JingyunLiang / SwinIR

SwinIR: Image Restoration Using Swin Transformer (official repository)

Home Page:https://arxiv.org/abs/2108.10257

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Training efficiency

Linfeng-Tang opened this issue · comments


Thanks for the great work again.

When training SWIR with the KAIR toolbox, I found that the CPU utilization was particularly high, while the GPU was always idle. And the training is particularly inefficient. I wonder if the author would be so kind as to tell me the GPU and CPU configurations used, and the training time?

Do you use DstributedDataParallel (DDP)?

Thank you very much for your reply. Actually, I did not use DstributedDataParallel (DDP). And, I ran: "python main_train_psnr.py --opt options/swinir/train_swinir_sr_lightweight.json".

This is the reason. DDP is much quicker than DP.

ezoic increase your site revenue