main_unet.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Standard Library\n",
    "import os\n",
    "import json\n",
    "\n",
    "## External Libraries\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "from torchvision import transforms\n",
    "from torch.autograd import Variable\n",
    "import torch.nn.functional as functional\n",
    "from torchsummary import summary\n",
    "import torchio as tio\n",
    "from torch.utils.data import Dataset, DataLoader, SubsetRandomSampler\n",
    "from torch.nn.parallel import DistributedDataParallel as DDP\n",
    "from skimage import io\n",
    "import SimpleITK as sitk\n",
    "import matplotlib.pyplot as plt\n",
    "import torch.nn.functional as F\n",
    "import nibabel as nib\n",
    "from HD_BET.utils import subdirs, subfiles\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "CREATE DIRECTORIES"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# General Data Directory ##TODO: Please fill in the appropriate directory\n",
    "data_dir = os.path.join(\"Registered\")\n",
    "# data_dir = os.path.join(dir, \"DATA\")\n",
    "# mask_dir = os.path.join(dir, \"MASK\")\n",
    "# label_dir = dir"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "CONVOLUTIONAL LAYER CALCULATIONS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Layer 1\n",
      "Output spatial dimensions: [145. 174. 145.], Total elements: 29266.800, Size: 0.117 GB\n",
      "Layer 2\n",
      "Output spatial dimensions: [72. 87. 72.], Total elements: 7216.128, Size: 0.029 GB\n",
      "Layer 3\n",
      "Output spatial dimensions: [36. 43. 36.], Total elements: 1783.296, Size: 0.007 GB\n",
      "Layer 4\n",
      "Output spatial dimensions: [18. 21. 18.], Total elements: 435.456, Size: 0.002 GB\n",
      "Layer 1m\n",
      "Output spatial dimensions: [36. 43. 36.], Total elements: 1783296.0, Size: 0.007133184 GB\n",
      "Layer 2m\n",
      "Output spatial dimensions: [72. 87. 72.], Total elements: 7216128.0, Size: 0.028864512 GB\n",
      "Layer 3m\n",
      "Output spatial dimensions: [145. 174. 145.], Total elements: 29266800.0, Size: 0.1170672 GB\n"
     ]
    }
   ],
   "source": [
    "# temp = sitk.Image()\n",
    "\n",
    "# for f in os.listdir(data_dir):\n",
    "#     img = sitk.ReadImage(os.path.join(data_dir, f))\n",
    "#     sz = img.GetSize()\n",
    "#     pad = [865-sz[0], 682-sz[1], 354-sz[2]]\n",
    "#     img = \n",
    "#     print(img.GetSize())\n",
    "    # img = torch.tensor(sitk.GetArrayFromImage(img))\n",
    "    # pad = (0, 864-img.size()[2], 0, 682-img.size()[1], 0, 354-img.size()[0])\n",
    "    # img = nn.functional.pad(img[None, :], pad, value=0)\n",
    "    \n",
    "# for g in os.listdir(mask_dir):\n",
    "#     mask = sitk.ReadImage(f)\n",
    "#     mask = torch.tensor(sitk.GetArrayFromImage(mask))\n",
    "#     pad = (0, 864-mask.size()[2], 0, 682-mask.size()[1], 0, 354-mask.size()[0])\n",
    "#     mask = nn.functional.pad(mask[None, :], pad, value=0)\n",
    "# layer 1, torch.Size([1, 8, 177, 341, 432])\n",
    "# layer 2, torch.Size([1, 16, 89, 171, 216])\n",
    "# layer 3, torch.Size([1, 32, 45, 86, 108])\n",
    "# layer 4, torch.Size([1, 64, 23, 43, 54])\n",
    "# us43 size torch.Size([1, 32, 46, 86, 108])\n",
    "stride = 1\n",
    "padding = 1\n",
    "dilation = 1\n",
    "kernel = 3\n",
    "poolsize = 2\n",
    "input_dim = np.array([182, 218, 182])\n",
    "input_dim = input_dim / 1.25\n",
    "# print(f\"Output spatial dimensions: {input_dim}, Total elements: {numelements/(1e3):.3f}, Size: {numelements*bytes_per_element/(1E9):.3f} GB\")\n",
    "# input_dim = np.array([288, 227, 118])\n",
    "# input_dim = np.array([96, 75, 39])\n",
    "# Conv3dTranpose:\n",
    "# output_dim = (input_dim - 1) * stride - 2 * padding + dilation * (kernel - 1) + output_padding + 1\n",
    "# Conv3d:\n",
    "numkernels = [8, 16, 32, 64]\n",
    "for k in range(len(numkernels)):\n",
    "    print(f'Layer {k+1}')\n",
    "    output_dim = np.floor((input_dim + 2*padding - dilation*(kernel - 1) - 1) / stride + 1)\n",
    "    numelements = np.prod(output_dim)*numkernels[k]\n",
    "    bytes_per_element = 4.0\n",
    "    print(f\"Output spatial dimensions: {output_dim}, Total elements: {numelements/(1e3):.3f}, Size: {numelements*bytes_per_element/(1E9):.3f} GB\")\n",
    "    if k < 3:\n",
    "        output_dim = np.floor(output_dim/poolsize) # max pooling\n",
    "    input_dim = output_dim\n",
    "\n",
    "kernel = 4\n",
    "stride = 2\n",
    "padding = 1\n",
    "output_padding = np.array([[0, 1, 0], [0, 1, 0], [1, 0, 1]])\n",
    "numkernels.reverse()\n",
    "numkernels = numkernels[1:]\n",
    "for k in range(len(numkernels)):\n",
    "    print(f'Layer {k+1}m')\n",
    "    output_dim = np.floor((input_dim - 1)*stride - 2*padding + dilation*(kernel - 1) + output_padding[k] + 1)\n",
    "    numelements = np.prod(output_dim)*numkernels[k]\n",
    "    bytes_per_element = 4.0\n",
    "    print(f\"Output spatial dimensions: {output_dim}, Total elements: {numelements}, Size: {numelements*bytes_per_element/(1E9)} GB\")\n",
    "    input_dim = output_dim"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "IMAGE DATASET CLASS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Image Transforms\n",
    "img_transform = transforms.Compose([\n",
    "        transforms.ToTensor(),\n",
    "])\n",
    "\n",
    "## Image Dataloader\n",
    "class ImageDataset(Dataset):\n",
    "    \n",
    "    \"\"\"\n",
    "    ImageDataset\n",
    "    \"\"\"\n",
    "    \n",
    "    def __init__(self,\n",
    "                 data_dir\n",
    "                 transforms=None):\n",
    "        \"\"\"\n",
    "        ##TODO: Add support for colorization dataset\n",
    "        \n",
    "        Args:\n",
    "            input_dir (str): Path to either colorization or segmentation directory\n",
    "            op (str): One of \"train\", \"val\", or \"test\" signifying the desired split\n",
    "            mask_json_path (str): Path to mapping.json file\n",
    "            transforms (list or None): Image transformations to apply upon loading.\n",
    "        \"\"\"\n",
    "        self.input_dir = data_dir\n",
    "        self.subjects = subdirs(self.input_dir)\n",
    "        # for dirname in subdirs(self.input_dir):\n",
    "        #     self.input_files.extend(subfiles(dirname))\n",
    "        # self.output_files = [ifile[:-7]+\"_bet.nii.gz\" for ifile in self.input_files]\n",
    "        self.transform = transforms\n",
    "\n",
    "    def __len__(self):\n",
    "        \"\"\"\n",
    "        \n",
    "        \"\"\"\n",
    "        return len(next(os.walk(self.input_dir))[2])\n",
    "\n",
    "    def __getitem__(self,\n",
    "                    idx):\n",
    "        \"\"\"\n",
    "        \n",
    "        \"\"\"\n",
    "\n",
    "        T1w = nib.load(os.path.join(f\"{self.subjects[idx]:03d}\", \"T1w_1mm.nii.gz\"))\n",
    "        T2w = nib.load(os.path.join(f\"{self.subjects[idx]:03d}\", \"T2w_align.nii.gz\"))\n",
    "        ADC = nib.load(os.path.join(f\"{self.subjects[idx]:03d}\", \"ADC_align.nii.gz\"))\n",
    "        FA = nib.load(os.path.join(f\"{self.subjects[idx]:03d}\", \"FA_align.nii.gz\"))\n",
    "        # Create tensors\n",
    "        T1wdata = torch.tensor(T1w.get_fdata)\n",
    "        T2wdata = torch.tensor(T2w.get_fdata)\n",
    "        FAdata = torch.tensor(FA.get_fdata)\n",
    "        ADCdata = torch.tensor(ADC.get_fdata)\n",
