[mlpack] branch svn-trunk updated (8c20ea8 -> 6101272)

Barak A. Pearlmutter barak+git at cs.nuim.ie
Sun Aug 17 08:21:50 UTC 2014


This is an automated email from the git hooks/post-receive script.

bap pushed a change to branch svn-trunk
in repository mlpack.

      from  8c20ea8   Rewrite NMFALSTest so that random failures should be less likely.
       new  445059c   Const parameters will cause this to be two billion times faster.
       new  a123d97   Save the correct labels; thanks Anand!
       new  e24e964   Add files and some preliminary code for R tree
       new  fad2e2e   Fix comment and clarify that it's pertaining to the runtime constructor, not the template parameters.
       new  e597e9a   It turns out the implementation of FurthestPointDistance() was just wrong. Thanks Andrew for pointing it out.
       new  f20c1bf   The test for FurthestPointDistance() was wrong also.
       new  087ae70   readability changes to BinarySpaceTree::FurthestPointDistance()
       new  57dc6ed   Add new contributor.
       new  7fa87c5   Remove the Pelleg-Moore k-means implementation; it is being replaced.
       new  9faf541   added more code for the rectangle type trees
       new  f98fec4   added module 'lmf'(Latent Matrix Factorization) to accommodate SVD based update rules alongside NMF based update rule. CF module is updated to use LMF module.
       new  367b49e   remove the silly folders I added when I meant to add txt files.
       new  b4790cb   now add the actual files the way they are supposed to exist.  No real code yet.
       new  ac788bb   Remove HasParentDistance trait, becase it wasn't used anywhere.
       new  62edcbe   added comment to base case.  Some more preliminaries for rectangle trees.
       new  dbfe468   add a short and hopefully useful explanation of what different files are for, focusing on BSP trees.
       new  63fc185   add a description of statistic
       new  f4ab5bb   Don't select points from clusters of size 1.
       new  efbd62b   more code for the RectangleTree.  Still not built yet.
       new  c9963e9   more R tree stuff.  Still no build
       new  bf8e283   more R tree stuff.  Still no build
       new  57e739a   Ensure that BaseCase() is called right after a node is scored.
       new  638a1b4   It's no longer reasonable to say the FASTLab developed mlpack, because first, the FASTLab no longer exists, and second, there's a lot of code from other places.  I should have removed this years ago, but I never noticed it until today...
       new  c93a8b4   What was this macro even for?
       new  4fa92d1   Refactor core.hpp into two files; one for non-mlpack includes, and one for everything else.
       new  46c6eed   Use prereqs.hpp to avoid weird include order dependency issues.
       new  6b04ba6   Use prereqs.hpp instead of core.hpp; core.hpp is required for the implementation, though.
       new  33d95b1   Include stdint, for uint32_t support.
       new  289a17d   Use prereqs.hpp instead of core.hpp; however, the implementation requires core.hpp.
       new  bde09c7   Fix header guard names.  This is an incredibly pedantic fix.
       new  97033eb   Remove unnecessary includes.
       new  ce06746   more R tree stuff.
       new  fe3b019   more stuff for R Trees
       new  70f5d69   Changing LMF(Latent Matrix Factorization) to AMF(Alternating Matrix Factorization). NMF and CF modules are updted.
       new  00fb2fb   more R Tree stuff
       new  59bb7bc   R tree stuff
       new  cf304d0   added code for accessing immediate child nodes (need to think of a way to rename this to be less confusing).  Some more quasi-code to split nodes and insert points.
       new  d7573a0   Trivial spelling fix.
       new  14f3bfc   Fix/update some comments, almost finish the splitting algorithm.  Several miscellaneous changes.
       new  549a8fb   change name of leafSize to maxLeafSize.  more stuff for the R-tree.  Some name changes, some more node splitting, a start on traversal.
       new  1f242aa   Typo which causes a segfault.
       new  1fc260f   Oops, this actually happens in two places.
       new  ad27d44   a few miscellanious small changes.  Added to CMake.
       new  c753900   add the missed files.
       new  9da2b8c   require kd-trees to have leafSize of at least 1.  Add an assert to ensure that the SingleTreeTraverser isn't called on a tree with only one node.
       new  8e7fec7   Rectangle Tree Traversal implementation.
       new  e39cf6d   change a comment to be more accurate.  Add on --r_tree option to the knn_all program.  Doesn't do anything yet.
       new  d7f04e7   comment out compilation.  Some small changes to support R tree in all_knn
       new  35db752   fixed some mistakes.  Still commented out to allow build.
       new  dffc2dd   * Modified AMF module so that now it uses tolerance checking rather   than minResidue checking * Added SVD batch learning
       new  984393c   * fixed include error
       new  383e30b   * Added momentum to SVD batch learning * AMF now calls Initialize on update rule before starting the optimization * Every update rule should now implement Initialize accepting data matrix   and rank
       new  bc3412b   more fixes.  Compilation still commented out.
       new  b6595ae   more of the same.
       new  b30e509   Inline the simple R tree descent heuristic.
       new  0e2b33c   Remove leafSize parameter from DTB constructor.
       new  88de233   Decision Stump added
       new  2a7a018   Fix #334 by ensuring vector accesses don't go out of bounds.
       new  2f16596   More bug fixes.
       new  b4f6095   Fixed armadillo issues, along with removing uninitialized and unused variables
       new  53fffdb   more bug fixes
       new  bc95ab6   keep the root node at the same address.
       new  69a7514   bug fixes
       new  b908875   Link against libstd++ when using Clang, and also use -Wall.
       new  f9e81fd   Fix for convergence, because sometimes the residue may increase (especially with ALS update rules).
       new  599af89   Make a note that this set of update rules often creates lots of NaNs when sparse matrices are used.
       new  096e707   Significant refactoring of NMF tests.  The sparse tests were generally invalid because NMF is not guaranteed to produce a unique decomposition.  Instead, now we test the Frobenius norm.  Also, the tolerances needed to be significantly adjusted because of the new convergence criteria in the AMF class.  Lastly, the SparseNMFRandomDiv test and SparseNMFDefaultTest have been removed because those update rules seem to often result in NaNs when sparse matrices are used as input.
       new  b96a620   Remove backup file from emacs or some other inferior editor that isn't vim.
       new  109f98a   First pass: comment standardization, fix header guard names, move .cpp to .hpp because it's all templated functions.
       new  766cba8   Use bool instead of int.
       new  6892081   Perceptron Added
       new  eaf1fcb   bug fixes for memory leaks
       new  3b6525e   bug fix.   had n_rows when I needed n_cols
       new  9417089   Minor changes to test.  const-correctness and comment normalization for Doxygen.
       new  93754b5   Decision Stump test fixed
       new  47402ce   Use const double where possible instead of having a variable used throughout the function.  This is only for clarity and consistency and is not likely to make any runtime difference.
       new  5a40b6a   For clarity, use separate split and binLabels objects instead of storing the label in the split matrix.  No casting is necessary anymore.
       new  b79a3ae   Remove unused variables.
