Skip to content

一些关于训练格式的疑惑 #30

@huangfu170

Description

@huangfu170

论文当中提到"Notably, both q and c can be images, text, or composed image-text data.",但按照文中提出的数据集,训练的时候不是固定是一张图片+一段文本搜索一张图吗?希望得到您的解惑

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