{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.7.6","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"gpu","dataSources":[{"sourceId":22307,"databundleVersionId":1502524,"sourceType":"competition"},{"sourceId":1421668,"sourceType":"datasetVersion","datasetId":832340},{"sourceId":1560596,"sourceType":"datasetVersion","datasetId":832396},{"sourceId":1579204,"sourceType":"datasetVersion","datasetId":933426},{"sourceId":1614468,"sourceType":"datasetVersion","datasetId":922491}],"dockerImageVersionId":30021,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":true}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"!cp ../input/gdcm-conda-install/gdcm.tar .\n!tar -xvzf gdcm.tar\n!conda install --offline ./gdcm/gdcm-2.8.9-py37h71b2a6d_0.tar.bz2\n!pip install ../input/fastai2-offline/timm-0.2.1-py3-none-any.whl","metadata":{"_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","trusted":true,"execution":{"iopub.status.busy":"2025-04-26T22:38:02.692425Z","iopub.execute_input":"2025-04-26T22:38:02.692664Z","iopub.status.idle":"2025-04-26T22:38:30.727761Z","shell.execute_reply.started":"2025-04-26T22:38:02.692642Z","shell.execute_reply":"2025-04-26T22:38:30.726886Z"}},"outputs":[{"name":"stdout","text":"gdcm/\ngdcm/conda-4.8.4-py37hc8dfbb8_2.tar.bz2\ngdcm/gdcm-2.8.9-py37h71b2a6d_0.tar.bz2\ngdcm/libjpeg-turbo-2.0.3-h516909a_1.tar.bz2\n\nDownloading and Extracting Packages\n######################################################################## | 100% \nPreparing transaction: done\nVerifying transaction: done\nExecuting transaction: done\nProcessing /kaggle/input/fastai2-offline/timm-0.2.1-py3-none-any.whl\nRequirement already satisfied: torch>=1.4 in /opt/conda/lib/python3.7/site-packages (from timm==0.2.1) (1.6.0)\nRequirement already satisfied: torchvision in /opt/conda/lib/python3.7/site-packages (from timm==0.2.1) (0.7.0)\nRequirement already satisfied: future in /opt/conda/lib/python3.7/site-packages (from torch>=1.4->timm==0.2.1) (0.18.2)\nRequirement already satisfied: numpy in /opt/conda/lib/python3.7/site-packages (from torch>=1.4->timm==0.2.1) (1.18.5)\nRequirement already satisfied: pillow>=4.1.1 in /opt/conda/lib/python3.7/site-packages (from torchvision->timm==0.2.1) (8.0.0)\nInstalling collected packages: timm\nSuccessfully installed timm-0.2.1\n\u001b[33mWARNING: You are using pip version 20.2.4; however, version 24.0 is available.\nYou should consider upgrading via the '/opt/conda/bin/python3.7 -m pip install --upgrade pip' command.\u001b[0m\n","output_type":"stream"}],"execution_count":1},{"cell_type":"code","source":"import numpy as np\nimport pandas as pd\nimport os\nimport pydicom\nimport matplotlib.pyplot as plt\nimport seaborn as sns\nfrom sklearn.metrics import confusion_matrix\nfrom albumentations import Compose, Normalize\nfrom albumentations.pytorch import ToTensor\nfrom torch.utils.data import Dataset, DataLoader\nimport torch\nimport torch.nn as nn\nimport torch.nn.functional as F\nfrom torch.cuda.amp import autocast\nimport random\nimport timm\nfrom timm.models.layers.adaptive_avgmax_pool import SelectAdaptivePool2d\nfrom collections import OrderedDict\nfrom tqdm import tqdm\nfrom sklearn.model_selection import train_test_split\nimport torch.optim as optim","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true,"execution":{"iopub.status.busy":"2025-04-26T22:38:33.852895Z","iopub.execute_input":"2025-04-26T22:38:33.853209Z","iopub.status.idle":"2025-04-26T22:38:36.953552Z","shell.execute_reply.started":"2025-04-26T22:38:33.853179Z","shell.execute_reply":"2025-04-26T22:38:36.952818Z"}},"outputs":[],"execution_count":2},{"cell_type":"code","source":"# Load train and test CSVs\ntrain_df = pd.read_csv('../input/rsna-str-pulmonary-embolism-detection/train.csv')\ntest_df = pd.read_csv('../input/rsna-str-pulmonary-embolism-detection/test.csv')\n\n# Combine unique StudyInstanceUIDs\nall_studies = pd.concat([train_df[['StudyInstanceUID']], test_df[['StudyInstanceUID']]])['StudyInstanceUID'].unique()\n\n# Randomly select 300 studies\nnp.random.seed(42)\nselected_studies = np.random.choice(all_studies, size=300, replace=False)\n\n# Split into 240 train and 60 test\nselected_studies_df = pd.DataFrame({'StudyInstanceUID': selected_studies})\ntrain_studies, test_studies = train_test_split(selected_studies_df, test_size=60, random_state=42)\n\n# Filter train and test data\ntrain_subset = train_df[train_df['StudyInstanceUID'].isin(train_studies['StudyInstanceUID'])]\ntest_subset = test_df[test_df['StudyInstanceUID'].isin(test_studies['StudyInstanceUID'])]\ntest_df = test_subset","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-26T22:38:41.026737Z","iopub.execute_input":"2025-04-26T22:38:41.02702Z","iopub.status.idle":"2025-04-26T22:38:44.39692Z","shell.execute_reply.started":"2025-04-26T22:38:41.026995Z","shell.execute_reply":"2025-04-26T22:38:44.396171Z"}},"outputs":[],"execution_count":3},{"cell_type":"code","source":"# Set plot style\nplt.style.use('seaborn')\nsns.set(font_scale=1.2)\n\n# Exam-level labels\nexam_labels = ['negative_exam_for_pe', 'indeterminate', 'chronic_pe', 'acute_and_chronic_pe', \n               'central_pe', 'leftsided_pe', 'rightsided_pe', 'rv_lv_ratio_gte_1', 'rv_lv_ratio_lt_1']\n\n# Plot distribution of exam-level labels in train_subset\nplt.figure(figsize=(20, 12))\ntrain_subset[exam_labels].mean().plot(kind='bar')\nplt.title('Mean Prevalence of Exam-Level Labels (Train Subset)')\nplt.xlabel('Label')\nplt.ylabel('Mean Prevalence')\nplt.xticks(rotation=45, ha='right')\nplt.tight_layout()\nplt.savefig('exam_label_distribution.png')\nplt.show()\nplt.close()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-26T22:38:48.42522Z","iopub.execute_input":"2025-04-26T22:38:48.425556Z","iopub.status.idle":"2025-04-26T22:38:48.853681Z","shell.execute_reply.started":"2025-04-26T22:38:48.425525Z","shell.execute_reply":"2025-04-26T22:38:48.852868Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 1440x864 with 1 Axes>","image/png":"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\n"},"metadata":{}}],"execution_count":4},{"cell_type":"code","source":"import matplotlib.pyplot as plt\nimport seaborn as sns\n\n# Plot distribution of pe_present_on_image\nplt.figure(figsize=(10, 6))\nsns.countplot(x='pe_present_on_image', data=train_subset)\nplt.title('Distribution of pe_present_on_image (Train Subset)')\nplt.xlabel('pe_present_on_image')\nplt.ylabel('Count')\nplt.savefig('pe_present_image_distribution.png')\nplt.show()\nplt.close()\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-26T22:38:54.375357Z","iopub.execute_input":"2025-04-26T22:38:54.375637Z","iopub.status.idle":"2025-04-26T22:38:54.544845Z","shell.execute_reply.started":"2025-04-26T22:38:54.375615Z","shell.execute_reply":"2025-04-26T22:38:54.544001Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 720x432 with 1 Axes>","image/png":"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\n"},"metadata":{}}],"execution_count":5},{"cell_type":"code","source":"# Count images per study\nimages_per_study = train_subset.groupby('StudyInstanceUID')['SOPInstanceUID'].count()\n\n# Plot histogram of images per study\nplt.figure(figsize=(10, 6))\nplt.hist(images_per_study, bins=50)\nplt.title('Distribution of Images per Study (Train Subset)')\nplt.xlabel('Number of Images')\nplt.ylabel('Count')\nplt.savefig('images_per_study_distribution.png')\nplt.show()\nplt.close()\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-26T22:38:58.166353Z","iopub.execute_input":"2025-04-26T22:38:58.166675Z","iopub.status.idle":"2025-04-26T22:38:58.561153Z","shell.execute_reply.started":"2025-04-26T22:38:58.166646Z","shell.execute_reply":"2025-04-26T22:38:58.560438Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 720x432 with 1 Axes>","image/png":"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\n"},"metadata":{}}],"execution_count":6},{"cell_type":"code","source":"def convert(dicom_path, is_train=False):\n    base_path = \"../input/rsna-str-pulmonary-embolism-detection/\"\n    dicom_path = os.path.join(base_path, \"train\" if is_train else \"test\", dicom_path)\n    series = os.listdir(dicom_path)[0]\n    instance = os.listdir(os.path.join(dicom_path, series))\n    stack_image = np.zeros(shape=(len(instance),512,512,3), dtype=np.uint8)\n    stack_zpos = np.zeros(shape=(len(instance),))\n    for i, ins in enumerate(instance):\n        f = pydicom.dcmread(os.path.join(dicom_path, series, ins))\n        image = f.pixel_array\n        image = image.astype(np.int16)\n        image[image <= -1000] = 0\n        _, _, intercept, slope = get_windowing(f)\n        image = image * slope + intercept\n        image_c1 = window(image, -600, 1500)\n        image_c2 = window(image, 100, 700)\n        image_c3 = window(image, 40, 400)\n        image = np.stack([image_c1, image_c2, image_c3], axis=-1)\n        image = image[:,:,::-1]\n        stack_image[i,...] = image\n        z_pos = f.ImagePositionPatient[-1]\n        stack_zpos[i] = z_pos\n    stack_image = stack_image[np.argsort(stack_zpos)]\n    instance = np.array(instance)[np.argsort(stack_zpos)]\n    return stack_image, instance\n\ndef get_first_of_dicom_field_as_int(x):\n    if type(x) == pydicom.multival.MultiValue: return int(x[0])\n    else: return int(x)\n\ndef get_windowing(data):\n    dicom_fields = [data[('0028','1050')].value, #window center\n                    data[('0028','1051')].value, #window width\n                    data[('0028','1052')].value, #intercept\n                    data[('0028','1053')].value] #slope\n    return [get_first_of_dicom_field_as_int(x) for x in dicom_fields]\n\ndef window(img, WL=50, WW=350):\n    upper, lower = WL+WW//2, WL-WW//2\n    X = np.clip(img.copy(), lower, upper)\n    X = X - np.min(X)\n    X = X / np.max(X)\n    X = (X*255.0).astype('uint8')\n    return X","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-26T22:39:03.454032Z","iopub.execute_input":"2025-04-26T22:39:03.454373Z","iopub.status.idle":"2025-04-26T22:39:03.469222Z","shell.execute_reply.started":"2025-04-26T22:39:03.454347Z","shell.execute_reply":"2025-04-26T22:39:03.468172Z"}},"outputs":[],"execution_count":7},{"cell_type":"code","source":"# Select a sample study for visualization\nsample_study = train_studies['StudyInstanceUID'].iloc[0]\nsample_dicom_path = os.path.join(\"../input/rsna-str-pulmonary-embolism-detection/train\", sample_study)\ntry:\n    series = os.listdir(sample_dicom_path)[0]\n    sample_instance = os.listdir(os.path.join(sample_dicom_path, series))[0]\n    sample_dicom_file = os.path.join(sample_dicom_path, series, sample_instance)\n\n    # Read raw DICOM\n    f = pydicom.dcmread(sample_dicom_file)\n    raw_image = f.pixel_array\n\n    # Plot raw image\n    plt.figure(figsize=(6, 6))\n    plt.imshow(raw_image, cmap='gray')\n    plt.title('Raw DICOM Image')\n    plt.axis('off')\n    plt.savefig('raw_dicom_image.png')\n    plt.show()\n    plt.close()\n\n    # Process image\n    stack_image, _ = convert(sample_study, is_train=True)\n    processed_image = stack_image[0]  # First slice, RGB\n\n    # Plot processed image (first channel)\n    plt.figure(figsize=(6, 6))\n    plt.imshow(processed_image[:, :, 0], cmap='gray')\n    plt.title('Processed DICOM Image (Channel 1)')\n    plt.axis('off')\n    plt.savefig('processed_dicom_image.png')\n    plt.show()\n    plt.close()\nexcept Exception as e:\n    print(f\"Warning: Failed to visualize DICOM for study {sample_study}: {e}\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-26T22:59:23.083141Z","iopub.execute_input":"2025-04-26T22:59:23.083498Z","iopub.status.idle":"2025-04-26T22:59:26.047409Z","shell.execute_reply.started":"2025-04-26T22:59:23.083459Z","shell.execute_reply":"2025-04-26T22:59:26.046593Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 432x432 with 1 Axes>","image/png":"iVBORw0KGgoAAAANSUhEUgAAAVQAAAFlCAYAAAC0tBC9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nOy9eYxsV3U1vquqq6q7unp4kw0EQxgfg7EdgcNz4MNAEpGgEAYHBzE4IQMZQCGOkZltI5AAK4CSQEhM+COxgADBmBAyEBAJIdhmiGWQgjAQMASwMTz7+b3Xc3f9/nha16tWrX3u7Wd/8PFTHanVVfeee84+++y99jr73HurNRqNRjEt0zIt0zItd7m0f9QCTMu0TMu0/P+lTAF1WqZlWqblbipTQJ2WaZmWabmbyhRQp2VapmVa7qYyBdRpmZZpmZa7qUwBdVqmZVqm5W4qU0CdlmmZlmm5m8oUUH8Myste9rI4ePBgHDx4MB760IfG4x73uLj44ovjlltu+aHL8sQnPrGS5fTTT4/HPvax8Ru/8Rvx/ve/P7a2tsbqPu95z4tXvvKVY8duu+22uPzyy+NJT3pSPOIRj4hzzjknnvOc58TVV189dv33v//9eO1rXxtPfOIT4/TTT49Dhw7FH/zBH8SXvvSlsfauuuqqOHjwYDzmMY+Jzc3NsXOHDx+O008/PQ4ePBif+9zn0jFdddVV8bCHPexkVTIt01KVKaD+mJRHPepR8alPfSr+7d/+Ld70pjfFl770pXjxi1/8I5Hlt3/7t+NTn/pUfOxjH4u3v/3tcejQoXjjG98YF1xwQayurqbX3XzzzfGMZzwjPvrRj8YLX/jC+OAHPxjvec974ld+5Vfine98Z3zlK1+JiIjvfve7cd5558X1118fl156aXz0ox+NK664ImZmZuL888+PT37yk2Ptdjqd6HQ68YlPfGLs+Ac+8IE45ZRT7n4FTMu0JGUKqD8mpdvtxoEDB+LUU0+Ns88+O84///y4/vrr49ixY1Wdt7zlLfGLv/iLceaZZ8a5554bl1xySRw9erQ6f+6558b73//+6vtLX/rSOHjwYNx0003VsSc84Qnxrne9qyjLYDCIAwcOxD3ucY94xCMeES94wQviyiuvjBtuuCHe+c53ptdddtllsbGxER/84Afjl3/5l+OBD3xg/ORP/mQ8/elPj6uuuirue9/7RkTEa17zmtja2oq/+Zu/iXPPPTfuda97xRlnnBFvfvOb49ChQ/Hyl7881tbWxto+77zz4n3ve1/1fTQaxd/93d/FeeedV6PZyQLGeu2118ZTnvKUOOOMM+K5z31u3HLLLfHZz342nva0p8VZZ50Vv/7rvz62SvjWt74VL3rRi+Kxj31snHnmmfGUpzwlrr766rG219bW4tWvfnU88pGPjLPPPjsuu+yyeNOb3hQ///M/P1bvIx/5SDz1qU+NRzziEfHEJz4xXv/618fKysquxzItP9wyBdQfw3LLLbfEv/zLv0Sn04l2+84p7Pf78drXvjY+8pGPxBve8Ib4zGc+E6973euq849+9KPjmmuuqb5fd911sXfv3urYN7/5zfjOd74Thw4d2rVMD33oQ+Oxj31s/PM//7M9f/vtt8e///u/x3Oe85xYWFiYON/tdmMwGMSRI0eqesPhcKLe7/zO78T3v//9+M///M+x48985jPjmmuuiW9/+9sREXHttdfG4cOH4xd+4Rd2PZaIiJ2dnXjb294Wr3vd6+I973lPfO9734sLL7ww/vRP/zQuu+yyePe73x0333xzvP71r6+uWVlZiXPOOSf+6q/+Kj784Q/H+eefH694xSvi2muvrer88R//cXz84x+Pyy+/PN773vfGwsJCvPvd7x7r+6qrrorLLrssnv/858c//uM/xhvf+Mb49Kc/HZdeeulJjWVafnhlCqg/JuUzn/lM/NRP/VSceeaZ8bjHPS4++9nPxq/92q/FYDCo6vz+7/9+POpRj4p73/vecc4558RFF10UH/nIR2JnZyciIg4dOlQ59ze+8Y247bbb4rnPfW4FqNdee20cOHAgHvCAB5yUjA960IPiW9/6lj33zW9+M3Z2duKBD3xgsY2bbropdnZ24kEPepA9j+u//vWvjx2/173uFT/zMz8TH/jAByIi4n3ve1889alPjX6/v9thRMQJhvuKV7wizjzzzHj4wx8e559/fnz+85+Pl770pXHWWWfFwx72sHjWs541FqAOHjwYz3nOc+IhD3lI3Oc+94nnPe95ce6558Y//MM/RMQJwP3bv/3buPDCC+Nnf/Zn4/73v39cdNFFE/p+61vfGn/0R38UT3va0+K0006Ls88+Oy655JL4+7//+zhy5MhJjWdafjhl5kctwLQ0K2eccUa88Y1vjPX19finf/qn+PSnPz2RQ/3oRz8af/3Xfx033XRTHD9+PHZ2dmJzczNuvfXWOPXUU+Occ86JH/zgB3HjjTfGf/3Xf8UjH/nI+D//5//ElVdeGaPRKK699tqTYqcoo9EoWq1Wei4i0vNaLyul63/1V381Xvva18azn/3s+Nd//dcKXE+mtFqtePCDH1x9379/f0ScAE0+dvvtt8f29nZ0Op1YXV2Nt73tbfGJT3wibr311tjc3IyNjY149KMfHREngsrm5macddZZY32dddZZVf738OHD8e1vfzve8IY3xOWXX17VgV5uuummOOOMM056XNPyf7dMAfXHpMzOzlY5xgc/+MHxjW98I17zmtdUS84bbrghXvziF8cLXvCCuPjii2NxcTFuuOGGeOlLX1rtft/znveM+9znPnHNNdfE9ddfH4cOHYrTTz89tre348tf/nJcd911ceGFF560jF/5ylfitNNOs+fue9/7Rrvdjq985SsT+UJX78Ybb7T1sHF1v/vdb+Lc4x//+HjNa14TL3nJS+LhD394HDx4MP73f//3pMbSbrej0+lU3wHk3W534hjA7vLLL4+Pf/zj8bKXvSzuf//7x9zcXLzhDW8Yy3Pzda5gNfHKV76yAmIu97jHPU5qPNPywynTJf+PaXnRi14UV199dXzxi1+MiIjPf/7zsWfPnrjwwgvjzDPPjPvd735x8803T1x36NChuOaaa+Izn/lMHDp0KNrtdjzqUY+KK6+8Mr7//e+fNEP90pe+FJ/61KfiyU9+sj2/vLwcj3vc4+Jd73rX2EYZyubmZqysrFT13v3ud08AUUTEX/zFX8T+/fvjMY95zMS5mZmZOO+88+Kaa66JZz7zmSc1jrtSPve5z8VTnvKUePKTnxwPechD4rTTTotvfOMb1fn73Oc+0e124/rrrx+77oYbbqg+79+/P+55z3vG17/+9bjvfe878XeyKYxp+eGUKaD+mJYHPOAB8fjHPz7e/OY3R8QJxnb48OF4//vfH9/61rfi6quvntjsiDgBqP/xH/8RGxsb8fCHP7w6dvXVV8e9733vuPe9713b98rKStx6661x8803xxe/+MW44oor4oILLohHPvKR8fznPz+97tJLL42ZmZl4xjOeER/+8Ifjq1/9atx0003xoQ99KM4777zqboNLLrkk2u12XHDBBfHJT34yvvvd78YXvvCFuOiii+K6666L17/+9TE7O2v7eOELXxjXXHNNPP3pT68dx91d7ne/+8XHP/7x+MIXvhBf/epX49WvfnV873vfq84PBoN41rOeFX/yJ38Sn/jEJ+LrX/96vOUtb4mvfe1rY6z1D//wD+PKK6+MP//zP48bb7wx/ud//ic+9rGPxSWXXPJDH9O07K5Ml/w/xuW3fuu34tnPfnZcc8018YQnPCF+93d/N97ylrfEyspKnH322XHxxRfHRRddNHbNoUOHYnt7O84+++xqSXvo0KHY2tpqzE7f8Y53xDve8Y7odruxtLQUBw8ejIsvvjie8YxnjC2TtdzrXveKD37wg3HFFVfEW9/61vjOd74Tw+EwHvCAB8Rv/uZvVhtRP/ETPxEf+MAH4u1vf3tceumlceutt8b8/Hz89E//dLz3ve8t3oTf7XZj7969jcZxd5eXv/zl8apXvSouuOCCGA6Hcf7558eTnvSksY26l7zkJbG+vh4XXXRRtNvt+KVf+qV4+tOfPnYnwNOe9rQYDofxjne8I/7yL/8yOp1OnHbaacVUybT8v1Fa0zf2T8u0/GjLBRdcEEtLS/Fnf/ZnP2pRpuUulilDnZZp+SGWL3/5y/Hf//3fcdZZZ8Xm5mZ86EMfiuuuuy6uuOKKH7Vo03I3lClDnZZp+SGWG2+8MV71qlfF1772tdjZ2Yn73//+8Xu/93vxcz/3cz9q0ablbihTQJ2WaZmWabmbynSXf1qmZVqm5W4qU0CdlmmZlmm5m0pxU6rVao09DcL3yuF7q9WK0WiUfi89jpj1yVkIbc/VzfrktvjRR7SDp1LwvdVqxc7OTjpOnGu32zEajSbqOpm5T7Shx9GOPnnD9dCnZmi4zmg0quppwTG8TIXraBuqEze/pTbr9J/pFm1lWSinH9WLtocCGd2cZfag+sM8cNtsc9wPt6dt8mfIo/LrGHXcTn6WidvSwuPkOcC1bl5cP+ovrg31Zb1e5861o+ecfbg51L7Vflxbei7DP7UvlCJD1Q6zznlAzpkygbNzpbb5PA8yq+PayGTIwL8EOKUx6jEXlDQgcHEyu6Jtluo1acuBXp0cdfPVxKAz0HFAqrp0bbtzJfDJZM/G6cZQVzJQyc45H8zq6XjqAENl4vnlwMDtZbbBx9yclmzKnXey1NlhNraTBVMOZCVM41ILqE4hToi6gXGbziHctXUTPBqNxiKF1snGVDfhrtQZQem4O68TpHrJgLuOney2qLGxXCV2ode6c03Bp9Qv13Elc+qmAOmuyUDLfXY6KrE6Pb8bMM6YWZ1dZkDkbBr1S8TDta3tZ/bixuPsT2Vx12R969ia2gBfyyufpmAaUQOoTRqoq1NyLAWJOkPAdVnUUblYoaqULDqVxuPYZGYkOobsfFZXx1onYxagsra5nSbGn+k7c/K6/l3wyK6pAzs3lqZy8fUlHbm+nHOr3TnHdjJnIL0bHfExtfU6u3bXNmG3JZmyNku2kNWrAzglJk2JUqYDHOMg+X8NUEsG0LRkEb9JfTXEDOBKhu9krwOCrG4TgN9NyZxbv5fGXJKjzrmagJOTpSSHOoNrw313Yyx91mUan8tA1tlJqd8saDYFLXyvC4KOQTrZmrJJV7Lx1bXngnv2OQNNHVerdeceA9fTMWeAryBahxk8ViVfdQEsK7ve5W+q6KbHXb0sMrpSUnCpvmvftVViia7fEihnRlCSvW7sziCaljrndHK4cTvn4Tb1mNPBbuZQ+8s2COqCy25sMgscDvDqAobWz44pmPCGXTZfJQDWOnU6z/zE2ZsDQb1W5zkLsCxbpru6oFYifHUBVH2K9V9XTuq2qVIHmUNpnczxS2yspKSsHtfPgA+feRc/i6JZ/yoDvtdF+8xos/ZKY8xKHcMt1ctkrgONkn60ZPOS6U312wTAmpzL6jYFLy0ZK3KylAI36tbp1N3B4eTJ2igRg1IgcfPkxtakDxwvzb0LXnV+pu1mMmfA6oKZK7WbUu57E+EV5XkQdQClAy9FDSeDq1cCI8dkFDRKUVe/N3VwJxv+KwtoMplZqWNqWbslxjQa1d/Ok7XXFJSUobnzGRCdjAxZaRIYs/b1NrCS/vHHt/6UxpYBgNZtEqSz85ms2nbd7Xhcmrat15SAlHVV10ZdAFDZd+N/uwJUFnw3oFo6nrFYJ0sGuqV2SmBVAhF3fRNmmDGLLHjUyeJA1cmZXevYURMmwHUz4Oeixrlblti0jyYsKwu8WXF2WtIZ1+H53a0NO/ma6qHOdlU2bdPZRdZnRoBKoKO6OxkQZZypI2F6javTVPdOrqZBe1dLfscUXR2dwIyl3pVIoef1+hJrqgMJFwlLpQ5gtI4aeqYHNqYS862bj+x7E7BpUq8pYNc5lcpWCpJN5qVJsC7ZXlaajKmpA2Ztl0qdLkt1mgRQnHdAXOorA+pSf3cHe1VmWsfO+b/zIedv7It3C6CWjM0xrmwwdSVjrg40MoNp4kC8m4icKc6XoqIbT2a8Drx2o4c6I2zCmFXGpmyj1K9elwUC/q7Xl4y/Lghq3+4cO5nTUx3D0/FkMrs2SwzJOb4DoSYsO2OYfE3d2OsISql/bouf9ss2CZ0uOVXgSEeJ9Wb2yJ+zsWdz5YKCI4FZqb1tqkR7S4PVUoqImaCZEp0Ssn4cS2GFucl3k7mbSKrylthlxsqcE5f6rNO/9pUFhCbOloFuFjRKAScz/Kw45uTG5OQqOWRTxsbf3aOqDtxL7ZfqZfOI43UPEDhgyJhcaax8LJOnDvzdGNUHHanZTXB3x50MJV9yY9dxlXb7axmqPvtcBxD6vQR2WeR3kaoEBu6aJhPchOk1ieBZcWPPJjSr26SPEjBkjtrUCdy1+rkEMk2BEnUzx2gaTLSdkv3ttl20oXJmTNExZdcW97UbOy6xKTdndUDKdTKWqfXr2iqNQd+LoHV3E2Tx3+m4dA3LpbjmdM6rWVcaPXrKTwk4Qyoxkwg/OarkbNnNMrg2HFstRavMMLOJaMJUS4yp7np3vAlLycbnxpZdV3LCUls83pLBZjo9Gbmza92qxcnr2GlJjgyY6gBOS5Og5UiL9u3abOKDJSZWB951rLRONq6rwQVgWqfnksxuvDoe1k0WWBzwu7nOAiOXRkt+FUzPqRAuypSAh4VmhZTk0XoZCGnd3UY9138me+ZwpSWGOkMJ3Epg4oKJ6t61XXL4JjJk19axmJI+mhgu13WliS6bAAdf7+ZZ5dHbhzKAy/SOtt1btLQ9Rzy4nVJQ4nO8hHU+XwKb3eiNgWtnZ8emLEpzXocRzo9KYFrSmcMNyF0qjZf8zsBhQKpsFkKNrImjaHFgWefsTjm7ASstDFh1DE2P7VbWJuzHgUHWr8qNYyU56wBWDZXruADs5qAELk1BLrveHc+YRzaeJkE6c8Im+uZ264iB6i2zRy0MBqqH7Brtr2RHpWtGo9HEPbURk8/F1z2QoG1n53A+m0sNEqVXRZZkKJVGz/I748iMX5mSu971o0ZRiix1xuDqZv3pWF07Oj5tpwm7K01GCRRdP6UA0cRxS+yQ5cnYk5M/+14H3KX+cH02j6U2m8hWZ2vcv47HsTzHYLLAUyd7yaZKflkHdnXnHDNDHRew6nBA62obrDcXQDJZsyCaBSH1n90SO/bP0nWNcqhZFHURyCk5i2wlRqSRJgO1TGZ3XPtq0qYzhBKT2k2UVV2VGIdzHJ0bp/cmxlPSV8b69FzdXJYCTB1gZgEwC3JNgmhpvrPjpbErmO4mcNXJ6q7l651umoCTu55tsA5MXXsZ2aoDU26T+8+CbyZXxk7rfL5kg6UAq6Xxfaj6Zu5sUHWRVRWcAR3340C9xAR0IrLbW1g+nfRS1Nb+uF9XmhijlhLQq9G7ZRVfWwfGTk4FrFYr/9UAJ19ToNA+XCkFjszeSmNpWuoCTZ2Nuf6zOlldldnNaUn+JuPFnKod4Xhd4MwCQ6k/9mN8Z7vWvG5dP04varMZuJfaVzkj7uJtU07ozPjrFOuAsc6pnSO5QWaggVKXTHalKZN1IOtYpruemaSOWetk/WVyqf5Kxek3679U6liZ6xfnSwCGzw64SmCltlAaS10Qqzuf9e8+ZzrWYFey/92AbymQ6fES8SiNN7NblaNEqhQEs0ClfpLdzuTA1BV3Xslbk8DR6Fn+uqjrlOZKU6NzgKjXKciVwCWTSYEvM/ISKDmjUBlUbo78TRiaayc73xQ4mwYGHY/2kYFgxkabsGLn7HXMqAlzznSTjYeZkwJaBnquPbVpHR8ChQKNY0R1fbuxNSEyOn5n9w4YS6VESNx88VvedPwlkK8bk749TmVkENeSsfas1DLUDFRL4MLnS4PQickiVwnU6sC7icwsk8rujDdryxmefq4zRNdGVo8jeDauOrkdoGfgqS//5fYzduL6rdOPyqz9ZCBR0kFJBqeLrI0SELnrHAhlTLPOYUu24WxBgaKOVGSyu5LJWadX9Tc3hsyWnE3tVmdO1gj/Rv66VJorjXb56yJAdq7pYFx/TkkKtBmzyBzDOaPWqzvHbWm/JbZVKhmj41IXIJyzZlE5CxLanwNb7TsDHAcgWR1tr2nAKYG46qYUlNlZM3DToO9kYrkyIuKIgmNvrm0+txsn575cm9xXBljZ96x/dw2PVdmo6sbJVALTUsCuw5oS0eLzJeBGqQXUksOhQz1WJ3CTicjqcx0HYE36wPm6IFAnu/bL17jJLoFLNq5M3lJQKQFHE2d1usyCE87X2YFrV2UuOYDTYQmEMyfSdkajyfuk+bOOq+kcOvBxOuRx6PWlsdQVZ4N1wd0FKC7uhzJxXRYcnAyo436OO/Nhd4zbVd9pQmS4ZLZZsklXGm1KuaiqneKzXsf/VVgnvGvrZAwqYvIJkDoZnKO64qIWK7+OcWSOyW1noJQVN+lNI7qOh3Wgx+oCrI5F65QYCH8vgW+p7yYgm4Gy++5kLQUlN39Z0TlRUCnpr4654bhjb01ASEHPBZo6ECvp0j3D7+ZF23f6VzlczplLFvSUUGh/TcC6dlPKdeKEK7Wh9XVAOhi9vs6x6gAs6zurkx0rjS9TunNsd23TMZbkKAEGl7pzJWZQ0kWTcyWgymTabXDV+ecNH5wvAY/K4vwgG1edXbkxqD/UBds6R3dzxdeUfrqIwbYUdJoEluz7bmwos+nMnjC2zO8xvuy1gbg2G6PDo7H+0zPUSNaZFqcMjYZ14MXHs0ndjVM50C7J7BTZ5JpSKT2+q22qvrT9EkC7e3Pr+iuBvdOhylAKbiort5nd01qyrRITasIiStdlYOpkzACsNG+ZLK5+Kfjo8YzocJslsFQ5VQb3bgI3t5lcdaUu0Dm/Qf/ZPLh6XKfkhzpG137p2rvtV0/1mX5XJ5vI7Jhrz13jHNcpJTMmxx64r5Ij6rE6tpW1l7WJ/9k9t6V+6vouOX/J6dTQS/Pn5jMrGQiXxtEUXEpBsEkur05vdSCgfdfpV4+VxuFsOPMBbd/JmYGpBn13TttV8HPjy/y7DsBKPprdLaDnSnJDB01AGOUu/eqpYyOZEnCuSXRwrKYu8u92ArIbxF1/2gb3WQf2pVIKPDymEhPA9U3kLY1Dj7v2smvcedcv67z0cuSmemwyz02ZkmszW1m4oKkMkI/pNTpObafuex1jcvbCdp3Zk5On6XtRcU0JyOvGl/l1ZmuZPZaYpc4Hy6xzVEoLZGVXu/yliXSozsdUIL6nUft0ym5iNO67TnD2dqw6wHZ9u4inE1MXYLJglIGMji0ba3aNlkzfel2TINoECJ1uXTCq+15qz41Rr8mcxN0EnoGDA9qMAGS61bHzMbV/F2CzoKvjLgEOy8nF3deqbenYs/NcNIerz/RnNleStck1KmepOIxoUho9KVXHYBxrq2NWJTbjjCprR2VyBoo6LFt2q0ypH22P+3IvxiixC9dPnYG7koFZCZjc3Q91xqvR3DELBwhNDDxz9oxVlfTSRGdN6zcBdSYRri4+Z8ebghDXryMPEfm7LnRs2m4TEuXs1F3vrnHBgOXL2sDnDItcvSb25/ShsuzGphpvSo1G+csTnEB17CVjhpnBoW5mhE0cGFGQQVXBohQQtJTkKX0vMcHMaR1wZU7ShMHwd10tuGtUvyxHBjw6jqYBo0kQrRtf1kbGruqOZ4DpAnYme4lplQA106fWKcnAdXRs7BdcT+2plKZh21ZbdatC53NZ/05e5++6umDy4ABb+8lA3OkxK41yqKwQffww6zAzUDcRmKgmr/FydfR7E7ZVejtQiQVk48rGWecoKre7thR0SjJzu9qOAyLNGZUcKQPH0hupnNw6V+ooupKoM3DnmFmf3J9zLmVWzqH1nKurcrqg1KTtjDmVwFTnC9dlgbEJOchsjWXhvkr1SiTD9e/SidyW4pOOE/WbpP6yUqrT+MZ+CMLHMhBz4IvCUUPbxDWZ8CUQjfB3Gjjj0+N1AJXVcbLUvR/WyaUOoXpWPey26FzpOPl/9rnkYHW6LMnF/x0wlYoDkd0wihLrcYHftedsUB2UdeM2eRxw1wU/NyZ3J4ibb2eTmW9zGxlwO11lOnRg18THVTcO4LGJqMSgFGSdDdbpuVR2vcvvFJ+BlDMix0i07dLOYsnBnVJcfY5yTV7r54xUFc3fldG5N9lkhgOn0DfsuL6dYWFcGbiVmIDTaZ0TOYMv6aZUSoacAYWCYtZmKai5txw5x9PNFHVGt7xkNtRUzpKP6DV1PgMZuM3stiGVhc+rnzu59Rr+rqQnG6ebZwd6WZ8OdNWu6oKCC4iZfrSc9G1TzrCbRngnYAbOrj6u4f9N2neKrjOobIx6PpvskoG49pt8Z1myNt14FLSdDBkbciBZcuYS43FjK12rQdrV0c86T1mA4YCXsUftT9vOApjTEdsiP9VTB+YqA+vGjSljWiVbgS5cv5m/8zHux/1H29lby7Kx6bxneioFKtZJhg86btduHTlo9BMorpMsl4nP6gRcL5vcOiCuk/VkFMHXlUDFyZoZepPJb8oMIvIbrPW6zPn5c8ai6owKpUk+tSmTcGNw5/halRff3Q+9ufrqnCWgYHmcnbjrMsBx353unZ2UwFxlrfvdeL1OfVSJBx93G1faZgmsI+60H5eew2ftQ3WPYOTmri5guMBZms86HNDS+PV9OqkZQGaTou26CUF/7nlc990dRxvuHromhqYyuiclMiDRa1HHOX7ddTpx2mcTRsqGVwoAbo5Ydu2/rj0GiwzAue+IyTl3Y88APzufAZNe667LQFdtGufdzz47v8BY1VkdqLGMdWCh9dw4nd4dkJZ0nV3jfMIBkdMnl0x/fN5hj2uPQZcBXXWfjS+TsVRmmlbMog87Tvb7MCWgcO1xf3XPfGeTyW01KVlg0HbrGINeo+fcjeOur8xYd3Z2JpLunKNlo+XfAcsADQXtOmZScm60hfl3xQG801VJvrpgqjJre/qiZWePpYDrSEC73R4bD/rg/jQou/Y6nU6MRncyJ2frrFvUd0GFASN7t4MG+ybF2ZGeL9l+Br9q5AcAACAASURBVMIaAJoQBye/1nMPC2R9apt1tlGykyKgssNqYzwAOKMDVW6r9FMEdYDd9LNeW2rPnVNQcQrX9jloNAEuBaoSkDnw1eJkUplZRgdou9G9XqNtMojw+PAZ17jln/bFuqi7eV71VqcX1aHqX5kMZNXdZC3u1zzrWBAHP+dv3G+r1Yqtra0xGTWgudfj6fjcW7icLlSHqr/MLlwArbuXPdMVxuTkd8DNWKTzosFI+3HkxH13pRFDVVbkFK2gyqUOZLIomdUtgUPE5KNt6qAZ8GUglsmlxsDXurEqCGl9Hgef151/Psdtl+R2AYHrZ0EzAzEGG9QBc2q329HpdMaui5gMvKg/MzNTXQ9ZMG7+r4GH5UBfXA9t6RgVdFmfzP4gD2RQ1okxqR1sbW1NyKR7DjzvLjCpjrXfTqdT6ZL1i/+qB51zBSFns03AVEsJwHROHOhnbWlwcmxe75d3tzC6DVndiHPB1MniShFQFXj4vzooC6JLUBY2i24lIMiYVAbuJTDM2i9NaIlh1IGqjjOTgWVX3XJ76kQMem6+VH/aHgra2N7eto/lAmjYqfHZbQhxe6pfXKeA3Gq1YmZmxgYH1uv29vYE2DDb3d7eHpsjnANAujyaLpVZj2g/c2TIo8f1O+tBHdvJosFe503zzpgj7k/lYTBQMK4D01LJbNf5gPqMnsv80NkZrm2CAdl98Qr6OO/y4nW6qGWo2kjmoDpQNWaczybNCeuUq+cdoJQmpTRxTi41UK6jy4pMT03GkoG2GhHn7Zxz6fXOSR1DYHbJc8dz7CI+B1A2TrTDLA/HGITd+FgfzPAUUFm3DKhqcwyo+M7jyXSiulSdZnXULhw4Zo6tMvF5lt0BGAdcPu/8shTwFWB4Xliv7nMWtFUWp8OsTW5Pz2nwV7l1vFm/Th8Ksq5/LY03pbLBsdDamTueRWw+7zZZ1Dhdnor/ZzIrW1JZ3HWuDTX4JnqqK9oXlnU4vr29bR0eyxsGNpaR21Ugi8gT9loAjp1Op2KS+NzpdOxxfN7Z2YmZmZkJZtpqtSpGCRn0OM6h4DiDJ+sA3/m40xmW5gAqgDX3yeDLgAudbW9vj8mP+mDmmSNnYMrz4AK7tuPytJAD9uPGrsUBRgZqmX2ov7t8qRt7xgIzwNfvrLPSPdZurwLnFYCdnupIUUSDTaksIrsB4bhTUBYlVMlZHXV0jfglUOfPbPSlNt13ru/ux8z05yIfPvN3XUqzvGwMvORFP5zL43aVZWZsSdlNu92OmZmZCgR7vV70er3odDrR7XZjdna2qs/1GLB1A0plZsN1QJLd88r6c4Cic+iCMW/qoDAo69KeAYLrbm1tVefRJo7jP/4AsgiMPGdsAy6to2PIgI3lVEBzLBM6cMth1b2ytSy/rrqqA1Nc48A0w4oS0eHjzhZ2C6Z1ZG1MV6MChcoYmDptE+Tma1n4EoA2LQpImaFxFOP6TcAR50rRWftWI1Ej0M/YnGFn5g0pPsag6+6sUJbnWA8+YxOp0+nE7OxsdLvdCkAZRMFEAaDMwlwagnWdLclZZyyPLplROp3OGItE0Q1RNx+sB2fDuN5tXjkbUyB0gAzA3dzcrEB3e3s7tra2YnNzM7a2tmJjY6P6zGNj3SGIsowKnpltsh45ZcTsnVm/zmHWlgY0vtaBrGvPnWOdOoB1x93YXVAqBSr8z8ZfF8wiagA1Y6c6uCzv5IBRJ4QnJQPWEpBxm6xonZysLY1IjkGxAyqzUj1l7MkV3rHl6wFUTl8lY1eZZmZmKqfBXKGvmZmZCiRnZ2ej3+9Hr9eLubm56jjLxuDCG07KbBgM2WkZcFmnDGCqX94gc4GJAQXysuNocGOdcz9bW1tVQCnt4NYFeoyd88Y8L8z+kSqADjc2Niomu7m5WX3f3Nys/tCOsmcFeQVXN3cKTHyNborxuDMdqG3qcdaHY6YlneMax1Y14PFcOFvQuk38S9ks+7aW2l1+N0DuuKT8UimBQlNQdcpRxqHXNIk+GbvhPrR/VbJjP9wmL+3YCdwS1hUH9nrdaDSqQHVmZiZ6vV4MBoPo9/sxOzsbs7OzMTMzE/1+fyJfGxFjDJSZkHNgx3pcfZZRwZTrO2MG4+aVxWg0Gtu95/ZRl+eGZdE7GjQXysHE5W2VFWN8AD/dRef5jrgzF91ut2N2draSHQwWf2traxWobmxsxMbGRqyvr1dpBe6LVzQ8F2qPbDccTBzT1RVClpPm+gp0jnRkWOHAkc9lpEn7LRG6OqBW++Pxlfxy15tSGcKrsBlrcwCXfS9FpKxwPdd2BkQRkwzX9emuU1B1qRLWCTNSBi0tKkNpvCh8LyeAs9frxfz8fPR6vQpIwVZRNJ+XLSFdekQBk3XgmKm2w/rh726XW4FDHcLVxWcNoCyrAjnXcWCq42S5maVqygPgzIGVby3DMczdzs5ODIfDiIgxgF1bW4v19fVYX18fSxdg9QFgzpijHsPYOe1U8l/nhy4Iom1th9uva8/JzLK41UiGPzy2jDiVwLSOLDa6DzUzXB6k5l64DRcZSwPPwIX7jPA5UzdJjpFqHR2zfi7pp8l5jJNZjW4kObkyZq7f4QxYus/Ozsb8/HwMBoPo9XrR7/cts2y1WpWj8rIZIAJgAJvj5T4ctwmThlHqvZIouisNsGIWyXW57ZJ9Mhvme2m1rjJl7kfrOdbNxxAwoV/Oi7rxaCDjgIs+sfnXarViOByOgevW1lasrq7G6upq9Z31B1vLVj/MasGa9RqVV/WbgSl/V390Rf0lI29cx12r9XisTk6VQQMRj6+UFmrEUGFEmdNzxxrFFRgy8ATYZBNYUrBOmsu7oTgGoorLmDG36WTKQJ/BlBmKy7GqYeIzMxxus9PpRL/fj36/H3Nzc7GwsFCx0m63OxGwGMhxHBsnpTnOWLfKisLAkBmgMkKdK2a3GfPnJb0CrBadu+xc5vCOqfA84jgHESzpFdD4Wr7ebbhExNimVKvVim63G91uN/r9fhUUkQ5YWVmJ1dXVWF9fH8vNKnA7eQD47kk3fIeMGZiV7EL1VMcOXfvuuDtfB7ZubCXZIspPSUU0uG0K/xUQVTgVyIFnk8E6BsvFRbysPZUlY75gEXWG4AA8c1IGIvfyC6dTDQYMJqgDpjIzMxODwSBmZ2djYWGhyo0CRNGeLlHZabQwi1N9MyjoBo4yP8158fXb29vV9dyPy03y/Oln1GPw4pynm4fSMW5fzzFr5u84VtIl61D7Ued07XC/ESeCHzN3tNXpdGJ+fj7m5+djOBzG5uZmrK2txfHjx2NlZWUs57qzszP2We2WiQl8Q1MsGbHRcyUMcNezDNyem1O+VvvTtktF5XV9Nm1r1zlUNzAGNT3Gx0vRKGMGbrA4rorNzml7Ohm8WcDHFIhZTrdbzNGb242489YXzdOCZSq4uv9gJOw4WMp3u93K0bApwrIqkKrMGTCw3pRJwdl4fkpzp4WPKVix3hU4Mzmdk+MzM2UOEmDn3AfPO4CR86gl5uXsjsfBY0UpsX8tyvjVHlutO+/gGAwGsbS0FOvr67G6ulox19XV1bG548+6EmK/yMCNZVPb5jZLYOrGw3054qJAnAVfR7YyvCmdy9gzl0Y39mfg5AbCQrtzOimOpTk2yUytLnJljCAbY0k+HXsWGZnZ4DNYGJZb3Aaf56JjBJAiJzocDmN+fr7amY+4c2OLUwkMCJq01+LGx2NUJ+a6Ls/JTJeZGorWZ4DTvlUmPoY62fsHOAiwDNkcQnaWJQNv9O/YqtqM2wF3/WTg7q7VIAk74/F1u93odDoxGAyqdNDa2locO3Ysjh8/Hqurq2MPHOBaZwPoQwNaiZSUAK2uKLnJNjEze2Vg5u91/XO7sPusH1ca3TaVgaqbbDUEd8wJr3V5MDpBLJsWB84OULXdktwuOnJ/yjqxuQJWyk6tDqoMDwwDN9EDRIfDYczOzk5s3MCpSk7LcjNb0CV85gAlXUeMMy8HbFw4H8/96d0AkHNmZmZs7hwQ67hRHOvVa9RZ+XvJbnWM+sgqZCkxeMdYNW8c4d9/Cl0rY0ZKpdVqVSsVBBM8sMHgurKyMgauWW7V2X8G9nxOdchjbkJU3BxzO6wPldfpUksJKxw+OZm41C75GTAytqadZQiu12qb6mQlBlwaVLYL7CInR90sKNSNl/sA4CnARowzD2aTvDvc7/djMBjE/Px8zM3NVbv0WNLjZm8urEeWH47IO/SOsTLTYmN0m1TsEOrw7ORcn9l7Vkpz7OyuZORcn8fNQYXbcPlc1oeOWwFaN/k0eGlbmtd2II6iO/4ZQCuT5Dp8WxbGhiCNTaxjx45VzHVzc9PeGYD/JXLiivp0Vlfn0tXnY1mggx5KYKrtNJFNccSVxm+byhgd19FzWX09pzTbCd900tR4s7E4B876ceccs+FlD851u91qPGCscEJc1+v1qk2mhYWFyti73W51zWg0GjN0Z3QaMLg4sEP/On9ueaV5O9c+t8l6djlDnNc3R7mgpnaA/jIbBOvOcq+chsicWIuTgcedMTe+1uV/NT+resra1aWsW4ozuGpgRd+4Q2Q4HFageuzYseouAegfQUNl0UDJduiCYEbMHOg6IufA1LFMliULviVZ3ViblEY5VP2sRp917JheHaBx+yVQ1b6zvrRuE9bq2isBLkf+TD4YJANbp9OJXq8XS0tL1UYTnl5iMN3Z2ameBef8ZKYDzVEy+LucJuvCOT3GBOfnvJ8CqcqiOmNGpnI4ZsdL2CzQseMpeON6DrSa52SdOgbOjJQZI9fJSomV6xg0B8nndWwK6rrZiGuyAAm7YnBdWlqK4XAYq6urcezYsThy5EisrKxUT2kxuKKvzE+4NAk2+KxBwOEOt6XyOB2WipPNgX7m11p29eipDs4dLw2kKcV2S01lyiUGqn3huwPkUjRz41cdMDjiOLNUvf+v3W6PPQa6sLAQy8vLVX4UBsI79dnSUXWmzg4Woy8TyYKKOp4CHPeL9nVJ6pL4PB+OoamOuSj4ucDmgN0xGH5YwQFXKafJdZQp8vyoXK7obXQuJZBt/PC8csFxvp2Kx6JpDd7MQtBBIO92uzE3Nxdzc3Nx9OjROHLkyBhjVb1maYssCGYkjf0msx/97Px7t4BYd4z1zwDuSqPbpiAs0/6MnWasCUVpvCq9BMxZdGDlYWIdsOoyic/rOHjcKhfLrktLGATnSNE3GOn8/HwsLCzE4uJiDAaD6Ha7ETGeV+WNG3VUpzeX89M6mj+FrhgcmaWy8eimFTu8A+9er1ed03qsaxckHBA5R3Q2ouPmOyF4HMyQ9RFcXkWU2K+bH7Uftj3VhZPHXa8BgfWl9Rl0VZ86/5oGwsta2u0Tt+EtLy9Xdwjccccdcccdd1RvxoJ9crBlv3BzlxEyli9rQ9vTecjad4DOQdoRKS1oXzHQlUaAquDn2EJp0I6ZaruuPb7GMUYXqfT6rP/SdSgMak4GVjSDp0ZZvP5ubm4uhsNhLC8vVzlSvpbvC2T9qGwOUOoYuOqNUwD4nzEqBgwH1K4fJ6/mLZnNZvOXsTV1CL5W9cZyc3H24ZbVfFxtk69npsZsK2PhylRZDwp2+qN8KNynmxNNdbA+dOwafGG3ANbBYBBHjhyJY8eOjb0BS9vM9IvPWWDM8EA/a9G5dvbg6nIpbVyxDCU5InZ5Y38d+9TjToBSBHKDdwpywKGRhq93TqvX8v9sUvk6ZXCjkc9LYRmFpf3i4mLFSMGGtH+nM/x3y1Ktx86M5/N1zBnD47E5h0A/bumpxqd54xLrQFEg4XyrvrPT2REvZTM7cRtJuNbJgz5dUGVds8wuAHHum5+X5764P2XvDMwKyiovSqvVmmDnJf9UgIU9zM3NVWmq22+/vcqv8huvcL0GiJJfafBjX3VEhtuHnriUgiuOcd3MB+oITFYaPXqaMcS6a1Rox7z4GseQHJusA3IubklZYnHKZN0Y+JyyUmZjvV4vhsNhLC4uVst7PLKJV7SxjDppWW6OnYTv08xevOHYCIBAWVKme6dX1i30AQDkNx2xLFxPj2eMRMGV84W8YVW3QYTCuWq3PHY547oAw+kBfSw0A3fNoeIzn2cwRl9qny4QqB7dphX6UPvieYJNg60uLi5WLx4HW8X7AhSIuZT8SevweR1rUzDlNnVMKgPXU7lZjhJOoRQBlZcSGiFK4Fqi3KU62m7WRylSMFjgu0YcNXSMVZWnSnbK1Ujeat35ZNPi4mLs2bOnWt5HlH8Xip0SMrFzl5aPWTBxzJL747paB283Yr25+XcrCT6nBuj0yayez2kfkBVBS8FUdeh0xOdYn8z2HKvmPCc+81y6wOvAISImgqmyan1MWcfBJCdLw0BOl+PXwnOLwsyW9Ts3Nzf2YvKjR4/GyspKbGxsjM2Ps4ds85CZvhurXqvyOnB049O6OJfpmr9n13Op3eXnpHzWeWkgetwpio+XjJ/bc5FQDdcZcnZcP7MxOpbF90/yBka32435+flYXl6OpaWlmJubq+qr8+Fax0icI6vRgaVFRPVWIx4j+lVAY6anGzLq4DpunMvmhMfEAUHfI6Ds0LEaPabAwJspep51zPpwgRx64gcwcJ7l0DQPt8Pjc8vwLFBowXn3qCx/5vmDL+l7VR375Jyr06lujnHQwotZcBx3p8zOzsbhw4fj+PHj1aOsvLGqgJixV80DO0Zbqs/XuLlm/To8c0Qxs52sFAGVQUMF4g4yRygJoddmA+b26q7FOVZABtzaZ+Zs7pyOCUDa6/ViYWEh9u7dW724JCJifX19Ij+mRTcWlEkAQLKdUGUWMBKXTlCw0/lzSynH2HReFJjAIiGPC1Cuf3fc6d/NQ9YGn2O70ftUm9oZ9+k2snA9AzGOYV7U+d28lsavaQXON0eMs322PZdKUl3yfKvtgkF3Op2Krc7MzMRtt90WR48erVIA2h7GqjaTganDD9VjxkadPp3dKEnJ+nHnXGm8KZUBKSM7jEdfttAE2bU4kObjjoI7RTqn0EgEY3QBwrXHY221WhUr3bNnT+zZsydmZ2fHmIwDUwVLxybY+JnN6bi5nr4pXhkgcq4cmXXMaI+P8b2bkJFfMp3lLxXIHKjrvaysF22DbQ2bbrokZaDkz+pIEePMnvtgmV3Rhw3QL9uI9sX+wvPCDu1emKOBE/2ovbMcqkMtrHu3EcZ2qukh1VG32439+/dHv9+PmZmZuOOOO6qfbeH8rubNMzDLbEZLpm8F7wwUHV6gnVIq4aQBVSNJ3SCziKDXO9rNA0S0ckzJMVcGyUyBLHdJIa6efocztdsnbtJfWlqKPXv2xPLycvT7/WpTBgw/051bBuKzY7OlhLyCqt7greNhWQCM/MZ+Lspw3BxwDhgAxvc1oj8XyFi/DOZguCyrApeCFQOCS19gd53TEmoPLqBkwKAMKyMCOq9Oz9kx9KNpGswdp340SKo8DBjusVf2QdUj2y70w9cvLi5WL9o5fPhwdVzfgsWpLdaP4oAyVw5CLE9JZyUwVZ2gZKmHJmXX70PVDtRh1ahUYRHjN/c7ZfDS1rGoDFQd21K5NaK7ZQCPhb9z/gjL/OXl5Thw4EDMz89Xhs3LHV36cN+8PI6YzGE6B+LiABZt8+4//1d9ZYHLsQZmKwA71g23yTossVfuG3MN1otzrEOet9HoxPsNWLaMZSBowK4c4+f/CrTaJtixY+c8xzpHpdca6ne0lR3LAoMLiBHjNqEgXGKJGD/LrblVHIMfdLvdOHz4cKysrEzI7Np1uXSX3nK6LhUebyng8XcHpipjVhrfNlUSIGOkTngFXXU+N7iMETv2pcyiLhrVTQgHAX7eezAYxPLycuzbty8Gg0HVN7/7lBmXtgd24JgZG4EalbIwHp/TLeTi9hig1SlQlBm775DbOQQvqZ0hs9wlu2GWq+wODq73dDJQ87hcLlAdVnWqMuM/AIUZNPpjXejmEsvAxdVzY8U4uK5jefqbYCABqsvsM/eDlYs+eKJjgW0hrwpZjh8/Xtkx24IbpwZlHmMJA/h6Pt4EQB2+qX0qXmVlV09K8eCygWSCZ+BaAksYkUYJx7ycfNxutlxQZqtjURY1MzMTi4uLsX///lhaWqoesYTjs+FnjCNj0GrwDATqcCqrc1KX+1IdcD03p7zU5SUiFw4I0BeAVFME3K/Kw+f0lwfgvABQHrO275gt10c/bD8OaHTuOTeHoiDDt5rxW/ChO+jKMSaWxwUxLZyD1Pwk6wHnWR+8omCdMGtnm3ApnywXjIdZTjnllOj3+3HLLbfE0aNHx1ZkWWBReUogqoCsIFpHllDX6cG14QimlkY5VOcQDIhOMZkyuKiAzrG0vRKbddGJo50CkQKpjg2Gj/Z6vV7FSpEvQuTljTgdXynB7ww20xUzQtePHuMb2NGHCyzaB+tNUw/M0t0GBqdFsntuuV9lJQzYqi8HoC5o6jKbGTRfh7Y5YOky2C2FWXYFGme3CAC85NZ5qWNg+i4FzekyiDs9oGRBVnPZGnA0mGS+zakX3FqFcscdd6TzWfJt7q9E5BwwlmTV9vBdr2sCpCi73uXnTp3yM0fBOfc5O6YDczS+tBxw53UcWYQDYMA4+v1+7N27N0455ZSYn5+PiBMGASAtsWwUNWS3o8ssT6/h9rJclLJO9MNsSdkJ5NCNLPTPYKEMRh3NMeZsBcM5T70zhNtw+UaAJN/ap08ouUctFWw1yCjjUr2rDiEbAwLPieqO+0dAxvGZmZlKD+4tXmhLVwmqnzqCwysoTnE4H+HvJYDl8xwIsc+A8d9+++2VvBi33j/r2m0CZkqcnOzaZnaN1m+SP43YxdumnNARnvVkJQM1tMP1HEjzuax9ZaconI/UsTGwKpAhX3rgwIHYu3dvzM3NVQYFw85uLOcljtt1R2Gnwnd1KI7i7BAAAReIHBCVNmI0SKIos2L5ud7JbKroccwFB5KMtaN/Xgk4e3Jycbsud4gxsyyQjdMOLIOuHtRO9QUn/Ke2UVrus9yZz2n6QuuxfdURFE1h8Xdcy+kEBlvoj5nqbbfdFhsbG9WKh69BkHU5bgd0et7VzbCEx+zadySgDtwbM1QWWAVlITIa7gzOOYK20yR6ZOdUZq3HcvHyHiwAL4PYv39/7Nu3r8qX8nPLLqqyMbsnlXSMypRKY3NLFMfgsVHDzs565D6dg/B1Ol8K3qxHFPSHZb/aim5SKIvMGJ3OMdsUOyHaqgNTyMVMU3XvigYuBVlXT4NXVnSeWq3xN3WpHvV2N37iS4mFO65BjG1Wb6vS7zz3Lm3FjBugurW1FbfffvuE/7tcPuvR4QynD0rBXll4BpZcn8fE190tgOqASc/xd/znJaMDQnagbIA8mCwqN4lO+K6GzWADR+90OrG4uBinnHJK9YYoZoc61iYFxo/xKmtV8HKsU+V1OkBh0ADIog7f0K23P0FWzhFz4Rv6Wa/o08nASzu+RudY9YB0BbMZbAC6+iwL5rPJ7UkcTBTIMVdYkkNe2PfW1la0Wq2xOwoYLLgdff6f7ZX1z2w9IsbuZGD94pcdME8MBAww+OzGrjatDy1ofhnXZEHaERfINRwO4x73uEdERNx+++2xublZLf/5ljJHvNRG2F8daYJ+M1JSYrDatsM3V3b1tiktji3ptS5nV+pPFcDnSuf1u1OSmxAHwvhZkgMHDoyBKZY1nPvT/hXw3dJV86RODxH+55EZENFWacx8W48GExSXY+U503HhBwPRht6io+kLTo8o6LvlpOsb55hF6tKX5xBAl7F+1IWOAVhcmGWzrCVWpO8E0DwvdMHgqWNBuxx8uX9lxwxYfJsU+tKUHNuRpjrURsB+9Y4JJko6l25VhqDTbp94AAB2gY0qTnPxXLLds5547KoTBVMlWFpH/zsAx/c6DDvpn0CpAzc28oyBZsDJfbtzKlsmAxuOgg0bBCfRe71e7Nu3Lw4cOBALCwsTifMsWnIbulxw+VnWneaPYOi6LOV+s00fZgvsYLiGGZQuod39ntwmgzTrkhkuOwL05kBfndnNGz6D/Tkmqg6jfek1apfMdKAXXf5m4MP2A71wPw4QAc4ISqxXfW8pn9NxcVHWyfaG+WQ5tV3UQTvQtRIAnjsQDNTjt2c5lspMu9vtxr59+ypZjh49Wv1mFT80wK+AdHPtdOFwScE4s2+tz+ccKLuy6yelMoBSBurqKTvBtUrFUZfPO9rt+tO2uS0GNAZBlH6/H3v27IlTTz01hsNh5UT89hxcD7nY+XR5p+DkxohxqXPx2FXHjokqq2TGonJGTN4+g/bYMVTnrA/NoaFNzYPt7OxUS2F1Yi46R+6OA/SvfTAQ8fJUHYvB2fUF2XTJC10yW9XvXBQIeAWA5S1sAwyW+2cS4gKCC9Bsn2xnCozcPgdyZbE6L9CdBl2MU9M0CuA4jjbb7RMbVVtbW2NvqOJ3MGh6xLFKxQ7oTPHBjU/tm8fE/1X/JZba6FdP6xDcdeAEV4Ed6LqSDazpi3UdEMKpEBV7vV7s378/Dhw4YMHUyaAGr/X4s+YpR6Px25RYJ7qEVgdxcvBSjx2Gdc9PZ0VEbG5uToASjyXiBJPip2xURqcL5Iq5HWfceg2fY5amzsuf4ZwZg0HfkEmdnwEGrJKv4/Fp0TG5gIdrlUXC/jidwPrlPnn+nT41r8v2xOkFtidO7XD7/KCC2iLrWnXOsimR0LoIsDMzM7Fv375q7CsrK2O+yfc9u3GzXTiMyhgl6rO9lNins7ms1C75HVPMGnQArMczI9WokIG0MzKNVNqGRjxWYrt94oW5uMd0bm4uIiY3UdgRNSEPUHbLC35CRsfivjOz5WjNToBjDLKO9bGe9EkrjAPXK7Ap++dgwcFRr2e2hB9zU6apDEeP4TinC3QH3j1R5oA+0wc7qbvHlWXNwAbn+Tq0D/kYsHj57eTi5TIvvxnAUFdZpWOzEeNpHk5b8aaiMlZlkpnPOfBBtVzTPgAAIABJREFUn04GPg799Hq92Lt3b6yvr8fW1lZsbGxUOkD/jnFmZE8xi7/z/DAGKMCqf6At91nLXXo5iiK7AqYKr8dUGdqHA98MzB39R+Eox2++AZiCmQ4Gg7E6zPB0ycZLQ5WXo78LQMqYXQTFOZcDdSkG1Tlk0/QE/zAgM1YdB+fWOPhwHwyGOqfIibnilnEZ2+B6YMmQPdMBH+exAjSynXbIoTpw7Bp1FXDcf+2T9cfXI1DxUlx1jMI5WNTT9ybAjvU+Yg1uOI5gzYEM9ZEv1XnOSBYHSE2h4Drob25uLg4cOBBbW1vxgx/8oPq5atgeP4DAQc0BnxLAjCCwXnFM66AdLiVCGbGLZ/ldUWAt0WWOdmpwrBxHqzPmqX3qJDOoqaF2Oide4nDKKafEgQMHKmaqmyjKLtzOu0ZtlgnslA2SjTKTn0FTJx91GPSYJTIb5/7U+fVdohlw6BxnkZ1zgpr+yOYK86HHeVdZGSTrGzpGPXY8BkPWoeZnXTvaH+/es6OW2JOyTRR+ByszMG5PgUNv61JQ5LmHDFhWq/1wesG1yXPIY+WxqQ9Dbw5YVZe4bjQ6sbzvdrsxHA4rUD18+PBYHlrTFrrKRLvORl3fWp/HxbIp3tTlTyMaACorL0NnBVJ2NqdgHYi2rwNxQO0KAxH6Qn+YBBjn3NxctZvPTz9lhc9pPR6v3oLE8jP70aJGj+98OxGclPvEGNUx0ZaCH7fFOtJg4fLOWTBUPXGqRPPAasBO5wr+3B/norkoa+fgpTp0c4XcKbfrnlZztujkRH/chuYhSySEwR6riiwgaf+OqLC94Dh0xYDFPoRAon27Prgf9RUOKBqMUKfdPnE71dbWVqyvr8exY8fG0iPK6t1rMVkXPA+KRQq+ToeqN9euK41vmyqBpF6TRQ7Nw2HyNNpETO7KOUbHxUUj/LGDt9vtmJ2djQMHDowxU2awPEbNDbJ87BRsnCwvCudYXaTTexfVKNmw2bDc43tqDCX9cFG5oJNerzfG4ngsPLcsA/ej7zNFHbzUGixE75lFURakeVUGTRcQmMVr4HIAzGwbbelTUJxSQV/6FBFk5znPfIgDDI+P5VJ7c0td/q82qTbGc4JrVG/8Dll+NaXLjXNbqlMel9bd2tqqVkvLy8uxuroam5ubsba2NoYjwA9mqjp+blt90AVxF9TcHCmmZaXRjf1Z5xmSK6jWtckDUfDlgfGx0rVsMDyRGZjqkzcomkfSjR0ePy9DVBanC9RxzIiX8prz5Q0pTmdgfMxgnQ4dWKmzOnn4Os6tqnHqxo0GBix3dZkPPeAc54p1njW/zBtAEeP3USrj5zYYdDQwct6P+1Kn4jQHX8fzwHlq3qB0zs1yqN0ouKB/Df4sH9pwKwVHBLRv9WekAphQKC6w3mEDPE5+VBayYw7xcypbW1vxve99L9bX1yfGAj3z+EsMVb+rrGynri/HdrOyqxdM1zFTDNBFrqZFB6SMVg3CKY4NgVlDv98fW+a3Wq1qJ5oZCYru3DtwYnBhp+L2XLREPWZlmFgFIgAp/iuTV3boUgtsCAou+lkZvQKzbmQxeOM7OxyPEW3ij4MwM0mMV1mFjkGBnvvjejweDfjMCFWvAFMNXAqGPC/OeREkdIPG2TFfwwxPUxc6Ls0vci7b2Z4GNQR4tmXWsfqH5pE5sLFfoI6OTX1uc3NzLJ+6srKSkh3YGKfytE/HPHmsGvTVt0ttZOWkXo6CAbnGHdNRQR0wa7slAHWGy+24/Nvs7Gzs27cvTjnllBgMBmPgFTGelK9j28renCwMxlyfGSYDgbIa9MNvPmfZlLmojplplIKcgqGCBgO2OjHGx/cN8hzwEo0ZIbfJS252EpVXdcDOiOt4TnW5qcFNAy+PT22H54uBBWDLMmbBmfWG59d5DrNlKAMV2ufxsjyuDb4fGPK7e3x1bnR1wGkaXWFETN5SxnfKMDvGn4IxjzUiYn5+Pk499dTY2tqKo0ePTuAH31LogmQGfBrMHC7VgWapFAFVc2oOLCEY/jvwdChfAliNOKW+8TkDpFarVb0YGrdGcQRX+TIAclFe5eYIqQ7FgMGOCFn0uEZ0fvSSwYpBiwNExAnj5/yli+T6aCWW5S54MRvguyFcnpLBh8eSgRfPmXtSSXWsY2Kw0QCs9oE2lT2pLaCO5hx1/nm+ORBpsFJn1uUqj1Hl0fnlz86GGWh1DAxoDLQu0HF9HiMHSbZNtqOI8Y093vR0IIg6ESduC1taWoqVlZVYX1+PjY2Nqo7aHd+RABn4v44jIxZo0wVbV9eVRkt+FdRNZB2D4nNc1wEuF3WKbDAqCxTX7XZjcXExTj311FhYWIhWq1X9tAb34YwpU6p7qxDrwOUu+fYVZqkKbNw/AwS+M2i12+0JoHX3lWqeiMeBMfCySndQuXCQZWfnAMTO6PJtqnv9rgwSetDdYpx3u/hqf/zH+sJ5BXt1KgUwlt2BN4OakgEEOs13O0KC89i44TFwPQ0WLp8KHbk5cAQA9SPGV1yaE3Xgi+PcF+uC9ciyccHTi2tra/GDH/ygSs9x2y4A6pgz5urmjdvSunq9K7va5S+BH58r1dPlUCaoAyonn1MWM6OFhYU45ZRTKjBlANPdce3PyVQCBHZuboOf1OJzcAzkidzSzcmgT17xEzD6dJcDJ74TQSM7Mw7HJpm1sbPpsr5k2Bk7dbrVFATXU0B0qwpe4UB2dUoGfWVipcKBW4EiA2cnP9fTYIC54GDKekZ7eheI5l+Z2WIZzsy01Rp/L4Gu9tQ2WZdoC7aXgTGPN3uija/tdDoxOzsbe/fujbW1tbjjjjsmwM+RBSVsfNyBruKWA1Luq1RO6o39WYf6ucQ+uDQBT9cGD14dp9M58ZO2Bw4ciD179owtDWBEnJtSUEEbek9pxGR0dNdjXAw2Ol4+j6IA4gyPnQB1ePmozoN+lPGp0zDzgEHzcp7TEaPRaOwJnGzuGNQ4d8e604CmelRm6kCK54bHxeNVsNXrnAzZfZj8n2Vlp1Mn57Fm7wxwNgfday6b6+gxxwRdsOWSESE+5myfr+d2OcCpf6rtqP5Go1G1SbW8vBxra2uxtrYW6+vrY/UZB7L9DJZFCRh/1oCnOnDsXsuuf0a6CeDhWh6sKi9D+9IxbYejMEq73a4i2549e2J5eTm63e4Y6ACQVF7Xrx53+VVn1AzKmew4zrJo21pgjJkxcBtZ8AOYMlvlex6hx4jJTYhSMM3mumSsmROz/fB/BnaWB/Ipy3JMO5OZ50IDqDJ5LMFVz85mcI17pwO+6wZLKZfpNjwV6HTOcC0+IyfJL07n4OnmzelJA6Kba7VTx9Z5jDxO/F9eXo6VlZVq6c91Ud+1q/JwDtUFBgVcFBf0XdnVfaiuOAdzynKRLKvn2IfKopsTKHgKau/evbF37177pn1tyzmrTpIDFXZeZoA4zyxE6zpdaJKdx+pkdblKFL4JWxPtrANejjErZoDVdty8OeboHIjbceNzbBLy8XfWBR9DXe0PbTuHLwV91xaCGr+U2s0Jt8UrI67rQB7zwnpSG8Z4+N0SPHa1dw1Q3EbEnY9cOyDnujoX+Aybhj25F1Kjb063YYz6IiDIjeMLCwuxf//+WF1dHXuKSufO+TbrQcHU4ZJep22VcLHRo6dZgQBq7E5Q91mFdEzCRTt1Uo7svV4vFhcXY+/evTE7O1tNihpVibXguCqXI3OEX17iuGubjQ4sh41se3t77MfhnA4gkxq+7l6qg2vQcs+rKwiiXdaDLt10TlgO1Hfzyec5R+iMVQMCy8jBSFkgbNMxJU2tjEaj6namTK/aJy+huThCoEyZdeN8gnWjrFD9CSsNBSTNz2f2yJs9uA4BQwsHXLb3dvvETfl4Y5Qy/VLA03ngVAePZ3FxMfbt2xcbGxvVC1RwLXxIdcjzwXK4wMptoY7anc6rll3dNsWdlDp0kYOv10GUjusAsojTbp+4cX9hYWHs10l515tBRsHSbUw5WbSO/nQGGwtu5HZ5NXaOUg4VY9ClGPpnNsxjwk4y2mDHdDoEq+DNCh1zKTByW24DiW1DCwc81okDHTAgfT0e+uC5YSBWR3JMiOdQx6W2xjlrHGP2qXV1brmwo7JdqOMyg0QqRlkX53xZNt6A1dy4jmln584n9CAP26PT+czMTPR6veotY6xTftGJ0wPPA8uMAgDHBhW/4Z/1l313c89zy+PguVby4OxMy66X/DqBpXoqaJM2cQ0PzF3DpdU6cXsUwHQ4HEa73R57CsoxD/dkDMukBuD65YnH231wbnZ2NrrdbszMzMTm5uZE7oc3q5jVcpTm/tUBVF9sLMqs2NlxTHXtojPPizqrC5wMYA6AVUatp4CsLNexELTDdy+4MWjwxHjBTPlpNHfrmGMyPE+sHwYsdU7VtY4nY74IlOwXGqx5vJzOUf2D1cJ3WK+YE9YL2mNwZPvvdE68pH12djba7XasrKyMpVi4f/Y7+KcCt44bdebm5mLPnj1x/PjxWF1dHbM1xQoHim4+HanKwLMEphENlvyuYcd0mtTVQTM7KzFRfC8Z2dzcXCwvL8fS0tLEY3Von3ebXY4IDqDGrqxRr9GXdHQ6neh2u9Hr9WI0unMnH0asLIb70eUvrsUGiC6lVBf6nY2Fc7TqYDjO84H+I/LbXByosiM5A9S5VkYKudTA1fBdikADPgOutsksEmDB+smcxzEYZlYaMJzd8BjZD5D73tm58921GhxRR8kK5gr3uLJNMrA7Nod6GD+A3d3lwLrjudja2qpAle9UgMzQD8+bAzwN2BgTrl9aWoqjR49WP53CuMJyKnnQ4K1zx3VZr3qsRPAaM1SN9M6RXNTVdnSAulHAfdVNIoyg3+/H4uJiLC8vR6/Xq4CLNwJQt+QoCuhsaDphDHBod2ZmJvr9/tjz0AriGXPj5RZ0ycsz1hmDqhpeHYtUIOVXt7l6KDq30AunFdw8MkNywUhtR2XhdnSOeBxqEwwoHMAzYNMdeO2bA4/WccRCdcIrIufEqIPlLV4gzaxbgzwv+1UeBjENvroqiph8G1pETLBhDrBsp/CBTqdTLcWxf7G+vj72LgQFcpZDV1+aEoAu8OrN1dXVOHr06Jj8SlS0DZ0z1qnmzTnNoXOblcY/gcKdZwaqzp21CaH5OxendFYEA2Sv14vhcBj79u2beEk0szKdPLSjyyQ+zzKr8elz671er2IU3D87uwNAZqM8NqcPp8dMh8oE2CE1n+r6YyB05zJ5NPAoCDBYKnPUwOAAV+3GgTXqsWOyPWjgUMbJgMz65JLlQ1kHDAQuULHOODWlAKM+6GyUC6eS8CN46AN+o/Lqk3qwRU0FYCz6YACIxPHjx2N7e7tKd4FNuzExQKufYw44b4yCF6gsLi7G2traBFC7ecgCdDZnPGbFu7sEqAqq3HDWiRuAnnM5S1aIgjefa7XuvG1lMBjEnj17YjgcRsT4Sxoi8jcv6Xgi/GYBf+Z8Ef73+/2JW1fc25iw+6k7rgy8vHngdA0ZlUmUmBfOZ+CoAU7Hq0wvY7pcV5kh+nJAzHLysp9B0eWQcZ06jJM3A2Zniw5oM3thgHG6gl45qHObCjRuPh348xI+C2Alx0dfuqRnuRnQmKmh316vF/1+vxrz1tZWrK2tVfsE6+vr1UZVRIytADiAqB2w3lwd6By/RXX8+PE4duzYhH2xHlh+xQFHOErAC9my0njJX+ewjiW4tpR1qkHoxDpQhoHi5v3FxcUxFsJMI2sjYlIxjjXzZpbes4l80c7OTmxsbIzpQaP85ubmxJNZiPQsO39Xfals7JTK1By70TnM9OwcNGOl6vAcINy8gQ3hO+owIHJREFeHd/OnbXLqRRmXW146AGH5WQ5dUmIestvfHHip7zhfc/bA7fM4+ZamEvFh2dTmtOj849YopL42NjbGNvLATLFRi7bd6wHVVtmHVE5c2263YzgcxtLSUgXkzp4zcMyCZF2pq3dSOVQWUDcxnMDOWd3gNUHt8qsRUYHZwsJC7Nmzp9qldJPkAEoLswOOyLz80o2nwWBQ3ZrEO/icUwKYgpmifwUCd2uVskp1cJ0LHZ9zJJ0jXWarw3P7GJvToQM7/OexoHBAabVaY0tS9KP2pgxbGQXblrMxvl7tTlMCbqOE2TOvKFRH3IbrzwEit+NSGywD65DrQCZ+XR/0rnlFnNfroTfYMoOhpqK2t7fj+PHj0e/3o9/vj+V8wVh5xQU/4QDGutZUgM49F9TBY6nHjx+PI0eOVBuLmb61lDBLA5DKlJWT+tVTdO4Gz4KWhNHB8IQpS1CjnpmZifn5+bH7TbHjh3rcT5PIi8jOLzJhg4ADYanT7XYrINUX6+K6zc3NKnIrc1Xgc8nzJhPIzsC6ZL07tqnHWefuuAMzDoQ6phLj0uCggMLXOMDXYJ71EzG+1FQgda8pRD3e9ee2dH5wjFkYL+8VjDQFxasLBhV17uxZftUp+igtSxVgMjtjEpARJ/z8c7fbrf7gS2gDP2/Cb3lTnWRFfZmD5s7OTszOzlYsdWdnZyLd5trWMWCMij+qqyag2vhtU9n5bHJUEFdfo7Vjsno97+ojb6qA4vJs3BdPNi/D1dF40lutVvT7/Wrja3Nzs3paA3IhSuP3cPjtO3zzeqY3pw91ePc9It+EUlbj+uPjDI6aP3OsmdtC/7zRgPbRlgMO3mFmoOEHGtCPBgtlgSwXHIyfk1edsxPy2HgzRcEfdfSWHYAHAC17+Q4zSma9PA60hyCfkRcFAW3LpVAyVsasF983NzerlZimR6Bj9IOXmeDea+hDyZJ+VvZfKrgWt4bNzMzE0tJSHD9+fOw2quw6174jABl+lRhvxF3IoerkZo6eteXquWjA55E3XVpaiuXl5YlbJXTTImMwnGfS5T2O6UZGr9eLubm5aLfbVVTm63DPKR6L49+CYtDQzQcUXpaxLhkU1RBZPu6r3W6P3c/I488iPqc8WEbHCFmPGYPUsfDz2zoWdSQOEsqO+Rqur6wCf/pEEQcdfmkx3w7EemQdaeBh3TObBbPTRzeZ1eJaZeksO6eC3OYRzsG2+HrWKcvFulT/wLjUnnRfgtNgLA9WbLq6Q98sD/ehaQlH5BzAoeDdHcilqn0o2VB7Zb25PtSWS6URQ3XRUTtxEUjrqqBZf8xmmD12u92Yn5+P5eXl6j43fgmIAlepaNRBgt2dn5mZidnZ2Yg4scSBfFjKRJxwTl7e6/KPdYXjjpVkekPhJSM7WJZDdcDrAAq6BjNUcFRm6ORX0OF+OLfH7UB2BTRcy/VVNtTTTQsGcAdU+M5pnojxW7UcIcjIguqb8+0M0Bz83f2cLKPzPTd/DLicv4SNZjlhXMvskFNXKAqm+s4Jp6udnZ2KcOA/30/L/s2EQ1dDaEsJBOsbb/2an5+P+fn5iduoVIeqB+cLbo4d9rlyUveh1kUQPuYEc2w26wsRmJ+GGg6H0Wq1xl7qwMtTN3iVRZeRCiBor9M58faqTqdTTRbSDq1Wq4qK/GyxjtW9XYdlUiOHQbAhMPOA8fEvB6A9vc6NmfXO/Tkd6pwwADLIKYNVYNR+MVb+z/U03YBrePmrjIc3jPjdBqw3DUasN/0O24uICYbt0hRon1cnqAeGxqDIuuaggKUsy4O6ACaWhZk2CpOT0ejOX9flthzgQn/b29tjvx6r87e5uTlxnkGfAQ/64+MOHPEZc8EgrPPFK4KdnZ3qJ1PwE9TQdxbseZ75O9d1/qdzpaV2U4qF5o6dcMwcVFHqTHrOMQBmj53OiRdGLy4uVolv1NFbYdyuZGl8AGxVXKvVqm7Yh1Gyk+Clt+zITm/MwFhGrqcvkGDjUfaZBS8dF44rS2XmoLpm9sRzouCm1/J4YXTQl24UOAfKWISOGfpQQGK98RKYl9Sop+8yVXtW9qSAxP2p3h2jwtzyex40naBBQwM82znLqi8gUR3DvgGc+uAAbAt5Ut6V52ALvep8gI1ixaZ9YIzOH7UtlhnXuBwwByYwa7DU4XAYq6urY09KOlKVgaljyWoLpVKbQ80ac1FDhcuKE0wNFG1jWT0YDGJpaalaerOTuo0LjUiqVDZaXuqrYyHBjl1WvDUcy3s34aoXzVvy+PiYYy+ZYaojOmOEnlguBSZ15gzc9D8+8y46rxZ4nkvsgK/lsenjo/jTjSYUMBb9lc8SC+HvfF531Fl+HFcAU6Dn/h0T5ECpZEKDDK5l/UAW3OMccedmKt/ryct46IbvF+XlNxc85cQpCvUntQt+1JXTCW7+dW7RDtIKGYBjflh/sId+vx9LS0tx7Nix2NjYsPrVOXdj0pQB27HTFZdGS34HpHrcOUwGaE5QVw8D6Pf71Q38zHgyZ60bU0R+O4aCBp4AwRJ7bW2t2k10YKRj088KVM7BHajhnOb4dOxwItxxoNezDPjTTQ1tV50c/fAYIBtvuNTNrebtcIzl1eABh9LUiW6CwE44IHEu0KU3mF1rsNElOcaoY1Pn076dPpRJaTt8nIMWpxE4Bwn/WFlZiZWVleqFQQz0zvYwJgV0jIGBHHpU/cFPAOzZrWk8ToCoFgQFADW3xSwVfSM9t7CwUKVB9EUyOl70w+2oHzvyk5UioOoyFh2UGlVGq/V1Eh0YasTCUp9fvpBNNLNNlZ+jXuYMLCMzUvSjCs/0wvV0eecAlnNPzGwyHfO13D9fl21GuGMK6BpY+JzOIQM3G6Bj5vjMTsSR3zk8lnS41jE1pGXwB9amtsibOCx7ltJytpttIOkccC48u/WJwULnhhmfjtdtIvH/lZWVOHLkiLVJHq/efeLAFXIogCooqx/xdQD9zG8ZIDntwN95HtgGOO+Lm/2RS3U/o65Ex5HEEpCW8K/xjf0Z49LCzMGBZh1lxjnkNQeDQezbty8Gg0F1Xm+4hqPUgbS7v5THw8bOyzc2Al5uZuPk+voYIIOyW9ZzPZ1kBVgFRQcEeowNRY2J2+XcI5/j77w805QLHF51zOe1L75e+1HngQOyI4KZ8cZNHQFwbEz16xgk69cFMNatu4NBx8Zt6mcNaHyPrAY1MES8pITTZAz0LoXg+talOwr7oLN/zBXrUXPACtZu3Pz0U2aHrP92+8StlQsLC7GyslJtXLu55f/uHNos2Y+WRu9DdeCBcxpxWSh3DZ9zjszsptvtxuLiYsVO+Z2ImKASkPIEOxaFz5pvjJi8C4Anj0HBbRi56Mag4ViW6oqd3NVnOR27dfPhioIhy6eMESVjZtCPswkOVPjPOT3Vq2MREZ4loT2uxxtAXJSpAoBZZgY5ZczMDBn8NSXgdKPg7TZqIibTKZwL1ByyyoG+8Ba2+fn56PV6qT/qBhUHKbYrzsU7n8U5Ps53SGhwYVvgecFx3djiuk5nHGyAG0ePHq0e+9a9DpVXZXBjLfWP0oihOubkonUmXMnBtB0sDTqdTvXyA+S82NAwOcwas5LJ4YCBAaY09szhta6eG43uvFuA2RbYLNfPdMjgwnWZLem8qP55M4nHzrIzMCiwwslYX+ycKgtfp5sKfIsQ10dOVoOULl3ZUbl91oMLbpqKYTakgYaBUJkvbFP1qcxVwZwBU4OJziPrgsGG7QD+AUBxuuXcK88VFwbTiKieQpqdnbUByn1mW0F/yOurznl+lThwYISudNOPdQzZcb86dvz5/nAXBDLf0TmtY6uNHz11inPgyqUOaF1/UCreJDU/Px8RMZE71X7dI6QR/ikilltvH0JhB80YooIZ6jPjVWDgdhy70rYyVqFRmxmftq/tsr7xPwMPZ1BqkAwyHPggExsqMym3sYC28B1z4fL5OIe2GKR4fJq2cA7FdZWda9B1S90mK5ZsTjQXynVYFgYmyKqpFdTnnDPa4brtdjt98xPra3t7u1o693q9CqARUOvSKkpO+DsDI7NRrsfBku2K55LnDec7nRO/lHrHHXdUb4JjX3ckiO2hhFOl0vhJKf7PhY3YOZ8aeakvsFP8cqn7OROl7mzMTgFw6kw5nLtRUOYxKOBkTukcWqOpsi6NnJnetV8GQs1Bqb6VEYHlMOgrMGlu1zFRx/T0nk18ZiPOAhO362TXpa2mZpi1ap8sJ4pjRbrjzG1qMNA2YcMqv5sP1OeAy/U4tcX96X3CDGz81jMuOnd8nt8/gPFymZ+fj1arVTFUgBPrW69TP+CUAQKC2rMDUW2HbVZ1qfPX6/ViaWkpVlZWxn5ZQlcM2TxlPlQC2kbP8jvhFSQdaCrAqGAZOxgMBtXPmUSM3/qAZZmyG/eTt2iT23cMDMf06RSn4IyxoX2eVFV+xjIZIAEaEZNLZwVu9MO3zOg1Cs4cMFQHGUhEjD8ZpEEFRq560TG4AJTZArMzrcNjYaDggIBjsBVOs2iOlFkc98cyqu6ZkXN/+vq8LKhoUOTvGqT4Wk2jZGPBnHHfqm8NUApiCI6DwWBs/jjFxrvoukLkz2w//K4DlZ37x1zhevTNOMC6wDg5NYg8Mh4Zdw9HsE1pW66uy8ei7PpH+pwgzundtQ6c1NHxCBk/XsqMAMpUcHMFQKNGqEatu5FZ0MDtO5mCdYmoemEdunyZc6RsjKijbEjraPrBjdsFDjZoDmAwWAVmB7iscxROuTAbZkBnGdRB2SHc8prnikFBx67gy/OneT3NcWrAy0BD23ABUYMq2wKu2dnZGXuKifUJcOM54c1blYHbVntzOW83j/wcP/rhMfFqizceOXfrUj3ogwOCFgZ23fTidAb0gtSh5lJ1jDonOk84zvpwpfFtU87BtTPHfNQxnEAc9RcWFmJ5eXni8dKI8WVLJgcK6igzwjGV27ES/syg6ZLXKosyD+4rYpzBqnxZBHSsxd1HiOVbBnDqaBiX1tflrQYCnl8uLnjoZo/mVXEdj0k3IdXQ8Z3BXgMtB1R9qosDoANylSFi8qdB+CGB0Wh8KcxjY7mZ0XMbGhDQjy6FWS7Wv9vQYtBEgHQESBno9vZ29Vb0ZK+CAAAgAElEQVR+TR8wsdFcLbNGJR+6Gal3d6gdIUgwMdJn9FEU+NlWFhYWYnFxMdbX16uxQV4eP1+vbatNZ2VXu/y7qec6d6ySlYWXn+BNUjxwnVRtU+txLktlgFwMTNoWOy2DsEZ7B3LaNl/Dy+os15cxGNWdBhaux7pmJoq+danmQBbF5bzUqPVaBg3dbOJr4DQKJtAD2mIA5H504w/1UVQHLCvXKTEyPcc2wTLxKkADuaZgeI5cn5h3XaYyyDAzY3lYz9ovmJv6Ad5vMBqd+KXSY8eOxcrKSgyHw5ibm5sIEMw+8TIS7lODGv+qBmTGBpsyXvYtXg0ww2X59Qk5lWFmZiaGw2EcO3asklNXEDr3XLheCUwjGgBqxipdPSeEtqPGA0PAbVL4fSgGU2U0aD9i8i1TmARO3LPDKgNk53BjUDkdU+dcXgnQMnBHGxk4KoPFeccO2bHQpgJ2KQfE7TAAM0Pn88oWHXN1Kwp+ezsAAS8lRhuaN+N2lYUoQ9d5B6BrIOCA6JhqNmYcV724/DPXd6kRXq5rmxwk1Fa5D84psr41+LFO+G6KnZ2d6neZNjc347bbbqvu6YROer3exKqBi1sNoi42vthmGCy5TbZfnhvVHxc+j+/YiIqI6hZMfVcxF9Wp+jpkv0s5VAUAByjcmXOyLNKjtNsn7htbWlqq2JAaaiabRideImg04/70ODseO6eeZ8Pm4+w06sgKJloccCnLUXl1ievOcZ9ZQOEx87xxG46N4jzrgxmae+iCx6b5QMfguF8Hatyesz21Q13ys/NxMEa+j98ZoGNlXUeM3xPM8rKtOaYLZsgsHNfzhlo2TrdkdhtAkIXnFzrh9AHAptvtVj8NvbOzUy2X+/1+VZd9Qsev9qwrPD4HmTUNpDbMtwUye+cUBAddzHGrdWJvZmFhIY4cOVK9OAXBm/tQMuBIUIml7uonUNRxnYGw8SlzUaGh9F6vF8vLy7GwsDCWo9OJ4EGxoeky370pnZ2Q2+RJ1ByUY39cGNAUqHTsPC5eKiobVxbCaQGebF6m63XQhW6aYIxaFAgyUHQBUp2I9aKMgnXF7XEQU1vCRiDrg/tjMGF50LeyIrTNG08cHJQVcn3VnbNNt8HivqM/vd3NrQR2dk5sSvFdL3j4hfWldx4oo+a5wpukdFMLfS4vL4/ZEH5kj4MM2kW/DN64RkFSH9Jhf1RbRR238ao4wZunrD/W99zcXMVSOXhpPdiKEiHFA1cavRwlY1X82bFQZQt8HSIK3mOI95yyATrFoegjbHjFngKZGjWcl2V3OZi6SOQCRDbpKosCNwOyc36N6jz5bAQRMQYGMDK+TxFzqrJzOzoWDQwK8A7sAIQaZPSzBiNd5mu/uIZtgvOKaMe9bMQFewd2Ghh0Xku5b5f6UQDSuyQwFt5wAYhCH7zB44K3zhMTDrUjDlY6FwhE+O00vKay1WpVO+VamO1GxMTGnG5Iqi9jrNlLtTXQQlYN9hyYlfHDBxYXF+PYsWNVQMGuv5IF7cv5piu7vm1Kz2VIrkrTusxOl5aWqpefsBJcO+wwbrmv16jDOIak+UrHMlgHji3zuLK+nf6cwyqw8bUsJzMsHoeyPOfcKgNH+kwuBU5mG7oi0aUyywhnZ4CHA+iGSbfbHevT5ZZRFCA0cKmOsTrQ/CXrFGNitoRxMQhwoINdak4XbStwsOyQm+ebZea2II+OLwuMaI8DD2QFmWm322MpCFzT6/XGNjIB1qw/3JbEcrD+HRjzeN350Wg09vPT7INMGhh8Oc3ADBipxcXFxYmb/VXPztfrwDRiF7v82qmbKBVGj7PT4m9xcXFsqY9oprkXjjxs8M6Y+LPKlR2DI2SgyyyDr1EA47FrXlL1puf4WKZbF+G5bwYTyA1j15wsA5qCkMqiy0jkGnUcjpHxfLIudRxoQ/PIPA/qNHorjrapbFRzzzqXAByWh+eSda15QFzLaZydnZ2xp5e4T3Z29AEnZ8at+UgGOs0D8nyoLfBSm+dwe3s7er1e1ScHAQ5Sg8Fg7I4Pfr0gByLn/yyDS9toygGF9ai60jHyeSU4GEu73a7uXNjY2BhrIwtCKM5mtZz0z0jXdazXs+HiO37BVN/Cj6KTg2PctkYhlU2jTQZuKqe2z/XZgRV41cGUUbp24IAaJd14NP8KA2VA0Hb4T0FA+9M2GGw0qHA7DMQOcDTwoI4+T64Gy0DG7Tj2zrrRAIPzcF4eK+tH5yYLgCqn8wfU05w+B4NWq1XlJ9FHxJ1PIDGhQJt8jO2OARHHeI42NzdjY2OjulVpY2Mjjh49GltbW7G0tFTlaNkG+Le5+L0MYKOYZ/Y/pC50fpicqI1p0MNx3WhS/+QAgLbZzlwail/vhyDjfEKLjseVXS35HXvjyXMGpQYJ5eONOLoR5SILJpaPo72I8ReZOOdzTpApxzmLWyazDtSRI+78NUZdwmpOOAM3OIiT27FG/DE4K/NSFucAhVcBfJzra78OpBVgdTMDY+C8YcZAFDw1n6r5Vl5aow/VjeqTWRbrW8FPGR6uRft8DdrhJbYyY/6BR32cU5evEVGlQABoHFSZ6ekmGl62DFABwzx+/PiE/yF3Cz/lJ+V0fnjDkMFWiQZ0rfOvbJvBeTQaRa/Xq37NdH19vWrLrSgQmHk+oX+2nU6nE8vLy9VbtNRWIY/Kz/OQlcb3oSo4qrM7tucEhAIGg0G1EcX10I9GHMc0FdgVlJUhKqhkjKIExDwWXX7zGBiUFDDYaTPAwB/fg8m6z5LknMtjuRj0FTT0+kwuBVi97Ujf0K7siUFbdadzgTwrX6+AxeMFgKhdsHw6Nv7OgMHzoPaLPC/qcf+sW025OP1n41cygdLtdqucYadz4jeUACIs987OzhjwQuf4SRL4HN6Z2ul0qnemqn45YCnjZqbJdsIPHXBRxqhjjYiJ38dqt0/8BBLaRFCAPbAe2ba4fSYoODYYDGJhYSGOHj06BsJsHyUfzUrtkt9NvjMEdTw9x9EE7FSf1+e6ejsJChyNIyI+s3Exi2FDzgCFwQuf0b8zDPxXdoL2ECH1vQPKVtX5dMwuR+wmO5sLLuhPmUbE+Nvt2YG0qHHh5m/WgbIw6ICXo/rWeR0Ds1mwCr7O3UqD+rxZxnJoUGV2yX1ySgDzA4fk4OACu84x94c+OM0D3St480YMrteNMf4ZdRSA7fr6egXA6JefVGq1WtHv92Pv3r1jPzYJoOIAwRuDDhQdoeHxKdFhXXAbDJBKFlqtEynC1dVVq2eXz9a+UAfMfjgcxmAwmHicVftvAqQotQyVAYOVqUywxFR1eYAf0tLckoIcR1yOis6hmFVw4UlysjsWw1HdjYuPuV1W9MtskZ0abbBj63GO/tkSSnWnbFXHiu+6Gw4nhK75Gt5M4Hdh8nwyKPM16I+BVgOBRn4dN8+hS3FwDpnBmc/DfhXcWF8azDjA4HpOh7BPqC51g4XnjPXJY9YgB92CsSk4qS5dKoz1z6BTmi8u3C4HErTLY9ZVg4Kv2+hzvqC64TnCI+krKyvWRznQ8a2CCsAgAbOzszEcDqu0B6dJNO3VtDS6D9WxKi4ltqSK6Xa71S5bxPhGlEZyB844x5/5emYS2g5HUZa9tNTjseh4WCfct+bqWBeZ4/N5BTOnYwYTFGZbPHdsGAhG7Ji6dOJoroGKnUwfD0aAQft8HyXLxG8gcg7KgKXXOnDW27YUbLgNnXdlRDw/rF8nD+tY9clApMtN1hfLDZ3j/mzWH8uud1ewjBibPoqqwZ19AWDCwQPt4GZ+vZuCwYof8XT2wnbL53jlyHlOkBoGOAZV3JjPcws9Mp5oMGOyMhqNqmf8b7/99kqnagsOTEsAW86w0sWOUSnQKG1GXY7ug8EglpaWKoNBTkQT0sxG2ClRdAI5KvKtFrycccxUI7s6sxqEc1Ru2wUBp6ssWGC8Crg8ybxswzVgNGoEmjpxDF6DQ6vVmnBaXgJGxMQySdkq24EatbMVXfpzfcwtAxj/+BoAiB2cl/uQmZe/morBcd7I0pWCA1gdC3TFNo82dDXDgKdBnIGr2+1Wy3fec3DMnK9XW9JHbxmYIAunCThAqt1wO/xoLGRGfWWJuI790hXIBXBFvhi5Y9Rh++IAywVj5/lEncFgEMPhcILAMZ5kBNKVxq/vc4UbZrrNDIoBCrlTvL2m3R5/65EqmBmfMiRlYq3WiZzQzs6d9/1xWkCZMh9jsGG5leXxZ72e282YpdOdRlNtk40A7fJmDQMmMyQeK+sI17HMMDCt40DfsVYUHT8DrPt575Iu0R6zYGU5mv6BPni5x3orzQnawivr9KERZjc4BmfP0glKKtixlXmpDhnIMUYGTpAFvY5vc9InsjD2tbW1MVvBZ35KjWVgubWonvCd32Oc2biCLbN4ZskbGxsxGAyqMczNzcXa2trEo67QMz+h54Cbcavf78dwOIwjR45MpCncarau7Po3pdgRlRE5FscMFTtriLIRkzdmM0PlNtmYdamLfMj8/Hxsb29XL0BwbdUVB3Q8bjZYFP0pXJ1A59SsHw1MLEtm9I6hQz5mlxkb5vnjp3/Qny63FOj4MwM0g4wuw7gN1QfncVkW11/E5EuO1R6c8WtOVguCPueYeWxqTw7cHUviTTJcz/as9o9NE31DE2TCW7l0bMz+oSPWJ8sOH0QAYuaIfpCSwdzgj+eEGTL6wuv89GESlgNpMXeegXg0OpEzXVhYiNnZ2VhbW4t+vz+xQeUCM28yu8CFc0hBrq+vj606dJx8bVYaM1TXiAIp12UjxFJlOBxWN/G75ZOLVty3Y6V4F8BgMIjBYBA7OzuVoh2g6HFtm+vosl+vQdG8mKuvDpQZAretDo16KheMk69zbAC6036YjepLf7lkAJIFUJYjY3IsHwMF13Ny8HiRa3QMXwOO6t/lO8F0dU6UsfGz69ndHJgbDZrKBiNiLG/KtsK6xTWaz9XPnHbQ49i/4NWP+m+73a7uXeX2dHxoh/Wu53gFoYFUfYJl4rbX19djeXm5SvX0+/1YW1ub+OE9nS8370wmdnZ2otfrxWAwiDvuuGNsBeHsxfk/l10t+dXZWbEKTAwmrVYr5ubmYnFxsYq67rldZhAqPBtHJfzMTMzNzVU/IoY38rjbUBx1V8Uo2LCDYHKUtWk/XDJAwjmVi/vU/B3rA6zPtav5YgY9Bi9mIAyE+js/mr9WwGF5IbPep8lLKNYJp4kAPNiB1flwwZHnBvWVNXH7uiRnIGCWxsxO50kBEPU40LFelO3weQUXgJfb0GT9c06W+1Jb4bnKwJA3CCPGd+P55c9oz7XB9sgsFmNRgsJF9erkHI1Gsbq6GktLSzE3Nxerq6vVqwVBANTWUDDv6qOaU52dnY1ut1u9E9b5ZjYGLo0AVdlOCaAci2u3T9wqhdwp8itO6Kw4w+j3+9XPTK+trcXs7OwE8EE+ZoqOvarz4jp2UmY7mYFyXZfGcJOlDsFyAfgc22BDHo3Gf36ZwZ6XT9yGMgI3JxrkVIaMQbOBl26P4/nlXW3eJOE5KuV2XQ4ceuK50PFBxkwHcDINANl8qz3wf5631dXVsXnd2tqKtbW1iDixmz03N2d36yPGVxzMAN1N/hmg4ru+Zg92yHOBfpUIIRBy3prTeGo3GhR1BeUCJvRy/PjxWFxcrO48wPP4mvvkx5n5ReYsP9sBdvsHg0Gsra3Ze+CdzK6c1JJfO1LA0XOzs7MVO+Xlj9LoktA8MXDawWAQ8/PzcezYscpwNjY2Jm5MV8BXAHDHuC9MDtpQp1YQVl0AKNyLguv6xmcXDHDu/2PtzZIbSZKsawU4gFNEMCuHSqkt9GZ6sb2Dln7ut15CiVRnxsAZBEngfwi5HscP1UBmf7+JUAA63M3U1NSuXlUzd++AOjIbqGPkuaYzHqcUDBR2lpxkZAMjgPHkSt/8DiLqn9tavKrMugluTks4Fx39eE+zQ2hvM7Njsl0ZvDx+ZPMBq/V6XTc3N3V/f1+73a5+++236d56jnNVvWKDnjvet/zy8jK7yyptk3wEELsX8HW6zm+pg2mKRCnMx9IRkAjYDgjG3K2zXC7r5uZmShv6cX/Ok3Z41EXAKdkbf319/ep82sRbBPBdt5524GSBOmaVW9qC/hHMzG00WPxuJXAR6unpqQ4PD2u1Wk15EIN12hnRdvelCx846G85AAJZd+dO17bB0qCZycEwzPs8fb7zlmmfjonj0gF7l7rIpPC4MFfqlAWBP/pwvpVt0/bYTmRnnpMTyPlMA0yAjMesH+qM/e5sh+Dt+rzQVlVT2ou2keeO5uEl1mHG4uXl+8vzsk0tY510DXXMdp3u4DkBwOjaUU9+t0Ng9GBGSsfaFbJI988YkX6t1+u6vr6un3766dWCVvplO4gsOd9kiGmR1Wo1hf0c8w6ch/3a96PZU/d7938GmoC6Wq1ehVJdGEYv5zwiQ66Tk5M6PDycXrx1cXFRVdVup+jkHjmDUarA/evOY30OjTsWGiNk/0Z6j4y87dbMqHN2nlBmIATdXOPXEPs8L7IQ+LLCThZiQ85ErapXoJpJ5vytJ090ax1wLMlweHtsGG9k5cQz++cYUa90Whw/j0O+R6d8YI5TEyEFZ2dnU/qK9RCkR6y6IwHRqRcuozeyV27F6upL+O25zoeMUAe0K+8Q4BgRqD3mOb7dbuvu7q7Oz8/r+Pj41eP3WDxHeZz6pG2uVqtarVZ1d3f3SueR+63yrm1TnjzdZHeDMbAwSRoOO0sjNIOp6nNABwcHU+6U7HSz2cy8C+vjX8e2c36nvM5x2Mt3HtrXu23rl+DbAYT1zAHn5OsiAF7j2/aYzujqIuCSeRhc2EdODvabILJYvL5PvHMsAT1OLjuUtENnQPnMMhiqEvwD5pGrcxzUPUGa55k5dw7WOjo9Pa2Tk5M6OTmp1Wo1c2ApdgwcI44T9cBcYogOATF63Ww2dXt7O21LSt0cG14T557Fzc1mUw8PD1M/OJadnmxHjkQMYMk7f/nypT58+FCbzWb2e9cGdW6g9xbIPNnq+vp6tthFud5iq+96fF+HzAaJ7pwkjpMLeisMqOo3I/v/3DmSfOnFxUUdHR3V9fV1+6CVETuk/CkeyHw30+zCTU7wDEIGgmFUB/Bmjfv0zTa7wfZEG7Vjdtf1l0yCCzgsZib5zn2UHoPOZjo5+Jsntp1OdMHVbzIQHmP96VeOeeXbeqET4p1L1G1ykna07N9qtZraPj09nRgznztKOTvWxf547FzMKnOMtpmtUl6tT2GqIaw247Ber+vq6moiOH4NTT55nWWz08r3EKfFYlE3NzdTJGoyYFbJ795uRqfDrZ3ePZD6PQe78u63nhpkTO05WGGUWYw6Ojp65SVo1GYOPO5Fk7Ca7XY7rfadnZ3NQhGyL8s3Yg6p26CQ/zM5GKKMwJn96waBuqMXJwB2uR6WnDvaf9mNDRmZx5OT0b8bJLs+55iBxjKlby50QJGzG6OMvXPEmUhkdLQfssnI7VQTAdk5Vk8ihq8MSWm/CYMNTAZqbprPw6ZfXl7q5ORkxmKZd/W4m62nT9Z1dGwbzm+J9OIQCL5cXOXCWFhw1fdbOZPeGxEiYsEor2l957yAaNgpx5+2xfGsmqeWujGIHo6Pj6d0Ap34e8C06i++RprHDKqeoGSnVPi+On0rWjexM6AxvOyVu7m5qcfHx1eMrer15uFR+x0zNMMy6I6KQzyHxczPUScGWbexbxJ1Rus2nZdjP60rnme2TbAiaBng2HbqYi7Qenfaw46R19K5jdIbTvtQfttHZO3AkZ+0CbIkht1VVQ8PD/Xw8FCHh4ez/dHsn1MeT09P064V5lED8LmBZbf7sfOB+nf4340lz6NOq2pKORBMedccr4tc+T/3xftONcsSe4jzyyfnGa8nUMaZePsWx6cjSXSAWdilbtKvo6OjOjs7m55oRf2MMILlzRyqAWgfkLAzx8fH9eHDh1lYs0+wTpEZULMKPkU8K/3ZbuL8rEE0bcQDj4oZH491/9tgRgZIEGGfyRw6hmlw8ZiwbTIktl31Y2LaGVJ3vJ6LH52DyqdX9w2A1D1X4e0cyLwMBOyT87zsPz93ux8LeQEhA3r6lH54XCkLgSxEIXJ7UgeI7u/v6+7ubroRJflR3oCy3W5nDwpKvpAOKMUvqPRcYv/4Oxmhz83x7qaK4+PjKSJkdBDdpv/ZZx47s3yOBmgX3feu5Lq3ts6xrm6uMA3EKOLo6Gjqh22wY84u71rl50T1BOV5MZLFYjEtRsXo6BVHd0XYkOlpOeHCUE9PT+vi4qJub2+nDb65jhTfAxqP1Q0wJ2TXX+dCo+jUY72kHhuxWVCXX+Zkct4vkz8yW3YajrfudPd4m7lS/wZzAkjkY740n96KEznYFu0hY5LIw7bFcbS+/N2OKdekbYJyxsL5e0cPnkwBGY5N1gvSfm6RzG6UvOU3CzfeyRBgSlmv11P6LE/cZ2Rg0sM+ZQwob84JU6PeXTLH/u3f/q0uLi7qv//7v+t///d/q+qHYx7phvljkxLqOuemPdo1f2OhLaYe2mxK7Ia7PezgqY8cS/43KUXOw/8nQB0xJXsUT/QYBb2dwSffOZij8IpKTXtHR0fTI73u7u6mVUyGeGyXk5oyewBGntOA69KFytZdx2DoJVNP5GBIx/M6PVkOy9TJ4v76+aWdfNYHDTvXRCYCvhkVgTafsZfuNct2GnZiBJaOQdNxjBgSJ4/tJDJS1+xfxpP9q6rpnVFJUa3X69psNnV+fl4XFxfTo+PomPgoxPQl9h67zsp9HmXHdr0hnnqKvBkT3uXUgccvv/xS//7v/17/+Mc/arvd1n/9139NDxEZzQ8X7sOmg2YdBFWTKAMn9WuyYJJhx8uUCI/RfrLLgdvDIs9o/qe8yVANnmZ7OU6DysMGuOcxSvSE6OqhQka5xBjX1dXV9ObGTjb+3ymR/3OAPekif6effTlaAwOPdbLxPBpWfutynf49MhlIzNTcH+c0O2+e35w6YK6vC/0jAyew2W9yWwENhs3O+ZkdeSw7MPTODOubeunk5uRk6Evb5+S8u7urL1++TM8dpX1li1KOhxXR0Vb9YP6RK88Q5r5WP7KPkUJuEshY+m6hFPaLi3K5ZXy3202r93xDq5mbU11M4XBsA1bOLVO3kYsOOUTDd9UZ7GgbSaVEX8Sk6CKOKGQt275ix87tj8qbgEqB/dkB13L5/b79PL/QyWN2oOr16puLJ/J2u50ezff4+FhXV1czMOU1uY5K6RTS5ZQMRDYesioOhic6B8NG1wGEFzhGsuR4V5dZf/eEdzJ/50r9mTZzLVlcitkNgTLXha0xfxgAYCHY8WnwbtPX2BnTZsJ+OU5m1WRN3KvrVIbTNRyL7XY7vaJ5vV7PQDfglxfi0XFw8nKMl8vlbE8sxzugwjnGXGfGPakFOoT0i2DM/kfmz58/13/8x39UVdX//M//THOPbJAgZccWu2AfuUrPt6aSOXPOdHVS55Hd7dMGOhKT4msCqMz1d2Ddlb/8PFQeZyMZjKOjo+nZhaPShWApo4R6jm82m7q6uqq7u7t6fHyc3ePNQeHktzLcF7MyhnyUmSkDAgaNicCWT7NztkvmkEkSGcx+oy/KTuOzTGb5ZJSU0TmktwxnsVhMuULn0rKtiWw53/lQHC7m8LNzhrQXRxbUEfXET+qawMVCuahz5lfpDFKib8rFh5wwWkvuOjYUvWddwBGcnUfaCGOM/qPvo6OjV+yVKZj0L/2JHPxO3W232/r8+XP953/+Z1XVjJmm77ZH2podJZ1ntoSZOEQ2j7Nt2hjCYyMG6TQKz+Xncrmcnj7FV3VTllF599OmRhWZlWUl04zFwrhjXZ3ddekgH4LCPwOWgSy/dWCY37qcJY9HHjNQDxABbMSMCfqs2yDH8zgmPm5joefugN+MLb95bKgrskvryeNnI8/kJehWzbf+2Aly4tgpdDZimTsZmQLoyALHlHcasS3aBdMX2Ud6fHxcJycnEyPN7zzfxMI5vxRuoGf/7fgXix93MPEGgZy7281fEcPrRikePlaQ6RDPN4bqHC/+lgVH7r3txjDtUTfGDs4RXsu5YFvwd49nSvajPjw8tLKNyrsZqv+nsXJws7E3HR8NklG/exJ+rnHy36yP8nhyGDQ96Q12ZgkxIk4C9oUAQLZC788+53sHzAY7e2n2KcX5vowDt/90bGxkuOwT28ykIBDzHLJRyuy9fHxnE9mT86Qe631jybYps1MtzENaf3Scu91uihTI+Kg/voU3MlD+7L/Oa3moA4InddU5V17HY7ybi3Inyqmav9vJ9acfKdlOxvPtwLpdBc515rsjNZcuKo3MXCxjv02eRrpxvdE3F647Elc1jxS506QjeV15862nbNAezaBwcPD9HvusvtuQmKva5yG68MMyuW2zQQ+k5e+YqdvsWGnn6TyxDTpkfb7TimBqnZsFsX6CdYq3qnBrE1MZnNC5zqDbycA++7vTBx63HDeopV4zQI+Lx5QgaLaXPjuUpYNkvd1WOO4H7dIYZmcG8ywGsaTPrK+zOdvVaFzyyaf8U8+eA3aYCa3tuPjOMr46nEBEe2K/yYCrfrxkMLbPHQy0fzq96JjpLNpEnEiXT++AstOFdUsiFJ2sVqvZbo8OV1z+EkP1b2Z3x8fHs3DfDKljUjQM1kv2YoU4XcAtH+l0N8EJuLmWTHckL9lO1Q925cWfgGX3LqBcZ2ZNfcYYOTE6z2hnZedCoOUWGl7P8J76MHPomACNy3qzE2X/CJJcvWbdIzZHHZnpG+TpmKMD6i138uQ4t2ylTwQbOqAuhUX95lzvmIj8TuUQQDwHmPPLOFHmnG+HSFClg/DWNUYNYYUpzOl6HB1tuDw+Ptb19XUtl8u6vLys4+PjaZ4y1UAixjFP8TxO/20PlG0kr1NMLDlOHfJ5BI+Pj6/qHJV370O112RHYv5zGDkAACAASURBVEBJ5AYYOvbStcHS5UUMLN1nx6YNnvmtY72ZqM5jkXHSi7o+npdQsduiYUbVsVQCByeEWbhZYwd+NhY7FoOowTLXesJbH/yf4x1dejU3fXL71A3HkCkRsvXOUYYNUbac1+0VtS1w/OM4kxdlYb8JVrydNCV96BaarHuCaQcC3TjQMXWLTJ57dDrcLZBw1+m0OBgvqEVeth1deVw7J+sFPjtZ99kgTnsy+PocR4607U4/3Bli2x2Vd90p1XXMBnx4eDg9x3FUPNE6ITvAy4B6s3XOf6t48nBi0mA74zVY7PNy6Q8NrwM+bl424Pu6tEdQ4YTnZPGYGJgjGyeg85ac0KPUTFdHwsPRmEYm9oVjwXPiPBzyRT/st0Pw7h5tTiiDlScO0zQEfq5Qe4IasKwjAs6I6TMiq6pXeVvaqB1IwJCgnT56a5jzk7vdbnqSE6MBtpX2k5ZhHjxjxnN5e21IBe9k3EduUmgHdiCeAyzUtUHSRMlv0Mh3pyty/D3l3c9DtZGZFeUe2I7hMTR22NQBohXdeSuzZyu+C7ndD+fgzOS6cNqFclonBD22S7ldD/eGchJRTu8rrZrvw+O96/ktvxvIuU8wuqbHNwvqmF9YHPvmNsLacw2N1nrNtcxh0n7shHg9WeRyuZzAgjlVjxHlIrjZedKO0+7BwcE0Vg7PIy9BtXMkne479kmG1dki6/YKOecEgTbn571Z7LfTHHxIPPvLsT04OJgtSkfe7Dv166tTeG4+ab8cbxeSjfw5leVF1BER4/E8JrQjBKPy7m1THRBkoA4PD2cPxd230d77S0dAxWvtvTuQ60ArZQTAPscD57rdn+53tuNz3V4mbxfmOp9jvVtm54BGAM7z8z/zUwZoh55hQxnLTs+pP/2O3LxDinqh40gISodCkOP4ELhtR2Ssy+VymtBkZjwvbZHddeMaeWL3lIHsnIsbHCvbukGLNsPfU/i9A2KCYQcquY7s3mSChKMbZ8rrPuausNzAECfsLVzUnx0k62e4TT26mMSxTpIr2zTH3tFo8qjpA/U2Kn9pH6q/UzF5VJ8VX9XfI20wJLOi0B6EjpV2xzrQZH2dAXfKHfW5Y+KUkazCsrC/3eC6nzQ0s0CfZ52QGdl4OXE6sKY+CPIEarNPg6XBKr9xL6rBkIBt0LAtUv7U44jDjKJjaBkPs1DK41A3/SfQUgbbVUcc0m6nP99FZidFfcTRURbrnTeNMLynY41uuI3RNkU5LYtTI5xPnvNkzl3k04Ep9ZD0Bm3BuqXc6XtshekdEr20w/xydMax6sq7bj113seTODTfe0k9ubgSb4Dh8ZFH7pihFcgQy9dRyeyjZc7xbmMxvToBJf9zo7w9nkP4qtev5bVM9rb8JOgzHKTxGkCjH4a+Bmq3y7Hy3WKctGyHf51HzySKLAZF2tho10Rk8/VdnpJM1Xp8eXmZtsjkf9sJbYLOPyTh5eVldndU5wDoZMnsEtF1tk/gS1uWKzJnLprxG+zI0jy+TOukTTpPyrVYLGZP948emXfkGNuJ2Q4oI52WbS6ydBGZx4t1sF6CZnd+Spfm6RxjyrtyqCNASseYa4gXoBfx9Z70biPXVb3el9mdw8nX5ZfM/DzBzSLTDvfjxThTn9mwdcb2zKSYx+xyaiwdQ8xxMwmyDOqDXtxgzfrs5Ci7ZXI/2J+0MXIo1DtzcN2+Qod3XQRhB0MdcDwZplbVLGQnyNGOqQs//cvt8+6fLjR0ntBkwWx93xywg7GeDNrUEZ2OZehCebZtG/RvZr6pl4VbA32O5/AovO6u86fnc+YwiVLnoGkfR0dHdXx83N5o0JV3hfzuCCdTHiSQ1X2HKexkhHFOlIWTlOEJPZQnCtvwpCSAdOzZIE3F+Yk2ZnIBNU4wA0zaJTBxApOdO9we6Z51dcDBNkYhb5gIgXrUpidXp/eOSTPMZR1hWlU1bQKn06KhdzlGsxmOiW3K+xaZoiBjst2EhTJSyXGWXJc3RxiwrMuR43TbvIb2wvG2bXIsDIxdPTnX0UZVzfrCNkbRI9mr51nqTv+fnp6mvams4+Xl5dUC30gvPGeky1GUQb2w2MZoj7m2u47l3U+bioD5y2Q9OPj+qmjeimfm6Dp53GA0Kmx7BFSe6J5oBjR7exvQW1SfjqWTtfPQ3WThcYaE1HU3Sfk7DdlyBXwCDvmNgGt2ZB1YzpGjok49tvkeoKLeq+Yb0pnf4vixHx4z9sP9IbgT2J3+YIhaVS0zsZPnw0i8tcgM0FvBbAN0cJG5c2a5jtugcu5or2uOJxdI0E3fc15nw3mu62hrJHVOm+yIkUNt6pkyRF6mtGynnBsdgzbZ4XkdwLLdPHcgfaMddeUvMVQPZhrPrWVusGM8Pk5F2BPk+OgOD7JMT6IRAzWIGniZ00sxMKV0HjT952Qw8HBwutyjZWPoyX7YucSjmsFGNuo48ndswg4gAOeQmMDdgQDtIedZrkw8PqKOY0Q9s66OXWZSOvS0M2UfyXazCOE7h/ideu/SWmnTr2rpojaOP4tTDUntRH6OFx1LgIp6zJhxDFNPbliIvgmyXeosD8mmLJTVqZTIz8VmOsHcBECbyvUcO46R52BK5+ip644c5rfokiTA88T1jcq7AdUDGaVky1QXWvo6d4KeLMdHodKozs7Dm5WYXXZsjINNOem9yWQ8EB0jNdCzTntg10GDGuUN7YzICmiAlMV6s7Pg1iYW68XOwgxwt5u/cYCTJiyuA1EDPNuz86M9RBfdq1M4JqPQmIBERuL9iym5nzyhvvXkHGXGhouWXXqK4E19d46cdjNKFzlKoAPv2qGeMk65hlFoxoAAS1s3QYq9Rp9uP7J1jpDAnLrjCCgz7WOEESnUifWa/tMeaOsdoKe864n9/uPKV5K27ERWLcloDKycgBSYBkdWZeWRUe6Tn0DVgR4LJzdBIzJ2/ejqM7ukh62q2QoxAbTrD50NDbVrmzkustwR+08dBF7203Wx35SNYJ7fAzY0vjAkb3UJiJg5dn3ipGZ6pBuPfX11moQLScmvBjDSDh92nfO9cyXj6MjD4SdtpXPgHSsPkLDfdox0cJx7DL/dd+cOIydtP/1xf+kUc303jzgWHFNuTUq7ZNWOcEw8qMdOvz6HkQbrMJHjQ2HolPax06o3ALVDbO5lc/6UHojK4yCkXv69Jex7GWr+t5H5XBquvWDHEANKZoVV8xVb15Xz2Hc7hHzv7gzxeXQ83QRlusH9Ibjmmm4vZhfe2Ni61ACBP8DDieGx98Qg8FN2TlCzv5EdEDipOwKIQ1/mdbnFhrk123l0SIdFQmAZI7+BmeBmlsh2uf/RY2JQpqNh3+m8MlaRLde7XjLi5XI527yfrWJ5KBJlixy0Iduy2Z8BjO13Tt1svSNOZuyxD44t9eVwfzSHu/KXcqj0PMl75W4Ioj+LgcdgQSByuOmQz0zVXo1GZUO2QkZenaE9B4L/m0mTTXRtmLnR6Dm4zH/x/PShY11hTs5D5nfKn89MYrIWhltmpGzbW4cIXB1z5hh6HNxOjnXMiv2g3ukcI1fVjyiJ50RmsrxsjVkul9Mqc/oY0Eg/bb+RhwDGMeue9k7HSV34NTS8hs8q5Zh6+xYL0yx23MylmjS5Tu5vdV8eHh7q6uqqVqtVnZ2dTQy660PmVmSK7u1A/T3X2k44RzsiZlZO3GIdXZ0mViNn05W/vG2qah7qeXtFx/AoPA0wv5upkGl5i1XnyT35qZRuoChfB4BmkayDCu6AktfZo3IwDfhcgSUToTPpJnZ+5108ZufO/xhw+eg6h8EEcU4stu2xZ5sGPI5xZPVWqVEIn9+iDzIth/6RwamDABefGp9ryaIYeeRZAh4/O6HozDZG0KJDoLwMe+3UTCA4NtYXbSHjHcew2/1IYZjhdYX9Zf95P/5u9+OmiRHDy/Wxc+8icP+6dA/7Z5swyyfrZlRA3Zq5sg32me2a0HTlL72kz3+8X7fr/EgwT7wogcLue3KRGWjqIRCyPefnOnCjUdIwOkZttmigycSlXDSoTFgbm7/TYVA3BBHq0AzZQBcQ8yps98mXDxLUM9YE9hgoj1E3Bp3U53CPfbPtUAe0t45RdXYS9hnbymJSJhv3PxIQuaI9umPLIEl9kXl6fBgisz6nJEag59A652csCPK0XToMR0XUPx+S7Tmbuk5PT185O+qA8yUpQuqADtLts00z0uios5+cP9JZZ4suJDHZ+WGM68pfupc/jUSI4+PjV+xmxPwoPHMnuZb1O7ShMiiPCxVOIxm1ReDkBDTgpm9pl09IZ/hMw7fXpg66QfGkyrk0fHtyT9D0xdu+aOBk/CMHQNCnXrsx6JgJjZwpH48J2asjDrMV6oE6tj6Zvsh16Qu3RnXOiq/IYNt0ho4AbPOjKII6t4061ULwpi2k/3ZodtLUHe05TiWOwiTD40NnwpsvoicDU445NdCREJOf0dylnplSIeBR99yXyzGg/lOX0ygjQkj9jgA45V2AOmIQNk5Pro5dspOpx3ei2OMRSBwGdeyA16Z04bmNm4NDlsVBpCFE0fT6+e4wjx47jJZysm85boAyE+V1HSP1ePGY++O+2njMrOzhKQvb7JgRr3e6KIbOfLDZHieCw7cOxFhvrmcfvd+UrNf3qtvezK4Chp3NehEw8tk++X/66VfEkHhQR7Q1thUw9RzkO6miA89B2pGBkm3YPuksPJ9McKiTbuzYbxb2J69+Xq/XU587+WkrHT4Z5I1jnk8sb97L39HpTCqu8DMHyInZKc6egcriNSwGxBGAcKLlOhpSBrkbSPZ7uZw/4JaTyPlgDgQZPOvlbw4LR2DQsWkCOcfDnpz1dsBrVp16OT4GEMoYOen1Oal4WygnOh0tJwttweenH3l1tR0WFzM91tFd+kRZcy31YTDqGFd0xA33bI9jGD12NyJ0kYqZKoE1fSZLtyOkTju5Mx6sw2Gw5eyYWfdcA9pY2vcWuXzv9NWxU88pzjfa0tnZWX369KkWi8W0WPbw8DDpstMDdZhzuvnLHL9TDC7vyqEaoZfL5fSa1W5zsoEyneGT6jtAI1iR7XDSd7IRXHg+leTwO9d5wz4Zz4jdUeEjhsbBoWwECerKRkmn490T9vQ+3xPP4DsaK4ZDnigcy5G3Jru1M3VdHUONzJxkdsRuw47bjM1gSkfBNlI32XP3OEnaLtMWbjv18E4kjzHHoJs3tJcsjLEPXDgj8+4YsnOmnLeUJ3p1Ks+A1AG/wZ8RFvtKvdMxjuZ5xsAkh04rTwvLa6pXq1Wt1+uZPbNdE0USsNQ5Yqf/TwzVHauqiZ3m4atVr18rnEHs2Ka9b4rB2V6Nq3k0UjOaFKcd/JuNjvVR1nwarM2QO+Do+tUxEx7v2LiZj9ugAec8go4NIcc7x2bdePJYXoIg9/c5NRDDZY7TbfP7bvfjzqGXl5fZ09PN3gyIBpOOHHA3hfXRbalhnVlETZ+YM+/G0eEwJzX1asaeayMbHUDaJSEI6LBfvD2VdsI842azmd14QsczAk7eIko7SirQOxFYF9MVnV0atDzvSSrSzuPjY63X6+kVLAFYnk8bIzO3PTKl0q3BjMqbOVSDRirPyhfBygbagVt+yzEKyGvsuWlIZn4GTLJLAhG/j9iiJ1IXArAugg1Z4siTkVFGdoN25Ah7pnclO8sxsmqOWwBpBJouZpbULfvrOtJWQmCuHCcHyT77wSEEOF5rxme7YPGEzrkGl/zuBbeOJUZ/HiezVgJe9/Bh2lQcjeujfrpxo42RRNi+O8dXNX8houvN9ey3757zrgGSiHwGF7gVrXOUOZ/Mk3LwN49dR5DCUKn77FCwk6OuMi7djgA6/y4CGpV3hfxWCtHb7Iys1JPWjKrzwt1ET+FKo5kbjbJjc/nfLJbnG9wJih1rzXcyOYN2N6Ft/F2brM/OY5TCIMPy9dQ9++t6qAfW2ekybfiVN3RYGbfoj2Aag+V+1M4x7DPiDnSrqp6enmYb+A2CYXcmAWSi3NbFsewe6zhKXdA+aHsEUMrFuu2g2V87Up7LsXt5eZn0UPXdaQVs2L63hHVgnnNTf/oRubO/16Soc5AeV85l63NEFvJbyJ33b0de1hVd2ZHSBviZOjpA78pffgUKBSINtrc2Q+vAtWN8DNc4qARphvIOndxGx4I5yJSFMjt8z3cyGzKFHHOfDcTs6wgsR3J1jop9MWMhI6GReyzJmj1ODAszQWmsnUNhPXxugW+hTH1h4tF7fsunAZ0Tn2mgjBnrcNjpW0Wpx+QpM0l3u92rB4R4nA26YW/sT8cY3Zdur2gHvnYgZJEck91uN3svV37rvuf/OLgwbTK0DkioAwKpWe+IpXoueoyd6jAmcA4eHR3NHpji9qhDtm32arKV8aUs+8qbgNop0+BJb0jB7OVoyATm1E1P4GsIMlYWizdRd4PbgRr76rq7evKdIW3HBC03/3dYR91GH77eOiPLTJ2UhSwm/fBTg9zv7n/LXlXTo9fMcjkGI48e3TkEjnyeZAwz7cz2bfpmuoEOxpM9gJTx5IOv+dZV9j/6d7oh13MM+ZxUsz3aSndbaADebZudsj8eKwKx66HTTX/s2Kgb10U7slzREceZY9iREZMC2sxoHjHE53wwA2U9jJJcL2+AMYnbV95966mNIGGCgZPFuRYPDnMyUQ7bs5H4dxu5ATnHOblTujAtyqeXH3lyt+OJ6rDH53mV2v1kDot9eGvDMu8U4u8pMbyAWXRRNb9hgSFoDI/Ma7ebv+spMgUUMqGYA44skZOTk6zMDoO6SZ1px6yeunEdefoVb78kWOU8bhM6OTmZZEx6IikO5+ycsjCw826tXMO3AnhyE7BGbC91pz3vo6UTIeB0T62is/Yj+QjavrHFdmRn6jlk5+eoIjLbNrzwnfPzTJEUMmyOAa/JJ+e60yYulHtU3rUoRUbQhZ0pXXLXCuJxsymzo7RPJXRMkO10v5mBdudQsayTn6wr3znwPt/bnapeM8zOA1M3ltPbP8hcyDDyojg/cCYyRD5OPuc3OZFZN9uLTHECBpGc43vKR3rtgNTsxzJwz2tkNYN3e0xJMG1Ah5k+ZHKadUXm0cOfc4wg63QV++gHN9u5ew50UR/1nRDYjpmvfzEYjsaDugyZ4nvkHBJ7rr9l7x0BcarHaZTIzz48Pz/XarV69TsjGYf+bIPRQRfBuW8u7wr5uwb5f4fYBrEc48TPsVHo2V2fYxxAe2O2ZbCjJ6YcBDq3Ge/YMWn2t/OeZlic7PmdE5jnsv8dk/Z2JTo9gyKZQIDCjMKAzvacO6V+eIdY1fxhK77DpwtnuwlkMI+8ZKedo+Ski5wEmaofkVM3QanPgLXz5nwoC3VfNd+KZVsMW08dvp3WobuBNQ6S9dnGU2deV5Lih6Okr0nb2Pl3jjztZBGIv0VmzmfnxdkuZe5ITBeJ0XboZOgk4sSjJ2+1JH4Rjwi4kcPOuCNdLu/eNsXO2svQe9vjGUD4G+uxp4lBO1TpQCul21rjp8Z3np/XkOlSRrMcntOFFa7PnrvzzmzPnnp0rVkKz6GB5DeOk42jA8pcZ5bDa/0gGIbv1D9TBJ4UBnTqvGq+z9kOyH2jHuhwuQWGNstJZuZKW+rYVUrYUWyQIXzHvu2cPJE9gXl+R2oik89Pv711y9cHKBk12pE4zDdJYR+jA+rL5MfjzbZoz+5/Ss47ODiY9qCmP09PT9MYjPbDd/POBIZztNOty19+fB/zYjRQssMIawCw8GasBEyGhhwIFneOiqC3JJu14XaydsYxWuXtwI7ncNAdalBmGxg9uHVHFmD92PhG9XuSG6QJBmZ7AQ/qhmyNgMn+u08E1xRuTaJcXvB0HpXvcDLrSeF4JyTmA0X4lHYyXqYBUgcfl2jwcYplxMK7J9UbwMmycl7H2vg9OV5HVN0cSgmYehyirxwzYbKjtZPh/5Tb86nTFf/vnAXHI4w/26ceHx9nfaPM1DGL26bclmlU3nWnVLwTGZG3o3TFHp112RC87cMdokwds6yaT+aq+aBxElOZDlcsP4GTv5uRjpRsg+rYRmSycfI31kdnwzZYD5mS+2S5IhPvg6duOe5kdA6HHLIyRxumYNbGsepCw9TliWwZDNrRhdMiBM+q+bYu6tOO1TqNrjy5qYd94MLxpi17rDp2HB1aTqYEmPoh4Kc4ciRQe26mPwzzLSOvp/P2edSv+2ad0JHaxm3HBwcHs4ejZKyzGGh50kYXWVr3XapiVP7SO6WoLBsBgacDGk8g5jk6lpTikKkDW4PdyANRJoZVHTNI3WRa6V+uMwv1HUwxWstipzFiM9SfQTdtbLfbafV6u93OmIb1ZuZpptvlcM1+UwjqrJslv3O7DcNFMj/Xzb4yf23deQxHOibI5Xi2Q2XcOhacfnTAZptOOxkTp2FYVwcm3SP7usiCuUn2P7IQGDsgSd+pC49fgIj68CuVWTobMGBbb93cr6pX88i/s+4QgYODg/rw4UNtNpvpFS2pK6VbvKR8tuec59to/88M1QwlA9Sxo6rXzKKrh+fSu3ShGgHNm5SpCCe/OyPsALdqrmQCbccY+Z1Amjq77x2wd3K6HTsbMrgMOL04Uya8lm0kTbMvX2nnYn3xt1yfCdC9+ygAYFvKp3VC4LPzos5oA3QMBBayC+4rpGwENDp56oN1pZ95KLVZWfrZsXjqm3OlIyMMYT2fqEcCWcDDusw1/O3x8XFGbmI3jkIWi8X0ECQyVD7oyKTKfaa+u/SdCZuBkzbBuZL2Hh8fa7fb1enpad3f38/C/RH4j+akHcI+8OzKXkAl0JF6B+g8IB3b64CU4Ug6yM7y/NVqVcfHx3V/fz8lnTsmFdlcIpcnpMEk/5tt5LoAEdmejaUzXudt+J06Yntsk8cNSOy3AdyGbgfgkDTndAaV9hnydrpnuLtczp/t4PGhjAzPnK/s8s+Rn5OfkQT7m+9+qDTbia7Dysjko/+0yTElg+VOh/zRkXhvaLfCb9mdh+6iKm6B6saQ40Vnataf671LgSv6tM0uCh3NPzqMfSBrp8M6u7RfyuPjY93e3tbh4WHd3d3NHoBNnbOeLqJivzNmbPc95c23nnqlsyuc8A43nbvJoHZhUgoH6Pj4uD59+lRnZ2f1+fPnyftwkAzCZB00MA+8+2R2Q3nCALrFqS5pbY/HBR6HDwRU3hrJejjZXDpgNOuwjJxEZDdcyHBag6BunZAZp146XctPh+J8r8fFeUHeTURgIcjmO8eL13AcmIIgsKZv3Ivapaqq5m9oSD83m00dHBxMKRluBeI8cfRB3fCP/QiQJjowEIcE2fZpeyQ1tsXOZpxWMFhRBtujwdYkwc6LujCoObJZr9f1+fPnWiy+PwuVUYJ1SBbO4r5WzRfhKPO+8q5FKQOehePxXMdJ0YWTow7RODi5P3z4UE9PT/Xly5fZxOtCVMpOcHQYRnk7Jplz0wYXMLifMHXQ8Dl5OsPr0gyc2OyPB5rjwk/qu3MKqZNhZYodhR0Er4+88eBkbgzPcq6ZXeeA3D+G2TZq7hElCHQMx/ZBB7tcLl/dq398fDxtiN9ut7O7mXj3DftB+QKO3Yq4baEL+x3y01lRj91NIx2p6PTtlEsHeHZG7mcHop737JOPe9yijxEuEBQ7ps95R4B2NNWNGXVH0mM9dv+7vAmo6aSVlt+tlPzfDXjnBVKcP4rwi8Viemjs6enpq1Cyq2/fMQICj+8zhsjHYoOi8aX4GJ1JBq0zSDNrgqNTL/lOI4i8u933l+2lPeqFedfIyknWsQLrjR7bYB75wqaq5ttXmIPj/k2CXsdUyT7tSNMmHaYZM52g66YzS7Gd5BrvxRx9p4PJb1x0SZ10PL6ZgONJG3BhPewLZaHdeG82xz6/UUdmwSNb4XV0jhyPfXXnOs4DX5dzvdWLO0sMfqmTfaaOWejM2I+3UgBvLkp1DRvxvSKY41X1ahJSMfbCHRuJF3p6eqqTk5M6Pj6uzWYzO3cfm6MsBAwDQgonTurlQ0A40BlQe88ALFmW80Q2SB7rmEv+7/bvmWF4QzX7FUBhXQZkM2DXzfaYWqiq2e2IATFOHjMz6z19dqqCdVX92DlgILfjY/6vY1i5RdcMiW/djVzRHVMD7IN3BdjpUafb7Y+n+XfhLm2fbbCfdhiRwU6A997zLQSdXOkbWaHbZ3ssBG/qoBvz1J36DMLsO8c9ttXV3QE026RuLfeIeY4I4KjsBVQLzmLP6Q5EEBo96XQXBtlQ0n62txwdHdXJyUnd39+37IQKpoemfLvd7tU90ykEQNbHie0tHQzJuCprhvv8/Dx8YpBltY47780J50kWOdgvFoMPjXkE6pyElI9jydcys+7Dw8NZPo9P8YlDjYO03qiXyMBw2myU6QjaFfVN3TCE7myBDxFJCZh2YM15QIedrTcBT94CSgfYORn/z3oMeNEB68x999n2lDbt9Lr6Wa9lMTj6XI4H5WTKzNdHDkdcHbbwmO+IZGGUbb1yrFzYVtfPrrzrTil7pUxahwr2jDnXnoQCd21FAc/Pz7XZbGYKOjk5mbHb1JMJ0AG02eFIDntN5uhGk5NK7jY+57zsSwzDI8NlWEzD40Z0T1i3yX6RabJvHYBW/QhLO1CjE+FYWpeRY7PZTPWaPZkVvZUHzAb6FIaoXLGOPAFnpgOoBwNfSmdP3YJOZEib3Nyfc+NM2CYjuBznhnPqh3IzR+1tX2Z2/GPbtNtuf3I+I6cdp8fdaQPrmYXydMTHIEx78nlsn/9zznJedyTQbNwO1Oc6ojGQd+VNhsp8CkGU360cg9Y+wW2c8Y45nvC+6vsEOzk5qdVqNaUB7JG6kH5E6e2Z6bH5mf7SYOhJ2ReCe6cT5gt5jSdD1fwhIx6XgApBL+24bqZWOEHTj1EuNG0RPNk/nm/A2m5/vGepG2+2323DIqvw+PB36iv2k+8EkeVyOa2MEwzSFzJHPCbSCAAAIABJREFUsz4yeo4jJ31u9+RYMlTt5KWsHL8wfbJgzkHq3I42cm42mxlbzbV0kCkGYNZPhuk/E5XIwTloQuC56LHtQIuydkyySylFdpII53ztdFhST7erZ1959yo/jZMd2ReqMlRyPsrCEbg5iHxy+GazqePj4ykF4IUUyp28J9v2IHEQaBQxQudBCZhVNWMoI6+dQgAxA0x/rXeDY8fCmT8y80i7vobM21FG1eunHxF0008CRM6lAXrfJ2UMWFAPqcuslec79E5fcocRwZN/XdjeTS6WrP5zaxL1MMrbcaw4nox68ukwlHOAjpLAnDHjvlfKHmDn9Tnf89L7nJ2mieOzzjkukakD6I5xdnrL+Bo4RyDNcTOzzW++tusDo5yMSQe8Hsf/M0O1AsgGKWheP5DjnARedbMwXeiSkgF9fn6u4+Pjenx8rMPDwzo5OZnl5WhAmdhmBt3KH706/3eJQdtDd/2I3hyO5Hzm3fLJwXeYbQbWAbuZMeWJLrrwiZPZBk8d5TcvnjCSoJ2M2HtKQIUhrDePc/GHDyKx/VXN3xHP9hyhZJJ2e1MDIrEXPhmry8/ROXHcGd7vA23rKw7Ir4CmTTMKIBiMSEXajUx8pTXlIRvz2HnrlHWazw54CeBkwdYF50L0ZRJkMsbo2DqmnC5m8qyTWNRdY7115U1AdcdN0akIgxmFZN7CYUDOtZfJQK3X67q4uJiM+/T0dLr9z17a9N/5RIcCoxDE7LcLRVisixSHoCMvzTCFuu7YOhmLQZHX09mQ1YwYMMGRrJAhIjfQE6A9bnRkZFkZk445UB8BjePj41osFrNtVrEL1kFml/A+9Vg+Op9uLNgWZWEaowulq34sNu12u1lOvYsaUjdDfIM/bcETuYsAbOOdA4jsab/qx3uhDPp2+FXVOrh9QNbZOa8xYBErTChMaKgvtpdriT+8znZLXeScbqfOW+VdD5imkNxLyMS9G+wA0t856ToFpe31ej2bnIeHh7PtUyPPt2/iE9gps4/lXF+f/zvPbJDmcQ5sBtyGwzY4SSIbDSVj4A3lZEcBIBez5+5mhJxHxkjDNivmODCM8zjRbjKulNkOjfr2Yg4jEKdFcl4X3qYd6seO3ky0u53WTprHI1fq9WIXwdTgRxZox0O9JgocLYpG98nJM/9Lu3LdkZfHOgBkO7nOtkCd8Nr3fLdcBu8RO++iLV7/Hhkyrp3j7cqbt55aYBpY8jUedA6WBTWt7nIbbH+7/Z5g32w2tVqtar1eT96UMtHzEITIHmzIVDBBzXkfG0J3vDP86CmFIGK5c5wGznuS0zf2iwZMVmR9pL85N4yPjivX2rF50nF1muEpPw3ofm8Sxz6hqJ8An5SO2dkoYiBzT71pg7qgjXZsh2OY87klrpvYtlnW2zEv2orB1KSENsaxp4Mhk7KjZolj8QvtRuzSc5r69jGmCKgfO9Kc37Funkcb6TCI9doJuJhI5bzoj2PnNrxuNCJLKW+G/L44hpvJ0YFnd92IzVW99gpmN8/Pz7Ver+vs7GyamKvVqg4PD1/lhQys/s42aITMn7HdGGaYDBXscM+TyNtUOkDlhHF+0DlrTnwDRa7nceqehknGyZAvfWK9lNVGzvN8fvppVrpcLqcXqjFPPJpAnDAEa/ZxuVxOd4SFfTG9YrBg6oE7IPi0LDPYyEIQipzeYUG2zIeLxImlpE3LmrrsFKmT9MHb8Xi+5SFTTURDvTJaSrGD7VIRtAVusTM4WraR87ZD6uZv56yoo7TrXRHOCft/9tnzscM6lzfvlBp5liiPwnMvJTua3y2gz02nOHg57+7uri4uLqY6CKgOmTsvyj50CuGEszGEmbzFODpPTAMJSGYim3XlmkxwbsLOZ8dWO9YWwDDI8ZFzBka25RVzguLIS6cd9i31dW1U/dga5nGh/h1mRz6G/txC51c4c0KxfgMzV/RTRoQh/WXKgmCVjfxcoMoj/3ge9WUmZaaX8Xdob0eRfnFnBOsapTSYzjEz9RiNdGR9G4Q9Tzsb7PpGuQ36tkenFEesn/VHB7me54Q85m8ka9U7GKoZDxvJKyRGq6YevFxvQOsmkoHp4eGhbm5upsWp3P2xXq9bFpf/XTePdd44BtDlC1lsxGzPuScyBeqFzInA6DAsOvJK+GLxY3uYQ332i2PAJ9UHvDkuAWHKTRlHE8MOKefENrwSz3HwuFAuys5rqZecyxQJr2OONcBExpZzn5+fX70WJXoPa+W4mnkdHBzUarWaxoFypd6wSoeSBLH877vrvBXN4Ba77Z776nFmLt9ON3WZOfJ/67cjF/6/c8pmlLaHEUkZtZvjOdaV0fEwdKalqD/Og1F516IUQYXhTpcfZO6GrGzECkdg6gF+fn6u29vbOj09nVgN5eo8Tvf/qLh/HFzqYsR2XBfBZN81ZuScaN1DYKwX5s+4X5J9otzs43Y7X2TkZOPk6cae53YMmMyWddMBGRBybfrBPa0xZhp7+kymkz6NiADZmsfA4aEnlseOKRyPHdkzn1lqQtHpg3J2euuAifrgOLG/dHrWgR2d56BJAwGZ51ge68044LnSyWbiQfsj3jhF5gVas1rPJcuRYhvbV971xH53LJ+bzWaWf2LuKYUDWvU6f+gO5Rx6whhsPP3Z2dkrb2GvRnk9sUbg5GPOm5it0sN7sDhgI7bAdrjQ5kHrDIGGRbkJaKmLE95shkBC4HQO03tQu9+yuMRz+VoPAl3nNNlHA5lv+sgtyYvFYmKWTGWweCKEaYaNp94stI5ANLpjeJxzX15epttuGS0QpJkzdd0ZI4+15TawcQwcynNMc82oX26XqRMzzo7E7HOYnqueRzlnH8vtwJj/dwt7GVfbM/v43vKWzaa8606p7thut5ve/X18fDzs9FsTh4J2d8Pw2s1mU1dXV9PGfg7uSFH78mgdWHaGGNDhpOYf2ZgB24yR13S5U8pLVpX6zDo8LkxVcALaeAjKZp+WzcZNnaQu5hCr6lU+3Uza+So6mQBPGIMny3K5rNPT04n5JTcZ/XoPKtlm+sfxsG6t08hB55NxzW/eQshzSSJG6amumIlyHO1MaXtmZB5L1umow2M8AmQDqAGT53uu8rzYmhmg56ZTaJ1MXdqC4xL9dXrv2Grq9+uX9o3buxhq11EyRh6nMTsn6LrZBjtslsmVw4eHh/r27VudnZ1NE8ogyoHLd+d6qbSq13e1+HcyEU6cnGcQpqFSpkxsAgzrsMFYR1ws8MZqT3TXN9KJdZ76OJFj9N3uAIMjWWv0xdSC5ePEy2b+1Wo1gWnngFInF+yyQBmQDdg5X20mQ1DONda770wiSBOcuDWMY8h8fOfEmbumzTlMpw0xPUIgpQ3aAdNGqEsDqtdFvEjKPrBe/2/S4zkyKgbMEbiToNCWSHg6NmwbcHE0ty+yZnn3xn5WRq+RhzDQ8DjoVAYNIUrpQk/Xwcn08vJSV1dX06ti93lhG3TnTT1gXZiTYrbhEL07n/ITWPiMVfbXE85GmTbC8myUHUOh9+ck9p08bp/jlzY5OXNNJhzzmOljdF81f/ZB9nWmjYDp0dFRrVarWZ6cYMSV+LTJyZ+oKX+xDfadD0DhCi7bqPo+4dOH6NG7JOygCJB2vF04bJvMd0dW7kfGw+caLD0/aBMd4yQAkUESnBwt5PpRnV0/u3GxzJEzxXVwnLqFcYJvR5hGxQSjc2qj8u7noVJA5uv8sGeDTidofsuAmf2yDV6b45kQ9kqjgegAtPPcLGYiBA/n6pz7ovweDDoenkvjcqog15JpdYXGzzt0yJ7Matk+nRsZece0eB37aybPV/B2IJHvR0dHdXZ2NlspT47+6OhotsrOiIVtxsnlHU55pXCAsNvCxQlvRsinU3mfJ8ePe1XZf25B42q9ZSeQeaytJztZgl76YMB1XwmaGaOOPdvWzHjNOF1GIEmbZL+7uokXo3aMA93vnEueV2yH+o8OY3v7+pryrn2oVi4Hdb1e12azmV5PEmPrvE7HtGgUNAjfhZPOkQ3k+lw3ktcKZJ0xKrOy5OLoKKrmQEgZuv6kntEgMLeUc/lOevazm0i5Pv0k2JHxRlY+OpD1kr3mGL18znV71jEZKt+/xMmbic87t1ar1fQ2hvz29PQ0OessQI0A3MCR43n9cUA1IOdIiSWgTQBzaqfLAdNZ0l44drQlgpvDadpj2nOE1bVtIB0xONbtCJCFbdkm/JsJxQhwSRzs2PN7Ptm/ri7KTaxgGuY9xaDNNrg9dNRXlnc9YJodYImxO0zuqH3n6XKMHrIDwQ6QfJ7bjDzOZbHOfcy2YwksNFbnY3Id9152RshBYl868CDoWc8sZux0QAS97i6sEYh7/2tAwLkojnHHsqOLxeJHGB9mSufiRSbqNIXgstlspglvhxHG672ZHveq+Z1MdEzWO3cysC3aQ/RoYLVdmzWmfgLFiIHREdO+eAMAx5VgYGfLNjobYnsjosQ+mnXyO8e1G4foMXJa/vSRkaPBuYuYR6Xrc5xQp7sRQar6i4BqYatqus8+x80e0yF2OEISaLgVJed1gGGDN+OkgliHB7kzHrfBSWtgIMPqwLwzSIO95QlImSGZefKTdXk7FlmNmYgNkeOV38NoGdLy3NSd3Ggc4+Pj40z/njA5dnZ2Nu0rTj0Bx8gQmf3k/tQTYOgmDZn7arWa2o18BwcH9fT0NJss1DcBlqyKzo36dEqLoXiO2UYiu5n+qHTgZpbasSjbOse5A0UWznvLb2CM/XY6yTHPa9YxAnJ+t65CTugUnPdNnXaEXX84p9kfy9KVd72kb98gPT091ePj4yRIF+La0PjJnJ6ZEmm/QS7K6TYkc2BGHr4D18jVGWsKV3urXud6CWxhTARke9yqmm6b9CRlsafkHVIZdO9icFsOjzudMKfHcaMd+BF27AtznBxnnhO58xhGs1myLU4esjZPZE+OUcoiW/xyLGDOELEDFI63w/RubhAkaeOOSmjX3eT1vOra8v/Oo7p+51opMwEv3/elzOw4DTi2MY7LqOwjQk47WP5OLzzPC7mdHLYjRy7WG8u7FqU40fhbQv77+/t6enqq1Wr1iqVWzZPZ6UTq4uRkB0c5Hcrl+ju2GQXQW9lQqPgOdPPdW2do4AbhqvmzUO1lWXfkCEg6zO3Yg1/J7IdsdCkJ1uO8nNkN+0HAs0dPG3mWpplkHAV1kbyp73gLKJJZchLlQeZ8Kr1tgLJxXMOik6dNqopj1+0aoY04tTMCRf5P/REg2E70w7Gjvm1n+wjOPvniDO206XA7MuOcru3Rc48Mr9vWZZsczT/nUNneCJg75zICXqepquYPtsncyuLkPieQ8q57+bm9xqCw3W7r4eGhHh8fp4cBhz12AnTg4DbfKqTlrHOUOLaB8RwbAgfe9YQJESBSPxc6zMpGfbaM2dPJXCr1xk8apG/tpUFTP+5ndOaJlH7b6Dt27Q37fieTmW5ALXnT6I730adeysH9v7wVdbfbtSDR5bXzeXZ2No2lbSH9NZNnXbvd/NXiHk/uj2W+uRsbAxjnDe3KsnhMXaI3g2Pqtb1YX96/y753jJT67QCYvzENMiIwtr3Ikf7Qxlhsy9QZx8qkhTrt+ufvo/KuJ/a7QnZqt9vV4+NjPTw81Pn5+av8T8dqU7hxu7uGrIaffi6lPTPBbjRI/o2D3jHmDgg5cAwFc50ngkGOOuY1HZDToMyI8xnQySPdyJZoDDaIkbF0QMKx4mQna+xWlwkYYafL5XLKm+YmEd6RZJDnpOfCTVecEkj/cmff8fFxPTw8zG5E8XWsiwzTuiFwdfeC21Y9wdnXbmw892h/HTNkn5l2IBGwDfktBSYoOUYw5G9p3+BE3VguOmnqYzQO7znPRIvFTs1z0n2L3ri7qKuX5V05VAMUw+ewi2y0j2dzTisKyUSnAXfeMZ3J/xkcGjfPiaIIGDa4yG+Q50QhO+Ek5rUjz5hzWDeP2UERKMkiRiyGv3UM0A6Gn3Fc3cTO/yMD7CYtGZjD1MhEYE17uRNqsVhMoRTHIPtOA7ZmPomYyEjZz/SRNpu6OImZTjDzIWsfpU6Wy+XkADodmhHSRhzKh9E6dcNwmfVz9dn2ZnbYOetu7FNsQzwn49vZvwG4s18y3vzmSNZyeS5wPPjJdjJ/u8c+kmjEnjqSkLri7CnzvvKufahRpg2Pxx8fH2cLIxbAA9t1Mtfl/DAae1t7PW6a7gx8xNI69tnlSuyp851GYRbB6zqjYR8M+LzeeULLxOvSf04qym1n0LEG5/IieyY33zhL/fFuqNRNcMhYcjHn/v5+Bhr5LTLmodF+IpYnCycsxzrtem+nr+sYP5nnaJKb2dMOqd+MHfVM4E1hP9kHMnQ6Dbad833Hmsfe4T7HkPOL7Dn1sh46QtZte2T/ze5NEtgfjoNtvYuCrQsWzg3ONdpm50g63b8Fqn9p2xSFS4NV3zu9Xq9rvV7X6elpe80+QWxkZBD5v8vFxbgNyARtX9uBqXMyncGN8oxVrx8QzGvMcHPc4OstY6Nj9tBmL/S8ZImUj46REyXM0GPlyRB2yb5Ez1no4bnRHRepki9NLpLslkyj6sc+ZT6dPgCdO6JyXjexKEPkzR1V2aFix5Y+245oezlOPXQsMP+nrwE91s3zmT7KMTsLA+sIFDrQpfw836EtHXT+p55cj4HO9ZtBd+DXzSOm0diumb/PiX1k3M1GXY/rStTQ5dNH5c13ShlcyE5TAqi3t7d1cXExy+EZ0Dpvw8lBBXrArDS27zDTbYwGpesn28lgpJgF09hY6DWrxmw15zKc9OB1xzrGEoP3rZ6Uhx43Ezv5z4xPd796+moD5HHfOWYASH3b7XYKpahrRyfRm0NhtjvqrwHPDoV9pu2l3WzeJ0MysDCEtT5sMw4zO9BL+6O9v56LHXMyO7ZDNpmgnTK6rJpHXqPHGo5A1PYfmbt2zXS7a7rotHNaxCc6GqfkqEvfRZZ5wR0l/78AagTsNuq7ZD/q8/PztIWG53ZewOFHJhtzsfbAbj95LD7NnMrpvGE3YJSJv3HQuLJLA+B1TENwID3wrJvA6bDbhkSwGuV6WVfXDzIoT0yyVvaH51MmT8guvE6dBEnvBjBQLpfLiUWy+K27vLZbYHIKikylYyt08CleaKK+WQdva7W9eyxNNDqHz//T19xs4egn5/J6tpHfmUM0wJlZd856xNrpXNIOnQLtt3MCHTPluGQc7Hj5aZBlHS7vAUjaWMeCu/ImoL7nWBp/eHio9Xo9PdwihkL2Q0Di5LDRjoA7naOhcLAJUjYqg6qBerSNJp8MA9MOgTS/7WOWHWPN8bAA6phM0QZD48wdP5S/YwjdMUYPHK/IaYdk/XO8MgHD8Mj+2B/qhOOcjffWKa+zsw04dzsK2BenLTgezA1Tf4l6aFf7QttOX3bSXPSKvnjXIHVjgMh11J37Qnv3GHnhKJ98Ctqoj8zPdoBk8GSoTFDna6w7HaUsFotpTpAt24ZSDwHXTDdtO23T2SZtn48nfQ8IvyuHasG6yZqw//7+vj5+/PjKeLuJzeMvLy+zTf72fCldeGfZuvcrdX1I/TzPiy3sn0M86oBgakC1EfB/65HgxrponGYNvp51crHILMR9IruywXMCs//5TBht1spzWA+BNotP+Z3jNbrejLPbVRK92RYDDJykHneyTAJpx4ype+qD491d77uYbANd3+lcTB66tlif7YoOmI4w9YwAiXZDp87zDHodOx0RNoKcI1bWZ93Y8bhev3U2tsD/2e7j4+P0xDP+PnKoVX9hHyonLgGR53L7FBcQmDLo8k1V/QvYDKZUTle22x8ry3wdS0fXCVw0TIMwGYCPdYbeXTMC9I4lENy6ULMbI4ZgZg+W0cCUMeIL4agrTxZOVgM5nR2BL78lNVNVs21MzKV64qQO30BAm8zEoG0FsJk+sa7DStNXE4XYdPSQcSE7Z9SV0oX9/I1AHXulnjm/aCN05hwjyt+x1bSV9jxu7FeOuUQmLpTlmG2BAO2+d2BuZ+30B2Vy/wnQTr8YII1bI3ad492CVJdqYXmToXaKtDA57+XlpW5vb+vu7m5211R+5yQeeeQIzUnujjuk8WTJZGX4FqVya5eZXNrM/7wufaZS2a6NrNOhvSjr5HkGV08sMhS2wXGxkREsbGgEiA50vUBCRhedsc4AJ5ksnSrtgdcFXJiLJZvu+hdAM2t2KEzWlNuk8xQqymZnQEYZ/XDsyXIjHwHKTorgyUlPh8e+EIA6kOR5nBsca+uchXeguU4TETqWkePuzqfdsi9mqqyDRKebi3ZgI/u3TtmO9Zk645zzaNLUY8fXlTfv5U9DNDYamAf6/v5+ttrPyUcvywkcJSTstwwEEX53yNR1NsoMiDIdYKNJIRsaMbIMuI3XMjp8psMwm+DvnbF1+bN8po+8zuDrEL+bpJ1hMuThai+NnvrOuc4Lpi7rKzJl7BnV5DqPI0N+RjkdK0wbdFZ0yAQVgmBkJKv2PmfKR+DnBEw6xOCbOqMvjzXnieeidcNiMIlOc33q85yyHaZ0zM9jHvlHc5DHTFAYQTidZiLQ3WTk8TADJfvuigkeiYKdgd8v5fLuHConYDeB0oGXl5d6eHiozWYzrfb73E6g0ZYNe9S04wFyzogDSyBxH2hwmWyUt2PlHDjWZ8MxcPt6y2embhbn9giMfp0Jr+n6m9+s78770nC7/Bbb5HajGCA37HdblMgiOoZcVdPCE68jwHpSdGmjEYvhZDX4G0RNJqgj94mOwTlFFkdYBlfL2aWUKENnt2TiI324mLkS+GxHHZP2+BKcMidyG/Bms6mHh4d222M3f933AH3nYJx+sI4ta+w2qSg62LfKu9966vwHO2WPeXd3V3d3d9NT/M1qbeSjgeg64LuHLKuZWNX89cWdAVqO9Jf1OnQwoHU66zws+9gx27Qx0i8nH9kmzx2N21t9dh2Uld65e6B41XdQ5jYn9i+AxS113OdJ0DBboQweE7LLOMTI6n4GdHMrofvAdFDXB5aA776oiN/3hYqRKekO2ir7S3Cm7dGWOI4cTwM0HajldV+sc9ocSQDHrnM2tqmLi4v65Zdf6ujoqP744496eHiYye/te2TV1p/73Y3DSDeRh5/ZBur23yp/eZXfIVfnbdfrdd3d3dXl5eXstRaj3A87SIDiokI62ymWynBoTgYaBXZsocur0nAsrxlCN0E5kVl3B3zMq3WF7CDX+8HePpf9N/OzTDyXwJ7+MNRhfprF45PrCZrem1j1/SHlYbFcrOryX5TZoSiZplNJ1E0crDfvG5jySXsjGDFktrNlHzqCEFl5NxjHhmMYnZkpjZy/22DqxXNpxFI7J2yyQpChTdPWOhKzWCzq4uKifv/99zo/P5+iD9uF5wdtqyMITmNk7Gm3rIvOgLa63W7r7u5uAlSTnH1M9V2LUt13Cm3Ez+LUer2eAHU02GZgHcjQcHgdlcz8GY272zDdeS0eM8BHFspE4xrpy+3auJhH5d04zO/RwRgczTJ8HvXbOTPripOfrMqAxvRC9i3yGIEx45K+pM/cIpeV56QDuhAxstJB+tbCyM39r2a5/LStEeB4LDYdZ8yJx/8NWNFF1fz1KmzfCy5xXsz1R8cjp0/7YiFIxMmQ/dtOUnfHeJ0353UjULaNpQ8B048fP9Z2+3170v39/auHmXTssWO++W7QNmuOjB0hoPPJePHhPSZMo/Lujf32Tp7INvaHh4e6vb2dXnExqp/1pqMGtarXeZD8ZsZDIMgE4VYcG8O+kN2KZB35n/nCXMM+0Tg7purcn9tysV7YF49DgMBy87tDGYKDc4pm/2RXGWM7GQLky8v3++5PTk5mYL3bfd+ZkVeTVNW0+OW6bBsE8rSdMbFTpd6Svzs7O6u7u7vZW1FzDdk1GSfHyLqjk+hYDYEpuovMzLkyOjATZX0cy1GuMro3UaBNesO/5zb72AF6Z1ueuwnz//GPf9SnT5+msPrbt2+1Xq9nEUFX3hNyU3br3PpjH5bL5bTVM/tPuzRT57hY3mSoBjGHP5zIPPf5+bnu7u7q06dPU97s8PBwdm8sQYcdHbVtQ/HkJcB2xkfvO+qnGaW9oj25WXMms7d75Xfve7Ueu8UhAjJ1vU9X+d+emIBAtshjOY9gwi1lLKvVagIsbqXqwvSnp6e6v7+fFiJYN20ioXiAturH/fdkpF7wopMkg6cOs6UrgBqWdHd3V1XzN9lGDwSjblITWNM2wcpO27bp37PbpWNDdijun+Vy/XQoozQVHVMYMx2jz+f8Mg7Qps/Ozur333+vn376aYpgrq+v69u3b9Mbbp3Djr5GBMOLoiYA7DMZvuc721qv19ObcnN9p9+uvDvk75C5Y0Q5/vLyUtfX13VxcTE9nd25LA5qF266foJbl//pZKQHdnsdsOWTwJ1jo4kxOmYvGPDwLXwjoKSROA3gFVuyAJ7H+jpHaGdmr8wQr2ODnKDduLhfm82mbm5upgc9mwXnLw+mCKD6TrHIE7vJ+6LsYDm+BwcHdXJyMjv37OxselKan9VAGzIIsl/RI5k4wYXjkGN8ySPHM8Ub00kG2C/PT4NnNwasg3bDfhqQrEvPlW7MCVwnJyf1+++/16dPn6bFp5ubm2kxKs7OOicBICGLHXA+26m6HywG7vyfdOXDw8PsOah2HqPyrkUpD2zHNqlMTp67u7tar9d1cnIyGyCzHQMdmQgnPllLFDfKiaVeshUyB3vTHDPDqJq/CtrXdp45MngiOv3ASZdrzGysf+bveH23+OD2udeXLJUhKHNvnuw05owNw+MRoAZkwnYeHx+nJ/czqsjvzF+NFurSN7J+MtzYaeQ8OTmZ9SN2lIWRsCRGHRkP5tRiD7QvjlPGz7f70rHFNpjuYp+yZcdbD8nAqeN8951hLARcy2znaydM2UgAeA7bI1ifnp7Wr7/+WhcXF9NCz2azqc+fP9fOe2l+AAAgAElEQVTDw8MrotLlTDkvravok33h3XQso9RfyvPz8/QmB6f83sNU33XrKQWI0dOLE9DMMq6vr+vDhw9Tno1gahrPyVf1+hUoPC+fnTGOwvQU56esJC/2pP8Eqlzb5SgJyGaiZggM7yif9Wl5DMy8jbJLReRaLt7Rq1fNV5oDOHluKZ1b+szVbvaNOcTUZfaXfKpf1+I8IkGHjoaLTg6z85SqMODI/PLyUvf39/X4+Di9cfX4+LjOz8/r5eWlbm5uptCfaQSy/eg7OiA4dXdU2e66fF6K2WYA2TtdOMdynK+PyTHfNMCcssG/s7cQJp5vZmsQ5rw6ODios7Oz+u233+r8/Lzu7++n9v7444+6vr6e3ZlI3XE+UnbiEPVKORi9+UYA45n7nmjF0TT7vy+X+659qJ2nq5rfmdJR4gh4c3NTHz58mL3l0kphvoftc/Lz2hQaCTvvujqwNVhyEti4WY/ZSAf47GM3eDluULOs3Z0pbCN1kfF0zqRjjTQwPxWfwMZQP0DiiR2A9F7OztmGQdghRU6CKPtg2+Cm/tSVazebzbRZPG0fHh5ODCmss6om4OXOAzMjT0wzPP7mbVVd9EGb4/hx4uZa26xZk8cotjUCUOdWCWCec5YrMnSOgTIHTMNMuej0559/1tevXycWTsfksWd/SLZcci3Pof1Qf3YiiWheXl6m203dJ86VLs+b8pe3TVmwbJEx3ae3yEZ/ru46HKYS7fnZvllXrnPS3F4tg5y6OUhUuj0Qj1tOyp7r0o6BmF7ddXuSuJ9mtqyXuUV6bzKmLhLgceqCYEq95bdMsuiSjJPXsS3nZWlLncNyxMFiHZD5E5yXy+UUwuf/sNeTk5NZKB3w9U0ClCt95liZNdlGyKw99rYJsnHKy/b8YJNubuQ4mS3/5zh7G5bnW2TwmLJNFsp+enpav/322wSmkeHPP/+sz58/T8ci74gAdSmfrq8kLgbjjqzlNzqizWZT9/f37Vxk2x5Hlne9UyrFXpKDZaPi9Q8PD9MC1fn5+YyZddeZlfKYlb8PiCg381odc9jHArsBNGATcDoD6YyVchMAc5x9dN/MmLwthhPKOu4mQmTfbrezpz91D/fgw7OXyx9PNme/+MmQnm0x1UH2amfLSCjnd0zFTNf1LZfLOjk5qY8fP9bLy8v0ptXn5+e6vb2d3d5KcLajdHEfcg0/I1/GebF4nfoyyJFxMcXjudPpqHM8+S3Hu7A2DJvtexsa67TNLZfft0b9+uuvdXx8XOv1erKhP//8s758+TK7A2lUbKPsD50ofx/ZRK4zM41O07/7+/vpPWcEZzqcTjaWv/ROKXeg6nUur1PE09NT3dzc1M3NzbS6yknTXcc6veKZ3+l1eVeVgcuOwEAaZkIGa0WanXcMptOTAc3XklV3zJf1sC4CVDwtQdlyR3bu72SuiSkDsj0CGB3P0dHRBErM1fGTzCB941hSD86be+yoYzI/PqaxY+UGQuo9DDahXmQzGEc/rJt3dHVOy49DNEkwW7LDtc1W/ciTmimb2IxIEImMwZa/8Vra0b58ZP4/OTmpX3/9tU5PT+vh4WFavP7y5cvETAluTut0W+74e6ejUTFRsi3yt+TXs9vDKbn3tvmujf0Wir/Te8bAOgZ0f39fX758qdPT07q4uHjVSd4bbTZi1strrRQq0SCV8zxguSaPc+NDZaNYy0K5WW8nc9qgAXug/FvV6wfGeNJELi8aEFz43eyDgBFm6vAufcpETqgcIOqAguNinUf+5K0ybtQxdR/g7FiZ33wandFR8AlWi8Vi2ra13W6ncfYWLUYYATHbQBbd+Oe78rrx53m2J37SKUWPDGspC23BtsLxMJh3dug8pG/jpQ1ansPDw/r48WOtVqtpsXWz2dSXL1/q6upqlptM3Sl0Yu5j5HLUYic0Sm2xvi4y3m5/3C7PLW+8lrreV979cBQ24uMx/FDnjnEmrLq9vZ32ArKTXafZRj7TqUwQD0qU7q08Vrb7mHDw8vKyttvvG73z1Cw7Cnr0fbozQyQ76q714OUaG4BlMYj794AfF10iP9kWF3WcLuAGem9+Z9tVP24LTV/35Rp5vSe/r2MhkJOtR9aOqS4Wi+kumBRuj/E+x/Q/70lj/2N77ANDzhGbS738nf0k4NopGywZkTnisY7Yn5G9GBx5TmSjk+TYLBbf78+/vLysxeJ7OuP+/r6+fv06u62Uts9x9Y0jBq/OJtx364uym3xRf9vtdnqOc+aAidt7y19+p5SP0ytSWfY0u933Ff+rq6s6Pz+fVv2Tr3NbNoqq+ebb/E+vZsDLINjwU2wgmWg///xzXV5e1s3NzSxMoRc1IwxAEQC7icR+uX0yyI6xMPRy3Zxs2+2P528SEMk8yF5HbDP15XrekOBbNSOf2ZfHjwzBW2aiC7dvcAxg+j32KWYZXoxMf99ystRT6gyw0gadJ6V9pD72Nbc1Vr1+saD3kaZt6u3p6Wl2N1nHxGivtsu017Gw/GZbpYysgxHH+fl5HRwc1N3dXV1fX08PGEkfuzvyImMAlSzf29C6zxQTnM7ZUz+x56qqx8fHV7cfmxTaCY3Km6+RHlHfLlQhQPhe5hjyzc1Nffv2rVar1ZRPpUGS8abjbNuD2SmNjJR1uWTQ8nvuK84tib/88kudnp7Obo9jOFlVM9ZC3dCQOzB3nwiU3CplgHbeLt/dBg0memCoxjHpJk7qdv4xYJ1JQpBJiBydE9AYKZyfn9dyuZyMmDJTjqr5I+uY1+bkCcAxvWEb7Vgx99d248NrYqfO5y+X329hpT7SX8rA61M6RuWJy7Hxqrx14zx4l0PM+R3LTHHKjVEA5WPkky1p//rXvyYgZZjud1XlWhMD2jSfUOcok/bpvcspXkTknKFzCDvtfrN+DLQuewGVITyFcmOm1rw7hQARlvr169fpif7pdHcXStoz4+uS/d3/OZ8sgHVnG0queXn5fttZFPrp06f66aef6uLiok5PT+vq6qqur68nJsv+jZTs3I+Zbc5x2MW+u4/dAO9jKQ7TeK3vdktZLn+swtOpZZy8aEVgjJ6Z+yToXV5e1sXFRV1fX0+3+AWkw77I0rp+u81O/xnro6OjOj8/n54vMWJmsV0Cn7cq8dyMYfTraMuskIVsnGPDXSjUp6MRpil87ig68jn7wJuLvPyNNkmW//T0VN++fZtd3wEUwZS67BxHZE27ZqzU4cgpWvfUeZz67e3t5ADodFgHx3EfS90LqGdnZ7Nb7hhG7DMyex4znORWTk5O6uzs7JVxEeScSOc90DweL+oJYFm7BR8ORkA1Mn/69KkODw/r119/rcvLyynBfn9/3y7iUA8dy0uxE8qx6I8vOeT5DLvtjWnkzqVRRqZIyHSoWy9GZALn7h1v9ud402Cph/RjtVrV2dlZnZycTO1vt98flMEUS1e8WJIccPScd0Wl7ufn52kh9Orqqtbr9ez5ANlH3e3nDcszsHSAFf2Z9VPnZJGjENI6s0M0QPH21s7x+vjIcbt9Xu/r+Lvz6bG/0fw1weC13f8GM0e0TFcZWD0/OM8yt25vb6dnS5DUdcx2NO4sewH1/Px8CuM2m82sUU8UN5jv9CQ5/vz8XFdXV3V2djY9rWjfCqmVTeBJ6fagsT6HdqOcas7NSvDLy0t9+vSpVqtVnZ6e1j/+8Y+6vLysz58/15cvX6YJyhDPRmvP1rFHh9kEGtbJ+hwCxqi464KgmnNp3Lvdj7ynmWVKtzXKOwu6MD39cr74+Ph4Ap/c/nlyclIPDw91cnJS//znP+v6+nqSyWNlhxKZP378OD2AY7Va1dPTU93e3s7SPmRPkTX9Tp2x27RFlpTfw6Q4ZtGvCQh/p/xdNNKRkC6XSEfHfcMExC7faLC2nVJX+9ITHUO3raV0tteNZydnxoh10aHYxm2XJivc5vnw8FBfv36dtswZqH0t+z4qewF1vV7X0dHR5PVfXl5md5VwRcyNe7CcbE++8uTkpD59+jQNTPeIOIKhQ5CubV9LJaWQRVGBkfH5+Xm61/jp6akuLy8n8D86OqrffvutTk5O6tu3b/Xt27fZ073zSeDnBGD7NubI3BknJ3/qZbpgNAkJuHzyPjfWx9j8Tq38OU9J3dIh8nF7lGe5/H6X0tnZWZ2entbJyUnd39/XbrebQPXg4KAuLy+n52MyN+Zigz89Pa1ffvml/v73v9fp6WktFotpQSR7Qulo0v/upWvRJxfiOCbpc3TqMDI51YArdWUQNQN16UgAbYghd2RiFGNQIbhZn26Pi2/sM6/tnBsdDttiHj0M0fvHaavWAcGeOuX4mN2y39Rx8Ofu7q6NKBjhkR2/Rfj2Aurd3d0EIMfHx9OEiPcPuPo1wuxsvpuxVlXd3t5OoX/YMMPcLndEZY0MsQtxnDKoer1CS3aYXFaePLNer+vy8rI+fPgw9eXs7Gx6tueff/45eydOB/JmomauHDBOal5LA+5ys7y26se2ILKy1ENvzu/MOROwzbL8So4uLbRYzB9Vd3R0VI+Pj7NH72232+n5k5x0dj52SgGO1MVjzNOtVquZ8+SEtt0Q+Mh4qF+CGAmFdZq+dw6Odpm0g8GCDtakgmDOc2PTPpd6y/ldVEhgy3zwzhDqoWPQ6QPTAWaZLJ4HOcfph+64HYFB1WMWmW9vb+vq6mr2fjHKQ4IykrErewE1oUvAM0wirPX4+LhNB7A4vKURPD091dXV1bQvlXlDTjZTcSqUHtEd39d5enuyC7ZNFvjy8jI9iSYPzQ7wBGT/+OOPKbkdGXJtt5l7FEZ0zomswx6zY+1h+44iukk/8vBp0yyrav4wFere4TT7fXBwUKenp1PKJ+1mlXi5XE4LBJStYxAOT5+enuru7q6urq6mR8JtNpspHN5uv2/eTomcjJyoQ4Jm+mEwCYNlHo96Iqh57KIr6oDzZMQgTRDI8ByJZItWZDGbZH8NUB2jZv9dOOa8dvSYzsjb5WwpCx0aV/0p01uskfOn6vv2yLBTp7PMlNk/g2xX3nUv/3K5nIwz7CLM7Pj4eEoH5EnXZAesK0pKBzMRrq+v6/T0dNoUTOZo6v+WJ+KAOP9mg8hEpRF0nj2GcXt7Oz3j9ePHj7O3un769Kl2u119+fKlbm9vZ3V4+5Lzb3YCnTGb+UVWD3BkNSsw0+X1dmAcK2+/sjzpT76TdTGNk7pXq1VdXFzMcn8EpLw/yQZMlp1UBVlugP/6+nq23zHsz4s3nW7cf5MA2h8ZYZezps0xOqIj8EIJ9Wlgo95dLxeCOL48z0585IT9nfPCc88REmWiHXaFeuhIAf/nOHSl6xNZeuSv+rEb4fr6enKWbodydHLvK28Cqr13NuDmTZXJrx4eHk4rt0wHZKJ0ub583t/fT/s/szk4E5EKYWe7RD1lrupzpl3/4jTMIBwWpc3n5+e6ubmpi4uL6e/k5KT+9re/TW3wnuWO+aU4L0mdUwb3xazcwMa2HDJSHx0T4rnexUBnwEnPhaguTA5D/Pbt25RDTc7WYXGAdt8NFQn149DjkOPYkuoImIYQEEw7NkiHyt0LHIucz6jBDJa6T/9TL51A5EjhWDpN43PplO0o2b88K7djXinWg3OOZpbUDdu2LVpmMkACcOYE03125PtYYucA0tbR0dGUL7+5uZke0EL9coxZ/4ixjsq7HjDdVRSlhpXmsWjHx8d1enpax8fH021+SQl0dSY0u76+nlZ/V6vV1I4T0gYMD7QVHFlH/bPTiMPgWyedf0lfrq+vp+e95pFwu92uTk5O2pVes+BRHqszVB73iiSNOIyJzNgTz4YeuQkUBALqsnNM7lPu9+/Yz9evX+vw8LB++eWXWq1W0/gHAA8ODurnn3+eVv2vrq5mYBsgSj7/06dPk1M7Pz+vqprdox/7u7+/n+03tMxV883nZLcd+IURcxtX+shUB893HtGAkfHmtR04dWk1jzEXkBzZmP0xokh7Hct1CE77pBPg/4yComOyZs5Bz1PP4dHaCvviKC9jGjD9+vXrdEMJdcr+Wj8u+4D1zadN7ZtQ7GS+bzabVwtYj4+Pk0fgqiqZRAbr6Oiofv7551l+iuCb67q9gTYEex8bVNfX/N5tsN5ut7NwcrfbTX27ubmZWFPazuPgGJ56gnShWK7JJMyAj4AsBstJwbxaJ3v2ambBMX1IntHGZDmrfqwCZwLn965PXOz7/PlzPT8/1++//16Hh4evXj+SfbiJfFJHcqur1ao+fvxYP//88+QMAqBPT0+zp66/vLxMudU8ms396HQbuz8+Pp76Gr3YMTrycmjcRQGdA+3y7N2KPdMzI3tgPftIB+3MqSkuSBGgO0JDRu1zMl/zO1Ns7CvthJjAvhGoR44xtk8MycOZbm5u9j4EqJOrk3NUFrs9cOtJOWJ9bIgdSTiWZHLAxw/zzbWr1aouLy/r73//e338+HEypAAxFWujHXXeXpiyU954LPeR/TFb3leY5ugAlcZg2Tsm7jCOToLfbVSUJf04Ozurv/3tb/W3v/1teoHi3d1d/fOf/6w//vjj1QNC+Om8HfXdMQzrlPe0f/z4cZIh5yZP//z8XCcnJ3V6ejo9Ieru7q4Wi+8P4fj48WNV1ZTXPjs7q7Ozs2liPz8/T+97z0vXvK+SYSGPe+zdL+e/83u3xauLQKLHgFiXIonz4Y0qXehJcGEaJ6Bt4pF6ukiF48tCG/VGfhIEHvP8GM2XEbbQbmzz+/bl5jtz1A8PD/Wvf/2rvn79OhG7bo3H+nHp7NjlzZC/Mxz/ZqG22+3EFvIe9uS5wlh5p1HqyHNTE/rnnm/Wz4513p//d4pn7sesY1RvF+5YDhpy2mLagJ7ZDJULGl2ujH1j6NSxnPSVudlMktVqVT///PP08Jc4uhgfQSayeYI57LLDta10Ez5j/fXr13p4eJhyqufn59OL+5JKeXx8nPLTHz58mBwvGTUdVRjq3d1d3d/fz57azy1F6SvHm2NhACFY+Y869EQl89tut9M8SLvZf8vzE+lx0c5sd+SQvSNhxLjo7GnHXlyLTjwfCNq5tgNS2kRHyHidnXKuYb0jJ8c5zPeUrdfr+vz588RM7Zy6udzJ/t7yrqdN0eBG3rz7Pyw0bCFh5unpaR0dHdVms5nuNArQbTaburq6mjz3xcXFDAxjCJ1ny/82am+NoZGY5XUK7hhKVR9eRVYm2Z0/q6rZNhnK1emzazu/j8Iwhl/L5ff75//xj39MNyTkfUtxel++fKk///xzWrhx3QyjaAtmDfzd4bVBterHYx1zT3UY69HRUT08PExboJIC2G63E1iGbWRxqup7Xjv7C+08ows6nshFmbpdDbYXTnbWxdA2fTczPj8/n24UoR4D+FlEW6/X9fDwMNvew4fPpBhk2Vb3nQSBtueFJtbvPpCpRp7o1vMt1xhEWUxY9jkQ10d5uZ0vi915/gYd5qh0hMaOZ19590v63JERg8onr+WWmM1mMwFr9rJmYSv51NwSltW5PJUqDG/ElrvCtEWnSHu/fNpT8/yqH8wlLx70RKVBpZ4O/Al6HcAzVWG2O1oZzV0zKXkn+t///vc6OTmZ8o2Pj491e3s73e3FfZpkwMzjRQ7rvNMX824cC55PJvjw8DDlV7NbJJP8/v5+0mUWmqL/7GO+urqaHmRsJkN7tB141wL76OLFHvY712TBarf7sfqfurNNkPuu49iyGLtcLqcdL7vd9wcKJXXBPHf6Q3ul8zarph5ybXTK/52TDGvu5k/HGO2wusLfOAe78eoWjCwDF/NeXl5mOdM4ZdczijQNoKnfTL0rf/l5qN0KYz47j5F6wpq4eEBQzTMJwzoCqoeHh/XTTz9NrwYm+4qSuknCPCu9W3deF7qyfx5EhktOsOd3Alpn0AyZzObcJ+crIzfPpy68Qhs9LxaLif3c39/Xn3/+OT0wJFuUFovFLFVBBuo8NB1o1Wv2bJDtGEo2oKed5Drz6uGsyDInHRs6ODiYdpZ8+/ZtAuMuJOT4OteYY6N8Z8aU15rRum9kqQS9xeL74trnz5/r8fFx9ghLLs4dHh5OkdxisZhtz/vzzz9nD5CJLGZ4lIc24mjGY5Y6GNrzvLdAxeQh9tPpLk6nsyfK4TGg/fMW6jy/4evXr7ObPLjATRky7tw33aW/IteITKa8651SDNlGYOOwuaPMKbwDK+9IX61WtVwup9cmhLJ/+fKlFotFXV5eTh59sVhM73lP8aqfgasL+Vi8UJT/eT6Pse6q+X5S9t3MJYPqDcW84yXAGJk7MHMEQOAjowzDyV1cy+WyHh8f6/r6esor0SkS3Ni/GF3O86Rx/6yDzhbSJ+ozgH93dzc9hDxvoiQrjDx5SPnV1dXUv0zCzj5tF2nbzteRwsh5GiypMzKd9I2MLuCfp2GFaSes597ZPECGuzLMMMmgGIZzfPLJczgHYsv5n8y0Y3kGdQIxj3O8Wbq8K22OhfMyeowtJLLN85Z5xx0fB5nzo5uDgx+vD3cUTlvootWuvGtj/4h55px9DXUeM5MiBsMV3Y8fP075vc1mMz0taLfb1U8//TTl0VJv92oFMjwqhQbHPpnhdl7bBprzWA+BjZOHA2hPHx3znNRXVdPGdE98yk4Q7vr08vJSV1dXky6TiwuYMqShblIcOnJSpn4zWYJm52DIBlIH9Xp3d1dHR0f18ePHCVTMVLKDJCybeqO81k13OyRt28BLPbgfHYNZLpdtKsjRHZ8t+/z8XBcXF3V2djZbN4gt5aaIRBP5rbNbLoLZ3sjMco7HMtfRqdp+abf5v5sbtK0ANPVie6e+GZlF13wuBFlp8qVhpXxLgLdRcWcFcYIppk63JhFdeXPb1HRi0+HOaDtGYG9PBaWdxWIx5Y1Wq1U9Pz9PoLpYLOrs7Kx++eWXiammDT66zDK50MOxf2Yrvob9sQ5Yuq1E+WRYQjlGrDmLG5nsnsC8u4i5ycgXGQ2UOW5QibfOBB4x+Y6ZGsg5aVJG0UPsxqvTu933h8/8+uuvU66XW64ODr4/mer5+bn+93//dwIZsvMR2HABzWPVAWjGwEyQ0Q9/91jkWFcHncti8X3BKmw1v93f30+5wDhCy9qNVZfiGv1mNpY/b7uyU/Cc6QhKN6/crlNknI/GithqUoM3Nzd1dXU1Ld4xisk6jAE0RCXpJPbbWEW5Is/oVdjvfkmfK+QxAgeP2Uh5nAIzDH55eZnY6sXFxbR39fHxsb58+VLL5ffnXia3dHh4OPMqHExOCHrtqvkqe4onPeUbDTCvow58tww9ZhcSWl4zNuePInu3tzafBCevuHdAS3104ZmZpnVBnRE4c45Xiqk773ne7XZTaoKr36k75yf0JSh1eX4usnWsiOBGdkb270lmwHGdo3PyezdGd3d3tV6vZ9cERG0Xb4GX8/J0zDmXYX1HFOxAPU7sK4vn0IiQdaBPveSThKTq+23JWaTj2ktuZQ6IMm3m248JpB1hGrHRkZOqekcOdZ8X7MKe0bW5fsRaIySfYZk7rbLFKsn87XY7y6lW/dhf6s4SCP1Yuhj7PkOit7RBWi8jY+P1kSEOgekE65iTkv0haHRP4Ml1zPmxjRieGVOuHbFT55PTN48pwY399+KNHXHHdh8eHur8/HzqZ3SWugKodKxs13lpOjMy5G5cR2BAOdhvjjP3vMbxObXgcYktJF3k8TdhyTGnakbnsa7I7NV7Og+Sjm6ud+sGlLljpp4fkZ/RHdMB/J7UyP39fV1fX8+2zi2Xy2mBMnIzyu3eOkIZKDv1Y9scYV3KXkA1k+qKAcCCdsY6AgAOevannp6eThu/46nz6LcPHz5Mq6QEq9TBdhyG0vgzIDlG1uQJzk/mCm1o1IEXc3LOaGJZvwbNTATv+TQLp7GT2Rr4M750MgYaTphuTKmXztMbXM2Ou7oiz2KxmNI8rIu3q/K4+0yQSjtMlYycR+RzhOIx7sbNekv0ZadkBtvJYhBK6XYlkG2l3shKQOnueEpf9jFuHu/apy2NSgf4vDYOM2W5XE7vA7u9va37+/tpN8hisZhy6bkNOYvd3IppbLJt034oI8/v9ODyrpB/NKAGm/zfTZZ9LNbnOAXw9PRU5+fn0zMCnp6e6suXL7XZbOry8nLKNwV4yEQjH8HFg+2JxcEky9hutzPvlzbINM3wyJQM8ARztuUxYN5vlCjnMwScj0o9nMD8jY7BYEdDo7PpJl3qd3hNO+pCZBs6S/rIt+juy1W6XjKwjCOdC4HRIExGz72qy+WPFxgSFLz9h7pzGiK/+zmoXehtZ9ix/Ogq9TqU5h1jqc8LRNEVdc2x4zgbgJhKS50coxFzzXEuHFH+7Cm9vb2d3UK8XC6nx4fm3IeHh1qv1216hG3Z6XscYwe85r3lzZC/Q2p70w4U/X8HqLmuA2AaWJ4QlNdn5MHWee7ox48fZ4/9I0C4PRtVZBtNSAKxvWbO67Z3ZBJxMlin6S8N1LmjnGcwI0ulQ7NhV80fYJHrPCH2le6uGDOR9IFAk2soQ/rou4nIpqwfg6D1UjXfz2r9dROG48TiV7h0bXeAwPo6xhaW9RbA0DGn0KlzDDg2LATdAIvri2z8TqcxIlDd/O4I0z6nTGKTucNteVU13UKcJ7o9Pj5OC055Il10lZ0PfEbIiFm/h6lSR9ENP/eVd98pxWMjdOf3zvi7TnSMjN+j+ND3PAgjObM8KDbbTrKyR0OiETrnY+WlXYfYDonsuTn5R4w83j/fu/N4l5PztnyrpnU3uuFiFJragNx31k/w7MLrrp8sI7baGTENPp/Pz8/TDSBh+lll5XazETh694BtmGE3dc36MlGdH2Z0QZm7OZLjdCgEJDJGM1ayXNsjr0lUt9u9fkW4HcE+AsQ+BLj2RQbup3/vyFVCe9oFGWlypIwK8ur5qppuzQ2Qsj06IRaPRa4xPrmPxrlRefedUlQGWdVImZ2n9vF9Rsf27W0DntmTGtXMPdwAACAASURBVKVuNpu6uLiYnqSfa81YDaZdGMLJU7X/aTkEbivbhs96+S4k6ma3203vp/dL87pcl8GUoWfnXTtgzDlkSNR/x/BGuvXkMetM/WZEBHHe7hhmElay231/qAjtIbemur0cM2vm/3Y0/DTodWPl60a/OQynI+mciwHP1+U8fmeKIzJ05KYDhQ5I9vWRc4d2zfmTsWREmDq8MHt/fz/lSfkA6IODg9mbcrlPnQ88caEeRox1H0CO6h0dr3onoNJIRuzOBvEWWzP7Ygc7OXLey8vLtLKbRavcXZVXwl5eXs6eVMVBdp/cj8hG1tN5Xk8csk/W14WLZBVdKM8wOm9G4HleOIiOzFo7D83zqW+Ccwe0+T6604Wye3GMbcYBdZOvA7mwtOTGnp+f6+joaFqQfH5+np4HYQblXLjbGgFWzsuEDls1mehsoivvCR875+R9zV2EEVDzfuyc0wHkSM6RTI7K9gFY5zQoN9crGG0kR5rn2AZIl8sfN0kESBn+W559WNXJa4dpEmK7eKu8e2M/G/ZEMyASaEYTvzue67vSMYXctscUwHb7/Ta9jx8/TiyWCW6/cM6Jef5G4+gYtxkjC1MMvmWU7ST8M9OwTrihuaqmfBwXFKKTji0xB0djI5B1/R05Q+rOhpc+ZrJ7NwL1nUVH7iyo+nGHS87N+bnl+OTkZNKdJxgntnVPnXYTiKDG1Ec3eSMbbcOMMN/pJLsbMbrSTWjbHcE/17DO/N/dbUcn7BTXXykGMY6xwZQkInc4JbznnXsc/6zax6mO8GUUNdsZ8LcR8/cxju9i0b/uvuoNQPVrgi1sStcxCtRNONbrDoza64B7uVxONwEkBZAFivPz8/rw4UOdnp5OK+VklF6p38eoLS+Nep/+UjeN2MbARLoHmvIEUMPOeYxt2qmxbeuPkyqAwuMdCFHmzulaR144yzlZeHh5eZlCOvaVhSkTAhJze92NA5HZjidj0y24MaowWCYSsaNyoUPpIgrK6PEwcPr2XI4/df0WG3NO3Hrl3OBWv/xu2TtQcnsZU45D9ofmsY3MkyZHyvP8iM9OVx2mjJwLdeZr3MduzsSxd+Uvbez3RPek84TpQLCrl3V1YOpr2F7Cwe329S6APAcxT+rJJEo9BAPvU7WX7TzwpETcBtotejGn51VkDzIBgvJxgYWsYrPZzJhwnMk+3XX/O++Y3zpAjezpN8GYdVt3riPPySWQ0lF1K9g5Tj1Thu4ap0u60Jn675hhiu+wy7iThbtttmOd0m6s+7TDjf5dJMP5lnq7NArlSdtdRNLlXDvyYICmXnjPPOV1njRjZyC9v7+fbYHqxsj/j9jq/6VQp9RPvo/Ku14jXTXfzuEQaMSo9jE9e5YRPWednUFW1ZSHSUifBxQHcL59+zYBa0JFe+Z9hSDHkIX1jLx01Y9FInrYThe8xobP7yO9cm8r67BsnQF6DN0H6ougmDY5uUfOkHWF6Z2fn88cRIpTM0wfsB7qddQX1jva0M/z6VwIMB2wMOpJPWRj3XX/X3vnth1FkizRLPWAkAT9OC/z/z82f9CDhAQN1HngWGnXlnlkQfOoWEtLVVlx8fDwMLe4ptuDbd2Ai9MFjE89u18Y7DwNEf16H7TP7ju4nVz+4fA8PUX5OU+avprFpmzIz8IyX5NkEuV6N5Br9kRZGc/2yv61hwstLIf8nIx2YctMB8CwFzPLmCg5K2iQaAs+19fX293d3XZzc3NqqG3bTu+E52uMAwpsvNYQPjXF/37fFdPHS3P4F/nJ9g6H83mZdimFOxLzyTOCfQyLsjXHlTzJVqxzgybl8rDfbRRAYdtz+DoBsJ+7wxMIrP8GcHyDKdvCW6YCCgao2All8La0xhZdH9c1wcDv58xjD4Qd2vy49ZZAXSSeGbTTse5s/1zFmMtdki/nSI/H4+ntDCFFE1A2LHFcAqXTs/7U5xS3hePx5ZRLwkWvkWYBrTO74mwMexiDkhmiK+L8myGyvG/fvp298TJ3TGZqIPN2Hz58OF28nHK8OJR3eXtl22Vv2/YCuLxo0OrGdAQi62uV3sDoE1UcHpuZtnpRHi/QsHMxD7YJWWO+czEwebPNmm14qoCBoEynQXntLJrdRRcEL89XWq8eLTUd2gEypB08PbIHsMx7D+QSz47Pv7WRlkcaHq3YaTS2/+3bt9PVm3x1d8gOL7vhWXs73YnrtakO68fAaxyZ1jT2WOne7xctSqUSe6h9ylTCT0DgSjQP3n5vIG6Wtm0/OkfYKt/PnbplbpX7G6e7Mtupp8jvIXbSRQbKa0be1O/GbvojUDKfla4C2kyXOHtzffwtHdrHJimb50E93HebGVAvcaRtWsRbttpin2VIerKt9pvbrrVLnLDla0P+ttBDB0q2zRFCbNDOhGXZ5pqsjeC0PsXyCb5tC12G9mGcGRXkiOi2Pb/uJod12pQNHR9lodyTrU4EpOFSs7Ppd+Y3LUpdtG1qz4gmjzCx2ZY+FbA49Cg0+j1WS4PLbece5h8Oh9PbNvnWgATeO+mFEDdgYwvsoDEaxvWQazIaA3abZ7XjYQdoaRpbbSyb7ca0nneLzGFf/EzgNutsbU4dt03h7mTNPiO/y/eiE5lws1ez0TYPbl2wPnvtTZtYDd1X/acFT32s9DzZs+0j0x1si4w+wkhzBDS64IIjtz7F6XoU5Dpa/knmxlJdb7PVPGuExHpy+KVV/gmtW5yA1wR6jL8CVqdvHmJibmaOMdZMcufYKvemZsjx+Ph4ml/NGzR9IUaCjc+dyI1LtsSOlOBV+T2dM5g1Jj8OKw08lLsNP92xGzszaLb6Ji0/T6BhcKE8+eyO0D4nNJbS2oi20obpee5nzIdD1WanTke2mTAdHW7yGuwbq6RMrN+29Vuq6Lg8L5o/tkve+pD7WwOUcTxZ/M1UW/ogj8M2QFw5gKYbx2t6a0CaYMyxXqY2mMLuotRKKDYmhyH8bQWeTRn0VszXn5lPYxmt3Jzzz7vRuc/227cfd4S+e/futP2K3pjA4QWmBM8xUT7rjOCX58zH+wCtL3YqbqlqZ/dXIwXqlt8bQ3O+bbGEekiH8mUdDA2czHrzGwGvjQhsf8w3zxzHDNKgPzF8MxuDVSMC/H4JWWEa5tOA0f3P6VjXyWGlfrz5KXUPiGbekzc7Jd3h8Hzr2ffv30+vwuY1i5SLYQLQVd/n7z8zCmrz82yDCZtY5i8N+T0BnWccSqyQ/2fYlStkw5/2VtpTtzLpZVMHvsqap6xSVnYC5NJaGqEvYqA8Kc9znIyzalwPbVYLDxzWEhjc2amHSQY/X41IGrCzDL/JNJ0znZppKVcA1Y7COrDjpTyeGmiOwmGaQiFrMxMncFpvrS5NjwZG5u050TgZszqX6zrx+TTvnjjeD5w4AcbsDc2WprRrXjPCd5XxonjL9DOs9FJATfpm481ep0XPqb8Zd37ppNRqQYWCNprdGGNT5MSQnN/0m1dlrdRpqJ6Okve6580AWX2MAeeC6wAs6+ZhWntP02T0zKcxotRjYkyJ4wn9pNm2eTuNAcROksFtyOf8nb8ZUJ026VYdvBksh6RedGL9PXpg/ciUzUxtNxO4U38rx0dQax1/1R/ct9pcssugbl0WdUUbIdngolFeDBgg9fV5ufkr8XOhM+dHWaepDzc2OD1vWMIy2nTNtp2PntPfprxs681+Vwx1d9uUBW9xJtCdGrnlMZU90XAbCjt2OhXTJR5liXLjWcNGc5NRbgmPYYWxxiPn0oZt6zc+tTq6gTjM9HRBfmcH9nzlKk+HyfuS5U7bgOzwGsgmr8RjnVeAwI4cHXD6ou09TdqJ/U9OjLbhPAm2jM/628m5f0ztTHuedDfphCeOmG+zhSlQb7RbhrDK2HtW4mMXOR6cvMJE+fZigvYl/beBmetkhjn93shd9MkyeBvbpD8DtJ3oFC6+HGUq7GeBknlMCmRaKorso9H4bXu5LWeSy/OumTt6+/btdn19vV1fX5+xg3T2zL++f//+5KXT+TltYAbqlfBJt643G5KMigZL/TTDa+mnNkmZ03SFAcOG58MJrAOB0a//nliG5aOO2pC61W2K01hTWP/EBDkfbLmpHx4icPBct0HIwL7qxB5BMC9eLsNyODf6/fvzoi2Pe3JnSWw8fctzoz4Uw3byUH1iogyt709guSJ5tNG00UTevHtj1aeurq7+2bapFGpwc9KmmMnL2MDNJPK8KckGMskwsZT81hhDDC5TATkYkP1yAZscm+SpKw4t00gsi3Vq0wHWieO6wxkQzGCTj4d3e8FTAiwvYDABlzfJs9406ImxMjT7mhhhY47+3XlPnYXpmwNPflOelwSDBctuv3uKIs7JAMq8Gsk4Hp9X6MNIw0TddmGjGa1lp8x0M34DMerK/d8se2qnPG/92fF+5fc9Z+b02/Zyeu+U1wpQjejNM6+GLwYsK5QVaJPOTE/jaXGmss4qq9/cKfyZjDTD/HjkXFiR+aTcE5BLoZlXe486P5vBEpi8MHg8rk/hGASSZ2OdzUDbTgGDcurPOdhmjM7L19YxuOOswMmdw3FX9WO97ABc5z05WxrKckk+Ce1o8YqVcS7ZTJQybdt2IgMcomcrE9stIJoQIE3c1MX3FzR9TPqa+nzDk0asHC5xYLQp48RefH5n+OVFqYnltZXRBANW4i+KegGYBsnWyVxh/8ZGpKexE2gNzDjcbsWjrBzmX139uKPTuwMsaz6bvVoWgtKk21ZX65Qr1C1Na0uCqFe2A85ZePKIwWn4mYzHgTppK+iNzTd7clszH9a/DelXIN7m0yZ7TJltMcwMOHH5vzE5L7K5LtZpnB5vbsp39sW8ip2y5oKSADFHJCsW2nTAZw1Up7QGvtberX0vccQTprU8J/l+GVBfJICRT2A6edX22wSQLUyNsGLH+WwGPIGL68nvmWPNySu+u4pzMP/6179OOwMykZ/LIAg+9Pg06Dzbtpf7VPfq6vbh1YIEvHxuToTBCzQMZEgBWhukF1U8JJ90bbMk++YOArcddUFH0UAqnxu7tIzJx3qf9N/A0flEPtbJIW1E1ksA5f9ch8ghPVfdU9c28vz7779Pi0tJY7vwEL/pgp/baMihkSbrlDprv016bo53ksMytDow/JY5VE/mr9iiw8QSV8HekM8dr3nQdCQOWZsjmNieOySHzG/evDmBa0CTrOB4PJ5A9M2bN9vNzc12d3d3uhwi+RIoCDZtLpSMkr+zPgaWifW09mqAMem9GTtBiUNq17Pl2YCbwN9kbotdjsO6tfyo0/aZejAQNwbTWPXE1JyHp3fixBPoeOPIA4CZ28zFQJQnIEqnxrReobfTa323sT22K8HXYMiQ8lbzx9T9RCicH7+7jVZMtY0gHGfbfgOgvkhYjM2heYsJlPfS28P7t2bwKe94PL7YaN46XnvuzkcjjY6yKyDDp217NtoYd1hqpg3CcJPGAGpPyTs6vWC1aiN36MYGkqff0NAMjt8npuByI6c7KnXIebmVUTNfL2wZMC2jwd11cRlT3aiz6G2ywcRrI6TI4uF80nDKrYFnFooiR5xMRgTJO+w1Dp8AaifXnEP01gDG9Xa7ub+4TWkXE8naI1ONdLW6OL8pbVuIsp38liH/qgNZwPZ789D8v0rHtJOHPKuYwLuBcPOAk4zO28wiRpxtV1zEylAqXjvx8j6sMFzuDzQAmYHzIIN/Y114Xynv8qT+OPyPjHxuRkjgZaflM7MO69zptu3lxvN2aCHf21RFgGU6VOCyOOVhG6Vu9vKzflaLb5E9beK4HMJnw3yG79zzSZkJoJ7nzjRA7M9zuq4z62eZGtgyD4Oq82/O3ETLMjTnP5WzV5epnSecWenkH9821TJtYPqznt/xJoBuzMX5TCDJfKcO6jzOlDQAfoIZRsAx75Kn0XD7VeZUA8JhrdyGlTzJSNPZ3JGtP7KLZlhN9ymD1xxu28v7C+xwJoZmUG9pWrDOuQoe3fE5ZfbQ2TZoYGd5HKpOR0Ob3VB/+U4GTr23UU/0m/nLbK7PdibO0Qc47bDIQMlCyUBdd7bRRCJan/Dv/N8Y3p7+Wtz0+Qn8DdB7uwb25HcfaXIn/PJZ/va5gd9KsAagTfCVNyEoNRBvaaPgqUO9UEYpb+o4lGOaGsl7cnL7Dm8pTyfh7eRXV893DJDp5i6BlVFYrrYdaNv6ufVpZZ7ANbHalY7MdPO7GV5jswzuVBPrdn4GOg8vm9yTA2I8gzHzbUBHp8T4YY3ZVJ9hPFfWA54E0siZaYAwUd4tGpnbVBLr5j7ivmL9rPSx5xwn9rjHFl2+5Uw9TdZY3z3ZVouvJguHw+H0FpAX+e0BqoFlUkrz/hRwAl/m4UowXw/xkl9byWUcy9q82dQIk1F5wt1GkPQ84pY9ftm3SpAkc02HyLCQ27U4lcC9rt7v6TbkMM/7Qq1vd/rWwfysAeLksMw2/GxaNOLvzo/OdgqJ4xNMbQHD3xsIxWG7Ll6Fp2ycxySIhlHGTgLKBNMAcNKGfXIxKW3hcif9T89SnxavkYqJiDQMsL6MCRM5WvVty0TnPeGO2WuzbefNdP94DtUZNmU0lJ9A9EyI0ljbdvmENYFpAsOWZvLGrbEpAw1p1RgEpcRNh8uCFFns4XA4gWo6XsoKU800QtIFoL0dxkNztk8DQE5DMP10KqoxvdV8o9uXbdwc5RQaS24jEccn4JjR0mkzxCEaPCN3nvGQA4E7UzNhj9xQn9+TJlNF3B1xPB7Phu/Mk3PWrLNBri2I2f5bHm6PqR0Zj3mznSfAdWjPW3mWzXJcggEt/1YOy2OdJgJz0cZ+e5rmVaz4VlF/nsLUIfbit06c/0xv42tluZNOq5xMx+dT41CHBNkwz4Ckz/5v2/MJFeYfQE36zL/ytbxtYYksK52UemoLTdTHZFBTcBsY1DkacjwDgPM14NuRNRnazoqw/sZIE2zfHGLTCYZ58ho72xMvGom8ZKDcD9qG7vnfVtAne5+YIevZ5lynKT/a1sQiL2WALZ7rMjlf26jTNPlXejFOGX/+8RzqVFAT8KwAPWsAzXybYhnnUs+0bX1epeXr/AhmzdtNnXTlNFZTDCzzcDicFrXCXH0kMB2M869J66kFAnXak0yIC197jizPGtDagxsUWzv42bSY5HLJHAPsbW7YaekoJwfNvKlr1oX65zak/PfceDo0V/eTFxls0keHdgoT6LhfUB9mlxPo5TvLcRrqrvW3lrY5qOYAWz/y5xVJYVj1wz3y5bgTqP6jC6at0Emp0+dpFXrV2I4/AWOL5+/TxHzzaK1x/GwC/Db/N3lUltUamUwqgEhw5ULF8f/ZkYeCBJHEJ8ByyoB58khpgNh6ss4iu+8BiF72vocxNubY2iNsMqCU4BV1Amqb5zVYJy2BjSDpz965QabNqRy2UVtEcp25LW5y2A04zBjTjq39PApzWDlJ6pjfI8Mk696q/ZS347e+Ylmbs3e9/H36bBl/G0O1Muj1E68pc+XJrIiVwpoi/Ll5WG/liMzeNWCvRMNznVj2BMquf2tk64xy+kBCngXsCIRcyMh2mfylI6fuyStslkBLMMgVhXyWPFJOvjN4asHbnlLHyORhMYedPOETphe5zWodojvOH1tOss3IkuG250Ib2Ec3/J8ymTfBl/bjqac4ionhme3RJhtDb9MlBh0+9+98PvUrxmH6VqZD6+sN6BsYN8JFZ5T/K+D2s0tk/qULplfBlVixvCkehbbyyNDsjV1G86T83BYUEt8yTobTZJ2czPQb87esTs+6e+hn9kIGmvlTzscmHXcS8O/z589niyoBLc7zEcQPh8PpvlgDSTtnn3rlOW8/4p2aAVQCCtsrgTKFrVJnBL18z593PeR5htsBfx7f9JRFyo8MtAmzWO7ciH7JqBvRYFjpcuXImVdjdS6rEZIJTJzHHhjtgSvr2OrDtpr6PZ9NLNhkqMV1f2yyTvXdtgsXpS5R3iQQG78ppDXqxFTbd4eV4bV88syGu/fdZW5bnwtaeXqyh6YHM4/WVHQY075FDunJ6DJM5bygDZe6I9BOALyqayuTTs/lRJaml4QAMstK/tv2kuGvFqxsC4nP/9ZVHET2lKYdGiNuDtT6sTwtUA9Nj6s6TeDR+lOzu1U/Yr7eKcC2NbHYy8u/rQjZNIc+tXXTRb63dslvnz9/fvF82y5gqM0jTELxudMYTFs5ztPDyVa5qUEa8DW21OozsQLGd32moc8ErimjAYbrOQ3j+J3ycaHmcPixCZmg1+ZjWda0CMPgCzhcB6adWIXnhrkwl05JtmidXF1dne7r9Av9GqvwQhwX6iY7pgxxBpyeSFyeNGJd01ZeSW7OmQBGHbR2WfU/57kC25bOJKCFBqzWedNFm8teEbVLZDH4GVwpm/v5qn7UGfVl+RmWgLpiio2J7Ak2AU0zMHq4VRmXNOgExBPoEpym7y7f6Ru7bCA4Narz8bOWj8GMvwcAKAMZq1mm/2fqwOVMTI9lR/YGtmTVbfGwXeixbdv29u3b7cOHD2dyXV1dne2OMGPOfx6eIGCSeXI7FBk167hyyvm+WoThd/5vzp358rdGVBrzolNswLmSy/Fdj0kHE+g7GEsm+97DAP6f+vAeMFte2nzWGqZ3cm3bBS/pS1gBYguOuwIoC98MisZj793kjQxWEtkc2QkZCevMNI3NTJ1qrwHNTB2fRmRdt3xXxmYwcCdOvMmQ/EdQ5hSCn0+r+5xjPBwO2/v377fD4XB2f2cMmHPBV1fPF6bk+devX08nz7y9LHOjnEMlWCZOnjWW054RKGmTyct3YExt3GxjAjcCp/OwjTh9s6lVmGzan6fQHMtEnPxscg4T8DfZmxx7aZvtZzQXEOV9G1NYAmqbt5uU1MCwKW419F95P5cxeVnnMQH/5M34+8Sw8my1wsnfCeI/67lZRxtdM/zJeTXA9G/TvsfE53CNQNLSeE40aWIDKfv6+nrbtm17enra7u/vz+YiY8y81Dv3yYZNf/v27ewttblMpE0XrNrBoEmH0Dr5ikg0G7cTXDnES214AqkptDLjHJo9XAKeU9jrb34+kSx/b4Dpcqch+Qp3CKIZ6fDSIi5ErsLukL8JZACajMMN0wzY39uG5Jb3z5Tf5vIYd/UbAaMZtPWyJw9lXhnt1OFdXpOhOb2Vs9q2l4snlttzqcm37Xvl823bKqjZMA+H85fMZdieIXxYAbcdBUBzsii/2RmyPNvuCqTas+acVmDoPJymEY5t62SG8S6RjfGm3TKNTDRQXQH2ZO9N/47fsGLSlevSZGjYtNL7tj2/S6tdGs+9xLHRf8RQI4SNtCG1ldOUYQNqjXaJYicwW4FpK+uSDtXAyt8b62ly+v9UrxbMCFt7TM5u5QwmZuL6ObiNmpPYs4Mwy3fv3m1//vln3QkS4Azr5Ap6vhPQ217Rqd1bG7I+Xq2m7K2MZh+T7l1mIx/WofU96ZW/T23nwHI8AnH9rJtVXvxvG2a9Vnn9TB9xPgF02ymBNJfEt3sxslOGF8lPYQmof/7552lOi8OnqfEnT8XKtjjuzN4U3oyiAbKfNxmSv+dDGzDy8wQWrQNNm78b2BJYDod+s3kDX3euST+U3eDgFUzXj/q3DD/DaiadHo8/5i4/ffq0ff36dXvz5s3pN97lSdC0TuwwrYsJ/KzLZgctrvVhAEtogOS2buUwTM5pFcd5uu0mO6Q9mIxMW7Fa/i7XfcRb8lq/avWNPO6Prm+rd+JmXzXnRNuBlsPh+W3G+eMI6dOnTy/KS1gC6n/+85/TGxOfnp5OQ6t0hFVwh7eCGKwYK3qar3Mal+mhPg1kMoImRwOxFcDYaJsOVs6GTsXyreT2b3uGOsnsvM0am0E3gJ8AwDrNcJ0b5QmgTc5pdbs5vEm2PXknYGplZKphspemg1W5rT2aU3JoBKUF2+e0Jas5T8oYpzf1Cds+px5CbOycXE/v7276bP+37XnqKKCYYX2ANM4ithcWen19fVrgjNN/eHjY7u/vfx1Q//3vf59uzHl6eto+ffq0PT4+nr2nm8yV82kRsoU97zyBqZXFQC+7l28+21hax3XnaHvoGqBNYOI0zmuSe/q9dfQ9tuW4K6CZ2DbzaEzPea3qTuY8XZA8OZK215AdsIGp6/H9+/ezlfmV01rpkelXQE5QaSxsCpPdMO0ENqv8Wv9aOWmzzCmvVt62nc/n7tn8JEtzGAFnXxCUlXmeGty28/e8cd40dvjw8LB9/Phxu7+/P+Hf09PTeOx023YA9f7+fru+vt7u7u62m5ub7c8//9y+fPmyPT4+nv5SQDtmZ4NZgZdDa6BmvFEkQ1N8y5v5T3FavhOzWbGNVu9LguuyBw7W8QR2zs+sZ9Up2KnYsfZktfMhgLDsth91soPm8CZ7avLthckx2ul40aftEphApzky/sY8LPeKua3qt1dOnrWN7av+2NpmKnOadnNagyd12I5cExzbZewB3AApmShZ6MPDw/bp06ez+2u5BbGFJaD+97//3d6+fbvd3t5u79+/3969e3d63/zx+OPy20wHPDw8nE0L8KILMs3WEK2RWgM4XmNXLb4buQHpxBhWDGSaC1t1jKmueW6W0jxxk8/PppXQls+0la3VYwLQBgpmai0QBB3PnWhPPgJc65zNqUyOxODgMii75Qvj3TvmmsC2mgB0cipTnq3Mqb+t4rHcJhvDymFP8byPmYcyvK7A9Dz158t9MleaPPM9C0uJdzz+mG7666+/TgD6+Pi4PTw8nL2OZk/XZ7IdF7/e3NycrXC9e/fuxFZvb29Piwjfvn3bvnz5cpoa4AvGuJ2FiwtNwdPpKHr7iaWumGHzmjSK1XycO6f3VE7G12SjgbUJf4MK86Kclm3VOZqerSfKQSOenMQeuHtTfwNKl03dNBbG+bYGug3QJ5mbHppc7bm/NwfcFhgd9pin49hRuf3tGFb1NCnw8xXAL8FkOmpCaQAACTtJREFUQRScf76TWWaOM+DXjt369Bu/81Qfb1HjynxwKEP4jx8/npjoly9f6gJomyOe3im1ZKh8h83Dw8NJwADqzc3N6R3zYbJB/bBVLmrxJEw7vWNwmAzSxjBtGWqN6s82ShpmY1tJszKsyDJ1luYEVh2hdeDV3KYdB/XcAM4yN+fUvq/2AybexPzYIZjnBOKN7TTAWIGI5WhzmW4Hg47zau0xnd13+e3/5PRW311/5rmno5VsjDPpZ4/guG/zSHOwJAc2yDCjQ9+Z4DqQrWYYHzx68+bNGR7d399v9/f3p6lKkr7I2XS7cowvdLtiqNxonYwPh/P9W/EAt7e32+3t7QlgOS/BqQHuGIhHSIWmfYQNjPKZd2Xy98Yc+ZzK83xX62ANdKf8DUyHw/MqsIPlWBkmF0+a15xCM0RuHXM9Gki0jjp11siaOk36dF52GAajBhATeLU2ak66OS/qOva4ZxOTzptMk4Nt8ky/Tfk1Jz05dpezB5CtLVaOgCHpuVAUvOCrZxI3L558fHw8u9aRYOzXtecU3bb9sJ9MRWZOlGs+ZKF7C5+sQ77/0n2oVr5PxXz58uWsgpkMvru7OyksnuLdu3cnQdrUgF+hy4pSHjb0dGaa//cMqMWZjIm/N5BueURvk8FPRt/kW82LNvbm/N0J3UFcdzuHSR+O32RpnbmV32RZpdljSE7bguu2x4j8G+u/+kyZW3s3sN9zKK0Ok5xNlr362Y4c32knhheMuL293e7u7k6M1Pc80BHnaPHT09MJZAOkNzc3p3yCA1+/fj2tyD8+Pm4fP358saiUurUTme4frpOdSAsXXd/XOiK9dqj509PTdnV1tf3vf/87gev19fVpISvs9e7ubnv//v1p7jUnYeJBwl45/zrthVyxRdejGVnzsqnTpeDMvJxnK69tLzNIrDqPZWqdxmCzB4h7nZ71msDUaSePb4Nd6auBm/NzPRrjoOxNRwmruV/KO5Xf5Gxz3e5PTN90vNdek1zTQZGpLzC/VfkttD6ZNRj2/217Po7MV9kcj8ez597WFEYbbNm27YQbmVrk1iauzF8i98qG9/pjwi6g2jhpsBSIgmSY/+nTpxcrbJwaCPXPsC3pwmAJsJx79aTxxM6aF7WymuG5vk7P+K4/47K8NlfoTrBqvIl5TQ29B/7WRQPQfD4cnm+ZXw2Z3Tkn1rhyUAZnhlUnpxOcWMTUIdg+01RKszU73jxrDnsC89Vwc0/uvXpa7yu22uJf+hufh0lyOH5zc3NawP769ev2+Pi4bdu2ffjw4YxMHA6H0wJS9o7e3t6eVvOvrq5OazpcTHp8fDxNH/oosmV2X2xOk3FX9lj1cFzEcmVbh28dqnXIw+H86FfA9Obm5uwcbS7DOB6fb0LP+8y5sMUpgnaqhuzZcrffqnJksG0bR5tnbGU042+ssbHDibVMXnRimBMb8bNWn2kB0GDcHI7LWTEq66XlSZ21zyyjfbdzdSfbYyV+zvnoydG0NnK8pge2J+Pn/8TGGVp9ViDS5LOe8lue8ZpFHtfMsD4j2Nx0n33tAczD4VA316evf/78+TQf+vnz57NpQvd716e14+SM3A7NQW7bL76krwHjSgCDwPTdi1tRfoCVQMsJa2/PilLDYA2w3kngExqrju0VYB9bbR2euroUUH38bgIq5tvaqcmzB6gTg7kERJxf29i+6rSJ08q4hHVNzyaZ20GEBlTTMWfGp93YIfIIqmVwJ530br3mWeyPAL6qF/tbfvOeX/8lXmPqJEbcM8o+zBNJ6bNcVErc7BLyEc+szWQYn+1MXsimvJMNrEBzIirNobR++48AdeUpHX/qqBOopHHYWNlOEXDl1qw3b96cDIQM1iD7999/n13sEjk8XdDAKPJ4nnivo7ie7pCNEdvrN5Y1NT71PoF6+07nwt8aiFN2tillvURGpqGOnRdX7fOdMtgJtnakLAb6yVm00JzZquNOHXFieSv5mm5d5vScnwl8Gf0FwLL2wV02PPXFPEiAeLTTL0psm/ADsunD3LhPAOXpS25p8iL1pGv3t6ltG4hady0tbe8fv/V0teJOYdpRQqZplfG+ziiYjZdpgru7u5N34y3a2/a8adceLQDr0w8ul/I1ltfqa8bS4rSO2fJpQNY6WFugWwEZO/zeBP2UlmW1tm3MbgLZFSNg3QigTYb2fZLP5TbdLLjFKW/rxMywPZsIxiXlNNuZQDfgx6k1rozzdTEJ6SfM1ySHlyvnL/Fc3wzZr66uTusjBtCcSuIWSp6yzHqKDx00HVE3LUztSntt4Dql3Wu7iwB1xX6aJ7DhNCbHeG0PKT0SG/qvv/46gegff/xx2hQcFhuQ/fDhw3Y8Hs/mYQLU8XxuQL8aY6pzYxWrBqUumNeUxoDEMho4TSA5eekVs5k6v+Nz7y7jux7t8yQX2VvLh3Lwv21s1fkmNuMyGlsxQLb8Hd/fadutri2PyYYCoD4ZxP8JtvEA5fF4PM1rMh8eAd2257fHGmgTl8P95Hs8Hk+HgrhVkkN5b6p3+3rqrTmV1l/29mlPuNUIz4RLLVzMUBkaq7OQrsylm2f93MqLJ/v48eOpUbM1g1MD2arBlcbI8f3799Nhg+wq8FataT429Zo26u81oOu3x0oTxwA2hcbmLpFvkudng41vjzE2gGqd5lLAnpzFHqBPIM/8W4djWn9fLXpOaSwLDxikXM5b8iYlg17Se4dMmGTeyXV9fX02rE8+Cdy+ZMAm89y2H4s1Hz9+PO3S4e10PoLuEWL0a2a6R8pST7fP1JaTPU6OsoHqFH4JUCcBEto5cAezUz5v7ICT8GS0eX3wp0+fzuaJ/vjjj7MdBNmKEQPMxPi2bWcsNsw1K4mZp+XpCr5mY9q0f6muGmg0Zu40zbBoUA0AJlaXZ+15k33FulehsbZWrz1QZ/mei2U+DaxbuU7TFh5b2Xy2WsSZ8li1lcsKUGYoncMzYaGxy23bzkCRF4Ew37xlNuCassl40294j2ji8W20YZ3pPwFTrmH4tNlem6xsbtV+7XvT6aUO1On2wk+dlHoNr+E1vIbXMIf1JM5reA2v4TW8hovDK6C+htfwGl7DbwqvgPoaXsNreA2/KbwC6mt4Da/hNfym8Aqor+E1vIbX8JvCK6C+htfwGl7Dbwr/BxHqR9oyBIqsAAAAAElFTkSuQmCC\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 432x432 with 1 