how do i split the data,uploaded by directory_iterator

Question

how do I split the dataset uploaded by

,
    fill_mode='nearest',from tensorflow.keras.preprocessing.image import (
    DirectoryIterator, ImageDataGenerator
)

# Your desired classes
desired_classes = [
    'Tomato_Bacterial_spot',
    'Tomato_Early_blight',
    'Tomato_Late_blight',
    'Tomato_Leaf_Mold',
    'Tomato_Septoria_leaf_spot',
    'Tomato_Spider_mites_Two_spotted_spider_mite',
    'Tomato__Target_Spot',
    'Tomato__Tomato_YellowLeaf__Curl_Virus',
    'Tomato__Tomato_mosaic_virus',
    'Tomato_healthy'
]
np.random.seed(seed)
dataset = DirectoryIterator(
    directory='/kaggle/input/plant-village/PlantVillage',
    image_data_generator=img_iterator,
    classes=desired_classes,
    target_size=(256,256),
    batch_size=8
)

I have done the visualizations but failed to split it into training,validation and test datasets. I'm trying this code but it is not working

how can i overcome this problem for spliting directory_iterator dataset to splits?