Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

pretrained model #4

Open
FuLy2002 opened this issue Oct 30, 2024 · 2 comments
Open

pretrained model #4

FuLy2002 opened this issue Oct 30, 2024 · 2 comments

Comments

@FuLy2002
Copy link

作者您好,感谢您的代码开源,我在使用您公布的预训练权重(LaDiC.bin)直接在测试集上测试时,在30步的情况下,发现各种评估指标与论文中的结果有微小的差距,但在CIDEr分数上差距比较明显,请问这是我的config.py(当前与仓库中的config.py保持不变)有问题还是有其他问题呢?

@wangyuchi369
Copy link
Owner

您好!如果只有微小差距的话说明整体Inference过程应该没有问题的,CIDEr明显一点也可能是本身range比较大。
有以下建议:

  1. 可以调整一些sampling时的超参数,例如classifier free guidance的weight等
  2. 我们采样时是用了Minimum Bayes Risk Decoding的,因为是后处理代码里好像没有放,您可以参考https://arxiv.org/pdf/2205.14217 的Section 5.2

@FuLy2002
Copy link
Author

好的,感谢您的解答和工作分享。

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants