Skip to content

已解决: RuntimeError: Input type (torch.cuda.FloatTensor) and weight type (torch.cuda.HalfTensor) should be the same #21

@dorannn

Description

@dorannn

在position coding时,输入固定float,在yolo训练val时会自动进行半精度训练,将position()输出类型转换成input就好
position(......).to(x.dtype)

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