Axes>","image/png":"iVBORw0KGgoAAAANSUhEUgAAAVQAAAFlCAYAAAC0tBC9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nOx9eZRlZXX9fvM8j/Wqqqurq7uq6a4egG66GRV1BQemBNEsRIUgyKAmulQIBvKTKYkQIIiKaFADicsBGzqKkWWICAQaUGSyx+qu+dWrN8/zu78/KvtwH91gAwU0cM9aLulXr17de9+95ztn7332p1MURYEWWmihhRavOfRv9gFooYUWWrxdQkuoWmihhRaLFFpC1UILLbRYpNASqhZaaKHFIoWWULXQQgstFim0hKqFFlposUihJdS3WIyMjODee+99sw9Di1cZnU4Hp59+Ov7rv/7rFf/uW/G7f8973oNvfvObr+kzSqUSjj32WOzYsWORjur1i0M6oV522WUYGRnByMgIVq1ahRNPPBFXXnklstnsm31oh2z87Gc/k2u2cuVKHHHEETjllFPw1a9+FWNjY13v3bZtG0ZGRjA3N9f1+oMPPojzzjsPmzZtwtq1a3HSSSfhyiuvxL59+7re96tf/Qof+9jHcMQRR2Dt2rU45ZRT8J3vfAfNZrPrfe95z3swMjKCO++8c7/jvfbaazEyMoJzzjnnZc9rMR7MQyHuvvtuKIqCk046qev13bt340tf+hJOOOEEjI6O4sQTT8Qll1yCxx577E060jcurr32Wpx55plYt24dVq1atd/PnU4nzjnnHPzjP/7jm3B0rywO6YQKABs2bMDDDz+MBx54AF/5yldw//3349JLLz3gexVF2e9hfieGwWDAww8/jIceeghbtmzBX//1X2NychKnnXYafvWrX73s795666248MIL0d/fj69//ev45S9/ieuuuw5msxk333yzvO+WW27BF77wBWzevBk/+tGP8Itf/ALnnnsuvve97+GCCy5Aq9Xq+txYLIYf//jHXa/V63Vs3boVvb29i3fyh3j84Ac/wEc+8hHodDp57aGHHsIZZ5yB+fl5XH311bjvvvvwrW99C+vWrcPf//3fv4lH+8ZEp9PBySefjLPOOusl3/MXf/EXeOKJJ7Br16438MheRSiHcFx66aXKJz/5ya7XvvnNbyorV65UqtWqcvfddyuHHXaY8uijjyqnnXaasnr1auWBBx5QisWicsUVVyibNm1SRkdHlT//8z9XHnrooa7PSaVSymWXXaYcffTRyujoqPJnf/Znyk9+8hP5+fj4uPKZz3xGOfLII5UNGzYo5557rrJjxw75ebFYVC677DLlmGOOUVavXq2ccMIJynXXXSc/f+KJJ5SPfvSjyvr165X169crp5xyivLb3/5Wfp5MJpVLL71U2bRpk7J+/Xrlox/9qPL44493HeOjjz6qnHzyycro6Khy8sknK48++qgyPDys3HPPPS95zXhNDhSf/exnlQ0bNiiFQkFRFEV57LHHlOHhYSUejyuKoijPPvusMjw8rHz7298+4O/ncrmu991+++37veepp55ShoeHlTvuuENeO/HEE5Wvfe1ryvr165U//OEP8vo999yjvO9971O+/OUv7/c9vzhOPPFE5Rvf+EbXv2+66SblyiuvVI444ghl8+bNyp133qnU63XlqquuUjZs2KAcd9xxyp133tn1Od///veVU089VVm/fr1yzDHHKH/zN3+jJBKJrvc88sgjXdd927Zt+133g/n+Xhx//OMfleHhYWVubk5eq1QqytFHH62cd955B/wdXnNFUZTh4WHlrrvuUr74xS8q69evV0444YT9voOtW7cqH/7wh5UjjjhCOeqoo5Tzzz9f2bt3r/x8ampKGR4eVn7xi18on/70p5W1a9cq73nPe5QtW7Z0fc7B/K1ms6nccsstyoknnqiMjo4qH/zgB5Uf/vCHXe958ff2cvFy966iKMrHPvYx5frrrz+oz3qz4pCvUF8cVqsVnU5HKqBOp4Prr78el156KX75y19i3bp1uPzyy/Hwww/j+uuvxz333IMjjjgCF154obS8tVoNZ599Nnbs2IEbbrgB9913H6644grYbDYAQCqVwllnnQW/349///d/x49+9CMMDg7iE5/4BDKZDADg5ptvxvPPP49vfvObuP/++3HTTTdhaGgIANBut3HxxRdj3bp12LJlC7Zs2YLPfvaz8vm1Wg2f+MQnUC6X8Z3vfAf33HMP3vWud+Hcc8+VY0wkErjwwgsxOjqKLVu24LLLLsO11177mq7d+eefj0KhgEceeeSAP7/33nths9lesv32eDwAgK1bt8Jms+GTn/zkfu9Zv349Nm/ejK1bt3a97nA48MEPfrCrSv3xj3+MM888s6taeyVx1113YenSpfjZz36Gj3/847jmmmtwySWXoK+vDz/96U9x9tln45prrsGePXu6fu/SSy/F1q1bceuttyIej+MLX/iC/CyRSOCiiy7C2rVrsWXLFlx++eX7tZoH8/0dKB5//HFEIhFEIhF57ZFHHkE6ncaFF154wN/hNWd84xvfwMaNG3HvvffivPPOww033NAFCzQaDVx88cXYsmULvve970Gv1+PTn/40Go1G1+f88z//M0499VRs3boVH/jAB3D55ZdjfHz8Ff2tv/u7v8P999+Pq666Cvfddx8uueQS3HDDDfjJT37yktfgtcTatWuxbdu21+WzFy3e7Iz+cvHiCnX37t3Ke9/7XuXMM89UFGVhRRseHlaeeOIJec/4+LgyPDys/OY3v+n6rNNPP1257LLLFEVRlB//+MfK6OioVGYvjltuuUX+BqPT6Sjvfe97le9973uKoijKhRdeqFx66aUH/P1cLqcMDw8rjz322AF/fvfddyvHH3+80mw2u17/+Mc/rlxzzTWKoijKjTfeqLz73e/ues8DDzzwmirUWq3WVVm+uEL91Kc+pZx88skv+dmMT33qU8opp5zykj+/+uqrlXXr1sm/WaU8/fTTyvr165Visajs2bNHWb16tVR6r6ZCveiii+Tf7XZbOfzww5VPf/rTXa9t2LBhvypVHc8//3xX1XjjjTcqJ554otJqteQ9Dz74YNd1P5jv70BxzTXXKB/+8Ie7Xrv99tuV4eFhJZvNvtzpK4qyUDVeffXVXa+ddNJJyg033PCSv5PNZpXh4WHlySefVBTlhQpV3UE0m01l/fr1XdXln/pbk5OTysjIiLJnz56u93z9619XTj31VPn3YlaoP/jBD5RNmzYd1Ge9WWF8sxP6n4rHH38chx9+ONrtNhqNBo4++mhcddVVXe9Zs2aN/DerkQ0bNnS9Z8OGDfjDH/4AAHj++eexfPlyRKPRA/7NZ599Fs8//zwOP/zwrtdrtRomJiYAAGeddRY+97nP4bnnnsPmzZtx/PHH4/jjj4der4fH48GZZ56J8847D5s3b8ZRRx2F973vfVi2bJl8fiqVwsaNG7s+v9FowGq1AgDGxsawZs0aGI0vfEVHHnnkwV20lwjl/3xwXqoiVBTloKpF5VX66axduxYDAwP4xS9+gb179+LEE09EMBh8VZ8FACtXrpT/1uv18Pv9GBkZ2e+1dDotr23btg2333479uzZg0KhIOcyMzODSCSCPXv2YHR0FAaDQX7nxffBwXx/B4p6vQ6LxdL12iu9lupzBoBIJIJUKiX/3r59O2699VZs3769i7ydnZ3tun/Un2M0GhEIBLo+50/9reeeew6KouDDH/5w13tarVbXtVvMsFgsqNfrr8tnL1Yc8gl17dq1+Kd/+icYDAaEw2GYzeaunxsMhv1u0gPFi5PFyyWOTqeDzZs348orr9zvZy6XCwBw/PHH43/+53/w8MMP4/HHH8eXv/xlDA8P4/vf/z4MBgOuueYafOITn8AjjzyCRx55BP/yL/+CK664An/5l3+JTqeDoaEh3Hrrrft9Ph/IAyW3V9saMwjoL1my5IA/HxwcxJNPPolGo7HfdX4l79u9ezcGBwcP+LtnnnkmfvjDHyIej+OGG254FWfxQqgXG2Dh+hzotU6nA2AhqVxwwQU47bTTcPHFF8Pn8yGRSOCcc87pIjP/1HU/mO/vQOHz+ZDP57te4yK7Z8+e/YqAA4XJZNrv2JiUq9Uq/uqv/gpHHnkkrrvuOoRCIQDAhz70of3I2pf7nIN5D///hz/8oUBZ6ve9HpHP5+Hz+V6Xz16sOOQxVKvVioGBAfT19b3sQ85YsWIFAODJJ5/sev13v/sdli9fDgBYvXo1du/evZ9ciDE6Ooo9e/YgEolgYGCg639+v1/e5/V6cfLJJ+Oqq67Ct7/9bTz++ONdeN3w8DDOPfdcfPe738UZZ5wh+OHo6CimpqbgdDr3+3zia8uXL8czzzyDdrvddQ6vJb773e/C6/XimGOOOeDPTz31VFSrVXz/+98/4M+ZDE455RRUq1X84Ac/2O89Tz/9NB577DGceuqpB/yM0047DRMTE3A4HDj22GNf3Ym8ynj22WdRq9Vw+eWX48gjj8SyZcv2q8qWL1+OZ599tuu6P/XUU13vOZjv70CxevVqjI+Pd+GZxx57LAKBAG677bYD/s6LE/DLxdjYGDKZDD7/+c9j8+bNGBoaQj6ff9UdxcvF6tWrAQDxeHy/a/BSC/ZrjZ07d2J0dPR1+ezFikM+ob7SWLJkCd7//vfjq1/9Kh566CGMjY3hmmuuwe7du3HeeecBAE4++WTEYjFcdNFF+N///V9MTU3h0UcfxX333QcAOPvss9Fut3HJJZfgySefxPT0NJ588kncdNNN+P3vfw8AuOmmm3D//fdj7969GB8fx3/+53/CbrcjFothYmIC119/PZ588knMzMzgqaeewu9+9zshrU499VT09fXhggsuwMMPP4zp6Wk8/fTT+Pa3v41f//rXABYghUwmgyuuuAJjY2N49NFHcdNNNx30dUgmk0gmk5icnMR///d/47zzzsMDDzyAa6+9Fk6n84C/s2bNGlxyySW48cYb8dWvfhVPPPGEHP91110nFfvatWtx4YUX4uabb8bXv/517NmzB9PT0/jZz36Giy66CEcffTTOPvvsA/4Np9OJ3/72t9i6dSv0+jf29hsYGIBOp8Mdd9yBqakp/PrXv8Y3vvGNrvecddZZSKfT+H//7/9hbGwMjz32mFx3Vl4H8/0dKDZt2gSdTodnnnlGXrPZbPiHf/gHbNu2Deeccw4efPBBTE1NYefOnfjXf/1XfOQjHzno84vFYjCbzbjzzjsxOTmJRx99FNdee+3rUjEODAzgjDPOwBVXXIF77rkHExMT2LFjB37605/i9ttvf0WfNTExge3bt2N2dhbAAmyxfft2lMtleY+iKHjyySfx7ne/ezFPY9HjkG/5X01ce+21+NrXvoYvfelLKJVKGB4exm233SYJzWaz4a677sL111+Pz3/+86hUKujt7cUFF1wAAAgGg/jRj36EG2+8EZ/5zGdQKpUQCoVw5JFHShtlNptxyy23YGZmBnq9Hocddhi+853vwOVyoVqtYmJiAl/4wheQyWTg9Xrx7ne/W/SzFosFd955J26++Wb87d/+LbLZLHw+H9auXYvjjz8ewAJeddttt+G6667DaaedhqVLl+IrX/nKnxTAAwsqg+OOOw46nQ42mw29vb3YuHEjLr/8crkGLxWf+9znMDo6irvuugsXX3wxarUaYrEYjj766C42/POf/zxWrlyJO++8E3fccQdarRaWLl2Kc845B+eee+5+7aI6CJu80bFy5UpcccUVuP3223Hbbbdh9erVuPzyy3H++efLeyKRCL71rW/huuuuw5YtWzA4OIgvfelLOP/88wVaOpjv70Dh8XjwoQ99CPfee29Xe/+ud71LEtFXvvIV5HI5BINBjI6O4pprrjno8/P7/bj++utx44034u6778bQ0BAuv/zyg7pnXk1cffXVuOOOO3DbbbdhenoaDocDK1aswMc+9rFX9Dl/93d/h8cff1z+ffrppwMA/u3f/g2bNm0CsIB9VyoVfOADH1i8E3gdQqe8Hv2AFlq8jeKJJ57A2Wefja1bt3aRXq8mJicnccYZZ+DnP//5y8IDWnTH+eefj40bN0rRc6jG27JC1UKL1xL/8R//gZUrVyIcDmNsbAzXXXcd1q1b95qTKbAASX31q1/F9PS0llAPMkqlEtavX/+6VdqLGVqFqoUWL4obbrgBP//5z5FKpRAKhXDMMcfgi1/84iHPMGvx5oeWULXQQgstFinediy/FlpoocWbFVpC1UILLbRYpHhZUur1mnjQQgsttHgrx0shpVqFqoUWWmixSKElVC200EKLRQotoWqhhRZaLFJoCVULLbTQYpFCS6haaKGFFosUWkLVQgsttFik0BKqFlpoocUihZZQtdBCCy0WKbSEqoUWWmixSKElVC200EKLRQotoWqhhRZaLFJoCVULLbTQYpFCS6haaKGFFosUWkLVQgsttFik0BKqFlpoocUihZZQtdBCCy0WKbSEqoUWWmixSKElVC200EKLRQotoWqhhRZaLFJoCVULLbTQYpFCS6haaKGFFosUWkLVQgsttFik0BKqFlpoocUihZZQtdBCCy0WKbSEqoUWWmixSKElVC200EKLRQrjm30AWrxzQqfTQafTwWQyQafTAQCMRiPMZjMURYFOp4PRaITBYEC73YZer5fXDQYDWq0WOp0O9Ho9DAYDAKDVasFoXLiNO50OOp0OTCYT2u022u22fFan05G/3263UavVoNfr0el0oCiKvKfVar1p10eLt35oCVWL1xRMUkyQBoMBfr8fDocDOp0OTqcTAODxeFCv11GpVOD1emE2m1Gr1QAAwWAQBoNBkqHdbkc2m4XZbIbZbEa9XofBYECj0QAAWK1WWK1WNBoN+Ty9Xo9isYhyuYyenh7U63VUq1XY7XY0Gg0YDAaYzWY0m020Wi3kcjlYrVYoiiJJtNPpIB6Pw2g0yuvtdhuZTAbVahXNZhONRgPtdhsAoCgKOp3OG33JtTiEQ6coivKSP/y/h0QLLRg2mw0Oh0MqyVAoBJ/PB4vFApPJBLfbjVarBZ1Oh3q9Drfb3ZWQ5ubm4PP54HA4UCwWUalUpEpttVrQ6/Ww2+2YnZ1Fp9NBLBZDo9FAuVyGzWaDz+eDTqdDoVCQ6lNRFJhMJmSzWdTrdXi9XhgMBlSrVXQ6HTidTjnmVCoFh8MBh8OBcrksVbDT6YTdbkcymZRzM5lMsNlsqNfraLfbWLJkCZrNJsbHx+FyuZBMJpHJZOByuZBKpVCr1dBqtdBoNFAsFrVq920cL5U2tYSqxX7BFttoNMJkMmFoaAh2ux2tVgs+nw8ulwsAkM/nYbVakc/n0Wg00Gq14HA40G63USqVUKvV4PF4UC6XkUqlYLfbUS6XpVpk9dloNOB2u2E0GlEul+F0OqX6tNvtkvRCoRBCoRCKxSL27NkDi8UCl8uF+fl52O121Go1NJtN2Gw2AECz2YTRaJT/AUC73YbFYgEA5HI5dDodWCwWmM1m+TmTND+LnxeNRlGv15HNZuHz+dBoNGA2m9HX14dWq4VarSbXrVwuo9lsolarIR6PIx6Po16vC6zwMo+dFm+BeKnvT2v5tQAAmM1m+Hw+eDwe2O12eL1e9PT0AIBUghMTE0in08hms8hkMgAWklahUEAgEECn00EqlYLRaIRer0ehUJCk1ul0UK1W4XK5kM/nUa/X4fP50Ol04Ha74XQ6USwW0Wg0UKvVYDabYTQape3u6elBKBQCAGm5a7WaJEC27/V6HYVCAX6/X2AFvkev10Ov1yOXy8FisUCn06FYLMLhcKDT6aBUKkGv10via7fbMJvNKBaLaDabqFQqaDQakmwJK0xNTcHpdApMEA6HEY1GoSgKrFYrRkZGkM1mUavVUKvVMDc3h06ng1qthpmZGVSr1TfnS9di0UNLqO/AYHVoMpkwMDAg7a5er4fL5UKtVkOhUEA8HkcymYROp4PZbEYul4Ner0er1UKpVEIgEJAk0Ww2YbVa0Ww20Ww2EQwG0el0UKlU5G+ZTCapFM1msyRYr9crCYeEEbCAlQJAuVxGq9VCJpNBLpeDoijyOdlsFgBQKpVgsVhgsVhQKpVgtVphsVhQKBTQarVgNpvh9/tRrVZRLpdhNBphsVjQaDTQ6XQENzWbzZLsrFYr6vU69Hq9LApM8PV6HQDkuPR6veCr7XYb2WwWxWIRTqcTNpsNtVpNFppYLIb+/n7odDrMzs7KOe/atQu5XA7tdluOS4u3VmgJ9R0SZrMZXq8XsVgMg4ODcLlcKBaLsFqtyGQySCQSgmfqdDrUajU4nU7Mzs5Cp9PBbrej3W4jFAohm83CarXCYDDAarXC5/NBURRUKhVh5Wu1GjqdjrTFoVBIqkCPx4Nms4lMJiN4p5qtr1ar0Ov1gqsqioJCoYBqtYpCoYBGowGfzwez2YxSqYRisYh6vQ673Q6r1SoJnwmMyZYYqKIoMBgMAm0QZrDZbDAajbBarTCbzXI+TMTNZlOOqdVqoVqtolqtwmazSfULAOl0Gm63G41GA/l8HoqioFwuS6IvFArIZrMwmUxSoQ8MDMDr9crfLJVKSKVSmJmZQblcfjNvHS1eQWgJ9W0aJpMJLpcLZrMZq1evxtDQEPL5PPR6PZrNJubn55HL5dBqtdBsNpFIJFCtVmEymRAIBGAymYRUqdVq8Pl8wpC3Wi3Y7XaRHJGhLxaLcLvdgpWymqzVanA4HKjX6yiXywgEAtDr9dDpdFLRWq1WlMtlmEwmlEolSapMfDwnt9uNVCqFRqMhZFOpVJKqjpUnj8dkMkm12mg00Gw24XA45LwcDgcURZHKludst9tRqVTQbDZRKpWg0+kES6XMi3/XZDLJdS+Xy0LQEYut1WpS6fJvJ5NJ1Go1KIoCu90ueG65XIbP50MgEEAwGEQ0GkWr1YLJZML09DTS6TSKxaLAHlocWqEl1LdR6PV6+Hw+LF26FOFwGEuXLkU8HofNZkO5XEY6nUYymRSssdVqwWq1SlJoNBrI5XKw2+0Ih8Oo1WqwWCzCorPyJJ7o8XhgMplQr9dRq9XgdrulPbZYLNLKGgwG1Ot1mM1mOJ1OSRzEOdPptBBHzWYTBoNBkh11q0ajEbVaDaVSScisarUKp9MpUAAAeL1e2Gw2ZDIZSU5+vx/lchkzMzPSupPhJ1ZLcsxutwN4Qd9arVbRarXg8XgEB2ZlzkRKrJfyLOK5tVoNJpNJNK9UK6ivu8VigdFoRDablevVbreRSqUALMAefX19cDgcCIfDAICpqSnkcjnE43EUi0UoiqKRXIdIaAn1bRBerxfhcBirVq1CKBRCvV5HLpfD5OQkJiYmAEDE60wkTFxMCqxm2ZabzWZJEGxNmQh0Op1oPE0mEywWC8rlsrT78/PzkmxbrRZcLpcQQazyiC8SMzQajXA4HGg2mzCZTFINW61WwVTZrlPepNPphOnncakrRh4zk6Xf74fJZEI+nxd9KokrVrc2mw2JRAKKooiaQafToVwuo91uS0seCAQk4ZZKJdHbmkymruqV8izCBaykTSYT9Hq9HDehDJ1OJ7AFk2u1WpWK1OPxwGaz4bDDDsOaNWtQLBaRyWQwOTkpOlwt3rzQEupbMDhRFIlEEAgEsGTJEiGL4vE4MpkM0um0PJRMYkxG1EsCC22vun0Ph8MolUoCD1DXqdPp0Gg0pG1lkjMYDCgUClLhUmrFqSUSK/V6HfV6HaFQCIVCATabDS6XSxJAtVqVhNzpdOD1eqXK41SUTqeD1+sFAFgsFrRaLZTLZbhcLsRiMZRKJSiKAofDAYPBgNnZWQALLXogEBBclKw+AMF5dTod3G63QAxq6RiVDCaTSc690WjA5XLBYDCgVCpJVW2xWORcvV4vHA6HqAMIjfCYqtWqXDNisFwM+H5FUUSeRmma0WgUUrBer8Pv92NwcBDtdhuTk5PYuXOnLABa5frGhpZQ30Kh1+sRDofh9/sxNDQEnU6HSqWCbDaLSqWCQqEgD2elUgEAaTn5usFgEMlQLpeTJMnPZ4XEypCtKj+r0+nAbrdLm280GlGpVGQklAmV7Wuz2cTk5KQkZcqurFZrF9NOUstisUj1ajAYkE6nRQNLbJIyIyZVvV4vlWm1WkU6ne6qqs1mMzwej5wvda5k8+v1OpxOp+CcOp1OhhPsdjsKhYIsLjabDaFQSH6f0128RuwAHA6H4LKsrDk5xokxwhTRaBRWqxWJRELkXQAERmAly/MkkUjFgtlsxsDAAKxWK6LRKMLhMOr1OpLJJOLxOObn57XE+gaFllAP8WBbOzAwgJ6eHiF3FEVBo9GQCR22mep20W63CyvNYOIxGo2iJa1Wq9JuVyoVuFwuaU0pgs/n86JVZYuuKIoQP/V6vWsMlSoBg8GA6elpABCCh3gsf4ewAmf1CTcwuXEx4N9st9sIBoMCWzSbTeRyOUSjUTQaDczNzcHlcsHpdErbnM/npf1nlcgkqGb7AYgGlYw/E6/6OrI6piqiWq0K+cVkziqf19vpdAoJyIECalcJXTDBq3WzrNTV1X+9Xpcx3Xa7LSQXpW8+nw99fX0YGhpCqVRCOp3G888/33X+Wix+aAn1EA2dTodwOIyenh74fD5EIhFkMhls374dnU4H4XAYJpMJuVxOCCEA0kaqTT+cTqeMXVLuZDab4XK5UK1Wu4gaVlxOp1PE9qwQyfaTYFJPJhEXJAbKYLvNyo2YIdlukk1Wq1UIKxqV6PV6wSqZ/FwulySoXC4Hs9ksVZper4fX60WxWBSCbn5+XvBYg8GASqWCUCiEdrst58UETpKLSYvTXrVaDcViUdQJ6oSkhiMsFotAApSN8ZgBiKrC4XAI9sy2nNVxs9nsMm/xeDyoVCoCVfB8qfHldZybm4PNZhOMOJ1Oy2hwT08PnE4ngsEgcrkc5ubmMDU1pSkFXofQEuohFkajER6PBxs2bBBccGxsTNp6ssGs2NQz8MTxDAaD4IHtdhs2mw0Wi0USa7lcRqPRkETI6osJkfIlNQPO/wcWmOdcLiesvnpElO06Ezd1mlrNIbYAACAASURBVMQWWVVSNwpAfrdSqUiCqtfrUh2rq0Nii7weDocDPp8PlUoFmUwGkUgE4XAYmUxGMNV2uy1SL7VmtFgswmazodlsCtRBcT+TqrqKZRLkcZGM8/v9gm9yIeDfbLVaCAaDco34WqvVkmPhd2QymYTA8nq9YgqjKIr8nLpaamaBBYKv0+nA4XCIkoCV7ezsrBBe/f39WLFiBQ477DDMzs5iz549iMfjAg9p8dpDS6iHSOj1ejEa6enpgV6vRzweF9yPlQsrQ4rVqbMkS57L5RCJRMTUg+QRp3uMRqO08zQ6yeVyYgRC6VQ+n0ckEhFNp8lkQrFYRC6Xg9/vF5zPZDJ1VUr8e8RFgRdINCZ7trAejwcOh0N0pc1mEy6XS0Y71cMCnLQiZMEkBEDm8KvVKhKJhEAIrOSoAmAS43ubzaZcQwAiwNfr9XL9iDf7/X64XC40m01xwqIQn5NS/BsAJPHyfFutlqgaCBXYbDbBqikb83g8cm3y+bwQYbwGAKTD4PWu1+twuVzyt4l7U85FxcHU1JS8brVacdRRR6FQKGB6eho7d+6U706LVx9aQn2TgzrDkZERVCoVTE1NYe/evQAAh8OBVqslDy4ryBcTKkyExFWBhSRGfaK6BSU5wipMLWRn5UXskbPtJF04CkpmmRUZK61OpwOPxyMJiX6knU4HmUymq6KORCLw+XwwmUxiHKLGZvl3eIzFYlESQaVS6Zqd57UqFosoFApwuVzQ6/VCCpEMKhaLqFar8lnRaFQIJzL6rDQBiPZ0bm5O/i6Pj2RcpVJBqVRCOByWqpAsv8PhECtCzv4T51bjxkyA1WoVuVxOpFYWiwVut1sw02KxCJfLJfAMu4hAIIBGo4FSqQS73S4wDLFdk8kklX8+n0e1WoVOp0M+n0c0GsWmTZuwbNkyufc4ZKDFKw/Nsf9NCqPRiOXLl2PTpk1Ys2YNAAgDPj09LZgaK0bKosxms2B1wAuEChMgHzKn0ykaxlqtJjP1aoNnVqVMBJxDdzgc8Pv9UjmRmHE4HFKFKYoiEibiedSeqn+PBBcVCBaLRQxYKEmiQxWraCZKkjyKosDtdsPv90On0wncQRiCVRgJLVbDiqLIdSOkQU9UJjNqb9mqU3WQz+flHBgcBNDr9chmsygUClIxt1otSbD0SaUEi0w/tahut1v0pOpjTCaTACCjtoR2uJDSbIbVKbFojrW+OIgTE1YgiVkul2VhyeVy2LVrFzKZDILBoNyP/D0tXlloFeobHAaDAbFYDO9617ugKArGx8exb98+WK1WZLNZKIoCj8cjhspqYw7qNzn6yMqUn1soFOBwOOR33G43ksmkJB0mCZfLJVUXPUOZZNrttiRGynJYwREXtFqtSKVSXQYhTFQ0FKF2kp/vcDjgdrulygMWSBpKhwhd8HNyuZwkHBJWFotFkj8AqWi9Xq/grFQZELuMRqNIp9MAAJ/Ph1KpJBaETIbqRGUwGERJQfzX5XJJ0qfPKeVnaqMW4IXOgMm+2WxK1WgwGEQHXK/X4fF45LyIPVNjSqkX1QBer7druILXt1wuizcsuwR2FG63WwYt1O5ZhHGq1SoymYxgqAaDAS6XCytXrsTSpUuRTCaxb98+zM3NabKrgwwtob5BYTAY0NPTg9WrV8tEz+TkJBKJhLyHDx5ZYGJaJFEoBle3dGpdJlvJQqEgY48kMxwOh4jnKXXiQ0Vmmu04k5nL5UI6nRZLPiY4bidSLpfFkMRutyOfz8PpdCIQCKBYLKJUKiGbzYrAnRABK1z1mClHOTkzr2b8CR/E43GEw2E5J7Ly0WhUcEdOLjGZEnc0m82w2+0IBoOCZRJTXrJkieg6mVApxyLGajabkUgkpDJWj59ykSCJxekwTlaxwqf+ldU0rQ2dTqdAOervm5IvLiRqGZjBYBByjt8NSS+2+IQwuPARfmFSVuOrRqMRhUIBADAzMwO9Xo+enh4MDQ3h2Wefxe7du7UprIMIzWD6DQiHw4GRkREsX74cOp0Ozz//vLghUYPIRKWWDVksFmnVg8GgtPXpdBrhcLiLdODvs/ogTsr/VmsrgYVqLZ1OS1tJ/JNTQO12G5FIBLlcDul0GlarFcFgUCowHj8ra8q21OOsNJXmAkKc0Gazwe/3S4tOjNfv9wt7nUgkpMoEXtgviqQVyaVKpSLHBCy0y5lMRggu9fXlgkNcmRguZVM0hKb/KSVlrFyZ+AmjUNNJ7Fk9zmuz2VAqlZDL5RAIBASqaDQa0nnwmHhOhAf4nVOHq7ZJJDRBDDSTySAUCsFms8kgBBM3x2C53QuvA9UVJCN57xAqUisN+vv74fV6USgUMDExgcnJyQPCC++00Aym34Qwm80Ih8MYGRkBAOzZs0f0jZxtt1gsXeyquh1jS6oWlrOSZFtN2zn1nkuslLidCB9kttLhcBhWq1W2K6GLPlt2kkCVSgU2mw3RaBRGoxHFYhFms1lwxXa7LRguAHGeUisAOFnFxMmHlpM/JpMJPp9PWl6aTtMbgOOVhAaYaIl35vN5ABCNJY2tmRjp/crJMB4HoRHCGTxGJtelS5eiXC5LEqVEzGAwYG5uDu12GzMzM5I8c7mcVJxsuTmBxSo3lUrJ9i/0fyUcQ73o3r17pQWnXSIJO15rHi+9ZJkgWfnyvLgIud1u8TMgaVgoFISk5Hgy7RAzmQzsdjsymYwYgTudTgwPD2P16tVSsWo61v1DS6ivU4TDYQwPD8voYrPZRDweFzE3cUrigzT7YCVJ9jgWi4lkiYmELkwkQjiRRHa9XC4LfjczMyNVIplbRVGE/SVhQt0qJVCU83Q6HQSDQZlqYpIhjsgHmPgnMVD1TqRerxe9vb2CcbLyJozBJD41NYVCoYB0Oi3C9pmZGYEkUqlUlxk0K0MmBaocmCjYwgOQ9p5VvqIoSCQSsFqt0uKrMVX+Ht2guGuA1WpFKBRCs9lEJBIBsFCZBgIBwWpZXXLMtFQqdSVwLlzES5mYKH0jgcdqNpPJiHk3K8pKpYJIJCILEzF2eguozbNJXHHRoCkMAPkeSYzR14BmL+wq2An19/dj7dq1sFgsmJycFINvLRZCS6iLHD6fD729vRgaGkKj0cDMzAySySQCgYDc2ExYfOg5O09mnyw+JTiUJpHN9fv9AF5oO9QVIdlwbjOSz+eFeabu1Ol0iueo3+8XXJAVMMkpu92OarWKfD4vmCEF8MR5aRBC0TyngtLpNLxeLzwejwwVqM2dOWM+OzsrVVMymZR20mazSTUPoEvypR6fZTLh1BZJFzWJRpZbfQ3YXtNOz263S0VbrVYxOTkp3wsXC7bLPp9PcGFgoTJ0uVyoVCqC6xqNRkxPTyOTySCfz8Ptdgt7TwNqyqE4AEGNKseGicPSgJtYKitxKgZYcXIKi+oCkoUU/HNBYeXKCTJ+l2qyTa3tpUKBJJbf74ff78eSJUswNTWFXbt2acMB/xdaQl2k0Ol06OvrwwknnIBarSatKQkBViWZTAaBQEBkMPTupI6UkzqBQKDLY5M4V7lcRigUgslkgtfrlQpKTSqVSiXRjQaDQRkf5QPKCpjyJlaulDVRh0myicQV7fq8Xq9gctx3qlAoyKy/w+GQihsAQqGQzJoTH56ZmcHevXtRr9eRSqW6LOzUGDL3dKLTFN2X2u22YIBss9VyJYvFIiqCZrMpJBclVBzF5OQZqzZiyrz27AKMRiPm5+cxMDAgiZCQCbFYl8uFQqEgxtcWiwWRSEQ0qRwKYLfBipaTY3TOUmtY+b1wbNXtdsPr9Uoi9fl8gnOTDCOhxYqX9ycVG+opO+LDVA1wAeIgCathLlxcdGmIs3HjRgwMDGDbtm2iJngnh5ZQFyHMZjOGh4cxNDQkWtJEIiFTNrRwI07JNp0PDsdB1Vt+8KFXVwh8eMnmqltHtmZqET2rYFY5rLK4o6her0cqlUKz2RQIggw2K01OGcViMUQiEfESbTabACBEFfFQaibD4TByuRyKxSLi8TimpqZEosM9qwqFgjhX0TCkWCwCQJcYXVEUZLNZGAwGDAwMiF8BFQC8nvl8Xqav+Bnq5MjpJ1b9rPB5rZxOJxKJRJcGk1VxpVJBPp8XzLvdbqO/vx96/cIeUdVqFaFQCBaLBVarVSrhUqmE4eFhWeRIss3NzaGvr08WN7WjFYCuBY/3DD+b/2aVqTZT4SKgJuF4PMlkEuFwuGtwg0QVYSMa3XARYfIF0LUbLTuY6elpOJ1OnHTSSRgbG8NTTz31jlYDaAn1NYROp0MkEsHq1avhdruxb98+wbuq1SoqlYrst0Q22Wq1yiw23Yf4QPCh5wPOG5wJqFKpCLuvPoZ2uw2/3y+aUwBCLFHDyASsnkjiA0lXey4AlBt5vV5hi4kd0jeU7k0krshUU5rFiZxqtYqJiQmRUTEZsn3lrqMcQaXrEllwQiIcWyWTzgqcW7MQ4yWeSrzW5/NJpaxeLNhqE+PmIsaFggMPatkRBf3pdBrBYBDJZBK5XA7T09Pw+/1wu90C1XAHV1aCfI0EH2EBVsVcSLLZLBwOB2ZnZ8WEhtI0Tn7Nzs6KbI7aY7fbjVqtJsoEtvCEIICFyS+PxyMSKd5z3OOL1oYGg0Eqey5OaoWCejHm9xAIBGSvsm3btomm+p0WWkJ9lWE2m7FixQps2LAB1WoV8/PzmJqaQrVaxdKlS6ViIubIVdvj8aBYLCKZTIp0iCQJK00y3NQSUoLD6jMSicBut4tIXL0PUSaT6drzyOfzIRwOy/YfTC4Oh0OqJWpHiYMODAxIC8xkUqstbH9M7avH4+k6J14DGq/kcjlkMhmxpCsWi11SI1ZfAGSMk4Sbun2nhtbr9YrhSTAYFE0rCSMAXeOh9A3lNbJarbKw0GWr2WwiFAohkUgIBqtebHgeJJl6enrQai3scqooClKpFObn52Gz2ZBKpWAymYR8Y8IkAcfqn8MNNIymxpYG3itXrkSz2cTAwAAKhQKSySR8Pp+02gAQiUREddFut2UR4PfOrgSAnJPaGIfXyul0CjxA/JiLU6PRkEk4taRPbRLDCnZ+fh6FQgGJRAIejwerVq3C3NwcxsfHuywP3wmhJdRXETabDccddxx8Ph+mp6elWuODSNJEjUmRkWbLSXE1EyNbvJ6enq4tP1ghUULEqk+v10ulCECck4hhkZhgFeTxeNDb29vVqvf19UnlQaae9nVkp0nYuN1umTOnBIttKPd0oj7zD3/4gxwnq6tcLidTTHa7HXNzc/D5fGg0GshmswgEAlId8XoCEJcqVvf1el1YcDLaJFCIQXIRY8LgNBWZ70AgIFWY0+mE2+3GM888I4mNyYh6T0IP6jFVtvBUEjSbTTzzzDOYnp6GyWRCX19f1x5aPp8P0WhUtqEBFghM4pX0UggEAjL+y+MnqafWx3LPKQCCi9OHtlarifaVfgwABKtnFc3vhsMBfJ04NJM4R36J1fPa8+fE/9PpNNLpNHw+H4aGhhCLxbB9+3YZqX0nhJZQX2H4/X4sW7YMXq8XuVwOiUQCmUxGSJNKpYJ4PC44HOfcyaTTeMJqtYpvJx94NSbKOWtWaJlMBjabTVrYQqEAt9stD58aV+Mcvvqze3t7EQqF4PV6ZeY/HA7DbDYLGcQ9l0iEEZNlkiZ77nQ6MT09jUQiIWOa6XRajEQmJyelzSQBwyqNWk5OD/HvEIqg1R/lVfV6HbOzs3C73YhEIigWi5ifnxf5mclkkm1VqD6goQw3taPFH2EGYAEbdrvdmJyclDaZxCKVEdT48ljVuGI+nxfGnLraYrEoBiZckCKRCHp7e1GpVDA8PCz6XjL0VBlQz0oYhhUy/V+5TQ21xISLarWaLGa8hhaLpcvrlcmXVTsldJTJtdttuN1uwWv5XRNbJUbObVWYuHmvsCqm7pjV7pIlS9Df34+HHnoIk5OTb/Sj+qaEllAPMvR6PYaHh3H44YcjnU5j3759Ug0QN2TFSR2l3+8X8okVJFlWjiyq901yOp3wer1iJkLmnOJ6m80mlW0mk8H8/Lw4MnEXTVY2RqMRgUAAsVgMfr8fdrsdTqcTfr8fExMTUmGpkyq3dSYODECqQm6pbDQakUwmMTc3h3w+j1KphFKphJmZGVQqFUkOxDTVrDpbQlY49XpdlAg8/lwuJ9gmpUGsQLkIxONxpNNp9PX1we12iyE28b5gMCjYMdtXai1dLpeMyVICZbfbxcZQDUdwft9oNIr7Pgk5tusUxNMakMdaKBSQyWTQ6XSQTqcRCoXEtm/JkiWSmGmxSEkXJVC89lzEuMjUajW4XC4MDg6KamBqakokdewquJAVCgUxpPH5fAAgiZGYNvF9Jk3i1NQYk8ikbyynxqrVKnw+n8AY6t0E5ufnJVFv3LgR0WgUzz333NteXqUl1IMIs9mMww47TAyfKWGhDZrL5ZJti3kTkXxhUmHrTgkPV3QmLLovJZNJOJ1OabOJQfKhIt7IFozMr9Vqhcfjgd/vh8fjEXMMYrZMGpzaYUWcSqVEgkNZDM+D7u9cOOgGNT8/j7179yKfz6Ovrw9zc3MibeJ5EQIhPscJMJ6n2+3uMoDmQkOIhGOj2Wy2K/FQQsZKs9N5YasRn88n0iCSgU8//TQajYbIxmq1GgYGBoTUIcwwPz8vgwTqQYVQKCT/Jr5IPJQmKzQoYTVJBpxMOhcVLqq8N1auXClmNeVyWe4F6kHpzs9FlIQQVQWc36eygl3C/Pw84vF410ipuroGINNThEb4N0nksSIn/ETlgNpcmwtjoVCQe5jXiIXF/Pw8hoaGsGbNGoyMjOCee+4RAu7tGFpC/RMRCARwwgknwOl0YteuXdKmA5A2me2SerthjoPyBuNDPjExITc5TT5IsNA2LplMytgqEyiNOKhTbbfb8Pl8sFqtGB4elgeQiYMJt9FoiA8oRzCdTqcQEMViUfAwsup8SOfn50UWMzc3J6YsPFeK4FOpVJcFn1qLyGTPRMeJJ4PBAI/HI0ww7f/UJtCsUIlhEmaxWq3o6emR6p7VGyvHSCSCvr4+SfJr165Ff3+/aC2JK6fTaUSjURkFDQaDcDgc2Lt3ryTM/v5+IaA4x66GVCh9o3UfkxKrWS4IvL75fF62fE4mkzAYDPD7/QiHw0JCkqyj5I7EI3WpNLXmQASldwMDA9DpdEgmk0in05iamsLMzIwMR3DYgXi0msAi8UmM1e/3y3YpVGGw4ubiTntIytXIIbjdbqnay+WyjNwODAzghBNOwEMPPSRV+NsttIT6MhGLxbBu3Tp4PB7Mzc2hXC5LciMY32q15IaknlPNDJOFZcXHjeSYUEjeFAoF+Hw+eY1B+Q9bUbLNZrNZSI7e3l5pU202m9i+tVotTE9PyzYnfCg4u015ER8sTusQS+PkTqPRQDKZFDcoYnzEhPn7rOJYcdKtyePxdIno1VIiAIIZs+JTC/w5C89jZfVaLpdl6oxtv9pwmdX9cccdJ6Yr1IpSrjU1NSUSKOLXPp9PcF2DwYBoNNpFFs3NzcmYJ1Ucer1e8GZOsdE8hnpULkQkgrhhIb9TyuNsNpvsEEsykRASIR/1fWexWAS2MBgMSKVSCIVCsqtCNBqVoQE69nc6HblHWbXyfqb+mLhrIBAQlzFixbwv+T8SamrnLPWxtVot7N69G8lkEna7HatWrcKuXbveltIqLaG+RCxduhTvf//7kU6n8dxzzyGTyQhWR+MS3qhsgyhTIonBSpKCaPXOmEyAtVpNfk73fVYeTCB2u11MmUnCOBwOhMNhaaH5QLMdVpuvUBFAuUulUkG5XBb5Dw2tSUiwdSf2SQyNuOj8/LzoOTudjiR0MvNqLE5tgcf2lVghK2OTySRMsjrJk5EHIAmc50DTaVbhAwMDiMViaDQaWLJkicjWQqGQWPVRe0mihY5NTKScrQeAqakp7N69G08//TQ2b96Mvr4+9PT0CFQxNzcn50APWy6gHOekbrRarcLlcokEingkJ8sqlQp27twJu90Or9cr9xfJLUqVAHR50pJg4vXm9eSWKP39/bDb7UilUgITkTgDIDKyarWKbDYrk1hcUNgV8RxZSCiKIiOulLapYRhyA+FwWIYWDAaD2EZyz7S9e/dibGzsbZVUtYT6otDpdFi9ejU2btyIVquFeDyO8fFxVKtVeL1eeQBYlbKi83g80karReLq6oStIcH9crmMXC4nzL1Op0MqlcKSJUtEB0q3pVgsBo/HI2bHrFoCgYDs9Nlut2UIAFi4saPRqEz7cE4cgCS8dDot1SfxT2BBezg1NSXVKEdJ2dazsmaLylaTo5FMpkykALo26eOEFZUM9EHl3+fDm06nhUgiJqnefZS7sJIIIzaq0+lEJkTCi5hxuVxGMBiUHUU50kqjmNnZWTz77LMoFAoCexx55JFYv349+vr6ZBiCwxt076dRSaVSQTAYhN1ux/z8PIAFHS+TH78nkj6zs7NSoXc6HUxNTcHhcGDJkiUy+MDFRQ33cFiA8jy1VI5VIxMtiT8a9ZBQnJiY6NLJspOhSQuxZHYVxJPpvUDlAL8z4rXU1VIN8GLj7UqlgsHBQTFZf7s4V2kJVRUmkwmDg4NYtWoVSqUSpqampErkza/eu4k6QbZhbNtYOZB08nq9yGazIrpn26tm+8mWkpAiwRSNRtHpdNDT0yNYYaFQEOw1n88jFotJorfZbEin09K2M6lRCkQmOJPJyMNIA2m1NAh4gYhg+0mhP3HNRqMhDx3bWE7mkJjgaGmr1ZKJHPXDy+vHEVLKkLj4sJqzWCyiMmBi5kRZvV4XWQ6hjFgshnK5jGKxKMmE0iYmZioqCNXwgVeP1vI6cNFUKwRY4YVCIakMM5kMAIjNXqPRkHl7su5slXntgAUZl9lsRjKZlCGH3bt3C/GYTqcRiUQwOjoqng52u13gFTU0wAWa9yTdwiwWC5YtWyYTXzt37hQzHWqIyeLPzs52jbpS+cFdBrggMLioUilBK0AeP4kr3rfxeBw+nw+bN29GKBTCtm3b3haVqpZQ/y9MJhNGRkZw7LHHIpFIYHp6GvF4XMyQOb304ikcYpxMjgT9qacEIGy2Xv/CFsB88MnIs3UmKx+NRgW34w6b/B1WLcViEeVyWTa5I6PMlpokA+U1MzMzkhA5w+9yuUTXyGDip+ogFArB5XIJW8+Hjg8bAJGOsWXkRBbbUxqB0PqOm+fZ7XbBprmY2O12Id9o+qyeGlPDEfwdwi8cnNi5cycAoKenRyrQVCoFvV4vI7WEQijl6XQ6iEajOProozExMYFGo4FVq1bB4/Ggv78fkUhEFANM5CRpVq5cibm5ORnZnJubk4WDyg+20M1mE/Pz8yKkJwZLeIekE1t3wku8T9Vjyg6HA7FYTBK42huXBJra1YoQid1uRzQaFaKS3Us+n5djo2aZLlp05OJ9G4lExA+AAwscriChyC6uWq2KmgSA3K/pdBpHHXUUjEYjHn300bd8paolVCzIolatWoX+/n7MzMwgl8tJpcmdJDlqSes9PsBk6FnNEXPkjV0ul5HNZsVsmBUZ2WZawdVqNUQiEcRiMSQSCcFr8/m8sP9krbmFBxMLb07inMFgEKVSCfv27RNsrN1uC1bKmXxKXgBIFUiogpUwH26v1wuj0ShYmtrnk0mAQm+eFxchkjokeGjMzErY4XCI9IntsNPpRCQSQTKZRLFYlGNgMiBOSX0ufVNpGZhIJIRpZtIn/seBCJ4zrxcHEFqtFjZu3CjTaVarVRYenh8XFKfTiXg8LgoFHisXrb6+Pni9XnHlYoLxeDzIZDJIp9MigWKSZKvM74O4b7PZxOTkpHicxuNxAJDEFIvFAEAWJ7bo/D7pl8BqPRwOi3dCoVDA3NwcgIVkR1csdizcn4rXh7g4nwe6p9Gchu5nVBGoFz2HwyFDCrOzs9DpdFi6dCkymQz++Mc/vqHP/mLHOz6hmkwmbNiwAR6PB8lkErOzs0KA0Km9WCzC4/GIMxDJnlKpJHsQcU7d6/VKFcLWju0t2x4KrHt6ehCJRDAwMCDtFjHSZDKJfD4vNnasSljpUO5EkobmGPl8Hvl8HtlsFpOTk5KImbw4c87WlXInVtGEL8h6Dw0NiWwKeMF7lcmSCZAYXqVSkesFvGAgQ00skw2vo9rpCIBUrjSYoRk3sVxuX+31ejEwMCAaSqPRiGAwiN7eXuTzeaRSKXg8HnHv7+3txcDAgMjauPDUajVks1mpHlkxk6Ti9BKnwNjqEqLhsRIaCoVCAkuwqu3v7xe4hFNtK1aswMzMjFTvhEsIJ3Ch5PfL9pqLN0nHTCaDffv2IZlMYmZmRjBkbtNNQkhtHEPZEzslkoEOhwMul0vIP7PZLA5hXGzYORDvL5fLcLvdcr14n3OcFnjBv4I+r+phDd6r/f39GBkZgd1ux1NPPfWWrVTf0QnVarVi/fr1WLZsGfbs2SO7O6qlQRwRbDabUkFxy12acTDpUWpEEbbT6YTH45GR1EajgUQiIc48ZOpZATAxUbBNrSEfAt7krA74eiKRkEr4xe1/pVKR6sBgMCAUConHJc0yqJcFIGL2SqWCXC4nbZo6iaulPsTxdDodenp6xIaOJFw0GhUZD2EBYCExq6Vb7XYb8/Pzsg0KK3xW5DTfoGaTkAsnoVhpcTFTLyLAwoPr9XoRDAaRz+exfft2gQ2SySSCwSB6enqEDFTPzzOJ8lpFo1HY7XYkk0kh3Lxer0ykcUNEp9MJl8uFvr4+mSwKBAIolUrixzo7O4uZmRlMTEzIPcaBBo6QsiJl1aomCalTNpvN2Ldvn8AA2WwWPT09IsNikubILe9X8gHUM/P+6+vrQyaTEfMT3udMnHQPo98BFz4uuGqtMklCchC8j6kbJn5fLpeluPjjH//4lvRWfccmVLvdjqOOOgrRaBRzc3Mihm61WqLhozbT5XLJ1s4U4DMJ8qacmZmRiotVIJMU9Xs0JAkGgxgcbux1+wAAIABJREFUHOzS/3FrCq7MrHyJjakNUwCIQUculxP7t3g8jmQyKVIpYmkul0uSDHWhDNrFqd2b+FBybDKVSgkWx/lxCsp501OMT0MQRVGkiqMVIR8cJjom9kKhgHK5LJgoEwmrOnWbXqlUxJNAPQ9PbJd4dE9PD5LJpEAUs7OzyGQyGBkZESJofHwcwWCwyz2/UqngoYcegtlsxvr166V6AyCjrdxOpdVqIZVKIRgMIhaLYW5uTvBmmofTTJzb31C3yetDu0TuVUXtL7dMyWQy8v1wJJkTeDwmVqxqzJJbzwCQQRH62nIgwefzCTlJMT4A8R7odBY2buzp6cHu3bsFmsjlcqKOACCVKb0DfD4fstls1y4BaukXg69xzDiVSiGXy2F0dBQWiwW///3v33JE1TsyoRqNRqxatQqBQADj4+Mi7QgEAshms104DzEotpu0j2P1Wq1WhS1nVUIHdxIqlFstWbJERNvqbZKJM1JHyukqPnD8W7T24xhosVjE9PS0VFqlUkkedh4bySx1y6gWyau9WVmpsu03m83o7e1FIpGQ6ontItvXYrEoI45kzp1Op0wO0WmKrlCUWjmdTiGq1N6xrHhZxanhAC42NMxWb1AHvGAoHYlEBL/2eDxYvny5nCsfdMqw1JUtDUZ2796NRqOBUCgkOCQhoE6nI+bhsVhMqj7uxeV0OtFut8UOkd6mO3bswODgoBwLRe2sCMPhsKgtSGby2pAc4oLEiThCGlxQKOXS6/WCs+dyOWSzWSFJiUVHo1FRfVAyRw9WLlb0HIjH47KA0JWfixIJJyZsLp6dTqfrWpMMM5vNQrSpcVYSXyR8g8EgBgYGBDN+q8Q7LqGazWasWbMGS5cuxfj4OKampiQBkKDgA869h3jzkA0mEcEkwTl0Ym3UPVI/6fF4MDg4iFgs1kU0AAsz1fTUpJRKvVcQqwkywWx96/U6ksmkbFzHtpTu9sTGWPmxXVUnZrozkeEnRMCkSrKM1avL5ZIKkO0mEzkxV8IcPp9PJoEqlQrq9ToikYgkYeJqiqIgGo3KljGsjBwOh/iPptNp5PN5WWw4ZcbFjtU3mWQAsqsnuwO2nLQ7JHZLzwDqgUl2UXhPaVQmk4HT6UQ0GsVvfvMbmfghpsj/J5bO9n9iYgI7duxAKpUStyy/3y/uYBTy22w2rFmzBrt378aePXsAQBYMdiLq43Y6nVi2bJmoPLhThMlkQiQSQTqdlqqVnQXvFUIlU1NT0pEwSRO+Ic6t3vmW37Hb7UYmk0EymeyCVegxS+9eAEK+Uu1A2SAJQi7s1WpVSN7JyUlEIhGsWbMGRqNRrsdbId5RCZWOUaOjo0ilUiiVSmK3RtzUbDbLbqAkFwAIy0os0ufzIZ/Py++wFVV7nfb09MhIJNlPtkFscV7sPsUJFY6qZrNZ6HQ6TE9PY2pqChaLRUY4WZmySna5XFi5ciXGxsZkbyO1+QqTFX+PUABHKSkjYhL1+/1SZVLsPTk5Kd6lHIWk5pAkHGfGSawNDg7CarVi165dACAjkgAE/wsGg8Judzod2WnT4/FISz49PY3x8XHZj56JiL6f09PTiMVisFqtmJubw+9+9zs899xzOP7446VLACBDGXTfImZJcmvjxo1SbTcaDczPzyOZTAop+Pvf/x4jIyOCt3MqiqoIAHINd+zYgZ07d8rUErFP7nZKpQRn+WkiTl8A9ZYwVAHw+6Ksirg9FQ3c5ppaXeAFqIiV+Z49e2QRpwaWhCqHKnhPsGjQ6XQYHBwUQT+3MQeA+fl5UaPQipAEF/CCHSAASaIkv+hxwKKmUqkIbLZixQqxxHwrtP/vqIS6bNkyHHHEEfKAUGvJ8UO1Kzk1gcTWAEibzXaUuCAA2ROKkiGaQbASIQ7XbrcFkyQswO1DwuEw9u3bJzdPOBwWd6dkMonp6WnUajUEg0FpwTi1RC9MtpokS8jiE++l9pNYJsX8bH0BiIUeKxVW20ajEeFwWCaVOOnFCp6VMuVbrPJZrfBa03iDDy8AYX7tdrv4zHK7lXA4LC2r3W6XnVIpiwoEAtIZqP1bWb2122184AMfkEWTsAmTAbuQSqUCt9uNww8/XGCVnTt3YmxsTMTzVqsV69atw8DAgLDq1CJzP3tCOWr1AStnYqS81jSTob2g3+/H0NAQJiYmJElx1BhA1wJPHwImxmq1Kh2QeuCEo7U6nU48CkiSKooiig2O8vJeBRYGDrjjLhcPRVG67BlZGNBPgNaRdNUiDMDvi/fOi20BuYBxV1dOmW3atAmPPfaYyMQO5XjHJNSBgQEceeSR4gjPiRafzyfbX5AxpYC/0+kIK862lqOXdPgJh8PS7gMLCay/vx9Lly5FX18fKpWKuApRmgRAyAnqL8vlsuC0iURCWHbqA5kIs9ms4LJ8aG02G4LBoLgLsZqmCQZnvAFIoiBeylFJjg2ShWWLVigUxL2J7THnv9miqaVhbHnZHrLKVRuJ8He4BTU/h60fsFBRjY+Pw+PxyN7vxOK8Xq/4tzKRms1mwU7NZjPWrl2LXC6HZ599Vog47h9FkoTVu9rxanBwEOFwGMlkEvv27cODDz6IRCIhsiuSVdzdtVqtykQd9ZmEIXhPdTodYdtJGvr9fpl4Y1uv1+sFdioWi1IhcsbeYrFgenpa8GAubJR+0duUOCtJTaAbf+YoKLstTi4BkF1oySlQXsb7nVAP988ym82iK2ZypbuY2vSbEjtWn2qFB6tVPnfU/tKuEADWrVsnxcyhHO+IhBqNRrFu3TohV/hQ5fN5cXgqFotd5iLEBiniZjVJN6hyuYx0Og2n04lQKIS+vj60Wi3EYjH09/cDgGCFFIGTHSdzz5ub24tMT09Dp9Mhm80in89LYhobGxPTaJ1Oh3g8LgmcFQdxV1Z6NLQgjsrqZ25uTsT8rHzUFRYrKiZgh8Mh21lw4VBrD7dv3y7eBqyepqamBMogFABAfA9Y6fBhohyN5Ia6cqWpCVtVdgQ0o261WrK1NmfMCX9s2rQJ69atk3OnUUmr1UIwGBQDcOKMLpcLoVAIAKTCp3ELCahsNivTWerqOp1OIxwOS0fg9/sxMzMjixalboQwuHhwJJT3i8FggNvtFvyYPq1UnVDKxX2xxsbGMDs7C6vVKtUyuxcSQKwAqZvmgkqculwuY3Z2Vs5JnVy52IXDYbnn7HY7AoGAkEx2ux1+v1/kVRwMUI8eEzpIpVJCRjE5q+9V7rFFI+1Go4Ht27ejr68Pxx57LH772992ubEdavG2T6ixWAwbN25ErVbD3r17ZS8d6vLUVQDbQILobNXppk/sSu0kRQB+xYoVWL58ueBs/Ey6VNXrddnfnsnZbDaL+TNHBefn54V0IlZLM2DOk3MWHYC0X/F4XPYRmpubk1l+JlSu+Kx2KBonywtAKgLih9VqFbFYDLFYDOFwGPPz85Jgli1bhmw2i/HxcSE/6GjFa0LmnLgayT/gBSyP+C6F5Vy86EzEZE9HLlb1FIlTxjY9PS1Who1GA2NjYwiFQiJb4i4LTP7cukVNXBHaIHPd09OD008/HZlMRsgSv98vgn1WXXSubzQamJ2dlWpOnUR5HsRLAYhkj2YrVEpwUet0OkIkUR4HQHwPZmdnxWVMzQM4HA5ZEAizEJrhgmS327v8e9UsPBN8Pp9HOp3GzMwMQqEQPB6PmO3QppCGOtwfi05cLExIaAKQDoznxuRLHJffB5Uv7ITYQR122GE4/PDD8dhjjx2yzP/bOqFarVZs2rQJFotFbML4AAOQ6gSA6P7IcnICh5UB8UT1uKTH45F5ezW2SDKJVUSr1ZI2i9gl27RisSimFtlsFtlsVuanOdnCypktEm94ukSxKiGupa76CCVQ48r2l4MB3K+JDyvhDUVRhKwg4QEstOK9vb3iKMX9nAg9UAJGfIwTP16vV3Bcv98vagHihyTyOEDAc6ZSgsmeu7XSdJkkHsdoo9EoarUadu3ahb6+PoyMjEhyUT+09XodiURCBivosUrSjri2uh1dt24dAHQZ5QCQqSOaSfN6ulyuLq8BGtnwulQqFaTTaej1epFuhUIh2YI6HA5j7dq1QvawayEOy/uQiV1tNEO1CrFsji3zu2S3REtBRVGQSCSQz+dl19tSqSS612w2K1U6J8R4XtQdu1wu2dlix44dyOfzXXaDVFDwnqa4H4C4klE3TUyZkEUul8O+ffvg8/mwYsUK7Nq165Akqd62CdVgMGDVqlXSanMqg8A8W3qC6TS14BQHAGlxOQ5J3IfVn8fjwZo1awQvJG5JcxImZ7U5BFsqWqnNz8/jmWeeEZcjrtjcoZOVmt/vFzyOCgHqNqenp2VlJ27Hh5SmJpzDttvtSCQSWLp0qfxN2gTmcjnRThKb5MPPnUjJyHK3V44Zkukn+cKWT6/Xo7e3VwiPaDQqDzsTFQ1bSJIQw2WLS9cjtu7cRnlsbAyTk5OiN9XpdEin03L83L+e225TAcBq2GQyyfw7H/h4PC4KEL/fL3pMLgSJRALFYlGIGHXLSuNwalV5D1GgTws+Xk+dTodwOCxtOSVvU1NTSCQSGB0dRW9vr8i9EokEpqamxGzabDYL2+/1euX7yuVyslBMTU2JxpcSObU5NuVLvHdpwB2Px7uwdPrgplIphMNhqcrZfdFgmiRWrVaTZ4pFAiEXQlSEiXjvsJXnVBn9JqjRZjW/atUqVKvVQ3Ljv7dlQjUajRgdHZX5fABdLu8vrrgoz+Gmd2wt6UxkNBpFU8kHlJWby+XqYlCJRQL/n7vzbG70vO7+HwTBApIgem9sy61arSRbsmMn4xlP2vtMvk7yhfIqbxKnOB7HGUleq612ucsCguidIAmSAEjweYHnd/amx36sx1E1ZjITr3ZZgOs+1zn/dmSaPyfAfnJyomq1KmmqlaQbhG11uVzmzUdCw+GJx+PW7RI2ze+CZxvWl5EVwqPX6xk7Xq/XVavV7Gfv9XoWKowsBxiAkY2OvdFoWBoR4zNkAw8OxRBsEjII4ojLAjyPUdOZ5wnGxueUy+VMuhWNRjUajZRKpUz7SKdP5GG1WrUukJ8RvJoxly6bSwxnHAWEJKX5+XnLVQCn5LNmHcp4PFY4HFaxWNTx8bEikYguLy/VbDbtrFB40QVLsj+n0+t2u/rFL36hWq2mq6srvfvuu6ZAAHtMpVKqVCrGAXDGmFQY3/n9yeQFOsBKjGmAM8KLc8DvRQeMNlmS5Rr4/X4NBgNTYUCKra+vW54tCWlOFyBTIg3KycnJLaUB7zG/DwQXE8a7775rU8a36fUnV1BdLpfu37+vH/zgB/qf//kfG6kk3epWYazBcbDtIf3hw+TfSbJ10dlsVuFw2PBYPM4cHMZGl8ulSqVimFClUrkVY0cyEd0G+GKlUjFDAA8HxAwpUKlUygo5lsJEImH21kajYYW/1+up3W5re3tbo9HI0pUgHIALFhame4bYrY5ts16v6+HDhxoOp+uiu92uNjc3bf8VIyrdF5iXJINYjo+PTZlAmhehJOl02lhnCgF/l8uNggsuPDs7awRSrVYzedpgMND+/r69z+CAGxsbxp4TXRgOhyXJIAW6LXDgjz/+WCsrKwYjkM8AZAOMI03hI5/Pp3a7bRAIuabkiNKpglPOz88rGo3aBcyFzTqdo6Mjvf3223YGYd1PTk7s93dCLtI0/AZJFjIvJi3OPnivs7BJsgwF3hNp2nA4i6XH47FGoNvtKplMKhaL2dZVyEqcWuD6Pp9Ph4eHVijb7bY9c/AFSPPYKoC9eWVlxS4M9Mjr6+t699139a//+q+3ZF7f9OtPrqAGAgHduXPHVlQgY6HbgXHHo894z8FBXoPUB4cKDx/aRcTOjNQItcEtJZkmsF6vmxuH/8aIRREmRIUOyRk551w/jU1vOBwqHo9ra2vLIAmIAlYiEyM4HA61vr6uyWRiOB4M8snJiUl/CDSBrcciiU0Ru2mn01Gj0dBoNLILC20j3TYdLRpI9LFcHFgMGet4T8knBVaB6CCeDyE+HSRaRyYKSWo0GpqdnbULEBJkMBhY0R8OhwqFQrcsqmCn7XbbpFB0skjbINLu3LljBgQuYggZZ8Tj2trarcAW5F583qRAYS+9ubnRkydPtLKyoq2tLXk8HsPKE4mElpeXtbOzo1arZUUbSR+dKvgu5BbdJ04oWH5wbEkGDzEVQfqhee73+7blgenp5OTklnyPz9VJJvr9fpssZmZmdHBwYBBYLBYzosyJ3VMg+Uy5tLgwiScMBoN6+PChPv74Y7tQvunXn1RBDQQC+su//EtNJhPt7e3ZAy7plveeBwe7Ifo3xkGyRfv9/i2mNp1Om6cbLAhml1Dd4XBoWaqM0dgqR6ORdZuMwIz4l5eXhrHyfxQvHuCZmenO9s3NTeVyOeXzeWNX6Qjn5+dvjUoQa/1+X6FQyLoJ9H6NRkNnZ2fmXHIuWyMpvl6vW27r5ubmLa9+LpezYowcDGyVbiUejxsemUgk1O12LfCaMQ8NpdP6CxRwdHQkv98vt9utbrerRCKhbDZrm00pOBBVhULBHkokcuypZ7VIo9EwKyTZBPwM7NuiUHu905Xe1WrVOmh0v5Kss0YxwvoQOtRWq2XmBKYRPP/8PTDk09NT5fN5PXr0yDpgIBBWxGQymVtuJTBMchQodpwtOkq6TlQSXHJMW9Vq1dbDgNfj/MMmvLi4aGeTbrlWqxnpCDyCVZeCDQfAIkjeV4g06XWT4bR8O/XNPLfz8/OmenjzzTfldrv1wQcffCtIqj+ZgurxePS9731PHo/HpCYURrJLSd8Hd6NDlKaFBAaSEYyRl/CKfD5vhaTT6ej8/FxHR0fmjqIzOzk5sc2khUJBOzs7FlIC5oWUhxg6Hki6NJji9fV1LS8vK5/P23/P5/M2OhHWMhgM1O/3FQ6HDfQnuJnfH82tkyGHRMF2yIFG2UCn2ev1TCqES4nOMpPJ2CgLsdbr9RSJRBSPx20XVqfTUbVaNbUBel0cMs7vzXgKeeQMMkbrSRhzJBKxzh6zRbfbNa2wJEug4n0lSKXRaJglmF1QKAsqlYpCoZCx/X6/3/BlLJN0sYQxBwIB2+XEZIQ6BFhkZWXFullUANI0C7ZWqyn/f8OWmUyGw6FWV1ctUhDylAtbmk5OGxsbOjk50fPnz015ALnIJEHnCOFEV08xckIu9Xr9lpZammb4Mu4TCiPJ4CcCZer1upF7oVDILuqtrS0zThCwDUnIahVGe35edKx8D37edrutSqWibDarSqWicrn81RaZL/D6kyioLpdLGxsbWlpa0suXLw14J6yC0ZuAEAonLiBuzmazaeML8WrJZFKpVErpdFqpVEq9Xu/WSFkoFJRMJu1w3b9/X6PRSJVKxQ4b4zoxgDDyR0dHt5hs7Hhkkk4m05UcKysr5h6BsKJbk2RZqOQBcNjpUCBT8JGTOsTr+vraxvKLiwvrHMm4DAaDRurhCELyBJaHxMbv9+vOnTsqFArWteDdPzw8tEJCVzgajeT3+xWNRhWLxazLwv2EDIhizc/f6/V0cHBgX4sxGoLE6/VqbW3NNgecn59bCA2bES4uLtTtdlUqlRSLxSz+j/Ugl5eXFgoCqYKkB5UFluObmxu1Wi3ThfL+DodDpdNp63I5a3yWfDbHx8d2STcaDV1eXiqTydjISwEl4T8YDFrcJGJ4cGAgk/n5edt4SgeP8YNzjlaaM3l9fa1qtWqravj8eZ7gB4LBoClVTk9Ptbu7q2QyaSu8Kfy43JaWlpRKpcw8AanJ8wsh6Sz2kqygopmFKESit7u7q3Q6rSdPntjk902+/iQKai6X04MHD9RqtWzsIjWJWDMYThhYn89nXQnSGcYiuoJ8Pq+trS35/X5bG4Kw/JNPPjHctNVqKRaLGbNJxwL5wgOOeNzn8+nm5sb82QRC+P1++f1+zc/Pm3UVzSB+e3SaztUdeMN5sJyaT96DaDRq5JEk08SCJaIgODs7UzqdVq1Ws5UWyGqAInA8gT+yaiUQCCiTyRic0O121Wg09OzZM9uu+bvGsmq1qhcvXsjlcpmJIBKJmIaU4A4UGgQ7o0ml8wWTlGSdIwQO79XOzo4uLy8VjUYtxYnPiq0CKysr2tzcVCKRUCaTuSUNo7gSDBIMBq3Afvzxxzo4ODBsVZIePHhgnTidtDQdu4F8Pv30Ux0fH+vFixdqNpv6i7/4C+vqwIoZoz/55BMbt5E+8XnX63X1+31rGlwul4rFojUO4NecF3B9Ol5+d84MqghIKeAsVAuSDKMHQmEzK80Acqtut2sXEg6uWCym4XCoer2uyWRiXf6dO3ds/Q8XGYV8YWHBlAasB7q8vFQikdCf/dmf6d///d/tjH8Tr+98QY1EInry5IlarZbp0hhrnYWIUZXtjjxkaOC4LbnN6SaxXMI8IobHUomuMhgMKpVKaW5uTs+ePbMug9UarLIAqyISDYhhdXXVYu+y2azhUYy5zWbTNlTiP4eVJpCC0Z6OGCKAxKpAIGAYZygUMqkYDwVdCtgrxY/wDTSJLC+MRCLK5XL2fpNANRgMtLe3p0KhYMXoD70gs8rlssrlstxutz788EMtLS1pc3NT29vbpj5gi+jFxYWZKxYXF607oXOimwVH7nQ6phSo1WqWcEVWgiQVi0Wtr6/bJcvYCYnZ6XQsThBP+s3NjQ4ODkxHy1mSZPF4jNaSLKCFDvzg4MAmFqANVph0Oh3LO6Wg7+zs2Nfm67K5gei71dVVu9B9Pp8ikYiN/aRuEYU4mUzk8/kUDodN2kTaFWeCs+Y0BdDh/rZh4Pz83LBfXGPdbtfUKUj74vG4TSLVatWMDS6Xy1K2nN05SgmMKkByp6enevnypXK5nDKZjDkiv4nXd7qgzs/P66233jLGGhcREiDkOniYkWU43T+MZ8iSEKavra1ZlB2iaVha8MfZ2Vnlcjklk0nbrIm9EKIBeIFOlrQkciUp7PiqY7GY4XsItlOplI34dNzNZlM3NzdKJBKq1WpqNpvyer3mf6aYJxIJXV5eGl7Huulut2ujGe4UOlRcSs1m09hdWGAKL1ZAuu0nT57o+PhYzWZTzWZTn3/++f/Kc82qGUby3/zmN3ahbG9v2xYD1A18Bug+kZXhWiuXy9bpOkOxpde+ffA8SRbCgoCfbvzk5ESHh4f2fRKJhGZnZ/Xs2TODmnh5vV69/fbbZmygcHq9XptEwMS73a7u37+vRCKhYDAon89nZopEIqHhcGh62clkomg0atmj7XZb7Xbbpg10ukiTMBrMzs7a+E7eABcuzihUL1youKRYKcPZkGS5qJCYXGTBYNAUFAsL02WAfr/fMlq5tMmXZbKiu2apJRpmNgMzeeLKQnnBz7C6umo280ql8kefvf/N6ztbUGdmZvTee+8pnU7r/ffftwcfvM+JF9GVYJXkgwZrcoZ7zM7Oam1tTblczjzlpJxfXV1pb29Po9FIiURCKysrymazFh+HzY+fA93r1dWVpffTMUhTSVQgENA777xjkXawvzDmfG12IqEbRSwfDAZNyH91dWUkDAWDdSbIeFgIxxg/Ozuro6Mj+5m8Xq8CgYCFlNTrdXO5EG2I7pXfp1AoSJriZJ9++qk+//xz63S+rBfWy0qlor29PcXjcXN78XPDKtO1ZzIZw47RN4LDhkIhSdNOjA6bHFIIJGAUYKThcKjDw0NVKhWTWu3s7Ojhw4dKJpOW2wBG+P3vf1937961iDrMIW632zaoQlixrZWOt9Vq3bLakoHr8/m0sbFhRgrp9QoSiJ92u23LAulGwerRARNwcnZ2Zs0FXSW/P1ADOm1ss8gApdcabQon3SxrXpDKQewxphcKBS0sLCgejyubzerBgwcqFov23vL3JdlKHxogniciKzEusDnj7bfftubn6359ZwtqMplUJBIxzRwjHh3k1dWVWQ5brZZ5rMFtGP94sNDtra6uKhqNKhqN6ubmRqenpyoWi7ZTZzgcGlHldrvtdpVkMqNut6tyuWy3MEv6cMTMzMwoHA5rY2PDDjAOIjokMCUgCrzddI0sYIPYQcwNLMGoRdcBGQIEwu+MkH80GhmLD3tOMjx7h+jW5+amu54+/vhjCz+ZTKaL43Z2dr70Yup8oSfd399XoVBQKpXSj370IxPce71e7e7uajwea2NjQzMzM7p7967efvttPX/+XJ9//rmlifGZk0OL+QB5FKNzsVhUrVaTz+ezbQhcTGCYyWRSPp/PLqTZ2Vm99dZbWl1dVblc1vn5ufL5vBV1pH0EupDE5CRPmaq8Xq8ZQRYWFoyMCwaD8vv9ppUGIwV7pRPkYqeY8nMzQfD/YyeFWOPfYSqAzESF4iQwEfwTpE02LSM7na1z2wVnfTweKxqNGtGG9I+4Rp47fk60sEioyN+tVqsaj8e6e/eu7t+/rw8++OBrD1Fx/8M//MM//L7/+I//+I9f44/yxV8+n0/f//73TSNJIAesM9gQQSGS7JAuLy/fCsgAfB+PxwqFQkZCHR8fazicbg9ttVoaj8cqFouWd8pYg0wEKQ/AfalUumW1m5+fNwkKS/pyuZxisZh6vZ7huZBpdKYIzNlT7wzgRTPJoWEUS6fTkmQkVCqVMukTYzMQAAWKzt3j8RhMgo0UjzzkGt0+l8R4PFar1dKHH374tQqsb25uTGOZSqV0c3OjV69e6d/+7d90cHCgFy9eGLHCtlSUDKzVQO7GmeC9bjQaJv3CkIGTDIUD+DjSupWVFbtwyBxAq0nnJskgCLSkkJHodzEkrK2tmbUVuR4jLhikJJNrUaBIMeNckE5FlxsMBnV9fa1er2fSQX53XIKcXcg4ulj+vF6va29vz/gHnhOekWw2a3pXyFhJxsQjFUOpQL7q4uKikb3AbAQNAS+BS6NuQRkjyZ5rrL+svPmyX7+vbH7nCiq4KSMyInWyQBlrRqORkTIwvXRWnU7H5C/OBWPxeNzWdXS7XWNwzlKiAAAgAElEQVQn+/2+Hjx4YEEWkCC/He5xc3Oj4+Njdbtd86kjT5GmY+mbb76pXC5nQm8eOB7Eer1ujOvc3JyazaYB9+g3V1ZWbAQGX+J7ITFy2mJh6CUZ4eTxeNTpdCx8A2wK3BXLJuQAFxQd1Hg81vLysoVof/rpp/Z3v+7XYDBQIpHQwsJ0zUqpVDJIo16v29pv5FCMhxgtJJmQny4RjBz7771792zHFpg87DiX9tXVNDeUbpV9TGCZyLKazaZ1hxAu+Py5xIBy5ubmDHd0u93GFbAuB8kRF4Pz5+HrUjSdiV7APTwvjNG47sB3ybIAaup2u9rf37di/NsvOAkY/sFgYDKoyWRiBRQdOMQacANJaMBtTscXWnG6YSATfgeIRbpiDBxfhTX195XN79TI73K5dPfuXcViMXPDEO47MzNdUYE1lHQawp9dLpfa7bZ6vZ55n6PRqNkJHzx4oHQ6bQwikWQIwjngMJcUqqWlJbXbbdMdguWyQ4qDl06nbfxHt7m0tKRYLGZdInhvtVo1Kci9e/eMfMFiCLmEWoGsUR6aSqWiVCqlUCikarWqo6MjRSIRW9tC4DR2UpfLZbAIcAYi+//6r/+yfUvODEsshb1eT++///43GlIxHo9NKfHbD89wONSrV690dnamn/zkJ3r8+LGRWp9++qkODw/V6XTMSkyoMpgzGN3KyooqlYpdmBQjDBXxeFyJRMLE/mig0ck69ZN0qm+++aaRQri2QqGQFcder2f4bzKZNCkUnTn4MRADEM/V1ZWF53DhX11dKZ1OW5i0U3nAJgrwX0m3soDD4bAWFhb07NkzlUqlPzhGn5ycaHd3V0+ePNFoNNLh4aFJuzA2IKOSZAsG+/2+0um0stms6WPJ2HW+507r9WQyMRILmAE44fHjx7p//75+9atffann7f/1+k51qPl83iRS7FmH7ZVkGOVwODQM1OVyGSPIjhq0cIxrEBqrq6smeQE/hYThpkeSxAfJ1+71epZWRZeBrhEZTCQSsc2f3NBcBARGwP5KMkwOKRR2TLpvVAJE7vHf2BWPddHn81mICXbQTqej1dVVSbJRfjQamZAdXNSZ1ypNk4jS6bTpGg8PD03G802+uAycATTOF7/HnTt3TLXBRgQsyUjmWIYXDAZvKQLq9bo9+ATKEOKB7peUJmIOJdkGBK/Xq9PTUx0eHqrVaimfz5v+ly2ixBBySft8PttNdnBwYGeL/VqsDieVTJK51Vqtlo3DXBbAV4zcfC0gi/n5eev26FTj8bjm5+f1wQcffCF8nNQubNF08lwIOPOcgdhMlUxJS0tLVhhnZmYsmwJYirwAyCmgO7IWnF8b7fCX+frOd6iLi4u2Ix5mkug9bmXcPYwPXq/XtJxYT4nSI20fITldA/8dHWq327XACZwjfA+Px2MPDSEpfMjz8/OWOwou61QUEFBC1ynJDgn2U/ZNMco4AXinywvoAgyK2LizszMTsKMZBBrgEEN60YW1Wi0jO3q9njHFTsjA6Zza3d39Vnioj46OzLP/+17ValX//M//rJ/+9KeWtJROp2+9P3RzFBmkboyqBO6gtQRz5O+vra3ZZReNRq1rplDOzMwoGo1qYWHByFO3261YLKZsNqter6ePP/5YbrfblCJg1gRxLy0tWW4rSWYkMkESgq3DkAP/cHFS4LGYkk9AIj+2U8ZySK0v+mq1WqrX67p7966du263a+E0zvQrSEIUDpPJxGR6FFUgCOeLhoTfjaAfNgEfHBwol8vp3r17t4KJvsrXd6ag4p4oFovGXCMz4VACvjOaMn4htnYm2MNUsldIktlNj4+PdXR0ZMWYsRopB1tFnU4nwHdcPJFIRJlM5hZeSiI5WZexWEzBYFC1Ws2E+BRdEtxZsscIjmuEjtvtdpsX3ummoruEcU2lUiqVSpbQ5Ay+gIAiJAPcikAS8gAuLy8VDAaVy+WUSCT0T//0Tybd+aZfdJB/6LWzs6NQKKQf/vCH2tjY0O7urtk8EdqDqZfLZdvBhfSMCxOW26mhRL7G+I4m2andBLuPRqMm1CcRjYLFFohIJGKKCyCqk5MTHR8fWwdLyAjfmzMGBg9hBLbqNAMQfs7vdHx8fMudheyQ34FL/ou8wIOdsBwF3aki4edjtRBqFc4aMAaJbuDHkIhMYZzZWCwmaTqROHMP3nvvPf385z//ShUo0nekoCJ8LxQKdmh6vZ4FKGNFw9csyQoeXSxecG47goiRTdGh7u/vW7HhweGDIiDa2RVIskLNOJZIJJTP5xWJRKwzRRtLlBlrlJPJpCVh8bs0m03D3pwBx3Nz0/31OLUovqQZkZx0enqqXC5n3RMPfj6fNxyZB5bRna7BuSDPuQKE/VAEPF9eXn5ndqU7Xxgj0NB6vV5tbm7q8vLSilQoFNL29rZevXplMBLnDgggm81akhUyJEZa9I9Ohr9arSqTySgcDpvUj0uPf8+IjRzQmfeKmB9lh1OneX5+rnq9bp8/mmfs1nx2KB7AYUm4cvrm6/X6rYxSNpum02k9evRI77///hdyvrF/CqNJq9XSZDLdKYV++ObmRslkUsfHxwaZ0UXCfUgyEozJkkkUUsrj8RiEwO9NmMt4PNbFxYVCoZCSyaQlhH1Vr299QY3H4/rBD35gQmdGTgTnuIdwJ7lcLiUSCWMCpWnoLtihJJOc+P1++8AZGQgrpvhy4PCZQzDEYjED12GFDw4OtLy8rK2tLWWzWbPrHRwcaGNjw3zhSJ8k2cgDQ4/es9VqaWFhQfl83rIBnJ0MBXQ0GtkOe6LR6GAXFxfV7XYtH5PimM/n5Xa7rZteW1tTu91Ws9m074GqAEE1ZoZkMimXy6Xd3d2vZYT6Ml8zMzPa3NxUKBRSuVy2SYJ1OOg9h8Ohjo6OlM/nVa/XDb/jYjo5OdGrV6+0srJiHU8kErHz1e/3VSgUTKKHhhPZHFMT7yvYK5cTXSvh3ATnAMFQpIF4WCGNuuPi4kKnp6cKBoOGKyLWJ/GKzFtW4zijHrkcksmk2Y0XFhZMN82zJ73WPXMZLy8vK5lM6vHjx4ZHY7W9vr7W7u6uPaN08URDEinI8+C08ZLFwHRFHaCL5TIgQU2SqW5QN7z55psWQPNVvb7VBdXj8ejdd9/V8vKy9vf3jZ2GYZZk4wnZpBxAt9ttpBK3ujN1nK6LkYBxH7yKcAZ0b2gBKWLD4fBW6tPZ2ZkSiYQCgYAtoSOSjDEfDBYXFg8Vf84Dht+60WjYHneng4bxjfBkkq44TNhdcePgDAMHpGvm6zp1ghTmWq0mr9drX4+EIyy2aCi/Ky90nffv3zc2HTycXNw7d+7o9PTUMgi4kDhDyOOIReTzm5+fN70w3VG5XDbbbzgc1sOHD3VwcKCDgwMrWGh5marQtaI06HQ6ttomFAqZVK7X6xnERDwlHvjxeGwQEEWXOEQ2ESALZIyHoCKhDEKVAovFU5pCb48fPza5Gc4mICLsvnT2kkzuNxgMjMjj71IMgVB8Pp81Ss71516v15oDCipBLBDB5EZgMXemrZXLZbO6FgqFr2yy+tYW1JmZGa2trenk5ESNRkMzMzMGaLOGBNyQRBqEzu1229hANJcUIL/fr1wuZ2M2h5eRPB6P299PJpP20IDVoN9j+yR21aWlJbMPcmtymNfW1mwsQx/KIaKgcWjpssE9R6ORCcd5WJDuIHTmQY7FYqpUKqZ0gARx7jIiQxWii0R4GFQ6I9LTi8WiyuWyFhcXtby8rM3NTROAf1debrdbjx8/1p//+Z8rGAzeUm40m01jkldWVrSxsWFJWzjFCoWCYdV8rh6Px2L3wCUbjYYqlYo2NjaUTqcNKz07O9OzZ8/sEuPiRL6HGoWLnpjIbrerdDotj8djRpFut2sSIfS2rC13WpshJ6Wp0ePm5sYkXhRgZHa47ej+mJRCoZAymcwtxxRNB2RtKpUynTQZEsALy8vLhtMyMZEXzBJECngmkzHYrdPp6PDwUJJsKvR6vcrlcpZaJsnspU6WnwQw5FPoprn8NzY2NB6Pv7LR/1tbUKPRqN577z1Vq1XV63XDOilmdHJ0pMgtWEgnyXSdHFRGfID1YDCo1dVVDYdDLS4u2qhAwUAPxx6myWSiWq1mYmonjuXMOmXMAdMJBoO3tKtOJxfOKGcyECYFOkUWlMHEcotDtFFUCbsgik56vXCtXq/L7/fbmhRyNUn7x+ZIBxONRlWr1WyVDAyzswB8V15LS0t68uSJBUSzYQHYBGcOxSmRSKjdbmt1dVUHBwcaDAbKZrOKRqO2Ofb6+toiF+myqtWqTU5MFDzwrF8hZxTrKmQfXw8JHNh9u922SQptKK43SCM4Av5/VpyDMSYSCVtbQ6fI+ej3++a7d9pEB4OBqTs4e+iY0VZz6QApsWjP4/EYTIAcENiEMZ9NC/wci4uLFrbC30PuhWYbuzdxkTihUGyAOQMVsNuMr8Ozt729bRtUv+zXt7Kgzs3N6e7du9ZNXV1NVzCwt+j09NTAcroNHCxoBMGSbm5uzCaKw4kODYeJpFvrS9DMQUyR7MQDASZF+pDf71cikbDula6PoGeKLKM4ukXn2g3GT3AyHkK6Hw4nKf/Ly8tqt9saDAZKJpOGrTrVBHQoXDQUZmQqSHMIwsBOyYqJYrFoXwPLJsoAuqKv2yv9x7yGw6FevHihp0+fGmwTi8V07949W3Xz4MEDuVwu1Wo1c+vQlfLeSbLxWZJJ8RqNhoLBoJFbTExut9suKGcXyDlxYqgej8eSwFgDTTFJpVKSXts0yQ+NRCJm/yUtjM25+PiJGiRDwu126/Dw0EZhEtWcmC1kGe49xmM2YPC5E3ZOkZZkGyOca3mISYR4wp8P2RsIBEwjDWb/28Hdk8lEqVTKPhsmQ7ITCAu6vLw0qACZFQYFoJjNzU1ls1nt7+9/6WftW1lQE4mEJOnFixfWRbHrBjtbq9WylRY+n0/Hx8cqFAomgsd2SvHh9iIsF3kQustgMGixev1+30Y61oDw8ID1YPvDAeKMtQP34YFpNBoaDAbWIc/Pz9sqZnSGxWJRmUzGNK6M/2SkXl1dWbfDwV9cXFSlUrGOErsd7xlECPiwJDWbTXuwEK/XajXbGsA+9n6/b0YFp1C81WppeXlZ29vbajabevXq1be+qA6HQ73//vs29qVSKaVSKQ2HQ8szYKphND0+PrauyOVyaW9vzy7GRqNhhhDOCTkRBKicnp6aDZXPk6JHuDmB1IFAQNvb24bJs4wQy6rb7VYikVChUFCj0dBwOFQsFrMiPzs7a7AYVmtkcJwRihWdpzO0GTwUHbUky7pl95pTKQN0gRGG9yOdTlsAOX+31+vp6OjIpkKmuouLC5OAOZ9HxnMmByeZSgoa3AQXkTNQxrnFlReKAJqoi4sLZbNZ1Wq1L52g+tYVVJbhNZtNI1XAAgn5DYfDqtVqJuIlhSccDttaCnzQkUjEUtVJpcrn87aPnI6AJWh0DBxs1AKDwUDNZtO+F24asE26ViAIiivWP/BeHkpW7jJqer1eG8lwbhFnNhqNbB8UayVcLpdp7lqtlq3sjUQi5s9HUI2mEMgAF8rFxYXJZLjVcd4QMEGCVr/ftzxUabrD6Ec/+pFardbvdCZ9215OAo3PkEKH2WE8HtvOMbSo9XrdcnIhDwkyQQPNeN7pdKzAMupubGwY4+7UO+OIYy0NmyI4WwSNOM8D5wnCxe/3W1fGxcC5Q6WCPx+iE9IM3B09N58vWlBs2E73lTMXlxAi2H6ceaws5zloNBpGCKHpZhpExM/PMplM17FHIhGtrq7q1atX9jswDTI9bm9vm5CfzAkKKoHrdKvgv5hgOp2O4vG41tfX9ezZsy/1nH2rCurCwoIePHhg1ke8u4zGlUrFOk0E50iK+JA5ONxwXq/XsFCi6rAYUpAQF6+srFjXSAxfPB63MQjRM7ioy+UyYovDzQeI/ZCvJ71WJAyHQ2WzWdup3ul0DCNmwR0wRjAYVK/XU6fTsSwApDuI/sH0gsGg4VooD3gfEPsj+MYxA87GaMXXcHrC6eDQKwJrYOv9NhRUtJVf5OXEILFHDgYDLS8vm5+90+noo48+0mg0UjKZNByu1WpZd08hQMoE/HT37l25XC7LbKD7RdhOUbx79659T2RDnFPUBfl8Xl6v15xgrCihMPT7/VsCeHTOSP/QFaOFPj09VaVSMc0z55citLCwcGuRIxcEUjwKH104UBhmmsvLS8OmsZjev39f19fXBsvxXlJQwYh5hlEILCwsqN1uG1k2Go2UzWaVyWSsWJdKJcOteY5J28LKymWFRBGjRjqdVqFQ+FLlf9+qgnrnzh1tbGzo4ODABLmMrmBZtPgccrBCmD6cFIuLiwqFQhZQMTc3p2w2a2w+Hevc3JxyuZy5T+gkYNsZu0kgJy4MicbCwoLd7DwcWAoDgYDdxCSeM1qPx2M9ffrU8MxOp2NbQrkowJeQcEmyYOyrqysdHh4qHo/bvqjV1VX1ej1Vq1VdXV1ZR7u2tmYxaIQl071DLjhDPmZmZlSpVHR2dqbl5WWFw2Ftbm5aUhNFeGlpSfF43AIzIDYobr+rwDnJLOd/dyYFSTLYhhGNkS+RSNi54MGUZOuYkRX9ru/LyIf4HmiJAgmuXCqVtLOzo2azqXA4rGKxaJcLWDUiemRm5+fnRmgxJeDugfgEUz0/P1ehUNDBwYHy+by2t7dtSWSn0zGzRS6Xk9vtNi0stloi/ZxLFOl0UWhEIhELIqHQP3361PBXJEpknNIcUNDb7fat5ZBwBahlnMEqfBYUv2QyaVOOM5+VMCC6W3S8ePf5rLFDc2HTQAyHQ1tLBIG3tbWlarWqYrFo8BwYLmoayCjWtCPfurq60ubmpj777LMvDbb61hTUaDSqN954w5LneXCcIL9z9S1sKV0XwneYyUQioVQqZclBEFa5XM7G/L29PWUyGdPG4djA0+yEBMbjsX1NsCawGXSIEAXOsYPiivGApXqETLMXqd/vq1gsKhwOW0I7jhBGdPSTOLPi8fgthxYGBlhXOuf19XXt7OzYeMvtT0eCnIoHii6bG357e9vcURgH0DkmEgm98847lngfCoUUi8VUq9VUKpWMDOGgZzIZc6Lx+UqyRDBiFskLgLE9Pj62oGPOAl0zDzxMOfudYJ0RqeMXl3Tr76+srOjOnTs6ODhQt9s1zBD3DaYG6bXKhI5wOBwaK+/3++3c8OA6L1H2PgG1PH/+3C5Luk4cQhA1OORwxDmF6xCs6FO5hBOJhNLptMLhsDKZjKkzarWaXr16ZRAHOQWcGcJW2N0VCoWswWBHFe8ZBYlwIGcuBTisM3gInBloDK6BzxelCQUQIhnYj8ns6urKtNiJRMJ0xEgUpSm8Ew6H7fMnFpDf7+rqyow1+XxejUbjC9mWv8jrW1FQ5+bmdO/ePV1dXalcLqtUKhm2I8lwFFZASLLCRLfGaEJqDlsVq9Wqjbl0pCwtI++UtCUCG2DDcTURStJoNMy+RxgFfnkOGyA6t7zTcECeJAWG9RAkY2EA4GFHeO/z+YzpRcAvyUgu/hw8l2KAYL/T6WhpaUlvvPGGdVt0zc7ujNQrSSb74lJBxrK8vKxqtWodejgcto0EdDVra2vKZDLa3t422Q8OGDA8sGM6FzoVJD0kM6HzZK0Ir0gkomg0qkqlotPTUyNa3G63tra2lMvlbKUJeLpzBTR4ME6iVqtlExBbZ9Hfrq+va2NjQ8Vi0WAAsFaIIOITwfq4/JAgcc4ZwweDgba3t7W1tWU6YzrgUChkEwPGDEKkkfiBPSJ1Yy12MpnUw4cPFY1GJckmNp6xyWSi999/3yyvxWLRdosxDi8sLGhra+sWLkuANFpRr9d7a8klFx4mEBoKSCLyaTHmOKMWYf15nini0WhUxWLxFkHc7XZNa8o4v7m5qclkYoHXXIIkufFesWgTQvnmZrqT7cGDB2o2m19Kl/qtKKjZbFZ+v197e3vWjZKvCRAPocOfQQRRmGAqsb5RSCga4/F0j1K5XDaXzObmpoHmRPo5F/txuNBuxmIxs3cWi0WtrKzYJlEi9yCgyL3kazIWgmVCZLC9c35+Xm+88YYKhYKGw6Ed9Gg0arcn0ILf71ej0VC/31ckErHij2YUIiwQCBh+GgqFjBRBNM3DiGV2ZmbG9IF0E4igcUmhXmCBHATLycmJyXjohIkvlGQymEgkYjIeRk8K7mAwUCqV0sXFhQ4PDxUIBLS0tKT5+XnDdRnjyJ9F8oPcjfR77IyMp5CaTrfQaDQycwQMOxmo7XbbCmw+n9e9e/fsPOA8czLkBCjzPbLZrE1OnF9WT19fX+uNN94wGAVdJu+fJCOHmBYgDIE/MGOEw+Fb0Y5MbMiC2NLAxfLmm28qGAzqZz/7mXq9non0kXHxfjFuQ/wAl0iy8ZxsVXBXMjVQHwCLoHXlQuOMEo8IfBEKhZTL5ez7k0NMyDvqAiA8JkG3220mIEn2+XBBE9uJ4YEiTwOVz+e1trb2pciovvGCGgqF9Oabb6rf75tQfWlpSScnJwaeo1fjDWN8YuTHUoZTCh0qIDi4K2ty+V748yFkpGmRaTabWllZUSQSMYwWHWu73bbVvuCEhFETPMEhww7I38G3TCcL28xBxYvMigoeLqylSEN+OwHo+PjYbHa4XIAxJNmtX6lUbKMm75/TjkvWJxcXuZxkpLKEjmAYQjfQAkLY8RA7jRU8JMfHxwah+P1+bW5uWkDwaDSyQum0D0pTFpnPk1E7Go2apMvj8dha7nq9biHfwCHxeNymGgwUROpdXV1ZjgF4Hbmv6XRamUzGcDqi+PjdITukqeusXC4rEAhod3dXk8nEskyBnig6ECe4iCjQc3Nzt7Bu3k/gluPjY7VaLcMWh8OhidtTqZQePXokj8ejSqViNlAgIi4Mn8+nx48f65e//OUtzDQSiWhtbc2wUz674XBozwPP1tzcnBV1tMv8XZ5RvncqlbIgH84Aq8zB3iHPuJRYq0PSGhMc0w08CIXT4/FYmEqv17vVgJHihn0cWInOuNVqWWP0v7VTf6MFlbAKhPCwoF6v124vRlyv12v7l0gnZ2zjjWYs5qHEAEBBA1sE/5Ombz6ramHGc7mcYbVo6WZmZkwhsLAwXZdCVBsjGbrBRCJxC6vp9/tKpVK2f8pJaA2HQ4XDYUmyQoytFVad7hdCY2Fh4ZbeFN85HXEul7OHhAuFogpWKslkWqxCrlarFoyMYNyJQfE7BgIBbW1tmS2YDptxHYlYtVqV3+83GyY/eyQSsQJDN4ufnvEMiIaA7n6/r9PTUyucjJ6M1D6fTycnJ6pWq4rH49bZ8FCxrTYcDhsOG4/HjcQh5IZpA+daPB7X6emprZOGjQcbpAHg5w2HwyoUCtrf31cymbQLtNvt2qZYziGwCmy303GFw+fmZro0sVQq3VrGx/YIScYbbG1t2QU7Ho+VTCZ1dHSk58+fKxQKKZ/Pm0vr0aNHqtfrFg7uzMYF+wTmYdRnHYskw1/T6bSdjfF4bFGDsPy815BKc3NzpoWdmZmxi4mfAXwaSArYAZgDeEKaaqrL5bKtJuLShZREGsekRRodbrDRaKSjoyO1221lMhkFAgG12+3/VU37RhP719bWtLW1ZWG54Jdo2xYXF+X3+02ig2gYfztMZiwWsyR8SRY0cnp6qlgsZp5sDiDgNnF8gUDAskBrtZqRRtyOYF50CvjesTI6Y8RIuCH53JnMz2gDVjwzM2MdIQwnhxJMEe0rB5SuBtaY3EluW0m3UnucjinS/JFkgZHycE4mE7XbbZNkAamAdTphDZhTLjwWzGEB9Pv9llaFA20wGBgxAGuM5IgHFhste+YhKdB7OsdQxkDMEktLS4pEIsZsk9gOKcPv6vP5DEN05t0yNSB1crlc2tjYsCkCmKLX61nuLF3ZaDQyQhJbI1AMgv9QKKROp6N/+Zd/seg6aWoPpkMl7NmpLS2Xyzo4ODCDBoWOeEDgHK/Xa+TZaDTSb37zG/3sZz9TqVQyNYIk6z5nZ2d1eHhozQUOLggpCGIcfDjluHTYvFutVi3s2hnqDoQEhjwYDCykiBQriDsUNU5zAlLFQCBgEyiqGoJnkBayHRblABIungc6WucmVX6uxcVFK8pscf1Dr99XNr+xDnVhYUH37t2zVCiIJW5QmGCkSDc3N7ZCmQ90cXHRpEaxWMxwsmq1auwr3Sy4KwA9qUncaNysECSsTVheXjZ89fLy0giOarUqSdahIo8KBoNqNBqGHdIVUcwePHig6+trw0ORhxFaAZiOy+Xo6Mj27wD6g/1gZ6WbQajujDSkM7++vlatVjOHDAcViVipVLJoQrA4l8tllwhGB6/Xa/gm7+39+/f1ySef2EXIz0rSFgJyhOAQcVwEjIxsKAgGg4aLQjAgkcNNg86Sh3c0mi6/g/Rh68JvryImOIOOjMuBr0EXlUwmFQqFlE6nTWqEpRPYBHnZ+fm5crmc/H6/YfXD4VD7+/uqVCo2LlcqFY3HY8XjcXtvnEYAyFiwVEJwdnZ2NDs7XXfN7ym91t4WCgWFQiHDuIPBoA4ODhSJRJRMJlUoFFSr1fTy5Uvl83n99V//taRpNCb5upLsd4QABPuXZG4lLt/Z2Vm7lJDs0WGjkeUiJt4SuM0ZBIMzEMKSbQMUZ4hQSSY95N9dXV0ZrgoWDm7rzBfgvQJzZS06l+tkMlGj0bDpBfjvj3l9YwWVYGVGuYuLC0UiEcMhwTh4GCXZlk9J1jHhj8e5Aht/eXmpra0tBYNBG63AItmxjgd+NBrZJtQ7d+7I5XKpWCxKej0K0ZXyoDLmUYApCBcXFwqHw8Zorq6uKhKJ6MWLFzbmYSfFw+90dCwtLVnmANs8U6mU5ufnVS6XjdFm/AFzoyBjCXQGP3D5MNIg/GeRYKfTUalUsmJAxwKkMjs7q2w2q/z/DQZm/AyHwzo4ONDOzo5NEfKNRLgAACAASURBVGCzfL502GCWyJIIeKEzkGS5oMiIwMfwuGObRRwPPASR6OwUebAhq0KhkEmE6HClabeGxIZudW5uzgKeceBJrw0BWKApbJFIxOAAvP2cG/JVJ5OJMpmMJNmKaQqE3+9XpVJRPp+/FefXbDZVrVZt+oHAhASiyHExLS0tGW6OrOjv/u7vVK1WdXBwoJcvXxqPAJ77+PFj26QqyfSlTrcVlzZbMJgipCkBhNIDC6mzK8XGzPeELKRgwuzTtNAccBZxqNF0EeLNxQzvwmVLEDdNA1MhzRPcB2YIJgM+o1wuZw3dH/P6RgoqWjsExU5mmg6Qm4vCxGFndGfUQ2LBm1qpVOzGArthNQpJ/YzBKAAA0ElqgkGEDXTebL1ez5hFJ4kBjntzc6NoNGpL8SRZCEm321WxWDQslT05POxgPzc3NwZbAD/QbYIDYWZAi8hh5v3s9XoGVfBAovPEA93v920LKx044Rrc3Jubm8Z8VyoVzc3NaX9/3z4DMGHSlZzOKmAaPkuwbIJgGO+azaaSyaTu379vNlouGKLmvF6vMpmMYbx06ojHkagBn7hcLjsL19fXFg8HOUY4M+ROqVSynAZJyuVySqVS9v5KMhlPtVqVy+WyaYlJiIsApxVGkbfeekvhcFj1el2j0ch0mZxL8EVcUExI/J1+v2+p/WRHYHaBV4AY5N+BRUPo+nw+vfXWW/ZZQPaRuNbpdAzGYuRlVIa0xEiB9pQmgxEbzTXPJmQrv48kM7Xw53To4Jp8D843HTk1gL8nyaRqOAb5mkwmaG2B01A1cOlDZgI7ODON/1j339deUGdnZ7W+vm5tNu05gmnCHwDjkczwYa+srNihIngWQTEfNLIOZBKM0hRddI7cuOFw2HAc1vNyIOl++PnoasGXotGoAoGAdnZ2LNsUXSlWwqOjI7lcLi0uLqpWq9mHT34jMWbgfefn59rf3zc4ARcQDy8kFw+xx+O5leCPNIiUrGazaSM41sSTkxO1223V63V7uLA8cvnAqq+trVn+rNfrNXswYR/pdNoStHjQ6vW6jWrn5+eWq0nngtTHGQkHdsiFiKD75ORER0dHWltbs2wFXHHIlHK5nMEXEFbOYozelgc1m82abZlsT1jnyWSiZDJpHbIzsBsMH1gCqdHl5aV2d3eVTCaVTCaVyWT04x//2GRmz58/t848GAzqk08+uZX4xKTCWM8GVNxShD2vr68bvAPuH4vFTN6H5PCzzz5TIBCw4kuxBt5CPA/UIckgDVLeeN+48Lnozs7ObGrgrMDi83Xm5qZ7pCB3+TOkbEBPPAvSa8khTRNKCGC36+tr64T9fr9xDqz1iUajNrWyjZaLnoWD/G/ygnm2ICLD4bDu3bunX/3qV39Ul/q1F9T19XXbIyPJPNBgV9j6uP3o1GjL8bOzk5vbEwsl+BjEEwzqeDw2dh3LJQ8jHZgk+7DAWegSGSMZdbjFq9WqPB6PkTG4YEajkXXU+MYRK7tcLjUajVsjC3KWubk5K7oU6HA4bPvZJZmawJkViS7wtxP/4/G4EW1XV1cGC9CB9/t9lctl6yooDmDKSHSi0aiur6+1vLysu3fvqtVqGetL+AbjKw8EhS0Wi9nlwMhNR83P8/LlS7MP89DC4CPvYmTn88T9gjgfTA/MWJL9PouLi0acEIHoXL0NgeT1evX8+XPTjHImUIuQ2Qnmx4PJuQoEAgbJINrHLx6LxexzIt8BpxSGlV6vpw8//FC//vWv9d577ymVSsnn82l9fd289ciXGI05/4VCwXDQzz//XBcXF/rbv/1bfe973zPSF4kSY/fLly9tkgJGW19fl8/ns86t0+mo0+nY+4qFFvUN7zGXERAOZwDJE40N0w2fLc0TdYHpBViARYiQm0jFsKny3+BbuFQfPHigSqWiw8NDO5+oKWheKOAQru1227bV/jELKL/Wgurz+fTDH/5Q/X5fjUbDDgO4DR+2JMNK0MJJMvaRzpMPiQ+tWq0qGo1a0UPzBmANLsf4GwqFjIWHuV1dXbWOFVYQ2xzfhxGaAoT0CfKGODPYYTzVHFyKADgx4Q3tdtukJoyDEDWA+HTNuKn4nUiNYhNmIpHQ+vq6jdkEBHOjLy4uKhaLWXGZmZmx5X8wsRA2mAyy2axdADhkgsGgOXiANHK5nB49emQyHIwXJycn5nBpNBqGR3IGnIldkIgUPbpPJF2BQMCKMX8XeKFQKGhubs6w7FqtZl0cFxXFG2KLCWRpacmi3YrFolZXV5XJZNRsNs0BhWyJcBRMFPxuQAf7+/va2Niwi4RGgYe/0+kY9stlVSqV9OGHH6rdbus///M/9fd///e6e/eulpaWTKRO4UCPSafv3EkGr8BnNxgMbOrCqLK7u6sPPvjAnk9WilAY6WIpPIRfc/641CDQUqmUMpmM+v2+aYUhj0ibotNGf8u6FrpW5HXwHmimJdk0S9F1vq8nJydqtVoG40Bc8e/RqzMBMpVxIaAmYfnn2tqahWL//7y+1oKaTCY1GAxsbTL2TUY33ERY8HhA6BwYXdA29no9k3NQJGnpOUCMT9y28XhclUrFJFZo5hjhbm5u7BAiUqYLkWRsJiQJFkkngQOrzEOA9g5WOZfL3QLlYX1fvnxp+8zR26XTaXtvOLwUkXK5bNgqmzRJ4XK73beWyJEkhRIACVav17MxFslINBrVYDAwHa80NR2k02nr6Jz2Vn7vUCikbrer0Wik3d1dg2ukacZto9HQycmJRTSCu/Fv/X6/xb8lk0m7VAnbnkymIcNOUThdu9/vVyqVsiVw8/PzNpLifII8okuqVqumC+X7oMpgQsHEwAob/judNoWZHARgBJ/Pp93dXQvyoQBXq1WFQqFb2aN0yMFgUM1mU/fv39fTp08NIsEyzWdzfHxs5BUWU3BRdMbLy8um7YWgK5VKFn5zeXmply9fmlQqnU5rY2PDiieFBMKLYBFnXB4qh3A4bKqFarV6S4XQ6XRsapFkOnIuMaYPJgfeE4jV8/NzdTodw17RyToVF5B3Z2dnymQyRqaRUhWJRExNVK1WrV5Ir2MdcTci3UylUvrss89ukbtf5PW1FVRCG9Cc0n4jVaELWV5ettEGjzi3WjqdVjQaNYKIA4ztjgATwipgbp2yDIoBEiuwSZhjHEF4qufm5lQqlYyxl6ajDJ0VXmG6Nro7ScbykntJ4IMTvsD7zDgL/kPKFQEPrVZLu7u7NjZyw/MzSlMSAecXEjP2zV9fXxurHovFDHcikevVq1f2oKGPhd2lCNfrdT148EChUMggC2fnizFDkrnNIAZOT0/VbDaVSqVsLM1kMqpUKrq4uDDL6/n5uW3EbLValnDFZAAW7vT6LywsGBvMum+6XHS3PIBYU4vFojlqPB6PMpmMFhcX7XcHM/V4PHrx4oUkmTYXxrvdbhvE0mq1NBqNtLa2ZnK0bDZrMjDn10PuNplMrMMnci+bzWpjY0OPHj2yvV6dTscCTfDQt9tte8/p8CRZaLgzxIXOn58LvPbNN9+0Tt/v96vT6VhI+6NHj2ws5xzQIbM8EGUN8XzOi5bzk0qljPhDqjQ/P69YLGYZAc4MDKCs2dlZ9Xo9y2DlIqQQttttS5dCsYO8kKAkplHODfpgnnPOL5PGxcWFyRvX19f/qFT/r6Wgzs3N6b333rMdPYxeTk0kJA7AOrcRuFwkElE4HLYFZsVi0SRLFCmwPAoWhxapB7cqgupKpWI3KAA/TDXfnyCVdrutFy9eaGVlxRaTcdvSnTJWMB7xEDGCXF1dWa4pFjgw4kajYQoDnB1ORQGjMGnySFf29vaUy+UkvV51Eo1GLZIQDeXW1pZ1Ac50I7oZus12u61SqWQXHRIeNsuWSiXTRHq9Xh0cHOiXv/ylXC6XvS9ObG12drq2moPNao9ms6lAIGCkymg0XQ+SSqU0Ho+1u7sr6XUADFpUOiWwW3Z5AQfAkkMIoj+kEBNrGI1GlUgkDK8NhUK6uZlmex4eHhrpI8kcVyhNzs7OzGyB3Znf5+7du1aYHz58aJI86XWwtSQbu6VplsXs7Kxh42wJoHh5PB6tra2ZiaPb7erk5ETxePxWwhLkWaFQMKlQLBYz3HFzc9NIIkJ9tre31el0VKlUbFdUtVrV9fV0hUoul9P8/Lz29/dt0ythMjwjn3/+uTwej8EbFD8we6SJELl09jy/TgcTGDiEMZs3MOnwPoJ7MhGSAQCWS2FGI07hdGL7cCzooJ2OrcvLS62vr6tSqZhB4Iu8vvKC6nK5dP/+fUUiEZVKJbNL0pJzS0jTtptVIxA2rC9Bw4jUgdAUihDJN8iRms2mXK5pqjmdrCTTmCEtoUjxPQm0ODw81MnJiTGHjGQQNfjEkWdgMWREgzRAcsPBgFElIBdN3MnJiaXhHB8fW/Yk7xdFmkJN50E3C6mBxhJsj7R4ulJ+fxQCjUZD3W5XjUZDi4uLttW02WzaZ9HpdOx9IDaRgkuXTKdPNwbJCHRCqhce/kQiYX58CDRSl0ajkR4/fixJ5siiiwG3KxaLdiGBE0sybXCr1bICCK4MbolECYkNIebOTg+VA8sNo9GobUiVZMWc/40k5/nz5woGg9rY2DCy6Pp6mlsaj8fVbDZtwhoMBtahAruwRBKskwtWmkJP9Xpd9Xpd/X5f0WhU/X7fICKwcCd+SPYBkBlFGSLo008/1fPnz+1ZkWQ4NbLE8XisarVqnT/kKjpy3IZcXpx9zinmDCRbaKcp6phE0EAzZjP2U/QgePmcJRne6/F4THcMJwNhReODqoHJhPcLZYoke66azaay2axJs77o6ysvqEtLSzbqVyoVuyWQZ4BJEYWHiJ6xHCEwzgwnQ4o8CMUADDAdAjghoQsA4LVazXIm8RRzWyFDYiThzURwDMvrjCpDisUH6PV6bTsqRRAbISMvBwy2lKAJsLVqtWohMPjHkW9xEM7Pz5XP51Wv19VoNEx+Mx6PVS6X1Wq1TCZVrVbNRvrs2TN7zySpUChYsDHWvmw2q+PjYxvhW63WrVUWjE03Nzfa3Ny0yDYMFM40IT4bQoeRavFePn/+XPl8XpIsa/Ozzz4ztpfxeTQaqVwuKxwOW4GH5EFRQecN7s4YKMm6JmAIus719XUzcoDVoj/l4e52u9rb27PoOFbovHz50tbTXF1d6aOPPtKdO3c0OztrIz45EJCepDdhrCiVSjo6OjK4izMoyf4ellSeH+SCFFWnxx2rLrg9ozGZCUxyo9Ho1lJIPhdUMCgSyKVAYoXLrlwu2/NBDizNEsQZFs/5+XmbzBDqk4mBdhktLOoFPk9+p2q1aqQiSzTJvSCPttlsqlKpKBqN2nMfCASMuT84OLDJhWaFIor0Dwx5PB4rn8//f/n7v9KC6nK5LJChXq8bi80v5wzwcKbTSLKuIx6PGxYFXjY/P2+JSs4UcW4ZmHI+TMgaxMFkejpdSUhB6Lr8fr+NJCzC4/DALA4GAxvJGGnQhNLxIC3iA8PpcnNzo62tLXm9Xr169epWgDQJRoxX0uvMSJxOyWTSLhkkV4i8S6WSBYbE43FdX1/r4ODAFshRlIEWkGZxw5MhG4vFbnWOdMTAKJFIxHIqIXWIEMTOilWRbhGdKcQQu4voFqXpRXl6empL4tDUons9PDw0FxxSOd4/py0RaIQgbwozk0Or1bLEevC6tbU1+z0YD7lw+fkhOjAvEOYDcYStmC4SiAIm3+Vy2XR0c3Ojo6Mjeb1e3blzR+fn57akj0wDLtLl5WV9+umnKhaL+vGPf2wKhNXVVcuawNsPR8HERefMz3hxcaFXr14ZGYNChJARxnuKuLOjg+hFd4qDjJ+Z4u3EcGk2SIFiLGciOTs70+XlpeGlFHOmE6RozkmTZ5XPhueFQB2iHv1+/y1imfQ5zglFFa4AfSu8CQHwX+T1lRbUYDCofD5vi8S4mRjb6BDRujFSxGIxkxZFo1EjgHizGSkRGnM4YYXL5bLJlmCOcdQg1uUmfvDggXWmMNCIlvFDe71e854jBGek5uACqEMe3Ny83u7oDDZB1cCtjLWP7pmLhQedA0XwA5bD4+NjW13CyMiIjrkB5YMkY2DRgBJBR0ISFx5s+dnZmarVqkajkaUZ4cIB02KcRLKFjIxuBgUCODk5Db+tp2XspTMAHvF6vXZhOrMLwETR66K/xToIrk0xHQ6HlnrUbreVSCTs/EEuxWIxjcdjRaNRzc7O6sWLF4Zdg42vrq4arDEcDo1cffnypeLxuFZWVmy199ramvL5vBVg3j8+T84tuCGyOogTdM/OHIRaraZf/OIXdmlw8XBh1et1e58hPUulktk+IXXa7bYODw8toIcJK5vNSpqmkNEBcqHCd/CzQOQybcHGAy9guqGRYGMFcAVjfLfbtTqAJhd+gcsFDTlwAPgq5DAMPklkp6enJutClcKkFw6H7SJgAuLZpDZQmK+vpxsQNjc39fTp0y9U876ygup2u/XOO+8ok8mY1hFAnZHAGebsdrttnAQbw8nAh0mxogNDo8iND2xAPijJPKwTQZ6FaHsymeizzz5TMBi0pBuA9vPzc8Ns0+m0MYVOe16n07GQDg5LMBi0bY3IjNDTspuKKDiS9MFvnT51ujwOEYv7nDgilw8pPnR9BEsvLy/bGo2TkxMjWTissMQ8nLhF6I6RtiHuz2Qy9j7d3NzY+8Atz8g6Pz9vJB/5sB6Pxx4qCIZ0Oq1Op2P2Sudomkwm1Wg0VCwW7YEGIuFzZVQlsZ78A3IZ6Hh5UHw+n4rFop49e2YkjhPjm5+ft6JA18Z5oMPlveXSwM3Dewomx8NMDgNWzIuLC5XLZbvcb26mCWe1Wk3r6+uWVrW/v69oNKpQKKRer6dCoWCrx4fDoT799FPD45GQhcNhy4dlZOUz9Pv9tkl1d3fXXHRgkVxy9Xpdw+HQYDXcYOQHoHGVpqaEfD6vXq+ns7MzvXjxQslkUqlUys6u9Jrk2d/f1+HhoZG9GxsbRobS/cbjcY1G09QvPke6RrIAKKZATM7u0TkNoVvFjccZQ2XCtEuCFn8OlkutQKmDjPL/9frKCiqdRblcvkWEIK6FkeXPYfNXV1eNwctkMnYwERpLr1NxpOliv3A4bPo5uoa5uTmLd6NgttttY9M3NzdvCZQp4hQV5CDIaxBFM64gtyFxHQE7hZTDHIlE1Gg07Hc4Pz+3NCG+JjIPxvpKpWLMOw9NIBAw3S03N0YDp6WObIHLy0tbaseeIGeaEjikJNVqNQUCAV1eXuro6EitVsvkRThfFhYWlMvllM/nLUwDfNbtdtsKiYWFBR0dHRlJ5na71Wq1lM1mFY/HlcvlzAHVbrftkPOQUSixtJ6dnenw8NAE8+CSKAe4gFwul2q1mm0gyGazymaz6na78vv99pmS7M50AflE5zMcDlUul01zen5+bvI9HjbcbKTlO4OJcZTt7e2Zm8wpv2HdDAWnWCzq5z//uUnFFhYWVCqVJL0mVSqVipk70um0Wq2WDg8PNTc3p3w+r5ubG+3u7mowGOjevXu3MMFEImFnEVjg6urK3hdpWvBqtZo5yrLZrG2xICC71+upVqtZcxQOh22J5OLiojmuxuOxjo+PLfeB5+Xy8tLMPJBGZF9wCXIRU7zq9bopMObm5uzvQ4CCGQNVSdPL5eTkxBK+kE7i92cLBrIszgGwHZMxU4kk+1y+sYLKqoVer6dyuWwd5sXFhYHQ4Dh0R5PJxEgQ9KiMADhf0HfyILMuha43k8nozp07djvTtTECsEsokUiY1Ar2HjIAWx/MOUv9EEeD1zEeYl0lZQoHBx/E6uqqfD6fJTo1Gg0rlHSBqBOQGUmybhPHCB0VcAT4LiMXf4/cUDpc9LqoGSgedOgI2MEFFxYWlMlkjLiju6GI8O/Ozs60t7d363IYDAZaXFy0QnN5eWm7n05OTlQqleT1Ttd683CvrKwomUwaOceZcU4Czo2zHs90p1etVrMuORKJWBJTr9dTLpdTOp22nWF0MGQnRCKRW3AKD8vh4aF17GCqXKbODhRSg+4VaRmaXiRF4HroTYl19Pl8t7peZ0zixcWFMpmMfD6fKpWKMpmMwuGwPvnkEz19+lS9Xk+BQECPHz/W+vq6SfaSyaRl60IOoYDBpXZwcKBnz57J4/Hozp07RtJhXHCmtzFF8pySL4rms9Pp3Hq/eI+BYpxh0hRB1Dyzs7MGQUlTNh8oiTBrtrgitYKAJNfD6RKkGXOqbHguqBPOrjqVStm0QzdPTgbFXpIRxH6/Xw8fPtR///d//8Ha95UUVJ/PZ4dmNBpZsPBw+HpXUigUsm2ekuwwMfYiV6AQc5tSRJGi8G8Bz8PhsBUIZCNLS0va29tTqVTS22+/bWEL8/PzqlQqWltbUzAY1NHRkeWBOrsOt9utcrmsfr9vYda4gGCr8fID7lMAeFCcYD0YH0XLyZxDrKEeQL6Ct53fC0Z2e3vb3CwEsSQSCROfM1JTqJeWltRoNCxzUpIymYz9/xQhJC4UVfAywqHxhcdisVtymcXFRW1vb9vPsru7q3Q6bcHhwA44XCaTier1upEeSGmYSCj0sPmIxunkU6mUfvrTn5qwn0uIlSXJZNJSwQhWgQEn5GQ0GqlQKFiEIRkJSO8YixHpHx0dmRyP6QrXEJ8hiWVOSGJ+ft6Io0KhYATWo0ePrEDs7e3pwYMHSqVS1jHNzU23MHz88cdWKCAIyWEg7Bn1ChmrjNuFQsGgCRoTzCFYvhHgj8djPXz40EZhzBWtVkvpdFqpVErFYlH1el2FQsHeB3B/Ji+28IbDYQvFoYHhwh2Px2YldxJX6K+Ba6SpbIzGoN/v6/j42J4d4B4aNP4+pgQumZmZGYuQBLpBCQEXIr3eEUZjFwwG9dlnn5nz7/e9vvSC6vV69b3vfU/Hx8dqNBo2EoGPtFotKzjgqBxWpB+MmLTikBvc/v1+394ksEkncD8zM2OWSqym+/v7Ft/HoeHQY+uEsZamWaC4rRgFcY5Uq1WLB5RkMIPH47HRGicPRJezUDhT+hG8s6oE/K1Wq5nmEw85Y30ymdSrV69UqVSs4+dA81AyDlK46MjPz89tSSEHEB8zRAFhFCcnJ3YpQCZtbW2Ziw29Hp8dqgM6VsiWhYUF28oJpOJc3nd4eGjjOz93v9836KBcLttnhwytUChofn6aGn9wcGAECz59LiOIL+AOyAs6V1KmnMvv+B7IboBkBoOBaTbB8OiS6FqduOHMzMwtl12lUtF//Md/mGYac8Rf/dVfGU5KZ472GdLx7bfflsvl0s7OjlmVKSYYOOi0S6XSrXF+NBqpVCpZsIoksx07Iw3R6+KNJ6ydCQPNsXPfFEVRkq20ZhJwaqMh0iD3mPiIz6MDZirh7wDTxWIxG8vhJICksBhDQjv16Zxbchv4Gpx7umWeLxopfg/yQPh9vvaCura2Zg4aVr465SQIrRk1eThjsZgJqNfX162zBQ+lUDIiIYMgFCUUCtkNSSoNncrz589vxczRVaLHGwwG1vpLMuE1B9Ttduutt94yhpTDA1YDzgtYjmIB/NZJDoC9QsxEIhHTi87NzZkPenl52QTYp6enxlKjfZWmeB2FAObd5/PpjTfesBGIMGh8z3x/OlY2GAAh5HI5C01hfOIAQyihfJibm7MwZayDdMvr6+sWvYc9F8wKvHtpaUn3799XrVZTvV63IGgS04lpJKMhk8moVCqpVCpZQbm5eR0VGI1GLX4QlvaTTz6xrw2Ovbm5afu5IOeQxlFM6F4hAhmBW62Whd1gMmE0zGQyajQaJvh3u9169OiR/H6/Xr58aYvydnZ29M477xjPMBgMzKn04MEDeb1e7e3tWTfHJbWwsGCWUPIpeC+k6fSBnrfX66nVauni4kKNRsN2fznlWphSgMgIbuYZpVsj/xVMkmChhw8f6uzszEhW1r1AeIVCIWtEGP+BBOLxuC3Go7nhIoEodC6TdOKc5GNwgROyxOdLoUYa9duh1PwZfA7qIWAytPA872dnZ8rn8xaaw6X5u15fakHFu0s3RhYjAl/GRD5IJCOnp6d2a9MxIk+hoNzc3NioQocAnjUej03czIoLvu7R0ZFGo5Hy+by2trYsMNlZLGB5GY3BciiG0lRKEgwGLfCEgo0UZTKZWJgtFwWQBrckukUKL8w63TQp64Q84zCSZAEZpFmR+M9oCnlEUIckK7JIocBOObCRSMTIC7p6tJ0ul0vpdNrwKWl60SBupwOfm5szJcDh4aGRe3t7e3K7p1spq9WqPTxIhShy7XbbjB2FQsHIA0gwphseXEmmuQUSANdlbC6VSvY97927ZxIvbKM7Ozu6urqybg1YBiiDnxMoBxwQCywECtsmOKu8RxQAMHS+xurqqq1G5nJ1u6frQpwOoBcvXmh2dlZ3795Vs9nU7Ox0YwJLFp35E4PBQH6/X5lMRolEQn6/3+R9WI2BBYgohOEm4yIUChlWC5Hb7Xa1v79v1lo6Q/69M2cAcpZYTZQJrKfGbEL+K1JCQlPQDKPfRfWCnfv6+lr1el3hcNhkVGDaOLyACrA9Q4SBWfM7wPxjq2ZKc67vcbvdpm6gKI9GI2WzWX300UfW0Pyu15daUJ88eaJYLGb4ECkyTnmO2+1WOp02HWmn07Fiy25vgHDkQWj0zs/P7cEA1xuNRqYdRTAO5thqtcyrT0gy4n7GHYomrT4Fn8LCLU1oBiyp2+02lhwTArcZHWcqlbKHAOEyY49zUSCHDLgAzM/n8ymRSFgIB6MrWQVE7bndbmNHIVQQSYPF8eAsLCyo0Wjo888/N60onSTvA5fi1dWV4b9ObJevLUn7+/tWlJFqUSycubYzM9Nwm4ODA1UqFSMeILJYpgjxBtGHCoBLCCJqMpmYXhKGnIeYC5OvwQgHiQdpl0qlrLNJp9Nmvx0Oh9rZ2VGj0TBiRJLJdpzvAZ0YI+RoNE2YX11d1RtvvKHLy0uViBNSZwAAIABJREFUSiUjOZHjQHbwAvoKhUI2Mt/c3Ojp06fa3Nw0pxzjKjGWuJqkKf798uVLO9NOQnZx8fXCy4uLCyUSCSsafE+eJYTuTmUNpCFwGd09WnFJdgFNJhM7s2Cd/DwsWvT5fAZ3IY2SZJcZ5xxJIrwJygngDjgJHJfOtfGoYBj/OcNMqTR4Ho/HEtqcDjUaN9QvOCgPDw9/bw380gpqKBSyneCvXr26xe4hO2AMRueG24fOjk6Rjg8LHb7zfr9vHxhCd0T+dKidTsce1MPDQ9Xrdd27d0/r6+u2doVxhp8LBxIaThh2ZECEl6AAgBxCPoJnmXED+xp77FOplBqNhgHj3KQUWogpvg8h1IxG+OMZO30+n46OjkwozgM5NzdnxB4FBdsuhQbLJITE5eWl4VTgcVtbW1pcXFSxWNTZ2Zm5y/b29jQ/P6+1tTX98pe/NDdZKBSy9w08zxmUQsfJAzQzM2Pd0Gg0st/3/v37urm50eHhoQUiM/Y6iYZyuax4PK6HDx9qb2/PpDzHx8cqFosql8uWUtRoNOz9WV1dVTAY1Pr6um0H5WeFqZ+ZmTHxudfrtchJyB26MXDwVCplZ51OPxaLKZvNam1tzWAE0paOj4/VbreVz+dVrVb161//WisrK9rc3DSS0uVymUMK0wFdGxbiy8tLHR4eamlpyfBBUqlQFrDCnOAgmg7nKh8yb7F5o5edm5vT+vq6ReNx1lE68HK5XGZY8fv9arfbarVaxugD0zmlS4zdPMt0gVxeyNS4CNBSYwBAiYATku+PE5BLAXkm/5vVO+QksO+LjpTOnWcRfTx4uMvlUi6X09HR0VdbUBkfcc7AwvIQANxDHNH1cBNmMhnrHpEyAUwTVYdMCX+7U3cpvRYPz83N6aOPPrJ/w8MLhgfwjdf68nK6RwnJ0/n5uR021kt3u107uIwMYDmM5hBCbrdb2WxWrVZLrVZLjUbjFobEQjNkKhRA5yGSZCQOjDvwCK4fQmXQi/K+4ZnnQKMuYCzvdDoKh8OG/TpZaWyyBwcHcrlcJrwmc5ZdSUdHR8pms2ZUIBd2bW1NrVZLvV7PSK3JZLqor9VqmSAfwuWtt96y9S8QJ3wu19fXBlWAU2JrZSMsq6jRwX7++ecW9gymjfQtl8vZ7qV0Om0/H+HLLDkEBoLxJoH/+vra1tKQ/3BxcWH2XgoNUwiFMRQKWbfFziKPx6OXL1/q1atX6na7VsiOj4+1u7ur/f19jcdj/c3f/I3u3r2rer1uTDSGEEJl6Pwgj+LxuC4vL03/DalIEpYkw16j0ajS6bSkKaRE0eKyaTQaOjg4sMLSbrdvjdNMnXSSGFp+O2gGyRRKlG63q1gsZqoQJ7kHNooHn2mHSEVnaDUTGR09jjxeTE504li0Ua9g8eW5oskCRqRB4nVxcWG8x1daUBcXF+3GJXzWKcVgROKNQuBPmlQqldL6+rr5zonzcuIlTukM21GdqeWzs7OGJTrj3FACUHQZHQGqucUZgcBjpWk3dHJyoqWlJStAbrfb7GyMfOPxNAWfmDVCKMAySbafm5suuGN8Z28Rtza4ID5oRhwOHJFyTosf+Zh7e3uqVqvmPOEWByaB3PB4PCZWh0xgNIZ8oOj7fD4dHh7eKrCJRMJ2pBPiwnsM85pOpw32WVpassT7vb09O5BcprFYzAgiiMZAICBJNjo2Gg3DpKUpWYVjBnVGu92+5di6vLzUxsaG1tfXrWMlK5YQDkiPYrF4y6vf7/cVj8etAMAJIBoPhUIWrzccDtVqtVSpVCTJUpdwVO3v71thYWTf29tTu902F1UoFNLy8rLW1tZULBYtwxVJE90YjcEnn3yi8Xisn/zkJ4aZgovX63XLNS2VSnZ+NzY2lM1m1Ww2rRlZXl421QbTFBkPyMXgBBiZMXJg/wwEAkqn04pEIiaPdKaVMUXRhIBx8jtxduBaUMSAcTLeezweI/ywH0Oy0jQsLS2ZegPMlgaIz5JGhgJL3rD0OjwHCGUymRhBCZxIJ/+VFtS7d+/K5/OZUFeSeWSdB5cOEqwN3GhxcdFcSm6326LCwM4oyMfHx5ZbiYaMD5gPuVqtmp4wFArZfhxGEx46vga6V4gLDjC3n8vluoUXgfGAoyLzwWFEcUIEjVuFYgUeyddkpOTQcOj44CGZDg4ODCsklwDQ3rn7ip8RjzvwBK4t7IccFOfGydnZWQvzwMf+5MkTWxxHsPD29rbhZs1mU81mUx6Px9ZUA7lEIhHrAM7OzrS+vm4ZBOz0cRZBLmFJ1pm8evVKjUbDzBixWMwwSx6Mfr+vSqWicrlsqgc2K3Ahx+NxXVxcWAwfDzJhMcvLy7cSnfgckE4RzMNFAwSDy4yCEAqFJMkkUfPz89rY2DACC4vp5uamUqmUpOlFT07svXv3zMdPJ0vXy9m6uLjQy5cvde/ePXNGQZRyhpF1kYeAy0ySWVT/D21vttzoeZ3hvhwAzhOIkSAIcOy51bJkx6qSU664KjlLVXKS68gd5GZykCvIQQ5cdiVyYk0tqQey2SQBEDMIcB4BkPsAft7+6drJLsvefWKXpO4G8f/f+tZ6pxUKhdRoNLwRFvaagjI1NaW5uTnHVSJhQ70CUcd2BIwjwDuQPeCVQUkeASuTk5P3Nh+AV1M/gjgtFmqyKMDnkbhdXFx4SzCRkpB9vAs8W4KtKaZkLXA2cUUhN5Q+KHs2Njb+11r4ZxdUbgoKkvRhKRv4RigUMqOG7k+SmVbAel4ibqahoSHl83kdHh66deeLCQYYcPjxWpO1SMcEs12tVo3jYWO7u7vz6tjNzU3LU5AWAV/gU+90Oh7b6P4QrSPFur3t72DHNirJmFs2m1UsFlOtVpMke/TBMlE14J+ng5yenvZOLXBgxvvNzU2/WGNjY14KyJgoSa9fv7agempqSo1GwxY/iD8kbJubmzo6OtLi4qLW1tb04sULvXr1ytgvMiVIQTZZYuuFvCAeLhwOu7hubW3dC0i+vLzUxsaGsSz2LCG1ev78uYko8FyyZMHOfvvb3+qbb77xIjmWG6JbxmrINAGmicaYgzk/P2+tqyTt7+/7YNXrdWsoK5WKXV4UFMbxRqNhuVUkElEmk9Hc3Jz29vb061//WgMDA/r444+tawXjzWazSqVSxnTp9Obn59VqtWyTJqaOPWmSTDZFo1E1Gg0dHBzcW30Czl6tVtVoNMwR8Cx5d9mGy7nudru2zoIjMxpDFtEg8F00Gg3rjyFfmRjA77lo+F/OKaM3kBDdMQUX2VMqlXLxJLAFQonu9Y/X0oyPj1snzBkNyjCxQlP0KejdbtddLJBVJpP5/6+gMraSLs6LhnNF6mOowVYfUgnpBF0MIDy2NVw6WPqQNAGQsxKX7hdyilSmeDzu6C7CkcHngBF4QcByCXnmFnz37p2xOm5T8DNJVgUgo+EmbrVa1m/e3t56/TR/Pi/t3t6elpaWTIyxTYBujZEymEbEODY2NubiDDZNp3d1deWiC/aJ0wTpSyKR0PX1td1lFxcXFsjncjl1u12nFaFu4HnRcdGpklYFvkl4NSw4ncH09LQajYYvXGykwYsPvBqXztzcnIrFosNE0NbSZYCb8lxZsgjevLS0dE/vyE6no6MjF0AE7mD/PEuMGBTe4JjKzwq0cnx8bEw7HA6rVqspmUyqUCjo9evXbgzevHmjv//7v5ckk7DIDBuNht68eeOfvdVq6eDgQI8ePfJl8U//9E+qVCpWgqAg4D3s9XoqlUoqlUr3lDCkqsGC8x0FQ8zRY4Nlsm2XZ0OaE+c9FotpcXHREAjTGRMkhBBT1eHhoWKxmENewDmRq/FuUfwGBgaMbdMUYR/m3QLe4jvg2dEwEbIuyT/ryMiIHVpg1FhS4QDY+BHEjUulknmbv3hBHRkZ0bNnzxzogBcYhluSrV/YvQiahXULkgeMGhAxMJSA1bFYzM4bMEbIhcvLS21ubiqfzzuUGr2f1O+GE4mE09y5gXAmkQHKwjFe3lQqZb0fQQuM1bjA0um0C20Qx8pkMv77Cc2ga+LGpBhls1mFQiF/R5lMRqVSyQw6wnNGWqAIvNVSv6NCHcFYOj4+7oAM2F862WAmJCP61NSUisWiuznE2khb6ESYACYmJkzoQCCBXeOkQdxNRxlcxXFwcKBKpaJEIqGf/vSnLtqQm3Q8dHSNRsPkG/gnLDSrxTc2Nmy5xTLJGIytmYVwnU4/jBtvPs8RqyasN8+cKMcgJsemzJWVFUnyBQCpuLa2pmw2q1/96lcuLtFo9N77T0H64osv9PLlS8ViMYVCIf3www8en5m+FhcX9fjxY09GdKL/+Z//qVqt5kWHEIpHR0feAkFHhrUZIoeNCnSdqHMoahCfKGvAoeEuWM7I99Lt9jNSG42GLzEUHqh9SJKjW+YXHSJxniSydTod7e7u6vT01Oli8BjAA0EVDdZ3oBlswr1ez0aRk5MTL2tk4mRaQuAPF8LPSBPxFy+oFAwcQ3QLyA7AoST5wSUSCTOREBQUGgonO9CRC/Gr0+kv0COujEINdtbpdMzAkarOIjGIBLBCsEcOGiMy+sher6doNGqxN4QZwPjY2Jgqlco9a21QdhEUX9OhkzTFz8J3BTZEl8KLPTMz42BloviQdtGVE5KM7Gx8fNybU+lEgvpBSfdG3tHRUZXLZVv7xsfHtba2ZpcbYv+rqytjfkAfLGCDEKK7GBoacj4A3QL2WPIW+CwUXw7WysqKMxVwi6GOODw8tI4RaAASEt/869evbZLAGUcnL8kuIbz1IyMjyufz91jeo6MjS3XA8hcXF7W8vKyRkRFjrRAkvIe8n0wH0WjUWtlIJKLHjx9rfHzciWHDw8P65JNPdHt7q3K57I2cvEu3t/0tE48fP9aDBw9sxBgZGdHZ2ZkXK/I+/+53v9Pp6ambgFqtZrwaBQ1dNxpO4A6IICzh09PT5igikYiSyaSOjo6s+IDUgbiRZPcakjemQL5XRnmmtFAopJ2dHZ9NijYuNUgnIAUcjOiGORNgnnTZQTKYugKxBPzAmM+Uy7vJGca6jCHi+vrazcT/ZT/9swpqPB63PGh4ePieVQxhLF8ItyOgL2MdXnRuBhZw4cGenJz0TSrJNxDWxXA4rFAopHw+71scnzSsXTA3kk6XlwmsDXYdFg8pDngcuBovFZ8f6xsPKJ1Oa2lpyXBCu912mHO73Va1WvVNTVfCjqS7uztrXbHawnZC8NE58j02m00Hco+Ojlp3h+Ts+++/91pcoAIcNDwj1BQcKFw0YKooLsD8WAN9eHioWq1mh1u9Xrfc5fT01EQGm25vb2/ttkH1kMlk3L0XCgUfvJmZGWOpBGeQOAQ8tLu7q+vrayWTSWORq6urxsGPj4/vWZTR7+LW4p/T2WSzWbPP/OzRaFTz8/O+AGD9eXe4bIaHh50OhRWU545Oli46mUzaANJoNDxBTE1N6bPPPjMUE4vFLF375S9/qZGREb1+/dpaa1LzGbFxGaE5xrKLlAh9KKuWpQ/bLSA0ySFmmmOSwYkH7IExB5iJTcPDw8NKp9N2J7Fd9/DwUMfHx3bvSbJqhk6TjhdbOhMvvAm15OjoSN9++60ePnzoZ0RtoaOFP0C1QYra+Pi4Go2G+R2UCaOjow4sOjw8NFEeJOo4t0Ep1V+soMLKc2PhvWWkwJPOaM0DhOlGzsC4yzZPvmhWLuNHluSFaYhtwT6wVCJLicfjSqfTvnUI16VbQhkAnkr3zGgfDJTlJYDEQR8KUST11zGg47y4uLinaSUWDp8yYyjdI7je8vKyg3rBlWBj6/X6vRUgyFeWlpYstqbz4fvgwtvY2LB0BZaaAxgMHEF2xViECB47IPgfe3soSKRLMZ7f3d3ZIgokw2GQ+ruHVldX/XPzck5NTSmfz+s3v/mNMpmMdy2dn58rl8s5WhBssF6va2VlxQf98ePHFq+Xy2V3ind3dyoUCqrX6x69wZHByTY2NozPEkb+7Nkzk2oHBwdmslGpgFnf3t7apQbxxVTzxzbLm5sb/fu///s9xxKrSIaHh7WxsaHHjx8rEok4QIiCUqvVNDQ0pFqtph9++EGlUsk4MmMp2k7eM5LW0EDToEQiEeX/kKeK2ePZs2fKZDK2XSPHo8Dx/OLxuDqd/uYF9qMBKywtLXkXG4UIRxcyPb5LzlY2m70XmILckAR/LmAs1NFo1NIyCD50wET70a0is6Mzb7VaXrRIqNLc3Jza7bYJaazWQUydRkCSp4+/aEEdGBjQs2fPLJVi9ALMJSFb0j3d2s3NjZl4xmRug7m5OTOX4KOEo6AKoEsisJhCR0wdejQ0c2ArePzJ8KQDA7uhyJObyFg1NjZmMgnNId0czD922ampKe3u7tr90+12722iBLdCXI6OFqwxHo+7Ozw4OLgnB+n1evaRVyoVY5FISvjsEGC4u8gLwNbZbDadaERqULfbX0VNMtfY2JjVEMTXMSqhbAjiv9fX1+6GTk5OtLi4qOnpaX333XcaGRmxz5x/n0qlFI1Gtbe3Z6aYDhELJD/fz372M9XrdVWrVXU6HcMSnU7HB/7hw4fuXI+Pj9VqtbSwsKCJiQl98803Lu7xeFzFYlH1el2ff/653T4XFxcejVlyODc3p1arpc3NTW83XV9fdyFdXFz0QYMYowuCJW42m9rZ2dHo6KhSqZRdTYy73W7X66UZcXl+MzMz2tjY8PcgSa9evbJBYmtry/8txZJOkhT92dlZ588ygtMoAA3xz+ikIR5pROh6cVOhJMCgw0RJwwS2XavVjL2iieY9pZgil+MsxeNxZx/QVfM8g/kKwCXhcNiQ1/v377W4uGgnIgQY+lYuQ0KLcEiC49PEMLGgHqApCAatIEf7ixbUUCjkTk2SbXe0yOAWQUsZLxvEALgZXypYFXgfLyvypfn5eYvZgzmjdGWSLH7OZDJu4S8vL1UoFDQ4OGhcBowTRwzB13SB19fXOjw8tMwDvzHFCv2qJJM+9XrdL14+nzcGhZQDbOz29tb45Lt379Tr9fTixQt/HhhIdLOTk/1Fhejk0ESiy52cnHSntru7K6m/5/309NS+8lKppIWFBfV6PXcpUj9FiJFnYGBA8XjchBMyEUYxLi8wSaAQ1nVIUrFYdCchyQw4OkmYb5heuterq6t7fniwSDo+9nnt7e1pYGBA6XTaeOLe3p7hB4gWEq5WV1eVy+VsHGDj6FdffaW7uzuna+FSAr7gANG5YqQgcITLhM6JoA70zgsLC5YSQvDwzLDncthJu0IyJ/W7OswldLqFQkH7+/vOEgZzn52ddbgJWQ8jIyNaXFxUKpXS1NSU3r17p3w+b9kchVDq8yBIlliTw/lgOSBFhWZjdHRUGxsbdgECD5BuxlQHJhmNRp3sX6lUHA4Ou08ATDwe96I/3gPOEF0vOR1oWcFBkXkFHZgEFlGUUdvEYjE/Nwg91vPg8uLPAfYAU+Xf/UULKl0bydwUSoI+wuGwMT8KLHFfwU4E4J0QBzrHYNYmLyFEELfX5uamCoWC1wFj3YR0wAVVqVRMdJGJ2m631Ww2lc1mzfQxvgYZdUbS8/Nz1Wo1d0gk/w8NDanb7arZbJq4YrSD5MKvz58PeXd5eelxB4Lo+PhY29vbisVi9xKwSKyiE2AUQqPL9x6LxUxWsbxsamrK46PUx4IikYgGBgY8XSBK5+eBLccRhhwIqALcmByDXq+nZDKplZUV7e/v+zKj4FMsEckzPkKAwPgy0n333XfeHAsuPT7eX82NoiIej+v09FTlctluMvSRqBKQ3qDQuLq68mhJeA9uGSRffG90fcjw+D6QAXEpcuFzAK+urlQul00uIue5uLhQPp93kUHmJemeauX4+NicBDgt7wLnBeVLLBbT2tqanjx54kKQSqVsia3X65Yx5nI5pzVtbm66M4WsnJ2d1cLCggYH+zF8EJYQQmNjY86jIEuX1elIxqampnzWTk5O/Fzu7u5sDoCkAhLsdruq1WoueBRPoBz2WN3e3iqZTJqvAB4MFl7UGMGQIEwzNBHYhiHKQqGQVRSnp6c2nACrcUHzrJi2/2IFFaEvIwF2LG4+yCVEwRxkQjJIu0dgz83HuEvLDZOGpOfw8FAjIyMWUjN+ttttjzAcerChoLCeAxV0aE1PT5u44LYLOkYAtSlmYJPguNxqR0dHikQi3ofVaDSsl8NFw7j96NEjJZNJ5fN5O62q1aouLi7cQaA84IAiZkb+QkIX207zf1hFQd7B2NiYdnZ2fCDRCe7v7/vnQbMIMYBlloVoktxZSjJzDiFEQUDeA8sPucL0AOOMPrPdbnszaCgUUiaTsd612WwaA7+6ujIxQFDIH+c3cMGBnWILZRq6vLzUzs6OWq2WnVNkBJRKJcvtwP/RWzLmIiECtpma+rBiGccXhzSdTmtyclKlUukeGSnJkhw+O2tiwECTyaTJNuyidKKzs7PuBDOZjI6OjpwdHLRbAqP1ev3NrlyyTBRoKWu1mokbJhMcerjK5ubmXDT593SlXDbowfk7yYglLWxoaEg//PCDU774RUPRbDbvde0Qj5Ls7+cCgfTkrNAhXl5e+pxxgZE6xn4ylANMleQVjI6O+udABXJ7+2HXGRcf55ALExfiX6ygIn2AYaSQMmqwYwjrJCA5pA3uDF5C/jngOD80gQzo9SBszs/P3XkSdkx6zvPnz92pIHpGeoM0gtGPkGTcI3Nzc2ZsOYyQKolEwt3NysqK5UQcKnSdFOGJiQlVq1XV63V3izjAwJcgynC1IF4mtJiMAUJJJHlEBhIJYmE40M7OzlQsFm1fvL6+9veNFpRsThK/UGZw6xN0A0jPmmlJWltb08nJiQscEYOMcFxig4ODevr0qSKRiL7//nsNDg4aSwUPA+4JKiGazaZyuZyGh4cd+MGFguQF6AC8nNUyMMATExPa3t5WsVjU9fW1nUs8A7oOsN9sNuvLEEMDFlqIPKR8JJUtLCxYMgTmRmeJ1hmfPlgpLDwQBmdDkjfrTkxMWIHB37u8vOxL/Pb29t6mXM4hz2xvb89TFB1bJpOxPIi0KoJViJmU5LH2+vraWkvULwQfBR1icCe4sqgDvE/tdlvFYtFT48TEhNbW1gwBcmlz0TGpQubh3oKX4ILDbozih4AdVCvSh31uFEi2csDUM1VgQ+W50RwAN9It85mYEP5iBXV1ddWAMLcqbT0aTw5pMI9yfn5eqVTKVR6tGJ3q6emp7X10VghrEYYjBdrd3dX+/r5arZbm5+ddHMGuIHrAJYMtPcHMHIbZ2Vl3EjiTEPjCzIOb5P+wAjeIIeInx8ZYr9eVTqctRQGKgMXd2dnRq1evvE56YGDA1s9gIs78/LwTsyBvwGEJJoFtpPMJxphJsjUVQoiQFCytFFVcL4xcZLiCO9PBtlotFQoFpVIp9Xo9k27tdtsdMwA/RAsvO90Wt3s2m7WW8u7uTslk0rkPXAAUhKdPn/o7prOAHEQRsrOzo+vra6dK4Z5KJpPGtrH58rNSCK+urrwlgkK7tLTk4GIOP1g76pJqtep3HnUFF1A0GvWlCsEYiUSUy+U8uZEcBQzBs1tZWdHIyIeV1iRekV6FkgJNJiQwFxx2bOROwTCfXC7ny4uLm5Gd6QQsm2683W7bmgwUwQUBvAeEcnt7q0qlYvgB9v3s7Ez7+/u+NCDQDg4OND8/r1wu50wBVDcUct4nGrhUKmUYABMRdQAYhvPI+wbWDHmGRj6YCRHEX7E+A99hoUUx8BcpqADQ5+fnOjvr7y6PRCJ+YdmlzQ6kyclJO3suLy/VbDa1sLBgrIUbB5wEzAutJEz77u6uMZXr62sVi0WHMeC3x2vMnwszzVK00dFRFy6wRcZ1RvLZ2VktLS0ZfObFazQaDkCR5M92fHysarXqEAlGp4ODA3cZvHwU+KDAfHl5WVdXHwKcAf0RD2P/A0LhZaOzOzk5sdEgkUjo8vLSxFFQA8llB9zBi8GLz2FDxgZ2FdQBgwMTyYewn9xZCuza2prDSsBT6YYIJUErybPhYllcXNTd3Z1x0VQqZd0ozpdEIqGFhQVbRzGTxGIxy22+/vprFwQuKhQPrFdZX193OhqBG7jB6ADBgt+/f39PanN2dmZWmYhGVB21Wk1HR0d6+PChMpmMXW10RMAWsVhMyWRS7XbbQSOYQHi3WG4IaTsxMaHd3V03JeiKQ6GQ9vf3NTQ05GKDHpkzCM+AC43OM3huwuGwJyygN4g4nh/hJnAMsOZIDvmzKFCTk5Nqt9sKh/ury3GGYdudn583EQbxFByx+V4gNzE10FWenp7aWUWjQzMQVMLwPEmT4wKCeOJihBcCjoBoDy6u/IuN/IuLi5qZmVGpVHK7TCGkSNFyU+SQW1Ac0N+x75vCODk5aX0bnlxwV7zWU1NTevv2rZPtuU2wmtIFXV1daXd3V5lMxh0rQn+KwcHBwb2wB9h5uhFeKMgH5EF0TaQ1EXgLxgbZxkvF5UCx3NjY8EuDxZSXkg4NUTsOoWDYMgQgioOVlRVvCkgmk8YfKXAImcGJcADxggbtmHRJfE9BiIbVFolEwk6it2/f3mPP6ZzYbUR3zJ/V6/W0vb3tVLD5+Xmtrq7eS+/pdDrKZrMmYYB3nj9/rvHxcYvaySkAAwU/v7u7Uzab9eEsl8vWBNO5p1Ipdbtd76D68ssvVSqV9PTpU9uNUVGcn58bH8Yqi3smHo8rFovp6dOnzh9lslhZWdHGxoY3uuLcI9QFCSD2TORfuN4IMTk7O9PS0pInEOCDtbU1FzVyM6rVqgqFgv8uJhmkXlyGw8PDljiiDgiu9gHDhwyjk19aWrI+Mxh2Mjw8bJyYXU3hcNhaYP5sYKVYLKZsNuuJBhyYiS6YBIXChokXD74kW45RF2SzWWcdU59GRkYMUWHIASrDtkvCGU0fQU90+LxLw8PDbqj+7IJKMAIdEl5xkt9h2hCL8+8plriouEkohmCFjOXkPwqWAAAgAElEQVQwovydWPfAarG9Bf316+vrSqfTur6+VqvVckI4mB6FHM0bXTWkQjQa1cLCgvWXkszKM64sLCw4aILABcYMigrrKeiQGb0AwoPQAqMr3Vm5XPaLQ5cJ9EC3TZgwMhAKJo4ksD9CXNAVhsNh+61nZ/t77CmAhCsDa9TrdUWjUZMGdDkcWLpWcDI6VQ47O4bQ90ESoLMFm8N1A9GDY4zvhWISi8WMrZHsNT4+rmw26+8OzJBRbWBgwO/KxcWFyRuKAGQiMpqNjQ0X43a7rVqtppubG7148cKXAaPt+fm5VSp0h4jBE4mE6vW6O1jeI1QbXNaJRMKW4Egk4uLWarVUr9c9+gKl0HgMDw/ryZMnxl0pgKVSyew2mmee39TUlIsjnTRjPxcDcZmcG4i+mZkZLSws+B3iTHIWsbVC4sE50OkCnwX/TEke4TkfFFWmWYwAxPkx3qN/5R2EqacOtNttf15cV/zsSOrormnsIMDgbbicaQ6C6VeYgCisf1ZBnZyc1KNHj4yz8YBhRyFUaNcZE7Dwwbox8gZvbUZo8EZYOrzYUv82yufz3mHEYWY8h2XHZcQIeH197YAN0m2CCeQkY+E/Pzw8dH5np9OxfpPCAmjOaEZxDm5DhDGFWYS5DoVC2t7eNhYM286oQhdLOk42m1W9Xtfm5qaWl5dtalhbW9Pa2pru7u4cqg2ednJyolAo5Ci9IKYMIM/YS2HnpWPKAFvj55mZmbETamZmRuvr60qlUtrc3FSn03FoCqJ0iB0IinA4rFKppEajca87ubm5UT6f19dff+0lgsi5bm5u7Gqq1+uWxNHVQYQND/ezOVEOzMzM6OTkxNtAIUSx69LpMJ6CYbIriu0BL1++VCKRUCQSMSQUi8UUj8e1sLCgUqlkCRjvINIjZEck08/Pz/twb2xs2HmEPpPFjPwKKlfgB5DggbuyW4lJAzIX7Sl5v5VKxcw6tl7WOgcvCYwhMO2YdlgZDQxF1w62ziSKZA0VAKtognkTKA0487i8gnnABMcwXdAYBbmaoLMKpQHacZ5rcMIKShaZKNF8S3K9wp0IFMm/o4hiUw+qh/6sggqegkA9GJAAIA+uQovPbQXLHHRPAHhjr+MFASQO+vkZ45DdNBoN32L8HvSZEFnYAsGKIJHokpeWltxtnpyc3Nu5w+04NTWlxcVFKwyw5SENevXqlaP26GQgPeLxuHWWMOWIkwnvwHnBS8MBwFe9uLiohYUFvX//Xp1Oxy+49CGEOZ/Pq1Ao+PdBgPBzQkTVajV9++23mp6eVjwed/wZmByQzebmpoMpzs7ObLsFTnj//r0PNB1kOBz27/niiy80OzurtbU1739PJBJKJBI+gJgXpA+JZGg/EZZz8LjUKpWKlSRsZWCfEx5yIKT19XVjl7xrMMHIkiAtK5WKJWd0f2traw5KGRoaUrPZtPMI9QZFj2eJDA+CDK//2dmZFRjANhMTEzbH0F13u11vIsC4QIEOhUJOZOv1enr+/LnzD4JJ8+QQcOHTrfF+0SUXi0WFw2G7EIFtIKQ4P0w+GEWKxaKlUhg/UA8cHh76jAadfjRcTLecUWoGrkE+N4Qx3SGcwPT0tLOC4XHYNBAK9RfyBTXruOe4zJgoKObkZHCWcrmcGzpqCJM0BRUI5M8uqEhRjo6ObNWk9YX5urm5sUMKFxKkCngrHQUdxv7+vgXjnU7HYDLjARIZSdre3rYbiG4DWymYJcJbigFfBGJ0xkDGdRhh9HtsSOU2ZnEaowKjBKz78PCwRc+A87yE4I0sG+PB8Wfyc0ciEetzEZPf3vYTiI6Pjw3gRyIRdyOVSsXEy7Nnz2yNxMq7srKira0tdbtdKw4IEsHme3h4qK2tLSfpX1xcKJPJ3HNhQWbx8gLsn52dKRqNmliB+Z6YmLD8pFwuG2fDhcSzRUs7MjLiThT8FMdPPp9Xq9WyZAp2GHkXSUgYBsCGOTQUEca5k5MTRx1WKhXrUCmOFxcXfg6xWEyFQsGpUT/72c90d3enb7/9VpL8/tI07O3t+TMlEgkTY8TmoZnFqXd4eKh6va5KpaLr62ulUil3vEw1Z2dnKpfLxjWZQoiEpGMNrtSGkKKDTKVSDibh2bx580alUsncAxMG+H8mk1Gr1bLgH7fY3Nyc1tfXvREYN9fFxYUODg60t7fnCwl8lkYMiziSJCZcGqxUKuXRHFwTAgwcGHItGFrNckkCVCCuqEnBMR3TAg0Y7wjfJ98fATlcAFwI/F6I5h9dUEdHRx2CHNwZRQHhLyb1iOIpyftyxsbG1Gq1PF6Qa8qYQGrO5eWlx+qrqyulUinF43Ht7e2pUqlYh4fGjO6YKDxeOB7szc2NSqWSQ3/x+wKi09mOjIwYTkin08afuNUgyFiOtrm56ZwCIviCzKKke2HVpDRR3GEjwfW63a5HFha7Ec4ACQQG1Gg0bCwAXOdyI2QkmUzq4cOHKhaL2t3d9cvz/PlzR6cB3wR3GHG4eOkYN9HWTk1N2TpMjGI4HHZAxdbWljsoPPb4zJvNpjErwrxRcGA9LBaLJtsikYj910w8HHJyLNE3gqmxWQFlB5pFChTPiMua6eLo6EhbW1s2QdCZfvrpp+62ut2uPvnkEx0dHalerzsuEsw4Go3q/PxciURCf/u3f6svvvjC7/LZ2Zni8bhKpZJWV1clyat6yKENFkXIHPBjPiOXXLvdNpdAh847VSgUPCVxqUCgoioAFqBDxSwD/IRigfXJuPsI5yY8HDMPCWFBuI8MUvSz1ATGac4ySh0cSdhy+VxcNMPDw4aWUNRwPjudjur1uvW36IKDnTFnEUwUcg6Oo9Vq+aKFFAUrRssuyTzLjy6o4JE3Nzf3FrQhvYD4QQIUdDZMT0+b9MEnfnZ2ZgH2H6sA+IJhytGD4fZAKH911d9WCslxcnLiEZPA49XVVY+WsKdoOiGEwNf+OB4OEB4bKyNIIpGwdvTm5sZbPoOF9Pz8/J49jZcLKRY/K9hcoVBw3B2MK4cdTSOe8LW1NS0vL/tn2t7eNhbHupFcLue0dtQAkFyVSsUd6uzsrAqFggkuDsTp6akzaxk/gTCwh6K5xVZI97O4uGjZDT51gP5cLqfT01MTLel02qwssrFaraaPPvrIzDUQCJg30qyZmRn98MMP2t/fNzMMYcXod3Fx4QuZQ4E+kYPOxUQRRfuZ/0NC2PHxsQvP2Vl/aykhOlxomBgGBga88QDcDrZekjH88/NzLS8vW9DOYee8AAmMjY1ZZTA9PW08lj8T5h+2m04dOR4xfiQ+UYQTicS9SyWdTvviOjk5Ua1W85nnvZM+KHfI1uh2uyoWi8Zv+dnoyiHbgCr4fQsLCy6ESCWDCge+F0huzj+jelCfzfd/c3Pj5gyikgYGCWLwOdCxMr4zITE1Hh0dOdEr+NmBMH90QR0aGrKMpVAomDCCJZTkh9lqte7FwzGKdzod7e3teXUBuBNWU2CEi4sLyyWmpqa0sLDgCDXMAaVS6Z5AOJ1Oa3193Ws+SJmiw6NDwOoI7hMcJ7DgAZhjWQMToqggO2k2m8Y/R0dHHdZCViSKiEQi4W2oa2trJmm4kVnTUi6X3UkxEiKq3tnZUbPZdG4susq9vT2PRhx8SX4mdMepVMqY1/v37/XmzRtlMhl3FdzyxCSy8ZRD22g0rBUOMqJBYTyXiPRhn1ihUDD8wWcKbsX85ptvND8/r5/97GeSPgSrYB0FF8vlcr4UGR/ZqBqNRn0gsUsGbZavX79WuVw2y05XTYB4MAgFWQyrRT777DNfmoRH8+8l2USCAYUFfBCNEJmJRML/rFgs6vj4WAsLC0qlUv6sGDTAiiH/wAE7nY4XQJbLZU1OTjqekm6JCzsYJALzj7OJs0EXxxZVSb7YIGowDAA5QWwxrlNMibucn593rGUul9Py8rJOT09VKpXU6XSc6yrJkBAkKeYaVD1ALgTn0LiBrRMDiJ2XZ4dV+qOPPrIMEAyfMCWaKoo3crvJyUnDNlzGTNacW6bvP0uHGgqFvOqX6C4AYYBvmFE6Cvz3kozTNBoNL7uiQ6NTub6+Vjqd1ujoqN69e2dLZDgcNns7MDBgprZYLLrwoS+8vb31mIZeFBxUkgFuOt5yuewHy3I3xo5kMumu5fr6WpVKxf8tBXRra8uYIP9tr9fzAwjiLoDzjA+ENSBfYlFaLpdTLpezdx1tKd/B6OioXr9+bfyYIkokH3bEdrvtMYmDQgrUxsaGFhYWtLW1dU/TR7Ei3AXCJplMGsJpNptKp9PG57jsKCoUQaIUwRC5JLa3tzU2NqaNjQ19/vnnHkspRoxgBEdDluFyQmJDp8H7dHl5eQ/75jnybhWLRYdjDAz0w5OxM+fzee3v76vdbjuV/9GjR/r444/9bNDP7u7u6vz83IcLhw7vT6fTudfZ0+0QuMJlen19rUKhYJIoHo+70CPBwxYaDByiC8PCDLONLJFNCDQlKCPgGqT+skFIWjpvsNhgxoake7ZLih9EDpGKoVDI6griLclRBc4aGBi4l8caDDOBj2i32858AL5g/B8bG7MmlukimNZPIxSLxfTs2TP9zd/8jfHx8/NzffPNN/4sGE2Qe42MjNghBsQVzCTmM8NBoF3/0QU1l8spnU5re3tbV1f9VRbsesImir8ZVxGCZ3RpBHRQRNDegVUwIvB7WLzFeIYEhXDadDrtVCWkH6TS0/5DhKGHZPQg8BlMh8+HJId0p6GhIS/1Ayti3xCANqPJ9va2er2eR29WZ4DVEEUH8cDlAM6XSCRUqVT0/v1747UUjlwup0QiYZcWLzY4MAlXLMnjRk6n0xobG9Pm5qaTiejEKT7VatUyGjpaMEAuC0kubBcXF9rb27sXyzcxMaGtrS2TQkjYVlZWfFAkuZM4PDxUoVCw9IXfC8ONqH5zc1O3t7fK5XIuTDwzBOBoOZmIkMygvyVEhE6GogL0US6X1Wq19Nlnn7lT/+STT7S8vOzULbqeUqmkdrttCd7+/r7tsly0dDTYbXHhcWGAHXPA19bWdHHRz2OliCDWp5PnfeGZt1ot61MpdOQ/UITz+bwGBwe1vLxsCR3dHuw93xtFicJOahgOuuHhYe+6p/CsrKwomUxqcnJS9XrdFzvhQRDAnGv03N1uf6UKl0GQUITowXbOShbedxb3UYA5H0FDwuTkpD7//HPHWfLOTUxM6N/+7d9Uq9W83htoAJcn0ALfEcUdiIpaAfzyowsqOktJfnEBnbkJwTqQriDEX15etr4QoB0IAJwJ2UXQsxuNRn1AiNR7/fq1vblEf7EsjXQYhO7BosfDpWPLZrOWx1CsCXcAkEfOxUgYi8VULBa9bjqIj+J750bjsuGw0ynRJfJdMJayq+bZs2e+GEhs4lbGIAExxmeKRCIqFouqVqtKpVJenUGXzDqVVqtlAi0Wi6lcLvvGZXXFu3fvbBEdGRnR48ePdXx8bAzx7u7u3tRAx3d4eGhiY25uzu8JRBRECBfazs6OEomEX3Z2Uh0fHxtiwUTBQQFTRi3R6XQcnMEhwFBAOPPFRX+DKzgnuloUB5ubm3b5NJtNffTRR3rx4oW7EazOExMTOjo60vz8vHdtEVWHvZilg4RzTE1NWU5WLBa1tbXlAG8KaTKZNBYNVkfBoTAS8MyI3W63nVOLXhQXIcQY0M3w8LCLDucAlntnZ0eDg4OGV1DdBAvZ+Pi4U7MkudMOPmdsncATqB2QoYHdM5VCBPd6Pa8bZ10PeuGbmxurMnDsoR6iY+eMMeYH06mCTYDUhyzX1taUSqW8bZcJmtolyU0YUAFEMLvZkLX9X+P+/2dBpT1GEoFejhcZhh2sptfrOawWeQSCYYTPrIRYWFiwfIY/jwME8xuNRk1OwGRyuxAQm8lktLKy4tuLMYw4L7SH+/v7ZnTBZklI4uWhYINlDQ8PG5eieyZRiJsVFjGVSjn3k89Xr9c1NjamdDptooyf++rqysWZ4sKYTVfEBcGW1qmpKWNBFDNCXCABT05OVK1WvQep1WpZ9YANE9yt0+no7du3evjwobLZrO1/Ozs7JoJwK01OTqparXraAPoAU8pmsy7ovV5/PQ3W2fn5edVqNf/sOImWlpa0tbVluZrU30EE9ir1ux1MIWC3LMvj0pibm/MIfX197a6J0R9H3PX1tdbW1rS6uuochEajocPDQz179ky5XM6HPygo5xxwWWWzWU1PT2t/f187Ozs6PDyUJENZXEzFYlHv3793k1GtVnV0dGScGpzyxYsXyufzDtAmwxYfPjGSQD2cGd4JZIwDAwMqFosOjWbVN24oJEAjIyO+pCYmJu4pDDAQBJseciFYY8L7GAqFvGYEqIcRGQgH6IpLoVQquTgNDw9rb29PKysrzung0qOIAZHQJDA9Bq20Uh+yYgXMH/+anp5WOp12TgWyKulDRCdWU84iTQQwAIUUM9GPKqgEFwS1cJK8ZI1bGiCZETgY2gBhRQdHmj4mAYoY5ABZjbBpzWZT796908HBgdl3ijo3JSMRQmUsdrxsV1dXXpoHBnZxcWF4gpcMwTG3GHgc3QXyHGQkfMGxWOze3qrJyUnF43FFo1Evx8OFgjWWlyUU6i8YvLi40OLiovWtjCO4OyqVivWopVLpnvRscnJShULBnQWM/tjYmJaXl/2z4q2HDBweHrabBT/15eWlu0NeKkgJviNWOdNRAT98+eWX/jv43MfHx3aJ7e7uKpVK+Xtj1GUqqVQqLuC8zHzG0dH+grvBwUEtLi4a4oAAevTokYaGhuyqwmVGAa5UKn7PCPbhe8pkMiZOk8mkJPndRNlBlsT09LR2d3fNoN/e9jclMFWw/51cUyYV8OpoNOoULljjeDxuzoFn9e7dO18yt7e3TkoiqhB7MqJ+GiB23jMdVatVd48HBweKx+MOpB4aGrINli6Pbp6LnVEYhQRcCQn9o6OjDsMJZlbg04ew4zlx0U1N9Tf55vN5P3OwbbSvwaQ5ilgkEtH+/r7140B/6E4pfMi8MPNEo1GHDKE8waBEVwr5TZgQ0yoTCxDL//Xr/yyod3d3XtdLceXGoW1H18f/IofhEMKu8yLww6KrxIJI+Eiwc8QFBXaDXhKXCyAzNy7FZXCwv1edjlmSWT0kU+FwWOvr63ZKdTr9nUW8oIzXPHTGP25xNJKdTj+5vtVqqVarmaEuFov3AjAQLSNcxpkSCoWUSqVUq9W8TpeQiWA0HiMenTKeZUivTqejcrns2DlwNXz37HFi5ELxgKU0KOGR5M+8uLhoOROh0jDhFJCbmxuVy+V72xtQbxB4TYgL6oRut6vvv//e4zqfe21tTYuLi8bdOBjE2N3e3mp/f9+OFQhJjBJYGaPRqDY2Nlzs+GdkhqIU+cUvfuHCgnSJFSODg4NKp9MaGRlRoVAwTIKumPSpoJYU8o7v4vr6WgcHB0qn0/r444/dDXGx3dzcqFarqVarqVQq3ZN0SR+2/DL5YPNk/XY+n3dTwTu6ubnpZHpY/eBEwUXA+Tw+Ptbc3Ny9/0+qEwoI8Fz4BmyfV1dXhtFKpZK2t7dVq9UspwviwkjryMUFl8TLD2wRxCzpwrk0qAdIBJFhjo+P++e7vb1VtVpVpVLRw4cPNT09bXUPeDHkGMQmK3CCdtpgcAo/L///RxXUVCplYifohBodHTUewboJfgVdKrzojP3gqZKMwWHXAzdh/xHdJ7gaHZ8kdx50SBShRqNhkf7Gxoby+byOjo40PT3tbqfZbOrsrL/zvVwum0Xkz726ujKBRpdM0QruOCdGjRAYWMSVlRWzsLe3t8rn885dxSgwPDxsWREWUFwlbH0EnmDMh7knU5b/LrheBBxsaKi/v/yHH34wmcH3zcGMRCJ6+/at3r5968MF7oktD1yQcZVE/aAfe2try4ROOp121whOSvfQ6XQ0MzOjBw8e6LvvvrOs6ObmxgHeKysrymQy6vV6HskJ7ADPA3/GiokkiVR7Sb4Ye72eDQcff/yxJif7iwqRyAFJUbw56OxSGh/vb4N9//691+Tw38ViMb1//96Qyurqqubm5pTP5/X27VsTkuivKRbACVguIVvhFyQ5kJziHY1GNTg46O632+06QLxer2tgoL8PjMseuARnG5cVnSvPFBiFDjEej7vhkGS5nSRfVsic6FaDzRfQCFZdSVaAjI2NWVIJLs97SQfaarU84UJIQeBycWGP5T2gQUFVgPqoXq9bW0sqGYFBnAG6YOANCG6yNpgsyWpgev3R1tOlpSUXUjAQPgwjoyQTPIi+6Va5kbgdeThSf+QLBraCC5KNODo66hv/6urKmaTDw8PGLolKoyCwDiQYNg0zyu8B+wmC971ez3uWGM0Za+iiwSM5yLC3YIispqDjQAuI3pRugQ5bkkXkECy4xebn512Uh4aGvOEVzAqMku4c9xbF/ubmxkRRt9v1c4FgIWhifHzcxAgZkRQILjqE3CSNUXz45ygNiMJDHsRqCpQHEGC9Xk+Li4sqlUp+lw4PDzU8PKyXL1/68PGcg/t8ghZCugzSyuiYYfq5PNE30uXS+fD3kCFAsvzPf/5zTx3dbtcd4suXL+9l3/L5yuWyxsfHbVPlz6TI1Ot1NZtNra+vm2AcHR316mlGaMToXEYso9vd3bX28v379447HBoa0rt377yaZmZmxvI8dLMnJyfmL1jzjNoB/gOctlgs2rhxd3fnVcpMEGiUb25unL2KiSJYgBOJhLF/zh3TJ98Vm2/D4bCfP6oFtKE0cnzPqDx4N2iCwuGwu89gTgDfNSYVchQIUeK5khVAY0VxpfNFNgdngtTrRxXU2dlZ5fN5OyyCLoFIJOKCwzhKig1yFcJCwApJu8EiSicWj8eN4aGZ4+9hNI3H43Y2cMuzHgPgGEKAAsiBiMVi9tsztjx8+FBjY2N6/fq1uypeagoGHXmv1/NYQwwY/ng65U6no1Kp5NAJiCJWvaCL7XQ6DhuBOEJSMzIy4jzRhYUFJZNJHR0dKZ1OG8NmXJb63TtyGSQoyIS63a5+/vOf+2VGDZDNZh2sjHNnZmbGGBM/e6fT0dzcnK6urnRwcOAQZLSzJycnWl5eViqVcjcGa8ufC751d3enmZkZLS0tqdFo+MAxvn733XdaWlrytFEul3V5eanl5WV3JziIEHVXq1XHId7c9NPhYYtfvXplYTY4OpgtGP3c3Jyi0ajGx8e1s7Njy2wkElGv19Pvf/9752NyeSwuLpp85eBtbGxoYmJC5XLZFwf2VaRWdG5Pnz7V+vq6u6pwOGx5GcYFoJyvv/7aAn6exd7enlqtlr777jtJfXcP7kC4DAJ9mAyQwAHlME4zjXAh93r9VT+QeMEV15C7jNjsd6MIcqYoxhiAgDbgC0hsYzMxihKmp+AlFZyq6KoJO6GZC0rbUM9gMyW8mkmXXF1G+CCsgqWZDpRLDvMH2GswKOVHFVRkFmBxPAS80lR40nYQ07IlE4soLx+4qyR/QXQVrVbLyTB80bC2dBcA3OAxt7e3HoMYgUOhkB84+AxjCMk/3JpBbzQBKOPj49bmkUoEQUJnWKlUvOeHDpUXTJLDmINyHgTLwYe1vLyst2/fqlwu+4UjLo6LhDFkYmJC79+/19DQkEdJ8F+6a/JLO51+WDbMLcwkB50i+/XXXztKDraUwkwg78LCgnK5nFZWVu49B7pYbni6N6yLuVzOONTs7KwlXlwUmUxGtVpN5XJZIyMjmpubUyaTsRoBy2+73db79++1trbmtKqBgQE9f/7czxAlCjZU0sHoRCmULLkLanf53FNTUyqXy/qP//gPTzB0dqlUyhMPUX+cDSQ2OPt2d3dNBCFAn5ub80YFDvX19bWDgQgBX1xcdBD6wcGBA5GTyaQePXpkFYUkF+9KpWL3FhfbxMSE+YWxsf6iQwoPzzZYOGDx6WjBSIMj9fb2tqT+BMm2AgheCh3nFd5leHjYnejFxYUxd2SLQG+ZTMYqG6AzoK6joyM3NhRNVA0UdCYdFEbUrmCgDj8bTD7qHmAWiiS1any8v8gTvoGzxgXzowoqXQZjOWM4nVKv17MfHrxlaWnJq0Z6vZ7dCDB94XDYYR78oLTq0WjUxUuS8Qz0a3h3g7pPulmwJEnWPfKCXV1dOVgFgBqGEDyQ8ZULADKH1v/bb7+1NpDxY2VlxQns6PQYkyn64DqSjMGen59brtPtdrW0tOTRjH1NSJYQfkOcISZHCzo2NqZMJuPD+uWXXyoWiymVSnlVNj8bAmwCpJGmgFOD8ZI4xCiEVxwMm2T0s7MzvX792qz53d2dIpGIx2+6GnBCbJzVatXkAhCHpHuytc3NTTPHQCTn5+fKZrM6OTmxlZnPSLJXrVbzZQ5sQqbAzMyMTSPI8cByp6amtLe3Z1XC2NiYVldXLZK/uLjQz3/+cy0sLPgiBoc/OjoyO86lUyqVzJDzGa+vr/X27Vsz2SMjIzYKhEIhr1M+PDy0+gKskIsCMurNmzee6IgyvLy8tEUZNQR6WH5O9J4sWaTpIHcAGIkud35+3gFHBJPQUXPmySAgXIR3bGJiwhGW4JyM3Sh2wHCZQrGUAoPg5ML3T6NRrVY1MDCg09NTm4xwvn366ae+pCncnHPCnZgYg5LP4AoXfj9ZBqhmuCh/VEFl1OdGReQaTO3hpcYVxcuB3QyWE6XA5GR/TzssaFD8ix6RkAKIpqDvGA0axTOfzxvgB3ifm5tTrVZzHB77YWjxDw8PjTmFw2FrVPkcY2NjDvjAqri3t6dSqaRYLGYd6/7+vq6vr61NBb/EBgtTzv54skfJ5iSYg5GDTM16vX4vixFXEKlBvHAUuK+++srfOXZQpFtInZaWltRsNn2BSXL3TcYrL0s4HL6Xy7q4uOjnQkcCg7u+vm67MN8hpMHFxYW/GwTa3PQDAwOKRCJORiI0GumPpHvp/u1225ASukDwO/A99jjt7Oy4iJfLZT18+NDs++PHj4EAta8AACAASURBVI3rUjhxfeVyOR0dHblrv7m58d/ZarX029/+1qQZY+D+/r4tu5gvxsfHtb+/76mABqHT6XiiQAtbrVbt3EulUnr79q03HYDp7ezs6Pz8XL/4xS8Uj8dVKBS8erperxt+GxgYsB4aETo6apoGAocgmgcHB+9pWsmq5WJl4kQbzXuJDRYCiBzRUqmkeDxuqAEFzh93fEiy2OFEsDtyPCYrznYw3B0FEDJOFDmQlaFQSMvLyxoa6ufAFotFFQoFQxySrG1lqSHTJZ0wHS5NCPAkErQfVVBJUKL9B6OhewNM5qWHOZXkF4z2HAYSbAsP9tHRkRqNxj38ErKEvfPo5EiAYdUBAD5darfb9QOHASedanV11d0eLDLOm0gk4i4TzEfqkwqwkKlUSslk0n+P9CHGizEZwB98GNkJUifcV+DLQR8xhWh3d9fBJOPj404lYjni7u6ub2SW3hUKBRcibJO8iGg7Ed4fHR2pUCg4v5TvCmkMrCid8tLSkl9mDvnR0ZGKxaLdQw8ePPC6iP/+7//W5OSk9ZxBWzDJ9WdnZ07p4rO+e/dO33//vRKJhCYnJ7W8vGzSkQDwx48fKxwOW9dJAhj+60KhoHg8rpmZGY/yT548cYcIOcrmXro6SB+wSvzpwECDg4N2fKG5vr291eHhoZO7CF958OCBOp2Oi9rk5KRqtZr29vbs8Nvb2/N+Lj6XJBOPvDtMSIj2y+Wy97pBnEpyEwEOm06n75kr2GWVSCScSgacxPPAhdXr9VStVrW1tXXPy04HSnwjk+kfX6QU5mQyaacjeQzkUXAxY3RBFgYhzUVJRgLEMbwGjQqXAhPy3d3dvRxlqU/Kvnr1yhJHlDM0gygMgjpTmsOhoSEnbqFawoD0owoq+kRsi8EUG4TF4Hd8UIBnLHGnp6fKZrP3vPfEvTFuoL27uLgweP/LX/7SGaPb29uq1+smng4ODtxZcGNwuNgAEBScDw4O6uHDh85HPD09lfQhVebZs2cOMrm6uvKIgdQKl8b09LRJDhhACiQSI34vIzBSn3Q6rXa77S2bY2Nj3hcOg761taXR0VHlcjnr9K6vrw2sx+NxxeNxd9i4wRirg8QDEiGmABjliYkJXzTYLNG6cigKhYKy2awymYy1kwQfoyfm+yd/ANsgekwK86effuqVJrzgi4uLur291VdffWVYhrg4yMl8Pu8OHiUBFyFdDAVnc3PTzi7cMmhIe72eNbIQqAcHB5bGgL0S+jM8PKxGo6FGo+G0q9XVVQ0P9xPC0BsvLS1pbW3NZBmecJoEDjh4JZAZAnIcQ2R4EnvH8wPywSnWarX0ww8/KJfLaXR01DI2pgDcbKgWSFTCDNDpdJwyxTm6vLx0DgW7uHZ2dnxpULQYg8E1IYuurq6Md0OCARsRPI40iYYqFAp5fEa9gUQRXBqlDVAF4StMmFyETBYUbMJsKJxSfxolTImfie+A9xoMnSaKi4E/FzyYRLgfrUMlWxS5DytVO52OR8CgoJ6XGVIFYT1jLZjP4uKinRQsacNWNjk5qU8//VRLS0vq9frBI2juuC0o2JADQayK2wdWlxsWJxHyIoo54x8kFAYAMEA6KdhH/vygSYAkdQoAFxAFpFgsutDwkgQF0sGABzzPSEQODw99qNDmgSMjCQPjQzrGs7u97W+qrFar9mSD78IeI4nhUNOFMo6hEaZTYHtC8OcpFAoepXK5nFc5A+vEYjGNjo4ac2WNRRC7JnBjYGDA9lqUEny3aFYjkYhxXb6/4B6nwcFBZ5qyAFLqj5t0aE+ePLlHMPF3XF9f6/j42Nto8cdziYEfgxMfHh6qXC47J7darZpsxcBCov4PP/xgohPxf7VatWSJDov3g+cdhJDIE4Ww4f0lL5VCHtRX00mCBTIZEV6CgWZra0s7Ozs2KoyMjLjLZSJst9va3d21fAhymhqwvr7uy6nT6W9LSCQSikajWlpaMpyCiwmVCoQTZiEkWawwouOkaeh0Oo5jlGQ4CxwciGFnZ8fNEDABjQrnFestn3t2dlZzc3P3AoMk+bzSBf/JBZWiSHHgIfBicvNxU4LH4GS6vLzUwsKCwXXWlUjy3ia6y6AWjI2RQ0ND1pg1Gg3VajVLhSiGExMTJpzoKJHtwCheXFyoXC4rFApZawlzzmGEaAOP+v77742zgtXc3t7qzZs3logAUgdND9y8iLWnpqa8hI0EH0KPYdsJwCWN6urqyvAH4x7CcpjeYFbrxcWF/eEwr4lEQpL07t07d5Tc+oynAwMDmp6eVjKZ1OvXr7W5uWm9MNMITD/yJ7A0AmkSiYSf+8XFhV6/fu3/hiJ7cnJi/Ljb7e8KCzpXkKThdEkmk6pWq2o0GlpdXXWoczgc1vv37325BDXRYNO8hxSl/f19Y+r46MfHx+0qm5ubUyQSsbyJqYzs2XC4v2BQkgkfoCAucHBKZISM+1w06J1597Feo0NeXV3V6Gg/CBl8k3ep2WwaaotGo3r37p0zCnCgjY6OKpPJaH193WQsxAv6Ty6vdrvtAsy54yxtb297FAdmm5mZcVDR7e2tCyyNDaoYuIPb21t3x+SaEuwiyRpR8pLRdaPSYZrlIkG2yJlmQqEhYNqcm5uzPfzNmze2zH/xxRf+s3lXmCCDuSClUsmZDORN0FGjBMKODUTzJxdUMCSKCdgggDE3O1ghmASBDIzxYCB3d3fuuJCyTExMeNxrt9vOReUXoztyH1xHANW8EGBqwZuzUql4IR2FGxYWuQ9+/cXFRXfKw8PD91LxIcDIBuVnB1wPhUL2Wu/t7en8/NwH9e7uTvl83i8GpAQPje+Gl0r6EPbLZ2EtN8UTvInvDM9/Pp93FiUi8SB5Q5FkzxWXXDwe99Tw7bffOnTi+vpav/nNb/Tw4UOtr6/7ZoeQofNjgwPPCisg2l9Jxt2ZYgYGBjyq0xlQfBOJhHJ/SPfne4C8ocvgkuI9YzwOJhE9efLEq535XniHX79+rbdv32p1dVWxWMyQVi6X0+DgoK27JIwFsbqgy+b6+lrT09Pe/3R+fq5arWYskKKDU2ppackFCZJodnZW7Xbb61To1o6PjzU6Oqrnz5/f2ykvyQqGcLi/oQEYJzhNYqfGcIA0jiYDnBXHFb7809NTE4PYVBuNhuEu9L+Qx+QigMliHw3KH0dGRtz9j42NuWvf3t72RQMRxiQZCoU8cdAM0Piw6ptwodnZWTP9X375pS4vL7W3t6f379/7zw1efEjyeJeDpDXbHYLRkOi/0db+qIJKIeWAglPhNqEQwJbiorm66u+CokUGT0N7SiHkw3LAgq4RfmUyGeX+sI0QbIpCTmGB7SbAgZFLkl0P6Pnu7u60uLjow8BtNTo6qlKppOvr/iK7sbExFYtFEwBjY2PWAHIwK5XKvcQeRpNyuexNjsAD3W7X4SmsfoZJn5+fNylCZ0r3DjSBRCwc7u++ocumI56ZmVE6nXZo8MHBgXq9nl68eOGXEqUGz48iMzg4qGw26z/rq6++MlCP42hpaUm3t7f+Lu/u7lz0yuWyJS7IX2Cok8mkxduEa+/t7VmGRdYCB4VcALDfcrnsvFY84eVyWc1mU1NTU/c8+OhP9/b2nBifzWbdyVCE+bxYeCkeNA2E4BCtx7tKahnkVrfbX4DIZU53xs+FuoSRc2VlxdMIJCPecVxOoVDI2bjtdtt6Z4jhpaUlpVIpd42saJZ07/+TDVwuly1LIjcD2zgdJ3GV4MDIquAkRkZGbA+m+EOInZ2d+d8Bl2EfRVoGkYjBg9pBtjHKCJZQHh0dqdlsGrqgqbm4uDCUhd4WWAP9K5sRUJNATqM1h7MBFgT7x6CDmoZmgb+LMR+X2Y8qqMFqjICZUYA8T0ZSgitgygkTQENIt4RQGAUA3ScfPmhPlWT5z9dff21ZFV0JQDv/DYQYBQNZFEk15+fn3lPP+MMtB27JDcYBIm6Qdp8EcnYMwUIjqSAVn/EWOGJ+ft6J8Lu7u94bz/caVBjg7KGgBTNawa9zuZwxTtarEFTCbiTS2fn9ZIWSZC7JWk9Gn1QqpWw2q/39fXebtVpN09PT2tjY0KNHjzQwMOAULtxgTC3IZOiuuShPTk68DgMfea/Xs/YRtjj4XMmaBYdFCjU5OaliseiOBWUHhT4ajSoWi6nRaNzD5C4vL5VKpe7ZQHu9fqD5l19+6b8T5QeQB3rGsbH+jqdUKmVpGzAOHALYOlMO8X5YQumSaFDAMNFE8pmCWxcqlYqnG9xVIyMj7kIZR7F944ufnp62WuH9+/de883nQENNgYD/wJFGFgJNTzqd9vqjarXqdxysFEySLbV02Uw10WhUMzMzNmdQdMEryeJgAuCCHxwcNAezvb1trSsXNwWPDpNUN949VCEQmTxLmi4mApo1Jp5isWjtOLWLQvu//Rr6l3/5l3/53/7lv/7rv5o9QwQNrtdut11g+Xek26fTaYP06EzBV2DlYM4uLy/NqE5NTWllZUWPHz+WJLPpv/71r7W/v2+mFJkHjHUkEjFGR2fHzqdEIuHMTHS1l5eXarfbXqvBS8m4ValUdHl5aWkLulm6BNQL7Mwi87PZbOrg4MCRYJJcOLjBuaFZybu0tOQCHbxAcE4xctVqtXvec0g67KC9Xj8Fiu8I6Q8dBLgzqgf+PG7lubk5NZtNlctlM65cmjw/fPUIxBHQEwZM0ES73XZmJqoKdKvhcNjOLdjq3d1dT0E/+clPDMdIcpFhvQeXsCQfNEgVYuEePHhwrzAStgO2yztJgaLA8TNNT08rn8/bbNLr9ZyXys+WSCSMU2cyGWPUdJ2jo6NO16cbpFgyKfHvwf9QkcTjca2vr7ubbjQaLmq4dsCM4/G4nj59ei9ch4IBFkt0INg809X19bXJpqmpKad88d6AT4KboimHbKQhYDILvk/Dw8Mm2Q4ODu5lMiCpoxGhkHGx8GxpgvgzaX64aEhhI5az0+kHj2cyGV8e/Aw0dEgiUQXRxEBWEZwDPEiEJZ+TafOf//mf/19r5v/ZoQYrNl8GoDyHHYYONgxLm6R7u77/eAMjGBvjNJ0LMiy+0JGREeVyOb18+VLNZtPuIFp1SLF6vW7RMsJxvkC+dA4QkgnGTH4OmECCIfg5kNNMTEx4sR7OLCyLQ0P9qL+PPvrI2kFuTcY5cDd2paNgCDLbg4ODqtfr3nlF10A3MDIyYugAq+bOzo6Ojo5sw+RyIQMWkwOuHLqH+fl5pdNpnZ+fK5lMKp1Oq1QqaXNzUy9fvnQouCTLcxDkT09PuwMkbxSsdm9vT4ODg85BpeA+ffrUNsVWq2U75+HhoQX5YGyxWMyJ8IVCwWu4NzY29Pvf/97Y3NDQkFZWVry3jDGQA4GsSeoTF9giuUBpAjAG7Ozs6M2bN7702Os1Pz/v4Ob9/X29efNGzWZTq6urfo8JkQGT63a7isfjLvrz8/PWX4+MjGhpackjKFsLUM10Oh2tra3p/Pzc0NLCwoI+++wzX1I3Nzf+jpl0GMUhgaLRqBPK+E74HsErIWOA7cLhsMNugM+CVk206agqcDfi0KLwsk8O8bwku8hweMEl8Cy57HBk8vehI5X6ag0kYr1e794ZQQuPqxPd7MBAf+dbEMoaHBx0Bw8vQudMY0YjGAqFHHBObfqTCyptMH56sKXBwUGD0EG8k46k2WwqlUrpwYMHGhsbc4Hjtrm6utLCwoKLNUHV+Plh1aUPHluwzVqtpoGBAa2ururx48dODIfIAW+DDKjX62acwZX4QtCsQRQMDQ2Z1WNU5MXkwNRqNY9MzWbTsixJ1qnt7++r1+u5AO/u7rqT7XT6GyAHBwftEuLhgpUhTQInbrfbisfj7jYolDCWiPjT6bSKxaLevHlj6RMa2qGhIe+VOjw8VKvV0u7urkKhkDY3N7WxsaHp6WkdHBxYisWLx02fTqd1dnamly9fKpvNmii4uenvjj86OnLU3crKivWDEIi8rEtLS+78UExQZA4ODmxBBjOPRqNOW2+1Wr4oc7mcVQt49FkdDGbPJRmNRm1/HRoacoSdJI+8wAsQU+Fw2KoS8k8JRuYXch4KvNQfRfkOMcDc3Nw4lb7ValkdQog1pN7Y2JgToFqtlubn523HRQXCn8t+JNxcExMTxrV3d3etGWfFM8UGjJf3jhAjCly323WACc0AEkfgMN73ZrOpYrHobQ/Iy/j3hOaAg/IdkWeL4gQnJJ0g5obT01NtbW153TkZBmitySEIZikDpfDnAQWBkQdxbd5dVBXBSMJg4SRCk/frRxVUnD1YAGnpGfFTqdQ9VpIKv7CwoGg06q4RHIMvE4aRm25hYcGJ9xQKftFF8gvNWL1eVzKZ9N4krGS5XM5aPjAQDhg4Cl5kgg7AnjggGBWGhoa8vuLly5d+kQjOAE9CEjU8PKxXr15ZI0fgBQRdLBbT2dmZ9vf3DQ8AmiNjmZiY0NLSkll/1BB0qGhC6Ur4uRipVlZWnJwDww4RyIGBFEFe8/Of/9wQyMXFharVqrsZAii63a4dbZKsGqjX6/r666+1sLDgRDCeF88EGzHYVzKZtAaTd4iORZLhiEwmY3MEJNH+/r4VBnNzc9apIpWCGOI5np+f33O2LC4uGtPD7lsqlXRz099l1Gw2XdAh+NDphkIhd7iSXFxKpZIlP3d3d8Z3IRNHRkbcPdKpHh8f+/nMzs56vQh2TooN2DuQ1dTUlLLZrLfuLi4u+j0eGekHX6dSKT18+NATBBjg5OSktb2EfQ8ODlpXenp66tXXkmxZpfAErbV3d/3dVphQMD7gtOJnwENPcby5udHi4qIxSd4ncHBs1ZxfeALUC4ODgzo8PPR54kxwoUQiETP+rVbLZxZCHNyWEHcmKP4dv+hMgzADhflHr0DhIdIu8xcG/z+CYfBUvPpTU1M+kNy6Un+tCsw6tzt//h8XT34BlBN6jHyqVqt5+yJ/Tq1WM0FBBwK5hL7w+vrajh+6IzphbsGLiwvLZLCkSTI0wUqQoO6QQ8Z4w/jGn4OdDjwqiGkdHx8bs0ErCTgOM8u2BHR+WDI5/NFo1BcXLzvfK37lt2/fan193eEhjE+S7H4DJ8dtRPeJbAxYAuz18vLSyUkrKyvGtQntIOmLiw1NLXKoXC5nbSjEBQeBZ0ZXTZdJuAmfHdhpe3tbJycnbgK4uHHvZDIZ45KQI2CaaFFRXvBzE6azv7/v6EKK+qtXrzQ1NaV0Om0Nq9TvbhiFIbWAWS4vL5XP55VIJHyoeUcosvy+oMd8fHxcFxf9zbM0ChRh3gf4CLpPujepn3sRzJhgDQuFg4uId3Z8fNwwB58LCIJOPx6P6+OPP7YjrtPpOPiHYp5MJm1Y4V2GHJLkWEsIKMZwXIVMvryTTBitVsufm8/WaDS8dZZ3hPcaFQLvMDIvApQgyvn++W8prrxnP9p6ysiEhhMME2FyqVRygC8aNsBh6cM4RNrL9PS0Rxh+QFLLm82mU3ywRfJi4oQCBAf3BCelEDIOkthDe48cgwcIrgvW2Gq13KVADCAxKRaLPsTgw6gYIpGIwzTQ53EIms2mMpmM90ERbYYbCPvt2dmZsT9uP0KJp6amvE4C6RgOKcZE2FUIJCQ4jHsQEwi8wV0LhYJOT08d3jE8PKxKpaJareb0djItx8bG1Gq1rLNk5OGFXFpasu6Xwo4a4N27d0qlUpqenrZtmK6AQ8TBJnSZn49VJ4Rwp9NpZbNZXVxcOACFIo4GlZSkV69eeXSV+vGO1WrV0MH19bUePXrkgsMozDNlLF9eXvYFmUqlTCINDw/bORckMTh0QDR0bsAF6HsfP35sfeXIyIjevXvn38v2hImJCYdzIPG7uLhwnsPHH3+soaEh7e3t2apLN4zygAAa9N68q2RqAIHNzs4qlUo5Ok+SFSMQiIODg5qZmVGz2fQqG5ohGg/UHtQA8G82wZLPQfdKIVtaWtLx8bF2dnZUKBRUrVbdEPEdQWhTnINci9TndyDf6vW69cOXl5eWxkGYkcnMGcFwwgQH2Ubd4L2EHPxRBZVQk0ajYclBUL9H0jwseTA9CcwJNhpZDJszcZgwfmG/LBaLyufzevr0af8DDg/7RQfHAJchdX99fV3dbld7e3v3MgfA7ugC+VLA5TjgfDYiAvHOM94Ew1AymYzOzs68WZSOl26PTp2LhqJBES6Xyx6DIK5yuZxSqZQF9+Vy2VkAFHjpA0nIjfzo0SO1221rEiFY8IlL8otDkaJbevr0qcXcjUbD60b4eWOxmDKZjN68eWOB+c3NjQt7EPPlcAHus5+I3WKIrOmcCMBh/TCWxHa7rRcvXmh3d1ejo6PWmNLRrqysmKwjnezk5ESFQkGSvJOL4Oqbm/5+sHw+r263a9Ls2bNnSiaT9mnzd9Hl4Txjbxi/NxaLaWZmRnt7ew6QAQtksR3nABvl7OysMpmMbm9v9f79e7/TWK6RMeGqQxXSarVULBbd5eESwtseXBWDvIvvgKmj3W67eIBNkv/ARNTr9dwI8T7TRRNWHgqFDAMRgch+MaCCs7Mzj/x0x4VCwZs85ufn1Wg0bB1GXgZnQag13wUdKBhrOBx2gBGXFwUW+AYilgxUiidFF5w0SJTyHdCRz8/PmzPguaObpZnkkv6TCyrVmO4CvPDsrL+6eHBw0O0/GYowe8EAgmC6Db8viFnd3d1peXnZmBg3PUUkl8uZPS8Wi9ZiAimglTw5ObHHnHG80+nnYOJZZzxHdlUuly2t4YCRKUBXHdTOQkI9fvzYkiFkJRSeYOYkDwySCwIBeISLS+qLscEcUQjgLKpUKhbVY0XN5/Oq1WrufNF3ElLRbDbdnbMOhKLc7fZzWNkeiT+eTu76+lpPnjzR+vq6SqWSSqWSEomEVlZWrD8Fi6NQ8dwLhYL90PziBX3y5IkqlYoxL8Tnl5eXqlarfj9wqkDIdDr9jQh7e3tOImNc7fV6yufzSiaTXqcCARqPx/WP//iPloQhcQtKbejo6E4vLy9VKBT0u9/9zsE1yLLI+Aw6aIAN2HRKQZ2cnDT7zQI+IKpCoaByuayVlRVrtcGqCQniLO3s7NgG/fbtW3W7XaeBLS8v3xO67+7u+mKDbGGiODw89EYGHFQ3Nzf67LPP/A5IHyL2ILJOT0/dJQK7ra+vexwnFImYxPwf4gi3t7fdlND8cEaAYoKSvCA/QxGFeOWz0NjNz8/754KkQmXA76dAEnsZjAgEXuNnBWNl8gVv5u9lOsVG/6MKKmL2WCzmZBeIo0gkYsYRUiF4M6fTaX8px8fH3lKIAH5oqL/Iq1qtmgHkC0mn03ry5Ik7MUnG3Lrdrk5PTx18S6dJN0pgBngsBS3IqmItI9E7FAr5xmMECO6YorNFwpHJZNxN87Ag6ehE0asyOiIzYyQBCyX9HAMBBz0oYwKHJnwD/SUMMfISMhDA77AUxmIxKxz4b/P5vK6urqw/5dmxqz7oYpuZmdHnn3+ucLgfObi6uuqQHGCcN2/e+MJAP5z/w7bYaDSqYrFo0gtcDlUHnyuXy9luiIyMMBrwfIipIHEYj8ct4yLJaH193TjaRx99pMHBQTUaDW1ubprsOzg4cO7E+fm58V46eogPMDyCkUn7r9frevz4sUX52WzWnRpSPUJ5gpfh+fm5N3nSwdGFYhnmEofwRVf7V3/1V/ewTkgaukQUDWhVOfy3t7e+iEldozi1220XKPDj7e1t5fN5Xx5HR0daXV01FEPDcXNz48uNQB2yicPhsL788ksrWZhIeW/AYcGjIalg8GmwUO7QzKDNBlMl3B3iipwNwn/u7u7cWFBQyQfgZwB+YWI+PDx0V356emr8G+fkjyqofOGA97TRVH9acFwN4XBYjUbDnUUqldLExIRWVlY8ruMA4WaAjQUbjEQiZs0pqNjLCHdGeH56emqfPKQH48Ld3Z0Tq7DRkSFwdXVl/zZFhKSlgYEBZbNZF8BgEAmifAiiw8NDjY2NueNhIdjs7KxjzSS5ECF8R+c6Pz+veDyuZrNp7Inv8vT01LufePiw/TzQUCjk2DU6Yrpa2FKS/vludnZ2rAPe3NzUu3fvfOtCXjx//twH+ezsTA8ePNDz589VLBa9A0nqdzJBydnOzo7DpguFgldv4JBaW1vT6OioNjc3dXvbD1TGuQb8wCaBlZUV22SZirhQcYwFLaONRkMnJyd68eKF1tbWFI1G9fnnnzvGsNPp6KOPPlIkEtF//dd/6fLy0ttAQ6GQ8vm8Dg8PlU6n9eDBA93d3ZlxJ++h2WwqnU77MJ+ennoZIQQiZ2NwcNAqDi79k5MTY5LdblfZbNbmEngKxmJGe4wahHCHw2F9/PHHdsUBz5BWzyVCUAwi+GASWrPZdJYu4yvnDePA4eGharWaGwuIQaSGFHRGaLo7fob5+XltbGxoeHhY//M//6N2u61arWbbN5cLwnuaJqYWLKuoIqT7hBCmAWAM3JrI2oKuM8xJXMCcm9vbWzvC6JaR69FAck6ZLCWZi/mTCyrjMQdVkuPWSOkGLJ6ennaRwnmC5AYSgIQZqQ8gk0zFKE6aT6VS0fb2tn76059KkkcnsI5QKORgiPX1dYvFq9WqU5/Gx8ed5hPMdgTsp4uNx+Pqdrt+iWEnEQfH43FjNgT58vdgC52ZmbE/HyyRMQTbJe4SijbFMZ/Pm4UuFArq9Xr2pPOwYaDRyOHDx6mFBbbX6ykWi6nT6fhzlEoldbv9CMFMJuMDhW4PAgbVRCjU38n1k5/8RA8fPnRxyOfzlj8dHByY7OJ7pDsdGxtTMpn06AUJkkgkzOSTCyt9kMvQTdFpAC+8fv3ajC8KhsHBQe8Je/nypUmD+fl5LS0t6aOPPnIiPj52OsOZmRn93d/9nQ0MXJizs7MOO4aQAEtdXl7Wzc2NSqWSiwv4KT8/QTVDQ0P+sxgTKYTwAOC3rC2vVCp6/vy5UqmULyuaGUi5kZERraysqFqtan9/3wYClhWOOZ3clQAAIABJREFUjo5qZWXFExYRmre3t952UKlU3Kyge0Umtbi46M8F8ba6umrWO5vNKhKJKBqNev1Ir9fzBXd4eOhnzkWFiYBwcazsYLqQpVzoTLnsVUOzCwmOU5D4QQijPyatSJfCWQcMFrS/0ykjBYQXgmPBgg75SwoYqpMfVVAXFxftM+cB85Bpwaenp83+Ub1p/xmnGFno6sCMwKTAaCXZHQTBEsQzwLuq1apDdAn7PT8/9wvECEswyejoqFl9SBgIkomJCSffBPWlAwP9hHQKGP5yEv/Rf/Jzc/lsb29bpIy0Bc0kDx/rKmwoagi+BxwsUv/WTqfT97S7yHgKhYLJAbDXw8NDffPNN0qn01pZWfE2AumDxI3sSMbxarWq4eFhxWIxZbNZra6uWtCdy+W0u7vrqDUIM5xx6XRax8fHSqfTmp2d1cuXLx3CzEqT5eVlTxm9Xs+CdLoMyE/2kW1vb2t/f9/ifzqgra0t++S73a7xx729PT1+/Fi/+tWv9A//8A8OQeHy53IjKOSTTz7R8vKy4ajb21sVCgU7/SiMdCqrq6uKRCLGur/55hsz7vAKkmyc4ELjneag03jc3d3du6TJVT07O3NH1Gq1tLm56e0NmUzGzQVdPlg9/w0kH2Qx+6T4DN99951KpZKL76NHj5TNZhUOh+2G6vX6209XV1fVaDS8zVWSNduI5jkXfIezs7P3Jk9JNnbAZbApghi/k5MTbz2A7KUJYzTnEr28vHQIUZAohhgMh/s5vzQrXPjohoN+f2CS4+Njh3ND/l5eXmp3d9eXz8HBwT0HFe/yn1xQ6ULx7XKrBHEeSXZQEIpxdHTkUS4SifimoYiRWsWLhiCcsfX8/Fw7Ozv667/+a7fXY2P9pWlIKng5O52Ozs7OvMs+GCXIgjpGnqurK68gZufN/v6+6vW6o9o4EGAzHLRoNGrcC03ewsKCJFnITDgKch46GTDY2dlZnZycWHoFFsroCkBPKDK3NIB7EAahOKJDJbSYEBTGKbr3vb09M+YXFxeeFoh6g3Hmz56ZmdHNzY1ev37tQ47YnC6sVqt5RMeKeH5+boUHu+pPTk5sVwWe+X9oO5PfxvO0/r+T2HHi7LHj2PGSOEslVV1LLzPNzPQB9EOIkbjwJ3DgAAg4IDgOEkck7vwDSCCQODAHJKQBNJrRrF3VXWuqsjuJ1+ybY8d2fof0613f9G965tc9EKkFPV2VxN/l+TzPe3vYqYWy4vPB2NgaEXHXajXdvXvXRBwv7TvvvOPxk+8N88vvy4sxNTWltbU1GzvoypFecfiDTaI+4GBlrEc6Njc35y4oHA6b4WZVDfIzMEEUKsGEqXa7rZ2dHfMKFFcsyWDNdLo41kZHRx04gp2Tra90zBsbGy4SHLiDg4NKpVI2j6DckWTTQjabdaoSqWnX19cql8uuBZVK5ZatO5fLaWpqSqFQyM0Wziy0peiveR8J0qFRAIcGV8XhhVEF8xD5G5K8hYB/gM14h3km6vW6FRx0s8HMCMwJPDfgvd3db1PeUN5wWHzpghoKhQwuw7yCS1BwONXoxgDXObUkWRKFx/ri4kKpVModBGk6aNW4UC9fvtSDBw/cFSN3ymaz2tnZ8aiJnxtSh6RtXn4wyk6n40LKywwxBdYUXBMRjUYdXwY2RSEEHCfyb3p62tkFwTGmUqno4cOHJpSQufT39zsEJSioZ6yWZI3o69evtbq6qlwuZ+kQYSu4SDqdjiYmJpTL5dwpP3v2zPhcs9n0Cw04T4Gu1+v66U9/qrGxMX3wwQfGsijGdLXZbNbdfFDKRbFnrGVxXU9Pj6GL+/fvW49MMQAWQjVxcXHhDobCsre3pzt37uj+/fsmGXG2UQBwjJ2cnGh1dVXxeNyqjOHhYbPU6GRrtZqSyaRyuZwP2UajYU0xjD/EHu47prX5+XnjibD5ZHFK8mTFe8AkVSwWLU3iQN7Z2dHAwIB/B9aZxONxuxCZFBCZt9ttH+bATHfu3LH8CQKqVCqpWCzaUsm6lbt37zoWEDkTq1DC4bCePHmiN2/eqNVqKZvN3spVgKRBmgbBhMwJDoNCyVYB0tLq9bohACZHHFiE+/Bu0XBhCeagDGrZJbkmUY8gmoLTTSgU8mSH+4la1Gg03LmCo/LnKPbAAOhav+jrl6ZN/e3f/q3DHLhRkEt4ajk5YMs/n1mIO4iugDEboBlWOJgoz4UiOader+v58+cqlUp68+aNarWa235OjqGhIRWLRWddtlotr4yFJENMDmkF6M1LgD2Qky7oeOrv77ekhp+3u7ur4+Njb1iEvJuYmHCX3ul0FI/H/SCGQiEvNwyHw5qYmNDJyYkODw9NAACRMNZcXV3dsvnSWbCUjvW7hKcglwGE7+3t1dLSku7fv++Fa4xqSK2wHeImQnIGbhT0vdP1M1qSgcrWUf4dQwYdNOoMyIN79+75Zby8vFS1WtXW1pY7JhQA2HXJQUA3StRbLpfT3bt3LWUD1wTn4zmLxWJaWlqyQwmlCBmrR0dHWl5eVqlUkiRj3cjwuGbSjV8f5j1ogQQXBBenEKNRBadDUgbOz/uCHpNJK1gMIW+Bu/je8ArghHSpuP4gY7u7uxWPx5VOp52Qtb6+fqtDrlQq2trasl2Z/W1oqMGRgfVQQ0Ca4XoiOKinp8de/8nJSaXTaW+AIK8AnBJdKwl0EF78d3BpJHFk0DLNpVIpZTIZT14cwEwUvO89PT3mJnj/IY1RDrTbbUMEjPtYmyORiP7kT/7kyxfUv//7v/e6BT5sUCbFD7i6urLvlwKAwBbGDZkFI1OlUrnFQjP+MD4hl9jZ2dHe3p4ODw91eHioWq1muQkPaKfTcbgyBQ8cBu1YNBrV7OysRkZGnG+KFpaVEOC3PT09SiaTTsXH+w5WA+sLbhu02VYqFdvtSLkHxyU5K5FIuJOv1+vWTpJT0NPT44BcZEsI1JGyBVdI0HF3dd2keuFtRsRNcaOrJGshGPpBFwCxEwqF7B1n3c3o6KgODw/t6olEItrf31dvb68ePXqkdDqt+fl5zczMaGhoyC8yMqK+vj5v7WQbK4TRq1evtLW15ZBw7ilpWrCsrLVhtKNokz9KEZB0yxwRTDFizCbQg2WFm5ub/h1wbe3v7+snP/mJyuWycV/W21A8eX7i8bgnB1LRCNQZGRkxWYPvHrIkEolocXHR94fii8Af6Afsbnd31xMBOlGIV7SVHAbIASkYyKKmpqYMrWHc+fTTT/Xy5UtfY0ha8OhQ6GapIIUIeR0xjsgiJZlFBzO+urpSPp/370TNgCiiOWJa5CDEYdhqtWwk4f+n4PK+3r171wQ4RZ3RnTrQbrcdJwnMCJfAAQERyLsY/Dlgu3/4h3/4C2vmrxz5BwcHLa0AzwCzikQiPl3pAru7u2+tC+Zi8eCji6N75aGnQCD34SHAMQIR0d3dbU91LBYzVgoRQFQXYwNeeyQekUjE6yQ4idDkEQaDYJxihf2Tk4pRGWwHnzanM6cieNru7q6XAyKtmZ+ft1gflw7EAlgkoy8PAj55xmJ0p2/evPFBgu6WrEti7SqVillSXCG40TY2NjQ6Oqrp6WnLgvr6+kyyTE9PG6tFbRFMc793757S6bQkWfOKiLvRaNzycf/sZz9zxkGlUtHKyop/T6LxgioEDmOCVDh4OGCkG6Irm81aOnP37l3t7e3pyZMnLrQoCTByUBx6enqUzWbdMUejUa2urlodgcsL8i+ZTNr7z5TRarX0+vVr9fT0WJOZTCadPJ9Op50fgGoimUwaM8TcEg6H9ejRI9VqNf3whz90Z4U55urqSlNTU5qbmzNeeXh46MaB55DGANwSrPvg4MDEZ9C/L91g148fP/aBjAaUZDAmBZyBzebN7i8gJ/S6wD0U9xcvXmhlZcUF7ezszGEsYKKtVsuflUmA6wrDDnSHVpfNB3TtdOVBLTEcDs8roz7QDE5FVERAJUyz1IZQKGQ4EVL0C2vmryqo/GVOHnBQvrC0Af4T5ssvBzHCCwVrR5Gm8PL30Zoi3+HCIGGQbsYtLJXBBwSHBF0cF59AZHDDo6Mj9ff3u1Csra05Fo+xAGkJFklJt04rYAnAerpsCiKsJIWJvVZra2sOFSGTlS6JfM6+vj7vuufn0KGUy2UX+42NDb9IjIt0WrxkkAuQZLxAjUZDMzMzdigRFoNsiM0EJycnWl5edkeZTqf14YcfmgkH+mi1Wnr33Xc9dfT29uqjjz5SPp/XT37yE+t3cTQ1Gg2HfRA+0+l0lEwmbcfkOYpGo3rw4IEymYxxu0wm4xUgRA7yYm5vb6unp0d379691flzv4JSPu51uVzW48ePbaTY3Nx0+Eo+n7flEiUEior9/X135sGQm6BEKJ1OO/meAHIskRxeQfyRKYzpC1adw7qvr0+FQsE2V1h4QmXAIhlvcSyCV4+Pj3viajabVtA8evRIH3/8sf/3qakpa5bL5bLlggS8Dw8Pe9rp7u62DZomCnyeIB1yDDhkgSCCDkxwaTpXmh4aLCZjdLtIIXm/wuGwiUkaB8LNSWADziOsJ1gfent7DZ+BT1OMkVbBDX3pgso35P/H5YTuj7xIOgmsXmQwNhoN7/1BO8pLDz4EE4fTSZJTzhnL0SeOjo7q4cOHfrjIXkWjB5aFk4RiBIEFXod1jYsFiA6Typiyvb3tcbu/v98rh4kiQ1AOFocciqQgPMD44+n4a7Watre3b8mp6IYp5EHWG6z6/PzcYvFwOOwMgvPzcw0MDNxyiMDMj42NeWyi2+Xe/PSnPzXAvrq6qrt376qvr08rKyt2xnR3d5vRTaVSmvksrm1kZMTkEnF1FD8OBeCNaDSqFy9eeOXGixcvPJ7NBDJN2cNEZ4xxBBvr7OysrYCEyyB7gywElkFHiVYTAf/V1ZULA+oU0tLwkUvSb/7mb7owjYyMWOiPXxxYCfcU0A6sN88hzjgUB8TnARUwnbBkj64NOI3nDdiC7419khG0Xq9bLw73wLQH+QquiIGm07nZa8aGitHRUf3O7/yOg3bK5bLhJ/gIrncwyu/09FRbW1ueLBOJhJLJpAPNybkAM726utL09LSVEbzHFEHyDCA2Yd8jkYiNP/AYnU5HY2Njmpyc1PDwsFqtlvezMdJjlQeCQSHDtWACZFJFyUBYEx0qRf0rW0+DNwCSh2xKOgoKwfX1tYFkdKZgSUGHAYJ5NGYIZrkQYKi8AKQ3EYALO4vrqqurS5VKRS9fvtRHH32kZrNpkT7C4pmZGb148UI7OzsWTyOiR+dZr9fNZFKUpJtOjhQjLIro5/j9eMH4u3xfYAggjVKp5FFienraCgVyIqPRmy0G1WpV1Wr1FpiOb5nij5Hh/v37Pt3Pzs40MDCgxcVFdz8EjfT396tYLLrzY5FfvV7Xw4cPNTk5qYWFBbvB+LMTExMex4F2kABB1HU6HRUKBWtwkcocHR1pdnbWBAzXF6KrVCpZVD8yMqIXL16oWCwqk8k4t5RDFxVET0+PJXxIqiDVlpeXdXZ2ptnZWQduoBsO7obq6+uzNOuHP/yhfvrTn94a7ei833//fR0fH6tQKFhqg/RoZGTEkwsQAZMAOtOrqytjkLjqkBzBL2xsbNhowk4r4i+7u7s1NTWlYrGoVqvlRDUKdV9fnzY2NjzVHR0dqVgsWhp1enrqJXw0RePj43r69KkuLi40Pz9vaRITGgx3LpfzZMh+MqA7Ot6Liws7k7DugpFiJmGpIT8fLJ8ihuNsb2/P35tNphRz6gz3saenx/vlsKgSmA2xBJlF6hZKFHIuBgYGbClFuhd0KzIJ0QjRQFK3vlJBxf3DS82DDZuPfKa3t9ci5Var5Y4VHIVoPsTjwZeLjqKrq8ukCGEVnDBXV1deMsboSWd7dnamw8NDbWxsKJ1O30oL7+rq8j72ICsZiUSM/wFL8GAhlQJfoYgdHx/bhcXPAKeSZGICsoJuB3CeLoGlbENDQ+7WLy4ubuFrqVTKDC2HD4U8SJ4RysJoDIaE7Id4QrBfCAzwz7m5OX9uUvERb2ezWV9/uttqtaoXL14Y68blRe4CI+np6amGhoac/cAailqtplAopK9//euWAxE0ze75WCxmQ8Dw8LCmp6c1Njam9fV1u+iOj4/tvCKRqNFoaHV11dh1sFvGSQRZR5e3u7ur73//+5bOMMqROCTJkkDcUsQtSvKYmc1mdXl5aTkbnZYkdz1MBdFoVDMzM2aX9/b2DJFMTExoYmJC+/v7Wlxc1OHhoSqViuEzujC2H0hyh0oKEqJ53ETYV+mw6Jx5fmq1miT5vabLYzqp1WrGLnmHgQ2wizNhQIQ2Gg198sknevz4sfmP4PRarVZVKpV8WCLIx9nGO0qzhayJwwkCmuIIsQmuiiaVbpPAHSRc1CX03sBbXLuglpXplEOS9+wrFVRSxEkXQurESMK4DKaBNS2bzXr0kOSO4PLyUjMzM5JupEgE0ZLWAxMLE88Jlk6nrflEKvLq1SsXD05zwi5YohaJRKyPRcd6enqqqampW90yJBd409DQkF1eHCK4gziNBwcHNTc3p0Kh4GuBhAUN7sDAgNlQAHQcIqVSyTgNhgkgiHq97k2QZKAy4tMFS1Iul3MBI0uBhx4JFzFv2IHff/99KwhQL3DAvXjxwrgfL/PMZ+HP+/v7Gh0d1UcffaRSqaRXr16p0WhocnJS+/v7GhgYsGKhXC7b/QJmWy6XNTAwoOXlZf3oRz+SJHfksLIQY/l83rAKOQsU+rOzM+3u7hp2gugLMueoPzY3N1Wv1y3nwqTy9OlTPX361Nhwb+/NVopisegmorv7JqQH4g8sNplMKpPJmAQpl8v+b51Ox+u9V1ZWVCgUNDMzo3Q67UNlZGREDx48UKlU0scff2wN7vb2thUm/DkKSat1k5o1Oztr+RyFkjByjCOxWMzjdzAIGYUEOHO5XFYikdD09LQJmWg0ajINpURvb68DdJA27e3tuTPF2vm1r33NB3Zw3z3W14ODA21tbRk2goCkq4R1p+vF2k2Dxu8i6db239HRUTsCQ6GQsXSKdTD8ni3AXA8OTuoXhRWIDZ178ODgHfxKBVWSbyhedHBVLHPgdZFIxMwoEgisqZ1OR4lEQs1m06c7L5MkZ34yEqCBo7Or1WoG+cnalGQwHKcVpFWz2by1twfLW7vd9mhAjFskEtEHH3zgMb6rq8tbJtln39fXZyAblq9er2t1dVVXV1fe8gpby3VjrKPQItEAlAebHBkZUTabNYQhve1sEP8XCgUL2MmdxM7Lgw4GiWWSMR1RMg8v2kGkUYynaGkJrri8vNT6+roDX7hnCKUpQMfHxxocHPQaYCzGXLNnz545L2FwcNDXjdUv09PTlt4MDw9rbGzM0AN20Hq9rpWVFX344Ye6c+eOJyQ6B6YJXngMCe12Wz//+c/18ccfq92+WbwHpMJ9yWQyXuHy5MkTb30I4sWMz11dXUqlUmq1Wrp//74nKHJ1sfICFRUKBb8H77//vubn591RESSEnRhJFqQVwSqkHBFKzQTAWIvGGT7hwYMHHmdRrnA9gFrISkWTi8nm9evXKhaLmpiYUD6f93MIXPXo0SO7oSBXcZiBbV9d3WwOxil4fHysarVq0wy4MJrcmZkZuxwxBzEtcI+x+YKD07jlcjk/rxyEcAscHgTrYB7iHQVGoWYAAwU3tEJkkxeAJfsrFVRwOLoBxn8qP+MkTiii5dg2iNsG5xC6REY9NHGcLD09PS6WXNhyuWxZAyEJqVTKjCApNt3d3bp3756+9a1vKRaL2ZXBumvCh2EPyYw8Pz+3KySdTjt6DgKKm4TkI0hOgOMUCgU/3KgWMCS8evXKMXlTU1O+BrxkSLrQ4MGew7Ii1AZDOjs709ramvb392/lbQadZPF4XG/evHGnhtSJPe68TODPQZgmkUi402BUrtfrVkAgyxobG7M1d3d31yEzJPK322397Gc/84tI99TVdbOkj7Br0rGIFwyFQlpZWTHWGtyECyaGhpLDPR6P6+DgQKurq1ZxEMrRaDRc9Ensgvw7OjqyQB+skAPq6OhIP/jBD5TL5RQOh0108L3m5uYk3Yz9xMMxCeCpf/fdd11AgKUIHIlEIvo//+f/OJMB+Id3ARx+dnbWCwYJtpbeJpgxQfFeQoBhgiAfuFKpqFAoWIUQDoedAjY4OOgQ7KurK42Pj2tiYkKTk5MujPl8XqFQyF75nZ0dE1XAfmyDZaoAKyV2T7rhJFZWVpx0dXh4qLW1NSsCgIJonmiAsJiDbxPcHY/HPRWgAkBRQZHl+Q4mVXHtMIegCgKqAgoEg724uPDk+Wt1qIzgyJJgIfkCf2o2m3ZgMCYRgoxtrN1u/z9EFWQA8W08AJJ8ugS99Szh2t7etsxEurFpQpZwMjF2N5tNDQwMaG1tTd3d3crn8/4zjGLBXfK4hBidAaYzmcwte2FQTE9XnkqljIu98847PmywtgVDTsLhsIsbSw1ZwSDJnTmuFjrP8/Nz22ETiYS97a1Wy6TNe++9p8HBQRcS1qrQSfJQoJ1st9va2toy0bOwsKBUKmXXCiqKZrOpZ8+eORKR609HhZ0UBwyuGDqEbDarjY0Nj1XgZPF4XIlEwiQGmBxSMjYgHB8fGz4A7wMOiMfjmpiYUK1Wc1GByMQxRJcJeYqUjKIZ1BAHrbDkRRSLRTPLrIShM0QWh/gd9rvT6WhxcVGRSETFYtHGDz73O++8o62tLeO5JJgFnVWRSMQsfiwWM57LqAoWycZclCUUFGybPB/AV5VK5dYaZXIguMbcw4cPHxrLR2eK24glhTyHk5OTzhNFWz08POz7QBcOiccBzWECBg7cyAFEUwXEMjo6qvHxcWdpoBUn0Ux6S4RDgtH9B/dYBQ8oJjM8/2wtiEQiPmh/WdrUL3VKffe73/UvgraTBxChNx8ekJ3xP2jlZIRHX8qJBnZCd4TbhBcZH/Tw8LDXTCOAfv36tTdgBpfh7ezs6Pj42HFkSHAYzRgl6vW6WdWgnx9AG68y40Y0GnUXBgiPnfbg4MC/J6HXdNz8b7ykQdkMmCOjH3uTwF7BuigKOGR4qFh3AVGFvKXRaNxy/6ysrHhEnpyctD0TJQIvbDabdUeHoBqNHmEp1WpVr1+/dmdC/gI6PkZdIgdJNpJkAqrZbGppacmrq/F98xwlEgnNzc2pt/dmsR4vHCw50wHqEGAUIg0RiKN7HBoa0tTUlP8d+y2HS6FQ0NnZmVebX19f2y8OlseoCURAF7q8vKznz5+b4Dg5OdH6+rr1l0hwEJyj9/2824hrFwqFfOCiqqHDI6Dn8vLSUrFSqeTxn4NLkn8//qHrRJrFxEJCFAJ3IJ1Go3Fr9QpFiFGfgJ3j42PbwYHzCBMhvhOegA40OM5DIlH8satSzJgm+J0ikYgymYzu3LnjdxWtLo0YMizCncLhsJ8NYCsObCYBGiyw36urK8duIgll11c0GtWf/dmfffmC+g//8A8qFov2zuNuYPQFq0BTR5ISJyRdAMw2oz9CZkkGf7k4nOyQTbT7l5eXtoMSsYZOEUyD0zORSOjhw4eWKx0dHalarer09NRdATcVworRhM/EyA/Oh/YOETpwAQJtRmFkHBwu+PIJPSGyj00FPDCcelgF37x5o+7um8Qj5ErEE7JRNKi1xSGGvRUhPYsNsd3ysBDBSMoW3WWn0/G6jZ2dHa9+vrq6cveF1bavr88mCw46lB9gTRyaOMyur6+ddIRHPvjv7AfjOcHy12zeJMPjF49Go7cOQjrQjY0NE0VYFSlYdGXI1drtttbX1/Xpp5+6YIFJctCfnZ25CNNI8JzwkkGqwfAzatJJxWIxw1dIu9BvY8c+ODhQoVAwOYqulmcbmAx9NPbnw8NDT34oI+jIYLBpapj+OASRSeESPDk5cU4EkxBFqlQq3cqjRQaIsw19Ke8/7zfdqSTrsoPBKsTh8XxIb1c4U0D5nSCLs9mslpaWbuU2cD8wW/Duoj4JWrTp2oOOp+vr6//H9orSCBklEGCn09Gf//mff/mC+s///M8mZrjJjLbY/xAOg6VyEZG1cBGRjlBQAHmDGCVLwMA7gvYzfMhDQ0Me2SCiKFR0sJAb/O+ckJzSFBJuMhgMxZ3TDs0bARpYF+koIepIsUmlUi4uwS2YuFe2t7dNDEA48Bnxre/s7Jjdz+Vympubc0Ax3a0kKxsYvbnuiLQxD9BxUKAwNmBUgFjEeRIcGU9PT7W7u+uUorOzM7O7Q0NDhnjoKPFzR6NRbWxs3FIMAP5DvvT392tyctK7xBj/cfBgae50OpbLwCqPjo562VuQyeX34/pjw0RDSifNWLe1taWdnR1FIhFbbnd2dvy9WCsCyRZ030WjUR0fH2t0dNSbbDc2NvTs2TPrTEdHR704knzSYPoZ5gcOTXIMaFDILiWYmkOfkHSmQQwpEGiSjDkjd7q4uPC14PPQrITDYb+7U1NTZsSJr2s0Gn4OLi4uVCgUtLq66qWSl5eXtw4dCFB00Z+PxAuK9ekIG42GyUdqBgeOJMsOIQ8Z74NhJnTaQYcZhHUQc+c6cbhSx9DA8+/UHnBwDpqurq4vLKi/kpTiJeaXoSPlJMWRAa7CGDEyMmIQGclGKpVSqVSyX35kZMSp9KQ0IRfB/tbT02MBb6vVUrlc9tgdTIsCLAa34qWdm5tTNpvV/v6+HUUER0AyQGpwUely+/r6dHh46IR0fg4RgHjUR0dH7TKC4aWbhRDiwRofH1cul9PW1pY7mu7ubhUKBeVyOc3Pz6tarTpRaXNzU9vb28pkMib0INtGRkY0NTWl6+trLS8v6/Hjx5YAweI3m031999suUQCAzSAYYIXl0zLSqWi1dVVkyKsbuaAxJyA9CoWi+nw8FCRSMRrPThkSBnitAcrSyQSunfvngYHB014ofEEW+NFgIAE3mi1Wt4NRWoU3WpQbH9xcaGJiQnrnIGomBzGxsZAuhEJAAAgAElEQVS0tLSko6Mjy+xYTy3d5BKwGQKtcr1e1507dzQ/P6/e3l53Z9KNFJBEMjo7TBZog0nFx4e/vr7uLv7x48eqVqtWHVD4W62bNSeVSsVpaj09PYaHJLlJ2d/ft+QuqJihAF1dvd1Ke35+7k2sjOjIH5GAoSGXbiYubMNMiEGjCjpTiix5FLyLXP+BgQEf6kCHZA4PDw/b0ALuy7OHxReJILwKDQjvLi4uLKQchFh0Kb7gowcHB4aKyI7Y29uzNI26gBWfmvOlCyqsFqsTYPDB60gYD0qEwHWwetLRXF9fa3d31yG46XTaOGNQQ0kAAyM2F4BO+eLiwo4Mdo4jwgbX6erqUjwet8YS0TwxaNzEbDarmZkZjYyMaHd31yn2jPtEl2FLBODu7u7Wzs6O8R7wYopKMMsRdxUOJkZxpBtkD+D0kd6uQ97e3vZpC7SBvlSSizmJXclkUvPz834ostmsLbKNRsMRazhJJFmryjiFvGZqasoSGLr28/NzvX792hrT8/Nzy48Q2JPmNDw8rK2tLY9VyWRSBwcHKhaLju0Lhu1MTk76QBobG7vFRiPbQpcKE8xYj6qDQrW7u+vCD2Fxfn7uDFAsjYz5YOC9vb3K5XIqFAouSEQvsq2UQ5Brn0wmVavVtLq6ekvjzHjINcZhQ+eF7x/CkqhCSf6cwDdYtjOZjP3yvNwkkrGlgOeDn83Yj7QMbBzMkELMO/bxxx8rk8n4AITxhlQNrnnGLIDGGgUN13dyctISJPTCwG4c+Di+gM4mJiY0NjbmIs7nTCQSWlhY0Mxn2bSM77xHEMF0wEFbK9ceVQxwS5DkpXFkOt7b2zMmTCfMNE29+UVfv3Tk//jjj32a4zjgpAf/A8QHA+FlbzabHgFDoZAj+Mgn5UPiQKLNhmEDjyO1HcZ+bGzMbG6QEEIYzw0CMmCtBqEljAjIqsBsq9WqsV66WXRpQWFv0E/MNUBSwejDzWGJG9eMzwyBws2nAwfDRJO6u7tr2yjYHXgnFlCgE+ROkCO7u7sOhYCp5qGFBGo2m3r9+rVarZbXJSN7Ah6QZCaYUbfRuNnDxCpjxqKenpuVz/fu3XPEHju6grbBDz/8ULOzs/bo04kxvSQSCePcuMp4qGH3cT0xOeAKW1tbs0aX643eGH0qEii68kePHmlgYMDkHYcC+Z9I146Ojrxjio6YkGJ0smtra+rp6bEaBRJjc3NTAwMDViNwSNGwPHv2TOVyWfl83npvXu5araZCoeAgD/4vB/3BwYFhKIgjcHc642C+RTQa9WF7dXXlKEOcjjzLwag+MPfP282ZeuhqS6WS3+G9vT03KsEgIMKmGfe5RgSkS2+X8IGzTk9Pe9UR+OvV1U0+LXg47ky+xsfHDdHQFVM3wK0h3CCoKM6QWsBpSEQ5fP7qr/7qF9bMX9qhgqmB93CycnPOzs5upVfju2UNB9BAq9XS4eGhdY1oVkmkpzhw0jEOERE2Pz/vdj7oE8e3v7q66pMGyREQAvt5yuWyJicnLRRHrUAALmM6XSg3khec4BX8zBwWHCRHR0daXFw0iUH6VD6fNzMKM352dqZarWbDAKcmQn50quBXOG2azaaWl5f98DGyJZNJB7pQsCEI9vb2rAzgMGCE39vbUzab1d7enn70ox/dwqgp5jDUyJqkt4v1tre37ZbCBgzjy5iFLIyid+fOHU1NTbkjphtHa7m9ve0xm5ePnN1EIqF0Ou1rxUHDS8PD3263HYQM2QB8AJQCKcHfL5VKKhQK7lqWlpZu/R1JJkc4oDjEgqlo19fXKpVKJsD4M+DojcbNdgde8Ovraz179kyvX79WMpnU06dP7U4rFAo3Xc9nXV+5XLaKYmJiwnvNGE0/n1McbFo4/GD5Ca5ut292fIFDM2qDg0oyxyDp1pqhUCjkleDUC3Bv9Lhce/BSyFSIRuArlAxkJYPJYnrJZDLOm5VkIwrEJkHtwdg9RnWkgGxG4LlENQMMQ35H8BoFk+xQhnzlDvWf/umfTNQEw0MQE3NCwfRKN4uw+IHRaNTC83w+b6yNboeRgkV/+/v7Zvb5/owarKeF5Ya9Gx8fV6FQ0NramsdsBLsQEPwzMDCgTCajdDrtFde8xJAyyCWC8qNwOOwOk3Ge0w/bI1IUnDp0ltzUIBPLDWXs4X8HEC8UCmaqSSLCYwwGVSqVLL+p1WrefsnLRZcDmwuxhn24Wq06TYgd80hLgGsYxSnIpCmBlwclX3SLFHL+/O7urk0gyWRSS0tLmp6eViqV8osXiURUqVRUrVZ1cHDg4ohdFkK03b5ZqcFUQ5cUJP+Gh4e1sLBgCIlJZHh4WIuLix6zGe87nY4eP35sUwS4GqqWoA0UUgUdKWYMSbpz547y+bwWFhZ09+5dZTIZ67CxztIdQ7yenZ0Zy+awxpsPgcuzfnR0ZAx1ZWXF36fRaNza24S4HdcR14esBFQPFAjMIz09PSqVSs5JIFOA5DcS1ID8wOdZE48tGN027z/FmeKEEoUCzXvebrdd1JrNpn8u4UaYO/hzKEaoF1wHiiIYfDBli4kXjBxlSzqdtusSzD6YicrzR6MWCoX0p3/6p7+wZv7SDpWAjc3NTXU6HVdnfOvd3d3OP8V+SKQXWCISGsBqMJJg9BxjCGAvmCKnPuzn4OCgl/HxUoXDYS0sLFj0DL7T1dWlYrGonp4ezc3NKZ/PS3rbXdHtXl5eeush/97T0+NsSYTvFFk0c4RKMPJPT087cIGgDUTJMOmA60SgodNDlE3HizsEPHd4eNi4oCSP64VCwVmrYFO8qNg3r69vMiCRJmFNpcND7hLEcGFrkR2FwzfJ/pVKxYdgMFicEZksBaQ7dBp9fX1KpVJ69913lUgkLL3iXkiyS4vp5/Ly0lpTXjhwRhju/f19M+yS7B7r6+vz74QpApNDUG+JyH5ubs6kBQW53W7r008/9WcEI8SiXCwWlUgkND8/77BnVtlcXl46MOXTTz/VkydPHAVIFx+NRu22gjM4OjrS6empyatQKGRfPYYJAmYoAExuTBCE+gB5QAxRfCGq8OqTJsY1ICcBuAzzCd306uqqrq+vtbCw4FAe0rCwTgdxZmAn1sfDngNVkb2AeuHq6sorh/hMQCcQTEjKgCdwQwKhQThJbzFYuuTT01NLAZmKgVCAsuhQkT/yvKMY+srC/r/7u79Tu902OXF+fu7wAT7Y5eWl16QgxufF52QAb0JAzt5tNGhggoxo3EA6D8Z42N/gTqTP548CyvO9pBtCBO0kMXe4H9LptLvTVCqlyclJs5GI9HlIwcoA4iG7MpmMvdBgvMExEWKH8ZAbKMkicrqSer3uIp/P5y3/KpVKCofD1g5yvRj36eRHR0ddHLjOEAx07+gIG42GXrx44RUjpD0FOzX0lKg6wMJYE0K8IuNjOBzW0tKSC0ytVlMmk9GDBw/0G7/xG8rlchofH7eFk4OH1TLkFuDbJq8U0wVdOwc0sABdCpg7kXVjY2MWolNAGJUheHgGMY+gjGg0Gma5EZqjvaUZmJycNP6M/ZMVN5AzZ2dnKhaL5gokWSoE1krOwvr6uiclCM2TkxPbUycnJ62lHBsbs4SR54yizd8FDw2GfGD3pjE6Pj42SUgXjcieDhDZWFB6hvQNjS0cBWM7+8qCmnIaI/5JJpOGHNi1NjQ0pDt37mh6etoW2OC1CKo5aF6A4pjiUDRwIGKrrdVqluGhUoJoYqsBEA0NThAqoUP+i7/4iy9fUL/zne8onU7r6urKpydYHKMDIz0EEQ8jBAsiaLRinEjozpDiQEBAbGWzWUsfJFlFwLjAic0DPDw87NEV/IPiQZeEZjIoHWJ04YWAjII9pqNDxE3IA7hKsVi0dhRM9PLyZnPj7OysixHEDL78YHxhEAxfX183bsSJur29bfgDjA95SKvVckI5oz9SKhQU2WzW6etABARh8ICn02nvnwfmAR8OLuiD7eb7TE1Nqaenx6TG5eWlfd7cC7aWBkc1DkCstdFo1Hvgj46OfOjgBOJecF8Z3SGpRkZGlEgkbkFJkqwU4WfSaSPlQd4GjEOBQtDd399v2IBFh9iLkd2VSiWLyXO5nKcEVqBgcYSkRG1SqVRUKpU8wWHDhBRh5Tn23N3dXQ0PD5sUBYaiuAWjEyGTuHdkY3Bg9/b2+tDigAQSOD09taabCYVDm2kKpQ0EaLVaNcQB5EMh4/BA50m3jxifbpNnaHZ2Vg8ePHDUJ9AiDRwwHOoIIBrpraIC/JbnoKury4Qd8Yo4Gaemptz98zmAGEnZorHr6blZ4PmXf/mXv7Bm/tKRnwsSHM9h+IJsMX78TCajq6sru4LoJliDAq53cXHhQs1oT9eLGB+HBGxoMpm0do50H14ctKTj4+PKZrNqtVrGnWDsh4aGvMaCVCC0hcSEEVwyPT2t4eHhW8lYYEGcaGRQ4ovmGrG1kxMRsXkqldLW1pYqlYpxTHScpPZgTmAUxyNON8XNxc1z584dwyF0wUHlBKw7GDUdycXFhQ8BXCx0COPj41pYWLiFl29tbRnjJQiYLoxOZHp6Wn19fdZhQk6iGSaucOaztSsXFxeKxWIOeyEeErKTjgAslrGdZ29xcVGbn616YZwHb7u6utL29rYzN/HYQ2RIN1g/mmlkO2T6YkVNp9MuuKurq9Z44sz6vCMKTSmQzOZnebGlUsmuM64BeRbLy8uqVCq6f/++4vG49dZ7e3vGnovFoqcBoBYCZzisLy8vvVqa340GhWcbITuFYW9vzx3fxMSEMXnpRh4FhMN7ixGAkRd5V9C5NjMz43g/agjaX1h9mq2DgwNtbm7aVRmLxYxnIo3qdDrGLVEaHR4eanV11UQo2HNQTkit4sCh0+T+SzL0QNHmueO6807TTEjylP1FX7+0Q/3rv/5rTUxMeCwDm+LicZLDnMPYHx8fWz+JQ4IwAyxcYHt0pNx0TiG+J4n4fKDg6Qo7jlSDi8RLjRYNYwL5i5AL6DWBIcAM+R3pAnhwg06saDTqER/skm6IBx6N5eXlpQkgMgGwj9IREvpBJ0U3APlBtCCkSDwev0W0sVSNm315een9T9LNpsxXr15ZYgNxNjQ0pM3NTa2vr1tcvbe3Z0kJcij2iBF4PDMzo8XFRRMzwb1b6XRaPT09Wl9ft4kCET1QDJ0qdkYKPZrTy8ubxXQI/8FaYdSRy7BOhzjEarV6S8bGPQHbDmYLQEAWi0WFQiF3o5VKxXIgYBgWHZbLZe3u7rrbJm6QQ6JarXpTLzpZcHLIW7oxshuChAzBKMFDHBdhMCQIWWEul/NIjHA9qCBAh8o1Y7rBPcaabIofUxGNEvplbOXn5+cOleF7Y/ahI4SjgPSRdEvxA3SHfptJc2FhQYuLi4ZUrq/f5pGiveVdlmRVDpMe7ifuDYcHJgRiBrmOmHxQi0AUU9/6+/vNoQTzdvv7+/XHf/zHX76gfuc739HU1JSmp6dvBWnwC3M6wKxRTGEsOQEAoEkMCnq0KQRcIE4RcFO+wNcoJrFYzC8jmFhQlkU+JkWxq6vLe9MZdehe9vb2zOjzYCKapkPhQaaoMVLhpSa4gm4BRYAky4vArpBmHR8f29bJiV2tVrW+vu4xkQBg2O+BgQErFOhegRl40cD5AObX1tacssTOd0Tj/f39FkWDAzMpUKjY/0NBAlMOxqZ1Oh1VKhU/8KzjyOVy1v6lUinNzs4aT8csIcmMPbhXMDuCkQ4ilEOaF5lCRLdLIeK/BX3xEIXd3d3ecQXxEwy7ef78uer1urXJEBkQdBwKXCdCXhhh6/W6izn3BLXGycmJA7A5bHh/MIPQnXFoMx1NTk6aQOnt7XUhA4bDwguWSdfFAcJ7jKaUzxAckxlr6XJZJ4NdmncTpQWyQKaH4HvPpJRKpYzjM+Xyzsbjcc3Ozho+gluBAGcqpCjDHwAhgP1yjcjYILMZEpmkK6ZcJl8URujdmR7AmWkMOWCur6+/mvWUQkYncXh46BvI/86DJOlW4RkaGjIxQOcH29xut306Ulwl+cYDM3DjKKYUN1JhAL1h/ev1utnEUChkmypraMm7PD09dZrQ8fGxMUiKEtITUq56enpcqCEQYMxhlYE2EHXzoHU6HXe0kF1o/oKgP78XHnA82CTwR6NRb0pFfA6mivURZjidTtuquLGxod3dXe3u7mpiYsIsJkJszBq9vb2an593sj22v0qlosHBQeVyOY/fyE4gGymGmCcYhbEuI2/pdDpWK+B4eu+995TJZFQulx0CjfU2mUwa26pUKspms8bPOJTo0rGXxmIxd3O40IKaQmL3UqmUHjx4YMKGyLfLy0sTZ4VCwYfx7OysfvCDH6irq0vf/OY31dPTo5///OduJrBSo7aYmZkxpxAM8UFBMTw87MkC7SYB7ZglYLnPzs6sn8Trj0+e5xE4BMgsn8/beUiBSaVS6urqci4o8Mrg4KA15sA51WrV3So5EpIc7EITw7RDx4z2FRMIRZbfkc9wdHSk6elp58MytcHLkKoVtIwDFXCNaZg4cIEWgarAkoPXDaju8vLSgTtcC/TyNI5ACGC4l5eXhiC+6OtXFlRGw+npaZ2dnZlwoBBychBEQRHkRCAhCKzv84w94zxdHdsfeREYOZFvsEkxkUhoZmZGw8PDlmpBdDEyoQstl8vGQw4ODjz6c5JzkWEpSZ8KZiJS7BlzOAlh77GKttttS8lIGUKqwTgUDIABy93Y2NDZ2ZneffddDQ8Pu3tGWUH4RSqVUrFYVKlUstmCkNypqSnj2BBCyLtarZvd8YlEwiPU8PCwx2rwufn5ecViMW1vb3tLQBAnZQpAB0tXyj3q7u7W5uamPv30U5XLZQ0PD+sb3/iGrY1YirGy/vd//7cymYw7zEwm404SJjYcvtmY+8477ygej7u7g5Sj20APjGYRhUatVjMpAsHVaDS0vLysSCSiVCrlHAmw36GhIcXjcROsfX193tSaTqd1fn6upaUlRSIRLyLs7u52kAmJbMAHZJgmk0kHZIPbIjBnQsA0QAHGaDEwMOAwk8nJSZsgQqGQM375PmydQMnCsw8mSsEGsiHxicloZmbGBzXjP85BGiycVWCOOCKlt/u2yL3Y2dnR5eWlr19vb6/F+pDWwB40WFjXLy4uboXdSG+T/NGr0t1SNOv1uo0QEIME9yC54+DH7cbUjKIEeALsGYcV0+9XKqhIDEhqglXjQ3PyIKKFdKLItdttyxIgi4LrjulkeOEgZzhZgl0qRBVtNysVDg8P3Q2BryBl2djYcKd8fX1t+2NXV5fu3r3r78fDg0OEkZL1CehBKTDIOxBcp1IppVIpC8T5/Rj9yKyE/MGBAd4XzCdghOHfx8bGtL29LUnWkNIZpVIppdPpW04Ziuj19bUXG/LfGZfpSrCGcjgUCgUn6BeLRV1dXbmLbjabVjX09fX5dwabomhlMplbQdnr6+tKpVJKJBJOSpqfn7dOeG9vz/KjWq3mEZTOs91uK5VKaX19XZubmxbld3d3O9yjXq9rcnLSTjTGXfBdZDVMMOfn52aVJTnkh46YLhw9LxANKWRMHmioE4mEXrx44XG1WCz6ea3VakokEu6w6dAgINPptCEX/sFSOz4+rtnZWUUiETPcqAuAdtBf9vf3W6UCt0DnGo/HbaMlhpDNuc1m00lYJHiNj48rnU6rUCgYz0ep0t/fr76+PsNQvDNwEsE0J2A0ftfFxUWNjo7eMh4AAQRJ72C0HwU0KF2C+KRRk+Q6BHbN8wzswc+CjIRkQ6s7MTFhmR7PBA4vah0wyBd9/VIM9W/+5m/M7mHX47SmoIJ3wT7zECIgDi6ZA/8BjCcCDUFwPp931wHGAiaFhgwZBisTwHGCeAde5Ww2a1mIJOMxkCD5fF6pVMq6UjR2dMWcvJz8fX19xkMl3Rp30Z9i/QS2QMjMC1Gr1VQul62XGxoaurX5EgiFEQU5yMDAgEdRGN+gDAXcC/KhVqtpd3fXWs9MJqOlpaVbKVjxeNyyLVJ3CEBmQR1+bxbvURCw/XGNUqmUdxDR7SAOR3LDKAX+l0gk3MkwHjIVcPgE4QOKDQoOErHogiAmcECRuB+0MRKyw+aFUChkqIMXDwwQXBs/Ok4kJg2mFw71oKSQRoMCyCqY2dlZ9fb2an193UlVaJnBWCErgalQzJDmTyI+kXwoOWKxmKclyGPkZvzu2D3pOi8u3i66lGR7LPIqZESQSExnw8PDvseoA5BDBmVSTJrj4+O6c+eOHj16pEwmo9nZWUsfCbEBx4WswnJL8A0NDq41PitTExMpMYMnJyfOZ+DvcW/IiyAXgjAWSYYp6FapN4z+4+Pj+qM/+qNfWDN/ZYcK+D44OGidVtDrC6Y5PDx8S9PJaMEYgcicXxwBPTcBIT92TDpYSbYQAhHw97GB0ukGg5rpUiYmJtRoNFQoFKyppMitr6+7pQf/CxZL0oh4WXGcwECCBXOyQVbwwCKOpnOHkY/FYnbjlEole9/JbsRZRBwcKVZ033RvqCzAf05OTrSxseGfD7GRSqWMnUKaYTWs1+va3t7W1dWVZmZmjBWdn58rl8splUp5jAKvAifltO7v79f6+rozUYvFolZXVy0XYyc9OBwYGwXy+vpaa2trFm8jwaPYIPhGmsV9zmQyLpw8N9yb3t5e5fN55fN5kxsrKyvGvKPRm/3vZNSirCC0hYLe19fnjNhSqXRrdTerlilONB10NvV63WEoQCSEbTP1sXLk6urKEAVBQslk0kVkYmJCU1NTZqyj0ajHZd6jYOI8jD3PNPcQglDSrY600Wi4EWL6CJKlTIDgkZBXBMYMDAxYG0yhAytmod69e/d07949H/jAKWTtjo+P2/2IEoL4PvBd8HHqBng6bkTsorzjwBvwKOD+Y2Njvr7wJ5gXkFHSqNEVI69D0fKVCqoka81wZIA7AKYPDAwokUg4iIM/i4MJsgXJDMwyUilWkdBtIaoOjhG08bwIFxcXxgAhg9BGkmEKiZTL5Szkx64Yi8W0u7ur09NTLS4u+iTn5GXEqNVqhgXC4bB1rrDzEEzpdNqnXzD9Bv3owMCA2VhJZo25DnTB6H2RmY2MjNzSygUtsIjaIW0ikYjZ8fHxcZMh4NXgtqR9QSTE43GrAMBJ+/v7rREk/aparaqrq8vEEFMEJzqwBDAGSwwhg7q6uhxKPTo6qs3NTS0vL7uYXV1deTEbAvtI5GaLZzBEeeazNcRMJdVq1aQE0jZ+DzoMChUwDM8KWDUjJzpiSX6Jkchhzezt7dXMzIw9/RRSnjFwSXJlCbdBn8r9RYQOfk2hxIa6vr5uHLdarer+/fuanJx0UT04OPBGCaSIV1dXDpzBDn5ycuKRHbIGrTK5BuFwWLVazUQxZF/QSUQEHgQauDPQG8ULmGVyclK5XE7379/Xw4cPvfH16dOn+td//VcvgozH4+YVeC8gnTkgSPOnG6YAol2+vr52sYT553fhXtK5MynTjEnygUxHTcEOZjHTOQe72a9UUPf39/X06VNlMhmFw2HjbrxIXV1dLgxccABmWMdgaAIsMkA+4yYnMXuEwATRVwI07+/v24bJnhewSkYIMEtkH+SEIvXCF84YRCuPzIOHLahnpLPo7e21f5vOly6em01n0mrdhGIHgxsYUbe2trS5uekM2GDeZnd3t0cubIGIkyGhSPUfHR018wsRQHGZmpqyTGZra0u1Ws3uNYKZs9msHj58qP39fV1eXioWi+nq6kpbW1vWb4KbT05OOvqMhzQWiymZTBoSIsVqcnLSThWgElLU0Y+Ctc3MzNxaEX5wcODV0nTHh4eHtoKyDhp5jCTj2sQNJpNJu3w45FGCTE1NmUBhbMXo0Wq1nAfLxBWNRpXL5fx5kIahEgBPxx4bCoU0MzPjQtVsNi0dQmrHfQPvRaaXz+d1fHys3d1dtVotzc/Pa3d3Vz/+8Y9dcKempkzEXF1d3QqH5tDj0E0kEk5nC+px0VbS2fMMIW1MJpN+HqPRqN89II92u+1rSuYpXTBuv2QyqQ8++EAPHz50o4HBYXh4+Fa4OLUEuIFnnuvGAYffH9sr+D8TDhAP7qggRMGYT5FFukg9Yy02hzU/E7kkDcqvVVAlaWVlxaty6ZAoEIzbjD7d3d0+NXnYOA3AOMA+IKm4oTg7GGm4sFj9wK0QVPf29tpHvL29bTkKH7rVatmZFIvF9Mknn+jFixcOmdje3tbq6qrOzs6Uy+W8ZgFyKbhuulgsuhjxsyE82u22crmcLi4uHMgdFCNDGlDMnz17Zt85hwte5/HxcVUqFa2trVleJcn7ucB8GUk2NzcVidzsZkJiA6ny5s0bY3mEqDDCcH8ohGyrZbEhzjLIrFQqdYv1Jvqt1WqZAa3X63r+/Lltoojog1m5HMwoEXZ2dgxRgPXNzs56dAOvDx5idGlzc3P+WaFQSK9evbJzCkgCvIwkr6dPn+rVq1d+hk5PT+0p5yW7vr52F7q5uenPjGrl8vJSmUzGJCD38ODgwHAQ5Ofk5KR2dna0vb2twcFBC8hx6TFthcNhbW9v+5B6+PChyuWyJzDIPrBPGgiKODiudEPiBCV9JKoxthPJiE7z5OTEihPgN3BQYutw2oGBY/ZJpVLWRFNE+/v7NT8/r/fff1937971vX/27Jn+7d/+TZVKxWw/zweWcmR8mGFOTk6Uy+U8huPPD0qmUAoFoyop2jisKOgcvkQiUpCDq5DQi1OkwfCB237tgkr7zGmCc4JxiqKBaPvz1i1OTMgX8FdWGvf395udxiHV29ur4+NjHR8fmywC7Ocm9Pf3K5lMWh2AjhFchC6UUJW7d+8qn8/r1atXKpVKuri42S65tbWlgYGbJV0jIyO3OkFafL4/2jiyRePxuJ49e2ZvO15rXkw+d61Wc9fGQ42VMBaLOTatXC47tg2DQVdXlwqFgp1dnOaMuOxIR9lQr9fNHiNJY9R8+PChU6FI2TgKraIAACAASURBVFpbW7OsJZfL+YGUblxGAwMDVjuAJRF9mM1mJd24sYrFojY2Njx6AQ+USiXfMzB4Erq6u7t1cnLih5X7DpsNWcdz2Ol0lEgkNDs764LAZAAshJzu/Pzc2GkikdDS0pKhGooZYy5KCF44cNeenh69fv1aW1tb6u7uNi6NdKi/v1+lUkknJycKh2+WHrZab3cdZTIZjY+Pq1arOYSGBgKXTzQatRKCETOYMZFIJNTV1aWjoyPFYrFbyhWE6nj1gQMYgYnXI/MX5USxWHQHT9ODxFCSySykZ+QLhMNhTU1NWabGdIrkbnR0VPfu3dNv/dZvuTOUpJ2dHX33u991nOL5+c1WX4hNnFR0vizZazabVkRgHqJrBAIMWrPBOFE3MIHyveBEaOBg9SGwmURR5yBF6+rqsqLl1y6odBIPHjzw2AxWBHHFOAS73W63ffMkudDBvkNmDQwMaHt722MvIwUjGVgs2A4kBS9mV1eX7X4TExP26gNEs6sIGVcoFNKdO3cUjUYtGzk4ONDLly+VTqd9GHBiRaNRB46EQiFrXYvFoorFohcSAikwamB/I18ROybynjdv3mh9fV3hcFj5fN6YFRMADzwH1NzcnFljrL6oCrAUNptNzc3NGX+enJz0f0MDiWSGB4rPRpIYXQhYKwTE8vKyIxlxv+zs7KharXpVM4RNrVbT2tqaRkZGlE6n9e6775psgMkHu8NuLN1gZ2DbiOUJD4aw4pnCLMAhQkgKvv6FhQVbCvksEFI4bnAUPX782EQm7DWHP880hB2jLQcU8Y9I/4BsQqGQjo+Ptbm56UAXoCrUFhxiEJNgm8imsHFC/nR1dVnWhfKjt7fXYdRIg5gc1tfXVSqV/MwxmUAKAyOhGadj5zCFYed7E0IErMX9w6qZTqf16NEjffjhhz4EJen169f693//d+82CxKtXDvUPKygoZvm0MT2TCAQ6hRCt4H9UIZATkLOSTJpTRFGKYTqhi45SAL39va62+eQ+bULKpjRwMCAstms9ZLIasDOiNii4KFjxN4Zjd6sJ242m8pms8a3IpGIWbe+vj4bCu7cuXOLtCBeLYhzIiUaHR11YAtdDUUCHPHi4mZxGyczhXdtbc0SEelmvGSEJ1np/Pzcuj4CYTgpcbQErxXSMWLK+vv7vYgNPIr0pr6+Pj1//lzh8E2+K6PI4uKiCxp5AhBbnU7HRKD0NogC1QSjEbAKOBSSpZOTE62urqpUKtkZNj097ei6ZrPpbhlXDtZMgmvA13AlHR4eqlgsOgPz4uJCm5ubzg0YHx+/paTA5RYM2RkZGfEBgxUY+Ve73XYc49XVlcdncN9IJKL33nvPnvp4PO77SNwjXQ0GFQpc0AWGBTIcDmtlZUVbW1tuGuiQhoeHfd/5b2ghOUQzmYwPC2RRKEGYzNBykpEAx1AsFtXV1eWu+/j42AYWRmE6f7ou2GiIJ7SvmFXoetFs83vgMpJkyIlnn8aCAwT4DgiBjnR0dFSzs7P6xje+4SmSwKDvfe97ev36tQ+0VqvljAs6aaR42Is5cFAJBVO7cARC7pEHEDQcAUXFYjGdnp5a/41oH2cXUAfkFsWUzpX7yZ//tVl+vgqFgjY2NpRMJnV8fKzt7W0L0GnFeVggm9Bv8XCh0ZTkBV50X7yMu7u7ds2cnJx4sdra2pqLDFs1sbryvVktws1AfhUKhTwWs4KYYJJms3kLY0EbC3wRlHSBsSH74UWnGPB/wV8hlygKQ0ND2t7e1u7urgYHB/1gg0chaua0R8MJeN7X16dMJuPrjfkBwJ0HmYNOkgvh7u6uyuWyOxsgGbA8MN5gtsHe3p4F+Hj6wbboeJF2HR4eqlwuKx6P6+HDh/rWt77le3pwcGDMK5lMWkg+NzenSCSilZUVbW5uKh6Pa35+3vIZuvwgibWwsOBO/fr6WtVqVfl8XisrK84ChdTE/cYzx2EMDs+BRCgKIdfghzC+wFqhUEi1Ws1CcIp5OBxWoVDwS4f6BVKWkZ7pClINBQhkriR3fvw8RnWmIGAprMioQJgAOFjS6bTfH5QzaF+DEAxONQ5hOmIK99XVldn2np4eLSwsSLqBBDAsTE1N6b333tM3v/nNW8V0dXVV//Iv/6JSqXQrRIm8XaRN1AHwULiPeDxuByIhLyhr6vW6ksmknxFgPVQZTDXkSeBgQ6VE5CIFmAOJ3ynY6WL+4T5+0devFPYHv3BJgOGAf9A2c7NptSVZWsNJQvAJej0E8XQcMHvJZFKZTMatPmN2LBYzyQAWy1jPqMUYgvQhGKwLtsnpi0Qnl8s5l5OOqVwu6/DwUKOjo15NAbYSNDkgbzo8PHSh5M9jraU72tjY0NXVldLptO7evWt9b29vr+7cuaPR0VFlMhlFo1F31AD+OGgikYgLI2MShRnRPnIUnD88SHTX/Du6TiQ24Hk8UPj3waK5fzz0dLHVatUWWbJg+dwYKSqVikfJvr4+HxaI+pHdoCGEnIjFYnr//fd17949zXy2pRYoYmBgwCvKKbJ8ZqaSYFANnR3XCAiCfAReLkwMjUZDm5+tMUEqSBrS+Pi4icJo9GbNMfIwnFCw67iKsCzTBXGoobqAIKJ4obnkAIX1psgFx3FGUw72k5MT48UURK4d0xUhMkBZeOTRipJQxn0Jh8N2TUpSNpvV4uKiPvzwQ33rW99yMW232/r444/1j//4jyoUCobNKIpBkwSQIJ36+Pi432t4DbS1HPzBtTQoinhmmCJ7e9/ml6JqwJaNwYYYUJoS+Jaenh43W1hReeYmJib0B3/wB7+wZn6pDlWStra2NDc356La19fnUxdMD2yMl5HlWdj0KJzValXn5+e6f/++Oz1wIggCgPxms+n1JwitefiRZzGG4oWnzUcbS/hFrVYz6XV4eGinEsEVi4uLHtNqtZqKxaJGR0c1MzPjlCscLWzuJLy21Wrp5cuXGh8f171799TV1WUNJScyOC6bOvld6FJevnxpoqFWqykajer+/fsOo+AFCrpiTk9PdXp6qlQqZTE8UhBwYE5hNnj29NysemGMOzg40M7OjsbHx50oxYtOsSM2j8xWJo5arabnz5+7s6AbZdUJcAa2wXK5rN///d8340/+AbGBFAbUIGCOrVZLL1688AvEYQj+Tsg1dlxsnugPkdPAnqN9JHAD99HCwoKOj4+1vr6ucrmsSCSiqakpra+vq9FoGDdFL0rwM/KhsbExlctlvyPoQQmWZnULZAwidjA+mgCYZSYlSbeeI2ycGBEQvqP5RvYEqTkxMWHlxdHRkUdpCE0KCpKpvr4+FYtFXV9fW74Iy41F9Xd/93f1zjvvmHy6uLjQ48eP9R//8R9+XiGvKKaSHDJO1CHySuSYdPqYeCBx6VIpdnx/DmYO1cPDQ9cheB94EPB4nnP+/2AwPT+/3W77PeYA/6KvL11Q2dj5zjvveO8SonEkQJw8dCHX19dmPRmf2YZZrVbdBaIa4DQAXA9aTwnCXV5eNn5Dmw7GC16ENo/RiGAO/N/tdlubn0Wo4aC6vr72zwDofvz4sWKxmDuY4+NjZyiiY724uFAqlVK1WnVRgQnHqbW7u6t4PO7d7YQWn57e7HMnEhHNZzgc1gcffGCrXF9fn/3XjObsqAqy4/zMZrOp+/fv37IA1+t1s9GoBVBqoIPd2dm5pcNDEQDJgIOmp+cmj/T58+fq6upyWj3MPsTY7OystYVHR0fa2trS9fXNtk8wv3A47OAMJh2eEz5vqVQytoUOFqwcrGx8fNws+dTUlPF41BBoMNPptLa2tvTmzRvnnBKUvLCwoFKppI2NDS0vL/vnIoHD6LK3t6d33nnH4UEcckwVyWRSAwMDngaI5SO8nAkDDJ5cVUluGoLQFoUC6EiS36OurpvdUufn505SozgMDg5alkVQDwoROjRW7fB9VlZWrJqAbGQkLpfLDpRJpVLK5XIupsfHx/rud7+rx48f+zlCoE9mbHd3t3kMcGcwVUZ2JHXYunkeYPYxpaCdDWY0r62t+ZlhSsT4EA6HHQGKYoYcCt4FCjzkOiR5MHHqf6ygttttPXv2TAsLC04lYhygWwqGxzLugM8w4jDi4eBgLEOWwE2gQNCx8jP4QlwNeI2sC187JAf6TYrg1taWzs7OLLIHP+Um8ntid+SGU+xRDFAAETifnp4qm80qlUo5Pg+vMgn5KBFevXplfLVWq5lAePTokV1kjOu8aITVsNIZrPD09NSxa7DjpOjglIFhZj0HqgsY5evra+XzeY+cfE+uDw8pqT4UV0bGnp4e5XI5r+yYnp7W/v6+1tbWfHDwvYrFov7zP/9TT5488XqZ2dlZLS4uqt1u6/nz54aYKpWKPvnkE2PxyFeAH3Z3d13MPvroI42MjFgPvL+/bwkXzxfRc8fHx9rZ2VGn0/FUArZIZOHa2po6nY5JWEwBzWbTf5bvTYcTJNeSyaSzGyBUONT43Fi1I5GID246X/BF/j7dG90ZPnxkRmSyBl1zpE6hzkBhQUNEQxI80Hnn0JHTcQOZRSIRvf/++1paWjKTvra2pqdPn2p5edmuN6ANitLY2JhDVtgIwMGOEoZmBOcgRTj4ngeDu+EeuL7B6QqTBe8KxC4HPPcaHTWdP/UJOIHrhmnhi76+FIbKF/pPtltSvD4fA4bTKfjScaMlGVgHfAejCzowGJkYLzgNYSe3t7ddsHFTYcPErcHq24ODA62vr7sowOq9fv3ae3yKxaJj55aWlpTNZu3Z5jRk9MK7T2dHslWz2VS1WtXm5qZvHEROf3+/pqamjCcjyMamidUQ54wkd/nkk+JTn5iYUKvV0uLiojFcxuJGo6FyueyXDnhkYOBmoVp3901m59zcnK8D4RIcShcXFxZf82IyGvb392t3d1cvX768JfWZmZlRq9VSsVj0bp69vT1tbGyoWCw6SAVh+MuXL9XVdbOxYWdnx6tZIPHoUPf29oxPshoG6R0sNM8Fub3IqMBIx8fHdXZ2ppWVFWNosP2dTsdbWrFd8v0qlcotDA35VSKR8DVnMgq+kNzXYIYrXZH0Nv9Xers9gGBunk+suMBn2Gl5N0jCAnKr1+tWq9AcNJtN60fRWO7v73tVOMUYAwCrv+kocT6Co6OqCTYVP/vZz/Rf//VfNiZABKJqwF0HXIBOGIglHo/bQIM9nSYmmO5E0QUCAiMmaAV7MvwJ6hTUP3t7e4ZiqGXkUqBVpY5RUyR56lhcXNTk5KS+/e1v/8La+KU7VOkte0dqzMbGhoMk6B6C2YuoAMihxL3CGhNOAU6hq6srvXjxwt0ENy+RSKjdbjvCjuV5CJcpOrhWMpmMRkdHtbGx4f05kGWMQuCMgNv4s/P5vH7v937PIuf5+Xk7rAgUYe8Tf6ZUKmlgYEBbW1vWxPFABJ0ljx8/tv62UCj4AKhUKnaHMa4g4wAbpnDyEPX39+vVq1cmpBBM85IODt6s3qZo4EUHY2a058CLx+MuHhMTE2bMMUqkUikXpbOzM3fiMPvDw8MqlUrq7e3V8vKyX/jLy5tlfYi0OfAmJyclyR5qtKGSbpkV6MogKih8HOzAFnRX3d3d/syQc4jvscF2Oh0HJnPIcE0l2bNPsYdkogDQMFB0KZZwCNPT0z6EsTSiEx4dHdX6+rp/JiRNOBy2tZqsz2w2q729PQfMoAqh42u329rZ2bHIHyIH6REBPei26dq4hgS4B7F2ujVJNjvwLKAR7+3tdePAc04hojGiUJEzUKvVDJ3FYjGvrAH/zGQytxbvXV5eOnMXeSGxiUjO+N3YI0aDBiSA3ha5FO8K91iSIQtwbbpzOITJyUkT6Gtra19YG79SQZVuRPpbW1se/cFQ8fUiEg46IDgVSKcBtwiuqCW1qlarmXEkdGR/f9+yi6AFFAgAZw2eb0D+o6Mjp6GPj49bg9npdKx9BVtjNAFD4YKzEfSHP/yhNjc33YW8fPlSjUbDLxiANqHNOzs72t3ddVhFsVi0hRMMkhcTQoboNYBymHNwnL6+PutiI5GI9vb2/KBvb2/7JRscHPR4yUs4MjKie/fu6eDgQPv7+7cyFjip6bRYjEfHGIvFVKvVvHCur6/P7i4KAMWZ4j4/P+8YwmCoBdNJEIPEOUcARblc9iGE4Pr4+Ni6UWLxcPp0Oh1Lr9AOYyFGacDhxAvPYrhyueyAGOntMjYOWwo+oyUdH9cfKAquIBaL+RlhVTPJXeT3gqeSGUrxD0oBK5WK4y1hq4GygGE40Cl+dIHX19futiiSFP2g1XNoaEjFYlHr6+vGCCF3enp6/H4Gbd/kGnANJBnqCG5JoEmhRuA02/wsIBxpHAoQ2HawdYhKtMzwJryjQZUAXf7JyYlhPgjXSqWi7e1tm4XApKkx4PY8bxRkuuZUKuVm49WrV//zBVW6sRreu3dPyWTSgnvYarRgiM5ZdCfJLx7jLoHFEB+MxZAYdI9cjFQqZdyPIIxgKAt47NbWlvV7wZENrJMHimg+WGFJxld/+7d/2xjM2NiYvv3tb2ttbU3Ly8t69eqVs0u3trYskEd6xIOFRIybhBccITN+4cHBQZXLZQvXwc4GBgZ8mIDncjIDyF9eXtrVQvcXjUZVLBZNfsTj8VsvGqMVqVzgdMAPZCzQcRWLRW1vb/taDQwMuNOp1+uGLLLZrD755BNJb9f6ZrNZFwNUGRsbG5KkBw8eeETjWtDx0BmjWQYjpNMguwEJEcE6jOtktOKIoggAaaC2QLLUbDatt2Wsn5iYsPwK1QS7qcDbIVMymcytqavZbNqtBjzGErzJyUmFw2FDFPw+yWTSyhMyHui0iOMLmkx4LoIaZgKL+HlAVtVqVdlsVp1OxxMAq1JYGcSfBYZDjlSv32xYzWQy+vrXv67BwUFLyuiEeeYoxkySEE88J/l83rpUuBWswhgGmLTQUQfZ/SD5zUQcCoWcd8AozxTMIYpWNwg9QgDz7tBEAVdQQ5LJpEqlkrHzX/T1lTBUvhCP0wFwA2AukXVQkACveQixTFIAIK4QcXOROZ25sWNjY8pms/bVU8iDBBFJPpHIzWKzRCLhzpAkI8Ybuju6MUYO/NCEpUhyYcrn8w4LRo4B9sReI1J8ECHT/cCoc426urqUz+ethgDzAWOFxUeQTpECq8KnzotAbiT4GZ1VcEngyMiIQy0SiYRj4YLRgGDKpVJJBwcH2tracqc6MTGhTCZj/A25W71edy4leOLAwIC+/vWvW1w9MTFhsm9nZ8eBJS9evNDl5aUmJiY0OTnp7pokMjA1WHc6OWR4ZC5w2KBbZrsua485qIrFogYHB23wYBsopCOfh+uNHpiXmoO51WpZpgW8AZnE+mrcfuCzZCxwaPH7UtSZFiBpkSuxYwxeAQce8h46XeADcM6Liws/JzD86MfZhEBXj7aV608BQmaXz+e1sLCgmZkZSXLgeFAzGgzgCaZbURS7urrcYdJEQSrxTpCviv0UTTsHfFCDy98hbhGIgGcevJwOGNVDEJ/lmvGuBTXJqE++973vaX9/X19UNn+tgirJe5FmZ2cNeKMnY+zm33kJsDYioIfhI3SFriMY4EGLzsNEsQEOYDQbGhq6tdUUdQCSB7DNoKlgcnJSs7Oz7nwgYU5PTz3e5nK5WykzsLjYZ9PptHEZCj0kDt0K7Cu4FR01RRQpDVgn5gCKPDAInxvwHcyQTonFZ0ARpCqh06QLhG0FFnn58qVOTk40Pz9v33K1WnVQMj8jnU4rl8tpaWlJCwsLdtUcHx9rdXVVFxcXdqXQQQwMDCidTls3WalULCt78+aNtre3TfKBE+JMazQaWlxc1NLSksbGxnzwtNttkxuMe0AJPC9095988omq1aqhKrA2Ajfa7badTxBWFCkORrq6tbU1HR4eqre318WVf+jK0EUSznF2duZDp1QqmSCiCKCRhhVHlnZ4eGhzBEvnKPqw7uDHOIDwxhP2QbNBSDtLMzc2Nnw4FQoFM/vgh0w5wedtZGREiUTCzyafGUkXhxlNEw4z+BQ8/8Gg96AtlkJPmD3wE5MjhZTrhxMNtx6NGDvdrq+vTVwx7fCe4uokExemn3R/6caGm8/ntbi4qEKhoCdPnqjT6fzvFVRGjLt371qiQ2o2pxW4JMQCHyrouqFTRJfGqgWKJX8mnU4745OoruCDTeFEa8kN4AFjRObkZTzAplgul81cMoZiQMD/z0MjyVbQyclJY31AG8FVIDxMXV1d3oUeCt2sQDk5ObE8iuJL0g36QE5xuo5gcpAkd1TxeNzrkcF1wY8B24kjA8fj4Z2YmHB3DFvKA5XL5Xzf8vm8U3iIIKTIYTEkvpFDamdnx4RhsVi8tV2SLgHpHV0VhYFnCLZ7Y2PDwTyQmIx1aDAhm9C9Bg8rFAFgsUwQ+Mjxa5NgxiFGZ0UiFKw6B/fW1pYk+RkBS0VjHQxzwbzCVMBUBQEIbg0mySQWDN1h5QlwCGug8aDTiXc6nVsLIxuNhg4ODgyhAVdwsCCngoVvtVpWyqTTaaVSKZ2fnzsXITgpIHEClkLmBXkXDI7hUGTvPdARRRZuAxwT+C/o9iJWEyVREJemwdvZ2XEzwrMBDAQExcEGpkqzSO6qJP34xz/2NftfK6iSjJFMTU0Zr6NtDvqogyQTFzuoM8MTz3iBPIuTZGRkRFNTU36ZGWmDJBLMLZmkQAgU7KB9FvyS8QJhMRZM2EoY4rW1NTUaDWNsfDHG5fN5S0VYaIgDCK0upz4Wuenpacuu8vm8IpGIx67t7W1vA2CbKQJlOg2sm0h3IpGIyT5i//iCEEKPB3aHoyQUCjmgu91ue20NODiQA3rf/f19FYtF349YLKZsNmvCBxCf4BVWtLx588akCCJv9nG1WjcZluiDWS/MSI6gnxcSi22wi6LL4oXlenKA0MFC1GBDxjxxfX3tURrTCWRMNBr1yptgXgLEVX9/vxKJhD8vdlGIOTSxTCwQN4QuM55zQF9fX5v4k27GcIo5zz9YIMYDtjwQZoJDDZb9yZMnWltb83NA5gSGE3BZcmYplLw3hEITccjozDRBB4nrq92+2VQAHhskjpCOQcryXABN7e/ve1rg+e7ru4laxHgTDH2BqGWV+8nJiQqFgp91SZZQARXxe1MfmNpisZiVQi9evNCzZ8/8Lv2vFlTkDTOfBfISrkzHyE0Ba4RR4+SBSIIwYMzF6YSvNmhN4+dQnEmfoSsDt+Hv8zDicunp6XF+Il0wnyUYsECxooNhhTG/A3+PB3p6elrZbNZbVIEiwH04VCh2R0dHKpVKvi4cFFgIIZBarZY+/vhjj1Dn5+dOr+c6wWbycKOr5OEDz8Jv3dvba6E2kixwM3SA19fXDk3GUUKxCIfDhlSAOBgv19fXdX5+bqJmaGjIUX88H+DWiUTCI3swtIY/x9SDMD2bzTo+ke4LTBFIRZKLMNMLExDOGw5srISw2HwPTBfoPIOJ+xwUXHeeQ2yn9XpdKysrdpVRAIP4JIUwFou5G6aTbLVazvikC+bgqlarhiCwdktyNwpngfW60Whoe3tba2trToTDeowllOkDayYuLwT2LAdMpVI2jrTbbU8nOPhoUNB/4kTiMKFzxFoLLAGBRsEtFos6Pj62zleS3XCkaQXJQZQJHErYoSmYQd03TUmQqKIpQzGE/XhmZkanp6f60Y9+5IIs/S8XVG5es9nU/Py8uz1eQEBtgHq6B2Q5jLOMZHRIPKiM29w8CmyrdRN6jdsCTzzaNUB35BaQOHSy6OaCtj46XhaooQSAdQzCGgjvGRUB2oOynWC6Frt+wIlgQJF3MXogKUPALMkPCWN/pVLxA8gLSBcP2wt2xYnd3d2tzc1NF+2xsTENDAwY8GeHE9eJaxIKhZzFSmff3d1tHPL6+toBN5KMn1Iw+N34HEFpFOQBJNri4qLu379vud3l5aX3DTEu9vX1+VCk2+Fn8jsEnx9YcVQewekIRcbc3Jyv9cXFhTKZjNLptEke4vB4bmgAgCUkGbrA/cehTm4F95xDCByeFxVMFQgBLfDl5aW1mnw+pi8agouLmw2vdMocqE+ePNHW1pa2trY88fAzOShpQHhWWeWDSyoajRr6IWcVCIsvSNJyuezCxOHd3d19awcYFmacgCcnJ1pbW7N+G7yXIg8EF1x9xPfFtce1YLLl2aEL5fdjuqKp4z5gmsAgAXE6OjqqH/zgB86b5euLyuavJZv6/BdLxSBwGNEYUZEuITTmtAffI3uRAoj9kaKHhIQxAPwSuAAbIezpq1evXChgiwn+CI4a4CfxeNynfqfT8aphHo5yuezRD+zrk08+sSD5a1/7mpnIcrmsFy9e6OzsTKOjo1pcXLS7iFO33W47+V2Ssz9xj3ENnj17ZhdHNpvVy5cvjRdOTU35Oh0fHyuTySgUChmPhUktlUp20NDh8ZnRa2J0GBwctOqCMONqtWrlBLpKihEPeafT0fr6uldxDA4OOmrw4ODAaUyM9oeHh7fi71jVsr+/75eaAkEhDK69wGdOsWN8J8ACDDwUCumDDz64lSuKHRO9JM8lOkTcUclkUjs7OyZByIdlpOUfXn4UCVgoIU6CGDZEy8OHD9XpdAwLjI2NOaYRSWEwSBkSF8eVdKMHxwFGqtTBwYEJLEJvwDLr9Zv12CgKeP6DrHtQMRGLxfw+J5NJNxmw8nTHTJeJRMKQDwcf14Z3i3sKNgzxBiQ1NjammZkZXyuINojV7e1tlctlr71BeM/YjluMd5yuNFgHgqlYNHzBZYtY6NfX17W8vPz/XQP/Rwvq5eWlnj9/rnQ67Tab1pzWGmG6JI/9YBp0ETDnnFIwlZxCdHxBZwZgPZ0dF6qvr++W7hWSiD/Deg/cK11dNxsBSPMeGhoy8TM9PW0mkLGL0/jy8mYjJlmXpVLJLwrQwMTEhC1+z549M5tIlwcGGEz458/wMuJb58VhsylkE6lZhJwwxmLhnZmZcZA04yMwAfITSLpG9oCatwAAIABJREFUo+Gsy8PDQ0cPgm+Nj49rcnLShZeXQ3rr0kKTXCqVdHh4qEQi4d1VMzMzJt6Qo+GZx4vNs4IoHcJteHjYOQ71ev0WZkbHiN4R3BHpF+aON2/eWOnw7Nkzq0fQLe/v7xuOKBaLHjfRpDJRtFot54JWq1VDEWgjgZZSqZQJDpYm8nmazeatpDAkZhsbGw4iQaZFAApKjN3dXZOZkIkkZAW7brpP9K4coKSnwdITFsTzgFqB64JKBBKHQzjInfDeghNj/OG9BHqT5OLL78PkB1SHcYPPEgw/IteClUE0VGChwImS/L4iVwxOlRxWsVjMJOz5+bm+//3vm0P4//nqukYE94v+Y4DQ+DJf8/Pz+sY3vuHABEZbxizwDE5eLhLEwZ07dxQO3ySldzo3wRR0Edyk/9vemS63dV/ZfmHiTIKYiIEgCEIUQ9mSx8hRYjs9uKpTXZ1UP0M/R79K9+f+0lU9VHV3JbcqqXiI5CmyJVniII4g5ongTAIk7gf2b+vQV/JNHNmR7LOrVGVLHICDc/Z/77XXWtvpIsXDCyaH3ntxcVG1Ws0GP1RyVD+Q/HGSPzt75BQEBgRdhQoQiR7QAzctFQE4JUIEKtrd3V1TgpXLZRuaLS0tmcM98AXcQ6a83KSsV0HCi60dRrqjo6PmSs4hEI1GTbWTSqWUzWYNx2OizICJ9np8fNwe/P7+flsvc3BwYNNvHkzw2mq1qr29PVNe8flAadre3ja57MDAgC5duqRUKmUDE5ytgAd6vZ5ee+015XI5rays6L333jPmAmopHpZarWYbXMfHx+398oBhkNLX12dOS2NjY9ra2tLvfvc7Y1gAGSSTSYNIXnrpJfn9fq2srJj9Hwmh0+mY7BW/VDBrvAHgPtJFeb2P9mex0QDaE85erPuBOkYyYIsFAyrgMZyZRkZGtLm5aZxMCpUvJxraYZRdJFrnYG16elqhUMjuh5OTExvOYeHoLJIoKvBfAEu9fPmyHSpra2uGa9JBORkADIQZdnFgjYyM2HuG0UEHycYGurm+vj7zb+UawtWGvtZsNi+YquDFfHp6qunpab388svyeDz6z//8zydWp09Km0+1QiXW1tY0Pj6ua9eumX8oDz0KIZRTTp0vb87n85l6A6YAQypWZgCWU75Do6L1rNVqZkohySy5nMR3Bhrc/JiywFOVZK1oJBKxARKYLomVdsXZMvj9fs3Ozhpwzz74F1980WzmcLGfmJgwygyVMbplTvRAIGBsAcjPTrz50qVLhkfCWoDUfHh4qPn5eTN54SbjhoYacnR0ZPScxcVFk0IyaFlaWjKqFnp8hknABKFQSKlUyiSXYLZUPAwkoNDQ8u7s7CgSiejSpUuanp42qd/o6KiuXLmigYEBLSwsaGBgwDid3W7XZLfSOTH+7/7u74z7yUZb+K6YjlDtRiIRzc7OSpJVRcFgUJlMxuAfOMIvvPCCWq3WBR4nzIb9/X1NTk7asj4UVDia8TnSCTCoZOAmydp2vBbg9YKnDwwM2ByhVqtpcHDQkgxshb6+Ps3OzqrZbJpclQMVqIYuiAPcya4JhUKam5szaSn3EsMcKHAkyHa7rcnJSRtm0ioDIzDwQiizu7urer2uvb09K0IGBgass6rX60aup4BCno26EFUVr5npP9eTZxFYI5VK2dyDA40OjmIOmqLz8Ll//76Wl5f/6Nz3jVSo0nll89d//demaMIP1Km9ZgKLsgavUK/Xq62tLauYnNO5aDRqLXwkEjGqE2Rk/hszFenctKRQKNiJRMvAyQUvENMLHGdI9gDVVAnciHxwyWTSAG+GUpKsGsG2jkkx2GMoFNLCwoIajYZhWkgBedDgyjqJ0ogXeHhfffVV2z2F4XW3273gsZpIJJRMJlUsFvXee+8ZCwMKEmToTqej1dVVo9/wp9frmf6amz8YDNpBgcs6aiMoSAsLC1pcXDQ+Zqdz7pg+PT1tlKlQKGSDih/+8IeWQJ3R6XT0/vvv6z/+4z+Mb8mhurOzo7m5Of3t3/6tVdbO72s2m7pz547+7d/+7QKMwEDj3r17WlpaMnhoeHhY77zzjq2Vlh6pr1jKxyAMrvH4+Lgk2bS71+tZomk0Grbzan9/36bhmLYUi0XDP7+MZSYSCWuVwZH9fr8dQAxxqO45CFAZ4p+BUQhVIFVct9s1n2AGlEhxmdJj0j44OGiDm0AgYAUG9w6vz+s9N+pGOUgSZXhGJ3h6eqqZmRkb8O3t7dk9zJohOhD4n+Fw2K4R9yVDWLooGB0oBJ08Z6pq8Fa8HliLxP39P//zP4YjPy6+1QpVOt9eeevWLRtM4JLEqgpOrG730UoMXGyoaME0Dw8P7cOCKsQ0H6XL8fGxXRQqYKhH0WjUfD2pao+OjswpCkoQHprs2IF7B8YSDod19epVq4xYOIeGntNyZGRE0WjUCMRO79dWq2XEeSha8HBpn6hKnDxLqlSm25L00ksv6Wc/+5n+5m/+xn4+E0zghy/H8PCw1tbWdOvWLcMCU6mUjo+P7bDDJQlfBgwqeB0cOpiJ8FBDRuf6DA0N6ebNmyqXy7p+/bq5fXGgkWBRoN24cUNXr1597EEeCAT05ptvyu/36ze/+Y2Wlpbse19++WX94he/UCQSeez3xeNx/fSnP1Wn07FBZadzbqQDDQ+TDSwRFxcXFQicr4T2+XwGZYyMjGhpacmEGXQCOFZFIhHlcjnDEGFlOD+Xubm5C/f93NycVXJU4KVSyQ59BodIaFHcVatVYzJQSODyRgUeDoeNjUEH1NfXZ4kJOS00KVpsKkx07MAIdHjZbFZ+v19ffPGFzT6gmMHwgdbIzIMhEdsYkALjKEe16vx3gkk87mJwTnu9nl0XDiInJ5nfy8/k/QP3kVRhGlWrVb333ntfmUy/Kr6xhCqd02c++OADvfnmm3azo2poNBpm+edUUe3s7NiJ71QC0WrgBSDJSnYmuyRZJsFUmrThtHDQV5B7gpMyjKGCYILu9XptOEGFSMKF50dS93g8KhaLRpnCzxISOHxNqgxubviztInQn6gEQqGQ+YjisfnGG2/o7//+7y8YbhOPS0pgf2BPVNlMhZG8gl2hjikUClpbWzPcyWnOm0qlNDIyoq2tLWvdFxcXzTyjv/98OyiVHqbbS0tLunHjhhleT0xMaHZ29iu7or6+Pr399tu2Hvro6Mh8EJwUnsdFf3+/fvSjH9m9d3JyYo5WThFIOBy2BXK/+tWvND09bX6z6+vryufzVk3H43Ht7OxcWLgIJsxAcm9vz7jWKI2o9hmAwZQYGRmx+z2dTuvg4MBYB5IMaviymxNVK5UeMurNzU2bD5yeniqdTlslyzPHocDPQSIKrsnwFTMh6EdsQMB82uv12ioZWBT4+uKgxiDPKW09ODhQrVYzuAiPVlggVKYIcnjW4EezGBTWC9UvdDAokXhrsKsNHjiwDTDVhx9+aJtvv058owlVOk+gv/vd73Tjxg2bFrJuwtl+QZ6HA8pEF7YAXDQepFQqJUnWgni9XlUqFSvnAddpVbrdrhmn0L6B/fFzgRVIOvBNfT6f2ZXhYs8piH6eFg38p1KpmKEHJ229Xlc+n5ff79eVK1cMr0EKyykLubtQKGhjY8OSDMMkqukvvvhCN27ceGxCJcCTP/vsM33xxRe6f/++qtWqPUw4+2D6QeXM9SWZY6y8v79vLR8LDiGCz8/Pq9frqdVqXZBh5nI5JRIJnZ2d6e7du6bHxxw8GAzqzTfftCnu/y8SiYS5jf0xAX8ZZZiTaM9CSB5IEhQeqFSkyJrBCcfGxmwQBX1rY2PDaFioxeDDsik0nU4bpuyEor5skwcFj4TsFM3w9fF4/ILIhEMMtg12d1R/HOwHBwemeefzRUEH64TrQ8WHXy+mMQy5ut2u+TGgBGQ1NQR8FFUUNmC+SMRhanAf8vk4Oy8nq8A5bKOAAcPlPnUqKPlZDEs5bOAb37x5U/fu3XtiO/+HxDeeUKVzm78PP/xQP//5z+1UZcKJ3hqSPx8SJyNgMe0KLbSTxIvax+lsDj7k9/sNFqAFpioOBoN22vNgj42NGd5UrVZtiru8vKxwOGwrOsCAWPPAcAcqDwM2TlRcjRj0sO8I0wawoF6vp2QyaUOnarWqkZERs3djws5k/V//9V+takNhBJ2kUChoc3NT7777rm0qAEbgUMlms8YVpc2vVCoKh8M2fOLB5CbnBoeHyJI4BglXr17Vu+++K0k2qWfrbbVatYegXq+r2+1qY2PjggrlmwqGj3fu3DHXo1qtpkAgYIcBAz+GNHhq8vBRwcfjcZvAc4+z1gSVVzgcNo4tHrAMEWOxmJHp+VxwuaIzY0NFvV7X1taWYfRs2XVWtE6hANxXzGUQ2IDnUokyY8BakXbeqZ6CYgjGz89kwMVcgZ8Nd5TDB4EFyRx8n3sew+6xsTHz2+Dn8uwhtiEPANs5lV1staAyBapjPiLJLPsQZoyPjyuXyykej+ujjz7SRx99ZAPOrxvfSkKVzm+4X//61/rJT36ibDarXC5n5swkExKFpAsGJjjBcJFoXZi243qDIoMqotFomOqEFgeKBK0W8kgAbkyeOdmcVYskbW1t2c4ksExJllDZmUNrQ3uB+UWr1VI+n9fCwoLi8bjm5ubsNfJQYSDCymioKOCS8Xjcfi6HCZxPBh9O53qAelohJsIej8cqR4xoxsfHjYJGpX96eqoXXnhB0WhUrVZLpVLJ3PD5vA4PD3X79m0NDw+rUqnYIALPT0jhk5OTqtfrxlfF3IP//ybj4ODAhAckcwZGHLzwiqvVqpmkwHWmpUTuuL+/L5/PZ8yG7e1tS3BUnFtbWzbEhHPsVO5Q6YL9cRgipfR4PEqlUiqXyzZsheKVy+WMmkgLDncWLBYbyoWFBRtAQS2LRCIqlUoqFoumgGLQCg2MRMXUnLa/1+tpc3PTcFLwU9SO7CsbHx9XIpEwLi7Tdqpy8NZisWj7uoDWgJUooIABGC5xP0OxcvoJkMDpWCnM6BRyuZwmJyc1MjKi27dv64MPPjBc+0+Jby2h9no9LS0taX9/Xz/96U/twSXBQSKGwgCuhA4e3C4cDtuH0e12rX0Ej4U+RZUJLgsdgw8J7trZ2Zm1rhsbG9rZ2THeJ8D8+Pi4rdtttVqqVquWIIeGhiwJSrIqGqpMPB631Rtra2s6PDw0Zx8S4tDQkBlII8lFCRSJRLS3t6fFxUUNDAxYZZRIJAwHphoNhUI2KKMtmpiYULPZVDabNZVWt9vVjRs35PV6bc8Wq1MYZuBSzkCNFvbk5MQgGfiVbCRgzcfS0tKFdhpV28nJiZaXl637YPLd6/X08ccf68aNG6azl2QmOU8rPvvsM21sbBiG6JQZS+dQgnNAyH0Gfr67u6vp6WkTFbBtFUksVoL8rN3dXeuWJiYmzMCZxMA1pGjgfdOyQrVCGAP9a39/X81mU8ViUT/5yU/M/m9vb894k8A4cK5jsZgdCtFo1ChpFBG7u7tqNBo2YHJKPBncQNmCY0p1Cn+W1fJwwznMGb5yeIK1Iy4ANmI4TEKne4Ryx3UAy+/v77cKl+qV6luSHQLAdlDWgsGgpqenlUgktLGxodu3bz+VZCp9g7Spr4pMJqM33nhDAwPna5E3NzfNvMOJl0KLodU5OjqySTNYWDQatXYRjIbERAUB3xHZIXgYC/FCoZBNJ2EXwDjw+/1KpVKGpaGTZr/55OSktSPNZtMs4SSZAw4nerlcNgoWfFYS96VLl6wNRBbpdIAaHBy0bYv7+/sGqDs/I9pE4AGv16tcLmeJnXUaR0fn+52wK2Q4Ag7IzxwZGVGpVDLSM25E8Hixr2OKTZJfXV3Vf/3XfykWi+ntt982WXG5XNbm5qYZVkOVu3Hjhtrttubm5hQMBu1QvX79ul5//fWvxIidwYP/uFheXtY//uM/6vbt21bRzM7OmrT50qVLhqNiYccAgwS7srJiBjx0AmCUcKrpTKAB0c0gvfT7/WYQg/KIh53kjKE5GyGgGo2Pj6vRaBiU1Wg0rBg4OTnR5uamPv30UzOvjkQimp+fVyKRkMfjsb1IdBWwaPB/hQcKrRGvY7BVqFD1et3uI6ecNpFIGJzGvzPw5f6ipT8+PrYDB18JoAFeH9WnUwREAYOpDYWYpAszD6drFNd9dHRUiUTCII6VlRV98MEH1lX8MfGktPnUzFH+mMCJe35+XpOTkzYcQT7HKcjUEroKDwuAOW+KpAS4TfvkNOeA1CvJDHIjkYh+8IMf2Ck8MzNjmA+nHSA9FBvAcHhszWbTZH7o0qlSMBdhOgveEwgEbJhGVY4Ujp+LdR4PP205XDlWc6CYgtwcDodNqhiJRIyNwFSTG9rv9xsbgYp6ZmbGKiHpkaih0+mYpyR0m1KpZMwAbO5wkPJ6vWZMTDWCWXgymbTWb2Dg3Kyazx2JJ2bawBz4jz4uwD3BzpwHTLd7vsfs888/1z/90z/ZcsSZmRm1220VCgWrCiHTo+pyOhQdHBwY1i098g9l8MK9gtABbM/j8Zj81CkLlWRdzcTEhHVjYIa1Wu3C4CYSidjvc1bEJC98GKDg0VZzKKN1BwvnnqVbowOB4sXrx4OhVqvZ54EDFIMoDj+SKn6kwChUsvwJBoMKh8P2XqFiUZHDiuCQx3vD4/FYJUvidWLDzGacXHQOIq5ZKBSy67CxsaF3331Xe3t7XyuHPSlt/lkqVGJyclJ/9Vd/ZRUge9RxHkLVwkSWYQ8u7FSuUCqABgCjT05O1G63DWyHi4pJAqYmp6enZj6C8Qe77FntC9bY19envr4+o5HQkly+fNkqC1Zg40/gtH5j4EP13e12LYmQ1KLRqOFnTOn39/f1wgsvmB1hPB5Xo9FQs9lULpfT2dmZyuWy+vr6lEqlFIvFVK1WraLhtK/X66bugj/r8XiUyWTsWuCUVKvVzP+TxJFIJOwBSKVSGh8fN2d/NPncN6x/9vl8unbtmnGDeWhIvg8fPtTg4KBef/11o+/cvXtXZ2dnisfjun79un7+859bK/f/i16vp1KppDt37ujmzZu6e/euVd69Xk9TU1MqFAq6d++eEomEXVdMa8DAwdigQ+GOFIvFrLvC8MdJ/8NDIRaLqVarWXKlc4CmBOwxPDxsFTswC9Jo2BdIkAOBgA2BGO6COUoybf/m5qatuQFGwhgce8W9vT2NjY3ZfcsBtL6+bhJtsEykohw6zCdweJMeqc2cHr2SLNHDKJEeiV54rjnYOTSAk7gG/H5+PgkcTi6YNHMI51AaEx3gscXFRS0tLZl45uvEM1WhEqwY4XRmag/+BmHciZ9AG2m326aGctpx8UA7VVi04FR8THHBpLa3ty1pId+k1T45OTFxArADLSBtBaoddNUQ/JnUQ+FiQs6NwjoIWj/2C9ECxeNxG+5wg6Ko4Zr0ej2bnJP0gBao7rmm/B7UQ61WS6Ojo/rBD36gyclJaz/Bhy9fvmzMCYQFYF7IHhlOUf2D07HpMhKJGCMAZsHu7q7d5PxMhgespcF1ipac5PNlBdWX4/j4WP/+7/+uf/7nf1axWDTfB6/Xa1NqZLEej8c2trL8b2BgwFacwwWGW8k1AG5hNQy0uU6no3Q6bRJe4CuSUrfbtU6Ge0CSqZS41gyUqC4ZKPLftNehUMgkqhQalUpFW1tbGhsb0/z8/AUVH58XxQo+DjwTYMq4i4GB0r5j8+j1eo2SCFQGfg+1juoauIuKv9lsXvDF5Tl0yqrxFkAg0u12rQNzrkKBK3t8fGwevhRLXD8OldnZWY2MjOjTTz/VJ598YtX5140npc1vbSj1pGg2m/r1r3+t+fl5zc/PXzAqODw8vEAyBlvhBqDdQZXi8/nMuYeqk+GOs7Lk5KJtkR7hnZLsQYLsy80OH5AbhMqXG5nWDaqTJMPG1tfX7WGiGibBMwVH9w8MwIOCxBZmgiRzmXIaMJMgut2u+W9y+k9MTBgdjZsdyzNgllgsZocFU++zszOVSiWDE0iuQ0ND5q9Jd+DkJDKs4BApFos28e10Opqentbg4KC2trYusA2QEJLctre3Va/XdevWLXk8Hr3zzjtPvJd6vZ7u3Lljdofb29sKh8PKZrNGHue+kWRkc/B67gMGRk5lDlQhlhRCMnc6egF7SLLPk2oNTB4oy2m7SIvK7wMfRaTCQIxB5tbWliVzSeb2xabc4+NjW3TI9WcKzyyB+2h/f986DqAfJu8PHz60NTMkLqdZCvBWsVg0a0soXRwedAXwxxnCMXeoVCrq6+vT9PS0dZ4UBshfMTNn6AzrglYeMxa6CihRoVDIOOMbGxt68OCB7t27Z8/5NxF/1gqV4EHvdDpKpVKmTIEWAc0JZUY8HjeXJ1xnAKIBvyFFs+ET7BBs6uDgwIY/TFjHxsYM1Ob0pfUG1yXZ7O3tGZAOTYj2FoUIhi/4ADBMOjs7Mz5cf3+/DV1omajCYS9Isgeg0WjI6/UqHo/b1BR+q5MLyvRdki0QpBLj4HHalzGAAoqgmoCi0m63Te3FjQwG6ExUmLrgN0C1DjwCvsZggmuaTqcVi8Xs4eP3lMtlPXz4UFtbW9rb29OVK1fs52JoQ9W6urqqX/7yl/bAgK/RbgNj0FJz30jndLilpSWrCsEoWbNxfHys4eFhpdNpa0PBbv1+vwqFgnFRpUcHKcwKBCZbW1uGKUMbYgcVnNhgMGgJiP1Jo6OjqlQqkmTdAUwO6Fy8b54bqmPYNKgGT09P7f1zCK6trVklTYKiE6J7dMJB4MUUKlToCEY8Ho8dznwvhz54MBxfqmeeC4aY7CZjCMr3U9BQFDgX/g0MDJg2f3Z21rwQbt26pYWFhaeWTJ/ZCpXo9Xp6+PChPJ7zxXBwMFEs4VcIAZkKDZ0/hGMwOKhH+FpSJdAG075RETiVIbhZdTodlUol9fX1WRUWj8cv2JdRHbPGmX07YGeQjoEsUJt4vV7VajUNDw9bwnH+99HRudkuZsNIPFGCgMOhsYbDy8BrZWVFm5ubmp2dtYoYo24qVx6KjY0Nm2xDFaPaBs+ampqywQzXkoeOwYpzeEg1lEwmJcmYDcA0KMPQ86+trdkwDss9SSZkgGWRz+eVTqeNGkbV1Ov1tLi4aDJQhBtUX8lk0g5FDoJMJqPNzU2zIUQCzBJFqGTIhavVqjmS+Xw+gwsCgYBSqZTJIHHPR3lH1QfJnnsL+hsdF6+XaTvTf+f3orSC9gVfmftSkn7wgx+YogjoAYiJZBeJRNRsNvXgwQPTssO9xYQI4QnvgbkECY2tAAhiIPtL5wPZQqFgUBYHEh0RBw9DLA45OjugA2YUrN0hISM+gIlA5RyLxZRMJm3fWrlc1m9+8xsVi8U/SQH1h8Yzk1Cl80p1YWFBrVZLb731lj2oYEAMQDidIKPTRgP+JxIJGyJQeUrn3gL7+/s22HEuYTs9PV/9DMCO2QeGJFQ57NThBiZJnZ6e2qI5ZxvV19d3ocqZmpqyypI2BRVWMBi01pOHKJ1O23SVm4lTnQqbg4TDxvk6JZnlXKvVUrFYNG4lBhtgdyRz6bwKot1EaomgANz45OTEDgkkvCQshoWhUEiSDNdmiwNmGrxfsGRI8Kh40MPjWcoDzaFI+06ChZpDZddqtdRoNLS1tWWDkXq9bu+ba3rp0iX5fD6DJDDMBsoIhUJWXTI4AXLCHwLYCc6n008iEAio3W4rk8lYJS/J7gswykQiYe9vf39fKysrxoDhXueaU0TArgC7paqEJSPJPAGo/DhUu92udWJ+v9/4tXQYk5OTRvJHF8+Ak27I7/fboBUxjtM8hsMbtysYBRQGsBboOlFYcajTpVH4OGELhnOZTMbWlAMR3r59W/fv37fv/zbimUqoRKVS0S9/+Utdv35d2WzWlFUPHjywoY6zJXTa65GokO852QFO4wXIyyRjhlpgXJKM6gMZv9lsqlAomJuPJOP8kcyc7Z0k00wTh4eHqtfrtpudyoQBFrQbXjPc07W1NQPah4aGTNudTqdtWsrGVNo96GC06NisMa2HFnNwcGDVVjgc1ubmpu7evSufz6dkMmnUJPZAgSEyEKMbwGcSaEGS8vm8DR0TiYRVrs4JLTxicGySFQffycn5XjBglV6vZ/gq1mzAOVQr8BuTyeQFdyKwvaOjI3vY+fzBIGEpkGSIdDpt0mQODqhuUP+453Bu4nDBKAYM0FktIqhwVr/4BDgrwrOzM4VCIatg4ThjQQnZHSiF6xSNRu0ehAWAqYvTVAUoBW4xz05/f7+R9mndkTAzi9jc3LzgaeE0cQZX7u/vtyEpzwgYKJ4GiUTC3k+z2TSZrvTIgpBODGMWVsszK9na2tL777+vfD7/jeKlj4tnMqFK5xXiBx98oMPDQ/3sZz8zjicKCk5TDE74QKTzagWHHbA6VCHgrtVq1cyTaVOYeHa7XbuZkXai02bgwh7wYDBolbQTn6WqYH0JVQ4kaqdMjpXFmAyzV4tKiEQI7xPJbTgcVrlcNmyWSgcjF6chcb1et0oSocPZ2ZmteqByOD4+Nlwa2z4egkAgoKtXr0o6v7nhEjK04GBBeQTTArtGjKYlXRj4MPVF6430FmUWGCOvjUoQPiOfeaPRsNaTRAtPVpLJNqGN4Y7PsBPMk3Ua/B3JGoiH1x0IBIyWxnAynU4bHursVuhYgAGcLBXYJnRJ4+Pjhl+jJuPAx2wFvBHSOt0YXFIS3dHRkTFpnLxVICNwZKegg4Ebv7NUKmliYsL2n0G+Z3sqnFVsNXl/JF1gDu5JWvmRkRHNzMxofHzctrviHAcDhs0I3LvMNHg/CGymp6dVKBR08+ZNLS4ufitS5sfFMzGUelIwBWTjYF9fn7U3GE0j6+OEh/rEoAkaCmR8p0EFE0R25FBl0ObBGODv0f6TxPAOAFOE7sSNzh8wLJI1DASSA9/L0KfX69nyQSSdsVjMCPW0OcANqMNYwAb8gTQXChckfaYjZTWLAAAgAElEQVTFAwMDikQiqlarJgdttVqGFWMODj4Fzgh2iTkHpGwebuS/Q0NDyuVyyv7venHUNwy/qFSj0aiy2awZ5bBKBBiCqoUhx6uvvmrDM7qOdrutDz74wP4fNgK8ZibfdDcMq7D+I4ExUXeS3alcqV4xVaH1pdspFotGpgdqQjnlxB35fY1Gw5ItcBUdEokHrvLZ2Zmp6FCmQZXb3d1VsVi0v9/b2zMuLdU+mPza2prp4RmcYZrNZ8gQCUMhhkIkWaha0A/39x8tUaQbcXKSUSLxernnuW/oVPAo4Do7sV9gFRg3qVTKqG27u7vK5/P66KOPdPfu3W+lxX/mh1JPik6no4cPH2p/f19vv/225ubmrGWj6hkaGjJKCpUPNzFcwFqtJo/HY0Mu2hU4lTyAyE9RmmAIDd7GMGBsbMw0xpzUHo/HNnuenZ2ZOQnekOiawYTBunDeJ4nt7++bD6TTbZyDhMk4nE/aQrwPJFklhfEMDzdGvRjoMlCDthMMBg2jpQqXZO0s7R4PU19fn8ECKFz6+/tVqVSshWs2m0Yr63Q6SiaTF5yC4ASiq4czypBMkoklcBnyeDy27E+S0YpI5lDncMVHGUSb+WWIB1MOPEDBAmnHJRkjgtfHwLKvr898DB4+fKjJyUlLfKOjo/b/GCrv7u4qHo9bUucegeeJcTfvAYyfqpChYa93bt3Hvq6DgwMlEgljnnA4MGyFA8pWBeceJqTYKANJpgwaj46OtLGxYZAbLBUgKqhNFAfBYFDxeNwKIEkmnT08PLRBoiTVajVLwjzTsFmAJGjnGcCB0XY6Hf32t7/V+vr6U9Pj/ynxTFeoztjb29Py8rJhhVQhVKO011QukqyiYU8TmwCcU8PBwUElk0mjLpEQMJagQmDq6qSFMLHs9Xqq1WqGldHy1ut1q3qgg0CHoZJBBUM7DGeQ9h02AfuIWPQ3MTGhlZUVO/2Pj48t6VOZs2YGPT7JMRQK2eoOj8dj0j3wPsQNVFlgd8graaE9Ho9VObRzYM1MpeGgomZzGoDQvhaLRfvspqamzPaQr4eDTDdy9epVO2ic9webVQ8ODjQ2NqZ2u61qtapEInGhG4FRAYRD0qFyomVnhz10OKg5VL68f9pgxBNcW3jFDBT9fr9dL+z3kEzyM+iO2NHF0IhEJslMdTiUYKzQeTBsBKah9aells5XicBZrlQqZowSCoXMAAbWgHOzBjxj5+QfnJTrB5QANk2i5UDnsHLi32DBkqzz4l4G+rhy5Yqy/7spF2XgzZs3tba29q1jpc9theqMk5MTffzxx1pfX9ebb75pa3cZWoAlMYmVZA81RsDQsgDc4QbyoZIM+/r6VK1WJZ2rMniYnFUgE1yqMQZEzqSB+/vJyYni8bg9WE5uI7xVkiNE807nfE0HU2vAfSaqGMVgnE0rBi2Lio1qFtcsn8+nQqFgBwdKq3g8btN5KhUCjJLEPTExoVKpZLDC4eGh8Sh5OGi3SSK4N/E+KpWKOTBls1mdnZ0pn88bzstAj2qSBxMepzOonnlYWXFNu4gNH5U5eCDLILE/PDw8NHVXo9Gw10L7CdSTyWTMj8DvP98kUSgU7GAmuZ6enhqrgKoQ/J/DE79cj+fcCR9uLOucoac5GS2tVsvafRgCoVDItgpQ9TOg63Q6Nhgrl8uSzg/E6elp68SYDwSDQc3MzBj+yvuky5Fk8A60N+iCJG8MhoC5JBkTheoaeTZafTpPFFw4wsViMVOk8ax8+umnunfv3p+seHra8VwlVEkmrXz33Xf1l3/5l8pms3aCs2QOxxx4eCcnJ8YrBA+iKqMy4UahgsHZH4cr1v2CL0kyzBGrvq2tLRsseL1eTU1NXZAKAs6DF4HLMZza29szOSgYLm0OGCrTYzZfInVF2uncB4RCCtyL6TNrW5gIg3+hpc7n81aFQI/CyIQpP1LcVqt1oXIKBAK2ox7VEcPCTqejqakplUolbW1tmdTR5/OpWq1eWKuMSU0mk7HDhqTM1NcZTv9NVEYYQ5fLZW1sbBicMTw8bE5HkkxxVigUlM1mDROnqqLSo1tYX19XJpNRKBS6sOqbNTYMFumEkFWCp3IPMoGvVqs2vKO95vDngGJQlM/nbSDIv3e7XW1tbZmtH9N97j3ob7ThAwMDKpfLyufzkmScU6CxDz/80FRPfCYM+vi5DCS5T+BbS49cn8CXwcthtdDhONeq0P3hD4DgZXBw0K51s9nU//k//0fLy8uGxT5r8dwlVKJWq+m///u/NT8/r1dffVVXrlxRtVo1sJ72hirNyZ+DyE2rfXJyYi2+JEucTD4hsdNicaOAJ6Kp5uZCu4/5BBzJdrttVSEsAlo8/h48ltaah5qbl8SMixGDMXBeTK1p1yRZ1TA0NKS1tTWrWmEVOFssDh88L6XzG9+piafCQXHD8C6TyUiS1tfXbe/V2dmZTXpJ2F6vV5FIxB7Qo6MjG4pBU2MK7Kw8ndXilwNOoiRrpxl4sEUTXBtJKvxdKkcqOHBKKjvwYe4bNjMAq3Cf8T0M0Bi60R6jgHIeqpJsoSMHG/cPHGjaduS/7XbbaFTQm7jfSNKot4CU2BSMZHR0dNSYCFDUwKxPTk4MMmCB3snJicl3nRU79wG8aUk27KTy5ICioOD7R0dH7e/o/MbGxpTNZu2z4HleWFjQ73//ez18+PBbb+//mHhuE6p03uZ9+umnqlarevXVV81Yg8ET1l88FGw8hHRNMkOvL8k+YOSPAwMD2tnZsRsbdRWViCQ7gamuUDxRGdCW1Ot1TU5OGnZEFUS1guUZVBba+fX1dVUqlQttFTvb2fSKsz/Tf95nq9WS3+83uIH3BYn/7OzMHkRaTkl2PahiGcDAraxUKsY5xQoRnJPrS0sIDMOwBM5wpVKxyg6/AuhQTJzBzBlMgCsiwCAuX76sO3fuGF7JfiKkxNFo9ILPLV3E9va2ksmkGbfgOAWeyOAPrjNmLYeHhyqVSuaeRODKPzQ0ZO5ftMoMb9gX5VQOUQE7pZVM0p1dBER+ft7JyYlV28ADULL4POBN12o1mxHgbAY3FyEDhzzwR6FQUCQSUSaTUbvd1vb2tlXq7GVCostnDK7LANcphWV4i0AlEAiYkToH4MzMjCnOsF988OCBJexnOf6s9n1PM/r7+/Xyyy8rl8uZbykkesjg3Cy0xNCJJNlJiDSVqoSTFnyTqvDg4MCqPiR60Ga4WfgeMLlqtWrMANo2pK8oiEi+JG5aVLBXpJ/O4QeKKKz8pqenDR/c2dlRPB63B5Pk6VzWBn4MnAG5nMOISgP6jtfrtYEOSY8BBA8sFQnDGRLQ6OioMpmM8YOdVCCUO2zARGJ8enq+sRN8MpfLaW5u7oI7171797SwsKCTkxOVy2XzLKjVaqpWqxoaGjIZq9d7vs4bw3FJ1trSojabTWMDQJejvXZ6QkjnHQ0Tc6e8lwSAWxhsFLqBYDBodCWSZ39/vxqNhiqVikZGRqwaxiaPw5TJvnReBCQSCXs/DOOc2Of29rZarZb6+/svLNpjGOlcLSLJBrKVSsVUbHhdUGjwuhGmMPCFdULhAGYKuwUPBsy6Q6GQLQvkmrdaLX388cdaXV21zuNZiielze9MQpVkbUgqldLLL79s2mja/lKpZEoPKlSqQPAuptN9fX3WVnGjkSSYOgIZcAJDAzo+PjYZqpPKMTIyop2dHdXrdbO2Q7IJKZqqgmTOUAssFKiCCS3eA2CcOPvX63VLBsg8OWhIXMg72ViwvLysqakpa7V5qE5OTizBhEIhG94x0AEDHRgY0PT0tCnWjo+PVS6XzaYP3TcuSHQHUGnq9bq10Syn4/DidULLCQQCisfjZtDB1kt22NdqNbPiAwP0er1KpVKmwqGVJhHgFREIBFQul7W3t2eUJRgcHCBIQFGP0dozSHKS+KlCMY7BlpIqGKwa2AKs9PT0VJlMxhYsIiA5PT0103Bwd+TFQA/1et0cqKAE0nlx/3O/Qq9DXsrhCrwFzk1xwUGJZwO0KxgGvGenlNmJyfP6U6mUZmdnbVBHwv3iiy90+/ZtGyA+i/G9SKiEx+PR5cuX9cYbb1hbgbVaoVC4YI3GTcJAB3I9yqdKpaJoNGo0Kao41ChULLSgSOoYGIG3gg3C1USGyU2UTqc1MjJyoa2amJgw2zhcnVBZQbui3YKviLKGdtuZmAuFgvr6+i54Rzq10oFAwHToLHvDss4pRoATyIAEHFQ634YQiUSMw+r0QmUA0el0DIcFF2XwA+E9nU5raGhIqVTK5Krgt/AgwTVZmwKeCJmcPe9Obb1Tzso6aUlWqXU6HSOck1RgAZDMisWiHd4cDpi3AN3wmjGX4fBzDmhqtZr5JEDlcvKocaJCeAL9jvae7wW/dA5S8Qjl8ARewkuY+5bBKAUBwphwOGwQF4UDPqVUjAxgUTFx2GFIjTE3Cjc8UzEbgs7Gofnw4UN9/vnntl/tWY4npc3nGkN9UvR65wsBq9WqXnjhBaXTaV2+fNlOUgZKtCZw+mhtuSlQmsAGYOAE3YqKVZI5M6FnZmCFuoMEzildLBZtsgsm2W63zSUe3FeSLYnb3t42IjytOQofbnL4tfwbElf8Jcvlsj2UVHHDw8OW7JmOSzKLP/YiQZ1yDmx4iGj1u92uUZfAsTE4QYIqySzyMDQBb1tfXzcruenpaW1ubqpYLMrv92tubk4jIyNaXV21hxs2BK9nZ2fHdlvV63WVy2Wtra1pYmLCdohRNcEhzefzisViFxy/+Dzhl4IvgtNXq1Xt7u7akkg6DkxYaF8xWqHrkWReDeDu4KK8LjB9sFoMRJxKOD4fDjcOZ/7NuaLE5/PZHABMH0EC75fkykJFjHecpt5OHi/XD0ybSb8k6+Y4SKn8R0ZGNDk5qb6+PqN5nZ2daW1tTe+//7559T7P8Z2sUJ3h8ZybLv/iF7/Q9PS0arWacfiYvoOHMrThZgATAxeiEsBmjDYJugg3EA5WVHqjo6NWVTqpLCi0GOBwysPbdA7USDgkeK/XawR6FFSsK5FkBwR74Wk/qSiomqg+aDV9Pp+mpqbs/fC6qVhJ6LT4/D8PKNJFVFEcGpLMPxa6D9Ug7askc9aiXXQyHvh98INzuZy58pOAqObw8EQZxDrkaDSqqakpo5XRbvJ6oIDxd7TE+CEEg0Gj0XFI0PbG43GjMLXbbV26dEnZbFarq6vG8gDXdg7dgJ0wFEdJBiOl0+kY9Qp2BgcsgxyGkU4mwtLSkra3t+2AdN7DUO24N4E2sIuMRCJqNBoXDIQ4KJ0kf94/5jtOJ6iTkxMTokAHRLDg8/m0sbGhhYUFFYvF52Lg5IzvVYXqDByR/uVf/sW8EjOZjBlrHByc769nKCTJWi5uOipCbOto/3u9nlGroGpR1Q4PD1t1Bv+TUxx2AKwAoINCoWDbAiDBk/Dj8bhhgWjbSSRMtamER0dHbZq6sbEhSTbAwd4NXI02lITb6XR09+5dw/sglKPjh/hfLBaN2hQIBGyRHJNnsGqwxEgkYrxHBg+Dg4Nqt9uqVCra3983pkIgELCV3oVCQclk0tzc4SfiVQpGTGLid2MUwmvmwacCcxLw6UCopjhwqB6ZevPeSEhQo6hC4dxyb6yurhphHsUUEAJtdywWM/tErq1TYgwrAVpVp9PRysqKdSsMUqFvIe4goQEpOXX6vG7uTQon7gUnjQ14hAPXydsm2fPa+Xt+//T0tO2rGh8ftz1oa2trun//vlG5vkvxna9QHxfhcFjXrl3TSy+9ZO0NID0VBF4B4FRwFqkUedCoDkiW2KCNjIyoWq2qXq9reHjYqjNI3wxTcADCPBhqFHgfQy5+Nriac9jATZ3NZtVsNu1ho6pA4ojMEhEElBlJ5t2JIxZsAJ/PZ05UVB/BYNAGSVjRoUuPx+Nm5kuCBTOkC4A+xHsnYeDnOTExYd61tJVcu2AwqFQqpWw2q2AwaHgz1wtDEfZ8cVDgNIWpN/4DYKr4mYbDYTtAJZknhCRT8JCcGaIBBTk/V+4hWm6YAZIMdtjd3bUhHdeE6wM7gkWMTMWpQmFAwIN2ms1wCDNdR7TC4YKnAIcsVDVJBu20221r7yORiH2eQAFOJd/IyIgmJiYumFCPjo5qbm5OiURC+/v7KhaL2t3d1crKyjOju/9T4ntboT4ums2mPvjgA92/f19XrlzR5OSkDRhQa3i9Xm1sbFibRDuPDyUPI+06Qw0SMes2nHJVKh7kn5ilUGVQMVMFMB0msfr9fiUSCR0cHBgvlOqWxMCNDlwBDschwFpr+Kd4X0KdcrrDww2sVqtmToLd4NTUlLrdrjY3N41WRLvtZBSMjo5eGLRhMsx0mmm59OgAZ/BEmz40NGRtsN/vtwTiHOB0u10TO1ABM2hB2AGbw+s937SKEQvTboaR0Kow8EDJ1t/fbxtqwZWpiMFJeR9wLlEo4QXRarVMmgl8BKUJfTtYPIcUgg1eJ39HYpd0odPAGQwRBdBWqVSy5Y8kbvieQF3SOdzB/jB4vRihwEEFHnJCVtls1naXSTL14MbGhr744guryr/L8b1MqJKMzFyr1WxV8PXr1zU2NqZwOGxYD2s3SqWS4ZtgRtyUtN/Yj2EqgmWgdM5FhHo0NjZmUIJ0ji0yKaUqY2qNfaAkGwaB7zr9BSTZFJ+vQcKHnj6dThu3EO4n+KxTITQ8PGz4LsYyfN3BwYEWFha0vr5uskrnlkukmMlkUgcHB7p165YZYD/uYQKT5D34fD7zUR0cHNTU1JS5OTn3jBWLRauQGQR6vV5r5Rmy4P2JJR++uKwsPzk534rAa2fQQ/VFR0BFzXVgDTdVXqVSMewdy0RwViSv+LWCSwMR4TOAFJMBE/glfM/9/X3rqIBbOFDpKqjgGXih3CIp1ut1u2e5tzwejzn1o5qCQijJHN04BDh8Sd6wDjwej3lJ9Pf36/PPP9ft27eN0fF9iO9ly/+kwMpvfn5eL774ojkUtdttu2HZ5unUWztb/+3tbQPlGXwx1MLo2kmzgnXg8/lMCAAflRbf7z/fKcWUHdUOZGr+mzZXklVakKsZuji9SsHUeE1UerwfBl+wHqjUwQEZFFEBoZo6PDw0ldrKyspXaq6DwaDm5+eNg0nVTkuK+fbExIRtBsAN3unORaLEGAWZJUM8XiOKK+APlvYNDAxcoGHRbYCVckhRxaKHhz0BaR62BQIPOMVOw2sSExU1zk5UnnQMHLj4PVDV46nK+4PznM/njXmAcg0ogBkAjApcrrhXub95fewQ29nZseFXPB43pkQ0GjU6GgPQdrttnqvlctkoiN/FcFv+PyBOTk5UKBRUqVS0uLioTCZju4aouBALLC0tGf6E2gdYgOk1GBYPN7Z4PCwHBwdmYQdmBUWF9n1yctJURewox+gXIjc4qSSr0vr7+9VsNi3ZAEGAc6LjJ6nSGoML8pDQ+ktSqVTS9va2YYFM4p1KFgZ7sAm+KplCzs9ms7YYcXJyUl6vV7lcTpFIxA6RZrNpLAjaehI8lWQ0GjWSPQYc2CCS3GBNcPhA7md3EiqrnZ0dG6jR/nKYSueHXjqdNhgEQvzIyIhhkNDwnFaGMERgEJBInabNqOygxHFIUbHu7u5adYrtHgY6vB8kvlCToOLhXQB8RRXNZ16r1exg5xpB7YNiNzMzo2AweEGYsrKyooWFBTOwfpb19t9kuBXqVwSa46tXr+q1116z1nRtbU3Ly8uqVqsXfCkhr+MyjhkGu+XB9iQZ6R2JJSou6E/QpsD2pHOaCjhrp9O5gKNhUjIxMaHJyUlbMEfS4WucO7X4XqhDcEepgHDCRymEBp7XA0eVpOX1ep+IkeGYRNUHLjo0NKTLly8brQjWBIcSlfvh4aHJK0ks6OgxhMGRihZaerRS2Zn8Q6GQ4b4o6dhyAFzT7XZVLBbtOpEY0dADN8RiMetiaM/7+vpULBZNlUTl7vQ0gDVB4sHFiQEpeDEqOO6vUChkxHe4v3xNLBYzI22wfoZEsE6AVGAzSLLPHikwfqzInlFhweRgyAZPd2NjQ5988old6+9DuBXq1whI/7///e9Vq9X04x//WLlcTgMDA5qamrJqABJ/o9EwCovH4zHCOYYhDA7QaDNEQLkELEBLSLIbGhoyJ61A4HxlMUKBbrdrXE2qFiotvpeEBMwAa4A2HooRlR0DLIYmPJC5XM4qMOzZhoeHNTk5aWuBwSuh4zBBxjh6c3PT4Akw3Gq1qsPDQxWLReOQkhCpKp04IgkjFAopm80a7xXckvdKJcZwhJ8HhIErPXxPtiQMDg5qcnLSYA+wcJ/PZ760VJ0sIYTbytJDOgdgE6AB1Hhce1gdGGOT5BgQ8VqBDJyqNlzSBgcH7cCORqPmWEZrj0ihXq8rHA4rHA5bUiUxgNvzWUciEV26dMmsJOmuMIh58OCBlpaWDId24zzcCvWPCK/Xq2w2q6tXryoWixlvkCpweXnZJJSYR1Blokvn75ngolxhMswGARITNzKtND+70WiYX6okcxDCwYjJ/NTUlPnFUg3xuxmsOZkCrMem5T85OdHOzo4ymYwZwEDBovLFPzYWi5l5BrQyWlzaTjxTJdkueOkRxIGklAqUa8VABu0/QxYStd/vN/iEZIKKp9c7962tVCqSZB0E14SqmusJFYopN8MmBmVUq/yRZFANBxKmM3QY3CeJREKjo6O2kZODgOEl3gAcqjA5kJ6yN6q/v99UZhMTE7aqPJVKaWxszJb+4f0qySpwjE4kGQTF9tqJiQkbNIGTsg2hUqmoVCqpUChYx/N9DbdCfQpxdnam1dVVlUolTU9P65VXXlE2m7VkyDDJqQDigebBAOhneAJnERkgyiUI3rjkY6OHWoeq18kakGTVCRtE2cHE1PdxWzPZRxQOh1Wr1SQ9WunM1zLtl2RtIbCEJBsQHRwcGG0Hezc2iDLIg0IGdktFyWFBQsKJy4mH8rolWWvt1IPTCeDSRBsL35UBINe7Wq2a4olhmM/nUywWM5L64eGhSXWBG3q9nlW0oVDIBkhwgCWZByqcUQjz4MAkcFaGIICA4oVEFZoYf89kngoaGIHXz+fNwUHiBzNmqAauGovFbOtCNBpVJpOxe3lzc1OFQkF37tyxpZhuPDncCvVPiGg0qtnZWc3NzSkej9vGzVarpVKpZID/zs6ONjY2rMVi2k+iaDabRvcBr6Odp0LCBQmp6f7+vkkdwb/A3Ri8jI6OWhVG2w9xHd29s82LxWJmFg0zAYNo6EKsfOHrnO0gk29aTLBDBl5UyEzmSRYYKiNfxQQF31AqZfxeSeashoGyQ7XJEkaSN1UltDaWB7KYb3NzU/V63eAav99vhxta/l6vZ1UlQyVadNyagHCOj481OTlpPFBEGyQpDqbe/1rcAU1A7RoaGrIVNXQwgUDAlFIkeKcEGtnv/v7+BRUYtn/AFE7sOhaLmaQW02kOuI2NDdXrdeXzeRtSufEovlduU99mIKt78cUXlcvllEwmNTg4qFarZQbDUFpo8eEwUmHQtne7XfO0BJvlgaVNwyHJ6SDEUAJ6DQ821TEPIg+9JHvwGWQwXSbJOjmwqGh4eEulkrXguHRJMh4nyjHWA1NxhkIhSbJ2looZgQFYMtc1GAwahEK7SmJmEh4MBs0V6ejoSJVKxaowWu5er2eS1tHRUeNS+v1+8y2o1Wra3d01Uj3VLE5j+DZgi4hJNoO2gYEBqzJhUtDm4zuL4xKfFd8HZAR17MtiBz5HEi6DULx9+ayBHahAOeCh942NjZm3A0YvXq/X9tgjRsDg5fs6qf9Dwk2o30JgbHHt2jUlk0mdnJzYhlIMQHA3qlarltyoWtFuQ8Znys0DwvCLwRC+rGjHO52OGXV4PB5rv3F1wviaag4SvvRIVkkix8AaviZyWoxQkCAy7ZVkq4tR4oCvYuYB3YvBBvStYrFopsMkBQw+SHwkYWSsfv+5+z8HC+8LxREJDbyTISE8VKAREhxDwOPjYzNMpnNAOODkngYCAaN6IYKo1WqGTyOwkGTSXCAghm7tdtv4vKjG+FoYIVDBoFwxpEJVRcVM4k4mkwqHwwbtwBkeGBiw7b44UbXbbeXzeT148EC1Ws0dLv0R4SbUbzFITpFIRNPT05qZmTFndnwwGc54PB5Vq1UznmACL8lwWKhVTvs/hkU8tI1GwwZlTpyLgcrh4aH5DEBvwQwY+hEcU1pMp8k2eKbTV9OpuCmVShocHLSfSfvp9Xo1MTFhQy+GSphBU4FRhUHzYTAESR/jZKpaGBjsO+I10wHgzuS0XARbjEajVo1zranKseULBoN2GOGdymHH30E7Oz4+NrNrfg8bGXhfJFmwU7YbkMQ45Kh0j46OFAwGzQQcGfP09LTR4ficoYEhxWWwKT1y76LKbzQaqtVqKhaLxmt2K9E/PtyE+mcKj+d8GVkymVQ6nVYmk7E2jwRVqVQutHq9Xu+CEw80JT4qWvtWq2W4KMocJJQMtUhqgUDA5ISoZwiEBdvb2xoeHrYqEa4j3gYMOjBY2d/f1/DwsPr7+w3zSyQSKhQKOjs7UywWu6D5xq291WrZdWHVNFjfl7msYL4HBwfmlAXt5+joSJOTkwqHwzYFBwtlAOhcecJ7oPUHOgGWkWTdAdZzx8fHth4E7jA4N4mSnzE2NnbBzHxoaMiuNV0Er4c2Hy9ajLcZRO7t7SmRSFilTqsPEwPDG7ogxBcML6mm19fXzaQH28rv83T+aYWbUJ+BwNYvnU4rkUgYpjg2NmYDJaq4Uqlk38OUGs4pratz6ru3t2et8cDAgEqlknw+nwqFggYHB23ogBcnSYp9QxDiWarnXM/h9Ds9PT1VoVDQ5OSk0cL4N5YLQtPBQITKkFUrTgPmdrttyYbqF2yUa0KlyXukQmMYMz4+bvg0UlnaejBkKn/oXlCS2NhAq07FzXVhUy2HzPb2tg4PDxWPx00kwTAM31KSsPRoN4Py5n4AAAxtSURBVBPiD5yaIPcPDg6qXC6b5BMaU7vdViwWs8MDTBSaFJgpkudAIGA47f7+vpaXl5XP57W9ve0m0G8g3IT6jIVzcj8zM6NkMmka+nQ6bRUYbSVyy3K5bI5NgUDAuIhbW1u2wsTr9Wpra8vI61QwWNuBx0E5YsMpSh94n1BxUB6x7JCqd3x83JIhevmhoaEL7kVs3IRbiWn0wMCAJQS//9w0mq/BU5ZKvtfrmVqLCoxhlaQL2CtEfWCRcrlsrlVM+xkcNRoN8zQg+Fpacl5XJBJRMplUp9Mxj9nT01PzvIUREIvF5PF4jL6EsECSDRiBBBiIcSCOjo4qHo+bE9nExIT6+vpsJXcwGDR8lQPA4/FobW1Ni4uLJrslubvxzYWbUJ+DYMCRSqU0MTFhbufwOf1+v7a3t23Cjomw0+WK6bYkG5JgLYiNH+T9crlsiZqfD66GWTUUJGSQJNJKpWLDLwYqk5OTNvTY3NzU/v6+wuGwEomEWfrRnmN0DL3r8PBQ5XLZ3KOQP7L2g4GdJJPFgp2SZBgwgUdTLVP54l9QrVbNvAXcFpx4YGDAzEDYVYWjF3Q3VGFUvaOjo1at0nrn8/kLaqdQKGTrSlCv8Xpx9RoaGlImkzEvAUxZcN+CcrW8vKxisWjrX57FraDf9XCJ/c9BdLtdlUol8zplmo37D7gf1anH49HMzIwODg5sJQlDpdHRUbOW83g8RjwnccAcYHMl/w/NCDMOzEmSyaTGxsZUqVSMQ4usFi6l8/cwYQf/RZmFRR3G1Bgxg7VSUdOGk7SY8oO/cr3AF09OTgxqwOGJrwO/hXLl8/kMawXGgIo0Pj6uYDCoZrNpVS94ZbVa1cnJibXmfr9fm5ubOj4+tiSP2i2TyRijAYw0m83aDirocgy4UIcx8a9Wq7br/vT0VCsrK0ZtgnPsxrMXbkJ9RsMpa2y329Zmoo1n1TKG0FSZkPYxowYygPNIpQOtq9lsWhLEJGV3d1cbGxvq9Xq2aRU3f0lGrgdjhdqDcTKJDijh6OjIKkKs7LDFy2Qytg651+tZ5e31eq3yIomhUIIojzSSvVJO9y+neQeeoAgEnK0/gzom+HjjMthBNBAIBJRIJAzXjUQi9jlgqMLXsc+KQwE+Loq3L9PNwGAfPnyoW7dumeE0PrJuPD/htvzPcSA9lc45sNFo1Aj1IyMjmpubsyThXHMNdQjDYfiUuDm1222rkBg0pdNpMxDx+/3mLTA2Nqbx8XHl83kblND2YmkInxPMV3qEIUciEW1vb6tQKNg2A6hRsVjMLAOj0agkWXvP0CabzdomVgy8ncsDYTWAleJrMDo6alN1TG4ODg702muvKZFI2IpspJn4ofLaoZUBW1D9Dg4OGkTSarWsmkQYgCUflLhms2n8XTxL3Xj2w8VQv4cBLxQ6FTgn0+3Z2VkdHBwYBspCNSzgWD4XDoc1MzOjzc1NMyWmdYfWJMkGWuygwq2KXUl8LetQnP6xmDf39/er1WopkUgol8tpfX3dWAbACGwXoG2m4m00GmYTiF0eFKZcLiefz6eHDx+adR7YptPVP5PJaGpqStvb2xds8sBKneICKFbOLaZwXqF5gfFiFYiAw43nO9yE6oakR7ZwYJrwHKPRqOLxuCQZFYkEeHZ2pnQ6bZgoeOfIyIhSqZRKpZLxKrlnYAlMTEyYq5Rzw6VT44/MkaqRNSxABshtUYrBI4Vrim0fWCbqMnxSy+WyZmZmNDQ0ZIosViBjYCLJhnaSDMtEbEFVifKKoVGr1VKhUDAGxJf/uPHdDDehuvEHB/xIjI09Ho8ikYjS6bTxHklEVI54gFK50b5Cm8L1itabIRjYq3ONjHTuXrWxsXEhsUFlwtaQoRmQxtnZmVWvtM6sbwYOoboslUqGx355mWG73TZvWgxt+FrcrlzXpe93uAnVjT9LQMtCT0+Sxj7v8PDQ2mX4piQwV1vuxrMabkJ1ww033HhK8aS06f2WX4cbbrjhxnc23ITqhhtuuPGUwk2obrjhhhtPKdyE6oYbbrjxlMJNqG644YYbTynchOqGG2648ZTCNUdxww033PgDor+/X8Fg0HxsHxduQnXDDTfceEKwyWJ2dlY3btzQO++8Y0Y9j/16l9jvhhtuuPEo2IybTqf1ox/9SNevX1csFtP6+rrW1tb0+eef61e/+tVjv/crEyorMNxwww03vsvBup5cLqe/+Iu/0FtvvaXJyUnt7u7q1q1b+u1vf6vPPvtM9Xpd0teUnv7DP/yDFhYWVCgU/h+3IDfccMON5zn6+/sVj8eVSqX0xhtv6K233tLAwICKxaJWV1e1srKimzdvql6v/z++El8roc7OzmpmZkaXL1/Wzs6OisWi1tbWVCqVzFDXDTfccON5CDYqTExM6Nq1a3rllVf00ksvaWJiQvv7+7p3755+9atf6cGDB6pWq19pv/i1zVFwCJqbm9O1a9eUSCS0vb2tfD6vWq2mWq1mC8PccMMNN561GBwc1NzcnGZnZ/XKK6/o6tWrikQiWl9f1/vvv6/79+/r3r172tnZ+YOLxKfiNsUOnqGhIb3yyiuanZ1Vf3+/qtWqSqWS1tfXVa/XbYeRG2644ca3Hfj5RqNRvf766/rhD3+o1157TaOjo3rw4IE+++wz/frXv1axWFStVvtav+Mbse/zer2amprS9PS0rly5ojfeeEM+n093797VJ598okKhoI2NDXdHuBtuuPGNhtfrVTwe1/z8vN5++21du3ZNPp9PzWZT+XxeDx8+1MrKin7/+98/FZ/db9wPtb+/X2NjY5qentbrr7+uXC6n8fFxNZtN3b9/X3fv3lW5XFaj0XATrBtuuPEnB7vRrl69qtdff10vvfSSQqGQ2u22lpeX9fHHH+vDDz/U9vb2Ux+of+sG02NjY8pkMpqfn9dPfvITTU9Pq9Fo6MGDB1pZWdFnn32mra0td/eOG2648UeF1+tVNpvVj370I73xxhuKRqOWU8rlslZXV1WtVr/R1/Bnc+xn3/n4+LiuX7+uubk5zczMyO/3q1wu6/79+1pcXLR1ui41yw033HCu7ZakYDBoe81mZmY0ODio3d1dFQoF3b9/X+12+1vNHc/MChS/3694PK5oNKp0Oq25uTnlcjmdnZ1pfX1d6+vrWlxc1Pr6ujqdjrrdrlvBuuHGdzzYwuv3+3Xp0iVdvXpVh4eHOj4+VjAY1MDAgAKBgPL5vFZXV1Wv19Vut/9sueGZSahf/vler1eDg4OanZ3VpUuXdPnyZSWTSe3t7Wl7e1v37t1TPp9XPp9Xu93+Rl+PG2648e2Fz+fT2NiY4vG4rl69qsuXL9vzX6vV9PHHH2tra0uVSkV37961dePPQoH1TCbUx8XQ0JDC4bCi0ah+/OMfK5PJqNfr6eDgQI1GQ/l8Xnfu3FGz2dTx8bFbwbrhxnMSXq9XQ0NDikQi+uEPf6hcLqd4PK7JyUklEglVKhV98skn+vTTT/Xw4cNnWp353CTUL4fX61UkElEulzOJ2PDwsBqNhiqViprNplZXV7WxsaF6ve4mVzfceIbC5/Mpm83q0qVLmpmZ0euvv64rV64oEAgYjWlpaUmrq6taXl5+blaHP7cJ9csxNjamwcFBJRIJvfXWW7p8+bJ8Pp92d3e1vb2tzc1NLSwsqF6v6/DwUPv7+67IwA03vuHw+XwaHh7W0NCQ8UHn5+cViUQ0MjKikZER7e7uamlpSZ999pkWFxe1u7ur3d3d5yaJOuM7k1C/HF6vV4lEQqlUSjMzM3r11VeVTCbV39+vRqNhQ66trS3l83m1Wi2XquWGG39CkBcCgYByuZxmZmaUSCT0yiuvKJPJaGhoSHt7e9rY2LDkubW1pWq1+p3hoH9nE6oz8B3o6+tTPB7Xiy++qFQqpVgspng8bnhMvV7Xhx9+qM3NTe3t7Wlvb0+7u7uu2Ysbbjwm/H6/QqGQ/Xn55Zf1wgsvKBKJyOfzqb+/X/V6Xevr68rn87p9+7a2trZ0fHys4+Pj72Tx8r1IqE+KwcFBJZNJjY6O6vLly8rlcpqYmNDIyIj8fr8ODg60tLSktbU1raysaHV19blsQ9xw42mEz+dTOp1WLpdTIpFQLpfT9PS0pqenJZ3nhVqtpps3b+qTTz5RvV5Xo9HQ9vb2dzJ5Pi6+VkJ1ww033HDjDw9366kbbrjhxlMKN6G64YYbbjylcBOqG2644cZTCjehuuGGG248pXATqhtuuOHGUwo3obrhhhtuPKX4v7RldiC0XvQfAAAAAElFTkSuQmCC\n"},"metadata":{}}],"execution_count":17},{"cell_type":"code","source":"class to_tensor_albu:\n    def __init__(self):\n        transformation = [Normalize(), ToTensorV2()]\n        self.transform = Compose(transformation)\n\n    def __call__(self, x):\n        return self.transform(image=x)['image']\n\nclass SeriesDataset(Dataset):\n    def __init__(self, images, labels=None):\n        self.images = images\n        self.labels = labels\n        self.to_tensor = to_tensor_albu()\n        \n    def __len__(self):\n        return len(self.images)\n\n    def __getitem__(self, idx):\n        image = self.images[idx]\n        image = self.to_tensor(image)\n        if self.labels is not None:\n            return image, self.labels[idx]\n        return image\n\nclass CatEmbeddingDataset(Dataset):\n    def __init__(self, images):\n        self.input = images\n        self.sl = 31\n        self.images = []\n        for i in range(len(images) - self.sl + 1):\n            self.images.append(images[i:i+31].unsqueeze(0))\n        self.images = torch.cat(self.images)\n        \n    def __len__(self):\n        return len(self.images - 30)\n\n    def __getitem__(self, idx):\n        image = self.images[idx]\n        return image        \n\ndef batch(iterable, n=1):\n    l = len(iterable)\n    for ndx in range(0, l, n):\n        yield iterable[ndx:min(ndx + n, l)]","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-26T22:59:58.078553Z","iopub.execute_input":"2025-04-26T22:59:58.078872Z","iopub.status.idle":"2025-04-26T22:59:58.089748Z","shell.execute_reply.started":"2025-04-26T22:59:58.078849Z","shell.execute_reply":"2025-04-26T22:59:58.08864Z"}},"outputs":[],"execution_count":18},{"cell_type":"code","source":"class NormSoftmax(nn.Module):\n    def __init__(self, in_features, out_features, temperature=1.):\n        super().__init__()\n        self.weight = nn.Parameter(torch.FloatTensor(in_features, out_features))\n        nn.init.xavier_uniform_(self.weight.data)\n        self.ln = nn.LayerNorm(in_features, elementwise_affine=False)\n        self.temperature = nn.Parameter(torch.Tensor([temperature]))\n\n    def forward(self, x):\n        x = self.ln(x)\n        x = torch.matmul(F.normalize(x), F.normalize(self.weight))\n        x = x / self.temperature\n        return x\n\nclass EfficientNet(nn.Module):\n    def __init__(self, name):\n        super().__init__()\n        backbone = timm.create_model(model_name=name, pretrained=False, in_chans=3)\n        self.conv_stem = backbone.conv_stem\n        self.bn1 = backbone.bn1\n        self.act1 = backbone.act1\n        for i in range(len(backbone.blocks)):\n            setattr(self, f\"block{i}\", backbone.blocks[i])\n        self.conv_head = backbone.conv_head\n        self.bn2 = backbone.bn2\n        self.act2 = backbone.act2\n        self.global_pool = SelectAdaptivePool2d(pool_type=\"avg\")\n        self.num_features = backbone.num_features\n        self.fc = nn.Linear(self.num_features, 7)\n        del backbone\n\n    def _features(self, x):\n        x = self.conv_stem(x)\n        x = self.bn1(x)\n        x = self.act1(x)\n        x = self.block0(x)\n        x = self.block1(x)\n        x = self.block2(x)\n        x = self.block3(x)\n        x = self.block4(x); b4 = x\n        x = self.block5(x); b5 = x\n        x = self.block6(x)\n        x = self.conv_head(x)\n        x = self.bn2(x)\n        x = self.act2(x)\n        return b4, b5, x\n\n    def forward(self, x):\n        with autocast():\n            b4, b5, x = self._features(x)\n            x = self.global_pool(x)\n            x = torch.flatten(x, 1)\n            logits = self.fc(x)\n            return logits","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-26T23:00:13.561213Z","iopub.execute_input":"2025-04-26T23:00:13.561528Z","iopub.status.idle":"2025-04-26T23:00:13.576272Z","shell.execute_reply.started":"2025-04-26T23:00:13.561502Z","shell.execute_reply":"2025-04-26T23:00:13.575379Z"}},"outputs":[],"execution_count":19},{"cell_type":"code","source":"class NewEmbeddingNet(nn.Module):\n    def __init__(self):\n        super().__init__()\n        # Minimal version to isolate error\n        self.fc = nn.Linear(10, 1)\n        # Original EmbeddingNet commented out until error is resolved\n        self.fw = nn.Sequential(\n            nn.Conv2d(1, 32, 7, 1, 3),\n            nn.Dropout(p=0.2),\n            nn.ReLU(inplace=True),\n            nn.Conv2d(32, 64, 5, 1, 2),\n            nn.Dropout(p=0.2),\n            nn.ReLU(inplace=True),\n            nn.Conv2d(64, 128, 3, 1, 1),\n            nn.Dropout(p=0.2),\n            nn.ReLU(inplace=True),\n            nn.Conv2d(128, 256, 3, 1, 1),\n            nn.Dropout(p=0.2),\n            nn.ReLU(inplace=True),\n            SelectAdaptivePool2d(pool_type=\"avg\"),\n            nn.Flatten(),\n            nn.Linear(256, 7)\n        )\n        self.conv_out = nn.Linear(7, 1)\n        self.lstm1 = nn.LSTM(7, 128, bidirectional=True, batch_first=True)\n        self.lstm2 = nn.LSTM(128 * 2, 128, bidirectional=True, batch_first=True)\n        self.linear1 = nn.Linear(128*2, 128*2)\n        self.linear2 = nn.Linear(128*2, 128*2)\n        self.linear = nn.Linear(128*2, 7)\n        self.lstm_out = nn.Linear(7, 1)\n        \n\n    def forward(self, x):\n        with autocast():\n            return self.fc(x)\n            \n            embedding_vector = x.squeeze(1).float()\n            logits1 = self.fw(x)\n            second_logits1 = self.conv_out(logits1)\n            h_lstm1, _ = self.lstm1(embedding_vector)\n            h_lstm2, _ = self.lstm2(h_lstm1)\n            h_conc_linear1 = F.relu(self.linear1(h_lstm1))\n            h_conc_linear2 = F.relu(self.linear2(h_lstm2))\n            hidden = h_lstm1 + h_lstm2 + h_conc_linear1 + h_conc_linear2\n            logits2 = self.linear(hidden)[:,-1,:]\n            second_logits2 = self.lstm_out(logits2)\n            return logits1, logits2, second_logits1, second_logits2\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-26T23:02:37.119282Z","iopub.execute_input":"2025-04-26T23:02:37.119592Z","iopub.status.idle":"2025-04-26T23:02:37.132032Z","shell.execute_reply.started":"2025-04-26T23:02:37.119566Z","shell.execute_reply":"2025-04-26T23:02:37.131192Z"}},"outputs":[],"execution_count":25},{"cell_type":"code","source":"\nclass SeriesEmbeddingNet(nn.Module):\n    def __init__(self):\n        super().__init__()\n        self.netA = timm.create_model(model_name=\"tf_efficientnet_b0\", pretrained=False, in_chans=1)\n        self.netA.classifier = NormSoftmax(self.netA.num_features, 9)\n        self.netB = timm.create_model(model_name=\"tf_efficientnet_b0\", pretrained=False, in_chans=1)\n        self.netB.classifier = NormSoftmax(self.netB.num_features, 9)\n        self.netC = timm.create_model(model_name=\"tf_efficientnet_b0\", pretrained=False, in_chans=1)\n        self.netC.classifier = NormSoftmax(self.netC.num_features, 9)\n    \n    def forward(self, x):\n        with autocast():\n            logits1 = self.netA(x)\n            logits2 = self.netB(x)\n            logits3 = self.netC(x)\n            return (logits1 + logits2 + logits3) / 3.","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-26T23:02:39.63699Z","iopub.execute_input":"2025-04-26T23:02:39.637309Z","iopub.status.idle":"2025-04-26T23:02:39.644851Z","shell.execute_reply.started":"2025-04-26T23:02:39.637283Z","shell.execute_reply":"2025-04-26T23:02:39.643829Z"}},"outputs":[],"execution_count":26},{"cell_type":"code","source":"try:\n    class MinimalNet(nn.Module):\n        def __init__(self):\n            super().__init__()\n    device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n    minimal_model = MinimalNet().to(device)\n    print(\"Minimal nn.Module (MinimalNet) initialized successfully\")\n    del minimal_model\nexcept Exception as e:\n    print(f\"Failed to initialize MinimalNet: {e}\")\n\n# Test nn.Module with Linear layer\ntry:\n    class TestNet(nn.Module):\n        def __init__(self):\n            super().__init__()\n            self.fc = nn.Linear(10, 1)\n    test_model = TestNet().to(device)\n    print(\"nn.Module with Linear layer (TestNet) initialized successfully\")\n    del test_model\nexcept Exception as e:\n    print(f\"Failed to initialize TestNet: {e}\")\n\n# Test NewEmbeddingNet (renamed from EmbeddingNet)\ntry:\n    embedding_net = NewEmbeddingNet().to(device)\n    print(\"NewEmbeddingNet initialized successfully\")\n    del embedding_net\nexcept Exception as e:\n    print(f\"Failed to initialize NewEmbeddingNet: {e}\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-26T23:02:47.66856Z","iopub.execute_input":"2025-04-26T23:02:47.668886Z","iopub.status.idle":"2025-04-26T23:02:48.293449Z","shell.execute_reply.started":"2025-04-26T23:02:47.668855Z","shell.execute_reply":"2025-04-26T23:02:48.292602Z"}},"outputs":[{"name":"stdout","text":"Minimal nn.Module (MinimalNet) initialized successfully\nnn.Module with Linear layer (TestNet) initialized successfully\nNewEmbeddingNet initialized successfully\n","output_type":"stream"}],"execution_count":27},{"cell_type":"code","source":"try:\n    import torch.nn as nn\n    print(\"torch.nn imported successfully\")\n    print(f\"nn.Module.__init__ exists: {hasattr(nn.Module, '__init__')}\")\nexcept Exception as e:\n    print(f\"Failed to import or inspect torch.nn: {e}\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-26T23:02:54.561674Z","iopub.execute_input":"2025-04-26T23:02:54.56199Z","iopub.status.idle":"2025-04-26T23:02:54.56668Z","shell.execute_reply.started":"2025-04-26T23:02:54.561962Z","shell.execute_reply":"2025-04-26T23:02:54.565882Z"}},"outputs":[{"name":"stdout","text":"torch.nn imported successfully\nnn.Module.__init__ exists: True\n","output_type":"stream"}],"execution_count":28},{"cell_type":"code","source":"# Check CUDA availability\ndevice = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n\n# Load models (skip EmbeddingNet until initialization fixed)\ntry:\n    backbone_b4f0 = EfficientNet(\"tf_efficientnet_b4\").to(device)\n    checkpoint = torch.load(\"../input/rsna-b4-folds/b4_sz512_fold0.pth\", map_location=\"cpu\").pop('state_dict')\n    checkpoint_surged = OrderedDict()\n    for k, v in checkpoint.items():\n        if \"second_\" not in k:\n            checkpoint_surged[k.replace(\"module.\",\"\")] = v\n    backbone_b4f0.load_state_dict(checkpoint_surged)\n    backbone_b4f0.eval()\n    print(\"backbone_b4f0 loaded successfully\")\n    del checkpoint, checkpoint_surged\nexcept Exception as e:\n    print(f\"Failed to load backbone_b4f0: {e}\")\n\ntry:\n    backbone_b4f1 = EfficientNet(\"tf_efficientnet_b5\").to(device)\n    checkpoint = torch.load(\"../input/rsnastrweights/b5_sz512_fold0.pth\", map_location=\"cpu\").pop('state_dict')\n    checkpoint_surged = OrderedDict()\n    for k, v in checkpoint.items():\n        if \"second_\" not in k:\n            checkpoint_surged[k.replace(\"module.\",\"\")] = v\n    backbone_b4f1.load_state_dict(checkpoint_surged)\n    backbone_b4f1.eval()\n    print(\"backbone_b4f1 loaded successfully\")\n    del checkpoint, checkpoint_surged\nexcept Exception as e:\n    print(f\"Failed to load backbone_b4f1: {e}\")\n\n# Skip EmbeddingNet loading\ntry:\n    embeddingnet_b4f0 = NewEmbeddingNet().to(device)  # Update to NewEmbeddingNet when ready\n    checkpoint = torch.load(\"../input/rsna-b4-folds/cnn_lstm_embeddings_b4_sz512_fold0.pth\", map_location=\"cpu\").pop('state_dict')\n    checkpoint_surged = OrderedDict()\n    for k, v in checkpoint.items():\n        checkpoint_surged[k.replace(\"module.\",\"\")] = v\n    embeddingnet_b4f0.load_state_dict(checkpoint_surged)\n    embeddingnet_b4f0.eval()\n    print(\"embeddingnet_b4f0 loaded successfully\")\n    del checkpoint, checkpoint_surged\nexcept Exception as e:\n    print(f\"Failed to load embeddingnet_b4f0: {e}\")\n\ntry:\n    embeddingnet_b4f1 = NewEmbeddingNet().to(device)  # Update to NewEmbeddingNet when ready\n    checkpoint = torch.load(\"../input/rsnastrweights/cnn_lstm_embeddings_b5_sz512_fold0.pth\", map_location=\"cpu\").pop('state_dict')\n    checkpoint_surged = OrderedDict()\n    for k, v in checkpoint.items():\n        checkpoint_surged[k.replace(\"module.\",\"\")] = v\n    embeddingnet_b4f1.load_state_dict(checkpoint_surged)\n    embeddingnet_b4f1.eval()\n    print(\"embeddingnet_b4f1 loaded successfully\")\n    del checkpoint, checkpoint_surged\nexcept Exception as e:\n    print(f\"Failed to load embeddingnet_b4f1: {e}\")\n\ntry:\n    seriesnet_b4f0 = SeriesEmbeddingNet().to(device)\n    checkpoint = torch.load(\"../input/rsna-b4-folds/triple_b0_series_b4_sz512_fold0.pth\", map_location=\"cpu\").pop('state_dict')\n    checkpoint_surged = OrderedDict()\n    for k, v in checkpoint.items():\n        checkpoint_surged[k.replace(\"module.\",\"\")] = v\n    seriesnet_b4f0.load_state_dict(checkpoint_surged)\n    seriesnet_b4f0.eval()\n    print(\"seriesnet_b4f0 loaded successfully\")\n    del checkpoint, checkpoint_surged\nexcept Exception as e:\n    print(f\"Failed to load seriesnet_b4f0: {e}\")\n\ntry:\n    seriesnet_b4f1 = SeriesEmbeddingNet().to(device)\n    checkpoint = torch.load(\"../input/rsnastrweights/triple_b0_series_b5_sz512_fold0.pth\", map_location=\"cpu\").pop('state_dict')\n    checkpoint_surged = OrderedDict()\n    for k, v in checkpoint.items():\n        checkpoint_surged[k.replace(\"module.\",\"\")] = v\n    seriesnet_b4f1.load_state_dict(checkpoint_surged)\n    seriesnet_b4f1.eval()\n    print(\"seriesnet_b4f1 loaded successfully\")\n    del checkpoint, checkpoint_surged\nexcept Exception as e:\n    print(f\"Failed to load seriesnet_b4f1: {e}\")\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-26T23:09:56.657102Z","iopub.execute_input":"2025-04-26T23:09:56.657408Z","iopub.status.idle":"2025-04-26T23:09:59.612417Z","shell.execute_reply.started":"2025-04-26T23:09:56.657382Z","shell.execute_reply":"2025-04-26T23:09:59.611646Z"}},"outputs":[{"name":"stdout","text":"backbone_b4f0 loaded successfully\nbackbone_b4f1 loaded successfully\nFailed to load embeddingnet_b4f0: Error(s) in loading state_dict for NewEmbeddingNet:\n\tMissing key(s) in state_dict: \"fc.weight\", \"fc.bias\". \nFailed to load embeddingnet_b4f1: Error(s) in loading state_dict for NewEmbeddingNet:\n\tMissing key(s) in state_dict: \"fc.weight\", \"fc.bias\". \nseriesnet_b4f0 loaded successfully\nseriesnet_b4f1 loaded successfully\n","output_type":"stream"}],"execution_count":32},{"cell_type":"code","source":"BATCH_SIZE = 32\ncriterion = nn.BCEWithLogitsLoss()\noptimizer_b4f0 = optim.Adam(backbone_b4f0.parameters(), lr=1e-4)\noptimizer_b4f1 = optim.Adam(backbone_b4f1.parameters(), lr=1e-4)\n\n# Training loop for backbone_b4f0\nbackbone_b4f0.train()\nfor study in tqdm(train_studies['StudyInstanceUID'].unique()):\n    try:\n        stack_image, instance_list = convert(study, is_train=True)\n        if len(instance_list) == 0:\n            print(f\"Skipping study {study}: No valid DICOM files\")\n            continue\n        study_labels = train_subset[train_subset['StudyInstanceUID'] == study]\n        labels = torch.tensor(study_labels['pe_present_on_image'].values, dtype=torch.float32).to(device)\n        dataset = SeriesDataset(stack_image, labels)\n        dataloader = DataLoader(dataset, batch_size=BATCH_SIZE, shuffle=True)\n        \n        for images, targets in dataloader:\n            images = images.to(device).float()\n            optimizer_b4f0.zero_grad()\n            with autocast():\n                outputs = backbone_b4f0(images)\n                loss = criterion(outputs, targets.unsqueeze(1))\n            loss.backward()\n            optimizer_b4f0.step()\n    except Exception as e:\n        print(f\"Warning: Failed to train on study {study}: {e}\")\n        continue\nbackbone_b4f0.eval()\n\n# Training loop for backbone_b4f1\nbackbone_b4f1.train()\nfor study in tqdm(train_studies['StudyInstanceUID'].unique()):\n    try:\n        stack_image, instance_list = convert(study, is_train=True)\n        if len(instance_list) == 0:\n            print(f\"Skipping study {study}: No valid DICOM files\")\n            continue\n        study_labels = train_subset[train_subset['StudyInstanceUID'] == study]\n        labels = torch.tensor(study_labels['pe_present_on_image'].values, dtype=torch.float32).to(device)\n        dataset = SeriesDataset(stack_image, labels)\n        dataloader = DataLoader(dataset, batch_size=BATCH_SIZE, shuffle=True)\n        \n        for images, targets in dataloader:\n            images = images.to(device).float()\n            optimizer_b4f1.zero_grad()\n            with autocast():\n                outputs = backbone_b4f1(images)\n                loss = criterion(outputs, targets.unsqueeze(1))\n            loss.backward()\n            optimizer_b4f1.step()\n    except Exception as e:\n        print(f\"Warning: Failed to train on study {study}: {e}\")\n        continue\nbackbone_b4f1.eval()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-26T23:04:37.440953Z","iopub.execute_input":"2025-04-26T23:04:37.441265Z","iopub.status.idle":"2025-04-26T23:07:09.227656Z","shell.execute_reply.started":"2025-04-26T23:04:37.44124Z","shell.execute_reply":"2025-04-26T23:07:09.225806Z"}},"outputs":[{"name":"stderr","text":"  0%|          | 1/240 [00:02<08:39,  2.17s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study a96af4056d5d: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":"  1%|          | 2/240 [00:06<11:43,  2.96s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study 4d07e1836d74: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":"  1%|▏         | 3/240 [00:13<16:30,  4.18s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study bed6309efd9c: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":"  2%|▏         | 4/240 [00:19<17:34,  4.47s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study 8617e3568481: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":"  2%|▏         | 5/240 [00:24<18:08,  4.63s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study 945d99d8682e: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":"  2%|▎         | 6/240 [00:28<17:18,  4.44s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study bb6ea46e5283: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":"  3%|▎         | 7/240 [00:31<15:50,  4.08s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study d1e3830db220: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":"  3%|▎         | 8/240 [00:35<15:40,  4.05s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study a52d115d9dde: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":"  4%|▍         | 9/240 [00:40<16:41,  4.34s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study 37d8f781387c: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":"  4%|▍         | 10/240 [00:45<17:10,  4.48s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study 9d223375dac7: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":"  5%|▍         | 11/240 [00:49<17:11,  4.50s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study bc6ff1c5061b: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":"  5%|▌         | 12/240 [00:55<18:27,  4.86s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study 5fab033df67e: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":"  5%|▌         | 13/240 [01:00<18:29,  4.89s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study fff1ef450040: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":"  6%|▌         | 14/240 [01:05<18:26,  4.89s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study 66e7a2394020: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":"  6%|▋         | 15/240 [01:10<18:23,  4.91s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study b82680f6d54e: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":"  7%|▋         | 16/240 [01:14<18:02,  4.83s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study 3509afe592f4: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":"  7%|▋         | 17/240 [01:18<16:47,  4.52s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study e32b54b94548: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":"  8%|▊         | 18/240 [01:22<15:59,  4.32s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study 35fe923f0708: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":"  8%|▊         | 19/240 [01:27<16:39,  4.52s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study 41feadf64832: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":"  8%|▊         | 20/240 [01:32<16:58,  4.63s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study 8f7cb4366552: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":"  9%|▉         | 21/240 [01:37<17:01,  4.67s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study c54ed30698a0: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":"  9%|▉         | 22/240 [01:41<17:07,  4.71s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study 96cd03b7e5b4: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":" 10%|▉         | 23/240 [01:46<17:00,  4.70s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study 565f32efb2f9: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":" 10%|█         | 24/240 [01:57<23:04,  6.41s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study ac16092ad220: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":" 10%|█         | 25/240 [02:01<21:23,  5.97s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study 44241eb40c01: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":" 11%|█         | 26/240 [02:06<20:01,  5.62s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study cb7c060e24b5: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":" 11%|█▏        | 27/240 [02:08<15:59,  4.51s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study af234b50ae62: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":" 12%|█▏        | 28/240 [02:12<15:33,  4.41s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study 9c661733ee10: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":" 12%|█▏        | 29/240 [02:23<22:01,  6.26s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study 396494573e34: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":" 12%|█▎        | 30/240 [02:28<20:40,  5.91s/it]","output_type":"stream"},{"name":"stdout","text":"Warning: Failed to train on study 96133ef5de06: name 'ToTensorV2' is not defined\n","output_type":"stream"},{"name":"stderr","text":" 12%|█▎        | 30/240 [02:31<17:42,  5.06s/it]\n","output_type":"stream"},{"traceback":["\u001b[0;31m---------------------------------------------------------------------------\u001b[0m","\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)","\u001b[0;32m<ipython-input-31-982f5ed208f5>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      8\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mstudy\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mtqdm\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrain_studies\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'StudyInstanceUID'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0munique\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      9\u001b[0m     \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 10\u001b[0;31m         \u001b[0mstack_image\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minstance_list\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mconvert\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mstudy\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mis_train\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     11\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minstance_list\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     12\u001b[0m             \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf\"Skipping study {study}: No valid DICOM files\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32m<ipython-input-7-9e83f55a49b1>\u001b[0m in \u001b[0;36mconvert\u001b[0;34m(dicom_path, is_train)\u001b[0m\n\u001b[1;32m      7\u001b[0m     \u001b[0mstack_zpos\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mzeros\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minstance\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      8\u001b[0m     \u001b[0;32mfor\u001b[0m \u001b[0mi\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mins\u001b[0m \u001b[0;32min\u001b[0m \u001b[0menumerate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minstance\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 9\u001b[0;31m         \u001b[0mf\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpydicom\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdcmread\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mos\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpath\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mjoin\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdicom_path\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mseries\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mins\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     10\u001b[0m         \u001b[0mimage\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpixel_array\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     11\u001b[0m         \u001b[0mimage\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mimage\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mastype\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mint16\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32m/opt/conda/lib/python3.7/site-packages/pydicom/filereader.py\u001b[0m in \u001b[0;36mdcmread\u001b[0;34m(fp, defer_size, stop_before_pixels, force, specific_tags)\u001b[0m\n\u001b[1;32m    869\u001b[0m     \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    870\u001b[0m         dataset = read_partial(fp, stop_when, defer_size=defer_size,\n\u001b[0;32m--> 871\u001b[0;31m                                force=force, specific_tags=specific_tags)\n\u001b[0m\u001b[1;32m    872\u001b[0m     \u001b[0;32mfinally\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    873\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mcaller_owns_file\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32m/opt/conda/lib/python3.7/site-packages/pydicom/filereader.py\u001b[0m in \u001b[0;36mread_partial\u001b[0;34m(fileobj, stop_when, defer_size, force, specific_tags)\u001b[0m\n\u001b[1;32m    666\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    667\u001b[0m     \u001b[0;31m# Read preamble (if present)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 668\u001b[0;31m     \u001b[0mpreamble\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mread_preamble\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfileobj\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mforce\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    669\u001b[0m     \u001b[0;31m# Read any File Meta Information group (0002,eeee) elements (if present)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    670\u001b[0m     \u001b[0mfile_meta_dataset\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_read_file_meta_info\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfileobj\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32m/opt/conda/lib/python3.7/site-packages/pydicom/filereader.py\u001b[0m in \u001b[0;36mread_preamble\u001b[0;34m(fp, force)\u001b[0m\n\u001b[1;32m    603\u001b[0m     \"\"\"\n\u001b[1;32m    604\u001b[0m     \u001b[0mlogger\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdebug\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Reading File Meta Information preamble...\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 605\u001b[0;31m     \u001b[0mpreamble\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m128\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    606\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mconfig\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdebugging\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    607\u001b[0m         \u001b[0msample\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mbytes2hex\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpreamble\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;36m8\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0;34m\"...\"\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mbytes2hex\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpreamble\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m8\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;31mKeyboardInterrupt\u001b[0m: "],"ename":"KeyboardInterrupt","evalue":"","output_type":"error"}],"execution_count":31},{"cell_type":"code","source":"","metadata":{"trusted":true},"outputs":[],"execution_count":null}]}