    "        \n",
    "#         # NORMALIZE\n",
    "#         mean = 54.4651\n",
    "#         std = 44.4922\n",
    "#         img = (img - mean) / std\n",
    "        \n",
    "#             # read the processed image\n",
    "#         ## Transform image and mask\n",
    "# #         if self.transform:\n",
    "#         if idx < 70:\n",
    "#             img, mask = self.img_transform(img, mask, augtype)\n",
    "# # #         print(img.size())\n",
    "#             widths = np.array([44, 85, 108])*2\n",
    "#             dims = 2*widths\n",
    "#         else:\n",
    "#             dims = [354, 682, 864]\n",
    "# #         print(dims)\n",
    "#         pad = (0, dims[2]-img.size()[2], 0, dims[1]-img.size()[1], 0, dims[0]-img.size()[0])\n",
    "#         img = nn.functional.pad(img[None, :], pad, value=0)\n",
    "#         mask = nn.functional.pad(mask[None, :], pad, value=0)\n",
    "#         print(img.size())\n",
    "        input = torch.cat([T1wdata[None], T2wdata[None]])\n",
    "        target = torch.cat([FAdata[None], ADCdata[None]])\n",
    "        input = nn.functional.interpolate(input[None], target[0].size())\n",
    "        # ## Use dictionary to output\n",
    "        return input[0], target\n",
    "\n",
    "    def img_transform(self,\n",
    "                      img,\n",
    "                      mask,\n",
    "                      aug):\n",
    "        \"\"\"\n",
    "        \n",
    "        \"\"\"\n",
    "#         buffer = 100\n",
    "#         print(img.shape)\n",
    "        tumoridx = torch.nonzero(mask).float() # result will be N(nonz px)x3(l,w,h)\n",
    "#         # print(tumoridx.size())\n",
    "\n",
    "#         # Get the centroid of the tumor and upper and lower bounds in each dimension\n",
    "        tcenter = torch.mean(tumoridx, dim=0) # result is 1x3(l,w,h)\n",
    "        widths = np.array([44, 85, 108])*2\n",
    "        xlow = int(max(0, tcenter[0] - widths[0]))\n",
    "        ylow = int(max(0, tcenter[1] - widths[1]))\n",
    "        zlow = int(max(0, tcenter[2] - widths[2]))\n",
    "        xhigh = int(min(img.shape[0], xlow + 2*widths[0]))\n",
    "        if xhigh - xlow < widths[0]*2:\n",
    "            xlow = xhigh - widths[0]*2\n",
    "        yhigh = int(min(img.shape[1], ylow + 2*widths[1]))\n",
    "        if yhigh - ylow < widths[1]*2:\n",
    "            ylow = yhigh - widths[1]*2\n",
    "        zhigh = int(min(img.shape[2], zlow + 2*widths[2]))\n",
    "        if zhigh - zlow < widths[2]*2:\n",
    "            zlow = zhigh - widths[2]*2\n",
    "        img = img[xlow:xhigh, ylow:yhigh, zlow:zhigh]\n",
    "        mask = mask[xlow:xhigh, ylow:yhigh, zlow:zhigh]\n",
    "#         print(img.shape, xlow, ylow, zlow)\n",
    "#         tmins = torch.min(tumoridx, dim=0)[0]\n",
    "#         tmaxes = torch.max(tumoridx, dim=0)[0]\n",
    "        \n",
    "#         torch.Size([349, 682, 865])\n",
    "#         torch.Size([121, 340, 432]) 228 108 255\n",
    "#         torch.Size([0, 70, 100]) [tensor([134]), tensor([13]), tensor([5])]\n",
    "        # Apply Random Transformations to Image and Mask\n",
    "        if aug == 0:\n",
    "            flipdim = torch.randint(low=0, high=3, size=(1,))\n",
    "            img = torch.flip(img, dims=[flipdim])\n",
    "            mask = torch.flip(mask, dims=[flipdim])\n",
    "#             print(f'flipping dim {flipdim}')\n",
    "        elif aug == 1:\n",
    "            cropsize = (40, 70, 100)\n",
    "            minbounds = [torch.randint(low=0, high=2*widths[0]-cropsize[0], size=(1,)),\n",
    "                         torch.randint(low=0, high=2*widths[1]-cropsize[1], size=(1,)),\n",
    "                         torch.randint(low=0, high=2*widths[2]-cropsize[2], size=(1,))]\n",
    "            img = img[minbounds[0]:(minbounds[0]+cropsize[0]), \n",
    "                      minbounds[1]:(minbounds[1]+cropsize[1]), \n",
    "                      minbounds[2]:(minbounds[2]+cropsize[2])]\n",
    "            mask = mask[minbounds[0]:(minbounds[0]+cropsize[0]), \n",
    "                        minbounds[1]:(minbounds[1]+cropsize[1]), \n",
    "                        minbounds[2]:(minbounds[2]+cropsize[2])]\n",
    "#             print(img.shape, minbounds)\n",
    "            resize = (int(cropsize[0]/cropsize[2]*max(2*widths)), int(cropsize[1]/cropsize[2]*max(2*widths)), max(2*widths))\n",
    "            img = nn.functional.interpolate(img[None, None, :, :, :].float(), size=resize, mode='trilinear')\n",
    "            mask = nn.functional.interpolate(mask[None, None, :, :, :].float(), size=resize, mode='trilinear')\n",
    "            img = img[0, 0]\n",
    "            mask = torch.round(mask[0, 0])\n",
    "#         else:\n",
    "#             rotatedim = torch.randint(low=0, high=3, size=(1,))\n",
    "#             for i in range(img.shape[rotatedim]):\n",
    "#                 img[:, :, i] = \n",
    "#             print(f'cropping {img.shape}')\n",
    "\n",
    "        return img, mask"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "UNET ARCHITECTURE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Functions for adding the convolution layer\n",
    "def add_conv_stage(dim_in,\n",
    "                   dim_out,\n",
    "                   kernel_size=3,\n",
    "                   stride=1,\n",
    "                   padding=1,\n",
    "                   dilation=1,\n",
    "                   bias=True,\n",
    "                   useBN=True):\n",
    "    \"\"\"\n",
    "    \n",
    "    \"\"\"\n",
    "    # Use batch normalization\n",
    "    if useBN:\n",
    "        return nn.Sequential(\n",
    "          nn.Conv3d(dim_in, dim_out, kernel_size=kernel_size, stride=stride, padding=padding, dilation=dilation, bias=bias),\n",
    "          nn.BatchNorm3d(dim_out),\n",
    "          nn.LeakyReLU(0.1),\n",
    "          nn.Conv3d(dim_out, dim_out, kernel_size=kernel_size, stride=stride, padding=padding, dilation=dilation, bias=bias),\n",
    "          nn.BatchNorm3d(dim_out),\n",
    "          nn.LeakyReLU(0.1)\n",
    "        )\n",
    "    # No batch normalization\n",
    "    else:\n",
    "        return nn.Sequential(\n",
    "          nn.Conv3d(dim_in, dim_out, kernel_size=kernel_size, stride=stride, padding=padding, dilation=dilation, bias=bias),\n",
    "          nn.ReLU(),\n",
    "          nn.Conv3d(dim_out, dim_out, kernel_size=kernel_size, stride=1, padding=padding, dilation=dilation, bias=bias),\n",
    "          nn.ReLU()\n",
    "        )\n",
    "\n",
    "## Upsampling\n",
    "def upsample(ch_coarse,\n",
    "             ch_fine,\n",
    "             stride=2,\n",
    "             padding=1,\n",
    "             output_padding=(0, 0, 0)):\n",
    "    \"\"\"\n",
    "    \n",
    "    \"\"\"\n",
    "    return nn.Sequential(\n",
    "                    nn.ConvTranspose3d(ch_coarse, ch_fine, kernel_size=4, stride=stride, padding=padding, output_padding=output_padding, bias=False),\n",
    "                    nn.ReLU())\n",
    "\n",
    "\n",
    "# U-Net\n",
    "class UNET(nn.Module):\n",
    "    \n",
    "    \"\"\"\n",
    "    \n",
    "    \"\"\"\n",
    "    def __init__(self, n_classes, useBN=True):\n",
    "        \"\"\"\n",
    "        Args:\n",
    "            n_classes (int): Number of classes\n",
    "            useBN (bool): Turn Batch Norm on or off. (Hint: Using BatchNorm might help you achieve better performance.)\n",
