SCM

SCM Repository

[matrix] Diff of /pkg/ChangeLog
ViewVC logotype

Diff of /pkg/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 1213, Tue Feb 7 23:44:07 2006 UTC revision 1420, Tue Aug 22 22:49:40 2006 UTC
# Line 1  Line 1 
1    2006-08-22  Douglas Bates  <bates@R-project.org>
2    
3            * src/Mutils.c (dup_mMatrix_as_dgeMatrix): updated for general
4            types of classed Matrices, matrices or numeric or logical vectors.
5    
6            * src/init.c: register dup_mMatrix_as_dgeMatrix for coercions.
7    
8    2006-08-21  Douglas Bates  <bates@R-project.org>
9    
10            * src/lmer.c (internal_mer_RZXinv): Fix memory leak caught by
11            valgrind.
12            * tests/matprod.R: Add tests to verify that 'solve' and '%*%' are
13            inverses.
14            * src/sparseQR.c (sparseQR_validate): add new validation test, fix
15            -Wall warnings.
16            * src/dppMatrix.c,dtrMatrix.c,dgCMatrix.c,dgeMatrix.c,dspMatrix.c:
17            Use dup_mMatrix_as_dgeMatrix.
18    
19    2006-08-20  Douglas Bates  <bates@R-project.org>
20    
21            * src/sparseQR.c: Fix thinko in sparseQR_resid_fitted.
22            * tests/sparseQR.R: Added
23            * man/sparseQR-class.Rd: Document methods for qr.* generics
24            * R/sparseQR.R: Return correct # of coefs; fix cut-and-paste errors
25    
26    2006-08-19  Douglas Bates  <bates@R-project.org>
27    
28            * NAMESPACE, R/sparseQR.R, src/init.c,sparseQR.[ch],SOURCES_C.mkf:
29            Added methods for sparseQR for qr.qy, qr.qty, qr.coef, qr.resid
30            and qr.fitted.
31            * src/Mutils.[ch]: Added dup_mMatrix_as_dgeMatrix utility
32            * src/dgCMatrix.c: Check for ordering before storing q in dgCMatrix_QR
33    
34    2006-08-18  Martin Maechler  <maechler@stat.math.ethz.ch>
35    
36            * R/AllGeneric.R: add "qr" and "chol" generics; via 'trick' since
37            the base version has no "..." argument.
38    
39            * R/sparseMatrix.R (prSpMatrix): fix the triangular unit diagonal case.
40    
41            * R/Matrix.R: define and
42            * NAMESPACE: export as.numeric() and as.logical() methods.
43                    Let's hope these do not badly slow down something...
44            Finally export the zapsmall() method.
45    
46    2006-08-17  Douglas Bates  <bates@R-project.org>
47    
48            * src/dgCMatrix.[ch] (dgCMatrix_matrix_solve), src/init.c,
49            R/dgCMatrix.R: solve methods for dgCMatrix and dense RHS.
50            * src/dtCMatrix.c :Remove code that is no longer used
51            * R/dtCMatrix.R: Use C code for diagU2N in CsparseMatrix classes
52    
53    2006-08-14  Douglas Bates  <bates@R-project.org>
54    
55            * src/Csparse.[ch],init.c (Csparse_to_logical): Added utilities
56            Csparse_to_logical and Csparse_symmetric_to_general.
57            * R/dgCMatrix.R,dsCMatrix.R,Csparse.R : Consolidate general
58            coercion methods between CsparseMatrix and TsparseMatrix.
59    
60    2006-08-14  Douglas Bates  <bates@R-project.org>
61    
62            * R/dtCMatrix.R,dgCMatrix.R,Csparse.R src/init.c,dgCMatrix.[ch],
63            Csparse.[ch],chm_common.c,dense.c : Use CHOLMOD code and methods
64            for CsparseMatrix, TsparseMatrix, denseMatrix or matrix when
65            possible.
