st235/Week3.ipynb

## Week3.ipynb
{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "hF5AYPeJRixk"
   },
   "source": [
    "# Scoring"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "nTEqYvSdPvAF"
   },
   "source": [
    "- The maximum number of points for this assignment is 10, the minimum number of points is 0.\n",
    "- You have one week to complete the assignment. Once the assignment is submitted you are not allowed to change it.\n",
    "- One week delay is penalized with 1 point.\n",
    "  - Example. The assignment is issued on the 1st January. The deadline without penalization is until 23:59 January 14th (anywhere on Earth). Student A submits his assignment on 22:51 January 6th and is not penalized; student B submits his assignment on 01:13 January 8th and is penalized with 1 point; student C submits his assignment on 3:56 January 16th and is penalized with 2 points and so on.\n",
    "- You have three weeks to compelte the assignment. After three weeks we will not accept solutions."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "oB9i1IlyhU0r"
   },
   "source": [
    "\n",
    "\n",
    "# Setting up the environment\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "sUIcN6uO5p-x",
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--2023-09-23 12:31:20--  https://github.com/iburenko/hse_course/raw/main/week3/data/corona_train_segm_dataset.npz\n",
      "Resolving github.com (github.com)... 140.82.121.4\n",
      "Connecting to github.com (github.com)|140.82.121.4|:443... connected.\n",
      "HTTP request sent, awaiting response... 302 Found\n",
      "Location: https://media.githubusercontent.com/media/iburenko/hse_course/main/week3/data/corona_train_segm_dataset.npz [following]\n",
      "--2023-09-23 12:31:20--  https://media.githubusercontent.com/media/iburenko/hse_course/main/week3/data/corona_train_segm_dataset.npz\n",
      "Resolving media.githubusercontent.com (media.githubusercontent.com)... 185.199.109.133, 185.199.110.133, 185.199.108.133, ...\n",
      "Connecting to media.githubusercontent.com (media.githubusercontent.com)|185.199.109.133|:443... connected.\n",
      "HTTP request sent, awaiting response... 200 OK\n",
      "Length: 339292009 (324M) [application/octet-stream]\n",
      "Saving to: ‘corona_train_segm_dataset.npz.3’\n",
      "\n",
      "corona_train_segm_d 100%[===================>] 323.57M  7.86MB/s    in 42s     \n",
      "\n",
      "2023-09-23 12:32:13 (7.79 MB/s) - ‘corona_train_segm_dataset.npz.3’ saved [339292009/339292009]\n",
      "\n",
      "--2023-09-23 12:32:13--  https://github.com/iburenko/hse_course/raw/main/week3/data/corona_val_segm_dataset.npz\n",
      "Resolving github.com (github.com)... 140.82.121.4\n",
      "Connecting to github.com (github.com)|140.82.121.4|:443... connected.\n",
      "HTTP request sent, awaiting response... 302 Found\n",
      "Location: https://media.githubusercontent.com/media/iburenko/hse_course/main/week3/data/corona_val_segm_dataset.npz [following]\n",
      "--2023-09-23 12:32:13--  https://media.githubusercontent.com/media/iburenko/hse_course/main/week3/data/corona_val_segm_dataset.npz\n",
      "Resolving media.githubusercontent.com (media.githubusercontent.com)... 185.199.109.133, 185.199.110.133, 185.199.108.133, ...\n",
      "Connecting to media.githubusercontent.com (media.githubusercontent.com)|185.199.109.133|:443... connected.\n",
      "HTTP request sent, awaiting response... 200 OK\n",
      "Length: 92458942 (88M) [application/octet-stream]\n",
      "Saving to: ‘corona_val_segm_dataset.npz.3’\n",
      "\n",
      "corona_val_segm_dat 100%[===================>]  88.17M  7.62MB/s    in 11s     \n",
      "\n",
      "2023-09-23 12:32:28 (7.87 MB/s) - ‘corona_val_segm_dataset.npz.3’ saved [92458942/92458942]\n",
      "\n",
      "--2023-09-23 12:32:28--  https://github.com/iburenko/hse_course/raw/main/week3/data/radiopedia_train_segm_dataset.npz\n",
      "Resolving github.com (github.com)... 140.82.121.4\n",
      "Connecting to github.com (github.com)|140.82.121.4|:443... connected.\n",
      "HTTP request sent, awaiting response... 302 Found\n",
      "Location: https://media.githubusercontent.com/media/iburenko/hse_course/main/week3/data/radiopedia_train_segm_dataset.npz [following]\n",
      "--2023-09-23 12:32:28--  https://media.githubusercontent.com/media/iburenko/hse_course/main/week3/data/radiopedia_train_segm_dataset.npz\n",
      "Resolving media.githubusercontent.com (media.githubusercontent.com)... 185.199.109.133, 185.199.110.133, 185.199.108.133, ...\n",
      "Connecting to media.githubusercontent.com (media.githubusercontent.com)|185.199.109.133|:443... connected.\n",
      "HTTP request sent, awaiting response... 200 OK\n",
      "Length: 102561953 (98M) [application/octet-stream]\n",
      "Saving to: ‘radiopedia_train_segm_dataset.npz.3’\n",
      "\n",
      "radiopedia_train_se 100%[===================>]  97.81M  7.91MB/s    in 12s     \n",
      "\n",
      "2023-09-23 12:32:46 (7.90 MB/s) - ‘radiopedia_train_segm_dataset.npz.3’ saved [102561953/102561953]\n",
      "\n",
      "--2023-09-23 12:32:46--  https://github.com/iburenko/hse_course/raw/main/week3/data/radiopedia_val_segm_dataset.npz\n",
      "Resolving github.com (github.com)... 140.82.121.3\n",
      "Connecting to github.com (github.com)|140.82.121.3|:443... connected.\n",
      "HTTP request sent, awaiting response... 302 Found\n",
      "Location: https://media.githubusercontent.com/media/iburenko/hse_course/main/week3/data/radiopedia_val_segm_dataset.npz [following]\n",
      "--2023-09-23 12:32:47--  https://media.githubusercontent.com/media/iburenko/hse_course/main/week3/data/radiopedia_val_segm_dataset.npz\n",
      "Resolving media.githubusercontent.com (media.githubusercontent.com)... 185.199.109.133, 185.199.110.133, 185.199.108.133, ...\n",
      "Connecting to media.githubusercontent.com (media.githubusercontent.com)|185.199.109.133|:443... connected.\n",
      "HTTP request sent, awaiting response... 200 OK\n",
      "Length: 17102978 (16M) [application/octet-stream]\n",
      "Saving to: ‘radiopedia_val_segm_dataset.npz.3’\n",
      "\n",
      "radiopedia_val_segm 100%[===================>]  16.31M  9.38MB/s    in 1.7s    \n",
      "\n",
      "2023-09-23 12:32:49 (9.38 MB/s) - ‘radiopedia_val_segm_dataset.npz.3’ saved [17102978/17102978]\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "# Download predefined datasets\n",
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "\n",
    "!wget https://github.com/iburenko/hse_course/raw/main/week3/data/corona_train_segm_dataset.npz\n",
    "!wget https://github.com/iburenko/hse_course/raw/main/week3/data/corona_val_segm_dataset.npz\n",
    "!wget https://github.com/iburenko/hse_course/raw/main/week3/data/radiopedia_train_segm_dataset.npz\n",
    "!wget https://github.com/iburenko/hse_course/raw/main/week3/data/radiopedia_val_segm_dataset.npz"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "id": "Uk6BK-XYB9UZ"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Defaulting to user installation because normal site-packages is not writeable\n",
      "Requirement already satisfied: segmentation_models_pytorch in /home/st235/.local/lib/python3.10/site-packages (0.3.3)\n",
      "Requirement already satisfied: torchvision>=0.5.0 in /home/st235/.local/lib/python3.10/site-packages (from segmentation_models_pytorch) (0.15.2)\n",
      "Requirement already satisfied: pretrainedmodels==0.7.4 in /home/st235/.local/lib/python3.10/site-packages (from segmentation_models_pytorch) (0.7.4)\n",
      "Requirement already satisfied: efficientnet-pytorch==0.7.1 in /home/st235/.local/lib/python3.10/site-packages (from segmentation_models_pytorch) (0.7.1)\n",
      "Requirement already satisfied: timm==0.9.2 in /home/st235/.local/lib/python3.10/site-packages (from segmentation_models_pytorch) (0.9.2)\n",
      "Requirement already satisfied: tqdm in /home/st235/.local/lib/python3.10/site-packages (from segmentation_models_pytorch) (4.66.1)\n",
      "Requirement already satisfied: pillow in /home/st235/.local/lib/python3.10/site-packages (from segmentation_models_pytorch) (10.0.0)\n",
      "Requirement already satisfied: torch in /home/st235/.local/lib/python3.10/site-packages (from efficientnet-pytorch==0.7.1->segmentation_models_pytorch) (2.0.1)\n",
      "Requirement already satisfied: munch in /home/st235/.local/lib/python3.10/site-packages (from pretrainedmodels==0.7.4->segmentation_models_pytorch) (4.0.0)\n",
      "Requirement already satisfied: pyyaml in /usr/lib/python3/dist-packages (from timm==0.9.2->segmentation_models_pytorch) (5.4.1)\n",
      "Requirement already satisfied: huggingface-hub in /home/st235/.local/lib/python3.10/site-packages (from timm==0.9.2->segmentation_models_pytorch) (0.17.1)\n",
      "Requirement already satisfied: safetensors in /home/st235/.local/lib/python3.10/site-packages (from timm==0.9.2->segmentation_models_pytorch) (0.3.3)\n",
      "Requirement already satisfied: numpy in /home/st235/.local/lib/python3.10/site-packages (from torchvision>=0.5.0->segmentation_models_pytorch) (1.25.2)\n",
      "Requirement already satisfied: requests in /home/st235/.local/lib/python3.10/site-packages (from torchvision>=0.5.0->segmentation_models_pytorch) (2.31.0)\n",
      "Requirement already satisfied: filelock in /home/st235/.local/lib/python3.10/site-packages (from torch->efficientnet-pytorch==0.7.1->segmentation_models_pytorch) (3.12.4)\n",
      "Requirement already satisfied: typing-extensions in /home/st235/.local/lib/python3.10/site-packages (from torch->efficientnet-pytorch==0.7.1->segmentation_models_pytorch) (4.7.1)\n",
      "Requirement already satisfied: sympy in /home/st235/.local/lib/python3.10/site-packages (from torch->efficientnet-pytorch==0.7.1->segmentation_models_pytorch) (1.12)\n",
      "Requirement already satisfied: networkx in /home/st235/.local/lib/python3.10/site-packages (from torch->efficientnet-pytorch==0.7.1->segmentation_models_pytorch) (3.1)\n",
      "Requirement already satisfied: jinja2 in /home/st235/.local/lib/python3.10/site-packages (from torch->efficientnet-pytorch==0.7.1->segmentation_models_pytorch) (3.1.2)\n",
      "Requirement already satisfied: nvidia-cuda-nvrtc-cu11==11.7.99 in /home/st235/.local/lib/python3.10/site-packages (from torch->efficientnet-pytorch==0.7.1->segmentation_models_pytorch) (11.7.99)\n",
      "Requirement already satisfied: nvidia-cuda-runtime-cu11==11.7.99 in /home/st235/.local/lib/python3.10/site-packages (from torch->efficientnet-pytorch==0.7.1->segmentation_models_pytorch) (11.7.99)\n",
      "Requirement already satisfied: nvidia-cuda-cupti-cu11==11.7.101 in /home/st235/.local/lib/python3.10/site-packages (from torch->efficientnet-pytorch==0.7.1->segmentation_models_pytorch) (11.7.101)\n",
      "Requirement already satisfied: nvidia-cudnn-cu11==8.5.0.96 in /home/st235/.local/lib/python3.10/site-packages (from torch->efficientnet-pytorch==0.7.1->segmentation_models_pytorch) (8.5.0.96)\n",
      "Requirement already satisfied: nvidia-cublas-cu11==11.10.3.66 in /home/st235/.local/lib/python3.10/site-packages (from torch->efficientnet-pytorch==0.7.1->segmentation_models_pytorch) (11.10.3.66)\n",
      "Requirement already satisfied: nvidia-cufft-cu11==10.9.0.58 in /home/st235/.local/lib/python3.10/site-packages (from torch->efficientnet-pytorch==0.7.1->segmentation_models_pytorch) (10.9.0.58)\n",
      "Requirement already satisfied: nvidia-curand-cu11==10.2.10.91 in /home/st235/.local/lib/python3.10/site-packages (from torch->efficientnet-pytorch==0.7.1->segmentation_models_pytorch) (10.2.10.91)\n",
      "Requirement already satisfied: nvidia-cusolver-cu11==11.4.0.1 in /home/st235/.local/lib/python3.10/site-packages (from torch->efficientnet-pytorch==0.7.1->segmentation_models_pytorch) (11.4.0.1)\n",
      "Requirement already satisfied: nvidia-cusparse-cu11==11.7.4.91 in /home/st235/.local/lib/python3.10/site-packages (from torch->efficientnet-pytorch==0.7.1->segmentation_models_pytorch) (11.7.4.91)\n",
      "Requirement already satisfied: nvidia-nccl-cu11==2.14.3 in /home/st235/.local/lib/python3.10/site-packages (from torch->efficientnet-pytorch==0.7.1->segmentation_models_pytorch) (2.14.3)\n",
      "Requirement already satisfied: nvidia-nvtx-cu11==11.7.91 in /home/st235/.local/lib/python3.10/site-packages (from torch->efficientnet-pytorch==0.7.1->segmentation_models_pytorch) (11.7.91)\n",
      "Requirement already satisfied: triton==2.0.0 in /home/st235/.local/lib/python3.10/site-packages (from torch->efficientnet-pytorch==0.7.1->segmentation_models_pytorch) (2.0.0)\n",
      "Requirement already satisfied: setuptools in /home/st235/.local/lib/python3.10/site-packages (from nvidia-cublas-cu11==11.10.3.66->torch->efficientnet-pytorch==0.7.1->segmentation_models_pytorch) (68.2.2)\n",
      "Requirement already satisfied: wheel in /home/st235/.local/lib/python3.10/site-packages (from nvidia-cublas-cu11==11.10.3.66->torch->efficientnet-pytorch==0.7.1->segmentation_models_pytorch) (0.41.2)\n",
      "Requirement already satisfied: cmake in /home/st235/.local/lib/python3.10/site-packages (from triton==2.0.0->torch->efficientnet-pytorch==0.7.1->segmentation_models_pytorch) (3.27.4.1)\n",
      "Requirement already satisfied: lit in /home/st235/.local/lib/python3.10/site-packages (from triton==2.0.0->torch->efficientnet-pytorch==0.7.1->segmentation_models_pytorch) (16.0.6)\n",
      "Requirement already satisfied: fsspec in /home/st235/.local/lib/python3.10/site-packages (from huggingface-hub->timm==0.9.2->segmentation_models_pytorch) (2023.9.1)\n",
      "Requirement already satisfied: packaging>=20.9 in /home/st235/.local/lib/python3.10/site-packages (from huggingface-hub->timm==0.9.2->segmentation_models_pytorch) (23.1)\n",
      "Requirement already satisfied: charset-normalizer<4,>=2 in /home/st235/.local/lib/python3.10/site-packages (from requests->torchvision>=0.5.0->segmentation_models_pytorch) (3.2.0)\n",