    "        \"\"\"\n",
    "        super(UNET, self).__init__()\n",
    "        torch.manual_seed(42)\n",
    "        # Downgrade stages\n",
    "        dilation = 1\n",
    "        stride = 1\n",
    "        \n",
    "        # self.conv1 = add_conv_stage(3, 32, useBN=useBN) # for segmentation training\n",
    "        self.conv1 = add_conv_stage(2, 8, stride=stride, dilation=dilation, useBN=useBN) # for colorization training\n",
    "        self.conv2 = add_conv_stage(8, 16, stride=stride, dilation=dilation, useBN=useBN)\n",
    "        self.conv3 = add_conv_stage(16, 32, stride=stride, dilation=dilation, useBN=useBN)\n",
    "        self.conv4 = add_conv_stage(32, 64, stride=stride, dilation=dilation, useBN=useBN)\n",
    "        # Upgrade stages\n",
    "        self.conv3m = add_conv_stage(64, 32, dilation=dilation, useBN=useBN)\n",
    "        self.conv2m = add_conv_stage(32,  16, dilation=dilation, useBN=useBN)\n",
    "        self.conv1m = add_conv_stage( 16,  8, dilation=dilation, useBN=useBN)\n",
    "        # Maxpool\n",
    "        self.max_pool = nn.MaxPool3d(2)\n",
    "        # self.max_pool = nn.Identity()\n",
    "        # Upsample layers\n",
    "        # [0, 1, 0], [0, 1, 0], [1, 0, 1]\n",
    "        self.upsample43 = upsample(64, 32, output_padding=(0, 1, 0))\n",
    "        self.upsample32 = upsample(32,  16, output_padding=(0, 1, 0))\n",
    "        self.upsample21 = upsample(16 ,  8, output_padding=(1, 0, 1))\n",
    "        # weight initialization\n",
    "        # You can have your own weight intialization. This is just an example.\n",
    "        for m in self.modules():\n",
    "            if isinstance(m, nn.Conv3d) or isinstance(m, nn.ConvTranspose3d):\n",
    "                if m.bias is not None:\n",
    "                    m.bias.data.zero_()\n",
    "        #TODO: Design your last layer & activations\n",
    "        # Segmentation w/o coloring:\n",
    "        # self.convFinal = nn.Sequential(nn.Conv2d(32, n_classes, kernel_size=1, stride=1, padding=0, bias=True),\n",
    "                                    #    nn.ReLU())\n",
    "        # Colorization and colorization+segmentation:\n",
    "        self.un_pool = nn.MaxUnpool3d(4)\n",
    "        self.convFinal = nn.Sequential(nn.Conv3d(2, n_classes, kernel_size=1, stride=1, padding=0, bias=True),\n",
    "                                       nn.ReLU())\n",
    "        # self.lastLin = nn.Linear(3, 6)\n",
    "        \n",
    "\n",
    "\n",
    "    def forward(self, x):\n",
    "        \"\"\"\n",
    "        Forward pass\n",
    "        \"\"\"\n",
    "#         print(x.size())\n",
    "        xsize = x.size()\n",
    "        if self.training:\n",
    "            scaled = self.avg_pooltrain(x)\n",
    "        else:\n",
    "            scaled = self.avg_pooleval(x)\n",
    "#         print(f'scaling down', {scaled.size()})\n",
    "        conv1_out = self.conv1(scaled)\n",
    "#         print(f'layer 1, {conv1_out.size()}')\n",
    "        conv2_out = self.conv2(self.max_pool(conv1_out))\n",
    "#         print(f'layer 2, {conv2_out.size()}')\n",
    "        conv3_out = self.conv3(self.max_pool(conv2_out))\n",
    "#         print(f'layer 3, {conv3_out.size()}')\n",
    "        conv4_out = self.conv4(self.max_pool(conv3_out))\n",
    "#         print(f'layer 4, {conv4_out.size()}')\n",
    "\n",
    "        us43 = self.upsample43(conv4_out)\n",
    "#         print(f'us43 size {us43.size()}')\n",
    "        conv4m_out_ = torch.cat((us43, conv3_out), 1)\n",
    "        conv3m_out  = self.conv3m(conv4m_out_)\n",
    "#         print(f'layer 3m, {conv3m_out.size()}')\n",
    "\n",
    "        us32 = self.upsample32(conv3m_out)\n",
    "#         print(f'us32 size {us32.size()}')\n",
    "        conv3m_out_ = torch.cat((us32, conv2_out), 1)\n",
    "        conv2m_out  = self.conv2m(conv3m_out_)\n",
    "#         print(f'layer 2m, {conv2m_out.size()}')\n",
    "\n",
    "        us21 = self.upsample21(conv2m_out)\n",
    "#         print(f'us21 size {us21.size()}')\n",
    "        conv2m_out_ = torch.cat((us21, conv1_out), 1)\n",
    "        conv1m_out  = self.conv1m(conv2m_out_)\n",
    "#         print(f'layer 1m, {conv1m_out.size()}')\n",
    "#         c1msize = conv1m_out.size()\n",
    "        #TODO: Design your last layer & activations\n",
    "        out = self.convFinal(conv1m_out)\n",
    "#         print(f'layer out, {out.size()}')\n",
    "\n",
    "        # outscaled = nn.functional.interpolate(out, size=xsize[-3:])\n",
    "        # print(f'layer out scaled, {outscaled.size()}')\n",
    "        \n",
    "        return out\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "DICE LOSS/SCORES"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "##TODO: Finish implementing the multi-class DICE score function\n",
    "def dice_score_image(prediction, target, n_classes):\n",
    "    '''\n",
    "      computer the mean dice score for a single image\n",
    "\n",
    "      Reminders: A false positive is a result that indicates a given condition exists, when it does not\n",
    "               A false negative is a test result that indicates that a condition does not hold, while in fact it does\n",
    "      Args:\n",
    "          prediction (tensor): predictied labels of the image\n",
    "          target (tensor): ground truth of the image\n",
    "          n_classes (int): number of classes\n",
    "    \n",
    "      Returns:\n",
    "          m_dice (float): Mean dice score over classes\n",
    "    '''\n",
    "    ## Should test image one by one\n",
    "    assert prediction.shape[0] == 1 #This line can not be deleted\n",
    "    ## TODO: Compute Dice Score for Each Class. Compute Mean Dice Score over Classes.\n",
    "#     dice_classes = torch.zeros(n_classes)\n",
    "    # print(prediction.shape, target.shape)\n",
    "#     print(prediction.shape)\n",
    "#     pd = pd[0].long()\n",
    "#     prediction = nn.functional.one_hot(prediction, num_classes=n_classes).permute(0, 4, 1, 2, 3)\n",
    "    \n",
    "    # If we want to scale down target\n",
    "#     pool = nn.AvgPool3d(4)\n",
    "#     target = pool(target)\n",
    "    # If we want to scale up prediction\n",
    "    \n",
    "#     target = torch.cat([target, 1-target], dim=1)\n",
    "#     print(prediction.shape, target.shape)\n",
    "#     prediction = prediction.float()\n",
    "#     target = target.float()\n",
    "#     TP_0 = 0\n",
    "#     TP_1 = 0\n",
    "#     FP_0 = 0\n",
    "#     FP_1 = 0\n",
    "#     FN_0 = 0\n",
    "#     FN_1 = 0\n",
    "    prediction = prediction.int()\n",
    "    TP_0 = (prediction[0]*target[0]).sum()\n",
    "    TP_1 = ((1-prediction[0])*(1-target[0])).sum()\n",
    "    FP_0 = (prediction[0]*(1-target[0])).sum()\n",
    "    FP_1 = ((1-prediction[0])*target[0]).sum()\n",
    "    FN_0 = FP_1 #((1-prediction[0])*target[0]).sum()\n",
    "    FN_1 = FP_0 #((prediction[0])*(1-target[0])).sum()\n",
    "#     for a in range(prediction.shape[1]):\n",
    "#         for b in range(prediction.shape[2]):\n",
    "#             for c in range(prediction.shape[3]):\n",
    "#                 TP_0 += (prediction[0, a, b, c]*target[0, 0, a, b, c])\n",
    "#                 TP_1 += ((1-prediction[0, a, b, c])*(1-target[0, 0, a, b, c]))\n",
    "#                 FP_1 += ((1-prediction[0, a, b, c])*target[0, 0, a, b, c])\n",
    "#                 FP_0 += (prediction[0, a, b, c]*(1-target[0, 0, a, b, c]))\n",
