[mlpack-git] [mlpack/mlpack] Density Estimation Tree made sparse-enabled (#802)
Ivan Georgiev
notifications at github.com
Tue Oct 18 06:24:44 EDT 2016
DET class can now work with Arma's sparse matrices. Also:
- The class is templated, so the matrix type becomes template argument.
- Couple of optimizations regarding matrix copying, incl. in dense matrix case.
- OpenMP-ification of tree building routine.
- Added few test cases with `arma::SpMat` data.
You can view, comment on, or merge this pull request online at:
https://github.com/mlpack/mlpack/pull/802
-- Commit Summary --
* - Mac gitignore changes.
* Merge remote-tracking branch 'upstream/master'
* - DTree class templated.
* - First successfull builtd.
* - DET changes propagated to tests.
* - DET templating ready and tests passing.
* - More sparse-matrix migration steps.
* - First successfull SpMat build.
* - Sparse DTree finally working.
* - All DET sparsification works.
* - OMP test line removed.
* Merge remote-tracking branch 'upstream/master'
* - DTree class templated.
* - First successfull builtd.
* - DET changes propagated to tests.
* - DET templating ready and tests passing.
* - More sparse-matrix migration steps.
* - First successfull SpMat build.
* - Sparse DTree finally working.
* - All DET sparsification works.
* - OMP test line removed.
* Merge remote-tracking branch 'origin/feature/det_sparse' into feature/det_sparse
-- File Changes --
M .gitignore (2)
M src/mlpack/core/arma_extend/Mat_extra_bones.hpp (9)
M src/mlpack/core/arma_extend/SpMat_extra_bones.hpp (8)
M src/mlpack/methods/det/CMakeLists.txt (6)
M src/mlpack/methods/det/det_main.cpp (5)
M src/mlpack/methods/det/dt_utils.hpp (23)
R src/mlpack/methods/det/dt_utils_impl.hpp (76)
M src/mlpack/methods/det/dtree.hpp (105)
R src/mlpack/methods/det/dtree_impl.hpp (410)
M src/mlpack/tests/det_test.cpp (162)
M src/mlpack/tests/serialization_test.cpp (15)
-- Patch Links --
https://github.com/mlpack/mlpack/pull/802.patch
https://github.com/mlpack/mlpack/pull/802.diff
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/mlpack/mlpack/pull/802
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.cc.gatech.edu/pipermail/mlpack-git/attachments/20161018/271764e9/attachment.html>
More information about the mlpack-git
mailing list