Move outcome_names from BenchmarkRunner to BenchmarkTestFunction
#3021
+44
−24
Conversation
This file contains 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
Summary:
**Context**: This will enable constructing the `BenchmarkRunner` based on the `BenchmarkProblem` and `BenchmarkMethod` rather than asking the user to provide it. In addition to making things simpler (it's weird that a runner is part of a problem!), that will enable the Runner to be aware of aspects of the method, such as parallelism.
This will also enable us to return metrics in a dict format (`{outcome_name: value}`) if we choose to do so in the future. That may be simpler since the data already gets processed into dicts by the runner.
Note that for problems based on BoTorch problems, names are usually already set programmatically, so that logic moves to the test problem.
**This diff**:
* Requires `outcome_names` on `BenchmarkTestFunction`
* Removes `outcome_names` as an argument from `BenchmarkRunner`
* Sets outcome names automatically on `BoTorchTestFunction` when they are not provided, following the convention used elsewhere.
Update usages:
* Remove `outcome_names` from calls to `BenchmarkRunner`
* Add `outcome_names` to calls to `BenchmarkTestFunction`, where needed; they are generally already present on surroagate test functions and can be constructed automatically for BoTorch-based problems.
Differential Revision: D65497700
facebook-github-bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D65497700 |
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #3021 +/- ##
=======================================
Coverage 95.60% 95.60%
=======================================
Files 483 483
Lines 49012 49024 +12
=======================================
+ Hits 46859 46871 +12
Misses 2153 2153 ☔ View full report in Codecov by Sentry. |
esantorella
added a commit
to esantorella/Ax
that referenced
this pull request
Nov 5, 2024
facebook#3021) Summary: **Context**: This will enable constructing the `BenchmarkRunner` based on the `BenchmarkProblem` and `BenchmarkMethod` rather than asking the user to provide it. In addition to making things simpler (it's weird that a runner is part of a problem!), that will enable the Runner to be aware of aspects of the method, such as parallelism. This will also enable us to return metrics in a dict format (`{outcome_name: value}`) if we choose to do so in the future. That may be simpler since the data already gets processed into dicts by the runner. Note that for problems based on BoTorch problems, names are usually already set programmatically, so that logic moves to the test problem. **This diff**: * Requires `outcome_names` on `BenchmarkTestFunction` * Removes `outcome_names` as an argument from `BenchmarkRunner` * Sets outcome names automatically on `BoTorchTestFunction` when they are not provided, following the convention used elsewhere. Update usages: * Remove `outcome_names` from calls to `BenchmarkRunner` * Add `outcome_names` to calls to `BenchmarkTestFunction`, where needed; they are generally already present on surroagate test functions and can be constructed automatically for BoTorch-based problems. Differential Revision: D65497700
esantorella
added a commit
to esantorella/Ax
that referenced
this pull request
Nov 5, 2024
facebook#3021) Summary: **Context**: This will enable constructing the `BenchmarkRunner` based on the `BenchmarkProblem` and `BenchmarkMethod` rather than asking the user to provide it. In addition to making things simpler (it's weird that a runner is part of a problem!), that will enable the Runner to be aware of aspects of the method, such as parallelism. This will also enable us to return metrics in a dict format (`{outcome_name: value}`) if we choose to do so in the future. That may be simpler since the data already gets processed into dicts by the runner. Note that for problems based on BoTorch problems, names are usually already set programmatically, so that logic moves to the test problem. **This diff**: * Requires `outcome_names` on `BenchmarkTestFunction` * Removes `outcome_names` as an argument from `BenchmarkRunner` * Sets outcome names automatically on `BoTorchTestFunction` when they are not provided, following the convention used elsewhere. Update usages: * Remove `outcome_names` from calls to `BenchmarkRunner` * Add `outcome_names` to calls to `BenchmarkTestFunction`, where needed; they are generally already present on surroagate test functions and can be constructed automatically for BoTorch-based problems. Differential Revision: D65497700
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.
Summary:
Context: This will enable constructing the
BenchmarkRunnerbased on theBenchmarkProblemandBenchmarkMethodrather than asking the user to provide it. In addition to making things simpler (it's weird that a runner is part of a problem!), that will enable the Runner to be aware of aspects of the method, such as parallelism.This will also enable us to return metrics in a dict format (
{outcome_name: value}) if we choose to do so in the future. That may be simpler since the data already gets processed into dicts by the runner.Note that for problems based on BoTorch problems, names are usually already set programmatically, so that logic moves to the test problem.
This diff:
outcome_namesonBenchmarkTestFunctionoutcome_namesas an argument fromBenchmarkRunnerBoTorchTestFunctionwhen they are not provided, following the convention used elsewhere.Update usages:
outcome_namesfrom calls toBenchmarkRunneroutcome_namesto calls toBenchmarkTestFunction, where needed; they are generally already present on surroagate test functions and can be constructed automatically for BoTorch-based problems.Differential Revision: D65497700