python-kerasHow can I use Python and Keras to create an image dataset from a directory?

Creating an image dataset from a directory with Python and Keras is a relatively straightforward process. The following example code block can be used to achieve this:

from keras.preprocessing.image import ImageDataGenerator

# Set the directory where the images are stored
data_dir = 'data/'

# Create the data generator
datagen = ImageDataGenerator(rescale=1./255)

# Load the images from the directory
data_generator = datagen.flow_from_directory(
    data_dir,
    target_size=(150, 150),
    batch_size=32,
    class_mode='categorical')

The code above will load the images from the directory specified by data_dir. The images will be resized to 150x150 pixels. The images will also be rescaled to a range of 0-1. The data_generator variable will contain the dataset.

Code explanation

• from keras.preprocessing.image import ImageDataGenerator - imports the ImageDataGenerator class from the keras.preprocessing.image module.
• data_dir = 'data/' - sets the directory where the images are stored.
• datagen = ImageDataGenerator(rescale=1./255) - creates the data generator, with the images rescaled to a range of 0-1.
• data_generator = datagen.flow_from_directory(data_dir, target_size=(150, 150), batch_size=32, class_mode='categorical') - loads the images from the directory, resizing them to 150x150 pixels, and returns the dataset in the data_generator variable.

Helpful links

• Keras Image Preprocessing
• Keras ImageDataGenerator Documentation

Edit this code on GitHub