matrices 2.0.0

An easy-to-use, high-performance matrix computation and linear algebra library with built-in parallel support.

2.0.0

• Redesign the package around high performance and script-friendly ergonomics, with Float32 matrix and vector types.
• Replace nested-list internals with contiguous typed-data storage and pure Dart SIMD kernels for dense matrix and vector hot paths.
• Add broad linear algebra coverage, including direct solvers, decompositions, eigen/SVD, pseudoinverse, least squares, PCA, iterative solvers, and Krylov methods.
• Add CSR sparse matrix support with dense conversion, sparse-vector products, sparse-dense multiplication, transpose, and JSON serialization.
• Align the 32-bit and 64-bit APIs across construction, arithmetic, statistics, serialization, solvers, decompositions, advanced algorithms, and sparse conversion.
• Add reproducible correctness, numerical robustness, performance regression, and benchmark-suite coverage.

1.2.7

• Documentation update.

1.2.6

• Bug fixes.

1.2.4

• Bug fixes.

1.2.3

• Bug fixes.

1.2.2

• Bug fixes.
• Add documentation webpage.

1.2.1

• Bug fixes.
• Add documentation webpage.

1.2.0

• Bug fixes.
• Major update; see the documentation.

1.1.5

• Bug fixes.
• Add rowCount, columnCount, and count.
• Add max and min.

1.1.4

• Bug fixes.
• Add fromFlattenedList.
• Add identity matrix support.

1.1.3

• Bug fixes.

1.1.2

• Bug fixes.
• Improve inverse.
• Standardize methods.

1.1.0

• Bug fixes.
• Remove emoji methods.
• Standardize methods.

1.0.11

• Bug fixes.
• Greatly improve determinant performance.

1.0.9

• Bug fixes.

1.0.8

• Update.

1.0.7

• Documentation update.

1.0.6

• Documentation update.

1.0.5

• Add English documentation.

1.0.4

• Documentation update.

1.0.3

• Bug fixes.

1.0.2

• Open source release.