https://kylewbanks.com/blog/loading-unlabeled-images-with-imagedatagenerator-flowfromdirectory-keras
from tensorflow.keras.preprocessing.image import ImageDataGenerator, load_img, img_to_array
datagen = ImageDataGenerator()
# In ./dataset directory, there is `test` directory.
test_data = datagen.flow_from_directory('./dataset', classes=['test'])
Found 1000 images belonging to 1 class.
The classes argument tells flow_from_directory that you only want to load images of the specified class(es), in this case the test “class”. Now test_data will contain only the one test “class”, allowing you to work with your test dataset just as you would the labeled training data from above.