FUSegNet_feature_visualization.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NtZrR6Lpamzt",
        "outputId": "0720d55f-bc7d-43d1-f88d-dcf4ab3f69dc"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ]
        }
      ],
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "wKYTa1YNahpR"
      },
      "outputs": [],
      "source": [
        "import sys\n",
        "sys.path.append('/content/drive/MyDrive/Foot_ulcer_torch')\n",
        "sys.path.append('/content/drive/MyDrive/Foot_ulcer_torch/sm_library')\n",
        "sys.path.append('/content/drive/MyDrive/Foot_ulcer_torch/utils')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "5QTR9txvZlTj"
      },
      "outputs": [],
      "source": [
        "# !pip install segmentation-models-pytorch --target='/content/drive/MyDrive/Foot_ulcer_torch/sm_library'\n",
        "# !pip install git+https://github.com/qubvel/segmentation_models.pytorch --target='/content/drive/MyDrive/Foot_ulcer_torch/sm_library'\n",
        "# !pip install albumentations --target='/content/drive/MyDrive/Foot_ulcer_torch/sm_library'\n",
        "# !pip install certifi --target='/content/drive/MyDrive/Foot_ulcer_torch/sm_library'"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "wWvUuJ7xZ17E"
      },
      "outputs": [],
      "source": [
        "import torch\n",
        "from torch.utils.data import DataLoader\n",
        "from torch.utils.data import Dataset as BaseDataset\n",
        "import albumentations as albu\n",
        "import cv2\n",
        "import numpy as np\n",
        "import segmentation_models_pytorch as smp\n",
        "from segmentation_models_pytorch.utils import metrics, losses, base\n",
        "import random\n",
        "import matplotlib.pyplot as plt\n",
        "import os\n",
        "from copy import deepcopy\n",
        "from datetime import datetime\n",
        "\n",
        "%matplotlib inline"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "u1TTozpn-D8o"
      },
      "source": [
        "## Dataloader"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "nx_iN5gGarSj"
      },
      "outputs": [],
      "source": [
        "class Dataset(BaseDataset):\n",
        "    \"\"\"CamVid Dataset. Read images, apply augmentation and preprocessing transformations.\n",
        "    \n",
        "    Args:\n",
        "        images_dir (str): path to images folder\n",
        "        masks_dir (str): path to segmentation masks folder\n",
        "        augmentation (albumentations.Compose): data transfromation pipeline \n",
        "            (e.g. flip, scale, etc.)\n",
        "        preprocessing (albumentations.Compose): data preprocessing \n",
        "            (e.g. noralization, shape manipulation, etc.)\n",
        "    \n",
        "    \"\"\"\n",
        "    \n",
        "    def __init__(\n",
        "            self, \n",
        "            list_IDs,\n",
        "            images_dir, \n",
        "            masks_dir, \n",
        "            augmentation=None, \n",
        "            preprocessing=None,\n",
        "    ):\n",
        "        self.ids = list_IDs\n",
        "        self.images_fps = [os.path.join(images_dir, image_id) for image_id in self.ids]\n",
        "        self.masks_fps = [os.path.join(masks_dir, image_id) for image_id in self.ids]\n",
        "              \n",
        "        self.augmentation = augmentation\n",
        "        self.preprocessing = preprocessing\n",
        "    \n",
        "    def __getitem__(self, i):\n",
        "        \n",
        "        # read data\n",
        "        image = cv2.imread(self.images_fps[i])\n",
        "        image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB) \n",
        "        mask = cv2.imread(self.masks_fps[i], 0) # graychannel # ----------------- pay attention ------------------ #\n",
        "        mask = mask/255.0   # converting mask to (0 and 1) # ----------------- pay attention ------------------ #\n",
        "        mask = np.expand_dims(mask, axis=-1)  # adding channel axis # ----------------- pay attention ------------------ #\n",
        "        \n",
        "        # apply augmentations\n",
        "        if self.augmentation:\n",
        "            sample = self.augmentation(image=image, mask=mask)\n",
        "            image, mask = sample['image'], sample['mask']\n",
        "        \n",
        "        # apply preprocessing\n",
        "        if self.preprocessing:\n",
        "            sample = self.preprocessing(image=image, mask=mask)\n",
        "            image, mask = sample['image'], sample['mask']\n",
        "            \n",
        "        return image, mask\n",
        "        \n",
        "    def __len__(self):\n",
        "        return len(self.ids)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yB8RKETkALtF"
      },
      "source": [
        "## Augmentation"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "T-ltFwaGAK3c"
      },
      "outputs": [],
      "source": [
        "def get_training_augmentation():\n",
        "    train_transform = [\n",
        "\n",
        "        albu.OneOf(\n",
        "            [\n",
        "                albu.HorizontalFlip(p=0.8),\n",
        "                albu.VerticalFlip(p=0.4),\n",
        "            ],\n",
        "            p=0.5,\n",
        "        ),\n",
        "        \n",
        "        albu.OneOf(\n",
        "            [\n",
        "                albu.ShiftScaleRotate(scale_limit=0.5, rotate_limit=0, shift_limit=0, p=1, border_mode=0), # scale only\n",
        "                albu.ShiftScaleRotate(scale_limit=0, rotate_limit=30, shift_limit=0, p=1, border_mode=0), # rotate only\n",
        "                albu.ShiftScaleRotate(scale_limit=0, rotate_limit=0, shift_limit=0.1, p=1, border_mode=0), # shift only\n",
        "                albu.ShiftScaleRotate(scale_limit=0.5, rotate_limit=30, shift_limit=0.1, p=1, border_mode=0), # affine transform\n",
        "            ],\n",
        "            p=0.9,\n",
        "        ),\n",
        "\n",
        "        # albu.PadIfNeeded(min_height=320, min_width=320, always_apply=True, border_mode=0),\n",
        "        # albu.RandomCrop(height=320, width=320, always_apply=True),\n",
        "\n",
        "        albu.OneOf(\n",
        "            [\n",
        "                albu.Perspective(p=1),\n",
        "                albu.GaussNoise(p=1),\n",
        "                albu.Sharpen(p=1),\n",
        "                albu.Blur(blur_limit=3, p=1),\n",
        "                albu.MotionBlur(blur_limit=3, p=1),\n",
        "            ],\n",
        "            p=0.2,\n",
        "        ),\n",
        "\n",
        "        albu.OneOf(\n",
        "            [\n",
        "                albu.CLAHE(p=1),\n",
        "                albu.RandomBrightnessContrast(brightness_limit=0.2, contrast_limit=0.2, p=1),\n",
        "                albu.RandomGamma(p=1),\n",
        "                albu.HueSaturationValue(p=1),\n",
        "            ],\n",
        "            p=0.2,\n",
        "        ),\n",
        "        \n",
        "    ]\n",
        "\n",
        "    return albu.Compose(train_transform, p=0.9) # 90% augmentation probability\n",
        "\n",
        "\n",
