BraTS_validation_cls_test.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import torch\n",
    "import torch.nn.functional as F\n",
    "import nibabel as nib\n",
    "from datasets.BRATS2018 import ToTensorVal\n",
    "from datasets.BRATS2018_3D import CenterCropBRATS3D, NormalizeBRATS3D\n",
    "from models.resnet3D import resnet50_3D\n",
    "from models.logistic import Logistic\n",
    "import os"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# helper functions\n",
    "def time_stamp() -> str:\n",
    "    ts = time.time()\n",
    "    time_stamp = datetime.datetime.fromtimestamp(ts).strftime('%Y-%m-%d %H:%M:%S')\n",
    "    return time_stamp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def classification_infer(case_name, val_dir, model):\n",
    "    case_dir = os.path.join(val_dir, case_name)\n",
    "    \n",
    "    t1 = nib.load(os.path.join(case_dir, case_name + '_t1.nii.gz')).get_data()\n",
    "    t1ce = nib.load(os.path.join(case_dir, case_name + '_t1ce.nii.gz')).get_data()\n",
    "    t2 = nib.load(os.path.join(case_dir, case_name + '_t2.nii.gz')).get_data()\n",
    "    flair = nib.load(os.path.join(case_dir, case_name + '_flair.nii.gz')).get_data()\n",
    "    \n",
    "    assert t1.shape == (240, 240, 155)\n",
    "    assert t1ce.shape == (240, 240, 155)\n",
    "    assert t2.shape == (240, 240, 155)\n",
    "    assert flair.shape == (240, 240, 155)\n",
    "    \n",
    "    sc = np.array([t1, t1ce, t2, flair])\n",
    "    assert sc.shape == (4, 240, 240, 155)\n",
    "    \n",
    "    # fake label for position\n",
    "    fake_label = np.array([1])\n",
    "    \n",
    "    center_crop = CenterCropBRATS3D()\n",
    "    normalize = NormalizeBRATS3D()\n",
    "    to_tensor = ToTensorVal()\n",
    "    \n",
    "    sc, fake_label = center_crop((sc, fake_label))\n",
    "    sc, fake_label = normalize((sc, fake_label))\n",
    "    sc = to_tensor(sc)\n",
    "    \n",
    "    # unsqueeze to 5-dimension array, NxCxHxWxD\n",
    "    sc = torch.unsqueeze(sc, dim=0)\n",
    "    sc = sc.to(device)\n",
    "    \n",
    "    with torch.no_grad():\n",
    "        label = model(sc)\n",
    "        label = torch.argmax(F.softmax(label, dim=1), dim=1, keepdim=True)\n",
    "    \n",
    "    # squeeze to rank-0 array\n",
    "    label = torch.squeeze(label)\n",
    "    label = label.cpu().numpy()\n",
    "    \n",
    "    return label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ResNet3D(\n",
       "  (conv1): Conv3d(4, 32, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1), bias=False)\n",
       "  (bn1): BatchNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "  (relu): ReLU(inplace)\n",
       "  (maxpool): MaxPool3d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)\n",
       "  (layer1): Sequential(\n",
       "    (0): Bottleneck3D(\n",
       "      (conv1): Conv3d(32, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn1): BatchNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv3d(32, 32, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), bias=False)\n",
       "      (bn2): BatchNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv3d(32, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn3): BatchNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace)\n",
       "      (downsample): Sequential(\n",
       "        (0): Conv3d(32, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "        (1): BatchNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      )\n",
       "    )\n",
       "    (1): Bottleneck3D(\n",
       "      (conv1): Conv3d(128, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn1): BatchNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv3d(32, 32, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), bias=False)\n",
       "      (bn2): BatchNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv3d(32, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn3): BatchNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace)\n",
       "    )\n",
