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

@@ -0,0 +1,22 @@
+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: '_'  }
+  - { key: readability-identifier-naming.EnumCase,          value: CamelCase  }
+  - { key: readability-identifier-naming.EnumConstant,          value: CamelCase  }
+  - { key: readability-identifier-naming.EnumConstantPrefix,          value: k  }
+  - { key: readability-identifier-naming.GlobalConstantCase,          value: CamelCase  }
+  - { key: readability-identifier-naming.GlobalConstantPrefix,          value: k  }
+  - { key: readability-identifier-naming.StaticConstantCase,          value: CamelCase  }
+  - { key: readability-identifier-naming.StaticConstantPrefix,          value: k  }
+  - { key: readability-identifier-naming.ConstexprVariableCase,          value: CamelCase  }
+  - { key: readability-identifier-naming.ConstexprVariablePrefix,          value: k  }
+  - { key: readability-identifier-naming.FunctionCase,          value: CamelCase  }
+  - { key: readability-identifier-naming.NamespaceCase,       value: lower_case }

.gitignore

@@ -15,17 +15,16 @@
 *.Rcheck
 *.rds
 *.tar.gz
-*txt*
 *conf
 *buffer
 *model
 *pyc
-*train
-*test
+*.train
+*.test
+*.tar
 *group
 *rar
 *vali
-*data
 *sdf
 Release
 *exe*
@@ -36,7 +35,6 @@ ipch
 *log
 Debug
 *suo
-*test*
 .Rhistory
 *.dll
 *i386
@@ -48,16 +46,12 @@ Debug
 *.cpage.col
 *.cpage
 *.Rproj
-./xgboost
 ./xgboost.mpi
 ./xgboost.mock
-rabit
 #.Rbuildignore
 R-package.Rproj
 *.cache*
-R-package/inst
-R-package/src
-#java
+# java
 java/xgboost4j/target
 java/xgboost4j/tmp
 java/xgboost4j-demo/target
@@ -65,9 +59,35 @@ java/xgboost4j-demo/data/
 java/xgboost4j-demo/tmp/
 java/xgboost4j-demo/model/
 nb-configuration*
-dmlc-core
 # Eclipse
 .project
 .cproject
 .pydevproject
 .settings/
+build
+config.mk
+/xgboost
+*.data
+build_plugin
+.idea
+recommonmark/
+tags
+*.iml
+*.class
+target
+*.swp
+
+# cpp tests and gcov generated files
+*.gcov
+*.gcda
+*.gcno
+build_tests
+/tests/cpp/xgboost_test
+
+.DS_Store
+lib/
+
+# spark
+metastore_db
+
+plugin/updater_gpu/test/cpp/data

.gitmodules

@@ -0,0 +1,12 @@
+[submodule "dmlc-core"]
+	path = dmlc-core
+	url = https://github.com/dmlc/dmlc-core
+[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

@@ -1,58 +1,91 @@
-sudo: true
+# disable sudo for container build.
+sudo: required
 
 # Enabling test on Linux and OS X
 os:
   - linux
   - osx
 
+osx_image: xcode8
+
+group: deprecated-2017Q4
+
 # Use Build Matrix to do lint and build seperately
 env:
   matrix:
-    - TASK=lint LINT_LANG=cpp
-    - TASK=lint LINT_LANG=python
-    - TASK=R-package CXX=g++
-    - TASK=python-package CXX=g++
-    - TASK=python-package3 CXX=g++
-    - TASK=java-package CXX=g++
-    - TASK=build CXX=g++
-    - TASK=build-with-dmlc CXX=g++
+    # code lint
+    - TASK=lint
+    # r package test
+    - TASK=r_test
+    # python package test
+    - TASK=python_test
+    - TASK=python_lightweight_test
+    # java package test
+    - TASK=java_test
+    # cmake test
+    - TASK=cmake_test
+    # c++ test
+    - TASK=cpp_test
 
-os:
-  - linux
-  - osx
+matrix:
+  exclude:
+    - os: osx
+      env: TASK=lint
+    - os: osx
+      env: TASK=cmake_test
+    - os: linux
+      env: TASK=r_test
+    - os: osx
+      env: TASK=python_lightweight_test
+    - os: osx
+      env: TASK=cpp_test
 
 # dependent apt packages
 addons:
   apt:
+    sources:
+      - llvm-toolchain-trusty-5.0
+      - ubuntu-toolchain-r-test
+      - george-edison55-precise-backports
     packages:
+      - clang
+      - clang-tidy-5.0
+      - cmake-data
       - doxygen
-      - libopenmpi-dev
       - wget
       - libcurl4-openssl-dev
       - unzip
-      - python-numpy
-      - python-scipy
+      - graphviz
+      - gcc-4.8
+      - g++-4.8
 
 before_install:
-  - scripts/travis_osx_install.sh
-  - git clone https://github.com/dmlc/dmlc-core
-  - export TRAVIS=dmlc-core/scripts/travis/
+  - source dmlc-core/scripts/travis/travis_setup_env.sh
   - export PYTHONPATH=${PYTHONPATH}:${PWD}/python-package
-  - source ${TRAVIS}/travis_setup_env.sh
+  - echo "MAVEN_OPTS='-Xmx2g -XX:MaxPermSize=1024m -XX:ReservedCodeCacheSize=512m -Dorg.slf4j.simpleLogger.defaultLogLevel=error'" > ~/.mavenrc
 
 install:
-  - pip install cpplint pylint --user `whoami`
+  - source tests/travis/setup.sh
 
+script:
+  - tests/travis/run_test.sh
 
-script: scripts/travis_script.sh
+cache:
+  directories:
+    - ${HOME}/.cache/usr
+    - ${HOME}/.cache/pip
 
+before_cache:
+  - dmlc-core/scripts/travis/travis_before_cache.sh
 
 after_failure:
-  - scripts/travis_after_failure.sh
+  - tests/travis/travis_after_failure.sh
 
+after_success:
+  - tree build
+  - bash <(curl -s https://codecov.io/bash) -a '-o src/ src/*.c'
 
 notifications:
   email:
     on_success: change
     on_failure: always
-

CHANGES.md

@@ -1,61 +0,0 @@
-Change Log
-==========
-
-xgboost-0.1
------------
-* Initial release
-
-xgboost-0.2x
-------------
-* Python module
-* Weighted samples instances
-* Initial version of pairwise rank
-
-xgboost-0.3
------------
-* Faster tree construction module
-  - Allows subsample columns during tree construction via ```bst:col_samplebytree=ratio```
-* Support for boosting from initial predictions
-* Experimental version of LambdaRank
-* Linear booster is now parallelized, using parallel coordinated descent.
-* Add [Code Guide](src/README.md) for customizing objective function and evaluation
-* Add R module
-
-xgboost-0.4
------------
-* Distributed version of xgboost that runs on YARN, scales to billions of examples
-* Direct save/load data and model from/to S3 and HDFS
-* Feature importance visualization in R module, by Michael Benesty
-* Predict leaf index
-* Poisson regression for counts data
-* Early stopping option in training
-* Native save load support in R and python
-  - xgboost models now can be saved using save/load in R
-  - xgboost python model is now pickable
-* sklearn wrapper is supported in python module
-* Experimental External memory version
-
-xgboost-0.47
-------------
-* Changes in R library
-  - fixed possible problem of poisson regression.
-  - switched from 0 to NA for missing values.
-  - exposed access to additional model parameters.
-* Changes in Python library
-  - throws exception instead of crash terminal when a parameter error happens.
-  - has importance plot and tree plot functions.
-  - accepts different learning rates for each boosting round.
-  - allows model training continuation from previously saved model.
-  - allows early stopping in CV.
-  - allows feval to return a list of tuples.
-  - allows eval_metric to handle additional format.
-  - improved compatibility in sklearn module.
-  - additional parameters added for sklearn wrapper.
-  - added pip installation functionality.
-  - supports more Pandas DataFrame dtypes. 
-  - added best_ntree_limit attribute, in addition to best_score and best_iteration.
-* Java api is ready for use
-* Added more test cases and continuous integration to make each build more robust.
-
-on going at master
-------------------

CITATION

@@ -0,0 +1,18 @@
+@inproceedings{Chen:2016:XST:2939672.2939785,
+ author = {Chen, Tianqi and Guestrin, Carlos},
+ title = {{XGBoost}: A Scalable Tree Boosting System},
+ booktitle = {Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining},
+ series = {KDD '16},
+ year = {2016},
+ isbn = {978-1-4503-4232-2},
+ location = {San Francisco, California, USA},
+ pages = {785--794},
+ numpages = {10},
+ url = {http://doi.acm.org/10.1145/2939672.2939785},
+ doi = {10.1145/2939672.2939785},
+ acmid = {2939785},
+ publisher = {ACM},
+ address = {New York, NY, USA},
+ keywords = {large-scale machine learning},
+}
+

CMakeLists.txt

@@ -0,0 +1,247 @@
+cmake_minimum_required (VERSION 3.2)
+project(xgboost)
+include(cmake/Utils.cmake)
+list(APPEND CMAKE_MODULE_PATH "${PROJECT_SOURCE_DIR}/cmake/modules")
+find_package(OpenMP)
+
+set_default_configuration_release()
+msvc_use_static_runtime()
+
+# Options
+option(USE_CUDA  "Build with GPU acceleration") 
+option(USE_AVX  "Build with AVX instructions. May not produce identical results due to approximate math." OFF) 
+option(USE_NCCL "Build using NCCL for multi-GPU. Also requires USE_CUDA") 
+option(JVM_BINDINGS "Build JVM bindings" OFF)
+option(GOOGLE_TEST "Build google tests" OFF)
+option(R_LIB "Build shared library for R package" OFF)
+set(GPU_COMPUTE_VER "" CACHE STRING
+  "Space separated list of compute versions to be built against, e.g. '35 61'")
+
+# Deprecation warning
+if(PLUGIN_UPDATER_GPU)
+  set(USE_CUDA ON)
+  message(WARNING "The option 'PLUGIN_UPDATER_GPU' is deprecated. Set 'USE_CUDA' instead.")
+endif()
+
+# Compiler flags
+set(CMAKE_CXX_STANDARD 11)
+set(CMAKE_CXX_STANDARD_REQUIRED ON)
+if(OpenMP_CXX_FOUND OR OPENMP_FOUND)
+  set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} ${OpenMP_CXX_FLAGS}")
+endif()
+set(CMAKE_POSITION_INDEPENDENT_CODE ON)
+if(MSVC)
+  # Multithreaded compilation
+  set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} /MP")
+else()
+  # Correct error for GCC 5 and cuda
+  set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -D_MWAITXINTRIN_H_INCLUDED -D_FORCE_INLINES")
+  # Performance
+  set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -funroll-loops")
+endif()
+
+# 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()
+
+
+# compiled code customizations for R package
+if(R_LIB)
+  add_definitions(
+    -DXGBOOST_STRICT_R_MODE=1
+    -DXGBOOST_CUSTOMIZE_GLOBAL_PRNG=1
+    -DDMLC_LOG_BEFORE_THROW=0
+    -DDMLC_DISABLE_STDIN=1
+    -DDMLC_LOG_CUSTOMIZE=1
+    -DRABIT_CUSTOMIZE_MSG_
+    -DRABIT_STRICT_CXX98_
+  )
+endif()
+
+include_directories (
+    ${PROJECT_SOURCE_DIR}/include
+    ${PROJECT_SOURCE_DIR}/dmlc-core/include
+    ${PROJECT_SOURCE_DIR}/rabit/include
+)
+
+file(GLOB_RECURSE SOURCES 
+    src/*.cc
+    src/*.h
+    include/*.h
+)
+
+# Only add main function for executable target
+list(REMOVE_ITEM SOURCES ${PROJECT_SOURCE_DIR}/src/cli_main.cc)
+
+file(GLOB_RECURSE CUDA_SOURCES
+    src/*.cu
+    src/*.cuh
+)
+
+# rabit
+# TODO: Create rabit cmakelists.txt
+set(RABIT_SOURCES
+    rabit/src/allreduce_base.cc
+    rabit/src/allreduce_robust.cc
+    rabit/src/engine.cc
+    rabit/src/c_api.cc
+)
+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})
+else()
+  add_library(rabit STATIC ${RABIT_SOURCES})
+endif()
+
+
+# dmlc-core
+add_subdirectory(dmlc-core)
+set(LINK_LIBRARIES dmlc rabit)
+
+
+if(USE_CUDA)
+  find_package(CUDA 8.0 REQUIRED)
+  cmake_minimum_required(VERSION 3.5)
+
+  add_definitions(-DXGBOOST_USE_CUDA)
+  
+  include_directories(cub)
+
+  if(USE_NCCL)
+    include_directories(nccl/src)
+    add_definitions(-DXGBOOST_USE_NCCL)
+  endif()
+
+  set(GENCODE_FLAGS "")
+  format_gencode_flags("${GPU_COMPUTE_VER}" GENCODE_FLAGS)
+  message("cuda architecture flags: ${GENCODE_FLAGS}")
+
+  set(CUDA_NVCC_FLAGS "${CUDA_NVCC_FLAGS};--expt-extended-lambda;--expt-relaxed-constexpr;${GENCODE_FLAGS};-lineinfo;")
+  if(NOT MSVC)
+    set(CUDA_NVCC_FLAGS "${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)
+    target_link_libraries(gpuxgboost nccl)
+  endif()
+  list(APPEND LINK_LIBRARIES gpuxgboost) 
+endif()
+
+
+# flags and sources for R-package
+if(R_LIB)
+  file(GLOB_RECURSE R_SOURCES
+    R-package/src/*.h
+    R-package/src/*.c
+    R-package/src/*.cc
+  )
+  list(APPEND SOURCES ${R_SOURCES})
+endif()
+
+add_library(objxgboost OBJECT ${SOURCES})
+
+
+# building shared library for R package
+if(R_LIB)
+  find_package(LibR REQUIRED)
+
+  list(APPEND LINK_LIBRARIES "${LIBR_CORE_LIBRARY}")
+  MESSAGE(STATUS "LIBR_CORE_LIBRARY " ${LIBR_CORE_LIBRARY})
+
+  include_directories(
+    "${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 "")
+
+  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
+else()
+  # Executable
+  add_executable(runxgboost $<TARGET_OBJECTS:objxgboost> src/cli_main.cc)
+  set_target_properties(runxgboost PROPERTIES
+    OUTPUT_NAME xgboost
+  )
+  set_output_directory(runxgboost ${PROJECT_SOURCE_DIR})
+  target_link_libraries(runxgboost ${LINK_LIBRARIES})
+
+  # Shared library
+  add_library(xgboost SHARED $<TARGET_OBJECTS:objxgboost>)
+  target_link_libraries(xgboost ${LINK_LIBRARIES})
+  set_output_directory(xgboost ${PROJECT_SOURCE_DIR}/lib)
+  if(MINGW)
+    # remove the 'lib' prefix to conform to windows convention for shared library names
+    set_target_properties(xgboost PROPERTIES PREFIX "")
+  endif()
+
+  #Ensure these two targets do not build simultaneously, as they produce outputs with conflicting names
+  add_dependencies(xgboost runxgboost)
+endif()
+
+
+# JVM
+if(JVM_BINDINGS)
+    find_package(JNI QUIET REQUIRED)
+
+    include_directories(${JNI_INCLUDE_DIRS} jvm-packages/xgboost4j/src/native)
+
+    add_library(xgboost4j SHARED
+        $<TARGET_OBJECTS:objxgboost>
+        jvm-packages/xgboost4j/src/native/xgboost4j.cpp)
+    set_output_directory(xgboost4j ${PROJECT_SOURCE_DIR}/lib)
+    target_link_libraries(xgboost4j
+        ${LINK_LIBRARIES}
+        ${JAVA_JVM_LIBRARY})
+endif()
+
+
+# Test
+if(GOOGLE_TEST)
+  find_package(GTest REQUIRED)
+  enable_testing()
+
+  file(GLOB_RECURSE TEST_SOURCES "tests/cpp/*.cc")
+  auto_source_group("${TEST_SOURCES}")
+  include_directories(${GTEST_INCLUDE_DIR})
+
+  if(USE_CUDA)
+    file(GLOB_RECURSE CUDA_TEST_SOURCES "tests/cpp/*.cu")
+    cuda_compile(CUDA_TEST_OBJS ${CUDA_TEST_SOURCES})
+  else()
+    set(CUDA_TEST_OBJS "")
+  endif()
+
+  add_executable(testxgboost ${TEST_SOURCES} ${CUDA_TEST_OBJS} $<TARGET_OBJECTS:objxgboost>)
+  set_output_directory(testxgboost ${PROJECT_SOURCE_DIR})
+  target_link_libraries(testxgboost ${GTEST_LIBRARIES} ${LINK_LIBRARIES})
+
+  add_test(TestXGBoost testxgboost)
+endif()
+
+
+# Group sources
+auto_source_group("${SOURCES}")

CONTRIBUTORS.md

@@ -2,28 +2,34 @@ Contributors of DMLC/XGBoost
 ============================
 XGBoost has been developed and used by a group of active community. Everyone is more than welcomed to is a great way to make the project better and more accessible to more users.
 
-Comitters
----------
+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 PhD working on large-scale machine learning, he is the creator of the project.
-* [Tong He](https://github.com/hetong007), Simon Fraser University
-  - Tong is a master student working on data mining, he is the maintainer of xgboost R package.
+* [Tong He](https://github.com/hetong007), Amazon AI
+  - 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).
 * [Michael Benesty](https://github.com/pommedeterresautee)
   - 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. 
+  - 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 Comitter
------------------
-XGBoost is a opensource project and we are actively looking for new comitters who are willing to help maintaining and lead the project.
+Become a Committer
+------------------
+XGBoost is a opensource project and we are actively looking for new committers who are willing to help maintaining and lead the project.
 Committers comes from contributors who:
 * Made substantial contribution to the project.
 * Willing to spent time on maintaining and lead the project.
 
-New committers will be proposed by current comitter memembers, with support from more than two of current comitters.
+New committers will be proposed by current committer members, with support from more than two of current committers.
 
 List of Contributors
 --------------------
@@ -37,19 +43,19 @@ List of Contributors
   - 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)
-* [Vadim Khotilovich](https://github.com/khotilov)
 * [Yangqing Men](https://github.com/yanqingmen)
   - 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
 * [Jamie Hall](https://github.com/nerdcha)
-  - Jamie is the initial creator of xgboost sklearn modue.
+  - 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.
 * [Hongliang Liu](https://github.com/phunterlau)
-  - Hongliang is the maintainer of xgboost python PyPI package for pip installation.
+* [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.
 * [Huayi Zhang](https://github.com/irachex)
@@ -59,3 +65,11 @@ List of Contributors
 * [Sam Thomson](https://github.com/sammthomson)
 * [ganesh-krishnan](https://github.com/ganesh-krishnan)
 * [Damien Carol](https://github.com/damiencarol)
+* [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)

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

@@ -0,0 +1,88 @@
+#!/usr/bin/groovy
+// -*- mode: groovy -*-
+// Jenkins pipeline
+// See documents at https://jenkins.io/doc/book/pipeline/jenkinsfile/
+
+import groovy.transform.Field
+
+/* Unrestricted tasks: tasks that do NOT generate artifacts */
+
+// Command to run command inside a docker container
+def dockerRun = 'tests/ci_build/ci_build.sh'
+// Utility functions
+@Field
+def utils
+
+def buildMatrix = [
+    [ "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" ],
+]
+
+pipeline {
+    // Each stage specify its own agent
+    agent none
+
+    // Setup common job properties
+    options {
+        ansiColor('xterm')
+        timestamps()
+        timeout(time: 120, unit: 'MINUTES')
+        buildDiscarder(logRotator(numToKeepStr: '10'))
+    }
+
+    // Build stages
+    stages {
+        stage('Jenkins: Get sources') {
+            agent {
+                label 'unrestricted'
+            }
+            steps {
+                script {
+                    utils = load('tests/ci_build/jenkins_tools.Groovy')
+                    utils.checkoutSrcs()
+                }
+                stash name: 'srcs', excludes: '.git/'
+                milestone label: 'Sources ready', ordinal: 1
+            }
+        }
+        stage('Jenkins: Build & Test') {
+            steps {
+                script {
+                    parallel (buildMatrix.findAll{it['enabled']}.collectEntries{ c ->
+                        def buildName = utils.getBuildName(c)
+                        utils.buildFactory(buildName, c, false, this.&buildPlatformCmake)
+                    })
+                }
+            }
+        }
+    }
+}
+
+/**
+ * Build platform and test it via cmake.
+ */
+def buildPlatformCmake(buildName, conf, nodeReq, dockerTarget) {
+    def opts = utils.cmakeOptions(conf)
+    // Destination dir for artifacts
+    def distDir = "dist/${buildName}"
+    def dockerArgs = ""
+    if(conf["withGpu"]){
+        dockerArgs = "--build-arg CUDA_VERSION=" + conf["cudaVersion"]
+    }
+    // Build node - this is returned result
+    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} ${dockerArgs} tests/ci_build/test_${dockerTarget}.sh
+        """
+    }
+}

Jenkinsfile-restricted

@@ -0,0 +1,121 @@
+#!/usr/bin/groovy
+// -*- mode: groovy -*-
+// Jenkins pipeline
+// See documents at https://jenkins.io/doc/book/pipeline/jenkinsfile/
+
+import groovy.transform.Field
+
+/* Restricted tasks: tasks generating artifacts, such as binary wheels and
+                     documentation */
+
+// Command to run command inside a docker container
+def dockerRun = 'tests/ci_build/ci_build.sh'
+// Utility functions
+@Field
+def utils
+
+def buildMatrix = [
+    [ "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": "8.0" ],
+    [ "enabled": true,  "os" : "linux", "withGpu": true, "withNccl": false, "withOmp": true, "pythonVersion": "2.7", "cudaVersion": "8.0" ],
+]
+
+pipeline {
+    // Each stage specify its own agent
+    agent none
+
+    // Setup common job properties
+    options {
+        ansiColor('xterm')
+        timestamps()
+        timeout(time: 120, unit: 'MINUTES')
+        buildDiscarder(logRotator(numToKeepStr: '10'))
+    }
+
+    // Build stages
+    stages {
+        stage('Jenkins: Get sources') {
+            agent {
+                label 'restricted'
+            }
+            steps {
+                script {
+                    utils = load('tests/ci_build/jenkins_tools.Groovy')
+                    utils.checkoutSrcs()
+                }
+                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 {
+                    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"
+                        }
+                    }
+                }
+            }
+        }
+
+        stage('Jenkins: Build artifacts') {
+            steps {
+                script {
+                    parallel (buildMatrix.findAll{it['enabled']}.collectEntries{ c ->
+                        def buildName = utils.getBuildName(c)
+                        utils.buildFactory(buildName, c, true, this.&buildPlatformCmake)
+                    })
+                }
+            }
+        }
+    }
+}
+
+/**
+ * Build platform and test it via cmake.
+ */
+def buildPlatformCmake(buildName, conf, nodeReq, dockerTarget) {
+    def opts = utils.cmakeOptions(conf)
+    // Destination dir for artifacts
+    def distDir = "dist/${buildName}"
+    def dockerArgs = ""
+    if(conf["withGpu"]){
+        dockerArgs = "--build-arg CUDA_VERSION=" + conf["cudaVersion"]
+    }
+    // Build node - this is returned result
+    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

@@ -1,9 +1,9 @@
-Copyright (c) 2014 by Contributors 
+Copyright (c) 2016 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

Makefile

@@ -1,18 +1,103 @@
-export CC  = $(if $(shell which gcc-5 2>/dev/null),gcc-5,gcc)
-export CXX = $(if $(shell which g++-5 2>/dev/null),g++-5,g++)
-
-export MPICXX = mpicxx
-export LDFLAGS= -pthread -lm
-export CFLAGS = -Wall -O3 -msse2  -Wno-unknown-pragmas -funroll-loops
-# java include path
-export JAVAINCFLAGS = -I${JAVA_HOME}/include -I./java
-
-ifeq ($(OS), Windows_NT)
-	export CXX = g++ -m64
-	export CC = gcc -m64
+ifndef config
+ifneq ("$(wildcard ./config.mk)","")
+	config = config.mk
+else
+	config = make/config.mk
+endif
 endif
 
-UNAME= $(shell uname)
+ifndef DMLC_CORE
+	DMLC_CORE = dmlc-core
+endif
+
+ifndef RABIT
+	RABIT = rabit
+endif
+
+ROOTDIR = $(CURDIR)
+
+# workarounds for some buggy old make & msys2 versions seen in windows
+ifeq (NA, $(shell test ! -d "$(ROOTDIR)" && echo NA ))
+        $(warning Attempting to fix non-existing ROOTDIR [$(ROOTDIR)])
+        ROOTDIR := $(shell pwd)
+        $(warning New ROOTDIR [$(ROOTDIR)] $(shell test -d "$(ROOTDIR)" && echo " is OK" ))
+endif
+MAKE_OK := $(shell "$(MAKE)" -v 2> /dev/null)
+ifndef MAKE_OK
+        $(warning Attempting to recover non-functional MAKE [$(MAKE)])
+        MAKE := $(shell which make 2> /dev/null)
+        MAKE_OK := $(shell "$(MAKE)" -v 2> /dev/null)
+endif
+$(warning MAKE [$(MAKE)] - $(if $(MAKE_OK),checked OK,PROBLEM))
+
+ifeq ($(OS), Windows_NT)
+	UNAME="Windows"
+else
+	UNAME=$(shell uname)
+endif
+
+include $(config)
+ifeq ($(USE_OPENMP), 0)
+	export NO_OPENMP = 1
+endif
+include $(DMLC_CORE)/make/dmlc.mk
+
+# include the plugins
+ifdef XGB_PLUGINS
+include $(XGB_PLUGINS)
+endif
+
+# set compiler defaults for OSX versus *nix
+# let people override either
+OS := $(shell uname)
+ifeq ($(OS), Darwin)
+ifndef CC
+export CC = $(if $(shell which clang), clang, gcc)
+endif
+ifndef CXX
+export CXX = $(if $(shell which clang++), clang++, g++)
+endif
+else
+# linux defaults
+ifndef CC
+export CC = gcc
+endif
+ifndef CXX
+export CXX = g++
+endif
+endif
+
+export LDFLAGS= -pthread -lm $(ADD_LDFLAGS) $(DMLC_LDFLAGS) $(PLUGIN_LDFLAGS)
+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${JAVA_HOME}/include -I./java
+
+ifeq ($(TEST_COVER), 1)
+	CFLAGS += -g -O0 -fprofile-arcs -ftest-coverage
+else
+	CFLAGS += -O3 -funroll-loops
+ifeq ($(USE_SSE), 1)
+	CFLAGS += -msse2
+endif
+endif
+
+ifndef LINT_LANG
+	LINT_LANG= "all"
+endif
+
+ifeq ($(UNAME), Windows)
+	XGBOOST_DYLIB = lib/xgboost.dll
+	JAVAINCFLAGS += -I${JAVA_HOME}/include/win32
+else
+ifeq ($(UNAME), Darwin)
+	XGBOOST_DYLIB = lib/libxgboost.dylib
+	CFLAGS += -fPIC
+else
+	XGBOOST_DYLIB = lib/libxgboost.so
+	CFLAGS += -fPIC
+endif
+endif
 
 ifeq ($(UNAME), Linux)
 	LDFLAGS += -lrt
@@ -23,192 +108,169 @@ ifeq ($(UNAME), Darwin)
 	JAVAINCFLAGS += -I${JAVA_HOME}/include/darwin
 endif
 
-ifeq ($(no_omp),1)
-	CFLAGS += -DDISABLE_OPENMP
+OPENMP_FLAGS =
+ifeq ($(USE_OPENMP), 1)
+	OPENMP_FLAGS = -fopenmp
 else
-	#CFLAGS += -fopenmp
-	ifeq ($(omp_mac_static),1)
-		#CFLAGS += -fopenmp -Bstatic
-		CFLAGS += -static-libgcc -static-libstdc++ -L. -fopenmp
-		#LDFLAGS += -Wl,--whole-archive -lpthread -Wl --no-whole-archive
-	else
-		CFLAGS += -fopenmp
-	endif
+	OPENMP_FLAGS = -DDISABLE_OPENMP
 endif
-
-
-# by default use c++11
-ifeq ($(cxx11),1)
-	CFLAGS += -std=c++11
-endif
-
-# handling dmlc
-ifdef dmlc
-	ifndef config
-		ifneq ("$(wildcard $(dmlc)/config.mk)","")
-			config = $(dmlc)/config.mk
-		else
-			config = $(dmlc)/make/config.mk
-		endif
-	endif
-	include $(config)
-	include $(dmlc)/make/dmlc.mk
-	LDFLAGS+= $(DMLC_LDFLAGS)
-	LIBDMLC=$(dmlc)/libdmlc.a
-else
-	LIBDMLC=dmlc_simple.o
-endif
-
-ifndef WITH_FPIC
-	WITH_FPIC = 1
-endif
-ifeq ($(WITH_FPIC), 1)
-	CFLAGS += -fPIC
-endif
-
-
-ifeq ($(OS), Windows_NT)
-	LIBRABIT = subtree/rabit/lib/librabit_empty.a
-	SLIB = wrapper/xgboost_wrapper.dll
-else
-	LIBRABIT = subtree/rabit/lib/librabit.a
-	SLIB = wrapper/libxgboostwrapper.so
-endif
-
-# java lib
-JLIB = java/libxgboost4j.so
+CFLAGS += $(OPENMP_FLAGS)
 
 # specify tensor path
-BIN = xgboost
-MOCKBIN = xgboost.mock
-OBJ = updater.o gbm.o io.o main.o dmlc_simple.o
-MPIBIN =
-ifeq ($(WITH_FPIC), 1)
-	TARGET = $(BIN) $(OBJ) $(SLIB)
-else
-	TARGET = $(BIN)
+.PHONY: clean all lint clean_all doxygen rcpplint pypack Rpack Rbuild Rcheck java pylint
+
+all: lib/libxgboost.a $(XGBOOST_DYLIB) xgboost
+
+$(DMLC_CORE)/libdmlc.a: $(wildcard $(DMLC_CORE)/src/*.cc $(DMLC_CORE)/src/*/*.cc)
+	+ cd $(DMLC_CORE); "$(MAKE)" libdmlc.a config=$(ROOTDIR)/$(config); cd $(ROOTDIR)
+
+$(RABIT)/lib/$(LIB_RABIT): $(wildcard $(RABIT)/src/*.cc)
+	+ cd $(RABIT); "$(MAKE)" lib/$(LIB_RABIT) USE_SSE=$(USE_SSE); cd $(ROOTDIR)
+
+jvm: jvm-packages/lib/libxgboost4j.so
+
+SRC = $(wildcard src/*.cc src/*/*.cc)
+ALL_OBJ = $(patsubst src/%.cc, build/%.o, $(SRC)) $(PLUGIN_OBJS)
+AMALGA_OBJ = amalgamation/xgboost-all0.o
+LIB_DEP = $(DMLC_CORE)/libdmlc.a $(RABIT)/lib/$(LIB_RABIT)
+ALL_DEP = $(filter-out build/cli_main.o, $(ALL_OBJ)) $(LIB_DEP)
+CLI_OBJ = build/cli_main.o
+include tests/cpp/xgboost_test.mk
+
+build/%.o: src/%.cc
+	@mkdir -p $(@D)
+	$(CXX) $(CFLAGS) -MM -MT build/$*.o $< >build/$*.d
+	$(CXX) -c $(CFLAGS) $< -o $@
+
+build_plugin/%.o: plugin/%.cc
+	@mkdir -p $(@D)
+	$(CXX) $(CFLAGS) -MM -MT build_plugin/$*.o $< >build_plugin/$*.d
+	$(CXX) -c $(CFLAGS) $< -o $@
+
+# The should be equivalent to $(ALL_OBJ)  except for build/cli_main.o
+amalgamation/xgboost-all0.o: amalgamation/xgboost-all0.cc
+	$(CXX) -c $(CFLAGS) $< -o $@
+
+# Equivalent to lib/libxgboost_all.so
+lib/libxgboost_all.so: $(AMALGA_OBJ) $(LIB_DEP)
+	@mkdir -p $(@D)
+	$(CXX) $(CFLAGS) -shared -o $@ $(filter %.o %.a, $^) $(LDFLAGS)
+
+lib/libxgboost.a: $(ALL_DEP)
+	@mkdir -p $(@D)
+	ar crv $@ $(filter %.o, $?)
+
+lib/xgboost.dll lib/libxgboost.so lib/libxgboost.dylib: $(ALL_DEP)
+	@mkdir -p $(@D)
+	$(CXX) $(CFLAGS) -shared -o $@ $(filter %.o %a,  $^) $(LDFLAGS)
+
+jvm-packages/lib/libxgboost4j.so: jvm-packages/xgboost4j/src/native/xgboost4j.cpp $(ALL_DEP)
+	@mkdir -p $(@D)
+	$(CXX) $(CFLAGS) $(JAVAINCFLAGS) -shared -o $@ $(filter %.cpp %.o %.a, $^) $(LDFLAGS)
+
+
+xgboost: $(CLI_OBJ) $(ALL_DEP)
+	$(CXX) $(CFLAGS) -o $@  $(filter %.o %.a, $^)  $(LDFLAGS)
+
+rcpplint:
+	python2 dmlc-core/scripts/lint.py xgboost ${LINT_LANG} R-package/src
+
+lint: rcpplint
+	python2 dmlc-core/scripts/lint.py xgboost ${LINT_LANG} include src plugin python-package
+
+pylint:
+	flake8 --ignore E501 python-package
+	flake8 --ignore E501 tests/python
+
+test: $(ALL_TEST)
+	$(ALL_TEST)
+
+check: test
+	./tests/cpp/xgboost_test
+
+ifeq ($(TEST_COVER), 1)
+cover: check
+	@- $(foreach COV_OBJ, $(COVER_OBJ), \
+		gcov -pbcul -o $(shell dirname $(COV_OBJ)) $(COV_OBJ) > gcov.log || cat gcov.log; \
+	)
 endif
 