        "def get_validation_augmentation():\n",
        "    \"\"\"Add paddings to make image shape divisible by 32\"\"\"\n",
        "    test_transform = [\n",
        "        # albu.PadIfNeeded(512, 512)\n",
        "    ]\n",
        "    return albu.Compose(test_transform)\n",
        "\n",
        "\n",
        "def to_tensor(x, **kwargs):\n",
        "    return x.transpose(2, 0, 1).astype('float32')\n",
        "\n",
        "\n",
        "def get_preprocessing(preprocessing_fn):\n",
        "    \"\"\"Construct preprocessing transform\n",
        "    \n",
        "    Args:\n",
        "        preprocessing_fn (callbale): data normalization function \n",
        "            (can be specific for each pretrained neural network)\n",
        "    Return:\n",
        "        transform: albumentations.Compose\n",
        "    \n",
        "    \"\"\"\n",
        "    \n",
        "    _transform = [\n",
        "        albu.Lambda(image=preprocessing_fn),\n",
        "        albu.Lambda(image=to_tensor, mask=to_tensor),\n",
        "    ]\n",
        "    return albu.Compose(_transform) "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6KRxrncX_p0v"
      },
      "source": [
        "## Split dataset"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "6SRrhpEIaGqz",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "e8d18da0-b991-44a1-ebe0-848eb0579d2d"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "No. of test images:  278\n"
          ]
        }
      ],
      "source": [
        "#%% Load dataset\n",
        "x_test_dir = '/content/drive/MyDrive/Dataset/azh_wound_care_center_dataset_patches/test/images'\n",
        "y_test_dir = '/content/drive/MyDrive/Dataset/azh_wound_care_center_dataset_patches/test/labels'\n",
        "\n",
        "list_IDs_test = os.listdir(x_test_dir)\n",
        "\n",
        "print('No. of test images: ', len(list_IDs_test))"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "list_IDs_test.index('b23232898bafcf304b2138af86be40b9_0.png')\n",
        "# list_IDs_test.index('fa6fb6f94a3ba6c9b98ba40a3e0d1175_2.png')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cNsxEtSPszAT",
        "outputId": "c6e9c3b8-c97a-4e64-d870-83c2773c8c2c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0"
            ]
          },
          "metadata": {},
          "execution_count": 61
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Z9xxDHfp_yFO"
      },
      "source": [
        "## Parameters"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Yq5o1vzBYxN0"
      },
      "outputs": [],
      "source": [
        "# Parameters\n",
        "BASE_MODEL = 'Unet+scse'\n",
        "ENCODER = 'efficientnet-b7'\n",
        "ENCODER_WEIGHTS = 'imagenet'\n",
        "BATCH_SIZE = 2\n",
        "n_classes = 1 \n",
        "ACTIVATION = 'sigmoid' # could be None for logits or 'softmax2d' for multiclass segmentation\n",
        "DEVICE = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
        "LR = 0.0001 # learning rate\n",
        "EPOCHS = 200\n",
        "WEIGHT_DECAY = 1e-5\n",
        "SAVE_WEIGHTS_ONLY = True\n",
        "TO_CATEGORICAL = False\n",
        "SAVE_BEST_MODEL = True\n",
        "SAVE_LAST_MODEL = True\n",
        "\n",
        "PERIOD = 10 # periodically save checkpoints\n",
        "RAW_PREDICTION = False # if true, then stores raw predictions (i.e. before applying threshold)\n",
        "RETRAIN = False\n",
        "\n",
        "# For early stopping \n",
        "EARLY_STOP = False # True to activate early stopping\n",
        "PATIENCE = 50 # for early stopping"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gkk8Gaove69f"
      },
      "source": [
        "# Build model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "5Ey12h9jMX1q"
      },
      "outputs": [],
      "source": [
        "import ssl\n",
        "ssl._create_default_https_context = ssl._create_unverified_context\n",
        "\n",
        "aux_params=dict(\n",
        "    classes=n_classes, \n",
        "    activation=ACTIVATION,\n",
        "    dropout=0.1, # dropout ratio, default is None\n",
        ")\n",
        "\n",
        "# create segmentation model with pretrained encoder\n",
        "model = smp.Unet(\n",
        "    encoder_name=ENCODER, \n",
        "    encoder_weights=ENCODER_WEIGHTS,     \n",
        "    # aux_params=aux_params,\n",
        "    classes=n_classes, \n",
        "    activation=ACTIVATION,\n",
        "    decoder_attention_type='scse',\n",
        ")\n",
        "\n",
        "preprocessing_fn = smp.encoders.get_preprocessing_fn(ENCODER, ENCODER_WEIGHTS)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7qcadi38Ac5H"
      },
      "source": [
        "## Loss, optimizer, metrics, and callbacks"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Loss function\n",
        "dice_loss = losses.DiceLoss()\n",
        "focal_loss = losses.FocalLoss() \n",
        "total_loss = base.SumOfLosses(dice_loss, focal_loss)\n",
        "\n",
        "# Metrics\n",
        "metrics = [\n",
        "    metrics.IoU(threshold=0.5),\n",
        "    metrics.Fscore(threshold=0.5),\n",
        "]\n",
        "\n",
        "# Optimizer\n",
        "optimizer = torch.optim.Adam([ \n",
        "    dict(params=model.parameters(), lr=LR, weight_decay=WEIGHT_DECAY),\n",
        "])\n",
        "\n",
        "# Learning rate scheduler\n",
        "scheduler = torch.optim.lr_scheduler.ReduceLROnPlateau(optimizer, \n",
        "                              factor=0.1,\n",
        "                              mode='min',\n",
        "                              patience=10,\n",
        "                              min_lr=0.00001,\n",
        "                              verbose=True,\n",
        "                              )"
      ],
      "metadata": {
        "id": "26g3jFOZyWmg"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zWhXVUtHEnT7"
      },
      "source": [
        "## Inference"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Jjfqn5YI5rnG"
      },
      "source": [
        "Load model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "EB41dkBfEpiD"
      },
      "outputs": [],
      "source": [
        "# load best saved checkpoint\n",
        "model_name = 'Unet_pscsev1_efficientnet-b7_2023-02-28_10-05-44' # PscSE\n",
        "# model_name = 'Unet+scse_additive_efficientnet-b7_2023-03-01_18-56-05' #scSE\n",
        "# model_name = 'Unet+scse_efficientnet-b7_2023-02-19_01-21-41'\n",
        "checkpoint_loc = '/content/drive/MyDrive/Foot_ulcer_torch/checkpoints_oldDfu/' + model_name\n",
        "\n",
        "checkpoint = torch.load(os.path.join(checkpoint_loc, 'best_model.pth'))\n",
        "model.load_state_dict(checkpoint['state_dict'])\n",
        "optimizer.load_state_dict(checkpoint['optimizer'])\n",
        "\n",
        "# Uncomment if to load periodically saved checkpoints\n",
        "# model = torch.load(os.path.join(checkpoint_loc, 'cp-0050.pth'))"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model"
      ],
      "metadata": {
        "id": "Ov7G8lFpNqCZ",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "da1761be-95ef-492f-ed67-58712dc3d689"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Unet(\n",
              "  (encoder): EfficientNetEncoder(\n",
