{"cells":[{"metadata":{"_uuid":"17ed98b52e46b8461539e6d93ef54798d0f4fa54"},"cell_type":"markdown","source":"Import libs"},{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true,"collapsed":true},"cell_type":"code","source":"# import libs\nimport glob, pylab, pandas as pd\nimport pydicom, numpy as np\nimport random\nimport json\nimport time\nimport copy\nimport pydicom\nimport torchvision\nimport sys\nimport torch\nimport torch.nn as nn\nimport torch.optim as optim\nfrom torch.optim import lr_scheduler\nfrom torch.autograd import Variable\nfrom PIL import Image, ImageDraw, ImageFont\nimport matplotlib.pyplot as plt\nfrom matplotlib import patches, patheffects\n\nfrom sklearn.model_selection import train_test_split\nfrom torchvision import datasets, transforms\nfrom torchvision import transforms\nfrom torch.utils.data import Dataset, DataLoader\nfrom torch.optim import lr_scheduler\nfrom pathlib import Path\n\n# from fastai.conv_learner import *\n# from fastai.dataset import *\n# from fastai.dataset import ImageClassifierData","execution_count":null,"outputs":[]},{"metadata":{"_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","trusted":true},"cell_type":"code","source":"# see input dir\n!ls ../input","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"3d7f0ddd940806b8ca11a50fa9241c3d1421e474"},"cell_type":"code","source":"PATH = Path('../input')\n# read training lables\ntrain_bb_df = pd.read_csv(PATH/'stage_1_train_labels.csv')\ntrain_bb_df.head()","execution_count":null,"outputs":[]},{"metadata":{"_uuid":"ec7735bf8d9f200d9489ae0c0ab60b0f510f412b"},"cell_type":"markdown","source":"row 4 shows, there is bouding box in image with given x & y. First 4 image don't have any bounding box\n### check if duplicate bounding box are present for any patient"},{"metadata":{"trusted":true,"_uuid":"3b98a272954161276884fd6a8c2605de9fa8991f"},"cell_type":"code","source":"train_bb_df['duplicate'] = train_bb_df.duplicated(['patientId'], keep=False)\n# see data\ntrain_bb_df[train_bb_df['duplicate']].head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"56ebd9ec72029a47bc1ff98eb6de4b756d03615e","scrolled":true},"cell_type":"code","source":"detailed_df = pd.read_csv(PATH/'stage_1_detailed_class_info.csv')\n# merge two df\nclass_df = train_bb_df.merge(detailed_df, on=\"patientId\")\n# class_df.head()\ncsv_df = class_df.filter(['patientId', 'Target'], )\ncsv_df = csv_df.set_index('patientId', )\n# detailed_df.head() , \nclass_df.head(10)\n# csv_df.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"21118a6d9577df6386ae3526aa4fd23094be7e75","collapsed":true},"cell_type":"code","source":"# these are dcm image files. Lets load them\n# ! ls {PATH}'/stage_1_test_images' | head","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"91eaaefbc070e9e912c84c917b3f5ffa852415fa"},"cell_type":"markdown","source":"Loading all train images in memory causes memory overflow. Kernel will restart. \nWe creating dataset to read images, then apply transformation."},{"metadata":{"trusted":true,"_uuid":"04d6963b59aec391ec051af33d8ba5d19d1b99cf","collapsed":true},"cell_type":"code","source":"class CDataset(Dataset):\n    def __init__(self, ds, img_dir, class_df, transform=None, ext=None): \n        self.ds = ds\n        self.img_dir = img_dir\n        self.class_df = class_df\n        self.ext = ext or '.dcm'\n        self.transform = transforms.Compose(transform) if transform else None\n        \n    def __len__(self): \n        return len(self.ds)\n    \n    def read_dicom_image(self, loc):\n        # return numpy array\n        img_arr = pydicom.read_file(loc.as_posix()).pixel_array\n        img_arr = img_arr/img_arr.max()\n        img_arr = (255*img_arr).clip(0, 255).astype(np.uint8)\n        img_arr = Image.fromarray(img_arr).convert('RGB') # model expects 3 channel image\n        return img_arr\n        \n    def __getitem__(self, i):\n        img = self.read_dicom_image(self.ds[i])\n        if self.transform:\n            img = self.transform(img)\n        patientId = self.ds[i].name.split('.')