The fine tune can't work

[toyer]

Thanks for sharing the tools. It helps me a lot. But I have some little question. When I load the model parameters trained before to continue my traininig, the accuracy can not improve after many epoches. I thought this is because the batch normalization parameters is sqrted when saved.

[Ligeng Zhu]

Sorry for the late rely. Do you still have the issue? If so, please attach a minimal case.

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[stale]

Close due to inactivity. Feel free to re-open it if necessary.