{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"## Large datasets\n![dataset.png](attachment:dataset.png)\n\n> As a Data Scientist or Kaggler, we crave for more data\n\nHow many times have you complained about not having good enough data for working on a particular problem? Plenty I'm sure.   \nHow many times have you complained about having too much data to work with? Maybe not many but it's still a better problem to have.\n\nThe most common resolution in the scenario of having too much data is using a part/sample of it that fits in the RAM available. But that leads to wastage of the unused data and sometimes loss of information. Many times there are ways to overcome this challenge without the need of subsampling. A single solution might not cater to all requirements and so different solutions can work in different scenarios.\n\nThis notebook aims to describe and summarize some of these techniques. The [Riiid! Answer Correctness Prediction](https://www.kaggle.com/c/riiid-test-answer-prediction) dataset is quite a nice sample to experiment on since the plain vanilla ***pd.read_csv*** will result in an out-of-memory error on Kaggle Notebooks. It has over 100 million rows and 10 columns.\n\nDifferent packages have their own way of reading data. The methods explored in the notebook (Default ***pandas*** and rest alphabetically):\n\n* [Pandas](#Method:-Pandas)\n* [Dask](#Method:-Dask)\n* [Datatable](#Method:-Datatable)\n* [Rapids](#Method:-Rapids)\n\nApart from methods of reading data from the raw csv files, it is also common to convert the dataset into another format which uses lesser disk space, is smaller in size and/or can be read faster for subsequent reads. The file types explored in the notebook (Default ***csv*** and rest alphabetically):\n\n* [csv](#Format:-csv)\n* [feather](#Format:-feather)\n* [hdf5](#Format:-hdf5)\n* [jay](#Format:-jay)\n* [parquet](#Format:-parquet)\n* [pickle](#Format:-pickle)\n\nNote that just reading data is not the end of the story. The final decision of which method to use should also consider the downstream tasks and processes of the data that will be required to run. But that is outside the scope of this notebook.\n\nYou will also find that for different datasets or different environments, there will be different methods that work best. So there is no clear winner as such.\n\nFeel free to share other approaches that can be added to this list.","metadata":{},"attachments":{"dataset.png":{"image/png":"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"}}},{"cell_type":"markdown","source":"## Methods\nBefore exploring various methods let's once confirm that reading the dataset using the default pandas setting fails.","metadata":{}},{"cell_type":"markdown","source":"## Method: Pandas\n![pandas.png](attachment:pandas.png)\n\n[Pandas](https://pandas.pydata.org/) is probably the most popular method of reading datasets and is also the default on Kaggle. It has a lot of options, flexibility and functions for reading and processing data.\n\nOne of the challenges with using pandas for reading large datasets is it's conservative nature while infering data types of the columns of a dataset often resulting in unnecessary large memory usage for the pandas dataframe. You can pre-define optimal data types of the columns (based on prior knowledge or sample inspection) and provide it explicitly while reading the dataset.\n\nThis is the method used in the [official starter notebook of the RiiiD competition](https://www.kaggle.com/sohier/competition-api-detailed-introduction) as well.\n\nDocumentation: https://pandas.pydata.org/docs/","metadata":{},"attachments":{"pandas.png":{"image/png":"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"}}},{"cell_type":"code","source":"import pandas as pd\nimport datatable as dt\n\n# confirming the default pandas doesn't work (running the below code should result in a memory error)\n# data = pd.read_csv(\"../input/riiid-test-answer-prediction/train.csv\")","metadata":{"execution":{"iopub.status.busy":"2021-06-02T10:08:32.294678Z","iopub.execute_input":"2021-06-02T10:08:32.295352Z","iopub.status.idle":"2021-06-02T10:08:32.374931Z","shell.execute_reply.started":"2021-06-02T10:08:32.295308Z","shell.execute_reply":"2021-06-02T10:08:32.374174Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"%%time\n\ndtypes = {\n    \"row_id\": \"int64\",\n    \"timestamp\": \"int64\",\n    \"user_id\": \"int32\",\n    \"content_id\": \"int16\",\n    \"content_type_id\": \"boolean\",\n    \"task_container_id\": \"int16\",\n    \"user_answer\": \"int8\",\n    \"answered_correctly\": \"int8\",\n    \"prior_question_elapsed_time\": \"float32\", \n    \"prior_question_had_explanation\": \"boolean\"\n}\n\ndata = pd.read_csv(\"../input/riiid-test-answer-prediction/train.csv\", dtype=dtypes)\n\nprint(\"Train size:\", data.shape)","metadata":{"execution":{"iopub.status.busy":"2021-06-02T08:36:34.302199Z","iopub.execute_input":"2021-06-02T08:36:34.302535Z","iopub.status.idle":"2021-06-02T08:44:20.694002Z","shell.execute_reply.started":"2021-06-02T08:36:34.302502Z","shell.execute_reply":"2021-06-02T08:44:20.69268Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data.head()","metadata":{"execution":{"iopub.status.busy":"2021-06-02T08:44:20.69625Z","iopub.execute_input":"2021-06-02T08:44:20.696601Z","iopub.status.idle":"2021-06-02T08:44:20.739031Z","shell.execute_reply.started":"2021-06-02T08:44:20.69656Z","shell.execute_reply":"2021-06-02T08:44:20.738044Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# free memory\ndel data\nimport gc\ngc.collect()","metadata":{"execution":{"iopub.status.busy":"2021-06-02T08:44:20.741023Z","iopub.execute_input":"2021-06-02T08:44:20.741339Z","iopub.status.idle":"2021-06-02T08:44:20.890517Z","shell.execute_reply.started":"2021-06-02T08:44:20.741308Z","shell.execute_reply":"2021-06-02T08:44:20.889187Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Method: Dask\n![dask.png](attachment:dask.png)\n\n[Dask](https://dask.org/) provides a framework to scale pandas workflows natively using a parallel processing architecture. For those of you who have used [Spark](https://spark.apache.org/), you will find an uncanny similarity between the two.\n\nDocumentation: https://docs.dask.org/en/latest/","metadata":{},"attachments":{"dask.png":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAWEAAACPCAMAAAAcGJqjAAABPlBMVEX///8QEBEAAADPYDuvr68NDQ7v7++kpKTqqm3qq25hYWHQYz3RZj/PXTvq6urOWjrR0dHopGnilF58fHwVFRbLy8vkmmLopWrg4ODmn2bgj1reiVbTa0LNVjjcg1IyMjLZfE7WdEdYWFjLUDVMTEy7u7uVlZX89vTj4+PVcUfNWC789O2Li4vooWD57uvMUSI9PT3RYjPLTCn35dhsbGwlJSbUazuRkZHmsaA4ODiBgYH03dfqvK7bfUTx0b3hkFLxzLH22bzelHnvw57kqZP25eHVeGrqrn3vzsXaiG/sxL3qt5ffiUzkoHbwyrXpr4X23szvuoLwwZLrqmHptZzhmXP55dHz0Kvhk2bxxZragVruwqXXeVLkpIPLTBfYdD7vuH3fmYnQZFLHPiHYhXvhpJ3HPCTUdmrYf2THNwD3HpUBAAAT8klEQVR4nO2d+0PiSBLHGRAB8YlPBBFUjIKgoDIoII7KjA903FF33Rldxp3dO+///wcu/Uqqk07SQVxuvHx/cGchQPikqK6qru74fJ48eXLURb9P4K1LaW73+xTeuFrty36fwhvXQbvtGfGr6mJ+/nO/z+Ft63I+5nni15QyPx+LeUb8itpuq4Q9I35FXSHCsU/9Po03rM8I8IxnxK+nGCY843ni11K+TQjPFPt9Jm9V14ywV514JV3MU8Iz+X6fyhvVvUb4sN+n8jaVTyDCCPBMvN/n8jZ1wNzwTHzmut8n8yb1qDmJeNwL2HosxYeLEpRwPB4veGNdT/UB/VFjNUD4sM+n9Lb062/o70UCEI57xYkeKn+DXIKSmGehhAp4ouAVJ3om5Zcv6D8HbTDQqYQP+3xab0hfbrC5XhIngRgjwhNeNNEr/X7zC/pPvkFCidi3S0LYiyZ6pA83YRxJfEtQwspnTHi2cNDvU3sb+hCOfMfBcIISvvAdEsITX/p9bm9CKuAINuGDNiHcLvq2m8hJzM6G+31yb0EfCpFIBJmw757asDq+KXFC2HPELxcCTLzwdYMQbqKKz2dK+EOfT+/n12FB9RER/M87DHg+hrvWVEesEg6Hf+/r2b0BHSPABhPG8cN2gRCO9Pf8fnp9KagMwzgW9iUSxIbv8f8plHDYc8QvUP4TttIwTueuGoRw84o8ORtHoUTYK028QNuzmCFxtUoiRQjH6LPHhHDYG+q61mFhgjDEkdptg3iJNsviDgrk2eM+nuJPLeUzdbQFPBvXaqQIYWbC6lCHnmZO2pNbXcdnSM42QcKxuxQh3NZmP5UCdtIkn/bkVsdNmrPNkmzuoEEJ41hY+RX9JcFaJOIFE+61HaflXzVnw6FCXgWMBzqygOM3/JcEGhEvXHOt/EWTzMKphAt4cs53lyaEE7hXLX+D7foLIXxjDNcG/daaWlzZWc2Oyp3I4BDUkvMLRgc2lnbBRw3t7G+8D2XH10YHRa5sekl/8znxO2aX4BmE5M7bQco3bZ5zYmKCTHVeIRPGhPGJ/k7yuGNqw8a+lEF/0Eb42y+FZCBn+auTdTh8bgcdFQy+Q9I+iuLeW9rZD02Pcx87AA4YF74lbyxTgxIn7aiD2DybhUM2jIm20tSEGxjm9g0Z/A4LxA8bA2KV8Dt7qd89uOE8QK744Yv8Q/ZHT/ptPpfRDu6Bjx3Q39+C8BA8A8dLLKPrWDumz9UXZrEDUFJpasKP+KDvlOkHlbAKOPKb4U2cCZOvPOBwMqN+/hXv7ExocFHmU9WPBW/iSHiAAzzpcL4SOrhrs2lO3G9yTK73XYMSnsdj2mH4BocSvmtK2FhdkyKMvqyDZ13lCdvakLLnF3/KSwiPBsEX8e864XOScnXfppNwpO1vgvrX206aECbZXD4cuSHRw3Uh/BLC6kmv2J7RouFt7NzEpBxgd4SXOBMW+xFpbd+isgMg3KQG7DtVAWPCKeIjvoQjNIvbfilhe8TjRmgcHYdje0E4xAHel6BoqfzVPcuJabtJ8xMLwWoaYFK0PCxoSPOUsHEyVJ6wrW8zm6W15x56BcJr8Gv492RAitWq3zVowsZsuBnX5uiLnQwjjBfGqHYbYdW0HhC2gTa4a3oXS8c9KAvYDWEYyQS79RHK6clXFSELdvEUkRqtHWoRjfIVAUYHNAj02QlWLsaEI46EDdGw+RtbBcZZATUrN8GN+e+CfgvhEFmWMDfO+jekmepqqXQzuoUywvPNb3rEqAKmhBu3+IGL+GxYK/XIEBZ9Wx6a1fAl+uH7LZIqzqH4pzZCA5pCqjYmJyeHllZWFnd3d2UJz3E+YlGWKoKmFGv16tFDqVTJEHyQcGL+AkTkytcKOSTVIKPc4QyqVzKiEoSDfsNzo+MhQ+Bq8fsbFf3w/VviLwUdigtzsyO8yPkIi5zaqOLT09nzQzSnwk0m19fXTYTb91ewiqMBTqfJKLc9E0eJMsvhtrsgjBTizNi/IzzZkIhw0L8mPNjv+HZC2RDe4E5Rth7R+iMXCESHh5NJBhgSTjUu+R40DJgckcCjXH4GEdbcsEbYJloTEvbNBTlHLfTE5nHO+rvCgc7NkGRNeJzzEfZhO1BxMxBVAQ8TwDzhRuLWUCIrrleYlZMwwofaLdFAx4647pawwcmJqMEAFx4rdBPAowSDLuYDLAkrvI8Q/3AEAoQ5G043Eo+mdVu1SoX5kQZhv91UCatOQnMK3RPmfIAwBtsH0KbgwCnyiJDwlBwLLEvCk135CPXKmGw4nc5kOve3NfN1Py0l2TXIUOO+iPFu2PeBErap/FgR9m0BKIIYTAFOwv8eZK/+VcGb9ZrwNAdY2kcICFcq6ccr4Qr7ailJCKcznRp9dQy74bDeCng4QQjbVC8tCcMIVlDRgcGwfxRYvNBN9JowF89bBuwCIS+BAQ9XKpXk+v3JqcX+BcpzKUkJZzIUsO+qSQjrU8vHBHBXhAc5IzU9PcQxHQVfWeQmekx4hzNhpxorVHETm/Bw8uikXrPZHKKWLCU1wi326CW1YZ3nF0KY1jJ1yRCGrs78O+SQDnBFLsHl6DFhLpl0KvvzYoS/2h9WzUXJYIichAb4uo0Ic/3CEWrDxplQKcJZ6IiNT4YMXjpkj7CnhEenOB/hauKIEX62Peg5F40SV52pfNUt/XI+hoK1WZBe0IHue1eEYcBmcnWg6IIzCFjmEgS8XLRm9+UMEhJ+ycRRjRI+szmmvjkSjdLBsHKkP66aMCI8W9DDOlZa+258DynCMKXwT/NPzcFxDj8HkZsLtb3MOPiJIxcJIlItRwgfWR/xXB7R0r7SCXjmfh674YmIHtj9Sm3Y1FUlRxhCM1gKSFmDu/gDB+z9ACTsYjZNQHiUy4V2XU4unxLC0SeL55Wn8gICTAhXTsFTaG8fRBiuAWXBmqkJXo4wHOoMMT30CRv0LY1WbfUCFfH0nEmjQlSQMA1Rll7gI3y+c0yYt02g6kJ5bGxkhER0pQcu2LhPUMIgN/liEaxJEgb1V0M5bFrwo4cWb7bSLY6LSO92V1ZNcR78ZSxuIS1yb+R6crlKCddFT54vlBcQYELYcBWuUvOYcPwQPDhLCZs6tOUIDxhGM10wGN4zH00dB9R7QyHOohNmy2CT0OeSQjYsN3UxufxECZ+anzr9+GN5YWEBEVaPKSX5Q/KJBF6PH58A341Vh82NgXKEQbjGB52DoNSj+Q9Y1Te7iWkDYQv5/UPcxRmwfVkXE0fP1AHUjE+c//ljeZkSDkQDm2eGdOQxhQnzG3aw2uV30+e4J8zVfriEWvOfQ7ZuAsawttS2FItPMh/axeTyMCXc4h5V/l5GfDFgRDg3bLRxvAYJ2TC3X8fxrMVAJ0kY2B2f1C0J3Qd0E36TmzA2r1hzgx9lR9hqOsVWJJRIluBjtb9+/Gt5WSc8lnsynT5u90Fz/FyJ89OsuO7TDWH4bbhgWHecsP/GPMYru9KIQdZtS7iLPssWccPJdfjg3/9ZXgaEN59NLsR3m04RG+b2kshTwDfm7uGX2fB7izFtyHJkROImJewE82A7wsEpFzU1KhIOJ/mEo/UvQLg8cm5+Wa2j2jCa5W9yLK8t3fALCcNcD7pCzj2b3y7rl7RiUGG298Mu6sJUJzSUqHKP/pshVvmKwjgFzTIhws1v3OPHVvnGC2MJUTBsfFdhKjC3Z9fdqiuoB2EOsYSo2G+rMzrQ8QPZ+Q9K+E9hmOw7ymAnQRcza1IixEuYSpe+F8bDOxbumXcTwuafgRW988S6DwZMRHE/C1Mvh/s+n6Q4lKA2XBY4CFVXnTQZ6AxbZqOuwLAwGn5ZTqdAS92fG9c1twqjCXHBYC27sTO0srfIa5cnp/8A4EVeWULik0OX4USLhhJJw+MfCeKysFrRqqQJ4XneRyAnIexZQ5IjDI1VH7f50pZlr5CrmQd1HITdxfr1hIRJFjPIxSQue6nOLSprzE0sCF5TzGQoYeM2rbNkgYEgVpMlDMwF8FqRG6+cmrtNguOn5pPMhPk+A1GRyUZPwoFOFSVcNsdpvuckJWy8S8+vtP1duJZOivDglCgNnpMMulzNTyKFRE5cQJjvNXLVfeF7IGVJc878sUxCCRN63xHuqkLd71eGZz5PEMLCFbdShGHLiY5LPjdzmRBMyxI21C9dTNO1yETzcNJ0VeokaV42TS5VS+u0NdO443u+YOMk5AjDxjD9IMnVGO7dxBwc1NiDQsKDXI3DhcOv5qymkFq0LDFmqPdUS2gyFBG+M76ErfO6Ef6IpAjviaIy6cUCgrk9e8kTVs2dmwqVbql6Jt0oZjfs8/1JCBsccT2HJkOxkzA52whd0SzebE2GcFY4uMuud3nn2k24IGzou5RN7Vi/j9kNq2NgmRDmUo56bpgSbphK7NcFEkqI0g2f664q8Et0szzBTdO0O8K8r5JN7eo5QvhB8Nw5rQ7DiLi+GaWEO+Z9Fz9NEMLfxZ/lsjNQLwJn+XFcIPgqya5pIleE1/xdfNADmeEUztEhR6y64QWQi1Q3A8OEcOfWdPw12RwlYrX/jDPhrPh3yK1NmRKIW0Hoqmww7oawIe+R+rW0cpSwwEn4lDFMeKysjVsnKPDAXT8dwSX5FKc2bLHtgSPhEG8jzKNyvVR7g6NmjXNHyHxxpg3paA2Lb12TmRGtku734ZLw2Y+kOqwNdUf4aES4YwojyO4zKGcOG7tamRwIZ7c4wPpozTUViwcyYOXSiysMpyRFmK/oy6R26zZOAg91mDAZ6ooPKmBCuGJMsZE+0Z3sLHfu4L5OiNPq5ErQuBhJS2K50U+cTA0IQxBHDXJ1CbusmYmLGyW61043CeFcS/j0+Q/sJXKYfy2AAGPCQsAHbCc7y+0YufV05kVt73hpNcJxIQWD4MWDpa/Qe7zCKzuNNL6GpTmX0JS4nmND2LCczjG/OSNuOGrRE1grY8JjOBt5woAR4ZIIsDJLNlgKW2/a6mZNKMhL9yFhq98ldJCwfrtltUrRfFXF1UvTB/LZs0NqV9wMYMI5cZHdVyyTaVAcTNQZYSFguqHorN02a24IgxIOXK5h2QwybeEmhDU54TpUkKXZEh51k9phs4xGo+aaBBWdaN5E/z6nhMU+Oz9DN3a22T+pu3XNXJZnGYnBBR6wR1Cy6vmOcy62hA0hpW01XtnEyKIlq5ZANW8mhNEVqOGjAxb2/okStrtNRHdr85dgmGBtMftiN+GCsP6DtydsWJEk6L3XdEJNWBgMY31cxt0om2ggxC4lkBO0Xqk6YNuwRWwKpy72l9gBLwIP2wws3HioXx9pwrAbzYGwIXu2nrVTyK8+GrDufX8iXYGb6BIgi88lxUGHwrZhK5jW3wHJEg7C2XouGLbrKt2Dtq49Kks4CFE6EeYXVlo3Cp6Q6CuaszRh399l3LOGDVe9IDmrHu7PM/QOBrY3OpEjHPTvQpAQnO1ajFVhyCFJOMgFBU6EJXdSKgaIDQdERR+qOiWM5puVaE5Q4cQ6aFLC9tvrSxBWI6jdVehouHzYtidvTRg2b0k1S/j5VNGRsCF7tvhpHdHYwMKzYp0TL4FHN0XQWEW0zbYSdLgbkt2egVSL+9N8mrQPn7VPh/dgoMveZW11SNss0HKzEL9hZ0C4o52YsMLvjyGs+tPQIGDjhRnhgKXxEs2wmyE53DNNCdlITbymBSc6AI+xf/txeCj3VoPj2YHVnZXFKXHasWMcq6YnNe1YXNXxSSjhmT1QwBYmfH2Hfqun5TFsw7aEPzd/plumDa7NjaMMOpsNra6uqtm0q3KyC9EMLRAQLfEqXt03GozwiANhtOrWu7GiSXSYU03YHH7VHlONVKoBbdhiDQ3SdZNth+ndVxHqjJqwKf6q3aY7ZOsOFBaclvECAxvCxTa77Z93mx6oczbM5WBkVDw9SVfYJlWUMFlgYOkl8jF268qfwgn/Y9J9hIZOpftcYTtQYcKI/bkDYeWS3brSu5M7JxZH4GSjWDs/OXtIVira9jOEMDoQE7as9qBl44ywd9dKqCcWRwTOjtZzuVKpNGzeBSyFjqxTGxb3ELObg6qAJ/7B0//f16kGmFTU2Q5KHOEGbm2vUsLimPmyrW3p6o1yQMWAAbCJcLqRoluAPdkQVi61TXNj3igH9cABNhHOdDqPByzCeFpGgAOi6ptyn6I7jpIb/3liOsrZEO50vj5Ce30m6203zWlJMaED9m5rC1XPBYSEk2owkTmqG1COLKBQIrppepvrFLuhV6x5+M+c+U+iGg+YEI6q0YRKV7B/3QImHDB1BH3Tt9b2AHMq5kwqJZ9PRHCRWmSgM9XoHxvavsSmdQb/5zqvVqsnJydPT08nJ9Vq/fTUbns11JFCCPPFi1aqoe1L3PZ88IukpnQ4lOCS5quOvnu5F0W8UCxYA+GF8sjuwpEwr/Xy5FYfyfYzIFi7yrDamwo4ZutiPDlLGcHhcKDExsHWXUUrDSXaj309uTehFqm/s7lSpQqqm4mGF0S8XHVSuwyQtrZ6hu2vjytDngvugf5aJnUfVB0+TcL6cePOu8tqL4Td8AiaLD190DcnR7U38zokT12IZHQjgeH6c44sQCKEO/eeh+iN6nSgi5aibGPnDFqp6Blwr/RMdrKLslkQTLhzL+5v9eReyh+EcIBt7Izqx1+9GK13ol2BAW3r7PVKpe6FED3U2fKYvu0wWqVYOfH49lIKWsKhE1b5elWI3uq8TPdsxcuekx7fnuvj8gIN1gK5dc//9l5FspJuZCSXO7JZhuCpa+E1MmO53HPdcw+vI9VJlMvPVS+9eC0V//NjxMP7mqr/7eH15MmTp97qv0Hq0qEul1V8AAAAAElFTkSuQmCC"}}},{"cell_type":"code","source":"import dask.dataframe as dd","metadata":{"execution":{"iopub.status.busy":"2021-06-02T09:20:56.67409Z","iopub.execute_input":"2021-06-02T09:20:56.674499Z","iopub.status.idle":"2021-06-02T09:20:59.759681Z","shell.execute_reply.started":"2021-06-02T09:20:56.67441Z","shell.execute_reply":"2021-06-02T09:20:59.758714Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"%%time\n\ndtypes = {\n    \"row_id\": \"int64\",\n    \"timestamp\": \"int64\",\n    \"user_id\": \"int32\",\n    \"content_id\": \"int16\",\n    \"content_type_id\": \"boolean\",\n    \"task_container_id\": \"int16\",\n    \"user_answer\": \"int8\",\n    \"answered_correctly\": \"int8\",\n    \"prior_question_elapsed_time\": \"float32\", \n    \"prior_question_had_explanation\": \"boolean\"\n}\n\ndata = dd.read_csv(\"../input/riiid-test-answer-prediction/train.csv\", dtype=dtypes).compute()\n\nprint(\"Train size:\", data.shape)","metadata":{"execution":{"iopub.status.busy":"2021-06-02T09:21:04.417286Z","iopub.execute_input":"2021-06-02T09:21:04.417716Z","iopub.status.idle":"2021-06-02T09:26:26.495883Z","shell.execute_reply.started":"2021-06-02T09:21:04.417683Z","shell.execute_reply":"2021-06-02T09:26:26.494724Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data.head()","metadata":{"execution":{"iopub.status.busy":"2021-06-02T09:29:53.267918Z","iopub.execute_input":"2021-06-02T09:29:53.26836Z","iopub.status.idle":"2021-06-02T09:29:53.317765Z","shell.execute_reply.started":"2021-06-02T09:29:53.268331Z","shell.execute_reply":"2021-06-02T09:29:53.316412Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# free memory\ndel data\nimport gc\ngc.collect()","metadata":{"execution":{"iopub.status.busy":"2021-06-02T09:29:58.899086Z","iopub.execute_input":"2021-06-02T09:29:58.899468Z","iopub.status.idle":"2021-06-02T09:29:59.045375Z","shell.execute_reply.started":"2021-06-02T09:29:58.899438Z","shell.execute_reply":"2021-06-02T09:29:59.044096Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Method: Datatable\n![py_datatable_logo.png](attachment:py_datatable_logo.png)\n\n[Datatable](https://github.com/h2oai/datatable) (heavily inspired by R's data.table) can read large datasets fairly quickly and is often faster than pandas. It is specifically meant for data processing of tabular datasets with emphasis on speed and support for large sized data.\n\nDocumentation: https://datatable.readthedocs.io/en/latest/index.html\n","metadata":{},"attachments":{"py_datatable_logo.png":{"image/png":"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"}}},{"cell_type":"code","source":"# datatable installation with internet\n# !pip install datatable==0.11.0 > /dev/null\n\n# datatable installation without internet\n# !pip install ../input/python-datatable/datatable-0.11.1-cp37-cp37m-manylinux2010_x86_64.whl > /dev/null\n\nimport datatable as dt","metadata":{"execution":{"iopub.status.busy":"2021-06-02T08:49:07.833238Z","iopub.execute_input":"2021-06-02T08:49:07.833706Z","iopub.status.idle":"2021-06-02T08:49:08.008294Z","shell.execute_reply.started":"2021-06-02T08:49:07.833671Z","shell.execute_reply":"2021-06-02T08:49:08.007029Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"%%time\n\ndata = dt.fread(\"../input/riiid-test-answer-prediction/train.csv\")\n\nprint(\"Train size:\", data.shape)","metadata":{"execution":{"iopub.status.busy":"2021-06-02T08:49:08.011742Z","iopub.execute_input":"2021-06-02T08:49:08.012109Z","iopub.status.idle":"2021-06-02T08:50:26.854047Z","shell.execute_reply.started":"2021-06-02T08:49:08.012078Z","shell.execute_reply":"2021-06-02T08:50:26.852612Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data.head()","metadata":{"execution":{"iopub.status.busy":"2021-06-02T08:50:26.856335Z","iopub.execute_input":"2021-06-02T08:50:26.856658Z","iopub.status.idle":"2021-06-02T08:50:26.878741Z","shell.execute_reply.started":"2021-06-02T08:50:26.856627Z","shell.execute_reply":"2021-06-02T08:50:26.877499Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# free memory\ndel data\nimport gc\ngc.collect()","metadata":{"execution":{"iopub.status.busy":"2021-06-02T08:50:26.880621Z","iopub.execute_input":"2021-06-02T08:50:26.881008Z","iopub.status.idle":"2021-06-02T08:50:26.998645Z","shell.execute_reply.started":"2021-06-02T08:50:26.880964Z","shell.execute_reply":"2021-06-02T08:50:26.997577Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Method: Rapids\n![rapids.png](attachment:rapids.png)\n\n[Rapids](https://rapids.ai/) is a great option to scale data processing on GPUs. With a lot of machine learning modelling moving to GPUs, Rapids enables to build end-to-end data science solutions on one or more GPUs.