Model post process for zero stage3 training #17187
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
pengwa
added
the
training
askhade
reviewed
Aug 22, 2023
orttraining/orttraining/python/training/ortmodule/_zero_stage3_compatibility.py
Outdated
Show resolved
Hide resolved
askhade
reviewed
Aug 22, 2023
orttraining/orttraining/python/training/ortmodule/_zero_stage3_compatibility.py
Outdated
Show resolved
Hide resolved
pengwa
force-pushed
the
pengwa/zero_post_process
branch
from
86704cf to
4e59594
Compare
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
| tensor_input_dtypes: List[torch.onnx.TensorProtoDataType], | ||
| ) -> Tuple[List[Optional[List[Union[int, str]]]], List[torch.onnx.TensorProtoDataType]]: | ||
| # output = input.matmul(weight.t()) | ||
| tensor_input_shapes[0] # input |
Check notice
Code scanning / CodeQL
Statement has no effect
Comment on lines
+51
to
+52
| # if ctx.current_step >= 0: | ||
| # print(f"{'='*6} Completed forward pass for STEP {ctx.current_step} {'='*6}") |
Check notice
Code scanning / CodeQL
Commented-out code
…pengwa/zero_post_process
askhade
reviewed
Sep 18, 2023
orttraining/orttraining/python/training/ortmodule/_graph_execution_manager.py
Show resolved
Hide resolved
|
Thank you a lot @askhade!!! |
kleiti
pushed a commit
to kleiti/onnxruntime
that referenced
this pull request
Mar 22, 2024
### Model post process for zero stage3 training This is the last change to make single GPU/Multiple GPUs run pass. Design details: https://microsoft.sharepoint.com/:p:/t/ONNX2/EfNfJ43necpIoPI6x5M2zvYBVbfjoPQmG4Boc_F7-tHm1w?e=ekQwA6&nav=eyJzSWQiOjMxNiwiY0lkIjoxMDE1Nzg3NDZ9 `PyTorch` runs with ZeROOffloadSubscriber: ``` model = prepare_model(...) from onnxruntime.training.utils.hooks import configure_ort_compatible_zero_stage3 configure_ort_compatible_zero_stage3() ``` `ORTModule` runs with ZeROOffloadSubscriber: ``` os.environ['ORTMODULE_ENABLE_ZERO_STAGE3'] = '1' from onnxruntime.training.ortmodule import ORTModule model = ORTModule(self.model) ``` It will be fairly easy to debug convergence issue if both ORT and PyTorch can run the same offload path. ### Motivation and Context <!-- - Why is this change required? What problem does it solve? - If it fixes an open issue, please link to the issue here. -->
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.
Model post process for zero stage3 training
This is the last change to make single GPU/Multiple GPUs run pass.
Design details: https://microsoft.sharepoint.com/:p:/t/ONNX2/EfNfJ43necpIoPI6x5M2zvYBVbfjoPQmG4Boc_F7-tHm1w?e=ekQwA6&nav=eyJzSWQiOjMxNiwiY0lkIjoxMDE1Nzg3NDZ9
PyTorchruns with ZeROOffloadSubscriber:ORTModuleruns with ZeROOffloadSubscriber:It will be fairly easy to debug convergence issue if both ORT and PyTorch can run the same offload path.
Motivation and Context