       new  e274306   Oops, missed one unused variable.
       new  79d2b21   Remove classLabels; it isn't necessary for the DecisionStump class to hold it.
       new  4bf2852   * Templatized termination of AMF matrix factorization * added 3 termination policies 1) SimpleResidueTermination 2) SimpleToleranceTermination 3) ValidationRMSETermination
       new  b2cdb65   Assume the response to be the last column of the input file, as pointed out in the description.
       new  b4ce20e   Fix ambiguous math reference error, to pass the test clang.
       new  ed93d58   Easy fix for #355.
       new  7e90db6   Document minor fix.
       new  fbbff22   Remove oneClass and defaultClass variables.  There is a shortcut that can be taken when all the labels are the same, but the Entropy() function does not appear to be working correctly.
       new  9c6b2e9   * Changes in AMF abstraction, Termination policy is made first template parameter * fixed minor formatting issues
       new  5f60af8   Removing old cosine_tree code.
       new  fee7956   Adding new cosine_tree code.
       new  cbb1297   Rectangle tree and tests.  Construction seems to work.
       new  b86180a   memory leak fix
       new  d231b19   Combined CosineNode and CosineTree classes.
       new  d3d63f8   Rewinding the code review
       new  c67420a   Fixed subvec calls.
       new  083e69e   change the tree to store size_t in the nodes and keep the dataset together.  Other misc changes.
       new  0d0c2f8   Very simple change to fix build on i386.
       new  656b1bc   Initial commit of simulated annealing optimizer from Zhihao Lou.
       new  4692277   Minor refactoring for brevity.
       new  f8dd102   rectangle tree traverser
       new  3e37bdb   fix build
       new  f92d532   commit to ease debugging
       new  8e97a67   Move LaplaceDistribution to mlpack::distribution and expand upon it, but then remove MoveDistributionType template parameter from SA class because really, the MoveControl() function is written specifically to work with the Laplace distribution.  I was incorrect to think it could be templatized so easily.  The implementation of the Laplace distribution has been inlined, too, which may help it be a little faster.
       new  7b8fde0   Create separate target for moving mlpack headers.  This should fix #322.
       new  06a0422   Remove now-unnecessary include.
       new  1f31d61   work arround for CMake bug.
       new  eb08229   Remove documentation for MoveDistributionType.
       new  901316b   Rearrange parameters -- maxIterations is probably the most likely to be changed, and also maxIterations tends to be the first parameter for other algorithms.
       new  0e9c71e   Rename parameters, clean up documentation; very minor tweaks.  No functionality changes.
       new  053419e   * faster implementation of SVDBatchWithMomentum * tolerance termination policy is modified according to new policy * test point selection in validation RMSE termination is shifted to constructor
       new  caba057   Refactor to avoid holding variables in the class itself (reduces the size of SA a bit).  Remove an unnecessary variable too.  Minor changes, no public API changes.
       new  95a9e90   Add comments to private functions in header file.
       new  2742d4f   Explicitly use arma::sign().
       new  204fe37   Adding QUIC-SVD.
       new  19029a0   arma::sign() doesn't exist in Armadillo pre-3.920.
       new  5fd519a   Why was svn:ignore set for cf.cpp?  Removed.
       new  75ffcdd   svn:ignore build directories.
       new  32fe1c6   Minor cleanups; no functionality changes.  Make some things const where possible.
       new  4c64d31   Removing files added by mistake.
       new  f6511ea   Set default CoolingSchedule.
       new  8cdae3f   Add some more tests for simulated annealing, including a highly nonconvex function (the Rastrigin function).
       new  25bb2e6   New test added. Improved entropy calculation.
       new  a48e791   Refactor tree construction so that arbitrary tree types can be constructed.
       new  8d2d78e   Refactor tests, use BOOST_REQUIRE_EQUAL(), and add a test for EMST using cover trees.
       new  4880f80   Oops, I did not mean to comment out the sparse k-means test.
       new  3ab629a   Removed a few tabs.
       new  bca691f   * modified termination policies * fast SVDBatch implementation
       new  3cb09d7   * error in SVDBatchTest resolved
       new  0414c36   Bump minimum required Boost version to 1.49 for boost.heap.
       new  a114ec2   Split RAQueryStat into its own class.
       new  265c8fd   Refactor RASearch so that it does not accept a leafSize parameter and can build arbitrary tree types.
       new  9c213ed   Fix constructor calls, and automatically construct a cover tree with the default RASearch constructor.
       new  4beaf83   First pass -- move files to match naming policy, change initialize() to Initialize(), standardize comment formatting, fix some Doxygen commands.  No serious functionality changes.
       new  577f872   Minor formatting changes according to the style guide (mostly, I think?).
       new  0a8a169   tree traversal
       new  30b836b   Added check to solve ticket 347. Note that this does not changeany functionality, just adds a warning.
       new  8e5d648   R tree traversal test code.
       new  dfc8372   * added SVD Incomplete incremental learning * added Termination Policy wrapper for SVD Incomplete Learning
       new  4a3613c   Minor refactoring of AMF class; mostly renaming for consistency and clarification of comments.
       new  eb12d66   Trivial spacing fixes.
       new  72754b4   Rename for slightly changed API.
       new  2d19ffc   Move warning to prereqs.hpp, because sometimes prereqs.hpp is included and core.hpp is not.
       new  a96d00b   Don't include <armadillo> explicitly, because <mlpack/core.hpp> does that already.
       new  ddfab04   Add new contributor.
       new  88af83c   Note that we now have simulated annealing.
       new  ab3a44d   Don't include <armadillo> before <mlpack/core.hpp>.
       new  ae6d77a   Disambiguate: math::RandomSeed() -> mlpack::math::RandomSeed().  Issue noted by Zhihao.
       new  c8a057b   Patch from Zhihao: sa_update.diff.
       new  29f126f   Return the correct type of the matrix, because it isn't necessarily dense.
       new  9045a55   clang complains when default parameters aren't part of the original declaration. Also vim removes trailing whitespaces, so this diff looks way longer and more complex than it actually is...
       new  fed0425   Add support for HMM initial states.  Slight modification of API for creating HMMs by hand -- the initial parameter is now required.  This change may affect some existing results, and the new results may not perfectly agree with MATLAB, but MATLAB does not have the flexibility to seriously support initial probabilities.  It is possible to set the initial parameters such that it will emulate MATLAB behavior right, probably by setting the initial probabilities to the fi [...]
       new  4f1ddd7   Test HMM initial probabilities.
       new  05dcc79   Document that #315 is fixed.
       new  5ec50c7   Fix some formatting, fix backwards entropy splitting, add getters/setters, and comment a little bit about the internal structure of the class.
       new  09617c4   Another test to make sure the correct splitting attribute is used.
       new  67a2a26   Include mlpack/core.hpp.
       new  016b20a   R tree now has dataset and indices
       new  b942dc8   Entropy calculation improved.
       new  db5c21a   Const-correctness and 80-character lines... very trivial fix, no functionality change.
       new  023d657   Fix some formatting issues; no functionality change.