              "    (_conv_stem): Conv2dStaticSamePadding(\n",
              "      3, 64, kernel_size=(3, 3), stride=(2, 2), bias=False\n",
              "      (static_padding): ZeroPad2d((0, 1, 0, 1))\n",
              "    )\n",
              "    (_bn0): BatchNorm2d(64, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "    (_blocks): ModuleList(\n",
              "      (0): MBConvBlock(\n",
              "        (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "          64, 64, kernel_size=(3, 3), stride=[1, 1], groups=64, bias=False\n",
              "          (static_padding): ZeroPad2d((1, 1, 1, 1))\n",
              "        )\n",
              "        (_bn1): BatchNorm2d(64, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_se_reduce): Conv2dStaticSamePadding(\n",
              "          64, 16, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_se_expand): Conv2dStaticSamePadding(\n",
              "          16, 64, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_project_conv): Conv2dStaticSamePadding(\n",
              "          64, 32, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_bn2): BatchNorm2d(32, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_swish): MemoryEfficientSwish()\n",
              "      )\n",
              "      (1-3): 3 x MBConvBlock(\n",
              "        (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "          32, 32, kernel_size=(3, 3), stride=(1, 1), groups=32, bias=False\n",
              "          (static_padding): ZeroPad2d((1, 1, 1, 1))\n",
              "        )\n",
              "        (_bn1): BatchNorm2d(32, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_se_reduce): Conv2dStaticSamePadding(\n",
              "          32, 8, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_se_expand): Conv2dStaticSamePadding(\n",
              "          8, 32, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_project_conv): Conv2dStaticSamePadding(\n",
              "          32, 32, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_bn2): BatchNorm2d(32, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_swish): MemoryEfficientSwish()\n",
              "      )\n",
              "      (4): MBConvBlock(\n",
              "        (_expand_conv): Conv2dStaticSamePadding(\n",
              "          32, 192, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_bn0): BatchNorm2d(192, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "          192, 192, kernel_size=(3, 3), stride=[2, 2], groups=192, bias=False\n",
              "          (static_padding): ZeroPad2d((0, 1, 0, 1))\n",
              "        )\n",
              "        (_bn1): BatchNorm2d(192, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_se_reduce): Conv2dStaticSamePadding(\n",
              "          192, 8, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_se_expand): Conv2dStaticSamePadding(\n",
              "          8, 192, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_project_conv): Conv2dStaticSamePadding(\n",
              "          192, 48, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_bn2): BatchNorm2d(48, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_swish): MemoryEfficientSwish()\n",
              "      )\n",
              "      (5-10): 6 x MBConvBlock(\n",
              "        (_expand_conv): Conv2dStaticSamePadding(\n",
              "          48, 288, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_bn0): BatchNorm2d(288, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "          288, 288, kernel_size=(3, 3), stride=(1, 1), groups=288, bias=False\n",
              "          (static_padding): ZeroPad2d((1, 1, 1, 1))\n",
              "        )\n",
              "        (_bn1): BatchNorm2d(288, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_se_reduce): Conv2dStaticSamePadding(\n",
              "          288, 12, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_se_expand): Conv2dStaticSamePadding(\n",
              "          12, 288, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_project_conv): Conv2dStaticSamePadding(\n",
              "          288, 48, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_bn2): BatchNorm2d(48, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_swish): MemoryEfficientSwish()\n",
              "      )\n",
              "      (11): MBConvBlock(\n",
              "        (_expand_conv): Conv2dStaticSamePadding(\n",
              "          48, 288, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_bn0): BatchNorm2d(288, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "          288, 288, kernel_size=(5, 5), stride=[2, 2], groups=288, bias=False\n",
              "          (static_padding): ZeroPad2d((1, 2, 1, 2))\n",
              "        )\n",
              "        (_bn1): BatchNorm2d(288, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_se_reduce): Conv2dStaticSamePadding(\n",
              "          288, 12, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_se_expand): Conv2dStaticSamePadding(\n",
              "          12, 288, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_project_conv): Conv2dStaticSamePadding(\n",
              "          288, 80, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_bn2): BatchNorm2d(80, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_swish): MemoryEfficientSwish()\n",
              "      )\n",
              "      (12-17): 6 x MBConvBlock(\n",
              "        (_expand_conv): Conv2dStaticSamePadding(\n",
              "          80, 480, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_bn0): BatchNorm2d(480, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "          480, 480, kernel_size=(5, 5), stride=(1, 1), groups=480, bias=False\n",
              "          (static_padding): ZeroPad2d((2, 2, 2, 2))\n",
              "        )\n",
              "        (_bn1): BatchNorm2d(480, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_se_reduce): Conv2dStaticSamePadding(\n",
              "          480, 20, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_se_expand): Conv2dStaticSamePadding(\n",
              "          20, 480, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_project_conv): Conv2dStaticSamePadding(\n",
              "          480, 80, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_bn2): BatchNorm2d(80, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_swish): MemoryEfficientSwish()\n",
              "      )\n",
              "      (18): MBConvBlock(\n",
              "        (_expand_conv): Conv2dStaticSamePadding(\n",
              "          80, 480, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_bn0): BatchNorm2d(480, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "          480, 480, kernel_size=(3, 3), stride=[2, 2], groups=480, bias=False\n",
              "          (static_padding): ZeroPad2d((1, 1, 1, 1))\n",
              "        )\n",
              "        (_bn1): BatchNorm2d(480, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_se_reduce): Conv2dStaticSamePadding(\n",
              "          480, 20, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_se_expand): Conv2dStaticSamePadding(\n",
