The R Project SVN R-packages

Rev

Show changed files | Details | Compare with Previous | Blame | RSS feed

Filtering Options

Rev Age Author Path Log message Diff
8465 598 d 3 h maechler /trunk/cluster/ tweaks from mv Fortran -> C: dysta() *returns*; dysta, dysta3 no longer exported  
8457 660 d 23 h maechler /trunk/cluster/ translate last remaining Fortran dysta.f and daisy.f to *.c adapting calls,headers  
7954 1930 d 4 h maechler /trunk/cluster/src/ new standards have reserved `*_t` var names -> use `*_typ`  
7509 3042 d 0 h maechler /trunk/cluster/ (from a year ago:) clara() with Jaccard metric, pam() to follow, not quiet yet; + small things  
7331 3426 d 6 h maechler /trunk/cluster/ switch (again!) mona to use C version; catch p=1 as error; cosmetic  
7184 3755 d 0 h maechler /trunk/cluster/ finalized the `correct.d` -- bug fix to clara(); ready for release  
6872 4199 d 20 h maechler /trunk/cluster/ basically revert r6871: do pass `n` for now  
6871 4199 d 20 h maechler /trunk/cluster/src/ cl_Pam() no need to pass `n`; can get it from the object  
6869 4200 d 1 h maechler /trunk/cluster/ pam() now uses .Call() instead of .C() -- (kept old code for now)  
6801 4343 d 22 h maechler /trunk/cluster/ agnes() and diana() finally get, respectively work with a "trace.lev" option.
Further: agnes(*, method="flexible", par.method = c(a1, a2, b, c)), i.e.,
length(alpha) == 4, finally works *correctly* {confirmed with single,complete tests}
 
6057 5290 d 6 h maechler /trunk/cluster/ pam(*, pamonce=.) incorporate the faster SWAP algorithms provided by Matthias Studer  
5978 5332 d 2 h maechler /trunk/cluster/ more twins.c code cleanup; export coefHier()  
5976 5332 d 5 h maechler /trunk/cluster/ twins.f -> twins.c -- using C instead of Fortran  
5758 5580 d 20 h maechler /trunk/cluster/ make clara() more pam()-like on C level --- notably allow "pamLike=TRUE" argument  
5634 5757 d 23 h maechler /trunk/cluster/ omit the "_" in cl_<name>() for fortran for now; Enhances  
5592 5881 d 22 h maechler /trunk/cluster/ rename some compiled routines, in order to allow registration of those names  
3721 7262 d 22 h maechler /trunk/cluster/ register dysta3 as C, not Fortran; ready to release 1.11.2  
3083 7404 d 0 h maechler /trunk/cluster/ many more fanny extensions  
3053 7414 d 15 h maechler /trunk/cluster/ translated fanny.f to fanny.c (and properly register)  
3012 7455 d 5 h maechler /trunk/cluster/src/ new sildist from Romain Francois  

Show All