       new  13f3025   Use bool instead of int for tracking convergence.
       new  c7ef548   Don't use arma::unique() because it's slow.
       new  2a64815   IsDistinct() improved.
       new  deed1e1   Very minor changes.
       new  600d289   Lengthen comments that weren't 80 columns long.  This may be the most trivial fix ever in my long, decorated history of trivial commits.
       new  41e0e9d   * added local minima storing functionality to termination policies
       new  1d0265c   Changes are part of perceptron code review, as discussed with Ryan
       new  7984071   Minor improvement. No major functionality changes
       new  94745c0   Add MinimumBoundDistance(). This represents the minimum distance between the center of a node and any edge of the bound.  Note that for ball bounds, this is equivalent to the furthest descendant distance.
       new  6ae18f1   Fix elusive bug that only occurred in particularly rare situations.
       new  e919505   Add MinWidth(), which is a better solution than having the tree calculate it by hand.
       new  5dc686b   Test MinWidth().
       new  4c6da87   Use the bound's cached MinWidth() for MinimumBoundDistance().
       new  cf12517   Use slightly safer Width().
       new  7ca914b   Oops, this needed to be divided by 2.
       new  b98d1cc   Adding Regularized SVD Code
       new  7c35ec2   Simple style changes for consistency.
       new  07da6fb   Don't save a rowvector into a colvector. Use the build in option of the save function to transpose the vector.
       new  f554c6d   Point deletion.  Massive refactoring of Children()[i] to Child(i).  Really minor point deletion test.
       new  ebe7866   Rename executable.
       new  92f7181   Refactor main executable.  Some further refactorization will need to be done when everything transitions to arma::Row<size_t> for labels.
       new  6d4f811   Rename executable.
       new  e37b73e   more tests.  Really silly bug fix.
       new  bf1be19   * Added convergence test for SVDBatch and SVDIncremental learning
       new  6576848   Minor updates
       new  7f608d2   Adaboost design issues, to be discussed, then changed later on
       new  4c049ec   * added svd incomplete incremental learning tests * combined functions IsConverged and Step of termination policies into IsConverged
       new  14de1b6   The size parameter is unused outside the constructor; also, commit this change to trunk this time...
       new  0d09c20   Minor changes
       new  9b62228   Minor changes to the macros in the *main.cpp files.
       new  59bd2cd   Adaboost and Perceptron modified (improved constructor), going for tests on one Weak L
       new  1314d5f   point deletion.  bug fix.  more detailed test.
       new  6c33f70   Actually calculate score and base case for first node combination.
       new  b545960   * remaned SVDIncrementalLearning to SVDIncompleteIncrementalLearning * added SVDCompleteIncremenal learning * added CompleteIncrementalTermination wrapper
       new  f23c031   R* tree split and descent heuristic.
       new  946d61a   Add nystroem method for kernel approximation.
       new  81ad58f   Add test for the nystroem method.
       new  9f9eb69   Embedding nystroem method into kernel pca method.
       new  9751e20   Outsource the matrix multipilcation.
       new  4d9f654   Avoid direct multiplication with a diagmat.
       new  03c58b1   Changes to implementation of adaboost. Implemented adaboost.m1
       new  4363fef   Adaboost.mh implemented
       new  ab7f442   Comment normalization.
       new  cd2fd00   R* tree split.  Default to using the R* tree in allknn
       new  93b5138   Refactor to be stricter with types (arma::Col<size_t> instead of arma::vec).
       new  d727f7c   BiBTex added to adaboost_impl.hpp
       new  f0af04a   Adding tests for Reg SVD.
       new  a35033e   Integrate nystroem method into the kernel_pca_main.cpp file.
       new  30b24a9   Improved nystroem localisation ('oe')
       new  cd98bf3   Bug fix.  Node splitting tests.
       new  519c998   Adaboost improved. Tests for the UCI iris dataset added.
       new  317f220   Adaboost test improved and now works. Improved adaboost.
       new  9cb84b0   Contribution from Yash to solve #250 and make BallBound usable.
       new  028b666   Tests for ball trees and BallBound<> by Yash, to solve #250.
       new  adee919   Point out that there is a known bug.
       new  a5b79f4   Fix normalization bug (transpose); Some more comments.
       new  a09a0e6   Some notes on some things I've done.
       new  e75989e   Additional things that have changed.  I incorrectly wrote in the last commit about stuff "I had done" but I'm not sure why, given that each of these contributions are from people who are other people.
       new  70509e2   Ball tree tests for EMST.
       new  c9f95d5   Initialize reverseStepCount to 0 to avoid memory issues.
       new  eec4ef4   Fix an uninitialized value.
       new  b5c3dc5   Increase tolerances a little bit, because they were failing in some cases.
       new  4183cd3   For each random dataset, ensure that the size of the implied user/item matrix is numUsers by numItems by manually setting the last element.  Some formatting and const-correctness fixes.  Also, increase the number of iterations for the optimization test since it didn't seem to be converging (hopefully the specific number of 10 passes over the data was not chosen for a particular reason).
       new  8e6e1de   Patch from Sumedh: fix some uninitialized values.
       new  60212a2   Patch from sumedh: Mat::row_col_iterator
       new  a06e3f6   Patch from Sumedh: more tests for SVD
       new  07320f0   Patch from Sumedh: tests for row_col_iterator
       new  72b0f06   R* tree bug fixes.
       new  ae82033   Propagate HISTORY.txt change to trunk.
       new  2b32e08   New tests added for adaboost.
       new  f2fe64e   code cleanup for R tree
       new  a71ca61   Update documentation.
       new  1707b53   Propagate documentation fixes to trunk.
       new  0509486   Merge in changes from release branch.  I guess I should have done all this before I branched.
       new  d29e7c4   Fix warnings.
       new  861328a   This warning only appears on i386.
       new  293e733   Fix botched LaTeX formula.
       new  c852f7e   Set date of release.
       new  a6216c2   More comprehensive tests.
       new  b7f6189   Fix typo.
       new  d608d1d   R tree clean up.  Change Child() to match BSP tree.  Copy constructor takes a second argument to specify whether it is a deep copy.
       new  6902f83   Don't do base cases before recursing.  This is slightly cleaner code and will properly handle the case where the tree is only one node deep.
       new  399b309   Extremely minor changes.
       new  2d5872c   More tests for Adaboost added, with tolerance for change in rt also provided.
       new  507f1e2   Refactor KMeans so that the actual Lloyd iteration step is separate, since there are many ways to do a Lloyd iteration.
       new  7b09546   Add some comments.
       new  06cca20   Refactor test for changed API.
       new  a9a9922   Further refactoring of KMeans.  Fix MaxVarianceNewCluster, and also change it so that assignments are not passed in (since the inner KMeans loop doesn't really keep track of these now).
       new  73d1efc   Fix changed API.
       new  39c4a4d   Explicit cast for older versions of Armadillo.
       new  21af7cc   Explicit cast to arma::vec for old versions of Armadillo.
       new  3213704   Yet another reverse-compatibility fix for Armadillo.
