step01_CNNmodel_02-CV0-Copy1.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import random\n",
    "import numpy as np\n",
    "import math\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import cv2\n",
    "import albumentations as albu\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "from torch.utils.data import DataLoader\n",
    "from torch.utils.data import Dataset as BaseDataset\n",
    "from tqdm.notebook import trange, tqdm\n",
    "from torch.utils.tensorboard import SummaryWriter\n",
    "from torchsampler import ImbalancedDatasetSampler\n",
    "from torchvision import models\n",
    "import torchvision.transforms as transforms\n",
    "#from torchsummary import summary\n",
    "import torch.nn.functional as F\n",
    "from sklearn.model_selection import StratifiedKFold\n",
    "from sklearn.metrics import roc_auc_score, classification_report, roc_curve, average_precision_score\n",
    "from sklearn.model_selection import GroupShuffleSplit\n",
    "from pydicom import dcmread\n",
    "import glob\n",
    "import pickle\n",
    "import scipy\n",
    "from efficientnet_pytorch import EfficientNet\n",
    "from datetime import datetime\n",
    "seed = 10\n",
    "np.random.seed(seed)\n",
    "random.seed(seed)\n",
    "os.environ['PYTHONHASHSEED'] = str(seed)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataDF = pd.read_csv('forTrainingDataFold.csv', dtype={'StudyInstanceUID': 'string', 'SeriesInstanceUID':'string', 'SOPInstanceUID':'string'})\n",
    "dataDF = dataDF.set_index('SOPInstanceUID')\n",
    "dataDF = dataDF.drop(['StudyInstanceUID','SeriesInstanceUID','rv_lv_ratio_lt_1'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>pe_present_on_image</th>\n",
       "      <th>acute</th>\n",
       "      <th>chronic</th>\n",
       "      <th>leftsided_pe</th>\n",
       "      <th>rightsided_pe</th>\n",
       "      <th>central_pe</th>\n",
       "      <th>rv_lv_ratio_gte_1</th>\n",
       "      <th>qa_motion</th>\n",
       "      <th>qa_contrast</th>\n",
       "      <th>flow_artifact</th>\n",
       "      <th>true_filling_defect_not_pe</th>\n",
       "      <th>negative_exam_for_pe</th>\n",
       "      <th>chronic_pe</th>\n",
       "      <th>acute_and_chronic_pe</th>\n",
       "      <th>indeterminate</th>\n",
       "      <th>img_pos</th>\n",
       "      <th>patient_position</th>\n",
       "      <th>fold</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SOPInstanceUID</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>c0f3cb036d06</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-234.5</td>\n",
       "      <td>HFS</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>f57ffd3883b6</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-252.5</td>\n",
       "      <td>HFS</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41220fda34a3</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-432.5</td>\n",
       "      <td>HFS</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13b685b4b14f</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-434.5</td>\n",
       "      <td>HFS</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>be0b7524ffb4</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-436.5</td>\n",
       "      <td>HFS</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                pe_present_on_image  acute  chronic  leftsided_pe  \\\n",
       "SOPInstanceUID                                                      \n",
       "c0f3cb036d06                      0      1        0             1   \n",
       "f57ffd3883b6                      0      1        0             1   \n",
       "41220fda34a3                      0      1        0             1   \n",
       "13b685b4b14f                      0      1        0             1   \n",
       "be0b7524ffb4                      0      1        0             1   \n",
       "\n",
       "                rightsided_pe  central_pe  rv_lv_ratio_gte_1  qa_motion  \\\n",
       "SOPInstanceUID                                                            \n",
       "c0f3cb036d06                1           0                  0          0   \n",
       "f57ffd3883b6                1           0                  0          0   \n",
