-
Notifications
You must be signed in to change notification settings - Fork 25.7k
Try to simplify FloorDiv axioms implications when needed during evaluations. #141267
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
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/141267
Note: Links to docs will display an error until the docs builds have been completed. ✅ You can merge normally! (4 Unrelated Failures)As of commit f3b78c1 with merge base 995e307 ( FLAKY - The following jobs failed but were likely due to flakiness present on trunk:
UNSTABLE - The following jobs failed but were likely due to flakiness present on trunk and has been marked as unstable:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
|
This pull request was exported from Phabricator. Differential Revision: D66307433 |
73bc344 to
063c514
Compare
|
This pull request was exported from Phabricator. Differential Revision: D66307433 |
063c514 to
e587f3c
Compare
|
This pull request was exported from Phabricator. Differential Revision: D66307433 |
e587f3c to
017db3f
Compare
|
This pull request was exported from Phabricator. Differential Revision: D66307433 |
017db3f to
0410ea6
Compare
|
This pull request was exported from Phabricator. Differential Revision: D66307433 |
0410ea6 to
4dfffa5
Compare
|
This pull request was exported from Phabricator. Differential Revision: D66307433 |
4dfffa5 to
68803fb
Compare
|
This pull request was exported from Phabricator. Differential Revision: D66307433 |
|
removing k.free_symbols.issubset(expr.free_symbols) remove the regression on torch rec. update the PR |
931e969 to
da0d573
Compare
|
This pull request was exported from Phabricator. Differential Revision: D66307433 |
da0d573 to
fd8829c
Compare
|
This pull request was exported from Phabricator. Differential Revision: D66307433 |
…is changed. (pytorch#141267) Summary: This very much the same solution proposed by bobrenjc93 except that it restrict it to expressions and axioms that have FloorDiv, since those are the only ones that could have became CleanDiv. and the only one that can changes as shape env changes. This also does not break torchrec benchmarks, it might be worth it to know why the generalization of this does break the torchrec benchmarks, but we could just be hitting another bug or NYI situation. ovearhead? None on ``` buck2 run fbcode//mode/opt fbcode//torchrec/distributed/tests:pt2_compile_benchmark -- --num-features=2000 ``` Differential Revision: D66307433 cc ezyang SherlockNoMad EikanWang jgong5 wenzhe-nrv D66307433 Test Plan: run existing tests. ``` TORCH_LOGS="+dynamo" TORCHDYNAMO_VERBOSE=1 buck2 run mode/opt //minimal_viable_ai/fire:light -- -d ~/fbsource/fbcode/hpc/ -d ~/fbsource/fbcode/hammer -d ~/fbsource/fbcode/minimal_viable_ai/ minimal_viable_ai/new_format/preproc_ta/experimental/pt2_export_train_preproc_roo_for_umia.py --start-ts 2024-11-10+10:00:00 --end-ts 2024-11-10+11:00:00 --steps 1 --preproc-type TA_accel_cpu_eager --export-archive-path /home/lsakka ``` ``` buck2 run fbcode//mode/opt fbcode//torchrec/distributed/tests:pt2_compile_benchmark -- --num-features=10 ``` Reviewed By: ezyang
fd8829c to
3bc41fc
Compare
|
This pull request was exported from Phabricator. Differential Revision: D66307433 |
3bc41fc to
00908d9
Compare
00908d9 to
176949b
Compare
|
This pull request was exported from Phabricator. Differential Revision: D66307433 |
1 similar comment
|
This pull request was exported from Phabricator. Differential Revision: D66307433 |
176949b to
b37ba44
Compare
|
This pull request was exported from Phabricator. Differential Revision: D66307433 |
…is changed. (pytorch#141267) Summary: This very much the same solution proposed by bobrenjc93 except that it restrict it to expressions and axioms that have FloorDiv, since those are the only ones that could have became CleanDiv. and the only one that can changes as shape env changes. This also does not break torchrec benchmarks, it might be worth it to know why the generalization of this does break the torchrec benchmarks, but we could just be hitting another bug or NYI situation. ovearhead? None on ``` buck2 run fbcode//mode/opt fbcode//torchrec/distributed/tests:pt2_compile_benchmark -- --num-features=2000 ``` Differential Revision: D66307433 cc ezyang SherlockNoMad EikanWang jgong5 wenzhe-nrv D66307433 Test Plan: run existing tests. ``` TORCH_LOGS="+dynamo" TORCHDYNAMO_VERBOSE=1 buck2 run mode/opt //minimal_viable_ai/fire:light -- -d ~/fbsource/fbcode/hpc/ -d ~/fbsource/fbcode/hammer -d ~/fbsource/fbcode/minimal_viable_ai/ minimal_viable_ai/new_format/preproc_ta/experimental/pt2_export_train_preproc_roo_for_umia.py --start-ts 2024-11-10+10:00:00 --end-ts 2024-11-10+11:00:00 --steps 1 --preproc-type TA_accel_cpu_eager --export-archive-path /home/lsakka ``` ``` buck2 run fbcode//mode/opt fbcode//torchrec/distributed/tests:pt2_compile_benchmark -- --num-features=10 ``` Reviewed By: ezyang
b37ba44 to
f3b78c1
Compare
|
This pull request was exported from Phabricator. Differential Revision: D66307433 |
|
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
…ations. (pytorch#141267) Summary: This very much the same solution proposed by bobrenjc93 except that it restrict it to expressions and axioms that have FloorDiv, since those are the only ones that could have became CleanDiv. and the only one that can changes as shape env changes. This also does not break torchrec benchmarks, it might be worth it to know why the generalization of this does break the torchrec benchmarks, but we could just be hitting another bug or NYI situation. ovearhead? None on ``` buck2 run fbcode//mode/opt fbcode//torchrec/distributed/tests:pt2_compile_benchmark -- --num-features=1000 ``` Differential Revision: D66307433 Pull Request resolved: pytorch#141267 Approved by: https://github.com/ezyang
Summary:
This very much the same solution proposed by bobrenjc93 except that it restrict it to expressions and axioms that have FloorDiv, since those are the only ones that could have became CleanDiv. and the only one that can changes as shape env changes.
This also does not break torchrec benchmarks, it might be worth it to know why the generalization of this does break the torchrec benchmarks, but we could just be hitting another bug or NYI situation.
ovearhead?
None on
Differential Revision: D66307433
cc @ezyang @SherlockNoMad @EikanWang @jgong5 @wenzhe-nrv @voznesenskym @penguinwu @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @jiayisunx @chenyang78 @kadeng @chauhang @amjames