dwww Home | Show directory contents | Find package

2012-01-07  Douglas Bates <bates@stat.wisc.edu>

        * man/GHrule.Rd, man/NelderMead-class.Rd, man/NelderMead.Rd,
          man/golden-class.Rd, man/golden.Rd, man/lmer.Rd, man/mkdevfun.Rd:
          Documentation updates.
        * R/lmer.R, R/profile.R, tests/lmer-1.R: Nlmer now working (for nAGQ=1).
          Updated tests and got profile working again.
        * R/AllClass.R, R/lmer.R, src/external.cpp, src/optimizer.h: Allowed
          setting options in the NelderMead optimizer. Switched so that this is
          the default in glmer.
        * R/AllClass.R, R/lmer.R, src/external.cpp, src/predModule.cpp: More
          work on nlmer.

2011-12-31  Douglas Bates <bates@stat.wisc.edu>

        * src/predModule.cpp: Add an explicit copy of the mapped sparse matrix
          to a sparse matrix in updateL

2011-12-28  Douglas Bates <bates@stat.wisc.edu>

        * R/lmer.R: Change maximum iterations error to a warning.
        * DESCRIPTION: New version number.
        * R/lmer.R: Allowed choice of optimizer in glmer at least.
        * R/AllClass.R: Extended and exported the NelderMead class
        * src/external.cpp, src/optimizer.cpp, src/optimizer.h: Removed
          debugging code, added methods to NelderMead to set convergence
          characteristics and exported same.

2011-12-22  Douglas Bates <bates@stat.wisc.edu>

        * NAMESPACE, R/AllClass.R, src/external.cpp, src/optimizer.cpp,
          src/optimizer.h: Tuned the Nelder-Mead optimizer a bit and got rid of
          some of the noise.
        * R/lmer.R, man/lmer.Rd, src/external.cpp: Incorporated aGQ code in the
          glmer function itself. Eliminated the doFit argument which is
          redundant when devFunOnly is also used.

2011-12-21  Douglas Bates <bates@stat.wisc.edu>

        * src/external.cpp: Added the glmerAGQ function and did some cleanup.
        * src/predModule.cpp: Delete some debugging code.
        * R/lmer.R: Updates in glmer, aGQ not yet added but soon will be.
        * R/GHrule.R: Rename the knots on the z scale as 'z', not 'k'
        * NAMESPACE: Export the golden class and class generator object

2011-12-16  Douglas Bates <bates@stat.wisc.edu>

        * src/respModule.cpp, src/respModule.h: Clean up declarations by
          declaring a typedef for Eigen::Map<Eigen::VectorXd>

2011-12-15  Douglas Bates <bates@stat.wisc.edu>

        * R/lmer.R: Adopt new version of the merPredD$new function that uses S.
        * R/GHrule.R: Added an R function, based on SparseGrid:::GQN to evaluate
          a matrix of nodes, weights and log-density for Gauss-Hermite
          quadrature against a Gaussian density.
        * R/AllClass.R, src/optimizer.h: Added a reverse-communication golden
          search algorithm class

2011-12-12  Douglas Bates <bates@stat.wisc.edu>

        * src/external.cpp: Fat-finger error. I meant to commit this file and
          not glmFamily.h
        * src/glmFamily.h, src/optimizer.cpp, src/optimizer.h: Added simple
          optimizers (Golden search for bounded 1-dimensional and Nelder-Mead
          for multidimensional).

2011-12-07  Douglas Bates <bates@stat.wisc.edu>

        * R/AllClass.R, R/lmer.R, src/external.cpp, src/predModule.cpp,
          src/predModule.h: Allow regeneration of pointers in merPredD objects.

2011-12-06  Douglas Bates <bates@stat.wisc.edu>

        * R/AllClass.R, R/lmer.R, src/external.cpp, src/predModule.cpp,
          src/predModule.h: Add many more fields to the merPredD class and allow
          for serialize/unserialize. Lower the default tolPwrss as per testing
          by Ben.

2011-12-05  Douglas Bates <bates@stat.wisc.edu>

        * DESCRIPTION, R/AllClass.R, R/lmer.R, man/lmer.Rd, man/mkdevfun.Rd,
          man/refitML.Rd, src/external.cpp, src/predModule.cpp,
          src/predModule.h, src/respModule.cpp, src/respModule.h: Use mapped
          objects in C++ structures so that serialize/unserialize will work.