       "41220fda34a3                1           0                  0          0   \n",
       "13b685b4b14f                1           0                  0          0   \n",
       "be0b7524ffb4                1           0                  0          0   \n",
       "\n",
       "                qa_contrast  flow_artifact  true_filling_defect_not_pe  \\\n",
       "SOPInstanceUID                                                           \n",
       "c0f3cb036d06              0              0                           0   \n",
       "f57ffd3883b6              0              0                           0   \n",
       "41220fda34a3              0              0                           0   \n",
       "13b685b4b14f              0              0                           0   \n",
       "be0b7524ffb4              0              0                           0   \n",
       "\n",
       "                negative_exam_for_pe  chronic_pe  acute_and_chronic_pe  \\\n",
       "SOPInstanceUID                                                           \n",
       "c0f3cb036d06                       0           0                     0   \n",
       "f57ffd3883b6                       0           0                     0   \n",
       "41220fda34a3                       0           0                     0   \n",
       "13b685b4b14f                       0           0                     0   \n",
       "be0b7524ffb4                       0           0                     0   \n",
       "\n",
       "                indeterminate  img_pos patient_position  fold  \n",
       "SOPInstanceUID                                                 \n",
       "c0f3cb036d06                0   -234.5              HFS     3  \n",
       "f57ffd3883b6                0   -252.5              HFS     3  \n",
       "41220fda34a3                0   -432.5              HFS     3  \n",
       "13b685b4b14f                0   -434.5              HFS     3  \n",
       "be0b7524ffb4                0   -436.5              HFS     3  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataDF.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Construct path mapping dict for jpg files \n",
    "listOfDCMfiles = glob.glob('data/train/*/*/*.dcm')\n",
    "\n",
    "imageID2pathDict = {}\n",
    "\n",
    "for eachPath in listOfDCMfiles:\n",
    "    imageID = eachPath.split('/')[-1][:-4]\n",
    "    imageID2pathDict[imageID] = eachPath"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load training labels\n",
    "gtLabelDict = pickle.load(open('data/CNNtrainingLabel.p','rb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0. 0. 0. 0. 0. 0. 0. 1.]\n"
     ]
    }
   ],
   "source": [
    "sampleImageID = list(gtLabelDict.keys())[5]\n",
    "print(gtLabelDict[sampleImageID])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# helper Functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Helper functions\n",
    "\n",
    "def get_training_augmentation():\n",
    "    train_transform = [\n",
    "        albu.ShiftScaleRotate(scale_limit=0.05, rotate_limit=30, shift_limit=0.05, p=1, border_mode=0),\n",
    "        albu.IAAAdditiveGaussianNoise(p=0.1),\n",
    "        albu.IAAPerspective(p=0.2),\n",
    "\n",
    "        albu.OneOf(\n",
    "            [\n",
    "                albu.CLAHE(p=1),\n",
    "                albu.RandomBrightness(p=1),\n",
    "                albu.RandomGamma(p=1),\n",
    "            ],\n",
    "            p=0.3,\n",
    "        ),\n",
    "\n",
    "        albu.OneOf(\n",
    "            [\n",
    "                albu.IAASharpen(p=1),\n",
    "                albu.Blur(blur_limit=3, p=1),\n",
    "                albu.MotionBlur(blur_limit=3, p=1),\n",
    "            ],\n",
    "            p=0.3,\n",
    "        ),\n",
    "\n",
    "        albu.OneOf(\n",
    "            [\n",
    "                albu.RandomContrast(p=1),\n",
    "                albu.HueSaturationValue(p=1),\n",
    "            ],\n",
    "            p=0.3,\n",
    "        ),\n",
    "    ]\n",
    "    return albu.Compose(train_transform)\n",
    "\n",
    "\n",
    "def get_validation_augmentation():\n",
