Fixes issue#39

[Leona-LYT]

Description

Fix IPython crash caused by pybind11 C++ object exposure and unsafe NumPy views in plqERM_Ridge.fit().

When running in IPython, clf.fit(...) implicitly returns self, which triggers sklearn’s _repr_mimebundle_. This introspects estimator attributes and attempts to access self.opt_result_, a pybind11-wrapped C++ object (rehline_result) without a proper __repr__, causing crashes.

Additionally, attributes such as coef_, _Lambda, _Gamma, and _xi were NumPy views into Eigen memory owned by the C++ object, creating potential dangling reference issues.

Fix

• Remove storage of raw C++ object (self.opt_result_)
• Replace Eigen-backed NumPy views with .copy() to ensure safe ownership
• Convert iterable outputs to Python-native types

Changes

• self.coef_ = result.beta.copy()
• self._Lambda = result.Lambda.copy()
• self._Gamma = result.Gamma.copy()
• self._xi = result.xi.copy()
• self.n_iter_ = result.niter
• self.dual_obj_ = list(result.dual_objfns)
• self.primal_obj_ = list(result.primal_objfns)
• Removed: self.opt_result_ = result

Impact

• Prevents IPython/Jupyter crashes
• Eliminates unsafe memory references
• Aligns with sklearn estimator design expectations

---

Related Issue

Closes #39

---

Type of Change

• Bug fix (non-breaking change that fixes an issue)

---

Checklist

• I have tested my changes locally
• Tests pass: pytest tests/ -v
• Code follows the style guidelines (PEP 8)
• Documentation has been updated (if applicable)
• Commits are properly formatted

---

Additional Notes

Workaround prior to this fix:

clf.fit(X=X, y=y);  # suppress IPython auto-display

[codecov]

Welcome to Codecov 🎉

Once you merge this PR into your default branch, you're all set! Codecov will compare coverage reports and display results in all future pull requests.

ℹ️ You can also turn on project coverage checks and project coverage reporting on Pull Request comment

Thanks for integrating Codecov - We've got you covered ☂️