{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.12.12","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"gpu","dataSources":[{"sourceId":10338,"databundleVersionId":862042,"sourceType":"competition"}],"dockerImageVersionId":31236,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":true}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"# **Set-Up**","metadata":{}},{"cell_type":"code","source":"%pip install torchio --q","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:42:39.039077Z","iopub.execute_input":"2026-01-14T13:42:39.039400Z","iopub.status.idle":"2026-01-14T13:42:43.952307Z","shell.execute_reply.started":"2026-01-14T13:42:39.039372Z","shell.execute_reply":"2026-01-14T13:42:43.951505Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import torch\nimport torch.nn as nn\nimport torch.nn.functional as F\nimport torch.optim as optim\nimport pytorch_lightning as pl\nimport torchio as tio \nimport torchvision\nimport torchvision.transforms as transforms\nfrom torch.utils.data import Dataset, DataLoader\n\nimport math\nimport os\nimport random\nimport numpy as np\nimport pandas as pd\nimport matplotlib.pyplot as plt\nfrom IPython.display import clear_output\nfrom tqdm.notebook import trange, tqdm\nfrom pathlib import Path\nfrom tqdm import tqdm\n\n%matplotlib inline","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:42:43.953806Z","iopub.execute_input":"2026-01-14T13:42:43.954063Z","iopub.status.idle":"2026-01-14T13:42:58.141927Z","shell.execute_reply.started":"2026-01-14T13:42:43.954034Z","shell.execute_reply":"2026-01-14T13:42:58.141106Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"# **Pre-Processing**","metadata":{}},{"cell_type":"code","source":"labels_df = pd.read_csv(\n    \"/kaggle/input/rsna-pneumonia-detection-challenge/stage_2_train_labels.csv\"\n)\n\nlabels_df = labels_df.groupby(\"patientId\")[\"Target\"].max().reset_index()\n\nprint(labels_df[\"Target\"].value_counts())\nprint(labels_df.info(verbose=True, show_counts=True))","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:42:58.142920Z","iopub.execute_input":"2026-01-14T13:42:58.143429Z","iopub.status.idle":"2026-01-14T13:42:58.241522Z","shell.execute_reply.started":"2026-01-14T13:42:58.143401Z","shell.execute_reply":"2026-01-14T13:42:58.240845Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"labels_df.head()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:42:58.243258Z","iopub.execute_input":"2026-01-14T13:42:58.243490Z","iopub.status.idle":"2026-01-14T13:42:58.261894Z","shell.execute_reply.started":"2026-01-14T13:42:58.243465Z","shell.execute_reply":"2026-01-14T13:42:58.261174Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"## Subjects","metadata":{}},{"cell_type":"code","source":"ROOT_PATH = Path(\"/kaggle/input/rsna-pneumonia-detection-challenge/stage_2_train_images/\")\npatient_dirs = list(ROOT_PATH.glob(\"*\"))\n\npatient_dirs[0] # debug","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:42:58.262677Z","iopub.execute_input":"2026-01-14T13:42:58.262910Z","iopub.status.idle":"2026-01-14T13:42:59.445215Z","shell.execute_reply.started":"2026-01-14T13:42:58.262886Z","shell.execute_reply":"2026-01-14T13:42:59.444441Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"def get_patient_label(patient_path: Path, labels_df: pd.DataFrame) -> int:\n    patientID = patient_path.stem\n    label = labels_df.loc[labels_df[\"patientId\"] == patientID, \"Target\"]\n    label = label.iloc[0] if not label.empty else None\n    return int(label)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:42:59.446132Z","iopub.execute_input":"2026-01-14T13:42:59.446457Z","iopub.status.idle":"2026-01-14T13:42:59.450795Z","shell.execute_reply.started":"2026-01-14T13:42:59.446429Z","shell.execute_reply":"2026-01-14T13:42:59.450061Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"subjects = []\nheights = []\nwidths = []\nlabels = []\n\nfor subject_path in tqdm(patient_dirs):\n    \n    img_path = subject_path\n    label = get_patient_label(subject_path, labels_df)\n\n    ct = tio.ScalarImage(img_path)\n    h, w, _ = ct.spatial_shape   \n\n    subject = tio.Subject(\n        CT = ct,\n        Label = torch.tensor(label, dtype=torch.long),\n        PatientID = subject_path.stem\n    )\n\n    subjects.append(subject)\n    heights.append(h)\n    widths.append(w)\n    labels.append(label)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:42:59.451731Z","iopub.execute_input":"2026-01-14T13:42:59.452085Z","iopub.status.idle":"2026-01-14T13:49:37.671783Z","shell.execute_reply.started":"2026-01-14T13:42:59.452058Z","shell.execute_reply":"2026-01-14T13:49:37.671050Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"print(type(subjects[15][\"CT\"]), subjects[15][\"CT\"])\nprint(type(subjects[15][\"Label\"]), subjects[15][\"Label\"])\nsubjects[15][\"CT\"].spatial_shape","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:49:37.672835Z","iopub.execute_input":"2026-01-14T13:49:37.673183Z","iopub.status.idle":"2026-01-14T13:49:37.711968Z","shell.execute_reply.started":"2026-01-14T13:49:37.673142Z","shell.execute_reply":"2026-01-14T13:49:37.711336Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Plot depth (dimensions)\nplt.figure(figsize=(18,5))\n\n# Plot height\nplt.subplot(1,2,1)\nplt.hist(heights, bins=20, color='lightgreen', edgecolor='black')\nplt.title(\"CT Height Distribution\")\nplt.xlabel(\"Height (pixels/voxels)\")\nplt.ylabel(\"Number of Subjects\")\n\n# Plot width\nplt.subplot(1,2,2)\nplt.hist(widths, bins=20, color='salmon', edgecolor='black')\nplt.title(\"CT Width Distribution\")\nplt.xlabel(\"Width (pixels/voxels)\")\nplt.ylabel(\"Number of Subjects\")\n\nplt.tight_layout()\nplt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:49:37.712745Z","iopub.execute_input":"2026-01-14T13:49:37.713359Z","iopub.status.idle":"2026-01-14T13:49:38.495264Z","shell.execute_reply.started":"2026-01-14T13:49:37.713329Z","shell.execute_reply":"2026-01-14T13:49:38.494515Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"img_size_og = subjects[15][\"CT\"].spatial_shape[0]","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:49:38.497313Z","iopub.execute_input":"2026-01-14T13:49:38.497817Z","iopub.status.idle":"2026-01-14T13:49:38.505368Z","shell.execute_reply.started":"2026-01-14T13:49:38.497753Z","shell.execute_reply":"2026-01-14T13:49:38.504669Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"## **Transforms**","metadata":{}},{"cell_type":"code","source":"process = tio.Compose([\n    tio.ToCanonical(),                        # step 1: fix orientation - RAS              \n    tio.RescaleIntensity((0, 1)),                      # step 2: normalize intensity\n    tio.Resize((356, 356, 1)),\n    tio.CropOrPad((256, 256, 1)),          \n])\n\naugmentation = tio.RandomAffine(scales=(0.9, 1.1), degrees=(-10, 10))\n\ntrain_transform = tio.Compose([process, augmentation])\nval_transform = tio.Compose([process])","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:49:38.506379Z","iopub.execute_input":"2026-01-14T13:49:38.506637Z","iopub.status.idle":"2026-01-14T13:49:38.519029Z","shell.execute_reply.started":"2026-01-14T13:49:38.506613Z","shell.execute_reply":"2026-01-14T13:49:38.518403Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"## **DataSet & DataLoader**","metadata":{}},{"cell_type":"code","source":"from sklearn.model_selection import train_test_split\n\n# (90/10)\ntrain_val_subjects, test_subjects = train_test_split(\n    subjects,\n    test_size=0.15,\n    stratify=labels,\n    random_state=42\n)\n\n# (80/20)\ntrain_subjects, val_subjects = train_test_split(\n    train_val_subjects,\n    test_size=0.2,\n    stratify=[s.Label.item() for s in train_val_subjects],\n    random_state=42\n)\n\n# Verify class distributions\ntrain_labels = [s.Label.item() for s in train_subjects]\nval_labels   = [s.Label.item() for s in val_subjects]\ntest_labels  = [s.Label.item() for s in test_subjects]\n\nprint(\"Train counts:\", np.bincount(train_labels))\nprint(\"Val counts:\", np.bincount(val_labels))\nprint(\"Test counts:\", np.bincount(test_labels))","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:49:38.519867Z","iopub.execute_input":"2026-01-14T13:49:38.520076Z","iopub.status.idle":"2026-01-14T13:49:38.625660Z","shell.execute_reply.started":"2026-01-14T13:49:38.520054Z","shell.execute_reply":"2026-01-14T13:49:38.624884Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"train_dataset = tio.SubjectsDataset(train_subjects, transform = train_transform) \nval_dataset = tio.SubjectsDataset(val_subjects, transform = val_transform)  \ntest_dataset = tio.SubjectsDataset(test_subjects, transform = val_transform)  ","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:52:29.967110Z","iopub.execute_input":"2026-01-14T13:52:29.967948Z","iopub.status.idle":"2026-01-14T13:52:29.976037Z","shell.execute_reply.started":"2026-01-14T13:52:29.967914Z","shell.execute_reply":"2026-01-14T13:52:29.975322Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"from