66    
67    2006-08-12  Douglas Bates  <bates@R-project.org>
68    
69            * src/chm_common.[ch],Csparse.[ch],Tsparse.[ch],CHMfactor.c,dtTMatrix.c,
70              dgTMatrix.c,dgCMatrix.c,dsCMatrix.c,dtTMatrix.c, R/dsTMatrix.R,
71              Tsparse.R,Csparse.R,dgTMatrix.R,dsCMatrix.R,lsCMatrix.R:
72              generalize conversions between TsparseMatrix, CsparseMatrix,
73              denseMatrix and matrix.  Preserve triangularity property and
74              propagate Dimnames in the easy cases.
75    
76    2006-08-10  Douglas Bates  <bates@R-project.org>
77    
78            * src/lmer.c: adjust checks in glmer_init for S4SXP
79            * tests/validObj.R: check of all(eq) should be all(eq@x) - worked
80            before S4SXP but for the wrong reasons.
81    
82    2006-08-08  Douglas Bates  <bates@R-project.org>
83    
84            * src/Csparse.c (Csparse_crossprod): tcrossprod result has stype = -1.
85            Later modified to return the upper triangle only.
86            * R/dgTMatrix.R: Remove vestigial crossprod and tcrossprod methods.
87    
88    2006-08-07  Douglas Bates  <bates@R-project.org>
89    
90            * src/Csparse.c (Csparse_crossprod): Set stype on result to avoid
91            R-level conversion to a sparse symmetric class.
92            * R/Tsparse.R,R/Csparse.R (crossprod and tcrossprod methods):
93            Remove conversion to sparse symmetric classes (now done in C code).
94    
95    2006-08-07  Martin Maechler  <maechler@stat.math.ethz.ch>
96    
97            * R/dgCMatrix.R: disable old crossprod and tcrossprod methods
98            * man/band.Rd:  adapt the \dontshow{} unit test.
99    
100    2006-08-07  Martin Maechler  <maechler@stat.math.ethz.ch>
101    
102            * DESCRIPTION (Version): 0.995-14  for CRAN
103    
104            * tests/other-pkgs.R: oops: library(Matrix) outside *if* !
105    
106            * R/sparseMatrix.R (Tsp2grNEL): fixed Tsparse -> graph coercion
107            * tests/other-pkgs.R:  more checks, sparseMatrix -> graph
108    
109            * R/Auxiliaries.R (as_Tsparse, as_Rsparse): new
110            * R/Tsparse.R (tril, triu, band): methods
111            * R/dgRMatrix.R (tril, ...):    ditto
112            * man/band.Rd:                  ditto
113    
114    2006-08-04  Martin Maechler  <maechler@stat.math.ethz.ch>
115    
116            * R/Matrix.R (head, tail): directly use utils:::head.matrix
117                    to be up-to-date automatically.
118    
119    2006-08-03  Martin Maechler  <maechler@stat.math.ethz.ch>
120    
121            * DESCRIPTION (Version): 0.995-13 to be released to CRAN
122    
123            * DESCRIPTION (Lazydata): no, instead of 'yes' because it fails for:
124            * data/CAex.R, inst/external/CAex_slots.rda: replacing data/CAex.rda
125            * data/KNex.R, inst/external/KNex_slots.rda: replacing data/KNex.rda
126            such that the S4 objects are always created by the current version
127            of R and 'Matrix' class definitions.
128    
129    2006-08-01  Douglas Bates  <bates@R-project.org>
130    
131            * R/lmer.R (LMEoptimize method), tests/lmer.R,
132            inst/external/test3comp.rda: Added warnings for convergence on
133            boundary and test cases.
134    
135            * src/lmer.c (mer_postVar): Modified to return the variances that
136            are marginal to the fixed effects, not conditional on them (which
137            is what the bVar slot contents represent).
138    
139    2006-07-31  Douglas Bates  <bates@R-project.org>
140    
141            * NAMESPACE, src/lmer.c, R/lmer.R (hatTrace):  Add the hatTrace
142            function which calls the C function mer_hat_trace2.
143    
144            * man/ranef.Rd: Include description of "postVar" argument and
145            producing a caterpillar plot.
146    
147    2006-07-31  Martin Maechler  <maechler@stat.math.ethz.ch>
148    
149            * NAMESPACE:    change  "correlation" to "corMatrix"
150            * R/AllClass.R: to avoid clash with S3 class in 'nlme'.
151            * R/dpoMatrix.R: ditto
152            * R/lmer.R:      ditto
153    
154    2006-07-28  Douglas Bates  <bates@R-project.org>
155    
156            * src/lmer.c (internal_mer_RZXinv): Split the calculation of the
157            RZXinv slot's contents into a separate internal function that can
158            be used in mer_hat_trace.
159    
160    2006-07-22  Martin Maechler  <maechler@stat.math.ethz.ch>
161    
162            * R/Matrix.R: Coercions "Matrix" -> (sparse|dense)Matrix
163            via new smart
164            * R/Auxiliaries.R (as_Csparse, as_dense, .M.shapse): new
165            utility functions.
166    
167    
168    2006-07-21  Martin Maechler  <maechler@stat.math.ethz.ch>
169    
170            * R/Csparse.R (tril, triu, band): do return *triangular* classed
171            matrices when appropriate; band() even symmetric ones.
172            (replCmat): extend to potentially all "CsparseMatrix"
173            * R/Tsparse.R (replTmat): extend to all "TsparseMatrix"; hence
174            allow subassignment for special sparse matrices.
175    
176            * R/Auxiliaries.R (as_geClass): factor out the  .M.kind() functionality
177    
178            * src/lmer.c (mer_MCMCsamp, glmer_MCMCsamp): new 'verbose'
179            argument; in glmer_*(): print only if(verbose).
180    
181            speed-optimize a few places by moving REAL(.) out of loops.
182    
183            * src/lmer.h, src/init.h, R/lmer.R:  related to above.
184    
185    2006-07-20  Martin Maechler  <maechler@stat.math.ethz.ch>
186    
187            * R/Matrix.R("["): disable ("Matrix", i = "logical", j = "missing"),
188                    since that wrongly triggers also for M[ logi , ]
189    
190            * R/denseMatrix.R: "[" methods now also work e.g. when indexing
191            a symmetric matrix that results in a non-symmetric one.
192            * R/Auxiliaries.R (as_geClass): new function used in "[" above.
193    
194            * R/dMatrix.R: make round(M) work as round(M, 0)
195    
196            * R/dgTMatrix.R (image): coordinate system and axis now use
197            1-based indices, not 0-based ones.
198    
199            * R/Tsparse.R (.ind.prep for "["): get rid of max(<empty>) warning.
200            * tests/indexing.R: test it.
201    
202            * NAMESPACE: export isSymmetric(); has been a generic in "base" for a while;
203            * man/isSymmetric-methods.Rd: and document it.
204    
205            * R/SparseM-conv.R: added coercion methods for some 'SparseM' matrices.
206            * man/SparseM-conv.Rd: docu them
207    
208            * tests/other-pkgs.R: renamed from tests/graph.R and add example
209                    for 'SparseM' conversions
210    
211    2006-07-17  Douglas Bates  <Douglas.Bates@R-project.org>
212    
213            * R/Matrix.R (head): added head() and tail() methods.
214    
215    2006-07-17  Martin Maechler  <maechler@stat.math.ethz.ch>
216    
217            * DESCRIPTION (Version): 0.995-12 released to CRAN
218    
219    2006-07-15  Martin Maechler  <maechler@stat.math.ethz.ch>
220    
221            * tests/simple.R: add check for correct dsT -> dgT coercion;
222                    add check for correct printing of symmetric sparse matrices.
223    
224            * R/Auxiliaries.R (non0ind): return *all* non-0 entry indices also
225            for sparse symmetric matrices.
226    
227            * src/dsTMatrix.c (dsTMatrix_as_dgTMatrix): do not copy the
228            diagonal twice.
229    
230    2006-07-11  Douglas Bates  <Douglas.Bates@R-project.org>
231    
232            * src/dsTMatrix.c (dsTMatrix_as_dgTMatrix): Fix a case of INTEGER
233            being applied to the x slot (detected by Brian Ripley).
234    
235    2006-07-10  Martin Maechler  <maechler@stat.math.ethz.ch>
236    
237            * src/dgCMatrix.c (dgCMatrix_validate): 'p' slot must have correct length.
238    
239            * R/Auxiliaries.R (isTriC): fix buglet (we were 1-based!)
