arthurdouillard / CVPR2021_PLOP

Official code of CVPR 2021's PLOP: Learning without Forgetting for Continual Semantic Segmentation

Home Page:https://arxiv.org/abs/2011.11390

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

About pretrained model

congwei45 opened this issue · comments

As you write in readme "Since the pretrained are made on multiple-gpus, they contain a prefix "module." in each key of the network. Please, be sure to remove them to be compatible with this code (simply rename them using key = key[7:]) (if you're working on single gpu). "

If I use 2 gpus, i.e., --nproc_per_node=2, do I still need to modify the pretrained network?

I don't think you need to do that in that case.

Thanks for your answer!