-ifndef LINT_LANG
-	LINT_LANG= "all"
-endif
+clean:
+	$(RM) -rf build build_plugin lib bin *~ */*~ */*/*~ */*/*/*~ */*.o */*/*.o */*/*/*.o #xgboost
+	$(RM) -rf build_tests *.gcov tests/cpp/xgboost_test
+	if [ -d "R-package/src" ]; then \
+		cd R-package/src; \
+		$(RM) -rf rabit src include dmlc-core amalgamation *.so *.dll; \
+		cd $(ROOTDIR); \
+	fi
 
-.PHONY: clean all mpi python Rpack lint
+clean_all: clean
+	cd $(DMLC_CORE); "$(MAKE)" clean; cd $(ROOTDIR)
+	cd $(RABIT); "$(MAKE)" clean; cd $(ROOTDIR)
 
-all: $(TARGET)
-mpi: $(MPIBIN)
+doxygen:
+	doxygen doc/Doxyfile
 
-python: wrapper/libxgboostwrapper.so
-# now the wrapper takes in two files. io and wrapper part
-updater.o: src/tree/updater.cpp  src/tree/*.hpp src/*.h src/tree/*.h src/utils/*.h
-dmlc_simple.o: src/io/dmlc_simple.cpp src/utils/*.h
-gbm.o: src/gbm/gbm.cpp src/gbm/*.hpp src/gbm/*.h
-io.o: src/io/io.cpp src/io/*.hpp src/utils/*.h src/learner/dmatrix.h src/*.h
-main.o: src/xgboost_main.cpp src/utils/*.h src/*.h src/learner/*.hpp src/learner/*.h
-xgboost:  updater.o gbm.o io.o main.o $(LIBRABIT) $(LIBDMLC)
-wrapper/xgboost_wrapper.dll wrapper/libxgboostwrapper.so: wrapper/xgboost_wrapper.cpp src/utils/*.h src/*.h src/learner/*.hpp src/learner/*.h  updater.o gbm.o io.o $(LIBRABIT) $(LIBDMLC)
+# create standalone python tar file.
+pypack: ${XGBOOST_DYLIB}
+	cp ${XGBOOST_DYLIB} python-package/xgboost
+	cd python-package; tar cf xgboost.tar xgboost; cd ..
 
-java: java/libxgboost4j.so
-java/libxgboost4j.so: java/xgboost4j_wrapper.cpp wrapper/xgboost_wrapper.cpp src/utils/*.h src/*.h src/learner/*.hpp src/learner/*.h  updater.o gbm.o io.o $(LIBRABIT) $(LIBDMLC)
+# create pip source dist (sdist) pack for PyPI
+pippack: clean_all
+	rm -rf xgboost-python
+# remove symlinked directories in python-package/xgboost
+	rm -rf python-package/xgboost/lib
+	rm -rf python-package/xgboost/dmlc-core
+	rm -rf python-package/xgboost/include
+	rm -rf python-package/xgboost/make
+	rm -rf python-package/xgboost/rabit
+	rm -rf python-package/xgboost/src
+	cp -r python-package xgboost-python
+	cp -r Makefile xgboost-python/xgboost/
+	cp -r make xgboost-python/xgboost/
+	cp -r src xgboost-python/xgboost/
+	cp -r tests xgboost-python/xgboost/
+	cp -r include xgboost-python/xgboost/
+	cp -r dmlc-core xgboost-python/xgboost/
+	cp -r rabit xgboost-python/xgboost/
+# Use setup_pip.py instead of setup.py
+	mv xgboost-python/setup_pip.py xgboost-python/setup.py
+# Build sdist tarball
+	cd xgboost-python; python setup.py sdist; mv dist/*.tar.gz ..; cd ..
 
-# dependency on rabit
-subtree/rabit/lib/librabit.a: subtree/rabit/src/engine.cc
-	+	cd subtree/rabit;make lib/librabit.a; cd ../..
-subtree/rabit/lib/librabit_empty.a: subtree/rabit/src/engine_empty.cc
-	+	cd subtree/rabit;make lib/librabit_empty.a; cd ../..
-subtree/rabit/lib/librabit_mock.a: subtree/rabit/src/engine_mock.cc
-	+	cd subtree/rabit;make lib/librabit_mock.a; cd ../..
-subtree/rabit/lib/librabit_mpi.a: subtree/rabit/src/engine_mpi.cc
-	+	cd subtree/rabit;make lib/librabit_mpi.a; cd ../..
-
-$(BIN) :
-	$(CXX) $(CFLAGS) -fPIC -o $@ $(filter %.cpp %.o %.c %.cc %.a, $^) $(LDFLAGS)
-
-$(MOCKBIN) :
-	$(CXX) $(CFLAGS) -o $@ $(filter %.cpp %.o %.c %.cc %.a, $^) $(LDFLAGS)
-
-$(SLIB) :
-	$(CXX) $(CFLAGS) -fPIC -shared -o $@ $(filter %.cpp %.o %.c %.a %.cc, $^) $(LDFLAGS) $(DLLFLAGS)
-
-$(JLIB) :
-	$(CXX) $(CFLAGS) -fPIC -shared -o $@ $(filter %.cpp %.o %.c %.a %.cc, $^) $(LDFLAGS)  $(JAVAINCFLAGS)
-
-$(OBJ) :
-	$(CXX) -c $(CFLAGS) -o $@ $(firstword $(filter %.cpp %.c %.cc, $^) )
-
-$(MPIOBJ) :
-	$(MPICXX) -c $(CFLAGS) -o $@ $(firstword $(filter %.cpp %.c, $^) )
-
-$(MPIBIN) :
-	$(MPICXX) $(CFLAGS) -o $@ $(filter %.cpp %.o %.c %.cc %.a, $^) $(LDFLAGS)
-
-install:
-	cp -f -r $(BIN)  $(INSTALL_PATH)
-
-Rpack:
-	make clean
-	cd subtree/rabit;make clean;cd ..
+# Script to make a clean installable R package.
+Rpack: clean_all
 	rm -rf xgboost xgboost*.tar.gz
 	cp -r R-package xgboost
 	rm -rf xgboost/src/*.o xgboost/src/*.so xgboost/src/*.dll
 	rm -rf xgboost/src/*/*.o
-	rm -rf subtree/rabit/src/*.o
 	rm -rf xgboost/demo/*.model xgboost/demo/*.buffer xgboost/demo/*.txt
 	rm -rf xgboost/demo/runall.R
 	cp -r src xgboost/src/src
-	mkdir xgboost/src/subtree
-	mkdir xgboost/src/subtree/rabit
-	cp -r subtree/rabit/include xgboost/src/subtree/rabit/include
-	cp -r subtree/rabit/src xgboost/src/subtree/rabit/src
-	rm -rf xgboost/src/subtree/rabit/src/*.o
-	mkdir xgboost/src/wrapper
-	cp  wrapper/xgboost_wrapper.h xgboost/src/wrapper
-	cp  wrapper/xgboost_wrapper.cpp xgboost/src/wrapper
+	cp -r include xgboost/src/include
+	cp -r amalgamation xgboost/src/amalgamation
+	mkdir -p xgboost/src/rabit
+	cp -r rabit/include xgboost/src/rabit/include
+	cp -r rabit/src xgboost/src/rabit/src
+	rm -rf xgboost/src/rabit/src/*.o
+	mkdir -p xgboost/src/dmlc-core
+	cp -r dmlc-core/include xgboost/src/dmlc-core/include
+	cp -r dmlc-core/src xgboost/src/dmlc-core/src
 	cp ./LICENSE xgboost
-	cat R-package/src/Makevars|sed '2s/.*/PKGROOT=./' > xgboost/src/Makevars
-	cp xgboost/src/Makevars xgboost/src/Makevars.win
-	# R CMD build --no-build-vignettes xgboost
-	# R CMD build xgboost
-	# rm -rf xgboost
-	# R CMD check --as-cran xgboost*.tar.gz
+	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@/$$\(SHLIB_OPENMP_CFLAGS\)/g' xgboost/src/Makevars.win
+	bash R-package/remove_warning_suppression_pragma.sh
+	rm xgboost/remove_warning_suppression_pragma.sh
 
-Rbuild:
-	make Rpack
-	R CMD build xgboost
+Rbuild: Rpack
+	R CMD build --no-build-vignettes xgboost
 	rm -rf xgboost
 
-Rcheck:
-	make Rbuild
-	R CMD check --as-cran xgboost*.tar.gz
+Rcheck: Rbuild
+	R CMD check xgboost*.tar.gz
 
-pythonpack:
-	#for pip maintainer only
-	cd subtree/rabit;make clean;cd ..
-	rm -rf xgboost-deploy xgboost*.tar.gz
-	cp -r python-package xgboost-deploy
-	#cp *.md xgboost-deploy/
-	cp LICENSE xgboost-deploy/
-	cp Makefile xgboost-deploy/xgboost
-	cp -r wrapper xgboost-deploy/xgboost
-	cp -r subtree xgboost-deploy/xgboost
-	cp -r multi-node xgboost-deploy/xgboost
-	cp -r windows xgboost-deploy/xgboost
-	cp -r src xgboost-deploy/xgboost
-	cp python-package/setup_pip.py xgboost-deploy/setup.py
-	#make python
-
-pythonbuild:
-	make pythonpack
-	python setup.py install
-
-pythoncheck:
-	make pythonbuild
-	python -c 'import xgboost;print xgboost.core.find_lib_path()'
-
-# lint requires dmlc to be in current folder
-lint:
-	dmlc-core/scripts/lint.py xgboost $(LINT_LANG) src wrapper R-package python-package
-
-clean:
-	$(RM) -rf $(OBJ) $(BIN) $(MPIBIN) $(MPIOBJ) $(SLIB) *.o  */*.o */*/*.o *~ */*~ */*/*~
-	cd subtree/rabit; make clean; cd ..
+-include build/*.d
+-include build/*/*.d
+-include build_plugin/*/*.d