      "Requirement already satisfied: idna<4,>=2.5 in /home/st235/.local/lib/python3.10/site-packages (from requests->torchvision>=0.5.0->segmentation_models_pytorch) (3.4)\n",
      "Requirement already satisfied: urllib3<3,>=1.21.1 in /home/st235/.local/lib/python3.10/site-packages (from requests->torchvision>=0.5.0->segmentation_models_pytorch) (2.0.4)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in /home/st235/.local/lib/python3.10/site-packages (from requests->torchvision>=0.5.0->segmentation_models_pytorch) (2023.7.22)\n",
      "Requirement already satisfied: MarkupSafe>=2.0 in /home/st235/.local/lib/python3.10/site-packages (from jinja2->torch->efficientnet-pytorch==0.7.1->segmentation_models_pytorch) (2.1.3)\n",
      "Requirement already satisfied: mpmath>=0.19 in /home/st235/.local/lib/python3.10/site-packages (from sympy->torch->efficientnet-pytorch==0.7.1->segmentation_models_pytorch) (1.3.0)\n"
     ]
    }
   ],
   "source": [
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "# Install required packages into the environment\n",
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "\n",
    "!pip install segmentation_models_pytorch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "id": "2tfurSvD477K"
   },
   "outputs": [],
   "source": [
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "# Import needed packages into the environment\n",
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "from torch.utils.data import Dataset, DataLoader\n",
    "\n",
    "from tqdm.notebook import tqdm\n",
    "\n",
    "import torch\n",
    "import numpy as np\n",
    "\n",
    "from segmentation_models_pytorch import Unet\n",
    "from segmentation_models_pytorch.losses import DiceLoss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'cuda'"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "device = 'cuda' if torch.cuda.is_available() else 'cpu'\n",
    "device"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "KOzsPn8fkWTk"
   },
   "source": [
    "# Load data into the environment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Returns basic statistic for the raw dataset.\n",
    "# Pre-calculated once to save time during training/validation.\n",
    "def calculate_statistics(raw_dataset):\n",
    "    images = raw_dataset['images']\n",
    "    min, max = images.min(), images.max()\n",
    "    mean, std = images.mean(), images.std()\n",
    "\n",
    "    return {\n",
    "        'min': min,\n",
    "        'max': max,\n",
    "        'mean': mean,\n",
    "        'std': std,\n",
    "    }"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def merge_datasets(dataset_a, dataset_b):\n",
    "    images_a = dataset_a['images']\n",
    "    masks_a = dataset_a['masks']\n",
    "    images_b = dataset_b['images']\n",
    "    masks_b = dataset_b['masks']\n",
    "\n",
    "    return {\n",
    "        'images': np.concatenate((images_a, images_b), axis=-1),\n",
    "        'masks': np.concatenate((masks_a, masks_b), axis=-1),\n",
    "    }"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "id": "GIedGJl_h1V0"
   },
   "outputs": [],
   "source": [
    "corona_train = np.load('corona_train_segm_dataset.npz')\n",
    "corona_val = np.load('corona_val_segm_dataset.npz')\n",
    "radio_train = np.load('radiopedia_train_segm_dataset.npz')\n",
    "radio_val = np.load('radiopedia_val_segm_dataset.npz')\n",
    "all_train = merge_datasets(corona_train, radio_train)\n",
    "all_val = merge_datasets(corona_val, radio_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Pre-calculated dataset stats.\n",
    "corona_train_stats = calculate_statistics(corona_train)\n",
    "corona_val_stats = calculate_statistics(corona_val)\n",
    "radio_train_stats = calculate_statistics(radio_train)\n",
    "radio_val_stats = calculate_statistics(radio_val)\n",
    "all_train_stats = calculate_statistics(all_train)\n",
    "all_val_stats = calculate_statistics(all_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "import random\n",
    "\n",
    "def show_stats(raw_dataset):\n",
    "    images, masks = raw_dataset['images'], raw_dataset['masks']\n",
    "    \n",
    "    assert images.shape == masks.shape\n",
    "\n",
    "    shape = images.shape\n",
    "    m = shape[-1]\n",
    "    \n",
    "    sample = random.randint(0, m)\n",
    "\n",
    "    print('Dataset shape:', shape)\n",
    "\n",
    "    print('Min image value', images.min())\n",
    "    print('Max image value', images.max())\n",
    "    print('Min mask value', masks.min())\n",
    "    print('Max mask value', masks.max())\n",
    "\n",
    "    print(f'Sample #{sample}')\n",
    "    _, axarr = plt.subplots(1,2)\n",
    "    axarr[0].imshow(images[:, :, sample])\n",
    "    axarr[1].imshow(masks[: , :, sample])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dataset shape: (224, 224, 2024)\n",
      "Min image value -1023.0\n",
      "Max image value 7314.4346\n",
      "Min mask value 0\n",
      "Max mask value 1\n",
      "Sample #422\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "show_stats(corona_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dataset shape: (224, 224, 801)\n",
      "Min image value 0.0\n",
      "Max image value 255.0\n",
      "Min mask value 0\n",
      "Max mask value 1\n",
      "Sample #151\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "show_stats(radio_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "f7z4d_l8kc6x"
   },
   "source": [
    "# Define Dataset class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Removes images with no instances in it.\n",
    "# Even if no classes are predicted dice score is\n",
    "# still 0 for such images which decrease the score.\n",
    "# As we are interested in segmentation quality\n",
    "# we can discard such examples.\n",
    "# Should be applied only during validation stage\n",
    "# as we still need them for training.\n",
    "def remove_images_without_positive_class(raw_dataset):\n",
    "    images = raw_dataset['images']\n",
    "    masks = raw_dataset['masks']\n",
    "\n",
    "    assert images.shape == masks.shape\n",
    "    \n",
    "    m = images.shape[-1]\n",
    "\n",
    "    elements_with_no_instances = masks.max(axis=(0,1)) == 0  \n",
    "    elements_with_instances = ~elements_with_no_instances\n",
    "\n",
    "    return {\n",
    "        'images': images[..., elements_with_instances],\n",
    "        'masks': masks[..., elements_with_instances],\n",
    "    }"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "id": "sMIshsi35zYv"
   },
   "outputs": [],
   "source": [
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "# Define a subclass of pytorch's Dataset class.\n",
    "# In this class we have to write code (at least) \n",
    "# for three ('virtual') functions:\n",
    "# __init__(self, *args, **kwargs)\n",
    "# __len__(self, *args, **kwargs)\n",
    "# __getitem__(self, idx, *args, **kwargs)\n",
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "\n",
    "\n",
    "class SegmentationDatasetNumpy(Dataset):\n",
    "    def __init__(self, train, dataset):\n",
    "        super().__init__()\n",
    "        assert train in ['train', 'val']\n",
    "        assert dataset in ['corona', 'radiopedia', 'all']\n",
    "        self.train = train\n",
    "        if self.train == 'train':\n",
    "            if dataset == 'corona':\n",
    "                data = corona_train\n",
    "            elif dataset == 'radiopedia':\n",
    "                data = radio_train\n",
    "            else:\n",
    "                data = all_train\n",
    "        else:\n",
    "            # val dataset.\n",
    "            if dataset == 'corona':\n",
    "                data = corona_val\n",
    "            elif dataset == 'radiopedia':\n",
    "                data = radio_val\n",
    "            else:\n",
    "                data = all_val \n",
    "            data = remove_images_without_positive_class(data)\n",
    "        self.all_data = data['images']\n",
    "        self.all_masks = data['masks']\n",
    "\n",
    "    def __len__(self):\n",
    "        return self.all_data.shape[-1]\n",
    "    \n",
    "    def __getitem__(self, idx):\n",
    "        return self.all_data[...,idx], self.all_masks[...,idx]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "UwtX2K8okhMc"
   },
   "source": [
    "# Initialize Dataset and DataLoader classes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "id": "fZkS0ppk6Ozz"
   },
   "outputs": [],
   "source": [
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "# initialize datasets for train/val corona/radiopedia data\n",
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "\n",
    "train_ds_corona = SegmentationDatasetNumpy('train', 'corona')\n",
    "val_ds_corona = SegmentationDatasetNumpy('val', 'corona')\n",
    "train_ds_radio = SegmentationDatasetNumpy('train', 'radiopedia')\n",
    "val_ds_radio = SegmentationDatasetNumpy('val', 'radiopedia')\n",
    "train_ds_corona_radio = SegmentationDatasetNumpy('train', 'all')\n",
    "val_ds_corona_radio = SegmentationDatasetNumpy('val', 'all')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "id": "xDds91bu5gPE"
   },
   "outputs": [],
   "source": [
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "# initialize DataLoader for train/val corona/radiopedia datasets\n",
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "\n",
    "train_loader_corona = DataLoader(train_ds_corona, batch_size=16, shuffle=True, num_workers=8) \n",
    "val_loader_corona = DataLoader(val_ds_corona, batch_size=16, shuffle=False, num_workers=8)\n",
    "train_loader_radio = DataLoader(train_ds_radio, batch_size=16, shuffle=True, num_workers=8) \n",
    "val_loader_radio = DataLoader(val_ds_radio, batch_size=16, shuffle=False, num_workers=8)\n",
    "train_loader_corona_radio = DataLoader(train_ds_corona_radio, batch_size=16, shuffle=True, num_workers=8) \n",
    "val_loader_corona_radio = DataLoader(val_ds_corona_radio, batch_size=16, shuffle=False, num_workers=8)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "PnGODKLZkmi9"
   },
   "source": [
    "# Define Dice score, the measure of the segmnetation quality"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "id": "peif1fYSlsQt"
   },
   "outputs": [],
   "source": [
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "# Define dice score for the positive class\n",
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "\n",
    "def calculate_dice(pred, y, eps=1e-8):\n",
    "    assert pred.shape == y.shape\n",
    "    assert y.min() == 0 and y.max() >= 0 and y.max() <= 1\n",
    "    \n",
    "    intersection = 2 * ((pred > 0) * y).sum(axis=(1,2,3))\n",
    "    union = ((pred > 0).int() + y).sum(axis=(1,2,3)) + eps\n",
    "    \n",
    "    return intersection / union"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "def visualise_inference(raw_dataset,\n",
    "                        samples,\n",
    "                        checkpoint_file='./checkpoint.pth'):\n",
    "    saved_model = load_model(checkpoint_file=checkpoint_file)\n",
    "    \n",
    "    print(f'Evaluating samples #{samples}')\n",
    "    images, masks = raw_dataset['images'], raw_dataset['masks']\n",
    "\n",
    "    stats = calculate_statistics(raw_dataset)\n",
    "    mean, std = stats['mean'], stats['std']\n",
    "\n",
    "    model.eval()\n",
    "    with torch.no_grad():\n",
    "            _, axarr = plt.subplots(len(samples),3)\n",
    "            for i, sample in enumerate(samples):\n",
    "                print(f'Inference for sample #{sample}')\n",
    "                \n",
    "                inputs, labels = images[:, :, sample], masks[:, :, sample]\n",
    "                inputs, labels = torch.Tensor(inputs).to(device), torch.Tensor(labels).to(device)\n",
    "                inputs, labels = inputs.unsqueeze(0), labels.unsqueeze(0)\n",
    "\n",
    "                inputs, labels = inputs.unsqueeze(1), labels.unsqueeze(1)\n",
    "                inputs = (inputs - mean) / (std + 1e-8)\n",
    "    \n",
    "                predictions = saved_model(inputs)\n",
    "                probabilities = torch.sigmoid(predictions).cpu().numpy()\n",
    "    \n",
    "                loss = criterion(predictions, labels.float())\n",
    "    \n",
    "                print(f'Loss: {loss.item()}')        \n",
    "                print(f'Dice: {calculate_dice(predictions.cpu(), labels.cpu()).item()}')\n",
    "    \n",
    "                axarr[i, 0].imshow(images[:, :, sample])\n",
    "                axarr[i, 1].imshow(masks[:, :, sample])\n",
    "                axarr[i, 2].imshow(probabilities[0, 0, :, :])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_model(checkpoint_file=None):\n",
    "    model = Unet(encoder_name='resnet34', encoder_weights=None, in_channels=1, classes=1)\n",
    "    if checkpoint_file is not None:\n",
    "        checkpoint = torch.load(checkpoint_file)\n",
    "        model.load_state_dict(checkpoint['model'])\n",
    "    model = model.to(device)\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "id": "ew-soN696udA"
   },
   "outputs": [],
   "source": [
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "# Define the network, criterion (loss function) and the optimizer\n",
    "# BEWARE! The choice of hyperparameters is very important!\n",
    "# Try different values for the learning rate (1e-3, 1e-4, 1e-5, ...);\n",
    "# Feel free to change the optimizer, you could try different encoder or\n",
    "# use Unet++ or other versions of Unet.\n",
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "\n",
    "model = load_model()\n",
    "\n",
    "criterion = DiceLoss('binary')\n",
    "optimizer = torch.optim.Adam(model.parameters(), lr=1e-3, weight_decay=1e-3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Train/val loop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "# Write the code for train and validation loops.\n",
    "# Pass a dataloader as an argument to train/val functions.\n",
    "# You could pass any number of arguments to train and val functions in\n",
    "# addition to the dataloader.\n",
