Skip to content
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

GH-15857: AutoML pipeline support #16041

Merged
merged 2 commits into from
Feb 15, 2024
Merged

GH-15857: AutoML pipeline support #16041

merged 2 commits into from
Feb 15, 2024

Conversation

sebhrusen
Copy link
Contributor

integrate TE in AutoML as a pipeline transformer: #15857

@sebhrusen sebhrusen added this to the 3.46.0.1 milestone Jan 29, 2024
@sebhrusen sebhrusen self-assigned this Jan 29, 2024
@sebhrusen sebhrusen linked an issue Jan 29, 2024 that may be closed by this pull request
Comment on lines +416 to +426
//TODO: given that a transformer can reference a model (e.g. TE),
// and multiple transformers can refer
// to the same model,
// then we should be careful when deleting a transformer (resp. an entire pipeline)
// as we may delete sth that is still in use by another transformer (resp. pipeline).
// --> ref count?

//TODO: in AutoML, the same transformations are likely to occur on multiple (sometimes all) models,
// especially if the transformers parameters are not tuned.
// But it also depends if the transformers are context(CV)-sensitive (e.g. Target Encoding).
// See `CachingTransformer` for some thoughts about this.
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

these are just things to keep in mind, but not currently needed

wendycwong
wendycwong previously approved these changes Feb 11, 2024
tomasfryda
tomasfryda previously approved these changes Feb 12, 2024
Copy link
Contributor

@tomasfryda tomasfryda left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks good to me, thank you @sebhrusen !

Base automatically changed from seb/gh-15856 to master February 12, 2024 23:44
@sebhrusen sebhrusen dismissed stale reviews from tomasfryda and wendycwong February 12, 2024 23:44

The base branch was changed.

@sebhrusen
Copy link
Contributor Author

@wendycwong, @tomasfryda this one is ready for merge.
Double checked the failing AutoML test (http://mr-0xg1:8080/job/h2o-3-pipeline/job/PR-16041/12/testReport/junit/ai.h2o.automl/AutoMLTest/Java_8_AutoML_JUnit___test_algos_have_default_parameters_enforcing_reproducibility/), ran in multiple times locally (single+multi nodes) without key leakage so don't know what happened there: looks like the test got interrupted.

@sebhrusen sebhrusen merged commit 17fa9ee into master Feb 15, 2024
64 of 68 checks passed
@sebhrusen sebhrusen deleted the seb/gh-15857 branch February 15, 2024 23:32
mn-mikke added a commit that referenced this pull request Feb 27, 2024
valenad1 added a commit that referenced this pull request Mar 8, 2024
valenad1 added a commit that referenced this pull request Mar 11, 2024
* Revert "GH-15857: cleanup legacy TE integration in ModelBuilder and AutoML (#16061)"

This reverts commit a8f309b.

* Revert "GH-15857: AutoML pipeline support (#16041)"

This reverts commit 17fa9ee.

* Revert "GH-15856: Grid pipeline support (#16040)"

This reverts commit b7ac670.

* Revert "GH-15855: core pipeline API (#16039)"

This reverts commit c15ea1e.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

AutoML Pipeline – Integration to AutoML
3 participants