NEWS.md

@@ -0,0 +1,279 @@
+XGBoost Change Log
+==================
+
+This file records the changes in xgboost library in reverse chronological order.
+
+## 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)
+* Tree dumps now use high precision in storing floating-point values (#3298)
+* Submodules `rabit` and `dmlc-core` have been brought up to date, bringing bug fixes (#3330, #3221).
+* GPU support
+  - Continuous integration tests for GPU code (#3294, #3309)
+  - GPU accelerated coordinate descent algorithm (#3178)
+  - Abstract 1D vector class now works with multiple GPUs (#3287)
+  - Generate PTX code for most recent architecture (#3316)
+  - Fix a memory bug on NVIDIA K80 cards (#3293)
+  - Address performance instability for single-GPU, multi-core machines (#3324)
+* Python package
+  - FreeBSD support (#3247)
+  - Validation of feature names in `Booster.predict()` is now optional (#3323)
+* Updated Sklearn API
+  - Validation sets now support instance weights (#2354)
+  - `XGBClassifier.predict_proba()` should not support `output_margin` option. (#3343) See BREAKING CHANGES below.
+* R package:
+  - Better handling of NULL in `print.xgb.Booster()` (#3338)
+  - Comply with CRAN policy by removing compiler warning suppression (#3329)
+  - Updated CRAN submission
+* JVM packages
+  - JVM packages will now use the same versioning scheme as other packages (#3253)
+  - Update Spark to 2.3 (#3254)
+  - Add scripts to cross-build and deploy artifacts (#3276, #3307)
+  - Fix a compilation error for Scala 2.10 (#3332)
+* BREAKING CHANGES
+  - `XGBClassifier.predict_proba()` no longer accepts paramter `output_margin`. The paramater makes no sense for `predict_proba()` because the method is to predict class probabilities, not raw margin scores.
+
+## v0.71 (2018.04.11)
+* This is a minor release, mainly motivated by issues concerning `pip install`, e.g. #2426, #3189, #3118, and #3194.
+  With this release, users of Linux and MacOS will be able to run `pip install` for the most part.
+* Refactored linear booster class (`gblinear`), so as to support multiple coordinate descent updaters (#3103, #3134). See BREAKING CHANGES below.
+* Fix slow training for multiclass classification with high number of classes (#3109)
+* Fix a corner case in approximate quantile sketch (#3167). Applicable for 'hist' and 'gpu_hist' algorithms
+* Fix memory leak in DMatrix (#3182)
+* New functionality
+  - Better linear booster class (#3103, #3134)
+  - Pairwise SHAP interaction effects (#3043)
+  - Cox loss (#3043)
+  - AUC-PR metric for ranking task (#3172)
+  - Monotonic constraints for 'hist' algorithm (#3085)
+* GPU support
+  - Create an abtract 1D vector class that moves data seamlessly between the main and GPU memory (#2935, #3116, #3068). This eliminates unnecessary PCIe data transfer during training time.
+  - Fix minor bugs (#3051, #3217)
+  - Fix compatibility error for CUDA 9.1 (#3218)
+* Python package:
+  - Correctly handle parameter `verbose_eval=0` (#3115)
+* R package:
+  - Eliminate segmentation fault on 32-bit Windows platform (#2994)
+* JVM packages
+  - Fix a memory bug involving double-freeing Booster objects (#3005, #3011)
+  - Handle empty partition in predict (#3014)
+  - Update docs and unify terminology (#3024)
+  - Delete cache files after job finishes (#3022)
+  - Compatibility fixes for latest Spark versions (#3062, #3093)
+* BREAKING CHANGES: Updated linear modelling algorithms. In particular L1/L2 regularisation penalties are now normalised to number of training examples. This makes the implementation consistent with sklearn/glmnet. L2 regularisation has also been removed from the intercept. To produce linear models with the old regularisation behaviour, the alpha/lambda regularisation parameters can be manually scaled by dividing them by the number of training examples.
+
+## v0.7 (2017.12.30)
+* **This version represents a major change from the last release (v0.6), which was released one year and half ago.**
+* Updated Sklearn API
+  - Add compatibility layer for scikit-learn v0.18: `sklearn.cross_validation` now deprecated
+  - Updated to allow use of all XGBoost parameters via `**kwargs`.
+  - Updated `nthread` to `n_jobs` and `seed` to `random_state` (as per Sklearn convention); `nthread` and `seed` are now marked as deprecated
+  - Updated to allow choice of Booster (`gbtree`, `gblinear`, or `dart`)
+  - `XGBRegressor` now supports instance weights (specify `sample_weight` parameter)
+  - Pass `n_jobs` parameter to the `DMatrix` constructor
+  - Add `xgb_model` parameter to `fit` method, to allow continuation of training
+* Refactored gbm to allow more friendly cache strategy
+  - Specialized some prediction routine
+* Robust `DMatrix` construction from a sparse matrix
+* Faster consturction of `DMatrix` from 2D NumPy matrices: elide copies, use of multiple threads
+* Automatically remove nan from input data when it is sparse.
+  - This can solve some of user reported problem of istart != hist.size
+* Fix the single-instance prediction function to obtain correct predictions
+* Minor fixes
+  - Thread local variable is upgraded so it is automatically freed at thread exit.
+  - Fix saving and loading `count::poisson` models
+  - Fix CalcDCG to use base-2 logarithm
+  - Messages are now written to stderr instead of stdout
+  - Keep built-in evaluations while using customized evaluation functions
+  - Use `bst_float` consistently to minimize type conversion
+  - Copy the base margin when slicing `DMatrix`
+  - Evaluation metrics are now saved to the model file
+  - Use `int32_t` explicitly when serializing version
+  - In distributed training, synchronize the number of features after loading a data matrix.
+* Migrate to C++11
+  - The current master version now requires C++11 enabled compiled(g++4.8 or higher)
+* Predictor interface was factored out (in a manner similar to the updater interface).
+* Makefile support for Solaris and ARM
+* Test code coverage using Codecov
+* Add CPP tests
+* Add `Dockerfile` and `Jenkinsfile` to support continuous integration for GPU code
+* New functionality
+  - Ability to adjust tree model's statistics to a new dataset without changing tree structures.
+  - Ability to extract feature contributions from individual predictions, as described in [here](http://blog.datadive.net/interpreting-random-forests/) and [here](https://arxiv.org/abs/1706.06060).
+  - Faster, histogram-based tree algorithm (`tree_method='hist'`) .
+  - GPU/CUDA accelerated tree algorithms (`tree_method='gpu_hist'` or `'gpu_exact'`), including the GPU-based predictor.
+  - Monotonic constraints: when other features are fixed, force the prediction to be monotonic increasing with respect to a certain specified feature.
+  - Faster gradient caculation using AVX SIMD
+  - Ability to export models in JSON format
+  - Support for Tweedie regression
+  - Additional dropout options for DART: binomial+1, epsilon
+  - Ability to update an existing model in-place: this is useful for many applications, such as determining feature importance
+* Python package:
+  - New parameters:
+    - `learning_rates` in `cv()`
+    - `shuffle` in `mknfold()`
+    - `max_features` and `show_values` in `plot_importance()`
+    - `sample_weight` in `XGBRegressor.fit()`
+  - Support binary wheel builds
+  - Fix `MultiIndex` detection to support Pandas 0.21.0 and higher
+  - Support metrics and evaluation sets whose names contain `-`
+  - Support feature maps when plotting trees
+  - 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 
+* R package:
+  - New parameters:
+    - `silent` in `xgb.DMatrix()`
+    - `use_int_id` in `xgb.model.dt.tree()`
+    - `predcontrib` in `predict()`
+    - `monotone_constraints` in `xgb.train()`
+  - Default value of the `save_period` parameter in `xgboost()` changed to NULL (consistent with `xgb.train()`).
+  - It's possible to custom-build the R package with GPU acceleration support.
+  - Enable JVM build for Mac OS X and Windows
+  - Integration with AppVeyor CI
+  - Improved safety for garbage collection
+  - Store numeric attributes with higher precision
+  - Easier installation for devel version
+  - Improved `xgb.plot.tree()`
+  - Various minor fixes to improve user experience and robustness
+  - Register native code to pass CRAN check
+  - Updated CRAN submission
+* JVM packages
+  - Add Spark pipeline persistence API
+  - Fix data persistence: loss evaluation on test data had wrongly used caches for training data.
+  - Clean external cache after training
+  - Implement early stopping
+  - Enable training of multiple models by distinguishing stage IDs
+  - Better Spark integration: support RDD / dataframe / dataset, integrate with Spark ML package
+  - XGBoost4j now supports ranking task
+  - Support training with missing data
+  - Refactor JVM package to separate regression and classification models to be consistent with other machine learning libraries
+  - Support XGBoost4j compilation on Windows
+  - Parameter tuning tool
+  - Publish source code for XGBoost4j to maven local repo
+  - Scala implementation of the Rabit tracker (drop-in replacement for the Java implementation)
+  - Better exception handling for the Rabit tracker
+  - Persist `num_class`, number of classes (for classification task)
+  - `XGBoostModel` now holds `BoosterParams`
+  - libxgboost4j is now part of CMake build
+  - Release `DMatrix` when no longer needed, to conserve memory
+  - Expose `baseMargin`, to allow initialization of boosting with predictions from an external model
+  - 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 
+* Documentation
+  - Better math notation for gradient boosting
+  - Updated build instructions for Mac OS X
+  - Template for GitHub issues
+  - Add `CITATION` file for citing XGBoost in scientific writing
+  - Fix dropdown menu in xgboost.readthedocs.io
+  - Document `updater_seq` parameter
+  - Style fixes for Python documentation
+  - Links to additional examples and tutorials
+  - Clarify installation requirements
+* Changes that break backward compatibility
+  - [#1519](https://github.com/dmlc/xgboost/pull/1519) XGBoost-spark no longer contains APIs for DMatrix; use the public booster interface instead.
+  - [#2476](https://github.com/dmlc/xgboost/pull/2476) `XGBoostModel.predict()` now has a different signature
+
+
+## v0.6 (2016.07.29)
+* Version 0.5 is skipped due to major improvements in the core
+* Major refactor of core library.
+  - Goal: more flexible and modular code as a portable library.
+  - Switch to use of c++11 standard code.
+  - Random number generator defaults to ```std::mt19937```.
+  - Share the data loading pipeline and logging module from dmlc-core.
+  - Enable registry pattern to allow optionally plugin of objective, metric, tree constructor, data loader.
+    - Future plugin modules can be put into xgboost/plugin and register back to the library.
+  - Remove most of the raw pointers to smart ptrs, for RAII safety.
+* Add official option to approximate algorithm `tree_method` to parameter.
+  - Change default behavior to switch to prefer faster algorithm.
+  - User will get a message when approximate algorithm is chosen.
+* Change library name to libxgboost.so
+* Backward compatiblity
+  - The binary buffer file is not backward compatible with previous version.
+  - The model file is backward compatible on 64 bit platforms.
+* The model file is compatible between 64/32 bit platforms(not yet tested).
+* External memory version and other advanced features will be exposed to R library as well on linux.
+  - Previously some of the features are blocked due to C++11 and threading limits.
+  - The windows version is still blocked due to Rtools do not support ```std::thread```.
+* rabit and dmlc-core are maintained through git submodule
+  - Anyone can open PR to update these dependencies now.
+* Improvements
+  - Rabit and xgboost libs are not thread-safe and use thread local PRNGs
+  - This could fix some of the previous problem which runs xgboost on multiple threads.
+* JVM Package
+  - Enable xgboost4j for java and scala
+  - XGBoost distributed now runs on Flink and Spark.
+* Support model attributes listing for meta data.
+  - https://github.com/dmlc/xgboost/pull/1198
+  - https://github.com/dmlc/xgboost/pull/1166
+* Support callback API
+  - https://github.com/dmlc/xgboost/issues/892
+  - https://github.com/dmlc/xgboost/pull/1211
+  - https://github.com/dmlc/xgboost/pull/1264
+* Support new booster DART(dropout in tree boosting)
+  - https://github.com/dmlc/xgboost/pull/1220
+* Add CMake build system
+  - https://github.com/dmlc/xgboost/pull/1314
+
+## v0.47 (2016.01.14)
+
+* Changes in R library
+  - fixed possible problem of poisson regression.
+  - switched from 0 to NA for missing values.
+  - exposed access to additional model parameters.
+* Changes in Python library
+  - throws exception instead of crash terminal when a parameter error happens.
+  - has importance plot and tree plot functions.
+  - accepts different learning rates for each boosting round.
+  - allows model training continuation from previously saved model.
+  - allows early stopping in CV.
+  - allows feval to return a list of tuples.
+  - allows eval_metric to handle additional format.
+  - improved compatibility in sklearn module.
+  - additional parameters added for sklearn wrapper.
+  - added pip installation functionality.
+  - supports more Pandas DataFrame dtypes.
+  - added best_ntree_limit attribute, in addition to best_score and best_iteration.
+* Java api is ready for use
+* Added more test cases and continuous integration to make each build more robust.
+
+## v0.4 (2015.05.11)
+
+* Distributed version of xgboost that runs on YARN, scales to billions of examples
+* Direct save/load data and model from/to S3 and HDFS
+* Feature importance visualization in R module, by Michael Benesty
+* Predict leaf index
+* Poisson regression for counts data
+* Early stopping option in training
+* Native save load support in R and python
+  - xgboost models now can be saved using save/load in R
+  - xgboost python model is now pickable
+* sklearn wrapper is supported in python module
+* Experimental External memory version
+
+
+## v0.3 (2014.09.07)
+
+* Faster tree construction module
+  - Allows subsample columns during tree construction via ```bst:col_samplebytree=ratio```
+* Support for boosting from initial predictions
+* Experimental version of LambdaRank
+* Linear booster is now parallelized, using parallel coordinated descent.
+* Add [Code Guide](src/README.md) for customizing objective function and evaluation
+* Add R module
+
+
+## v0.2x (2014.05.20)
+
+* Python module
+* Weighted samples instances
+* Initial version of pairwise rank
+
+
+## v0.1 (2014.03.26)
+
+* Initial release

R-package/DESCRIPTION

@@ -1,37 +1,64 @@
 Package: xgboost
 Type: Package
 Title: Extreme Gradient Boosting
-Version: 0.4-2
-Date: 2015-08-01
-Author: Tianqi Chen <tianqi.tchen@gmail.com>, Tong He <hetong007@gmail.com>,
-    Michael Benesty <michael@benesty.fr>
-Maintainer: Tong He <hetong007@gmail.com>
+Version: 0.71.1
+Date: 2018-05-11
+Authors@R: c(
+  person("Tianqi", "Chen", role = c("aut"),
+         email = "tianqi.tchen@gmail.com"),
+  person("Tong", "He", role = c("aut", "cre"),
+         email = "hetong007@gmail.com"),
+  person("Michael", "Benesty", role = c("aut"),
+         email = "michael@benesty.fr"),
+  person("Vadim", "Khotilovich", role = c("aut"),
+         email = "khotilovich@gmail.com"),
+  person("Yuan", "Tang", role = c("aut"),
+         email = "terrytangyuan@gmail.com",
+         comment = c(ORCID = "0000-0001-5243-233X")),
+  person("Hyunsu", "Cho", role = c("aut"),
+         email = "chohyu01@cs.washington.edu"),
+  person("Kailong", "Chen", role = c("aut")),
+  person("Rory", "Mitchell", role = c("aut")),
+  person("Ignacio", "Cano", role = c("aut")),
+  person("Tianyi", "Zhou", role = c("aut")),
+  person("Mu", "Li", role = c("aut")),
+  person("Junyuan", "Xie", role = c("aut")),
+  person("Min", "Lin", role = c("aut")),
+  person("Yifeng", "Geng", role = c("aut")),
+  person("Yutian", "Li", role = c("aut")),
+  person("XGBoost contributors", role = c("cph"),
+         comment = "base XGBoost implementation")
+  )
 Description: Extreme Gradient Boosting, which is an efficient implementation
-    of gradient boosting framework. This package is its R interface. The package
-    includes efficient linear model solver and tree learning algorithms. The package
-    can automatically do parallel computation on a single machine which could be
-    more than 10 times faster than existing gradient boosting packages. It supports
+    of the gradient boosting framework from Chen & Guestrin (2016) <doi:10.1145/2939672.2939785>.
+    This package is its R interface. The package includes efficient linear 
+    model solver and tree learning algorithms. The package can automatically 
+    do parallel computation on a single machine which could be more than 10 
+    times faster than existing gradient boosting packages. It supports
     various objective functions, including regression, classification and ranking.
     The package is made to be extensible, so that users are also allowed to define
     their own objectives easily.
 License: Apache License (== 2.0) | file LICENSE
 URL: https://github.com/dmlc/xgboost
 BugReports: https://github.com/dmlc/xgboost/issues
+NeedsCompilation: yes
 VignetteBuilder: knitr
 Suggests:
     knitr,
+    rmarkdown,
     ggplot2 (>= 1.0.1),
-    DiagrammeR (>= 0.8.1),
+    DiagrammeR (>= 0.9.0),
     Ckmeans.1d.dp (>= 3.3.1),
     vcd (>= 1.3),
     testthat,
     igraph (>= 1.0.1)
 Depends:
-    R (>= 2.10)
+    R (>= 3.3.0)
 Imports:
     Matrix (>= 1.1-0),
     methods,
     data.table (>= 1.9.6),
     magrittr (>= 1.5),
-    stringr (>= 0.6.2)
-RoxygenNote: 5.0.1
+    stringi (>= 0.5.2)
+RoxygenNote: 6.0.1
+SystemRequirements: GNU make, C++11

R-package/NAMESPACE

@@ -1,48 +1,86 @@
 # Generated by roxygen2: do not edit by hand
 
+S3method("[",xgb.DMatrix)
+S3method("dimnames<-",xgb.DMatrix)
+S3method(dim,xgb.DMatrix)
+S3method(dimnames,xgb.DMatrix)
+S3method(getinfo,xgb.DMatrix)
+S3method(predict,xgb.Booster)
+S3method(predict,xgb.Booster.handle)
+S3method(print,xgb.Booster)
+S3method(print,xgb.DMatrix)
+S3method(print,xgb.cv.synchronous)
+S3method(setinfo,xgb.DMatrix)
+S3method(slice,xgb.DMatrix)
+export("xgb.attr<-")
+export("xgb.attributes<-")
+export("xgb.parameters<-")
+export(cb.cv.predict)
+export(cb.early.stop)
+export(cb.evaluation.log)
+export(cb.gblinear.history)
+export(cb.print.evaluation)
+export(cb.reset.parameters)
+export(cb.save.model)
 export(getinfo)
 export(setinfo)
 export(slice)
+export(xgb.Booster.complete)
 export(xgb.DMatrix)
 export(xgb.DMatrix.save)
+export(xgb.attr)
+export(xgb.attributes)
 export(xgb.create.features)
 export(xgb.cv)
 export(xgb.dump)
+export(xgb.gblinear.history)
+export(xgb.ggplot.deepness)
+export(xgb.ggplot.importance)
 export(xgb.importance)
 export(xgb.load)
 export(xgb.model.dt.tree)
 export(xgb.plot.deepness)
 export(xgb.plot.importance)
 export(xgb.plot.multi.trees)
+export(xgb.plot.shap)
 export(xgb.plot.tree)
 export(xgb.save)
 export(xgb.save.raw)
 export(xgb.train)
 export(xgboost)
-exportMethods(nrow)
-exportMethods(predict)
 import(methods)
 importClassesFrom(Matrix,dgCMatrix)
 importClassesFrom(Matrix,dgeMatrix)
-importFrom(Matrix,cBind)
 importFrom(Matrix,colSums)
 importFrom(Matrix,sparse.model.matrix)
+importFrom(Matrix,sparseMatrix)
 importFrom(Matrix,sparseVector)
+importFrom(Matrix,t)
 importFrom(data.table,":=")
 importFrom(data.table,as.data.table)
-importFrom(data.table,copy)
 importFrom(data.table,data.table)
-importFrom(data.table,fread)
+importFrom(data.table,is.data.table)
 importFrom(data.table,rbindlist)
-importFrom(data.table,set)
+importFrom(data.table,setkey)
+importFrom(data.table,setkeyv)
 importFrom(data.table,setnames)
+importFrom(grDevices,rgb)
+importFrom(graphics,barplot)
+importFrom(graphics,grid)
+importFrom(graphics,lines)
+importFrom(graphics,par)
+importFrom(graphics,points)
+importFrom(graphics,title)
 importFrom(magrittr,"%>%")
-importFrom(magrittr,add)
-importFrom(magrittr,not)
-importFrom(stringr,str_detect)
-importFrom(stringr,str_extract)
-importFrom(stringr,str_extract_all)
-importFrom(stringr,str_match)
-importFrom(stringr,str_replace)
-importFrom(stringr,str_split)
-importFrom(stringr,str_trim)
+importFrom(stats,median)
+importFrom(stats,predict)
+importFrom(stringi,stri_detect_regex)
+importFrom(stringi,stri_match_first_regex)
+importFrom(stringi,stri_replace_all_regex)
+importFrom(stringi,stri_replace_first_regex)
+importFrom(stringi,stri_split_regex)
+importFrom(utils,head)
+importFrom(utils,object.size)
+importFrom(utils,str)
+importFrom(utils,tail)
+useDynLib(xgboost, .registration = TRUE)

