-
Notifications
You must be signed in to change notification settings - Fork 57
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
[torchlib] Remove adaptive_avg_pool implementation #1751
Conversation
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #1751 +/- ##
==========================================
- Coverage 75.02% 74.98% -0.04%
==========================================
Files 245 245
Lines 26417 26393 -24
Branches 4810 4804 -6
==========================================
- Hits 19820 19792 -28
- Misses 5677 5679 +2
- Partials 920 922 +2 ☔ View full report in Codecov by Sentry. |
Test Results 24 files ± 0 24 suites ±0 2h 58m 34s ⏱️ - 11m 27s For more details on these failures, see this check. Results for commit c94031a. ± Comparison against base commit 874365e. This pull request removes 1864 and adds 392 tests. Note that renamed tests count towards both.
This pull request removes 125 skipped tests and adds 109 skipped tests. Note that renamed tests count towards both.
♻️ This comment has been updated with latest results. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What ops do we get after decomposition? I recall 1D is correct, but 2d and 3d is not passing test_fx_op_consistency.py
Good point. I can test that. In any case globalavgpool is incorrect so we need to either improve the implementation or fix decomp |
Remove adaptive_avg_pool implementation because our implementation using GlobalAveragePool is incorrect. We can rely on torch decomp instead.