We are excited to announce that linear-algebra has reached a Stable Beta state. This version is designed for reliability and is suitable for industrial use and testing.

We have conducted a thorough audit of the codebase, ensuring high test coverage (106 tests across both mutable and immutable modules) and correcting inconsistencies between documentation and implementation.

• Mutable & Immutable Support: Full suites for both Matrix and Vector types.
• Advanced Operations: Includes determinant, inverse, rank, eigenvalues (QR algorithm), and more.
• Zero-Cost Abstractions: Efficient Transpose views that allow operations without deep copying.
• Correctness First: Rigorous testing including edge cases (empty matrices, single-element vectors, etc.).

Comprehensive API documentation is available and up-to-date with the latest implementation. You can access it at mooncakes.io or browse the doc/ directory in this repository.

We provide documentation in multiple languages to support our global community:

• 🇺🇸 English (doc/en_US)
• 🇨🇳 简体中文 (doc/zh_CN)
• 🇯🇵 日本語 (doc/ja_JP)

• Renaming:
  • map_row() / map_col() -> map_row_inplace() / map_col_inplace() for clarity.
  • eachij() -> each_row_col()
• New Features:
  • Vector::dot added for dot product calculations.
  • Transpose view now supports direct indexing (t[row][col]).
• Fixes:
  • Corrected edge case behavior for 0x0 matrices (determinant is now 1).