1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
|
2018-09-15 Martin Mächler <maechler@stat.math.ethz.ch>
* man/logm.Rd: use try(.) [for solaris where needed]
2017-04-06 Martin Maechler <maechler@stat.math.ethz.ch>
* DESCRIPTION (Version): 0.999-3
* src/expm-eigen.c (do_expm_eigen): fix possible PROTECT stack imbalance
* src/expm.c (do_expm): ditto; both thanks to T.Kalibera's `expm.so.bcheck`
2017-03-28 Martin Maechler <maechler@stat.math.ethz.ch>
* R/sqrtm.R (sqrtm): simple modification so it works for 1x1 matrix
* R/logm.Higham08.R (rootS): (ditto)
* src/init.c: "finish" registration
2016-12-13 Martin Maechler <maechler@stat.math.ethz.ch>
* DESCRIPTION (Version): 0.999-1
* src/matpow.c (matpow): use size_t etc, trying to prevent integer
overflow when n^2 > .Machine$integer.max, i.e., n >= 46341
2015-10-06 Martin Maechler <maechler@stat.math.ethz.ch>
* DESCRIPTION (Version): 0.999-0
* R/balance.R (balance): also add 'job = "N"' (i.e., 'none') option.
* src/R_dgebal.c (R_dgebal): catch +/-Inf in matrix avoiding
infinite loop in Lapack's DGEBAL(); thanks to Terry Thernau.
2014-09-11 Martin Maechler <maechler@stat.math.ethz.ch>
* NAMESPACE: no longer export logm.Higham08
* R/expm2.R (expm.AlMoHi09): new expm(x, method = "AlMohy-Hi09")
* src/matexp_MH09.c: thanks to Drew Schmidt.
2014-09-03 Martin Maechler <maechler@stat.math.ethz.ch>
* src/matpow.c: need to carefully REPROTECT x_
2011-11-08 Martin Maechler <maechler@stat.math.ethz.ch>
* man/balance.Rd: use new name balance() instead of dgebal()
* R/balance.R:
2011-10-28 Martin Maechler <maechler@stat.math.ethz.ch>
* src/matexp.f (matexpRBS): do not call STOP ..
* NAMESPACE: import rowSums from Matrix (because of its
sparseMatrix method).
* R/expm.R (expm): auto-coerce sparse matrices to dense, when
"needed".
2011-10-27 Martin Maechler <maechler@stat.math.ethz.ch>
* R/expm_vec.R (expAtv): finally add the functionality for
'exp(A*t) %*% v' Ravi had translated from EXPOKIT in August.
2011-08-27 Martin Maechler <maechler@stat.math.ethz.ch>
* R/expm.R (expm): add new method "PadeRBS" (Roger B. Sidje),
interfacing to the corresponding Fortran code.
* src/init.c, src/expm.h: also namespace-ify the Fortran symbols.
2011-08-26 Martin Maechler <maechler@stat.math.ethz.ch>
* NAMESPACE: also import solve() from Matrix, for sparse expm()
2011-03-30 Martin Maechler <maechler@stat.math.ethz.ch>
* DESCRIPTION (Version): 0.98-4
(Suggests): RColorBrewer, sfsmisc -- used in examples
* inst/doc/expm.Rnw: replace the few non-ASCII chars by LaTeX.
2010-08-12 Martin Maechler <maechler@stat.math.ethz.ch>
* R/logm.Higham08.R (logm.Higham08): force argument to Schur(.) to
be dense for now (working around infelicitous Matrix method).
(logm.Higham08): catch case of infinite loop
* man/logm.Rd: add example of the above.
2010-07-19 Martin Maechler <maechler@stat.math.ethz.ch>
* DESCRIPTION (Version): 0.98-2
* src/matpow.c (R_matpow): fix modify-argument bug in %^%.
Thanks to stackflow-user "gd047" for reporting it.
2010-07-08 Martin Maechler <maechler@stat.math.ethz.ch>
* DESCRIPTION (Version): 0.98-1, for (first!) release to CRAN
2009-06-06 Martin Maechler <maechler@stat.math.ethz.ch>
* DESCRIPTION (Version): 0.96-2
* tests/exact-ex.R: adapt last test, to also pass on 32-bit Mac OSX
2009-06-02 Martin Maechler <maechler@stat.math.ethz.ch>
* tests/exact-ex.R: move print() to see Mac OSX error
2009-03-02 Martin Maechler <maechler@stat.math.ethz.ch>
* R/sqrtm.R (sqrtm): allow negative eigenvalues, with a
message iff "verbose"; use sqrt(<complex>) in that case.
