9kihibxxBU0XlXRVhhsp7auofaNSoB1YNGoK7xo7DfePCcJ9Ycr2YNnYc7h03DneNn4hJXTqgY5cp0GQybqXiCQCsMPMoMjMvFo6njmkLGbbrRQGjeMtM167jEBPzHUwma3p4AvDf27XfVWVxFStvXxXk1qz5Fg0atEeLps2w/b//Rf6xYyUq9LbKvSut69auRcSoUVgWHIxVY8ZgVUhIiW3NuHGY6eeHxTNm4LPdu0sEOQkzt6iGKC9LSqCKSKCiMCMC/wv24vyZt7AqeioaNHbHkOghSDqWJF7uUo+oqC5ie7yEmVLAjFXZ34Xcy7vw0aEnMXLEEPj7pyI+/ivo9XniTSffdrKZTNblunUnMW3aU/Dzmw6ddjEu/vUBlPzdLuliplpn6NtoztuN++9biw4dWqDTqE5YvHOxcDVj/RTbB6fKrN8AmIn5NOZq4cuKyoUQorcYkHw0GePuGYeuw4Zh7NjNWLXiENalnoFWw4xTZqSn/4n4uG8wY8Zz6D10Gryjh2PFR8utvyScaLot7nqqGsyoVoz09Evw9k6Ed7+5GDZgXmHjuqM2D8MGzsfAPrMQEnK3SGtsDZQnaBQFjOK/qwp9yZaZ4o8v/jz2YSY2VoWZygnoL34+xc/d6mKmYOrUrahbtwlCRozEV/v3V7l4GRb2VN3D5s2YgZihQ/HizJnYPn8+ts+d67C9HRWFd5YuhdHPD2E+Ptj99tvCxU4Ai534GQkzVUTjk9OUErhFJVBRmBElPywHcWD/o9YsuZ08MP6B8dAX6CvFBV7CTKlhZicsebtw+eJuvP7anfD374/hw2PBdKdr1nyKtUm/Ijn5KOLjv8XSpbsQFnYnAgLCsS51Hn75+RUoBfug5O5wSRczFWYsV3ZAKTiAt9+6F0FBg9CkWxOMvy/MWm/GyRRdnJJcKduqGMwIGYg5G5B6MhUT/zkB3UP7YWDAHISGbcL0Gc9jZsRLGD/pAQwLjkG3kcPgnzocqz6PvmGptKsezFgVbVookpNPIz7hKOISjiK2lI194xOPYt263wszfpUeZKj4V1+YsVqHsrLMGNB/MerUcUdGXAKOf/cdzFXMxcwWZpbMmYOHJk/GWY0Glo0bYcnJgeKgWbKyoNxzD96MjERkcDD2vvMOlFOnHLrYSZi5RTVEeVlSAlVEAk6BGfMhPPpoOnr37oRuYV2xdMeSSksqJWGmlDBzVeHP3Yn8vAPYu+dfmB0ZBB/v0QgYvgIhIQaMGbMegaOS4OszFePDgnD/fXE4fmwblIL9Lutepl4XlyJuhoWNvnsBy5dNwu0N6mDA4gFIv5Ah4kCcAhuFcSO2blglul9VxCZUFWGG18t5KwboL+ux4vAyBBh90X2mF7qG9EOX4P7oNqkHhsQPROT/zUTa2TQYUHkB/0Xve1WFGYKFwQCYjEqZm9HIhB7FWx9KslRUZ5ihS15Kyhk0bNgNbnfUxUuPPYaLR45UacvMotmzcee4cTielIQCvR4FjL/Sau22vMxMmHNy8L9p0zBz1Cjs2b5dwkwVUejkNKUEqqsEKgoz9PRhrPnaxEg0beqBoSuHIvGXRBFD7GwXM+ooEmbKCDNXlX7Lh7hy4X289cZm6HVzsWjBaETNCUR8bDgefywZP3z7PBTzPigFe13eIqMCjdXHkcWN3sedm9fAw8NdBGytPLwCWXBCRXlFJzJ66cx6EWOiL1wyy1elAU1VhZlCoKFsTMgSRTAzcjMQf2QNYn5ahdQ/UmAyZyFLybaCJmV7g/5VZZixxo+ocSNlXZbNIqNCTvWEGcrKLAoMz569DXXq1EWvbt3xJeNF7LhWuVJcTNG52FpmCDN3h4XhVEoKFKMRilbrkHLNGg2wYQNemT4dMwMDJcxUV+1QXreUQBWSQEVghrokM/eePdUyD+IAACAASURBVPU6wicFoHbd2gjKCkLmJeuL8cqI6ZUwUw6YofJPf0CYd0NRPgaUT0T9GCgHAPH9MGA+AJKpuKnlPIcKGTdyacm1Zp94441/YNiwPsLVbOKDE5DNuiXm8gdsaalsm3XQ5+uh/1MP7VmtaPo/dNDn6sU+bWVATVWGmUI4ITyw6WGASfzLggFGAYBiXyXHyBSFpKoMMypg3MhldYYZvf4CBg9agVq16iB+2XKc/OFHmI8fd+hiVRQkXOG7hJkqpIXJqUoJSAlUWAIVgRk1XubLz59BcNBgNPBsgPCnwqGHvtJeuUqYqQBoEFTMVz4QYMN1NrpqcRuhQMnb5dIxMsVBEudvKdiH7757UWSgqFO/DvrM6Q1NrqZ81pNCa4zebID2dy2WHV6KMfcFY0jiAAyO64egzSMwf8dcrDu5DoZ8A3QWJweH3QIwcxUm6KJXCDaV8Wbj6nlK+HUjYcbhi/jrEgNUX5hRsHbtMdR1b49atWpj21NbcZkFJ6tgsUwJMxXWjeQAUgJSAlVIAhWDmQ9E9t73378ffr690da3Dea+NrdSM+NKmKkAzBQHA1V9G4EMBbuRe+l9PHh/Atzc6qDN0DaI/iLa6mpWBuuMULgVHQx5Bqz8ZCW6zeiEOvU90KZpEPp7xWBQj0R0bDUR9Rq0Rfvgtoh8LQLaSzrnZuS6lWCmBMgoLYxUtJ+EGQkzji1bdDGzQK+/gjlzXhd1dTzbd8ARFso8fqLKWWVoGZIwU4W0MDlVKQEpgQpLoOIw8xGee9aIvn07o/uk7li+Z3mlJpOSMCNh5jrrkbAw5e/FO+/cD1+/Xmjs2QiT/jkROUqOCLIqlTJMiFD00F7SYtYrkWjUrSX69FiC1Wu+gMFwEVnZV0QzGi8hIf5HDPfVo3mXHgjeGIT08+nOAxoJM05HIAkzEmYcwQzjkhgvwwKjPj7xqFGjBtYsWozfjx6FmfVljh2DK7iOlWUOEmYqrBvJAaQEpASqkAScATP/+lcqvLw6oO/sPlh9eLWEmXLc/9tOnz6FpKS1aNjQHStXTELBlQ+ExUFYHiTAXAcwthYlusiZ8/fh++//h5g101CnXh30mO4FbT7jNkrp80iIyNNh7ltz0WZQd4zwN0KnPweDoUDUSdHqzNDqrOt6vRl6w58In/QEOg30R+jdodDlFiYLcFJgO93klh9ejvk75ouWcCQBuoIyJh6glQkGxP4Ui4W7F4pxor+MRvrFwmxvTpprqWDR6YhS+rNKmCkbzFDx1+uBjIwrWLLkABYs2Cna2rUnxc+CIzAoaR/BgfVrFizYjfnzd2Dlys+RkXFZnK+kYytvvxVmYmK+Rf36nVGzZk28/vRW5J46VSXjZaRlphx/geUhUgJSAlVaAs6AmTvvjEPnzm0wcPFAxH0ZK2GmHE+EhJkKABuBjwU+L/31Hp54PA0NPNzQvE9zLDu4TLiaaSyOC2jSvcxoMSL2qzUYsmIYenvNR0bmaegNZps6HVR4bJreDI32dwSNykbPCX5Y8F6UyNRFxbn0araDngSNfJ0VkghKZgd9Szojj80rHItjVhOIUSUmYabsMGMFBz7vedDpcguXzipqyZ8rddz8wp+r8s6xYscRrvhyQqv5E1OnbEXt2nfAq0tXnPzyS2usTBXMZCZhphx/geUhUgJSAlVaAs6AmU2b1qBTp1YYtGQg4r6SMFOeB0LCTAVghlYaAo05bw/27XscwcGD4N7cHaO0I5GNUmQ1o1XGrEfESzPRxccfcyLeQnaOAo2GSlfxypIIks6yYOWywxgYMBejTCOgUyvFOgsWOA4zplU0axrHUcdy1txKAigX2i9hpvhn2N6z/fftBBpCjLNAhnOxHZPrFZlfRY8lzFiQlPQzevaMQF03N2TExSP3zJkq62JWHMzcFRaGE8nJMBsMor6MRauFvVaQmQnL+vV4WdaZKc/fcnmMlICUwE2QgDNgxmqZaY2Bi6Rlpry3UMKME2DGkr8Hx4+9DpNhCWrVqYW2Pm2Renad43gWumIpBqz7fR1Ct4ShZ98IpKacFwUHrb709pQlBXqDgvT0iwgMMmHgIl8kHkuwZr9wZuphFUQqCgfOGqei87gJx0uYsfcMl2Y7YUNtpelf2j6VMWZpz632s85Bq83DggXvwd29OZo0aoSP33oL5jNnYKmCsTJqXI1tzMzi2bNx74QJOJ2WBiU721prhvVm7DSLXg9l82Zsi4jArKAgWWemvH/V5XFSAlICN0wCzoCZf/6TMTPt0W9OX6yRMTPluncSZioIM8I6k78LeZd34K0370aHDi1Qv1V9zHp1FnKQY7fmjHAxU0xIPJKAEWljMGRgHAxGvoWmoqMqPfaWrLKuYOL4hzEgYhSiP1/uXFezm6D4q65ZRZeUE6FANHXdbP1etK+rfZcwY+/5rd7brS5mFqxbdwKBgUa43eGGsKDR1nTMakawW8DNbEFEBIxBQfgiOhonkpJwPCEBxxMT7bZj8fE4mZGBJ8aNw5Thw7H77behnDrlMBGCSJJw4gQKfvsNXp07o3HjxtixYwfy8/OhKEq5/ijKg6QEpASkBEorAWfAzLPPGmQ2s9IK3E4/CTPOgBnWzcnfg++/ewlzo8aIRAB95/eByWwSSnhxsSJUdI0WExJ+jsPw5FB4D1oLk4lKXskwo9UpMBgVTJ7wOAbMCMKKT5YImCkpRsfVlH1H8xEQo2iF9SrLkg2TkgWjYhJLfjcqhcUwnWmNcjLESZip3tBi/6WENV4mOvojtGs3HM2bNsUjW+4C/vwT5qNHq1wGM9Uqc9XNrBDEYhYvxsT+/RHt54fEESMQP3y4wxY3fDiSAgMxpVcvTA4MxKH33oNFwoydP91ys5SAlIArSKDiMPMh3tl+L3x8eqHtsLaY9/o8EUtN/cGRjlTefTI1sxOUfttMYLfSujVF8278+ft7ePKJDLi510HDzh6I+S4GBjMLXF7/2FktM0aknE5GUPZY9O23AJmZV6DXl8INRq+ALiqhoXdiwNxhiPt5DUxKIThVyuN//fwrdUuhaxrhJfV0CpbuXYSIF6Yh/KmJmP7vcCz6YB7WMstavl7Ajs5FgUbCjISZ64GGIKMgI+MvTJ36NOq6NcaAXr3xy4cfQTlzxqEVwhYaqsL6W889h7v1emxMTy9125SRgfVpadh63304/vnnUFg41IGVSlpmXEGdk3OQEqi+EqgQzOTugGI5iM8/24qgwEFo2KUhpj49VZTrKO4luDP0LgkzEmYcpmlW8nehIG8XPvlkKwYN7oY67nUQvCEYxlyj1TpTDGSwvgxTIYc/FY6ug4KxfOmnyMouIQGAxgKjyYL42J/gHRANv2Q/ZIiUx4YSs4Wp1g7+kPDcbFy/ur2YOTrjh6dMYyha6C16aHO1mP/2XIzSBmHQnEAMCBuPASFT0H/MBAycGYyA1GDMemkWko8mFwJNmc5yQ65UwoyEmaIwQxczg8GC2LivMWDAQjRu2FDUlrGcPQtLFbfKFAUsWpnyjxxBXjka3cZKU2dHwkz1VSLllUsJuIIEKgIzogSKZR9OHf8/TAkPwB2NbsfojaOReSXzqn7mbM1GwoyEGccwQ1ezvN04ffodmIxLULN2TbTo1xxJR5KgV7ONFVWhhRuVEcv2L0Xfmb4Y4a+FwXgRej3ryhSNn6E1htvMMBivYGr4k+gzJhjT/zNFuF45NEkWWjoMFoNV8TfroMnTWOvhWPQwWIzWH5yi87vR3zlPiw6Zf2UifGs4uocORd8BkQgN2oLZM7dhyYL9mBe5HeNDH8KgoUvRK2QkQjeNRcKP8eC1Ecqc/YNfkfEkzEiY+TvMWK0yGs1lRET8Fw0bdkSPrt3w/osvAn/8DvNvv1VpF7OiMIMTJ1Du5sAaY3seCTOuoM7JOUgJVF8JVBRmWN4j78pOxMRMR6NG9eC92htrf1trTepUCa5mEmYkzDiEGWua5l3Iu7ILhw48hjZtmoqK3lTK9Rf11notVNZt/xVaIdLPpyP8scnwHOiNWZGvQKf7S9SaYdYyHd3O9FYlSNSf0V/CosU70c9vBoavC0DysSQYFCO0RccuPA8VfGGBMeuRciwZKw4tR9QbczDzxQhEvDILC96fj9gv1yD99/RKexPwt2u2vf6i67QSXdJixgsz0NjTEyP9s5C09jeYsgtEymqmrVZbauoJTBj/ELp5B2KUYSTSzqRZLToEGltrkx25lHpORedYhu8SZiTM2MIMX0bQKpOQ8CN8feNRv149zJw0GZeO/Gp1p6rCWcxsAUNdJ2hUpKnjOFpKmKm+SqS8cikBV5BARWCGeqPlitXV7KGHUtCjRwd4TeqOZXuWVlrhTAkzEmZKATO0zuzC2bPvITo6HLfVuA2eYzyRfCS5eJiB1cWLVoWkX5MwQjsS7Xv6YmbEv5Gc/CMyM/+AVpcHrS4fmZl/ITX1V8yb9xp6DZmKAYu9seLQMhEU79AqY9FBc0ErCnkGmULQY8IQeA7wQ+eeQfDsMwpdhnlj4EI/hD8xBYk/JYi6N5Xlq+kQIAgd+Tqs/nw1mvZriRE+OTBlX4bRSIuUGRqNpbBxndYpC/SGi5g+fSs8vQch/JlJMJmzrNnjLHroLHoBN3RZE9dzE6BGwoyEGVuYsdbMycecOa+gZYv+okjm8//6F5Q//kDBkSPlVvwdKfu3+j4JM66gzsk5SAlUXwlUFGbMhTCza+c/ERQ4GE26N0b4I5OFR49D3a4ML1ZtdS8JMxJmSgEzuwTM5F7Zg13v34969dxQo1YNRG6LFBYHu0UoGcNiMSD1eCqCsgPRskdHDByyCOFTnsKChTuxaNEezJjxPPz849G+dz94xwzFqk9XCYuM3TEJShYt9Jf1mPW/WegY0APduk3G+NCHsWLpZ1ibcALxsUcwf84H8PfOQLvu3hi4bDBiv40Vrmg30mVLWI/MemSezcConJFo2TQApmyztUK6cK27XikWQGMwIzXlOAKDs+A5rjPSzqdbEy5c0UH7lxbaP7XQXaLrGuGmsIBnOX8B2P4yKO26hJnr79vflfvqs98aK6MgOfk4hg9Px+2334Hxo0fjzFdfwXL6NAqOHYPl+PFyNUcB8hJmqq+SJa9cSkBKoPIlUFGYYdyMYt6Ls6fewOKFE3C7Wx0MXT0UqWdTK0UXkzAjYaZEmFEztNE689efHyBsvDdq166Ffgv6IvW3VGH1sAsJhUH5TBiw9KMlGJzYB638W6BZ77Zo1rMtWvo0R99lXpj9xkxk/p55NTWxPcWainSWkoW526PQ1LMtgkZkYd2641i/AcjKBowma8teD2SvL8CypQfRucskeE7shLSzaTfWQlMIc4nfJaD9qE4IC34MGzfTIlNQQs0dq/vdggXvo+OAIYh6dxYyjmci6p05mPhEGMY/Ng4R/zcTa75fA815DfT5N9ZKI2Gm+sBKSZBGFzOj0YwFC95Ehw4j0KVjJ9ybs15YZfJ/+w2/HzmCs2Vs544cwfkjR3D5FkscUBYAk5aZylfW5BmkBKQE7EugojAjXM1EVrP9eOCBJHTr1g6dQzpj0XuLkGUp9DZx4ktYCTMSZkoFMyJN85WdyM8/gNdf24j69euidv3aWPDBAmjztI4tBIz1IIQgGxuwSbhHJZ1JxNpTCdDkZSJHWY9s5FyLDbH3gAs40CPjjwx49PTAyIAcaHWnYTBaXbSoWPFNMZtGaxZuXKxxExvzJVq19odvxiBsUDbaLfhpD6DKvZ1Z1Sw6LPtoKZp37YyVy74SWd2sCQ/sK8TWGAQzYmO+QJ+BEWgZ0BCtfNqifv2OaNM8EO1ahKJRw17w6NgUA5b3wbJDS6HLs2ZyuxGudBJm7N+7kpT/W2k/f86Yjjkt7Q+MHKlFzZq3Y8zIkfjhwEEUnDuHH/btw5LZszFlwgTRwsePR0mNfadNnAgWpnzmgQeA8+dvuWxopYEaCTP2lSy5R0pASqDyJeAUmKGrmfkgPv7oKUyZMgJuTe/A8DR/aK5oHOuM9nRAB9slzEiYKRXMqNYZpmrOvbwTPj49Uef22hi8cjBSfkkpGUQK3cP4wOmhh5H/FJNYp3LM5lARJxAVaJGTn4OQB0eiRVMfJCf/DJ2eVg5mQ7tewaSyJfZpLyNy1uuo37YBkk6udZhYoNzgUtwPGeecr8WCHfPRrF13rI0/KQqIWud1/XzVa1BhJiHxWwwYsBSNPLwwbfJ/kJpyFkZTPkxZjDW6gEULd8PLcw6a9miHyU9NguaSRiQ7sGslK26O5dgmYcb+vVPvYXVY8jk1mRQsWrgdnp2C0aFtW2SnZ0A5d064lx3Ytg0N7rgDyQMH4s7AQGwZMaLEdueoUcjw9YV/y5ZYGBUFXLhw62VDK0VGMwkzla+syTNICUgJ2JeAM2AG+TuhFOxCfu5OZGctR9OmHugc2hnL9i1DtpINZxZElzAjYaZMMCOCupSP8OxWLZo1bYg7Gt+BOa/PEXVlSh2/Ud6sXBYdDFeMaBvaBKNH3Qmd7vfC+BNCiz0Fk1nTzFi79idhnZm4NRQ55vVW60wlB89bs63psOTQEjT17ILVK79DVhYKU1Hbm691P113YmM/Qb/+SxEYZMCWO3OFxckKbhbxRtxozIdW/ydCgu5E+4H9MPnJiVaXs0qOo5EwY//e2X8Ob61jVKtMevofGD48FbVru2Pi6BB8tWsXQJg5ehQfvvEGurVsiSOxsbi0YQMumUy4lJVlv3F/Tg6+SUtDyrBhWLJokYSZ336DV+fOaNy4MXbs2IH8/HwoimJfA5F7pASkBKQEnCAB58DMLlhyd8BScAB7dz+CSRP9cUeTO+AT733VOuOsl68SZiTMlAlmlLxdsOTvxp+/v4uQkKGo634HhkQPRuL3iZVbE4XgoeiQ9uc6eHT3QNTsd6HX50KnL1q35u9Ko1XpMiMz8xQG9luNQYk9kJO/QVh5HFqCymGxuM6qU+gWF/dtHNr4d0R42H+wYRNhxewAvjh/uu9YsHjxDnh6TkBk5KtYv97qSndNWbZandgvU/M7/H3T0HuqDxbvXlRyfZ4KXpuEmb8/Y9fuSfXZTqtMVpaCeXO3oWOHEejUrj026QzIO3FCBPsXHD8uYKZzs2b4Pjoa+SYT8nU6FOh0MNtp+Vot8gwGfJ+cjERfXyxeuFDCjIQZJ6hlcggpASmBskrAWTBDrx7WnLn01/u45+54tGnXFK0Ht0LEixFW64xZU0GNxKp5SZiRMFMmmOGDabXOfIhnt+rQrl1z1GtZDxH/iYDmsvP9IK8CgoAZPVLOJaNBx0ZYvPAgdPr8Qhcz+5YZFWa02rPw99Wi55KOyM5bf0NgRs1mln4qHb5pvujQaiKy11sTAFAZLE4JtrqYWZCWdgZjx/4DnTqNRnLyCRhYm+c6dzqOZVUqVyz9CH39ZiJ4QzC0lRw/I2Gm+kBL8c+oIp5HWmX8fONx++0NMHVcGD794AMoZ8/CfPQobGHmu+homI1GmLVaKGw6XbHNotWiwGDAj0lJSJAwIwpzFkiYKasOJvtLCUgJOEECzoQZC4uvF+zFl1/8G0sWTUCdurXhNdFLZJk1KSZrmEEFkUbCjISZMsOMSLlXsAdnT7+JCROGwb2eGwYuHYjYL2MrzzojXML0SPsrDQ26N8Cc2W9Dr78iLBi0ZBSndHGbCjMZGafQv+8KDEnphfU30DIjXO9ydVi6fyk8PFtiQugTyM7JFW5itNAQRq41M/QGFiDMw5w5L6NLlzCMGXMXTCZrv+Kv0RqEnZl5GQEBWgxeOhzxv8TCWb8grsKkzS8aCTPVHWassTJz525D+/b+6NyhI/6Rs15kH2MaZqZULgozFpMJluIf4KvCJMyYDQb8lJwsYYaFRk+cgIQZJ2hlcggpASmBMkvAmTAjEkjl7cKVizvx6iubMcyvN9ybu4tyHJkXMp2SqlnCjISZMsOMsM6IlHuH8PRTmejSpQ0atG+AKU9MQebFTJEMwOkuXISZwpiZjpNaYaR/NnS68yIehsBiX0+yxswkJv6MVm19MOU/Yci+QTEzAgQK553xVybGPTAOzVr3xqTxjyA9/Tyyciwiu5kpSxHL7BwFGs0fiIz8H3r3ioK3dwySkn4T1dUdXSNhiGNMGPsvDJwZjKX7FyJbyam0rG0SZq7q3w6eu1uzD581g8GC1NSzGDo0Gm5ujRAZHo5Pd+yA5cwZkXlMoRJu42ZGy4yEmeMoTRYztY9MAFBm3UseICUgJeBECTgTZoSrGa0z+btx5tSbeOD+JHTo2AKNOjfEmLtDYSgwVDh5kYQZCTPlghlB2gV7cfL464iICEYDD3f0ntULKz9cWTnWGUKBWYesgmyEPToabdoPRVzc19AbCgotG9cDDRUvZjvTaP/AlPBn0KRXSyT+lgCDxeg4c5qNFaI4y0SZtxXOPfV0GkZvGI0uw3zh7RuL6eHPYvnig4iJ/hYrlx7GrOn/Q4B/Gtq3D4CX1wysXPkJjMbi3Mv+rijzOrOyFYRPeAoDp43Bwh1zJczYTQjxd9nZh2DZrzjZ0JrIDGYsdtuy5QB07eSJf955Jy4eOQLLiROgEi5hpmzgogKM7VLCjBO1MjmUlICUQJkl4GyYsQWan354CakpUWjYpB5a9GuOKU+Gw2gxCqAR3izl0MEkzEiYKRfMXLXOmA/ghf+Y0LdPZ9RvXR9jt4xF2rm0ClN2scCg6GAwG5B8LBkdR3fAsGFpSE4+CqOJ0GKNH6FiL5rO6tevN1zEksXvo2vv0RhhGiaKNRGKbvQ/ET9jsdbImfZcOIZGB6D/xDEYFByBIYELMShwNgaEjcegecPhObor+g9ZgLR1F4TbmSM3OiqctNoYTQrGhd6HgZGBWPHxEmQp2U7xQy1OTtIyUz1Bh88ZY7fWrj2CPn0iUbduIyyLisJXe/dBOX0aFrpG0c2sPJYZvV7E0ViMRuFmxgQAIpvZxYtQWDzzxAn7rRSpjm1BoSqsS5gps+4lD5ASkBJwogQqA2asQLML+Vd24bNPn8GihePhVq8OWg1uiWnPTIM2V2t9Gc5SHWX8J2FGwky5YUbEzuTvwbnTbyN65TSRQ9wzxBML/p+98wCL4ur6eFQQFFGw94oFuzRREBQVCyrYe0NRFAugCCiwFVCTGBNNj4mJ+cybxCSaYuxKs/fYG1FRFBArdXfn/z3nLoMrAREEZeGuz31mmLlz596zs+757Wl7ZhT7gSzsASZFOlwIx/htY9CiV1f0dZFj/rxTCAtLY1CjDCeXKw0rpBkUnIwpk7ehh/1kdJ/eFUv/XQKFoNAWayriB6Wweb3KeQIauSBnNXZ8ry3GxG3jMPTTwRi4pj8Gr3fFuC2j4XNqHtw+HoqOvYZhzqwTCI/QICys4OxnooIZHPwQvRyWoudCR51aOkX/D+GV1qGRQA45ApOD0GZgGxhUqQW3oV+wBAUyGSn6/7WS5fcrPz+mX1BEPxJQyvChQ9fDzKw1ulp2xM9ff43MO3e0VpkcqHgVmKEkAJDJkLZiBeJmzsSG4cPxlZsbvhw+HCtdXdG/RQv07tULGz79FF++/z6+WrPmhbbhgw9Abe/PPyPpwgUI9+4xq5A+gMqrzJHDTAlqZXwoLgEugSJLoLRgBlnRoKy4GWnROHpkI8aO6Qvj6oao360e3L8egeCkYOZ2RhaaoqRt5jDDYabYMEOUTTnEqcLrnl3r4OzcHdXqGMMxyAHL7i4rkaCu/yjXOTEoMrUCHptGoIObHWydZ2OE+5eYMW03vL0OY86sWEyauBUDBkeia9/hsJlri/kn5pV6yuL/zDUfYGIfTkHKCoaS9URbOJQARwmlJhyRmpXwipmFrmMd4OIcCUU4pZ/OPwEAgYzoYjZl4k506jMMbp8NhUwlL1U3upK0zBD8FLVVZAgqqqxkJeRupwUZAYsWXULLli4wrWGOED8/3DxzFpr797XWk6LADGU1UyiQsHQpltnbo0+TJvDq1g1zunWDZ9euGGtpibGdO2Nunz7wcnDAHJ1Gf891coKLpSXmTpiAo3v3sixqGrLglBMrDYeZIute/AIuAS6BEpRAqcEMpWqm+BkGNDE4dnQjxo3tB1NTY5hZ1EL/yP4smZQ0Xap1O8upS1iYfsVhhsPMa8EMeygph/jTKEgks9C4cR00tmvMUjXLs+VaKwgBSAn+ywUCKOB5YDrsl/SEpZsdOvdzg5XLJPToNw6WLv3QabwVXD8cgCU3/MsEyOjKgICANVE2ghRhlG9dLUNwcjCGrhsKC6u+mDp1D+SKTMjkZJ15bvGQyQRt5jOFAN/FV9DDxgtWsx3gc3Y+AyQaW/d+Jbn/ujBDCnlYWDbmzDmOKVN2YdLkna/cJk/ZhclTdmLevLMIDU17QSblE3K0NYd8fC5g8tTdENdfqMym7MSkKTsxefIueHoeRFiY6jWSFWjjtuTyTPR1lsDEpCEc7eywf9vvUOXEyehCRH6WGUrNTNnKxNTMbF+hwHU/P3hbWWGutTUuLlyIi4sWsXZ18WJco30fn/+0C/Pn46KfHwJsbDB98GDs274dmtRUlhJadx76vM9hpgS1Mj4UlwCXQJElUJowkxs/kxmDrMyDuHjhJ8yYMZiV+jAyM4SNtw08d3kyK400W5qbVOpllhoOMxUIZphbGDPxkZkv+rUghh5GsTHrjOoIThz7FqNHOcO4RlV0ndYVvld9WVBXaSjW9FBTY7EhGRLMPzcXY34ZgSGfucBtw0BM2TUeSxL8mcWDXMtKYw4lCQjiWLQmckVbfGkxevo6oE2nIZg+fQeCg+9AIqGUzmTJIOVShZCQB/DxOQU7e1+0d7PBpO2TIFPJStUqQ/N8HZjRusUBQUEP0anjJHRu2gOObezhZNGr0NbHwh7OHZzQtoElbG0WYenS6yw1N41ZPkGG4qE0UMgzYNtzGTo07grH9g7o07YX+hQiLzrv2LY3bFrYoHnz/ggOfpLz3BRdVjQHCvqftdX9lwAAIABJREFUM+cI6tTpALOaZlgfGYnkq1ehyce96wWYqVcP13x8oA4PZ8UyKT2zIJOxNM2U4eyGvz/8bG0hc3aGEB4OISJCu1UqIVCjv/M2Ov7ee1g7aBDmubtj/99/c5gpsqrCL+AS4BLgEihYAqUNM1qgiYE6k2rQHGJZzqRST3Tt0gZVqxqimUNTDFk7BAtOLkDIwxBIVc+hhmXLpR+CdX8MVoUhEpHop+zH0j4vWLgA169fh1qtBq1FX1/vJCXdR0DAUtSqVR3zvEdAlXEAUMWWqBIvKvNlfauFmBios2KhztQ2TVYsBFaZ9TmUFHcd5P9IJkN1Vhw+/TgAFhZNYNbGDIPeGwRJmoRlIROV9ZLekmItE2QIRzhWCquwUljNWoSwkoEMU76F0rNSlPR62HiU6EAjZzBo52uHJh27oW9/CaZP3w2f+eewcMEFzJ59EMPdP0fbLkPR1r0TpuyaAikVy4Ss1Cwy4lpfF2YUciAgIBmtmjsioq8jjnvNxFlvL5ydOxtnvQtup+d44h/fxfDt3hZdOoyCr995KOT5u+CVD7gRIJWoEa5IR5PW47DUqiMOzfPCmVeQFcnxpPds/OA+BDVMmmFZ4MNiwgzBD8n4MTp3nsIKZLq7DsbZuLh8QYasIbow06pePVzx8UG6UokMieR5CwtDhkKBK76+WGRtjTBHR1ZUk2CHimuS5aagpgoLY3D0Xv/+mDtiBIcZff2W5vPmEuASKLMSeBMwI+qc9IM4NIcAzWH89GM4nJ17oH59cxjVqop2I9rB4xsPLPpnEUJTQpiewzKeUSInTQ7QkGcLwYwQiX4KDjO5lgZRwOVhK4IG5fdOT4vDo0exePgwFs+exkGVHccAh/qAmo61paj7WuvMIdyM34oFPqNhbFwVLZxbwDPaE0pNyVR4FZXpvFtmpdFImJsWuWpRk6i1lpu8ffXlb9FCE/owFKN/9kAr92aoZ9kUjdtYoamFPeq3t0CTfo3h8q4T/G/6saQCBHVvYn0lBTPtW7rgxzHjkK1QApErgfAIICKywCYow4G167CmvwOsu0yCr9+FigMzFpOxbkA/PI5YCYHaS2VF1oxIqCIiccZ7LmqbtS8mzFBMlhrh4SqMGrUZxsbmqGNWG39s+h5pt2/npmLO686lCzONzczw3fDh2DdzJvZNnYr9OY3tz5iBzaNGYVy7dpA4OrJsZmR6ZMkBXkKjzEUtIgLvc5gps4oQnxiXAJeAfkvgTcIM6ZtMhySPIc0JxN/YConEE926WaBWLRMY1ayKFv2aY9D7gzDnkBeCE4MgfSSFPEMOmVrrjRKWHYZIVST6yTnMvJYyX1Tl/030J4sMWUuePY1F/LWf8NsWOVZHzoJMMgUbNwTgzMmNSE2Ngkp1iFlWXtf1jB5GjeoIdu5YB5d+PVDV1BBdpnXB8ofLmc8jKehvQtkuN/cQtAFwEYiEQq2E761FmBY9AZP3jcW8C3MQmhEGdg6KImX9eF35lBTMtG3RF9+6e+BZSAhzQaJf5F/WMkNDoV71LlY59USPzhMrGMxMxLt9HZAqlSJLIoWqEFnRebKEHPKcAfNaFsWCGXLfk8lUWLr0NmrVsoChQVWE+voi8epVqPOJlRGhRrh7F+o7d3AxKgqOtrbo2qEDunfogG55mm2XLuxc63r1IOvTh8XTkL8gh5m72sxslOL69m20b90a5ubmiIqKQnZ2NgRB0G8tic+eS4BLoMxL4E3DjKgTazKiIGiOADiK06c2ITBwCjp2bAkTE2MYGhqgev3qsBhmgQEr+2PqzqlYfG0xgh4EIeRJCFY9XQUXpQt3MxOFqe9b5lZGLmTqQ7hy5WcELB2Dpk0bwNS0Ixo2GoHGjUehZk1b1KxZBx7udoiL+RwZ6XEQsuNeyw2P7ksWoIy0KHz0oR9qmZnAvK05hn02jJn/WIA7x5miSUBMDCBIKd8ZVmIVa+GIYC5lBBbUXhdQinJ9ScLMdx4jkR4ayuIoKDj8ZcEv5F6E1e9htXNPWFVAmHmvnyMey2RQS2W5in9+8iIYIHetbKkUR2bNLCbMaN3LZLInsLFZjMqVDdDNsjOuHjvGQEbzClnDKLtYRnw80m7ceLHFx+PR5cu4cfIktm/ZginDh3PLTB558gQAZV7X4xPkEijXEnhbMEP693MrzVEAx3Dt6s/4eL0/+ve3Qo0axqhUqRJr71R6B5WNKsO0rSlaDWkFa08rNLZrBINqBliwgMfM6LWFRgQZssjs2vEBnJ1s0aKFGyZP3omQkMcID09DREQaZLJnmO9zDl27LkSL5i3x5efLkPpgDwTV6wENs85kH8Lli1swz9sDlSpXQtNeTbHw0kJWH4YDTVGw4cW+DFzUUkiovWGA0Z0Jh5mXMld+fFHMYzoxMxYT8aZgRptQQQOZLANTpuxElSpG7Ivj940bkXb3LghkmLKdRwEXLTMvbMmCo9vo2gcPcOPYMYwbPhx1TE3RyMQEMienXEDjlhlumSnXWiJfHJeAHkjgbcKMaFDQUHIA+pFcdRBZWUfwIGUPjh/dgE8/DcD0GUPQsWMLVDM2ZHpmZYPKqGJYBYZVDdn3FYeZ14gbEd+At7llMKM6jL171mHo4P7o0WMhgoNvQ6HIzEljSyl+tWl+yYVELkvF0CGfwLKDDb7dGMpSLAvZlCih+DE0lAwgKyMW27d/CBub9jAyM0L3Wd0gzZTm+jfqKsd8X78kwGGmfMOMNgW4Cn5+V9GokT2qVKmCuZOnIDk+nrmPvTLI5EAP9RebOiEBmgcP8E9sLGYPGwZ/W1tM79yZW2bygCGTMXcz0wOVj0+RS6B8SqAswIyoS7MEUwQ22THIzorD06dxSEnZj9u3tuL0iW+w6+8PsPHr5fjk00C4utoxlzR/f3/E03cWz2ZWfGVefAPe9JYoVlAdZkH4/n4TYG09FX5+N6BQaHIAhpQwch8R07PSVgWZLBX2vVZgpMcwxER/Do3qMGis4s5f624Wh5T7O7HuQz+Y1DRGrZa1MOxzN+5u9kYdwkoHkjjMlF+Y0cbJqBEUmAB7+yWoWrU62rdug0uHDiErMRGaO3deqzAlwQxZZs7GxmKBhwcL4qfUzDwBwN0X5MphpnwqiHxVXAL6IoGyBDOki9IP7CwDL2XOzdQ2+luVHYOM9Bg8ebwfjx6dgDRsNurWNUNgYCBu3rzJYaa4ivzbvI7BjPooftmixOBBo+Dh8R0iIlQsI1FBtThYBflwDebNO4cePUbho7V+ePZkf467WfGBjoBGlRmLixd+hvfcEahsWBkNezTE3KNzmbvZ23STKh0Vv+KMymGmfMKMFmQ0CAlJxUiPb2Fi0gBmNWth86efIispiYEMU7LzWBFecCsr5BzF0Ygws5DDzAsAoytHDjP6ovLxeXIJlE8JlDWY0dWtmQdSTq1EAhqy3GSnR0GtPg2FbC7q1TVDUFAQbt26xWFGV3B6sc+oNRZZ6dGIjJgLR0cv+PhcQng4pVcly0xBChhZZ8g/Pgt2dkux0Gcarl7+iWUlex3rDMmMHrD0p1E4sP9zODh0ZkFZbT3aIuRRiNbdjHKE8396JwEOMwV9lkr6+JuMmdFabKkw68yZ+9GooS1qmZpi0azZeHT7NtRFiZN5CdDowgxZZqhWDFlmwhwcoA4LY7VlCq0zQ1ntwsPxnosLrzNTPvUoviouAS6BtyyBsgwzeXVygpvstAMQNCehkHrlwgy3zOhh3AwjVfUhPErZgaX+09CnTzCr+i2Xa5WUgmEGCAvTIDJSQP/+H2LqlMk4fOgraFTHWEaJvA9NUf6mOVFxztQHe/HdxlDUb2CGqrWqov9qF8jSZdpimmIFV71T6SsuhnGYKWloKWi8Nwkz9IOGBosWXUCP7rNhUr0mBjo54+KhQ9AkJ7+2e5lodcgPZpbY2UHq7AwhPBxCRIR2S/sFNI1cDs2772LtoEGY7+HBi2a+ZaWH355LgEug/ElAn2CG9FJV+gEAp6CUzeEwUxRFvaz1ZTCjOYyU+9vhu3AqnJzCEBKSQTXodGJk8leawsLUWLlSg0GDPsPkSVMQG0NxM68PMyQj8nNUZ8XiTsJfCFgyAZWqVIJZm1qYumOqFmjECq4cZvRGAiUJM1RnJi0khKUS1uT8Ml9Q5fes0FBoqM6Ms12FTM38bl9HPKSUyxJprhWjIFmRdSPzFevMiO5ly5YlwHXgapiaNkGHtm3xy4avIaSkgMW5vMTaIoLKq2zzgxkfGxvM6tEDJ+bPx0kfH+2W9vNp1OfM/Pn4x9cX/jY2mD5kCPZt3w5Naiqb56vMQR/6cDez8qcc8hVxCeiTBDjMlI13652kpPsICFiKWrWqY573CKgyDgAqytJV/MD2sgYwuvPRWmYO4tnjPVgeNAOOjn4ICEiCQkFuZmSdyR9k6Di5oUVECHByisSsmVNw6uS30KiOvrZlRpwfy26WHoN/zvyA/i5WqFTpHbR2bQ2fkz4sw5mUu5vpDciQPaqkYKZdSxdsHj0GGVIZNMpwaBQKaBTKAptKJofm/bV436U3rDpPqmBFMydjbX9nPFKGQ60oXFYapRJZSiVOzvFCbbN2BRbNFEEmJOQhxoz+Dg0adEfjBg0Q6r8ET2/fhuYlxTGLAwW6MOPj7o4PXV2xwtERA1u2ZO5m5HL2suZvZ4cZXbtids+eGNS5M+ZPnoyje/cy6KKxizOnsngNh5myoUjwWXAJVFQJcJgpG+98hYOZ51aQg/h4nR+cnCZjxowYRERoWAKAgmCGlBlyRVu+/CGsenhhRfBsJN75A4LqUImBn9bdLBrpz2Kx/a81aN+uGapUqQzr+Vbwv+YPqUoKKXc30xugKSmYadPCGR+6DsSlRQtw3d8X1/0W4brf4gLbtcULEb9iBYLtOqOb5dgKBjMTEGpvhX/8/XDN3xfXXiYnv8W45r8YV/wX48/xo1HTtGUBMCO6oGZi+rQdaNPaFbXNzDBj/ARWHJO5l5UwIIgwQ6mZ5w4fjg3DhyPK0xMfDRmCDwYNemlbM2gQ1g4bBqemTTHAwQEhy5fj759/xr1z5yDcv//qtW9KyMpUmhDEYaZsKBJ8FlwCFVUCHGbKxjtfIWFGm5r5KKKjPsXYMaPQt284pNLHkMnUOUkAXrTQEMiQVYbiZcaO3Ybevdyw+XspsjO0KfBEy0pJbCnjhDorGpkZsfjog8Vo2NAcVWsZYuC7AxF4LxAyjYwDjZ7gTEnAzLLAB2jfzh3DO9rC39EVy/oMKrQtdXRFcP8RcGnZFlbdPOG/5ArkcoL1F5/rgsBdP49roFRkoGPXeRjQuiOWOA9BgNNgBLxEXnQuwEnbpnXvjfr1rREUTP8PiPCitdLSZ5/kN3/+aXTtOh2mJuZwGzAAUdt+LzVLRy7MxMVhtpsbvh85ElAoILxCI8ud8N57COrZE7L583Hp7FkIDx+WK4gRAYnDTNlQJPgsuAQqqgQ4zJSNd75CwgzFp2iy4/AodQ8+eH8xevYcgUmTtkEmewq5XM2UGVJonjc1U2YWLjyPzp3Gw2/xFFy++BM02YcgvEadmYLghyw0asoJnhaDuXOHo2bN6jCzqIWxP47VZjjjQKMXOPO6MENxXCEh6Rgx4hvYO4TAtlcwbF6x2fVegV4OKzBm7E8IXp4CqaxwN0r9hBjRLZQspypMGL8NvRzDQOu37bW8UHlRH2o9e4dg8OB1CA2lork0phb8RPeypUtvwcEhiMXJ2Hbvgc2ffQZNCaVhFhVz3W1+MCPIZBAkkkKbimKqIiMRYGODkLlzceHECahpriVsPdKd79va5zBTNhQJPgsugYoqgQJhRhVb7BqEBemGr3ucdEtKACAIPAFAmXtzivvmstgZ1SFcvvQzlgVMhlWPIZg06VcEBd1EWNgzVjyTshZRCtYVK+5h3rxj6N59FtxHDERs9GdQZ8WVmHtZfmvQZEZB0FBhz20YOMAG1aoZoYVLC8zcNROSZxLw+JmynyXtdWBGFywUCiBcWfSmVAIKOXKSW4hKf/neyuUArbs48iI568pdCzQahIY+xJAhH6BOnfZo2aw5VoVJ8OzmTQj37pWatUMXZmblWGYYzEilEAppKolECzO2tgxmLp48qc20xmGmoupbfN1cAlwCpSSB/GHGH4I6Tlu8khI8laGWCzNyns2s3AHNlcu/IDhoKjpaWsHBwR8TJ27F3Lkn4e19BtOn74frwFVo27YPxo/tj6OHN7Ail0J2LJD1OsUyX7xWrNqqCzbqDAKa44iN+gx2tpYwNDRAl0ldMP/IfMgyZSzAvOyr9BV3hiUFM6KbI7k7Fb2VZ9eyvPBBSTq0LqFFl5NWts9hhuRGVtp0jBv3Ixo3tkVNU1P4zfXG7bP/QChlS4cuzMweNgybRo2CWi6HWiYrtGVJpVCtXIkAOzuEenuDw0wpaTF8WC4BLoEKL4G8MNOkcR2sW+cPjToOalVsGWsxIL1SozkNJcFMPW3RTF5nRg/rzOjCAu1rLTQH8eRRFHbvXIOZM/qja1cLtGjeHs2bW6JDh7YY5maFb78OxIOU3YDmEKNsFDPbG4uJyYyGOiOapWPWziEG6kxqdOzFLHLqDCpwdAIbvwlBx44tYVClCmzm2WDxhcWQq+QcaMqww1lJwcxzBfu/yjs/VzoykUjUUCpVrDBms2a9YWxkhGljx+F0dDQ0KSml7rKlCzNzRozAV2PGIFWhQKpcXniTyfBw9WossbdH6Lx5HGYqvLrFBcAlwCVQWhIQYebatWtwd3dH/frmCAudgcsXNuPSuf8rc+382U24cmUrfBePR+3aNREUFIRbt25BrVaD1qKvrwoZM5Mf0GgIzHAcGvUJPH6wA5f+2Yizp75C0t3foVEfZUWGhOzXcy1jrmNE6+pjyMo6joz0WGSk7UdGRhyysk5ArToOQRUDSlAgzpH5OBLkaI4iInwuWrRogCrGldE7sDeW3V0GuYYDTVm1/XCYKR3QKG2Ao3pSSqWAhQsvonnzvqhSpSqGDRiAmL/+emPuWiLMXDh4kCUb6N2wITzbt8fMdu0KbTPatcPszp3RslYt+MyciUvczUxfv5/5vLkEuATKuAREmLl69SqDmUqVKsHIyAA1TY1hWgYbzcvMrDrz9KG5Llu2DNwyUw4sM7rQQLBBDZqDEISjEIRjEITDoPovzOUrj9VEvPZVttoYmON4/HAvtv0qx1yvAejj0A5WPVrB2ckSC3yG4O+/ViI7MxYQjoEsMuK4As0rOxZpT6OwdMlE1K9nBuN6Ruin7IeQpyGQCTJIBEkZtlGUVdwo3XlxmNE/mCGQobiboKAUWFgMh4FBdfS0ssKOLVugvn8fmrt3Sy1ORjeIXnPnDpCUhJQLF7BhzRos9PLCXE/PV27zZs2C79y5+H3jRjy4erXcpWQWZcUTAJRxTY9Pj0ugnEtAhJn4+HhMnz4dDRs2RKNG1BqV6da0aVM2V6VSiYSEBG6ZERXu8rQla4gmp1Esy+uuTQSZPbs+wrCh9mjV0g7W1osxbNhGjBmzFUOHfoXu3b3RvFk3eLjb48ypTYBw4gWg0WRFQVAdxN2EP+E1exhMaxrDvK0ZhqwbAplKxoGmDKIchxn9ghltCmaKlclEt26eMDY2R7vWbfDz118j7dYtCG8IZERFnbYET+n37+NJcjIeF6FR/ydJScgkAEtMLDdFMnVlQ/scZsq5psiXxyWgJxLIzs5GYmIirl+/ztqNGzdQllt8/A2Qa1xKSgpUKpWeSLngaXI3s1K2LmlB5gR+2aKEvX139HUOhI/PUYSGJkEqfQKZLI1tQ0LuY+7cw+hpvwDW1h1w5PBXEDTHtJainDmS+5k6+yD+ObMZEycMQNVqBmho3RAjv/WAQlCwDGfcQlO61paijM5hRn9gRkzBrFCo0cs+ACYmDdCgTn18vfZDpF67xoCAKc5vuJAkARRZadTFbMyS9IbnnBc4SvNvDjMFf7nzM1wCXAJvVgJkpdG3JgjCmxVSKd2Nw0wpwgwrzqk5irOnNsGpTzcMHhSJpUuvQibLZAU6KfWztlHWJDWk0gwsWXIevXsHwN7eEilJO1k9HHJzE61DNGZmehyio77AiBGOMKhmgKa9m2D05tFQCkoONGXIQsNhRj9ghkBGKqXPYhb69lXCtEYTVDc2wXsSKe5cuAB1YuIbcS0rSOknhb3YrRyDDLfMlJJWwIflEuASKLYECA70qRV7oWXsQg4zpQwz0BzHQh832NhMx+LFZyGXZzHFSatAkRKlbc8VqgwsXnwGDRr0xkcfeCMj/QjLdCa6u2ld4GLw7EkUtv+1FoMG9WRA07xPM4z9aSwHGg4zeWql6AdQlHZQf0HjixYZieQpBg58F6amzWBctRokSwIQf+YMskVXpnIOBQWBVFk/zi0zZUyj4NPhEuAS4BJ4CxLgMFNKMEPQQUUv797aiq5d2mPcuF8RFvYUMinVsyCA+a+SqT2uglz2BC4u69Hb3gKpKXtBmdZoPNE6owWaWDx6sA9bf3sPAwfYwrC6IVr2a4HJf05mQBOmDoNUKIpTFO9b0hLglpn/PuP5Pfdv65j286ZGSMhDuLl9AjOz1qhcuQqW+SzAtdOnkZVjESnrCn1Fnh+HmbegNfBbcglwCXAJlDEJcJgpJZhh2c+EU9izYxVatuyJhQvP5biS5Q8yzxU6DeSybHh7n0Pt2vVx6dwmqFlK6OeuZgQ1ItAQ7Gz5eRX6OveAoYkhWru2xvTd0xGuCYdULeVZzt6ipYbDTNmFGdEis2JFCjw8NqBu3Y4wNDTEAs9ZuHryJLISE1msCiirGLfKlLgMCEIoFqio7nPiNeJ7wmGmjGkUfDpcAlwCXAJvQQIcZkoBZsglTJURBUE4i283+KFt2+EICLjO4mO0bmWFKXkCli1LQY0arRG1dy2yMg+y9NC61hldoElJ2oP//RCB3r06MwuNxRALzNrrCYWaJwUoaWtLUcbjMFPYc/52zosgExychFGjvkHDhlaoXq0aZkyYwEAm+949LchwiClxiGEQQjFISUkQkpOL3ui6+/dz58Vh5i1oDfyWXAJcAlwCZUwCHGZKAWYINLSWmbP47pslsLBww9Klrw4zMpkIMy0Rs/+jHJiJhRg3I7qbPQeaOBDQ/N/3CtjZWqKKURW0GdwaM/fMhDRLqnU34y5nb9xGUzFg5nnclxj/VbTtmwYagf2oEBiYiBEjvkCjRrYwrVEDk8eMwblDh6FOTuYgU4oQJyQmIun8eez4+Wd8/+WXrG364gsU1sS+v3z3Hc5GRQE5SRk4zJQxjYJPh0uAS4BL4C1IgMNMKcEMy2QmnETU3jVo0cIKPj5nIJVSxrJXcDOTZ8Nr9inUrVsPVy/+AHV2LISXzPO5y9k+fL9JDmurDqhUpRJaDWyFqTumIiwtJ36GA80bBZqKADME3gq5BgpFEZtcA7m8sM9CSYMOgQz9UHAHw4Z9jAYNesCsZi1MGj0ap6NjoElNhSYhIfdXf9GViW/vlohMWJ2ee/dwaMcOjBs0CGNsbTHHxQWznJwwu5Dm1bcvpjo4YKSDA1aGhAD37+e6qBHYqG7fRvvWrWFubo6oqChQzYfyknL0LegF/JZcAlwCXAJ6JQEOMy+BBF0LSFH3yYoiqA8h+d52BhejRm1GaOgj9quwNvD4v4qaGJAskz2Gg+NKuPS1xKPU/czFrLD70/00WbF48iiKuZxZ9WiHSpUqoZlTU0z9aypCnoawwpo8KUBRHMVer295hhmCgtDQdCxefBWesw5gpud+zChCmznrAHwWnEVoKKUp/+9n4XkMWUmc01qPKA16QEAChgz5EPXqdUYd89qYNn48zsTEQEhNhfr27RJR2jn85A8/DGbu38fOX37B4B498HH//jjk7Y3YGTNwcObMAlscnfPywm9jx2KejQ0WzJ7NYUav1Aw+WS4BLgEugdKVAIeZUoIZgg8qmAnhJJYHj0P37uPg43MEclk6S81MFhqCF91Gx2SydMybfwT16nfDN18tQWbG0RdSM78MashCo86MQXraIWz9dTW6dbWAgUEVNO7dmGU5W/F4BWQaGc9y9obsM+UVZiQSsqposHTpv+jrrIRFYxv0bOcCmzZOsGnj/ArNCZbNe8Pa2gdLlyZCoUCBGf5KAmrEWk5Ll95m6Zfr1LFE3dp1MHPSZFw8coSDTCm6leUFO4p32bN1KyY5OeGQpyeENWugiYiAEBlZcKPz776LO8HBeNfNDX7z5nGYKV29gI/OJcAlwCWgVxLgMFPKMCOoj+DG1V/Q38Ua/fouh68vuZul5xTNpGKZVKxPLJqZjkWLT6B7j1kYNKgrqyWjIRcznbTML4MZOkcWGnVGNFTZx7Bz+4cMaAyrGqCRbQOM+2kclqcsZ1nOpJrXszrwqwuXQHmGGaVCg0W+52HfwxOTLBrjr2mT8Ps4D/w5YWSB7Y8JI/HHhFH4Y/xILLGzQkeLQVi4OB7hSoIZTb7pyl8fZugHg2dYsuQa+vaVoba5BRrWr4+5M2biyokT0Dx4wCwyLPbiDSr1eZX8ivI3wczu337DOAcH7J8yBeqICGSHhkITFlZgU4WFQSWXI97PDwpXV/h6e3OY0Ss1g0+WS4BLgEugdCXAYaYUYUa0zgia49i/bz2cnXrA1nYWPD33Y8WKRMhkDyGTPYFUmorg4NuYPn0XunaZgD6OnXD18hYIwhFQ7E1hAJPfeW0CgpPYu2c9rK3bw8jIEGZta2HYZ8MQlBgEWbYMpGwXrpLzHsWVQPmHmYtwsV+Idwc4QVi7HkJEJISVq17eVq1m538cPwk9O3lgQanBDLmWaRAWloUpU7ajY8cJqFatPuqYm8N3zhzc/OccVA8fslgLtZgmOAdmONjk7yZWEsClCzMHpk4FVq6EWiJ5qS+hRiKBoFTi5pIlUHKYKV2NgI/OJcAlwCWghxLgMFPKMKMLNKdPbsLM6QPQulVHWFiMgYODEgMGfIKU20VFAAAgAElEQVRevaRo3Xo4WrVqi/nz3HDn1h+AcAzqjAPFAhkRblTpByAIJ3HsyEY4O3VD9epVYVzfCH3lfbHs5jLIVXIONKWIc+UfZi6gb8/5UPaxgypiJTJDQsB+Radf0gtqEgmyw8Lwfx4jYWs5otRgJixMhfDwbMydewzNmzvinXfeYa118+bwnTsXO3/6CVcPHcKTW7eQkZICVUoKhJyUzBRMToDDoabkoYbDjB5qCXzKXAJcAlwCZVwCHGbeAMw8B5qj0KgO4fjRzxEZPhXjxthhsGsnTJnYGx+8Nxv/nN4I4BggHNbG25TA3FTp+yEIJxB/fRvGje0LU1NjVDaqDOu51vC/5M+Ka4apw3hxzVKAmooCM+FOPSGsXM0A5mW+YgL5jMlkoO0Po0aVCsxQDBqBTGSkAC+vODRu3AuVKhnA0MAQBgYGLCkGJcagZm5mhkH9+iFixQrs/+03JF+4wNzOhGfPgEePgKQklqaZwU1CwvPsWdwdrdiJEjjMlHGNgE+PS4BLgEtADyXAYaYEgEG0hBS2FTKjWYpldfYRZGYeQ1raYTx7Foe0NPr7ONSqwyw+RlOEGJnC7knnycJDYyfe3Q4/v/Fo0MCc1aLpOKYjvI94IxzhLCmAROBuZ8V1KcvvOg4z0hfch0obZrRxNwJ5LmHixN9Qr15HVK5sgJ5WVvj+009w6O/t+GrNGsyeNAmd27dHVUMt4BgbGaGGiQka1a+P0W5ueF8qRcy2bXhw4QKrO4MnT3Lhhqw1lPWMGqtGz8GmSGDDYUYPtQQ+ZS4BLgEugTIuAQ4zbxBmCCxYCuVMAowYqDOjoaF99rd2/1XgpDh9KPZGrYpD8v0dWBk5D23aNIFhNUO06NsC434eB7laDrnA3c7yg5LiHuMw8+ZghkCGsqLJ5WqWsczcvDUDmeEDB+LAH38gLSEB2ffv41liIh7cvIl/T51CzNat+Gz1agY3XTp0QJXKlVHN2BhmtWqhQb166NSuHSaNGoUPlUpEb9uGpHPnoKHq848f/xduEhJ4sc1XADsOM2VcI+DT4xLgEuAS0EMJcJh5wzAjggirQ5MVw2JitPvFC/QXx3uVLWVFE7JjkZK0E19vCIGdrSUMqhmgQff66L+qP8t0Fi6Eg9zOiqvAl9vrBClzxSNAofYq9Xo4zJQ+zGhTm6uhVKqxZMlNdO/uBVPTJgxkZk+ejKP79iGdrCh37+Y2dc5+1p07eHTrFhIuX8bFuDjs/OEHfCCXY9q4cejcoQNMqleHqYkJy37WpmVL9LKygueECVi/ciVi/vwT98+fZ65owuPHEMgtLTmZuaKJlhsec/PfmBsOM3qoJfApcwlwCXAJlHEJcJh5SzDzKvBRGn0YOGXH4kHybvz++xq4u/eBkYkharWqCStvK/he9AUBTa7SXsGxhlzvSBYyQQYFlAgXIqAUwiGHQuuaR3BTgHseh5nShBlKuUxpzamlYebMfWjVajCMjMxR17wOFEFBOH/0KDLIYnL3bm68CwEGa3SMoCYxEerERLafmZCA5Bs3cO3MGRzfswdbvvoK4cHBmODhgY7t2qGWqSnMatZEi6ZN0cXSEv379MHC2bPxxdq1rKr9g4sXISQlMbChIpwUc8PukZAAdU7MTUlkBNPnMTjMlHGNgE+PS4BLgEtADyXAYaaCwQwBEnN1y4rF44f7ERe7AYsWjkWDRuaoVscYHdw7YPrOaVAICq3bmbrixtEwGBHkDGAkWRIsueMPn4veWHhlPgIfLGOueQQ2ZKXJD2g4zJQOzJA1htIuy+UCli9PxsiRG9CkiR0MDarDqlNnfPPRR7h9/jyycoClIAtJLtjcucPiPoTERGgoo9m9e6D99Fu3cPfyZfxz+DD2/fYbNn74IUJ8fTFm2DAGNzVr1GCpntu1aYNeNjZwGzAAQQsXYtOnn+Lo7t0soYCKxkpNhfDggRZ0CKAIsCpoQgFdmKE6M5rISJbdjtIvF9RUEglUCgXi/f15nRk9VDL4lLkEuAS4BEpbAhxmKiDMaIEmGpqsWGSmRePSxZ+waqUPunRuxdzOWjg1x9D1QxGcHAyFWgEqsJmfsl5uXcogZdYYpaDEitTlGP/raPQO7oVuU3qi65g+6Dq2N7rPskW/lc7wjJ4OSYYEckGhdT/TsWRxmCl5mCGQIWuMVJoJH5+zcHIKQf36nVmsy8ghQ1jK5Sfx8czaQgH6RbFikBUlF3Bon+Dm/n1tS0xE2r//4va5czgRFYU/N2/Gx5GRWOo9D+6DB6ND27YsoUD9unXRNcdqM9HDA5KAAPzw1Vc4tncfki5chIrGpTTQBcBNUearj30ZzGzdigl9+iBuxgwI770HjVIJITy8wEbnNatW4XZgIFYOGQLfefN40czS1gz4+FwCXAJcAnokAQ4zFRRmngNNDNTZsUi88yc2bgzDwAG2qGJYGeZtzdHT145lO5NmSJmbFQMaoTwjjHZttE6FoMSii4vgGOSIzoNc0LOfN1wHr4H7iI0YPvxLuLgqYOUyGV1HO2LwusEITAzUWrJ0XM4qEsxoVq4q9Bd29su7VMqKJG4eWbTUzFprjJBrjaFsZZ07TULNms3QsF59LJg1C4d27Wa1YghCCrLGFAUAaAyWsUx0TSO4oXTNSUlQ37vH4ObfM2dwcPdu/LhhA3NJ854xA4P79UPbVq1QvVo1lkjAqmtXDHd1hff06VgtkeC377/H6ehopFy6pHWBS0mBhgCHkguQ5YbuW07d0miNu379FcNtbLB5+HBcDQjA5UWLcPUl7cqiRbji749oT08sdXTEwjlzOMzokZLBp8olwCXAJVDaEuAwU4FhRgQalhhAFYsnj/Zh5471mD59KBo2NIeBqQEsR1li3A/jEZQUxNyqyruVhkBGrpFjSfwSWHlZoYvNKIwcsQGLF15ESEgaZDJycVIjODgVc72OYMDASHRwHAAXhQtWPFzBoE9MDlARYKZfTx9E9u0FzbvvQy2TQ6NQFNKU7PwPY8bCztL9lYpmPo+NUcPX9zKGDPkAzZr1Qc0atWFvbY3IkFBcOX4cGgrAz7GuFAVaXrWvCDcMNOg+5JJGcJOczKw3z+Ljcf3kSez/8098s24dli9ejKljx6KfgwMsWrZk8TaNGjRgcx7v7o6AefPw6erV2LFlC84fPIjUK1e0MTY0Hq3l/v3c9ZQXuCGZxfz1F0Y4O2OqtTWWDxqEwP79EfSSxs4PHIgFffpglK0tpAEBHGZKWzPg43MJcAlwCeiRBDjMVHCYIaARoYbczrIyDuLy5S1QKLxgbd0ehoYGaGTTCANXD8SCMwsQ9izsucJe3qw0tB6NFNJ0OXoF26NN+4GY5bkXMvlTKJXk3kQgQw2QyQUowzVYseI+PEZ8hZbdesJ907Dc5Alk46kIMONk643l9t2QGBaGuwFLkLgsoMB2l84FLgNtP3Vzg1X7YS+FmefWGA1CQlLh6bkf9va+qF3bglk8JowciS3ffINH8fEQCADesDXjBbghlzaKj6F5kAvZ/ft4cuMGLh85gp2//IJPVq+G35w5rI6Ng60tWjVrxhIKNGnYEM69emHm+PGQ+Pvj+/Xrsf+PP3Dp2DGkXrsGFVlpclzdCAREWBMtRq8KYmWmX2Iibp06hY0ffYTIkBCsDgvDytDQQtuqnH7rIiNZDSDkWN+YBS4xEVTYtH3r1jA3N0dUVBSys7MhCIIefRXzqXIJcAlwCXAJFFcCHGY4zDCY0QJNDKgejaA+hMz0KPzvh3C4uTmgdm1TmDQyQfcZ3TH5t8kIvBsImUoGmUZWvmJpNFLIs+WYc9wLJvUaYNLEPyFXPINUpmaZs0SQIZiRSCmblprVNVm+PBEu/Veiad9mCLqvdTdjSQE0EsghR2ByENoMbAODKrXgNvQLFrwuk1E9SS0Y0Xj61LQ1XTTw9b+M3rYL4NK8OWQDh0HiPADSvgMLaa6QOg/EGMtu6NzOHYv9bkKpQI58n8vheWxMNvz8rsLD40u0aTMINU3rw7JdO/jP98Hxffu1QfWk7CckFCk+pjSUewY3usH9ZFkh9zFKAJCUhKfXr+PyoUP463//w/tyObymTMGgfv1g3a0bmjZuzGJumjZsCBdHR+aStio0FFu+/hqHdu/G9dOn8fDGDS3ckEWI1qwDN3Rvpti/Qq2X0lh7kcZMTGTgB3KrK06j63PWyWGmuF/9/DouAS4BLoHyIwEOMxxmcmFGtNJoMqOA7FgImmM4dfJ7zPMeCUvLljA2rvrcSnNqAcKehDGgYW5V+m6lofmrpZA/lcM+zBqtGo+ERPqQuZS9HDo0kMpU8PE5g0atrTH6t+FYiVWQqLUpncsnzGitVEFB9+DuvgmdOs9E5y6z2LZTp5l4WevYaSaode7siYEDP0BgYCrkcoIZLdiJ1hgK8l+x4gFmzYpi1phatVqgbu06cO3bF5+/vwaJly5BeJCSG7AvKrdlaSvCTW5aZoIbsto8fMjcyAhuLsTF4ddNmyAPDmZWG0e7nixTWsP69WFUtSqaNW4M13794OftjU9XrcLf//sfzsTG4taFC3jy779gNXMovTSBTU6KaRFsyjLcsLmRNas4TQfY2DjcMlN+NBK+Ei4BLgEugWJIgMMMh5n/wEyu21lmFATNUTxK3YuvvlqOAQNsUbduLVSra4yOYzsyK03Q3SBIs3PcswgI9BVqaN4aGUIfhKKRSz0M7v8lFMq0QmGGWRDkagQH3YaV9XxYL7XEarwHiar8wkxBViQZud8Vob04jtaFTyZTQyrNQkBAPEaM+AKNGvVEDRNztG7RAvM8Z+Hwjp3IFi0TZcAaUxR4IsWbLEgENyzTGgEIwc2jR2ybFh+PczEx+PGrr7Dc1xdDXFzQvXNntva6tWuzhAIEN8MGDMDyRYuYS1rcH3/gyokTuHf9Op6JxUFzauew+joU25MTR0TbosxXH/pymCnGtz6/hEuAS4BLoJxJgMMMh5l8YUbXSiOo4wDhGM6e3Qxf33Ho0L45qlUzYhnPnEKc4HPKB5LHYdp4E4o50eN/gcmBMLM0x9SJByBXZDOXsBeV7ueuUHRc6w6lRkhIMgb0fxftJzfHKqyGRF1+Y2byk0dRIEbs+3wcAhlKt6xBWFgKvLzi0L3bDBgbm6G2mRkGOjvj+08/Q9K1awC5beXUatEHZftlc8yFG4IQArN79wBySXv8mG2f3biBc1FR+OHzz7Fk/nyWSKBNy5agJAK1atZEdWNjNGnQACNcXSELCMAvGzbg7L59zGrz4NYtZJCcchqz4OSADc2JQUA5gBsOM+VMI+HL4RLgEuASKIYEOMxwmHkpzLxgpVEfQXpaNLZtXYUhg+1halodhtUM0Mi2Idy/HKFNT5wp12uoCUoOgnlHc0wevxdyRdarwYycAtST0d9lNTpMbVnu3cyeQ8iLYPfqx7VWGK37HkGMGjJZGvz8rsPFRY5qxvVhbGSMBnXrInDRIlw/dozFnrCClm84yP9lMFLS50gxJ6uNOgduKOCf4AZPnrDt0xs3cPbAAXy7fj3me3rCrkcP1KtTh2VJq2ZsjCpVqqBR/fqs7k3kihXY8cMPuHH0KFLj4/Hs3j1kU+Y1smrl3IclEdBzoOEwU4xvfX4JlwCXAJdAOZMAhxkOM4XCzAtWGlUMs9L8G/8HwsO90b59MxgYVEHVGoZoNbAlpm6fCnmWHEqNMjduRG8sNeRmlhqGJq514er8cY6bmeqlgfqim1lQ0C107z4HNoGdKqybWVFghpIIEMQolWosX34f7u5foXbtTqhcqTKLFXHrPxD7t25FGgWIkxJejiGmICgiRZ25penCzcOHwNOnwKNHrM7N6X37sGHtWsyaOBEd2rRhrmhVq1aFQZUqqFSpEpo3bYoxbm74QCpF7NatuH/uHDLv34eK4nbIxY3gJiGBZQPLje3RI8DhMFPONBK+HC4BLgEugWJIgMMMh5lXhhldK40m+yCysw7h9Mnv4OMzCubm1VGpciUYmxmj67Qu8Dnrg0hNBJRQshTFlKa4TMfT5CQAUD5TwinSDk1qD4REmgqpRAVpTnB6/so6JQDQYP78M6jTtCMm7hzDLDNh5TgBQP5yKNxKQ+An1owJDwfCwh5h0qRtaN3aFZUrG6BKFQN0atsBmz75FI9u3mSFJNU5Cn1BCn9FOi6614mWG5YKOjUV6sePkZ2aimfXr+PUrl348r33MG3sWFbbhpIIVK5cGQYGBjAyNETLZs0w3sMD6yIicHj7djy6ehUgQHr2jAESZRcjqFHduqUXAMlhphjf+vwSLgEuAS6BciYBDjMcZooEM7lWmqxoaLJioMqMQ0ryPuza+SHGjOmLSlUqwcDYANXrVUfvJb3gc3YBFFkKBjWs4GZZhhpBCplahsXXfWFkVg3jRv0KpfIZpFIVwsLIkqCrsAsIC6PUzCosX54AB8cQNO7bAGFpUm0dnnJcZ+ZFOejKJP/95xAjgCAmNPQRpk7bgS5dJrK4GENDYzRt2AgRK0KQcPYsMsgSQ7Ee+pRu+C1YMxjc3LnD0jUT9FHtFVVSEtKTk/E4IQEJZ84gZus2rAuPwOTRo2HRqhVL/2xkZATTGjVQx9wcbVu1wpSxY/Hx6tU4unMXHl+9yiw2zPpDkEMpoMkt7fZtBjkED2UpkQCHmXKmkfDlcAlwCXAJFEMCHGY4zBQLZkSoEbJioM6MQUZ6NG7c+AObvpXCybk73nnnHWalMbcwh0NAb8w7NQ+yDFnZhhpBylzjlFnhGPRRf9RpZIFp03ZAGZ7GlHCK8SDLgkSqYQU0IyIFhITew2DXtWjcuQ2m7pqMcCGCjVGei2a+KszoQkxEBEHMY0ybthPduk1DzVrNUbVqDTSsVx8LZ3vh8M6dSL19WxvcrxOoXpEsL8VdqwgXpNgTBBLYiCmbsxIT8SghAYlXruB8XBz++r//AxWgHO/uzuCG4mxqmJigft26aNG0KYvDmTt9Or5auxbHd+3CoytXWBpplpSA4CYpiSUPEK1DDCbeAsiJsuIwU4xvfX4JlwCXAJdAOZMAhxkOM68FM1rXsxgI2TFQZcTg8eP9OHfuf1j30RI4OHZBVSNDVK9bHQ16NIBjoAPmHPKC5JkE4UI4s2CQ+xlzQSsrGdBy3M1CU0LRR9IHTdt2R3/nSCyYfwFSaSbCIwQow4GQFU8wc1oUbHr4oJVdN4z81gPSZzKtK11OempaV3msM1MYzDDgk2gglwuIiBAQEvIYU6b8ie7dZ6BuXUsYGdVC00aN4TVlKv7YvBm3L1xAdk7WLVJO37aCLCrK+roVZcgsNwQ3OhnN0hMSkHzjBm6cPYvju3fj5y+/hHLZMowZNoxZacgtzaxWLTRv0oTVu3Ht0wd+c+di4/r1rP/Dy5eB5GRtOmlKTvCW4YbDTDnTSPhyuAS4BLgEiiEBDjMcZl4bZpiVJisGQlY0hOxYqLNikJK8B8ePf4vVq3zQp09X1DCtBpOGJmhi3wSOyxwwc99MrEhdAaWghEJQQCLkQE0OCLzVpAGCBHKNDIG3A+H+pTs6j+yJtt1d0a3rLNhbB8HOKgCdukxG255OsPFywOTfJyPkQQgrIErrEOdesWAmx2olEaBUaltg4D1MmvgLevSYhYYNezCXspbNWmDW5Cn46euvcfXESaRRcLtOsUd9BYiyOm8CGl24ocKaJG/W7t5F2q1buHv5Ms4fPYrY33/Hxg8/ZDVuPIYMYTE3NWvUYBnT2rVpw6w2HoMHI5hq3Hz+OU7u24eHV65o4ebhQ1Yrh9zS6H5vynLDYaYY3/oV5BKNRgO1Wg1BECrIisvGMkneXO5l472oSLPgMMNhpmRgJkeODGiyYqDJjoOQHYd7iTsRG/sFIiK8GdTUNKvOoKZZr6aw87HDxF8mICAhgAENgQ0lCZBQ8LwOFIhw8Ea3DGjkDLjmHPbCyO9GwUUxGI5LB6JPoCtc3xuK8b+Mx8J/FiAsLew/IENzrQgwI1phZDIwiCFrDKVY9vDYgC5dJqBRwx7MEtO6RSvMmDARmz77HBePHMXTm7dYmmBWtV6sAv8W3ZXKKoyU9LxywSYn9oVlM6O4GMoal5iIZ//+i1vnz+NEdDT+2rwZ6yMjmWVmaP/+rHgnxdo0qFcPXSwtWd2bCR4ekAcF4edvv8WZ6GhtzA0BDdXLoWxptE+WISoUWgoZ6fQBZki5K4sKdWFzEmGAtmJf2tf9u6SVJbpPSYxP4+i2kp7n64xH6yNlnxrt6/Mrv/eqrMpdn+VMcxflWtx10PXl+cVhhsNMicLM81iaaECEGlUcEu9ux57d66CQz8GgQT1Rv745jM2N0cSuMay9rDHq25HwvbQYUpWUpXWWC3It2LzFhAEEVDJBBrmgQGh6KJbdXQq/q4vhf90XQcmBkGSRBUehdZfLB77KK8yIsTAUQ6RQUNMgNDQN8+cfx9ChH6JTp3GoV68TDA2M0bl9B8yePAUb13+Ms7FxePLvv1oFl6wx3KUMJQ0rRR1PfA8oyJ+uZXBDCRhykjA8vnED10+dQuzOnfjhyy+xcsUKeE+bxgqZtm7eHNWMjNC4YUPYdu+O4a6umDN5Mt6Xy/H75s04FxvHLDdqeq+p2Cm1EoYbDjOlp56Q8pNXWX1dhaqw2ZbE+OIY169fR0xMDGhbFl7ivGhLcs0r27Iwx6LOQVyT7nUk76ioKFy7du21FXDdcfk+l8DLJMBhhsNMqcCMLtRo3c8OAqo43E/8GwcOfIbVqxZg9Ki+aNmyEaqaGKJ+l3roNrUrBn0wCLP3z0ZQUhCzdpALGgHFW3NDI7c3ynImyLTWI40SCo0SImyxdNMFuMaVL5jRupERwJD1hSBGIsnEkiXxmDr1T/Trp4ClpQfMzFqBCjj2srGFj6cn/u+LL3Hh8BHmTkaKLHMp4xDz1iGmIOghOCCwYZnk8sCNipIJXL+OK8eOYd/vv+ObdesQsngxZowfz6w0lEDAuGpVFm/jYGuL8SPcscTbm2VK+3vLFlw4fBgPr14FwY0mJUWbWIAKg76G5aasw0x6ejru3LmDxMREZGZmvuy7+I2ey08J1Z2ASqViEHDo0CGcP38eT548QVZWFu7fv49bt24hLS2NdadxSvKVmpqKf//9Fw+pDlIxrBbiui5duoSwsDCEhITg7NmzZUappmfg6tWrDLJiY2MRHx/PLDQlKcM3NRY9IzT/u3fvIjs7O1fG586dQ3BwMJYvXw7aF9+TNzWv8nqf15Hj61yrL/LkMMNhplRhRhdqQDE1qoMQ1Ifw+OFeHD2yEWvXLsH48QPQpUsbmJgaw6RRdVgMtoBDgAPG/zgOi/5ZhJAnISxlslwjZ1DBXNHIElIARJSKOxq5v+XE9RCkvIobnP7DDAGLwIqGagGGsripEBycDG/vI/Dw+Aq2tvPQvHkfVK9eDw3q1Uf/Pk7w9/bGb99vxq2zZ0HZtAQKGOeWmDILMAWBDbPU5NT5YS5iBDcEpMnJ2hpA9+7hybVruHT4MHb88gs+XrUKC2fNYskEetnYoGmjRiwFNEFO3969GfRIly7Ft+vXY/+ff+LKiRN4TL+aE9zQmJRM4N69IrmllVWYEX91v3jxIt5//318/PHHuHnzJlPsiqOkl4RCISo0BCLHjx/HN998g7179+Yqm+J5Urj//vtv+Pr6YuTIkQgICMDhw4fZ/D/99FOsWLECR48eZVMqqbWI4/z2229YunQpdu/eDQLBorxo/jQOAZdcLoebmxuTewpZBN+iq584LwLa7777Dl5eXhgxYgRrBFtPnz5l8yvKWt9mX1GWDx48gJ+fH1avXo17VHw3x9pEIPrZZ58x+UskEgY7dI6u46/iSYCe4f/973/YtGkTk6f4Hoij5f2bjovHHj9+jP3797Nnjz734jnx2vKy5TDDYeaNwIwu1DBLjSoWgvow0p9F4+KFn/DN12GYO2cEnJy6o3GTunin0juo26kuus/ojiEfDcHM3TPhf80foc+0gfZ5weZV4KJUICc33P+/o+sjzOi6kMlklJGMAEaDFSsews/3ArPCDBgQCUvL0cwKU6O6CYufGOXmBvmyQOzd9jtTUOmXd4FclbgVRi8hpiDAYZYbMf4lB27IfUzz4AGrSfPo2jWci43F1v/7P6wOC8PsSZMwxMUF1l27oknDhqhlaopWzZvDtW9fzJ8xg/X57ZtvcGjPHsSfOYOn8fHaGkM5cTwi3NB9xaY7t7IKM2IA9J49e9C/f3+4u7szgCAFg869jRfdm5RKUjwjIyPRsmVLzJgxgyk9oiJKfY4cOcIU7T59+mDChAlYtmwZU4b++ecfzJkzB46Ojti2bRtbQkmtRRyHftFv2LAh1qxZg0ePHuUqZK8iL1oDjfP999+zORJ03aYEI29RkRZlnpSUhA0bNqBnz54YNGgQZs6cialTp2LhwoUgZZP66ctLfFYIzs3NzZmsL1++nAszdJ6eMbKMOTs7M2gmyw0d56+iSYCeC2pnzpxB9+7d0aFDB+bCR7IUPzM0othPd3Q6Rv3I3c/b2xvt2rWDQqFgXXSv1b1Gn/c5zHCYeaMwows1mswoaLJjGNSosw8jNXkn9u79BBLZbLi69kTHjq1Qs2Z1GJlVRVOHprD37QmPb9zhFeOFpdeXai02GhlzR5NpnqdFZmDzJq02BQCNfsDMc+sLuZBpAUbNrDEhIQ/h738JM2bsgZvbethYz0GDBlYwMamNZo2boLeNLWZOmIiNH3/Cfp2nFMAU+M0KLVKwd0JCuVLkdZVovn+Xvbci3FD2MgYbFBNDgf85zwFlOzt94ADLXicNCMAkDw8429ujc4cOLJFA9WrV0KZFCwwfMADL5s3DJ5GR2PHjjzgVE8MSETz991+oyA2NfvmlRhY+es504Ib+Vt2+jfatWzPlivz1dd1e6As6vy/70vziJmWB7kmWjwEDBsDDwwMnTpxgx96WIiEqN8nJyfjiiy+YUh0YGMjmRDeRqLcAACAASURBVEoPNepDQNG+fXt8+OGHzMpB/enXdrJ4EMSQdYbcuOhF15TES5QJWSoaN25cZJgR504ucASOrq6uzLojrqsk5licMUSZU+wOPQdOTk44duwYc9MjWCMgIHctfXqJMhVhhoDlCtWjynl+xPP07A8dOpQ9++RaJ8pCn9ZaVuZKnzf6P4QsevT8iDKm+ZEllWCZoFj8HNFxUd70PpGFbNiwYcxSSed0+5WVNb7uPDjMcJh5KzDzHGpioMmMZg3qOAiaY9CoDuPf+K347jsZZs4YAnv7Tiy2pkaNajA2N0KTXo1h72uPUd+PgvfRuVj27zKEPgxhyQMIamRqLeCI8SxvC27KJsz8F15kMjWDGIkkGyEhDxAQcA1eXvTr7BewtpqNOnU6oXq1WmhQtx6rPeI2YCCUwcsR8+dfLIaCZa5KSWHKrSZHqeXKvlbZr2hyEOEmN3MZuY5RZjOKgUhKwqPLl3F63z5s/vxzLFu0CG4DBjCrjUWrVqhjbg4DAwOWEppSQ4f4+eHbdesQ9ccfuHj8OBKvXmWxV2JBUBZ7Q2BDPvu3bhUIM/QlT8p4XsB53S9Pup4UhryNjpOyQMdJoRMtM/nBTN5r6e+XvfLrn/eagvqIx0l5JuWHfk1PoB8cdNZBihG5Z1lbW2P79u25SpN4D3GMguYonhe3BfXTvSf1pTnRlmCmUaNGRYYZUd5ffvklOnXqxH6B1nV9ym8e4hx1ty/rJ54T+4t/F7QV+1Gs0bfffotmzZoxlz1RERXP614vHtPd6p7X3Rf76B6j/YKO5z0n9hO3ecfR/VvsQ1uSNa1BhBkCNF2YoT50ntzQlEolOnfujI8++ogdK44SrXtv3X3d+enu6/bR3dft8yr7dK340h1H3BfPFbQV+4nbgvrRcbGP7la3P30uycJCchbj1cRnno6RWx9ZgQmQSfbiODQGPX8Ut0f96HOf936698m7rzuO7jnxuLjVPfe29jnMcJh5qzAjQg1tyf2MWWuyogHhMATNCWSlR+PMqU34+OMAjB3bD127tEHjxnVBYGNUqyoa2DSA3QJbuG8YAc9YTyy5vgQhSSGQpOfE1GikkKpzWk4wvwg5pR1z8/ZhRhvvQhaX500DqZQsL9qtRPIMK1bcg6/vZUyfvg9ubh+jRw9PmJt3hJFRddQxr43WLVuin4MjAhcuwp/fb8bds/9o42BISSV3IB23o4qmvPP1FgxtogVFnZAAagQ6rA4NPTf0pZucjEdXr+LE3r3Y+PHHWDh7Nnrb2qKDhQWLt6lVsyaMjIxYSujRw4ZBERyMX7/5Bif37sW/587hwc2byCALTWIissgy06aN1jJz4AD7Aqcve2rkJ75161YWjExf7OKX/et+6dIXOY1P41ET7ycep+3LYIbOi9eJY9CWjuf3EscV7yNu6RrxVVAf6lvQuHQtnaNxCProl3Y7Ozvs3LmTAaDoIiSOLYKHeE9xK56ne4nrKeie4v3ENYigSa5hRXUzo7GoUYzNqFGjmFWJYm8KkyWdF+cqbvPOV5wn9RXlJM45b19RDuJWvJZiYihmqkmTJpDJZEym4nMojpt3bN255XcfcWzd68X70jFqea+jv3XXKd6jsPWI9xL70ftP+wXBjLgWuu6XX35hMEOxV0V1GxTHyTtPcQ3ienW34hrFPuKc85OH7nV598U1i9flnQP9/bKX7jzEOdCx/F66fXXvU1B/cQzxffj999/RtWtX5g5KCRnE44VdT/d62Tro+vz65J0v9SnsXuKcS3PLYYbDTJmBGV2wIahRZxwAVDGA5hggnEDa0yicOvkd1q9fijFj+6J9+6aoV9eMgY1B1SowqmOE1kNaoZ+8HyZvn4xF1xZh+b3lkD6SQp4uh0wlgzQHblgtGzHts+iSJsJOAW5j/42KefmRNwszIrAgB1w00NaAIXDRwgsF78tkaZBIHiIoKBE+PucxbtwWODoGoVkzJ1SpUgPGRkasAnyjBg1g1707li1YiB0//oykCxdZ0DcePXqh6jtTUHl9GO5O94rPgGi5YUU1c+AGVHDzyRMgNRWPrlzB0V278MWaNfCaMgVWXbqwWBuy2phUrw6TatWY5Ybq26yRSrHnxx8Rf+wYkq5eRTsdN7OMjIxcoKFUsf7+/pg9ezYo7oMUSVIwivsSv+RJaSAFjSwAFNxN7ljPnj1jCgCNT/1eBjOkBNCvrHQdZYSicShrWEFzI4VfvB/1J7cv+lsEAbofNXFMmhM1+oWcjumOS3OnY3Qtveg6kgvBDP3STjDz119/sT4kS7qW5kt9CBp0x6Lr6Rwdo/nTvGgtunPLK2vqS+NQ9jKanyi34sAM3ZtelLWsR48ezB0nryuO7v1prXQNQQbNlWREAdbiOum8+CI50dxIVnRcnLe4NrFfflvqT/IlmYowQ+4+NFbe90OUH82Jfj3XnROd050T3Uv3/ct7ju5JshWfC3FudA1ZKekcjUnrFeVP9837ntJ1NDb1pbHoWnqvaEvPwctghq6hRj8kjBs3Dra2tiyJBI2X333EOebd6s5T/JzRnGk+dE73RWPTGkUZip8p6k9rzSsn3Wvz7tM4JCeyiNB8ab3i+um9p/N5709j0D2o0fyon/hZoGeAxsh7Df1Nx6kvrU/8XIsyFudMW5oLzUk8Rvegv+lzSs89ZY+j/+vo2aLjNEdxTtSXjovH6L70N42Zd06iLKgvyY36iC/qS/enzznNV/ycU19xXmLfN73lMMNhpkzCjAg2orWGwEZQxwLCUUA4iayMWFy59D98tzEEs2a5oUsXiq8xQfXqxjA0NEClSu/ApLkJ2rm3RT95X0zYOgELzi/A8pTlkD+TQ5mphEKlACUSIAsNgYdEldOoaOcrZix7Gc6UPsyQyxhBixoSiYo1srhQAUuq/aJUqqBUZkIme4agoCTMn3cGY8f+hD59lqNVq0GoXr0xqlSpwgCmhokJ6tetg0EuLlAuD8GB37ayX801BC85AEPB/PQLu5i2l1slCrZKcNm8mmwIbuiZUt26xSx8ZLlhz9vTp7lwc+zvv/HZqlWYPnYsOrVvj5qmpqBYm6qGhqy1ad4cU0ePZskFqKgnZeIi5YG+hMUvWYoRadu2LVN0T548yZSNoihV4hczfWHTdaTgnj59mmXNIl90ioeYPn06cyeiuA26N/UtCGZoXuT6QcHqnp6eGDhwIPOH/+CDD1i6W3HeNAY1UngoCDg8PJytgVzXxo4di4iICOZDT8oK9SNl84cffmDgRnEj5DK2aNEi/PTTT8yljPrQ2JT++EeKTTp1ii2NjlMaZspwZmlpCQsLC+bytXnzZiZPWhMpWJTVjH4JJqWLXnQdKTgkExqT3IloXuTbTxnFaHyaO/UTX9SXAI5+tadMaZQRi7I0kTyKEzND66EXuXK1adMGQUFBTNGmeeVV1MT10/zJNWf06NEsvsbHxwc7duxg86LxRKWNlEN6j3799VemJNN1ZOUj96nC4kDoXjdu3GCyp0QLFCxPsqGsVCR7khf1IXnQ+0e/qlM80iRKmjFkCBYsWIA//viDvafUR5wTzY/60rxIvuJ7L8qXgJ3mSM8nnaMXXUvzpeeLxqT3ktZLslq8eDGThfirPs1JbHQdPesEJZS1jJ6lVatWsfgvkgWtKa+bmXg/upbcGAlQ6X355JNPctcrzvVlW7qenhOyEFISiuHDhzO5LFmyhD2H4meMxhDlSPKmzzolV6DPFMVP0bWk8BMwiH1fdl86R+87yengwYMMdOPi4ti6af30GSRwzgsnosxoXvS+rFy5kj1fNAd6rumzowtV1J/GINnSZ4WsVzRnSrpBMqaMgWJ/ggfKSEafGYI1elGmRPqb4t5atGjBPnMEzfSZpbnTGuhFY1B69Z9//pm5qtExksWWLVvY/0+0T3PRfdHf9JmncaKjo3Ofvf9v7zzgo6rSNi5dEFGKbe3oWkB0VaQIghIUBBuKIsonTUEQFQQR0TQIJBS/ddeGa18/ZV372lGRkNAEBUWwYEGB0FFUIJnMzPP9/nc47DAmIXUyybyH3+VO7pz6nHPnPs9533Mu4xChzXo66svvC5sKkL8bh5F5hedbkZ9NzJiYiWkx40RNuBsawoaNAxRcJGmpFFii33+epYXZM3TftFvUo0cbHXhgPdWsWcMTNfvtt584ah9Y23NLO3vY2brssUs1KGuQRv00Ssl5SZoUnKQpwSnKUIbSla40pWmC/it0kvxJSsoPHZ5VB8HjRE+4hcdZeRQSSORx1+ZxOuHCE1S71kHq2eMfnosXgiPk+sU5/AitaQmJlJBQSUryKykp3ztCwgUXMd73IqWlSZMnSxkZHEFNmhRUUlKe7rhjtQYNzFTPng/rjDMGqnHjlqpZs4GHA1jUqFHDex8M5HDEoBv10hNP6adPPpU2b5Z4hwQ//LiQrVmjPbPoxZx9NyJfPCJvOP0RJ0/crFlTsLj5+WdtX7VKi955Rw9MnqxrLr9cRx9xhDeWGc/uPoeIQ2YgqxAFiBsLaNkSlzU57NIFMeChCzEs7sOXeBzM9D755JM67LDDvMmAZs2a6ZhjjtEBBxygmjVreqQCQkmIFDMQNMpkvQokkMkE1oiwyxViq27duh65pt7EdfEhImeccYYOPPBAz22nXbt2Ov7447307EoGIeHo27evGjVq5NUH60qrVq08womLEySMAB6Ih4YNG3qE2bUfEkX+tAE83cGsOmKId4ZAytkVCUFDOsgYAaFFPOrP2hDIFXi0b9/e2zSAuK49kHjycb9D1OPggw/21umAA5/ZztqRLFc/r6AC/oNgESDmhx56qB544AFPzJLOpXWfIZq834U1QbQTXBBv9erV89rL7mKsTaCPyJd2gjXijM+QTXA55JBDNGfOHC9/V3541Vy5iEOwoCwO12ba+Nxzz3llkI7xSJ3q1KnjbYDAWGjSpImHBeMWYgp+EEbILPmSHyIM4U4d+J7ABg5Nmzb1trimrwmkY/MG+uSCCy7wRAmfESOMQerVvXt3734hH5cf6R9//HGvDbSb+4c04Ny7d2/tv//+3v0UvmaG8hzekHCIL3ix1TfB1dP7o4D/+J70zPojhCiXcYkr1VlnneW1u0GDBsKVEKLu7mEINf3EGOR+ZCzRt9wPbGiBiCRf4hcWXL8hvLgvW7Ro4U0egBV9Q13AnU0qIPlO0JCOemMRQUyRrn79+jrqqKO8XQPBCQwQOMR1IpMxhdAjLmu9uGepO+ORSRJEEwHhgvsnvzXuZaT333+/1w/Uyf328ZmDdrN+jID4QCyRlnYRGDPcx9SJDVPob9cGzlhtyJ9xxBjjGriBMb8pjAPGNb9B1B2sEZ3Ecf3nFRTF/0zMmJipMmJmL2HjC20c4N+VKX8u76+ZJ79/kXJzP9Zvv83VD6te1Bv/maa0SUN1Ve9OOu74Q1W3Xm3VqFVDterWUp36dVT3gLqqf0h9NftLU7W45lR1uuc89Z55lQZnDdbotaM17udxnlvaJE3yBA5Ch3+Td//juhM9qUrdyzmNTQcQP7i2jVk/Vs27NvfETI+LZ3hiI2RJwaKCYPmvixgix4kUuAfHf8WKlJ4eupaSkq/x47dr1KjVGjx4nnr1+qc6dmQXIrZMZr3LQapb9wDVrl1ftWrVVY0atdSscRN1attWI28aqmf+/qCWzHpfm7/6Wjs3bZJv27bQez7WrfN2htqzxsEEjLmQVeIYYN3NXhbBDRvk37JFebgm5eQo5/PPNefVVzUtMVEHNmiwh2xAPHjIQrwgQP369fOsJ5AEHsQ8kJmZ5CHtHuRFPXfdg55Zaog9JBjywSJ5rCGQNmaqIZnOukJ+kWLGlcesZ0pKyp5ZcggqeUDeyBuLBbPMBIjssGHDPLKLCCEuB+Uye4tFAnKE4EHY4FK3cuXKPfEgmhANXK8on3Ig0gcddJAXl2uQEGbRya9169YewUIMkg/lIywQhEOHDvWIEu0mgB3iDsIPWSIN2yFDgt98802P/ONmhJWAgPWFdSOQQ8QZBJtyOSC91B+yOH369D8IEi+DAv6j/gQsGogE3uXi3HHcd5ypK+1DdEFU6UfqCR60kW2nIanMotM/BKx8kH/i098QT4QW1j3GQlGBMsEGoUeeiBOsd2AKOeU74vAZLNh2l/EDRtQJ4ojVhPGAoCE+RJFyaSNjHKsWfUPb+I6AKxvrjrBIODED6cY6AXkFI9oEUcYNip3Wzj33XK9P6APGFXkxJhhTtBviythE0FI/0iDGqQN5YfUJJ7G0i4NxSV0pF6sU1/YVXFrEDFZGxClWI8Y8GCD8wJKxQrnUkzS0G3LNltCMQeLTfvBlLBZnBz5XPyYr6GvIPCJ/9uzZXlmMhy5dunhjFOGBqx3tdm1HQLDhAcIQyxd9Q52x8HAfIAIRFPQX2JAX5WBhAXd3X2Ml5V1LWKaoE5MfWFpJ7yx69B3XuVeaN2/uWWOpJ+MGwUO5lAOOWHCZfEGYEsgTiwr58XtFObTBjSPGBmOVe4U8iU9bmZzg9/Ppp5/2xg7jFEwQM9dff713XxCXvKIdTMyYmKmSYmYvYZPH5gFZCuTNlX/XXO8c8M/Trl3ztH17tjZt+kDffvOi5nz4gB55+E7dMuJKdelypo46qplq16mlmrVrqnb92qrXqJ4aNGughoc3VKOjGqnxyY113EXHquV1LdRh3Lnq8feL1fv5qzTgw/4atvRmjVg5QiO/H6lxW+9Scn6S55qG2xnCZqImepadjGCGEnck6c8X/1m1ax+syy9/XOnpQe/wxEoaLmEhAZOailXFp3HjftaYMet0++2rNGzYMg0alKlr+7yknj0eVMcO49Sy5fU66qgL1LhJCzVqdJQaNjxcDRo0Vb16jTzxUqNGbTU96CCddVor9b2ilyaPv0evPf20ln/0kXJWrNA23uC9foN8bHWbkyM/rj673/pu1oI/WgsMk9jBZI+42T1eGb95Gzdq+9q13poZCDoPaQ4Wg/OQhzhzIG4gysxcMnsLeYR8u8BDuLDgHtCQWGakmdVEDDk3ENIRB6LJAx7yyN+FiRkIGETHpcdiwDUsOrxPAnLJjCkBK02fPn08MQbJcYH8IUSUyWdIDWScNkFkXICgUA5lQDKoH2KG2W7IsMuHeMzYMgPMzDKzxlzjIA5kEMLPLLcTM5SPaxkE+dFHH/UIjiMylInbCyQMksx12gTuEDhmtonDdVcvXO7Iq7jvmaFeBOrI+1sgzczYO4LLdy5/CC5El5lv6szss2sfZ8ghLkG0Dzcw8kAEnn/++R5pdxY98KPdrmyHc+SZ7127wIG6gbcrk+8IgwcP9kQ3BBFC676nDIgpQhZBgdAhODHDWC5MzEBcEcSRYgZxjzCjH2gfB2MViwFWBF5YinAgIPBwq8LShjBibBCfetNvuF5xvyFmCrPMgAH9QR4Qd+pO2Bd2xAEH4kPw+QwelAtGEG0mJhBbLk/epYLogrA7bF0+pCP9voKrF+8EQhjhRoUY4t6hDuTD387ygiCmfgSu85vDe5gYP8SlHhxgzG6G3NuIMAQG4497AYsc73ZyZXMmPukpk/SIFqyCTBg4MeP6D7cyLDHOvZLrpHMHZWGVZfw5aw31RWRjAWT7bNziXF0pHysWbUSA8TvFwe8K/f3ggw/umWggH+qJ4OQ+YQyT3rXFAyZK/5mYMTFT5cXM3sJmrteeIJabvCz5c7HgZHluabm7svTLz3OUkzNL3333mj795Gm9+9Z9euqJu5WcNFD9+l2k885rpT8d2VQN6tcJWXH2r6W6DeuoftP6anhEQzU6upEaN2+spic1VbOTm6lZi2Y67KzDdHSXo3V8z+N10pUn6dSrT9GZN52hs4acpfPGd1SHuzqo8QmNVaNGHZ3050vUuVOiOnS4S23a3K6zzxquVq0GqmXLfjrxxCt05JGddehhZ+qQQ1qoWbNT1LTpSWrcuLkOOugYNWx4hOrXb6q6dQ9UrVr1VK9uPR3WrKnObHmaenXvoTHDbtGDkzP0+lNPa9Hbb+vrRYv008ovtWX1au1AsDDLjXjZffDZ7ThlC/ljh7CbeNp3XzBeObDa8O6Z3Ij3zLCImplExAfrB1j8z2w1s4ocEEHOEBaIGgvAefgX9hDmOt8jYCBMzLzygIcAuIc3Z0cIXF6FiRniOmLIs97lATHCygHJcIKBdrB+AjKCGxntgdw4EufqwMw7RAnLyrRp07zZYIiMI3GubgWJGcgT30MUmXGHyOAvzzXawhmyhphByLm6QchYv4MrD2SbWXOIMi5YnHFVIT+sRRBrSCKz3cziQoLAIBw3yDOEraRiBiwQgIg5BAh5uuCwZXacOLgrQQgdHnzv+oJycVPECoEFwokZrvEdFh/wDM/flRN5duWCN2TPiRnKde2mfxiDzIBDKF09SEs8SDRWAgQY1kX+ZoxgjWMM47rFOHR9RB0QbAWJGawTkFEsleRB/rSFM/2C8MfNDmsHga1+IdCIOPrT1ZuyqCfCG+EJiY0UMw4L2sE44p5hXHOfEYqDH/Eok/JcGv5mzEHOETNYOWg/gTFPPzEeId6IB8ojjcPTi1jEf65eCCLuJaxJ/Ja4etBuxi2WDlzdEIy4cRHoJ6wyw4cP96yqrk9cWqwfWOhwOUOAMJZwi2O88xJVRB84uvuaurg8ChIz7p5lYgJRxNoZfitI49JxLkjM0A7GEmvGELH8rvE3wdUTgeQmIciXuLiEItQi73PEOr87rCWj/tTdYellGoX/TMyYmKlWYmZvYRPa7plNBBA3IYGTLX9etgK+bPl9Wdq1K1O//DxbOTnvadWq1/XZZ88rO/NhvfPmNE/kPPj3Ubpn/A26achluuKKjmrfvoVatjxOJ5xwhA49pJHq1q3liZ6adWuq1v61VPuA2qrTsI7qNa6n/RvvrwMOO8A7atUL+SVjPQlZUg5V/frNtP/+TVS3Li5hWFUaqGbNOtq/Xl01bXyQjjvqKO/lgryc8pKuF2rAtX111y236r6kFD311/v12pNPKfOVV72tar9askQ/ffmVtnz/g3asWRuytCBa1q/3yF5hLxs04rxv4mwYxS5GnqiJeM8MM/48mB0ZmDVrljf7CIlwPuWsE8AyAWnHjQTR44hk5HPXPZh5SPPQhnxA6rkOUQkPLi4kgs9FiRm+Jx4z4bikQCghaRBXSBmCgfwhTwgVSBrkEeKCWwezwMySIhKIx8w5i6OZBScPXJdIwzUwccS1MDFDfcLFDOTX1ZH8I8UM32EBoq5gi+seRNkdzDiDL/WhDoiD9PR0T1RiOSM9mBPIn79LK2YgtAgCiBnuOeHBtQFLEy5bWBKcIOE7AvXgM6QOsQO5hKzTL1iqEGSstaGe9FlxAvlxRIoZrrm+wLqBuIM04xrkxg35E4/yEK/0OaIB4owQYQa8JGKG8hAViBna4/B27S5IzODiBTln7QYuj9SH+jkMcEUqiZhhzJZUzDissJqBP26WuPkxmcA9jEh0kwpuXRzCCSHCDl9Y6xITEz3rJnlxFBXc91gwyAMrLGKG67SdA8wg8wh4+s1ZShAqTJhQHvekK8+lZRwg+LEKY0Xh9wSx3KFDB0+E08eMYcpk3Q0CwmFdmJgh70gxQxp3UN+CxIz7fXTudOyERnkEJiGw1oCdcw9FZLk1XeH3N5+5zjhBQCOknWscdYtmMDFjYqbaiplwYcNnT9QgbHYf3rX83cImb578efMU8M3zRI8vb65+3T5bW7d8oJ9+eltff83uMDO1YMETysx8RB9+8IDeeDVdzz59rx7/xzg98tAYTc0YpqTEgRo39nqNHXudRo3so1F3XKshQy7TMcccptq1aqlLhw4affMwjb/tdiWOukMTxt6lSePu1vSkZD04abKevO9/NfPhh/T6U0/p/RdfVOabb2rBrFn6NCtLXy75RD9+scLbhvbX1au92ejAuhxW+HlvR/fvfks6790It7i4WWwj5bFLyq1vStc3jG3Gf36YZYZZZB7WED9cn3BzguDgPsRn1n0ws8puUJAS1lFA4h35iHwAu+sQFLcolhlsAqQhMhAfEsE5Usw4cgKR4TtmcdkJDaJK3TggiM4yQx6kQTTgu866AMgOJALygJXEESPiMavOrDIuN7j1MOsPgUf4II7Ir6Rihnw5ChMzWEPAFtct1kdAOMGWw33GjYZ1MawToj4QUupSkJiB3JfWMkPe4ZYZynDlsCMZ1hFwceXyHYH+oo3MjuMKhOUIS4kTMxA71roQn3jFCa5sxIezzGCh4roTMwgv1n9gccOty40b8nfpEWHMkkN6WctTlJghDYSStRDhbmbhYoaxRjxCUWKGvqRfcb2j7whuLHAuqZgpiWWG+lE31ugg8NnJDEFJ30G0wQMxx5hzliksFcRnRzjcFbEoIeoZn+BLf5JvUf3ncIkUM67t9A/14jcDMYNlxQk98GaiA/y5xwjk58rkGpgiZrCo0Y8c/AYh7nHp4r529xOiiHV4pC9IzNCnfBcuZhgf4X1EfQsSM2BFWiZSGFe4xvG7SdvAFHxxYcQKRn4IRSYrmCwIv7fd/e2uMSnE5EtlBBMzJmbiRsxEiptwgRNgEwEndBA++dnepgKIm4BvvgJ5C7xzELGTj+DBwjNXu3bO0Y7fZuvX7R9q08Z3tXbNW1q9+g2t/uE/+u7b1/Ttd2/q82XPqtN5Z6hhgwaadM+9+n7ZZ/px+Rda+8UK5Xz5pTZ89ZW2fLNKv3z3vXb8+KP3AkD/2tDLAIO7BUqAncXCjw0bvJcFOhLsBIs7u+t2Lh1JNtyqBm5/EDMHH6y5WVnerCckgbUAEGhEDaSQmXBIOQ943DwIPKx5sBcW+I4DgcRMJkQRaw5EgcOldySCuFznHClmuIYowtICMYNwQVqYaYdsczBDy5oS4hDfHZAErBusoUGYQPhx7YD8sLgfgkK5WGggHyzShujccMMNnuhhVyJEG6QK8hG+ZsaRm4Isp5HmowAAIABJREFUM65dkWIGvBBlkH9clPieOnKAFQefwZn8mbFFxCAocImhXY6QObwgeaVxM4OE4RJF3lgy+JvgsKMNWOiwIjGTzEw/37lyXT2wvBEHkclstBMzWALAlDTkVZzgyi5IzFA/vsf6Q3shk4gl6uPw5kw8+goSyXbNiNpIMUN/u3TkieCNdDMrjZhhLDK+EA+03dWHsqgX63mKssy49rs1M4iocJJfGIYuHYKBiQdcA7GEYoVkjQr1Yp0RbqLOMuP6kTpi/aFu3P9YnBD/TGZwn5E3cQoLfE+IFDOkce0GS9aU4HKF+yprsQhYHSmHMYz4c2k4gxf3FsIMMcM21wgF8iQ/7mvc4hDifIfwY4MBNgYgFFfMhLuZkTdHQWKG+lAvymaTCe5fftsQ1PxOYEXDnczVnetYefmt4T4u6D7nGvk5DL2KR/E/EzMmZuJazBQkcNw1J24QOnuJnd0bDniua7tFjydw8ucrkL9gz+HPm6+Af7G2bX5PF13YRo0aNtSj06fLjyjBDzdcnPBwRaDsFilYViDUTpx4awP4gdz9JnV33Uh31SDd1k8V008Fipm5cz2iABmFLEIUINQ8mMMP99B156Keu8QhLesIIBpujQPXXXr3mYc65J34kWKGMpgNxbKDSwpEFQsS5ILALDNEIlzMuDqHlwNJgZCzMxjuNpA8N4saHg9BwY5hzGrjLoTlBjHFjGp5iBnaiTBhYTltIe/I4Agb9aOeEG0sUJA72kZ9ORPYtQuCjOgBC4dpZJ7hf7s4bJLg6kEfuOsufywJuAUxW48LlbvuztSHheUIB3Z4QiS4NTPlKWaoO5jQZsi9czNiMTt/h9cbcsqaLkgyWy4zThAzbs0M7nvg5MgpedMfWHvKapnBmoYQQKy6bY1df1Eem1G4DQAYV+476uDaALllXDABgHWQdHxXVOB7xhVCBcsj23gjThg/BM5YMhAFCBxnmXGYhudPXNancD9hRXHfuXNkPdx1t2aG9jvhS/soA0HJZAblYwVinBCYYGC3LyytWNzIy2ECDghCBAEWHSYauEZ+rkxXF8QCfY7AYDKGgJghX4Svc2tzwoEJCywpbgMAynQH+RcmZlzZuJPSx1gNaTf3JpZJLE7kQ59RJ35rEHD0J/0TGcgvfBxGfl/Rf5uYMTFjYqaUY4Ad1JzgceeQ8MlUIDdT+Tszxa5qG3PeVNeE1jrwgAM0IyPDc4kJFyd8duIk/GwEuGIIsOFafXD9g5hp3Njz+XZiwpEJCIMjF+HXivuAdendlsqQS2aHcauATEEwESXM/uPv7lxPEDOQkCuuuMKbeaU8RBYEjVlOrEUIAEQHM8oQWlzMIHFseUpglhcLAX9DbikLiwj5MKuKmMFFBKsT6fFzJy+IHAQdqw+zvBAzZpGdZQZS5dyewIs2Eh/hw+5Zbs2MI0aQMWbKcT2CuBEgL7jvsD4BosaaE3CgPdQTEkqdwAHyRVxm+tlRC8ECnsSFEGKJQsix/qYkYgYSRYDYY5mBBJOf6zPX75BQsABfrDgQTnACD/rLLf5nByu3Pob+ZI0NVpGyWmYg9OBNcJhSNxarI7BwpcKqhGsRY4L+BQesI7i5USfi0/cQdNyssAzQZ2CAgIMkU1cWpyNqGJsEsGfNDEKR9pAPwYkLt3aCNUsIE75nrQjumKwdwQUK8UKfMjGAOxebUrBDIP1emJghPoKDthOfQNsLC5TLwZjgxZEIS3ZzY5zQPnBBZDGGIi0z9CfCnbrR/9STscgmAbhb8u4plz/ngoK7jkWHNGCPkHP3E+uaELqICtaJcP+7ewfRwBhknRj3iXvhJv1DPNalOKsu/eL6kXqH39cIFyyD5IOlkOAsM4zvSDGD5Yu6IMSx0CK+mMRh/IA19aIfKdvtZuZEB+2ln1ncj+UMXPmtIH74byj9Ae6IHiZz+G1A4BDH3eeMSdpM3MoIJmZKSWTd7L2ds0wMFTKGEDYKLtDmDW/rwq7neGLm0SlTvB2YINQeEavEd3gYqa8+pD5e+7IwMQN5CyeMjqSU5SFLHuQLMYEYMXOKLz4zluziw2wqxI41BpB2ApYcriFmcA8jQB7wS8dCwXfMMkOQcA8iLdYFZ5khPuIEkklcxAuuKhA9rDsQLlyUIDAQFQgJu16xGJk4+O9D3FnUztbNEEInZrDMkB/tcoTMiRlc3SC/fOdwxI0MywliBuFBcKQIQu5INySasqgnC+kRBxBa4iIomEmGfFF3/PIREeAHGUNoODETLki8wgr5j3wJzjUI0hi+doDvXDsgjexUBlEET3CibFxrmN1GyGE9op6kQUSCKQKB9rt8CqnKXpeJywHebs0MLkgEMHXfMx6wooArooXPvOQQQoy4Y8E+C8eddRGyiHhlnLDJA/hhqYG0Q4JZ28AMeriYoX8RzlzHtZGyCWDHZ2bkGc9uNzM3yw45xZqFoGEM06f0Le1AFGGdADNHnF2+bswgzBiD9DXrzfjeiU+vAgX85+LgTkb73foNyoVkI+DIDxFPnowT2sH9w1hnogDXUqwaYMBkAmIbiwJ5U7fCgqu/2xCBNVhYYbhHcf8CX8rm/mRMI65cnpwRfNwjlAdejC/azz3J2MY6RBzqwL3G/Y7Vz93XuKrhUoflF0sjAoJ8EdK4pSJmsOxyzfUd7mlY98CJMYOFhnoyJsAaQccYJy0TLQR3z7hxCFbck4hdxnq4ZYn4lIdo4R7jt4R7h/uIsefucyw6TBbwG1MZwcRMIUTURIqJlLKOARMzJhbiVWREq91FiRkewOUZyI8DMsk7GiBKEAwIJQQSQoErDQ98ZkMJCBje3YG/PrOrpIdYsj4C0gNpIS2EEKIDEWGWFJEEkSQ+bnLMnHONeBAlyoToYW1hlhSRxcw9JBNyRL6QDhYUQ+YgHMyeE7DYILKwBECyKcMRI8gZAgNhQfkER0wRAlgKIIsQMgJkifQQGMgTxI+ZW+pI/agLpIf1Dy4fZnSZcaZerh20HWIEppBArDmQJ/LmKCo4ckwZ5Af5RnC68lxa8iEuli6IMaSN9TOu71hbxMw+QoZ4pCdP8MDFi/aTB9eLE1zdESHMnrvd50jr2uTiIL7AtlevXl7f0Q5EKsQdUh9Omikf8QphBWMOBCPjA9KKmGNcQFzpawL9S3/SZoSxK99hhziFcDMe3HoPyiE9VjVEKTP21It+w8qCxQLXRQQQFj/iu3z5zIEQR/gzDt3LF12ZhWEYjgkiH/FFuZTPeIKwM3EAecYSyDghT8Q3Vhzqh7CmX+lfxhaiB/Hg8i6qbL5zwhi3MSYrGJOUj2UCwelEFOW6PDnTZu55+oB60G7S8DuBYMZy5uJjmQFv7hl3X3PPcv+AN2OR+4o8GR9Y9fht4LfFlcWZsYFgYm0R9x14sQEC6akf7catlLRgRHB9QN7kgUWHcQGmYO7K4DuCK4+xjGBhLDPuKM/1C79xTOK4vL2EUfzPxIyJGbOsVNAYMDFjYiZapD5ey4mmmHEPdfeAh2hAXFh/wawvC44hmU4YEB+ixeJZZlMhAi4PHvgQBmbL2eEIMonYQJTg8oQVAJISHp+8cSVjYS7CAasBJJL6uIM64dqFwCFfRAEEA+uAiwOBgbDjisLMuSuDM/XClQbhRZrwgIijPGaJ3XcuTxcP0sU6E/CgbcSFEBOP4OKTF0QX1xTqSHsRedSH+iN4ikuKXJ7gBbmGEFI26V25rn7ub3DGmoDIgOjRd65Mlx9pXP/RJ9TZtcHlV5wzeJM37YrE2+VHmdSXOkE4EQpYX5woDq8Tn8kT1yf6mbGAexr5gzVlgTtlEc+VQZ/hgsV4dIG8CBBe+gDhTD8QXJmMZwQOroX0F9t8u7Hs8gMbl5dLS3voB8guwhkLlMvTK2Af/xGXeiHSKZfF8NxvlI0QYPyCQXg/477FuAIPxj9CkDoWt+9cG5yYQWhw34EN45R7j4kMF8+dI5vCWKQelI+lDwsL14gfmSb8vmYs4toYeV9Tf8Ygfez6x+HMmX7j94A6ghd5ukD/8TdpwZPg6uDO3A+4r4GV6yeX3p1dXM7UD7dX6st45ffCjQkXP9pnEzMVRGTLOqtv6au+ZcjEjImZeBUZ0Wp3tMWMe0CHP9j57A6+d9+5uAWdXRyXrqCzS+e+c3m7v905nLAWFsfFdXm6eOF/u88FxXXfFZbOpSnqHJlHceOGpyvssyO0vDgRFyBcgiBurozwdOHX3GfOBLB0n10aF8f9XdpzZL7h+bgyIs/Ecdci44d/5+KEn8Pju8/ue/d3+JnvIkN4fPfZxXN/F5YG8k4/sP4LNy3iu7Eamaawv10Z4WcXl2vhITxO5GfiRcYPT+s+uzgFbQAQmaeL69K6c0Hx3LXI9rvrkefwvNxnzsSLDJFpC4pTWNrwvFw+4dciPxPH3SMufuQ5Mk20/jYxY2LGLDMVNAZMzJiYiRapj9dyKkvM8IDmIQ6JDj+4Fhl4+DsCEP5dZHpHdIobn3JdGpcveXItvE58jqyXixeZnnwKK5/vXLrI/Fz5kWXzd2Fxw+vo6rGv/F05kWdXDjPEuCLhBsMMdUEYubSRdS2onq4+Ln+XtqRnl09BZbi8XJxwXPYVPzyuq6PLJzItfxeGR2FpXN0Kw4rrHOHBXcMawboPFpU7awbllyS4Ort2kjfXCqtvZHzXXq4XJ7h4TszgsoilgnywXrj89pWXq5+rN2eXd2RarofHK6wMh2tkev6OzMPh5OK6+hRVB9IUVU+XF2cX19U7srzwuNH6bGKmgoisWVaqvmWlrH1oYsbETLyKjGi1uzLFTLQe0lZO8RBwhIr1DKyXYAOEwlx7ipejxSopAo404+4G/qx1wVrmrpc0v2jHd2Sf9UZseoAYQ8wQGF8WYhcBEzMmZswyU0FjwMSMiZlokfp4LcfETOySi2jXDLLJgYBhkTiLqllDUNzZ5mjXtzqWB/7gzVokdoGjH1jDg0hwQiGW2+3qyOJ5dhpj4wd2SyOYmInlnpNMzFQQkS3rrL6lr/qWHRMzJmbiVWREq90mZmKbYES7dhBOCCmbGLA7GYQ0MzPTI6JGRiu2Nxz27HTH7l/syOV28HMioWJrUH650xbcylg8b6FqIGBixsSMWWYqaAyYmDExEy1SH6/lmJipGkQjmrWEOENC2UqZ3d8coY5mHeK5LIQkgobdseiHqiZkXN9R76pad9eGeDqbmKkgIhtuWeHt8LwR3r+LY87ug7fEzzUhEQX8w/simp9NzJiYiVeREa12m5iJJ7pS/LZCQnF3YktbZtgtRA8B8Ab3qixkooeWlVReCJiYqWAyjYhRfraCgcUKBpYqGFi2+/hU8i9U0JclxE40SbaVFR0XNhMzJmaiRerjtRwTM+VFBapfPgga5/pU/VoXuy0y3GO3b6pzzUzMVKCYQaQE/Uvk25mpz5Y+pRdmJurBB27Vo4+M0ltvZmj19y8pP2++gvkLTdBUYD9UlngzMWNiJl5FRrTabWKmOtOTsrfN3ITKjmFpcjDcS4OapSkLAiZmKoJEI2KwuPg/VuacGbpj5NW6pGcPJXTpq4SEIUroMlhdE65Sn6t76n/vG6EvV/5bQaw0eVkSR0XUyfKMOq4mZkzMRIvUx2s5JmbK8vi3tIaAIWAIVA8ETMxUAMkP5GUpd9c8vTBzgi66sLM6dbpZl1/+hG64YZZuummBBt+Yrb59/6Pu3aepc6c+unnoNcqeO0NB/wKz0FRAf1SWODQxY2ImXkVGtNptYqZ6EBFrhSFgCBgCZUHAxEw5k2eEzM7fszTrvfvVuvXp6tw5SSNGLFNycq7S0gKaNCl0TJwY1PjxW9Sv3xvq0GGgBg64TMs/+z9zOSvn/qgsIUO5JmZMzESL1MdrOSZmyvL4t7SGgCFgCFQPBCpVzHhrSrydvtjtK3RU5cXwuIn5c7P1/Xev69KeHdSq1VAlJm5RWppfKSlBJScHwo6gJkwIauLEXA0c+KHatr1G48Ze7wmhgC875HJWjYh9ZYqKyirbxIyJmXgVGdFqt4mZ6kFErBWGgCFgCJQFgUoRM07EBH3ZCvjmK+BbuPtY4P1dFXf4ok2BvGz9tj1T/545QQcf3EKjR6/2hExyckjMpKRI4QfiJjU1oKSkX9Wr17Nq26a9li5+Qgos9rZxriwSbuWWz7olEzMmZqJF6uO1HBMzZXn8W1pDwBAwBKoHAlEXM56QwfKQP0+7dszR+py39fVXL2vlihf13beva9uW94VlIuCbJ1WhLYtplz9vntaueU/Db75CJ588TNOnS0lJ+XsJmHAxw2cEzcSJfg0f/olat75O9//vUCm4zHsXjYmK8hEVlYWjiRkTM/EqMqLVbhMz1YOIWCsMAUPAECgLAlEVM856seO3TH3z9Uv6z2tTNSX9Vt02op+G3nStxo4ZqEcfvkvzs2Zofc4sYbmpKoLGa5uPtw2/ooQubdS9+7OaNg3XsuKJmTF3rlaXhGQNuekiKfiJ95LNyiLhVm75iCgTMyZmokXq47UcEzNlefxbWkPAEDAEqgcCURMzISGTpe2/ZGrWew9o0MBeatmig0466Wq1bj1O7dun6vTTR6h5825qc05bpU8arh9/fENB37wqs8NXMD9bK1fM1Dmtz9CVV76tKVOwvOBitrd7WfjfycmsnQlo3Lj16tZtmq69toMU/LjKtNmET+HCx8SMiZl4FRnRareJmepBRKwVhoAhYAiUBYGoiRmIXe6uBfr3v9LUts05+stfhmrw4EXeAvipU+VZMTIygrr77m26/PLndMQRbXVd3y7avGmW8nNjf0E8i/9xjfv665fV5YK2uvji5zV1agksM2NWK6FLsoYMvlAKLgnthGUbAET93TDlKc5MzJiYiRapj9dyTMyU5fFvaQ0BQ8AQqB4IREXM+HdlKhhcrAXZM3TmmS2UkJCuxMT1Sk/HYhFUUpLfW1uCFYMdvjIy/Bo9+ls1anSGbh3eQ9u3zxNbHss3N2bJrbdmJneefvrpXQ258RKd2uI2TZ/uxEywUOuMWzNzyy2f6pxz+mn61MFS8DPl75wTs20tT8JfnfMyMWNiJl5FRrTabWKmehARa4UhYAgYAmVBoMLFDDuTBXxZys9bqCsub6127e/QnXd+q4kT8z0Rg5tVpNsV60wmTfJrxIgvVaNGHc398G/K2xUSNIiGWCTAITe6bP32a6ZeeD5JDRocraSk3zVhgt9b5B/eRvc51PaAUlLy1KfPSzrzzNO17JPHPcuMPzczJtsZi9jHap1MzJiYiRapj9dyTMyU5fFvaQ0BQ8AQqB4IVLiYwSrDgvZF8x/RKSefof79ZyslxaeUFEj83kImnOQjaNLTd6h580EaNuRibd40x3uHS6yKGQi152qWl62ffnpLnc9rqRP+PFRTp+70rE1Yn/Z+z0zo7/T0oIYMmaeWLXvp9tuvkD9/kQIxKthiVTTEar3iQcxAJgPr1smd+VzcwyOiOSZ44lWIlEe7TcxUDyJirTAEDAFDoCwIVLiY8dylgsv1t78O1RlnXKuRt68o0loREjSIHL8mT/bp8stf09lnnayfVr8e2t0shteRhKwzc5W7M0vzs/+hww9totZn36l7792qKVOCmjxZSksLHRkZUnp6QAMGZOnUU67UVb066YfvXlEwf4Gtl4nhPi6JcKr2Ymb9emnjRmnz5tIdmzZJ5GGCxjAo5RgwMVOWx7+lNQQMAUOgeiBQ4WImZJn5TKNGXqoOHcdq7NjVxRAzrKUJKC0toME3LteJJ5ykr1Y8p0B+tnBbKwmhjHbckHUmSzt3ZumDWX/TeR3P0XHHdVdCwt80aNAi3X77jxox4ltde+27at36Tp1ycmcNv/lqfbbsWfl9uNLFphtdtHGsDuVVdzHjX7tWv65apQ3Ll2v955+X6CDNL19/rcDatSZoSknkTQTmeBZBBHH+mjU6uXlzNW7cWJmZmfL5fAoGg9XjKW2tMAQMAUPAECgSgQoXM/k7cTP7XHcgZjrcqbFjfyiRmLnxxuU64fgT9fXK58TWx7EuZiDhbp1Q7q65WvrpM5qSPkJX9uqpNm266i9n9NCZf+muTuddqMGDeuuZp5P0/Xcvy5cXSlcdSLy1ISS4q6uY8VzKNm7Umi++0F+Tk3XRuefqmq5ddeUFF+zzuKpLF111wQW6tFMn3TV8uDZ+8YVn1TGXM3O3K404M8tMkc93+9IQMAQMgbhAIApiZo4nZnAzO/30PrvdzAKKXPjv1sv8wc3ssld0ztmn6McfQm5mVUHMeIImb25IeOVna33Ou1r88TN658379MqLk/T6y+ma/cEDWvHFv/Tztg+kKiLSTKSUzCpYXcWMtyZm0yatWLJEI/v3V+8TT9Sr/fvrxauv1st9+uz7uOYa3dGunXp37arvlyyRtm4NWWjMQmHuZiUcAyZm4oKnWCMNAUPAECgSgQoXMyE3syVaOO9htWx5tvrfMMvbvQs3ssIFDWtm8jV58k4dd3x/3TL0Em3e9FHMbwDwB7K/W9AEfPMV8C3wtpf25WYqP3eu904arnsvBcV1ztt6umRk+Q/lxbgLXrzVt1qLmc2btXzxYo3p31/3tG2rvIwM5SYmKi8pqegjOdn7/v969VLfiy7St4sXm5gpIYEvjQWjuqYxMVPk892+NAQMAUMgLhCocDETeplktn79ZY6u75ugc9uP0qhR3ygtjXewIGhCogZhEzrY5cyvtLRdGjhwjg4++Gi99/Z92rUj2xMDsbybWWFknTp7By5oWGE4cCvbfb2wdHa9aou7eBAzdw4YoOT27RWcNk3+lBQFU1OLPiZM8L7/11VX6bpu3UzMmJApkzXKxExc8BRrpCFgCBgCRSJQ4WIGQh7IzVTQ/7FmvXOfOnXsrK5d03THHT94C/wnTgx4Wxenpga9M4v+U1N3aMiQuTr55Cs0bOgl2rb1Q/k98l+1yS1YmICp+n1YXJEZL2ImqV07BadMUX5SkoIImqKO1FQFUlI088ordZ1ZZspE5KurtaUk7TIxU+Tz3b40BAwBQyAuEIiKmIHAB/KylLcrW/+YMVoXdu2pLl0S1b//+xo9+kfdc89vSkzcpbvv3qZbb/1CvXs/q7Zt+qnPNQla9fVLys+b76UvLom0ePEjGGK5r+NKzEyd6omZvd6Au/dCOE/kCMuNiRkTMeVkkTIxExc8xRppCBgChkCRCERFzEA4PYtE/jzt/H2OZj6frGv7XKrzO/dVt24puuyyx9Sr1zO65JK/K6HLKCV0uVS33XqlvlzxLwX8i+y9KzG4FibSTY6NGbzDXOf2bB1uYiaFPdb3HIgYEzO2a1lJLC/7imtipsjnu31pCBgChkBcIBA1MRMuaBRYoC9XztSMR+7QzUMu05W9uurySy/UtddcqHF39dEbr6drx2+zPdc0XNRiefY9XusWzGMN07w9Gxj4fbwfh2tY0eZ5fYaAjVd8aLeJmf8KGUSNiRkTMvsSJyX93sRMXPAUa6QhYAgYAkUiEFUxA8ELEdy5kn+RgoElytuVpZ+3vKMtG97U77+8LwUWKhj4RPLPM4tMLFpkPAtMtnbtmKv1Oe9p5Yp/6+OFjyk78yEtW/K0Vn3zmrZt/WjPdtPlsZV2yApU9VznTMyYmCkpObf4JRN8JmaKfL7bl4aAIWAIxAUCURczbqYei4u3MUB+thRcIAUXSoH5nthhO+d4n9V3OMXSOeRKlq1tWz/QnI9mKDlxsLp166TTTjtLJ590plqf3UZ9rummRx4eqx9+eFV5ubt3oCulKKM81loFckPn8hBG0cTTxIyJGRMnJRMnJcXLxExc8BRrpCFgCBgCRSJQaWImnFR662lsrUVMu2R5feSbpy0b39P0aSN0eqszdfzxV+rCCx9T//4LddOQz9W37/tq336iDjnkHJ3bvqXmZ/1Du3a6LbWLb1kJ5M2VHwHjnyefb4Hy8hYq37fA29IagcBRFcSuiRkTMyUl5xa/ZOLHxEyRz3f70hAwBAyBuEAgJsRMuLCxz8Un/dHCCuEQ8GVr529zddfYPjr++Hbq0eNppaXt0LRpUnq6NHmylJEhTZ0aVGLidp122u1q0uRAT9Dk7txtWdnHGhrK8SNWAh8rL2+xNqx9XQuzH9T770zRkoUztGXjW/LlLVIgf9GezQaihUFpyjEx80cxw7oZ25q5ZITdBE7heJmYiQueYo00BAwBQ6BIBEzMlNIFqjTktqqm8dwBA59qxsMj1eq0jrryypmaNMnvvdw0KSlfycn+sBeg+sU7g6ZPz9Mpp45Qx3Nb6rtVryjgY7OAwoVayKUsW3m5i7Rw3gz1uaadjjn6cDVpcqIaN26lpk2b6+STj9Ytw3to8aLHFMhfoKB/XkxbaOJBzIwZMECJbdsqPyNDu+6919uemffNFHokJ8uXlKTnevVSX3vPjG3RXMYtmk3MFPl8ty8NAUPAEIgLBEzMFEGwq6r4KM96ey5d/oXKWfO6undrp4u6TdPdd2/1XnCanBwI33l39+egJ2xSU/0aP36rmjY9XY89OlY/b5sd2v2sAOuMs/xs2fyBMtKH6LDD/qTWrUeqf/85Gj16ncaN26xRo1arb9+3dNppg3R6q7/o7/ffpu2/sONdaJ1Veba5vPKKBzEzbvBgTT7/fOmhh7wXZ3qmOsx1RRzBqVP10vXXq3+PHvp28WJp61YF1q41Yl9GYh+PFhwTM3HBU6yRhoAhYAgUiYCJGRMzRa7V8TZjCCzVyy+mql3bS3TDDe9q4kQsMcEChIx7pUhQWGymTQuqdetE9buuh7755jX58/4oPDwhkzdPGzfM0vSpw3Xssafo+n7/UXLyRqWm5npWntRUXk+CcNqp8eNz1LPHIzr7rARNmzJUO3/L9FzgyKe8REh55VPtxczHH+vGPn3UumlTpZ1/vpLbtlVqu3ZFH+3bK6VdO/U64QR179xZ3338sYkZEzGlFrImZop8vtuXhoAhYAjEBQImZkw79vMfAAANkUlEQVTMFCkC8nfOUTC4TGkT+6tTp1t0yy3LlJYWcisLex/iH4QNrmdTpgR11VVvq3279vpkyTN73knjxAICJMhanN8z9dYb96lFi7N01VX/Ulra70pJ8SklJdzyg3gKCIvPvfduUbdu9ymhS3e9+vJkBf2LvZ3xXL6xcq6uYgYCGdy4UetXrtQz9/9NQ/r21agbbtCt11+v2/Z19OunW/v10/Drr9dfU1O1deVKadMmeaTUSH2pSX0sWGW8cbFunWdlw9JWrIP4u9OUZgx4adavV/6aNTq5eXM1btxYmZmZ8vl8CgaDcfEQt0YaAoaAIRDvCJiYMTFTpJjxLDPBTzX2zt5KSBivkSO/UVpawHMlK1rMBJQ+WRo0aKlOb9VGC7NnhBbth+EN2Q/4FmjV16/othF9dcopAzVhwnZPsIRc2CKtPyEXNsq/7bZvdF7HkRp+cy9t//kDBfNjb/1MdRUzjjj71q7Vxm++0VfLlmnl0qVaUYLjy2XLtO7LL5Vv7mVVWsC4seCdN2xQcPNmBbdsKdWhDRtKjIWJmXinMNZ+Q8AQMAQkEzNh5DpWZvRjqR4hMbNMSYnX6YLzR+m221YUS8wkJfmVkRHUdddlqvXZbbV40ROeFSZ8EwBehsnGAHNmP6zO53XR5Ze/pClTAiJt4UIJgcORp0sve1IXX9xTSz5+XMHAYlHXWMKuuosZCGxgwwb5N26Uf9Omkh0bN3pp9yLDZpkpMZmPBfw8i8zmzVq1YIEm3XOPxt9xh+4eNarYxz2jRytxzBgt/fBD+davV6AE48DEjNEYQ8AQMAQMARMzJmaKFADOzezpJ8eqfburNXhwptIm7tsyw5qZ6dODOq/T/brisq5aueJFBSLWzATzs+XblamX/z1ZrU67SCNGfKdJk0LWl8LFjDyr0KRJAfX7n3fVJeE6vfhCioKBpSZmSkACy4sEQyaL5U5UgNuRR0Qroc7l1XbLJ7Rlsucm9ssveuv553Va8+a6vVNnJXbtqsSEhGId4y68UCceeqieeeAB7UAY5+QU2+3QxIyRGEPAEDAEDAETMyZmihQz3rbM/o+14vN/6oLzO+mSS2coMfEXb0F+YZsA4CI2YUJAKam/6uijuykj7WZtWP/eXruZeetl/PO0a8dHeu6fKWrR4hLdeedWTZyIWIl0L3MbC4TO5M/W0AMHzVFCwkA99cRYBQOfmZgxYVAlLRtVXRT5162Tf/t2vfDEEzrtkEM0e/BgrRo5Uqtuv71Yx6e33aZTGzbUg1Om6HcsfCZmjJkYAoaAIWAIlAABEzMmZooUM7htBXxZyt2RraR7+6vDuX01YMD7mjiRncaw0IQsKQiM0BH0hMzkyX517fqIzjrzdM3Peky5O1nsn7XnXTOemMnPVt6uTL30Qppnmbn11tVKSyu+ZeZ//uc9JXS5Ti/MTFIwsMzEjIkZEzOVMAb8a9fK/8sv+veTT6rdn/6kb++8U4HJkxVMSyvyCKSlKTBpkrampqpt06Z6aNo0EzMleHhbVEPAEDAEDIEQAiZmTMzsW8zkZiqYv0jLP3tegwZcqvPOu0kDB32oexN/1sSJAW8NDWeOCRPyNW5cjnr3fkHHH3e2HvzbSG3b8qECedmSb+/tkxE0gbx5mv3Bg+rY4QL17v3G7jUz+UWsmcE6w65mPl1xxT/VvVtPLZo/Q8HAEhMzlUBkq7pVweofchUrCw7hYuacww/XqlGj5E9JUSA5eZ9HfkqKtiYmqvXBB+uhqVNNzBgzMQQMAUPAECgxAiZmTMzsU8x41hm2Uc5fpLlzHtaAAZerQ8frdEWvx3TTTdm67bYvNHLkVxoxYpkGDHhP3bun6/RWHZSSdIN+3vK+AnlZQrhELs73xIxvgb768kUNG3q1Tj99mNLSdiglpfCtn7H+sJvZqFE/qHPnsbrpxsu0bfN7MfnyzHjYAKAsJNjSll1IxAKG4WKmzeGH67tRoxTk5VBFLXzb/V0gJUXbEhN1jomZEj+8LYEhYAgYAoZACAETMyZm/iAyIkWH9zc7jyFo/Av1xfKZSk0ZpISEzmrTppc6dhyuTp3uUMcOg9W2bTdd0rOD/vn0PfLtyt69ZXJWgWV4rma+bP22/SO9/OJknXrq2bq271tKS8OFLd97YWZo57LQGprUVFzY/EpK3qqelzys8zv30Mznkuw9M2aRMfeyShwDJmaMThgChoAhYAhUJgImZkzMFCg0ChQ0rHnZbaHJy52vr798Xs89e7eSE/tqzOhemjrlRr391hRtxVIS+FTyZRdokQnP27PO5M3T2jVve2tyjj22jQYPnq2kpM1KSUHQ4FLGEdqO+Z571uuqq57VWWf1VNK9A/XLtg9DL+MswPITXk5lfDbLTPWwPMSC9SOW62BipjIf4Va2IWAIGAKGgIkZEzMlEjOIAm+HM3ALfCwFl+0+lioY/EwKLpWCC+TfNUfKK9giEyksQu+bydb69e9ozOg++tOfTlJCQoaGDVuqu+5aq/Hj12vs2LW68cb5OvfccWrZorVSkwdofc7bnqUIQRSZZyz8bWLGxEwsi5DyqpuJGSMShoAhYAgYApWJgIkZEzOlFgKIGkQLL6vk4J00/F0qccFmAL5s7dwxX6++kqYLzj9FRxxxvI48sqOOO7abjjiijY499jj1vvJcvf3GFO34nU0JFiiYG1svygwXUSZmTMyUl2CI5XxMzFTmI9zKNgQMAUPAEDAxY2Km1GImnLiXy2fW5eRmSVoq6WN9ufwZvf5qimb+31i99foE/bDqX8rLXShpiXjhZjA3Ni0yDgsTMyZmYlmElFfdTMwYkTAEDAFDwBCoTARMzJiYiR0xs7sv/LlzPeuO379QPt9i+fKWeOeAf4H3zhtvI4Iq0G8mZkzMlJdgiOV8TMxU5iPcyjYEDAFDwBAwMVMFSLGb6Y+3M6IFYRM62Hzgvy/drApYmJgxMRPLIqS86hYuZrz3zIwc6b1nxp+crH0d+cnJ2nLPPfaeGeMihoAhYAgYAqVGwMSMiZmYs8yECxVv9zR2UCvmZgLhaSv7s4kZEzPlJRhiOZ9wMdPhyCP13bhxCmZkKJieXqxjW1qa2jdrpoemTbOXZpb6UW4JDQFDwBCIXwRMzJiYiWkxU9mCpCzlm5gxMRPLIqS86ubEzAuPP64TmjTRi9dco/kDB2r+gAH7POYNGKD3buivY+rV0wMZGSZm4peLWMsNAUPAECg1AiZmTMyYmKmgMWBixsRMeQmGWM4nsG6dAj//rI9efkXtzzxTfTp0UP9OnXRDMY9+nTrrjD//WS8//rh2bdwof06OguvWFetFqF689euVv2aNTm7eXI0bN1ZmZqZ8Pp+CwWCpH4yW0BAwBAwBQ6DqIGBipoKIbFlm9C1t8d5PE+s4mZgxMRPLIqS86hZEfGzcqE1frNC7L76o/8yc6R2vz5ypfR0u7hszZ2rt0qXyr18v8itu3UzMVB2yYTU1BAwBQ6CiEDAxY2LGLDMVNAZMzBSflBaXvFq82MU0iBDZuLH0x/r1xRYxbhyYmKkoamD5GgKGgCFQdRAwMVNBRDbWrQZWv4q3/piYiV3i7ciwncuvjzwLzW4XMURGaY6S9oeJmapDNqymhoAhYAhUFAImZkzMmGWmgsaAiZnyI8olJbkWPz6wNzFTUdTA8jUEDAFDoOogsLeYGXqZ8nfNUTCfd3rE9tvVzbJQ8ZYFw7hsGCNmgoEF2rzhLV3Y9RwdeMABenTKFAXWrvXWBXhErATrA4ygxwdBt34uXj/vsfzk5NgGAFWHc1hNDQFDwBAodwT2FjM37y1mgr7QSwrtbDjYGCj5GPivmHm7UDGzxzUH9xw7DAMbAyUbA7iz2W5m5U4MLENDwBAwBKoSAnvETKNGDTTMs8xkKpCfLd6+HvBl2WEY2Bgo7RjInauAH8vM3mKG93IEcnLkbWnL2Q7DwMZA6cYA20KbmKlKnMPqaggYAoZAuSOw38aNG3XnnWPUsOH+6nPN+Vr26VNavuyfWv7pM1q+1A7DwMZAacfA5588o+WfP6esOQ+rfbvTdED9+koeNUqff/ihln/0kZbPnm2HYWBjoKxj4KOPtOyDD3TcUUep8cEHK3POHPny8uw9M+VOFyxDQ8AQMARiEwFPzIwZM0Y1a9bU/vXq6NBDG+nQQ3YffLbDMLAxULoxwH10aCM1bdJQderU1n777aeGDRro0CZN7DAMbAyU4xg4rGlT1ahRQ40aNdKcOXOUZ2ImNhmH1coQMAQMgQpAYC8xA9nigcDZDsPAxoCNARsDNgaqwhhwzy0TMxXAEixLQ8AQMARiHIH9Yrx+Vj1DwBAwBAwBQ8AQMAQMAUPAEDAECkTAxEyBsNhFQ8AQMAQMAUPAEDAEDAFDwBCIdQRMzMR6D1n9DAFDwBAwBAwBQ8AQMAQMAUOgQARMzBQIi100BAwBQ8AQMAQMAUPAEDAEDIFYR8DETKz3kNXPEDAEDAFDwBAwBAwBQ8AQMAQKRMDETIGw2EVDwBAwBAwBQ8AQMAQMAUPAEIh1BEzMxHoPWf0MAUPAEDAEDAFDwBAwBAwBQ6BABEzMFAiLXTQEDAFDwBAwBAwBQ8AQMAQMgVhHwMRMrPeQ1c8QMAQMAUPAEDAEDAFDwBAwBApE4P8BjYWe7MFzlVcAAAAASUVORK5CYII="}}},{"cell_type":"markdown","source":"# 2. Đề xuất hướng làm\n- <b> Tiền xử lý dữ liệu </b> (Bỏ dấu, chữ số, từ dừng, rút gọn, tách từ) <br>\n  Unicode: Chuyển về dạng unicode <br>\n  Lowercase : Chuyển về dạng in thường. <br>\n  Punctuation, Remove Number: Bỏ dấu, chữ số. <br>\n  Tokenize, Stopwords: Tách từ, Các từ dừng. <br>\n  Lemmatizers: Rút gọn từ về dạng ngắn gọn.<br>\n\n=> Loại bỏ các từ không ý nghĩa, giúp ta xác định được các từ cần thiết để phân loại từ đó thuộc Label 1 hay 0. \n\n- <b> Trích xuất dữ liệu </b>   \nSử dụng `CountVectorizer` để trích xuất các từ, biến words thành dạng vectors ở dạng `Bag-of-Words` bằng cách đếm số lần xuất hiện của các từ trong bộ dữ liệu.\n\n- <b> Huấn luyện mô hình </b>  \nÁp dụng mô hình học máy `Logistic Regression`, `LinearSVC` để huấn luyện, dự đoán.\nSử dụng `OverSampling` thử xem kết quả như thế nào với mô hình trên.