    "#                 FN_0 += ((1-prediction[0, a, b, c])*target[0, 0, a, b, c])\n",
    "#                 FN_1 += ((1-(1-prediction[0, a, b, c]))*(1-target[0, 0, a, b, c]))\n",
    "    if TP_0 == 0 and FN_0 == 0:\n",
    "        dice_score0 = 0 if FP_0 > 0 else 1\n",
    "    else:\n",
    "        dice_score0 = 2*TP_0 / (2*TP_0 + FP_0 + FN_0)\n",
    "    if TP_1 == 0 and FN_1 == 0:\n",
    "        dice_score1 = 0 if FP_1 > 0 else 1\n",
    "    else:\n",
    "        dice_score1 = 2*TP_1 / (2*TP_1 + FP_1 + FN_1)\n",
    "    print(f'Class 0 TP: {TP_0}, FP: {FP_0}, FN: {FN_0}, DSC: {dice_score0}')\n",
    "    print(f'Class 1 TP: {TP_1}, FP: {FP_1}, FN: {FN_1}, DSC: {dice_score1}')\n",
    "#     for cl in range(n_classes):\n",
    "#         TP = (prediction[0, cl]*target[0, cl]).sum()\n",
    "#         FP = (prediction[0, cl]*(1-target[0, cl])).sum()\n",
    "#         FN = ((1-prediction[0, cl])*target[0, cl]).sum()\n",
    "        #When there is no grount truth of the class in this image\n",
    "        #Give 1 dice score if False Positive pixel number is 0, \n",
    "        #give 0 dice score if False Positive pixel number is not 0 (> 0).\n",
    "#         if TP == 0 and FN == 0:\n",
    "#             dice_classes[cl] = 0 if FP > 0 else 1\n",
    "#         else:\n",
    "#             dice_classes[cl] = 2*TP / (2*TP + FP + FN)\n",
    "    return (dice_score0 + dice_score1) / 2\n",
    "\n",
    "def dice_score_dataset(model, dataloader, num_classes, use_gpu=False):\n",
    "    \"\"\"\n",
    "    Compute the mean dice score on a set of data.\n",
    "    \n",
    "    Note that multiclass dice score can be defined as the mean over classes of binary\n",
    "    dice score. Dice score is computed per image. Mean dice score over the dataset is the dice\n",
    "    score averaged across all images.\n",
    "    \n",
    "    Reminders: A false positive is a result that indicates a given condition exists, when it does not\n",
    "               A false negative is a test result that indicates that a condition does not hold, while in fact it does\n",
    "     \n",
    "    Args:\n",
    "        model (UNET class): Your trained model\n",
    "        dataloader (DataLoader): Dataset for evaluation\n",
    "        num_classes (int): Number of classes\n",
    "    \n",
    "    Returns:\n",
    "        m_dice (float): Mean dice score over the input dataset\n",
    "    \"\"\"\n",
    "    ## Number of Batches and Cache over Dataset \n",
    "    n_batches = len(dataloader)\n",
    "    scores = np.zeros(n_batches)\n",
    "    # when using pretraining:\n",
    "    # grayed = transforms.Grayscale()\n",
    "    ## Evaluate\n",
    "    device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")\n",
    "    model.eval()\n",
    "    idx = 0\n",
    "    for data in dataloader:\n",
    "        ## Format Data\n",
    "        img, target, label = data\n",
    "        if use_gpu:\n",
    "            img = img.to(device)\n",
    "            target = target.to(device)\n",
    "\n",
    "        img = img.float()\n",
    "        target = target.short()\n",
    "        ## Make Predictions\n",
    "        # out = model(grayed(img)) # when using colorization pretraining\n",
    "        out = model(img) # no colorization pretraining\n",
    "        n_classes = out.shape[1]\n",
    "        # Scale up to be same size as target\n",
    "#         print(out.shape)\n",
    "        out = nn.functional.interpolate(out, size=tuple(target.size()[-3:]), mode='trilinear')\n",
    "#         print(out.shape)\n",
    "#         out = out.cpu()\n",
    "        out = torch.argmax(out, dim = 1) # [1, w, h, d]\n",
    "        target = target.int()\n",
    "#         out = out.cuda()\n",
    "        scores[idx] = dice_score_image(out, target, n_classes)\n",
    "        idx += 1\n",
    "    ## Average Dice Score Over Images\n",
    "    m_dice = scores.mean()\n",
    "    return m_dice\n",
    "\n",
    "## TODO: Implement DICE loss, \n",
    "#  It should conform to to how we computer the dice score.\n",
    "class DICELoss(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(DICELoss, self).__init__()\n",
    "\n",
    "    def forward(self, output, target):\n",
    "        # alpha = 10.0\n",
    "        prediction = nn.functional.softmax(output, dim=1)\n",
    "        dice_classes = torch.zeros(n_classes)\n",
    "        # Scale down the target when calculating loss\n",
    "#         pool = nn.AvgPool3d(8)\n",
    "#         target = pool(target)\n",
    "#         target = torch.cat([target, 1-target])\n",
    "#         print(prediction.shape, target.shape)\n",
    "#         print(torch.min(prediction[0]), torch.min(target), torch.min(1-target))\n",
    "        prediction = prediction.float()\n",
    "        target = target.float()\n",
    "#         print(torch.max(prediction[0]))\n",
    "        TP_0 = (prediction[:, 0]*target[:, 0]).sum(dim=(1, 2, 3))\n",
    "        TP_1 = (prediction[:, 1]*(1-target[:, 0])).sum(dim=(1, 2, 3))\n",
    "        FP_1 = (prediction[:, 1]*target[:, 0]).sum(dim=(1, 2, 3))\n",
    "        FP_0 = (prediction[:, 0]*(1-target[:, 0])).sum(dim=(1, 2, 3))\n",
    "        FN_0 = ((1-prediction[:, 0])*target[:, 0]).sum(dim=(1, 2, 3))\n",
    "        FN_1 = ((1-prediction[:, 1])*(1-target[:, 0])).sum(dim=(1, 2, 3))\n",
    "#         print(TP_0, FN_0, FP_0)\n",
    "        # print(f\"Prediction size: {prediction.shape}, Concatenated target size: {target.shape}\")\n",
    "#         for cl in range(n_classes):\n",
    "#             # print(f\"Prediction of class {cl} shape: {prediction[cl].shape}, Target of class {cl} shape: {target.shape}\")\n",
    "#             TP = (prediction[cl]*target[cl]).sum()\n",
    "#             FP = (prediction[cl]*(1-target[cl])).sum()\n",
    "#             FN = ((1-prediction[cl])*target[cl]).sum()\n",
    "#             dice_classes[cl] = 2*TP / (2*TP + FP + FN)\n",
    "        dice_score0 = 2*TP_0 / (2*TP_0 + FP_0 + FN_0)\n",
    "        dice_score1 = 2*TP_1 / (2*TP_1 + FP_1 + FN_1)\n",
    "        return 1 - (dice_score0 + dice_score1) / 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "class FocalLoss(nn.Module):\n",
    "    def __init__(self, gamma=2, alpha=None, reduction='sum'):\n",
    "        super(FocalLoss, self).__init__()\n",
    "        self.gamma = gamma\n",
    "        self.alpha = alpha\n",
    "        self.reduction = reduction\n",
    "\n",
    "    def forward(self, inputs, targets):\n",
    "#         if self.alpha is not None:\n",
    "#             alpha = self.alpha[target]\n",
    "#             alpha = alpha.view(-1, 1)\n",
    "#             alpha = torch.gather(alpha, 0, target.view(-1, 1))\n",
    "#         else:\n",
    "#             alpha = 1\n",
    "\n",
    "        p = torch.nn.functional.softmax(inputs, dim=1)\n",
    "        targets = torch.cat((targets, 1-targets), dim=1)\n",
    "        ce_loss = F.binary_cross_entropy_with_logits(inputs, targets, reduction=\"none\")\n",
    "        p_t = p * targets + (1 - p) * (1 - targets)\n",
    "        loss = ce_loss * ((1 - p_t) ** self.gamma)\n",
    "        if self.alpha >= 0:\n",
    "            alpha_t = self.alpha * targets + (1 - self.alpha) * (1 - targets)\n",
    "            loss = alpha_t * loss\n",
    "\n",
    "        # Check reduction option and return loss accordingly\n",
    "        if self.reduction == \"none\":\n",
    "            pass\n",