    "# train function should return the mean loss over the entire dataset;\n",
    "# val function should return two parameters: mean dice score and mean loss \n",
    "# (over the entire validation dataset).\n",
    "#\n",
    "# If you feel confident, you can rewrite given code and change train/val\n",
    "# function (say, make them methods of some class). This is not necessary\n",
    "# and is absolutely up to you. If you change the code of train/val function\n",
    "# you should check that everything works correctly by yourself, \n",
    "# we won't do this!\n",
    "# We will just run the entire notebook and check cells' outputs.\n",
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "\n",
    "def train(train_loader, stats, *args, **kwargs):\n",
    "    model.train()\n",
    "\n",
    "    mean, std = stats['mean'], stats['std']\n",
    "    \n",
    "    losses = []\n",
    "\n",
    "    for i, data in enumerate(tqdm(train_loader)):\n",
    "        inputs, labels = data\n",
    "        inputs, labels = inputs.to(device), labels.to(device)\n",
    "        inputs, labels = inputs.unsqueeze(1), labels.unsqueeze(1)\n",
    "        inputs = (inputs - mean) / (std + 1e-8)\n",
    "\n",
    "        optimizer.zero_grad()  \n",
    "        \n",
    "        predictions = model(inputs)\n",
    " \n",
    "        loss = criterion(predictions, labels.float())\n",
    "        loss.backward()\n",
    "        \n",
    "        optimizer.step()\n",
    "\n",
    "        losses.append(loss.item())\n",
    "        \n",
    "    mean_loss = np.mean(losses)\n",
    "    return mean_loss\n",
    "        \n",
    "def validation(val_loader, stats, *args, **kwargs):\n",
    "    model.eval()\n",
    "\n",
    "    mean, std = stats['mean'], stats['std']\n",
    "\n",
    "    losses = []\n",
    "    dices = []\n",
    "\n",
    "    with torch.no_grad():\n",
    "        for i, data in enumerate(tqdm(val_loader)):\n",
    "            inputs, labels = data\n",
    "            inputs, labels = inputs.to(device), labels.to(device)\n",
    "            inputs, labels = inputs.unsqueeze(1), labels.unsqueeze(1)\n",
    "            inputs = (inputs - mean) / (std + 1e-8)\n",
    "\n",
    "            predictions = model(inputs)\n",
    "\n",
    "            loss = criterion(predictions, labels.float())\n",
    "\n",
    "            losses.append(loss.item())\n",
    "            dices.extend(calculate_dice(predictions.cpu(), labels.cpu()).tolist())\n",
    "\n",
    "    mean_dice = np.mean(dices)\n",
    "    mean_loss = np.mean(losses)\n",
    "    return mean_dice, mean_loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "id": "-UzhwkjC6kFF"
   },
   "outputs": [],
   "source": [
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "# Define train/val loop\n",
    "# \n",
    "# For each epoch save the mean value of the loss function for the train loop.\n",
    "#\n",
    "# After each train loop save the mean values of the loss function \n",
    "# and dice score for the dataset which you ***did use*** to train \n",
    "# the neural network.\n",
    "#\n",
    "# After each train loop save the mean values of the loss function \n",
    "# and dice score for the dataset which you ***did not use*** to train \n",
    "# the neural network.\n",
    "# \n",
    "# Save the model with the best result obtained on a validation\n",
    "# dataset (the same as was used for training).\n",
    "#\n",
    "# Hint: do not forget to normalize data!\n",
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "\n",
    "def perform_training(train_loader,\n",
    "                     train_stats,\n",
    "                     val_domain_loader, \n",
    "                     val_domain_stats,\n",
    "                     val_shift_loader,\n",
    "                     val_shift_stats,\n",
    "                     epochs=50,\n",
    "                     checkpoint_file='./checkpoint.pth'):\n",
    "    best_dice = 0.0\n",
    "    \n",
    "    train_losses = np.zeros(epochs)\n",
    "    val_losses = np.zeros(epochs)\n",
    "    val_dices_domain = np.zeros(epochs)\n",
    "    val_dices_shift = np.zeros(epochs)\n",
    "\n",
    "    for epoch in range(epochs):\n",
    "        print(f'Epoch {epoch}')\n",
    "    \n",
    "        train_loss = train(train_loader, train_stats)\n",
    "        train_losses[epoch] = train_loss\n",
    "    \n",
    "        domain_dice, domain_loss = validation(val_domain_loader, val_domain_stats)\n",
    "        val_losses[epoch] = domain_loss\n",
    "        val_dices_domain[epoch] = domain_dice\n",
    "\n",
    "        if val_shift_loader is not None:\n",
    "            shift_dice, shift_loss = validation(val_shift_loader, val_shift_stats)\n",
    "            val_dices_shift[epoch] = shift_dice\n",
    "    \n",
    "        print(f'Train loss: {train_loss:.3f}')\n",
    "        print(f'Domain loss: {domain_loss:.3f} dice: {domain_dice:.4f}')\n",
    "\n",
    "        if val_shift_loader is not None:\n",
    "            print(f'Shift loss: {shift_loss:.3f} dice: {shift_dice:.4f}')\n",
    "\n",
    "        if domain_dice > best_dice:\n",
    "            best_dice = domain_dice\n",
    "\n",
    "            checkpoint = {\n",
    "                'epoch': epoch,\n",
    "                'best_dice': best_dice,\n",
    "                'model': model.state_dict(),\n",
    "            }\n",
    "\n",
    "            if checkpoint_file is not None:\n",
    "                print('********* New dice found, saving... *********')\n",
    "                torch.save(checkpoint, checkpoint_file)   \n",
    "\n",
    "    return train_losses, val_losses, val_dices_domain, val_dices_shift"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Reset model and optimizer to \"clear\"\n",
    "# state of the previous experiment.\n",
    "model = load_model()\n",
    "\n",
    "criterion = DiceLoss('binary')\n",
    "optimizer = torch.optim.Adam(model.parameters(), lr=1e-3, weight_decay=1e-3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8cfcd24a1ab54555a7aeb631c5ba61a6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "de3432d80104454bb03b86f4702692c1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "26ed040735f24b2f84eb1ca97ae308f4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.580\n",
      "Domain loss: 0.405 dice: 0.5782\n",
      "Shift loss: 0.414 dice: 0.5392\n",
      "********* New dice found, saving... *********\n",
      "Epoch 1\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0816a1aba6a34e03b56e8a808a73153d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6798ec51a12340588af458b0a5869be3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4e06535ecb5b4ffb98a5cf35339ce709",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.247\n",
      "Domain loss: 0.360 dice: 0.6144\n",
      "Shift loss: 0.400 dice: 0.5526\n",
      "********* New dice found, saving... *********\n",
      "Epoch 2\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "de83d420e5bf4b829fe0d0e29dc9b9a6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e7d0b0ea660147b5986a26b8cd776a46",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c40903e9e28a42d88da914f010d670ac",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.227\n",
      "Domain loss: 0.429 dice: 0.5505\n",
      "Shift loss: 0.366 dice: 0.5888\n",
      "Epoch 3\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0377a2b1cff140b9973d4e59f4d8a93a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d3a479af1e84449fb8934f5fa37ebed6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b5f3c5768ca94929b36390c21e2248cc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.235\n",
      "Domain loss: 0.362 dice: 0.6202\n",
      "Shift loss: 0.400 dice: 0.5711\n",
      "********* New dice found, saving... *********\n",
      "Epoch 4\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0020e352dc4f478cac8f0f46d16ed8ff",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f89126d7fad24083a407cda84eeb6108",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cd7defd4bb0d470f9d1bca5d4be1512f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.205\n",
      "Domain loss: 0.332 dice: 0.6413\n",
      "Shift loss: 0.406 dice: 0.5556\n",
      "********* New dice found, saving... *********\n",
      "Epoch 5\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8d271ecb2528476b8efce69b54fcf4d2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1cb6a624710a47a4b64b848cbe713430",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e7bfe83bc2c34bbbbaaf4734b02e601e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.215\n",
      "Domain loss: 0.389 dice: 0.5898\n",
      "Shift loss: 0.396 dice: 0.5728\n",
      "Epoch 6\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6f4edb01192644bca7c34cf0b660319b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1a0da717478c44d2ae929ce4d89c7bf0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7a3fcb8d8ae5409fa424ee255612c8ed",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.197\n",
      "Domain loss: 0.336 dice: 0.6471\n",
      "Shift loss: 0.415 dice: 0.5599\n",
      "********* New dice found, saving... *********\n",
      "Epoch 7\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "711bfbcba57a423f81c244b28f3c5288",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "acfad81cbd8040388c7e8b0c179bf821",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "07554bbc6a0641bab18874d60ee49e4b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.194\n",
      "Domain loss: 0.356 dice: 0.6225\n",
      "Shift loss: 0.440 dice: 0.5425\n",
      "Epoch 8\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8e6b07c6cbd748a58ccb26ccc6b3b930",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0f5f52ee5a4745ac9a96c55f7e9bc7c8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "65765da6b6f744ce838d93108fd4c446",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.211\n",
      "Domain loss: 0.397 dice: 0.5782\n",
      "Shift loss: 0.350 dice: 0.5967\n",
      "Epoch 9\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b51419e082ca4d4293c9f2c321527929",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "edb73e19a9be4e028091cb6f955bd35d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "78bcab99a62e4cf9a6e07335b2818e1d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.195\n",
      "Domain loss: 0.347 dice: 0.6259\n",
      "Shift loss: 0.371 dice: 0.5848\n",
      "Epoch 10\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3898235eda5547a8b7c184e35741c579",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1086c5180e3e4021b2683261cfda00ee",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3603a5cf587042309a4f1f853849fc79",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.191\n",
      "Domain loss: 0.359 dice: 0.6220\n",
      "Shift loss: 0.464 dice: 0.5146\n",
      "Epoch 11\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a0c1bdb66ea847b985cba14cc68ac41a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a43d94fa7c3448ab9d77daabe288a563",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d1685cea3b49494e8a7445bdbca7d6cf",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.201\n",
      "Domain loss: 0.965 dice: 0.0352\n",
      "Shift loss: 0.402 dice: 0.5595\n",
      "Epoch 12\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3f6f3c1b4e8a44fdbf09e9f7648146a7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1ca784db794249ada6e615de57e799c5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "77de4582c98943a5a49be52db0397954",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.187\n",
      "Domain loss: 0.350 dice: 0.6216\n",
      "Shift loss: 0.324 dice: 0.6287\n",
      "Epoch 13\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7ccf36910b75476e8a528f3ef67fb486",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3f56f00cd37f45d2987366a8403c67de",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e15e6f1ea870476e92eb2195bdc85e2a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.180\n",
      "Domain loss: 0.318 dice: 0.6515\n",
      "Shift loss: 0.376 dice: 0.5900\n",
      "********* New dice found, saving... *********\n",
      "Epoch 14\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "48794c5958f7468dbc27459bdebf0ae1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d3d5a7cc06da4f29b0e5253777a17941",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2dbe863aa06c482892a6b63625fe47b5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.170\n",
      "Domain loss: 0.383 dice: 0.5942\n",
      "Shift loss: 0.438 dice: 0.5375\n",
      "Epoch 15\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "70e6d62534e5412a89f9625de727cd4b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f084eddd137c4a63b835030a31b1cb3f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "94f50750aab04488b6a905e315a02b05",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.175\n",
      "Domain loss: 0.383 dice: 0.5940\n",
      "Shift loss: 0.334 dice: 0.6241\n",
      "Epoch 16\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b3ddb27193c84c71885fc6ea7400e5be",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "45902b78d3c14650aeffe6fa3affba3f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e197d6bba6e04a839153df242182b22a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.170\n",
      "Domain loss: 0.334 dice: 0.6376\n",
      "Shift loss: 0.358 dice: 0.6145\n",
      "Epoch 17\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "88abf55168844c4787b3a828ad528c4c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b557d0f0716d4fc3844fd9649ab1bf5c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "050c6febc1564e4dbec444189131af3f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.181\n",
      "Domain loss: 0.336 dice: 0.6381\n",
      "Shift loss: 0.426 dice: 0.5444\n",
      "Epoch 18\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a7f5a2a3591a4db5ae98ea75b2d8f381",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ee097092c9964e6cb6afecd1cd4519c5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "851d1b1932344877a148f94773bac413",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.167\n",
