r-scikit-learn 0.1.0__tar.gz → 0.1.1__tar.gz

r_scikit_learn-0.1.1/CHANGELOG.md

@@ -0,0 +1,23 @@
+# Changelog
+
+All notable changes to r-scikit-learn are documented here. Release tags and
+published package versions are immutable.
+
+## Unreleased
+
+## 0.1.1 - 2026-06-15
+
+- Added wheel and source-distribution installation testing across supported
+  operating systems and Python versions.
+- Added a numerical-safety fallback for ill-conditioned tall least-squares
+  problems.
+- Added TestPyPI, cross-platform benchmark, and immutable manual release
+  workflows.
+
+## 0.1.0
+
+- Added Rust-powered preprocessing, categorical encoding, sparse
+  infrastructure, composition, metrics, model selection, and linear models.
+- Added Linux, macOS, and Windows wheel builds for Python 3.10 through 3.13.
+- Added Rust-native tall-matrix least squares and multinomial logistic
+  optimization.

{r_scikit_learn-0.1.0 → r_scikit_learn-0.1.1}/Cargo.lock

@@ -998,7 +998,7 @@ dependencies = [
 
 [[package]]
 name = "r-scikit-learn-core"
-version = "0.1.0"
+version = "0.1.1"
 dependencies = [
  "faer",
  "nalgebra",

{r_scikit_learn-0.1.0 → r_scikit_learn-0.1.1}/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "r-scikit-learn-core"
-version = "0.1.0"
+version = "0.1.1"
 edition = "2021"
 license = "MIT"
 description = "Rust computational core for r-scikit-learn"
@@ -9,6 +9,7 @@ repository = "https://github.com/rishib42/r-scikit-learn"
 include = [
   "/Cargo.lock",
   "/Cargo.toml",
+  "/CHANGELOG.md",
   "/LICENSE",
   "/README.md",
   "/benches/*.py",

{r_scikit_learn-0.1.0 → r_scikit_learn-0.1.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: r-scikit-learn
-Version: 0.1.0
+Version: 0.1.1
 Classifier: Development Status :: 3 - Alpha
 Classifier: License :: OSI Approved :: MIT License
 Classifier: Programming Language :: Python :: 3
@@ -12,6 +12,7 @@ Classifier: Programming Language :: Rust
 Classifier: Typing :: Typed
 Requires-Dist: numpy>=1.23
 Requires-Dist: scipy>=1.10
+Requires-Dist: hypothesis>=6.100,<7 ; extra == 'dev'
 Requires-Dist: maturin>=1.9,<2.0 ; extra == 'dev'
 Requires-Dist: pytest>=8 ; extra == 'dev'
 Requires-Dist: ruff>=0.11 ; extra == 'dev'
@@ -25,6 +26,7 @@ Author: r-scikit-learn contributors
 License-Expression: MIT
 Requires-Python: >=3.10
 Description-Content-Type: text/markdown; charset=UTF-8; variant=GFM
+Project-URL: Changelog, https://github.com/rishib42/r-scikit-learn/blob/main/CHANGELOG.md
 Project-URL: Homepage, https://github.com/rishib42/r-scikit-learn
 Project-URL: Issues, https://github.com/rishib42/r-scikit-learn/issues
 Project-URL: Repository, https://github.com/rishib42/r-scikit-learn
@@ -34,7 +36,7 @@ Project-URL: Repository, https://github.com/rishib42/r-scikit-learn
 Fast, familiar machine-learning building blocks powered by safe Rust. 🦀
 
 `r-scikit-learn` combines a Rust computational core with lightweight,
-scikit-learn-style Python estimators. Version 0.1.0 includes:
+scikit-learn-style Python estimators. Version 0.1.1 includes:
 
 - Preprocessing, categorical encoding, and missing-value imputation
 - Pipelines and column transformers
@@ -327,14 +329,22 @@ Substantial numerical loops release the Python GIL.
 
