Separate out restricted and unrestricted tasks (#3736)

* Fix #3730: scikit-learn 0.20 compatibility fix

sklearn.cross_validation has been removed from scikit-learn 0.20,
so replace it with sklearn.model_selection

* Display test names for Python tests for clarity

.clang-tidy

@@ -1,10 +1,11 @@
-Checks: 'modernize-*,-modernize-make-*,-modernize-use-auto,-modernize-raw-string-literal,google-*,-google-default-arguments,-clang-diagnostic-#pragma-messages,readability-identifier-naming'
+Checks: 'modernize-*,-modernize-make-*,-modernize-raw-string-literal,google-*,-google-default-arguments,-clang-diagnostic-#pragma-messages,readability-identifier-naming'
 CheckOptions:
   - { key: readability-identifier-naming.ClassCase,           value: CamelCase  }
   - { key: readability-identifier-naming.StructCase,          value: CamelCase  }
   - { key: readability-identifier-naming.TypeAliasCase,          value: CamelCase  }
   - { key: readability-identifier-naming.TypedefCase,          value: CamelCase  }
   - { key: readability-identifier-naming.TypeTemplateParameterCase,          value: CamelCase  }
+  - { key: readability-identifier-naming.LocalVariableCase,          value: lower_case  }
   - { key: readability-identifier-naming.MemberCase,          value: lower_case  }
   - { key: readability-identifier-naming.PrivateMemberSuffix,           value: '_'  }
   - { key: readability-identifier-naming.ProtectedMemberSuffix,           value: '_'  }

.editorconfig

@@ -1,11 +0,0 @@
-root = true
-
-[*]
-charset=utf-8
-indent_style = space
-indent_size = 2
-insert_final_newline = true
-
-[*.py]
-indent_style = space
-indent_size = 4

.github/ISSUE_TEMPLATE.md

@@ -1,7 +0,0 @@
-Thanks for participating in the XGBoost community! We use https://discuss.xgboost.ai for any general usage questions and discussions. The issue tracker is used for actionable items such as feature proposals discussion, roadmaps, and bug tracking.  You are always welcomed to post on the forum first :)
-
-Issues that are inactive for a period of time may get closed. We adopt this policy so that we won't lose track of actionable issues that may fall at the bottom of the pile. Feel free to reopen a new one if you feel there is an additional problem that needs attention when an old one gets closed.
-
-For bug reports, to help the developer act on the issues, please include a description of your environment, preferably a minimum script to reproduce the problem.
-
-For feature proposals, list clear, small actionable items so we can track the progress of the change.

.github/lock.yml

@@ -1,32 +0,0 @@
-# Configuration for lock-threads - https://github.com/dessant/lock-threads
-
-# Number of days of inactivity before a closed issue or pull request is locked
-daysUntilLock: 90
-
-# Issues and pull requests with these labels will not be locked. Set to `[]` to disable
-exemptLabels:
-  - feature-request
-
-# Label to add before locking, such as `outdated`. Set to `false` to disable
-lockLabel: false
-
-# Comment to post before locking. Set to `false` to disable
-lockComment: false
-
-# Assign `resolved` as the reason for locking. Set to `false` to disable
-setLockReason: true
-
-# Limit to only `issues` or `pulls`
-# only: issues
-
-# Optionally, specify configuration settings just for `issues` or `pulls`
-# issues:
-#   exemptLabels:
-#     - help-wanted
-#   lockLabel: outdated
-
-# pulls:
-#   daysUntilLock: 30
-
-# Repository to extend settings from
-# _extends: repo

.gitignore

@@ -91,4 +91,3 @@ lib/
 metastore_db
 
 plugin/updater_gpu/test/cpp/data
-/include/xgboost/build_config.h

.gitmodules

@@ -4,6 +4,9 @@
 [submodule "rabit"]
 	path = rabit
 	url = https://github.com/dmlc/rabit
+[submodule "nccl"]
+	path = nccl
+	url = https://github.com/dmlc/nccl
 [submodule "cub"]
 	path = cub
 	url = https://github.com/NVlabs/cub

.travis.yml

@@ -6,7 +6,9 @@ os:
   - linux
   - osx
 
-osx_image: xcode9.3
+osx_image: xcode8
+
+group: deprecated-2017Q4
 
 # Use Build Matrix to do lint and build seperately
 env:
@@ -24,10 +26,6 @@ env:
     - TASK=cmake_test
     # c++ test
     - TASK=cpp_test
-    # distributed test
-    - TASK=distributed_test
-    # address sanitizer test
-    - TASK=sanitizer_test
 
 matrix:
   exclude:
@@ -41,10 +39,6 @@ matrix:
       env: TASK=python_lightweight_test
     - os: osx
       env: TASK=cpp_test
-    - os: osx
-      env: TASK=distributed_test
-    - os: osx
-      env: TASK=sanitizer_test
 
 # dependent apt packages
 addons:
@@ -64,13 +58,6 @@ addons:
       - graphviz
       - gcc-4.8
       - g++-4.8
-      - gcc-7
-      - g++-7
-  homebrew:
-    packages:
-      - gcc@7
-      - graphviz
-    update: true
 
 before_install:
   - source dmlc-core/scripts/travis/travis_setup_env.sh

CMakeLists.txt

@@ -8,31 +8,19 @@ set_default_configuration_release()
 msvc_use_static_runtime()
 
 # Options
-## GPUs
-option(USE_CUDA  "Build with GPU acceleration" OFF)
-option(USE_NCCL  "Build with multiple GPUs support" OFF)
+option(USE_CUDA  "Build with GPU acceleration") 
+option(USE_AVX  "Build with AVX instructions. May not produce identical results due to approximate math." OFF) 
+option(USE_NCCL "Build using NCCL for multi-GPU. Also requires USE_CUDA") 
+option(JVM_BINDINGS "Build JVM bindings" OFF)
+option(GOOGLE_TEST "Build google tests" OFF)
+option(R_LIB "Build shared library for R package" OFF)
 set(GPU_COMPUTE_VER "" CACHE STRING
   "Space separated list of compute versions to be built against, e.g. '35 61'")
 
-## Bindings
-option(JVM_BINDINGS "Build JVM bindings" OFF)
-option(R_LIB "Build shared library for R package" OFF)
-
-## Devs
-option(USE_SANITIZER "Use santizer flags" OFF)
-option(SANITIZER_PATH "Path to sanitizes.")
-set(ENABLED_SANITIZERS "address" "leak" CACHE STRING
-  "Semicolon separated list of sanitizer names. E.g 'address;leak'. Supported sanitizers are
-address, leak and thread.")
-option(GOOGLE_TEST "Build google tests" OFF)
-
-# Plugins
-option(PLUGIN_LZ4 "Build lz4 plugin" OFF)
-option(PLUGIN_DENSE_PARSER "Build dense parser plugin" OFF)
-
 # Deprecation warning
-if(USE_AVX)
-  message(WARNING "The option 'USE_AVX' is deprecated as experimental AVX features have been removed from xgboost.")
+if(PLUGIN_UPDATER_GPU)
+  set(USE_CUDA ON)
+  message(WARNING "The option 'PLUGIN_UPDATER_GPU' is deprecated. Set 'USE_CUDA' instead.")
 endif()
 
 # Compiler flags
@@ -51,42 +39,17 @@ else()
   # Performance
   set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -funroll-loops")
 endif()
-if(WIN32 AND MINGW)
-  set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -static-libstdc++")
+
+# AVX
+if(USE_AVX)
+  if(MSVC)
+    set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} /arch:AVX")
+  else()
+    set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mavx")
+  endif()
+  add_definitions(-DXGBOOST_USE_AVX)
 endif()
 
-# Check existence of software pre-fetching
-include(CheckCXXSourceCompiles)
-check_cxx_source_compiles("
-#include <xmmintrin.h>
-int main() {
-  char data = 0;
-  const char* address = &data;
-  _mm_prefetch(address, _MM_HINT_NTA);
-  return 0;
-}
-" XGBOOST_MM_PREFETCH_PRESENT)
-check_cxx_source_compiles("
-int main() {
-  char data = 0;
-  const char* address = &data;
-  __builtin_prefetch(address, 0, 0);
-  return 0;
-}
-" XGBOOST_BUILTIN_PREFETCH_PRESENT)
-
-# Sanitizer
-if(USE_SANITIZER)
-  include(cmake/Sanitizer.cmake)
-  enable_sanitizers("${ENABLED_SANITIZERS}")
-endif(USE_SANITIZER)
-
-# dmlc-core
-add_subdirectory(dmlc-core)
-set(LINK_LIBRARIES dmlc rabit)
-
-# enable custom logging
-add_definitions(-DDMLC_LOG_CUSTOMIZE=1)
 
 # compiled code customizations for R package
 if(R_LIB)
@@ -101,20 +64,13 @@ if(R_LIB)
   )
 endif()
 
-# Gather source files
 include_directories (
     ${PROJECT_SOURCE_DIR}/include
     ${PROJECT_SOURCE_DIR}/dmlc-core/include
     ${PROJECT_SOURCE_DIR}/rabit/include
 )
 
-# Generate configurable header
-set(CMAKE_LOCAL "${PROJECT_SOURCE_DIR}/cmake")
-set(INCLUDE_ROOT "${PROJECT_SOURCE_DIR}/include")
-message(STATUS "${CMAKE_LOCAL}/build_config.h.in -> ${INCLUDE_ROOT}/xgboost/build_config.h")
-configure_file("${CMAKE_LOCAL}/build_config.h.in" "${INCLUDE_ROOT}/xgboost/build_config.h")
-
-file(GLOB_RECURSE SOURCES
+file(GLOB_RECURSE SOURCES 
     src/*.cc
     src/*.h
     include/*.h
@@ -128,17 +84,8 @@ file(GLOB_RECURSE CUDA_SOURCES
     src/*.cuh
 )
 
-# Add plugins to source files
-if(PLUGIN_LZ4)
-  list(APPEND SOURCES plugin/lz4/sparse_page_lz4_format.cc)
-  link_libraries(lz4)
-endif()
-if(PLUGIN_DENSE_PARSER)
-  list(APPEND SOURCES plugin/dense_parser/dense_libsvm.cc)
-endif()
-
 # rabit
-# TODO: Use CMakeLists.txt from rabit.
+# TODO: Create rabit cmakelists.txt
 set(RABIT_SOURCES
     rabit/src/allreduce_base.cc
     rabit/src/allreduce_robust.cc
@@ -149,7 +96,6 @@ set(RABIT_EMPTY_SOURCES
     rabit/src/engine_empty.cc
     rabit/src/c_api.cc
 )
-
 if(MINGW OR R_LIB)
   # build a dummy rabit library
   add_library(rabit STATIC ${RABIT_EMPTY_SOURCES})
@@ -157,21 +103,22 @@ else()
   add_library(rabit STATIC ${RABIT_SOURCES})
 endif()
 
-if (GENERATE_COMPILATION_DATABASE)
-  set(CMAKE_EXPORT_COMPILE_COMMANDS ON)
-endif (GENERATE_COMPILATION_DATABASE)
 
-if(USE_CUDA AND (NOT GENERATE_COMPILATION_DATABASE))
+# dmlc-core
+add_subdirectory(dmlc-core)
+set(LINK_LIBRARIES dmlc rabit)
+
+
+if(USE_CUDA)
   find_package(CUDA 8.0 REQUIRED)
   cmake_minimum_required(VERSION 3.5)
 
   add_definitions(-DXGBOOST_USE_CUDA)
-
+  
   include_directories(cub)
 
   if(USE_NCCL)
-    find_package(Nccl REQUIRED)
-    cuda_include_directories(${NCCL_INCLUDE_DIR})
+    include_directories(nccl/src)
     add_definitions(-DXGBOOST_USE_NCCL)
   endif()
 
@@ -184,46 +131,16 @@ if(USE_CUDA AND (NOT GENERATE_COMPILATION_DATABASE))
     set(CUDA_NVCC_FLAGS "${CUDA_NVCC_FLAGS};-Xcompiler -fPIC; -Xcompiler -Werror; -std=c++11")
   endif()
 
+  if(USE_NCCL)
+    add_subdirectory(nccl)
+  endif()
+
   cuda_add_library(gpuxgboost ${CUDA_SOURCES} STATIC)
-
+  
   if(USE_NCCL)
-    link_directories(${NCCL_LIBRARY})
-    target_link_libraries(gpuxgboost ${NCCL_LIB_NAME})
+    target_link_libraries(gpuxgboost nccl)
   endif()
-  list(APPEND LINK_LIBRARIES gpuxgboost)
-
-elseif (USE_CUDA AND GENERATE_COMPILATION_DATABASE)
-  # Enable CUDA language to generate a compilation database.
-  cmake_minimum_required(VERSION 3.8)
-
-  find_package(CUDA 8.0 REQUIRED)
-  enable_language(CUDA)
-  set(CMAKE_CUDA_COMPILER clang++)
-  set(CUDA_SEPARABLE_COMPILATION ON)
-  if (NOT CLANG_CUDA_GENCODE)
-    set(CLANG_CUDA_GENCODE "--cuda-gpu-arch=sm_35")
-  endif (NOT CLANG_CUDA_GENCODE)
-  set(CMAKE_CUDA_FLAGS " -Wno-deprecated  ${CLANG_CUDA_GENCODE} -fPIC ${GENCODE} -std=c++11 -x cuda")
-  message(STATUS "CMAKE_CUDA_FLAGS: ${CMAKE_CUDA_FLAGS}")
-
-  add_library(gpuxgboost STATIC ${CUDA_SOURCES})
-
-  if(USE_NCCL)
-    find_package(Nccl REQUIRED)
-    target_include_directories(gpuxgboost PUBLIC ${NCCL_INCLUDE_DIR})
-    target_compile_definitions(gpuxgboost PUBLIC -DXGBOOST_USE_NCCL)
-    target_link_libraries(gpuxgboost PUBLIC ${NCCL_LIB_NAME})
-  endif()
-
-  target_compile_definitions(gpuxgboost PUBLIC -DXGBOOST_USE_CUDA)
-  # A hack for CMake to make arguments valid for clang++
-  string(REPLACE "-x cu" "-x cuda" CMAKE_CUDA_COMPILE_PTX_COMPILATION
-    ${CMAKE_CUDA_COMPILE_PTX_COMPILATION})
-  string(REPLACE "-x cu" "-x cuda" CMAKE_CUDA_COMPILE_WHOLE_COMPILATION
-    ${CMAKE_CUDA_COMPILE_WHOLE_COMPILATION})
-  string(REPLACE "-x cu" "-x cuda" CMAKE_CUDA_COMPILE_SEPARABLE_COMPILATION
-    ${CMAKE_CUDA_COMPILE_SEPARABLE_COMPILATION})
-  target_include_directories(gpuxgboost PUBLIC cub)
+  list(APPEND LINK_LIBRARIES gpuxgboost) 
 endif()
 
 
@@ -239,6 +156,7 @@ endif()
 
 add_library(objxgboost OBJECT ${SOURCES})
 
+
 # building shared library for R package
 if(R_LIB)
   find_package(LibR REQUIRED)
@@ -246,25 +164,22 @@ if(R_LIB)
   list(APPEND LINK_LIBRARIES "${LIBR_CORE_LIBRARY}")
   MESSAGE(STATUS "LIBR_CORE_LIBRARY " ${LIBR_CORE_LIBRARY})
 
-  # Shared library target for the R package
-  add_library(xgboost SHARED $<TARGET_OBJECTS:objxgboost>)
-  include_directories(xgboost
+  include_directories(
     "${LIBR_INCLUDE_DIRS}"
     "${PROJECT_SOURCE_DIR}"
   )
 
+  # Shared library target for the R package
+  add_library(xgboost SHARED $<TARGET_OBJECTS:objxgboost>)
   target_link_libraries(xgboost ${LINK_LIBRARIES})
   # R uses no lib prefix in shared library names of its packages
   set_target_properties(xgboost PROPERTIES PREFIX "")
-  if(APPLE)
-    set_target_properties(xgboost PROPERTIES SUFFIX ".so")
-  endif()
 
   setup_rpackage_install_target(xgboost ${CMAKE_CURRENT_BINARY_DIR})
   # use a dummy location for any other remaining installs
   set(CMAKE_INSTALL_PREFIX "${CMAKE_CURRENT_BINARY_DIR}/dummy_inst")
 
-  # main targets: shared library & exe
+# main targets: shared library & exe
 else()
   # Executable
   add_executable(runxgboost $<TARGET_OBJECTS:objxgboost> src/cli_main.cc)
@@ -287,53 +202,41 @@ else()
   add_dependencies(xgboost runxgboost)
 endif()
 
+
 # JVM
 if(JVM_BINDINGS)
     find_package(JNI QUIET REQUIRED)
 
+    include_directories(${JNI_INCLUDE_DIRS} jvm-packages/xgboost4j/src/native)
+
     add_library(xgboost4j SHARED
-      $<TARGET_OBJECTS:objxgboost>
-      jvm-packages/xgboost4j/src/native/xgboost4j.cpp)
-    target_include_directories(xgboost4j
-      PRIVATE ${JNI_INCLUDE_DIRS}
-      PRIVATE jvm-packages/xgboost4j/src/native)
-    target_link_libraries(xgboost4j
-      ${LINK_LIBRARIES}
-      ${JAVA_JVM_LIBRARY})
+        $<TARGET_OBJECTS:objxgboost>
+        jvm-packages/xgboost4j/src/native/xgboost4j.cpp)
     set_output_directory(xgboost4j ${PROJECT_SOURCE_DIR}/lib)
+    target_link_libraries(xgboost4j
+        ${LINK_LIBRARIES}
+        ${JAVA_JVM_LIBRARY})
 endif()
 
 
 # Test
 if(GOOGLE_TEST)
-  enable_testing()
   find_package(GTest REQUIRED)
+  enable_testing()
 
   file(GLOB_RECURSE TEST_SOURCES "tests/cpp/*.cc")
   auto_source_group("${TEST_SOURCES}")
+  include_directories(${GTEST_INCLUDE_DIR})
 
-  if(USE_CUDA AND (NOT GENERATE_COMPILATION_DATABASE))
+  if(USE_CUDA)
     file(GLOB_RECURSE CUDA_TEST_SOURCES "tests/cpp/*.cu")
-    cuda_include_directories(${GTEST_INCLUDE_DIRS})
     cuda_compile(CUDA_TEST_OBJS ${CUDA_TEST_SOURCES})
-  elseif (USE_CUDA AND GENERATE_COMPILATION_DATABASE)
-    file(GLOB_RECURSE CUDA_TEST_SOURCES "tests/cpp/*.cu")
   else()
     set(CUDA_TEST_OBJS "")
   endif()
 
-  if (USE_CUDA AND GENERATE_COMPILATION_DATABASE)
-    add_executable(testxgboost ${TEST_SOURCES} ${CUDA_TEST_SOURCES}
-      $<TARGET_OBJECTS:objxgboost>)
-    target_include_directories(testxgboost PRIVATE cub)
-  else ()
-    add_executable(testxgboost ${TEST_SOURCES} ${CUDA_TEST_OBJS}
-      $<TARGET_OBJECTS:objxgboost>)
-  endif ()
-
+  add_executable(testxgboost ${TEST_SOURCES} ${CUDA_TEST_OBJS} $<TARGET_OBJECTS:objxgboost>)
   set_output_directory(testxgboost ${PROJECT_SOURCE_DIR})
-  target_include_directories(testxgboost
-    PRIVATE ${GTEST_INCLUDE_DIRS})
   target_link_libraries(testxgboost ${GTEST_LIBRARIES} ${LINK_LIBRARIES})
 
   add_test(TestXGBoost testxgboost)

CONTRIBUTORS.md

@@ -6,30 +6,21 @@ Committers
 ----------
 Committers are people who have made substantial contribution to the project and granted write access to the project.
 * [Tianqi Chen](https://github.com/tqchen), University of Washington
-  - Tianqi is a Ph.D. student working on large-scale machine learning. He is the creator of the project.
+  - Tianqi is a PhD working on large-scale machine learning, he is the creator of the project.
 * [Tong He](https://github.com/hetong007), Amazon AI
-  - Tong is an applied scientist in Amazon AI. He is the maintainer of XGBoost R package.
+  - Tong is an applied scientist in Amazon AI, he is the maintainer of xgboost R package.
 * [Vadim Khotilovich](https://github.com/khotilov)
   - Vadim contributes many improvements in R and core packages.
 * [Bing Xu](https://github.com/antinucleon)
-  - Bing is the original creator of XGBoost Python package and currently the maintainer of [XGBoost.jl](https://github.com/antinucleon/XGBoost.jl).
+  - Bing is the original creator of xgboost python package and currently the maintainer of [XGBoost.jl](https://github.com/antinucleon/XGBoost.jl).
 * [Michael Benesty](https://github.com/pommedeterresautee)
-  - Michael is a lawyer and data scientist in France. He is the creator of XGBoost interactive analysis module in R.
-* [Yuan Tang](https://github.com/terrytangyuan), Ant Financial
-  - Yuan is a software engineer in Ant Financial. He contributed mostly in R and Python packages.
-* [Nan Zhu](https://github.com/CodingCat), Uber
-  - Nan is a software engineer in Uber. He contributed mostly in JVM packages.
-* [Sergei Lebedev](https://github.com/superbobry), Criteo
-  - Sergei is a software engineer in Criteo. He contributed mostly in JVM packages.
-* [Hongliang Liu](https://github.com/phunterlau)
-* [Scott Lundberg](http://scottlundberg.com/), University of Washington
-  - Scott is a Ph.D. student at University of Washington. He is the creator of SHAP, a unified approach to explain the output of machine learning models such as decision tree ensembles. He also helps maintain the XGBoost Julia package.
-* [Rory Mitchell](https://github.com/RAMitchell), University of Waikato
-  - Rory is a Ph.D. student at University of Waikato. He is the original creator of the GPU training algorithms. He improved the CMake build system and continuous integration. 
-* [Hyunsu Cho](http://hyunsu-cho.io/), Amazon AI
-  - Hyunsu is an applied scientist in Amazon AI. He is the maintainer of the XGBoost Python package. He also manages the Jenkins continuous integration system (https://xgboost-ci.net/). He is the initial author of the CPU 'hist' updater. 
-* [Jiaming](https://github.com/trivialfis)
-  - Jiaming contributed to the GPU algorithms. He has also introduced new abstractions to improve the quality of the C++ codebase.
+  - Micheal is a lawyer, data scientist in France, he is the creator of xgboost interactive analysis module in R.
+* [Yuan Tang](https://github.com/terrytangyuan)
+  - Yuan is a data scientist in Chicago, US. He contributed mostly in R and Python packages.
+* [Nan Zhu](https://github.com/CodingCat)
+  - Nan is a software engineer in Microsoft. He contributed mostly in JVM packages.
+* [Sergei Lebedev](https://github.com/superbobry)
+  - Serget is a software engineer in Criteo. He contributed mostly in JVM packages.
 
 Become a Committer
 ------------------
@@ -45,25 +36,28 @@ List of Contributors
 * [Full List of Contributors](https://github.com/dmlc/xgboost/graphs/contributors)
   - To contributors: please add your name to the list when you submit a patch to the project:)
 * [Kailong Chen](https://github.com/kalenhaha)
-  - Kailong is an early contributor of XGBoost, he is creator of ranking objectives in XGBoost.
+  - Kailong is an early contributor of xgboost, he is creator of ranking objectives in xgboost.
 * [Skipper Seabold](https://github.com/jseabold)
-  - Skipper is the major contributor to the scikit-learn module of XGBoost.
+  - Skipper is the major contributor to the scikit-learn module of xgboost.
 * [Zygmunt Zając](https://github.com/zygmuntz)
   - Zygmunt is the master behind the early stopping feature frequently used by kagglers.
 * [Ajinkya Kale](https://github.com/ajkl)
 * [Boliang Chen](https://github.com/cblsjtu)
 * [Yangqing Men](https://github.com/yanqingmen)
-  - Yangqing is the creator of XGBoost java package.
+  - Yangqing is the creator of xgboost java package.
 * [Engpeng Yao](https://github.com/yepyao)
 * [Giulio](https://github.com/giuliohome)
-  - Giulio is the creator of Windows project of XGBoost
+  - Giulio is the creator of windows project of xgboost
 * [Jamie Hall](https://github.com/nerdcha)
-  - Jamie is the initial creator of XGBoost scikit-learn module.
+  - Jamie is the initial creator of xgboost sklearn module.
 * [Yen-Ying Lee](https://github.com/white1033)
 * [Masaaki Horikoshi](https://github.com/sinhrks)
-  - Masaaki is the initial creator of XGBoost Python plotting module.
+  - Masaaki is the initial creator of xgboost python plotting module.
+* [Hongliang Liu](https://github.com/phunterlau)
+* [Hyunsu Cho](http://hyunsu-cho.io/)
+  - Hyunsu is the maintainer of the XGBoost Python package. He is in charge of submitting the Python package to Python Package Index (PyPI). He is also the initial author of the CPU 'hist' updater.
 * [daiyl0320](https://github.com/daiyl0320)
-  - daiyl0320 contributed patch to XGBoost distributed version more robust, and scales stably on TB scale datasets.
+  - daiyl0320 contributed patch to xgboost distributed version more robust, and scales stably on TB scale datasets.
 * [Huayi Zhang](https://github.com/irachex)
 * [Johan Manders](https://github.com/johanmanders)
 * [yoori](https://github.com/yoori)
@@ -74,17 +68,8 @@ List of Contributors
 * [Alex Bain](https://github.com/convexquad)
 * [Baltazar Bieniek](https://github.com/bbieniek)
 * [Adam Pocock](https://github.com/Craigacp)
+* [Rory Mitchell](https://github.com/RAMitchell)
+  - Rory is the author of the GPU plugin and also contributed the cmake build system and windows continuous integration
 * [Gideon Whitehead](https://github.com/gaw89)
 * [Yi-Lin Juang](https://github.com/frankyjuang)
 * [Andrew Hannigan](https://github.com/andrewhannigan)
-* [Andy Adinets](https://github.com/canonizer)
-* [Henry Gouk](https://github.com/henrygouk)
-* [Pierre de Sahb](https://github.com/pdesahb)
-* [liuliang01](https://github.com/liuliang01)
-  - liuliang01 added support for the qid column for LibSVM input format. This makes ranking task easier in distributed setting.
-* [Andrew Thia](https://github.com/BlueTea88)
-  - Andrew Thia implemented feature interaction constraints
-* [Wei Tian](https://github.com/weitian)
-* [Chen Qin](https://github.com/chenqin)
-* [Sam Wilkinson](https://samwilkinson.io)
-* [Matthew Jones](https://github.com/mt-jones)

ISSUE_TEMPLATE.md

@@ -0,0 +1,44 @@
+For bugs or installation issues, please provide the following information.
+The more information you provide, the more easily we will be able to offer
+help and advice.
+
+## Environment info
+Operating System:
+
+Compiler:
+
+Package used (python/R/jvm/C++):
+
+`xgboost` version used:
+
+If installing from source, please provide 
+
+1. The commit hash (`git rev-parse HEAD`)
+2. Logs will be helpful (If logs are large, please upload as attachment).
+
+If you are using jvm package, please 
+
+1. add [jvm-packages] in the title to make it quickly be identified
+2. the gcc version and distribution
+
+If you are using python package, please provide
+
+1. The python version and distribution
+2. The command to install `xgboost` if you are not installing from source
+
+If you are using R package, please provide
+
+1. The R `sessionInfo()`
+2. The command to install `xgboost` if you are not installing from source
+
+## Steps to reproduce
+
+1.
+2.
+3.
+
+## What have you tried?
+
+1.
+2.
+3.

Jenkinsfile

@@ -14,8 +14,7 @@ def dockerRun = 'tests/ci_build/ci_build.sh'
 def utils
 
 def buildMatrix = [
-    [ "enabled": true,  "os" : "linux", "withGpu": true, "withNccl": true,  "withOmp": true, "pythonVersion": "2.7", "cudaVersion": "9.2", "multiGpu": true],
-    [ "enabled": true,  "os" : "linux", "withGpu": true, "withNccl": true,  "withOmp": true, "pythonVersion": "2.7", "cudaVersion": "9.2" ],
+    [ "enabled": true,  "os" : "linux", "withGpu": true, "withNccl": true,  "withOmp": true, "pythonVersion": "2.7", "cudaVersion": "9.1" ],
     [ "enabled": true,  "os" : "linux", "withGpu": true, "withNccl": true,  "withOmp": true, "pythonVersion": "2.7", "cudaVersion": "8.0" ],
     [ "enabled": true,  "os" : "linux", "withGpu": true, "withNccl": false, "withOmp": true, "pythonVersion": "2.7", "cudaVersion": "8.0" ],
 ]
@@ -53,7 +52,7 @@ pipeline {
                     parallel (buildMatrix.findAll{it['enabled']}.collectEntries{ c ->
                         def buildName = utils.getBuildName(c)
                         utils.buildFactory(buildName, c, false, this.&buildPlatformCmake)
-                    } + [ "clang-tidy" : { buildClangTidyJob() } ])
+                    })
                 }
             }
         }
@@ -68,60 +67,22 @@ def buildPlatformCmake(buildName, conf, nodeReq, dockerTarget) {
     // Destination dir for artifacts
     def distDir = "dist/${buildName}"
     def dockerArgs = ""
-    if (conf["withGpu"]) {
+    if(conf["withGpu"]){
         dockerArgs = "--build-arg CUDA_VERSION=" + conf["cudaVersion"]
     }
-    def test_suite = conf["withGpu"] ? (conf["multiGpu"] ? "mgpu" : "gpu") : "cpu"
     // Build node - this is returned result
-    retry(1) {
-        node(nodeReq) {
-            unstash name: 'srcs'
-            echo """
-            |===== XGBoost CMake build =====
-            |  dockerTarget: ${dockerTarget}
-            |  cmakeOpts   : ${opts}
-            |=========================
-            """.stripMargin('|')
-            // Invoke command inside docker
-            sh """
-            ${dockerRun} ${dockerTarget} ${dockerArgs} tests/ci_build/build_via_cmake.sh ${opts}
-            ${dockerRun} ${dockerTarget} ${dockerArgs} tests/ci_build/test_${test_suite}.sh
-            """
-            if (!conf["multiGpu"]) {
-                sh """
-                ${dockerRun} ${dockerTarget} ${dockerArgs} bash -c "cd python-package; rm -f dist/*; python setup.py bdist_wheel --universal"
-                rm -rf "${distDir}"; mkdir -p "${distDir}/py"
-                cp xgboost "${distDir}"
-                cp -r python-package/dist "${distDir}/py"
-                # Test the wheel for compatibility on a barebones CPU container
-                ${dockerRun} release ${dockerArgs} bash -c " \
-                    pip install --user python-package/dist/xgboost-*-none-any.whl && \
-		    pytest -v --fulltrace -s tests/python"
-                # Test the wheel for compatibility on CUDA 10.0 container
-                ${dockerRun} gpu --build-arg CUDA_VERSION=10.0 bash -c " \
-                    pip install --user python-package/dist/xgboost-*-none-any.whl && \
-		    pytest -v -s --fulltrace -m '(not mgpu) and (not slow)' tests/python-gpu"
-                """
-            }
-        }
-    }
-}
-
-/**
- * Run a clang-tidy job on a GPU machine
- */
-def buildClangTidyJob() {
-    def nodeReq = "linux && gpu && unrestricted"
     node(nodeReq) {
         unstash name: 'srcs'
-        echo "Running clang-tidy job..."
+        echo """
+        |===== XGBoost CMake build =====
+        |  dockerTarget: ${dockerTarget}
+        |  cmakeOpts   : ${opts}
+        |=========================
+        """.stripMargin('|')
         // Invoke command inside docker
-        // Install Google Test and Python yaml
-        dockerTarget = "clang_tidy"
-        dockerArgs = "--build-arg CUDA_VERSION=9.2"
         sh """
-        ${dockerRun} ${dockerTarget} ${dockerArgs} tests/ci_build/clang_tidy.sh
+        ${dockerRun} ${dockerTarget} ${dockerArgs} tests/ci_build/build_via_cmake.sh ${opts}
+        ${dockerRun} ${dockerTarget} ${dockerArgs} tests/ci_build/test_${dockerTarget}.sh
         """
-      }
-  }
-
+    }
+}

Jenkinsfile-restricted

@@ -13,10 +13,6 @@ def dockerRun = 'tests/ci_build/ci_build.sh'
 // Utility functions
 @Field
 def utils
-@Field
-def commit_id
-@Field
-def branch_name
 
 def buildMatrix = [
     [ "enabled": true,  "os" : "linux", "withGpu": true, "withNccl": true,  "withOmp": true, "pythonVersion": "2.7", "cudaVersion": "9.2" ],
@@ -46,28 +42,27 @@ pipeline {
                 script {
                     utils = load('tests/ci_build/jenkins_tools.Groovy')
                     utils.checkoutSrcs()
-                    commit_id = "${GIT_COMMIT}"
-                    branch_name = "${GIT_LOCAL_BRANCH}"
                 }
                 stash name: 'srcs', excludes: '.git/'
                 milestone label: 'Sources ready', ordinal: 1
             }
         }
         stage('Jenkins: Build doc') {
+            agent {
+                label 'linux && cpu && restricted'
+            }
             steps {
+                unstash name: 'srcs'
                 script {
-                    retry(1) {
-                        node('linux && cpu && restricted') {
-                            unstash name: 'srcs'
-                            echo 'Building doc...'
-                            dir ('jvm-packages') {
-                                sh "bash ./build_doc.sh ${commit_id}"
-                                archiveArtifacts artifacts: "${commit_id}.tar.bz2", allowEmptyArchive: true
-                                echo 'Deploying doc...'
-                                withAWS(credentials:'xgboost-doc-bucket') {
-                                    s3Upload file: "${commit_id}.tar.bz2", bucket: 'xgboost-docs', acl: 'PublicRead', path: "${branch_name}.tar.bz2"
-                                }
-                            }
+                    def commit_id = "${GIT_COMMIT}"
+                    def branch_name = "${GIT_LOCAL_BRANCH}"
+                    echo 'Building doc...'
+                    dir ('jvm-packages') {
+                        sh "bash ./build_doc.sh ${commit_id}"
+                        archiveArtifacts artifacts: "${commit_id}.tar.bz2", allowEmptyArchive: true
+                        echo 'Deploying doc...'
+                        withAWS(credentials:'xgboost-doc-bucket') {
+                            s3Upload file: "${commit_id}.tar.bz2", bucket: 'xgboost-docs', acl: 'PublicRead', path: "${branch_name}.tar.bz2"
                         }
                     }
                 }
@@ -99,25 +94,28 @@ def buildPlatformCmake(buildName, conf, nodeReq, dockerTarget) {
         dockerArgs = "--build-arg CUDA_VERSION=" + conf["cudaVersion"]
     }
     // Build node - this is returned result
-    retry(1) {
-        node(nodeReq) {
-            unstash name: 'srcs'
-            echo """
-            |===== XGBoost CMake build =====
-            |  dockerTarget: ${dockerTarget}
-            |  cmakeOpts   : ${opts}
-            |=========================
-            """.stripMargin('|')
-            // Invoke command inside docker
-            sh """
-            ${dockerRun} ${dockerTarget} ${dockerArgs} tests/ci_build/build_via_cmake.sh ${opts}
-            ${dockerRun} ${dockerTarget} ${dockerArgs} bash -c "cd python-package; rm -f dist/*; python setup.py bdist_wheel --universal"
-            rm -rf "${distDir}"; mkdir -p "${distDir}/py"
-            cp xgboost "${distDir}"
-            cp -r lib "${distDir}"
-            cp -r python-package/dist "${distDir}/py"
-            """
-            archiveArtifacts artifacts: "${distDir}/**/*.*", allowEmptyArchive: true
-        }
+    node(nodeReq) {
+        unstash name: 'srcs'
+        echo """
+        |===== XGBoost CMake build =====
+        |  dockerTarget: ${dockerTarget}
+        |  cmakeOpts   : ${opts}
+        |=========================
+        """.stripMargin('|')
+        // Invoke command inside docker
+        sh """
+        ${dockerRun} ${dockerTarget} ${dockerArgs} tests/ci_build/build_via_cmake.sh ${opts}
+        ${dockerRun} ${dockerTarget} ${dockerArgs} bash -c "cd python-package; rm -f dist/*; python setup.py bdist_wheel --universal"
+        rm -rf "${distDir}"; mkdir -p "${distDir}/py"
+        cp xgboost "${distDir}"
+        cp -r lib "${distDir}"
+        cp -r python-package/dist "${distDir}/py"
+        # Test the wheel for compatibility on a barebones CPU container
+        ${dockerRun} release ${dockerArgs} bash -c " \
+            auditwheel show xgboost-*-py2-none-any.whl
+            pip install --user python-package/dist/xgboost-*-none-any.whl && \
+            python -m nose tests/python"
+        """
+        archiveArtifacts artifacts: "${distDir}/**/*.*", allowEmptyArchive: true
     }
 }

LICENSE

@@ -1,201 +1,13 @@
-                                 Apache License
-                           Version 2.0, January 2004
-                        http://www.apache.org/licenses/
+Copyright (c) 2016 by Contributors
 
