{"cells":[{"metadata":{},"cell_type":"markdown","source":"Dataset available here: https://www.kaggle.com/xhlulu/panda-resized-train-data-512x512","execution_count":null},{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"cell_type":"code","source":"import os\n\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\nimport openslide\nimport matplotlib.pyplot as plt\nfrom PIL import Image\nimport cv2\nfrom tqdm.notebook import tqdm\nimport skimage.io\nfrom skimage.transform import resize, rescale","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Load dataframe","execution_count":null},{"metadata":{"_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","trusted":true},"cell_type":"code","source":"train_labels = pd.read_csv('/kaggle/input/prostate-cancer-grade-assessment/train.csv')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_labels.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"data_dir = '/kaggle/input/prostate-cancer-grade-assessment/train_images/'","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"mask_dir = '/kaggle/input/prostate-cancer-grade-assessment/train_label_masks/'\nmask_files = os.listdir(mask_dir)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Speed tests","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"img_id = train_labels.image_id[0]\npath = data_dir + img_id + '.tiff'","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"%time biopsy = openslide.OpenSlide(path)\n%time biopsy2 = skimage.io.MultiImage(path)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"%timeit img = biopsy.get_thumbnail(size=(512, 512))\n%timeit out = resize(biopsy2[-1], (512, 512))\n%timeit out = cv2.resize(biopsy2[-1], (512, 512))\n%timeit out = Image.fromarray(biopsy2[-1]).resize((512, 512))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"out = cv2.resize(biopsy2[-1], (512, 512))\n\n%timeit Image.fromarray(out).save(img_id+'.png')\n%timeit cv2.imwrite(img_id+'.png', out)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"Conclusion: skimage is fastest for loading, cv2 is fastest for resizing and saving.","execution_count":null},{"metadata":{},"cell_type":"markdown","source":"### Try loading masks","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"mask = skimage.io.MultiImage(mask_dir + mask_files[1])\nimg = skimage.io.MultiImage(data_dir + mask_files[1].replace(\"_mask\", \"\"))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"mask[-1].shape, img[-1].shape","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Start here","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"save_dir_isup_grade_0 = \"/kaggle/isup_grade_0\"\nsave_dir_isup_grade_1 = \"/kaggle/isup_grade_1\"\nsave_dir_isup_grade_2 = \"/kaggle/isup_grade_2\"\nsave_dir_isup_grade_3 = \"/kaggle/isup_grade_3\"\nsave_dir_isup_grade_4 = \"/kaggle/isup_grade_4\"\nsave_dir_isup_grade_5 = \"/kaggle/isup_grade_5\"\n\nos.makedirs(save_dir_isup_grade_0, exist_ok=True)\nos.makedirs(save_dir_isup_grade_1, exist_ok=True)\nos.makedirs(save_dir_isup_grade_2, exist_ok=True)\nos.makedirs(save_dir_isup_grade_3, exist_ok=True)\nos.makedirs(save_dir_isup_grade_4, exist_ok=True)\nos.makedirs(save_dir_isup_grade_5, exist_ok=True)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"for img_id, isup_grade in tqdm(zip(train_labels.image_id, train_labels.isup_grade)):\n    load_path = data_dir + img_id + '.tiff'\n    if(isup_grade == 0): save_path = save_dir_isup_grade_0 + img_id + '.png'\n    if(isup_grade == 1): save_path = save_dir_isup_grade_1 + img_id + '.png'\n    if(isup_grade == 2): save_path = save_dir_isup_grade_2 + img_id + '.png'\n    if(isup_grade == 3): save_path = save_dir_isup_grade_3 + img_id + '.png'\n    if(isup_grade == 4): save_path = save_dir_isup_grade_4 + img_id + '.png'\n    if(isup_grade == 5): save_path = save_dir_isup_grade_5 + img_id + '.png'\n        \n    biopsy = skimage.io.MultiImage(load_path)\n    img = cv2.resize(biopsy[-1], (512, 512))\n    cv2.imwrite(save_path, img)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"save_dir_mask_isup_grade_0 = \"/kaggle/isup_grade_mask_0\"\nsave_dir_mask_isup_grade_1 = \"/kaggle/isup_grade_mask_1\"\nsave_dir_mask_isup_grade_2 = \"/kaggle/isup_grade_mask_2\"\nsave_dir_mask_isup_grade_3 = \"/kaggle/isup_grade_mask_3\"\nsave_dir_mask_isup_grade_4 = \"/kaggle/isup_grade_mask_4\"\nsave_dir_mask_isup_grade_5 = \"/kaggle/isup_grade_mask_5\"\n\nos.makedirs(save_dir_mask_isup_grade_0, exist_ok=True)\nos.makedirs(save_dir_mask_isup_grade_1, exist_ok=True)\nos.makedirs(save_dir_mask_isup_grade_2, exist_ok=True)\nos.makedirs(save_dir_mask_isup_grade_3, exist_ok=True)\nos.makedirs(save_dir_mask_isup_grade_4, exist_ok=True)\nos.makedirs(save_dir_mask_isup_grade_5, exist_ok=True)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"'''\nfor mask_file in tqdm(mask_files):\n    load_path = mask_dir + mask_file\n    save_path = save_mask_dir + mask_file.replace('.tiff', '.png')\n    \n    mask = skimage.io.MultiImage(load_path)\n    img = cv2.resize(mask[-1], (512, 512))\n    cv2.imwrite(save_path, img)\n'''","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"!tar -czf isup_grade_0.tar.gz ../isup_grade_0*.png\n!tar -czf isup_grade_1.tar.gz ../isup_grade_1*.png\n!tar -czf isup_grade_2.tar.gz ../isup_grade_2*.png\n!tar -czf isup_grade_3.tar.gz ../isup_grade_3*.png\n!tar -czf isup_grade_4.tar.gz ../isup_grade_4*.png\n!tar -czf isup_grade_5.tar.gz ../isup_grade_5*.png","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"!tar -czf train_label_isup_grade_0_masks.tar.gz ../isup_grade_mask_0/*.png\n!tar -czf train_label_isup_grade_1_masks.tar.gz ../isup_grade_mask_1/*.png\n!tar -czf train_label_isup_grade_2_masks.tar.gz ../isup_grade_mask_2/*.png\n!tar -czf train_label_isup_grade_3_masks.tar.gz ../isup_grade_mask_3/*.png\n!tar -czf train_label_isup_grade_4_masks.tar.gz ../isup_grade_mask_4/*.png\n!tar -czf train_label_isup_grade_5_masks.tar.gz ../isup_grade_mask_5/*.png","execution_count":null,"outputs":[]}],"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat":4,"nbformat_minor":4}