 ## Release
 
-1. Run all development checks and build a release wheel.
-2. Install the wheel into a clean virtual environment and run the import smoke
-   test.
-3. Verify the distribution name on PyPI.
-4. Tag the release as `v0.1.0` and push the tag.
-5. Approve the GitHub Actions Trusted Publishing environment.
-
-The release workflow uses PyPI Trusted Publishing and contains no API token.
+1. Update the matching versions in `pyproject.toml`, `Cargo.toml`, and
+   `python/rsklearn/__init__.py`, then update `CHANGELOG.md`.
+2. Push the release commit and wait for CI, including manylinux and sdist
+   installation checks, to pass.
+3. Run the manual TestPyPI workflow and verify its distributions.
+4. Run the manual Release workflow with the version number without a `v`
+   prefix.
+5. Approve the PyPI environment if required.
+
+The release workflow refuses existing versions, installs every wheel on
+Python 3.10-3.13 across Linux, macOS, and Windows, verifies sdist installation,
+publishes through PyPI Trusted Publishing, creates the immutable GitHub tag and
+release, attaches artifacts, and verifies installation from PyPI. No API token
+is stored in the repository. Configure separate `pypi` and `testpypi` GitHub
+environments and matching Trusted Publishers for `release.yml` and
+`test-pypi.yml`, respectively.
 
 ## Roadmap
 

{r_scikit_learn-0.1.0 → r_scikit_learn-0.1.1}/README.md

@@ -3,7 +3,7 @@
 Fast, familiar machine-learning building blocks powered by safe Rust. 🦀
 
 `r-scikit-learn` combines a Rust computational core with lightweight,
-scikit-learn-style Python estimators. Version 0.1.0 includes:
+scikit-learn-style Python estimators. Version 0.1.1 includes:
 
 - Preprocessing, categorical encoding, and missing-value imputation
 - Pipelines and column transformers
@@ -296,14 +296,22 @@ Substantial numerical loops release the Python GIL.
 
 ## Release
 
-1. Run all development checks and build a release wheel.
-2. Install the wheel into a clean virtual environment and run the import smoke
-   test.
-3. Verify the distribution name on PyPI.
-4. Tag the release as `v0.1.0` and push the tag.
-5. Approve the GitHub Actions Trusted Publishing environment.
-
-The release workflow uses PyPI Trusted Publishing and contains no API token.
+1. Update the matching versions in `pyproject.toml`, `Cargo.toml`, and
+   `python/rsklearn/__init__.py`, then update `CHANGELOG.md`.
+2. Push the release commit and wait for CI, including manylinux and sdist
+   installation checks, to pass.
+3. Run the manual TestPyPI workflow and verify its distributions.
+4. Run the manual Release workflow with the version number without a `v`
+   prefix.
+5. Approve the PyPI environment if required.
+
+The release workflow refuses existing versions, installs every wheel on
+Python 3.10-3.13 across Linux, macOS, and Windows, verifies sdist installation,
+publishes through PyPI Trusted Publishing, creates the immutable GitHub tag and
+release, attaches artifacts, and verifies installation from PyPI. No API token
+is stored in the repository. Configure separate `pypi` and `testpypi` GitHub
+environments and matching Trusted Publishers for `release.yml` and
+`test-pypi.yml`, respectively.
 
 ## Roadmap
 

{r_scikit_learn-0.1.0 → r_scikit_learn-0.1.1}/benches/benchmark_linear_models.py

@@ -10,6 +10,8 @@ from collections.abc import Callable
 
 import numpy as np
 import rsklearn.linear_model as rlinear
+import scipy
+import sklearn
 import sklearn.linear_model as slinear
 from rsklearn import _core
 
@@ -65,6 +67,10 @@ def main() -> None:
         )
     print(f"Python: {sys.executable}")
     print(f"Rust extension: {_core.__file__} ({profile})")
+    print(
+        f"Dependencies: numpy {np.__version__}, scipy {scipy.__version__}, "
+        f"scikit-learn {sklearn.__version__}"
+    )
     rng = np.random.default_rng(20260614)
     X = rng.normal(size=(args.samples, args.features))
     coefficients = rng.normal(size=args.features)

{r_scikit_learn-0.1.0 → r_scikit_learn-0.1.1}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "maturin"
 
 [project]
 name = "r-scikit-learn"
-version = "0.1.0"
+version = "0.1.1"
 description = "High-performance scikit-learn-style machine learning powered by safe Rust"
 readme = "README.md"
 requires-python = ">=3.10"
@@ -26,6 +26,7 @@ dependencies = ["numpy>=1.23", "scipy>=1.10"]
 
 [project.optional-dependencies]
 dev = [
+  "hypothesis>=6.100,<7",
   "maturin>=1.9,<2.0",
   "pytest>=8",
   "ruff>=0.11",
@@ -36,6 +37,7 @@ dev = [
 Homepage = "https://github.com/rishib42/r-scikit-learn"
 Repository = "https://github.com/rishib42/r-scikit-learn"
 Issues = "https://github.com/rishib42/r-scikit-learn/issues"
+Changelog = "https://github.com/rishib42/r-scikit-learn/blob/main/CHANGELOG.md"
 