    "    \"\"\"Add paddings to make image shape divisible by 32\"\"\"\n",
    "    '''\n",
    "    test_transform = [\n",
    "        albu.PadIfNeeded(384, 480)\n",
    "    ]\n",
    "    '''\n",
    "    return albu.Compose(test_transform)\n",
    "\n",
    "\n",
    "def to_tensor(x, **kwargs):\n",
    "    return x.transpose(2, 0, 1).astype('float32')\n",
    "\n",
    "data_transform = transforms.Compose([\n",
    "        transforms.ToTensor(),\n",
    "        transforms.Normalize(mean=[0.485, 0.456, 0.406],\n",
    "                             std=[0.229, 0.224, 0.225])\n",
    "    ])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def window(img, WL=50, WW=350):\n",
    "    upper, lower = WL+WW//2, WL-WW//2\n",
    "    X = np.clip(img.copy(), lower, upper)\n",
    "    X = X - np.min(X)\n",
    "    X = X / np.max(X)\n",
    "    X = (X*255.0).astype('uint8')\n",
    "    return X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "class Dataset(BaseDataset):\n",
    "    def __init__(\n",
    "            self, \n",
    "            dataframe=None, \n",
    "            augmentation=None,\n",
    "            transform=None,\n",
    "            dirPath=None,\n",
    "    ):\n",
    "        self.dataframe = dataframe\n",
    "        self.ids = self.dataframe.index.values.tolist()     \n",
    "        self.augmentation = augmentation\n",
    "        self.transform=transform\n",
    "        self.dirPath = dirPath\n",
    "    \n",
    "    def __getitem__(self, i):\n",
    "        thisID = self.ids[i]\n",
    "        \n",
    "        #jpgPath = imageID2pathDict[thisID]\n",
    "        #image = cv2.imread(jpgPath)\n",
    "        \n",
    "        dcmPath = imageID2pathDict[thisID]\n",
    "        dcm_data = dcmread(dcmPath)\n",
    "        image = dcm_data.pixel_array * int(dcm_data.RescaleSlope) + int(dcm_data.RescaleIntercept)\n",
    "        image = np.stack([window(image, WL=-600, WW=1500),\n",
    "                          window(image, WL=40, WW=400),\n",
    "                          window(image, WL=100, WW=700)], 2)\n",
    "        \n",
    "        target = gtLabelDict[thisID]    \n",
    "        target = target.astype(np.float32)\n",
    "        \n",
    "        # apply augmentations\n",
    "        if self.augmentation:\n",
    "            sample = self.augmentation(image=image)\n",
    "            image = sample['image']\n",
    "        \n",
    "        '''\n",
    "        # apply preprocessing\n",
    "        if self.preprocessing:\n",
    "            sample = self.preprocessing(image=image)\n",
    "            image = sample['image']\n",
    "        '''\n",
    "        \n",
    "        image = image.astype(np.float32)\n",
    "        #image = np.rollaxis(image, -1, 0)  \n",
    "        #image = image.transpose((2, 0, 1))\n",
    "        \n",
    "        if self.transform:\n",
    "            image = self.transform(image)\n",
    "        \n",
    "        return image, target\n",
    "        \n",
    "    def __len__(self):\n",
    "        return len(self.ids)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "def resnet50_model():\n",
    "    myModel = models.resnet50(pretrained=True)\n",
    "    num_ftrs = myModel.fc.in_features\n",
    "    myModel.fc = nn.Sequential(\n",
    "        nn.Linear(num_ftrs, 256),\n",
    "        nn.ReLU(),\n",
    "        nn.Dropout(p = 0.2),\n",
    "        nn.Linear(256, 64),\n",
    "        nn.ReLU(),\n",
    "        nn.Linear(64, 8),\n",
    "        )\n",
    "    return myModel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Custom weighted loss function\n",
    "class customWeightedBCEwithLogits(nn.Module):\n",
    "    def __init__(self, PE_pos_weight = 3.0, other_pos_weight = [30.0, 30.0, 3.0, 3.0, 3.0, 1.2, 0.5]):\n",
    "        super(customWeightedBCEwithLogits, self).__init__()\n",
    "        self.image_PE_PosWeight = torch.tensor(PE_pos_weight, requires_grad=False).cuda()\n",
    "        self.otherLabels_PosWeight = torch.tensor(other_pos_weight, requires_grad=False).cuda()\n",
    "        self.criterion1 = nn.BCEWithLogitsLoss(pos_weight=self.image_PE_PosWeight)\n",