R-package/R/callbacks.R

@@ -0,0 +1,831 @@
+#' Callback closures for booster training.
+#'
+#' These are used to perform various service tasks either during boosting iterations or at the end.
+#' This approach helps to modularize many of such tasks without bloating the main training methods, 
+#' and it offers .
+#' 
+#' @details
+#' By default, a callback function is run after each boosting iteration.
+#' An R-attribute \code{is_pre_iteration} could be set for a callback to define a pre-iteration function.
+#' 
+#' When a callback function has \code{finalize} parameter, its finalizer part will also be run after 
+#' the boosting is completed.
+#' 
+#' WARNING: side-effects!!! Be aware that these callback functions access and modify things in 
+#' the environment from which they are called from, which is a fairly uncommon thing to do in R.
+#' 
+#' To write a custom callback closure, make sure you first understand the main concepts about R envoronments.
+#' Check either R documentation on \code{\link[base]{environment}} or the 
+#' \href{http://adv-r.had.co.nz/Environments.html}{Environments chapter} from the "Advanced R" 
+#' book by Hadley Wickham. Further, the best option is to read the code of some of the existing callbacks -
+#' choose ones that do something similar to what you want to achieve. Also, you would need to get familiar 
+#' with the objects available inside of the \code{xgb.train} and \code{xgb.cv} internal environments.
+#' 
+#' @seealso
+#' \code{\link{cb.print.evaluation}},
+#' \code{\link{cb.evaluation.log}},
+#' \code{\link{cb.reset.parameters}},
+#' \code{\link{cb.early.stop}},
+#' \code{\link{cb.save.model}},
+#' \code{\link{cb.cv.predict}},
+#' \code{\link{xgb.train}},
+#' \code{\link{xgb.cv}}
+#' 
+#' @name callbacks
+NULL
+
+#
+# Callbacks -------------------------------------------------------------------
+# 
+
+#' Callback closure for printing the result of evaluation
+#' 
+#' @param period  results would be printed every number of periods
+#' @param showsd  whether standard deviations should be printed (when available)
+#' 
+#' @details
+#' The callback function prints the result of evaluation at every \code{period} iterations.
+#' The initial and the last iteration's evaluations are always printed.
+#' 
+#' Callback function expects the following values to be set in its calling frame:
+#' \code{bst_evaluation} (also \code{bst_evaluation_err} when available),
+#' \code{iteration},
+#' \code{begin_iteration},
+#' \code{end_iteration}.
+#' 
+#' @seealso
+#' \code{\link{callbacks}}
+#' 
+#' @export
+cb.print.evaluation <- function(period = 1, showsd = TRUE) {
+  
+  callback <- function(env = parent.frame()) {
+    if (length(env$bst_evaluation) == 0 ||
+        period == 0 ||
+        NVL(env$rank, 0) != 0 )
+      return()
+    
+    i <- env$iteration 
+    if ((i-1) %% period == 0 ||
+        i == env$begin_iteration ||
+        i == env$end_iteration) {
+      stdev <- if (showsd) env$bst_evaluation_err else NULL
+      msg <- format.eval.string(i, env$bst_evaluation, stdev)
+      cat(msg, '\n')
+    }
+  }
+  attr(callback, 'call') <- match.call()
+  attr(callback, 'name') <- 'cb.print.evaluation'
+  callback
+}
+
+
+#' Callback closure for logging the evaluation history
+#' 
+#' @details
+#' This callback function appends the current iteration evaluation results \code{bst_evaluation}
+#' available in the calling parent frame to the \code{evaluation_log} list in a calling frame.
+#' 
+#' The finalizer callback (called with \code{finalize = TURE} in the end) converts 
+#' the \code{evaluation_log} list into a final data.table.
+#' 
+#' The iteration evaluation result \code{bst_evaluation} must be a named numeric vector. 
+#' 
+#' Note: in the column names of the final data.table, the dash '-' character is replaced with 
+#' the underscore '_' in order to make the column names more like regular R identifiers.
+#' 
+#' Callback function expects the following values to be set in its calling frame:
+#' \code{evaluation_log},
+#' \code{bst_evaluation},
+#' \code{iteration}.
+#' 
+#' @seealso
+#' \code{\link{callbacks}}
+#' 
+#' @export
+cb.evaluation.log <- function() {
+
+  mnames <- NULL
+  
+  init <- function(env) {
+    if (!is.list(env$evaluation_log))
+      stop("'evaluation_log' has to be a list")
+    mnames <<- names(env$bst_evaluation)
+    if (is.null(mnames) || any(mnames == ""))
+      stop("bst_evaluation must have non-empty names")
+    
+    mnames <<- gsub('-', '_', names(env$bst_evaluation))
+    if(!is.null(env$bst_evaluation_err))
+      mnames <<- c(paste0(mnames, '_mean'), paste0(mnames, '_std'))
+  }
+  
+  finalizer <- function(env) {
+    env$evaluation_log <- as.data.table(t(simplify2array(env$evaluation_log)))
+    setnames(env$evaluation_log, c('iter', mnames))
+    
+    if(!is.null(env$bst_evaluation_err)) {
+      # rearrange col order from _mean,_mean,...,_std,_std,...
+      # to be _mean,_std,_mean,_std,...
+      len <- length(mnames)
+      means <- mnames[seq_len(len/2)]
+      stds <- mnames[(len/2 + 1):len]
+      cnames <- numeric(len)
+      cnames[c(TRUE, FALSE)] <- means
+      cnames[c(FALSE, TRUE)] <- stds
+      env$evaluation_log <- env$evaluation_log[, c('iter', cnames), with = FALSE]
+    }
+  }
+  
+  callback <- function(env = parent.frame(), finalize = FALSE) {
+    if (is.null(mnames))
+      init(env)
+
+    if (finalize)
+      return(finalizer(env))
+    
+    ev <- env$bst_evaluation
+    if(!is.null(env$bst_evaluation_err))
+      ev <- c(ev, env$bst_evaluation_err)
+    env$evaluation_log <- c(env$evaluation_log, 
+                            list(c(iter = env$iteration, ev)))
+  }
+  attr(callback, 'call') <- match.call()
+  attr(callback, 'name') <- 'cb.evaluation.log'
+  callback
+}
+
+#' Callback closure for restetting the booster's parameters at each iteration.
+#' 
+#' @param new_params a list where each element corresponds to a parameter that needs to be reset.
+#'        Each element's value must be either a vector of values of length \code{nrounds} 
+#'        to be set at each iteration, 
+#'        or a function of two parameters \code{learning_rates(iteration, nrounds)} 
+#'        which returns a new parameter value by using the current iteration number 
+#'        and the total number of boosting rounds.
+#' 
+#' @details 
+#' This is a "pre-iteration" callback function used to reset booster's parameters
+#' at the beginning of each iteration.
+#' 
+#' Note that when training is resumed from some previous model, and a function is used to 
+#' reset a parameter value, the \code{nround} argument in this function would be the 
+#' the number of boosting rounds in the current training.
+#'
+#' Callback function expects the following values to be set in its calling frame:
+#' \code{bst} or \code{bst_folds},
+#' \code{iteration},
+#' \code{begin_iteration},
+#' \code{end_iteration}.
+#' 
+#' @seealso
+#' \code{\link{callbacks}}
+#' 
+#' @export
+cb.reset.parameters <- function(new_params) {
+
+  if (typeof(new_params) != "list") 
+    stop("'new_params' must be a list")
+  pnames <- gsub("\\.", "_", names(new_params))
+  nrounds <- NULL
+  
+  # run some checks in the begining
+  init <- function(env) {
+    nrounds <<- env$end_iteration - env$begin_iteration + 1
+    
+    if (is.null(env$bst) && is.null(env$bst_folds))
+      stop("Parent frame has neither 'bst' nor 'bst_folds'")
+    
+    # Some parameters are not allowed to be changed,
+    # since changing them would simply wreck some chaos
+    not_allowed <- pnames %in% 
+      c('num_class', 'num_output_group', 'size_leaf_vector', 'updater_seq')
+    if (any(not_allowed))
+      stop('Parameters ', paste(pnames[not_allowed]), " cannot be changed during boosting.")
+    
+    for (n in pnames) {
+      p <- new_params[[n]]
+      if (is.function(p)) {
+        if (length(formals(p)) != 2)
+          stop("Parameter '", n, "' is a function but not of two arguments")
+      } else if (is.numeric(p) || is.character(p)) {
+        if (length(p) != nrounds)
+          stop("Length of '", n, "' has to be equal to 'nrounds'")
+      } else {
+        stop("Parameter '", n, "' is not a function or a vector")
+      }
+    }
+  }
+  
+  callback <- function(env = parent.frame()) {
+    if (is.null(nrounds))
+      init(env)
+    
+    i <- env$iteration
+    pars <- lapply(new_params, function(p) {
+      if (is.function(p))
+        return(p(i, nrounds))
+      p[i]
+    })
+    
+    if (!is.null(env$bst)) {
+      xgb.parameters(env$bst$handle) <- pars
+    } else {
+      for (fd in env$bst_folds)
+        xgb.parameters(fd$bst) <- pars
+    }
+  }
+  attr(callback, 'is_pre_iteration') <- TRUE
+  attr(callback, 'call') <- match.call()
+  attr(callback, 'name') <- 'cb.reset.parameters'
+  callback
+}
+
+
+#' Callback closure to activate the early stopping.
+#' 
+#' @param stopping_rounds The number of rounds with no improvement in 
+#'        the evaluation metric in order to stop the training.
+#' @param maximize whether to maximize the evaluation metric
+#' @param metric_name the name of an evaluation column to use as a criteria for early
+#'        stopping. If not set, the last column would be used.
+#'        Let's say the test data in \code{watchlist} was labelled as \code{dtest}, 
+#'        and one wants to use the AUC in test data for early stopping regardless of where 
+#'        it is in the \code{watchlist}, then one of the following would need to be set:
+#'        \code{metric_name='dtest-auc'} or \code{metric_name='dtest_auc'}.
+#'        All dash '-' characters in metric names are considered equivalent to '_'.
+#' @param verbose whether to print the early stopping information.
+#' 
+#' @details
+#' This callback function determines the condition for early stopping 
+#' by setting the \code{stop_condition = TRUE} flag in its calling frame.
+#' 
+#' The following additional fields are assigned to the model's R object:
+#' \itemize{
+#' \item \code{best_score} the evaluation score at the best iteration
+#' \item \code{best_iteration} at which boosting iteration the best score has occurred (1-based index)
+#' \item \code{best_ntreelimit} to use with the \code{ntreelimit} parameter in \code{predict}.
+#'      It differs from \code{best_iteration} in multiclass or random forest settings.
+#' }
+#' 
+#' The Same values are also stored as xgb-attributes:
+#' \itemize{
+#' \item \code{best_iteration} is stored as a 0-based iteration index (for interoperability of binary models)
+#' \item \code{best_msg} message string is also stored.
+#' }
+#' 
+#' At least one data element is required in the evaluation watchlist for early stopping to work.
+#'
+#' Callback function expects the following values to be set in its calling frame:
+#' \code{stop_condition},
+#' \code{bst_evaluation},
+#' \code{rank},
+#' \code{bst} (or \code{bst_folds} and \code{basket}),
+#' \code{iteration},
+#' \code{begin_iteration},
+#' \code{end_iteration},
+#' \code{num_parallel_tree}.
+#' 
+#' @seealso
+#' \code{\link{callbacks}},
+#' \code{\link{xgb.attr}}
+#' 
+#' @export
+cb.early.stop <- function(stopping_rounds, maximize = FALSE, 
+                          metric_name = NULL, verbose = TRUE) {
+  # state variables
+  best_iteration <- -1
+  best_ntreelimit <- -1
+  best_score <- Inf
+  best_msg <- NULL
+  metric_idx <- 1
+  
+  init <- function(env) {
+    if (length(env$bst_evaluation) == 0)
+      stop("For early stopping, watchlist must have at least one element")
+    
+    eval_names <- gsub('-', '_', names(env$bst_evaluation))
+    if (!is.null(metric_name)) {
+      metric_idx <<- which(gsub('-', '_', metric_name) == eval_names)
+      if (length(metric_idx) == 0)
+        stop("'metric_name' for early stopping is not one of the following:\n",
+             paste(eval_names, collapse = ' '), '\n')
+    }
+    if (is.null(metric_name) &&
+        length(env$bst_evaluation) > 1) {
+      metric_idx <<- length(eval_names)
+      if (verbose)
+        cat('Multiple eval metrics are present. Will use ', 
+            eval_names[metric_idx], ' for early stopping.\n', sep = '')
+    }
+    
+    metric_name <<- eval_names[metric_idx]
+    
+    # maximize is usually NULL when not set in xgb.train and built-in metrics
+    if (is.null(maximize))
+      maximize <<- grepl('(_auc|_map|_ndcg)', metric_name)
+
+    if (verbose && NVL(env$rank, 0) == 0)
+      cat("Will train until ", metric_name, " hasn't improved in ", 
+          stopping_rounds, " rounds.\n\n", sep = '')
+
+    best_iteration <<- 1
+    if (maximize) best_score <<- -Inf
+    
+    env$stop_condition <- FALSE
+    
+    if (!is.null(env$bst)) {
+      if (!inherits(env$bst, 'xgb.Booster'))
+        stop("'bst' in the parent frame must be an 'xgb.Booster'")
+      if (!is.null(best_score <- xgb.attr(env$bst$handle, 'best_score'))) {
+        best_score <<- as.numeric(best_score)
+        best_iteration <<- as.numeric(xgb.attr(env$bst$handle, 'best_iteration')) + 1
+        best_msg <<- as.numeric(xgb.attr(env$bst$handle, 'best_msg'))
+      } else {
+        xgb.attributes(env$bst$handle) <- list(best_iteration = best_iteration - 1,
+                                               best_score = best_score)
+      }
+    } else if (is.null(env$bst_folds) || is.null(env$basket)) {
+      stop("Parent frame has neither 'bst' nor ('bst_folds' and 'basket')")
+    }
+  }
+  
+  finalizer <- function(env) {
+    if (!is.null(env$bst)) {
+      attr_best_score = as.numeric(xgb.attr(env$bst$handle, 'best_score'))
+      if (best_score != attr_best_score)
+        stop("Inconsistent 'best_score' values between the closure state: ", best_score,
+             " and the xgb.attr: ", attr_best_score)
+      env$bst$best_iteration = best_iteration
+      env$bst$best_ntreelimit = best_ntreelimit
+      env$bst$best_score = best_score
+    } else {
+      env$basket$best_iteration <- best_iteration
+      env$basket$best_ntreelimit <- best_ntreelimit
+    }
+  }
+
+  callback <- function(env = parent.frame(), finalize = FALSE) {
+    if (best_iteration < 0)
+      init(env)
+    
+    if (finalize)
+      return(finalizer(env))
+    
+    i <- env$iteration
+    score = env$bst_evaluation[metric_idx]
+    
+    if (( maximize && score > best_score) ||
+        (!maximize && score < best_score)) {
+      
+      best_msg <<- format.eval.string(i, env$bst_evaluation, env$bst_evaluation_err)
+      best_score <<- score
+      best_iteration <<- i
+      best_ntreelimit <<- best_iteration * env$num_parallel_tree
+      # save the property to attributes, so they will occur in checkpoint
+      if (!is.null(env$bst)) {
+        xgb.attributes(env$bst) <- list(
+          best_iteration = best_iteration - 1, # convert to 0-based index
+          best_score = best_score,
+          best_msg = best_msg,
+          best_ntreelimit = best_ntreelimit)
+      }
+    } else if (i - best_iteration >= stopping_rounds) {
+      env$stop_condition <- TRUE
+      env$end_iteration <- i
+      if (verbose && NVL(env$rank, 0) == 0)
+        cat("Stopping. Best iteration:\n", best_msg, "\n\n", sep = '')
+    }
+  }
+  attr(callback, 'call') <- match.call()
+  attr(callback, 'name') <- 'cb.early.stop'
+  callback
+}
+
+
+#' Callback closure for saving a model file.
+#' 
+#' @param save_period save the model to disk after every 
+#'        \code{save_period} iterations; 0 means save the model at the end.
+#' @param save_name the name or path for the saved model file.
+#'        It can contain a \code{\link[base]{sprintf}} formatting specifier 
+#'        to include the integer iteration number in the file name.
+#'        E.g., with \code{save_name} = 'xgboost_%04d.model', 
+#'        the file saved at iteration 50 would be named "xgboost_0050.model".
+#' 
+#' @details 
+#' This callback function allows to save an xgb-model file, either periodically after each \code{save_period}'s or at the end.
+#' 
+#' Callback function expects the following values to be set in its calling frame:
+#' \code{bst},
+#' \code{iteration},
+#' \code{begin_iteration},
+#' \code{end_iteration}.
+#' 
+#' @seealso
+#' \code{\link{callbacks}}
+#' 
+#' @export
+cb.save.model <- function(save_period = 0, save_name = "xgboost.model") {
+  
+  if (save_period < 0)
+    stop("'save_period' cannot be negative")
+
+  callback <- function(env = parent.frame()) {
+    if (is.null(env$bst))
+      stop("'save_model' callback requires the 'bst' booster object in its calling frame")
+    
+    if ((save_period > 0 && (env$iteration - env$begin_iteration) %% save_period == 0) ||
+        (save_period == 0 && env$iteration == env$end_iteration))
+      xgb.save(env$bst, sprintf(save_name, env$iteration))
+  }
+  attr(callback, 'call') <- match.call()
+  attr(callback, 'name') <- 'cb.save.model'
+  callback
+}
+
+
+#' Callback closure for returning cross-validation based predictions.
+#' 
+#' @param save_models a flag for whether to save the folds' models.
+#' 
+#' @details 
+#' This callback function saves predictions for all of the test folds,
+#' and also allows to save the folds' models.
+#' 
+#' It is a "finalizer" callback and it uses early stopping information whenever it is available,
+#' thus it must be run after the early stopping callback if the early stopping is used.
+#' 
+#' Callback function expects the following values to be set in its calling frame:
+#' \code{bst_folds},
+#' \code{basket},
+#' \code{data},
+#' \code{end_iteration},
+#' \code{params},
+#' \code{num_parallel_tree},
+#' \code{num_class}.
+#' 
+#' @return 
+#' Predictions are returned inside of the \code{pred} element, which is either a vector or a matrix,
+#' depending on the number of prediction outputs per data row. The order of predictions corresponds 
+#' to the order of rows in the original dataset. Note that when a custom \code{folds} list is 
+#' provided in \code{xgb.cv}, the predictions would only be returned properly when this list is a 
+#' non-overlapping list of k sets of indices, as in a standard k-fold CV. The predictions would not be 
+#' meaningful when user-profided folds have overlapping indices as in, e.g., random sampling splits.
+#' When some of the indices in the training dataset are not included into user-provided \code{folds},
+#' their prediction value would be \code{NA}.
+#' 
+#' @seealso
+#' \code{\link{callbacks}}
+#' 
+#' @export
+cb.cv.predict <- function(save_models = FALSE) {
+  
+  finalizer <- function(env) {
+    if (is.null(env$basket) || is.null(env$bst_folds))
+      stop("'cb.cv.predict' callback requires 'basket' and 'bst_folds' lists in its calling frame")
+    
+    N <- nrow(env$data)
+    pred <- 
+      if (env$num_class > 1) {
+        matrix(NA_real_, N, env$num_class)
+      } else {
+        rep(NA_real_, N)
+      }
+
+    ntreelimit <- NVL(env$basket$best_ntreelimit, 
+                      env$end_iteration * env$num_parallel_tree)
+    if (NVL(env$params[['booster']], '') == 'gblinear') {
+      ntreelimit <- 0 # must be 0 for gblinear
+    }
+    for (fd in env$bst_folds) {
+      pr <- predict(fd$bst, fd$watchlist[[2]], ntreelimit = ntreelimit, reshape = TRUE)
+      if (is.matrix(pred)) {
+        pred[fd$index,] <- pr
+      } else {
+        pred[fd$index] <- pr
+      }
+    }
+    env$basket$pred <- pred
+    if (save_models) {
+      env$basket$models <- lapply(env$bst_folds, function(fd) {
+        xgb.attr(fd$bst, 'niter') <- env$end_iteration - 1
+        xgb.Booster.complete(xgb.handleToBooster(fd$bst), saveraw = TRUE)
+      })
+    }
+  }
+
+  callback <- function(env = parent.frame(), finalize = FALSE) {
+    if (finalize)
+      return(finalizer(env))
+  }
+  attr(callback, 'call') <- match.call()
+  attr(callback, 'name') <- 'cb.cv.predict'
+  callback
+}
+
+
+#' Callback closure for collecting the model coefficients history of a gblinear booster
+#' during its training.
+#'
+#' @param sparse when set to FALSE/TURE, a dense/sparse matrix is used to store the result.
+#'       Sparse format is useful when one expects only a subset of coefficients to be non-zero,
+#'       when using the "thrifty" feature selector with fairly small number of top features
+#'       selected per iteration.
+#'
+#' @details
+#' To keep things fast and simple, gblinear booster does not internally store the history of linear
+#' model coefficients at each boosting iteration. This callback provides a workaround for storing
+#' the coefficients' path, by extracting them after each training iteration.
+#'
+#' Callback function expects the following values to be set in its calling frame:
+#' \code{bst} (or \code{bst_folds}).
+#'
+#' @return
+#' Results are stored in the \code{coefs} element of the closure.
+#' The \code{\link{xgb.gblinear.history}} convenience function provides an easy way to access it.
+#' With \code{xgb.train}, it is either a dense of a sparse matrix.
+#' While with \code{xgb.cv}, it is a list (an element per each fold) of such matrices.
+#'
+#' @seealso
+#' \code{\link{callbacks}}, \code{\link{xgb.gblinear.history}}.
+#'
+#' @examples
+#' #### Binary classification:
+#' #
+#' # In the iris dataset, it is hard to linearly separate Versicolor class from the rest
+#' # without considering the 2nd order interactions:
+#' require(magrittr)
+#' x <- model.matrix(Species ~ .^2, iris)[,-1]
+#' colnames(x)
+#' dtrain <- xgb.DMatrix(scale(x), label = 1*(iris$Species == "versicolor"))
+#' param <- list(booster = "gblinear", objective = "reg:logistic", eval_metric = "auc",
+#'               lambda = 0.0003, alpha = 0.0003, nthread = 2)
+#' # For 'shotgun', which is a default linear updater, using high eta values may result in
+#' # unstable behaviour in some datasets. With this simple dataset, however, the high learning
+#' # rate does not break the convergence, but allows us to illustrate the typical pattern of
+#' # "stochastic explosion" behaviour of this lock-free algorithm at early boosting iterations.
+#' bst <- xgb.train(param, dtrain, list(tr=dtrain), nrounds = 200, eta = 1.,
+#'                  callbacks = list(cb.gblinear.history()))
+#' # Extract the coefficients' path and plot them vs boosting iteration number:
+#' coef_path <- xgb.gblinear.history(bst)
+#' matplot(coef_path, type = 'l')
+#' 
+#' # With the deterministic coordinate descent updater, it is safer to use higher learning rates.
+#' # Will try the classical componentwise boosting which selects a single best feature per round:
+#' bst <- xgb.train(param, dtrain, list(tr=dtrain), nrounds = 200, eta = 0.8,
+#'                  updater = 'coord_descent', feature_selector = 'thrifty', top_k = 1,
+#'                  callbacks = list(cb.gblinear.history()))
+#' xgb.gblinear.history(bst) %>% matplot(type = 'l')
+#' #  Componentwise boosting is known to have similar effect to Lasso regularization.
+#' # Try experimenting with various values of top_k, eta, nrounds,
+#' # as well as different feature_selectors.
+#'
+#' # For xgb.cv:
+#' bst <- xgb.cv(param, dtrain, nfold = 5, nrounds = 100, eta = 0.8,
+#'              callbacks = list(cb.gblinear.history()))
+#' # coefficients in the CV fold #3
+#' xgb.gblinear.history(bst)[[3]] %>% matplot(type = 'l')
+#'
+#' 
+#' #### Multiclass classification:
+#' #
+#' dtrain <- xgb.DMatrix(scale(x), label = as.numeric(iris$Species) - 1)
+#' param <- list(booster = "gblinear", objective = "multi:softprob", num_class = 3,
+#'               lambda = 0.0003, alpha = 0.0003, nthread = 2)
+#' # For the default linear updater 'shotgun' it sometimes is helpful
+#' # to use smaller eta to reduce instability
+#' bst <- xgb.train(param, dtrain, list(tr=dtrain), nrounds = 70, eta = 0.5,
+#'                  callbacks = list(cb.gblinear.history()))
+#' # Will plot the coefficient paths separately for each class:
+#' xgb.gblinear.history(bst, class_index = 0) %>% matplot(type = 'l')
+#' xgb.gblinear.history(bst, class_index = 1) %>% matplot(type = 'l')
+#' xgb.gblinear.history(bst, class_index = 2) %>% matplot(type = 'l')
+#'
+#' # CV:
+#' bst <- xgb.cv(param, dtrain, nfold = 5, nrounds = 70, eta = 0.5,
+#'               callbacks = list(cb.gblinear.history(FALSE)))
+#' # 1st forld of 1st class
+#' xgb.gblinear.history(bst, class_index = 0)[[1]] %>% matplot(type = 'l')
+#'
+#' @export
+cb.gblinear.history <- function(sparse=FALSE) {
+  coefs <- NULL
+
+  init <- function(env) {
+    if (!is.null(env$bst)) { # xgb.train:
+      coef_path <- list()
+    } else if (!is.null(env$bst_folds)) { # xgb.cv:
+      coef_path <- rep(list(), length(env$bst_folds))
+    } else stop("Parent frame has neither 'bst' nor 'bst_folds'")
+  }
+
+  # convert from list to (sparse) matrix
+  list2mat <- function(coef_list) {
+    if (sparse) {
+      coef_mat <- sparseMatrix(x = unlist(lapply(coef_list, slot, "x")),
+                               i = unlist(lapply(coef_list, slot, "i")),
+                               p = c(0, cumsum(sapply(coef_list, function(x) length(x@x)))),
+                               dims = c(length(coef_list[[1]]), length(coef_list)))
+      return(t(coef_mat))
+    } else {
+      return(do.call(rbind, coef_list))
+    }
+  }
+
+  finalizer <- function(env) {
+    if (length(coefs) == 0)
+      return()
+    if (!is.null(env$bst)) { # # xgb.train:
+      coefs <<- list2mat(coefs)
+    } else { # xgb.cv:
+      # first lapply transposes the list
+      coefs <<- lapply(seq_along(coefs[[1]]), function(i) lapply(coefs, "[[", i)) %>%
+                lapply(function(x) list2mat(x))
+    }
+  }
+
+  extract.coef <- function(env) {
+    if (!is.null(env$bst)) { # # xgb.train:
+      cf <- as.numeric(grep('(booster|bias|weigh)', xgb.dump(env$bst), invert = TRUE, value = TRUE))
+      if (sparse) cf <- as(cf, "sparseVector")
+    } else { # xgb.cv:
+      cf <- vector("list", length(env$bst_folds))
+      for (i in seq_along(env$bst_folds)) {
+        dmp <- xgb.dump(xgb.handleToBooster(env$bst_folds[[i]]$bst))
+        cf[[i]] <- as.numeric(grep('(booster|bias|weigh)', dmp, invert = TRUE, value = TRUE))
+        if (sparse) cf[[i]] <- as(cf[[i]], "sparseVector")
+      }
+    }
+    cf
+  }
+
+  callback <- function(env = parent.frame(), finalize = FALSE) {
+    if (is.null(coefs)) init(env)
+    if (finalize) return(finalizer(env))
+    cf <- extract.coef(env)
+    coefs <<- c(coefs, list(cf))
+  }
+
+  attr(callback, 'call') <- match.call()
+  attr(callback, 'name') <- 'cb.gblinear.history'
+  callback
+}
+
+#' Extract gblinear coefficients history.
+#'
+#' A helper function to extract the matrix of linear coefficients' history
+#' from a gblinear model created while using the \code{cb.gblinear.history()}
+#' callback.
+#'
+#' @param model either an \code{xgb.Booster} or a result of \code{xgb.cv()}, trained
+#'        using the \code{cb.gblinear.history()} callback.
+#' @param class_index zero-based class index to extract the coefficients for only that
+#'        specific class in a multinomial multiclass model. When it is NULL, all the
+#'        coeffients are returned. Has no effect in non-multiclass models.
+#'
+#' @return 
+#' For an \code{xgb.train} result, a matrix (either dense or sparse) with the columns
+#' corresponding to iteration's coefficients (in the order as \code{xgb.dump()} would
+#' return) and the rows corresponding to boosting iterations.
+#'
+#' For an \code{xgb.cv} result, a list of such matrices is returned with the elements
+#' corresponding to CV folds.
+#'
+#' @export
+xgb.gblinear.history <- function(model, class_index = NULL) {
+
+  if (!(inherits(model, "xgb.Booster") ||
+        inherits(model, "xgb.cv.synchronous")))
+    stop("model must be an object of either xgb.Booster or xgb.cv.synchronous class")
+  is_cv <- inherits(model, "xgb.cv.synchronous")
+
+  if (is.null(model[["callbacks"]]) || is.null(model$callbacks[["cb.gblinear.history"]]))
+    stop("model must be trained while using the cb.gblinear.history() callback")
+
+  if (!is_cv) {
+    # extract num_class & num_feat from the internal model
+    dmp <- xgb.dump(model)
+    if(length(dmp) < 2 || dmp[2] != "bias:")
+      stop("It does not appear to be a gblinear model")
+    dmp <- dmp[-c(1,2)]
+    n <- which(dmp == 'weight:')
+    if(length(n) != 1)
+      stop("It does not appear to be a gblinear model")
+    num_class <- n - 1
+    num_feat <- (length(dmp) - 4) / num_class
+  } else {
+    # in case of CV, the object is expected to have this info
+    if (model$params$booster != "gblinear")
+      stop("It does not appear to be a gblinear model")
+    num_class <- NVL(model$params$num_class, 1)
+    num_feat <- model$nfeatures
+    if (is.null(num_feat))
+      stop("This xgb.cv result does not have nfeatures info")
+  }
+
+  if (!is.null(class_index) &&
+      num_class > 1 &&
+      (class_index[1] < 0 || class_index[1] >= num_class))
+    stop("class_index has to be within [0,", num_class - 1, "]")
+
+  coef_path <- environment(model$callbacks$cb.gblinear.history)[["coefs"]]
+  if (!is.null(class_index) && num_class > 1) {
+    coef_path <- if (is.list(coef_path)) {
+      lapply(coef_path, 
+             function(x) x[, seq(1 + class_index, by=num_class, length.out=num_feat)])
+    } else {
+      coef_path <- coef_path[, seq(1 + class_index, by=num_class, length.out=num_feat)]
+    }
+  }
+  coef_path
+}
+
+
+#
+# Internal utility functions for callbacks ------------------------------------
+# 
+
+# Format the evaluation metric string
+format.eval.string <- function(iter, eval_res, eval_err = NULL) {
+  if (length(eval_res) == 0)
+    stop('no evaluation results')
+  enames <- names(eval_res)
+  if (is.null(enames))
+    stop('evaluation results must have names')
+  iter <- sprintf('[%d]\t', iter)
+  if (!is.null(eval_err)) {
+    if (length(eval_res) != length(eval_err))
+      stop('eval_res & eval_err lengths mismatch')
+    res <- paste0(sprintf("%s:%f+%f", enames, eval_res, eval_err), collapse = '\t')
+  } else {
+    res <- paste0(sprintf("%s:%f", enames, eval_res), collapse = '\t')
+  }
+  return(paste0(iter, res))
+}
+
+# Extract callback names from the list of callbacks
+callback.names <- function(cb_list) {
+  unlist(lapply(cb_list, function(x) attr(x, 'name')))
+}
+
+# Extract callback calls from the list of callbacks
+callback.calls <- function(cb_list) {
+  unlist(lapply(cb_list, function(x) attr(x, 'call')))
+}
+
+# Add a callback cb to the list and make sure that 
+# cb.early.stop and cb.cv.predict are at the end of the list
+# with cb.cv.predict being the last (when present)
+add.cb <- function(cb_list, cb) {
+  cb_list <- c(cb_list, cb)
+  names(cb_list) <- callback.names(cb_list)
+  if ('cb.early.stop' %in% names(cb_list)) {
+    cb_list <- c(cb_list, cb_list['cb.early.stop'])
+    # this removes only the first one
+    cb_list['cb.early.stop'] <- NULL 
+  }
+  if ('cb.cv.predict' %in% names(cb_list)) {
+    cb_list <- c(cb_list, cb_list['cb.cv.predict'])
+    cb_list['cb.cv.predict'] <- NULL 
+  }
+  cb_list
+}
+
+# Sort callbacks list into categories
+categorize.callbacks <- function(cb_list) {
+  list(
+    pre_iter = Filter(function(x) {
+        pre <- attr(x, 'is_pre_iteration')
+        !is.null(pre) && pre 
+      }, cb_list),
+    post_iter = Filter(function(x) {
+        pre <- attr(x, 'is_pre_iteration')
+        is.null(pre) || !pre
+      }, cb_list),
+    finalize = Filter(function(x) {
+        'finalize' %in% names(formals(x))
+      }, cb_list)
+  )
+}
+
+# Check whether all callback functions with names given by 'query_names' are present in the 'cb_list'.
+has.callbacks <- function(cb_list, query_names) {
+  if (length(cb_list) < length(query_names))
+    return(FALSE)
+  if (!is.list(cb_list) ||
+      any(sapply(cb_list, class) != 'function')) {
+    stop('`cb_list` must be a list of callback functions')
+  }
+  cb_names <- callback.names(cb_list)
+  if (!is.character(cb_names) ||
+      length(cb_names) != length(cb_list) ||
+      any(cb_names == "")) {
+    stop('All callbacks in the `cb_list` must have a non-empty `name` attribute')
+  }
+  if (!is.character(query_names) ||
+      length(query_names) == 0 ||
+      any(query_names == "")) {
+    stop('query_names must be a non-empty vector of non-empty character names')
+  }
+  return(all(query_names %in% cb_names))
+}