\n\nDocumentation: https://docs.rapids.ai/","metadata":{},"attachments":{"rapids.png":{"image/png":"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"}}},{"cell_type":"code","source":"# rapids installation (make sure to turn on GPU)\nimport sys\n!cp ../input/rapids/rapids.0.19.0 /opt/conda/envs/rapids.tar.gz\n!cd /opt/conda/envs/ && tar -xzvf rapids.tar.gz > /dev/null\nsys.path = [\"/opt/conda/envs/rapids/lib/python3.7/site-packages\"] + sys.path\nsys.path = [\"/opt/conda/envs/rapids/lib/python3.7\"] + sys.path\nsys.path = [\"/opt/conda/envs/rapids/lib\"] + sys.path \n!cp /opt/conda/envs/rapids/lib/libxgboost.so /opt/conda/lib/","metadata":{"execution":{"iopub.status.busy":"2021-06-02T08:50:27.000095Z","iopub.execute_input":"2021-06-02T08:50:27.000412Z","iopub.status.idle":"2021-06-02T08:52:25.239523Z","shell.execute_reply.started":"2021-06-02T08:50:27.000379Z","shell.execute_reply":"2021-06-02T08:52:25.238272Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import cudf","metadata":{"execution":{"iopub.status.busy":"2021-06-02T09:27:20.651244Z","iopub.execute_input":"2021-06-02T09:27:20.651606Z","iopub.status.idle":"2021-06-02T09:27:22.451917Z","shell.execute_reply.started":"2021-06-02T09:27:20.651576Z","shell.execute_reply":"2021-06-02T09:27:22.450896Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"%%time\n\ndata = cudf.read_csv(\"../input/riiid-test-answer-prediction/train.csv\")\n\nprint(\"Train size:\", data.shape)","metadata":{"execution":{"iopub.status.busy":"2021-06-02T09:27:24.734629Z","iopub.execute_input":"2021-06-02T09:27:24.735014Z","iopub.status.idle":"2021-06-02T09:27:57.374126Z","shell.execute_reply.started":"2021-06-02T09:27:24.734977Z","shell.execute_reply":"2021-06-02T09:27:57.372771Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data.head()","metadata":{"execution":{"iopub.status.busy":"2021-06-02T09:28:04.554775Z","iopub.execute_input":"2021-06-02T09:28:04.555201Z","iopub.status.idle":"2021-06-02T09:28:04.629749Z","shell.execute_reply.started":"2021-06-02T09:28:04.555172Z","shell.execute_reply":"2021-06-02T09:28:04.628421Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## File Formats\nIt is common to convert a dataset into a format which is easier or faster to read or smaller in size to store. There are various formats in which datasets can be stored though not all will be readable across different packages. Let's look at how these datasets can be converted into different formats.\n\nMost of them are available in pandas: https://pandas.pydata.org/pandas-docs/stable/user_guide/io.html\n","metadata":{}},{"cell_type":"code","source":"# reading data from csv using datatable and converting to pandas\n# data = dt.fread(\"../input/riiid-test-answer-prediction/train.csv\").to_pandas()\n\n# writing dataset as csv\n# data.to_csv(\"riiid_train.csv\", index=False)\n\n# writing dataset as hdf5\n# data.to_hdf(\"riiid_train.h5\", \"riiid_train\")\n\n# writing dataset as feather\n# data.to_feather(\"riiid_train.feather\")\n\n# writing dataset as parquet\n# data.to_parquet(\"riiid_train.parquet\")\n\n# writing dataset as pickle\ndata.to_pickle(\"riiid_train.pkl.gzip\")\n\n# writing dataset as jay\n# dt.Frame(data).to_jay(\"riiid_train.jay\")","metadata":{"execution":{"iopub.status.busy":"2021-06-02T10:03:37.804729Z","iopub.execute_input":"2021-06-02T10:03:37.805089Z","iopub.status.idle":"2021-06-02T10:03:45.746781Z","shell.execute_reply.started":"2021-06-02T10:03:37.805059Z","shell.execute_reply":"2021-06-02T10:03:45.745199Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# free memory\n# del data\n# import gc\n# gc.collect()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Reading Different file formats","metadata":{}},{"cell_type":"markdown","source":"## Format: csv\nMost Kaggle datasets are available in csv format and is pretty much the standard format in which datasets are shared. Almost all methods can be used to read data from csv.\n\nRead more: https://en.wikipedia.org/wiki/Comma-separated_values","metadata":{}},{"cell_type":"code","source":"%%time\n\ndtypes = {\n    \"row_id\": \"int64\",\n    \"timestamp\": \"int64\",\n    \"user_id\": \"int32\",\n    \"content_id\": \"int16\",\n    \"content_type_id\": \"boolean\",\n    \"task_container_id\": \"int16\",\n    \"user_answer\": \"int8\",\n    \"answered_correctly\": \"int8\",\n    \"prior_question_elapsed_time\": \"float32\", \n    \"prior_question_had_explanation\": \"boolean\"\n}\n\ndata = pd.read_csv(\"../input/riiid-test-answer-prediction/train.csv\", dtype=dtypes)\n\nprint(\"Train size:\", data.shape)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Format: feather\nIt is common to store data in feather (binary) format specifically for pandas. It significantly improves reading speed of datasets.\n\nRead more: https://arrow.apache.org/docs/python/feather.html","metadata":{}},{"cell_type":"code","source":"%%time\n\ndata = pd.read_feather(\"../input/riiid-train-data-multiple-formats/riiid_train.feather\")\n\nprint(\"Train size:\", data.shape)","metadata":{"execution":{"iopub.status.busy":"2021-06-02T10:01:30.232339Z","iopub.execute_input":"2021-06-02T10:01:30.232708Z","iopub.status.idle":"2021-06-02T10:01:54.748804Z","shell.execute_reply.started":"2021-06-02T10:01:30.232673Z","shell.execute_reply":"2021-06-02T10:01:54.747777Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Format: hdf5\nHDF5 is a high-performance data management suite to store, manage and process large and complex data.\n\nRead more: https://www.hdfgroup.org/solutions/hdf5","metadata":{}},{"cell_type":"code","source":"%%time\n\ndata = pd.read_hdf(\"../input/riiid-train-data-multiple-formats/riiid_train.h5\", \"riiid_train\")\n\nprint(\"Train size:\", data.shape)","metadata":{"execution":{"iopub.status.busy":"2021-06-02T10:04:29.655915Z","iopub.execute_input":"2021-06-02T10:04:29.656337Z","iopub.status.idle":"2021-06-02T10:05:14.382864Z","shell.execute_reply.started":"2021-06-02T10:04:29.656302Z","shell.execute_reply":"2021-06-02T10:05:14.381623Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Format: jay\nDatatable uses .jay (binary) format which makes reading datasets blazing fast. An example notebook is shared [here](https://www.kaggle.com/rohanrao/riiid-with-blazing-fast-rid) and also shown below which reads the entire dataset in less than a second!\n\nRead more: https://datatable.readthedocs.io/en/latest/api/frame/to_jay.html","metadata":{}},{"cell_type":"code","source":"import datatable as dt","metadata":{"execution":{"iopub.status.busy":"2021-06-02T10:06:10.564212Z","iopub.execute_input":"2021-06-02T10:06:10.564709Z","iopub.status.idle":"2021-06-02T10:06:10.651486Z","shell.execute_reply.started":"2021-06-02T10:06:10.564675Z","shell.execute_reply":"2021-06-02T10:06:10.650310Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"%%time\n\ndata = dt.fread(\"../input/riiid-train-data-multiple-formats/riiid_train.jay\")\n\nprint(\"Train size:\", data.shape)","metadata":{"execution":{"iopub.status.busy":"2021-06-02T10:06:12.135219Z","iopub.execute_input":"2021-06-02T10:06:12.135640Z","iopub.status.idle":"2021-06-02T10:06:12.163278Z","shell.execute_reply.started":"2021-06-02T10:06:12.135604Z","shell.execute_reply":"2021-06-02T10:06:12.161871Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Format: parquet\nIn the Hadoop ecosystem, parquet was popularly used as the primary file format for tabular datasets and is now extensively used with Spark. It has become more available and efficient over the years and is also supported by pandas.\n\nRead more: https://parquet.apache.org/documentation/latest/","metadata":{}},{"cell_type":"code","source":"%%time\n\ndata = pd.read_parquet(\"../input/riiid-train-data-multiple-formats/riiid_train.parquet\")\n\nprint(\"Train size:\", data.shape)","metadata":{"execution":{"iopub.status.busy":"2021-06-02T10:06:19.537968Z","iopub.execute_input":"2021-06-02T10:06:19.538468Z","iopub.status.idle":"2021-06-02T10:07:01.391021Z","shell.execute_reply.started":"2021-06-02T10:06:19.538427Z","shell.execute_reply":"2021-06-02T10:07:01.389865Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Format: pickle\nPython objects can be stored in the form of pickle files and pandas has inbuilt functions to read and write dataframes as pickle objects.\n\nRead more: https://docs.python.org/3/library/pickle.html","metadata":{}},{"cell_type":"code","source":"%%time\n\ndata = pd.read_pickle(\"../input/riiid-train-data-multiple-formats/riiid_train.pkl.gzip\")\n\nprint(\"Train size:\", data.shape)","metadata":{"execution":{"iopub.status.busy":"2021-06-02T10:08:42.213820Z","iopub.execute_input":"2021-06-02T10:08:42.214553Z","iopub.status.idle":"2021-06-02T10:09:10.573997Z","shell.execute_reply.started":"2021-06-02T10:08:42.214497Z","shell.execute_reply":"2021-06-02T10:09:10.572617Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## There is no winner or loser\n![choice.png](attachment:choice.png)\n\nEach method has it's own set of pros and cons. Some examples are:   \n* **Pandas** requires a lot more RAM to handle large datasets.