{"cells":[{"cell_type":"markdown","metadata":{"_cell_guid":"7b2fd960-5e74-33e0-bca6-c06c83bbb53a"},"source":"ddd"},{"cell_type":"code","execution_count":null,"metadata":{"_cell_guid":"ab1ffa9f-c974-288a-33f3-398db92bc9e3"},"outputs":[],"source":"import subprocess\n# Get all the data path in train\ntrain_data_dir = \"../input/train/\"\ncomplished_process_for_folders = subprocess.run(['ls', train_data_dir],\\\n                                          stdout = subprocess.PIPE, encoding = \"utf8\")\nprint(complished_process_for_folders.stdout)"},{"cell_type":"code","execution_count":null,"metadata":{"_cell_guid":"2e4c3f8d-a836-39c6-1e55-0f94e2191ba7"},"outputs":[],"source":"import pandas as pd\n# Get folders for 3 types\ntype_folders = complished_process_for_folders.stdout.strip().split('\\n')\n\n# Get image names \nimage_names = pd.DataFrame()\nfor folder in type_folders:\n    complised_process_for_images = subprocess.run(['ls', train_data_dir + folder +'/'],\\\n                                                 stdout = subprocess.PIPE, encoding = \"utf8\")\n    image_names_in_folder = complised_process_for_images.stdout.strip().split('\\n')\n    df = pd.DataFrame({folder: image_names_in_folder})\n    image_names = pd.concat([image_names, df], axis = 1)\n    \nimage_names"},{"cell_type":"code","execution_count":null,"metadata":{"_cell_guid":"dcc14036-e1e8-7cb0-1d69-c9ed11728fba"},"outputs":[],"source":"# Number of image files of each type\nimage_names.count()"},{"cell_type":"code","execution_count":null,"metadata":{"_cell_guid":"e8e14be4-2f2d-267c-c08a-670774db23d6"},"outputs":[],"source":""},{"cell_type":"code","execution_count":null,"metadata":{"_cell_guid":"bea32038-25fd-f428-19c6-d707a05ac0ed"},"outputs":[],"source":"import cv2\n# check the shape of each image\nimage_shapes = pd.DataFrame()\nfor image_type in image_names.columns:\n    image_shape_temp = []\n    for i in range(image_names[image_type].count()):\n        image_dir = train_data_dir + image_type + '/' + image_names[image_type][i]\n        img_shape = (cv2.imread(image_dir,cv2.IMREAD_UNCHANGED)).shape\n        image_shape_temp.append(img_shape)\n    image_shapes = pd.concat([image_shapes, pd.DataFrame({image_type: image_shape_temp})], axis = 1)\nimage_shapes"},{"cell_type":"code","execution_count":null,"metadata":{"_cell_guid":"d3345109-4111-0faf-0c80-28961db059c3"},"outputs":[],"source":"for image_type in image_shapes.columns:\n    image_shapes[image_type].value_counts()"},{"cell_type":"code","execution_count":null,"metadata":{"_cell_guid":"569a5532-c1eb-db21-a94f-5e671f1c9068"},"outputs":[],"source":""}],"metadata":{"_change_revision":0,"_is_fork":false,"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":"3.6.0"}},"nbformat":4,"nbformat_minor":0}