The lecturer is saying
you can think of machine learning, deep learning in general as modeling a function that takes an input that can be an image, a speech, a natural language or a CSV file, give it to a box and get an output that can be classification
in this context, there are 3 steps in which the function do its job
1. takes an input 2. gives it to a box 3. gets an output
so, the "give" and "get" are not grammatical, am I right about this?