    "        elif self.reduction == \"mean\":\n",
    "            loss = loss.mean(dim=(1, 2, 3, 4))\n",
    "        elif self.reduction == \"sum\":\n",
    "            loss = loss.sum(dim=(1, 2, 3, 4))\n",
    "        else:\n",
    "            raise ValueError(\n",
    "                f\"Invalid Value for arg 'reduction': '{self.reduction} \\n Supported reduction modes: 'none', 'mean', 'sum'\"\n",
    "            )\n",
    "        return loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def MAE_score(model, dataloader, use_gpu):\n",
    "    device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")\n",
    "    model.eval()\n",
    "    scores = 0\n",
    "    for img, target in dataloader:\n",
    "        ## Format Data\n",
    "        if use_gpu:\n",
    "            img = img.to(device)\n",
    "            target = target.to(device)\n",
    "\n",
    "        # img = img.float()\n",
    "        # target = target.short()\n",
    "        ## Make Predictions\n",
    "        # out = model(grayed(img)) # when using colorization pretraining\n",
    "        out = model(img) # no colorization pretraining\n",
    "        # Scale up to be same size as target\n",
    "#         print(out.shape)\n",
    "        # out = nn.functional.interpolate(out, size=tuple(target.size()[-3:]), mode='trilinear')\n",
    "#         print(out.shape)\n",
    "#         out = out.cpu()\n",
    "        # out = torch.argmax(out, dim = 1) # [1, w, h, d]\n",
    "        # target = target.int()\n",
    "#         out = out.cuda()\n",
    "        scores += torch.mean(torch.abs(out - target))\n",
    "    ## Average Dice Score Over Images\n",
    "    return scores / len(dataloader)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "CREATE HYPERPARAMS AND DATALOADERS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "100\n"
     ]
    }
   ],
   "source": [
    "## Batch Size\n",
    "train_batch_size = 4\n",
    "validation_batch_size = 4\n",
    "\n",
    "## Learning Rate\n",
    "learning_rate = 0.001\n",
    "\n",
    "# Epochs (Consider setting high and implementing early stopping)\n",
    "num_epochs = 20\n",
    "\n",
    "## Initialize your unet\n",
    "n_classes = 2\n",
    "gen = torch.Generator()\n",
    "gen.manual_seed(42)\n",
    "\n",
    "trainmask = range(0, 140)\n",
    "valmask = range(140, 170)\n",
    "testmask = range(170, 200)\n",
    "train_dataset=ImageDataset(input_dir=data_dir)\n",
    "print(len(train_dataset))\n",
    "train_dataloader = DataLoader(train_dataset, batch_size=train_batch_size, sampler=SubsetRandomSampler(trainmask), generator=gen)\n",
    "val_dataloader = DataLoader(train_dataset, batch_size=validation_batch_size, sampler=SubsetRandomSampler(valmask), generator=gen)\n",
    "test_dataloader = DataLoader(train_dataset, batch_size=1, sampler=SubsetRandomSampler(testmask), generator=gen)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# # Get the amount of tumor voxels\n",
    "# w_tum = 0\n",
    "# for i, (inputs, targets, labels) in enumerate(train_dataloader):\n",
    "#     w_tum += targets.sum()\n",
    "#     print(w_tum)\n",
    "# w_tum = w_tum / (len(trainmask)*torch.numel(targets[0]))\n",
    "# print(w_tum)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# # Calculate percentage of tumor voxels\n",
    "# w_tum = 22660855 / (len(trainmask)*torch.numel(targets[0]))\n",
    "# print(torch.numel(targets[0]))\n",
    "# print(w_tum)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# classes = np.zeros(10)\n",
    "# classes[0:2] = classes[0:2] + 1\n",
    "# weights = sklearn.utils.class_weight.compute_class_weight('balanced', [0, 1], classes)\n",
    "# print(weights)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# print(outputs[:, 0].shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# with torch.no_grad():\n",
    "#     sumvx = 0\n",
    "#     numvx = 0\n",
    "#     for i, (inputs, targets, labels) in enumerate(train_dataloader):\n",
    "#         sumvx += inputs.sum()\n",
    "#         numvx += torch.numel(inputs)\n",
    "#     meanvx = sumvx / numvx\n",
    "#     print(meanvx)\n",
    "#     sumsq = 0\n",
    "#     for i, (inputs, targets, labels) in enumerate(train_dataloader):\n",
    "#         sumsq += torch.pow(inputs - meanvx, 2).sum()\n",
    "#     stdvx = torch.sqrt(sumsq/numvx)\n",
    "#     print(stdvx)\n",
    "    "
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "TRAINING PROCEDURE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Start Training...\n",
      "\n",
      "EPOCH 1 of 20\n",
      "\n",
      "torch.Size([353, 682, 865])\n",
      "torch.Size([176, 340, 432]) 136 0 281\n",
      "torch.Size([40, 70, 100]) [tensor([107]), tensor([123]), tensor([235])]\n",
      "torch.Size([354, 682, 865])\n",
      "torch.Size([176, 340, 432]) 75 0 265\n",
      "torch.Size([40, 70, 100]) [tensor([78]), tensor([27]), tensor([27])]\n",
      "torch.Size([348, 546, 843])\n",
      "torch.Size([176, 340, 432]) 103 0 411\n",
      "torch.Size([40, 70, 100]) [tensor([25]), tensor([155]), tensor([24])]\n",
      "torch.Size([348, 682, 865])\n",
      "torch.Size([176, 340, 432]) 136 0 433\n",
      "torch.Size([40, 70, 100]) [tensor([52]), tensor([3]), tensor([107])]\n",
      "0\n",
      "torch.Size([330, 608, 865])\n",
      "torch.Size([176, 340, 432]) 146 0 69\n",
      "torch.Size([40, 70, 100]) [tensor([13]), tensor([73]), tensor([262])]\n",
      "torch.Size([330, 608, 865])\n",
      "torch.Size([176, 340, 432]) 61 0 248\n",
      "torch.Size([40, 70, 100]) [tensor([101]), tensor([155]), tensor([246])]\n",
      "torch.Size([349, 682, 865])\n",
      "torch.Size([176, 340, 432]) 138 0 234\n",
      "torch.Size([40, 70, 100]) [tensor([13]), tensor([107]), tensor([236])]\n",
      "torch.Size([330, 608, 865])\n",
      "torch.Size([176, 340, 432]) 154 0 138\n",
      "torch.Size([40, 70, 100]) [tensor([93]), tensor([239]), tensor([62])]\n",
      "1\n",
      "torch.Size([354, 682, 865])\n",
      "torch.Size([176, 340, 432]) 144 0 178\n",
      "torch.Size([40, 70, 100]) [tensor([71]), tensor([227]), tensor([213])]\n",
      "torch.Size([352, 682, 865])\n",
      "torch.Size([176, 340, 432]) 81 0 86\n",
      "torch.Size([40, 70, 100]) [tensor([37]), tensor([74]), tensor([86])]\n",
      "torch.Size([330, 608, 865])\n",
      "torch.Size([176, 340, 432]) 78 0 228\n",
      "torch.Size([40, 70, 100]) [tensor([48]), tensor([166]), tensor([165])]\n",
      "torch.Size([348, 682, 865])\n",
      "torch.Size([176, 340, 432]) 117 0 245\n",
      "torch.Size([40, 70, 100]) [tensor([6]), tensor([17]), tensor([126])]\n",
      "2\n",
      "torch.Size([348, 682, 865])\n",
      "torch.Size([176, 340, 432]) 141 41 243\n",
      "torch.Size([40, 70, 100]) [tensor([1]), tensor([25]), tensor([271])]\n",
      "torch.Size([349, 682, 865])\n",
      "torch.Size([176, 340, 432]) 0 0 137\n",
      "torch.Size([40, 70, 100]) [tensor([120]), tensor([254]), tensor([158])]\n",
      "torch.Size([353, 546, 843])\n",
      "torch.Size([176, 340, 432]) 64 0 49\n",
      "torch.Size([40, 70, 100]) [tensor([90]), tensor([29]), tensor([245])]\n",
      "torch.Size([349, 682, 865])\n",
      "torch.Size([176, 340, 432]) 137 56 276\n",
      "torch.Size([40, 70, 100]) [tensor([16]), tensor([27]), tensor([88])]\n",
      "3\n",
      "torch.Size([349, 682, 865])\n",
      "torch.Size([176, 340, 432]) 46 0 406\n",
      "torch.Size([40, 70, 100]) [tensor([135]), tensor([78]), tensor([290])]\n",