       "    (2): Bottleneck3D(\n",
       "      (conv1): Conv3d(128, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn1): BatchNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv3d(32, 32, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), bias=False)\n",
       "      (bn2): BatchNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv3d(32, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn3): BatchNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace)\n",
       "    )\n",
       "  )\n",
       "  (layer2): Sequential(\n",
       "    (0): Bottleneck3D(\n",
       "      (conv1): Conv3d(128, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn1): BatchNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1), bias=False)\n",
       "      (bn2): BatchNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv3d(64, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn3): BatchNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace)\n",
       "      (downsample): Sequential(\n",
       "        (0): Conv3d(128, 256, kernel_size=(1, 1, 1), stride=(2, 2, 2), bias=False)\n",
       "        (1): BatchNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      )\n",
       "    )\n",
       "    (1): Bottleneck3D(\n",
       "      (conv1): Conv3d(256, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn1): BatchNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), bias=False)\n",
       "      (bn2): BatchNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv3d(64, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn3): BatchNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace)\n",
       "    )\n",
       "    (2): Bottleneck3D(\n",
       "      (conv1): Conv3d(256, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn1): BatchNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), bias=False)\n",
       "      (bn2): BatchNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv3d(64, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn3): BatchNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace)\n",
       "    )\n",
       "    (3): Bottleneck3D(\n",
       "      (conv1): Conv3d(256, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn1): BatchNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), bias=False)\n",
       "      (bn2): BatchNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv3d(64, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn3): BatchNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace)\n",
       "    )\n",
       "  )\n",
       "  (layer3): Sequential(\n",
       "    (0): Bottleneck3D(\n",
       "      (conv1): Conv3d(256, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn1): BatchNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1), bias=False)\n",
       "      (bn2): BatchNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv3d(128, 512, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn3): BatchNorm3d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace)\n",
       "      (downsample): Sequential(\n",
       "        (0): Conv3d(256, 512, kernel_size=(1, 1, 1), stride=(2, 2, 2), bias=False)\n",
       "        (1): BatchNorm3d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      )\n",
       "    )\n",
       "    (1): Bottleneck3D(\n",
       "      (conv1): Conv3d(512, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn1): BatchNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), bias=False)\n",
       "      (bn2): BatchNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv3d(128, 512, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn3): BatchNorm3d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace)\n",
       "    )\n",
       "    (2): Bottleneck3D(\n",
       "      (conv1): Conv3d(512, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn1): BatchNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), bias=False)\n",
       "      (bn2): BatchNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv3d(128, 512, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn3): BatchNorm3d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace)\n",
