* Add scikit-learn tests

Goal is to pass scikit-learn's check_estimator() for XGBClassifier,
...

* Document LambdaMART objectives

* Distinguish between pairwise and listwise objectives

* Fix CRAN check by removing reference to std::cerr

* Mask tests that fail on 32-bit Windows R

* add interaction constraints

* enable both interaction and monotonic constraints at the same...

- previously, vec_ in DeviceShard wasn't updated on copy; as a result,
the shards continued t...

* Fix #3648: XGBClassifier.predict() should return margin scores when output_margin=True

* Fi...

* add R-interface for SHAP interactions

* update docs for new roxygen version

* Replaced std::vector with HostDeviceVector in MetaInfo and SparsePage.

- added distribution...

* Fixed issue 3605.

- https://github.com/dmlc/xgboost/issues/3605

* Fixed the bug in a bet...

* Add google test for a column sampling, restore metainfo tests

* Update metainfo test for vi...

* Remove thrust copy calls

* Fix histogram memory usage

* Cap extreme histogram memory us...

* Remove the use of old NDevices* functions.
* Use GPUSet in timer.h.

* add back train method but mark as deprecated

* add back train method but mark as deprecated...

* Document fact that GPU doesn't support external memory

* Document CUDA requirement

* Add XGBRanker to Python API doc

* Show inherited members of XGBRegressor in API doc, since ...

* add back train method but mark as deprecated

* add back train method but mark as deprecated...

* added xgbranker

* fixed predict method and ranking test

* reformatted code in accordance...

* Add travis sanitizers tests.

* Add gcc-7 in Travis.
* Add SANITIZER_PATH for CMake.
* Ena...

* Add JSON model dump functionality

* Fix lint

Close #3597.

for some reason this part was missing in master branch????

Fix #3587.

* Add basic Span class based on ISO++20.

* Use Span<Entry const> instead of Inst in SparsePag...

* Up versions

* Write release note for 0.80

This pull request amends the broken #3062 allow Spark 2.2 to work.

Please note this won't wor...

* Clarify multi-GPU training, binary wheels, Pandas integration

* Add a note about multi-GPU ...

* Adding Java/Scala doc build to Jenkins CI

* Deploy built doc to S3 bucket

* Build doc on...

* Refactor parts of fast histogram utilities

* Removed byte packing from column matrix

* Revert "Fix #3485, #3540: Don't use dropout for predicting test sets (#3556)"

This reverts ...

* add back train method but mark as deprecated

* add back train method but mark as deprecated...

Fix #3559
Fix #3562

During removal of FindGTest.cmake, also

* Fix gtest include dirs.
* Remove some blanks and u...

* Fix #3485, #3540: Don't use dropout for predicting test sets

Dropout (for DART) should only...

* Fix #3545: XGDMatrixCreateFromCSCEx silently discards empty trailing rows

Description: The ...

The base margin will need to have length `[num_class] * [number of data points]`.
Otherwise, th...

* Fix #3402: wrong fid crashes distributed algorithm

The bug was introduced by the recent DMa...

* add back train method but mark as deprecated

* add back train method but mark as deprecated...

* __CUDA__ is defined in clang. Making the change won't make clang
compile xgboost, but syntax ...

This bring many goodies, including:

* Ability to specify delimiter and weight_column for CSV ...

* add back train method but mark as deprecated

* add back train method but mark as deprecated...

* add back train method but mark as deprecated

* add back train method but mark as deprecated...

* Fix bug of using list(x) function when x is string

list('abcdcba') = ['a', 'b', 'c', 'd', '...

Move building dmlc-core before adding DMLC_LOG_CUSTOMIZE.

Fix #3520.

Just fixing a minor typo

* Expand histogram memory dynamically to prevent large allocations for large tree depths (e.g. >...

* Save max_delta_step as an extra attribute of Booster

Fixes #3509 and #3026, where `max_delt...

* Added finding quantiles on GPU.

- this includes datasets where weights are assigned to data...

Add `'total_gain'` and `'total_cover'` as possible `importance_type`
arguments to `Booster.get_...

* Added configuration for python into .editorconfig

* Fixed forgotten change in the number of...

* Improved library loading a bit

* Fixed indentation.

* Fixes according to the discussion
...

* Fix relpath in setup.py on Windows

Fixes #3480.

* Use only one lib file; use 4 space indent

* Change doc build to reST exclusively

* Rewrite Intro doc in reST; create toctree

* Updat...

Remove calls to thrust copy, fix indexing bug

* Fail GPU CI after test failure

* Fix GPU linear tests

* Reduced number of GPU tests to s...

* Allow multiple split constraints

* Replace RidgePenalty with ElasticNet

* Add test for c...

* Support single instance prediction.

* Address comments.

* Upgrading to NCCL2

* Part - II of NCCL2 upgradation

- Doc updates to build with nccl2
...

- Needed by the issue #3404
- as v9.1 doesn't have a nccl2 release

* python sklearn api: defaulting to best_ntree_limit if defined, otherwise current behaviour

...

* add back train method but mark as deprecated

* add back train method but mark as deprecated...

* Add regression test

* Add callback interface to re-direct console output

* Exempt TrackerLogger from custom loggi...