    "        self.criterion2 = nn.BCEWithLogitsLoss(pos_weight=self.otherLabels_PosWeight)\n",
    "\n",
    "    def forward(self, inputs, targets):\n",
    "        loss1 = self.criterion1(inputs[:,0:1], targets[:,0:1])\n",
    "        loss2 = self.criterion2(inputs[:,1:], targets[:,1:])\n",
    "        return loss1+loss2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "def train_loop(model, train_loader, valid_loader):\n",
    "    # Train one epoch\n",
    "    train_total = train_correct = train_cost = 0\n",
    "    model.train()\n",
    "    for x, y in tqdm(train_loader):\n",
    "        x = x.cuda()\n",
    "        y = y.cuda()\n",
    "        optimizer.zero_grad()\n",
    "        z = model(x)\n",
    "        train_total += y.size(0)\n",
    "        train_correct += ((torch.sigmoid(z[:,0])>0.5) == (y[:,0]>0.5)).sum().item()\n",
    "        loss = customLoss(z, y)\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "        train_cost += loss.item()\n",
    "    return train_cost/train_total, train_correct/train_total\n",
    "\n",
    "def valid_loop(model, train_loader, valid_loader):\n",
    "    # Evaluate on validation  data \n",
    "    val_total = val_correct = val_cost = 0\n",
    "    model.eval()\n",
    "    with torch.no_grad():\n",
    "        for x_val, y_val in tqdm(valid_loader):\n",
    "            x_val = x_val.cuda()\n",
    "            y_val = y_val.cuda()\n",
    "            z = model(x_val)\n",
    "            val_total += y_val.size(0)\n",
    "            val_correct += ((torch.sigmoid(z[:,0])>0.5) == (y_val[:,0]>0.5)).sum().item()\n",
    "            loss = customLoss(z, y_val)\n",
    "            val_cost += loss.item()\n",
    "    return val_cost/val_total, val_correct/val_total\n",
    "\n",
    "def main_loop(n_epochs, model, train_loader, valid_loader):\n",
    "    for epoch in range(n_epochs):\n",
    "        print('epoch ' + str(epoch) + ':')\n",
    "        train_avgCost, train_acc = train_loop(model, train_loader, valid_loader)\n",
    "        val_avgCost, val_acc = valid_loop(model, train_loader, valid_loader)\n",
    "        print('train_cost: %.4f, train_acc: %.4f, val_cost: %.4f, val_acc: %.4f'\\\n",
    "              % (train_avgCost, train_acc, val_avgCost, val_acc))\n",
    "        now = datetime.now().strftime(\"%Y%m%d_%H%M\")\n",
    "        modelPath = 'models/CNNmodel/CNNmodel_01_cv0_epoch' + str(epoch) + '_' + now +'.pth'\n",
    "        print('saving: ',modelPath)\n",
    "        torch.save(model, modelPath)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "myModel = resnet50_model()\n",
    "myModel = myModel.cuda()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Prepare train variables and parameters\n",
    "col_names = ['train_cost', 'train_acc', 'val_cost', 'val_acc']\n",
    "resultsDF = pd.DataFrame(columns = col_names)\n",
    "\n",
    "epochCount = 0\n",
    "optimizer =torch.optim.Adam(myModel.parameters(), lr=0.00005)\n",
    "\n",
    "customLoss = customWeightedBCEwithLogits()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# prepare dataset and dataloader\n",
    "preTrainDF = dataDF[dataDF['fold']==4]\n",
    "trainDF = dataDF[dataDF['fold']!=0]\n",
    "valDF = dataDF[dataDF['fold']==0]\n",
    "\n",
    "my_pretrain_dataset = Dataset(\n",
    "    dataframe= preTrainDF,\n",
    "    augmentation=get_training_augmentation(),\n",
    "    transform=data_transform,\n",
    ")\n",
    "\n",
    "my_train_dataset = Dataset(\n",
    "    dataframe= trainDF,\n",
    "    augmentation=get_training_augmentation(),\n",
    "    transform=data_transform,\n",
    ")\n",
    "\n",
    "my_val_dataset = Dataset(\n",
    "    dataframe= valDF,\n",
    "    augmentation=None,\n",
    "    transform=data_transform,\n",
    ")\n",
    "\n",
    "myPreTrainLoader = DataLoader(my_pretrain_dataset, batch_size=48, shuffle=True, num_workers=4)\n",
    "myTrainLoader = DataLoader(my_train_dataset, batch_size=42, shuffle=True, num_workers=4)\n",
    "myValidLoader = DataLoader(my_val_dataset, batch_size=42, shuffle=True, num_workers=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1429585\n",
      "label: [0. 0. 0. 0. 0. 0. 0. 1.]\n",
      "(8,)\n",