      "torch.Size([349, 682, 865])\n",
      "torch.Size([176, 340, 432]) 173 113 423\n",
      "torch.Size([40, 70, 100]) [tensor([2]), tensor([67]), tensor([66])]\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m/tmp/ipykernel_4153139/2272225242.py\u001b[0m in \u001b[0;36m?\u001b[0;34m()\u001b[0m\n\u001b[1;32m     44\u001b[0m \u001b[0;31m# for i in range(100):\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     45\u001b[0m \u001b[0;31m#     img, mask, label = train_dataset.__getitem__(i)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     46\u001b[0m \u001b[0;31m#     # print(img.size())\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     47\u001b[0m \u001b[0;31m#     imgshapes[i] = np.array(img.size())[1:]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 48\u001b[0;31m \u001b[0;31m#     # print(label)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     49\u001b[0m \u001b[0;31m# plt.figure()\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     50\u001b[0m \u001b[0;31m# for j in range(imgshapes.shape[1]):\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     51\u001b[0m \u001b[0;31m#     plt.plot(imgshapes[:, j])\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/.local/lib/python3.8/site-packages/torch/utils/data/dataloader.py\u001b[0m in \u001b[0;36m?\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    643\u001b[0m                     warn_msg += (\"For multiprocessing data-loading, this could be caused by not properly configuring the \"\n\u001b[1;32m    644\u001b[0m                                  \u001b[0;34m\"IterableDataset replica at each worker. Please see \"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    645\u001b[0m                                  \"https://pytorch.org/docs/stable/data.html#torch.utils.data.IterableDataset for examples.\")\n\u001b[1;32m    646\u001b[0m                 \u001b[0mwarnings\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwarn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mwarn_msg\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 647\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m~/.local/lib/python3.8/site-packages/torch/utils/data/dataloader.py\u001b[0m in \u001b[0;36m?\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    676\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_next_data\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    677\u001b[0m         \u001b[0mindex\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_next_index\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m  \u001b[0;31m# may raise StopIteration\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 678\u001b[0;31m         \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_dataset_fetcher\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfetch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mindex\u001b[0m\u001b[0;34m)\u001b[0m  \u001b[0;31m# may raise StopIteration\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    679\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_pin_memory\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    680\u001b[0m             \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_utils\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpin_memory\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpin_memory\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_pin_memory_device\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    681\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/.local/lib/python3.8/site-packages/torch/utils/data/_utils/fetch.py\u001b[0m in \u001b[0;36m?\u001b[0;34m(self, possibly_batched_index)\u001b[0m\n\u001b[1;32m     47\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mauto_collation\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     48\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mhasattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdataset\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m\"__getitems__\"\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdataset\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__getitems__\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     49\u001b[0m                 \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdataset\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__getitems__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpossibly_batched_index\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     50\u001b[0m             \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 51\u001b[0;31m                 \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdataset\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0midx\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0midx\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mpossibly_batched_index\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     52\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     53\u001b[0m             \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdataset\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mpossibly_batched_index\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     54\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcollate_fn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/.local/lib/python3.8/site-packages/torch/utils/data/_utils/fetch.py\u001b[0m in \u001b[0;36m?\u001b[0;34m(.0)\u001b[0m\n\u001b[0;32m---> 51\u001b[0;31m     \u001b[0;32mdef\u001b[0m \u001b[0mfetch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mpossibly_batched_index\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     52\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mauto_collation\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     53\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mhasattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdataset\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m\"__getitems__\"\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdataset\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__getitems__\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     54\u001b[0m                 \u001b[0mdata\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdataset\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__getitems__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpossibly_batched_index\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/tmp/ipykernel_4153139/3457135585.py\u001b[0m in \u001b[0;36m?\u001b[0;34m(self, idx)\u001b[0m\n\u001b[1;32m     51\u001b[0m \u001b[0;31m#             pic = int(idx / 3) # Which image to select\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     52\u001b[0m \u001b[0;31m#         else:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     53\u001b[0m \u001b[0;31m#             pic = idx - 140\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     54\u001b[0m         \u001b[0maugtype\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrandint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlow\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhigh\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m3\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msize\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;31m# Which augmentation to apply\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 55\u001b[0;31m         \u001b[0mimg\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0msitk\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mReadImage\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf\"{self.input_dir}/DATA_{idx:03d}.nrrd\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     56\u001b[0m         \u001b[0mmask\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0msitk\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mReadImage\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf\"{self.mask_dir}/MASK_{idx:03d}.nrrd\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     57\u001b[0m         \u001b[0;31m# sz = img.GetSize()\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     58\u001b[0m         \u001b[0;31m# temp = sitk.Image(int(sz[0]/scale), int(sz[1]/scale), int(sz[2]/scale), sitk.sitkUInt8)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/.local/lib/python3.8/site-packages/SimpleITK/extra.py\u001b[0m in \u001b[0;36m?