\n* **Dask** can be slow at times especially with transformations that cannot be parallelized.\n* **Datatable** doesn't have a very exhaustive set of data processing functions.\n* **Rapids** is not useful if you don't have a GPU.\n\nSo it's a good idea to explore various options and finally choose whichever appropriately fits the requirements. I strongly believe in not marrying a technology and continuously adapting to newer ideas, better approaches and ultimately the best possible solutions for building data science pipelines.\n\nEven in my personal experience I've found different approaches working well on different datasets. So don't shy away from experimentation.\n\n> Data Science is blooming under the blessings of open source packages and communities","metadata":{},"attachments":{"choice.png":{"image/png":"iVBORw0KGgoAAAANSUhEUgAABLQAAAOmCAIAAAB9gFvqAAAACXBIWXMAAAsTAAALEwEAmpwYAAAgAElEQVR42uzdd3SVd4Ln6eqa6TkzPXtmJ/yx22dmd2bP9s6cnbOnnSt0lavKuRzLdjkAImOwwRiDAza2MY6YYHKOBgyInLMIkkAgEIggiwwimCByFkJI2pdSNUURFe7Vvfe9z3O+Z053TxkL3feV7qeudH8/KQcAACDp/cSnAAAAAHEIAACAOAQAAEAcAgAAIA4BAAAQhwAAAIhDAAAAxCEAAADiEAAAAHEIAACAOAQAAEAcAgAAIA4BAAAQhwAAAIhDAAAAxCEAAADiEAAAAHEIAACAOAQAAEAcAgAAIA4BAAAQhwAAAIhDAAAAxCEAAADiEAAAAHEIAACAOAQAAEAcAgAAIA4BAAAQhwAAAIhDAAAAxCEAAADiEAAAAHEIAACAOAQAAEAcAgAAIA4BAAAQhwAAAIhDAAAAxCEAAADiEAAAAHEIAACAOAQAAEAcAgAAIA4BAAAQhwAAAIhDAAAAxCEAAADiEAAAAHEIAACAOAQAAEAcAgAAIA4BAAAQhwAAAIhDAAAAxCEAAADiEAAAAHEIAACAOAQAAEAcAgAAIA4BAAAQhwAAAIhDAAAAxCEAAADiEAAAAHEIAACAOAQAAEAcAgAAIA4BAAAQhwAAAIhDAAAAxCEAAADiEAAAAHEIAACAOAQAAEAcAgAAIA4BAAAQhwAAAIhDAAAAxCEAQEgcPHh4zuxFnTv3bv7aO08/Ve/Xv3q2Yg/++tkXX2jyVusOPXsOXrYs6+zZcz5XgDgEAAib48dPjBwxvk6d1+/6x4cqszGjJ/mkAeIQACA89u7Z3+nTbg/c/3glszDYt90H+rwB4hAAICROnz7Tq9eQ++97rPJZGOzDD78qLS312QPEIQBAGKzMynnk4T9WKQuDNW/+bnFxic8eIA4BABLexYvFnTv3rmoWBquf0ur8+Qs+gYA4BABIeKdPn2napG01yvCZp+sfO3bCJxAQhwAACW///gN/eK5hNcrw7rseHjJ4zOTJsys2c+aCzIxVeXlbDh0q9FkFxCEAQCL58ceDjz/2SjXK8Pb75S+eatDgzc8/7zFt2txDB7QiIA4BAOLY0aNHn346JeJleOOefaZ+l2/6rs/NKysr82kHxCEAQBw5ffrMiy80qYUyvHa/f6JO377DDxw45PMPiEMAgNgrKSl5rVm7Wi7Da39Z8Z12n65du9EDAYhDAIBY+vKLHrEqw2vXsEHr1atzPRyAOAQAiIERI8bFQxleXauW7Xft2uNxAcQhAEDtWbQo/e67HoqrOAx27z2Pftt94Llz5z1AgDgEAIi6/PxtP//Z7+OtDK99u5qcNes9TIA4BACIokMHCh9+6MW4LcN/fq+ah3r0GHTxYrHHCxCHAACRV1RUVOfV1+O8DK/ujy823bZtp0cNEIcAABH2aceuiVKGFXvg/sdnz17ogQPEIQBAxKSOn5ZYZXh1vXoNKS0t9QgC4hAAoKbW5+bdd++jCRqHwdq17ehXEAFxCABQI0ePHn/0kZcStwwr1qxpW6dcAOIQAKCaSkpKGjV8K9HLsGJNmrx9/vwFjykgDgEAquybzn1CkIUP3P/4hx9+tXbtRg8oIA4BAKosM2NVomfhs8/UHzFi3MmTJz2agDgEAKiOw4ePPPjgc4mbhe+/93lOzgaPIyAOAQCqr7S09LVm7RL6NcM6r76+Zct2DyUgDgEAqm/UqAkh+FXD++97bNq0uR5NQBwCAFRHQcG+B+5/IhzvUBps4IDvPKaAOAQAqJrS0tLGjd4KTRlWbMSIcR5ZQBwCAFTBuHFTQ1aGwe6+66GMjJUeXEAcAgBUytGjR3/5i6fCF4fBHnzwuePHT3iIAXEIAHBnHT78OpRlWLGPOnztIQbEIQDAHazPzQtxGVbM4RaAOAQAuJ3S0tJXXm4e+jh8951OHmtAHAIA3NLcuYtDX4bB7rn7kcOHj3i4AXEIAHATpaWlf3iuYTLEoWMtAHEIAHBLM2cuSJIyDFa3zusecUAcAgBJbd26jYsWpt+4p59OSZ44DHb69BkXA4hDAIDkNXny7KSKwFstK2uNiwHEIQBA8rp4sfjhh14Uh4MHj3YxgDgEAEhqQRclXMs1e77RI796JoJ/YOs3O7gSQBwCACS1EydO/eyBJxIrDn9+32MbBo/o9ub79979SET+wGefqe9KAHEIAJDsvv66d8K9ePjiYy8XzZ6zbeSYOk/Wqfmf9oufP+kyAHEIAJDs9u798e67Hk64PuzU7O3yBQsuzZ036N2ONX8J8ezZc64EEIcAAMmuzVsfRTzeWr7U9Inf/iGqfTi3c4+gD4NtGDziyd8+X5M/ateuPS4DEIcAAMmrqKioT++h997zaMTLrXDSlLL58zcNHdnzrQ+e+t0L0YjDXz7wxJ6xqRV9eGb6jPfrt6z2H7Vp02YXA4hDAIAklbNm/XPPNohGtj3z8IsVzXZ1O0ePHfRux+cfe7ky//g9dz28ccjI4B9Z1K3PyA8++6JFuxZ/bPL0QzcpzJcef+XinDlX/y1jP/nqnmr9iOyGDT+4HkAcAgAknePHT3T48OsI1uDzj7507f/62Z9+IfCm2/bd9wPf+eSFv/7P37gGz6SUzr/+nz09fUbesO/mfdNz0LufdGjcOuXper/5xZNfv/7Otf+ZjF79/6nq77+6bt1GVwWIQwCAJFJWVjZ92rwHH3wusi8Vbho6stnzja7+r7P/+bcBb7Pdo8cOfb/Ty79/9VZ/5tQvut7xDwl2atqMsvnzr/2/7Bj1/WO/frZKH3/OmvWuDRCHAADJoqBgX9MmbSP+Q6SP/vrZkrlzj02e+ug/J9mh1EmV6bqK7RubOuKDz+o+Xe+6P/a3v3jq5NTplf9zrt2esamP/KoKfbg+N8/lAeIQACD8iotLBg8e/cD9j18XRfff99iTv6/7xxeb1iQOg7S7+q6h993z6NMPvVC9ojuQOnH0R180eCbl6tEanzdvW70/KtiuMeMe/tXTXVu+F/wPJ6dNX9l3UMoN/Xl127fvcpGAOAQACLm8vC1vte7QpHGbDh9+3bvXkPHjpi1duiI/f9uRI8cq/gNpaRmV7MB7737kd798+tr/y5O/ff7CrNlXk2xcx68/bdqm2kVXscMTJ4/95Ksmf2gQ/Os2DR1Z7T/n1LQZ1/6vxXPmNn+x8U3/XgcPHnadgDgEAEh2I0eMr2Qc1n2q7u9/85fTC+++6+GsvoNu/DHRGsbh1R2dPGXdoOGR+tOCHZow6YF7b3Jux9mz51wGIA4BAJLd55/3qGQcDn2/07Vx2POtDyJYbrWz9244DvGB+x8vKytzGYA4BABIdm3f/qQyZXj3XQ//OH7C1Ths/mLjy/PmJVwcjvn4y+tPZXy6vmsAxCEAAOXNmlbqLUwbP9cgiKuKOHz193XOzZyVcGUYbMaX3a/7ezVv/q5rAMQhAADlr7zcvDJxOO+bnhVx+MzDLx6bPDURyzDYyA8+u+7v9WnHrq4BEIcAAJS//NJrdyzDR371zKW5V36ItOnzDQ+kTkzQMgz25ivNrvurDRs21jUA4hAAgPI6r75+xzgc9G7Hirg6M31G4pbhztFj7/nnQxSvbumS5a4BEIcAAJSn1Gt5+zK8/55Hjk6ekrhNWLHT02e88vs6N/7t9u7Z7xoAcQgAQHmrlu1vH4cfNmyV6GWYO2jYtYdwXHOOxROlpaWuARCHAACUf/zxN7ePw41DRyZ6HM7tfPOzHBs3essFAIhDAIArevcacpsyrPtU3UQvw2BHJ0+5+4bfNgzWo8cgFwAgDgEArli4cNkdT7AIwV56/JUb/3ZLFme6AABxCABwRWHh0VuV4VO/e+HyvHnhiMOh73e68S945MgxFwAgDgEA/uzZZ+rfNA5TO3YORxkG2zduwnU/Wfr0U/U89IA4BAD4i++/n3xjGf7ul08XzZ4TmjgM1vT5htf+BT/q8LWHHhCHAAB/cfbsuX/65dPXxeHg9zqGqQyDpXXvc+1fcMqU2R56QBwCAPyVESPGXRtOv7j/8ZNTp4csDi/Pm/f0Qy9c/TsePnzE4w6IQwCAv1JaWtqkydtXw6nbm++HrAwrNuur7hV/wZdfes2DDohDAICb2L//wG8e/EMQTvfe/cih1EmhjMPS+fOff+zl4O/Yp/dQjzggDgEAbm7btp2/++3zH7UM58uGFVvQo18Qh+vWbfRwA+IQAOCWdu0s2JWdE+I4LM3Nfe/dz0pLSz3WgDgEALitS5fKFy4MbR/++GNxcYkHGRCHAACVsH59OMtw0aIr6QsgDgEAKqWwMJxxuNGvGgLiEACg8kpLy5cuDWEcHj3qsQXEIQBAVezaFbYyXLHCowqIQwCAKiopKV+8OFRxeOiQRxUQhwAAVbd9e3jKMDOz3PEVgDgEAKiOS5fC85uHhYUeT0AcAgBU14EDYSjDjAyPJCAOAQBqZvXqMBxvWOLge0AcAgDUxLlz5WlpCd+Hu3d7JAFxCABQMwcPJnwcpqeXl5V5JAFxCABQM3l5jrIAxCEAQNK7fLk8Ozux43DVKg8jIA4BAGrs0qXyFSsSuw9PnvQwAuIQAKDGioqu/PJe4sbhhg0eQ0AcAgBEqA8T+vXDCxc8hoA4BACIhEuXrvz+XoLG4bZtHkBAHAIAREhJSfn69QkZh4sXX/ngAcQhAEDE/Phj+aJFideHe/d66ABxCAAQUadOlS9fnmBxmJl5+XLpkcLTHj1AHAIARE5ZWXlBQXlaWgL14bGCQ+vX7tmx7VBxsR8xBXEIAEAEXbhw5aCIBInDXWu2BnEYbNOGvSdPnPPogTgEACCizp8v37y5fOHCeC7D0oWLNubsrojDiu3ZfeRyyWWPHohDAAAi6sKF8p07yzMz4zMOTy3LurYMK/bDxv1nzjgCEcQhAADRcPJk+dat5cuWxVUc7s3aeGMc/mkFu3cV7tx98Oy5Ig8diEMAACKtqKh8/vw4KcOyBQs2rdl1izjcs3D+hlfrdX21Xpe27w4bMHDOwkW5R46c8gCCOAQAIELi5riLM0syb1WGwebOzv1THF67Lh91HJ2eseny5VIPI4hDAABqZvPmOInD/Ss23CYOZ85Yc0Mc/nmtWg/MWbvdIwniEACAGjh8OC7icOHCH9bsvE0cTpucfas4rNigIfO8hAjikATWd//Wl/OWh2av/LB85pH9HlYAEsmlS/EQh+cWZ9ymDINNnJB1+zgM1qffTI8niEMSVf38rJ8sGx+yPbx+8Q9nT3pwAUgYK1fGPA4PZK67fRyOG5txxzgMNmr0Yq8fgjhEHMbR/nZZapvtOadKLnmIAUgA+fkxj8PNq7ffPg5Hj1pamTgM9mabQXPn5RQX+y4M4hBxGDf7T8un9N63pbTUf38JQHzbvz+2ZXhh8bLbl2Gw4cPTKhmHFWvSvPeQ4fO3bNlfVuYBBnGIOIyP3ZczP+vkEY81APHrzJnYxuGhzLV3jMMhgxdUKQ6vrk3bIWPGLg0qsbRUJoI4RBzGen+zdFyD/KzDFy94xAGIR6Wl5WlpMYzDrdnb7hiH/QfMq14cXt1rr/ft02/W0mUbjx497TEHcYg4jOX+bcbETrs3Xrx82eMOQNxZvTpWZVi8aMkdyzBY7z6zahiH165128FDhs9flb317Fn/1S2IQ8RhjPbfs2fNPfqjhx6A+LJ5c6zisDBjTWXisEePGRGMw6urk9K1w8ejx6Wmr1u389y5IhcCiEPEYW3vmY3pu86fdQEAEC/27YtVHG7P3lKZOOzSdWo04vDGVxT7DZidtmT9vn1HvJMNiEPEYS3tX6VfOe7iTEmxywCA2DtxIiZlWLIwbX1OQWXi8KuvJ9ZCHF67Fi379+ozfe68nB07DhQXl7hGQBwiDqO7/5w1bfTBXf6rSQBi7NKl8vnzaz8Oj6ZnV6YMg332+fhajsNrV7d+t3feG9ZvwOygFbds2V9SohVBHCIOo7Pf5KZtOHPC9QBArJw6ef5c5sraj8Odq/IrGYefdBwbwzi8bg0a9+jY6fuRo9LSMzbt23fEmcYgDhGHkdxPl45vkJ91pPiiqwKAWlZ88dKmDXt/zKnt96S5vHDRhsr9TGmwDzuMip84vG6Nmva8thVdUSAOEYcR2H/InNx735bLfgUegNpSVla2bcvBoL7y1u6u5Tg8vmxVJcsw2HvtR8ZtHN74y4pdu0+ZNGV5ztrtzlQEcYg4rNHuXjMv48RhlwcAtWD/3uNXA+z84ozajMPdWXmVj8O27YYnShxet2bN+3zZOXXM2KXLV+Sv37B74OC5P/541IUH4hBxWLXjLvZccNwFAFF06uT53Gt+sPPA8nW1VoalCxdtzNld+Ths/dbgBI3DG/dBh++8pQ2IQ8Rh1fZ3GRM67d5YdNn3DwAi7+LFSxvX7702wPLX7Ki1ODy1LKvyZRjs9VYDQhOHwSZNXu4KBHGIOKzy/o+s6aMP7nK1ABBBpaWlW/MP3NhgFxan104c7s3aWKU4fK1F3zDFYUrD7n64FMQh4rCae3j94h/OnnTNABAR+/Ycu2mDHcysjZ8sLVuwYNOaXVWKw8ZNe4UpDoN99sU470AH4hBxWM397bLUNttzTpVccuUAUBMnjp+7VYNtXr29FuLwzJLMKpVhsJSG34YsDoMtX5HvagRxiDis/v7T8im9921x3i4A1VNUdGlj7t7bZNiFxcuiHYf7V2yoahzWSemWCL3XpUr/+eZv9Dt7rsg1CeIQcVij3ZczP+ukY3YBqJrLl0u33OxXDa/docy10Y3DhQt/WLOzSmWYk707fC8bVmz4yIUuSxCHiMOa7m+WjmuQn3Xoov/GEYDK2ltw9I4lFu2fLD23OKOqLxuuytoR1jisk9J1/37vTAPiEHEYif3bjImddm+8ePmyywmA2ygrK//hh72VjLGitCj+ZOmB5blVjcMVGVvCGofBevWZ7voEcYg4jNj+e/asuUd/dEUBcFMnTpz5usukQYPnVzLGoveTpecWZ+RV8WdKgy1bmh/iOHy1Xpft2w+4SkEcikNxGMk9szF91/mzrisArrUqe2vjZlfOgZg6ZVUlY2zL6m3RKMNj6dkbcgqqWobB0hZuCnUcdu3SbbILFcShOBSHEd6/Sr9y3MWZkmJXFwAnT57t9u3UqwWSmb658j1WlLY0kgcbLly4r+rvUHp18+fmhjsOX63XxW8egjgUh+IwKvvPWdNGH9zlZF2AZLYqe2uz5n2u5kfDJj1zc6rQY4cj95OllxYt3r5qS7XLMNjsWWvDHoddBw+d56IFcSgOtVy09pvctA1nTrjMAJLNqVPnevSadl17vPf+yCr12Nbsqv1kaenCRbf6JcMfqv5Lhtdt+rTVoY/DlIbdT570uyEgDsWhRW0/XTq+QX7WkeKLLjaAJLFu3c4WLfvf5Lfauk6tapJdrPRPlpYtuHJ0YbCCrE3H0rOLFy2p4S8ZXrdJE7NCH4fBps1Y6QIGcSgOLbr7D5mTe+/bcrnMz5kChNnZc0VDhs+/VXiMGL64qkl2OGNNJePw9NIV1/2z+Wt27Fi1ueZZWLHU8cuTIQ7bvTfMZQziUBxabezuNfMyThx21QGE0vr1u95o1f824TFrRk5Vk2xb9tZKxuHurLxIdeBN9/2YZckQh8H27Cl0MYM4FIdWe8dd7LngVxoAwuP8+Yu3ecHw6lau2FaNKqvMT5aWLErbEM0yDDZy5JIkicNxqekuaRCH4tBqb3+XMaHT7o1Fl0tcgQCJbuPG3S1bD7hjcrzWom/1qqwwI+eOcViYsWZ9lONw2NBFSRKHb78z1FUN4jBJvTRj0v86qp/FZP8wdkiXJYvXr883s3jYnj37fVOgSoqKi8emVvaHLTt8NLp6VbZ59fY7xuGW7G3RjsNBA+cnSRwGO3r0tMsbxGHSKSsr++2Dr5iZWbBOHXv6vkDlbd26v027IZXvjR49Z1Y7zM4uybxNGZ5bkhHtMgzWr9/c5InD5SvyXeEgDsWhmZk4hDsoLr5U+RcMr27MmPRqh1lBVt5t4nDfig21EIe9es5Knjj8ftxS1zmIQ3FoZiYO4Xa2bf+x7bvDqtEb8+etr3aYbcgpKFmUdquD7