typing import Tuple, List\n\ndef collate_subjects(batch: List) -> Tuple[torch.Tensor, torch.Tensor]:\n    images = torch.stack([s.CT.data.squeeze(-1) for s in batch])\n    labels = torch.tensor([s.Label.item() for s in batch])\n    return images, labels","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:52:30.256315Z","iopub.execute_input":"2026-01-14T13:52:30.257127Z","iopub.status.idle":"2026-01-14T13:52:30.261813Z","shell.execute_reply.started":"2026-01-14T13:52:30.257093Z","shell.execute_reply":"2026-01-14T13:52:30.261110Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"train_loader = DataLoader(train_dataset, batch_size=64, num_workers=4, collate_fn=collate_subjects, shuffle=True, pin_memory = True)\nval_loader = DataLoader(val_dataset, batch_size=64, num_workers=4, collate_fn=collate_subjects)\ntest_loader = DataLoader(test_dataset, batch_size=64, num_workers=4, collate_fn=collate_subjects)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:52:37.366646Z","iopub.execute_input":"2026-01-14T13:52:37.367297Z","iopub.status.idle":"2026-01-14T13:52:37.371937Z","shell.execute_reply.started":"2026-01-14T13:52:37.367251Z","shell.execute_reply":"2026-01-14T13:52:37.371286Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"x, y = next(iter(train_loader))\nprint(f\"Images shape fresh off the loader: {x.shape}\")\nprint(f\"Labels shape fresh off the loader: {y.shape}\")\nprint(f\"Labels corresponding to {y.shape[0]} images in the batch: \" + str(y))","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:52:37.599163Z","iopub.execute_input":"2026-01-14T13:52:37.599903Z","iopub.status.idle":"2026-01-14T13:52:48.927045Z","shell.execute_reply.started":"2026-01-14T13:52:37.599872Z","shell.execute_reply":"2026-01-14T13:52:48.926250Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"# **Model**","metadata":{}},{"cell_type":"markdown","source":"![vit.png](attachment:d6dbbcb0-b038-485c-99bc-353028cf4828.png)","metadata":{},"attachments":{"8d1a447c-dc89-45d0-8c88-d9d6d1c0f0b0.png":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAwAAAAJECAMAAACW4bNyAAAAnFBMVEXl5eV8enoWGBL6+voBAQELCwnNvdH////44uH406qnoaGtq61EZqFOcaw4NjWal5i4tbZjYmORjo1bWlrAv8JycHDv8PCGg4MjIx3Z2dnNy8svLidNTExEQkE1V5JSU1ZagLpvaWTo1NNfeafFq4pwh7FDQyntyaKClbnx3Nu4rp+snIPQvr1gWEOSpMbevZiWgWjW3epVZ4asvdhxdJAVAACFUklEQVR42uyZbY+zKhOA0YFMYhPwFQUlpv//Tz7MsKXteXru7t7ZnA+FKxsbxxE1mYu3FVipFIzASqVgqgCVoqkCVIqmClApmipApWiqAJWiqQJUiqYKUCmaKkClaKoAlaKpAlSKpgpQKZoqQKVoqgCVoqkCVIqmClApmipApWhKFAAA/4EF/C8BIex/9GUMWPiX7MpnCzA3idUjQtdEOoDumDw+AiYMDr9D3zxzwb/AmkOfh7H46/jVXZrE6OhB6R3tGBaBbm2eGG3fWSyezxbAja2MXGeBCL7b1eDBb1JOgA/MMWkB/AaNlPtigow/ZtFSDvhzoFFycVotgL8LXKYZ/KKklGqdLX2x2VTjsI+hBmcl22m8xqvtuAYld2HHxWHpfLYACA2VQ4+JIVikQpD0e+cSI+c3BRgcwhDbPADFov5GAN/GF4qHXeDv0gUfj/akEnfI2DAAoomRAXqlPWCnpNwcWtPG54M5ijfgwwWgcsvduw0dV4hqDT4igto6/A6LFpgFQEHl9WO6aOTsN3lY/FXm7YIR7u+lQcbpmY5BxWt92yPeBEBoSEA7HaXPgj5dAK7W3SHRa/61vQd8QqTIe5aV8m4C4Pg3AqwkAPjO4a8igrZI2CDzEGeCzd/XH/ZBAPRBsC1r4SvhTxeAO0RlMALXBeLRRoDPhRMAIkcI67wAxJwGfJ4xM13KAviVs6gl+9WiE/Z+OwdyyzHHxr+F5mQWUsx/XbYEonAi35qupfsSILy7t06XHt7NqAUTRpJi3Oox3t9lHvFRANuIeBTbVvgk6OMFuHeIaT4wTyGEBuLpVZ/DEKW4nDEyYkSM4ZjC4AGxnyjornu7rxa/AACMZAFgHqgxse7b6RD66ZzCfnQx3w1HCFcx6nYbHGV2IUzDcs5i2uLd+ugQbXeEKZw9IIolpg+i01uY+cmGn9yIeN+2nR4j4Ac9heMCaNeYfbo+bPvl4TMNJtxO636gb+UhrzupQQTARwEwqRTkikXz8QKgUVK2fZ4P2E5x+bqgjBWmPcCOKu3niFMFF882KjJDaWfQUqoRMPM0AiD0bbRriQc5Ytfuzrog1QII8ybltuhA62uLYFrtrD9Vb02QMWP0aBe1zfbStisgeB3TV3JjEPzkI2jO0yE2znXcb8pYv7UdojukbNc9mZ2YlbzJAI28zfMHHkPWeL5gIguQgFPqslcBny+ASB0izwcIt3H5jrL1VCMHoG9ZADtIOfL1zSMHt85Sd6oFPpEEyI23g9mjJVbLdsbYLjdsqc4bEdukqNCSatHpnstTxTxYVYxRHpU0TDEauiBjzH09mQy4RiHjT0f38u5N4LehEtdmehBg5L2uhE9CoggpNKuXAjADvW3JfL4Atw7RUz+aBaBufBHxbM0C9IqLyOp4loITIJx088sRgLAkgIf+vAoR211ThXUsAK+9DZUvNbb1gDDOWQCX0pGqWJAAnDdNHkmAE/gxrU/lu7JY2lKeuvBCWjYglrPP78SNMpTE1nZBIL4RYJHygiXz+QKg36hDTPOBuwBX7nIFAN4EWFJlwiHlJlIQELMAd54F4DkEAEKzaY/wKIBA7GQ85We2iweALACZYfD2VJhStQMgCzA9C9DwW2kn3BlPWAB1SdkJCFmA28L/YicS7L0ABkumAAFSh+h1j0wSAPs2RpW+ZAEgfNXgSSWZBfjzCCCSAAw4IfpGyxcC2ElG2sVhFmDhLj/NZkwSwCHxWgChSdgQ9m1bkgA9PmAfBeCHy2mOPkbeCjBiyRQgQOoQp2AxkgWAseWiNPAlgNUPAvSPArRvRwAGxKj12rwQgJetRHA3AdiiC6bZzPhWADZt844QbwVAI2P2OdE71hHgDxQiAHeIakUiC4AwT21aHSQB4PjxCPAsQL/LfcaXUyDEKIeMNPDPEWD99giQE14IANOTAG6XPEmKvBHgKlVdA3w8RnEtRbIA0AtSQOXOHrHhJBZhFz8eAbjdBl4L4HpAYWJuyAKkPBahnd8LAEeu+ZcCrM+BJS2DI28EmOouUAG4nfdViCzASdVnr7mzR5zbvAu0wM/XAGPaezTyhQBzEFSImzyyAG6/7wLZ9wJwu5OlVi6QBbjTq6fdHN+yjZE/CwCh/h+gBBqabjBZgINrsnvo7GG5/R9gd5iDx/+NAHBKLtosgEBiJS3cHMgDyxrtItWtiW6NwGqtAEsqXjAqtk+xtse8C0R4FiCHehYgrYIb57pjpu95nrnwZYN3IPfszwKwpvkKvVL9T3AJ+G0XmBAr9YG9PdQ5Ox9aA47mC/sMKAalvWjU3se0JgXnqIsaLWas59Wz+h97d9ubuM40cBycsaIpUkqfqMFEK0tW5oVlzXz/T3cT03J1z91leUjZtJmfdCAEFvZo+ctxSKAccb+6t/3dJZDnfrUrHwu719WdAbD3q3Ufy9P62brH9fretOvquenXbHaJvBq3nN1Zd1fV1dIBwEs5JG82t/tX9v2rrGYvfW67e/rb4JxfVOVzY/j1vKo/eIRQf7Cwh+M816HsBavKKREN7Dwt1+/3GT0WaAqqcNge2Ly8hJeXx9nL42N4ard3Vb0sa+6r/jCJX9un7bwcePD2sMf+er6pD9avL6G/6+V1t3IzL4t31f5wH9/M16u5b15m63LHfL0oT7PYPNzNH7bty7reP2F5xmr5sHu18q6fvZa1JbRlef77twdung9/u81L45uHXair/V2vs/qDWdN8DGK1PQx5i8MzVP0fLC+/qPfu9GjQehJmq0MLe/WqOpya+7Zmp6x6Wyrerz6o3n18tmK1+7PlKaqPf7xY1dWqrK+rI6/26Ssf7jk8xeHRv1m6u0//j397hnf13upXO5E3wNQDmIL7dlOfowxa094FpAH8JNXiYVOd9/5/mvz7XwP4STZlWnCy5av+62sAP0pV1WdYTXz+qwEopQGoSdMA1KRpAGrSNAA1aRqAmjQNQE2aBqAmbegAqmpV6I8vqG9hwABW6+X9/GnbNr3218vjYjPTDNS4DRTAanP/0BgLhbUWCuO3r0uNQI3YEAGsNq+tKe/3LgdJhUjY+r4I6x8WOtFQYzW7/hnutgbANCFFJvyIOMnWWbDNfKPDgBqlawOYPTYWTCeR8FPEKTsAFzQBNUbXBbC6ayy4EAmPIJbOgnlZawJqdK4JoHreWnDC+FeUWgv+ftrfQKPG6IoAVq8OTGA8CaUG7C/dDlIjc3kA61/WthFPRRwM+DstQI3KpQFUzw0YITwDxQbMq24GqTG5MIBq4cAnPBNna1+0ADUiFwZwZ6CLeDYK1j5oAWo8LgtgYSAzXkIMBC1AjcZFATw7yIyXEWPnOhNWY3FJAOsGOsYLkVh7Vys1DhcEsHoAH/FilMHpN/KpkTg/gOremoRX4A62Og1Q43B+AGsPAa8SjZ32LxOq8Tg7gOoFPON1AviJ/y6JGouzA9g4K3gl9jD1HyZRIzE7fwDoCK8lOgSocTg3gHUZAK7FfuK/T67G4twA7sEzXi/ojiA1CmcGsNpCwAFEM/EfKFcjcWYAa2cjDoAa3QZSQ7lhAAvrGYcQ4EH3A6lh3DCAV8iEQxBodBKghnG7AKonCDiIOPnf6FejcF4AqxYEf8cxpUiIzMg7hKdhZ3UWrAZxwwBmjU34EYv3OeQcUhYK3jnB05C3y1qpPxtlAN5G/CA25choip2FgOwBMp6GGruolfqzbxBA9PA2InAHAakFCHga6vS0GHXUKANwHwOgFg4HBkW3DyBrAOo7uSaAZOEwJyb5MAJQjJER3xaZ+LBAGoAalSsCoAwACd8wv48AJF3O3uX+kbHbLbctY2yb3ULHGoAalSsCYA9gIx68jwBiPWGy0DGyNxE5e+bOJqTgNAA1LtcE4D4JICNtwSZk118KmIjImZO1qV/QANS4XBFAdJ+PAMGahOzBCgqAF8ZIyYLrF3QOoMZlyBEA93MATpFjMACC0QDYRmgfixfSvUBqXK4IgNrDJPi3EQBZulYMWEESAwA2EIrrF7KOAGpcrggABQDCfwLI5ePhjtmVXaSUtgbARKSYDYBNGoAalWsCYA/QEe7F9DYCcANWkF1/mSJS6gCkvzeWu4cKoJo9Lwa31N8xm5prAsBUtnOK2MnbCBDt+9a/oOQyWbZJMpVgZKgA1nNvv4B7etYEJuWqACg5MMJELJ1QGREyRQPgQ2sAcgxGCKPxLLa/37k4UABLD9b5wTkL5lELmJLLAyhidsY1XZMjYjkc2guJsaaNwdomStfm3DQRpdktdD7hMAFsHHhhGhynbK0WMCXXBFBwTJIi4Q7x/mif2N+mGBmZiGN/JzNxv2KYAFZb6Bi/BAWr3109JRcEcIWBAlhaE/GLcAcvOgRMx7cMYP6HU/P1dH01iQB+geCXiVZP15+Q7xhAtf3SAIzRAKZDA9AAJm2AAOgN/gkxaQBqnAYIIErbdDm3WRh3KEX8iFJ2UQNQ4zRAAEgZWo6ptW1fQDKZ8CPeWg1AjdQQAWCAXN7T5X3JWfB3eWQBEJIGoAYPAAUyfibcJgAOuesy4bvY3xSW3VWbhQ9r265fkSWSBqCGDCCXK06CSFFYckSMEtKtRgBKBmzCdwHAJOrXlus3xC1YSZKd7aIGMHlDBUBEqRzsmbLNyMG40FlBaSOLu1EAyO7DCQrsLXjeLZAvKw8ClL9P9OCjBjB1AwXgcu5Ml3An2YzI3iRKHF1CpHyrAMg7OAwB4j14+iQAARvLtYWgAUzdQAG0MUreFxD7AKhxvFvO5fJ2ATTBQKa3F5D2eADRQNYApm7IvUABHB8C8IxI3tONJsEFNymDidhLPv4tAKsBqAEDwGis/BYAO0+3HQFSshDKYg70eQB02AQCKxrA1A0ZABv4TwDe8I1HAOrAcZniRjo6AlBy0LEGMHVDBiDWxN8CwAzSX9p0sxEAxZY7c6YjI0AQycZk1t2gkzfQoRAdcxRnhBCT3SKyd/0DozMSk4ccbzUCIDfgGdkn/PMIAF3b6gdharCD4bpyMFwbIiHG3LTC0nWhf2TqXBNCFr7ZCIACICgt4bE5AJEeCqEGGwH2Pt44rCBmLEu3GgGQHTRl4dgcQI8FUj/whJgyAmAAmzv6PACOGoD6sQFwCSAasIKfByCiAaifG4Av6zM4Li8Bnn8PIPmIGDQANcYAqtnzXXXd0aA2U3kLB9yJDkwsN8GluJOy8YyUAYQ0ADWqAKrZct6YdnXV+QCd8yEiUrmQ7J3PwuW66Xa8d4Fi8M51EjUANVQAnPYi4UfESQgpihwPoLz7XxoLAFcFQNyjfgl3eA/5I0LiQneDqsECoNRaF3LrOyH8nyjGMcZsMx0LoJotXryFol3pOcHqzfcJAFGgI6TkrOAH5MvRoeZIANV68eAtvGs0APXu2wXwfnVAzfEAqvXdg7Pwgb+/O0WjAagxBpCgIdzhyPT3AF7nDi6kAagRBkBvZ8XnIDnzXwO4X989OfiN2Z7EaABqZAH4FFN2+f08yGTk75tAdV39p4FmVZ1A5wBqdAGYnDvjpQTQMbLJJwTQKw1Y+Bd7gUjPB1DDbQIRxc6WLR+mKPbUAA4NXL0XKIW8I/xh6QiK0mXSAKZuuAAOJwVTFEnm5ACK/Q6h60YATg6sMCFSbAACEx5BqYNOA5i8QQMgDwFRmoTxrAAODTxcE8DhKLjDl18dJxqAGjYAdiDIrsXzAyiqWa0BqDffLwAK1kWMpmESm+nMAIqhAhANQN0sAIoZTJDQlq+G4852WZzJnJxJxGKbSLcO4DACsASJ9DbrlbK4Q2m3qAGooY4GlSJFKjdDiJRC7Fcnjv0l/aMRgCRL8CYQIucuxWwEdzjnlLLXANRozgcYKoDIxdsIIB2/75sSyIjs+kIoN/1lqwGonxYAOF+YEgD7QDsdtITBBkRq+tViRecA6kcGYFIscgkg2SbveNMwcmKk6G1C6kzUANSPDOD3OUAAiQW/zXyDswmjMawBqB8dQHgPAA9SDsxNH4DVANRXBEDMhEg0ohEgY0GE4gIh7QOwugmkBg+AZdvmnCXH0YwAAi5iTyL7fkUJgB3oJFgNHUDyXphYvE2jCSA62DIiSSY2NiGy211SBs8lgEYDmLyhAojOx7eFfxYAdWDifilD6TBY8CFsu4jsoUuSHYTE0e2WozRgQtQAJm6gALizggUF+UcBpOyMaYURY9gtdcJIwVkwbflMzFmTY7AmEKbGGpeD2wprABM3UADJOsa9GBH57fgHiolE+DYBMEfen46/X4jUrzx8X1fsFyjt1/Y3mEnPCJu8gQLI0BIecBNYnCAH0wRv0xkB6PcCqTffKQDqIOP/iE1ITUcUjUssrAGosfqSAGJg5MYzsis7WjQANVYDBdBCS//5XlxfAug0ADVmwwSAATzjAUmO3GkAavwGCiAam/BAXETSANQ3MFAAFKBjLGIq5wLvLjQANXoDBYCcbRuJiCUwdVaiOJOYjf+GATgNYDqGCgBJOt+9fzObM20K/X/eCw8aQPEE4QsDsG5Wq6kYJoCCYoqRsMeRkCIT79DwAbxCR/hVBNpVrabiO54QU28OU+7hsYfXWk3GtwygCuC+qABuodEpwIR8ywDqWQsmpzg88eCWtZqO7xlAvX6yYM3gLECzrGo1Hd80gHq1ePLuNAbAuNP49lG