search

CVMI-Lab / UHDM

(ECCV2022) This is the official PyTorch implementation of ECCV2022 paper: Towards Efficient and Scale-Robust Ultra-High-Definition Image Demoireing
Apache License 2.0
191 stars 25 forks source link

Training time & Nums of GPUs used #22

Closed FireWallDragonDarkFluid closed 11 months ago

FireWallDragonDarkFluid commented 11 months ago

Hi! Wondering the traing time for UHDM and TIP 2018 dataset, since in the paper it was mentioned that only a single 3090 GPU was used for training?

Or, I misunderstood the paper's statement. How many 3090s were used for training?

Thanks a lot.

XinYu-Andy commented 11 months ago

Hi! Wondering the traing time for UHDM and TIP 2018 dataset, since in the paper it was mentioned that only a single 3090 GPU was used for training?

Or, I misunderstood the paper's statement. How many 3090s were used for training?

Thanks a lot.

We only use one GPU for both training and inference. You can find the detailed training configurations in our paper.