In technical ML articles, should one report input layer shape as part of their model architecture?
I’m a student in natural sciences, working on a project that uses machine learning. I’m working with a neural network model created by a different research group, implemented using haiku.