[FIXED] training model with image data generator , giving warning

Issue

I preprocess my data with image data generator, and when apply model.compile and model.fit it shows me this message

"/usr/local/lib/python3.10/dist-packages/keras/src/preprocessing/image.py:1862: UserWarning: This ImageDataGenerator specifies featurewise_center, but it hasn't been fit on any training data. Fit it first by calling .fit(numpy_data). warnings.warn("

and then the model begin training what is it meaning the warning? and when try to add featurewise_center in image_data_gen, it gives me error

train_datagen = ImageDataGenerator(
    main_dir,
    rescale=1. / 255,
    shear_range=0.2,
    zoom_range=0.2,
    width_shift_range=0.1,
    height_shift_range=0.1,
    brightness_range=[0.2,1.0],
    horizontal_flip=True,
    validation_split=0.2)
test_datagen = ImageDataGenerator(rescale=1. / 255)
val_datagen = ImageDataGenerator(rescale=1. / 255)

and apply image _data_flow

train_generator = train_datagen.flow_from_directory(directory = main_dir,subset="training",target_size=(img_w,img_h),
                                                    batch_size = batch_size,class_mode="categorical")
validation_generator = train_datagen.flow_from_directory(
    directory = main_dir,
    target_size=(img_w,img_h),
    batch_size=batch_size,
    class_mode="categorical",subset="validation")

Solution

You have to remove main_dir in ImageDataGenerator. The first positional argument of ImageDataGenerator is featurewise_center:

train_datagen = ImageDataGenerator(
    rescale=1. / 255,
    shear_range=0.2,
    zoom_range=0.2,
    width_shift_range=0.1,
    height_shift_range=0.1,
    brightness_range=[0.2,1.0],
    horizontal_flip=True,
    validation_split=0.2)
test_datagen = ImageDataGenerator(rescale=1. / 255)
val_datagen = ImageDataGenerator(rescale=1. / 255)

Answered By - Corralien