It is a text classification task. I have converted the text to tokens and then padded them so they are of same length using tensorflow Tokenizer and pad_sequences. y_train