-   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
 
-   1. Definitions.
+   http://www.apache.org/licenses/LICENSE-2.0
 
-      "License" shall mean the terms and conditions for use, reproduction,
-      and distribution as defined by Sections 1 through 9 of this document.
-
-      "Licensor" shall mean the copyright owner or entity authorized by
-      the copyright owner that is granting the License.
-
-      "Legal Entity" shall mean the union of the acting entity and all
-      other entities that control, are controlled by, or are under common
-      control with that entity. For the purposes of this definition,
-      "control" means (i) the power, direct or indirect, to cause the
-      direction or management of such entity, whether by contract or
-      otherwise, or (ii) ownership of fifty percent (50%) or more of the
-      outstanding shares, or (iii) beneficial ownership of such entity.
-
-      "You" (or "Your") shall mean an individual or Legal Entity
-      exercising permissions granted by this License.
-
-      "Source" form shall mean the preferred form for making modifications,
-      including but not limited to software source code, documentation
-      source, and configuration files.
-
-      "Object" form shall mean any form resulting from mechanical
-      transformation or translation of a Source form, including but
-      not limited to compiled object code, generated documentation,
-      and conversions to other media types.
-
-      "Work" shall mean the work of authorship, whether in Source or
-      Object form, made available under the License, as indicated by a
-      copyright notice that is included in or attached to the work
-      (an example is provided in the Appendix below).
-
-      "Derivative Works" shall mean any work, whether in Source or Object
-      form, that is based on (or derived from) the Work and for which the
-      editorial revisions, annotations, elaborations, or other modifications
-      represent, as a whole, an original work of authorship. For the purposes
-      of this License, Derivative Works shall not include works that remain
-      separable from, or merely link (or bind by name) to the interfaces of,
-      the Work and Derivative Works thereof.
-
-      "Contribution" shall mean any work of authorship, including
-      the original version of the Work and any modifications or additions
-      to that Work or Derivative Works thereof, that is intentionally
-      submitted to Licensor for inclusion in the Work by the copyright owner
-      or by an individual or Legal Entity authorized to submit on behalf of
-      the copyright owner. For the purposes of this definition, "submitted"
-      means any form of electronic, verbal, or written communication sent
-      to the Licensor or its representatives, including but not limited to
-      communication on electronic mailing lists, source code control systems,
-      and issue tracking systems that are managed by, or on behalf of, the
-      Licensor for the purpose of discussing and improving the Work, but
-      excluding communication that is conspicuously marked or otherwise
-      designated in writing by the copyright owner as "Not a Contribution."
-
-      "Contributor" shall mean Licensor and any individual or Legal Entity
-      on behalf of whom a Contribution has been received by Licensor and
-      subsequently incorporated within the Work.
-
-   2. Grant of Copyright License. Subject to the terms and conditions of
-      this License, each Contributor hereby grants to You a perpetual,
-      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
-      copyright license to reproduce, prepare Derivative Works of,
-      publicly display, publicly perform, sublicense, and distribute the
-      Work and such Derivative Works in Source or Object form.
-
-   3. Grant of Patent License. Subject to the terms and conditions of
-      this License, each Contributor hereby grants to You a perpetual,
-      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
-      (except as stated in this section) patent license to make, have made,
-      use, offer to sell, sell, import, and otherwise transfer the Work,
-      where such license applies only to those patent claims licensable
-      by such Contributor that are necessarily infringed by their
-      Contribution(s) alone or by combination of their Contribution(s)
-      with the Work to which such Contribution(s) was submitted. If You
-      institute patent litigation against any entity (including a
-      cross-claim or counterclaim in a lawsuit) alleging that the Work
-      or a Contribution incorporated within the Work constitutes direct
-      or contributory patent infringement, then any patent licenses
-      granted to You under this License for that Work shall terminate
-      as of the date such litigation is filed.
-
-   4. Redistribution. You may reproduce and distribute copies of the
-      Work or Derivative Works thereof in any medium, with or without
-      modifications, and in Source or Object form, provided that You
-      meet the following conditions:
-
-      (a) You must give any other recipients of the Work or
-          Derivative Works a copy of this License; and
-
-      (b) You must cause any modified files to carry prominent notices
-          stating that You changed the files; and
-
-      (c) You must retain, in the Source form of any Derivative Works
-          that You distribute, all copyright, patent, trademark, and
-          attribution notices from the Source form of the Work,
-          excluding those notices that do not pertain to any part of
-          the Derivative Works; and
-
-      (d) If the Work includes a "NOTICE" text file as part of its
-          distribution, then any Derivative Works that You distribute must
-          include a readable copy of the attribution notices contained
-          within such NOTICE file, excluding those notices that do not
-          pertain to any part of the Derivative Works, in at least one
-          of the following places: within a NOTICE text file distributed
-          as part of the Derivative Works; within the Source form or
-          documentation, if provided along with the Derivative Works; or,
-          within a display generated by the Derivative Works, if and
-          wherever such third-party notices normally appear. The contents
-          of the NOTICE file are for informational purposes only and
-          do not modify the License. You may add Your own attribution
-          notices within Derivative Works that You distribute, alongside
-          or as an addendum to the NOTICE text from the Work, provided
-          that such additional attribution notices cannot be construed
-          as modifying the License.
-
-      You may add Your own copyright statement to Your modifications and
-      may provide additional or different license terms and conditions
-      for use, reproduction, or distribution of Your modifications, or
-      for any such Derivative Works as a whole, provided Your use,
-      reproduction, and distribution of the Work otherwise complies with
-      the conditions stated in this License.
-
-   5. Submission of Contributions. Unless You explicitly state otherwise,
-      any Contribution intentionally submitted for inclusion in the Work
-      by You to the Licensor shall be under the terms and conditions of
-      this License, without any additional terms or conditions.
-      Notwithstanding the above, nothing herein shall supersede or modify
-      the terms of any separate license agreement you may have executed
-      with Licensor regarding such Contributions.
-
-   6. Trademarks. This License does not grant permission to use the trade
-      names, trademarks, service marks, or product names of the Licensor,
-      except as required for reasonable and customary use in describing the
-      origin of the Work and reproducing the content of the NOTICE file.
-
-   7. Disclaimer of Warranty. Unless required by applicable law or
-      agreed to in writing, Licensor provides the Work (and each
-      Contributor provides its Contributions) on an "AS IS" BASIS,
-      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
-      implied, including, without limitation, any warranties or conditions
-      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
-      PARTICULAR PURPOSE. You are solely responsible for determining the
-      appropriateness of using or redistributing the Work and assume any
-      risks associated with Your exercise of permissions under this License.
-
-   8. Limitation of Liability. In no event and under no legal theory,
-      whether in tort (including negligence), contract, or otherwise,
-      unless required by applicable law (such as deliberate and grossly
-      negligent acts) or agreed to in writing, shall any Contributor be
-      liable to You for damages, including any direct, indirect, special,
-      incidental, or consequential damages of any character arising as a
-      result of this License or out of the use or inability to use the
-      Work (including but not limited to damages for loss of goodwill,
-      work stoppage, computer failure or malfunction, or any and all
-      other commercial damages or losses), even if such Contributor
-      has been advised of the possibility of such damages.
-
-   9. Accepting Warranty or Additional Liability. While redistributing
-      the Work or Derivative Works thereof, You may choose to offer,
-      and charge a fee for, acceptance of support, warranty, indemnity,
-      or other liability obligations and/or rights consistent with this
-      License. However, in accepting such obligations, You may act only
-      on Your own behalf and on Your sole responsibility, not on behalf
-      of any other Contributor, and only if You agree to indemnify,
-      defend, and hold each Contributor harmless for any liability
-      incurred by, or claims asserted against, such Contributor by reason
-      of your accepting any such warranty or additional liability.
-
-   END OF TERMS AND CONDITIONS
-
-   APPENDIX: How to apply the Apache License to your work.
-
-      To apply the Apache License to your work, attach the following
-      boilerplate notice, with the fields enclosed by brackets "{}"
-      replaced with your own identifying information. (Don't include
-      the brackets!)  The text should be enclosed in the appropriate
-      comment syntax for the file format. We also recommend that a
-      file or class name and description of purpose be included on the
-      same "printed page" as the copyright notice for easier
-      identification within third-party archives.
-
-   Copyright (c) 2018 by Contributors
-
-   Licensed under the Apache License, Version 2.0 (the "License");
-   you may not use this file except in compliance with the License.
-   You may obtain a copy of the License at
-
-       http://www.apache.org/licenses/LICENSE-2.0
-
-   Unless required by applicable law or agreed to in writing, software
-   distributed under the License is distributed on an "AS IS" BASIS,
-   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-   See the License for the specific language governing permissions and
-   limitations under the License.
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License.

Makefile

@@ -68,7 +68,7 @@ endif
 endif
 
 export LDFLAGS= -pthread -lm $(ADD_LDFLAGS) $(DMLC_LDFLAGS) $(PLUGIN_LDFLAGS)
-export CFLAGS= -DDMLC_LOG_CUSTOMIZE=1 -std=c++11 -Wall -Wno-unknown-pragmas -Iinclude $(ADD_CFLAGS) $(PLUGIN_CFLAGS)
+export CFLAGS=  -std=c++11 -Wall -Wno-unknown-pragmas -Iinclude $(ADD_CFLAGS) $(PLUGIN_CFLAGS)
 CFLAGS += -I$(DMLC_CORE)/include -I$(RABIT)/include -I$(GTEST_PATH)/include
 #java include path
 export JAVAINCFLAGS = -I${JAVA_HOME}/include -I./java
@@ -260,8 +260,7 @@ Rpack: clean_all
 	cp ./LICENSE xgboost
 	cat R-package/src/Makevars.in|sed '2s/.*/PKGROOT=./' | sed '3s/.*/ENABLE_STD_THREAD=0/' > xgboost/src/Makevars.in
 	cp xgboost/src/Makevars.in xgboost/src/Makevars.win
-	sed -i -e 's/@OPENMP_CXXFLAGS@/$$\(SHLIB_OPENMP_CXXFLAGS\)/g' xgboost/src/Makevars.win
-	sed -i -e 's/-pthread/$$\(SHLIB_PTHREAD_FLAGS\)/g' xgboost/src/Makevars.win
+	sed -i -e 's/@OPENMP_CXXFLAGS@/$$\(SHLIB_OPENMP_CFLAGS\)/g' xgboost/src/Makevars.win
 	bash R-package/remove_warning_suppression_pragma.sh
 	rm xgboost/remove_warning_suppression_pragma.sh
 

NEWS.md

@@ -3,383 +3,6 @@ XGBoost Change Log
 
 This file records the changes in xgboost library in reverse chronological order.
 
-## v0.82 (2019.03.03)
-This release is packed with many new features and bug fixes.
-
-### Roadmap: better performance scaling for multi-core CPUs (#3957)
-* Poor performance scaling of the `hist` algorithm for multi-core CPUs has been under investigation (#3810). #3957 marks an important step toward better performance scaling, by using software pre-fetching and replacing STL vectors with C-style arrays. Special thanks to @Laurae2 and @SmirnovEgorRu.
-* See #3810 for latest progress on this roadmap.
-
-### New feature: Distributed Fast Histogram Algorithm (`hist`) (#4011, #4102, #4140, #4128)
-* It is now possible to run the `hist` algorithm in distributed setting. Special thanks to @CodingCat. The benefits include:
-  1. Faster local computation via feature binning
-  2. Support for monotonic constraints and feature interaction constraints
-  3. Simpler codebase than `approx`, allowing for future improvement
-* Depth-wise tree growing is now performed in a separate code path, so that cross-node syncronization is performed only once per level.
-
-### New feature: Multi-Node, Multi-GPU training (#4095)
-* Distributed training is now able to utilize clusters equipped with NVIDIA GPUs. In particular, the rabit AllReduce layer will communicate GPU device information. Special thanks to @mt-jones, @RAMitchell, @rongou, @trivialfis, @canonizer, and @jeffdk.
-* Resource management systems will be able to assign a rank for each GPU in the cluster.
-* In Dask, users will be able to construct a collection of XGBoost processes over an inhomogeneous device cluster (i.e. workers with different number and/or kinds of GPUs).
-
-### New feature: Multiple validation datasets in XGBoost4J-Spark (#3904, #3910)
-* You can now track the performance of the model during training with multiple evaluation datasets. By specifying `eval_sets` or call `setEvalSets` over a `XGBoostClassifier` or `XGBoostRegressor`, you can pass in multiple evaluation datasets typed as a `Map` from `String` to `DataFrame`. Special thanks to @CodingCat.
-* See the usage of multiple validation datasets [here](https://github.com/dmlc/xgboost/blob/0c1d5f1120c0a159f2567b267f0ec4ffadee00d0/jvm-packages/xgboost4j-example/src/main/scala/ml/dmlc/xgboost4j/scala/example/spark/SparkTraining.scala#L66-L78)
-
-### New feature: Additional metric functions for GPUs (#3952)
-* Element-wise metrics have been ported to GPU: `rmse`, `mae`, `logloss`, `poisson-nloglik`, `gamma-deviance`, `gamma-nloglik`, `error`, `tweedie-nloglik`. Special thanks to @trivialfis and @RAMitchell.
-* With supported metrics, XGBoost will select the correct devices based on your system and `n_gpus` parameter.
-
-### New feature: Column sampling at individual nodes (splits) (#3971)
-* Columns (features) can now be sampled at individual tree nodes, in addition to per-tree and per-level sampling. To enable per-node sampling, set `colsample_bynode` parameter, which represents the fraction of columns sampled at each node. This parameter is set to 1.0 by default (i.e. no sampling per node). Special thanks to @canonizer.
-* The `colsample_bynode` parameter works cumulatively with other `colsample_by*` parameters: for example, `{'colsample_bynode':0.5, 'colsample_bytree':0.5}` with 100 columns will give 25 features to choose from at each split.
-
-### Major API change: consistent logging level via `verbosity` (#3982, #4002, #4138)
-* XGBoost now allows fine-grained control over logging. You can set `verbosity` to 0 (silent), 1 (warning), 2 (info), and 3 (debug). This is useful for controlling the amount of logging outputs. Special thanks to @trivialfis.
-* Parameters `silent` and `debug_verbose` are now deprecated.
-* Note: Sometimes XGBoost tries to change configurations based on heuristics, which is displayed as warning message.  If there's unexpected behaviour, please try to increase value of verbosity.
-
-### Major bug fix: external memory (#4040, #4193)
-* Clarify object ownership in multi-threaded prefetcher, to avoid memory error.
-* Correctly merge two column batches (which uses [CSC layout](https://en.wikipedia.org/wiki/Sparse_matrix#Compressed_sparse_column_(CSC_or_CCS))).
-* Add unit tests for external memory.
-* Special thanks to @trivialfis and @hcho3.
-
-### Major bug fix: early stopping fixed in XGBoost4J and XGBoost4J-Spark (#3928, #4176)
-* Early stopping in XGBoost4J and XGBoost4J-Spark is now consistent with its counterpart in the Python package. Training stops if the current iteration is `earlyStoppingSteps` away from the best iteration. If there are multiple evaluation sets, only the last one is used to determinate early stop.
-* See the updated documentation [here](https://xgboost.readthedocs.io/en/release_0.82/jvm/xgboost4j_spark_tutorial.html#early-stopping)
-* Special thanks to @CodingCat, @yanboliang, and @mingyang.
-
-### Major bug fix: infrequent features should not crash distributed training (#4045)
-* For infrequently occuring features, some partitions may not get any instance. This scenario used to crash distributed training due to mal-formed ranges. The problem has now been fixed.
-* In practice, one-hot-encoded categorical variables tend to produce rare features, particularly when the cardinality is high.
-* Special thanks to @CodingCat.
-
-### Performance improvements
-* Faster, more space-efficient radix sorting in `gpu_hist` (#3895)
-* Subtraction trick in histogram calculation in `gpu_hist` (#3945)
-* More performant re-partition in XGBoost4J-Spark (#4049)
-
-### Bug-fixes
-* Fix semantics of `gpu_id` when running multiple XGBoost processes on a multi-GPU machine (#3851)
-* Fix page storage path for external memory on Windows (#3869)
-* Fix configuration setup so that DART utilizes GPU (#4024)
-* Eliminate NAN values from SHAP prediction (#3943)
-* Prevent empty quantile sketches in `hist` (#4155)
-* Enable running objectives with 0 GPU (#3878)
-* Parameters are no longer dependent on system locale (#3891, #3907)
-* Use consistent data type in the GPU coordinate descent code (#3917)
-* Remove undefined behavior in the CLI config parser on the ARM platform (#3976)
-* Initialize counters in GPU AllReduce (#3987)
-* Prevent deadlocks in GPU AllReduce (#4113)
-* Load correct values from sliced NumPy arrays (#4147, #4165)
-* Fix incorrect GPU device selection (#4161)
-* Make feature binning logic in `hist` aware of query groups when running a ranking task (#4115). For ranking task, query groups are weighted, not individual instances.
-* Generate correct C++ exception type for `LOG(FATAL)` macro (#4159)
-* Python package
-  - Python package should run on system without `PATH` environment variable (#3845)
-  - Fix `coef_` and `intercept_` signature to be compatible with `sklearn.RFECV` (#3873)
-  - Use UTF-8 encoding in Python package README, to support non-English locale (#3867)
-  - Add AUC-PR to list of metrics to maximize for early stopping (#3936)
-  - Allow loading pickles without `self.booster` attribute, for backward compatibility (#3938, #3944)
-  - White-list DART for feature importances (#4073)
-  - Update usage of [h2oai/datatable](https://github.com/h2oai/datatable) (#4123)
-* XGBoost4J-Spark
-  - Address scalability issue in prediction (#4033)
-  - Enforce the use of per-group weights for ranking task (#4118)
-  - Fix vector size of `rawPredictionCol` in `XGBoostClassificationModel` (#3932)
-  - More robust error handling in Spark tracker (#4046, #4108)
-  - Fix return type of `setEvalSets` (#4105)
-  - Return correct value of `getMaxLeaves` (#4114)
-
-### API changes
-* Add experimental parameter `single_precision_histogram` to use single-precision histograms for the `gpu_hist` algorithm (#3965)
-* Python package
-  - Add option to select type of feature importances in the scikit-learn inferface (#3876)
-  - Add `trees_to_df()` method to dump decision trees as Pandas data frame (#4153)
-  - Add options to control node shapes in the GraphViz plotting function (#3859)
-  - Add `xgb_model` option to `XGBClassifier`, to load previously saved model (#4092)
-  - Passing lists into `DMatrix` is now deprecated (#3970)
-* XGBoost4J
-  - Support multiple feature importance features (#3801)
-
-### Maintenance: Refactor C++ code for legibility and maintainability
-* Refactor `hist` algorithm code and add unit tests (#3836)
-* Minor refactoring of split evaluator in `gpu_hist` (#3889)
-* Removed unused leaf vector field in the tree model (#3989)
-* Simplify the tree representation by combining `TreeModel` and `RegTree` classes (#3995)
-* Simplify and harden tree expansion code (#4008, #4015)
-* De-duplicate parameter classes in the linear model algorithms (#4013)
-* Robust handling of ranges with C++20 span in `gpu_exact` and `gpu_coord_descent` (#4020, #4029)
-* Simplify tree training code (#3825). Also use Span class for robust handling of ranges.
-
-### Maintenance: testing, continuous integration, build system
-* Disallow `std::regex` since it's not supported by GCC 4.8.x (#3870)
-* Add multi-GPU tests for coordinate descent algorithm for linear models (#3893, #3974)
-* Enforce naming style in Python lint (#3896)
-* Refactor Python tests (#3897, #3901): Use pytest exclusively, display full trace upon failure
-* Address `DeprecationWarning` when using Python collections (#3909)
-* Use correct group for maven site plugin (#3937)
-* Jenkins CI is now using on-demand EC2 instances exclusively, due to unreliability of Spot instances (#3948)
-* Better GPU performance logging (#3945)
-* Fix GPU tests on machines with only 1 GPU (#4053)
-* Eliminate CRAN check warnings and notes (#3988)
-* Add unit tests for tree serialization (#3989)
-* Add unit tests for tree fitting functions in `hist` (#4155)
-* Add a unit test for `gpu_exact` algorithm (#4020)
-* Correct JVM CMake GPU flag (#4071)
-* Fix failing Travis CI on Mac (#4086)
-* Speed up Jenkins by not compiling CMake (#4099)
-* Analyze C++ and CUDA code using clang-tidy, as part of Jenkins CI pipeline (#4034)
-* Fix broken R test: Install Homebrew GCC (#4142)
-* Check for empty datasets in GPU unit tests (#4151)
-* Fix Windows compilation (#4139)
-* Comply with latest convention of cpplint (#4157)
-* Fix a unit test in `gpu_hist` (#4158)
-* Speed up data generation in Python tests (#4164)
-
-### Usability Improvements
-* Add link to [InfoWorld 2019 Technology of the Year Award](https://www.infoworld.com/article/3336072/application-development/infoworlds-2019-technology-of-the-year-award-winners.html) (#4116)
-* Remove outdated AWS YARN tutorial (#3885)
-* Document current limitation in number of features (#3886)
-* Remove unnecessary warning when `gblinear` is selected (#3888)
-* Document limitation of CSV parser: header not supported (#3934)
-* Log training parameters in XGBoost4J-Spark (#4091)
-* Clarify early stopping behavior in the scikit-learn interface (#3967)
-* Clarify behavior of `max_depth` parameter (#4078)
-* Revise Python docstrings for ranking task (#4121). In particular, weights must be per-group in learning-to-rank setting.
-* Document parameter `num_parallel_tree` (#4022)
-* Add Jenkins status badge (#4090)
-* Warn users against using internal functions of `Booster` object (#4066)
-* Reformat `benchmark_tree.py` to comply with Python style convention (#4126)
-* Clarify a comment in `objectiveTrait` (#4174)
-* Fix typos and broken links in documentation (#3890, #3872, #3902, #3919, #3975, #4027, #4156, #4167)
-
-### Acknowledgement
-**Contributors** (in no particular order): Jiaming Yuan (@trivialfis), Hyunsu Cho (@hcho3), Nan Zhu (@CodingCat), Rory Mitchell (@RAMitchell), Yanbo Liang (@yanboliang), Andy Adinets (@canonizer), Tong He (@hetong007), Yuan Tang (@terrytangyuan)
-
-**First-time Contributors** (in no particular order): Jelle Zijlstra (@JelleZijlstra), Jiacheng Xu (@jiachengxu), @ajing, Kashif Rasul (@kashif), @theycallhimavi, Joey Gao (@pjgao), Prabakaran Kumaresshan (@nixphix), Huafeng Wang (@huafengw), @lyxthe, Sam Wilkinson (@scwilkinson), Tatsuhito Kato (@stabacov), Shayak Banerjee (@shayakbanerjee), Kodi Arfer (@Kodiologist), @KyleLi1985, Egor Smirnov (@SmirnovEgorRu), @tmitanitky, Pasha Stetsenko (@st-pasha), Kenichi Nagahara (@keni-chi), Abhai Kollara Dilip (@abhaikollara), Patrick Ford (@pford221), @hshujuan, Matthew Jones (@mt-jones), Thejaswi Rao (@teju85), Adam November (@anovember)
-
-**First-time Reviewers** (in no particular order): Mingyang Hu (@mingyang), Theodore Vasiloudis (@thvasilo), Jakub Troszok (@troszok), Rong Ou (@rongou), @Denisevi4, Matthew Jones (@mt-jones), Jeff Kaplan (@jeffdk)
-
-## v0.81 (2018.11.04)
-### New feature: feature interaction constraints
-* Users are now able to control which features (independent variables) are allowed to interact by specifying feature interaction constraints (#3466).
-* [Tutorial](https://xgboost.readthedocs.io/en/release_0.81/tutorials/feature_interaction_constraint.html) is available, as well as [R](https://github.com/dmlc/xgboost/blob/9254c58e4dfff6a59dc0829a2ceb02e45ed17cd0/R-package/demo/interaction_constraints.R) and [Python](https://github.com/dmlc/xgboost/blob/9254c58e4dfff6a59dc0829a2ceb02e45ed17cd0/tests/python/test_interaction_constraints.py) examples.
-
-### New feature: learning to rank using scikit-learn interface
-* Learning to rank task is now available for the scikit-learn interface of the Python package (#3560, #3848). It is now possible to integrate the XGBoost ranking model into the scikit-learn learning pipeline.
-* Examples of using `XGBRanker` class is found at [demo/rank/rank_sklearn.py](https://github.com/dmlc/xgboost/blob/24a268a2e3cb17302db3d72da8f04016b7d352d9/demo/rank/rank_sklearn.py).
-
-### New feature: R interface for SHAP interactions
-* SHAP (SHapley Additive exPlanations) is a unified approach to explain the output of any machine learning model. Previously, this feature was only available from the Python package; now it is available from the R package as well (#3636).
-
-### New feature: GPU predictor now use multiple GPUs to predict
-* GPU predictor is now able to utilize multiple GPUs at once to accelerate prediction (#3738)
-
-### New feature: Scale distributed XGBoost to large-scale clusters
-* Fix OS file descriptor limit assertion error on large cluster (#3835, dmlc/rabit#73) by replacing `select()` based AllReduce/Broadcast with `poll()` based implementation.
-* Mitigate tracker "thundering herd" issue on large cluster. Add exponential backoff retry when workers connect to tracker.
-* With this change, we were able to scale to 1.5k executors on a 12 billion row dataset after some tweaks here and there.
-
-### New feature: Additional objective functions for GPUs
-* New objective functions ported to GPU: `hinge`, `multi:softmax`, `multi:softprob`, `count:poisson`, `reg:gamma`, `"reg:tweedie`.
-* With supported objectives, XGBoost will select the correct devices based on your system and `n_gpus` parameter.
-
-### Major bug fix: learning to rank with XGBoost4J-Spark
-* Previously, `repartitionForData` would shuffle data and lose ordering necessary for ranking task.
-* To fix this issue, data points within each RDD partition is explicitly group by their group (query session) IDs (#3654). Also handle empty RDD partition carefully (#3750).
-
-### Major bug fix: early stopping fixed in XGBoost4J-Spark
-* Earlier implementation of early stopping had incorrect semantics and didn't let users to specify direction for optimizing (maximize / minimize)
-* A parameter `maximize_evaluation_metrics` is defined so as to tell whether a metric should be maximized or minimized as part of early stopping criteria (#3808). Also early stopping now has correct semantics.
-
-### API changes
-* Column sampling by level (`colsample_bylevel`) is now functional for `hist` algorithm (#3635, #3862)
-* GPU tag `gpu:` for regression objectives are now deprecated. XGBoost will select the correct devices automatically (#3643)
-* Add `disable_default_eval_metric` parameter to disable default metric (#3606)
-* Experimental AVX support for gradient computation is removed (#3752)
-* XGBoost4J-Spark
-  - Add `rank:ndcg` and `rank:map` to supported objectives (#3697)
-* Python package
-  - Add `callbacks` argument to `fit()` function of sciki-learn API (#3682)
-  - Add `XGBRanker` to scikit-learn interface (#3560, #3848)
-  - Add `validate_features` argument to `predict()` function of scikit-learn API (#3653)
-  - Allow scikit-learn grid search over parameters specified as keyword arguments (#3791)
-  - Add `coef_` and `intercept_` as properties of scikit-learn wrapper (#3855). Some scikit-learn functions expect these properties.
-
-### Performance improvements
-* Address very high GPU memory usage for large data (#3635)
-* Fix performance regression within `EvaluateSplits()` of `gpu_hist` algorithm. (#3680)
-
-### Bug-fixes
-* Fix a problem in GPU quantile sketch with tiny instance weights. (#3628)
-* Fix copy constructor for `HostDeviceVectorImpl` to prevent dangling pointers (#3657)
-* Fix a bug in partitioned file loading (#3673)
-* Fixed an uninitialized pointer in `gpu_hist` (#3703)
-* Reshared data among GPUs when number of GPUs is changed (#3721)
-* Add back `max_delta_step` to split evaluation (#3668)
-* Do not round up integer thresholds for integer features in JSON dump (#3717)
-* Use `dmlc::TemporaryDirectory` to handle temporaries in cross-platform way (#3783)
-* Fix accuracy problem with `gpu_hist` when `min_child_weight` and `lambda` are set to 0 (#3793)
-* Make sure that `tree_method` parameter is recognized and not silently ignored (#3849)
-* XGBoost4J-Spark
-  - Make sure `thresholds` are considered when executing `predict()` method (#3577)
-  - Avoid losing precision when computing probabilities by converting to `Double` early (#3576)
-  - `getTreeLimit()` should return `Int` (#3602)
-  - Fix checkpoint serialization on HDFS (#3614)
-  - Throw `ControlThrowable` instead of `InterruptedException` so that it is properly re-thrown (#3632)
-  - Remove extraneous output to stdout (#3665)
-  - Allow specification of task type for custom objectives and evaluations (#3646)
-  - Fix distributed updater check (#3739)
-  - Fix issue when spark job execution thread cannot return before we execute `first()` (#3758)
-* Python package
-  - Fix accessing `DMatrix.handle` before it is set (#3599)
-  - `XGBClassifier.predict()` should return margin scores when `output_margin` is set to true (#3651)
-  - Early stopping callback should maximize metric of form `NDCG@n-` (#3685)
-  - Preserve feature names when slicing `DMatrix` (#3766)
-* R package
-  - Replace `nround` with `nrounds` to match actual parameter (#3592)
-  - Amend `xgb.createFolds` to handle classes of a single element (#3630)
-  - Fix buggy random generator and make `colsample_bytree` functional (#3781)
-
-### Maintenance: testing, continuous integration, build system
-* Add sanitizers tests to Travis CI (#3557)
-* Add NumPy, Matplotlib, Graphviz as requirements for doc build (#3669)
-* Comply with CRAN submission policy (#3660, #3728)
-* Remove copy-paste error in JVM test suite (#3692)
-* Disable flaky tests in `R-package/tests/testthat/test_update.R` (#3723)
-* Make Python tests compatible with scikit-learn 0.20 release (#3731)
-* Separate out restricted and unrestricted tasks, so that pull requests don't build downloadable artifacts (#3736)
-* Add multi-GPU unit test environment (#3741)
-* Allow plug-ins to be built by CMake (#3752)
-* Test wheel compatibility on CPU containers for pull requests (#3762)
-* Fix broken doc build due to Matplotlib 3.0 release (#3764)
-* Produce `xgboost.so` for XGBoost-R on Mac OSX, so that `make install` works (#3767)
-* Retry Jenkins CI tests up to 3 times to improve reliability (#3769, #3769, #3775, #3776, #3777)
-* Add basic unit tests for `gpu_hist` algorithm (#3785)
-* Fix Python environment for distributed unit tests (#3806)
-* Test wheels on CUDA 10.0 container for compatibility (#3838)
-* Fix JVM doc build (#3853)
-
-### Maintenance: Refactor C++ code for legibility and maintainability
-* Merge generic device helper functions into `GPUSet` class (#3626)
-* Re-factor column sampling logic into `ColumnSampler` class (#3635, #3637)
-* Replace `std::vector` with `HostDeviceVector` in `MetaInfo` and `SparsePage` (#3446)
-* Simplify `DMatrix` class (#3395)
-* De-duplicate CPU/GPU code using `Transform` class (#3643, #3751)
-* Remove obsoleted `QuantileHistMaker` class (#3761)
-* Remove obsoleted `NoConstraint` class (#3792)
-
-### Other Features
-* C++20-compliant Span class for safe pointer indexing (#3548, #3588)
-* Add helper functions to manipulate multiple GPU devices (#3693)
-* XGBoost4J-Spark
-  - Allow specifying host ip from the `xgboost-tracker.properties file` (#3833). This comes in handy when `hosts` files doesn't correctly define localhost.
-
-### Usability Improvements
-* Add reference to GitHub repository in `pom.xml` of JVM packages (#3589)
-* Add R demo of multi-class classification (#3695)
-* Document JSON dump functionality (#3600, #3603)
-* Document CUDA requirement and lack of external memory for GPU algorithms (#3624)
-* Document LambdaMART objectives, both pairwise and listwise (#3672)
-* Document `aucpr` evaluation metric (#3687)
-* Document gblinear parameters: `feature_selector` and `top_k` (#3780)
-* Add instructions for using MinGW-built XGBoost with Python. (#3774)
-* Removed nonexistent parameter `use_buffer` from documentation (#3610)
-* Update Python API doc to include all classes and members (#3619, #3682)
-* Fix typos and broken links in documentation (#3618, #3640, #3676, #3713, #3759, #3784, #3843, #3852)
-* Binary classification demo should produce LIBSVM with 0-based indexing (#3652)
-* Process data once for Python and CLI examples of learning to rank (#3666)
-* Include full text of Apache 2.0 license in the repository (#3698)
-* Save predictor parameters in model file (#3856)
-* JVM packages
-  - Let users specify feature names when calling `getModelDump` and `getFeatureScore` (#3733)
-  - Warn the user about the lack of over-the-wire encryption (#3667)
-  - Fix errors in examples (#3719)
-  - Document choice of trackers (#3831)
-  - Document that vanilla Apache Spark is required (#3854)
-* Python package
-  - Document that custom objective can't contain colon (:) (#3601)
-  - Show a better error message for failed library loading (#3690)
-  - Document that feature importance is unavailable for non-tree learners (#3765)
-  - Document behavior of `get_fscore()` for zero-importance features (#3763)
-  - Recommend pickling as the way to save `XGBClassifier` / `XGBRegressor` / `XGBRanker` (#3829)
-* R package
-  - Enlarge variable importance plot to make it more visible (#3820)
-
-### BREAKING CHANGES
-* External memory page files have changed, breaking backwards compatibility for temporary storage used during external memory training. This only affects external memory users upgrading their xgboost version - we recommend clearing all `*.page` files before resuming training. Model serialization is unaffected.
-
-### Known issues
-* Quantile sketcher fails to produce any quantile for some edge cases (#2943)
-* The `hist` algorithm leaks memory when used with learning rate decay callback (#3579)
-* Using custom evaluation funciton together with early stopping causes assertion failure in XGBoost4J-Spark (#3595)
-* Early stopping doesn't work with `gblinear` learner (#3789)
-* Label and weight vectors are not reshared upon the change in number of GPUs (#3794). To get around this issue, delete the `DMatrix` object and re-load.
-* The `DMatrix` Python objects are initialized with incorrect values when given array slices (#3841)
-* The `gpu_id` parameter is broken and not yet properly supported (#3850)
-
-### Acknowledgement
-**Contributors** (in no particular order): Hyunsu Cho (@hcho3), Jiaming Yuan (@trivialfis), Nan Zhu (@CodingCat), Rory Mitchell (@RAMitchell), Andy Adinets (@canonizer), Vadim Khotilovich (@khotilov), Sergei Lebedev (@superbobry)
-
-**First-time Contributors** (in no particular order): Matthew Tovbin (@tovbinm), Jakob Richter (@jakob-r), Grace Lam (@grace-lam), Grant W Schneider (@grantschneider), Andrew Thia (@BlueTea88), Sergei Chipiga (@schipiga), Joseph Bradley (@jkbradley), Chen Qin (@chenqin), Jerry Lin (@linjer), Dmitriy Rybalko (@rdtft), Michael Mui (@mmui), Takahiro Kojima (@515hikaru), Bruce Zhao (@BruceZhaoR), Wei Tian (@weitian), Saumya Bhatnagar (@Sam1301), Juzer Shakir (@JuzerShakir), Zhao Hang (@cleghom), Jonathan Friedman (@jontonsoup), Bruno Tremblay (@meztez), Boris Filippov (@frenzykryger), @Shiki-H, @mrgutkun, @gorogm, @htgeis, @jakehoare, @zengxy, @KOLANICH
-
-**First-time Reviewers** (in no particular order): Nikita Titov (@StrikerRUS), Xiangrui Meng (@mengxr), Nirmal Borah (@Nirmal-Neel)
-
-
-## v0.80 (2018.08.13)
-* **JVM packages received a major upgrade**: To consolidate the APIs and improve the user experience, we refactored the design of XGBoost4J-Spark in a significant manner. (#3387)
-  - Consolidated APIs: It is now much easier to integrate XGBoost models into a Spark ML pipeline. Users can control behaviors like output leaf prediction results by setting corresponding column names. Training is now more consistent with other Estimators in Spark MLLIB: there is now one single method `fit()` to train decision trees.
-  - Better user experience: we refactored the parameters relevant modules in XGBoost4J-Spark to provide both camel-case (Spark ML style) and underscore (XGBoost style) parameters
-  - A brand-new tutorial is [available](https://xgboost.readthedocs.io/en/release_0.80/jvm/xgboost4j_spark_tutorial.html) for XGBoost4J-Spark.
-  - Latest API documentation is now hosted at https://xgboost.readthedocs.io/.
-* XGBoost documentation now keeps track of multiple versions:
-  - Latest master: https://xgboost.readthedocs.io/en/latest
-  - 0.80 stable: https://xgboost.readthedocs.io/en/release_0.80
-  - 0.72 stable: https://xgboost.readthedocs.io/en/release_0.72
-* Support for per-group weights in ranking objective (#3379)
-* Fix inaccurate decimal parsing (#3546)
-* New functionality
-  - Query ID column support in LIBSVM data files (#2749). This is convenient for performing ranking task in distributed setting.
-  - Hinge loss for binary classification (`binary:hinge`) (#3477)
-  - Ability to specify delimiter and instance weight column for CSV files (#3546)
-  - Ability to use 1-based indexing instead of 0-based (#3546)
-* GPU support
-  - Quantile sketch, binning, and index compression are now performed on GPU, eliminating PCIe transfer for 'gpu_hist' algorithm (#3319, #3393)
-  - Upgrade to NCCL2 for multi-GPU training (#3404).
-  - Use shared memory atomics for faster training (#3384).
-  - Dynamically allocate GPU memory, to prevent large allocations for deep trees (#3519)
-  - Fix memory copy bug for large files (#3472)
-* Python package
-  - Importing data from Python datatable (#3272)
-  - Pre-built binary wheels available for 64-bit Linux and Windows (#3424, #3443)
-  - Add new importance measures 'total_gain', 'total_cover' (#3498)
-  - Sklearn API now supports saving and loading models (#3192)
-  - Arbitrary cross validation fold indices (#3353)
-  - `predict()` function in Sklearn API uses `best_ntree_limit` if available, to make early stopping easier to use (#3445)
-  - Informational messages are now directed to Python's `print()` rather than standard output (#3438). This way, messages appear inside Jupyter notebooks.
-* R package
-  - Oracle Solaris support, per CRAN policy (#3372)
-* JVM packages
-  - Single-instance prediction (#3464)
-  - Pre-built JARs are now available from Maven Central (#3401)
-  - Add NULL pointer check (#3021)
-  - Consider `spark.task.cpus` when controlling parallelism (#3530)
-  - Handle missing values in prediction (#3529)
-  - Eliminate outputs of `System.out` (#3572)
-* Refactored C++ DMatrix class for simplicity and de-duplication (#3301)
-* Refactored C++ histogram facilities (#3564)
-* Refactored constraints / regularization mechanism for split finding (#3335, #3429). Users may specify an elastic net (L2 + L1 regularization) on leaf weights as well as monotonic constraints on test nodes. The refactor will be useful for a future addition of feature interaction constraints.
-* Statically link `libstdc++` for MinGW32 (#3430)
-* Enable loading from `group`, `base_margin` and `weight` (see [here](http://xgboost.readthedocs.io/en/latest/tutorials/input_format.html#auxiliary-files-for-additional-information)) for Python, R, and JVM packages (#3431)
-* Fix model saving for `count:possion` so that `max_delta_step` doesn't get truncated (#3515)
-* Fix loading of sparse CSC matrix (#3553)
-* Fix incorrect handling of `base_score` parameter for Tweedie regression (#3295)
-
-## v0.72.1 (2018.07.08)
-This version is only applicable for the Python package. The content is identical to that of v0.72.
-
 ## v0.72 (2018.06.01)
 * Starting with this release, we plan to make a new release every two months. See #3252 for more details.
 * Fix a pathological behavior (near-zero second-order gradients) in multiclass objective (#3304)
@@ -498,7 +121,7 @@ This version is only applicable for the Python package. The content is identical
   - Compatibility fix for Python 2.6
   - Call `print_evaluation` callback at last iteration
   - Use appropriate integer types when calling native code, to prevent truncation and memory error
-  - Fix shared library loading on Mac OS X
+  - Fix shared library loading on Mac OS X 
 * R package:
   - New parameters:
     - `silent` in `xgb.DMatrix()`
@@ -539,7 +162,7 @@ This version is only applicable for the Python package. The content is identical
   - Support instance weights
   - Use `SparkParallelismTracker` to prevent jobs from hanging forever
   - Expose train-time evaluation metrics via `XGBoostModel.summary`
-  - Option to specify `host-ip` explicitly in the Rabit tracker
+  - Option to specify `host-ip` explicitly in the Rabit tracker 
 * Documentation
   - Better math notation for gradient boosting
   - Updated build instructions for Mac OS X