2011-11-29  Douglas Bates <bates@stat.wisc.edu>

        * R/profile.R: More fixing of labels. Fix up strip names to
          reflect sigma, not log(sigma).

2011-11-29  Douglas Bates <bates@stat.wisc.edu>

        * R/profile.R: Profiling now works for lmer models with variance
          components only, including the case of a single fixed-effects
          parameter. Labeling of some plots needs to be corrected.
        * src/predModule.cpp, src/predModule.h: Modifications for the 0-column X
          matrix case used in profiling.
        * src/Gauss_Hermite.cpp, src/Gauss_Hermite.h, src/external.cpp: Move
          Gaussian quadrature knot/weight calculations to the Gqr package.

2011-11-28  Douglas Bates <bates@stat.wisc.edu>

        * R/profile.R: Got profile working for cases where the number of
          fixed-effects is greater than 1. Need a "do nothing gracefully" clause
          for p == 1.
        * R/lmer.R: Use the value of emptyenv(), not the funtion itself. Add ...
          to refitML generic and method
        * R/lmer.R: Use separate mkdevfun in glmer. Switch devFunOnly to an
          integer value where 1 and 2 refer to different stages.

2011-11-17  Douglas Bates <bates@stat.wisc.edu>

        * R/AllClass.R: Correct and extend the allInfo method for the glmResp
          class.
        * ToDo: Need to come up with a crafty way of creating a copy of the
          reference class object with the external pointer. It may work to just
          copy it then set the PTR to be a NULL pointer.

2011-11-16  Douglas Bates <bates@stat.wisc.edu>

        * src/Gauss_Hermite.cpp, src/Gauss_Hermite.h, src/external.cpp: Add
          another method of creating Gaussian quadrature rules. Export the GHQ
          class.
        * tests/lmer-1.R: Comment out enough tests to make R CMD check succeed.
        * R/lmer.R: Clean up some functions and methods that were causing
          problems with the tests.
        * man/glmFamily-class.Rd, man/glmFamily.Rd, man/lmResp-class.Rd,
          man/lmResp.Rd: Document the class generator object and the reference
          class (for now).
        * NAMESPACE: Remove redundant S3 method vcov.summary.mer

2011-11-04  Douglas Bates <bates@stat.wisc.edu>

        * src/external.cpp: Correct the argument counts for externally visible
          glmFamily accessors
        * man/sleepstudy.Rd: disable a test - need to check what rcond(fm@re)
          did in lme4a
        * man/reTrms.Rd: No longer used
        * man/profile-methods.Rd: Don't run examples until the code has been
          fixed
        * man/merPredD-class.Rd, man/merPredD.Rd: Try to document reference
          class generator object this way
        * inst/unitTests/runit.compDev.R, inst/unitTests/runit.link.R,
          inst/unitTests/runit.lmerResp.R: update tests (compDev is inert at
          present but could be activated for glmer tests)

2011-11-02  Douglas Bates <bates@stat.wisc.edu>

        * DESCRIPTION: Bump the version number.
        * NAMESPACE, man/glmFamily-class.Rd: Export and document the glmFamily
          reference class and generator object
        * src/Makevars, src/Makevars.win: Add a Makevars.win file. Make the
          factory-fresh setting of CPPFLAGS suppress warnings.
        * man/lmerMod-class.Rd, man/merMod-class.Rd, man/merPredD-class.Rd:
          incorporate documentation for [gn]lmerMod-class in the merMod-class
          docs; document merPredD-class
        * inst/doc/PLSvGLS.Rnw, inst/doc/PLSvGLS.pdf, inst/doc/Theory.Rnw,
          inst/doc/Theory.pdf, inst/doc/lme4.bib, vignettes,
          vignettes/PLSvGLS.Rnw, vignettes/PLSvGLS.pdf, vignettes/Theory.Rnw,
          vignettes/Theory.pdf, vignettes/lme4.bib: Move vignette sources to a
          vignettes directory per recommendations for R-2.14.0

2011-10-18  Douglas Bates <bates@stat.wisc.edu>

        * man/mkdevfun.Rd: Cleanup.
        * NAMESPACE: Export S3 method for devcomp.
        * man/lmResp-class.Rd: Document response reference classes and
          constructor objects
        * man/chmSp.Rd, man/deFeMod.Rd, man/feModule-class.Rd, man/glmFamily.Rd,
          man/glmerResp.Rd, man/lmerResp-class.Rd, man/lmerResp.Rd,
          man/reModule-class.Rd, man/reModule.Rd: Remove documentation of
          no-longer-used classes from Rcpp modules
        * man/cbpp.Rd: From R-2.14.0 on getCall is imported from the stats
          package.
        * R/lmer.R: Getter and setter for REML field in lmerResp should be reml,
          not REML.