       "    )\n",
       "    (3): Bottleneck3D(\n",
       "      (conv1): Conv3d(512, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn1): BatchNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), bias=False)\n",
       "      (bn2): BatchNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv3d(128, 512, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn3): BatchNorm3d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace)\n",
       "    )\n",
       "    (4): Bottleneck3D(\n",
       "      (conv1): Conv3d(512, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn1): BatchNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), bias=False)\n",
       "      (bn2): BatchNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv3d(128, 512, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn3): BatchNorm3d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace)\n",
       "    )\n",
       "    (5): Bottleneck3D(\n",
       "      (conv1): Conv3d(512, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn1): BatchNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), bias=False)\n",
       "      (bn2): BatchNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv3d(128, 512, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn3): BatchNorm3d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace)\n",
       "    )\n",
       "  )\n",
       "  (layer4): Sequential(\n",
       "    (0): Bottleneck3D(\n",
       "      (conv1): Conv3d(512, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn1): BatchNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1), bias=False)\n",
       "      (bn2): BatchNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv3d(256, 1024, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn3): BatchNorm3d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace)\n",
       "      (downsample): Sequential(\n",
       "        (0): Conv3d(512, 1024, kernel_size=(1, 1, 1), stride=(2, 2, 2), bias=False)\n",
       "        (1): BatchNorm3d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      )\n",
       "    )\n",
       "    (1): Bottleneck3D(\n",
       "      (conv1): Conv3d(1024, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn1): BatchNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), bias=False)\n",
       "      (bn2): BatchNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv3d(256, 1024, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn3): BatchNorm3d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace)\n",
       "    )\n",
       "    (2): Bottleneck3D(\n",
       "      (conv1): Conv3d(1024, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn1): BatchNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), bias=False)\n",
       "      (bn2): BatchNorm3d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv3d(256, 1024, kernel_size=(1, 1, 1), stride=(1, 1, 1), bias=False)\n",
       "      (bn3): BatchNorm3d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace)\n",
       "    )\n",
       "  )\n",
       "  (avgpool): AdaptiveAvgPool3d(output_size=(1, 1, 1))\n",
       "  (fc): Linear(in_features=1024, out_features=2, bias=True)\n",
       ")"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "case_list = sorted(os.listdir('BRATS2018_Validation/'))\n",
    "val_dir = 'BRATS2018_Validation'\n",
    "device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')\n",
    "model = resnet50_3D(num_classes=2)\n",
    "#model = Logistic(volume=(4, 160, 160, 144), num_classes=2)\n",
    "model.load_state_dict(torch.load('../ResNet50-3D-32-CLS/trained_model.pt', map_location=device))\n",
    "model.to(device)\n",
    "model.eval()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## test classification for BraTS2018 validation set"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### resnet50 3D 16 lr 7e-4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Brats18_CBICA_AAM_1 - HGG\n",
