{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.14","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"nvidiaTeslaT4","dataSources":[{"sourceId":86142,"databundleVersionId":9786425,"sourceType":"competition"}],"dockerImageVersionId":30787,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":true}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"****Here I uploaded the code for the approach I tried but couldn't complete the model training due to certain time constraints of my own. The major issue I faced in the end was that this code was utilizing my CPU only and not my GPU at all, which resulted in exhaustion of the allocated memory space. ****\n\n**To solve this problem, I attempted to implement the prefetch approach but couldn't finish the training in due time. Below is the code for the approach I followed:** ","metadata":{}},{"cell_type":"markdown","source":"# Approach Overview\n\nStep 1: Load the dataset: Match the train images with their corresponding label images and JSON files.\n\nStep 2: Preprocess the data: Preprocess both the training and label images, normalize them, and prepare them for input into the model.\n\nStep 3: Create a data generator: Since the dataset is large, we will use a generator to load images batch by batch, avoiding memory overload.\n\nStep 4: Define the U-Net model: Build the model using EfficientNet as the backbone.\n\nStep 5: Compile the model: Define the loss functions, optimizer, and metrics.\n\nStep 6: Train the model: Train the model using the generator and validate it.\n\nStep 7: Evaluate the model: Check the performance on validation data.\n","metadata":{}},{"cell_type":"code","source":"train_dir = '/kaggle/input/iitg-ai-overnight-hackathon-2024/dataset/dataset/train'\nlabel_dir = '/kaggle/input/iitg-ai-overnight-hackathon-2024/dataset/dataset/labels'","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2024-10-10T12:55:06.032545Z","iopub.execute_input":"2024-10-10T12:55:06.032956Z","iopub.status.idle":"2024-10-10T12:55:06.037907Z","shell.execute_reply.started":"2024-10-10T12:55:06.032893Z","shell.execute_reply":"2024-10-10T12:55:06.036699Z"},"trusted":true},"execution_count":2,"outputs":[]},{"cell_type":"code","source":"import os\nimport cv2\nimport numpy as np\nfrom tensorflow.keras.utils import to_categorical\nfrom sklearn.model_selection import train_test_split","metadata":{"execution":{"iopub.status.busy":"2024-10-10T12:55:06.043441Z","iopub.execute_input":"2024-10-10T12:55:06.044258Z","iopub.status.idle":"2024-10-10T12:55:21.892339Z","shell.execute_reply.started":"2024-10-10T12:55:06.044202Z","shell.execute_reply":"2024-10-10T12:55:21.89143Z"},"trusted":true},"execution_count":3,"outputs":[]},{"cell_type":"code","source":"import tensorflow as tf\nfrom tensorflow.keras.applications import EfficientNetB0\nfrom tensorflow.keras.layers import Conv2D, UpSampling2D, concatenate, Input\nfrom tensorflow.keras.models import Model","metadata":{"execution":{"iopub.status.busy":"2024-10-10T12:55:21.894038Z","iopub.execute_input":"2024-10-10T12:55:21.894619Z","iopub.status.idle":"2024-10-10T12:55:21.907133Z","shell.execute_reply.started":"2024-10-10T12:55:21.894581Z","shell.execute_reply":"2024-10-10T12:55:21.906328Z"},"trusted":true},"execution_count":4,"outputs":[]},{"cell_type":"markdown","source":"# Defining Label Class","metadata":{}},{"cell_type":"code","source":"# Define the Label class\nclass Label:\n    def __init__(self, name, id, csId, csTrainId, level4id, level3Id, category, level2Id, level1Id, hasInstances, ignoreInEval, color):\n        self.name = name\n        self.id = id\n        self.csId = csId\n        self.csTrainId = csTrainId\n        self.level4id = level4id\n        self.level3Id = level3Id\n        self.category = category\n        self.level2Id = level2Id\n        self.level1Id = level1Id\n        self.hasInstances = hasInstances\n        self.ignoreInEval = ignoreInEval\n        self.color = color\n\n# Define your label definitions here\n# Define your label definitions here\nlabels = [\n    Label('road', 0, 7, 0, 0, 0, 'drivable', 0, 0, False, False, (128, 64, 128)),\n    Label('parking', 1, 9, 255, 1, 1, 'drivable', 1, 0, False, False, (250, 170, 160)),\n    Label('drivable fallback', 2, 255, 255, 2, 1, 'drivable', 1, 0, False, False, (81, 0, 81)),\n    Label('sidewalk', 3, 8, 1, 3, 2, 'non-drivable', 2, 1, False, False, (244, 35, 232)),\n    Label('rail track', 4, 10, 255, 3, 3, 'non-drivable', 3, 1, False, False, (230, 150, 140)),\n    Label('non-drivable fallback', 5, 255, 9, 4, 3, 'non-drivable', 3, 1, False, False, (152, 251, 152)),\n    Label('person', 6, 24, 11, 5, 4, 'living-thing', 4, 2, True, False, (220, 20, 60)),\n    Label('animal', 7, 255, 255, 6, 4, 'living-thing', 4, 2, True, True, (246, 198, 145)),\n    Label('rider', 8, 25, 12, 7, 5, 'living-thing', 5, 2, True, False, (255, 0, 0)),\n    Label('motorcycle', 9, 32, 17, 8, 6, '2-wheeler', 6, 3, True, False, (0, 0, 230)),\n    Label('bicycle', 10, 33, 18, 9, 7, '2-wheeler', 6, 3, True, False, (119, 11, 32)),\n    Label('autorickshaw', 11, 255, 255, 10, 8, 'autorickshaw', 7, 3, True, False, (255, 204, 54)),\n    Label('car', 12, 26, 13, 11, 9, 'car', 7, 3, True, False, (0, 0, 142)),\n    Label('truck', 13, 27, 14, 12, 10, 'large-vehicle', 8, 3, True, False, (0, 0, 70)),\n    Label('bus', 14, 28, 15, 13, 11, 'large-vehicle', 8, 3, True, False, (0, 60, 100)),\n    Label('caravan', 15, 29, 255, 14, 12, 'large-vehicle', 8, 3, True, True, (0, 0, 90)),\n    Label('trailer', 16, 30, 255, 15, 12, 'large-vehicle', 8, 3, True, True, (0, 0, 110)),\n    Label('train', 17, 31, 16, 15, 12, 'large-vehicle', 8, 3, True, True, (0, 80, 100)),\n    Label('vehicle fallback', 18, 355, 255, 15, 12, 'large-vehicle', 8, 3, True, False, (136, 143, 153)),\n    Label('curb', 19, 255, 255, 16, 13, 'barrier', 9, 4, False, False, (220, 190, 40)),\n    Label('wall', 20, 12, 3, 17, 14, 'barrier', 9, 4, False, False, (102, 102, 156)),\n    Label('fence', 21, 13, 4, 18, 15, 'barrier', 10, 4, False, False, (190, 153, 153)),\n    Label('guard rail', 22, 14, 255, 19, 16, 'barrier', 10, 4, False, False, (180, 165, 180)),\n    Label('billboard', 23, 255, 255, 20, 17, 'structures', 11, 4, False, False, (174, 64, 67)),\n    Label('traffic sign', 24, 20, 7, 21, 18, 'structures', 11, 4, False, False, (220, 220, 0)),\n    Label('traffic light', 25, 19, 6, 22, 19, 'structures', 11, 4, False, False, (250, 170, 30)),\n    Label('pole', 26, 17, 5, 23, 20, 'structures', 12, 4, False, False, (153, 153, 153)),\n    Label('polegroup', 27, 18, 255, 23, 20, 'structures', 12, 4, False, False, (153, 153, 153)),\n    Label('obs-str-bar-fallback', 28, 255, 255, 24, 21, 'structures', 12, 4, False, False, (169, 187, 214)),\n    Label('building', 29, 11, 2, 25, 22, 'construction', 13, 5, False, False, (70, 70, 70)),\n    Label('bridge', 30, 15, 255, 26, 23, 'construction', 13, 5, False, False, (150, 100, 100)),\n    Label('tunnel', 31, 16, 255, 26, 23, 'construction', 13, 5, False, False, (150, 120, 90)),\n    Label('vegetation', 32, 21, 8, 27, 24, 'vegetation', 14, 5, False, False, (107, 142, 35)),\n    Label('sky', 33, 23, 10, 28, 25, 'sky', 15, 6, False, False, (70, 130, 180)),\n    Label('fallback background', 34, 255, 255, 29, 25, 'object fallback', 15, 6, False, False, (169, 187, 214)),\n    Label('unlabeled', 35, 0, 255, 255, 255, 'void', 255, 255, False, True, (0, 0, 0)),\n    Label('ego vehicle', 36, 1, 255, 255, 255, 'void', 255, 255, False, True, (0, 0, 0)),\n    Label('rectification border', 37, 2, 255, 255, 255, 'void', 255, 255, False, True, (0, 0, 0)),\n    Label('out of roi', 38, 3, 255, 255, 255, 'void', 255, 255, False, True, (0, 0, 0)),\n    Label('license plate', 39, 255, 255, 255, 255, 'vehicle', 255, 255, False, True, (0, 0, 142)),\n]\n","metadata":{"execution":{"iopub.status.busy":"2024-10-10T12:55:21.908616Z","iopub.execute_input":"2024-10-10T12:55:21.908925Z","iopub.status.idle":"2024-10-10T12:55:21.940112Z","shell.execute_reply.started":"2024-10-10T12:55:21.908877Z","shell.execute_reply":"2024-10-10T12:55:21.939226Z"},"trusted":true},"execution_count":5,"outputs":[]},{"cell_type":"markdown","source":"# Colour to class mapping","metadata":{}},{"cell_type":"code","source":"def get_color_to_class_mapping():\n    \"\"\"\n    Map the RGB color values from the segmented images to their corresponding class IDs.\n    \"\"\"\n    color_to_class_map = {tuple(label.color): label.id for label in labels}\n    return color_to_class_map\n\n# Create the mapping\ncolor_to_class_map = get_color_to_class_mapping()\n\n# Print the mapping for verification\nfor color, class_id in color_to_class_map.items():\n#     print(f\"Color {color} corresponds to Class ID {class_id}\")\n    print(f\"{color}: {class_id},\")","metadata":{"execution":{"iopub.status.busy":"2024-10-10T12:55:21.942513Z","iopub.execute_input":"2024-10-10T12:55:21.942818Z","iopub.status.idle":"2024-10-10T12:55:21.95633Z","shell.execute_reply.started":"2024-10-10T12:55:21.942786Z","shell.execute_reply":"2024-10-10T12:55:21.955363Z"},"trusted":true},"execution_count":6,"outputs":[{"name":"stdout","text":"(128, 64, 128): 0,\n(250, 170, 160): 1,\n(81, 0, 81): 2,\n(244, 35, 232): 3,\n(230, 150, 140): 4,\n(152, 251, 152): 5,\n(220, 20, 60): 6,\n(246, 198, 145): 7,\n(255, 0, 0): 8,\n(0, 0, 230): 9,\n(119, 11, 32): 10,\n(255, 204, 54): 11,\n(0, 0, 142): 39,\n(0, 0, 70): 13,\n(0, 60, 100): 14,\n(0, 0, 90): 15,\n(0, 0, 110): 16,\n(0, 80, 100): 17,\n(136, 143, 153): 18,\n(220, 190, 40): 19,\n(102, 102, 156): 20,\n(190, 153, 153): 21,\n(180, 165, 180): 22,\n(174, 64, 67): 23,\n(220, 220, 0): 24,\n(250, 170, 30): 25,\n(153, 153, 153): 27,\n(169, 187, 214): 34,\n(70, 70, 70): 29,\n(150, 100, 100): 30,\n(150, 120, 90): 31,\n(107, 142, 35): 32,\n(70, 130, 180): 33,\n(0, 0, 0): 38,\n","output_type":"stream"}]},{"cell_type":"markdown","source":"clss id 35, 36, 37 , 38 here has samee color value wich we are classifying as \n\ncls id 12 and 39 have same colour code so we better classify it as 12\n\nfor id 24 and 38 ,m we choose the class 24\n\nout of 26 and 27 we choose 26","metadata":{}},{"cell_type":"code","source":"COLOR_TO_CLASS = {\n    \n    (128, 64, 128): 0,\n    (250, 170, 160): 1,\n    (81, 0, 81): 2,\n    (244, 35, 232): 3,\n    (230, 150, 140): 4,\n    (152, 251, 152): 5,\n    (220, 20, 60): 6,\n    (246, 198, 145): 7,\n(255, 0, 0): 8,\n(0, 0, 230): 9,\n(119, 11, 32): 10,\n(255, 204, 54): 11,\n(0, 0, 142): 12,  # same as 39\n(0, 0, 70): 13,\n(0, 60, 100): 14,\n(0, 0, 90): 15,\n(0, 0, 110): 16,\n(0, 80, 100): 17,\n(136, 143, 153): 18,\n(220, 190, 40): 19,\n(102, 102, 156): 20,\n(190, 153, 153): 21,\n(180, 165, 180): 22,\n(174, 64, 67): 23,\n(220, 220, 0): 24,\n(250, 170, 30): 25,\n(153, 153, 153): 26,  # same as 27\n(169, 187, 214): 34,\n(70, 70, 70): 29,\n(150, 100, 100): 30,\n(150, 120, 90): 31,\n(107, 142, 35): 32,\n(70, 130, 180): 33,  \n(0, 0, 0): 38, # for 34-38\n}","metadata":{"execution":{"iopub.status.busy":"2024-10-10T12:55:21.957451Z","iopub.execute_input":"2024-10-10T12:55:21.957777Z","iopub.status.idle":"2024-10-10T12:55:21.969043Z","shell.execute_reply.started":"2024-10-10T12:55:21.957744Z","shell.execute_reply":"2024-10-10T12:55:21.968161Z"},"trusted":true},"execution_count":7,"outputs":[]},{"cell_type":"code","source":"print(len(COLOR_TO_CLASS))","metadata":{"execution":{"iopub.status.busy":"2024-10-10T12:55:21.970284Z","iopub.execute_input":"2024-10-10T12:55:21.970967Z","iopub.status.idle":"2024-10-10T12:55:21.983337Z","shell.execute_reply.started":"2024-10-10T12:55:21.970905Z","shell.execute_reply":"2024-10-10T12:55:21.982331Z"},"trusted":true},"execution_count":8,"outputs":[{"name":"stdout","text":"34\n","output_type":"stream"}]},{"cell_type":"markdown","source":"# Pairing Train and Label Images","metadata":{}},{"cell_type":"code","source":"def get_image_pairs(train_dir, label_dir):\n    \"\"\"\n    Pair the train images with their corresponding label images \n    (segmented images) in the label directory.\n    \n    train_dir: path to the directory containing the RGB images.\n    label_dir: path to the directory containing the segmented images.\n    \n    Returns:\n    - train_files: list of file paths to RGB images.\n    - label_segmented_files: list of file paths to segmented images.\n    \"\"\"\n    train_files = []\n    label_segmented_files = []\n    \n    for root, _, files in os.walk(train_dir):\n        for file in files:\n            if file.endswith('.jpg'):  # Assuming RGB images are in .jpg format\n                common_part = file.split('_')[0]  # Extract common part from filename\n                \n                # Train image path\n                train_image_path = os.path.join(root, file)\n                \n                # Corresponding segmented image in label directory\n                segmented_image_path = os.path.join(label_dir, root.split('/')[-1], f\"{common_part}_gtFine_labelColors.png\")\n                \n                # Append only if the corresponding files exist\n                if os.path.exists(segmented_image_path):\n                    train_files.append(train_image_path)\n                    label_segmented_files.append(segmented_image_path)\n                else:\n                    print(f\"Warning: Missing segmented image for {common_part}\")\n    \n    return train_files, label_segmented_files\n\n\ntrain_files, label_segmented_files = get_image_pairs(train_dir, label_dir)\nprint(f\"Found {len(train_files)} training pairs.\")\nprint(f\"Found {len(label_segmented_files)} training pairs.\")","metadata":{"execution":{"iopub.status.busy":"2024-10-10T12:55:21.984505Z","iopub.execute_input":"2024-10-10T12:55:21.985152Z","iopub.status.idle":"2024-10-10T12:55:39.082334Z","shell.execute_reply.started":"2024-10-10T12:55:21.985106Z","shell.execute_reply":"2024-10-10T12:55:39.081064Z"},"trusted":true},"execution_count":9,"outputs":[{"name":"stdout","text":"Found 7034 training pairs.\nFound 7034 training pairs.\n","output_type":"stream"}]},{"cell_type":"markdown","source":"# Splitting the Dataset into Training and Validation","metadata":{}},{"cell_type":"code","source":"from sklearn.model_selection import train_test_split\n\n# Split the dataset into training and validation sets (80% train, 20% validation)\ntrain_files, val_files, label_train_files, label_val_files = train_test_split(\n    train_files, label_segmented_files, test_size=0.2, random_state=42\n)\n\nprint(f\"Training set size: {len(train_files)}\")\nprint(f\"Validation set size: {len(val_files)}\")","metadata":{"execution":{"iopub.status.busy":"2024-10-10T12:55:39.083702Z","iopub.execute_input":"2024-10-10T12:55:39.084165Z","iopub.status.idle":"2024-10-10T12:55:39.101024Z","shell.execute_reply.started":"2024-10-10T12:55:39.084115Z","shell.execute_reply":"2024-10-10T12:55:39.099727Z"},"trusted":true},"execution_count":10,"outputs":[{"name":"stdout","text":"Training set size: 5627\nValidation set size: 1407\n","output_type":"stream"}]},{"cell_type":"markdown","source":"#  Preprocessing the RGB and Segmented Images","metadata":{}},{"cell_type":"code","source":"def preprocess_rgb_image(image_path, target_size=(1080, 1920)):\n    \"\"\"\n    Preprocess an RGB image: resize and normalize.\n    \n    image_path: Path to the image file.\n    target_size: Tuple (height, width).\n    \n    Returns:\n    - preprocessed RGB image.\n    \"\"\"\n    image = cv2.imread(image_path)\n    image = cv2.resize(image, (target_size[1], target_size[0]))  # Resize to target size\n    image = image / 255.0  # Normalize the pixel values to [0, 1]\n    return image\n\ndef preprocess_segmented_image(image_path, target_size=(1080, 1920)):\n    \"\"\"\n    Preprocess a segmented image: resize and map colors to class IDs.\n    \n    image_path: Path to the segmented image file.\n    target_size: Tuple (height, width).\n    \n    Returns:\n    - Class-mapped image.\n    \"\"\"\n    image = cv2.imread(image_path)\n    image = cv2.resize(image, (target_size[1], target_size[0]))  # Resize\n    class_mapped_image = np.zeros((target_size[0], target_size[1]), dtype=np.uint8)  # Initialize empty class-mapped image\n\n    # Loop through each pixel and map the RGB color to a class ID\n    for color, class_id in COLOR_TO_CLASS.items():\n        mask = np.all(image == color, axis=-1)  # Create a mask for matching pixels\n        class_mapped_image[mask] = class_id\n\n    return class_mapped_image","metadata":{"execution":{"iopub.status.busy":"2024-10-10T12:55:39.102499Z","iopub.execute_input":"2024-10-10T12:55:39.10296Z","iopub.status.idle":"2024-10-10T12:55:39.114776Z","shell.execute_reply.started":"2024-10-10T12:55:39.102875Z","shell.execute_reply":"2024-10-10T12:55:39.113545Z"},"trusted":true},"execution_count":11,"outputs":[]},{"cell_type":"code","source":"abb = preprocess_segmented_image(label_segmented_files[0], target_size=(1080, 1920))\nabb.shape","metadata":{"execution":{"iopub.status.busy":"2024-10-10T12:55:39.119658Z","iopub.execute_input":"2024-10-10T12:55:39.120081Z","iopub.status.idle":"2024-10-10T12:55:41.891759Z","shell.execute_reply.started":"2024-10-10T12:55:39.120024Z","shell.execute_reply":"2024-10-10T12:55:41.89058Z"},"trusted":true},"execution_count":12,"outputs":[{"execution_count":12,"output_type":"execute_result","data":{"text/plain":"(1080, 1920)"},"metadata":{}}]},{"cell_type":"markdown","source":"# Data Generator for Efficient Data Loading","metadata":{}},{"cell_type":"code","source":"# from tensorflow.keras.utils import to_categorical\n\n# def data_generator(train_files, label_files, batch_size=8, img_size=(1080, 1920), num_classes=40):\n#     \"\"\"\n#     Generator function to yield batches of data.\n    \n#     train_files: List of paths to RGB images.\n#     label_files: List of paths to segmented images.\n#     batch_size: Batch size for the generator.\n#     img_size: Size to which images are resized.\n#     num_classes: Number of classes for segmentation.\n    \n#     Yields:\n#     - Batch of preprocessed RGB images.\n#     - Batch of one-hot encoded class-mapped labels.\n#     \"\"\"\n#     num_samples = len(train_files)\n    \n#     while True:\n#         for offset in range(0, num_samples, batch_size):\n#             batch_train_files = train_files[offset:offset + batch_size]\n#             batch_label_files = label_files[offset:offset + batch_size]\n            \n#             train_images = []\n#             label_images = []\n            \n#             for i in range(len(batch_train_files)):\n#                 train_img = preprocess_rgb_image(batch_train_files[i], img_size)\n#                 label_img = preprocess_segmented_image(batch_label_files[i], img_size)\n                \n#                 train_images.append(train_img)\n#                 label_images.append(to_categorical(label_img, num_classes=num_classes))  # One-hot encoding for classes\n            \n#             yield np.array(train_images), np.array(label_images)\n","metadata":{"execution":{"iopub.status.busy":"2024-10-10T12:55:41.893194Z","iopub.execute_input":"2024-10-10T12:55:41.89354Z","iopub.status.idle":"2024-10-10T12:55:41.900028Z","shell.execute_reply.started":"2024-10-10T12:55:41.893506Z","shell.execute_reply":"2024-10-10T12:55:41.898945Z"},"trusted":true},"execution_count":13,"outputs":[]},{"cell_type":"code","source":"reduced_size=(540, 960)","metadata":{"execution":{"iopub.status.busy":"2024-10-10T12:55:41.90155Z","iopub.execute_input":"2024-10-10T12:55:41.902034Z","iopub.status.idle":"2024-10-10T12:55:41.934057Z","shell.execute_reply.started":"2024-10-10T12:55:41.901981Z","shell.execute_reply":"2024-10-10T12:55:41.932928Z"},"trusted":true},"execution_count":14,"outputs":[]},{"cell_type":"code","source":"def data_generator(train_files, label_files, batch_size=8, img_size=(1080, 1920), num_classes=40):\n    \"\"\"\n    Generator function to yield batches of data.\n    \n    train_files: List of paths to RGB images.\n    label_files: List of paths to segmented images.\n    batch_size: Batch size for the generator.\n    img_size: Size to which images are resized.\n    num_classes: Number of classes for segmentation (not needed for output shape now).\n    \n    Yields:\n    - Batch of preprocessed RGB images of shape (1080, 1920, 3).\n    - Batch of class-mapped images (segmentation masks) of shape (1080, 1920, 1).\n    \"\"\"\n    num_samples = len(train_files)\n    \n    while True:\n        for offset in range(0, num_samples, batch_size):\n            batch_train_files = train_files[offset:offset + batch_size]\n            batch_label_files = label_files[offset:offset + batch_size]\n            \n            train_images = []\n            label_images = []\n            \n            for i in range(len(batch_train_files)):\n                # Preprocess the RGB image (1080, 1920, 3)\n                train_img = preprocess_rgb_image(batch_train_files[i], img_size)  # (H, W, 3)\n                \n                # Preprocess the segmented image and keep it as class-labeled single channel (1080, 1920, 1)\n                label_img = preprocess_segmented_image(batch_label_files[i], img_size)  # (H, W) with class IDs\n                \n                # Expand dimensions to make the label image (1080, 1920, 1)\n                label_img = np.expand_dims(label_img, axis=-1)  # Convert (H, W) -> (H, W, 1)\n                \n                train_images.append(train_img)\n                label_images.append(label_img)\n            \n            # Convert list of images into numpy arrays and yield the batch\n            yield np.array(train_images), np.array(label_images)","metadata":{"execution":{"iopub.status.busy":"2024-10-10T12:55:41.935362Z","iopub.execute_input":"2024-10-10T12:55:41.93575Z","iopub.status.idle":"2024-10-10T12:55:41.946615Z","shell.execute_reply.started":"2024-10-10T12:55:41.935713Z","shell.execute_reply":"2024-10-10T12:55:41.944638Z"},"trusted":true},"execution_count":15,"outputs":[]},{"cell_type":"markdown","source":"#  Building the U-Net Model","metadata":{}},{"cell_type":"code","source":"import tensorflow as tf\nfrom tensorflow.keras import layers, models\nfrom tensorflow.keras.applications import EfficientNetB0","metadata":{"execution":{"iopub.status.busy":"2024-10-10T12:55:41.947966Z","iopub.execute_input":"2024-10-10T12:55:41.948271Z","iopub.status.idle":"2024-10-10T12:55:41.966126Z","shell.execute_reply.started":"2024-10-10T12:55:41.948239Z","shell.execute_reply":"2024-10-10T12:55:41.965044Z"},"trusted":true},"execution_count":16,"outputs":[]},{"cell_type":"code","source":"def build_fcn(input_shape):\n    inputs = layers.Input(shape=input_shape)\n    \n    # EfficientNet as the backbone (without the fully connected top layer)\n    backbone = EfficientNetB0(include_top=False, weights='imagenet', input_tensor=inputs)\n    \n    # Extracting feature maps from different levels for skip connections\n    skip1 = backbone.get_layer('block2a_expand_activation').output  # First skip connection\n    skip2 = backbone.get_layer('block3a_expand_activation').output  # Second skip connection\n    skip3 = backbone.get_layer('block4a_expand_activation').output  # Third skip connection\n    skip4 = backbone.get_layer('block6a_expand_activation').output  # Fourth skip connection\n    \n    # Bottom layer of EfficientNet (lowest resolution, deepest features)\n    bottom = backbone.get_layer('top_activation').output  # Bottom layer\n    \n    print(f\"Bottom layer shape: {bottom.shape}\")\n    print(f\"Skip connection shapes: skip1={skip1.shape}, skip2={skip2.shape}, skip3={skip3.shape}, skip4={skip4.shape}\")\n\n    # Decoder path (Upsampling + Convolutions)\n    up4 = layers.UpSampling2D(size=(2, 2))(bottom)\n    up4 = layers.Conv2D(512, (3, 3), activation='relu', padding='same')(up4)\n    print(f\"up4 shape after upsampling and conv: {up4.shape}\")\n    \n    up4 = layers.concatenate([up4, skip4])  # Concatenate with the corresponding skip connection\n    print(f\"up4 shape after concatenation: {up4.shape}\")\n    \n    up3 = layers.UpSampling2D(size=(2, 2))(up4)\n    up3 = layers.Conv2D(256, (3, 3), activation='relu', padding='same')(up3)\n    print(f\"up3 shape after upsampling and conv: {up3.shape}\")\n    \n    up3 = layers.Cropping2D(cropping=((1, 0), (0, 0)))(up3)  # Crop to match skip3 shape\n    up3 = layers.concatenate([up3, skip3])  # Concatenate with the corresponding skip connection\n    print(f\"up3 shape after concatenation: {up3.shape}\")\n    \n    up2 = layers.UpSampling2D(size=(2, 2))(up3)\n    up2 = layers.Conv2D(128, (3, 3), activation='relu', padding='same')(up2)\n    print(f\"up2 shape after upsampling and conv: {up2.shape}\")\n    \n    up2 = layers.concatenate([up2, skip2])  # Concatenate with the corresponding skip connection\n    print(f\"up2 shape after concatenation: {up2.shape}\")\n    \n    up1 = layers.UpSampling2D(size=(2, 2))(up2)\n    up1 = layers.Conv2D(64, (3, 3), activation='relu', padding='same')(up1)\n    print(f\"up1 shape after upsampling and conv: {up1.shape}\")\n    \n    up1 = layers.concatenate([up1, skip1])  # Concatenate with the corresponding skip connection\n    print(f\"up1 shape after concatenation: {up1.shape}\")\n    \n    # Final upsampling to match input size\n    up0 = layers.UpSampling2D(size=(2, 2))(up1)\n    up0 = layers.Conv2D(32, (3, 3), activation='relu', padding='same')(up0)\n    print(f\"up0 shape after final upsampling and conv: {up0.shape}\")\n\n    # Output layer: 1x1 Convolution to predict the class ID for each pixel\n    outputs = layers.Conv2D(1, (1, 1), activation='linear', padding='same')(up0)  # Change num_classes to 1\n    print(f\"{outputs.shape}\")\n    \n    model = models.Model(inputs=inputs, outputs=outputs)\n\n    return model\n\n# Input shape and build the model\ninput_shape = (1080, 1920, 3)  # RGB image input shape\nmodel = build_fcn(input_shape)","metadata":{"execution":{"iopub.status.busy":"2024-10-10T12:55:41.967473Z","iopub.execute_input":"2024-10-10T12:55:41.968257Z","iopub.status.idle":"2024-10-10T12:55:44.92885Z","shell.execute_reply.started":"2024-10-10T12:55:41.968196Z","shell.execute_reply":"2024-10-10T12:55:44.927826Z"},"trusted":true},"execution_count":17,"outputs":[{"name":"stdout","text":"Downloading data from https://storage.googleapis.com/keras-applications/efficientnetb0_notop.h5\n\u001b[1m16705208/16705208\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 0us/step\nBottom layer shape: (None, 34, 60, 1280)\nSkip connection shapes: skip1=(None, 540, 960, 96), skip2=(None, 270, 480, 144), skip3=(None, 135, 240, 240), skip4=(None, 68, 120, 672)\nup4 shape after upsampling and conv: (None, 68, 120, 512)\nup4 shape after concatenation: (None, 68, 120, 1184)\nup3 shape after upsampling and conv: (None, 136, 240, 256)\nup3 shape after concatenation: (None, 135, 240, 496)\nup2 shape after upsampling and conv: (None, 270, 480, 128)\nup2 shape after concatenation: (None, 270, 480, 272)\nup1 shape after upsampling and conv: (None, 540, 960, 64)\nup1 shape after concatenation: (None, 540, 960, 160)\nup0 shape after final upsampling and conv: (None, 1080, 1920, 32)\n(None, 1080, 1920, 1)\n","output_type":"stream"}]},{"cell_type":"code","source":"# Summary of the