2011-10-17  Douglas Bates <bates@stat.wisc.edu>

        * src/respModule.cpp, src/respModule.h: Change working weights to be a
          matrix.
        * src/predModule.cpp: Thinko regarding 0-based indices.

2011-10-14  Douglas Bates <bates@stat.wisc.edu>

        * DESCRIPTION, NAMESPACE, R/AllClass.R, R/lmer.R, src/external.cpp,
          src/predModule.cpp, src/predModule.h, src/respModule.cpp,
          src/respModule.h: nlmer added. Increment calculation for PWRSS works
          but not yet incorporated in R code.

2011-10-03  Douglas Bates <bates@stat.wisc.edu>

        * R/lmer.R: Small but important change in checking for object weights in
          environment rho in mkRespMod2

2011-09-30  Douglas Bates <bates@stat.wisc.edu>

        * R/AllClass.R, R/lmer.R: Use inheritance on reference classes, being
          careful about package installation. Create a cleaner version of
          mkRespMod2.
        * NAMESPACE: export more classes
        * DESCRIPTION: new version and date

2011-09-29  Douglas Bates <bates@stat.wisc.edu>

        * R/lmer.R: Modify mkRespMod for nlmer response.
        * src/external.cpp: Add isNullExtPtr .Call'able function for Ptr/ptr
          formulation.
        * NAMESPACE, R/AllClass.R: new organization of response classes. Use
          Ptr/ptr formulation to generate external pointers.

2011-09-27  Douglas Bates <bates@stat.wisc.edu>

        * R/lmer.R: More components/slots/etc. available in getME()
        * R/AllClass.R, R/lmer.R: Reinstitute the Gp slot in merMod objects

2011-09-23  Martin Maechler  <maechler@stat.math.ethz.ch>

        * R/lmer.R, man/ranef.Rd: *must* use lme4Eigen::: inside devFun()
          [evaluated from C++]
        * man/refitML.Rd: add doc. + example
        * man/Pastes.Rd, man/Penicillin.Rd: reactivate the image(L) plots

2011-09-22  Martin Maechler  <maechler@stat.math.ethz.ch>

        * NAMESPACE, R/AllClass.R, R/lmer.R: tolPwrss - for now

2011-09-22  Douglas Bates <bates@stat.wisc.edu>

        * data/VerbAgg.rda: Correct labels on Gender, use camelCase consistently
          on item labels.
        * data/sleepstudy.rda: switch to bzip2 compression for smaller file
          size.

2011-09-21  Douglas Bates <bates@stat.wisc.edu>

        * src/external.cpp: Added a function to check on the SIMD instruction
          sets in use by Eigen.

2011-09-20  Douglas Bates <bates@stat.wisc.edu>

        * R/lmer.R: Do the same number of pwrssUpdate calls with compDev=TRUE or
          FALSE
        * src/external.cpp, src/predModule.cpp, src/predModule.h: Remove
          debugging code. solve and solveU methods for merPredD class now return
          the numerator of the convergence criterion.
        * R/lmer.R: Need to pass the pointer, not the reference class object.
          Modifications to glmer, pwrssUpdate, etc. to get glmer working.

2011-09-20  Martin Maechler  <maechler@stat.math.ethz.ch>

        * R/lmer.R, man/lmer.Rd, src/external.cpp: non-hardcorded PWRSS
          tolerance

2011-09-19  Douglas Bates <bates@stat.wisc.edu>

        * NAMESPACE, R/AllClass.R, R/lmer.R, src/external.cpp, src/glmFamily.h,
          src/respModule.cpp: created a glmFamily reference class and allowed
          for compiled version of glmer update using working residuals and
          weights (not currently working properly, use compDev=FALSE argument to
          glmer)
        * DESCRIPTION: new version and date
        * src/external.cpp, src/predModule.cpp, src/predModule.h,
          src/respModule.cpp, src/respModule.h: Moved all externally .Call'able
          function definitions to external.cpp. external.h and init.cpp are no
          longer needed.

