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.

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

Run optimizer inside parallel do if the forward pass contains parallel do. #8620

Closed
wants to merge 5 commits into from

Conversation

helinwang
Copy link
Contributor

@helinwang helinwang commented Feb 28, 2018

Fixes: #8592

Tested work on parallel do version (modified test_recognize_digits.py) with or without NCCL (both converges), also tested on non-parallel version fit a line.

Ready for review to see if you guys agree with the general direction, and I need to do more testing. Thanks!

if (!exists && platform::is_gpu_place(place)) {
auto *dst = sub_scope->Var(param)->GetMutable<LoDTensor>();
framework::TensorCopy(src, place, dst);
}
Copy link

@tonyyang-svail tonyyang-svail Mar 1, 2018

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

parallel_do_op.cc:150:  Tab found; better to use spaces  [whitespace/tab] [1]
parallel_do_op.cc:151:  Tab found; better to use spaces  [whitespace/tab] [1]
parallel_do_op.cc:152:  Tab found; better to use spaces  [whitespace/tab] [1]
parallel_do_op.cc:153:  Tab found; better to use spaces  [whitespace/tab] [1]
parallel_do_op.cc:156:  Tab found; better to use spaces  [whitespace/tab] [1]

@helinwang
Copy link
Contributor Author

Closing, this is no longer needed.

@helinwang helinwang closed this May 2, 2018
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants