Skip to content

Question about training setup #3

Open
@congzp

Description

@congzp

Hi, when training downstream tasks, are the weights of the Vision encoder frozen, or does it participate in full fine-tuning training?

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions