{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"### **Dependencies**","metadata":{}},{"cell_type":"code","source":"import tensorflow as tf\nfrom tensorflow import keras\nfrom tensorflow.keras import layers\nimport tensorflow_addons as tfa\n\nimport matplotlib.pyplot as plt\nimport numpy as np\nimport re\n\ntry:\n    tpu = tf.distribute.cluster_resolver.TPUClusterResolver()\n    print('Device:', tpu.master())\n    tf.config.experimental_connect_to_cluster(tpu)\n    tf.tpu.experimental.initialize_tpu_system(tpu)\n    strategy = tf.distribute.experimental.TPUStrategy(tpu)\nexcept:\n    strategy = tf.distribute.get_strategy()\nprint('Number of replicas:', strategy.num_replicas_in_sync)\n\nAUTOTUNE = tf.data.experimental.AUTOTUNE","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:30:48.633260Z","iopub.execute_input":"2021-07-22T10:30:48.633640Z","iopub.status.idle":"2021-07-22T10:30:53.183046Z","shell.execute_reply.started":"2021-07-22T10:30:48.633602Z","shell.execute_reply":"2021-07-22T10:30:53.182236Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### **Parameters**","metadata":{}},{"cell_type":"code","source":"HEIGHT = 256\nWIDTH = 256\nCHANNELS = 3\nBATCH_SIZE = 1\nEPOCHS = 25\nGENERATOR_LR = 2e-4\nDISCRIMINATOR_LR = 2e-4\n\nHEIGHT_RESIZE = 128\nWIDTH_RESIZE = 128\nTRANSFORMER_BLOCKS = 6","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:30:53.184521Z","iopub.execute_input":"2021-07-22T10:30:53.184892Z","iopub.status.idle":"2021-07-22T10:30:53.192068Z","shell.execute_reply.started":"2021-07-22T10:30:53.184853Z","shell.execute_reply":"2021-07-22T10:30:53.191304Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### **Load data**","metadata":{}},{"cell_type":"code","source":"GCS_PATH = '../input/gan-getting-started'\nMONET_FILENAMES = tf.io.gfile.glob(str(GCS_PATH + '/monet_tfrec/*.tfrec'))\nPHOTO_FILENAMES = tf.io.gfile.glob(str(GCS_PATH + '/photo_tfrec/*.tfrec'))\n\ndef count_data_items(filenames):\n    n = [int(re.compile(r\"-([0-9]*)\\.\").search(filename).group(1)) for filename in filenames]\n    return np.sum(n)\n\nn_monet_samples = count_data_items(MONET_FILENAMES)\nn_photo_samples = count_data_items(PHOTO_FILENAMES)\n\nprint(f'Monet TFRecord files: {len(MONET_FILENAMES)}')\nprint(f'Monet image files: {n_monet_samples}')\nprint(f'Photo TFRecord files: {len(PHOTO_FILENAMES)}')\nprint(f'Photo image files: {n_photo_samples}')","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:30:53.193958Z","iopub.execute_input":"2021-07-22T10:30:53.194366Z","iopub.status.idle":"2021-07-22T10:30:53.216848Z","shell.execute_reply.started":"2021-07-22T10:30:53.194326Z","shell.execute_reply":"2021-07-22T10:30:53.215989Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def normalize_img(image):\n    image = tf.cast(image, dtype=tf.float32)\n    # Map values in the range [-1, 1]\n    return (image / 127.5) - 1.0\n\n# takes an image in tensor datatype and returns after reshaping it.\ndef decode_image(image):\n    image = tf.image.decode_jpeg(image, channels=CHANNELS)\n    image = tf.reshape(image, [HEIGHT, WIDTH, CHANNELS])\n    return image\n\n# takes a file and parse to get required variables.\ndef read_tfrecord(example):\n    tfrecord_format = {\n        \"image_name\": tf.io.FixedLenFeature([], tf.string),\n        \"image\": tf.io.FixedLenFeature([], tf.string),\n        \"target\": tf.io.FixedLenFeature([], tf.string)\n    }\n    example = tf.io.parse_single_example(example, tfrecord_format)\n    image = decode_image(example['image'])\n    return image\n\n# reads the TFRecords file and maps with the help of read_tfrecord function.\ndef load_dataset(filenames):\n    dataset = tf.data.TFRecordDataset(filenames)\n    dataset = dataset.map(read_tfrecord, num_parallel_calls=AUTOTUNE)\n    return dataset\n    \ndef get_dataset(filenames, augment=None, repeat=True, shuffle=True, batch_size=1):\n    dataset = load_dataset(filenames)\n\n    if augment:\n        dataset = dataset.map(augment, num_parallel_calls=AUTOTUNE)\n\n    dataset = dataset.map(normalize_img, num_parallel_calls=AUTOTUNE)\n\n    if repeat:\n        dataset = dataset.repeat()\n    if shuffle:\n        dataset = dataset.shuffle(512)\n        \n    dataset = dataset.batch(batch_size)\n    dataset = dataset.cache()\n    dataset = dataset.prefetch(AUTOTUNE)\n    \n    return dataset","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:30:53.219248Z","iopub.execute_input":"2021-07-22T10:30:53.219498Z","iopub.status.idle":"2021-07-22T10:30:53.231206Z","shell.execute_reply.started":"2021-07-22T10:30:53.219474Z","shell.execute_reply":"2021-07-22T10:30:53.229800Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"monet_ds = get_dataset(MONET_FILENAMES,augment=None, repeat=False, shuffle=False, batch_size=BATCH_SIZE)\nphoto_ds = get_dataset(PHOTO_FILENAMES,augment=None, repeat=False, shuffle=False, batch_size=BATCH_SIZE)","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:30:54.312146Z","iopub.execute_input":"2021-07-22T10:30:54.312459Z","iopub.status.idle":"2021-07-22T10:30:56.314451Z","shell.execute_reply.started":"2021-07-22T10:30:54.312431Z","shell.execute_reply":"2021-07-22T10:30:56.313636Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"example_monet = next(iter(monet_ds))\nexample_photo = next(iter(photo_ds))","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:30:56.315863Z","iopub.execute_input":"2021-07-22T10:30:56.316162Z","iopub.status.idle":"2021-07-22T10:30:56.474128Z","shell.execute_reply.started":"2021-07-22T10:30:56.316128Z","shell.execute_reply":"2021-07-22T10:30:56.473283Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plt.subplot(121)\nplt.title('Photo')\nplt.imshow(example_photo[0] * 0.5 + 0.5)\n\nplt.subplot(122)\nplt.title('Monet')\nplt.imshow(example_monet[0] * 0.5 + 0.5)","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:30:57.268075Z","iopub.execute_input":"2021-07-22T10:30:57.268423Z","iopub.status.idle":"2021-07-22T10:30:57.879501Z","shell.execute_reply.started":"2021-07-22T10:30:57.268393Z","shell.execute_reply":"2021-07-22T10:30:57.878623Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### **Build the generator**","metadata":{}},{"cell_type":"markdown","source":"The `generator` is responsible for generating images from a specific domain. `CycleGAN` architecture has two generators, in this context we will have one `generator` that will take `photos` and generate `Monet paints`, and the other `generator` will take `Monet paintings` and `generate photos.`","metadata":{}},{"cell_type":"markdown","source":"![image.png](attachment:042e0541-9060-4de0-94b6-5f95bd25104b.png)\nAn architecture for a CycleGAN generator. As you can see above, the representation size shrinks in the encoder phase, stays constant in the transformer phase, and expands again in the decoder phase. The representation size that each layer outputs is listed below it, in terms of the input image size, k. On each layer is listed the number of filters, the size of those filters, and the stride. Each layer is followed by an instance normalization and ReLU activation.","metadata":{},"attachments":{"042e0541-9060-4de0-94b6-5f95bd25104b.png":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAA2sAAAIjCAYAAACdyYMlAAAgAElEQVR4AeydB5gUxRaFJeecc84555yTYkZBxYwBRRADgmDOEUVBwICKgjkrIGZ9RswBFQUTIkGCgCD33VO71cwuu93bw85Us3vm+4ZpZnpmuu+erjl/3apbBwlvjAAjkLQIbNu2TTZs2CAbN27knTGgBqgBaoAaOKA1sHnz5qT9fvKLGIHcGoGDcuuJ87wZARcRIKwRUgnq1AA1QA3kFA0Q1lw4CX5nbosAYS23/cV5vk4jQFijScspJo3nQS1TA9QAYc2ppeCX55IIENZyyR+apxmNCBDWaG5ocKkBaoAayCkaIKxFw1vwKHJ2BAhrOfvvy7OLWAQIazRpOcWk8TyoZWqAGiCsRcxk8HByZAQIaznyz8qTimoECGs0NzS41AA1QA3kFA0Q1qLqNnhcOSkChLWc9NfkuUQ+AoQ1mrScYtJ4HtQyNUANENYibzt4gDkgAoS1HPBH5CkcOBEgrNHc0OBSA9QANZBTNEBYO3D8B4/0wI0AYe3A/dvxyA/ACBDWaNJyiknjeVDL1AA1QFg7AI0ID/mAiwBh7YD7k/GAD+QIENZobmhwqQFqgBrIKRogrB3IjoTHfqBEgLB2oPyleJw5IgKENZq0nGLSeB7UMjVADRDWcoQ14UlEPAKEtYj/gXh4OSsChDWaGxpcaoAaoAZyigYIaznLo/BsohkBwlo0/y48qhwaAcIaTVpOMWk8D2qZGqAGCGs51KzwtCIVAcJapP4cPJicHgHCGs0NDS41QA1QAzlFA4S1nO5aeH5RiABhLQp/BR5DrokAYY0mLaeYNJ4HtUwNUAOEtVxjX3iiDiNAWHMYfH517osAYY3mhgaXGqAGqIGcogHCWu7zMTzj5EeAsJb8mPMbc3EECGs0aTnFpPE8qGVqgBogrOViQ8NTT1oECGtJCzW/iBEQIazR3NDgUgPUADWQUzRAWKOzYQQSHwHCWuJjzG9gBLwIENZo0nKKSeN5UMvUADVAWPN+3rnBCCQsAoS1hIU293zwb7/9JkuWLJGlS5dmeMdrf/75Z6QC8t9//8knn3wib731luzcuTNpx0ZYo7k5UAzu119/LU899VTgHdf9X3/9JVE4rxUrVshdd90lt9xyi3zzzTeROKYoxIXHwHYnURqIB9bw+/vOO++k8Quvvvqq4Ppdu3at7NmzJ2m/yWG/6N9//5W3335bPvroI8F58MYIJCMChLVkRDmHf8fUqVPloIMOMvc8efJI+jtee+ONNyIVhX/++UeaNGkijRs3JqxtpJFJlJE5UD/377//lvPPP9+7ru31ndHjgAEDIgFFH3zwgdSoUcMcc+HCheWll16KxHEdqBrgcbNdzIoG4oG1P/74Q8qUKZPGM6BtKViwoFStWlWuvfZa2bJlS6Q8gz2YX375RYoXLy6HH344Yc0GhY8JjwBhLeEhzvlf0K9fP9PI3nnnnfLss8/uc3/++ecFDXqUestWr14tJUqUkOOPPz6pfyBm1miAsmKAXO+zadMmee2112ThwoXm/sgjj0jPnj2NuULnjH0ej+hljsLxnnHGGZI3b165/PLL5f3335fff//d+XG5jgu/n+1NojUQD6whq4ZrdciQIfLMM8+Y++LFi2XGjBlSpUoV085Mnz49Up7BGgW0iwDLG264wT7FR0Yg4REgrCU8xDn7C/BDUL16dalcubIxRpmdLUANwweQ0cLQge3bt8uqVavkxx9/NNvp34f9MXRy5cqVgmGWu3btSr+Lacgx/Or77783++DzM7rhs7Cf/azdu3eb4RdocOfMmeO9xe6Hz8N3Y7/YG44bx49jwTbMIPZNv1/se9JvE9ZonhJtnrLj8zds2JAGdPD/tm3bmh5lZLBiv2P9+vXmWsA1g21cZxgitG7dOsH7fvrpJzPk+NNPPzXXaexn47rE9Y19cf/yyy/Ne9GZErsfvg892hgm9cUXX5ihUvYY8L1oR1q2bCl16tSR7777Ls2wTPu5GPaMY8Ex2vfiEa/jGPA8Puuzzz6Tb7/9VtD7jzuOA23Vxx9/bK53e1y4/u3x4DNiPxPb9rPweb/++mua88Fn4HV8L7bxiOPDeaT/HP6fbUaUNRAPrM2aNcsAzx133JHmJxK/we+++66UKlVKypUrJ2vWrPFex28urkf85uJaxf8zuuE3FtcRrvWtW7dmtIt5r/0sXIeZ/Ybb33m0abjGcQOkwTvEjhbCfriGcWy4nmOPDeeEqRbwPNhGvPB58cQtw5Phk7kiAoS1XPFnTtxJwogUKFBA+vTpk2mDh29HI3XllVdKixYtzHySNm3amGEQGArRq1cvr1HGfmj0Tj31VAOBpUuXlkqVKskhhxxiGjh7JjB5kyZNkpo1a4rdp3fv3qaXH59hvxON+nnnnWeGR+EHoGLFijJ69GiZrr126Nl77733zL74PAz7wudhv2rVqpnvhGmzN8zhwdBJzIc599xzTQ9gly5dMgRJ+570j4Q1Gq8oG6/Mjg3XQYUKFQwM4fqM3Q/XBUDpwgsvlMmTJ5thTBhe/NVXX5lrr1atWuYaxbWO6+f+++/3wOWxxx4zn3nJJZcIMvTly5eXkiVLSoMGDeSFF14w3wMzhWuuYcOGps2AievcubMgY48MIL4T122+fPmkUKFCUq9ePUEvPa79V155Rbp162Y+F+0EhkmOHz/eGDmcA95/6623SqNGjeTBBx+UHj16mGNF9hC9/sOHDxds4304LrQLd999t7z44osGXnFOOJ6DDz7YGDV8Jr736aefNsdYtmxZc8z169eX66+/3oNM7IM2pH379vLQQw9J69atDQhjOza23GZ7EXUNxAMdo0aNkvz588vrr7+e/ifS+IgjjzzSABHADTd0dpx22mmmbcHvM67H4447zvgG/N7jDiCaP3++NG3aVHDd4Q6/gVEBFp6w3yrteDnmmGPM77f9rT/iiCNMJ0+sdwDsYT90RNv9LrroIjnssMPMNY12EDd83pgxY7z90N6NGzfOXOt4HZ3Ixx57rAwePNi0MWgD0Rah04s3RiCrESCsZTVS3C/DCCxYsMA0qqeffrppONH7HXtHbxQaQDSWMELokSpatKigMb7sssuke/fu5rnrrrvOfD56pgByMF4YE37jjTfK2LFjTcOOuTFokNFIYkgW9kHDiZ6uE088UTBPBY04Glnc0PMNA4j9Ro4cafbDcRYpUsQM2wS44fjQ+zZ06FCzHx6vueYawX74PJhC/BjhHNDo4/ibNWtmGmsYudtuu818V1b/IazRfEXdfGV0fCgSBO2j0wSgEbsPoAnzVAFYGFqMa+7iiy+Wo48+2nSIwKSgowadJjApMD+AP8yLAzjhczFXBVAFM3TooYea96HzBfs88MAD5tpExwg+B5ADSMK1jiwa2g47RBMdPzB1yOwhaw7AQmfPhAkTDHQhO4jvwz7oAcfn4zqGcYTJQmYObQqgCWYP8Id24qSTTpJTTjnFHCfOEeDav39/mTJlirRq1cp8Jgqb4DMBoJjTguFcZ555pvlewBg6h2bPnm32AYACcPFZ2A8G86ijjhJkH2Njy222F1HXQFhYw+gU/IaikwPFRDK6XXDBBeaawm8u2puOHTuaaw9QhflsADVcs2iP4AlwR7uA3/rmzZubodAYUok2AtcYMuD4DUd2HJ0+8AAnn3yy6UDBZ6D9Qvtj58lhf0AV9sO1Dx+C6xPfiXu7du0MhMHr4NrFfugExrHB2+DzTjjhBON7cPyYh4c2A/dOnTqZtgTP88YIZDUChLWsRor7ZRgBZMBgfooVK2ZMDYxN7B2NKoYYIHMFkwZThp51O6wRjScav4kTJ5qGDQ0oGrqrrrrKNMD4Urx/xIgRpqFD44598Z0wSjt27DDHBRiEQcTz99xzj3kOPf34/7Rp09Lsh6wankfjjM8GcOI7sX/s5+H/aPwxRh0Nvf08ZA1w3HY4pPmyLP5DWKP5irr5Sn98yD4BiHDNwADFvo7XYFDwGrJOyIYBRDCcERkugI+dOwZzAkODTDzmlKFNAFzhveiJhvEB7Pz8888mUwaYwT4AHuyDuS32u9FBU7duXVm2bJl5z6WXXmr2QVERfAayfTBIaHMAmngO70W2r3bt2uY1DJVCxw8gE5+PXnTAH44T54E2Ae9/+eWXzXM4Dxgz7AtjiGFU+Mwnn3zSPIdjwjmgswnvQ6VZ+72AMBg1nC86iHB8AE58BzL6OGd8r93fnicf2V5EXQNhYQ1aB0Ch88X+3sb+fOK3Fj4A1xnmwKN9QUfHTTfd5PkGZKvgLdDxi2sJFWnRruA3HdegvT3++OPmeWTcAHTojMV++L8d+ojhiWin4EP+97//mf2GDRtmOmruu+8+bz8cq8344fhww+gevG/u3LnefoBRvB/XN4ZzY3gzfA86fjCvHx4AfgXnyRsjkNUIENayGinut08E0GCixwuNFXq8MBQg/R2NHW7o6cZ+GA6ARtI2VG+++aZplNEQr9LhBGjE0fOEBs/ug/fDVMFoIWtmhzek/5HAd6CBx5Aq7I8hVegBQ+Ys9rMwtAL7obcdDTh6vfG9ixYtMsYO5g53a0IBc3g/hnqioV++fDkOKa4bYY3mK+rmK/3xIfuEbBc6LmB+Yl+3rwE67r33Xu81AMnnn39uMkUYagz4uf32200WCdACKMK1jPmugDpc+/ZzkV1HBg6ZK3w+MlYwa4CzmTNnevPJ8B58DyAHnTloF2DcAJCASlzjV1xxhfm/hSC0Gciywzxh7hsgCh1NADZ8HvbD+9FZhPcDwHBceB7zZwCAyMDh2O3xzps3z+yLLAAKruB9GBUAiLN3LIGAoZB4LyAPwyhxTsge4hzs8dnP5CPbiQNFA+l/h4N+GNH5gWsE2e2MbvhN7qWdGmhT0FGKIc7IjgPc7G8zHpGpx+dgKKW9plG4JPYGH4GOI3QgAcTQhsGrpIcldCDjs9AhhM/A7zw6b/C7H+sd0BZhP/gaXMfwGOj8wfVsjw3giNFA2A+ghn2xjefSf17ssXKbEfCLAGHNLzp8zTcCtrGCCQGE+N1g5NBgYYhQbOMHMwTTggYOjS72sUMiM/o8VJ7DPsiWpb/ZKk0333yzaeQz+ywYJ7yGOSowaBiPDpCEaUt/R28YhkTBTMGoYY4JfkxizyH9cfj9n7BGE3agmDB7nOj4wJAgdGigw8Q+j0dkmwA6MCy2KAiuFVzLgwYNMtkkgBdACsOeYMDQGYN9kHnCdYdeakCZ/Vxkx2CqkDnH88imA7pg2tBW4DpEJg096HgP5rPgNWS0kInDZ6OnHEMRMxpWOHDgQNPrjUn+mNuGtgBZc/v9eD9gC732MFv2eXQsoT3A8CYAHZ7H8WGYFIZBYejUWWedZT4P/0/fluA5tJU4H8xfw/fCJNrP5yPbhgNRA2FhzWbB0cmR0Q2drrh2MCQZmXNcc4Cn9NcTrk90usA74LrC/DR0IGd2QzYN1xx+/9PfUOgEr8FDIPuFbcx3jb3hNx+jctA2oV0AnKH9wjFkdGyYRoH9zjnnHNNuARZ5YwTijQBhLd7I8X3GbMF8oVcbPVV+N8wpg1lBj3rsDe8FLMH02flvDz/8cOwuJhOHseRoiG3PNRre9DcMScDxwOyhCAAaXGTLYm9ocNFjhh8AFEdB7z8aXAzPgnnM6A4DiEYXPxj4jnhBDcdBWKMhO9AMGa4RmBEMSwS4AWbsOQBQAEWAI2S48DwWo0YGCc9jyPITTzxhris7JAnXEGDH9lLjWrbwg89GB441VfZ78DzaDmTnMAwZ1zky6HgemTsYI/TUY3/AEObCImuXvroiPgPDNTFnBp1NGIKIz0Kvvf0uC6coOhJ7vjaDhrlwdl+0DZizBrOIzwPk2Q4etEPp76ggh/ciXogp2pXYeNrP5SPbiQNFA2FgDaNqMEQQnS7oSEl/w1BDzHXF9Y/OGiyUjW0UEcrotxmdt+hwReYNQxzT+xD83tqKkCgUhN/w9NCEY8J70RmFrD7m0aI9QbsXe8NnobMWnUW4ZuEtcGyYR5vRsSFDh9hgaCbaIvw998c7xB4Lt3NfBAhrue9vnm1njCpqaKzQU+Z3w9wu9HrD0NgJvNgfDRkMEYYqYriCbZjR+MU2augJw34YF26zZ2jI7Q37oscf80QweRiNIjJ4ODaYO3vDfng/TCQmHsMg/vDDD+b/mDwc2yuHYRMoPIBeMTTmyK7h89KDpP3srD4S1mjCDhQTZo8T1xKABvM1LFTZ1wBiuC5QLMi+9txzz5neZ8xntRkzQA2K9mBfmByYHcxFg3nCsGT7eXgenSkwTjBzaAsAXhjKZPdBzzvmg6BgEYYQogAJPhcdNNgH34VOIBQX+fDDD733YV8YNuwL0MO+yLKhswjDJ+3nWzjF0Cp7/Dg3HC960ZE1tICFdgdZQ1SDxPsBjOj8wVAvuw+eRzuEYkXorQdMoj1DRtLO57PfzUe2DweaBsLAGvbFsGfAC65T+zuPR1xjuH7Q1sArIHOOua24njDk0e6L31p0wqCQETLumHOGz0RhEfy+2hs+H8MpcR3jt90uF4Bsuv0sPOJaRQcLfu/hVdCWASZxndsbIBCdSngen2nfh2PFMcTe0H5gH/gjdGBh2Hffvn1jd+E2IxA6AoS10CHjGxABNF7IRsH4YLI9YAbDCmPvmMMBOMMwJQyBwrwXgI+9oZcbPVgnnHCCeQrGBY04qqM9+uijxkBhvDcgDD1n6PVC4weTA4MEeEMvPqrRYZgWPgvfj9sq7W3Dd2IoBcpoo/ochhzhs3DMGKKFhhlDGtGQorHGPBWYL/SSofcPwx3QwOMGo4l9MM9lf26ENZqxA8mMwUDZwjrIeFkgwzkAZPAaDAsAzZ4Xske4VtCjDCOE4UIYHgjThWsU89cALCgwgE4TDJ+078Xz6HABzKDDBFXYcL2imADME+AL89EAeYBBHA8WwwZE4XstIOG6hbHCnDDMJ0H2DZUmsR8yhGhL8L1oH9CRhO+1x2B7zGPPF5+LXnW0HyiQYPfFeeP4MCwbx4KOJZwnOqAwXw1mE8OyAYRot9A5BAgFjGLujM1G2s/jI9uHA00DYWANnSJoAwBX6GTB7/W9OkUC1zTmjuNaAqghY4YbrikMb0THC+ar4vceQyPRtuBaxm87wAkFitAOIWuPETMYsowqtPgNx/WHfdDhgyHZuA5xneKz4AkAU7jbZXwwJx2f3bVrV9Pm4JgxNQPDLnF8tqMYbQbm0cJnwIvAY6CtQXuCfdFe2XlyGD7JGyOwPxEgrO1P9HLxe5EJg+lB44VGEsYo/R2AhQbNzjNL3wNlTRHKWdsbnsMwAzSyaNTxiMYbBg8NLm5oTGHm7D4wbjB9ADw7DAL7YtgSGmFrEvF5aIBxvKgcaW9o3Dt06JDm89CoYx8MywBgopAK5uXAoO7PjbBGM3YgmTH0TqPjAtc5Mt+xxw6AQacHslwwPvY1dLog04T2ANcm7rimsW4ZPgemDPtjcj46Suz78IhhgQAZZNfw+YAq9KDjGoaBwjWMz8N1DOixEIXPj4UoDEnEcCYMNcT+gEe8H8eL6x3fBfDDcyiKZDNo+Dw7pwaQaY8N89tQ5RbHggwdnsd77L4wjXgOgInsHcylbXfwiLYSJhP7YF/EAVlD+/l8ZLtwoGogDKzhNxrtAvQf6xcwRQIdJxjJgqyZveF3HNCDzlj7e4/rCddibHVIXPsoAIZ9cK3jmkc7giJi8Cq4wRvYDtvYz0JGDh20+C7ckYXDNYwROPgctDnYRucSjhvDuXHDvvAlALbYz0NbhKHS+D6AKN6DYda8MQL7EwHC2v5ELxe/FwCDRR0zGqttn8NwJWSv0HuM52CgYm+YD4Ln8SOFGxo/3FBpDT1fmPSLrBkMEG62McU2IBDzQbAPhmmhlzz2dbsNswcARM+XLReO78R3x95g0gCB+DzMi4OZtJ+Bc0WpfphDC4Ox7w2zTVijKTuQTBmuXXS2YPgeIAzXiT1++xqGFluAsa+hciKMGXqkYVwwtw3XNTpmYHbQFgCGcF3Z9+ARWXhk4wBU9rvwHMwO1jTE4tjYtsVFAJP4fvSkp89S2ePD96N3HW0JPst+H44R34VhS/Y5fCeOFceGNsI+b48XbZo9LrwGuEQ7FLsvvhc96ugsQmYQbVls9hDgh/fg0X4+H9kuHKgaCANraEOsP7CPmNuFay52WKT9TcVvMG74vUcbhOwafu8t0NnXsQ9G8aAtwHBojPTBKBj8Xsfug/1WrVrl+QtkxvG9uMXuh/fh9x7ZP8ylR3uEY0dbaDts7f5oUwBtODZk03Gt2xvaKbwnTIzse/nICMRGgLAWGw1u5/gI2AY2/Ylm9nz6/fb3/4Q1mrID1ZSFOW4LNBjGhLt9r922r9vn/R7tvrGfZZ/ze599Ld732ffvz6M93/35DL6XbUaUNZAMEMnK73NW9tnf3++M3u/qezM6Fj6XcyNAWMu5f1ueWQQjQFij8Yqy8eKxUZ/UADUQRgPJgLUI/pTzkBiBpEaAsJbUcPPLcnsECGs0QmGMEPelXqgBaiDKGiCs5XZXw/NPRgQIa8mIMr+DEUiNAGGNxivKxovHRn1SA9RAGA0Q1mhvGIHER4CwlvgY8xsYAS8ChDUaoTBGiPtSL9QANRBlDRDWvJ93bjACCYsAYS1hoeUHMwL7RoCwRuMVZePFY6M+qQFqIIwGCGv7/s7zGUYguyNAWMvuiPLzGAGfCBDWaITCGCHuS71QA9RAlDVAWPP5wedLjEA2RYCwlk2B5McwAlmJAGGNxivKxovHRn1SA9RAGA0Q1rLyy899GIH9iwBhbf/ix3czAqEiQFijEQpjhLgv9UINUANR1gBhLZQF4M6MQFwRIKzFFTa+iRGILwKENRqvKBsvHhv1SQ1QA2E0QFiLzwvwXYxAmAgQ1sJEi/syAvsZAcIajVAYI8R9qRdqgBqIsgYIa/tpCvh2RiALESCsZSFI3IURyK4IENZovKJsvHhs1Cc1QA2E0QBhLbvcAT+HEcg8AoS1zGPDVxiBbI8AYY1GKIwR4r7UCzVADURZA4S1bLcJ/EBGYJ8IENb2CQmfYAQSFwHCGo1XlI0Xj436pAaogTAaIKwlzi/wkxkBGwHCmo0EHxmBJESAsEYjFMYIcV/qhRqgBqKsAcJaEowDvyLXR4CwluslwAAkMwKENRqvKBsvHhv1SQ1QA2E0QFhLpoPgd+XWCBDWcutfnuftJAKENRqhMEaI+1Iv1AA1EGUNENacWAl+aS6LAGEtl/3BebpuI0BYo/GKsvHisVGf1AA1EEYDhDW3noLfnjsikGVY27Nnj+Cu/+TYu3eOueNvz7N0EAHCGo1QGCPEfakXaoAaiLIGCGvxG4kUO71Hdu7eLlv/3SRbd26QLTnkjnPZunOT7Ni9zbADzjXsbceuPbJh6zrZs+t7kRx6373rB1m3ZbP8959/dLIMayvWfC1b/75TZPstOfa+YeNs+fK3Nf4R46uMwH5EgLBG4xVl48Vjoz6pAWogjAYIa/Ebgm07d8kbq16Vt355SJaveUBeXX1/jrrjnN745UF546cXZMO27aED9deWP2T9+qtlz7ZLRHLo/b9tU2XN2lmy498dvvHJMqz9s2WuyD/TNGiX6mPOu6ec1zT5Z+sTvgHji4zA/kSAsEYjFMYIcV/qhRqgBqKsAcJa/I5g0461suznebLk5/mydPU8veMxJ91Tz+3ne2Tt1h9DB2rPrm8NqO3JgcxhOQrntuefy2TPf5t945MlWDPZy+136wfmPEizAcOjOb8di8Scr2/Y+CIjEF8ECGs0XlE2Xjw26pMaoAbCaICwFp8XwLt+27oyh8FZxqC5RIH0h00fhw/Urm9SMmo5nD1EYU2yA9ZMhBXWYsEmx24rrPHGCCQqAoQ1GqEwRoj7Ui/UADUQZQ0Q1uJ3C79s/jrXwNrKDe+HDxRhzYtZljJrZm/Cmhc0bjAC8UaAsEbjFWXjxWOjPqkBaiCMBghr8boBkTWbv8k1sPYtYc0n4cXMmk9wMhnSycxa/C0P3xkYAcIajVAYI8R9qRdqgBqIsgYIa4E/+5nukJtg7TvCmg+PENZ8gkNYy7QF4QsJiwBhjcYrysaLx0Z9UgPUQBgNENbitwuEtYDYcRikFyAOg0w/cZGZNU8c3Mj+CBDWaITCGCHuS71QA9RAlDVAWIvfJxDWAmJHWPMCRFgjrHli4EbiI0BYo/GKsvHisVGf1AA1EEYDhLX4fQNhLSB2hDUvQIQ1wponBm4kPgKENRqhMEaI+1Iv1AA1EGUNENbi9w2EtYDYEda8ABHWCGueGLiR+AgQ1mi8omy8eGzUJzVADYTRAGEtft9AWAuIHWHNCxBhjbDmiYEbiY8AYY1GKIwR4r7UCzVADURZA4S1+H0DYS0gdoQ1L0CENcKaJwZuJD4ChDUarygbLx4b9UkNUANhNEBYi983ENYCYkdY8wJEWCOseWLgRuIjQFijEQpjhLgv9UINUANR1gBhLX7fQFgLiB1hzQsQYY2w5omBG4mPAGGNxivKxovHRn1SA9RAGA0Q1uL3DYS1gNgR1rwAEdYIa54YuJH4CBDWaITCGCHuS71QA9RAlDVAWIvfNxDWAmJHWPMCRFgjrHli4EbiI0BYo/GKsvHisVGf1AA1EEYDhLX4fQNhLSB2hDUvQIQ1wponBm4kPgKENRqhMEaI+1Iv1AA1EGUNENbi9w2EtYDYEda8ABHWCGueGLiR+AgQ1mi8omy8eGzUJzVADYTRAGEtft9AWAuIHWHNCxBhjbDmiYEbiY8AYY1GKIwR4r7UCzVADURZA4S1+H0DYS0gdoQ1L0CENcKaJwZuJD4ChDUarygbLx4b9UkNUANhNEBYi983ENYCYkdY8wJEWCOseWLgRuIjQFijEQpjhLgv9UINUANR1gBhLX7fQFgLiB1hzQsQYY2w5omBG4mPAGGNxivKxovHRn1SA9RAGA0Q1uL3Db6a3moAACAASURBVIS1gNgR1rwAEdYIa54YuJH4CBDWaITCGCHuS71QA9RAlDVAWIvfNxDWAmJHWPMCRFgjrHli4EbiI0BYo/GKsvHisVGf1AA1EEYDhLX4fQNhLSB2hDUvQIQ1wponBm4kPgKENRqhMEaI+1Iv1AA1EGUNENbi9w2EtYDYEda8ABHWCGueGLiR+AgQ1mi8omy8eGzUJzVADYTRAGEtft9AWAuIHWHNCxBhjbDmiYEbiY8AYY1GKIwR4r7UCzVADURZA4S1+H0DYS0gdoQ1L0CENcKaJwZuJD4ChDUarygbLx4b9UkNUANhNEBYi983ENYCYkdY8wJEWCOseWLgRuIjQFijEQpjhLgv9UINUANR1gBhLX7fQFgLiB1hzQsQYY2w5omBG4mPAGGNxivKxovHRn1SA9RAGA0Q1uL3DYS1gNgR1rwAEdYIa54YuJH4CBDWaITCGCHuS71QA9RAlDVAWIvfNxDWAmJHWPMCRFgjrHli4EbiI0BYo/GKsvHisVGf1AA1EEYDhLX4fQNhLSB2hDUvQIQ1wponBm4kPgKENRqhMEaI+1Iv1AA1EGUNENbi9w2EtYDYEda8ABHWCGueGLiR+AgQ1mi8omy8eGzUJzVADYTRAGEtft9AWAuIHWHNCxBhjbDmiYEbiY8AYY1GKIwR4r7UCzVADURZA4S1+H0DYS0gdoQ1L0CENcKaJwZuJD4ChDUarygbLx4b9UkNUANhNEBYi983ENYCYkdY8wJEWCOseWLgRuIjQFijEQpjhLgv9UINUANR1gBhLX7fQFgLiB1hzQsQYY2w5omBG4mPAGGNxivKxovHRn1SA9RAGA0Q1uL3DYS1gNgR1rwAEdYIa54YuJH4CBDWaITCGCHuS71QA9RAlDVAWIvfNxDWAmJHWPMCRFgjrHli4EbiI0BYo/GKsvHisVGf1AA1EEYDhLX4fQNhLSB2hDUvQIQ1wponBm4kPgKENRqhMEaI+1Iv1AA1EGUNENbi9w2EtYDYEda8ABHWCGueGLiR+AgQ1mi8omy8eGzUJzVADYTRAGEtft9AWAuIHWHNCxBhjbDmiYEbiY8AYY1GKIwR4r7UCzVADURZA4S1+H0DYS0gdoQ1L0CENcKaJwZuJD4ChDUarygbLx4b9UkNUANhNEBYi983ENYCYkdY8wJEWCOseWLgRuIjQFijEQpjhLgv9UINUANR1gBhLX7fQFgLiB1hzQsQYY2w5omBG4mPAGGNxivKxovHRn1SA9RAGA0Q1uL3DYS1gNgR1rwAEdYIa54YuJH4CBDWaITCGCHuS71QA9RAlDVAWIvfNxDWAmJHWPMCRFgjrHli4EbiI0BYo/GKsvHisVGf1AA1EEYDhLX4fQNhLSB2hDUvQIQ1wponBm4kPgKENRqhMEaI+1Iv1AA1EGUNENbi9w2EtYDYEda8ABHWCGueGLiR+AgQ1mi8omy8eGzUJzVADYTRAGEtft9AWAuIHWHNCxBhjbDmiYEbiY8AYY1GKIwR4r7UCzVADURZA4S1+H0DYS0gdoQ1L0CENcKaJwZuJD4ChDUarygbLx4b9UkNUANhNEBYi983ENYCYkdY8wJEWCOseWLgRuIjQFijEQpjhLgv9UINUANR1gBhLX7fQFgLiB1hzQsQYY2w5omBG4mPAGGNxivKxovHRn1SA9RAGA0Q1uL3DYS1gNgR1rwAEdYIa54YuJH4CBDWaITCGCHuS71QA9RAlDVAWIvfNxDWAmJHWPMCRFgjrHli4EbiI0BYo/GKsvHisVGf1AA1EEYDhLX4fQNhLSB2hDUvQIQ1wponBm4kPgKENRqhMEaI+1Iv1AA1EGUNENbi9w2EtYDYEda8ABHWCGueGLiR+AgQ1mi8omy8eGzUJzVADYTRAGEtft9AWAuIHWHNCxBhjbDmiYEbiY8AYY1GKIwR4r7UCzVADURZA4S1+H0DYS0gdoQ1L0CENcKaJwZuJD4ChDUarygbLx4b9UkNUANhNEBYi983ENYCYkdY8wJEWCOseWLgRuIjQFijEQpjhLgv9UINUANR1gBhLX7fQFgLiB1hzQsQYY2w5omBG4mPAGGNxivKxovHRn1SA9RAGA0Q1uL3DYS1gNgR1rwAEdYIa54YuJH4CBDWaITCGCHuS71QA9RAlDVAWIvfNxDWAmJHWPMCRFgjrHli4EbiI0BYo/GKsvHisVGf1AA1EEYDhLX4fQNhLSB2hDUvQIQ1wponBm4kPgKENRqhMEaI+1Iv1AA1EGUNENbi9w2EtYDYEda8ABHWCGueGLiR+AgQ1mi8omy8eGzUJzVADYTRAGEtft9AWAuIHWHNCxBhjbDmiYEbiY8AYY1GKIwR4r7UCzVADURZA4S1+H0DYS0gdoQ1L0CENcKaJwZuJD4ChDUarygbLx4b9UkNUANhNEBYi983ENYCYkdY8wJEWCOseWLgRuIjQFijEQpjhLgv9UINUANR1gBhLX7fQFgLiB1hzQsQYY2w5omBG4mPAGGNxivKxovHRn1SA9RAGA0Q1uL3DYS1gNgR1rwAEdYIa54YuJH4CBDWaITCGCHuS71QA9RAlDVAWIvfNxDWAmJHWPMCRFgjrHli4EbiI0BYo/GKsvHisVGf1AA1EEYDhLX4fQNhLSB2hDUvQIQ1wponBm4kPgKENRqhMEaI+1Iv1AA1EGUNENbi9w2EtYDYEda8ABHWCGueGLiR+AgQ1mi8omy8eGzUJzVADYTRAGEtft9AWAuIHWHNCxBhjbDmiYEbiY8AYY1GKIwR4r7UCzVADURZA4S1+H0DYS0gdoQ1L0CENcKaJwZuJD4ChDUarygbLx4b9UkNUANhNEBYi983ENYCYkdY8wJEWCOseWLgRuIjQFijEQpjhLgv9UINUANR1gBhLX7fQFgLiB1hzQsQYY2w5omBG4mPAGGNxivKxovHRn1SA9RAGA0Q1uL3DYS1gNgR1rwAEdYIa54YuJH4CBDWaITCGCHuS71QA9RAlDVAWIvfNxDWAmJHWPMCRFgjrHli4EbiI0BYo/GKsvHisVGf1AA1EEYDhLX4fQNhLSB2hDUvQIQ1wponBm4kPgKENRqhMEaI+1Iv1AA1EGUNENbi9w2EtYDYEda8ABHWCGueGLiR+AgQ1mi8omy8eGzUJzVADYTRAGEtft9AWAuIHWHNCxBhjbDmiYEbiY8AYY1GKIwR4r7UCzVADURZA4S1+H0DYS0gdoQ1L0CENcKaJwZuJD4ChDUarygbLx4b9UkNUANhNEBYi983ENYCYkdY8wJEWCOseWLgRuIjQFijEQpjhLgv9UINUANR1gBhLX7fQFgLiB1hzQsQYY2w5omBG4mPAGGNxivKxovHRn1SA9RAGA0Q1uL3DYS1gNgR1rwAEdYIa54YuJH4CBDWaITCGCHuS71QA9RAlDVAWIvfNxDWAmJHWPMCRFgjrHli4EbiI0BYo/GKsvHisVGf1AA1EEYDhLX4fQNhLSB2hDUvQIQ1wponBm4kPgKENRqhMEaI+1Iv1AA1EGUNENbi9w2EtYDYEda8ABHWCGueGLiR+AgQ1mi8omy8eGzUJzVADYTRAGEtft9AWAuIHWHNCxBhjbDmiYEbiY8AYY1GKIwR4r7UCzVADURZA4S1+H0DYS0gdoQ1L0CENcKaJwZuJD4ChDUarygbLx4b9UkNUANhNEBYi983ENYCYkdY8wJEWCOseWLgRuIjQFijEQpjhLgv9UINUANR1gBhLX7fQFgLiB1hzQsQYY2w5omBG4mPAGGNxivKxovHRn1SA9RAGA0Q1uL3DYS1gNgR1rwAEdYIa54YuJH4CBDWaITCGCHuS71QA9RAlDVAWIvfNxDWAmJHWPMCRFgjrHli4EbiI0BYo/GKsvHisVGf1AA1EEYDhLX4fQNhLSB2hDUvQIQ1wponBm4kPgKENRqhMEaI+1Iv1AA1EGUNENbi9w2EtYDYEda8ABHWCGueGLiR+AgQ1mi8omy8eGzUJzVADYTRAGEtft9AWAuIHWHNCxBhjbDmiYEbiY8AYY1GKIwR4r7UCzVADURZA4S1+H0DYS0gdoQ1L0CENcKaJwZuJD4ChDUarygbLx4b9UkNUANhNEBYi983ENYCYkdY8wJEWCOseWLgRuIjQFijEQpjhLgv9UINUANR1gBhLX7fQFgLiB1hzQsQYY2w5omBG4mPAGGNxivKxovHRn1SA9RAGA0Q1uL3DYS1gNgR1rwAEdYIa54YuJH4CBDWaITCGCHuS71QA9RAlDVAWIvfNxDWAmJHWPMCRFgjrHli4EbiI0BYo/GKsvHisVGf1AA1EEYDhLX4fQNhLSB2hDUvQIQ1wponBm4kPgKENRqhMEaI+1Iv1AA1EGUNENbi9w2EtYDYEda8ABHWCGueGLiR+AgQ1mi8omy8eGzUJzVADYTRAGEtft9AWAuIHWHNCxBhjbDmiYEbiY8AYY1GKIwR4r7UCzVADURZA4S1+H0DYS0gdoQ1L0CENcKaJwZuJD4ChDUarygbLx4b9UkNUANhNEBYi983ENYCYkdY8wJEWCOseWLgRuIjQFijEQpjhLgv9UINUANR1gBhLX7fQFgLiB1hzQsQYY2w5omBG4mPAGGNxivKxovHRn1SA9RAGA0Q1uL3DYS1gNgR1rwAEdayC9b27BHZs1PvO3L4Heeo58pbXBEgrNEIhTFC3Jd6oQaogShrIDNY26M+YceurbL67y/ly/VvyufrlstnOeiO8/l6/dvy25aV8u9u+L7wloCwFhAzwpoXIMJatsHaNpF/7tT7NTn8fpPIf2s9AXEjXAQIazReUTZePDbqkxqgBsJoIDNY27lru3zwx1Oy5Od5Ofq+9Of58sW6ZbL7v93hzIDuTVgLCBlhzQsQYS27YO2/jQpp0/V+ac6+b7tEZNcPnoC4ES4ChDUaoTBGiPtSL9QANRBlDWQGa2u3rlJImy9LV+f8+6urH5DNO9aHMwO6N2EtIGSENS9AhLX0cLVjkRecUBuEtVDhyq07E9ZovKJsvHhs1Cc1QA2E0UBmsLbq7xUmo5YbYG3Z6ntl3T9rQtsawlpAyAhrXoAIa4S1cJlAZta8iyeeDcIajVAYI8R9qRdqgBqIsgYyg7UfNn6ca2BtqcLan9t+Cm0JCGsBISOseQEirBHWCGve5ZD4DcIajVeUjRePjfqkBqiBMBrIDNa+z2WwtpawlumQV8xb/G7D++ENFmHNixlhjbBGWPMuh8RvENZohMIYIe5LvVAD1ECUNUBYw5y8e4WwlvncRMJaUC2Ly7Rw32ZfA0pYI6wR1nwvkex9kbBG4xVl48Vjoz6pAWogjAYIa4S1oHmJhDXCWjjQSA9mGf2fBUb8Y8o5a/tFb4Q1GqEwRoj7Ui/UADUQZQ0Q1ghrhLUgGAt6nZk1f/AgrIWPD2GNsLaR5inK5onHRn1SA9RAsjRAWCOsEdaCYCzodcJaeBhhZs0/ZoQ1whphTZJlhPg9NN3UADUQZQ0Q1ghrhLUgGAt6nbDmDx7MrIWPD2GNsEZYI6xRA9QANUANqAYIa4Q1wloQjAW9TlgLDyPMrPnHjLBGWKNJo0mjBqgBaoAaIKyllqtnNUg/YGOBEcKaP1hklDkLeo6w5h9TwhphjSaNJo0aoAaoAWqAsEZYW43Mov+dsEZY8weLIDDL6HXCmn9MCWuENZo0mjRqgBqgBqgBwhphLQDUAHKENcKaP1hkBGNBzxHW/GNKWCOs0aTRpFED1AA1QA0Q1ghrhDV/zxzEHOZ1zlkLH0TCmn/MCGuENZo0mjRqgBqgBqgBwhphjbDm75kJa0FpxThfJ6z5C4+wRlijSaNJowaoAWqAGiCsEdYIa/6embAWJ4wFBY6w5i88whphjSaNJo0aoAaoAWqAsEZYI6z5e+Yg5jCvcxhk+CAS1vxjRlgjrNGk0aRRA9QANUANENYIa4Q1f89MWGNmTbIkgmyOE2GNsEaTRpNGDVAD1AA1QFgjrBHWCGtOYISZNX/hEdYIazRpNGnUADVADVADhDXCGmHN3zNnKanCYZDhg0hY848ZYY2wRpNGk0YNUAPUADVAWCOsEdb8PTNhLZuH99mAEtb8hUdYI6zRpNGkUQPUADVADRDWCGuENX/PbNnC95GZtfBBJKz5x4ywRlijSaNJowaoAWqAGiCsEdYIa/6e2RfSbNKJsBY+iIQ1/5gR1ghrNGk0adQANUANUAOENcIaYc3fMxPWLJFm8yNhzV94hDXCGk0aTRo1QA1QA9QAYY2wRljz98yEtWyGNBtQwpq/8AhrhDWaNJo0aoAaoAaoAcIaYY2w5u+ZLVv4PnIYZPggEtb8Y0ZYI6zRpNGkUQPUADVADRDWCGuENX/P7AtpNulEWAsfRMKaf8wIa4Q1mjSaNGqAGqAGqAHCGmGNsObvmQlrlkiz+ZGw5i88whphjSaNJo0aoAaoAWqAsEZYI6z5e2bCWjZDmg0oYc1feIQ1whpNGk0aNUANUAPUAGGNsEZY8/fMli18HzkMMnwQCWv+MSOsEdZo0mjSqAFqgBqgBghrhDXCmr9n9oU0m3QirIUPImHNP2aENcIaTRpNGjVADVAD1ABhjbBGWPP3zIQ1S6TZ/EhY8xceYY2wRpNGk0YNUAPUADVAWCOsEdb8PTNhLZshzQaUsOYvPMIaYY0mjSaNGqAGqAFqgLBGWCOs+Xtmyxa+jxwGGT6IhDX/mBHWCGs0aTRp1AA1QA1QA4Q1whphzd8z+0KaTToR1sIHkbDmHzPCGmGNJo0mjRqgBqgBaoCwRlgjrPl7ZsKaJdJsfiSs+QuPsEZYo0mjSaMGqAFqgBogrBHWCGv+npmwls2QZgNKWPMXHmGNsEaTRpNGDVAD1AA1QFgjrBHW/D2zZQvfRw6DDB9Ewpp/zAhrhDWaNJo0aoAaoAaoAcIaYY2w5u+ZfSHNJp0Ia+GDSFjzjxlhjbBGk0aTRg1QA9QANUBYI6wR1vw9M2HNEmk2PxLW/IVHWCOs0aTRpFED1AA1QA0Q1ghrhDV/z0xYy2ZIswElrPkLj7BGWKNJo0mjBqgBaoAaIKwR1ghr/p7ZsoXvI4dBhg8iYc0/ZoQ1whpNGk0aNUANUAPUAGGNsEZY8/fMvpBmk06EtfBBJKz5x4ywRlijSaNJowaoAWqAGiCsEdYIa/6embBmiTSbHwlr/sIjrBHWaNJo0qgBaoAaoAYIa4Q1wpq/ZyasZTOk2YAS1vyFR1gjrNGk0aRRA9QANUANENYIa4Q1f89s2cL3kcMgwweRsOYfM8IaYY0mjSaNGqAGqAFqgLBGWCOs+XtmX0izSSfCWvggEtb8Y0ZYI6zRpNGkUQPUADVADRDWCGuENX/PTFizRJrNj4Q1f+ER1ghrNGk0adQANUANUAOENcIaYc3fMxPWshnSbEAJa/7CI6wR1mjSaNKoAWqAGqAGCGuENcKav2e2bOH7yGGQ4YNIWPOPGWGNsEaTRpNGDVAD1AA1QFgjrBHW/D2zL6TZpBNhLXwQCWv+MSOsEdZo0mjSqAFqgBqgBghrhDXCmr9nJqxZIs3mR8Kav/AIa4Q1mjSaNGqAGqAGqAHCGmGNsObvmQlr2QxpNqCENX/hEdYIazRpNGnUADVADVADhDXCGmHN3zNbtvB95DDI8EEkrPnHjLBGWKNJo0mjBqgBaoAaIKwR1ghr/p7ZF9Js0omwFj6IhDX/mO0PrO3ZI7Jnl97/zeH33Xp+eq4Z3LZt2yYbNmzI/Ed+w1+ycf3anH/fsD7zGPiYwA0bNXbr/8z58VENmHP1icXGzF5DbHOFhtbtR4x4nWWqH9UVr7ONwe0TrzMTo82bN2fwSyfy/caPZcnP81JhZn4Of7xX1m77KcM4+D25ZvM3OTwuKX936OC7De/7hSLj13Z9IwLPmSXgseBzID4S1sL/kQlr/jHbH1jb/Z1+9k16vz5n37c/qLCmwJbBzQ/WNmzcJNs/v1F2v3qU/PfmmBx6Hy27lx0p275boIZwU7AhSg8kf62WXW+dIv+9fkwOjY/+3d84Vna9dqxs+uOL0PHZoPHavPrVFA29MTrnxui1UbLzgwtk44Z1ccRok+z4/IbccZ2tfJDXWWZtqXedfRmHhnCdLcsd19mHF/peZ4Q1AAlhbalPho2wFgSQhDV/8MiI1glr/jHbH1j7913tJZnq//kZ/U0OqOem6fldm5I5DA1rf8uut8eJLB+Rw+/DZeeHl6mJ/Du0Sdr0+7eyZ+kw2YMYvTo858VJz8mc29JBsvmnt0LHB7C27euHRJYNzXmxsddF6t/9v+VHpWQQ0wN9wP+hu11vnZ5z4xMTp50fXr4f19nQ3HGd/fx2fNfZVwtyx3X22tG+1xlhjbDmB2p4jbBGWMt+409Y848pYc0/PgYs9wPW3hufMyHEGshUyNq54qr4TOTalWogD84FRnuobF7zTnwm8rtHVEPDcnaMFNj+e/NYXxOZ2TA/A2vvnZ1LrrNr4rzOvstF19m7cV1n/3y7MJdcZ6P1OtOh55l0ghDWCGuEtSAYC3qdmbUsGOt0QSSs+ceMsOYfn+yAtViwyaHbOz8hrPlmUF8lrPnGR6+L/Ya1HHptxcZt5wrCWmw89tk21xlhbZ+4xFwb/71JWAuCEQ6DTJmbllmcmFlLxxn7jBYjrGXBWKcLImHNP2aENf/4ENaylNEhrAUMdSWsBeqIsBagITXchLWAGBHWsnCdEdYyg5C9z3PO2t5Y7AtuhLV0nEFYCwpIFl4nrPnDSMRgbc+2afLf1pQ7tnF3XzloP4dBxvRq+vV4HsivEdayYiI5DNJP465hbQ+GYqbesY37hmcGyUdzesi3C/rI9pfdzxuMGqyZmC3TuOndxszvb5zw1whrhDWfwhh+AJL2NcJa2nikBTbCWhB7MLMW3rgT1vxj5gjWAGH/rL9YPtBKgE8+epR8++mZsnvrVHnx6WPlyMOaSt/etWXSOZ3lG33ePbC5hTWYIBjFv58bLP8u0blLqSbypes7yeIZ7eTz+b1k5xK3RtI1rO18ZZiJz45UQ/3PS0Plrdu7ycJpbeW9Wd1k6wtDAk1MQo2kw8zaPvpZPlw2PjtYXrmhkyya3lY+nddTdr7iVj+IvUtY2/biELlrQnM5omdlObpPVXlrZjdZcmNnqVulqBTMn0eKF8kn3VuUkS/0WkuoTgI6dtzBmoLr04Pk9Vu7yLNXd5TfHx9g2qQ5k1rKkI4VZGininLNqY1l3VODnMZHHMPaLi2WtPn5IQI94brD/z+b10seubStvHpzZ43PQPO8Sw1xGGRa8MgYSghrGcclJXaENcKaP1jsk2oMCpi+HjFYA3j89cv58sOX42XDr5M1i+S4mqIjWNu67iI59ujmUrhwfsmXL49UrlRMbr1hkJQsUVDKly8qTRuXl2JF80vdOqXlx6/HZ78uQmnJHazt1h/7xy5rJ/3blZNmtYvLqcNrynfay9+lWRnJr3HLm+cgKVY4r5x3RB2nPf/OYE0N0Qd395CR3StJ8zolZFjnimqoe8tJQ2tIoQJ5TXwKFcgjw/X5tU8OdGckHcEaMh5PX9lBBrQvb/Rz4uAaRj89Wpb19FNU9XPWIbUEgOvWRO5ngZEA0Mn83IbL1DH1jVbqVCki1coXlmrlCkvtykWkXtUicsO4pnLp8Q2kbIkC0rNlGY2TO/B3BWs/PNRX2jcqKQUUXPPlzSNtGpSUK09upNsHSc2KRaRh9WJGT0M7VXCrI0ewBjBbs6i/nDKshrSpX1K6NS9jYP/mM5tK8cL5jLYQu5Z1S8jHmqkV7TDJXI8BWfq4dZ7yuYQ1wpofiGXlNcJaEHswsxbetDuEta3rLpQF8w+VCeM7yZw7hsvWvy7S/4+URg3KSZkyhaVxw3Jy242aLfnb4SKBjmDtqUVHSx79oT/lxDby+MKjpG+v2pJX/1+/bhn58uNxJibPPT5KihUrIJdO6ek4u+YO1j6e01NKaK9+lbKFpFOT0uaHv0G1YlKkYF6ZdlwDef7aDnJIt0pSQMHt2as7ODMBrmDtL+3tb1i9qIkR4lO+ZAGpVamIFFZAO35gNXnxuo4y8cg6xkhOPa6+yGuJNUKZGjBHsIasa6li+aRSmYJGP9BSg2pFTXwuPra+vHBtRzlcs0n51YA/plnaTI9/Pw1iVj7XVWYNWcbaqpkjelU2wx5hujs1KWUyam/e3jU1JsNl9qQWUlB1hZhm5XwSsY8LWAOInHt4bdPmXK2A9sDFraVO5aIGQND2rH1ioAHYq1LhbZlmkBJx7ln6TEewtktHPCArmz/fQdKuYSmNTxGpWLqgXnv5TUb2ySvayy1nNZXSxQvI4I7lnY6EIKwR1rICZH77ENYIa+FhLChD4gjWtm+YIoeMaCR5NPNRQrNFyB4dPrKxgbQmmjU66fjW0rF9VQMs9889JPvPOygu9nUXsLZjukw8p7M0VGjd8ueF5tx/+vYcKa5gds0VfUW2T0+Bs+2XyvChDWWE3ndtdpmBdARrChbTFDAqlCooX9/f28wLwbA+ZNPO1EyIvJ4CHpufHyyNaxaT00bUNHNusmRqstl8u4K1Z67qYED1KTVDMJUr7ulpMiCdFdzsHCNkl0b1rSIttFfbPpf0GLmANdXPlSc3lDJqEAEYmI/1xGXtTWbkZM08WnDFkC30+ANuEaukxyZVi65g7aeF/aSMmmpcW/bc7zy3ucmw/a3XFp6Dtr59oI+U0mz/S9oBYPdL9qMLWNuhQ2RxPR3br1qKZlRXT1/Z3vy2Lb+lixcLzO+rVr6Q3HB6E09byY6Pq2GQq1RDpVVDl41tYIY+rn1igIG2ctp59NMj/VJipHGbeU4zKVk0397nsrkdzkq8CWuENT8Qy8prhDXCWvZDiyNYe3PpWCmgw7Cma1ZozcoJct/sg6VwoXxmWN/P356rQHKpbPhtsvToVkM6dagmu1xl1xzB2oSzO0mzphVkiw6HRAERwG0fpoWd0QAAIABJREFUza498chRezWgMTr80MYyZFD9XAtrMNA9dciamaumP+zIApRXeHt4ahvPJMF0H9GrigzvUlF2OzLbrmDtrvNamKwR5tDAUCMWfdqUM8NFzf9hhvS5G89oKjUrFTaZk6wYmmzfxwWsKcxj2CwyjjDciMcWnbtXpVwhmX9hqzT6GTOgmgxoV94YzWw/9ywaUlew9pfOIwJkXDCqrl4/KevZfXlvb4G2/tVhyDYemCNauGAeef+u7t5z9rVkPbqCtY6NS8tJQ1IAHzrCsMiWdUuaIbX23DfpnFoMHb36lEa5DtbsCAjM6bPt0CU6tBadILbtxvNv39FNYS1/6lBIN1l+whphLStA5rcPYe0AgDXMufr+i7Nl5ednOx6aFhSs1NddwJpCxj2zhkuF8kXkl+8nGPhA3ABlxx3b0lQ7NBUOdb/rr+ovNWuUMvPXnFQ9dAFrCmcL7z/MZBtvvm6g/Ll6kuzeMlW26TDRXZtThoQiXu+9dpKUK1vYZOHwfyfxMRlId5m1y09saDJrMEcwRfjhxwR/DNWyJmnLC4OlWa3icnouzKxhmCOGgD5/zd4hoG/p0LWPZvdIMU0aM2SLRvevaoxTbsusXadZDmTWvtHMLMzirqXDNVYd5Wfb26/x+Ucza63rIbNWPVdm1hCXsYOqK4ilzN1747aUoY94HncUzcC8P8zLqq9DSJFBstdesh9dwBpicJpCP+bsQTsodASoRdEem4lFIY25k1uaOWzPmeHYbkDEVWZt9aOanS2R38zjQwYbuvj6vt6miA/iZ3UySzsASjGzJulB4OVV98iCt66TuUuvlKe/vEMAA+n3Se7/WWDEL96EtSD+iMCcNcyv6tuntpQpXVjuuGWw7Ng4xaGJDgqYvu4C1tTgL37oCClevKB8+v7pKfHRoX3XX91fZt8xLA3kXjCxqzSsX1aHA6ZkmJIOJA5gDeCF+XuHDG9oso/Vq5WQVV+f4+kIsRipQ0gxfLRKlWLy9SdneK8lPT4uYU1/8D/TSmIYSlOvalHBnCsYJfvDjwp+z6gxQkYNFetgKO1ryX50lVmDcW5Vr6QxSmfo0NCVD6ZALc4fJukNBbezD61tgG7K6Nw3Zw3DZzF3BnNoMEcNmVmrDYD/cwq5KM4C4EVlSPuai0dXmTWc69onB5hhoKWL5zdz1+SNg00soKHbxjczQ/5qVCgsz1zV3mmMXMAa4rPywT7SVDuE0M501eIZNluE1755oLd0bJwyxw8VM7ekVkJ0oSFXsIZOEGSnUcxoVN+qpriIHWaMYiJfKbhdq9UyMVRykFbPRPVaJ/HRv5fLzBpM/pyXL5OjzxgiPYe1lwnXHi9Pfj5TBhzRVYqVLCKFihSQKjXLywU3nyyvrJrrENgIa4S1LPCF8YcZ7RcRWOvZvaZUqlhMqlQurpmiFvLFR+NMZsSNkc4oUDHPOYK1n745x8SoQ7sq8uiDh8v6XyanQJrJEE2TP36aKIsWHG6yb6OOaqYw4ihz5ADWrE4AbE88cqRMu6iHrNV42Of//uMCGTSgrpyqxUcs7ObKzJr+sMIsouQzqvnVVcMdm1HDxH4Y8UplCpke3VgDlWwj4ArWjFlUIIFRqlGxsGComj13xK5z09I6qT+/jB1c3WlGJMVEJn+dNQOst3XRogYVTBGNHx9OnT+j2lqvxVmq6pBIaGj6CQ2cl+93CWvQDIYQ/6HDaX94OC3wr5jbUx7Xiqy49rA0hNPrbMU1smHj37Jx48ZQ901rv5M9y1MA1F4fWX2EhrAv5mEhe4ZhjrsVTuz7P9F5ohiqjWF/GI5sn3fy6KjACIAMZfkvHFVPGlQvKlM0FnthbYRM0Gq9WP6hb9tyZv6xk9jo3xDf6xLW7l1+lZSrVFqKFissFauVk4KFC0jb7k0kf36NzchOcvIFh0vDlrUV2grKbU9OIaxly5pxWRn2GW4fZtZiGCNDYIsQrGEe0ZvLxkq9OmWkbJnCMmlCZ/nqkzNllw5ns6Y7Eo8OYM2CBbJrtWuVkoMOOkieXny0Fxe83r9vHa1+mEca1CvjNnPkDNZS4VSHgmKoKIZBWr0gPvg/FscG1L70zOg02Ui7X/IeHQ2DTP1xhVnCMCOYazzaH3rMQ3r3zu5msjqe3/DMYAN39vVkPrqCtRQjmbIoL7KOseXn8dqHs3uYIYAw2KgcaYcoJTM25rtczFlL1Q/W5cMSENBPLGhgbT7oZ9XCvin60iylNeZJj48xka5K96eYWJz7b48NMIVGYq8zG4ufH+lvOgWwdIZ9LtmPrjJrVhcY9vjbY/3T6ASvAd5wbX2i1WtXzHVXLdNVZi0lPinDZrG0w8Zn9w6V3aMgh0XVUfwIxXwwty92LmSyNeQM1tbMlxMnHyZFiheWmxdfaIY7nnzh4cYf9Tmkk2bS7jFwtvD9Gw3QHXnaIIfDIZlZY2YtCMj8Xo8QrA0drD1H/84wxTPGndrOlFgvocP+sJjx7JnD5C0tsPGzVvj7N3UOUvKMdboAOoA1nGsKsE0z87Fef+UEEycbA7x2zeV9ZeZNg+XXH84zYJKyf7pjz5DYs3kfZ7CWeh47p8sJx7WUqZpd27z2Qg/KsP7cG0vGmiIkyLA5i4/5G7iFNfNjrsUibtI1e1D4wBqnlB/54fK3LsI6Y2xDmaZrQaV9LXnzRlzBWqzRefzy9mb4EcDEPo947FRQe+iSNnJw10ppYMXuk5RHl7AGaFP9oHQ4htWm1UhK0RFkSyZrViDta8nTD/4GrjNrOIbvdRgtljc4e2RtA/eIB8ANcyPr6Xw1ZEdyI6zZawRzPnvoWnMocJQG/LXj6N4LWpnhyLFVNe37kvboLLO291pBG43FwpGFjb2eALpYbL2vFkD643F3az66grVla+6V/od2kVadG8nLP6aA2WMrbpOiJQrLxbefKksV5iwgdBvURroNaiuv/ORqKGQ0YQ1DQ19YOVte+mGOQ5Cdb777uw3vS+jbrm9E4DmT4W2dfkcEYc2CyYdvnyLjz+wodWuXNhkjVEIsWaKQLH/5eLd/GEewFivGj989TTAsMg1w7JhuKkJiCOnFk7vlrmqQsReRxuGqy/pIIa2U2b9fHfnq4zNk0+8XyGVTe5k5f9WrlTQVItPELvb9SdmOAqwdLEf3qWrWW3tQwQNGyfRk39ND+rQuZ9aowyK1seYgaQZJTW4UYG3mOc3NOnTTT2ioAJuSZfxFh66hih0WpMWCva6qZboaBulp4PWD5TgtIIIhj/dqJUjMmYFWPpvX01SBxOLGGMbmSj84zijAGjKzF+vcxmK6kDGqaL5+a1etElnPFB9BgREMh8wo6+bFWc8jkduuMmv2nNDuHKbriWHB+TMOriV/6kLzKFYzun81b40xzJO0+yf9MQKwhmw+itWM0Zhg6CyuKWT90T6X0OIimH+MLH/SY5OqTXewNl+OGjdYKtesII99cqsBM4AHsmtzl17hgRqew1DIIcf0dAgkbmENkDr/1atkyszTZPrss+QpLbry8Hs3yojje5vYtO/VXM667Fh57tu7vLhZ0E3Go+thkD9+NV5u12QH/LNbb5hZ8iSisGbB5D/NGK3/5XxZ8uxoueWGgTJ2TCv59H+n5XpYu+X6gboQdllZ+vwYrxIkCrUsmDfSzPtr37aqVwXRxjJpj64zawpbOzdNkYfuO1SqVi4hKDaCqpkw14eOaCzffXaWXoyZXRDJej4CsKY/tF/e28uAGSaxo+f/rgnNjfkuo3OybjyjianO5soARAHW/tQ5fAAzFEEYomC26NK2giqZ+D+MpcvebOewpvr5VgtBoDw/4jFOK4feo4s8V9b5jli4F8UPtmiG1pV+8L1RgDUcBzKzyKTV1DmQiBXWNUS1yF8XowLr3rlaLmLlGtbk1RGmI2TacQ0UaPOaCquAj6La2Tb56LqyXudtYditM+iPAKwB+K89rYlZT61Z7eKyeHo7GdShglnfEI9faDvuLD7mOhstG9f/memcx82bN2eYMPl+48f7DU83L7pQChYqIK27NJaLbjtFnv/ubvOZS1bPE1SEnPPK5XLI2L6ST+ewXXzbqYLnkwEf+36HW1i79O4zpUTpYroOr7Y/eu8xtL20791c8hdQ2G9SQ6rVqSh5tSjUseOHy5Kfkh8j17D23BOjzPBZ+EVsx06jSZp39k0URATWMNTxYK3kh2GQvoFB9khPyCn5RiCz9sFbp5iKj8V00WeUqv9Nhz6ecVo7nVSbRxo3KicvK9w6i1EEYA0awfy0B+aONBkizPHr3bOWrFszyV9fvhdLdoJcNGAN5hALYJ97WB3TUCFOzeuUkA/u7pFqst0ZySjAGuKDrMc957c062EhPmUVZBdrlcOU9bPcxScKsIb4oNw6TDVig3vjGsV03lo3LT6B2DiMjzGRbuesWfiChpbc2NmsGZZPjRKyjmeNTMkiuTTZOD7nsKbHgBhgaPE5h9X2dHSeFtBAdVobQ2ePEYA1nDtihMJQVcoWNDHKrxrCUOPtL6d0iOxR6HUVI1eZNQARhvGdrRmh8lXKmCIiC/93owdjiz6+VUqVLa5L+uSVPod0dJY1SgE3d7D2/Ld3S+2GWjCrfhW57J6z5ZwrxmilzKJakTaPnHX5aDOE9Okv75ReIzpIybIlZPFHt3gx3Bc69w4tzc7XogJrqBpeqlQhOe+czmb0Gnyk+859eM8IwBqg4gdNQf7w5fhAI401s3a6WuzZGnnHsGYh7Lcfz5Oxx7UypeorlC8qBQvmk/HjOpgCGk6BNgKwhmGPV0zvLSVLFjIFa0Yd2VSK6sKhWDD83eUnKcjtLT7i2zlg/+bZ/hgNWMOcB1SmG6iVIfNrrxqqHKKaHxY4xqR2Vz/++N5IwJoaJGQ/kEVDKXoseVCsUF6Zpkse/PW0uzki5u9iTGTyq0HGagIG8ov5vWRY54pGP2V03axKZQrKnPNbmMIHsfu62I5CZm3bi0Nl0lF1pYgOY2tSs7iuadhB54M2MPPYULoe/8/NwyChIRSkOW5AdZN17KZl/DEHC8MisU7fj1pJE/u40I/5zgjAGs4fwx6v1zUOS+ocR1xnhbRjdsyAqqbSqNP4aFvtEtYsMDzx6e0y++UZZv6Vfe7Zr2fJhGuOlxsfnSwvfj/bvAYosK8n99EdrD3w1rVmGQMMgbTnfNjJA6WCAu5Tn+9dg27m01OlcNFCMvPpS7z97P6JfowKrH2z4kyZo8thVa5UzNwvmNRV/vf6yWnqH7jxjBGBNXvyW9ddJA/r4sYZrbW24bfJMnF8Z3lj6QmBUGc/LyGPjmHNnpMd9likSH7T01a+XBFN3x7jPn3rGta0wMi5Z3U0GTWsu4axyIjVYw8fKbW0kiZ6TebdpT2V2klgY5n8xwjA2msj5IGLW5sFsssqhGBx1S/VeA/RkuwYqnVYj0pqlPo5M0pRgLX/zepuDDaM44Qj6goWEj/70FrGSHbQdaDenqkZJFdG0jWsqX5Q+AHLPADysXYYFu0d0TVljb6D9RFr1DmLjzGR7jNrKDCC+WqjdH6oXSYDnSTIkjSvXUKHjOaTlQ/1cQYjrjNrmOvYtZlmRXQo9uSj65l1+7ZqhcNrdBhtSY1NrUqFTQXW3Axrv+sSB/11uDGysiO0qBHm8N16dlNdnD6/iQ+KHbkE/ijAGoqNXDLzdFms2TQY/1goe+mHe2TyTSfJGdNHpXk+0QCS9vPdwdqDb18nxUsV1WGiWnQltXT/1FnjpHP/VvLS93O85+54RmGtmMKaPtr9kvUYFVhb+fnZxhuu/PwsOWVsG1PoEPUP2rWpLGed1l7uvn2ofPzOqQ78YwRgLdYob9SMSLWqutba6Jay9ueUIWsw1V98OM5kRTDMZtkLxzk02ZqOjACsrVtzvpx+Slsz7LFl84oy/+4RWuWwvADcLjq/myCOsXFN6nYEYO2Usa1l+iW95J/1KQuDWzD7/ouzdb21ehq7dg4utthhlBGANS0QMbJbZTPs8d079kIHMgEzdI0sZALO1+Ftrsx2FGDtpjOaGhhBARYYbBhGLFb78LQ2pjALCkY4M0muYU31c1TvqgZm37itq6cTzK+5SgsfYP4Rhvq50g/+VlHIrP26eIDMmtBCsCyGBQ4bE6zBdsqwGvK9dgLY15L96BrWUA2yj2bSsIA6rrE9+ndDfHBfrkDbQIuwPKJzRZMdF+/7IpBZQ6cRqoai0JFdSgTDHt/R4cat6pWU8qUKyB8KdN4xawyTuR0FWHv11/sMfKCQyOyXZqTOW5svj6+4XYaP7mXmaR1zlg5HzoWZNQyDrNO4utRrWkPueHaqIOP4sg4fRQVIwBjmqD31xUxdl66zlCxTXBZ9mHuHQQLWrF/evWWafPLuqXLe+E5SX5fEyqeja8AgPbvXcpAUiRis7do8Va6c0ccMX2uhEILKjyjbj6wR7jdc01+2b7jYC6YNalIfIwBrd9wy2Ax/BPn/uVqhVtcW+1PhdpwCHHoBRgxt6EBMqTDiGtZ0yKK3YLhuA9RQwn/Dr5NNxhYA99UnZxDWNDPy5BXttfpa2h95Y5Z0rhF6/u+7qLUzsx0FWHtHIdZWorMG25qg7xb0lstO1OtM5yPZ55L66BrWVD9PX6nDrlMXLbbxgX4wV+1NXTgbc/3s80mNTapZdQ1rWCsMc0Ixrw9xQOVQlGF/SMvUv3BtR/lV1xeDfpxVFNU4uYY1VKDdoOvx2fmNiNnGZwebOzpCsAbbL6YQS3IBxNNrBGBttVbHXHZT5zTXkr2usNYhlllJiaGbGEUB1gAdV917rlSpWd4Ax8TrxsoNj0yWuk2qS4GC+WXQUd3l0Q9uzpWwhthc++AkM68Pi4YffHxfAdwaUFN4Pf/Gk8yQSBQbOX7iIU5iFKXMmuUJ28kvWivj7z8ukC+1svgTC4+SRQ8e7sA/RgzWECRM6HtjyQnSrEkFAx6oXNNHi0OseO80BwGKzYakbkcA1iCWh+491AzvQ8ysqDCnD68deVgTrQbpaF6WI1hDDABpjy44XEv3a2P04nFmfuNMBdu2rSubnpHBA+uZDgAbL3tRJv/RbWYNP/T4cf9JTYCZm6b/x9yQ285uZn74MY8GQ5FMT7e+5hmXJPbYuoY1VDJECfFNqYvRwhTdd2FrmapV6zDsaK2WGHcZH5cFRmL1g0V5YbShpZnjm5ve/6ev6iBbFFCcxke16hLW1mkVw/GH1pbW9UtKu4alNBvbVjs/WplKmagGiTmidSoXkaVqwl1cX/Y73cHacDOseK4CPeZifaEQi3lZ5x1ZV7O1xQTz+cYMqGbWqcOxWjixx520R8ewhswjhs+u1cq0iAH+/4xeX6ieeffEFmaNPvfXmbtqkAAOCx1LtdLjg+9cL90HtzWVDQFplaqXkxlaVOOl1HXYcmNmDeeMO2Jzwc0nm9L9GDZq4zbp+hOlz8EdZeqd4+TF1GybjWuyHl3D2uvKHI0alDNFRZLvBzPgDPX1aY8jgrAGI43MR9fO1b3KUKOObCYoqOHeZGsAHcMaYrBbC2TYWABu16w8Vz7SdC0WxEbxjE3aC2BfT/sHTy+ABPzfEayhqEifXnuriaGgCIaEFtThfLV1rlrP7jWlbJnCUq5cYaOvpMclzcXnDtZ2adW1O89tIY3VEGGNrME6R+2jOT3MAr22ot9BaiaP7l3FaZEIV7AGQ4TKfe0alpSKWiyjpQ4zQoZtWJeKXnuEOLVrUFJWP9rPndF2lFlDJggZMxjqCqofzKWBfqAnTz8an0O7V1JgG+wuPi5hTTV05iG1TIn+tqoTFBbBcgbldW5om/ol5LEZ7WWeLvhcVdsi/N9lnFzB2ue6mHpNnY9mNVOrUhEZr1Vp86h2WtYtIV2aljZzQzs1KWXAP2lwlr5DyhGsoR36dkEfGdihvFTRok91qhQ1w0EvOraeiZGNW5WyheS1W7uYDhNXMYpKZg2G/4nPdNjjmN6erjD8b9bzlzrJFqUFHXdz1uxxID7PfjNLHv/kNgNq9nn7iIXFL587XrCwuH0uWY8uYc166i1/XmhGpOH/mflnPI8pWnvUdyfXQ0YM1nZtmaoT+HTYY/kiUkHvd94yRK69sp+ULlVY6tUtLY89dISXTUpuoGKgxiGsrf7uXLnkwh4yYlhDmXRuZwOwV87oLeXKFjHVIMvpUNFzdCFxDPtzFh9HsIY15jCm+OrL+8pHb58qJ2qlTPygtW9bJaVCpg4V/UgXWsdw2olaljWzizE5cXMHa8v1hx0FRNo3KmXWeqqmhhEmspxWGMOwR1T3Q9ls9PwvmKLDIB2VX3cFa+jFRnnsWrom1omDq5siECiggTl8l+lwI6xNd6euR4c5JGccXNOdSXIEawBXFINoqxmjsRqfGhVS9INiB/MmtzJrPl0wqq6pngmos8PbXBhJV5m1v54aJNXUYEMfKKCBjBEWUMfaYZ8ppJhY6DDSx3RBbFyLgF0X8cF3uoA1ZIJwbaG66qM6H+3N27uaBcPRXp82vKZs03mPgJX7NROJZQ6e0Uy/q/ikZLDfzXQNsY0bN2b42gZ9/p9vF+oacfENk96BuXyty0qpovnMAuF9WpfTCqL5zeLYo/tXlU/m9DQZtjqVi0rnpqWcVu+NCqzd8thFKcMeC+SXg4/rI1hbrFqdSlK0RGEZe/6hgvL0gIJkAUja73EPazieuUuuMDG6SYeIYt4ankNMMER0yKgeUqxEEUFBkrTHnphy/bHf4RLW0ng+9YkoQPfs46PSrVU8TVAA8aZrBwjK+yd/HbaIwRqyI1WrlDBZtY9SK67AVL/3+km6sHFVKahV2d5cNtYdiCA74gjW/lYA69yxmhka2qB+WSlRvKABkeK61tqg/nXljpsHyzFHNzPAcsPVutiqii6NCNNkdhL4mgtY0zHFZ5/RQTDP8Z/1KXMa//h5ooH8W2/Q+RAaCwNnOy6VIw5rKhgO6WyYqPk7OII1NYgoGoJe7N91rhEM0as3dzFgNmV0fRF9HaYIRqF9oxQzDmPlwii5grXFM9oZMHt7Zldz3oC3auULmQzSv5qVTInFcDldF4HGUK2UYYAO5oq4gDXVx9Qx9c3yDqayoernLa2ICeBAeXqrH8Spa7PSMqpvVafzsVzB2qqF2sGombTYwhgA1xraAbA5NdtoMicP9DEZNyyY7eIaw3e6gDXTvmjm+tRh2tnxesq1s1zbIYAZOgNsLAC5GCqKgjVWW/a1pD06yqx9p1m1kgpqs7RjCFrBsOL+bcuZaw9DIu35P3hJaymh+6HqqH0u2Y9RgDXMweo6oLWZl4VKh8gSAQaw7lr/w1Rb+fMaYMvtsPbohzcbWCtavJCcfOHhptjItQ9OlJoNqpi5fSjGguxbLEglYztKsHaNdvgXTS3WZ0epffPpmTJ8SAPjlUaPauFg+aeIwdpmTUNeNaOvoEx/etBY/wsqILYz89nSv5bU/zuCtSXPj9Hqj3lN5hFAskQXvgaooQok4gUY2b5xigwdVF/atKosOzddsk8MkxInR7B2/nldpaGOOQbU4jyx/MNhhzQ2yxnEnjcWX0cBFmRxY59P7rY7WMPaRT1blhULHjBEWBtroVY4jP2RP1KHQQ7X9bNcFT9wBWt36RIGiMfvj/c3JglmEuvQAc5gmkyMFFpu1EqRNdV8O5vY7wLWNBYw2KiCCcONWGBuY3XNrmE+lqcfjQ/mGw1oV85dtUw9NlewBk1AG+M0s/avDhtFXDD/8QWFstjr6fHL25ss5cf39NwbO93Xi2MStl3Ami3XjwwRzhXXFeY89m5Vzsxjs+ePQiPI3GJOW26DtY81c1ZCs/ev60gIs+C1XlPTtUpvq3olvLYbcXt7Zndd4iC/fOwwOxsFWFv2y71y+rSjZd6yK2OyZ8iizTNVD8+96jg57ZKjYl5LfLYoLeREI7OGY1qkwNZ3ZCf1k/mkYctaZm21ilXLytS7zvDm9qU99sTHKjKwpt5xnRbtO+mE1pJfk0Po2L939gipWb2kqbY+5YJuOs3Ixci1CMAa5ly9ueQEWf7S8QoYU3wzQpir9a8W0UiusU6XhXIBa5oZwkJ9FSsUld/t3D19rke3mnLi8TpUzY6f3T5dbrpugNRQYa3/9Xw3cXIBa3qBPadp6/zawz9B11j77IPTzXBZzN8DxOKOoaGLdRht8eIF5NIpPVMybSbLle7vm5TnHMGaGqMbxunCqjqf7z0t+7xTMyCouAZjgAIa1jj9ptXX0KN97mG19wJKEoyjNWl4dAVry2/pYobwzZvcUrZruXUYom81A4JsCYaE4v/IpmEhaBSOQAxjjztp2y5gTQ3jzHOamXXD3tRy/TDdgI9PFDYw9A/njvis1TLiDasXNUPaUO0vaTFJp1FXsCZaVv2CUfVMLyx0gqF+FjYQH6zXZ5aF0Dl/GI6MrImrGLmANcRgsmb4sf7cnec2T1n4WnVkCmXoNQZNYdF5LB6eXwuMoRKiq/i4Ggb5m458qFC6gJyjbTDmNOI6+lXb5c91mDrih3hg/ujlJzUy6625rJjpGtZg9FEY4/nv7pZXdGgf/v/MV3fKzYu18//+CXL/G1frkL97dGHsOYQ1hTXE55H3b5K6WsofQ49RBRILZiN2yYY0+31RgjXwBYY5zp013IzmQ4yqVilulg0zI7T+SfZ8NXjUCMAaFizupdUeCxXMJ+eo0d74+75ZNadwlt68u4A1PYZnHhslhQvnl9dfSV0UXGEN8PHCU8d44AEhIfvYtHF52fbXRbkG1nDeyCRO1PUwSpQoaIaIYk01q5u/tSekfTtN82tPSauWFeWX7yd4r9l9kvvoDtZ+UuioX62olC2RXxe+rizrUk02fvwBIViMFhP8YaQwl8SVSXIFa4jBoA4VdG5IHumr6z99NHvvfCKYpPkXtjKZNgzZul0XgrYmPOlxcgFrqpFfdFhoYy2YgXl8h3SrpOs77R2ShUp1yIK00aIaRQrldWuy9VhdwtrCnmhHAAAgAElEQVRmrSZ6rV5LzesUlxOHVNfhfgebawkauuWsZjp/La900aGiH87u7uwag2ZdwBq+F0A/RIsbYf5j89rFvWwRXlsxt6cZeox5s8f0rWLWqLOA4uY6S/6cNYDrRcfWlwLaAdmhcWkznzi2rUEV0TH9q0lh/U07fmA1xxlsd9UgARhXzBsvnfu1kgYtaslhJw+QBW9dJy06NTQQkjdfXjMP69jxww2sWThI/mM0MmuYp3bNgvN03TWdUqNl/Ice09PErYguhn30GUN0XbrbZIlmI5MdnyjBGvzktzrs8VAdmQXPWE1BDQXrzp/Q2SyRlVyvaJMJEYE1VOpDxb46tUqbbNFrLx+fbnKfPeAIPDqCtT9+mmhiVKtGSbn+qn7y6/cp1TEN6au4sJjfTboOHUR12snak5seMpP1f0eZNZwvli74RJd4QLERrKtmY4Cs2lgtOHKtjkVGVVE8b+KWrJjs8z3uYA2mB2uHnTSkhnTUXv3YHlnMhUDlsUY1ipniIrFDtpJtklzBWoqRHGhKYyNztuzmvb36iB2q1GFoFoYkmWUP1FwmOzbm+xzBGmKA+TQoBNFe44Pha/b8kZ1FbBpUKybztdrh7mWOso6pfxNnsJb6/YgVFsOOHSqL51BFFNlILA+B2OE5G8NkP7qANXu+uH7e0PX4Hp7aJs3wUBTxOVaHSEJDWKMu2TFJ833mOks+rOEYMMR4tpbn796irFx5csM0HUMTtAgUqvliuDHmH6c55lT9Jes5l5m1u16YLkWLF5ayFUopdNTWIX0FpWb9KoL1xI49e5hcNuds6di3peTXMv7XLpiYdAjZCz3RgLX7X79Gimi8ylUuLTPmnGXm9j2m1SGHHNPDzOtr0rauqai597gTPwQS3xUlWHvluTFSrWoJ9dIF5MrpveW3HyaYRFIh7WBr3aqSvP3qiZ63tB4z8Y8RgrVhgxvIN5+eJZ07VJMimkE68vCmsvSF40wZegyVTHwwsgiCjmAN54/150yhFc1CPr34aC8mAI9uXWqY4iLdulQ3pfydxcshrPme887pvkNsfd+7D2xlUSuZvs8trJkfcR3ShmFssQVEMCQSw/026Tw2VPGzpipZP/qx3+MS1mx8sECvnXOE5zAM8udH+8s6XWNN1GC7jI+rdda8c07VTyzQ71I4M/oxa9M5jo/+vVzDWqyeo7rtAtb8YuHpKzUT6bdvUl5zCGvm/PQ6Q1sM6Lfni3YIHWu/PTYgDeTa15P96ArWlq2ZL6M1Y1aybHGdq3aFGep4/o0nSp48eWSEVoNc9ksKaDz1xR1SpVYFGTm2X64fBok5fR37tJB7X7vagCsgCbCEDOWFt54i1bV6Zm6vBjnt4h7SoEFZeemZ0V6nPoZFLn7ocKmpyZLeOhIwV1aDxDBIZNaGDtZqdP/OMAUibr9psNSqWcqkIFH58MzT2+ucreHymg4B3KLlM5NrrNMZc0ewZjNBKC7yw5fj02SO8NrjC4+Ul1VcKNLiND5RhbVMwSnd3zcp+7mDtWT/kMf7fc5hLcm906Hj5CizFvo4HcaRsBac9Y0arEVOX65hzeH1k9W/hTNY06Ii/Q7tLK26NPaqPz75+UwpUbqYWeA5NjuEhbK7Dmwjr/zkal5WNDJrL/2oc/tW3u1lGDHPD2X7sT4dYvOQLpyN52Jjl4ztKGXW/vfGyd5UGeu792xL8Yk/fjVeZmrlddRDSK7PjlBmzcIaAoAAYV7RHbrOGsrVY65WHl2oF/dXtCpicoOUzsw7gjWn5xwGYAhrWdAnYS3ICBDWAow2Yc3LNGSmJcJagIYUBAhrATEirGXhOnMzZw2ZtTHnjpCyFUvJQ+/eYADjJS00Mv3uM+Xh91L+D9B45utZUktL0x98fN9cm1nDXDUUXDlh0khTFfPRD2+Re3TNtTbdm0jZSjq0v25lk3l8SmE3GXCW/jtcw9pnH55ukkKYgrUX0NKO6Et5PqVgnd0neb48orDmBUCLaKBwxFcfnSFPPHKkID2JiX/e62EgIrv2Jaz5x5+w5h8fo0PCWmYG2z5PWMuKiXwnwwV5M1uoF89jsd5t3z0S92K99u9zIDwS1gI0RFgLBBFX1SAPhOvLHqOrzBoM/13PX2oKiNTSeWonX5SybpgFAVSIvO6hSdJNs2r5tEz95XPHOwGRlONxm1mbcM3xpuAK1pvLqxVWUYClUes6gsIiHXu3kGbt60teLegzfExvJ1UhXcPac0+MMpUxS5cqZKpAYumn5ANZusRQGmaJAKxhcWKshzXmmBYiO2dkbnQV3GSHzjvSE3AaRMJa5n8jiIuw5h8fwlqwQYKJ/OQqBYu/Q8PIprUrdW5ZSuU9ayZy5CMza4E6ijKsYW4WCkigaiSqaHpztZI87I2ZtQCgZWYtC9eZm8waIAgm/6r7zpXmHRtIhaplTEl6C2uLPrrFzGcrVaa4ycAh62ZfS/6jO1h7+kuds1ezghkueveLM0yhlQpVymqNg7ymyAhi8dIP98jhpwyUoiWKaFbyxqTHKSqwNvGczlK+XBEZPrSBvL7kBLOcmB0C6TRJFIXS/QjAxt8vkI2/XZAFk+tHnkl6jbDm/3eKMKz99PV4ueu2oTLvrhGCccnu1uyLdmYN5hEl2dc9NTBNSe1kQk+UYQ2FWNZqkRFM8k8x2gGGLxEGPMKwhgWzXesHWo0yrEFDxw2qrmv1lZQ+rcvJ3MlaPVML2iTzGsN3RRXWAK/v3NFNsEj9gimt5Rtd69AJ0EYY1rD2GtpoXGtYq88snp2ItibgM11m1lKgK2WdtcUf3+rNXcPzL6ycLTctulDuf/Nah3PVbDVFd7B23xvXGAhD9UcLqaPHj5BK1cvpenSzvKGhqKxZpHghufWJi7397P6JfowKrH332VmCapDNmpY31SAPHtZQFj14hPy+aqLbJFFUYM0tsYaEvAjDGjKOqFKDO7ad9AhEGNaeeexok+rGouH16pYxE0UFGVuT7UrmY7Rh7bmrO0hVLeFfr1pRGa6L+qJUe9JNZIQza1jUuEH1olK7cmHp1qyMLLmxc/KNZIRh7ZUbOkk16KdqEbOW1tf39U66fqIPa8Ola/My3sLQWC7jf7OSv+ZalGFt0lF1TXvdql4JsxzE2zO7ObrO3JTuD2pz/3lpqHRvWVZqViwsiNGsCS2cVId0D2sWiKL86A7WMJ8PRVcmXT/WA7Mr5p8rA4/oajJqFsRuffwis/TBrOcuzbWwhiWw4AfX/3K+TJ/SUyqUL2raoOpayn/4kAYy9aLusuS5vVUik+cdIzAMMnknm01mPMKw9tuq88x6YhhSeqkKDZVrkj5kNMKw9o3Od7xwYjd58tGj5JtPzpRZtw7V7Fqyq/pAh9GGteev6WDW8Jl4ZF254sRGcsrQGmlK/AeZiOx4PcqZtR8f7it1qxaVfm3LyyPT2kp/fdxsljtIYoYtwrC2RGGtUplCctbI2mZhaCzau1szSdmhizCfEeXMGrJEf+j6WKt1KYhdS4bJX5ohQXYkzPllx75RhrUXr+soF4yqJ5/O6ynoAHjhmo6EtZgsF7L63RT4m+gi9Wizu2rHkYuOEcJaViDRHay9+P0cadm5oVSsWlYuuu1UeVjhDQtfI5sFUHvu27tl7tIrdJhkI7PPE5/enuthzXLJr1rocM6dw2RQ/7pSuVJxKVasgPTpWTt3lu63QTlgHiMMays1hYvKmbV1cfGzx3WQPr1qC0r9JzW2EYU1FKoxsYjNpMVuZ5BdQ4Zy56YpCYhftGEN5hHGEcPZsA6bizk1UYY1xASL9WJRX5hu9HBjSFJ2mOcsf0aEYQ1D/NLoR+PjYghbVGAN575NY7BN56hh+5dF/eWZKzvIR7N76lqHe9fPyvLfPsaw7+97ogpruKawDmTK+SX52oqNr7nOoplZM7pSTWHuI7bRTrsYSktYizasAchmvzRDFw2vZRa+HnRUd3n11/sMkAHYzph+jBQuUtAUG5mo2belqRBnM27JeIzKMEibWYNnjk10YImxvzTbtuK90+T9N09O81py/DUza+GNeIRhbfuGi+Xt5SfJJyqofxVOPnn3VNmuVW2SI6bUzGUEYA2LqH/09ilypy79gO0P3z5Veulafq1bVtYV6fsEAtiuzZfIIwsOl4H96ki7NpXllBPbyNcrzkxzga5eea7071tHvvr4jDjiGw1Yww/8vwpmvy7ub0wRDPaTl7fXOSJtZJVmj/D6/prBeN8fBVjDHJD/dKHn3zQDgsWxAbGv3dJF5p7fUlbc09O8Fu/57ff7IgBriI/VjzWOTyuEPHBxG/nhIUdzjGKMdhRgDXF59eYu0lbnpi2c2kZ+XTTAZEIOOuggKZg/j1yumevYhen3Wxcx55+Vz3IPa8MNlL10XSdZqFlqHPNDl7SRNvVLSpempeWJy9pF4DpzC2u4zqCjv54eZIAfHUOfzu0pcya2kNdv7eoc+Alr0Yc1ANfTujg4yvdjHh+WPcBzgKSpd46TI08dKLc+frH5v824JQPS7He4hrV3XztJunSqLvB1SfXLGSQJMv7+AwTWQLjIjGzTLJHzkpoRgTXEBCAGGMH2Hz9NlC8+HGeKteD/uGf8R0+BKmSN1q2ZJN9/cbbpMcD/Y/fH+9etOT98Zi4CsLZE1+JDurpxw3JmEfUO7apKqZKFpHOHalKwQF5ZMO+QNOcae97Ynnf3CMmX9yBdmL2kdGxfVUqWKChVqxQXXNB23691GGV+NVvvvbH3Ofta8GM0YG31o/1kQPvy0rZBKdMje+qwGmZ8dn4t4Vu7chH5ytFcIxi2KMAa4PW04TWlctlC8ttjA+SuCc217PFBUkD/7sWL5JOnrmjvDGZTSoq7Ld3/62P9ZUinCtKybgkDbWePrOXpp0aFwvKZDl/LCjAkap8owBoKQNSqVFhqVyoib93e1QwLRdty1SmN5LgB1aR08fzy8yP9Mo3TOi1k89n8Xlm6Y3hu2A4W17AG8Lj65EamvR3dv5rpJEJMamm8mtcuIaWK5Zev7898ziPev1M7UbJyR8dC2PhEoXQ/hhDfclZTqVC6oCy7qbO8q4VXEBe00wXyHyQzTmjoFPijDmsvr7pHntehfi/oYs+v/JQy9M9CQvIe3Q2DtOcYC2Gx28vW3Cuv6gLj2C/2efu+ZDy6hDXjl//Reg86JSbINwd7u9SkRZYhLKv7H0CwdsWlvaVHt5oyeEA9uf2mwQbanAQuArAGsHpg7kgZpLFY+/MkWf7S8VK9WgkpokMgWzSrKJ99cLoHFRnFCJA3/ZJe5j1FixQQFNy49OIesnXdRd77sN2wfllZMH+k91xGn7XPc65hbcelMvLgRlK/Xln54qNxsuJ/p0nBgvnkuiv7yba/LjIaOvqIppmOOd6uHQKtWlaUrp2rm4XZt2+YIu9r5UjEonnTCvLn6kkmHgc8rL02QsYfWluKFsonN4xrIt9r0YySRfPJqL5VBL3clcsUlCmj62VqbjDkBoU1ntd5JEH3F67tKH+HnNMVBVh7SDMhMESnKMT+qaYZUNKsdnFZrtm1rs1KS8+WZQRD/jKCDWRLYKqey0J8ntP5JqsXZW7YM/p857Cm+jlfC0AUKZjXgMeqhf2kTIn8cmiPSjq/qLNUK19IzjuiTqb6wRDbpVnUz/Oqn43PDsowzhnGJjW7FAVYQ9EQZNFsEZqeWhCic5PSZsgaIKSkmm4Y8AzPQ2N8tUJdYY0x4hx0H9apYugCE65hDUP4ALMju1c219id2iGCjOPrt3aRVQv7SrmSBbRwRnMRjUVGMVoxt5cc3LWSDNNCSEF3wDE6YDL6nEyfi8AwSOgEnUP92pYzYH/SkOoG1p7VQlDjRmhnkrbVa58ckOl5fXlvL22HOgS202irVmjGLizQRh3Wrl94vllHrE3XJqY8/eKPb0n6nKylq93DWjKgK97vcAlr+3jYbAetrAKZ334HCqxp9mjMsS3Mj96MqT2ldatK8sTCI/eBiJ+/OUfuu+cQuXdO8P1xfX/6bFKW/mgRgLW3Xj1RihTJbyoa/vrDeTpcr67JHE0c31kqVSomRx7WdJ/Y7D23abLw/kNNT2a7tlVkwvhOmv6tpmtu5JFJ53YWrHuHfbf8n73rAI+i+r7SS2gJLYRA6L33kkDohN6kd+kdpUkvgg2QIr0rIk2kiFQLKshPxYINULBgl6qigsj9n3M3s2w2W2Y2gVn9m+9732xmZmfn3X1v9p537z3npwmqN8Eo0+33+hpM8cdsBmu8/yro1329KkC3b5qsQnFoGkTTSC4if0yTgf0qS2ydKGEOsqd+EfzmzhUiCx6Hc+hS0/bK/h5q8wljauN9U+SfDtZuAUxULpZVutSP0B/nHTMr65g4CoBBx6VjbB5pXTu3V+fvG6RzZc+cRucknVFfLQ3G1ntwArw6RPHOtetx28EaVuxHAWwUzRuitWmfbawn6dKkkAVY4ZYjrWTekFIaLbm8p6nHfnGln2lcvuziemzd+PIer+NqkwSv7U6DhH1qgtCgLZxsRjdIcJAyxT3y8hMAHnCse4BUpGnVnF7B7A/PNVYSG1cbeHvNaOb/lllnSgwGsEZijNS4fwIPEorkgmNN4h7a6MzTsRpZI5FGgu/WZT58i4juuC6F1LaRiFZueLCCbJjguR3AZ/G78HYtT/vtBmvfbmukQMSR7thCo43FIkPkD9b4oXGBZFK3ol7B2sfr66r8AcdOCrQyBTNrCiXTKN0bswiu7vU8Xz3ZRvcFAVjb9VAVjeaTlZeLZGSi5XObC0KUN2CUjWmRHvtwpKUMae2IeHubX677ezeNtDyGgh2sTV4yUJ/DsS2rSl20Zl3r2hBB+g+s+QJy/xawxsjcrz+NhxTZWFONGYOe/NDE+/4pYA0G+B5Mh6c+GCq/ITpCWs0LSNFz79DubZ1UG8F4+BDQZPTSypXNLYyauF/D7/9BANYmjY9WUPbl6RHy3bnRkhWq66MBuuTGdJk+ua4UKZRNrv443mPfCFDbtioh5dH/K9C3EwCa64i0jRxaTdIhykJ6e9rgnwrWBGOFEdiW0Mf4BZOGQLZ0yRwaNSRAqxOdT9q1KeEVqP8KuxUrml0G96/sPEfD5Lju+PtrSqZMaTWS+cl7g//RaZBMCSpXKJNG1+g43gfGRzqDV+LBRxsANV9g7SZquUiQQMDH+bZoeGmNODHq5N5exSr5LxadJLvBGleXuWpdqWgWOEEtZN248kihTaG1arTXw/2LA6yll8u7PUd8+P4vEW3qDKeK9qETxHo3d9vwf9Y0Mc3So7Pl4rgnOG4zWGNEsQrqsDhuCF7pEHKFn3U1tE+nenkA1nJ4BWt0NBkVYZSJ9pk3uKRH2xj2shqZpa2CAax9si5Wx83D/UvItF5FJQUALQEcwQj3ZUiXUhj5SPDdun3nrJUk0CgL4ELb+jrX6jG7wdpFjJdQpD0uHFZaiVcYke0bhzGFfp4HEQsjawuHlvbZb5L9NKwMXSTYkuyaJCbx1P6paZCbEOHPlD6VaqqRGZNjZuUD5XQcHFlYUzMiPkANrbfv/sLOJjKzbzHNEqiIlPfXFtbyOtcYxfu3RdaoJbbu1dmy/b0Fqr1G0edDdz0d8j+w9v8BrJHErh6CASyb8dcYFNjzXGePfnpiHBKkYI1ok+iUN0yxbKb5nTgGsgw/4IrEEMeP9JW8SAkMDU0v7745QE6DIdFT+wICyX//lrBOK7GB4qNFrmFRm8EagcMgRIcYFROk/L2wozMiIkgbOdQL/0+TRfObanojAa2n/hCwMMWvb09EnhBpMs6hvWNjogBsHKl+/1iwhmjYY7MbIPUxpdabMY1t/AO15OoP46VXj3IaZVu2qJmz30b/XbcjBleT9ACuo4ZVj1exd6x+cLWkds1Ite+sabEajfzH1qzBGbq3bh5oYWWUxSPKSBiiZFytJTvdQaSnsW5kco+iXn+4jR90Op6swXkd9TjenAVjv/Ee439fW7vBGu+NDiTTjwhEqWNUAlpYv+5rKh+urStlsYJft0KYV0p6o690OHNmTSMz+hTz6XDy84z3+LKL85jNYI3ONGuMWFv05Mgy2kdG2Th+XgYApZNNnSxvfTL2kyCBkafD3lIBXYCL8R6nDVyOedpnN1jj/d6APRil5hwhUGMaJFkhH+xWRNKkukdB7W3WQ89AjNd5uF8JjRR56mdS9tkN1ghEuTDE8cI5RpvsQ9orKeg5v0LSp5T3fAAR7TvG4rbplTVN9HtEL5Nij0TvDYLI2luIKnOhaBLGTNvo3ApKTz0VK9+AGKoTntl5sqcV1jYmuneX+cGIHKn9KTcir7Xyea6v63g6FkyRtYNfrpadINIgMNj3+QpZsmeKPLF9guw4STr623T1noADozu7P12qNPZP7p4sm9+al0hMm0Lba1+eraDP0zW87/v3gDXa+Nn/zZXFuyaprfZ8uixRpPLZ/82TZwmKTTJL/lsiawyGbAE5XXHwJXARkpT/UybEeGwMupBnwtX39P466MDaFPkExA1xjYvI9El1lFCkMTrLTjOVjXT0NwDIvHVIRaABQGZMqSvZwzJ4jS55e7+p/TaDNd7jNNiGmg8fojatc4fSEg6EzvQ9gty2qNcqWyYXgO2DHu1EQpKe3cshehQmP37pqL9y9HuK1mZlBpkGo0qXAExyZM+gZBum7GIAWpvTIHmvV38YJ30ARosUDpXmcUWUKOXsx8MlPDyT3Ne7gnMhwFu/GLVt3RIrkaidyJIlrXyJ9Frj3LPQrqtVM586XikBkv+xYA0/5EyhCcvkSGVkvRr/Z6pNOsw1Cq2ewiqrpx9n131M7QpDrdLRxf7Bmuv7/L0OBrBGlsxSUZn0+UNnewXY1+j0FMmbUR0mMh/66wcd7arFs8pskCj4O9fScbvBGsbP28ujJQecbD6fufJv1BkxwkFR9Q/X+I4Ysb/UFyOYfXm+l7otF4fTkn3wPrvBGu+X4PUTzCMyiC5AhIikPtxPAeO5qBW9ZLIWj2Px3ZXeoydWbWOcbzdYE8yP0wAejBxyXo1GiijB6/YZlUDsk1YXTG4iCusPqJ/Gc6sDFp8ueYl0G/21vLUZrLHffx6Ik3Z1wnWeca4xms3IdNuYcF0EMEswMrRNlDStlvNfC9Z2ffyktOpZT6rFllWQ0Gt0a1DVp9Lf6tJVisjWE75r1Z59a65UqlNK0qRLre/JnhsSSDO7yoFzq5w1bstBgZ8+Y1pZdWimc593gObKEhncYO3AF6tl/BP3yaCpnWXiogGy9Z35HvtHUDVmbl/JnjsrbJRCUqdNLRVrl5T1r85xnv/Kt+uldtNKEh1XKRHY9WarfwJYY6CDmWzff3G/kvsp3jD8Xrft6y/1xgJLKlk4F5kmKJsxsrM8bQ3f0vc2yMAamR7rgGI9G9L6WGu0c2tHnTSsPxrYr5J2/p1j/ZyOs7fOHQYbYIniYAKMj855Oy+g/UEA1t5FlDE0W3qkP6ZX+wwZUFlrsAhMWHs2ezqcbJd6K/d+GmyJBGwUzz7/2SjnYHpsTkNJB0KOFs2KQrMt1T8OrDFaOu/RxjL34UZy8dsxzlTG65horO+7aTKayigtpQ+e39xRGSVdbUjyle2bOsjIIdXk3MfD/I5H1/c6Xj8icuuGePq7du2aXLp0SS5fvpyoXbp8Vf46PjxZnH46AXSSWC/DNBujsJwr048NKCGfPVNfP8eXk8RjdKTOgoWOukiWHSEfjrjdYI1948o1yR+2TqukaUNMo7p+qJk62W8ujVYH0pd9DHswusZ0L+P/ZNnaDNbY7zNwkB2U6xUBJGLUHiRieRTj5xSOsZ/+7EOnkyyG1/Yl7/jhZwcDWDv2ZC0Fo5QzENQP6XcPKnZNZ0REKFnGgo955O/6doM1pkeP61JY59gvex16YbxnEoEwWsTx428MOfuIdFzn6yTYJME1bAZrvBeS67yBxbAdkFWhPt+vLzZFP1vI6rHlZPdsLGDjuZTgnr30ncykrBE0c66Vc4IhskYq+l4PtAFbbwpp2jFao2JZQkOUWOS+Ce0lQ8Z0MmhKJ6+RHoKFhu1qKlBr2T1W+j3YQd+bKnVKGTWnp/N9y/ZOk9Rwwv9tYG3/2VVSoHiEAjCKYxctk98juKV4dnrYslDJSOkPG7Xt01AyZQ2RgiXyAuA5wDDBWo0G5aRW44pIN13tBHHegBr3/xPA2oeIgkXkySx5IzJp1tb+XV0ViCX276bK7wiWFCoYqmSIno5b3xdkYO0bqIWzbor6WAQbA/pW0rxPMvD9hMhRrpwZoSaOAmofQITIlQj4Muqx+Nq6UTykPrqiZpvBGvvEvFgSpAwZWAVgK0br9wh0721XUh4H2PIlhM3VAAIavr961Qi178v7ejjtRJCyYnFzJS8h8Fu7Ag6Fa//9vbY7soaxEdekiJKEMNpo6d499c3bWOP++GPWx5n9YI1Ocq0yoVKtRFaNFvHH2ekUxTuRzv+9/Phb+UG3em4wgLUxnQopXTaZDhPcfxDYx242SI6N2ArZpQKIHAyg5RwvwWAfjNlgAGvnUUOVC5TrTDn+0ynybD9IM8az3WCNtYisUyPTIZ9Jxn0FzTYIwBoJRsgEumeOCxGNAfhteDa7fzfBANaY4lisXAGp3aSi7Du7Uh5aN1KB2+NggjwMIFe9fjmJbVXNK3X/c+8vkuzh2aTDgCYKHEhnv+uTJRodIhhZcWC6go5/K1hjauOEBf1l+Mxu8vyHi1V77ek3Hk0AtGiTwdM6S8bMGTQVlP8TaD22aYxkypJBGnWoJZRJ+LeCtVPQ2i1VIod0bF9aF/HbtCyu2X+efcwpsv2ZDhbSHP3gjt+DDKyRYS8NVjLeeKmPGqFCudzSrGkRdYovfjsWtUJZHEjVpdbKs6H8dTwJx4MArM2ZUV/rzr4964iI0QYEDMyXNQscCNoI8BhtunZxQgJQw2v8DID82sFe8jVITCzZ2G6wBlscfrGHaqOR3axD/KsAACAASURBVJLpnJbu3xNgS/Z99oM1rsxO6FpEi9W5auv+A2z3/3aDNfb/Oaxks+Zx0fAyQWcfu8GavNpCpkLfiQyZL4EgxQnUgsB5NMau3WCNNmHU4zGkO5J+vy/S10juwJpHahiyfbrBwfBn3PPd3toN1sheyXq8jKhNI8sl04zvtg18fl4QgLXPNtbXmuJeTSJNR9F89imZ52gwgDXWp0UVi5COg5qqZli7+xpJthyZFXgQUNRpVlkjPQQlniI8m5ECmSVbiKb4EdwZ55CMJFfeMKmOSBHr1Za9+O+LrDGqtfvjJXIQqZBGv40tjxmvqcXWbUQLiYjKKbsBZI1j3DISmQYpkXM2jJZXvvt3RtZYQsRAEAMa/vxtVz/c9XXgvmiQgTVGQrIhvW/MqJqyaT3o5eEoLVkQp8LPz2/pqI7TXhBqBN7hJIA0w2E3CdZu3bolv/32m/Tq1UtGjx4tv/7yjfwNgJQc906dOdbQHNjdLVmulxz35LyGBbBGG508eVJiY2Nl69atcvPPY5gEAZC+GN8NtyBZWbOspVQom0tz/KOiskr92ALSvg00VuIbCUaSZwIF+n2aB2tMiaSN6tWrJwsXPSl/Hh2abI7xCdTAlC0I1qIc6TV1jVo8r8DxNtoZ0NXb5YSbBWtGyuj69eulTp068vqhHaCedokSJsExYRpkGy3oB9kMaPyfxwo3a6sM+7y9PMYyzXWyOVEW0iBpozfeeEOio6Nlw4YN8tupZ/G9mkud8nW/H6yuKxXBlknR8DlwuN3Hz6mn7Bs/vG8rYI02+uCDD3SeLVq8BPNsSLKM/XObGmj9J0E/641ILx+C9PLMIK5hy5E1rdBOvux8J4+ZBWvGPFu3bp3OszcOJ888I5glkVFuMInyN61kVCYIreeS9qjRYiOhhq2LSRbAGm3EtmPHDqlVu7Yc3jAK0cIAhLjdnllMCX0QmpdpUEN9b2y4poySkMd4DpHc6U/URt7JceLr2mbBGn/v2ZYvXy7NmzWXQyee05omw+k3QEFAW0R5KkBHjbVpJBTJGREmNRuWl/2oN3v69UckDDVW7fs1dgIM98/Ye2a5RBWNkGqIwO04ufj2fQG4TVzUH5wJqWXg5E6yeOekO5oGSfv8+uuv0r17dxk7Zqyc/uFEsjJX0tbUmRuM2jRGCfecWiatetST8Hw5pFy1YkrI4vX7gC2mLR8CvzylTF81TJg6aZz7wullUrpqEYksFC4bjz2qkcw7kQZJ+9y8eVNmzZol9957r3x3/ojc+m1isvpzRiDjsw+HqpQVM9VY9sLSmmOQzQqImNDVRw3odRCBNTrPNMJwkIjwh42EIoUKZIMw8WjogrVH/VRq1c+6/L3NkRKTYO3vv/+WH374QSIjIzG4U0mvnu3kChgJkwMkfIcCx9oguSCBxg6AWBJgsOjRaCTISI7PcQIwK4PLIljbtm0bmBvTSkhIiKxdOcK5ahHQZ/M+r0+TB8fWVrbMqpUjhK0KWyVHI4vmI7Ma2GcftaU1sPbiiy9C4y2DpE2XTh4fXE4Z5pIMolDf0bVBXmU7pAgtHSVGSUgOwVVu0kNPQC1Jkj/HzfHw9aPveswKWLty5YoMGjRI8+3zReSQN5dEJ4uTRJZDahjRPqnBUsetYR/aKKZsmFdqete+3JHXFsHa2rVrnfNs1czO8hdq75L03WL89IEcAdkyPY0f2mlkO++i2HfEJm5jzSpYe+GFF+LnWXro6JVHFCOJNsL9XEGa35qx5ZVqnXTr7m0tJCECkSVILvtZAWucZwMHDtR5lj9vDjmOus2kghGSibQDUUYVkPBURUo2yXj42mjUSnvpDpDPmLafRbDGWuPJkycjBS8l0l/TyYHHoH0HmYukzLV3VsRIWJbU+nymZiX9Iz6f+Qxii8Ri2w87fLNBmu6v2xwy8z6rYK1Tp046hgoVzydrXnpIUxMNp98dRFn5f/rKoZI2fRr8TjraQ2tHClP5QnNmkSyhmeTJPZOdUSJP173/0d4KxPIXzSMd+jVC9GipghECPkbqeF2CwVR3sGaNfuN3330nERER6je26948QRTL031b2fcCwFn5miU0RfShtSNk3Pz7dBGpfI3ikjtfdtSi5VNGTG/XJNNmyYqFJF36tBptnAHQZqRDrjo0S0FfRIFckgMppbXvQM0awdr169d1UY0EJzVrFIX/OzxZAdQFMKm3blFccoG47+I3Y2XG5Lpqo4wZU0tIxjTy/JZ7bQiSBBFYo3POND6y9q1CXdrMKbHy3vH+6lRT+2vsqFryhQsrX8DOvBXg4elck2CNg4pt3759kj17dl2N6N6lDMgqkg7YNqxuLVH5s+gAYkEtddYoV8A0UTayaTJUa4uNTII12oYPJk68sWPHgnkxtWqYrVvRylI6p6c+chwxXH3twoPyybuD5cTRfvIL9NMoZ8D9VtJFeX0C3ytgmOR7+ZoyAEwh9fTZ5vaZA2vGSu3PP/8sM2bMQD1nOkmPRYzHBpZMFkeSbH3U+PLWSBySFCfDzA+9t3PMgjU6RxcvXpRz585Jo0aN1EmKRA3Mm0tqy60krmqTuc/QQfNkI7Lz2WUfs2mQHEO0z48//ijDhw93zDP86KyfUF5lB5Jy/9QH82QXYx/Tt5JyfW9jw+x+s2DNdZ5Nnz5d5xlrhOZC+y05ABvvl8Q9FD4f0AKF+1MrKXENKekNEWuzduJ5TJ88+HgNte1Xmxso8USg9V5mwZrrPGvYsKHOs3zQZTy+NGnzjP2hZh8jbJd3N1WiI44rAlgSjnC/YSOz3zuvR71I2oR6kGTc5Nbs+xOcZwGsGTb65ptvpEOHDuoQk+mU8iZJAbW0BTUajXnlvn1jcW21U4L7DgB0Bfp+s2CNhFr8zf/666+levXqDtBfJI/WP3lLT/QGGtz3E+xRR20qIj89wQI5a+1w1E+tlqePPipNO8cg2jbeGQVyf6/x//7PV8rExQOlVKXCINiIQgRqgRPc7T29HAQlHSUPUgBJsHEnCEYMn5FbLhyFhYVJCoD+xvfWFoKs5AC0S/dO1XE5DLVpZLmsXq+cRBbMrdHER56+X5kuV4Dx0rCJp+0zxx4TkrDkx3fX8/7WmnbKe2NbDcBWs1EFyYyUUjJCmv1e+d4zl97yxLmWaB/H0CeffCIlS5aEX51CalSLkK8/G5k8C/DgIaD8FYNFXPRntl9BBI1qVo+Ud97oB2KRvKrdSx/SnK8XaPaV+/uCDKx9+ekIjabNno60kOvT1RgaIXIhc7BqIL6fuabcur62eh3n+RbBGgfW3r17MfGyY5LcI71Am0+WyqSEUl9CTdbg/lW8NuqMsc/Oe/YEOu/UPpNgzZiBfDDduHFDU0UJ2Lh6sXFdG8uAyrWv/J5fP9xLGtQrIFmzpJM8oOx/6/X7pEypnMoy+tcv1ibaz4hUVq0SoZOVgK1uTJTsiRcPd/1c86/NgTX++BuO5IULF2TSpElgRMUKKyI81P5KqiNJR4mOwHMzKkv3RnmVBOEMWPyWIqIUiF4RxWmp2fb99kbqgC0bXVY+QyplIE6AFbBm2Ons2bNIqa2LKGEKoSNJavmkOEm8b1Kvk5KetNeNQS9OR3IxbE8dKLMOttF/ns+6HK6Ukw1wx4wqchAr78ZxS1sLkTXDPj/99JMMHkwx91QSgpX5Z8ACejOJq/5k89s5q4r0bJxXIySfYfwswfgh1byl/sDWZBRlNPMbsGeSZXTlA2VR0xVr+TrG55oFa4Z9ONc4zyZOnOiYZ4g0Lx5ROslRSOpkUc8wJXTWmAo5qXtR6d88v6ZDso9WxhHPpWZf3fIg5MDrzVMqqZg9QY3RbytbK2DNsBPnWd26dXSeUeLjHaYDJ2FhhOBq3fgKUr5IZiXSaAgtMI6pKsWzyGvQ4bNiH/adEiSM0JGBlcL0JfOHyFfxkglWbKPnWgRrho3Onz8vbVrGacYC5S3IKpsUGxF4noQg9uQeRSA2j5ohsEAyKssUSM5hK/26hXrlQwD7SliC5xDH53osJChDKexn5Vo81wpY4+8+f/MJ2CpULK+ArQBqzTYcefh26iFqzDyBBF/7CDwqxZRC6mMFTX3kuQa4eQm1Vq7/+7oOjxHkMYLkqYaLem0rDswQgjd/10l43Bx1vwHY6Dfu2rVLsoVmA2BLIXEAnExZNAt+En72bXsSxJLNkhpyz3/4pNbjte5VX221BvpxIVkyytwt40z1jQCS92R8lmFvRiI3AiSvPwI6f4Aw47ivrRWwZtiIgK1o0aLqV1P/9lvQ6tP3M++HuQMix/+9upWTihXC1QclIzgDIgyQyI3pMnNqrBQvSqb5hDwPSf1M/+8PMrDG6AXF5Ihif7uQPMagMDZFkSmozXTBwQOqJE1/zSRYMx5KxsDau3c32CxD9Ivv0bWsRmoCGVj6HoJX1GaRJXPX1k4gXWmiKwBE/qSV5zH/X/7tgcprnnxroHyLlFMyHJL15vOPAqGkxzUtgjXDToywjR83XMWsMyLUvBp1Z0Yhp5W+8NzXANSos1e+rIOgJmeOjHLm5DD8eBZXttFd2zpZss/3X94vOcFEehTEN5QAKICVlmc3tLN0jYR9MA/W+ONvOAAXLlyUOaOaqEPD1DONsAFMWHVmjB/jH3c00h9+kh9kQbpfXqTT/A+pTdRcKweBWq7aG+ea2dIhopg2wQgdb9ah7IJTYea97udYBWuGjc6dfl/iauRWxzgfHEk6boE6SSQ7GNW+oOrOhUKPjuLNTLFkXU1OMPxR0Nn9vn39T4e6dIFMqh0lr7eSJlVzyKBWUYE5SRbBmmEfRthGDoYGDFM6kUJFh4/CxIGMIVKBt6yZS6O9TBcl6+Hby2IkFGOA+nSk5PdlD/dj3wHkU5yd3xnT4wi4t0A2wf08s/9bAWvu82z2iEaOeQbARj00ilsHZqMmqh9WoXAWTYkrFplRwRoXQ7o0iNBxRHBqtk+8h1l9i0k9MHEqWEOULn/u9AFHVqyCNWMcnTv9njSt7phnBGyMZAcyz9iHhVj8oLYjtdYojF2/Yg45udqoh0wrX6Duz6x9eB7ruZiCyPHJxQOmdzNLwMo1nOcGANYMG/34/tP4jimInkK/Z8pcBLI4QqBGmv4sWMjkc4iLvkyn7VAnj6Zlc3HEytjkuX3iIqUzBLUF6cwP9y+hrK5OaQmLgC0QsAa8Jsc/OSwlEcWiWHwUUg+TkhJJ6n5GoDJlzSjrXpltCiD4Ag/Jf8wcWDP8IcNv3Lh9rWQJy6x+Y5OOtWUXiEAMUBTIPc7fNl7B3xPPTRCmMKZEZGrKssFqr2krhmga6Yr9M5JsP+Meja2/ew0ErNFGn358REqXyq5jqEa1vOrjMxASiG9NH43v69qxjJYZ0RemlBVTHz+jPwy/etzomlISjJC/IUMuoU932592389rvnd8AHSN79f3fHxisN6n+3m+/w8isMYOMRq0flVrpMOlkU4dSsmRgz0hkj1YWOj32cmhcLiHWgZxWza21zAmkfA7SIcjEEgSpbsFsJZg4v19GYQg3YXAgUi9S6cyChoDGVR8D1XSo/Jl0WsxFHwUhY8ForJJdK188hVAqe8vPuHAotB4FdRyLQZxifw1Hfm6xWQoZAGsgj79zADAmmGnP6+9IRPGQOsEP9whIWlk1dIAABsmWCvcf9Eiocp0uWF1GwVuZBNloWj1qnmlK2x/00LkMRjAGh2Ai5euyB8gGJmH9CwCLDoh/KENyJHEiuqIdgW05mjbtMqyYFhpyQOiiMtIH3pjUW0VO57eu6glJ4Bpg9mxinxihQOsURh595y7C9au/HBGLiKdiiCCTgBpwY+iP4GkRFJ/Lg1q1Wb0LiaH5tZQ5516dGefqS+VQaxBunFGgJyOnR8nh2CtXKHMGhUlWIuDQO2Q1ncPrDnGEDT8Pt4k4zsXQt9SaHRnDRxBy4AN42cspA1Ym/bMpIqyFFFUkmVc2NlEjgPQ5gZwG2+x5pFgLSeuYYC1qNwZ7jpYuz3PhsjjZHHEs4hAgtpxgcyzo0hRYzTt0NzqynRIsD6pW1EF6GSFJMnIy/Nqmh5DwQLWrvxwWudZixqOeRYJYH0MfbU6zxiZJdgj8PgdWnt94/IpEGWU5/Nn6kk2LAIQmFiJ+gQDWOM4ugYin1/3NpHuDfNi8SgF5kca2fdINcugltFlLqB1rhchJ5G6zucqbUIBcDJERuZMp+LyZp9DHEMU1mbNMsEax3YlPM/uJljjb/7nl07Is/+bK6VQB8Zndb5CuWUVdLwOfmkuGuMOAJbtmwamwlxSrGwBYf0aNcE2vPawsz33/sIkAR33z7P2v3mwRtvwj4D2618+lTnrR0lm1NxxDDVqX1NrysyCIPd7fA6pouGoTcuZJwwgEIuO2DJ91CF0nU2KlM4vez5dmmSw5v65/v63AtYctrmlEdq/b5ySU+8NlNIlc+gYqoYMKAZlAs5cg/84Z2Z9yZwprcyZUU8KwqeuUilcy17oY0flz6r+pRW//Q+UzJQumVOegi8q4FQg4d2UB5Fdg88y76cHEVjjTZOAo2L5cI1+8AeOLROMRoZItixIaXt5fw8LHZwqW6F1ULBgNg1bEqwRBNxNsOYcWABrf1+bJi/u7ALw4BCz7t6lrAouWx1Yx4/0VWr6mNr5ZcHjTSQ9tOlOQCj72fXtkG6ZQfr0rGDJRgRrjELxWgRrzeOKyiCIkN9NsKZ2uvEm6sImyv0jaiBVC8XTGVLLU2taW0qJvPnrJCUV6dEVP/BYCXEFa/yfaagNMFnYZ7MTJVjAGkWxb7w5XOsvZt9XHNGRlEjXSiFPDCllOSXy5uEWuppKB4A0/ivw409WPwfZQQuwIIZrswJG7AZrdJCu/vgZIg4tkQLVWNMWuTBCZ4apPqxbMbsCzfOGtS0gRZG+9vv+OKRUxmhRPxkQ6Tg+AgenQHgGFaw16yQFA1i7BBv9dnozxkucjAbDZWrYhxG2jROtpURytZ8afW1q51abrkN6J51ROpAcT1y1JxhliptZ+wQLWHPMs2E6Xh7qe3ueLRhqPSXyKGQx6Ih+vC42EVj7AMA/BLZnPZJZGwUDWHPMszM6zy5gnjWqnEMXDQnYGFm1Ms++g0gzSWo2I0LIKJsrWPttX5yUQArj1J4OcGvWRsEA1jjPfj8N1tWXmslvELHuGJtHbcTFrMNY+LFio+3TKysT5OmnYrX+joQiCmCxOMQ0SGZFcCyZtY/dYI2/9fz7/PK7mtb3LOjxqZHGecI6qPWvzrGc7kdaeep8hWROr34jfUeSjYRA/4saYNzffWSrfxRYo43OA6wdAnidBTIQ6r3RRowgBpoSSVD0MGrTylQtKkVQlzdtxVAlCGmCaxYvX1AW75pkOnXRHwCzctwqWKNt+Hfrr1OIhk3UoE6xIqjxg31qVENKJCSpAi0Fot5zKYArXou+9XPPdlDOA5bThMF3P3Kgl2nfkT4ms/qKFcku61cilRLlXXWi88ukcdH/bLDGTi1d2Ewem93Qazt/ZqQlQwUK1kjf6bEhssYVDyNMbXoLsHYLYI0DaO/zXcA044iwdQPpyOXvxllaCZgwpraKPnNAEqRlAFMmw6wC/bmJY6OVJZKEGmbBSKBgzaN9EFm79ddZ6/aBUW8BrN36bZISeYx/oDZSIlMpYFu9FHpFV3FdRBT99gmArCOisgyJMyXUFaxdRZotV2Aotm7qWpho/LykgDWPNvr9EdjnukcbXbt2zVmnZtSrObcAa38dHw4HwFE4zzRIsoIZKZGsr7ICRmqVzoZao0j9kXcFa3TE65YLkx6oY+Nrs05AoGDtllLtk27/dvvzvdly6fIV77ZAfZHTLi6vGVm79UprtQMdyWbVczodSUY5rDhJ47oUkhL5MmlKnjtYY1SJjiTr9MzaJ1Cw5moX52s4gb98fcyjDTzZxXXfNYA1OpHXwXY4rnNhAH4HqxwjbLxHc2OIothhWufI/icAa3C6mdJGIWgrYD8pYM1pF5cx9PdrXeXyxR+t2wjj7q/jw5zzjJEHRrI5zxhtMzvPaEemOzJiPRjprj/saCSlozLJxG5FEHGJQ91aPgnNnFrYb7NjiNcMNA3Sk43+fO/hJMyzVo55hnRDAnMujDD1+E2ksZqdZ0yXLhyRQR7oWAgpggnBGgEIn2+syzNrH56XFLCWyEacZ+fftD6G8Ey6dnoT7MP51BxkKU0RySLD3+0IGxcyzMy1vYjGcfx9gfo7kqW4grU9SDNn1O2MBfkHfmagkbVE9sF8+/u1bj7n2S+//JLot07gQxGs0UlnI2AjsYcB2BgVY30Wj5lx+slIyGjagIkdvbR7ZcGOB01fz8xnejrH6I/79vBXa+XH375MZAff/qMjsqbXImBbM0KyMiUSETYCtl0fP2m5PytRb8fo3OJdk1U3zugDa/QoX8D/zdrceK/VrbttHP+vltMkGLHqWytYc/iGn74/WMqUzin3AGRVqxohX4ADw2pKJP1ClgG9sKOL7NvZVUkOSSZCbLIKfigz/eR3B/+FX1803n/8V4I17TxCg2TbexGAhrVmK59sIczxJEGEwcBixdEOBKzx+saXnMjZ/iNAsHbzNljj9Q/u6RZfw3aPdL63NPo8zhSA4HtHDq2mg5KvGS10gjUAFQ3dFsgqVwAAzQ6mQMDabRu5gVqArVs3AgRr199UuxPQkm1xIlYfHCmRqWX54uamGS4PvdBdI6i06+jh1UEykla2buwgLZsXBTV3atm/q6tp2yQFrHm10TWAtb89gzVq85G9z9W5Nl4zDZKRNUaOuArNH/v5iKqpI4kI20OItpl1JBkd4op1GFZ6yVZGwhI6lT/AuXxqYgVJlzaFLB1VBk7SnQVrrHNx1LrcBmp0CP58d7amfRp9N7u9DLD298sOJ5JOyaXdjaVVrdyqb8X0Ia5E8/PMOH97kMLJND/a5mVEP+g4ErRR94nX6lQvj6X0rEDAGvvgyT4EW1e/PupxnPiy1UU4kYysMV2NY4j6TA92hYZTfEokV+xJ6+/XPhg/D/UrrjWKex+uJsvvZxpkGhCLNJJnp1ZUW80bXMr/dVxSRwMFa95sdBNg7dKFHwKwEefZMOc8u4F55pp6TE05symRvLelo8pqCi3r1pj2WCJ/JqSdZdVoCdOYeQ6bX5vDVjwvELDmdZ69NyfAeXY6wTxjJFtTjxHRYOrxGybnGfvD6CW157hwwDrOysWyyhZE2lg3yxRJ6h2asY1xTqBgzaONdJ5ZXxThPLt2CmBNo/l8XjeXXxFh6wEiHgISRthewLzx9yyifb54tr7Wgw4Ek+hp1ODRvsswpj5/pr7ULhOK+tAQuYYopNF/f1teMxCw5tE+eFbfBFi7dMH7osjVq1c9gBSRz5AGaQAyOuwkvWDEh1GxSKRErjw4E8fNgTUDYOw7u1K1wvqMaSvdhoOAB9d8eONoZVO0CipePLNCZq0boUQcJB2ZjXTEre884RU88h5c++MAIQ4weghg7YffvvBgB0c6nzfQ9vVVRNZw3UPxwJWi00xfJG19w7Y1LNP6L9kzRcdfn3FtkwWU8d5Ya0ggTPtuBFMkWSW535O9E9rEsI3Dbqcv/g/2EWs2QhokF/jpa7ERaJUt40jNrlI5j5z7BP4SatjM+sI8777eFeGbZ5RvEAxJ8L74tEV+ToL98aDM275/LVj7+rNREl07n67ScdJ2aFsSBBqk0kR9DnJJrYY2AwVrT4H9hVoLrZoXS9haVpFWrVqjtbLY4pDrGn891FSxLow6cuwjG0lHrl+ZaGoQkHwjS2YU8wPAvnP0dmSNBC21akRq3ZqVARUIWGPUikAokX1or5YNLdom3pYtY25fr0VRad60iLJD0j4Z8GNuhtSD/eYYefKJOK0PNOzLLesF5z3cUCevFfsEGlnj9zl5QvTtPhljqUUp2KelRxs1b95cmjVr5rW1jCmo4IMAhI0OEkk92D/OEQIL/hj7+8Hm8QtYEWcqGx0lkotQdJU1NXTcyTZ3Gau4Zq5jnBNIZI0kFKzpMPpjbFvElvRqA1/2ada0gbSqHe68XstaubQGhkX5tBFB1vugTDfu2deWbJs9URPCiArrp+hkkVyEDnf+XBnklEWmwkDAGiN3JDkx7HJ7m0uaN44OyEbNG1ROcL1mNXIqKKV9NJIBR9mXXYxjHB+14CwS0DIFjlp0ZTh+YK+apUM1FdU418w2ULD2J4hgWAd22zaOudEqOlKaxTUNyEbu86wF5lnWkFQ6hph6TJZIM/OM53AcPQ92Q+ryRYEMhGOpdtlQeXZKRY3amrGNcQ6vFwhYOwfH/s7PszD8bjvmGZ8nZlPzOMYHo3aTi04cg0YrHhmiEgVG381uAwVrP2Kh6j7UzCUcR5xnMQGNIfd5xuuSUZbzhH0MBZnOawv9p8DyO180vIzWKPO5kyb1PZqCnTs0nYTgWUTmTDNj0bAfzw0ErLFetyOi5Qntg7nmZ57xN83dX2oNH6pBXF2p2biCs9UA5XupyoU1MkL75CucR51/T06/p32MDLXu3UDSpHP8HubKGyYLnntQfUnSyJOcw9P7vO3beOxxTT1cvn+aUPQ5NEcWmQ2w5O18ApGxc/tILZc+Gf3jvibNGyWyg7tdEv7fWho1q++0Ty3Yh7T4uSOz6/jh71GTe6NVnNrbPbnu5/29+NkKiY6rDB20UNhmgtL3ewJQru/z9Zrv7TiwKUSwy6tdqFVHKQDu9/Q+SijwuzDs4rqtH1dHOC4S2sCPn92y/m3/Cj41+Qoqls/tfIYww+rC+QdM+dUG2NqEciKW4SxbhMh3PAg0jgWy/VeCNRbiNaxfUFP8WIwXWydKwdolkEMM7l9ZoyIn3x5kyfCBgrXomvlUfJJfWuKWGvsCabevlQrXNX7Y+GBirdl35xxsMf4GBPXCKPYcnjsERCyl1TkaAkIQFleS+p4MkVaKFwMBayR7YREmBToT2of/pwrMPhAPTw1H7/b1Ujgnh3jUUgAAIABJREFUHW3UshlqFxA99Gcf6qtdA5voeWhv7Hmus6bW7kVYm2yX3E+iEStgjTWOhQuFyvFX+2o6JldunoOivb/7oEg5V2gS24jjIJDxg/fQ3q4NzlEK2Ib2YatZKhSrtebAGn+8SW/N1f3mcNirFs+mP8QLQTZyOb72yPiBN7MNBKyRPp5gKEGf0L9UqQK0EceQm33I4mjYh/Tpjw4oaToiRkbIpx6sIO3rhEO0N5uu/D+INDZSgltxkGi/QMDaKaQ3saYnQZ+c/Qt0nsG2zmvA9i72oZ3ojJklGyBgYzouwUwVjJ+WNXPLfETULmFcWYnK0j6BgjWmGpKdM0GfnP0L1EZuY9JtntUCGGVk0t+8+BVAZOdDVZSZkJEJ1vSxMU2QW4K4q0iR83cd4zjH3PbplWRmn+I6/pjaS6DDFELjHE/bpxEtv9vzjCmjnu7FdR/7Qz01zjPKYTCixpRs6hvyGcRj15Ga6/oef6/53iIRIcI0aJIBkaWUKYT+3kd6fZLJJB5HgY4hD9dymWt8bnMhxsxc43jh/VFihWOvbvkwGdgyv5I5WX0O8XzLYA2R9NmIpHMhJrF9uM+6jVLBT+Bz3mgp8ZqptM5nNV5PXjLIo9PvDgTIBtl3PJxs0NJTZ637yJYKSJ77YJGMfqQnRJzTyMjZ3b2CCPfr8f9n3nxc9cKWQ3eMYC0sZ1ZhZMvTudzHqBoFo43+JNoG4Bd5slEKlzGUFVE2Vz04b/dm7CehSzTAEq9LUJuvcLiKZFP2QKUPAAa3npjvtY/GdYytgrVBTaVGAxewhqioJ7D28rfrpVWPeslqH4evifEX76fTr2aqsTGG6E++fri3X1/N1Zejb9u0UWHFG/37VpQ1y1vC37sXtWuOduRgL0tBo38lWCOLSzoU9C1ZGKfONIkuGFljUd730FDInj2DUrpbASKBgrXTHwxVxkWyLiZoz4ySLVu2om2x1javw3U66LU2P9VOnl7bRsqXc6wAkPRk3UqkbpkIrxrnUJOuH8K1+cEISfKV3AButRFV27mloyVCDg7SQMAaU1Jf3t8zoW3UVm1ly6bF1mxj2HLTQ7heO6eN5iIKlhnMoJx4pUCX6sgXnup74l2fBoKS6mobI23WOV4A9Lp0LA0mnjqmbG1MYEoIUKz9j0sOkMfQ+tUf/aeZMsJHBqEE40dt1BX22eTRRk899ZSsXbvWc1u3XjbNbqoODZ0a1nKQtjlnNoeN8mPF/viS2uZABH58t6FwncXr/PEmkGD9iBJC4NiaceWUwc6KI3AFjvsjAH50unm9x+HEn3rKt04WV9VJ78/+uLaNj3WWteivV1t4sdG6VQtxncrOa21GSh4dolR0tgHUGIlkypY/x43Hjy+NliVIBWVqKZ3hP2Af9osppPuhj0baeyvAmLTd/L5IdkIHjSCQqZa+7oXvoQiuq2309ZQKsmHZLMv2oT2fWjgS16uo1+QYolOdCXUvnGeMrJrSNoMNdsysosQQHCMcN67jZwNEt19ErY2V8cO6HgLp89RZg72ZekjGRF/24TF+BxQGdrfR5hnRsnbNKus20nnWxHk92uhJRNKY5kkbMTJGiQszfft8Y33JhMj1K/gOXfvBVN9Xod/HaJIpe6OffD8/kyCRfeZr/T/+tev13V8TNO56yNM86xLYPFuZeJ6NBLssFyA5zwj4HWQzvnW7/oREQ+MqOWX1GBBCxffRYH688kJTqYKUSNZsOY8Z5/jYMiWQVP2sl2Rd5hmkDlIKwt81CBg5ZhOMI51nD1kfQ5xnC0bgWo55xmuSRGVy9yIKeDiO6gBwfbfdf4onNQsfwxz96fnG+r3zPtkffvdMi2S6Mb9ff/0zjvN9XCTg/KWtX1tYS5aNLuN3AYvfJzU5E9iH/fIzzzZs2ODht26rLFr7iExeOsjZJj05UBqDJIQ1RwRtjBq56ngZAMHTlkCpSJn8iGpVVH20EbN7SPbwbMpsSOBQOaa01Gtd3XRaJT/DKljj56yFVtkUlz4Z/ZuydLCsfGqJBzv49iOXrn/CaR8C1/EL+knBEpH6HMoQkl7GP9HfIzDyZCPuo/4ZQVnFWiWlYm00bCvUgnRDfKvRoJyyQ3p7v/t+K2CN79124gmZumzI7T652GrBmjmy1apvvWmBi8/YXjZBUqklMpg4vwjUxt1fS0tqDF/O7xapjtR5ZuZaenBBEI9wSx/daE0bF3GWZPm9HvzrfyVYoxPMGiXWc9HRvg3WpqlGQQ5En0g24XS+/eSK0pB0sreCvp8EFdQlWwdGlt/hdPsyMgERHW1uk79mbbKTQIODiTVUTGtkfw0g5uveeIxhXYJJgggKNn/y7hA5f2YU0ignyXfIs6VGmpV0UYpEExQ/s66tCv8xSjcL4n++olgOG/Gek7dm7Rbyi3nvBOdM6eSPP6NaH70zCPsne/3e/rj8IIBRb3njpd5Sv24UIo955OV9PSD/0MvZWMuWP19W1doza2t/34Wv4w4beRhDSaxZo6PHGghGNZpXz6WaYmRyPIrCftPU2aBsbo10QbI+upNAECBQJ4tsgfxhN37k/W15rqfm632MNjiK8JO/Zo0OLZ3qvNnBvoqHdxOIyV5EDZvZPtEJKpgng67uu/ehT9NIdSRZz+R+zNv/rrbhOcb/3s43znHYKKF9kqtmjc5izVLZdAwVyZtRKPrNz/N1T3oM46djbIQ09cD4SLtXLJJFBrWMMm1ro6+GTVy3/u6F5zrAS0IbJbVmzTnP4KQ2I4EGnEjOM2qK+Ztn6yAyXK9idkRBIGqL95EavQHEnm+37EpQQzbI84EKNvsALO428zrPkqFmjbY/vrS2phhznnFM0LHn9+J+H8b/FxB5pU4hJTGYmswoEQWwuc9oz4ChlIssz0E83nif/633z/T3Xtoo0VxLlpo1B8D+eH1d1dyjjWpgzn2LuefLRsb98hnGcfIhaPuNfdxyfG5ExJQah4wuux7z9dp9bhn/+3oPj3kbQ8lRs8batIefGq2RLM6X+m2qA6gtlYNfrTYV5TnwxSopXCqfUOCZZCOuYI0goXIdgLVW1e44WLtzNWtrZN/ZVdJhQBMFsukypJVx8+9TAW9PUSx3UGX8z3ONtvWd+bLkhSmQT4BANsS+KQJuHDPO97fl+R1NRtZ4LW/24f6k1qzRh2YQhNJfqZBqTG4HAiVffqMn/41skOTI8Nboe1vxH/9EOUzLZsVk97bOSt3fu0d51Ua2gmPk9yCi7mfnWQfFCEq7NiWEuljNULPUoW0JBTLzH22sSPkECDU8GdjTPqdBXfUMXF+bAHuJrmtRZ02MP2WDdKDsCSBOYdiWsgRkmCFY4r0679fPfS2e31RKQDz8F9SoJbg/sEGOwAClqDijZQmO+bgmP5dAz4hEUfiZ4Nbs+xOcBzZIuXnW6LW1bTwbJCcLFen50KZavAOosajTe2Hn10h5ZMohQ+B8H1dW+Nq9ZcIKyVqEtRPcsw/b3JnzHhGwsHi0DdkgSY3tqZFSXNkgAdRYb0agxn5SPJgMbARw/n54v97SEOmOWaUkSA644s8UO7IakvTAaIXhtNNJYpTMaiqbvx989+N0ONz38f/r75MN8qpHO3iyjbHvCqj7b73SSh2Lo0/W0lo8An6yQjItyl/qGqNnjArQFkyfYspPsciE9uH/rO0i+LvT9vFkG90Hsd5fzx+zbB/a6doZBxvk+a0NNKVK51m+EAdQ8+Fg83MZNa1eMpuOH9busd6xBN5rjB1uCfqYWqmU6xYIarz2FZ8byLFARLFpH8c8G4a55JhnZDrUeQayC2rImZlnJGqhnQhaHfbNpLZmLZ82pCvXAePq5B5FTUV9Aum/63u8zzOyQVqfZ1d/JOuqY56RcIcLIpxnzaG7xsi1v3m2Z041jTDRqaJtCWD42rVx7jG99SMT0VXXvgb62qONAhTF5jgidb9gDBHknFxTR58jBPwkBSFpii8b8Tk+CVG44phbBVDnyDFUEFIhrvOM845EJSRh+dVCKm2y2ocg7nWyQf7k9VlENkhPfwYbJB31h9aN1PowRtRIwf8CgBqBgD+wYBwndX/zbrGSJwp6VtBWGzarm0bWyHK4EOQXGTOllwGTOlq6ptXImnEvnrfWddZos/O/nNJ7fhH1ePcSqCFVlBG1sfP6AqitsgyueG/Um2uJdETW4PF6/R7sIE06xUiVOmVkHWQTPN//Wo/7rYI1b98p95+5BDZIq3/xbJD0XckzQQFr+tajkV1FkjorQQtXX4+6zNsh+9UZZUb0RZnVNXNKrLwL9nWzfrpxPfWvcS9O/xq+NgMtxnFz2yACa44bniLbYCCCmArlwiUiIpMKWjdtDHpp0LgTkZqmcL9TDniAYO3WzUtA+ZPlgZHUEEuhfWSU7zbrpHcgQtuQrIIEK8MHV5UYCF9nRph2ICjo+b/RGBErCNISRqSsDihzA8ZPCiJtTrAG6n6rf8oCBDbIL08NV+Fq/jiVLpUTUcPBTiDrq09cQWFUjfnJ9esWkEoVw+WlF3vIqwd63W4He8mHiNAFDESTbUz5Bmtk8zPAh+uWThXZIEkBztQh2ohaX3QgjZV+/sj7+jH+cUdjGQBWMVL2kxSCq9okHujRKNLZeGxC1yJY8WeKju/r+fqspBy7/l6AYO0H6qy1wup8TQdQg41agWSEYs10sv3dE9OyHkTfaQ+y0pE8gxpJrvahpAE12DSdMUAg4e8+/B5XJ9I6WLt0yaGz9sWmeiCXcUR9yhTMJJ+sj1Vw4g/s046MmHGMRIHsIHdoWumWaPzklXEQxGZdn13jh/ZLClgjGyTnGSUIDEdZUx/hePuzET+baaFGHRaFvqmtRSIU7jMax5q/+ep3HCRx/F1/PzCwRokMstJSI45kIgQhraG5xxpYM/PsZ4yjV5+opQQifAb1g0gzI2q8ntGYnnduUwNczxGxv9O28Hj9AMEaddauUWcN739vZQwWMBwaWbFIfSRjKhcCPH5e/PfJcbEKqaGsUYurlksXz5phcS7hcyhSn+VMZ2R009f17uSxwMEa2CARzSHlfuZsIbqw2rRzTMCiz0+9/oiCtVwRYVK0LAhrEH1ifVZm6JMVQurg5rfneQQc3sBJMIA1skESqLW9r6ECK4LO8Qv6J4hQebt/T/tfOLVMajQsB925DNIGZCyseSNYm7ZiiOTKm13ZOF847aDw9/R+931WwZr7+43/AwVr1Fm7+etEWbmkhaYpMjOP0lbkvwgUqFH2iuUyTIEMh7ZaOmCPd97oJyURSMqO7L4jB3taBFom/Ga/vmXQgbWpij4PvdBNYmOitB4rPHcmALfc8vicBkrpn2ygwq9xvBg4ALD2999/y/U/L6CWihE1iNAC/W9YZU3smambjRsWlnyRjho1rkBF5Mmk/3Of0YoUCZXlYLGxFmL10tdAbBQgWPv771vyzZcHpTpIUti3YkXDFKgZIWxfQC3BmEBd2sonmwujj4FO1gTXC8QGft8TGFgjdf+PhwYqyQVtRApnOpCO9BRzDg2dAF3RRa0CHYFeYDukY8l9rs3hRJq75p1wBAIBawS5dCKPPhmjMgR0sltgpZ9EBf6cI6MP7LfRlowsK5WKZVGNJFfb8LVxjsNOgUV+jM8MaBsgWLt48ZKcObpOqhWHXg/GEFfuSezAMcT78NcfHmf/WetCAWPKF3gdP/F2Cqh/SQQh/MxAwZrOs4MDlLmPNlJRdYvzzOgzARpZ9KxQqxvvNbNlDSVJOPx9b96uFQhY03n2/WnUCUZjnqVVMEtiGbPzzPVeef/UnXtpfg2vYMP1fG/9uGP7AwVrWBShztrJNTFSCKnUfBaRDfR7RKaN/hhbT/fOY2yca+9ALoTZE1xQ8fYc8nSNu7UvILB2izprJ+LFnjOqfVivlhSxZzr8616ZLfVRm5a3AFiSs2eWfIXCpUW3uvL0G49YjkLZDdZI5//FxQ+lbV8HUCP4fHBRf8upjwYg4vbJ3aDuxzNt9CO9tJ4vAnbqN6GDpo7OhI4b0ytXHZppGtTaCdZon5s3PpV1y1sIM6YYURszqmZ8RM17yYxP/w6Zd4ufAPs8/PQlYBTft6urSmOdBunIV6dHSDmQyzVvWtQG3zLIwBojHq8CtRrq46zHIqMew5lshxEpuQAmRJ/G9usoJxGYWARrBGp//PGHjB07xhlRW+NSo2YWhBB4XPpurNaoUUutQFRW1ZD4+esx4toMMWyz1012W1oEa4a+yPnz56VGjer60C5WLEw+jo+ombo/TDCOjafXtNGcZdbebVrfVl8zhzlBwznHj/Q1HXmk3b//YjSYOkfre07+b6A8+lADDbkHPhatgTVDN+vzz89K4+jSmjLEqJiDTMTa6vw1pPnRuWNBOhnSvtrcQP/nPvdGR9OXQ+HuDLD2zah/Y9TgK1z/8h6A5gBWfa2CNdqI+nT79+6SPC4pWWQltNIH6o4ZdmDK32cb6+H/ps59xjFur+2zZh86WkpQAiDx1+EWusLOSCfBjpV7VLtbBGvGGProo4+kSuVKOs8I1IyImvt36e1/koiw7xw/X2LssLnaxPU1I0hW+sVxYrAakkCBY5O1T4GMH96/VbBm2Ojzzz+XRrVL6Tyjk6wRNURCzPblzNP1pTw0wsoV8t+qICWZ0SNv9nbdTzuwXm5ar2I69j5Aah3rDRldZw0l55rr+WZeWwVrxjzb98JO1O85Uh/JBOqvRs31Xv4CQNsOooqnJ1WUjWibIGFAVlj+796o+WiGyZHXZxojr00GSGqacV5RQ3I29Cd3g8TojwMBgNoAwJpho+MvLgUrZQZND2Xq43caUTMXAeO9G3OJ0UqywrJG2djnuuVc1AUU2MDVzt5ecxzzOcQtxxSjnN+A1EdJWLDP2/u87bcK1vibT79o/dblGuFJiXRXArW9iOjQ+XcFF1ZeL983XSj6TAFqCkZveXu+ApLDYIpcvGuSrD/ysKXrb333CanTvIo8BaBHyvt6LavJk9Aps3JPt8+1lgZJ+/z+++8ydMRgZ0Rt3BO3a9QCtRPTTZn6SCDKmkADrNFGtF3GzOmF9Pq379tz+qNxnPdx/6O9pMeo1voesmXGIaXS6v3xfCtpkArUbt6U9euWIvhBJtKUCIbUkOtXJ5r27Tz5lvSbu3UuqwEipiq+imws6hiTIVL+nApyuhgpjhKkX38af5dxSJCBtcvfj5M84SGybkWrBJEhkliQKCQ96iMO7e1m2kg0POvgmCtKp5tsk68d6i3fQMst4KiLBbDGAcUJN3r0aAym1Moww4ia2dRHT4OJ+1jg+OLOrs4cWG/nBbqf9WsBpwpaAGsGUDt37pxUrVpVHciyUKD/FMKG/O6M5rcf16dJ29YlYN90kiWL75YZ54wYApY6XN/fdXnO2uWtVLycKahnQYCTB2FxpuQy3E4dOH9kNZ4/wxpYIwg5deqUxMbGaroRV2qZgmc29dH5IwsGQqYq5UYdCNPXfDXWa01BPY0pBxU/8AQ1g1tFyXNIybkKFrcOdfJIFshI0NldB9ZEU9dxcTasgDXDOdq9eze+n3BNGWoT7UjJsvS5iBbNQZ2eL7u4HmOKHEGX074u959gH+xDh2tar6KoQ6FNW2qtUhh0lnKCYZBiwATGCd7j7VrGfgtgzQAh7733nlSsWFFSoj6IQIJMnQ5yF5MRVJCK3IuUUI4NVzt4es1zxnYqbPJ7bwHAUh+ppVGyCU47GUK71I9Q4oRIRI9X3F8uIMBmFawZ86xu3brOefb2cuvz7CsQhjDiyPRZf40ppGZFn4+hLpVkEvlQo8R01IaVs2vNKaNarKGchAiVpTGEsWQFrBnzbNeuXUgPcsyztphnVhdEfgNbY7HIjPp84DPCV2NNJBkIzfSLc4ikSHwPATb1x9JBE4//c6V8Ru9iJsejC+ixCNaMufbKK6+gJCFKx1G9Ctm11tNsdJ99fW1RLWX59TS33Pdx0YWLPqZshEUQSiP0aZpPn11kdmQaKrU62+C34WfUQpu5jus5VsCa4WRv27ZNsoVmUwr5Zl3raETNqoNvAAZuSQtPLbPYllU1RdD1GGvimAbZeXCcJRDB+6EYNrfG64N47Xpt86/NgzXaiLXrw4cPV/swZfHBRQMCTn10vcfVh2dJWsgYkFlyz6cuYA02HDoDUaSQdLIJQM71Pb5e0y60LxvPOwSiGMNmvt7nfozXMQvWaJ+//vpLli9fDqCWAf5YSpk4rraQyMOMX+fZJ7sdxGE5EbP5HMEjF7CGoEDfnhWkcsU8Gjzydx1vxxl4+mfWrDEqsre7tABNP2vT6ASXL5dL6TfJoOJoRaVG9bxaiMyoiDcjuO4nQCPoq4q0unMAeqxlyp3LkRedNyKzRvBczzf92iRY44C6ceOGE6hlypRBIzy8L8sDCjbaiGhRAgFqiP8l+N8QXMZ2FMSqb5osYOT9vPlKX9lO3TB8zj6AwPqxBTTlcte2TgosTduGUU0LYI2rR998841UqVIFq48ppVTJ3KpAbzn1EZ9LNkzWt5lpZMw08x1cQGQ3L+omq1TKIx++PVAY0WTR++an2svSBXH6kCBjkCX7aOTXPFjjjz/BrDqQsBFpw99CWoyV1EfnDyvAyPL7y8qYjoXgSPtuPIdU32bADlO8apXJJhnw0NyE1XGyuJE0oAO0yaKxokzHnfTTzvswAIePrRWwRif74MGDkisXmTFTKiDlKrQV58i4twNYifdnG+P4AujRGZFE4/2etly9Htm+gI4dst4xmhWSIaVUR1SkExgVqX/FGjtP7/W6zwJYo30+/vhjKV++vM6z0gWz6Uq92dRH5z1g/KweW870+NkOFj8z44ekLrEVwpTGfg2uv31aZR0/BNzUkyKoZZTNeR8+xo3rOVbAmjHP6tSpozbiPHs7wHnGPhtRZjNbMzZi2umwNgWUjZJkFYxokuBlFMAJxxfrKKuXyGoZ9FsBaxxHBw4ccM4zfj+BzDPeLxd3OA/8tY/XxQop/F2/V2+vGXWkYDmB/k/PN1LtPwIRZiD0b55PiTh4v97e73G/RbBGGx07dkzy58+PxccUwho11j4a37Gx9fhZLuOakg9j/DyjjecQiXwYafN3TY6hp+MZNhnpu4iodbF8GYWSLwNbRCngd5BKuYBVl3vydn0rYI2/+Xv27IHcEOpl8XvGGrU90DAznH13J97f/5sRPavboqqKRWfLkVnCcmWB5lc51GWVd7bKdUtDfy219BnT1hJY8/fZ1o6bA2v0G69fv+4AatCuywi/cRKAVSAAyNP97ft8pdqK5CK9YY+w3NkkrkuMatPxs+o0qyL7z60yDdY8fUYg+6yAtZuIqK1ZswY1aiHAC6nia9Ssaed69dngA2+DL5weGnQL5zaRPTs6a2SN+s57n++CcZtOBvWrBF95mim/j+SBDBDt2e7QPt4Brba6KPGKa1IErPfdTfmhjnsNhsgajLMTIs5lEFEhoQTZ+xi9KFc2l+aHMkeUjarkDE9e/cFc+PFtFARSILpEsRwaUWvTqrhkRdTlIVDSlyieXQ1281f/0ZVEX6pJsMaHElP7QkNDEe3JIuvWLkAkzEK0SJ36eLQPGy0BOChbOpep1hEMNmbYIAlYCGhZQEmgzLRTAtlQ6LZRv43bT8GCk8gGrvfm/toCWOODibpPqfBQKlGihHx6cj0Gb2C5xuwvI4LsEycIX3trBvumv359+M5AdRopAcDrNm5YSCoDuFFv7btzoyRnjoyq6+HvOomPWwNr27dv18hs3sh88r+VSC2LJznw9uPpbb86CvjB5o+2qWbih5qfxZonEgw8CQ0xEkpQQJqpWUyNok4WaaeZjuTtvjztNwvW6GRfuXJFOnfurE52XIMacgnpYGadItfPdr7HrH14ngkbMfWUaat9UOPFVCNqrXHF/10QDzDFkrZ6CKla+p2YuJ5+pgWwRhstWrRI51mxYsXk/b2zMYYsRvJwX3dq/HwOx52LIPOGlNLx061hhNZj/gxdKUb/GBnxp0fn6XuwCta42s8MiMh8+eWtVUzhNJ/66Pr5XLxgFIyAhGmjBAv831N7E+RAZurZCHDaxoRLQ9D/y2utdCGFdVC8JjX7ZvUtrmyDTP9zvRd/r82CNWOederUSedZs4YQrU7CPKPQtZGKxy3nhbfGvvvrB49T95BROkaZWD8XjuwBglkeOwBdxFCAfn4nZq7lPMcCWDNsxIgIFx+jy4crpb7zuWJ2bhvnJfNziM8XMptWKJxFI2iMGpO1lZE2HmPWBdk8mYLp7L9xLz62ZsEaf+/ZGjdurM+iVl0aywtnliUJQD1z7HEpVamQFCoZqSyJjAwVLB6pkTRG07ThWKXoko4USUR/AgETSX+PObBGv/HLL7+UrFmzant85QwnUCOgSfp9rAVV/1yp3aSiAliOUy4qpEmXRslYCH6T4zOsXsMsWOP4+fPPP6VSpUoAamlkzOgO8PMcQM3MAnxiX+x2RM04xogayQzTYfGZ+sVKMAXivvQgHCEuIVu5ca6vLe+HgC8NUjQ731taziOjLwe0onOEZVQ/m+zlX3w63NS1goa6/y8Y57ef6QDfr0QZq5Y0l2sXH9R93M927cKDplMXaaT5jzaSsNAM8gVSH1lblBOGYXhTkDK3fHFziQAgpO6aL2N7PGYCrBkPJYZqKZ69f/9++fvmJTj809Tp93hdd9Dj9j8HEFM6zbTfYTszA5cMkwTJ0TXzqTbbsxDrJlimUju12ghGFjyOlUiARdP3bBKs0UZ8MHElcvHixXL27Fmw2b+J+yZYswii8Z0yX7lh/YJa7zh9Ul2JrpUPrJn5EzXuf2w29G0wRvz16QRoWglCGD1jzRojs8ORQikgMaEmYNas6VTDz991Eh83B9b4489G0D9//nx58/j/4qn7Taatuf/A4gd536PVtT6EqUgUW14/oYLXpqmWJmoYyNpGO52GY80USFK3k7yEP/hMCSO9NMW3rTil0tRsAAAgAElEQVQAZsGaUq3DRm+//bbMnTsXD8QTqN9AGrV7303+TwC1HhpZFGbmqr4v+5B+3IwjSecxFGlGTPHj+Z3qRUjJqExwTh3MgGSdZCTzToA1jh/OMc6vxx57TN5//30QH2xWSnHLNsL4IehmnRHrZF4Fe58v+6gemYnxQ/Fxjp8P19bR69I2Herm0e/wW9QNMjK7cVIFy9+pWbCWeJ69hXk2TMFpII42FygyYoGCrKsEB4wMklXUU+Mxf8LxxvfUG2RAZQtlVjDStUGELgBQr4zONR1tandZTac1C9aMefbWW2+5zDNzixXG/RtbAljeL58RHEfsS0zZUI+tFqQODs/zTj5iXJPb5aPL6rOGz6BXwCrJiDWzA3jsuZmVdQ6eQz2b63v8vrYA1gwbnTx5Uh599FH5+thiXVjz+xkenk2sl6VW37urYlQMe4OP5/SzEG2/tt9/ZI3jhHISHEcE/HOhI8n0Wa0JxNwehDR2kqAYdbVm79ssWCNDNH/3T5w4IatXr5aPvj9qWkfNm/NPfbYXEJljTRkBSGyrqmCTXALq/2UJ2oH4aFFyAR5v9+N9v3+w5uo3bt68WbNFyAbJe06u+6Ydtr+3EPWBy2T14YdkxqphMgvEIiRm2QvWSR6jJID3ftwZsGsWrHEM0W88cuSIbNy4UW788RF8uYl+fbnEPphvn5apisw2u693RWlYr6B07VRGFsIX/lkxg3/fkZ9HX53s7I0bFJKvzoyQZYuaKXDbv7ubfHxikJbsrKGMlCn/Ohgiay6ghBEPOsaBEzc4vgA64pMnxEjRImGaxkfSEhYgvghmFxqRRBREtd+dNYeQE3zRJsAaB1SiP+isye/TAh5U1J0jgOIAYG3f5x8P89qI4P82AUYufDNWwkBFSt0x2mwYZAAiIzPL1R/HaU5utap5FQQRnCSwgct3lmi/SbCWyD7c4QRrvidSos/E/U0cG61poT98eb88AgIQThCyZyZq2L9oHlLkTNiH18oBwNq6RXHp1b2cRtkorP0TSF1oK660MD0y0f34so8eMwfWXGn71RFw6qyZW2lO9GOLmqNGlXM40hIhyMoUvPRYPfLUGPkh/boZZ5Wghrps6yeUl63TKmFV+R556kGHc80aNtaLHIUWU6L78eCkGOeYBWt0tF3tZOisGdextIXDQvIG9p2gdvGIMurMeLIP9zGVyMwqNIExI2sj2heU91bVUYdyKFLaeG8kDsiCOqRFw0vfEbDmahvjNXXWqP9kyTb8ruDgUduP4JtpeD0hb+DNNtw/vF0BU+PnzNP1JA0WiahNtvOhKjqWVoK2nOD1hYeradTtFR9sgd76YRasGXYxtrd11gKbZwQLe0BoQdBGJ5rCz/sxnjy1Q4/XEI4Pb31w3c/aLdqJuoiMiNwHunuSuJDRlfunY+xaAvz4Ts2CNfd5Zuisud6f2dcEa0xVJJhl3/s1y6/Mm41RA+reGElk/ZaZa5Pyn2mQTB+sDZDHcUrhac4xptky5Y/1kGau5TzHAlhzt5Ghs+a8lo/nnfs5BJtpUt8jM/sURzpuNDQxPT+nOc8iUNf5w3Mmas1ebaGRtRoQav8coJV6myS44Rj6BRFZkuKwxtJMarfr/ZoFawQjrn+GzlpyAYO1L8+WDdBYSy5g43pfvOaeT5aKAfpcj5l77R+sudrGeE2dNXPXNwei1gOU5S2UW5a/OC3RdRftmCiRBcNl03HzNWvu98YUShKXWP0OzII19zEkoO7X0hu/PpdJnxLA6QDA1NyHG8kNZGeR2+IGcAnLhegHP7uhHdjWm5kqD/oG5TbUfdux+V69TvcuZZXl/A8EU367AE3pkjll2qQ6el3/PmQQgLVLACGskapeDTn3flq1qhFy/DX/NUJ0xJctbqaEE68d6iW9upVDlC29hiEZdSJKLlI4FIwuE6w72naANQygOTPqS7as6WX/rm4aGWTU0FurB50xM2mQVyGqzYjj/McaI710nAptM12UA4fMk0yFnDm1run8XB1wSQRrElAa5BSdYEx95SoFo16ffTjMsWJBoOneTK1kTFVAN3NyLBwApgrco3nG1HOjOCL/v69XBS0UNQP8Ek7GpIE11x9Kq69PrqmL9Ls6cgOrrF+A1OF9gAfu89Rcaaa9fQ7B3O9wuGpC5JfpamyUFPgOTtKaceWQepRaowF0YL1dw9N+s2DNcLCNbZLAGpwpghDq1rEG5Cek4b0HW51c7dk+rCsxA2YZTRvRrqCu9NOBTJ82hbyCqCajeEyBZHE/a3c82cHrPgtpkIZtuFX9p0DBGuzz0bq6cmIFxg9S16ij9j7t42X8OPSkfAMe2o81a3XLZ9fIE+uwmL72NSKyzyCCFwZ7EZwwpdarLbw4wUkFa1Y/zzj/nRUxmmpG4pAfEFXtD1Bi+fv10Kcbh5rJjD7FpFLRrNISwu2sA6VdaJ/eYIM0W9tl3Ce3ZsGa6xji6yvxotiu1zL7moyrXNRRYWjMDYJaJchAOidTOhM13KeZefY7gHHbaAfxCdNqp6CWi++j+DjrabkYYOY6CfphAay52ojzLClgjbV1RxfX1vHDNFku8nh7DnGxzMyiEYH802DX5MIao9UUIn90QAkVMa+B5zcBP6P/CfrvYRy6HzcL1gwAYmyTCtZe+oZC2HWlRMVCUtJPK1GhoAyd2c00iCCbJIWnKai97+xK6TGypcoAlK9RXBbvnGT6OrfBjL1gjcQiM9cMV7Hw1KjzGjDpXpVNmLV2RPx2uLS7r5GkS59W0yRv37d3EEiAtQygb+DkjsJauFWHZkm56sUkf9E8KuDN6J2Z6/Acs2DNGDvObTKBNfpwzF5jwGjSuGhlfLyCoIhrKQ2PtwJXRJ3oKFPkfj9+9YDqIa9d0VLIf0B/uiewCP3RH8ByT997AXxvc/VvQQDWaJAObUtKXOMifluTRoWFqWkJHV/PiJm5oKS3z5AhjTrWA++rpF9EcxCZUOtsFmrXzFwn0Tl2gDWApyMHeylgIyMhbUD6eG9tA5TcDZKORPfvsgLB1YI2LYtrel8DAOZU0N/YCKr785+PVkCbFpOaUUlf10h07NqkgESxdfIhshYIWGM/yoO9hysXvJ/uqG0kUyMnF1dG3BsnJluie3exjXGM12b07PDeHtDwGqfvIVkNi00J3IzzrG3tAWs3AdC4wjwdrIR0iCZ1LwJWwiKmUvncf6AT/u8gDKBQNFnwXopPW3oYzIr31skjJ+GMJTwfDpkfJ8AWsAbHjkX7Tavm1AjFFhBdMAJgpqbId39aaA0N2SBZz0exW65ev4xoUT18H4weWaHe1s+yAawRdDLyMR7slYyykV2PTJYG3b5vG/j6zsEG+Wx9RB4LSEeMn/2oL2Lt4xNDS0k7ON90VAO5tl1gjWQxJNhhiigjJIzUkpqe6WbujYCX6bBm+ncrfs4w1dGIfnCrQtTYmrmG+zl2gDWCTKYoEjgwylYSYHPuoJIKpNx1xPi/FYDF+lDW8RHoOOzaQjYjTfD40tqBPedsAms7kb5ZOCKjErgQ6DdFrZnlFE4Pz1gC/mUgmboXacZT8Tzis+381gYKaBcNL2M5BZLjyU6w1nFQU6kaW1aq1vPdqsSWkQce620KZBF4xMQ5SI7GPN5HZoPunouzxcpGSe582RF9yi3Pf7jYNBBxABZ7wVonMGGmSZta0qZLrc8mAjYCs9stjTJBVqhdQtNKzYCsBQCyIVkyQEw7THZ99KSUr1lCr1G+ZnFJDR6EAQBxZq4TDGCNgZuY2iAhAqDKkjmtYoTIvNDpw/9Gy4v/SYDI+jMzqYssM2IgitepG5Nf/evd2zuDVXwE2MuLg90+tbz9hv/gk8O3DAKwRqeZaJaOtZlmlnKf1z351kCZNa2ezHukkaZW0vEe3L+yLF3YTBhhs+Zgx4NCm8DaCgg9U/CPg+TNV7Hig8JFDgYdNIwUuTYPoMNTX2kjsifWqZ1fawV7dCuL8CxWkhC9C0UkctL4aP0Ms8BGP8MGsCZ/TpV2St2fFuyhRTE5soCxLET3tW9TUtwbc4fN9inReZCR8GRLa/vsAWt0ergqz9oWOpFccWZ7A46NJ/KDL6D/ZMZRMs7h1uFEwsGCs83oi2PF15rDRQfAFrCGzx0OwElpg+dB+81oWJ6wdFoz48k+HyDiZgZkudrHQfXvcKxpK9qItnJ3ov3+bwNYI4Cirhdr7Fjv2AApaqyBeQNpap7sQ70ro++++mOco+NHpRAc4yUp44efZxdYY5SoaF4wD2PxKx1+3Anc0iINn9Ed97RRRlWZpufLPnfymB1gjSQoBRFRzp8rvdauMRpfpmBmaV83XBczuKBhNEbKXjeZBunRTpDJ8Lgf48PUfpvA2vNIH2cEjKRNq7G4kx6An3WbnuYZa4vNAn6jz1xgMeYdn2EkdzH+N84xu7ULrNHJJ0siU+/MtIPxNPz+AMT6I3NAvpEK0bRWiBitkAZta4BlMqtGnJ7YPkGp7594boJpIOL4PHvBGklFlr04VaYuG6ygbczcPpoKyX1GW3VwJvTplgDQeo+mGbZjVLNJx2jJER4qK6DP9hTSTymo3XlIM9l3bqXKKBD4mWWWtDuyRtK4GZPrynCUt1StDNkYkBEOQIBn2KCqtxuOTZ1YR05BOsucvzdFAys1q0Wqfz0I2OMP1MJtebqDZAfRCLPlSOGfyMf06L8HAVjjzTJycQZpa0zdoxjyBwBZntr7/xsghuizOWN5cKxNpsB5vb4NYI1fJgdOsaLZVQCb+bRRQPxkbzQEw123BL1mBoBxDs//8csHFDSz34x2MpWQ4NarHTwOKNjbDrCGe/nonUFSv14BKVUiB/KEU4PWNY2UBOunp8YJZ/Tdcv+89dvSfnvAGtNg5g0uqU4AHUh/jUxqgf6Am/2h93aeXWDtMOqLMmdM5dc2tB1BC9NJvfXhju63A6xh/CweARpspEv5Gzs8TpkCu8YPbW8XWGOfSSyy6oGyMqFrYV2VH9ImSuaD7dK9LUStIklC7uhY8QFM7ABrtA8Jh6pCaqAENMIISnJigaRkVEiixjqzvYg822UfsQGs0T5fb2mokTUz84wsl98/18g2G9kJ1lijthrpd/uRqvj0sUeV8ZHizp4aRbIJCgzA4W07H6LQKbDQsv7VORpliioWIXUhjs33PvPmXMkSlkmmrxzq9zoJr28vWDPuZcs786XjoDgAz3kW7z8hgKPUQpU6pVUG4JVv18uUpYP1Obd071R5GUCu67BmYOjMp6Qlxmf72toN1px+IFIUH5vTUNni6VMnCIJYxA6Gj8mgClMiDX+apV9fnR4hN3+zIuMVBGCNjIzh4SEaLmReZ5FCqH2Bo+2pZURK42HQqDsNa8lB9gDcAnm/DWCN/V0Lin2GYHPlDIHWQ3plbcwbgRBtZOLWrElRUzVrd8SONoA1Y1JwMjA1kWHqZkiDZNriryBkcW+chMZ77ogN/I4rm8AanDZGulhTQwKEioiyVS+ZVZnSGElybydRi2SXs20XWGN/yWpJqvgejfOitiONPDulYiLb0FaMTpphg7wjjqYdYA3jh6m0rLWjfaohqlaxaBYVLXYfO/yf6Yt2jR87wZrz+wa4Zd0eSX0+WF3HQf6BfUoC4rqFXe2ykx1gzbAPozkEqiXyZxLqe5FshOyQ7u0vRJ+N99z1rQ1gjX3keGC66L5HqwnTyUk2QmDvaZ4RzJIk5K7bBvfJz7QLrBEUVIouJdmhFbb1xBNSB4AqPaI76TOmS9QY9ek2vIUpsMY6rHuQ9vjoMw/I/G3jNco2YnYPeen8Opm3ZSxo7lPrfl/gI/Ex+8AabTN4amdHm9ZFxa8HY+vcZxzDdvis7rIT9XqJ7z8hWOPx2k0rSanKhWUnUiCpz5YrMrumhzKCWa1+OaRFFlcQbeZadoO1qz+OFxLKsTFQceqDIc7/jf3GlmDr7vuPQQDW6Fxve6aDvLy/h6bcsRZoO/731HjeT0Co9jjY8WDPJrBGOz21uo2yM9aqESmZM6XVlM4RQ6uJe3scKwMGir/rtrIBrLn38dT7Q0CNOtiGCWV2QcA+sOb8QQczJH/8nwTjoV2OovNe4n/0Xf+3E6zpfcCZJnvfWNRkkRDB9d6C4rVNYM3Zd4yfpaPKaKQoGMcP79OuyJrTRrgH2oagxDZQ72Fuud6fnWBNvyOkAnPxiKDW9b6C5rWNYM2wAdOJGaUmeDP2BdPWLrB2+PxaeXTTA4hyDVPq/kU7J8rU5UM8t2VDQIAx0xRYI/jIXzhccuYJQwuVzNlCZCOidiToiCiQS/Lkzyk7PlhkCtDcBir2gTUyP2ZFNDBrWGa/LTcA16Y3/bNBvgTbs54vVepUUqQ06rFSpZSW3WM1khbXpY4C3PvGtzdtI1vBGqJpD4ysoVT6WZD+6K/FIRBw9/3rIABrFJgj+GCKI7XVxj+A+odXetsLyHxFRmwCa65IftVS1I5UjwSbpTmBcHcwc0f/DwKwdkf752tsmD5mP1gLVgfbcELsAmvG5we7fRzpWccSyBa4stF5e51UNsh/jH2CBKwZ9grWrd1gLVjt4rwvm8Ca8/PdwHYwPpfsAmsHkZZYr3V1jQaxhmrknB4OEhFor90GSYkjQv6OHfp6jczdMk5KQmw7X5E8cv+jvfV67fs3BojLIw8/PVoOm0inTPg59oG1fdCg23jsMXn66KN+2zM4jymlCe/dsw33nlkBBsmGkg82YUok00y3vbsAhCPZtc6PjJpmrsNzbAVrSG98cWcXrUdjiYy/tnZ5KyWtu7t+ZhCAtc8+HKoaaE+AwvI0CvdIsU8qeeZ4kt7SvWkeqWmn2Gykw8J5NoE114FBcpTLoNZ3BXCux219/R9YM7HQYD9Y8+YMBMt+u8FasNjB633YHVlzcyK93qeN5wVDZC0Y7eJ6T/+BNT9EI0EG1ly/u2B5bRdYo+YZ68li4irJ5rfmaUpk9QblHaABqX/b3NruT83pfxE4EEDw+hTcdgCONfL8ycWyB9cwwIVjv2cgk/iYfWCNaYm8b4I21prxtdcGjTSek/j+vfQTqaEUICfRC9/DLe1u/G/2OnaCNfWjWY+GCJupZrF2LXn88SAAaxR5rgHxZYpWM7WPFKmktPQUisyE4y/9P61ZS54v3AIoDRQQ/wfW/gNryeCg/wfWzDiR9kXWgsVR9HUf/4E1P2MI8/Q/sObHRv+BNb+pl3aBNabi3TugiZKBZAhhHX9KpOWllIyZ0qNlSNB4vPuIlqbSIM0CDGvn2QfWSMBSqESk9JvQXkXDi5UrIIVL5fPYSlUqrMDXWt+8ADmAN7PXsROskUTkZeAKBozOfzZSjoFtna+9tW0bO/z/jKwR1X55aqSQTn3S+BhJB32GdtAgmA1ay4em10vUzn4yzIQzfAdBSRBE1oIauAUhWCPrjhndubtn1/8ia76cbB77D6yZcSL/A2u+xpHdYI0paxSqNuQdKP5N0pVDj9eQEyBpoYizr/u/G8eCEay50srfDRv4/Iz/wJrfMWoXWCMQ2PPJUpm0eCCASAcJz59DChTLK33GtpW+49olaH3GtpPHnx3z/xKsrT/ysLI2Umdu0/HHJaZZZVDrV/TY6rWqJtvfX2AaZJkFY/7OsxWsIaI2alh1DRJRQ/eh6bH6mkEjT61xg0LK5Hj3/EXimSCIrDk7DHTLKFvDegVlz3OdRa4jJPmHh4aIj63pf/+BNd9g2SawxjHxzWej5Mkn4uSdow7h9E/fGywD+1VSweyyZXLJkIFV5NN3h/i+/0AjipbeZy9Yo9YXi9Wv7HFQhlNkl/o968aVh5BveXUkKZ7q04lJhuiZr+vbCdboXFNomHTYdLjpPFKYdhOEjSloTRZIMtf5uv87fszGNEja47vtjVSwl/2kxtPxpdGyfnx5WYf29vIYy7pPd8JedoK1r7c0kG4NIqRW6VD5HWPlFBhG65YPU4Fs0rFTFJrHPl5nXTA+OW1lF1jjvDoJncLHIYbNeXbzpeYqEk99NYpkVyuRTR66r7j8BM265Oyv5WvZCNZoIzJjUkidzyT+/83Whsq+uhzC1mRkvbirse0EUXaCNQMEvPLdBtRPNYLOV5wCDTI3ujfjXHu29kXWqLE2EmyWe08vlz1Ic5y/dbxuSbn/sodmh31sBWvw3T7/aBgEqvspm/hPX90PgrpBcvbjYR7bt+dG24BBggmsxYMw1mPdhEi2E8RZcoLvYETNuI8gA2uMGH2HwXMGtX/fg3aUAuO22s4msEagVrZ0LlWe37G5o06yggWyqQRE7Zr5VDeDqbWFsO/MSbOihndqPNkH1riaP65zIQnLklp2gF6doKN3k0gV600L4VW2kPQpIQhdwFY6aLvAGln7CMiicmeQMZ2wgob/HxtYUihenAb6YhQ1pqPdtFpO+el5Gx1Jm8AaKcIndS8i2bOkkc1TKyE61Ez6N8+vgs+0C8dPxnQpZVCrKD1m2TlOxkUAu8Aa5THaRIdjHqWSsZ0KawStQaXskgX6fZN7FFVnm3TsuUPTKij5da99wN8usPbSvBo6hrJnTSOfb6wvuyElkhHPnVzQW2sIqYOqxbPqPIurnksBi23jyCawRmB2ZEFNqVgki8SUDRPKF9BGFBJPDV06Poeod1imQGZ5b2UdWwFtMIA1Aox9n6/URsffDsDh+zPtA2uz1oxAeVEKWfPSLGXEzBiSThbtnKS1aaxPc2922M9WsIaF/j49y0uLuKIaJGKWX/u2JRW42epLG5hDt0EG1oLHMD6cdBvBGslWpkNlvU3L4vLwrAYqEL5obhPVWkuXLpXkh1A2mWwIeG2zpR1gDVHZBY83kfSwwdaN7XWSPTyzgf7/AqK0FP0miH3tUG8JDU0nIyF3YGt09nf7wNr26VVUhLZN7XA5hxXbtYimpUSdaC8AtqOLa6HVVvBGR2DjxAq2OQF2gbVPEUHLEpJaKkGD7sjCWtDHqgvHKIVElw2VA49Vl7eXRcvMPsXVWRrSOsqhmZWMAMO0U2oTWNszuxrGTwppUSOXfAZK8U2TK+n46dowr7yxCEy+iNASvKVOdY+sBug13Z87YEO7wBqjjrkAxGb0LqaU/V9tbqBgn1pidMJpk1svt1RhaALcN2Ezu+xkB1hzgNncUiA8g7yzPFoYxSeYLZQno0awuUDCaP8ySENQMJvaYnbZxw5RbPb1FwD4soUyS94c6fUZfWl3EykUkQGLSOll85RKcmJFjGZChOP3jBHa/486a77Bkfl6qbtzHfvA2vJ907GInQqsjYhaoyaNvBDFyxdQQhbq1Lm2GiBo2Qrh7Ltjk9vfka1gDf5jr27lJE94Jtn6dHvp2bUcfOmssm9nV3npRejrubUTx/rb4D/+B9asgxqbwBoBWL26BSQlHKUcOTKqQDZXAkjKUrJEDhk2qIrE1M4vTLFZsgArtRiAtgA2O8AaUmb731dJqlaOcK6G9OhaVmpUz+vUw7h1Dfb4c6qumDSILWifaLiuktgE1qAdNrhVfikVlQmr1U3VcWRUrXSBTAl+7KkrRrHjLvUjbNOHsgusbUAaaOaMqdVplCMtZfnosrrif+bpei4OYwvp1zyfFInMCKeqqct+P3VmyQlI7ABrGD+j2heUonlDkELr6Pcg6D8VhR1+Q7qW4VDfQGSJYutto3NrCqmx/25v7QJrXz5bX0IRid2CyCP7/NnGepItc2qheLGrDT7dgIUBRNv2Ip3Ndf/dfG0HWCOwKAcgwrFEgXDOoWIYQ6Pvxf+Yc0b/ub9AeHqZ2hMr3jjP2H9XtzrP3gxIIuP308+KvByYoPdHSI/NnCGVbJmGMYS+vw1Qmw6LRgpc8b8B+p9+sIJmQnCR6a7axeVZFiyRtbsNMKx9nn1gjayWAyd3krLVikqB4hHqI5JFs0TFQokahaw3vz3v/x1YO7inm4SFZVD/mj40GxexPbXYOgWcfuXd87H/A2vWAY1NYO2VAz0VoC1dGCcXvhkj2zd1EEbTmNb31emRSjlKnbqmjQpLpQrhEBifZL1vCcKuAYI9O8AaCkQnT4iRyLxZ5Luzo7Tfs6bFKohlHaQxoRhdi42JkuZxRWxOF7UPrPWNi5SapUJV6Jk/+Py/Pla1jR9/xw9+C2lZK5c0R/SEq9x2OAF2gbVVY8pKzqxptTaETtLKB8pputqFnS4pj9g/G/U0+XKm19o2O+xji84aHOnBrfNLZUQdmf7IfjO6WLtMqNYcOe0A+3SoGy4NK+UQRlGc+10cvLuxzy6wxoWQEvlCpFn1nPIjaq7+OBAnJbFAQlsxYsS+s+5v3uCSqGFLKe+uiLHNRnaAtetI6YutEAb75BK+/gv1aqzn694oAuDm9vOGdbU5kSb5yIAS/+/A2ger6yhYOzy3ho4NRtIyA9i/hci+69x59Ymakgmgjsdd99/N1/+BtdsRIu8Azj6wZtwT9dNW7J8u2XJkkYXPT1Qh8RdB5+/eDiZRp874PCtbWyNr8HsdRIcjkH3VS7p2KqNZavt3dZWX9/dM1N5987/ImtOpNpxrbk9Be+3UB0NsCDt6ASt2gDVEyVYtaSHhuUPke6RC0i43AMYYUeuGgfX3b1MctsN58x5ppGmRFwHoXO14117bAdZgjxNH+0m2rOkQfYySN0G9egSTLkf2DDJicFWQigzWOrVpSBFNA+fosdkNMJ68fL+41p23lU1gDU7i6rHlta5ow4QKWgfBNMhQrPofXxKtgO3WKy30hz8MNUkjsfKdEMTdvdVtu8DaOyDHSIe6qwc6FtJo0ck1dTWyxpS+mwAetMf3SHOrXCwr6mqygUjDJjBiR2QN42fjxIqSBql7rOu7AUf7WZCuZEUU6Q2kjNI2HD90NFl7NKBFftsis3RW7QJr/OytiIgwMsII0ux+xWVUh4L/x95ZwEtRdQFcpVOkuyQkFJAQC0RRUQEBAxFFAQvBQkExQAwUG7tRVEDFDgxCUVHBTkQUBEUwKD+ZYh4AACAASURBVBP95Hz3fx93mLfs23m7vNnZ9/bs77dv5k3fs2dmzv+euDYHC0AZP7iZHNOttpQxuX2H713DA990GtjuXFHAGue+3YQ4knd1igmZXTSlm9xyZivZsVwxufXM1rLEhNd+ZHTo6G617LOK8Gx3vWmfRuRZI+yRkMdDDPCvNAVY1puqouTRjujb0AN+OkzobKtcoURO55KRa9rlY++zgbL215/y9D5u2LBB4n2WrP0gtOqM5K/d++plcvvzY42n6IbQzpN/IIke1rjWFxbfKfe8cpkpNuLGj8sPaIa/TSbAmrX7jA39/JMD5MpL97fpM+Hbgvm1NzPAswbR4vGI90VQPQ9tKof1aGo8RRfabf4z20cqwChgzcjhMTO2Q5kyxeXdeUNzegGMN61atbKyz1715M81OTlqyJJwyBbNq9rKmpHIKSJYo+1PTDtKqlcva8utNtl5JxMiWsK6s5nuaIqL7GDyH447prWsX31+tDoUUc4axjQv/T7GQCRX5uCO1YRCB3WqlpKqBs4GmOp1A7vXlmrG0MZrhMEUxcufc0YBa8gHjwcFNACSDqbIweVDmtuckNIG4MjTGmrCH5vWLWdDj6Zf0i4y+UThWUM+VKc7ej9jRBv5UAgC/cGIrGzuMQDkuAPrWE9k7SqlbOhWVPrDeaOCNQutRlbPXNFBupjiEGUMlBBOQ/4VzyC+GNiDDqrjVdSMSk5RwRqgT3gjwEooaLN6ZW3BDMKPdjK6RJEaYO46433EO0uOXyQyigjW0CHCaJENuWp0Hh1rns/oEDlqp5jw404tKpnn1HZywbFNIg43jg7WMPJvfGKMdO3Z0YT5NZMRlx0rT3w0SfbusbuULFXChrVVqVFJRlx+nBmo+Z60h/dtgbnMgLUt1xM+gCVzrqhhzdnK2JE2hSiqNKI8nQUZAGuEqQ0Z1NaEphkoi/kCatVMflaNaiakpEcT+yW5zwk2kmlEsPb9krNNmF8Fadyokgw/raO02bW6zVcracalO+fMzvLW3MFyy/U97LJThpg496iULSJYQxe40b43YZD339lLkEHf3s2l+/6NzLh9u8hZIzrJM4/393LaItEd70aMzrOGwbPe5IJQMrtzy0rWaKxocrSoXEdhjVrGyO67b03Bo8S2GA1RGElRwJprJ8bhI8aDdIip+Ai47mjkQqgRcqpqQiQJ2Xre5Bk5o9ztl9ZpRJ412kiY342nt7RGI9CRS38ql5LepjPgAzOOGNtGpT+cOypY8+sBUEK1Q8LVZoxvbypBdpDXTJW/lTO65w4dNdfr3y9d81HBGu0jxPoLk7fHs+jEHvWk557V5aAOVW1nwCWmauZbpvAK1Vmj1KGoCowgH9o9x1TNHGiK91BoBA82zyE8tgAtlSInjWhlPdzp0pd454kyDPJu4yWqUKmcVKpSXuo3qSVlTKXDVh2ayA5mcOzu/faSYWP7y257NJXiJYvbcdaSAYiC3VZhLZE8MwXWorULE3nZMgDWyLM6qHtj6wGpYTxFeNEonMG312HNpLqBtapVtiz/ZOGpWQdrlvb/vMRWp2mzaw0LsOSlzXrxODn2aFMxy1TuI7yP4iOdOtS2g4xHpnQRw5prN6Gh/264SNatGm3K+J8hS788w3ggL4xWdzIA1rYYPj1tUZEVZjyoT+7vKgvMOFkktVOOPifPKFojKUpYyzFIelojiDGgGGONXBHC+5abcY7Is0GOW2QZgaEdEaxtafNm/THycPrzqdGj1U8emBH6w28YFawBGGtNGJu/Qh+eoa9NkZr7TNjxhQObmDDSNvLd9O7R6pCRUZSwtkWXzG9lZEZVyJ+fPli+MmGR5KtRqCYeHKR1WUSeNdq4yXyZArW/mHEfvzL6Q6EROkK+mbq//GWGYfHLMK1y2XxtnDMyWFtxv5wwso8BtQq2LD0hfqdfOsDakt0O30MorAEg4Gmr1aCa9DnxgAjDIRXWFNYSwVjQugyANYzr9avOl/PO6izVDaxdMa6b/PHLBbZghvxvvC1T39tAmxhYEVNIgu1z4CWocSGtj8iz5tqMbJYtOtMLc/zd/P+s8RhdM6G7PDfjGFmzclS0QBIRrL0//yRhfAxXWOV/v11swyLbtatp4bZq1TI2ZPTN2SdGqz8W2KLzrBEG6QZ7znm59zQG0kFyv8llO/3whjKiT0PrAaAoQlQvf84bFaxhYDPwrL8wBrkhT5vy4eeZMbOGHlJP7hq5mzWeIjWUIoI1KvRRmp5Bep3+MID4A+e3leFGd4Yf3sBWsGM8vyj1h3NHBWtUxqSAhhu6AD158rIOtnANYX6EsjElJ+mNmyPMx+I+++gqWbN2fZ75RmvXro27bt3qxQYmeqf0GwNmjPHIPeWqPHKPXWMKiTQzIcZU0sTD39/krC2dtn9K5ygw3YsI1nj+AKz+++hfE6I9b9KeMu6EZnaolYmntJBvTH5fpM+hCGFt9veTZf8+nWX3fVt6YPbwW9dI8RJmfMPrh8osA3MOEPY9ZHfZ68C2dkwxtyy908yENTxaQC1jraVXHlt+G86rnrUgXskQWAPA8ITMNNVXWrWoKnvvWVfmm7A+QvkO79lMgDUGf3Zek0inEcGag1TAFk/R8sVn2/HUvEIZgKyR1waTj/XzivOiA5IoYM20+5Ybe9hxMn77yYC+0Scq+VAts2HDHeW0k9vLyYPbST0zDl01A22fLjwtYl2KDtZuOaO1GUOsou31pxjEt+Zlv9vOFezYK+StEdbGOFpHmGp+/nLsBWb8mJd7fo4VFay9MWlvm8OHEcR1IoN+JiyUsWkYGJswSMaga9ukgvGMHJCvtuSnvUlvEwWsba6OuWujCnYwda4ZGbRvVtHKh4Gynf703quGHSsq6XblUz/yc9yoYI1Brhkz7NrTWlr9WGW8swxm3KROWdsRwjAQMy5tb8GEsbTIA8xPe8LYJgpYw1PEoOE9O9cwpfp726I040z+GgDLYNhnHdHQDhtCeC2h2nQwhdH2fB0zIlijimhToy8WaM09AbwRGkquKKGQhGdTSbRW5ZIy54Y9o5OPubaoPGuzDYz1P62HVK21kzz81kSh2uH5N51kn0U9B3Y1AJIzOPYzn98mDZrWll7HdVPPmgEjP5SRx9d2r13kjMsH5lru3yYd85kAazhE/l57oaxdOdpLl/nt5/PtOGtT7utjmcRfXTy9HJJBsOYavvLbc+Rkk2+El23C+P1l//0aKqxZb8xYmT1zkLRrU1PKlTV5IhVL2vDRj989dQuYGWAbM2pvO97aPyYE0Mk0rdOIYI2x5eqYnD4Ha31772LLr3796XArH25EqorWqVVBzj1rzy0y2yzbtMooogIjGCcYkBiSf5mebHpkKSfOoM+3ndVaVpkQSLxKV560izUKJhtvSb4MmgI0sN35ooK1ucbwoSDE4ocMrBk4oWomoHZGv4ayzIAJYaLTTeI/UDLEeNmcZ8Bdd9qmEcEasF/XFJ9xID/yqMZWVyaNaCmrTAjkD48fKNeYHCQqat5hxqhLmzzi6GCmwNo84z2jYM2zMeOpMcg6BTYiLeQTgWcNWKMDBKAH1tY+l1P58ECTr7bOzKMzbPPMFR2lhOkYefHq3OPTpVWnIoI1AL/qjiVMVdH2Vh4f3tPF5hUf3LGqfGrC1dcYbzZ61cp0nFCZ1g0JkVbZbL7nooI1IOKWpy+S0mVLSV0z4HOnbrtKuQplpGrNSra4SP9hh8i4u4ZLl0Pb20GhL7vvjAiBJDrP2lOf3ipnTzjeFF8ZuNV3uAkbJeePwbBZf/ZVgwS4TQeg+c8RNaxhH74x60TpcdDOsosp0Ef9jM9NFfEj+7WwKUbYANSHOLxXc/n1+yii1zIQ1jCaqQz5kCFZRhQnH0s9a2Pl5+XnSqMGlcyA2GVk2CnthUGfK1YoJc2bVpHljLMGdBhYO/P0TmactVrRDfqcAbDGjdeqZVU59aTdc0OZkQ9jaBxycJaOs2Zern5YoxR9G+NVO8pU93OJ/ADcvyYni7GzTjZltbeEu+XPI1ZQxkJGwJoZV2zooXWlSe1yQilt2uZCjs4xpdjxMPnDlAqq7fk6TgbAGrLouMuO1uim55//+aJL+7erYitDMp+v9mw2/Apy20yBtdmmSATVDSlR72/fV1MYFLt4pDASlWfND2vfm3A/PEYMC+E6P9AjPI5NaptOWzP0gVvul19a5jMA1pDFIxe1MzpUzOaH+tv95GXtrcc/ynDRKGGN8L1R1w+Rxi3qSvU6lU1RkT3l/jlXSMv2Tayn1oYdm7DIw0/YXyjn7weE9M5HB2uTX5sgpQzQIgsqZJYtX8b3LW2Wby/FTEEWllM5c+o716ZdTlHD2kozJFa9uhVlp51Kyz571pMqZoDsBvUrmoHoi8kwE5n16vMDrW0Nj0wYb57jaS/gl6Gw5rwceEJGnbOXqe5nQiWMAe6WRzqNKAzy2cePsYT/wtPH5uTuGfB47ZUTjFKVlqP6tcwZEyLLYa2WGYduxddny1/Gld3H9IBQBfJfk7vm9IWCI/t1aSB9zTpy2tzy9E+jC4OMB2t4R/wGAPOM/9TbDIxND3fsunT8nymwdtKh9WQPAyR/v5IzmLFtu/G4XWW8j+QcOYhLh0xynSNDYK2TkQ3eWQxK7/qMfI7er7at7OfP/fPWhwBm8Y4dLayVkSO61JQXruokc031R8adY2w6JyfkcsPpLWx5egqzxLv+dCyLEtYYOgQgW2281fXNvXSd8chugbKeAsThXbrBVB7dsjy9nUZRVYP0e9bQmalmPEPC1MkX9evF60a3qA6J582/PJ3zUcKaA64XFt8lT31yixfm+OyXt8vEqefJ2DuGmeIjV0SekzVreXSwRqjj2DtPl1r1q0rzNo3kmmmjZPqC6834c9dbMKtZr6oJJz1EHjXj0T323g1miIP0569FCmsGvB6Z3Nemzcx+8XgRwxoL3xxqnCElbTVx7EZsRGzGXqZC/R4d63i1EdJnO2YQrP1kPEcIigIagNliE7424bJuMnhQWznv7D1lrhlJ3BWPSJ+A4iT9RQFrRpkYFJt8qx++OcfKBxnxvfvWw6x7dvqUflntWbvzlkNtz1F90ztywH6NpPMedey4dC89a+DW3GiE114ypouR1Q5ysxniwBaswRsZyTdaWCtXegc5f8DOthDEoIPqSosG5WW1CbvhBY9hQOVDwvwuGNDEMy7T+fLnXFHCWvFi29kQx8kXtLHjiDFY7zu37W3kk+M5WmFCRdubvL99Wlc2uSQ+SEkThNjfIkJYY/yrUf0b23DQk4z3ldwaiiE4/aFaHXAS5aDqXEtUsEaxjKO61pKGNc34c+Y+Km+GxWCcNcYuJBeJIi3cd4Tb9jLl6v+OamB17rOIwiAZk4/Q0Ob1ytnxC+n4aGjG6/t2KnmgPeXzyftZGSKjhWageqvz6by/3Lki9qwxfMgtJkT9oQvb2iFE7jfVRB3wA7qDD64rNcy9hl5FJaOoYQ24eOjNiXKTGW8NMHvRgJuDODdlPSACFLhl6Z1GB2uuzQ/Pv0a69e4k1WpVlrOuPN4MjH2nLS5Sp1ENGTKqX0RyycmhixbWxsnEK7tbzxohjjmOoUukk4GyUef4UmaMM2TsRV2kaZPKsuGndI/VmyGw9vn7p0lLE7bGwMV4Rha+dZLJPypvc0UqG7dkaeP+J170uqtMqWMDKNEY2JsN+yhgzQDF668MMnHX28vUB/rKJipjboaMv8yA2AeaoQ/qm+IZSz4bkZ1hkEYWq5eda6thjr+4q+39aNWimg0TveQC86I3NxnATxjAQd13lrU/RhFz7AfD6GDtuQmdZE8zqCrGEaFHyITv9cNyeq8Zt6eaKaJRuUJxibTH/8MrU6xS93XKVeowdijT393kzjStw8DXxWT7zfI5wuTYMGD2m7fsbXL+ytiBjcldi8pAimJQbNpKntVeJkQWw5oCB05/JgzNCVW7c+Su1ngEcD+4u0t08jHXGhWsYUz/N/swoUrmp2a8QnKvLjEFNI49oI7NeaT66i4GUo7qUksA/8h0yMgoClhDPhRZmX7J7nKeGex5/3ZVbfEVFxJK4aMOJg+LTpPRx+wcmXff/i4Rwdo6U1RlgBkEu1XD8jbM0T2HmpgBstc918N2juDVphOAAbOjDTeOblDsl0wlw8EGNMi74llEmN/eB7ezYOaHsWZtGsrRJofNgYt/XXrmo4M1f/sIBT3rykGyY+XyNpdvyhtXi8LaWJkx9UjLGHNmDvJs63mvniCfGTZxzGE9a4c1lT071ZWN69NdEyIDYA0B9OnVzI6ldu/tvWw1FoqKVDcDYT8x7Sj54duz5fP3hpkBjnexnpJsHGcNMPvNDB6+7971pWyZ4jLQ5F09N6O/p1SLPjpdGpt8tiY772Ty1WpkX87aZnC1AGvADDj700As4E9lTJa/bkJGn39ygPHcjvHk5oA3/dNoYM32yJoQNYwhwvcY9+k5U/Rg/OBm8pIxwgk1umxwc5uDRElxtrf7GKMu3QZlFJ41r61GDhtMRb9l0w6wldYIwwJCMIheubazdNmtss0hIefP2yfdMorAs2bbamSD54OiEAz2TJjf5UOay/Nmiv5cc2oLOaxzdZlrBoFm+8jkEyGs+e8V2m/zPpGbyYG0/28GuShl464xClhz57ZTIxP0Bm8j9xsDrqNf00xHyHzTMRLpPcY9HRGsWdkYueClpWjPO2YczLvO3VXGn9hc/piZ82zax3SaXGHuvd/N/7lkmuZnUZSetWtMqGOxYsWkTefmdow1qkBScGRX8//Tn93qeYt2bllPjj61R9bDWg6s3id3vTRedt2jmVSrXVlKmAHDh4zOYs+asQ+pCdFyl6o2V43q4QyP5exCYO0Vk7N24vFtbCXoiVcY73825qytN6Xmd25cSUaPNOVnjZH96/fnWXfkhaP3ySWQ74zwdjQ9/nfdYgbKTLugfF6RiDxrKM7SL8+UoSe0tV60wcebSn1GXo76KUd/qCmcgXeyfbssKzCyGdYA/+WLz5J3Xh8qiz4aboc2QG7e8AZGb1j/owmJdHJzN2R6p9HAmv+FjqHI4M4YRl4+CMtMSFZOnlG0hnYUsBYrn3+MfGyPNYa2+SIzZIOM2DZSYzsCWIuVj9UfI5NM1B+uNSrPGufGCzvrus5y9hGNZLCpGnqrCWVjwGe/DJc/2t16llY9EV0IW5Swxn1EhxEgstSEP+YKKTb3G/ce69cbkPPLLa3zUcLaZugib9jKxvccQjb+QdfTKpPN1+XOGRWszV4xWfqddKDUMDlXMz6cJPwPjFx066nWwzZgxGHy6uby/QprW0r2O+/icyavD69k/Sa15LxrB3tg6/fEpWuea1q8ZoEk/fl3kckxKxgv1xem+uPAAa2lYYMd5c05J3qwhl3ZZrfqUqVKGTljWEdTcTzdIZDwRwZ41hgXrJmJASVMDQjDg9TMVDi8wpTt90PZj0vPMcIqLbfcYB7cWQhrDi5QHAa+JsfPAQbr+DJGBCGlb80ZHB2MRFEN0gAZcrnpuoOlRvVyNny2jPFAUkUUOHNyAm7x2p52Uvvo5GPBMjpYAzAIQRrWu4Hsu2tlOdIUQXhpYidrXLqXL9Pzjm4kd5+7pSCCf1065qOEtTXP9rADz3ZrW0XIGbnbFIb4Y2buAiMMeHxWv0ZbyS0dsrHniAjW0J9vp+5vB09HfwgPff6qjraCqNd2Y1SOOXZnueXM1pECbZSw9ti43W1Z/nImhJ/CEISr7bvrTmZ4gy1g9q6BlJJmiIMvHtwvMhiJCtbItzrpsHpSoawJNTay2cnk9gG2LHd6tNZ4/6uZAiOPGlm6ZWmfRghrlOO/y3j0eQZ1M9VViYD42RRj8cuA/GKK+URW6Mi8K6KEtQOP2Et27dTUjLF2jwcbrxjDv+/gA6SsKeN/x4vj7HKFNTPwtAFXyvI7gH3hqzvkzpnj5OqHR5rppfLi11vn+mULrDkbmvGecSD9teZCz25k2fvzT7ZpRtiZnj252UmQnv8zANYQ0qCBu5kwyDLyzOP95e91Fwpeo12aVbXggdDWrRotF5+/jy3j/9rLW2JK0yMkn1eNHydCz1p+2os82c5N87NPgW4TEay9+/oQW2a1Q/vacrVxUzMIdvnyJWWvznVlzQ+bc9Q2jjMDrtez4/hFJp+IYY1CEAzOS75RywblbbU1ctduGt5qS96DMbZ3NwMdn23K00flPYoK1uitPrRTdZuTRt5aPTOYMYb2iT3q2vAjayiZ0K3hfRrY8Y3+MZ42v/GUtvmIYA3YaFG/vM13RH8oJMJYWIQ/kqfl5LNnq0pyaq/6kekP1xEVrP358iF2WIddG5WX9+/a145deMuZrWwOJIVHNm6uLJrNsHajCS3mvjrcjLWG7vQylWfJURvep+Fmj38vCyAVDcxNM7ltabuvYrxGUYVBEjo71uQ5ArIUpqGIT/EdtpM9WlTKVRGSQbN3LFdcftxcICoKOUUFa4DE6eMGSBkT9njjExdYYHNeoyc+mmQ9RpTwf/T9GyTbYe3ht3KKi7Tds7mBsrvl3lmXS+uOTU2F8eI216+EqQnRxgyOPWXeVR70pgvUOE8meNYK1A4ucJDLAFhDQFTqO3D/RrZ0JmUxj+i7i40NZcwDjOvmzarYsvW4IPEeRVrJL8NhLXKFiwLWjMds7IVdrFftO8acIzzU6BVJo2XNOEbnnNFZ6B0RhTW5cXhLKW16858yL3l6sZdN21+ONAZkWVMhkuIR9mVvYK19lsLavEl7GqNoe9uLTZI/xSDGnWBeagZoMSotvBpYG2GMSgajzSpYM3px+zmtLajh7UB/lj96gCmcUduWoH/uyo45+mPks1cWwxoyoQokpfo949kY3wBbCQMkU8bkDDafrbBGeDGeIjzXFlyNXjG248XHN7WexhmX5gwEjbcoW2GNAebpCDnajIFJB8l68yx6+KKcipBHda3pDYINrFUqn72wNu3d66SeGRCbMcIoLHLD4+d7sHHDY+dLpSoVpEGz2lK+YpmszVmbabyOHbu1ljLlSguhoc8vusOAWhMpV7GsHHdmL7nkjmEy6JzDpeJO5aSNgbkXFt/pyTBdwKawFuMU2gr2MgDWnIeD0L777+olBx3Q2IZB1q5dQWrWKC+NG1WyoWv33G5itC2oBTUq5PURwdo3n4+QaVOOkGkPBn8pV//f7xG5ayOCNQYK371dTeu+djq1yQDa6JF7SflyJWT+3MEKa8YowkPUuWUlm3flvGZUrWvXpKL9UjiCHKRshTXCjhjbiTGenKFNbs2QQ+rKTsYo+vyBrrZQRFbCmoEwvGXtzLAF/nyZtSZslLHoqFz3yzMmTCvLYY0xsmrsVNJ2jNic0M3eGsbqo/gKJf2/m36AZCusoTs8b0YY7zS64u6z3w38M5g65fx/MuF+2QxrjJtW3oyf9qrJe3TyEQP8t56VA/x0lvD8znZYw8hnEOwD+naW+k1N7tV1Q2TWii3l4G95+iJpazxGZcuXlv7DsrPAyCNvXysVdiwnp1x0tB1v7oHXr5LS5UrJiMsGelCGHEffMNTk+hU3xUcu9ZYrrIXME1tBWV7nywBYi/UEYWhvWHW+/GhGFF9pxhQjhM0a31HmqfkFGgWsmbbfa2CV0dNdqWzm8/p2NKGA/2weyC9WvqH/HwWsGfncfdthplpoMZk7c5D8u+FiLwwU/dm1dXXpsHstq1N7d87iMEgDYReY8dV2rl3WlhX3jABjTDK4aplSO8i4E5vacLZshbWnLusgZY0cPry7yxYjyciHEusNTLn6Q/aoZnu1sxLWjP5caryM9U1oKMa0g330aP6te9swv/NNrhrhkNnsWaMIDdBRp2opa0z/ZDwjtlCNkRNDQwByB5j1j1zczoQjZ1/OGrI4Zv/a0syEGX9nKkBSQCNHl3rKe2ZMNcL6hvSoZ++5bPWsffPI/tZj9sAFbb37DBmRx8Zg4jzDVxgP7tNXZLdnDchwQIHHiMGx3f9u3cxv7pYpb14teOHcMrdN+qbRle6fPHeC9Txe+cBZVjb3vHKZyecrLdf7vJDIAUgrYyDu+umjPRmmSz78LlEXGAndNvZzRNLzGQhrmS0wQ71RwJr5YfEqPjfjGKlUqbTUqlleFrwxVD58+5S4X0r527C/pBUiL6pPYnkUsGbaCdy3asnYaiVtGO2sF47zCtHgVatZo5xdX7162azOWZtzQ2cTBrmD9NmnhilDv4cHbeRIXHRcUwtsV5gxs5rXL5eVOWsrTIW+2lVKSVvT8/+oKR9OSJuD2sdNeBYFI04wA9H2NfLLujBIAxtv3bKXHcy5p/EQMdzDLyZMFPlgSDLsA4MYX3piM5OzVT5rc9aQx0fGM9LC3EPFTacaniJPTqb66oumoE8dk4eErMgdzcYCI3QOVTJQVqdqaRl6aD1bth8d4nv72bua51Ax2dNEABQzYaPZmLMGlO3XtrLUrFxKbjOVRIF8B/yfT+5q5baXGS+TEO1sDoNMFiayEdae/vQWqWkqZnY5rIM8+fHNQhVI/u9vxp1zhVleXnqPLd9fqkxJuXf25QprUdjOCc+psJZ8ZZeIYM1B7AiTt8cA2BvXFUy5UnfcAptGBGt4XynXP2jgrtK8aWWZZCpDUjWU5XwXzBsq3bs1skMbnDLUhJCYZQXW5oQ3WTzQjaYaJIYQvf6Mi8ULHkNy0oicwbAxMBmvh8IZeNjw4GZbgRHkQ9jaE+M7SH3jRSO5f8D+tbzS9HgAyNmqbKr7sY6Be7MpZw35MO7VRJO7R05WMZPbN/GUXTz5MCbUOaYoDZ5J9CdbC4wgJ76rTd7RIxe1k+tOa5GryiHrljzczRaQ6NGxqgUV1yGQ7mlU1SDpHJpzw55ykBmAvnndcrLwjn29ThHuM6pp7rZzRZsryuDZ6ZaLd76oqkEa+Xx8b1cbjr6DKSxCiDFholwX+jPr+s7S2HjXKNLCvZitBUaSS8/0jgAAIABJREFUhbXoto/Os0abR10/1Iw/V1Ka7dZQTr34aOl13H52fDXy/IaY8v1de3WUkqVLyH5mOnPJFu9kuuSlnrV4dqJ/mcJa8gZ7xLA27cF+dvA+hTW/IvvGUvvThNH+dL78/ssF3m/rwAyZff3JcCH/zy2LBtiigTVngGya00sWmXLh043niNwIt5wp40O9ZgY0PuuIhnLfqDbWMPCvT9d8VNUgMYTEyGfljO42xMgrurLZSGL9Z/d3teGi5x3d2OvtTpdcvPNEVA2S86M/i83wD+TNELbmXZNZR+//vEl7ycijGlkPiZWnWe7fJl3zUVWDzHf7TFipvGbAznQQRCWnqGANHUJOtHvtcz2EoiNObk4Wv5nx1T65r6v8ElOu3m2XlmlEsOZkQGGRV6/bw3r5nYxy1vU0g2V3l5tGtDL5tPVknZFhWuQR516OshrkxEfOleHjjzX5V4m/bHPTE2OyMgwSEHpl2b12/LkW7RpLcTMINp2N25vONvetUKmc9Oi/jy3tny5A859HYS23Pbu1Xaqw5hn0WwsnD+FFCGsABkMZMMZatLCRh2zwMEXkWcv377fZCxat/LYF1kZs80vZGQLu5e7/35sn8R9jMs7LOR3LUoa1VYuN8du7QK/bk8lm49K2P2L5yOxDZMPy+bJ27dqkvmvM9n98Nd0A6RbjONnf0y8P9vX/781HLR9zXf/NO1bW/ro6KfkgzzVr18u/bw8vUB1KVsbp2n7jhxNse5PVo3Uh3GfpanNS57H32dsp6NBa+fOrqdt0n+W6zryexSz3FWnJtU+ant1B99mGDRvijnW8ZO0H2wRPDILNeGrFiud48ncotoMtpFHeFNOI/VL58Pize2/T+fxwkfx8tJ41rne2Kbwyc8ndMvm1CXLttPPk4ttOk7F3nC7XPTpaHl14gwW65Nu1ZbDtbdlXYS2BTW1tVoW1QgVryQJJJNtvE6y9a2AvyvDEoBumINbTvokim/6J+wL7448/ZM2aNXGNA4zIf944xRgAhOtFB1Khn9uAxN8LxqVmRP74lfz3ypaBdUO/1oh+h02vHCgblr0RV08SGd4W1r54WDbNiq4nPi2/iblH/jf7SFn7S4qwNu/k7LjPFo5P8T5bJJteKeI6ZO7tnPvszdTus8+nmPusiD+LuM/mHJWwUyQsWAMOGMh51PVDpLgZL6xTt13l0fdukKmmkMhW33euk6c+uzWrYQ15AUXbAlVh7auwFmRbKqwprCWdbxWgVNsCa//9aAq4mN7Iv6cU7e/G2QbW/ksB1oxXZOkM2fj2ufLPexcW0e8Y2Tj/XPnt+9nGiEzOa2QhZc0q+fujq+SfBecXUfmY333BBbJx4cUGRJamZESuX/2RbHxntPyzcEzRldG7o+TPL++UtWt+TUlGf3z7eJbcZ3O24T6bUPTvs/e24T5blSX32aK7Et5nYcIaRj7A1rxNQ9n3kPYy94cHMhJGZi2P3rMWFmgVxHEV1gLs6j8V1hTWMgXWNm2KCy/ZtjCRZ23t2jXGsFona9ZtsL3heNqK5Ne2D1CL72FM5DlauyYnjG3N2qIso5y20daEsogLu+iQkVFW6NA6Y0SmKiO9zxLqlt5nAfee3mdOf8KENUCBcMjDT9hfeg/qJnMU1jIUVhOHSyqsKawlD2NB8BJhzlokYY1B8ohdvy2etWwjszjtTQxrqRieuo8zGnSquqA6oDqgOpBeHQgb1gA2KhjiYSsIL084x1DPWiK5KqwprCmsxcJU2P8rrMVBsPwvUlhLryGhhpvKW3VAdUB1IDwdSAesJQKBzFinsJbod1BYU1hTWAsbzmKPr7CWfzKLs6XCWnhGgxpkKlvVAdUB1YH06oDCGiGACmsKa0FAlmi95qwlD3MaBplYZgprcRAs/4sU1tJrSKjhpvJWHVAdUB0ITwcU1hTWEoEa69SzlgjUWKewlhg8Yr1G/K+wllhmCmv5J7M4WyqshWc0qEGmslUdUB1QHUivDiisKawprAXBWNB6hbXE4KGwlrx8FNbiIFj+FymspdeQUMNN5a06oDqgOhCeDiisKawprAXBWNB6hbXkYUQ9a4llprCWfzKLs6XCWnhGgxpkKlvVAdUB1YH06oDCmsKawloQjAWtV1hLDB7qWUtePgprcRAs/4sU1tJrSKjhpvJWHVAdUB0ITwcU1hTWFNaCYCxovcJa8jCinrXEMlNYyz+ZxdlSYS08o0ENMpWt6oDqgOpAenVAYU1hTWEtCMaC1iusJQYP9awlLx+FtTgIlv9FCmvpNSTUcFN5qw6oDqgOhKcDCmsKawprQTAWtF5hLXkYUc9aYpkprOWfzOJsqbAWntGgBpnKVnVAdUB1IL06oLCmsKawFgRjQesV1hKDh3rWkpePwlocBMv/IoW19BoSaripvFUHVAdUB8LTAYU1hTWFtSAYC1qvsJY8jKhnLbHMFNbyT2ZxtlRYC89oUINMZas6oDqgOpBeHVBYU1hTWAuCsaD1CmuJwUM9a8nLR2EtDoLlf5HCWnoNCTXcVN6qA6oDqgPh6YDCmsKawloQjAWtV1hLHkbUs5ZYZoUU1jZt2iT/+9//hGmUn0yGtTVr1sgvv/wiv/76q6hxE9+4URnFl4tfX5AROqR6lLesVEZ5y8avS06PkJd/uc5vkR8y4rkdlYwU1hTWFNaCYCxovcJaYvBQz1ry8skAWHvjjTfk3nvvlT///DPf3PXxxx9Lnz595J133slzn40bN8qiRYvk559/znObbV2RLlh77bXX5IYbbpDly5fn28j56KOPpHfv3vLYY4/luQ+yee+99+TLL7/Mc5vCYkgtWLBArrvuOvn666/z3ZavvvpKjjjiCLnrrrtk3bp1cffDcPrwww/l008/jbu+sMiH66QN119/vXz22Wf5bst3330nAwcOlIkTJ8r69evj7oeB+cknnwg6F5WRWVC/w5IlS+y9hj7l95g//vijnHLKKXLRRRfluQ9y+fzzz+X9998v9DLiOXTzzTfLnDlz8mxvrOy4j0aOHClnnHGGhZHY9e5/ntkLFy5MuI3bNpOn6MTtt98uzz77bL5lhI5ceumlcuKJJwr759U+dPTdd9+Vn376Kc9t8to3aLnCmsKawloQjAWtV1hLHkbUs5ZYZhkAa4cccog0btxY8npJxALVv//+K4cffrg0b948TxDD6zZhwgQpW7asTJo0KfYQBfZ/OmANiDjuuOOkSpUqAlwEvWxZj/E8aNAgqVGjRp6QwTY33XSTlC9fXs4888w8DfH8nC8Ttjn//POlTJky1hjOz/VgGI0aNUoqVqwo8+fPjytXZDRlyhTZaaed5Jhjjin0MsLALlasmMyePTtue2PlhowA4NKlS8tLL70Udx+2ee6556R69epy4IEH5gm9scfO1P/p3Nh+++1l+vTpcdsb77rRkZIlS8ojjzyS5z6vv/661K9fX9q3bx+KkR3vusJaNmvWLCujG2+8Mc/2xp77+eeft3rEPuhM7Hr+B2R32WUX2XnnnWXFihVxt4m3XyYuoxMMnRg9enS+nxt0XFaoUEEuuOCCPGX0xRdfSMeOHe2zffHixQUuo7zew0vWfiCvfnefBBnyRWP9ZFn9x7Kk7YQVGxZlhXzQg8VrFiQtH/l3kQg2ZzzHSpFaprCW/I+ssJZYZhHDGj31TZo0kZ49e9qwxvzc/RgKpUqVst64eNsTGnnffffZbZxhGm+7gliWDlhbtWqVNfDatWsnq1evztfLGcMQUL3kkkvibo+xNG3aNAtq2223nZVXXp6lTDSEYq+JXvsePXpIo0aN5Pvvv4/b5th98AIBwMOGDcvTMJo5c6ZUrlxZkNHVV19d6EHkhBNOsG3Or/eRHvwGDRrIkUcemaen46233pI6depYGZ177rn5kn3sb5Ep/3MPYFwnA/3o26677ipdu3bN0xuCrjVr1szKCC8lnQCZ0uZkrwMZXXPNNbLDDjvkG/p5bnXp0kVat26dZ3QAHVEdOnSwMjrggANsR1yy15ZJ2z/wwAO2LYkiG/zXyzMML3+9evUkLwhbtmyZ7RDhebTbbrvlqW/+4yY7r7CmnrUg4FZYU89aYrBIhbwV1hLLNGJY4wUNeI0da5Q/Hx/gaO+995Y99thDmI/9/PfffzJjxgzrLaFXE48I4SRhfdIBa8hoxx13FAxtXqRBL18MI+C3adOmsnTp0rjbP/300xZCkD1fwo6CjpvJ6zGY69atK4cddli+erExlocMGSK1atWyoWnx2kaIF+sx3DFMX3311UItI4xBjGXuHebjtTl2Gd5KdO/tt9+OC7SECtLZUq5cOWuY0gEQe4zC9D8gAlRw76xcuTJfbbn22mutjrz44otbbU+nCCGndLQ4GRGGWphkEnutdLABFXjtv/nmm3y1ZfLkydbL9NBDD8XdnuN069bNdjABIoRLFubOI659xIgRtjOM8OBYGcb+j544zyPRDrHr+R9PY79+/ayuIaMBAwbk61kX71iJlimsKawprAXBWNB69awlBo94MKewllhmEcPao48+ao28V155JRdPYUwDFM8884wHZXjMHnzwQRtKE7s9O7MegwlAI2yQnsdOnTrJ33//HVohknTAGjkPhGXdfffduQwYXt6EamEA+XPZgFUA7J577on70n/55ZdtyNrBBx8s++yzjw1B5Vh5hSYlerFnyjq8O8D5lVdemavNgDryQG/8hiUghueR7WPbzf+ERTZs2FA6d+5s8/6qVq1qe7tjt82U9ufnOoCGSpUqWSPS3w7yXrjXkBEhVu5YADz3EjlG/u3deoxQPEp8jz/+eCvPwp6PRecG0I9R7Pd+AbcvvPCCldEHH3zgyYNcT7bv379/ru2djPBgco/h8T3ttNNsCCqRAW59YZz+8MMP0qpVK9tpRs6rawPyokMDPeJ+dMvxBrVs2VK6d+8eN/yTZxedS8Af4diASDIhqO48mTTlnsLTSkinP/eM+4hQR2TkD0VmG/SEENl4kQF0HPBOo+MEkOV9cOutt+Z6HxRU+xXWFNYU1oJgLGi9wlpi8FBYS14+EcMaYVPE6FPEwH14WQES5MkQevbPP//YVfRWtmjRwhrPbpnbhym9/OTNHHroobbIBOFrgwcP9m9S4PPpgDVy70qUKCFvvvmmZwBhVGJQsvyss84SQiV5WfPSx3NCTgNGlf8FjqGAjAhrYz2hWcwfdNBBobz0/ecOe56QI7xf/mR+9AjvWfHixa1X0uXAYGCSJ4lH6Ntvv80lI66TIhB4oDC0KCzSpk0ba0T5jfew2xPG8enIcNDvjo/enHPOOVaPyAN1QEtbXc5jvGIkyA0PN6BCkZ/999/fQn9+vVHu/Jk2xYOIvlDkwXl2MLwvu+wy+zzab7/9rH5wL/ElZJKcR2QQ2xb0D08vzyEg5uijjxag38k4dvvC8j/3B3mup556qufZAWaBB57lAIffU09hJJ7ldBLFthH9A/Q5Hh13wBqdKBSQit22MP3P+wz47Nu3rycj7qmHH37YhiGTb03RKNcmOtxcziN65ZYzRf+4R5EhBUt4JxLeD/T5tyuoeYU1hTWFtSAYC1qvsJY8jKhnLbHMIoQ1gIvcBHrm//rrL+v94iWPoVy7dm3b2+8vzU/4EKFEJG77P2xDlTt6r+nNxBifN2+eNUwx4sP8hA1rvOCPOuooKw8MAF7kGIbAFl6P2267LVdI2x133GFf+o8//vhWL3IMIHrEkTfy4jgYCOPHj/cM04J64af7OHh/MJrxDCEj2gpAYDziPcPgcddEDgmQi6zcMjcl5HTPPfe0uoR88JzQm3366advta3bpzBMAY/LL7/cthsg4ZppK8YkYZ7nnXdeLg8A3h+Ws4+DFtdOPCXkB2KM4uFGL503Kr/hle5YmTSlnegEnh28aFwbnSIAP4byySefnKvoBc8h7kE6S/ztQP/oKCFMDU8m+gaUOG+UXxf9+xWWeTzVyIiiKsgMKMXbgzcfIPV3gKBj5GAhi1jdQCbctzzTqcbKc9t5o2I7mgqLbNx1AmJAP6CKjOhE4znLPUVHpN+Dzf2D55F3ITJxx2CKTNx+QBrvA7y4vB/9cvbvs63zCmsKawprQTAWtF5hLTF4qGcteflECGsYNRh8hHfwoWeV/wldxNjm42CNFxr5Q4QSUenR/yEZm5cd+2E48CGPhJclUBLmJ2xY4+WNBwjwQF5PPPGENX7omaWnmmXu5UwxCKqoEVLES94tZ4pxQPI+uTiu15vhEvC0kCvh37awzSMDfns8YOTrARDIh8p7GJZ+GeFdQw4AmT88iTbj8UDOGEIuRIkqh8iIgjWFTS7+68VgxJuIJ5WwM8LUdt99d+uJJp8II9BtjwwxKPFic9+55UyRH8VGKMzy1FNP2XUcC/gF7PzbFrZ5ZETJdDxhlI8n3BF4ANbJI/JDFvLiuUVhldhhL7hnKeNPRwGyRf94DuE94vnFeQqbbPzXy/AEQIcb8oPnDcCFNzIWNs4++2wLrLHDIPB8GjNmjN0PoEGegHGsN8p/3sIyz++NBwzvF50ePFfQFWSGPHhH+dtyxRVX2HU8t/zLkQmFXPA0InNkhnwpVLPvvvvm2ta/37bOK6wprCmsBcFY0HqFteRhRD1riWUWIazxsgeo8A7xwuZlRigWxmLsh5ccBiJA4v8QdoWXCQgB2oA7YI6eXHp0eXE64PPvV1DzYcMaYXgYQhR6QE4YgISfuaR1ByJMx40bZ1/s/vAaXtwYQSTv16xZM1foDQnwGKKENbnjbOuLPor90QmMvBNMARZ6+9ETvIeMQ8T1uLYxdZ5HcrT814oBRdEEPCEulBKjGgDB+5hXaX//MTJ5nvYB/YS8AllABp5ooNTJx10/gEubgXm3jCn3JaFv6CNyZj9khEwBWsDWv31hm6d9QD/FQOgI4ZmCXqErsTJCbjyvYnMe8R4BIXjiADwHwc4bBbwVZlijfb169bK6w3MGWfEMQR9cW93vTqcQ62JzHtmOzjRkdPHFF3set7lz51rAYV1hlhG6ctJJJ9mQV2REwRqA65ZbbvHa6vQJ0KdzKDbnkfV0ELEfgO86Ctz7gCFHnJwLeqqwprCmsBYEY0HrFdYSg4d61pKXT0SwBkDdf//91sjjZYZxSGgNBmLsh15pQtxcxUj25YvhgPGJ4UlhAz4sZ3Bteh8JKwn7Ezas4UlDLuShEWbEPL2s7mXvXtTICMOSkC3WufV4QoAQwrWcMc06DCYgl5DTWC+cO2ZhmWIQka/mKu4BDuQqxhqPhA0BLBibtNnJyHlC8Hy4anWsw2Ak3I8QP5fvVlhkEnudGHkYx3ig0QXkhactNjSN8DPneaQjxMkIeTH2kx9CWEdlQHKOMMoJeYs9b2H6n2IgeNUY85GODWSELIA4fzswnBlPDu8tIaFuHbLEE4KMGDKD/50ekdvGM44OKrd9YZyiHxTe4Z7AS4uMgNpYbxFtZ4gCoiH8nkfuSZ7xQAhDZjgIQRaMh8nxCnsBFtrOs4j8RGSDhw19ii3HjywIH6WDyK8X6AzPfe4pxnb0RwC4glxhVl1VWFNYU1gLgrGg9QprycOIetYSyyxCWONlDXzwIgPcMH4IvcJIdh/y2uh1xMjmxe4+lOinAAKGOTkj5IXwIuNLCAoGE8n9LKeHF4gL4xMmrCGHCy+80MqIFzc5ewAWBhBhWs7Y4+WOLKtVq5YrMZ/98Z4h42OPPdaGZNGzz5diABwTCET2FJ/gOO6YhWVKGxlklzZiAOIJw5CmbS7c07UFAxogI5/RLWN/PJLoEfqCLJyMqL6JrNFJDEw8UrEA6I6T6VMgFBkB/OSnkQfJPRIbeoXBzDZ+zyMyQl/wglOxjh5/JyPkgkGK8U7eEfdbYYV/xtSjjXzxjOBFxNCOrUzIwNeAF3JANu63p+3IlCq0VGJ1MkKngD5AEDlOnTp1q3BBd4xMn+KtxrMKVBEOy/OJe4d2OVnwHGEAde5H4NUtp214JIkOIFz7zjvv9GSErOh4Yx+ec3jqYgEw02Xjrg84xbvP/UabkAEyIkzUbcMUWQJqsV4yOh55lvPsYRB7p0dMeReinzznyMdeujT+0Cz+8yQ7r7CmsKawFgRjQesV1hKDh3rWkpdPRLD277//2qphhCo6mMJQ5KXm964xuDMGkH8Z0PX7779bo4jeSypA+r8Y5Lwo8SCwnJdlWJ+wYY0cKtrBEAaAAi9oZISX0RlB5AzR5liPG0YzuVkYDsgJA8B9MZiQER5LlhXWXBo8OxQ1wMijHDa92hSHwKABYJ2hQtgo7Rw6dGgu4MKwpHpoPBkBfMgI2bIvHkonc3fcwjDlmunQIK+M0Dz0Ap1BZrQJmdEOwkkJjcTz6JaxHBmRyxVPRugmMuJYyMjlVhYGucReI88JIAR4p2MInQGwyA9yuVgABF4TQpH9HjdkRPgj28feayzjnuU5howAN3e82GvI9P8d9FO1EY8POoOHDY+t8z4jO8r0s8w/pAhtAz4AlFgZoVvIHghGRnTchQEi6ZAvIbS0hTw12o/OtG3b1nZouMHo0Re8ZnRy+DveuD68ZvFkhMzoPODZxjw5ufEqtW5rGxXWFNYU1oJgLGi9wlryMKKetcQyiwjWCLHCiMGIdB8MRIxFeqYxwhkfjVBGCiHwAvF7x8hLIySHl2HsF28SxiMGKet+++03d4oCn4YJa1w7L+Q+ffp4gMGLn0IahH7y4kdmrMdgihdmgyHAC93/JUeNkDZe/HgDWBdbSGJbX/jp2h+jF30BSjESMYKYAg0YPAxPAORS8AGDkP/918b2hO/55cM8MsJ4B0TwGLGssBqPwBn3EQawC22k3fTSAxB0iPC/8zzGlgRnHUZ5rIz4n04UZHTVVVdZmbGdX76FZR4dcWF7/M60GcjlWcJ98uSTT9p2kTeK59GFFLv2sT1htvFkREgboIwHBb3iPmV7t29hmrqhVgi7djKiWiEwyn2CzIhuALrwKMa2jecM+8bKiaEN6BQhfJl1PLcKo4xo/8SJEy1Q+e8jl9fJEBBsg0ebnEfum9h2Ar3xZMT7jFB3IgDoSKBoVBiefoU1hTWFtSAYC1qvsJYYPNSzlrx8IoI1xgyj95GQIj+EUWiEFz9eEnoYMYzwKiXzYSwkxshiOICwP2HCGmEyGD3+8Ble9ITnOQ8A4WrICK8k62Jf/LHGEv8DwgAeXkcMzHjbFJZlGCx4CYEx2uWum5BFDGQ8hoQ9YghSzCAZGXFMQr4wnNxxC+PUVVLlN/fLCEijU4NwNnLa8GggL9qYHz1iOwx1evoxJAujbNw14yUj3JUwT7eMKVVpgXwK9KBrdCbhjQSA8yMjtiEkkmdaLOD5z1NY5vEo4jHzexUBdDqV6EQCspiSh+zPtQpqH0V9kFFhr7rK/dW7d29vqBXXbgCM/GA61YBR9In/k+kk452ZjqqrCmsKawprQTAWtF5hLXkYUc9aYplFBGvABS8eegj9H4wgDCK8AHwp8LBx40b/Jgnn6WmkuhYhKOS1hf0JC9aACue1cNUJ3YsfjyLVDmkn4TV4Hlnm1gdN8a4gW8qSO29U0D6Zuh5YxbNDLz4yc9eJMUn7CNPDACdHJpnBiJEL3jrGxnLeKHfswjalhx+gwlPolxFeWXLXADby/IB37sf8ts95dQnlotBGfuAlv8dO93YAOUBPPp//3LSJaoY8qwiXpWPAPzi9f9t48+xPOC5e3sJedZWOHfKL8UD6PTroFDlUdCAhIzqPkhkOhP3pkKJjiqqrhVmPeFa44lZ+oKeNzrvGsBi0lbw8//0YT3/cMrbDqwvQJiNbt38yU4U1hTWFtSAYC1qvsJYYPNSzlrx8IoI1ksupIkfuWeyHvCwKaQBtDOCbzAdoIYeJUtnp+IQJa/Qy4w2JDS3jxY1HkuIghEMSypjMy5gwL0Lg3ICtyeybadsSYsY4T7GQgYzwZJBvhIwoOpPMtWN0kaflvHHJ7Jtp21LUARnFG34AryMhksgIL1kyhjLGKGGChDLn1+jMNNm466EaHx4RCu24ZW6K1xEIAUqHDx9uZZRfObEdRTjIlQRu3TEL45TCGf369fMqpro20EZCOxlgHQ8buVjAXH5lhO4w4DP7cd+54xbGKZ0WtIPcUP89gSzwrjGkDNU0qcRKh1IyMgL2eB+EHQ2hsKawprAWBGNB6xXWkocR9awlllkEsOYPe4wFKrcObxova7xjblnstkH/p7pf0HH968OCNV7ihNTw4ow1WliHIYCxjHGAEZjfl747Fsfl+Mnu5/bPhKmTQyIZsQ1w6u/lzu+1Z4OMnCyQEd5E939+p8iIb2HWI9dW2uE3sFnu2sVyZJRKYZC87mN33sI0dTJycvFfO+3keZRM+KOTcVGTUV7PVuRH6GOyUIq80yUjrjHeZ8naD+TV7+6TIEO+aKyfLKv/WBZPDAmXrdiwKCvkgx4sXrMgoSzirvx3kQg2ZzzHSpFaprCW/I+ssJZYZhHAWtybuJAuDAvW/EaQzq9NGiJUZioz1QHVAdWB5HVAYU09a0HArbCmnrXEYJEKeSusJZapwto2YaLCWvLGgBpQKjPVAdUB1YHM1AGFNYU1hbUgGAtar561xOARD+YU1hLLTGFNYc2Ee6nhpDJQHVAdUB1QHVBYU1hTWAuCsaD1CmuJwUNhLXn5KKwprCmsKayqDqgOqA6oDhgdUFhTWFNYC4KxoPUKa8nDiHrWEstMYU1hTY00NdJUB1QHVAdUBxTWNhcI0QIjiYBNc9YU1hKDRTzPWdAyhbXEMlVYU1hTI02NNNUB1QHVAdUBhTWFteV4FhN/FdYU1hKDRRCYxVuvsJZYpgprCmtqpKmRpjqgOqA6oDqgsKawFgBqgJzCmsJaYrCIB2NByxTWEstUYU1hTY00NdJUB1QHVAdUBxTWFNYU1hLbzEHMYddrzlryQlRYSywzhTWFNTXS1EhTHVAdUB1QHVBYU1hTWEuGTyoGAAAgAElEQVRsMyusBbkVU1yvsJZY8RTWFNbUSFMjTXVAdUB1QHVAYU1hTWEtsc2ssJYijAUJTmEtseIprCmsqZGmRprqgOqA6oDqgMKawprCWmKbOYg57HoNg0xeiApriWWmsKawpkaaGmmqA6oDqgOqAwprCmsKa4ltZoU19axJvpSggOWksKawpkaaGmmqA6oDqgOqAwprCmsKawprkcCIetYSK57CmsKaGmlqpKkOqA6oDqgOKKwprCmsJbaZ8+VU0TDI5IWosJZYZgprCmtqpKmRpjqgOqA6oDqgsKawprCW2GZWWCvg8D4nUIW1xIqnsKawpkaaGmmqA6oDqgOqAwprCmsKa4ltZscWCafqWUteiApriWWmsKawpkaaGmmqA6oDqgOqAwlg7Zu1H8ir3923GWbuL+LTyfLTH8uStg1WbFhUxOWS87ujB4vXLEhaPvLvIhFszoSgE5LjJq3nVFhL/kdWWEssM4W15B84vj3++OMPWbNmjb7k1dBTHVAdUB1QHSj0OrBhwwbfG27L7LfrPswqWPv5j+VbGp/POYW1AEEprHkC2s6bC5r5687ERnxaKTREklZYS/w7K6wF3SkJ1yusrS30xslaNTD1N1QdUB1QHbA6kBes/fDbV1kDa7OXPyDr/lqd8N0fb6XCWjyp+JYprHnCUFiLhcxUYW2T6V368yrzvSQx7MSer1D9b9r2x6Ui/1vhKZDOJCcBhTWFNYU91QHVAdWBoqIDecHan/9skDd+mCavLi/aoZC0b+GqZ+Wf/21MzhgwWyusBYhMYc0TkMJaLCylDGv/ify3yoDMsiL+/UFk07+eAulMchJQWFMjragYadoO1WXVAdWBvGBt06ZNsuavH+SD1S/JWz88Lm98P13mFaHvm6Ytb62cIZ/8PEd+3/ir0N5kPwprARJTWPMEpLBWULDmiVRnVAJ5S0BhTY0bNXBVB1QHVAeKig4kgjXehJs2/Sf//rfReJ7+lo1F6Et7aJeDNDfN++2/9RqFta1lkmuJwponDoU1hTVPGXQmfAkorKmRVlSMNG2H6rLqgOpAXrAW/tu08J9BYS3gN1RY8wSksKaw5imDzoQvAYU1NW7UwFUdUB1QHSgqOqCwlrrdoLAWIDuFNU9ACmsKa54y6Ez4ElBYUyOtqBhp2g7VZdUB1QGFtdTtBoW1ANkprHkCUlhTWPOUQWfCl4DCmho3auCqDqgOqA4UFR1QWEvdblBYC5CdwponIIU1hTVPGXQmfAkorKmRVlSMNG2H6rLqgOqAwlrqdoPCWoDsFNY8ASmsKax5yqAz4UtAYU2NGzVwVQdUB1QHiooOKKylbjcorAXITmHNE5DCmsKapww6E74EFNbUSCsqRpq2Q3VZdUB1QGEtdbtBYS1AdgprnoAU1hTWPGXQmfAloLCmxo0auKoDqgOqA0VFBxTWUrcbFNYCZKew5glIYU1hzVMGnQlfAgpraqQVFSNN26G6rDqgOqCwlrrdoLAWIDuFNU9ACmsKa54y6Ez4ElBYU+NGDVzVAdUB1YGiogMKa6nbDQprAbJTWPMEpLCmsOYpg86ELwGFNTXSioqRpu1QXVYdUB1QWEvdblBYC5CdwponIIU1hTVPGXQmfAkorKlxowau6oDqgOpAUdEBhbXU7QaFtQDZKax5AlJYU1jzlEFnwpeAwpoaaUXFSNN2qC6rDqgOKKylbjcorAXITmHNE5DCmsKapww6E74EFNbUuFEDV3VAdUB1oKjogMJa6naDwlqA7BTWPAEprCmsecqgM+FLQGFNjbSiYqRpO1SXVQdUBxTWUrcbFNYCZKew5glIYU1hzVMGnQlfAgpratyogas6oDqgOlBUdEBhLXW7QWEtQHYKa56AFNYU1jxl0JnwJaCwpkZaUTHStB2qy6oDqgMKa6nbDQprAbJTWPMEpLCWYbDGg+/tt9+WX375xfuRgmY2btwoN954ozzwwAPepps2bZJ//vlHVq1aJStXrhS2YZl+opWAwpoaN2rgqg6oDqgOFBUdSBesffPNN7JgwQL5999/8/0SZ58xY8bIokWLvH2wg3777TdZsWKFrFmzRv73v/9FZhsprHk/S/wZhTVPLgprGQZrzzzzjBQrVkxeeeUV70cKmpk5c6aUKlVKHn74YbvpX3/9JY888oi0b99eqlevLjVq1JCmTZvK+PHj5eeffw46nK4PUQLpg7Vf5PcNb8of62eZ75w0fWfL7+tfk7VrfzZfNcZUBqoDqgOqA0VdB9IFaz169LB2DO/Q/HyAsOOOO04aNGhgO6yBtB9++EHOOeccadiwoVSrVk1q1aol++yzj2B3JQOB+Tl/frZRWAuQksKaJyCFtQyDtQsvvFAqVKgg3377rfcjJZr5/fffZY899pCuXbsKDzE8aGeccYaFt549e8o999wj06ZNk0GDBknJkiXlqKOOkj///DPRIXVdiBJIF6ytW7dC/vvzUpE/L07r978/LpZ1a5cprCmsqg6oDqgOZIEOpAPW1q1bZwHriCOOsJ6w/LyiX3vtNSlTpozccccddnO8bK1bt5bKlSvLqFGjrF10++23S7t27aR06dL2/3RHHymsBfySCmuegBTWMgzW9ttvP2nRooX8/fff3o+UaObBBx+0EOY8cc8995wUL15czj77bMHD5j70Gp177rnWa0cogX6ikUC6YG29gTX580rzHZvG7yWy6Y9xBta+UyMtC4y0ou4x0PapV0x1IFgH0gFrX375pe2Avvzyy/P14qbT+oADDpDddttN1q9fb71mxxxzjJQvX15mzZol//33n3ccUkWaNGkiu+++e9q9awpr3s8Qf0ZhzZOLwlqsMfv3Y55w0j1D/HSdOnXk+OOPz/Uwobdn6dKl8umnn8qvv/7qXRZ5bYQ39unTx+ansaJ37972GKtXr94qDptcOEIs77vvPu8YOpNeCRRtWBursKaQpqCuOqA6kEU6kA5Ymz59umy33Xby6quv5nphEyX0+eefyxdffOHZQGwwY8YM24n9xBNP2O2JVALUhg0btpVdxAZEI+FxA9zS+VFYC5C2wponIIW1DIK1jz76yHrFJk2a5P1AeMTuv/9+m3u2//77C658PgAcvUyETH744Yd2GQ+utm3bysCBA+OGCvCg23777eWpp56y2+uf9EtAYS24p1Z7s1VGqgOqA6oDhUMH0gFrI0aMkB133FG+//57+9LG/qFDmo7tnXbaSS655BKbAsJKQibbtGkj5LjhYeMze/ZsqVixohB5FO9zwgknSO3ata0XLt76sJYprAVIVmHNE5DCWgbBGtUcgak33njD/kCEMV566aU27rpfv365en3wtFE8hIcYHx5egB09TFQ54n//B5Aj3hu4W7x4sX+VzqdRAgprhcMAUUNRfyfVAdUB1YFgHQgb1gCuLl262Nwyl9pBpzXLALC77rorV/giOWrkqs2fP997s/M7fvzxx0KOf+yHiCW8auT4U5QknR+FtQBpK6x5AlJYyyBYO+mkk+xD46effrLhjieeeKJNfL3gggty5Z8Rb33WWWdZWFu2bJn3Y8abAdp4QFG+FhDkWP547Xj76LLwJKCwFvzyVwNJZaQ6oDqgOlA4dCBsWCPdg4rWVHbkA4SR/kGVR3L1sXFc5zTeNpbjcQuq7sg+X331lY1GqlSpkrz77rvhvfjzOLLCWh6CcYsV1pwkRGEtQ2CN3qMOHTrIXnvtJZ988oktJ1ulShVbjj+2t4ewR3qUrrrqKu8h5f2ivhn2e++99+TAAw+0yblnnnmmHV/Et4nOplkCCmuFwwBRQ1F/J9UB1QHVgWAdCBvWgKgddtjBVnVkeCLson333deCVuzrm85owiKJMEr0YZw10kvq1q1rc/yffvrphLZUomNtyzqFtQDpKax5AlJYyxBY+/HHH61XjbHR6DUimfaQQw7ZygsGgBESyTZ5jZlGjxHjiYwcOdLGee+8884W+hgkWz/RSkBhLffLnxwExrgZO3asTfK+5pprhHEDSfQuDIYSg62+8MILgod7W66XwkGEP8+ZM0foSd6WY6W6L88TKqXRc02xo3jHWblypbz44ovy1ltv5blNvP10WW69V3moPIqKDoQNa4Q1AmtHHnmk7XTGNgKu+DiPGvN4yQA1yvL7l7POfbCf5s2bJ1TdZmxaCrJ99tlnbnXapwprASJXWPMEpLCWIbDGA4QHEg+iww47TA4++GApV66cjbP2fi0zgzHFQ+buu+/2L/bm8dAxIHbjxo2t940S/oAgn7weYN7OOhO6BBTWcow0YOD111+3YwQSnkuVUsYB5B5g6Am8weQYZLJBQyL7xIkT7T1L1bFkrvXZZ5+Viy66SIA99uMeJSm+WbNmtqMlmWMV1LbfmoppNWvWlG7dutmOoHjHZdgPnlFdzbiOeQFdvP10mcKJ6kDR1IEwYY2UjZNPPtk+c7CHRo8ebTu1u3fvnqv6IxA2dOhQ6yWjozr2g+1DiCQ5/mXLlrXDI02dOtUrQBK7fbr+V1gLkLTCmicghbUMgbXrrrvOGqrkouGiX7hwoU2SJW+NBxEfcs9w/zMINkZ/7IeX4emnn25hDuD74IMPYjfR/yOWgMJajsHCMBKEoABp5BdQqRQ4w8vGUBQAHF7mpaaQTqYaecAankCu9cknn0zqOs877zwLpXjTaB+GxCmnnCJUJYvKq4isGTqE8YnwssWTO7AGTFOZVmGtaBrf8X53Xaa/dV46ECas0fnMoNUUU6ODC+iiA5pnkBtbllc60QDA3M033xz3DY89RaVsctMo2uaGQIq6A1thLe7PtWWhwponC4W1DIA1eo+OPfZYqVevnvcQYdnRRx9txwYhh42HCh6z0qVLy0svveT9gG4GoDv11FOtd4KHkYO5qB9G7vp0miMBhbW1VsepvIWHhhwDQMVvCOBlOvTQQ+16PMhAEeuBA3pNqd7Fl5A8PzAwz74U6GH+66+/FobDIETRbccUGOLrlrlzE37I/v4wRF7qVE/lOEz5320fD9Y4N8fwb8f2tNEt59znnHOOvVcpKe2uhfbEaxNt5hlAmzmG/7qZZ5lrs7vW7777Ltd2Tn5sSz4HYExFNf91FjSscW1ch/u9CHl11+5kzXU7ebprBBSdrNw6tmM8JeSAPNxypsiWL8dmynbI0b+NzitsqA6EowNhwhrPDEr2n3vuuZ4JwXOLcEeij4A5vrwvGACbZ3KszcNzrVGjRrbwCNEcmVRgTWHN+1njzyiseXJRWMsAWKMcbevWrW0ctffLmBm8D4SGnXbaafYh1KpVKxsiSe5Z7AOJXqcSJUrI8OHDbRUkHkjxvrH7+c+n8+FLQGFtrfX40gtK1S5C75wB7zemXn75ZRsKOWHCBLseA37y5Mn2PqlatapUq1bN9rg++uijHlyRs9CpUycbKkO4C547SjKTs3nDDTdYbxFGPLmgbAfYuHPykr/22mulZcuWdiwelnM8PNvuOHidqEjmQhdjYW39+vWCx2z33XeXL7/80js2xsyQIUOsR/z999+362kDHjmMCDxZGCAMrQHEAiqcnzYzgD3PBpLqXZsff/xxr82cp2PHjrbaK151d61NmjSRm266ydsOKJsyZYotYkRlNeRSv359681jqA/OV5CwhpEFiHMdtJVvixYt5LbbbrPXhNFEuwhdcr8BU3SB6AJCQpEV/5Mfh4y4bo6DUfbYY4/ZdcicTio8fa+99pp9hjJeEnl1/uPqfDiGuspV5RomrHFPExrP/e7/8KzD3iF9hHsdO4mwxlj7hk5sOsIp5c+28Wwityx2X//5wppXWAuQrMKaJyCFtQyANYwkjFeKLPg/lJ6lmAg9Sxif5cuXF8KQYj88kEi+xVPRq1cva2BiZMZ+qQbJeGv6iU4CCmtr5aGHHrK6Si5CPFBzyzDEASIMQrzFhEwCI+wH/JBfxUv6xhtvtNvgwaFKKvcJ6wYPHmzvG4x3wmZ4qXNMOjS4V+69917PoMcjA8ABBEAaIAcUcE5AAIAAGDgOnSYAWyysYbRgGHAvcy3OkCWsmRw8QnCo5ErYJ+ACrHFMOmOANcJ0WO48RzwPOD8ed0IkaTPXR5sJ9+H4eMgYO9G1mW1oX61atey14oXnOknIZz9yWXmWkITP+ZDDFVdcYeVSULAGGHINtI+wbfJMhg0bZkGL66QTCphDjkAjCf5OVkuWLLG/HcCLx3Hu3LkWUmk3Q5sgB3SA40yfPl2QOSHfyLxhw4b2WUlVXRde6o6rU4UK1YFwdCBMWOPZTo4+zwg/TPF85nnKc5VnBQVD4tk2RFWwHZ64WHvI/z8dgVF8FNYCpK6w5glIYS0DYI0eZHrH6UWK/WDYYFTRs05vkstf829HGMCgQYPsQ4v47ry+hFX+/fff/l11Ps0SyHZYAxyuvPJKCwmXX365Z6QnMqQAEox6PFvvvPOOBTyAjsqJLMd45+XtYA14AVLYhvM5ODz//PMtlODVoVf28MMP9zxPlIcmxPiYY46x+7AtIAMcOE8X8MA9yHLCcgA/f85afmANTyIwgwcOeOJa+J9z+GENqMPAoG1cG23hS4EhluMVA2wcrAGp5P25Nj/44IP2OiliAixy3XT6sL+TNXkeyIGKaFx7QcEasAno0p7ly5dbeXJO97sDWVzn+PHj7TVOmjTJ24ZQb+SLXAmVxBijbYSLsg/H4brxMgLXeB+BNfbBK4o8AW9k6tqp03CMdJWryhUdCBPW6KQj3JH3pv8DuNGZRXQGz4J4thPb804gxx+gy8suYjkdd1F8FNYCpK6w5glIYS0DYM37NWJmXE8SYY/09mMc6qdwSyDbYQ0dvuyyy6xxzTQvgw/D3BnnDrbwbgFfbh3zeKkIk3nuuec8WOtqKhX6c6EotMM2eGY4P+soXkJIHfcV18CYhXiCKHDCesY7BPrIVfNfIzlTQA8GAKCQCqxxPDxbwJrzAPlhjVDNBx54wMoIKIxtM947PG54CrkerhPvH9fjrpXxFWkPXjuMKXqm6fjhPIROArrIk20whgC6goI1IIsIAICT8zElTJvfBahirCSuk9+Fa8f7xj6000USAKH8NvxGeMoIi6WgAF+Ak2NR1Y3CAQ7WCBl17depgoTqQHp0ICxYc/ZPojc+zzw6rejEzs/2iY61rfsnOnZe6xTW8pLM5uUKa56AFNYyGNa8X0lniowEsh3WMMjpxcRoJ1/JAVmsYYWnDJggx8nBHePtsL/blnkgi2PdeuutFkjwwvTv3z/XcQEuP6yxP/lp7MfAqLzw9957b5tfRXgeX3KtyCfDM+TOx5T8LsY4pEeX+TBgDcAZN26cvT5/gRXOD2wStsi1Iw8HawCcX5YOUMnnYh+KrRCOSOghIYV4JAGhMGCN66CnGy8lcqKSG6FInJPrBtbYBu8XXj0HXYSW0ktONVDgjVw1QqAAU3JO/F9Al/3woAJrzAOofhn4fzedT4/hrnLOPjmHBWtF5qWfoCEKawmEwyqFNU9ACmsKa54y6Ez4Esh2WMOYw8ODEU6OEbla8Qw8imFg2JOfRn4W81dffXUuWANC8FABHHjf8B4BawMGDMhltMeDNSAHgKCKGEY+OU/ADNcCLBHGR94Xvbb+6yOMkXBMoIft8gNr7A8Mcj725/9EnjWOS0EU2gxU+gGVNjPYPW0mod7BGh5GP6j4YY3leKwAVgCKMERAiHbj2SpozxrVGAnTxAN58cUXW68a1wN4OlhzMp0xY4ZtC3DKetpF0RjW87s5zxvbMZZd7JecFGCN3z32t3Ln0Gn2AYT+5un7zRXWUrcbFNYCZKew5glIYU1hzVMGnQlfAgpra22YYefOna3hTk6CP3wPI4ucJ0LjHIQR+gZoEOpHPpIzxNiOcEbAjzwml7OWH1jDc0PlRXJBx4wZ4+WPOY8P3h3OCRC58zGlKhmeHiCP//2wBkjhLST3jdw6tx9eQgAvv7BGGCQhjrSfnK3YNpMLhpeJc7ictUSwRngj4LnLLrtYb6C7Lrxf5KwVNKwBW1w7RUWQiTsfOSZ+WEPWeDGbN29u80kOOugg6810ww6wjkqeDBTulrljAbuuiqaDtbzA3+2j0/QZ8Crr7JG1wlrqdoPCWoDsFNY8ASmsKax5yqAz4UtAYS3HiHnhhRcsvGDU4z2jrDKeseeff96W1seo79Gjh60ICKwAasATYX145t58800ZOHCgNf7xGgFfycAaxiRFODgmcEUOmh+KgCWAiAIfePnw8hC+R1gf23OdgIgf1vCAMbg9104lSnK1ACIKmbDMD2t4nGg7RTeoeAiY+AuMcC1UOOP6gB7aTH4bIMqxjjrqKNvm/MAaoIOni7DHp556ysoJDxXn41i0HaDjC1QCQbEA7YxvNyj2rrvuKtOmTYv7JbQUoO3SpYvNKaOAElXdaD/no7CMkzUyA9hZTlvxGjpPIjDH0A3IqW/fvjbPDk8inlbCHrlOQikV1rIHDJwe6jRzfnOFtdTtBoW1ANkprHkCUlhTWPOUQWfCl4DCWs5YWhjZeGDwqjhgohw7nh48ZcAIgzc7owzw6d69uy1H73KXKKNPzpPbLhGsAQ+uwIg7JiGJQAygMHHiRO9czrtGvhxDAHAewiSZkn8FLLiCGH5Y47jkXVEQA8BgH66VNlLt1Q9rlNInTJDrwntEMQ0/rHEsQgcBEs7r2uyqWLpx3IJgjQIjQCVwCGSS64WcubZ99tnHhqIif0rtJwNryIxrj/1yrXgj+a04LlDFuWgrw4rQDvaZOXOmJ29+W9azrd8jye9AXiBtYD/0wukIFdzYFkMRDylhkOpZyxwD3t1jOi36v4nCWup2g8JagOwU1jwBKawprHnKoDPhS0BhLbfxQp4RHitK5Z999tl2IGdK0FOREWPd/yXsEY8Q3ig8LlQYJL8Lg5DtmKcqJEY8/ztDkbBCqjxSOdAtc/vg+QKcyH3yr3PrASYKeeD9Ad7cQM1uW7yB7O9AgfPSJsbtIQ/rzjvvtNUXuSa8ca5dwB6hm1wX4MKwAFQ5JOSTda7deNzI16KoCG2mff420zaWufL+7rpYznWRl8YyzktFRaAULxfzyBMPJcuYsg3X4oYTcMfyT7kejouHLt7XyRJvHtCGF40qjbQVbxqAzvld7h7HRnYUFsET59ruPydePnSC/D1+e47B8d02XLuTrVum09z3mcpD5RGWDiispW43KKwFyE5hzROQwprCmqcMOhO+BBTWchtNDqrw/vD1h8DFGhduW7ZJtF3sfgXxvztfMsdKZZ/Y4xd0m/2yc+dy1+nO5ZZvy9Qdyx2bY8UuA8wARFdM5Z577vEALN65/dfujhVvO12W+x5Teag8wtQBhbXU7QaFtQDZKax5AlJYU1jzlEFnwpeAwpoaTmEaToXp2HjcKCpCaGObNm22KiJSmNqi16r3dbbqgMJa6naDwlqA7BTWPAEprCmsecqgM+FLQGFNjbpsNepi233XXXdJixYtbPEYCszErtf/9V5RHch8HVBYS91uUFgLkJ3CmicghTWFNU8ZdCZ8CSisZb7xoQZi+L8RYYx8yb9zIY1uqvIPX/4qY5VxQemAwlrqdoPCWoDsFNY8ASmsKax5yqAz4UtAYU2NpIIykvQ4qkuqA6oDUeuAwlrqdoPCWoDsFNY8ASmsKax5yqAz4UtAYU2Nq6iNKz2/6qDqgOpAQemAwlrqdoPCWoDsFNY8ASmsKax5yqAz4UtAYU2NpIIykvQ4qkuqA6oDUeuAwlrqdoPCWoDsFNY8ASmsKax5yqAz4UsgnbC26c/LRf68xHzHpu276Y+xsm7t1mOWRW1Q6PnVqFUdUB1QHSh4HVBYS91uUFgLkJ3CmicghbVYQ/bvxzzh6IxKoKAlkC5YW7t2tWz8/T759/dJ5ntbmr63mHPebqr6rdTKfmZcMTUMVQaqA6oDRV0HFNZStxIU1gJkp7DmCUhhTWHNUwadCV8C6YM1BiHmuy7t36JunGj71ABXHVAdUB3I0QGFtdTtBoW1ANkprHkCUlhTWPOUQWfCl0A6YU2NCTUoVQdUB1QHVAfC1AGFtdTtBoW1ANkprHkCUlhTWPOUQWfCl4DCmhpOYRpOemzVL9UB1YF06oDCWup2g8JagOwU1jwBKawprHnKoDPhS0BhTQ2pdBpSei7VN9UB1YEwdUBhLXW7QWEtQHYKa56AFNYU1jxl0JnwJaCwpoZTmIaTHlv1S3VAdSCdOqCwlrrdoLAWIDuFNU9ACmsKa54y6Ez4ElBYU0MqnYaUnkv1TXVAdSBMHVBYS91uUFgLkJ3CmicghTWFNU8ZdCZ8CSisqeEUpuGkx1b9Uh1QHUinDiispW43KKwFyE5hzROQwprCmqcMOhO+BBTW1JBKpyGl51J9Ux1QHQhTBxTWUrcbFNYCZKew5glIYU1hzVMGnQlfAgprajiFaTjpsVW/VAdUB9KpAwprqdsNCmsBslNY8wSksKaw5imDzoQvAYU1NaTSaUjpuVTfVAdUB8LUAYW11O0GhbUA2SmseQJSWFNY85RBZ8KXgMKaGk5hGk56bNUv1QHVgXTqgMJa6naDwlqA7BTWPAEprCmsecqgM+FLQGEtx5Bas2aN8P3111+3+rI8LGPDnTPoHPndLpXrzM+x3TZ++bCMb+w53bbx1sVum+z/7pxcRxjHT/Z6dHsFEdWBzNIBhbXU7QaFtQDZKax5AlJYU1jzlEFnwpeAwlqOobFixQrp27evHHDAAbm+3bt3lyOPPFIuv/xy+fLLLwscEGbNmiWc4+mnn5Z169ZtBT4Ygix//PHH7Xavv/563G1SNRg59tSpU+XAAw+U+fPn53nsyy67LJdckBPX3bt3bxk5cqQsWLDA2/fzzz+Xww47TG655RZvWarXF7vfzz//bM83cOBAWblyZYEfP/Z8+n9mGeL6e+jvEaQDCmup2w0KawGyU1jzBKSwprDmKYPOhC8BhbUc4+ebb76RnXbaSajoQIMAACAASURBVIoXLy4NGjSQRo0a2S/zFStWlO22206aNWsmH374YYECwpQpU6RYsWJy5513JoS1q6++2l7DU089VaDnB9auvPJK2X777eW5557L89gAKzKoWbOmJ5v69et7smH+rbfesvsDbmXKlJHhw4fnebwggyuv9T/99JN06dLF/kbfffddgR8/r/PqcoUE1YHCoQMKa6nbDQprAbJTWPMEpLCmsOYpg86EL4F0wdoaE7a2btUyWb/6a1m3enHo3/WcY9VSWfPrL/ky6B2sNWnSRJgHBNz3s88+kyOOOMLCyqmnnpqv4+XXsMML9cADD8gnn3yS53EBqkyBtWeeecaTC/JBNieccIKVzYABA2wIqcJa4TBq86ujup3+noVJBxTWUrcbFNYCZKew5glIYU1hzVMGnQlfAumCtbUrl8j/5p4sm17rZ7590/L939whsu77L/OEIL8B4mCtefPmQqidW+fyoubNm2eBpFu3bt568qbwJk2cOFEuuugiefDBB2XZsmXevhyDUD0AhzDKsWPHWjD76quvvHDKxYsX2xDHRYsWeftxDI7FMYG0t99+WyZMmGDPj2cNeGPZE088Id9//723H8u5TpavWrXKLucaP/jgA7njjjvs8QhnJOTShRAm61kjbDNWNm+++ab1Dnbt2tXKJh6sIUfAlOsYM2aMbRchnb/8khum2e7999+XG2+8US688EIbSsnxaAfnzcuz9vHHH8tjjz0mL774oretu06dKmyoDmSPDiispW43KKwFyE5hzROQwprCmqcMOhO+BNIFa+t//ExkTk+Rub3S951zqGz4/n0PLhIZbHnBmtvnhRdesLBEjhaAwfeaa66RChUqyA477GDDJwkT3G+//eTrr7+25/zhhx+kX79+3voSJUrYY+yyyy4WSAAloIz9br/9dgthH330key55552GeGRHHvHHXeUzp0722XAGsbISSedZM+5cOFCr30s79Onj5QuXdrLryPMsmrVqjbMkeUck5DHIUOGWLDaFlhDNsDVtGnT7LWdeOKJcT1rbANAElJKWwk15RrKli1rAdYPlngZq1WrZrdDXmxXqVIlmTRpkj12PFh75513BI9oqVKl5KabblJYM7+L01udqiyyTQcU1lK3GxTWAmSnsOYJSGFNYc1TBp0JXwLpg7VPDawdlj5QAwotrL2XL8PVwdrOO+9sPUB4uvjiDXr22Welffv2FjLuvfdeC1WPPvqo/Z9CG7Nnz5b33ntPRo0aJQDG4YcfboGBoiDACcU2gCq8PxdccIFdBmxhVABTbIPHCY8exwOoTjvtNFvwA+8TxTSAFrZzsHbyySdLyZIl7XGdMcbxgEMgiGvHQwcg1a1b13qdvvjiC8EzBiyWL19ePv30U9uWZHLWzj33XLn55pvt97rrrpMzzjjDwmCLFi2st49rifWs4dmrXr26hS72Ie8PeNttt91sm+6++277G+GlJD+QvDiWIa8nn3xSGjZsaJcjw1hYoz2sr1y5sgVfrRKpcOLuB51mpy4orKVuNyisBchOYc0TkMKawpqnDDoTvgQU1nIMGgdreLIoNILxzxeowRPEF9ACFvCq7b333hZAAB5nFLIcUCtXrpyFN6ohAlhUmQSe8DARfkhIJDCyfv16D9YoMEIIIwDGsZ23iWMT6tixY8ekYY1zcj3XX3+9d41cP9fDdeGRStazBkg6eTDPcfjiUQSmaGMsrDlAveKKK7zroF2AKPJt27atrF69Wk4//XR7LGTDdTm5Tp48Wdq1ayczZsywQEuBEQqaAMx16tSxvwNAzbndPjrNTkNdf3f93RXWUrcbFNYCZKew5glIYU1hzVMGnQlfAgpruWENz84xxxwjFMvge/TRR1uYAOI6dOggS5culW+//dZ6qwhP7Nmzpy1fT3gkYIRnDi8YgIFnqHHjxhZA8G6xnjA9crJcDpbzrAFr999/v9123LhxFuSc4Qm4jB492q5LxrPG/pwHKMMjeNZZZ8k+++xjwwVThbVbb71V5syZY72JeLXwHlIpkjbjfQQs/bAGQPXq1ct6+1juByoAjX3wuuH1w6sIKFO0xG3npg7enGcNqHVVOvldfvzxRwU183s7ndGpyiJbdUBhLXW7QWEtQHYKa56AFNYU1jxl0JnwJaCwlhvWKDACIPDCd19A4JBDDrGwBFBRIASvG+GGhPK1adPG++IlwgtELhrGEhBCoYx9993Xyx0jz41iI4CUH9bIywKibrjhhlyeJUDlqquuyhesAYQuDBJw6t+/v/VekbdG3tuxxx4re+yxhz1WKp41f4ERZwzSjr322suGgL766qtbwRqeMMDW74VkXxf2iTwAWGRHCCReTndspvwe7utgDTm1atXKwifzl156qQd4/n11XqFFdSC7dEBhLXW7QWEtQHYKa56AFNYU1jxl0JnwJaCwtjWs+atBYugBS3i+gAI8XMAbUNeyZUvrZeN/voQuUiDkpZdesvli5Lu98cYbNsQPLxKQR25ajRo1rFeI9X5Yw2uGB498NudJcoamK4/vPGsMIYB36d133/XAhusmhBJY41x407hmvIP8zzVwPBduWFCwxrW6QirPP//8VrB23HHH2Ty8l19+2btWrmP58uXWE1mvXj0rR3L7GJ/N3ya2w5NHG8hzo43AX5UqVWzYJZAH4AGDbOfkpdPsMtD199bf2+mAwlrqdoPCWoDsFNY8ASmsKax5yqAz4UtAYS0Y1jACKCIC+ABS5Jodf/zxNncL2MKzhOcHkAA4gCWKaFCMg/DA6dOnW4hgGzxDBx10kAUtwM7BGhAHUAEewBzA4rxJlOknVJDzA2ucH28dx546daq3HV4vcsA4P0MCUPCEfQBN2sDxOAeQyfL58+fbYyVTYMR51ty1MZ07d66t4Ii3kfHi/GGQnPeuu+6y5yNk1A0ZgMwoz891kEPHcQj/5H9gkvw/liEvYI/lM2fO3KrACMenaAnr8V76hzJgnX5VBqoD2aUDCmup2w0KawGyU1jzBKSwprDmKYPOhC8BhbX8wRoAAhB0797dQhkeHfLQKCZCbhthjUAYRTeo3ghsEBJIiB8hiMOGDRMKbOAhckVE8Mb5YQ0IYxvWUzgDaGE/zkNRD84PrOHJAlyoPFm7dm3r7WM7II/rcZ61hx9+2AJdo0aNLNyRs0aJe47PsSi1D9wkA2sUOkEG7sv/tJFjXnzxxRZcY2GNIQzIR8Nr2KlTJ1uohTw3Su3TTmAUg5ohDwBJZEjYKcc79NBD7X78j7xcGCRVLhmUm/3wGB588MG2TePHj7eQpwZ6dhno+nvr7+10QGEtdbtBYS1AdgprnoAU1hTWPGXQmfAloLCWY+RQ5h6QAA4ALffid9MlS5bYnC/yvb788ksLBHi/KJ4BTOH5atq0qQUnBxHsS8VCwvZc1ULACa8cHijWA1+U0sdDBoThncPL1rp1awtfQAkhhnjS2M55ttju2muvlWbNmlmvFtURzzzzTBk5cqTNmeN6CctkAGqKnnB9gNqgQYNsPh3XypeKkZyP0vvu2K7N/inHJvQz9st1cn1AJyDFPngMyT8DnNwxKMzCtVFwhWtBZsgOUHNeOrZFLsiHawU+2R4Q5TpZT7vxtDE4+YoVK7zjA4i77767zcdz27pz61QNedWB7NEBhbXU7QaFtQDZKax5AlJYU1jzlEFnwpeAwlqOEQMwEJrnQhpjjbt461kG2AFG5J8R4scyty/zfAEMimawDXla/nOw3v8/+7IMLxLFSfzbx9sOrxXVE4FNt95N3bGAGo7lvz6Akn3cNfr3cdfvn7rrZLvYrzuG295ty9Rdg5tyLcgBmblQR/9+bjuuFSimfZzPv02847tlQe1wx9Fp9hjv+ltn12+tsJa63aCwFiA7hTVPQAprCmueMuhM+BJQWMsuQ0YNV/29VQdUB4qyDiispW43KKwFyE5hzROQwprCmqcMOhO+BBTW1HAryoabtk31W3Ugu3RAYS11u0FhLUB2CmuegBTWFNY8ZdCZ8CWgsJZdhowarvp7qw6oDhRlHVBYS91uUFgLkJ3CmicghTWFNU8ZdCZ8CSisqeFWlA03bZvqt+pAdumAwlrqdoPCWoDsFNY8ASmsKax5yqAz4UtAYS27DBk1XPX3Vh1QHSjKOqCwlrrdoLAWIDuFNU9ACmsKa54y6Ez4ElBYU8OtKBtu2jbVb9WB7NIBhbXU7QaFtQDZKax5AlJYU1jzlEFnwpeAwlpuQ4YS8IwXFjvWGsspwc86pvxfWIxArtV9t/Wa3XGibL//GvgtYn+rbW2j7p/7nlB5qDwKkw4orKVuNyisBchOYc0TkMKawpqnDDoTvgQU1nIbYh9++KF07drVDmaNgQIYfPrpp3L55ZdLz5497cDYDJx99tlnyxtvvFEoQIGBpkePHi2vvvpqUoBJ2xnvjDHfHJwx9tn5558vL7zwQlLHKkhj75133pHzzjtP3nzzTbn44ovtQN+rV6+O7HoKsm16rNz3o8pD5ZGsDiispW43KKwFyE5hzROQwprCmqcMOhO+BBTWthhDeGn69u0rTZs2lW+//dYa/0899ZQ0aNBAtt9+e6lbt67stttu0rhxYylWrJhUrFhRrrnmmowHtueff95e/5VXXinr1q3LN9QwIPXhhx8uQ4YMsR5FjCaAb4cddpCxY8cmdaxkDa5E20+ePFm22247ufvuu+Wll16SsmXLyoMPPpjvdiU6tq7bcj+oLFQWhVEHFNZStxsU1gJkp7DmCUhhTWHNUwadCV8CCmtbDLKXX35ZSpQoIdddd50Fkffee0+qV68uxYsXl5EjR8rXX38tv/76q3z33Xdy//33S+XKlaVUqVIyc+bMjAYFvGAAVrKwtmLFCmnUqJHstdde4jxXn332WeSeNT+s4fE74IADpF27dtYLWBiNS73mLfegykJlsa06oLCWut2gsBYgO4U1T0AKawprnjLoTPgSUFjLMY6AsIEDB1oAA0iAgAEDBlgPDiF35EW5UEA3vf322+16tnMeK9YtW7bMenweeugheeaZZ2Tx4sXevhgiS5culQULFtjwwo8//tiGXE6fPl0IweQ6OAbXsHDhQlm1alUuECQskX2XLFlil7MtUDVr1izhfE888YR88cUX9jjO6PHD2vr1663XkGPgOXPbMOWY7777rj0n1/jaa69JnTp1pG3btvLWW29ZWOV63n//fdtGt69rM9A6ZcoU22bAluVum3htfvTRR+Wjjz7Kda3ss3z5cpkzZ45MnTpVkAthj8CiO54f1mjPfffdZ2H0scce887nzqtTNf5VB7JLBxTWUrcbFNYCZKew5glIYU1hzVMGnQlfAgprOYbMN998Y71oeGkAL8AFrxpf4CfW4HNQcc8998gjjzziwRzeObw8eOgIlWSKdwqQAvg49sSJE6V06dJyzjnnSNWqVa3njm3x1N16660WSk4++WQpWbKkzJgxwzs35+R8LJ80aZI91vz582Xfffe1yzgGXsBatWrJzTff7IUu+mENQ4bQTTyChEf620VoI9dFLhjeRbYh/BOvHOccMWKEBbjy5cvbHD7awjURitimTZtcbSZUFLkAn2yHV8+1uUqVKl6bmb/jjju86wAKO3bsaM/NOWlPuXLl5NRTT/W8e35Y49j8Psju2GOPzQV+/rbpfHYZ7Pp7Z+/vrbCWut2gsBYgO4U1T0AKawprnjLoTPgSUFjLMWrwCgE7F1xwgYULPDuA1j777JNvAMAbVrt2bQslgA1etUsvvVR22mknm1c1e/ZsD1zIuQJeBg8ebLe78cYbbQ5cvXr1rOcNDxmQFOu169OnjwBLH3zwgd2uVatWFmiOP/54Ib8OyALWgBy8UhitsbA2YcIE6xF89tlnPUhiuwsvvNAup3AKEAhUAlP169e3xwVE8eAhJ5ezRvESzkdbzjzzTNuWcePGSaVKlSxkzZ07V/B+jR8/3h6b7YYOHSqc+/rrr5cKFSrYnEC8kVTa7NatmwW1UaNGCecD+Jo0aWKP//bbb9vrjYU1gBA5AIix3kI12rPXaNffPjt/e4W11O0GhbUA2SmseQJSWFNY85RBZ8KXgMJajkFz2223WZggjA8jD48WsHTEEUcIL/8gww8P07nnnmuPMWbMGG97PD/ABYBDFUnnZQLW+vfvn6s4CeAGIBKKSGhjs2bNbBgiXj/O/9VXX1l4Ovjggy1A4l3jOBQA4fxsw/FffPFFCzcdOnSwAJQsrM2bN88eC/BxOWsUX/l/e2cCblVVxfFKs7DUyolS1DJCnINyKEcME8IsFWfFrHCgskHTEvTTcKak1KQCQnNAcUCeE6AgWDkkJQiKmsgoCDxyIkWB3fttWod9D+/e+97l3XvP8L/fd79z7xn2Xnvtdd9bv7PWXof2KTASwtrZZ5/tZaAqo+kIGYYOHerPo4Im3w3WSDUFruxcIJMIGimfrAXcd999PaDaeNiiT8bZ0NDgr4vDGjDIPBGBA2KtbW3z6axr3vM974K1yv0GwVoZ3QnWIgUJ1gRrkTHoQ/U1IFhb49gQAQMIiE7h7BEVA0q6d+/uI0PNOYCABGvIeAMz3bp185Ei1mGF57MGi6gPbyCMlED6ooIhIMO5bCmJz34iW3wH/khDJLrE9xEjRvjjrNHiGkAHuCMKaP0hExGqvfbay4MdpfYrhbXmCoyEsEZaJ485oComETaTgS3gRRVNomJAn8HazTffXDBmA1wieaZPHpUwbtw4d+2117q+ffu6rbfe2o/bIoFxWEM3pEnGdRHKo8/5duA1//mZf8Fa5X6DYK2M7gRrkYIEa4K1yBj0ofoaEKytcWJ4DlkYvSG6RcRn1113LVplEJg56KCDXI8ePXzBDdZttW/f3hfiCJ1DinKwDouUQgpvGKxZpIhzAQ5SMJGBNWPsY/0WMrAWi++U0WcNHQVL+H7ooYf6lEgKfvDd3kAU55JySCGRlsKaRbAsslYO1oDC3XbbzadBWvTPZODZbF26dHFbbrmlX/9nsEbUz85hzBRvAUiBNSJupEbusMMOPo0TSEP/vNFLKVjjuXdEQhmrta/tWpuQLqSLvNiAYK1yv0GwVkZ3grVIQYI1wVpkDPpQfQ0I1tY4cVdffbUHAioU4tQAKp06dfLQMHr06GYBgMIawBTRN0CDLevJ4vDEeiyLMs2fPz+CtbDARxzWiDIRrdtnn308DAGPgBrgxjFkPPXUU7188YddUzmxa9euvngJqZNxWLv88ssL4Ie2aJMoFlDUUlgDCi2aSCXL0Bmk+iPr70jlJPLYElhjfRtFTTp37uxTOQFbxkKxlHKwRiooczFp0qQCOUKZ9FnAIhvIvg0I1ir3GwRrZXQnWIsUJFgTrEXGoA/V10DtYG26cxN6OTfx8Nq9J/R0b84vjDoVc9ao1ggQkHpn5xhgULoeWLL9bAGJAw44wF8zZMgQf4x1W7RBoQ1b4wXE0SZRnyOPPNJH0CyyVgrW6AOAozokkafjjjvOtxGCI/3SH2vfrMQ/0EUJe8CF4ijIEcIa67vskQODBw+OUhJnzpzpHwZOe3FY23vvvaNKjGEaZAiYRLbCMdM2Y+7du7fvw3RZLLJG8RBbN4gekRMdEKEjSohcNvbm0iAPO+wwn45JCmU4T/qcfedcc6w5Dm1AsFa53yBYK6M7wVqkIMGaYC0yBn2ovgZqBWuvv/qSWzWhT+1ArQkKV004yb0+b0aLnHeiYRtvvLGPVhkosNaM6oSAApExqhwSeaOaIul/7KdoCECBswDwUJCDtVOsJ2NNGuX5qXi42WabRWvRWgprtDljxgyfPklfrP8i4meOCfIRQQOKABogBtDhcQCkQNr6uzisAWNEsIh8AUiUzgfIqCBJPwZrAOBOO+3kKzsyDtIQ49UgKZuPboDDPn36+DFTdIQIIxUhSeVEn+VgjTRIYIx1gh07dvQFSihSgn4ZH3L169fPP4MtDmvoH92QLmnQajrSVo68bCBfNiBYq9xvEKyV0Z1gLVKQYE2wFhmDPlRfA7WCtWWNTc8YawK2NxZOb3o/W4P3dN8f/bbEWSPdjuqJlIDns10za9Ysnx7I2itghigXMMZzvSjPH1+rxQOcKbrRrl07Dx5AEWBHFI2oF9EonnNG5ULSKK0ftpTDB3IAF9tPZI7y/bRHMQ4DSTtOYQ8qLtIeoAM0AS5ECrmW83gsAUVA6Jf+SV9kfRwwxZgYD+vLzjjjDA+WABbXcS6FVxg78PfDH/7QP2cN8Bw4cKA/znnIS5QxHDPr94BEGzOpl8hI4RCTnfYvuOAC3yd6A7qoiAncIhe6Yz7QCwBqaY4UKaGt4cOHexmoJAloU2SE/qx9bfPlpGu+Nd/YgGCtcr9BsFZGd4K1SEGCNcFaZAz6UH0N1ArW0uBIkboHuBA9CuUFAGzt16hRo3z1RYp8GBjEt0R3AAhgDPBhnRrt2Xms4eKh2xToCPux/ZZOaNdQTZG0y3jUyNqjHUrW81wyomJUYrRr2XKc/mjf+uNaonbISEETZASWOM/65xyAj/Vn7OccjoVtmQzIRjET2gPe4mNmDFwXH7PtByCRje3UqVM91LEOjuP0QYVNdMn19EVbbA0oATmeY2fj01aOu2wgnzYgWKvcbxCsldGdYC1SkGBNsBYZgz5UXwOCtbUODVEyKhGSwmgQgsMXfg4dwGL7w3Nq8bkSOSq5ph5jQc5SspIKSkGSQw45ZB0QrIW86mPt70e6kC6SYAOCtcr9BsFaGd0J1iIFCdYEa5Ex6EP1NSBYW+tgEaWh8iBpfpYKmATnQzKsnaO4Lmy+Hn30UUXVmm4sxPWj79JJ3mxAsFa53yBYK6M7wVqkIMGaYC0yBn2ovgYEa4XOHKmCAED4oOm8OTtpGS8pmiNHjvQPDUfmUhG4tIxJchb+HqUP6aO1NiBYq9xvEKyV0Z1gLVKQYE2wFhmDPlRfA4K14s6QnP/iummtA1Xt8zVX6ZmratuC2s+3LQjWKvcbBGtldCdYixQkWBOsRcagD9XXgGAt346NHFvNv2xANpAlGxCsVe43CNbK6E6wFilIsCZYi4xBH6qvAcGaHLUsOWoai+xZNpBvGxCsVe43CNbK6E6wFilIsCZYi4xBH6qvAcFavh0bObaaf9mAbCBLNiBYq9xvEKyV0Z1gLVKQYE2wFhmDPlRfA4I1OWpZctQ0FtmzbCDfNiBYq9xvEKyV0Z1gLVKQYE2wFhmDPlRfA4K1fDs2cmw1/7IB2UCWbECwVrnfMP/Nme7hucMz/x4/Z5h7adlTrVeUYC3SmWBNsBYZgz5UXwOCNTlqWXLUNBbZs2wg3zYgWKvcb1i0fFYTqA3LAawNdbPfmNZqRa187wW3anl/5+J+esa+r1x+iVu16s2S+mkxrC1uvLVJYQP+/74wg8ob4FY3jW/JsodKKkwHpYH10YBgLd+OjRxbzb9sQDaQJRsQrFXuEby94j9uwpzhbnyGo2t+bLOHusblC1qtqNfebHQLF1/jVi/PInOsGdPKprG9tOBG985775XUT4th7cXFje6tt8c5t6Ihs+9lb0xwrzS+VVJhOigNrI8GBGty1LLkqGkssmfZQL5tQLBWuUfw7vur3fSFM92zi8e7pxY1uCcX3pupN2Oaunicm75omnvr3ZWtVtT7q5xbvmK5W71qqXMZfa9a2cRW765wq1eXVk+LYC1spEx7pXtL8NFwXOF4EyyyREuhBgRr+XZs5Nhq/mUDsoEs2YBgbX0ckTWe5+omp3PlqpVN7/cz9l7ZBCFNxOVfoZe9PjrL57UtgrV8qkajlgbaXgO1grUlja+5KQvGuScWjKnZ++kFY93ixoUuS46IxiLHWjYgG5ANFLcBwVrb+wlqURqIa0CwFteIvksDVdRArWDt1aWzHRWYallpavycoW7ekpcFa/8p7tjI6ZNuZAOygSzZgGCtig6DmpYG/q8BwZpMQRqooQZqB2tz6gJr8wVrglXBqmxANpAbGxCs1dCBUFe51YBgLbdTr4HXQwOCNd1Vz9JddY1F9iwbyLcNCNbq4Umoz7xpQLCWtxnXeOuqAcFavh0bObaaf9mAbCBLNiBYq6tLoc5zogHBWk4mWsNMhgYEa3LUsuSoaSyyZ9lAvm1AsJYM30JSZFsDgrVsz69GlzANCNby7djIsdX8ywZkA1myAcFawpwMiZNJDQjWMjmtGlRSNZB3WGtsbHQLFiwo+3711VfdsmXLcrNIv62dN/Q8d+5c97e//c2NHTvWPfnkk27OnDmp0emSJUu8jbBdX90sXbq0zdqqVJbFixd7GZClWBuvvfZa2XOKXav9AsB62YBgLanehuTKkgYEa1maTY0l8RrIO6w98cQT7tOf/rTbaqutSr67dOniFi7UM9ta64ABuLNnz3YDBgxwn/3sZ93GG2/sPvzhD7uPfexjbvvtt3cXXHCBe+WVV4oCQ2v7q8b5r7/+urv55pu9fYwcOdLxvaX9TJ8+3TU0NBSM8c4773Rbb721GzFiRKvaammf5c5D/t/+9reuffv2bty4cUXH8r3vfc/LOXHixKLnlOtLxwVttbYBwVri3Q4JmAENCNYyMIkaQno0kHdYmzZtmjvhhBPc8ccf79+HH36422CDDdwWW2zhjj322Gj/j3/8Y9cWUZVaOy717m/evHkOnX7gAx9wn/vc59wvf/lL98c//tH94he/cDvuuKPff9RRRyUahIGbP/3pT17W4cOHtxiwuO7yyy93H/nIR9wDDzwQAc9tt93m7as1bbXlPCLXpZde6j74wQ96kCzW9tFHH+3H/PDDD0eyFztX+wVlSbEBwVp6/A9Jml4NCNbSO3eSPIUayDOsWVojzqu9n332Wbfpppu6ffbZx5EiZvvZcn74JrWPd3wf19l+c2DCc/jM8ebO4/xSx+Pt8L1YO2Hfpfor12fYjvVHn3y2Y81t0dnAgQO9w/+lL33JPfPMMxHocOz55593xh7qpAAAGMBJREFUe+65pz9+6623FrRF26Vk5nj4Rp5QpvAYn0P5wmO2n33F+kPWOKw11wZtxfcDRR/60IfcfffdF+lr/vz5burUqY6t9W/XFpPBjoftx8dcrC3TC9dyDuNpa1ij7eZkD+W1/k3O8Fi4z9qx4+Ex28eW83jz2c7RVtAoWEuhIyKRU6cBwVrqpkwCp1kDeYa15hw7nGiDNdb1hOcAcj/5yU/cpEmT3JgxY9yJJ57oI0Q4jFOmTPFRoyOOOMJ97Wtf88f+8pe/RNE4IkxElW688UZHGhyRvEMOOcSddtppjsiFOZz0SardKaec4ts58sgjvWP9wgsveFmI7l199dUegkhhI+J32GGHOaJTQ4YMWSdCRermsGHDfJSQ/ogW/vnPf3aLFi2KxkafwNLJJ58c9UlE6MUXX4zOQb4ZM2a4888/3/fXq1cvd9FFF7l///vfkeyhrvjMmrQddtjBpzyis/hxvjNW0vFIuQMi2Mf6QODomGOO8To67rjjvN5MZs6744473DnnnOOefvrpSCYieFdeeaVD14zpiiuucD//+c8LUhBp/5FHHvF6Y8t31mYxfuYTHRFRuv766yOYisPaG2+84eX+2c9+VqAjziNahlwzZ870KZ4HHHCAj2D17t3b9e/f37f5+OOP+/4fe+yxSCeMmYgj5yEDY77pppuieaJtdHXuuef6MZ933nnu61//uvvmN7/p7SEEP+aca7Gh7t27u2984xteR0SRGS9ttSWsMVbsBRs0/f3ud7/z692ee+45/5vhO78T+rc3c8jv6V//+pe3IdY0/uY3v3H8hg499FD3gx/8wK9ttPPp56c//alP3XzwwQf9+LgeGLVztF2r37zqQrCWZo9EsqdFA4K1tMyU5MyEBgRrhc5NKVgbP368d7y//e1vu0022cSnSwI///jHP/z6q49+9KNul112cV27dvXAx9osS3XDaf3EJz7htttuO3/s85//vNt1113dRhtt5D71qU95pxQn+rLLLvP7WEP35S9/2a/zIl2tW7du3tEHWFg/R1+0x9on2iFtk/TNU089NXLwAQBSPNlPe7vttps/b8MNN/SOLk49fV5yySV+HRlt0SeARZ84+rSB0wdYdOzY0ffLGElhJGJ08MEHu5dffrlZZxkYIv0RBz4OvuZI4mgDdRR5AQjZotO4zOjSxoYzdvrpp/u2v/CFL/gx7bHHHu4zn/mM38cx2v/Od77jvwMu1h/AgE7QwaOPPurlOvPMM73OP/nJT3odAY/0DzQAEOgojKy99dZbjms4B/CytpELwCTtkbWQyMQcoYMtt9zSRxGBW4CdfTfccINvG9AC0miPOWCeNt98cz8n3/3ud/180jZAy7wwD8w37bPekrbOOussrz/GB8AwPvTBfGJrnLPvvvt6/bYlrM2aNcvttddevj/mgv6wNeRkPSJzy1pFxg9sma6wB34n7CfCSjvYG3Ij78477+z1iC0a6P/1r3/1OuJGAXOFvgB0pScX/g0zHed1K1jLhGuiQSRcA4K1hE+QxMuWBgRrhY5OOVgDUHBEDzroIPfQQw+5l156yRHlYN8111zjnX8c5ttvv907kzjvOA/AGg4m11944YUevAAVolM40hTgwOkExABBHFOcLYpzED0BVnBagTWcXK4hakO0hHYARvYjB33jkBNp4ztOOtEL5CISRUoi+4mwEVXafffdPVQYeOA49+zZ0wMMcnAO0UKKglBoA0eb6BVRHpOd/kLnkO9Dhw71x/v27VtwLDwv/pmIFm3ut99+Pm0SmZ966imvF3RHtBJ9nnHGGf48QJXKkoAeaZbA8Lbbbuv1RvSFawAh2jF9cvyLX/yiHxdRTnQLHPz973/35xFBBRyQgygmkbQ4rAFHXAeU2RiQC9AEpCkswtwRTUMGIncUUkFO4JG2gTXaJurE9wMPPNCnRyIrYyJFlGtvueUWP+bvf//7/jzmC53Q1j//+U8/XsZNf0RgDQyJhCIbwETb3Bjg/LaENcATW+rXr18E9uhxs80283NG/2Yn2JvpCltDT4Az4+U3QTsAHvCKjXF+u3bt/I0KvmOLwBy6+spXvuLuv/9+D3rowdrVtvDvWR71IVjLlo+i0SRTA4K1ZM6LpMqoBgRrhc5NS2CNSIFFk3B8SU8jrQyAAaZwmoEaHEucfiIxBmvABdEjc6KAKCIERI2ANSITfKcAB441jivwcPfdd3tn32CN6E2YRkd7FLGgT9LeADgcdBxiIkgcN6eW9Ekcd2SjfcCF6wAL5GEfEEifOPoACf2R5sZ3oIM3QELRkE6dOhWMib7Qy+DBg71jTeoa323MxbY45F/96ld9xUgbm8kMeAFHQCTXA2s496R4WnuAz7e+9S0PlURriByib6JVltJ51113+etIl6RtYBogYqy0Y/0BQ6TDAkzI1VpYA5QYM5FS2gcsTM4Q1phz1kcCwjzWIJSBa5gXoke0BawxZgDJ2mLMRJo+/vGPe1BjXn71q1+5e++919sAc8k8cXMByGEuaaut0iAnTJjgbzigb6CeSCT7uDGBbSAnYMb4sDf0i8xAGXoZPXq0hzyiaUTm+J0wBt78jhg7wIbcBmvbbLNNQZTOdKFt4d+yvOpDsJZRZ0XDSpQGBGuJmg4Jk3UNCNYKHZxysIazTGqaOfVscbgpIAFAkBZIiiAOJs5xHNZwrC3Kw7U40kDISSed5KGOSA9OLv3ggAML9Me6Nq4zWOMc4DCUg4gYYMI1pNvhAHMeDnR4Ho5whw4d3E477eQhi0icpT4S1QPeAAOcbvokKoRjDbxwnb2JUAETpPoxjtA5BAgs3Y/1edZ/eA6fGc8999zjSDHFOQeEkbu5seGkk4IJFKBr+jaooy2cNKJo6B6nHxlI8WQeKJPPOawRZBxE4YAwUviIRBGRMhnZkv5JBIs0Q/RaLVhjbniEAcALWIUyMIf0jxzIypxgK0SuTI+MmXWNPBLB0gzZAqOk6xKpJaWSGwBtDWvIin1MnjzZsX6PdWbYFHZLXwZr/D5YX4dtoXcgmhRO5pIxczOCSBw3BLApsy+2jAvZqaAJrPGZ6KX9hkwP2hb+HcuzPgRrWfdaNL4kaECwloRZkAy50YBgrdDJKQdrOKGkbJlTjVNElAInGpigMAQVEAEVnNY4rOFYh9eGsIaTQXtEKUghpNCFlbcnMkFUCFAh3dFgLXTKDNZY80R6JudwPUAQngcUkTZngMAx4IYCF0TPuI5xIj8RGmANeCR9knVgRAHDN2vE6C/sg89ABZBHiiHwFT/Od4uWALlAJOBChCV+PmPDkactAA9YQ+ekC1q7cVhjP+mh6I4ImoER4EDk0XTJGisr4GJtAWsABevXiKK2BNaIGKE/opktjawhE1AC5AAu1j9b+mXdGdFBZAXWiIgyJjsvDmvYTufOnT3IkipIARpsCcBpa1hDBiAKCAPYSdclnZd9wL/BGuexdpP+gUgAG9Cm4A5AbbDG/Pbp06fAtrAzbI71jwZrjCn8DZkutC38W5ZXfQjWcuO+aKB11ECiYI0fPRWr+MerlzSQRQ0I1godnJbAGjBmziLONtEgIiCkEOKw44DSDhGe1sAa0ZPf//73vlokbdAW0IKDC/RQ8IIoBbBG25Y2Z04Z0TeceWCE8/bff38fmcDJtXPYkhZJFIN1aMjPOjHSAK1P9gGg9Al8kBLI+VQX5G+ivWmLNWTAALKHfaAfi07RDiAYj4bwnWIYOPGkYAJQrK8DEm39nLWJs87YqHzJvpbCGm0yToCMNVBEZpCXNugfIGYfkVHriy1zCYAAbIwthDXG/6Mf/cgDB1Eluw6IJI21NbDGPBHdA3hC8KRNoo0AKWmtfC8HawDnr3/9a69P1okxPrNHK7bSlmmQyI59YxukqdIXb+Yd6A5hDRAH4Ej5PPvss/0cU0GVcZEWTDSVmwxE3cy+SB9mXSgP8AZcDdao8Gk617bw75f0sSbCnsX/1VkdE5kA3ODhb4dehRpgOQLLLN5+++3CAwn4lihY484wd5S5K6iXNJBFDQjWCp2d1sIaTiTRFxxR/rDi2BOlogw+EMK6MZxm9rGOp1RkDacV8CNCR/SBtgAAonQABbBksEbbOMrIi4PLloqR7AdG+MdH4Qr+flGNkfQz2mIdHIVJAKg//OEPfg0aUTbewJ/1yVowziEFk32AH0CBLAAkbZGyCTT26NEjgte4s0j6IcBBmtt1113nUx1pjxS/QYMG+agXfRNdARYp3c4YSKljTPRDURTWshlo4cy3FNaARkCLaw2qwyggUUPkA4CBJfoj2gkYIwdgjlwhrKFbCo9wnEgScALgIjtthbDG4wNsTogOIk+4Zi1sCxDFhpCB6BngitxAJmMuB2usy2MtGP1RuIZoHNE6QJxUT/YD9IynNWvWcKS4YYlewjdRvL333tuDFzDNvPJ7QCfYDsCGbrAJ+iRVkvFQ/TRuMzzugGuIttEGsnOTgMgwETd0J1gr/FsV/63p+xr98FvRKz0a4MYX/x/4v6JXoQZ4nAxLG/i/n7RXomCNBfL8c+EOoF7SQBY1IFgrdIBaC2tEbkj/whEGBogckNbGuihAhhS8iy++uEWwhpNBcQicVtYZ0RaRGiJNRHmIMljqHkDAeh5AkfRE0uUAMwpsmIMM/AGHtEfkj7Y4n++kBQJ5wAOOPfsoB299Ijdwac49fdMGURQiIPTJOUSsJk6cWDTSgQNP1Ax5SX0j3Q8IwQmnT67nzqE5msAFoMRYrPQ8/6w4l7VvyAzgtBTWaBeIoQ3miH9+jJk3x9An69j4O8/40RFwQP9EKCmoEoc1vgO9nE+0j3VhpCoy31xrsEb7gA79Mn/oDEgNYY22cFIopBGOGb0wZuQF3loCa9yhHjt2rO8f22AsRAb5THvIYSmyrYE17IDr428ifkAW7aJf5pUIGXpgvIArazFN10T1kIXzKcBjc86WqCARRsbMowmAQPpDv0TWaEOwVvi3KtSfPq/VjWAtPZ7KihUrfPEj/i6988476RG8BpL+97//9evHuVH53nvv1aDH1nWRKFgjRQaHgX8UekkDWdSAYG3tP3kcHu7qkzJGhAIQC50gUuMABiIV5oByHAecdD6iaES3KOtOZOyqq67yUSycTdLAiFKRphZeS8SAKBypiDju3EGjVDuQRfl6qvhxnQERjjtRIBxa1pOxFom1SaT6EcUBZkKZgRFuOhG1wZlme+2113pI4TxkoU/SGSlIYX3yTC97vpXJy00rZCHKhlycgwMe9tfcZyKLVKqkTDsAQcEM9MSzytCpyWHXMkaiVMASMhOFQT+AH+egJ9IqARkrqsF+II6xsu6JdXnWLv3z0GZ0RQTI+rEt80zkDPigPyKW6NKgl/OooMn1pCZau+wjTdT0SkQTueif+UZvzAdprIwdfQF/FG6hLVIHTQbGjG2EY6a0fzhmHsUA/FhlS65lzOiKtV20TZ8UjOEGAv/kkY8IK1FD9gGF2Jw9mL3U/CEPchZ7A/mAJjcYiN4SsUWOhoYGv86RdYhUArUxYovYLRE3048dY8ucUQUV20B2isVQLZL5Y1xE9Zhzfh/hdfpc+Dcs7/oQrKXHU+G3zU0Zftd6FWqAG5fc9MWfSOIrMbBGjiiL2fkn9P7770e6Wr16tb+ryj9a/mnqJQ2kWQOCtXUdHRxg3s05PTgCwEL8mIEW4GPXss+cbc4v1q61iUPK29rnWtqz7xwzWKMIh8EETjDAYec1Jxv/FDmPbfy8cn3G22tOrvg54Xfat+/Ij9zxcdlx2yJjMZlNXvTG+dZ+fL+1xRbdx8+343ad9RfXpR2Pz5NdH+oVua0fO277QrsI2wrPo++wvWJjC/fHx0Z/HKcd2uY7W8bH2/aZnNaWycGWfdYu5zX3DseD3NiF9cWWOQ7btEdGsGatuT5NLq4z2U0Wa6c5vdkxbdf9W5ZHnWAjeqVDA9wsIquBm2Xhi6gSNyjJ6ODvTB5f/L0kC4G6GeGLtbxkvHC8nmvZEgNr3KUkZYM0D3sBbaxfI8WDu6mkbuglDaRZA4K1dDk4IayFz2vLo1OmMafDdlkLyRo/UjJZ91gqmqc5TcecJnmeBGvp8UiANNLjWaNrL270kIVAyj3r2fBR8vgiG4SUeta724sbWWT3sJ8shHqmjiYG1nggKVTLM4B4QfqsPUFJpEeiNF5E2vSSBtKqAcFaupwjIg6kPLL+ySJrSXacJFu67Ksa80UhG+6eA2r8DyWqViyyVo3+1Wa+bFCwlg5vBN+Zx76EBTQIklBcinXapHcnca1WLbS7cuVKnwZO4TKLLFIYi2UKBItIt+ecer4SA2s8S4nIGutRuJvN85OsWADgppc0kAUNCNbS5cjg5LK255WmNVGW0iZnNF1zmLf5Yu0Fazip+Cmbla1W2/4Fa+nwTACxPffc06+BXrVqlX8uJ0uPABTWOOc5EIJfRrEm1lHzYq01j0Jh+QPpoeim3vpJBKyR7tirVy//0FiqflEhjYpY5I7Wm2bT8TOUlGnRgGBNzlO1nSe1LxszG1BETbZgtlCtrWAtHd4HN3GIEvFMSHxrqutSEZaiUbzqDSP11CJLrKiqS9VeHn+DnihgRnGopOgmEbDGAj7oHpKFbkmHpILZu+++m2sDqqfxqu/qaECwJuepWk6T2pVtyQZkA7W2AcFadXyFtm518uTJ/nElVBYGTPCzL2qqMJtnSDMdjxkzxuuGJQ+s3UM3VJVO0isRsEaJYFIgee4NJbR5mCfPfxk1alSSdCVZpIH11oBgTc5UrZ0p9Sebkw3IBqplA4K19XYLatIAjwYBQlhedNlll/kMtm222cavxa6JAAnuZMCAAV43PCx80KBB/vmYBI5II0/KKxGwxnNqADUgjdKYLOTHiAjR4tzqJQ1kRQOCNTlN1XKa1K5sSzYgG6i1DQjW0uGdHHHEEf45YhTxY3nRXXfd5f1uUv/y/uIZlaSFUp6f9XzDhg3zuuEZqUl5JQLWzjvvPAfRzp07NwrJ8pBcoms8pFMvaSArGhCsyZmqtTOl/mRzsgHZQLVsQLCWfO8Ev4NiGT179oyeY0xgpGvXrq5Dhw7+OYvJH0V1JOR3se222/oHhQNqvNjXqVMn/+ZzEl51hzXyZffff3//TJiwbCiLIdu3b++PqRpkEkxFMrSFBmoFa0sbl7jnFz3ppi2cXLP3c4uecEsa1zwYuFqOgdqV0ykbkA3IBpJjA4K1tvAMqtsGVdZZatS/f/+Cjm655RYfQSItMq+vadOmuY022sgNHjy4QAXXX3+9180NN9xQsL9eX+oOazw/jec+9O3bN4qqmTJY/Eh0raGhwXZpKw2kWgO1gjU5M8lxZjQXmgvZgGwgqzYgWEu+S0KGGuvVeJ5x+CK61qVLF7fddtv5R2aFx/Ly+aabbvJQxuNOwhe/V6JrnTt39s+pDI/V43PdYY3Uxx49eri77767YPxE3ObNm+fDtgMHDiw4pi/SQFo1IFiT05ZVp03jkm3LBvJnA4K15HsjI0eO9A+/XrBgwTrCsnaNNVuTJk1a51gedgwdOtQ/X23p0qXrDBeQ69atm5syZco6x2q9o+6wFg64VAnRUsfCNvRZGkiyBgRr+XNm5MBqzmUDsoGs2oBgLckeR+nnp8X96vj3ZI9s/aUrNd5Sx9a/59a3kChYa734ukIaSJcGBGty2rLqtGlcsm3ZQP5sQLCWLh9E0qZTA4K1dM6bpE6pBgRr+XNm5MBqzmUDsoGs2oBgLaXOiMROlQYEa6maLgmbdg0I1uS0ZdVp07hk27KB/NmAYC3tXonkT4MGBGtpmCXJmBkNCNby58zIgdWcywZkA1m1AcFaZtwTDSTBGhCsJXhyJFr2NCBYk9OWVadN45JtywbyZwOCtez5KRpR8jQgWEvenEiiDGtAsJY/Z0YOrOZcNiAbyKoNCNYy7LBoaInRgGAtMVMhQfKgAcGanLasOm0al2xbNpA/GxCs5cFz0RjrrQHBWr1nQP3nSgOCtfw5M3JgNeeyAdlAVm1AsJYrF0aDrZMGBGt1Ury6zacGBGty2rLqtGlcsm3ZQP5sQLCWT19Go66tBgRrtdW3esu5BgRr+XNm5MBqzmUDsoGs2oBgLedOjYZfEw0I1mqiZnUiDazRgGBNTltWnTaNS7YtG8ifDQjW5N1IA9XXgGCt+jpWD9JApAHBWv6cGTmwmnPZgGwgqzYgWIv+veuDNFA1DQjWqqZaNSwNrKsBwZqctqw6bRqXbFs2kD8bEKyt+39ee6SBttaAYK2tNar2pIESGhCs5c+ZkQOrOZcNyAayagOCtRL/8HVIGmgjDQjW2kiRakYaaIkGBGty2rLqtGlcsm3ZQP5sQLDWkv/8OkcaWD8NCNbWT3+6WhpolQYEa/lzZuTAas5lA7KBrNqAYK1VLoBOlgYq0oBgrSK16SJpoDINCNbktGXVadO4ZNuygfzZgGCtMl9AV0kDrdHA/wDk7ajZgDm+sQAAAABJRU5ErkJggg=="}}},{"cell_type":"code","source":"conv_initializer = tf.random_normal_initializer(mean=0.0, stddev=0.02)\ngamma_initializer = tf.keras.initializers.RandomNormal(mean=0.0, stddev=0.02)\n\n# Encoder block: Apply convolutional filters while also reducing data resolution and increasing features.\ndef encoder_block(input_layer, filters, size, stride, apply_instancenorm=True, activation = layers.ReLU()):\n\n    block = layers.Conv2D(filters, kernel_size=size, strides=stride, padding=\"same\", use_bias=False, kernel_initializer=conv_initializer)(input_layer)\n\n    if apply_instancenorm:\n        block = tfa.layers.InstanceNormalization(gamma_initializer=gamma_initializer)(block)\n\n    block = activation(block)\n    return block","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:31:08.743436Z","iopub.execute_input":"2021-07-22T10:31:08.743785Z","iopub.status.idle":"2021-07-22T10:31:08.756511Z","shell.execute_reply.started":"2021-07-22T10:31:08.743754Z","shell.execute_reply":"2021-07-22T10:31:08.755479Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Transformer block: Apply convolutional filters to find relevant data patterns and keeps features constant.\ndef transform_block(input_layer, filters, size, stride):\n\n    block = layers.Conv2D(filters, kernel_size=size, strides=stride, padding=\"same\", use_bias=False, kernel_initializer=conv_initializer)(input_layer)\n    block = layers.ReLU()(block)\n    block = layers.Conv2D(filters, kernel_size=size, strides=stride, padding=\"same\", use_bias=False, kernel_initializer=conv_initializer)(input_layer)\n    block = layers.Add()([block, input_layer])   \n\n    return block","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:31:10.407612Z","iopub.execute_input":"2021-07-22T10:31:10.407936Z","iopub.status.idle":"2021-07-22T10:31:10.413581Z","shell.execute_reply.started":"2021-07-22T10:31:10.407908Z","shell.execute_reply":"2021-07-22T10:31:10.412475Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Decoder block: Apply convolutional filters while also increasing data resolution and decreasing features.\ndef decoder_block(input_layer, filters, size, stride, apply_instancenorm=True, activation = layers.ReLU()):\n\n    block = layers.Conv2DTranspose(filters, kernel_size=size, strides=stride, padding=\"same\", use_bias=False, kernel_initializer=conv_initializer)(input_layer)\n\n    if apply_instancenorm:\n        block = tfa.layers.InstanceNormalization(gamma_initializer=gamma_initializer)(block)\n\n    block = activation(block)\n    return block","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:31:12.409729Z","iopub.execute_input":"2021-07-22T10:31:12.410085Z","iopub.status.idle":"2021-07-22T10:31:12.418299Z","shell.execute_reply.started":"2021-07-22T10:31:12.410055Z","shell.execute_reply":"2021-07-22T10:31:12.417173Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def Generator():\n\n    inputs = layers.Input(shape=[HEIGHT, WIDTH, CHANNELS], name='input_image')\n\n    # Encoders\n    enc_1 = encoder_block(inputs, 64,  7, 1, apply_instancenorm=False, activation=layers.ReLU())\n    enc_2 = encoder_block(enc_1, 128, 3, 2, apply_instancenorm=True, activation=layers.ReLU())\n    enc_3 = encoder_block(enc_2, 256, 3, 2, apply_instancenorm=True, activation=layers.ReLU())\n\n    temp = enc_3\n    # Transforms\n    for i in range(TRANSFORMER_BLOCKS):\n        temp = transform_block(temp, 256, 3, 1)\n    \n    # Decoders\n    dec_1 = decoder_block(temp, 128, 3, 2, apply_instancenorm=True, activation=layers.ReLU())\n    dec_2 = decoder_block(dec_1, 64, 3, 2, apply_instancenorm=True, activation=layers.ReLU())\n    \n    outputs = layers.Conv2DTranspose(CHANNELS, 7, strides=1, padding='same', activation='tanh')(dec_2)\n\n    return tf.keras.Model(inputs=inputs, outputs=outputs)","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:31:14.462977Z","iopub.execute_input":"2021-07-22T10:31:14.463313Z","iopub.status.idle":"2021-07-22T10:31:14.473096Z","shell.execute_reply.started":"2021-07-22T10:31:14.463284Z","shell.execute_reply":"2021-07-22T10:31:14.472073Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### **Build the discriminator**","metadata":{}},{"cell_type":"markdown","source":"![image.png](attachment:e279261b-5a92-4b89-9134-f528682c5ff9.png)\nAn example architecture for a PatchGAN discriminator. PatchGAN is a fully convolutional network, that takes in an image, and produces a matrix of probabilities, each referring to the probability of the corresponding “patch” of the image being “real” (as opposed to generated). The representation size that each layer outputs is listed below it, in terms of the input image size, k. On each layer is listed the number of filters, the size of those filters, and the stride.","metadata":{},"attachments":{"e279261b-5a92-4b89-9134-f528682c5ff9.png":{"image/png":"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"}}},{"cell_type":"code","source":"def Discriminator():\n\n    inputs = layers.Input(shape=[HEIGHT, WIDTH, CHANNELS], name='input_image')\n\n    temp = encoder_block(inputs, 64, 4, 2, apply_instancenorm=False, activation=layers.LeakyReLU(0.2))  \n    temp = encoder_block(temp, 128, 4, 2, apply_instancenorm=True, activation=layers.LeakyReLU(0.2))\n    temp = encoder_block(temp, 256, 4, 2, apply_instancenorm=True, activation=layers.LeakyReLU(0.2))\n    temp = encoder_block(temp, 512, 4, 1, apply_instancenorm=True, activation=layers.LeakyReLU(0.2))\n\n    outputs = layers.Conv2D(1, kernel_size=4, strides=1, padding=\"valid\", kernel_initializer=conv_initializer)(temp)\n\n    return tf.keras.Model(inputs=inputs, outputs=outputs)","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:31:19.512614Z","iopub.execute_input":"2021-07-22T10:31:19.512930Z","iopub.status.idle":"2021-07-22T10:31:19.519367Z","shell.execute_reply.started":"2021-07-22T10:31:19.512902Z","shell.execute_reply":"2021-07-22T10:31:19.518466Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"with strategy.scope():\n    monet_generator = Generator() # transforms photos to Monet-esque paintings\n    photo_generator = Generator() # transforms Monet paintings to be more like photos\n\n    monet_discriminator = Discriminator() # differentiates real Monet paintings and generated Monet paintings\n    photo_discriminator = Discriminator() # differentiates real photos and generated photos","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:31:21.566145Z","iopub.execute_input":"2021-07-22T10:31:21.566493Z","iopub.status.idle":"2021-07-22T10:31:22.873152Z","shell.execute_reply.started":"2021-07-22T10:31:21.566462Z","shell.execute_reply":"2021-07-22T10:31:22.872150Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"to_monet = monet_generator(example_photo)\n\nplt.subplot(1, 2, 1)\nplt.title(\"Original Photo\")\nplt.imshow(example_photo[0] * 0.5 + 0.5)\n\nplt.subplot(1, 2, 2)\nplt.title(\"Monet-esque Photo\")\nplt.imshow(to_monet[0] * 0.5 + 0.5)\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:31:23.144582Z","iopub.execute_input":"2021-07-22T10:31:23.144972Z","iopub.status.idle":"2021-07-22T10:31:29.558937Z","shell.execute_reply.started":"2021-07-22T10:31:23.144941Z","shell.execute_reply":"2021-07-22T10:31:29.558057Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"monet_discriminator.summary()","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:31:29.560658Z","iopub.execute_input":"2021-07-22T10:31:29.561035Z","iopub.status.idle":"2021-07-22T10:31:29.574060Z","shell.execute_reply.started":"2021-07-22T10:31:29.560994Z","shell.execute_reply":"2021-07-22T10:31:29.573144Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"monet_generator.summary()","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:31:29.575854Z","iopub.execute_input":"2021-07-22T10:31:29.576256Z","iopub.status.idle":"2021-07-22T10:31:29.598106Z","shell.execute_reply.started":"2021-07-22T10:31:29.576218Z","shell.execute_reply":"2021-07-22T10:31:29.597325Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### **Build the CycleGAN model**","metadata":{}},{"cell_type":"code","source":"class CycleGan(keras.Model):\n    def __init__(\n        self,\n        monet_generator,\n        photo_generator,\n        monet_discriminator,\n        photo_discriminator,\n        lambda_cycle=10,\n    ):\n        super(CycleGan, self).