keras.preprocessing.text.Tokenizer methods for tensorflow.data.TextLineDataset

Question

I have a very large text corpus which I am loading with:

text_ds = tf.data.TextLineDataset(list_files)

and cannot use the methods keras