W&B resume ddp from run link fix (#2579)
* W&B resume ddp from run link fix * Native DDP W&B support for training, resuming
This commit is contained in:
1 parent
dc51e80b00
commit
518c09578e
2 files changed
+54
-16
No files matched your search
@@ -33,7 +33,7 @@ from utils.google_utils import attempt_download
|
||||
from utils.loss import ComputeLoss
|
||||
from utils.plots import plot_images, plot_labels, plot_results, plot_evolution
|
||||
from utils.torch_utils import ModelEMA, select_device, intersect_dicts, torch_distributed_zero_first, is_parallel
|
||||
from utils.wandb_logging.wandb_utils import WandbLogger, resume_and_get_id
|
||||
from utils.wandb_logging.wandb_utils import WandbLogger, check_wandb_resume
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
|
||||
@@ -496,7 +496,7 @@ if __name__ == '__main__':
|
||||
check_requirements()
|
||||
|
||||
# Resume
|
||||
wandb_run = resume_and_get_id(opt)
|
||||
wandb_run = check_wandb_resume(opt)
|
||||
if opt.resume and not wandb_run: # resume an interrupted run
|
||||
ckpt = opt.resume if isinstance(opt.resume, str) else get_latest_run() # specified or most recent path
|
||||
assert os.path.isfile(ckpt), 'ERROR: --resume checkpoint does not exist'
|
||||
|
||||
Reference in new issue
Block a user