\n\n","metadata":{}},{"cell_type":"markdown","source":"## 2.1. Tiền xử lý\n\nCode ở đây em tải về bộ từ điển định nghĩa sẵn trong NLTK  ","metadata":{}},{"cell_type":"code","source":"import re\nimport nltk\nimport string\nfrom unidecode import unidecode\nfrom nltk.tokenize import word_tokenize\nfrom nltk.corpus import stopwords\nfrom nltk.stem import PorterStemmer, WordNetLemmatizer\n\ncontraction_dict = {\"dont\": \"do not\", \"aint\": \"is not\", \"isnt\": \"is not\", \"doesnt\": \"does not\"\n, \"cant\": \"cannot\", \"mustnt\": \"must not\", \"ll\":\"will\" , \"re\": \"are\" ,\"ll\": \"will\", \"wont\": \"will not\" ,\"hasnt\": \"has not\"\n, \"havent\": \"have not\", \"arent\": \"are not\", \"ain't\": \"is not\", \"aren't\": \"are not\"\n,\"can't\": \"cannot\", \"‘cause\": \"because\", \"could've\": \"could have\"\n, \"couldn't\": \"could not\", \"didn't\": \"did not\", \"doesn't\": \"does not\", \"don't\": \"do not\"\n, \"hadn't\": \"had not\", \"hasn't\": \"has not\", \"haven't\": \"have not\", \"he'd\": \"he would\"\n,\"he'll\": \"he will\", \"he's\": \"he is\", \"how'd\": \"how did\", \"how'd'y\": \"how do you\"\n, \"how'll\": \"how will\", \"how's\": \"how is\", \"I'd\": \"I would\", \"I'd've\": \"I would have\"\n, \"I'll\": \"I will\", \"I'll've\": \"I will have\",\"I'm\": \"I am\", \"Iam\": \"I am\", \"I've\": \"I have\"\n, \"i'd\": \"i would\", \"i'd've\": \"i would have\", \"i'll\": \"i will\", \"i'll've\": \"i will have\"\n,\"i'm\": \"i am\", \"i've\": \"i have\", \"isn't\": \"is not\", \"it'd\": \"it would\", \"it'd've\": \"it would have\"\n, \"it'll\": \"it will\", \"it'll've\": \"it will have\",\"it's\": \"it is\", \"let's\": \"let us\"\n, \"ma'am\": \"madam\", \"mayn't\": \"may not\", \"might've\": \"might have\"\n,\"mightn't\": \"might not\",\"mightn't've\": \"might not have\", \"must've\": \"must have\"\n, \"mustn't\": \"must not\", \"mustn't've\": \"must not have\", \"needn't\": \"need not\"\n, \"needn't've\": \"need not have\",\"o'clock\": \"of the clock\", \"oughtn't\": \"ought not\"\n, \"oughtn't've\": \"ought not have\", \"shan't\": \"shall not\", \"sha'n't\": \"shall not\"\n, \"shan't've\": \"shall not have\", \"she'd\": \"she would\", \"she'd've\": \"she would have\"\n, \"she'll\": \"she will\", \"she'll've\": \"she will have\", \"she's\": \"she is\"\n, \"should've\": \"should have\", \"shouldn't\": \"should not\", \"shouldn't've\": \"should not have\"\n, \"so've\": \"so have\",\"so's\": \"so as\", \"this's\": \"this is\",\"that'd\": \"that would\", \n\"that'd've\": \"that would have\", \"that's\": \"that is\", \"there'd\": \"there would\", \n\"there'd've\": \"there would have\", \"there's\": \"there is\", \"here's\": \"here is\",\"they'd\": \"they would\"\n, \"they'd've\": \"they would have\", \"they'll\": \"they will\", \"they'll've\": \"they will have\", \n\"they're\": \"they are\", \"they've\": \"they have\", \"to've\": \"to have\", \"wasn't\": \"was not\",\n\"we'd\": \"we would\", \"we'd've\": \"we would have\", \"we'll\": \"we will\", \n\"we'll've\": \"we will have\", \"we're\": \"we are\", \"we've\": \"we have\", \n\"weren't\": \"were not\", \"what'll\": \"what will\", \"what'll've\": \"what will have\", \n\"what're\": \"what are\", \"what's\": \"what is\", \"what've\": \"what have\", \"when's\": \"when is\",\n\"when've\": \"when have\", \"where'd\": \"where did\", \"where's\": \"where is\", \"where've\": \"where have\", \n\"who'll\": \"who will\", \"who'll've\": \"who will have\", \"who's\": \"who is\", \"who've\": \"who have\",\n\"why's\": \"why is\", \"why've\": \"why have\", \"will've\": \"will have\", \"won't\": \"will not\", \"won't've\": \"will not have\",\n\"would've\": \"would have\", \"wouldn't\": \"would not\", \"wouldn't've\": \"would not have\", \"y'all\": \"you all\", \n\"y'all'd\": \"you all would\",\"y'all'd've\": \"you all would have\",\"y'all're\": \"you all are\",\n\"y'all've\": \"you all have\", \"you'd\": \"you would\", \"you'd've\": \"you would have\", \n\"you'll\": \"you will\", \"youll\":\"you will\", \"you'll've\": \"you will have\", \"you're\": \"you are\", \"you've\": \"you have\"}\n\n\nnltk.download('stopwords')\nnltk.download('punkt')\nnltk.download('wordnet')\nnltk_stopwords = stopwords.words('english')\n\nnltk_stopwords.remove('not')\n\nstemmer = PorterStemmer()\nlemmatizer = WordNetLemmatizer()","metadata":{"execution":{"iopub.status.busy":"2021-06-10T16:24:54.764896Z","iopub.execute_input":"2021-06-10T16:24:54.765282Z","iopub.status.idle":"2021-06-10T16:25:54.881781Z","shell.execute_reply.started":"2021-06-10T16:24:54.76524Z","shell.execute_reply":"2021-06-10T16:25:54.880708Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Định nghĩa hàm `clean()` để tiền xử lý các dữ liệu theo các bước đã nêu ở trên","metadata":{}},{"cell_type":"code","source":"def clean(text):        \n    # chuyển từ về dạng unicode \n    text = unidecode(text).encode(\"ascii\")\n    text = str(text, \"ascii\")\n\n    # chuyển về dạng chữ thường, bỏ các link, kí tự đặc biệt, chữ số.\n    text = text.lower()\n    text = re.sub('https?://\\S+|www\\.\\S+', '', text)\n    text = re.sub('<.*?>+', '', text)\n    text = re.sub('[%s]' % re.escape(string.punctuation), '', text)  \n    text = re.sub('\\n', '', text)\n    text = re.sub('[’“”…]', ' ', text)  \n    text = ''.join(i for i in text if not i.isdigit())\n\n    # Chuyển các từ viết tắt trong từ điển về dạng thường\n    tokens = word_tokenize(text)\n    tokens = [contraction_dict.get(token) if (contraction_dict.get(token) != None) else token for token in tokens]\n    text = \" \".join(tokens)\n\n    # Bỏ các từ chứa ở trong stop-words   \n    tokens = word_tokenize(text)\n    tokens_without_sw = [word for word in tokens if not word in nltk_stopwords]\n\n    # Chuyển từ về dạng số nhiều về dạng thường\n    text = [lemmatizer.lemmatize(word) for word in tokens_without_sw ] \n    text = \" \".join(text)\n\n    return text\n","metadata":{"execution":{"iopub.status.busy":"2021-06-10T16:25:54.88309Z","iopub.execute_input":"2021-06-10T16:25:54.883448Z","iopub.status.idle":"2021-06-10T16:25:54.892935Z","shell.execute_reply.started":"2021-06-10T16:25:54.883402Z","shell.execute_reply":"2021-06-10T16:25:54.892049Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"word = clean(\"有毒 they're you'll aren't mean in chinese takes taking taked\")\nprint(word)","metadata":{"execution":{"iopub.status.busy":"2021-06-10T16:25:54.893814Z","iopub.execute_input":"2021-06-10T16:25:54.894076Z","iopub.status.idle":"2021-06-10T16:25:54.914597Z","shell.execute_reply.started":"2021-06-10T16:25:54.89405Z","shell.execute_reply":"2021-06-10T16:25:54.913518Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_data['clean_questions'] = train_data['question_text'].apply(clean)\ntrain_data","metadata":{"execution":{"iopub.status.busy":"2021-06-10T16:25:54.915653Z","iopub.execute_input":"2021-06-10T16:25:54.915989Z","iopub.status.idle":"2021-06-10T16:36:03.703211Z","shell.execute_reply.started":"2021-06-10T16:25:54.915957Z","shell.execute_reply":"2021-06-10T16:36:03.702165Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X = train_data['clean_questions']\ny = train_data['target']\n\nX.shape,y.shape","metadata":{"execution":{"iopub.status.busy":"2021-06-10T16:36:03.704427Z","iopub.execute_input":"2021-06-10T16:36:03.704722Z","iopub.status.idle":"2021-06-10T16:36:03.728434Z","shell.execute_reply.started":"2021-06-10T16:36:03.704693Z","shell.execute_reply":"2021-06-10T16:36:03.727271Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"non_toxic_data = X[y == 0]\ntoxic_data = X[y == 1]\n\nprint(\"Dữ liệu Label 0 (Tiền xử lý):\\n \", non_toxic_data[:10])\nprint('\\n---------------------------\\n')\nprint(\"Dữ liệu Label 1 (Tiền xử lý):\\n \",toxic_data[:10])","metadata":{"execution":{"iopub.status.busy":"2021-06-10T16:36:03.729859Z","iopub.execute_input":"2021-06-10T16:36:03.730187Z","iopub.status.idle":"2021-06-10T16:36:03.813501Z","shell.execute_reply.started":"2021-06-10T16:36:03.730155Z","shell.execute_reply":"2021-06-10T16:36:03.812442Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from collections import Counter\n\n# top 30 words in a non toxic data\np = Counter(\" \".join(non_toxic_data).split()).most_common(30)\nrslt = pd.DataFrame(p, columns=['Word', 'Frequency'])\n\nrslt.plot(x='Word',kind = \"barh\", figsize=(12,10), title=\"Top 30 từ sử dụng nhiều nhất của label 1\")","metadata":{"execution":{"iopub.status.busy":"2021-06-10T16:36:03.814683Z","iopub.execute_input":"2021-06-10T16:36:03.814953Z","iopub.status.idle":"2021-06-10T16:36:07.740263Z","shell.execute_reply.started":"2021-06-10T16:36:03.814925Z","shell.execute_reply":"2021-06-10T16:36:07.739397Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# top 30 words in a toxic data\np = Counter(\" \".join(toxic_data).split()).most_common(30)\nrslt = pd.DataFrame(p, columns=['Word', 'Frequency'])\n\nrslt.plot(x='Word',kind = \"barh\", figsize=(12,10), title=\"Top 30 từ sử dụng nhiều nhất của label 1\")","metadata":{"execution":{"iopub.status.busy":"2021-06-10T16:36:07.741513Z","iopub.execute_input":"2021-06-10T16:36:07.741788Z","iopub.status.idle":"2021-06-10T16:36:08.395222Z","shell.execute_reply.started":"2021-06-10T16:36:07.741761Z","shell.execute_reply":"2021-06-10T16:36:08.394136Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Tách dữ liệu để huấn luyện mô hình, dự đoán kết quả\n\n- Dữ liệu train = 80% \n- Dữ liệu test = 20%\n","metadata":{}},{"cell_type":"code","source":"from sklearn.model_selection import train_test_split\n\nX_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 1)\n\nprint(X_test.shape, y_test.shape)\nprint(X_train.shape, y_train.shape)","metadata":{"execution":{"iopub.status.busy":"2021-06-10T16:36:08.396573Z","iopub.execute_input":"2021-06-10T16:36:08.396862Z","iopub.status.idle":"2021-06-10T16:36:08.728275Z","shell.execute_reply.started":"2021-06-10T16:36:08.396835Z","shell.execute_reply":"2021-06-10T16:36:08.727278Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## 2.2. Vector hóa dữ liệu\n\n- CountVectorizer N-gram = (1,3) \n \n=> Sử dụng `CountVectorizer` với n-gram trong khoảng [1,3] để trích xuất được trong khoảng 2 từ và 3 từ liên tiếp nhau.\n- Trích xuất thêm dữ liệu để hiểu câu từ trong ngữ cảnh nào hơn. <br>\nGiả sử như: ta có từ `germany` thì sẽ chưa hiểu nó là thuộc label nào. <br>\nNhưng nếu mà ta trích được `germany hitler` => khả năng cao câu hỏi thuộc label 1.","metadata":{}},{"cell_type":"code","source":"from sklearn.feature_extraction.text import CountVectorizer\n\ncount_vectorizer = CountVectorizer(analyzer=\"word\", ngram_range=(1,3))","metadata":{"execution":{"iopub.status.busy":"2021-06-10T16:36:08.729499Z","iopub.execute_input":"2021-06-10T16:36:08.729812Z","iopub.status.idle":"2021-06-10T16:36:08.734477Z","shell.execute_reply.started":"2021-06-10T16:36:08.729784Z","shell.execute_reply":"2021-06-10T16:36:08.733291Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Count Vectorizer\ncount_train = count_vectorizer.fit(X)\n\nX_train_vec = count_train.transform(X_train)\nX_test_vec = count_train.transform(X_test)","metadata":{"execution":{"iopub.status.busy":"2021-06-10T16:36:08.737028Z","iopub.execute_input":"2021-06-10T16:36:08.737334Z","iopub.status.idle":"2021-06-10T16:38:33.99294Z","shell.execute_reply.started":"2021-06-10T16:36:08.737306Z","shell.execute_reply":"2021-06-10T16:38:33.991871Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"- Thông thường `CountVectorizer` chỉ cần bigram là đủ.","metadata":{}},{"cell_type":"code","source":"def get_evaluation(predictions, test):\n    print('Recall: ', recall_score(predictions, test))\n    print('F1 score :', f1_score(predictions, test), '\\n')\n    print(classification_report(test, predictions))","metadata":{"execution":{"iopub.status.busy":"2021-06-10T16:38:33.994611Z","iopub.execute_input":"2021-06-10T16:38:33.99504Z","iopub.status.idle":"2021-06-10T16:38:34.000361Z","shell.execute_reply.started":"2021-06-10T16:38:33.994996Z","shell.execute_reply":"2021-06-10T16:38:33.999164Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## 2.3. Huấn luyện mô hình","metadata":{}},{"cell_type":"code","source":"# Thư viện\nfrom sklearn.linear_model import LogisticRegression\nfrom sklearn.svm import LinearSVC\nfrom imblearn.over_sampling import RandomOverSampler\nimport numpy as np\n\nfrom sklearn.metrics import accuracy_score, recall_score ,f1_score\nfrom sklearn.metrics import classification_report","metadata":{"execution":{"iopub.status.busy":"2021-06-10T18:19:52.075697Z","iopub.execute_input":"2021-06-10T18:19:52.076073Z","iopub.status.idle":"2021-06-10T18:19:52.081497Z","shell.execute_reply.started":"2021-06-10T18:19:52.07604Z","shell.execute_reply":"2021-06-10T18:19:52.080503Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<h2> Mô hình Logistic Regression </h2>\n- Count Vectorizer với N-grams = [1,3] <br>      \n","metadata":{}},{"cell_type":"code","source":"lr = LogisticRegression(class_weight = 'balanced', random_state=0)\nlr.fit(X_train_vec, y_train)\ny_pred = lr.predict(X_test_vec)","metadata":{"execution":{"iopub.status.busy":"2021-06-10T18:12:49.781613Z","iopub.execute_input":"2021-06-10T18:12:49.782022Z","iopub.status.idle":"2021-06-10T18:16:22.576598Z","shell.execute_reply.started":"2021-06-10T18:12:49.78199Z","shell.execute_reply":"2021-06-10T18:16:22.575607Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(\"Logistic Regression with Count Vectorizer\\n\")\nget_evaluation(y_pred, y_test)","metadata":{"execution":{"iopub.status.busy":"2021-06-10T18:16:22.578093Z","iopub.execute_input":"2021-06-10T18:16:22.578388Z","iopub.status.idle":"2021-06-10T18:16:23.245559Z","shell.execute_reply.started":"2021-06-10T18:16:22.578359Z","shell.execute_reply":"2021-06-10T18:16:23.244372Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<h2> Mô hình SVC </h2>\n- Count Vectorizer với N-grams = [1,3] <br>      \n","metadata":{}},{"cell_type":"code","source":"lsvc = LinearSVC(random_state=3, tol=0.01, loss='hinge', C=1, verbose=2, class_weight='balanced')\n\nlsvc.fit(X_train_vec, y_train) \ny_pred2 = lsvc.predict(X_test_vec)","metadata":{"execution":{"iopub.status.busy":"2021-06-10T16:48:08.595031Z","iopub.execute_input":"2021-06-10T16:48:08.595452Z","iopub.status.idle":"2021-06-10T16:49:36.814541Z","shell.execute_reply.started":"2021-06-10T16:48:08.595416Z","shell.execute_reply":"2021-06-10T16:49:36.813475Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print('-----------\\n')\nprint(\"SVC with Count Vectorizer\\n\")\nget_evaluation(y_pred2, y_test)","metadata":{"execution":{"iopub.status.busy":"2021-06-10T16:49:36.816018Z","iopub.execute_input":"2021-06-10T16:49:36.816317Z","iopub.status.idle":"2021-06-10T16:49:37.460812Z","shell.execute_reply.started":"2021-06-10T16:49:36.816285Z","shell.execute_reply":"2021-06-10T16:49:37.459713Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<h2> Mô hình SVC + LR + Oversampling</h2>\n- Count Vectorizer với N-grams = [1,3] <br>      \n","metadata":{}},{"cell_type":"code","source":"# define oversampling + undersampling strategy\noversample = RandomOverSampler(sampling_strategy=0.5)\n\n# fit and apply the transform\nX_over, y_over = oversample.fit_resample(X_train_vec, y_train)","metadata":{"execution":{"iopub.status.busy":"2021-06-10T17:45:28.456786Z","iopub.execute_input":"2021-06-10T17:45:28.457275Z","iopub.status.idle":"2021-06-10T17:45:29.665531Z","shell.execute_reply.started":"2021-06-10T17:45:28.457237Z","shell.execute_reply":"2021-06-10T17:45:29.66454Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from collections import Counter\n\nprint(Counter(y_over))\ny_over.shape","metadata":{"execution":{"iopub.status.busy":"2021-06-10T17:45:31.464933Z","iopub.execute_input":"2021-06-10T17:45:31.465295Z","iopub.status.idle":"2021-06-10T17:45:31.725794Z","shell.execute_reply.started":"2021-06-10T17:45:31.465264Z","shell.execute_reply":"2021-06-10T17:45:31.724631Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"lr2 = LogisticRegression(class_weight='balanced')\nlsvc2 = LinearSVC(class_weight='balanced')\n\nlr2.fit(X_over, y_over)\ny_pred3 = lr2.predict(X_test_vec)","metadata":{"execution":{"iopub.status.busy":"2021-06-10T17:46:51.202615Z","iopub.execute_input":"2021-06-10T17:46:51.203101Z","iopub.status.idle":"2021-06-10T17:50:42.96393Z","shell.execute_reply.started":"2021-06-10T17:46:51.203049Z","shell.execute_reply":"2021-06-10T17:50:42.963003Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"lsvc2.fit(X_over, y_over)\ny_pred4 = lsvc2.predict(X_test_vec)","metadata":{"execution":{"iopub.status.busy":"2021-06-10T17:50:42.965567Z","iopub.execute_input":"2021-06-10T17:50:42.96587Z","iopub.status.idle":"2021-06-10T17:59:37.608395Z","shell.execute_reply.started":"2021-06-10T17:50:42.965839Z","shell.execute_reply":"2021-06-10T17:59:37.607382Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print('Logistic Regression - Count Vectorizer - Oversampling')\nget_evaluation(y_pred3, y_test)","metadata":{"execution":{"iopub.status.busy":"2021-06-10T17:59:37.610288Z","iopub.execute_input":"2021-06-10T17:59:37.610592Z","iopub.status.idle":"2021-06-10T17:59:38.271715Z","shell.execute_reply.started":"2021-06-10T17:59:37.61056Z","shell.execute_reply":"2021-06-10T17:59:38.270777Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print('Linear SVC - Count Vectorizer - Oversampling')\nget_evaluation(y_pred4, y_test)","metadata":{"execution":{"iopub.status.busy":"2021-06-10T17:59:38.27287Z","iopub.execute_input":"2021-06-10T17:59:38.273156Z","iopub.status.idle":"2021-06-10T17:59:38.921121Z","shell.execute_reply.started":"2021-06-10T17:59:38.273129Z","shell.execute_reply":"2021-06-10T17:59:38.920016Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## So sánh mô hình\n\n- Nhận xét: Kết quả mô hình không thay đổi sau lần OverSampling ","metadata":{}},{"cell_type":"code","source":"print(\"Logistic Regression with Count Vectorizer\\n\")\nget_evaluation(y_pred, y_test)\nprint('-------------\\n')\nprint(\"SVC with Count Vectorizer\\n\")\nget_evaluation(y_pred2, y_test)\nprint('-------------\\n')\nprint('Logistic Regression - Count Vectorizer - Oversampling')\nget_evaluation(y_pred3, y_test)\nprint('-------------\\n')\nprint('Linear SVC - Count Vectorizer - Oversampling')\nget_evaluation(y_pred4, y_test)","metadata":{"execution":{"iopub.status.busy":"2021-06-10T18:17:38.454024Z","iopub.execute_input":"2021-06-10T18:17:38.454422Z","iopub.status.idle":"2021-06-10T18:17:41.054392Z","shell.execute_reply.started":"2021-06-10T18:17:38.454388Z","shell.execute_reply":"2021-06-10T18:17:41.053412Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<h2> Chọn mô hình Logistic Regression đầu tiên để kiểm tra. </h2>","metadata":{}},{"cell_type":"markdown","source":"\n","metadata":{}},{"cell_type":"markdown","source":"# 3. Tạo submission","metadata":{}},{"cell_type":"code","source":"test_data['clean_questions'] = test_data['question_text'].apply(clean)\ntest_data","metadata":{"execution":{"iopub.status.busy":"2021-06-10T14:48:29.892641Z","iopub.execute_input":"2021-06-10T14:48:29.892992Z","iopub.status.idle":"2021-06-10T14:51:25.091341Z","shell.execute_reply.started":"2021-06-10T14:48:29.892961Z","shell.execute_reply":"2021-06-10T14:51:25.090314Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X = test_data['clean_questions']\n\nX.shape","metadata":{"execution":{"iopub.status.busy":"2021-06-10T14:51:25.093088Z","iopub.execute_input":"2021-06-10T14:51:25.093437Z","iopub.status.idle":"2021-06-10T14:51:25.099747Z","shell.execute_reply.started":"2021-06-10T14:51:25.093405Z","shell.execute_reply":"2021-06-10T14:51:25.099039Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_vec_test = count_vectorizer.transform(X)","metadata":{"execution":{"iopub.status.busy":"2021-06-10T14:51:25.101447Z","iopub.execute_input":"2021-06-10T14:51:25.101876Z","iopub.status.idle":"2021-06-10T14:51:37.112774Z","shell.execute_reply.started":"2021-06-10T14:51:25.101833Z","shell.execute_reply":"2021-06-10T14:51:37.111818Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"predictions = lr.predict(X_vec_test)","metadata":{"execution":{"iopub.status.busy":"2021-06-10T14:51:37.114414Z","iopub.execute_input":"2021-06-10T14:51:37.114854Z","iopub.status.idle":"2021-06-10T14:51:37.163405Z","shell.execute_reply.started":"2021-06-10T14:51:37.114807Z","shell.execute_reply":"2021-06-10T14:51:37.162499Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"predictions.shape","metadata":{"execution":{"iopub.status.busy":"2021-06-10T14:51:37.164507Z","iopub.execute_input":"2021-06-10T14:51:37.164771Z","iopub.status.idle":"2021-06-10T14:51:37.17028Z","shell.execute_reply.started":"2021-06-10T14:51:37.164745Z","shell.execute_reply":"2021-06-10T14:51:37.169358Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"test_data['prediction'] = predictions\n\nresults = test_data[['qid', 'prediction']]\nresults.to_csv('submission.csv', index=False)","metadata":{"execution":{"iopub.status.busy":"2021-06-10T14:51:37.17147Z","iopub.execute_input":"2021-06-10T14:51:37.17177Z","iopub.status.idle":"2021-06-10T14:51:38.144662Z","shell.execute_reply.started":"2021-06-10T14:51:37.171742Z","shell.execute_reply":"2021-06-10T14:51:38.143737Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]}]}