R-package/DESCRIPTION

@@ -1,8 +1,8 @@
 Package: xgboost
 Type: Package
 Title: Extreme Gradient Boosting
-Version: 0.82.0.1
-Date: 2019-03-11
+Version: 0.71.1
+Date: 2018-05-11
 Authors@R: c(
   person("Tianqi", "Chen", role = c("aut"),
          email = "tianqi.tchen@gmail.com"),
@@ -51,7 +51,6 @@ Suggests:
     Ckmeans.1d.dp (>= 3.3.1),
     vcd (>= 1.3),
     testthat,
-    lintr,
     igraph (>= 1.0.1)
 Depends:
     R (>= 3.3.0)
@@ -61,5 +60,5 @@ Imports:
     data.table (>= 1.9.6),
     magrittr (>= 1.5),
     stringi (>= 0.5.2)
-RoxygenNote: 6.1.0
+RoxygenNote: 6.0.1
 SystemRequirements: GNU make, C++11

R-package/R/callbacks.R

@@ -168,7 +168,7 @@ cb.evaluation.log <- function() {
 #' at the beginning of each iteration.
 #' 
 #' Note that when training is resumed from some previous model, and a function is used to 
-#' reset a parameter value, the \code{nrounds} argument in this function would be the 
+#' reset a parameter value, the \code{nround} argument in this function would be the 
 #' the number of boosting rounds in the current training.
 #'
 #' Callback function expects the following values to be set in its calling frame:

R-package/R/utils.R

@@ -74,19 +74,6 @@ check.booster.params <- function(params, ...) {
         params[['monotone_constraints']] = vec2str
   }
   
-  # interaction constraints parser (convert from list of column indices to string)
-  if (!is.null(params[['interaction_constraints']]) && 
-      typeof(params[['interaction_constraints']]) != "character"){
-    # check input class
-    if (class(params[['interaction_constraints']]) != 'list') stop('interaction_constraints should be class list')
-    if (!all(unique(sapply(params[['interaction_constraints']], class)) %in% c('numeric','integer'))) {
-      stop('interaction_constraints should be a list of numeric/integer vectors')
-    }
-
-    # recast parameter as string
-    interaction_constraints <- sapply(params[['interaction_constraints']], function(x) paste0('[', paste(x, collapse=','), ']'))
-    params[['interaction_constraints']] <- paste0('[', paste(interaction_constraints, collapse=','), ']')
-  }
   return(params)
 }
 
@@ -275,8 +262,7 @@ xgb.createFolds <- function(y, k = 10)
       ## add enough random integers to get  length(seqVector) == numInClass[i]
       if (numInClass[i] %% k > 0) seqVector <- c(seqVector, sample.int(k, numInClass[i] %% k))
       ## shuffle the integers for fold assignment and assign to this classes's data
-      ## seqVector[sample.int(length(seqVector))] is used to handle length(seqVector) == 1
-      foldVector[y == dimnames(numInClass)$y[i]] <- seqVector[sample.int(length(seqVector))]
+      foldVector[y == dimnames(numInClass)$y[i]] <- sample(seqVector)
     }
   } else {
     foldVector <- seq(along = y)

R-package/R/xgb.Booster.R

@@ -129,13 +129,11 @@ xgb.Booster.complete <- function(object, saveraw = TRUE) {
 #'        logistic regression would result in predictions for log-odds instead of probabilities.
 #' @param ntreelimit limit the number of model's trees or boosting iterations used in prediction (see Details).
 #'        It will use all the trees by default (\code{NULL} value).
-#' @param predleaf whether predict leaf index.
-#' @param predcontrib whether to return feature contributions to individual predictions (see Details).
+#' @param predleaf whether predict leaf index instead.
+#' @param predcontrib whether to return feature contributions to individual predictions instead (see Details).
 #' @param approxcontrib whether to use a fast approximation for feature contributions (see Details).
-#' @param predinteraction whether to return contributions of feature interactions to individual predictions (see Details).
 #' @param reshape whether to reshape the vector of predictions to a matrix form when there are several
-#'        prediction outputs per case. This option has no effect when either of predleaf, predcontrib,
-#'        or predinteraction flags is TRUE.
+#'        prediction outputs per case. This option has no effect when \code{predleaf = TRUE}.
 #' @param ... Parameters passed to \code{predict.xgb.Booster}
 #'
 #' @details
@@ -160,11 +158,6 @@ xgb.Booster.complete <- function(object, saveraw = TRUE) {
 #' Setting \code{approxcontrib = TRUE} approximates these values following the idea explained
 #' in \url{http://blog.datadive.net/interpreting-random-forests/}.
 #'
-#' With \code{predinteraction = TRUE}, SHAP values of contributions of interaction of each pair of features
-#' are computed. Note that this operation might be rather expensive in terms of compute and memory.
-#' Since it quadratically depends on the number of features, it is recommended to perfom selection
-#' of the most important features first. See below about the format of the returned results.
-#'
 #' @return
 #' For regression or binary classification, it returns a vector of length \code{nrows(newdata)}.
 #' For multiclass classification, either a \code{num_class * nrows(newdata)} vector or
@@ -180,14 +173,6 @@ xgb.Booster.complete <- function(object, saveraw = TRUE) {
 #' such a matrix. The contribution values are on the scale of untransformed margin
 #' (e.g., for binary classification would mean that the contributions are log-odds deviations from bias).
 #'
-#' When \code{predinteraction = TRUE} and it is not a multiclass setting, the output is a 3d array with
-#' dimensions \code{c(nrow, num_features + 1, num_features + 1)}. The off-diagonal (in the last two dimensions)
-#' elements represent different features interaction contributions. The array is symmetric WRT the last
-#' two dimensions. The "+ 1" columns corresponds to bias. Summing this array along the last dimension should
-#' produce practically the same result as predict with \code{predcontrib = TRUE}.
-#' For a multiclass case, a list of \code{num_class} elements is returned, where each element is
-#' such an array.
-#'
 #' @seealso
 #' \code{\link{xgb.train}}.
 #' 
@@ -284,8 +269,7 @@ xgb.Booster.complete <- function(object, saveraw = TRUE) {
 #' @rdname predict.xgb.Booster
 #' @export
 predict.xgb.Booster <- function(object, newdata, missing = NA, outputmargin = FALSE, ntreelimit = NULL,
-                                predleaf = FALSE, predcontrib = FALSE, approxcontrib = FALSE, predinteraction = FALSE,
-                                reshape = FALSE, ...) {
+                                predleaf = FALSE, predcontrib = FALSE, approxcontrib = FALSE, reshape = FALSE, ...) {
 
   object <- xgb.Booster.complete(object, saveraw = FALSE)
   if (!inherits(newdata, "xgb.DMatrix"))
@@ -301,8 +285,7 @@ predict.xgb.Booster <- function(object, newdata, missing = NA, outputmargin = FA
   if (ntreelimit < 0)
     stop("ntreelimit cannot be negative")
 
-  option <- 0L + 1L * as.logical(outputmargin) + 2L * as.logical(predleaf) + 4L * as.logical(predcontrib) +
-    8L * as.logical(approxcontrib) + 16L * as.logical(predinteraction)
+  option <- 0L + 1L * as.logical(outputmargin) + 2L * as.logical(predleaf) + 4L * as.logical(predcontrib) + 8L * as.logical(approxcontrib)
 
   ret <- .Call(XGBoosterPredict_R, object$handle, newdata, option[1], as.integer(ntreelimit))
 
@@ -322,28 +305,17 @@ predict.xgb.Booster <- function(object, newdata, missing = NA, outputmargin = FA
   } else if (predcontrib) {
     n_col1 <- ncol(newdata) + 1
     n_group <- npred_per_case / n_col1
-    cnames <- if (!is.null(colnames(newdata))) c(colnames(newdata), "BIAS") else NULL
+    dnames <- if (!is.null(colnames(newdata))) list(NULL, c(colnames(newdata), "BIAS")) else NULL
     ret <- if (n_ret == n_row) {
-      matrix(ret, ncol = 1, dimnames = list(NULL, cnames))
+      matrix(ret, ncol = 1, dimnames = dnames)
     } else if (n_group == 1) {
-      matrix(ret, nrow = n_row, byrow = TRUE, dimnames = list(NULL, cnames))
+      matrix(ret, nrow = n_row, byrow = TRUE, dimnames = dnames)
     } else {
-      arr <- array(ret, c(n_col1, n_group, n_row),
-                   dimnames = list(cnames, NULL, NULL)) %>% aperm(c(2,3,1)) # [group, row, col]
-      lapply(seq_len(n_group), function(g) arr[g,,])
-    }
-  } else if (predinteraction) {
-    n_col1 <- ncol(newdata) + 1
-    n_group <- npred_per_case / n_col1^2
-    cnames <- if (!is.null(colnames(newdata))) c(colnames(newdata), "BIAS") else NULL
-    ret <- if (n_ret == n_row) {
-      matrix(ret, ncol = 1, dimnames = list(NULL, cnames))
-    } else if (n_group == 1) {
-      array(ret, c(n_col1, n_col1, n_row), dimnames = list(cnames, cnames, NULL)) %>% aperm(c(3,1,2))
-    } else {
-      arr <- array(ret, c(n_col1, n_col1, n_group, n_row),
-                   dimnames = list(cnames, cnames, NULL, NULL)) %>% aperm(c(3,4,1,2)) # [group, row, col1, col2]
-      lapply(seq_len(n_group), function(g) arr[g,,,])
+      grp_mask <- rep(seq_len(n_col1), n_row) +
+        rep((seq_len(n_row) - 1) * n_col1 * n_group, each = n_col1)
+      lapply(seq_len(n_group), function(g) {
+        matrix(ret[grp_mask + n_col1 * (g - 1)], nrow = n_row, byrow = TRUE, dimnames = dnames)
+      })
     }
   } else if (reshape && npred_per_case > 1) {
     ret <- matrix(ret, nrow = n_row, byrow = TRUE)

R-package/R/xgb.create.features.R

@@ -52,9 +52,9 @@
 #' dtest <- xgb.DMatrix(data = agaricus.test$data, label = agaricus.test$label)
 #'
 #' param <- list(max_depth=2, eta=1, silent=1, objective='binary:logistic')
-#' nrounds = 4
+#' nround = 4
 #'
-#' bst = xgb.train(params = param, data = dtrain, nrounds = nrounds, nthread = 2)
+#' bst = xgb.train(params = param, data = dtrain, nrounds = nround, nthread = 2)
 #' 
 #' # Model accuracy without new features
 #' accuracy.before <- sum((predict(bst, agaricus.test$data) >= 0.5) == agaricus.test$label) /
@@ -68,7 +68,7 @@
 #' new.dtrain <- xgb.DMatrix(data = new.features.train, label = agaricus.train$label)
 #' new.dtest <- xgb.DMatrix(data = new.features.test, label = agaricus.test$label)
 #' watchlist <- list(train = new.dtrain)
-#' bst <- xgb.train(params = param, data = new.dtrain, nrounds = nrounds, nthread = 2)
+#' bst <- xgb.train(params = param, data = new.dtrain, nrounds = nround, nthread = 2)
 #' 
 #' # Model accuracy with new features
 #' accuracy.after <- sum((predict(bst, new.dtest) >= 0.5) == agaricus.test$label) /

R-package/R/xgb.ggplot.R

@@ -22,7 +22,7 @@ xgb.ggplot.importance <- function(importance_matrix = NULL, top_n = NULL, measur
 
   plot <-
     ggplot2::ggplot(importance_matrix, 
-                    ggplot2::aes(x = factor(Feature, levels = rev(Feature)), y = Importance, width = 0.5),
+                    ggplot2::aes(x = factor(Feature, levels = rev(Feature)), y = Importance, width = 0.05),
                     environment = environment()) + 
     ggplot2::geom_bar(ggplot2::aes(fill = Cluster), stat = "identity", position = "identity") + 
     ggplot2::coord_flip() + 

R-package/R/xgb.plot.deepness.R

@@ -27,7 +27,7 @@
 #' a tree's median absolute leaf weight changes through the iterations.
 #'
 #' This function was inspired by the blog post
-#' \url{https://github.com/aysent/random-forest-leaf-visualization}.
+#' \url{http://aysent.github.io/2015/11/08/random-forest-leaf-visualization.html}.
 #' 
 #' @return
 #' 

R-package/R/xgb.plot.shap.R

@@ -212,7 +212,6 @@ xgb.plot.shap <- function(data, shap_contrib = NULL, features = NULL, top_n = 1,
   }
   if (plot && which == "2d") {
     # TODO
-    warning("Bivariate plotting is currently not available.")
   }
   invisible(list(data = data, shap_contrib = shap_contrib))
 }

R-package/R/xgb.train.R

@@ -22,11 +22,10 @@
 #'   \item \code{gamma} minimum loss reduction required to make a further partition on a leaf node of the tree. the larger, the more conservative the algorithm will be. 
 #'   \item \code{max_depth} maximum depth of a tree. Default: 6
 #'   \item \code{min_child_weight} minimum sum of instance weight (hessian) needed in a child. If the tree partition step results in a leaf node with the sum of instance weight less than min_child_weight, then the building process will give up further partitioning. In linear regression mode, this simply corresponds to minimum number of instances needed to be in each node. The larger, the more conservative the algorithm will be. Default: 1
-#'   \item \code{subsample} subsample ratio of the training instance. Setting it to 0.5 means that xgboost randomly collected half of the data instances to grow trees and this will prevent overfitting. It makes computation shorter (because less data to analyse). It is advised to use this parameter with \code{eta} and increase \code{nrounds}. Default: 1 
+#'   \item \code{subsample} subsample ratio of the training instance. Setting it to 0.5 means that xgboost randomly collected half of the data instances to grow trees and this will prevent overfitting. It makes computation shorter (because less data to analyse). It is advised to use this parameter with \code{eta} and increase \code{nround}. Default: 1 
 #'   \item \code{colsample_bytree} subsample ratio of columns when constructing each tree. Default: 1
 #'   \item \code{num_parallel_tree} Experimental parameter. number of trees to grow per round. Useful to test Random Forest through Xgboost (set \code{colsample_bytree < 1}, \code{subsample  < 1}  and \code{round = 1}) accordingly. Default: 1
 #'   \item \code{monotone_constraints} A numerical vector consists of \code{1}, \code{0} and \code{-1} with its length equals to the number of features in the training data. \code{1} is increasing, \code{-1} is decreasing and \code{0} is no constraint.
-#'   \item \code{interaction_constraints} A list of vectors specifying feature indices of permitted interactions. Each item of the list represents one permitted interaction where specified features are allowed to interact with each other. Feature index values should start from \code{0} (\code{0} references the first column).  Leave argument unspecified for no interaction constraints.
 #' }
 #' 
 #' 2.2. Parameter for Linear Booster

R-package/configure

@@ -1667,12 +1667,12 @@ OPENMP_CXXFLAGS=""
 
 if test `uname -s` = "Linux"
 then
-  OPENMP_CXXFLAGS="\$(SHLIB_OPENMP_CXXFLAGS)"
+  OPENMP_CXXFLAGS="\$(SHLIB_OPENMP_CFLAGS)"
 fi
 
 if test `uname -s` = "Darwin"
 then
-  OPENMP_CXXFLAGS="\$(SHLIB_OPENMP_CXXFLAGS)"
+  OPENMP_CXXFLAGS="\$(SHLIB_OPENMP_CFLAGS)"
   ac_pkg_openmp=no
   { $as_echo "$as_me:${as_lineno-$LINENO}: checking whether OpenMP will work in a package" >&5
 $as_echo_n "checking whether OpenMP will work in a package... " >&6; }

R-package/configure.ac

@@ -8,12 +8,12 @@ OPENMP_CXXFLAGS=""
 
 if test `uname -s` = "Linux"
 then
-  OPENMP_CXXFLAGS="\$(SHLIB_OPENMP_CXXFLAGS)"
+  OPENMP_CXXFLAGS="\$(SHLIB_OPENMP_CFLAGS)"
 fi
 
 if test `uname -s` = "Darwin"
 then
-  OPENMP_CXXFLAGS="\$(SHLIB_OPENMP_CXXFLAGS)"
+  OPENMP_CXXFLAGS="\$(SHLIB_OPENMP_CFLAGS)"
   ac_pkg_openmp=no
   AC_MSG_CHECKING([whether OpenMP will work in a package])
   AC_LANG_CONFTEST(

R-package/demo/00Index

@@ -11,5 +11,4 @@ early_stopping                  Early Stop in training
 poisson_regression              Poisson Regression on count data
 tweedie_regression              Tweddie Regression
 gpu_accelerated                 GPU-accelerated tree building algorithms
-interaction_constraints         Interaction constraints among features
 

R-package/demo/cross_validation.R

@@ -5,20 +5,20 @@ data(agaricus.test, package='xgboost')
 dtrain <- xgb.DMatrix(agaricus.train$data, label = agaricus.train$label)
 dtest <- xgb.DMatrix(agaricus.test$data, label = agaricus.test$label)
 
-nrounds <- 2
+nround <- 2
 param <- list(max_depth=2, eta=1, silent=1, nthread=2, objective='binary:logistic')
 
 cat('running cross validation\n')
 # do cross validation, this will print result out as
 # [iteration]  metric_name:mean_value+std_value
 # std_value is standard deviation of the metric
-xgb.cv(param, dtrain, nrounds, nfold=5, metrics={'error'})
+xgb.cv(param, dtrain, nround, nfold=5, metrics={'error'})
 
 cat('running cross validation, disable standard deviation display\n')
 # do cross validation, this will print result out as
 # [iteration]  metric_name:mean_value+std_value
 # std_value is standard deviation of the metric
-xgb.cv(param, dtrain, nrounds, nfold=5,
+xgb.cv(param, dtrain, nround, nfold=5,
        metrics='error', showsd = FALSE)
 
 ###
@@ -43,9 +43,9 @@ evalerror <- function(preds, dtrain) {
 param <- list(max_depth=2, eta=1, silent=1,
               objective = logregobj, eval_metric = evalerror)
 # train with customized objective
-xgb.cv(params = param, data = dtrain, nrounds = nrounds, nfold = 5)
+xgb.cv(params = param, data = dtrain, nrounds = nround, nfold = 5)
 
 # do cross validation with prediction values for each fold
-res <- xgb.cv(params = param, data = dtrain, nrounds = nrounds, nfold = 5, prediction = TRUE)
+res <- xgb.cv(params = param, data = dtrain, nrounds = nround, nfold = 5, prediction = TRUE)
 res$evaluation_log
 length(res$pred)

R-package/demo/custom_objective.R

@@ -33,7 +33,7 @@ evalerror <- function(preds, dtrain) {
   return(list(metric = "error", value = err))
 }
 
-param <- list(max_depth=2, eta=1, nthread = 2, verbosity=0, 
+param <- list(max_depth=2, eta=1, nthread = 2, silent=1, 
               objective=logregobj, eval_metric=evalerror)
 print ('start training with user customized objective')
 # training with customized objective, we can also do step by step training
@@ -57,7 +57,7 @@ logregobjattr <- function(preds, dtrain) {
   hess <- preds * (1 - preds)
   return(list(grad = grad, hess = hess))
 }
-param <- list(max_depth=2, eta=1, nthread = 2, verbosity=0, 
+param <- list(max_depth=2, eta=1, nthread = 2, silent=1, 
               objective=logregobjattr, eval_metric=evalerror)
 print ('start training with user customized objective, with additional attributes in DMatrix')
 # training with customized objective, we can also do step by step training

R-package/demo/early_stopping.R

@@ -7,7 +7,7 @@ dtest <- xgb.DMatrix(agaricus.test$data, label = agaricus.test$label)
 # note: for customized objective function, we leave objective as default
 # note: what we are getting is margin value in prediction
 # you must know what you are doing
-param <- list(max_depth=2, eta=1, nthread=2, verbosity=0)
+param <- list(max_depth=2, eta=1, nthread = 2, silent=1)
 watchlist <- list(eval = dtest)
 num_round <- 20
 # user define objective function, given prediction, return gradient and second order gradient
@@ -32,9 +32,9 @@ evalerror <- function(preds, dtrain) {
 }
 print ('start training with early Stopping setting')
 
-bst <- xgb.train(param, dtrain, num_round, watchlist,
+bst <- xgb.train(param, dtrain, num_round, watchlist, 
                  objective = logregobj, eval_metric = evalerror, maximize = FALSE,
                  early_stopping_round = 3)
-bst <- xgb.cv(param, dtrain, num_round, nfold = 5,
+bst <- xgb.cv(param, dtrain, num_round, nfold = 5, 
               objective = logregobj, eval_metric = evalerror,
               maximize = FALSE, early_stopping_rounds = 3)

R-package/demo/interaction_constraints.R

@@ -1,105 +0,0 @@
-library(xgboost)
-library(data.table)
-
-set.seed(1024)
-
-# Function to obtain a list of interactions fitted in trees, requires input of maximum depth
-treeInteractions <- function(input_tree, input_max_depth){
-  trees <- copy(input_tree)  # copy tree input to prevent overwriting
-  if (input_max_depth < 2) return(list())  # no interactions if max depth < 2
-  if (nrow(input_tree) == 1) return(list())
-
-  # Attach parent nodes
-  for (i in 2:input_max_depth){
-    if (i == 2) trees[, ID_merge:=ID] else trees[, ID_merge:=get(paste0('parent_',i-2))]
-    parents_left <- trees[!is.na(Split), list(i.id=ID, i.feature=Feature, ID_merge=Yes)]
-    parents_right <- trees[!is.na(Split), list(i.id=ID, i.feature=Feature, ID_merge=No)]
-
-    setorderv(trees, 'ID_merge')
-    setorderv(parents_left, 'ID_merge')
-    setorderv(parents_right, 'ID_merge')
-
-    trees <- merge(trees, parents_left, by='ID_merge', all.x=T)
-    trees[!is.na(i.id), c(paste0('parent_', i-1), paste0('parent_feat_', i-1)):=list(i.id, i.feature)]
-    trees[, c('i.id','i.feature'):=NULL]
-
-    trees <- merge(trees, parents_right, by='ID_merge', all.x=T)
-    trees[!is.na(i.id), c(paste0('parent_', i-1), paste0('parent_feat_', i-1)):=list(i.id, i.feature)]
-    trees[, c('i.id','i.feature'):=NULL]
-  }
-
-  # Extract nodes with interactions
-  interaction_trees <- trees[!is.na(Split) & !is.na(parent_1), 
-                             c('Feature',paste0('parent_feat_',1:(input_max_depth-1))), with=F]
-  interaction_trees_split <- split(interaction_trees, 1:nrow(interaction_trees))
-  interaction_list <- lapply(interaction_trees_split, as.character)
-
-  # Remove NAs (no parent interaction)
-  interaction_list <- lapply(interaction_list, function(x) x[!is.na(x)])
-
-  # Remove non-interactions (same variable)
-  interaction_list <- lapply(interaction_list, unique)  # remove same variables
-  interaction_length <- sapply(interaction_list, length)
-  interaction_list <- interaction_list[interaction_length > 1]
-  interaction_list <- unique(lapply(interaction_list, sort))
-  return(interaction_list)
-}
-
-# Generate sample data
-x <- list()
-for (i in 1:10){
-  x[[i]] = i*rnorm(1000, 10)
-}
-x <- as.data.table(x)
-
-y = -1*x[, rowSums(.SD)] + x[['V1']]*x[['V2']] + x[['V3']]*x[['V4']]*x[['V5']] + rnorm(1000, 0.001) + 3*sin(x[['V7']])
-
-train = as.matrix(x)
-
-# Interaction constraint list (column names form)
-interaction_list <- list(c('V1','V2'),c('V3','V4','V5'))
-
-# Convert interaction constraint list into feature index form
-cols2ids <- function(object, col_names) {
-  LUT <- seq_along(col_names) - 1
-  names(LUT) <- col_names
-  rapply(object, function(x) LUT[x], classes="character", how="replace")
-}
-interaction_list_fid = cols2ids(interaction_list, colnames(train))
-
-# Fit model with interaction constraints
-bst = xgboost(data = train, label = y, max_depth = 4,
-              eta = 0.1, nthread = 2, nrounds = 1000,
-              interaction_constraints = interaction_list_fid)
-
-bst_tree <- xgb.model.dt.tree(colnames(train), bst)
-bst_interactions <- treeInteractions(bst_tree, 4)  # interactions constrained to combinations of V1*V2 and V3*V4*V5
-
-# Fit model without interaction constraints
-bst2 = xgboost(data = train, label = y, max_depth = 4,
-               eta = 0.1, nthread = 2, nrounds = 1000)
-
-bst2_tree <- xgb.model.dt.tree(colnames(train), bst2)
-bst2_interactions <- treeInteractions(bst2_tree, 4)  # much more interactions
-
-# Fit model with both interaction and monotonicity constraints
-bst3 = xgboost(data = train, label = y, max_depth = 4,
-               eta = 0.1, nthread = 2, nrounds = 1000,
-               interaction_constraints = interaction_list_fid,
-               monotone_constraints = c(-1,0,0,0,0,0,0,0,0,0))
-
-bst3_tree <- xgb.model.dt.tree(colnames(train), bst3)
-bst3_interactions <- treeInteractions(bst3_tree, 4)  # interactions still constrained to combinations of V1*V2 and V3*V4*V5
-
-# Show monotonic constraints still apply by checking scores after incrementing V1
-x1 <- sort(unique(x[['V1']]))
-for (i in 1:length(x1)){
-  testdata <- copy(x[, -c('V1')])
-  testdata[['V1']] <- x1[i]
-  testdata <- testdata[, paste0('V',1:10), with=F]
-  pred <- predict(bst3, as.matrix(testdata))
-  
-  # Should not print out anything due to monotonic constraints
-  if (i > 1) if (any(pred > prev_pred)) print(i)
-  prev_pred <- pred 
-}

R-package/demo/predict_first_ntree.R

@@ -7,10 +7,10 @@ dtest <- xgb.DMatrix(agaricus.test$data, label = agaricus.test$label)
 
 param <- list(max_depth=2, eta=1, silent=1, objective='binary:logistic')
 watchlist <- list(eval = dtest, train = dtrain)
-nrounds = 2
+nround = 2
 
 # training the model for two rounds
-bst = xgb.train(param, dtrain, nrounds, nthread = 2, watchlist)
+bst = xgb.train(param, dtrain, nround, nthread = 2, watchlist)
 cat('start testing prediction from first n trees\n')
 labels <- getinfo(dtest,'label')
 

R-package/demo/predict_leaf_indices.R

@@ -11,10 +11,10 @@ dtrain <- xgb.DMatrix(data = agaricus.train$data, label = agaricus.train$label)
 dtest <- xgb.DMatrix(data = agaricus.test$data, label = agaricus.test$label)
 
 param <- list(max_depth=2, eta=1, silent=1, objective='binary:logistic')
-nrounds = 4
+nround = 4
 
 # training the model for two rounds
-bst = xgb.train(params = param, data = dtrain, nrounds = nrounds, nthread = 2)
+bst = xgb.train(params = param, data = dtrain, nrounds = nround, nthread = 2)
 
 # Model accuracy without new features
 accuracy.before <- sum((predict(bst, agaricus.test$data) >= 0.5) == agaricus.test$label) / length(agaricus.test$label)
@@ -43,7 +43,7 @@ new.features.test <- create.new.tree.features(bst, agaricus.test$data)
 new.dtrain <- xgb.DMatrix(data = new.features.train, label = agaricus.train$label)
 new.dtest <- xgb.DMatrix(data = new.features.test, label = agaricus.test$label)
 watchlist <- list(train = new.dtrain)
-bst <- xgb.train(params = param, data = new.dtrain, nrounds = nrounds, nthread = 2)
+bst <- xgb.train(params = param, data = new.dtrain, nrounds = nround, nthread = 2)
 
 # Model accuracy with new features
 accuracy.after <- sum((predict(bst, new.dtest) >= 0.5) == agaricus.test$label) / length(agaricus.test$label)

R-package/man/cb.reset.parameters.Rd

@@ -22,7 +22,7 @@ This is a "pre-iteration" callback function used to reset booster's parameters
 at the beginning of each iteration.
 
 Note that when training is resumed from some previous model, and a function is used to 
-reset a parameter value, the \code{nrounds} argument in this function would be the 
+reset a parameter value, the \code{nround} argument in this function would be the 
 the number of boosting rounds in the current training.
 
 Callback function expects the following values to be set in its calling frame:

R-package/man/predict.xgb.Booster.Rd

@@ -7,8 +7,7 @@
 \usage{
 \method{predict}{xgb.Booster}(object, newdata, missing = NA,
   outputmargin = FALSE, ntreelimit = NULL, predleaf = FALSE,
-  predcontrib = FALSE, approxcontrib = FALSE,
-  predinteraction = FALSE, reshape = FALSE, ...)
+  predcontrib = FALSE, approxcontrib = FALSE, reshape = FALSE, ...)
 
 \method{predict}{xgb.Booster.handle}(object, ...)
 }
@@ -27,17 +26,14 @@ logistic regression would result in predictions for log-odds instead of probabil
 \item{ntreelimit}{limit the number of model's trees or boosting iterations used in prediction (see Details).
 It will use all the trees by default (\code{NULL} value).}
 
-\item{predleaf}{whether predict leaf index.}
+\item{predleaf}{whether predict leaf index instead.}
 
-\item{predcontrib}{whether to return feature contributions to individual predictions (see Details).}
+\item{predcontrib}{whether to return feature contributions to individual predictions instead (see Details).}
 
 \item{approxcontrib}{whether to use a fast approximation for feature contributions (see Details).}
 
-\item{predinteraction}{whether to return contributions of feature interactions to individual predictions (see Details).}
-
 \item{reshape}{whether to reshape the vector of predictions to a matrix form when there are several
-prediction outputs per case. This option has no effect when either of predleaf, predcontrib,
-or predinteraction flags is TRUE.}
+prediction outputs per case. This option has no effect when \code{predleaf = TRUE}.}
 
 \item{...}{Parameters passed to \code{predict.xgb.Booster}}
 }
@@ -55,14 +51,6 @@ When \code{predcontrib = TRUE} and it is not a multiclass setting, the output is
 For a multiclass case, a list of \code{num_class} elements is returned, where each element is
 such a matrix. The contribution values are on the scale of untransformed margin
 (e.g., for binary classification would mean that the contributions are log-odds deviations from bias).
-
-When \code{predinteraction = TRUE} and it is not a multiclass setting, the output is a 3d array with
-dimensions \code{c(nrow, num_features + 1, num_features + 1)}. The off-diagonal (in the last two dimensions)
-elements represent different features interaction contributions. The array is symmetric WRT the last
-two dimensions. The "+ 1" columns corresponds to bias. Summing this array along the last dimension should
-produce practically the same result as predict with \code{predcontrib = TRUE}.
-For a multiclass case, a list of \code{num_class} elements is returned, where each element is
-such an array.
 }
 \description{
 Predicted values based on either xgboost model or model handle object.
@@ -88,11 +76,6 @@ values (Lundberg 2017) that sum to the difference between the expected output
 of the model and the current prediction (where the hessian weights are used to compute the expectations).
 Setting \code{approxcontrib = TRUE} approximates these values following the idea explained
 in \url{http://blog.datadive.net/interpreting-random-forests/}.
-
-With \code{predinteraction = TRUE}, SHAP values of contributions of interaction of each pair of features
-are computed. Note that this operation might be rather expensive in terms of compute and memory.
-Since it quadratically depends on the number of features, it is recommended to perfom selection
-of the most important features first. See below about the format of the returned results.
 }
 \examples{
 ## binary classification:

R-package/man/xgb.create.features.Rd

@@ -63,9 +63,9 @@ dtrain <- xgb.DMatrix(data = agaricus.train$data, label = agaricus.train$label)
 dtest <- xgb.DMatrix(data = agaricus.test$data, label = agaricus.test$label)
 
 param <- list(max_depth=2, eta=1, silent=1, objective='binary:logistic')
-nrounds = 4
+nround = 4
 
-bst = xgb.train(params = param, data = dtrain, nrounds = nrounds, nthread = 2)
+bst = xgb.train(params = param, data = dtrain, nrounds = nround, nthread = 2)
 
 # Model accuracy without new features
 accuracy.before <- sum((predict(bst, agaricus.test$data) >= 0.5) == agaricus.test$label) /
@@ -79,7 +79,7 @@ new.features.test <- xgb.create.features(model = bst, agaricus.test$data)
 new.dtrain <- xgb.DMatrix(data = new.features.train, label = agaricus.train$label)
 new.dtest <- xgb.DMatrix(data = new.features.test, label = agaricus.test$label)
 watchlist <- list(train = new.dtrain)
-bst <- xgb.train(params = param, data = new.dtrain, nrounds = nrounds, nthread = 2)
+bst <- xgb.train(params = param, data = new.dtrain, nrounds = nround, nthread = 2)
 
 # Model accuracy with new features
 accuracy.after <- sum((predict(bst, new.dtest) >= 0.5) == agaricus.test$label) /

R-package/man/xgb.cv.Rd

@@ -4,12 +4,11 @@
 \alias{xgb.cv}
 \title{Cross Validation}
 \usage{
-xgb.cv(params = list(), data, nrounds, nfold, label = NULL,
-  missing = NA, prediction = FALSE, showsd = TRUE,
-  metrics = list(), obj = NULL, feval = NULL, stratified = TRUE,
-  folds = NULL, verbose = TRUE, print_every_n = 1L,
-  early_stopping_rounds = NULL, maximize = NULL, callbacks = list(),
-  ...)
+xgb.cv(params = list(), data, nrounds, nfold, label = NULL, missing = NA,
+  prediction = FALSE, showsd = TRUE, metrics = list(), obj = NULL,
+  feval = NULL, stratified = TRUE, folds = NULL, verbose = TRUE,
+  print_every_n = 1L, early_stopping_rounds = NULL, maximize = NULL,
+  callbacks = list(), ...)
 }
 \arguments{
 \item{params}{the list of parameters. Commonly used ones are:

R-package/man/xgb.dump.Rd

@@ -44,8 +44,8 @@ test <- agaricus.test
 bst <- xgboost(data = train$data, label = train$label, max_depth = 2, 
                eta = 1, nthread = 2, nrounds = 2, objective = "binary:logistic")
 # save the model in file 'xgb.model.dump'
-dump_path = file.path(tempdir(), 'model.dump')
-xgb.dump(bst, dump_path, with_stats = TRUE)
+dump.path = file.path(tempdir(), 'model.dump')
+xgb.dump(bst, dump.path, with_stats = TRUE)
 
 # print the model without saving it to a file
 print(xgb.dump(bst, with_stats = TRUE))

R-package/man/xgb.plot.deepness.Rd

@@ -5,11 +5,11 @@
 \alias{xgb.plot.deepness}
 \title{Plot model trees deepness}
 \usage{
-xgb.ggplot.deepness(model = NULL, which = c("2x1", "max.depth",
-  "med.depth", "med.weight"))
+xgb.ggplot.deepness(model = NULL, which = c("2x1", "max.depth", "med.depth",
+  "med.weight"))
 
-xgb.plot.deepness(model = NULL, which = c("2x1", "max.depth",
-  "med.depth", "med.weight"), plot = TRUE, ...)
+xgb.plot.deepness(model = NULL, which = c("2x1", "max.depth", "med.depth",
+  "med.weight"), plot = TRUE, ...)
 }
 \arguments{
 \item{model}{either an \code{xgb.Booster} model generated by the \code{xgb.train} function
@@ -50,7 +50,7 @@ per tree with respect to tree number are created. And \code{which="med.weight"}
 a tree's median absolute leaf weight changes through the iterations.
 