model\nmodel.summary()","metadata":{"execution":{"iopub.status.busy":"2024-10-10T12:55:44.930394Z","iopub.execute_input":"2024-10-10T12:55:44.930829Z","iopub.status.idle":"2024-10-10T12:55:45.295197Z","shell.execute_reply.started":"2024-10-10T12:55:44.930781Z","shell.execute_reply":"2024-10-10T12:55:45.294247Z"},"trusted":true},"execution_count":18,"outputs":[{"output_type":"display_data","data":{"text/plain":"\u001b[1mModel: \"functional_1\"\u001b[0m\n","text/html":"<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"functional_1\"</span>\n</pre>\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"┏━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┓\n┃\u001b[1m \u001b[0m\u001b[1mLayer (type)       \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape     \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m   Param #\u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mConnected to     \u001b[0m\u001b[1m \u001b[0m┃\n┡━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━┩\n│ input_layer         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1080\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ -                 │\n│ (\u001b[38;5;33mInputLayer\u001b[0m)        │ \u001b[38;5;34m1920\u001b[0m, \u001b[38;5;34m3\u001b[0m)          │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ rescaling           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1080\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ input_layer[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n│ (\u001b[38;5;33mRescaling\u001b[0m)         │ \u001b[38;5;34m1920\u001b[0m, \u001b[38;5;34m3\u001b[0m)          │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ normalization       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1080\u001b[0m,      │          \u001b[38;5;34m7\u001b[0m │ rescaling[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n│ (\u001b[38;5;33mNormalization\u001b[0m)     │ \u001b[38;5;34m1920\u001b[0m, \u001b[38;5;34m3\u001b[0m)          │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ rescaling_1         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1080\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ normalization[\u001b[38;5;34m0\u001b[0m]… │\n│ (\u001b[38;5;33mRescaling\u001b[0m)         │ \u001b[38;5;34m1920\u001b[0m, \u001b[38;5;34m3\u001b[0m)          │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ stem_conv_pad       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1081\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ rescaling_1[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n│ (\u001b[38;5;33mZeroPadding2D\u001b[0m)     │ \u001b[38;5;34m1921\u001b[0m, \u001b[38;5;34m3\u001b[0m)          │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ stem_conv (\u001b[38;5;33mConv2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m540\u001b[0m, \u001b[38;5;34m960\u001b[0m,  │        \u001b[38;5;34m864\u001b[0m │ stem_conv_pad[\u001b[38;5;34m0\u001b[0m]… │\n│                     │ \u001b[38;5;34m32\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ stem_bn             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m540\u001b[0m, \u001b[38;5;34m960\u001b[0m,  │        \u001b[38;5;34m128\u001b[0m │ stem_conv[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]   │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m32\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ stem_activation     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m540\u001b[0m, \u001b[38;5;34m960\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ stem_bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]     │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m32\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block1a_dwconv      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m540\u001b[0m, \u001b[38;5;34m960\u001b[0m,  │        \u001b[38;5;34m288\u001b[0m │ stem_activation[\u001b[38;5;34m…\u001b[0m │\n│ (\u001b[38;5;33mDepthwiseConv2D\u001b[0m)   │ \u001b[38;5;34m32\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block1a_bn          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m540\u001b[0m, \u001b[38;5;34m960\u001b[0m,  │        \u001b[38;5;34m128\u001b[0m │ block1a_dwconv[\u001b[38;5;34m0\u001b[0m… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m32\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block1a_activation  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m540\u001b[0m, \u001b[38;5;34m960\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ block1a_bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m32\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block1a_se_squeeze  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m)        │          \u001b[38;5;34m0\u001b[0m │ block1a_activati… │\n│ (\u001b[38;5;33mGlobalAveragePool…\u001b[0m │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block1a_se_reshape  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m32\u001b[0m)  │          \u001b[38;5;34m0\u001b[0m │ block1a_se_squee… │\n│ (\u001b[38;5;33mReshape\u001b[0m)           │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block1a_se_reduce   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m8\u001b[0m)   │        \u001b[38;5;34m264\u001b[0m │ block1a_se_resha… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block1a_se_expand   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m32\u001b[0m)  │        \u001b[38;5;34m288\u001b[0m │ block1a_se_reduc… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block1a_se_excite   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m540\u001b[0m, \u001b[38;5;34m960\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ block1a_activati… │\n│ (\u001b[38;5;33mMultiply\u001b[0m)          │ \u001b[38;5;34m32\u001b[0m)               │            │ block1a_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block1a_project_co… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m540\u001b[0m, \u001b[38;5;34m960\u001b[0m,  │        \u001b[38;5;34m512\u001b[0m │ block1a_se_excit… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m16\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block1a_project_bn  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m540\u001b[0m, \u001b[38;5;34m960\u001b[0m,  │         \u001b[38;5;34m64\u001b[0m │ block1a_project_… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m16\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_expand_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m540\u001b[0m, \u001b[38;5;34m960\u001b[0m,  │      \u001b[38;5;34m1,536\u001b[0m │ block1a_project_… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m96\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_expand_bn   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m540\u001b[0m, \u001b[38;5;34m960\u001b[0m,  │        \u001b[38;5;34m384\u001b[0m │ block2a_expand_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m96\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_expand_act… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m540\u001b[0m, \u001b[38;5;34m960\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ block2a_expand_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m96\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_dwconv_pad  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m541\u001b[0m, \u001b[38;5;34m961\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ block2a_expand_a… │\n│ (\u001b[38;5;33mZeroPadding2D\u001b[0m)     │ \u001b[38;5;34m96\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_dwconv      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m270\u001b[0m, \u001b[38;5;34m480\u001b[0m,  │        \u001b[38;5;34m864\u001b[0m │ block2a_dwconv_p… │\n│ (\u001b[38;5;33mDepthwiseConv2D\u001b[0m)   │ \u001b[38;5;34m96\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_bn          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m270\u001b[0m, \u001b[38;5;34m480\u001b[0m,  │        \u001b[38;5;34m384\u001b[0m │ block2a_dwconv[\u001b[38;5;34m0\u001b[0m… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m96\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_activation  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m270\u001b[0m, \u001b[38;5;34m480\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ block2a_bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m96\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_se_squeeze  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m96\u001b[0m)        │          \u001b[38;5;34m0\u001b[0m │ block2a_activati… │\n│ (\u001b[38;5;33mGlobalAveragePool…\u001b[0m │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_se_reshape  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m96\u001b[0m)  │          \u001b[38;5;34m0\u001b[0m │ block2a_se_squee… │\n│ (\u001b[38;5;33mReshape\u001b[0m)           │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_se_reduce   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m4\u001b[0m)   │        \u001b[38;5;34m388\u001b[0m │ block2a_se_resha… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_se_expand   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m96\u001b[0m)  │        \u001b[38;5;34m480\u001b[0m │ block2a_se_reduc… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_se_excite   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m270\u001b[0m, \u001b[38;5;34m480\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ block2a_activati… │\n│ (\u001b[38;5;33mMultiply\u001b[0m)          │ \u001b[38;5;34m96\u001b[0m)               │            │ block2a_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_project_co… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m270\u001b[0m, \u001b[38;5;34m480\u001b[0m,  │      \u001b[38;5;34m2,304\u001b[0m │ block2a_se_excit… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m24\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_project_bn  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m270\u001b[0m, \u001b[38;5;34m480\u001b[0m,  │         \u001b[38;5;34m96\u001b[0m │ block2a_project_… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m24\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_expand_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m270\u001b[0m, \u001b[38;5;34m480\u001b[0m,  │      \u001b[38;5;34m3,456\u001b[0m │ block2a_project_… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m144\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_expand_bn   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m270\u001b[0m, \u001b[38;5;34m480\u001b[0m,  │        \u001b[38;5;34m576\u001b[0m │ block2b_expand_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m144\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_expand_act… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m270\u001b[0m, \u001b[38;5;34m480\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ block2b_expand_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m144\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_dwconv      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m270\u001b[0m, \u001b[38;5;34m480\u001b[0m,  │      \u001b[38;5;34m1,296\u001b[0m │ block2b_expand_a… │\n│ (\u001b[38;5;33mDepthwiseConv2D\u001b[0m)   │ \u001b[38;5;34m144\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_bn          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m270\u001b[0m, \u001b[38;5;34m480\u001b[0m,  │        \u001b[38;5;34m576\u001b[0m │ block2b_dwconv[\u001b[38;5;34m0\u001b[0m… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m144\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_activation  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m270\u001b[0m, \u001b[38;5;34m480\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ block2b_bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m144\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_se_squeeze  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m144\u001b[0m)       │          \u001b[38;5;34m0\u001b[0m │ block2b_activati… │\n│ (\u001b[38;5;33mGlobalAveragePool…\u001b[0m │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_se_reshape  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m144\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ block2b_se_squee… │\n│ (\u001b[38;5;33mReshape\u001b[0m)           │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_se_reduce   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m6\u001b[0m)   │        \u001b[38;5;34m870\u001b[0m │ block2b_se_resha… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_se_expand   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m144\u001b[0m) │      \u001b[38;5;34m1,008\u001b[0m │ block2b_se_reduc… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_se_excite   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m270\u001b[0m, \u001b[38;5;34m480\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ block2b_activati… │\n│ (\u001b[38;5;33mMultiply\u001b[0m)          │ \u001b[38;5;34m144\u001b[0m)              │            │ block2b_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_project_co… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m270\u001b[0m, \u001b[38;5;34m480\u001b[0m,  │      \u001b[38;5;34m3,456\u001b[0m │ block2b_se_excit… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m24\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_project_bn  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m270\u001b[0m, \u001b[38;5;34m480\u001b[0m,  │         \u001b[38;5;34m96\u001b[0m │ block2b_project_… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m24\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_drop        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m270\u001b[0m, \u001b[38;5;34m480\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ block2b_project_… │\n│ (\u001b[38;5;33mDropout\u001b[0m)           │ \u001b[38;5;34m24\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_add (\u001b[38;5;33mAdd\u001b[0m)   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m270\u001b[0m, \u001b[38;5;34m480\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ block2b_drop[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n│                     │ \u001b[38;5;34m24\u001b[0m)               │            │ block2a_project_… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_expand_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m270\u001b[0m, \u001b[38;5;34m480\u001b[0m,  │      \u001b[38;5;34m3,456\u001b[0m │ block2b_add[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m144\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_expand_bn   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m270\u001b[0m, \u001b[38;5;34m480\u001b[0m,  │        \u001b[38;5;34m576\u001b[0m │ block3a_expand_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m144\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_expand_act… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m270\u001b[0m, \u001b[38;5;34m480\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ block3a_expand_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m144\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_dwconv_pad  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m273\u001b[0m, \u001b[38;5;34m483\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ block3a_expand_a… │\n│ (\u001b[38;5;33mZeroPadding2D\u001b[0m)     │ \u001b[38;5;34m144\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_dwconv      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m135\u001b[0m, \u001b[38;5;34m240\u001b[0m,  │      \u001b[38;5;34m3,600\u001b[0m │ block3a_dwconv_p… │\n│ (\u001b[38;5;33mDepthwiseConv2D\u001b[0m)   │ \u001b[38;5;34m144\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_bn          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m135\u001b[0m, \u001b[38;5;34m240\u001b[0m,  │        \u001b[38;5;34m576\u001b[0m │ block3a_dwconv[\u001b[38;5;34m0\u001b[0m… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m144\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_activation  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m135\u001b[0m, \u001b[38;5;34m240\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ block3a_bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m144\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_se_squeeze  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m144\u001b[0m)       │          \u001b[38;5;34m0\u001b[0m │ block3a_activati… │\n│ (\u001b[38;5;33mGlobalAveragePool…\u001b[0m │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_se_reshape  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m144\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ block3a_se_squee… │\n│ (\u001b[38;5;33mReshape\u001b[0m)           │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_se_reduce   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m6\u001b[0m)   │        \u001b[38;5;34m870\u001b[0m │ block3a_se_resha… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_se_expand   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m144\u001b[0m) │      \u001b[38;5;34m1,008\u001b[0m │ block3a_se_reduc… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_se_excite   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m135\u001b[0m, \u001b[38;5;34m240\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ block3a_activati… │\n│ (\u001b[38;5;33mMultiply\u001b[0m)          │ \u001b[38;5;34m144\u001b[0m)              │            │ block3a_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_project_co… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m135\u001b[0m, \u001b[38;5;34m240\u001b[0m,  │      \u001b[38;5;34m5,760\u001b[0m │ block3a_se_excit… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m40\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_project_bn  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m135\u001b[0m, \u001b[38;5;34m240\u001b[0m,  │        \u001b[38;5;34m160\u001b[0m │ block3a_project_… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m40\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_expand_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m135\u001b[0m, \u001b[38;5;34m240\u001b[0m,  │      \u001b[38;5;34m9,600\u001b[0m │ block3a_project_… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m240\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_expand_bn   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m135\u001b[0m, \u001b[38;5;34m240\u001b[0m,  │        \u001b[38;5;34m960\u001b[0m │ block3b_expand_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m240\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_expand_act… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m135\u001b[0m, \u001b[38;5;34m240\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ block3b_expand_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m240\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_dwconv      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m135\u001b[0m, \u001b[38;5;34m240\u001b[0m,  │      \u001b[38;5;34m6,000\u001b[0m │ block3b_expand_a… │\n│ (\u001b[38;5;33mDepthwiseConv2D\u001b[0m)   │ \u001b[38;5;34m240\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_bn          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m135\u001b[0m, \u001b[38;5;34m240\u001b[0m,  │        \u001b[38;5;34m960\u001b[0m │ block3b_dwconv[\u001b[38;5;34m0\u001b[0m… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m240\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_activation  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m135\u001b[0m, \u001b[38;5;34m240\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ block3b_bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m240\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_se_squeeze  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m240\u001b[0m)       │          \u001b[38;5;34m0\u001b[0m │ block3b_activati… │\n│ (\u001b[38;5;33mGlobalAveragePool…\u001b[0m │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_se_reshape  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m240\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ block3b_se_squee… │\n│ (\u001b[38;5;33mReshape\u001b[0m)           │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_se_reduce   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m10\u001b[0m)  │      \u001b[38;5;34m2,410\u001b[0m │ block3b_se_resha… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_se_expand   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m240\u001b[0m) │      \u001b[38;5;34m2,640\u001b[0m │ block3b_se_reduc… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_se_excite   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m135\u001b[0m, \u001b[38;5;34m240\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ block3b_activati… │\n│ (\u001b[38;5;33mMultiply\u001b[0m)          │ \u001b[38;5;34m240\u001b[0m)              │            │ block3b_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_project_co… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m135\u001b[0m, \u001b[38;5;34m240\u001b[0m,  │      \u001b[38;5;34m9,600\u001b[0m │ block3b_se_excit… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m40\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_project_bn  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m135\u001b[0m, \u001b[38;5;34m240\u001b[0m,  │        \u001b[38;5;34m160\u001b[0m │ block3b_project_… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m40\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_drop        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m135\u001b[0m, \u001b[38;5;34m240\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ block3b_project_… │\n│ (\u001b[38;5;33mDropout\u001b[0m)           │ \u001b[38;5;34m40\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_add (\u001b[38;5;33mAdd\u001b[0m)   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m135\u001b[0m, \u001b[38;5;34m240\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ block3b_drop[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n│                     │ \u001b[38;5;34m40\u001b[0m)               │            │ block3a_project_… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_expand_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m135\u001b[0m, \u001b[38;5;34m240\u001b[0m,  │      \u001b[38;5;34m9,600\u001b[0m │ block3b_add[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m240\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_expand_bn   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m135\u001b[0m, \u001b[38;5;34m240\u001b[0m,  │        \u001b[38;5;34m960\u001b[0m │ block4a_expand_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m240\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_expand_act… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m135\u001b[0m, \u001b[38;5;34m240\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ block4a_expand_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m240\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_dwconv_pad  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m137\u001b[0m, \u001b[38;5;34m241\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ block4a_expand_a… │\n│ (\u001b[38;5;33mZeroPadding2D\u001b[0m)     │ \u001b[38;5;34m240\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_dwconv      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │      \u001b[38;5;34m2,160\u001b[0m │ block4a_dwconv_p… │\n│ (\u001b[38;5;33mDepthwiseConv2D\u001b[0m)   │ \u001b[38;5;34m240\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_bn          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │        \u001b[38;5;34m960\u001b[0m │ block4a_dwconv[\u001b[38;5;34m0\u001b[0m… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m240\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_activation  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block4a_bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m240\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_se_squeeze  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m240\u001b[0m)       │          \u001b[38;5;34m0\u001b[0m │ block4a_activati… │\n│ (\u001b[38;5;33mGlobalAveragePool…\u001b[0m │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_se_reshape  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m240\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ block4a_se_squee… │\n│ (\u001b[38;5;33mReshape\u001b[0m)           │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_se_reduce   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m10\u001b[0m)  │      \u001b[38;5;34m2,410\u001b[0m │ block4a_se_resha… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_se_expand   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m240\u001b[0m) │      \u001b[38;5;34m2,640\u001b[0m │ block4a_se_reduc… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_se_excite   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block4a_activati… │\n│ (\u001b[38;5;33mMultiply\u001b[0m)          │ \u001b[38;5;34m240\u001b[0m)              │            │ block4a_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_project_co… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │     \u001b[38;5;34m19,200\u001b[0m │ block4a_se_excit… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m80\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_project_bn  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │        \u001b[38;5;34m320\u001b[0m │ block4a_project_… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m80\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_expand_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │     \u001b[38;5;34m38,400\u001b[0m │ block4a_project_… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m480\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_expand_bn   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │      \u001b[38;5;34m1,920\u001b[0m │ block4b_expand_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m480\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_expand_act… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block4b_expand_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m480\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_dwconv      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │      \u001b[38;5;34m4,320\u001b[0m │ block4b_expand_a… │\n│ (\u001b[38;5;33mDepthwiseConv2D\u001b[0m)   │ \u001b[38;5;34m480\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_bn          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │      \u001b[38;5;34m1,920\u001b[0m │ block4b_dwconv[\u001b[38;5;34m0\u001b[0m… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m480\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_activation  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block4b_bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m480\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_se_squeeze  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m480\u001b[0m)       │          \u001b[38;5;34m0\u001b[0m │ block4b_activati… │\n│ (\u001b[38;5;33mGlobalAveragePool…\u001b[0m │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_se_reshape  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m480\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ block4b_se_squee… │\n│ (\u001b[38;5;33mReshape\u001b[0m)           │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_se_reduce   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m20\u001b[0m)  │      \u001b[38;5;34m9,620\u001b[0m │ block4b_se_resha… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_se_expand   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m480\u001b[0m) │     \u001b[38;5;34m10,080\u001b[0m │ block4b_se_reduc… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_se_excite   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block4b_activati… │\n│ (\u001b[38;5;33mMultiply\u001b[0m)          │ \u001b[38;5;34m480\u001b[0m)              │            │ block4b_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_project_co… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │     \u001b[38;5;34m38,400\u001b[0m │ block4b_se_excit… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m80\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_project_bn  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │        \u001b[38;5;34m320\u001b[0m │ block4b_project_… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m80\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_drop        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block4b_project_… │\n│ (\u001b[38;5;33mDropout\u001b[0m)           │ \u001b[38;5;34m80\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_add (\u001b[38;5;33mAdd\u001b[0m)   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block4b_drop[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n│                     │ \u001b[38;5;34m80\u001b[0m)               │            │ block4a_project_… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_expand_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │     \u001b[38;5;34m38,400\u001b[0m │ block4b_add[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m480\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_expand_bn   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │      \u001b[38;5;34m1,920\u001b[0m │ block4c_expand_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m480\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_expand_act… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block4c_expand_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m480\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_dwconv      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │      \u001b[38;5;34m4,320\u001b[0m │ block4c_expand_a… │\n│ (\u001b[38;5;33mDepthwiseConv2D\u001b[0m)   │ \u001b[38;5;34m480\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_bn          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │      \u001b[38;5;34m1,920\u001b[0m │ block4c_dwconv[\u001b[38;5;34m0\u001b[0m… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m480\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_activation  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block4c_bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m480\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_se_squeeze  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m480\u001b[0m)       │          \u001b[38;5;34m0\u001b[0m │ block4c_activati… │\n│ (\u001b[38;5;33mGlobalAveragePool…\u001b[0m │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_se_reshape  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m480\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ block4c_se_squee… │\n│ (\u001b[38;5;33mReshape\u001b[0m)           │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_se_reduce   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m20\u001b[0m)  │      \u001b[38;5;34m9,620\u001b[0m │ block4c_se_resha… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_se_expand   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m480\u001b[0m) │     \u001b[38;5;34m10,080\u001b[0m │ block4c_se_reduc… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_se_excite   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block4c_activati… │\n│ (\u001b[38;5;33mMultiply\u001b[0m)          │ \u001b[38;5;34m480\u001b[0m)              │            │ block4c_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_project_co… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │     \u001b[38;5;34m38,400\u001b[0m │ block4c_se_excit… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m80\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_project_bn  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │        \u001b[38;5;34m320\u001b[0m │ block4c_project_… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m80\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_drop        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block4c_project_… │\n│ (\u001b[38;5;33mDropout\u001b[0m)           │ \u001b[38;5;34m80\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_add (\u001b[38;5;33mAdd\u001b[0m)   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block4c_drop[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n│                     │ \u001b[38;5;34m80\u001b[0m)               │            │ block4b_add[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5a_expand_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │     \u001b[38;5;34m38,400\u001b[0m │ block4c_add[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m480\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5a_expand_bn   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │      \u001b[38;5;34m1,920\u001b[0m │ block5a_expand_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m480\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5a_expand_act… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block5a_expand_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m480\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5a_dwconv      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │     \u001b[38;5;34m12,000\u001b[0m │ block5a_expand_a… │\n│ (\u001b[38;5;33mDepthwiseConv2D\u001b[0m)   │ \u001b[38;5;34m480\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5a_bn          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │      \u001b[38;5;34m1,920\u001b[0m │ block5a_dwconv[\u001b[38;5;34m0\u001b[0m… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m480\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5a_activation  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block5a_bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m480\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5a_se_squeeze  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m480\u001b[0m)       │          \u001b[38;5;34m0\u001b[0m │ block5a_activati… │\n│ (\u001b[38;5;33mGlobalAveragePool…\u001b[0m │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5a_se_reshape  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m480\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ block5a_se_squee… │\n│ (\u001b[38;5;33mReshape\u001b[0m)           │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5a_se_reduce   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m20\u001b[0m)  │      \u001b[38;5;34m9,620\u001b[0m │ block5a_se_resha… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5a_se_expand   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m480\u001b[0m) │     \u001b[38;5;34m10,080\u001b[0m │ block5a_se_reduc… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5a_se_excite   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block5a_activati… │\n│ (\u001b[38;5;33mMultiply\u001b[0m)          │ \u001b[38;5;34m480\u001b[0m)              │            │ block5a_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5a_project_co… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │     \u001b[38;5;34m53,760\u001b[0m │ block5a_se_excit… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m112\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5a_project_bn  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │        \u001b[38;5;34m448\u001b[0m │ block5a_project_… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m112\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_expand_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │     \u001b[38;5;34m75,264\u001b[0m │ block5a_project_… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m672\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_expand_bn   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │      \u001b[38;5;34m2,688\u001b[0m │ block5b_expand_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m672\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_expand_act… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block5b_expand_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m672\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_dwconv      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │     \u001b[38;5;34m16,800\u001b[0m │ block5b_expand_a… │\n│ (\u001b[38;5;33mDepthwiseConv2D\u001b[0m)   │ \u001b[38;5;34m672\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_bn          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │      \u001b[38;5;34m2,688\u001b[0m │ block5b_dwconv[\u001b[38;5;34m0\u001b[0m… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m672\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_activation  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block5b_bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m672\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_se_squeeze  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m672\u001b[0m)       │          \u001b[38;5;34m0\u001b[0m │ block5b_activati… │\n│ (\u001b[38;5;33mGlobalAveragePool…\u001b[0m │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_se_reshape  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m672\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ block5b_se_squee… │\n│ (\u001b[38;5;33mReshape\u001b[0m)           │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_se_reduce   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m28\u001b[0m)  │     \u001b[38;5;34m18,844\u001b[0m │ block5b_se_resha… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_se_expand   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m672\u001b[0m) │     \u001b[38;5;34m19,488\u001b[0m │ block5b_se_reduc… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_se_excite   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block5b_activati… │\n│ (\u001b[38;5;33mMultiply\u001b[0m)          │ \u001b[38;5;34m672\u001b[0m)              │            │ block5b_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_project_co… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │     \u001b[38;5;34m75,264\u001b[0m │ block5b_se_excit… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m112\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_project_bn  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │        \u001b[38;5;34m448\u001b[0m │ block5b_project_… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m112\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_drop        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block5b_project_… │\n│ (\u001b[38;5;33mDropout\u001b[0m)           │ \u001b[38;5;34m112\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_add (\u001b[38;5;33mAdd\u001b[0m)   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block5b_drop[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n│                     │ \u001b[38;5;34m112\u001b[0m)              │            │ block5a_project_… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_expand_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │     \u001b[38;5;34m75,264\u001b[0m │ block5b_add[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m672\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_expand_bn   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │      \u001b[38;5;34m2,688\u001b[0m │ block5c_expand_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m672\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_expand_act… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block5c_expand_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m672\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_dwconv      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │     \u001b[38;5;34m16,800\u001b[0m │ block5c_expand_a… │\n│ (\u001b[38;5;33mDepthwiseConv2D\u001b[0m)   │ \u001b[38;5;34m672\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_bn          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │      \u001b[38;5;34m2,688\u001b[0m │ block5c_dwconv[\u001b[38;5;34m0\u001b[0m… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m672\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_activation  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block5c_bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m672\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_se_squeeze  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m672\u001b[0m)       │          \u001b[38;5;34m0\u001b[0m │ block5c_activati… │\n│ (\u001b[38;5;33mGlobalAveragePool…\u001b[0m │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_se_reshape  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m672\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ block5c_se_squee… │\n│ (\u001b[38;5;33mReshape\u001b[0m)           │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_se_reduce   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m28\u001b[0m)  │     \u001b[38;5;34m18,844\u001b[0m │ block5c_se_resha… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_se_expand   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m672\u001b[0m) │     \u001b[38;5;34m19,488\u001b[0m │ block5c_se_reduc… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_se_excite   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block5c_activati… │\n│ (\u001b[38;5;33mMultiply\u001b[0m)          │ \u001b[38;5;34m672\u001b[0m)              │            │ block5c_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_project_co… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │     \u001b[38;5;34m75,264\u001b[0m │ block5c_se_excit… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m112\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_project_bn  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │        \u001b[38;5;34m448\u001b[0m │ block5c_project_… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m112\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_drop        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block5c_project_… │\n│ (\u001b[38;5;33mDropout\u001b[0m)           │ \u001b[38;5;34m112\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_add (\u001b[38;5;33mAdd\u001b[0m)   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block5c_drop[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n│                     │ \u001b[38;5;34m112\u001b[0m)              │            │ block5b_add[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_expand_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │     \u001b[38;5;34m75,264\u001b[0m │ block5c_add[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m672\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_expand_bn   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │      \u001b[38;5;34m2,688\u001b[0m │ block6a_expand_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m672\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_expand_act… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block6a_expand_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m672\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_dwconv_pad  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m71\u001b[0m, \u001b[38;5;34m123\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ block6a_expand_a… │\n│ (\u001b[38;5;33mZeroPadding2D\u001b[0m)     │ \u001b[38;5;34m672\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_dwconv      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │     \u001b[38;5;34m16,800\u001b[0m │ block6a_dwconv_p… │\n│ (\u001b[38;5;33mDepthwiseConv2D\u001b[0m)   │ \u001b[38;5;34m672\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_bn          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │      \u001b[38;5;34m2,688\u001b[0m │ block6a_dwconv[\u001b[38;5;34m0\u001b[0m… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m672\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_activation  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ block6a_bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m672\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_se_squeeze  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m672\u001b[0m)       │          \u001b[38;5;34m0\u001b[0m │ block6a_activati… │\n│ (\u001b[38;5;33mGlobalAveragePool…\u001b[0m │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_se_reshape  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m672\u001b[0m) │          \u001b[38;5;34m0\u001b[0m │ block6a_se_squee… │\n│ (\u001b[38;5;33mReshape\u001b[0m)           │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_se_reduce   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m28\u001b[0m)  │     \u001b[38;5;34m18,844\u001b[0m │ block6a_se_resha… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_se_expand   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m672\u001b[0m) │     \u001b[38;5;34m19,488\u001b[0m │ block6a_se_reduc… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_se_excite   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ block6a_activati… │\n│ (\u001b[38;5;33mMultiply\u001b[0m)          │ \u001b[38;5;34m672\u001b[0m)              │            │ block6a_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_project_co… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │    \u001b[38;5;34m129,024\u001b[0m │ block6a_se_excit… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m192\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_project_bn  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │        \u001b[38;5;34m768\u001b[0m │ block6a_project_… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_expand_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │    \u001b[38;5;34m221,184\u001b[0m │ block6a_project_… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_expand_bn   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │      \u001b[38;5;34m4,608\u001b[0m │ block6b_expand_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_expand_act… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ block6b_expand_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_dwconv      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │     \u001b[38;5;34m28,800\u001b[0m │ block6b_expand_a… │\n│ (\u001b[38;5;33mDepthwiseConv2D\u001b[0m)   │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_bn          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │      \u001b[38;5;34m4,608\u001b[0m │ block6b_dwconv[\u001b[38;5;34m0\u001b[0m… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_activation  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ block6b_bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_se_squeeze  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1152\u001b[0m)      │          \u001b[38;5;34m0\u001b[0m │ block6b_activati… │\n│ (\u001b[38;5;33mGlobalAveragePool…\u001b[0m │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_se_reshape  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ block6b_se_squee… │\n│ (\u001b[38;5;33mReshape\u001b[0m)           │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_se_reduce   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m48\u001b[0m)  │     \u001b[38;5;34m55,344\u001b[0m │ block6b_se_resha… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_se_expand   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m,      │     \u001b[38;5;34m56,448\u001b[0m │ block6b_se_reduc… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_se_excite   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ block6b_activati… │\n│ (\u001b[38;5;33mMultiply\u001b[0m)          │ \u001b[38;5;34m1152\u001b[0m)             │            │ block6b_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_project_co… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │    \u001b[38;5;34m221,184\u001b[0m │ block6b_se_excit… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m192\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_project_bn  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │        \u001b[38;5;34m768\u001b[0m │ block6b_project_… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_drop        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ block6b_project_… │\n│ (\u001b[38;5;33mDropout\u001b[0m)           │ \u001b[38;5;34m192\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_add (\u001b[38;5;33mAdd\u001b[0m)   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ block6b_drop[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n│                     │ \u001b[38;5;34m192\u001b[0m)              │            │ block6a_project_… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_expand_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │    \u001b[38;5;34m221,184\u001b[0m │ block6b_add[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_expand_bn   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │      \u001b[38;5;34m4,608\u001b[0m │ block6c_expand_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_expand_act… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ block6c_expand_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_dwconv      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │     \u001b[38;5;34m28,800\u001b[0m │ block6c_expand_a… │\n│ (\u001b[38;5;33mDepthwiseConv2D\u001b[0m)   │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_bn          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │      \u001b[38;5;34m4,608\u001b[0m │ block6c_dwconv[\u001b[38;5;34m0\u001b[0m… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_activation  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ block6c_bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_se_squeeze  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1152\u001b[0m)      │          \u001b[38;5;34m0\u001b[0m │ block6c_activati… │\n│ (\u001b[38;5;33mGlobalAveragePool…\u001b[0m │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_se_reshape  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ block6c_se_squee… │\n│ (\u001b[38;5;33mReshape\u001b[0m)           │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_se_reduce   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m48\u001b[0m)  │     \u001b[38;5;34m55,344\u001b[0m │ block6c_se_resha… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_se_expand   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m,      │     \u001b[38;5;34m56,448\u001b[0m │ block6c_se_reduc… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_se_excite   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ block6c_activati… │\n│ (\u001b[38;5;33mMultiply\u001b[0m)          │ \u001b[38;5;34m1152\u001b[0m)             │            │ block6c_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_project_co… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │    \u001b[38;5;34m221,184\u001b[0m │ block6c_se_excit… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m192\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_project_bn  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │        \u001b[38;5;34m768\u001b[0m │ block6c_project_… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_drop        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ block6c_project_… │\n│ (\u001b[38;5;33mDropout\u001b[0m)           │ \u001b[38;5;34m192\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_add (\u001b[38;5;33mAdd\u001b[0m)   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ block6c_drop[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n│                     │ \u001b[38;5;34m192\u001b[0m)              │            │ block6b_add[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_expand_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │    \u001b[38;5;34m221,184\u001b[0m │ block6c_add[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_expand_bn   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │      \u001b[38;5;34m4,608\u001b[0m │ block6d_expand_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_expand_act… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ block6d_expand_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_dwconv      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │     \u001b[38;5;34m28,800\u001b[0m │ block6d_expand_a… │\n│ (\u001b[38;5;33mDepthwiseConv2D\u001b[0m)   │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_bn          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │      \u001b[38;5;34m4,608\u001b[0m │ block6d_dwconv[\u001b[38;5;34m0\u001b[0m… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_activation  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ block6d_bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_se_squeeze  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1152\u001b[0m)      │          \u001b[38;5;34m0\u001b[0m │ block6d_activati… │\n│ (\u001b[38;5;33mGlobalAveragePool…\u001b[0m │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_se_reshape  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ block6d_se_squee… │\n│ (\u001b[38;5;33mReshape\u001b[0m)           │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_se_reduce   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m48\u001b[0m)  │     \u001b[38;5;34m55,344\u001b[0m │ block6d_se_resha… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_se_expand   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m,      │     \u001b[38;5;34m56,448\u001b[0m │ block6d_se_reduc… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_se_excite   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ block6d_activati… │\n│ (\u001b[38;5;33mMultiply\u001b[0m)          │ \u001b[38;5;34m1152\u001b[0m)             │            │ block6d_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_project_co… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │    \u001b[38;5;34m221,184\u001b[0m │ block6d_se_excit… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m192\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_project_bn  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │        \u001b[38;5;34m768\u001b[0m │ block6d_project_… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m192\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_drop        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ block6d_project_… │\n│ (\u001b[38;5;33mDropout\u001b[0m)           │ \u001b[38;5;34m192\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_add (\u001b[38;5;33mAdd\u001b[0m)   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ block6d_drop[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n│                     │ \u001b[38;5;34m192\u001b[0m)              │            │ block6c_add[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block7a_expand_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │    \u001b[38;5;34m221,184\u001b[0m │ block6d_add[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block7a_expand_bn   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │      \u001b[38;5;34m4,608\u001b[0m │ block7a_expand_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block7a_expand_act… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ block7a_expand_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block7a_dwconv      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │     \u001b[38;5;34m10,368\u001b[0m │ block7a_expand_a… │\n│ (\u001b[38;5;33mDepthwiseConv2D\u001b[0m)   │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block7a_bn          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │      \u001b[38;5;34m4,608\u001b[0m │ block7a_dwconv[\u001b[38;5;34m0\u001b[0m… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block7a_activation  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ block7a_bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]  │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block7a_se_squeeze  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1152\u001b[0m)      │          \u001b[38;5;34m0\u001b[0m │ block7a_activati… │\n│ (\u001b[38;5;33mGlobalAveragePool…\u001b[0m │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block7a_se_reshape  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ block7a_se_squee… │\n│ (\u001b[38;5;33mReshape\u001b[0m)           │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block7a_se_reduce   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m48\u001b[0m)  │     \u001b[38;5;34m55,344\u001b[0m │ block7a_se_resha… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block7a_se_expand   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m,      │     \u001b[38;5;34m56,448\u001b[0m │ block7a_se_reduc… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m1152\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block7a_se_excite   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ block7a_activati… │\n│ (\u001b[38;5;33mMultiply\u001b[0m)          │ \u001b[38;5;34m1152\u001b[0m)             │            │ block7a_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block7a_project_co… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │    \u001b[38;5;34m368,640\u001b[0m │ block7a_se_excit… │\n│ (\u001b[38;5;33mConv2D\u001b[0m)            │ \u001b[38;5;34m320\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block7a_project_bn  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │      \u001b[38;5;34m1,280\u001b[0m │ block7a_project_… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m320\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ top_conv (\u001b[38;5;33mConv2D\u001b[0m)   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │    \u001b[38;5;34m409,600\u001b[0m │ block7a_project_… │\n│                     │ \u001b[38;5;34m1280\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ top_bn              │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │      \u001b[38;5;34m5,120\u001b[0m │ top_conv[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]    │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m1280\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ top_activation      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m34\u001b[0m, \u001b[38;5;34m60\u001b[0m,    │          \u001b[38;5;34m0\u001b[0m │ top_bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]      │\n│ (\u001b[38;5;33mActivation\u001b[0m)        │ \u001b[38;5;34m1280\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ up_sampling2d       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ top_activation[\u001b[38;5;34m0\u001b[0m… │\n│ (\u001b[38;5;33mUpSampling2D\u001b[0m)      │ \u001b[38;5;34m1280\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2d (\u001b[38;5;33mConv2D\u001b[0m)     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │  \u001b[38;5;34m5,898,752\u001b[0m │ up_sampling2d[\u001b[38;5;34m0\u001b[0m]… │\n│                     │ \u001b[38;5;34m512\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ concatenate         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m68\u001b[0m, \u001b[38;5;34m120\u001b[0m,   │          \u001b[38;5;34m0\u001b[0m │ conv2d[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m],     │\n│ (\u001b[38;5;33mConcatenate\u001b[0m)       │ \u001b[38;5;34m1184\u001b[0m)             │            │ block6a_expand_a… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ up_sampling2d_1     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m136\u001b[0m, \u001b[38;5;34m240\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ concatenate[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n│ (\u001b[38;5;33mUpSampling2D\u001b[0m)      │ \u001b[38;5;34m1184\u001b[0m)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2d_1 (\u001b[38;5;33mConv2D\u001b[0m)   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m136\u001b[0m, \u001b[38;5;34m240\u001b[0m,  │  \u001b[38;5;34m2,728,192\u001b[0m │ up_sampling2d_1[\u001b[38;5;34m…\u001b[0m │\n│                     │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ cropping2d          │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m135\u001b[0m, \u001b[38;5;34m240\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ conv2d_1[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]    │\n│ (\u001b[38;5;33mCropping2D\u001b[0m)        │ \u001b[38;5;34m256\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ concatenate_1       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m135\u001b[0m, \u001b[38;5;34m240\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ cropping2d[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m], │\n│ (\u001b[38;5;33mConcatenate\u001b[0m)       │ \u001b[38;5;34m496\u001b[0m)              │            │ block4a_expand_a… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ up_sampling2d_2     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m270\u001b[0m, \u001b[38;5;34m480\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ concatenate_1[\u001b[38;5;34m0\u001b[0m]… │\n│ (\u001b[38;5;33mUpSampling2D\u001b[0m)      │ \u001b[38;5;34m496\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2d_2 (\u001b[38;5;33mConv2D\u001b[0m)   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m270\u001b[0m, \u001b[38;5;34m480\u001b[0m,  │    \u001b[38;5;34m571,520\u001b[0m │ up_sampling2d_2[\u001b[38;5;34m…\u001b[0m │\n│                     │ \u001b[38;5;34m128\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ concatenate_2       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m270\u001b[0m, \u001b[38;5;34m480\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ conv2d_2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m],   │\n│ (\u001b[38;5;33mConcatenate\u001b[0m)       │ \u001b[38;5;34m272\u001b[0m)              │            │ block3a_expand_a… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ up_sampling2d_3     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m540\u001b[0m, \u001b[38;5;34m960\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ concatenate_2[\u001b[38;5;34m0\u001b[0m]… │\n│ (\u001b[38;5;33mUpSampling2D\u001b[0m)      │ \u001b[38;5;34m272\u001b[0m)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2d_3 (\u001b[38;5;33mConv2D\u001b[0m)   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m540\u001b[0m, \u001b[38;5;34m960\u001b[0m,  │    \u001b[38;5;34m156,736\u001b[0m │ up_sampling2d_3[\u001b[38;5;34m…\u001b[0m │\n│                     │ \u001b[38;5;34m64\u001b[0m)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ concatenate_3       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m540\u001b[0m, \u001b[38;5;34m960\u001b[0m,  │          \u001b[38;5;34m0\u001b[0m │ conv2d_3[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m],   │\n│ (\u001b[38;5;33mConcatenate\u001b[0m)       │ \u001b[38;5;34m160\u001b[0m)              │            │ block2a_expand_a… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ up_sampling2d_4     │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1080\u001b[0m,      │          \u001b[38;5;34m0\u001b[0m │ concatenate_3[\u001b[38;5;34m0\u001b[0m]… │\n│ (\u001b[38;5;33mUpSampling2D\u001b[0m)      │ \u001b[38;5;34m1920\u001b[0m, \u001b[38;5;34m160\u001b[0m)        │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2d_4 (\u001b[38;5;33mConv2D\u001b[0m)   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1080\u001b[0m,      │     \u001b[38;5;34m46,112\u001b[0m │ up_sampling2d_4[\u001b[38;5;34m…\u001b[0m │\n│                     │ \u001b[38;5;34m1920\u001b[0m, \u001b[38;5;34m32\u001b[0m)         │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2d_5 (\u001b[38;5;33mConv2D\u001b[0m)   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1080\u001b[0m,      │         \u001b[38;5;34m33\u001b[0m │ conv2d_4[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m]    │\n│                     │ \u001b[38;5;34m1920\u001b[0m, \u001b[38;5;34m1\u001b[0m)          │            │                   │\n└─────────────────────┴───────────────────┴────────────┴───────────────────┘\n","text/html":"<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┓\n┃<span style=\"font-weight: bold\"> Layer (type)        </span>┃<span style=\"font-weight: bold\"> Output Shape      </span>┃<span style=\"font-weight: bold\">    Param # </span>┃<span style=\"font-weight: bold\"> Connected to      </span>┃\n┡━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━┩\n│ input_layer         │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1080</span>,      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ -                 │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1920</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)          │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ rescaling           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1080</span>,      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ input_layer[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>] │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Rescaling</span>)         │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1920</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)          │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ normalization       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1080</span>,      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">7</span> │ rescaling[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Normalization</span>)     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1920</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)          │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ rescaling_1         │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1080</span>,      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ normalization[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Rescaling</span>)         │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1920</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)          │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ stem_conv_pad       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1081</span>,      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ rescaling_1[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>] │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">ZeroPadding2D</span>)     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1921</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)          │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ stem_conv (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">540</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">960</span>,  │        <span style=\"color: #00af00; text-decoration-color: #00af00\">864</span> │ stem_conv_pad[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ stem_bn             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">540</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">960</span>,  │        <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span> │ stem_conv[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]   │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ stem_activation     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">540</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">960</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ stem_bn[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]     │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block1a_dwconv      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">540</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">960</span>,  │        <span style=\"color: #00af00; text-decoration-color: #00af00\">288</span> │ stem_activation[<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">DepthwiseConv2D</span>)   │ <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block1a_bn          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">540</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">960</span>,  │        <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span> │ block1a_dwconv[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block1a_activation  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">540</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">960</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block1a_bn[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block1a_se_squeeze  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)        │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block1a_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GlobalAveragePool…</span> │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block1a_se_reshape  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block1a_se_squee… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Reshape</span>)           │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block1a_se_reduce   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>)   │        <span style=\"color: #00af00; text-decoration-color: #00af00\">264</span> │ block1a_se_resha… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block1a_se_expand   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)  │        <span style=\"color: #00af00; text-decoration-color: #00af00\">288</span> │ block1a_se_reduc… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block1a_se_excite   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">540</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">960</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block1a_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Multiply</span>)          │ <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)               │            │ block1a_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block1a_project_co… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">540</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">960</span>,  │        <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span> │ block1a_se_excit… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block1a_project_bn  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">540</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">960</span>,  │         <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span> │ block1a_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_expand_conv │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">540</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">960</span>,  │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,536</span> │ block1a_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">96</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_expand_bn   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">540</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">960</span>,  │        <span style=\"color: #00af00; text-decoration-color: #00af00\">384</span> │ block2a_expand_c… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">96</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_expand_act… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">540</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">960</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block2a_expand_b… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">96</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_dwconv_pad  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">541</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">961</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block2a_expand_a… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">ZeroPadding2D</span>)     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">96</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_dwconv      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">270</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>,  │        <span style=\"color: #00af00; text-decoration-color: #00af00\">864</span> │ block2a_dwconv_p… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">DepthwiseConv2D</span>)   │ <span style=\"color: #00af00; text-decoration-color: #00af00\">96</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_bn          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">270</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>,  │        <span style=\"color: #00af00; text-decoration-color: #00af00\">384</span> │ block2a_dwconv[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">96</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_activation  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">270</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block2a_bn[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">96</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_se_squeeze  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">96</span>)        │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block2a_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GlobalAveragePool…</span> │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_se_reshape  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">96</span>)  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block2a_se_squee… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Reshape</span>)           │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_se_reduce   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">4</span>)   │        <span style=\"color: #00af00; text-decoration-color: #00af00\">388</span> │ block2a_se_resha… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_se_expand   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">96</span>)  │        <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span> │ block2a_se_reduc… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_se_excite   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">270</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block2a_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Multiply</span>)          │ <span style=\"color: #00af00; text-decoration-color: #00af00\">96</span>)               │            │ block2a_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_project_co… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">270</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>,  │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,304</span> │ block2a_se_excit… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">24</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2a_project_bn  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">270</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>,  │         <span style=\"color: #00af00; text-decoration-color: #00af00\">96</span> │ block2a_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">24</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_expand_conv │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">270</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>,  │      <span style=\"color: #00af00; text-decoration-color: #00af00\">3,456</span> │ block2a_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_expand_bn   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">270</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>,  │        <span style=\"color: #00af00; text-decoration-color: #00af00\">576</span> │ block2b_expand_c… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_expand_act… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">270</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block2b_expand_b… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_dwconv      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">270</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>,  │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,296</span> │ block2b_expand_a… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">DepthwiseConv2D</span>)   │ <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_bn          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">270</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>,  │        <span style=\"color: #00af00; text-decoration-color: #00af00\">576</span> │ block2b_dwconv[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_activation  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">270</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block2b_bn[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_se_squeeze  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>)       │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block2b_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GlobalAveragePool…</span> │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_se_reshape  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block2b_se_squee… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Reshape</span>)           │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_se_reduce   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">6</span>)   │        <span style=\"color: #00af00; text-decoration-color: #00af00\">870</span> │ block2b_se_resha… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_se_expand   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,008</span> │ block2b_se_reduc… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_se_excite   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">270</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block2b_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Multiply</span>)          │ <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>)              │            │ block2b_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_project_co… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">270</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>,  │      <span style=\"color: #00af00; text-decoration-color: #00af00\">3,456</span> │ block2b_se_excit… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">24</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_project_bn  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">270</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>,  │         <span style=\"color: #00af00; text-decoration-color: #00af00\">96</span> │ block2b_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">24</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_drop        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">270</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block2b_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)           │ <span style=\"color: #00af00; text-decoration-color: #00af00\">24</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block2b_add (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</span>)   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">270</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block2b_drop[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">24</span>)               │            │ block2a_project_… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_expand_conv │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">270</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>,  │      <span style=\"color: #00af00; text-decoration-color: #00af00\">3,456</span> │ block2b_add[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>] │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_expand_bn   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">270</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>,  │        <span style=\"color: #00af00; text-decoration-color: #00af00\">576</span> │ block3a_expand_c… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_expand_act… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">270</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block3a_expand_b… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_dwconv_pad  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">273</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">483</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block3a_expand_a… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">ZeroPadding2D</span>)     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_dwconv      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">135</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>,  │      <span style=\"color: #00af00; text-decoration-color: #00af00\">3,600</span> │ block3a_dwconv_p… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">DepthwiseConv2D</span>)   │ <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_bn          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">135</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>,  │        <span style=\"color: #00af00; text-decoration-color: #00af00\">576</span> │ block3a_dwconv[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_activation  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">135</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block3a_bn[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_se_squeeze  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>)       │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block3a_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GlobalAveragePool…</span> │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_se_reshape  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block3a_se_squee… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Reshape</span>)           │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_se_reduce   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">6</span>)   │        <span style=\"color: #00af00; text-decoration-color: #00af00\">870</span> │ block3a_se_resha… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_se_expand   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,008</span> │ block3a_se_reduc… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_se_excite   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">135</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block3a_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Multiply</span>)          │ <span style=\"color: #00af00; text-decoration-color: #00af00\">144</span>)              │            │ block3a_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_project_co… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">135</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>,  │      <span style=\"color: #00af00; text-decoration-color: #00af00\">5,760</span> │ block3a_se_excit… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">40</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3a_project_bn  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">135</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>,  │        <span style=\"color: #00af00; text-decoration-color: #00af00\">160</span> │ block3a_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">40</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_expand_conv │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">135</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>,  │      <span style=\"color: #00af00; text-decoration-color: #00af00\">9,600</span> │ block3a_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_expand_bn   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">135</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>,  │        <span style=\"color: #00af00; text-decoration-color: #00af00\">960</span> │ block3b_expand_c… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_expand_act… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">135</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block3b_expand_b… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_dwconv      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">135</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>,  │      <span style=\"color: #00af00; text-decoration-color: #00af00\">6,000</span> │ block3b_expand_a… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">DepthwiseConv2D</span>)   │ <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_bn          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">135</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>,  │        <span style=\"color: #00af00; text-decoration-color: #00af00\">960</span> │ block3b_dwconv[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_activation  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">135</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block3b_bn[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_se_squeeze  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>)       │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block3b_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GlobalAveragePool…</span> │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_se_reshape  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block3b_se_squee… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Reshape</span>)           │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_se_reduce   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">10</span>)  │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,410</span> │ block3b_se_resha… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_se_expand   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,640</span> │ block3b_se_reduc… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_se_excite   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">135</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block3b_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Multiply</span>)          │ <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>)              │            │ block3b_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_project_co… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">135</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>,  │      <span style=\"color: #00af00; text-decoration-color: #00af00\">9,600</span> │ block3b_se_excit… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">40</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_project_bn  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">135</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>,  │        <span style=\"color: #00af00; text-decoration-color: #00af00\">160</span> │ block3b_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">40</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_drop        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">135</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block3b_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)           │ <span style=\"color: #00af00; text-decoration-color: #00af00\">40</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block3b_add (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</span>)   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">135</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block3b_drop[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">40</span>)               │            │ block3a_project_… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_expand_conv │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">135</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>,  │      <span style=\"color: #00af00; text-decoration-color: #00af00\">9,600</span> │ block3b_add[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>] │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_expand_bn   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">135</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>,  │        <span style=\"color: #00af00; text-decoration-color: #00af00\">960</span> │ block4a_expand_c… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_expand_act… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">135</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block4a_expand_b… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_dwconv_pad  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">137</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">241</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block4a_expand_a… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">ZeroPadding2D</span>)     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_dwconv      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,160</span> │ block4a_dwconv_p… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">DepthwiseConv2D</span>)   │ <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_bn          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │        <span style=\"color: #00af00; text-decoration-color: #00af00\">960</span> │ block4a_dwconv[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_activation  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block4a_bn[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_se_squeeze  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>)       │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block4a_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GlobalAveragePool…</span> │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_se_reshape  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block4a_se_squee… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Reshape</span>)           │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_se_reduce   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">10</span>)  │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,410</span> │ block4a_se_resha… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_se_expand   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>) │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,640</span> │ block4a_se_reduc… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_se_excite   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block4a_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Multiply</span>)          │ <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>)              │            │ block4a_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_project_co… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │     <span style=\"color: #00af00; text-decoration-color: #00af00\">19,200</span> │ block4a_se_excit… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">80</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4a_project_bn  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │        <span style=\"color: #00af00; text-decoration-color: #00af00\">320</span> │ block4a_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">80</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_expand_conv │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │     <span style=\"color: #00af00; text-decoration-color: #00af00\">38,400</span> │ block4a_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_expand_bn   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,920</span> │ block4b_expand_c… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_expand_act… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block4b_expand_b… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_dwconv      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │      <span style=\"color: #00af00; text-decoration-color: #00af00\">4,320</span> │ block4b_expand_a… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">DepthwiseConv2D</span>)   │ <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_bn          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,920</span> │ block4b_dwconv[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_activation  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block4b_bn[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_se_squeeze  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>)       │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block4b_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GlobalAveragePool…</span> │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_se_reshape  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block4b_se_squee… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Reshape</span>)           │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_se_reduce   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">20</span>)  │      <span style=\"color: #00af00; text-decoration-color: #00af00\">9,620</span> │ block4b_se_resha… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_se_expand   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>) │     <span style=\"color: #00af00; text-decoration-color: #00af00\">10,080</span> │ block4b_se_reduc… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_se_excite   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block4b_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Multiply</span>)          │ <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>)              │            │ block4b_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_project_co… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │     <span style=\"color: #00af00; text-decoration-color: #00af00\">38,400</span> │ block4b_se_excit… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">80</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_project_bn  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │        <span style=\"color: #00af00; text-decoration-color: #00af00\">320</span> │ block4b_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">80</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_drop        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block4b_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)           │ <span style=\"color: #00af00; text-decoration-color: #00af00\">80</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4b_add (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</span>)   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block4b_drop[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">80</span>)               │            │ block4a_project_… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_expand_conv │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │     <span style=\"color: #00af00; text-decoration-color: #00af00\">38,400</span> │ block4b_add[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>] │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_expand_bn   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,920</span> │ block4c_expand_c… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_expand_act… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block4c_expand_b… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_dwconv      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │      <span style=\"color: #00af00; text-decoration-color: #00af00\">4,320</span> │ block4c_expand_a… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">DepthwiseConv2D</span>)   │ <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_bn          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,920</span> │ block4c_dwconv[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_activation  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block4c_bn[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_se_squeeze  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>)       │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block4c_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GlobalAveragePool…</span> │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_se_reshape  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block4c_se_squee… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Reshape</span>)           │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_se_reduce   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">20</span>)  │      <span style=\"color: #00af00; text-decoration-color: #00af00\">9,620</span> │ block4c_se_resha… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_se_expand   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>) │     <span style=\"color: #00af00; text-decoration-color: #00af00\">10,080</span> │ block4c_se_reduc… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_se_excite   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block4c_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Multiply</span>)          │ <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>)              │            │ block4c_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_project_co… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │     <span style=\"color: #00af00; text-decoration-color: #00af00\">38,400</span> │ block4c_se_excit… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">80</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_project_bn  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │        <span style=\"color: #00af00; text-decoration-color: #00af00\">320</span> │ block4c_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">80</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_drop        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block4c_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)           │ <span style=\"color: #00af00; text-decoration-color: #00af00\">80</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block4c_add (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</span>)   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block4c_drop[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">80</span>)               │            │ block4b_add[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>] │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5a_expand_conv │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │     <span style=\"color: #00af00; text-decoration-color: #00af00\">38,400</span> │ block4c_add[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>] │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5a_expand_bn   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,920</span> │ block5a_expand_c… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5a_expand_act… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block5a_expand_b… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5a_dwconv      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │     <span style=\"color: #00af00; text-decoration-color: #00af00\">12,000</span> │ block5a_expand_a… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">DepthwiseConv2D</span>)   │ <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5a_bn          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,920</span> │ block5a_dwconv[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5a_activation  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block5a_bn[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5a_se_squeeze  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>)       │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block5a_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GlobalAveragePool…</span> │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5a_se_reshape  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block5a_se_squee… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Reshape</span>)           │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5a_se_reduce   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">20</span>)  │      <span style=\"color: #00af00; text-decoration-color: #00af00\">9,620</span> │ block5a_se_resha… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5a_se_expand   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>) │     <span style=\"color: #00af00; text-decoration-color: #00af00\">10,080</span> │ block5a_se_reduc… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5a_se_excite   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block5a_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Multiply</span>)          │ <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>)              │            │ block5a_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5a_project_co… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │     <span style=\"color: #00af00; text-decoration-color: #00af00\">53,760</span> │ block5a_se_excit… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5a_project_bn  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │        <span style=\"color: #00af00; text-decoration-color: #00af00\">448</span> │ block5a_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_expand_conv │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │     <span style=\"color: #00af00; text-decoration-color: #00af00\">75,264</span> │ block5a_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_expand_bn   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,688</span> │ block5b_expand_c… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_expand_act… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block5b_expand_b… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_dwconv      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │     <span style=\"color: #00af00; text-decoration-color: #00af00\">16,800</span> │ block5b_expand_a… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">DepthwiseConv2D</span>)   │ <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_bn          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,688</span> │ block5b_dwconv[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_activation  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block5b_bn[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_se_squeeze  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>)       │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block5b_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GlobalAveragePool…</span> │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_se_reshape  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block5b_se_squee… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Reshape</span>)           │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_se_reduce   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>)  │     <span style=\"color: #00af00; text-decoration-color: #00af00\">18,844</span> │ block5b_se_resha… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_se_expand   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>) │     <span style=\"color: #00af00; text-decoration-color: #00af00\">19,488</span> │ block5b_se_reduc… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_se_excite   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block5b_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Multiply</span>)          │ <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>)              │            │ block5b_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_project_co… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │     <span style=\"color: #00af00; text-decoration-color: #00af00\">75,264</span> │ block5b_se_excit… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_project_bn  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │        <span style=\"color: #00af00; text-decoration-color: #00af00\">448</span> │ block5b_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_drop        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block5b_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)           │ <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5b_add (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</span>)   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block5b_drop[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>)              │            │ block5a_project_… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_expand_conv │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │     <span style=\"color: #00af00; text-decoration-color: #00af00\">75,264</span> │ block5b_add[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>] │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_expand_bn   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,688</span> │ block5c_expand_c… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_expand_act… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block5c_expand_b… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_dwconv      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │     <span style=\"color: #00af00; text-decoration-color: #00af00\">16,800</span> │ block5c_expand_a… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">DepthwiseConv2D</span>)   │ <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_bn          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,688</span> │ block5c_dwconv[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_activation  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block5c_bn[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_se_squeeze  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>)       │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block5c_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GlobalAveragePool…</span> │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_se_reshape  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block5c_se_squee… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Reshape</span>)           │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_se_reduce   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>)  │     <span style=\"color: #00af00; text-decoration-color: #00af00\">18,844</span> │ block5c_se_resha… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_se_expand   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>) │     <span style=\"color: #00af00; text-decoration-color: #00af00\">19,488</span> │ block5c_se_reduc… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_se_excite   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block5c_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Multiply</span>)          │ <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>)              │            │ block5c_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_project_co… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │     <span style=\"color: #00af00; text-decoration-color: #00af00\">75,264</span> │ block5c_se_excit… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_project_bn  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │        <span style=\"color: #00af00; text-decoration-color: #00af00\">448</span> │ block5c_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_drop        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block5c_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)           │ <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block5c_add (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</span>)   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block5c_drop[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>)              │            │ block5b_add[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>] │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_expand_conv │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │     <span style=\"color: #00af00; text-decoration-color: #00af00\">75,264</span> │ block5c_add[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>] │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_expand_bn   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,688</span> │ block6a_expand_c… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_expand_act… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6a_expand_b… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_dwconv_pad  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">71</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">123</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6a_expand_a… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">ZeroPadding2D</span>)     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_dwconv      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │     <span style=\"color: #00af00; text-decoration-color: #00af00\">16,800</span> │ block6a_dwconv_p… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">DepthwiseConv2D</span>)   │ <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_bn          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │      <span style=\"color: #00af00; text-decoration-color: #00af00\">2,688</span> │ block6a_dwconv[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_activation  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6a_bn[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_se_squeeze  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>)       │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6a_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GlobalAveragePool…</span> │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_se_reshape  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>) │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6a_se_squee… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Reshape</span>)           │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_se_reduce   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>)  │     <span style=\"color: #00af00; text-decoration-color: #00af00\">18,844</span> │ block6a_se_resha… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_se_expand   