datasets<\/em>\u201d subdirectory. The datasets are stored in a compressed format, but may also include additional metadata.<\/p>\nAfter the first call to a dataset-specific load function and the dataset is downloaded, the dataset does not need to be downloaded again. Subsequent calls will load the dataset immediately from disk.<\/p>\n
The load functions return two tuples, the first containing the input and output elements for samples in the training dataset, and the second containing the input and output elements for samples in the test dataset. The splits between train and test datasets often follow a standard split, used when benchmarking algorithms on the dataset.<\/p>\n
The standard idiom for loading the datasets is as follows:<\/p>\n
...\r\n# load dataset\r\n(trainX, trainy), (testX, testy) = load_data()<\/pre>\nEach of the train and test X<\/em> and y<\/em> elements are NumPy arrays of pixel or class values respectively.<\/p>\nTwo of the datasets contain grayscale images and two contain color images. The shape of the grayscale images must be converted from two-dimensional to three-dimensional arrays to match the preferred channel ordering of Keras. For example:<\/p>\n
# reshape grayscale images to have a single channel\r\nwidth, height, channels = trainX.shape[1], trainX.shape[2], 1\r\ntrainX = trainX.reshape((trainX.shape[0], width, height, channels))\r\ntestX = testX.reshape((testX.shape[0], width, height, channels))<\/pre>\nBoth grayscale and color image pixel data are stored as unsigned integer values with values between 0 and 255.<\/p>\n
Before modeling, the image data will need to be rescaled, e.g. such as normalization to the range 0-1 and perhaps further standardized. For example:<\/p>\n
# normalize pixel values\r\ntrainX = trainX.astype('float32') \/ 255\r\ntestX = testX.astype('float32') \/ 255<\/pre>\nThe output elements of each sample (y<\/em>) are stored as class integer values. Each problem is a multi-class classification problem (more than two classes); as such, it is common practice to one hot encode the class values prior to modeling. This can be achieved using the to_categorical()<\/em> function provided by Keras; for example:<\/p>\n...\r\n# one hot encode target values\r\ntrainy = to_categorical(trainy)\r\ntesty = to_categorical(testy)<\/pre>\nNow that we are familiar with the idioms for working with the standard computer vision datasets provided by Keras, let\u2019s take a closer look at each dataset in turn.<\/p>\n
Note, the examples in this tutorial assume that you have internet access and may download the datasets the first time each example is run on your system. The download speed will depend on the speed of your internet connection and you are recommended to run the examples from the command line.<\/p>\n
<\/div>\n
<\/p>\nWant Results with Deep Learning for Computer Vision?<\/h3>\n
Take my free 7-day email crash course now (with sample code).<\/p>\n
Click to sign-up and also get a free PDF Ebook version of the course.<\/p>\n
Download Your FREE Mini-Course<\/a>
![\"How](\"https:\/\/machinelearningmastery.com\/wp-content\/uploads\/2019\/04\/How-to-Load-and-Visualize-Standard-Computer-Vision-Datasets-With-Keras.jpg\")
<\/p>\n