Skip to content

q(image), k,v(condition) channel dismatch #24

@internationalwe

Description

@internationalwe

q(image), k,v(condition) channel is not matching in cross attention
how to make the channels of q and k the same?

  1. put k,v in to zero matrix(Same as the shape of the image)

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