      "Domain loss: 0.307 dice: 0.6627\n",
      "Shift loss: 0.347 dice: 0.6211\n",
      "********* New dice found, saving... *********\n",
      "Epoch 19\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4065c5f01604446495e8f258593e469d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "51da739da6e149099bc0cbfbbf55ee75",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7717da8bf43841499e3b33212a3bdef4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.165\n",
      "Domain loss: 0.335 dice: 0.6405\n",
      "Shift loss: 0.376 dice: 0.5976\n",
      "Epoch 20\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "eeb772c449a3488691bd48698a45bc52",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7c26e2eecac34ab9b92ecfb580d9868f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "34e3841f3ca1461f97bd86855e9a9276",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.160\n",
      "Domain loss: 0.381 dice: 0.5932\n",
      "Shift loss: 0.382 dice: 0.5845\n",
      "Epoch 21\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4469ffb7bab0458991d6330073d64ee8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fcfd289686f94eda9a7ccec9d19b27fb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0e782d7273fd4d9ea697c67d4cc11c4d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.177\n",
      "Domain loss: 0.613 dice: 0.3692\n",
      "Shift loss: 0.576 dice: 0.4019\n",
      "Epoch 22\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e6fda161c8244c4eb819bf998924c5be",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ffcd572643f34712a2f1575190ae3f94",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "929bcfd1c0324d5aaf2515adbfc0a2af",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.167\n",
      "Domain loss: 0.273 dice: 0.7029\n",
      "Shift loss: 0.303 dice: 0.6492\n",
      "********* New dice found, saving... *********\n",
      "Epoch 23\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "75fc8dcb7881450cabbbd7ff62e3cd7b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d25193f8acea4c4bb4cfb6173f9a5f04",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "308e253ef93945c2a683560054e45caf",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.158\n",
      "Domain loss: 0.311 dice: 0.6629\n",
      "Shift loss: 0.407 dice: 0.5631\n",
      "Epoch 24\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7eab6c6a30e94110a5170a98869ee5c3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b8f06bb0ee8842b884c96a211924e3d9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "51780c04aa92436abcbcc6b4b05004d2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.168\n",
      "Domain loss: 0.297 dice: 0.6763\n",
      "Shift loss: 0.388 dice: 0.5708\n",
      "Epoch 25\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dba2951920a74f6d9fcc259da7d8c80d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "198dece001924f8cb46341ef2bd1cb3b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5a5a4eefff644b0aaaec088cab63fab1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.159\n",
      "Domain loss: 0.331 dice: 0.6443\n",
      "Shift loss: 0.364 dice: 0.6058\n",
      "Epoch 26\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2db10eaf517d4ff6b21244cce522315f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4abe5051c15b487a9789ab2f331cc6a2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e9d0e690464945f094c01329d87b0a42",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.158\n",
      "Domain loss: 0.340 dice: 0.6325\n",
      "Shift loss: 0.380 dice: 0.5956\n",
      "Epoch 27\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3a06abba8dc445af8621cd458387b445",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7778f2a8d9c04751939af7e03465e0d2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2a969d49ae534d3a893539aabb68637f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.182\n",
      "Domain loss: 0.341 dice: 0.6289\n",
      "Shift loss: 0.408 dice: 0.5656\n",
      "Epoch 28\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b8f8f79a11fd477a94e79a67c3aab0ac",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "267115406ed846c4ab7512ca709baf01",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "69c4c67aceeb487c818c4f752a331f4a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.152\n",
      "Domain loss: 0.323 dice: 0.6599\n",
      "Shift loss: 0.402 dice: 0.5674\n",
      "Epoch 29\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dcf527dabc42472ebc04541c2cd4ceda",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "91029d4e84f9458f984c1b468e18da75",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "06388ca6a16c4c0f9d1a08a93bd32048",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.151\n",
      "Domain loss: 0.325 dice: 0.6505\n",
      "Shift loss: 0.332 dice: 0.6268\n",
      "Epoch 30\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b36084d9a8fd46bdb47093e2c24dea3e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8de7e8f06fa440a7a8a46e09312552c0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f4872a54bcda4a9e8c0bfc87c2ff242a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.148\n",
      "Domain loss: 0.362 dice: 0.6128\n",
      "Shift loss: 0.305 dice: 0.6493\n",
      "Epoch 31\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c0dd3f87acfc4ae5b51456bbbe3f394e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e5b92262abf842e99f75aecde7c3ab5a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fbde4522c92f4c46943d234515c87827",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.146\n",
      "Domain loss: 0.416 dice: 0.5662\n",
      "Shift loss: 0.394 dice: 0.5676\n",
      "Epoch 32\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d3c483ad5abd41b884f1380d40345de2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d0af4903527d43739e562d233c29f024",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e0fe7da241e74a00a59053a1d3a6392c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.158\n",
      "Domain loss: 0.329 dice: 0.6400\n",
      "Shift loss: 0.389 dice: 0.5830\n",
      "Epoch 33\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0d3d078f8f0c42479f110125dca704f4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c3ac30bbdef54f24a298296072dcb1c5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b2d9af33e032433998189e9b9b6af533",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.145\n",
      "Domain loss: 0.307 dice: 0.6624\n",
      "Shift loss: 0.434 dice: 0.5486\n",
      "Epoch 34\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b59c5e6ea2354d1881c401102dff1549",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7e7152f904f94956bbc3fb134e5b4837",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1d15e9c44ac448c3bed65c8ee316e5c1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.144\n",
      "Domain loss: 0.337 dice: 0.6390\n",
      "Shift loss: 0.440 dice: 0.5291\n",
      "Epoch 35\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "84490a8293cd49f29c27e4b49633ec3c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "65730119daaf4efdbe6942006dfb8d9f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "abc88a792a2b44a2a7aed59ff21ee4a9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.154\n",
      "Domain loss: 0.322 dice: 0.6441\n",
      "Shift loss: 0.397 dice: 0.5783\n",
      "Epoch 36\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ca7fe1a7840649e6868625c804d99d86",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4fd4dfc0d74541bb8abc699723c5ddf4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0e98734bf9a94a63837465e078c04acb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.164\n",
      "Domain loss: 0.532 dice: 0.4613\n",
      "Shift loss: 0.378 dice: 0.5835\n",
      "Epoch 37\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5f46068a54be4e88a34fb87fc1c4b25f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "39df1ef5e29545f19317d911ea41e4b0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dec0c43fecd247bbb431e8c2cbc6bb08",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.143\n",
      "Domain loss: 0.330 dice: 0.6412\n",
      "Shift loss: 0.387 dice: 0.5761\n",
      "Epoch 38\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e79cd9e3642c4a85ab502df694258b3b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0fafc750be7e44a6bde2dfaa52a68679",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "303a4da3beee4a4cba4849a195313119",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.141\n",
      "Domain loss: 0.383 dice: 0.5949\n",
      "Shift loss: 0.449 dice: 0.5251\n",
      "Epoch 39\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6fd091da8d37401781a3f4a55cff5647",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "125c5d571e0e47a49ce153eaf19dcb72",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "28c184e57da744c7bb1e347dbfefb2ff",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.139\n",
      "Domain loss: 0.291 dice: 0.6793\n",
      "Shift loss: 0.370 dice: 0.5955\n",
      "Epoch 40\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c4e14e9b52f2497a97a9a59259b2aa8f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fd37115e7d5f4a9893fd1df428bc1da3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2b7d6a2458ba4ae48d1060076e5df57e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.143\n",
      "Domain loss: 0.333 dice: 0.6418\n",
      "Shift loss: 0.430 dice: 0.5539\n",
      "Epoch 41\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "afbf77c4d3a340c7a48ff734d53c8de0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ee9cb58f3f664e1a88fa3de532bd4c9f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6f6caffce76844429db881a72c5f9a44",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.138\n",
      "Domain loss: 0.374 dice: 0.5992\n",
      "Shift loss: 0.327 dice: 0.6253\n",
      "Epoch 42\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "955bac7a15464dff8150cf6542f0f08c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b2f324017e794021b254a1004f44b7cd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1740bd38e3384f1597166226ac6478cb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.150\n",
      "Domain loss: 0.380 dice: 0.5979\n",
      "Shift loss: 0.352 dice: 0.6103\n",
      "Epoch 43\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "61ce349114b4443a89b39e89bd441f4e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a5177d9a88994434a68086f6e5e6f283",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3211613955434b6189c1da44106e99a5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.139\n",
      "Domain loss: 0.471 dice: 0.5167\n",
      "Shift loss: 0.368 dice: 0.5857\n",
      "Epoch 44\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0d2ecc7d0f20408396eb12d7714132df",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "87d1abdf437b44fdb5475dd20669f5a7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "67fa979e1c9d4e798548236866fc0dd4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.148\n",
      "Domain loss: 0.320 dice: 0.6514\n",
      "Shift loss: 0.427 dice: 0.5531\n",
      "Epoch 45\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7f05a7cefa8242b1a83dff3e303afb90",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "62e7b6370b1f4d72a6931ed8dbe9ff12",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b38eecdbabca47a0984ddb5034b6c2cc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.141\n",
      "Domain loss: 0.331 dice: 0.6332\n",
      "Shift loss: 0.419 dice: 0.5561\n",
      "Epoch 46\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ec32cf5d485e46bb9f8fbe2985aa6a41",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "35fd00780a0c4232a0d4bf8f64f17eb1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9e88f9e78b3849139e8c85672ff11ab8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.138\n",
      "Domain loss: 0.315 dice: 0.6569\n",
      "Shift loss: 0.399 dice: 0.5723\n",
      "Epoch 47\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "51b009d89c0f4f39a8ee04806383fa10",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "60c8ec3bbf2446699dd02deb9ac3161b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4101725ca0b747e38c0c4f7a7f30420f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.135\n",
      "Domain loss: 0.327 dice: 0.6471\n",
      "Shift loss: 0.439 dice: 0.5332\n",
      "Epoch 48\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c37e3aac820e4b598e9a270b929f8d2c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ee00954c1d0c4526915a723e4c0e8c74",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ebaaa6aef800485e85cea45e9fa036c3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.140\n",
      "Domain loss: 0.316 dice: 0.6561\n",
      "Shift loss: 0.352 dice: 0.6071\n",
      "Epoch 49\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "765bc59c94bf45c0b3a153e8af9e460b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/127 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8515daa691cf4d9883f9abbce2452042",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6421cb390295432eaaa604eb26e83232",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.138\n",
      "Domain loss: 0.377 dice: 0.6067\n",
      "Shift loss: 0.308 dice: 0.6435\n"
     ]
    }
   ],
   "source": [
    "train_losses, val_losses, val_dices_domain, val_dices_shift = perform_training(train_loader=train_loader_corona,\n",
    "                                                                               train_stats=corona_train_stats,\n",
    "                                                                               val_domain_loader=val_loader_corona,\n",
    "                                                                               val_domain_stats=corona_val_stats,\n",
    "                                                                               val_shift_loader=val_loader_radio,\n",
    "                                                                               val_shift_stats=radio_val_stats,\n",
    "                                                                               epochs=50,\n",
    "                                                                               checkpoint_file='./train_corona_checkpoint.pth')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluating samples #[1, 183, 184, 412, 511, 555]\n",
      "Inference for sample #1\n",
      "Loss: 0.0\n",
      "Dice: 0.0\n",
      "Inference for sample #183\n",
      "Loss: 0.3746340274810791\n",
      "Dice: 0.6240713000297546\n",
      "Inference for sample #184\n",
      "Loss: 0.3856949210166931\n",
      "Dice: 0.6177884340286255\n",
      "Inference for sample #412\n",
      "Loss: 0.1817275881767273\n",
      "Dice: 0.819198489189148\n",
      "Inference for sample #511\n",
      "Loss: 0.12454557418823242\n",
      "Dice: 0.8802589178085327\n",
      "Inference for sample #555\n",
      "Loss: 0.0\n",
      "Dice: 0.0\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 18 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualise_inference(corona_val, samples=[1, 183, 184, 412, 511, 555], checkpoint_file='./train_corona_checkpoint.pth')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "id": "05-3Czwzs0v9"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best dice score 0.7028849429099371\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b2b378f419f948d198819854575105be",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Domain dice 0.7028849429099371, loss 0.2734685508828414\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1e85bdd2107446649ff99c30e3c57594",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shift dice 0.6491841837623209, loss 0.30343161310468403\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "# (maximum 2 points)\n",
    "# Output the best obtained dice score and \n",
    "# plot dice scores for corona and radiopedia validation datasets when trained\n",
    "# on ***corona*** dataset.\n",
    "# You should obtain (best) mean_dice (for corona dataset) > 0.6.\n",
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "\n",
    "# Generalises normal (neither good or bad) on radiopedia,\n",
    "# though loss is quite big.\n",
    "print(f\"Best dice score {val_dices_domain.max()}\")\n",
    "\n",
    "_, axises = plt.subplots(1,2)\n",
    "\n",
    "axises[0].plot(val_dices_domain)\n",
    "axises[1].plot(val_dices_shift)\n",
    "\n",
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "# (maximum 2 points)\n",
    "# Load the best model obtained while training \n",
    "# and run the validation loop one more time.\n",
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "\n",
    "model = load_model(checkpoint_file='./train_corona_checkpoint.pth')\n",
    "\n",
    "domain_dice, domain_loss = validation(val_loader_corona, corona_val_stats)\n",
    "\n",
    "print(f\"Domain dice {domain_dice}, loss {domain_loss}\")\n",
    "\n",
    "shift_dice, shift_loss = validation(val_loader_radio, radio_val_stats)\n",
    "\n",
    "print(f\"Shift dice {shift_dice}, loss {shift_loss}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Reset model and optimizer to \"clear\"\n",
    "# results of the previous experiment.\n",
    "model = load_model()\n",
    "\n",
    "criterion = DiceLoss('binary')\n",
    "optimizer = torch.optim.Adam(model.parameters(), lr=1e-3, weight_decay=1e-3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2eb58316516c4b2a8639d2a30ec80bb3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2e1e1de78cfe456f8966efdeea9e8a2f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c0f039f8597b4cf9b1fe512bd8197df9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.834\n",
      "Domain loss: 0.709 dice: 0.3258\n",
      "Shift loss: 0.969 dice: 0.0310\n",
      "********* New dice found, saving... *********\n",
      "Epoch 1\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "79d99a8557974de69036cf1cebf6d735",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "119a101699f748098eaadabbec9b1bbd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "76965402645d43ca993ba128162e9c99",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.445\n",
      "Domain loss: 0.571 dice: 0.3950\n",
      "Shift loss: 0.845 dice: 0.1424\n",
      "********* New dice found, saving... *********\n",
      "Epoch 2\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "353d5afba0f64fb390df608f83302cf6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c8527ed2b07a4f77a246de142e2fee4b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "93bd651a2dc84119a08c626b209e9a73",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.384\n",
      "Domain loss: 0.429 dice: 0.5298\n",
      "Shift loss: 0.703 dice: 0.2929\n",
      "********* New dice found, saving... *********\n",
      "Epoch 3\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d67550eae0c44de689e6cdace3e63f43",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ba400c3568794277808f3cc5a5115ed8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ad6bcfa12f96437aa8caed01afd38912",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.348\n",
      "Domain loss: 0.411 dice: 0.5378\n",
      "Shift loss: 0.697 dice: 0.2995\n",
      "********* New dice found, saving... *********\n",
      "Epoch 4\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1fafc164ac084f0b922e5afaa1760596",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "eb812db725ca4ca088a59539c6309f3b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7bd40c82ec884ecdb3efdb3312d4564a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.283\n",
      "Domain loss: 1.000 dice: 0.0000\n",
      "Shift loss: 1.000 dice: 0.0000\n",
      "Epoch 5\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fe6c29ad14324b859a109ac1087541b8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "939f68be38c14b64b2a970b37f3a6a9b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0d40e41832df44a39852c5db9f56d7c6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.328\n",
      "Domain loss: 0.363 dice: 0.5786\n",
      "Shift loss: 0.643 dice: 0.3420\n",
      "********* New dice found, saving... *********\n",
      "Epoch 6\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1e3c42b9dd05427d8e04844330b648e8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c330229738524c47aed53383cd9e4bb2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "375ed8347d7440c5a75805c4121a39c4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.266\n",
      "Domain loss: 0.539 dice: 0.4372\n",
      "Shift loss: 0.674 dice: 0.3334\n",
      "Epoch 7\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1b75131ca2ec46c78c7350060ec2b0a6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7e80e36a3a504ca5847a0d3a1efe16ee",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "484c0ae79794447f9e0220be5cb2b464",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.344\n",
      "Domain loss: 0.795 dice: 0.2060\n",
      "Shift loss: 0.999 dice: 0.0000\n",
      "Epoch 8\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "14d765ac521f47589f6abaad17acefff",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3e8899c03e64446c92de18d97b007fcc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f9f1836a813f4731a0c987fad810ef71",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.297\n",
      "Domain loss: 0.671 dice: 0.3148\n",
      "Shift loss: 0.909 dice: 0.0781\n",
      "Epoch 9\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8713dd6b0f48404591a2e7d56d266a0d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6bd86d0d1c304ed791682a8a4b9638c5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bf0227230cb442db939bb3d1e94f2950",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.287\n",
      "Domain loss: 0.514 dice: 0.4261\n",
      "Shift loss: 0.835 dice: 0.1458\n",
      "Epoch 10\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5419a76ca5f643f3a98ab9b11fed44f5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "91cd13045258487da22c2f1c6fd1634e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "58777338612c400ea986316a0ab48756",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.275\n",
      "Domain loss: 0.428 dice: 0.5180\n",
      "Shift loss: 0.645 dice: 0.3265\n",
      "Epoch 11\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ef007e172faf4db0ba5f35abc28f51eb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "10a42d3cb2574aa98f4212d6c6b15cf5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bb40e092346c46b9a10b815bc0c90800",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.245\n",
      "Domain loss: 0.322 dice: 0.6337\n",
      "Shift loss: 0.581 dice: 0.4019\n",
      "********* New dice found, saving... *********\n",
      "Epoch 12\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "26284375b0bb47279b9d5317858d908d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2e121c292759446fbff32b140bd15267",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "46c01ad0533443aead928c0756224fd5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.278\n",
      "Domain loss: 0.506 dice: 0.4531\n",
      "Shift loss: 0.825 dice: 0.1589\n",
      "Epoch 13\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4cb268c84cb94fcf9ed751bd2fcc88a0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3c6f75a9eefb473385f40b10375cd881",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "eb8a4fefff4346ba9ea90a2bac30746f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.248\n",
      "Domain loss: 0.475 dice: 0.4918\n",
      "Shift loss: 0.698 dice: 0.2834\n",
      "Epoch 14\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ca4ca1b03e7d44d5ae14e98809e91f8e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0041d0db70b34a619f232d004174c13b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "40106ccc08a340fe822d73811a6dd7b5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.240\n",
      "Domain loss: 0.708 dice: 0.2561\n",
      "Shift loss: 0.867 dice: 0.1104\n",
      "Epoch 15\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "320c61405864423c85f2ae84de7502e2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "03660bfd3caf4815ac8b51aaabb0767a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "869efc7dd2fe4214aaa0c040a7efee78",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.231\n",
      "Domain loss: 0.338 dice: 0.6136\n",
      "Shift loss: 0.708 dice: 0.2645\n",
      "Epoch 16\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "51b006b336a84bdbafc3607a0c97b5f5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "00eae5cc9d9545d48f115c9b4f7a2ecb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a16223d40cce425eb80540cbb58b8133",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.228\n",
      "Domain loss: 0.380 dice: 0.5420\n",
      "Shift loss: 0.709 dice: 0.2659\n",
      "Epoch 17\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4a536b2ec9c745c686e65ddef75b13f5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9840428756bc4f36b1dbfe0742bdc7c4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d2c59151498a4b709a1b26ed8302b53f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.212\n",
      "Domain loss: 0.349 dice: 0.5990\n",
      "Shift loss: 0.594 dice: 0.3772\n",
      "Epoch 18\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "198a5ef4319e4868a30302ee212f01a6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2f6ed01e5b3848609c69a1d536f12fb0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e9cf5385a9d44efd8cb384a162f68622",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.208\n",
      "Domain loss: 0.428 dice: 0.5172\n",
      "Shift loss: 0.619 dice: 0.3561\n",
      "Epoch 19\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6219dd8002a0477aa4371bfb7a27f9fa",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "012a04efa7c241948b012f17d4f1cfa8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "eac25c264d1048c9bea57aaf45472f94",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.198\n",
      "Domain loss: 0.305 dice: 0.6243\n",
      "Shift loss: 0.528 dice: 0.4410\n",
      "Epoch 20\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0b895453543744ba9ce030f7bf504617",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "49a746f8e23444979e6c9e17928e6fae",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c3db2bf297cb42baa9e1694aded07821",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.193\n",
      "Domain loss: 0.417 dice: 0.5165\n",
      "Shift loss: 0.735 dice: 0.2349\n",
      "Epoch 21\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "72d0e9976e7048439b1bc09096330562",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3b9fd5219063429da3dad516d84b92d1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f8b096e2a6bc4b8f87e3b34508aaf1ae",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.206\n",
      "Domain loss: 0.495 dice: 0.4433\n",
      "Shift loss: 0.865 dice: 0.1120\n",
      "Epoch 22\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "407a952e455b47daba21a858341bfdb8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cb71353090834e3085738633496873c8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ec753cb7ee6e46089efd99acc1bbaa54",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.222\n",
      "Domain loss: 0.456 dice: 0.4927\n",
      "Shift loss: 0.716 dice: 0.2597\n",
      "Epoch 23\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e48b903026b34059b2a0f98004456357",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "802bf6a50775446a9e27c90feefb0094",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "725171accc814e6cb7783ef39d49e09a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.236\n",
      "Domain loss: 0.345 dice: 0.6045\n",
      "Shift loss: 0.640 dice: 0.3379\n",
      "Epoch 24\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d37fee8bf8c24e469b6f3631d483ebf5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "db16137d476c4ca9abfb77996b022086",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "783809ed21234ebba04f8cd2b6f776eb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.232\n",