R-package/R/getinfo.xgb.DMatrix.R

@@ -1,55 +0,0 @@
-setClass('xgb.DMatrix')
-
-#' Get information of an xgb.DMatrix object
-#' 
-#' Get information of an xgb.DMatrix object
-#' 
-#' The information can be one of the following:
-#' 
-#' \itemize{
-#'     \item \code{label}: label Xgboost learn from ;
-#'     \item \code{weight}: to do a weight rescale ;
-#'     \item \code{base_margin}: base margin is the base prediction Xgboost will boost from ;
-#'     \item \code{nrow}: number of rows of the \code{xgb.DMatrix}.
-#' }
-#' 
-#' @examples
-#' data(agaricus.train, package='xgboost')
-#' train <- agaricus.train
-#' dtrain <- xgb.DMatrix(train$data, label=train$label)
-#' labels <- getinfo(dtrain, 'label')
-#' setinfo(dtrain, 'label', 1-labels)
-#' labels2 <- getinfo(dtrain, 'label')
-#' stopifnot(all(labels2 == 1-labels))
-#' @rdname getinfo
-#' @export
-getinfo <- function(object, ...){
-    UseMethod("getinfo")
-}
-
-
-
-#' @param object Object of class \code{xgb.DMatrix}
-#' @param name the name of the field to get
-#' @param ... other parameters
-#' @rdname getinfo
-#' @method getinfo xgb.DMatrix
-setMethod("getinfo", signature = "xgb.DMatrix",
-          definition = function(object, name) {
-              if (typeof(name) != "character") {
-                  stop("xgb.getinfo: name must be character")
-              }
-              if (class(object) != "xgb.DMatrix") {
-                  stop("xgb.setinfo: first argument dtrain must be xgb.DMatrix")
-              }
-              if (name != "label" && name != "weight" &&
-                      name != "base_margin" && name != "nrow") {
-                  stop(paste("xgb.getinfo: unknown info name", name))
-              }
-              if (name != "nrow"){
-                  ret <- .Call("XGDMatrixGetInfo_R", object, name, PACKAGE = "xgboost")
-              } else {
-                  ret <- xgb.numrow(object)
-              }
-              return(ret)
-          })

R-package/R/nrow.xgb.DMatrix.R

@@ -1,19 +0,0 @@
-setGeneric("nrow")
-
-#' @title Number of xgb.DMatrix rows
-#' @description \code{nrow} return the number of rows present in the \code{xgb.DMatrix}.
-#' @param x Object of class \code{xgb.DMatrix}
-#' 
-#' @examples
-#' data(agaricus.train, package='xgboost')
-#' train <- agaricus.train
-#' dtrain <- xgb.DMatrix(train$data, label=train$label)
-#' stopifnot(nrow(dtrain) == nrow(train$data))
-#' 
-#' @export
-setMethod("nrow",
-          signature = "xgb.DMatrix",
-          definition = function(x) {
-            xgb.numrow(x)
-          }
-)

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

@@ -1,80 +0,0 @@
-setClass("xgb.Booster.handle")
-setClass("xgb.Booster",
-         slots = c(handle = "xgb.Booster.handle",
-                   raw = "raw"))
-
-#' Predict method for eXtreme Gradient Boosting model
-#' 
-#' Predicted values based on xgboost model object.
-#' 
-#' @param object Object of class "xgb.Boost"
-#' @param newdata takes \code{matrix}, \code{dgCMatrix}, local data file or 
-#'   \code{xgb.DMatrix}. 
-#' @param missing Missing is only used when input is dense matrix, pick a float 
-#'     value that represents missing value. Sometime a data use 0 or other extreme value to represents missing values.
-#' @param outputmargin whether the prediction should be shown in the original
-#'   value of sum of functions, when outputmargin=TRUE, the prediction is 
-#'   untransformed margin value. In logistic regression, outputmargin=T will
-#'   output value before logistic transformation.
-#' @param ntreelimit limit number of trees used in prediction, this parameter is
-#'  only valid for gbtree, but not for gblinear. set it to be value bigger 
-#'  than 0. It will use all trees by default.
-#' @param predleaf whether predict leaf index instead. If set to TRUE, the output will be a matrix object.
-#' 
-#' @details  
-#' The option \code{ntreelimit} purpose is to let the user train a model with lots 
-#' of trees but use only the first trees for prediction to avoid overfitting 
-#' (without having to train a new model with less trees).
-#' 
-#' The option \code{predleaf} purpose is inspired from §3.1 of the paper 
-#' \code{Practical Lessons from Predicting Clicks on Ads at Facebook}.
-#' The idea is to use the model as a generator of new features which capture non linear link 
-#' from original features.
-#' 
-#' @examples
-#' data(agaricus.train, package='xgboost')
-#' data(agaricus.test, package='xgboost')
-#' train <- agaricus.train
-#' test <- agaricus.test
-#' bst <- xgboost(data = train$data, label = train$label, max.depth = 2, 
-#'                eta = 1, nthread = 2, nround = 2,objective = "binary:logistic")
-#' pred <- predict(bst, test$data)
-#' @export
-setMethod("predict", signature = "xgb.Booster",
-          definition = function(object, newdata, missing = NA,
-                                outputmargin = FALSE, ntreelimit = NULL, predleaf = FALSE) {
-  if (class(object) != "xgb.Booster"){
-    stop("predict: model in prediction must be of class xgb.Booster")
-  } else {
-    object <- xgb.Booster.check(object, saveraw = FALSE)
-  }
-  if (class(newdata) != "xgb.DMatrix") {
-    newdata <- xgb.DMatrix(newdata, missing = missing)
-  }
-  if (is.null(ntreelimit)) {
-    ntreelimit <- 0
-  } else {
-    if (ntreelimit < 1){
-      stop("predict: ntreelimit must be equal to or greater than 1")
-    }
-  }
-  option <- 0
-  if (outputmargin) {
-    option <- option + 1
-  }
-  if (predleaf) {
-    option <- option + 2
-  }
-  ret <- .Call("XGBoosterPredict_R", object$handle, newdata, as.integer(option),
-               as.integer(ntreelimit), PACKAGE = "xgboost")
-  if (predleaf){
-      len <- getinfo(newdata, "nrow")
-      if (length(ret) == len){
-          ret <- matrix(ret,ncol = 1)
-      } else {
-          ret <- matrix(ret, ncol = len)
-          ret <- t(ret)
-      }
-  }
-  return(ret)
-})

R-package/R/predict.xgb.Booster.handle.R

@@ -1,18 +0,0 @@
-#' Predict method for eXtreme Gradient Boosting model handle
-#' 
-#' Predicted values based on xgb.Booster.handle object.
-#' 
-#' @param object Object of class "xgb.Boost.handle"
-#' @param ... Parameters pass to \code{predict.xgb.Booster}
-#' 
-setMethod("predict", signature = "xgb.Booster.handle",
-          definition = function(object, ...) {
-  if (class(object) != "xgb.Booster.handle"){
-    stop("predict: model in prediction must be of class xgb.Booster.handle")
-  }
-
-  bst <- xgb.handleToBooster(object)
-
-  ret <- predict(bst, ...)
-  return(ret)
-})

R-package/R/setinfo.xgb.DMatrix.R

@@ -1,37 +0,0 @@
-#' Set information of an xgb.DMatrix object
-#' 
-#' Set information of an xgb.DMatrix object
-#' 
-#' It can be one of the following:
-#' 
-#' \itemize{
-#'     \item \code{label}: label Xgboost learn from ;
-#'     \item \code{weight}: to do a weight rescale ;
-#'     \item \code{base_margin}: base margin is the base prediction Xgboost will boost from ;
-#'     \item \code{group}.
-#' }
-#' 
-#' @examples
-#' data(agaricus.train, package='xgboost')
-#' train <- agaricus.train
-#' dtrain <- xgb.DMatrix(train$data, label=train$label)
-#' labels <- getinfo(dtrain, 'label')
-#' setinfo(dtrain, 'label', 1-labels)
-#' labels2 <- getinfo(dtrain, 'label')
-#' stopifnot(all(labels2 == 1-labels))
-#' @rdname setinfo
-#' @export
-setinfo <- function(object, ...){
-  UseMethod("setinfo")
-}
-
-#' @param object Object of class "xgb.DMatrix"
-#' @param name the name of the field to get
-#' @param info the specific field of information to set
-#' @param ... other parameters
-#' @rdname setinfo
-#' @method setinfo xgb.DMatrix
-setMethod("setinfo", signature = "xgb.DMatrix",
-          definition = function(object, name, info) {
-            xgb.setinfo(object, name, info)
-          })

R-package/R/slice.xgb.DMatrix.R

@@ -1,44 +0,0 @@
-setClass('xgb.DMatrix')
-
-#' Get a new DMatrix containing the specified rows of
-#' orginal xgb.DMatrix object
-#'
-#' Get a new DMatrix containing the specified rows of
-#' orginal xgb.DMatrix object
-#' 
-#' @examples
-#' data(agaricus.train, package='xgboost')
-#' train <- agaricus.train
-#' dtrain <- xgb.DMatrix(train$data, label=train$label)
-#' dsub <- slice(dtrain, 1:3)
-#' @rdname slice
-#' @export
-slice <- function(object, ...){
-    UseMethod("slice")
-}
-
-#' @param object Object of class "xgb.DMatrix"
-#' @param idxset a integer vector of indices of rows needed
-#' @param ... other parameters
-#' @rdname slice
-#' @method slice xgb.DMatrix
-setMethod("slice", signature = "xgb.DMatrix",
-          definition = function(object, idxset, ...) {
-              if (class(object) != "xgb.DMatrix") {
-                  stop("slice: first argument dtrain must be xgb.DMatrix")
-              }
-              ret <- .Call("XGDMatrixSliceDMatrix_R", object, idxset,
-                           PACKAGE = "xgboost")
-
-              attr_list <- attributes(object)
-              nr <- xgb.numrow(object)
-              len <- sapply(attr_list,length)
-              ind <- which(len == nr)
-              if (length(ind) > 0) {
-                  nms <- names(attr_list)[ind]
-                  for (i in 1:length(ind)) {
-                    attr(ret,nms[i]) <- attr(object,nms[i])[idxset]
-                  }
-              }
-              return(structure(ret, class = "xgb.DMatrix"))
-          })

R-package/R/utils.R

@@ -1,305 +1,231 @@
-  #' @importClassesFrom Matrix dgCMatrix dgeMatrix
-#' @import methods
+#
+# This file is for the low level reuseable utility functions
+# that are not supposed to be visibe to a user.
+#
 
-# depends on matrix
-.onLoad <- function(libname, pkgname) {
-  library.dynam("xgboost", pkgname, libname)
-}
-.onUnload <- function(libpath) {
-  library.dynam.unload("xgboost", libpath)
+#
+# General helper utilities ----------------------------------------------------
+#
+
+# SQL-style NVL shortcut.
+NVL <- function(x, val) {
+  if (is.null(x))
+    return(val)
+  if (is.vector(x)) {
+    x[is.na(x)] <- val
+    return(x)
+  }
+  if (typeof(x) == 'closure')
+    return(x)
+  stop("typeof(x) == ", typeof(x), " is not supported by NVL")
 }
 
-# set information into dmatrix, this mutate dmatrix
-xgb.setinfo <- function(dmat, name, info) {
-  if (class(dmat) != "xgb.DMatrix") {
-    stop("xgb.setinfo: first argument dtrain must be xgb.DMatrix")
+
+#
+# Low-level functions for boosting --------------------------------------------
+#
+
+# Merges booster params with whatever is provided in ...
+# plus runs some checks
+check.booster.params <- function(params, ...) {
+  if (typeof(params) != "list") 
+    stop("params must be a list")
+  
+  # in R interface, allow for '.' instead of '_' in parameter names
+  names(params) <- gsub("\\.", "_", names(params))
+  
+  # merge parameters from the params and the dots-expansion
+  dot_params <- list(...)
+  names(dot_params) <- gsub("\\.", "_", names(dot_params))
+  if (length(intersect(names(params),
+                       names(dot_params))) > 0)
+    stop("Same parameters in 'params' and in the call are not allowed. Please check your 'params' list.")
+  params <- c(params, dot_params)
+  
+  # providing a parameter multiple times makes sense only for 'eval_metric'
+  name_freqs <- table(names(params))
+  multi_names <- setdiff(names(name_freqs[name_freqs > 1]), 'eval_metric')
+  if (length(multi_names) > 0) {
+    warning("The following parameters were provided multiple times:\n\t",
+            paste(multi_names, collapse = ', '), "\n  Only the last value for each of them will be used.\n")
+    # While xgboost internals would choose the last value for a multiple-times parameter, 
+    # enforce it here in R as well (b/c multi-parameters might be used further in R code, 
+    # and R takes the 1st value when multiple elements with the same name are present in a list).
+    for (n in multi_names) {
+      del_idx <- which(n == names(params))
+      del_idx <- del_idx[-length(del_idx)]
+      params[[del_idx]] <- NULL
+    }
   }
-  if (name == "label") {
-    if (length(info) != xgb.numrow(dmat))
-      stop("The length of labels must equal to the number of rows in the input data")
-    .Call("XGDMatrixSetInfo_R", dmat, name, as.numeric(info),
-          PACKAGE = "xgboost")
-    return(TRUE)
+  
+  # for multiclass, expect num_class to be set
+  if (typeof(params[['objective']]) == "character" &&
+      substr(NVL(params[['objective']], 'x'), 1, 6) == 'multi:' &&
+      as.numeric(NVL(params[['num_class']], 0)) < 2) {
+        stop("'num_class' > 1 parameter must be set for multiclass classification")
   }
-  if (name == "weight") {
-    if (length(info) != xgb.numrow(dmat))
-      stop("The length of weights must equal to the number of rows in the input data")
-    .Call("XGDMatrixSetInfo_R", dmat, name, as.numeric(info),
-          PACKAGE = "xgboost")
-    return(TRUE)
+  
+  # monotone_constraints parser
+  
+  if (!is.null(params[['monotone_constraints']]) &&
+      typeof(params[['monotone_constraints']]) != "character") {
+        vec2str = paste(params[['monotone_constraints']], collapse = ',')
+        vec2str = paste0('(', vec2str, ')')
+        params[['monotone_constraints']] = vec2str
   }
-  if (name == "base_margin") {
-    # if (length(info)!=xgb.numrow(dmat))
-    #   stop("The length of base margin must equal to the number of rows in the input data")
-    .Call("XGDMatrixSetInfo_R", dmat, name, as.numeric(info),
-          PACKAGE = "xgboost")
-    return(TRUE)
-  }
-  if (name == "group") {
-    if (sum(info) != xgb.numrow(dmat))
-      stop("The sum of groups must equal to the number of rows in the input data")
-    .Call("XGDMatrixSetInfo_R", dmat, name, as.integer(info),
-          PACKAGE = "xgboost")
-    return(TRUE)
-  }
-  stop(paste("xgb.setinfo: unknown info name", name))
-  return(FALSE)
+  
+  return(params)
 }
 
