Apply new input validation to metrics.trustworthiness

[csadorf]

Applies new input validation system to metrics.trustworthiness.

Part of #7998

[copy-pr-bot]

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 3e37655e-ae75-4dee-9da2-06f05661ac7d

📥 Commits

Reviewing files that changed from the base of the PR and between 1678d35 and 8c1b111.

📒 Files selected for processing (2)

• python/cuml/cuml/metrics/trustworthiness.pyx
• python/cuml/tests/test_trustworthiness.py

---

📝 Walkthrough

Summary by CodeRabbit

• Bug Fixes

  • Improved input validation and clearer error reporting for invalid sample counts, neighbor constraints, unsupported metrics, and dimensionality.

• Changes

  • Default behavior now converts input dtypes automatically, reducing dtype-related errors for the trustworthiness metric.

• Tests

  • Expanded coverage with targeted tests for invalid neighbors, mismatched sample counts, dtype conversion behavior, 1D inputs, and unknown metrics.

Walkthrough

The PR refactors the trustworthiness() function to replace input_to_cuml_array with check_array/check_consistent_length, extracts device pointers from validated arrays, changes convert_dtype default to True, adjusts neighbor validation to use validated n_samples, updates the backend call, and adds tests covering invalid n_neighbors, mismatched rows, dtype conversion, 1D inputs, and unknown metrics.

Changes

Trustworthiness Input Validation Refactor

Layer / File(s)	Summary
Imports & Documentation python/cuml/cuml/metrics/trustworthiness.pyx	Replace input_to_cuml_array import with check_array and check_consistent_length; update docstring to set convert_dtype=True default.
Input Validation & Pointer Setup python/cuml/cuml/metrics/trustworthiness.pyx	Validate/convert X and X_embedded to contiguous np.float32, derive n_samples, n_features, n_components, enforce consistent sample counts, and extract device pointers d_X_ptr, d_X_embedded_ptr.
Neighbor Bounds & Backend Call python/cuml/cuml/metrics/trustworthiness.pyx	Use validated n_samples for n_neighbors bounds check; call trustworthiness_score with extracted device pointers and adjusted dimensional args; keep handle.sync().
Tests / Validation Coverage python/cuml/tests/test_trustworthiness.py	Replace generic invalid-input test with parameterized test_trustworthiness_invalid_n_neighbors; add tests for mismatched rows, convert_dtype=False rejecting float64, 1D input rejection, and unsupported metric handling.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~22 minutes

Possibly related PRs

• rapidsai/cuml#8057: Performs the same input_to_cuml_array → check_array refactor for a different metric (kl_divergence).
• rapidsai/cuml#8043: Migrates array validation to check_array/check_consistent_length and updates pointer extraction patterns across metrics.
• rapidsai/cuml#8059: Similar metrics migration applying check_array-based validation and .data.ptr extraction in related metric code.

Suggested labels

Cython / Python

Suggested reviewers

• jcrist
• betatim

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title accurately and concisely describes the main change: applying new input validation to the metrics.trustworthiness function.
Description check	✅ Passed	The description is directly related to the changeset, explaining that it applies the new input validation system to metrics.trustworthiness as part of issue #7998.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

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

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

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

[coderabbitai]

Actionable comments posted: 2

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@python/cuml/cuml/metrics/trustworthiness.pyx`:
- Around line 83-104: The code currently calls check_array on X and X_embedded
before validating the requested metric; move or add an early validation of the
metric (e.g., in the trustworthiness function) to raise a clear ValueError for
unsupported metric values before performing any check_array conversions so the
heavy allocations for X_m and X_m2 (and their data.ptr conversions d_X_ptr /
d_X_embedded_ptr) are avoided on invalid input; ensure the validation references
the same metric parameter name used in the function signature and matches the
allowed metric set.
- Around line 105-106: Update the n_neighbors validation in the trustworthiness
implementation so it matches scikit-learn: in the trustworthiness function (in
python/cuml/cuml/metrics/trustworthiness.pyx) replace the current check that
compares n_neighbors to n_samples with a check that raises a ValueError when
n_neighbors >= n_samples / 2 (i.e., require n_neighbors < n_samples / 2); keep
the error message clear and reference n_samples and n_neighbors for context.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 3273cf85-51fc-4220-9130-d5693a42f877

📥 Commits

Reviewing files that changed from the base of the PR and between b51c429 and 1678d35.

📒 Files selected for processing (2)

• python/cuml/cuml/metrics/trustworthiness.pyx
• python/cuml/tests/test_trustworthiness.py

[jcrist]

[jcrist]

/merge