\u001b[0;34m(fileName, outputPixelType, imageIO)\u001b[0m\n\u001b[1;32m    351\u001b[0m         \u001b[0mreader\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mSetFileNames\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mstr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mname\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mfileName\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    352\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    353\u001b[0m     \u001b[0mreader\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mSetImageIO\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mimageIO\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    354\u001b[0m     \u001b[0mreader\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mSetOutputPixelType\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutputPixelType\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 355\u001b[0;31m     \u001b[0;32mreturn\u001b[0m \u001b[0mreader\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mExecute\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m~/.local/lib/python3.8/site-packages/SimpleITK/SimpleITK.py\u001b[0m in \u001b[0;36m?\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m   8434\u001b[0m         \u001b[0mtype\u001b[0m \u001b[0mto\u001b[0m \u001b[0mbe\u001b[0m \u001b[0msame\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mthe\u001b[0m \u001b[0mfile\u001b[0m\u001b[0;34m.\u001b[0m \u001b[0mIf\u001b[0m \u001b[0mthe\u001b[0m \u001b[0mpixel\u001b[0m \u001b[0mtype\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0mspecified\u001b[0m \u001b[0mthen\u001b[0m \u001b[0mthe\u001b[0m \u001b[0mitk\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mConvertPixelBuffer\u001b[0m \u001b[0mwill\u001b[0m \u001b[0mbe\u001b[0m \u001b[0mused\u001b[0m \u001b[0mto\u001b[0m \u001b[0mconvert\u001b[0m \u001b[0mthe\u001b[0m \u001b[0mpixels\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   8435\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   8436\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   8437\u001b[0m         \"\"\"\n\u001b[0;32m-> 8438\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0m_SimpleITK\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mImageFileReader_Execute\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "model = UNET(n_classes)\n",
    "# summary(model, (865, 608, 354))\n",
    "\n",
    "# Initialize Dataloaders\n",
    "if torch.cuda.is_available():\n",
    "    device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")\n",
    "    model = nn.DataParallel(model)\n",
    "    use_gpu = True\n",
    "else:\n",
    "    use_gpu = False\n",
    "model = model.to(device)\n",
    "# summary(model, (1, 354, 682, 865))\n",
    "\n",
    "optimizer = torch.optim.Adam(model.parameters(),lr=learning_rate)\n",
    "scheduler = torch.optim.lr_scheduler.StepLR(optimizer, step_size=10, gamma=0.1)\n",
    "# loss = FocalLoss(gamma=2, alpha=0.25, reduction='mean')\n",
    "loss = nn.L1Loss(reduction = \"sum\")\n",
    "# weights = torch.tensor([.0001*2, .9999*2])\n",
    "# weights = weights.to(device)\n",
    "# loss = nn.CrossEntropyLoss(weight=weights)\n",
    "\n",
    "trainloss = torch.zeros(num_epochs)\n",
    "valloss = torch.zeros(num_epochs)\n",
    "testscores = torch.zeros(num_epochs)\n",
    "\n",
    "print(\"Start Training...\")\n",
    "for epoch in range(num_epochs):\n",
    "    ########################### Training #####################################\n",
    "    print(\"\\nEPOCH \" +str(epoch+1)+\" of \"+str(num_epochs)+\"\\n\")\n",
    "    model.train()\n",
    "    currloss = 0.0\n",
    "    # TODO: Design your own training section\n",
    "    for i, (inputs, targets, labels) in enumerate(train_dataloader):\n",
    "        print(f\"Batch {i}\")\n",
    "        # print(inputs.size(), targets.size(), labels.size())\n",
    "        optimizer.zero_grad() \n",
    "        if use_gpu:\n",
    "            inputs = inputs.to(device)\n",
    "            targets = targets.to(device)\n",
    "        # inputs = inputs.float()\n",
    "        # targets = targets.long()\n",
    "        outputs = model(inputs)\n",
    "        batchloss = loss(outputs, targets)\n",
    "        print(batchloss / outputs.shape[0])\n",
    "        batchloss.backward()\n",
    "        optimizer.step()\n",
    "        \n",
    "        currloss += batchloss.item()\n",
    "\n",
    "    inputs = torch.tensor([])\n",
    "    targets = torch.tensor([])\n",
    "    labels = torch.tensor([])\n",
    "    trainloss[epoch] = currloss / len(trainmask)\n",
    "    scheduler.step()\n",
    "    ########################### Validation #####################################\n",
    "    # TODO: Design your own validation section\n",
    "    model.eval()\n",
    "    with torch.no_grad():\n",
    "        currloss = 0.0\n",
    "        for i, (inputs, targets, labels) in enumerate(val_dataloader):\n",
    "            if use_gpu:\n",
    "#                 inputs = inputs.cuda()\n",
    "#                 targets = targets.cuda()\n",
    "                inputs = inputs.to(device)\n",
    "                targets = targets.to(device)\n",
    "#             inputs = inputs.float()\n",
    "# #             pool = nn.MaxPool3d(8)\n",
    "#             targets = nn.functional.max_pool3d(targets.half(), 8)\n",
    "#             targets = targets.long()\n",
    "            outputs = model(inputs)\n",
    "#             batchloss = loss(outputs, torch.squeeze(targets)).sum()\n",
    "            batchloss = loss(outputs, targets)\n",
    "#             batchloss = 0.0\n",
    "#             for j in range(inputs.shape[0]):\n",
    "#                 batchloss += loss(outputs[j], targets[j], n_classes)\n",
    "            currloss += batchloss.item()\n",
    "        inputs = torch.tensor([])\n",
    "        targets = torch.tensor([])\n",
    "        labels = torch.tensor([])\n",
    "        valloss[epoch] = currloss / len(valmask)\n",
    "        testscores[epoch] = MAE_score(model, test_dataloader, use_gpu)\n",
    "    print(f'Training Loss: {trainloss[epoch]}, Validation loss: {valloss[epoch]}, Test score: {testscores[epoch]}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4190900224 0\n"
     ]
    }
   ],
   "source": [
    "print(torch.cuda.memory_allocated(0), torch.cuda.memory_allocated(1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2\n"
     ]
    }
   ],
   "source": [
    "print(torch.cuda.device_count())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tensor([296.8278], device='cuda:0') tensor(296.8278, device='cuda:0')\n"