3f6bluwZQV9VsfZqlMcv1aWa6/2dqvm0Ap9voJ1vqc9MIYK0BqM9NIwAdAdQfTCMAHQHUH2gAaup+fgCVBqCmHIDOAdQRZwbgv2EAOgKoP/iXATQagBqBswNIOAjydlF/PZ0DqCEDWDVDBcDOPtc3oHMAdcS5AVS/QPAIFkEkEdpfHMHGbOob0E0gdcTZAbxAxiOS9bz/DS52Jv7lkbP6BjQAddSsPss9NIR/xkEIKQRCksB4hMB2VX89nQOoQQPYGBNxALSFeX0LOgdQR5wdwKoBGWgOvKxvQTeB1BFnB1DNoSW8nkAzq29BA1BHzerzPA+yDUQdzKv6BnQOoIYNYLWFjFdL9lY7QXUOoI6a1Wda2OuHAMrwVNU3oZtA6ojzAyhDAF0/ADzXt6EBqKNm9bkWV38vITfwUNU3oXMANXQAqwCe8QoUwK/rG9E5gDpqVp9t3Vy3ESTG3te3optA6ohLAigbQYHwUtHBw6q+FQ1AHXFZANWjtYIXih7aWX0rOgdQwwZQrF7ACOElYgPNpr4dnQOoo2b1JVYPYC8qIHrwz/UN6SaQOmpWX2QVrA2M50oOmtu+/zUAddSsvsxqbqGNeBYKBtpNfVMagDpqVl+ounfghPB0sbX2YV3f3MZpAOpPZvWlqucWbBsJT8PiwD2u6qFUm/v5aR523c3VZN0/r4YPoJjNDZh8UgIsHuz2uaqHsnlwoNQpzHZZDRzAYRD4ZcH9NQFiaSz4x1k9mOcGbPP0oNRftQbcXTV8AMVq0VowrTDhn3DKDsDN1/Vw1g00d+uZUn+3Xm7BLAcO4KBaLbYGwG0lMv4/xCl7C7Z5/T/27ai1eRsK4/iRJjGQQZZkudaRZIyhsIsx+n7/TzfZTU36royUNqQhz+9C2BY56YX/bkkJtwS/zR+jKpEALpMXNT9dKYDmKQqrldKlrr+en19e/mpeXl6e//71j5VtIw0d/fmtOOmeAC7lknbXC6ChKOakVatAplQ2SertXBZv2sP/mzldMgFcihclrhpA88RuGuYipX4lk63C5Kd29387o2YCuJxQ/toB7J6Ic3SuczEztXv/SnoEAJ8iji+hXzWAc9939yMAuMMA3iAA+D8IAAEAAkAAgAAQACAABAAIAAEAAkAAgAAQACAABAAIAAEAAkAAgAAQACAAAAQAgAAAEAAAAgBAAAAIAAABACAAAAQADw4BwENDAPDQEAA8NAQADw0BwENDAPDQEAA8NAQADw0BwENDADfF8Vymz8hhXHumL8rxHH0Gx76je4cAbirIVKxNMtltrUyXc3Pok57oi6IoUtp5U+TMdLk4Ji3o3iGAmxJSuJgXlbocjbVMF+M6M/Ul0FexUDpkzjnHwTJ9QrYKAfww9xbA6Lmtg0qZiLqZ6WJOLm3NTF8WlO5oFyvTZywI4Ke5twCEIToCYM90sV4t9D36IwA2TJ9REcBPc28BZD4CaCIRE/F+zLnt7ZiJ9pPzyzypynx2acdvE/cxpznH5tnE4/J5AMzv3+/36ZSP430uL0p8PPmOIIDbOwLgOFXX22KIp2H0c89EuR9WnmyqjprOC+HHTGawKg3DRORW7+timIidWLIoc8xmXXJXS5k4C5usoYaNr9ZO20QzhFjT+J8/gbrQlhgW0aakJdKm836ZRT4d+nk4tRIG78WsxL5xNnnaJt9TAQjg9o4AzKK1WIoS7GVHXKUh7q2y6yqsmpkolkBkSqQcRzXHmKkvInOcpWAOrQlRkzbdouTog5By8iJ4XbbhYXG5K1pQHKUa/Kwt//YbgAfRllBU9aOwatn2p9lwrmrgNqD1xM7KrRJereMs9B5AqC47q0fOb5MRwEUQwPsAmiiV5xxylDISBeWJaFLJbDvStWPdFh7jfrm2NaatC3JJ90S91hO7wOSkbit5VToitqrtde1wm5gicVXWcd/REYCytbFaUDPur+q0zEQm9dRWnTK5tBfRSdl2Q0tznyyIXDH7ZOn2yZHv658DCODWzgPI6fQwXgYm6lVta9if/Wy1IRLKM5HLRwCjEqcJM1Onj45k3G9Ky/ueaEvqu66btOq3CyvtjgDGvu/DmAQ1Qi28/ySOuNptIIuJadCBGq5tO5eSj0+BVnmaHLbJnu4MAri59wE42jBTNouazwPoiZxUs2GitwDYqkDN/mg/D6Cdvb2yTR8pl+TXdfV+6PYLH34MKqbTHbEFUNrFKCvTZj83p/ct2ejKb58CsZWvk715nXxnEMCNfRwA5elf9s6wt1kcBsAOOIoEUgg0lDiAUP//nzzCsd5u63Z19Oq0Ln6QUJfggFw/JNAPG5vpkwAwVKgv/i5AnE8BWl25TzNA948APliCk9Tw8DWoaz4I0OjtLcqHU4AGg1e43X8HiGF+O0YE4CECfC1AYxVB91kAaFeN1r8TQEHCVdV3M0Bqid8KcPJBALxrE0NS7XwgGHG9zwAxhCgC/Bh+iQBtsAQPBIgENAS80CkArbjREaDn+N0MENMN/KClbwSgIb4X4JhXEhTJ4u2cAVaa8P4MkA7s4MCRCMBCBPhSgDFVNwyfBBhSb1PN8RQAJj0fkYMe4bsZgBZcj75u+uoZINGu9F6AOB+fgZSHEe0xgEIFrT6KnpIAtJ0djRIBeIgAb2wY/AcBZk9xw5XoXwKoVF80pwZ1CBBXPaWmZfbvBHAf3gLtUZ3GpY1+WuIjARpIkLN9OvdGb++jlNZb612vALzVQzrG2gi0onV704x9hOZtZC9LIB4iwN/ENqBWMVV5p1HF4207zttFab10scfZEe0tI4EKDVE3N0BxweAIwNkwRK9sCxAV6o6OhYxOH/bI0BJ4i0sE6jVWIdg2lTmunuCE3Iq4DTvTJWAHcUHrjyuZCOiiUYfQU7omO3fR31aXDJsx9Ldt3yui8T6yt2j9K/0KJgL8AKgb+76/TR6g2T+NQzKgW1YV6bbvUu/kvdp7GmomNamxBfDTre9VQwBxuuzB8RxnTAY0au9rj7+Vi/uRtyHu3Ztdk11pqH5ycOJUf3C8yRwjdbc0XGpN56FhscthVRJsO86U8OO69m5SjuAc2adL+Xvkl0IE+JkQHftHHZ9biDPmHX4QET0ejF7sti8CCIIIIAgigFA2IoBQNCKAUDQigFA0IoBQNCKAUDQigFA0IoBQNCKAUDQigFA0IoBQNCKAUDQigFA0IoBQNCKAUDQigFA0IoBQNCKAUDQigFA0IoBQNCKAUDQigFA0IoBQNCKAUDQigFA0IoBQNCKAUDQigFA0IoBQNCKAUDQigFA0IoBQNCKAUDS/TIBGBBCYAvTmAa8qgMP5df9fm/D/QxdU5gGvKgAF3YAgPIsPujUPeFUB6h7XCILwHNSjhd8kgIkBVyerIOEpfK+rxjziZQUwbcBqUWqSTbb/2NQlYDXV5hGvK0DtF42C8AzaNrX5ZQIYc3VTf3mSZbvksC1/Kox/hfy4n5mMrMBl+ZPZGFswX/DKAiTq57brcDV1xgZDnRMXG35MF02dsw3AjagHyjrVdcpLYp2RRefS/g9lMfFbBXiSOEeTg7Nk+NTDcjVMrutgcgDr+CGtMflJ5ONm4mawHw2H/CwWIsCgs8qrHnWb9TVU0TCJVZKGT6tvGSF1VhJxMFmMujE8YJ7BMMjPYhkCXBdcc8oLLPZ1ztfAL5UJkzRs6hvOkBuSn0R+FrmOamcyGNhZLEOAWOWVl9NZtTIgu1SuK2Jn+MCM2mWE5CeRj9fIXAPVPaLiOJOfxTIE6BBzpu9aIeo251aJVWRXFy51lqI48uLaIyQviZ3JYORmMTmK9prlKG68uCIEqDdEzFkcWsSMNRAFfqkMiBjIcKlHRJyBHWLBMEla5zl6TVm8sQKTo9qbREYWRYBHFZkzffuKXV6JDrmlkqqLL82pKHNBAzNmVVc8kkiGjdfcLNY3xJw1UJ2yqHlZLEKADjFrDaQQ2bP3Od9gRezqwq02TJzGnbFmh6jMJHaGjUKepKejGU/cscKdCyuLJQhQXxBz1kDXFRM3ZkIp8EulQ8yZvWuFyFzQ1CMmLDsZG+Y5erXIlbQ9HK28YdL9xb7V7rqpA0Gwd3UkkDCfxsYglP6szvs/37WdbUMTDjeLeqVb6vlzqlEm8m5mvAtSqYspAK+O5O9AtAHxd6ACgbsDyRUi2K/LKwjAmi/pzzWRv2H3yA0pZRTGDx7kDAHI6uLfEIASIrBl6kZg2+sxb0BYtrtglfyIMtflGs+5SwE3o4Sc10XagPg7EG1A3NcWf0EAyJG8HYg2IO70JjNzdyBFEmPZ5mJer3KAU+6iGcXfsHUF3JCWCE8jm9vFFIAtMgP8O5kulBM7UIHAuc4povz7dRtRUb/djApOucuak+7qBTOklFH+aws9kwzLFIAtSgSCYulagBM7kGyAYCzbXdBJprmY1yttQPwdSJGOuWHTHsl4pqWM0sjmZpS/A10/APIGBM5zKV0o/B0oc8DOW4HM0BBGAO71OsApd8kVzrmLhhTjzStllD+lFADBZSkAz47k70CT4NuL5g1hlYyhwd+BtuYSPed25bvLmjPuoiHFeuqgDYj/2kKvD12ZAkCgC+XMDqQAuDsQzRvWdU4RZe9AFFHmDtQjwJkNW+FJd43ACylllLBqZkb5I/v6AciFByIKD8b4lt1GN76ty5aN7E2r1BsJZ9ioje7NhWbcSDrJiPVGlzPcpdeN8L3E9eahcIyRXYiHbslSAH7CTlM/Ndj1/q9l/ObWy/oZhyC378vCx2snlP8zvdnZLGhaUdVBo5mV5bj0QSeZEg9uMxrsqBmMdjC7SN0ohSl53aAu4r2LaQI8YaCpyITsoD0hyyrsuZoaK32mMgWzZEvWU83IYfg4A7myu2iFIFuyu5j+R1gKwB8fAJkCcIQUgKsHIE0ABv7DAMgUgB9IE+DPRpoAf1QA0gQ4RArA1QOQJsAhUgCuHoA0AQ6RngGuHoA0AQ6RJsDVA5AmwCFSAK4egDQBDpECcPUApAlwiPQMcPUApAlwiDQBrh6ANAEOkQJw9QCkCXCIFICrByBNgN+N/N8DIDM72UxvKW0XGKZMHqm0nfreZnJLTQ5bq+WXmsxLJqt/oRS6miiWLodqyr7UxdOFCrZUDgtRx1WFErZUAx1RR8J4RPnLqRcYiNqt6vWIWSlETRSjtBiA7G8PQDDyFBy5pW7YeOpANLWzM0K4Je/1D2pcDAIK1xXZV7I+X4xAFNWtJpnu88oAoCHqBVnZuCAxyzhJoorOCQCi9kAfirq5tZIotToEIGqnqB+na0q9oRCAqH3ofqi8TphqqPWGQkCivhLWt8pgFPbyQYkg9I3dO6JVa6zKrcpKoto5dGND7enGOZRGPw5x5TiDGNSk/+cBkE///p2wRReMbDaNsSo6x1QNUTuqIXhWCBG80UUTZrkJ1nLOIOBSyt1foYm2NcbrcI4/8OQp9LJIrb180ZQLAgYNApjBPijnNZ7K98+oi2qjG7NAqUDdTwjOUy9FCQCk0y21DNRN3IuKBw7UXmy6R1XCl0AUVUX92RP2Kx0x6BsbqRmpNIzUE3TrIEpEKKHVPynqu1N637KjeZRWFTIee3AIEWLeD/cbAZB1G6BU+VX26ENKH31hcf+Sotd78nJ4dL4f1G+MAHlCUGOUjpTbUMXuufsKsBpUUZYqXwQY39FpAVzG+vPb9FncDIhxR1h72ZIXZV0X4+xlvpTagZjbevIy1cVvejrg6Mmu9ZJSDRXCMt2pW/E5ffusxyVSr9ADgml+6nC1H5mn3PA9nPB77qnOPhW1Od0qQLQyUnN7L6oxIPZ6L0sHIurKe1WFjNSqPmNVq6f2r4PCgJjHovbCfEFwtZTKU10U1u0soOqfjNIguEbFqhoH2GQfWYdY5feqhjv1CjuH3yvq2s4A5vpDKwPo1mH8h70zUW/bVhawwKXuByTERhDccBjKdT5LVnNP3fd/tzszhERqSWq5TuuTaCRRFERQADk/ZgYgIaZDkuHBeLmsjloNIIsJnXT+2BteGr62+PYEI5UqgGcui6QvL5S+y8hPa1qaeyeBt7eSlSY14UJwOjDjipI6x3cgfJyO1Zm0XdZ59cskpYTaiSZktVn/Oslmp/Pcp5eymfKXKDxkCS/rrOCHbKrPE3GsJAZ0l++zlB42byFJ7zYxz9rUWWgu6X+uxT6fcknWNy5P3G7/U2vY1XCkK00BlZpLB0WRWOBF6Yb8EgFYhblWEmolFJRpu9nn24asVpfASbIgf9mLgd8qeZL36lA1DiVqj2qls2RUh1qBddKNzRKGRSTZMUiqzk9zDxsd8glAhlU+z4NsN3QscE+huR4AkjJHlVwVx/8zkMqwtBPDtwGguUEGmu+pONsO9x0qrIWnGXG65u6tpLJZMquJ0HhAdZaw9qDIY5676ixbD1r7yyyqz7oh69Svs6xZnpQXsolltiGroWq7ZTZ7Ur02yXW5yCO7LCQ5nPFZVHchXhdJ7n5ZiEmykOdmM2fb8Do3x/1X3VGlLJauOCqdRoU8kVXIgjg5GF3Wt4t8bZ9FbTgBrleLjAJ+rc7ssmq74ng2Lpknflkrn+TI6bJWJsn5hd6o2izzuTzRee7WczbeXTOjyur4FCEAqT6ex0jVxV8BkC7PmiMAmuTiRHotb7BNyzx5S+2buUCpOTmgLMnDyQH1+fm0Ngq2+WUpAr1Q+etS1n2m07/IpiBbrY6y7brMHDcMQR3lMXmW9eujPDKv23PWhvIoH86KO2LFllVbTllVJrk5yoFWOdkeV6o4n06I5zU/Phg11Gp3UqsLp9VknTjKyJMM7NtRRpXk6qgl1L8ciYZaseNaufP5tJo6Z0fZyh6PxtFR3EZwXg2Ag8JXaSu4alLU/6yjlZXiakUAtI0Qy4zNYNL5REcAapyKCvYiSMvTUpRNm7Zt26QVy7Oxxf6aKeJUQsHepu9Kih2ul1WXub88oHpuhGZa+1+OBWtPx3MWDur1gmx2c5zNH/XSV13uj7OgV85/PVVLeX7K5XE+VWeJOs63q3O11McAyJww0x+XbmPOtUtn9pezgzj+eiwu0+k5o+4kI6ilP863GTK2VLVEHOcQCbG2lF1y5iOLvFPH+WSedSf5HB341wNgs8xVLnDZ1SbFCCwZfJqKYpR10SIAFK41y3w6PbEAtK9y5TppOgdhiis4D0Ml6yRUbciyzirV0dxHqrDCFiqtfJL0rMvzV41aqbxWJwe0Pjug6vyADnDqzg6o/fVUvU7ypRezmZNs/GhgpsFTfiw2S6CEJ4yyU78ca3YsIevWJ/nCEhyd6ZMcPM/YSekUUH2mx+wkn8kyeZJP5mfgVEVuztqDfHuS0S9nvhNZd8Jo2WXF5qw5UGe2Jvxyep7PWp4tVO31AFQYCTUiL1Ypo5CgyzroRVdJ3VQByBiy3DUmy8w5ALMFSFcMY84R6oyHNRW5AyW1KeysqKD2mU8nW3LX1nmJJ7lFp6nmDSVeLzIrfjlrX8OJmqy703kOUyjcuSbrCwC8IJv89fw8fBMADXU9yTNm7gyA7hyA4hSA/giAMzx5kvlTAOCEnuhxyPwpAHnGT7E+B2DVgZE6NTm5Oslolu0yz4ryJEuRhXMATn1WD6b3zB66X88sR/NaAHKrQ/ANuINJC7hlAvDGjpuqh+Knrpakthguj7E2VVWVuYZlOluA2tliEGmZZB4/163PalFVKgKgMkxPLewp1VndoL+kUwAgpFBDoP5aOW8ZSE36swP6DQCUV1cDUEo9OH4dAMpby9TLAVALALhWLwAg3QNQSgPC1CsAEH5k6koAuIlSvhwA5UfNBBX2KgA406MsIwDsDQFI1HRFQMXFqrQzAKCiA6Q2KQXBCwCaPoQiS0IIcrYAfbMCo4EAQSKDpcCxF5HenQFA9uWOlgBATwCINwGgvAoAxbpEXAtAaZM8zxL3FQDEJQBE6Lok6/hrLIAqEnGNBWB5BtKXFwHYfQMAYQvNy6ssAGnwJLV4MQCyGDnXnbnWArDCc18M6s0BmP2PxvdMzwC0SSz+GQCh6+os6brOL2MAlJQhAKjSpmI4QGjScwDqCEBNAPxbFkB5nV8PgCu89EC/ebkFKK0TJS+y4TUWYIQyXmEBVOiHYbDyahdIdp253gWSnR3d6HQdXmwBVNFTU9Wp6yyATBj1nNrvB0Bb5D41CwuQgKJeAiBdrVYit9DkV8teIJTZAvCqUkOe1c0ZANUbWwCptfbqWgtAx/FqAETPpy7G/uUAcDbpVVFebwFMX19lAUwf3ewrLYCsaxnzySsAcDIGpe7FMYDMNWbROb/OAoSEjnyXyO8FQKpxzUcAulW1CtPoWHUKwOUgeFqdYwDZglOVJO0MQLo6jQHewgK4mglZBNKY7w6A9DHeK8oXA6DKSR/76y2A6GVxjQUoQy/V1UEwEc1eEwSL/QAMfzEAJqf+WpvwqyyASGr6sSHTbwuAyrO83PeI5b4MWSYrACDhpuE59tEIWfUxCNZfAaCyh3G01FEvUM5SpisgqVhRU4+/M8Celr1AHfYCYT520QJU6UsAiEZR5nZpATb4+h4AqDJ2FfZXxAAkvpbXxwDal1fFADLJ8s7yqy2Ay+AHy1f3AoWifDEAos4ZLofyKgsg8wkAnYW3BCBVfZLUg6BSSjAGGgfdm9QkecfT1NR5UrBGFEnNVq5OinIGIPj0sEPfJUnHpgJM4wCsSlnBOCtEZeoE9rXq87xvaU8VjgNwG1RaMfhOtFCG82ugUinLVfqXAJQFnQjQEz5bAN5vf5XwensAorCcXdkNKoK/vhfIa6jYNRZgurC4NldaANVlPeuL3rwOAFG7K7pBXZ5r2Q/qul4gnuMJJlTLt7QADUjbNFOs24qySkvR4KpqU3hraFCXNqlwOWdNq2o+itNeKIF2I3CHzapRsJLSdzj4q1bpKv7YqhTlCvcPsqLE8yt2vNXOQJ5vA8APptHNFsBm/Tpk7vsB0HfqKgAUq/NwdS8Q7wUCcFUvkOIax1ausQBUkcC8ThJ3ZQwwe0BXjAOMeKV0ed04AOq8f0MA3vwC/fRS2kvyfXWr1FurgQGvmuobAPh4VF0WZgugimJXhN13A4DX5rqBMKOLPCvUdRagHMwvV1oAElln9joL4DNSYZ8n8jUWoC/KawBQcCxqf+04gM0KOiZZ8WIX6N3cEHO1RAA0POA5enGRAQJAwwLFZ10ZLcCa73bOuFbJ9fcBoOxHWF4ZA5gkY9dZAAe/cq0FIDGQ/SoLQDE9GVD9CgBEPV4zEqwGa7osYVeOA6iQF4xZgPvHBiDdS0UWgCCAJePtWVBMANgIgMlqFQFwWWJdwSxo3PcBwOnyFZdCuMxeZQFk4aWUpk68VNdYAHTp6+tGggmAGGFeDwBL5BUAlH2h8KLr63qBQEo/9KPscvNjA9AY46fHaPX8QAZkW50DMGT9KQCy66UZDB8K/l0A8Fa95logmQ9XWQCWJ/jAWwjZldcCFVdZADIZ0YkJ5fUABKLn5d2gI7XnmX7NtUA+L9T/EADp9Vlap6PMqj89nCnPAJgtgJ8B2KzXst8GuV5/lxhAatL/8loAeOKusgDCo7A6cV5cZwHKor8uBhCxnCwbrrcAotbXXAzHsjgu2F9pAWL44H/53gA0EhthIw3I2abUQpcXNRcyiPSwB57eVca1VwNAf35Z4pMtlJ9i4YsukIvWm80xABz3JKyLxLx9EDzrv2LlFTEAiq/5teMAr4wBeG2uHAfoc48Zbe6vB4Al8hoAfKanQ26vtwClzSHz9wagwjudCqE465L2/FuWZf7iXnmyr20FBSwqHBS2kPBKSc3B/Zm1/xwAE3VlpKMUAXD5uNE5+ysARJ3zawGQXbAoBXuxBeBalmj03fUjwVf2AoleC4wydXnlSLCsybMogroagNglfMVAGG3vEnm1BRADdZ9+fxfIZjTmm5bdhRZc5BcBoPvBQro/OnQxXZ27vw/A5T6gCMDciqN6HizAWra/tnz9FwBwl+cjL68CQHT5JLV4MQA+T3rvNSjaqyxA/XILIJOsdl6z8uprgXzXS2kDv3YcYPaAXn41aNcbzgp27dWggoVg4P2fAWAED6ZK3SsBqAiAtOT4U68H4Fz7zwH4RUevsBZXXgtUcmmMvBIAZaLwK64G9do6qX555f0Asnz5OIAcrfbiNTfECDY6o14xEiyYuAoAuu/AMX7tOIAcHRyJfxIAJe7aFd21W5VcVXcrwduUAGANF800ORGm0QovWwSAVlWDAFSYFX36toKstD3siJer1eplAMza/y0ARId2m+XuiqtBv98dYeJd3xG2+3HuCPv+AKROwipeydPTNEzlUGe1IgCCLvJOpKCjnZedT3Gup3rQCQKQYgZdAwArlyT6rg1J4oeEsrZDbUxdFy9yjNKWtP8vAACRwUrWTc7hJQD+B2+JVFcDkP5P3hJZvlMAdJb1wiWSNga1l43OgACVZH2FAAwVzsLUwGpRVSFRdzzPRcVzBAA2MpVK4g01FoufDS0kDHincMALTe2LS/oCAECUYR4s6ruwAD/APcHyZgFslhWuyCMA2KEjs0zG22QwBiC1NtM9wzZzq5B1qykGqIasbuDoLAHIOSVUsIFN0x63fQN55R1hF2eFMBdmhUheNCuE/6tZIfhJnuF8Vgj7klkhikuzQvA5y3hpVgh3Gc9vY+3fcFYI9tezQpxPYvCCWSGSzJ17d2/sAq36YwA4AtBFAHCpQaE7xorclnRVPwGAadVXAIDF2wHwulkhSNjZAR2zM/WSoCtn2cqzbMMJbw4s3VJLTttXhVPwnSHKT0p4Pi+QuDgvULnUkqI81cez5tVnp4qcjll/Pi+QPs63AbrOPNThjLiQZezMSPk5S3vWHPDkJdPYiLPmwOTXT4tyfRDMOWyafgUABUVoE0Aap5WHD306AQBpxdcASFlerJouZ+ndm4jKa/GaeYHKPDGn2bLMHGtlOJuwsEwuZEu2p0opl1rCsuL43LE8O2XU52dtV2pPYdNQQrs+mUOtWC21K/fn093x49J1Zx145/Pd8To7ZU3VuThvfk6nlDNJdjpdFU+Sculunc53N5zNd7c+n8jsfAI0Fc4mr9uMc8vzBjEAAdC0dxVbfd0CuKaOjXn7MgsAsXLt2CCru7eRy1Ptuc2xf3F+QKuTaTRFyIoxS/jm6Dycz1cIafIkWzieUnQXTmbRbOrjCTRxSts60+ujOT6z8xZBJblWx5M+2iR3y3wmOWbNHU+gKcDS9lm95HPXX5i6tRqODwYvsFbF7myKz1NZwVbiZNpT2HCZcdsdzygn8twtCChdnus895vjOVlPWDufAlPZLKlPmgOezIy+kQXAX5Z9dRkASMgFEI3uRdWuijkGCF+PAe7aweMELG8lqcwTtnSAWI5zg5q/OKA0r2vN1D6XgU9qNWSJ221irm3AqcQvZXP7bMp3kA2TfLvPJuFTeZyHJzT38jx9b7+CpLBdRzXeuYRmeT4/53mQ+3zc5ple+Twf1GYxf/VYnShk4sS+dFgpmvLaHyrFMenuTFo6GHOt0KrDwuxLuDbwqb1w9Mt6Mce2cjXQpWAh9xlbX58AV7E873msVcl7nOUZITjUStn8kn9QYR/MvmqlhIoaUHiLrM1zbOvqzQBIwTLZKk1XstZ3DQFAcz2sugiAw9B3qGg2aFV6jqdLrDx8alKe57KSeVa3aZugWWIHAKo+s1yo9q0QmA8MCR9AJxqcgXt3UJPLB5T+DKDQhgsuHc5sL9JppnVrtkrx+JcBF7LBKautl1z66QaXlPbUjRKySReTZomT6OfFaCAPG2DVNpSUBMaV2hoLSbq5VDWDGznIZ1ifZLlbYV9znI9/6wf41q1O+BygBgOLpaNKNVDgTk+ViknnQlWItepymtO/xOODtdrKsaCkC5KqAjJqymghY99OSYXjCjN2oO0n4FQsoVpxaVyAVVZRUj14Hv9zIfHVJY0doWo9kzFfLdPUJIDfbr0BaQGlrD/V6tcD0OAdvgUzrK9rWXm8a7cZksQ2vEs6CXX0VvrONnSW8qQzUAuX1L3r6mAqmu5zLOrCN3QTcQt7sY2ErBwMRZbneQKH5Y1k5eKB4RLUhHSCkrRRux2p1+UDmq7Y4h9iNJWnkkVO5ZuS0ou/Z7rDNrVrKMlDElQqJl0YztBJzEJ/4XE56VIDO8wbBUH51HHSOTTFXLqxTfcFzhcFvngw/KJWbKoVO9Qq7yDpsjSuPpRn+seaNCYlMekcmv5QhaRXKSaJs6RzqXiYT44tU8xWwGoxaBvqnM78mwFQ4S26LTxgWaW4XFUNCCwgLV01OLczNuPxnuE4xNtW9F9pNChMqxVlXWGmmJXXCemcfZsoeD4wSRJ14vDfQvukiweUSm6Grk7qrvcHXV8JapaKISZdVBauQ1d3hZVNut8Txwa3npPOEDC2gDzBidWcNHRZPiddkqr0A+brmarmpB5sRkw6l5UYp9KZuVJ8hErlc9LFgxFrpRe1khoPxpx0OaOcqqZ5ky6SQCMp6WKtWN/VddGz8lArBUlZ1sWky7JSDouIJ6eK+BkoIAia4PIqjXopLOm8TOFxvcQ8h3PR92XbtqIObwTAfGBQvVQ1q8Ei6bJc/pdIHMbwf/m/iKvV6nibtOly8Y1s6ZxnThJ510DSN6WifCfRQRYg6ZrSVQYchJh0VT6XaUy6PmOPR/GbtaqqEyySBAD9tpz9UtUKndWmPK/ce7kh5lhwPl2KJc0b7pROQXMaW8ek6+T2P8ELea//E7ysyP/a/wRXvtBGegde73uVGwB7ea8ALOV/DQBwOvAC0fer/jcAbgD83JLeAPh3AEhvALwPuQFwswA/tdwAuFmAn1puANwswM8stxjgZgF+brkBcLMAP7XcALhZgJ9abgDcLMDPLLcY4GYBfm65AfATW4DqNSVI0xdvucjyjiStlrLCq0GvF7oa9EoRedFU1wtd2vnds5DQ1aCvkquPIl0Niu9vdRTTVwDg+/hFKphj3nvmV3+FErflC6EzdtpZY2xz924EKhCKpSRZXbxC8qy7Nkud5fR+fb7ku2eJkrwyX5FffRS77PojSPkuH8VgZXMdAJUvfWh9OW0w5BIv3O/b5RaXcslcvFDTXNJMKyxp796JpArvnbnJjycZ8MTTawBoikIYV8vpk8uJBLWgqLyo6W0CyS8SmcSd8fcDAM6iMkp+kx9OpOtoWoPLsrqoyoJZWVZLACq8MX61SqtVBYD4VQopVQPLpqmOAIgJ6apZ0RvkwmV1SE0JgBV9Fkk7ZfnXQ4Gmy3q4f/4mP6LsdJ5cdM+/5gKNvfFjrxYApLzF+6zZyrsmNXkvG97LsStbJ4xuFgC0xli9AudGcqvuGjbwYFvWc1cMAILxYpQpAGCGLpSTBUiV8T2r7v5VSV1WtL/e5AeVdZ8N6csBWAXPrQocVgmA0RhWtJjeNwzem9qkFU/G0jS6T1d0V2MEYDU2aZmwO941qRuqlU24kCuWyJVK+F1bqzterO5kblrRhRUB0LpVyuHLf1Noetdfb/LDyjZJ2hcCQIkpqw83abvclKWwmL+se4Ff1wbVnWMfkUgbmuE8AsALyU0xVK2sqrGo7liNPyBpfix/tzKrVNbtFAOYXBAAbpDc1+zfdYLaBKbXu8kPKzTP9EVZXU5uRBrXYgxQNuQoJOUBgJrThpwvAXBD24CkdyshLALQLQG4qxRnewDKxCAAaWCY418ekFM5TIh7kx9WaPbsi/JtxZsBSFOwCJWrbbUEIBW2qcICgDFUtHFrVcoOAJi9BdCyEhEA3AECgNmvHRG7AXCTqwGQd+dyBQB3la9SKXgi0xkA0uojAEyisEt9Zfv0jp1ZANM1dwcAyq4hC6A7+FCJa6LgGwA3+ccASDWN7DfO3ZWuAnVtQZd105SJRFvQt6IeGiKghKig6TqpPE+Huiz7rqxcvbcATc3ufMJbl/AVuv6VM+mdhBWVBC6YunuN3AC4yXcGoBJD0IwxG1TjTJrKgP2gBWt9cC18LKxiBU9xTz6wJlV9F2R1V4Ze8oIpHeTqrnGw5GFsmiF4FWyzcqPxEGW0LphVykM3qGs8oBsAN/nnLEC1l7u0SvEjLNNVldIKrsGLtDemVJCCH1YppE9pKSxSXMCX8Kro6+o4yzVyA+DtZLPefPWrN/uNV2Z7HwD8HHI1AJsdR9mu/+YJ2yjOuZR8u1t/K//m+yjMZstsYFSH9ZYvZL3embF9E9XbyXEX19dbqitXu/VfcqmY2dwAuCjvAoBfFUvy3PI5l7KvOmPKJXnS29B1dvvV/HLY/uUpNoO6+sdlsVVdbkg5uc3zerAgoU646fPkTUZG+GJHaxnyvLN90QX/bbrWvsvsDYDL8j4AwD99O1IRl3XrV47RFGto8WyWmK9tEr6pDeu4m/Han267YQMaGuHaJZleb0DW2267UckZAJv1/u0KOd4RzzMPPyDrLOzOK7HMpn80ANJvJf67IfArAbBZvWzGVPCbV56bYj1p8FdHo03g31BjtsHd+LC9unXO9cJ1apNsjOvjFnA4A2BrplzmuivR6sWOtnlGuWUOv3VWiaW4HwuARppzJW+NpPdK4f+OvDgSfkcArI8SNr++HgAQdvxHji/d9VqHzStjAJ/p5Y7qg1Lu1hcA2BUsvv1tADbrmiq9rMT6PQGQnvwXQ4UfvyUVZPhWM96GcLaDtKw15Vn5pLyrenb3YnmXAKxbSm7Xv67V/M9kh4BvHVdxk81mt9ucAmCyTOKXkL1dZqD19fE+cDulyHCwvMAAmraJ8WP8eUzCLHMp531QecfMrtvDty0BQIWGBQFAH+Lu2j4b2/Vm12duvd7MO6I94WZ7c7j8fUg9AkAQAGTuunXcdFEJEKhWu4kArM92SvuELTffGYBVKIJfKLQIxfhNN4WHULTf+D7VoTqnJuhppwoB8PzuGnlnAGx2PoBDvZa24zwkyUBfKW37uuAUo/bMhKAw2tSFYkkiTwEYs3y7M8HuQgIJa987VwS+wZPvCkZ9NhigBkXtqrOu76ADx3RZ0lu13o6dIQVygxuLXsHmkMR2fZIUu0Wp+TCyvvBrjKy7rB4GfwbAls0A7LQ1rnBQHpdk3TDuRnpbw46wMD1so1jR78Y6r+UUvtqhqMep+oNmekwuuEC7Ous3a9Z7XwzwbawEkmDd0I1rAkD1dbLcqVvjHgZvdFDfG4BOnFgAN6TftgBt3S4+xct74ijABAB+wOS456oiAPB7AIA2x0Uat4pbX5R3CMDaFFnYwFuSDSOXXcbwZIXtZmPyWqHTO4JCJGDotzrLR9fn5gSAHQaGssiK0XX1bm0LBWlF7jegYXXm0HEOO1Aa+gPqXQBtaPFXNiYroGEGpEi1dgGjy21dcyANiqQlD6Brh0L7WkIpdW5BhddjNmBjPrtAdg2ys+MBgHWA5WbMJaxaavqnN9DZHgrjMH5XA/2l5pjXmGEc15s2EErbzm8w2wUXiGW53LDc4IHRm1gJ2LjXawz5IQl2Pe9UOzA9Bf7t865jWCr+3QFQ0TuXvKE1RgCkpZQtrjRcqgo3VFJMGzTdHoCKe62btHVMjYVvTB9KAKCQfcFW6PizQQIWykvTAQArY6RLyrR0BtatckHDVitIHKW6zMB7BAA0AgHAszeu44d1GGP3zGajMwerXbKj3hYGfsp6AcBuTaq/Q7cccNmpjc/9pDAJtuUD5m47M3Hi4beJGYaqLbNiM+3GwHKc4ghDmIxZWCN2c3CtkgE33nXEn8vsMg6N98vXmY4A7Jcyt5DLZmyqNqM9SFwSdZwQX3f407LbIaqo9OsirOknj10gv16rMU/cZhNyibvAIy9z3HbjKLfBDw4PR9ypmZKxk4BTwc0/BEClA7NFeXcAQBbMFfBVE7S3uoINBq/75giA1Iu06YaqCZ1oTMLaph/SVNemZIlLG7dKZaLu1ACI1DpNnUlTn5epqh3+T6xv8eaY1DnYR/gfBID0EHUCNHSbDB6ky3BdT6dToQrjcnlu8qLo6sK3017WRA1tQ/TQTznUi9GD1NmwUYmjbyXslCMAIH1mcN+o7ISJQQXXx1EocOhpZYQyLQGIFqBXILyYAdgwtsFf7k8B8FQYVmd2qmys+aavGaSPeS4hF6OinLhANVS2Gzg0+3Js4WvEMwKwq+2GqqUOQfAAO12H2k875fhbCpqW9p8BQHRNWjGZ7gFIpQBVhmZbFKu7lahI6ytxDEBTSCEC5LR9Nd02w4pVqgvAZagbNggBVFS9Ab4KnZZdE2OAwk23SK46Bi9/OWwgeZ8AyGgBCIAt7A+VxJBIJANcEpfk21kh53PTKbWPJw026rPWkKZGAGzODAnfePoRkoMFIAA41YN2ajenABBXU0aeQVoE4CwG4G4uAejpzvTZKQCbIdkX5giAtu68Idlthlxe6gViUNd232+1k2M+AyBzdtoLpDMDe+hM3Cke32SEDP8IACnr0+ibRwuwUsYEm9613fTPr6swyLY6doFEp0qQigBoCIBuNWmzSVTvSpCmqfkUBPtutQSgBgAKBx/gRxy8Lsk7BeDIAgjc35jx5djmyHfdOQBzELwEYJvHbfwBgE1IDoZjnAE4tgBmPzZhs/4SAF3MqPJEnVuACMB6uwBgO7LduQWg2CDKAgDIhasRtgsAKIwBDrJjbqsWFsBnFwFQSdjMeUKedXLzzwCwUEECoNLByx7W0pb1BatA7c1Q2OYYgJo+pxcA4EkZb39pkwiAOwJAIgBEQsHbsbxc5HcIwAZWzJkFGDM2dxKFoQWdeCkAKsl5/DwDkMsZAHfZAsg9OPqyBehiRpV8wwKQRBfIo7LJCwDk/CIAh3hjA7/lL1iAJbrFuD5ygRgU+gQASwB0bdwp7pfVWWL+GRfIo9a27V0EgG5qT/WQUmeN6uAD+DVNMEcAlIlJ4WN7yQKElS0Aj1S1tUsJAJOU5xYAg2CuYMPL8v4A2LHNSQxALtBes8G/CaD6VwBA2oPC5hjA7tVj29Ke44iVnC3AApzhcgww4O5JES/FAOcA8BwqdQYA7UhP2/L1kQvURUrXnK7f+BYAuw5VfwmAzGA9VisC0JMLlPO40w18g0MR3fofAaDs+KoNBpS8GfumSRv4LIp+dcfZqmohOm51UzX9MQCrUMtW+FU6HGKAbooBVpaDERjK1ohU16Jqu37V1kNT+VxWq0KDBaAYwN2lo2/br98n/N4A2Dh2KQbYJrknFRjbXZJTR8lLAdhQBw752ts9ADJLJLlH41ol2dQpDuBxyh4twLpHxUTU6t05AKhoNa373J/3AiXnADgsTiyUzhzF0cSBzxJOHpLbLABAMArynLzZMFDn0yBYLADgeb0IggsKgjO9xiwuAkC1wgXtdO3lrxz4ot60fwSAlIc+uNWd6kNRhL5NWTEwXvi0sf3Qe2j+HbyPqxkAlDbUhW7Ssg9lJTvXNLYQleoZ9wJafd91AcJqSLSs12XKi6BZ8A2A1a5cJ4Ew26SuBgnqcpnfCQB9Vsex0jHhBxeomCwAtFGojfm4VSZwPLXjeueTXLabZS/QPCgaxWeBPuymplTV4yYCgBsmbLv1QWFCPkjO+h2yVu/WarMJZDO2dY07l4khT8lOGqgOPzbkjkoZ1vT9AMtZ5bPhFABQ4u1aoQKuEYZ1u6O33W7XZTUUhgXS3g4BoJrzJCuM4tqusc44VsDy3Lez0mdstga5X+/GPFHthk+VcFneS6jhlso2GVSPP5AVUnELO5U9/lRf/AMWACWdJm1LVyTg+sBHGsiC9FU6T/UWAYhCmejyCcxY0QokQrRMOdpV3KiirPC5qiA5bp7i5itftmUp2fvtBt2ANmdZYa0dQp117VqFLOHrNQct2m12NkskqEDIsjzpzAb75/MuGGBGK5Zner4QYgcfc6fW0yfciyR1UaEzivfU0BMAOLCEu6OB5baHXHmxjQD1suV1FhTslXeFBOY8/KYqskKRBrqDBra2dmprhx1khP3VvN0XZKuzLDG7TSyWy3MGFemypNAyyQuJBq2wO3wL+EaFwRJQZanmwOPGJ5g+YNMARapD0EnS84gfD1nWydh6Y/vQ9bzLOrZpsRIcR+gySJRU9no6nAF2yhKobWJbJHvctbL4/uMApn2pC55ShhZvdn8z8XoabXvHACi5FyPldrPmsMZ39LalL1G11nLUZkdn22i/A2eI7Sjjdn3QO4nCp8+UXarNtO6cx9Xod4O0Ro+mjVuOelrdqNFtNzu+30srnTPkMWwpaSdBdst7YByjn5+y0C/En6ayx63ih50b+XpjRg5Acc1gP/DmybWBwsg1/v6y5vCRaRers/N63LYI1fI3uNrD6LUBR3E0sDVUQm2wzI72HsveLndKRdttufGwxXcGoLL9YK7RPjH0ffOGANSOC2P+oSDg3d8SCQD4X39MITbe3z3B6VWX4rz9pTuVcJqp6posPzYA0A1yk//l+wGuk394oqx3DcDG8M2m724z9/5kAHxDfi4AdnXieXGbufcGwEF+LgA2JoSR3/654AbAQX4uAHDw/6b+NwAW8rMBcJMbAEdyA+AmNwDeWMobAD+0rC//P8ANgL00daJ+vckPK7v64t/33wDYS9Vn4y3U/HHF590lTb8BcBCeJ/JGwI8qvM783QW5AXCQaswSt1vf5AeUHauz/qKi3wCYZaXzLOmKm/x4kmT50NzdAPgLqXhI8iPJsiy/Vijb988z5/t+Wd7kYFDGf/loJIX5iprfADiSqlViIbzItLheZJ0zcaWwvObiFaKz4hVZXvVTQ9aLV8n1R9HkiRSvEJZdPIqq/aqS3wD4lqQ2M6/Itiry8to8Ki+q1xRRZn16dZYhfc3BYJm7e42kw9VHsUkSUsu3O4o3AG4AkNwAuAFwA+C9A5C+EwBSkCr9piy2fKsN35HcAPipLUD65z0IY0Y+3F+Wh/hrf3x1g/h8+POOpHp8eHh4fPzy5RFWUD7hglbi8ss7IiC9AfBzW4DH/6BYL2B5UT78cUfSXPiKnrig5x6Aj//58OHjA8j9hw//OXrE58M7AuBmAX5yC/AFNXjQhj9/BYCPEwDpH2fqH1UfF/TcA3APOv7h/uER7MIHEvxMC9L/GwBXyY8KwLuyAIOX4v4vLMASgLnlX2BwsACo9h/RJwICPkatjwsi4QbANfKjAvCOLMCzkKptgYSLcv9VC4DPuKCVJQD3KB/xdWj8o/a/MwtwiwF+cgvw5T/Psm2atm3vX2wBSJEjATMDH04AADPw8f7jx6j3+9d7A+BmAX5uC3D3i9eybZu2KcX91ywASYvODuk6CWp5VPy4cgIAaj4hcKT+NwCulR8VgPdiAZTtTdmClGUr7r9lAbClp+gWBBak4VH7zyzAR9B8+hIsAG4X5QbADYBJ3o8FkKE3qmzhodr2EgIfVxFY1PF70PxHEOzlBw7I0XmeKFgCMDf+HxYu0LsE4BYD/OQWgNtBlmWpSiVadIW+fDh1gWLeP8ixj+pP8vhIduA5OkL3ewAePkyb7gG4WQCUGwBLeTcWIOV6YII7xgUvSyCgUQ9LBEBbV7EX6B6dn1n9BbyIANwINX0G4B49ILQPNwBmuQHwPi2ADMPIxmC9lEopigbU/UH90ePZW4Do/hACsw24vx8AAGJgAQDBck8BwAV5vAHwcvlRAXgvFsB0w+hsEUYulCoVWYH28cNB/R8e6dcIgGgDZgiIgGcKApYAfMRNP8ELCJjlExAxrdyuBbpCflQA3osF8IVl3tteM8MVZ0aoEgkAPSWHfwnACQFEAQCgnz8eA/AJAIAIGGSp/x/RJFBw/PDnOwLgZgF+bgtwJ7VRreLejY553WtJRqAEDSZNh8UeAHRpCAESen+Ex/3g7k8AQM3H1zII/oR5MPnT45cbAFfIjwrAe7EAAIAQoPEGAGC26JkqUR5R/78gAjMA6NNEM4ALfAIAz8E+k5rPAIDawyM+aZ1wmoLjxy83AK6RHxWAd2IBUu6FkOj/G8ml0ZZ5LowRXyIAsNgD8Dhd2xYZgBe9P9ugPXX3LAHAz/ML9R929gmzfoJd3mKAK+RHBeC9WAAAQCnBpeCibZUwzDE2aCNij+cMwIq6fKYOT5KP9P48DNp71PMTAGYrgG7/Q7QA6Dm9qyD4ZgF+bgtwx0fPpWZcGY/xr5KeudAz8YhybAEiARMDMaD9AABYxkDLjwH4OENAjj80/RgDgIAbdesGvUJ+VADeiwUojXfO9qY0UgkuyraUcrTOyEcUbLmXAEQCUKejlj/3/eAYpcwALE0APkjuwQWa1m4AXCE/KgDvxAKk3EvvmfPc0xURNBogpGf73p7HJQA08huVm+QZAOgG7WDluBeI+Ijqj88o9+8OgFsM8JNbADno0UjpGUa/CgiQHBDg5h4FKVjGACiRgPsJgOfnvrAEwGwBKEAmAqbHQm4W4O5a+VEBeDcWwFqrRyaBATd6yfUoyAyQgiMBi14gEFLtqNlkAf4z9Nay+6UFwA7STzBAPGv/hxsAdzcA3qcF4FprZMAZabwxTA+mVCD88Z5i3TkGmNT/EAbHEGDo+946htq+AACCXhwcONP9DzcArpUfFYD3YgEEY07bIgxWj6OXQIGXoiyV4Pzx4SMSMF8Md08qTI97UOWo/0M/aD0+LwEAXx8sQFT49+0C3WKAn9wCCG+8d6EYtLaDdsxIY7wBBoAAcY+qvgfgY+zZoaYdWIAn9gGFECCvfl7EAKTlg0Ze3j0ANwvws1sAx7w3zDmnNSry6PyEgNO8fMSLePb3A8zdP/NQ2PMQ6jpYXQynADz3lt3fAJjlBsC7tQDOOY92QFvnRj3YcfRGema9AgCWFiAGvjTpDyl3BEC7wt4fA/BBD/8rAOhEviJbFZLrAUjCqwDgub0agASzXC8sZ3evO4r5tUexqbtXAVAm4S0tgJLCMDQAjGnNvHMhWK1HdIWY8c+nFiACgDI5Ov0EgAatX4wD0Ajx+BUAPr0vAJp29ZpsbXv1aaja5lU1X7XN67N8/3wkcBSvP4J0NN72KF4/EqzKUnDjPdNaexA3jG6EdWbQDLAIAFmAyQY8xxEt+jz0XRI0G9wZAH0PgfGZ7tPrXwOg+oOkaZrV6o+/kr1+p7Dpi2WV7nPFzyB/rF70U5SrmmS1iu/VLCuSZjWtYa5lteaaney7ibLCSp8WcEXp+E3z1RrBY/FTtDFlorV9QenTvrRpLHk656LvYAFZqQ4ryoyb0ZLeYgUPBYT1g2AmkPiL9MIch02+Sd9fAkAMSAYAGAwIjHeEAHpC/H5pAUj/3X3s3cHFBIB37AgA7AMqgnu+vz/Vflr+axfDffkPidXO+8tzoZ7O9Usu3WJ6378Smks7Hi4aR1kOHh4GUD6efPpyF+WPEu/GUMI7ScPysmxBaCEMnhnmmJFcShNdTpT0Szy+9/d2sHDFySd60gMvVrQo2jE9whmJBcRu7U+fPt1rq+3Q23Ec7z99TeDyRfwpksajtzBOr3F0jBTG+D70vVdQVMWn0sq23TX7AgrNsFvFy60czdaMfOvD4Jwpjdlyv91ys/Veec+5c26fqxLSOa/dqEfsodTeGMaM4V5ywwUfJVeQcQihtzb4b6nUtyfGUqoEUaWSzEnuGWPTgXZ4XJy/X8QAhMCzfY59obiIFsD6YwBgiDgEBplPDQA9/20AhsEJAQB8Ww4A3P/n5XJ/0C+iBa9+fZz1H9JO54mk1wGABk6FkMz1vTZKcW5USfpEmmUMfMUcCDDw8PFAaPq4t7QfQNsn23xI6PseCIAFapLVz5/2BVxRe2T7Iqm7wVIvxvI00eOAwKGAKyggFkwaVBISBEH38MvMM4RBcg7kYskXADgpFEcAnH+SjG+ZBXaMYuy3p3G7Nf43z5TxQkDFD7lK5Z0fPaAmrdPaefho8GGc4U8WcHl6MqEIPYh5NQAlAECihPFKcUnKjwzoYEetn08twGA1Htl4lAHALmin/f2xC/QM+GhocJZHNR7adwCAZvzlAKQIwLfNRXzAMwIQp1GK95USAfB5fsbt4+vYAjDrRosECAIACFDSCByZKRWeJO+YG7GDYQlAbJGe7WiWjc4HbYcBniFY59io3f2nA6F0QnQfkqSwY0BwotZPK/snXMb76SH+FAEwiyIasOWWehh6bzwIAoHiHGv3BVTOSU4AcO+33AgEQXqjDNs+eb6FN09rPITZApSCOeaMd55j56ThfgTunv6LoHk2aA8A/LcPoQf6Xg9AQ/MigrT7mYGU4FJygTDAyvJyaBLv5eMshk0mQ4jHxz9idb/gVkSqgU2/xEdcoPz5LwMQrDf8/m9agHli1NlrWloAVG0EYPaBaPv4nEnA17KBVQ7HFUdrmeTck44xq7UXgiMMCIIxjt0vLcC+yX8GtBGAWWikEnXEMq9BZgBWpOAQqiVJGF3Q97OL+ml+RCuwKGAURQ8UYZ3iaGc8fVZCCI5OmvGHeFoxJyUBoOT4xKXaypFxb4Q8AGA8rAkRejdbAO4YOj3AChAPWzNNlseg3+V/7x1GrjQOZfvXuUAk6V16kOXHORHlx/iLpL0FYEKqhwjDKy0A6f5sAeL7bAH+IOX+SPfOxTn0lmq/VP9jCyCcHUcclXRcogUQ0uj+9x4JaMsWBZGYLS4BcO8no8yMFI9HAPQDeqn9oL0vNDuxAARIkvSaADhr/6MXdH8EgFJCldIYKeEdylJK7ZQYmZTk/Ue7QIuDBRAYKGAQ88RNz7hQxju2nQDYGjkBwJ+8Ev3ADhagNZoAYF4ydH6EH6WUT+igMOOt9v6//r8hFGEYXgHA6yU9+fTVn35PnZ0zAM+P3Ag4UV9ebwEOur9s/OlxbAHQAMx3UoMRODYAJGcWQDhNw5LWIwBl6a3WPYjjoqSAABbKSIi5on4hAM+MuuZ8WRq8n3tpAYYhJHkYLGOFhWxLC/ApAjDoaAGOHvNzCcDkMbvBAqEC11vhfGnQBXL4MSJAsp5DG/TiQJ44dwxj/JH99wlglf4JAfD+yRjOmRDW+tkCOI1tvTROeuDAKFgR0njNDCAxAcACyF8BcPuXyAUAz3Jy+Jp2EQSEqyzAh1n5zx6fZguAKo76T945dQfMNgAWi7VjAJjDPpVBc0EAsMH2ddEPjk8uUEutrgAtX1qAB9R/w0EdFbweZgCsBQASBMAXgzcPxxbg07NFAAY36HsCYu4/2i/gtQRASAWi6zrpCIC2VN4oC4RaQ4pP8ys4iSsLZ4ZLLwRqv9fUgWQtbK5EqbZbeHL+hLQ7wbX2cwzgHQHgnfBGGg8AGAxSmTdAhR2Bif8CAEXdDz27AfAyAJ6dbNoJgMfrLQDp/Ky6Fwg4swCxBwiUDUmIsrQAJwCokjOvQ+iZ4DxagJDUYXASAJj6gy4A8Aj751yoVpWqbb7czwAMQ8gnAHrrTywAAhDqpO8BAPbpTPtpcRwDlIND3991dd05NfEopQghwN6FIPvAum4wLchqX8Dd9jfDxPbpyY66B7bRxetH74ELzoXYKoVPCQC4JQBmpI4lp4WX0nshmVFCSrAWTBrspTceASgAAHcD4CXSSKcNBvrwKtXj/ZW9QLFjB57kz9Bjfp7FABQCHO4fugcC5kj4CILjIFh4w6y1XkQXSDtswi2LQbBgXp0C8OHZg7MljRTKCFk24mE/5vjBDtTPgwAMg+cPD8cA0LUsvXXa+dPQl5T/DICeoZKzouuCE0Kht2+k6JEAJ0oKCjyopClBZgB++804sd0+jc4BAE6jW4f9/5CEzycUuQUAmPb7XJUyHgDA4J07aQgArzgZA+3HPmimP38OBQIwvLkFSOERl7PMaen8iRLON6THvElMjs9/S8qx7325DyWbVtx/eJkFmPyZDzGeXY5tzfaAHrOHEYe6YsNPV5CfmYBLAJRCSqatZhEAM3rWd0FLwUG9pHcDsICh7nMZM6XGauqCZF6aXvtSGLnveL63/QEAdwTAHxMAqLrYJI8sBgHLJz1OACBP308AqBJFSnSBBmumj0qGYvAzACC77fa/IwLgPABg+yJoNzxtfxNq+39beKL8xhEAb2cAhPGeMRzAM6N03gMeXhg7sKcn7W3R9Z9rsENFEcLfsQAVDTOXJboFbRxsbqgeitTkzz+ruOGfkwD2f84iuABRChP33aCUTqIg9YL88W8hwEMYvFKEQCnaBhF4mQUA9T9MDv/wSHOiTkJXR80N+8NsARYTAhxNlDQHwpcBUNKHz0GbCABnjBvdM+oGFVo71w8GAHgelgAYySVjzDvrvTCjeaCwgC5K3ANgjPXy8eEkCKZxMq2dtvaZ2v35edECBC2UUn5vAYQnCzACR4jGrgQRoei9WgLQSm60QP+dP/1udaiD8wjEVkX1xyUBYJYASEMAjI71HgAWyns5hs+9I1eq7mqUrus+/x0AGmxyvLXkXVEUTwMvkiYIgrePn46uBsV+A3fwMD/Q8esHjUZtnh4dvgDKu2DPhxdR/r2BMBlAeRTWEUS00x8iXEZgeSkEqf9ycmwQeDv8Q8jzwQn6+HhoYD9GzwckmoALTf9lALjXdpQRACUNN85OAOCIkNPaw5fOLQFgRiku8awZWC0VhAXUot/rAYPgbhiYcYNZAPAHavuztoPV2mqHY55L1yeunMcAWsHuTdHVZAFMzwEA5ULo0QVS2LqoopsmGJzHAYJlWgjnpNiOAwDQsy0Tzgs5+C12iOrtk5VbJgVnZq+rKy6AaTN677UGto1slB9B3bvPrg92qOsEpAYOun7wrwUgBQCYdra3hksK3aWZbgrmhJ//eBgJjhN+Pmu2H21HD3PADgDNnHZLAD5SbKV7AOC0d+3+X/yneAGg8qmbWpWyJAKa9uHDBfWfAUgfpu7Mh3lS7CiEAHhRz1H9wd85AYCuhdgTMDMwo3ARALo+9wAAF9I77ScARub96LgUSvq5mx2UamSwNUaSxnupFEzaPY1sAQB9koOKMAkAcH4AoEIFf9bjYHsLADmr76PWzwbgHAAVEIBShqILo1DK91IZeA794FBzSPoCYFhaAGUJgFJILu2ge4DHjKLkwgS2teP2v/3T+BnQxR3IAwBSsP53Z7F0NvzuJVNKd3UNOv8ZePucTFJDymfG/4YFUI6icgYAKCWMxcaHWkmFUcf9wgLcEwDMzL1s6GF2YXBsOAEAKpnYsQerOo8t/vsWgI8OgzipRyMETQNMBKgzI4D6+steUx6Xs2JPfw2i4LEgIKryEQDU8TnlRADgw7n2vwSAkoux760kAAQzdtBMwCfGD72MrbdaD6NjUgl4cu8ZBsV4wO+xF6jrAjZSUnIhlhfDIQBuCHVvRz9ad3+q9hctgEMtl30oguYAQOCllCW3oXdjMIqGiG04AaAdAQAOXxquO1CLutBDEK0AAAwCYMLnroYvmJTM7au18r93oOuf+545bOXN2GvyejAxhDpJ8jwCUNu/A4BwdMGdRwsgNRzsfuQq/m+ekfM9wdimQFMogORIALUvRYIAFMwcAfDcQ6l071DpTx7/qgs0WMfM2APkfGqvEIFTP2jqv5/HmZb6jw8U4iBOltcv/gF2BoAkBsxnTtC3XSA2uqUFcH0YULHbUhmpC2sE+gqzBeAMG0mrMUDw6GoLaQzzD9j06MH2AaS32jHGxeMRAPeOjV3SDSPT1kDC1+TPJQBCRACkUtwrZWQp9gCgcA0feKlUc8il+xEBaKWw9aBtXVgAoBRc9n6rx+3TZ1Jr65XQ+gCArSkVit6Bonejw0G7nFr9LhSwlk8EwFqQfwcAzzSOuxAA1nZFP0pRTo4yFwsAMKx6kMh4+TgPNHYEQAC7MAOA15jUUNPBzeOLC//y8V8DoAsWe+KKYBkabIGVaY//G42Gr1DaUwuADEQM9kIAPPekxacAfIQHNMMPZ/8UQkJ3VOzX52vN1GQBnHNGcM49AeC1tRMAJZcuaALAuQMA0mnH2WAHDRCMmFVwJc0DxQA96j+OUzEmmWNzEDwBMOourwf9dQCgCucDYTJMAJAyGL7bhnpwnrR+p/gYBlgtZwBaBnhw1ZZSaPgxDQDYwFvBufXb0W1dbNidkLY/AFAkIOTvkNp/1rbLQTJQ+/pzl+QotAm8xr8TBHPP7MAEAQCFq3snUS2U4ccAYGezlEJx3qqHPQA9AeAhzxkAvYtmNbqU7yAIrgvrEIBi0ILOHgkagUZ9iL5/7Ox5oE6taAEiArNEaxAJuD/o9dwNOvk/lBVIOAXg06In9ePDAgCFGm8cuUCcmyMAMHIXGt0hDgCMMwBaO2lojkuAABlgnisO7iv28wMB2FOumSyFdwsAHhAANxZ5jUGwlS8DAIukeN8XYZRQWoFuQsnDpDXlDh6K4bCdUHMQ3EjLnJgASAar687qIKEyYvRbxp4+kxrXXdBhPHDddAlJ1w9JPil9vVB6eCMc4AN897cAEJJhCEAAOD8UvQFYlWR25Fwc7gdotKa+ZtjYK0nmNQJQAAD6FIAQgh01WoCZgBhefXcAqrKpvmoBHGPYZzF6I7E7g0mhJgTABsR7WEj/Hw4u9pf7D+cEkEw+EN4hcWYBCACyAmeDwKTzNFf2PnCYASAXaAlAqzieCitFBGAMFlaFMP+dXaBRM8UNTfDndB+dIcdwNtYBdH8PgBLSLFygewTAm5DXWjOmL1uAh2cP6XMBuRMgEvZZaDQGNLG+2upgR6NQEA/DnPdiYQG4Y5DPGMNdByUJvbYBiOWAMWzK6oQEuAijngGIuv45JDl925G/j4uo/JGH+m9aACV13xuyAMobqTXnopTajQNeNs+mvGmjR8O5x04Iz418wG62AwDGsQUAwAY6VRaD6w+g8CcG4PsD4EYvLjEgw8iEUtI47bCZlMZCrRQRgL1BsfUn/T9YgOoLNtVEwOQJkcw4AADs46kFSPH+oYUsQwD4YjIA1LH8CfbzyxIAPjq3DIKFMq6fLYDXTgig9rff1nsAdk+jb2k41XgjzUgTfNhBI4EIAF0toxlXnMvH2QI8g3p77ocEAfCjMZ8uiXbLq0Eb6ThApq2z6IqJCMBOMWkMFyheQpp0Qc5RujLMeCkZA0a1kyA6dKOXnlK4T0iRc9T1EGYAon5HxU9wCTItagSA3CES/VoA6AJcLj36lQBAKblxjgAYsVHgypiYt3E2tkreG6UAgBgD5AiAZfJxBgB6FywI9i6f32n3/QFYOav1JQb4YBSoOqoRG1HPdBEsIkF+kEIlj38LAnrezhYACPhIRmDxB1FR0AIMDDc4sQAEwNz3GRH4SIv4bwkkGFAsLIBSxk18GgAAjRQA8EwASDQPgiEAHAFoDgD89l/Tbn/77YkgkMYDBNpqZ9jzvQbdL6gXyHMlxBIA/+lRKdiyqB0C4C/7QM/PyyC4MVaj+1PYokY/R3hjqPOmZ5zDuvc+jPg+FN7MAxV9YI5JgYQwhj6UK2rrjQ297fVQ5CSxR6fdF7CmxEtSd58P4UBOov+WC8SNH8kCYJBlLQOwS8BVj5wryau4C2OBe2eADgG0QzxAl5IMeHQ1kMwXANzfjyPOGu2p+Zg9oH/MAiB99pwB2TMjRNk2LfcjEM7GUPQjL+NlvKjX5NgQAs1sAVAiAdStiYJL2v4+BOs+XrAAJMTArP7U/tOvfJq8o08IQHvwFbz3buQgzHMhhUSV54YNBwvARztGADb7XLvfnhCA7W/AADxBJF0zKbGxctoGAgA9lGMAvpSlpHkBRwCAmTMAHnFh3cPSAugCpCtwGDZY67jznDHutfXOD8Oz5yIwwVHDDRN7AHxRaAYAgHCngVDWd3URPncgddcle01HfT4A0NT5OQGw9WeU3z9HC5BlRMJrASCjiz6bji4QhuEIQKuMC05w4Q8WgGkMsjR6QMAyIDABQBeTjGjLZgAwuCpotgj/8C9ZAKsvMMC1wJIrdDQY3bOmKRpAmy2VQgBI/0n+OFgAUnmKD6LAhvv3B/YMAHiKgz8eWYBZ5ZfvNDSG0TMSBU+6rmLuZXTjSJPS0K21niMAHh2i/cCF4rrX5xbgSaIF2IL+EwTRG2KeSYn8aJy+0pO/Ig4AeK9KzhA4ipuN5w9n2o/G3D8uLIDqDqqYdL3zkknFtPcO7AzTAzNfHgsAVKkxMHsAgOElmyghhA6lBuloUDf0g62pLc+mRTJbgKWPU9dR83//fRyZkawGWLL94+9YAIVa7w8AcGe9wDEL7noEgPmDC6S9YU6Ta+ONFPKe7qjrsQpMcsb+PHQbAgBjkQft7BkA9/czAFX7faREUOmBzwUDnEZxBJdGlELBhkJivMnQjo9ePd6TAcDnEgD4FAlYQvAxvj/boXeG0dcfL1oAXMzvaEJQ66dxsk/4i3MvUOtGi4IxrB4pkhTMS3TY8KzgubKD5hMA61MAnpAAfCcGJKdZG6SUHO0BMEQEHAAwphRoJdgI0I0au0ij6tMzLu6ZUI8zACKJCglvXW8HZ4TyA+PSej85hwxsKg6vBdvxfQFxAAueUboOb2QvCu28NxLEJhMBGT6T2QJgYo7+TlR9kNEbLhSI7JMuySjPG7hABwBKIRgCIGDN9QxWzMICGG5oFnW6eoo57GVDDxMBwJb1YAEQABPQAuhzANgMQONG/T1k1HZ+ELFsOu9cS4EACCk5p0shEH7mGNPBOvElBracls0CADyzk/5jBz7pP4UF8GEAsd7TtZ4nFiDK8s9CKAD+FGMAyD0NFMwAeKbtLNo6xkYnscCCYgAhAQ0uBACwXe0L2G63vN2i2sMzekLwLpgbkQFyb4SZbq/iBwAEpAEbQrrRgUjJ7lHpo0QEHgwdsBmAOkn2AWldhC5o7YbhS2mG5+dn/QxOEBtI2wu7AGAkx4VG48a+Z6T32mI1uPFQys8LPrqDBSio1V+0+3vlF5LpAKYo4oQRgfs7LpBkjjnPOZcIgxvYBIBBKNR2W8UNn5ykIIsulBu1tY7+Ii90AQGA3AsAHqSxda+9OwfAzQCsjP8+gojOT+2MWtFPCov1pH9DVpJJTlaAGwwGsFtbYMsv4XU8DoCp0QuK2ny4zQXesA0YHHvGj6cxADX7s/rTkvChyXbmP81ZjAMIIECDYMF7O4wIM5oCb8h188w55gFftZ27QZHndus9IaDIF0JToMAMoHohBswwR+tstgCMGYHerEdzwwCyh0P7/0grYKgEpwEGcXCB6iTq6aR4dYHXI+tQFNQQhFD04XdYdsE5dgiCuWN4EyRDv85ZWADA3qL++xF0u4fXpOf4OAAA61FGzCm2SigBTwN5Qv+5Tj5jzs+Y9W/cElkC+R7DEgkEoAvEBqaEaEvhtUcA/i8esV+3hoPdjQxI75iR2F71BYDtEcuDC/TwpZROY2zs/OMpAHrxL5Hp9xGKAeJDO1Oi9pOU3jjHdB9GVUqvFKfhmlYA00PQ3vA42vsAsowBKOlwbws8F406XgpPAFDauQVYRr8EAAkYScQoyvKOMKYtdlChVzJYiyvW2hgZ0LmSxhvOSwRg1i9ZopvzZMyTMoZMwHaLKAADxjP0cWiuG4CnmgMlAAv9I4mWgnHFHxaNP83hgadZa2etnGOA2RlH3SvQpy+Kou5C6AMkhv53jUELA2bnAo4jbf47JDJvYIE2FwtFeBPUDAVB3+tqJQ1hH2caItkKgUiS2JqZg6hXA6CctiMSIEHK0nimGedCoQvklxag+u0J7G4UhACLhySMQ689GqY2bmhMyan3gbral9qP4TF7nAGI98i87YPO7V77JWr/TLtR0qP5cgJqqqY5nmhuAyE9Hv9lP//sAkUmohcUW/75T9JCGDR7BipOYgB6HBCIn0+nyiNZdoMap0FQYfWgnZT0cTJm2I5OptrIcrs9FJAz+CyE/K/Hdn4cn7bkAinkYIKAvnFGMT9bAK8ZQ9VDi46drkpAy08PXOB1Ho5mFYKHngGI2k+tNa0AEck0XnUm230BHX7qKGPs04+C60AUBcTFZ4AH6GkOBWSOZp+T4oIoMdaLL8pXAyCmsFbrKWByeDjMBAAFA7MLBABsKb6alk9bT60K8ujxtIgmAgCrAK0fgW/mWWxU4vOBCfXlH+kFsprJlrR/Fm4Zup9maqBKEkBAihZQMB4vodzLmQsUh64muY8gPCMAXW21hRXU7+WlECSw0fwAOULg3AIo6vVxkwWwjpWlRxZGEPhI4ZdDvWX8iR8AkMBJyUf/BGTgSPDoGU62oFD742u7ffJGKOncoYHlbrqNzIEwdAAAgOkc4an68ucXMwz95NgU5qAu3amW15MkUceXchQDdPX0dR478A/vGS1pnXr4Fy4QCKDB1EVhtVQCHiB/B4CW7r+zKIPFmImadVEqj3HBEoAtYPGETUsMs7bceAcyMupKcb7cH1ncA+eQ5Aw3D9Gm4hIDTM6N+O7jAFoz3lZnPyN7QN1TrwhzFAMAAAIJEGWjcHQ7ChJwBgB+vezRASEAQqedZR+e74+7QdHVxwyHLJfVH1NnC4BASqMJUKcNQGrIfYGnduSQTH02GhA4BMGCpp2y7ok79jQGS72o3ott7BrFpXgyEiFnBwCEh31N7i8iYLj4ghE5DGzAG1oA34OEAjAIBwBU9MtHcGpQOeO1CaDgv5/p/zIIrsFlx3gVgIld94den/0oGNmCBQANZCFhcRY6Wi4A6LgiAv6eBWiU13rEQAkQoF4ZokBKNI2Gq62MeRvvMWZ/Mk9bOWGwnTxM6kTwwMAeAKYdDcLAF2YfWeEDFAmR0dp8bwAMaf+58B5tugOyAQI67RKjMa9KASulos7+ePPjAoCD+pOXQ0EAvuMLAAAF0VrDPbqYe7YAqP1oNP7z/DH6P6cmYP5wfDm0kKEIjgszSgCAMzQAFKeWSkZzAICbuRdIamOEYOyJs/HJOPcEp2krDJfy6Wm7hbANT5Y0HACQrJoBMH505H4b40Zk7oEcoGegH+QLCzjfAj3NPBIsYa+cpn6LJMBC08xtv59YhhkAWxMogAtdxokodPvr+dE20GfIQV7RAQDYkhJYW56IUmgpAQCBghi0r78UQkk3uZxWO8n0OPUhMoehgDGzBWi84TQEjB1zT3hkt9EWUJowTO3Vz3mGSNDot3iInQo4+KPtgJYGQvbvK1V6OZ1brS2KY55z6amh1UWQ04x+6kvs6yQT0Px/e9fC3DiOo4Oe27sqck8gQbHhiaVV6WbGtSV1tnpz//+/nT5QL3d6ZifVlU3VhZ8dPZ2O28ZHACQeOwGK/BsOxxYoBFiQIv8GApycYCgAaxb4XMT/JY4WmnfRoK1nyygJkkssEAsGaCGwIY/FQMo8nwiQRP005nnSPFsOcZpnUi80K+nsnH1ZxhGnOwFa8UHtu4uZ7DsjWSZn4xBFLPhbyxoPamDtPsD/iMyy/TALCz31/a279SluNhB2ZtTvBBgu1zLCQw3gVgGOLthB9IHvEuDK++CPnw1erxJ2h+DxRzLCWMdsH4O0XIpeRyJo2TxGkjsCjAr3dohWygva1TA54cAZBNj67pEoLEt1yypKocCqU/sh9fRuGWHRJlWGbkgxi4iyF3jrUmr5eeeXxoBrQcNzPgDEf+2Kdup/bAQYOgCUeoZaOBEAMALsw//vkuC+8FpwGIx84EKANuSmTy6wtI+eRhP/UdtpOplAoj4QzfMYZx5p4ggeBMngwUw0C/tJPIMbZwJ4sTURUnKkvA4KMUKyRf/+3K3in84aQFhK4VemjNESAgyYINtgXiZKuxy3yVMQoCxpYcDHC0ER7A+H2MKbsTliga6rinAm7ybs2x6HdM0BJ5gh+hEC4ANPXWRHKq0nU7ej82ryn7LuBMjCHh+nCsTbSnnNM0gAArgguw8Q1ISfsCtq9ZeyxIowZEDfjQBkJlh3ta6YCzIXh0DYm/Hh0BwTi10LznWnbObShH7vkYk95L/YCDaNcKcBECwL2sTnu9H+e7jXABjcMmXfupUAXocu+ZZx4pR0AX1LAPJBdJ7zaASYE5SBZ53dyBMUAdZwPJMynwggZEudjgOzqtdcpghGaARmtQnevsf/T3eXMSbLL+hGJQTF9bcixMUR3mKVcZRoOAhwzPpc8VyPi+Djp2zxfKEBLtF57yDx+MEzFMRPA6jxwxoAy+t5GFxQYSx/DUlZfculB4JM4SCAFAI4Uh5nFUczNAH06+xkD/77CZONpZeDkmZdjIOUmRetmiEpaRjekQAiqpSaawNjqHRA4AWkKuYFgAP8i4n5URXiCGp4ti3IYNrACLBW1skxLjfOBCgi/7Jb5ksi3FWGg6Uf45kATDG6lsvJmBJDG0zzEW1MnB0IAFlnkonHeR6zZ5on9WhB4bKbZi9dlIMAwWbvaBzYefXCQUihYAgeoaXIgNsFepRFwX8XVVaYbfawO2Qb20IDoO9up1AIXN9RZn/2s/WaHb4kQGKY4rKJfSGApSR8atQvcCL6I0nxyOoRpCyLMIaYodNQZoEW5CN45EFF8SFjokHncRZ1CgJgM81OVfDOCwFEvS9LLCUGXGJWbK2p8NDROybFl8Ji5rfG1NnATcpsORwuMDKZfJm6OapC7PJvO9MP9mPlX3rENQ459el7BPjluU9rFY0/wH35/SD9rQEByAiA3PEMAihWLByNHbfungAi2QXMfWLMBwHyPMfRY0c8OZqm7NzsZYjqDgLY10KpgwfgxXmRqYSPkhNbNI6DKbgFBwEsSinlkR/bnNNTj+EcKLP5ZZ5n2QKfDifYRPwwdQx2tKuAPR70IABCIZo+mV5azSAARg8rOzU9xUp4R/IjGgCGT9oJQElb9rgY1bM7nGAR8iyQeZI5z0peZ+Fp2YAAYxZ8IYUASqKRPbN4JcpEGB8pD92AVjrvRwCf87r0o1r6wo7gATggTlg1dwO3bZH4bzWAmTyrg7Alez0nJKxe+0TxOqT/PhNgLZyUukT5pQ74XQ1QYv5tqc7TZgKNsfgACxxrkm8I8BhcdsHxjJVgxnhEs8uj9zLPJPAIigbgMctBAO+kG0XHnkjIQ6DE2YLBrNMkwqosY9d3UJRyROvhzfTmNj3ZIkCZDN2FHubPJum8vcHhMHRwUCZCT4K/8QE0anYfoKR2MrMzyXcijLVgg1vAQkTZvsH0AwSwHKSnzrWFAF6yBpR6D5opnNcBnJAXnVkc6DdT9jrPMqkwCJDjToDASHTrspBjvBaqVpggfMkGlPcjAAlngFRYKA7mUpoiGAljIOUoAQS40wDHpM+wE6D4BQiHvXy69pHyNcVf7glg9xHOYCrgT2qAYFlfRBRatxEgNs0AEygYAaQQILidAC3LqOzDNFthIJrhDlB2HkaR/OpWAjiOWU8E8JIslUXHMeKbIp4KePp1pJkhXF2CzZp41wBpkZX+1iEUwnJ0YfoYDpNmOzoT4ASEMRsDcFSwKhHUxFY+FiqK6f8CsH9EV9nHnMaCHyJAMBOoZSMAmwnkYIkKtc6HnQBeMns3O1cMH9BhZplIZhBA80EAl0ehjnJmFnKT+QiTshDCxt/TB/DqvIXX55hVKaZs+vPaDJboILY6oPzLCw2whXT+bcsMg0jDF94IkPM15t/uCVDq6KUh0p/XAFaOyQsRFR8ggBC57wbXOjUC6NgVDXAKh55mKyRnMa4OattbvXGn8+SMABMVAkQ9RYN6jiRMHT2l6JjzKJNbCSBNR6ASxoa0gHcN8HQzgABdqU1iQr0/zhfuNYBdLQdHoD8uwNBB8ZMsbkE4bLTvCr8r/qWtBmKCyvADleEe8ImzjpsG4K5fCaBjviPAJNlNk/NOWJyfbBZo4om0EGDXAJ+Do8w65KfEjrLDh+rc1z47xA2lBfJeBAgugNhKhI8wjUntw+ybUrE7Zi3Bj/T8fQ0AJxiwMg9FwLuh+YROydRkuicAYMtkiZ5/p2Fyed7NAgFeSLM/NACllMwJhg+Q+0ZaL204m0AT+S9fgg017AsRvF+eDqX3heZZjQAp3RFAIolopzRG53iMPLGIzE44Nf0g0zR97Swo+2QCpadkEdbCzEJNGb9tND+I8NIEsqvrnUs39pctzb25NU9ZxWp2FSu/PaapwgntKvyUgX3kB35QA+DvsmTXMhsBYtpMoC6GOwI4BGXYZ8zz5ByznyY/q8xiBODHXXXRKDpojuxydixAalAPezYCvJsGaL3Z2GC8Uk5jKg3OxFyDzRuwREJ6qQHwyBD8AmPASoAuZklZ700g0xHPA9w4RMq9FP+DBWcC4P0VAnjyXAqiZ5SVcmInOvRGgGn+sv7STw4vD8JutgLMpEXnihPI1Sxunp3z4qCmTwRAoUXH0rHQCN0X2TtRFlWJQyGA9F1ZujwlLKhKWQYQanZ/Fiy4m9BfcOcDHL7vlZxGhI8+RSIRdrDrw1TEvD0I8NMR+zBB+IUoG2JMu/T/bXl2y/O3H/IBvEjcCUDwAdBDSFIM3oV5D4UgCnCymJRkWoDPl2cbb4SZdgLgC3QcWUeGLjDvUvIwYBSyiVB6RwLYZ7wmAeWkosUfWMOaYiwTo/7vdxpgUwHPMOfBhG2DshgrATK9JMAvz3/trk3Mz+YxfJ8D9xpgnfakTL71atXrHj0xZXOCkcMmsddlG8J85AMQZe+J1CYyLcHbvp8pymyyyuyd+yoswqdZILQedl4Ss2S2rEC3umyCSFS8GAQAA05OsIVWkyr1o1yLKV/Efx38j82ZAPvdy6XP9geyTH5ypbj4CWcChC8m/95DgMhqvcRD+gGbVvxxH0CUMgdmh5KxTKTicTUmhNnts0CkGpQsDLqoQDjBGRpA2dzKPRjOsTomYXJOCwHEam+pUNd376gB/tND+AsJYKhFYVaiwgFVqIHOaggqnzXAHgD3nPJq+wA4WAjQX0AAHV4SAOsGQ9Nn/Q0EeGH72JEd3xPA3glxcOrZCCGeMpGsJ7FRaIAvR04wJ6UQNPMsZtKB1LMKe5Jf5xFf0kzCY7SUgKP2rGMQQEnFEQeWQgAmVs2YvFLREbHeKR4EaFP39GQ2kPaD5N2LLY8dOLwnANIIrD6jOJgRW4R/wANow7r7cjaBfAm+AefuR34M/Ov+RwgAH2BtbOmDMkPsJXUdeeiCFINzZwJIkFzWFSWrQH3OWUhn4hGh6lsskK5dxJUgX9ASyqIxkgh1zXsS4BE25orgYfj54hLkTLT6A9chLhif/3qvAew5DPEZcn3uFN4XAlAiuScA8Pw89IO1G39h+5y7MX6jARY4a6VYfABWRekBZfDWy9BoixvTQYAs6lsdZNIoTIPMpBj6vepMef5KMpO4sURR7ASgbFdKirkEdiSwk0SdzQ9YZv6IVd87DTCORF9VFjz1o0vGgE36z4bQ5RwMl9ERQ9i5CeYODJrwAu323AngmHUz+e+FH+iwiT/mBAOtiKiKM9PY5iBEs9h6pCiCLDZGPuBFkqeZnJNJCOMTT4RVAPakcgqG0xEUKKpSbH7O0o7EsaYujVHekQBrLURz/VlxCg1bsgVlQR4ipOL5vBK8CfwzCGBHBuxRetCa9cdRv0eAYSj9J+iF7YOfrSndUXbHAd54GlSdtguc0y7lUTngRHLUwHomgMsqPkiUSbLMPLopp1lHDiq/6koAxSKgEh0EEB1TTiZYUHpkQexQC84UdlYWyUaAu1ggFTaznbu+o3z5y2HyA5dPa72HbqRTo+xN+IEi7WfgzMZh53nvLgyGAuNL4TekeL6oP9IhBnj4vKQSfv7pM/CwnGH3YDuLrDRd8YBbeH7GER4Pn3/GZjkDtmQotJlB25mfbYd/envBcga8W8eyf6Ip+Arr5/m4HremBD34YQduCd3bw6H/9x8bRBi7M0REpZjZIZx6gH+2uyy84h9/hH/u8mUiuMC39tZsg9z94CW0OPXBElbDcv1Iime//E+chi8TSIJWSmIxwm6xovnLl2W3kN45iJg7Blhvf8s5Z2/fAR4V8r744N0aGCJqHhLGNsMDqn+WPCK7Q3kHbRD2qBL6ePQI+xljzoaf7UvYDoByWoKxnNvXAeyMzxA8RH6T+2sLQm2S9+eAPuDY4Nh29+3BcWE7Pn7nT8D+MdttsAsbcOv38d2/VU7xs163fxLPu1Zrx80F61G5cjxxcX/B+Q2WS3c43jq256uG43P7Axxv6h7rDTzOWeHlxouPfTv/nb/x7cdfCVBRUQlQUVEJUFFRCVBR8WcI8PB4AB65NQn++WGdA8JmmwVa7j/gPnz2B/TI32d78Iu4uL5weckO3DQ8lJ9HLOf//F8VFRvemQCPzKKqtCNjXc0Hb0HY7rQOQIR4lImzY51oyORZJoSOT+RU3Bi3nODAmbItmY9ZlERUWJxm550gwQIpkRUV/z788UrwlnMPoXekGtOguldMUrcTIHNAdh3NEPws6lhndTPN5GZxQ9oI0DrKhOdIKkSi+pVIKIqbvyIhZqgEqPg3449jgdZCK5Z7wW5II7dlYY5ZZCeAEnsn00Tq3CRZxDFN4hzNyrP4Ie4EELJFR80kjpRFhEhpIFHq+j4NVAlQ8Qq8PQEkE7ERILPPOSMKF4QA9pxgz+Icz85SjYhmgSkk87wSII17UjwUhygpCQjgRLOqUlL9+hXBVZUAFf9W/Ot8gHy9Nhog9eoIhcBaHLOqX/Y7AZw471VpVMopEkL5ENHKMztRP+adAFQq+4lYYFVgIVHRKNAATdMNuRKg4hV4Ux/Ao2EuXS8gwALxse8HCTgZm44fH3cCTLPyNI/IE7X0KWHvkCKGx0wubwT43Go2D0JY1IlaFj+JJhIa+0qAilfg7TUAR7Q6bq69WvSTWgKqQAG02nQ4gLRaf4Bfiadfc4699STPk9UQ+BUPN6ujvJVFCUIWXamsRgDJKmNOTZdu16YSoOIVeHsN4JIPAQSg0HrXSqCuS9ICciYAwrOJPSvnLjXNOCq6lGyY1WveUyJFS7EWzeq88tg9fc1jaqxwRnPtKwEqXoE31gAuuRC4EEApqBf4ABalzU0Hu2gzgcaY2Ss7GVLTkIowJyGax5mI1TvZCOCckmYiS7V45OF6vXapG5rLmieR6jRoxSvwxiZQ8qF1XdPnEDIFRaHQTAoCuL6jsDvBnymOslzk1MWmGaO0ToY55znFW5/JuT0lUvLt9tSlnG6dko7Xi6XD3S5r6aNPHVcCVLwCb6sBovPI7umzD8xBPQgQx9KJsqMjSflBUyEAeqJdU6+tkwQCoOvNbfSct4yw8QqJ79KSHTTk4foJuPbXkjx9WY6lEqDiFXjbaVBkMjkjAJbDNPjlUDmEL61LHR0m0GdOI4dHI0B3TZ22nmH9iDXJ6ViTX/9WbyVfetsNXUmXu9wupXwAECsBKl6BtyWAgACDaYAFCjKMHCw2giKx3wngc+bgmfsmpiYO5NiNSppKB/vUjW79W42ZOn0Deb9eT5UCtprAT5UAFa/A2wbDKdu6Vp8tF1t9K6q8HHtsNKnfTCAmZc7KVk0n9oOKkgj3ZWC/9R3vBMCFvX/UQgb8LC+rBKh4Jd6eAG1MXb/gOhCz8MguZxZS4YQLfd57hKWBCKRgNOaI1ybHHrXOL5Bqs+3lIAAuHTAv+FIaAYIuXSVAxSvwtk5wcynoE7G6rBrjkLtn8QP54LtM6+8+ds0QsxKNvZWFvxpuDSTa8F0ClLbK6KXWL7KPvoCVABWvwJsTQLdKXtchZvZjZs35F11apPWJNPdpWNNXHq0Q/LU0xWzQG7+U/TFxN8HeCHAtbu8Vor/3ufd0uZgKqASoeBXengCXiz2bpk+xe/b5eQGKbkHcu9RtBOhvt+4JrXP6rMI6sPeseeyMFwBvs0AQfeBpHFWQWOOZUoMOgcDtOlYCVLwCb0oAWSt5Acu+6fqmt04uffdk/URl8wGopHixZGFmGVRkjMtjwRO2e2lEGg2giSv1Hzkv590ljUAc6zpAxSvwtrFAYkM1TBUclFau5dLWE3zzAfACXLz1t2Yf9W89mPL0lMZxb5TNSkTK/oDDRq9sh861lQAVr8DbaoDbCRfD9a7p/dPhA9zMo4WJb/iPsgGwEnDb1gGeTJdQ+AbSsFVF9b4SoOIVeFsCTKXar9INWGdsbgeuuw9g8/rXRdDXlpbF1d2avy5Hmw9wKw3uW8PR5YBBgAWVABWvwFsT4ItYa0ERxeOp62H13I6O97eNAO21LGlB2q9213yHowPsMQ1qBBATfTyWJ8C9gAAo2loJUPEKvO1CWIndVxljSj1sF5NqAOJ+i3nzAbaegNfSCbY0uj+DdwL8ZW1w78NUunvjIbCKPFAJUPEKvHFCTNdj3ocUBUuua7fvsgGaGA8TaLuLwOaCc//XewKM1kQsLPBlG/TTAA8AjQArASpegTfVAD6OMcXswzg+dSbhxci5rKte134nwCHz5vvuJ6tPsBOgx29lVWG/97Z04iRmLm3+6jRoxSvwxgTIOQ4pOvd0u2xCD+wdv6+bD3BZBd7k/2jwjR9TDI0rL/ycMymzcxB/3pp7Oy7tuGKMNSOs4hV4awI8rfP7t+vR3+9s3jSbBricNMDR6dsmgdApmnjvEwxX9wwoAF0bkQ7DUCvDVbwCbz0Nilkf63uZm73HN56bjdO81AC3/rJLfsqkUhZ8t07x7b3ws9Iu/EAtjFXxCrw1AQKRqgqTxMvR8/LU/v7QALiDS5csY38S/WBo24etjPQ68w+7hyjfC3/VABWvwts7wSmNRNrltenrKvrASw2AQNBu5MmxuGLbhwMPJw1QutpTjntT+6OzdyVAxSvw1iZQ6fstqdOhyP8dBS6XfRaoabqx9MgM31j5LZ7tbgLB6jm6uw4HuudUTaCKd8Af9gm2CRuXupivewoXgEi3YSTeUyLZeev1Wrq6AnaAB6ydU1kU2sR/wbBS4NlOqxNc8Q74132CH02WS+DmtnzVlpYwW4eY5fB3YK1fHvcOMTh+bL/F6UpdCa74Abxhj7AqmBX/P1Gb5FV8aFQCVHxoVAJUfGhUAlR8aFQCVHxoVAJUfGhUAlR8aFQCVHxoVAJUfGhUAlR8aFQCVHxoVAJUfGhUAlR8aFQCVHxoVAJUfGhUAlR8aFQCVHxoVAJUfGhUAlR8aFQCVHxoVAJUfGhUAlR8aFQCVHxoVAJUfGhUAlR8aPwfIz1qu1rgVskAAAAASUVORK5CYII="},"d6dbbcb0-b038-485c-99bc-353028cf4828.png":{"image/png":"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"}}},{"cell_type":"code","source":"#@title Variables\n\npatch_size = 16\nattention_heads = 4\nnum_blocks = 8\nhidden_size = 32\n\nLEARNING_RATE = 3e-4\n\nimg_size = x.shape[2]\nprint(f\"Image Size fresh off the loader: {img_size}\")\n\nchannels_in = x.shape[1]\nprint(f\"Input Channels: {channels_in}\")\n\nnum_classes = 1\nprint(f\"Output Channels: {num_classes}\")\n\nnum_tokens = (img_size//patch_size)**2 # num_patches\nprint(f\"Number of patches/tokens: {num_tokens}\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:52:57.295303Z","iopub.execute_input":"2026-01-14T13:52:57.295626Z","iopub.status.idle":"2026-01-14T13:52:57.301433Z","shell.execute_reply.started":"2026-01-14T13:52:57.295590Z","shell.execute_reply":"2026-01-14T13:52:57.300694Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"## **Patch Embedding**","metadata":{}},{"cell_type":"code","source":"class PatchEmbedding(nn.Module):\n    def __init__(self, channels_in, channels_out, patch_size): # c_out = hidden_size\n        super().__init__()\n        self.patch_embedder_2d = nn.Conv2d(\n            channels_in,\n            channels_out, # num of kernel-filters\n            kernel_size = patch_size,\n            stride = patch_size\n        ) \n\n    def forward(self, x: torch.Tensor) -> torch.Tensor: # x: (b, c_in, h, w)\n        x = self.patch_embedder_2d(x) # x: (b, c_out, h/p, w/p)\n        x = x.flatten(2) # x: (b, c_out, h/p * w/p)\n        x = x.transpose(1,2) # x: b -> tokens -> token_dim\n        return x","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:53:53.584588Z","iopub.execute_input":"2026-01-14T13:53:53.584935Z","iopub.status.idle":"2026-01-14T13:53:53.590484Z","shell.execute_reply.started":"2026-01-14T13:53:53.584903Z","shell.execute_reply":"2026-01-14T13:53:53.589784Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"## **Transformer Encoder**","metadata":{}},{"cell_type":"code","source":"class TransformerEncoder(nn.Module):\n    def __init__(self, hidden_size, num_heads):  # hidden_size = token_dim\n        super().__init__()\n        # norm -> mha -(+)-> norm -> mlp -(+)-> return\n        self.layernorm1 = nn.LayerNorm(hidden_size) \n        self.mha = nn.MultiheadAttention(hidden_size, num_heads, batch_first = True) # mha_dim = tokens_dim/num_heads (should be divisible)\n        self.layernorm2 = nn.LayerNorm(hidden_size)\n        self.mlp = nn.Sequential(\n            nn.Linear(hidden_size, hidden_size*2),\n            nn.GELU(),\n            nn.Linear(hidden_size*2, hidden_size)\n        )\n\n    def forward(self, x: torch.Tensor) -> torch.Tensor:\n        \n        res1 = x\n        x = self.layernorm1(x)\n        x = self.mha(x, x, x)[0] # mha(query, key, value)[0] = context_vector at zeroth index\n        x = x + res1\n        \n        res2 = x\n        x = self.layernorm2(x)\n        x = self.mlp(x)\n        x = x + res2\n        \n        return x","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:53:56.170352Z","iopub.execute_input":"2026-01-14T13:53:56.171097Z","iopub.status.idle":"2026-01-14T13:53:56.176632Z","shell.execute_reply.started":"2026-01-14T13:53:56.171066Z","shell.execute_reply":"2026-01-14T13:53:56.175977Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"## **MLP Head**","metadata":{}},{"cell_type":"code","source":"class MLPHead(nn.Module):\n    def __init__(self, hidden_size, num_classes):\n        super().__init__()\n        self.layernorm = nn.LayerNorm(hidden_size)\n        self.mlp = nn.Linear(hidden_size, num_classes)\n\n    def forward(self, x: torch.Tensor) -> torch.Tensor:\n        x = self.layernorm(x)\n        x = self.mlp(x)\n        return x","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:53:58.520424Z","iopub.execute_input":"2026-01-14T13:53:58.521072Z","iopub.status.idle":"2026-01-14T13:53:58.525571Z","shell.execute_reply.started":"2026-01-14T13:53:58.521039Z","shell.execute_reply":"2026-01-14T13:53:58.524842Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"## **Vision Transformer**","metadata":{}},{"cell_type":"code","source":"class VisionTransformer(nn.Module):\n    def __init__(self, channels_in, num_classes, hidden_size, patch_size, num_tokens, num_heads, num_blocks):\n        super().__init__()\n        \n        self.patch_embedding = PatchEmbedding(channels_in, hidden_size, patch_size)\n        \n        self.cls_token = nn.Parameter(torch.randn(1, 1, hidden_size))\n        self.pos_embedding = nn.Parameter(torch.randn(1, num_tokens + 1, hidden_size))\n\n        self.transformer_blocks = nn.Sequential(*[TransformerEncoder(hidden_size, num_heads) for _ in range(num_blocks)])\n        self.mlp_head = MLPHead(hidden_size, num_classes)    \n\n    def forward(self, x):\n        bs = x.shape[0] \n        x = self.patch_embedding(x)\n        \n        cls_token = self.cls_token.expand(bs, -1, -1)\n    \n        x = torch.cat((cls_token, x), dim=1) + self.pos_embedding\n\n        x = self.transformer_blocks(x)\n        \n        cls_token_ = x[:, 0] # (+1) zeroth-CLS vector across all batches\n\n        cls_logits = self.mlp_head(cls_token_)\n        \n        return cls_logits","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:54:02.648497Z","iopub.execute_input":"2026-01-14T13:54:02.649149Z","iopub.status.idle":"2026-01-14T13:54:02.655496Z","shell.execute_reply.started":"2026-01-14T13:54:02.649117Z","shell.execute_reply":"2026-01-14T13:54:02.654800Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"### **Two tensors can be added if:**\n\n1. They have the same number of dimensions\n\n2. For each dimension (from right to left):\n\n    * Sizes are equal or\n\n    * One of them is 1\n  \n**They add because:**\n\nDimension 1 broadcasts across batch, i.e. One tensor is automatically reused for every item in the batch, without copying it in