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>) │     <span style=\"color: #00af00; text-decoration-color: #00af00\">19,488</span> │ block6a_se_reduc… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_se_excite   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6a_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Multiply</span>)          │ <span style=\"color: #00af00; text-decoration-color: #00af00\">672</span>)              │            │ block6a_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_project_co… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │    <span style=\"color: #00af00; text-decoration-color: #00af00\">129,024</span> │ block6a_se_excit… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">192</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6a_project_bn  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │        <span style=\"color: #00af00; text-decoration-color: #00af00\">768</span> │ block6a_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">192</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_expand_conv │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │    <span style=\"color: #00af00; text-decoration-color: #00af00\">221,184</span> │ block6a_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_expand_bn   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │      <span style=\"color: #00af00; text-decoration-color: #00af00\">4,608</span> │ block6b_expand_c… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_expand_act… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6b_expand_b… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_dwconv      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │     <span style=\"color: #00af00; text-decoration-color: #00af00\">28,800</span> │ block6b_expand_a… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">DepthwiseConv2D</span>)   │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_bn          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │      <span style=\"color: #00af00; text-decoration-color: #00af00\">4,608</span> │ block6b_dwconv[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_activation  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6b_bn[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_se_squeeze  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6b_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GlobalAveragePool…</span> │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_se_reshape  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>,      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6b_se_squee… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Reshape</span>)           │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_se_reduce   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">48</span>)  │     <span style=\"color: #00af00; text-decoration-color: #00af00\">55,344</span> │ block6b_se_resha… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_se_expand   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>,      │     <span style=\"color: #00af00; text-decoration-color: #00af00\">56,448</span> │ block6b_se_reduc… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_se_excite   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6b_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Multiply</span>)          │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │ block6b_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_project_co… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │    <span style=\"color: #00af00; text-decoration-color: #00af00\">221,184</span> │ block6b_se_excit… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">192</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_project_bn  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │        <span style=\"color: #00af00; text-decoration-color: #00af00\">768</span> │ block6b_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">192</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_drop        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6b_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)           │ <span style=\"color: #00af00; text-decoration-color: #00af00\">192</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6b_add (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</span>)   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6b_drop[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">192</span>)              │            │ block6a_project_… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_expand_conv │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │    <span style=\"color: #00af00; text-decoration-color: #00af00\">221,184</span> │ block6b_add[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>] │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_expand_bn   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │      <span style=\"color: #00af00; text-decoration-color: #00af00\">4,608</span> │ block6c_expand_c… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_expand_act… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6c_expand_b… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_dwconv      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │     <span style=\"color: #00af00; text-decoration-color: #00af00\">28,800</span> │ block6c_expand_a… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">DepthwiseConv2D</span>)   │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_bn          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │      <span style=\"color: #00af00; text-decoration-color: #00af00\">4,608</span> │ block6c_dwconv[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_activation  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6c_bn[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_se_squeeze  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6c_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GlobalAveragePool…</span> │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_se_reshape  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>,      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6c_se_squee… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Reshape</span>)           │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_se_reduce   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">48</span>)  │     <span style=\"color: #00af00; text-decoration-color: #00af00\">55,344</span> │ block6c_se_resha… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_se_expand   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>,      │     <span style=\"color: #00af00; text-decoration-color: #00af00\">56,448</span> │ block6c_se_reduc… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_se_excite   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6c_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Multiply</span>)          │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │ block6c_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_project_co… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │    <span style=\"color: #00af00; text-decoration-color: #00af00\">221,184</span> │ block6c_se_excit… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">192</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_project_bn  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │        <span style=\"color: #00af00; text-decoration-color: #00af00\">768</span> │ block6c_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">192</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_drop        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6c_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)           │ <span style=\"color: #00af00; text-decoration-color: #00af00\">192</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6c_add (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</span>)   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6c_drop[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">192</span>)              │            │ block6b_add[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>] │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_expand_conv │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │    <span style=\"color: #00af00; text-decoration-color: #00af00\">221,184</span> │ block6c_add[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>] │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_expand_bn   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │      <span style=\"color: #00af00; text-decoration-color: #00af00\">4,608</span> │ block6d_expand_c… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_expand_act… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6d_expand_b… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_dwconv      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │     <span style=\"color: #00af00; text-decoration-color: #00af00\">28,800</span> │ block6d_expand_a… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">DepthwiseConv2D</span>)   │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_bn          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │      <span style=\"color: #00af00; text-decoration-color: #00af00\">4,608</span> │ block6d_dwconv[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_activation  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6d_bn[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_se_squeeze  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6d_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GlobalAveragePool…</span> │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_se_reshape  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>,      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6d_se_squee… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Reshape</span>)           │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_se_reduce   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">48</span>)  │     <span style=\"color: #00af00; text-decoration-color: #00af00\">55,344</span> │ block6d_se_resha… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_se_expand   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>,      │     <span style=\"color: #00af00; text-decoration-color: #00af00\">56,448</span> │ block6d_se_reduc… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_se_excite   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6d_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Multiply</span>)          │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │ block6d_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_project_co… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │    <span style=\"color: #00af00; text-decoration-color: #00af00\">221,184</span> │ block6d_se_excit… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">192</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_project_bn  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │        <span style=\"color: #00af00; text-decoration-color: #00af00\">768</span> │ block6d_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">192</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_drop        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6d_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)           │ <span style=\"color: #00af00; text-decoration-color: #00af00\">192</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block6d_add (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Add</span>)   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block6d_drop[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">192</span>)              │            │ block6c_add[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>] │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block7a_expand_conv │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │    <span style=\"color: #00af00; text-decoration-color: #00af00\">221,184</span> │ block6d_add[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>] │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block7a_expand_bn   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │      <span style=\"color: #00af00; text-decoration-color: #00af00\">4,608</span> │ block7a_expand_c… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block7a_expand_act… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block7a_expand_b… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block7a_dwconv      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │     <span style=\"color: #00af00; text-decoration-color: #00af00\">10,368</span> │ block7a_expand_a… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">DepthwiseConv2D</span>)   │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block7a_bn          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │      <span style=\"color: #00af00; text-decoration-color: #00af00\">4,608</span> │ block7a_dwconv[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block7a_activation  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block7a_bn[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]  │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block7a_se_squeeze  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block7a_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GlobalAveragePool…</span> │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block7a_se_reshape  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>,      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block7a_se_squee… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Reshape</span>)           │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block7a_se_reduce   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">48</span>)  │     <span style=\"color: #00af00; text-decoration-color: #00af00\">55,344</span> │ block7a_se_resha… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block7a_se_expand   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>,      │     <span style=\"color: #00af00; text-decoration-color: #00af00\">56,448</span> │ block7a_se_reduc… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block7a_se_excite   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ block7a_activati… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Multiply</span>)          │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1152</span>)             │            │ block7a_se_expan… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block7a_project_co… │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │    <span style=\"color: #00af00; text-decoration-color: #00af00\">368,640</span> │ block7a_se_excit… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)            │ <span style=\"color: #00af00; text-decoration-color: #00af00\">320</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ block7a_project_bn  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │      <span style=\"color: #00af00; text-decoration-color: #00af00\">1,280</span> │ block7a_project_… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">320</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ top_conv (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │    <span style=\"color: #00af00; text-decoration-color: #00af00\">409,600</span> │ block7a_project_… │\n│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1280</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ top_bn              │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │      <span style=\"color: #00af00; text-decoration-color: #00af00\">5,120</span> │ top_conv[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]    │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalizatio…</span> │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1280</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ top_activation      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">34</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">60</span>,    │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ top_bn[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]      │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Activation</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1280</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ up_sampling2d       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ top_activation[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">UpSampling2D</span>)      │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1280</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2d (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │  <span style=\"color: #00af00; text-decoration-color: #00af00\">5,898,752</span> │ up_sampling2d[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ concatenate         │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">68</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">120</span>,   │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ conv2d[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>],     │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Concatenate</span>)       │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1184</span>)             │            │ block6a_expand_a… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ up_sampling2d_1     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">136</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ concatenate[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>] │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">UpSampling2D</span>)      │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1184</span>)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2d_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">136</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>,  │  <span style=\"color: #00af00; text-decoration-color: #00af00\">2,728,192</span> │ up_sampling2d_1[<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ cropping2d          │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">135</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ conv2d_1[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]    │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Cropping2D</span>)        │ <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ concatenate_1       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">135</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">240</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ cropping2d[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>], │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Concatenate</span>)       │ <span style=\"color: #00af00; text-decoration-color: #00af00\">496</span>)              │            │ block4a_expand_a… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ up_sampling2d_2     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">270</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ concatenate_1[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">UpSampling2D</span>)      │ <span style=\"color: #00af00; text-decoration-color: #00af00\">496</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2d_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">270</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>,  │    <span style=\"color: #00af00; text-decoration-color: #00af00\">571,520</span> │ up_sampling2d_2[<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ concatenate_2       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">270</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">480</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ conv2d_2[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>],   │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Concatenate</span>)       │ <span style=\"color: #00af00; text-decoration-color: #00af00\">272</span>)              │            │ block3a_expand_a… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ up_sampling2d_3     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">540</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">960</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ concatenate_2[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">UpSampling2D</span>)      │ <span style=\"color: #00af00; text-decoration-color: #00af00\">272</span>)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2d_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">540</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">960</span>,  │    <span style=\"color: #00af00; text-decoration-color: #00af00\">156,736</span> │ up_sampling2d_3[<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ concatenate_3       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">540</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">960</span>,  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ conv2d_3[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>],   │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Concatenate</span>)       │ <span style=\"color: #00af00; text-decoration-color: #00af00\">160</span>)              │            │ block2a_expand_a… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ up_sampling2d_4     │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1080</span>,      │          <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │ concatenate_3[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]… │\n│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">UpSampling2D</span>)      │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1920</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">160</span>)        │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2d_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1080</span>,      │     <span style=\"color: #00af00; text-decoration-color: #00af00\">46,112</span> │ up_sampling2d_4[<span style=\"color: #00af00; text-decoration-color: #00af00\">…</span> │\n│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1920</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)         │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2d_5 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1080</span>,      │         <span style=\"color: #00af00; text-decoration-color: #00af00\">33</span> │ conv2d_4[<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>][<span style=\"color: #00af00; text-decoration-color: #00af00\">0</span>]    │\n│                     │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1920</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>)          │            │                   │\n└─────────────────────┴───────────────────┴────────────┴───────────────────┘\n</pre>\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"\u001b[1m Total params: \u001b[0m\u001b[38;5;34m13,450,916\u001b[0m (51.31 MB)\n","text/html":"<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">13,450,916</span> (51.31 MB)\n</pre>\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m13,408,893\u001b[0m (51.15 MB)\n","text/html":"<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">13,408,893</span> (51.15 MB)\n</pre>\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m42,023\u001b[0m (164.16 KB)\n","text/html":"<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">42,023</span> (164.16 KB)\n</pre>\n"},"metadata":{}}]},{"cell_type":"markdown","source":"# Dice Loss and Focal Loss","metadata":{}},{"cell_type":"code","source":"def dice_loss(y_true, y_pred, smooth=1e-6):\n    y_true_f = tf.reshape(y_true, [-1])\n    y_pred_f = tf.reshape(y_pred, [-1])\n    intersection = tf.reduce_sum(y_true_f * y_pred_f)\n    return 1 - (2. * intersection + smooth) / (tf.reduce_sum(y_true_f) + tf.reduce_sum(y_pred_f) + smooth)","metadata":{"execution":{"iopub.status.busy":"2024-10-10T12:55:45.296464Z","iopub.execute_input":"2024-10-10T12:55:45.296784Z","iopub.status.idle":"2024-10-10T12:55:45.302483Z","shell.execute_reply.started":"2024-10-10T12:55:45.296751Z","shell.execute_reply":"2024-10-10T12:55:45.301562Z"},"trusted":true},"execution_count":19,"outputs":[]},{"cell_type":"code","source":"def focal_loss(y_true, y_pred, alpha=0.25, gamma=2.0):\n    y_true = tf.cast(y_true, tf.float32)\n    y_pred = tf.clip_by_value(y_pred, tf.keras.backend.epsilon(), 1 - tf.keras.backend.epsilon())\n    cross_entropy = -y_true * tf.math.log(y_pred)\n    weight = alpha * tf.math.pow(1 - y_pred, gamma)\n    loss = weight * cross_entropy\n    return tf.reduce_sum(loss, axis=-1)","metadata":{"execution":{"iopub.status.busy":"2024-10-10T12:55:45.303752Z","iopub.execute_input":"2024-10-10T12:55:45.304117Z","iopub.status.idle":"2024-10-10T12:55:45.313632Z","shell.execute_reply.started":"2024-10-10T12:55:45.304071Z","shell.execute_reply":"2024-10-10T12:55:45.312537Z"},"trusted":true},"execution_count":20,"outputs":[]},{"cell_type":"code","source":"def combined_dice_focal_loss(y_true, y_pred, dice_weight=0.5, focal_weight=0.5):\n    \"\"\"\n    Combine Dice loss and Focal loss into a single loss function.\n\n    y_true: Ground truth labels.\n    y_pred: Model predictions.\n    dice_weight: Weight for Dice loss.\n    focal_weight: Weight for Focal loss.\n\n    Returns:\n    - Combined loss.