BIT-DA/RIPU

训练程序不支持断点训练

licongguan opened this issue · 3 comments

if cfg.resume: 训练时加上resume 以及模型,程序还是从头开始训练

Hi @licongguan

Sorry about it. Here, cfg.resume is only used to load checkpoint (specifically load model parameters). If you want to resume the whole process, iteration and optimizers should be loaded and modify some lines accordingly.

I believe the issue at hand was addressed, as such I'm closing this. Feel free to ask if you have further questions.

Hi, I get it, but it would be better if you could keep optimizing your code so it can apply reverting the whole process, iteration and optimizer.