       new  870d1fa   Return statements for operator++() and operator--().
       new  470a5e5   Fix out-of-order initialization warnings.
       new  59527ad   Fix error in random sample generation.
       new  f80384b   Add LaplaceDistribution to the list of default-ly included distributions.
       new  7817399   Adaboost now works with Decision Stumps; added tests for the same and extended tests for Perceptron too.
       new  4ace3ee   Minor code cleanups.
       new  7ab1939   * added plain SVD factorization - wrapper of arma::svd for CF module
       new  9f564cf   * added tests for PlainSVD
       new  cd7a15b   Removed phi, added gmm_convert_main.cpp
       new  26e527b   Removed phi, added gmm_convert_main.cpp
       new  99eb6fc   Changes to work with new, hierarchical GMMs
       new  e711d30   Changes to work with new, hierarchical GMMs
       new  f40b87e   utility for converting legacy GMM XML files to new format
       new  2d24ce4   Hierarchical GMMs store params in GaussianDistributions. Makes code clearer and simplifies Save/Load.
       new  148368c   Changes to work with new, hierarchical GMMsHierarchical GMMs store params in GaussianDistributions. Makes code clearer and simplifies Save/Load.
       new  aac8645   Hierarchical model support
       new  003da75   Hierarchical model support
       new  b8738a6   Hierarchical model support
       new  6e50fce   Hierarchical model support
       new  caad14a   Hierarchical model support
       new  df9c858   Implemented Save, Load
       new  124ad9a   Implemented Save, Load
       new  b355d89   Implemented Save, Load
       new  985c19d   Implemented Save, Load
       new  d351c4a   Added hmm_convert (legacy file converter util)
       new  4b0320b   Hierchical model support
       new  3f90d52   Implemented Save, Load for hierarchical model support
       new  7a6be9e   Implemented Save, Load for hierarchical model support
       new  55d117a   Added legacy file conversion
       new  d3cdde3   Legacy file conversion, deprecated old SaveHMM, LoadHMM functions
       new  22bfcd4   Legacy file converter util
       new  4d49b10   Added Save, Load tests
       new  b97ee0a   Updated Save, Load tests and updated tests for hierarchical format
       new  126bdb6   Remove duplicate declaration of ReadFile() to fix the failing build.
       new  c8938c0   Remove const identifier since it gets ignored and throws a warning.
       new  c956001   Minor formatting changes for row_col_iterators, mostly to adhere to the Whitesmiths style.
       new  8ee1e45   Test sparse iterators, when we can.
       new  15c5023   Point deletion bug fix.
       new  a996917   Fix incorrect check (how did this happen?).
       new  dee6bf6   Oops, I fixed this backwards.  The actual error is the output.
       new  4b20f83   * changed row_col_iterator::operator-- implementation * added documentation to termination policies * minor fix of PlainSVD module
       new  cc92921   * modified PlainSVD module to return normalized frobenius norm * modified PlainSVD tests * added row_col_iterator operator-- tests
       new  d7ded32   * minor changes
       new  6434e3b   GMM::Save() now adds type information to XML files
       new  cef9359   phi.hpp no longer used, see GaussianDistribution::Probability()
       new  ece3f7c   removed #include <mlpack/methods/gmm/phi.hpp> which included only unused functions
       new  fab5419   Add header comments and clean up BiBTeX citation a bit.
       new  59b4995   Add header comments, fix header guard naming to be in line with the rest of the files.
       new  4d2499d   Add some const and fix some formatting.
       new  f22cac6   Change names of functions.
       new  7f10b4b   Some formatting fixes.
       new  6e75b3c   Simplify a matrix calculation.
       new  f1dda22   Set ARMA_USE_U64S64 in hopes that the types of problems Gilles is having on the mailing list will be solved.
       new  c56d84a   Refactoring and optimizations on Adaboost.
       new  4397d06   added Save(), Load(), Type() to LaplaceDistribution
       new  896d40b   * changed PlainSVD to SVDwrapper * SVDwrapper is templatized to support other SVD factorizers * Added simple typedefs for simple API
       new  040918d   Return *this for operator--().
       new  36adb2d   Fix odd tabbing issue.
       new  63a9063   * added documentation to all update rules
       new  1af3081   X tree commit
       new  05224a7   First pass: make lines 80 characters long, tabs to spaces, and bracket surgery (or something).
       new  6942ef0   Typing failure.
       new  7427219   Don't add row_col_iterator support after 4.349 (currently svn trunk) since Conrad accepted our patches.
       new  da2f8ba   When Conrad took the patch, he stripped out a lot of compatibility between iterators, so... take the tests depending on that functionality out.
       new  b3b83f4   First pass: make things 80 characters, minor style fixes.
       new  0bf8a9b   How did I accidentally remove two lines?  I'm not actually sure.
       new  ed50e1c   Center the reconstructed approximation and the kernel matrix.
       new  504be03   Scale the transformed data matrix.
       new  08224bf   Use the maxIterartion as template parameter.
       new  d02aa8a   X tree
       new  b669269   Decision Stumps modified, along with adding Classify() function to AdaBoost. Other minor changes (renaming).
       new  e596d8b   Minor formatting change, and use zeros() instead of fill().
       new  6101272   Made Reg SVD work with CF.

The 324 revisions listed above as "new" are entirely new to this
repository and will be described in separate emails.  The revisions
listed as "adds" were already present in the repository and have only
been added to this reference.