\n    \"\"\"\n    dice = dice_loss(y_true, y_pred)\n    focal = focal_loss(y_true, y_pred)\n    combined_loss = dice_weight * dice + focal_weight * focal\n    return tf.cast(combined_loss, tf.float32)","metadata":{"execution":{"iopub.status.busy":"2024-10-10T12:55:45.314948Z","iopub.execute_input":"2024-10-10T12:55:45.315349Z","iopub.status.idle":"2024-10-10T12:55:45.325129Z","shell.execute_reply.started":"2024-10-10T12:55:45.315287Z","shell.execute_reply":"2024-10-10T12:55:45.324237Z"},"trusted":true},"execution_count":21,"outputs":[]},{"cell_type":"markdown","source":"# Evaluation Metrics","metadata":{}},{"cell_type":"markdown","source":"**Dice Coefficients**","metadata":{}},{"cell_type":"code","source":"def dice_coefficient(y_true, y_pred, smooth=1e-6):\n    # Cast both y_true and y_pred to float32 to ensure they have the same data type\n    y_true_f = tf.cast(tf.reshape(y_true, [-1]), tf.float32)\n    y_pred_f = tf.cast(tf.reshape(y_pred, [-1]), tf.float32)\n    intersection = tf.reduce_sum(y_true_f * y_pred_f)\n    return (2. * intersection + smooth) / (tf.reduce_sum(y_true_f) + tf.reduce_sum(y_pred_f) + smooth)","metadata":{"execution":{"iopub.status.busy":"2024-10-10T12:55:45.326375Z","iopub.execute_input":"2024-10-10T12:55:45.326737Z","iopub.status.idle":"2024-10-10T12:55:45.343749Z","shell.execute_reply.started":"2024-10-10T12:55:45.326698Z","shell.execute_reply":"2024-10-10T12:55:45.342696Z"},"trusted":true},"execution_count":22,"outputs":[]},{"cell_type":"markdown","source":"**F score with beta parameter**","metadata":{}},{"cell_type":"code","source":"def precision(y_true, y_pred):\n    y_true = tf.cast(y_true, tf.float32)  # Ensure y_true is of type float32\n    y_pred = tf.cast(y_pred, tf.float32)  # Ensure y_pred is of type float32\n    true_positives = tf.reduce_sum(tf.cast(y_true * y_pred, tf.float32))\n    predicted_positives = tf.reduce_sum(tf.cast(y_pred, tf.float32))\n    precision = true_positives / (predicted_positives + tf.keras.backend.epsilon())\n    return precision\n\ndef recall(y_true, y_pred):\n    y_true = tf.cast(y_true, tf.float32)  # Ensure y_true is of type float32\n    y_pred = tf.cast(y_pred, tf.float32)  # Ensure y_pred is of type float32\n    true_positives = tf.reduce_sum(tf.cast(y_true * y_pred, tf.float32))\n    possible_positives = tf.reduce_sum(tf.cast(y_true, tf.float32))\n    recall = true_positives / (possible_positives + tf.keras.backend.epsilon())\n    return recall\n\ndef f_score(y_true, y_pred, beta=1):\n    y_true = tf.cast(y_true, tf.float32)  # Ensure y_true is of type float32\n    y_pred = tf.cast(y_pred, tf.float32)  # Ensure y_pred is of type float32\n    prec = precision(y_true, y_pred)\n    rec = recall(y_true, y_pred)\n    beta_squared = beta ** 2\n    return (1 + beta_squared) * (prec * rec) / (beta_squared * prec + rec + tf.keras.backend.epsilon())","metadata":{"execution":{"iopub.status.busy":"2024-10-10T12:55:45.345049Z","iopub.execute_input":"2024-10-10T12:55:45.345339Z","iopub.status.idle":"2024-10-10T12:55:45.355777Z","shell.execute_reply.started":"2024-10-10T12:55:45.345307Z","shell.execute_reply":"2024-10-10T12:55:45.35472Z"},"trusted":true},"execution_count":23,"outputs":[]},{"cell_type":"markdown","source":"**Mean Dice Coefficient with F-Score**","metadata":{}},{"cell_type":"code","source":"def mean_dice_fscore(y_true, y_pred, beta=1):\n    dice = dice_coefficient(y_true, y_pred)\n    fscore = f_score(y_true, y_pred, beta=beta)\n    mean_metric = (dice + fscore) / 2.0  # Averaging Dice Coefficient and F-Score\n    return tf.cast(mean_metric, tf.float32) ","metadata":{"execution":{"iopub.status.busy":"2024-10-10T12:55:45.357357Z","iopub.execute_input":"2024-10-10T12:55:45.357727Z","iopub.status.idle":"2024-10-10T12:55:45.369526Z","shell.execute_reply.started":"2024-10-10T12:55:45.357685Z","shell.execute_reply":"2024-10-10T12:55:45.368539Z"},"trusted":true},"execution_count":24,"outputs":[]},{"cell_type":"markdown","source":"# Compiling and Training the Model","metadata":{}},{"cell_type":"code","source":"# model.compile(optimizer='adam',\n#               loss=[dice_loss, focal_loss],  # You can combine both losses or choose one\n#               metrics=[mean_dice_fscore])    # Use the Mean Dice F-Score as the evaluation metric","metadata":{"execution":{"iopub.status.busy":"2024-10-10T12:55:45.37081Z","iopub.execute_input":"2024-10-10T12:55:45.371161Z","iopub.status.idle":"2024-10-10T12:55:45.379513Z","shell.execute_reply.started":"2024-10-10T12:55:45.371127Z","shell.execute_reply":"2024-10-10T12:55:45.378555Z"},"trusted":true},"execution_count":25,"outputs":[]},{"cell_type":"code","source":"# Compile the model with the combined loss\nmodel.compile(optimizer='adam', loss=[combined_dice_focal_loss], metrics=[mean_dice_fscore])","metadata":{"execution":{"iopub.status.busy":"2024-10-10T12:55:45.380894Z","iopub.execute_input":"2024-10-10T12:55:45.381326Z","iopub.status.idle":"2024-10-10T12:55:45.399063Z","shell.execute_reply.started":"2024-10-10T12:55:45.38128Z","shell.execute_reply":"2024-10-10T12:55:45.397971Z"},"trusted":true},"execution_count":26,"outputs":[]},{"cell_type":"code","source":"# Set the batch size and image dimensions\nbatch_size = 4\nimg_size = (1080, 1920)  # Resized image dimensions\nnum_classes = 34  # Number of segmentation classes\n\n# Define the training and validation generators\ntrain_gen = data_generator(train_files, label_train_files, \n                           batch_size=batch_size, \n                           img_size=img_size, \n                           num_classes=num_classes)\n\nval_gen = data_generator(val_files, label_val_files, \n                         batch_size=batch_size, \n                         img_size=img_size, \n                         num_classes=num_classes)\n\n","metadata":{"execution":{"iopub.status.busy":"2024-10-10T12:55:45.400376Z","iopub.execute_input":"2024-10-10T12:55:45.400696Z","iopub.status.idle":"2024-10-10T12:55:45.406869Z","shell.execute_reply.started":"2024-10-10T12:55:45.400662Z","shell.execute_reply":"2024-10-10T12:55:45.405897Z"},"trusted":true},"execution_count":27,"outputs":[]},{"cell_type":"code","source":"# Train the model\nhistory = model.fit(train_gen, \n                    steps_per_epoch=len(train_files) // batch_size, \n                    validation_data=val_gen, \n                    validation_steps=len(val_files) // batch_size, \n                    epochs=20)","metadata":{"execution":{"iopub.status.busy":"2024-10-10T12:55:45.408279Z","iopub.execute_input":"2024-10-10T12:55:45.408971Z","iopub.status.idle":"2024-10-10T13:01:11.123767Z","shell.execute_reply.started":"2024-10-10T12:55:45.408896Z","shell.execute_reply":"2024-10-10T13:01:11.122247Z"},"trusted":true},"execution_count":28,"outputs":[{"name":"stdout","text":"Epoch 1/20\n","output_type":"stream"},{"name":"stderr","text":"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\nI0000 00:00:1728565006.919830     102 service.cc:145] XLA service 0x79bc54002880 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\nI0000 00:00:1728565006.919887     102 service.cc:153]   StreamExecutor device (0): Tesla T4, Compute Capability 7.5\nI0000 00:00:1728565006.919891     102 service.cc:153]   StreamExecutor device (1): Tesla T4, Compute Capability 7.5\n2024-10-10 12:57:40.027209: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng11{k2=3,k3=0} for conv (f32[4,256,136,240]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,1184,136,240]{3,2,1,0}, f32[256,1184,3,3]{3,2,1,0}, f32[256]{0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBiasActivationForward\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kRelu\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:57:40.146729: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.119635885s\nTrying algorithm eng11{k2=3,k3=0} for conv (f32[4,256,136,240]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,1184,136,240]{3,2,1,0}, f32[256,1184,3,3]{3,2,1,0}, f32[256]{0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBiasActivationForward\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kRelu\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:57:41.146994: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng13{} for conv (f32[4,256,136,240]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,1184,136,240]{3,2,1,0}, f32[256,1184,3,3]{3,2,1,0}, f32[256]{0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBiasActivationForward\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kRelu\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:57:41.244269: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.097448655s\nTrying algorithm eng13{} for conv (f32[4,256,136,240]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,1184,136,240]{3,2,1,0}, f32[256,1184,3,3]{3,2,1,0}, f32[256]{0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBiasActivationForward\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kRelu\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:57:54.733121: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng11{k2=2,k3=0} for conv (f32[4,64,540,960]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,272,540,960]{3,2,1,0}, f32[64,272,3,3]{3,2,1,0}, f32[64]{0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBiasActivationForward\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kRelu\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:57:54.888464: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.155459912s\nTrying algorithm eng11{k2=2,k3=0} for conv (f32[4,64,540,960]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,272,540,960]{3,2,1,0}, f32[64,272,3,3]{3,2,1,0}, f32[64]{0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBiasActivationForward\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kRelu\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:57:56.837605: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng13{} for conv (f32[4,64,540,960]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,272,540,960]{3,2,1,0}, f32[64,272,3,3]{3,2,1,0}, f32[64]{0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBiasActivationForward\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kRelu\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:57:56.871638: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.034136402s\nTrying algorithm eng13{} for conv (f32[4,64,540,960]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,272,540,960]{3,2,1,0}, f32[64,272,3,3]{3,2,1,0}, f32[64]{0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBiasActivationForward\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kRelu\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:01.835148: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng20{k2=0,k4=1,k5=1,k6=0,k7=0,k19=0} for conv (f32[4,32,1080,1920]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,160,1080,1920]{3,2,1,0}, f32[32,160,3,3]{3,2,1,0}, f32[32]{0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBiasActivationForward\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kRelu\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:02.373251: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.538210699s\nTrying algorithm eng20{k2=0,k4=1,k5=1,k6=0,k7=0,k19=0} for conv (f32[4,32,1080,1920]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,160,1080,1920]{3,2,1,0}, f32[32,160,3,3]{3,2,1,0}, f32[32]{0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBiasActivationForward\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kRelu\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:06.892518: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng11{k2=2,k3=0} for conv (f32[4,32,1080,1920]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,160,1080,1920]{3,2,1,0}, f32[32,160,3,3]{3,2,1,0}, f32[32]{0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBiasActivationForward\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kRelu\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:08.517546: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 2.625203388s\nTrying algorithm eng11{k2=2,k3=0} for conv (f32[4,32,1080,1920]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,160,1080,1920]{3,2,1,0}, f32[32,160,3,3]{3,2,1,0}, f32[32]{0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBiasActivationForward\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kRelu\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:09.517752: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng11{k2=3,k3=0} for conv (f32[4,32,1080,1920]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,160,1080,1920]{3,2,1,0}, f32[32,160,3,3]{3,2,1,0}, f32[32]{0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBiasActivationForward\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kRelu\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:09.724267: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.206610189s\nTrying algorithm eng11{k2=3,k3=0} for conv (f32[4,32,1080,1920]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,160,1080,1920]{3,2,1,0}, f32[32,160,3,3]{3,2,1,0}, f32[32]{0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBiasActivationForward\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kRelu\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:10.724535: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng13{} for conv (f32[4,32,1080,1920]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,160,1080,1920]{3,2,1,0}, f32[32,160,3,3]{3,2,1,0}, f32[32]{0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBiasActivationForward\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kRelu\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:11.187028: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.462664876s\nTrying algorithm eng13{} for conv (f32[4,32,1080,1920]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,160,1080,1920]{3,2,1,0}, f32[32,160,3,3]{3,2,1,0}, f32[32]{0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBiasActivationForward\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kRelu\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:12.187256: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng0{} for conv (f32[4,32,1080,1920]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,160,1080,1920]{3,2,1,0}, f32[32,160,3,3]{3,2,1,0}, f32[32]{0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBiasActivationForward\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kRelu\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:22.101117: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 10.913957297s\nTrying algorithm eng0{} for conv (f32[4,32,1080,1920]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,160,1080,1920]{3,2,1,0}, f32[32,160,3,3]{3,2,1,0}, f32[32]{0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBiasActivationForward\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kRelu\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:27.753125: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng25{k2=2,k3=0} for conv (f32[4,160,1080,1920]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,32,1080,1920]{3,2,1,0}, f32[32,160,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:28.738375: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.985337309s\nTrying algorithm eng25{k2=2,k3=0} for conv (f32[4,160,1080,1920]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,32,1080,1920]{3,2,1,0}, f32[32,160,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:29.738697: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng25{k2=1,k3=0} for conv (f32[4,160,1080,1920]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,32,1080,1920]{3,2,1,0}, f32[32,160,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:30.611431: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.872926654s\nTrying algorithm eng25{k2=1,k3=0} for conv (f32[4,160,1080,1920]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,32,1080,1920]{3,2,1,0}, f32[32,160,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:31.611646: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng25{k2=0,k3=0} for conv (f32[4,160,1080,1920]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,32,1080,1920]{3,2,1,0}, f32[32,160,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:32.415735: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.804182645s\nTrying algorithm eng25{k2=0,k3=0} for conv (f32[4,160,1080,1920]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,32,1080,1920]{3,2,1,0}, f32[32,160,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:34.213165: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng5{} for conv (f32[4,160,1080,1920]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,32,1080,1920]{3,2,1,0}, f32[32,160,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:35.188036: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.975007384s\nTrying algorithm eng5{} for conv (f32[4,160,1080,1920]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,32,1080,1920]{3,2,1,0}, f32[32,160,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:36.188353: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng50{} for conv (f32[4,160,1080,1920]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,32,1080,1920]{3,2,1,0}, f32[32,160,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:37.169082: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.980902423s\nTrying algorithm eng50{} for conv (f32[4,160,1080,1920]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,32,1080,1920]{3,2,1,0}, f32[32,160,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:41.927529: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.000018422s\nTrying algorithm eng25{k2=1,k3=0} for conv (f32[4,272,540,960]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,64,540,960]{3,2,1,0}, f32[64,272,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:42.927812: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng25{k2=2,k3=0} for conv (f32[4,272,540,960]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,64,540,960]{3,2,1,0}, f32[64,272,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:43.056986: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.129355672s\nTrying algorithm eng25{k2=2,k3=0} for conv (f32[4,272,540,960]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,64,540,960]{3,2,1,0}, f32[64,272,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:44.057279: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng5{} for conv (f32[4,272,540,960]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,64,540,960]{3,2,1,0}, f32[64,272,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:44.447419: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.39033469s\nTrying algorithm eng5{} for conv (f32[4,272,540,960]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,64,540,960]{3,2,1,0}, f32[64,272,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:45.447721: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng50{} for conv (f32[4,272,540,960]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,64,540,960]{3,2,1,0}, f32[64,272,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:45.834627: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.387089951s\nTrying algorithm eng50{} for conv (f32[4,272,540,960]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,64,540,960]{3,2,1,0}, f32[64,272,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:51.084103: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng5{} for conv (f32[4,496,270,480]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,128,270,480]{3,2,1,0}, f32[128,496,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:51.118529: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.034516183s\nTrying algorithm eng5{} for conv (f32[4,496,270,480]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,128,270,480]{3,2,1,0}, f32[128,496,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:52.118749: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng50{} for conv (f32[4,496,270,480]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,128,270,480]{3,2,1,0}, f32[128,496,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:52.153470: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.034822914s\nTrying algorithm eng50{} for conv (f32[4,496,270,480]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,128,270,480]{3,2,1,0}, f32[128,496,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:56.965842: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng25{k2=2,k3=0} for conv (f32[4,1184,136,240]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,256,136,240]{3,2,1,0}, f32[256,1184,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:57.254336: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.288604328s\nTrying algorithm eng25{k2=2,k3=0} for conv (f32[4,1184,136,240]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,256,136,240]{3,2,1,0}, f32[256,1184,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:58.872297: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng5{} for conv (f32[4,1184,136,240]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,256,136,240]{3,2,1,0}, f32[256,1184,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:58:59.149457: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.277357025s\nTrying algorithm eng5{} for conv (f32[4,1184,136,240]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,256,136,240]{3,2,1,0}, f32[256,1184,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:00.149813: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng50{} for conv (f32[4,1184,136,240]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,256,136,240]{3,2,1,0}, f32[256,1184,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:00.427804: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.278227448s\nTrying algorithm eng50{} for conv (f32[4,1184,136,240]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,256,136,240]{3,2,1,0}, f32[256,1184,3,3]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardInput\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:11.729154: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng4{} for conv (f32[4,144,273,483]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,144,277,487]{3,2,1,0}, f32[144,1,5,5]{3,2,1,0}), window={size=5x5}, dim_labels=bf01_oi01->bf01, feature_group_count=144, custom_call_target=\"__cudnn$convForward\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:12.028037: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.299057922s\nTrying algorithm eng4{} for conv (f32[4,144,273,483]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,144,277,487]{3,2,1,0}, f32[144,1,5,5]{3,2,1,0}), window={size=5x5}, dim_labels=bf01_oi01->bf01, feature_group_count=144, custom_call_target=\"__cudnn$convForward\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:15.147017: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng4{} for conv (f32[4,96,541,961]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,96,543,963]{3,2,1,0}, f32[96,1,3,3]{3,2,1,0}), window={size=3x3}, dim_labels=bf01_oi01->bf01, feature_group_count=96, custom_call_target=\"__cudnn$convForward\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:17.147115: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 3.000240019s\nTrying algorithm eng4{} for conv (f32[4,96,541,961]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,96,543,963]{3,2,1,0}, f32[96,1,3,3]{3,2,1,0}), window={size=3x3}, dim_labels=bf01_oi01->bf01, feature_group_count=96, custom_call_target=\"__cudnn$convForward\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:38.823830: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng21{k2=2,k4=3,k5=0,k6=0,k7=0} for conv (f32[256,1184,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,1184,136,240]{3,2,1,0}, f32[4,256,136,240]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:38.854611: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.030896776s\nTrying algorithm eng21{k2=2,k4=3,k5=0,k6=0,k7=0} for conv (f32[256,1184,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,1184,136,240]{3,2,1,0}, f32[4,256,136,240]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:39.854852: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng1{k2=8,k3=0} for conv (f32[256,1184,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,1184,136,240]{3,2,1,0}, f32[4,256,136,240]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:40.179293: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.324565122s\nTrying algorithm eng1{k2=8,k3=0} for conv (f32[256,1184,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,1184,136,240]{3,2,1,0}, f32[4,256,136,240]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:41.179511: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng1{k2=6,k3=0} for conv (f32[256,1184,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,1184,136,240]{3,2,1,0}, f32[4,256,136,240]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:41.633682: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.454275073s\nTrying algorithm eng1{k2=6,k3=0} for conv (f32[256,1184,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,1184,136,240]{3,2,1,0}, f32[4,256,136,240]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:44.710685: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng20{k2=8,k3=0} for conv (f32[128,496,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,496,270,480]{3,2,1,0}, f32[4,128,270,480]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:44.788440: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.07800428s\nTrying algorithm eng20{k2=8,k3=0} for conv (f32[128,496,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,496,270,480]{3,2,1,0}, f32[4,128,270,480]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:46.522934: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng20{k2=6,k3=0} for conv (f32[128,496,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,496,270,480]{3,2,1,0}, f32[4,128,270,480]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:46.756663: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.233943783s\nTrying algorithm eng20{k2=6,k3=0} for conv (f32[128,496,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,496,270,480]{3,2,1,0}, f32[4,128,270,480]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:47.757012: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng1{k2=8,k3=0} for conv (f32[128,496,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,496,270,480]{3,2,1,0}, f32[4,128,270,480]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:47.918678: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.161903652s\nTrying algorithm eng1{k2=8,k3=0} for conv (f32[128,496,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,496,270,480]{3,2,1,0}, f32[4,128,270,480]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:48.918986: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng1{k2=6,k3=0} for conv (f32[128,496,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,496,270,480]{3,2,1,0}, f32[4,128,270,480]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:49.238070: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.31927209s\nTrying algorithm eng1{k2=6,k3=0} for conv (f32[128,496,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,496,270,480]{3,2,1,0}, f32[4,128,270,480]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:53.884429: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng20{k2=8,k3=0} for conv (f32[64,272,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,272,540,960]{3,2,1,0}, f32[4,64,540,960]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:54.072686: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.188440447s\nTrying algorithm eng20{k2=8,k3=0} for conv (f32[64,272,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,272,540,960]{3,2,1,0}, f32[4,64,540,960]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:55.072945: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng20{k2=0,k3=0} for conv (f32[64,272,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,272,540,960]{3,2,1,0}, f32[4,64,540,960]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:55.419412: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.346598443s\nTrying algorithm eng20{k2=0,k3=0} for conv (f32[64,272,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,272,540,960]{3,2,1,0}, f32[4,64,540,960]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:56.419641: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng20{k2=6,k3=0} for conv (f32[64,272,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,272,540,960]{3,2,1,0}, f32[4,64,540,960]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:56.760607: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.341080667s\nTrying algorithm eng20{k2=6,k3=0} for conv (f32[64,272,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,272,540,960]{3,2,1,0}, f32[4,64,540,960]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:57.760837: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng1{k2=8,k3=0} for conv (f32[64,272,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,272,540,960]{3,2,1,0}, f32[4,64,540,960]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:58.077617: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.316889564s\nTrying