2011-09-19  Douglas Bates <bates@stat.wisc.edu>

        * R/lmer.R: Clean up printMer and summary.merMod
        * src/external.cpp, src/external.h, src/lmer.cpp, src/lmer.h: rename
          lmer.[h,cpp] to external.[h,cpp]
        * R/lmer.R: Initial iterations in glmer (the nAGQ=0L part) now working.
          Later iterations still need work.

2011-09-16  Douglas Bates <bates@stat.wisc.edu>

        * R/lmer.R: Use fac argument in call to sqrL() method. glmer now
          iterates but does not yet converge.
        * src/predModule.cpp: Add the - u_i on the rhs of the equation in the
          solve method.
        * R/AllClass.R, src/init.cpp, src/predModule.cpp, src/predModule.h: New
          'allInfo' method for glmerResp reference class. Added .Call'able
          extractors for Utr and Vtr to merPredD class.

2011-09-14  Douglas Bates <bates@stat.wisc.edu>

        * R/lmer.R: Modified glmer to return a meaningful object when
          doFit=FALSE.
        * R/AllClass.R, src/init.cpp, src/respModule.cpp, src/respModule.h:
          Added methods and new classes for response modules. Modified names of
          .Call'able functions to make them easier to understand.
        * src/predModule.cpp: Use d_LamtUt instead of recomputing d_Lambdat *
          d_Ut
        * inst/unitTests/runit.link.R: Modified link/muEta/variance unit tests
          to use a glmerResp object.

2011-09-13  Douglas Bates <bates@stat.wisc.edu>

        * src/Makevars: Add more explanation about the -DNDEBUG compiler flag
          and when to use it.
        * R/AllClass.R, R/lmer.R, src/init.cpp, src/predModule.cpp,
          src/predModule.h, src/respModule.cpp, src/respModule.h: Many more
          methods added to the reference class definitions, as well as
          documentation. glmer is now working in the sense that it doesn't
          segfault, however it is not yet performing the calculations correctly.
        * DESCRIPTION: Remove RcppModule specification.
2011-09-12  Douglas Bates <bates@stat.wisc.edu>

        * src/respModule.h: Clean up calculation of working residuals.
        * R/lmer.R: Still working on glmer, not there yet.
        * R/AllClass.R, src/init.cpp, src/predModule.cpp, src/predModule.h,
          src/respModule.cpp, src/respModule.h: Added many more R-callable entry
          points in respModule and predModule, incorporating them as methods in
          the reference class definitions.
        * src/Makevars: Allow for suppression of assert statements with -DNDEBUG
          (R CMD check complains)

2011-09-12  Martin Maechler  <maechler@stat.math.ethz.ch>

        * DESCRIPTION, NAMESPACE, R/AllClass.R, R/lmer.R,
          inst/unitTests/runTests.R, tests/doRunit.R, tests/drop.R,
          tests/extras.R, tests/lmer-1.Rout.save, tests/lmer.R,
          tests/lmer2_ex.R, tests/nlmer-conv.R, tests/nlmer.R, tests/throw.R:
          more updates
        * man/Dyestuff.Rd, man/VarCorr.Rd, man/cbpp.Rd, man/getME.Rd,
          man/lmer.Rd, man/lmerMod-class.Rd, man/merMod-class.Rd,
          man/mkdevfun.Rd, man/ranef.Rd, man/reModule-class.Rd: considerably
          more documentation

2011-09-11  Douglas Bates <bates@stat.wisc.edu>

        * NAMESPACE, R/AllClass.R, R/lmer.R, src/init.cpp, src/respModule.cpp,
          src/respModule.h: Added glmerResp class and began glmer function.
          Still needs work.

