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.

Sign up for GitHub

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

Jump to bottom

[ENH] Adding leaf node samples to be stored when "quantile" tree is turned on #45

Merged
merged 21 commits into from
Jun 15, 2023
Merged

[ENH] Adding leaf node samples to be stored when "quantile" tree is turned on #45

merged 21 commits into from
Jun 15, 2023

Conversation

adam2392
Copy link
Collaborator

Reference Issues/PRs

Addresses the quantile-trees part of: neurodata/treeple#29

What does this implement/fix? Explain your changes.

  1. Stores for each leaf node a 2D numpy array of the y-samples (remember y is (n_samples, n_outputs))
  2. Does this all the way in Criterion
  3. Only supports supervised tree/splitter/criterion

Any other comments?

@adam2392
Adding quantiles
83f807b 
Signed-off-by: Adam Li <adam2392@gmail.com>
adam2392 and others added 11 commits June 13, 2023 22:59
@adam2392
Adding quantiles predict method
7506308 
Signed-off-by: Adam Li <adam2392@gmail.com>
@adrinjalali @jeremiedbb
MNT mark non-standard inverse_transform X args as non-metadata (SLEP6) (
83cb686 
scikit-learn#26503)

Co-authored-by: jeremiedbb <jeremiedbb@yahoo.fr>
@jeremiedbb @ogrisel
Fix PowerTransformer leaves constant feature unchanged (scikit-learn#…
e5df5fe 
…26566)

Co-authored-by: Olivier Grisel <olivier.grisel@ensta.org>
@thomasjpfan
FIX Ensure that index is correct when global transform_output is pand…
96878ba 
…as (scikit-learn#26454)
@jeremiedbb
MAINT Param validation: decorate all estimators with _fit_context (sc…
9c266cf 
…ikit-learn#26473)
@glemaitre
FIX use np.nan instead of None for missing marker in fetch_openml (sc…
f721c6d 
…ikit-learn#26579)
@jeremiedbb
DOC Update author tables for 1.3 (scikit-learn#26572)
e99b421
@jeremiedbb
MAINT reorder and clean up what's new 1.3 (scikit-learn#26567)
03ab632
@thomasjpfan
CLN Renames missing_values_in_feature_mask (scikit-learn#26580)
15f7cfb
@glemaitre @jeremiedbb @ogrisel
FEA add ValidationCurveDisplay in model_selection module (scikit-lear…
8cac52f 
…n#25120)

Co-authored-by: Jérémie du Boisberranger <34657725+jeremiedbb@users.noreply.github.com>
Co-authored-by: Olivier Grisel <olivier.grisel@ensta.org>
@adam2392
Working quantile tree and quantile forest
34c4221 
Signed-off-by: Adam Li <adam2392@gmail.com>
@adam2392
Copy link
Collaborator Author

@sampan501 and @PSSF23 you both might be interested.

This is a WIP, but almost done addition of code into the sklearn-fork to enable all trees to optionally store all leaf-node samples. This is useful for:

  1. quantile regression
  2. implementing procedures where we use samples in the leaves (e.g. trying to marginalize a forest, etc.)

@adam2392 adam2392 merged commit 21ccb30 into submodulev2 Jun 15, 2023
1 of 4 checks passed
@adam2392 adam2392 deleted the quantiles branch June 15, 2023 14:35
PSSF23
PSSF23 reviewed Jun 16, 2023
View reviewed changes
Copy link
Member

@PSSF23 PSSF23 left a comment

Choose a reason for hiding this comment

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

@adam2392 predict_quantiles functions have check_input documentation but the param is not in the function. Meanwhile get_leaf_node_samples has the param but it's not documented. These might cause the errors below:

https://app.circleci.com/pipelines/github/neurodata/scikit-tree/463/workflows/b92f9a94-dbb3-4b9d-b1f0-5a75c9941062/jobs/771

@PSSF23 PSSF23 mentioned this pull request Jun 16, 2023
Merged
5 tasks
@adam2392
Copy link
Collaborator Author

I pushed a commit 3f5cb65

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@PSSF23 PSSF23 PSSF23 left review comments

@sampan501 sampan501 Awaiting requested review from sampan501

Assignees
No one assigned
Labels
cython module:tree
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

None yet
8 participants
@adam2392 @PSSF23 @adrinjalali @jeremiedbb @thomasjpfan @glemaitre @rprkh @StefanieSenger