      "Brats18_CBICA_ABT_1 - HGG\n",
      "Brats18_CBICA_ALA_1 - HGG\n",
      "Brats18_CBICA_ALT_1 - HGG\n",
      "Brats18_CBICA_ALV_1 - HGG\n",
      "Brats18_CBICA_ALZ_1 - HGG\n",
      "Brats18_CBICA_AMF_1 - HGG\n",
      "Brats18_CBICA_AMU_1 - HGG\n",
      "Brats18_CBICA_ANK_1 - HGG\n",
      "Brats18_CBICA_APM_1 - HGG\n",
      "Brats18_CBICA_AQE_1 - HGG\n",
      "Brats18_CBICA_ARR_1 - HGG\n",
      "Brats18_CBICA_ATW_1 - HGG\n",
      "Brats18_CBICA_AUC_1 - HGG\n",
      "Brats18_CBICA_AUE_1 - HGG\n",
      "Brats18_CBICA_AZA_1 - HGG\n",
      "Brats18_CBICA_BHF_1 - HGG\n",
      "Brats18_CBICA_BHN_1 - HGG\n",
      "Brats18_CBICA_BKY_1 - HGG\n",
      "Brats18_CBICA_BLI_1 - HGG\n",
      "Brats18_CBICA_BLK_1 - HGG\n",
      "Brats18_MDA_1012_1 - HGG\n",
      "Brats18_MDA_1015_1 - HGG\n",
      "Brats18_MDA_1081_1 - HGG\n",
      "Brats18_MDA_907_1 - HGG\n",
      "Brats18_MDA_922_1 - HGG\n",
      "Brats18_TCIA02_230_1 - HGG\n",
      "Brats18_TCIA02_400_1 - HGG\n",
      "Brats18_TCIA03_216_1 - HGG\n",
      "Brats18_TCIA03_288_1 - HGG\n",
      "Brats18_TCIA03_313_1 - HGG\n",
      "Brats18_TCIA03_604_1 - HGG\n",
      "Brats18_TCIA04_212_1 - HGG\n",
      "Brats18_TCIA04_253_1 - HGG\n",
      "Brats18_TCIA07_600_1 - HGG\n",
      "Brats18_TCIA07_601_1 - HGG\n",
      "Brats18_TCIA07_602_1 - HGG\n",
      "Brats18_TCIA09_248_1 - LGG\n",
      "Brats18_TCIA10_195_1 - LGG\n",
      "Brats18_TCIA10_311_1 - LGG\n",
      "Brats18_TCIA10_609_1 - LGG\n",
      "Brats18_TCIA11_612_1 - HGG\n",
      "Brats18_TCIA12_613_1 - HGG\n",
      "Brats18_TCIA13_610_1 - HGG\n",
      "Brats18_TCIA13_611_1 - HGG\n",
      "Brats18_TCIA13_617_1 - LGG\n",
      "Brats18_TCIA13_636_1 - HGG\n",
      "Brats18_TCIA13_638_1 - LGG\n",
      "Brats18_TCIA13_646_1 - HGG\n",
      "Brats18_TCIA13_652_1 - LGG\n",
      "Brats18_UAB_3446_1 - HGG\n",
      "Brats18_UAB_3448_1 - HGG\n",
      "Brats18_UAB_3449_1 - HGG\n",
      "Brats18_UAB_3454_1 - HGG\n",
      "Brats18_UAB_3455_1 - HGG\n",
      "Brats18_UAB_3456_1 - HGG\n",
      "Brats18_UAB_3490_1 - HGG\n",
      "Brats18_UAB_3498_1 - HGG\n",
      "Brats18_UAB_3499_1 - LGG\n",
      "Brats18_WashU_S036_1 - HGG\n",
      "Brats18_WashU_S037_1 - LGG\n",
      "Brats18_WashU_S041_1 - HGG\n",
      "Brats18_WashU_W033_1 - HGG\n",
      "Brats18_WashU_W038_1 - HGG\n",
      "Brats18_WashU_W047_1 - HGG\n",
      "Brats18_WashU_W053_1 - HGG\n"
     ]
    }
   ],
   "source": [
    "for case_name in case_list:\n",
    "    label = classification_infer(case_name, val_dir, model)\n",
    "    if label == 0.:\n",
    "        label = 'LGG'\n",
    "    else:\n",
    "        label = 'HGG'\n",
    "        \n",
    "    print(f'{case_name} - {label}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### resnet50 3D 16 lr 1e-3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Brats18_CBICA_AAM_1 - 1\n",
      "Brats18_CBICA_ABT_1 - 1\n",
      "Brats18_CBICA_ALA_1 - 1\n",
      "Brats18_CBICA_ALT_1 - 1\n",
      "Brats18_CBICA_ALV_1 - 1\n",
      "Brats18_CBICA_ALZ_1 - 1\n",
      "Brats18_CBICA_AMF_1 - 1\n",
      "Brats18_CBICA_AMU_1 - 1\n",
      "Brats18_CBICA_ANK_1 - 1\n",
      "Brats18_CBICA_APM_1 - 1\n",
      "Brats18_CBICA_AQE_1 - 1\n",
      "Brats18_CBICA_ARR_1 - 1\n",
      "Brats18_CBICA_ATW_1 - 1\n",
      "Brats18_CBICA_AUC_1 - 1\n",
      "Brats18_CBICA_AUE_1 - 1\n",
      "Brats18_CBICA_AZA_1 - 1\n",
      "Brats18_CBICA_BHF_1 - 1\n",
      "Brats18_CBICA_BHN_1 - 1\n",
      "Brats18_CBICA_BKY_1 - 1\n",
      "Brats18_CBICA_BLI_1 - 1\n",
      "Brats18_CBICA_BLK_1 - 1\n",
      "Brats18_MDA_1012_1 - 1\n",
      "Brats18_MDA_1015_1 - 1\n",
      "Brats18_MDA_1081_1 - 1\n",
      "Brats18_MDA_907_1 - 1\n",
      "Brats18_MDA_922_1 - 1\n",
      "Brats18_TCIA02_230_1 - 1\n",
      "Brats18_TCIA02_400_1 - 1\n",
      "Brats18_TCIA03_216_1 - 1\n",
      "Brats18_TCIA03_288_1 - 1\n",
      "Brats18_TCIA03_313_1 - 1\n",
      "Brats18_TCIA03_604_1 - 1\n",
      "Brats18_TCIA04_212_1 - 1\n",
      "Brats18_TCIA04_253_1 - 1\n",