 This function was inspired by the blog post
-\url{https://github.com/aysent/random-forest-leaf-visualization}.
+\url{http://aysent.github.io/2015/11/08/random-forest-leaf-visualization.html}.
 }
 \examples{
 

R-package/man/xgb.plot.importance.Rd

@@ -9,8 +9,8 @@ xgb.ggplot.importance(importance_matrix = NULL, top_n = NULL,
   measure = NULL, rel_to_first = FALSE, n_clusters = c(1:10), ...)
 
 xgb.plot.importance(importance_matrix = NULL, top_n = NULL,
-  measure = NULL, rel_to_first = FALSE, left_margin = 10,
-  cex = NULL, plot = TRUE, ...)
+  measure = NULL, rel_to_first = FALSE, left_margin = 10, cex = NULL,
+  plot = TRUE, ...)
 }
 \arguments{
 \item{importance_matrix}{a \code{data.table} returned by \code{\link{xgb.importance}}.}

R-package/man/xgb.plot.shap.Rd

@@ -6,8 +6,8 @@
 \usage{
 xgb.plot.shap(data, shap_contrib = NULL, features = NULL, top_n = 1,
   model = NULL, trees = NULL, target_class = NULL,
-  approxcontrib = FALSE, subsample = NULL, n_col = 1, col = rgb(0,
-  0, 1, 0.2), pch = ".", discrete_n_uniq = 5, discrete_jitter = 0.01,
+  approxcontrib = FALSE, subsample = NULL, n_col = 1, col = rgb(0, 0, 1,
+  0.2), pch = ".", discrete_n_uniq = 5, discrete_jitter = 0.01,
   ylab = "SHAP", plot_NA = TRUE, col_NA = rgb(0.7, 0, 1, 0.6),
   pch_NA = ".", pos_NA = 1.07, plot_loess = TRUE, col_loess = 2,
   span_loess = 0.5, which = c("1d", "2d"), plot = TRUE, ...)

R-package/man/xgb.train.Rd

@@ -5,17 +5,15 @@
 \alias{xgboost}
 \title{eXtreme Gradient Boosting Training}
 \usage{
-xgb.train(params = list(), data, nrounds, watchlist = list(),
-  obj = NULL, feval = NULL, verbose = 1, print_every_n = 1L,
+xgb.train(params = list(), data, nrounds, watchlist = list(), obj = NULL,
+  feval = NULL, verbose = 1, print_every_n = 1L,
   early_stopping_rounds = NULL, maximize = NULL, save_period = NULL,
-  save_name = "xgboost.model", xgb_model = NULL, callbacks = list(),
-  ...)
+  save_name = "xgboost.model", xgb_model = NULL, callbacks = list(), ...)
 
 xgboost(data = NULL, label = NULL, missing = NA, weight = NULL,
   params = list(), nrounds, verbose = 1, print_every_n = 1L,
   early_stopping_rounds = NULL, maximize = NULL, save_period = NULL,
-  save_name = "xgboost.model", xgb_model = NULL, callbacks = list(),
-  ...)
+  save_name = "xgboost.model", xgb_model = NULL, callbacks = list(), ...)
 }
 \arguments{
 \item{params}{the list of parameters. 
@@ -37,7 +35,7 @@ xgboost(data = NULL, label = NULL, missing = NA, weight = NULL,
   \item \code{gamma} minimum loss reduction required to make a further partition on a leaf node of the tree. the larger, the more conservative the algorithm will be. 
   \item \code{max_depth} maximum depth of a tree. Default: 6
   \item \code{min_child_weight} minimum sum of instance weight (hessian) needed in a child. If the tree partition step results in a leaf node with the sum of instance weight less than min_child_weight, then the building process will give up further partitioning. In linear regression mode, this simply corresponds to minimum number of instances needed to be in each node. The larger, the more conservative the algorithm will be. Default: 1
-  \item \code{subsample} subsample ratio of the training instance. Setting it to 0.5 means that xgboost randomly collected half of the data instances to grow trees and this will prevent overfitting. It makes computation shorter (because less data to analyse). It is advised to use this parameter with \code{eta} and increase \code{nrounds}. Default: 1 
+  \item \code{subsample} subsample ratio of the training instance. Setting it to 0.5 means that xgboost randomly collected half of the data instances to grow trees and this will prevent overfitting. It makes computation shorter (because less data to analyse). It is advised to use this parameter with \code{eta} and increase \code{nround}. Default: 1 
   \item \code{colsample_bytree} subsample ratio of columns when constructing each tree. Default: 1
   \item \code{num_parallel_tree} Experimental parameter. number of trees to grow per round. Useful to test Random Forest through Xgboost (set \code{colsample_bytree < 1}, \code{subsample  < 1}  and \code{round = 1}) accordingly. Default: 1
   \item \code{monotone_constraints} A numerical vector consists of \code{1}, \code{0} and \code{-1} with its length equals to the number of features in the training data. \code{1} is increasing, \code{-1} is decreasing and \code{0} is no constraint.

R-package/src/Makevars.in

@@ -17,8 +17,8 @@ endif
 $(foreach v, $(XGB_RFLAGS), $(warning $(v)))
 
 PKG_CPPFLAGS=  -I$(PKGROOT)/include -I$(PKGROOT)/dmlc-core/include -I$(PKGROOT)/rabit/include -I$(PKGROOT) $(XGB_RFLAGS)
-PKG_CXXFLAGS= @OPENMP_CXXFLAGS@ -pthread
-PKG_LIBS = @OPENMP_CXXFLAGS@ -pthread
+PKG_CXXFLAGS= @OPENMP_CXXFLAGS@ $(SHLIB_PTHREAD_FLAGS)
+PKG_LIBS = @OPENMP_CXXFLAGS@ $(SHLIB_PTHREAD_FLAGS)
 OBJECTS= ./xgboost_R.o ./xgboost_custom.o ./xgboost_assert.o ./init.o\
          $(PKGROOT)/amalgamation/xgboost-all0.o $(PKGROOT)/amalgamation/dmlc-minimum0.o\
          $(PKGROOT)/rabit/src/engine_empty.o $(PKGROOT)/rabit/src/c_api.o

R-package/src/Makevars.win

@@ -29,8 +29,8 @@ endif
 $(foreach v, $(XGB_RFLAGS), $(warning $(v)))
 
 PKG_CPPFLAGS=  -I$(PKGROOT)/include -I$(PKGROOT)/dmlc-core/include -I$(PKGROOT)/rabit/include -I$(PKGROOT) $(XGB_RFLAGS)
-PKG_CXXFLAGS= $(SHLIB_OPENMP_CXXFLAGS) $(SHLIB_PTHREAD_FLAGS)
-PKG_LIBS = $(SHLIB_OPENMP_CXXFLAGS) $(SHLIB_PTHREAD_FLAGS)
+PKG_CXXFLAGS= $(SHLIB_OPENMP_CFLAGS) $(SHLIB_PTHREAD_FLAGS)
+PKG_LIBS = $(SHLIB_OPENMP_CFLAGS) $(SHLIB_PTHREAD_FLAGS)
 OBJECTS= ./xgboost_R.o ./xgboost_custom.o ./xgboost_assert.o ./init.o\
          $(PKGROOT)/amalgamation/xgboost-all0.o $(PKGROOT)/amalgamation/dmlc-minimum0.o\
          $(PKGROOT)/rabit/src/engine_empty.o $(PKGROOT)/rabit/src/c_api.o

R-package/src/init.c

@@ -1,5 +1,5 @@
 /* Copyright (c) 2015 by Contributors
- *
+ * 
  * This file was initially generated using the following R command:
  * tools::package_native_routine_registration_skeleton('.', con = 'src/init.c', character_only = F)
  * and edited to conform to xgboost C linter requirements. For details, see
@@ -10,7 +10,7 @@
 #include <stdlib.h>
 #include <R_ext/Rdynload.h>
 
-/* FIXME:
+/* FIXME: 
 Check these declarations against the C/Fortran source code.
 */
 
@@ -70,7 +70,7 @@ static const R_CallMethodDef CallEntries[] = {
 
 #if defined(_WIN32)
 __declspec(dllexport)
-#endif  // defined(_WIN32)
+#endif
 void R_init_xgboost(DllInfo *dll) {
   R_registerRoutines(dll, NULL, CallEntries, NULL, NULL);
   R_useDynamicSymbols(dll, FALSE);

R-package/src/xgboost_custom.cc

@@ -32,10 +32,7 @@ extern "C" {
 
 namespace xgboost {
 ConsoleLogger::~ConsoleLogger() {
-  if (cur_verbosity_ == LogVerbosity::kIgnore ||
-      cur_verbosity_ <= global_verbosity_) {
-    dmlc::CustomLogMessage::Log(log_stream_.str());
-  }
+  dmlc::CustomLogMessage::Log(log_stream_.str());
 }
 TrackerLogger::~TrackerLogger() {
   dmlc::CustomLogMessage::Log(log_stream_.str());
@@ -49,11 +46,10 @@ namespace common {
 bool CheckNAN(double v) {
   return ISNAN(v);
 }
-#if !defined(XGBOOST_USE_CUDA)
 double LogGamma(double v) {
   return lgammafn(v);
 }
-#endif  // !defined(XGBOOST_USE_CUDA)
+
 // customize random engine.
 void CustomGlobalRandomEngine::seed(CustomGlobalRandomEngine::result_type val) {
   // ignore the seed

R-package/tests/testthat/test_basic.R

@@ -182,7 +182,7 @@ test_that("xgb.cv works", {
   expect_is(cv, 'xgb.cv.synchronous')
   expect_false(is.null(cv$evaluation_log))
   expect_lt(cv$evaluation_log[, min(test_error_mean)], 0.03)
-  expect_lt(cv$evaluation_log[, min(test_error_std)], 0.008)
+  expect_lt(cv$evaluation_log[, min(test_error_std)], 0.004)
   expect_equal(cv$niter, 2)
   expect_false(is.null(cv$folds) && is.list(cv$folds))
   expect_length(cv$folds, 5)
@@ -223,42 +223,3 @@ test_that("train and predict with non-strict classes", {
   expect_error(pr <- predict(bst, train_dense), regexp = NA)
   expect_equal(pr0, pr)
 })
-
-test_that("max_delta_step works", {
-  dtrain <- xgb.DMatrix(agaricus.train$data, label = agaricus.train$label)
-  watchlist <- list(train = dtrain)
-  param <- list(objective = "binary:logistic", eval_metric="logloss", max_depth = 2, nthread = 2, eta = 0.5)
-  nrounds = 5
-  # model with no restriction on max_delta_step
-  bst1 <- xgb.train(param, dtrain, nrounds, watchlist, verbose = 1)
-  # model with restricted max_delta_step
-  bst2 <- xgb.train(param, dtrain, nrounds, watchlist, verbose = 1, max_delta_step = 1)
-  # the no-restriction model is expected to have consistently lower loss during the initial interations
-  expect_true(all(bst1$evaluation_log$train_logloss < bst2$evaluation_log$train_logloss))
-  expect_lt(mean(bst1$evaluation_log$train_logloss)/mean(bst2$evaluation_log$train_logloss), 0.8)
-})
-
-test_that("colsample_bytree works", {
-  # Randomly generate data matrix by sampling from uniform distribution [-1, 1]
-  set.seed(1)
-  train_x <- matrix(runif(1000, min = -1, max = 1), ncol = 100)
-  train_y <- as.numeric(rowSums(train_x) > 0)
-  test_x <- matrix(runif(1000, min = -1, max = 1), ncol = 100)
-  test_y <- as.numeric(rowSums(test_x) > 0)
-  colnames(train_x) <- paste0("Feature_", sprintf("%03d", 1:100))
-  colnames(test_x) <- paste0("Feature_", sprintf("%03d", 1:100))
-   dtrain <- xgb.DMatrix(train_x, label = train_y)
-  dtest <- xgb.DMatrix(test_x, label = test_y)
-  watchlist <- list(train = dtrain, eval = dtest)
-   # Use colsample_bytree = 0.01, so that roughly one out of 100 features is
-  # chosen for each tree
-  param <- list(max_depth = 2, eta = 0, silent = 1, nthread = 2,
-                colsample_bytree = 0.01, objective = "binary:logistic",
-                eval_metric = "auc")
-   set.seed(2)
-  bst <- xgb.train(param, dtrain, nrounds = 100, watchlist, verbose = 0)
-  xgb.importance(model = bst)
-  # If colsample_bytree works properly, a variety of features should be used
-  # in the 100 trees
-  expect_gte(nrow(xgb.importance(model = bst)), 30)
-})

R-package/tests/testthat/test_callbacks.R

@@ -282,7 +282,7 @@ test_that("prediction in xgb.cv works for gblinear too", {
 })
 
 test_that("prediction in early-stopping xgb.cv works", {
-  set.seed(11)
+  set.seed(1)
   expect_output(
     cv <- xgb.cv(param, dtrain, nfold = 5, eta = 0.1, nrounds = 20,
                  early_stopping_rounds = 5, maximize = FALSE, prediction = TRUE)

R-package/tests/testthat/test_dmatrix.R

@@ -77,18 +77,6 @@ test_that("xgb.DMatrix: slice, dim", {
   expect_equal(getinfo(dsub1, 'label'), getinfo(dsub2, 'label'))
 })
 
-test_that("xgb.DMatrix: slice, trailing empty rows", {
-  data(agaricus.train, package='xgboost')
-  train_data <- agaricus.train$data
-  train_label <- agaricus.train$label
-  dtrain <- xgb.DMatrix(data=train_data, label=train_label)
-  slice(dtrain, 6513L)
-  train_data[6513, ] <- 0
-  dtrain <- xgb.DMatrix(data=train_data, label=train_label)
-  slice(dtrain, 6513L)
-  expect_equal(nrow(dtrain), 6513)
-})
-
 test_that("xgb.DMatrix: colnames", {
   dtest <- xgb.DMatrix(test_data, label=test_label)
   expect_equal(colnames(dtest), colnames(test_data))

R-package/tests/testthat/test_gc_safety.R

@@ -9,7 +9,7 @@ test_that("train and prediction when gctorture is on", {
   test <- agaricus.test
   gctorture(TRUE)
   bst <- xgboost(data = train$data, label = train$label, max.depth = 2,
-  eta = 1, nthread = 2, nrounds = 2, objective = "binary:logistic")
+  eta = 1, nthread = 2, nround = 2, objective = "binary:logistic")
   pred <- predict(bst, test$data)
   gctorture(FALSE)
 })

R-package/tests/testthat/test_helpers.R

@@ -7,9 +7,6 @@ require(vcd, quietly = TRUE)
 
 float_tolerance = 5e-6
 
-# disable some tests for Win32
-win32_flag = .Platform$OS.type == "windows" && .Machine$sizeof.pointer != 8
-
 set.seed(1982)
 data(Arthritis)
 df <- data.table(Arthritis, keep.rownames = F)
@@ -44,8 +41,7 @@ mbst.GLM <- xgboost(data = as.matrix(iris[, -5]), label = mlabel, verbose = 0,
 
 
 test_that("xgb.dump works", {
-  if (!win32_flag)
-    expect_length(xgb.dump(bst.Tree), 200)
+  expect_length(xgb.dump(bst.Tree), 200)
   dump_file = file.path(tempdir(), 'xgb.model.dump')
   expect_true(xgb.dump(bst.Tree, dump_file, with_stats = T))
   expect_true(file.exists(dump_file))
@@ -54,8 +50,7 @@ test_that("xgb.dump works", {
   # JSON format
   dmp <- xgb.dump(bst.Tree, dump_format = "json")
   expect_length(dmp, 1)
-  if (!win32_flag)
-    expect_length(grep('nodeid', strsplit(dmp, '\n')[[1]]), 188)
+  expect_length(grep('nodeid', strsplit(dmp, '\n')[[1]]), 188)
 })
 
 test_that("xgb.dump works for gblinear", {
@@ -215,8 +210,7 @@ test_that("xgb.model.dt.tree works with and without feature names", {
   names.dt.trees <- c("Tree", "Node", "ID", "Feature", "Split", "Yes", "No", "Missing", "Quality", "Cover")
   dt.tree <- xgb.model.dt.tree(feature_names = feature.names, model = bst.Tree)
   expect_equal(names.dt.trees, names(dt.tree))
-  if (!win32_flag)
-    expect_equal(dim(dt.tree), c(188, 10))
+  expect_equal(dim(dt.tree), c(188, 10))
   expect_output(str(dt.tree), 'Feature.*\\"Age\\"')
 
   dt.tree.0 <- xgb.model.dt.tree(model = bst.Tree)
@@ -242,8 +236,7 @@ test_that("xgb.model.dt.tree throws error for gblinear", {
 
 test_that("xgb.importance works with and without feature names", {
   importance.Tree <- xgb.importance(feature_names = feature.names, model = bst.Tree)
-  if (!win32_flag)
-    expect_equal(dim(importance.Tree), c(7, 4))
+  expect_equal(dim(importance.Tree), c(7, 4))
   expect_equal(colnames(importance.Tree), c("Feature", "Gain", "Cover", "Frequency"))
   expect_output(str(importance.Tree), 'Feature.*\\"Age\\"')
 

R-package/tests/testthat/test_interaction_constraints.R

@@ -1,38 +0,0 @@
-require(xgboost)
-
-context("interaction constraints")
-
-set.seed(1024)
-x1 <- rnorm(1000, 1)
-x2 <- rnorm(1000, 1)
-x3 <- sample(c(1,2,3), size=1000, replace=TRUE)
-y <- x1 + x2 + x3 + x1*x2*x3 + rnorm(1000, 0.001) + 3*sin(x1)
-train <- matrix(c(x1,x2,x3), ncol = 3)
-
-test_that("interaction constraints for regression", {
-  # Fit a model that only allows interaction between x1 and x2
-  bst <- xgboost(data = train, label = y, max_depth = 3,
-                 eta = 0.1, nthread = 2, nrounds = 100, verbose = 0,
-                 interaction_constraints = list(c(0,1)))
-  
-  # Set all observations to have the same x3 values then increment
-  #  by the same amount
-	preds <- lapply(c(1,2,3), function(x){
-		tmat <- matrix(c(x1,x2,rep(x,1000)), ncol=3)
-		return(predict(bst, tmat))
-	})
-
-  # Check incrementing x3 has the same effect on all observations
-  #   since x3 is constrained to be independent of x1 and x2
-  #   and all observations start off from the same x3 value
-	diff1 <- preds[[2]] - preds[[1]]
-	test1 <- all(abs(diff1 - diff1[1]) < 1e-4)
-	
-	diff2 <- preds[[3]] - preds[[2]]
-	test2 <- all(abs(diff2 - diff2[1]) < 1e-4)
-	
-  expect_true({
-    test1 & test2
-  }, "Interaction Contraint Satisfied")
-  
-})

R-package/tests/testthat/test_interactions.R

@@ -1,141 +0,0 @@
-context('Test prediction of feature interactions')
-
-require(xgboost)
-require(magrittr)
-
-set.seed(123)
-
-test_that("predict feature interactions works", {
-  # simulate some binary data and a linear outcome with an interaction term
-  N <- 1000
-  P <- 5
-  X <- matrix(rbinom(N * P, 1, 0.5), ncol=P, dimnames = list(NULL, letters[1:P]))
-  # center the data (as contributions are computed WRT feature means)
-  X <- scale(X, scale=FALSE)
-
-  # outcome without any interactions, without any noise:
-  f <- function(x) 2 * x[, 1] - 3 * x[, 2]
-  # outcome with interactions, without noise:
-  f_int <- function(x) f(x) + 2 * x[, 2] * x[, 3]
-  # outcome with interactions, with noise:
-  #f_int_noise <- function(x) f_int(x) + rnorm(N, 0, 0.3)
-
-  y <- f_int(X)
-
-  dm <- xgb.DMatrix(X, label = y)
-  param <- list(eta=0.1, max_depth=4, base_score=mean(y), lambda=0, nthread=2)
-  b <- xgb.train(param, dm, 100)
-  
-  pred = predict(b, dm, outputmargin=TRUE)
-
-  # SHAP contributions:
-  cont <- predict(b, dm, predcontrib=TRUE)
-  expect_equal(dim(cont), c(N, P+1))
-  # make sure for each row they add up to marginal predictions
-  max(abs(rowSums(cont) - pred)) %>% expect_lt(0.001)
-  # Hand-construct the 'ground truth' feature contributions:
-  gt_cont <- cbind(
-      2. * X[, 1],
-     -3. * X[, 2] + 1. * X[, 2] * X[, 3], # attribute a HALF of the interaction term to feature #2
-      1. * X[, 2] * X[, 3]               # and another HALF of the interaction term to feature #3
-     )
-  gt_cont <- cbind(gt_cont, matrix(0, nrow=N, ncol=P + 1 - 3))
-  # These should be relatively close:
-  expect_lt(max(abs(cont - gt_cont)), 0.05)
-
-
-  # SHAP interaction contributions:
-  intr <- predict(b, dm, predinteraction=TRUE)
-  expect_equal(dim(intr), c(N, P+1, P+1))
-  # check assigned colnames
-  cn <- c(letters[1:P], "BIAS")
-  expect_equal(dimnames(intr), list(NULL, cn, cn))
-
-  # check the symmetry
-  max(abs(aperm(intr, c(1,3,2)) - intr)) %>% expect_lt(0.00001)
-
-  # sums WRT columns must be close to feature contributions
-  max(abs(apply(intr, c(1,2), sum) - cont)) %>% expect_lt(0.00001)
-
-  # diagonal terms for features 3,4,5 must be close to zero
-  Reduce(max, sapply(3:P, function(i) max(abs(intr[, i, i])))) %>% expect_lt(0.05)
-
-  # BIAS must have no interactions
-  max(abs(intr[, 1:P, P+1])) %>% expect_lt(0.00001)
-
-  # interactions other than 2 x 3 must be close to zero
-  intr23 <- intr
-  intr23[,2,3] <- 0
-  Reduce(max, sapply(1:P, function(i) max(abs(intr23[, i, (i+1):(P+1)])))) %>% expect_lt(0.05)
-
-  # Construct the 'ground truth' contributions of interactions directly from the linear terms:
-  gt_intr <- array(0, c(N, P+1, P+1))
-  gt_intr[,2,3] <- 1. * X[, 2] * X[, 3] # attribute a HALF of the interaction term to each symmetric element
-  gt_intr[,3,2] <- gt_intr[, 2, 3]
-  # merge-in the diagonal based on 'ground truth' feature contributions
-  intr_diag = gt_cont - apply(gt_intr, c(1,2), sum)
-  for(j in seq_len(P)) {
-    gt_intr[,j,j] = intr_diag[,j]
-  }
-  # These should be relatively close:
-  expect_lt(max(abs(intr - gt_intr)), 0.1)
-})
-
-test_that("SHAP contribution values are not NAN", {
-  d <- data.frame(
-    x1 = c(-2.3, 1.4, 5.9, 2, 2.5, 0.3, -3.6, -0.2, 0.5, -2.8, -4.6, 3.3, -1.2,
-           -1.1, -2.3, 0.4, -1.5, -0.2, -1, 3.7),
-    x2 = c(291.179171, 269.198331, 289.942097, 283.191669, 269.673332,
-           294.158346, 287.255835, 291.530838, 285.899586, 269.290833,
-           268.649586, 291.530841, 280.074593, 269.484168, 293.94042,
-           294.327506, 296.20709, 295.441669, 283.16792, 270.227085),
-    y = c(9, 15, 5.7, 9.2, 22.4, 5, 9, 3.2, 7.2, 13.1, 7.8, 16.9, 6.5, 22.1,
-          5.3, 10.4, 11.1, 13.9, 11, 20.5),
-    fold = c(2, 2, 2, 1, 2, 2, 1, 2, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2))
-
-  ivs <- c("x1", "x2")
-
-  fit <- xgboost(
-    verbose = 0,
-    params = list(
-      objective = "reg:linear",
-      eval_metric = "rmse"),
-    data = as.matrix(subset(d, fold == 2)[, ivs]),
-    label = subset(d, fold == 2)$y,
-    nthread = 1,
-    nrounds = 3)
-
-  shaps <- as.data.frame(predict(fit,
-    newdata = as.matrix(subset(d, fold == 1)[, ivs]),
-    predcontrib = T))
-  result <- cbind(shaps, sum = rowSums(shaps), pred = predict(fit,
-      newdata = as.matrix(subset(d, fold == 1)[, ivs])))
-
-  expect_true(identical(TRUE, all.equal(result$sum, result$pred, tol = 1e-6)))
-})
-
-
-test_that("multiclass feature interactions work", {
-  dm <- xgb.DMatrix(as.matrix(iris[,-5]), label=as.numeric(iris$Species)-1)
-  param <- list(eta=0.1, max_depth=4, objective='multi:softprob', num_class=3)
-  b <- xgb.train(param, dm, 40)
-  pred = predict(b, dm, outputmargin=TRUE) %>% array(c(3, 150)) %>% t
-
-  # SHAP contributions:
-  cont <- predict(b, dm, predcontrib=TRUE)
-  expect_length(cont, 3)
-  # rewrap them as a 3d array
-  cont <- unlist(cont) %>% array(c(150, 5, 3))
-  # make sure for each row they add up to marginal predictions
-  max(abs(apply(cont, c(1,3), sum) - pred)) %>% expect_lt(0.001)
-
-  # SHAP interaction contributions:
-  intr <- predict(b, dm, predinteraction=TRUE)
-  expect_length(intr, 3)
-  # rewrap them as a 4d array
-  intr <- unlist(intr) %>% array(c(150, 5, 5, 3)) %>% aperm(c(4, 1, 2, 3)) # [grp, row, col, col]
-  # check the symmetry
-  max(abs(aperm(intr, c(1,2,4,3)) - intr)) %>% expect_lt(0.00001)
-  # sums WRT columns must be close to feature contributions
-  max(abs(apply(intr, c(1,2,3), sum) - aperm(cont, c(3,1,2)))) %>% expect_lt(0.00001)
-})

R-package/tests/testthat/test_update.R

@@ -7,10 +7,6 @@ data(agaricus.test, package = 'xgboost')
 dtrain <- xgb.DMatrix(agaricus.train$data, label = agaricus.train$label)
 dtest <- xgb.DMatrix(agaricus.test$data, label = agaricus.test$label)
 
-# Disable flaky tests for 32-bit Windows.
-# See https://github.com/dmlc/xgboost/issues/3720
-win32_flag = .Platform$OS.type == "windows" && .Machine$sizeof.pointer != 8
-
 test_that("updating the model works", {
   watchlist = list(train = dtrain, test = dtest)
 
@@ -33,9 +29,7 @@ test_that("updating the model works", {
   tr1r <- xgb.model.dt.tree(model = bst1r)
   # all should be the same when no subsampling
   expect_equal(bst1$evaluation_log, bst1r$evaluation_log)
-  if (!win32_flag) {
-    expect_equal(tr1, tr1r, tolerance = 0.00001, check.attributes = FALSE)
-  }
+  expect_equal(tr1, tr1r, tolerance = 0.00001, check.attributes = FALSE)
 
   # the same boosting with subsampling with an extra 'refresh' updater:
   p2r <- modifyList(p2, list(updater = 'grow_colmaker,prune,refresh', refresh_leaf = FALSE))
@@ -44,9 +38,7 @@ test_that("updating the model works", {
   tr2r <- xgb.model.dt.tree(model = bst2r)
   # should be the same evaluation but different gains and larger cover
   expect_equal(bst2$evaluation_log, bst2r$evaluation_log)
-  if (!win32_flag) {
-    expect_equal(tr2[Feature == 'Leaf']$Quality, tr2r[Feature == 'Leaf']$Quality)
-  }
+  expect_equal(tr2[Feature == 'Leaf']$Quality, tr2r[Feature == 'Leaf']$Quality)
   expect_gt(sum(abs(tr2[Feature != 'Leaf']$Quality - tr2r[Feature != 'Leaf']$Quality)), 100)
   expect_gt(sum(tr2r$Cover) / sum(tr2$Cover), 1.5)
 
@@ -69,9 +61,7 @@ test_that("updating the model works", {
   expect_gt(sum(tr2u$Cover) / sum(tr2$Cover), 1.5)
   # the results should be the same as for the model with an extra 'refresh' updater
   expect_equal(bst2r$evaluation_log, bst2u$evaluation_log)
-  if (!win32_flag) {
-    expect_equal(tr2r, tr2u, tolerance = 0.00001, check.attributes = FALSE)
-  }
+  expect_equal(tr2r, tr2u, tolerance = 0.00001, check.attributes = FALSE)
   
   # process type 'update' for no-subsampling model, refreshing only the tree stats from TEST data:
   p1ut <- modifyList(p1, list(process_type = 'update', updater = 'refresh', refresh_leaf = FALSE))

README.md

@@ -1,34 +1,51 @@
 <img src=https://raw.githubusercontent.com/dmlc/dmlc.github.io/master/img/logo-m/xgboost.png width=135/>  eXtreme Gradient Boosting
 ===========
-[![Build Status](https://xgboost-ci.net/job/xgboost/job/master/badge/icon?style=plastic)](https://xgboost-ci.net/blue/organizations/jenkins/xgboost/activity)
 [![Build Status](https://travis-ci.org/dmlc/xgboost.svg?branch=master)](https://travis-ci.org/dmlc/xgboost)
 [![Build Status](https://ci.appveyor.com/api/projects/status/5ypa8vaed6kpmli8?svg=true)](https://ci.appveyor.com/project/tqchen/xgboost)
 [![Documentation Status](https://readthedocs.org/projects/xgboost/badge/?version=latest)](https://xgboost.readthedocs.org)
 [![GitHub license](http://dmlc.github.io/img/apache2.svg)](./LICENSE)
 [![CRAN Status Badge](http://www.r-pkg.org/badges/version/xgboost)](http://cran.r-project.org/web/packages/xgboost)
 [![PyPI version](https://badge.fury.io/py/xgboost.svg)](https://pypi.python.org/pypi/xgboost/)
+[![Gitter chat for developers at https://gitter.im/dmlc/xgboost](https://badges.gitter.im/Join%20Chat.svg)](https://gitter.im/dmlc/xgboost?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge&utm_content=badge)
 
-[Community](https://xgboost.ai/community) |
 [Documentation](https://xgboost.readthedocs.org) |
 [Resources](demo/README.md) |
-[Contributors](CONTRIBUTORS.md) |
-[Release Notes](NEWS.md)
+[Installation](https://xgboost.readthedocs.org/en/latest/build.html) |
+[Release Notes](NEWS.md) |
+[RoadMap](https://github.com/dmlc/xgboost/issues/873)
 
 XGBoost is an optimized distributed gradient boosting library designed to be highly ***efficient***, ***flexible*** and ***portable***.
 It implements machine learning algorithms under the [Gradient Boosting](https://en.wikipedia.org/wiki/Gradient_boosting) framework.
 XGBoost provides a parallel tree boosting (also known as GBDT, GBM) that solve many data science problems in a fast and accurate way.
 The same code runs on major distributed environment (Hadoop, SGE, MPI) and can solve problems beyond billions of examples.
 
+What's New
+----------
+* [XGBoost GPU support with fast histogram algorithm](https://github.com/dmlc/xgboost/tree/master/plugin/updater_gpu)
+* [XGBoost4J: Portable Distributed XGboost in Spark, Flink and Dataflow](http://dmlc.ml/2016/03/14/xgboost4j-portable-distributed-xgboost-in-spark-flink-and-dataflow.html), see [JVM-Package](https://github.com/dmlc/xgboost/tree/master/jvm-packages)
+* [Story and Lessons Behind the Evolution of XGBoost](http://homes.cs.washington.edu/~tqchen/2016/03/10/story-and-lessons-behind-the-evolution-of-xgboost.html)
+* [Tutorial: Distributed XGBoost on AWS with YARN](https://xgboost.readthedocs.io/en/latest/tutorials/aws_yarn.html)
+* [XGBoost brick](NEWS.md) Release
+
+Ask a Question
+--------------
+* For reporting bugs please use the [xgboost/issues](https://github.com/dmlc/xgboost/issues) page.
+* For generic questions or to share your experience using XGBoost please use the [XGBoost User Group](https://groups.google.com/forum/#!forum/xgboost-user/)
+
+Help to Make XGBoost Better
+---------------------------
+XGBoost has been developed and used by a group of active community members. Your help is very valuable to make the package better for everyone.
+- Check out [call for contributions](https://github.com/dmlc/xgboost/issues?q=is%3Aissue+label%3Acall-for-contribution+is%3Aopen) and [Roadmap](https://github.com/dmlc/xgboost/issues/873) to see what can be improved, or open an issue if you want something.
+- Contribute to the [documents and examples](https://github.com/dmlc/xgboost/blob/master/doc/) to share your experience with other users.
+- Add your stories and experience to [Awesome XGBoost](demo/README.md).
+- Please add your name to [CONTRIBUTORS.md](CONTRIBUTORS.md) and after your patch has been merged.
+  - Please also update [NEWS.md](NEWS.md) on changes and improvements in API and docs.
+
 License
 -------
 © Contributors, 2016. Licensed under an [Apache-2](https://github.com/dmlc/xgboost/blob/master/LICENSE) license.
 
