ml_linalg 10.0.4

SIMD-based linear algebra (1 operation on 4 float32 values, 1 operation on 2 float64 values)

Changelog

10.0.4

• pubspec.yaml: sdk version constraint changed from >=2.3.0 to >=2.2.0

10.0.3

• Float32Matrix.columns: empty list supported as a source
• Float32Matrix.rows: empty list supported as a source

10.0.2

• Float32Matrix.fromList: empty list supported as a source

10.0.1

• Links in README corrected

10.0.0

• Vector:
  • removed possibility to mutate an instance of the Vector:
    • isMutable flag removed
    • []= operator removed
• Matrix:
  • removed possibility to mutate an instance of the Matrix:
    • setColumn method removed
  • insertColumns method added
  • Matrix.from renamed to Matrix.fromList
  • Matrix.fromFlattened renamed to Matrix.fromFlattenedList
  • benchmarks added for checking matrix initialization performance
• Performance enhancements
• Dart 2.3.0 supported

9.0.0

• Vector:
  • Distance enum added
  • distanceTo method refactored
  • cosine distance calculation added

8.0.0

• Matrix:
  • rows constructor renamed to fromRows
  • columns constructor renamed to fromColumns
  • flattened constructor renamed to fromFlattened
  • rows getter added
  • columns getter added

7.0.0

• Matrix: ZRange support (instead of the library built-in Range)

6.1.1

• Vector: unique method refactored

6.1.0

• Vector:
  • normalize method added to interface and supported in Float32x4Vector
  • rescale method added to interface and supported in Float32x4Vector
• VectorBase: cache for popular operations implemented

6.0.2

• Matrix: MatrixMixin corrected

6.0.1

• Vector class refactored (get rid of redundant mixins)

6.0.0

• prefix ML removed from entities' names
• Float32x4Vector: equality operator override added
• Matrix: uniqueRows method added

5.5.1

• MLMatrix: fixed bug in setColumn method when rows cache was not cleared

5.5.0

• MLMatrix: setColumn method implemented

5.4.0

• MLMatrix: dummy for setColumn method added

5.3.0

• MLMatrix: frobenius norm calculation added

5.2.0

• MLMatrix: / operator added
• MLMatrix: rowsMap method added
• MLMatrix: columnsMap method added

5.1.0

• max and min methods added for matrix

5.0.1

• Travis integration added
• dartfmt task added

5.0.0

• MLVector and MLMatrix now don't have generic type parameters

4.2.0

• Static factories converted into abstract factories

4.1.0

• toString method specified for matrix mixin
• examples for vector operations fixed

4.0.0

• Vector type removed (there are no longer column and row vectors)
• Matrix's method getColumnVector renamed to getColumn
• Matrix method getRowVector renamed to getRow
• Public api documentation for MLMatrix<E> added

3.5.0

• Mutable vectors supported in matrix

3.4.0

• Add possibility to create mutable vectors
• Add support for value assignment via []= operator

3.0.3

• readme contacts section updated
• build_runner dependency updated

3.0.2

• readme badge corrected

3.0.1

• readme updated

3.0.0

• vectorizedMap vector's method improved: batch boundary indexes may be passed into a mapper function
• columnsMap and rowsMap matrix's method combined into one method - vectorizedMap
• Public factories Float32x4VectorFactory and Float32x4MatrixFactory renamed into Float32x4Vector and Float32x4Matrix
• copy vector's method removed

2.3.0

• pick method added to matrix api: it's possible now to combine a new matrix from different segments of a source matrix

2.2.2

• README.md updated (table of content)

2.2.1

• Travis integration added

2.2.0

• Support matrices in vector operations

2.1.0

• Column and row vectors added

2.0.0

• Unnecessary generic type argument removed from MLMatrix class
• Matrix logic split into separate mixins

1.3.0

• MLVectorMixin added, benchmark directory reorganized

1.2.0

• Map functions added to matrix

1.1.0

• Float32x4MatrixFactory extended

1.0.1

• Readme updated

1.0.0

• Library public release