      "image shape: torch.Size([3, 512, 512])\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Sanity Check\n",
    "\n",
    "print(my_train_dataset.__len__())\n",
    "\n",
    "oneItem = my_pretrain_dataset.__getitem__(35)\n",
    "print('label:', oneItem[1])\n",
    "print(oneItem[1].shape)\n",
    "print('image shape:', oneItem[0].shape)\n",
    "for eachInd in range(3):\n",
    "    plt.figure()\n",
    "    plt.imshow(oneItem[0][eachInd,:,:], cmap='gray')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# pre-Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "block index: 0 conv1 weight\n",
      "block index: 1 bn1 weight\n",
      "block index: 1 bn1 bias\n",
      "block index: 4 layer1 0.conv1.weight\n",
      "block index: 4 layer1 0.bn1.weight\n",
      "block index: 4 layer1 0.bn1.bias\n",
      "block index: 4 layer1 0.conv2.weight\n",
      "block index: 4 layer1 0.bn2.weight\n",
      "block index: 4 layer1 0.bn2.bias\n",
      "block index: 4 layer1 0.conv3.weight\n",
      "block index: 4 layer1 0.bn3.weight\n",
      "block index: 4 layer1 0.bn3.bias\n",
      "block index: 4 layer1 0.downsample.0.weight\n",
      "block index: 4 layer1 0.downsample.1.weight\n",
      "block index: 4 layer1 0.downsample.1.bias\n",
      "block index: 4 layer1 1.conv1.weight\n",
      "block index: 4 layer1 1.bn1.weight\n",
      "block index: 4 layer1 1.bn1.bias\n",
      "block index: 4 layer1 1.conv2.weight\n",
      "block index: 4 layer1 1.bn2.weight\n",
      "block index: 4 layer1 1.bn2.bias\n",
      "block index: 4 layer1 1.conv3.weight\n",
      "block index: 4 layer1 1.bn3.weight\n",
      "block index: 4 layer1 1.bn3.bias\n",
      "block index: 4 layer1 2.conv1.weight\n",
      "block index: 4 layer1 2.bn1.weight\n",
      "block index: 4 layer1 2.bn1.bias\n",
      "block index: 4 layer1 2.conv2.weight\n",
      "block index: 4 layer1 2.bn2.weight\n",
      "block index: 4 layer1 2.bn2.bias\n",
      "block index: 4 layer1 2.conv3.weight\n",
      "block index: 4 layer1 2.bn3.weight\n",
      "block index: 4 layer1 2.bn3.bias\n",
      "block index: 5 layer2 0.conv1.weight\n",
      "block index: 5 layer2 0.bn1.weight\n",
      "block index: 5 layer2 0.bn1.bias\n",
      "block index: 5 layer2 0.conv2.weight\n",
      "block index: 5 layer2 0.bn2.weight\n",
      "block index: 5 layer2 0.bn2.bias\n",
      "block index: 5 layer2 0.conv3.weight\n",
      "block index: 5 layer2 0.bn3.weight\n",
      "block index: 5 layer2 0.bn3.bias\n",
      "block index: 5 layer2 0.downsample.0.weight\n",
      "block index: 5 layer2 0.downsample.1.weight\n",
      "block index: 5 layer2 0.downsample.1.bias\n",
      "block index: 5 layer2 1.conv1.weight\n",
      "block index: 5 layer2 1.bn1.weight\n",
      "block index: 5 layer2 1.bn1.bias\n",
      "block index: 5 layer2 1.conv2.weight\n",
      "block index: 5 layer2 1.bn2.weight\n",
      "block index: 5 layer2 1.bn2.bias\n",
      "block index: 5 layer2 1.conv3.weight\n",
      "block index: 5 layer2 1.bn3.weight\n",
      "block index: 5 layer2 1.bn3.bias\n",
      "block index: 5 layer2 2.conv1.weight\n",
      "block index: 5 layer2 2.bn1.weight\n",
      "block index: 5 layer2 2.bn1.bias\n",
      "block index: 5 layer2 2.conv2.weight\n",
      "block index: 5 layer2 2.bn2.weight\n",
      "block index: 5 layer2 2.bn2.bias\n",
      "block index: 5 layer2 2.conv3.weight\n",
      "block index: 5 layer2 2.bn3.weight\n",
      "block index: 5 layer2 2.bn3.bias\n",
      "block index: 5 layer2 3.conv1.weight\n",
      "block index: 5 layer2 3.bn1.weight\n",
      "block index: 5 layer2 3.bn1.bias\n",
      "block index: 5 layer2 3.conv2.weight\n",
      "block index: 5 layer2 3.bn2.weight\n",
      "block index: 5 layer2 3.bn2.bias\n",
      "block index: 5 layer2 3.conv3.weight\n",
      "block index: 5 layer2 3.bn3.weight\n",
      "block index: 5 layer2 3.bn3.bias\n",
      "block index: 6 layer3 0.conv1.weight\n",
      "block index: 6 layer3 0.bn1.weight\n",
      "block index: 6 layer3 0.bn1.bias\n",
      "block index: 6 layer3 0.conv2.weight\n",
      "block index: 6 layer3 0.bn2.weight\n",
      "block index: 6 layer3 0.bn2.bias\n",
      "block index: 6 layer3 0.conv3.weight\n",