 [tool.maturin]
 python-source = "python"

{r_scikit_learn-0.1.0 → r_scikit_learn-0.1.1}/python/rsklearn/__init__.py

@@ -45,4 +45,4 @@ __all__ = [
     "make_column_transformer",
     "make_pipeline",
 ]
-__version__ = "0.1.0"
+__version__ = "0.1.1"

{r_scikit_learn-0.1.0 → r_scikit_learn-0.1.1}/python/rsklearn/linear_model/_least_squares.py

@@ -12,6 +12,26 @@ from rsklearn.base import BaseEstimator, RegressorMixin
 
 from ._base import LinearModel, validate_regression_fit
 
+# Normal equations square the condition number. This cutoff limits the
+# resulting float64 error amplification before selecting the fast Gram path.
+_GRAM_MIN_SINGULAR_RATIO = np.finfo(np.float64).eps ** 0.25
+_GRAM_RANK_RESOLUTION = np.sqrt(np.finfo(np.float64).eps)
+
+
+def _tall_solution_is_stable(singular: np.ndarray, rank: int, tolerance: float) -> bool:
+    """Return whether normal-equation accuracy is reliable for this spectrum."""
+    if rank == 0 or singular.size == 0 or not np.isfinite(singular).all():
+        return False
+    if rank < singular.size and tolerance < _GRAM_RANK_RESOLUTION:
+        return False
+    largest = singular[0]
+    smallest_retained = singular[rank - 1]
+    return (
+        largest > 0
+        and smallest_retained > 0
+        and smallest_retained / largest >= _GRAM_MIN_SINGULAR_RATIO
+    )
+
 
 def _fit_lstsq(
     X: np.ndarray,
@@ -22,7 +42,9 @@ def _fit_lstsq(
 ) -> tuple[np.ndarray, np.ndarray, int, np.ndarray]:
     """Solve unregularized least squares through a shape-aware dense backend."""
     if X.shape[0] >= 4 * X.shape[1]:
-        return _core.linear_fit_tall(X, y, weights, fit_intercept, tolerance)
+        tall_fit = _core.linear_fit_tall(X, y, weights, fit_intercept, tolerance)
+        if _tall_solution_is_stable(tall_fit[3], tall_fit[2], tolerance):
+            return tall_fit
     uniform_weights = np.all(weights == weights[0])
     if fit_intercept:
         if uniform_weights:

r_scikit_learn-0.1.1/tests/release_smoke.py

@@ -0,0 +1,28 @@
+"""Minimal installed-distribution smoke test used by release workflows."""
+
+from __future__ import annotations
+
+import numpy as np
+import rsklearn
+from rsklearn.linear_model import LinearRegression, LogisticRegression
+from rsklearn.preprocessing import OneHotEncoder, StandardScaler
+
+
+def main() -> None:
+    X = np.asarray([[1.0, 2.0], [2.0, 1.0], [3.0, 4.0], [4.0, 3.0]])
+    regression = LinearRegression().fit(X, [3.0, 3.0, 7.0, 7.0])
+    np.testing.assert_allclose(regression.predict(X), [3.0, 3.0, 7.0, 7.0])
+
+    classification = LogisticRegression(max_iter=500).fit(X, [0, 0, 1, 1])
+    np.testing.assert_array_equal(classification.predict(X), [0, 0, 1, 1])
+
+    scaled = StandardScaler().fit_transform(X)
+    np.testing.assert_allclose(scaled.mean(axis=0), 0.0, atol=1e-12)
+
+    encoded = OneHotEncoder().fit_transform([["a"], ["b"], ["a"]])
+    assert encoded.shape == (3, 2)
+    assert rsklearn.__version__
+
+
+if __name__ == "__main__":
+    main()

{r_scikit_learn-0.1.0 → r_scikit_learn-0.1.1}/tests/test_linear_model_parity.py

@@ -1,12 +1,16 @@
 import numpy as np
 import pytest
+from hypothesis import given, settings
+from hypothesis import strategies as st
 from rsklearn.linear_model import (
     ElasticNet,
     Lasso,
     LinearRegression,
     LogisticRegression,
     Ridge,
+    _least_squares,
 )
+from rsklearn.linear_model._least_squares import _tall_solution_is_stable
 from scipy import linalg
 
 sklearn_linear = pytest.importorskip("sklearn.linear_model")
@@ -60,6 +64,84 @@ def test_tall_linear_regression_matches_svd_near_rank_deficiency(perturbation):
     np.testing.assert_allclose(ours.predict(X), expected, rtol=1e-7, atol=5e-9)
 