-Contribute to XGBoost
----------------------
-XGBoost has been developed and used by a group of active community members. Your help is very valuable to make the package better for everyone.
-Checkout the [Community Page](https://xgboost.ai/community)
-
 Reference
 ---------
-- Tianqi Chen and Carlos Guestrin. [XGBoost: A Scalable Tree Boosting System](http://arxiv.org/abs/1603.02754). In 22nd SIGKDD Conference on Knowledge Discovery and Data Mining, 2016
-- XGBoost originates from research project at University of Washington.
+- Tianqi Chen and Carlos Guestrin. [XGBoost: A Scalable Tree Boosting System](http://arxiv.org/abs/1603.02754). In 22nd SIGKDD Conference on Knowledge Discovery and Data Mining, 2016 
+- XGBoost originates from research project at University of Washington, see also the [Project Page at UW](http://dmlc.cs.washington.edu/xgboost.html).

amalgamation/xgboost-all0.cc

@@ -20,7 +20,6 @@
 #include "../src/objective/regression_obj.cc"
 #include "../src/objective/multiclass_obj.cc"
 #include "../src/objective/rank_obj.cc"
-#include "../src/objective/hinge.cc"
 
 // gbms
 #include "../src/gbm/gbm.cc"
@@ -44,11 +43,10 @@
 #endif
 
 // tress
-#include "../src/tree/split_evaluator.cc"
 #include "../src/tree/tree_model.cc"
 #include "../src/tree/tree_updater.cc"
 #include "../src/tree/updater_colmaker.cc"
-#include "../src/tree/updater_quantile_hist.cc"
+#include "../src/tree/updater_fast_hist.cc"
 #include "../src/tree/updater_prune.cc"
 #include "../src/tree/updater_refresh.cc"
 #include "../src/tree/updater_sync.cc"

appveyor.yml

@@ -44,18 +44,16 @@ install:
     - set DO_PYTHON=off
     - if /i "%target%" == "mingw" set DO_PYTHON=on
     - if /i "%target%_%ver%_%configuration%" == "msvc_2015_Release" set DO_PYTHON=on
-    - if /i "%DO_PYTHON%" == "on" conda install -y numpy scipy pandas matplotlib pytest scikit-learn graphviz python-graphviz
+    - if /i "%DO_PYTHON%" == "on" conda install -y numpy scipy pandas matplotlib nose scikit-learn graphviz python-graphviz
     # R: based on https://github.com/krlmlr/r-appveyor
     - ps: |
         if($env:target -eq 'rmingw' -or $env:target -eq 'rmsvc') {
           #$ErrorActionPreference = "Stop"
-          Invoke-WebRequest https://raw.githubusercontent.com/krlmlr/r-appveyor/master/scripts/appveyor-tool.ps1 -OutFile "$Env:TEMP\appveyor-tool.ps1"
+          Invoke-WebRequest http://raw.github.com/krlmlr/r-appveyor/master/scripts/appveyor-tool.ps1 -OutFile "$Env:TEMP\appveyor-tool.ps1"
           Import-Module "$Env:TEMP\appveyor-tool.ps1"
           Bootstrap
-          $DEPS = "c('data.table','magrittr','stringi','ggplot2','DiagrammeR','Ckmeans.1d.dp','vcd','testthat','lintr','knitr','rmarkdown')"
+          $DEPS = "c('data.table','magrittr','stringi','ggplot2','DiagrammeR','Ckmeans.1d.dp','vcd','testthat','igraph','knitr','rmarkdown')"
           cmd.exe /c "R.exe -q -e ""install.packages($DEPS, repos='$CRAN', type='both')"" 2>&1"
-          $BINARY_DEPS = "c('XML','igraph')"
-          cmd.exe /c "R.exe -q -e ""install.packages($BINARY_DEPS, repos='$CRAN', type='win.binary')"" 2>&1"
         }
 
 build_script:
@@ -96,7 +94,7 @@ build_script:
 
 test_script:
     - cd %APPVEYOR_BUILD_FOLDER%
-    - if /i "%DO_PYTHON%" == "on" python -m pytest tests/python
+    - if /i "%DO_PYTHON%" == "on" python -m nose tests/python
     # mingw R package: run the R check (which includes unit tests), and also keep the built binary package
     - if /i "%target%" == "rmingw" (
         set _R_CHECK_CRAN_INCOMING_=FALSE&&

cmake/Sanitizer.cmake

@@ -1,58 +0,0 @@
-# Set appropriate compiler and linker flags for sanitizers.
-#
-# Usage of this module:
-#  enable_sanitizers("address;leak")
-
-# Add flags
-macro(enable_sanitizer santizer)
-  if(${santizer} MATCHES "address")
-    find_package(ASan REQUIRED)
-    set(SAN_COMPILE_FLAGS "${SAN_COMPILE_FLAGS} -fsanitize=address")
-    link_libraries(${ASan_LIBRARY})
-
-  elseif(${santizer} MATCHES "thread")
-    find_package(TSan REQUIRED)
-    set(SAN_COMPILE_FLAGS "${SAN_COMPILE_FLAGS} -fsanitize=thread")
-    link_libraries(${TSan_LIBRARY})
-
-  elseif(${santizer} MATCHES "leak")
-    find_package(LSan REQUIRED)
-    set(SAN_COMPILE_FLAGS "${SAN_COMPILE_FLAGS} -fsanitize=leak")
-    link_libraries(${LSan_LIBRARY})
-
-  else()
-    message(FATAL_ERROR "Santizer ${santizer} not supported.")
-  endif()
-endmacro()
-
-macro(enable_sanitizers SANITIZERS)
-  # Check sanitizers compatibility.
-  # Idealy, we should use if(san IN_LIST SANITIZERS) ... endif()
-  # But I haven't figure out how to make it work.
-  foreach ( _san ${SANITIZERS} )
-    string(TOLOWER ${_san} _san)
-    if (_san MATCHES "thread")
-      if (${_use_other_sanitizers})
-        message(FATAL_ERROR
-          "thread sanitizer is not compatible with ${_san} sanitizer.")
-      endif()
-      set(_use_thread_sanitizer 1)
-    else ()
-      if (${_use_thread_sanitizer})
-        message(FATAL_ERROR
-          "${_san} sanitizer is not compatible with thread sanitizer.")
-      endif()
-      set(_use_other_sanitizers 1)
-    endif()
-  endforeach()
-
-  message("Sanitizers: ${SANITIZERS}")
-
-  foreach( _san ${SANITIZERS} )
-    string(TOLOWER ${_san} _san)
-    enable_sanitizer(${_san})
-  endforeach()
-  message("Sanitizers compile flags: ${SAN_COMPILE_FLAGS}")
-  set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} ${SAN_COMPILE_FLAGS}")
-  set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} ${SAN_COMPILE_FLAGS}")
-endmacro()

cmake/build_config.h.in

@@ -1,11 +0,0 @@
-/*!
- * Copyright 2019 by Contributors
- * \file build_config.h
- */
-#ifndef XGBOOST_BUILD_CONFIG_H_
-#define XGBOOST_BUILD_CONFIG_H_
-
-#cmakedefine XGBOOST_MM_PREFETCH_PRESENT
-#cmakedefine XGBOOST_BUILTIN_PREFETCH_PRESENT
-
-#endif  // XGBOOST_BUILD_CONFIG_H_

cmake/modules/FindASan.cmake

@@ -1,13 +0,0 @@
-set(ASan_LIB_NAME ASan)
-
-find_library(ASan_LIBRARY
-  NAMES libasan.so libasan.so.4 libasan.so.3 libasan.so.2 libasan.so.1 libasan.so.0
-  PATHS ${SANITIZER_PATH} /usr/lib64 /usr/lib /usr/local/lib64 /usr/local/lib ${CMAKE_PREFIX_PATH}/lib)
-
-include(FindPackageHandleStandardArgs)
-find_package_handle_standard_args(ASan DEFAULT_MSG
-  ASan_LIBRARY)
-
-mark_as_advanced(
-  ASan_LIBRARY
-  ASan_LIB_NAME)

cmake/modules/FindGTest.cmake

@@ -0,0 +1,79 @@
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+# Tries to find GTest headers and libraries.
+#
+# Usage of this module as follows:
+#
+#  find_package(GTest)
+#
+# Variables used by this module, they can change the default behaviour and need
+# to be set before calling find_package:
+#
+#  GTest_HOME - When set, this path is inspected instead of standard library
+#                locations as the root of the GTest installation.
+#                The environment variable GTEST_HOME overrides this veriable.
+#
+# This module defines
+#  GTEST_INCLUDE_DIR, directory containing headers
+#  GTEST_LIBS, directory containing gtest libraries
+#  GTEST_STATIC_LIB, path to libgtest.a
+#  GTEST_SHARED_LIB, path to libgtest's shared library
+#  GTEST_FOUND, whether gtest has been found
+
+find_path(GTEST_INCLUDE_DIR NAMES gtest/gtest.h gtest.h PATHS ${CMAKE_SOURCE_DIR}/gtest/include NO_DEFAULT_PATH)
+find_library(GTEST_LIBRARIES NAMES gtest PATHS ${CMAKE_SOURCE_DIR}/gtest/lib NO_DEFAULT_PATH)
+
+if (GTEST_INCLUDE_DIR )
+  message(STATUS "Found the GTest includes: ${GTEST_INCLUDE_DIR}")
+endif ()
+
+
+if (GTEST_INCLUDE_DIR AND GTEST_LIBRARIES)
+  set(GTEST_FOUND TRUE)
+  get_filename_component( GTEST_LIBS ${GTEST_LIBRARIES} PATH )
+  set(GTEST_LIB_NAME gtest)
+  set(GTEST_STATIC_LIB ${GTEST_LIBS}/${CMAKE_STATIC_LIBRARY_PREFIX}${GTEST_LIB_NAME}${CMAKE_STATIC_LIBRARY_SUFFIX})
+  set(GTEST_MAIN_STATIC_LIB ${GTEST_LIBS}/${CMAKE_STATIC_LIBRARY_PREFIX}${GTEST_LIB_NAME}_main${CMAKE_STATIC_LIBRARY_SUFFIX})
+  set(GTEST_SHARED_LIB ${GTEST_LIBS}/${CMAKE_SHARED_LIBRARY_PREFIX}${GTEST_LIB_NAME}${CMAKE_SHARED_LIBRARY_SUFFIX})
+else ()
+  set(GTEST_FOUND FALSE)
+endif ()
+
+if (GTEST_FOUND)
+  if (NOT GTest_FIND_QUIETLY)
+    message(STATUS "Found the GTest library: ${GTEST_LIBRARIES}")
+  endif ()
+else ()
+  if (NOT GTest_FIND_QUIETLY)
+    set(GTEST_ERR_MSG "Could not find the GTest library. Looked in ")
+    if ( _gtest_roots )
+      set(GTEST_ERR_MSG "${GTEST_ERR_MSG} in ${_gtest_roots}.")
+    else ()
+      set(GTEST_ERR_MSG "${GTEST_ERR_MSG} system search paths.")
+    endif ()
+    if (GTest_FIND_REQUIRED)
+      message(FATAL_ERROR "${GTEST_ERR_MSG}")
+    else (GTest_FIND_REQUIRED)
+      message(STATUS "${GTEST_ERR_MSG}")
+    endif (GTest_FIND_REQUIRED)
+  endif ()
+endif ()
+
+mark_as_advanced(
+  GTEST_INCLUDE_DIR
+  GTEST_LIBS
+  GTEST_LIBRARIES
+  GTEST_STATIC_LIB
+  GTEST_SHARED_LIB
+)

cmake/modules/FindLSan.cmake

@@ -1,13 +0,0 @@
-set(LSan_LIB_NAME lsan)
-
-find_library(LSan_LIBRARY
-  NAMES liblsan.so liblsan.so.0 liblsan.so.0.0.0
-  PATHS ${SANITIZER_PATH} /usr/lib64 /usr/lib /usr/local/lib64 /usr/local/lib ${CMAKE_PREFIX_PATH}/lib)
-
-include(FindPackageHandleStandardArgs)
-find_package_handle_standard_args(LSan DEFAULT_MSG
-  LSan_LIBRARY)
-
-mark_as_advanced(
-  LSan_LIBRARY
-  LSan_LIB_NAME)

cmake/modules/FindNccl.cmake

@@ -1,58 +0,0 @@
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-# http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-#
-# Tries to find NCCL headers and libraries.
-#
-# Usage of this module as follows:
-#
-#  find_package(NCCL)
-#
-# Variables used by this module, they can change the default behaviour and need
-# to be set before calling find_package:
-#
-#  NCCL_ROOT - When set, this path is inspected instead of standard library
-#              locations as the root of the NCCL installation.
-#              The environment variable NCCL_ROOT overrides this veriable.
-#
-# This module defines
-#  Nccl_FOUND, whether nccl has been found
-#  NCCL_INCLUDE_DIR, directory containing header
-#  NCCL_LIBRARY, directory containing nccl library
-#  NCCL_LIB_NAME, nccl library name
-#
-# This module assumes that the user has already called find_package(CUDA)
-
-
-set(NCCL_LIB_NAME nccl_static)
-
-find_path(NCCL_INCLUDE_DIR
-  NAMES nccl.h
-  PATHS $ENV{NCCL_ROOT}/include ${NCCL_ROOT}/include ${CUDA_INCLUDE_DIRS} /usr/include)
-
-find_library(NCCL_LIBRARY
-  NAMES ${NCCL_LIB_NAME}
-  PATHS $ENV{NCCL_ROOT}/lib ${NCCL_ROOT}/lib ${CUDA_INCLUDE_DIRS}/../lib /usr/lib)
-
-if (NCCL_INCLUDE_DIR AND NCCL_LIBRARY)
-  get_filename_component(NCCL_LIBRARY ${NCCL_LIBRARY} PATH)
-endif ()
-
-include(FindPackageHandleStandardArgs)
-find_package_handle_standard_args(Nccl DEFAULT_MSG
-                                  NCCL_INCLUDE_DIR NCCL_LIBRARY)
-
-mark_as_advanced(
-  NCCL_INCLUDE_DIR
-  NCCL_LIBRARY
-  NCCL_LIB_NAME
-)

cmake/modules/FindTSan.cmake

@@ -1,13 +0,0 @@
-set(TSan_LIB_NAME tsan)
-
-find_library(TSan_LIBRARY
-  NAMES libtsan.so libtsan.so.0 libtsan.so.0.0.0
-  PATHS ${SANITIZER_PATH} /usr/lib64 /usr/lib /usr/local/lib64 /usr/local/lib ${CMAKE_PREFIX_PATH}/lib)
-
-include(FindPackageHandleStandardArgs)
-find_package_handle_standard_args(TSan DEFAULT_MSG
-  TSan_LIBRARY)
-
-mark_as_advanced(
-  TSan_LIBRARY
-  TSan_LIB_NAME)

demo/README.md

@@ -135,7 +135,6 @@ Send a PR to add a one sentence description:)
 
 ## Awards
 - [John Chambers Award](http://stat-computing.org/awards/jmc/winners.html) - 2016 Winner: XGBoost R Package, by Tong He (Simon Fraser University) and Tianqi Chen (University of Washington)
-- [InfoWorld’s 2019 Technology of the Year Award](https://www.infoworld.com/article/3336072/application-development/infoworlds-2019-technology-of-the-year-award-winners.html)
 
 ## Windows Binaries
 Unofficial windows binaries and instructions on how to use them are hosted on [Guido Tapia's blog](http://www.picnet.com.au/blogs/guido/post/2016/09/22/xgboost-windows-x64-binaries-for-download/)

demo/binary_classification/README.md

@@ -62,7 +62,7 @@ test:data = "agaricus.txt.test"
 We use the tree booster and logistic regression objective in our setting. This indicates that we accomplish our task using classic gradient boosting regression tree(GBRT), which is a promising method for binary classification.
 
 The parameters shown in the example gives the most common ones that are needed to use xgboost.
-If you are interested in more parameter settings, the complete parameter settings and detailed descriptions are [here](../../doc/parameter.rst). Besides putting the parameters in the configuration file, we can set them by passing them as arguments as below:
+If you are interested in more parameter settings, the complete parameter settings and detailed descriptions are [here](../../doc/parameter.md). Besides putting the parameters in the configuration file, we can set them by passing them as arguments as below:
 
 ```
 ../../xgboost mushroom.conf max_depth=6

demo/binary_classification/mapfeat.py

@@ -18,7 +18,7 @@ def loadfmap( fname ):
             if it.strip() == '':
                 continue
             k , v = it.split('=')
-            fmap[ idx ][ v ] = len(nmap)
+            fmap[ idx ][ v ] = len(nmap) + 1
             nmap[ len(nmap) ] = ftype+'='+k
     return fmap, nmap
 

demo/guide-python/custom_objective.py

@@ -33,9 +33,9 @@ def logregobj(preds, dtrain):
 # Take this in mind when you use the customization, and maybe you need write customized evaluation function
 def evalerror(preds, dtrain):
     labels = dtrain.get_label()
-    # return a pair metric_name, result. The metric name must not contain a colon (:) or a space
+    # return a pair metric_name, result. The metric name must not contain a colon (:)
     # since preds are margin(before logistic transformation, cutoff at 0)
-    return 'my-error', float(sum(labels != (preds > 0.0))) / len(labels)
+    return 'error', float(sum(labels != (preds > 0.0))) / len(labels)
 
 # training with customized objective, we can also do step by step training
 # simply look at xgboost.py's implementation of train

demo/guide-python/runall.sh

@@ -1,5 +1,5 @@
 #!/bin/bash
-export PYTHONPATH=$PYTHONPATH:../../python-package
+export PYTHONPATH=PYTHONPATH:../../python-package
 python basic_walkthrough.py
 python custom_objective.py
 python boost_from_prediction.py

demo/kaggle-higgs/higgs-train.R

@@ -24,9 +24,9 @@ param <- list("objective" = "binary:logitraw",
               "silent" = 1,
               "nthread" = 16)
 watchlist <- list("train" = xgmat)
-nrounds = 120
+nround = 120
 print ("loading data end, start to boost trees")
-bst = xgb.train(param, xgmat, nrounds, watchlist );
+bst = xgb.train(param, xgmat, nround, watchlist );
 # save out model
 xgb.save(bst, "higgs.model")
 print ('finish training')

demo/kaggle-higgs/speedtest.R

@@ -39,9 +39,9 @@ for (i in 1:length(threads)){
                   "silent" = 1,
                   "nthread" = thread)
     watchlist <- list("train" = xgmat)
-    nrounds = 120
+    nround = 120
     print ("loading data end, start to boost trees")
-    bst = xgb.train(param, xgmat, nrounds, watchlist );
+    bst = xgb.train(param, xgmat, nround, watchlist );
     # save out model
     xgb.save(bst, "higgs.model")
     print ('finish training')

demo/kaggle-otto/README.MD

@@ -1,4 +1,4 @@
-Benchmark for Otto Group Competition
+Benckmark for Otto Group Competition
 =========
 
 This is a folder containing the benchmark for the [Otto Group Competition on Kaggle](http://www.kaggle.com/c/otto-group-product-classification-challenge).
@@ -20,3 +20,5 @@ devtools::install_github('tqchen/xgboost',subdir='R-package')
 ```
 
 Windows users may need to install [RTools](http://cran.r-project.org/bin/windows/Rtools/) first.
+
+

demo/kaggle-otto/otto_train_pred.R

@@ -23,13 +23,13 @@ param <- list("objective" = "multi:softprob",
               "nthread" = 8)
 
 # Run Cross Validation
-cv.nrounds = 50
+cv.nround = 50
 bst.cv = xgb.cv(param=param, data = x[trind,], label = y, 
-                nfold = 3, nrounds=cv.nrounds)
+                nfold = 3, nrounds=cv.nround)
 
 # Train the model
-nrounds = 50
-bst = xgboost(param=param, data = x[trind,], label = y, nrounds=nrounds)
+nround = 50
+bst = xgboost(param=param, data = x[trind,], label = y, nrounds=nround)
 
 # Make prediction
 pred = predict(bst,x[teind,])

demo/kaggle-otto/understandingXGBoostModel.Rmd

@@ -121,19 +121,19 @@ param <- list("objective" = "multi:softprob",
               "eval_metric" = "mlogloss",
               "num_class" = numberOfClasses)
 
-cv.nrounds <- 5
+cv.nround <- 5
 cv.nfold <- 3
 
 bst.cv = xgb.cv(param=param, data = trainMatrix, label = y,
-                nfold = cv.nfold, nrounds = cv.nrounds)
+                nfold = cv.nfold, nrounds = cv.nround)
 ```
 > As we can see the error rate is low on the test dataset (for a 5mn trained model).
 
 Finally, we are ready to train the real model!!!
 
 ```{r modelTraining}
-nrounds = 50
-bst = xgboost(param=param, data = trainMatrix, label = y, nrounds=nrounds)
+nround = 50
+bst = xgboost(param=param, data = trainMatrix, label = y, nrounds=nround)
 ```
 
 Model understanding
@@ -142,7 +142,7 @@ Model understanding
 Feature importance
 ------------------
 
-So far, we have built a model made of **`r nrounds`** trees.
+So far, we have built a model made of **`r nround`** trees.
 
 To build a tree, the dataset is divided recursively several times. At the end of the process, you get groups of observations (here, these observations are properties regarding **Otto** products).
 

demo/multiclass_classification/README.md

@@ -1,10 +1,10 @@
 Demonstrating how to use XGBoost accomplish Multi-Class classification task on [UCI Dermatology dataset](https://archive.ics.uci.edu/ml/datasets/Dermatology)
 
-Make sure you make xgboost python module in ../../python
+Make sure you make make xgboost python module in ../../python
 
 1. Run runexp.sh
 ```bash
 ./runexp.sh
 ```
 
-**R version** please see the `train.R`.
+

demo/multiclass_classification/train.R

@@ -1,64 +0,0 @@
-library(data.table)
-library(xgboost)
-
-if (!file.exists("./dermatology.data")) {
-  download.file(
-    "https://archive.ics.uci.edu/ml/machine-learning-databases/dermatology/dermatology.data",
-    "dermatology.data",
-    method = "curl"
-  )
-}
-
-df <- fread("dermatology.data", sep = ",", header = FALSE)
-
-df[, `:=`(V34 = as.integer(ifelse(V34 == "?", 0L, V34)),
-          V35 = V35 - 1L)]
-
-idx <- sample(nrow(df), size = round(0.7 * nrow(df)), replace = FALSE)
-
-train <- df[idx,]
-test <- df[-idx,]
-
-train_x <- train[, 1:34]
-train_y <- train[, V35]
-
-test_x <- test[, 1:34]
-test_y <- test[, V35]
-
-xg_train <- xgb.DMatrix(data = as.matrix(train_x), label = train_y)
-xg_test = xgb.DMatrix(as.matrix(test_x), label = test_y)
-
-params <- list(
-  objective = 'multi:softmax',
-  num_class = 6,
-  max_depth = 6,
-  nthread = 4,
-  eta = 0.1
-)
-
-watchlist = list(train = xg_train, test = xg_test)
-
-bst <- xgb.train(
-  params = params,
-  data = xg_train,
-  watchlist = watchlist,
-  nrounds = 5
-)
-
-pred <- predict(bst, xg_test)
-error_rate <- sum(pred != test_y) / length(test_y)
-print(paste("Test error using softmax =", error_rate))
-
-# do the same thing again, but output probabilities
-params$objective <- 'multi:softprob'
-bst <- xgb.train(params, xg_train, nrounds = 5, watchlist)
-
-pred_prob <- predict(bst, xg_test)
-
-pred_mat <- matrix(pred_prob, ncol = 6, byrow = TRUE)
-# validation
-# rowSums(pred_mat)
-
-pred_label <- apply(pred_mat, 1, which.max) - 1L
-error_rate = sum(pred_label != test_y) / length(test_y)
-print(paste("Test error using softprob =", error_rate))

demo/rank/README.md

@@ -1,6 +1,6 @@
 Learning to rank
 ====
-XGBoost supports accomplishing ranking tasks. In ranking scenario, data are often grouped and we need the [group information file](../../doc/tutorials/input_format.rst#group-input-format) to specify ranking tasks. The model used in XGBoost for ranking is the LambdaRank, this function is not yet completed. Currently, we provide pairwise rank.
+XGBoost supports accomplishing ranking tasks. In ranking scenario, data are often grouped and we need the [group information file](../../doc/input_format.md#group-input-format) to specify ranking tasks. The model used in XGBoost for ranking is the LambdaRank, this function is not yet completed. Currently, we provide pairwise rank.
 
 ### Parameters
 The configuration setting is similar to the regression and binary classification setting, except user need to specify the objectives:
@@ -14,28 +14,8 @@ For more usage details please refer to the [binary classification demo](../binar
 
 Instructions
 ====
-The dataset for ranking demo is from LETOR04 MQ2008 fold1.
-Before running the examples, you need to get the data by running:
+The dataset for ranking demo is from LETOR04 MQ2008 fold1,
+You can use the following command to run the example
 
-```
-./wgetdata.sh
-```
-
-### Command Line
-Run the example:
-```
-./runexp.sh
-```
-
-### Python
-There are two ways of doing ranking in python.  
-
-Run the example using `xgboost.train`:
-```
-python rank.py
-```
-
-Run the example using `XGBRanker`:
-```
-python rank_sklearn.py
-```
+Get the data: ./wgetdata.sh
+Run the example: ./runexp.sh

demo/rank/rank.py

@@ -1,41 +0,0 @@
-#!/usr/bin/python
-import xgboost as xgb
-from xgboost import DMatrix
-from sklearn.datasets import load_svmlight_file
-
-
-#  This script demonstrate how to do ranking with xgboost.train
-x_train, y_train = load_svmlight_file("mq2008.train")
-x_valid, y_valid = load_svmlight_file("mq2008.vali")
-x_test, y_test = load_svmlight_file("mq2008.test")
-
-group_train = []
-with open("mq2008.train.group", "r") as f:
-    data = f.readlines()
-    for line in data:
-        group_train.append(int(line.split("\n")[0]))
-
-group_valid = []
-with open("mq2008.vali.group", "r") as f:
-    data = f.readlines()
-    for line in data:
-        group_valid.append(int(line.split("\n")[0]))
-
-group_test = []
-with open("mq2008.test.group", "r") as f:
-    data = f.readlines()
-    for line in data:
-        group_test.append(int(line.split("\n")[0]))
-
-train_dmatrix = DMatrix(x_train, y_train)
-valid_dmatrix = DMatrix(x_valid, y_valid)
-test_dmatrix = DMatrix(x_test)
-
-train_dmatrix.set_group(group_train)
-valid_dmatrix.set_group(group_valid)
-
-params = {'objective': 'rank:pairwise', 'eta': 0.1, 'gamma': 1.0,
-               'min_child_weight': 0.1, 'max_depth': 6}
-xgb_model = xgb.train(params, train_dmatrix, num_boost_round=4,
-                           evals=[(valid_dmatrix, 'validation')])
-pred = xgb_model.predict(test_dmatrix)

demo/rank/rank_sklearn.py

@@ -1,35 +0,0 @@
-#!/usr/bin/python
-import xgboost as xgb
-from sklearn.datasets import load_svmlight_file
-
-
-#  This script demonstrate how to do ranking with XGBRanker
-x_train, y_train = load_svmlight_file("mq2008.train")
-x_valid, y_valid = load_svmlight_file("mq2008.vali")
-x_test, y_test = load_svmlight_file("mq2008.test")
-
-group_train = []
-with open("mq2008.train.group", "r") as f:
-    data = f.readlines()
-    for line in data:
-        group_train.append(int(line.split("\n")[0]))
-
-group_valid = []
-with open("mq2008.vali.group", "r") as f:
-    data = f.readlines()
-    for line in data:
-        group_valid.append(int(line.split("\n")[0]))
-
-group_test = []
-with open("mq2008.test.group", "r") as f:
-    data = f.readlines()
-    for line in data:
-        group_test.append(int(line.split("\n")[0]))
-
-params = {'objective': 'rank:pairwise', 'learning_rate': 0.1,
-          'gamma': 1.0, 'min_child_weight': 0.1,
-          'max_depth': 6, 'n_estimators': 4}
-model = xgb.sklearn.XGBRanker(**params)
-model.fit(x_train, y_train, group_train,
-          eval_set=[(x_valid, y_valid)], eval_group=[group_valid])
-pred = model.predict(x_test)

demo/rank/runexp.sh

@@ -1,5 +1,11 @@
-#!/bin/bash
+python trans_data.py train.txt mq2008.train mq2008.train.group
+
+python trans_data.py test.txt mq2008.test mq2008.test.group
+
+python trans_data.py vali.txt mq2008.vali mq2008.vali.group
 
 ../../xgboost mq2008.conf
 
 ../../xgboost mq2008.conf task=pred model_in=0004.model
+
+

demo/rank/wgetdata.sh

@@ -1,10 +1,4 @@
 #!/bin/bash
-wget https://s3-us-west-2.amazonaws.com/xgboost-examples/MQ2008.rar
+wget http://research.microsoft.com/en-us/um/beijing/projects/letor/LETOR4.0/Data/MQ2008.rar
 unrar x MQ2008.rar
 mv -f MQ2008/Fold1/*.txt .
-
-python trans_data.py train.txt mq2008.train mq2008.train.group
-
-python trans_data.py test.txt mq2008.test mq2008.test.group
-
-python trans_data.py vali.txt mq2008.vali mq2008.vali.group

doc/R-package/discoverYourData.md

@@ -222,7 +222,7 @@ The code below is very usual. For more information, you can look at the document
 
 ```r
 bst <- xgboost(data = sparse_matrix, label = output_vector, max.depth = 4,
-               eta = 1, nthread = 2, nrounds = 10,objective = "binary:logistic")
+               eta = 1, nthread = 2, nround = 10,objective = "binary:logistic")
 ```
 
 ```
@@ -244,7 +244,7 @@ A model which fits too well may [overfit](http://en.wikipedia.org/wiki/Overfitti
 
 > Here you can see the numbers decrease until line 7 and then increase.
 >
-> It probably means we are overfitting. To fix that I should reduce the number of rounds to `nrounds = 4`. I will let things like that because I don't really care for the purpose of this example :-)
+> It probably means we are overfitting. To fix that I should reduce the number of rounds to `nround = 4`. I will let things like that because I don't really care for the purpose of this example :-)
 
 Feature importance
 ------------------
@@ -448,7 +448,7 @@ train <- agaricus.train
 test <- agaricus.test
 
 #Random Forest™ - 1000 trees
-bst <- xgboost(data = train$data, label = train$label, max.depth = 4, num_parallel_tree = 1000, subsample = 0.5, colsample_bytree =0.5, nrounds = 1, objective = "binary:logistic")
+bst <- xgboost(data = train$data, label = train$label, max.depth = 4, num_parallel_tree = 1000, subsample = 0.5, colsample_bytree =0.5, nround = 1, objective = "binary:logistic")
 ```
 
 ```
@@ -457,7 +457,7 @@ bst <- xgboost(data = train$data, label = train$label, max.depth = 4, num_parall
 
 ```r
 #Boosting - 3 rounds
-bst <- xgboost(data = train$data, label = train$label, max.depth = 4, nrounds = 3, objective = "binary:logistic")
+bst <- xgboost(data = train$data, label = train$label, max.depth = 4, nround = 3, objective = "binary:logistic")
 ```
 
 ```

doc/R-package/xgboostPresentation.md

@@ -176,13 +176,13 @@ In a *sparse* matrix, cells containing `0` are not stored in memory. Therefore,
 We will train decision tree model using the following parameters:
 
 * `objective = "binary:logistic"`: we will train a binary classification model ;
-* `max.depth = 2`: the trees won't be deep, because our case is very simple ;
+* `max.deph = 2`: the trees won't be deep, because our case is very simple ;
 * `nthread = 2`: the number of cpu threads we are going to use;
-* `nrounds = 2`: there will be two passes on the data, the second one will enhance the model by further reducing the difference between ground truth and prediction.
+* `nround = 2`: there will be two passes on the data, the second one will enhance the model by further reducing the difference between ground truth and prediction.
 
 
 ```r
-bstSparse <- xgboost(data = train$data, label = train$label, max.depth = 2, eta = 1, nthread = 2, nrounds = 2, objective = "binary:logistic")
+bstSparse <- xgboost(data = train$data, label = train$label, max.depth = 2, eta = 1, nthread = 2, nround = 2, objective = "binary:logistic")
 ```
 
 ```
@@ -200,7 +200,7 @@ Alternatively, you can put your dataset in a *dense* matrix, i.e. a basic **R**
 
 
 ```r
-bstDense <- xgboost(data = as.matrix(train$data), label = train$label, max.depth = 2, eta = 1, nthread = 2, nrounds = 2, objective = "binary:logistic")
+bstDense <- xgboost(data = as.matrix(train$data), label = train$label, max.depth = 2, eta = 1, nthread = 2, nround = 2, objective = "binary:logistic")
 ```
 
 ```
@@ -215,7 +215,7 @@ bstDense <- xgboost(data = as.matrix(train$data), label = train$label, max.depth
 
 ```r
 dtrain <- xgb.DMatrix(data = train$data, label = train$label)
-bstDMatrix <- xgboost(data = dtrain, max.depth = 2, eta = 1, nthread = 2, nrounds = 2, objective = "binary:logistic")
+bstDMatrix <- xgboost(data = dtrain, max.depth = 2, eta = 1, nthread = 2, nround = 2, objective = "binary:logistic")
 ```
 
 ```
@@ -232,13 +232,13 @@ One of the simplest way to see the training progress is to set the `verbose` opt
 
 ```r
 # verbose = 0, no message
-bst <- xgboost(data = dtrain, max.depth = 2, eta = 1, nthread = 2, nrounds = 2, objective = "binary:logistic", verbose = 0)
+bst <- xgboost(data = dtrain, max.depth = 2, eta = 1, nthread = 2, nround = 2, objective = "binary:logistic", verbose = 0)
 ```
 
 
 ```r
 # verbose = 1, print evaluation metric
-bst <- xgboost(data = dtrain, max.depth = 2, eta = 1, nthread = 2, nrounds = 2, objective = "binary:logistic", verbose = 1)
+bst <- xgboost(data = dtrain, max.depth = 2, eta = 1, nthread = 2, nround = 2, objective = "binary:logistic", verbose = 1)
 ```
 
 ```
@@ -249,7 +249,7 @@ bst <- xgboost(data = dtrain, max.depth = 2, eta = 1, nthread = 2, nrounds = 2,
 
 ```r
 # verbose = 2, also print information about tree
-bst <- xgboost(data = dtrain, max.depth = 2, eta = 1, nthread = 2, nrounds = 2, objective = "binary:logistic", verbose = 2)
+bst <- xgboost(data = dtrain, max.depth = 2, eta = 1, nthread = 2, nround = 2, objective = "binary:logistic", verbose = 2)
 ```
 
 ```
@@ -372,7 +372,7 @@ For the purpose of this example, we use `watchlist` parameter. It is a list of `
 ```r
 watchlist <- list(train=dtrain, test=dtest)
 
-bst <- xgb.train(data=dtrain, max.depth=2, eta=1, nthread = 2, nrounds=2, watchlist=watchlist, objective = "binary:logistic")
+bst <- xgb.train(data=dtrain, max.depth=2, eta=1, nthread = 2, nround=2, watchlist=watchlist, objective = "binary:logistic")
 ```
 
 ```
@@ -380,7 +380,7 @@ bst <- xgb.train(data=dtrain, max.depth=2, eta=1, nthread = 2, nrounds=2, watchl
 ## [1]	train-error:0.022263	test-error:0.021726
 ```
 
-**XGBoost** has computed at each round the same average error metric than seen above (we set `nrounds` to 2, that is why we have two lines). Obviously, the `train-error` number is related to the training dataset (the one the algorithm learns from) and the `test-error` number to the test dataset.
+**XGBoost** has computed at each round the same average error metric than seen above (we set `nround` to 2, that is why we have two lines). Obviously, the `train-error` number is related to the training dataset (the one the algorithm learns from) and the `test-error` number to the test dataset.
 
 Both training and test error related metrics are very similar, and in some way, it makes sense: what we have learned from the training dataset matches the observations from the test dataset.
 
