I have unformatted text documents that contain semi standard sentences/paragraphs that I would like to pull out so that I can format the document with the correct information in the correct place. I am thinking I would need to do some sort of attention layer to return the start and end tokens but can’t find any examples.
Think: I have an empty document template with multiple sections that need to be filled in, and an unformatted document with all of the information in text form. How would you split up the text file so that all of the information can be placed in to the template.