Summary of changes:
 CMakeLists.txt                                     |   37 +-
 COPYRIGHT.txt                                      |    2 +
 HISTORY.txt                                        |   27 +-
 doc/guide/build.hpp                                |    2 +-
 doc/tutorials/emst/emst.txt                        |   12 +-
 doc/tutorials/range_search/range_search.txt        |    9 +-
 src/mlpack/CMakeLists.txt                          |    7 +-
 src/mlpack/core.hpp                                |   48 +-
 src/mlpack/core/arma_extend/CMakeLists.txt         |    2 +
 src/mlpack/core/arma_extend/Mat_extra_bones.hpp    |  141 ++
 src/mlpack/core/arma_extend/Mat_extra_meat.hpp     |  477 ++++++
 src/mlpack/core/arma_extend/SpMat_extra_bones.hpp  |   18 +
 src/mlpack/core/arma_extend/SpMat_extra_meat.hpp   |   37 +
 src/mlpack/core/arma_extend/arma_extend.hpp        |    9 +
 src/mlpack/core/data/normalize_labels.hpp          |    2 +-
 src/mlpack/core/data/normalize_labels_impl.hpp     |    3 +-
 src/mlpack/core/dists/CMakeLists.txt               |    2 +
 src/mlpack/core/dists/discrete_distribution.cpp    |    8 +
 src/mlpack/core/dists/discrete_distribution.hpp    |    9 +-
 src/mlpack/core/dists/gaussian_distribution.cpp    |   40 +
 src/mlpack/core/dists/gaussian_distribution.hpp    |   56 +-
 src/mlpack/core/dists/laplace_distribution.cpp     |  104 ++
 src/mlpack/core/dists/laplace_distribution.hpp     |  151 ++
 src/mlpack/core/kernels/gaussian_kernel.hpp        |    6 +-
 src/mlpack/core/math/lin_alg.cpp                   |    3 +-
 src/mlpack/core/math/lin_alg.hpp                   |    2 +-
 src/mlpack/core/math/random.hpp                    |    5 +-
 src/mlpack/core/optimizers/CMakeLists.txt          |    1 +
 .../core/optimizers/{sgd => sa}/CMakeLists.txt     |    7 +-
 .../core/optimizers/sa/exponential_schedule.hpp    |   64 +
 src/mlpack/core/optimizers/sa/sa.hpp               |  225 +++
 src/mlpack/core/optimizers/sa/sa_impl.hpp          |  229 +++
 src/mlpack/core/optimizers/sgd/sgd.hpp             |    2 +
 src/mlpack/core/optimizers/sgd/sgd_impl.hpp        |    1 +
 src/mlpack/core/tree/CMakeLists.txt                |   21 +-
 src/mlpack/core/tree/TREE_EXPLANATION.txt          |  101 ++
 src/mlpack/core/tree/ballbound.hpp                 |   72 +-
 src/mlpack/core/tree/ballbound_impl.hpp            |  173 ++-
 src/mlpack/core/tree/binary_space_tree.hpp         |    2 +
 .../tree/binary_space_tree/binary_space_tree.hpp   |   63 +-
 .../binary_space_tree/binary_space_tree_impl.hpp   |   62 +-
 src/mlpack/core/tree/binary_space_tree/traits.hpp  |    6 -
 src/mlpack/core/tree/cosine_tree/cosine_tree.cpp   |  425 ++++++
 src/mlpack/core/tree/cosine_tree/cosine_tree.hpp   |  277 +++-
 .../core/tree/cosine_tree/cosine_tree_builder.hpp  |  108 --
 .../tree/cosine_tree/cosine_tree_builder_impl.hpp  |  172 ---
 .../core/tree/cosine_tree/cosine_tree_impl.hpp     |  108 --
 src/mlpack/core/tree/cover_tree.hpp                |    2 +
 src/mlpack/core/tree/cover_tree/cover_tree.hpp     |    4 +
 .../tree/cover_tree/dual_tree_traverser_impl.hpp   |    7 +-
 src/mlpack/core/tree/cover_tree/traits.hpp         |    7 -
 src/mlpack/core/tree/hrectbound.hpp                |   10 +-
 src/mlpack/core/tree/hrectbound_impl.hpp           |   25 +-
 src/mlpack/core/tree/rectangle_tree.hpp            |   27 +
 .../dual_tree_traverser.hpp                        |   38 +-
 .../rectangle_tree/dual_tree_traverser_impl.hpp    |   49 +
 .../tree/rectangle_tree/hilbert_r_tree_split.hpp   |    1 +
 .../rectangle_tree/hilbert_r_tree_split_impl.hpp   |    1 +
 .../r_star_tree_descent_heuristic.hpp              |   44 +
 .../r_star_tree_descent_heuristic_impl.hpp         |  149 ++
 .../core/tree/rectangle_tree/r_star_tree_split.hpp |   73 +
 .../tree/rectangle_tree/r_star_tree_split_impl.hpp |  596 ++++++++
 .../rectangle_tree/r_tree_descent_heuristic.hpp    |   58 +
 .../r_tree_descent_heuristic_impl.hpp              |  146 ++
 .../core/tree/rectangle_tree/r_tree_split.hpp      |   90 ++
 .../core/tree/rectangle_tree/r_tree_split_impl.hpp |  528 +++++++
 .../core/tree/rectangle_tree/rectangle_tree.hpp    |  563 +++++++
 .../tree/rectangle_tree/rectangle_tree_impl.hpp    |  861 +++++++++++
 .../tree/rectangle_tree/single_tree_traverser.hpp  |   74 +
 .../rectangle_tree/single_tree_traverser_impl.hpp  |   77 +
 src/mlpack/core/tree/rectangle_tree/traits.hpp     |   52 +
 .../core/tree/rectangle_tree/x_tree_split.hpp      |   78 +
 .../core/tree/rectangle_tree/x_tree_split_impl.hpp |  715 +++++++++
 src/mlpack/core/tree/tree_traits.hpp               |   16 +-
 src/mlpack/core/util/log.hpp                       |    4 +-
 src/mlpack/core/util/option.hpp                    |    4 +-
 src/mlpack/core/util/option_impl.hpp               |    4 +-
 src/mlpack/core/util/prefixedoutstream.hpp         |    4 +-
 src/mlpack/core/util/prefixedoutstream_impl.hpp    |    4 +-
 src/mlpack/core/util/save_restore_utility.cpp      |   74 +-
 src/mlpack/core/util/save_restore_utility.hpp      |   58 +-
 src/mlpack/core/util/save_restore_utility_impl.hpp |   10 +-
 src/mlpack/methods/CMakeLists.txt                  |    8 +
 .../methods/{fastmks => adaboost}/CMakeLists.txt   |   17 +-
 src/mlpack/methods/adaboost/adaboost.hpp           |   91 ++
 src/mlpack/methods/adaboost/adaboost_impl.hpp      |  267 ++++
 src/mlpack/methods/adaboost/adaboost_main.cpp      |  147 ++
 src/mlpack/methods/{cf => amf}/CMakeLists.txt      |   14 +-
 src/mlpack/methods/amf/amf.hpp                     |  262 ++++
 src/mlpack/methods/amf/amf_impl.hpp                |   74 +
 .../methods/{nmf/nmf_main.cpp => amf/amf_main.cpp} |   71 +-
 .../methods/{cf => amf/init_rules}/CMakeLists.txt  |   12 +-
 .../{nmf => amf/init_rules}/random_acol_init.hpp   |    8 +-
 .../{nmf => amf/init_rules}/random_init.hpp        |    8 +-
 .../termination_policies}/CMakeLists.txt           |   15 +-
 .../complete_incremental_termination.hpp           |  108 ++
 .../incomplete_incremental_termination.hpp         |   92 ++
 .../simple_residue_termination.hpp                 |  115 ++
 .../simple_tolerance_termination.hpp               |  191 +++
 .../validation_RMSE_termination.hpp                |  236 +++
 .../{cf => amf/update_rules}/CMakeLists.txt        |   16 +-
 .../update_rules/nmf_als.hpp}                      |   69 +-
 .../methods/amf/update_rules/nmf_mult_dist.hpp     |   80 +
 .../update_rules/nmf_mult_div.hpp}                 |   82 +-
 .../amf/update_rules/svd_batch_learning.hpp        |  241 +++
 .../svd_complete_incremental_learning.hpp          |  262 ++++
 .../svd_incomplete_incremental_learning.hpp        |  202 +++
 src/mlpack/methods/cf/CMakeLists.txt               |    2 +
 src/mlpack/methods/cf/cf.hpp                       |   62 +-
 src/mlpack/methods/cf/cf_impl.hpp                  |  133 +-
 src/mlpack/methods/cf/svd_wrapper.hpp              |   73 +
 src/mlpack/methods/cf/svd_wrapper_impl.hpp         |  128 ++
 .../{fastmks => decision_stump}/CMakeLists.txt     |   16 +-
 .../methods/decision_stump/decision_stump.hpp      |  183 +++
 .../methods/decision_stump/decision_stump_impl.hpp |  470 ++++++
 .../methods/decision_stump/decision_stump_main.cpp |  113 ++
 src/mlpack/methods/det/dt_utils.cpp                |   11 +-
 src/mlpack/methods/emst/dtb.hpp                    |    6 +-
 src/mlpack/methods/emst/dtb_impl.hpp               |  150 +-
 src/mlpack/methods/emst/dtb_rules.hpp              |   19 +-
 src/mlpack/methods/emst/dtb_rules_impl.hpp         |   11 +-
 src/mlpack/methods/emst/dtb_stat.hpp               |   14 +-
 src/mlpack/methods/emst/emst_main.cpp              |   36 +-
 src/mlpack/methods/emst/union_find.hpp             |    3 +-
 src/mlpack/methods/fastmks/fastmks_main.cpp        |    7 -
 src/mlpack/methods/gmm/CMakeLists.txt              |   15 +-
 src/mlpack/methods/gmm/em_fit.hpp                  |   14 +-
 src/mlpack/methods/gmm/em_fit_impl.hpp             |  102 +-
 src/mlpack/methods/gmm/gmm.hpp                     |  110 +-
 src/mlpack/methods/gmm/gmm_convert_main.cpp        |   63 +
 src/mlpack/methods/gmm/gmm_impl.hpp                |  238 ++-
 src/mlpack/methods/gmm/phi.hpp                     |  147 --
 src/mlpack/methods/hmm/CMakeLists.txt              |    9 +-
 src/mlpack/methods/hmm/hmm.hpp                     |   39 +-
 src/mlpack/methods/hmm/hmm_convert_main.cpp        |   77 +
 src/mlpack/methods/hmm/hmm_generate_main.cpp       |   26 +-
 src/mlpack/methods/hmm/hmm_impl.hpp                |   75 +-
 src/mlpack/methods/hmm/hmm_util.hpp                |   19 +-
 src/mlpack/methods/hmm/hmm_util_impl.hpp           |  127 +-
 src/mlpack/methods/kernel_pca/CMakeLists.txt       |    2 +
 src/mlpack/methods/kernel_pca/kernel_pca.hpp       |   33 +-
 src/mlpack/methods/kernel_pca/kernel_pca_impl.hpp  |  102 +-
 src/mlpack/methods/kernel_pca/kernel_pca_main.cpp  |  102 +-
 .../{cf => kernel_pca/kernel_rules}/CMakeLists.txt |   12 +-
 .../kernel_pca/kernel_rules/naive_method.hpp       |   89 ++
 .../kernel_pca/kernel_rules/nystroem_method.hpp    |   79 +
 src/mlpack/methods/kmeans/CMakeLists.txt           |    2 +
 src/mlpack/methods/kmeans/allow_empty_clusters.hpp |    6 +-
 src/mlpack/methods/kmeans/kmeans.hpp               |   33 +-
 src/mlpack/methods/kmeans/kmeans_impl.hpp          |  641 ++------
 src/mlpack/methods/kmeans/kmeans_main.cpp          |   26 +-
 .../methods/kmeans/max_variance_new_cluster.hpp    |    6 +-
 .../kmeans/max_variance_new_cluster_impl.hpp       |   42 +-
 src/mlpack/methods/kmeans/naive_kmeans.hpp         |   60 +
 src/mlpack/methods/kmeans/naive_kmeans_impl.hpp    |   70 +
 src/mlpack/methods/lars/lars.hpp                   |    5 +-
 .../linear_regression/linear_regression_main.cpp   |    3 +-
 .../logistic_regression_main.cpp                   |   14 +-
 .../methods/naive_bayes/naive_bayes_classifier.hpp |    1 -
 src/mlpack/methods/naive_bayes/nbc_main.cpp        |    4 +-
 src/mlpack/methods/neighbor_search/allknn_main.cpp |  216 ++-
 .../methods/neighbor_search/neighbor_search.hpp    |    1 +
 .../neighbor_search/neighbor_search_impl.hpp       |    9 +-
 .../neighbor_search/neighbor_search_rules.hpp      |    9 +-
 .../neighbor_search/neighbor_search_rules_impl.hpp |   16 +-
 src/mlpack/methods/nmf/CMakeLists.txt              |   21 -
 src/mlpack/methods/nmf/mult_dist_update_rules.hpp  |   81 -
 src/mlpack/methods/nmf/nmf.hpp                     |  158 --
 src/mlpack/methods/nmf/nmf_impl.hpp                |  111 --
 src/mlpack/methods/nmf/nmf_main.cpp                |   37 +-
 .../methods/{cf => nystroem_method}/CMakeLists.txt |   15 +-
 .../methods/nystroem_method/kmeans_selection.hpp   |   51 +
 .../methods/nystroem_method/nystroem_method.hpp    |   81 +
 .../nystroem_method/nystroem_method_impl.hpp       |   91 ++
 .../methods/nystroem_method/ordered_selection.hpp  |   38 +
 .../methods/nystroem_method/random_selection.hpp   |   40 +
 .../methods/{fastmks => perceptron}/CMakeLists.txt |   19 +-
 .../initialization_methods}/CMakeLists.txt         |   12 +-
 .../initialization_methods/random_init.hpp         |   35 +
 .../initialization_methods/zero_init.hpp           |   37 +
 .../learning_policies}/CMakeLists.txt              |   11 +-
 .../learning_policies/simple_weight_update.hpp     |   58 +
 src/mlpack/methods/perceptron/perceptron.hpp       |   94 ++
 src/mlpack/methods/perceptron/perceptron_impl.hpp  |  168 +++
 src/mlpack/methods/perceptron/perceptron_main.cpp  |  124 ++
 .../CMakeLists.txt                                 |    6 +-
 src/mlpack/methods/quic_svd/quic_svd.hpp           |   70 +
 src/mlpack/methods/quic_svd/quic_svd_impl.hpp      |   82 +
 src/mlpack/methods/rann/CMakeLists.txt             |   10 +-