[0]\n        kls = self.class_df[self.class_df['patientId'] == patientId]\n        return img, kls.iloc[0].Target","execution_count":null,"outputs":[]},{"metadata":{"_uuid":"795d7230a56322376d6ee877c25bff2b6afac414"},"cell_type":"markdown","source":"We are randomly sampling 400 images. This for testing our code."},{"metadata":{"trusted":true,"scrolled":true,"_uuid":"57df580cddb34df670cf0a970ceedebedfa34cd7","collapsed":true},"cell_type":"code","source":"# img_dir = PATH/'stage_1_train_images'\nimg_dir = PATH/'stage_1_train_images'\nsample = random.sample(list(img_dir.iterdir()), 400) # sample\n# sample = list(img_dir.iterdir())\ntrain, test = train_test_split(sample)\n\ntransform = [transforms.Resize(224), transforms.RandomHorizontalFlip() , transforms.ToTensor()]\ntrain_ds = CDataset(train, img_dir, class_df, transform=transform)\ntest_ds = CDataset(test, img_dir, class_df, transform=transform)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"82e6bdbdca617b04c89d1f19e977f31e1ea9d2ca","collapsed":true},"cell_type":"code","source":"batch_size=32\nsz=224\ntrain_dl = DataLoader(train_ds, batch_size=batch_size,)\ntest_dl = DataLoader(test_ds, batch_size=batch_size)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"e19fc577289cbe43bddcaabf1ed4b080df2c5278","collapsed":true},"cell_type":"code","source":"def show_img(im, figsize=None, ax=None):\n    if not ax: \n        fig,ax = plt.subplots(figsize=figsize)\n    ax.imshow(im)\n    ax.get_xaxis().set_visible(False)\n    ax.get_yaxis().set_visible(False)\n    return ax\n\ndef draw_outline(o, lw):\n  o.set_path_effects([patheffects.Stroke(\n      linewidth=lw, foreground='black'), patheffects.Normal()])\n\ndef draw_rect(ax, b):\n    patch = ax.add_patch(patches.Rectangle(b[:2], *b[-2:], fill=False, edgecolor='white', lw=2))\n    draw_outline(patch, 4)\n\ndef draw_text(ax, xy, txt, sz=14):\n    text = ax.text(*xy, txt, verticalalignment='top', color='white', fontsize=sz, weight='bold')\n    draw_outline(text, 1)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"24e8a3a3507a0dd0f8796337a1a8d09b4a1d3657"},"cell_type":"code","source":"image, klass = next(iter(train_dl))\nfig, axes = plt.subplots(1, 4, figsize=(12, 2))\nfor i,ax in enumerate(axes.flat):\n    image, klass\n#     ima=image[i].numpy().transpose((1, 2, 0))\n    ima=image[i][0]\n    b = klass[i]\n    ax = show_img(ima, ax=ax)\n    draw_text(ax, (0,0), b)\n\nplt.tight_layout()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"f85109b99572e4bdec0594c83d7a3b5aa3a2e271","collapsed":true},"cell_type":"code","source":"use_gpu = torch.cuda.is_available()\ndataloaders = {'train': train_dl, 'val':test_dl}\n               \ndef train_model(model, criterion, optimizer, scheduler, num_epochs=25):\n    since = time.time()\n\n    best_model_wts = copy.deepcopy(model.state_dict())\n    best_acc = 0.0\n\n    for epoch in range(num_epochs):\n        print('Epoch {}/{}'.format(epoch, num_epochs - 1))\n        print('-' * 10)\n\n        # Each epoch has a training and validation phase\n        for phase in ['train', 'val']:\n            if phase == 'train':\n                scheduler.step()\n                model.train(True)  # Set model to training mode\n            else:\n                model.train(False)  # Set model to evaluate mode\n\n            running_loss = 0.0\n            running_corrects = 0\n\n            # Iterate over data.