search

raoyongming / DenseCLIP

[CVPR 2022] DenseCLIP: Language-Guided Dense Prediction with Context-Aware Prompting
505 stars 38 forks source link

question on the device #44

Closed duoxiangqinzuo closed 8 months ago

duoxiangqinzuo commented 11 months ago

Which and how many gpus do you use？How long does it take to complete a training session？

raoyongming commented 11 months ago

Hi, thanks for your interest in our work. In our experiments, we used 8x 3090 GPUs for ResNets and 8x A100 40G GPUs for ViT models. It would take roughly a day to complete training.