240    
241    2006-07-08  Martin Maechler  <maechler@stat.math.ethz.ch>
242    
243            * src/lgCMatrix.c (lgCMatrix_diag): new function
244            * R/lgCMatrix.R (diag): for new method
245    
246            * R/AllClass.R (TsparseMatrix): do use Tsparse_validate
247                 ==> construction of illegal "*gTMatrix" via new() should
248                 now be much less easy:
249            * tests/Class+Meth.R: assertError() for some illegal "dgT*"
250    
251            * R/Matrix.R (Matrix): Matrix(0, nrow,ncol) now "goes sparse"
252                    directly.
253            * man/Matrix.Rd: documents it.
254    
255    2006-07-06  Douglas Bates  <bates@R-project.org>
256    
257            * src/pedigree.c (pedigree_inbreeding): Correction in
258            initialization.  This function is not currently being used and is
259            not fully tested.
260    
261            * NAMESPACE, R/{AllClass.R,lmer.R}, src/{init.c,lmer.c}:
262            Introduced the glmer class.  Added code for mcmcmsamp on glmer
263            objects.  Modified validity check on pedigree objects to account
264            for nonparallel patterns of missingness of parents.
265    
266            * man/{lmer-class.Rd,mcmcsamp.Rd}: Update documentation for glmer
267            class.
268    
269    2006-07-01  Martin Maechler  <maechler@stat.math.ethz.ch>
270    
271            * R/pMatrix.R: coercion pMatrix -> Tsparse
272    
273    2006-06-12  Douglas Bates  <bates@R-project.org>
274    
275            * DESCRIPTION (Version): 0.995-11 released to CRAN
276    
277            * R/lmer.R (mcmcsamp method): Corrected arrangments of names on
278            the output from mcmcsamp.
279    
280    2006-06-10  Douglas Bates  <bates@R-project.org>
281    
282            * R/lmer.R (simulestimate): added C code for calculating the trace
283            of the hat matrix.
284    
285    2006-06-09  Martin Maechler  <maechler@stat.math.ethz.ch>
286    
287            * R/diagMatrix.R (setAs): define coercion methods to sparse matrix
288            classes.
289            * R/sparseMatrix.R etc: multiplication of diagonal and sparse
290    
291    2006-06-08  Martin Maechler  <maechler@stat.math.ethz.ch>
292    
293            * R/dgTMatrix.R (colSums): etc. All four of
294            colSums(), rowSums(), colMeans(), rowMeans() now should work for
295            all "Matrices".
296    
297    2006-06-01  Douglas Bates  <bates@stat.wisc.edu>
298    
299            * R/lmer.R (panel.ci): Add a reference line at zero and a
300            background grid to the qqmath plot of ranef.lmer.
301    
302    2006-05-30  Douglas Bates  <bates@stat.wisc.edu>
303    
304            * R/lmer.R (expandSlash): Functions (non-exported) to allow nested
305            grouping factors to be specified in the formula as (1|foo/bar).
306    
307    2006-05-27  Douglas Bates  <bates@stat.wisc.edu>
308    
309            * R/lmer.R (findbars and others): Change check of is.numeric to
310            !is.language to resolve the bug reported by Jacob Wegelin.
311    
312            * src/pedigree.c (pedigree_inbreeding): Initial implementation of
313            code to evaluate inbreeding coefficients without calculating T,
314            based on code in Sargolzaei and Iwaisaki's paper.
315    
316    2006-05-27  Douglas Bates  <bates@stat.wisc.edu>
317    
318            * R/{lmer.R,AllGeneric.R}, src/{init.c,lmer.[ch]}: Added local
319            generic and mer methods for isNested and denomDF. This denomDF was
320            an attempt to emulate that in lme but I don't think that makes
321            sense.  Use the trace of the hat matrix instead.
322    
323    2006-05-17  Martin Maechler  <maechler@stat.math.ethz.ch>
324    
325            * R/sparseMatrix.R: Matrix <-> graph methods: can no longer use
326            the C code depending on a slot structure that's no longer valid.
327            * src/dgTMatrix.c: ditto (also: src/init.c src/dgTMatrix.h)
328    
329    
330    2006-05-17  Douglas Bates  <bates@stat.wisc.edu>
331    
332            *  R/{AllGeneric.R,Csparse.R},man/band.Rd,NAMESPACE: changed name
333            of lowerTriMatrix generic and methods to tril (also upper to triu)
334            and added a general band extractor.
335    
336    2006-05-16  Douglas Bates  <bates@stat.wisc.edu>
337    
338            * R/pedigree.R (pedigree): Replace sire and dam values outside the
339            allowable range with NAs. Added a corresponding check in the
340            validity check for the pedigree class.
341    
342            * R/[CT]sparse.R ([t]crossprod): The result of single-argument
343            crossprod methods now inherits from symmetricMatrix.
344    
345    2006-05-15  Douglas Bates  <bates@stat.wisc.edu>
346    
347            * R/AllGeneric.R (lowerTriMatrix): Added (but did not export)
348            generics lowerTriMatrix and upperTriMatrix along with methods for
349            the Csparse virtual class.  Also added a C function Csparse_band
350            that implements these methods by calling cholmod_band.
351    
352    2006-05-15  Martin Maechler  <maechler@stat.math.ethz.ch>
353    
354            * R/Tsparse.R ("["): column or row subsetting; @Dimnames[k] got
355            erased when it was NULL.  This led to invalid subselections!
356    
357    2006-04-25  Douglas Bates  <bates@stat.wisc.edu>
358    
359            * R/dtCMatrix.R: avoid coercion of dtCMatrix object to dgCMatrix
360            in method for "t" so as not to lose the unit diagonal property.
361    
362    2006-04-19  Douglas Bates  <bates@stat.wisc.edu>
363    
364            * R/lmer.R, R/AllGeneric.R, NAMESPACE: Remove the postVar generic
365            and methods.  This is now an option to the ranef method for the
366            mer class.
367    
368            * src/cs_utils.c: Ensure that the nz component is -1 for a
369            compressed column-oriented matrix.  Minor formatting cleanup.
370    
371            * man/lmer-class.Rd: Document the qqmath method for ranef.lmer
372            objects.
373    
374    2006-04-19  Martin Maechler  <maechler@stat.math.ethz.ch>
375    
376            * R/Auxiliaries.R (diagU2N): new for the solve() methods in
377            * R/dtCMatrix.R: where dgC -> dgT coercion now preserves diag = "U".
378    
379    2006-04-15  Douglas Bates  <bates@stat.wisc.edu>
380    
381            * src/cs.[ch],src/cs_utils.[ch] : Added Tim Davis' CSparse library
382            in cs.[ch] and utilities to interface to that code in cs_utils.[ch].
383    
384            * R/dtCMatrix.R, src/dtCMatrix.[ch] : CSparse-based solve methods
385            for the dtCMatrix class.
386    
387    2006-04-12  Douglas Bates  <bates@stat.wisc.edu>
388    
389            * R/pedigree.R, R/AllClass.R, NAMESPACE: added a pedigree class
390            and methods for it.
391    
392    2006-04-12  Martin Maechler  <maechler@stat.math.ethz.ch>
393    
394            * R/dgCMatrix.R: add  storage.mode(.) <- "double" for "matrix"
395            arguments, such that M %*% 1:6 now works
396            * Tests/matprod.R: test the above
397    
398    2006-04-03  Douglas Bates  <bates@stat.wisc.edu>
399    
400            * R/lmer.R (qqmath,ranef.lmer-method): added a qqmath method for
401            the ranef.lmer class.
402    
403            * R/AllClass.R, NAMESPACE, R/lmer.R: Added a postVar generic and
404            methods to extract the posterior variances from the bVar slot.
405    
406    2006-03-30  Martin Maechler  <maechler@stat.math.ethz.ch>
407    
408            * R/dtCMatrix.R: allow coercion from  dgC* to triangular (dtC*)
409            * R/dsCMatrix.R: and symmetric (dsC*) 'Csparse' matrices.
410    
411            * R/Tsparse.R: Tsparse* -> Csparse* coercion now works and is tested in
412            * tests/simple: (extended)
413    
414            * R/sparseMatrix.R (isTriangular): now using much improved
415            * R/Auxiliaries.R (isTriC): new triangularity check for *CMatrix
416    
417    2006-03-23  Douglas Bates  <bates@stat.wisc.edu>
418    
419            * src/dsyMatrix.c (dsyMatrix_as_dspMatrix): Propagate DimNames
420            (problem report from Franklin Parlamis).  This should be done
421            generally.