 src/mlpack/methods/rann/ra_query_stat.hpp          |   67 +
 src/mlpack/methods/rann/ra_search.hpp              |   63 +-
 src/mlpack/methods/rann/ra_search_impl.hpp         |  194 +--
 .../CMakeLists.txt                                 |    8 +-
 .../methods/regularized_svd/regularized_svd.hpp    |   84 ++
 .../regularized_svd/regularized_svd_function.cpp   |  181 +++
 .../regularized_svd/regularized_svd_function.hpp   |  115 ++
 .../regularized_svd/regularized_svd_impl.hpp       |   52 +
 src/mlpack/prereqs.hpp                             |   51 +
 src/mlpack/tests/CMakeLists.txt                    |   12 +
 src/mlpack/tests/adaboost_test.cpp                 |  566 +++++++
 src/mlpack/tests/allkfn_test.cpp                   |  149 ++
 src/mlpack/tests/allknn_test.cpp                   | 1565 ++++++++++----------
 src/mlpack/tests/allkrann_search_test.cpp          |  145 +-
 src/mlpack/tests/arma_extend_test.cpp              |  213 +++
 src/mlpack/tests/armadillo_svd_test.cpp            |   52 +
 src/mlpack/tests/cf_test.cpp                       |   27 -
 src/mlpack/tests/cosine_tree_test.cpp              |  186 +++
 src/mlpack/tests/data/german.csv                   | 1000 +++++++++++++
 src/mlpack/tests/data/iris.txt                     |  150 ++
 src/mlpack/tests/data/iris_labels.txt              |  150 ++
 src/mlpack/tests/data/nonlinsepdata.txt            |  200 +++
 src/mlpack/tests/data/nonlinsepdata_labels.txt     |  200 +++
 src/mlpack/tests/data/vc2.txt                      |  310 ++++
 src/mlpack/tests/data/vc2_labels.txt               |  310 ++++
 src/mlpack/tests/decision_stump_test.cpp           |  310 ++++
 src/mlpack/tests/emst_test.cpp                     |  210 ++-
 src/mlpack/tests/gmm_test.cpp                      |  293 ++--
 src/mlpack/tests/hmm_test.cpp                      |  359 ++++-
 src/mlpack/tests/kernel_pca_test.cpp               |   70 +-
 src/mlpack/tests/kmeans_test.cpp                   |   37 +-
 src/mlpack/tests/lars_test.cpp                     |    6 +-
 src/mlpack/tests/local_coordinate_coding_test.cpp  |    4 +-
 src/mlpack/tests/nmf_test.cpp                      |  280 ++--
 src/mlpack/tests/nystroem_method_test.cpp          |  182 +++
 src/mlpack/tests/perceptron_test.cpp               |  169 +++
 src/mlpack/tests/quic_svd_test.cpp                 |   42 +
 src/mlpack/tests/radical_test.cpp                  |    3 +-
 src/mlpack/tests/range_search_test.cpp             |  248 +++-
 src/mlpack/tests/rectangle_tree_test.cpp           |  974 ++++++++++++
 src/mlpack/tests/regularized_svd_test.cpp          |  240 +++
 src/mlpack/tests/sa_test.cpp                       |  117 ++
 src/mlpack/tests/svd_batch_test.cpp                |  175 +++
 src/mlpack/tests/svd_incremental_test.cpp          |  119 ++
 src/mlpack/tests/to_string_test.cpp                |    6 +-
 src/mlpack/tests/tree_test.cpp                     |  291 ++--
 src/mlpack/tests/tree_traits_test.cpp              |   18 +-
 236 files changed, 21734 insertions(+), 4479 deletions(-)
 create mode 100644 src/mlpack/core/arma_extend/Mat_extra_bones.hpp
 create mode 100644 src/mlpack/core/arma_extend/Mat_extra_meat.hpp
 create mode 100644 src/mlpack/core/dists/laplace_distribution.cpp
 create mode 100644 src/mlpack/core/dists/laplace_distribution.hpp
 copy src/mlpack/core/optimizers/{sgd => sa}/CMakeLists.txt (74%)
 create mode 100644 src/mlpack/core/optimizers/sa/exponential_schedule.hpp
 create mode 100644 src/mlpack/core/optimizers/sa/sa.hpp
 create mode 100644 src/mlpack/core/optimizers/sa/sa_impl.hpp
 create mode 100644 src/mlpack/core/tree/TREE_EXPLANATION.txt
 create mode 100644 src/mlpack/core/tree/cosine_tree/cosine_tree.cpp
 delete mode 100644 src/mlpack/core/tree/cosine_tree/cosine_tree_builder.hpp
 delete mode 100644 src/mlpack/core/tree/cosine_tree/cosine_tree_builder_impl.hpp
 delete mode 100644 src/mlpack/core/tree/cosine_tree/cosine_tree_impl.hpp
 create mode 100644 src/mlpack/core/tree/rectangle_tree.hpp
 copy src/mlpack/core/tree/{binary_space_tree => rectangle_tree}/dual_tree_traverser.hpp (69%)
 create mode 100644 src/mlpack/core/tree/rectangle_tree/dual_tree_traverser_impl.hpp
 create mode 100644 src/mlpack/core/tree/rectangle_tree/hilbert_r_tree_split.hpp
 create mode 100644 src/mlpack/core/tree/rectangle_tree/hilbert_r_tree_split_impl.hpp
 create mode 100644 src/mlpack/core/tree/rectangle_tree/r_star_tree_descent_heuristic.hpp
 create mode 100644 src/mlpack/core/tree/rectangle_tree/r_star_tree_descent_heuristic_impl.hpp
 create mode 100644 src/mlpack/core/tree/rectangle_tree/r_star_tree_split.hpp
 create mode 100644 src/mlpack/core/tree/rectangle_tree/r_star_tree_split_impl.hpp
 create mode 100644 src/mlpack/core/tree/rectangle_tree/r_tree_descent_heuristic.hpp
 create mode 100644 src/mlpack/core/tree/rectangle_tree/r_tree_descent_heuristic_impl.hpp
 create mode 100644 src/mlpack/core/tree/rectangle_tree/r_tree_split.hpp
 create mode 100644 src/mlpack/core/tree/rectangle_tree/r_tree_split_impl.hpp
 create mode 100644 src/mlpack/core/tree/rectangle_tree/rectangle_tree.hpp
 create mode 100644 src/mlpack/core/tree/rectangle_tree/rectangle_tree_impl.hpp
 create mode 100644 src/mlpack/core/tree/rectangle_tree/single_tree_traverser.hpp
 create mode 100644 src/mlpack/core/tree/rectangle_tree/single_tree_traverser_impl.hpp
 create mode 100644 src/mlpack/core/tree/rectangle_tree/traits.hpp
 create mode 100644 src/mlpack/core/tree/rectangle_tree/x_tree_split.hpp
 create mode 100644 src/mlpack/core/tree/rectangle_tree/x_tree_split_impl.hpp
 copy src/mlpack/methods/{fastmks => adaboost}/CMakeLists.txt (68%)
 create mode 100644 src/mlpack/methods/adaboost/adaboost.hpp
 create mode 100644 src/mlpack/methods/adaboost/adaboost_impl.hpp
 create mode 100644 src/mlpack/methods/adaboost/adaboost_main.cpp
 copy src/mlpack/methods/{cf => amf}/CMakeLists.txt (75%)