              "          20, 480, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_project_conv): Conv2dStaticSamePadding(\n",
              "          480, 160, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_bn2): BatchNorm2d(160, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_swish): MemoryEfficientSwish()\n",
              "      )\n",
              "      (19-27): 9 x MBConvBlock(\n",
              "        (_expand_conv): Conv2dStaticSamePadding(\n",
              "          160, 960, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_bn0): BatchNorm2d(960, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "          960, 960, kernel_size=(3, 3), stride=(1, 1), groups=960, bias=False\n",
              "          (static_padding): ZeroPad2d((1, 1, 1, 1))\n",
              "        )\n",
              "        (_bn1): BatchNorm2d(960, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_se_reduce): Conv2dStaticSamePadding(\n",
              "          960, 40, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_se_expand): Conv2dStaticSamePadding(\n",
              "          40, 960, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_project_conv): Conv2dStaticSamePadding(\n",
              "          960, 160, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_bn2): BatchNorm2d(160, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_swish): MemoryEfficientSwish()\n",
              "      )\n",
              "      (28): MBConvBlock(\n",
              "        (_expand_conv): Conv2dStaticSamePadding(\n",
              "          160, 960, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_bn0): BatchNorm2d(960, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "          960, 960, kernel_size=(5, 5), stride=[1, 1], groups=960, bias=False\n",
              "          (static_padding): ZeroPad2d((2, 2, 2, 2))\n",
              "        )\n",
              "        (_bn1): BatchNorm2d(960, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_se_reduce): Conv2dStaticSamePadding(\n",
              "          960, 40, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_se_expand): Conv2dStaticSamePadding(\n",
              "          40, 960, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_project_conv): Conv2dStaticSamePadding(\n",
              "          960, 224, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_bn2): BatchNorm2d(224, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_swish): MemoryEfficientSwish()\n",
              "      )\n",
              "      (29-37): 9 x MBConvBlock(\n",
              "        (_expand_conv): Conv2dStaticSamePadding(\n",
              "          224, 1344, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_bn0): BatchNorm2d(1344, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "          1344, 1344, kernel_size=(5, 5), stride=(1, 1), groups=1344, bias=False\n",
              "          (static_padding): ZeroPad2d((2, 2, 2, 2))\n",
              "        )\n",
              "        (_bn1): BatchNorm2d(1344, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_se_reduce): Conv2dStaticSamePadding(\n",
              "          1344, 56, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_se_expand): Conv2dStaticSamePadding(\n",
              "          56, 1344, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_project_conv): Conv2dStaticSamePadding(\n",
              "          1344, 224, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_bn2): BatchNorm2d(224, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_swish): MemoryEfficientSwish()\n",
              "      )\n",
              "      (38): MBConvBlock(\n",
              "        (_expand_conv): Conv2dStaticSamePadding(\n",
              "          224, 1344, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_bn0): BatchNorm2d(1344, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "          1344, 1344, kernel_size=(5, 5), stride=[2, 2], groups=1344, bias=False\n",
              "          (static_padding): ZeroPad2d((1, 2, 1, 2))\n",
              "        )\n",
              "        (_bn1): BatchNorm2d(1344, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_se_reduce): Conv2dStaticSamePadding(\n",
              "          1344, 56, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_se_expand): Conv2dStaticSamePadding(\n",
              "          56, 1344, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_project_conv): Conv2dStaticSamePadding(\n",
              "          1344, 384, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_bn2): BatchNorm2d(384, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_swish): MemoryEfficientSwish()\n",
              "      )\n",
              "      (39-50): 12 x MBConvBlock(\n",
              "        (_expand_conv): Conv2dStaticSamePadding(\n",
              "          384, 2304, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_bn0): BatchNorm2d(2304, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "          2304, 2304, kernel_size=(5, 5), stride=(1, 1), groups=2304, bias=False\n",
              "          (static_padding): ZeroPad2d((2, 2, 2, 2))\n",
              "        )\n",
              "        (_bn1): BatchNorm2d(2304, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_se_reduce): Conv2dStaticSamePadding(\n",
              "          2304, 96, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_se_expand): Conv2dStaticSamePadding(\n",
              "          96, 2304, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_project_conv): Conv2dStaticSamePadding(\n",
              "          2304, 384, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_bn2): BatchNorm2d(384, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_swish): MemoryEfficientSwish()\n",
              "      )\n",
              "      (51): MBConvBlock(\n",
              "        (_expand_conv): Conv2dStaticSamePadding(\n",
              "          384, 2304, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_bn0): BatchNorm2d(2304, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "          2304, 2304, kernel_size=(3, 3), stride=[1, 1], groups=2304, bias=False\n",
              "          (static_padding): ZeroPad2d((1, 1, 1, 1))\n",
              "        )\n",
              "        (_bn1): BatchNorm2d(2304, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_se_reduce): Conv2dStaticSamePadding(\n",
              "          2304, 96, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_se_expand): Conv2dStaticSamePadding(\n",
              "          96, 2304, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_project_conv): Conv2dStaticSamePadding(\n",
              "          2304, 640, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_bn2): BatchNorm2d(640, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_swish): MemoryEfficientSwish()\n",
              "      )\n",
              "      (52-54): 3 x MBConvBlock(\n",
              "        (_expand_conv): Conv2dStaticSamePadding(\n",
              "          640, 3840, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_bn0): BatchNorm2d(3840, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "          3840, 3840, kernel_size=(3, 3), stride=(1, 1), groups=3840, bias=False\n",
              "          (static_padding): ZeroPad2d((1, 1, 1, 1))\n",
