Fix optimization direction in tune_mcgrad_params #202
Closed
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
flinder
requested review from
Lorenzo-Perini,
TaXxER and
danielha1
as code owners
meta-cla
bot
added
the
CLA Signed
|
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #202 +/- ##
=======================================
Coverage ? 94.92%
=======================================
Files ? 9
Lines ? 1753
Branches ? 0
=======================================
Hits ? 1664
Misses ? 89
Partials ? 0
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
flinder
force-pushed
the
export-D91915530
branch
from
f01783c to
e39f664
Compare
Summary: Add a `use_model_predictions` parameter to `tune_mcgrad_params` with `False` as the default. When False, returns the actual observed best trial parameters. When True, uses the Bayesian surrogate model's predicted best parameters. The default is False because with few tuning trials, the surrogate model has high uncertainty and may predict a different optimum than the best observed trial. Users who want the model-predicted best can set this to True. Reviewed By: Lorenzo-Perini Differential Revision: D91884588
Summary: D90397011 migrated from the deprecated AxClient to the new Client API. However, the migration lost the `minimize=True` setting for the objective. In the new Ax API, the objective string convention is: - No prefix = maximize (default) - "-" prefix = minimize The old code had `ObjectiveProperties(minimize=True)`, but the new code used `objective="normalized_entropy"` without the minus sign, causing Ax to maximize instead of minimize. This resulted in selecting parameters with the worst (highest) normalized entropy instead of the best (lowest). This explains customer reports of bad parameters and MCBoost returning 0 rounds since the Jan 13th landing of D90397011. Reviewed By: leepface, TaXxER Differential Revision: D91915530
flinder
force-pushed
the
export-D91915530
branch
from
e39f664 to
47ba371
Compare
|
This pull request has been merged in 2986799. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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.
Summary:
D90397011 migrated from the deprecated AxClient to the new Client API. However,
the migration lost the
minimize=Truesetting for the objective.In the new Ax API, the objective string convention is:
The old code had
ObjectiveProperties(minimize=True), but the new code usedobjective="normalized_entropy"without the minus sign, causing Ax to maximizeinstead of minimize. This resulted in selecting parameters with the worst
(highest) normalized entropy instead of the best (lowest).
This explains customer reports of bad parameters and MCBoost returning 0 rounds
since the Jan 13th landing of D90397011.
Differential Revision: D91915530