Validate that X is 2 dimensional

[jcrist]

sklearn requires that X is 2-dimensional, and errors nicely otherwise.

cuml doesn't uniformly require that X is 2-dimensional. Some estimators error on non-2D-X, but they usually do so accidentally, not as part of the input validation code. This leads to a bunch of xfailed tests in our sklearn compatibility test suite (both in the cuml.accel upstream tests, as well as in test_sklearn_compatibility.py).

This PR:

• Adds a deprecation warning to all estimators when X is non-2-dimensional. In 26.06 we'll remove the deprecation warning and error instead.
• If cuml.accel is enabled, this warning is an error matching the sklearn error message instead. This lets us un-xfail a bunch of tests right now.
• Updates our test suite to not trigger the warning, ensuring we're always passing in 2-dimensional X in tests. This was most common for TargetEncoder, only a few other locations needed it. Since it was so common for TargetEncoder, I added a deprecation test there as well to check that everything still worked on 1D inputs.
• Updates reflect to support reset="type", for setting the type on fit-like functions alone (and not n_features_in_/feature_names_in_). This was needed for 2 "transformers" sklearn (and cuml) supports that are meant to operate on y instead of X (LabelEncoder and LabelBinarizer). These estimators operate on y alone and shouldn't support n_features_in_/feature_names_in_. They also shouldn't validate that the array input is 2 dimensional, since y can be 1D. This is a stop-gap solution as we refactor our validation functions - in the long run we might remove reset entirely from reflect and instead move setting the input type to the validation functions (with the reflect decorator only remaining for coercing outputs).

As per our deprecation policy, I've marked this PR as "breaking" since it adds a new deprecation warning around non-2-dimensional X. All prior working code should continue to work, users providing 1D X should just see a warning.

[csadorf]

LGTM!

[coderabbitai]

📝 Walkthrough

Summary by CodeRabbit

• Bug Fixes

  • Enforced stricter 2D array input validation across estimators; 1D inputs now trigger deprecation warnings.
  • Enhanced error messages for invalid input shapes and edge cases.

• Documentation

  • Updated TargetEncoder examples to demonstrate 2D DataFrame inputs.

• Tests

  • Added validation tests for features infrastructure behavior.
  • Updated test inputs to comply with 2D array requirements.

Walkthrough

This PR enforces 2D input validation across cuML's input handling pipeline, introduces a new "type" reset mode in the reflect decorator for fine-grained feature tracking control, updates preprocessing examples and tests to use 2D inputs, and re-enables previously skipped tests by removing xfail entries.

Changes

Cohort / File(s)	Summary
Feature Validation & Reset Semantics python/cuml/cuml/internals/outputs.py, python/cuml/cuml/internals/validation.py	Updated reflect() to support reset="type" (in addition to bool), with conditional logic that skips certain feature checks for "type" mode. Strengthened _get_n_features() with explicit 2D validation, detailed error messages, and FutureWarning deprecation for non-2D inputs when acceleration is disabled.
Preprocessor & Decorator Updates python/cuml/cuml/preprocessing/label.py, python/cuml/cuml/preprocessing/TargetEncoder.py	Changed LabelBinarizer.fit decorator from reset=True to reset="type". Updated TargetEncoder docstring example to demonstrate 2D DataFrame input instead of 1D Series.
Core Algorithm Tests python/cuml/tests/test_coordinate_descent.py, python/cuml/tests/test_dbscan.py, python/cuml/tests/test_tsne.py, python/cuml/tests/test_validation.py	Reshaped test inputs to 2D arrays; updated validation tests to assert 3D input raises ValueError and added parametrized test for 1D input deprecation warnings.
Feature Attribution Tests python/cuml/tests/test_label_binarizer.py, python/cuml/tests/test_label_encoder.py	Added new tests verifying LabelBinarizer and LabelEncoder do not attach n_features_in_ attribute, confirming feature infrastructure exclusion.
Integration & Encoder Tests python/cuml/tests/test_target_encoder.py	Refactored TargetEncoder tests to use 2D DataFrame inputs for X and 1D Series for y; added deprecation warning expectations and updated output assertions for new input pathways.
Test Configuration python/cuml_accel_tests/upstream/scikit-learn/xfail-list.yaml	Removed multiple xfail entries to re-enable tests; updated TSNE validation marker from cuml_accel_tsne_validation_on_init to cuml_accel_tsne_validations.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~25 minutes

Possibly related PRs

• #7877: Modifies cuml.internals.validation with overlapping feature/2D input validation logic; this PR's reflect() wrapper now conditions check_features() calls affecting both changes.

Suggested labels

Cython / Python

Suggested reviewers

• dantegd
• viclafargue
• csadorf

🚥 Pre-merge checks | ✅ 2 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 37.93% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (2 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title 'Validate that X is 2 dimensional' directly summarizes the main change: enforcing validation that input array X must be 2-dimensional across all estimators.
Description check	✅ Passed	The description thoroughly explains the rationale, implementation details, and changes made to enforce 2D X validation, including deprecation warnings, test updates, and the reflect function modifications.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment

📝 Coding Plan

• Generate coding plan for human review comments

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

Tip

You can customize the high-level summary generated by CodeRabbit.

Configure the reviews.high_level_summary_instructions setting to provide custom instructions for generating the high-level summary.

[coderabbitai]

Actionable comments posted: 3

🤖 Prompt for all review comments with AI agents

Verify each finding against the current code and only fix it if needed.

Inline comments:
In `@python/cuml/cuml/internals/validation.py`:
- Around line 107-112: The deprecation warning emitted in
cuml/internals/validation.py uses warnings.warn without a stacklevel, so it
points to this internal function instead of the user's callsite; update the
warnings.warn call in this module to pass an appropriate stacklevel (e.g.,
stacklevel=2 or 3) so the warning points to the user's code that passed non-2D
input (i.e., modify the warnings.warn(...) invocation in this validation code to
include stacklevel=<n>).

In `@python/cuml/tests/test_target_encoder.py`:
- Around line 36-38: Fix the typo in the test comment: change "tarnsform" to
"transform" above the pytest.warns block that asserts FutureWarning for
encoder.transform(df.category); update the comment text so it correctly reads
"Warns in transform".

In `@python/cuml/tests/test_tsne.py`:
- Around line 231-233: The current test_components_exception uses np.array([[]])
which is empty and can raise for zero features instead of testing
TSNE(n_components=3).fit; update the test to pass a minimally valid 2D X with at
least one sample and fewer features than n_components (e.g., shape (1,1) or
(2,1)) so the ValueError originates from the n_components check in TSNE.fit, and
(optionally) assert the raised exception message mentions n_components to ensure
the correct error path in TSNE.fit is exercised.

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: 5febae2b-ff2c-488a-ab2c-b760fb4a38b6

📥 Commits

Reviewing files that changed from the base of the PR and between abc17be and 10ce094.

📒 Files selected for processing (12)

• python/cuml/cuml/internals/outputs.py
• python/cuml/cuml/internals/validation.py
• python/cuml/cuml/preprocessing/TargetEncoder.py
• python/cuml/cuml/preprocessing/label.py
• python/cuml/cuml_accel_tests/upstream/scikit-learn/xfail-list.yaml
• python/cuml/tests/test_coordinate_descent.py
• python/cuml/tests/test_dbscan.py
• python/cuml/tests/test_label_binarizer.py
• python/cuml/tests/test_label_encoder.py
• python/cuml/tests/test_target_encoder.py
• python/cuml/tests/test_tsne.py
• python/cuml/tests/test_validation.py

[jcrist]

/merge