422    
423    2006-03-21  Douglas Bates  <bates@stat.wisc.edu>
424    
425            * R/AllClass.R,lmer.R: Change the name of the lmer.ranef class to
426            ranef.lmer (like summary.lmer).  Add the coef.lmer class and
427            update the plot methods.
428    
429    2006-03-20  Douglas Bates  <bates@stat.wisc.edu>
430    
431            * R/lmer.R (resid and ranef methods):  Added methods for the
432            "residuals" and "resid" generic but only for linear mixed model
433            fits.  Changed the ranef method to return a list of data frames so
434            that the plot methods now work.
435    
436    2006-03-16  Douglas Bates  <bates@bates2>
437    
438            * src/dpoMatrix.c (dpoMatrix_chol): Require n > 0 in call to
439            dpotrf - otherwise the BLAS on Mac OS X complains and quits.
440    
441            * DESCRIPTION (Date): New release
442    
443    2006-03-15  Martin Maechler  <maechler@stat.math.ethz.ch>
444    
445            * DESCRIPTION (Version): 0.995-6  -- to be released to CRAN
446    
447            * data/KNex.rda: replacing 'mm' and 'y' by KNex <- list(mm=mm, y=y)
448            * man/KNex.Rd: and other help files
449            * tests/*.R: several ones needed adaption
450            * inst/doc/Comparisons.Rnw: ditto
451    
452    2006-03-11  Martin Maechler  <maechler@stat.math.ethz.ch>
453    
454            * R/dgCMatrix.R (replCmat): "[<-" methods for dgCMatrix
455    
456            * tests/indexing.R: tests for new [<- methods for sparse matrices.
457    
458    2006-03-10  Martin Maechler  <maechler@stat.math.ethz.ch>
459    
460            * R/dgTMatrix.R (replTmat): "[<-" methods for dgTMatrix
461    
462            * R/Tsparse.R (.ind.prep): fix out-of-range indexing
463    
464    2006-03-08  Martin Maechler  <maechler@stat.math.ethz.ch>
465    
466            * R/dMatrix.R: enable things like  M [ M < 10 ]
467            * R/dgeMatrix.R: implement it
468    
469    2006-03-06  Martin Maechler  <maechler@stat.math.ethz.ch>
470    
471            * R/AllClass.R: define "summary.mer" and "*.lmer"
472            * R/lmer.R (summary): summary(<mer>) computes & returns the above;
473            * R/lmer.R (show): now works with summary()
474    
475    2006-03-04  Martin Maechler  <maechler@stat.math.ethz.ch>
476    
477            * R/dgCMatrix.R: finally direct "Arith"  dgC o dgC
478            * R/Auxiliaries.R (WhichintersectInd): and other utilities to
479                    support the above
480    
481  2006-02-07  Douglas Bates  <bates@stat.wisc.edu>  2006-02-07  Douglas Bates  <bates@stat.wisc.edu>
482    
483          * R/lmer.R (lmer): fix initial values of offset and weights for glm.fit.          * R/lmer.R (lmer): fix initial values of offset and weights for glm.fit.
# Line 171  Line 651 
651  2005-08-26  Martin Maechler  <maechler@stat.math.ethz.ch>  2005-08-26  Martin Maechler  <maechler@stat.math.ethz.ch>
652    
653          * R/Matrix.R:  added first "[<-" methods; not yet functional          * R/Matrix.R:  added first "[<-" methods; not yet functional
654          * R/denseMatrix.R: dito          * R/denseMatrix.R: ditto
655          * man/Subassign-methods.Rd: new help file for these          * man/Subassign-methods.Rd: new help file for these
656    
657  2005-08-25  Martin Maechler  <maechler@stat.math.ethz.ch>  2005-08-25  Martin Maechler  <maechler@stat.math.ethz.ch>

Legend:
Removed from v.1213  
changed lines
  Added in v.1420

root@r-forge.r-project.org
ViewVC Help
Powered by ViewVC 1.0.0  
Thanks to:
Vienna University of Economics and Business Powered By FusionForge