Open xLuge opened 3 years ago
I train deeplab on 4gpus,each gpu 16g, but Each EPOCH training time takes more than an hour.Is this normal?
The training using DataParallel is really slow (due to syncbn). You may try using train_dist.py, which is much faster.
train_dist.py
I train deeplab on 4gpus,each gpu 16g, but Each EPOCH training time takes more than an hour.Is this normal?