@@ -390,7 +390,7 @@ For a better understanding of the learning progression, you may want to have som
 
 
 ```r
-bst <- xgb.train(data=dtrain, max.depth=2, eta=1, nthread = 2, nrounds=2, watchlist=watchlist, eval.metric = "error", eval.metric = "logloss", objective = "binary:logistic")
+bst <- xgb.train(data=dtrain, max.depth=2, eta=1, nthread = 2, nround=2, watchlist=watchlist, eval.metric = "error", eval.metric = "logloss", objective = "binary:logistic")
 ```
 
 ```
@@ -407,7 +407,7 @@ Until now, all the learnings we have performed were based on boosting trees. **X
 
 
 ```r
-bst <- xgb.train(data=dtrain, booster = "gblinear", max.depth=2, nthread = 2, nrounds=2, watchlist=watchlist, eval.metric = "error", eval.metric = "logloss", objective = "binary:logistic")
+bst <- xgb.train(data=dtrain, booster = "gblinear", max.depth=2, nthread = 2, nround=2, watchlist=watchlist, eval.metric = "error", eval.metric = "logloss", objective = "binary:logistic")
 ```
 
 ```
@@ -445,7 +445,7 @@ dtrain2 <- xgb.DMatrix("dtrain.buffer")
 ```
 
 ```r
-bst <- xgb.train(data=dtrain2, max.depth=2, eta=1, nthread = 2, nrounds=2, watchlist=watchlist, objective = "binary:logistic")
+bst <- xgb.train(data=dtrain2, max.depth=2, eta=1, nthread = 2, nround=2, watchlist=watchlist, objective = "binary:logistic")
 ```
 
 ```
@@ -576,8 +576,8 @@ print(class(rawVec))
 bst3 <- xgb.load(rawVec)
 pred3 <- predict(bst3, test$data)
 
-# pred3 should be identical to pred
-print(paste("sum(abs(pred3-pred))=", sum(abs(pred3-pred))))
+# pred2 should be identical to pred
+print(paste("sum(abs(pred3-pred))=", sum(abs(pred2-pred))))
 ```
 
 ```

doc/build.md

@@ -0,0 +1,377 @@
+Installation Guide
+==================
+
+This page gives instructions on how to build and install the xgboost package from
+scratch on various systems. It consists of two steps:
+
+1. First build the shared library from the C++ codes (`libxgboost.so` for Linux/OSX and `xgboost.dll` for Windows).
+   - Exception: for R-package installation please directly refer to the R package section.
+2. Then install the language packages (e.g. Python Package).
+
+***Important*** the newest version of xgboost uses submodule to maintain packages. So when you clone the repo, remember to use the recursive option as follows.
+```bash
+git clone --recursive https://github.com/dmlc/xgboost
+```
+For windows users who use github tools, you can open the git shell, and type the following command.
+```bash
+git submodule init
+git submodule update
+```
+
+Please refer to [Trouble Shooting Section](#trouble-shooting) first if you had any problem
+during installation. If the instructions do not work for you, please feel free
+to ask questions at [xgboost/issues](https://github.com/dmlc/xgboost/issues), or
+even better to send pull request if you can fix the problem.
+
+## Contents
+- [Build the Shared Library](#build-the-shared-library)
+  - [Building on Ubuntu/Debian](#building-on-ubuntu-debian)
+  - [Building on macOS](#building-on-macos)
+  - [Building on Windows](#building-on-windows)
+  - [Building with GPU support](#building-with-gpu-support)
+  - [Windows Binaries](#windows-binaries)
+  - [Customized Building](#customized-building)
+- [Python Package Installation](#python-package-installation)
+- [R Package Installation](#r-package-installation)
+- [Trouble Shooting](#trouble-shooting)
+
+## Build the Shared Library
+
+Our goal is to build the shared library:
+- On Linux/OSX the target library is `libxgboost.so`
+- On Windows the target library is `xgboost.dll`
+
+The minimal building requirement is
+
+- A recent c++ compiler supporting C++ 11 (g++-4.8 or higher)
+
+We can edit `make/config.mk` to change the compile options, and then build by
+`make`. If everything goes well, we can go to the specific language installation section.
+
+### Building on Ubuntu/Debian
+
+On Ubuntu, one builds xgboost by
+
+```bash
+git clone --recursive https://github.com/dmlc/xgboost
+cd xgboost; make -j4
+```
+
+### Building on macOS
+
+**Install with pip - simple method**
+
+First, make sure you obtained *gcc-5* (newer version does not work with this method yet). Note: installation of `gcc` can take a while (~ 30 minutes)
+
+```bash
+brew install gcc5
+```
+
+You might need to run the following command with `sudo` if you run into some permission errors:
+
+```bash
+pip install xgboost
+```
+
+**Build from the source code - advanced method**
+
+First, obtain gcc-7.x.x with brew (https://brew.sh/) if you want multi-threaded version, otherwise, Clang is ok if OpenMP / multi-threaded is not required. Note: installation of `gcc` can take a while (~ 30 minutes)
+
+```bash
+brew install gcc
+```
+
+Now, clone the repository
+
+```bash
+git clone --recursive https://github.com/dmlc/xgboost
+cd xgboost; cp make/config.mk ./config.mk
+```
+
+Open config.mk and uncomment these two lines
+
+```config.mk
+export CC = gcc
+export CXX = g++
+```
+
+and replace these two lines into(5 or 6 or 7; depending on your gcc-version)
+
+```config.mk
+export CC = gcc-7
+export CXX = g++-7
+```
+
+To find your gcc version
+
+```bash
+gcc-version
+```
+
+and build using the following commands
+
+```bash
+make -j4
+```
+head over to `Python Package Installation` for the next steps
+
+### Building on Windows
+You need to first clone the xgboost repo with recursive option clone the submodules.
+If you are using github tools, you can open the git-shell, and type the following command.
+We recommend using [Git for Windows](https://git-for-windows.github.io/)
+because it brings a standard bash shell. This will highly ease the installation process.
+
+```bash
+git submodule init
+git submodule update
+```
+
+XGBoost support both build by MSVC or MinGW. Here is how you can build xgboost library using MinGW.
+
+After installing [Git for Windows](https://git-for-windows.github.io/), you should have a shortcut `Git Bash`.
+All the following steps are in the `Git Bash`.
+
+In MinGW, `make` command comes with the name `mingw32-make`. You can add the following line into the `.bashrc` file.
+```bash
+alias make='mingw32-make'
+```
+(On 64-bit Windows, you should get [mingw64](https://sourceforge.net/projects/mingw-w64/) instead.) Make sure
+that the path to MinGW is in the system PATH.
+
+To build with MinGW, type:
+
+```bash
+cp make/mingw64.mk config.mk; make -j4
+```
+
+To build with Visual Studio 2013 use cmake. Make sure you have a recent version of cmake added to your path and then from the xgboost directory:
+
+```bash
+mkdir build
+cd build
+cmake .. -G"Visual Studio 12 2013 Win64"
+```
+
+This specifies an out of source build using the MSVC 12 64 bit generator. Open the .sln file in the build directory and build with Visual Studio. To use the Python module you can copy `xgboost.dll` into python-package\xgboost.
+
+Other versions of Visual Studio may work but are untested.
+
+### Building with GPU support
+
+XGBoost can be built with GPU support for both Linux and Windows using cmake. GPU support works with the Python package as well as the CLI version. See [Installing R package with GPU support](#installing-r-package-with-gpu-support) for special instructions for R.
+
+An up-to-date version of the CUDA toolkit is required.
+
+From the command line on Linux starting from the xgboost directory:
+
+```bash
+$ mkdir build
+$ cd build
+$ cmake .. -DUSE_CUDA=ON
+$ make -j
+```
+**Windows requirements** for GPU build: only Visual C++ 2015 or 2013 with CUDA v8.0 were fully tested. Either install Visual C++ 2015 Build Tools separately, or as a part of Visual Studio 2015. If you already have Visual Studio 2017, the Visual C++ 2015 Toolchain componenet has to be installed using the VS 2017 Installer. Likely, you would need to use the VS2015 x64 Native Tools command prompt to run the cmake commands given below. In some situations, however, things run just fine from MSYS2 bash command line.
+
+On Windows, using cmake, see what options for Generators you have for cmake, and choose one with [arch] replaced by Win64:
+```bash
+cmake -help
+```
+Then run cmake as:
+```bash
+$ mkdir build
+$ cd build
+$ cmake .. -G"Visual Studio 14 2015 Win64" -DUSE_CUDA=ON
+```
+To speed up compilation, compute version specific to your GPU could be passed to cmake as, e.g., `-DGPU_COMPUTE_VER=50`.
+The above cmake configuration run will create an xgboost.sln solution file in the build directory. Build this solution in release mode as a x64 build, either from Visual studio or from command line:
+```
+cmake --build . --target xgboost --config Release
+```
+If build seems to use only a single process, you might try to append an option like ` -- /m:6` to the above command.
+
+### Windows Binaries
+
+After the build process successfully ends, you will find a `xgboost.dll` library file inside `./lib/` folder, copy this file to the the API package folder like `python-package/xgboost` if you are using *python* API. And you are good to follow the below instructions.
+
+Unofficial windows binaries and instructions on how to use them are hosted on [Guido Tapia's blog](http://www.picnet.com.au/blogs/guido/post/2016/09/22/xgboost-windows-x64-binaries-for-download/)
+
+### Customized Building
+
+The configuration of xgboost can be modified by ```config.mk```
+- modify configuration on various distributed filesystem such as HDFS/Amazon S3/...
+- First copy [make/config.mk](../make/config.mk) to the project root, on which
+  any local modification will be ignored by git, then modify the according flags.
+
+
+
+## Python Package Installation
+
+The python package is located at [python-package](../python-package).
+There are several ways to install the package:
+
+1. Install system-widely, which requires root permission
+
+   ```bash
+   cd python-package; sudo python setup.py install
+   ```
+
+   You will however need Python `distutils` module for this to
+   work. It is often part of the core python package or it can be installed using your
+   package manager, e.g. in Debian use
+
+   ```bash
+   sudo apt-get install python-setuptools
+   ```
+
+   *NOTE: If you recompiled xgboost, then you need to reinstall it again to
+    make the new library take effect*
+
+2. Only set the environment variable `PYTHONPATH` to tell python where to find
+   the library. For example, assume we cloned `xgboost` on the home directory
+   `~`. then we can added the following line in `~/.bashrc`.
+    It is ***recommended for developers*** who may change the codes. The changes will be immediately reflected once you pulled the code and rebuild the project (no need to call ```setup``` again)
+
+    ```bash
+    export PYTHONPATH=~/xgboost/python-package
+    ```
+
+3. Install only for the current user.
+
+    ```bash
+    cd python-package; python setup.py develop --user
+    ```
+
+4. If you are installing the latest xgboost version which requires compilation, add MinGW to the system PATH:
+
+    ```python
+    import os
+    os.environ['PATH'] = os.environ['PATH'] + ';C:\\Program Files\\mingw-w64\\x86_64-5.3.0-posix-seh-rt_v4-rev0\\mingw64\\bin'
+    ```
+
+## R Package Installation
+
+### Installing pre-packaged version
+
+You can install xgboost from CRAN just like any other R package:
+
+```r
+install.packages("xgboost")
+```
+
+Or you can install it from our weekly updated drat repo:
+
+```r
+install.packages("drat", repos="https://cran.rstudio.com")
+drat:::addRepo("dmlc")
+install.packages("xgboost", repos="http://dmlc.ml/drat/", type = "source")
+```
+
+For OSX users, single threaded version will be installed. To install multi-threaded version,
+first follow [Building on OSX](#building-on-osx) to get the OpenMP enabled compiler, then:
+
+- Set the `Makevars` file in highest piority for R.
+
+  The point is, there are three `Makevars` : `~/.R/Makevars`, `xgboost/R-package/src/Makevars`, and `/usr/local/Cellar/r/3.2.0/R.framework/Resources/etc/Makeconf` (the last one obtained by running `file.path(R.home("etc"), "Makeconf")` in R), and `SHLIB_OPENMP_CXXFLAGS` is not set by default!! After trying, it seems that the first one has highest piority (surprise!).
+
+  Then inside R, run
+
+  ```R
+  install.packages("drat", repos="https://cran.rstudio.com")
+  drat:::addRepo("dmlc")
+  install.packages("xgboost", repos="http://dmlc.ml/drat/", type = "source")
+  ```
+
+### Installing the development version
+
+Make sure you have installed git and a recent C++ compiler supporting C++11 (e.g., g++-4.8 or higher).
+On Windows, Rtools must be installed, and its bin directory has to be added to PATH during the installation.
+And see the previous subsection for an OSX tip.
+
+Due to the use of git-submodules, `devtools::install_github` can no longer be used to install the latest version of R package.
+Thus, one has to run git to check out the code first:
+
+```bash
+git clone --recursive https://github.com/dmlc/xgboost
+cd xgboost
+git submodule init
+git submodule update
+cd R-package
+R CMD INSTALL .
+```
+
+If the last line fails because of "R: command not found", it means that R was not set up to run from command line.
+In this case, just start R as you would normally do and run the following:
+
+```r
+setwd('wherever/you/cloned/it/xgboost/R-package/')
+install.packages('.', repos = NULL, type="source")
+```
+
+The package could also be built and installed with cmake (and Visual C++ 2015 on Windows) using instructions from the next section, but without GPU support (omit the `-DUSE_CUDA=ON` cmake parameter).
+
+If all fails, try [building the shared library](#build-the-shared-library) to see whether a problem is specific to R package or not.
+
+### Installing R package with GPU support
+
+The procedure and requirements are similar as in [Building with GPU support](#building-with-gpu-support), so make sure to read it first.
+
+On Linux, starting from the xgboost directory:
+
+```bash
+mkdir build
+cd build
+cmake .. -DUSE_CUDA=ON -DR_LIB=ON
+make install -j
+```
+When default target is used, an R package shared library would be built in the `build` area.
+The `install` target, in addition, assembles the package files with this shared library under `build/R-package`, and runs `R CMD INSTALL`.
+
+On Windows, cmake with Visual C++ Build Tools (or Visual Studio) has to be used to build an R package with GPU support. Rtools must also be installed (perhaps, some other MinGW distributions with `gendef.exe` and `dlltool.exe` would work, but that was not tested).
+```bash
+mkdir build
+cd build
+cmake .. -G"Visual Studio 14 2015 Win64" -DUSE_CUDA=ON -DR_LIB=ON
+cmake --build . --target install --config Release
+```
+When `--target xgboost` is used, an R package dll would be built under `build/Release`.
+The `--target install`, in addition, assembles the package files with this dll under `build/R-package`, and runs `R CMD INSTALL`.
+
+If cmake can't find your R during the configuration step, you might provide the location of its executable to cmake like this: `-DLIBR_EXECUTABLE="C:/Program Files/R/R-3.4.1/bin/x64/R.exe"`.
+
+If on Windows you get a "permission denied" error when trying to write to ...Program Files/R/... during the package installation, create a `.Rprofile` file in your personal home directory (if you don't already have one in there), and add a line to it which specifies the location of your R packages user library, like the following:
+```r
+.libPaths( unique(c("C:/Users/USERNAME/Documents/R/win-library/3.4", .libPaths())))
+```
+You might find the exact location by running `.libPaths()` in R GUI or RStudio.
+
+## Trouble Shooting
+
+1. **Compile failed after `git pull`**
+
+   Please first update the submodules, clean all and recompile:
+
+   ```bash
+   git submodule update && make clean_all && make -j4
+   ```
+
+2. **Compile failed after `config.mk` is modified**
+
+   Need to clean all first:
+
+    ```bash
+    make clean_all && make -j4
+    ```
+
+
+3. **Makefile: dmlc-core/make/dmlc.mk: No such file or directory**
+
+   We need to recursively clone the submodule, you can do:
+
+    ```bash
+    git submodule init
+    git submodule update
+    ```
+    Alternatively, do another clone
+    ```bash
+    git clone https://github.com/dmlc/xgboost --recursive
+    ```

doc/build.rst

@@ -4,17 +4,15 @@ Installation Guide
 
 .. note:: Pre-built binary wheel for Python
 
-  If you are planning to use Python, consider installing XGBoost from a pre-built binary wheel, available from Python Package Index (PyPI). You may download and install it by running
+  If you are planning to use Python on a Linux system, consider installing XGBoost from a pre-built binary wheel. The wheel is available from Python Package Index (PyPI). You may download and install it by running
 
   .. code-block:: bash
 
-    # Ensure that you are downloading one of the following:
-    #   * xgboost-{version}-py2.py3-none-manylinux1_x86_64.whl
-    #   * xgboost-{version}-py2.py3-none-win_amd64.whl
+    # Ensure that you are downloading xgboost-{version}-py2.py3-none-manylinux1_x86_64.whl
     pip3 install xgboost
 
-  * The binary wheel will support GPU algorithms (`gpu_exact`, `gpu_hist`) on machines with NVIDIA GPUs. Please note that **training with multiple GPUs is only supported for Linux platform**. See :doc:`gpu/index`.
-  * Currently, we provide binary wheels for 64-bit Linux and Windows.
+  * This package will support GPU algorithms (`gpu_exact`, `gpu_hist`) on machines with NVIDIA GPUs.
+  * Currently, PyPI has a binary wheel only for 64-bit Linux.
 
 ****************************
 Building XGBoost from source
@@ -70,21 +68,19 @@ Our goal is to build the shared library:
 The minimal building requirement is
 
 - A recent C++ compiler supporting C++11 (g++-4.8 or higher)
-- CMake 3.2 or higher
+
+We can edit ``make/config.mk`` to change the compile options, and then build by
+``make``. If everything goes well, we can go to the specific language installation section.
 
 Building on Ubuntu/Debian
 =========================
 
-On Ubuntu, one builds XGBoost by running CMake:
+On Ubuntu, one builds XGBoost by running
 
 .. code-block:: bash
 
   git clone --recursive https://github.com/dmlc/xgboost
-  cd xgboost
-  mkdir build
-  cd build
-  cmake ..
-  make -j4
+  cd xgboost; make -j4
 
 Building on OSX
 ===============
@@ -92,11 +88,11 @@ Building on OSX
 Install with pip: simple method
 --------------------------------
 
-First, obtain ``gcc-8`` with Homebrew (https://brew.sh/) to enable multi-threading (i.e. using multiple CPU threads for training). The default Apple Clang compiler does not support OpenMP, so using the default compiler would have disabled multi-threading.
+First, make sure you obtained ``gcc-5`` (newer version does not work with this method yet). Note: installation of ``gcc`` can take a while (~ 30 minutes).
 
 .. code-block:: bash
 
-  brew install gcc@8
+  brew install gcc@5
 
 Then install XGBoost with ``pip``:
 
@@ -104,30 +100,42 @@ Then install XGBoost with ``pip``:
 
   pip3 install xgboost
 
-You might need to run the command with ``--user`` flag if you run into permission errors.
+You might need to run the command with ``sudo`` if you run into permission errors.
 
 Build from the source code - advanced method
 --------------------------------------------
 
-Obtain ``gcc-8`` from Homebrew:
+First, obtain ``gcc-7`` with homebrew (https://brew.sh/) if you want multi-threaded version. Clang is okay if multithreading is not required. Note: installation of ``gcc`` can take a while (~ 30 minutes).
 
 .. code-block:: bash
 
-  brew install gcc@8
+  brew install gcc@7
 
-Now clone the repository:
+Now, clone the repository:
 
 .. code-block:: bash
 
   git clone --recursive https://github.com/dmlc/xgboost
+  cd xgboost; cp make/config.mk ./config.mk
 
-Create the ``build/`` directory and invoke CMake. Make sure to add ``CC=gcc-8 CXX=g++-8`` so that Homebrew GCC is selected. After invoking CMake, you can build XGBoost with ``make``:
+Open ``config.mk`` and uncomment these two lines:
+
+.. code-block:: bash
+
+  export CC = gcc
+  export CXX = g++
+
+and replace these two lines as follows: (specify the GCC version)
+
+.. code-block:: bash
+
+  export CC = gcc-7
+  export CXX = g++-7
+
+Now, you may build XGBoost using the following command:
 
 .. code-block:: bash
 
-  mkdir build
-  cd build
-  CC=gcc-8 CXX=g++-8 cmake ..
   make -j4
 
 You may now continue to `Python Package Installation`_.
@@ -144,20 +152,6 @@ We recommend you use `Git for Windows <https://git-for-windows.github.io/>`_, as
 
 XGBoost support compilation with Microsoft Visual Studio and MinGW.
 
-Compile XGBoost with Microsoft Visual Studio
---------------------------------------------
-To build with Visual Studio, we will need CMake. Make sure to install a recent version of CMake. Then run the following from the root of the XGBoost directory:
-
-.. code-block:: bash
-
-  mkdir build
-  cd build
-  cmake .. -G"Visual Studio 14 2015 Win64"
-
-This specifies an out of source build using the Visual Studio 64 bit generator. (Change the ``-G`` option appropriately if you have a different version of Visual Studio installed.) Open the ``.sln`` file in the build directory and build with Visual Studio.
-
-After the build process successfully ends, you will find a ``xgboost.dll`` library file inside ``./lib/`` folder.
-
 Compile XGBoost using MinGW
 ---------------------------
 After installing `Git for Windows <https://git-for-windows.github.io/>`_, you should have a shortcut named ``Git Bash``. You should run all subsequent steps in ``Git Bash``.
@@ -177,9 +171,21 @@ To build with MinGW, type:
 
   cp make/mingw64.mk config.mk; make -j4
 
-See :ref:`mingw_python` for buildilng XGBoost for Python.
+Compile XGBoost with Microsoft Visual Studio
+--------------------------------------------
+To build with Visual Studio, we will need CMake. Make sure to install a recent version of CMake. Then run the following from the root of the XGBoost directory:
 
-.. _build_gpu_support:
+.. code-block:: bash
+
+  mkdir build
+  cd build
+  cmake .. -G"Visual Studio 12 2013 Win64"
+
+This specifies an out of source build using the MSVC 12 64 bit generator. Open the ``.sln`` file in the build directory and build with Visual Studio. To use the Python module you can copy ``xgboost.dll`` into ``python-package/xgboost``.
+
+After the build process successfully ends, you will find a ``xgboost.dll`` library file inside ``./lib/`` folder, copy this file to the the API package folder like ``python-package/xgboost`` if you are using Python API.
+
+Unofficial windows binaries and instructions on how to use them are hosted on `Guido Tapia's blog <http://www.picnet.com.au/blogs/guido/post/2016/09/22/xgboost-windows-x64-binaries-for-download/>`_.
 
 Building with GPU support
 =========================
@@ -187,27 +193,26 @@ XGBoost can be built with GPU support for both Linux and Windows using CMake. GP
 
 An up-to-date version of the CUDA toolkit is required.
 
-From the command line on Linux starting from the XGBoost directory:
+From the command line on Linux starting from the xgboost directory:
 
 .. code-block:: bash
 
   mkdir build
   cd build
   cmake .. -DUSE_CUDA=ON
-  make -j4
+  make -j
 
-.. note:: Enabling multi-GPU training
+.. note:: Windows requirements for GPU build
 
-  By default, multi-GPU training is disabled and only a single GPU will be used. To enable multi-GPU training, set the option ``USE_NCCL=ON``. Multi-GPU training depends on NCCL2, available at `this link <https://developer.nvidia.com/nccl>`_. Since NCCL2 is only available for Linux machines, **multi-GPU training is available only for Linux**.
+  Only Visual C++ 2015 or 2013 with CUDA v8.0 were fully tested. Either install Visual C++ 2015 Build Tools separately, or as a part of Visual Studio 2015. If you already have Visual Studio 2017, the Visual C++ 2015 Toolchain componenet has to be installed using the VS 2017 Installer. Likely, you would need to use the VS2015 x64 Native Tools command prompt to run the cmake commands given below. In some situations, however, things run just fine from MSYS2 bash command line.
 
-  .. code-block:: bash
+On Windows, see what options for generators you have for CMake, and choose one with ``[arch]`` replaced with Win64:
 
-    mkdir build
-    cd build
-    cmake .. -DUSE_CUDA=ON -DUSE_NCCL=ON -DNCCL_ROOT=/path/to/nccl2
-    make -j4
+.. code-block:: bash
 
-On Windows, run CMake as follows:
+  cmake -help
+
+Then run CMake as follows:
 
 .. code-block:: bash
 
@@ -215,15 +220,13 @@ On Windows, run CMake as follows:
   cd build
   cmake .. -G"Visual Studio 14 2015 Win64" -DUSE_CUDA=ON
 
-(Change the ``-G`` option appropriately if you have a different version of Visual Studio installed.)
-
 .. note:: Visual Studio 2017 Win64 Generator may not work
 
   Choosing the Visual Studio 2017 generator may cause compilation failure. When it happens, specify the 2015 compiler by adding the ``-T`` option:
 
   .. code-block:: bash
 
-    make .. -G"Visual Studio 15 2017 Win64" -T v140,cuda=8.0 -DUSE_CUDA=ON
+    make .. -G"Visual Studio 15 2017 Win64" -T v140,cuda=8.0 -DR_LIB=ON -DUSE_CUDA=ON
 
 To speed up compilation, the compute version specific to your GPU could be passed to cmake as, e.g., ``-DGPU_COMPUTE_VER=50``.
 The above cmake configuration run will create an ``xgboost.sln`` solution file in the build directory. Build this solution in release mode as a x64 build, either from Visual studio or from command line:
@@ -237,9 +240,7 @@ To speed up compilation, run multiple jobs in parallel by appending option ``--
 Customized Building
 ===================
 
-We recommend the use of CMake for most use cases. See the full range of building options in CMakeLists.txt.
-
-Alternatively, you may use Makefile. The Makefile uses a configuration file ``config.mk``, which lets you modify several compilation flags:
+The configuration file ``config.mk`` modifies several compilation flags:
 - Whether to enable support for various distributed filesystems such as HDFS and Amazon S3
 - Which compiler to use
 - And some more
@@ -249,7 +250,7 @@ To customize, first copy ``make/config.mk`` to the project root and then modify
 Python Package Installation
 ===========================
 
-The Python package is located at ``python-package/``.
+The python package is located at ``python-package/``.
 There are several ways to install the package:
 
 1. Install system-wide, which requires root permission:
@@ -259,7 +260,7 @@ There are several ways to install the package:
   cd python-package; sudo python setup.py install
 
 You will however need Python ``distutils`` module for this to
-work. It is often part of the core Python package or it can be installed using your
+work. It is often part of the core python package or it can be installed using your
 package manager, e.g. in Debian use
 
 .. code-block:: bash
@@ -270,10 +271,10 @@ package manager, e.g. in Debian use
 
   If you recompiled XGBoost, then you need to reinstall it again to make the new library take effect.
 
-2. Only set the environment variable ``PYTHONPATH`` to tell Python where to find
-   the library. For example, assume we cloned ``xgboost`` on the home directory
-   ``~``. then we can added the following line in ``~/.bashrc``.
-   This option is **recommended for developers** who change the code frequently. The changes will be immediately reflected once you pulled the code and rebuild the project (no need to call ``setup`` again).
+2. Only set the environment variable ``PYTHONPATH`` to tell python where to find
+   the library. For example, assume we cloned `xgboost` on the home directory
+   `~`. then we can added the following line in `~/.bashrc`.
+   This option is **recommended for developers** who change the code frequently. The changes will be immediately reflected once you pulled the code and rebuild the project (no need to call ``setup`` again)
 
 .. code-block:: bash
 
@@ -285,23 +286,12 @@ package manager, e.g. in Debian use
 
   cd python-package; python setup.py develop --user
 
-.. _mingw_python:
+4. If you are installing the latest XGBoost version which requires compilation, add MinGW to the system PATH:
 
-Building XGBoost library for Python for Windows with MinGW-w64 (Advanced)
--------------------------------------------------------------------------
+.. code-block:: bash
 
-Windows versions of Python are built with Microsoft Visual Studio. Usually Python binary modules are built with the same compiler the interpreter is built with. However, you may not be able to use Visual Studio, for following reasons:
-
-1. VS is proprietary and commercial software. Microsoft provides a freeware "Community" edition, but its licensing terms impose restrictions as to where and how it can be used.
-2. Visual Studio contains telemetry, as documented in `Microsoft Visual Studio Licensing Terms <https://visualstudio.microsoft.com/license-terms/mt736442/>`_. Running software with telemetry may be against the policy of your organization.
-
-So you may want to build XGBoost with GCC own your own risk. This presents some difficulties because MSVC uses Microsoft runtime and MinGW-w64 uses own runtime, and the runtimes have different incompatible memory allocators. But in fact this setup is usable if you know how to deal with it. Here is some experience.
-
-1. The Python interpreter will crash on exit if XGBoost was used. This is usually not a big issue.
-2. ``-O3`` is OK.
-3. ``-mtune=native`` is also OK.
-4. Don't use ``-march=native`` gcc flag. Using it causes the Python interpreter to crash if the DLL was actually used.
-5. You may need to provide the lib with the runtime libs. If ``mingw32/bin`` is not in ``PATH``, build a wheel (``python setup.py bdist_wheel``), open it with an archiver and put the needed dlls to the directory where ``xgboost.dll`` is situated. Then you can install the wheel with ``pip``.
+    import os
+    os.environ['PATH'] = os.environ['PATH'] + ';C:\\Program Files\\mingw-w64\\x86_64-5.3.0-posix-seh-rt_v4-rev0\\mingw64\\bin'
 
 R Package Installation
 ======================
@@ -315,13 +305,35 @@ You can install xgboost from CRAN just like any other R package:
 
   install.packages("xgboost")
 
-For OSX users, single-threaded version will be installed. So only one thread will be used for training. To enable use of multiple threads (and utilize capacity of multi-core CPUs), see the section :ref:`osx_multithread` to install XGBoost from source.
+Or you can install it from our weekly updated drat repo:
+
+.. code-block:: R
+
+  install.packages("drat", repos="https://cran.rstudio.com")
+  drat:::addRepo("dmlc")
+  install.packages("xgboost", repos="http://dmlc.ml/drat/", type = "source")
+
+For OSX users, single threaded version will be installed. To install multi-threaded version,
+first follow `Building on OSX`_ to get the OpenMP enabled compiler. Then
+
+- Set the ``Makevars`` file in highest piority for R.
+
+  The point is, there are three ``Makevars`` : ``~/.R/Makevars``, ``xgboost/R-package/src/Makevars``, and ``/usr/local/Cellar/r/3.2.0/R.framework/Resources/etc/Makeconf`` (the last one obtained by running ``file.path(R.home("etc"), "Makeconf")`` in R), and ``SHLIB_OPENMP_CXXFLAGS`` is not set by default!! After trying, it seems that the first one has highest piority (surprise!).
+
+  Then inside R, run
+
+.. code-block:: R
+
+  install.packages("drat", repos="https://cran.rstudio.com")
+  drat:::addRepo("dmlc")
+  install.packages("xgboost", repos="http://dmlc.ml/drat/", type = "source")
 
 Installing the development version
 ----------------------------------
 
 Make sure you have installed git and a recent C++ compiler supporting C++11 (e.g., g++-4.8 or higher).
-On Windows, Rtools must be installed, and its bin directory has to be added to ``PATH`` during the installation.
+On Windows, Rtools must be installed, and its bin directory has to be added to PATH during the installation.
+And see the previous subsection for an OSX tip.
 
 Due to the use of git-submodules, ``devtools::install_github`` can no longer be used to install the latest version of R package.
 Thus, one has to run git to check out the code first:
@@ -343,39 +355,10 @@ In this case, just start R as you would normally do and run the following:
   setwd('wherever/you/cloned/it/xgboost/R-package/')
   install.packages('.', repos = NULL, type="source")
 
-The package could also be built and installed with CMake (and Visual C++ 2015 on Windows) using instructions from :ref:`r_gpu_support`, but without GPU support (omit the ``-DUSE_CUDA=ON`` cmake parameter).
+The package could also be built and installed with cmake (and Visual C++ 2015 on Windows) using instructions from the next section, but without GPU support (omit the ``-DUSE_CUDA=ON`` cmake parameter).
 
 If all fails, try `Building the shared library`_ to see whether a problem is specific to R package or not.
 
-.. _osx_multithread:
-
-Installing R package on Mac OSX with multi-threading
-----------------------------------------------------
-
-First, obtain ``gcc-8`` with Homebrew (https://brew.sh/) to enable multi-threading (i.e. using multiple CPU threads for training). The default Apple Clang compiler does not support OpenMP, so using the default compiler would have disabled multi-threading.
-
-.. code-block:: bash
-
-  brew install gcc@8
-
-Now, clone the repository:
-
-.. code-block:: bash
-
-  git clone --recursive https://github.com/dmlc/xgboost
-
-Create the ``build/`` directory and invoke CMake with option ``R_LIB=ON``. Make sure to add ``CC=gcc-8 CXX=g++-8`` so that Homebrew GCC is selected. After invoking CMake, you can install the R package by running ``make`` and ``make install``:
-
-.. code-block:: bash
-
-  mkdir build
-  cd build
-  CC=gcc-7 CXX=g++-7 cmake .. -DR_LIB=ON
-  make -j4
-  make install
-
-.. _r_gpu_support:
-
 Installing R package with GPU support
 -------------------------------------
 
@@ -391,9 +374,9 @@ On Linux, starting from the XGBoost directory type:
   make install -j
 
 When default target is used, an R package shared library would be built in the ``build`` area.
-The ``install`` target, in addition, assembles the package files with this shared library under ``build/R-package`` and runs ``R CMD INSTALL``.
+The ``install`` target, in addition, assembles the package files with this shared library under ``build/R-package``, and runs ``R CMD INSTALL``.
 
-On Windows, CMake with Visual C++ Build Tools (or Visual Studio) has to be used to build an R package with GPU support. Rtools must also be installed (perhaps, some other MinGW distributions with ``gendef.exe`` and ``dlltool.exe`` would work, but that was not tested).
+On Windows, cmake with Visual C++ Build Tools (or Visual Studio) has to be used to build an R package with GPU support. Rtools must also be installed (perhaps, some other MinGW distributions with ``gendef.exe`` and ``dlltool.exe`` would work, but that was not tested).
 
 .. code-block:: bash
 
@@ -402,8 +385,8 @@ On Windows, CMake with Visual C++ Build Tools (or Visual Studio) has to be used
   cmake .. -G"Visual Studio 14 2015 Win64" -DUSE_CUDA=ON -DR_LIB=ON
   cmake --build . --target install --config Release
 
-When ``--target xgboost`` is used, an R package DLL would be built under ``build/Release``.
-The ``--target install``, in addition, assembles the package files with this dll under ``build/R-package`` and runs ``R CMD INSTALL``.
+When ``--target xgboost`` is used, an R package dll would be built under ``build/Release``.
+The ``--target install``, in addition, assembles the package files with this dll under ``build/R-package``, and runs ``R CMD INSTALL``.
 
 If cmake can't find your R during the configuration step, you might provide the location of its executable to cmake like this: ``-DLIBR_EXECUTABLE="C:/Program Files/R/R-3.4.1/bin/x64/R.exe"``.
 

doc/conf.py

@@ -41,7 +41,7 @@ sys.path.insert(0, curr_path)
 
 # -- mock out modules
 import mock
-MOCK_MODULES = ['scipy', 'scipy.sparse', 'sklearn', 'pandas']
+MOCK_MODULES = ['numpy', 'scipy', 'scipy.sparse', 'sklearn', 'matplotlib', 'pandas', 'graphviz']
 for mod_name in MOCK_MODULES:
   sys.modules[mod_name] = mock.Mock()
 
@@ -62,7 +62,6 @@ release = xgboost.__version__
 # Add any Sphinx extension module names here, as strings. They can be
 # extensions coming with Sphinx (named 'sphinx.ext.*') or your custom ones
 extensions = [
-    'matplotlib.sphinxext.plot_directive',
     'sphinx.ext.autodoc',
     'sphinx.ext.napoleon',
     'sphinx.ext.mathjax',
@@ -70,11 +69,6 @@ extensions = [
     'breathe'
 ]
 
-graphviz_output_format = 'png'
-plot_formats = [('svg', 300), ('png', 100), ('hires.png', 300)]
-plot_html_show_source_link = False
-plot_html_show_formats = False
-
 # Breathe extension variables
 breathe_projects = {"xgboost": "doxyxml/"}
 breathe_default_project = "xgboost"
@@ -156,7 +150,7 @@ extensions.append("guzzle_sphinx_theme")
 # Guzzle theme options (see theme.conf for more information)
 html_theme_options = {
     # Set the name of the project to appear in the sidebar
-    "project_nav_name": "XGBoost"
+    "project_nav_name": "XGBoost (0.72)"
 }
 
 html_sidebars = {

doc/contribute.rst

@@ -18,8 +18,6 @@ Everyone is more than welcome to contribute. It is a way to make the project bet
 
 * `Documents`_
 * `Testcases`_
-* `Sanitizers`_
-* `clang-tidy`_
 * `Examples`_
 * `Core Library`_
 * `Python Package`_
@@ -123,79 +121,6 @@ Testcases
 * All the testcases are in `tests <https://github.com/dmlc/xgboost/tree/master/tests>`_.
 * We use python nose for python test cases.
 
