Support custom trainer and backend #91
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
added 3 commits
December 3, 2022 01:40
yhna940
reviewed
Dec 15, 2022
| class _CustomTorchBackend(_TorchBackend): | ||
| share_cuda_visible_devices: bool = True | ||
|
|
||
| def on_start(self, worker_group: WorkerGroup, |
There was a problem hiding this comment.
If the process group is not initialized, how about initializing it here without throwing an error?
There was a problem hiding this comment.
I think we have no way to know whether the process group is initialized in Task.run method.
yhna940
approved these changes
Dec 15, 2022
2 tasks
nijkah
reviewed
Dec 15, 2022
nijkah
reviewed
Dec 15, 2022
Co-authored-by: Hakjin Lee <nijkah@gmail.com> Signed-off-by: Junhwa Song <ethan9867@gmail.com>
nijkah
approved these changes
Dec 15, 2022
KKIEEK
commented
Dec 15, 2022
Signed-off-by: Junhwa Song <ethan9867@gmail.com>
KKIEEK
force-pushed
the
v2.1.0/custom
branch
from
c0b83ab to
9e54600
Compare
KKIEEK
added a commit
that referenced
this pull request
Dec 19, 2022
* Bump ray from 1.9.1 to 2.1.0 * Fix deprecated warning * Refactor * Fix modules * Fix requirements * Fix test code * Support custom trainer and backend (#91) * Upgrade MMTask (#97) * Fix minor (#100) * Fix blocking issue at test_tasks.py * Support single GPU tuning * Bump FLAML to v1.0.14 to avoid deprecated warning * Supplement documentations (#102) * Support resume (#104) Co-authored-by: Younghwan Na <100389977+yhna940@users.noreply.github.com> Co-authored-by: Hakjin Lee <nijkah@gmail.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Motivation
Since MM-based repositories already use
torch.distributed.init_process_group, usingTorchTrainerfor DDP in ray framework causes theRuntimeError("trying to initialize the default process group twice!").To solve this problem, I introduced a custom backend modified from here.