How can we perform different text preprocessing methods other than Text Vectorizer ?

[ashikshafi08]

You can do the kind of processing you're after using the TextVectorization layer if you like: https://www.tensorflow.org/api_docs/python/tf/keras/layers/experimental/preprocessing/TextVectorization

See the standardize parameter where you can pass in your own function.

This function you can define to preprocess text in a certain way (e.g. leave the "@" symbols in if you need). You could try different experiments and see what suits you best.