              "        )\n",
              "        (_bn1): BatchNorm2d(3840, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_se_reduce): Conv2dStaticSamePadding(\n",
              "          3840, 160, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_se_expand): Conv2dStaticSamePadding(\n",
              "          160, 3840, kernel_size=(1, 1), stride=(1, 1)\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_project_conv): Conv2dStaticSamePadding(\n",
              "          3840, 640, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "          (static_padding): Identity()\n",
              "        )\n",
              "        (_bn2): BatchNorm2d(640, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "        (_swish): MemoryEfficientSwish()\n",
              "      )\n",
              "    )\n",
              "    (_conv_head): Conv2dStaticSamePadding(\n",
              "      640, 2560, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "      (static_padding): Identity()\n",
              "    )\n",
              "    (_bn1): BatchNorm2d(2560, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "    (_avg_pooling): AdaptiveAvgPool2d(output_size=1)\n",
              "    (_dropout): Dropout(p=0.5, inplace=False)\n",
              "    (_swish): MemoryEfficientSwish()\n",
              "  )\n",
              "  (decoder): UnetDecoder(\n",
              "    (center): Identity()\n",
              "    (blocks): ModuleList(\n",
              "      (0): DecoderBlock(\n",
              "        (conv1): Conv2dReLU(\n",
              "          (0): Conv2d(864, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "          (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          (2): ReLU(inplace=True)\n",
              "        )\n",
              "        (attention1): SCSEModule0(\n",
              "          (cSE): Sequential(\n",
              "            (0): AdaptiveAvgPool2d(output_size=1)\n",
              "            (1): Conv2d(864, 54, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (2): ReLU(inplace=True)\n",
              "            (3): Conv2d(54, 864, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (4): Sigmoid()\n",
              "          )\n",
              "          (sSE): Sequential(\n",
              "            (0): Conv2d(864, 1, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (1): Sigmoid()\n",
              "          )\n",
              "          (if_concat): Sequential(\n",
              "            (0): Conv2d(1728, 864, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (1): BatchNorm2d(864, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (conv2): Conv2dReLU(\n",
              "          (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "          (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          (2): ReLU(inplace=True)\n",
              "        )\n",
              "        (attention2): SCSEModule0(\n",
              "          (cSE): Sequential(\n",
              "            (0): AdaptiveAvgPool2d(output_size=1)\n",
              "            (1): Conv2d(864, 54, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (2): ReLU(inplace=True)\n",
              "            (3): Conv2d(54, 864, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (4): Sigmoid()\n",
              "          )\n",
              "          (sSE): Sequential(\n",
              "            (0): Conv2d(864, 1, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (1): Sigmoid()\n",
              "          )\n",
              "          (if_concat): Sequential(\n",
              "            (0): Conv2d(1728, 864, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (1): BatchNorm2d(864, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "      )\n",
              "      (1): DecoderBlock(\n",
              "        (conv1): Conv2dReLU(\n",
              "          (0): Conv2d(336, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "          (1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          (2): ReLU(inplace=True)\n",
              "        )\n",
              "        (attention1): SCSEModule0(\n",
              "          (cSE): Sequential(\n",
              "            (0): AdaptiveAvgPool2d(output_size=1)\n",
              "            (1): Conv2d(336, 21, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (2): ReLU(inplace=True)\n",
              "            (3): Conv2d(21, 336, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (4): Sigmoid()\n",
              "          )\n",
              "          (sSE): Sequential(\n",
              "            (0): Conv2d(336, 1, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (1): Sigmoid()\n",
              "          )\n",
              "          (if_concat): Sequential(\n",
              "            (0): Conv2d(672, 336, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (1): BatchNorm2d(336, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (conv2): Conv2dReLU(\n",
              "          (0): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "          (1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          (2): ReLU(inplace=True)\n",
              "        )\n",
              "        (attention2): SCSEModule0(\n",
              "          (cSE): Sequential(\n",
              "            (0): AdaptiveAvgPool2d(output_size=1)\n",
              "            (1): Conv2d(336, 21, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (2): ReLU(inplace=True)\n",
              "            (3): Conv2d(21, 336, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (4): Sigmoid()\n",
              "          )\n",
              "          (sSE): Sequential(\n",
              "            (0): Conv2d(336, 1, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (1): Sigmoid()\n",
              "          )\n",
              "          (if_concat): Sequential(\n",
              "            (0): Conv2d(672, 336, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (1): BatchNorm2d(336, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "      )\n",
              "      (2): DecoderBlock(\n",
              "        (conv1): Conv2dReLU(\n",
              "          (0): Conv2d(176, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "          (1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          (2): ReLU(inplace=True)\n",
              "        )\n",
              "        (attention1): SCSEModule0(\n",
              "          (cSE): Sequential(\n",
              "            (0): AdaptiveAvgPool2d(output_size=1)\n",
              "            (1): Conv2d(176, 11, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (2): ReLU(inplace=True)\n",
              "            (3): Conv2d(11, 176, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (4): Sigmoid()\n",
              "          )\n",
              "          (sSE): Sequential(\n",
              "            (0): Conv2d(176, 1, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (1): Sigmoid()\n",
              "          )\n",
              "          (if_concat): Sequential(\n",
              "            (0): Conv2d(352, 176, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (1): BatchNorm2d(176, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (conv2): Conv2dReLU(\n",
