The following problems occur in the preprocessing of Chinese-English translation

A post was merged into an existing topic: I want to ask whether the training set verification set data needs word segmentation