torchvision

It would be pretty helpful to see what layers are frozen and what are not. There is already an indication of how many params are trainable and what is not, but also viewing per layer setting would be useful. Thanks!

Feature Request

Description of Problem:

We've taken a fairly risky/bad approach by just resorting to setting function params to arbitrary defaults when they aren't passed in. While this works sometimes, I think we've overused this approach and should really trim parts of it down.

Potential Solutions:

Eliminate optional parameters in any non-public function, and actually throw ex