The Sequence Scope: Using Transformers in Mainstream Deep Learning Applications