2011-09-11  Douglas Bates <bates@stat.wisc.edu>

        * src/predModule.[h,cpp]: Wrote native C++ L() method. Somehow calling
          ::M_chm_factor_as_SEXP was messing up memory.
        * tests/vcov-etc.R, tests/lmer-conv.R: Test using this package not lme4a
        * R/AllClass.R, src/init.cpp, src/predModule.cpp, src/predModule.h: Add
          capability of extracting L from a merPredD object

2011-09-10  Martin Maechler  <maechler@stat.math.ethz.ch>

        * DESCRIPTION, NAMESPACE, R/lmer.R, man/Pastes.Rd, man/Penicillin.Rd,
          man/bootMer.Rd, man/getL.Rd, man/getME.Rd, man/merMod-class.Rd:
          getME() for all three(!) versions of lme4 -- deprecate getL() where it
          exists

2011-09-02  Douglas Bates <bates@stat.wisc.edu>

        * src/init.cpp, src/predModule.cpp, src/predModule.h: Drop the check for
          diagonal Lambda. Cache the value of LamtUt and update it in place with
          new method updateLamtUt.

2011-08-31  Douglas Bates <bates@stat.wisc.edu>

        * src/init.cpp, src/predModule.cpp, src/predModule.h: Finally got a
          work-around for the "pruning of sparse product" problem. Also added a
          diagonalLambda method.
        * R/AllClass.R: Add a "fitted" method to the lmerResp reference class.
          Re-arrange code.

2011-08-30  Douglas Bates <bates@stat.wisc.edu>

        * src/predModule.cpp, src/predModule.h: Drop some debugging code.
          Isolate the calculation of LamtUt to try to find out what goes wrong
          in there.
        * R/lmer.R: Update the ranef method for merPredD storing Lambdat, not
          Lambda

2011-08-29  Douglas Bates <bates@stat.wisc.edu>

        * R/AllGeneric.R, R/utilities.R, src/init.cpp, src/predModule.cpp,
          src/respModule.cpp, src/respModule.h: Code cleanup - remove exception
          declarations, use specific 'using' statements within blocks instead of
          'using namespace' statements.

2011-08-28  Douglas Bates <bates@stat.wisc.edu>

        * NAMESPACE, src/eigen.h, src/glmFamily.cpp, src/glmFamily.h,
          src/lmer.cpp, src/predModule.cpp, src/predModule.h,
          src/respModule.cpp, src/respModule.h: Initialize d_delb and d_delu in
          predModule to zero (not doing so was causing hard-to-debug errors).
          Remove exception declarations as suggested for Rcpp. Make declaration
          of short names for Eigen classes and constants namespace-specific.

2011-08-11  Douglas Bates <bates@stat.wisc.edu>

        * src/predModule.cpp: Using solveInPlace to try to avoid memory
          problems.

2011-07-31  Douglas Bates <bates@stat.wisc.edu>

        * src/lmer.cpp, src/predModule.cpp, src/predModule.h: Trying to track
          dowm memory problem in the evaluation of delu. It looks like it is in
          the CholmodSupport.h file in Eigen but my intended fix apparently
          didn't succeed.
        * R/AllClass.R: Modified according to John Chambers' suggestions.

2011-07-29  Douglas Bates <bates@stat.wisc.edu>

        * src/respModule.cpp, src/respModule.h: Flailing around trying to find
          the source of the memory problems. valgrind claims there is an
          uninitialized value being used in a log call within the Laplace method
          but I can't find it.
        * src/predModule.h: Remove unneeded template keywords (caught by an old
          compiler).
        * R/AllClass.R: Realized that the initialize method should return an
          object.
        * R/AllClass.R, R/lmer.R, src/init.cpp, src/predModule.cpp,
          src/predModule.h: Switch to Lambdat and Zt and the CHOLMOD
          factorization. Special code for diagonal Lambda.
        * src/predModule.cpp: One more attempt. Still failing on the boundary
          when the number of nonzeros changes.

2011-07-28  Douglas Bates <bates@stat.wisc.edu>

        * src/init.cpp, src/predModule.cpp, src/predModule.h: First shot at a
          version with the Eigen/Matrix/CHOLMOD links. Compiles and loads but
          not yet tested. Checking in so I can access it from home.
        * src/init.cpp, src/lmer.cpp, src/predModule.cpp, src/predModule.h: More
          getter methods defined in C++ code and reference class. Remove some
          debugging code.
        * R/AllClass.R, R/AllGeneric.R, R/lmList.R, R/lmer.R: More changes to
          use S3 methods when dispatching on the first argument only.
        * DESCRIPTION, NAMESPACE: Change dependencies and imports.

2011-07-26  Douglas Bates <bates@stat.wisc.edu>

        * * initialize this archive.

Generated by dwww version 1.15 on Thu May 23 12:40:27 CEST 2024.