{ "cells": [ { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "%reload_ext autoreload\n", "%autoreload 2\n", "%matplotlib inline" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "from fastai.vision import *\n", "from fastai.metrics import error_rate" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "path = Path('Database')" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[PosixPath('Database/GoodPLR_0.404_7581.jpg'),\n", " PosixPath('Database/GoodPLR_0.433_2440.jpg'),\n", " PosixPath('Database/GoodPLR_0.352_5594.jpg'),\n", " PosixPath('Database/GoodPLR_0.46_5072.jpg'),\n", " PosixPath('Database/GoodPLR_0.37_914.jpg'),\n", " PosixPath('Database/GoodPLR_0.404_840.jpg'),\n", " PosixPath('Database/GoodPLR_0.404_9135.jpg'),\n", " PosixPath('Database/GoodPLR_0.509_6291.jpg'),\n", " PosixPath('Database/GoodPLR_0.389_3233.jpg'),\n", " PosixPath('Database/GoodPLR_0.46_1675.jpg'),\n", " PosixPath('Database/GoodPLR_0.391_4683.jpg'),\n", " PosixPath('Database/GoodPLR_0.439_7106.jpg'),\n", " PosixPath('Database/GoodPLR_0.442_3577.jpg'),\n", " PosixPath('Database/GoodPLR_0.259_2725.jpg'),\n", " PosixPath('Database/GoodPLR_0.463_3786.jpg'),\n", " PosixPath('Database/GoodPLR_0.52_3953.jpg'),\n", " PosixPath('Database/GoodPLR_0.434_7464.jpg'),\n", " PosixPath('Database/GoodPLR_0.349_8029.jpg'),\n", " PosixPath('Database/GoodPLR_0.421_4339.jpg'),\n", " PosixPath('Database/GoodPLR_0.339_9193.jpg'),\n", " PosixPath('Database/GoodPLR_0.404_3004.jpg'),\n", " PosixPath('Database/GoodPLR_0.36_8549.jpg'),\n", " PosixPath('Database/GoodPLR_0.362_2952.jpg'),\n", " PosixPath('Database/GoodPLR_0.37_478.jpg'),\n", " PosixPath('Database/GoodPLR_0.421_1363.jpg'),\n", " PosixPath('Database/GoodPLR_0.46_2342.jpg'),\n", " PosixPath('Database/GoodPLR_0.46_6261.jpg'),\n", " PosixPath('Database/GoodPLR_0.341_3736.jpg'),\n", " PosixPath('Database/GoodPLR_0.472_1775.jpg'),\n", " PosixPath('Database/GoodPLR_0.396_2996.jpg'),\n", " PosixPath('Database/GoodPLR_0.3_7677.jpg'),\n", " PosixPath('Database/GoodPLR_0.611_8898.jpg'),\n", " PosixPath('Database/GoodPLR_0.545_1715.jpg'),\n", " PosixPath('Database/GoodPLR_0.404_4433.jpg'),\n", " PosixPath('Database/GoodPLR_0.451_9843.jpg'),\n", " PosixPath('Database/GoodPLR_0.469_621.jpg'),\n", " PosixPath('Database/GoodPLR_0.457_649.jpg'),\n", " PosixPath('Database/GoodPLR_0.451_3790.jpg'),\n", " PosixPath('Database/GoodPLR_0.527_3408.jpg'),\n", " PosixPath('Database/GoodPLR_0.412_8090.jpg'),\n", " PosixPath('Database/GoodPLR_0.439_5332.jpg'),\n", " PosixPath('Database/GoodPLR_0.509_528.jpg'),\n", " PosixPath('Database/GoodPLR_0.453_97.jpg'),\n", " PosixPath('Database/GoodPLR_0.421_5869.jpg'),\n", " PosixPath('Database/GoodPLR_0.358_9200.jpg'),\n", " PosixPath('Database/GoodPLR_0.385_8919.jpg'),\n", " PosixPath('Database/GoodPLR_0.433_4170.jpg'),\n", " PosixPath('Database/GoodPLR_0.294_7891.jpg'),\n", " PosixPath('Database/GoodPLR_0.345_4944.jpg'),\n", " PosixPath('Database/GoodPLR_0.429_3374.jpg'),\n", " PosixPath('Database/GoodPLR_0.345_1363.jpg'),\n", " PosixPath('Database/GoodPLR_0.358_2210.jpg'),\n", " PosixPath('Database/GoodPLR_0.358_3472.jpg'),\n", " PosixPath('Database/GoodPLR_0.42_186.jpg'),\n", " PosixPath('Database/GoodPLR_0.35_9730.jpg'),\n", " PosixPath('Database/GoodPLR_0.345_2904.jpg'),\n", " PosixPath('Database/GoodPLR_0.396_866.jpg'),\n", " PosixPath('Database/GoodPLR_0.321_2785.jpg'),\n", " PosixPath('Database/GoodPLR_0.472_7271.jpg'),\n", " PosixPath('Database/GoodPLR_0.383_4186.jpg'),\n", " PosixPath('Database/GoodPLR_0.256_7193.jpg'),\n", " PosixPath('Database/GoodPLR_0.5_8621.jpg'),\n", " PosixPath('Database/GoodPLR_0.389_387.jpg'),\n", " PosixPath('Database/GoodPLR_0.468_8549.jpg'),\n", " PosixPath('Database/GoodPLR_0.393_4466.jpg'),\n", " PosixPath('Database/GoodPLR_0.446_2972.jpg'),\n", " PosixPath('Database/GoodPLR_0.56_5872.jpg'),\n", " PosixPath('Database/GoodPLR_0.48_5767.jpg'),\n", " PosixPath('Database/GoodPLR_0.4_3042.jpg'),\n", " PosixPath('Database/GoodPLR_0.434_2704.jpg'),\n", " PosixPath('Database/GoodPLR_0.442_8469.jpg'),\n", " PosixPath('Database/GoodPLR_0.491_4330.jpg'),\n", " PosixPath('Database/GoodPLR_0.471_462.jpg'),\n", " PosixPath('Database/GoodPLR_0.472_7153.jpg'),\n", " PosixPath('Database/GoodPLR_0.358_9380.jpg'),\n", " PosixPath('Database/GoodPLR_0.426_4216.jpg'),\n", " PosixPath('Database/GoodPLR_0.238_1857.jpg'),\n", " PosixPath('Database/GoodPLR_0.426_4146.jpg'),\n", " PosixPath('Database/GoodPLR_0.441_5527.jpg'),\n", " PosixPath('Database/GoodPLR_0.537_4572.jpg'),\n", " PosixPath('Database/GoodPLR_0.327_121.jpg'),\n", " PosixPath('Database/GoodPLR_0.412_6757.jpg'),\n", " PosixPath('Database/GoodPLR_0.35_4755.jpg'),\n", " PosixPath('Database/GoodPLR_0.391_6700.jpg'),\n", " PosixPath('Database/GoodPLR_0.408_5190.jpg'),\n", " PosixPath('Database/GoodPLR_0.442_2803.jpg'),\n", " PosixPath('Database/GoodPLR_0.462_1921.jpg'),\n", " PosixPath('Database/GoodPLR_0.417_3765.jpg'),\n", " PosixPath('Database/GoodPLR_0.446_5492.jpg'),\n", " PosixPath('Database/GoodPLR_0.438_8798.jpg'),\n", " PosixPath('Database/GoodPLR_0.537_8547.jpg'),\n", " PosixPath('Database/GoodPLR_0.404_7475.jpg'),\n", " PosixPath('Database/GoodPLR_0.481_7662.jpg'),\n", " PosixPath('Database/GoodPLR_0.439_7177.jpg'),\n", " PosixPath('Database/GoodPLR_0.455_2387.jpg'),\n", " PosixPath('Database/GoodPLR_0.473_3151.jpg'),\n", " PosixPath('Database/GoodPLR_0.365_7357.jpg'),\n", " PosixPath('Database/GoodPLR_0.418_53.jpg'),\n", " PosixPath('Database/GoodPLR_0.442_1477.jpg'),\n", " PosixPath('Database/GoodPLR_0.527_4367.jpg'),\n", " PosixPath('Database/GoodPLR_0.478_1642.jpg'),\n", " PosixPath('Database/GoodPLR_0.404_2266.jpg'),\n", " PosixPath('Database/GoodPLR_0.46_8846.jpg'),\n", " PosixPath('Database/GoodPLR_0.468_1945.jpg'),\n", " PosixPath('Database/GoodPLR_0.449_1899.jpg'),\n", " PosixPath('Database/GoodPLR_0.509_704.jpg'),\n", " PosixPath('Database/GoodPLR_0.522_2888.jpg'),\n", " PosixPath('Database/GoodPLR_0.357_238.jpg'),\n", " PosixPath('Database/GoodPLR_0.429_3697.jpg'),\n", " PosixPath('Database/GoodPLR_0.529_9426.jpg'),\n", " PosixPath('Database/GoodPLR_0.472_7543.jpg'),\n", " PosixPath('Database/GoodPLR_0.479_3398.jpg'),\n", " PosixPath('Database/GoodPLR_0.48_7067.jpg'),\n", " PosixPath('Database/GoodPLR_0.451_9265.jpg'),\n", " PosixPath('Database/GoodPLR_0.345_9173.jpg'),\n", " PosixPath('Database/GoodPLR_0.377_7544.jpg'),\n", " PosixPath('Database/GoodPLR_0.483_8133.jpg'),\n", " PosixPath('Database/GoodPLR_0.5_5843.jpg'),\n", " PosixPath('Database/GoodPLR_0.385_4206.jpg'),\n", " PosixPath('Database/GoodPLR_0.365_9072.jpg'),\n", " PosixPath('Database/GoodPLR_0.489_666.jpg'),\n", " PosixPath('Database/GoodPLR_0.468_5861.jpg'),\n", " PosixPath('Database/GoodPLR_0.39_9028.jpg'),\n", " PosixPath('Database/GoodPLR_0.463_9787.jpg'),\n", " PosixPath('Database/models'),\n", " PosixPath('Database/GoodPLR_0.389_6505.jpg'),\n", " PosixPath('Database/GoodPLR_0.457_3367.jpg'),\n", " PosixPath('Database/GoodPLR_0.396_4852.jpg'),\n", " PosixPath('Database/GoodPLR_0.32_3518.jpg'),\n", " PosixPath('Database/GoodPLR_0.473_6279.jpg'),\n", " PosixPath('Database/GoodPLR_0.489_619.jpg'),\n", " PosixPath('Database/GoodPLR_0.357_8261.jpg'),\n", " PosixPath('Database/GoodPLR_0.4_3232.jpg'),\n", " PosixPath('Database/GoodPLR_0.446_24.jpg'),\n", " PosixPath('Database/GoodPLR_0.418_2412.jpg'),\n", " PosixPath('Database/GoodPLR_0.4_2501.jpg'),\n", " PosixPath('Database/.ipynb_checkpoints'),\n", " PosixPath('Database/GoodPLR_0.434_266.jpg'),\n", " PosixPath('Database/GoodPLR_0.4_1797.jpg'),\n", " PosixPath('Database/GoodPLR_0.315_5310.jpg'),\n", " PosixPath('Database/GoodPLR_0.462_200.jpg'),\n", " PosixPath('Database/GoodPLR_0.365_5661.jpg'),\n", " PosixPath('Database/GoodPLR_0.442_1653.jpg'),\n", " PosixPath('Database/GoodPLR_0.5_5120.jpg'),\n", " PosixPath('Database/GoodPLR_0.52_6453.jpg'),\n", " PosixPath('Database/GoodPLR_0.421_815.jpg'),\n", " PosixPath('Database/GoodPLR_0.5_2574.jpg'),\n", " PosixPath('Database/GoodPLR_0.31_7665.jpg'),\n", " PosixPath('Database/GoodPLR_0.32_791.jpg'),\n", " PosixPath('Database/GoodPLR_0.345_6004.jpg'),\n", " PosixPath('Database/GoodPLR_0.48_4147.jpg'),\n", " PosixPath('Database/GoodPLR_0.365_7663.jpg'),\n", " PosixPath('Database/GoodPLR_0.442_9326.jpg'),\n", " PosixPath('Database/GoodPLR_0.442_9435.jpg'),\n", " PosixPath('Database/GoodPLR_0.49_192.jpg'),\n", " PosixPath('Database/GoodPLR_0.39_8593.jpg'),\n", " PosixPath('Database/GoodPLR_0.42_3708.jpg'),\n", " PosixPath('Database/GoodPLR_0.479_4228.jpg'),\n", " PosixPath('Database/GoodPLR_0.426_654.jpg'),\n", " PosixPath('Database/GoodPLR_0.327_9716.jpg'),\n", " PosixPath('Database/GoodPLR_0.492_4189.jpg'),\n", " PosixPath('Database/GoodPLR_0.4_8430.jpg'),\n", " PosixPath('Database/GoodPLR_0.46_8323.jpg'),\n", " PosixPath('Database/GoodPLR_0.222_2094.jpg'),\n", " PosixPath('Database/GoodPLR_0.529_7364.jpg'),\n", " PosixPath('Database/GoodPLR_0.345_2954.jpg'),\n", " PosixPath('Database/GoodPLR_0.42_4819.jpg'),\n", " PosixPath('Database/GoodPLR_0.482_5835.jpg'),\n", " PosixPath('Database/GoodPLR_0.412_1458.jpg'),\n", " PosixPath('Database/GoodPLR_0.444_2873.jpg'),\n", " PosixPath('Database/GoodPLR_0.404_4915.jpg'),\n", " PosixPath('Database/GoodPLR_0.532_7177.jpg'),\n", " PosixPath('Database/GoodPLR_0.418_2342.jpg'),\n", " PosixPath('Database/GoodPLR_0.396_97.jpg'),\n", " PosixPath('Database/GoodPLR_0.25_2925.jpg'),\n", " PosixPath('Database/GoodPLR_0.448_3465.jpg'),\n", " PosixPath('Database/GoodPLR_0.463_8105.jpg'),\n", " PosixPath('Database/GoodPLR_0.404_3682.jpg'),\n", " PosixPath('Database/GoodPLR_0.345_8376.jpg'),\n", " PosixPath('Database/GoodPLR_0.46_3063.jpg'),\n", " PosixPath('Database/GoodPLR_0.389_470.jpg'),\n", " PosixPath('Database/GoodPLR_0.442_9570.jpg'),\n", " PosixPath('Database/GoodPLR_0.455_3528.jpg'),\n", " PosixPath('Database/GoodPLR_0.431_4040.jpg'),\n", " PosixPath('Database/GoodPLR_0.426_2715.jpg'),\n", " PosixPath('Database/GoodPLR_0.415_1981.jpg'),\n", " PosixPath('Database/GoodPLR_0.42_3102.jpg'),\n", " PosixPath('Database/GoodPLR_0.4_10.jpg'),\n", " PosixPath('Database/GoodPLR_0.418_7903.jpg'),\n", " PosixPath('Database/GoodPLR_0.421_4499.jpg'),\n", " PosixPath('Database/GoodPLR_0.489_8875.jpg')]" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "path.ls()" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[PosixPath('Database/GoodPLR_0.404_7581.jpg'),\n", " PosixPath('Database/GoodPLR_0.433_2440.jpg'),\n", " PosixPath('Database/GoodPLR_0.352_5594.jpg'),\n", " PosixPath('Database/GoodPLR_0.46_5072.jpg'),\n", " PosixPath('Database/GoodPLR_0.37_914.jpg')]" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "fnames = get_image_files(path)\n", "fnames[:5]" ] }, { "cell_type": "code", "execution_count": 44, "metadata": {}, "outputs": [], "source": [ "# variable a will determine the nearest digit that the float will be rounded to. \n", "# small a will result in many classes\n", "def round_nearest(x, a):\n", " return round(x / a) * a" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [], "source": [ "#pat = r'/([^/]+)_\\d+.jpg$'\n", "def get_float_labels(y):\n", " return (float(y.parts[-1].split('_')[1]))\n", " \n", "\n", "\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "??ImageDataBunch.from_name_func" ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[PosixPath('Database/GoodPLR_0.404_7581.jpg'),\n", " PosixPath('Database/GoodPLR_0.433_2440.jpg'),\n", " PosixPath('Database/GoodPLR_0.352_5594.jpg'),\n", " PosixPath('Database/GoodPLR_0.46_5072.jpg'),\n", " PosixPath('Database/GoodPLR_0.37_914.jpg'),\n", " PosixPath('Database/GoodPLR_0.404_840.jpg'),\n", " PosixPath('Database/GoodPLR_0.404_9135.jpg'),\n", " PosixPath('Database/GoodPLR_0.509_6291.jpg'),\n", " PosixPath('Database/GoodPLR_0.389_3233.jpg'),\n", " PosixPath('Database/GoodPLR_0.46_1675.jpg'),\n", " PosixPath('Database/GoodPLR_0.391_4683.jpg'),\n", " PosixPath('Database/GoodPLR_0.439_7106.jpg'),\n", " PosixPath('Database/GoodPLR_0.442_3577.jpg'),\n", " PosixPath('Database/GoodPLR_0.259_2725.jpg'),\n", " PosixPath('Database/GoodPLR_0.463_3786.jpg'),\n", " PosixPath('Database/GoodPLR_0.52_3953.jpg'),\n", " PosixPath('Database/GoodPLR_0.434_7464.jpg'),\n", " PosixPath('Database/GoodPLR_0.349_8029.jpg'),\n", " PosixPath('Database/GoodPLR_0.421_4339.jpg'),\n", " PosixPath('Database/GoodPLR_0.339_9193.jpg'),\n", " PosixPath('Database/GoodPLR_0.404_3004.jpg'),\n", " PosixPath('Database/GoodPLR_0.36_8549.jpg'),\n", " PosixPath('Database/GoodPLR_0.362_2952.jpg'),\n", " PosixPath('Database/GoodPLR_0.37_478.jpg'),\n", " PosixPath('Database/GoodPLR_0.421_1363.jpg'),\n", " PosixPath('Database/GoodPLR_0.46_2342.jpg'),\n", " PosixPath('Database/GoodPLR_0.46_6261.jpg'),\n", " PosixPath('Database/GoodPLR_0.341_3736.jpg'),\n", " PosixPath('Database/GoodPLR_0.472_1775.jpg'),\n", " PosixPath('Database/GoodPLR_0.396_2996.jpg'),\n", " PosixPath('Database/GoodPLR_0.3_7677.jpg'),\n", " PosixPath('Database/GoodPLR_0.611_8898.jpg'),\n", " PosixPath('Database/GoodPLR_0.545_1715.jpg'),\n", " PosixPath('Database/GoodPLR_0.404_4433.jpg'),\n", " PosixPath('Database/GoodPLR_0.451_9843.jpg'),\n", " PosixPath('Database/GoodPLR_0.469_621.jpg'),\n", " PosixPath('Database/GoodPLR_0.457_649.jpg'),\n", " PosixPath('Database/GoodPLR_0.451_3790.jpg'),\n", " PosixPath('Database/GoodPLR_0.527_3408.jpg'),\n", " PosixPath('Database/GoodPLR_0.412_8090.jpg'),\n", " PosixPath('Database/GoodPLR_0.439_5332.jpg'),\n", " PosixPath('Database/GoodPLR_0.509_528.jpg'),\n", " PosixPath('Database/GoodPLR_0.453_97.jpg'),\n", " PosixPath('Database/GoodPLR_0.421_5869.jpg'),\n", " PosixPath('Database/GoodPLR_0.358_9200.jpg'),\n", " PosixPath('Database/GoodPLR_0.385_8919.jpg'),\n", " PosixPath('Database/GoodPLR_0.433_4170.jpg'),\n", " PosixPath('Database/GoodPLR_0.294_7891.jpg'),\n", " PosixPath('Database/GoodPLR_0.345_4944.jpg'),\n", " PosixPath('Database/GoodPLR_0.429_3374.jpg'),\n", " PosixPath('Database/GoodPLR_0.345_1363.jpg'),\n", " PosixPath('Database/GoodPLR_0.358_2210.jpg'),\n", " PosixPath('Database/GoodPLR_0.358_3472.jpg'),\n", " PosixPath('Database/GoodPLR_0.42_186.jpg'),\n", " PosixPath('Database/GoodPLR_0.35_9730.jpg'),\n", " PosixPath('Database/GoodPLR_0.345_2904.jpg'),\n", " PosixPath('Database/GoodPLR_0.396_866.jpg'),\n", " PosixPath('Database/GoodPLR_0.321_2785.jpg'),\n", " PosixPath('Database/GoodPLR_0.472_7271.jpg'),\n", " PosixPath('Database/GoodPLR_0.383_4186.jpg'),\n", " PosixPath('Database/GoodPLR_0.256_7193.jpg'),\n", " PosixPath('Database/GoodPLR_0.5_8621.jpg'),\n", " PosixPath('Database/GoodPLR_0.389_387.jpg'),\n", " PosixPath('Database/GoodPLR_0.468_8549.jpg'),\n", " PosixPath('Database/GoodPLR_0.393_4466.jpg'),\n", " PosixPath('Database/GoodPLR_0.446_2972.jpg'),\n", " PosixPath('Database/GoodPLR_0.56_5872.jpg'),\n", " PosixPath('Database/GoodPLR_0.48_5767.jpg'),\n", " PosixPath('Database/GoodPLR_0.4_3042.jpg'),\n", " PosixPath('Database/GoodPLR_0.434_2704.jpg'),\n", " PosixPath('Database/GoodPLR_0.442_8469.jpg'),\n", " PosixPath('Database/GoodPLR_0.491_4330.jpg'),\n", " PosixPath('Database/GoodPLR_0.471_462.jpg'),\n", " PosixPath('Database/GoodPLR_0.472_7153.jpg'),\n", " PosixPath('Database/GoodPLR_0.358_9380.jpg'),\n", " PosixPath('Database/GoodPLR_0.426_4216.jpg'),\n", " PosixPath('Database/GoodPLR_0.238_1857.jpg'),\n", " PosixPath('Database/GoodPLR_0.426_4146.jpg'),\n", " PosixPath('Database/GoodPLR_0.441_5527.jpg'),\n", " PosixPath('Database/GoodPLR_0.537_4572.jpg'),\n", " PosixPath('Database/GoodPLR_0.327_121.jpg'),\n", " PosixPath('Database/GoodPLR_0.412_6757.jpg'),\n", " PosixPath('Database/GoodPLR_0.35_4755.jpg'),\n", " PosixPath('Database/GoodPLR_0.391_6700.jpg'),\n", " PosixPath('Database/GoodPLR_0.408_5190.jpg'),\n", " PosixPath('Database/GoodPLR_0.442_2803.jpg'),\n", " PosixPath('Database/GoodPLR_0.462_1921.jpg'),\n", " PosixPath('Database/GoodPLR_0.417_3765.jpg'),\n", " PosixPath('Database/GoodPLR_0.446_5492.jpg'),\n", " PosixPath('Database/GoodPLR_0.438_8798.jpg'),\n", " PosixPath('Database/GoodPLR_0.537_8547.jpg'),\n", " PosixPath('Database/GoodPLR_0.404_7475.jpg'),\n", " PosixPath('Database/GoodPLR_0.481_7662.jpg'),\n", " PosixPath('Database/GoodPLR_0.439_7177.jpg'),\n", " PosixPath('Database/GoodPLR_0.455_2387.jpg'),\n", " PosixPath('Database/GoodPLR_0.473_3151.jpg'),\n", " PosixPath('Database/GoodPLR_0.365_7357.jpg'),\n", " PosixPath('Database/GoodPLR_0.418_53.jpg'),\n", " PosixPath('Database/GoodPLR_0.442_1477.jpg'),\n", " PosixPath('Database/GoodPLR_0.527_4367.jpg'),\n", " PosixPath('Database/GoodPLR_0.478_1642.jpg'),\n", " PosixPath('Database/GoodPLR_0.404_2266.jpg'),\n", " PosixPath('Database/GoodPLR_0.46_8846.jpg'),\n", " PosixPath('Database/GoodPLR_0.468_1945.jpg'),\n", " PosixPath('Database/GoodPLR_0.449_1899.jpg'),\n", " PosixPath('Database/GoodPLR_0.509_704.jpg'),\n", " PosixPath('Database/GoodPLR_0.522_2888.jpg'),\n", " PosixPath('Database/GoodPLR_0.357_238.jpg'),\n", " PosixPath('Database/GoodPLR_0.429_3697.jpg'),\n", " PosixPath('Database/GoodPLR_0.529_9426.jpg'),\n", " PosixPath('Database/GoodPLR_0.472_7543.jpg'),\n", " PosixPath('Database/GoodPLR_0.479_3398.jpg'),\n", " PosixPath('Database/GoodPLR_0.48_7067.jpg'),\n", " PosixPath('Database/GoodPLR_0.451_9265.jpg'),\n", " PosixPath('Database/GoodPLR_0.345_9173.jpg'),\n", " PosixPath('Database/GoodPLR_0.377_7544.jpg'),\n", " PosixPath('Database/GoodPLR_0.483_8133.jpg'),\n", " PosixPath('Database/GoodPLR_0.5_5843.jpg'),\n", " PosixPath('Database/GoodPLR_0.385_4206.jpg'),\n", " PosixPath('Database/GoodPLR_0.365_9072.jpg'),\n", " PosixPath('Database/GoodPLR_0.489_666.jpg'),\n", " PosixPath('Database/GoodPLR_0.468_5861.jpg'),\n", " PosixPath('Database/GoodPLR_0.39_9028.jpg'),\n", " PosixPath('Database/GoodPLR_0.463_9787.jpg'),\n", " PosixPath('Database/GoodPLR_0.389_6505.jpg'),\n", " PosixPath('Database/GoodPLR_0.457_3367.jpg'),\n", " PosixPath('Database/GoodPLR_0.396_4852.jpg'),\n", " PosixPath('Database/GoodPLR_0.32_3518.jpg'),\n", " PosixPath('Database/GoodPLR_0.473_6279.jpg'),\n", " PosixPath('Database/GoodPLR_0.489_619.jpg'),\n", " PosixPath('Database/GoodPLR_0.357_8261.jpg'),\n", " PosixPath('Database/GoodPLR_0.4_3232.jpg'),\n", " PosixPath('Database/GoodPLR_0.446_24.jpg'),\n", " PosixPath('Database/GoodPLR_0.418_2412.jpg'),\n", " PosixPath('Database/GoodPLR_0.4_2501.jpg'),\n", " PosixPath('Database/GoodPLR_0.434_266.jpg'),\n", " PosixPath('Database/GoodPLR_0.4_1797.jpg'),\n", " PosixPath('Database/GoodPLR_0.315_5310.jpg'),\n", " PosixPath('Database/GoodPLR_0.462_200.jpg'),\n", " PosixPath('Database/GoodPLR_0.365_5661.jpg'),\n", " PosixPath('Database/GoodPLR_0.442_1653.jpg'),\n", " PosixPath('Database/GoodPLR_0.5_5120.jpg'),\n", " PosixPath('Database/GoodPLR_0.52_6453.jpg'),\n", " PosixPath('Database/GoodPLR_0.421_815.jpg'),\n", " PosixPath('Database/GoodPLR_0.5_2574.jpg'),\n", " PosixPath('Database/GoodPLR_0.31_7665.jpg'),\n", " PosixPath('Database/GoodPLR_0.32_791.jpg'),\n", " PosixPath('Database/GoodPLR_0.345_6004.jpg'),\n", " PosixPath('Database/GoodPLR_0.48_4147.jpg'),\n", " PosixPath('Database/GoodPLR_0.365_7663.jpg'),\n", " PosixPath('Database/GoodPLR_0.442_9326.jpg'),\n", " PosixPath('Database/GoodPLR_0.442_9435.jpg'),\n", " PosixPath('Database/GoodPLR_0.49_192.jpg'),\n", " PosixPath('Database/GoodPLR_0.39_8593.jpg'),\n", " PosixPath('Database/GoodPLR_0.42_3708.jpg'),\n", " PosixPath('Database/GoodPLR_0.479_4228.jpg'),\n", " PosixPath('Database/GoodPLR_0.426_654.jpg'),\n", " PosixPath('Database/GoodPLR_0.327_9716.jpg'),\n", " PosixPath('Database/GoodPLR_0.492_4189.jpg'),\n", " PosixPath('Database/GoodPLR_0.4_8430.jpg'),\n", " PosixPath('Database/GoodPLR_0.46_8323.jpg'),\n", " PosixPath('Database/GoodPLR_0.222_2094.jpg'),\n", " PosixPath('Database/GoodPLR_0.529_7364.jpg'),\n", " PosixPath('Database/GoodPLR_0.345_2954.jpg'),\n", " PosixPath('Database/GoodPLR_0.42_4819.jpg'),\n", " PosixPath('Database/GoodPLR_0.482_5835.jpg'),\n", " PosixPath('Database/GoodPLR_0.412_1458.jpg'),\n", " PosixPath('Database/GoodPLR_0.444_2873.jpg'),\n", " PosixPath('Database/GoodPLR_0.404_4915.jpg'),\n", " PosixPath('Database/GoodPLR_0.532_7177.jpg'),\n", " PosixPath('Database/GoodPLR_0.418_2342.jpg'),\n", " PosixPath('Database/GoodPLR_0.396_97.jpg'),\n", " PosixPath('Database/GoodPLR_0.25_2925.jpg'),\n", " PosixPath('Database/GoodPLR_0.448_3465.jpg'),\n", " PosixPath('Database/GoodPLR_0.463_8105.jpg'),\n", " PosixPath('Database/GoodPLR_0.404_3682.jpg'),\n", " PosixPath('Database/GoodPLR_0.345_8376.jpg'),\n", " PosixPath('Database/GoodPLR_0.46_3063.jpg'),\n", " PosixPath('Database/GoodPLR_0.389_470.jpg'),\n", " PosixPath('Database/GoodPLR_0.442_9570.jpg'),\n", " PosixPath('Database/GoodPLR_0.455_3528.jpg'),\n", " PosixPath('Database/GoodPLR_0.431_4040.jpg'),\n", " PosixPath('Database/GoodPLR_0.426_2715.jpg'),\n", " PosixPath('Database/GoodPLR_0.415_1981.jpg'),\n", " PosixPath('Database/GoodPLR_0.42_3102.jpg'),\n", " PosixPath('Database/GoodPLR_0.4_10.jpg'),\n", " PosixPath('Database/GoodPLR_0.418_7903.jpg'),\n", " PosixPath('Database/GoodPLR_0.421_4499.jpg'),\n", " PosixPath('Database/GoodPLR_0.489_8875.jpg')]" ] }, "execution_count": 14, "metadata": {}, "output_type": "execute_result" } ], "source": [ "np.random.seed(42)\n", "bs = 40\n", "bs = 40\n", "tfms = get_transforms(do_flip=False)\n", "fnames = get_image_files(path)\n", "fnames" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "data = ImageDataBunch.from_name_func(path, fnames, get_float_labels, ds_tfms=get_transforms(), size=224,\n", " ).normalize(imagenet_stats)" ] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [], "source": [ "data = (ImageItemList(fnames)\n", " .random_split_by_pct(0.3)\n", " .label_from_func(get_float_labels)\n", " .transform(tfms, size=224)\n", " .databunch())" ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [], "source": [ "data.bs = bs" ] }, { "cell_type": "code", "execution_count": 19, "metadata": {}, "outputs": [], "source": [ "learn = create_cnn(data, models.resnet50, metrics=[mean_squared_error])" ] }, { "cell_type": "code", "execution_count": 20, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAcsAAAGoCAYAAAApYs6zAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvOIA7rQAAIABJREFUeJzsvWuMZdl13/fb+zzvq249uqv6MT3TnJHIESWRokGKkm05lpIP/hIYSeAPkWM4CPwpyBsIYgcxggQJgjgJHDiw80AAO4gRxwgQG7aiOEAcSJQlhiKi4VAazgzJ6Z6eru6ud933vee582Gftc++xekZSiLFovouoNBdt+499zzWXo//+q+1lTGGjWxkIxvZyEY28nzRP+gT2MhGNrKRjWzkusvGWW5kIxvZyEY28jGycZYb2chGNrKRjXyMbJzlRjaykY1sZCMfIxtnuZGNbGQjG9nIx8jGWW5kIxvZyEY28jGycZYb2chGNrKRjXyMbJzlc0QptauU+ntKqblS6pFS6hef875/Syn1QCk1UUo9VUr9VaVU6P39jyqlflMpNVVKfV0p9cf/4K5iIz9M8t3qnPf+WCn1jlLq8Mrr/4NS6l2lVK2U+pev/E0ppf4TpdQTpdRYKfUrSqkf/z5czkZ+SOV7oYdKqRtKqV9XSp0rpUZKqS8rpf6Y9/eksZVPlVKXSqm/oZSKvp/X9fuVjbN8vvx1IAcOgD8L/LfPMSr/EPgjxpgt4CeAzwL/BlilA/4B8F8A28BfAf6hUmrn+3/6G/khlO9W50T+XeDkQ15/E/hXgd/6kL/9GeBfAX4O2AW+DPzPv49z3sgfPvle6OEMq2c3gR3gP8faPkkk/iLweazN/CTwR4D/4Ht1Ad8P2TjLDxGlVA/4F4C/bIyZGWP+Cdbp/bmr7zXGvGeMGclHgRr4keb3PwocG2P+N2NMZYz528Ap8M9/3y9iIz9U8rvRueb9nwD+JeA/u/o3Y8xfN8b8Y2D1IR/9BPBPjDEPjDEV8LeBT3+PLmMjP+TyvdJDY8zKGPOuMabG2sUK6zR3m7f8s8BfM8ZcGGNOgb+Gda7XVjbO8sPlk0BljPmm99qbwIdGV0qpX1RKTYAzbGb538ufmp+1t2OjqY1sxJfflc4B/w3w7wPL3+X3/K/AjyilPtnAXn8e+Ee/25PdyB9a+Z7qoVLq69ig7R8A/6MxRjLQq7ZRAS8ppYa/j3P/vsrGWX649IHxldfGwODD3myM+V8aGPaTwH8HHDd/+g3gjlLqX1RKRUqpPw+8BnS/P6e9kR9i+a51Tin1zwGhMebv/R6+5xnwa8C7WAP3Z4B/+/dwnI384ZTvqR4aYz4DbAG/CPwT70//J/BvKqVuKqVu0ZSuuMa2ceMsP1xm2AfsyxYw/agPGWO+BbwF/I3m93PgTwP/DtaB/ing/wYOn3eMjbyw8l3pXAOT/RXgX/89fs9/CHwBuAekwH8E/D9KqWtrpDbyByrfcz1sINm/A/xFpdRnm5f/U+AN4GvYpOLvAwUfXoO/FrJxlh8u3wRCpdSPeq99FusIP05CbPYIgDHmV40xXzDG7GJx/08Bv/m9PNmN/KGQ71bnfhS4D/yaUuoI+N+B20qpI6XU/e/iez4L/F1jzKExpjTG/C1sLWlTt9wIfH/1MAJeBTDGLI0x/5ox5q4x5lXgHPj/mjr6tZSNs/wQMcbMsQ//P1ZK9RrK85/mQ1iDSqm/oJTab/7/aeAvAf/Y+/vnGgh2C/gvgUNjzP/1B3EdG/nhkd+Fzv0ONiv8qebnL2BRi58CHoOj8qfYOlCklEqVUrLWvwr8GaXUgVJKK6X+HNaIffv7e4Ub+WGQ75UeKqV+Rin1xxtd7Cil/j0su/YrAEqpu0qpO00r088AfxmLelxfMcZsfj7kB8va+vvAHPgA+MXm9Z8DZt77/iZWSebA+9g2kdT7+9/BYv5j4O8C+z/oa9v8XM+f71bnrnzmT2IDMP+1XwHMlZ8/2fwtxbYGPAMm2PaSP/WDvvbNz/X5+V7oIfBPYYlBU+AC+FXgT3h//xONvVxg6+d/9gd93R/3o5oT38hGNrKRjWxkI8+RDQy7kY1sZCMb2cjHyMZZbmQjG9nIRjbyMbJxlhvZyEY2spGNfIxsnOVGNrKRjWxkIx8j4ce/5fsvRyfPTJompGnCbDYjCAJWq4w8L8iyjCiKqOuaLMswxrBarVgsFhhjKIqCorDv01pTlqX7SdOUPM+p65q6rgGoqoqqqlitVsRx7JhOYRhSFIV7b1EUACilMMY0nzcoBUVRNK8rtNYYY9Bao7UmyzLqukZrTVVVKKXI85wwDFFKEUURxhh3TUEQUNc1Ydg+CjkXjzmGMYaqqtz3VVXlrs0Y4645CAKKonDXWRtjB9bWNVEUuXOSa1JKEQQBf/W/+q+vjuV7oeSX/9H/YeR+l2UJwGQyYblckqYpaZoSxzF1XVOWJVproigijmPAPrMkSTDGuH+LogBldUihnH7GcUxV2XayOI7J8oyqrFgul/T7fXSg0UqzylZEYcQHH3xAFEXs7OwQhiHGGLrdrjsPpRSW8KoIw5AwDFkul3Q6HcqydM9YKSjLiiRJGl1ShGGE1ookSamqkrquqaqafr9PWYqe4/RQKdXonNXZ2hiqsqI2NUmcsFotKYrS6aUONBio64osy4niCK0UeV6gA01dVYCiLEt+9ot/9IXWQYBvvPPbBuxzMsbaqyzL6Ha77t6LbZH/i94GQeBsZKfTQWtNXuRO9+Tzog++LVBKuddFr+S4YieDIACsLuR57taA/K2qKsIwXHuf6E0QBO7YURRTVxU6sOshSRJnn/3vjcKIstFJax8VVVW6dQSgUNTGrkl5X1EUa9cna01su9j32tSYurWlsvb/6Z//Zz5UD6+Fs5TFKw4wCAK2t7c5PHzinIAYqbIsnQJVVYUxxv1fDINSVjnkpmmt3Q1SyhqUNE0bpSuQh2ANDo3xaUWOaUyNMbgHUVU1VVWtOXNxmvLwfWWr69o9FN/RifGUz4qCyXF8h55lmTVSjfOTxSFK+x0K3yiMf02yyKwBVe77XnQ5Ozvjm9/8Jvfu3WN/f5/JZAJAt9ulrmsuLi54++23SZKET33qU/T6PWeI5J6LQ3XPUNm/yfPWWmMwjUFLOT+/IE5itNIMBgOiyO5SlOc5/V6Pqqp47bXX3PN/7733WCwWvPTSS0RRRK/Xc99r9aWiKAq01iyXS5Ik8Z691YO8yNFKs1isqKoZVVURBAG3b9/GAEkYUdVWv309yYucUAfoQKOwhrIsS3Rkx3xWVUkUx9R1q9Pz+dw6yCAgTROMgaLIXaAYRa3ub4Q151IU1talabq2fsVGSpAutkVsg9iCJEnc+32b5DtdP3D3v9//HFinHQahdTDNsevaOhsdaDpph6IsyLO8+Q4NGHdcwCUNEoSJ/S7Lcs1hyzlKsAmtzfL/r7WmLEq3NsQpin3USrvz9RMg+wVQFdXacT9OD6+Fs7Q3ChdlnJ2doXXgnORyaWf05nnuskh5COIQJaIQR+lHNxJxiDMF1pyS/dcQBBpj6u/IvsThSkbnO7myLMnzfO1aoHW44ujkmP7Dkffmee4UJgxDd/6iEBKx+X+P45gsy1zEJk5bvudq5CjR2Po1s3GWjWituXfvHgcHB84Y3b9/n9VqxXQ6dQ7oi1/8Iv1+n263y7vvvsvDhw/5qZ/6KW7cuOECNDFMSimCMHCviSMUA7FcrhgMBk5nbVZXohvnVJYV3W7XGcE8z7l//z5FURBFEbP5jMvLS/r9vgsagyAgCAMCYxGL+XzOcDhcCyaTOCEMQ7rdDqDo9XvOWK2WK5RW9lg6QGurs2EUNuvBXAmyrNGL4witbZYsKEkYhgwGA6bTqc04woCqKAH72dlsRqfbQSlFp9v5gT376yg2A6rcWs/znDiOvUC9XHOUYvuiKHLrX/RGnJLopO9s5bsAZ2vE2YqI3UEADP88TU0ap6yyFcbY301pHMJhj62cLZvNZvT7fXdc0XvfYfkSBqE7P7s+2uBf9J0alNZo3aJvzv7Ron+yhuR+rSUmgUbX+ju+f+1cfs9P83sodvGHLBYLxuMxSinm8wWrVUZZlsznc6CNiHznBbgbKIvUv7nykMSh+VCmMfVahGOMhT/8CCcMQ1arFUmSkGU2Cy3Lek0JRfkkxZdjikLK8UShr8Ic0Dp1UQgxSEEQOKfsv9ePpnz4UIIEUQyBc+W8/OsVByqfeZFlMpkQRRFaa5IkoSgK8jwnTVMXeUvWGEU2w79//z537951yMF4PGaxWDZOCBe1b21tuQUqAY1IEAZopZhOZzx9+pQsy7h79y5JkhAEAcvl0sFV4mxFL3Z3d52e1tU69J8kCYY2a1gsF/R6PbYGW40Rq1BaE+jAQaFBFBAnIb1ez36varLg2jjHXZX23G0AZzPFMAzIc6vbdVURBNaxVrXVszRNbdC7WLr1OBgM3HooioIiL9hIm1Fpbe9pmqYATveA70AwtNbOOQHOWYjDlCzfIQR5ThRFTh/FduZ57myE6JigIlJisolJRFHka58TB2tMsJYMOP2sDXEcufJBHMfOnts3W0crtkicqNgtuW57rLq1W83n8lXusnD/c22ZonX64kMkmFgul2sB6/PkWjjLPM9ZrRaUZemMVFnajLIsS1fvkwcuEbw4KVECuYnQ1iblc/7/PyyTkihZ/i+wmlKKbre75nTkbxK9iNJ+WObo/4hze140dTXbk2NLxu0/aPk+P/O8mtlAC0mI8sjnfYdZbzJLt3iKsqRf1wyHQwfV+4tVns1qtWQ0GrkSQF3XDIfb9Ps9lFLMZjNWqxVlUfL06VO3iLtdO698MBgQBAGPHz9GKcXBwQEvvfQS3V7X1mE8mAys8ZDIPoxCtLLGSyllIdhAE+sYMGgdUNcVVV0TRxFaB3R7XcqiREmmGIZNNK4JdERR2HUWaBu8drtd6qbmXVUVYRSisKUIrTVlVZJlK9I0pSgMYWhPLssMYO9HmiYURUkUR+RZjtKNI1BtkGbfl66hMy+yyJrMsoxOp+PWuticq3oh7xcbKLrolwJgPStL03StdCOonNgSCcyulrIcx8NLWMSeiX2UYMh3hPZ1HErY6/Xcdbj3aTC1Pd8kSVitVg4NFP0Q+ykZdBiEFGXhbNvVEojcD3H+fsIh11sUxRpf5WoJzpdr4SyXyyVVVbibaeuSbbYmxl+ich/Hl0Um2ZfAtFLfjOPYRT8SgfmwhNbqO5yV/50S/UpUdVUB5XO+Eud57og28sBFAtUSgvzMF9ZrpX4hH3DZre8IkyRxWarvHOU1uSe+g5Wfq1Dwiy6SpdUK6sI+L2MMy+WSVZGjUYRNYBaGNvvqD7eo8mKtVi4R/tbW1poxksVeFAVhFPLB+x9wc/8mn/nsZzg+OnYZbFmVBDpw2W0YBhbtCFQDi9pnXValIxcZYxxcpZps0NZGLTyVpikoiKOIqMkyqqomDkPCICCMIhK3nlrDjAf7l2VJVVeANZZJEpHnJUVZNgiNBmxgqbVmPB67IKyuKkd0Euctn5G1+FFG6kWSKIqatR6jVOsAJQtcz9ZaG+E7S7EB8q84T7GZ1k7m9PuDxva2BBgfgpV//RKWn3FWZYUK2wTElb5qg1EtotYGmLbs4M4X5YI3+R6wts63/a2o9YC/XoeXfefoEy99O7tcLt15+pl6XVtC1bV3ltPp1Mt4DLajpXLF6uFwyGAw4OnTp0Bbp5RI5/j42JF+VqvVWhQlv/vOAmicrK2VigKBdsogzlUehkAZV2FXqR36NUmfgSXQhqtdNjWfyqsfyAK4+q/WGmoDCnQcs1wuCUMLVwu5wlduH1b14QTJJEVR/MUh1/uii9TFdRhQl5VzelEUsbWzTZHnKAPL1dIFaEEUogwu6xcWbJqmLBYL+v0+o9GIsiyZTqcopej3+wQ64O5LdwnDkMlk4iJ9iabFANrFHgAVunEsjoRgwFARNe+fTueO5GOJMxGDTkqRFyRpgsKum7qyxIeKiiLPCRqYL4oistWKINCUDZu6P+jbTKWsiKKQuqopqwpT19TGEDTnsVpl7hpqUxOEAXESgzFNzdMaQq0sJ8Cqq2X7hkHonPtGoCgq6hrSNFlDrAQhcgGXR/KT4BhwsLtkUT773nYZWEfU7fZc1gZteUqcrNQuMayxriUJkOP7yYEcz2Ao8oI0SRsymKKqW4g0DCxSqLRaP8eG/XqVjCQBgI8K+n+XazbGWNSiaq9dPivk0SiKmM1mnJ2dsbe3h6FlEUvZ6nlyLZylj09LFBDHMUVR0Ol0HLV3b2+P5XK5VttcLBbOaUjxVghAYnj8DM2POIQJK9G/McrddJ/eL9GGT7rxYYsPy+zku/zUH1qlVMo0hkN539H+X2qQGtUYxrb2I8fOc4vTS8Yp359lGdDCwOIU5V76DvrjcPoXRTqdjiXjlCU6ikmSxAUXQRCQ1zVaB6SJNS5KKfKyINQtCxYsUWA+n6MDS2aQyP7mzZsO6pHASpAQX+cXi4Vr+RA4StjiZVUShZF7flpZxiQK+v2+M6JRFBKGEaa2rMWqLIiTmG7cceeexDF5UVmdKEpMWWMqQxBZne32uiyXK6sfVd3UqQrAEEWxc+zGwNbWFkVZ2M8HAWVRunOmrgkCq2NBGFCVODZkFEbUdUVdbZyliNgYnyHq20EJ1P1kQIIk3151O13Hupb3i50U+yK2yuq4Zan6ZEFjDGEUrjleS+aKXUZa1/Vaa5wEjGEYYWiJjT6CRmPqxNHJ8cuqXGPy+3a3LNtyl5/0+FyRILBMbVSL9snxkjhhPBm718uydN0Lcs8MZq2F76pcC2cpF1yWJdvb2yyXS4wx7O/vu4hAWH2TyYTZbMZisQBYi6J8ppcfhfiZoP+dQdDCGfbzbV3Av9k+3Cq1RImw5HVRFv9BXoVJTG1AjqdbOMV+pvbqmh6E0fSg6XD9POS7fQxfvlMIHMaA1i3OL5mJj/3LMV50OTo6Immi4k6SUpQFipbAINCSODgAHQTkDeFF0BGllKspCgze7XZZrpaEQegMTl7kxFFMr9dzTEcJ5FarFVHUsp8lCHSkhChsdAqMgUDbHso2+g9djdGeT0QYBDbYMoYwDFguV8Rx6oI+mpYoVE0Y2RqjbhAWpRSmVHb7EgOrLLeBpTEoVHO+EZWpXN9brJtov6m/llXpMgwxUEEYUOd1AzWXz3s0L5z4AS205Zba1ERhu86hDeTlnrr7qGwPotiptexLa4IgBM85iK0S/fRtjQ+l+gSyPM/JsswxsqXFyjpWCwtrte685JhyvuJoQfrXG7JSELos09osW4eXi5MaqJyfoGMSFLQkUNx9ERuYJAkvv/yyrbNi+4Qd6vIRQdu1cJZSdI7jmH6/76JSeUhgb+q3vvUtVquVe2hu0XmkHx+ukCiojTyUy7qU0k3txBAE4uTazM4vdouiQIvjy9/8iEkUQpzzGjmj2SnJwg9QNcrpjBVtJgq6df61IQgDuEJO8h3tdDolaIyhVRDB8HEKZr+jRqmoqcW2Be+PiqZeFHnllVcoqhJT2Tp2nhcuKFGmJmgifIBerwfYeqBpdGuxWLBarSyBpyHhBEFAr9dzOlKWJZeXl65HTbJRqSVJICNGR2B90WlrhEJ0EBAGIUkSNU7T6nIUaVfz3+/dRDeBWZJYCNXQDKioahJhSIYhxpEiLOs1W60cuaSuKuraANYIL5crOt2ODe5q1dQxcVG6qQ11VVE0QZ4Y1aqqqJqBCHVdNfXK0q2ZTWZpxWfrX5VAr9fafEe2XC4d1HoV8fLrf63TKtd0TmyAnwnKZyQ7k6Bf7JCsByGr+axxOTfJFv0sMQxDygYRabNk7bJSY7DBanN+VxMO4/rhbckgSZO1IS4SNIp9C4LAZY0CY7sESyvbNqLb1q7nybWwkr1ez9ZygoDZbMZkMiEIAsbjsWPH+swsnzoNLeNToDOfnu9HYDJ1xEZO1gjYRapQyjZ0+wolD1Je8zNLX2FF4USuknoAqqpGN5G+oaVI+3DvukNWqMa51rQMNLkmOQeftDSfz12dTeumTqRDF4WJw5TITJR7k1naRavCgHy5cnqRZRmLxQIdhaRR7IheMrnHYDBNVDscDhkOhw5hANYWrtDrb906oKosdJXE7cAAH8LykYra1E2tzxDFEaauKZuaqhgiO33HuCEFVsdoWj9qMI2elRWlQMBFQVW3U01MbWGwIs8Aw2g0bgLCgjhOqOsWqirKwrWQiN5L6UKMKtjMsc2UI1TYwH9RRFmUtu4ZNvXyDc8MwOmKn+X5NUs/k/TLKUIIE+fmt2YYYxzTVAK3uq5RKAf515598UmCfj3Uz+CEMKR16+TEEcp3yLnIMaFNMvzJVvY67fmgrN6GcQPR1zJ9TbmWLbHVAErbUpzWduqVrDc/kGiPwRo5yNr3EgzOjn4Uf+NaOEtoRyhdXFyQZZnrrfRTf2gjn6vZm+/YhIBhjG3UlocpN1icRJ63SuIrix/V+fVUv/7nGwk/47tahJbMQClbe1RKgbLN3ldrom39cH3iilZtj6Zco4/py+KS7/ZZwNJP1dZoae5BTFlWLvp/0cXWyDPSNKaq22wnjiMLe5YVWgfNGDgLiUn98qqxEbkKCVlITBxcshaly//9yN7qCo4AU+S2H3kNeUGR5Rn93tZaZK2VJsub0YuBIm+CzjCwddkiLzAoqtJOzrJoQwGqpiobHW4M12QyxZLfrE+rq/o7UBWwa6HX6xHHMVEUki/sOpRe5RaxKSlLy5ANdEM425BhgbakI9mYOMV2xGYbrPv1R3GYsN6PboyxvbSqXkPj/F7L1rG03yXHkXOCdmiBVtq2ejROWKZE+efoJxvOwVbrWafW2gVdQWCn7Ug2agPBECFgQgulSlLjrwGw/Za2vt5O8/ETAVlfxlgyUVWuM2CvDmO4KtfCWYZhyHg8ZrlckmWZyw6BtUzRp0f7i9THvaGFUaPINpwWRdHUhNpGccv6aqEM2y/WZq5yjKtRlq/Molgyi1Ve8x2qrds0VP6qdtG5fUji8CT6AaUCJMzWDfmo9o4pjl3O2y/G+8bWkqJKlxFchVCsItauHvKii81+NGUT4UZhSJqkLJczWxPuJIRBbDM9rYjjtq9VdM+PmAPdwlJ1XVOb2jWIt2SJ0ukijhEaNN+h3Xl1Oh1W2Yq0k6I8eEwr29sYRXZyTqfToW6IQELtD4KA+XxuXy+rpnZqa5ZF0Trqto6kSbodV8Koqoput09d167uWlYlcR07JydM3aIoGI/HAGslkX6/TxAGRGHUQIXassEb8pptR/kBPPRrKL5z8W2JH0D5zkCCcD+hAJzN1FoThAGqbj8nx/Jbz8ASr8IwwjtMa3OqtldbaYUpjXNYdWX124cxZdqQnLN8TusAU9eOCauD9Z5zn9xjjEVRLBGstoFjbevk4hTzPCdO1gc0uB74sk2shOXrEyOhDQAEnv0ouRbOcjQa2QbuJkKHtq3BT+kFajXGsFgs3I3wa0J+Kh1FQRPFlhSFIUk6TeSgCQLbJ+bj2/KgxEEKJdrH2+W7pF1EsjrgSmF5PeM1xqAFcqrbDODqYvAjG6UsiaRojKw0ocN6fbT9LkVVtf2BQhyRyM/WZDVBYOHYMAws/LGJ6hmPxxhTNRBr6mptNqDBMk/LwlLpK9tbaMw6kxVgubADzIUMJMYvCiNXg27RgYCykqkkLfIRNm0aEgjJ8U1tyIuMMIps3TG1E3Wq0qBUxXQydQ56uVhSNTDWfD53jrPf77Na5YDm4OAGeZFbir/WzOczSxjyglKZPxxFIUFgnZ4YTtF90UOBAsuqbWmSeq4YLMCt0TiO6XQ6a0jNiy5pmq4RCP0WNb9dQ+61aaBsVTfZZDO710eefFtjGcxtwiABkc8P8WueLgloWiz84/iICdjgPwhaWy36o5RFMOwaaWqOxmaEhcfAFea5yFXnV3v1SkObAQvJ00+gJEmR40i3hE9W8rkofuvM8+RaaOh8Pl/DwiWFz/PcERDi2NLV/aHpEqXLBfrK0el0iGMbSS0WC8ewrarK9RqFYeiMnJ8hSAFYaPuA+36BOIQV6bO85L0CafmRlSikTwDyyUNXaxB1XVOjMA3d2ycC+cou5+v/zYes5/M5ShnStLsWTZkmoteaj8TpXxRJkoT5fGrHvOkFW8OtBma1wZUs+GfPnmFMTZat2N294Xosq7py9Re/p1cHXtuS0uRV7hyg0gpV2W/w9b/Ii7Wak+h7VVeURUnQjEErC9klpIE1wfUVL1d2iLoMck/TlF63R1EWbO9sUzZDuoUxWTd9cEVREscRWTNxB2MNuNXVqoGjLWTmZypi3OM4Jg1SlosFMnTAr1tJ4CgIkl8D3kibJPhGW/RAbIlf3vEDEBlmXxSFSyr8mrgcX3ancTZWB6Bwdsbv1ZZA2+dpSMJgnXbqMkUJspzTbvTEd66iD/Y62xnY0DhHGVRBe+5BEDQlrHUbLzNz/VquvVbt3us7xavX4Y7VoDofN0XqWjhL6XnxnYpE0xbCMm53DV9RhG4vkYz0mo3HY2azGUliYYHt7W1HNPAZhvIAV6vVWpYnStTOO7Q32h+c7js/UaqrRWRgDcf3j+UrPoA0a/uYvwosGcjUNWGovWvX35HFSjFca+WyA4mWZrMZRdNTpxrobl25NqllmiZ0u2mjA8Kuzjk9PWMwGBDHNXGccO/ePXSgmsZ8LPO1qUm3z8TODxZkoq2XK3f/jTEoA2EYOYcxGo0slFq3k6OUskzHOIkdpV70dLVaAbhSgE/82ruxR11Z3er3ei6C7qQd6rolpNWmrWWtsiVJbEfUKa3Q6LUSRxhGjsFq21Usk7BsRuVVddVkulVDCqqdc7zKUhT2pAwS2QRsVvw+Qx9du2o3JAtbJwWaNefjE3PE1viEQj/h8LNLyT7te2uMaUtMcn5yrkGg6fT7zOczFywBNni8MnQAGtqiSw580pByWwmKoxR/IOLfCzmHqlyvd8ta822af20+0ufeY7BEH34IJvj47R4+1KCUYmtri/F47CBFv7AtOzIIGchmUcph0yCj9Gp2d/dYLJZcXFysMRa3trZQSjGfz8nznE6nw/b2tjM+0Crd1ta5C3ssAAAgAElEQVQWz549c/CaMMLkd8loBXby65jrkCtrBsS+t42OfAzfvxf+Qohihan9ojrIXt5yfHs/be+SRIwy1NuGaqKAG0Nls6q8WcBtwHPn7h2KvCAMY5SiMS4VlTGEQWC3AWqY1QKVCzwpM4VlwEaLgAiRokLrtvWk2+s61iLNLjyShVZlRVEX2Ik4xg08AFuz7/a6a6QupRQ3b920uthsoyRQmJ3FvGJra4te1GtYss00l2ZEsjItpV5QEqm1i+ETp6d0M3bS4K65quw8WSlZ+OIbJH/Nb6QZFWdaRrTvDMUG+I7kKtRqjHFTkXxnIAiCsGwFoRNH5tcWfdKLVgoVhGv2KI5iOt2Og+NXq2XrdBvkIQoj1w9s9+ZsB84oZfkkFkZt90ethERZ1Y4lbYdrWNuY5XYnHLnGuq6Jk7Y85Q+ngZarIcmHlOlkxrJv441ps/rnybVwln70IScv0a5s7yPGRpyhhc3ma31FUndUqoXDbLO4dUbdbpckSbi8vHTOebFYsLW1RZ7nzulCO10CWBvq2+l0uLy8dAXyfr/P6empM66y8PM8d3sNAmvKJv//MCq4fHcLHdj6otwb+WyRFxizvv+bQLN+9NdmM1aWy2UzzFuhZS1t2BU281MQx4mrn9vNkCuKIidJuiit6HV7KCXsVYOO211tJPL2aztRLJNvLONPS6ZpWhRCAkDTpKpidESP6mZvSaVsC5KfsYlT7qQd184is2XnM1uDlPpWFEXM5zPCMLTGrq5ZLW12GkURy2LpSBFVVRJgg626Ns4YlVVpna9uBrs3u5KIkcrzrMlGzJpu+yKBpBgvf0286CLwtjHGtdNIRgUtI5RmqL3YibXWEiV5WgvhovxNFdre6qqsSDvpFQRAuayTAC/TtfqpXfbbwrcYKKqCIGzKSbTlJK0VWkceuUZsOZ4DbaFSvFoimDWW69VrvGpDhffiJx5BELiALc/tXq61aqBj2mlIEhA+T66FsxQIxod8ZNSYvzGyPECJdAeDgYNwhAkqyuSzRoXtJBF1XdvxYxLRr1arNSbp6emp2x1CHqYMZ5ddy61hyJnNZu7cZTzZcDjk4uLCnbcPrfrQydUoyC84tw6upXQbUzlYwv6+PjvRh3bldbkPEoTIKMBer0dVtd/7oktdN9OOysoNFIDEReFikGwbSYlSmjgOXGYvwU4URS6L84M3C483UbRP8KKd8mSDFtVknc12XnVF4Q1r9yGwTrcDxhuSoCyLN0nStXMuy8KhJkHQ9Bkbg4qUKytIFig/qIZqX9s+TakNhWGI0dYYlkXpBmn7mY5jaupWp3099DMZ0dVNZmlFGu99JEp0Adr1WlXrw1KuolHWbqz3brfvsX8Pwwjd6G5ZlgyHQy4vL50Ndjodru+T204DauurgQ4oipwolKy2JssE7tRkWeGet7VL6yNAjWk3yrC2bb0+WxQFcRR719e26Yn9V8pODErixI5fbN6XF20roZ+QCYtYfIQEFc+Ta+EsfUZoFEUkScL29jaXl5cO1pKHJFh6p2OZrTIvVYranU7HRfLiHMMw5OLigjRN3QSg8/NzAGazmcPAZV8zgPPzc3Z2dnj27JnrrRM6tJB/FouFa8zd29tzI/jm87mrD4pCi+NcLBbOsfnXfdVYXC3oQ+tAlVIkSepIHdAqthg0/5hiAEWhZJKMJUFtiBXQNkgrpRxhQYJMYfyJg2iHWfuZfdtvmWhLyZ9MJm4aj09Gi6KIumw3oIV2IftEBMlW/e2P4iQmbBxeHMVrbUxBoEEpZvOZRZpMs6kt7abA/rVUVUWcxG2dJwhAGeaLOXFk538WeeHmLwtq0W7XZRgOh669a7VaEUYhcWTLJVVpNyn2g7ayLC2hoiF3+Dq6kRZR87MjcV6iF1f7AW2Tft2UU9bHaELLbgVLMqtNTVmUtm0DqFfWIZ2dnRE1NXR/NnKapm5DZ6uT7WSwMAgtShLhkJiqrFz7U13XKNa3AWxRBxtEWv3264zrAw2c/Wq+05/WI/ZM1qTvQN1UKWOoaMmNfteF2Ez/s8+Ta+Es5WHKotra2mJ7e9tlfNLAnWWZm/YzHA6Zz+dMJhPADpLu9XoURcFkMuH09NRduNyIxWLhakh1bbcDm8/nDZ1+5YzBcrl0Gav8rarsrh2dTofpdIoxhslkQp7nbG9vr9Hj/bFP/u4lPuvvO6AT1ie++M6yjdb9ovV3jgjziUXy/X5dy4c6REnWCt0vsMiMSKnhygxWaZQWA+JvxLyOBGiUss9Eenu3trYA1oIU6RGTSFYyQJ9J6BMQZFFLAJQm6RrcK5CdXR/WucZJTFkUdLu9BvJtZ7imaeJG+YEhz3JQkGc5q9GYsrR10aPplE6nw3w+59nTZ6yyFQcHB3S7XVbL3EHEglTYzLmmKDR0cHM3BaJdrpbIrvU+0iFllY1YEWd5lasgQZM/+lCev7UNNmMUxEnm7fo6JPXlOIidnRMdlucirUx+H6fYQiGiCRomSF5d10wmEzodi6wJ8te2urW9lKKvYu/FZlVV6wCTpONskqwzuQ6/p1xe82voLpv0pr75aIcLIBpimgz5t9wTW0N9nlwbLY2iyEWn0+mU0WjkFEeclsy9LIqCg4MDRqMRi8WC09NTt9XMwcEBR0dHPH361O2d5iJaY/uTxEAJi1AWvDygLMuYTqfMZjOXGRRFwWw2YzgcUpZlA2NaByzOUpp8q6qyOzE0SiNba6Vp6qjY/sbWPgvXj25ayEARBNqRgHyoVgyPz4ATBfTZdOK4/QxelG4zwccSK8RRCVNaNRNuQEoFLYQj97Y2tctIxeFpra1+NK0evkGSLM4ncNgfO5TCJ4+BfcZpmlLVlYOhwjBwRAWB5NwISFM3NHtLPLLDFmx/W9kMjBbjMZ/Pmzrm3BnfxWLutryra5vF7OzuMJ1OmUwmTCYTVquMd999h5/8iZ+k15f9Eu2wa9tila8hFmLQRG99kooLIjYBG9DCrDKA3M8yBYoVdEscVdsLG7mRcUK0uWpj/IxK9FzKUaKnYkuhnb4UxzFFXrghGZLdyT6/AtPP53NrV8sChe1JjqPEDeQIAjvS1DpSyJs2KUPlNi6IY4vWpWmCbljXojN5nrs1J9mkMWbN/vrQrD9gRtaWRY+EXRySZXmTtX50SepaOEu/AVeMOrSbkQqh59atWxwdHbmFu7W15SDSi4sLnj17xltvveUMh2SI3a5lCo5GI/I8p9/vAzQtFcXabNVut8v+/r6LSmRIuSiNGLi6rh0EJdN/wJIp3n//fQcbS3Yo393tdl0dQgxFt9t1BlKUV8Qa2YCqamtGEkmKo/SzC1FspxRNUOA7UB+7l9dedGmDFEEHbN8atAZLa7vQoygiiiPnCINQuwVZFEUzDMAyZOX52IwtcEYD2h0i/GzS1aCambCyyKU1qa2/F07vhHgmQVNZ2nF8Fu0ArW0UL3tqSkb7+PFjoN2PUwzPxcWIfr9HWVasVjlnZxc8ePCAWwe3CMKAw8NDXn31VQ6fPKXXtKXYgLZHUeRsbW07wy4ZsQQKPvR6tZ65kfVB5lrpNeMPbU3TR5BEJ3w0wu9tldKVoGqScIjdlZqo6K/W2o00lERCstC6rt3MbsnWbAuTYbFYrtU2RZ9Wy8yRe8QeCpoQx17wrxVlWbBa5c0Wd5owCi2z1mN5iy5dJUr6fBW5Pljf8cSSl/yZs809Kgy1qj9SD6+Fs/RJMKIc4lBcnSQMefbsGVEU0el0+MpXvuKyxMvLS+esptOphYqaXbnPz885Pz/n9u3bhGHIbDZjPB67h/f06VOSJHHF5cPDQ7a3t8nz3LFehQ02HA4dOejy8tL1YopSGWPY29vj4OCA9957z7IUm7qgZHWj0Yh+v+++Yzweu9cEm5dagVKqqXVJv55tG7D3yu6F6UN3fv+UGE6JQsWw+pmnn1m/6OL6DuugIb0UKK2aulvlNjfu9Xru3ulAEwZhc1/b3rBABw3JwR7bQrnaGSUJlPzeSKnLyzPzIXif3OEjBgINi+5K4CcBUl7Ydg5pGcqyjPPzc5RSPHr/Ea++9ioA33z3m6SdlG9/+9tsb+/w7NlT9vf3m0xzwdnZGU+fPqUsS8bjMTdv3uSN33oDpRSf/NQnATg7O2N7uM3l6JK7d+86RGZ3d7dtMWkyg729PVd68SerbKSt3Uqw7RMYfUJUWbTzn6uqsvNfjbEs5ibpEIfh8yLqumbQHzCdTdc5DrQDAOw0K4s6TSYTl6EJGuYHO1mWubUj5ac0Tdc4JmJvZN34xMfWadk5rVUFUWSHClRVRbbKWJlVuxabmmin03H3S3ruJRDzB3pMp1P6/b71KVqRN6UKe/4AyvXNi61/nlwLZ3l5eUm323WGvdvt0u/3OTk5cYMETk9PGQ6HjvAzmUw4OztzRqbf77uC7nw+p65rRqMRSilmsxnvv/8+OzsWTrq4uHAZlg97hWHoICmtNYPBwO2dKYZiMBi4mtTx8TGDwQDAQasCt3Y6HdfLJkQNMRDSqykPVfbprGs73ODs7IwkSZqB1C2pyTr4zIuO2h3QRSkFUhUDK4be7/n0/5Vzf9FF7oH0d1WVbaq2fV0tnC2QvjCkszpbg4lc75tqswSpUfr1FHGWV0lZ8jlxio6dCu1O8817JUCUBS5Bkt9GJYHhyckJl5eXXFxcsFgsGA6H/MZv/AYHBwc8fvyYm/s3OT4+pq5rLi4u2N3d5eLiArCZw82b+85APXjwgLquef3111ksFraev1g6UshoNKIoCk5PTykKu3GxGMeqsgZMdN+HqDfSZoJ+9udD9pL5SIYo2VZtZMs3G7wvFos1zoQ4sKqqyPLMI6lZ1G46nbqgTfR7uVw2tq9gsVw2UKw9XtLUztM0RWlFElsItN/vO3tnjO3vVQ1rX/RUkJW8yF2LSRjGGCNBvG3jMldGL4ZhCCWo0NpauxbtGtraGl6p4eIQGbHvomcC5QY68FAdWwb5yGfzfXvqvwuRiFNIMpPJZC3tln7K8XhMFEVcXl5yenrKarVag1WlhigParlc8pM/+ZOsViuOjo7cDFqBAcQAQutcdnZ2XAYmTliK3BJViUO/desW4/GYx48fO+hzNBpxcHBAp9Mhz3PSNCWKImazGVVV0e12GY1GTKdTFouFIyYdHBw4gzYYDFzGe/PmTXd/rHJrisJO+4njZK2+KVGWGH5RML+mKf/3CT4bggVkeU7SzF91xknZYfa2VmlcXWiZrYiCdjamT1rwyVlX4XSgGafXRtQ+FK61drM9Yb031zduWZaRpilhs2VRXdeOoCS1I2Ps4ILlcsl0MuXwyROePX3KjZs3uLy8BGwZ4tmzZ7z66qs8eO8Bn/zkJ/nSl77Ez//8z1OWJfP5nK2tLU5PTxs0xAaro8sRP/bpH2M+n9Pr9Xjy5An7+/v0ej0ODw8JgoDJeMLF5SXD4bBZHwlbW4OmZi8chdjVmjY6aMUvqyja/SCvkn2EMCg6J+PtsixHa2UHE5gW7ZDWNxnb2el0nO5UVb2GMs3n82a7tnaQRhiEmNqwv3/AaDyyZQFTM9wa2vmuup2EliQJi8WcILCboKugnYpWVSVp2pQMlEK5IKpE+juDoNV1aZ2SklwQtOvD1ktDlxiBDTa63a5jXGutWa6WxM0We0VZ2A20TUsIsgFp+LEI27XQ0DRNHezz9ttvs1wu137EOaVpynK5ZDabuU10JQI7PT3l7OwMrbWrZy6XS375l3/Z0fbjOHa1TJ+a7NcJxEGLcknTd7/fd1lmnucui+x2u26hC1Sxvb1Np9Oh2+1yenrKjRs3MMbQ7XZZLBYMBgOiKOLevXuMRiMuLy85ODggDEP29/eZz+e89tprDqbqdDoepVzYaq2T88kjftZ4NRr1GW6Ai7o2gqOwC4svSRNqY1nHYJ2c3XNV0+v3KbN87b5neebmWgIus6vqZksj2vpSVa3v8i6ZZFW3zhbawf1CMAOaZ6tYrubOuWbZsoHuI9ey8vTpU1dmqKqKb37zm3zxi1/knXfeodvp8qVf/RL9QZ9Pf/rTHB4e8v6j9zk6PuLTn/40RVFSVxUPHjzk9q1bdDod+t0ex0fHrLIVg60tOp0O3/rWt+j3+1xcXFho99Ejtre3OXp2zN6NXYbDLWazOY8ePaLT6bCzs8O3v/VtvvgzX3SbSyexZRh/3Ma7L4oIhO7vtOQTVnzegd9F0HI+mm23lHEkFmGlir2LoojJZOJely6BoixYLVeMx2OXrEjPba/bc+1yvW5vLVBcLBZ0tjoO2RqPx+7v0I7etIiHPVfb9xuvMXbz3GaCqyx3vcpBEFDrCqXaQSvyfimNCLRbVSXj8dxeT1mQNANG4jhibuxev0orCm2TobIqicJ2pKnPrP0wuRbO8vz8nKIouHv3Ljdu3GA8Hrs6pBBo5EKE/ervwXdxceGyQXF40+mUwWBgH2TjbARfF0jAd4x+9iWYv0C04nSEzSoPyK+nCmtRcHLJ6Hq9nmsIl82Ze70ejx8/djVJqV0NBgNGoxG9Xo+HDx9y8+ZN3nzzTT7zmc8QRRF7e3tk2bIxpvbhynn6JAlZYHI9Ak34tH0/Wv0onP5FkSiMKEr7HLo9O7JNhzaaVlq5GrMxtgbpN9lrrdwGzQLPCtIQBIHNEBrH6frdGicrk6H8Oo7AR36dyW8fqOu23in9oEVRcH5+yWQy4eL8gsvRJUfPjuj2ujx58oTh1pAss87z9u3bbreHy4tL3n33Xba3d+h2O6RpyuXFBYvlgjzLmM6mdtbyxBLdagz9fp+nT5/aiP7SZhl5lnPv3j2+9rU3WS0WLFcLhltbJFHshiYALFdLzs7OyLKMO3fucFFc0Ol01hiYL7q0hDK99uzFvtm6pV5DjiTo8vkfRVHQ6/Ucr6LIC5bl0qFukmiIs12uliwXC8IoJImThi0bEMfWAR4cHNgTbNi2tsSUub5xIRhaHa7QQeAyOGHs+qzUPC+aslLgmN11syFBWbS1UWNAxuXZ322LixyrKG0gOZvNSNLU2mvsLPDlcslwe4hWbUeAoH0+0dIY05BBO899LtfCWSqlGI/HazNeJYvb2dlhNpuhtR0IvlqtOD4+ds5UitwSJcnClHrfcDh0JAmJIAaDgRt4ILCERHMCaUBL4/d74KBl6fpFdInaZMTddDptDGJFmnbXshCBV9cp+wv29vaYTCYkScJLL73El770JX7hF36Bhw8fEgQB+/v7hGHYZMbtVmTiwK/WIcUZQtuMKxm0/CvX96KLM06VbdgOgxAdhk122dY8jDHU1FDZCSZhGNnai1nfTsi1aJTt3qvSZwi4Z+Uv1qvPzNcZfzpOnluHK7XPLMs4Pj7m4YOH1tAlCVprnj57yu7uLvP5nMFgwHg8dgbjpbsv0e31ODo+YjDYambNFrz99tsMh0MG/QFKS29mShRYU5Gtlq6/WWvN4eEhBrsX4ipbsb095CzPyTM70erJkydESczl5SVbW1sNZ+CS+WzG3t4eDx8+ZDFf8IWf/sIf/EO/hiI6JoY8CALCIGwnKjXBlrT1SGnAwaVePVs2lhBnKLBjURTO5kmWZ4ydyqSDgDRJUUrR6XSaoS04kowLAuOAy8tLRx6UfUwlWNRak2e2vukP/JdWJevg7dB+rS3/Quys1N6LQhjfOWVZuZ1qhsMhZ2dn7OzsWJvf7fHo0SNu3rxJEGjyvCERBXbD9KNnR+zu7nJ2dsbu7u4aShOGoatf2sD1+UjbtXCWRVG4BS20diG+SL1QGFpSd8zznOl0ys2bNx1MmmWZyyIF8rp79y6PHj1qDZZHxAF7w6QFREg1Ps1ZHAq0GLeQE3x6skR1gqtL1KabrWuSJHFGzBjDrVu33Li8yWTCG2+8we3bt0nTlOPjY/b39/n85z/P4eEhw+GQw8ND3n77G+zt7QKKorDTg3q9nos+pT4hkaYPKYhi+LCe3xf4ooubkhO0tPlYK5YNUUIi5jzP0VG7UaxMJvEdqojfvO2TFOSz8q8Ec34rkAQ2fjsJ4KZEZVnmssiT01OOj464c/cuZVlydnZGHMV0u1201o59+v77Dx3TejabkR0eknZSVqsMY2oO9veZz+bcuXvHEUSUagYh1DWz2Zz+lq2xLxdLur2um895eXlp7xGKKIop65LziwvyVUZZVw1h6JIf/dEfZblcuBprFEXEScyv/dqvff8f8g+B+IxnWa9F2Y6789GjupL+7JC6zp1+SpAluiIohNjN1WrlUC6LXFR0OralTfe0I/nkec7Ozs5afV2+Q7gjP/7jP87jx4/p9Xqcn59TliWnp6fcvHmTmzdv8uDhA85OzwjDwDH5xUl/4hOfABTn5yOMsZtQJEnJaDQmCDRBEDbnOmOVZezu7rKzswPA7du3GU/GnJ6c8BM/8RO8/MrL5E2/pKwngYKzPOP09JQoitaIT0HQzII1QGB5Cde+dUQ2Kvb7zqA1Nsvl0vWITSYTZ+hffvnltRF0ElVJxFWWpY18G7KDjKYD3EMTpyk1FMkuh8Ohc85Si/QzSzFgokBi+Nr+obhxSiVFUQKKNF0fsi2Tiba3tx3LdjabMRqNXO1VdnEoy5KTkxOePXvGrVu3uXv3rvu7wLziLEX8AQw+BCuQiU8OetHFDSMIQlQIRbPfozBAxYHZKTxtFi8Dnl3dUQKrqh3oLE7PZzYKEUeCMWkDEQKZZAvyuwRZAi1Jj+R7771HkiScX1zQH/TZ3dllsVjywdkHTbSvmEzGpGnK2dkZQWCHZJjasDXcaoxyQJJ0GW4PKSuLigjTXKHcyDyltIOUZUC8ZD1KKzppB4Oh1+1y+PSJDQzynCiJGY/HbjZt2WwBNhqPGI3GRFG4CdgakcxKAnW/ru2jD5IJ2vcXLRrRzJWV2rvYquVqyXw2Z76Ys1quXOC+Wq1IOzIEIHV2Vtoxtre3mc1mzdzuwJWjptMpSZIwm8145ZVXeOutt5yDll7yr//218lWmRszCrC7u8uDBw8YDoecnJ5weXHpuiF2d3edTXx8eMjWYMslI3FkmbimthDy2dkZr776CT7xiVebmmvpkBxJlAAm4wmdTsdlq+PxmH6/T5LEGCLqZd30VIeuPvs8uRbOUiCB8/NzxxyVRmsxUDK4XBhLe3t7DjL1e3YkoxOH6ff7yBB0wfmF4CLMVTFccly/31EMFbAGzYqSflhTsDjvPF/vf5NFAC38JzWqJEmYTCYcHx87ODgIAu7evWshrSjirbfeIggCB+NKVtLv99fmSPrXdBXek+h1U6+0EoZ2FFhd124nhVor2/hlWAsyarM+fUfm/QqkCu2WSH6kKs9BUAh5TZwksKZrEuDYbeYq1/f28MFDHj9+zO07t3n//Uf89E9/gZOTE6qqYjafOaOwvb3NZDJ2LPFer2eby1Hs3th1E1mGw22Usr2SZ2dnHB+fMJ1OCILABWNJZElyaBgMBigUaScliWJQ7TQera1D7Xa6TCYT7t6+w+n5GWmSMplMuHXrFnVdcXZ2xnA4JE0T11qzkXZuquiH3yvtiwRmYvukX1X2kZQtByXQ9jM0afIvioJut0sURW67Q9FZaeWbzWZ0Oh2WqyVZZuh0usxmlsl/48YNlssl3/rWt5pkwtYhlVKcnp6glObZ0TN63Z4dMjCdua3BhMF6eXlJXuRERcg777zDwa0DC4kGAQ8evMd83u4KtbOzQ5zEdNJOU15Y0Omk1MZu6yVZMo0NPj4+4t69lxvbv7JDE+qK4+Njer0eW1tb3ibZNoALg+e7xGvhLGWoubBUBRoVHFyMTJ7nbG1trRW7JQIXo+TPQtW6nUghqbfUTvxdGaQGKseQ2kqapg4WERjEz379KfnWeRrqut1ZXt4nGH4ct6QNiR7FkYvxjaKI119/nclkwmKxcG0CWmtu3rxJmqb0+30WiwXHx8fMZjO3/2av12tw+2ANiriaRfqN7X6940WWILQZT1XZwCYIbI9lkVtKu913z6C0RtMuTp+wAy1i4Ts/6XuUDFQCKvsMIMuWFEXZtAWVyGi6PMtZNZsj17WdyPP06VOLMBw9Y7FYsL9/k6997WsMt4Z88MEHvP7669S1nUZlakNe2K3iBDmpa8PejT27xraHBM2IuvF4xHvvvefWheioBIgSECRJwic/+Ul+5Vd+hZdeumc3gdZe/ZuAmwcHJJ2U7cU2s4noZ8lytXIku16vz2g0oq5r9vb2uLi8+EE89msnkkn5LUU+AkEzHF/07mqjv1+fXDX3WxKNqqrZ2toi0DbQTpKENE3pdrtuQwlhwQrSMpvNODk5pqpqDg4O+M3f/H+ZTifcvn2bszP7unzXw4ePrFPb3SEMAp4+feZ2YZI66dHRkctaf+uN3+LVT7xKoAOePTtCKcVvf/233fzvTqdDFNl2FkFqxqMxatsSgbKGrCPJTxRGJEnMfD6j3x9w7+V74CUTcWyTnLIonW3t9Xt00g5Jmtgh8uXquc/mWjhLqdNprTk4OHDj7AC3u4g4J6EzC+YseLw/Jk6MmNCLRWlkD8zz83OWyyUnJyeWpPDBIdPprHEkmv6gz87uDnt7e3zqU59qYY8r2aA4PHE8AlPUNWvTKyTL81tUfKajHEcgGIE6/CBA6oyDwYAwDLl58ybT6ZSdnR3SNOX27ds8fPiQKIrY3t52Dla+32fSXYVdNwgYTf0sbAKemizLUaGG2jYrG6OaLbMsASjQes0BSq3I1UI8HWxJGW1A17YD2GHOpnGGElzZjHJFUZR88OgDev0eo9GYp0+fUBQF28Ntlqsl/YHdT/WVV15hNp9xdHTE4eEhvV6fxWJuo/E4RinN5cUl2zvbjoUtgVSeZ81EKlsuCJRy+20Wma1PoRVhHNHt9kjTlL29PS4vL9jdfY26sqxsGU02m8+4ceMmjx9/QJIm9jz7fXq9LhfnF+zv73Pnzm3e+/Z7DLYG5HneMi1fcPF5F7Kbi9gKyTIlEfCJeaIsTLEAACAASURBVD4z1t9OULYg7PV6jsCmlZ27miSp2xhA7KMEclKTNMbw7NkRr776KicnJ4RhzO7uDcaTGePRmLzISZOEx48PuXv3Lu+//5B33n7HzSmuynYDCkFhgiDgG29/g1sHt/j1X/91Pve5z5HlGVVZEoaBG5wBrS2/f/8+h4eH3Lt3j8l0glbabWIxGAzcZtNgAwKloCwKVDPcAHCETOmSmM0sCqOVJoli8rrdIvLD5Fo4y16v50bJSQY1Go1c1iNOUoy8zcLWN8H1hwwHjVPwJ+BLDXC1Wrnmf3HKGrt7tz1eyXg0Yjqbcnh4yOXlJa+99hrb29soY3cOr2oLR4mDs4X4Fuwuy4I4TtYcoDwoiXJgfcC0wC9+S4cMMJDfBTYWRuxLL70EwEsvveRqpNPplPl87qZk+PdNji+QXrvYroUa/EBF4HKt7SbNRtkaXU3VECgaNqKpMXWNgKuif35d2D+m/0z9MYRtG0jtWlJMVbsNlMuyYLlcMRqNuRxdMl/MbXDXtKjs7++7OciDwYDHjw+JotDW2IuSoIneB4MBsgnAYGvA7u4uWtspV2Fg33N5OeLk5ATAXZvPzK3rGrT9/W/9T3+Tv/QX/xJf+MIX+KVf+iVmszk7O9ukndQyYFMLa21tbdm66NYWdXOtSZLwyiuvgLLrNEkTDvYPuLi8IFttRi5C2+YV6MDN+wW7jtM0dR0DPk9CELSqrByJpyxLl7lbPgOOsJXnGf3+gCCw22+dn5+zvb29xra9HF2SxDHjBjqXQPzRo0fcunXAbDbn5OSYfn/AxXyB1povf/nLLqh//PgD0rTjslZBZ0SfDg7scIM4jlmullxe2EEzP/uzP+vmf/f7NhAMgoA333yTwdaAk5MTPvOZzzTBnZ1v/NqPvMawmeBj0cDKZdpBoJEW3rqu2v7QyNo865QNGkXS+ej2pWthJYMgYHd31/0+GNhFnWUZo9HITbzZ3t72mu9Dqip3CiMzN8MwpK4q0sRe+HK5BGxddDabcfj4kLPzM+eEMAaUAYSJuD6n9uHDh1xcXPD6669z59ZtCzslbT+mVeYapQKUCjCm3QpGaNLioPwZmfLQxMj68Jy0wkgwIJCw36As/UJlaYfIyxSg6XTKcDhc+x5xlvI9Uq+VjHYjeNm9vU95VUJdN/tC2m2P6rqmNDVRM1R/fbPaNiCROrs/nNq1ndTtBsjyryWBFa42//jxY7a3h8znM771zfe49/I9vvzlL/Paq6+hA830aMrjx48JgoDR5ciOFFMKrXu2VSXQaK24desWALOZncSztTVsAq2Vi/TzLOf99x8212ENmt/Thlx/bQ3Lz/2xn8Ng+OxnP8tyueT4+Ihew4rNi5yOsTWw09MTu++lVnS7nQaOtvVgyQi01mzv7NAf9Hn65OkP4KlfPxE9kr0b/fU7m83WCGOAY/EXhUWuVtmKuqrdlCaB36WVTqagyVjO0Wjkhr3ItKY8t2Poloulg3GFuJNlGb/z1lt0O12yLKcsR3z5y19mMBiQJLEjCHU6Xdc+J0RNsVdKKZ48eQLYNfX4g8dkWcbe3h5vvPGGa3NaLO2Y0V6vx/379/nGN77BcDh0Pejn5+fs7+8zm84cM1gy8cFg0Ny7wGWWqLYMlcSJCz6yLG/6qoM1guRVuRbOUibrSI+kRNzSVyO1Eqk/ivgGarlctozDpmfHPtDMjZr7+te/zsXZOTXrmRbGHiuO273ifKByNBrx1a9+lZ/+/Be4ffs2cZq4hw4yKk+7tpMgkH3UWvarz5QUR+wzJK/WuMQR7+3tMZ/PqarK7bAiTlh6PafTKXVtd0GROoRsM/byyy+7SUAC39iaXLAZYu1JW39UzmECBKgm4m56WOvazeEUh2fnX37nXpRSn5OATgIV8Cb8VFXDvLV1lCeHTwijkOPjI37nd94iCEImvzPhlVde4eu//XXu37/PvXv3ePNrb6K0QmnF/v4BB/v7nF+cu2zi1q1bhGHEfD6j1+06Y2aDR+Ou98nTp65OWdcGhUVP/GzZ9eTWld0YOo756le/ijGG4+Njbty4wWAwIE1TZrOp01nbOxcSRU1w0Bi0O3fuEMcxW1tDHj54wN6NPT71+qd+MA/+momPLPlkH4EyhUTmN/dfXl462FbmWEdR5KBXYb3mheV8iE6ORiNnB2QKGljOxpMnT5jP55ydndLr9akqO9N6PB6TdlLeeOMNtNbs7u46HkmeFyRpigKyLHf2SgJEsYV2rJ1ec/w2oy6bQMtuudjtdOnspjx7duTGL0qv8Hg8dts5CkkzDENu37ndkHuO2RpucWPvhm0Hq2uShjMi5QcJGGezGZfjEYN68JH8jWvhLAFXixMF8Qky4ljszMGFM/g+lCh1o8ViQRQ0RWrT9io9fvyY89OzhrkXgABpnlcUQ6YDTVGuz2Ss67qJ+LdJOqlr6fBZfK6J2OuTE0ck7/uwuqX/mjDfJMMMArsZtmQdfiO79IXKAIezszMAHj16xOuvv+6mHwl5yd8aqoUpgk12SctwFoemQk1d1hBo1xAOTT+m0i6oAdzsStETn/izWlrCgNST5G/SDC4DLISxenx8zJ27dxwyUNc1o/GIXr/H9vY2YRBycXFhDV4D7y/mc+coy6KkP+gThba+6khe2gaRURhRNvtfLuYLzs/P1gI1vECyrXMbyqrV56IosbveGLe2Pve5zzn2+nxuR/F1Ox2qqrR7hRqbVYruKaXYv3mT2XzG/v6+Q4BedPF5F2IPpZ64XC7ppB2UVq7vXNj/VozLroSz4EoDCqIgcnZLkCXpkRUmdbfT4eTkpBnrmVHXxu1venp6wuPHhy4AD4KA0eiyHVWoaPS63fYwiiJXuxSUTWyyIDlyzXGcUJaFG2NntzW0bSi9Xs/6haqkLOzUHukXzbLMcTfe+p23XBD7yiuvcPTsiPufuE+3Y7dBVLqxyV4poCgK51eUvuaZpe17SRxrVKLarGECykISZyHO9OqoJ59UU5YlZW0Nn2V0ndiFXpaEzSa6SikX7Yphsg9Qu+n5fjPu0dER9+/fZ2dv1zmaTifFmBqt7a20hqBtVJcI0H1f4wR98b8f2sktopB+hiLGXBaB3AcJJOT4Qu1+8OABOzs7bYNutqJyhi9Yy3xeZFmbYqQgimPyKrNtDXVFGEaW9VnZ1gjJIKEdTyaZpY862L14ldNleWaCnBwdHTkn/d5771EUBScnJ4xGlygFQRBy/xOf4M033+Tzn/88WisePHhAURTs3djj7My2ZQy3hm5iihgoMI4ssspWpKpDkVsSQ40dGSbEJOscbaBmaEfs2cBUk+WCtrQO1WdMShuCbXdqxogF6/2AgPu+qqzoD+y2dHEUbxjZjYgOyb0XGyKDAAyGomnPkPfmedaUCYxrV4vj2GWiMp9YOA9S2hG0TgK1bJUxn9maaFEWnF9cMLoc2U2/pxNOT05dtmgnp9lReLJxdJqm9Lo9Li8vHEs2DO0w9aSbOAau6L+9hmYqUGOrgybRgRYy7fV6TKdTXn3tVR5/8Jg7t++wWMz5sdd/jHfffZeiKLhx4wZ1XfHKK6+wXC45uHVAWZZ88MEHPH3ylKqueOXlV6wjjlvExA7RsN8n26E9T66FsxTYUKIRmX4/Ho+9XbXbHkF/E1HrADIXhUnDd9mMILu4uOCdd96x2w0phXZ9kzaqMbTOSo4nNT1hb/kw5TvvvMNwZ5udnZ0m47WzQeu6HY0nG/8a0xbs5dgCJft9ev53Cizqk0GEvCN9l6Lgct1y/7pdO1ZvtVq5nVvu3LnjhiZbpWiHIVjD1yrLiywSlOhAE6mQkJCCgrIqqSpD3AlYZSuSMHL1ZBF/s3L/NXm+eZ6zWq6cQRiPJ8RxxAcffMDFxYXN+A2cnJzwIz/yI3zlK1+h3+81+6F2efTofXq9LtPJhLRj9yVN0oTbt2///+y9yZIkWXYdeHS2efTZYx6yMiOHAqtYZGYVCmQBRTQ+AAviB7BrdvdHYI8FtljhAwpogi1NtgjIREmhIU1hTcixEjF5RPhs5mbmNurYi/vO1WdemQlyAUhQIp5IZoR7mKmpqT69w7nnngvXdXH//huYTi8RRmafVGSfUAqSUNvp6akQ1VwHyIDFfKH7TxnSRaFTHQCUwh3mYcks1RjfNzV3x8XFxYUS8eQ5YH8vzLUpiWxFUQAOEIURqtUqRqORKrO86otBl9qxzEgvuq46R7s9RIIMB65XKnex35r6r/wdyWSz2UxVdATOXKAoxFk8f/5cmNcXI5ycnOjQegCm91JGHvb6PQkMkxSuC/zWb/0W/vf/7f/Av/2Df4vt7R08ffpUvweTGoHpp1rLp9xonueqb2y30wEQCT4DI0ehtC3RWT969Ag3b97EYDCA67nodDp4/PgxAOjEna2tLd3fBwcHcD0XtWoNcIBet2dfecTx6uXPLHlR7WjdjrCAkv1lN+xq5GIiL7ZoMBtlmm5vmsD3UJC4kZWGwoZ/Hc/FaDTSKM2GUqezqfZqCqTAPs+S3BEEHlwXaxGU7Rh5LGbJtgwaIVy+zhYx4HGiKFK2L1/L7ISqGmEYamMyMXqBlV0EQUkRt8kCr/JKU2lKLnKBSZkFAmUvm+/5ajj0YTb31JawY+M17z8dZ5qmOD4+RrVaxempiJ6TpCNOMMXFxQUqlQparTaWyxWePTtAlslot+2dbXz88ceKJDx9+hTvvPOu0OQN/b9SrQAFyjmHJkOMVzF63R4c18F0OkMYBohXKyAvTMDnosgzeG75Hn6/LMsMCzyD5wgBCLkDp8jhQtjnyaq8Lo7nYbVYwPcDcKSc57lwHP/XAg1b7Pv1EpF+OgvHcXTmIlDaAvZNsr3CcRwUmUDdnU5HbST3KO2F7/uYTMaK2s1mU8RxgsePH6Pf7yMvcownY1xeXiqcTgY/7XNu9spysbQY/lJ/hiNOajgcqv1i8M7kQ8oCrooX0JHy+anVayp6Pp/PkWc5vFB0seN4hWazhagiwf6t27fw6OEj9Ho9ySCdQ+zu7GpbYK/XRRAEWK6WGJwPUBTSW9xskPxD+b0YFxcjLBYLbGxsfuW9eSmcpQ030MBcVTthhnXVAfHvNGi2/moQBDg5OVEnm+c5XJSFczpBO3N0HEdaRFwXeZqZ10uE4wBaYyJslGWSnVGAt+yVLKHVq20FtsOnwwTKNg/CDyzU2xqiQAlHc1IDr1EJIzu6QV+8eIGdnR0VnSeMwpqaXXt7lZcdzQPQ4MUWvigj+fI9dm2a0T4dJfVkgXLWqhzfwa9+9Su4rkiMDYYD7O3toVKJcHJ8YhCNBZJEzmV3Z1cclecp+tDtdo36iog/s4YdhREKFDotxfNlH3m+B2Ryv2u1qsj6WUbV3qOO4+o+AkrdXJsnIO8rv3eW5yKCEMeIDBNdxL59PdZ8Ppdaqnm2+Xlker9eAtnbgTEAa3hCKTrAElEQ+PA8CbCrUVWQET9QaLzZbKoK02g0wmw211Yi1so5l3S5XGK5WCoblXYhjEJ4BqKv1aRm2qw14TiOYegWqmNN8RfaZcqEsr8yL3J4KIUX+LwR/UjTFKFRi+JeF2RRmL83b/qo1eqKmty7dxdPnjzFzZs3cH4+wOHhobQr5TkGgyHSVBIIzltttZrIsgz1ekMTLT/wsLu7p9n2V62XwlmWjqc04EzfWaMjnESHYDsSbipCaWSJ0miR7Ufmq03HBkqJKX42nTCXMASFDZiazUrmK1loURSp2kkYBmoIbMkzQih2fyYX/84COKFnJR1ZtU9mKYzGbaPHKfSEc1arFQ4ODrSPyh6FZI/uetWXDe8DZQDCe3Y1++b1taN+u6a8Wq2QJilWsUiIDYdDfPzRx7h2/Rp+9rOf4Z133sHz58/x0Ucf4dq163j8+DHSNMV7772HL774Ap1OG0+ePJF95rlotpo4OztDURR468234LouWq0W4ABhYBG1HAceHZ1TtsTwGRIx60LZ07azZM2oKHLcvXsX3W4X/+2//Td9DZGc9esmI5Mo0lAxNfxarabXyc5wXc+1nK0wERuNxus9aJZty2jP7ERCyi9lEOYZiJb20vc8zOczndiRpsKyns/nWCwX6hwvp5e4uLjQ2uazZ89U/Pz58+dIUrGrlUoFURiVBMFCBkG3O22MTfvdbDbFH//xH+Pk5ATz+VzRPCY9tE9hGErN0HQrMAmyIWYia3AA3yslQ9M0Q7vdwGKxxHK5EslFx4EfBLh1+xaWiwU2NzdV8awwKluTyURHNm5ubmI4HGJzcxOL5RIb/Q3U6zV4npQX1Ed8xXopdigdFB+Yq9GuDQlx+jUjLDsb43Fms5k+oHSudqTMz+T6sgfVbuewX0cYhBAoIBt7Pp9b0Nx6Qzo/jw6fWS+zTpsddlVk4Sp8y2yH/wGluAGvlT1RpVqtqhyeRFsDdQw0ZK8N1a8HbPbekexRxNG1BmkELxgdExJjTejo6FizsLOzM/z1h3+NO3fv4PPPP8ed2/Ln1uYWNjc3sb29BcdxcPfuPTx+/Bi7O7uI4xj377+BTqejjejNZhP37t3D5tYmOt2O1hHl/jkI/ED7Ql13XbuYjFg7Uw6CcC1gcl0PMDWiP/3TP8W/+3f/qzpRu0yxju4YqDrwJQOhRFuWg6rU9vPNMUh0BoEf6HDs16tkyzN71AkgBmWQPVeoM7Szc5lC1NDrS3s0mUwwmUwwHo0xvBB1nOFgCMcBTk5OFFHJslLeMIkTdW58DohmhWGI2VT6E/ksjEYjHB0dYTSSWufF8ALD4RCj0RjD4RDDwRCD84G2n2RZpr2fJPHQplWrVVNKCDXg4zUJwxD1Rl0Z157nGfKSTGDav7aP0WiEhml3WS6XuLi4wHgs3z3LMjx6/EjnbC6WSxQF4LlGQ9xKkq6ul2KX2uQaAGsRhh3Z57n0uNHQA1AZOQDaO0PyDPs2bejMzvQIgV7tJ7sK0wKl0yKUyYySRpPvvdonZTtqOky7XniVXERtRtvJ05kCJfOS52Q7TR7DrvXyutGBUqSecOyXBQWv4mJGWBSFIAgmcy/LAeIAdJSXxYBlxsUa8Wg0wvb2lgZ0P/3pT/GD3/6B7otqrYrxeIK/+Zu/QRiGeP78ORw4mIzHaLfbyPIMg8EA8/kMd+7cwfb2Nna2d7Baxdjc3AIAE3gJ1Z3MV8d1kBpj63mldrLWj0wpAQAq1QrCMFBjJXugZPdeu7aPn//8F/rz1T1it5tQLYhZkONI/2durimdZ8lDcMFxTZtbmyrW8XqV15qoWRAE0jqWlMMklkupF7IWSUiSPAWOqWInwGw2w3w+x2g0wmQ8wXK5QJyIViydn7TuBVpfD0KpLbLdzPM8hJH8vUCh84c5Meqzzz7DYDDEixeHYhdXkkwUxo4VEJ3i1WqF8/NzLBYLFW5nEFatVnViiSRBOZI40WdNhmOIGEi324Xv+7icXAKAJiK1Wk0yR9Of3+11dW+dnZ5hMBig1WphOBxKOS1hwmME6f2vrp2/FDCs7WTKiyIaiXRGtgOgg7jqjPheDi2lKgUp1IA83IRQmdXZvV82uYYOje8jPXpnZ0cNaabGqTw3GiAew2a+AiXUepXIw2ySf+cmYQbD62PretpGhg6a0SA/n/DCfD7XYnin09Ea2GtDVQYVNrnFrnVfdZKspzPyn06n0sZRqaBWqyNJEpycnuD46Bj/7Df+GQ4ODjCbzfDgwQMcHx/jjTfu49GjR5LlJSnu3L2LIPDxySefGHLPEmdnZ+j3+ui0O1itVrh//55+Nkk8/LlaqWq5QXtFnXLwN2uYjmEzMqrf39/H48ePDVu1LBf8h//wf+GP/uiPUKvV1sojXNzTcB3s7e+h3e4gNGxYaQHzUGCdmOIWrj5/QeBrpuu4r0lmXEwUaPcozEL4NIoiIRf6nj6/1WpVW5Om0yk67Q7Ozs4wnYl+K23f6ekplssler0eNjc2EMex6qRWTL8h6+60q57n6kAJCYBES9VxHFXYefHiCAsjeecYUTQXjr7OhQPHwLAUyjg+PsZkIoLsnU4bvi+a1uwbpWwfy1t23TZNUngV2TsbGxvqlIWIJue7ubkJpwCmc1GvGo/HWK1WGA6HuLi4wObW5pr9r9UqcF0g9746f3wpMks6Btsxkd1Jx3LVKfG1hF3pUAgpsWF/fQxL2bAbBMGacLAdKduwkyxxdEki08d5bnR0PP+rMOjV78TMku+jo3JdVyeRA2X2a/dU8nVXh8Da5ByeNzNp3/d1HiO/OyWogiBQxqddn31VFxmtNOYAADZMm21Q1k9SNSwkS8xmM+zs7KDVasF1HZycnGA4HKK/0VejdXFxgZOTU5ycnGBwPsCtW7cwm81wMRrhxYvn6qjfevNNhGGI999/H3v7e8jyDNdvXFcShOd5aiy576JKhDAItZ+O0KwKfRiHJFR/MThhKIa30WiAW51Ixh/+4R+i1WqtBXFAuY8ZXLiuDD/wfU/6+QyM5jjl/pfswYhw5GVdKC9yDUBf70FZ3Ic22kUbZ/dQUki82+2qrbu8lCxrNBop7DqfzzEYDnB6eopKpaK1vrqxN81m0wTyJdmPLWpixxxlXxd52akQx1KTpxMqETKA4gjynyQltVoN+/t7+Jf/8l+YACzHbDYzOtYRGo0G2u0Wbt68qXu60WgoQYx2zR6Dd5UwadveWr0G12SavV5PFaYAaalaGXj20aNHWlogse0r780/wv3+H142DZ8PYBRF2jdoa2sySqIzY6bFh5yGjpus1WqpDqXrelisYpycnePBgwcKAfAi87N5/DLaLTM2ykXZtRw7krezXjuyp2gCWYvMDtmbZkv58dzt7JRZJB8gO7u0z9eGZe2NBUAZttVqFY8fP1aR7ddLmJuuCxRFhjxPkSQrFLkRmCjK1iVOKqC0GGFtjiGaTCZGVGCE3Z1dPHv2DJ999hnOB+e4d+8evvGNbyCJEzx6/Aij0QhZluH27Vu4e/cunjx+gv39fRyfHOPGjZsq67W/v6/ENACGhFOs7RESejyfKIED13XW9kAURYgqIoeYZikWywUqtSq2drYRRCFc30NW5MjzQp+9sqxQZtQ2R+DWrVtSd3RddXrMfIESBfKNcw+jUM+HgaOSR14vvbZ2iUSIeZHWDJntt9ttDZ5U8CITB0bm9WQywfmZKHt1Oh2pdUOGKVerNRWgYK1apuDIfaKtoRgCa5hhGGJpJpscHx+b8y7tN79HmmZIEimZzWYznJ2dI0kS9Ho9NBoyvSZJElSq4sTZ9uL7nqr4uK5h/TslK3u5WsrMS5LNHJgsVsr4juuYGqSv4+koIs/kYD4XAtJwOMTBwYFRtcq+dvjzS+Es7UyLDxAdFZmtQOkgCRXwYW42m2vOgwVrOg/OgfQ8EWwfXQzxox/9CG+++aY5AQdZkcP1AxSOq38WcAEIRR6Og529XWxsbJTMQ5TMWttB2o6Wm53nZTtl112vVdpZp32MqzM7eUz+/SoUbWfh9nkURYFGowEAKroexzGm0+k/5u39n2LZg3LtrJ1jjgiHkenKSJYRK6Gj5XKJhw8fYm9vH3/30d+h0+lguVzinXfewe3bt/H//Kf/pI7T931ERgHl2bPn6HQ7uH3rNobDC9y4cR1xHK+J4vu+QJeOGU4tZDcPvlE9kaZxX2uWsu987VtjQFeJKuh1ewgMFNvr9rCzsyNBXFA6MNvRuu56JksjurO9A9c4bD6/YRgijELtn5SpLaUYRhiI0Q38cuTeVZbtq7ps/oWNLvkmIKH0peeX3AtqQwPA5eQSl5eXGI1lsPJ8Pke1VtXko16ra99kq9UCChhot23kOuU+RGGkNUruU6o1zWYznJ6c4vPPPxebk62PniuTjAJMNESLdownT56gXq9rwuC6roj/uy6WS3mu6vUGgjBAEPjqoClgURQFcmPfaG/Jmg3CEJ7nG9KPgyAsEcubN2/quDra05MTadM6OjrCp598qsjIV62XwlnS6S0WizW4klEMHWF50Ucqgccvx8iBF5A/u66Lra0t7O/vI8syHB0e4tnBgUbFrDsBkl0QinAcB65XOu0wCtFut42qSlULyryZdHSM9OkU7d5Hwp+lU3QsY1Q6X/4738NaLL+b3ZtJI2PXbO1snMdh36DtCJrNJgC8dpYAGFlTmo0PPSXEbIPOWgfr1qvVStmL8/kcvW4Pq5XUhmq1Gq5du4a7d+6iKAocn4gowb179xBFEXpGOjEIfNy4cdOItAdwHBe1Wg21Wk32gNFiJXQJwKiNmH5Kwyx1nLK1SPZOpvvY8zzp44Mo6FSrFYSBEEW2t7dlXwc+HM/V/3IUKEzWCteB47mA68ALfNy9Lw6/VqtqhuO6MjPRRjwcB0qcYCuM53kKDWv98/VSFMouDQFQ8goDniiUwIPTk+JEWNijkYxb46xJz/PQqDfURoncXGICGwmAaNdYL82yTGeT9vt9dW5k4ALAeDxWJynayemaveE5i2g6Fc1yHB+fYDyZrLUDVioVeL5MUhqPxwCkPaVarak9DMJArw9M/ZTPJB287/soUCDwfcMS9xCGsi+TJMHNWzeVQFcUhQq4OI6I2Tx5/ASr1VePinspnCWjWKl1lOOrWFcs57BJ9P/ixQscHR2taXBebZlg5FutVlGpVPDmm28iDAO02k38m//ld9HpdPBnf/ZnMtRWH9YCWSYzDUVoPYfjAZ7noN/va/Qts9TiNcfFjQhAIRG7rspi/FV2K49h4+98WHht6DDteiozWbvm6Hme1awcrDl+jpTicUgiqNVqr6XGYMGFnr9GcLAjzclkIlGuqXVfJQV9/tnnCMMQu3u7+OKLL3Dj+g38/Gc/x3e+8x0UkAjW931897vfVah2Op0iCiPcvXsXYRggzwv0+31UoghBIEObeX68l1x2rcaBgzCSnrnAD9b2CQAjVScwU2YJaMARdqPv+7i2f81IpXnaHE9nR7KO1MFruHPnLnZ3d01fZYF6MV7nDgAAIABJREFUra77mjMtuS/pIBkIMwAG2E/ovyaZmWVLJ7I043meEhVZSrEnEC2XMpZrsVzg/PxcM0pm7Zw4AwgisWUGbWeG3c1EZD6fi6iBCXyIUNCucA9eXl7qax2IvitttmSX0EBT2lhKp0aSjd3hwF743IyGc83ra3UZ9+Z6oh9MkpEjm1adbRDK4OciL4woB6zg1lFf0Kg3cO3aNfS6Pezv76vPSdIEk8kUz58/x3/9//7rV96blwL7IExJlQdS8emAWNcjHbher6+RX+yIhjU4GjqyyqrVKh48eIBPPvlERKj7/TW1lTXmoFVvpL7qzs6OZrgkzQDrIuh2rbEoyl5GOkp+Fzt75DEIufBn/psNSbP4z+Oz4M/38DVkePHfbVYngDXd2zWD+govmwHNGrIEGzkWiyWCwEen01EnRMiW2plZluH73/8+fvHLX2BjYwPf+ta38KMf/Qh/8Ad/AMdxMDdswbfeegur1Qrf/OY38bd/+7cYjUa4du2a3qvJZCyEG1fEBYo8RwaIA3NKJKLcb4ZxHQQiUGAkyPzCX4PrfU+i7SIv4PkyrJx71fM8JHGCTrcDP3ARxykePXqEeBXDizw9juM42N7ext7uHirVimbXrusizVLNfknOYJRuv5+6pQ7KNi2OZnq9ymeT+qd0iNx3cbxSZZtOp4Pj42MRGbi8xIvnL0QDGyUZLYxCg9JVNdCLKpGqLXW7Xcznc3VenuvBq3C84RJJmiJLiZxISWK5XGK1kr5M1zCaYc5P7qODLEsBFPB9m61vhqXnuda0kySRfZMXWC1XGtjLORpyo+shd3INUl3XRRAGipJ4jofClc9exSt5fSh7jIQ8BmxbW1tCqjOykizz3bt3T9tgvmq9FM7Sll+jsSchhiNlqPPq+z7u37+vQ46LolC1HmLRTLHTNFXHKlJJN7Gzs4Nnz57h008/RZJxQK8PFwUy08cJlD2b+/v72N/fR6vV0hvF/iab3Wo7bht6ZSBAFi6/G42KTQxiJE8olf1nNixjk37o/GjoCbWy4fdq/c2GsmngR6PRa3IFSlIVEQwGKGz2pqapPf/z4OAA3W4Xm5ubOD09xePHj/HgwQP8+//z3yPNUvz+7/8+FouFCtn/+Mc/xgcffIBWq4WPP/oYe3t7+I3f+A0URbGmeNJqtQA4Zj8WiPyyLs4H335eSkZ0oft9PB5LtlmRWiUFqeEI8YfPWVGI0Ee9XjcBQgHX9ZUhK1JpEoDGscxD5HPH8/U9X3tTK1FFyReLxQKe56nTrNVqmlW6ngvfLYO915mlLO6FRqOhk1y4F3k/XNfFfD7DYCB9g7/61a9wMRor07jVamnvpWSVkSJsrusgjERWcWtrA0HgI0ml2b9aqyJexXBcIdmkmY/FfKGJDO3aYjFHksRGCrQU8RD7IkgcVNKuUOINA/08z7GKV0hS+fns9Aw72zvae0lUzPcDJMk6emiXEzhTFpApKYSny97JkrNB6b8wDPDgwQMcHh7iYnSB6eUUrVYLB88OcP36dVWe+rL1UjhLW2aMRoFYdlEUOlHD7k3sdrvqrK7W/Wy5N6CU/GLGsL29jaIocHh4iMnkUi6oI0NvXd/H1tYWarWaMml5AVmrpMEC1gUVbKdn051Z16RDt/s+bedqE3jsehijfxuq5Yaysw1uJLsGZGvr2lkvM19Gcq8X9N7y+jIb4pxG3iMK9ff7fYGzF0usViv0+j0cHR0hqkT41tvfApnPaZri2bPnaLfb2N/fh+/7ePjwobBcIXqgSZLo6CVBMEoGtdw3kaHj3rbRD95bZnpRVEEQzJEmZf8wYTJ7b3JveL4PQCL7ZqsFzwtMRO+g1W5jPl+o6EK73TFZaal32+v1gBRwHIn4kzgBB3rZwWAYhEpU4fPI7/h6lYtBEQMa1hn5rBeFQJfD4RBPnjw2A5qbCMyosygK0WxKC1OlUkEQBkq+AqTeuLm5Cd93sVwuBNYEdO6oVxg0IZUgPcuztV7vLCuDdEG7Ug02bdsj+0wGqovNKQdkZEaQvVqtKneC9hIwrNY12wqVsvMsyUTaXccBlqaFxXUdBAGHYMi/LxYLndbj+R62tjZxPjjXyU29bg8Lw5D9qvVSOMvlcqlfnLADHQQdC2despeIDsxmo/I45aYq9GEMggBxvFQY99q1a9jd3cNweIHZbAoYh9XpdJDmmW6udrutx+LDbTseGiO7RkOjyoI5GVj1ev3X+hptR2dnpHa9woZo+fNVQoTtBG1yT0lUyfV87azgdZ+lLN8PFKWAIwy/IBB2HQWnPd/DZDLR+q+9T6fTKTzPw0cffYT33vsmdnd3kKYpJpMJ+v0+fvKTn+C3f/u3EYYhzs+FDVutVrFYLIQ5ijpmmGkDuJyTIW+ZNpA8L6F52zhdrVtXKmVrEhEFwvCeWyqmiOMtIPV5B51uB4HvI8tzVKs11OtVJHGi31O4A6aNxpqD6fs+8ixH4ZSOkVMlZC/KtRTVmXJsGL8jz+f1YqBcKpApwdC6n/P5HJPJJZ4/f4HJZIZazfQQGpibmSDHU5EMtFwuUa1W0e/1RboRhWEyC0GokGKjCaQTtZ/kiggJKC1LPAVVykopT+rRUg6UdklsW2mX2Y6VpqmO+/I8T2roRYHZbI7QsKfTLFWH7Ac+XKuzgG0jReHDSziBaZ0wxqA1SRJ4rqvCBVubW4iiCJPJBLdu38LgfKCO+8vWS+EsScfnA8SoE4A6m3q9jl6vZ+o6E4zH47UiN4C1bIBYNPF41qIqlQDz+Vwj9O3tTfj+NuJY5KSEfu+t3WjXddVQ2H1tZOzSAX2ZuHun0xHShWHB2k7RrrUyo+H7yv42Zw2SBUrHyoyBDxGXDSfa350GVfUfrcb6V31Np6InHFWF7bdaLH8tE78wEl/dbldhU4HEpN/y6OgI9+/fx/7+Ps7Pz1GtitrTX/7lX+Jf/+sfYG9vD6vlCqvVSlRYXA/NRlOJDaTyc1IIYKJrV+j14vBKByXKJWVPI7PSPM91cgQA4xChCIod8Pm+B9cNkKUSDFB/U/ZODtfzkKYZ5it5Zoq8gBcIqYKQse/5cCIHi6XMx1wsF7r/5PMzVKs11Oo1JfG5xmjlRa5liteLfAVf7VCJpnmI49RklRm++OLvcXZ2biERxnGYpMNmvxL6rtfrImNnJogkyQqZkxm2KuvNnkHzlvCt0VxCPhOh/W63axxUGaDV63X8yZ/8CX7v934Pd+7csWrrpTKU+Ybms1DWZZNyok+e5ULmcUUBKC1SZVf7vjB0a9Waoihplpo+3Qgzwwuw9WzjJNZAUWx4KciyubUJzxNRDc/1sLezuzbL9ep6KZgdjUZDafaMRolfk1larVaVEeY4jmodrlYrbQxnvZKanHQwVLJhrbFerystnxFzo9FAs9lUyTAAa9lXiZc7mnWyJmkrDNnwGOtdzEJs9ixQEnnszIAGmrUoG0Yjc8x24vwdUPY5ldMJVhoV2tkknSu1Jl9nluWUB+lfLGt63IOLxQJhGGJrawtFUawxEw8PD9HtdvGdf/4dqcGcnaLTaSPwhZR2eHiIvd1dJEmCX/3qV7hz5444B5nuJs7P9EpWqhUtP/Besf2Dqj2OY02gQaGtGBQHkHJBgMDn9J1UW2DEscoHM3hj9L5aiXTfbDbT71aJKnA9cWyuMTRpIr9necE1GrUAEK9iZXJyz4VhaNi9gRo8Om6gLL28Xuu95iT7RFEEz/VQq1V1L1arVbz/wfvo9/varxiFkSYa3J9EHvzA11qfDW+yZYr7jy0ZHglhyqEwPAnj+EjmkoArQKPRwAcffICLi4s16F0jOWvRTvK/Wq2m+3tlxA5cp6zF+0HZuhVFAqPyewV+6dApyOE40mPJGqZ8H9ckZC6S1KB2LtBqtcVGez6azQaylx2G5aZgFklDT6YqIDetUqmg0WhgNpuZCd9L1Utk9lStVtcyN24McXqFOqtKpYLlcmmIDStlG9qZBB0ijQqjX4pGc1MzyyDF2q5NcjIAUCro2FCrDeHSQdqEJ6AUi7ennFB44Wr/JFA207MeacN6AkfHa1MNXpMryog8y3NkmTDvqEXp+z6Wq5VOKiAr2XVdHB8dYzwe4+2338boYqT36OjoSLWN7969i+Vqifl8jv5GX2ovUUUfeEAMQpqmCsO2Wi0sFmULkuwHqc24jgvH843eqxgk35efAzcwwWUFWSazV6UNJEccy+8dQ7pI4gR+4GvGkOcZwiA0xkTgXDa/+74vKiecvJJYLQ5xonUt9rzFcaLPVKPRQK1e02e7COz5rl8jmfIKrqIohUPs9rPlaqnP7Hg8wg9/+EN8+OGHKAqqe7maVARhSSbU3utchii7bhmoC5QZKKTpeZ5K2tFWtdttzOczZaJSAcwz2r/M2mazGYIgwH/8j/+3fhdRfUqNNAHM/wukOUUWHOzv74utzstxbSVpMwXgIF7Fit5R2IJi/LJ3xRFyeg3bS4Iw0GHui8UC7U4bAWv+bgHHDeG6PpIkxfjyAlmWoVZ/yQk+YWgLjYfqGOkgqAkICKwJCJR4cnKy5qwAcSzM4gCoU2P0ypter9etOkmBovh1DVc+6HRmdKBkddmtHFmWaRbcbDbRaDQU/rha07SjKmYQAB8OCDyS5VitYmUS2gLH9vcFSoKHnXnSEbO+QIh4NpvpZ9rO4FVfHLHVaLcwN5nVGus5FKgyNY4rTVN8+smnuHP3Dnb3djGfzzEej9FqtdBqtbCxsYHLy0t89Hcf4fu/9X0sFgtcXl6KgtRohF6vpwiGbdR4L9iSQYKb54kqSZbncP1ylJPrAK5jyA6+Kw7QLzWBibLY+032uwhtEGEgiSg3TlngWRezmVwXx3UQ+ZEaUdvB5ijVfpKEhi1HlgH1el0HEPu+D98rSxGixel8WfLxyi67nhv4IjPneR7G4zHOz88xm03x7rvv4cMP/wuOjo5FvzdNAIMolfNqS04Fr7VdspFykcCuJXkwh+exfu/ovhFbLOP9GrU6Qj/A5sYGTk5OFNIXof83AEhvpbSVmLY3x0EBIM8yUAihyHM0Wm3cu38PWSY6xdVKSXwUe5sjSUpGfxgGqNVrqFVrSNNMEY8kSY1cXWqepaysW5qyAxWvHCPwDqeA43jwPeltrlSkVMYe5C+9N/94t/2/f63X8UrNSDq8MAyxXC7VWTFq6na7ePbsGU5OTjSqAsTw0QAlSaL9hHQQnU4Hg8HAqgcmiKKKZmw25RgoWzKCgLUVV2uAdF72fEjWCZiJMmMs2WTZGvHGJvkAos3JTcKpFvwuNhHCJjLxOMTjGfHxMxk0iHBxuAYtXyULvYqrFJQulDym5JVchhsXZkDsfD7HX//1X+N73/seDg8Psbu7J861EmFnR4g9s+kMv/jFL9BsNXF6cgrXc3F2di6SY2Z233Q6VdIXUAY9sl9yDeiYWbquq7qXmmW6JTMwz3PNBAkp1+t1xHGC2WyKIAgRr2LAgRIxbCiWy/dL5ILi0noO1jxZIfoYVMMpW8DyPFct22q1qixyW4WKx3BdX6ZZvC4FAMCatCfnNlYrQgR79uwZgiDAT3/6U4zHEw38wyDEYrFArVY37GZPYVPaUt5j/o73e7lcaSnKdWU4OO8LSZNs/K9Vq1KHdhzsX9vH8OJCVIJcb43db5eiXFdUoIByJGKayRD6b37zm9KZkBcmKBCijtojU/emHdvY3DCwqqOEIY4jcyDQMAwszISBDjOqRNp2k2YpjIgVCjO9Z8NMYXH9r0bZXgpnyQdT6n+lsgcAJcyw9sdI2fM8dDodXL9+HQcHBzg8PMSjR4+sqCnRSJ1OYXNzE/fv38fPf/7ztVmUMvR0pUaBxiZNBcYq4VjXPPBSVGfN0+7vbLfbKtxrGz/eONYhlUJN4+I4SJMEaRpjOp1pb6ktxweUij9cNJIsajuOY3rj5HqyN5WbmBAPoVsGAq/6IhO7UqliOZuv9frO53PkALI0xS9/+UtsbW3he9/7HhzHQa/Xw3K50EwwjmPtdXz77bfxV3/1V3jrrbcwGAzwzjtvg8pJ8SqWkoFRUQnCQCfRi7hAYCCvmtZO7RYjigoAJtjMcpWPs5ERJY8BWMWCWDjmPJkRMOMoigJOIDT9Eu3ILMgUYpwKmfuXF7lkNRAiR5mhipNk/VWGSq8LeFwlrNG4vuqLAUUQBBgMBvA8DyenJzg8PMTxybEQdYy+KwMpKnFxD7J2TYTNlv7kvQZKQhiV0hqNhiH/leUrG0nwgwBJmorKTxhiZ3cHL56/MExViqsUYGuRfAiMY5K/ZJkEdG+++aa2/+VFjtAP4ToOKhVP9xHrohzVVavWjDSejzRJDXmIHJVU6rHsAEA5aMN1XGWB2yieSEVCnwN7n3/ZeimcJS+GQARlsZ+GnJE+Yc4wFFkvOpF2u423334bi8UC4/FYxX6XyyWePn2qDbUHBwf47ne/ix/84Af48z//c4Ug+Xl23xchz+VyqXCc5zmYzxea/U4mE3ieh263i2aziWq1WiqUmIyPTel08nZbx2w2U6knWzSAN5lZtD1FHCjZjGz45iawHSkzonq9bjHTyiCgJI/kX6ta8aos13MROAHsLIsUeM/zcHJ2iuVsjvfeew+1Wg2r1QrPnj1Tws/5+Tlu3bqlWfzjx4+xs7Mrs08N2YakC0Jhg/MBsjxDr9dDpVLBaDTC/v4+tra3cHx0rM4wMqLkGtylGQoUFhxKaKscM2cbRzrPyXiCy+klkDFIEpUVYD1DTdMSCrPrqtx/eZZjmS1L0liaoYDJKvIC8IB6o67jo1Tlx1KVstEMfd/rhUqlgul0imqlqm0is9kMz58/k+vsF2ZObxPVWlWl72z7wCSBKJdNoKJzoe3zfWHedjodjMdjtQ9SzhKJOgocyBzUALFpJ9rY2EAYhnj0949AXe2iyJBlbIkTR+i5HGnooVKNsLO7g7t376LZbABwhPGaS3tRtVbF6EL0bleLlZ5rvV5HrS6Bo+u4ylotQL1kGHQj0azaRibZeuO4jlzHQjqBizQ1/fWeHCt7yZ1lp9MxD2WhjDz7xjNro5QXyTGl0oOv0CMZi/z5zp07WC6XGA6HGA6H+PTTT/H+++/r8a8O+LXZqnRAJMEkSa7/HsexjsuhuDphEQ5tZQTNLBKQm8h+UKVDm2jSFl2gMeH3Y6YYRZE6OmlAjjSjsclDNtPW7tHkd6Oj9Dzva1UrXpXFOY9ZXLI4Pc/DYDBEmibY2dpCdC1am3jT72+oYfKNUcnzHOfn53jjjTewWCzw4MEDTC4ncFwHp6en2Nvb00xge2dbRQuKokCv18N8Psfh80NsbW9hOByq3jBVXSjazkkNniu9aWkikyiKzG4IL9uZCPdKz1qCxcLUIq29TlIPYTE6sCRJlJ2YphnYL2fPkOUzQjZuq9VCEITg1BF+lk2643nmea5s2ld9JbGgaMvVUmvj0+kURSGiKAJfp5gvFto/6XlSt4usvkReZ2aptJGsict+KJ0m0QDXc4FMnFAQhsjSTPd1EAbwPOF9tNstjCdj3LlzC6PhBcbj8VoAfjUgcj0X21tbuHvvLppmhmqWleQ1JkFU31ksFvpc+IEoSgV+YCD7rCRDJiX0v1wuTVYtryGtKKpEOmgcgLGXK6xWsX5/Eh9f+swSgCHylD2StsMgXOm6MolhPp+jVqvpBHFSirkR7IiYzaj9fl8dyunpKb7//e+rUWR7AMcwTadTzdg8z9PIDSjVc/iZ9XpdMX+7LkDI7GokzXOynbXneVpLpO6jUvpNP5TvB5rR2KxaZqN8kKife5WKT0fM8+J3AfBrr30V1+XlpQ7GjeMYo9EIvu+j02lrhG5rCGeZzBgEhETGZuo8z7XF6dGjR7h58xam00tsbGwgz3M8efIEvV5vDea3a6NBECCqRPrQ3rx5A/O5CGRzMTNdrVa4du0anj59apyNPRQAWiLgz0RjlovSENvtREEQ6t7j/tOe3Cwx50wlGRkgneXls+m6rpJ5ONXCNcN/HaxP8fHcMnt/vcpVoIDrCFJA+3Z6dqptD0XhiMA+gKLIURSOyhGqnSnW52JyzzLwYpYv0obrUKwDM3KrgPY5Oo4Dz/fgpg4KpzDtTgVq1SqqlQq+8Y1v4OnTpxgOL5Ak5f4pigK+J45uY2MDDx480Lo7gLK04LkII7GflOtcrpYawDYaDZlsY9itju8hSRNNcCSjzFSJiGUK17TKhIFklRRbn81miGNhFtNuy/XA16JsL4WzZMYoTmOuDzixZGZGbHlgDbDRaGiPJet0NEJ2NkZmLR0goQtGx4QmSCKigbAFD+r1+tpYMLseaRMWaAwUsjKO0+7DZI2QEXYcx2g0GphMJmuGU96XGGhvrufDiIvHY32CRW1m3Pwcfn9ujOVyqe0vrO++6ov3hfXBZrP5a86G93o2m6ncHR2FrbBC8tfh4SHeeustRFGoGWEYhhiaSJxOkyw8EtBYL+31etjd3cX5+QDVqsByT58+1T0QBAHG4wm6nS4uRhcqss29IYFkgDzPUJi5hZ12B/NwDseV78FA1CZoZFmq10EzEs9H5mZrAZfWtVwPrudq/zJnx9ozNHls/pxkiT5Hr+uV5fLNmDURIChwMbrA8dGx2EAHqoRE4h5rb5ubm6jXGxiPx8rxsO2RYwVzdJZAbqBWV+ePVioVxEkMH76ieWmaaq8jEGCxWGIymUgi4Drob27A8Vzs7O3i9PRU93mtVsPOzjY2NjYxGo1UkSjPc1ViSxJB6BxAVYWWywWyNNXgsRJFCEPJDlnmAFDWNbMyKODz5zjCsq7Vq1I/dUTOjy1cLH/5voc8lwRnuVoi/poRXS+Fs4yiKiqVmhG7bQGQKPjy8lIjerLEbBk4GpaylumbrK9mxik1le3FdpHlcolKpYL5fL4GmZIiTYNHY2HDRuX07jIT46YFYG3McgrJYrHQmieANWNKkgOzw16vh9lspptEXldR4whINMaC/VXII8syba2hiLWdRfJ7sReVGe5rNqzUvZlVdbtdffhsWIakC05fB6BOod1ua33T932cnp5iZ3sHL168wMbGBhxIkLa9va3ONUkSzGYzANB7QuFsDpQ+Px/g8PAF7ty9i0a9gSdPnmjQw2eDTPA333wTz58/x2KxQKVSMVkrFbGomRmIVmgYaS8pn4FWq6Ww6OXlpcL7eZ4jKRLdQyWc6qp8XaPRQLVSFcjLXBNqD3MPsq3LvrYl4vJ6DwKAzH8UR3l5eYlnB8+kZteoIzawYZImRtgB5p6GcFwX3/nn38Hf/L9/o0GQTeyTcpOrSURZkhH1m1azhSyVAG25WmJh9IB9X2ZEUlCdTftRFOnMS9SF+NUu2tje3tae3Fazidiwpbe2ttQe0vZp9un7mM3mGrTNFwsslxKo7e3todFoyCg4rE+XkmPkSC2bSrso5QAhl7me9HGmaYqVOe5qtUK324FvBpADkrSl1a9OHF4KZ0lijR2JMnon05NG/SoFnQVqudArNTJ0UK1WA1mWoV6vYzQaoVarGJhWVCocp4RtCW0SA+dxgZIcw4efTtPuA7UjaMK/xMLZbD6fzxVqtaM/GiVOG7DJOnbk7Tjsnyp79ABoPYKGmJ8FlMxinqvdV0nH/Kov+3rTiJNpyP3G+3p2doZWq6WBjdD2axr4yLVfoNFsoNVsiTFYzdZ6H58+fYqTk1M0GnVcv35d2ytoSD777DN84xvfwIsXL5CmKZ4/e45ms4kgCNDtdjEYDABAIc88z3F6eiqtQ4slrl+/jhcvXmggyYkU8SpGFAocVwCIwgjL1RJhUJYRqIesrMQsNyQK1wSqMuIrDEJUqlVR5wkDhFGopQcGY5oF5KXR9jwPeSYC3cwCpMb0etkQ+2QyQZqVvZO+L5NcGnWxaf1+H+PJGI4DTC+ngCOMZ2A9YCe3AYAVnORrgXIQBGg0RcWsPWupUlCtXoMDgWEdCIrgB1Kf52e5jotGvYH5fK62q16rmT7clQ5uns6m0qtMopdbSt8lSazIYRInihZGUaStgpTp4/nLn6n0fzgOPE9qoxxFRt1YP/CxWsp4sQIF0iw1ZStDhgx8JfZ8XdD2UjjLq8VoOskgCBQqJK2ZUSmp7/1+X1lclMkTAezAQEwparWK1hYpx7S5uYnpdIqLiws9Hut+QImn26m9Ha1x2f2UhLSY5VEOjM27hIonkwk6nY7WSqvVqo4iK5X1vbWNXVLt2QtaziPkxmdWYDtFkjAYkdkkH1um71VfDLxsJIEwFBeDmk6no61LjiNaxdQp5nEajbpMd9jaxHK5RLPR1NYOOtt33nlHnYktRhBFEe7duwc/KB/PxWKBh3//EHlhNGRRBplJkqBer2NzcxNPnz5Fr9/D4eEh5vM5NjY2cP/+ffz4xz+W7+a6KHJCx4E0ifseKpEEkUma4PLycj1gC8uyAlm3UaWCIPCN2EBJtANgZY7rwv6SWSxRNQOjBVKLEYZf3wz+Ki0+u1kmY9bilYxFy7JUn11mlvP5HPVaHdPZFL/9g+9iMBhgNp2h1+thsVisIQGUqbPlMlnq4c+B4UXU6w00GyKisVwsNQDnkHAifTwfxxFdYCYEUj7K1V6naYrYj6XeaCF0QSCDnZeLJZIkxmLBVr4MjUYDW1tbaLdbKIqSxctAtkTVPPiB1OtZs7Q5L67rYj6bay3S9zypx7qutpfY5YEg+GqX+FI4S8JJADTSXK1WinHzprdaLTQa4nCOjo7gONLEX61GehP6/T7m8zkuL2X0FrX/JHsM0Wy2AQBJkqFebwJwMZlMEIah1qmGw6FCFWSKJYkYEQ4A5iYDsAaf2RJ23EhbW1s4Pj5GvV5Hu91GGIaYzWZotVpKVOJ8znq9juFwqNEg4ToJIuaGjFE6bkZYjByprcjrwZoD6wh0prVaDUmcrEX/r/Ji7REotUptyJ3kAM5y5D2ezWbC1LNEp8/OznQSe1EUSsw6OjzYKBthAAAgAElEQVQCAPT6PZ2RyQd/sVgY8fUabt26gcVC6oTT6RSPHz9Bp9MWTdZqBR9//DEuLi7w7W9/W5EXatCy1YgzDPM8x8OHD9HtdhEEASaTCVDI/L/ZbAbXLYWn+/0+7t69i08+/gRHx0e6V7jUYbqemXnor8FfzGBd10VWlEGYGuy8QBgEmEwu0et1AaAsKdRfSy4CUmaaTqe4vLxU0QpmVmdnZ6hWqspxCIJAByf/5//yn6VGV6tpSxtLVpRCtNXGgJIxTSdEZItMVBLQ4iSGu3CVKGNP20lSaYerG0GE8Xis+4H2hlyAMChrrYA8Z8uFZLDL1VJH1G1s9LG5uYVOp4MgCFEU5VCBOC73VRAGEH6Po6x+OnE6wDJZkEHRKBzUG3UksVwLssSLQLgccfySs2FplMIwVAk2e+gpWzLiOFb2IG8YABUycF1XWbE0QIy+i6LAxcUFOI2EG6Tb7WI6nWpWGoYhOp2O9kQya6RDJLGCkUq9Xsd0OlUnHwQBTk9PFTIjk3JjYwPT6VSl8MbjsWYujLKn06nOdyNxSTdlkiCKQgSBjzhOf43xyk3I1hKyHa/WLXkti2JdCupVX8yiSHQpm/+NAzXalWTMlpFoKVFnE78cR8Qh8lxkwpI4QbfXVZk7u25HHWHWDIfDC60ZpkmK3d0djMdj7O3tqUF68OABsizD+fk5Njc3Ncrf3d1FEAR4+vSpcYJCoa9WqkizxFD2M2ABNBtNVd4BhAj27NkzHJ+cGNKFY2pUhh3reljFK80MiNb4vowxi+MEnMNpw3u+72M2naHdacP3fbTaLTx79gz7+/v6DHP+5Su/rmRRdTOp5fLyUjMmji3k6wBoZq4/m+CPNW1JPqRFzM4mxYGU5LVqtYJWqyXZYBJjPBkjSzP4UZlRkljjG7H1NE1ViIPH4evCKFwrY4iMooy6I8mG4iue75nWv210Oh0jNZoDKMUC7PZB9nDymtRqVVPHlWc2zUo7mZsxcWmWqtg/UJI4mQV/HX/jpXCWvu/rRWCrhd6UvBzPIo60fLg9L4DnBXCcUtKNhBoq/rD3aDwe4/bt26jValgul3jx4gXyPMd8Psfm5ibG47E64KIolJ1qU/r5O2ZmjFy4GQgNNBoNDAYD9Ho9rSe12239DhzZZWevVP1hTZNOmNl1vV7HarUw16ZUYGENlVkNF8+JDtNuO2G2Scmz1/T9sm5tDyL3PBec8MEM0WYUknoOlLP5oihSeJbCzo1GA5VKFVFUvvaqEAZ/Xi6WGAwGeP+D9/H0yVMcHh6iWq2i3W6vTT65vJyi1+tid3cXRVHg7//+V3jzzbfQ7/ewu7eH58+fG+RFiDie78FxCxX9yPMcTkWyvTzPdNBvGIbY3t5CluVI4hiuEei2oXuJ6BOdScie5igMVW6Pht3zPBwdHamjZ1Bybf8aBsMBtra21lCaV30laYIkTXB+fg7Pc7FYLOGaTF6cmUjfxUkMEQIQ+TbXEeKPs3Bw69YtnJ6eIEkoSCHyiMvlCp6fqj6vrfJEJ+F5vkkKRBx/dDHCIB5IcOc6SLNUyEV5gcItNCHgfWW90HVF1i6OY5l1aqEx0+lU+nwdB7P5zMyOFVu9u7uLjY0NLYl4vo8iT+F5bplVGp6J53pwPEdFaoSD4qjIzWq1wnw5lxFyDjTAKOVLgzJgADQT/qr1UjhLGm0aLNbvmFUx8pCbAGSZiODW63W0Wm2VaGKUQoYh+8fIwuMNiKIIW1tbGI/H+rntdlujc8JG7OfM8xzT6RSO42Bzc1NrV3TwpYCvGJVWq6WKF0EQKOTGKSRxHKPZbGr2RyfNrJctL4zwSASS83PhOIUek+fKTIXZDmFaoKSMK+RQlBPN+fpXfdmBh81qvlqnvhqRlvR9ediOj4/RbDaVwMCxc3QIDMYYGAKy/2ezGU5PT7G/v4/7rfs4PTnF+fk59vb28OTJEywWC1y7dk3h+yiKcHp6ikajgZ/97Gd4++23TEZ5gNFojC+++AK3bt1Ap9NFtyuw7+PHD3Fw8Bzdblc5AFEkhsOJHFUs4j5xnIZ+V2YGeTXX7641cDgK8WZ5hiiU8gHnVY5GI+zu7uk19X0fqADXr1/HYrHAjRs38OGHH/6T3u+XdS2XS5yfnePg4EDa29xCmfsSMK9QFBLIzJMUUSTXeTqdSq0wzfD555+LzTDkM8/3dFyb67p6j4iUVavVch87MCIWDbiuh9VSkITBcID5bI4llki8BJ7voVqpqgNcrZYqjCBJhOnd9QPkrqiETWdTk9VlWMWxZNG5JCbtdhsbGxvo9rqo1+rIskSFahw3QJZnyPMEjgO4joN6s6VJgt3P7gemrx2FjulK0xQXFxdiS6sVBH5g/vSFSew4iCIpD1BA48vWS+Es43hpnBkVKXyrhUPmuC0WCwM7lKo+QRDB83zUDTuM8ARrMPZDzj43svQ2Nzfh+z7G47E6HTvCtfsQGbkAZYpPh2k7dRoPbkAAqi1r9+sx7beNMJ2t3TPJWmOZefhwnAJFker3tVU5bMUenpPdV2VHkWUkWRJ+XuXFLJzLzva4XFcyNKrnMPCwrzuzKWaZdA4245rICe8bZc1u3rypLT8vXrxAEASYTqfY2dnB1taWwlZ06J1OR+tb5+fnWK0SrVN2u1ITHI/HqJsJONvbWzg+PsVkMsHOzg6azSbqjTqGg6EGUKxf8/xlxFEOx5HAirVv+/q4nofR6EIn0NOAMRPu9Xo4Pz9TsQa2biVJgslkgo8++giNeuOf4ja/9CvLMgwGA23fAWCCLw+z+Qz1Wt30g0Mb8ou8nFXLOuVoPEKtWlvrh42iCNVaVdqYklKsoNPp4OzsTB0OALVD/X5PSzlDd4DLS5ES3dragh/4mF5OdV/LPp6tQaFxHCNJE8RmgpKtUJZnMqFpd3cX/X7PlK5CrcPLs2K6E0z26fkefM+X+Zeup+ccJ7GMl4vLFqc8Y387zMi7OdIsk7q6mVhSqXgoitygbv7XKkm9FM7SdT3jLEPF12nkBdLxEUUhojCE64gzS1Yx4niJyUQg0TRLEBvCCrNIahwSXqODWK1WODs70wfe7kcCSiGBZrOpbSh2uwoApXPTyNh1QTogFpvt1hEaz2azqTAsxc5tgQBmjmxNIJxsQ2IASrjCK3VlPU8mUqRpOXOQ9Q07g7KZYK9XSavnsu+pTWKxSRI2PHl6eqpKUfagcdaQuSdK2MrFfL7AbDbD5uYmACiSAUBbptrt9posI0lE/Lff/M3fhOMURoEnwdOnT9FsNnF8fIyiEHWTra0teJ6gHmEYYjAY4NatW7h+4zrGIwkYL4YXaLVbWntlVj0cDtDv90VuzHHwySef6AxPKl/t7Ozo30ejkWEytnF4eAjXkWn0ZLoD0JJDo9HAyckJbty48U9yj1/2JRDlQgMt3of5Yo4ojKSNJ5e+WjqFpBCbwb5Z2jzHcVCpVrCYL3Dnzp212bpAuXfr9Tp6vR7+7u/+TvcmJQ3DMFJ96aIo4AcB5rM5JpMJRqMRAGnXIByPooTqL6eX8D1flYgY6LFToNVsotPpYnt7WxGYq8E8tYt5TIrE83mS1+eoRGUbXp7nOlggiiJTr5RpOqnWdlMNLFl+CEMzkeQr1kvhLKvViomIfBSFiAtkWQ7XNSogaWnk6UQ3NzdhJvMhTWM4AAJfaMGMSuqNOoqpzN3jfEG7AC4ZmIui8AAUSnWn42m1Wri8vNQswNYOZLRVr9cBlJAdNzhf1+/3MZvJFBHKjdGh8hg2ych29GS4si5WypIJbMoshcehIc0yGnD5PTNkW/nHvpavnWVZc5SI04Hrhnp9i6IopbY4xtYp9U2zLMPl5aWKqrNXljAXACXF8PVBEODs7AyVSgXXrl/T3rIkSTAej7UcAcgg6dxMazg+PkalUkWv18XGxgYODg5w9+5ddDotPHz4CHUjXNBo1PHmm99AGEbwfdkvlPBjbemzzz7DT3/6U9WxrVQrKvYuKIwMie73+4iiCooih5M7uH3rNlrtFgaDAbrdHg4OnuJycomd3R384he/wI0bNzRTt4PNarWqezCOY1xcXODBgwc4PDyE47yWXASg2R9rxa4rMxjbnTZ8Mz+0YuBPOg1C4q7rYnI5MTB4IA35FtuZQieHR4e4tn9tLeiezWa4fv06lsulBj0sZ1G8REoCjxHHCY5PjjG9nGKxXKBY5Wo7lVS4lIBrNp3BcaV+T/Z/GIZrf69UK3AdV8tlsj+rivZJO0d5rqvVSlvsaPPCEIp8yIDssbR4+Z4q/4idZLtLYNnYAr4fIk0z+F/jEl8KZ5kXGVzPQZ6nmqkBknHmKWnOPoAcruvrRRFWVgrPd+E43lp2lWUZJuOJGXUkjpFUec4JTJIEfuDCK1y4rozeyvMccSxF9flcapGMgK/24dl9mLaAAM+Nv2s0GiqxxPojo3abEcllsyoJLbPGZcN9QAlJs/bK7JN/2gaf19VmfNrn+SovBmNCHpCWJC72FtrIwlVSCoMSkqfWSC6+p4aEDGjS/2u1GrI0033w9OlTpGmK/f19bG9vYzKZoN1uwfelJtioN6RR3JzH3t4eiqLAYDDQvuP9/X10Om2tUZFB+/TpY9y79wbCMMRPfvITfPDBB+j1evjwww9x7949Hd/GTIDiGVJ3j/Dpp5/i5s2bqDfqePjwEfb2dnF2doZer4etrS188smnioawfvvpp5/id37nd+B5ItQ9mUzw8OFDVCoV3L592/y+hn6/9093s1/itVgsUK1UZayU2V+VSgQ/CJQpy+DD3oO0B81mE3meYzgcYmNDeBh0Lq1WC47j4Nr+NWSZQJy2kAkdFQVV7Ek3cRzjd3/3d5EkCQaDAf7iL/4CFxdDTGczJHE5OGK5kuyyQAHHddDpdjSBoQ2V2Zg1+IEoBEVhCMcphf9JVrRhYYBTgBYIQ0FtTk5O4HqCrszncwu6ldF5WSbzV3PD6M6M8IBkzD6CUEZ9yfeN4fk+8D/D8Ge52SK8CwgryylkuKdknIUht8h7uFkAH64rIg6+K8o2eZajGtVQCatYxQskRaosQmaW/X4f5+fnyLIEDqixyoGhpTYtYTM5pxIG9TxPs0ZmD4QB7HohHVwYhlqEtyEu+2caTC67Dsrf0+nx/TYcQXiQ77PZxDx2ySQue5Fer/Leuo6DtCip9aw/0hjxevHasx7MlhIGT0QDfN9HpSK1dUKplUoFZ2dnygAcj8dYLBYYDAa4ceMGjo6OVHeWte/ZbIaz0zPcuHkDWZ7B93yFb6fTKabTKbrdLpbLJRqNBlqtltl3Yvg+++wzvPHGPURRhIcPH+Lu3btotVoiXNDfwNHREd58801tt/rss8/w7rvvrhGcTk9O8d577wmRJxUG5tOnT7G3t4fd3V3cvXsH4/EY+/v7+Pzzz9FsNvGvfutfKdu7KAr0+33Jwje3NLPY399bE394lRefb1vVrCgKhGaCC20Kg2tbjKQoSpnEGzduqMZ1v9+H40jvehzHePHihRJdKMTCljkej1KiLDdFUYCTkxM8e/YMSZKg1+siDEPsuA4m44kRJ4+1JY3nqZJ5RSl64Ae+fidZEqQSWrYFWShnNx6NMRwOTfDYxrVr17QNJc9FYUqQjg5kMLpjVKccLW+VffEGkXM9OIHM3nQ9F2mS6+ivL1svhbMMg0iT7LwwDf1W2l0603JOWUm/L+nxaVr+G2D6CsMqfJ8jsDzAygQJzU4mI/M+gS6DoKHN/zoqxjRb2zVC9h3ZNSa2k/A1dm2rXq+vtcewlaMoCs0QNcPB+lDcL1PqsenY9mt5nQCs1VqvXhs60tetI1iL0IsiRWoUU2xCgt0exIeQgZDrudqvxr1KZSZl6pmHe2BqgKPRSNqDGnXM53Pcu3cPFxcXuHv3rqqb5JkEOJPJBHfu3pE9Z5rDGTEvl0s4cHBw8AzL5RIHBwf49re/hX6/hzRNcXBwoIS24XCIw8NDvP/+++h2O3j+/AWarRZq9ZqqSnF99tlneOeddzCfz9HtdlFvSP21Xq/jrbfe0r7g/f19fPLxJ7h56xayVGpTG/0NzBdzHJ+c4PbtW2sylN1uF1vbZctIv9//p7/hL+niviIq4bgyBYT8B9/30Wq1lO/AVifaFgAIA9mPHAidJAk6nQ6ePxcmNBWg7PolHbPMesyQZYmK/Uub2xLn5+dod9qYTWfY3d0zTfwxWs2WEotW8UoJNI4jRCI6PnsINQBRtMpZk2QpyajpuI5K/tkdBLdu3cJ0KqSi0XiEZqOBoigwHo21P361Es4AJRyThMM4KNcYWDVYyS4LReNe8szSKcoWB8cBXEdk4sU4uZBrLeNoxOCLkywdRK7GTJhYwnDixG/PLcdkOXDgm1TU9wK0W210Om3MZlNlzFYqQrIg8QYolXH4gNsSXjbTlZGhDbmxRYQQCrNBOlMWmW1HZqtscPqDbbjt6IuLAUGWrUO1PD7Pl9DLP9SE+yot7g+bNczoWOqZiUV8WJ+Nx/vP7MgmYACSXXq+0PCTJBEVliTG3t4eFgsh+Ozt7WE4HGJvfw+JGR9Uq9WQFRlGoxHu3buL1SpeO9/NzU0cHR2h2Wyi0+lgY1PmuG5ubmJzcxOOIzM0Ly8v8fbbb8P1XDx69AQ//OEPsVgsjPRcVfcFB52Px2NsbW2JfNpshsPDQ+0PLhmwLr73m98Tdmu1gv63+3j8+DF6fZnJmWYy2efGjRu6z6vVKvzA1z7UIAzUEbzeh7J8X64PRfk9z0NYF2ifaAEZ/UVR6GB4BtxFYeaQOpAB0nmGIAixXC6xtbWFZqOJyeVEbYptR4IgQNgJMZ9PcfboDNvbu/pMTKdTNBotxKsYYRSi6TQN8Qc6g5hynTYa5gc+kjjR/UO7OZtNES8l2VgsBXoGZExjtVrF+fkZ4jjBzs4O0jSB7/mo1qrwPcl2B4MBrl+7rmMX6406fE/Y3FEk80DzPEOSpAZ5NMkSoOik43AQtOjTFsU6efLX7s0/9s3/H12SUZZOo3yIHJTfQ5xi6QBL1qLjsG4J8K22E7r6UPqej+VyoVGbGDxXoQm+z3VdbeOwszgbouNrSx3XsrZFB2nDpsB69khoj+0vVwXabeiUtU8adzpyx4FmyILTZ4iiUOnmV+FYO9J7lRcnGqhqkhFZLqF1BnTrQ4w9QyrjXnFdF5VqZY1oxukMdksPIAojjiMM1el0KvJefsn4i+MYs9kM29vbWCyXUscygtbsJy4KkeAj6jAYDHDz5k0UhShWDYdDvPvuu9jdlfFJKIATo9DDua0nJydoNJrY2tpCv99X4sPz5yLefv/+faxWKzx48EAJalEU4cmTJ6jWqkow2dvbw+npKa5fvw7PE5WWWk3aF1TU3/WUuOagrPu/Hv4si4FwKdHprwm10BlVooqIBJjnmHXwPM8RRALRck8DQLvVRpqlmM1ncBxHZe0AB7VqFWmWYbVammNIoMOgL89zbGxs4IsvHuLatWsiTm4CbgpSUMXp4uICRVGoOlNRFPAqHnzfw+npKXxfEBPP83F8LCxoZskAUKvVsVjMMRgMNSvm+ZIhnBc59vf38ezZAer1BqJKqVpWthz6Og9TuC5li5zrrjPdHTOZR/r3X3IY9uvWlzkV+f3aiyzQNjeO8ssj1avHKwzV2UVZ42NmZmtf2jMouXSCgtnUvGE0Hsx2Cd3aMmB0cMxcmTlyg9rMVaAkoNA508jwnG2oUK9ETpYXJ0aUrDX7OryO6qXRubCcIIlUYqw8xHGi0xNQQAkwjuuos2S9JjCwf2QG2joQ5RNqY7quqZM4bjl1oxBhDBIlGNBUa1XNIvI8RzUQluDBwQF6vd6a2gkVrGq1Gl68eIH5fK4Iw/HxMT75+BO8+dabpt+xgslEJNTu3LmD8/Nz0wpQ9t9eXl5ie3sbw+FQAz0aZO5nZqZ8VtgiZTOByey1+1IBaDbE5+b1koyd5R7Wj4lI2bNq0ywFzCUr8kL3SK1Ww3Q61bpjt9tFlsow5eViKcxTYz867Q4WywWSNEGRy3vn8znOz0/NrNVA7cRqtcDt27fV/kRRhNlshqdPn2BnZxeDwQCVSoRWq4lGo6mJR5LEABwz/3VDM9U4jtFoNtBoNPDpp5/i29/+NmazGQaDAXZ2dvDuu+9iFa/w5PFjbO/soBE0lOORpwLTbmxsiqN1PfluKt4iPekUbhcH6JZom4VOcpJOUTjwvHUfc3W9VM7Sjth/3Tn+w5Gn/Z4vyyLXH9QMQMme9T3pUSsKwDFv46ZiJsfj2k3+PKbneZjP52vFdzpDQnSsB9Hw0FGyoZdtIYweabTtxmLdMMaB207Sdrj2ecq/i8NkZksYBsBrQwUhlMXxQoMfIgJFUSA2jc6r5UphV7IGef1p4FSAIPARx4kq+RR5jgKwUAYzJsmTDHJrc0tp/kVRaL9XrVZDXuQIi1CG1y4XGI8nuHHjhgoAMJBrNiX6Pjo6QqfTUdENUWSp487dO3jjjTdU3GB7extplmK5WKJWqyLLcty+fRu//OUv4bouPvjgA2k3Cfw1VKPdaSungJ/Pfcqh7EmSoNVqrQVztXoNaSKZu63PDPz3Pd+vwpL2kPzXVL3Y/kXuhAPJxrmfKGAgYwhrWvMkk98mFDpwUK/VsVguzL4NEATybzdv3MRiMUOz2USt1sBwODSsaulIGAwGqlWbZRnefvsdXF5eYn9/H0dHR5hOp9jY2NR+WgprZFmGx48fa8vHvXv38Pz5c4wnY3znX3wHi/kCnU5nbeh6rVZD9a0qfvnLX2pd1J7MRDtLIQ/upTTLgEKE1kVAxNOAI81SeK7YPdphIUD+w9OXXgr8zXZqX3XCdk3Q/s/+d9tBXv2Zx57P5wodqHM1guLVqkTxtVpN5xPaajqEaumIbIk7bnApKCdqVFkXsOtftiCArf5Dp8v32J9vt6bYhB2+3q6b8jj8u3yeNOTy+nJqAc/zVV9smLaDELKUOTKu1KD0jEiEq7U3eyyQ7/tSGzfOU6Z0uMizDAUkcIoiGb5MyS/J0HIURS79jpWSHOS6MvHBcWXE1dbWlvap5XmuAuzz+ULPYbFYYLlc4s6dO2i2RKBgb28PeSHKMMPhEKvVCtPLqZGJXMmczOFAA8qNjQ08f/4cKKB7MYoi+F6ZKRJmZVbD99IRco8DknGr4ooZRGAjJ69XSeSjcyTKRORiFa/UZoVBqJB94AeqRTyfzzEajZSdWhTQJv04jv9/9t48WLP0vuv7PGd/17vf7p7u6Vk0qyRL2MYuF1iOXQrYJBCCqWzGJC4ZRLHEdlyFQzkkdsCQwlSMq+wApiD8gRNIpRySkIIKFcoJju2IINuyJKyRZjSLppfbt+/+Lmd/8sfv+T3vuXeWHjme0ZX7fLtu3b7vet7zPuf5bd/f9yfDnPOlc7JkluhisWCxWHDn7h12d64QRYmfZqOlqfv373sSzfr6OtPplKOjI7//yGBzGYO4WCx8r+ZwOGQ8HvP8+5/n677u6yTCCwOefPJJkjgmDMJzAjFKbNIs3fPvf57j42N/banua9M2nUBgNWYr0DJXJ1bqclYU3WtcI/a3w6VZoV0jB280mheNRPfnYkR58XldqEFUNE2NCQLCUH6iKGaRL8+lw9Qwd5v4NSWqggXqLeui7raC6GfSaEVFBLoTUboEE31u93Pruem2emjNR19bI0/5XI1/jy5pRZUy9DV1DNjDjm7k3pVsUw9X+9CCIPCi9zIEWaawR/HqIo/j2I2wkintJjCuBp56qS7AbzJb21vn17OLOLVmo0LUWNjZ2fGODsBwNCRwGrQvvvgim5ub3Lhxg62tLdq2ZXNrU6aLIJ72ciFi/Nkg81MZ9LOdnp6yWCxYX19nMBj4tKsyJVX837cCuKin6wB0hbW7fXJRFPmG98hFMboGu87gww7dI8bjMcYY77iBGLwwDLl65aqUfdrG9zVGsZDIYFWy0VLC8fGRa8ELVkbXZa8Wi7l3dpIk4fj4iOFoRGACjo+PvTSntfacRnWWZWxvb7O5uek1sEejEc8++ywAt+/cZjwRzdd79+75QCUMQ97//vfz+pe+RNu2Tig+YO/enhc016yKrp0kTnwmAiNrKc1Sn6UIAtP5zI0j66yCKa3d+pY+2/q9VLNCuo++ndN2KdKwF9Otb2cw4c3qlxfrkG+MLru/67pyeW1LEIgYgoq1Q+BTTnpCVTBd36ur+9k0MlqmrmpfqNZNpXscstGqKPXKQOpxac2z286ht8HKGOqXW9e19y61n0pfR49VCUB6H+DTI+II1F7t52GHpmnU0VFprC7BInBOVZwktE3rjIakt7VOqYYxCAOZ7oClcunYMAgoilKiznjVd9nUDSYyjjW7kl5s2savs8VCBthuunYQNVggtenT01Nu3LjB448/7h2wwWDAcDDks5/9LM899yytW8/j8dhHLovFHF3zw+GQxXJBGMiQAt0ot7a2CCNpU1CCjxrDPM8ZZEICWSzmvuVFHbDupuXXcLtic7+ZhOPDjIvZDG23AShKEb147UuveedCjVhXbSdNRRYvTmJOTk4wxjCbzc49Xhn9YRhyfHzCcDhwI+imHB4e+rKSksh0VqY6S3Ec++hVGf4aTKxvSAvSYr5gMplw5coVee8z66O5zGVOjk+OyfOc7e1t8mXur8GyWO2lbdv69djlbHT7Trs2QcbEtX5YtcDSNLWYY4uf4nKe6PP2a/BSGMsuLhrCN0P3/rd6/JsZ09XJFKYoQMsqCpQTF/gwv2swu/2IF6M126gm62oD0AUEK2k9OVa8koSvb3W+7G5UqYaxa6i7RhZWajzqGWnfVbcdRb1MPfamqZ2TEDsS0KVbBu85zp2zwHidYYyrfbjvUiQLK2fEIAgS2rbxKj0+7dghmUXxau6fps279XMTGKlV2ZYwCAjC0JcK4jimKAuCQFpFVDg7DEM/Lmk+n3Ht2jVOT0/9e6pQQtM0bG9vE1K7CZoAACAASURBVDvjVZWVPzYhk0gJQHVFR6MRy8VS0m9p4pVQjJFJ9FqiODk58eUDWGU9VGg+iiLfR9xd1/qj0UyXLNRDoPuClHTKc/W4wAQQSJQZxeJM61QS1ZsGtw/kAVVZOSH22LNdtXdcjVHTNOT5SpmqbVvfRpTnMuTi7OyM1rY+INBylhrfe/f2WHdkojAIfRSn2Yu6qVlfX+fk9IQszVhbE4UpzY4NB0Pfh2nbVd+58kA0EgY8L8Nn2Qw0tRLnpO3QGLAtNG3joknXNujY6xg5hxhhaItQQfC2keWlSMO+WQ2ye9+b/Va8GZHnovF8Y73TDfU1TiHCLZzRaOTrRJqG0otbPWp9HW3r0PfTDeCiWLEW4Fd9eaFP53Xnq3WVYXTz6JJ1gDekgfW5el/386rn342AoyhiNBo58keMTGwZ9Skw8BtE48gBeu6yNCOJJcWYpAmDwVC83CQWGnxgiJOEuqr9c9I0cXW+gCAwJLGbaG8l7a+EGFgRebpRVlkIwy8bSHuAwZBmGWAJjCFxo4e0RtW2EoVmg8yVEiI/aiuOY65eu0pZlD5Fqr1qoBrHkr5K05Smw2Ct3frWdgBJ1cWUlbyWEkfUwbBWCCrabtCNDDQ6D6Pw3PWhBrOPLAXiIEUu+9TQNDXz+dynP9Vg6R7UTSOq0dT9rCxkHmlVVwTOmddmfRUgyPPci6/r87vpcZX77O4v2kKnE2Tk2CIv26iRZxQJI9Vi/bzUOFqR4770+pfY2dlhMpn4SLqrPqbnQ5nnsKqFt7ZdER3LqsP7MGgg5PkDziCGoZQ42rb1s1hVvU3ey9K2l1xI/a3Sq9373swAXrzvrS648/fL5HgpDjcrgxMGLApRQqmKyp/Abm+cslnTND2XCtPUSXfD7aZPtb6k76cRqho/jSK6efNuJNuNLvWzay2jK47QTcPq62n9Y7mUgr4KLRgTsLGxdq4m8jAjCIJzkw/iJPYknbZtfV9bVanRkVSrpqjCRCJLHawrhhLCSBh53nA4b1Ybx7NoNRMzMIZQe8s6PYjZIKOuasBigsDtBUqIkBF2xv2r3dCBMAoZDYfs7d3j0ZuPesdL16x6/J6k5mo+89nc6ymnaUZV1aIV6zz0JEn9ehFvX66r4WjI2emZT5Np5Ngl+hhjJLJllZL1dXve6Cg/jBAd1sJxKEJ33VtvoPQ707q3jkxTqKSdRmZpKGPWluXS7xFN25xzslUDOAhkLmY366GGWPW0te6pYuuAr2ECnhS0s7NDEIigitZatfbdWlmHg47i2WKx8APOLauMXtOKvquWzwLX/2zsamBFkmR+fbWt9FFqScS2jY84RV98xSPxx+MyQ2pA3wqXwlgq3u5A36pO2b3/rW4/n7aVJlXcybKs5ktagyxUE/m+JjVYmnaNooCyLM5FnV1jdVFUXVMVOvJGDaS+Xpc63zXQXXaY1hn18yh762I64mLRWt9H36urJ6npm250+jBjNB55A6UMVnAp9Sz10WYYhrKxi8oi1j0+CIyvkwRBAC7t3jaiEBJo2w7Gs0vl+2oIwpA4lFpfWZTnGqq1fiOEIfc9WZ3rGmEtaCO1pn8BJ4m3mlyiMw91rYaBiBus2otavymWjnGZ50sfxUSRKLEMBgOvutNlEBd58YbshkqJdcsMcgwro63GtCeZCYSnUHsnQpwKAL3OQ0+60tKPnj9dLyotqI65GjltLYPVftY1lN0hAICPKLs9nyqaAqt9RA0ucG7ggzri+r13hTt0qo2mktM0Oedgqb6s6ei8hlFE2zTIfNXV3qvHofusSOZJdG3da0qE2vprSPdDOS8iSBBGIdevP/KW382lCCneKv3aTc12o8M3S9l2T173NbqPPxfatytjt1yK1xUSEoXyheqi6bJTwzDw6VfdZKqqpK5LH0l2iRfdXkxjDFVVu/TqaniwHoOfotK2/v2apvGenqZXuuzXLmMTVuLyujh1cYdh6KXK9DnaPN4Vin+YkSapJy/oRZ0kMkO0qcXz7Do23nAZgwXn9UsfnLBXXX+sO+fd+p585/p0iRQlVebGMrmBAdZaP/E9MIGXpNIWoMa9j7KrlZmrm5euU01jqeECSDOJEFfqKYbWSmuCBeq6onIb687ODkXemajiDGsYudRXGJxrt+lGkZqV6ZYjwnA1CADE6PelgBWMCVgscmSSUoxGlrIHND5iU26F9h8ul0u/r2h2So2P1irVSGRZ5meKSkYl8AZNhjDHngGt6d5u1NntkVUnrG0lRVxWpXfi1fHXY9J9uKplbUShpGyn0zWn4qTO+/kWQa1lWlXQYtXbrOl9WY/BuetBSXm+vQ4xiltbW7z//e/n53/+5/nYx74XgGeffZaTk9O3/F4uhbHspmH1d5cMcNE4vplh7D6v+7iLz9PRS3Jy5USPRiOybOgEsUP/patiRtcAiZGKz9UhuwZRPa+LkWddN4h6jvELU9Op3c99sTagQu26SLuPVQPd3bz1/9q0q4u0W1/VFhc93u57PswYjUZeREDGIq3GF6mzFEba5iBp1MAELkILiZOYuq7QCe8YQ+2yE1Z1il36rHbeszfCVuuX1hmk1Xpq6kZYtqGSZ1bvn7jWH10bYjBXo+NUCkydJy/A7VJosauneok/1/cZBCFpkpJlrs3K4PvatO5kOC/31+2d1PMG+OsJVk7w+T7gkG6Z4WGGnqeuc9HlTVi7yg54Jz88P5rQz5R0xsLLCxrDfD6XKDGStRCYwMkSDvz3VBQFoZslPB6PmUwmnmWrGYgwDP0e2W29AHwWRQ2ZpkS1Vl/XNVEYMZ1KS5PUwpVhrutx1Y6na0mNYxiFvv6ut8dxDC6yBIjjlCCMuHr1Eb72a38np6enfPzjH+cTn/gEH/zAB12r1Rf4tm/7Nv7W3/oZoOWzn/00R0cHb/ndXKr825tFOG9Mo751T+WbPd5TnwAxlBaLjH1pXN2xyHMgcFGCPSdx1yUfyBcR0rYrAxkEhrpeUeO7/ZWr1EDwhvSHbmD6mK6XpmkR9fwvpmu77SF6YQyHQ99Col6oijHrRafjnvQC1OJ8H1lyrt0jCANCt4FL72Ak/WDh+QntqraiJBzbWkLXImJxYvZhKKmjuhbP2UWVgQkIA309MIGkYMH4KDSKXE+Ykyhsz7W0SK1Tpzp0HarIbyRSW9WevLYRKTBd03kherPa4iJpKWECB4EQQ7S/LWplqxiPx7S2dT1uwtwlhrZtfMpXa21dZy7LUp8uvlib7972sEMNj5J5xEBI7VgjdOU4qFOi0WE3qtR9Q1OvOrljOBzSNA179/ao6/qcQpAa6qaRVqbud1gUhTuelbOojo+8hyjrBCFUZYlMgZK9TiZwyW/sqpdUX6urpV2WJUmc+EDFT0Sy+Aya7p3D4ZDd3R22trb543/843zxiy/xJ//kn+SjH/0ot27dYjabsb+/z927dxiPx/zMz/wMAJ/5zGfeEGjpnnnp2bBdXEyd6m0XH/Nmv99807eIkVy1UiibKnCGJE1TqrqkKJY0TXWuZtiN5ryn34kq5f3PH5tnbHXSEIqu9werAcw6YFU3yq7HqEa76/l1awJBEDCfz1lfX/cDfPW99P6LqUB974u3PayIoogojLwRtFhP4KnqWtKWjlouhrGlqkpPRggC6YW0rXxfsVfeiQmjlcZwUzcrVShXe8bIGKY4lsG4WntstfYcakoppOkyBt2/qq6IYp2OUnuVkunampsd2Drx61hcR2MIQxF3B1yELKWJumn8tefVTgxeas1aYeTK+jeeFKGRgK43jUIjFwWocdcNvKtcdfF6epjRVfzSlHpRSJlHmbDGfR/KVtW2Hy0ZdWfWapZJR1/p9JIuuSvPc1+f1P1RU691LQQva2X+pD6nLCunGlZTVUL0yrLEs6GNgSSNiGOpCRrTSp38wp7ULQmBEISUMS2On0wbuXLlCh/+8If5pV/6Jb7ne76H1157leeff56qrvnkJz/J937vx/hLf+kvcf3GdX7jN36Ds9kJdV0gsqb23L6sddwu38OYEIsQhN7yu3m3v/zfLLqpzYu3v9lj4XxUqQYuCHRAsnUnR9JJYImjBEIoi4o0zQjUCzKhF6EuisJfzG3buE2PTsSnr9uRWwqCcxFbtxbZTYd2o8ZVXXOl99r9rWmNLplHo0atUR0dHbnPvGra1ci1m05eORlCPul73CQdGjiGa9M0K9F0B2utN1pt04Ijx9i2xQRSKxE1pJimaUnTxBuvKAyx7nXBUlYlBlxP44IsTSmKUrRj3Rr0a9kRiYJAtEBNI0ayLEsZeVRWjFyzuDEidmBbYeVq03lZ1wRB6IUQYOWkxXHkjjelzZe0bi3ESUybC3M3QGqhWZb6HuEgsN6714Z4XedtKxtjEAXE8arOj7Fuvemg9Zo4Tnymo4dAa8PqPPvMQbvq+e5u/hdTsfrd6nmXdPjK4T85OfGpT41UlZijWa2maRxRqECGU6wyEbqHyN5iSJLV2EHRWBU2OBaslT3SmIgwPL9/a3pYuQJVJQSyp973FL/4S7/IT/3UT3Hz5k3+9J/+02xubvLpT3+aj3zkm/mmb/omrLV86lOfcmWsVVYOXGTcWjCGsiqd/J0jPDa119XVkVx1LWI02kr4VrgUxvLNUq36W41F93HnI8323PO7pBmLGAQTAshcy6apRSfRGIqqoHXDR5uydF4INM1KqOB8LaABVt6JGjz5ootz9YLuYu7+7r6u3t6N+roRtW6A3bqiqgnpwgbxMpUxqx6ZFvI13epZkB0ihf6t46ceZhjkPGqUJnU7A46NpwIAdOrpYRhgCUnixNPaS0dqWea5Y/PJNqWqKjrQWUkZujFK6nfVjiJ1yZYoCjrXgJVpIWdnPnKN4oCyKgjDSBR6nAYoBieVJ3VQyUQ0jEYZs/nc67RGcURQN35cWBCGhCE0tdss3UQKNeBdx09Z5LqO2rZlPBlTV7Vnyla11L+iyAloGOmFCwJDFCfUrmbfR5YCTaGqM1xWpd+vuiUbZbhqSxJwzhHX19Ks1ErHWp0aURKrKjGqw+GAxWLhWlKEYFYUsqcoOSaOY6/oYy1UVUEQrGb3WiuqUUVRkCTZitDl1kuaply5coXd3St87GPfwz/+x/+E7//+7+dP/ak/xd27d7lz5w6LxYJP/fqnyAYZP/iDP0jbtty5c4fbt28jPfHn989V2Us5Icq61n00oGktxjRg7Ipd26qMqdoLXIr4rVuYLsUuqXl2OB8lXkyBKuTDaZ3mPJGnrpUU42Y9hisyTZ4v0RYOOamaAtVG1JXmq0ZbVbUSBmiaVXFb1UmshdyJEq88m9YvyKpafbldI6eN25pOvRhVdl9PF3t34Xcl7fS1QKIVbQnRwneWZed6ouQ81f7C0WjhYUbkUpIigo6TyoqxVjIRojaCGA2nnmLbViLJWqKmpqlcTbIzri1JiKOIfDnH2payaFydSOs+IgIdBAFNXfnn1nUjAgB1JZGpFRafCaT+o/pTxhinQGJ9r2Lg0rphEBCYkMrVpowxLBZLL9PXtC1rwzWOjg7l9Ts0f62ZhUFAbSwGvd6k7hnHybk1q4a9yJcuqyEbtc4FlQ3dUjqtW1G4kpYVdfB6vNHQKcPZ31ZX0K6yR13GKKx0dq21lGWOxVLXUBQLx0Je7QHKhtW6oWr7Kqu2bhrfI7lYLM5lp8AwGk08T0IZurqXbW1usrm5xd/523+bn/7pn+aZZ5/l7/3s36Ouaj73ud/g43/iT/Cxj30vQRDwqU99yq8B3Xu9EpltHWEtoK61T73yNVG5WK03cvoaXVa2O7NeS9ZihXluW3ftrM7j22U4LoWx/Lmf+zm+5mu+hueee+5N7+9+EM2hj8dDf7uOaOlOCJG6SUtVFh3Duxrmq8ZSyT9VVZxLccDKoNSO5iwbG+T5giCApqlcz5NGvOKhtG1zwetpXe2h8OnU09NTHwFaa/3IrDSNWSzE69NUiBpPhbIaYSVQ3b1IumxDHfmlNQIt5gsJo3BG/V38cr9KYG0NuPqtM0xNWxMGMVEocl1CcxdCi/RBGnchizJT6+oyFhFzVma1arJ2BcSzwUA2JY3KrCVOEpq6JkkT2lbWShJLWjdOxDhXZUmcSH2y7fTaGdPStJbWNkgvsXjV2orSug0oimO3UTQMBymLxYy21bqOIU0TTs/mrE2n1LWlamqw1tdwNc1sjLQ4WKyPfEQjWVJpy+XCO3W6ySVJAk4n14t9xGFfBuigGyQYI/2FvhatafkL+3lZla5317jSio5AW81WzbLB6jVASFq1DHlWg5dlA/J8SRQlUmu2sZe9kz7iblpV5mVOp1OuX3uEX/mVX+Enf/InWSwW/MAP/ADPP/sc9/b3+d0f+WZ+90e+mbqu+dznPucjZO2l1P1Y96Ruq56uC2tlD9YyhgoTeMKQl+HD7cGr99BWPv3R0pQ8feWYaEbw7Yhm5iJ5pkePHj169OhxHn3uo0ePHj169HgAemPZo0ePHj16PAC9sezRo0ePHj0egN5Y9ujRo0ePHg9Abyx79OjRo0ePB6A3lj169OjRo8cD0BvLHj169OjR4wHojWWPHj169OjxAPTGskePHj169HgAemPZo0ePHj16PAC9sezRo0ePHj0egN5Y9ujRo0ePHg9Abyx79OjRo0ePB6A3lj169OjRo8cD0BvLHj169OjR4wHojWWPHj169OjxAPTGskePHj169HgAemPZo0ePHj16PAC9sezRo0ePHj0egN5Y9ujRo0ePHg9Abyx79OjRo0ePB6A3lj169OjRo8cD0BvLHj169OjR4wHojWWPHj169OjxAPTGskePHj169HgAemPZo0ePHj16PAC9sezRo0ePHj0egN5Y9ujRo0ePHg9Abyx79OjRo0ePB6A3lu8QxphNY8w/NMbMjTGvGmO+6wGPT4wxnzPGvP5eHWOPr2680zVmjPkBY8wXjTGnxpjbxpi/ZoyJOvf/vDFm393/KWPMH+zcZ4wx/5kx5jV3/z8wxkzfi8/X46sDX8Y6/LPGmM8YY86MMS8bY/7shft/lzHmX7j7f90Y881v8Tp/1xhjjTFPvRuf57cKvbF85/hvgBK4AvwR4G8YYz7wNo//s8C99+LAevy2wTtdY/8I+Dpr7RT4IPBh4Ps6938/cM3d/3HgZ40x19x9/yHwR4HfDTwCDICfehc+S4+vXrzTdWiQ9bQBfAfwZ4wx/z6IwQX+V+CvAuvAjwP/yBizce4FxIC+7136HL+l6I3lO4AxZgT8YeA/t9bOrLX/N7IQ/uhbPP4J4LuB/+q9O8oeX834ctaYtfYla+2xPhVogac69/+6tbbWP4EYeNT9/QeAv2Ot/ZK1dgb8FeDfM8YM343P1eOrC1/mOvxxa+2vWGtra+0LwP+COGEAvwvYs9b+j9baxlr7s8A+8J2d94oQR+3PvLuf6rcGvbF8Z3gGaKy1n+/c9ingrSLLnwJ+GFi+2wfW47cNvqw1Zoz5LmPMKXAfiSx/5sL9/5sxJgc+AfyfwL/Uu9wPnb9T4Onfgs/Q46sfX+5eB0h6H/gI8Fm9ifPrTG/7YOfv/wT459baX///dcTvEXpj+c4wBk4u3HYCTC4+0Bjzh4DIWvsP34sD6/HbBu94jQFYa/97l2Z9BvibwN6F+3+/e+6/Afzv1trW3fVPgD9mjHncGLMG/Kfu9j6y7AFf5jrs4EcRe/J33d+/BDxijPkPjDGxMeY/QtKtQwBjzKPAnwD+i9+i437X0RvLd4YZcJEEMQXOuje4FMaPA//xe3RcPX774B2tsYuw1n4B8eb/+pvcV1lr/wnw7caYf8vd/N8Cfx+JNj8L/Ly7vSei9YDfxDo0xvwZpHb5b1prCwBr7QHwB4EfRBy57wD+D1br7CeBv2CtvWiYLy16Y/nO8HkgMsZ0U1UfZpVyUDwNPA78gjHmLvA/AdeMMXeNMY+/B8fZ46sX73SNvRki3p4k4e+31rbW2h+x1j5urb3hXv+W++nR48tah8aYjwF/Dviotfacw2Wt/b+std9grd1Eap7PAv/C3f1R4K+6vfGuu+2XH9Rl8BWFtbb/eQc/wD9APPIRUsQ+AT5w4TERcLXz853Abff/8Cv9Gfqfy/3zTtaYe9wfA3bd/9+PbGQ/4f5+Dvh9CMs1RohmJcKeBdhEDKdxz/0M8PGv9Gfvfy7Pz5exDv8IcBd4/i1e52vdGpwikeQvdu7bvbBXWuCbgMFX+vO/5Xn5Sh/AV8uP22T+Z2AOvAZ8l7v9I8DsLZ7zrcDrX+lj73++On7e6RpD6kJ77nGvIPT8zN33PELqOQOOgf8X+EOd5z4DvAAsgFeBH/xKf+7+53L9fBnr8GWgQlK3+vM3O/f/fWdoT4D/QR28t3hPCzz1lf7sb/dj3IH26NGjR48ePd4Cfc2yR48ePXr0eAB6Y9mjR48ePXo8AL2x7NGjR48ePR6A3lj26NGjR48eD0D04Ie8+/gLP/Zf2iiKsNbSNA3GGPI8x1pLkiQYY7DWsr+/T57n/u+2bRkOhwRBQFmWiOIStG1LGIYARFFEnucsFguatiGKIgITcHp2SlPL33EcYwy0rbxfGIbUdU3btljbYi3kec5gOKAsSuI4pmka0jQlTVPCMMRaK8+xLbaVY0vTlCRNaOoGgKIoyLKMNE0ZDodYa5nP57RtS1VVcmxBQBzHWGuJIvl6giAgiiKMMQRBQNu2tG1LnucEQYDFspjPWS7l3OjxF0XBaDQiyzJ/fvRc6WsCxHHMX//pv3FRmuqhwp/74R+ybWt54YUXeO6557h69SpN03Dv3j329/eJooi2bdne3mZjY4PNzU2WyyUnJ8fcuXOXmzcf5f79A1566SWyLKMoCm7fus03ftM3sr29zedf+Dzj8ZjrN65jrWVvb4/xeMzOzg7ra+vkec7JyTEga/vKlSts72zTNA2HB4fUTU0URjRNw3g8ZjgaMjubcXh4yCOPPML29hZVVbGzu8vZ2RnGGPbu7nF2dkaapWRpxmg8Yj6fA2AwGGMoyoIb12/QtA3WWoo8J45jWmsJwxBjjF+fL7zwAlEU8dRTT9E2DW3bkmUZ88WC5WLJfD5nY2PDr7MolnNWliV3bt9huVwyGo+YTqYMBgMWizlFUTBdWyMKI77z3/7DD/UaBFgul55xqfvcRXzyk5/kox/9KN/1Xd/FX/yLf5Gtra1z9+s+2JM3f3MYDAZvug4vhbE8OjryxiaOYwCqqqJpGsIwJAgCkiRhd3eX+XxOWZYURUFRFOR5TlmWAN5AqkEqipzbt297o2ACQ1VWYlCCkCAOqOuaIAgIo5AojvwCDQIJuufzJW3bEkURVVl5YzocDggCeb+yKimLkjRNqaua+XxO0zREUUiaZgyHQ+I4Znt7C2OM20wgSVIGg4E3fE0jRrVrHJMkoWkaLJYkTrwzEYQBSZt4Q20wDAZD7xikacp0OvXGN8tS9NoZDAZYtxkKLbrlYUdrLcfHx2xtbXHzsZvcunWLwARcvXqV6dqUV15+hevXr/Pkk08yX8ypqpLPf/7zHBwcsLu7wyc+8S945plnuH7jOvkyB+D3/N7fw2w243Of+xwf/vCHvXP0ud/4HE8/8zSP3bxJnhfs3dsTxzBOmE6npFnKld1dzmYz5rM5g+GANE2xbctwOALk+hgOh+zu7mCCgLa1hFFEmqTsze9iLUwmY9bW10jimMViwcHBAXEcM3LrcbFY8Oyzz3pH7fTklGwwAGsJANuK87rMl5yenIpxTFNx0NoW685bHMUUQUFRFpRVSRiGRETUtRh4g+Hq1atUVcW9e3v86q/9Kk8+8STrG+sMRyNu3brFxvrGW385DxG6BrJr7P78n//z/MRP/ATf933fx4/8yI9wfHz8hufpc3oj+e7gUrSO/IE/+PutMYY4jomTGCw0znNdW1sDIE1TxuMxTdtQVzV1XVOWJct8icF4gxcGYlzVcDZNQ9M0VHVFUzeMRqM3RG9FUVCWJbPZTDaT0cgbyDzPyYtcjKoJfDQZRZGPBAEf6cLKaMdxTByHYCAMJNqTyC+iba0z6uIctK1Er3Ecu+OTz2Gt8QZ9sVjQti1BELBcLmmahtnsDGthNBqRJAmAN8ht2/pIU2+r69ofq373VVXx1/7rn3yovfr/7h/8rP2VT/4Kjz32GKPRiL29Peq6ZnNzk/FkTGAMX3r9db7w+S9Q1zUf+tDXcOXKVf7ZP/tnRFHE2toa07Upo+GIe/v3GA1HEhnWNcfHJ7S2ZTQccTY740Nf8yHmizn37t3j6pUr5HlBnMSsr61hTMDt27e5fv06g+GALE05OTllMBy4tS3fa9M0tM5JCsKAppHrAuO+fwzZIOPo6IiyLH1UnKYpZVFSVSVZNsAYWC5zwjBkMByIM+ne4/7+fV5++WXW19f5wAc+wGx2Rpplcj2VJadnZyyXS/+6dVXTto2LKmPyPOfw4JCqkmsrywZcv35drt2qZD6fMxqOGI6GlEXJv/OH/92Heg2CRJa/9mu/xg/90A/x9NNP82M/9mNsbJx3JPrI8d3FpY8sJ5PJKpJqa7/RazoT4PT0lKqqvEExxjAejb0hAFlIYRQSxzFhEPqUmDGGoih8Ok3Tp3kuUUCWZYxGI/GwT098FDYejwGIx7GP+DR1Ox6PfdSoRll/usbTGEPTNMSxGNmqqrDWkuc5dV26tOyYyWRCGIYURUHTVC5iDpzRlGi3qioWiwVxHDMcDllbW/PGWc+JGlIx1jFhGFKWpY9K1RFQh6S/6OCLL31R1mBguHX7lk8Vtm3LCy+8wP/zy59gd3eHj3zLR0iShJdefIlXX3uN5557jmvXrmKBk5MTXnrpJTY3N8HA3bt3ybKMJElobcv169fJspSyLEmTlK3NLYIw5Oq1q1hrefXVV2mblitXdhkOh0ymE9IkJc8LkiShqiqMDcEY4jghjEIW8wVYODk+IUkTyqIkDAMscOfOHdbW19VGmgAAIABJREFU1tjY2PBOpxhSiGL5/m3TMp6MiaOIZZ5jDMxmM269foswDHn6maeJwpC8yBmNx7RNy3w55+DggPX1Ndama0RxtMpQGENZVdy5c5cwDJnP59y4cZ2dnR2aRjIoR0eHDIZD1qZTwFBXFU1Tv8238/Dgh3/4h/nRH/1R/uk//afnbu8jx688LoWxtFjqpia2sdRmogioGQ5HPmUYBAG7u7s+BXt6eooxxhuvqqrEOx4MAKlb1nXN0dGRT+Our69746Jp3uFwyOnpKYvFwhvO0WhMWZZMp1OGwyFPPvmkpDpNQBxLeqlpGs7OzlgsFoxGI9I0pW1ayqpcfS63sPX9QWqok8mEpmneEEFqPXI4HFLXWhu1/vbpdIoxhsVisVKVcGkbjZCttcRxTJIk3njWtWxEYRh6R0EjzB6CX/21X+Xbvu3bSJOUxx9/3Edvv/6pXyfLMn7vt/8exqMRRVHwysuvsLY25Ru+8Rso8oLDw0P27+9zenrq11dgDOPxGIPhkUeusebqkrP5jCIvGI2GbG5uyncXRpycnjAajVibrpFmKWEkDs58NiMbZGDBti1xmlJVFUVZ0C4bzmYz4igmzVLSJCUKI+bzOVVdsbG5wfuffz8A9w/u0zYtxsDVq9c4PT3l+PiYJE3FgbKWuqo5OTnBYtne3mYymZAXOWtrawSdrEQQBDz55JOSrWka6romDEP29u4xX8wZDCRtvLUpZYdlvqSqKu7fv89kMhFHNQixQNPUFPOCg4P7X9kFcEnwl//yX/b/7w3k5cKlMJaP3XzM1dUyBoOMwKUskyQhyzLiOPYbvtYf19fXvTHY3NwEZFGdnZ35yFMNB+DTqxod1nVDWUrN89q1a964RFFEWZaUZUmWZcB5whBISjhLMzY3N6nr2hvPJE3IBhnGGE+i0RSoRpxJknhjJnVJqOuKwESUlUR/+tzxeExdtz5C1d9Zlvl0rNZrfe3V1XibpvEOAeDTxYPBgLIUQzyfzzk+Oca8Yezcw4fv/u7v5u6dO7z44otcuXKFra0tZmcznnr6KVlPxnB8fMLO7g67O7sslgs++S8/ycnJCYPBgDAK+dZv/VaqsuLk5ITFYsHTTz/FfL5gmS+xWCaTiavjhQwGQkwzgSEvcl+Tb2qJ9sMgdFFiSBRFLBcL0izj8PCQqq7Zu3uX4WjE7s4uWZZxcnLC0dEhFphOJqytrVNVlZCGjCEKQ6q2Jc/F0cyL3EWiBRbY2toiiiLCSKLBNEkJw4Ct0RYYXA1enNLBcOCyI6281jJ3pKRdssEAA5RlyWKxpLUtZVEyXZuyubHBMs8p8oL9/X3uH9zna3/H72CQZWxvbz3gG3r40BvIy4VLYSx3r+wSR7E3jLrZKyM2jmMGg4E3OMYYhsOh30istd64adpUDV8cx2RphsV6RqhGoWmaesJOkiQ+Natp1LquJfVlDE3d0LRigNQoaXSoKWNloGp0W9f1uZStknY0GpTnyLEs3YajUaK1gfu8rX9sEIS0rRg/ay1VVQH4KEhJO5rm1fdSg71YLDhzdaazszOGwyEb6xveg32Y8eorrzKdTvnIt3yEg4MD6qpmY2ODIAwwGGazM8Iw5PTklC996TXa1rK1tUWaJkRRzAc/+EFe/uLLZFnK1tYWW1ubviSwvrbO7Vu3SdKEMAgwbv0kaeLXBxbPpl7fWKfIc+f4GIaDIXt7ezTHx5wcn7Czs8Pv+NqvJYoiFosFYRiyvbNNVZa0VoxyWZYMR0PW1zc4Pj52DmBFlqXUdcV4PMa2Le1oxOxsRlEU3Nu7RxiFDIdDyZS4dH3bNhAIGSwMQpq2IQgMdQ1ZmjEYDDxxaLnMSdKU2XzOfD5jurbGxvqGX49hEJLnS7a2tnjiicel7moC3jgnuEePy4VLYSyzNGN9fd1HQd0WkrZtPfEGZEMZDof+glbjN51OfepVDZy2nSyWC3/x6wbgDaljzqrhUwNXlhVpEvo0sG0tRSlpTq2Bag2xLCUCkMh44Ik/YRgyGo38Z9Hj0ueJMayxtiUMA5qm9Z9fCEgtEkgbrJWUVRxH/v00utSoUSNahdZklfqvNH5te9FzqCnihxlXrlyRdVA3svlPYsByfHzCF178AsvFksl0QhInPPnk+3z6fTKdMB6PuXv3Lu976n0+CxIYw2KxJEkTTk9PieOYe3v3eP755xi49btYLEiShLqqMYEhTVMGA1lDtrU0bcN8Nuf1W58Vwx0EPPW+9zEcjmhtS9s0zM7OwFowhizL2NralnXaSlr17t27YGA4GBJFFXVdYcGzt2dnpzRNw97eHutra6RpJs5q25DEiRCGAkPQirMZRiHUUjqJ4og0TWitBQt5UXB6doo9sYwnY65cvUoURkRRSFlWYPDORNtK1BQEAdas2Oc9elxWXApjmaapJ634VghXj9SozVrLdCozSTWdpYZTIyfbSmQ2HI48YcC3Wri0prR9DEmSxBsJjWQB/1rj8ZimbsiLnKIomM/nvo1DjWqWZZ5k07YtBokkDAYM59Ow9ep5atRXtQhhyA6HUo9q6sYzEmGVfg0C41O6XXq5/q3pY22r0V5OfYxsxgN/PvRz62d/mLG2NmX//n3yZc6jj96gqip+4Rd+keFoyFPve4qj4yOfih+OhmxsbLC+vkaaZezf2+frv/7ryZdLZrM5bdOSVyVt03B4dMhrr73GM888w7f8a99CvlzyyPXrLJdLiqIAhMkcxRGL+VxY0ctc2iz277E2XWN7e9uv/aZpmc3OGI/HnMxmgLRvhCZwGYoFxgSu7l05sljL2dkp6xsbVJVkUfJlTlHk7N/bZzKdcO3qNabTiXOqLGVZSAQcCLnHYIiTmLZpIYLABMIzcO1MxyfHNHXNZCxEvboR9rjPwIQBgTEEYUjbChmobhoiRz47PDj8ynzxPXq8Q1wKY6mMQWMMJycnvsViPB4TBAHT6dTXKjWFOh6PwYrXW5alkClcqlGMkfEtIhqVNs2qdUTrnVojlMhPqO95nvvUkr5/mqRYrPS7OcOkx6mpUzWkgRE2Yts0nnQDnDOSmkI2JvYRtEasGAkWVFxg1Q8Jbdv4lK/WJYui8IZ5PBZW7UVSgL5W91iUnNHXRuDWrdvuO8j5zGc+w87OLt/x+76DV155hfFkzM3HbrJYLNjc2CAvCrCW8XhMURRsb2+xd/cut27fpq5qskHG2ekZW9tbPHbzMT74gQ8wnky4t3dP2jkOD2naltFo5DIZkjrf2NwkDELOZmekacLNR2/6LEMcRyyXS98e1TQNzzz9DHf37lKVFct8SRCGjMdCysmXS45PThgNR4zHI5pmIKzTuuZoPpcMTFny6KM3qN26PTk9palr0ixjMBzK2mhbSRE3tZQMooizszOKopCsxIY4ndtb27RtQ+WcwSSV2nzbtFSVe426gaahta0vsbzy8itEceTZuj16XFZcCmM5GAyYz+e+dphlQpKZTCY+AlSDpExO6dvKGGUjhoMhRSmGJo4TBoOMPC8IAkMUSTotyzIfYWmkZlsrLEKXzoVVLbBbixyNRkynU5bL5TlGoB6nbgpN25yLGFvbnovsAhN4Q60EpG6EeDGqBnzUZzAkSUxVrQr/+jpKRNJUr76uKgbpe3V7T/Xz6rl92KHOVlkNuHbtmndsHrt5U85zGDJwPYZKdCnKkvliwQsvvMB4NOZDH/oQn/70pxkMBjz99NOURcFkMiXLMsJAUvpZljGbnZENBmxtbbG/v89kMuVsdkZT1yyrJW3TUDcNxrSMhkPPVs0GA5aLpe/bvXXrlqRCs9Qr/GgqfrFYsrW5yWg8ZjFfkGYplXPwVOxj69o16qryZYi2aYhc/TQMQwwwny/Y3Nzk4PAAW1nmszmbm5vSr1mWBMZgggCwWBsQhSFEEQaccIE6eS1JkrBcLjk6OuL+/fsEYcCNG9dJkvQcga5Hj8uIS2EsrbV+IynLksFg4NmpWsuIwgiMRJZZlvHII4/4aExlt+q6pm0q/9wszZiuTT0xQVOjWvNT5ms3ytLWDq0JlmUp5J5O6tMY42uT5waENtZHifqjqbass9Hqa6hxPLdROLtVOeJP9z3846ycD32MHluX5APCHO5Gp1or1fcFzn2uhxlXr12VcB5D0zaMnFJO0zakScru7g6vvPKqc4YgiRM++S8/yc7ONt/+7d/OL//SL/Piiy/yxONPkA0yyqIgThKaViKpuq5ZW1/j8OCQKI7ZWN8gz3OyNBPpupmwuPNcDGxrW9ama7S2ZT6fY4DxZMIgWxHd1HEsy4I4irh+/QYvv/wya+trbG5syLWUDVzLiKFtGuI4YXMjEwWoOMZkmUuLStZEe4bbRnomx5Mxx8fHVFXFeDxhfV0l8GTNhGFIUZaoilSWDVjmS6JI2sAMgIHlcsFiCacnp0ymE57depYokpSsKlT16HGZcSmMpdYRtS4j3vfsnBJPFEUMB0PyQlmCoSMPROciJSWtgMEYXEpUGHfWtkRhRDgIvTHTyEoMpZBqsHjJMoz8SGSXdFKisrFa23pjq4Zdf+ttGolGUUwUhT51240sxdC3ftPo1mq1BqqbZN3U5/onQfZ5Zdt2jaGeQ404u6+t/Z8aaT/MGA4G1C6Kj8KQsqpYLpYMhgOapubVV19jvphz5/YdPvqvf5Q8z/mO3/cdYC2HB4c8//zzhKE4I1VZEceJME6xbG5sMl/MvVDEeDyidAYmiiOODg+5+dhj7N/bZ2dnh/lCorcgCJidzbj56E2Ojo+oqord3R2Ojo6JopDFYukyC5KZee1LrzEajSiLko31Daq64uT4hOnalPl8znA4IkkkTZoNMupKlHSicKVBrBJ14oTVWGAynTCxE3HOAmcsbevUhCzD4ZCmbjwBLggCUb7Kc/b397n2yDW2t7exwPraGrWr33e5AkHYE3x6XG5cCmN58+ZN31SvqVelv9d1TVmUPi2q6bKuQs5kMnGEhtozRDWd2bTS2yiRqbRqQNegtTRNLU3SpgEMJjAMhgOXbqsJAuPrlG3bivftWkm6yjj6vtrzKHqtAwxGNqVOXVOPXTcLNZSaAr4oTadtIE3TgIXWtsTOe1epMcsbU63dGqn+v6t2pGSghx1K2Nre3na1x4rtnW0R4Hcp9jRN+Z3f8Dupykp6Bpc5e3t3SZKEsixIs5QgCMkGcl4n0wlBEHB0fMQgk/7W6dqUyXjMxsYmd+7cke8jEdFza1uieMTAqU6FQSD10Ht7kk1wrNIwDKXlIpBSxXKxdCSzxpUdBK2rD85nc6JYLvUkianrxvXWaiakIU4SRsMhbWuZzWfusQkWyJdLWmsl5erWeWiED6Cp2CCW6yZ277+3t8doNOKxxx+TPtRAHFTbtmSDgU8HB1Z6TXv0uOy4FMZy6OoyGllqxKWN+wN3ccVRTNO2tG3jSSxpkgqN3hkfZdAGQSg9bbFZXajWehafRmAAQRD7FKzqaiobVR5jO1Gc8e0YwLmoUsUGfN2wFWWiboSsRlfTvAoTGF8rstbKpBLZz3xU3NSy+UkdSF5bjXPTrqJYNYgaPSuUzKT14W5rycOOwWCICQxffPllmrpmbX3dC9IrkSwKQ4YDyVzcPzhgf3/fN/MHy4DRcMTJyTEb65uUReEcPSFfLZYL6qrmsZuPUVUVr776Ktceucat12+xtrZGUeQiYg6SNXGOVd00PHrjUe7d22MymZIXOU3bkLmUaRInBCPX41uKQ4aBs7MzsixlY32D9fV1Tk6OWeY5ZVURmMC/T5qEXhd2uRAmbRLHlGVF7ZzIs7MZ48mY0GVeTCDXiESXEh3GUURrLfv39zHAtatXAWHqdlub4iShcq+t6E7A6dHjsuJSrNDRUBiqcgG1rm3MeCZgGISsTdcwgRGxZpcq9XqvcXQuzai1vqquZCJCXfiL+rwgQH2ucb+bnmw7YgCrcV0WY1a1wKZekXa6zNJz0aURI2hdL5q1eIMamMDXYTXaFGKFoQ1b/zo444iRCFRYipYgDDw7V1No3QhYf8MqgpXWmiGw2qR6cgW8fut1tjY3fcvQ6ckp4/GILE1ZLnMCp3m6WC7EYDUNV3avEIYByzxnMp5Ir+LGuieCTaZTJuOJd1Ba56wcHR2xtb3FfDbn6rWrHNw/IE2FpBUGIctqyXg85mx2xmK+4OjwyE3RKQiDkPFoTBAGlEUpx1I3JEnMIBP1qziOOS6OZYRbGProOI5iceDqmqquGA5kHRhjWLrHmMBSFqVIUFY1eSMCAk3r6phutFddyYSgqpQ+Y13DWZqRpAm2lWyGOHFQVyVJmhI0luVCsjVgSAeptNHkfXajx+XGpTCWYRS6tExNEseEoRgrlQNTtp5p3WiqKKGuG2zbkleFr5VEYeQjwqaW3q5uW0YYRt7YqWRc21rqujqnYnNR0UY949J57rpZ1E19zgiBGL5uLVGf37bW1SvxBhkMVVWeq92ANHyrcdYUqrJ4/RSVjhh7t+6pBljrY13Foi7DV9O7fTO44MruFYajocjanZyIkLmTjZsv5pyenpJacexGoxHHJ8dUVUnbiiydAb7u67/Op1ZtaxkNR0wmEz772c8QxzGTyVTWeJIwn80AEb1fX1+XLEFrOZudSQalaRgNh04iceKcsYqqkgzIzs4OeSTScW0rZJyN9Q3Ozs4Io1CUepqGKIy8nN5sNqNtW3Z2dsSgNzVN3XjloMFApo4EYcDp6SnDwdAT04zr8ZW2JiHlFEUh8ndOorJtWyY7E5bLpc9q6NocjkZUrk4bJ5LJMciM1yiMaEzf69vjcuNSGEtNj2ZZRl1VEhHGMVEsSjhFUUj0GQZyYbWNFzBYD9ZpbeuNjeqqalS2WMhUBiHnSA9k27a+H7NrGLupUSEhvFEYwA+ZtqsUrDGBJ+FoGlgjYI1Oq0qiXE2hqrHq1li7qjyrSHSlZgT41pqLE04A71Qoaac7uUVZtfo6akD1/w87gjBgPptRNw07Ozv+vN8/uM/Q6biOx2PW1tY4Oz31UfxwJGSdtfV1mrpmOplycHhAEsfcuXsHa1uef/793N27KypQRYkJDHfv3GV9Y0OEL/KcNE1obMt0MpXvxmm0BsaQJKFfWyrecXJywmw2c+lhGWhe1RVxEmOc4pO2kjR1Q97mRFHEdG3KYrmgLEqptTal1OQbw3w2Jy9yN3t1m7qqfe9vEsYEcSTrumn9mhxPxr4MMBqNyJe5b7vRFhVj3HWD/nbzWt3frWkdE7lHj8uLS2EsAR8dDocZYRT7ekw2kGZqkK4KvZijUITHbWT9xuVJLR35MD/tI3RT26sS0xhP59doMTABpRU5LzVmed5lkFoRBHCtBV6Or2mx1C7yE2USNT6aHtW+SSXwaHpU65NdL7wr87dK/b6RPORTtheMX1fhR42l9mF2jau2kfS6sIKBa6GIk5jQtRdFbnrIfDH346b0+1PSymK+wALXrl7j/sF9wiDksZuP8fkvfJ4ojDg6Pubo6AjceY6QGak60eb09JTFcsFyucQEhtFQJpvMZjOeePIJqrLk+PiEtfU1sjRjNp8TBFI22NjcoG0kIjw7O+Ps9MwLBxhjmM1mgGXkpp+oAo8asSiOiBsR3qgXSy9qYegMSY9Cp4wlJXRZd4Efm1fXFbGrr3qjv6wwxhKFBoylqsRZK6uKIIgl++Ki2qOjQ3Z3d/356dHjsuJSGMsuy7WuZSPSCFLTVlVdEZqAKBqsNF5d6wiIYYqimLqqCKLACwSokVmUC/9+0p4hG04Sy6zByqVim0JYroDrNVtFkE3TSieJI+Mslgsf+V5MpSphyLNdOyQaNYo6akvrhpoqXSwWnhTUjWz1s2h9VIlC3fdtneKKps3UeOr/1RBrKhboU7FAnheMRiO2t7a5t7+HwbrUeUTTRJIlaK2P0EfDEYdHh2RpxhNPPMFsdspwIPrDL770eW5cv+HXgEz/OPECEYPhwAtZRKFEWGvraxR5zunpKesb6yRxIuPlnJLV8fEx1kobymAw4ODwgKosSZJUIkOc8EUQkiYpZ2dn1HXN1tamU5NqOT05FYcqECdvkMl1dnJyQhRH5/qNw2Al5K8a59ZaJ25Qud7dAGuRlhtraeoaXIZE+iu1Xi/lBn0N1UW2WLa2tp2oQY8elxuXwliORmOsY9ZNxmOCMKQqK3DNz2VVelEBWDFKgyCQEUCd1Gs3MtMUlO9rbC0WnQ+5GoUF+HqmtZYwEiNVtZV/vGxEQvaJgsjXmNpOX+QqShQiUNsIc1Cjw+Vy6aO6pmnIsszXOJfLpRdTAHzbjBybpEu1xaM7hksE160f8KznR42oMQGRqwlrb1tXdg+MI1s83FAh/3/+C/+cp596H2Gn788YaOuWOInJ85yqlF7Kzc1NjIHDw0PCyJC6nt0ru1ck/Q+UlQx6Hk/GYC1VVWOMjIyrkXaVsirJ0ow0SZnPFxR5webGJlktU3iKomA4kD7ku3fvcnR0yPXrNzg7O3PiBCLIQRx7Ry8IRaZRok5pBdF6fhzFlEXJ4dGhCGWEIU3T0ioJyAoTzZiAwTClyHPPIFcHrywLkkTIZbaxwhzOUgJjsHZVUxeynlxPUSwOsDF4PoFco/U5h69Hj8uIS2EsresZNIERL7gVbzZ29T9pwWjOEXHUAwZWKjat9ZFgN2ryyjsuIpQUpESsSoFv3GBcqT0a5KEr2rumLS8q8FR15Y9Ro7m2leNqTUtkIsCcSxWrQb84IUSPu0sS0g1KjaI6AsulNKTrbzWAcRx7KTJhBRsf+aqST7dtptuv+jBD2MMF3/iN38js7JTuKLTakatsLSSyJJFpHIvFnLKq2N7aJolDTk5OOD09JRsMuHH9JgeHBwQmYL6YSzo8zUTGDsPx0TEmMKxN1xiPx+zd22M8GnP12lXmMyEUBUHAwcF9kiRlMplw3w1Ink7XODg8kEk6lcyUlOyEpPvzIvdGbz6fMxqNpL6JEOTapiGKhM1dN5KhSJOEqiqpqpIojqkqWX/z2cyLadSVqAW5q4nlckHortFskHkGNohIeutKExjhCiyXCzCr9SgvJHmenpF9udHlOzysuBTGEkTeTeXnAB8pnatFus1eZeO6bR+AZ8Rq/6GkWlcM0La1xFHkRJ5z6roRRqMTJkiShLquPUlB+xq1dURrfJqu0tv0/10m6ptFueptd0UMsCv1EiXn6GfxurDus2s0GUWRbz9R4124vj6NXouioKorkjihLEvW19fPze/Ux3SdiYcZRSFTNg7uH9C2tZtjCbP5nDiWcWZxEnN2esZgOOCxm4/x0ksvMh6NOTw85PHHHyUvChLX/7hYLKRXNlxNtZnPZ1y5cpUvfvGLPPHEE5ycnoCB+XxO5kZjLeZz8nzJ1atXOTw8ZGtrm/v39z0JbjKZ+LWjdUIdQqDqVvky96xrYwLCMKIoC5q6pqpq4iReCX3YTkbEWqqyoihKojhiPpNarQWaQtb4fD73jlvbNMSulr+SyJOMSlEWonHrhNl1o62dqIK2XrXWEkexd056XE7oXvtO94qu1vZvF1wKY6np0aqspIeSFaHGiwl0GpeTJJH+RmcIw0AUTfSCk541aatI4sRT1efzufR0lYXvVWzaxvfFLRZLGu2pdKzVroFTdI2lCghYLNastGN9XcaxWpWxqxvTRSWdbvuHEjD0vbReqe+rtwP+sZ60EUUy/zOJ/XFXVcnt27cBmE6njEYjbzB1zufDjiyTFKiwOoeErnY8mUwYDkakrt/y0ZuPcnZ6yiuvvELjnKDhcMjx8bG0l6Qp4/HYy9uphmucitrNnTt3GA4H0gMZBL6/cTqdMJ/PieKYwXDIF19+iWtXr1IUS65c2eXu3btsbGzIEOoooqpEs1bLCUWRMx5PnPqOpWktSSxG9PDoEFWeCsJARBfW1ti/vw9AHMUsF0uappayR1mSDTKCwLDMlz76DNw6Xcznfu5lWQlZzRaFG2HX+rplFEakWQZWnOHGZUua2slLOgJVVVdecq/He4+uVGfXyF10pL0qWufxXXT3MX3smxnYi1Hq2z32MuFSrNAwkpqJtSJIoLMoNSILw9D3DArl3HgN1SBw2q9A7FpNZO5j4KeRWGupnBGqmxpbK8lA5Ld0hmRVV+cWjbaJ6I9GugpPvnEGvstiVWOqwgewihbPaWJ2WKka7ak0nY4I02PS1+1G32oo4zj2aTD9v9a7RqMxVVWxXC45PDxkb2+PKIr8CLSBU455mFEUJYvlgkdvPMp4PCQIDAeHUtMrq4rp2hphGDEajnjllVdYW1vj6rWrvPbqa9x87CavvPISo9GYpm0p8oKN9XXu3t0jTmIWy4UXQq9rYWqfnZ2RDTIGgyFNU6+m7sQJ+/v3yLKUg4P71HVNkjzijPnM1wyDQIhGkYvyqkJeI1/mBGEg2ZFB6+ukbdtQ5IVP8Z+cnDqFnoB5PSNway4wUpaYzWa+RUoHCchahtC1sARhIHX7Vuq588UCg/Gvq1dKYAInJCLXoSpVla6mHoRSTujx3qBrrNpWvmvplx14NbU3S7nqbYvFQkYkvsnrdqFO/kUjqK1sulepUb4oVXrZDOelMJZKXOl6LU0jk9rDKPSTELS3sq6Fdq50+O5oK22vUMOU5zmLdkHd1OdSpoBsBO2q1qmvo20YXS1XGYorxlubsTWdWlWVT4uqwbR2VYfsGsfuZ9QWE02neqF054EXZXFeqk+Frl0KWg2pMRKJ67nR86Ai8XpudOi1NpQfHh463dv3/ju/bGiamiefeJK1tTX27u0xn80JQnG4hLgSSA0yDNje3qYsSu7v7zMaDTk5PmY4HDObzZ3iz5LlMmdza5OT4xPiKGa+mGPPzqTO7doyTGDY3trm8PCAummwwCAMCKPIsaLnDEcj7u3fk/VR167Xc5179+6z54xxHEUsHfFIFZ3yIvcDq8uiFMNmLflS1IeiOHJ9mIiEZCMykha8wyY9yY0Y0ko1hgNqt36yLPP3ofL1AAAgAElEQVQp1KIsZF07w6hOrTBvG+e4lX5q0GK58DwFKtuvwXcBb2VwuhmtIAhYW1ujLEsODg5omobpdOr3pq7WtO47aiirSvrhxaFL3KDxxnMsQAyoSpPqfqhlLeVe6KhD3as0UHinn+e9wqUwll1PRyPIMAxJ09RT9bEuXRvICR2Px9If6XripC8yd39X5HlO09S+zqlRohod/eJCQvK88K+vmpXeaLnBt8YYaRJ3hk8F1AG/IVmnIbtq0VhFi2pY9f/KxO0+Xg1vN+3aBu05Twwgz3N/7rrHsNKSNf61gXOGVT04neEptc5elGBnZ5f5fI61LWenpyRJwmgkkffh4SGP3XyMOI55/UuvM51O/TldLBaMxiOiMBYRcxMQxTEG2N/fZzQasVwsxbGLhOxlrfXi4a+++grD4YjdnV3KsmTv7l2MCcjzgqKsyPNDsixjOBoRhBHHx8e8fusWbQtFnlOUIdYxsKtKomMsJKlosBZt4daa1OxVZGCxXAixx4358jrGxjg1HhkqIJubOIYqgG7blsBdU2EQegF/beeyCDdALltLFIbkhdRMdbJP0zQimZfE58oLPX7z6PZkd0mOsMrS6R540fAkScL29jb/6l/9K4bDodTRM2Fj7+7uUlUV0+nUG7H79++7uamS6RgMBpycnACyP+3s7Pi5w4OBjJV78sknhROS5z4jNhgMVhrM4IeKr/rbV/vfVzrS/P/Ye5MmOdLrWvD4FB4RHnPkjEwMVQWAVaRE02ObsU1Sa6eNFtJWW/0Faf3eTgv9C5lMJjP9A73Xm34mU7NbTZZEVQFkYUoACeQc8+Czey+un+seYLFaJTX58gn4zFBAAZmREe7X73DuuefeiGBZJenwZst2+KJEL/btcR6xmoVwn+W66LlUpeAAbJBvtLpDaTDy70JOoDB0HIv0nvSCQn2NMrgkpT5s0YdMkmQj8FV7k1XCT5IIeSRJRC5P9DahBl7tE5DUAwj0IWSlUmyA/U3+TCYPVeH26uFnZp+S3/NBwQciU4ccvV4Plm2j2fRQr7s4OXmDfr+PFy+e48GDh3j69ClgiDgGyTXr1Qp5Lqu9SJpybEn6sjTD7t4uxuMxojiGW5NFzWkobNHVaoXVeo3xZIxup4skTRFHAVrtFnw/KDRiV5hMZ6ILXCx5TpIQWQ4YsYEwipDEcTGmITBqcB0oLO+6LtarFSiaERaiCuPRWHa8GrJcOk3k2ak36oijEu2oktr4Z7F/UwOw53lwHBtOrQYDgG07sGxBZ4jH5nmOMBLGLben8JlM0+SX35wP5191qu0eHZ2r9Bf553f7jfRX9LlRFGF/f19VojzPw2g0wnK5RLPZhGUJ89v3fZyfn2tQDIIAvV4Pk8kEn3zyCeI4xvn5OdrtNiiiwoSfW6W4ZQqAVqlJkmiApZ+9CWzcGxEs2T+zTAtplqFWkFOq8Cw3kFSryOqAc1m1EScvKyvOO3KFkGM7CKMQcZxgsZjrDWRQLEUIChi3yJ7qdbdYkxRguYy0l2pbFmqOo33B2WwGz/M2KmMaalWP1TRMNLyG/rtq4FaMt1qVctyFvUpK61XfO4k+pmluMGtZETPJ4HUFAM/zfn03+4Yet+6i5tTgeR7SNMV8PsdyuUS9UcdsNkMUhXjy5IleU6/poeW1EIYh/MCX2cU4hud5wpy1HcznEuAWy4WO8sxmM90JOV8sAADReo2aU8Ozy6dIswx1t475Yo4gCDCZTOB5nkjRBT7CMEQUxwAyzGcLPHv2HO12Gzu7OzANEy9evMDtO7cxHo1VAMG2bUwnU2RZhm63q0pSX375Jba2ttDutOHYTjGqMlICmCyDtoW0UxzDMJS9nSQpoijUZ499VxLkCNvFUQTPa2mrg0kwIMLxbBl8OP/2U21BsY1E9Iq+gveDPiIIAoxGI6zXa9y+fRuTyQRpmmJ3dxcXFxe4vLyE67o4Pz/H5eXlxuv4vg/P8/Dq1StkWaa7SwFJ8p8+fQrXdbFarZSJH4ahjkTdu3cP8/kc29vbsG0bW1tb6h/b7baSLlllEnFkgk8/9ssC/6/i3Ihg6TjyoFi2BYgYiF4INoiZJbPSZAVZlawDANM0NEs1DFP7IsTJOezvum4B1QobloE4TVNEYSRSenmOLM20Hxn4a9iOg2azqdsjGo2GYvKGIfsrwzDEZDLRPiGXL4dhqNUf4YjlcqlGWCXn0HAAqNpJloqmLclGJOiUq8akD+oHvoocEN6gcVVfixX9h32WBYKd53j0+JE4BMdSsX3HcVCLhFlab4h04Hw2h2Vb2Nnexnw+h2GaWplNxhPM53P0+32sVivs7u5gNBrr6y2XS2FQ58ICz7MMiyxDGIrNzedzkdULIzSbTZyenuL66grNpofReFTIF2aII+kPNZsNPH70CLu7e1itVnj18hW8loeLswtsbW0Voywr5Cj1g2ezGe7eu4vrq2t4LQ9RFCIMI3ieh0ePHuHBwwcF5CaQrWVaJbRvAHEaw3ZsldZjMhnH8u9pXLYxpO0wEYZ2UWHYtgPHsWHbzoa284fz7Q+LBgZCwzA02MRxrIIb5CpQUeqrr74CIPuEx+OxoCYAXr9+LXY8mWB/fx/n5+c4Pj7GRx99hOPjY9Uobrfb+NnPfgbLsnB1daXtpX6/j8vLS+zv7yMIAi0eoihSQf9er4eTkxNYxVYc3/exvb2N09NTPHz4UPugQRDoliT2PXm+bq3brxKqvRHBstlsbMwsGkahNmOU8IFosAK+76s+Jx/gVqst7Ls8RxiFWn0Rplyv1wpp+msflm1pFkUYlFkxiRecJUvNtCBs+MgzmRkjtEVjmEwmMlOWSi+VwgBsmJOYJD9nicViiU6ngyzL0GiIJul6vRZ1mKJpbtmyqqsKUTuOow3xaq8niiJZcl0kHaZh6kjMcrlEnMTF/sOaiKynJTxMiON9P2IrGR7cf6DZsu8LtM/ZSdd1cX5xjsFggHrdBQwD8/kCs9kMbr0Os9g7OhwOEcWRasq+OD5Gr9vF69evkSSxzC0mCaJIlgYkSaKrrhaLOaazGVqeh9lsjuPjY9y+cxtpkiAI36LmOFguFnBqNmy7hnqjjsVyiXa7g8urS4zHYxnvKNCQ6+trDIYDnJ9fYGdnB+PxWKGtGEIIenPyBru7u8hzUeY5un2E2XQGAwbm8zkuLy9lhKbYyiLJWlwE7RymSU3iRPuUAFQaMs/l2YyjCGmawa27yBP57JYZqWLWh/PtDwMlCwHqTnOErNvtyiaaopJcLBYYjUZ4/PgxTNPERx99hKurK1xeXmrw8jwP0+kUl5eXMAwDx8fHCIIA4/EYjx49Ul9FeJY+mUWB74vOMCUX6/U6PvnkE1xeXsLzPEwmExwfH+PVq1eIokhfh9XtYDBAHMfodrsK/TabTS0Mms3mBunz6xi3v4pzI4Ilh+wXi6X27ExT9vXBgM5fksBDiS/XlWw/iSWDouPhRSQGzos4nc6QJJIxzecCvw6HQ4U7yWrliqE4jjEcDtHt9hCGAaaTMfwgQJJkWCzGAGRsxDItdFptrAMfZ2dnSNMU3W4HvV4PL1++RLPZxM7ODizbQstpo93uYDYXtZeXL18iTmLcvXMXBwcHMAwDb9++AXLAcWWDRBiG+p4A6Hs1Lamcm82mfP6oZNdWmbB5liMIQvh+sAHjVleBve9nd28Ps+kUT548QavdUudjWRbWq1UhhygbPrI0w3K5gtfy0Ol2cH5xjvv3H+Bf/uVfYFomrq+vYRbC6wCQxDFOT08l41+vEYYhRqNxkeFPkaaZsktZEdTrdViWhfFYhMavrwv1nm4Hs+kU3V4HnXYXXtPD559/jlu3bqHRaODu3Tv48Y8/x87ONprNJq6vr1ErnpPj42McHh4q/Nbr9bC1tYVarYaXL19iMOgjjhOMRiMMBgO8ffsWy+USd+7cQZqm+Pwnn+Ozzz5TKO6LL77A/fv31S7dupDG2K8FAMM0YRsGkiSFbTuouSLSnlu5qGUVBLoojL7+xnw4G4cIFlmlvu8rYcYwJLmxLEuJNa9fv4ZlWbi+vsbnn3+OIAhwcXGB3//938erV6/w6NEj7OzsYD6fYzQaIc9zbG1t4fnz51itVnjz5g2ur69x9+5d/NVf/ZUm557nKdeh2gv1PFlLt16vdY8rmfdbW1u4urrCeDxGr1iuPhqNsFgscHZ2hsePHyOKInz++ecK1dq2Ddd11b5JrguCQJE7onD1ivjFryJ43ohgGYaRMvYsy1YiAgwoVGrbtuDWRbUVhqJSwgqJLD8GSN/3Ma+sUmJfJIpK1ZpGo4HLy0vdb2kYBi4uLhRKdRwHV1dXklU3m+h2eqi5gTLFCHuu12ucnZ0hgyxWns1mmM3m6Ha7+PTTT/H27VucnJzg9u3bqLmiEtTv9bG9tY2trS2cnZ3hpz/9KbrdLobDIYbDobz/2RydTgedTgeDwQCvT15Lb2i0kvfT7SIIAlxeXpTLg4ueUBzHMEyjWAxcU3LRer3eEHcgoep9P5PJBEkcw3VrMg8YyUo1skttx8B0NgUgKEQQBnBqDp4/e44wDPFP//xPApEDSFNJ7JIkQRRHWK9WCMMIaZrg9OwMYRDi6dOnGA6H6PV6WCwWuLi4wP379zEajbC9s13YbIbxeILT01P0+32cn5/LqEcBtaVpCj9YwbZsjMdj3Lp1C8vlCshz9Pt9nJ6eij1OZ9oOuL6+Rr/fVxiNDnI2m6PVamlCulqtMBwOEYYhzs7OcH11jU8/+xSvXr1CrVbDcDjE7du38eTJE3zyycfw/QBZlsqC5+L5cmy7QIRyDbBZlml/Ms1SRHG5I/bD+fpTHdpnf1DHywrCISvKxWKhleTZ2RkajYb6wS+++AL37t3Dxx9/jMePH+Pq6kpFNNg6Wq1WGI/HePbsGdbrNd6+fYvDw0M8e/YMQMm+b7VaG1wMy7JQq9XQ6/X061j9+b6Pk5MTNBoNbTu5rovd3V202+0NrWzTNPH69WtFR+ijRqMRDg4O0O/3YZqmJpPkerB1xa//VRQANyJYpgVDVCBHG6YBpDkKHN5EmkYKfaaZ9IXYe+GFT9MUi8VC2HxuHXmeYzKZFLOEslLIsiztoQwGg2I8wMOiIFqQAMFsp9FoII5jzOdzNBoNNNw6anUXnudtVKD9fh+rxRJhHGn/c7lc4uLiArdu3cLu7i6WyyXm83nR26mrIW1tbWF3dxcHtw7w4tkLvHl9gtO3b9HpdtFsNoVFGUWo1+vYGm5hsZDlwDTqTqeDvb19rFYrLJcL5ADMQIIfja/qXE3ThB/IKANVkT4ES3FINdfFdDpVgpQf+MgK2FC0TX006o1CDNyAvxZ1m5bnodn0hDqf55gvFkjiImFbzPHzn/0cb968wf379zEej3WujQgKBdEXi4U6xDAIi+CS4uLiEvV6Hd/97nfx1VdfIQjWuHP3Dv6vH/3fcBwRd8/SrLA9C2mW4uTkBFtbW5jNZphOptg/2NekkSLvFxcX8DxPBBb2djGdTtFut+E4DqbTKYJAdmCenZ3hwYMHCvMZhoGf/vSn2N3dxeHhIY6PBT0JwxC3bt2SPlOjqXA/RemNzECe5TCLvbJ5LgpSeZZBKbMfzi+cqoBANThSyIHQO6FWJiW8j+PxGKPRCI1GQ/uOX331FWq1mhYG7C3+/Oc/1ymAKJK5WL4Gi5adnZ2N+Ur2slldck6XAYx9yePjY+R5jqOjI9i2razXsFB/sm0b3W4X47Ggduv1GsvlcmPO/eLiQsdc4jhGu91GEAQ6usI22NfNaf57z40IlpZdkgcsjlUkCXw/ULIOZ9qqjp0svOpYxng8hu/7WK/X6HQ6yLMc4+kYhmGg1+vpIl9ujTdNU1iLlPmq1/XGz+fzUvQZwPZwCx104RYBs9vtamUp0BMAQ/YItloteJ6H58+fo1arYXd3F47jYDKZqFFU4YRBf4D+/9LHm9cnOH55jNVqhaurK3mdlvQQxpMxWh5Xmcn4yvHxMdy6i+2tbbRabUxnUzRbTd1heXV1hSgK0e8PYDtioFkqOwlX6xWajeaHOUsIVLparnB0eKQD81EUYjaf4+GDh/jqyVewLAur9UqJWgXgg+l0ijg+FQcThtLTCUP89//jv6Pb7eLVq1fodrtCy4/FEXU6HdiOrQFyuVxhMplgNBrh5OQEeZ5juVgq6/vVq1eYz+e4urpCHEf4b//1f0fdrSMIAnk/lo3rq2tkeYooijEajfDixQvNwqezKRr1BhbLBfr9vqIw19fXaLVaiKII87kgGawy9vb20O12cXl5qX2wwWAgcorF9wZBgMOjQ+1bvXnzBgcH+3j9+jX6/X7xPMlezDiOhWeQipCGaVrCGcjxobL8/zjkJxBBI7+jmvAyEVoul5hOp5hOpzg+PsZwOMTPfvYz/PCHP8SjR4/w/LmgIc1mE9vb23j+/DnOzs7w9OlTTYaITlUVwwCg0WhsjKeFYagEQVabbPNUNzHRjsmKrc7WkuFKn1kVQqCvW61WuLy8hOM4ePz4MR4+fIg0TRUp6XQ6KozgOI4ycKujfP/ecyOCZd2tAwZUYxMgkScqtoDIhWvUG4iTuICNZlrS+2sfpmWqITHwHh8fK1wRRRG++uordDod0bz0Ax3k54YGZkCs/Jh5kXW78tdIslThhkZDaPm9Xk8NglJQnudhMBhgd3dXg/dyudT+7LroXRmGyIORPeu6Lj777Ls6xzeZTkRjs+0iTRNtnvMa1Bt1fPLxJ1gsFnj16pU4nRxKCR8Oh+j2ekiSpBDTLreeNBoNrLKlspHf59Nsetje2YFjO5hMxpj4E9QcB7Zt4+nTpwiK3lCn2xXkYLHEfDHXEafrq2s8efoEBgw8ffIU3/+t7+P09BSDwQDrtY+dnR0EYYBup4vJZIKTkxPcvXsXk8kE6/Ual5eXAtcW8owU5XdqDkzDxHQ6000l7bYH0yx7zlmaAbYQu/IsU1sGhBQ2m83gNT0s5gukWabwFqG7s9Mz2I7MQ2ZZhmazifV6jclkgm63izRNcXFxgb29Pe0Z+b6My5ydn2FnZxt5luvXf/nlIxwcHCBJkyJ4HhROU7gGOfJi7llIQnESI4k/zFl+3WFCzaQ6K+4fq8o8zzEajYStX7Dr5/M5rq+v8eWXX+L+/fv40Y9+JMvE793Dq1evNKgOh0N89dVXePv2rY5j9Pt94UMUwXC5XCoKQr4DfR39SLvdBgCt/tjLZFBN01TtzbIs+L6PiwtharPY4eunaYr9/X2EYYjZbKavV6vVtKd/dXWlSYJhSLJ6dHSkfrHRaOD6+hrb29vo9/v/v0GzNyJY2o6t84RxJKMPhBEAFPOO4UZDebFYCAyaZvADH7btFNT1GIvFAvP5XKHKIJABbfn/ETyvVWZIRqlFyF++7wskbJfrrGgkfjHrxlnPVqulcKdlyaZ5UYIph2g7nQ7a7Tba7TZmsxniOEaz2YTneUokqW4LoRE4joNGvaGNc8MwEMUREq4XM4Us8ejRl+j2euh2u/B9X1ls9TpnBKNCJq1bSJhlwkDMMwRBgvUHXU4VPid6cbB/IH2f8zM4diGzmKYYj0YII0niVqsVloslRuMRfvrPP0Wj0UC/34Mf+Bhdj3D//n28fPkSk/EYv/mbv6HXfj6fYzAY4Pr6WkeaOEYidlZIGxbiGlkmuqqGaaLZbCixgexpmV+0kSQxajWnINMUq+EKzU+jsE/HcRCEgQZGw5CVdICBOIlwfn6uUnVv3rxBnufwPE+rhyRJ0Ol0EASBCDlkOVartY7FhGGI/f19hYHbnXbRV7/EYDiAaVlIg1SDZJaLYMjXjQF8OFDmvzCVS7ar7/sKk1I8Is9zPHnyBK9evYLvy+aa+XyO2WyGW7du4dWrVzg7O9NK8YsvvsBqtcK9e/f03tm2LYIVhd90XVdFXjzP0/vEUZTq7DtH7JiscVUgg6HrulpZTiYThVIZVInqMWiyWuaYH39ulcSTZZmidVUZPs/zdJxlOBwq0vbvOTfCQqMogr/2EYSBzkGyx0adQd/3EYaBioKLyo70L1utFkajESaTMabTKdbrNebzuc4attttrP01GllD+5fV+URedJb8hAjIxGLGwiBWjhWUMEI1oHJWkuMovu/DKeYzqzfeNGXlEiHki4sLZBAhhDCO4NSEVegHvn7OTruDPBciUbvdRhzH0rPIcg3SrXar2D1YXkuyLO/fv6/aj69fvy5WU32AwKICyrz/4D6SOMH19RXCKEKeZRIUiuRnWazQms/nePbsOV68eIHhcIjRaIQ7d+9gPpe++Wg0QpIkOqfW9JpYLUVOj1R5QMhtWZbCLMagZIOOyMbNQpkxS5J0gx4v83MpkAP+2ken21EnEviBjobI+FRcJF520fsp1t4VmsiOW4NhFGMeMHSGOc9zhEGoffHLy0scHBwgDEN0u7JU+vDwEG/fvsVqtdKtKOPxGP1+H0dHRzg9PcV6vcbh4SF2d3dxcnKC6WSKer2OVqulKAiAD33z4lQT9ziOVWKTQVOUvCztDdIHzudzTCYT/OhHP8JqtcLBwQFevXqF8Xisc4qnp6eKYHHmkbKhDJbr9VqZpQw65FowePFXFS4lmldVDGILi7bLP5OAeX19jf39fQ2Q/GwANgg8QClMw0DKZ4FfwzEVjvxRQo/vRca9/n0B80YES6HRl/NBHHtgI5ujE4vFApPJRC8ooSIy+MIwxOvXr4s5HFsZgIDAUaFLWS5H1SWyTAgccaEn6ziOrPSKhTnKG+M45fsCoFUg/7/ZbG7s2ATKAXA6L/YDqO7D5ja1WlutFuJQrkOSpYV4dYZGvQHDkP7jfD6XirMQQ6AQcpIkCMIAs+kM09kUdbcuiUbNQV70Y5fLJZ4/fy6920JUvboV5X0+W1vb8H0fX/1cepOL5QJJHCNJUviBVN7r1Rqj0TV+/vOvYNkW/IIt2Ol20Gw2MRnLjFin08FiscCdO3dgmib6/T7CINT5RJIpmo0mUFSTrZaHRqMhwhRWDWmSwK255TxZTZjgktTlQG4ijELUG/UNpRzTspBnOWzXRhTJPtMwCqWdAdn3atQNBEFYOEuvWA0me165/YeLoqn3GQalxux6LdfD8zzEUYxXr17j448/UnEO9kEPDg5wfn6ujj6KIqxXa9w6vKXw2ng8RqPR+KDgUxwm40TICMP6vo9Go6GJOkdGfN/HarXC48eP8cUXXygs2uv18JOf/ARJkqi263g81tlLy7LQ6XQ2NKopHMCqlSRGBrlq35QkS2q5AtBgVVUQY6+Sn42V52w2Q71eV/5GVTa0XsyycwMKAK2s2RabzYThTdQuSRIsFgtF4Kpbqh49eoRut4ujoyPs7+/re/m2sOyNCJZhWFZaLOdXqxWm06lewPV6LZT9gtBA2KfT6eDFixdotWQ2jk3mJEnhed5GFcdeYr1ex3K5RM2tie5ssbGDGVCj3kBolKuydGylCIRk5JqWiLjzZzBb5i8GSOoc8u9In6dE3XK5VFw9Qw7TttC0mtqnoDMjxTuOIwRh2TDn99ZqNbS8lvapDMNAnmbodLqYzuRaLpYL/TzL5VKX+b7vZzqdYjAYCLyUSX98WjzoWZpisVhiMp3gR//nj9Dr9QqW6DlWyxVWyxX29vbw9u1bDAYDnJ+fY39/X5Ojjz76SAe95/M5Aj8oVyFlBkzDVPRE+kIm4ligcjowIcTIvKJhAJZV2BlEBtEs5BzzONclzjJ+JdVikiRIiyXNor1sA8YvZv9sZTQaDWV2p2mKre0tXFxc4N69e1itVlqtbG1vKQFtMBjg9PQUDx48wHw+R5ZlODw8xHg8hmVZ2rs8Pj7G1tYWOp0Otre38fLlK3Q67f/BFnBzDgmODJr882KxULEJBsvpdIrXr1/j+fPnKgTw8OFDnJ6eqp/r9Xq4vLyEaZoqhs7kZDgcIggC1Ov1Ute4YFiz18h5YfZOWVkSYuWM57viABzbIzJXnRFlscE2FCvZ6npBMl3J5aDfJMpWlfBLkgTNZrNsW0B8PlndRA7Zjvq3iGDciGD57gVmNiPjEEttKE8mEyXLLBbi9B88eKBahbIzr5Q84rButS9UXdPFG0n2K2Xsqj0hNqirItIGNoXfqzh9o9HQfyO8wPfDG04iEbVnsyzTEZdqJlUVEOaCbNMykaWynoxjDvVGHc1GU4Pm7t6uQrRv377FdCrQl+d5uLq8Utbl1taWshvf+2MAs/kMbs2VRdkGEEYRFosFloslXr9+jYuLC71Hb96+wdnZOX7z+7+J9Xqt1frp6Sn29/ext7+H4xfHiOMYBwcHiJMYy+VS7Nf3i4Ao/aYc+Ub2bsDQuV/OrwE5koQbaURuDLkgJsvFAl7Lkw9RBFBKzmV5BtuxsPZXmkjS+SSxiLJnZq72nqYZLMvWBFAEGERF5e2btzg6OtKkrdFo4ODgACcnJ0jiBM+ePcOdO3dU8Wc0GmmwNE0Tg8FAK4KT1ydoek3cu3cPtZqjxL73/bAnuV6vlTBDxinbKmSoBkGAZ8+e4eXLlzoz3mjIYvFXr16h2WxiMBgoiSzPc713nE+kVGhVg5oBh7/TT00mEw2obEuRaAPI3DYhXCbyVX1a/qzqYonVagXbttFut6W3bhgb/tnzPE3aAGjArMYMFie1Wk2RGDJjJ5OJjrlMJhO8fPkS3/3ud/9NymU3IljyooZhqAEgCEIsF0vM5jOsVitMphOsVzJQ3+/39eF+8uSJltsUQWdmwhkhVlAc5jVMA91uVwKMW0On3VE8nGU+DZJZEwCd62TAA6D6i+wLAtDMrbrRg++RN4mZDueUqhRtrU7NAjorZpIAYe6GabZh0Ov1GlEYwiz6VP1eX5X+d3Z2sF6v8eLFC8yL1VPsgVxdXamhvu8niRMEfoDz5TkMAH4QYLVc4osvvkQQBri+ulbYy6k56Ha66HY7ODo6LFjVci+fP3+OXr+HTruNbqPa7H0AACAASURBVLeLly9fak+lbBlIh7DZbKq9MqNnPxqA6gkz4w6DGGkWw7KcoockcHG9LpC7ZUpgzQu4NbNzGFmq6IjYYrlMHVkhYmGa+gwsFgvkxfNDG7NtcTYH+/v6XiQZE+d99+5dPH/6DHvDPZyfn6Pb7apucpzE2N3dxWw6w8nrE3gtD54nvxbzBabTqTr4DwfqR0zTRBAEGmxIsGKF+fTpU8xmMzx58gSu6+Lx48cAgKOjIzx//hyu6+LevXtKhKTUIYk3nHFkxVdtx1QTdqJ6VTi1yu/I81z71fQr/BpKc9KO6JOrfcZOp6PvgyhX9XWqfBGSMokCDodD/TzsTbKNR44G22C0r/Pzc7iuiwcPHnzre3MjgiUzKDarSYxZrpa4vLzEarVSVYpGQ4bCq9Va+VCXjWp5ECc4Oz3DYiH6nXkue//M4uE3TQOuW8Ng0MfBwS3cunVLYNNmQ4MnAOR5hiiKyyqv+HlVw2DmBGAje+KNr24f4fcwkyO+z7+XSjiFbdmIERcGkwEw1Fmx6s3zvJhTKwZ30wxXV1eqp0i69/e+9z2cncm1yLIM5+fnG1ns+36CglGdpSmWyxVW6xWWBaua2xKYOHmFAMHR7SOYpoiCv3jxAp999hkePHigI0HNZhOBLzDSYrHQh54PNBELBsvq7ta42CFJ2ykXONvIM4Fao8hHngOr1bqwS3FQpiXEMcOAvqbodkZo1JsqdUjSj2EaCtNLpZlpMJZELkGem6jZDs7OznD79u2i71+H6wqUu7e3C8txsL29jVevXuHo6AhRFGE8Ege9XCzQ6rQxHo+RZRk++eQTdbjT8QQr/4MNAsBisRCpz8KXEK7k4oXVaoXRaIQwDPH48WOs12u8fv0ab9++VfibQvlk9RPtYi+QSTt7ldXF8/SfALQPWA2UQIkEMqhyDhKABkK2rYjmsSAyDENZt7TLMAxl61SlEGFAZv+RX8+gyEKB14k9VwZl27YVUSNySAF49m7Jnv3XnhsRLFerNdK03J5NyEFk42aapWp2vAZqNRuGkcNxLIRh6ViyLMO0kAgbj8eYTidFn6cksUhGlCHPDQSBj7enPq6urnFxeYHvfOc7Eiidmjoxw7BQqxnFVpJELzqDpeM48AMfbs1VB1Blf1XVLKpKF4QMAGjGxddjY5/Qr60B3txQ6RB4Itfvp0GlmTwwJA5lWYajoyNcXFwgCHw0Gnfx9u2pVNHZh40PcZJgsVhivV7j5M0JTk9PsVqutC0ger9d/XOWZXBrLuI4guc1cffuXSwWCwyHQ7x9+xbcOzrcGmqFkGUZ0iRFo9nQ8RAiDmEQqo34/hqWbYsDykUWjv+WW7m+zmwqY0EwDIHmi9eCUbQAnBq8lqetANtykGayrs40TREDsSyRnLNsDIYDXTqQ57nMPlYkJdM0hdduKSxWb9RhWzYs00S328M68DeW/nLkJAgC7OzsYDQZwzItXYNG2LpZbwAfppcAQPdJkmRDu1kul6r3mmUZ/vmf/xnL5RKTyUTnJHU7UrGpg4gRiYwyRRArC58IA4MxgxgDVb1eV+m5ryMBVv+eYya00+qShmqAo/+injU5HQzehIT5WgzmRFgIxwZBoJt93u3xWpalEqa0fQoUAMJP4GL2b0NuvBHBcr1ebTC8ZrMZLi4uVAaMTWmSIggNOI6rF4pZyPHxC1ycX+L66qoIwCWuzZsFZLBtUxwHJCMPggAnr08wuh7hhz/8Iba2ttBoNPQis/LjxaWQsLB4hVBBh1KFVkmeIRYPQJvL/J03FygDOTOkam+BmRpn6Vi9WpYFwzR05KBWc4BcMq/vfe97+K3f+i385V/+JdrtdrEpxUeSxGi12lgulx8qS0CIOqsVXrx4gadPn6pj2draQrfbLRiEJiaTCWo1B/fu3ZPrZhjIc6G6zxdz3fxuWqI4tb29rfBrkiRFoMw24HgAQjbLAcexEUcRgsCH69YRBL6Syji/WV0OALy7lsiAYQCT8QQA0Ak6ePjwYaG/LOMxcSLohAkDcZqos2LPyLZFa9ayLXV6lFJzG3X9O7PoqXqehyRJUStkAA8ODvD06VMlX1iWBcswYdccXF1dI4oivH37Fjs7O/joo4/wTz/5HOm37B/9Rz3z+Vx5DGwfAVK9U0OYMp7sITJQdrtdfP755+j1erhz545Kd7bbbVxcXCjDtaoVTf9Bf8OfTeJNda4R2AyQ1T9XgzGh1CovpHo8z9Migv6y1Wrp9APHRqqEo2qbgvZuWZaOy1FIg0lYdUSGVevV1RUMw8B6vcb19TVu3779C+/tm86NCJbMIJiFyu+SanI2hvRkZTpZtrL74kR2/8kC0mtcX18V0FG+cUMJWzo1EWt3bPndMiwl6fiBPMj1eh2NRgMANEAzq6NhVJvghmHIJvvKfk0apbxvmaVjpsXAXQ2YDIimYcFAAsMwC/gs3jDYLBeRbRoc93OapgHLtosBdVmE/eMf/z/4+7//e9y+fVuvR71ex/7+XSwW0gvmtpL3+VxeXuD09AyXV5eaxbIfcnp6iluHt1ArRo5ct162BNJUt79Qnec3fuM3cHx8DMM0dDynSkYQpxTCskpClwQVsc96o4EkTZBlRCgsBP4Mk+lUGIgAaA1Vu5CgWggYGGLvs9lcKPVND4a0wEXMP45hWzaaLWGHz+dzHex2XVdJX+QS8FBOjOS5PI/gOjKrWavVcHx8jDt37+D+/ft48uQJjo6O5PszeQ739/ekal+tVC6y3W6jPxz8Om7zjT90+NVenWWJAP7V1RWSJMHz5891ltwt1sft7OxgMpngzp07CIJA57fr9boGRhJy2McDSjiTf5bdp2tlTW/4nUzUnRqNBlar1YYMHgAVKKi2oQBsSPLRV1LsoBqU6ZuIvHFried52tdkMAagwZmIG9//9fW1Vs304UQefd/XkRlWtdXzTZXmjZgE5iAsMXKBCsQZVYMWvyYIAkRhJKu7skyZouPJGNejEZIkQ1708d694dVA5Xkt9Pt9hTMtywHyXDFthaOKRnH1PTDTkco113EMGmG1aqDDY6CtBnBWi3xAAHGElmkByNW5btCzYWg/Kcsy1Nxa8X5quhDaMExEcYSa6+Lhw4eaKAhRyMDLly9xenqqBvu+nxfHxzLaMZORh063s7GVpebUMJvNMdwabpCzHMeBVUjP1dyaLocebg1FxhESYJCX/UNAxkB474ksRFGspJ4kSZEkKdyai/lshulshoRksRy63irPMl1cLbQh+ZWlObIsR1qIXXBOr9FswHEcdDoddLtddDodJEmC3d1dRLE4jxy5EtSAgnCGspKgHRHSS9JEK8iDgwNMxqKocu/ePZyfnyNJYoXF8jzH7u4u3EK03jRlp+J8Pv/13vAbeuivKEDAmcLZbIbXr1/j5cuXODk5wWq10lbN3t4e6nXRCeZiBgq68LXod6p+jEGpSpqhL6wSeHhs28Zf/MVf4Pnz5/i7v/u7jQqTv0ha45IKBnPCpHyeOCnguu7GNhoGQtEJtzZm5Wl31R4uFcoIU7MyJpLDfipRSs63kxn+7vkmhuyNqCyZPbRaLVxfX2uWS3YqACyXS71I0+lUMvy6q4FsMpng+uoacZzAsm0gS4EiB+fn1xI+Fz3W+XyugUb6mjmyPIO/XutGhjRNpYy3pYKLwwh+GGglKJma/ABmbFUJJ8LD1Yb1JiRcbvxmBRkhRJzK58iLnyFGZsAw7Iosmqwxo3RZnueoOTUsFwsYBlB3RWxhsZgjz8vsKssypFmsfakP6inAolDeoUPf29vTObbbt28jzVL0+30AwGw2Q7fXVf1WoEAtshzNRlOF2C3Lwmw2ky0MBaRJJwSIQzANU5VL6OBoP3mWYx2LbmxV9xgoFOrkJ8v/8795LgiDxZ8HzKczOJZAWYPhQIOe5dg6egSIRjPHE4BiE0vRBzMtC42aqzPBHB0h89Ysfp5bdzWRBIBbt27h8uIC3XYHtaIPxp5ZEAR48tVXuHf3Hm7duvWrvsX/Uxz2e4FysfPp6SkeP36Ms7OzYk2boTO7hM7X6zWOjo40CLGS4/wuofQqy7+6jahaof0yf9DpdPDHf/zHyPMcvV5vY3yjiphVBSbYb6wy+gEoSgeUaAsr0Op749/TLqWn76v/ZZDnBijbluUEnEjg2JLjOLh165Ymiru7u9rW+9eOkNwILyk32N1gPRHL5p+Z/Vi2pYoNyEuYNgxCxEkGE6WYtGGU80AAyurMKOHPKIyRyew28jyDZcrSacJPWS6s1axY0JvnmUKfNMoq9braCwBK6JeZ9y/D+/l3pmkiNzZnlspgKcLvplEGYfpMGvlyuYRhlME1z3OEhYZkmpa93SRO9HOMJ+Nf162+sYcQ0v7+vpAPolghScMUoQrCS81mA5ZpqUNyi4SLvWOreDiZ0JTMv5J1SF1Pav2qziug7GjTshCEAZIkVaeQF0iGmNfmg54X8KtR0Tvm88PRlWp/X/v9Bcua70GQnlzp/0QzDJTPG1CMEEDaAmyJGDDQaDZwfHwsUHVTqtnZYgYgg+NYWukYhoGm5+H5ixe4urr6Nd7tm3voJ8iCXSwWePnypXI4qDLGypLP8Pb2Ntrttq6oqkKM5CRUq0ne+yo5psqIpa+q9gkNw8Cf/umfIkkS/M7v/M4vMFbZa5RnwtUgx2BMf8mf5bquMvbfnX3n97Aypd1xfIsSfCTgUe6OBDWgZBbX63X0+330Cv1sznSy9favPTciWEqWkSrO7HkeAGxcOOLZtmWiXq8pkylNykHawF+rxiV36FUZWSXcKFvaDVOqzqox0LkxWFHkgCMgsphaYF8A6jD5s+g06aRYDVRFDN59X9WKc0P8wCgFDzRZMC0gyysVjVU4SEMDszwQlhqZUThVflZWLnTOnXbn13CXb/ZJ0xRbW0M4NUe0dZMYgR9g0B9siO27rovBcCg24tbg1lygCCUMlMJcjvVh5vcaRmlrJIEZhqEJmW3burJKet9h0U8Wa2MwJImHr0m4XsaL5JRECANpmiHL0g3xbQBK2gnDEMvlEnEcF78nSJLNSiNHjjgpVmwVzjZHjpSthYLda9s2TMPAvXv3REUK4kiHw0HxWuU6Ka78Gm4Nv5XT+o98yBAlinZxcYEXL17g4uIC0+kUaZqi0Whoss5AcHh4iCzLFP0oba70C2TZ89+rqmJM0PiLX08/axiygPlv/uZv8Mknn+jcIu8bk0MGTAZhvi6DJ0mPDJh8Dsq2VtniYF+Togz8MwPrbDbTCptJHuHbLBN1t16vp2pR76oXfdv2042AYavZC+ECVd7Jsw3leonvso8xTUtyjNxwA4aRF/3BpCA5WL9YuWV5ESglIzYg+/VylMQIjlzw5vN9IsuLCqN0gqyI6fDUwVQ+E38+UPZN5ZQyUTQY3vAyQBezR2kKFNJ8uQHk0ryC49R+oYLNslTHbQQ2E3WWkjwAGKYM43+oLKWS7/Z6gjqkGTyvJevaGnUksQhZSHCQB3W9Eqar4ZaJjmEa2psMQtnT2mg0iky/VHmyCoQAKIUurIL8FUYh6vUGzIIctMyXEggLJyI5mjwH4viqVeQvJn95ActmWSkvxoyfZIzpdIrhcKgtjTwXdneWlz0mOrckTRBHkvHblg1iG3mewTBNmKZRqABZ2Nrewnw2x9bWVkHVb2nywGqbc8BVEtH7fGgPFEZ/+fIlxuMx5vO5PtuEH8mA5T5SQql5nqvvWiwWcF0Xi8VC20bkXVCcgIGMgZT6rFkm8qFEJpjE870AZUHDfZIAlMVPRK8apKuIHFXF6GOZjLLCZcJAUg8lTym6Th/KarrVaqHVamEwGOhoDGXxAHkuWckC+FZMWOCGBEvbspW1xY3XWaE8IlqXkrkqmysTp0BSDSCq8vP5HMv5AqaZaZNXstohoihEGMqcpGlayPJUH9oslZ8hu5tNbG9vwy0qy+pskrAfE8QFbCyOx9AHn1/PKpB4+7tZHoANOIxZHpUmmChUg7EyzwzIcDrEUF3Xhe3YcGwHcRIjjuIim3Phug2F/NI00/cCAGlaJgUkorzPZ29/T/u3vIedTgdZLj0Y27GRpWUCEoWSwNmWLdUd72UmkoSNegOr5UofaNsuBq1h6+xilmUKlTE5CsMQOYAkEYGJRlPWaxnv2JBRUFvTNNvouwg6ISQuQ3+ODG9zAXOSpKjXXUVPhsOhfO6ohIFt20YUlwPflmXBgPT6c+SKUJSMcVaVYpu2LSMqXstDEicYDIeYTaeoFyosQCnMwaD54UD71svlEv/wD/+A8/Nz3VfJAMA1agcHB2i327orktuEeL/CMNTq0/M8HUOpwqus5t4VTeHXsK/Xbrdxfn5eKVpKVbK9vT1V2GHlSJ3WKtuUQT7LZHFEr9dDp9PRVhNQ2gQr7NVKxgq5V5OEH9OUBQXD4RCu62J/f18JRdwUVZW/63Q6ik5y/vLbBssbAcPmFTYpM4RGo6kXkD060xL82Wt6Kp3EQVUq7UtAALIsx5dffomzs3P84R/+obJHTbNc+yJfWzSZUQha14Ru3B8M9KIzy4miSLMkQhhcFQaUGX0VTgDKXimdGinO1ey/+rVZlsJxZFUT4VrDNGCYJpxaDTClWqi5ApmwB8uv41gDIQc6o2qVa1mWLh2mvNr7fHhtJCv34NZlVMksGNVZmsK2ZRGzVUDclmXBdmRUhwEGAGpOCWMBUIflOA5Mq2QeEoLiv1mWhXanLYHMkQDtOA4MFD1py0KtJglYt9spkrEyKFZheKkCOBBeOrjye+TZStOKqHUxm5skCaI42tjTKqSwdMN+FD6rOcqgJPEoz4sFBcVz5zgO3HpdyWglAlPIQNrfznH9Rz1xHGMymeDi4mIjULZaLUwmE31Wu90uhsOh+g0AakcMjrRpktbot6qtGBYoAJQoRJIO7bldSDfu7Ozo/Hmj0UCn01HJORIbObNb5WhUWw4MUp7nodvt6v/TxvI8VxEPTiVQ15afudVq4d69e9jb20O73cbu7i62t7d1A0l1x7Bt27qVqaopW+WP8LNX///rzo2oLHkRBYIoHU8Q+Prwmqap2xmYmUgslQffNE3cvXcXp2/e4OLiEr6/KhRwgP/8n/8L/vqv/7p4SC1kGZS0Ixeo3BIyHAyxt7+HltdSxzYYDCqU/kRWXsWRGhONrorHV2EKNpzpUJg9VfuYNGL5XhN5Lv+eFrsMkUAZZaZhwrJs1Gquzk+xv8q1Tu9WtXz9amCmYX1gwwq6AUNGMfr9gVb6/F02eEADYNNrCvzulPNqlmnCrAtBzKnJTCazWNoH52Z5r2kT1WAWRRHcuot6o64r2api/3/7t3+L7e1t/PCH/2uR/Il0Y5YDBuF/o2ApGsWYiWnAa3mapJmmUdEbFZulcoppyqgMAF2zRIjM9wMUI8/y+uQVJLJf07JlnKqcw8wRhgEsS9CP1WqlEpWOQ2GGb6bsv09nMpnoxiUh6xkb0ne0GRJVlsulVkwAdNyiqghGVAyAQq9s71QTf34tfUdVND1JErTbbURRpMpAVXEN2ij9G4PmuwQiQGyKVSUXShMxYYCk+DunEQDZkNJsNrUHyc9ahXPZWgBKjW7DMJQ1zMDOYEm747X4Jju8EcGS+n2GYaDb7SCOo6KSC+D76wISqGv2QlybgueOI+onlmnhP/2nH+D582d48+Ytfu/3/jewJ8i+XZKkyHPAtCwhUxgA8hSe18JwOMTBwT62trc1s6qOgoRhoNAuqc8MlDSYatO82tSulvzMshjoiMnz54gBF0zD4iZalqUjI45jKQRC2JDyUIZhFPs8y9dTIQezHGfJ8wzrtVCwP2wdKYlXaZbDqTkwE0miBKovhCGKKo1sYyYfsuJNZltdRwguhiELpZUohlI3k4f/VrUFMlIbjSbyPEO328Xh0SHenLxBnuf4kz/5E/zgBz8oHElSvLbYi1k4Lmlrlr/zudrZ2QEAhecoOE37I5mDfTGeakD3vKbancBcJozCLquKP3TeZEGSGOfWXZHmM4E0SQqCmqEO7n0/o9EIs9kMP/3pTzEajdRGLi4usLOzo3OEd+/eVWWvqv8JgkArOTp+7r7lLCZVckjUok8jcsbE7t2VWaxCuU6rOjdJf1clEhHK9X1fx0dYRR4cHOhuYfZbJ5OJJgisPre3tzEYDLC7u6vvodfrASgraf4cVqnkafBnEQ6mr2QSXOWD8Fp8EzR7IyyUmRCAQqDZLS6Ei0ajHB1J0xiAqTNnkuUL7Eq1iizL8J3PvoP+oI/jFy8xmUyKCxNhI2nghnbDwGBrCzvbAjFsbW1p1sGZJ2Yw7XZHIQzZwbkp8ks6NLFyZoEAB7uF+Vhtpit5pyJYIE7VQI4SwlWY1ZJAJ/Nv6cbnZp+0mllxdpNN/Spcx/fMaul9PlV4hg9MlpIYJj05ER8ALNOAZdlSVeZMQvjAxYBhwKnVhJFcK+fEmGkTOeHDGUWRLtL1fR+9Xg/L5RL1eh2r5QqD/gDrtY/xaIQ/+IM/gGEY+PM//3OIIlSZUGXFDJQBIBdsFnZhu3u7e+j1egiCQAWnuYVEPmO+0Rs3DVMJPuVCAakwlFFYjKkABmq1ElER0pwkDNVZOsu2kMc5YBZEFrsmX5fr4/jen9FohPPzc7x58wau62IymajzB2R0otVqodlsKjLA53c+n5etK7NUweHqwX6/r2vm6FNpj1RmqraU+CxUW0hVW2CQZAUKQJWDaCvUsWZAAqDMVAoYUFwgSZKNKpKCNL1eD71eT98PfSyJn6w+q9KiTMD4XFDqj33danuB7Tz6/F+2henGBEs6cw5Bc8ZIshyBJgXOLJmjeW4Wsm5lwGg0GohiEx9/8jGGW9v4+Vc/x+nbU5GSg5AdckBnfPr9Ho6ObqPb7Sp5gxeRAS9JEhXtLR/+QmkjjmBn9obxsMlcZbmyn2jmpt5AZlCkQldZh4A46zRLSygWdGouwjCAbZdzS6yCJas04Lp1nZHL81yNhVW8fAYDa7+cS3qfjzj2vNjsIdWizNQKBGsULGl+MWFvQrSWJQxXC7bsVQXQaNQ3BK6BTT1gealSpKJer+Pg4ACj0Viz/KYn7L2Dg324ros/+7M/g2ma+OKLLwAYG8FIFSoAIM9hFe/3088+VUiLS36jMNKRFcrsVfvohmUgjcUOKfSepqn2yW3bKsQITE30pBKXQGk7tiZivl/YFxPUSrJoGLKg/EPCJuf09BSnp6e4vr7WxGQwGGA0Gun9Gw6HSqYBoM93lU3Kw/EgJtKUjaOPUzJXEXiqBES+Fn0YZx6rfoWVGmFfJvW0S8L7fC/VfuJyuUSaykiT4zjo9Xrqg0ksIsOVUCpJjUzqyG4lEliFWgkNs1Kuvk+gFGmpVqDf1JK6EcGSPR1mLy2vhTAIseqUbEKAFOSy6ouiTKnvVcmkuluHaRrY2dnBcDhUR8EMgn0k9gN4cfn3QMn04nvjDBphJhpZHpc9SVYIhCOqhB5Wn9WeJYNjNRsEWE3mSJIUNdQQ5eUCVRqoEELkNW3HRt2tq7MjIakarA1ThAiqjDiBZMLNivs9PVmeS+WTl+M9pmkWi5mtwr6Ked2CpSrfl5X94UxeQwkxSYJBf4DZdKbzjFQ0CcMIzWZDbcW2pWeapRmGwwHiuFWwt0tdz93dHVycnRd7SEvUQElipom0qFa9Vgu9Xg/tdlvXtdHmSCYyTRO2ZatcHVASd9JUVsQxoCZpCVGJ46zDceyC/WsXNiRznbZtyXgWDIWKZctPOQ+cZqkoIOVAbhkbsO/7fH784x/j9PRUAx8JLPP5HDs7Ozg4OCjJZYXPoh2QqEVRckKUVLw5ODiAYRi4urpSoQLLsuD7vmoYV5EVoHwOiEplWaa9UsKZ7IPSl7HIIFmI9lSv17G3t4fBYKBzmo7j4MGDBwrXh2GIjz/+WAsIkono/2g/rDr5NVXSJFAiRapAVVTZnL9kscIiiwnGN7UDbkSwlAe+gKbiBDW3hnanjdVqpcoi8qADtm0qoYdVHNcC8QLL3KGBNM1143wcx1KCJ7HOhpFFy71vlPLyvCby3NBZtKrSBStBMZwUJAfleY5mobtZKraUYsjMeNibZaBkVghA33+apsjeYcjatq2rtNjI5uuwf8RMnj+XztE0xSnR2Hx/jbTI5l23/qFfBIHj80LsO8tTLdIkAJoFIlEIPlgSMGEBNatk2PHfeO1rrgvLWqNWq20semZCxkyd9ykMw+L7RaqxKh5hGIZsuhkOsLW1hclkgsvLK7UDADAsCy3PQ73uYjAcoNVqbWT8tIX5fK5yZe/2aZR9bRRjIhWCWJqkqNcbcOuuOsqa61R662Xvl/2gLOPy8lgTNqNQoRLeUbG1xfkwvgTIcuLlcolut6uJt+/7aLWEU9Hr9bQa472tBksA2tcEyvEwVldbW1s65M/gUW3L8HXYl2YhwK9htWbb9sYarXdhTxYmeS7L6RuNBvr9vgjLFGMm7KOyakySBL1eD1kmgu18Jmzb1nVvVda57/vKCCZvwzAM7ckT4gVK8RiKqfM1qskqf9YvOzfCS1LCiw9tkiZoNptotVtIs7TYJnIFICsIAVzvUjK3WO4TPpBAKk6JWU+e53BsgXtoLBTWZSYnju4XZZeqcCodCpcxEwIls4uVJLO7r2OkAr+osiGbSSQLQyFnVlUIsgu5NMMob3K1SuD7q/Yv+fmrJA02v99lqb3Ph/ehXCxeEa1Iqz3LAqLNcri2bHqR5dsiQmDAAAzAdmyYqYlmIfdGAWgAWi2maYp6o65bPkzTFFYpANuuYWtLxgRWq5XaodeSkYAD7wCtTgtc5xZHMZpeS20ZgIpU8/NVxalllZiv9kraf5V0VmVy06HUaqWsmVt3i2JSFLMo1EEHLMGedg8khXOPonJzBAlJpvmv3yv4H/nQ1+zs7OD6+hp37tzBdDrFzs4OPM/TZ53PLXW0GQxlXVqy0ZMDoMHNMAysVitdsovb6wAAIABJREFUFMFnnwUDq1He76o/Y0JYFU6pVrDVFg8rUS6pp2g/k7R+v6+KOqxALy8vcXR0tCFUUB1r63a7WCwWG0QhVoaEbXkNOafp+77Cto1GQ/u39LlV+LlKhvy6cyOCpWVbMLJSao4XoNvt6gdwHAdXVxeFcThFECH5wtigA1cJLbYts2l0AApLGqWOK4XZaVwMhllWlubV76kGGBKSCBewV1ntfVbfE51IlWij5AyUDLIojjZGBhhkicVXjZmvz8BdkjewIehQzSYdx8FoNFIG3ft+pC8uoyNSRQob2bYsWMUOSMu2CianKVBjniNJZTOI2p4puIVjlpDQ9vY2DBg6ZM0Alue5yilWae+u62I2mwEotUJFuNzFZLLUWeThcEuzYSZ+pmlisVgoUaJqdzK/3NDKxLZLce0qSkFHl+cyJxlGoSZmjUYDtUIxSjbjADCgPVaZWy7F4tMsE7E+o4R4OfNrmAYsWEjzctTqfT9xHOPo6Ai3b9/Wftynn36KxWKhgQbABrLGVhIl4RRNAjR5YlAzTVMH+d+8eYPr62tFNgjjskX1LrTJn1ntCfK+UxTm5ORkYzTl9u3bePDgAY6OjnDr1i1N1Bno+Z5I0lmv17i6ukK/30e/34fnebrQIs9zVeThHDkDO30fma9k4LKqrraoqoUFUb1qj/WXnRsRLBntmYFaZgbLtNButTVjFe2/dtH7CZBlQjvnXCYPMyyZ85GeievWEceb+y15oapZk8hBlUE7jpMKeaFUl+CNA6BQCWEDBlVCJNUgVqUuVxV6mBAw6AdBoFqjtVoNOWQ7SPW9VzUOCcXQ4VZlp2gglLrqdruIogjT6VSCOckp7/kRJimkSjQNZEkGmMJgjhPZ/ZilgmzUXBGCAMreSC5RD6bB3rTcF27O8VqeQGdJjDAIN+jqYRjKvFijDrfm4vz8TIWe5YF30O229aG3bVvk99KS7FCFpvi6rDSq9HjTlC0nVXiXNksnx+dEvs9Wu3QcBzW3pglimmWwrVIKMssymFZZHZumqXqyWZYiyUWj1nGKkZlCGk8c1/+Q237jjuM4ODw8xP379+G6Lu7du7dBhqmOhfC+VGeCKWBAbgUnB6rseyoBMeGfTqcbY2a0nUajoXBmtepin5K+cDgcKvpH9u3h4SEODw+xt7eH4XAIz/O0iuNu4sPDQ1X7od+zbVk8ziqYpJ6q3bKgYNVaPaYpG0h831cUkBBudWqh6ieZFFRZvV93bkSwBLgMVHqETs1GHAvzDkYu/Q7DVJxehvazInst4SUGvFIdRPqbWVZCp8x4mM3TAFTSyyh7lTquYRqwrVKVgpBpo9FUqIBsLP6c6uLnd0WEaehVmERJGgUcTIiUlG6yJVlRVhlcfIiqPVUaH1+HDwsNttvt6pZxVqHv86HztkyzcOCizsSAx1VUQBFEHJnTzdIMqZnCyi3YNUcJM2YRRGpuDd1OB0ks8nW+78Nf+1LFVliMQRjArdUQxxHa7Y5uVRA7tgvGqpA3sixDzakhSEUsgNk6nSKXOff7fXn4iyQgDEOkWboRYAm1al+ycLIlnBarTXKki2o7jm2Dc8wSWOVZk4qjZECmxXNqmjIOVQo9bC4S+HCAXq+H73//++h0Ovj0008xGAxwdnaGIAhUlpBoE4OlbEMqkarZbIZ2u60+RkehKkkTZ2rpf7jMudrbribnwKbAClGSOI6xs7OjSdwPfvAD9Pt9DerkesRxrL12y7Kws7ODnZ2djR7k1tYWsizD7u4uWq2WStwxgFGqjn1J2hD/zGKGn4XBkFU1PwOvW7XXSz/5TZMBNyNYCpYKxxGBchnLKIabXRdGtwsDcoF5MxeLGWxbgkicxEjWibIJmY3kuVEYiLFRybFaJPxENlWVqEHc27Is3QTBhxwAPE/gAEJbbHZX+wnVAMigVoVW3yX7KNyKsl9aJe7QsJgMVOERVpvVqpdBu+oQ2dMlZEiN0/f9aPAws8IGSvH7NE1gmOXC2lIFR5ImAwX8mHMzhzzcNUeEMNx6HZ1uF2maCUmnqBjZ02PAESRDAgtJEV7Lg2M7mEwmcN0aajVL5b8AqEOYTqfo9Xo6+0lnSVsksYJjSBQkoJ0C2AiaVUSFBKVWqwXLNIvNKqxoHEjjMkeWo0gyZP9lFscFH4ELDkxwfRhnMWm735TRv09nb28Pd+7cwfn5Of7oj/4I//iP/wgAG/dlg7gHbPQQSXRhVcXqiwl2dYzC8zyFVmezGa6vrxWWZFDlUm7aLBP4agDr9XrY39/Hb//2byMMQ8xmM9RqNfT7fYxGI4RhiDt37ujoHVffxXEMz/PQ6/W0smWPkv1P/n117I29f14T+jXaM1t5LCz4NfSN786TkjHM5OKXnRsRLG1HaPl5nukFaTSaxY0VR2VZIi3WbrexXCzx4vgZgiDEshCrZmbAIe/JdAKv6SmcVB3ClexjqVkwqdOWZSlblplSFEWwTEsDJxvclGrijA+h22qWXDrUTAPXJvmhhDbe7WFWRYdZtbLfpGSUihEQ2uXnBKAQMI2kOsRbhd0+OCogKlAN1SjNy8FsADDMDI7taJ+xGkgEfZAF4pZlwbSsgiGaolZzkWbFfNyWwFWWbWE6mWAyCQtB/hyBHxQVXwzDADqdFhzHVXlFAOj2elgu52i1msXPz5Bl4iR7vQ5s2wL73lW7oO6rEuiKxJGQK6tJoJRUq0JmIq3WQqvVVlavZOdO2aLIs4LhWujDKpNYhAgyBk3TgJGLznNViSDHh4QNAH73d39XV6Y9e/ZMkaWdnR0dCWOfkKLp1TYRKywGD6JSq9VKlb14v5mMHx0d4ejoSNWDptMpViuRC12tVhvIloiztHF4eIjhcKgznyQadjodJEmCyWSCW7du4cGDB5qsc4641+spqjcYDHT+khVgmsqWExJ4GPhol6wueR1YvRJepbRflUBJf8o+a5WwWS1ivonseCOCpeyfNADIHkAuyQVkW4NpiGJKmqS6dPcglOFty7axXCw16yKUYFkWVquV4vZAOeMF5KjVHMRx2cuj5JNhGPrg8gblucwyNuoNuG4NzaanK2mq1H8AGpBptNXxj6oYQBUG4PexWuB7kqShUVyHsndUDXjMhvjZmFVWmbIMlKwo+YCtVptzrO/zCYPCRiJhdsZk+RVsTsuwEBdbOJjVu24NURzDsA1kqegah2EoTNhi5ZZpCYml5bXgm2v0+z2BjoIQQTHoz1nGIPCLsZQErtvcgCs7nY6gAJmBLAPiOIVhWLCdmi4akITLAKXldLyjVkOe5Rv3WmykMvdYkN3eJa+R4s++UlmJ5vpzgEKRp2YVbGImCg7iYr2Zkcv3GJBASUiWiM2Hvrmcra0tBEGA3d1dXF5ewjBEvnJnZwfr9RpBEGA0Gm30mgm1Vgkt1emATqejSAb9GX1PlWxWr9d1NSHVgNi7ZIHQarVgWRa2t7cV7arX6xiPx+pjyJ9gQN7d3dVtIfSNlKyjT2LSLvPzkbYTqgUERz6qyBmwuSOY1WsV1atC/LQ1wxD0RgmVReC98QQfPrgAGaFUODGAPENmSNbq1usiH2aZyDLJVFYrH2dnZ3oj2Ne0HQkunDkqM64yqzAqi5OpyVnd5l3tB4mxtCvM1+YGe7YKq1qWpTJoOsiebe4GJEOVxsvv5/Ug9ABA3wODIB8SvibZjdXKsro8lf/PpbL8e7LePoyOQJmFOXL467WiCsiBzM7g5BI4nZr0/iy7MkaSi56sEBQs5FkGWJas1jIM5IVdu/U6Wq02kiQtbDHGaDRGnldWJCWJCg5UR3yY5BimAaTQLLreqG84hSzLN5xDtWfJ3jR7PqZZticAbAgDUCuWs3AS4Ir5ZttSEg9QwP1OId4RxXAcW1nYnEllMmkVywniWJjqpmXAcSzUqurs7/FZLpe4e/cuLi4uNogo7J0HQYBGo6HEHVaaeZ4rR4FQJkeHfN/XQEj/RL9A4YMgCHStlmEYmM/nsuh8MNAEiagfk3kmU+xBtlottNttjMdjjEYjZFmGra0t7Y9W19HxvbDqZGBlJc1qkqf6uYAyuHMGn+ggAzb9IJ8FBkKiKQz05RggNlpYX3duRLAsVbqk9yEXEQWkmsCy7AIytZGmhqzosm10u334vo9+v4/ZbKYLU6nYA+AXxIalj1mw/eJI+yiOI0FYjKncR8nZHLKqpGfI3gtXaskFZpZWhUENo7z47zbNqywsBr9qRVj9+mpvqfp6zBbZiyAlezabYb1eYzAY6LoeBsogCGSRcVL2YN/38/A7D2GZJkbjMT7++GOYhokkFXin7rqICyfhr9cwiwc7S1NEcQzHdrBYLAQaK5CRLBMIMzdzFe03DQP1Rh0DcwCnSObqjQaWi6Uu1zUMB3leQqXsj0ZxJNrCxQOvwut+oE6ISjyWYakTqtVqiIMSimfPh0kbbYeJFntbO7s7aLfaaHpNNBvipCjKIEiQtZGsGaYBIzO0qq7bJaGi6gRZrdRqNRimgVYh58eZ6Pf97O7u4vr6Wu+VaZpaaZGRTMYrZywZKIbDobKveV+ZkANQzVPONjIA8+/IpqUtspeZpikGgwF2dnZ0qH+1WmE8HuPg4ACr1Uoh2ocPH6JWq+Hk5ASPHj3SgE0926871bYV1XXe/Xeifiwo6DPZl2232xsFCVDOjrJFRZ/OgMixGjJn/6dgw2ZZpsPNFEaXofAMZvHgV/tElm3Bzh04aQa7XewQLERym80m5vM5oihCEAa60PZd42EmzgDpGq5mS+xXciSERlXNdrKsutam/Cx0BlUWmQTFGmRZb6oZDZ0VA+XXDQFXiRYMnAAUElOYuMLEbTab2N7eRpqmyiiThnsPVD8yLVn2y2D7vp9eMdPbbrexWq1Rr7u6fzFOxIakgnJwdXUt2X0UFnBqoNm9bdvwAx9RsTmnUag6WaaFJE1Rd10ZpcgzxHECFNUp+8dxHGM2m6mYeqfTASDENZ1dLOyYtlDVtyz7V6Kp/G5yVrU3/jwGzuoYVMtrod1pa5+2yjCP4xg5yn2dKATTVbayEBjIC9UjkQQsVK1sS7dNyC5QA7blgILu7/sJggB5nuv2DbZVqpKdXGHFe8rn3vd9dLtdnYPkveIar/l8rsgZ7z2Z/KvVCldXVzrMb9t2oZ3dx8HBAba3t7FarWAYBra3t/F7v/d7+MlPfoI4jnF4eKhMaSaN3W4XH330Ebrdru4ZBjZ9GBMC2lF1Qw+wuboQKElH79ow7ZsVeHUK4F2FKgZFBs0sy7BcLnV88Zs0im9EsCQzM01SffCqOHSapbKmukB+skzmMOuNOtIk1Q3Z4f/L3pvF2pKd932/tWre4xnvfHsg2RPZFEWLgsSmBktCEsg2DBjwixLkIUgMIwPyYChAkCBBbNgvebWSCMiTEAgO4gdDBmIjkixYtiIGEkVxUIuD0OyB7O7b555xTzWvlYe1vtp1Drv7ynaLPN23/uiL02efPdSuWrW+6f/9v7IkjmLm8zlFUbDZbC4tFkkz9SdxyGKU9g5JEfSNptZOQzOKo0usWrmYEu73GYqysfX/vz/hQX5uI95tRCkpCbnwci7Es+pL+0n+XWoEUo8Q7116jIRSLelZpRRJ7KS0hOb/OCPNEmJ/A6WZF5YIZCJIhsUxTuMoIIoDinLDjcNbSLAvBrMsK5qmpr10jdz80SRNXIpWQZZmlKPSk4jcbbhcLju2oNRniqJgd3cHY1pcu9TWQZIWo34dRmj0fS1PSedJ+q1PiBNCRJIk3TSINEvZ291zRs+XE4IwoKndOo6iqKvrBzogCDR096umqmu00oSR25RD//3iJCZLM6dE47kIdV11kfgAOD095fT0FICLi4sufdm2LUdHR13p5Pz8vHPe8zxnuVyyt7dHnufd+pCyiwhUyLparVadOMbJyUk3yPnu3bsAXT+jMYadnZ1OuKVPjEmShNu3b3NycsKnPvWpjvMhe04cx91wZ1mL/ahN9tD3ymz165H9/VX2beGmiDZuP5UKW4Uz6Yzo77NS3+2zbOU8X/vWEeV7JftpRzkpsGV1gpN8q62fLWjA6K20W5a6WWuzdkZd1ZRV2dWdmqahNS7d1NRNd7Nfpa6L9+zSEjVab/vBjDEdQaEfGW7rMdv8t2xY/c8QI9YnUAgrTLyquqq79IKkaWGrtSnepkSfgj6t3BjTybb1NRCTJOlIT3VTU+SF32QHNux8tkuapZRFwf7BAW+/9TZRFHHn7i5f+9rXePbZZ7l54xY60NxtGlarJUmc8fD4iN2dOXv7u6zXa8bjqavr+JYci2W9WtO0DavlEtOrl4hT5uTnnFdelhXHxw9Zrzed5/3pT3+KP/3Tb/DwYdnd8H3HSdahpOhEmECMI1yuR4oQQBi6IcJyDPP53A2d9nMCRU9Ya91FkmGw7WcLQ9FJdu0vxhh0GJDqLdNyvjN3zlySujSuZ9GiXA+nqxMr9JDdAOg27aOjoy4CevDggZf7pKsLZlnmHaldPv7xjwN0ZJujoyPyPAfo2Po7OztsNhsuLi66+uTBwQHz+Zzbt293BmcymXRjwqQf8+mnn2Y6nXJwcNAZbhEjEJGV+/fvfx9JS7Jf/dph/3vKMV+tE8p+28+K9IMRqVcKp6RvEMUACzFJappiI/rqQlLrFIdDApb3wvUwlloRKfdFhOIus/aCQDtx9Th2dHR8vdFuR1yJco3lsmi50i7X3TZbxqHzugxyffoeST9VIIbGXajt49Ja0AkheI/YenZgGAZdxNlnrvadAEmTdKlgT+4xxhBGYRcF9JmKYmDFa+zXMfsko64Xsw07xlefbi3K/lEUkW/ygdzjMd+ZE2g3OWS5XHaGYzKZ8Jkf+Qz37t3jjTfe8K0TLXu7+xyfHPPcc89xfnbKaDxmPpuxWKxIYkegmozGnROnlKI9POzSVKenp4R+43nyiSf5nd/5F9y6fZvc91fKZAhrLV/60h9TNzVJklKWi0tjmGQjUkp1tfl+DVvWh2w+kioVYoeMTIoiNw4siiK00pecx/763ab9HGnJpbNAB5rSj4IK49B77AkysUXGnSm9lWYMtJs+4qQGhzQsuOyCkHustR0HQ3oZJUM0nU7Z39/nqaeeYjqddo64rAGA27dvdwLm0iL32muvoZTiM5/5TNdKIgxYIQlJu8bBwQEvvPAC8/kc2PbuSn1yNpuhtebw8PA9iTFi5Prp1z6uGsL+a+T/rz6vP1NTniMOY5+Y2Z/KI5Oa5Dz1SxaydwpZ6r1wLYxlN9qIgLZy0w1EpecqU8la1/gtj2vt2HYiT2et07OUvp6qKtEarNW0rUg+2e6nLCypI12+aNsLAVspOyE5GGuI4qg7LinGy4WQx/sekaRA5Xd5H/kuHZuWrd6m1rpTbpFzIcZaPk82QUnhysK52owsHpakTSRV97hD0oM3b97ije++weHBIeA2iDAIOT4+ZjTKumsRBIGb2djU3Lp1q7vR0izj6J0jTk5PeOH5FwjCgKR1/WTnF+fESUJVlty6dYvK1402mzVPPvkUP/ETP0HTNDz99NP86q/+KheLC59tqIjimJPj464x/CpzTzIzzriZjkAE24yHNGtLJDmdTR1ZLo6Io9j3Siri2Pd3sq0VGWv8+C9Rj9pOYgkCjTWG/f0DtFbgI8nQ1yld24onAeFTu0qDVlicMyx14ccdJycn7O7uEgQBb731FpPJhDt37jCfz0nTlDfffJObN292ZB1huYZhyGaz6dpFDg8Pef755zut1ziOeeWVV5hMJjz33HM8+eST3Zroi+hLX+3BwQF7e3u+x3balYhkXJbgiSeeeE9DCVzaG9/vOX+ev/WNaL/zoP9YXzlIIsV+S0ifLCQRqDDOJUB5L1wLY6m08gQJl5JtjR/UWfmeL4s3KEIDjrF2m7cWI+cYde7ixEHs6yWR63kzljBwf5NRVxL9VVWFwXRGxZ0w1dWjlPJRoTds1rq+y20KrMFaMKZBKdsZMyHrSON6P1qU9Ji1loDtRSrLqrt4o9Foq1Xq+6aCYKvgIlHkVY9SjKRsoBINuF6+ovtdCuTiqT3O2JnNO8diPpmRRBFx7Byb6XTKbDZjtVpRlgVFWbC3u9f1vSmtOD095datW4Shq8vdv3+fOI754he/yI/8yI+4Wal5TBIntGmKUpqqrrqWpY+PxpycHGOs5cEX3+bZ554FYLlYEicxDx48YGdnh4ODQ1/LgqqqL2nBStq2Na1zQL1jJJPlw2hrLJM46a57GLpWj8AGW8GMQBMoxxcwravtkyT+/2NkIk6SJoRBgNZBl3XR2omsV3XVOaFV6epdTdt0o9AkhReEwSDm7/H000/z8OFDiqLgM5/5TFfzW6/XNE3DU089xWg0omkaptNplzEQVn1Zlnz2s5/l5s2bXd1Oa83p6SmTyYTPf/7z7O7uMp/Pu6gLXJ1SIkYxQDKIWaLOPnfi3SLCd0P/M/5tcDX9Kr+LY9iPWsX5F4dSnidRpWRYpF4p9f1+p8G1Z8PWtfQXmo5NJ2nR1rRd5KUUneFr7bYH0hpnvITVJRtQFEWYwHbMQNH1pHFpU0llxrHTpe2fKBm1JLUhMXK23Ua6UpMU9aFtmjXsmK1Xa0zvptoDdI+7Ru7LAtcS3cr4ndlsRlVVrFar7nsqpTrD6s7VZYUKdz5Vx/qSz3i/HP3jBB1olusV49GIg8MDtNIcHx9T1iW3b9/pmrp3d/e2xglLXuSul2w0oq6bbtCu1m4U0he+8AW+/e1vc3h4yOHBAX/yJy+7VJFWmNawu+c2rqZp+Pa3vs3tO7dp6oa9vT3A98oFQUcwW1xcsPIju8qyBOvS/0WRk+cFxjq9Whn31jQNcZI4HVdfDxPSTmvcgGfnSLnBBLJxaK19b3PS1RvTZDtzUiLrMAi6NKojBFkqL3NnjcXg17oXmm/qhjzfUJaVd9Qimqbl+OT4h3Ldrxvatu2iSBHBB2fMxAErioLZbNYZsziOO3LYxz/+8Y5c03eUj46OOmFzGfUlBkLS8XEcdyLr4kSLU9+vLV79+ReJ/mdc5bNIVk2cRNnPge7YpSVG6vuSogW+j1PST82+G66FsVyvV73fXJ0Si09ztV1Uae32iwihoW3dQNq2bXzhegxsT5ZjC2q0dsoiWEtFdYlW3E9hiScnJ35b97Fe0EB3xyEGysmMhWgtUmPbHjTYXvC+J7TVJ1R+7JO6ZFSbpmGz2XT6jfJaMcB96SrpJ0Jv1YH6z5XI1k0uqLvvd1VF6HHGerMhL5zAuTHW9dTGITp0G87F4oLpZNqlr2VzunnjFiLTGMcteVH6Nesitq985St84hOfYL1es8k3GOsmtR/eOKTIC6q6Yj6f861vfovDQyc0fXJy4kgb1pLETijB9fiKOMbcibUrN9VjtVz5NQh106CAoizdWDDfn+n+86IYVqaFbMdywVZ9SmTLqqrEDSHYZjBQitBPIjHWjTPTgStLuL/j2m3ampPjEwB2dt3A4rIoOT07xbSmSx8WZUFZlB1j9nGHkGqEECjXZm9vr8sQCEtf2kDA9QzevXu3ExEQ0ouUdLIs69K3kraVlKo44f09QyaD9AXIrxP6tdA+V0P+JudIokypS8ooxqZpOnYwbMlI7/c9r8UKFYq0THsQAkwYRl3Etu1ddGLN1kLrpy6Ix29My2q17PQ0jd02XctdLAQiMSJiKPtN20BnTLXSvnZjuyg28AtqO0/NHbf0isqG47wU+adwWp5ubqLxDeRirKqyukTkkXOwXq+BbaR4dVqD9EPJ5IF+Hr+fmhEGmfQD9tV+BpIPlGVBGLiUfOvTNW3rZjFeLC6IwtDL0bn5jmEU0tQNVeVSlm1rybIxi+UK68fHQctzzz9DEmcsFguUUnzsYx/rzn2WZbz62qscHx/z8PghTzzxBFFZ8rGPfYzT01OeuP8Ef/Tl3+WF519gZ2eHOAldy0ASUfhUkt207O3PCYN4G+EpN8UE8O0ZNRaoyhKtt71qeZ47Q9Y0ZKNRtznXXp82imLixBHCUIooignDgKbxUosW6Na9ywyJolHbtuzu7RIGTp3n9PS0q7MFQcBytfRqP64HVTatxx0ioym1RkkrHh4edhu97DvyuzwnSRJ2d3e72p0YyqIouHHjRmcs+ySZq0YVLqc+rwv6USXQBTYSlPSNpqRjpT2m35MsZEdJ08p7y75/7VtHRIFEwmqJqoR8Yo0bshsEMqW7RYyXnKzlcsn9+/d54YUXWK1WHB0d8corr3RDd5U3sEo56rwbeWQ7r1suhjNGANu2iyAIUKiOidsZGIuvf/qLqLbKKyJaIO8pvY5ShHcRcUoYxsxn236konBCClLHBdc2slqvCIOt1yeiCbCljEtqQiJHoUML5P36XuS7KQM9jphMJiivsSrSgTrQaHzLkPfwZb25erVLUbVNw6ouKcsNURRgjZOYm8YzqrpivVrzxBNPoLXmwTtvkyYpVV2zXq8pi5KPf+zjHBwcOFp/XbGX7HLv7l1e+c4rfPrTn6apG04uzrixt08UuejiG9/8Nk8+8SRaBcSxWwdhFBF7dZYwCLm4uCAbZZ1xm/q66zvvvMMnnvmEZzla0vmOa7MqK05O3UDwyWTCeDxik2+wxpHq0szNhe0Ue1AURd6RS9brNXt7e0wnE45PTqhK18N5fHzCeDzm4uKCqqrY399jOpti2hZjneCI9Pg97hADsL+/3xnBLHPkM5lTKaQViTolShQNV3GkxSAuFgueeeaZ7u9y7/dLNdcdV7N0AokuxbEQxqso/Mjz+yO8JECSiSx9Yyl76rvhWhhLrFPlGWWjjszSZ2leZpYCFF2vmFKqE4n+zne+w9tvv91FU2EYupSR3p5oSZ1KPdLNu7zcgtG2jiDRmpY4irvX9xVz5CSbxkWVEqU5Q+r6Mx1Zqe48PaHsi3qGM44lb739FtOpYybGcdIt6Ol02mkx7u/vc3Jy0qUXiqJwEyyC7RxDYcURAUJMAAAgAElEQVQKI66ryUZxV/s17Vb9Yivs/kO45tcMaZqxWC6IwuiS9qmxplsr4lxFYURdV8RpSu37tqIeiausS+LEbXpxFMNYOWZr2xImkSeTKdI04sUXX8BaQxhpZrtzjLWMJxNGWdaJkAc6oGxrzs7PXZqtabh39x5pmmBMxHQ65c233uL09LSbNHHv3j3mO3PKomRnZ85oNOa1117FAs899xxgWa3XWGOIfGqv1CV37rr6bJokfOMb3+TmzRukWUaWpf73m12qerVaURQFr732Gg8ePOCZZ55huVh292qaJmSjjDt37nByesLbb7/N8y88v2UwKkWgnODBMFPVQSnHK5BRW6J2JKIp0vDfL+eIIZGNXqJPSaFKVCp7h3wOXL/06r8JZC/u962LGo98TxEtkPplP5P2biIJ154Na9kaxnE6drPu1FaUQKKf7UkJtyIB3lg27Tb1KTMmtXaDZ9Ms/b4+yk5QQLkooPaqI1vijyNhiPHujCxb5Z2Oqdsan35tHSswiDqmrKPOK+q67dSAnKSfIooSsszVJzabDSenp0RR2I1BAi7pN+7u7nYXfTQadZGjpHO7Xk1fi+zXHuS7trbdGvqumD1Yy6ZpuppkEsfkbU6g/TDw2qWuJ5MJy+WCxrc5tK0XC9eapq49e9bN1giCyKVLi4IsSzxrFtrWdqpU1nFeHEFMKdI45mBvn6qquHvnbhd9ZlmGynPWStai7jZTpZzo9c7ODvPZnD/7sz9zTl1rOD876+oyb7/9gNFoRJq6rITFsre7R1HkrJYrN+W+rlyazjuKu7u7VFXFa6+9znK15FOf/BTHJ8fUVcXB4SG379ymyF1f3rPPPuuHpIeEYcA7R0fkRcHZ6SnLxZL1xolhnByfcPv2bZR2UU1ZOIY3H+JN+4OEzDsVXWoRIJFa5XQ6ZbVaXUrJZlnWzT/tRsYp5bMl6pLg+UcFErBIsCBGXzJ0EnBIECEBizGGoii6YEqymbBtPXkvXAtjWdc1GpeDX61Xl06C0Pn7El3gyApJnPip9Jowimi9wZTXG2MwyqUzdeDYfcKstdb1Y0qzt5zsy0Z1OxtQKUUUR13BWNIcMu8NnNSXRKXu9+2EctenJMOtVbeo5TOd+pCTqDo9Pe2YbpImlly89HJKz5ywIrf1021aop9yUGpr9PspGpeaeO88/eMCRYttVXdOkyRFxlBFYYSxhuVy4WrmQUAcxU4wo2kpyoI4illv1uzt7jGd6m78mVszrh43Ho1oWv9+xrBYLjwRKCbSioP9A2f8lgtOT05IY6fTul6tCbXmE594hvOLc+7dvQfQ9RvPd3Yo8pxWGV544QWiOMK0LZOJYz0WZUmSpJ5Nu+DrX/869+7dI4pdpDGbzchGbjNN/Lqom4YHDx7w2quvcefOHT4x/QTT6ZTNZsNo/6DT+4y9Kkocx4wn424W7N2799y9qvASf06Kb7PeEEYhX//a17l16xbz+fcLZz/OEFWv/vBiYwzz+bwLFubzeZd6lahJhAPG43FH+BMhgvl8/qGOIK+iHwH2pRzlO0tw1W91kZLTZrPp6rmS0pb1Jz2a74VrYSxv3rzp9QjdeKO8yGnqhijezors9/nUdY0JTBcFrjfrS8r3fYZn6McFNU1DZbbtITLOxbFEl92xVJVjD47GI8qy6lKUckJH2YimbTg6OuqMkhBv5ALlmxxNgAinBx293k2Hl+8ixktrjY6Fvhx2PUB9qnM/1SqphDiKCeKtgIEsCtm8hEgi57C/KfUL3ENkCVGYoJSmMW2n8uGMm0+9++yHQnWkGYiIk7i7hnu7e10/rMzIk3q5Swc16CAkjV1UMJ/O/M1ecnp+xnK1dDNT45hIu5q0DoMuugTXQjAajUizlMU7C7735ve4e+cus7lrJ3p45GTR4jgmSROUZ9Tu7uySpAlt0/LUU08ynkw6daG6rryBNb7FCrA4UsiNGyyWS6f8Ajz3/PNeK/SMJEkZjTL29/e8Q+b0YI3fhLoygXdKR9nIGde25ZOf+qSr2ZYVZVV2k00ed0jzv1LbkVYy7agfLYkil9a6SzuKsyy9ksCl+/+jAPmOwruQc9UXghHDKF0FQooUJruwYqW22a/zXntjaa0lDDSr9RI3LmsrP+TSEm2XNnRfzrHy+mG2zNOThdEnr7hFp1itlrRNS5K61GRRFF1KQ7wvOeEuzRVfaviPooiiLDqjs16vWS6X3L17t6eUY8hG2SVhAjcVYhvtRnrr/UhqVxpm4zhEKTph+L7YQN9QyuQH6ZkTElGfBdtnzfbz9PKe/bT0gG0NxzVij1ivNiittxqTpsUqi0J3/W7L1ZKduRtkK06KzByUm1No/mVZEAQhTVP70UqFE+dvW27evAUKzs/OHTloNHZscGXZ3dmlqivapmF/b4833nidOI558qkn2d3bxRpDmmacn50znU544oknOT4+Ji9yl+XQkOcb3vjuG6xXK1566SWMMbz66qtMJhOK0jltjvUKsd881suVi5yLnP39fQAeHj1ksVy4IcWl0xh1KdWgaxuRdR7HEWV5eeZmkqQETU2eF0wmE+qqIk5imqFmCdCtF7nXRXigL5F5NesmRkPqdJJlEtGVj4qhhG0JTAKdPrlHokUxpMIREYMocztFH1b2PXk+8L6182thLKuywgLj8dQr1bvidBxFnJ6dcnR03NUYp5Mp09m0y8c7A2c59lqao/GoSxEtF0vXb+lrTPP5jLKsWC4XlGXplVnmFEXeacyOx2MePnzoozrNbDalLErWmzVN3fiT73rjdnd3vapL2Q1E7evJincXhqEfMbYtIvc9oM5w2a36jjHm0nR62LazyE0hxr3PujXGdBPAxeGQG0cWl0SrYrAHBR8ATVlWRIkbG1WWBVma0lrr+iM9gzrLRtRVRdNUGNMQRyHL1cI7bEknUXhxccF8PqcsS/J8Q11XhFEEWOI0Jc9zstGYumloTEte5GSpYyuuN2tOqwpQxEnsSEJAlo1YrddOfELBOl97LVaX3XjyqScpi8JHbAXj8XjrGAUB+3t7PPfss4RRxHde+Q5t23JxftG1WJ2fn7lzEDsRdcnUjEYj4iTuWOEuY9GgtOb05JT9/T2M3UqnhdqJ81sMaeqH/OIGRBdFTu1rSmEUemfv303l5aMEmQpkjOlKMcJ0F+MIW6MqmSshsojxFEdZotOPgkPczy7KoGq4PKVKvrc8tlgsun15tVp1ij59zobsm486R9fCWN66fYtQB15pxLEFN5sNm9yN2BIlC2MMt27eIoojRpkjwZRVyXg0Zn9/j83atWUorRiPx+zu7LLJN0RhxGK5ZLlccOPwkLt37/Dyyy9zfn7OaDTqmoC11rSNW3xhEHbjwmRKvLXbArJSziju7e1xcbFAe3KDm1O4ncYNdG0l2K1ObH8ihECMnlCc+zdH27ToYNu+cLU39L1o4P2oUuqXUvCWm2qoGTmpwiSJaExLVVfEkWshkohdiBdBEEAUY8rGTykp2ZnPyfMNWru6NcpH+lXp5N2ExGMMaHfdW2swbUschrRJ2olLTCYTmqb2rNa4a32y1s0jtH4IeuPbNwLfD1r7lPtkPKFpGvb29inKgkAHaD/cPEtTUIq33nyrY1j2Bwwc3rjR1WeVUrRNg/HqWHVd09Q1VVUzGmUEXphgd2+Xum6o6qrTzJV7yYn2OI8/9U5m6Gv3Um9N09Sxzn/YC+CaQKZ6iACB3P9yv5dl2dUzJarq910ClxyPj4KRFPTVyvpDAuR8SX97vx1OyFESNPSJkOKAiJP3qH3wWhjLMAzJkpFPF1gCFTAdT3j19WPyzcYPjJ1y585tN26rbfnem98D3AmUgvgoG7Gzu8tms+bs7MylTj2ZQuYLnl+cE8cJ9+7d6/RoZc5jp8majVitV6Rp2jFIA+0YjLWX8gLf42MsiY9em7bp6otixPoi1k29HUMW6AATXBl02vhxNmwjP/GSkjTp0q+wFZZ3n+UawmGrEiQeZ98YaqUwjR/d5aXNUFsN3McZKtDUbcNmtWY+n2/JKdaiFCyWC7I0ZbVaEWiN0qFP7ySUZY3WEdYq1usNk8mUuqm7azqbzdx0+c2GyWRCVZW4nk6cwo5WtLUjhq1WSyaTKWmacXp22rUPXFy4CHA0HrmWpDx30URVAdaxUANX39zZ2aEo3XzNOImdAHoYUBYlTdvStg1R5KUiraGqa8mgduWGuqq8MtblsUdplmH82hOy2ng8IjNZxyEQLz0Ko20Lk2cNKxQG11LjNnyDRRMM2Q1gayylRinMTtn8d3Z2Ogfk6vCED1Pf5PvhqijC1b0UtuRJ+c59lR7hb8hAaqDbO/s8EQlG5Pn9AOfdcC2MZZaNUNYdbJ5vOtLMU088SZ7nvr8wBGOxKKIg4sbhDRbLBdPpmLOzC0bZqEs5Sf/kYnGGMa3TrzQtxlhOTo4Jw8AzzKauh62syQs3QLWuG5cWwqnqCCzOwFnrJqR007axRHGE0i49JU3E4vnBdnKJDnS3kRRl0dUTpXbZGi8X5ltZ+vRoeV2gg65BfusN2UtqPELykYgBvPYsvjLnjalVMr9zsJaL5cIZCdOwXK98Or0ljlNCL4SRpq4OqVG01jGci7xgf3+foii8QdmyQ/Mi79qARqORu2GrEqUCZrMJ69Wa1XLFeDJmuTwnyyaufLB0vYqjbNSN6wrDgPv37vPOO+8wm80oS6detbO7y/n5OZlP7Y5GI9drNh5hxy6abY3rr51OJ06BKPTeuWkJg9iXKkSf2RL6tJ5W2qegW5I4RSe6cwJE3aqsSicsoNzkHmu2bF9RvJKpQE79p8HYnu5yZbp7Z8B2MpE4wpJGlcHKYgT6TM+qqrq6XT8l+WFF1/GgtuPl5Pv1e93lb/3XyKAJYQf3W+uCICBJkkv1Ttkn5bXvN6LrWlR+67qibko2+ZpslKEDzcXinLppSEcZN27dJM835EXheiGtJfXDjKVGdHR01M2zdELhVVc7dEZrqy1bVTVt03omXkVrJAqj85YtbgNwrSLbaQ5V5SI78Wz6F83J3mmqqqRpaqIo7B67enHFG5I0gu0pEvXl5+R3qVe6eYB0tVGB6HrKwrjqXfYNpyygq+/xOGM+m3N+ds7e7h5xHLPZbHwjuBMFsNjusdVmwXq5IkCTxgkXZ+ckUUxT1WR+LNx6vUYr10IitZPpdMp0Z+4H8eZUVUWWpV2jedu2NG3Dzs6OM9xNzXK58IxSTZ7nJEnC+fk5Yz+HcrN2xzSZTrsWqGyUgXXRpohc6EBjxEMP3YgtNy7LbUBhEHZzLNvWEHgNWambyTSfMNw+T4agK+XSwaHv+5MaqKRzvYfm1r5SKLb3Txi59/so9QD+u0Bqav37s18qEvKh7BH9NrSPQlR5FVeNYX8vlQhRomzpoUyShM1m0ymiCaknz90915cG7da0DzSuvdzdyckxzpjBar0my1KatuXo4VFX4M6ylNffeI293X3SLGUymTDKUsCALajjkMXinGbkWj7Oz0+dbqdWvh3DeVtpmpLnG5yItMJ4uToXkXpB3Wab83aLz4XnVVl1mqAifO7WpvRdRtRV3c2Fc5Mc9PelCUSDMwhCZIiu5OP7RrVvKPtqFZIu7l/sfipBbpyyLIFtWqOLUNkKucvxPe4oy5zRKGGxOAMU2cgZJa01s/mOE54AFosz3yzuIjhjDfOdOefnZxhjKaqSNI1p2oai8I34J0fcvn2b9WZDEmc8PHvIzs6OI4VJejPNKIqK+WzeZS1KL2SutRtKvVhcMJvP0Vqx2eSMx2MKX9tcr9ZdDaZL6fve4rIqt6QwYzBS70qk/u4cvaap0T4LExi3eSgg8Gzeuq69KlXTRT9S43deeu10dZXqDKm1lgCfHfEblA6Cbi1qpVChpl4PbFiAPM+ZTqddRClOsKRl5RrDtn9Q/r+rFX9IId9BSgH9vezqc4BLnJA+b6NPguwHCOB72rOM3JcxJG0rLTvXXpTAMbkqqrJiXebUtZsecuPGAccPj3nze98lil2z7nqzZrVeuc3BNp1hcvMvK07z086zDQMXETR17STMWoMxVW9heeNh6Oj9Mr5K64CwF5bLjD9rbTftXQyppFK1dtJdMn6pE5e2utvwWttekqirq5YwDNA9xmrfGPZH6YgBlbqlvMfVgnafZCEG8VKDbhBeev1A8HGj36bTuc8qGE5PTsiyEU3jhjNHccR6vUJpIf0UfiBy68cZJR1pJc9LysoRf9y6CLi4WFDXFWdnF+zt7pLnOWEUMspGLBYLP+XjrOupPDs74/DwkLzISeKEk5NjtHY1SYVid3eHo6OHxL5fLApDgkB3/cZBoCnK0keXMUVedOvX+qHN4NZAkiVeiDv1a9cNX8cztUPPvAwD3wbl06nKs3WxXrou0DRtizWecOTr40maYlqDNU7xKo5dq4hL17ryiLRzPe7oK/DAtl1M7ue+1J38vY+r9b4PE97rO11Nyyq1FR8ALhlKeR9JqY7HY9q27aLLNE27OjBsZxrLOZcA491wLYzlcrlCa+UGybbOwjd1zdnZOcZaZj7/7KIr14C/WCzQgfMK8twp7tdNS54XSPNiR3JBUVWN9zp8qwWaqmx8vdGgjAyWjvxkkaD7WdfNJQPUqQOZbYpVmH9dvUBtKc1CCJLHZFGI4e3rFcp3vBpZaqU7coYMoZbny3vJe0h02k+zXk3/Wms76v6gywln5+ekSeoJNJoodoYjSVzdsKorRtmIPF97qSxXfwyjiLouKcuC6XTWkVmUdlM6LhYXBN4QuykIFVVVUtWVI8BEEdko9aorkGUJi8U5WluWqwvyzYb08AZt2xDHEW1rCCM3jcYxZxtM0FCWOTu7u2Sj1LNmDeNxxmaTExD6VK5Ll0rrRtvUGAVt4/qAm7ohDv0Io7ohSzOMaanrhrIoaI0j5tRNjVZ9cQ3blRtoW6I4Rgvb29+v0g8sBltYZRbfl9kMDhvQsZSFGS0Rlty30mjvhjC47ftqO9mHDZf2TbYGv1+2En6GRNcyb7OfTQE65npRuJq+tIvIOdpsNsB2JFdf9EbsxXvhWhhLV2dzouKO3NN2tTeJ7CRMrsoSUN9nrIqiuFTr66dR+2ODukkl2snlydT2TiUIRRAGl4xaGAYoFW5HvgQhta29QPl2RExAgLUNSolRqpFJFkLaEfSL1JJqkwiwnyK9Cpko7/o4t6lWuZEkKu23lkiEaoxBs/VY+6nhxx078znGk1Gaxt2YdWHQuqWtW+bzHS4uzp1yCgqlNUmivcIPKB1SFCWbjRMylzUUR7GvMbmat4v6QszKrcHzi7NugPJ8PvcT6xNW6zWmbZlOp5xfnBMEmrp2sl5ZljrVKQUHB3t897tvMBqN2KzXtK3pSEVlUbihzyiSJHY1y7phPHIyeKt1zXqzJvOtCFEc0raVF1bXFEUOSqEDp1oURhFt06A1xLHjC9RVhbGWQGuwFoVLJWqtu4HTpb83t97/Nl1mjMHqoX1JIPvCVZanGAdJxwq7sy9M8GEl9lw9bqnJdiWFnnxdv/0NtnJ30jYjr+1n3SRokbFnsu76JSg579e+Zrm7u+uJBq73DHCMz36+2s+5VBKkiei51u5GlVSo2jLtur6aqnbecVl0EkeSbm2ahrwsupOeb3KquurEdmVcmBSLW68FKu8hfxd1ISkgS9hflqVLMdcNm9WG1jRYa9hsclcD0oo0SXsjtxSjcYZWypEzvH5nFLr+tDAInBauTxsHQYBpzXbKhc+5i9ByX/3DAkkYoXVAkrjn9Xu5HmfEcczLL/8pX/rSl/jt3/4XfPvb3+K177zKfL7Dxz72cV588UVefPFFfvEX/wNu3rwJyt2oeeGExJfLJa1yg50j34O4WCyYzqauv9ZYUk/ACYKI2dS1d2gVdpveZlMQxQlN3XDzxi0ePHgHazXGwP7+DaqqYr1Zu5mZ1mLqinfeecerk8xYLhcoLXNYXT2rKEuiMCRJUjcDM4Cyytnd2eHsvGY2m7lRWcZ0MnXr9ZpRlhF44oNsZcY6gXXtNy3HDA8JleqYs1EYMZkklIVb636KLDrQTq0nTroNz1onVCD19gF0rOl+z6Tc0y7Fn3dSd50DrLfD4f8i8W4G+d/GSPcDGQk0hNXbZ672s2YS7EgQ0LbtpZT1doLSVjNW2golQyd7oRhQeUw4JDLe671wLYyltZJGbCjK0m0ucGlMUlM3hFGIstsTUXlj4zx9N/WjP0eybdoeLd3dtvJc8dx0GDIOx13hd2dnp3u+I/9sL6qIpl8dGt2v+8mxBYFrLxHVk7YjEm1f10+/Xh0+2k8NyEix/iQTea78Ll6RHJccg0TmcpxBoH3azP8DqmZIw56cnPLxT3ycZ555hl/6pV/yIgOKMHBtIzvzXc7Pz0hTN0EkjF2LSFVXjq04GvloKuL111/jq1/9Gl/+8pd5+eWX+cM//EOOj0+4d+8en//8T/Lii5/iR3/0R/m5n/t52rZlOnaTQYIo5GJxgUJxsVgw35mzuLjAAsfHx4zHYwLtxNtdNqXqlEy0DimrhiRO0CogzTLOz88YZRMslrJsaBtL60sCJ6fnJD6N1bSuvUopRVm5NRuEYSd0oDyDtT/U3LTOg2/93E9rLHHsarlVjWPTegPsapqWJImp6opGJvAoRdPUWFydc8C2z7XfLlb6Qd+w7TkUA3lVsrLfrnb1fa9O6OjjUbXO93vd+z2//7p+dkGMVRzHLJfLLr3qyG5bA9dnwTrS5Fa72Vp76XX97JpE27LPdoS03vO+8Y1v8C//5b/kD//wD/nd3/1dnn/+eX78x3+cX/u1X3vX73QtjKUbCuuiuMlkgvXKKdZY8sIp6Cdpglaaqq7RYUDr60WuBuNYem7qh+v/ArpU2FYyzhnOgK3GqqvPuP7GLTsVhOreD+HBLca6qi8ZLxfYqs5YudRqg1KXh0BLakE8KFkI/d6ofqG6bxgVroVBPEkpakvjcp8h1meTwfZGcdToy2Sg/s/HGWmWuqHMPlvQNI1LS5qWbJyxKdakIxcZrvMlk3BOWblNrGkbZ1yDgChsuHnrFn/tiSf46Z/+ab+exZlxaXlRDRFHKQxDfu///T2+8tWv8Ccv/ynf+sY3+drXvs6NGzd4+mNP8wu/8PN8+lMv8rnP/Ri3bt/2WZCa2XTKerMhCAJOz06ZTqbdOl0sFgRB6CdPbKhrV6fURvvxYBalQpQyzKZu9JjFolVAa9uutmOt7dKpgJvJGYa0rSUMVDcUXWlL5VVSwiDo6ppBFHU9nK0xRGGIwrWnyD7rzmHLgK2TLE7zeDwG8FrCeRdN9dnvfebn1Qiz338I3x/NdU7zlcevor8X/Xmi2KuGUt5D9r1+i0ffsMl3E8MY+HYkmaQi30WmP0mU2f+biKQXRcGv//qv8/u///v8wR/8Afv7+3z+85/npZde4nOf+xwvvvgizz//PH/7b//tLlp/P4KPug4F4W/92TetU1tIusnreINZVZWr91W194JdDj/PCy8u4CZ5JGkKFs7OzjpZOHAzByWdKxqxwoRVikveh1ZuI3EesxuWrHCGbpNvLnlG/QWlFTS+P9MxXOsuTSfi6oEOqXzNSeqt/UXSN2haa1c78yPH2qbtjst9p200KsfUl76TWqhcfPn/viKQew/Xhwrw9//eP3is82CrzYXdeIWd1WpJnhfs7e2x3qwdWzNJUYCxrp0ijCLyzaarGRVlyXzmiGh13frrq9E66JrGxckRR6YoCld+aFrqtiHLMje4vCqZjJ2RXW/WPgUfMJ/NaduGMAh55ZVX+Of//J/x8st/yje+9U2++pWvkmUZTz/9NM88+wy/8Au/wI/92I/x/PPP+zXgiDqLxQW5n0EpxlvapsqqIo7iLqPSNI1PargIMAgCUArrSWjWZ2qEBazYRhpl6Wq3TdtQ182lwQaBdvq3XgESrR0D+Rf//b/2WK9BgK9//et2my1w+0zfuZIh0F1pxd/rUteUsV2yF8k/uBx9yt4jbRN9w9lvQ+k7dP166rsRcgR9Yk7/PUVyTtKdQroRwy8RYd9oSn9pn78in2Gt5Utf+hK/8zu/wz/+x/+Y1WrFZz/7Wf7W3/pbfO5zn+vUjvozfWUflGhdBFzkuxpjeOqpp951HV4LYzlgwIABAwZcZww0yAEDBgwYMOARGIzlgAEDBgwY8AgMxnLAgAEDBgx4BAZjOWDAgAEDBjwCg7EcMGDAgAEDHoHBWA4YMGDAgAGPwGAsBwwYMGDAgEdgMJYDBgwYMGDAIzAYywEDBgwYMOARGIzlgAEDBgwY8AgMxnLAgAEDBgx4BAZjOWDAgAEDBjwCg7EcMGDAgAEDHoHBWA4YMGDAgAGPwGAsBwwYMGDAgEdgMJYDBgwYMGDAIzAYywEDBgwYMOARGIzlgAEDBgwY8AgMxnLAgAEDBgx4BAZjOWDAgAEDBjwCg7EcMGDAgAEDHoHBWA4YMGDAgAGPwGAsBwwYMGDAgEdgMJYDBgwYMGDAIzAYywEDBgwYMOARGIzlgAEDBgwY8AgMxnLAgAEDBgx4BAZjOWDAgAEDBjwCg7EcMGDAgAEDHoHBWHoopfaUUv9EKbVWSr2ulPoPH/H8WCn1TaXU93qPPauU+g2l1EOl1KlS6v9RSj3X+/uL/rFjpZT9i/w+Az78+CDWpH88UEr9faXUW0qppVLqj5VSO3+xRz/go4AfxL74YcFgLLf4X4AKuAn8R8D/ppT61Ps8/78Bjq48tgP8U+A5/z5/APxG7+818H8B/+kHdMwDPtr4INYkwN8FXgI+D8yA/xgoPthDHfARxQ9iX/xQQFk7BDhKqTFwBrxorf22f+z/AN601v637/L8p4F/Bvwd4H+31t57j/fdA06AA2vtSe/xTwB/Zq1VH/iXGfCRwAe1JpVSu8B3gc9Ya1/5QR3/gA8/ftD74nXHEFk6PAu0siA8vgq8lwf1D4H/Dsgf8eV6ghAAACAASURBVL4/Azz4MC2IAdcGH9Sa/DTQAH9TKfVAKfVtpdR/+YEf7YCPIoZ9sYfBWDpMgIsrj10A06tPVEr9DSC01v6T93tDpdQ9XArj73xQBzngscIHtSbvAXPcxvc08DeB/0kp9e99sIc74COIYV/sIfxhH8A1wQpXy+ljBiz7D/i0xP8M/JX3ezOl1CHwm8D/aq39Rx/gcQ54fPBBrUnx8v+etTYHvqaU+j/983/rgzvcAR9BDPtiD4OxdPg2ECqlnrHW/pl/7DPAy1ee9wzwFPCvlVIAMTBXSj0AftJa+5qvEf0m8E+ttf/gB3L0Az6K+EDWJPA1/7yBnDDg3xTDvtjDQPDx8N62Bf4z4EdxheqXrLUv954TAge9l70E/Arwl4CHwBj4beAPrLX/1bt8hgIS4GO4BZcB1lpb/kV8pwEfbnwQa9Ja2yql/hXwDeC/xq293wV+yVr7L34gX2TAhxY/iH3xw4KhZrnFf4EzXkfAPwL+c2vty0qpn1ZKrQCstY219oH8A04B439vgb8B/DjwnyilVr1/T/jPeBKXFpOFlgPf+sF9xQEfMnwQaxLgl3Br7wT4v4H/YTCUA/6c+EHsix8KDJHlgAEDBgwY8AgMkeWAAQMGDBjwCAzGcsCAAQMGDHgEBmM5YMCAAQMGPAKDsRwwYMCAAQMegWvRZ/nHX/uyjaKIMHSHY0xLmkbEcUxVVTRNi1KK6WxCUzeUVUUYRiRJRlkUFEWBUooojkmShCiKePvNt4iiiJuHh1RVhdYBrWlpMUShJgxDrLEEKqGuS6IoRClFa2vKssRai7WWMI6w1hJFEZt1gdIBWZYCUBY5cRJRFA1hGBLHEdZCWayx1mKMoShqoigmjEICrVFKsV5v2GzWTCYTlNKsVktGozFplrJZb1gsFqRpQhi6zz47PWU8maCUomkaAKy1zOczlssVdVVRVpU73jBEAdZC27aEUYi1lrIswUIQaKyFuqnRWtM2LXEc8TM/9Zcfa53a//Hv/vf2O698B61DPvWpT3Lr9m0mkwlf/cpXmU4n3Lt3n8PDA9IswxhDoAPSNKVpatI0I4pCtA6o65pXX32VqqrIRhl7u3sApGlCkqQopciLnMViQRRGVFVFnMTcunmLKIoYj8ecnJxQljnrdc69e/cYjUaUVcnezh7/+vf+FXEY8ZM/+XkMlouLC5IkxlpDVVVcXCzZ398nSRIAAq0pqxJQVFVFFEUopVguz2lbS5aN2NnZYblcssk3JHFC0zbMZ3OapuHi4oI0TVmuluSbnNu3b5MXOQrFer0mjuNurYdhiDGGOI4Zj8c8ePA2SmnG4zHWWjabDXmekyQJYRjQGsN6tWY6HXN+fsZP/PgXHus1CPCVr/6x1YHGGndOoyjCYonjhLZ1+6BpW4w1GGPAgg4CwjDorm3TNARBQFmWJLFbB3meM55MSOKY45NjnnziSdbrNcYaAPd51tI0DZPJhKapyTc5Bwd7KK3J8wKtNMfHx9y7f5ezs3PG4zGr1Yqmabh//z7LxZL1Zk1TNxhrKIrc7z0FYRihlCYKI4wxtMYQxxF5XtC2DVpptNYUZeHsgAUUVFWN1oo4TqiqEmssSmuapiaK3P2jUERxRNu0GGOwWIIgBGup6sq9RimUVrRNi7UWrTX41aaUuze00jRNw1/9K3/tXdfhtTCW4/GIsqyo65ogcIYsy7LuZqzrBmMMp8enzHZmxFHE9773FtPpjPnODlEUEcUxURRRFiV1XTObu5v99OKU8XiEjgNsA4GJyNc5YWhI0wRjLaApywqwKAVhGLiTqxTGWiygtXKbQs9YgQZr0UoRRe5UbtZryjJnNM6oihqlLNYvyKIoMMZQliVhGFEUJVpr4jgmzRJWyxWVN3ry7/zsDGMtVVkRRmF3IzR1zWa9IUkSijxHKwVaM5mMKfKCTZ67DSwKieOYfJODgqZsSJKEQAfd9zDG/OAv+jXDz/7MX+ZnfuZnOTk55d7de9R1jQ40L730Eq+//jovv/wyX/jCFxhP3HlzhrJxNyXQNC3gNrEXPvkCRV5wenbKt771LebzOTdu3mBnZ5e6rpnP5uzs7FBXNda6GzmOY4qyYLVaoRRYLJPJhOl0ysOHR3z963/CaDzip77wUyzOLzg+PuZ7b73JrVu3mEzG1HWJ1orDw0PSJEUHmvPzc9I0JU1GKK3IshGmbTm/OCcMY8JQsVwuOTs7I4oi7ty5i1aK1hi+9703MMaSpikXFxfcuXOH4+NjNpsNdV1TVRX37t1jsVhQVqXbpKylrmuKIqesSuI4wRjjPi8ICMKQMAwpy5JN3hLogDAKWC6X6CD4IV796wNrW+I4ZbVaE8fOsARhQNu2VFXln2Pd31rDeDLpnleVlTMUOugcfK01Fst8Z+4c9c2a3Z1dTk9PObxxgzzf0DQteZUThiFt2xIEmiQZkyQJZ2dn3Lx5k7OiYDweg4Lz83OKIvd7iCKOYpaLJYvFAmMNcRRjjMHELdY7UWEQgNJsNjlt25LEsQ9KDApFEIa0bYNCMR6NfYCjMcZQ1zVN4+6Vpm3Q1jn81lrnFFqc0VegtAZraJsGFIRhRF1VtMYQKI3SmrZpsG2LDjRt22KNxWJpTPu+yh3Xwlg2deM8KO/ZaG0oyhKUoiwrgkCjlKYGyqJiNp+yt7fL6ck5eV6wszNH1TWRNyajcUYcx6zXG7ANTdNgcUZhFMdMpxOMsdR1QxgqtA4A4z9bY42hrkqiOKKqaqI4pmlaRqMUVIi1UBQlgdZUVQtKu+dFbjMIgowwCBhPRt6zN1RlyWg8oqmdsSrLiqqqCIIApWC5cB5a0zRgLcfHJxweHjCZTt3GrRRhFJEmCcvlCmMtmzwnrGt0EDjvb5Nz0V6Q50VnvCv/Oc7pqNnd3WW5XHhjr7xR/mFe/euBumm6aKw1bbfRzGZTPvOZH+FHP/ujrJZLfuUf/gqf+9zn+PxLnycKwy6Ct9YSJzGxjmmbliRJODw4YDabEYYhURh2kZ3W2v1U7mbVWrNar5zTF0XuplcWYxSbzYb795/g5s1bhGHIcrnky1/+Mrdu3eLFF19ktVqxWq3Z3Z1zdna2jfZa2zmd/esbeoMVRZGLNsZjyrIkjmNef+M1xuMxhweHTKZTdnd2OTs7Y2dnh4cPH3bRyng8JggCzs7OnCNoLXEc0/hz2LYhm82G8WhMFEUEQUBRFuSbDXGcEPpz0bYNTeUyHFoNxhJclLher9FK0TTtNkJMUpq6JpQMWNMShiF1XRHHEavVivF4TF03BGHojJBye5vWGudLa+Iopm5ql+HwmYS6XjGbzaiqkiAIOD8/99FXQKAVi+WS6XRKWbh1grVkWUZV19R15SPFlulsyvHxCaDI0pS2bZjtTnl4fEwUx2jlvotpW9IspTgr3D6kIIpCyrJAaZd5McagdYBpDVppjHHrWevAGeBA0RrXRmyNJQgCWt9WbIzby5XSaG06h8M91wAWYyxKu4DIKpyT6B2F98K1MJbpKGO1XHljE7DZ5OR5jtbu5pZIKU0T9g/2Wa3WGNMym8+oyorT0zOCIGBcjN37pc7bsP6kpWlCHMfOEDaWqqxp25ZslGKtwbStv6ldmrQoDHXT0LaGvf0d8qKiKAqSZMRqsSDLUpLEGVCtnMeMtWw2OQqIk4DVauOiy7omiTNGo4yL8wu01sx35jTtBbYwFHlOGEVYY8jznOVySaC1T2m59GkQBN1NFMcROtDESUztjeBqucJYwxNP3ufs9Iy6aYji2HlM1nlNrV+gy9WStm0xfoFFUTRElkCaJLStIU1TwiDAWEvbNt6AuVTjZDLll3/5l4miiN/4jd/gr//1v87p6Ql10xBHEVrPCJLA3dSBpmkbxiOXglyv18zmM7cGrcs2aK1QOkShGGUZTdOSpAllWXL37j2Oj08pioI8z1mvV+R5wf379/nZn/1ZrHWb33w257d++zfZ29/nqSefYjabuYyIMSRJwne/+zpN0xDHKZPJhDiOCcOQPF+z2biRlk3TEIQBSZwQBu54qqpiuVxSVSVlWVJVFXfu3OHi4oLlcunvq5TlcslsNuuyJpt846OJiKapafymjrUkSUoYhaxXLv23u7vHarVwG51pH3GFHg9UZUkQunSqc8Rc9iKKIuowJAgCduY7nF+c+2hRU1cVN27cpCwL6romjqMuCpWyS5EXjMYjDg4OOHroxk2GYcTp2Smz6QyLpaoqbt64QRiGXCwufATr9t/T5SlRFDEZuxTterMijjO01gRBQBzHnJ6cohRsNmvyfEOgA9747neZjMecn58TBBFlWTAajVhcLMiyjCRNODo6oqrccbdt49Kt1qWe4yTu0qvr9YbAp0+10qAUdd3QNg3Kl7jatvFRJ4ClaYxP2zpD6PZC50g0dUMYhZi2QQfuvmzb916H18JYLi4usBZM27JcLplMJoDbDADnAQBJkjivJE0JxmNOT8+pm4bZbEqaZVhjmM1nnJ9dYIG6bhhnI8IgxNYKAkVRFLTWkKQBKrRgocorX3NyOeuqqgh1QjoKqWsXlcZRTF3WZElCHEbY1qANncdWlSV15YzwZuM8bBfV1WxsSVmU/mK2XeQXJ7FLvxUFRdWAhdls1tVqlVJMpxPyvKCpa+I4pm0aan+M4/GI1WpFnMSURck777ibQCvtUto+HWNMSxC4Gq011hn3pqVp3Wd6PcfHGl/84v/H/fv3GI3GBGGINpaDgwNef/01ptMZQRB0NXWtNX/1r/4V/uiP/ojxZMwnX/gkSqmuZheGPj3lHZbxeOwjSsXrr79OWbq1MB6P2d/f74zrdDqlaZyBfvPNN8kylwpbr9fs7u5hzEn3vsZYlAVjW37u536eMIxom4ZXvvMKSilGoxFJknD37n2WywvCMGa1WlJWOa3nAERRRJIkzGdzirK4VNPcrF3taXd31ztXhtdffw2tA6I4IsB56nEcE4QBtnR1oJCQIA58tA060CitGI8n7n2sM+Lz+ZyiyLeRjx7WIEAYRWAVIhZjrcW0LW3b+AgsoqorwiAkTVPyTU5d16xXK6LYZS2KoqBtjYvWipIgDJhMpyRJTFVXncHd5BuCIGQ8GXdO+enpKbPZzAUvceT2DonovDFxdeeU6XTKyckpk/GYOEmIoxgUPHz4kLpuiCINKJarDWEYkKUp+WbDarUkyzKiKODmjZscHT0EjE+XNr4W2bDZuPKSDjRYX64rnDKosQaNdk5noNFKU9WVdzKMN6Z00bXx2R+p+ypFV6Pt9j/L+67Da2EsQRFFgbs59/ZIkoQ8zwGDwi2Qu/fu+DTUytXpipLNZkOWjpjOZlSlq//Vdc10OiEIAy7OL5hkmV8EZ8zns24jgZbVck1rDCGRNyouzz+dOmNtaVmvN2Rpig41sa9PtT4ShdrVD43h6OgIcMataVwtYXGxZDKds16tnYfUGuLEbaBFUTCZTIiiiKIoyLLU5dn9+6epJxGVlSdwWKqqpqprFBAEAavVmkC71EbTNC4SimOUVi510bZUndFsHMkHMI0jqGDdudXvk3p4XPCFL3wBpRUXFxfOAYljHrzzgNl87s4flsXFBZPpFFtYgjDgs3/ps2it+c3f+k1+6gs/5VLwYdBtZkEQ+FR+Sdu2NI3l6aef8lFXw2w2Y7lcOnJOXZPnjhCRZik3btzAtoazszPefPstkjTxKVUDKOq6YrFYkCQJSikmE1cXev7Z5zHGbbSuXm6JwsQbLY1pG2fggoCqalitV11KDhRpmvLmW2+SZSMevPOgu6eyLEPrMW3ralBaKSyWNE149TuvcuvWLdI0Zb1Z07aO8IZSBMqdD4XqfrZtw9HREda686iMoa6rH+LVvz5wZJcQ6xkukjoty4pAOydEodBBwPHDY4Iw6LIYpjSMshHrzYY4jgh0wO7hLhfnF5RFQRhua5nnF+cUecHe3h5nZ2cURXEpI1KUhSsXFSVxklAURVdKCHTAarmmrluapubh8bEvP2lmszmT8cTVtitHbjTGYEzL2fk5e/v7rFcLqspl644eHvvUsPUOgXKEHADljGKow27/swhfxNJaZ/jqqkZpRaCDjrCEwpWv/DnTgab0NsJanLPZ2UgL/mVtc80jyyzLqKuK0XiE8enIOI7IMhdhjSJXOzTGsr+/i9KOkHPz5g3m811OT89cjU4pwNV5giDg8MYB1m8c+/v7HXno3G8ycZy5qCxfO2Zo61KvRVF2Fz8Kw85DcR659hfA+E0mp2kbJpOMpmnIRjFlAfmmoG1bTo5PuHP3NmVRcnp6Stu66HI8HgOKoig5O3UplTRLMa1LAQMUee5SqlFEmqaOGVw3nslrsMZS1iVBoD07zlIWJdPplKIoKMsKHTiPK0tTgv+fvTft1SS5zsSeiMg9893uXktXdXU3KYki2JLggTSkBIzhAQzYn/VfDPtXGcZgBNgWB5oZjjRDjrizm13Vtdyqu7577pmx+MOJiFsNkBx/sbuA6uCHZld13XvrzciIc57zLEGErqU5ARMMWZZCKoVxHL++h/+OrO12C6WJeLDdbDHKkchicgQMcP/+PUynMwRCII4Tv1/6fsBf//VfYzad4s3FBQIhEMUx5rO5h3qIoU0w0TjSLMVBs47gk6UZmLDzF85wefnazxQfP36MoR/Q9Z2d5dA+nh8QYSiOY1/5n5+f44MPPkDTEREnEIJgVgTQSoNxASnpZCiKAmqvIKXExeUlFvMF4pgu5TzPfcHmSGeb7RZ5loMLjjCMoLTC2ekZptMZpJJo6gZFXiAIAj//d3MhYuQCUklkWWZZiVQk6P8GseJ9WlEUgYFDG4Msy/xhb2Asm10gL3JstztkWerJLcJ26HXTIIljJGmC/X6Pm+sbMMYQ29GUtPsPYEjSxBKxIsxnc2KUKoWmbpCmmZ2JjpZbwfHwg4e4vLyCVgqLg4W9fBIo1WDoewzjgLKsAGP85TWMI3VxSoNxhrquvsKe7rqe1A1CwHABGCrwqaCj7z+OI+0X0GfSNC2U3V9SUWdoRiq8CKHU9kLUvmvs2s4ilAzaXbLjiDAI6R0HvsKQ/V3r3bgsMyLkNE391q/SpRLagTIxBunSVFojLzIYbbDdbj10laUpMaDCEFFEUhTGGPQg7cGmEAQCURgi4EQ8gKJDrKp3SNMMDABjAaTUVOUJRv+dMdhu94jjyG4kAc5DuqD7HlXfIE5imIAjy1NUZQeAIc8zjKNEWVb2omUIgpgOEyEw9APiJLZsQo1+GCDH0c8px7alobqdKQA0axzHkTYDpwuX2HK9ldhMMI4D8iKHkspCGQJN3dBBJ0JIJTGMdBA6evn7vLIsQ9/3KIoCJyenJLsJAmhLlGBgVm4jAFjqOQLkeUCXHwwe3H8AYwyub67x5Zdf4vDwEPfv30dVVURi4TRrDoLAHxqOcKO1JlhIjRiVxGQyRRTFFlaibsCRtOI4pplOGGG320IIAWElQsfHx1itlpCa0InOwmtEEonsDLLD0dEjCC6w3+/Rti0+ePgBtNG+cIrCCGVZekISHagzjJIYi44N/Nsvfos4iiGEQNs2VkJFc1etNabTKfq+Q5Kk6PseWZphHIj1Po70Mw59hyJPv9bn/64sYrorMM49OjRa6ZvjVDRN44lbUUQKAHDmz0HOiQl9cHAAIQSapsGkmBBhcuj96CXPc+x3eyL97UsIISzSQaObKI58FxqGIdarNThn4DzA9fUNoij03RvnHPfO7qHve99AtF2LJE6I9QoDPdIlxhixxwNBezYQQBgKhCxA348wakDfDQDognMNjzHwBabjYgQigAHQ9g1inrh+yctR3J8NwsCyaunXtWXhukYIwH+Tu/FOXJZd26LvB7Rti8lk4qsCpWAPJyBJEyhJbT9jDK1S6PsBjHEcHR/RX5QxwBjkeYZhGNC2BD2MY4++o8HyqGmmx5kAAwcMQxxmAGqIgEMOdDGLIEQQMIBrIt2EAZq+RpyF4BHAhKCDI+AwA8N0OoHhDHkUEzwaCIwjzYDqqgIAr4Xa7/aIk9hrJuM4RlWWSNKUmMFWu9b3PcKQILuqqpBlVCB0fQfBqWNgnGZYwzggZgnCKMR6vUaapFgtV3j0+BG22y16K1Nx7EutNLTSCESAT//8e1/Xo39n1osXLzCfz712MrAkiyiIPKGAcfYWQUCBMYAxkhm51ogxRheP1gijEC9evMA//MM/4M//4s/x+NFjv8fDMMRoNW11UyMKI8hhQJ5l/mfinGM+n2O93vjLfD6b4+r6ClprlPs9zs7u2Reffq4oIjJb0zXEArRzqDRNcXh4iN1uxHx+iLquaZ9bkpe7zKWUqOsa1zfXNHOF8XP1IAgxm80xjiP2+z2RScKQvr+m2VCW5R7uKoocSikcHh6hsu9AVVVI05TGH0GApm6g1OC72Pd9KUXFvrtItJ3JAQyBCNB2LdIk9d1a27SIk4QKk9kU+90ewj7LQARQWhEnouswm81oD83n2Gy2FhofcXV1jSgKsdlsfLGU5wXW641l0pNMpUePru2gjbaKgRFKKwRBiLZp0LSNHyXIkYhJbdd5DTuhg8rOYTWkUojCEEorNHYMoBV1iPR3bQBlLHSqkcQJXcJtCyFsI6Q0lFLExrZsYa3pXBuGwcv+5EhFntYaGoTOaKM9suMuVTfC+F3rnbgs65raeIAuHaU06qqyEg0yHAiUsuYEdCBFcYzIGHuoEfO1bom8IARR85umsVR8DS4Esa000LU9koRBG23nSwzagCQeFnYIggB5lEMrYBhHyKbF3EKlQz9AqcETdpI4Qi1HpCLCdrND1/WQkh7ybre3hA+qppRUaJoWRyfHaJsWjDMsb5c0AzKGmK5hCKlIhxZGIQCgqRsyZOh7KK2guUaSJsiyDOW+RBInVNGNkjRtQiBNU1RVhbIskaUpwSrNCMWIkk5U/wj/+J/+Cf/T//g/f4074Otf3/6jbyPPcoJ95Ijz16+hFX3Gi/kCqYW8eMj8zFobDWYhpSiOoJWm2Y6VfzR1g3v37uFv//Zv7YywR9+PvgMwxqCsSsAAneosYqD8xWXAUNck4l+tlsjzHLfLWywWC3DGkcSxPZQ4DAMYJ7OJyB4YQghorXHvHlX8DAyTydyaAzToOpIzUefZYRylh0+VkkhTurj3+72d0QPb7QZaE8rxwQcf4PrmGkM/IIxDiIAY25PJxMqjemgtsd6s/WEUJ7FnHGqjMcoRxmiU5f7refDv2AodGmblGE7iFYZk4kB7iy4JxhiyPCNCn5RomgbFpCBNtkVFgiCkM6KsYLTBdDJF27TgnCNOYgjOEVuNrxA081NKoetaMAYcHR5iGAY0TYOu7cAFx9iPYFb+xCQjg4AoxH5fWq17BzAGzkl/rpTCOIw0TzSACOj7jMOIKI6A0SCyBgMK9OsDBhRFjqZpaL8YahLCMPL6S2XHCcYYT1qTUiIQgddcaiWhFRWzSivfTWpNJCA5SijbUTL2h7vLd+SyrCGEQJLE2O12ligRIEliSDn4edwwDEjTlA4HrTGZTgANz/hy7FGlFWBIOB4HIYzV0rTtiCQl+Khua6RZiraqsVjMoYSlz3OGNE2o8h9HtG2HzWaHw8UBwiiBMVR9xwnNgPb7ElonmExylPvK/53CUKAoJliv19jvaoxSwhhiRh4dH6HvOtze3HiXntGytNznoY0BszpIomQT2zKKI8RxhNVyjaZuPHzy4OEDvHr5irrXKEZZ7slRZiCCT9t2EIJ7CFcIYU0OFLT6RjqSJinKsrTi/RQff/Sx/8w550jT1MqZyBHHCe3DKLIsv5ZINIykGGDEBJVS+lmQEAEWiwNveBEEAdIkRdd1XnuZpZl3MUnTFLvtFoYRkaYocgQBvScOcnekjVGOuL29BQBkSYpu6HB8fIooiizUWmMYenT20jw8pO6SWKmdhfUVuVzpO32f0z4bGMhWoshz2yFoXF5e0uEUBFgsFlitV16mdHl5iSAMvIPR1fUVzk7PYIzGcrmEUlTV3zu7h9evXyP+BoYFYA9xwe3oBf7zdbNfd5Ge3buHfUnvvhAcfKB55Z25CsN2u0UUxcgzmj8qqTCbz8g4II6xXC7BBUfT0nw7y3JUZYU4iUknayUfNFvsPBEwz3LP14jiCHVVI0kT74ATWtczgthJh1kUBcIoRLkvySBACDKbMdoyWoU/88ABaGC32/sz0RhD88yBxlmCc4wWSpXWzIBxaoDasfXoGWMMUkvfSdLol76PZtqPWYw2ALcY7u9Z78Rlmaapp/ZyLhAExCDUWmM2mUJKhTAILfxF1k1KaXTVQM46gmPoadY3WOunOIlJuMtoWA7BUA8DtputZY4qMO0cbIydh5I7DgC0TYuyrFDXNY6PT0hczji6oSdBbdtAKonJpECaJSjLGsKKz12FXte1t5FKigyzfIK2bdE0NTgXyPOc5o6KKiKlNQ4OFijLyruvbLdbAHTwdl2LMIqw2+08DBuEBDdcXlzaTlcDwwjGuJ8bSSlp1pXEULaiMsb4oXeSfDOz1Ibma47Qw2xlTNR5gVGOVBErcighcTbsS2gsIUygk8ofcIEIoED7WnDhiQTaaGIkC8vU48x3brv9jmRHw0DdoCWZxVGMvh+hFJlpRFGC7X4DGI22GXB0fIzT01Mi1RjA2Hf++nqJtmuRZymKovAHlBOxX15dotyXePLkCTnpWCi2bVuSHlgbsP1uj8lkgu2Wfr4ojhCIwLNlq6pC17Y4PT3Dm4s3tOfkiJVeeUH4arXyms48IULccrVEEAbY7b7pLAFipytNbjScczBFl6NgAv0wQCuFKI5xcXFhbd0EFvMFzbTD0F8ccRJDJIQuMc4Qc9pTNzc36Lseh0eHUIrIlLPZDG3bYbVaIopiVFUFwYU3uJhMJpbRSlaGURwRmWtCzYDWBAe7yygIA8/Q55yDgS7WfVkiCkPAn7kkhTOGbO0IHjEkCREGAQLbAfeYlAAAIABJREFUZNC7MoxENBMCGCU1RBr0e5xREcvA/JyfMXuRWvMPuggJ4VCakDvXkBgYKHlHCPpd6524LIMggLDuKc6ibTIpaO7Y0IdeqRphSOwpYrJGSNMUYAalnb8kSYIojohiLSXaqkWeE5Q0DqRNHIYBR0dHiIPQVtZTcM4wSppLKanAwRAFIQ7mcxzO5/RDaoN2aD2T0UHbYRSirhrEUYT9vgTSxEKnBlIqGAYsTo4wjiN6y2Y0hipGBtpYsPRnZRR2u72Vr0y8sDhJYoyjtF+DKrV9Q4zecRgRR7Gfpzk7PYJg6PGOI8EkzoFDM9o4buYpR/n/49N+N5fvoLRBFIdezE0a2sG6oDAv6zHGQITWdcbP2On3tNFer0WsQI5xJE0rWSlySDbCkdjeZu65zkFr7Z2XAhFAcIHNZu0ZgrPZDEpJLGaHmE3pZ/M+n5Y2z8BwcnRMGltmsFwuIaUdMwiG29sGXdfhwcMHePHiBbIsA2OAELSf7t+7j+fPnwOggtbN0Jk9c5wRgTH0tQ0Mbm5uPDkqy2yXbC9VYwwm0wkCEWC325ENY9V6zeA3i5aTfTj/Zi4EijzHvtwjnRR+7su5AANwc3vjz8NhGADGEMcxpEWrdrsdiqLAzc0NwiC0F6VC0zTQSmG33QEMOD09xdXVNeIoxsHhgffyvb1dIrX7uus61BWdt9vd1sKqNF+dFBOUVUnjoyAkL1alvRnApCjofRqJnzJ2LRUH0jLyDeA2l4NaXZestfF7yBWzUtG5xRmHZs6xiGMcyNXMecA6D10iAdsu0lnkgcFYoiSMefft7tKUoKgsy5BlKfZ7cvUQgUDdtwijAAELEccR4jzCfrcHD4F2oIs1y+9IEdr6/A3tQBvGzoeCMMDiYOE/7LIs6QADQ5oK1LsKRZEjEgGGQaK3f05bu7zQQpZdV9sLhx5E13aYTAtUZU0/XxJjvSJHoaLIUbY1Wgtp7MqNJ9koSX63kSbReRAE1nRBo5UtjN0cs9mMDs6xtIQm2hDzBZlfwxmmW5o3LNFDa+qAmqbxBA0RCGszRgbrJIOIv6nqAey2Ow8bUpEyWls2orYv5gsv23CMVJJmBNY1xWkXidRltIYRNCshs2jqUIUI7Mwz8p3/2+QaYjkaNE1Ns2mt0TQN4kWMw8MjbDZr5PnUWs0ZaBi0VQkeCGRphuVqSc4ocYyuaa2m2Hh04fDwCGHoZugEpeZZ8RWGdRAGCFmIN29ee2KR0+iSXpQOKWHhQjdfg6ZLP01T6/5De00rjcPDQ6xWSytLSHF2dopnz760cinlDUje96W0AjcCXd9bVydpZXQpikmBuq4pmEGQFtMYgw8++ABXl1foe1tchSEMgCxLcXV5RYz5tsMwjEjiBDfXN3aUFMNoMrhPsxTjMOL4+AhxFBPsLpU1rki8KUxd16h7Ui1EUYwwCJBm9LydtMl1lKMk2YcLrSirEkmSkg3eZIrlagXOlCXVWBjVOjnJUX5ltghm/LyRcwHpgiCUQhAFnpiklLKXtPSaSsbI71gNNN93BR5TzMuw3OhEv+sOPuM4WMErmUFLSQNdJRWyLEe535LnpNZo6hbjKDFfpORv2XRoxwZ5ntmHS8YFaRjbmU4EgIHDWHwcvtIvitwa9ZIlnnthg4CD88i38kTVJ+P1PM9JLSJHdF2NvMhRVy3kqBAnCXbbCmFIJIaqKgm2CwI0uxJRRB1j07Q4OFxgebvEOEp88MFD3FzfevPgvu8RhSQx2G133lXIHdAE5TYQARGAjDaemEI+itrrVSlhokSWZbbCA1RLB5gCdSDOLeR9XvMF6SLdbCjNUnDGEUYhFhEVaO5CJHiWOslA0BzZdZeUNEM2co7oo+2lJ3tpbRNpH7U2lUHKEUqF/mIUgQDnAmlK/rHz+Ry73Q4AqCu1ZvpFUXgWbtu2SNMUs9kMARcAIy9hV307dvX19TU4h7Xbo45vubr1tmWMs7cs+UiLOQwDsiyFUtpfgHEcgTnXFPs/qSXMYICBJBAMNJd1zkT0LkWoqsqT77I0A+fcs2Xf90UzcmbRLgmAmKNX11dIk9QXNNpo/2z2+z2iOEKWZdjtdmCcYbUqcXRIKoEkJjMLPlA3tlgscHV1hSwl1QDzXRwA40YB5EU7SioIaQxERd/p6anX/VZ1haquaN5Zlb5gNMYgtXuGujoODrLzMzDYbDcQ1uSdrBBjUhcw51dNF2gQhDSi08ZzAKTlf5BkhfyvyRTdFZ60h+VIyJkx5P3Kw+Ar/Az39ZRWtiOV+AMo7LtxWUqpEFlrpHEc7wT4UhIsKgIi1owGYAJpVtDsZuisXoheuLYpYZIE82JiHz4HY3YuxMh4NxABoGg+RVqv0RsVE0NU3bGtlAKlxWhytIjo0mysdq1tRxSTAJGIwQTDOEjLJOs9YYlzQYN0S+Do+wFGa7x5feHtl/a7vaWEJygm5N/Z1LU3V3ApA13fkSuPtcKLEGGwaS1OqhAIF3OmvfiXc+4Zvu6loEosRNdV31yWgIdljDEesibKO4ccRyKDtS1pBbMMQtx5D0cRCfQdjBpaqr2HjUBxVm5GTW4k5IQyKoJjRzn6y2UcBkIHDDmy0KyPGIgkKmc23aOjfRwI5FGOrusg5YBJPsebN+fQRiO1aE1V1Xjw4AFev37t57BZmqG3c/wopkIg5ERsGxTtl9V6RYefNogTMkFPsxRt05KdmaHIJBKL1wT9gSG0xLVh6FHVlWdortcbHB4eom1bHB4eevu0UX7j4APAQvU0v6Y0DND5ZBOStCEbO2f6r20BYzR5qc6mM8TWFKOsSswXcwRBgOVySchdmmGUI/IiR9d3VnfIvPmAm683TYPAetIC8C5OwzCgrmp0fQ+tlU8qOThYYBh6iuAy2mrKue/kAHjZSCACku8F3Lr8hNYfWHuHKi4sSujSRPwckkNYT2XqYskoRmuDOIoglYLL97ozGTBeC804h3nLpceQj7stQLlND/rd6524LKfTiYVEafBLlbY1ttUaSZYijiJq+ZPEe5yGgbCsvQH7jox7Q/shkuyS+eGw1hJd00EwbrMwqZJ3GjPNgW4kUkVe5JBqwK7cIcsT0pJBAYwq5oBx7MsSWmtcvLlEYTdhxBnSPEVdEaxaVQRXJEniN2NTN746B4gYst0Q42y73XpIdBxHTKdT7Ha7u4u26/2Q3QnTAZqLaQu9Gk5dpqPnu+7YvQhu7sYYJbqE4d0F+j6v1EJeMN4ICr3NygPIBDwKQ9/ZGRiEYXT32doWz2XnjXKEUMKnGBwdHWGU9FzKsvRJOm5W6Z6tk22cnBCTtSxL79sZBqH9eYDZbAbAoCwrO4+6wmw2xXx+AAaGhw8/gDEaL169wGJBhtubzcZ715KEqcbR0QnGYbAIy4DcWpU504EwDKBFgEENPubNCc/n8zkFHMSJrfbJkL7tOvR9B2Oog+3aFnleYL/f2wtz7ZmLWtE7mMTfzCwBIC9yNHVj/VoHcMYQxZFHDty+oairEmdnZ16u0/cDqqrCZrsF58ThUEphu91iv99jPl9guVri+PgY2+0WsdWEz2ZkhO9MMgQXaLuWEpzSzL8XwzD4qEDA+OcbhiEuLy/BwNC0DbFibQiB1prImZZh7fImszwjVCug2C1ttDVeoHEAYwyCM8uQhefmMGuBB8UQCOG1pACzemMgDCM0bUPFq7kj/GhNXaz/OezXVlpZIqb+g+OAd+Ky7PvBMlhTW1UrGmBbwfR+s8WgDOJJZE2F3eHOwe3l4FhXzsqJ2YutbSuSTAQMRTYFQA+YBsZUUSkYhEEIHlkShw0IzbIMSRyBcSDQAYZxoI4jiijgVGtkWQYNg9V2jcP5AlLpOwF2WVvmWE0HSZrQ7FM5s+AeQgSYzqYwWkOIHJUlK0VRBGl9PAnSrbwhdxiRsQF5MFAlKEfqap1JtyP3OEjDVW+O5OM2h4M03vfVNA29LIx7pxTBuZ8xOmTBMe6UVmAx88JoZt9o99ycmf0o7y5QYwwg4HWZLpAbgCX09GCM5oCr9QqBIMr+ZDIhskzXEYktosOz7WpEIXV7jx8/IVq8MfiPP/r3+MEPfgCA4cNHH4IxjpEpHB0d2e6T5o9BEOL8/JxmSiAobLVekS2ahae0Uj4DMLTxTpyRVeJoPXCHcfhK9FwgAmimrf9xjlFKa2pgDRwYILidl2+3fs75zSI4nXGOuq7JRGWUXiEQhiHKskSe55ByxNnZGa4uryjFSEo0DSXbvHnzxs4XGzuzhjeuj6IIL56/oLzgpkFgnb3cmQTYiLrAuvMIAW19e4dhtCYWETiXMGYA54EnK2o7Q82L3PIwCOliFtVzxBytyeglsO+Wi92iCbxxjSFpNYl344tK4y45RppiF8TuLkylFbjmb1161DCRztf47/92o8ENx6hGn5Tz+9Y7cVlOpoW3c6r2JeDEthY+DOMIeZKSBMR2USJkYNxAGgXDgSzMEYXOFYdZKHX07hFBCKiR+RmR6yx5QMNiV2UEjH6OUUpEaQipJVRPess4jpDEKbZlicxWgE40yxiwqyo6xFpieRWTAlopHB8f+wOjLCuAAdk0Q93UYJxjmk1QTAq8fPkKnLn4LYKH+47ikcKIWIjDMGC1WnkdlgtAdV3J27mgrrJzqRHO6NtlyLnN+03qCBVQzjWF28pTSWXjtyIfaOv+6SpwihSiw8RlQ3LOvaDcWIo84wxqVPSCK42u7XwsVRRGdnZMjD0HzxtjfPyVk504TeYwDiiKDNPpDGmSou1aK6lS+O/+xb/wBaRDT2JrnQeQi06SpGCsI4mW0ciznGBhy7Rt6sbPLQPOEaVE9OGMI81SNHXjD1fOmCf6OAajGwt0lhhEsyRKbtE9EX6urq/cp++JHe/7Ilg/RiDooum7HlVJcCrJyJhHlYZhQJImWK1WKPICwzhit93hwYOH2G23VhfMrJm5IqIMgE++9QlWq5WVCgkEQmA+n2MYB9R1jXGwFwZj1hyG5oVxTM946HtKQLG+qk4nCVuU1w1p2I02NrLNWDISvWdJnECFipKYOCO5nRx9d+mSb9wYg0zcQ89eJRc3TiQf+99JJe24KqKkKIuicc5A/uwULtF2rZeaeGIaqHgzfv/+7vVOXJYuODZkAnmaYjKZQI0SxlWiUQAFjpAHGIYG0IBgADQDB0eeEOPPGGkNxHub3Sf8nK7vBsot4wxNU0OICQlYmx677RaLgwW587TUebkLqRsHBAEgbJUPCEynEwt9FCjLCscnR2jqFr0Vk0fWZSOOY1y8uUBZloiTGHEUk+C3brC8XdosNY31ZoPdfm8NpqXvBhmIaDKbzxAGARkgGO11T0kSA4x58okrLoyVjbjqSVhHn7fnlq5Sc3/mfV/S6rAYv0sxIP9eIr44o4BxIPPltm0RhIEnHgBA17VUFKu7iCVYkb5SkoJ5rYi860Y/U1aSjADGcURgAh9mC0Edr4th45ZCD1ClPQ4d1qsN5vMFwb4WOSjLPUYpicSl72j5IiCTiuOTQ4RBiNvbG2hDhgxKK+xWO8hxtGMBijUaLSvRGOD09ATr9QZt074Fl8ETK5ynaNeRZeOkmJDRA2PQIAF5bw2331y8wTiOODg4IBOOP2Az9j4t5wM9yhHSMo+LyYQcxezMe7CwuVaUCUqIxwDK0o2wXq8xKQo0bUO6ckXZo8WkgNEG+/0eTdNgsTjAbrtFPwwYt2R1F4YRHj58iNdvXkNwjiInqYqBsYSZOzapYpT/qqRFtDjzSSVuTwshwAQZfSRpiqosYQB0VWeTmKyH8bazGndj4XmaVzqzAWddpy0ZRylyKlOaogYpn5ffzV0t12OwOmFt6PcpnJze9cA6EPm/m7V//H3rnbgsGeMW4pE+Rd15Z8IAXBtoJaEChjjOobUCg7YvMbXXAFUtWRYhyyL/71JSfmSSxOT1Cob5fEZC2K6HUhJZnsFlnCklsdu1WCwWCOIQ2O0AKGR5hqoi0o0zmHbQ2sWbS2ilfGpKmlHO5Ha7hQGxwoIgwGQ6oQ1ti4NxoDmmO2wDa3rsXSns16+r2nczMIAyCoXVLIVhSDop+3Wds4b7TN6GHJzuyB10jqHoutT3eeV5jrZpkVrmYGs7vziI0XU1hAgg5QCYu0Bux0rljHursXEYAQ47d6eCZpQjjKZuv+87xElibeqEvUjVHUz1lrNI13V0IYOY4Z0crXcwOQIZTV3I6zev/eEAAwgB8l4V5NW6Wt1SNd9S5f3q1SvrHSowmcx96Pgwkql+Yw3RBRfI7F5mCcNuv/fRcm7U4A6mYejpM2zJK1SOErvdzsuanLbN7ek0Ibax86j1ePR7vhz0arS2UiPqoG5ubvy8OkkTCC7oefQdZtMZpFIILLTKOaXoTIoJuOBYrVYAyBRCcDJs4ZxjtaSCHYZCkomUpbHerDGZTFBXNfblnlI8cHeRE/OUOBdUWHJ/8UlreuGQkEAEiBOK+NpsN+SxzO7miEYbrwOtmwaQrtCXyLLcW3h6BaS542HQmED71BWnWQ6C4Cu/To49sERPC9EawGWEGm0QhAEVfewdn1lSZakxmeRgykDqAWoYqeIZDWazKQQLPI7uiDtuNgncYeGuwnCrbbs7HWcRI+AhYeBGo5ekj6NkEw0YBR4IMK2Q2MgtZwPGWGdnmRSzRDT/loTjYYjahj8DwJXNd+OWJWg0JUosFnMYENV5v9vBwODo+BBxHOP66oaSSUCOOnWtwAPybXSYvRBEsVdKoyorqtosG7goJtiXe3DBURSFLzqccNx1SG9fnncSgW8OKiez6fqOJCNhCIyk93JzcWETDphNLQh44D0phRAo96XXSRrAvoCKWH7a2DQE5Yky7oJ0kLiyTGcRCPCBNjZVzRr7fg8GSt5pGzo8jJaoawpQpnkr6cnkeKdBG4YLDMOAYjIBA0dZlWhq6vbCMMLl5bV/Z4IgRB/2PqyAc0GJKYJju9siz3NkaYZ9ub+LioNBJCLrGDXQzIjRXnNMSXp/uN9ngQjQtg2KghytnC7vmwXvnhPFMfqeIHcHvZNlovAFWhEQW/Xq+opMTMYRXUfa2rqpcXh0aJNCuNVgxpDj6AmIYRQSGUbW1EVOCmy3W4RhhO124zNvubhTB4Rh4CFMt+8JdQkRRTHiOELbdphMCqzXa49o0XtEMqi6qlHXZJFH7ljSssEpVlIqei/GYfTjBGMsN8V9b9yhN97qbtRerwzAW9txFzph/4ybbzp42N0rWql3P6KLPkhynG9shp+bdUxnE/ISDA3GsbMvlfnKy6eUBuPk8cUtfBbYKsnpDQGaB0hrbacZDaJdjNd+t7GuEHThXl5ceeYZ5wbzxRQvnp9byz3C7wXnSJLEd2u9hTOnkym6rqUNaeeGTV3j9mZJ0K81fE+SBNvNFtrOprRSODo5xs31DWZzSgjo7Mzy3v17KPclzu7dw8WbCzDGMJ1NcXh4gOdfviCDAoCcVfrOEogGP8t0BAznC+s2lIMC3/dltLY2Y8xXpi59xMCm39i3mQu6lIZxIJagJboopVDX9Vv7gYyfnZbV6b9gSUHDOEDaoOS+6yxV3tCFC0NIQ9t6j8v9fm+/l7DzSOXTF5QlvglBlbTggubo+z3Nv4fBQsEjtIYNp+6sgFt5Vmrf9/6dEAFpJYui8NIWmmUSI9sYYDqd+pm9lCOiMELbEpvWjR+ooDDeMcp1Fw6ivZPYfLNgqIja7bbEIrWhB4NlZs+to5iLnOqHAaenp5BS2sQmg6qskGe5P1uI4EdjpDRJ0Q+9vZAIfpRSIgyIPKS1xna7sT+K8eidC+2GJfHEcYyimGAcBkLMGMPt7a3PtWya2hZTtOIkQREENq6LiDXKyvQ4Y5ZrQTpMKSW0MRjGwV+MznRgVFS0OpmJZ5+PozcxcFAxg5WDWRYsLLnMpY84ct4oR29K8IcAjnfisnTEg2EcUUwKAECep5CjQt92SKIEfdN5KNJ5mfpDnkm0XW87sh5pklmSi7LsQbqw6rZEkU1xu1phtphju9kiThLUZYU0zyAHN0eitHppHXH6XuLVy0toTa4VgnPvZ+nYWWmWebxcKsqq5JyjKivqQqywNxQhDg4WqKoaVVkiyylzcr1ee81k3TQeLjCGGGLbLSWbO3ejOCaN5csXr6h7NJrgCnO3mZ0e01kBEgzY+3nm224b7/vSVtPGDGhubH9NG42+Jm2lVgaB4FBqBBgdHlGYQPYSvaJM0cgSrChyiFATIj/AX2pkOsDRWnIYweHcMvwkpBzRNK2Xc1BCROJRDK0Nup4kG84JKIoohMAxqImsM3jWNgzDMEjIkb5fWIS+i47iiBLibVU9jgO0MeADJVPc3t7eHTowdr4L+3claVRR5OAW6UgTyqd19nsO7ndQsbHzI6U0+r5DluXfXJZ2ubGMEMKS+iiaS0mFw8NDXF9fI0koIzLNUuuwNGC33SIIA7LAc+YqYQBtSTdd22E+n2NflphMJlitVtDa2CLHngGKSDtccCtlMpTxy4CmaREHoYcyu66/i3njAlxQ1+iIX86jNhAC2lAGZ2PlbcJKPkitQChLVZHRv1sOUmUM9t0ixjmxzAePkAkb4eiaAGbfSxiACU52NJbIQ05AVAQ4H1ttNLjh/hI17Pfvw3fisiwmOYwBxmHAOEhiUhkDrSUZDsA66wd3GkLGuYe6rq6XOD4+gAEQJwF4oCF4CDC6oLQxGKXCdH4AIzUmeYF6V1pxLMPB4QHAgNYYy3qN0Tatj9aiTL4DtC3px5yvolLK63rSNKHkcnviENY+Ikmck1BwRwZqWitVMN64IIpC6xE74OTkxIvf67pGklDq+TiMODs7xXa7Q1WVyPMCUUxOQ01DhzPNbVP0fQtAWy2fRNP0MMayvowjndwxwd735SpNMAOlCJ5xLjRccGIWCg4TCi9i5rbi74fef65d1/uurO86SKVgLOXe6VtHSUznuq7JzJ7Z7m2kfNau68in0grUlVa+81TWwiu0B5cjZzk7OjlKxFHsI8TuHx1BKYXX56+t5dyKHIcAKKlQzAsi7ATCF58GNvNV0BihbVsLI9NnFIWETPRDB84C1E1tuxKFPM+RJCm6rgWsfjSy+axJkpD+zZovEDlDoW5qb6bxvi9nOg8wdF2HKIwse5lYrxSuHaGqaxxlOaIowsXlBV0eLekX05QKq7IsUViDFhEElpmqfcAzyevo0rp/7z5ubm7IkUmTZjEOY4uuBDBGQzkLaU4MaNjLKY5jpIlFQ8IAbduRwsA6i0lFNo5pkuLi4sL7v8qWvJjf1mC6os3NM0UgbPCFhuAcyijyd7VnmBzvJC9vzynv4GsiDcEAitEF7BAe8xZZzv3znSf4OAF/ksTUPe2Iesy4wG5fYW7DcnvdIogijKOCAAMPBaqmwdHRMbgQ2Kx3mM4K9P2ActiRVjLPaKZjMfembtA1LdKCuk8SWJMejJyDRu8e1LUtMstAy4sCF28uyKDXziCFEIiiCLkV2J6cnuD81Wsb8LwDbEd3cnKMcl/i4GABALi+uraHR4y6ogQSqSSUlABj6NYbmh9stmiaBrfDrXdzabsO+3KPIs/QdjWUlBSzpRUADc4pDb2pG08kcjIE2oV3hutuw/2hDfI+LYdGuJghqSXUQAWFK6yU/X0pHRxkO3TczUTAyFIwCAOPFjjjCIKoGl/Vugu574lsJpx8xXpWKk3Btp3V3xE9Pia7szzHcrXys8wwCqkoq0qfct+1HdbrNf1+GIJzht12h5PTE697DKOQDiKl8fDhA7x+88YyyHusxpWH7BkHjFYIrMctYwpdN2A6m9riUaOua5IyFMTgnM1nyHSKg4NDuIg4N3dyh9o3BJ+7pRVJbuQocXB44PWWURRh25Emtdt0+OCDD7BcrmDsmCkIAoLJmxZnZ2fYrM8RxRF22y0hGzBeO6tsZqrRxuacAi9fvYQQgXUyI1QrCiPc3N5QdmpKgdNOhsQ5mVwMw4CjwyPsyz2SJMFiscDFxYV17jEoy9IGQYwo9yXJ/4y+s+1TLg6u857ZQRhCSkoFGTuC/50ZhntnHNvWyagEF77gc+ed12Va6NiZLox27GcsoQ6AJ1SqPyBheicuy+1mC6UUsjT1cUcUy0XWWuvdDtAGQcQRJ2TLpc2IUJA5gMgzyFFiMs0BMM+m7fsedVXDZec5sktgHTEAYDKdYLvZWaccm/VoD6pxHFGWJZSU+Pyzz5EkZCjM7AvedR2qsrKbKMLL5y8hggCr5crqfUa0TYvnXz5HYmcFMMR4cx1hnCQEn40jhqFHEAh8+OQxnn/5AtoYVGUJxjnSJMV6vYReanuRN9aEwCVicHvR935eOQ7Kyx+c1tKxF+nf74wZ3vflng0A/2IKW2RJJTGAZkZv+6aOVqztPlep5FsuQAzrzdp3ptwK+0c7H2KCgRsy0RCBsEYcme0W6VBwQbzjMNAlqjXqpqFxRV7g5atXiOMYZ2dn6G0aT1VXcFmUURR5m8TLi0skCSE288Ucm/WGcluDEG3TQjEilt3c3tIhNYz2QLmD64kMEUBKDW2oi1Sa4uzGgZJthBAIFYU5CxFgv98hzTJcX18jzwskSWKtIBMkSUIGBt/MzP0Kw9DP1HbbHc0XJRVRk6LwDcBmswFnDCKOEA0RyS+sJ/F6vfYmLVxwaAv1E/HMFspW8hQGgQXDCCbX1tgEAJTa4VuffAvL1dIXaS46ses6CrBgwHa3tSOwCjc31zYtqkE/DHY80HkURGsqqNq2hRC0v4aeUkiUVnDB6lVFkqmqqnD/3n2s1is8e/YM3/nOd3xy1BdffIFnz57h+9//Poq8wO3S5rlmmbdRfPz4sUVuJsitWkFr45NGjCEVwh0z/R3XWYYhB4NB35DEQ47SGqNTJuR8MSNB90iMU5EKdC1BVUlCL9tqtUJeFHDu+O5QU1ojioV3CZFSoqnpAm2bxnrB0jyHMfqAp9MpBnvYaKUgLJuWjLMFZrPwehd6AAAgAElEQVQZbq5vSFpg47OCgC7nIoqQFzmGngbfi8UCxSTHl0+/BGPA4dEBQRpKYbPeQMkBxlAyuRACTdPi17/6DGlKgcLEvDXou/Er+ZSOHeYchxgndtcwjnZOYLAbdjCGIn/cJiC4Q3lXf+GIK+/5cgeI+yxCW7A5lIF8d+8kN0FA4bVNU8N5uzrHHTezdOQurpy7FM2f3LOaTqdgdl7jtXUjzRxHKf1l7SKHOAX52SSTFnFESMznn32O+Xxm54RkvP3q5St8//vfx3q9wsuXL/Hi+QtinE8nmEwmuH//PjbbDaqqsqYc8JAXzF3kmNYKLmvV+b0qRZ2Bu6C5oJlPGIb0fgrhoTIqEhTqsUGSpLhd3uLg4ADGGNy/fw/X19cYhh5Zln+dj/+dWcYYH83liqUkSdB3PRVtw0h7Q9BM2plB1HWNYlKgLEtIO7/rh96bjAN3MKNhhmB2xiG1RCgIdWjaBlprxDG5QrWNxOs35zAG6LsWB4eHOD05xtXVFYrJBFdXV1jMieHvGguXv6us4TmhXtoSFXtsNhus12s8f/4l+mHA4cEB0jTDxeUF9vs97p3d8wlMbdvi+uba68LznJyB4jjG+fk5Hj16hCzL8OrVKxRFgQcPHuCXv/olHtx/4PNgX7x4gePjY5++1DSN9yRO04wg3CiEHrSHh3/feicuy7pucHh0gAAhekVMqPV6g0lRABbaCqIAbdNAhAxN3dqwZgo0LssdGCOmnnMsSdIE2hiUuz0xHIXwl6PbNEmaoihyMjPfl971pipLhDZhou97T1wAgLP7Z8Rq1RpKEnQ6nU5tmjmnSzaOwaDBucHFxWsfQaTkiPV6TT6PSUIwRVVBjhJ13XiJCPmHUnC0VGS0MA4jBkv2GIYBQghsN1vc3t5iuVoiy3IwAIeHh5hMJ9isN1it11gtl5BS4qOPP4KUEvP5An3X4eXLl3j56iX+6i//ysI07/caenJJGvqBpEcGVkJhQ3gBv3fGcUDbKXRt52foFCZL0JMLgibNI3227rJ0LyPnJOMgb8q76C+lFLjk1ouzh5SkUxxs/JKL8xrlaOeDCXa7LcqqogPFRmoxRqzwH//4J4jCCJeXl5BK4sH9B/joo49w/vocQz+gqkpMZzMcHR7h5voGaUamIPtrSqnP8gxMa5S7HRbzBVy2oJQSUhHl33nm7nc7uLzNOCGZgivqlFYeLtxtt4jjBL/85a8wn8+/ws5+35dUNJ9jAB4/fIznL56TpSCjLN48z1GWFWbTKaqqRDjSWVfXDeqa/FC54DaA3snsqGAezADGGSIRYRhHwHb0zvwgCAQeP3qC5y9eII5C33klcQLOqIB8dX4OpRR2+z36rkNT1xjG0Sd8DP2AUUpst1uUZYWh77Ev93j27Blub29wdHQMgMw2ttstjQisP7eUEq9evcJsPgNAfs1VWSFLM8xmZBB/eXmJjz/+GGma2vQb4oTUdY2u6/DRk4+w3+89y9olsSQxsbOn0yl+9rOfYTabYTqdYrFYeL26c/n5fYu9Cyy069VrI6XEdrnH/OgAQ99bogPJSFyVYaSCNBJRTFCFY9m1bYc8z5AXOXa7EuMgfWCqo0Mzxq0xATHE6qpCXTeknbTwEjOk3yyKAruqxDiOmBQFkTSMwTgMGIYRWZYijEI0Fn5yVZ6xWrkojvDkyQd48eIV+m5AkqbEBquJhOO6E6fjdP/fEW5WyxXquvbzrbIscXlxidvbW291lRcFAKKJ7/Y7TOwgP44izOZzYlWOo59djaPEfr+D1ppYcbs9yRZEgLN79/Bv/82/fa/7y//j3/zvRtt3gS5MuuTI0zTwKR9KkSNO13eQo7RsVeE9e43da/6StGQC519pDOm6tCaighxHiLd8fB1SEASCrlUD1E2N7373u7i6ukJVVei7HlxwXF1e4l/9q/8eX3zxBX7605/i9PQU3/3ud3H//n388N/9kEYAz59b3a1AlqW4/+A+Li8ufDTT6ckZptMZdrsdbm5ucHx8jC+++IJIH/fvgTGO9XqN6XQKqSTmszkdzlUJYUk53LIbHbliGOnzi0LS3o2SXI/CKASMwXyxwM52INPZDJOiwMnJCf7lX/7gvd6DAPCbz39tpFQ4WCx8oLZSGgtrWs8AZHmGQASo6grz+dzviVGOVnpCZBjXpTJGuaoUSg/v/kNifeZnmozR9yosQhdGEbbbHeI4JuZq02K9WQOAD6FX9gyrmxrX19e4urrCcrnExeUFXcBCWFSLWNSTYmITahpPjsuz3EvhdtutnSmSR3OW5dhuN8jzwktjhmHAo0ePPLmtLEssl0ucnJyQ521NhLPZbIq+J8jWvbdSSpydncEYgzdv3uDg4ABFUSAMQ1xcXODJkyf4X/+X/+137sN3orOEIXeIo+NDGE4u+3034OT0xOY3diTLkBoioigszhnCKKIEBaWhDVDua3AmYAx1fFRt9KSJFORpCQDbzQ5t29lBMUGawor4szQl554KXudGCfIM0+mEGLFdh+125/VP0+kUaZpguVx5vc7Pf/ZrRHGEuqqxXK28Vg4gQpPD7Z10YL/bY1/usdvucH5+jqZtsN/tcXR8hK3NtGSc/rzSCn3bUnpKFCGNExQWonDROGEYIrSFAQNDFIY4OjzC7fIWlxeX3guyHTvv8PE+r86yTbkgCUfXdTSDHC0pQipLwlK4vr5GZufkWUYexs4I3M1mnJVhwIlJKJX0dnQwNilGkmGBHgby71TamxUIIzBKuqCffvHFXRqE9fv8v/7Pv8f3v/99/PjHP8Zms7EFZYOnz57ihz/8IfI8x/XNtQ333iHLMvz857/AD/76B9CaoSwbMjhoW1xeXvmqvrAX19OnT32s2HK5xOMPH4MzjhcvXyAKI5ycHGKUg0+KYIyhLum/V1IiTUjWopRCbAlJQ997iVUYRjQb7nrriPRuHEVf9+q73rOVw4gi/ZKEmoUoDNEPA/p+QGcoXmt5u6TowbfkYM4HuO97YhkzZ9R/N4d2PRLnzJtZkNZxwHq9sobpCbq2w3a7hVJ0MTprud4SDa+vr/H69Wssl0ucn59bElfg9e7TydRbzJVlibZtce/ePWRZhqdPn3rnHyft+PDDJ3j58qX11ybns8LaJjrP5eOjY/LFtQzh6XQKKSUuLi7w6aef4vz1Ob733e+hqik39eXLl/j4449xcXGB09NTbDZrHB+d2LFX46PuptMp9vv9730278QOHaXEUPeIowRRkEArjcmUKO3KGjKTIJbgBQNtWYMKnAVgTKCpO1DkVWihshFd2yEIQyRJgqqqsbxd+YxIgIgJURwhthdlIARR2Wv6bwKrVdxttwgt3u1noTaQlHGGstyTc4oc0feD7xhdWvwdvX8gGzSp8OzLL/HixQusVkuM44jXr18jiRPkOWnO6OtWePT4Efa7PYJAYOhp+B5Goc9RFCLCfD4n1pmdyXLBoUZp3V5ahGmKUUooJfHgwUPs93tIOeL+vfu4ur5C+Qc2yPu2HNnJWN1q13V+9kMzSEXevNZ/1y1loXbHLi33pGcjhr3tJAFPtjKSOlClFBIft2asXZjBiBGvzl9hHEZ8+OETVGWFYRyw2+7w85//HJ/+2afWbSXE6zevUVc1Pv30U/z6V79GEATY7/d4+fIlPv30U3Rdh9/85jc4OzvDl8++RFEU6Psef/Inf4zNZoNnz56R8X9R4J/+6Z9w//59bLdbXF9f48MPP4QxBr/65a+QJAnCKERd1UiS72EyydG1Lbqup3CDQKBpCa2ZTokwkeUZjSEYHeZKKjSqweHBATlR1RXCIMSbN6+/jsf9zi03X6zqCuNADmOTosDOxpsB1hd4dHNzOk+ck5TroJRWliUdoR/IAJ9xY3NXOUhWRmYAvfWTHkcaNRhtMKoRfU9B8y4vs6prtG2Dpmnx+vU5Li8vSZb0+jUODw9weHToCz6jDXTbkCdscpeAwjjD+etznJ2e4Vvf+hZubm+wXq9xsDjA7e0t8jzH6ekpLi8v0PcD4nhElmVI0xSr1RJBGGK332EYejx69BgnJye4urpCGJIBzHJ5iz/9zp/i899+jslkgjdv3uBv/uZv8M///M/45JNP8OzZM/s5ErGR7AVHr9P8QyHk78ZlORCE2XYtoiCylOc95allKdqGsPjJbIL9bo8wjkg3VNdYrWqb2u6Yewr3H9zD7c0STdN4eYijE7v2fhwo0DdJKDbr9qZFYJ3z65rsnwBLV9YGjX3Ym+3GOpwQbNa2LQZ7QXLOfUBw13XW1H2AVgq3yyVevHiBtm1QlRVBejaJZBwlphZGVZKcdtq+gxAcq9WaLJsY/4onJxcCUisEAUMUx8i09t6LXdcjz4j51XaUg8c5g9ZU7R0dHVKmplSoysrDM+/z6rre08vruiaW34agQucOUjc1kph0sWEQkl2WLZwc7ZyIEzTrI+RhhNa9N6xQSvmQaYAOlSiMLMGLo22p0zeGIuK++6ffxU9/+lPM5zOsVyv87Gc/x1/8xV/g888+x5OPnuC3n/8WYRgiz3P83d/9HR49eoTKpt+4ff/jH/8YZ2dniKIIv/71b7DdbvH48SNIKfGj//gjLA4OLORF853Ly0ucHJ8gz3P84ue/wEcff4S2bT15Kcsy/Kcf/SM++uQj7CxM5+ZDi8WCHLg6cpFq6gYiEF5r6cgr2+2O5q5ZRs4/Ufx1Pv53ajlHpSAMEYUhkRTDEFEUoZgUeP36DWazmYeyhRBYLBa4vrmGSxwyMIij2JsIKMvYpkg16jDJlUf5PFPGSNNJrFoGOUrcLm+9bd1/+fF/wWa9QZ7nWG/WiCOCZ4uCskpPTk6xWq18LFiWZbi+ucbJyQkODg+w3Wyx221x/959LJdLKEUe14ILrNdrzOYzvHr1CmEY4sGDh+i6DldXl5jNKMD69PQMyxWd68dHRDQ6OTnB/fv3sV6vMZ/PoZTG9fU1vv3tb2O/3+OP/uiP8JOf/AQff/wxhBB49OgRxnG0geo0/nLkT8YYdrvd730u78RlGQQBmoakHJvNFiKif49di64U5os5MaqkhLJJ3M7myGiN0WaqLQ4WePXy3M+ZAOog7y7JAVoHSJIEWitEUYjLyysb86IAe6l2XQcXzuviX5zfa9u13gTdMVMnkwkuLi7Q9z3Oz89xfn6Op0+f+tgmF6N0sDjw3ekwjJ5FKyU8fTlJEg8Bdh397EFIMBwZH1MFmqUZyqqC0Qaz+Qxd36GqKpT7PfI8J3gwoXTzNE3BQo5h6FGWCtPJBJt6g+99+j18+ezLr/HpvxtrGEkaoo1GWZUo9yWKoqBfZ8zbtDkSFufMPy+axziG4kijAdzR9AGb/BLcSTLGUWK5XOLo+MjPp43WuLq+xnazxccff4Srq2t89tlnePLRE/T9gB/+8N/hBz/4Pp4+fQouOP7Dv/8POD09xfPnz/H48WPUdYPdboemaTCOElVV40c/+pHd6xqfffYZ7t+/h3/9r/8H/OQnP8HJ6QkYY/jkk49ppt40CEMabbx+8xpFUWAynWC1XJH4PInJJCNO0A8Dbq5vcHNzi+l0gu12hwcP7uPevTOUVYkkSZCmCbquR6hDT0xSUkLEsf+8giAEs6bu3yx4dEIrDa0k9l2H6WRC5JlxIP9nTi412ljLNqOxXq9tohAVcJxZG0ZLPBTMktiSBPP5nGaPBpjP5wTRjmRiQEX+gNubG4xyxK9//Wv85je/wZMnT3B9dUUG/ZatOp1Ove44TReoqtKeXzHm8wVevXqFOCIHqDAIMZlMIAKB29tbHB4eomlqrDdrfOdPvuP5AGmaYrm8Beccs9kMsznJnKbTKfI8x6SgxKeLiwssFgtCL7IMZ2dnePXqFc7PX+Hhww88yafrOhwcHODly5coigJPnjwBQEbzcRx7iV+WZTg/P0eS/P4Q8nfistzvK0RRhKEfAabRN3c5lMba+m3WW/Rd5x+WY0IBNNCezabWb9JWRgaYTieU5J4klM6x24FZL0IHj2pNXZtWGpqRbpG+pgLnRPyhGSpdgoeHh6jr2jv49H2Pru1wcXGBq6trXF5e4Pr6mtiUI4UARxGxzxgYttst5osFGSbbwfqdj62yZJwRcRTbA6/CpCC6/363RxInaJoGMYsRRiEOFgsiMSmNs9MzrMM1ttstqqpEkRdI0xRlVWIynRC7dhzJjkpKHB4dYrfd4ZNPPvkan/67saSkpIeqqr0huksXYRwe9iINsPRGF+T6wywBbLQFngbTGs1QQ1m0AIyh23c+U+/F8xf44z/+YztbbxDFEYa+x09+/BP82Z/9GYZhxLNnz/DmzRs8evwI//Un/5WCe798ge1+h8OjQwxdj89+/RvkkwK/+MUvcHZ2BiUVMWIHClQfhsHnWipJob4vXrwAYwz/+R//M2azGT7//HOEYYiTkxMAwNXVFbIss9U/JacopRD1EYIgtPrOyOpGR1xfX2M6nYJzjs8//y2yPEMYBNhut1gsFv5QHyXpQ6lgIH1pWZVk4ae+ybME4MdIxihMp1O0beuF+E3dILVs57KswJ0blyQLRaMNlCXtAbCmA4PXUgoh0HaUMBPFEbq+w6tXr8gMveuw2++w3+/x4sULioWLInL1GQZ8+eUz2vNKe+tMgNJ6Li4uvNwEIGu8LMtxcHBAyUvG4HZ5i+OjI0wnUzAwtG2DBw8e4ujoGE+fPsWDBw/8XPLJk4+w3W5R1zUODw+RpSk22y2CIMB0SjPQ5XIJzjlub2/9Bfjw4UNUVYWHDx96ZKPIC+IPaI2bmxucnp4ijmN89NFH+Pu//7/xV3/5L7FcLfGrX/0SH3/8CZ4+ffr7n83/x8/+/9UKwxhpmkKIAU3TelZgXhSorEG4tk4MVA2EAJh3ziG5hfZeqGEYWjF27+HRpmmQxLHPK+y6jvD8jkKdx1H63DcnqB6GAVVZ+SgmpRSub64h7SXWNA1ubm7x2We/wcXFBfa7PRYHC3DGPSHDaPIgNFojimN0HaUsTGdTbNdbMhY2d2bASimcnJyiqirkeY7Vek00bxApJE1SSGswQO47FP213+/x/7D3Jk12ZNeZ4Hevz+5vjAmIABCBGTmTSSWVpEQWJZGiyrSQyaoX1Yvqfde6619o27VRWf8CdVl1W2swo1jSoilSmclMCZlIJqbEFEAEEPMbfb7ei3PO9QeKye5NizAD3IyWYCYi4sVzf/ec851v8DwfnW4Xs/kMJ8cnCMMIAfuEiiA5DEN2vCAouj8YYDJ5tbMs8sLGVJE3p4brhpawFYSBlfII1AU03Hy1dPMsy0h3VpSQIOR79+5jaWkJcRzj3v17SJIEb775BsqqwmQyxc9+9jN8+3e+jf/jv/2fePPNN7C3twftaNy5cweXLl3CZ59+hi+//BLvv/8+Prv+GaIkws0vbqIqCL50HBdXr17F2toaNBSyPMdoTKQex3ForeA4cLWD8YgOxOXlZTx6+BCAgmGW7v6zfWxsrGPz7Dk8efIEy0vLpBkuWo/ZpJNgb28fw+GQXqdyECcx3n77bTJODwJMxhMMh0M8uP8A1167hn6/b41CpEnUjraG9SVKrHKhfnU1zM8gLW2v30M6T9HpdGBqY60HO0mCk9GJXT8lYYLJeMJG/bC7QzTEQqWeTqEx4j9MjY6I9+fzOR4/foz9/X2Mx2NorfHkyRMcHx/DNBTWXbNVnsthzdLU+z4hB+PxGGfObGD78WNsbGwgCAImztUwDfnMjsdj1nKGePz4MVzXxZkzZ7C7u4uiKDAYklH8xsYGZrMZdnd30e120e108fjxY5w+fcoaW2it0e128ejRI1y5cgVKKbz55pvsWOUzYjmHaQzW1tbQNA1u3bqFt956C3Vd4wc/+CFu376FlZUVzOdz3L9//8WfLNN0bgud6zrI85pyKDlkNstyq5/MsszizCIdUQDCMECR5xb6ms+INShTqHgPmrrhgjK3YtfpdIamoR2BuLcAuQ1wHo/HKPICs3mbZ/nl3S/x85//HEkn4X0L+YHO2QR9MBzY4N1G0d6LiBXkmxj4Afr9PsbjEeqaLJiKokRVlbh79w62tragXRfHJyc4Pj7GjIkQg+EQSpFVWsbxY7UhurfYQK2urmI0GmE8GmF9YwMhw2HkdauI2ebUaAA0eY5et/cbvf8vwiW7YN/zmShhYAyZAYgGU8T6QUgBt5LuobVGpQhSn/N+PctS3Lp1G0tLS1RAphOUVYlTp07h3r17uHnzJl5//XXcuXMHV65cwT/85B/w+7//ezg6OsJsPsNf/V9/hbfeegtPHj/BeDxGr9/DT37yE2yd24RyyKrs7bffRm/QJ5Z0A9sAigOVMCSDIOCoLQfX//k6mqbB9vY2Q8o1s7QJGr579y60djAcDrDCFpDWxacsCdUIaLfV7/ctRX8ymeDzzz/H5uYmHj58iI2NDXz00UcoigKnT59GEASIkxi+51PQehgSwzPw4cDB4cHBb/YBeEEuWTtpjoHLsgzT2ZRIZkWJuibT/dF4ZCVJZUWxW6KxLIoSWimOcWtgmgplmVsDfDQNZvMUaZri5PgE1z+9jtOnTyNNUxwc7OPg4JCCoH0P6+unMRqPEQQhPdcpwaVJnFiELQhDTKdTXLhwAdvb24iiCF988YUl5RjTYG2NlA0NGkyn0+fyeMX7dnNzkzXvIxwdHyGJE5w9e9Z62a6uruL4+IT3l6dQ1zWyLMPW1haRhJaWiZOxvILHTx5DQaHX69kd/GAwwNbmFj75p09w9epVAMD58+cxGo2wvr5uJ+Svul6IYum6HiTluyhKC0XmeU5WTUygoEapsbtCx3FQcpEdjcYoi4Ljj2qWkygEvLv0+aa4nofDQ9rBpPMUnu/ZLLWCmWWe52M0Im/Z4+IYaUoP1t27XyJN59jf2yeB/3BIdHi2mvIYQtEeufbXdW0hWKI9G0t6iOII4wkllgsRA4DNrXvyZAeNIicMpSnQOdekpYqiCEmckPE6aylFLE75lgk818Ph0SFBu67HOzENz3MBTouoqwqN1paB9zJftLOu2H6NOnG2EuDQbSLkEBpBzZvY2JVlCdcjo+qypKDejz74EN/5zncQhAFu37mDr7/7dVz/5+t466238MEHH+BP//RP8ezZM6b0FyiLEkfHR7h+/TquXbsGz/dwmyHNOEnQ7XZx5coVVEWJm7dv4fyF81hdXUXNZKE0k5QSepbF8quqKkvOefTgof1d65r2riJnIp2dIhP5psbx8QnqxuDUqVPodDoYDAZQDXB4fIQkSWBqahpMbXBwcIAwDHH58mX0+31sP9pmow7PsiW3traw92wPW1tbBBtqx+6MaO//G7jpL+CllCJijqnh+wFOTk5IMsL2i5Vp0CjJQVU2WNxogrolxN4Ywxmn9FxXtWHjcTpX53MqwuPJGDs7T9Drtf6+eZEjDEKWRTlI4hi9Xs9qjn3ft9ahvk+ETGPoOZjNZrwvpNcwGpHEbjQ6wdOnu3A9D2eZvNPtdmg6DQK79iDypMbSsG+ZvTHLjeq6xnA4xGh0wl/fxf7+PgaDAef35pYbEEWkhY+iyE6u8/kca2trWF5eBtCm4ayuruLH//3H+Pa3vo07d+585b15IYqlSDnE1LzXI/GpSDUEUsgNBdxqrdHrdsksADS16aqC4n8/nUzgcFctb4hhJxEoolOLpZ10KmiAvCCmqna0jbMqigJPnz5Dlqb48u5dhGFIaQlVhZCnziRJiALPnqySKShWVAp0QOVZmx94dEi7xTlDZLLYF0g2TedQWpFhNZR1509rIheNxxNsbKxDByE4d4beG9fhgu8hjmJkKQfIckSPdjQ8RXRx35NG4aXXgtN73TitdyQ77dQ8eRlDz6C40DiuYwX46ZyK0XQ6ZbnFFq5evYo4jjGdTrG2toqHDx6iaRr89V//Df7gD/4ADx8+tHuX7e1tmMbgzq07dod4ZuMM3njzDfzt3/4Yf/qnfwLTkKxkNpkijmOsra3BKGLXQqL62OfXY0cUeU5938e9e/eRZpn1HV1sruRqGti4NmMM5SHWBleuXqE1SE3ZgiQHCC2hCTMiTFy8eJHsxdZW8eW9L7G6uoKjoyPMZ3NcuHAB4zE1h65HZu95nqPX69KLf2UPC4CUAUTiM6jrEr7v2QDkqq7sPwHA1BUAmiDRcK4vowRinA7AInGz6QxHR4eYTqfw/BDj0Qiu56LX61v2tOM4iKMYrusiScgs4OTkhCLqDO2v6fmBbRSbpkGWZtg/2MfBwT7Aum4oajSrusI8TcmUoNvFzz/+CF//2tdxyGYXWZa1RCM+P9M0tT7WSZLQSurwEFVdIUk61jO30+ng8PAQFy5cIMLc8gocl9jBaZoy8tG3/yuKAufPn8dsNoPneaxb1vjtb/42rl+/jnfeeecr780LUSwpHVxDac0i6rEN0JXdnOvRS3VdF1DKeqCS7ylNcI7roNOJEYYuDvYPoZXD01dlHSnCIMRkOsFoNEKRF9ZgXDpwkZpMp1NMJ1PkeY6PP/kY589fwHgy5t0p7A2RDifNUvuwhhHlzUlCOQDUTW1zLfM8x97ennXNXzzg6KIDjKR3DVQDO1VLLBgAbG8/wenTaxT1xfvYqqpQlAWSOEGv30eazpHEie3ijTHsDaks01OCtl/my3DWntizURag4RBlBdUoS7Ro2Ji/RAlTG9y/fx+9Xg9LS0P0uz3cf/AAs/kMxZ3bODkZkY1eUeDSpUtIsxQ//+jnuHL1ynP3/uRkhDiOcPnyJfz8o4/xbO8ZOt0O/tN/+l9QMvw5mU5w7+6XePvtt1HXldW/KY5aAmp4oWcbPWm8dnd3KX1CU7af1prtIot/UTBbFjkFrE+mE2w/2sb6xjo1qf0eTo5PLLzb7XbRH/RhaoOPPvoIv/O7v4O1tTX83X//OywtLeHZ02fYOLNhyXDzOaWSaEfj8qXL1Lx53quYOL6U1nC0Rl5V0KzdRV44cmIAACAASURBVNOw1yp1FMLOlhirBm08VZ4b1HXVhinwJJllGSaTKY6Pj3F8fIiNjbNkVRj4uHTpEra3twHA8hr2D/axvr6Ouq4Zbp9ieXkZo9HIOuzIHnpnZwePth9SelJF7HDaR5PphsiCiqLAaDSC7/n47MYNDIdDBL4PzWYbSZLAcRz0ej2MxiO74hAuh+d51uauQYPDw0PrxpPnOU6fXsd8PkMn6KDT6RCqVxQ4PDzE2toaNW7zuYVwi6LA8tIyDTaej3fffZfImV9xvRD4W+D71nLJmv0K05A9v+TDSdExHuq6QFnl5OPquqjqGmmaYXf3KY6PT5DlOQWUzueo2AmlKiscHOwjnafWXi7LMtuBzGdz7O3t4bPPbpDBcymwrsF0OkEcxzZrEOzo3+11aSeqHYRhYF1aNE8dnueyRIX2mkfHx3j27BntC1nXJ2xFALZ4A+BMw4Yz2ZSFWUR6UpYl9p7tAYCl5gdhiMAPCNrltAJhEOdFAe04nDai7de9Sn0gdMMPfMsarMrKTpRGiiOHg3seufI8uH8fH3/8MZaWl1iDCHzyySfY3d1FGIbY2d0lmVFZYnl5GT/+8Y/haAeD4QBffPEFsizH5cuXceHiBQyHA7z3zfewu/sUZ86ewfe//318+9vfhlL0TMzmMzx88BDntjbt9CBCdNpl1dZjVQ4Pgcr29/ft7ygh0yQxAFwOopZLdpiai1fTNNbhSd6POIktd8DzPGhFBe/db7xLzPAnO+h2u1hZWaH8TN+374kctHdu36EDfDzBeDy2a4iX/fJcF540teCpsKHwYqVhEQ5aSylrfE9sbtKZW5Z+lhNx5d59TKdT1FWFPM8oSYnfb4paUy0HhM+ZwA/sROp6LhMwHYtYABQqXZQFtrcfIefBg4YcNnn5pZB53/fp92P287OnT/H06TObG0xJJERGHA6G1jTd8zzEcWy1yb7vI4kTLC8v4+DgAN1uF7MZnfNitq6VRhiENoyAzvAplpaWaHXl+7h8+TKqmly4lleWEYYhzp0795X35oUoltLZioGwVgqO65K7QlXCYwNpilXRmM3ncB2i8IchdS2igfQ8Dysry/A9D45DzFeltbWzo4DowBbJLKO8v93dXdz4/AZ2d3ahFbFOXZemwDihFJG6aolC4K4uiiKCtzgfzg8C3iMYnhbJNxRNAwXg5OSYHrZGQSnHQq/yUHmex0xM1xZFY2qbBSi5lPbB5Yeuaag7kuUPmR2UVtumtCZWGh+qMrkTW/gVDFuUBYo8t4bKcg/RtE2NzxFUwoL+8MOPsLm5icPDQ6yuruKf/umfcO/ePZw/v4XtR9tI56k1u9/f30eWZbhx43OMx2NkaYazZ88gL3KMRiO89833sLGxge9973vo9XrUJDFZY3d3l/7+ubOWla0U6zy5uGVZRhAn0/clDinPc+TsdVvX7S5W1hNNAxhQyK52NOqmQcNp8oRqaJi6wcHePhqevo1p7LNom9uG0k6iiPxn/+iP/ghbW1vYOr+F1dUVPHjwEEtLSzb9oa5JG3jr1i1qWvPsN3r/X5TLNIT6DAZ9ZGnOXtDtMwjA3gMb1t20fxY3mtlshpNjSpUReHM2n6HksOSSp0KtFTyXDA+kEQ+CgEhpbE8nDb2kHBEJTiPPczx88MAWRsMadTnD66qCqx1oKLhKo8oLlHmBPM3Q7/bgKI3dJ0+YTEfD0Xg8ssYggtwtatmLoiDuCWs9ZdpViuRdAt2mWWrDqAeDAQ4PD3H27FnLeify0aH9vEge6GLj+MvXC1Esq5qKgWa7Jumo8jyHx6w+m7ZQlrw3JI1Pt9eB0hpJkiCKIhRFifv3HqBkFwutNaaTCUYnI6QZFceD/QMcHR7h0cNH+NGPfoTbt2/j4OAA83lK0g+tKWkECmEQII4i+zBR2kLNYl96cA4PDmB4IozCEGFASeUSWupygklRFhZSELu0hm+QsOAWD0L5vcFOGysrKwCAuiZChEKD2jSYzWfI8mzBocjBYDCwUxLlYYZMCoA9iOk9pUnzZb8I2qK9svhrts0J7dNnszn29vfw6NE2PvzwQ/yb7/0bPHjwAFVZ4Uc/+hFu3ryJr33ta7hx43Oi1lcl3vvt95BlGW7evIVz587B4X3419/9OjY3N/E73/42tra28PWvvY1zZ88iSRJ873vfw9qpNQz6AwSBjytXr+DM2TMYDig8/DliTkOFTQ66sqLpdzAYwPO85w6x1hO0YSkB/d5atKJoG1f6rdtG7smTHaRpxlC1YV10YX1hBaFQbHWmlMLyyhB/+Ic/wJ07d/Fv/+0P8Y1vvIs0nWNnZwee5+Ls2TNYX1/HfD7/tc4pL9NFZK2MJqWapEU0MVIRqGqa4MTAXGDW2hhkeYbZdIb9vT083n6E/YN9a515eHiIvb09y2Im3gdpFhtQJiXAqEpFk1en07EmKY6jrctUkiRwXReHhwdWKqUaAKaB77ho6gaudgDTkH6WizkhCySpGo8niKKYdLsPH9gVk+eRrSgAm2pDCgLav0oOqkgJZeoEgDzPyAAmiqwnsmjsO90OHjx4gF6/Z92kRqMRhaWzEbvjOpb886uuF6JYdrsdaE15fRKNpaBQMmyoACv+LsuSluAOuaHMpnM7AUqWWsWayyLPLeV6Np+R6w53Frs7u9jb38PnNz5Hp9PF4+3H2Fhfh9YOL7OpaB0fH9NexfdQG3KY8BY8QaVjr2qyqZOlttbaQq2aYY7RaGThMurEa5vi3SbG0+8aRfQAiGeoUgrPnj2DRNlUVQmt2r3uaDSGaai7dD0PpjFkVt00DFtLTBhNlwItUpf6CgIL/ICz/2jCkhxH0dOKtOng4AAff/wJVldXcfvWbSs5GY1GuHbtGr744gsopdDv9zEYDKwdXRD4tkv+1re+hTffehNnz57F3t4zXLlyiaUe9Ix1e10sLy3xlEtFyPd8GN5xK26IFEubXMdFFJE+rK5q3pXnC1pP2CaMVhpgOFbbQqdYk6eov1ugfLUTDRlnGNs4er5nnyFp+Cr++U3TIPB9LC8v47vf/S729w/w6WefIo4jrK2tYWtrC0EQ0mRwMsJDZuq+7FcQBCQxy7KWbMXvP8GMHTiOhh/4qOp2VVBVJebzFKPxCLP5DNPZDFVZ2L3iZDKxaEIUxSiryjr+KAXMpjMbGC7WeNK0i+UnNdX0ZBRFgaOjI1Rsz7k43bqus9CYgR1+KHh8ZWXFNlzE61AYj8fE4eCBgxixNIgsDZdgTNNKavg9kTit9me60IqnX5DWfjAY2CErDENcuHABT58+tYxbQUV8rjkA6a2/6nohiuV8nrb+mtxB+IGPgDVe2qFfNkszxHGMTpemySzLbEBq0xisnVqD57qIo9i69FQlGYgbLqDjUetQIc4XWiksL5MA23E0XCYcKMbnyY2fGGdC8jD8MIiMpSyJ8SUHi9zwmkXtcoBJV9MeQvRAWngX9D0z/rC0SSX/8mCywbx1jclkTGHFZcEwMaxAWeZGxQ8O0atdaO2wK80rb1jRR4pf7mw+R5pSGnwQBpw4k+In//dPcO7cWaytrWF3Zxee5+HTTz9Ft9vFw4ePMJ+n+O33v4lPP/0Mm5ubONg/QJ7nRJIYT/D9738f586dY33vGJtbm3AdiuMS8hZ92P1WCsA2fICypghymAHg4qWt2F+IHUorm3qvtAb4+8meMgxJHhCGARltq/ZhkWdUrobj5+qqthZjnkeZh7KbkueyrNoAaGMqnNs8h/e++R6GgwHKssTbb7+Nk5MTG+dENmTn//Vu9gt8CUPT8HkjGaqyR57NpnYfXVf0Xud8XmRZitHJCADb5ZmGCTnUNHU6HV4nkNsOoUsE4UuykhQhYbkKPCnFSohjo9EIeVFYwxTHdRAEIVlzWsP3di+aZTnW1k7BccRqtAGZt1Ns1sHhgSXxSAET+HU4HNr3RvyJrW6eVw1KKcQ88Upz4bGfruu4dgBbW1tDnuWYTMZsRcrqBdNw6Hr5lffmhSiWURRZH1fHFZs70lOev7DFRSlHGBG8OedUkDhJUHHI6dHRMXae7ODo+BiTyRTT6cxmEM7nKUajEX7605/ir//mrynFw3Pheh6yjB68s2fPstMO7aYoaFdhNp0hjmJ7SERRzIdChTzLMJ3N0OuSDMXR2nbkQRigy8Jb0QtRh0YF0RiCm13HsYefTJySj2lMY1ltYrUGpdAokDBdAY4CYBqGZAq4jkuwGmgSmM/mSNM5sjxDwXpOwxIcIk2Fr4QjaIslTWxAJ0k4B5J24zdv3cI//uM/4oc//CGyLMMnn3yC0+uncfvWbQDU6c7nM3z3u9/Bf/2v/w2nTq3hb3/0t/j3/+O/R54XiOMY/+F/+g9455130Ov3sLa6all5nkf5rGVVYjAYWoao67lWAyvynppTIWQ/L1IC0dfJ3xNzfN/3RZJnXydAiEgQ+Hj06CHOn78Ax3UEl7WQLHhFIE0bPZ+GdXMEcS0WSrEia4yB49DXyGEVxzGuXbuKN954HTdu3MDGxgYePXqEXo9gMc/z/7Vu9Qt9HewfYDKZUOPGE6NWGmgU8qxAXQNVZTCfTdHAMHxZMQcjw5OdJzC1weHREaq6pj2242JpuIRut8uTK9lnimG/VtpOYRlDnmJqEYaUhCSfhW63C8/zMJvNeIBgvgU38k1DbFpiXPv2HHVd0loSumY43EHbhKaTkxOKJWRPV9lJlmVpd6LT6dQ+93KOPb+3rclOsSR+ScNyK6WV1UW7rovBcADf89m2r7TTLNBKp37V9UIUy3Q+JwizadCIVIRf/O7OLqqqtJ12FIUMk1FMTVmWmLPGUXFHlWUp5vMM8zkRNubzOba3t3FyfALXbU2dPV7+drod+p5Nw6xc1ZoEcJcjB4YQJ0xtUFYkX9Ga3DbkYKNdkUfuJI4Lx6HCLJ0b0GpL5RSTmyQ/lxoGgzCK4Hoe8rKEw4fn4qIf9nsqK0A2nJ8ojE6aPjnpgg9NmVzLsnxlSgAg4kasrmskScdO22ma4su7X+Lhg4fQWuMXv/gFXN+D47nY3zvE6fUNW/DefPNN/MNPf4o33ngdP/jBD3DttWv44IMP8M1vvoc//uM/tp28MAaVduC5AWazlGN4FRnm8zPVsG/x4gQZSIA5H0xNQ2HSdMB4FsUQra1lTxt5ZkgS4nku/vzP/9xKS4SmLwQNucwCJKc1afCsZ+7CymHx/weczOI4Lhd9Da3IZu3MmTPY3NzEa6+/BsP2aSejk1eZqnyRxtugrgw3HcT0r6oKNe+EFfMOROI2m82sa49SCscnROwRQk4QBLyW8djD1UEUxfb5AAj+FehevkaMX6SYkO48R1VRYy76dYAlfdxiUZiA+xx7n+w8KxoSGoM8p2QTLQ5YVWVZ0XlOpLfRaIQoiuz3WmTuy7MqVqZ23WAkP5V+jtINPNcBKXCYG6M14iS2P1d0057n2c/Jr7w3///d9v/vl+EbnxcFTNPYrlTMlbvdLubzOWshM9byUD6k6wVoQExZgS/l4ZrP57h58yb+5q//BtuPtjGfz9HpJPAD8g1M0xRlUSAIKCw15igrrZVNqRdyBBRFYUkhl7FeKYUwDBCF0XM3nswIFN8AY/euAqkIscLwHqEt4G2HXfNy3A98uy8C2ulgsbMypraxWwINi4De83zWpBIEnBe0y5XEFnz18/HSXFrTHkj2cQ0oeebB/Qe4e/euFUCvr6/jwf0H+Na3voU0S3F0dGA/sA8fkvHAm2++ib/8y7/ChYsXMJvN8Mabb9D6IOlAKQ2X6foA2+fxdF/XFXxu+uQzQIeCsZOd6D4BIWNUNi+1ach+T6ZSIZP5PgnEyRyD9pYA8PrrrxN8l6ZwXMfu6eXnyvKygYHSQNyJAa0wGAzQ77cWiYYnySzLWNagAWXY7SWCUg6MAbR2YYzC1tYWFBQ2N8/hwoULOH/+PK5du/aveLdf3MvUhsOYazbGaGzocs33WgLJy6LA8fExQe1M5NFaY29vj+BKhhZdz7UTXGMaiz6JE490R4KiLEraZDigMGiPodZWbiT8DWmwpLlyXdJE9np9xHECYJGY1nIzXJcY3TKA9PtkkCCpJmhgByffp7PRNokONWpJklgNuWadquu0BiPlwqRccUScnNNBEKBiFm2RF4SwfMX1QhRLBXAuWkIyC9AEp5itKW4LQRBguDSEAhkZ+AuL69l8jtmMltST8QTz+Rz37t3DJx9/gjAiBlVZlej3+9aX8PDgEL//B79vu+IGRH8XIbDjOgijiOBS10EYBEjYnNp21bxvkr2m47oIwsDCE41qGapra6cAtF0RYHhf6cL3A3S7PY63GdifIZZldm+h2gU37YXoz0mcwHGJbOTyNNsfDOD5voWV6b8bhiAi+AzjvaqVtBMWogLQ4Pbt29jZ2cGDBw/snmZvbx/GGGxubeKv/vKvYEyFXq9nWdjnz5/Hv/sf/h0ePHiAP/qjH8JzPfzgD3+A4ZCg1SDwOU0HdlcuOjKlNDrdLpMo6I5IoRTfYc/zEMURAEDyWYUNKOxm66rDcgOtNdbX1y3ELwhDUZT43vd+D9euXcPKygodFNppmzJuEA1/j82tLeuuQsWYQtWTOGHz6Xa/6rouwiCAo9toMzG21o5Gms5ZnhCSZo5j7F5dsJpKKR5CHqw46KGuiZcwHo+Rc5NkmAkr06R4qboeufBIg1XXNcknysKGUsznc4tyCQO/2yUHNTHMkDNIyIlBEBCvg88fQJ414V6QFjmOY/zFX/yFtaFrM13J37auKxjOc1UKKMvCwrhN02A8HqOsSjvhaq0tUcgaM5jG/m6OdihGDxS4ThpSzWQeANBwXd9Ojx4/w4LG/LpCSV/9AlxKa3hCiGECgzGGWKl1bTVFZVni+PDIMrTmaYrADzCbzjAcDC1MenJygi+//BK/+PwXiOIIUUg+gSsrq1wI0TrdJImdAGVXioZ2PtRhK3ieC9dxrQ9iA4ZrFeCwhpOgJ2U1mp1ul6YVz7c4exzH1loPkH8QGy0IiO34Z3/2Z9jaOr8AFzsLFmXaTj6LRgJN0zANPGQzeHrNhnV1sj/QigwdSPJCPrY1J0687JfWRO7JsgzXr39K0qJHj3B0SDmB49EYQIOiLHDnzh14vo9v/vY3bWj0u+9+He+//z6uX7+O9957D1euXEGWZ0hi8uldZEcLu0/J9AbAkd2kls5bTChazS0RcIh04Xqu7e7F81MkSdIxS/G3k6Cin0OdOa0ryrLE/v4+TRD87CveWQoikiQJut3Oc/tJOYCFCCL7yU6nw0Q5xx5wUszFpUdIHKdOnbL/vtd7ZeYPwAaKU2BzbXfQhpuYxjRsNkHxgYskF+JvVCTK12Qs4Lqu/YwLAUwaKTljpPHOsszCnpK+IZCsNGbkJjVDw1IW1/XgOq6VqEATkUK5Dt565x2c3TyH4fIyHPdfmsVRAc1hmgaeR8lPQNsIyoAA4Dn41aoJGIWUrwFbPmrmjjhcvJXSz32drPHkPVFK0bTN5/RXXS+E3Z3AraaRbodMoQV6XV5ZxvHRMWHyWiPLiP2UZ20C/e7uLsbjMR4/foL9/T3a27C2x3EcPHv6DBcuXoAwCgFgabiEJE5a6yiRcPDB2fAmyXU9e2BVbEygYKD4jdVQcMPQTqeOdqCbBoaX0gIjBIFvBbKOdqwOFAB8P8B//I//M37v9/4NPvzwQ9y8eRMk/M1s0RT2K9BOp1Lg4zii94edkKR7EoZnoSmUWEHB0WIWvmAU/pJfSmnMZhNMJuRZOh6PcXhwiDknY4RRiOFwiM8+/czCsXvP9lFVJfq9HjY3N3H9+nVcvXoVg8EA0+kU66fX7fOnDBFwqOulolFXtUVVAILzG24MxRO0aahoNiUdRt1uj/SOM4LPa1OzaX9r1ycHoHThYvFVlhVSJmbQbqhA27I1MOwIZKRzZzh20QxByBuL+yM5bMMwhO97MDUl+Ni/1xi4joMav+QM47q0l3W0Jfi97Jd4EpvGsF81Ef3kPErT1BIG5d5JkwfAHv7GNHb/PZvNrCG55/lEaqlbvawwXaMoRMCrJuE6AO39DcLAugYNl4Y4PDqkgaasUJYKRV1Bc3lP5yne/fq7xO3ghoqYvTWfoyUTuzw7xQ6GA7RuZ7AMXfraysKsi4W0KisEUchnGP1d7WqAG9LAC+0EK0OKPNe0xqDdvmjSw+CrI7peiMlS0kPKokCeU/HrdjvwPErXbhgzX9S9kSSEbqrrudjZ2cGdO3cwHo8QBnTT4yhC4AdQTHzxXI/ggShuadFcMBRrLGh301pK+UFg942eT/tH6UIchzIrrWk7GxsDkr9JhgujMQmuHe3QpOF5thAD4IehxJ07d6CUwo0bn2NxWW7qGg1P1jWbWYs0RDr6KI656BvLlhQoh7I16TF2HU7Q4LgppZX9gL7M13RKtmvPnj2zuX6O62Bz8xxWVlYs/Li0tIRLly7B8zw8fbqLy5cv4zvf/V1keYaNjXVcunwRQIOiyLGystJCqQsWZWRdJnsb6vSlQauqEg1Phx6jEnJ/5NDyPA+eS8W0tbwrrEeyENIWtXKO4+DMmQ1YRynQnlum3EYmyqaxbGqlFJaWljjRobB+mwCeS8qxWjfPZVSDLPAcMRNRmvXHjt0tWTkWKCTdeUUyAyCfd/qslxxIXteVLZTSVIhpC5oG09nU8jWo6JGJShhGberL5hY/K/RzLBnR9TAej5mYFbYw+wJxS/aWdVVjNpvSOqvXQxzHz2kU0VB2Zm0MHNfB7tNdvP+t9zGfz1BWpSW3AW26kud5Vi7oeZ49S4UXIlIRrclZSBpCKZaydqCvoWGAJkYXSmurKlAceeZoGRgcO53Lcx+wv/ZXXS/ESFEbYn3a7kEpC2/RwxFhNpsBoFBd6YREXnL79m3cuXMX6+un4Xk+R3TVKKsK3R6FhkZRTHsire3uUoJ9taMthu5oB2D9pOu58HjP4nkeexpKYgMVQ1kol4o0lcJUqw1NoHEckwM/6FDpdrtwPQ8PHty3VmUKQJpm+OlPf4b/8l/+N3z22acWXlGKjM4XodfaAOTd2WA4HGBtdY2h6QKGdW5hGCLPC/hMmZbfhxiTvEtqNMN2r0wJ/vmfr2NzcxM/++nPcPbcWZxeP42yoPig+/fv2+SMNE3x7NkzrK+v4/XXX8dvvfdbiMIIs9kEcRyhFWMrdLodjvwiZqNRJOeRg8B2upp1cWh9WR1HW82XTAtKUfxalmfIixxIwS5QDe/JW9KX6xHDj54DQmDyPMeVK5ehlLI6R2ruhMyj4HDztbq2ipWVFWoS2cBayB++7yPLKUFHnKi01lYiQg4vPrQm27MgJKaxox1kGZE3KIKKvHhlqnl10VXVrc8rWN86m80s0aauCU1wPQdpmiHPcsznKcIwpFCHNMXq6iqABrPZHKurq4iTGPOU9pN1VRFpDHQO+L7PlnEtouYtnFnS4EjWo4R2Ly0tsTQtbWFNy3wt8Rf/+18gCilvt+RnhybWGq7r2PPe81xo10Wv20NjGhR1YZ8zgYBl1+m6nmWwEnKhUFRl+wx6wsKl1Qr9TGEQt+xd2VOKkYsgJr/Oo/iFKJayjzTGoCorhFFgE9dFn+g4DsVNcUcku8nRaITtR9tYGg4trON6dIB0OgkaFuZeunTJTmLSlQlUVbPw13UoAVwxM7KuasbAHXg+d/O6FesS+UFZz9iyLOFA2RtRcdoJ6SPB7hg14ijC0tISTk5OFhbUFfYP9vG//uf/TDdQAcpRvNtStrsSIwOlFPr9Hs6eOWunX891YZhw5LouTSpMDojiiLsulgIw/FpVFQxeFcsoCvHBBx9g48wGvvOd34Xrebj+z9fx5PETdLtdnD59Gt1eF59++ikuXryItVMreP211wEF7Dx5gs2tC0ADlFWJw8ND9Ad9S0JQjexaNKyJfSOuVMwm9QgSCsMQaIjB57ku8ryg7yEHW+Cj2+na4F55LmRvJNOAwJyGySKSghNwUVpZW0XcSSxMm2c5/MBHHCfo8M7R4722wGWe7/H0W9nu3A+oOZWdfBRG7V5f04Frn10QAa5kwgYVAtqvClnlZb9kapLGhFivJci8hBqv+WyO2XzOOY/kTkZFxVgUQQ7/qqqwsb5hvx8A+EGAqqL3WzgQcr+znPaWUgBFyyvftyiKlrC4uobP927YBKiGn1tV13C0tkWuLApUdQ3dCBO7nQypGfNx6vRp9Ho9S5y0lnYMwcq5Jz/fNAaqUQvkSllNUd9X1RWTLxv7vpLCwrSfJ0XfU8wLyCf7qxGOF6JYSkWnZS4xUmezuX0jyJGksm9enueYTCbkd/hsD8sry/Smu2Tz1jSE8fuej3ma4vTp0wjDyMKXrkOdkfhbioB7MdOwMXSIhWEI13FQccH0K98GrmpF8hbTGJRpbp0gGhGHKyIKGfZIrFFzgGtFgaWOi92nu3STlYIpcjQsV4FqD5lqoSsDYB+o1ZVVVHUNk2WIWCCeZRk9qKYhA3pe0AtUYgwVStM0gDF2Qn/Zr709iiT6xm99A4eHh/jZ3/09tNZYXVvFfD7Hm2+9iQ8/+BB/8id/gl6vh53dx2iaBsfHxzhz9iw87ng93m8HDFdWrDkritJ+EMWk3HE9OTuIQME+rkLqAgT1IOKY67lojKGiFsXkarUAUy3qxgDYDl+sv+SgEUlJp9NBlmU4c+YMsiyzEwQAO8UQnBdaAlxu2iZT/p1IYaI4QhRHTN13LcTlciZrXuTwPR/pPLWwWppxIsUrUwIAsMzOBlwoS/IYpr1bu2LJs5yNxUsbE+g6rjU+j8II4/EYg8GAmhQrHSPWqdYKSgMwdG6GYYg0Te3PFy2vIFpyfkjy0nQ6RZzEOHXqNB4+fGClclq11opW4uH5UKpExRpOaQh837drpo2NDXu+SRMvpgSLlnbyDKOh2lBzXfB977nzUbvEOm8My1Ua2EIp0hwFRRarPI3KWu6rRHSBLAAAIABJREFUrheiWNLilqe8msbu0WgEMBFBOw6SJMbTp5Q1trOzg5tf3KRuNokBkOF5gwaeQ6niSisEIX2v9fV1C00JBCD7HRnbgyAgH0UuuMYh3L2qaQqN4sjGfNV1jdLQgrpCA5MbptnzjtATz0WCCRxHodYa83QOUxt4Pj2cWpHl2HQ6xfHJMQgBUHa5Lw8bIHCvg26vxy4aBHlVdQ0F2hUopnYrJkJ5LLbViglCbJHXNEQAEnaj+/9CmX4ZruFwiNXVVdz78ks8erSNOI6xtLSEJ0+e4Nq1a/j8xud48803EEURRqMRrl65gtlshlNra1Tc2PbN9VysqlVorUjvy/tAj9mrdU2RStrRKIsSjktyDTEz15qS3l0QOuL7PptLUDft+5RC0e12MRgOcHx0bKONBDlRWqHIi+eKnLiYuK5roaayLK2GmeQvJA0RZqo0qmJaLakMYRQijiIrcwmCEK7roNftIQojJEkHURSyQxVNt3It+nvK665Rc+zTq4uIOspKRYqcZB51TTaMwoLWjsZ0MoXvk+RJch7n6RxLS8soysJK5IC2wSZRfoCiyAV5R9JJkBfUBI3H4wWtZJt8JJMlyeMcDIcDVFWN1dVVHB0dYTaZoi5LFHXBhga8S89zmgJVa6YiJCIqfA7eeedr/DlorJGCFC3ZUwrCKJp4aQy1oxGy7Ev2p1LgJatVzlFhyuZ5zmQ6PGfcUeSFhWV/1fVCFEvPc1HXBuPRBINhH8fHJ+gkCQp+g/I8R56RW0VRFHj06BHCiPD5xW5EdJKu6yGOYmKEOQ7Bq01pu3cazZWFdAUisukNrGV0HAeBH7SaNt+Dx91QbUjvJJME2FzA0Q5qEce6DsMDCkrVbOtHe00FhaTTgZtn1kl/nqaYTiYQ43S9AGUopbCyvIJ+v0+2f9bwV6A3SicRmNXUFaqyBBQQRTF8oVQbDcDAwLB+SSN+tS+yH8qTkxGWlpZwfHyMk5MTvP/++3j69Cl6vR7On6cA8FOn1izN3GUyC5SC67soi9LClXEUIc8LdjypoHUNAolaiy6thN3XgL6Nsm46xEaU/EwQ7MQ4UxRFDMMVUAo2EUSIXSIRWaTCy0EDUKEUlxYx6BDLMwnDTZLERr25rmuhXs8lLTR9K3pNYRjxc0x7LdHcNWjQ1G1OpojtKRi4tqjSqzxLunw/QJal1KRnuW1URKtaliXKhnXXWsN1NJKk8xwzXtjzADUrACzXYvE8oQaFINgoimyKjBRYeUYEtQDorIyjGKOitCbt586dw5PtbRwfn3Bz1FjuhgHzCRsDz6Vs3aIooBwN1/cwHC4hYTKnTJyyH5XmTwq3BJP/8rTruM5z/04Im3K+L8pORCvqOA4c37XIC0Dn/Wwy+8p780IUS8N6LtdzMZ3OGJ7JcHJyAs8lE95Ot4M8z/Hk8RM8ebKDixcvAoDdz4jmpmkauEpjNpuiriv0ul1LCwZgST6LYlzLrmI94yI13vU8hCa0iSNlUaIocmTsxF+yI4RE3TRNA8URWtTF058JBgmsaDZN56ydauB6HpIkQRAESJIEVUUsR6J+8+TLZAjXdVHkhe2+Ap8YXGJ5VjFcQ4vThrR8jPu3+yxlA6K1o3+lBuplu157/TVsb2/j4sULuHPnLpIkoQzA2QyTyQTvvPMO9vf3EIaR3bsNBgOC/BtCRqqq4nvhWlaedjS04d1Js6ADYxICeVa6RJRRbhv0q1qLO0mfJ7N0YpemSBFFEZF38txOjDIFALD3XBx8gDbnUogirkf3XpAWAAtuJ5XNPKw4zSGKoue+t3T0URzR9Ow4CNmSUlIu6DPlQNXESRDfToe7fGr4XqEbAMVMie9pycSVuiJCj+i7xycnMDU1ROLxK4XA933EcYTIyimwsL82SNMUcRTDdR3+HsqGcsdJaXNFZRKVhl2g1UX5hmgxk06Cra3zqOt7GM8mqKoSEjIuA0pZlihqep5q3q32+n1sbGw8p/8UJq9MylIsPc+zz+MiPLwInYrDFZSCY1mtDYxpodW2qRBol9ZkjqOQpumvvTcvxCm5duoU9p6RNjJLU5uO4Xs+iiJHXVe48dkNPHq0jTzPcO3aVSI1CIzoyLRI3XORF9bFRjolo4312XQccqkQWIEcHiorVpVCqflwEgN3rbT1TqzrGuPxhJwiPIenC9k5eBZmcLRGUXKKuEM5mXVVIfCpuJLeEqgdDd/x4QcBHE2CZBHVQhE8I7BqkZbwfe/5bouFuHEU09SbF/ZBakDxPiQhIaOFmh+4OIrRSZLf4N1/Ma7AD8j+Lc0QxTFGJyNsnNnAzpMdfPe738Xh0SHOnTuHNM0wm80QRwHSNKWiWlZ8r1xLICM2XwClq9aNSawIDVHlqXGhAosG1EixHkyBWIpKK3geTXtFUVhdYhD4WBoOiQJf18iy3B6O4ngikGfTAA3D73Vd2sPPD4i4UzBzkAwEyLZRplMhJC3mrBKc1uqSgzAgCDaK7P6yrmhyrCsOdm5IbgIFnJycsFSmspIV0fS97FfJE1tZlZY5amqD2XyGMIwwHo+4OFKElUzkgR/g6PgIp06dsoe+UgTXykBhd4Qh7T5J/+rZ3aAUI0EbjGl9uhezdqWI+j5lTzraQdLr4N3fehd7e/s4ODjA8fERSnb9qVk5IGSi4XCISxcvYmlp2d573/OtJpQMzul5IA15bF9TWZZWUiIKhSiKEAYhmX+wk5U841RQHYa3Hetd7LoO5vPUkomePn2K1dVVJlP96uuFKJajkxGWloYIwgAP7j9EVZaYpynqqsJ8niLPczx6+AiOS64fwpRaFE5Lt2z4MIECwjCwFnCABCETiWjRZd6Ydtdnap7K0BJs5EErUMDnCTLwAwRBTuJeBTZSd+zNCZlmnecZ5Q+CiB0EC9MuVIg2jSE3FRLjkqmBqwh6S9M54ihGt9PladaQmbxeSKYQuAKwE0wYhgw2y+HZOr6gAaq6ROAHCKOIZAcv+RXFEebzOe7evYvV1VWsrCzD1AanT5/GdDpFr9dDnhdWFG7qkgkyOYIwpBxHJng5rtMGAjgOlEiOQFpDOZiky5XGpmF0xDRigMEGFNx1a6VRNYIQuKSjY1grTTO7b3I4FMDuezgQejwaw/dd+3MVCJHQ7BJEB2WNIGjJHVppjtzilBPerYpsyg98q7kDYNm9pm4ZnVVVIZ2ntonLspTWE462aM6vI1a8TFdtyEjciF7a0TA1ODarsESoMIwIpWAbANkpB0GA6XTKUGM7bcp0JnmPBuDGh1ydAj9AGZRIkoQdgKgYpmlqDf0b1RIB4zi2JhVpliJwgueGkDNnzmA0GuHZs6dQ/NqjKEa328XKyoqVDY3HYwyHQ8yZ3St7djmzpei1BFBYREPOQLub18rWBLno89YGc1ChpD06ZQaTI9La2ikYU2M2f8Fh2JPjE2QsuO12Ozg6OkYcxzg5PsF0OsXNL24iL3Ks9FbsG0s1QRiiisfzBko13AkZBAE9WNJly5+FLCPZlFVFuqYgDGBcAwix4jkWquKHgya/PIlhmobgVNNY0wDZR5YNmWJX4kziOtbKyhhj2ZPEiiSNkUyTtalR1/Q6ut0u0fQdh0KcTc3mCh6Jf/lDQNZ1BnUtgm9jpSKOS0YEtTGc/lBaDWgYBq+YiCAXlPv37uP06dOoqgphGOHzz2+g3x+g1+9hNBqhKiue6igpZP/gCAoKKysr6PV7zED2gFrCmsmViohXtEeiTA8Fn51sMnZoajXFHhpQF60qZZs5pduUBaUVPIcOvsAnw+nj42N0Oh3M07lFPxqH2ietFLK8QL/fp8gm12NJEr0eY2hX5jExTQ4mpRQb7ROURhpk3q865NO5vLSEfr9PHXsYoiorFKZAbWomxJUoypIOYGbnkta5gqscaJaf/Lq0h5fpEjarrFKaRtgUDTdVHly3NZaQ3FLJ5pWpD6CiolW7pzNM6qNdpYbrtrpCITlGUWQ1nQLpy+tpyULtmkokd1KQq7rC8vIyOVitr2N1ddX+fGkIkyTBdDpFY+jPMr3KpLgIsS42UhLV9csTrvBVFl+njeiqiGBnjGH9r8v+34TaVcymrWtq6JaXlr/y3rwQxdLzSctTlCWyNCNq/pMn2N3dxfb2YyitsLy8guVl+kXkDX7+Q90wlORxNiHgeT4c7aCsSutHKdCUZAAKyUKcI6IogljtETxRWYqxeLtmml6j53nwPQ+j8Qh5WdmpoixLmupU61gBNs92XRdplqKuDcIotD+/dB0b9GqMQRSGTKzwaFKoDeIkRp7lLOo2VrCrtLb5lGCIWDMbVpjB8nqUUjb0tz8YEJzrvBCPwW/0Ojo6xNWrV7B/QHmC8/kcm1tbmM8oD3B5adkWqk7SaQ0tWCYhEg9jDBwN1kl6diepVMOTJxB2OvxMurwnJINzrcmmrK5qzKYzi4SIB6hmP1ZL2lGwkV2XLl/C/v4+jg4pvX6ezqkAuZQG4jCa0e11n5tkU4ailpaXuJmrkBe5JRmZxkC7DtI8o88TkyOSJEa/30e/37dEHoqKKti/NKWdPsNmvkeTRJJ0eFImadPR4VHbZLy6oNDacZIbl8JkMkWRF+gP+hzq3JJbZHecFzkGg4Gd9gQy1ywjIzSC9NxCfln8PhLj5fs+ut2uJVPS93DsTp523vT8iKxoUeImcKowcUWSIjpzwyYpnU7HEseiKOK9aWkJSDIJin5d/ixkHQnW6HV7CMNwgSVLWcANpOFo4LkUbSdfHycxfTbUgvuU6yEMSQr2VdcLcUqeOr2Gp7skCwmjENuPHmM8HmNnZwdFTmLpKApt8ZEuhSy3yONQOo2yLOF7beiokCUE0qJuhTruf2G91bRsMoF6tVLQ/EEWFxJhyAopKMtzcp4QTRDjn8ZQ+KjjunBlN2kMW/ApSxNvQAekVgradaBAnpAEo1Tw/RgOk3UEfoAiJZ7S2u44Xc/jqB6aRqRoGiF9NLUtlknSsc3CK19OYGVlFY8ePbKOTkEQoCxK9Po9MqmuSgRBaKORPI9ICKQl1DYhYjafQaFBknSQq4K8h7mIqsbAcT3kRQ7P9VA3QBj6fK+M3YUrBRTsVKWMstpL+jtsUydNVl3a+5fEMYwxZOZRldYeTaa2pjHI0sze9zCiSTBJEqytreL27TttRJNurHGHdOxFWbJhf4AoismxCMB8TmSjtEwpLqwBx8pREaSsWraSrCkhw3OJMby8vIT9gwOUr0wJALAnLEPTPmu0Z7Mpup0uFTfXgSn/pf4xDEP7DHquZ5sagMxNtCa4VqY4+f8i66hryoHMvMwWrul0ap9JMVgX4xY5JyWceWlpCZPJBJ1Oxw4xnufZmECBTI+Pj+x/E1vGxelS5CVSxOzKgFEJS0yTYi11gTkjVVmiadq9ve+FrXOZVQ44KOqCLe8MF2TSb87T+VfemxeiWD7ZfoLBcAjf9/B4+wkODg7w8MFDpGlKzjNKIYpi+6YIPEAjdWEhKjIHBqeBtBZhVsgKefOfNyKXN1Y6Dd/3SYLiy0LZRVHkFhZwlYtYxRYe7ZU9uC5ZT5W8nHcdD4bZhjL9NcZQEggXOLHZE+H64o5UflfabYKKI++LoADPcdnZB5aFiYac+PUChAOQ5ZjADZ1uF91Ol6AchmSst+NLfFXsJFMx+cpxHCwtLcFxHAz6AysU9z3fvtdyyDRNjePjY9vhatXYDlmyJpUyiDjeDQxXkQerMEHp/go71eWiKs+7pCo0hqzDiMnoWJNyl3eYxhBcl2apNTyYp/PWMQiwBVlE6+fPn8fBwQFOTk4sNX/R5CDm1x34AZIkRhwnSJIEAOnoPM+3EVtRyMk3lg1es+ib2ZEuey77Ph48fMBpOOq5HNeX+RLbysYYpPMU48kYnU7Hom9y2NN5pizx0Pd8asbryvqbLjLgAVpbea5H8iZ25pFMXtd1rHREiIOBHxDKwKsqx3FQlKSNDALf7u/FO3kwGMAYY03dc+ZYSGGrqgq9Xt+ypCW2TQoj0PoTy7ks5+LinwUu7iRkJxmEgWVfL+p1Zfqk/TqtG0RT7nkeQcENUJYVNtbXcXB4gDMbZ77y3rwQxbLTSVCWBabTGXZ2dqzui3ZDM6yurlpsWsgq8oF2Pfc5obNSoH0fOyvJXlKYXXIjZEKTxbfV6Ag9mllViwQfefjEXSUIaUKsO5zKrTRS0GFU5Dk834eq6ftrR6M0NdnbgXwfxYlCtjVCRvI9H2hgGV9yiaWddHu+x3tVAOC0CklwcBwH4N+vKOlD4AcBh18HVk+nlGLd3Mt9jUYjlhhprG+ce440IU1MbcijM4oiC5Gausb+wQGm0ylBYEGIy5cvoEFrT0adM01/Ne9Q2udIQf5B94gYjkYbNu6nOC3DjMI8z+E69MynWWo/KyRnKa2UoJOQZd14NLa+sIJmCOvR4zxYSbHIsgzz+dwiK+L6EgQBur0uYg7zjaIQcZxA3FnEKs1xHCr2TdOuIZgYNB6NcMjxer7vYW3tFK5cuQJHU5IPXvF7AMCiBMRWb2F+xcYiruO00jH+/CZJQjs705qLK2b+2xWLtVdUz8W7gR3NoKiYiv/rot2eFKtW9032iVEUk1UhWynK14mkRJ4fC8UzG1bY1fLMoAEjJ5WFZhfZsPKeCHoohdL3fQvfF0VpEUYpqJK7WVeVtWgEHB6qWri32+3g4PAAnU6nPYx/xfVCFMsoDvB09wB7e3vYfrSNLM8QBmQAvb6+zvh1YzPUtE1soAdDFr8A7LRHHpscpmot3dqA2kWoQP4sIlwA9t8LXr/4940xCFTAUAftq3yPUsfTKMLxyTHr34p2Wc+HYVEXaKo2RYIkMh5pL+dzFmo7nPHmodPpWMPhqq4RBgFNJVpz/htTRpQif0bWYtamRpWRJ2MSx4jiGP0e4fua38N+r0/OHa9OKnSSBHVtMBgO6DDiYNmDgwMEQWCLR9SLuPvP8fnnn6Pb62JrcwsbGxt2X+RooKobWzyrsgJUg6e7T7G6dgp5nttkB1XTFLloFkCsRh9VXcMLuZHT1NUHPnmANg1FvpHsqUaW5+Q9zIdVr5ejKHKkS8sYj8e8b8qtxET2T77v4+TkBIdHh7h46SIcTiWRhBHH0cS85AYTqpW1UGGk909pBUcT6YdIbQqj0Qi9fg/7e3twHBevvfYairJAlmZ2R1ZWFbYfbePSpYu/uZv/Al0Ca8rk0+10uQkhYo9M8CLvkfdfdtpFkfMzwnpCSFCyZh6Ha2VxUG3oAyEToX0W7YDBEiBpphb/W55ndgqV1y6oX1XScyiQqezNJXA6DCMILFrXNTS/zkXd5CLK5jLBTNzPPN+z03HDpgdU/DwyxmiIC5POUwRhAM/xiezouGgag6qo0O10LfI4HAz5/fjqe/NCFMtnz/YwTzM83X0KpUkcKvEvbeQK2YBVVf2cxhKA7W7IUsnYRa/kZLZ7l5pJGm28lUwNcpnaPKe3lJ8hSRHSkcjER+wtwHHI6kwir6bTGbSeI8tyFGUJH2SjJxOLVpoJH+SmUxYlsxxp11PkOcIoao0OtIbT0A7UNNJBagD0+0gnWNZ0CNac0yYFMo5jhFEEl6cAj/MSZWJ+2S/HcdHphHanVpUVer0ehktDW7zEp/Po6BDjyQRrp9YQBiH29/cxXBoiCiOCRxn68lwPdV1hb28fcUzTWZ7lCKPQRjFVNcmjkiSBEzBjGgZlpSz8ZWqDShEcvxhSLnv1oshBTiyad50Vv94ISVJhaWnJShIgNoporKbMcR0yfudGk6B5jvbSDhyWl0i4gMC/EhMmRZNRZURxjAOOOAv8ACsrqxiNTpDnOT7++GO89833cHJygsFwgCiM8Pbbb6F4tTcHAEynM5sJ2uUAebDhvKOFUd3Yc840ZOPmKAdi7wa05iuiIyd0AlYeJPaHgLLkQDLH13bQiGPaJVLYNDVoi1ahdV3bRl4p0n0KjO8xc1/QOfGUlbNoEVYV7ogU4sWpc3Fq1Q69NtFViiF6baoWaeTfm15b+32I++Ezx4XIdmmWIQxp+pQmEOqrq+ULUSynkymePNnFZDohzeVwyb5peZEjjhNLZRY5iIQwi1WY7EfE4xKgB8x2UYQzoDEGzQL1WdLbFwXXi9RlwdQXl+lSQGUnKotqpZ5nwAa+j4kzRZHn9FA1NAWSywRYNiLaUIcmTinqDu01izy3cKB0cJ6iwluVXCCrCnVFzih1Q+9BwJmenU4HvX7PQofys9tsOdrhvuzXYDigpIXaoNvpQhiqnksONoq1iKPDQziOi62tLdvMAbAHBB1igKkblKbEyegEnuchSbpwHI2dnacwxuD4+Bjnz2+RVZnTYDKZoIuuZcSqsrJwmcNxbPS8tY2esCaFpGYa89zhoBRnsPI+v5N0UJYFAEJRmqixuyjTGERhhDRLW1he0TOvFcF0Wim4IUH4ZUFuL452cHJ8gqSTAAx/1VWFbpfkLE1j0Ov37GT+ztfeQc6H1C9+8Qu8+/V3URSlfbZf9mt5eQnPnj1Dj7WrRV6grAltKJWY8St7rsnERSYG9D20JsKZmLCgodg4x4a/txDtYpatsPmF2Q1FZiZCZPTcBmmWPrdXBNrdaI2WLOl5HtzGtUqEpmnQ7/ctgge0O/SW2KYsV0OzG1sUkf5ZSJuCyEhxFwKnrBVclxs3R1lCWWv1R2uWLM0QhBQKbbkspgF0Syz6VdcL8YROplM8e/oMDYC1U2sAYOnOdSX7v3bB3eppPD4oSFdZFiWuvXYNt764haIsbLdCf1cYiQowaK24GoPAI+hVCqR8f9lhSucmD8iiI4ZMvQIFyzI7jmKUZUkG7GmGNE2RZRmKktiCxhhAK9buOXCZBau0xjBJiLLPh9I8nVsomabGiozTeQnegCKmut0OtEvElDiKESdEzJDYMjJUh7U+m01nmKeptTl7ma90nnLjI5CoAq1TGivvmc1mGA6XLFVfPoS+70E7Dq0PmEnouR5u3LiBK1ev4JOPP8GVK1eQJAm2zm9hOp1ic/MciqLE9vY2wjDEYDjA3v4eZrMZzp8/jyIvSAKlFU6fOoUsy0H2Zi0hQikFA4Ok07E7ec2wbM6sVMnKdFwHOUNhQuMvqxJO5Vj2ttYaSZxA/gWxy3My2mbJEgDs7e1hZWUFcRzj7//+73H+/Hn0ez1MZzN0Ol1oR2N7extXr14lwkUDPLh/H2fOnMXaKn2+tVbIC0q/OD4eQb8imQEAT2rCNCVdLyoyK9ANnYcunz/apcSMVm/eQpcNGjt5VXXryAMNO1UuFisxyRekTUIeRJtYFAXG47GVcsgO/PmEkeetQoWrMZ/P7X5wkT8iryHPKY1GJlFB9aqKdphhGHJ8XMxkMypbi8giOaI1rSymIaJR4BPc7zmEAPocqE5QcAilaGCQs30RZfzl64Uolg8fPEQDA6Vci3cLo7U2ZJdkmKklRWMRhgXIrefS5Ys4f34LH37wIRUy0y6oFzsIayKuHRhtnhPmtno3Y2+a6M1s/qNpaf7yvWWyFd2Rx1Z9SivEcYw0TTGbzZClZEydFzkHrVJkjvwqiou0Bj2wJ/MRTo6P2RWmsZ2idhxUNQmRu90e1tdPk6wlCOCyfk+YbcACM45DV6nIu+gkHUyn03/V+/0iXqJZpVQWwDQ1PNdFVXEsXJUjzzNEMXlW1lWFqNOB47itRpBjgIo8x80vvsD6+jom4wn+8Id/iDwjD9SyLDHo97Gzs4ssz3Dx4kW4noubX9zEpUsXcfr0aQwHA3z55T1sbm6iKApMJlN+jb7NcXVdF1VNWk5JHxkMBhSSrtjrlU3e/ZDYqr5Inni/k8v+0lQ2oFzszqCYTOS5mE6mJPdga7GDgwNsbGygrEpcvnwZjqORFwWTLqi739rcQlEUWFtbw7Nnz/Duu99AlqW4efMWXnvtNbiui8uXLuPul3exurJiG7iX/SJSH8niyG6zRb6EyKgdDaU1qkqga2VlHEopW3Qahmhdx7We1MAvIWOMjNggem4UxQVNnJkW4VL5+izL2ilPtTI0q6fU+rkzUs4da8PZtNFhsqf95Rg5OcMEfqViW1u4mN6fdsKUNZjWZPhe8lAkmcAFIysefzbshMz15tddL0SxTOcpPN+H5/lWS9Q0DbTvAEVBJs5aoapad3gpqFq3ous333oDWZZjeXkZh4eHkOBkUxvUksYBYwk2cIGmbGxaOD1gBlXVWKeRxYXz4m5Pa82JD893aS3D1rX4eZ4TMSMMQ0wnU+RFgZANrYuSJgjaVRGJaf/gAEWeI45jzGczhgxqOC6RLLqdLmuHXCRJAj/wEQQhZR16HsCQsOMwDLHwgErXSA8N0OkSTPuyXyLxAci42gERBzzPRZpm2NnZwblz5xD4PmtlfSZ/MSTK0W2OozGbzZHlOaeTOBYun86mCMMQn3zyT1hZWcag38fx8REePHiI119/HU0DPHn8BI8fP8alixfhui5ORicYj8dYX1/HbD7HZDwBQPZ8k8MJ+ut9AMD+/j4RkKKQyGEuhRL0B30cHh6i0+0ACrh16xa2traQpik63Q5u/D/tvdmPZVd25vedebrzGGNOETknh6TEIlldZA0slVtuqG29SQIs9KsFQ4AAG/4j1PKDANtwv3Q3Cu029FSW0Gq0BUkltVqqslScM5lTRGbMce+NO5959MPae0fSLlY+uZkAzwIKTLIiY7rn7mGt7/t9n36Gy1cuQ1U1LJcLNBpNyLKEpetiPqPQ9DRNMB6fodvpQtd1XLiwSZmGnov+Sh+qouL45BiXLl76Qu6saZnwfR95nrEZlYZXX30V4/EYtm1B1TT0ez34QUDv57Lo92aYNGZhB3pZkpGD0pQKdpgu+HojycwqwnN4C/FMclINUAi9RJblwvMKspYL+wgn89B8lL6uZFInjSfOAOeggOd96s9bjs69lYro4MVxLDbcJEnEjZW/7mlKUYjP3/D44YxvmLIsC0WtaZrUMVNU+AH/PvvYAAAgAElEQVSh8vhGnGd0s86LHIHvw7RMMXIjH+s5yej5n6tAwUILfn69FJulU6kKMgW/hvNbHPfP5EnBIATMVCudWx84jOAf/v5n+OyTz9BoNjCbzUhSndLsKc+pRcV9hXQ7yESMS5qk4tTGiSJxHCOOEzEnBSAkyPzhoBubAiD/wsNE/1/2BcUWJJA6MiVKiud5SFNSZWWsNSGBQNPdTgedTgf/7k//FO+99y4xEA1T3DC5qZf/DviGzckUWZYhTVLohgGV/XsUxahUHJiWCcM04Tg25vMFdnef/ud/0V+2Kp4PIQcgyWSUT4EwCnH16rZ4zkzD/MIcm0PzKe5Hhed7ePPNNxEEISaTCbrdDqIoxuhsBHfp4o033kCl4sD3AxwcHOD27VvwWKbkxUsX4S6XWCyW+OSTT/H2228DBbC7s4Pt7atoXKAYrSePn+D69evIsgyz2YwdGGUcHBzCNE2sr68hY+1afioPwxDbV7fx53/+53jn7XeQJilu374NXdcxX8zRbDXo2YWGVquF//PDP0YQ+Ljzyis4ODjExUuXCLogy2h32mi1mrh37z6KokAQBFhdWWUtuwSGacBdumi32/B9H45Tge3YyNIU/ZU+ojDC0l2iWq3Cdhx47pczOb9OValUkDGSF1kpWD5kliHLUiaSsZAj/4JPkvJpNcRxJJ5LWksp5Jhj6cRYSVGY2BDiY2RFAo9S44KxIqVAAM6WNg0TaZZiuVwKS1Kapv+fUQ7/2rZtU+tYJW7scrnE1e2rePMbb2IymeDP/uzP4Ae+EPLw9ZzSU2yhRNc1ogVRJnHK1jPq3NVrdcTJF50HBbPsOY6DyWSCao10GxJozU6L9BxGw4RpACih50vqyxu0/xmL5m7n80B+IwK4JUIVNzmuDuQzSS619j2fEiMsCw8+fyAM5nRiIXVjwDxmgR8gDCOEUYgoihAEofiz63oIwhDL5RKe58NdLjEZT+B5HoKAoO6u58LzPLF5R1EEd+nCcz24rgvf91nb1cViQeg03/fEx6asX28aJkzTIhFOrYZmo4Fer4cLFy9gY2MDtmPj13/919HtdtGoN0SMl23b0A1dzDUBMLrLeZs1SShX8eT4GGdnZySPbjbFiUwCWGqGjV/6pTe+wlf/5Sje7uJ4rywlT2ySJqjX6zgn6FCLSDd0SoCJY9FirVQqUDUNaZKyYF4iT00mUzgVB4Ef4O7du1CZqOv+/fvImE+zUW9AURWEQYgCwOnpKd5++23IiowLFy7g8PAI1WoVSZLg9OQUcRLDcRxoGt3qbty4jojdZjcvbEKSiEL07OkzdDodJHGCKCQkmmmYODs7w2AwwDvvvA1FkXHt2jX8xZ//BTrtFm7cuI4gCFGv1/Ebv/Eb+O53vovFYoGr21fxp//uT9FqNvH06VNomo61tTUslgvIioxvfvOb0HUda+trqNfqQklpWTab3RP3lCtqHceBaZgwdB39fv8rff1fllJkhUXznSv/+QwTAFPhF1Dlc7FhluUsMJ4+rmB8Xy4AE4Bx1mXjkAtqj0rCMwyWdvM8uEViallVoeg1Djt3HAcVp4JqpQrLss/X7ee6bBxJyp0DaUKAi739PfzhH/4hfvjDH2J1dRUB003wta1er6PiVOA4BL8gS4sqLjaaRkjTarVKKUtZhvl8LoSdHIxw8eJFmKaJVquFZ8+ewnNdmqEzy0wURQhYXmvKBJK/yBnwUtwsgfPFSmVyewDIU+Yvy2LIEsVKRVEkgNa8581bpjtPdpBlGdrtNpI4YUDyjCUnpOJhipMEqqIgy8hzo2mcssJgA2x+xTFhkixRFA271Z2naUvCgykYg4qKQApEWzWKInHTJG+axnyiBYMrMDWjqonBu1NxxMkwDMPzWQVTwknMSylecJbM0u11AdBctlYldu2FCxeQ5RmOj47PDb8JwY5NwxQtkq99SYCM8zlPURDkeblcns+N2IKiKfQ75Gn2YRQiCiNYdgLLsnDlyhUkCQkiOp0ubMfG+OwMly5dgm3bmC/m8Dwf8/kc9XoN3V4PtXoNhm5QnBvzSpomRcM9fPgQ77//PtlWxhM8fvIY/8UPfiA29uFwiI31ddy8eYslPQxwfHyEa9evw3ZsErEZFOFkGiZm8xl+5Qe/gg8/+ACD4RCPHz/Bt771Lr7z3e+h0WzDtm2MJ2NiNB8fo1qt4fKVy4iiCG+9/RZWVlbxyaef4s7tO/jOd74D0zSwt7cvOLHNRhOng1N0e13s7+2jXq8RJ1fLhHI9TihGL46J41mrWV/xA/BylCTLMFQNsUQajTTLkDNYgJRKYi4uVK44vw1SqDYfY0FsmgAFivM1iRSpKWSdAcc5RYpfndgtK2NJOmBWI10i4o4SKWLD4iKeojDFWsdbqnxOCdDaylFyWZah1Wrh8uXL+P73v4+DgwOGjtSEkEfoL1TKeOVENkVRkaYJKk4FAB0AJFlCvV5HFEWYzWaI4ximaeLg8BA7uztk61M1uJ6L6WyK2XSGeqMuWsGSxLI/XzC3fCk2Sy4D1g0dnNMnSRQ0y8UGkiSJuJg0TZHFOXNfFMJYm2c5JFlCGITnIgddZw8bzSQVRYGepF/AIHHFKL+S8xOSxm4JdKKRoGmqUFgxRAlkpqgV8wWFk4ToNGdbtlh8AUpxUFVFnMSKnNBniqJCAqCaJH2OkxiTyRSL+Rx+4OPq9lXRNjANCtcN/ACVShW2bePg4BBgsIVWuwXP9Zhsm37uS5cvwXEcSs9gN26+ETQaza/olX+JihmbFUWBlEuwbQuSLNM8lyXcCHUpe1bG4zNUKlU0m03hURwMBwIB1mg0EIYhTo5P8ODBA7z55i9jPp8hTRPM5nO8//771CFgXRTuXXz48CFef+01atubBo6Pj3H79i381V//Ner1Ot791rtI0wyL5QK+7+Pu63eh6wZa7RZ+8pO/w42bN9FqtzA+G2MwGOB73/seNtY38Ld/97c4PT3Fe+++h+9+57vQVA0//elP8c4338FgMEC9Xsdnn32GW7duYXNjE1tXthAEAWazKf7Jf/lP8NFHH+Gb77yDn/3sZ+h0OjAMYnOapok333wT/+H/+g/QGKZPApFl7t59HZ9//gCLxYIxni1K6slytNttLJdLSJKEldWVr/gBeDmKH55VtnHEcUI2OZm6b8Kjyw7nHIyeM4IPX/cAALJMB8A8R8LGBtwLC4mlD8mKWJuAc3N/UXCyJreXnYeWmxZpKBzHQZ7lcD33ueCJBL7ni0AKPr/mM0xu11AUBdPpFH/0R3+Era0tcfjnm6Is83xgGbqsYuku0Wg0hIWLzzE5U1nXyMZl1Ah9l2U50jRBo9HAeDzGlSuXGbPYwPo6fY+k1ZDY187F+O/L6qXYLGmWyBSmrEWjKJTyMRgMnntB6XXOsow2TjBgLkO38eGxxCwZ1Ps2IEnnZm6ZEUqKgrIlRfo8OwXxYfb57ZE2T95KEKZfZg4WNhJNFxJtvoGjAGSV2nZpQokPeZ4KI7mikEdKYaR827aFjLpgIo1OpwPP86CqKobDIQI/QKfbQb/fp9NUHAFQsLm5gSAIUKvV8I9/9R9jNpvhs08/w5PHT+C6LsMC5rh06RJ6vR5kWcZwOEK324Fllad6urmrQrTF5fJc+KNIMsIwEs9WgQIbm5vCy5ZlGeIkxmK+wOYmCWDyjDyG0+kM73//fQS+jzD00W63EIQBdEMXMye6HchYuktsbmwIsQUK4O7duwjCEJcvXcLJyQkMFqu2s7sDWWLzw2YLf/e3f4et7W2gAOMNE69TYuq/GzduwDAMjMdj/OhHP0IURbj7+l36Pms13Lh+Aw8fPoBjOwiqAS5dvIT/+Df/EdvbV/HBBx/g2rVr+PiTT9BoNBEPh5BkGbs7O+h0upBlCRcuXIBlWvjggw9w69YtSqgochHma5oG43fmrEVtYbFcII5iDE4HWO2tfdWPwVde5+ky1KVK5RRJkgkLDh8PiMOb8C3SLkg3d0X4MPMsRy4xZja7DAgWNnC+kWXsFik0ECnz2PJOC22eXCtSyOfzQWqlE0fW930BLODqWL4J0T8d8TX5zZFfMoiFTSI5LrwpigJLdwnTNDGZTNBsNCEz4Af/ORSV/t1gdLOcBQ7omoF6vY611TWcDgbo9/sEEGGjiOlsCsMwqc0sLIMvOZSAL0irq6s4PDgEChp093pd7O3tCcoMz3iMk5iRbXKB+DoXtBBlxPd9JGkiksKzjLie/X4fSZxCY9FUXECUJtROI/Nsg6HvdPF385xaWbztK3B4kgxJkREmlLxA6tdI/GycU2gapni44zgh4UenRw97moKzHPnHpEkKy6ZN7NneHkzLxLVr14SnM02JDJPnGRaLBSHOqjW02218+smnMAwDa+trgoTEN1wuTAKA7e0t4Z8qi7oMkNjtkanm8oxahxkgUIdcXacqlCSjMAXgcDikOYpto2k0aX6cpIhZpyQMQ7TbHWRZiorjIIoiFtJNi4Kuqjg9PcX21rbANJoGzVw++eQT3Lx5E81mE4pMm2oYhFhfX0e308W9e/ewubkp/JNBEODhw0d47733cHRE4QSVSgWKoqBer+NP/vhP8Nu//duwHRthEDAF+Rlu3ryFf/jZP+Ctt94CAIFOUxQFzWYD+wf7MC0Tx8fH+LVf+zXs7e+h3Wqh1Wrj8ePHUBUV3/3Od3H/8/tot9piUUzTFJcvX8HZ2ZmwBgwGAwERUdTSZwnQMxaGoRAjSmwT5KMYnm3J25ucrcpnlqTpyGhn46ESiipEPJBojMPXsDynrpimapCyFBwJioJoYfxiIYkxUMFUtl+0XfB8SMu0kKSJsNgJnCgkQRXiKlfy9AKu54K6Yik0Nq/lY6csy0TQNN88ZVkieAYLfJYlGZ1OB4PhAJZpIYwiSBJjd7NosV6vC9/30G63yDNsErAlDELkuvYFu82X1UuxWfK2we7uLkzDpIUgInK953mivcp76NSDVkUqR5omwhybh3Qjq1ardNowSUCQsA3VNE0cHh7Btmljq9frbBNUBRidxAe2+OUlSYLhcAjbdmCZpjCoJ3ECSPhCoC6R7BnBQtWg69TbBwq0Wk0cHQ3huS5u3rqJIIjEyYmDgU3TxP7+AXRdw+raKh4/eozNjQ00m01xIuPhrN0uGcOn0ylkiTLiJpOJ6MX3ej14rodQDQWf0TANLOaUZOB5nsiv+7pXXuTIC5kRaNgbR8kFH/h5zxoPreXmat/34Xs+1tbW4PmemONkDFhumRZm0ynW1tcQRyHiOEaj2QT4qV6SkRS0gF24cAFZlqFSoVSPk9MTrK6skI1kNkOr3cJ0NsPu7i7e+sZb4nmdzab4pV/+JSzmC1QrVUzGY9TrdXS7XSRpAhQFtra28Md/8sdYXV3Fa6+9ho2NDRiGgX6/j48++gidTgeu6+Ltt95GvV5HkiS4dJnmrFevXoWu6zg9HaDVasKyLYEsW11dw8OHDwnZx2ZWvuej1WyKOValWhF80TRNRcsuZ4HnRNkqi+sXCgByTrYvfnvkIh9un+AAfG7PUNhtjNtEojilm6DGulyKjILhPCmeqhDh0NxHK74HZgfhKR6cqsOVslnOohHxRZIUt/PpGt0u8dxtVFXJGWAYBvKcOjFRRH5z3dChKobwLOc5eZ7DMIRtO/B9D41GA7qhM/fBuc5FlmVMZ9PnEqlkYTVRFAVJlohRk+u60FRyOjiOI/CPfK770hN8xuMJi2shdJaiqrAtC48fPaYcNUbF5ych8i8S9Yb/N77hEhSADlZRFImrPgqg3+9BkmTcuXNb9MVns5mwmTTqDTgVBycnJ5hNZzBMA51Oh26pSYL9/T3Yto2LFy+i0+2IpIk4igmCnqUUeirJCMIQx8dHqFYc9Po9zOcLLOYuLl28jOFohCCI0Gw2oaoq2u22MJXfvn0L08kUFy5ewHA4xJvfeBOrqyuQZQXjyRiyLOPW7VtI4gSj0QjVWhXtdhu6rmM6m6LRbIhN1XVddLod1GpVBEGINE0QhhEuXrwI3/dgmhZms9kvPE19XYor/5IkEW14SZahKQp0TSNYhSILI3+WpoAkIQpJqt/r9UBtsHOohappCIMA/ZU+bMtCnhUwDAtJksE0LBiGhjiJYOg68lzG3v4+rl29xsKRafHZ399Hp93G1WtX8eGHH2KxWGDzwibefvttSOCRXgWuXbuORw8fYW1tDVevXsWPfvQj/N7v/Z7oKsgyHfAuXbyEwWCAN964i8ViDsuycXh4iJs3b+LZs2e4ceMGFosF5vM5Go0G1tbW8ODBA1y8eBGHh4fY2NjAaDTC7du3cXBwgLX1NQyHAxRFgWq1iiAIcDY+w8bGBhNZHKBaraLf62MymQg+aBxHaLXa8HyPUXy+PHT361RpkopbDj9McIUqv11y8WKSJGKDkCWZICigrhTN9DRYpvnczU8Wc0GOuROdM5bUJNi/zGpCVjQiRWUJm4Uyn2ae08VEUzVhyaPbK8QYjG/qGTsU8VQdVSFij6pq0DUdQRgI2wzZQwhS4NjUgVFVVTBlJQmo1+vwXA8c+5hmqdgLhFUPEhynAl/yWDanLER7hm5ga2sLDz5/wCxWrLX9C9gYL8Vmubq6gsViQScaXWOnIroqE8j5PCH+eS4h/+Xw9ASV0VT4KYyrWvkJA6DWbRwnzPtlo9vtot1uIwxD7B/sI4kjdHs99PodBEGITz75BJIko9lsoF6vCylyGIbia2xsbIgkbkmSMBwOURQF+v0+LMtEFJHyL8szfPzxx7h56yb+0T/6FlM3jnH//n00m03Yto0PPvgQtuNge3sLgwEtQp9//gBZluHSpYtQWERNr99Ds93E7s4urly5LPB3nufBMAzEcUy0fVWD71PuXAE6lUYM0h4EAarVaglSB86TaVjrSpUVMUcvcM7V5L9T3m7P8gyNeoNGBYzHqmqaEO34foDV1VXiFBc5kIN51s5Px6ZpwTQlaOw0zI3ltmPDsW0829vDN978Bv79n/57vPfuu6jV61jMF6JVFYQhLl26hM8ffC7QYq+8+gqqlSo838NsNsViscTa+hpu3ryJDz/6EK+88oo4LAKF2ByXyyUajSbyPKdWaruN9fV1UunmGa5fu46LFy/ipz/9KRaLBfr9PhaLBTSNbr48JODatWs4ODhAp93B0fERWk0KB+bdGoul8+RZzmhXpSIbgIhd46MWvhlwFShvuXJdB99AJVmGIRlMR6GKvFzgORY1IDY23tIUwH+2mXFFLf9nltMIioc/AxAtVtGKhQRd0SFA7gXlZqYZ41WzLFiDzazp2kzfV6/bxmg0Oo/HYgIn3qHgLVvuo+ReUZ5dKRUSLNOC69GNsZBoLELfZ4I0Scjf6brgmZarK6uYzWZ4+vSpYMQmSfIF3/3Pq5disxRZgM/loD3v1+EnEt7/ft50C9CtQFM1oZrln4f34/mDNR6PsVgs4Tg2ms0GAGAymSBNMyTsWt7cWCeYQZFjOBjijTfeEP423udP01QEnfKBNm+NuEuitMxnc6xvrOPw8BCO48CydPzWf/NbuPfpPTQadaxtbCDLUlzZukRYsH4Px0dH+OjDj3D7zm3823/7f+DGjRtYLpdYW1ujAFhNQ55neLq7izCMROrE6WCA9fV1qJqKWr2GVrOJs/EYtWoVi+USlE6ig0c7CVp/mgm+7de9dE1DUQCKRgeeKKaEDgonlinQmy0WBQpkWQrP81gLmy8SEKd03r6XmDqRLy4cs0WCFwuarmPpLpHEKdbW1sWMiN8sqjVS4/7Jn/wxfvVXfxXNZoug55ZFoA1GSaGZ4GUc7O+z1u0qRmcjVKtVhFEERVUwm80wOB2g2+lCURV4rgdFUdDt9VCtVpHnOYIgQBCcoF6v49bNW7h37x6yNIXCcI7T6RR5nuPK1hXsPXsmGKGVShXT2RRFnqPRbOLo6AhhRG3/breL8XhMvtSIDnH1egPu0RGpEIuyBctLlqjDocgKJI1mhIZhIk0JeMHFP/yyQOtSioLNuFVFpf+pLAVJOgcS8Dm4pmoiuYjfBHkoQJZlUDRK+1BUBYZqMAoTawcz6H2hAPnzcz4GKQdIQSsrMmI3FkpZRVFQrVLcGGdcUzdsJkA0iqIAGVleuGdeUVRcungRp4NTRGEkoDEZI+0YzBlgmda5n1Q6Z97mRU44z6JgLVziwlZrVZycnJCSXZLF6IUfcn9evRSbJSeMcLwcQOnVhqELlRK/HXIGa8RwcAAtRhqLtXmeoRiGRFDRNB2GoaPT6WB1dRXz+Vx4DitOBWvra4L+8HR3B7Isodfr4dXXX0UUJuJ7eD7xmzxQFIfV7XZRq9VwdHQkUrvXN9bhez5WV1dx+fJlGIaOjz74EJ1uG4PBAMPhAKZlIvAjWLaFg/0DJq838N6338M33voGVtdWEYUhZtMZjo6O8Mprr+LRw0fo9fui/ZfnOdqdNo4Oj2AYOm7duYWMmeJd16MkCNOEqihwKhWMRqMv8Brplml+lS//S1GyoiCJGX84p2BtztlUZVp8OJBeY5SaWq0mTsAACQ8qlQrAVNKqrNJcSJGZZJ57xSglgmeU6rqOo6MTXLl8BVlGWDigQOD76HW7GI1GeOWVV1Cv1zGZTADQTbfT7sD1XFQqFZEasrm5iX/xv/0L/M7v/LdoNpt48uQJGvUGqtUqsizD3t4e7ty5jb29PVzdvorlcoEkTnB2dobRaISbN29iPp8L0/fmhU3sPdtDq9VCtVqF67qo1+v0rBuGgKrzMPJmq4nAD2BaJvr9PnZ2dnD71m18/PHHJC5qkB/uyc4TSrl/LiOzLAj7XM5AA/xCYFuW8I3zTTKOY+iafn6IKwqxWXACGqf3+LGPJE5gWiZkXUaRnbd1UQBxFItUDx67pmnU5aswUD+J1RJAkpCmibDFccEPt63EcYxgEUCRFXieh/X1dYoWZOklYRTCcSqIowjtdgeTyRiqRGvs6ekptra2MRwNRejA7u4u0Z8YClU3DSwWczi2A9u2oGkaIjWCpmoIGPqU21TESAWSOLByWEKTzdRN0xSH1F90bXgpNktd1+lFZ8Gk5wIZ9QsvPldzJkkimIHiVsRuAXxAyyXLq6ur8BlKLAgCJElC+X4ZqUiJziJjd3cXg8EQuq7gwsULQmVWrVVFlp8sy6g4DjzPh6qquHptGx9/9DE0TcfZGc1pdnd3MZvNkSQxVldX0e93EcchnIoJBDk8z8PmhXX4foA4irC2voZarYY97OGzTz/DlSuXUa3Spjb5dILLly6iUatjbWUFKQqsrKxAliXMiwJb21eQpRkePXgERZKgygoe3nuIfr+Hy5cu4v6DhxT2LMs0PwtDZGkKlSvcJImyEMubJSRAsIlpxiMhimJouiY2uiSnVo3OPk7TyYer6RokWUbOBC+KqtAJXKZEiDRJILMWPT/BOrbD5kckmKjX63RCtgyB85JkWsTa7Q50XUetVsNkMkG9Xhf+xDiKUKlUsXQJP1ZxHPzu7/4uZrMZDg8PEcexCCMnJXWEonDQ6/bwZOcJ2u022q02HMcRKlrelj86OkQc0/utxnBhi8WCMW4pQ7bRaIi2nm0TUi+MQvR6Pezu7sLzPTx79hSVKs3ykyRBHMUwDZNBwzVsXtjEfD7/Sl//l6X4Qq/IChPOkJebazIogk09J5spPH6wELM34JxjLckSDNUQpv80Tc9bvVEMVSMRosaAKVzIpqka2VA0hVJPCuJsW5ZJ467nlKiSKomWcZFTC1at0kHx+rXrDIpOboIsy5EsF7BtC4HvY39/H7VaFUVeYDqdMgEYtU5lSUa1WoWqqjg7O8PJyQk2NjYEWq8QVhlNhJkrisw6OQUUZgeUZRmqriFNEiE248I9roilWWrO3o8/v16K/odl2WIzkiRJ5DAahiHy+XjbVdNUgVziN03DMMR8kmPoFEVBr9dDUQDLpYsoiqjNykQ1YRiKfLWnT58ijmPcvHkDr9+9+xzPtRBzKn7yLQARF7a/f0DqXFVBq93EaDTEcDhErVbFtWvX8Prd11FAgmU7WC58qKoO07QxGU8xn89hmCZQ5BiPh3jw+X1sX93G63dfx9HRMUzDwJ1XbyOOYiHC8ZYu2p0Wur0url7bxmw8Q+D6qFWrWF9fx/b2FtbX1yApMobjM6xvrEHXNTQaDRRMBGUYBs3EWKvCsmwkJcQaPLmB3yDjmCU4PEcg4QkJtHlKwr/I8YEaS7PnyS+KokDX6I1t6IZAjnFBAs3W6VDGjdbJcxFNS9ZpSFOyRHGqUJImzLAeQ2MKatuyIUsy/v7v/wGHhwdwXVekfliWhW63y6T5OlZWVph53EOz0cR0NhWL79HRkRBL8N/JtWvX2I0ZYlbvB2SVMk0TmxubbJHL0Gq2IEHC0fERfd+aTnDqApgv5gjDUPwOtre3UK1WcXpyymanZWlMQc/XO97a5Lg3WZIBti7x9ZI7AThNTFDOZO5XBOk0AqJGyRKNreqNBmrVKnu2qfuR5zl5OdlhmihVvBNF802+PlLcG//6JCDjYp44ihEEPo5PjhHFEQLfp1GRqnxhPsitIqZlQtd11Gt1BEEA27Yxm88QBAEW8wVs28Zrr76GK5cpYEBVeAcyxsbGutgUBemMzWMhUUC2LLF8S9Z+VlUaaaEgi0mW5VAU+eVXw5om5eRxTyKX3idxIjatcxyeymT5bLDNlF5hGEKSJHS7HWQZsQGn0yk0TcXq6gp6vR6hyYJQ3FT5FZxM3U202m1MxmMxZyIvJvWxdV2Hosooigxh6MMyLURhhEqlAsvS8bN/+BBhGOPV117F2toayZknE1TYBt3udnB6cip+5tuv3CGFW5rio48+wre/+x2ELDNQlmTUaw0gpRiaXq+LOI4xn86BvIBTr6DIcpyenEJmM7a1C+s4Pj4BAKysrmAynTKowxCWTS2MjC3EmkbD96Io4Pu+aPV8nYvjE/mfZUkWM8V2LRIAACAASURBVHDKiDyX09PzqZzP1BVFbK6KqghEIaf6qGzOnRc5NFljb/SCUHQsEs7QDTYzSaCq5yK1boe8xjyoN8sz2JaNxXKBJElg27Zozeq6BtM00el20ev2sFgsUKvVkKYJFssFVlZWoSgqxuMzKIqC69dvYDQa4dbNW9jZfQJV1XDnzh3BNs6LHFeuXMF0OsXu7g6KAgLov7G+gePjY+R5juFoiG63g7OzMxwcHkCWaARCimKaG2UJmekVndrctWoNo9GZMNv/olnR16k4UpMSNgioImbYzKoGCahVa1i6S+HLdj0PMrtoaKomdB8c86kzX7llWkIg83waDN9MRsMRavWamMFzFIxlWdQt0WjLUJmH2PVcxijWUBS2+LqWZZPVqdVi2McO7t27B1VRacQhS7h27RpGoxGiKIKmaWg1W9hb7gMF0Gg0UHEqWN9Yx7Nnz0Rber5Y0HzVkERr9+HDR2KWSzF0qYj4kqXz6DD+vfGkE03TkaYJA4vQ6I9fIn5evRSbJT0YGtvdFdFX5tmQPGjZMOiUWhQ5EomG4IpCIam2bcN1XaQpeR1b7Rb6K32EAUHRTcvEbDoTn6/f78M0TcxmlGRfgDydpFQsWIsXMC0LpqmJzTUIQtiWBR5UPZtOKZtvextBEGJrawvNVoM9xIBtO6jVG5CkHBubG4jCEJPJFNPJFLVaDZBkNOoNrHR72N8/gAIZK/01ZofJRRZmEITY3NyA53nQJBVzbw7btpBLwIULmzgbjanlqih4/PAx+it9VGtVWJaF5WIB07JQ73ZweHBEirMsP4ccl+MiYS9KWTwRVyESeIC8W1mefWFUkBUpJGY54RzXPKcIJL6I8VxVgIhRz5vIdU1DmIUAZCGw4Nl8UUwtziRJ0Gg2aATBNuQ4joUiN03JrpQkMYajES5evCi6EVx4BkjIswzj8Rhra2tI0xSTyQT7B/uoOBWMzkYi1Pp0cErQ9SjCxsaGyM9stdrY26c5ZxCQIXxjY4M80WGIxWJJNCyJNAjxIsb21hYePnpEQPokga5r6Pa6OBudUSqJ69KCHsfQWbD01714Zq2s0HPF55ayJCNBIpStHgtmoK6EgmqVbv78Rkjz8JQ5CNj6yjiwWZpR1JtEWMeUHcziJEGlUvnCIdqyLChs7eMKVNu2kSYpkjQ5dylAEtAWd+lC0zSsrPRxdjbG9es3CHFYEKB96S7ZM5sIFXQQBGIuznMsZ/MZJDaykGQJ1UoVw9GQPdsGWe6mU0RRyILJJSRJjGq1hgVr63PdAU9jev53miU00+TUn6LIoWj6l700L0kb1rYANki2LEu0ZDkSyzAMQX+gHEc63VI+mvGFCK1ms4GtrSswDZMptxTU6jW4S1JlSZIE27EpqoqpSW3HhqEb9PVsC1kO5Dmgahqq1SpWV1dg2zZqtRqqlQqarSaubF0i+kkUQVF0OI4D27ZhWSaWiyVGozNadOUCkpQzUQ317dutJqq2g4Nn+5iNJ9hYW4ckSahWq9B1E3meYrmcQVHohkyHAWItVirUhpvMZlB1DWvrqxiNRqg36KQ+GU9EWOrB/iGKnL/ZJAwHQ3pIWAuEt/7KvZLOCzTP4xxLmd0oz2HUgh8MnvquImctsCzPxFwIAnSfE9JOkqBplIOpMOUfsY5DsbhxOoos04xKlmUo7HMYhsG8dxKSOBYAa8/zUOQFM/inIv2dDn8FC45eUnuYzawfPPgcR0eHmC/mIgJpOBwKVFkSUzC5ygKIJ5MJeCxevVaHptHiMhwOsVwukLHDaRD4qNWqkBUJQeij0+7g5OQUlmkJD97W1jZ542QJy8VSaBUkSRaG+K978a4aN8vzZCVuF+F+4DiOqRXLOl/nHZBCxB2apkleXJ5ElNLmFAS0uays9FGtVKEbOntOgWazKUAassxpYQkURRX2PICeb03T2SZzbq1a6a8IDya17CXs7u5gNBqh02lDkRXcff0uatWaECeZBqWY+L6PxXKJ/soKJFlCo95Aq0XfTxInaLZaACQ4tgNZkuhGqmrkz2TvQz5iq1arZOWS5XMtAvOrcjU6h9LwjZ6rgb+sXoqbpcpeCJ1Bv9M0hWKfR9VIMkGZkziBXBTQLA2KTDdKTSOavC7LWK+vQ1aod25ZlrhuR1GISqUCx7EAFHAqNVIWjsdYX1/DfLGAyX5ZiiKL05XMTmTTKd0+l0sXjUYNUZhA00xMJ6cwDBW+72PJyBCQJDSbDew+eYpWq444DCBX6pAlin3KWWpFlueo1Ks0c7IqWC59tNsdxHGEJIngui5mC5prNpp1uEsXikItPt8P0O60IEkyppMZJEnC4HSARr0ORVUxn89xfHwiWhB5nsP1PGZWphuSLBGPNIrisg0L0OLz3AwoL8hrJssyNIX8vHxuKGTuKCDJ59abMAygsHmKzHia52HmhMfjGa080UZjHE1VVQl7iBxByLoXzAtnW7ZAMfLOh+u5cGwHYRggCAJ0u11GRsnF17AtG0mSYMbk+XxRbDaaODk5QaVREQK6xXwh/nxh8wKOj4+xWCzgOA77WTI4jonFYiZsBWkaY+m6qDg2gsBnkPgUWZZiGETQGI+YhxjMZhMEgcdaixl8P2FzXmKRliUCPwQ0QJaJtqNrOoIsYN0O9dwlIMnIZfpYfiCXFRUa645ousbaugoMQxUCmCzLBeOYjxkURUUYkWUjSRIoqSJElgAQ+MQzLnIaIWxubuKDDz6Apqliw6SLCAkULcvGt771LRwdHmFjcwMffPAB/MDH7Tu3ocgK9g/2iQHuOLh48YI4RP7tf/pbbF7YRKfTweHhIWyHmNnjszPomgbTNBDHCTY3N3D//n0UgCBhxTGhRvM8ZaOBMRRFZa1pWSSocE/8bDYDGMCGezy/rF6KzVLXNRIysM1NnMp1HZAk6OzUqSgyTMuCqiqwLBvTyQSqRmg7mlkCFYeQdBSmS4njqkabnu/50Bnr8tHDR9i+uoUoIvm1rLAFTTPhui4818Wt27fg+wF8L4RuFKhW64jjDE6lwk7dMzTqVVSrdVzY2EASJzh4to+1tTX0e134oYtWqwk/8FAUgKypgsYv5xI810OzpePk5Bhra2s4OjlCzNpuVtXG2WgESDkGJyH6qys4OjyCZVH+5Ww6g6pp8FwXOpvzhlGE6ckpKhUHOYCU4bIi1h7UVAVFkSHPUlTrNSyXOTRNhVRulpDYopQkCRRVgQpVsDn5zJxbhiSQWjbhQghIbA5ETE7TpK4GAOGXNJjlSGUet4LdIrOMzOA8tR0gFV+W52QXyCD+WxQR2J2LO/j8Pooi+L6H0WiIdrtNmxubR2mahk6nA4AM777nwddJEGdZFsbjMWH8WHK8bug4PDwUsA1FJs9dGHooAOqkyBTb5Pk+VEVDkqTY2rqGnZ0dtNtdTKYTxHGI5dLH9vY2FEXG4dERM9HTnOnVV1+FbTv4yU9/gmq1ivg5nvLXufjBi9sceJszCENwbqvwoLM0HEWWxXzYNh0xi1TYzZCHR1NHRCfAQXIeYcgh7Px2xWem7VabcHrsNVM1lUVoOcJN8Pprr8PzPQwHQziOA8/z0Ol00Gq1MJvNKH7LtHB8fIL33ntPiINOT0+x82QHlm1R/iq7LQ8GA9QbdfR6fZimgZOTU9i2Bcdx0Gq1YFkWG7eRPc4yLaEapt+FClmhdKfZdEptVwZFkHN2GGZcZ1mWkRYp/bzsAPt8gMb/u16KzdI06cbHZb1hGEKRzwHoNC+MWV+aps5pmqBSrUJVFfh+wHxFCSS5APICFcdhGygJN4IghKqpzOeV4tLlS9B1HUsW0lupOshzSnw3DIOAvGFE7NRaDUEQYDqZwrItxLEK23HQbrfhLhfodnuQJaKhbGxs0Iuma0Ck4cmTPdRqNTgVE+54AV03aDPLc/RW+rBME5ZtI5cYH1FTkSYRgiBgWKcaJFnD8HQgfJVZljKYcgxVoxd9ymanjUYdMVOb5UWOPEsggTZK0zIobzMIMF8shO2hLAhBj6IoQsigyAoJABgNhfvEINHGxd9gzxOm6PVLhDgIIlAXQqwmSzJykMcSRQHZoLabbVvwPBemaUGWJaRZBllSEcURs/yokGUgDHx2oCSYxtXtLXz88UcwTR15ngrUF1/oeNtJURU0my0s3SWKvECz0cRiuSD+pmVBlg0mx0/RaHThugvUanXUahVkOdFQOCPU9zxGKirQ7nQQhiG63Q4GwyFZC1QVzZUmBsMBTMNkUVMpwFquT5/uM7VxgWajjSAMvupH4KWonCHg+EbIuz4Sm7epukkRhpAgKecDlCIlywfYTVFVVfH3yWdIf58sJVwhKkPRFZHqxA+E3LoXRiEajQYm44nQj/R6faysrOD09BSDwSnCMIJtWXjnnXfw5MkTyqJUVCRJjHq9jqe7T+HYDtI0wU9+8hOsrKwiDAIcHByg2WoiyzL87Gc/w8bGBmbTGQHWHQeKIpOALMuQxAl8ycd8MYdpmmwWTyknsiJDk2lmqqgqfN9DAa7QBTTdYHD5nHXUSMzD29uSLEOWGIu2yH4ho/il2Cz5i2qotNDkWY5ur4vpZIpqrYbpZMLUqJT5mKapuHr7fgBVUWCYBqq1Kk5Pjlk7ShKkfF2zoagKgiCAqipQNR1xlGB8NoasKMiQwvN8yJIkYAU8fVvXySRcqThYW1/DydExJpOJSPFQNRWTyRRBSGGnbkBUlEKiDb2/0kccE0Ba13U0Ww24rofFfI48LxCFIVRNw8pKF81WFYf7x8zcm6NabcDzQuR5AEgFJOTw3AXdlhUVaZbBMk0kcczaqjlsx0FwdiZQbXmek+E+SZAsYmZ1UYUPNU2S0mYJnIP6ZWrFy0wxWDChDl+00jSB8hyWTlUVARLg4d1g6kWaR8rM7Fw893rQ56c8VVnwaJcu0Z/SJIUkUWxRkiQwVDrQcD6oYZjPzThlHBwesO6ABF034PkeDEOnG2xWUK6hXCAIPISBB1lRoCoKnu2RwnV1tQ9N1XBycgzyfNYwm03hOBUEgYfBcCBmVRzppxsG+XMVmUKr4xjz+QKWSX5m8rDVsFx6CMMYqqJhNlug2+lhNp8hz3NCPDoUNB2F5c0SAEMOumJGyZ89HjeIgpJwTJM6YJqu072QCVj4PwGei0mvG7frqBJBxDkSTgJ7Dtl4Zn1jHYPBUIwANjY20Kg3EIQhTk9PkGUpBqenODw8EArUxXKB/YMDzOdzzOZzzGczOJUKTIMOX3t7e+TJlGVMJmOy8VUcBH4AwyRC0NERbYyz+QzVahWTyQTLpQvbtuFUHNy5fYfSc0ZntGZnKfJMgiTJME2yUW1sbGB/fw9h6AvgAndN2JYthGlRHJ1bbhiwgB87eMv559VLsVlapgmeTcmVcYHnEzkiy2A7jqDO6IaO2EuQRREajTqWSxe1eg2+5+Hk+JT53HQUUDCfz0mS3KbEjiiKsFyQUivLaV7U63fhLl1EUUw3WvaA6qYuaBRFTkrG3Z3dc48RGw7PZgWSNBOiDIe1SI+PTrCy2meeOQmLhQcU1BtPYjKpV6sVNJtNnJ6c4PDwEADzWTG81Ww2o8xNGQAKdHpdnBydIkly5Cp9zcXCY55kovRPxmeIY6LyhGGIomBQYzYnJaW+hCKXYDsV+J4vbDlf5xIJ83kOsFkvWFYfB/XTBnUe7s0T5oviXIXIb44kWqGke1VWxedTFPn883MUrQCv02FR1cg7HIUhIOMcgSfJLD1CYokMvC1M4g7bsaEw9Bcd+kjlnWUUAWdbhHmczeZQNRpLQCowGU+hKApWVtYxGg0RBEQyWiyWODw8hmEaQEGtPVlSIMkKiiSHomhQJOBsfAZVUQW8gN9QRiMSlNm2jfl8hjfeeBOffvoJNJlEHbUKefzkQoLWKjscACi4Hgw+wMAESZKgUqkiCkMiPxU5ut0uPN9DnmUwTBNxEkOR1fNNkAlgcqkQim65kJGmJHrkViRZknH3jbuIogjPnj3DaHQGDmlVFAWff/45KhVKRLp8+TI8z0Oaprh79w1UHAf7+/tYXVvF2dkZLl+5jJWVVYB1MkdnQ2R5hoODA+RZzubyCSpOBZVqFVEUYjwe4/Lly2g1W/jss89gmiaDF2RwbBuqpiIMQvzN3/wN2u026o060jQjbUeaoMhzrPRXsPt0F3t7zxgIXoWq6kwARb+HOIlQgASTRE4jfQF/E0qSBFVSfyEn+6XYLOts08uzDIqq0vU8TeH71JqhPDx688dxgiIvUKk4mM8ph7HIc6RpxmTSOVzXQ63WwJWtyxicDmHoJnZ3aaODDBimDtcNEUcRTk8G0DWNBrwFpY3rpglJKpjKLEeuAgt3KTBUpmkgCmNMJhMW5+UgiWPESYLBySlUTRMJ8tPpDCgKNJtNhGEIVSVFGbeh5LmHLKdgXNMwkOU5Yz0WqFYrUBQCVCuKguHgTMi7bZsSQyS5oJslU7zGcUQtPGZ4D0IShvT6LYzPxsJjxRVmsiyLOLSve9HNjcg8KIA8TwRHlysO0zSBycRjiqIgismry43caZo+F1V1HqtEOagRwBYvVaVNNi+YZF1RIEuALANFnlHbTAJ0Fv12roSkrkNR5CKfMk1TJEkscH00m9IQhpQT2ag34Pke0jRHmhaQZRVhENHHRCEkdiMlcEcF9XodJycnSLOMeaAlxgaVqQUo0YhkdXUV0+kUnueh4lTgea5Iu6lUHBhGA4ZuoFGvQ9U0PN3dxd7Tp/iLv/gLfPbZPezs7CAMI2xvb+HmzRv4V//qh1/hq/9yVK1eQxRFuHTxEu7dv8eUnNn5WIkpNw8PD2EapvBLmoYpGMScrKNqGgomMKMbqkJGfEBwY6Mwws6THdEu73a7wsNLhKgYs3SGtbU19Lo9nElnlEkqSdA1A9euXkecRlhbW8Pe3j729vZw+eJFeF4Ap+JQXKCi4snOE7TaLRFld3BwgPWNDayuruLJkyeCMcxJbJyqU6/XkaYp5vM5ojhGMBiSY0LXyYusyNjZ3QEA8Tm4XYsYzhnrEtF66Ng2/MCnmzvDUfJOUBqnL//NkjxaNPNp1+uYTaeoN+piZhlHEWt9SXBdF+12C612GyfHJzBNg8n9CQxt2QY7FVOb0zB0kkrrpAR0HBvLpQvDsNBoNDAaDum/BS5USYJhalBkIAhDZCnlCtbaTUzHM0Q+SeoJPk4ttDiKRNIHigJOpULtXkXBcDjChYsXcHRwCM/1kRcZAj+AxhYmKUlQrVagMhNxARm+H4iwU98PWCSUgjRJkKYJDENFlsXwPHqTKIqCXFHY0FpFkdFiqmkyUhY6rKoKJuOJaD1AAlbXVtFsNvAHv/8/4d/87/8G//XJr3/Vj8FXWnmRQ5NUgKXMi+BagQ2TRfsxiWPhGeNxXbzdapiGWLA4qoz3xYqCNiv+hiQakMQ2ZBZ7xDoocRzDNC2ayzBDuqQCCRMZqZqGIAjBhRnCZxYnYp5DN8JCBBUoioLhkBabjIlAwL4ux0gCYNgwhWxQrivarLJMkIxqpQrLNKEbBhzbwf3P7+Ov//qv8eMf/yXu3/8cT3efQtd1XLl8GTdu3MAPfvADvPLKK9ja2sLm5ibee+/bIi3IMAw8fvwYchn+DABCLHY6OMWbv/wmHjx8ANd1MRwNhTMABW12RQGmkCdONk8OSdKEhTlnJFxjEPPZdIasIBxdu91GFEUk3opCFoNYYDgaAqAIrHq9jv39fSRJivGYAid45+Dw6BCypKBWq+KTTz7Bq6++CqDAcrnEcDiEqqoYTymSLY5jZGkGTdMRxTPmYc5wdHjIYrpUzKYzyIqCZrMJ1yXhTqvdotFVs4m//7//Hs1eE4ZhYDAcPDciYfFjrOWaxAn6vT6GoxEADntnhwdFQRgGQs3OY/ZkSRLvs5deDZtlGVrtFsIgwGAwQK1Wg+t66Pd7ODig1I7+Sg97T/cYyDyj3EuFSCFJnMCpOCicAp7rwrapjck5l+7ShSRJqNcbSJIUSZKj16tjNptjfXOD5jtSjiiM4FjEft3avorlYokgCDEdkSS+UnUQhhGSJMXa2irOzsbs9qfg+PiEGcvPI2Y0RcbJ8QniJKENXZORZTnyMCRsnqpiMV+wwXSEXq8O1/XgugvBa1VUGTIb5Oe5BN8LaHYBIC8KNq9lQIaEZrkKU0gqqoLLlzfh+T7+l//5f8UPf/iv8fu//8/xT/+rf4rZbIYPfvYh3v/++/je+9/7al74l6g4hIJj3/gpXdWovag8B6mWZBlKQW9SVVVFGjuH7Esytb45q5j2xgKKosGybEYsIdELT5KRFRX0PiV8YpJmsGwVElKEQUQUpiyDbVNQNwcgaLpGG7CsooAETTOQFwUubF7E6Gwk5rBZRj+jpmkCJ1mr1aCpGuaLOe7fv48f//jHuHfvHh4+fIT79+6h0+3iypUr+Pa338WdO6/gldt3sL29jSgmnFoQBJjNSPT25jfexO07t1lSfYg8y6DrNI/SDR1e6OOjTz5Cq9Umm8lige2tbQxHI3T6XRweHH6Fr/7LU6qqor+ygp2dJ/jwww9JiMO8lZIk4erVqzg+PhYZj4EfQFGovco/xjLJq25aFnzfO7fjsQNJlpMncX19HeMxZeQapoFLly7g4cNH6HZ7aDaaWFldAQrg+PgYRZHDdZeUfqJrOD46Rr/XB1BBr9uD71E3IQwD+GEAb+li6+o24jjGeDyBpmoIQ0Ik7uzs4PqNGwjDgDbiszPhg+Tf2507NKM8PDjE3rM91Op1zOZzALTZx1EsIhpb7SZOT05RoECtVsN4MkGencdt0fw1R8Buy3nG4smY/zRj7+XiuaSrn/va/P//8r+4VEZ98FyP9ZMJHjCZTJGlFD/jueSdXC6XCIKAtStI7epUKqhUHYxGZ2i1mphNJ6g3Gmi123i68xRpliHPMgRBxFJINBweHsEyTcymU/IXxTFMBkCIohjz2Ryz2Rz9lR6WC8D3MvRX+nj2dA+yLCFOEsymU9j2Gk5PB7Btm1pr4MZ1MFsKzZtSqUDNriDPqeevyhJb8DKokoIiT3F6cgSVLWYh2+jlTIGkKwjZrFNRFMiKgu2rW3j88DEKFAj8gNkJFDSbDew8foJ/+S//Ne7fv48/+IN/jtuv3sY/+2e/jd/8zd9AHMe4d+8eAMCyTEiyXIorAOHb1XWNtWNTQg/KigjGlRl9h894+ebJ44TyPIeh60hYi1aSJUgFv1UWJKxirVk+Ay0Y4EBi88i8yKFChSLRbVJliK4ojMg7zIHVoJutxkYInXYbqqqx51CFYzuYz+b4/NE9/OVf/iUePXqMx4+f4NGjR9jc3MClK5fx7rvv4tXXXsWNGzfw1ltv4Y27byCMQoEKo/cgKSF5W+6z+/eIYpTlRBhKU9iWLYgvfuBjY30DOztP0Gg0EYQB3ZSjCLKs4Pj4GC02Zjk5PRGRdjwy7+teSZpgMDglUY9JrdIkSQQQ49mzZ8w8nwNIBbyCDk4K8xFKcF0PjmOT4IdFxxV5gV6/j8HglA5NukFt8kYD47MxoTFNU6TOPH7yGLdu3WJhzIQKXVtbR7PZQLPZwuNHj/Dw4UM0mjW02k0sBwMKbHYcnA3PMD4b4+T0hGauToWRm0hVy2+VOaOUceHY5uYmPv30U3z88ccwLWotG6aBWq2KJI6RZiks20IYRfRnyyIKkAQRpcf/TpKw90qRE1hdUVlQwjmhDTiHP0hcIPUl9VJslo1WE9PJFCurK5hNpyhYm8EwdDgsHmY4GMJgODfLsjAaDFGpVhHHMRaLBWSZwnOnkymyrBCtTR6KXK3V0Gw14TjEL3z2dA+yI8OybJydndEvNcuBjFpgUUiUi/HZBJZlwnYcHB4ciVnf2XCEaq2G5dLFla3LOD0+RZblWC4o/aG/0sd8NkPB0lLyJMPZaAzd0KAqJN3Os4QWopSlUoBac4uFy/x7soAg5xmpKJvNBiaTCT779B40VUW/14Pnevid3/nvAAC/9Vu/iV/5wffxP/yP/z3iOIGiyHi6+wySLAlJOEBiKk3X4Xs+SpslGNrQhCzJiKKQCbmopcVP5GmSChGQyn6PhGIkw7NUSIB0nqfHFYW808DnKPzzUQtJpjdxkQtrCvduqrIC0zTQcRwRvmvoBh4/eYzHjx7jr/7qr/Dw4UM8ffoUpyen2NjcwJUrV3Dj5g18+71v486t23j//V/BO+98k4J4sxyL5ZzyW5OECFRBCNMw8ejRIwaeBpqNJiRJgu/7AEgYJyuyaMVqqgo/DjCZTtBuNeEzSDb/HJPpBHfuvIJPPv0E169dw9nZGQyd8lQ73RY810VRZNi6chmz2UxEMJVFm5ppmDBNgt8PBgNi7UYhZEnG2to6Tk9PANBzRVKcgjX6SckdxQl0jVSvBQpqy+oaC5WOYNsORsMRnj17hizLsL6+Ac/zEfguOQ6yQozFer0eAj/AcrmE53kIAp9EYtMprl+/jtOTEyzcBXaePKEUGV2HrmvYvrqFJKUDZ6fTFToJ16Uu33wxF4fNzQubogsyOB2g0+5A02nMYNmWwDemLNVnOp1BkWUxtx8Oh+BBB3GcMH8yzf2prSMDyAWdh5TqAMADsFVkWUT+0+zLBT5SGfxbVllllVVWWb+4yjtFWWWVVVZZZb2gys2yrLLKKqussl5Q5WZZVllllVVWWS+ocrMsq6yyyiqrrBdUuVmWVVZZZZVV1guq3CzLKqusssoq6wVVbpZllVVWWWWV9YIqN8uyyiqrrLLKekGVm2VZZZVVVlllvaDKzbKsssoqq6yyXlDlZllWWWWVVVZZL6hysyyrrLLKKqusF1S5WZZVVllllVXWC6rcLMsqq6yyyirrBVVulmWVVVZZZZX1gio3y7LKKqusssp6QZWbZVlllVVWWWW9oMrNsqyyyiqrrLJeUOVmWVZZZZVVVlkvqHKzLKusssoqq6wXVLlZllVWWWWVVdYLqtwsyyqrrLLKXIypAAAAABZJREFUKusFVW6WZZVVVllllfWC+n8AgIVgquwC0dIAAAAASUVORK5CYII=\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "data.show_batch(rows=3, figsize=(7,6))" ] }, { "cell_type": "code", "execution_count": 21, "metadata": {}, "outputs": [], "source": [ "learn = create_cnn(data, models.resnet50,metrics=error_rate)" ] }, { "cell_type": "code", "execution_count": 50, "metadata": { "scrolled": false }, "outputs": [ { "data": { "text/plain": [ "Sequential(\n", " (0): Sequential(\n", " (0): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)\n", " (1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (2): ReLU(inplace)\n", " (3): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)\n", " (4): Sequential(\n", " (0): Bottleneck(\n", " (conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace)\n", " (downsample): Sequential(\n", " (0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " )\n", " )\n", " (1): Bottleneck(\n", " (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace)\n", " )\n", " (2): Bottleneck(\n", " (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace)\n", " )\n", " )\n", " (5): Sequential(\n", " (0): Bottleneck(\n", " (conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace)\n", " (downsample): Sequential(\n", " (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)\n", " (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " )\n", " )\n", " (1): Bottleneck(\n", " (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace)\n", " )\n", " (2): Bottleneck(\n", " (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace)\n", " )\n", " (3): Bottleneck(\n", " (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace)\n", " )\n", " )\n", " (6): Sequential(\n", " (0): Bottleneck(\n", " (conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace)\n", " (downsample): Sequential(\n", " (0): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False)\n", " (1): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " )\n", " )\n", " (1): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace)\n", " )\n", " (2): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace)\n", " )\n", " (3): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace)\n", " )\n", " (4): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace)\n", " )\n", " (5): Bottleneck(\n", " (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace)\n", " )\n", " )\n", " (7): Sequential(\n", " (0): Bottleneck(\n", " (conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace)\n", " (downsample): Sequential(\n", " (0): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False)\n", " (1): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " )\n", " )\n", " (1): Bottleneck(\n", " (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace)\n", " )\n", " (2): Bottleneck(\n", " (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", " (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n", " (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (relu): ReLU(inplace)\n", " )\n", " )\n", " )\n", " (1): Sequential(\n", " (0): AdaptiveConcatPool2d(\n", " (ap): AdaptiveAvgPool2d(output_size=1)\n", " (mp): AdaptiveMaxPool2d(output_size=1)\n", " )\n", " (1): Flatten()\n", " (2): BatchNorm1d(4096, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (3): Dropout(p=0.25)\n", " (4): Linear(in_features=4096, out_features=512, bias=True)\n", " (5): ReLU(inplace)\n", " (6): BatchNorm1d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n", " (7): Dropout(p=0.5)\n", " (8): Linear(in_features=512, out_features=5, bias=True)\n", " )\n", ")" ] }, "execution_count": 50, "metadata": {}, "output_type": "execute_result" } ], "source": [ "learn.model" ] }, { "cell_type": "code", "execution_count": 22, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "
\n", " \n", " \n", " 0.00% [0/4 00:00<00:00]\n", "
\n", " \n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_losserror_rate
\n", "\n", "\n", "
\n", " \n", " \n", " Interrupted\n", "
\n", " " ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "ename": "RuntimeError", "evalue": "Expected object of scalar type Long but got scalar type Float for argument #2 'other'", "output_type": "error", "traceback": [ "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", "\u001b[0;31mRuntimeError\u001b[0m Traceback (most recent call last)", "\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mlearn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit_one_cycle\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m4\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m", "\u001b[0;32m~/anaconda3/lib/python3.7/site-packages/fastai/train.py\u001b[0m in \u001b[0;36mfit_one_cycle\u001b[0;34m(learn, cyc_len, max_lr, moms, div_factor, pct_start, wd, callbacks, tot_epochs, start_epoch)\u001b[0m\n\u001b[1;32m 20\u001b[0m callbacks.append(OneCycleScheduler(learn, max_lr, moms=moms, div_factor=div_factor, pct_start=pct_start, tot_epochs=tot_epochs, \n\u001b[1;32m 21\u001b[0m start_epoch=start_epoch))\n\u001b[0;32m---> 22\u001b[0;31m \u001b[0mlearn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcyc_len\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmax_lr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mwd\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mwd\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcallbacks\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcallbacks\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 23\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 24\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mlr_find\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlearn\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mLearner\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstart_lr\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mFloats\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m1e-7\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mend_lr\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mFloats\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m10\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnum_it\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mint\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m100\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstop_div\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mbool\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mwd\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mfloat\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m~/anaconda3/lib/python3.7/site-packages/fastai/basic_train.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, epochs, lr, wd, callbacks)\u001b[0m\n\u001b[1;32m 176\u001b[0m \u001b[0mcallbacks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mcb\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mcb\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcallback_fns\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mlistify\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcallbacks\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 177\u001b[0m fit(epochs, self.model, self.loss_func, opt=self.opt, data=self.data, metrics=self.metrics,\n\u001b[0;32m--> 178\u001b[0;31m callbacks=self.callbacks+callbacks)\n\u001b[0m\u001b[1;32m 179\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 180\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mcreate_opt\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlr\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mFloats\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mwd\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mFloats\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m0.\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m->\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m~/anaconda3/lib/python3.7/site-packages/fastai/utils/mem.py\u001b[0m in \u001b[0;36mwrapper\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m 78\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 79\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 80\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 81\u001b[0m \u001b[0;32mexcept\u001b[0m \u001b[0mException\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 82\u001b[0m if (\"CUDA out of memory\" in str(e) or\n", "\u001b[0;32m~/anaconda3/lib/python3.7/site-packages/fastai/basic_train.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(epochs, model, loss_func, opt, data, callbacks, metrics)\u001b[0m\n\u001b[1;32m 93\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mempty_val\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 94\u001b[0m val_loss = validate(model, data.valid_dl, loss_func=loss_func,\n\u001b[0;32m---> 95\u001b[0;31m cb_handler=cb_handler, pbar=pbar)\n\u001b[0m\u001b[1;32m 96\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mval_loss\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 97\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mcb_handler\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mon_epoch_end\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mval_loss\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;32mbreak\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m~/anaconda3/lib/python3.7/site-packages/fastai/basic_train.py\u001b[0m in \u001b[0;36mvalidate\u001b[0;34m(model, dl, loss_func, cb_handler, pbar, average, n_batch)\u001b[0m\n\u001b[1;32m 55\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mis_listy\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0myb\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0myb\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0myb\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 56\u001b[0m \u001b[0mnums\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0myb\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 57\u001b[0;31m \u001b[0;32mif\u001b[0m \u001b[0mcb_handler\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mcb_handler\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mon_batch_end\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mval_losses\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;32mbreak\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 58\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mn_batch\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnums\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m>=\u001b[0m\u001b[0mn_batch\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;32mbreak\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 59\u001b[0m \u001b[0mnums\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0marray\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnums\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdtype\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfloat32\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m~/anaconda3/lib/python3.7/site-packages/fastai/callback.py\u001b[0m in \u001b[0;36mon_batch_end\u001b[0;34m(self, loss)\u001b[0m\n\u001b[1;32m 257\u001b[0m \u001b[0;34m\"Handle end of processing one batch with `loss`.\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 258\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstate_dict\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'last_loss'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mloss\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 259\u001b[0;31m \u001b[0mstop\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0many\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'batch_end'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstate_dict\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'train'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 260\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstate_dict\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'train'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 261\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstate_dict\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'iteration'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m~/anaconda3/lib/python3.7/site-packages/fastai/callback.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, cb_name, call_mets, **kwargs)\u001b[0m\n\u001b[1;32m 198\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m__call__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcb_name\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcall_mets\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m->\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 199\u001b[0m \u001b[0;34m\"Call through to all of the `CallbakHandler` functions.\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 200\u001b[0;31m \u001b[0;32mif\u001b[0m \u001b[0mcall_mets\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mgetattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmet\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34mf'on_{cb_name}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m**\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstate_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mmet\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmetrics\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 201\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mgetattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcb\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34mf'on_{cb_name}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m**\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstate_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mcb\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcallbacks\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 202\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m~/anaconda3/lib/python3.7/site-packages/fastai/callback.py\u001b[0m in \u001b[0;36m\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m 198\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m__call__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcb_name\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcall_mets\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m->\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 199\u001b[0m \u001b[0;34m\"Call through to all of the `CallbakHandler` functions.\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 200\u001b[0;31m \u001b[0;32mif\u001b[0m \u001b[0mcall_mets\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mgetattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmet\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34mf'on_{cb_name}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m**\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstate_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mmet\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmetrics\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 201\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mgetattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcb\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34mf'on_{cb_name}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m**\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstate_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mcb\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcallbacks\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 202\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m~/anaconda3/lib/python3.7/site-packages/fastai/callback.py\u001b[0m in \u001b[0;36mon_batch_end\u001b[0;34m(self, last_output, last_target, **kwargs)\u001b[0m\n\u001b[1;32m 292\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mis_listy\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlast_target\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mlast_target\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mlast_target\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 293\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcount\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0mlast_target\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msize\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 294\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mval\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0mlast_target\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msize\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m*\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfunc\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlast_output\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0mlast_target\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdetach\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcpu\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 295\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 296\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mon_epoch_end\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m~/anaconda3/lib/python3.7/site-packages/fastai/metrics.py\u001b[0m in \u001b[0;36merror_rate\u001b[0;34m(input, targs)\u001b[0m\n\u001b[1;32m 42\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0merror_rate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mTensor\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtargs\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mTensor\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m->\u001b[0m\u001b[0mRank0Tensor\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 43\u001b[0m \u001b[0;34m\"1 - `accuracy`\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 44\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0;36m1\u001b[0m \u001b[0;34m-\u001b[0m \u001b[0maccuracy\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 45\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 46\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mdice\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mTensor\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtargs\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mTensor\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0miou\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mbool\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m->\u001b[0m\u001b[0mRank0Tensor\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m~/anaconda3/lib/python3.7/site-packages/fastai/metrics.py\u001b[0m in \u001b[0;36maccuracy\u001b[0;34m(input, targs)\u001b[0m\n\u001b[1;32m 27\u001b[0m \u001b[0minput\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0margmax\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdim\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mview\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mn\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 28\u001b[0m \u001b[0mtargs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtargs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mview\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mn\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 29\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m==\u001b[0m\u001b[0mtargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfloat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmean\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 30\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 31\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0maccuracy_thresh\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0my_pred\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mTensor\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_true\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mTensor\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mthresh\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mfloat\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m0.5\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msigmoid\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mbool\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m->\u001b[0m\u001b[0mRank0Tensor\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;31mRuntimeError\u001b[0m: Expected object of scalar type Long but got scalar type Float for argument #2 'other'" ] } ], "source": [ "learn.fit_one_cycle(4)" ] }, { "cell_type": "code", "execution_count": 52, "metadata": {}, "outputs": [], "source": [ "interp = ClassificationInterpretation.from_learner(learn)" ] }, { "cell_type": "code", "execution_count": 53, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "interp.plot_confusion_matrix(figsize=(12,12), dpi=60)" ] }, { "cell_type": "code", "execution_count": 23, "metadata": { "scrolled": true }, "outputs": [ { "data": { "text/html": [], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "LR Finder is complete, type {learner_name}.recorder.plot() to see the graph.\n" ] } ], "source": [ "learn.lr_find()" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "learn.recorder.plot()" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "learn.save('PLRstage1')" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "learn.unfreeze()" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "learn.fit_one_cycle(2, max_lr=slice(3e-4,2e-2))" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "learn.save('stage-2')" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "learn.load('PLRstage1')" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "import os, random" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "randomchoice = random.choice(os.listdir(\"/home/ubuntu/Eyetrain/Database/\"))\n", "randomchoice" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "img = open_image('/home/ubuntu/Eyetrain/Database/'+randomchoice)\n", "img" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "log_preds = learn.predict(img)\n", "log_preds" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "log_preds[0]" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "e^0.268269" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "pytorch_fastai", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.7.1" }, "varInspector": { "cols": { "lenName": 16, "lenType": 16, "lenVar": 40 }, "kernels_config": { "python": { "delete_cmd_postfix": "", "delete_cmd_prefix": "del ", "library": "var_list.py", "varRefreshCmd": "print(var_dic_list())" }, "r": { "delete_cmd_postfix": ") ", "delete_cmd_prefix": "rm(", "library": "var_list.r", "varRefreshCmd": "cat(var_dic_list()) " } }, "types_to_exclude": [ "module", "function", "builtin_function_or_method", "instance", "_Feature" ], "window_display": false } }, "nbformat": 4, "nbformat_minor": 2 }