Passing GSoC Midterms
Hey everyone! Weeks 7 and 8 are done, and the biggest news first: I passed the midterm evaluation🎉. Half of GSoC is behind me now. These two weeks were less about writing new classes and more about going back and making the ones I already had a lot stricter. Here is what happened. PR #1920: The RatioEstimatorBuilder This PR adds the builder for the NRE family, so NRE_A , NRE_B , NRE_C and BNRE…
The GSoC midterm evaluations have concluded, and the highlights are positive. The reporter has successfully passed the midterm assessment, marking a significant milestone halfway through the program. The last two weeks have focused on enhancing existing classes rather than developing entirely new ones.
PR #1920 introduces a builder for the NRE family, unifying the interfaces of NRE_A, NRE_B, NRE_C, and BNRE. This PR also includes improvements to the base class, which was deemed an opportune time to address validation gaps. By implementing these enhancements early, future builders will benefit from the improvements without requiring extensive retrofits.
Improvements include raising errors for invalid literal values during object construction and invalid model-incompatible kwargs, providing more immediate feedback to users. Additionally, configs are now immutable, eliminating issues arising from attempts to modify them after initialization. A custom __repr__ was also added, offering a concise representation of the config that omits irrelevant fields.
The PR also addresses the renaming of the z-score parameters across different families, improving clarity and consistency. Furthermore, the reporter introduces a shift in workflow for the second half of GSoC. Instead of directly implementing designs provided by their mentor, Jan Teusen, the reporter will now write short design proposals beforehand. This change aims to reduce the risk of design mistakes and streamline the review process.
Despite a slower pace in week 8 due to other commitments, the reporter remains on track, outlining plans for the upcoming weeks. The next step involves addressing the hardening review for PR #1920 and commencing work on the vector field estimator design proposal.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.