      "Domain loss: 0.445 dice: 0.4944\n",
      "Shift loss: 0.699 dice: 0.2689\n",
      "Epoch 25\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "64dc726a45914bf29f44aa531a4798b0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3eef3a755a254fc597b15f9d854cdc09",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3a822908aaf44bfd8207e705a4edadce",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.226\n",
      "Domain loss: 0.359 dice: 0.5851\n",
      "Shift loss: 0.577 dice: 0.4017\n",
      "Epoch 26\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ed00d712bccf491a9c5fbf4c351e7266",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e679adc53346496a8b3550798f4727ed",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2f3e12b01d7d49e1b353aa4a30e8831d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.219\n",
      "Domain loss: 0.301 dice: 0.6498\n",
      "Shift loss: 0.549 dice: 0.4243\n",
      "********* New dice found, saving... *********\n",
      "Epoch 27\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2125947348ad4857a25289b19c9872cb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1c7c5910e4874e56a7a6aa299e1d127e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "558fde468f7844049b69923ff549577d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.190\n",
      "Domain loss: 0.454 dice: 0.4932\n",
      "Shift loss: 0.637 dice: 0.3307\n",
      "Epoch 28\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "da1f7b44dc86496f920ceca6fa4aee00",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "aab00b4481364da593435ec6f5bf5cb0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6ab6835c070147c8a5f8a76cfaf768a1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.205\n",
      "Domain loss: 0.308 dice: 0.6344\n",
      "Shift loss: 0.552 dice: 0.4121\n",
      "Epoch 29\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f4c2c587eb1b426d9185c860ff60d509",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "07cd64b8964542cb87c1ee1fefa9e26c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d279df8d3ca44dfda064aeac34035175",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.213\n",
      "Domain loss: 1.000 dice: 0.0000\n",
      "Shift loss: 1.000 dice: 0.0000\n",
      "Epoch 30\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e7979723a9fd4474bbd960676c3a03f2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "46008bd8b2494222975d315b4d5bcd3e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a622500a8d034802b5af6a04db326b95",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.385\n",
      "Domain loss: 0.409 dice: 0.5404\n",
      "Shift loss: 0.747 dice: 0.2301\n",
      "Epoch 31\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2d2041bef7de4a208d2da158a6b181fc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b224cfecb6da42f89b4e1c37de578bd3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9af120e076c744599ed269a9004c5a96",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.351\n",
      "Domain loss: 0.442 dice: 0.4828\n",
      "Shift loss: 0.831 dice: 0.1456\n",
      "Epoch 32\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "152c6771897241bf84fdd1199ee25f6b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fdc5bffcd3c94af5801c55a0a4699402",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b7b608aa012c4623bbf5d7a0e7311b50",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.277\n",
      "Domain loss: 0.412 dice: 0.5214\n",
      "Shift loss: 0.767 dice: 0.2081\n",
      "Epoch 33\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3d2dbd729a54470a87826bf94445723a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "db0e942461444656a42a86807e8f3a2f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "098a1ed826ba42ed9c87fe79350ec84e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.270\n",
      "Domain loss: 0.358 dice: 0.5621\n",
      "Shift loss: 0.662 dice: 0.3066\n",
      "Epoch 34\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "535728fb14b04b899477ac2e9ada6661",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fd28fb44e58d45e68dfe4dc6921a91a6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "aa4f017776954ba5b628808e696ed784",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.234\n",
      "Domain loss: 0.528 dice: 0.4217\n",
      "Shift loss: 0.813 dice: 0.1651\n",
      "Epoch 35\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "38e7dd348f254dd4b7e894fae93c788c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "07fb2ef662624c4c9e32928f82041a5f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9777df51561142c4a5cc95680b42d65e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.238\n",
      "Domain loss: 0.438 dice: 0.4906\n",
      "Shift loss: 0.711 dice: 0.2591\n",
      "Epoch 36\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "611bbe866fa6401e8bfac07aa3ce2e68",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6313119a49b64e39ab43e99a8c861523",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "84fdab3c530840f8ae07fc70d73d3d62",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.229\n",
      "Domain loss: 0.327 dice: 0.6076\n",
      "Shift loss: 0.602 dice: 0.3692\n",
      "Epoch 37\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7ab072c311c04dd8b8fb4d5f46319a5e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "18ae6745c11747cbb9984347d0436d34",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e8991bc9c5a34c0394a466337d5f915f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.229\n",
      "Domain loss: 0.403 dice: 0.5254\n",
      "Shift loss: 0.645 dice: 0.3253\n",
      "Epoch 38\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3abea34f76b844809756b5634b0d9335",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b0ba1cfb04f84be5b1222804457948ec",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "37b1d0866ec14dc88ab5802761c6f342",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.201\n",
      "Domain loss: 0.384 dice: 0.5448\n",
      "Shift loss: 0.708 dice: 0.2641\n",
      "Epoch 39\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b44ca3fd41ea4eca8940d7e8f9290962",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c2632696de9a46a89fdc07dd8d5293ea",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9cf9ea6a89384575ade98a5aa00bbf26",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.201\n",
      "Domain loss: 0.405 dice: 0.5320\n",
      "Shift loss: 0.646 dice: 0.3222\n",
      "Epoch 40\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9f2e18aeeafe44b2a03b74064ef1aef2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "71eb00d6b6e74c598869795fb738f2d2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dfa7d73ec49e40d5a321c1a7879bb373",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.191\n",
      "Domain loss: 0.405 dice: 0.5343\n",
      "Shift loss: 0.514 dice: 0.4462\n",
      "Epoch 41\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "db7debc025464a45a0f4dbb80800ccbd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2d075d77fb6145bc87cf2ccde7fdf71c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9a37ca56da7447d9baf49072574957da",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.207\n",
      "Domain loss: 0.395 dice: 0.5374\n",
      "Shift loss: 0.567 dice: 0.3946\n",
      "Epoch 42\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d1566ab15ed54a8889869f81385c9575",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "216352b65d3b44d48169bd0494157f40",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a5946d95e803429ba391b0c257901598",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.195\n",
      "Domain loss: 0.321 dice: 0.6195\n",
      "Shift loss: 0.457 dice: 0.5037\n",
      "Epoch 43\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7eec0a31d70949b0ac4cae6a91989240",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "51516bd2b8df48579678bdc7255881c2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7f6ac2d854444d559a99d98fb2e1e074",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.183\n",
      "Domain loss: 0.373 dice: 0.5655\n",
      "Shift loss: 0.564 dice: 0.4012\n",
      "Epoch 44\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a53a1ae911da4138a0a7d86798dfcf4f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c4a205ee31e64da9994e922ddf8592fa",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b259b41369574a409ea02d4d05af1755",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.177\n",
      "Domain loss: 0.401 dice: 0.5313\n",
      "Shift loss: 0.619 dice: 0.3500\n",
      "Epoch 45\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "00eabe58ca4a4c9c9880cadc8e43cbad",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dcb6bc8610374c0eac54f3764667030e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5104af53b09547f0883516f4e5033e2b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.200\n",
      "Domain loss: 0.304 dice: 0.6357\n",
      "Shift loss: 0.512 dice: 0.4638\n",
      "Epoch 46\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a67030a01eda49929be40e549e5686a1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c336d9a1e3fe415bb0ff50482eaf5668",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5094be24861e4607b2c8b852d0dc4634",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.195\n",
      "Domain loss: 0.440 dice: 0.4927\n",
      "Shift loss: 0.678 dice: 0.2871\n",
      "Epoch 47\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "13725cba5f184583a3bcf15614803287",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "059971385c024706b5bc893e0ddc3be7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6ed09cda717a43f99dcfab75313bdb70",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.204\n",
      "Domain loss: 0.491 dice: 0.4410\n",
      "Shift loss: 0.725 dice: 0.2504\n",
      "Epoch 48\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6e0b39ffcbc24e1c8493150eaf442f67",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d4ecab1449fb4fdeaaab8b5f9853e72f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d474b2f2e1174f45bde75f7c964bf0ee",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.208\n",
      "Domain loss: 0.353 dice: 0.5938\n",
      "Shift loss: 0.639 dice: 0.3248\n",
      "Epoch 49\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2a452d30719d483e9aa6c9c7342ba799",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/51 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b48612584ba24ce2adada07e2ac19996",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1e81e4738a0b4ae29dd072aaf1c4dc86",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.195\n",
      "Domain loss: 0.363 dice: 0.5728\n",
      "Shift loss: 0.676 dice: 0.2896\n"
     ]
    }
   ],
   "source": [
    "train_losses, val_losses, val_dices_domain, val_dices_shift = perform_training(train_loader=train_loader_radio,\n",
    "                                                                               train_stats=radio_train_stats,\n",
    "                                                                               val_domain_loader=val_loader_radio,\n",
    "                                                                               val_domain_stats=radio_val_stats,\n",
    "                                                                               val_shift_loader=val_loader_corona,\n",
    "                                                                               val_shift_stats=corona_val_stats,\n",
    "                                                                               epochs=50,\n",
    "                                                                               checkpoint_file='./train_radio_checkpoint.pth')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluating samples #[1, 22, 50, 74, 98]\n",
      "Inference for sample #1\n",
      "Loss: 0.9998569488525391\n",
      "Dice: 0.0\n",
      "Inference for sample #22\n",
      "Loss: 0.7598164081573486\n",
      "Dice: 0.24583333730697632\n",
      "Inference for sample #50\n",
      "Loss: 0.0\n",
      "Dice: 0.0\n",
      "Inference for sample #74\n",
      "Loss: 0.5623503923416138\n",
      "Dice: 0.4378317594528198\n",
      "Inference for sample #98\n",
      "Loss: 0.3890116810798645\n",
      "Dice: 0.6130634546279907\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 15 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualise_inference(radio_val, samples=[1, 22, 50, 74, 98], checkpoint_file='./train_radio_checkpoint.pth')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "id": "2FX2ZOW8tIB0"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best dice score 0.6497524966538515\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cbdb3081365946a18c9b4d2382580e2d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Domain dice 0.6497524966538515, loss 0.3014616881098066\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ee51b520847a4e07ba97b2b762314d58",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shift dice 0.4242634596904456, loss 0.5494188854568883\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "# (maximum 2 points)\n",
    "# Output the best obtained dice score and \n",
    "# plot dice scores for corona and radiopedia validation datasets when trained\n",
    "# on ***radiopedia*** dataset\n",
    "# You should obtain (best) mean_dice (for radio dataset) > 0.3.\n",
    "#\n",
    "# ***NotaBene***: training on radiopedia is much less stable compared to\n",
    "# training on corona dataset. Nevertheless, you should be able to score \n",
    "# dice measure > 0.3.\n",
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "\n",
    "# Generalises really bad on corona if trained\n",
    "# on radiopedia.\n",
    "print(f\"Best dice score {val_dices_domain.max()}\")\n",
    "\n",
    "_, axises = plt.subplots(1,2)\n",
    "\n",
    "axises[0].plot(val_dices_domain)\n",
    "axises[1].plot(val_dices_shift)\n",
    "\n",
    "\n",
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "# (maximum 2 points)\n",