      "block index: 6 layer3 0.bn3.weight\n",
      "block index: 6 layer3 0.bn3.bias\n",
      "block index: 6 layer3 0.downsample.0.weight\n",
      "block index: 6 layer3 0.downsample.1.weight\n",
      "block index: 6 layer3 0.downsample.1.bias\n",
      "block index: 6 layer3 1.conv1.weight\n",
      "block index: 6 layer3 1.bn1.weight\n",
      "block index: 6 layer3 1.bn1.bias\n",
      "block index: 6 layer3 1.conv2.weight\n",
      "block index: 6 layer3 1.bn2.weight\n",
      "block index: 6 layer3 1.bn2.bias\n",
      "block index: 6 layer3 1.conv3.weight\n",
      "block index: 6 layer3 1.bn3.weight\n",
      "block index: 6 layer3 1.bn3.bias\n",
      "block index: 6 layer3 2.conv1.weight\n",
      "block index: 6 layer3 2.bn1.weight\n",
      "block index: 6 layer3 2.bn1.bias\n",
      "block index: 6 layer3 2.conv2.weight\n",
      "block index: 6 layer3 2.bn2.weight\n",
      "block index: 6 layer3 2.bn2.bias\n",
      "block index: 6 layer3 2.conv3.weight\n",
      "block index: 6 layer3 2.bn3.weight\n",
      "block index: 6 layer3 2.bn3.bias\n",
      "block index: 6 layer3 3.conv1.weight\n",
      "block index: 6 layer3 3.bn1.weight\n",
      "block index: 6 layer3 3.bn1.bias\n",
      "block index: 6 layer3 3.conv2.weight\n",
      "block index: 6 layer3 3.bn2.weight\n",
      "block index: 6 layer3 3.bn2.bias\n",
      "block index: 6 layer3 3.conv3.weight\n",
      "block index: 6 layer3 3.bn3.weight\n",
      "block index: 6 layer3 3.bn3.bias\n",
      "block index: 6 layer3 4.conv1.weight\n",
      "block index: 6 layer3 4.bn1.weight\n",
      "block index: 6 layer3 4.bn1.bias\n",
      "block index: 6 layer3 4.conv2.weight\n",
      "block index: 6 layer3 4.bn2.weight\n",
      "block index: 6 layer3 4.bn2.bias\n",
      "block index: 6 layer3 4.conv3.weight\n",
      "block index: 6 layer3 4.bn3.weight\n",
      "block index: 6 layer3 4.bn3.bias\n",
      "block index: 6 layer3 5.conv1.weight\n",
      "block index: 6 layer3 5.bn1.weight\n",
      "block index: 6 layer3 5.bn1.bias\n",
      "block index: 6 layer3 5.conv2.weight\n",
      "block index: 6 layer3 5.bn2.weight\n",
      "block index: 6 layer3 5.bn2.bias\n",
      "block index: 6 layer3 5.conv3.weight\n",
      "block index: 6 layer3 5.bn3.weight\n",
      "block index: 6 layer3 5.bn3.bias\n",
      "block index: 7 layer4 0.conv1.weight\n",
      "block index: 7 layer4 0.bn1.weight\n",
      "block index: 7 layer4 0.bn1.bias\n",
      "block index: 7 layer4 0.conv2.weight\n",
      "block index: 7 layer4 0.bn2.weight\n",
      "block index: 7 layer4 0.bn2.bias\n",
      "block index: 7 layer4 0.conv3.weight\n",
      "block index: 7 layer4 0.bn3.weight\n",
      "block index: 7 layer4 0.bn3.bias\n",
      "block index: 7 layer4 0.downsample.0.weight\n",
      "block index: 7 layer4 0.downsample.1.weight\n",
      "block index: 7 layer4 0.downsample.1.bias\n",
      "block index: 7 layer4 1.conv1.weight\n",
      "block index: 7 layer4 1.bn1.weight\n",
      "block index: 7 layer4 1.bn1.bias\n",
      "block index: 7 layer4 1.conv2.weight\n",
      "block index: 7 layer4 1.bn2.weight\n",
      "block index: 7 layer4 1.bn2.bias\n",
      "block index: 7 layer4 1.conv3.weight\n",
      "block index: 7 layer4 1.bn3.weight\n",
      "block index: 7 layer4 1.bn3.bias\n",
      "block index: 7 layer4 2.conv1.weight\n",
      "block index: 7 layer4 2.bn1.weight\n",
      "block index: 7 layer4 2.bn1.bias\n",
      "block index: 7 layer4 2.conv2.weight\n",
      "block index: 7 layer4 2.bn2.weight\n",
      "block index: 7 layer4 2.bn2.bias\n",
      "block index: 7 layer4 2.conv3.weight\n",
      "block index: 7 layer4 2.bn3.weight\n",
      "block index: 7 layer4 2.bn3.bias\n",
      "block index: 9 fc 0.weight\n",
      "block index: 9 fc 0.bias\n",
      "block index: 9 fc 3.weight\n",
      "block index: 9 fc 3.bias\n",
      "block index: 9 fc 5.weight\n",
      "block index: 9 fc 5.bias\n"
     ]
    }
   ],
   "source": [
    "ind = 0\n",
    "\n",
    "for name, child in myModel.named_children():\n",
    "    for name2, params in child.named_parameters():\n",
    "        print('block index:', str(ind), name, name2)\n",