              "          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "          (1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          (2): ReLU(inplace=True)\n",
              "        )\n",
              "        (attention2): SCSEModule0(\n",
              "          (cSE): Sequential(\n",
              "            (0): AdaptiveAvgPool2d(output_size=1)\n",
              "            (1): Conv2d(176, 11, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (2): ReLU(inplace=True)\n",
              "            (3): Conv2d(11, 176, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (4): Sigmoid()\n",
              "          )\n",
              "          (sSE): Sequential(\n",
              "            (0): Conv2d(176, 1, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (1): Sigmoid()\n",
              "          )\n",
              "          (if_concat): Sequential(\n",
              "            (0): Conv2d(352, 176, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (1): BatchNorm2d(176, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "      )\n",
              "      (3): DecoderBlock(\n",
              "        (conv1): Conv2dReLU(\n",
              "          (0): Conv2d(128, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "          (1): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          (2): ReLU(inplace=True)\n",
              "        )\n",
              "        (attention1): SCSEModule0(\n",
              "          (cSE): Sequential(\n",
              "            (0): AdaptiveAvgPool2d(output_size=1)\n",
              "            (1): Conv2d(128, 8, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (2): ReLU(inplace=True)\n",
              "            (3): Conv2d(8, 128, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (4): Sigmoid()\n",
              "          )\n",
              "          (sSE): Sequential(\n",
              "            (0): Conv2d(128, 1, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (1): Sigmoid()\n",
              "          )\n",
              "          (if_concat): Sequential(\n",
              "            (0): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (conv2): Conv2dReLU(\n",
              "          (0): Conv2d(32, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "          (1): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          (2): ReLU(inplace=True)\n",
              "        )\n",
              "        (attention2): SCSEModule0(\n",
              "          (cSE): Sequential(\n",
              "            (0): AdaptiveAvgPool2d(output_size=1)\n",
              "            (1): Conv2d(128, 8, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (2): ReLU(inplace=True)\n",
              "            (3): Conv2d(8, 128, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (4): Sigmoid()\n",
              "          )\n",
              "          (sSE): Sequential(\n",
              "            (0): Conv2d(128, 1, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (1): Sigmoid()\n",
              "          )\n",
              "          (if_concat): Sequential(\n",
              "            (0): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "      )\n",
              "      (4): DecoderBlock(\n",
              "        (conv1): Conv2dReLU(\n",
              "          (0): Conv2d(32, 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "          (1): BatchNorm2d(16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          (2): ReLU(inplace=True)\n",
              "        )\n",
              "        (attention1): SCSEModule0(\n",
              "          (cSE): Sequential(\n",
              "            (0): AdaptiveAvgPool2d(output_size=1)\n",
              "            (1): Conv2d(32, 2, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (2): ReLU(inplace=True)\n",
              "            (3): Conv2d(2, 32, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (4): Sigmoid()\n",
              "          )\n",
              "          (sSE): Sequential(\n",
              "            (0): Conv2d(32, 1, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (1): Sigmoid()\n",
              "          )\n",
              "          (if_concat): Sequential(\n",
              "            (0): Conv2d(64, 32, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (1): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (conv2): Conv2dReLU(\n",
              "          (0): Conv2d(16, 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "          (1): BatchNorm2d(16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          (2): ReLU(inplace=True)\n",
              "        )\n",
              "        (attention2): SCSEModule0(\n",
              "          (cSE): Sequential(\n",
              "            (0): AdaptiveAvgPool2d(output_size=1)\n",
              "            (1): Conv2d(32, 2, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (2): ReLU(inplace=True)\n",
              "            (3): Conv2d(2, 32, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (4): Sigmoid()\n",
              "          )\n",
              "          (sSE): Sequential(\n",
              "            (0): Conv2d(32, 1, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (1): Sigmoid()\n",
              "          )\n",
              "          (if_concat): Sequential(\n",
              "            (0): Conv2d(64, 32, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (1): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "      )\n",
              "    )\n",
              "  )\n",
              "  (segmentation_head): SegmentationHead(\n",
              "    (0): Conv2d(16, 1, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
              "    (1): Identity()\n",
              "    (2): Activation(\n",
              "      (activation): Sigmoid()\n",
              "    )\n",
              "  )\n",
              ")"
            ]
          },
          "metadata": {},
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from torchsummary import summary "
      ],
      "metadata": {
        "id": "wR2x4rikOn7t"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model.to(\"cuda\")\n",
        "summary(model, (3, 224, 224))"
      ],
      "metadata": {
        "id": "GSt97ZcDOply"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Extract all layers\n",
        "layers = list(model.children())"
      ],
      "metadata": {
        "id": "-QMBMTc-jzlS"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Extract a layer\n",
        "model.encoder._conv_stem"
      ],
      "metadata": {
        "id": "0XXlDh7xkXon",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d1310bcd-aff9-4ab0-c748-acff75b6385c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Conv2dStaticSamePadding(\n",
              "  3, 64, kernel_size=(3, 3), stride=(2, 2), bias=False\n",
              "  (static_padding): ZeroPad2d((0, 1, 0, 1))\n",
              ")"
            ]
          },
          "metadata": {},
          "execution_count": 23
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model.decoder.blocks[0].conv1[2]"
      ],
      "metadata": {
