Skip to content

训练时验证过程中模型性能与测试阶段不一致 #754

@kesssssa

Description

@kesssssa

我使用Basicsr超分框架训练模型,使用Set5数据集作为验证集,训练过程中模型在Set5数据集上的性能较好,但训练结束后我将性能较好的模型用于推理,使用basicsr test.py推理Set5数据集,发现性能达不到验证时的结果,例如:某一轮模型在Set5数据集上验证时性能为PSNR32,而我将这轮模型用于test.py推理Set5数据集,PSNR只能达到31.7左右,这是为什么

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