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

修复bert 1x1下的时间统计bug #171

Merged
merged 1 commit into from
Jul 27, 2023
Merged

修复bert 1x1下的时间统计bug #171

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
5 changes: 3 additions & 2 deletions training/kunlunxin/bert-pytorch/extern/trainer_adapter.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,8 +7,9 @@
from torch.cuda.amp import GradScaler
from torch.nn.parallel import DistributedDataParallel as DDP
from torch.optim import Optimizer
from torch_xmlir.optimizer import FusedLAMB

import torch_xmlir
from torch_xmlir.optimizer import FusedLAMB
import torch_xmlir.core.xpu_model as xm

import utils
Expand Down Expand Up @@ -62,7 +63,7 @@ def backward(step: int,
optimizer: Optimizer,
grad_scaler: GradScaler = None):
loss.backward()

torch_xmlir.xpu.xpu_synchronize()
update_step = step % config.gradient_accumulation_steps == 0
if update_step:
update_model_params(loss, optimizer, grad_scaler)
Expand Down