summaryrefslogtreecommitdiff
path: root/CHANGES.md
diff options
context:
space:
mode:
Diffstat (limited to 'CHANGES.md')
-rw-r--r--CHANGES.md136
1 files changed, 136 insertions, 0 deletions
diff --git a/CHANGES.md b/CHANGES.md
new file mode 100644
index 0000000..bb1045f
--- /dev/null
+++ b/CHANGES.md
@@ -0,0 +1,136 @@
10.11.2.0
2========
3
4- geigSH' (symmetric generalized eigensystem)
5
6- mapVectorWithIndex
7
8
90.11.1.0
10========
11
12- exported Mul
13
14- mapMatrixWithIndex{,M,M_}
15
160.11.0.0
17========
18
19- flag -fvector default = True
20
21- invlndet (inverse and log of determinant)
22
23- step, cond
24
25- find
26
27- assoc, accum
28
290.10.0.0
30========
31
32- Module reorganization
33
34- Support for Float and Complex Float elements (excluding LAPACK computations)
35
36- Binary instances for Vector and Matrix
37
38- optimiseMult
39
40- mapVectorM, mapVectorWithIndexM, unzipVectorWith, and related functions.
41
42- diagRect admits diagonal vectors of any length without producing an error,
43 and takes an additional argument for the off-diagonal elements.
44
45- different signatures in some functions
46
470.9.3.0
48=======
49
50- flag -fvector to optionally use Data.Vector.Storable.Vector
51 without any conversion.
52
53- Simpler module structure.
54
55- toBlocks, toBlocksEvery
56
57- cholSolve, mbCholSH
58
59- GSL Nonlinear Least-Squares fitting using Levenberg-Marquardt.
60
61- GSL special functions moved to separate package hmatrix-special.
62
63- Added offset of Vector, allowing fast, noncopy subVector (slice).
64 Vector is now identical to Roman Leshchinskiy's Data.Vector.Storable.Vector,
65 so we can convert from/to them in O(1).
66
67- Removed Data.Packed.Convert, see examples/vector.hs
68
690.8.3.0
70=======
71
72- odeSolve
73
74- Matrix arithmetic automatically replicates matrix with single row/column
75
76- latexFormat, dispcf
77
780.8.2.0
79=======
80
81- fromRows/fromColumns now automatically expand vectors of dim 1
82 to match the common dimension.
83 fromBlocks also replicates single row/column matrices.
84 Previously all dimensions had to be exactly the same.
85
86- display utilities: dispf, disps, vecdisp
87
88- scalar
89
90- minimizeV, minimizeVD, using Vector instead of lists.
91
920.8.1.0
93=======
94
95- runBenchmarks
96
970.8.0.0
98=======
99
100- singularValues, fullSVD, thinSVD, compactSVD, leftSV, rightSV
101 and complete interface to [d|z]gesdd.
102 Algorithms based on the SVD of large matrices can now be
103 significantly faster.
104
105- eigenvalues, eigenvaluesSH
106
107- linearSolveLS, rq
108
1090.7.2.0
110=======
111
112- ranksv
113
1140.7.1.0
115=======
116
117- buildVector/buildMatrix
118
119- removed NFData instances
120
1210.6.0.0
122=======
123
124- added randomVector, gaussianSample, uniformSample, meanCov
125
126- added rankSVD, nullspaceSVD
127
128- rank, nullspacePrec, and economy svd defined in terms of ranksvd.
129
130- economy svd now admits zero rank matrices and return a "degenerate
131 rank 1" decomposition with zero singular value.
132
133- added NFData instances for Matrix and Vector.
134
135- liftVector, liftVector2 replaced by mapVector, zipVector.
136