algorithm eng1{k2=8,k3=0} for conv (f32[64,272,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,272,540,960]{3,2,1,0}, f32[4,64,540,960]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:59.077976: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng1{k2=6,k3=0} for conv (f32[64,272,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,272,540,960]{3,2,1,0}, f32[4,64,540,960]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 12:59:59.525371: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.447629289s\nTrying algorithm eng1{k2=6,k3=0} for conv (f32[64,272,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,272,540,960]{3,2,1,0}, f32[4,64,540,960]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 13:00:00.525691: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng1{k2=0,k3=0} for conv (f32[64,272,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,272,540,960]{3,2,1,0}, f32[4,64,540,960]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 13:00:00.918959: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.393456647s\nTrying algorithm eng1{k2=0,k3=0} for conv (f32[64,272,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,272,540,960]{3,2,1,0}, f32[4,64,540,960]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 13:00:01.919239: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng20{k2=2,k3=0} for conv (f32[64,272,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,272,540,960]{3,2,1,0}, f32[4,64,540,960]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 13:00:02.679084: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.760001814s\nTrying algorithm eng20{k2=2,k3=0} for conv (f32[64,272,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,272,540,960]{3,2,1,0}, f32[4,64,540,960]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 13:00:03.679386: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng1{k2=2,k3=0} for conv (f32[64,272,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,272,540,960]{3,2,1,0}, f32[4,64,540,960]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 13:00:04.527504: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.848282216s\nTrying algorithm eng1{k2=2,k3=0} for conv (f32[64,272,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,272,540,960]{3,2,1,0}, f32[4,64,540,960]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 13:00:07.126433: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng20{k2=8,k3=0} for conv (f32[32,160,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,160,1080,1920]{3,2,1,0}, f32[4,32,1080,1920]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 13:00:07.568894: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.442569111s\nTrying algorithm eng20{k2=8,k3=0} for conv (f32[32,160,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,160,1080,1920]{3,2,1,0}, f32[4,32,1080,1920]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 13:00:08.569221: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng20{k2=6,k3=0} for conv (f32[32,160,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,160,1080,1920]{3,2,1,0}, f32[4,32,1080,1920]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 13:00:09.203175: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.634159022s\nTrying algorithm eng20{k2=6,k3=0} for conv (f32[32,160,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,160,1080,1920]{3,2,1,0}, f32[4,32,1080,1920]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 13:00:10.203522: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng1{k2=8,k3=0} for conv (f32[32,160,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,160,1080,1920]{3,2,1,0}, f32[4,32,1080,1920]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 13:00:10.869516: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.666210817s\nTrying algorithm eng1{k2=8,k3=0} for conv (f32[32,160,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,160,1080,1920]{3,2,1,0}, f32[4,32,1080,1920]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 13:00:11.869850: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng1{k2=6,k3=0} for conv (f32[32,160,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,160,1080,1920]{3,2,1,0}, f32[4,32,1080,1920]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 13:00:12.598546: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 1.728908032s\nTrying algorithm eng1{k2=6,k3=0} for conv (f32[32,160,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,160,1080,1920]{3,2,1,0}, f32[4,32,1080,1920]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 13:00:13.598893: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng20{k2=0,k3=0} for conv (f32[32,160,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,160,1080,1920]{3,2,1,0}, f32[4,32,1080,1920]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 13:00:15.535816: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 2.937143268s\nTrying algorithm eng20{k2=0,k3=0} for conv (f32[32,160,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,160,1080,1920]{3,2,1,0}, f32[4,32,1080,1920]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 13:00:16.536139: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng1{k2=0,k3=0} for conv (f32[32,160,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,160,1080,1920]{3,2,1,0}, f32[4,32,1080,1920]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 13:00:18.620498: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 3.084537804s\nTrying algorithm eng1{k2=0,k3=0} for conv (f32[32,160,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,160,1080,1920]{3,2,1,0}, f32[4,32,1080,1920]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 13:00:19.620744: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng20{k2=2,k3=0} for conv (f32[32,160,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,160,1080,1920]{3,2,1,0}, f32[4,32,1080,1920]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 13:00:22.563770: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 3.943126548s\nTrying algorithm eng20{k2=2,k3=0} for conv (f32[32,160,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,160,1080,1920]{3,2,1,0}, f32[4,32,1080,1920]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 13:00:23.564105: E external/local_xla/xla/service/slow_operation_alarm.cc:65] Trying algorithm eng1{k2=2,k3=0} for conv (f32[32,160,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,160,1080,1920]{3,2,1,0}, f32[4,32,1080,1920]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\n2024-10-10 13:00:26.716936: E external/local_xla/xla/service/slow_operation_alarm.cc:133] The operation took 4.153024914s\nTrying algorithm eng1{k2=2,k3=0} for conv (f32[32,160,3,3]{3,2,1,0}, u8[0]{0}) custom-call(f32[4,160,1080,1920]{3,2,1,0}, f32[4,32,1080,1920]{3,2,1,0}), window={size=3x3 pad=1_1x1_1}, dim_labels=bf01_oi01->bf01, custom_call_target=\"__cudnn$convBackwardFilter\", backend_config={\"operation_queue_id\":\"0\",\"wait_on_operation_queues\":[],\"cudnn_conv_backend_config\":{\"conv_result_scale\":1,\"activation_mode\":\"kNone\",\"side_input_scale\":0,\"leakyrelu_alpha\":0}} is taking a while...\nI0000 00:00:1728565260.244065     102 device_compiler.h:188] Compiled cluster using XLA!  This line is logged at most once for the lifetime of the process.\n","output_type":"stream"},{"traceback":["\u001b[0;31m---------------------------------------------------------------------------\u001b[0m","\u001b[0;31mResourceExhaustedError\u001b[0m                    Traceback (most recent call last)","Cell \u001b[0;32mIn[28], line 2\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[38;5;66;03m# Train the model\u001b[39;00m\n\u001b[0;32m----> 2\u001b[0m history \u001b[38;5;241m=\u001b[39m \u001b[43mmodel\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfit\u001b[49m\u001b[43m(\u001b[49m\u001b[43mtrain_gen\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\n\u001b[1;32m      3\u001b[0m \u001b[43m                    \u001b[49m\u001b[43msteps_per_epoch\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43mlen\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43mtrain_files\u001b[49m\u001b[43m)\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m/\u001b[39;49m\u001b[38;5;241;43m/\u001b[39;49m\u001b[43m \u001b[49m\u001b[43mbatch_size\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\n\u001b[1;32m      4\u001b[0m \u001b[43m                    \u001b[49m\u001b[43mvalidation_data\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mval_gen\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\n\u001b[1;32m      5\u001b[0m \u001b[43m                    \u001b[49m\u001b[43mvalidation_steps\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43mlen\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43mval_files\u001b[49m\u001b[43m)\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m/\u001b[39;49m\u001b[38;5;241;43m/\u001b[39;49m\u001b[43m \u001b[49m\u001b[43mbatch_size\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\n\u001b[1;32m      6\u001b[0m \u001b[43m                    \u001b[49m\u001b[43mepochs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m20\u001b[39;49m\u001b[43m)\u001b[49m\n","File \u001b[0;32m/opt/conda/lib/python3.10/site-packages/keras/src/utils/traceback_utils.py:122\u001b[0m, in \u001b[0;36mfilter_traceback.<locals>.error_handler\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m    119\u001b[0m     filtered_tb \u001b[38;5;241m=\u001b[39m _process_traceback_frames(e\u001b[38;5;241m.\u001b[39m__traceback__)\n\u001b[1;32m    120\u001b[0m     \u001b[38;5;66;03m# To get the full stack trace, call:\u001b[39;00m\n\u001b[1;32m    121\u001b[0m     \u001b[38;5;66;03m# `keras.config.disable_traceback_filtering()`\u001b[39;00m\n\u001b[0;32m--> 122\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m e\u001b[38;5;241m.\u001b[39mwith_traceback(filtered_tb) \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[1;32m    123\u001b[0m \u001b[38;5;28;01mfinally\u001b[39;00m:\n\u001b[1;32m    124\u001b[0m     \u001b[38;5;28;01mdel\u001b[39;00m filtered_tb\n","File \u001b[0;32m/opt/conda/lib/python3.10/site-packages/tensorflow/python/eager/execute.py:53\u001b[0m, in \u001b[0;36mquick_execute\u001b[0;34m(op_name, num_outputs, inputs, attrs, ctx, name)\u001b[0m\n\u001b[1;32m     51\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[1;32m     52\u001b[0m   ctx\u001b[38;5;241m.\u001b[39mensure_initialized()\n\u001b[0;32m---> 53\u001b[0m   tensors \u001b[38;5;241m=\u001b[39m pywrap_tfe\u001b[38;5;241m.\u001b[39mTFE_Py_Execute(ctx\u001b[38;5;241m.\u001b[39m_handle, device_name, op_name,\n\u001b[1;32m     54\u001b[0m                                       inputs, attrs, num_outputs)\n\u001b[1;32m     55\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m core\u001b[38;5;241m.\u001b[39m_NotOkStatusException \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[1;32m     56\u001b[0m   \u001b[38;5;28;01mif\u001b[39;00m name \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n","\u001b[0;31mResourceExhaustedError\u001b[0m: Graph execution error:\n\nDetected at node StatefulPartitionedCall defined at (most recent call last):\n  File \"/opt/conda/lib/python3.10/runpy.py\", line 196, in _run_module_as_main\n\n  File \"/opt/conda/lib/python3.10/runpy.py\", line 86, in _run_code\n\n  File \"/opt/conda/lib/python3.10/site-packages/ipykernel_launcher.py\", line 18, in <module>\n\n  File \"/opt/conda/lib/python3.10/site-packages/traitlets/config/application.py\", line 1075, in launch_instance\n\n  File \"/opt/conda/lib/python3.10/site-packages/ipykernel/kernelapp.py\", line 739, in start\n\n  File \"/opt/conda/lib/python3.10/site-packages/tornado/platform/asyncio.py\", line 205, in start\n\n  File \"/opt/conda/lib/python3.10/asyncio/base_events.py\", line 603, in run_forever\n\n  File \"/opt/conda/lib/python3.10/asyncio/base_events.py\", line 1909, in _run_once\n\n  File \"/opt/conda/lib/python3.10/asyncio/events.py\", line 80, in _run\n\n  File \"/opt/conda/lib/python3.10/site-packages/ipykernel/kernelbase.py\", line 545, in dispatch_queue\n\n  File \"/opt/conda/lib/python3.10/site-packages/ipykernel/kernelbase.py\", line 534, in process_one\n\n  File \"/opt/conda/lib/python3.10/site-packages/ipykernel/kernelbase.py\", line 437, in dispatch_shell\n\n  File \"/opt/conda/lib/python3.10/site-packages/ipykernel/ipkernel.py\", line 362, in execute_request\n\n  File \"/opt/conda/lib/python3.10/site-packages/ipykernel/kernelbase.py\", line 778, in execute_request\n\n  File \"/opt/conda/lib/python3.10/site-packages/ipykernel/ipkernel.py\", line 449, in do_execute\n\n  File \"/opt/conda/lib/python3.10/site-packages/ipykernel/zmqshell.py\", line 549, in run_cell\n\n  File \"/opt/conda/lib/python3.10/site-packages/IPython/core/interactiveshell.py\", line 3051, in run_cell\n\n  File \"/opt/conda/lib/python3.10/site-packages/IPython/core/interactiveshell.py\", line 3106, in _run_cell\n\n  File \"/opt/conda/lib/python3.10/site-packages/IPython/core/async_helpers.py\", line 129, in _pseudo_sync_runner\n\n  File \"/opt/conda/lib/python3.10/site-packages/IPython/core/interactiveshell.py\", line 3311, in run_cell_async\n\n  File \"/opt/conda/lib/python3.10/site-packages/IPython/core/interactiveshell.py\", line 3493, in run_ast_nodes\n\n  File \"/opt/conda/lib/python3.10/site-packages/IPython/core/interactiveshell.py\", line 3553, in run_code\n\n  File \"/tmp/ipykernel_30/1019388919.py\", line 2, in <module>\n\n  File \"/opt/conda/lib/python3.10/site-packages/keras/src/utils/traceback_utils.py\", line 117, in error_handler\n\n  File \"/opt/conda/lib/python3.10/site-packages/keras/src/backend/tensorflow/trainer.py\", line 314, in fit\n\n  File \"/opt/conda/lib/python3.10/site-packages/keras/src/backend/tensorflow/trainer.py\", line 117, in one_step_on_iterator\n\nOut of memory while trying to allocate 16523089264 bytes.\nBufferAssignment OOM Debugging.\nBufferAssignment stats:\n             parameter allocation:  351.37MiB\n              constant allocation:       316B\n        maybe_live_out allocation:  153.62MiB\n     preallocated temp allocation:   15.39GiB\n  preallocated temp fragmentation:     6.6KiB (0.00%)\n                 total allocation:   15.73GiB\n              total fragmentation:   690.0KiB (0.00%)\nPeak buffers:\n\tBuffer 1:\n\t\tSize: 5.84GiB\n\t\tOperator: op_type=\"Conv2DBackpropFilter\" op_name=\"gradient_tape/functional_1_1/conv2d_3_1/convolution/Conv2DBackpropFilter\" source_file=\"/opt/conda/lib/python3.10/site-packages/tensorflow/python/framework/ops.py\" source_line=1177\n\t\tXLA Label: custom-call\n\t\tShape: u8[6273073152]\n\t\t==========================\n\n\tBuffer 2:\n\t\tSize: 2.10GiB\n\t\tXLA Label: fusion\n\t\tShape: f32[4,272,540,960]\n\t\t==========================\n\n\tBuffer 3:\n\t\tSize: 1.24GiB\n\t\tOperator: op_type=\"Sum\" op_name=\"gradient_tape/functional_1_1/up_sampling2d_4_1/Repeat/Sum\" source_file=\"/opt/conda/lib/python3.10/site-packages/tensorflow/python/framework/ops.py\" source_line=1177\n\t\tXLA Label: fusion\n\t\tShape: f32[1,4,160,540,960]\n\t\t==========================\n\n\tBuffer 4:\n\t\tSize: 1012.50MiB\n\t\tOperator: op_type=\"Sum\" op_name=\"gradient_tape/functional_1_1/conv2d_4_1/add/Sum\" source_file=\"/opt/conda/lib/python3.10/site-packages/tensorflow/python/framework/ops.py\" source_line=1177\n\t\tXLA Label: fusion\n\t\tShape: f32[4,32,1080,1920]\n\t\t==========================\n\n\tBuffer 5:\n\t\tSize: 759.38MiB\n\t\tOperator: op_type=\"Conv2D\" op_name=\"functional_1_1/block2a_expand_conv_1/convolution\" source_file=\"/opt/conda/lib/python3.10/site-packages/tensorflow/python/framework/ops.py\" source_line=1177\n\t\tXLA Label: custom-call\n\t\tShape: f32[4,96,540,960]\n\t\t==========================\n\n\tBuffer 6:\n\t\tSize: 506.25MiB\n\t\tOperator: op_type=\"Sum\" op_name=\"gradient_tape/functional_1_1/conv2d_3_1/add/Sum\" source_file=\"/opt/conda/lib/python3.10/site-packages/tensorflow/python/framework/ops.py\" source_line=1177\n\t\tXLA Label: fusion\n\t\tShape: f32[4,64,540,960]\n\t\t==========================\n\n\tBuffer 7:\n\t\tSize: 284.77MiB\n\t\tOperator: op_type=\"Conv2D\" op_name=\"functional_1_1/block3a_expand_conv_1/convolution\" source_file=\"/opt/conda/lib/python3.10/site-packages/tensorflow/python/framework/ops.py\" source_line=1177\n\t\tXLA Label: custom-call\n\t\tShape: f32[4,144,270,480]\n\t\t==========================\n\n\tBuffer 8:\n\t\tSize: 284.77MiB\n\t\tOperator: op_type=\"DepthwiseConv2dNative\" op_name=\"functional_1_1/block2b_dwconv_1/depthwise\" source_file=\"/opt/conda/lib/python3.10/site-packages/tensorflow/python/framework/ops.py\" source_line=1177\n\t\tXLA Label: custom-call\n\t\tShape: f32[4,144,270,480]\n\t\t==========================\n\n\tBuffer 9:\n\t\tSize: 284.77MiB\n\t\tOperator: op_type=\"Conv2D\" op_name=\"functional_1_1/block2b_expand_conv_1/convolution\" source_file=\"/opt/conda/lib/python3.10/site-packages/tensorflow/python/framework/ops.py\" source_line=1177\n\t\tXLA Label: custom-call\n\t\tShape: f32[4,144,270,480]\n\t\t==========================\n\n\tBuffer 10:\n\t\tSize: 253.12MiB\n\t\tOperator: op_type=\"Conv2D\" op_name=\"functional_1_1/conv2d_2_1/convolution\" source_file=\"/opt/conda/lib/python3.10/site-packages/tensorflow/python/framework/ops.py\" source_line=1177\n\t\tXLA Label: custom-call\n\t\tShape: f32[4,128,270,480]\n\t\t==========================\n\n\tBuffer 11:\n\t\tSize: 253.12MiB\n\t\tOperator: op_type=\"DepthwiseConv2dNative\" op_name=\"functional_1_1/block1a_dwconv_1/depthwise\" source_file=\"/opt/conda/lib/python3.10/site-packages/tensorflow/python/framework/ops.py\" source_line=1177\n\t\tXLA Label: custom-call\n\t\tShape: f32[4,32,540,960]\n\t\t==========================\n\n\tBuffer 12:\n\t\tSize: 253.12MiB\n\t\tOperator: op_type=\"Conv2D\" op_name=\"functional_1_1/stem_conv_1/convolution\" source_file=\"/opt/conda/lib/python3.10/site-packages/tensorflow/python/framework/ops.py\" source_line=1177\n\t\tXLA Label: custom-call\n\t\tShape: f32[4,32,540,960]\n\t\t==========================\n\n\tBuffer 13:\n\t\tSize: 189.84MiB\n\t\tOperator: op_type=\"DepthwiseConv2dNative\" op_name=\"functional_1_1/block2a_dwconv_1/depthwise\" source_file=\"/opt/conda/lib/python3.10/site-packages/tensorflow/python/framework/ops.py\" source_line=1177\n\t\tXLA Label: custom-call\n\t\tShape: f32[4,96,270,480]\n\t\t==========================\n\n\tBuffer 14:\n\t\tSize: 189.84MiB\n\t\tOperator: op_name=\"XLA_Args\"\n\t\tEntry Parameter Subshape: f64[4,1080,1920,3]\n\t\t==========================\n\n\tBuffer 15:\n\t\tSize: 127.50MiB\n\t\tOperator: op_type=\"Conv2D\" op_name=\"functional_1_1/conv2d_1_2/convolution\" source_file=\"/opt/conda/lib/python3.10/site-packages/tensorflow/python/framework/ops.py\" source_line=1177\n\t\tXLA Label: custom-call\n\t\tShape: f32[4,256,136,240]\n\t\t==========================\n\n\n\t [[{{node StatefulPartitionedCall}}]]\nHint: If you want to see a list of allocated tensors when OOM happens, add report_tensor_allocations_upon_oom to RunOptions for current allocation info. This isn't available when running in Eager mode.\n [Op:__inference_one_step_on_iterator_51487]"],"ename":"ResourceExhaustedError","evalue":"Graph execution error:\n\nDetected at node StatefulPartitionedCall defined at (most recent call last):\n  File \"/opt/conda/lib/python3.10/runpy.py\", line 196, in _run_module_as_main\n\n  File \"/opt/conda/lib/python3.10/runpy.py\", line 86, in _run_code\n\n  File \"/opt/conda/lib/python3.10/site-packages/ipykernel_launcher.py\", line 18, in <module>\n\n  File \"/opt/conda/lib/python3.10/site-packages/traitlets/config/application.py\", line 1075, in launch_instance\n\n  File \"/opt/conda/lib/python3.10/site-packages/ipykernel/kernelapp.py\", line 739, in start\n\n  File \"/opt/conda/lib/python3.10/site-packages/tornado/platform/asyncio.py\", line 205, in start\n\n  File \"/opt/conda/lib/python3.10/asyncio/base_events.py\", line 603, in run_forever\n\n  File \"/opt/conda/lib/python3.10/asyncio/base_events.py\", line 1909, in _run_once\n\n  File \"/opt/conda/lib/python3.10/asyncio/events.py\", line 80, in _run\n\n  File \"/opt/conda/lib/python3.10/site-packages/ipykernel/kernelbase.py\", line 545, in dispatch_queue\n\n  File \"/opt/conda/lib/python3.10/site-packages/ipykernel/kernelbase.py\", line 534, in process_one\n\n  File \"/opt/conda/lib/python3.10/site-packages/ipykernel/kernelbase.py\", line 437, in dispatch_shell\n\n  File \"/opt/conda/lib/python3.10/site-packages/ipykernel/ipkernel.py\", line 362, in execute_request\n\n  File \"/opt/conda/lib/python3.10/site-packages/ipykernel/kernelbase.py\", line 778, in execute_request\n\n  File \"/opt/conda/lib/python3.10/site-packages/ipykernel/ipkernel.py\", line 449, in do_execute\n\n  File \"/opt/conda/lib/python3.10/site-packages/ipykernel/zmqshell.py\", line 549, in run_cell\n\n  File \"/opt/conda/lib/python3.10/site-packages/IPython/core/interactiveshell.py\", line 3051, in run_cell\n\n  File \"/opt/conda/lib/python3.10/site-packages/IPython/core/interactiveshell.py\", line 3106, in _run_cell\n\n  File \"/opt/conda/lib/python3.10/site-packages/IPython/core/async_helpers.py\", line 129, in _pseudo_sync_runner\n\n  File \"/opt/conda/lib/python3.10/site-packages/IPython/core/interactiveshell.py\", line 3311, in run_cell_async\n\n  File \"/opt/conda/lib/python3.10/site-packages/IPython/core/interactiveshell.py\", line 3493, in run_ast_nodes\n\n  File \"/opt/conda/lib/python3.10/site-packages/IPython/core/interactiveshell.py\", line 3553, in run_code\n\n  File \"/tmp/ipykernel_30/1019388919.py\", line 2, in <module>\n\n  File \"/opt/conda/lib/python3.10/site-packages/keras/src/utils/traceback_utils.py\", line 117, in error_handler\n\n  File \"/opt/conda/lib/python3.10/site-packages/keras/src/backend/tensorflow/trainer.py\", line 314, in fit\n\n  File \"/opt/conda/lib/python3.10/site-packages/keras/src/backend/tensorflow/trainer.py\", line 117, in one_step_on_iterator\n\nOut of memory while trying to allocate 16523089264 bytes.\nBufferAssignment OOM Debugging.\nBufferAssignment stats:\n             parameter allocation:  351.37MiB\n              constant allocation:       316B\n        maybe_live_out allocation:  153.62MiB\n     preallocated temp allocation:   15.39GiB\n  preallocated temp fragmentation:     6.6KiB (0.00%)\n                 total allocation:   15.73GiB\n              total fragmentation:   690.0KiB (0.00%)\nPeak buffers:\n\tBuffer 1:\n\t\tSize: 5.84GiB\n\t\tOperator: op_type=\"Conv2DBackpropFilter\" op_name=\"gradient_tape/functional_1_1/conv2d_3_1/convolution/Conv2DBackpropFilter\" source_file=\"/opt/conda/lib/python3.10/site-packages/tensorflow/python/framework/ops.py\" source_line=1177\n\t\tXLA Label: custom-call\n\t\tShape: u8[6273073152]\n\t\t==========================\n\n\tBuffer 2:\n\t\tSize: 2.10GiB\n\t\tXLA Label: fusion\n\t\tShape: f32[4,272,540,960]\n\t\t==========================\n\n\tBuffer 3:\n\t\tSize: 1.24GiB\n\t\tOperator: op_type=\"Sum\" op_name=\"gradient_tape/functional_1_1/up_sampling2d_4_1/Repeat/Sum\" source_file=\"/opt/conda/lib/python3.10/site-packages/tensorflow/python/framework/ops.py\" source_line=1177\n\t\tXLA Label: fusion\n\t\tShape: f32[1,4,160,540,960]\n\t\t==========================\n\n\tBuffer 4:\n\t\tSize: 1012.50MiB\n\t\tOperator: op_type=\"Sum\" op_name=\"gradient_tape/functional_1_1/conv2d_4_1/add/Sum\" source_file=\"/opt/conda/lib/python3.10/site-packages/tensorflow/python/framework/ops.py\" source_line=1177\n\t\tXLA Label: fusion\n\t\tShape: f32[4,32,1080,1920]\n\t\t==========================\n\n\tBuffer 5:\n\t\tSize: 759.38MiB\n\t\tOperator: op_type=\"Conv2D\" op_name=\"functional_1_1/block2a_expand_conv_1/convolution\" source_file=\"/opt/conda/lib/python3.10/site-packages/tensorflow/python/framework/ops.py\" source_line=1177\n\t\tXLA Label: custom-call\n\t\tShape: f32[4,96,540,960]\n\t\t==========================\n\n\tBuffer 6:\n\t\tSize: 506.25MiB\n\t\tOperator: op_type=\"Sum\" op_name=\"gradient_tape/functional_1_1/conv2d_3_1/add/Sum\" source_file=\"/opt/conda/lib/python3.10/site-packages/tensorflow/python/framework/ops.py\" source_line=1177\n\t\tXLA Label: fusion\n\t\tShape: f32[4,64,540,960]\n\t\t==========================\n\n\tBuffer 7:\n\t\tSize: 284.77MiB\n\t\tOperator: op_type=\"Conv2D\" op_name=\"functional_1_1/block3a_expand_conv_1/convolution\" source_file=\"/opt/conda/lib/python3.10/site-packages/tensorflow/python/framework/ops.py\" source_line=1177\n\t\tXLA Label: custom-call\n\t\tShape: f32[4,144,270,480]\n\t\t==========================\n\n\tBuffer 8:\n\t\tSize: 284.77MiB\n\t\tOperator: op_type=\"DepthwiseConv2dNative\" op_name=\"functional_1_1/block2b_dwconv_1/depthwise\" source_file=\"/opt/conda/lib/python3.10/site-packages/tensorflow/python/framework/ops.py\" source_line=1177\n\t\tXLA Label: custom-call\n\t\tShape: f32[4,144,270,480]\n\t\t==========================\n\n\tBuffer 9:\n\t\tSize: 284.77MiB\n\t\tOperator: op_type=\"Conv2D\" op_name=\"functional_1_1/block2b_expand_conv_1/convolution\" source_file=\"/opt/conda/lib/python3.10/site-packages/tensorflow/python/framework/ops.py\" source_line=1177\n\t\tXLA Label: custom-call\n\t\tShape: f32[4,144,270,480]\n\t\t==========================\n\n\tBuffer 10:\n\t\tSize: 253.12MiB\n\t\tOperator: op_type=\"Conv2D\" op_name=\"functional_1_1/conv2d_2_1/convolution\" source_file=\"/opt/conda/lib/python3.10/site-packages/tensorflow/python/framework/ops.py\" source_line=1177\n\t\tXLA Label: custom-call\n\t\tShape: f32[4,128,270,480]\n\t\t==========================\n\n\tBuffer 11:\n\t\tSize: 253.12MiB\n\t\tOperator: op_type=\"DepthwiseConv2dNative\" op_name=\"functional_1_1/block1a_dwconv_1/depthwise\" source_file=\"/opt/conda/lib/python3.10/site-packages/tensorflow/python/framework/ops.py\" source_line=1177\n\t\tXLA Label: custom-call\n\t\tShape: f32[4,32,540,960]\n\t\t==========================\n\n\tBuffer 12:\n\t\tSize: 253.12MiB\n\t\tOperator: op_type=\"Conv2D\" op_name=\"functional_1_1/stem_conv_1/convolution\" source_file=\"/opt/conda/lib/python3.10/site-packages/tensorflow/python/framework/ops.py\" source_line=1177\n\t\tXLA Label: custom-call\n\t\tShape: f32[4,32,540,960]\n\t\t==========================\n\n\tBuffer 13:\n\t\tSize: 189.84MiB\n\t\tOperator: op_type=\"DepthwiseConv2dNative\" op_name=\"functional_1_1/block2a_dwconv_1/depthwise\" source_file=\"/opt/conda/lib/python3.10/site-packages/tensorflow/python/framework/ops.py\" source_line=1177\n\t\tXLA Label: custom-call\n\t\tShape: f32[4,96,270,480]\n\t\t==========================\n\n\tBuffer 14:\n\t\tSize: 189.84MiB\n\t\tOperator: op_name=\"XLA_Args\"\n\t\tEntry Parameter Subshape: f64[4,1080,1920,3]\n\t\t==========================\n\n\tBuffer 15:\n\t\tSize: 127.50MiB\n\t\tOperator: op_type=\"Conv2D\" op_name=\"functional_1_1/conv2d_1_2/convolution\" source_file=\"/opt/conda/lib/python3.10/site-packages/tensorflow/python/framework/ops.py\" source_line=1177\n\t\tXLA Label: custom-call\n\t\tShape: f32[4,256,136,240]\n\t\t==========================\n\n\n\t [[{{node StatefulPartitionedCall}}]]\nHint: If you want to see a list of allocated tensors when OOM happens, add report_tensor_allocations_upon_oom to RunOptions for current allocation info. This isn't available when running in Eager mode.\n [Op:__inference_one_step_on_iterator_51487]","output_type":"error"}]}]}