search

kunzhan / DSSN

ACM MM 2023: Improving semi-supervised semantic segmentation with dual-level Siamese structure network
http://arxiv.org/abs/2307.13938
17 stars 1 forks source link

Requirement of GPU #1

Closed Hugo-cell111 closed 11 months ago

Hugo-cell111 commented 11 months ago

Hi! May I ask how many GPUs and what kind of GPU are needed in your work? Thanks!

zhibotian commented 11 months ago

Thank you for your interest in our work. For Pacal VOC 2012, we employed 4 NVIDIA RTX 3090 GPUs for both ResNet50 and ResNet101 models. In the case of Cityscapes, 4 NVIDIA RTX 3090 GPUs were used for ResNet50, whereas ResNet101 required 8 NVIDIA RTX 3090 GPUs. Further details can be found through the run bash scripts available at https://github.com/kunzhan/DSSN/blob/efaee86d3c023abc6b72dcda22ab1017ff71f1d4/voc/scripts/train.sh#L20 or from the output log files at https://github.com/kunzhan/DSSN/blob/efaee86d3c023abc6b72dcda22ab1017ff71f1d4/voc/exp/pascal/DSSN/r101/1_16/20230416_102707.log#L19

Hugo-cell111 commented 11 months ago

Thanks!