    "# Load the best model obtained while training \n",
    "# and run the validation loop one more time.\n",
    "# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n",
    "\n",
    "model = load_model(checkpoint_file='./train_radio_checkpoint.pth')\n",
    "\n",
    "domain_dice, domain_loss = validation(val_loader_radio, radio_val_stats)\n",
    "\n",
    "print(f\"Domain dice {domain_dice}, loss {domain_loss}\")\n",
    "\n",
    "shift_dice, shift_loss = validation(val_loader_corona, corona_val_stats)\n",
    "\n",
    "print(f\"Shift dice {shift_dice}, loss {shift_loss}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Reset model and optimizer to \"clear\n",
    "# results of the previous experiment.\n",
    "model = load_model()\n",
    "\n",
    "criterion = DiceLoss('binary')\n",
    "optimizer = torch.optim.Adam(model.parameters(), lr=1e-3, weight_decay=1e-3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "53a7018f32fb4c30b2a67a164c64c00b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7e2e410e05224feeba2332aeb107efd3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.702\n",
      "Domain loss: 0.610 dice: 0.3854\n",
      "********* New dice found, saving... *********\n",
      "Epoch 1\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "befba43e95ef48ebad89b60f4aca0f09",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ed10643405f0463fb7318b47081dafa9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.497\n",
      "Domain loss: 0.581 dice: 0.4129\n",
      "********* New dice found, saving... *********\n",
      "Epoch 2\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dfd9b2a594344e4d998dced3539672f2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cd099a2c21a64ff1994a2468198957ae",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.468\n",
      "Domain loss: 0.565 dice: 0.4333\n",
      "********* New dice found, saving... *********\n",
      "Epoch 3\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "054097e7580b4f39b652d640b7b392a4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f70156e0286c441c92313f057342c509",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.457\n",
      "Domain loss: 0.581 dice: 0.4133\n",
      "Epoch 4\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "04f28e4a97de406ebccf762a2ad1996a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "122c46d11cf340f39236d45eb5845459",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.454\n",
      "Domain loss: 1.000 dice: 0.0000\n",
      "Epoch 5\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9fadc9b33bd44b38855a42caf35e7cca",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1ddbaf3f56f44f238cb4de17e4f3ae76",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.445\n",
      "Domain loss: 0.697 dice: 0.3065\n",
      "Epoch 6\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1b10d98e15c74d01a9a15215431bf690",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "857c5c6672b14bd7b471fc3c2867bb94",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.452\n",
      "Domain loss: 0.638 dice: 0.3644\n",
      "Epoch 7\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7fd212cd9b454b499a717a4a4afc5412",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b41b717985a44c059295354660c51c2f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.448\n",
      "Domain loss: 0.599 dice: 0.3977\n",
      "Epoch 8\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bf1c9e8dca5841689163dca17b06ed26",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8579d907edcd43af983718f1e0d305de",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.445\n",
      "Domain loss: 0.627 dice: 0.3661\n",
      "Epoch 9\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "77db65ad9c1f47889c3ee6ba137a3e29",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "12356a8c04d24d40b7d294af66a900f0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.430\n",
      "Domain loss: 0.740 dice: 0.2610\n",
      "Epoch 10\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "195e53cec1cb4befa639a5339c51704e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6499ef3f289549dea3c9b3b7446496fa",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.429\n",
      "Domain loss: 0.548 dice: 0.4448\n",
      "********* New dice found, saving... *********\n",
      "Epoch 11\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7e46285b91df43be95f970257856cb69",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4010525d37844fd8bc11f4993f7587a6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.426\n",
      "Domain loss: 0.532 dice: 0.4588\n",
      "********* New dice found, saving... *********\n",
      "Epoch 12\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7cb161a226bf411ea196866c43a64eaa",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d1d09b9d30d6417590ed6903d6ac110f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.432\n",
      "Domain loss: 0.691 dice: 0.3106\n",
      "Epoch 13\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "40d7ae48e04741aaba1d931ea5ee6941",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e59cbe4f0f28447f8b1161e5228e14a2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.432\n",
      "Domain loss: 0.738 dice: 0.2634\n",
      "Epoch 14\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b5042987d030437882aff7c18fcd3a03",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "214f188aa7ad4b2fa4c2a1e0a80f1692",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.431\n",
      "Domain loss: 0.541 dice: 0.4510\n",
      "Epoch 15\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "86b44d882f7842bfa7f0cab06a07a4b3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f0947377b7dc4571a39774bcdbedce10",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.417\n",
      "Domain loss: 0.530 dice: 0.4608\n",
      "********* New dice found, saving... *********\n",
      "Epoch 16\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9257e0ca172d47008544c451c7af62bc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a63dbcdd23254ebbbe0ef550b196dbc5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.428\n",
      "Domain loss: 0.613 dice: 0.3875\n",
      "Epoch 17\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d756fa41f41d482ea4e50af83ae38622",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c9565635169c415aa0bfa9eae0af57b8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.409\n",
      "Domain loss: 0.533 dice: 0.4631\n",
      "********* New dice found, saving... *********\n",
      "Epoch 18\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "270b947781e94de19b15c1fc1052739e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bfae262e8bb74be1bf90e5df8e0d5db2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.418\n",
      "Domain loss: 0.565 dice: 0.4289\n",
      "Epoch 19\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "71f21e99480645e2abea34f5cf96af21",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "942001142f614fd78a873e015957615f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.431\n",
      "Domain loss: 0.530 dice: 0.4618\n",
      "Epoch 20\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8249b332360a4642aeb5f2588a889a25",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "097b401619a140c7b3b13d82aa5f560b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.423\n",
      "Domain loss: 0.528 dice: 0.4621\n",
      "Epoch 21\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b04db8b4fb574727819926a810be9481",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0abe26b4971445abb174cd786e4b68c3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.420\n",
      "Domain loss: 0.647 dice: 0.3530\n",
      "Epoch 22\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "04bbdfb4a09b42f682e10a9bbf390daf",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "428a79937906499fb529777260b4a91c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.398\n",
      "Domain loss: 0.511 dice: 0.4774\n",
      "********* New dice found, saving... *********\n",
      "Epoch 23\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4c0e6f5c35e94b9a982d6a5eb7812dba",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "aac8782d6d224551b42c3776a8418e74",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.407\n",
      "Domain loss: 0.620 dice: 0.3773\n",
      "Epoch 24\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "edd44313af7a4ef79abf3b48d332b272",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2d63e514ebe9462996d435f40765ff00",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.405\n",
      "Domain loss: 0.519 dice: 0.4660\n",
      "Epoch 25\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "59567475c598489db529a6c1ffb3b27f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7f61e797dca2498898419f9cc17b8dd7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.387\n",
      "Domain loss: 0.548 dice: 0.4458\n",
      "Epoch 26\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e12ae0b620894819922db1f4143c3b3a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a99517c070e8487f8b380c3f875751c6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.413\n",
      "Domain loss: 0.575 dice: 0.4193\n",
      "Epoch 27\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f09fc49f27de41e7b36c000e7308fe12",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1d9b02531d4d4f63ae25434e4574a750",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.412\n",
      "Domain loss: 0.569 dice: 0.4236\n",
      "Epoch 28\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f05587709c354c2181e50109e8ce9880",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "79502214ee864cceb588324104562535",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.402\n",
      "Domain loss: 0.509 dice: 0.4814\n",
      "********* New dice found, saving... *********\n",
      "Epoch 29\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2961dd795d4c4f8cb6dea81f3bfa20f3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "839165bafe004d0db47f502d01327694",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.401\n",
      "Domain loss: 0.656 dice: 0.3425\n",
      "Epoch 30\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cbf28727968b44e0a1874e7eef0579f6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7a023bf009dd485e9dcf17f6b0e323a5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.399\n",
      "Domain loss: 0.515 dice: 0.4770\n",
      "Epoch 31\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "03678add6d0149b2a2469d02f2803725",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1a02a16b82cc4ec493235a29b94f2031",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.392\n",
      "Domain loss: 0.645 dice: 0.3498\n",
      "Epoch 32\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5560c5000b794859a13b8957574495b0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2f1dd1eb40fa4aa58b3916c69f2c319e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.395\n",
      "Domain loss: 0.546 dice: 0.4485\n",
      "Epoch 33\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7f5f3400e2bd4290aa510fee42d6622a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dd8796b54ac647359009b75f042c72e8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.417\n",
      "Domain loss: 0.505 dice: 0.4903\n",
      "********* New dice found, saving... *********\n",
      "Epoch 34\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c0ded46dd7cd4a47aa945eb83ae57005",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6bd92da0a3f847a39a3df52f6795a0d8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.387\n",
      "Domain loss: 0.539 dice: 0.4541\n",
      "Epoch 35\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6f306d0411d540ae93bad4b061ce3b2b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "95c3651e684b48ed848ddd140819dae8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.387\n",
      "Domain loss: 0.585 dice: 0.4142\n",
      "Epoch 36\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1482f2026abc4031be2654bfaee0690b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "feb9a2e8847a4345a5f95bb9128cdf4a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.389\n",
      "Domain loss: 0.502 dice: 0.4884\n",
      "Epoch 37\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c42769e9a41248f7b92c30603d567c23",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a85e34ae5e90459b9fda0fce8e0df46d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.398\n",
      "Domain loss: 0.531 dice: 0.4604\n",
      "Epoch 38\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "708cc85826f24d90aebdb86ac02b79ce",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "17304d5f0e3e4463bb95866fcda368e3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.391\n",
      "Domain loss: 0.537 dice: 0.4586\n",
      "Epoch 39\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "373410e0baf74144b7d61b27abb608c8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a55f8f0be49e4f3b8a4e3d0f84d9eb99",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.390\n",
      "Domain loss: 0.536 dice: 0.4598\n",
      "Epoch 40\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5146496731794262b2f080094b63932c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ad8191480315494987066c3d65596866",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.386\n",
      "Domain loss: 0.507 dice: 0.4844\n",
      "Epoch 41\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8332df4a716249b7937820c5c688c1d5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1c14a3bcdb2b4d8999f19a648c5c0553",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.401\n",
      "Domain loss: 0.523 dice: 0.4667\n",
      "Epoch 42\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "202c49ec481a46ea8cc989e3b029bd81",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "96c426f60ad94e8aa9a735d7e400f2ce",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.381\n",
      "Domain loss: 0.543 dice: 0.4528\n",
