[Usage] how to train model with only one GPU
SidneyRey opened this issue · comments
SidneyRey commented
I used “pods_train --num-gpus 1” to train OTA, but got error: RuntimeError: Default process group has not been initialized, please make sure to call init_process_group.
how to train detection model with only one GPU?
thank you
SidneyRey commented
sorry,
training with one GPU, do not need norm_sync.