 create mode 100644 src/mlpack/methods/amf/amf.hpp
 create mode 100644 src/mlpack/methods/amf/amf_impl.hpp
 copy src/mlpack/methods/{nmf/nmf_main.cpp => amf/amf_main.cpp} (64%)
 copy src/mlpack/methods/{cf => amf/init_rules}/CMakeLists.txt (75%)
 rename src/mlpack/methods/{nmf => amf/init_rules}/random_acol_init.hpp (93%)
 rename src/mlpack/methods/{nmf => amf/init_rules}/random_init.hpp (87%)
 copy src/mlpack/methods/{cf => amf/termination_policies}/CMakeLists.txt (69%)
 create mode 100644 src/mlpack/methods/amf/termination_policies/complete_incremental_termination.hpp
 create mode 100644 src/mlpack/methods/amf/termination_policies/incomplete_incremental_termination.hpp
 create mode 100644 src/mlpack/methods/amf/termination_policies/simple_residue_termination.hpp
 create mode 100644 src/mlpack/methods/amf/termination_policies/simple_tolerance_termination.hpp
 create mode 100644 src/mlpack/methods/amf/termination_policies/validation_RMSE_termination.hpp
 copy src/mlpack/methods/{cf => amf/update_rules}/CMakeLists.txt (72%)
 rename src/mlpack/methods/{nmf/als_update_rules.hpp => amf/update_rules/nmf_als.hpp} (54%)
 create mode 100644 src/mlpack/methods/amf/update_rules/nmf_mult_dist.hpp
 rename src/mlpack/methods/{nmf/mult_div_update_rules.hpp => amf/update_rules/nmf_mult_div.hpp} (56%)
 create mode 100644 src/mlpack/methods/amf/update_rules/svd_batch_learning.hpp
 create mode 100644 src/mlpack/methods/amf/update_rules/svd_complete_incremental_learning.hpp
 create mode 100644 src/mlpack/methods/amf/update_rules/svd_incomplete_incremental_learning.hpp
 create mode 100644 src/mlpack/methods/cf/svd_wrapper.hpp
 create mode 100644 src/mlpack/methods/cf/svd_wrapper_impl.hpp
 copy src/mlpack/methods/{fastmks => decision_stump}/CMakeLists.txt (65%)
 create mode 100644 src/mlpack/methods/decision_stump/decision_stump.hpp
 create mode 100644 src/mlpack/methods/decision_stump/decision_stump_impl.hpp
 create mode 100644 src/mlpack/methods/decision_stump/decision_stump_main.cpp
 create mode 100644 src/mlpack/methods/gmm/gmm_convert_main.cpp
 delete mode 100644 src/mlpack/methods/gmm/phi.hpp
 create mode 100644 src/mlpack/methods/hmm/hmm_convert_main.cpp
 copy src/mlpack/methods/{cf => kernel_pca/kernel_rules}/CMakeLists.txt (75%)
 create mode 100644 src/mlpack/methods/kernel_pca/kernel_rules/naive_method.hpp
 create mode 100644 src/mlpack/methods/kernel_pca/kernel_rules/nystroem_method.hpp
 create mode 100644 src/mlpack/methods/kmeans/naive_kmeans.hpp
 create mode 100644 src/mlpack/methods/kmeans/naive_kmeans_impl.hpp
 delete mode 100644 src/mlpack/methods/nmf/mult_dist_update_rules.hpp
 delete mode 100644 src/mlpack/methods/nmf/nmf.hpp
 delete mode 100644 src/mlpack/methods/nmf/nmf_impl.hpp
 copy src/mlpack/methods/{cf => nystroem_method}/CMakeLists.txt (75%)
 create mode 100644 src/mlpack/methods/nystroem_method/kmeans_selection.hpp
 create mode 100644 src/mlpack/methods/nystroem_method/nystroem_method.hpp
 create mode 100644 src/mlpack/methods/nystroem_method/nystroem_method_impl.hpp
 create mode 100644 src/mlpack/methods/nystroem_method/ordered_selection.hpp
 create mode 100644 src/mlpack/methods/nystroem_method/random_selection.hpp
 copy src/mlpack/methods/{fastmks => perceptron}/CMakeLists.txt (60%)
 copy src/mlpack/methods/{cf => perceptron/initialization_methods}/CMakeLists.txt (75%)
 create mode 100644 src/mlpack/methods/perceptron/initialization_methods/random_init.hpp
 create mode 100644 src/mlpack/methods/perceptron/initialization_methods/zero_init.hpp
 copy src/mlpack/methods/{cf => perceptron/learning_policies}/CMakeLists.txt (75%)
 create mode 100644 src/mlpack/methods/perceptron/learning_policies/simple_weight_update.hpp
 create mode 100644 src/mlpack/methods/perceptron/perceptron.hpp
 create mode 100644 src/mlpack/methods/perceptron/perceptron_impl.hpp
 create mode 100644 src/mlpack/methods/perceptron/perceptron_main.cpp
 copy src/mlpack/methods/{sparse_autoencoder => quic_svd}/CMakeLists.txt (77%)
 create mode 100644 src/mlpack/methods/quic_svd/quic_svd.hpp
 create mode 100644 src/mlpack/methods/quic_svd/quic_svd_impl.hpp
 create mode 100644 src/mlpack/methods/rann/ra_query_stat.hpp
 copy src/mlpack/methods/{sparse_autoencoder => regularized_svd}/CMakeLists.txt (77%)
 create mode 100644 src/mlpack/methods/regularized_svd/regularized_svd.hpp
 create mode 100644 src/mlpack/methods/regularized_svd/regularized_svd_function.cpp
 create mode 100644 src/mlpack/methods/regularized_svd/regularized_svd_function.hpp
 create mode 100644 src/mlpack/methods/regularized_svd/regularized_svd_impl.hpp
 create mode 100644 src/mlpack/prereqs.hpp
 create mode 100644 src/mlpack/tests/adaboost_test.cpp
 create mode 100644 src/mlpack/tests/armadillo_svd_test.cpp
 create mode 100644 src/mlpack/tests/cosine_tree_test.cpp
 create mode 100644 src/mlpack/tests/data/german.csv
 create mode 100644 src/mlpack/tests/data/iris.txt
 create mode 100644 src/mlpack/tests/data/iris_labels.txt
 create mode 100644 src/mlpack/tests/data/nonlinsepdata.txt
 create mode 100644 src/mlpack/tests/data/nonlinsepdata_labels.txt
 create mode 100644 src/mlpack/tests/data/vc2.txt
 create mode 100644 src/mlpack/tests/data/vc2_labels.txt
 create mode 100644 src/mlpack/tests/decision_stump_test.cpp
 create mode 100644 src/mlpack/tests/nystroem_method_test.cpp
 create mode 100644 src/mlpack/tests/perceptron_test.cpp
 create mode 100644 src/mlpack/tests/quic_svd_test.cpp
 create mode 100644 src/mlpack/tests/rectangle_tree_test.cpp
 create mode 100644 src/mlpack/tests/regularized_svd_test.cpp
 create mode 100644 src/mlpack/tests/sa_test.cpp
 create mode 100644 src/mlpack/tests/svd_batch_test.cpp
 create mode 100644 src/mlpack/tests/svd_incremental_test.cpp

-- 
Alioth's /usr/local/bin/git-commit-notice on /srv/git.debian.org/git/debian-science/packages/mlpack.git



More information about the debian-science-commits mailing list