memory.","metadata":{}},{"cell_type":"markdown","source":"# **Train**","metadata":{}},{"cell_type":"code","source":"device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\nmodel = VisionTransformer(channels_in, num_classes, hidden_size, patch_size, num_tokens, attention_heads, num_blocks).to(device)\noptimizer = torch.optim.AdamW(model.parameters(), lr = LEARNING_RATE)\ncriterion = nn.BCEWithLogitsLoss()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:55:03.224726Z","iopub.execute_input":"2026-01-14T13:55:03.225502Z","iopub.status.idle":"2026-01-14T13:55:03.239538Z","shell.execute_reply.started":"2026-01-14T13:55:03.225470Z","shell.execute_reply":"2026-01-14T13:55:03.238920Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# print(model)\nx = torch.randn(8, 1, 256, 256).to(device) \ny = model(x)\nprint(y.shape)\nprint(f\"How likely (logits) the model predicts each image in the batch to be pneumonic: \\n{y}\")\n# y.argmax(dim=0)\n\nprint(\"Our batch size is 64, yet this works. batch size is <= 64\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:55:08.330273Z","iopub.execute_input":"2026-01-14T13:55:08.330539Z","iopub.status.idle":"2026-01-14T13:55:09.307320Z","shell.execute_reply.started":"2026-01-14T13:55:08.330514Z","shell.execute_reply":"2026-01-14T13:55:09.306367Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"num_model_params = 0\nfor param in model.parameters():\n    num_model_params += param.flatten().shape[0]\n\nprint(\"-This Model Has %d (Approximately %d Million) Parameters!\" % (num_model_params, num_model_params//1e6))","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:55:13.384260Z","iopub.execute_input":"2026-01-14T13:55:13.384579Z","iopub.status.idle":"2026-01-14T13:55:13.390585Z","shell.execute_reply.started":"2026-01-14T13:55:13.384550Z","shell.execute_reply":"2026-01-14T13:55:13.389832Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"from sklearn.metrics import accuracy_score, roc_curve, auc\nimport time","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:55:13.684901Z","iopub.execute_input":"2026-01-14T13:55:13.685497Z","iopub.status.idle":"2026-01-14T13:55:13.688908Z","shell.execute_reply.started":"2026-01-14T13:55:13.685465Z","shell.execute_reply":"2026-01-14T13:55:13.688168Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"def validate(model, val_loader, device):\n    model.eval()\n    all_preds = []\n    all_labels = []\n\n    with torch.no_grad():\n        for images, labels in val_loader:\n            images, labels = images.to(device), labels.to(device)\n            labels = labels.float().unsqueeze(1)  # (B,1)\n\n            logits = model(images)\n            probs = torch.sigmoid(logits)\n            preds = (probs > 0.5).long()\n\n            all_preds.append(preds.cpu())\n            all_labels.append(labels.cpu().long())\n\n    all_preds = torch.cat(all_preds)\n    all_labels = torch.cat(all_labels)\n\n    val_acc = accuracy_score(\n        all_labels.numpy(),\n        all_preds.numpy()\n    ) * 100\n\n    return val_acc","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:55:46.890180Z","iopub.execute_input":"2026-01-14T13:55:46.890504Z","iopub.status.idle":"2026-01-14T13:55:46.896398Z","shell.execute_reply.started":"2026-01-14T13:55:46.890475Z","shell.execute_reply":"2026-01-14T13:55:46.895591Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"epochs = 30\n\npatience = 5\ntolerance = 0.005\n\nbest_val_acc = 0.0\nstale_epochs = 0\n\nprint(\"Training started...\")\nfor epoch in range(epochs):\n    model.train()\n    total_loss = 0\n    all_preds = []\n    all_labels = []\n\n    START = time.time()\n    for batch_idx, (images, labels) in enumerate(train_loader):\n        images, labels = images.to(device, non_blocking=True), labels.to(device, non_blocking=True)\n        labels = labels.float().unsqueeze(1)  # (B,1)\n\n        optimizer.zero_grad()\n        logits = model(images)                 # (B,1)\n        \n        loss = criterion(logits, labels)\n        loss.backward()\n        optimizer.step()\n\n        total_loss += loss.item()\n\n        # --- compute predictions ---\n        probs = torch.sigmoid(logits)          # (B,1)\n        preds = (probs > 0.5).long()           # binary 0/1\n        all_preds.append(preds.cpu())\n        all_labels.append(labels.cpu().long())\n\n    TRAIN_TIME = time.time()\n    # --- epoch metrics ---\n    all_preds_tensor = torch.cat(all_preds, dim=0)\n    all_labels_tensor = torch.cat(all_labels, dim=0)\n\n    epoch_acc = accuracy_score(all_labels_tensor.numpy(), all_preds_tensor.numpy()) * 100\n\n    # ---- validation ----\n    val_acc = validate(model, val_loader, device)\n    VAL_TIME = time.time()\n\n    print(\n        f\"===> Epoch {epoch+1}: \"\n        f\"Loss = {total_loss:.4f} | \"\n        f\"Epoch Acc = {epoch_acc:.2f}% | \"\n        f\"Val Acc = {val_acc:.2f}%\"\n    )\n\n    # ---- early stopping ----\n    if val_acc > best_val_acc + tolerance:\n        best_val_acc = val_acc\n        stale_epochs = 0\n\n        torch.save(\n            {\n                \"epoch\": epoch + 1,\n                \"model_state_dict\": model.state_dict(),\n                \"optimizer_state_dict\": optimizer.state_dict(),\n                \"val_acc\": best_val_acc,\n                \"loss\": total_loss,\n            },\n            \"best_model.pth\"\n        )\n    \n        print(f\"Validation improved to {best_val_acc:.2f}%\")\n    else:\n        stale_epochs += 1\n        print(f\"No significant improvement ({stale_epochs}/{patience})\")\n\n    if stale_epochs >= patience:\n        print(\"Early stopping triggered.\")\n        break\n\n    print(f\"Train time: {int((TRAIN_TIME - START)/60.0)}.{int((TRAIN_TIME - START)%60.0)} mins | Val time: {int((VAL_TIME - TRAIN_TIME)/60.0)}.{int((VAL_TIME - TRAIN_TIME)%60.0)} mins\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:58:41.091341Z","iopub.execute_input":"2026-01-14T13:58:41.091666Z","iopub.status.idle":"2026-01-14T14:07:47.310606Z","shell.execute_reply.started":"2026-01-14T13:58:41.091637Z","shell.execute_reply":"2026-01-14T14:07:47.308961Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"# **Evaluate**","metadata":{}},{"cell_type":"code","source":"checkpoint = torch.load(\"/kaggle/working/best_model.pth\", map_location=device)\n\nmodel.load_state_dict(checkpoint[\"model_state_dict\"])","metadata":{"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"model.eval()\nall_preds, all_labels, all_probs = [], [], []\nval_loss = 0.0\n\nwith torch.no_grad():\n    for images, labels in test_loader:\n        images, labels = images.to(device), labels.to(device)\n        labels = labels.float().unsqueeze(1)   # (B,1)\n\n        logits = model(images)                 # (B,1)\n        loss = criterion(logits, labels)\n        val_loss += loss.item()\n\n        probs = torch.sigmoid(logits)          \n        preds = (probs > 0.5).long()           \n\n        all_probs.append(probs.cpu().view(-1))     # (B,)\n        all_preds.append(preds.cpu().view(-1))     # (B,)\n        all_labels.append(labels.cpu().view(-1))   # (B,)\n\n# concatenate\nall_probs = torch.cat(all_probs).numpy()\nall_preds = torch.cat(all_preds).numpy()\nall_labels = torch.cat(all_labels).numpy()\n\n# accuracy\nval_acc = accuracy_score(all_labels, all_preds) * 100\nprint(f\"Validation Loss: {val_loss:.4f} | Validation Accuracy: {val_acc:.2f}%\")\n\n# ROC + AUC\nfpr, tpr, thresholds = roc_curve(all_labels, all_probs)\nroc_auc = auc(fpr, tpr)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:55:53.305377Z","iopub.execute_input":"2026-01-14T13:55:53.305659Z","iopub.status.idle":"2026-01-14T13:57:24.697498Z","shell.execute_reply.started":"2026-01-14T13:55:53.305632Z","shell.execute_reply":"2026-01-14T13:57:24.696566Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"idx_tpr = np.argmax(tpr)\nidx_fpr = np.argmin(fpr)\n\nidx = idx_tpr\nbest_thresh = thresholds[idx]\n\n# print(f\"TPR = {tpr[idx]:.2f} | FPR: {fpr[idx]:.2f}\")\n# print(f\"Best Threshold Value: {best_thresh:.2f}\")\n\nplt.figure(figsize=(6,6))\nplt.plot(fpr, tpr, label=f'ROC curve (AUC = {roc_auc:.3f})')\nplt.plot([0, 1], [0, 1], linestyle='--')\nplt.xlabel('False Positive Rate')\nplt.ylabel('True Positive Rate')\nplt.title('Receiver Operating Characteristic (ROC) Curve')\nplt.legend(loc='lower right')\nplt.grid(True)\nplt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-14T13:57:24.699384Z","iopub.execute_input":"2026-01-14T13:57:24.699698Z","iopub.status.idle":"2026-01-14T13:57:24.841246Z","shell.execute_reply.started":"2026-01-14T13:57:24.699661Z","shell.execute_reply":"2026-01-14T13:57:24.840606Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"!pip freeze > requirements.txt","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-02-11T11:58:53.876685Z","iopub.execute_input":"2026-02-11T11:58:53.876886Z","iopub.status.idle":"2026-02-11T11:58:57.348711Z","shell.execute_reply.started":"2026-02-11T11:58:53.876864Z","shell.execute_reply":"2026-02-11T11:58:57.347806Z"}},"outputs":[],"execution_count":null}]}