Use upstream PyTorch cost models instead of local copy #305
Conversation
This file contains hidden or 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
DEPENDENCY: Requires pytorch/pytorch branch `cost-model-consolidation` which adds `all_to_all_cost` and `include_compute_cost` to `torch.distributed.tensor._collective_utils`. Changes: - Remove local `all_to_all_cost` function (now imported from PyTorch) - Remove copy-pasted `redistribute_cost` function - Use PyTorch's `redistribute_cost` with `include_compute_cost=True` - Deprecate `order` parameter (PyTorch now uses _gen_transform_infos for optimal ordering) - Add comprehensive tests for cost model functions This consolidation: - Reduces code duplication between PyTorch and Autoparallel - Ensures cost models stay in sync across projects - Removes the 5x hack in all_to_all_cost (calibration moved to PyTorch) Authored with Claude.
meta-cla
bot
added
the
CLA Signed
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.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.
DEPENDENCY: Requires pytorch/pytorch branch
cost-model-consolidationwhich addsall_to_all_costandinclude_compute_costtotorch.distributed.tensor._collective_utils.Changes:
all_to_all_costfunction (now imported from PyTorch)redistribute_costfunctionredistribute_costwithinclude_compute_cost=Trueorderparameter (PyTorch now uses _gen_transform_infos for optimal ordering)This consolidation:
Authored with Claude.