zfl7K6FOOzWfVrcR12XSP1RX3ZOdamDOBSHZmbiEG7u8uXSSZOX10mpTmzUSemWk72rJm1WmHnzt6U5nr6qFsow2DffTEmeVw6bNO/tZGIQh+LQzEwcwk3s2VPY/sOR1Y6Nlm8OrGGb3eptabZH7pj72+/LryYmTxwGKyw86bIHcSgOzczEIfxFScnlGbNW1WvQvSal8WmncTXPsxvfluZi2tLaKcNgwV8hqeJwVfYWFz+IQ3FoZiYO4c/27Tvy4cejal4a/frNrXme7cnadF0cHlyeW2tx+NEn3ydVHHpPGhCH4tDMTBzCX14wTGnYPSKlMSF1ec3z7Lq3pSlbsPCHNTtrLQ7bfzgqqeLw/Q9GugtAHIpDMzNxSLI7dOjEp5+NjWBpLE7Li0ihXfu2NKeXrqi1Mgz2znvDkyoOgx09etq9AOJQHJqZiUOS9jtjedqS9Q0a94hgY9Rr8G1uTkFECu3at6XZnZVXm3HYpu3QZIvD4EpwR4A4FIdmZuKQZFRYePKzL8ZHvDGCrIpgpFW8LU3JorQNtViGwVq2HpRscdi9x1Q3BYhDcWhmJg5JxhcMGzbpGY3G+PKriRGMtIq3pSnMWLO+duOw+Rv9ki0O6zf69vz5i+4OxCHi0MxMHJJELxh2+GR09Bpj8KAFEYy0irel2ZK9rZbjsOlrfZItDoNlZOa5QRCHiEMzM3FIsmjUtFdUA2PqlFWR7bSClXm1XIbBIvt7mImyzl0muUEQh4hDMzNxSLKo3yi62ZOZvrn2Wy7iq9egexLGYd363U6ePOseQRwiDs3MxCFJoU5Kt+jVRaMmPXNzEr4Mc3MKkrAMKzZ/wVr3COIQcWhmJg4Jv/PnL0Y1Ld57f2QIXjZcs2pn0sbhJ52+d5sgDhGHZmbikPArLDwZ1bTo0nVqCOJw5fJtSRuHr9brElwk7hTEIeKwmtu0aeu5c+fNzKK3fn1HicNwO3zxwr6ic7XwL9qx80BU02LE8MUhiMPM9M1JHIddp01f6ZZEHCIOq7kdO/b43AJRNXjQOHEYYoUXL/x/q+f815UzCi5E/b1A1q/fFdWumDUzJwRxuDQtL5nj8J33h7srEYeIQ3EIiENiU4Y/WTY+WC30YebyH6LaFStXbA9BHC6avzGZ4zBYQcFh9ybiEHEoDgFxSGzKsHb6cN78nOgVxWst+oWgDIPNnZOb5HE4ZuxStyfiEHEoDgFxSC05fPHC//zrMqzY/7Nq1sXLl6P0L504OTN6RfHhx6PDEYczZ+QkeRy+3qp/aWmZmxRxiDgUh4A4jF+7du1t0ujdWG3qlPmR+ovc+JphxX66dPyog7ui9wkcPnJR9IqiZ6+Z4YjDaZOzkzwOg23MK/C1F3GIOBSHgDiMX/n5O4IPvu3bn3/+We/a35IlWQldhoFefWZELye+H7MsHHE4cUKWOBw8dJ6vvYhDxKE4BMRhvMfhunV5iftoxrAMA192To1eTsyftz4ccThubKY4bNaiT0nJZV9+EYeIQ3EIiENxGMIyDLTv8F2UWqJOSrec7F3hiMNRo5eJw2B5P3iGgzhEHIpDQByKwzCWYaBV64FRColWbw4MRxkGGzF8sTIMNnzkIl9+EYeIQ3EIiENxGMIyDDRo1CNKIfFpp3GhicMhQxYqw2Bt2g7x5RdxiDgUh4A4FIchLMPi4kvRC4l+/eaGJg77D5inDCt29OhpX4ERh4hDcQiIQ3EYqjIMHD9+JnoVMSF1RWjisE+f2bKwYstX5PsKjDhEHIpDQByKw1CVYWDPnsLoVcSStLzQxGGPHjNkYcWGDV/gKzDiEHEoDgFxKA5DVYaBvLyCKCVEvQbf5uYUhCYOu3abKgsr1unzcb4CIw4Rh1XYl5/36dN7pJlZ9Nb8tQ/EYcLFYbyVYWBV9tYoJcTbbYeGpgyDfd15kiz885vQth7oCSTiEHFoZhbyicNkK8PAwrTcKCXEV19NDFMcfvZ5qiysWKOmPT2BRBwiDs3MxKE4DFUZBqZOy4pSQgwZvCBMcdjx07Gy8OqKioo9h0QcIg7NzMShOAxPGQZGjYnW2e7TJmeHKQ47fDRaE17dgQPHPIdEHCIOzczEoTgMTxkG+g+s/AkNXarUD5npW8IUh++1H6kJr25NzjbPIRGHiEMzM3EoDsNThoFvuk6Kzq+l9QpTGQZr984wTXh1Eydneg6JOEQcmpmJQ3EYnjIMfNQxKj8t+d77I0MWh2+1GaIJr+6bbpM9h0QcIg7NzMShOAxPGQbatI1K83TtNjVkcfhGq4Ga8OpatOznOSTiEHFoZiYOxWF4yjDQpHnvaMTDyBGLQxaHr7Xopwmv3cGDxz2NRBwiDs3MxKE4DEkZlpaWRakcZs3MCVkcNm7aWxBeuyVLN3gaiThEHJqZiUNxGIYyDJw+fS5K5bByxfaQxWH9hj0E4bUbMHCOp5GIQ8ShmZk4FIdhKMPAjz8ejUY2vNaiX8jKMFidlG6C8Nq1bjPI00jEIeLQzEwcisMqlOG/WBanZRjYsnV/NLKhw0djQlaGa1fvVoM37ujR055JIg4Rh3feJx917/rNILN4m/gxcVj7ZTg6XsswsCZnWzSaoWevmSGLw+ysHVLwxi1fke+ZJOIQcXjn7dixx+eWOCR+TBzeKu3KavABHL544X8mYBkGlizbGI1m+H7MspDF4YqMrVLwxg0bvsA3VsQh4lAcIg5NHIYkDjedPfFQ7uKfLh3fa9+WZCvDwMxZ2dFohgXz1ocsDtOX5EvBG/fOe8N8Y0UcIg7FIeLQxGEY4nDn+TN/lzGhouX+VXrq6lNHk6oMA2NTl0U8GOqkdMvJ3hWyOFy8cJMUvNm6nDp1zvdWxCHiUBwiDk0cJnwcdti5/tqi+79WzjxRUpw8ZQhU3wUAACAASURBVBgYPHRexIOh1ZsDw/dWpfPnrZeCN1326q2+tyIOEYfiEHFo4jDh4/CtbTnXdd3zmzIq+cuHISjDQPceUyNeC592Gh++OJwza60OvOmGj1zkeyviEHEoDhGHJg4TPg6/O7jzxrqrzC8fhqMMA59+NjbitdCv35zwxeGMaat14E3X9p2hvrciDhGH4hBxaOIw4ePw9KXif5c56brA+9tlqStPHbnNPxWRUytKS0tPnTpbUnI5tp+6d94bFvFamJC6InxxOHnSSh14qxUWnvTtFXGIOBSHiEMTh4kdh4H2O3NvzLz/tnLGrX75MFKvGQbfJoKPedXK3Nh+6pq/0S/iqbAkLS98cTghdbkIvOUjvmyjb6+IQ8ShOEQcmjhM+Dg8U1L89yum3Rh7N/3lwwj+NGmcxGG9Bt0j2wkpDb/NzSkIXxyO/T5dBN5qffrN8u0VcYg4FIeIQxOHCR+HgfnHDvzN0nF3/OXDiPw0aVzF4fnzFyPeCW+3Gxa+Mgz23cglIvBWe+31vmVlvsEiDhGH4hBxaOIw8eMw0Grbmhur79qTDyP+DjTxEIeFhScj3glffTUxlHE4fNgiEXib7dlT6Dss4hBxKA4RhyYOwxCHl8pKf5ObdmP7/Z8rpx8rLorsa4bxE4c7dh6IeCQMGbwglHE4aNB8BXibzZ6z2ndYxCHiUBwiDk0chiEOA4cuFv3nrJv88uFTG5fd6jXDMYd2Vz/M4iAO16/fFfFImDolO5Rx2K/fHAV4m3XuMsl3WMQh4lAcIg5NHIYkDgOrTx391+kTbuzAiJdhnMRh5vIfIh4Jy9O3hDIOe/WapQBvswaNehQXl/gmizhEHN5827btLoH4I35MHN7epCN7f7r0zmVY85Pu4yEO587LiWwhNGraK5RlGKz7t9MV4O2Xn7/Xs0rEIeLQzEwchicOA533/BDtMoyTOJwwKTOyefBe+5FhjcNvukyRf7ffhIkZnlUiDhGHZmbiMFRxGGi+JTtKP00aV3E4fOTCyOZB127TwhqHX3w5Uf7dfh91HO1ZJeIQcWhmJg7DFoclpaWPrF8cpdcM4ycOe/WZEdk8GDlySVjjsNNn4+Xf7VcnpevpM+c9sUQcIg7NzMRhqOIwcLy4+B9WzYpSGcZJHH7ZOTWyeTBrZk5Y4/DjT76PsxjrEod9mLVysyeWiEPEoZmZOAxbHF75Y8+e/F8yJkb2p0njKg7bfzgysm2wKmtHWOOw/YejvDZ4xw0aPNcTS8Qh4tDMTByGMA4DQw7siEYZxkkctmw9IIJh0Pz1vmEtw2DvvjdC+91xb7TqX1bmqSXiEHFoZiYOwxiHl8vK5hz9MRofczzEYYNGPSIYBh0+GhPiOHy77VDtV5nt3XfEc0vEIeLQzEwchjAOoyfmcVhcfCmyVdCr56wQx2GrtwYJv0r93unsbM8tEYeIQzMzcSgOEykOjx07HdkqGPt9eojj8PU3+gu/yuzLzqmeWyIOEYdmZuJQHCZSHBbsORzZKlgwf0OI47Bp8z7CrzJLadi9qOiip5eIQ8ShmZk4FIcJE4d5eQURPeOuW072rhDHYcMmPYVfJbcud6enl4hDxKGZmTgUhwkTh6uyt0awB1q2HhTiMgyW0rC76qvkRo5K8/QScYg4NDMTh+IwYeJwYVpuBHug02fjQ1yGuTkFdVJUX2XX9p2hnl4iDhGHf7XlmWsKCvZbPG/e3GVKwEwcJm0cjk9Nj2AP9O83N8RxuCZ7p+Sr0goLT3qGiThEHP5lwXd9n9s4t3ZtnhIwE4dJG4e9+kyPYAxMSF0R4jhcuWKb3qvSFqblepqBOEQcikNxaCYOxWFixGH7D0dGMAaWpuWFOA4z07fovSqte4+pnmYgDhGH4lAcmolDcZgYcdgocm+/mdLw29ycghDH4dLFeXqvSmvYpGdxcYlnGohDxKE4FIdm4lAcxnscnjx5NoIl8Ha7YeF+q9KFCzbovaouL6/AMw3EIeJQHIpDM3EoDuM9Djdv2RfBDPjq64nhjsN5c9eJvapuzNilnmkgDhGH4lAcmolDcRjvcbh02cYIZsCQwQvCHYezZuSIvarunfeGeaaBOEQcikNxaCYOxWG8x2Fkz7GYMmlluONw6pRssedAC8Qh4lAcikMzcZgwcdggpe3rzTskyho3fDeGcRjZcyxmTFsd7jicNDFL6VXnQItFDrRAHCIOxaE4NBOHtWv//kMff9Q9Ebdt2+6YfMYie45F6ONw/LhMpVeNde0+xZMNxCHiUByKQzNxSFyL4DkWyRCHo0ctU3rVWIPGPYqLL7ndEIeIQ3EoDs3EIXEqsudYJEMcjhyxWOlVbxs27nbHIQ4Rh+JQHJqJQ+JUZM+xSIY4HDJkocyr3kaNXuyOQxwiDsWhODQTh8SpyJ5jkQxxOGDAPJlXvb39zlB3HOIQcSgOxaGZOCROjZ+QLg6rtL59Zsu8au/QoeNuOsQhyR6Hfft8N3bsdIvndes6WAmYicMkFNlzLJIhDnv0mKHxqr1589e66RCHJHscmpmJQ+JTZM+xSIY47Nptmsar9jp3meSmQxwiDs3MxCHxKLLnWCRDHHb+epLGq/ZSGn5bVFzsvkMcIg7NzMQh8SXi51gkQxx+/kWqxqvJ1uXudOshDhGHZmbikPgS8XMskiEOO346VuDVZCO+W+TWQxwiDs3MxCHxJeLnWCRDHH748WiBV5O1bjPIrYc4RByamYlD4kvEz7FIhjh8v/1IgVfDHThwzN2HOEQcmpmJQ+JIxM+xSIY4bPfuMHVXw82Zu8bdhzhEHJqZiUPiSPsO34nDqq7124PVXQ33ZecJ7j7EIeLQzEwcEkcifo5FMsThG28OUHc1XL0G3c9fuOgGRBwiDs3MxCFxIRrnWCRDHLZpO1Td1Xw5a7e7BxGHiEMzM3FIXIjGORbJEId9+82RdjXfsOEL3IOIQ8ShmZk4JC5E4xyLZIjDRfM3Sruar2XrAe5BxCHi0MxMHBIXonGORTLEYbCWbw5UdzXf3n1H3IaIQ8ShmZk4JPaicY5FksRhr56zpF0ELpVZq9yGiEPEoZmZOCT2onGORZLE4dw5udKu5vv0s7FuQ8Qh4tDMTBwSe9E4xyJJ4jA3Z0/z1/upuxquTkrX02fOuxMRh4hDMzNxSCxF6RyLJInDYN26T1d3Nd/yFfluRsQhca1BSlszMwvWv99o3xTCasuW/eKwJps1I0fa1Xx9+s1yMyIOAQBiKUrnWCRPHObmFDRt3iexSqxJ897x9iE1bd67pOSy+xFxCAAQM1E6xyJ54jDYN12mJFYcfvbF+DopcfdRbdmy3/2IOAQAiJkonWORVHE4berqxIrDAQPnDBoyL94+qnGp6e5HxCEAQMxE6RyLpIrDdWsKGjftlUBx+N3otJMnz0bpXWqrvXffH+5+RBwCAMRM9AoheeIw2JdfTUygOJw0eXnw0E+esiLePrDCwpNuScQhAEAMRO8ci2SLw8mTshIoDufOywke/eLiS61aD4yrD2zBwnXuSsQhAEAMRO8ci2SLw5zs3Q0b9UiUOFyWsaniAgj+h7j6wLJXb3VXIg4BAGIgeudYJFscBvvs89REicOctdsrLoDS0tIPPx4VJx/VO+8PDz4edyXiEAAgBqJ3jkUSxuH4cZmJEof5m/devQby8/fGyUeVtXKzWxJxCAAQG9E7xyIJ43D1yh0pDb9NiDjcu/fItZdBt2+nxvxDavfeMC8bIg4BAGImeudYJGEcBvv4k7EJEYdHj56+9jI4dOhEvQbdY/shLV+R735EHAIAxExUT7pLwjj8fsyyhIjDoqKL110JI0ctiuHH06btkJKSy+5HxCEAQGxE9RyL5IzDVVk7Yv4S3B1Xt363srLrL4Y589bE8ENK/+d3TwVxCAAQA1E9xyI54zDYhx1GxXkcNmvR57orYf/+o/Ubxey3Jd9qO9jLhohDAIBYiuo5Fkkbh9+NXBLncdiiZb9rL4Pi4pL3PxgZw49nybKNbkbEIQBALEX1HIukjcOszG11UrrFeR8W7Dl89TL4bnRaDD+SVq0HlpSUuBkRhwAAsRTVcyySNg6Dvdd+ZJzH4YRJmRXXwIaNu1+t1yWGH0nakvXuRMQhAECMRfUci2SOw+HDFsV5HL7z/vDgAjh9+lyLlv1j+GG0bD3Ay4aIQwCA2IvqORbJHIeZ6VvqpMT7aRb79x/t2n1KbD+GBQvXuQ0RhwAAMRbtcyySOQ6DtXtnWJzHYczLsEXL/sXFl9yJiEMAgBibMnWFOIzeBg2eH+dx+GqsX9ucNz/HbYg4BACIsVOnzjVq2jPaz/5nzcxJ2jhctjQ/3uMw1sdpeNkQcQgAEHtDhtbG61qZ6ZuTNg6DtX57sAi81WbPWe02RBwCAMTY/v1H69aP+kF8KQ2/zc0pSOY47N9vrgi86Zq16FNUdNGdiDgEAIixr7+ZWAsB8FabIclchsEWLdyoA2/+y6izVrkNEYcAADGWl1dQOwHwaadxSR6HwVq2HiQFr3/ZsHmf8xe8bIg4BACIqdLSsvfaj6idBujVc5Y47N17lhq8btOmr3QnIg4BAGIsbcn6WmuA775bIg7nz8lVg9euSbNeZ88VuRMRhwAAsVRUdLFFy/61lgEzZ6wRh8Fef6O/Jry6yVNWuBMRhwAAMTZhUmZtZkCSn2Nxdd9+O10TVqxxs17nvGyIOAQAiK3jx880aNyj1jLAORZXN2tmjiys2MTJme5ExCEAQIwNGDinNjPAORZXF0Ry0+Z9lWGjpj3Pnr3gTkQcksB2Xziz9szxMK24tNTDCpBsCvYcruUScI7FtevSdao4HD8h3Z2IOCSx1c/P+smy8WHaP6yaNfvofo8sQFL5snNqLZeAcyyu3fRpq5O8DBs07nH69Dl3IuIQcRiPe3T9ks3nTnl8AZJBztrttR8DzrG4duvW7G7crHcyx+HYccvciYhDxGH87m+XpbbZnnP6UrFHGSDESkouv/P+8NqPAedYXLevO09K3pcNG/U4dcrLhohDxGHc7+9XTBt9cFeZRxogpBYsXBuTHnCOxXWbPGll0sbh6O+XuBMRh4jDhNkDOfOzTx31cAOEzPkLF5u/0a/2Y6BOSjfnWFy3nOzdtXmUSPwspeG3x4+fcTMiDhGHibSfLh3fID+r8KL3mAYIj3Gp6bH6MUI1eOM+/yI1CePwu9Fp7kTEIeIwIffvMyf33relxHEXAInv2LHTQaTFpAfe/+A7KXjjJqQuT76XDbt72RBxiDhM7P2P7NkLjh3w6AMktD79ZsYoCbo4x+KmW5O9s36j5PrJ0uEjF7kTEYeIwzDsmY3pBRfOugYAEtGOHQeCSIvZTxI6x+IW6/jp2OQpw7r1ux054twsxCHiMCz7N+kTPti5/mxJiSsBILF8+lksI8Q5Frfa92PSkycOhw5f4E5EHCIOw7b/kjV99MFdLgaARLEqe2tsq8A5FrdadtaOeg26J8nLhoWFJ92MiEPEYTj3UO7iTWdPuCQA4lxJSUmbdkNie3SBcyxusw4fjUmGOBw0ZJ6bEXGIOAzz/uWy8S22rj5aXOTCAIhbc+auiW0VvNVmiAK8zUZ9tzT0ZVgnpevBg8fdjIhDxGH49x+XXznu4nJZmcsDIN6cO1fUrHmf2IbBp53GKcDbbOXybXXrh/wnSwcMnONmRBwiDpNo9+bMW3Gy0BUCEFdGjVkc8zBwjsUd1/6D78L9suGBA8fcjIhDxGFy7W+Wjns5b/m+onOuE4B4UFh4MqVh7F+Sco7FHTd8eFqI47Bv/1luRsQh4jBJ928zJnbavfHi5cuuFoDY6tFrWjy0gXMs7rjM9C11UrqFNQ737jviZkQcIg6Tev+watbso/tdMACxsm3bjzE89d45FlXdO+8ND2UZ9uozw82IOEQc2pU9un7J5nOnXDYAtaysrPyTTt/HQxs4x6KSGzJ4QRjjsMuePd6PAHGIOLR/3t8uS22zPef0pWIXD0CtWb4iP07ywDkWlVz6kvw6KWGLwx69prkZEYeIQ7t+f79i2uiDuxx2AVALSkpK3mo7OE7ywDkWlV+btkND9rLhrt2H3I+IQ8Sh3XwP5MzPPnXUVQQQVTNmrYqjXzlzjkWl13/AvDDFYbdvp7oZEYeIQ7vdfrp0fIP8rMKLF1xLANFw+vT5xs16xU8hOMei8luclhemONyx84D7EXGIOLQ7799nTu69b0tJaakrCiCyho9cFFeF4ByLKq3VW4PCUYbfdJvsZkQcIg6tCvsf2bMXHPNfKwJEzIEDx+rWj6/j8pxjUaX17TM7HHG4ZavjrBCHiEOr+p7ZmF5w4axLC6DmunafEleF4ByLqm7BvPUhKMOvv5noZkQcIg6tmvs36RM+2Ln+bEmJCwyg2n7I3xNvkZDM51jk5uxekZEfLH3JhoXzc4LNmrZ8+pT06ZOXjRszN9jokbOGDpocrG+vsb2+HRXsqy8GBGvQqFuix+HmLfvcj4hDxKHVaP8la/rog7tcYwDVUFpa+uHHo+ItEjp+OjYeOm3Nqh3XddrcWSuDTps0Ia2i074bMaOi0wb0Gx9EWrcuwyo6rVPH3h99+G2w99/95u02n7d645PXmrUP1rhhu7p1Wgd7+Y+v/+G5JsGe+n39xx555be/eeGffvlMsHvufvSuf3yoenv08ZYJXYZffJXqfkQcIg4tMnsod/GmsydcaQBVsixjUxx2wnXnWFRE2nWFdjXSriu0GyOtotBujLTrCu1qpN1/3+PVLrQY7ue/eCGh4zA/f6/7EXGIOLSI7V8uG99i6+qjxUWuN4DKKC6+1Kr1wDjshMcef+2eux9OxEKL6R5+8eXPErQMO3b63v2IOEQcWuT3H5dfOe7iclmZqw7g9qZMXRGfqfDLX/5R7FVjjz/xZoLG4aa8Avcj4pAkkpK34idLxlut7d418zKPHy4F4k+Z/+4mPpw6da5R057xmQr33/ek0qvGgqhOxDL85NMx7kcQh0kkeCb02wdfMTOzYJ069vR9IR4MGTo/PlPhlbqd775L6VXzJ0v/+MqXCReH6zd4VzkQh+LQzEwcEiP79x+Nt1Pvr+75Fz+RedXeE0+2vsUntkt8PtwfdRztfgRxKA7NzMQhMfP1NxPj9nWkp555R+NVe7/61cu3//S2e3dYk9d6x8/DvW7dTvcjiENxaGYmDomNvLyCeP4hw8cee0PjVXt33/XIS69+davPbbMWfdes2pmTvWvE8LTmr/eL+WP94cej/A4yiENxaGYmDomN0tKy99qPiOc4/M3vGmi8muzJp9++1ed29KilV0+PDBJx4ID5DRrH8k2J1uRsc0uCOBSHZmbikNhIW7I+zt+exDkWNdyvf13npp/Y1m8Pzs0puBqHFUtfkt/+w1ExeaDbfzjSy4YgDsWhmZk4JDaKii62aNk/zuPQORY1/snSR1+u0/nGT+y0qauvK8OK5ebs6ddvbp2U2n6gV2VvdUuCOBSHZmbikNiYMDEjzsvQORYR+snSdjf8dt/om5bh1U2etLJB4x619kC/8/7w0tJStySIQ3FoZiYOiYHjx8/U5rN/51jEcA8+WO/az2qdlG4LF2y4fRwGWzR/Y/M3auldarJWbnZLgjgUh2Zm4lAcxsaAgXPi/zx051hEZPfc89jLdb65+lnt/M3kO5ZhxTKWbXm95YCoP9ApXS9dKnFLgjgUh2Zm4lAcxkDBnsPxX4bOsYjgnn723YpPaYPGPVdkbKlkHP6pDze//kbUfzF169b97koQh+LQzEwcisMY+LJzakLEoXMsIrXf/KZ+xae0X785lS/Dii1b8sNrLfpG9YGeMnWFuxLEoTg0MxOH4rC25azdnhBl6ByLSP5k6b2Pv1qvS/XiMNjC+RvqN4rSb6he+ag++2K8GxPEoTg0MxOH4rBWlZRcfuf94YkSh86xiOCefe794FOa0vDbjGWbq9GHEydkvVqvW5Qe6HoNuhcVFbs9QRyKQzMzcUjtWbBwbaKUoXMsIrvf/a5hxSd26rDqvHgYbED/udF7uDdu3O32BHEoDs3MxCG15PyFi7V2OIFzLOJt9937RN2ULvN7jy1fsGB79pZqxGFuzp6PPxkbpYd7XGq6OxTEoTg0MxOH1JKxqcsSpQydYxHx3X/PI8sHjAnKMNj5xenVe/Fw5fJtTZtH5c1pOnw82h0K4lAcmpmJQ2rDsWOnGzTqkUBx6ByLCO7n9z22vPfAijKs2J6sTdXrw2lTV9dJifzDHfyZ584VuU9BHIrDyGzqlPlZK9aamUVvnTr2EIeJq0+/mXf+Nb/63V5u0M05FiHbr372+9zBw68tw2CX0pZszNldvT7s1m1aNB7xNTnb3KcgDsVhZLZjxx6fWyCqBg8aJw4T1I4dB16t16VBox5ffj1h8KAFkyZmzZ+bG2zqlFVDhizs+OnYeg2+fSWl638dPux//37kS/HRh86xiMwJh794Mm/Yd9eVYcUOZa6tXhyuyd7ZomX/iD/iY8YudauCOBSH4hAQh0TXgEFz5s7LOX36fN6GfTd9ur9q5Y6fzZr6k2Xjg8VJHzrHouZ79NfP7hj1/U3LMFjpwkX5q3dUrw//dLJFhB/xjz7xa4cgDsWhOATEIdF0+XJpsOB/2L/v+E2f6K9bW/DkikUVZRgnfegci5rvmYdfPJQ66VZlWLETy1ZWLw6DRfydS+vW71ZUdNENC+JQHIpDQBwSXRfOF69fd5On+GvXFjy2/K/KsGJ/P2bky/W7OcciQffEb/9wIHXi7cuwYtU71iJYxtLN9Rp8G9nHfVNegVsVxKE4FIeAOEwkhw4W9uoxPFbLWrG2Gh/z9m2HKvOa4dX934OHveIciwQtw9/84cfxEypThjU51iJYr16zIvu4T5q83FdaEIfiUBwC4jCR5OfvCD74555p9scX3qj9jR83s6of8PFjZ6tUhi+sWDJ4yELnWCTiHn/wuX3jKluGfz7WYsXG6sXhqqwdjZv2juDj/vmX432lBXEoDsUhIA4TLw7XrctLiI/28uXS/E37q1SG63IKgv9Mly5TnGORcO9NumvMuCqVYQ2PtRg6dFEEH/f6jb4tKSnxxRbEoTgUh4A4FIdRdPHipROFJ3ev2RJkQGXKMFhO9u6WrQc5xyJR9rN7H7vxPMNKrtrHWgQXyWst+kXwod+2/UdfbEEcikNxCIhDcRhNZ86Up6dfeZlo4aI/ps+6YxlWbOaMnDopVXhm/0brAc6xiMnuuevhxd/2rV4ZVhxr8cOandXrw+HDIvni4YxZq3yxBXEoDsUhIA7FYdQcP16elhY0wOUFC+otmVLJMqzYV19NrPwz+46dvneORUw2sdM31S7DGh5rsXb1rmYt+kYqDrt0m+yLLYhDcSgOAXEoDqOjsLB80aLqlWGwFRlbK3liwaDBc19v1d85FrW/L5u3q2EZ1vBYiyGDF0QqDps171NW5sst4hBxKA4BcSgOI+7AgYrn/bcpw5eWL7pVGVasS9epd3xO36hJz/37jzrHovbX8Ln6l+bOi0gcVvtYi4z0zXVSInYq5qFDx329RRwiDsUhIA7FYUT9+OMdy/C1xVMvLlq8efX22zz1X7Y0v2797nf8VbH8/L3OsajlPfrrZ49MmhKRMqz2sRa5OXs++zw1gr92uHxFvq+3iEPEoTgExKE4jJyCgvL58+9YhqV/SoKitKWbbnuYwedfTLjNs/nWbQcXF5csWLjWORa1+yY0D60dOCyCZVi9Yy369J0T2feqHTV6sa+3iEPEYY2Wmjpr2dJVZmbR20cduonDhInDrVsr85ph6TVVcHrpitsEwLy5627zbH71mm3Bv3PYiIXOsajN9X67Q2TLsBrHWkxIXVGl97OtzD7p9L2ni4hDxKGZWdJNHMZJGVZsb9btfqSw5ZsDb/pU/vMvx1f8azt9Ps45FrW2Fx97+eLsOdGIw9KFi/JX76hMGS6cv6F+ox4RP+UypeG3ly+XesaIOEQcmpmJQ3FYM9u2Va8Mr/wjC9Py1uy6VQn07XeTnx6sk9J1z57Cin9zs+Z9nGNRa6cabh05OhplWPljLVau2PZ6ywERL8OK7d59yDNGxCHi0MxMHIrDGti+vdplWLHCjDW3ioElizbd+CR+2PAFFf/m48fPRKQKnGNRmX3Rol30yjDYwUkz58zKuU0ZrltT8H7776JUhsEWpeV6xog4RByamYlDcVhdu3bVsAyv/EjhgoU/rNl5qyRo03botc/gmzTrdfr0uYp/+YaNu51jUTv72b2PHZ44OXplWLZgQae2/du9Oyw359anm3SZEr0yvHJm5pB5njEiDhGHZmbiUBxWy48/Vv69SW+/woxbvmQ0YMC8a5/Bz56z+uq/P/ifnWNROxvQ7uOovmw4t+f3FY/F92NufuzhiOFpUS3DYO+1H+EZI+IQcWhmJg7FYdUdPVq+cGFEyjBYycK0DWsLbloFc2b95T1L27QdUlxccvVDGDR4rnMsamE/v++x09NnRK8Md4+fXr/Bn4+zf61F3zWrrn8ZedbMnHoNukc7DuukdC0uvuRJI+IQcWhmJg7FYVWcOlW+aFGkyrBiu1b+cIvjzgsaNulZ8fR93bqd134UH30y2jkWtbBPm7aJXhkWzZ7b9o3e1z4c/frNvfYCSF+S3+S13tEuQ+9JA+JQHJqZiUNxWHVFReXLlkW2DIOdXJZ10zg8dvTsN90mB0/cv/gq9a+/tZU3+udodI5FVJfRq3+0ftVw/oLeHw258VSJjKWbKx791St3vNVmSO2UYbCMzDxPGhGHiEMzM3EoDivn8uXyVasiXoZX/sBFN/nJ0v17jwX/zllzVtet323vviPXfiBHjpyKSA+8UrfzXf/4sAK8zQkWRdE52zDYxC6jbn6O5RcT/vSi8Z5On42vtTIMNjZ1mSeNiEPEoZmZOBSHlZOXF40yrNjWVVuuLcNtWw+Wll45l3z37kMjR6VddDBIlwAAIABJREFU94Gs37DLORa1sD888scoleHygal1Um7563+zZub06TunNssw2DfdJnvSiDhEHJqZiUNxWAn79kWvDIMd2LDjahn+sGn/pUt/fu+ZIBHPnSu67mOZNz/HORa1sMbPNYhGGWb2HvDCCx/c5nFp/nq/W6Vj9PZmm0GeNCIOEYdmZuJQHN7J+fPlaWm3KcPmNSvD8sWLTx0/W1GGG9btOXf24u0/nOEjFzrHohbW4JmUyL9m2HvgA/c++tvfNazl9qvEupw/f9HzRsQh4rA6e/mlVil125iZRW/PPNVEHMZFHJaWlmdnR7EMgxUUXLx46c9vQnPkzB0/os+/SnWORS3swZ8/Wfan0ywj+JphUIbBn3zfvU8EMRZvfbh9+wHPGxGHiMPqbMeOPT63QFQNHjROHMZFHO7aFd0yzMws/9OvF27M3VvxJjR39Ear/s6xqJ2tGTA0UmU45bMu993z6NU/+dk/fBhvcbhk6QZfeBGHiENxCIjD0MbhmZLiT3dv/H+zZ6cdr9YxbhcuXFq06OXFNy/DNxZPLat5Nhw/XvGv+nH/8Yo3obm98xcuRupFJ+dY3HFNn29YOr+mD/HF2XM+afLWdX/yQw81jrc4HP39El94EYeIQ3EIiMNwxuGpkkv/PXtWRcj9byumHrx4vqp/wqWNG6Nbhvn5Vf2Qdu48GKlzLO6+S/7deX3e7lCTh3jjkJEvP/HKjX/sffc9GW8/Wdq9x1RfeBGHiENxCIjDcMZhlz351+bc73LTLpeVVaEMT516KaplmJ5efulSVf9SGZl5zrGo5XVq9va5mbOq+vienj7jixbt7rl1gf/hhY/iKg7bd/jOF17EIeJQHALiMJxx2GrbmuuiruOujZX8Z0tKS19aPjuKZRjsxIlqfDYmTMxwjkXt77FfP5vasfP5yiXi0clT+rf76De/uMNP7T70SNO4isPGzXr5wos4RByKQ0AchjMO++zbel3X/XTp+IXHDt7xH7xcVlZ3U0a03oGmYjt3Vu+z0avPdOdYxGq/+tnvP2zYam7nHvvGTbjujUwvz5u3c/TYCZ92bvVys3vuergyf9r9Dzwdb792ePaGQzVBHCIOxSEgDsMQhwUXzv6LG+rujr98WFJa+nLe8ui+ZrhmTXkl3nvmptp/ONI5FvGwf3rgiecefrHuU3WDPfPwi9e+E2nl94cXP4mrONy//6ivvYhDxKE4BMRhCOMw8OoPN8m82/zyYaTKcN/4SY1eaLFp2Oibn11R9V81vKpR057OsQjNHnmkeVzF4Q/5nt4gDhGH4hAQhyGNw71F5/8uY8KNpffp7pv88mFQjPXyV0Tkp0l3jkkNPubs/sOv//9avLj83LlqfypOnzkfqQxwjkU87Gc/ey6u4jBr5WZfexGHiENxCIjDcMZhoOtfv2fp1V8+vO7kw8j+NOnN43DRoqunGlbPjp0HnGMRsr3wx0/jJw7nL1jray/iEHEoDgFxGNo4LC0t/V1u2k9u+8uHEf89w5vEYVCGx47V8FORtXKzcyxCtkcfez1+4nD6jFW+9iIOEYfiEBCHoY3DwOGLF/5L1vQbw++3f/rlwwj+NOkt4zASZRiYMWuVcyxCtp///Pn4icOJkzN97UUcIg7FISAOwxyHgdWnjv7r9Jv88mGHneuj8d6kfxWHS5dW70jDGw0dvsA5FqHbwy++/FmcxOGY75f42os4RByKQ0AchjwOA2MO7b5pBN50r61Pr8l5hn+Jw5Uryy9ciNSn4qvOE5xjEb499nhLcQjiEHEIIA5rLw4D7+1YV5kyfGPr6isnXWzcWMM4XDVxZvnlyxH8VLRpO8Q5FuHbL37xYpzEYerEDF97EYeIwyqvSaN333j9IzOz6O3F51uIw4jHYZB8t/r1wr/8nuGW7NKKE+qD/7e6fbhj8owrcbgyN4Kfh+CjSmnY3TkW4dvddz30x1e+iIc4nDxlhaeOiEPEoZlZUkwcBooul/zTuoV3LsOrDh8uz8ioQhmmp5fv2bNje0HE4/DYsdOROsfirn98WJLF1R7/fet4iMNFabmeOiIOEYdmZuIwWeIwcKT44n9bOePGMmy5dU3ZLV6zKy8sLF+/vjwt7ZZNGPx/5eZeKck/teWOHXsiHoeROuTQORZxuF/+08vxEIcbNu721BFxiDg0MxOHSRSHV/7Msyf/XeakO7xmeNNKPH26/Mcfg/gr37KlfOvW8u3by/fvLz91qvyv/9loxGHO2u3OsQjxe5a+9MpXMY/DwsKTnjoiDhGHZmbiMLniMDDjyL6fLr3Ta4bVfpUvCnGYtmS9cyxCvN8/2Sa2Zdim3RDPGxGHiEMzM3GYjHEYaLd9bWVfM4yDOJwydYVzLEK8X/3q1djG4djUZZ43Ig4Rh2Zm4jBJ4/BsSclHuzaUReFjjkYcDh+5yDkWoX7P0kdervN1rMqwafPep8+c97wRcYg4NDMTh0kah9ETjTjs0WuacyzCvSefbhuTMnyzzaCtW/d70og4RByamYlDcZgYcdix0/cROcfi7rtkWJzuwQfr1mYT1q3frdu3U9eu21FaWuYZI+IQcWhmJg7FYcLE4VttBzvHIty75+5HX6nzTbSbsHGzXr36TM/IzDt79oIniiAOxaGZmTgUh4kXh02a9XKORegXPEBRCMIubdoN6dVnxtx5Obt3H/I6IYhDcWhmZuIwseOwbv1uzrFIgp8sTalhB77Rqv+nn40dMHDOpCnLMzLztm7df/78RU8IQRwiDs3MxGFI4rCouNg5Fsnxk6WPPfrY64890eqJ37d+8qm3n3q6XcU6fT66V5/pvfvOHDJ8frAxY5eOTV02dVrWwrTcVdlb837YU7Dn8LFjp4uLSzz3A3FIbcfhiOETZ85Ms3he/36jlYCZOAxNHJ46dc45Fsm8KZNne1IH4pA4jcPgu77PbZxbuzZPCZiJw9DE4bFjp51jkczr1WuI7+wgDhGHiEMzcSgOy48ePe0ci2TeO+0+9Z0dxCHiEHFoJg7FYfmRI6ecY5HMe+mPTX1nB3GIOEQcmolDcVh+4sQZ51gk8372wBPBUxrf3EEcIg4Rh2biMOnfrbSo2DkWSb7CwqO+uYM4RBwiDs3EoXMOI3DO4W9/11BiJe5ycjb45g7iEHGIODQTh+KwvO07Q51jkcybNm2ub+4gDhGHiEMzcXjlgx/Qb3SirPNX/SMeh917TK3xORZPSazE3YD+I31zB3GIOEQcmiV1HG7btvu5Z5ol4oIvRBH8PCxZusE5Fsm8Tp92880dxCHiEHFoltRxSIXzFy6+1XawcyySdi3faO8uAHGIOEQcmolD/tSH5y/OmrO6qonYpt2QHr0m/eyBZ/WVow4BcSgOxaE4NDNxGCq7dh+aMi2rS7fJrd4aWCflr1Kwbv1ub7Tq/+lnY0eOSluWsamw8GTwn9+790dxleh78MHnXPkgDhGHiEMzccgtlZSUnDtXFOzIkVOnTp276X+mqKhIXIVgRUUXXfAgDhGHiEMzcUiNPPhrP1aa8Nu//4ArGcQhcReHC+ZnbNq01eJ5UybPUwJm4pCrXnyhibhK9K1bt9GVDOKQuItDMzNxSGJ5vcV74irRN3/+ElcyiEPEoZmZOKRGPvmki7hK9I0ePdGVDOIQcWhmJg6pkb59h4urRN+33Qe6kkEcIg7NzMQhNZI6fpq4SvS1f/8LVzKIQ8ShmZk4pEaWLM4UV4m+Jo3buJJBHCIOzczEITWyadNmcZXoe/qpeq5kEIeIQzMzcUiNHDpUGOJqevml1375i6dCH4cP3P+EKxnEIeLQzEwcUiMlJSV33/VwWKup5Rvt3377k2R48fDkydMuZhCHiEMzM3FIjTz80B/DmkwffPDltGlzkyEOt2/f5UoGcYg4NDMTh9RInVdfD2syffVVr2PHTtx9V/jjcMWK1a5kEIeIQzMzcUiNvNW6Q1iTqV+/4cFfsH79VqGPwxnT57mSQRwiDs3MxCE18uUXPcKaTKNHTwz+gsOGjQ19HI4YMc6VDOIQcWhmJg6pkUGDRoU1maZPu/J62o4du0Mfh9269XclgzhEHJqZiUNqZOrUOWFNpqVLllf8HZ9+OiXccfhB+y9cySAOEYdmZuKQGsnMzA5rMuXkbKj4O3bt2i/ccfhas3auZBCHiEMzM3FIjWzdsiOsybRt286Kv2N29rpwx+Hzf2jkSgZxiDg0MxOH1MjJkyfDmkyHDhVW/B1LSkp+/atnQxyHD/76WVcyiEPEoZmZOKSm3xMfuP/xUCZTUVHR1b/mB+2/CPeLh8XFJS5mEIdU06lTp83MLNj580W+KSS53z9R5/9v786Do677Bd/Pnao5M3Xm3qo7NTN/nTpTdWuqbtX95woouKICAQREBCKCiggY2R4UjYIiCqggiyxuCAgi8EQQBGWRTXYNQaMIskRWwYAsUTAEDTxtciZnOOXxCEqS7k5/f7+8XvX55znlEfvz+4bud3W6f/GLpasbZf32MX744dp4x+Gvb5MC4hAAoJa6dx8Qv1hqdmvH3z7GsrJzjRpmxTgOd+0scpJBHAIAJCX3seHxi6UOl3xHy4O9H41xHG7atMVJBnEIAJCUsWNieKeH7t0H/O5hzp2zMMZxuGjRcicZxCEAQFJmzsyLXywN6P/k7x5mcfGxGMfh9GlznGQQhwAASVm2bHX8YumpJ0dd+kg73vlAXOPwxdEvO8kgDgEAkhLLe8SPefGVSx/py5OnxzUOH88d4SSDOAQASMrBg4fjF0tTXp916SPd/uWuuMZhzwcedpJBHAIAJKWs7Fz8Ymnu3IWXPtKKiormzTrFMg7vaN/dSQZxCACQrOuvaxuzWFqyZNVlH+mzz4yNZRzecH07xxjEIQBAsu5o3z1msbRhQ/5lH+m6tZvj+pul5eXlTjKIQwCApPTuNShmpbRt21eXfaRVBdWkcetYxuHRo985ySAOAQCS8tSTo2JWSgcOfPNHD3ZA/ydjGYc7dux2kkEcAgAkZeLEqTErpZMnS/7owS5YsCSWcbh+3cdOMohDAICkzJ27MHYfwDv/Rw+2qhtjGYfvLVzmJIM4BABIyprVG+OUSY2vaf3nj7dr1z7xi8Pp0+c6ySAOAQCSsm3bV3HKpBbNO//5433jjbfjF4djx77qJIM4BABIytGj38Upkzp17Pnnj3fPnr3xi8Mnn3zBSQZxCACQlAsXEnHKpAd6DLziQ76tddeYxWGfhx53kkEcAgAk6+amHWKTSQ8PHHrFx/vCC5NiFod3ZT/oGIM4BABIVnbnXrHJpKeffvGKj/fjj7fGLA6zWmQ7xiAOAQCS1a/v4Nhk0rhxr13x8Z4/f+H669rGKQ6vbpRVWVnpJIM4BABIyvBnx8Umk6ZOnV2dh/zYo8/G7M3DsrJzTjKIQwCApLz26szYNNI7eYur85CXfLAyZnF45HCxkwziEAAgKe++uyQ2jbR82ZrqPOTTp39s2KBFnOJw+5e7nGQQhwAASVm//pPYNNLmTQXVfNQ97h8YpzisuohOMohDAICk7NpZFJ830LZX9w20t2a+E6c4fO+9ZU4yiEMAgKSUlJTEppEOHTpSzUd98ODhOMXhm2/+1UkGcQgAkJSKiorYfADv++9PV/+Bt7/9vnp1Dw9AHAIAXEFWi+x4NNKFC4nqP+qXxk+JTRw+9eQoxxjEIQBAsrp17RODQLru2jY1etSFn30Zmzjs2+cJxxjEITW2rKR4zOHdsZmxh3cf/KnMZQUgGY88MiwGgdSqZZcaPeqKioqmTe+IRxx2uSvHMQZxSI3dtzv/3214J07zdxvnPbyv8GzigosLQO288MKkGARSdudeNX3gTz05Kh5x2DLrLscYxCHi8F/mH/IXz/7uYKULDEDNTZ82JwaB1KvnoJo+8FWr1scjDq+5uqVjDOIQcfhv5uZtH20/e9pVBqBG3l+8IgaBNOiRYTV94GVl565ulBWPPqx6LE4yiEPE4b+Zf7/+ne67809dOO9aA1BNn3zyaQzq6NlnxtbisT/0UG484vDIkaNOMohDxOFl5r9sXjj526JfKv2eKQBXtnfvgRjU0YSXptTiseflLYpHHG7fvstJBnGIOPzDafDZik2nT7joAPy5M2dKY1BH06fPrcVjP37sZDzicMOGfCcZxCHi8Apz+46Nh392uwsA/kzja1pHvY7mz/+gdo89u3OvGMThokXLHWMQh4jDK8/fb5o//NCO8l8SDgAAl9W2Tbeo19GKFWtr99hffXVGDOJwxow8xxjEIeKwuvOP+e/P/u6gMwDApXrcPzDqdfTJJ5/W7rF/9dWeGMTh+HGvO8YgDhGHNZvmX67dVXbGSQDgt554fGTU62jnzqLaPfbKysqsFtlRf/hDnxrlGIM4RBzWeP7DhnkP7yv8MfE35wGAi8aNey3y93I4XFzrhz9ixPioP/x+fQc7xiAOEYe1nP/68XuTvy2qqKhwKgCYNWt+1Ovo9Okfa/3wN2zIj/rD73p3H8cYxCHiMKm5unBl/plTDgZAPffh8o+iXkeJRO2/d628vLxJ42h/X2urll0cYxCHiMNk5/9Yn9d9d/6J8z87HgD11mefbYt0Gt1wfbskN/DwwKGR3kDja1o5xiAOEYepmf+86d3hh3ac/+UXhwSgHjp8+NtIp9FtrbsmuYH33lsW9fdOz537yUkGcYg4TNn8v1uXflhy1DkBqG/Ky8sj3UVd7spJcgMlJT80uCracVhcfMxJBnGIOEzx3L5j48GfypwWgHrlhuvbRbeLch58LPkN3HtPv0jH4Y4dux1jEIeIw9TP323859tdnE1ccGYA6okOHXpEt4tyHxue/AamT5sT6TjcuDHfMQZxiDhM1/xD/uLZ3x2sdG4A6oGcnNzodtGIEeOT38DXXx+IdBy+v3iFYwziEHGY3rl520fbz552eADi7emho6PbRRMnTk3JEtrc1i26S5g5M88xBnGIOEz7/Pv173TfnX/qwnlHCCCuJk+aFt0umjEjNV304uiXo7uEl8ZPcYxBHCIO62j+y+aFk78t+qXS75kCxFBe3qLodtHChctSsoQt+YXRXcLTQ0c7xiAOEYd1Og0+W7Hp9AlnCSBm1qzZGN0uWrN6Y0qWkEgkbrzh9oguoX+/wY4xiEPEYWZud3H4Z7e7AIiP7V/uim4cFhQUpmoPj+eOiOgSunbt4xiDOEQcZmb+ftP84Yd2lP+ScK4AYuD4sZPRjcPdu/emag/Llq2O6BJateriGIM4RBxmcv4x//3Z3x10tACiLpFINLgqqnFYXHwsVXs4c6a0YYMWUVxC42taO8YgDhGHmZ/mX67dVXbGAQOItFtvuTOicVhaejaFe+jZ85GI7uGnn352jEEcIg4zP/9hw7yH9xX+mPibYwYQUXdlPxjFImpwVbOKiooU7uHtt+d7BxXEIeLQJDv/9eP3Jn9blNonaQDqxoD+T0axiJre1D61ezhy5GhE4/Crr/Y4xiAOqa47XnntH1583qR7rp0wYdzsBQsXfGiMCXa2FnzpSYHfGTFifBSLqF3be1K+ig533B/FVWzatMUxBnFItVRWVt7StIsxxpiqGf7MRM8L/M6U12dF8hYOd6f+Fg4TJ06N4io+eH+FYwziEHFojDHikGQtXLgsikXU56HHU76KL77YEcVVzJo13zEGcYg4NMYYcUiyNm7Mj2IRPfH4yJSvoqKi4uamHSK3ismTpjnGIA4Rh8YYIw5J1p49e6MYh88/NyEd23j66Rcjt4oRI8Y7xiAOEYfGGCMOSVZJyQ9RjMOXJ09PxzbWrNkYuVUMemSYYwziEHFojDHikGRVVFQ0apgVvQ/avTUvHds4d+6nxte0itYqej7wsGMM4hBxaIwx4pAUaNWyS+TicNGi5WnaRr++g6O1ijs79HCGQRwiDo0xRhySAvfd2z9ycfjRR5vStI35896P1iqa3drRGQZxiDg0xhhxSAo8OuiZyMXhp59uS9M2Tpw4Fa1VNGzQourVjmMM4hBxaIwx4pBkjR49OXJxWFS0L30L6XJXTrS2UVp61jEGcYg4NMYYcUiy3nzzr5GLw+PHTqZvIVNenxWtbRw5ctQxBnGIODTGGHFIspZ8sDJycVhWdi59C9m9O2L3ftyxY7djDOIQcWiMMeKQZG3JL/Qpu9+9eGiZdVeEFrJp0xbHGMQh4tAYY8QhyTpw4JtoxeEtN9+Z7p08/9yECC1kyZJVjjGIQ8ShMcaIQ5JVWno2WnHY/vb70r2TzZsKIrSQ2bPfdYxBHCIOjTFGHJIC1za5LUItdO89/dK9kPPnL1x/XduoLOSVl990hkEcIg6NMUYckgLt2t0boTjs13dwHexk0CPDorKQ50ZOcIZBHJKxOOzUsU+X7P7GGJO+adO6hzikzvTs+UiE4nDIkOfrYCfvL14RlYU89uizzjCIQzIWh/v3H7ZbIK2mvpEnDqkzQwY/F6E4HDVqch3s5PvvTze4qnkkFtKr5yBnGMQh4hAQh+KQFHhp/JQIxeGrr86om7V07z4gEgvp1LGnMwziEHEIiENxSArMnv2uL+e81IwZeZFYSPNmnZ1hEIeIQ0AcisOo+uTMyY9+OB7If8yKFWsjFIfvL15RN2vZv/9QJBbSqGFW1QseP1MgDhGHgDgUh9GTf+bU/7VpwX/cOG95ydEQ/ns+/3xHhOJw/bqP62wzt7e7LxI7KSs758cKxCHiEBCH4jCSZfjvNrxTNYH04ZEjRyMUh4WF2+tsM2PHvhqJnRQXH/OTBeIQcQiIQ3EY1TIMpw/Ly8sjFId79x6os81s3fpFJHayc2eRHy4Qh4hDQBzGPA4TicSPP5ZmasrLL6TwsXxy5uTvyvDi/GP+++W/JDK755tubB+VODxx4lRdHr9IbObjj7f6axPEIeIQEIcxj8Pdu/en42/yas70ae+k6oFc+p7hxfnvH7+34+zpjO+5U8eeUYnD8vLyutxMJG4CuWzZan9tgjhEHALisF7E4euvzVm2dG3dz96vD9aHMqzS56HHI1GGVzfKquPNfPhhBL7Kde7chf7aBHGIOATEYb2Iwy++2Bndqxl+GVYZNmxMRO7p16mON1NWdq6qSANfy2uvzvTXJohDxCEgDsWhMkyBV16ZEYk47NChR90v58Hejwa+lhee93VTIA4Rh4A4FIfKMBXmvbM4EnHYvfuAul/O3DkLA1/L47kj/LUJ4hBxCIhDcagMU2Dt2s2RiMMB/Z+s++UUFx8LfC0P9n7UX5sgDhGHgDgUh8owBb76ak8k4nDoU6Mysp+Odz4Q8lqys3v7axPEIZmJwwd7Dxk44FljjEnfZHfqKw6jG4eRK8Mqx4+fjEQcjnnxlYzs5+XJ00NeS1aLbK/6QBySmTg0xpgojjhUhn8ikUg0uKp5+HE45fVZGdnP9i93hbyWxte08qoPxCHi0BhjxKEyTI3mzTq5od8fqaioCHw/P/30sxd+IA4Rh8YYIw6VYQp0vbtP+HG4ZMmqTO3n2WfGhryZY8eOe+EH4hBxaIwx4lAZpsDAvzwVfhxu3Jifqf2sC/sLXXfv3uuFH4hDxKExxohDZZgCz42cEH4cbtv2Vab2U15e3qRx62A3k5//mRd+IA4Rh8YYIw6VYQq88cbb4cfhgQPfZHBFA/o/GexmPvxwrRd+IA4Rh8YYIw6VYQq8996y8OOwpKQkgytasGBJsJvJy1vkhR+IQ8ShMcaIQ2WYAps3bw0/DsvLz2dwRSdPlrjJB4hDxKExxohDcRjnMqzyddH+wMuw8TWtM76lrl0D/U7X0aMne+EH4hBxaIwx4lAZpsAPP5wOPA5bNO+c8S0F+8nMwU8854UfiEPEoTHGiENlmJonzWuubhlyHHbq2DPjWyoq2hfmch56KNcLPxCHiENjjBGHyjA1bmvdNeQ4fKDHQFv6o+lyV44XfiAOyUAcTn0j772FK4wJbcSPEYfKMEnduw8IOQ4fHjg0hC2NGjU5wOW0atXFCz8Qh2QgDvfvP2y3BEj8GHGoDJOU+9jwkONw2LAxIWzpk08+DXA5TRq39jwI4hBxCOLQiENlmBpjXnwl5DgcN+61ELZ0/vyF669rG+R9Pso9FYI4RByCODTiUBmmwMyZeSHH4dSpswNZVJhvsR4/ftJTIYhDxCGIQyMOL68skVhw6si5xN9q/R/wyZmTly3D/xa7MqyydOnqkOPwnbzFgSxqyZJVAe6nqGifp0IQh4hDEIdGHF7G9GP7//Omd6tC7v49W5RhdWzd+kXIcbh82ZpAFnX69I8NG7QIbT8FBZ97KgRxiDgEcWjE4e9tOH3itzn31ncHavpvqD+/TfqrgwcPhxyHmzdvDWdXD/QYGNp+Vq5c56kQxCHiEMShEYe/N2jf578tuv+0cf72mhRdPSzDKmVl50KOw+3bd4Wzq1lvzQttP/PeWeypEMQh4hDEoRGHv/fE/m2/67r/b+uyan74sL79Nulvhfk9nBfn0KEj4SwqwHdZ33jjbU+FIA4RhyAOjTj8vaUlxZfWXXU+fFify7DKHe27BxuH338f1v7b335fUPsZ8+IrngpBHCIOQRwacfh7v1RW/j9bllzaeH/+4cOU/DbpuXM/bVhf8MP3Z6K48N69BgUbhxcuJILa1UvjpwS1nyFDnvdUCOIQcQji0IjDy3jz2P5LM+/vN83fWXb5bEvVe4ZVTxNV/80FW7ZFceFPPvlCmGV43bVtQttV4WdfBrWivn2e8FQI4hBxCOLQiMPLqKioaPL5qktj77IfPkzhb5NGOg4nTHgjzDhs1bJLaLtKJBJNm94Rzoq63t3HUyGIQ8QhiEMjDv+g034q/T//960Ofzfddn3y238std9NGuk4nDtnYZhxmJ3dO8B1PfXkqHBW1LxZJ0+FIA4RhyAOjTj8QzMu98ulv/3wYcq/gSbScbhs2eow47Bnz0cCXNeqVevDWVGDq4L7WCaIQ+Ifh2s/+qRozwFjQhsL0+2qAAAgAElEQVTxY8ThH+m/97NL2+/inQ/TcT/DSMdhULXz2xn0yLAA13Xu3E/XXN0ynC0VFx/z8g/EIXUah8YYIw6jFYd/q6xo+sWaSwvwfxYsScddKyIdh+vWbg4zDp99ZmyYG3voodxwtlRYuN3LPxCHiENjjBGHf+bE+Z//x5b3L/v7pSm/n2Gk43DzpoIw43D8uNfD3NiMGXnhbOnD5R95+QfiEHFojDHi8Er/qrIz//fmhX9ehv89FXe6j3QcFhQUhhmH06bOCXNja1ZvDGdLM2fmefkH4hBxaIwx4vDKNpw+8Xcb56W1DKMeh4WF28OMw7y8RaEe0b3hbGnUqMle/oE4RBwaY4w4rJaZxw6k6bdJ4xGH27fvCjMOly5dHebGSkvPhrOlhwcO9fIPxCHi0BhjxGF15e7/Ik3vGcYgDoN6H+y3s2FDfrBLu+nG9oFs6e4uOV7+gThEHBpjjDisroqKitt3bExTGUY9DvftOxhmHH7++Y5gl9b17j6BbOmWm+/08g/EIeLQGGPEYQ38mPjb/yxYktrfJo1HHH7zzbdhxuHevQeCXVpu7vBwFlVeXu4VIIhDxKExxojDGth4+kQ6yjDqcXj06HdhxuHxYyeDXdqkSdPCWdSRw8VeAYI4RBwaY4w4rJni8nPp+NdGOg5LSkrCjMOysnPBLm3hwmXhLOrTT7d5BQjiEHFojDHiMAiRjsPTp38MsAwbXNW86jk92KUFdXPIYL/WFcQh4tAYY8ShOIySsrJzAcZh05vah7y04uJj4exq+vS5XgGCOKTu4rDZLV2b32pMcCN+jDgUh8k7f/5CgHHY5rZuIS8tkUg0apgVyK6ef26CV4AgDqm7OKx61rdbAiR+jDgUhyl56gwwDrvcFfrt+9q26RbIrv4y4ClPiCAOEYeIQ/1jxKE4TIFw3gT7dXr3GhT40nJycgPZVXZ2b0+IIA4Rh4hD/WPEoThMgWub3BZaHA56ZFjgSxs5coLPZ4I4RByCODTiUBzGKg5vurF9aHH4zLAxgS9txoy8cNZ17txPnhNBHCIOEYfGiENxmKxmt3YMLQ7Hjn018KWtWb0xnHUdOnTEcyKIQ8Qh4tAYcSgOk9WqZZfQ4nDK67OCP6h7w1lXQcHnnhNBHCIOEYfGiENxmKx2be8JLQ7nzlkY+NJKS8+Gs64P3l/hORHEIeIQcWiMOBSHyerQoUdocRiJ2gnns5pTp872nAjiEHGIODRGHIrDZN2V/WBocbhu7ebw99b17j6BrGvEiPGeE0EcIg4Rh8aIQ3GYrHu69Q0tDj/7LALLzM0dHsi6+vcb7DkRxCHiEHFoTDzjsCq0fv65PCqze9e+SMdhj/sHhhaHe/bsDX9vkyZNC2RdnTr29JwI4hBxiDg0Jp5xGMWJbhw+2PvR0OKwuPhY+HtbuHBZIOu64fp2nhNBHCIOEYfGxC0Of/jhzKL3VkZxjh8viehfJv36Dg4tDktLz4a/t4KCwnA2VlZ2ztMiiEPEIeLQmFjFIXXv4YFDQ4vDioqK8PdWXHwsnI0dOPCNkwziEHGIODRGHJKU3MeGB1WG11/XNhJ7SyQSjRpmBbK0/PzPnGQQh9RFHD7x+KgRwycZE9qIHyMOSYkhQ54PKg5bteoSldW1bdMtkKUtWrTcSQZxSF3EoTHGiENibNiwMUHFYXbnXlFZXU5ObiBLm/L6LCcZxCHi0BhjxCFJGTlyQlBx2POBh62upjP82XFOMohDxKExxohDkjJ69OSg4vDhgUOjsroZM/ICWVqfhx53kkEcIg6NMUYckpSXxk8JKg6HPjUqKqtbs3pjIEvr0KGHkwziEHFojDHikKS8PHl6UHH44uiXo7K63bv3BrK0665t4ySDOEQcGmOMOCQpU16fFVQcvvbqzKisrrT0bDh7q/qPcZhBHCIOjTFGHFJ706fNCSoO3357foS2d9ON7QPZ2759Bx1mEIeIQ2OMEYfU3qy35gUVh9G6ZV/Xu/sEsrfNm7c6zCAOEYfGGCMOqb25cxcGFYerV2+I0PZyc4cHsreFC5c5zCAOEYfGGCMOqb358z8IKg4LCgojtL1Jk6b5rCaIQ8ShMcaIQ+Jg0aLlQcXhrp1FEdreggVLAtnbsGFjHGYQh4hDY4wRh9Te0qWrg4rDI4eLI7S9goLCQPaW8+BjDjOIQ8ShMcaIQ2pv1ar1QcXhmTNnIrS94uJjgeyt/e33OcwgDhGHxhgjDqm9dWs3BxWHFy4kIrS9RCLRqGFWCHtr0ri1wwziEHFojDHikNrbvKkgnDK8tsltkVtg2zbdAtne6dM/Os8gDhGHxhgjDqmlcD41VzVZLbIjt8CcnNxAtvd10X7nGcQh4tAYY8QhtVRYuD2cOOx45wORW+DIkRMC2d7GjfnOM4hD/tWI4ZOMMcZUzcIFH3pSoDq2b98VThx27z4gcgucMSMvkO3Nn/+B8wziEACglnbv3htOHPbvPyRyCwzn615feflN5xnEIQBALe3bdzCcOBwy5Hl1XesZ+tQo5xnEIQBALX3zzbfhxOELL0yK3AJLS88Gsr1ePQc5zyAOAQBq6ejR78KJw4j+YuRNN7YPYXvt2t7jPIM4BACopZMnS8KJw7dmvhPFHXa9u08I27vm6paVlZWONIhDAIDaOH36x3DicMGCJVHcYW7u8EAW+MMPpx1pEIcAALVRVnYunDhcuXJdFHc4adK0QBa4e/deRxrEIQBAbZw/fyGcOPzkk0+juMMFC5YEssD16z52pEEcAgDURmVlZThxuGPH7ijusKCgMJAFvpO32JEGcQgAUEuNGmYF0jaHDh2J4gKLi48FssBJk6Y5zyAOAQBq6domtwXSNqdOfR/FBSYSiUACe8iQ551nEIcAALUUyG36qqa8/HxEd9i2TbcQFvhAj4HOM4hDAIBaanZrx0Bu0xfdHebk5Iawwza3dXOeQRwCANRSq5ZdQgib5s06RXeHI0dOCGGHVzfKqqiocKRBHAIA1Ea7tveEEDZ3tO8e3R3OmJHnc5sgDgEAoq1Dhx4hVM299/SL7g5XrVofSBzu3FnkSIM4BACojezs3iFUTd8+T0R3h7t37w0kDj/6aJMjDeIQAKA27unWN4SqeeLxkdHdYWnp2UDicO7chY40iEMAgNrocf/AEKrmuZETIr3GQO4IMuGlKY40iEMAgNp4sPejIVTNxIlTI73Grnf38QYsiEMAgAjr13dwCFUzffrcSK8xN3d4CGvs3n2AIw3iEACgNh4eODSEqpk/7/1Ir3HSpGkhrLFVqy6ONIhDAIDayH0siLe8Plz+UaTXuGDBkhDW2LBBi4qKCqcaxCEAQI0NGfJ8CFWzeVNBpNe4Jb8wkC8sPXHilFMN4hAAoMaGDRsTQtJ8uW1npNdYXHwskDjcvn2XUw3iEACgxkaOnBBC0uzffyjSa0wkEo0aZoWwyTWrNzrVIA4BAGps9OjJfhkyJdq26RbCJufMXuBUgzgEAKix8eNeDyFpfvrp56hvMicnN4RNjhv3mlMN4hAAoMZenjw94z3TqGFWDDYZyC/o5j423KkGcQgAUGNTXp+V8Z65uWmHGGxyxoy8EOLwvnv7O9UgDgEAamz6tDkZ75l27e6NwSZXrVofQhxmtch2qkEcAgDU2Ky35mW8Z7p27RODTe7aWRRCHDa4qnkikXCwQRwCANTM3LkLM94zOTm5MdhkaenZQG51ePzYSQcbxCEAQM3Mn/+B71BJlZtubB9CHG7b9pWDDeIQAKBmFi1anvGYGf7suHgss+vdfUKIw5Ur1znYIA4BAGpm6dLVGY+Zl8ZPiccyc3OHhxCHs2bNd7BBHAIA1MzKlesyHjNTp86OxzInTZoWQhyOefEVBxvEIQBAzaxbuznjMZOXtygey1ywYEkIcTjokWEONohDAICa2bypIOMxs3Tp6ngsc0t+YQhx2C0WtwYBcQgAUKcKCjLfMxs25MdjmcXFx0KIw+bNOjnYIA4BAGqmsHB7xmOm6r8hHstMJBKNGmaF0Ifnz19wtkEcAgDUwPbtuzJeMl9/fSA2+2zbplsIcVhcfMzZBnEIAFADu3fvzXjJHD92Mjb7zMnJDSEOY/NmLIhDAIA6sm/fwYyXTFnZudjsc+TICSHE4YfLP3K2QRwCANTAN998m9mMaXBVs8rKytjsc8aMvBDicObMPGcbxCEAQA0cPfpdZjPmphvbx2mfq1atDyEOR42a7GyDOAQAqIGTJ0symzFtbusWp33u2lkUQhw+/PDTzjaIQwCAGjhz5kxmM6bLXTlx2mdp6dkQ4vDuLjnONohDAIAaKCs7l9mM6d1rUMxWetON7TMeh7fcfKezDeIQAKAGzp+/kNmMeeSRYTFbade7+4Tw5mF5ebnjDeIQAKC6KisrM9sww4aNidlKc3OHhxCHRw4XO94gDgEAaqBRw6wMNszYsa/GbJ+TJk0LIQ4/+2ybsw3iEACgBq5tclsGG2bK67Nits8FC5aEEIdLl652tkEcAgDUQGa/QGXunIUx2+eW/MIQ4nD69LnONohDAIAaaHZrxww2zAfvr4jZPo8cORpCHL7w/ERnG8QhAEANtGrZJYMNs3bt5pjtM5FIZPZjnBfnLwOecrZBHAIA1EC7tvdksGE+/TSG35vStk23jMdhdnZvZxvEIQBADXTo0CODDbNnz974rTQnJzfjcdj0pvbONohDAIAayM7uncGGKS4+Fr+Vjhw5IYSPHZ4795PjDeIQAKC6unXtk8GAOXOmNH4rnTEjL4Q4PHToiOMN4hAAoLp63D8wgwFTUVERv5WuWrU+hDgsKPjc8QZxCABQXQ/2fjRT9XL9dW1judJdO4tCiMP43SYExCEAQBr16zs4U/XSqlWXWK70zJnSEOJw6tTZjjeIQwCA6np44NBM1UvnTr3iutWbbmyf8TgcOXKC4w3iEACgunIfG56penmgx8C4brXr3X0yHof9+w12vEEcAgBU15Ahz2eqXgb+5anYJnfu8IzHYaeOPR1vEIcAANU1bNiYTNXL0KdGxXWrkyZNy3gc3nB9O8cbxCEAQHVl8I7tL45+Oa5bXbBgSQjfSVNWds4JB3EIAFAto0dPzlS6vPrqjLhudUt+YQhxePDAN044iEMAgGoZP+71TKXLrFnz47rVI0eOhhCH+fmfOeEgDgEAqmVy5j4d9957y+K61UQi0ahhVsbjcNGi5U44iEMAgGqZ8vqsTKXL6tUbYrzYtm26ZTwOqy6uEw7iEACgWqZPm5OpdCkoKIzxYnNycjMeh8OfHeeEgzgEAKiWWW/Ny1S67NpZFOPFZvBrYH+dPg897oSDOAQAqJa5cxdmKl0OH/42xoudMSMv43HYoUMPJxzEIQBAtcyf/0Gm0uWHH07HeLGrVq3PeBxed20bJxzEIQBAtSxatDxT6XLhQiLGi921syiEu1mUlp51yEEcAgBc2dKlqzMSLdc2uS3eiz1zpjSEONy376BDDuIQAODKVq5cl5FoadG8c+x3e9ON7TMeh5s3b3XIQRwCAFzZurWbky+QGxq3vvm6Nu2adezQonO3tt16duj+UOeej93TZ3D3/s899OiL/XInP/LUtCeefWvIiAUjxiya+dc1qzfG+z4WF3W9u0/G43DhwmUOOYhDAIAr27btq6qGeaDHwJyc3EcHPfPE4yNHjpwwevTkSZOmTZ06e+bzL1Xl3OLnx60Z9/LGia9tfXXqjulv7Z7xdnHe/OPzFpS+/8H5Zcv/adWqms3hw/Vkt7m5wzMeh6+9OtMhB3EIAJC0wsIat98V58CBerK8qsDOeBwOGzbGKQZxCACQtE8+SX0cfv11PVneggVLMh6HOQ8+5hSDOAQASNratamPw50768nytuQXZjwO299+n1MM4hAAIDmJROrLsGq+/LKe7O/IkaMZj8MmjVs7yCAOAQCSU1aWljj87LN6E9eJhg1aZLwPz5wpdZZBHAIAJOH48bTE4ZYt9WeFbW7rlvE4/Lpov7MM4hAAIAn79qUlDjdtqj8rzMnJrcsOvLpRVts23ar+0JEjJ8ydu7CgoLC4+FhFRYWzDOIQACAJ27alJQ7Xras/K6yKtHR1YMMWOhDEIQBAndi4MS1xWDX1xowZecl34DUNW7Rr1vGhzj2fe+jRvw57YeurU4vz5v/y1U4nFMQhAED6lZenqwyrJpGoJ1tctWp98h1YsfJyO/zqK4cUxCEAQPql6dtoLs7PP9eTLe7aWZSCDqzfdwQBcQgAkFF79qQxDs+erSdbPHOm9PffEzNrXo078LLzxRcOKYhDAID0+/jjNMbh6dP1Z5HHj5+srKz81//900+p2WFhoUMK4hAAIM1+/vmfVq5MYxyePFl/d1tWlpodfvqpcwriEAAgzb79No1lWDXHjtXf3ZaWpmaHW7Y4pyAOAQDS7PPP0xuHhw/X392ePp2aHX7yiXMK4hAAIJ0SiX9avTq9cXjgQP1d7/ffp2aHmzY5qiAOAQDS6dix9JZh1Xz9df1d76lTqdnhhg2OKohDAIB0SvfvlFbNzp31d70nTqRmh+vWOaogDgEA0ubnn9Oahf9yc7/6fAP3775LzTLXrHFaQRwCAKTN/v1pjcOVYyY+fl/fMxvq8efljh5N2T4BcQgAkC6bNqU1Dl97dOhV/3+zljd3yM//rJ5uOIW3CamsdGBBHAIApEFJSbo/bTjonj5VcVg1Da5qPu+dxfVxyYcPp2yff/ubMwviEAAgDbZtS/cHDlvc2P5iHF6cSZOmVda3t78OHUrZSs+fd2ZBHAIApFpZWbrfNtz62rTfluHFee3VmfVrzwcOpGylP/3k2II4BABIgYqKii+37Zw/7/157yzePGf+uSVL0xqHIx8cdGkcVs3ixR/Wo6Xv25eylVb1PCAOAQCSkUgk5sxecFvrrr+NtKsbZvW68/7ZQ587teC9lJfhj4s/uLHJbZeNw0YNs7Z/uau+rP7rr1O21dJSJxnEIQBA7X311Z47O/S4bKddnIZXNe+X3WvLK2+kMA4nDhzyJ39ihzvuLy+vH5+g27MnZVs9fdphBnEIAFBL7+QtvrpR1p902m+nW7t7tk+dmXzGHJ+3oMnVLf/8z5o4cWq9uAA7d6YsDr//3nkGcQgAUGOVlZVjx75azSz8dRpc1XzCX4b8smJFrRumcuXKAV16X/EPatigRXHxsfhfhh07UhaHp0451SAOAQBqXIbPDBtT0zL8dR6/r1/Fylo2zOyhz1XzTxk5ckL8r8SXX6YsDo8fd7BBHAIA1My4ca/VugwvztIXxtciYLZPnXlNwxbV/COubpR1/PjJmF+JL75IWRweO+ZggzgEAKiB6dPmJFmG//yZwIFDalove9+a0/TaNjX6U6ZNnRPzi1FYmLI4/PZbZxvEIQBAdS1cuCz5Mqya9S+9UqN0OTQnr9kN7Wr6p2Rn94759fj005TF4eHDjjeIQwCAavniix2NGmYlX4bXN259fvny6nfLtqkzbr6+be3+rG++ifUbYgUF1f8in9L3P/izf+bQISccxCEAwJWdPFnSvFnnlLxt+Eyvh6tfhivGTLrijSv+ZObP/yDOVyU//8+3d27J0plDRvTs0L1RgxYX7znZMSt7dN/cy9xTZP9+hxzEIQDAFSQSiQd6DExJGVbN51PerE4Wnl+2fGy/xxtc1TyZP2vEiPFxvjAff/wnC9ww8bWsm9r/0WaqivHjyVMqV678l39+796L/8qKigoHHsQhAMDlTXhpSqrK8Pbmnf41SP549syc3TErO/k/rlvXPnG+MBs3/tECFz8/rmE1urrXnfcX583/5/+XoqKqf9/x4yd7PvBwQcHnzjyIQwCA39u69YsGVzVLVRy++cSzV/xNyHEDnmiY3BuGv84N17eL87XZsOGyO/z0tWnVX2CTRi3fGjLil5271q//pGnTO6r+L/36DnbsQRwCAPwbpaVnW2bdlaoyvLph1qkF7/1RFv6yYsWi58a2/OPfhKzFVGVtZWVlbC/P2rWXrvGH9xbX4ptdO97W9bf/c+/eAw4/iEMAgH/19NDRKUy1x+/rd9ksrFi5csWYSR1adE7hn/XrlJWdi+3lWbPm0mU+99CjyS/t6adfdPhBHAIA/Ist+YWp7bRPX5t26S+Rvjv8xdubdUxHFl6ckydLYnuFLinDw3+dl5LfyL26UdaJE6f8CIA4BAD4p/Ly8nZt70lhpN2Zlf3br6LZPePt5x569IbGrdOXhRfn1Knv43mFKisvjcPRfXNTtbeJE6f6KQBxCADUdwUFhXdlP5jaSMt7ZlRVvXwz952ZQ0Zkt+qS7ib8dX744XQ8L1IicenNP1IY2zfecHucfyMXxCEAwJ/bu/dAv76D0xFpIx8clJK7U9R0SkvPxvNSXbjwuzj8ePKU1K5u9ux3/USAOMyYuXMWG2OMMSYj88aUud27/6Xu4y3dM2PG/Fher/f++v7v4vCFVHwVze/ePHx71kI/GlGZ7dv3qAlxGCu3NO1ijDHGmDqfu65tfHv8svDixPWqdW3f+7dlWLlyZav/fZfC1M4N197hByQqM+PN+WpCHMZK/37DjDHGGFOX0/Xu/k0at4lrGVZNXC/c8Cde+G0cfps3Px3bu/662/2MRGWWLVunJsQhAEBtHD78bf/+Q2KchVVzbZPb4nr5Zr3x9uP39T29aPHFOFw1dlKadlhQ8LkfFhCHAEA8lZeXT3l91jVXt4x3GVbNrbfcGdeL+PQTI//5AV7fbvmLE6ricPIjT6Vph/36DvYjA+IQAIihjRvzW2bdFfssvDht23SL63Xsmt3714fZJ7tnjzu6p2+Ne/ce8IMD4hAAiI/S0rMjR06oJ1l4cbp3HxDLS1lRUXFt6m5peMV5+ukX/fiAOAQAYmLN6o233HxnvSrDqsl9bHgsr+aRI0frco2NGmZ9990JP0QgDgGAaCspKRn0yLD6loUXZ+yYV2N5Tdev/6SONzlx4lQ/SiAOAYCoqqysfPfdJddf17Z+lmHVzJyZF8srO3363DreZNUpKis752cKxCEAED0lJSX9+w2ut1l4cZYtWx3LiztkyPN1v8zZs9/1YwXiEACImI8+2nRz0w71vAyr5uCBb2J5fbN/81WldTatWnVJJBJ+uEAcAgDRUF5ePmLEeFlYNVV5XFlZGcurfH/3v2RkpcuXrfEjBuIQAIiAb775NiPvKYU2DRu0qMqnJR+sjOuFnvDSlIwsNrtzr7j2NohDACA+Vq1aX5+/e6Zq7uzQY8yLr2zYkB/7r0756KNNmVpyQcHnftYQhwAAgaqsrJzy+qz6GYS33nJnbu7whQuXHT92sv5c8ZKSkkwtvF/fwX7iEIcAACE6d+6n+nYbw2ub3JaTkztzZt7u3Xvr7W85tm51d6b2/3XRfj93iEMAgLCcOHGqc6de9edjhFNen/XFFzt8Z2aVwU88l6lr8fTTL9o/4hAAICDFxcfatb0n3k3Ytk23kSMnrFm9sbT0rCv+W3/963uZuiiNGmZ9990JlwBxCAAQhK+L9jdv1inq7df0pvadO/Ua0P/Jp4eOrorAsWNefeH5icOfHffsM2OXLFl18mSJC/1Hdu4syuCFmzRpmkuAOAQAyLy9ew/cdGP7KNZgy6y7ch8bPuuteZ99ti323ymaVolEoknj1pm6jtdf19blQxwCAGRYcfGxFs07R+63Q6dMmVVUtM/lS6Ee9w/M4DWdPftdlwBxCACQMd9/f7pdu3sjlIX9+w3euvULd05PhwkT3sjsm8C+GQhxCACQGRUVFTk5uVHJwp4PPLx7915XLX3Wrt2c2Uv84fKPXAXEIQBABsx6a14ksvCmG9vLhjpQUvJDZi90dude3hNGHAIA1LWvi/Zf3Sgr/DLscf9A9zmoM61b3Z3Zy11QUOgqIA4BAOpUJH6h9PHcEefPX3Cx6sydHXpk9op37tTrwgWfPEQcAgDUlaKifeGX4YQJb1RUVLhYdaltm24Zv+5z5yx0IRCHAAB1ZNKkaSFn4XXXtlmzeqPLVPc63HF/CB8xPXOm1LVAHAIA1IWePR8JtgwHP/Hc0aPfuUYZkd25VwhnYNKkaa4F4hAAoE4aILt3gFmYk5P71Vd7XJ0Muqdb3xBOQpPGrU+f/tHlQBwCAKTdwwOHhtOE11/XdtSoyQcPfOO6ZFz/foMDORXTps5xORCHAABpl/HbnVfN1Y2y/jLgqSVLVpWVnXNFAjH82XGBxGFWi2xfR4Q4BACoCyOGj8/Ii/6mTe94PHfEkg9WlpaedRVCM3Xq7HDeUnbPQ8QhAEBdqKysfGvmO42vaV033z/Zr+/gmTPzdu/e6+2gkIXwlvKvM/zZca4I4hAAoI6cPFkyedK0du3uTfHbgze173H/wNGjJy9duvrQoSNVIWrVkXD48LfhxGGrll1cEcQhAEBd27//0Ky35j355At3d8lp0rgGbye2aN75vnv75+YOHz/u9XnvLN669YtTp763z4iqyvhbbr4znD48dOiIi4I4BADIZCGUlJQcPPDNl9t2btq0Zc3qjb+dLfmFO3cWHT787Q8/nE4kEtYVM7mPDQ8nDletWu+KIA4BACAD3n13SThx+MYbb7siiEMAAMiAkpIfGjZo4TtpQBwCAFDf9e83OJA4HDZsjMuBOAQAgMxYuXJdIHH49NMvuhyIQwAAyIyKior2t98XQhy+8vKbLgfiEAAAMuaD91f4tlIQhwAA1HeJROLODj0yHodHjhx1LRCHAACQSbt2FmX2a0u7du3jKiAOAQAg86a8PiuDcbhm9UaXAHEIAACZl0gkevcalJEyrPpzKysrXQLEIQAABKG09Gx2du86LsPmzTp/990Jy0ccAgBAQM6cKe3WtU+dlWGzWzsWFe2zdsQhAAAEp7y8/MknX6ibL6EpLj5m4YhDAAAI1/Jla5re1D5NWXjdtW2mT5tz4ULCnhGHAAAQujNnSl8aP6VJ49YpzMKq4Hx58vSSkoRkZ/UAAAHeSURBVB+sF3EIAAARS8S3Zr7TqWPPZJrw5qYdBj/x3Nq1m8vLz1sp4hAAACLs4MHDeXmLcnOHt2t3b8MGLa70HaSd7u/+l+dGTliwYElR0b6KigoLBHEIAACAOAQAAEAcAgAAIA4BAAAQhwAAAIhDAAAAxCEAAADiEAAAAHEIAACAOAQAAEAcAgAAIA4BAAAQhwAAAIhDAAAAxCEAAADiEAAAAHEIAACAOAQAAEAcAgAAIA4BAAAQhwAAAIhDAAAAxCEAAADiEAAAAHEIAACAOAQAAEAcAgAAIA4BAAAQhwAAAIhDAAAAxCEAAADiEAAAAHEIAACAOAQAAEAcAgAAIA4BAAAQhwAAAIhDAAAAxCEAAADiEAAAAHEIAACAOAQAAEAcAgAAIA4BAAAQhwAAAIhDAAAAxCEAAADiEAAAAHEIAACAOAQAAEAcAgAAIA4BAAAQhwAAAIhDAAAAxCEAAADiEAAAAHEIAACAOAQAAEAcAgAAIA4BAAAQhwAAAIhDAAAAxCEAAADiEAAAAHEIAACAOAQAAEAcAgAAIA4BAAAQhwAAAIhDAAAAxCEAAADiEAAAAHEIAACAOAQAAEAcAgAAIA4BAAAQhwAAAIhDAAAAxCEAAADiEAAAgDT4XwemywuCSp+mAAAAAElFTkSuQmCC"}}}]}