    "    ind = ind +1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Freeze everything except block index 9\n",
    "trainBlock = [9]\n",
    "\n",
    "ind = 0\n",
    "for name, child in myModel.named_children():\n",
    "    if ind not in trainBlock:\n",
    "        for name2, params in child.named_parameters():\n",
    "            params.requires_grad = False\n",
    "    ind = ind +1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "epoch 0:\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "db770c3d2e0a4a01bd8c1fc75a03c65d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, max=7461.0), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4e2aa21682b24f6ea711318f39da6214",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, max=8596.0), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "train_cost: 0.0147, train_acc: 0.9477, val_cost: 0.0171, val_acc: 0.9443\n",
      "saving:  models/CNNmodel/CNNmodel_01_cv0_epoch0_20201013_0636.pth\n"
     ]
    }
   ],
   "source": [
    "main_loop(1, myModel, myPreTrainLoader, myValidLoader)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Train for 3 more epochs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "#myModel = torch.load('models/CNNmodel/CNNmodel_01_epoch0_20201008_0038.pth')\n",
    "#myModel.cuda()\n",
    "#torch.cuda.empty_cache()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Unfreeze everything before further training\n",
    "for name, child in myModel.named_children():\n",
    "    for name2, params in child.named_parameters():\n",
    "        params.requires_grad = True"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "epoch 0:\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0684facba435481c92a612541cd15a6f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, max=34038.0), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "20804e9e24174d84b98bd1a42eb28b74",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, max=8596.0), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "train_cost: 0.0112, train_acc: 0.9582, val_cost: 0.0111, val_acc: 0.9607\n",
      "saving:  models/CNNmodel/CNNmodel_01_cv0_epoch0_20201013_1523.pth\n",
      "epoch 1:\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ea16219b179f4bd1b792c3fd1d1dab3e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, max=34038.0), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c81e122b7e8c4cb8bac15dcffbd55d97",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, max=8596.0), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "train_cost: 0.0091, train_acc: 0.9657, val_cost: 0.0115, val_acc: 0.9639\n",
      "saving:  models/CNNmodel/CNNmodel_01_cv0_epoch1_20201014_0012.pth\n",
      "epoch 2:\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6ba6854ad04749efbd04dac62d825e94",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, max=34038.0), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "69cc6c3591d043bd8cab131cd2f4d24a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, max=8596.0), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "train_cost: 0.0079, train_acc: 0.9693, val_cost: 0.0121, val_acc: 0.9660\n",
      "saving:  models/CNNmodel/CNNmodel_01_cv0_epoch2_20201014_0901.pth\n"
     ]
    }
   ],
   "source": [
    "# Train for 3 more epochs\n",
    "main_loop(3, myModel, myTrainLoader, myValidLoader)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "epoch 0:\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a85600901351494cb7cdfa0ed9c4b0c9",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, max=34038.0), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dcddeaa81a564cf484ba9425848cd2c5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, max=8596.0), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "train_cost: 0.0069, train_acc: 0.9730, val_cost: 0.0140, val_acc: 0.9581\n",
      "saving:  models/CNNmodel/CNNmodel_01_cv0_epoch0_20201014_2159.pth\n"
     ]
    }
   ],
   "source": [
    "# Train for 3 more epochs\n",
    "main_loop(1, myModel, myTrainLoader, myValidLoader)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "pyTorchEnv37",
   "language": "python",
   "name": "pytorchenv37"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}