Upstream updates + data set merging + cost effective gradient boosting#2

Merged

alisterw merged 69 commits intoG-Research:masterfrom rvb:master

Feb 13, 2019

+2,202-550

Commits on Jan 17, 2019

[ci] removed temp brew hotfix and deprecated sudo option (microsoft#1951 )

StrikerRUS
authored and
chivee
committed

Commits on Jan 18, 2019

Commits on Jan 19, 2019

[R] Fix multiclass demo (microsoft#1940 )

maximilianeber
authored and
Laurae2
committed

Commits on Jan 20, 2019

Commits on Jan 21, 2019

[docs] Added the links to the libraries used (microsoft#1962 )
StrikerRUS
authored

Commits on Jan 22, 2019

[docs] fixed minor typos in documentation (microsoft#1959 )

jameslamb
authored and
StrikerRUS
committed

Commits on Jan 23, 2019

support to override some parameters in Dataset (microsoft#1876 )
guolinke
authored

Commits on Jan 26, 2019

Fix build on macOS Mojave (microsoft#1923 )
henry0312
authored

Commits on Jan 27, 2019

Commits on Jan 28, 2019

Found error from microsoft#1939 (microsoft#1974 )
Laurae2
authored

Commits on Jan 29, 2019

fix more edge cases in mape (microsoft#1977 )
guolinke
authored

Commits on Jan 31, 2019

fix microsoft#1981
guolinke
committed

Commits on Feb 2, 2019

[python] added OpenMP options for python-package installation (microsoft#1975 )

StrikerRUS
authored and
henry0312
committed
improved model loading routines (microsoft#1979 )

StrikerRUS
authored and
guolinke
committed
[ci] refined command status check (microsoft#1980 )

StrikerRUS
authored and
guolinke
committed
cpplint whitespaces and new lines (microsoft#1986 )

StrikerRUS
authored and
guolinke
committed

Commits on Feb 3, 2019

fix microsoft#1994
guolinke
committed
[docs] Fixed OpenCL Debian package name typo (microsoft#1995 )

washcycle
authored and
StrikerRUS
committed

Commits on Feb 4, 2019

Commits on Feb 5, 2019

Commits on Feb 6, 2019

Initial attempt to implement appending features in-memory to another data set
Remco Bras
committed
Implement Dataset.dump_text, and fix small bug in appending of group bin boundaries.
Remco Bras
committed
Add basic tests for validation logic for add_features_from.
Remco Bras
committed
Remove various internal mapping items from dataset text dumps
Remco Bras
committed
Add test that add_features_from results in identical data sets according to dump_text.
Remco Bras
committed
Add test that booster behaviour after using add_features_from matches that of training on the full data
Remco Bras
committed
Expose feature_penalty and monotone_types/constraints via get_field
Remco Bras
committed
Add tests that add_features correctly handles feature_penalty and monotone_constraints.
Remco Bras
committed
Ensure add_features_from properly frees the added dataset and add unit test for this
Remco Bras
committed
Remove some obsolete TODOs
Remco Bras
committed
Tidy up DumpTextFile by using a single iterator for each feature
Remco Bras
committed
Factor out offsetting logic in AddFeaturesFrom
Remco Bras
committed
Remove obsolete TODO
Remco Bras
committed
Remove another TODO
Remco Bras
committed
Add documentation for newly-added methods.
Remco Bras
committed

Commits on Feb 8, 2019

Initial work towards add_data_from
Remco Bras
committed
Fix bug where dense bin copy of num_data_ wasn't updated
Remco Bras
committed
Small bug fix in dense_bin.hpp, initial implementation of Merge for 4-bits bin.
Remco Bras
committed
Add unit test for dense bin case of add_data_from, and refactor tests slightly.
Remco Bras
committed
Initial implementation of Merge for sparse bins and unit tests for it.
Remco Bras
committed
Ensure we test merging sparse data sets after loading them from binary
Remco Bras
committed
Add labels to text dumps.
Remco Bras
committed
Add weights to text dumps.
Remco Bras
committed
Ensure add_data_from properly merges labels.
Remco Bras
committed
Ensure metadata appends weights correctly, and unit test for it.
Remco Bras
committed
Implement metadata merging for query bits
Remco Bras
committed
Check datasets are aligned before merging.
Remco Bras
committed
Add test that booster behaviour is preserved by add_data_from.
Remco Bras
committed

Commits on Feb 12, 2019

Add configuration parameters for CEGB.
Remco Bras
committed
Add skeleton CEGB tree learner
Remco Bras
committed
Track features used in CEGB tree learner.
Remco Bras
committed
Pull CEGB tradeoff and coupled feature penalty from config.
Remco Bras
committed
Implement finding best splits for CEGB
Remco Bras
committed
Set proper defaults for cegb parameters.
Remco Bras
committed
Ensure sanity checks don't switch off CEGB.
Remco Bras
committed
Implement per-data-point feature penalties in CEGB.
Remco Bras
committed
Implement split penalty and remove unused parameters.
Remco Bras
committed
Merge changes from CEGB tree learner into serial tree learner
Remco Bras
committed
Represent features_used_in_data by a bitset, to reduce the memory overhead of CEGB, and add sanity checks for the lengths of the penalty vectors.
Remco Bras
committed