     ]
    }
   ],
   "source": [
    "prediction = nn.functional.softmax(outputs[0], dim=0)\n",
    "# Scale down the target when calculating loss\n",
    "pool = nn.AvgPool3d(4)\n",
    "target = pool(targets[0])\n",
    "TP1 = (prediction[0]*target).sum(1).sum(1).sum(1)\n",
    "TP2 = (prediction[0]*target).sum()\n",
    "print(TP1, TP2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7fae9cc59f70>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plt.plot(trainloss)\n",
    "# plt.plot(valloss)\n",
    "# plt.plot(testscores)\n",
    "\n",
    "ax1 = plt.subplot()\n",
    "ax1.plot(trainloss[0:9], 'red')\n",
    "ax1.plot(valloss[0:9], 'blue')\n",
    "ax2 = ax1.twinx()\n",
    "ax2.plot(testscores[0:9], 'green')\n",
    "ax1.legend([\"training\", \"validation\"])\n",
    "ax2.legend([\"test\"])\n",
    "# print(trainloss[0:5], valloss[0:5], testscores[0:5])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([1, 2, 188, 227, 288])\n"
     ]
    }
   ],
   "source": [
    "targets= torch.zeros(1, 1, 188, 227, 288)\n",
    "targets = torch.cat([targets, 1-targets], dim=1)\n",
    "print(targets.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "torch.cuda.empty_cache()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "np.save(\"ImageSizes.npy\", imgshapes)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "VISUALIZATION"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "## GET STATISTICS ON WHERE THE TUMORS LIE WITHIN THE SEGMENTATION MASKS\n",
    "\n",
    "torch.manual_seed(43) # need this when we retrieve\n",
    "stats = torch.zeros(len(train_dataloader)*train_batch_size, 9) # datapts x 6(cl,cw,ch,ml,mw,mh)\n",
    "for i, (inputs, targets, labels) in enumerate(train_dataloader):\n",
    "    for j in range(targets.size()[0]):\n",
    "        tg = targets[j]\n",
    "        # print(tg.size())\n",
    "\n",
    "        # Get indexes of the voxels with 1 (tumor label)\n",
    "        tumoridx = torch.nonzero(tg)[:, -3:].float() # result will be N(nonz px)x3(l,w,h)\n",
    "        # print(tumoridx.size())\n",
    "\n",
    "        # Get the centroid of the tumor and upper and lower bounds in each dimension\n",
    "        tcenter = torch.mean(tumoridx, dim=0) # result is 1x3(l,w,h)\n",
    "        tmins = torch.min(tumoridx, dim=0)[0]\n",
    "        tmaxes = torch.max(tumoridx, dim=0)[0]\n",
    "\n",
    "        stats[i*train_batch_size+j] = torch.cat((tmins, tcenter, tmaxes))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x17aae22f550>"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x400 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## PLOT TUMOR LOCATION STATS \n",
    "\n",
    "fig, axs = plt.subplots(1, 3, layout=\"constrained\", figsize=(10, 4))\n",
    "titles = [\"X\", \"Y\", \"Z\"]\n",
    "maxsizes = [354, 682, 865]\n",
    "\n",
    "axs[0].set_ylabel(\"Voxel Index\")\n",
    "for k in range(3):\n",
    "    axs[k].plot(stats[:, k], color='red', linestyle='-', label=\"min\")\n",
    "    axs[k].plot(stats[:, k+3], color='blue', linestyle='--', alpha=0.3, label='center')\n",
    "    axs[k].plot(stats[:, k+6], color='green', linestyle='-', label='max')\n",
    "    axs[k].set_title(titles[k])\n",
    "    axs[k].set_xlabel(\"Training sample\")\n",
    "    axs[k].axhline(y=maxsizes[k], color='black', linestyle='--', label='maxsize')\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Analyzing images for patient 069\n",
      "Image size: (108, 85, 44)\n",
      "1.0 1.0\n",
      "Analyzing images for patient 070\n",
      "Image size: (108, 85, 44)\n",
      "1.0 1.0\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 960x720 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 960x720 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## PLOT SELECTED TRAINING SAMPLES ALONG WITH THEIR SEGMENTATION MASK\n",
    "\n",
    "dir = os.path.join(\"Train-20230419T152017Z-001\", \"Train\")\n",
    "data_dir = os.path.join(dir, \"DATA\")\n",
    "mask_dir = os.path.join(dir, \"MASK\")\n",
    "label_dir = dir\n",
    "\n",
    "slices = np.array([i/10 for i in range(3, 8)]) # slices to plot for validation (given as percentage through the height)\n",
    "showresults = [\"069\", \"070\"] # choose the directories to plot the results for\n",
    "for f in os.listdir(data_dir):\n",
    "    if f[-8:-5] in showresults:\n",
    "        # Show current directory\n",
    "        print(f\"Analyzing images for patient {f[-8:-5]}\")\n",
    "        maskname = os.path.join(mask_dir, f\"MASK{f[-9:]}\")\n",
    "#         img = sitk.ReadImage(os.path.join(data_dir, f))\n",
    "        mask = sitk.ReadImage(maskname)# * 255\n",
    "        idx = int(f[-8:-5])\n",
    "        img = outputs[idx-69]\n",
    "        img = torch.argmin(img, dim=0)\n",
    "        img = img.float().cpu()\n",
    "#         img = torch.tensor(sitk.GetArrayFromImage(img))\n",
    "        mask = torch.tensor(sitk.GetArrayFromImage(mask))\n",
    "        pool = nn.MaxPool3d(8)\n",
    "#         img = pool(img[None, None, :].float())\n",
    "        mask = pool(mask[None, None, :].float())\n",
    "#         img = img[0, 0]\n",
    "#         mask = mask[0, 0]\n",
    "        img_idx = torch.nonzeros(img)\n",
    "        \n",
    "        img = sitk.GetImageFromArray(img.numpy())\n",
    "        mask = sitk.GetImageFromArray(mask[0, 0].numpy())\n",
    "        # scale = 3\n",
    "        # sz = img.GetSize()\n",
    "        # temp = sitk.Image(int(sz[0]/scale), int(sz[1]/scale), int(sz[2]/scale), sitk.sitkUInt8)\n",
    "        # img = sitk.Resample(img, temp, transform=sitk.ScaleTransform(3, (1, 1, 1)))\n",
    "        # mask = sitk.Resample(mask, temp, transform=sitk.ScaleTransform(3, (1, 1, 1)))\n",
    "        # Confirm sizes are correct\n",
    "        sz = img.GetSize()\n",
    "        print(f\"Image size: {sz}\")\n",
    "\n",
    "        # Generate tiles of multiple top-down slices to plot\n",
    "        Xslices = [img[int(s*sz[0]), :, :] for s in slices] + [mask[int(s*sz[0]), :, :] for s in slices]\n",
    "        Yslices = [img[:, int(s*sz[1]), :] for s in slices] + [mask[:, int(s*sz[1]), :] for s in slices]\n",
    "        Zslices = [img[:, :, int(s*sz[2])] for s in slices] + [mask[:, :, int(s*sz[2])] for s in slices]\n",
    "        print(np.max(sitk.GetArrayFromImage(img)), np.max(sitk.GetArrayFromImage(mask)))\n",
    "        # print(type(mask), type(img))\n",
    "        \n",
    "        # Convert tiles to arrays for plotting\n",
    "        Xarr = sitk.GetArrayFromImage(sitk.Tile(Xslices, [len(slices), 2]))\n",
    "        Yarr = sitk.GetArrayFromImage(sitk.Tile(Yslices, [len(slices), 2]))\n",
    "        Zarr = sitk.GetArrayFromImage(sitk.Tile(Zslices, [len(slices), 2]))\n",
    "        # print(Xarr.shape, Yarr.shape, Zarr.shape)\n",
    "\n",
    "        # Plot tiles\n",
    "        fig, axs = plt.subplots(3, 1, constrained_layout=True)\n",
    "        # plt.figure()\n",
    "        axs[0].imshow(Xarr, cmap='gray')\n",
    "        axs[0].set_title(f\"X={sz[0]*slices}\")\n",
    "        axs[1].imshow(Yarr, cmap='gray')\n",
    "        axs[1].set_title(f\"Y={sz[1]*slices}\")\n",
    "        axs[2].imshow(Zarr, cmap='gray')\n",
    "        axs[2].set_title(f\"Z={sz[2]*slices}\")\n",
    "        fig.suptitle(f\"Images for patient {f[-8:-5]}\")\n",
    "        fig.set_dpi(150)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "dlhw",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}