\n            data_loader = dataloaders[phase]\n            for data in data_loader:\n                # get the inputs\n                inputs, labels = data\n\n                # wrap them in Variable\n                if use_gpu:\n                    inputs = Variable(inputs.cuda(),requires_grad=True)\n                    labels = Variable(labels.cuda())\n                else:\n                    inputs, labels = Variable(inputs), Variable(labels)\n\n                # zero the parameter gradients\n                optimizer.zero_grad()\n\n                # forward\n                outputs = model(inputs)\n                _, preds = torch.max(outputs.data, 1)\n                loss = criterion(outputs, labels)\n\n                # backward + optimize only if in training phase\n                if phase == 'train':\n                    loss.backward()\n                    optimizer.step()\n\n                # statistics\n                running_loss += loss.data[0] * inputs.size(0)\n                running_corrects += torch.sum(preds == labels.data)\n\n            epoch_loss = running_loss / len(data_loader.dataset)\n            epoch_acc = running_corrects / len(data_loader.dataset)\n            \n            print('{} Loss: {:.4f} Acc: {:.4f}'.format(\n                phase, epoch_loss, epoch_acc))\n\n            # deep copy the model\n            if phase == 'val' and epoch_acc > best_acc:\n                best_acc = epoch_acc\n                best_model_wts = copy.deepcopy(model.state_dict())\n\n        print()\n\n    time_elapsed = time.time() - since\n    print('Training complete in {:.0f}m {:.0f}s'.format(\n        time_elapsed // 60, time_elapsed % 60))\n    print('Best val Acc: {:4f}'.format(best_acc))\n\n    # load best model weights\n    model.load_state_dict(best_model_wts)\n    return model","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"9bcf0020aea5a3622cc61f441444182ed1214595","scrolled":true,"collapsed":true},"cell_type":"code","source":"device = torch.cuda.set_device(0)\n\nmodel_ft = torchvision.models.resnet18(pretrained=True)\n\nnum_ftrs = model_ft.fc.in_features\nmodel_ft.fc = nn.Linear(num_ftrs, 2)\ncriterion = nn.CrossEntropyLoss()\nmodel_ft = model_ft.cuda()\n\n# Observe that all parameters are being optimized\noptimizer_ft = optim.Adam(model_ft.parameters())\n\n# Decay LR by a factor of 0.1 every 7 epochs\nexp_lr_scheduler = lr_scheduler.StepLR(optimizer_ft, step_size=7, gamma=0.1)\nsince = time.time()","execution_count":null,"outputs":[]},{"metadata":{"_uuid":"f875cd84fab289f39640ad5ec088d90f65975b1d"},"cell_type":"markdown","source":"Lets train our model and check classification accuracy"},{"metadata":{"trusted":true,"_uuid":"7f2c0c0d9787c4ea7ce227f8df17eeb9699beb12"},"cell_type":"code","source":"model_ft = train_model(model_ft, criterion, optimizer_ft, exp_lr_scheduler, num_epochs=3)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"fbce099f733786ffeacd0e70916e610edbe0a1d1"},"cell_type":"code","source":"! nvidia-smi ","execution_count":null,"outputs":[]},{"metadata":{"_uuid":"5f3938fbf35d269ff9b109a2f553998674fd857a"},"cell_type":"markdown","source":"1. With resnet 18, transformation.RandomResizedCrop(224) & random sample of 400 images we reached to 85%\n1. With resnet18, only horizontal flip transformation & random sample of 400 images we reached to 84%\n1. <img src=\"attachment:image.png\" width=\"400\">\n1. With resnet18, Adam optimizer, val loss reduces to 85%","attachments":{"image.png":{"image/png":"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"}}},{"metadata":{"trusted":true,"collapsed":true,"_uuid":"01d3eee040e5121c3a839e46d56a8cc74b4be3cb"},"cell_type":"code","source":"","execution_count":null,"outputs":[]}],"metadata":{"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"name":"python","version":"3.6.4","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"}},"nbformat":4,"nbformat_minor":1}