      "Epoch 43\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f35173c97eb541758de5f1750c978d30",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ad670b1425464f6ab3b17093d974ded9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.394\n",
      "Domain loss: 0.551 dice: 0.4432\n",
      "Epoch 44\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7d5d25023b9e4e7c99064f36a234e8df",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "02ee69a76d324983bf0474b8f991352e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.387\n",
      "Domain loss: 0.513 dice: 0.4752\n",
      "Epoch 45\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "03731d6794be4e9ebd8461f60c3b6ced",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f09dd893b5794a659d9f31a530252b4a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.393\n",
      "Domain loss: 0.496 dice: 0.4950\n",
      "********* New dice found, saving... *********\n",
      "Epoch 46\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "513f0e89413b4e3b927d3c897b4bcfc4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6e84b9e695a64dcf921eb79e5574b03f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.384\n",
      "Domain loss: 0.506 dice: 0.4810\n",
      "Epoch 47\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "943a997b168a426a924f45ca532d588a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "49a40e58743243a2918e752a85ac3079",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.377\n",
      "Domain loss: 0.539 dice: 0.4549\n",
      "Epoch 48\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "84cc8bce64d14262b20bee9db4a62051",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dd1f751621074daf94d51a67c2b5938e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.377\n",
      "Domain loss: 0.521 dice: 0.4731\n",
      "Epoch 49\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d158493a1ab543feade4cce25d1f61e8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ff01197046ad41679cccac25023aeb4e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.389\n",
      "Domain loss: 0.567 dice: 0.4306\n",
      "Epoch 50\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7fade6d66d2e4f808b54f1d9a64d58b8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "99fcca2ecee9460487d084d3364e0c08",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.388\n",
      "Domain loss: 0.568 dice: 0.4249\n",
      "Epoch 51\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dcfabae673c0489686e8d28d1e2eae8e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d7f805289ba24301abf176132a8c93c8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.375\n",
      "Domain loss: 0.548 dice: 0.4466\n",
      "Epoch 52\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ff82cdaa975d40308d736273ce460217",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bf39061969904ba2ad94008d1724209e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.383\n",
      "Domain loss: 0.515 dice: 0.4763\n",
      "Epoch 53\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "80ce5b1ecdb147d1b70169864fedf2d7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fdbebf548f1c4921810b427e077c6d8d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.384\n",
      "Domain loss: 0.618 dice: 0.3806\n",
      "Epoch 54\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "899c6b07834d427d942812c8982ae2f5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b4e83e25a117450eb56f754e27bea78b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.367\n",
      "Domain loss: 0.551 dice: 0.4398\n",
      "Epoch 55\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f5fa0f75d9fc42b688bae1fc6538a5e8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "42042ba8cfc8485a8bc57332bc89f83a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.376\n",
      "Domain loss: 0.515 dice: 0.4775\n",
      "Epoch 56\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d8d0756c89954597a3ca4611e5c75ff9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2b1dffffb65c452985d81f893593eff5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.377\n",
      "Domain loss: 0.566 dice: 0.4290\n",
      "Epoch 57\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "65b9637d9ba349e0b267e32b334eb0de",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9c07d01417a547b39e776edd65c58e3d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.387\n",
      "Domain loss: 0.536 dice: 0.4544\n",
      "Epoch 58\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "28a0f243e2c2483a93da8f32fa3e9a9b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8bad76503fd04cc1b3f527d9bce5a7ac",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.379\n",
      "Domain loss: 0.551 dice: 0.4440\n",
      "Epoch 59\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fb3757141d0b443182b8f680b64498dd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "652d6971bf0d4fa7acfa14532fe9e661",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.387\n",
      "Domain loss: 0.703 dice: 0.2990\n",
      "Epoch 60\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ae2e533da36e455eb516d1a9484756d2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c671e9f32edf41a284b6c223679df733",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.410\n",
      "Domain loss: 0.544 dice: 0.4467\n",
      "Epoch 61\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a2e6f18deba0418cb1a5b7f9b4eb3958",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "90cdb48531ed486ea9b948c11177043f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.375\n",
      "Domain loss: 0.665 dice: 0.3359\n",
      "Epoch 62\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "050102d57c3345969f86764888b2c5fe",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "52a2a384e2044cf0a90afafb2e8d45d6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.380\n",
      "Domain loss: 0.559 dice: 0.4358\n",
      "Epoch 63\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "eb439c7eddff478d86998c7165c9ea89",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d9c528975a0d4e549fa956c43c95c4a8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.375\n",
      "Domain loss: 0.568 dice: 0.4296\n",
      "Epoch 64\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6d1f46056dd045239c3c7b2f04ea123c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d0fa05d630a24c6582011207a7df81cb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.381\n",
      "Domain loss: 0.566 dice: 0.4278\n",
      "Epoch 65\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0152a1c265b945e6b561a9f572bcccda",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1bc50bf7b42e42e6aeee6a9112374c39",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.376\n",
      "Domain loss: 0.550 dice: 0.4452\n",
      "Epoch 66\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6dccb08986484dc980c1c8a0126f4032",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7e175c3bb30842e1a7c72e6c4cf46c24",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.371\n",
      "Domain loss: 0.546 dice: 0.4475\n",
      "Epoch 67\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6239d74ef461434fa84591a85edf884c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a075e220d59d439fbf40e42408ec60e5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.386\n",
      "Domain loss: 0.597 dice: 0.3988\n",
      "Epoch 68\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e472ba8aa2d54ab8a4a429a2cc08ea62",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a79b88ddb46e4de5a1edebc67e4fd37d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.374\n",
      "Domain loss: 0.531 dice: 0.4627\n",
      "Epoch 69\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5862e4aeb83e4de0a4a96def70a91905",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7f492d4107af4f899e02ddf3c7eb88ec",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.376\n",
      "Domain loss: 0.530 dice: 0.4649\n",
      "Epoch 70\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8298fc561aa54a9dac2bff158de228eb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "51af58a470464427a04aa330a791723e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.401\n",
      "Domain loss: 0.610 dice: 0.3794\n",
      "Epoch 71\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2201fd5a578e43889d8757ac52877550",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7589a471a7bf49d2a909074bc1438fd0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.411\n",
      "Domain loss: 0.601 dice: 0.3961\n",
      "Epoch 72\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c454dcac55324eacbdc1b4af4e48624d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "503835c046ba47738cca5595e895defc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.393\n",
      "Domain loss: 0.527 dice: 0.4662\n",
      "Epoch 73\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f8ee3f40699149d195cc0d444574df83",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "12a3036b5b934ce0aa52e48e9c296562",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.380\n",
      "Domain loss: 0.627 dice: 0.3701\n",
      "Epoch 74\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "831b7fd68fdf49b085edc107b1b06a63",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "70f291640eac4de49731d4aab090f557",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.378\n",
      "Domain loss: 0.552 dice: 0.4461\n",
      "Epoch 75\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5337d255964744ad811befaf6f422c7b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1aba74b52a20401a98b4210d8365647a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.367\n",
      "Domain loss: 0.526 dice: 0.4633\n",
      "Epoch 76\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "46853e314ab4433fa1c6f63780dd818f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f518cd623e3a45efaacc2b116280b2b8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.369\n",
      "Domain loss: 0.558 dice: 0.4380\n",
      "Epoch 77\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "726154a2f511446281cd16b9bc214ce9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d1b604344c39440ab3808f73d880406b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.371\n",
      "Domain loss: 0.565 dice: 0.4325\n",
      "Epoch 78\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "aa310d2458e3476c8d05473d28c32b1b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "687351fc67af4149b313e741b19249a0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.380\n",
      "Domain loss: 0.532 dice: 0.4589\n",
      "Epoch 79\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9ebc1f8ad5014f41bb7e2db9e8d22aa8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/177 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fc654929927344708dd2a44e5eea0f8d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train loss: 0.367\n",
      "Domain loss: 0.519 dice: 0.4680\n"
     ]
    }
   ],
   "source": [
    "train_losses, val_losses, val_dices_domain, _ = perform_training(train_loader=train_loader_corona_radio,\n",
    "                                                                               train_stats=all_train_stats,\n",
    "                                                                               val_domain_loader=val_loader_corona_radio,\n",
    "                                                                               val_domain_stats=all_val_stats,\n",
    "                                                                               val_shift_loader=None,\n",
    "                                                                               val_shift_stats=None,\n",
    "                                                                               epochs=80,\n",
    "                                                                               checkpoint_file='./train_all_checkpoint.pth')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluating samples #[1, 183, 184, 412, 511, 555]\n",
      "Inference for sample #1\n",
      "Loss: 0.0\n",
      "Dice: 0.0\n",
      "Inference for sample #183\n",
      "Loss: 0.5432459115982056\n",
      "Dice: 0.44483986496925354\n",
      "Inference for sample #184\n",
      "Loss: 0.627865195274353\n",
      "Dice: 0.3649425208568573\n",
      "Inference for sample #412\n",
      "Loss: 0.2037556767463684\n",
      "Dice: 0.7988826632499695\n",
      "Inference for sample #511\n",
      "Loss: 0.16989648342132568\n",
      "Dice: 0.8322147727012634\n",
      "Inference for sample #555\n",
      "Loss: 0.0\n",
      "Dice: 0.0\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 18 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualise_inference(corona_val, samples=[1, 183, 184, 412, 511, 555], checkpoint_file='./train_all_checkpoint.pth')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluating samples #[4, 12, 120]\n",
      "Inference for sample #4\n",
      "Loss: 0.9999983906745911\n",
      "Dice: 0.0\n",
      "Inference for sample #12\n",
      "Loss: 0.8099150657653809\n",
      "Dice: 0.1932203322649002\n",
      "Inference for sample #120\n",
      "Loss: 0.8053988814353943\n",
      "Dice: 0.19537274539470673\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 9 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualise_inference(radio_val, samples=[4, 12, 120], checkpoint_file='./train_all_checkpoint.pth')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "id": "RN4_nwEDtKAc"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best dice score 0.4950425619049473\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "80366dab80904fa5814d15caba2a391c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Radiopedia dice 0.3500028753714671, loss 0.632238507270813\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "95e945edd6944223bf64bc456c4f44b5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/19 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Corona dice 0.4245450758523252, loss 0.5647179484367371\n"
     ]
    },
    {
     "data": {
      "image/png": "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