__init__()\n        self.m_gen = monet_generator\n        self.p_gen = photo_generator\n        self.m_disc = monet_discriminator\n        self.p_disc = photo_discriminator\n        self.lambda_cycle = lambda_cycle\n        \n    def compile(\n        self,\n        m_gen_optimizer,\n        p_gen_optimizer,\n        m_disc_optimizer,\n        p_disc_optimizer,\n        gen_loss_fn,\n        disc_loss_fn,\n        cycle_loss_fn,\n        identity_loss_fn\n    ):\n        super(CycleGan, self).compile()\n        self.m_gen_optimizer = m_gen_optimizer\n        self.p_gen_optimizer = p_gen_optimizer\n        self.m_disc_optimizer = m_disc_optimizer\n        self.p_disc_optimizer = p_disc_optimizer\n        self.gen_loss_fn = gen_loss_fn\n        self.disc_loss_fn = disc_loss_fn\n        self.cycle_loss_fn = cycle_loss_fn\n        self.identity_loss_fn = identity_loss_fn\n        \n    def train_step(self, batch_data):\n        real_monet, real_photo = batch_data\n        \n        with tf.GradientTape(persistent=True) as tape:\n            # photo to monet back to photo\n            fake_monet = self.m_gen(real_photo, training=True)\n            cycled_photo = self.p_gen(fake_monet, training=True)\n\n            # monet to photo back to monet\n            fake_photo = self.p_gen(real_monet, training=True)\n            cycled_monet = self.m_gen(fake_photo, training=True)\n\n            # generating itself\n            same_monet = self.m_gen(real_monet, training=True)\n            same_photo = self.p_gen(real_photo, training=True)\n\n            # discriminator used to check, inputing real images\n            disc_real_monet = self.m_disc(real_monet, training=True)\n            disc_real_photo = self.p_disc(real_photo, training=True)\n\n            # discriminator used to check, inputing fake images\n            disc_fake_monet = self.m_disc(fake_monet, training=True)\n            disc_fake_photo = self.p_disc(fake_photo, training=True)\n\n            # evaluates generator loss\n            monet_gen_loss = self.gen_loss_fn(disc_fake_monet)\n            photo_gen_loss = self.gen_loss_fn(disc_fake_photo)\n\n            # evaluates total cycle consistency loss\n            total_cycle_loss = self.cycle_loss_fn(real_monet, cycled_monet, self.lambda_cycle) + self.cycle_loss_fn(real_photo, cycled_photo, self.lambda_cycle)\n\n            # evaluates total generator loss\n            total_monet_gen_loss = monet_gen_loss + total_cycle_loss + self.identity_loss_fn(real_monet, same_monet, self.lambda_cycle)\n            total_photo_gen_loss = photo_gen_loss + total_cycle_loss + self.identity_loss_fn(real_photo, same_photo, self.lambda_cycle)\n\n            # evaluates discriminator loss\n            monet_disc_loss = self.disc_loss_fn(disc_real_monet, disc_fake_monet)\n            photo_disc_loss = self.disc_loss_fn(disc_real_photo, disc_fake_photo)\n\n        # Calculate the gradients for generator and discriminator\n        monet_generator_gradients = tape.gradient(total_monet_gen_loss,\n                                                  self.m_gen.trainable_variables)\n        photo_generator_gradients = tape.gradient(total_photo_gen_loss,\n                                                  self.p_gen.trainable_variables)\n\n        monet_discriminator_gradients = tape.gradient(monet_disc_loss,\n                                                      self.m_disc.trainable_variables)\n        photo_discriminator_gradients = tape.gradient(photo_disc_loss,\n                                                      self.p_disc.trainable_variables)\n\n        # Apply the gradients to the optimizer\n        self.m_gen_optimizer.apply_gradients(zip(monet_generator_gradients,\n                                                 self.m_gen.trainable_variables))\n\n        self.p_gen_optimizer.apply_gradients(zip(photo_generator_gradients,\n                                                 self.p_gen.trainable_variables))\n\n        self.m_disc_optimizer.apply_gradients(zip(monet_discriminator_gradients,\n                                                  self.m_disc.trainable_variables))\n\n        self.p_disc_optimizer.apply_gradients(zip(photo_discriminator_gradients,\n                                                  self.p_disc.trainable_variables))\n        \n        return {\n            \"monet_gen_loss\": total_monet_gen_loss,\n            \"photo_gen_loss\": total_photo_gen_loss,\n            \"monet_disc_loss\": monet_disc_loss,\n            \"photo_disc_loss\": photo_disc_loss\n        }","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:31:29.599410Z","iopub.execute_input":"2021-07-22T10:31:29.599756Z","iopub.status.idle":"2021-07-22T10:31:29.618559Z","shell.execute_reply.started":"2021-07-22T10:31:29.599721Z","shell.execute_reply":"2021-07-22T10:31:29.617718Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### **Define loss functions**","metadata":{}},{"cell_type":"code","source":"with strategy.scope():\n    def discriminator_loss(real, generated):\n        real_loss = tf.keras.losses.BinaryCrossentropy(from_logits=True, reduction=tf.keras.losses.Reduction.NONE)(tf.ones_like(real), real)\n\n        generated_loss = tf.keras.losses.BinaryCrossentropy(from_logits=True, reduction=tf.keras.losses.Reduction.NONE)(tf.zeros_like(generated), generated)\n\n        total_disc_loss = real_loss + generated_loss\n\n        return total_disc_loss * 0.5","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:31:30.634709Z","iopub.execute_input":"2021-07-22T10:31:30.635029Z","iopub.status.idle":"2021-07-22T10:31:30.641314Z","shell.execute_reply.started":"2021-07-22T10:31:30.635000Z","shell.execute_reply":"2021-07-22T10:31:30.640321Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"with strategy.scope():\n    def generator_loss(generated):\n        return tf.keras.losses.BinaryCrossentropy(from_logits=True, reduction=tf.keras.losses.Reduction.NONE)(tf.ones_like(generated), generated)","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:31:31.812704Z","iopub.execute_input":"2021-07-22T10:31:31.813052Z","iopub.status.idle":"2021-07-22T10:31:31.818562Z","shell.execute_reply.started":"2021-07-22T10:31:31.813016Z","shell.execute_reply":"2021-07-22T10:31:31.817288Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"with strategy.scope():\n    def calc_cycle_loss(real_image, cycled_image, LAMBDA):\n        loss1 = tf.reduce_mean(tf.abs(real_image - cycled_image))\n\n        return LAMBDA * loss1","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:31:32.876483Z","iopub.execute_input":"2021-07-22T10:31:32.876890Z","iopub.status.idle":"2021-07-22T10:31:32.882867Z","shell.execute_reply.started":"2021-07-22T10:31:32.876855Z","shell.execute_reply":"2021-07-22T10:31:32.881873Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"with strategy.scope():\n    def identity_loss(real_image, same_image, LAMBDA):\n        loss = tf.reduce_mean(tf.abs(real_image - same_image))\n        return LAMBDA * 0.5 * loss","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:31:34.368241Z","iopub.execute_input":"2021-07-22T10:31:34.368613Z","iopub.status.idle":"2021-07-22T10:31:34.374011Z","shell.execute_reply.started":"2021-07-22T10:31:34.368562Z","shell.execute_reply":"2021-07-22T10:31:34.372853Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### **Train the CycleGAN**","metadata":{}},{"cell_type":"code","source":"with strategy.scope():\n    monet_generator_optimizer = tf.keras.optimizers.Adam(GENERATOR_LR, beta_1=0.5)\n    photo_generator_optimizer = tf.keras.optimizers.Adam(GENERATOR_LR, beta_1=0.5)\n\n    monet_discriminator_optimizer = tf.keras.optimizers.Adam(DISCRIMINATOR_LR, beta_1=0.5)\n    photo_discriminator_optimizer = tf.keras.optimizers.Adam(DISCRIMINATOR_LR, beta_1=0.5)","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:31:36.107751Z","iopub.execute_input":"2021-07-22T10:31:36.108104Z","iopub.status.idle":"2021-07-22T10:31:36.114010Z","shell.execute_reply.started":"2021-07-22T10:31:36.108074Z","shell.execute_reply":"2021-07-22T10:31:36.113144Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"with strategy.scope():\n    cycle_gan_model = CycleGan(\n        monet_generator, photo_generator, monet_discriminator, photo_discriminator\n    )\n\n    cycle_gan_model.compile(\n        m_gen_optimizer = monet_generator_optimizer,\n        p_gen_optimizer = photo_generator_optimizer,\n        m_disc_optimizer = monet_discriminator_optimizer,\n        p_disc_optimizer = photo_discriminator_optimizer,\n        gen_loss_fn = generator_loss,\n        disc_loss_fn = discriminator_loss,\n        cycle_loss_fn = calc_cycle_loss,\n        identity_loss_fn = identity_loss\n    )","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:31:38.663478Z","iopub.execute_input":"2021-07-22T10:31:38.663879Z","iopub.status.idle":"2021-07-22T10:31:38.690054Z","shell.execute_reply.started":"2021-07-22T10:31:38.663846Z","shell.execute_reply":"2021-07-22T10:31:38.689246Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"history = cycle_gan_model.fit(\n    tf.data.Dataset.zip((monet_ds, photo_ds)),\n    epochs=EPOCHS\n    # steps_per_epoch=(max(n_monet_samples, n_photo_samples)//BATCH_SIZE)\n)","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:31:43.767569Z","iopub.execute_input":"2021-07-22T10:31:43.767917Z","iopub.status.idle":"2021-07-22T10:54:30.365245Z","shell.execute_reply.started":"2021-07-22T10:31:43.767886Z","shell.execute_reply":"2021-07-22T10:54:30.364350Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### **Save Models and History**","metadata":{}},{"cell_type":"code","source":"import pickle\n\nmonet_generator.save('gen_monet.h5')\nphoto_generator.save('gen_photo.h5')\nmonet_discriminator.save('disc_monet.h5')\nphoto_discriminator.save('disc_photo.h5')\n\nwith open('history.pkl','wb') as f:\n    pickle.dump(history.history, f)","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:54:30.367327Z","iopub.execute_input":"2021-07-22T10:54:30.367624Z","iopub.status.idle":"2021-07-22T10:54:30.562596Z","shell.execute_reply.started":"2021-07-22T10:54:30.367574Z","shell.execute_reply":"2021-07-22T10:54:30.561729Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### **Load Models and History**","metadata":{}},{"cell_type":"code","source":"import pickle\n\ngen_monet = tf.keras.models.load_model('./gen_monet.h5')\ngen_photo = tf.keras.models.load_model('./gen_photo.h5')\ndisc_monet = tf.keras.models.load_model('./disc_monet.h5')\ndisc_photo = tf.keras.models.load_model('./disc_photo.h5')\n\nwith (open(\"./history.pkl\", \"rb\")) as openfile:\n    history = pickle.load(openfile)","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:54:30.564505Z","iopub.execute_input":"2021-07-22T10:54:30.564878Z","iopub.status.idle":"2021-07-22T10:54:31.492374Z","shell.execute_reply.started":"2021-07-22T10:54:30.564841Z","shell.execute_reply":"2021-07-22T10:54:31.491514Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### **Visualize predictions**","metadata":{}},{"cell_type":"code","source":"def display_generated_samples(ds, monet_generator, photo_generator,n_samples):\n    ds_iter = iter(ds)\n    f,ax = plt.subplots(n_samples,3, figsize=(22,n_samples*6))\n    for n_sample in range(n_samples):\n        example_sample = next(ds_iter)\n        generated_sample = monet_generator.predict(example_sample)\n        cycledphoto_sample = photo_generator.predict(generated_sample)\n \n        ax[n_sample,0].set_title(\"Input image\")\n        ax[n_sample,0].axis('off')\n        ax[n_sample,0].imshow(example_sample[0]  * 0.5 + 0.5)\n \n        ax[n_sample,1].set_title(\"Generated image\")\n        ax[n_sample,1].axis('off')\n        ax[n_sample,1].imshow(generated_sample[0] * 0.5 + 0.5)\n \n        ax[n_sample,2].set_title(\"Cycled image\")\n        ax[n_sample,2].axis('off')\n        ax[n_sample,2].imshow(cycledphoto_sample[0] * 0.5 + 0.5)","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:54:31.494781Z","iopub.execute_input":"2021-07-22T10:54:31.495366Z","iopub.status.idle":"2021-07-22T10:54:31.503457Z","shell.execute_reply.started":"2021-07-22T10:54:31.495329Z","shell.execute_reply":"2021-07-22T10:54:31.502594Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"display_generated_samples(photo_ds, gen_monet, gen_photo, 8)","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:54:31.506100Z","iopub.execute_input":"2021-07-22T10:54:31.506422Z","iopub.status.idle":"2021-07-22T10:54:35.942916Z","shell.execute_reply.started":"2021-07-22T10:54:31.506359Z","shell.execute_reply":"2021-07-22T10:54:35.941864Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def plot_acc_and_loss(history, load=False):\n    monet_g = []\n    photo_g = []\n    monet_d = []\n    photo_d = []\n\n    if load == False:\n        history = history.history\n    \n    for i in range(np.array(history[\"monet_gen_loss\"]).shape[0]):\n            monet_g.append(np.array(history[\"monet_gen_loss\"][i]).squeeze().mean())\n            photo_g.append(np.array(history[\"photo_gen_loss\"][i]).squeeze().mean())\n            monet_d.append(np.array(history[\"monet_disc_loss\"][i]).squeeze().mean())\n            photo_d.append(np.array(history[\"photo_disc_loss\"][i]).squeeze().mean())\n    \n    fig, axs = plt.subplots(1, 2, figsize=(15, 5))\n    axs[0].plot(monet_g,label=\"Monet\")\n    axs[0].plot(photo_g,label=\"Photo\")\n    axs[0].set_title(\"generator loss\")\n    axs[0].legend()\n\n    axs[1].plot(monet_d, label=\"Monet\")\n    axs[1].plot(photo_d,label=\"Photo\")\n    axs[1].set_title(\"discriminator loss\")\n    axs[1].legend()\n\n    plt.show()","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:54:35.944370Z","iopub.execute_input":"2021-07-22T10:54:35.944902Z","iopub.status.idle":"2021-07-22T10:54:35.956692Z","shell.execute_reply.started":"2021-07-22T10:54:35.944866Z","shell.execute_reply":"2021-07-22T10:54:35.955964Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plot_acc_and_loss(history, load=True)","metadata":{"execution":{"iopub.status.busy":"2021-07-22T10:54:35.958106Z","iopub.execute_input":"2021-07-22T10:54:35.958717Z","iopub.status.idle":"2021-07-22T10:54:36.256021Z","shell.execute_reply.started":"2021-07-22T10:54:35.958674Z","shell.execute_reply":"2021-07-22T10:54:36.255148Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### 1. Add skip connections","metadata":{}},{"cell_type":"code","source":"def Generator_with_skip_connections():\n\n    inputs = layers.Input(shape=[HEIGHT, WIDTH, CHANNELS], name='input_image')\n\n    # Encoders\n    enc_1 = encoder_block(inputs, 64,  7, 1, apply_instancenorm=False, activation=layers.ReLU())\n    enc_2 = encoder_block(enc_1, 128, 3, 2, apply_instancenorm=True, activation=layers.ReLU())\n    enc_3 = encoder_block(enc_2, 256, 3, 2, apply_instancenorm=True, activation=layers.ReLU())\n\n    x = enc_3\n    # Transforms\n    for i in range(TRANSFORMER_BLOCKS):\n        x = transform_block(x, 256, 3, 1)\n    \n    # Decoders\n\n    x_skip = layers.Concatenate(name='enc_dec_skip_1')([x, enc_3]) # encoder - decoder skip connection\n    \n    dec_1 = decoder_block(x_skip, 128, 3, 2, apply_instancenorm=True) \n    x_skip = layers.Concatenate(name='enc_dec_skip_2')([dec_1, enc_2]) # encoder - decoder skip connection\n    \n    dec_2 = decoder_block(x_skip, 64,  3, 2, apply_instancenorm=True) \n    x_skip = layers.Concatenate(name='enc_dec_skip_3')([dec_2, enc_1]) # encoder - decoder skip connection\n\n    outputs = layers.Conv2DTranspose(CHANNELS, 7, strides=1, padding='same', activation='tanh')(x_skip)\n\n    return tf.keras.Model(inputs=inputs, outputs=outputs)\n\ndef create_adam_optimizers():\n\n    monet_generator_optimizer = tf.keras.optimizers.Adam(GENERATOR_LR, beta_1=0.5)\n    photo_generator_optimizer = tf.keras.optimizers.Adam(GENERATOR_LR, beta_1=0.5)\n    monet_discriminator_optimizer = tf.keras.optimizers.Adam(DISCRIMINATOR_LR, beta_1=0.5)\n    photo_discriminator_optimizer = tf.keras.optimizers.Adam(DISCRIMINATOR_LR, beta_1=0.5)\n\n    return monet_generator_optimizer, photo_generator_optimizer, monet_discriminator_optimizer, photo_discriminator_optimizer\n\ndef create_cycle_gan_model(monet_generator, photo_generator,monet_discriminator, photo_discriminator, \n                           monet_generator_optimizer, photo_generator_optimizer, monet_discriminator_optimizer, photo_discriminator_optimizer):\n  \n    cycle_gan_model = CycleGan(\n      monet_generator, photo_generator, monet_discriminator, photo_discriminator\n    )\n\n    cycle_gan_model.compile(\n        m_gen_optimizer = monet_generator_optimizer,\n        p_gen_optimizer = photo_generator_optimizer,\n        m_disc_optimizer = monet_discriminator_optimizer,\n        p_disc_optimizer = photo_discriminator_optimizer,\n        gen_loss_fn = generator_loss,\n        disc_loss_fn = discriminator_loss,\n        cycle_loss_fn = calc_cycle_loss,\n        identity_loss_fn = identity_loss\n    )\n\n    return cycle_gan_model","metadata":{"execution":{"iopub.status.busy":"2021-07-22T11:17:04.562788Z","iopub.execute_input":"2021-07-22T11:17:04.563175Z","iopub.status.idle":"2021-07-22T11:17:04.578388Z","shell.execute_reply.started":"2021-07-22T11:17:04.563144Z","shell.execute_reply":"2021-07-22T11:17:04.577452Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"monet_generator = Generator_with_skip_connections() # transforms photos to Monet-esque paintings\nphoto_generator = Generator_with_skip_connections() # transforms Monet paintings to be more like photos\nmonet_discriminator = Discriminator() # differentiates real Monet paintings and generated Monet paintings\nphoto_discriminator = Discriminator() # differentiates real photos and generated photos\n\nmonet_generator_optimizer, photo_generator_optimizer, monet_discriminator_optimizer, photo_discriminator_optimizer = create_adam_optimizers()\n\ncycle_gan_model = create_cycle_gan_model(monet_generator, photo_generator,monet_discriminator, photo_discriminator, \n                                         monet_generator_optimizer, photo_generator_optimizer, monet_discriminator_optimizer, photo_discriminator_optimizer)\n\nhistory = cycle_gan_model.fit(\n    tf.data.Dataset.zip((monet_ds, photo_ds)),\n    epochs=EPOCHS\n)","metadata":{"execution":{"iopub.status.busy":"2021-07-22T11:17:09.149258Z","iopub.execute_input":"2021-07-22T11:17:09.149575Z","iopub.status.idle":"2021-07-22T11:43:20.562069Z","shell.execute_reply.started":"2021-07-22T11:17:09.149546Z","shell.execute_reply":"2021-07-22T11:43:20.560927Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"display_generated_samples(photo_ds, monet_generator, photo_generator, 8)","metadata":{"execution":{"iopub.status.busy":"2021-07-22T11:43:20.564295Z","iopub.execute_input":"2021-07-22T11:43:20.564557Z","iopub.status.idle":"2021-07-22T11:43:24.653751Z","shell.execute_reply.started":"2021-07-22T11:43:20.564530Z","shell.execute_reply":"2021-07-22T11:43:24.646573Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"monet_generator.save('gen_monet_with_norm.h5')\nphoto_generator.save('gen_photo_with_norm.h5')\nmonet_discriminator.save('disc_monet_with_norm.h5')\nphoto_discriminator.save('disc_photo_with_norm.h5')","metadata":{"execution":{"iopub.status.busy":"2021-07-22T11:43:24.655680Z","iopub.execute_input":"2021-07-22T11:43:24.656048Z","iopub.status.idle":"2021-07-22T11:43:24.864863Z","shell.execute_reply.started":"2021-07-22T11:43:24.656013Z","shell.execute_reply":"2021-07-22T11:43:24.864020Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plot_acc_and_loss(history, load=False)","metadata":{"execution":{"iopub.status.busy":"2021-07-22T11:43:24.868245Z","iopub.execute_input":"2021-07-22T11:43:24.868519Z","iopub.status.idle":"2021-07-22T11:43:25.176641Z","shell.execute_reply.started":"2021-07-22T11:43:24.868493Z","shell.execute_reply":"2021-07-22T11:43:25.175829Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### 2. Data augmentations (flips, rotations, and crops)","metadata":{}},{"cell_type":"code","source":"monet_generator = Generator_with_skip_connections() # transforms photos to Monet-esque paintings\nphoto_generator = Generator_with_skip_connections() # transforms Monet paintings to be more like photos\nmonet_discriminator = Discriminator() # differentiates real Monet paintings and generated Monet paintings\nphoto_discriminator = Discriminator() # differentiates real photos and generated photos\n\nmonet_generator_optimizer, photo_generator_optimizer, monet_discriminator_optimizer, photo_discriminator_optimizer = create_adam_optimizers()\n\ncycle_gan_model = create_cycle_gan_model(monet_generator, photo_generator,monet_discriminator, photo_discriminator, \n                                         monet_generator_optimizer, photo_generator_optimizer, monet_discriminator_optimizer, photo_discriminator_optimizer)\n\nHEIGHT_RESIZE = 256\nWIDTH_RESIZE = 256\n\ndef data_augment(image):\n    p_spatial = tf.random.uniform([], 0, 1.0, dtype=tf.float32)\n    p_rotate = tf.random.uniform([], 0, 1.0, dtype=tf.float32)\n    \n    # 90º rotations\n    if p_rotate > .8:\n        image = tf.image.rot90(image, k=3) # rotate 270º\n    elif p_rotate > .6:\n        image = tf.image.rot90(image, k=2) # rotate 180º\n    elif p_rotate > .4:\n        image = tf.image.rot90(image, k=1) # rotate 90º\n        \n    # Flips\n    image = tf.image.random_flip_left_right(image)\n    image = tf.image.random_flip_up_down(image)\n    if p_spatial > .75:\n        image = tf.image.transpose(image)\n            \n    # Train on crops\n    image = tf.image.random_crop(image, size=[HEIGHT_RESIZE, WIDTH_RESIZE, CHANNELS])\n        \n    return image\n\nmonet_ds = get_dataset(MONET_FILENAMES,augment=data_augment, repeat=False, shuffle=True, batch_size=BATCH_SIZE)\nphoto_ds = get_dataset(PHOTO_FILENAMES,augment=data_augment, repeat=False, shuffle=True, batch_size=BATCH_SIZE)\n\nhistory = cycle_gan_model.fit(\n    tf.data.Dataset.zip((monet_ds, photo_ds)),\n    epochs=EPOCHS\n    # steps_per_epoch=(max(n_monet_samples, n_photo_samples)//BATCH_SIZE)\n)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"display_generated_samples(photo_ds, monet_generator, photo_generator, 8)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"monet_generator.save('gen_monet_with_data_augment.h5')\nphoto_generator.save('gen_photo_with_data_augment.h5')\nmonet_discriminator.save('disc_monet_with_data_augment.h5')\nphoto_discriminator.save('disc_photo_with_data_augment.h5')","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plot_acc_and_loss(history, load=False)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### 3. Change batch size from 1 to 4","metadata":{}},{"cell_type":"code","source":"monet_generator = Generator_with_skip_connections() # transforms photos to Monet-esque paintings\nphoto_generator = Generator_with_skip_connections() # transforms Monet paintings to be more like photos\nmonet_discriminator = Discriminator() # differentiates real Monet paintings and generated Monet paintings\nphoto_discriminator = Discriminator() # differentiates real photos and generated photos\n\nBATCH_SIZE = 4\n\nmonet_generator_optimizer, photo_generator_optimizer, monet_discriminator_optimizer, photo_discriminator_optimizer = create_adam_optimizers()\n\ncycle_gan_model = create_cycle_gan_model(monet_generator, photo_generator,monet_discriminator, photo_discriminator, \n                                         monet_generator_optimizer, photo_generator_optimizer, monet_discriminator_optimizer, photo_discriminator_optimizer)\n\nmonet_ds = get_dataset(MONET_FILENAMES,augment=data_augment, repeat=False, shuffle=True, batch_size=BATCH_SIZE)\nphoto_ds = get_dataset(PHOTO_FILENAMES,augment=data_augment, repeat=False, shuffle=True, batch_size=BATCH_SIZE)\n\nhistory = cycle_gan_model.fit(\n    tf.data.Dataset.zip((monet_ds, photo_ds)),\n    epochs=EPOCHS\n    # steps_per_epoch=(max(n_monet_samples, n_photo_samples)//BATCH_SIZE)\n)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"display_generated_samples(photo_ds, monet_generator, photo_generator, 8)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"monet_generator.save('gen_monet_with_batch_size_4.h5')\nphoto_generator.save('gen_photo_with_batch_size_4.h5')\nmonet_discriminator.save('disc_monet_batch_size_4.h5')\nphoto_discriminator.save('disc_photo_batch_size_4.h5')","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plot_acc_and_loss(history, load=False)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### 4. Change generator and discriminator learning rate from 2e-4 to 0.01","metadata":{}},{"cell_type":"code","source":"GENERATOR_LR = 0.01\nDISCRIMINATOR_LR = 0.01\nEPOCHS = 25\nBATCH_SIZE = 1\n\nmonet_generator = Generator_with_skip_connections() # transforms photos to Monet-esque paintings\nphoto_generator = Generator_with_skip_connections() # transforms Monet paintings to be more like photos\nmonet_discriminator = Discriminator() # differentiates real Monet paintings and generated Monet paintings\nphoto_discriminator = Discriminator() # differentiates real photos and generated photos\n\nmonet_generator_optimizer, photo_generator_optimizer, monet_discriminator_optimizer, photo_discriminator_optimizer = create_adam_optimizers()\n\ncycle_gan_model = create_cycle_gan_model(monet_generator, photo_generator,monet_discriminator, photo_discriminator, \n                                         monet_generator_optimizer, photo_generator_optimizer, monet_discriminator_optimizer, photo_discriminator_optimizer)\n\nmonet_ds = get_dataset(MONET_FILENAMES,augment=None, repeat=False, shuffle=True, batch_size=BATCH_SIZE)\nphoto_ds = get_dataset(PHOTO_FILENAMES,augment=None, repeat=False, shuffle=True, batch_size=BATCH_SIZE)\n\nhistory = cycle_gan_model.fit(\n    tf.data.Dataset.zip((monet_ds, photo_ds)),\n    epochs=EPOCHS\n    # steps_per_epoch=(max(n_monet_samples, n_photo_samples)//BATCH_SIZE)\n)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"display_generated_samples(photo_ds, monet_generator, photo_generator, 8)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"monet_generator.save('gen_monet_with_lr_0.01.h5')\nphoto_generator.save('gen_photo_with_lr_0.01.h5')\nmonet_discriminator.save('disc_monet_lr_0.01.h5')\nphoto_discriminator.save('disc_photo_lr_0.01.h5')","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plot_acc_and_loss(history, load=False)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### 5. Change optimizer from Adam to SGD","metadata":{}},{"cell_type":"code","source":"GENERATOR_LR = 2e-4\nDISCRIMINATOR_LR = 2e-4\n\nBATCH_SIZE = 1\n\nmonet_generator = Generator_with_skip_connections() # transforms photos to Monet-esque paintings\nphoto_generator = Generator_with_skip_connections() # transforms Monet paintings to be more like photos\nmonet_discriminator = Discriminator() # differentiates real Monet paintings and generated Monet paintings\nphoto_discriminator = Discriminator() # differentiates real photos and generated photos\n\n\ndef create_SGD_optimizers():\n\n    monet_generator_optimizer = tf.keras.optimizers.SGD(GENERATOR_LR)\n    photo_generator_optimizer = tf.keras.optimizers.SGD(GENERATOR_LR)\n    monet_discriminator_optimizer = tf.keras.optimizers.SGD(DISCRIMINATOR_LR)\n    photo_discriminator_optimizer = tf.keras.optimizers.SGD(DISCRIMINATOR_LR)\n\n    return monet_generator_optimizer, photo_generator_optimizer, monet_discriminator_optimizer, photo_discriminator_optimizer\n\nmonet_generator_optimizer, photo_generator_optimizer, monet_discriminator_optimizer, photo_discriminator_optimizer = create_SGD_optimizers()\n\ncycle_gan_model = create_cycle_gan_model(monet_generator, photo_generator,monet_discriminator, photo_discriminator, \n                                         monet_generator_optimizer, photo_generator_optimizer, monet_discriminator_optimizer, photo_discriminator_optimizer)\n\nmonet_ds = get_dataset(MONET_FILENAMES,augment=data_augment, repeat=False, shuffle=True, batch_size=BATCH_SIZE)\nphoto_ds = get_dataset(PHOTO_FILENAMES,augment=data_augment, repeat=False, shuffle=True, batch_size=BATCH_SIZE)\n\nhistory = cycle_gan_model.fit(\n    tf.data.Dataset.zip((monet_ds, photo_ds)),\n    epochs=EPOCHS\n    # steps_per_epoch=(max(n_monet_samples, n_photo_samples)//BATCH_SIZE)\n)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"display_generated_samples(photo_ds, monet_generator, photo_generator, 8)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"monet_generator.save('gen_monet_with_lr_0.01.h5')\nphoto_generator.save('gen_photo_with_lr_0.01.h5')\nmonet_discriminator.save('disc_monet_lr_0.01.h5')\nphoto_discriminator.save('disc_photo_lr_0.01.h5')","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plot_acc_and_loss(history, load=False)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### 6. Change Generator and Discriminator Architecture\n","metadata":{}},{"cell_type":"code","source":"conv_initializer = tf.random_normal_initializer(mean=0.0, stddev=0.02)\ngamma_initializer = tf.keras.initializers.RandomNormal(mean=0.0, stddev=0.02)\n\ndef downsample_block(filters, size, stride=2, apply_instancenorm=True, activation = layers.LeakyReLU()):\n\n    result = keras.Sequential()\n    result.add(layers.Conv2D(filters, kernel_size=size, strides=stride, padding=\"same\", use_bias=False, kernel_initializer=conv_initializer))\n\n    if apply_instancenorm:\n        result.add(tfa.layers.InstanceNormalization(gamma_initializer=gamma_initializer))\n\n    result.add(activation)\n\n    return result\n\ndef upsample_block(filters, size, stride=2, apply_dropout=False, apply_instancenorm=True, activation = layers.ReLU()):\n\n    result = keras.Sequential()\n    result.add(layers.Conv2DTranspose(filters, kernel_size=size, strides=stride, padding=\"same\", use_bias=False, kernel_initializer=conv_initializer))\n\n    if apply_instancenorm:\n        result.add(tfa.layers.InstanceNormalization(gamma_initializer=gamma_initializer))\n\n    if apply_dropout:\n        result.add(layers.Dropout(0.5))\n\n    result.add(activation)\n\n    return result\n\ndef Generator_3():\n    inputs = layers.Input(shape=[HEIGHT, WIDTH, CHANNELS], name='input_image')\n\n    down_stack = [\n        downsample_block(64, 4, apply_instancenorm=False), \n        downsample_block(128, 4), \n        downsample_block(256, 4), \n        downsample_block(512, 4), \n        downsample_block(512, 4), \n        downsample_block(512, 4), \n        downsample_block(512, 4), \n        downsample_block(512, 4), \n    ]\n\n    up_stack = [\n        upsample_block(512, 4, apply_dropout=True), \n        upsample_block(512, 4, apply_dropout=True), \n        upsample_block(512, 4, apply_dropout=True), \n        upsample_block(512, 4), \n        upsample_block(256, 4), \n        upsample_block(128, 4), \n        upsample_block(64, 4), \n    ]\n\n    last = layers.Conv2DTranspose(CHANNELS, 4,\n                                  strides=2,\n                                  padding='same',\n                                  kernel_initializer=conv_initializer,\n                                  activation='tanh')\n    x = inputs\n\n    # Downsampling through the model\n    skips = []\n    for down in down_stack:\n        x = down(x)\n        skips.append(x)\n\n    skips = reversed(skips[:-1])\n\n    # Upsampling and establishing the skip connections\n    for up, skip in zip(up_stack, skips):\n        x = up(x)\n        x = layers.Concatenate()([x, skip])\n\n    x = last(x)\n\n    return keras.Model(inputs=inputs, outputs=x)\n\n\ndef Discriminator_3():\n\n    inputs = layers.Input(shape=[HEIGHT, WIDTH, CHANNELS], name='input_image')\n\n    temp = inputs\n\n    down1 = downsample_block(64, 4, apply_instancenorm=False)(temp) \n    down2 = downsample_block(128, 4)(down1)\n    down3 = downsample_block(256, 4)(down2)\n\n    zero_pad1 = layers.ZeroPadding2D()(down3)\n    conv = layers.Conv2D(512, 4, strides=1,\n                         kernel_initializer=conv_initializer,\n                         use_bias=False)(zero_pad1)\n\n    norm1 = tfa.layers.InstanceNormalization(gamma_initializer=gamma_initializer)(conv)\n\n    leaky_relu = layers.LeakyReLU()(norm1)\n\n    zero_pad2 = layers.ZeroPadding2D()(leaky_relu)\n\n    last = layers.Conv2D(1, 4, strides=1,\n                         kernel_initializer=conv_initializer)(zero_pad2)\n\n    return tf.keras.Model(inputs=inputs, outputs=last)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"monet_generator = Generator_3() # transforms photos to Monet-esque paintings\nphoto_generator = Generator_3() # transforms Monet paintings to be more like photos\nmonet_discriminator = Discriminator_3() # differentiates real Monet paintings and generated Monet paintings\nphoto_discriminator = Discriminator_3() # differentiates real photos and generated photos","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"monet_generator_optimizer, photo_generator_optimizer, monet_discriminator_optimizer, photo_discriminator_optimizer = create_adam_optimizers()\n\ncycle_gan_model = create_cycle_gan_model(monet_generator, photo_generator,monet_discriminator, photo_discriminator, \n                                         monet_generator_optimizer, photo_generator_optimizer, monet_discriminator_optimizer, photo_discriminator_optimizer)\n\nmonet_ds = get_dataset(MONET_FILENAMES,augment=None, repeat=False, shuffle=True, batch_size=BATCH_SIZE)\nphoto_ds = get_dataset(PHOTO_FILENAMES,augment=None, repeat=False, shuffle=True, batch_size=BATCH_SIZE)\n\nhistory = cycle_gan_model.fit(\n    tf.data.Dataset.zip((monet_ds, photo_ds)),\n    epochs=EPOCHS\n    # steps_per_epoch=(max(n_monet_samples, n_photo_samples)//BATCH_SIZE)\n)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"display_generated_samples(photo_ds, monet_generator, photo_generator, 8)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"monet_generator.save('gen_monet_with_new_architecture.h5')\nphoto_generator.save('gen_photo_with_new_architecture.h5')\nmonet_discriminator.save('disc_monet_with_new_architecture.h5')\nphoto_discriminator.save('disc_photo_with_new_architecture.h5')","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plot_acc_and_loss(history, load=False)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### 7. Change the droupout value from 0.5 to 0.3 and change the epochs  value from 25 to 100","metadata":{}},{"cell_type":"code","source":"def downsample_block(filters, size, stride=2, apply_instancenorm=True, activation = layers.LeakyReLU()):\n\n    conv_initializer = tf.random_normal_initializer(mean=0.0, stddev=0.02)\n    gamma_initializer = tf.keras.initializers.RandomNormal(mean=0.0, stddev=0.02)\n\n    result = keras.Sequential()\n    result.add(layers.Conv2D(filters, kernel_size=size, strides=stride, padding=\"same\", use_bias=False, kernel_initializer=conv_initializer))\n\n    if apply_instancenorm:\n        result.add(tfa.layers.InstanceNormalization(gamma_initializer=gamma_initializer))\n\n    result.add(activation)\n\n    return result\n\ndef upsample_block(filters, size, stride=2, apply_dropout=False, apply_instancenorm=True, activation = layers.ReLU()):\n\n    conv_initializer = tf.random_normal_initializer(mean=0.0, stddev=0.02)\n    gamma_initializer = tf.keras.initializers.RandomNormal(mean=0.0, stddev=0.02)\n\n    result = keras.Sequential()\n    result.add(layers.Conv2DTranspose(filters, kernel_size=size, strides=stride, padding=\"same\", use_bias=False, kernel_initializer=conv_initializer))\n\n    if apply_instancenorm:\n        result.add(tfa.layers.InstanceNormalization(gamma_initializer=gamma_initializer))\n\n    if apply_dropout:\n        result.add(layers.Dropout(0.3))\n\n    result.add(activation)\n\n    return result\n\ndef Generator_with_new_dropout():\n    inputs = layers.Input(shape=[HEIGHT, WIDTH, CHANNELS], name='input_image')\n\n    down_stack = [\n        downsample_block(64, 4, apply_instancenorm=False), \n        downsample_block(128, 4), \n        downsample_block(256, 4), \n        downsample_block(512, 4), \n        downsample_block(512, 4), \n        downsample_block(512, 4), \n        downsample_block(512, 4), \n        downsample_block(512, 4), \n    ]\n\n    up_stack = [\n        upsample_block(512, 4, apply_dropout=True), \n        upsample_block(512, 4, apply_dropout=True), \n        upsample_block(512, 4, apply_dropout=True), \n        upsample_block(512, 4), \n        upsample_block(256, 4), \n        upsample_block(128, 4), \n        upsample_block(64, 4), \n    ]\n    conv_initializer = tf.random_normal_initializer(mean=0.0, stddev=0.02)\n    last = layers.Conv2DTranspose(CHANNELS, 4,\n                                  strides=2,\n                                  padding='same',\n                                  kernel_initializer=conv_initializer,\n                                  activation='tanh')\n    x = inputs\n\n    # Downsampling through the model\n    skips = []\n    for down in down_stack:\n        x = down(x)\n        skips.append(x)\n\n    skips = reversed(skips[:-1])\n\n    # Upsampling and establishing the skip connections\n    for up, skip in zip(up_stack, skips):\n        x = up(x)\n        x = layers.Concatenate()([x, skip])\n\n    x = last(x)\n\n    return keras.Model(inputs=inputs, outputs=x)\n\n\ndef Discriminator_with_new_dropout():\n\n    conv_initializer = tf.random_normal_initializer(mean=0.0, stddev=0.02)\n    gamma_initializer = tf.keras.initializers.RandomNormal(mean=0.0, stddev=0.02)\n\n\n    inputs = layers.Input(shape=[HEIGHT, WIDTH, CHANNELS], name='input_image')\n\n    temp = inputs\n\n    down1 = downsample_block(64, 4, apply_instancenorm=False)(temp) \n    down2 = downsample_block(128, 4)(down1)\n    down3 = downsample_block(256, 4)(down2)\n\n    zero_pad1 = layers.ZeroPadding2D()(down3)\n    conv = layers.Conv2D(512, 4, strides=1,\n                         kernel_initializer=conv_initializer,\n                         use_bias=False)(zero_pad1)\n\n    norm1 = tfa.layers.InstanceNormalization(gamma_initializer=gamma_initializer)(conv)\n\n    leaky_relu = layers.LeakyReLU()(norm1)\n\n    zero_pad2 = layers.ZeroPadding2D()(leaky_relu)\n\n    last = layers.Conv2D(1, 4, strides=1,\n                         kernel_initializer=conv_initializer)(zero_pad2)\n\n    return tf.keras.Model(inputs=inputs, outputs=last)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"monet_generator = Generator_with_new_dropout() # transforms photos to Monet-esque paintings\nphoto_generator = Generator_with_new_dropout() # transforms Monet paintings to be more like photos\nmonet_discriminator = Discriminator_with_new_dropout() # differentiates real Monet paintings and generated Monet paintings\nphoto_discriminator = Discriminator_with_new_dropout() # differentiates real photos and generated photos","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"GENERATOR_LR = 2e-4\nDISCRIMINATOR_LR = 2e-4\nBATCH_SIZE = 1\nEPOCHS = 100\n\n\nmonet_generator_optimizer, photo_generator_optimizer, monet_discriminator_optimizer, photo_discriminator_optimizer = create_adam_optimizers()\n\ncycle_gan_model = create_cycle_gan_model(monet_generator, photo_generator,monet_discriminator, photo_discriminator, \n                                         monet_generator_optimizer, photo_generator_optimizer, monet_discriminator_optimizer, photo_discriminator_optimizer)\n\nmonet_ds = get_dataset(MONET_FILENAMES,augment=None, repeat=False, shuffle=True, batch_size=BATCH_SIZE)\nphoto_ds = get_dataset(PHOTO_FILENAMES,augment=None, repeat=False, shuffle=True, batch_size=BATCH_SIZE)\n\nhistory = cycle_gan_model.fit(\n    tf.data.Dataset.zip((monet_ds, photo_ds)),\n    epochs=EPOCHS\n    # steps_per_epoch=(max(n_monet_samples, n_photo_samples)//BATCH_SIZE)\n)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"display_generated_samples(photo_ds, monet_generator, photo_generator, 8)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"monet_generator.save('gen_monet_with_new_droupout.h5')\nphoto_generator.save('gen_photo_with_new_droupout.h5')\nmonet_discriminator.save('disc_monet_with_new_droupout.h5')\nphoto_discriminator.save('disc_photo_with_new_droupout.h5')","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plot_acc_and_loss(history, load=False)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### 8. Using external data (1367 files)","metadata":{}},{"cell_type":"code","source":"EXT_MONET_FILENAMES  = tf.io.gfile.glob('../input/monet-tfrecords-256x256/monet*.tfrec')\nALL_MONET_FILENAMES = EXT_MONET_FILENAMES + MONET_FILENAMES","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def read_tfrecord(example):\n    tfrecord_format = {\n        \"image\": tf.io.FixedLenFeature([], tf.string),\n    }\n    example = tf.io.parse_single_example(example, tfrecord_format)\n    image = decode_image(example['image'])\n    return image","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"n_ext_monet_samples  = count_data_items(EXT_MONET_FILENAMES )\nn_all_monet_samples  = count_data_items(ALL_MONET_FILENAMES )\n\nprint(f'Photo TFRecord files: {len(PHOTO_FILENAMES)}')\nprint(f'Photo image files: {n_photo_samples}')\n\nprint(f'Monet TFRecord files: {len(MONET_FILENAMES)}')\nprint(f'Monet image files: {n_monet_samples}')\n\nprint(f'External Monet TFRecord files: {len(EXT_MONET_FILENAMES )}')\nprint(f'External Monet image files: {n_ext_monet_samples }')\n\nprint(f'All Monet TFRecord files: {len(ALL_MONET_FILENAMES )}')\nprint(f'All Monet image files: {n_all_monet_samples }')","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"monet_generator = Generator_with_new_dropout() # transforms photos to Monet-esque paintings\nphoto_generator = Generator_with_new_dropout() # transforms Monet paintings to be more like photos\nmonet_discriminator = Discriminator_with_new_dropout() # differentiates real Monet paintings and generated Monet paintings\nphoto_discriminator = Discriminator_with_new_dropout() # differentiates real photos and generated photos","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"GENERATOR_LR = 2e-4\nDISCRIMINATOR_LR = 2e-4\nBATCH_SIZE = 1\nEPOCHS = 70\n\nmonet_generator_optimizer, photo_generator_optimizer, monet_discriminator_optimizer, photo_discriminator_optimizer = create_adam_optimizers()\n\ncycle_gan_model = create_cycle_gan_model(monet_generator, photo_generator,monet_discriminator, photo_discriminator, \n                                         monet_generator_optimizer, photo_generator_optimizer, monet_discriminator_optimizer, photo_discriminator_optimizer)\n\nmonet_ds = get_dataset(ALL_MONET_FILENAMES,augment=None, repeat=False, shuffle=False, batch_size=BATCH_SIZE)\nphoto_ds = get_dataset(PHOTO_FILENAMES,augment=None, repeat=False, shuffle=False, batch_size=BATCH_SIZE)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def display_generated_samples_grid(ds, n_samples):\n    cols = 5\n    rows = int(n_samples/cols) + round(n_samples%cols)\n    ds_iter = iter(ds)\n    f,ax = plt.subplots(rows ,cols, figsize=(cols*3,rows*3))\n    for n_sample in range(n_samples):\n        example_sample = next(ds_iter)\n\n        ax[int(n_sample/cols),int(n_sample%cols)].axis('off')\n        ax[int(n_sample/cols),int(n_sample%cols)].imshow(example_sample[0]  * 0.5 + 0.5)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"display_generated_samples_grid(monet_ds,15)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# split monet data to train and validatiion\ntrain_monet_size = int(0.9 * n_all_monet_samples)\nval_monet_size = int(0.1 * n_all_monet_samples)\ntrain_monet_dataset = monet_ds.take(train_monet_size)\nvalidation_monet_dataset = monet_ds.skip(train_monet_size)\n\n# split photo data to train and validatiion\ntrain_photo_size = int(0.8 * n_photo_samples)\nval_photo_size = int(0.1 * n_photo_samples)\ntest_photo_size = int(0.1 * n_photo_samples)\n\ntrain_photo_dataset = photo_ds.take(train_photo_size)\ntest_photo_dataset = photo_ds.skip(train_photo_size)\nvalidation_photo_dataset = test_photo_dataset.skip(val_photo_size)\ntest_photo_dataset = test_photo_dataset.take(val_photo_size)\n\nhistory = cycle_gan_model.fit(\n    tf.data.Dataset.zip((train_monet_dataset, train_photo_dataset)),\n    epochs=EPOCHS,\n    # validation_data=tf.data.Dataset.zip((validation_monet_dataset, validation_photo_dataset))\n    steps_per_epoch=(n_all_monet_samples / 300)\n)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"display_generated_samples(test_photo_dataset, monet_generator, photo_generator, 8)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plot_acc_and_loss(history, load=False)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"monet_generator.save('gen_monet_with_external_data.h5')\nphoto_generator.save('gen_photo_with_external_data.h5')\nmonet_discriminator.save('disc_monet_with_external_data.h5')\nphoto_discriminator.save('disc_photo_with_external_data.h5')","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Make predictions and create a submission file","metadata":{}},{"cell_type":"code","source":"def predict_and_save(input_ds, generator_model, output_path):\n    i = 1\n    for img in input_ds:\n        prediction = generator_model(img, training=False)[0].numpy() # make predition\n        prediction = (prediction * 127.5 + 127.5).astype(np.uint8)   # re-scale\n        im = PIL.Image.fromarray(prediction)\n        im.save(f'{output_path}{str(i)}.jpg')\n        i += 1","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import os, shutil,PIL\n\n! mkdir images\npredict_and_save(photo_ds, monet_generator, './images/')","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"shutil.make_archive('../output/kaggle/working/images/', 'zip', './images')\n\nprint(f\"Generated samples: {len([name for name in os.listdir('./images/') if os.path.isfile(os.path.join('./images/', name))])}\")","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import glob\nprint(glob.glob(\"../output/kaggle/working/*\"))","metadata":{"trusted":true},"execution_count":null,"outputs":[]}]}