Skip to content

A Question About Training Loss #4

Description

@johannwyh

Hi, Lingzhi !

I have a question about your model training loss. As is stated in your paper, you sum up the batch loss to get a loss number.

My question is that, did you divide the loss by the batch size? If not, does it mean that my learning rate should be inversely proportional to the batch size?

Thanks a lot! By the way, FaceShifter is awesome and hope we can surpass it!

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions