Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

add error if grads nonfinite #1

Merged
merged 1 commit into from
Jul 29, 2022
Merged

add error if grads nonfinite #1

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
26 changes: 17 additions & 9 deletions util/misc.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -45,7 +45,8 @@ def synchronize_between_processes(self):
"""
if not is_dist_avail_and_initialized():
return
t = torch.tensor([self.count, self.total], dtype=torch.float64, device='cuda')
t = torch.tensor([self.count, self.total],
dtype=torch.float64, device='cuda')
dist.barrier()
dist.all_reduce(t)
t = t.tolist()
Expand Down Expand Up @@ -218,7 +219,8 @@ def init_distributed_mode(args):
args.rank = int(os.environ['OMPI_COMM_WORLD_RANK'])
args.world_size = int(os.environ['OMPI_COMM_WORLD_SIZE'])
args.gpu = int(os.environ['OMPI_COMM_WORLD_LOCAL_RANK'])
args.dist_url = "tcp://%s:%s" % (os.environ['MASTER_ADDR'], os.environ['MASTER_PORT'])
args.dist_url = "tcp://%s:%s" % (
os.environ['MASTER_ADDR'], os.environ['MASTER_PORT'])
os.environ['LOCAL_RANK'] = str(args.gpu)
os.environ['RANK'] = str(args.rank)
os.environ['WORLD_SIZE'] = str(args.world_size)
Expand Down Expand Up @@ -254,13 +256,16 @@ class NativeScalerWithGradNormCount:
def __init__(self):
self._scaler = torch.cuda.amp.GradScaler()

def __call__(self, loss, optimizer, clip_grad=None, parameters=None, create_graph=False, update_grad=True):
def __call__(self, loss, optimizer, clip_grad=None, parameters=None, create_graph=False, update_grad=True,
error_if_nonfinite: bool = False):
self._scaler.scale(loss).backward(create_graph=create_graph)
if update_grad:
if clip_grad is not None:
assert parameters is not None
self._scaler.unscale_(optimizer) # unscale the gradients of optimizer's assigned params in-place
norm = torch.nn.utils.clip_grad_norm_(parameters, clip_grad)
# unscale the gradients of optimizer's assigned params in-place
self._scaler.unscale_(optimizer)
norm = torch.nn.utils.clip_grad_norm_(
parameters, clip_grad, error_if_nonfinite=error_if_nonfinite)
else:
self._scaler.unscale_(optimizer)
norm = get_grad_norm_(parameters)
Expand All @@ -286,9 +291,11 @@ def get_grad_norm_(parameters, norm_type: float = 2.0) -> torch.Tensor:
return torch.tensor(0.)
device = parameters[0].grad.device
if norm_type == inf:
total_norm = max(p.grad.detach().abs().max().to(device) for p in parameters)
total_norm = max(p.grad.detach().abs().max().to(device)
for p in parameters)
else:
total_norm = torch.norm(torch.stack([torch.norm(p.grad.detach(), norm_type).to(device) for p in parameters]), norm_type)
total_norm = torch.norm(torch.stack([torch.norm(
p.grad.detach(), norm_type).to(device) for p in parameters]), norm_type)
return total_norm


Expand All @@ -309,7 +316,8 @@ def save_model(args, epoch, model, model_without_ddp, optimizer, loss_scaler):
save_on_master(to_save, checkpoint_path)
else:
client_state = {'epoch': epoch}
model.save_checkpoint(save_dir=args.output_dir, tag="checkpoint-%s" % epoch_name, client_state=client_state)
model.save_checkpoint(save_dir=args.output_dir, tag="checkpoint-%s" %
epoch_name, client_state=client_state)


def load_model(args, model_without_ddp, optimizer, loss_scaler):
Expand Down Expand Up @@ -337,4 +345,4 @@ def all_reduce_mean(x):
x_reduce /= world_size
return x_reduce.item()
else:
return x
return x