        "id": "xOjrfEy7p_df",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "05dcc0e3-40ae-4089-b337-5a3263c15ac1"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "ReLU(inplace=True)"
            ]
          },
          "metadata": {},
          "execution_count": 24
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model.decoder.blocks[0].attention2"
      ],
      "metadata": {
        "id": "akNFVIxJSHTm",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "fce7e95c-d563-497f-e600-80298009c52a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "SCSEModule0(\n",
              "  (cSE): Sequential(\n",
              "    (0): AdaptiveAvgPool2d(output_size=1)\n",
              "    (1): Conv2d(864, 54, kernel_size=(1, 1), stride=(1, 1))\n",
              "    (2): ReLU(inplace=True)\n",
              "    (3): Conv2d(54, 864, kernel_size=(1, 1), stride=(1, 1))\n",
              "    (4): Sigmoid()\n",
              "  )\n",
              "  (sSE): Sequential(\n",
              "    (0): Conv2d(864, 1, kernel_size=(1, 1), stride=(1, 1))\n",
              "    (1): Sigmoid()\n",
              "  )\n",
              "  (if_concat): Sequential(\n",
              "    (0): Conv2d(1728, 864, kernel_size=(1, 1), stride=(1, 1))\n",
              "    (1): BatchNorm2d(864, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "  )\n",
              ")"
            ]
          },
          "metadata": {},
          "execution_count": 29
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model"
      ],
      "metadata": {
        "id": "a2yUkTD6Rq73"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Create a hook\n",
        "# (Ref: https://discuss.pytorch.org/t/how-can-l-load-my-best-model-as-a-feature-extractor-evaluator/17254/6,\n",
        "# https://kozodoi.me/python/deep%20learning/pytorch/tutorial/2021/05/27/extracting-features.html)\n",
        "\n",
        "int_feat = {} # intermediate features\n",
        "def intermediate_features(name):\n",
        "    def hook(model, input, output):\n",
        "        int_feat[name] = output.detach()\n",
        "    return hook\n",
        "\n",
        "# model.decoder.blocks[0].conv1[2].register_forward_hook(intermediate_features('feat'))\n",
        "\n",
        "model.decoder.blocks[0].attention1.register_forward_hook(intermediate_features('b0_attn1')) # max-out\n",
        "model.decoder.blocks[0].attention2.register_forward_hook(intermediate_features('b0_attn2')) # additive\n",
        "\n",
        "model.decoder.blocks[0].conv1.register_forward_hook(intermediate_features('b0_conv1'))\n",
        "\n",
        "model.decoder.blocks[1].attention1.register_forward_hook(intermediate_features('b1_attn1')) # max-out\n",
        "model.decoder.blocks[1].attention2.register_forward_hook(intermediate_features('b1_attn2')) # additive\n",
        "\n",
        "model.decoder.blocks[1].conv1.register_forward_hook(intermediate_features('b1_conv1'))\n",
        "\n",
        "model.decoder.blocks[2].attention1.register_forward_hook(intermediate_features('b2_attn1')) # max-out\n",
        "model.decoder.blocks[2].attention2.register_forward_hook(intermediate_features('b2_attn2')) # additive\n",
        "\n",
        "model.decoder.blocks[2].conv1.register_forward_hook(intermediate_features('b2_conv1'))\n",
        "\n",
        "model.decoder.blocks[3].attention1.register_forward_hook(intermediate_features('b3_attn1')) # max-out\n",
        "model.decoder.blocks[3].attention2.register_forward_hook(intermediate_features('b3_attn2')) # additive\n",
        "\n",
        "model.decoder.blocks[3].conv1.register_forward_hook(intermediate_features('b3_conv1'))\n",
        "\n",
        "# model.decoder.blocks[4].attention1.register_forward_hook(intermediate_features('b4_attn1')) # last stage don't have max-out\n",
        "model.decoder.blocks[4].attention2.register_forward_hook(intermediate_features('b4_attn2')) # additive\n",
        "\n",
        "model.decoder.blocks[4].conv1.register_forward_hook(intermediate_features('b4_conv1')) # final output\n"
      ],
      "metadata": {
        "id": "ANG9UgorrxqF",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "7735772f-f8a3-42eb-8f96-16a000433d70"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<torch.utils.hooks.RemovableHandle at 0x7fe961fc2970>"
            ]
          },
          "metadata": {},
          "execution_count": 39
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_dataset = Dataset(\n",
        "    list_IDs_test,\n",
        "    x_test_dir, \n",
        "    y_test_dir, \n",
        "    augmentation=get_validation_augmentation(), \n",
        "    preprocessing=get_preprocessing(preprocessing_fn),\n",
        ")\n",
        "\n",
        "test_dataloader = DataLoader(test_dataset, \n",
        "                             batch_size=1, \n",
        "                             shuffle=False, \n",
        "                             num_workers=6)\n",
        "\n",
        "it = iter(test_dataloader)"
      ],
      "metadata": {
        "id": "9wyw3EcNs1U5"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "img, mask = next(it)"
      ],
      "metadata": {
        "id": "vosK4a7etYa4"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "img.size()"
      ],
      "metadata": {
        "id": "5IClPXuOvJeB",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "e367b818-6c06-4874-c4c5-5c12b8f848ff"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "torch.Size([1, 3, 224, 224])"
            ]
          },
          "metadata": {},
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "img2 = torch.permute(torch.squeeze(img), [-1, 1, 0]) # convert from CHW to HWC"
      ],
      "metadata": {
        "id": "ON-NMtqvtnsv"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "img2.size()"
      ],
      "metadata": {
        "id": "-5PcsVg9uCWX"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plt.imshow(img2)"
      ],
      "metadata": {
        "id": "m2hNfBdyuJdf",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 469
        },
        "outputId": "4c04bc73-ff92-4ee7-9e67-d90ccbc6cb18"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7fb3b7f47550>"
            ]
          },
          "metadata": {},
          "execution_count": 53
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Run from Dataset class\n",
        "idx = 273 #0 #273 # 12 # 10\n",
        "\n",
        "# Read original image\n",
        "org_img = cv2.imread(os.path.join(x_test_dir, list_IDs_test[idx]))[:,:,::-1]\n",
        "\n",
        "# Read standardize image and mask using the Dataset class\n",
        "img, mask = test_dataset[idx] # get image and mask\n",
        "    \n",
        "img = torch.from_numpy(img).to(DEVICE).unsqueeze(0) # convert to tensor and add batch axis\n",
        "\n",
        "print('Image size:', img.size())"
      ],
      "metadata": {
        "id": "9z85hkT9rtGZ",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ac192b42-04e0-4a73-b44d-2628da245b29"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Image size: torch.Size([1, 3, 224, 224])\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "fig, ax = plt.subplots(1,2)\n",
        "ax[0].imshow(org_img)\n",
        "ax[1].imshow(img.squeeze().permute([1, 2, 0]).cpu())"
      ],
      "metadata": {
        "id": "zTOCtugwsuGz",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 326
