search

ethnhe / FFB6D

[CVPR2021 Oral] FFB6D: A Full Flow Bidirectional Fusion Network for 6D Pose Estimation.
MIT License
295 stars 72 forks source link

How to run the code on the single GPU? #37

Closed LZL-CS closed 2 years ago

LZL-CS commented 3 years ago

Hi,

Thank you very much for your nice work! I have a look at your code and your model run with 8 GPUs, but I have only one GPU, may I know how can modify the code to run the code?

Kind regards,

cwkim1993 commented 3 years ago

You can train with single GPU by changing line in train_lm.sh file like below.. python -m torch.distributed.launch --nproc_per_node=1 train_lm.py --gpu "0" --gpus 1

LZL-CS commented 3 years ago

You can train with single GPU by changing line in train_lm.sh file like below.. python -m torch.distributed.launch --nproc_per_node=1 train_lm.py --gpu "0" --gpus 1

Ok, I'll have a try, thanks for your reply!