-**********
-Sanitizers
-**********
-
-By default, sanitizers are bundled in GCC and Clang/LLVM. One can enable
-sanitizers with GCC >= 4.8 or LLVM >= 3.1, But some distributions might package
-sanitizers separately.  Here is a list of supported sanitizers with
-corresponding library names:
-
-- Address sanitizer: libasan
-- Leak sanitizer:    liblsan
-- Thread sanitizer:  libtsan
-
-Memory sanitizer is exclusive to LLVM, hence not supported in XGBoost.
-
-How to build XGBoost with sanitizers
-====================================
-One can build XGBoost with sanitizer support by specifying -DUSE_SANITIZER=ON.
-By default, address sanitizer and leak sanitizer are used when you turn the
-USE_SANITIZER flag on.  You can always change the default by providing a
-semicolon separated list of sanitizers to ENABLED_SANITIZERS.  Note that thread
-sanitizer is not compatible with the other two sanitizers.
-
-  .. code-block:: bash
-
-    cmake -DUSE_SANITIZER=ON -DENABLED_SANITIZERS="address;leak" /path/to/xgboost
-
-By default, CMake will search regular system paths for sanitizers, you can also
-supply a specified SANITIZER_PATH.
-
-  .. code-block:: bash
-
-    cmake -DUSE_SANITIZER=ON -DENABLED_SANITIZERS="address;leak" \
-    -DSANITIZER_PATH=/path/to/sanitizers /path/to/xgboost
-
-How to use sanitizers with CUDA support
-=======================================
-Runing XGBoost on CUDA with address sanitizer (asan) will raise memory error.
-To use asan with CUDA correctly, you need to configure asan via ASAN_OPTIONS
-environment variable:
-
-  .. code-block:: bash
-
-    ASAN_OPTIONS=protect_shadow_gap=0 ../testxgboost
-
-For details, please consult `official documentation <https://github.com/google/sanitizers/wiki>`_ for sanitizers.
-
-**********
-clang-tidy
-**********
-To run clang-tidy on both C++ and CUDA source code,  run the following command
-from the top level source tree:
-
-  .. code-black:: bash
-    cd /path/to/xgboost/
-    python3 tests/ci_build/tidy.py --gtest-path=/path/to/google-test
-
-The script requires the full path of Google Test library via the ``--gtest-path`` argument.
-
-Also, the script accepts two optional integer arguments, namely ``--cpp`` and ``--cuda``.
-By default they are both set to 1.  If you want to exclude CUDA source from
-linting, use:
-
-  .. code-black:: bash
-    cd /path/to/xgboost/
-    python3 tests/ci_build/tidy.py --cuda=0
-
-Similarly, if you want to exclude C++ source from linting:
-
-  .. code-black:: bash
-    cd /path/to/xgboost/
-    python3 tests/ci_build/tidy.py --cpp=0
-
 ********
 Examples
 ********

doc/get_started.rst

@@ -42,7 +42,7 @@ R
   train <- agaricus.train
   test <- agaricus.test
   # fit model
-  bst <- xgboost(data = train$data, label = train$label, max.depth = 2, eta = 1, nrounds = 2,
+  bst <- xgboost(data = train$data, label = train$label, max.depth = 2, eta = 1, nround = 2,
                  nthread = 2, objective = "binary:logistic")
   # predict
   pred <- predict(bst, test$data)

doc/get_started/index.md

@@ -0,0 +1,79 @@
+# Get Started with XGBoost
+
+This is a quick start tutorial showing snippets for you to quickly try out xgboost
+on the demo dataset on a binary classification task.
+
+## Links to Helpful Other Resources
+- See [Installation Guide](../build.md) on how to install xgboost.
+- See [How to pages](../how_to/index.md) on various tips on using xgboost.
+- See [Tutorials](../tutorials/index.md) on tutorials on specific tasks.
+- See [Learning to use XGBoost by Examples](../../demo) for more code examples.
+
+## Python
+```python
+import xgboost as xgb
+# read in data
+dtrain = xgb.DMatrix('demo/data/agaricus.txt.train')
+dtest = xgb.DMatrix('demo/data/agaricus.txt.test')
+# specify parameters via map
+param = {'max_depth':2, 'eta':1, 'silent':1, 'objective':'binary:logistic' }
+num_round = 2
+bst = xgb.train(param, dtrain, num_round)
+# make prediction
+preds = bst.predict(dtest)
+```
+
+## R
+
+```r
+# load data
+data(agaricus.train, package='xgboost')
+data(agaricus.test, package='xgboost')
+train <- agaricus.train
+test <- agaricus.test
+# fit model
+bst <- xgboost(data = train$data, label = train$label, max.depth = 2, eta = 1, nround = 2,
+               nthread = 2, objective = "binary:logistic")
+# predict
+pred <- predict(bst, test$data)
+```
+
+## Julia
+```julia
+using XGBoost
+# read data
+train_X, train_Y = readlibsvm("demo/data/agaricus.txt.train", (6513, 126))
+test_X, test_Y = readlibsvm("demo/data/agaricus.txt.test", (1611, 126))
+# fit model
+num_round = 2
+bst = xgboost(train_X, num_round, label=train_Y, eta=1, max_depth=2)
+# predict
+pred = predict(bst, test_X)
+```
+
+## Scala
+```scala
+import ml.dmlc.xgboost4j.scala.DMatrix
+import ml.dmlc.xgboost4j.scala.XGBoost
+
+object XGBoostScalaExample {
+  def main(args: Array[String]) {
+    // read trainining data, available at xgboost/demo/data
+    val trainData =
+      new DMatrix("/path/to/agaricus.txt.train")
+    // define parameters
+    val paramMap = List(
+      "eta" -> 0.1,
+      "max_depth" -> 2,
+      "objective" -> "binary:logistic").toMap
+    // number of iterations
+    val round = 2
+    // train the model
+    val model = XGBoost.train(trainData, paramMap, round)
+    // run prediction
+    val predTrain = model.predict(trainData)
+    // save model to the file.
+    model.saveModel("/local/path/to/model")
+  }
+}
+```

doc/gpu/index.rst

@@ -14,11 +14,11 @@ To install GPU support, checkout the :doc:`/build`.
 *********************************************
 CUDA Accelerated Tree Construction Algorithms
 *********************************************
-Tree construction (training) and prediction can be accelerated with CUDA-capable GPUs.
+This plugin adds GPU accelerated tree construction and prediction algorithms to XGBoost.
 
 Usage
 =====
-Specify the ``tree_method`` parameter as one of the following algorithms.
+Specify the ``tree_method`` parameter as one of the following algorithms. 
 
 Algorithms
 ----------
@@ -31,49 +31,41 @@ Algorithms
 | gpu_hist     | Equivalent to the XGBoost fast histogram algorithm. Much faster and uses considerably less memory. NOTE: Will run very slowly on GPUs older than Pascal architecture. |
 +--------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------+
 
-Supported parameters
+Supported parameters 
 --------------------
 
-.. |tick| unicode:: U+2714
-.. |cross| unicode:: U+2718
+.. |tick| unicode:: U+2714 
+.. |cross| unicode:: U+2718 
 
-+--------------------------------+---------------+--------------+
-| parameter                      | ``gpu_exact`` | ``gpu_hist`` |
-+================================+===============+==============+
-| ``subsample``                  | |cross|       | |tick|       |
-+--------------------------------+---------------+--------------+
-| ``colsample_bytree``           | |cross|       | |tick|       |
-+--------------------------------+---------------+--------------+
-| ``colsample_bylevel``          | |cross|       | |tick|       |
-+--------------------------------+---------------+--------------+
-| ``max_bin``                    | |cross|       | |tick|       |
-+--------------------------------+---------------+--------------+
-| ``gpu_id``                     | |tick|        | |tick|       |
-+--------------------------------+---------------+--------------+
-| ``n_gpus``                     | |cross|       | |tick|       |
-+--------------------------------+---------------+--------------+
-| ``predictor``                  | |tick|        | |tick|       |
-+--------------------------------+---------------+--------------+
-| ``grow_policy``                | |cross|       | |tick|       |
-+--------------------------------+---------------+--------------+
-| ``monotone_constraints``       | |cross|       | |tick|       |
-+--------------------------------+---------------+--------------+
-| ``single_precision_histogram`` | |cross|       | |tick|       |
-+--------------------------------+---------------+--------------+
++--------------------------+---------------+--------------+
+| parameter                | ``gpu_exact`` | ``gpu_hist`` |
++==========================+===============+==============+
+| ``subsample``            | |cross|       | |tick|       |
++--------------------------+---------------+--------------+
+| ``colsample_bytree``     | |cross|       | |tick|       |
++--------------------------+---------------+--------------+
+| ``colsample_bylevel``    | |cross|       | |tick|       |
++--------------------------+---------------+--------------+
+| ``max_bin``              | |cross|       | |tick|       |
++--------------------------+---------------+--------------+
+| ``gpu_id``               | |tick|        | |tick|       |
++--------------------------+---------------+--------------+
+| ``n_gpus``               | |cross|       | |tick|       |
++--------------------------+---------------+--------------+
+| ``predictor``            | |tick|        | |tick|       |
++--------------------------+---------------+--------------+
+| ``grow_policy``          | |cross|       | |tick|       |
++--------------------------+---------------+--------------+
+| ``monotone_constraints`` | |cross|       | |tick|       |
++--------------------------+---------------+--------------+
 
 GPU accelerated prediction is enabled by default for the above mentioned ``tree_method`` parameters but can be switched to CPU prediction by setting ``predictor`` to ``cpu_predictor``. This could be useful if you want to conserve GPU memory. Likewise when using CPU algorithms, GPU accelerated prediction can be enabled by setting ``predictor`` to ``gpu_predictor``.
 
-The experimental parameter ``single_precision_histogram`` can be set to True to enable building histograms using single precision. This may improve speed, in particular on older architectures.
-
 The device ordinal can be selected using the ``gpu_id`` parameter, which defaults to 0.
 
-Multiple GPUs can be used with the ``gpu_hist`` tree method using the ``n_gpus`` parameter. which defaults to 1. If this is set to -1 all available GPUs will be used.  If ``gpu_id`` is specified as non-zero, the selected gpu devices will be from ``gpu_id`` to ``gpu_id+n_gpus``, please note that ``gpu_id+n_gpus`` must be less than or equal to the number of available GPUs on your system.  As with GPU vs. CPU, multi-GPU will not always be faster than a single GPU due to PCI bus bandwidth that can limit performance.
+Multiple GPUs can be used with the ``gpu_hist`` tree method using the ``n_gpus`` parameter. which defaults to 1. If this is set to -1 all available GPUs will be used.  If ``gpu_id`` is specified as non-zero, the gpu device order is ``mod(gpu_id + i) % n_visible_devices`` for ``i=0`` to ``n_gpus-1``.  As with GPU vs. CPU, multi-GPU will not always be faster than a single GPU due to PCI bus bandwidth that can limit performance.
 
-.. note:: Enabling multi-GPU training
-
-  Default installation may not enable multi-GPU training. To use multiple GPUs, make sure to read :ref:`build_gpu_support`.
-
-The GPU algorithms currently work with CLI, Python and R packages. See :doc:`/build` for details.
+This plugin currently works with the CLI, python and R - see :doc:`/build` for details.
 
 .. code-block:: python
   :caption: Python example
@@ -82,95 +74,6 @@ The GPU algorithms currently work with CLI, Python and R packages. See :doc:`/bu
   param['max_bin'] = 16
   param['tree_method'] = 'gpu_hist'
 
-Objective functions
-===================
-Most of the objective functions implemented in XGBoost can be run on GPU.  Following table shows current support status.
-
-.. |tick| unicode:: U+2714
-.. |cross| unicode:: U+2718
-
-+-----------------+-------------+
-| Objectives      | GPU support |
-+-----------------+-------------+
-| reg:linear      | |tick|      |
-+-----------------+-------------+
-| reg:logistic    | |tick|      |
-+-----------------+-------------+
-| binary:logistic | |tick|      |
-+-----------------+-------------+
-| binary:logitraw | |tick|      |
-+-----------------+-------------+
-| binary:hinge    | |tick|      |
-+-----------------+-------------+
-| count:poisson   | |tick|      |
-+-----------------+-------------+
-| reg:gamma       | |tick|      |
-+-----------------+-------------+
-| reg:tweedie     | |tick|      |
-+-----------------+-------------+
-| multi:softmax   | |tick|      |
-+-----------------+-------------+
-| multi:softprob  | |tick|      |
-+-----------------+-------------+
-| survival:cox    | |cross|     |
-+-----------------+-------------+
-| rank:pairwise   | |cross|     |
-+-----------------+-------------+
-| rank:ndcg       | |cross|     |
-+-----------------+-------------+
-| rank:map        | |cross|     |
-+-----------------+-------------+
-
-For multi-gpu support, objective functions also honor the ``n_gpus`` parameter,
-which, by default is set to 1.  To disable running objectives on GPU, just set
-``n_gpus`` to 0.
-
-Metric functions
-===================
-Following table shows current support status for evaluation metrics on the GPU.
-
-.. |tick| unicode:: U+2714
-.. |cross| unicode:: U+2718
-
-+-----------------+-------------+
-| Metric          | GPU Support |
-+=================+=============+
-| rmse            | |tick|      |
-+-----------------+-------------+
-| mae             | |tick|      |
-+-----------------+-------------+
-| logloss         | |tick|      |
-+-----------------+-------------+
-| error           | |tick|      |
-+-----------------+-------------+
-| merror          | |cross|     |
-+-----------------+-------------+
-| mlogloss        | |cross|     |
-+-----------------+-------------+
-| auc             | |cross|     |
-+-----------------+-------------+
-| aucpr           | |cross|     |
-+-----------------+-------------+
-| ndcg            | |cross|     |
-+-----------------+-------------+
-| map             | |cross|     |
-+-----------------+-------------+
-| poisson-nloglik | |tick|      |
-+-----------------+-------------+
-| gamma-nloglik   | |tick|      |
-+-----------------+-------------+
-| cox-nloglik     | |cross|     |
-+-----------------+-------------+
-| gamma-deviance  | |tick|      |
-+-----------------+-------------+
-| tweedie-nloglik | |tick|      |
-+-----------------+-------------+
-
-As for objective functions, metrics honor the ``n_gpus`` parameter,
-which, by default is set to 1.  To disable running metrics on GPU, just set
-``n_gpus`` to 0.
-
-
 Benchmarks
 ==========
 You can run benchmarks on synthetic data for binary classification:
@@ -202,16 +105,13 @@ References
 
 `Nvidia Parallel Forall: Gradient Boosting, Decision Trees and XGBoost with CUDA <https://devblogs.nvidia.com/parallelforall/gradient-boosting-decision-trees-xgboost-cuda/>`_
 
-Contributors
+Authors
 =======
-Many thanks to the following contributors (alphabetical order):
-* Andrey Adinets
-* Jiaming Yuan
-* Jonathan C. McKinney
-* Matthew Jones
-* Philip Cho
 * Rory Mitchell
+* Jonathan C. McKinney
 * Shankara Rao Thejaswi Nanditale
 * Vinay Deshpande
+* ... and the rest of the H2O.ai and NVIDIA team.
 
 Please report bugs to the user forum https://discuss.xgboost.ai/.
+

doc/how_to/index.md

@@ -0,0 +1,17 @@
+# XGBoost How To
+
+This page contains guidelines to use and develop XGBoost.
+
+## Installation
+- [How to Install XGBoost](../build.md)
+
+## Use XGBoost in Specific Ways
+- [Parameter tuning guide](param_tuning.md)
+- [Use out of core computation for large dataset](external_memory.md)
+- [Use XGBoost GPU algorithms](../gpu/index.md)
+
+## Develop and Hack XGBoost
+- [Contribute to XGBoost](contribute.md)
+
+## Frequently Ask Questions
+- [FAQ](../faq.md)

doc/input_format.md

@@ -0,0 +1,56 @@
+Text Input Format of DMatrix
+============================
+
+## Basic Input Format
+As we have mentioned, XGBoost takes LibSVM format. For training or predicting, XGBoost takes an instance file with the format as below:
+
+train.txt
+```
+1 101:1.2 102:0.03
+0 1:2.1 10001:300 10002:400
+0 0:1.3 1:0.3
+1 0:0.01 1:0.3
+0 0:0.2 1:0.3
+```
+Each line represent a single instance, and in the first line '1' is the instance label,'101' and '102' are feature indices, '1.2' and '0.03' are feature values. In the binary classification case, '1' is used to indicate positive samples, and '0' is used to indicate negative samples. We also support probability values in [0,1] as label, to indicate the probability of the instance being positive.
+
+Additional Information
+----------------------
+Note: these additional information are only applicable to single machine version of the package.
+
+### Group Input Format
+As XGBoost supports accomplishing [ranking task](../demo/rank), we support the group input format. In ranking task, instances are categorized into different groups in real world scenarios, for example, in the learning to rank web pages scenario, the web page instances are grouped by their queries. Except the instance file mentioned in the group input format, XGBoost need an file indicating the group information. For example, if the instance file is the "train.txt" shown above,
+and the group file is as below:
+
+train.txt.group
+```
+2
+3
+```
+This means that, the data set contains 5 instances, and the first two instances are in a group and the other three are in another group. The numbers in the group file are actually indicating the number of instances in each group in the instance file in order.
+While configuration, you do not have to indicate the path of the group file. If the instance file name is "xxx", XGBoost will check whether there is a file named "xxx.group" in the same directory and decides whether to read the data as group input format.
+
+### Instance Weight File
+XGBoost supports providing each instance an weight to differentiate the importance of instances. For example, if we provide an instance weight file for the "train.txt" file in the example as below:
+
+train.txt.weight
+```
+1
+0.5
+0.5
+1
+0.5
+```
+It means that XGBoost will emphasize more on the first and fourth instance， that is to say positive instances while training.
+The configuration is similar to configuring the group information. If the instance file name is "xxx", XGBoost will check whether there is a file named "xxx.weight" in the same directory and if there is, will use the weights while training models. Weights will be included into an "xxx.buffer" file that is created by XGBoost automatically. If you want to update the weights, you need to delete the "xxx.buffer" file prior to launching XGBoost.
+
+### Initial Margin file
+XGBoost supports providing each instance an initial margin prediction. For example, if we have a initial prediction using logistic regression for "train.txt" file, we can create the following file:
+
+train.txt.base_margin
+```
+-0.4
+1.0
+3.4
+```
+XGBoost will take these values as initial margin prediction and boost from that. An important note about base_margin is that it should be margin prediction before transformation, so if you are doing logistic loss, you will need to put in value before logistic transformation. If you are using XGBoost predictor, use pred_margin=1 to output margin values.

doc/jvm/index.rst

@@ -58,11 +58,10 @@ For sbt, please add the repository and dependency in build.sbt as following:
 
 If you want to use XGBoost4J-Spark, replace ``xgboost4j`` with ``xgboost4j-spark``.
 
-.. note:: XGBoost4J-Spark requires Apache Spark 2.3+
+.. note:: Spark 2.0 Required
 
-  XGBoost4J-Spark now requires **Apache Spark 2.3+**. Latest versions of XGBoost4J-Spark uses facilities of `org.apache.spark.ml.param.shared` extensively to provide for a tight integration with Spark MLLIB framework, and these facilities are not fully available on earlier versions of Spark.
+  After integrating with Dataframe/Dataset APIs of Spark 2.0, XGBoost4J-Spark only supports compile with Spark 2.x. You can build XGBoost4J-Spark as a component of XGBoost4J by running ``mvn package``, and you can specify the version of spark with ``mvn -Dspark.version=2.0.0 package``.   (To continue working with Spark 1.x, the users are supposed to update pom.xml by modifying the properties like ``spark.version``, ``scala.version``, and ``scala.binary.version``. Users also need to change the implementation by replacing ``SparkSession`` with ``SQLContext`` and the type of API parameters from ``Dataset[_]`` to ``Dataframe``)
 
-  Also, make sure to install Spark directly from `Apache website <https://spark.apache.org/>`_. **Upstream XGBoost is not guaranteed to work with third-party distributions of Spark, such as Cloudera Spark.** Consult appropriate third parties to obtain their distribution of XGBoost.
 
 Installation from maven repo
 ============================
@@ -149,7 +148,6 @@ Contents
   :maxdepth: 2
 
   java_intro
-  XGBoost4J-Spark Tutorial <xgboost4j_spark_tutorial>
   Code Examples <https://github.com/dmlc/xgboost/tree/master/jvm-packages/xgboost4j-example>
   XGBoost4J Java API <javadocs/index>
   XGBoost4J Scala API <scaladocs/xgboost4j/index>

doc/jvm/xgboost4j_spark_tutorial.rst

@@ -1,535 +0,0 @@
-#######################################
-XGBoost4J-Spark Tutorial (version 0.8+)
-#######################################
-
-**XGBoost4J-Spark** is a project aiming to seamlessly integrate XGBoost and Apache Spark by fitting XGBoost to Apache Spark's MLLIB framework. With the integration, user can not only uses the high-performant algorithm implementation of XGBoost, but also leverages the powerful  data processing engine of Spark for:
-
-* Feature Engineering: feature extraction, transformation, dimensionality reduction, and selection, etc.
-* Pipelines: constructing, evaluating, and tuning ML Pipelines
-* Persistence: persist and load machine learning models and even whole Pipelines
-
-This tutorial is to cover the end-to-end process to build a machine learning pipeline with XGBoost4J-Spark. We will discuss
-
-* Using Spark to preprocess data to fit to XGBoost/XGBoost4J-Spark's data interface
-* Training a XGBoost model with XGBoost4J-Spark
-* Serving XGBoost model (prediction) with Spark
-* Building a Machine Learning Pipeline with XGBoost4J-Spark
-* Running XGBoost4J-Spark in Production
-
-.. contents::
-  :backlinks: none
-  :local:
-
-********************************************
-Build an ML Application with XGBoost4J-Spark
-********************************************
-
-Refer to XGBoost4J-Spark Dependency
-===================================
-
-Before we go into the tour of how to use XGBoost4J-Spark, we would bring a brief introduction about how to build a machine learning application with XGBoost4J-Spark. The first thing you need to do is to refer to the dependency in Maven Central.
-
-You can add the following dependency in your ``pom.xml``.
-
-.. code-block:: xml
-
-  <dependency>
-    <groupId>ml.dmlc</groupId>
-    <artifactId>xgboost4j-spark</artifactId>
-    <version>latest_version_num</version>
-  </dependency>
-
-For the latest release version number, please check `here <https://github.com/dmlc/xgboost/releases>`_.
-
-We also publish some functionalities which would be included in the coming release in the form of snapshot version. To access these functionalities, you can add dependency to the snapshot artifacts. We publish snapshot version in github-based repo, so you can add the following repo in ``pom.xml``:
-
-.. code-block:: xml
-
-  <repository>
-    <id>XGBoost4J-Spark Snapshot Repo</id>
-    <name>XGBoost4J-Spark Snapshot Repo</name>
-    <url>https://raw.githubusercontent.com/CodingCat/xgboost/maven-repo/</url>
-  </repository>
-
-and then refer to the snapshot dependency by adding:
-
-.. code-block:: xml
-
-  <dependency>
-      <groupId>ml.dmlc</groupId>
-      <artifactId>xgboost4j-spark</artifactId>
-      <version>next_version_num-SNAPSHOT</version>
-  </dependency>
-
-.. note:: XGBoost4J-Spark requires Apache Spark 2.3+
-
-  XGBoost4J-Spark now requires **Apache Spark 2.3+**. Latest versions of XGBoost4J-Spark uses facilities of `org.apache.spark.ml.param.shared` extensively to provide for a tight integration with Spark MLLIB framework, and these facilities are not fully available on earlier versions of Spark.
-
-  Also, make sure to install Spark directly from `Apache website <https://spark.apache.org/>`_. **Upstream XGBoost is not guaranteed to work with third-party distributions of Spark, such as Cloudera Spark.** Consult appropriate third parties to obtain their distribution of XGBoost.
-
-Installation from maven repo
-
-.. note:: Use of Python in XGBoost4J-Spark
-
-  By default, we use the tracker in `dmlc-core <https://github.com/dmlc/dmlc-core/tree/master/tracker>`_ to drive the training with XGBoost4J-Spark. It requires Python 2.7+. We also have an experimental Scala version of tracker which can be enabled by passing the parameter ``tracker_conf`` as ``scala``.
-
-Data Preparation
-================
-
-As aforementioned, XGBoost4J-Spark seamlessly integrates Spark and XGBoost. The integration enables
-users to apply various types of transformation over the training/test datasets with the convenient
-and powerful data processing framework, Spark.
-
-In this section, we use `Iris <https://archive.ics.uci.edu/ml/datasets/iris>`_ dataset as an example to
-showcase how we use Spark to transform raw dataset and make it fit to the data interface of XGBoost.
-
-Iris dataset is shipped in CSV format. Each instance contains 4 features, "sepal length", "sepal width",
-"petal length" and "petal width". In addition, it contains the "class" columnm, which is essentially the label with three possible values: "Iris Setosa", "Iris Versicolour" and "Iris Virginica".
-
-Read Dataset with Spark's Built-In Reader
------------------------------------------
-
-The first thing in data transformation is to load the dataset as Spark's structured data abstraction, DataFrame.
-
-.. code-block:: scala
-
-  import org.apache.spark.sql.SparkSession
-  import org.apache.spark.sql.types.{DoubleType, StringType, StructField, StructType}
-
-  val spark = SparkSession.builder().getOrCreate()
-  val schema = new StructType(Array(
-    StructField("sepal length", DoubleType, true),
-    StructField("sepal width", DoubleType, true),
-    StructField("petal length", DoubleType, true),
-    StructField("petal width", DoubleType, true),
-    StructField("class", StringType, true)))
-  val rawInput = spark.read.schema(schema).csv("input_path")
-
-At the first line, we create a instance of `SparkSession <http://spark.apache.org/docs/latest/sql-programming-guide.html#starting-point-sparksession>`_ which is the entry of any Spark program working with DataFrame. The ``schema`` variable defines the schema of DataFrame wrapping Iris data. With this explicitly set schema, we can define the columns' name as well as their types; otherwise the column name would be the default ones derived by Spark, such as ``_col0``, etc. Finally, we can use Spark's built-in csv reader to load Iris csv file as a DataFrame named ``rawInput``.
-
-Spark also contains many built-in readers for other format. The latest version of Spark supports CSV, JSON, Parquet, and LIBSVM.
-
-Transform Raw Iris Dataset
---------------------------
-
-To make Iris dataset be recognizable to XGBoost, we need to
-
-1. Transform String-typed label, i.e. "class", to Double-typed label.
-2. Assemble the feature columns as a vector to fit to the data interface of Spark ML framework.
-
-To convert String-typed label to Double, we can use Spark's built-in feature transformer `StringIndexer <https://spark.apache.org/docs/2.3.1/api/scala/index.html#org.apache.spark.ml.feature.StringIndexer>`_.
-
-.. code-block:: scala
-
-  import org.apache.spark.ml.feature.StringIndexer
-  val stringIndexer = new StringIndexer().
-    setInputCol("class").
-    setOutputCol("classIndex").
-    fit(rawInput)
-  val labelTransformed = stringIndexer.transform(rawInput).drop("class")
-
-With a newly created StringIndexer instance:
-
-1. we set input column, i.e. the column containing String-typed label
-2. we set output column, i.e. the column to contain the Double-typed label.
-3. Then we ``fit`` StringIndex with our input DataFrame ``rawInput``, so that Spark internals can get information like total number of distinct values, etc.
-
-Now we have a StringIndexer which is ready to be applied to our input DataFrame. To execute the transformation logic of StringIndexer, we ``transform`` the input DataFrame ``rawInput`` and to keep a concise DataFrame,
-we drop the column "class" and only keeps the feature columns and the transformed Double-typed label column (in the last line of the above code snippet).
-
-The ``fit`` and ``transform`` are two key operations in MLLIB. Basically, ``fit`` produces a "transformer", e.g. StringIndexer, and each transformer applies ``transform`` method on DataFrame to add new column(s) containing transformed features/labels or prediction results, etc. To understand more about ``fit`` and ``transform``, You can find more details in `here <http://spark.apache.org/docs/latest/ml-pipeline.html#pipeline-components>`_.
-
-Similarly, we can use another transformer, `VectorAssembler <https://spark.apache.org/docs/2.3.1/api/scala/index.html#org.apache.spark.ml.feature.VectorAssembler>`_, to assemble feature columns "sepal length", "sepal width", "petal length" and "petal width" as a vector.
-
-.. code-block:: scala
-
-  import org.apache.spark.ml.feature.VectorAssembler
-  val vectorAssembler = new VectorAssembler().
-    setInputCols(Array("sepal length", "sepal width", "petal length", "petal width")).
-    setOutputCol("features")
-  val xgbInput = vectorAssembler.transform(labelTransformed).select("features", "classIndex")
-
-Now, we have a DataFrame containing only two columns, "features" which contains vector-represented
-"sepal length", "sepal width", "petal length" and "petal width" and "classIndex" which has Double-typed
-labels. A DataFrame like this (containing vector-represented features and numeric labels) can be fed to XGBoost4J-Spark's training engine directly.
-
-Training
-========
-
-XGBoost supports both regression and classification. While we use Iris dataset in this tutorial to show how we use XGBoost/XGBoost4J-Spark to resolve a multi-classes classification problem, the usage in Regression is very similar to classification.
-
-To train a XGBoost model for classification, we need to claim a XGBoostClassifier first:
-
-.. code-block:: scala
-
-  import ml.dmlc.xgboost4j.scala.spark.XGBoostClassifier
-  val xgbParam = Map("eta" -> 0.1f,
-        "max_depth" -> 2,
-        "objective" -> "multi:softprob",
-        "num_class" -> 3,
-        "num_round" -> 100,
-        "num_workers" -> 2)
-  val xgbClassifier = new XGBoostClassifier(xgbParam).
-        setFeaturesCol("features").
-        setLabelCol("classIndex")
-
-The available parameters for training a XGBoost model can be found in :doc:`here </parameter>`. In XGBoost4J-Spark, we support not only the default set of parameters but also the camel-case variant of these parameters to keep consistent with Spark's MLLIB parameters.
-
-Specifically, each parameter in :doc:`this page </parameter>` has its
-equivalent form in XGBoost4J-Spark with camel case. For example, to set ``max_depth`` for each tree, you can pass parameter just like what we did in the above code snippet (as ``max_depth`` wrapped in a Map), or you can do it through setters in XGBoostClassifer:
-
-.. code-block:: scala
-
-  val xgbClassifier = new XGBoostClassifier().
-    setFeaturesCol("features").
-    setLabelCol("classIndex")
-  xgbClassifier.setMaxDepth(2)
-
-After we set XGBoostClassifier parameters and feature/label column, we can build a transformer, XGBoostClassificationModel by fitting XGBoostClassifier with the input DataFrame. This ``fit`` operation is essentially the training process and the generated model can then be used in prediction.
-
-.. code-block:: scala
-
-  val xgbClassificationModel = xgbClassifier.fit(xgbInput)
-
-Early Stopping
-----------------
-
-Early stopping is a feature to prevent the unnecessary training iterations. By specifying ``num_early_stopping_rounds`` or directly call ``setNumEarlyStoppingRounds`` over a XGBoostClassifier or XGBoostRegressor, we can define number of rounds if the evaluation metric going away from the best iteration and early stop training iterations.
-
-In additional to ``num_early_stopping_rounds``, you also need to define ``maximize_evaluation_metrics`` or call ``setMaximizeEvaluationMetrics`` to specify whether you want to maximize or minimize the metrics in training.
-
-For example, we need to maximize the evaluation metrics (set ``maximize_evaluation_metrics`` with true), and set ``num_early_stopping_rounds`` with 5. The evaluation metric of 10th iteration is the maximum one until now. In the following iterations, if there is no evaluation metric greater than the 10th iteration's (best one), the traning would be early stopped at 15th iteration.  
-
-Training with Evaluation Sets
-----------------
-
-You can also monitor the performance of the model during training with multiple evaluation datasets. By specifying ``eval_sets`` or call ``setEvalSets`` over a XGBoostClassifier or XGBoostRegressor, you can pass in multiple evaluation datasets typed as a Map from String to DataFrame.
-
-Prediction
-==========
-
-XGBoost4j-Spark supports two ways for model serving: batch prediction and single instance prediction.
-
-Batch Prediction
-----------------
-
-When we get a model, either XGBoostClassificationModel or XGBoostRegressionModel, it takes a DataFrame, read the column containing feature vectors, predict for each feature vector, and output a new DataFrame with the following columns by default:
-
-* XGBoostClassificationModel will output margins (``rawPredictionCol``), probabilities(``probabilityCol``) and the eventual prediction labels (``predictionCol``) for each possible label.
-* XGBoostRegressionModel will output prediction label(``predictionCol``).
-
-Batch prediction expects the user to pass the testset in the form of a DataFrame. XGBoost4J-Spark starts a XGBoost worker for each partition of DataFrame for parallel prediction and generates prediction results for the whole DataFrame in a batch.
-
-.. code-block:: scala
-
-  val xgbClassificationModel = xgbClassifier.fit(xgbInput)
-  val results = xgbClassificationModel.transform(testSet)
-
-With the above code snippet, we get a result DataFrame, result containing margin, probability for each class and the prediction for each instance
-
-.. code-block:: none
-
-  +-----------------+----------+--------------------+--------------------+----------+
-  |         features|classIndex|       rawPrediction|         probability|prediction|
-  +-----------------+----------+--------------------+--------------------+----------+
-  |[5.1,3.5,1.4,0.2]|       0.0|[3.45569849014282...|[0.99579632282257...|       0.0|
-  |[4.9,3.0,1.4,0.2]|       0.0|[3.45569849014282...|[0.99618089199066...|       0.0|
-  |[4.7,3.2,1.3,0.2]|       0.0|[3.45569849014282...|[0.99643349647521...|       0.0|
-  |[4.6,3.1,1.5,0.2]|       0.0|[3.45569849014282...|[0.99636095762252...|       0.0|
-  |[5.0,3.6,1.4,0.2]|       0.0|[3.45569849014282...|[0.99579632282257...|       0.0|
-  |[5.4,3.9,1.7,0.4]|       0.0|[3.45569849014282...|[0.99428516626358...|       0.0|
-  |[4.6,3.4,1.4,0.3]|       0.0|[3.45569849014282...|[0.99643349647521...|       0.0|
-  |[5.0,3.4,1.5,0.2]|       0.0|[3.45569849014282...|[0.99579632282257...|       0.0|
-  |[4.4,2.9,1.4,0.2]|       0.0|[3.45569849014282...|[0.99618089199066...|       0.0|
-  |[4.9,3.1,1.5,0.1]|       0.0|[3.45569849014282...|[0.99636095762252...|       0.0|
-  |[5.4,3.7,1.5,0.2]|       0.0|[3.45569849014282...|[0.99428516626358...|       0.0|
-  |[4.8,3.4,1.6,0.2]|       0.0|[3.45569849014282...|[0.99643349647521...|       0.0|
-  |[4.8,3.0,1.4,0.1]|       0.0|[3.45569849014282...|[0.99618089199066...|       0.0|
-  |[4.3,3.0,1.1,0.1]|       0.0|[3.45569849014282...|[0.99618089199066...|       0.0|
-  |[5.8,4.0,1.2,0.2]|       0.0|[3.45569849014282...|[0.97809928655624...|       0.0|
-  |[5.7,4.4,1.5,0.4]|       0.0|[3.45569849014282...|[0.97809928655624...|       0.0|
-  |[5.4,3.9,1.3,0.4]|       0.0|[3.45569849014282...|[0.99428516626358...|       0.0|
-  |[5.1,3.5,1.4,0.3]|       0.0|[3.45569849014282...|[0.99579632282257...|       0.0|
-  |[5.7,3.8,1.7,0.3]|       0.0|[3.45569849014282...|[0.97809928655624...|       0.0|
-  |[5.1,3.8,1.5,0.3]|       0.0|[3.45569849014282...|[0.99579632282257...|       0.0|
-  +-----------------+----------+--------------------+--------------------+----------+
-
-Single instance prediction
---------------------------
-
-XGBoostClassificationModel or XGBoostRegressionModel support make prediction on single instance as well.
-It accepts a single Vector as feature, and output the prediction label.
-
-However, the overhead of single-instance prediction is high due to the internal overhead of XGBoost, use it carefully!
-
-.. code-block:: scala
-
-  val features = xgbInput.head().getAs[Vector]("features")
-  val result = xgbClassificationModel.predict(features)
-
-Model Persistence
-=================
-
-Model and pipeline persistence
-------------------------------
-
-A data scientist produces an ML model and hands it over to an engineering team for deployment in a production environment. Reversely, a trained model may be used by data scientists, for example as a baseline, across the process of data exploration. So it's important to support model persistence to make the models available across usage scenarios and programming languages.
-
-XGBoost4j-Spark supports saving and loading XGBoostClassifier/XGBoostClassificationModel and XGBoostRegressor/XGBoostRegressionModel. It also supports saving and loading a ML pipeline which includes these estimators and models.
-
-We can save the XGBoostClassificationModel to file system:
-
-.. code-block:: scala
-
-  val xgbClassificationModelPath = "/tmp/xgbClassificationModel"
-  xgbClassificationModel.write.overwrite().save(xgbClassificationModelPath)
-
-and then loading the model in another session:
-
-.. code-block:: scala
-
-  import ml.dmlc.xgboost4j.scala.spark.XGBoostClassificationModel
-
-  val xgbClassificationModel2 = XGBoostClassificationModel.load(xgbClassificationModelPath)
-  xgbClassificationModel2.transform(xgbInput)
-
-With regards to ML pipeline save and load, please refer the next section.
-
-Interact with Other Bindings of XGBoost
----------------------------------------
-After we train a model with XGBoost4j-Spark on massive dataset, sometimes we want to do model serving in single machine or integrate it with other single node libraries for further processing. XGBoost4j-Spark supports export model to local by:
-
-.. code-block:: scala
-
-  val nativeModelPath = "/tmp/nativeModel"
-  xgbClassificationModel.nativeBooster.saveModel(nativeModelPath)
-
-Then we can load this model with single node Python XGBoost:
-
-.. code-block:: python
-
-  import xgboost as xgb
-  bst = xgb.Booster({'nthread': 4})
-  bst.load_model(nativeModelPath)
-
-.. note:: Using HDFS and S3 for exporting the models with nativeBooster.saveModel()
-
-  When interacting with other language bindings, XGBoost also supports saving-models-to and loading-models-from file systems other than the local one. You can use HDFS and S3 by prefixing the path with ``hdfs://`` and ``s3://`` respectively. However, for this capability, you must do **one** of the following:
-
-  1. Build XGBoost4J-Spark with the steps described in `here <https://xgboost.readthedocs.io/en/latest/jvm/index.html#installation-from-source>`_, but turning `USE_HDFS <https://github.com/dmlc/xgboost/blob/e939192978a0c152ad7b49b744630e99d54cffa8/jvm-packages/create_jni.py#L18>`_ (or USE_S3, etc. in the same place) switch on. With this approach, you can reuse the above code example by replacing "nativeModelPath" with a HDFS path.
-
-     - However, if you build with USE_HDFS, etc. you have to ensure that the involved shared object file, e.g. libhdfs.so, is put in the LIBRARY_PATH of your cluster. To avoid the complicated cluster environment configuration, choose the other option.
-
-  2. Use bindings of HDFS, S3, etc. to pass model files around. Here are the steps (taking HDFS as an example):
-
-     - Create a new file with
-
-       .. code-block:: scala
-
-         val outputStream = fs.create("hdfs_path")
-
-       where "fs" is an instance of `org.apache.hadoop.fs.FileSystem <https://hadoop.apache.org/docs/stable/api/org/apache/hadoop/fs/FileSystem.html>`_ class in Hadoop.
-
-     - Pass the returned OutputStream in the first step to nativeBooster.saveModel():
-
-       .. code-block:: scala
-
-         xgbClassificationModel.nativeBooster.saveModel(outputStream)
-
-     - Download file in other languages from HDFS and load with the pre-built (without the requirement of libhdfs.so) version of XGBoost. (The function "download_from_hdfs" is a helper function to be implemented by the user)
-
-       .. code-block:: python
-
-         import xgboost as xgb
-         bst = xgb.Booster({'nthread': 4})
-         local_path = download_from_hdfs("hdfs_path")
-         bst.load_model(local_path)
-
-.. note:: Consistency issue between XGBoost4J-Spark and other bindings
-
-  There is a consistency issue between XGBoost4J-Spark and other language bindings of XGBoost.
-
-  When users use Spark to load training/test data in LIBSVM format with the following code snippet:
-
-  .. code-block:: scala
-
-    spark.read.format("libsvm").load("trainingset_libsvm")
-
-  Spark assumes that the dataset is using 1-based indexing (feature indices staring with 1). However, when you do prediction with other bindings of XGBoost (e.g. Python API of XGBoost), XGBoost assumes that the dataset is using 0-based indexing (feature indices starting with 0) by default. It creates a pitfall for the users who train model with Spark but predict with the dataset in the same format in other bindings of XGBoost. The solution is to transform the dataset to 0-based indexing before you predict with, for example, Python API, or you append ``?indexing_mode=1`` to your file path when loading with DMatirx. For example in Python:
-
-  .. code-block:: python
-
-    xgb.DMatrix('test.libsvm?indexing_mode=1')
-
-*******************************************
-Building a ML Pipeline with XGBoost4J-Spark
-*******************************************
-
-Basic ML Pipeline
-=================
-
-Spark ML pipeline can combine multiple algorithms or functions into a single pipeline.
-It covers from feature extraction, transformation, selection to model training and prediction.
-XGBoost4j-Spark makes it feasible to embed XGBoost into such a pipeline seamlessly.
-The following example shows how to build such a pipeline consisting of Spark MLlib feature transformer
-and XGBoostClassifier estimator.
-
-We still use `Iris <https://archive.ics.uci.edu/ml/datasets/iris>`_ dataset and the ``rawInput`` DataFrame.
-First we need to split the dataset into training and test dataset.
-
-.. code-block:: scala
-
-  val Array(training, test) = rawInput.randomSplit(Array(0.8, 0.2), 123)
-
-The we build the ML pipeline which includes 4 stages:
-
-* Assemble all features into a single vector column.
-* From string label to indexed double label.
-* Use XGBoostClassifier to train classification model.
-* Convert indexed double label back to original string label.
-
-We have shown the first three steps in the earlier sections, and the last step is finished with a new transformer `IndexToString <https://spark.apache.org/docs/2.3.1/api/scala/index.html#org.apache.spark.ml.feature.IndexToString>`_:
-
-.. code-block:: scala
-
-	val labelConverter = new IndexToString()
-        .setInputCol("prediction")
-        .setOutputCol("realLabel")
-        .setLabels(stringIndexer.labels)
-
-We need to organize these steps as a Pipeline in Spark ML framework and evaluate the whole pipeline to get a PipelineModel:
-
-.. code-block:: scala
-
-  import org.apache.spark.ml.feature._
-  import org.apache.spark.ml.Pipeline
-
-  val pipeline = new Pipeline()
-      .setStages(Array(assembler, stringIndexer, booster, labelConverter))
-  val model = pipeline.fit(training)
-
-After we get the PipelineModel, we can make prediction on the test dataset and evaluate the model accuracy.
-
-.. code-block:: scala
-
-  import org.apache.spark.ml.evaluation.MulticlassClassificationEvaluator
-
-  val prediction = model.transform(test)
-  val evaluator = new MulticlassClassificationEvaluator()
-  val accuracy = evaluator.evaluate(prediction)
-
-Pipeline with Hyper-parameter Tunning
-=====================================
-The most critical operation to maximize the power of XGBoost is to select the optimal parameters for the model. Tuning parameters manually is a tedious and labor-consuming process. With the latest version of XGBoost4J-Spark, we can utilize the Spark model selecting tool to automate this process.
-
-The following example shows the code snippet utilizing CrossValidation and MulticlassClassificationEvaluator
-to search the optimal combination of two XGBoost parameters, ``max_depth`` and ``eta``. (See :doc:`/parameter`.)
-The model producing the maximum accuracy defined by MulticlassClassificationEvaluator is selected and used to generate the prediction for the test set.
-
-.. code-block:: scala
-
-  import org.apache.spark.ml.tuning._
-  import org.apache.spark.ml.PipelineModel
-  import ml.dmlc.xgboost4j.scala.spark.XGBoostClassificationModel
-
-  val paramGrid = new ParamGridBuilder()
-      .addGrid(booster.maxDepth, Array(3, 8))
-      .addGrid(booster.eta, Array(0.2, 0.6))
-      .build()
-  val cv = new CrossValidator()
-      .setEstimator(pipeline)
-      .setEvaluator(evaluator)
-      .setEstimatorParamMaps(paramGrid)
-      .setNumFolds(3)
-
-  val cvModel = cv.fit(training)
-
-  val bestModel = cvModel.bestModel.asInstanceOf[PipelineModel].stages(2)
-      .asInstanceOf[XGBoostClassificationModel]
-  bestModel.extractParamMap()
-
-*********************************
-Run XGBoost4J-Spark in Production
-*********************************
-
-XGBoost4J-Spark is one of the most important steps to bring XGBoost to production environment easier. In this section, we introduce three key features to run XGBoost4J-Spark in production.
-
-Parallel/Distributed Training
-=============================
-The massive size of training dataset is one of the most significant characteristics in production environment. To ensure that training in XGBoost scales with the data size, XGBoost4J-Spark bridges the distributed/parallel processing framework of Spark and the parallel/distributed training mechanism of XGBoost.
-
-In XGBoost4J-Spark, each XGBoost worker is wrapped by a Spark task and the training dataset in Spark's memory space is fed to XGBoost workers in a transparent approach to the user.
-
-In the code snippet where we build XGBoostClassifier, we set parameter ``num_workers`` (or ``numWorkers``).
-This parameter controls how many parallel workers we want to have when training a XGBoostClassificationModel.
-
-.. note:: Regarding OpenMP optimization
-
-  By default, we allocate a core per each XGBoost worker. Therefore, the OpenMP optimization within each XGBoost worker does not take effect and the parallelization of training is achieved
-  by running multiple workers (i.e. Spark tasks) at the same time.
-
-  If you do want OpenMP optimization, you have to
-
-  1. set ``nthread`` to a value larger than 1 when creating XGBoostClassifier/XGBoostRegressor
-  2. set ``spark.task.cpus`` in Spark to the same value as ``nthread``
-
-Gang Scheduling
-===============
-XGBoost uses `AllReduce <http://mpitutorial.com/tutorials/mpi-reduce-and-allreduce/>`_.
-algorithm to synchronize the stats, e.g. histogram values, of each worker during training. Therefore XGBoost4J-Spark requires that all of ``nthread * numWorkers`` cores should be available before the training runs.
-
-In the production environment where many users share the same cluster, it's hard to guarantee that your XGBoost4J-Spark application can get all requested resources for every run. By default, the communication layer in XGBoost will block the whole application when it requires more resources to be available. This process usually brings unnecessary resource waste as it keeps the ready resources and try to claim more. Additionally, this usually happens silently and does not bring the attention of users.
-
-XGBoost4J-Spark allows the user to setup a timeout threshold for claiming resources from the cluster. If the application cannot get enough resources within this time period, the application would fail instead of wasting resources for hanging long. To enable this feature, you can set with XGBoostClassifier/XGBoostRegressor:
-
-.. code-block:: scala
-
-  xgbClassifier.setTimeoutRequestWorkers(60000L)
-
-or pass in ``timeout_request_workers`` in ``xgbParamMap`` when building XGBoostClassifier:
-
-.. code-block:: scala
-
-  val xgbParam = Map("eta" -> 0.1f,
-     "max_depth" -> 2,
-     "objective" -> "multi:softprob",
-     "num_class" -> 3,
-     "num_round" -> 100,
-     "num_workers" -> 2,
-     "timeout_request_workers" -> 60000L)
-  val xgbClassifier = new XGBoostClassifier(xgbParam).
-      setFeaturesCol("features").
-      setLabelCol("classIndex")
-
-If XGBoost4J-Spark cannot get enough resources for running two XGBoost workers, the application would fail. Users can have external mechanism to monitor the status of application and get notified for such case.
-
-Checkpoint During Training
-==========================
-
-Transient failures are also commonly seen in production environment. To simplify the design of XGBoost,
-we stop training if any of the distributed workers fail. However, if the training fails after having been through a long time, it would be a great waste of resources.
-
-We support creating checkpoint during training to facilitate more efficient recovery from failture. To enable this feature, you can set how many iterations we build each checkpoint with ``setCheckpointInterval`` and the location of checkpoints with ``setCheckpointPath``:
-
-.. code-block:: scala
-
-  xgbClassifier.setCheckpointInterval(2)
-  xgbClassifier.setCheckpointPath("/checkpoint_path")
-
-An equivalent way is to pass in parameters in XGBoostClassifier's constructor:
-
-.. code-block:: scala
-
-  val xgbParam = Map("eta" -> 0.1f,
-     "max_depth" -> 2,
-     "objective" -> "multi:softprob",
-     "num_class" -> 3,
-     "num_round" -> 100,
-     "num_workers" -> 2,
-     "checkpoint_path" -> "/checkpoints_path",
-     "checkpoint_interval" -> 2)
-  val xgbClassifier = new XGBoostClassifier(xgbParam).
-      setFeaturesCol("features").
-      setLabelCol("classIndex")
-
-If the training failed during these 100 rounds, the next run of training would start by reading the latest checkpoint file in ``/checkpoints_path`` and start from the iteration when the checkpoint was built until to next failure or the specified 100 rounds.