 
+@given(
+    rows=st.integers(min_value=80, max_value=240),
+    columns=st.integers(min_value=2, max_value=12),
+    log_condition=st.floats(
+        min_value=0.0, max_value=12.0, allow_nan=False, allow_infinity=False
+    ),
+    weighted=st.booleans(),
+    fit_intercept=st.booleans(),
+)
+@settings(max_examples=40, deadline=None)
+def test_linear_regression_matches_svd_across_condition_numbers(
+    rows, columns, log_condition, weighted, fit_intercept
+):
+    rows = max(rows, 4 * columns)
+    rng = np.random.default_rng(
+        rows * 10_000 + columns * 100 + int(log_condition * 10) + int(weighted)
+    )
+    left, _ = np.linalg.qr(rng.normal(size=(rows, columns)))
+    right, _ = np.linalg.qr(rng.normal(size=(columns, columns)))
+    singular = np.geomspace(1.0, 10.0**-log_condition, columns)
+    X = np.ascontiguousarray((left * singular) @ right.T)
+    y = rng.normal(size=rows)
+    weights = rng.uniform(0.1, 2.0, size=rows) if weighted else None
+    ours = LinearRegression(tol=1e-10, fit_intercept=fit_intercept).fit(
+        X, y, sample_weight=weights
+    )
+    reference_weights = np.ones(rows) if weights is None else weights
+    if fit_intercept:
+        x_mean = np.average(X, axis=0, weights=reference_weights)
+        y_mean = np.average(y, weights=reference_weights)
+    else:
+        x_mean = np.zeros(columns)
+        y_mean = 0.0
+    root_weights = np.sqrt(reference_weights)
+    coefficients, _, _, _ = linalg.lstsq(
+        (X - x_mean) * root_weights[:, None],
+        (y - y_mean) * root_weights,
+        cond=1e-10,
+        check_finite=False,
+        lapack_driver="gelsd",
+    )
+    expected = X @ coefficients + y_mean - coefficients @ x_mean
+    np.testing.assert_allclose(ours.predict(X), expected, rtol=1e-7, atol=1e-9)
+
+
+def test_tall_solution_stability_gate_rejects_unsafe_spectra():
+    assert _tall_solution_is_stable(np.asarray([10.0, 1.0]), 2, 1e-6)
+    assert not _tall_solution_is_stable(np.asarray([10.0, 1e-10]), 2, 1e-6)
+    assert _tall_solution_is_stable(np.asarray([10.0, 1.0, 0.0]), 2, 1e-6)
+    assert not _tall_solution_is_stable(np.asarray([10.0, 1.0, 0.0]), 2, 1e-10)
+    assert not _tall_solution_is_stable(np.asarray([0.0, 0.0]), 0, 1e-6)
+
+
+@pytest.mark.parametrize(
+    "log_condition, expects_fallback", [(2.0, False), (10.0, True)]
+)
+def test_tall_linear_regression_falls_back_only_when_numerically_unsafe(
+    monkeypatch, log_condition, expects_fallback
+):
+    rng = np.random.default_rng(1234)
+    left, _ = np.linalg.qr(rng.normal(size=(1_000, 5)))
+    right, _ = np.linalg.qr(rng.normal(size=(5, 5)))
+    X = np.ascontiguousarray(
+        (left * np.geomspace(1.0, 10.0**-log_condition, 5)) @ right.T
+    )
+    original = _least_squares.linalg.lstsq
+    calls = 0
+
+    def tracked_lstsq(*args, **kwargs):
+        nonlocal calls
+        calls += 1
+        return original(*args, **kwargs)
+
+    monkeypatch.setattr(_least_squares.linalg, "lstsq", tracked_lstsq)
+    LinearRegression(tol=1e-10).fit(X, rng.normal(size=X.shape[0]))
+    assert bool(calls) is expects_fallback
+
+
 @pytest.mark.parametrize("alpha", [0.0, 0.1, 10.0])
 @pytest.mark.parametrize("fit_intercept", [True, False])
 def test_ridge_matches_scikit_learn_svd(alpha, fit_intercept):