* R/logm.Higham08.R (logm.Higham08): allow negative eigenvalues,
as in sqrtm.
* R/expmCond-all.R (.expmCond.1, .expmFrechet2008.26): and others;
cosmetic (speedy!) improvements; dating of Feb.23.
2009-02-28 Martin Maechler <maechler@stat.math.ethz.ch>
* DESCRIPTION (Version, Depends): 0.96-1, Matrix
* src/logm-eigen.c (logm_eigen): *do* signal an error, not just
print; as the help page has always said.
* man/logm.Rd: comment 'order' and 'trySym' out; introduce
method = "Higham08" and make it the default (!)
* R/logm.R: ditto
* R/logm.Higham08.R: new logm.Higham08() from Michael Stadelmann's thesis.
* R/sqrtm.R: new sqrtm(), ditto
* NAMESPACE: add here.
* tests/log+sqrt.R: tests for these; notably small non-diagonalizable cases
* R/expm2.R (expm.Higham08): renamed 'expm2' to expm.Higham08
2009-02-19 Martin Maechler <maechler@stat.math.ethz.ch>
* DESCRIPTION (Version): 0.95-1
* R/expmCond-all.R: only public expmCond(*, method=.) function,
instead of expmCond1Est() etc.
* man/expmCond.Rd: ditto
* tests/expm-Cond.R: ditto
2009-02-19 Martin Maechler <maechler@stat.math.ethz.ch>
* DESCRIPTION (Maintainer): myself, for the moment; need to get
E-mail feedback from win-builder.
* tests/expm-Cond.R: new tests for expmCond*()
* tests/exact-ex.R: factor the utilities out into new file; add expm2()
* demo/exact-fn.R: containing only (function + data) definitions
2009-02-18 Martin Maechler <maechler@stat.math.ethz.ch>
* R/expmCond-all.R (expmFrechet): method "blockEnlarge", also for testing
* man/expmFrechet.Rd: the default method
2009-02-17 Martin Maechler <maechler@stat.math.ethz.ch>
* DESCRIPTION (Version): 0.95-0
* R/expm2.R (expm2): new algorithms from Michael Stadelmann's
Master thesis work
* R/expmCond-all.R (expmCond, ..): Exponential Condition Number estimation
* man/expm2.Rd, man/expmCond.Rd, man/expmFrechet.Rd: docu
* NAMESPACE: export new functions
2009-01-30 Martin Maechler <maechler@stat.math.ethz.ch>
* src/R_dgebal.c (R_dgebal): fix to work also for integer matrix()
* src/expm.c (do_expm),
* src/expm-eigen.c (do_expm_eigen): ditto
* tests/bal-ex.R: and test for it.
2008-05-13 Christophe Dutang <christophe.dutang@ensimag.fr>
* Use Lapack function zgecon in "hybrid_Eigen_Ward" to test the
singularity.
* Start the matrix logarithm logm.
2008-03-20 Vincent Goulet <vincent.goulet@act.ulaval.ca>
* New function matpow() as alias to operator %^%.
* R/matpow.R: miscellaneous fixes and prettyfication of comments
2008-03-01 Martin Maechler <maechler@stat.math.ethz.ch>
* R/expm.R (expm): rename "Eigen" to "R_Eigen"
* src/expm.c (expm): get "1bal" case right.
* tests/compare-bal-expm.R: minimal testing of "2bal" vs "1bal"
2008-02-29 Martin Maechler <maechler@stat.math.ethz.ch>
* src/expm.c (expm): second argument 'precond_kind'
* src/expm.h: update for new second argument
* src/expm-eigen.h: implement workaround for Lapack.h's zlange() typo
2008-02-27 Martin Maechler <maechler@stat.math.ethz.ch>
* DESCRIPTION (Version): 0.9-1 -- new version for the occasion
* src/expm.c (expm): HOORAY !! --- fixed the octave bug:
using simpler __ and correct __ code for back-permuting
* R/expm.R (expm): add method = "R_Ward77"
which works *correctly* !
2008-02-25 Martin Maechler <maechler@stat.math.ethz.ch>
* R/matpower.R ("%^%"): add R interface to
* src/matpow.c
2008-02-23 Martin Maechler <maechler@stat.math.ethz.ch>
* R/expm.R (expm): add methods from David Firth's "mexp"
and my modifications, of Fortran code in
* src/mexp-common.f
* src/matrexp.f
* src/matrexpO.f
|