doc/parameter.rst

@@ -12,10 +12,6 @@ Before running XGBoost, we must set three types of parameters: general parameter
 
   In R-package, you can use ``.`` (dot) to replace underscore in the parameters, for example, you can use ``max.depth`` to indicate ``max_depth``. The underscore parameters are also valid in R.
 
-.. contents::
-  :backlinks: none
-  :local:
-
 ******************
 General Parameters
 ******************
@@ -23,25 +19,14 @@ General Parameters
 
   - Which booster to use. Can be ``gbtree``, ``gblinear`` or ``dart``; ``gbtree`` and ``dart`` use tree based models while ``gblinear`` uses linear functions.
 
-* ``silent`` [default=0] [Deprecated]
+* ``silent`` [default=0]
 
-  - Deprecated.  Please use ``verbosity`` instead.
-
-* ``verbosity`` [default=1]
-
-  - Verbosity of printing messages.  Valid values are 0 (silent),
-    1 (warning), 2 (info), 3 (debug).  Sometimes XGBoost tries to change
-    configurations based on heuristics, which is displayed as warning message.
-    If there's unexpected behaviour, please try to increase value of verbosity.
+  - 0 means printing running messages, 1 means silent mode
 
 * ``nthread`` [default to maximum number of threads available if not set]
 
   - Number of parallel threads used to run XGBoost
 
-* ``disable_default_eval_metric`` [default=0]
-
-  - Flag to disable default metric. Set to >0 to disable.
-
 * ``num_pbuffer`` [set automatically by XGBoost, no need to be set by user]
 
   - Size of prediction buffer, normally set to number of training instances. The buffers are used to save the prediction results of last boosting step.
@@ -64,8 +49,8 @@ Parameters for Tree Booster
 
 * ``max_depth`` [default=6]
 
-  - Maximum depth of a tree. Increasing this value will make the model more complex and more likely to overfit. 0 is only accepted in ``lossguided`` growing policy when tree_method is set as ``hist`` and it indicates no limit on depth. Beware that XGBoost aggressively consumes memory when training a deep tree.
-  - range: [0,∞] (0 is only accepted in ``lossguided`` growing policy when tree_method is set as ``hist``)
+  - Maximum depth of a tree. Increasing this value will make the model more complex and more likely to overfit. 0 indicates no limit. Note that limit is required when ``grow_policy`` is set of ``depthwise``.
+  - range: [0,∞]
 
 * ``min_child_weight`` [default=1]
 
@@ -82,22 +67,15 @@ Parameters for Tree Booster
   - Subsample ratio of the training instances. Setting it to 0.5 means that XGBoost would randomly sample half of the training data prior to growing trees. and this will prevent overfitting. Subsampling will occur once in every boosting iteration.
   - range: (0,1]
 
-* ``colsample_bytree``, ``colsample_bylevel``, ``colsample_bynode`` [default=1]
-  - This is a family of parameters for subsampling of columns.
-  - All ``colsample_by*`` parameters have a range of (0, 1], the default value of 1, and
-    specify the fraction of columns to be subsampled.
-  - ``colsample_bytree`` is the subsample ratio of columns when constructing each
-    tree. Subsampling occurs once for every tree constructed.
-  - ``colsample_bylevel`` is the subsample ratio of columns for each level. Subsampling
-    occurs once for every new depth level reached in a tree. Columns are subsampled from
-    the set of columns chosen for the current tree.
-  - ``colsample_bynode`` is the subsample ratio of columns for each node
-    (split). Subsampling occurs once every time a new split is evaluated. Columns are
-    subsampled from the set of columns chosen for the current level.
-  - ``colsample_by*`` parameters work cumulatively. For instance,
-    the combination ``{'colsample_bytree':0.5, 'colsample_bylevel':0.5,
-    'colsample_bynode':0.5}`` with 64 features will leave 4 features to choose from at
-    each split.
+* ``colsample_bytree`` [default=1]
+
+  - Subsample ratio of columns when constructing each tree. Subsampling will occur once in every boosting iteration.
+  - range: (0,1]
+
+* ``colsample_bylevel`` [default=1]
+
+  - Subsample ratio of columns for each split, in each level. Subsampling will occur each time a new split is made. This paramter has no effect when ``tree_method`` is set to ``hist``.
+  - range: (0,1]
 
 * ``lambda`` [default=1, alias: ``reg_lambda``]
 
@@ -110,7 +88,7 @@ Parameters for Tree Booster
 * ``tree_method`` string [default= ``auto``]
 
   - The tree construction algorithm used in XGBoost. See description in the `reference paper <http://arxiv.org/abs/1603.02754>`_.
-  - XGBoost supports ``hist`` and ``approx`` for distributed training and only support ``approx`` for external memory version.
+  - Distributed and external memory version only support ``tree_method=approx``.
   - Choices: ``auto``, ``exact``, ``approx``, ``hist``, ``gpu_exact``, ``gpu_hist``
 
     - ``auto``: Use heuristic to choose the fastest method.
@@ -152,7 +130,7 @@ Parameters for Tree Booster
     - ``refresh``: refreshes tree's statistics and/or leaf values based on the current data. Note that no random subsampling of data rows is performed.
     - ``prune``: prunes the splits where loss < min_split_loss (or gamma).
 
-  - In a distributed setting, the implicit updater sequence value would be adjusted to ``grow_histmaker,prune`` by default, and you can set ``tree_method`` as ``hist`` to use ``grow_histmaker``. 
+  - In a distributed setting, the implicit updater sequence value would be adjusted to ``grow_histmaker,prune``.
 
 * ``refresh_leaf`` [default=1]
 
@@ -170,7 +148,7 @@ Parameters for Tree Booster
 
   - Controls a way new nodes are added to the tree.
   - Currently supported only if ``tree_method`` is set to ``hist``.
-  - Choices: ``depthwise``, ``lossguide``
+  - Choices: ``depthwise``, ```lossguide``
 
     - ``depthwise``: split at nodes closest to the root.
     - ``lossguide``: split at nodes with highest loss change.
@@ -192,24 +170,8 @@ Parameters for Tree Booster
     - ``cpu_predictor``: Multicore CPU prediction algorithm.
     - ``gpu_predictor``: Prediction using GPU. Default when ``tree_method`` is ``gpu_exact`` or ``gpu_hist``.
 
-* ``num_parallel_tree``, [default=1]
-  - Number of parallel trees constructed during each iteration. This
-     option is used to support boosted random forest
-
 Additional parameters for Dart Booster (``booster=dart``)
 =========================================================
-
-.. note:: Using ``predict()`` with DART booster
-
-  If the booster object is DART type, ``predict()`` will perform dropouts, i.e. only
-  some of the trees will be evaluated. This will produce incorrect results if ``data`` is
-  not the training data. To obtain correct results on test sets, set ``ntree_limit`` to
-  a nonzero value, e.g.
-
-  .. code-block:: python
-
-    preds = bst.predict(dtest, ntree_limit=num_round)
-
 * ``sample_type`` [default= ``uniform``]
 
   - Type of sampling algorithm.
@@ -250,7 +212,7 @@ Additional parameters for Dart Booster (``booster=dart``)
   - range: [0.0, 1.0]
 
 Parameters for Linear Booster (``booster=gblinear``)
-====================================================
+==================================================
 * ``lambda`` [default=0, alias: ``reg_lambda``]
 
   - L2 regularization term on weights. Increasing this value will make model more conservative. Normalised to number of training examples.
@@ -263,22 +225,8 @@ Parameters for Linear Booster (``booster=gblinear``)
 
   - Choice of algorithm to fit linear model
 
-    - ``shotgun``: Parallel coordinate descent algorithm based on shotgun algorithm. Uses 'hogwild' parallelism and therefore produces a nondeterministic solution on each run.
-    - ``coord_descent``: Ordinary coordinate descent algorithm. Also multithreaded but still produces a deterministic solution.
-
-* ``feature_selector`` [default= ``cyclic``]
-
-  - Feature selection and ordering method
-
-    * ``cyclic``: Deterministic selection by cycling through features one at a time.
-    * ``shuffle``: Similar to ``cyclic`` but with random feature shuffling prior to each update.
-    * ``random``: A random (with replacement) coordinate selector.
-    * ``greedy``: Select coordinate with the greatest gradient magnitude.  It has ``O(num_feature^2)`` complexity. It is fully deterministic. It allows restricting the selection to ``top_k`` features per group with the largest magnitude of univariate weight change, by setting the ``top_k`` parameter. Doing so would reduce the complexity to ``O(num_feature*top_k)``.
-    * ``thrifty``: Thrifty, approximately-greedy feature selector. Prior to cyclic updates, reorders features in descending magnitude of their univariate weight changes. This operation is multithreaded and is a linear complexity approximation of the quadratic greedy selection. It allows restricting the selection to ``top_k`` features per group with the largest magnitude of univariate weight change, by setting the ``top_k`` parameter.
-
-* ``top_k`` [default=0]
-
-  - The number of top features to select in ``greedy`` and ``thrifty`` feature selector. The value of 0 means using all the features.
+    - ``shotgun``: Parallel coordinate descent algorithm based on shotgun algorithm. Uses 'hogwild' parallelism and therefore produces a nondeterministic solution on each run. 
+    - ``coord_descent``: Ordinary coordinate descent algorithm. Also multithreaded but still produces a deterministic solution. 
 
 Parameters for Tweedie Regression (``objective=reg:tweedie``)
 =============================================================
@@ -300,7 +248,9 @@ Specify the learning task and the corresponding learning objective. The objectiv
   - ``reg:logistic``: logistic regression
   - ``binary:logistic``: logistic regression for binary classification, output probability
   - ``binary:logitraw``: logistic regression for binary classification, output score before logistic transformation
-  - ``binary:hinge``: hinge loss for binary classification. This makes predictions of 0 or 1, rather than producing probabilities.
+  - ``gpu:reg:linear``, ``gpu:reg:logistic``, ``gpu:binary:logistic``, ``gpu:binary:logitraw``: versions
+    of the corresponding objective functions evaluated on the GPU; note that like the GPU histogram algorithm,
+    they can only be used when the entire training session uses the same dataset
   - ``count:poisson`` --poisson regression for count data, output mean of poisson distribution
 
     - ``max_delta_step`` is set to 0.7 by default in poisson regression (used to safeguard optimization)
@@ -309,9 +259,7 @@ Specify the learning task and the corresponding learning objective. The objectiv
     Note that predictions are returned on the hazard ratio scale (i.e., as HR = exp(marginal_prediction) in the proportional hazard function ``h(t) = h0(t) * HR``).
   - ``multi:softmax``: set XGBoost to do multiclass classification using the softmax objective, you also need to set num_class(number of classes)
   - ``multi:softprob``: same as softmax, but output a vector of ``ndata * nclass``, which can be further reshaped to ``ndata * nclass`` matrix. The result contains predicted probability of each data point belonging to each class.
-  - ``rank:pairwise``: Use LambdaMART to perform pairwise ranking where the pairwise loss is minimized
-  - ``rank:ndcg``: Use LambdaMART to perform list-wise ranking where `Normalized Discounted Cumulative Gain (NDCG) <http://en.wikipedia.org/wiki/NDCG>`_ is maximized
-  - ``rank:map``: Use LambdaMART to perform list-wise ranking where `Mean Average Precision (MAP) <http://en.wikipedia.org/wiki/Mean_average_precision#Mean_average_precision>`_ is maximized
+  - ``rank:pairwise``: set XGBoost to do ranking task by minimizing the pairwise loss
   - ``reg:gamma``: gamma regression with log-link. Output is a mean of gamma distribution. It might be useful, e.g., for modeling insurance claims severity, or for any outcome that might be `gamma-distributed <https://en.wikipedia.org/wiki/Gamma_distribution#Applications>`_.
   - ``reg:tweedie``: Tweedie regression with log-link. It might be useful, e.g., for modeling total loss in insurance, or for any outcome that might be `Tweedie-distributed <https://en.wikipedia.org/wiki/Tweedie_distribution#Applications>`_.
 
@@ -334,9 +282,8 @@ Specify the learning task and the corresponding learning objective. The objectiv
     - ``merror``: Multiclass classification error rate. It is calculated as ``#(wrong cases)/#(all cases)``.
     - ``mlogloss``: `Multiclass logloss <http://scikit-learn.org/stable/modules/generated/sklearn.metrics.log_loss.html>`_.
     - ``auc``: `Area under the curve <http://en.wikipedia.org/wiki/Receiver_operating_characteristic#Area_under_curve>`_
-    - ``aucpr``: `Area under the PR curve <https://en.wikipedia.org/wiki/Precision_and_recall>`_
     - ``ndcg``: `Normalized Discounted Cumulative Gain <http://en.wikipedia.org/wiki/NDCG>`_
-    - ``map``: `Mean Average Precision <http://en.wikipedia.org/wiki/Mean_average_precision#Mean_average_precision>`_
+    - ``map``: `Mean average precision <http://en.wikipedia.org/wiki/Mean_average_precision#Mean_average_precision>`_
     - ``ndcg@n``, ``map@n``: 'n' can be assigned as an integer to cut off the top positions in the lists for evaluation.
     - ``ndcg-``, ``map-``, ``ndcg@n-``, ``map@n-``: In XGBoost, NDCG and MAP will evaluate the score of a list without any positive samples as 1. By adding "-" in the evaluation metric XGBoost will evaluate these score as 0 to be consistent under some conditions.
     - ``poisson-nloglik``: negative log-likelihood for Poisson regression

doc/python/python_api.rst

@@ -39,10 +39,6 @@ Scikit-Learn API
     :members:
     :inherited-members:
     :show-inheritance:
-.. autoclass:: xgboost.XGBRanker
-    :members:
-    :inherited-members:
-    :show-inheritance:
 
 Plotting API
 ------------
@@ -53,15 +49,3 @@ Plotting API
 .. autofunction:: xgboost.plot_tree
 
 .. autofunction:: xgboost.to_graphviz
-
-.. _callback_api:
-
-Callback API
-------------
-.. autofunction:: xgboost.callback.print_evaluation
-
-.. autofunction:: xgboost.callback.record_evaluation
-
-.. autofunction:: xgboost.callback.reset_learning_rate
-
-.. autofunction:: xgboost.callback.early_stop

doc/python/python_intro.rst

@@ -25,8 +25,7 @@ The XGBoost python module is able to load data from:
 - LibSVM text format file
 - Comma-separated values (CSV) file
 - NumPy 2D array
-- SciPy 2D sparse array
-- Pandas data frame, and
+- SciPy 2D sparse array, and
 - XGBoost binary buffer file.
 
 (See :doc:`/tutorials/input_format` for detailed description of text input format.)
@@ -48,15 +47,9 @@ The data is stored in a :py:class:`DMatrix <xgboost.DMatrix>` object.
     dtrain = xgb.DMatrix('train.csv?format=csv&label_column=0')
     dtest = xgb.DMatrix('test.csv?format=csv&label_column=0')
 
-  .. note:: Categorical features not supported
-
-    Note that XGBoost does not support categorical features; if your data contains
-    categorical features, load it as a NumPy array first and then perform
-    `one-hot encoding <http://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.OneHotEncoder.html>`_.
-
-  .. note:: Use Pandas to load CSV files with headers
-
-    Currently, the DMLC data parser cannot parse CSV files with headers. Use Pandas (see below) to read CSV files with headers.
+  (Note that XGBoost does not support categorical features; if your data contains
+  categorical features, load it as a NumPy array first and then perform
+  `one-hot encoding <http://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.OneHotEncoder.html>`_.)
 
 * To load a NumPy array into :py:class:`DMatrix <xgboost.DMatrix>`:
 
@@ -73,14 +66,6 @@ The data is stored in a :py:class:`DMatrix <xgboost.DMatrix>` object.
     csr = scipy.sparse.csr_matrix((dat, (row, col)))
     dtrain = xgb.DMatrix(csr)
 
-* To load a Pandas data frame into :py:class:`DMatrix <xgboost.DMatrix>`:
-
-  .. code-block:: python
-
-    data = pandas.DataFrame(np.arange(12).reshape((4,3)), columns=['a', 'b', 'c'])
-    label = pandas.DataFrame(np.random.randint(2, size=4))
-    dtrain = xgb.DMatrix(data, label=label)
-
 * Saving :py:class:`DMatrix <xgboost.DMatrix>` into a XGBoost binary file will make loading faster:
 
   .. code-block:: python
@@ -101,10 +86,6 @@ The data is stored in a :py:class:`DMatrix <xgboost.DMatrix>` object.
     w = np.random.rand(5, 1)
     dtrain = xgb.DMatrix(data, label=label, missing=-999.0, weight=w)
 
-When performing ranking tasks, the number of weights should be equal
-to number of groups.
-
-
 Setting Parameters
 ------------------
 XGBoost can use either a list of pairs or a dictionary to set :doc:`parameters </parameter>`. For instance:
@@ -165,10 +146,6 @@ A saved model can be loaded as follows:
   bst = xgb.Booster({'nthread': 4})  # init model
   bst.load_model('model.bin')  # load data
 
-Methods including `update` and `boost` from `xgboost.Booster` are designed for
-internal usage only.  The wrapper function `xgboost.train` does some
-pre-configuration including setting up caches and some other parameters.
-
 Early Stopping
 --------------
 If you have a validation set, you can use early stopping to find the optimal number of boosting rounds.
@@ -223,3 +200,4 @@ When you use ``IPython``, you can use the :py:meth:`xgboost.to_graphviz` functio
 .. code-block:: python
 
   xgb.to_graphviz(bst, num_trees=2)
+

doc/requirements.txt

@@ -3,6 +3,3 @@ mock
 guzzle_sphinx_theme
 breathe
 sh>=1.12.14
-matplotlib>=2.1
-graphviz
-numpy

doc/tutorials/aws_yarn.rst

@@ -1,8 +1,216 @@
 ###############################
 Distributed XGBoost YARN on AWS
 ###############################
-[This page is under construction.]
+This is a step-by-step tutorial on how to setup and run distributed `XGBoost <https://github.com/dmlc/xgboost>`_
+on an AWS EC2 cluster. Distributed XGBoost runs on various platforms such as MPI, SGE and Hadoop YARN.
+In this tutorial, we use YARN as an example since this is a widely used solution for distributed computing.
 
-.. note:: XGBoost with Spark
+.. note:: XGBoost on Spark
 
-  If you are preprocessing training data with Spark, consider using :doc:`XGBoost4J-Spark </jvm/xgboost4j_spark_tutorial>`.
+  If you are preprocessing training data with Spark, you may want to look at `XGBoost4J-Spark <https://xgboost.ai/2016/10/26/a-full-integration-of-xgboost-and-spark.html>`_, which supports distributed training on Resilient Distributed Dataset (RDD).
+
+************
+Prerequisite
+************
+We need to get a `AWS key-pair <http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/ec2-key-pairs.html>`_
+to access the AWS services. Let us assume that we are using a key ``mykey`` and  the corresponding permission file ``mypem.pem``.
+
+We also need `AWS credentials <https://docs.aws.amazon.com/cli/latest/userguide/cli-chap-getting-started.html>`_,
+which includes an ``ACCESS_KEY_ID`` and a ``SECRET_ACCESS_KEY``.
+
+Finally, we will need a S3 bucket to host the data and the model, ``s3://mybucket/``
+
+***************************
+Setup a Hadoop YARN Cluster
+***************************
+This sections shows how to start a Hadoop YARN cluster from scratch.
+You can skip this step if you have already have one.
+We will be using `yarn-ec2 <https://github.com/tqchen/yarn-ec2>`_ to start the cluster.
+
+We can first clone the yarn-ec2 script by the following command.
+
+.. code-block:: bash
+
+  git clone https://github.com/tqchen/yarn-ec2
+
+To use the script, we must set the environment variables ``AWS_ACCESS_KEY_ID`` and
+``AWS_SECRET_ACCESS_KEY`` properly. This can be done by adding the following two lines in
+``~/.bashrc`` (replacing the strings with the correct ones)
+
+.. code-block:: bash
+
+  export AWS_ACCESS_KEY_ID=[your access ID]
+  export AWS_SECRET_ACCESS_KEY=[your secret access key]
+
+Now we can launch a master machine of the cluster from EC2:
+
+.. code-block:: bash
+
+  ./yarn-ec2 -k mykey -i mypem.pem launch xgboost
+
+Wait a few mininutes till the master machine gets up.
+
+After the master machine gets up, we can query the public DNS of the master machine using the following command.
+
+.. code-block:: bash
+
+  ./yarn-ec2 -k mykey -i mypem.pem get-master xgboost
+
+It will show the public DNS of the master machine like ``ec2-xx-xx-xx.us-west-2.compute.amazonaws.com``
+Now we can open the browser, and type (replace the DNS with the master DNS)
+
+.. code-block:: none
+
+  ec2-xx-xx-xx.us-west-2.compute.amazonaws.com:8088
+
+This will show the job tracker of the YARN cluster. Note that we may have to wait a few minutes before the master finishes bootstrapping and starts the
+job tracker.
+
+After the master machine gets up, we can freely add more slave machines to the cluster.
+The following command add m3.xlarge instances to the cluster.
+
+.. code-block:: bash
+
+  ./yarn-ec2 -k mykey -i mypem.pem -t m3.xlarge -s 2 addslave xgboost
+
+We can also choose to add two spot instances
+
+.. code-block:: bash
+
+  ./yarn-ec2 -k mykey -i mypem.pem -t m3.xlarge -s 2 addspot xgboost
+
+The slave machines will start up, bootstrap  and report to the master.
+You can check if the slave machines are connected by clicking on the Nodes link on the job tracker.
+Or simply type the following URL (replace DNS ith the master DNS)
+
+.. code-block:: none
+
+  ec2-xx-xx-xx.us-west-2.compute.amazonaws.com:8088/cluster/nodes
+
+One thing we should note is that not all the links in the job tracker work.
+This is due to that many of them use the private IP of AWS, which can only be accessed by EC2.
+We can use ssh proxy to access these packages.
+Now that we have set up a cluster with one master and two slaves, we are ready to run the experiment.
+
+*********************
+Build XGBoost with S3
+*********************
+We can log into the master machine by the following command.
+
+.. code-block:: bash
+
+  ./yarn-ec2 -k mykey -i mypem.pem login xgboost
+
+We will be using S3 to host the data and the result model, so the data won't get lost after the cluster shutdown.
+To do so, we will need to build XGBoost with S3 support. The only thing we need to do is to set ``USE_S3``
+variable to be true. This can be achieved by the following command.
+
+.. code-block:: bash
+
+  git clone --recursive https://github.com/dmlc/xgboost
+  cd xgboost
+  cp make/config.mk config.mk
+  echo "USE_S3=1" >> config.mk
+  make -j4
+
+Now we have built the XGBoost with S3 support. You can also enable HDFS support if you plan to store data on HDFS by turning on ``USE_HDFS`` option.
+XGBoost also relies on the environment variable to access S3, so you will need to add the following two lines to ``~/.bashrc`` (replacing the strings with the correct ones)
+on the master machine as well.
+
+.. code-block:: bash
+
+  export AWS_ACCESS_KEY_ID=AKIAIOSFODNN7EXAMPLE
+  export AWS_SECRET_ACCESS_KEY=wJalrXUtnFEMI/K7MDENG/bPxRfiCYEXAMPLEKEY
+  export BUCKET=mybucket
+
+*******************
+Host the Data on S3
+*******************
+In this example, we will copy the example dataset in XGBoost to the S3 bucket as input.
+In normal usecases, the dataset is usually created from existing distributed processing pipeline.
+We can use `s3cmd <http://s3tools.org/s3cmd>`_ to copy the data into mybucket (replace ``${BUCKET}`` with the real bucket name).
+
+.. code-block:: bash
+
+  cd xgboost
+  s3cmd put demo/data/agaricus.txt.train s3://${BUCKET}/xgb-demo/train/
+  s3cmd put demo/data/agaricus.txt.test s3://${BUCKET}/xgb-demo/test/
+
+***************
+Submit the Jobs
+***************
+Now everything is ready, we can submit the XGBoost distributed job to the YARN cluster.
+We will use the `dmlc-submit <https://github.com/dmlc/dmlc-core/tree/master/tracker>`_ script to submit the job.
+
+Now we can run the following script in the distributed training folder (replace ``${BUCKET}`` with the real bucket name)
+
+.. code-block:: bash
+
+  cd xgboost/demo/distributed-training
+  # Use dmlc-submit to submit the job.
+  ../../dmlc-core/tracker/dmlc-submit --cluster=yarn --num-workers=2 --worker-cores=2\
+      ../../xgboost mushroom.aws.conf nthread=2\
+      data=s3://${BUCKET}/xgb-demo/train\
+      eval[test]=s3://${BUCKET}/xgb-demo/test\
+      model_dir=s3://${BUCKET}/xgb-demo/model
+
+All the configurations such as ``data`` and ``model_dir`` can also be directly written into the configuration file.
+Note that we only specified the folder path to the file, instead of the file name.
+XGBoost will read in all the files in that folder as the training and evaluation data.
+
+In this command, we are using two workers, and each worker uses two running threads.
+XGBoost can benefit from using multiple cores in each worker.
+A common choice of working cores can range from 4 to 8.
+The trained model will be saved into the specified model folder. You can browse the model folder.
+
+.. code-block:: bash
+
+  s3cmd ls s3://${BUCKET}/xgb-demo/model/
+
+The following is an example output from distributed training.
+
+.. code-block:: none
+
+  16/02/26 05:41:59 INFO dmlc.Client: jobname=DMLC[nworker=2]:xgboost,username=ubuntu
+  16/02/26 05:41:59 INFO dmlc.Client: Submitting application application_1456461717456_0015
+  16/02/26 05:41:59 INFO impl.YarnClientImpl: Submitted application application_1456461717456_0015
+  2016-02-26 05:42:05,230 INFO @tracker All of 2 nodes getting started
+  2016-02-26 05:42:14,027 INFO [05:42:14] [0]  test-error:0.016139        train-error:0.014433
+  2016-02-26 05:42:14,186 INFO [05:42:14] [1]  test-error:0.000000        train-error:0.001228
+  2016-02-26 05:42:14,947 INFO @tracker All nodes finishes job
+  2016-02-26 05:42:14,948 INFO @tracker 9.71754479408 secs between node start and job finish
+  Application application_1456461717456_0015 finished with state FINISHED at 1456465335961
+
+*****************
+Analyze the Model
+*****************
+After the model is trained, we can analyse the learnt model and use it for future prediction tasks.
+XGBoost is a portable framework, meaning the models in all platforms are *exchangeable*.
+This means we can load the trained model in python/R/Julia and take benefit of data science pipelines
+in these languages to do model analysis and prediction.
+
+For example, you can use `this IPython notebook <https://github.com/dmlc/xgboost/tree/master/demo/distributed-training/plot_model.ipynb>`_
+to plot feature importance and visualize the learnt model.
+
+***************
+Troubleshooting
+***************
+
+If you encounter a problem, the best way might be to use the following command
+to get logs of stdout and stderr of the containers and check what causes the problem.
+
+.. code-block:: bash
+
+  yarn logs -applicationId yourAppId
+
+*****************
+Future Directions
+*****************
+You have learned to use distributed XGBoost on YARN in this tutorial.
+XGBoost is a portable and scalable framework for gradient boosting.
+You can check out more examples and resources in the `resources page <https://github.com/dmlc/xgboost/blob/master/demo/README.md>`_.
+
+The project goal is to make the best scalable machine learning solution available to all platforms.
+The API is designed to be able to portable, and the same code can also run on other platforms such as MPI and SGE.
+XGBoost is actively evolving and we are working on even more exciting features
+such as distributed XGBoost python/R package.