        },
        "outputId": "8675349a-ac1a-4e8b-8ab3-54feb7c8bea3"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7fe9444b7e20>"
            ]
          },
          "metadata": {},
          "execution_count": 60
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "output = model(img.to(DEVICE))\n",
        "\n",
        "b0_attn1 = int_feat[\"b0_attn1\"]\n",
        "b0_attn2 = int_feat[\"b0_attn2\"]\n",
        "b0_conv1 = int_feat[\"b0_conv1\"]\n",
        "\n",
        "b1_attn1 = int_feat[\"b1_attn1\"]\n",
        "b1_attn2 = int_feat[\"b1_attn2\"]\n",
        "b1_conv1 = int_feat[\"b1_conv1\"]\n",
        "\n",
        "b2_attn1 = int_feat[\"b2_attn1\"]\n",
        "b2_attn2 = int_feat[\"b2_attn2\"]\n",
        "b2_conv1 = int_feat[\"b2_conv1\"]\n",
        "\n",
        "b3_attn1 = int_feat[\"b3_attn1\"]\n",
        "b3_attn2 = int_feat[\"b3_attn2\"]\n",
        "b3_conv1 = int_feat[\"b3_conv1\"]\n",
        "\n",
        "# b4_attn1 = int_feat[\"b4_attn1\"]\n",
        "b4_attn2 = int_feat[\"b4_attn2\"]\n",
        "b4_conv1 = int_feat[\"b4_conv1\"]"
      ],
      "metadata": {
        "id": "JSIpcwc0uNfY"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def avg_output(input):\n",
        "  \"\"\"\n",
        "  Take the average along the channel axis.\n",
        "\n",
        "  Input\n",
        "  -------\n",
        "  input: A tensor. Size: batch x ch x H x W\n",
        "\n",
        "  Output\n",
        "  -------\n",
        "  output: A tensor. Returns average image.\n",
        "\n",
        "  \"\"\"\n",
        "\n",
        "  output = torch.squeeze(input)\n",
        "  output = torch.sum(output, dim=0)\n",
        "  output = output / output.size()[0]\n",
        "\n",
        "  return output"
      ],
      "metadata": {
        "id": "fjMsdp6tvvRY"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Create output for visualization\n",
        "# Output shape -> batch, channel, height, width\n",
        "\n",
        "# int_output = torch.squeeze(int_out) \n",
        "# int_output = torch.sum(int_output, dim=0)\n",
        "# int_output = int_output / int_output.size()[0]\n",
        "\n",
        "b0_attn = avg_output(b0_attn1 + b0_attn2)\n",
        "b0_conv1 = avg_output(b0_conv1)\n",
        "\n",
        "b1_attn = avg_output(b1_attn1 + b1_attn2)\n",
        "b1_conv1 = avg_output(b1_conv1)\n",
        "\n",
        "b2_attn = avg_output(b2_attn1 + b2_attn2)\n",
        "b2_conv1 = avg_output(b2_conv1)\n",
        "\n",
        "b3_attn = avg_output(b3_attn1 + b3_attn2)\n",
        "b3_conv1 = avg_output(b3_conv1)\n",
        "\n",
        "b4_attn = avg_output(b4_attn2)\n",
        "b4_conv1 = avg_output(b4_conv1)"
      ],
      "metadata": {
        "id": "lQg9VfhhweHx"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# int_output\n",
        "b4_conv1.shape"
      ],
      "metadata": {
        "id": "tr99h7bkxcoI",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "7fca8a20-b4af-430e-cd26-086f2992b934"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "torch.Size([224, 224])"
            ]
          },
          "metadata": {},
          "execution_count": 64
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "fig, ax = plt.subplots(2, 6, figsize=(13, 5))\n",
        "\n",
        "ax[0][0].imshow(b0_attn.cpu(), cmap='gray')\n",
        "ax[0][0].set_xticklabels([]); ax[0][0].set_yticklabels([])\n",
        "ax[0][0].set_title('Attention 4')\n",
        "\n",
        "ax[0][1].imshow(b1_attn.cpu(), cmap='gray')\n",
        "ax[0][1].set_xticklabels([]); ax[0][1].set_yticklabels([])\n",
        "ax[0][1].set_title('Attention 3')\n",
        "\n",
        "ax[0][2].imshow(b2_attn.cpu(), cmap='gray')\n",
        "ax[0][2].set_xticklabels([]); ax[0][2].set_yticklabels([])\n",
        "ax[0][2].set_title('Attention 2')\n",
        "\n",
        "ax[0][3].imshow(b3_attn.cpu(), cmap='gray')\n",
        "ax[0][3].set_xticklabels([]); ax[0][3].set_yticklabels([])\n",
        "ax[0][3].set_title('Attention 1')\n",
        "\n",
        "ax[0][4].imshow(b4_attn.cpu(), cmap='gray')\n",
        "ax[0][4].set_xticklabels([]); ax[0][4].set_yticklabels([])\n",
        "ax[0][4].set_title('Attention 0')\n",
        "\n",
        "ax[1][0].imshow(b0_conv1.cpu(), cmap='gray')\n",
        "ax[1][0].set_xticklabels([]); ax[1][0].set_yticklabels([])\n",
        "ax[1][0].set_title('Decoder 4')\n",
        "\n",
        "ax[1][1].imshow(b1_conv1.cpu(), cmap='gray')\n",
        "ax[1][1].set_xticklabels([]); ax[1][1].set_yticklabels([])\n",
        "ax[1][1].set_title('Decoder 3')\n",
        "\n",
        "ax[1][2].imshow(b2_conv1.cpu(), cmap='gray')\n",
        "ax[1][2].set_xticklabels([]); ax[1][2].set_yticklabels([])\n",
        "ax[1][2].set_title('Decoder 2')\n",
        "\n",
        "ax[1][3].imshow(b3_conv1.cpu(), cmap='gray')\n",
        "ax[1][3].set_xticklabels([]); ax[1][3].set_yticklabels([])\n",
        "ax[1][3].set_title('Decoder 1')\n",
        "\n",
        "ax[1][4].imshow(output.squeeze().detach().cpu(), cmap='gray')\n",
        "# ax[1][4].imshow(b4_conv1.cpu(), cmap='gray')\n",
        "ax[1][4].set_xticklabels([]); ax[1][4].set_yticklabels([])\n",
        "ax[1][4].set_title('Decoder 0')\n",
        "\n",
        "ax[0][5].imshow(org_img)\n",
        "ax[0][5].set_xticklabels([]); ax[0][5].set_yticklabels([])\n",
        "ax[0][5].set_title('Original image')\n",
        "ax[1][5].imshow(np.squeeze(mask), cmap='gray')\n",
        "ax[1][5].set_xticklabels([]); ax[1][5].set_yticklabels([])\n",
        "ax[1][5].set_title('Ground truth')\n",
        "\n",
        "fig.tight_layout()\n",
        "fig.subplots_adjust(wspace=0.02, hspace=0.1)\n",
        "\n",
        "save_fig_dir = \"/content/drive/MyDrive/Foot_ulcer_torch/plots_oldDfu/attention_output\"\n",
        "if not os.path.exists(save_fig_dir): os.makedirs(save_fig_dir)\n",
        "\n",
        "fig.savefig(os.path.join(save_fig_dir, model_name + '_idx_' + str(idx) + '.png'))"
      ],
      "metadata": {
        "id": "DRJYloC2jOq6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 514
        },
        "outputId": "effd897e-178b-497f-ae56-92565f16c826"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1300x500 with 12 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "output.squeeze().cpu().size()"
      ],
      "metadata": {
        "id": "WGjk7UXzqfsK",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d4e00784-1a78-4837-e039-891933489818"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "torch.Size([224, 224])"
            ]
          },
          "metadata": {},
          "execution_count": 69
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.imshow(b4_conv1.cpu(), cmap='gray')"
      ],
      "metadata": {
        "id": "rG6WWLGXxf4H",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 451
        },
        "outputId": "35bc39df-1952-4687-91a1-53defda0bb62"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7fb3b63f1040>"
            ]
          },
          "metadata": {},
          "execution_count": 70
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "machine_shape": "hm",
      "provenance": []
    },
    "gpuClass": "standard",
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}