v0.1

v0.1.0

Released August 27, 2022

Changes

  • First beta release!

  • Fixed PyPI package metadata.

Contributors

v0.1.1

Released September 2, 2022

Changes

  • Added support for NumPy 1.23. (#414)

  • Added seed keyword argument to random_prime(). (#409)

    >>> galois.random_prime(100, seed=1)
2218840874040723579228056294021
>>> galois.random_prime(100, seed=1)
2218840874040723579228056294021

  • Deployed documentation to https://mhostetter.github.io/galois/latest/ with GitHub Pages. (#408)

Contributors

v0.1.2

Released November 9, 2022

Changes

  • Fixed major inefficiency when dividing an array by a scalar or smaller (broadcasted) array. (#429)

    In [1]: import galois

In [2]: GF = galois.GF(31**5)

In [3]: x = GF.Random(10_000, seed=1); x
Out[3]:
GF([13546990, 14653018, 21619804, ..., 15507037, 24669161, 19116362],
   order=31^5)

In [4]: y = GF.Random(1, seed=2); y
Out[4]: GF([23979074], order=31^5)

# v0.1.1
In [5]: %timeit x / y
261 ms ± 5.67 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)

# v0.1.2
In [5]: %timeit x / y
8.23 ms ± 51 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)

  • Optimized lagrange_poly() by adding a custom JIT-compilable routine. (#432)

    In [1]: import galois

In [2]: GF = galois.GF(13693)

In [3]: x = GF.Random(100, seed=1)

In [4]: y = GF.Random(100, seed=2)

# v0.1.1
In [5]: %timeit galois.lagrange_poly(x, y)
2.85 s ± 3.25 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)

# v0.1.2
In [5]: %timeit galois.lagrange_poly(x, y)
4.77 ms ± 190 µs per loop (mean ± std. dev. of 7 runs, 1 loop each)

  • Added ability in FieldArray.row_reduce() to solve for an identity matrix on the right side of a matrix using the eye keyword argument. (#426)

    >>> import galois
>>> GF = galois.GF(31)
>>> A = GF([[16, 12, 1, 25], [1, 10, 27, 29], [1, 0, 3, 19]])
>>> A.row_reduce()
GF([[ 1,  0,  0, 11],
    [ 0,  1,  0,  7],
    [ 0,  0,  1, 13]], order=31)
>>> A.row_reduce(eye="right")
GF([[ 5,  1,  0,  0],
    [27,  0,  1,  0],
    [17,  0,  0,  1]], order=31)

  • Removed comma separators in FieldArray.__str__() to be consistent with NumPy’s use of str() and repr(). (#432)

    >>> import galois
>>> GF = galois.GF(3**5, display="power")
>>> x = GF.Random((3, 4), seed=1)
>>> x
GF([[α^185, α^193,  α^49, α^231],
    [ α^81,  α^60,   α^5,  α^41],
    [ α^50, α^161, α^151, α^171]], order=3^5)
>>> print(x)
[[α^185 α^193  α^49 α^231]
 [ α^81  α^60   α^5  α^41]
 [ α^50 α^161 α^151 α^171]]

  • Modernized type annotations to use abbreviated notation. For example, a | b instead of Union[a, b]. (#418)

  • Added Self type annotation where appropriate. (#420)

  • Updated documentation and improved examples. (#424, #430)

Contributors