Profile method 1 stabilisation and default#1713
Draft
Doresic wants to merge 4 commits into
Draft
Conversation
|
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## profile-multi_start_profiling #1713 +/- ##
=================================================================
+ Coverage 84.10% 84.12% +0.01%
=================================================================
Files 164 164
Lines 14646 14661 +15
=================================================================
+ Hits 12318 12333 +15
Misses 2328 2328 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
…abilisation_and_default
…abilisation_and_default
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Profiling with extrapolation can be unstable when parameters that do not actually move with the profiled parameter still get extrapolated. In that case, the recent profile history mostly reflects optimizer noise, and extrapolating from it can make the profile walk bounce around or fail.
I changed the defaults toward more conservative behavior:
next_guess_methodfromadaptive_step_order_1toadaptive_step_order_0reg_orderfrom4to1correlation_threshold(default0.7) foradaptive_step_regressionThe idea behind the correlation filter is simple: only extrapolate parameters that are actually moving together with the profiled one. If a parameter is not moving with the profiled parameter in the current region, we now keep it at its current value and let the optimizer decide instead of extrapolating noise.
I also clarified the docstrings around
adaptive_step_order_1vsadaptive_step_regression, kept regression rank reduction local to each parameter, and extended the related option validation.This also moves the defaults toward the benchmark-backed
o0 n6 mediumconfiguration.