      "Brats18_TCIA07_600_1 - 1\n",
      "Brats18_TCIA07_601_1 - 1\n",
      "Brats18_TCIA07_602_1 - 1\n",
      "Brats18_TCIA09_248_1 - 0\n",
      "Brats18_TCIA10_195_1 - 0\n",
      "Brats18_TCIA10_311_1 - 0\n",
      "Brats18_TCIA10_609_1 - 0\n",
      "Brats18_TCIA11_612_1 - 1\n",
      "Brats18_TCIA12_613_1 - 1\n",
      "Brats18_TCIA13_610_1 - 1\n",
      "Brats18_TCIA13_611_1 - 1\n",
      "Brats18_TCIA13_617_1 - 0\n",
      "Brats18_TCIA13_636_1 - 1\n",
      "Brats18_TCIA13_638_1 - 0\n",
      "Brats18_TCIA13_646_1 - 1\n",
      "Brats18_TCIA13_652_1 - 0\n",
      "Brats18_UAB_3446_1 - 1\n",
      "Brats18_UAB_3448_1 - 1\n",
      "Brats18_UAB_3449_1 - 1\n",
      "Brats18_UAB_3454_1 - 1\n",
      "Brats18_UAB_3455_1 - 1\n",
      "Brats18_UAB_3456_1 - 1\n",
      "Brats18_UAB_3490_1 - 1\n",
      "Brats18_UAB_3498_1 - 1\n",
      "Brats18_UAB_3499_1 - 0\n",
      "Brats18_WashU_S036_1 - 1\n",
      "Brats18_WashU_S037_1 - 0\n",
      "Brats18_WashU_S041_1 - 1\n",
      "Brats18_WashU_W033_1 - 1\n",
      "Brats18_WashU_W038_1 - 0\n",
      "Brats18_WashU_W047_1 - 1\n",
      "Brats18_WashU_W053_1 - 1\n"
     ]
    }
   ],
   "source": [
    "for case_name in case_list:\n",
    "    label = classification_infer(case_name, val_dir, model)\n",
    "    print(f'{case_name} - {label}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### resnet50 3D 32"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Brats18_CBICA_AAM_1 - 1\n",
      "Brats18_CBICA_ABT_1 - 1\n",
      "Brats18_CBICA_ALA_1 - 1\n",
      "Brats18_CBICA_ALT_1 - 1\n",
      "Brats18_CBICA_ALV_1 - 1\n",
      "Brats18_CBICA_ALZ_1 - 1\n",
      "Brats18_CBICA_AMF_1 - 1\n",
      "Brats18_CBICA_AMU_1 - 1\n",
      "Brats18_CBICA_ANK_1 - 1\n",
      "Brats18_CBICA_APM_1 - 1\n",
      "Brats18_CBICA_AQE_1 - 1\n",
      "Brats18_CBICA_ARR_1 - 1\n",
      "Brats18_CBICA_ATW_1 - 1\n",
      "Brats18_CBICA_AUC_1 - 1\n",
      "Brats18_CBICA_AUE_1 - 1\n",
      "Brats18_CBICA_AZA_1 - 1\n",
      "Brats18_CBICA_BHF_1 - 1\n",
      "Brats18_CBICA_BHN_1 - 1\n",
      "Brats18_CBICA_BKY_1 - 1\n",
      "Brats18_CBICA_BLI_1 - 1\n",
      "Brats18_CBICA_BLK_1 - 1\n",
      "Brats18_MDA_1012_1 - 1\n",
      "Brats18_MDA_1015_1 - 1\n",
      "Brats18_MDA_1081_1 - 1\n",
      "Brats18_MDA_907_1 - 1\n",
      "Brats18_MDA_922_1 - 1\n",
      "Brats18_TCIA02_230_1 - 1\n",
      "Brats18_TCIA02_400_1 - 1\n",
      "Brats18_TCIA03_216_1 - 1\n",
      "Brats18_TCIA03_288_1 - 1\n",
      "Brats18_TCIA03_313_1 - 1\n",
      "Brats18_TCIA03_604_1 - 1\n",
      "Brats18_TCIA04_212_1 - 1\n",
      "Brats18_TCIA04_253_1 - 1\n",
      "Brats18_TCIA07_600_1 - 1\n",
      "Brats18_TCIA07_601_1 - 1\n",
      "Brats18_TCIA07_602_1 - 1\n",
      "Brats18_TCIA09_248_1 - 0\n",
      "Brats18_TCIA10_195_1 - 0\n",
      "Brats18_TCIA10_311_1 - 0\n",
      "Brats18_TCIA10_609_1 - 0\n",
      "Brats18_TCIA11_612_1 - 1\n",
      "Brats18_TCIA12_613_1 - 1\n",
      "Brats18_TCIA13_610_1 - 1\n",
      "Brats18_TCIA13_611_1 - 1\n",
      "Brats18_TCIA13_617_1 - 0\n",
      "Brats18_TCIA13_636_1 - 1\n",
      "Brats18_TCIA13_638_1 - 0\n",
      "Brats18_TCIA13_646_1 - 1\n",
      "Brats18_TCIA13_652_1 - 0\n",
      "Brats18_UAB_3446_1 - 1\n",
      "Brats18_UAB_3448_1 - 1\n",
      "Brats18_UAB_3449_1 - 1\n",
      "Brats18_UAB_3454_1 - 1\n",
      "Brats18_UAB_3455_1 - 1\n",
      "Brats18_UAB_3456_1 - 1\n",
      "Brats18_UAB_3490_1 - 1\n",
      "Brats18_UAB_3498_1 - 1\n",
      "Brats18_UAB_3499_1 - 0\n",
      "Brats18_WashU_S036_1 - 1\n",
      "Brats18_WashU_S037_1 - 0\n",
      "Brats18_WashU_S041_1 - 1\n",
      "Brats18_WashU_W033_1 - 1\n",
      "Brats18_WashU_W038_1 - 1\n",
      "Brats18_WashU_W047_1 - 1\n",
      "Brats18_WashU_W053_1 - 1\n"
     ]
    }
   ],
   "source": [
    "for case_name in case_list:\n",
    "    label = classification_infer(case_name, val_dir, model)\n",
    "    print(f'{case_name} - {label}')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.6.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}