Huge Time Cost When Running the Code

YudeWang / SEAM

Self-supervised Equivariant Attention Mechanism for Weakly Supervised Semantic Segmentation, CVPR 2020 (Oral)

MIT License

539 stars 97 forks source link

Huge Time Cost When Running the Code #35

Closed hyalvin closed 2 years ago

hyalvin commented 2 years ago

Hi, thanks to your work! But I have got a problem when running this code with an 8*GPU(A100) server, it just stuck on this two line model = torch.nn.DataParallel(model).cuda() for iter, pack in enumerate(train_data_loader): And also it cost a lot of time to run evert option in the traning process like F.interpolate I wonder if there is something wrong with my conda env?

hyalvin commented 2 years ago

OK, I have solved this problem with the help of my mentor. It must be that when creating a new conda env, the version of CUDA in this env is not matched with that in the server home, and thus when moving the model to GPU, something wrong happened, what a terrible problem!