-# construct a Booster from cachelist
-xgb.Booster <- function(params = list(), cachelist = list(), modelfile = NULL) {
-  if (typeof(cachelist) != "list") {
-    stop("xgb.Booster: only accepts list of DMatrix as cachelist")
+
+# Performs some checks related to custom objective function.
+# WARNING: has side-effects and can modify 'params' and 'obj' in its calling frame
+check.custom.obj <- function(env = parent.frame()) {
+  if (!is.null(env$params[['objective']]) && !is.null(env$obj))
+    stop("Setting objectives in 'params' and 'obj' at the same time is not allowed")
+  
+  if (!is.null(env$obj) && typeof(env$obj) != 'closure')
+    stop("'obj' must be a function")
+  
+  # handle the case when custom objective function was provided through params
+  if (!is.null(env$params[['objective']]) &&
+      typeof(env$params$objective) == 'closure') {
+    env$obj <- env$params$objective
+    env$params$objective <- NULL
   }
-  for (dm in cachelist) {
-    if (class(dm) != "xgb.DMatrix") {
-      stop("xgb.Booster: only accepts list of DMatrix as cachelist")
-    }
-  }
-  handle <- .Call("XGBoosterCreate_R", cachelist, PACKAGE = "xgboost")
-  if (length(params) != 0) {
-    for (i in 1:length(params)) {
-      p <- params[i]
-      .Call("XGBoosterSetParam_R", handle, gsub("\\.", "_", names(p)), as.character(p),
-            PACKAGE = "xgboost")
-    }
-  }
-  if (!is.null(modelfile)) {
-    if (typeof(modelfile) == "character") {
-      .Call("XGBoosterLoadModel_R", handle, modelfile, PACKAGE = "xgboost")
-    } else if (typeof(modelfile) == "raw") {
-      .Call("XGBoosterLoadModelFromRaw_R", handle, modelfile, PACKAGE = "xgboost")
-    } else {
-      stop("xgb.Booster: modelfile must be character or raw vector")
-    }
-  }
-  return(structure(handle, class = "xgb.Booster.handle"))
 }
 
-# convert xgb.Booster.handle to xgb.Booster
-xgb.handleToBooster <- function(handle, raw = NULL)
-{
-  bst <- list(handle = handle, raw = raw)
-  class(bst) <- "xgb.Booster"
-  return(bst)
+# Performs some checks related to custom evaluation function.
+# WARNING: has side-effects and can modify 'params' and 'feval' in its calling frame
+check.custom.eval <- function(env = parent.frame()) {
+  if (!is.null(env$params[['eval_metric']]) && !is.null(env$feval))
+    stop("Setting evaluation metrics in 'params' and 'feval' at the same time is not allowed")
+  
+  if (!is.null(env$feval) && typeof(env$feval) != 'closure')
+    stop("'feval' must be a function")
+  
+  # handle a situation when custom eval function was provided through params
+  if (!is.null(env$params[['eval_metric']]) &&
+      typeof(env$params$eval_metric) == 'closure') {
+    env$feval <- env$params$eval_metric
+    env$params$eval_metric <- NULL
+  }
+  
+  # require maximize to be set when custom feval and early stopping are used together
+  if (!is.null(env$feval) &&
+      is.null(env$maximize) && (
+        !is.null(env$early_stopping_rounds) || 
+        has.callbacks(env$callbacks, 'cb.early.stop')))
+    stop("Please set 'maximize' to indicate whether the evaluation metric needs to be maximized or not")
 }
 
-# Check whether an xgb.Booster object is complete
-xgb.Booster.check <- function(bst, saveraw = TRUE)
-{
-  isnull <- is.null(bst$handle)
-  if (!isnull) {
-    isnull <- .Call("XGCheckNullPtr_R", bst$handle, PACKAGE="xgboost")
-  }
-  if (isnull) {
-    bst$handle <- xgb.Booster(modelfile = bst$raw)
-  } else {
-    if (is.null(bst$raw) && saveraw)
-      bst$raw <- xgb.save.raw(bst$handle)
-  }
-  return(bst)
-}
 
-## ----the following are low level iteratively function, not needed if
-## you do not want to use them ---------------------------------------
-# get dmatrix from data, label
-xgb.get.DMatrix <- function(data, label = NULL, missing = NA, weight = NULL) {
-  inClass <- class(data)
-  if (inClass == "dgCMatrix" || inClass == "matrix") {
-    if (is.null(label)) {
-      stop("xgboost: need label when data is a matrix")
-    }
-    dtrain <- xgb.DMatrix(data, label = label, missing = missing)
-    if (!is.null(weight)){
-      xgb.setinfo(dtrain, "weight", weight)
-    }
-  } else {
-    if (!is.null(label)) {
-      warning("xgboost: label will be ignored.")
-    }
-    if (inClass == "character") {
-      dtrain <- xgb.DMatrix(data)
-    } else if (inClass == "xgb.DMatrix") {
-      dtrain <- data
-    } else if (inClass == "data.frame") {
-      stop("xgboost only support numerical matrix input,
-           use 'data.matrix' to transform the data.")
-    } else {
-      stop("xgboost: Invalid input of data")
-    }
+# Update a booster handle for an iteration with dtrain data
+xgb.iter.update <- function(booster_handle, dtrain, iter, obj = NULL) {
+  if (!identical(class(booster_handle), "xgb.Booster.handle")) {
+    stop("booster_handle must be of xgb.Booster.handle class")
   }
-  return (dtrain)
-}
-xgb.numrow <- function(dmat) {
-  nrow <- .Call("XGDMatrixNumRow_R", dmat, PACKAGE="xgboost")
-  return(nrow)
-}
-# iteratively update booster with customized statistics
-xgb.iter.boost <- function(booster, dtrain, gpair) {
-  if (class(booster) != "xgb.Booster.handle") {
-    stop("xgb.iter.update: first argument must be type xgb.Booster.handle")
-  }
-  if (class(dtrain) != "xgb.DMatrix") {
-    stop("xgb.iter.update: second argument must be type xgb.DMatrix")
-  }
-  .Call("XGBoosterBoostOneIter_R", booster, dtrain, gpair$grad, gpair$hess, PACKAGE = "xgboost")
-  return(TRUE)
-}
-
-# iteratively update booster with dtrain
-xgb.iter.update <- function(booster, dtrain, iter, obj = NULL) {
-  if (class(booster) != "xgb.Booster.handle") {
-    stop("xgb.iter.update: first argument must be type xgb.Booster.handle")
-  }
-  if (class(dtrain) != "xgb.DMatrix") {
-    stop("xgb.iter.update: second argument must be type xgb.DMatrix")
+  if (!inherits(dtrain, "xgb.DMatrix")) {
+    stop("dtrain must be of xgb.DMatrix class")
   }
 
   if (is.null(obj)) {
-    .Call("XGBoosterUpdateOneIter_R", booster, as.integer(iter), dtrain,
-          PACKAGE = "xgboost")
-    } else {
-    pred <- predict(booster, dtrain)
+    .Call(XGBoosterUpdateOneIter_R, booster_handle, as.integer(iter), dtrain)
+  } else {
+    pred <- predict(booster_handle, dtrain)
     gpair <- obj(pred, dtrain)
-    succ <- xgb.iter.boost(booster, dtrain, gpair)
+    .Call(XGBoosterBoostOneIter_R, booster_handle, dtrain, gpair$grad, gpair$hess)
   }
   return(TRUE)
 }
 
-# iteratively evaluate one iteration
-xgb.iter.eval <- function(booster, watchlist, iter, feval = NULL, prediction = FALSE) {
-  if (class(booster) != "xgb.Booster.handle") {
-    stop("xgb.eval: first argument must be type xgb.Booster")
-  }
-  if (typeof(watchlist) != "list") {
-    stop("xgb.eval: only accepts list of DMatrix as watchlist")
-  }
-  for (w in watchlist) {
-    if (class(w) != "xgb.DMatrix") {
-      stop("xgb.eval: watch list can only contain xgb.DMatrix")
-    }
-  }
-  if (length(watchlist) != 0) {
-    if (is.null(feval)) {
-      evnames <- list()
-      for (i in 1:length(watchlist)) {
-        w <- watchlist[i]
-        if (length(names(w)) == 0) {
-          stop("xgb.eval: name tag must be presented for every elements in watchlist")
-        }
-        evnames <- append(evnames, names(w))
-      }
-      msg <- .Call("XGBoosterEvalOneIter_R", booster, as.integer(iter), watchlist,
-                   evnames, PACKAGE = "xgboost")
-    } else {
-      msg <- paste("[", iter, "]", sep="")
-      for (j in 1:length(watchlist)) {
-        w <- watchlist[j]
-        if (length(names(w)) == 0) {
-          stop("xgb.eval: name tag must be presented for every elements in watchlist")
-        }
-        preds <- predict(booster, w[[1]])
-        ret <- feval(preds, w[[1]])
-        msg <- paste(msg, "\t", names(w), "-", ret$metric, ":", ret$value, sep="")
-      }
-    }
+
+# Evaluate one iteration.
+# Returns a named vector of evaluation metrics 
+# with the names in a 'datasetname-metricname' format.
+xgb.iter.eval <- function(booster_handle, watchlist, iter, feval = NULL) {
+  if (!identical(class(booster_handle), "xgb.Booster.handle"))
+    stop("class of booster_handle must be xgb.Booster.handle")
+
+  if (length(watchlist) == 0) 
+    return(NULL)
+  
+  evnames <- names(watchlist)
+  if (is.null(feval)) {
+    msg <- .Call(XGBoosterEvalOneIter_R, booster_handle, as.integer(iter), watchlist, as.list(evnames))
+    msg <- stri_split_regex(msg, '(\\s+|:|\\s+)')[[1]][-1]
+    res <- as.numeric(msg[c(FALSE,TRUE)]) # even indices are the values
+    names(res) <- msg[c(TRUE,FALSE)]      # odds are the names
   } else {
-    msg <- ""
+    res <- sapply(seq_along(watchlist), function(j) {
+      w <- watchlist[[j]]
+      preds <- predict(booster_handle, w) # predict using all trees
+      eval_res <- feval(preds, w)
+      out <- eval_res$value
+      names(out) <- paste0(evnames[j], "-", eval_res$metric)
+      out
+    })
   }
-  if (prediction){
-    preds <- predict(booster,watchlist[[2]])
-    return(list(msg,preds))
-  }
-  return(msg)
+  return(res)
 }
 
-#------------------------------------------
-# helper functions for cross validation
+
 #
-xgb.cv.mknfold <- function(dall, nfold, param, stratified, folds) {
-  if (nfold <= 1) {
-    stop("nfold must be bigger than 1")
+# Helper functions for cross validation ---------------------------------------
+#
+
+# Generates random (stratified if needed) CV folds
+generate.cv.folds <- function(nfold, nrows, stratified, label, params) {
+  
+  # cannot do it for rank
+  if (exists('objective', where = params) &&
+      is.character(params$objective) &&
+      strtrim(params$objective, 5) == 'rank:') {
+    stop("\n\tAutomatic generation of CV-folds is not implemented for ranking!\n",
+         "\tConsider providing pre-computed CV-folds through the 'folds=' parameter.\n")
   }
-  if(is.null(folds)) {
-    if (exists('objective', where=param) && is.character(param$objective) &&
-        strtrim(param[['objective']], 5) == 'rank:') {
-      stop("\tAutomatic creation of CV-folds is not implemented for ranking!\n",
-           "\tConsider providing pre-computed CV-folds through the folds parameter.")
-    }
-    y <- getinfo(dall, 'label')
-    randidx <- sample(1 : xgb.numrow(dall))
-    if (stratified & length(y) == length(randidx)) {
-      y <- y[randidx]
-      #
-      # WARNING: some heuristic logic is employed to identify classification setting!
-      #
-      # For classification, need to convert y labels to factor before making the folds,
-      # and then do stratification by factor levels.
-      # For regression, leave y numeric and do stratification by quantiles.
-      if (exists('objective', where=param) && is.character(param$objective)) {
-        # If 'objective' provided in params, assume that y is a classification label
-        # unless objective is reg:linear
-        if (param[['objective']] != 'reg:linear') y <- factor(y)
-      } else {
-        # If no 'objective' given in params, it means that user either wants to use
-        # the default 'reg:linear' objective or has provided a custom obj function.
-        # Here, assume classification setting when y has 5 or less unique values:
-        if (length(unique(y)) <= 5) y <- factor(y)
-      }
-      folds <- xgb.createFolds(y, nfold)
+  # shuffle
+  rnd_idx <- sample.int(nrows)
+  if (stratified &&
+      length(label) == length(rnd_idx)) {
+    y <- label[rnd_idx]
+    # WARNING: some heuristic logic is employed to identify classification setting!
+    #  - For classification, need to convert y labels to factor before making the folds,
+    #    and then do stratification by factor levels.
+    #  - For regression, leave y numeric and do stratification by quantiles.
+    if (exists('objective', where = params) &&
+        is.character(params$objective)) {
+      # If 'objective' provided in params, assume that y is a classification label
+      # unless objective is reg:linear
+      if (params$objective != 'reg:linear')
+        y <- factor(y)
     } else {
-      # make simple non-stratified folds
-      kstep <- length(randidx) %/% nfold
-      folds <- list()
-      for (i in 1:(nfold - 1)) {
-        folds[[i]] <- randidx[1:kstep]
-        randidx <- setdiff(randidx, folds[[i]])
-      }
-      folds[[nfold]] <- randidx
+      # If no 'objective' given in params, it means that user either wants to use
+      # the default 'reg:linear' objective or has provided a custom obj function.
+      # Here, assume classification setting when y has 5 or less unique values:
+      if (length(unique(y)) <= 5)
+        y <- factor(y)
     }
-  }
-  ret <- list()
-  for (k in 1:nfold) {
-    dtest <- slice(dall, folds[[k]])
-    didx <- c()
-    for (i in 1:nfold) {
-      if (i != k) {
-        didx <- append(didx, folds[[i]])
-      }
+    folds <- xgb.createFolds(y, nfold)
+  } else {
+    # make simple non-stratified folds
+    kstep <- length(rnd_idx) %/% nfold
+    folds <- list()
+    for (i in seq_len(nfold - 1)) {
+      folds[[i]] <- rnd_idx[seq_len(kstep)]
+      rnd_idx <- rnd_idx[-seq_len(kstep)]
     }
-    dtrain <- slice(dall, didx)
-    bst <- xgb.Booster(param, list(dtrain, dtest))
-    watchlist <- list(train=dtrain, test=dtest)
-    ret[[k]] <- list(dtrain=dtrain, booster=bst, watchlist=watchlist, index=folds[[k]])
+    folds[[nfold]] <- rnd_idx
   }
-  return (ret)
+  return(folds)
 }
 
-xgb.cv.aggcv <- function(res, showsd = TRUE) {
-  header <- res[[1]]
-  ret <- header[1]
-  for (i in 2:length(header)) {
-    kv <- strsplit(header[i], ":")[[1]]
-    ret <- paste(ret, "\t", kv[1], ":", sep="")
-    stats <- c()
-    stats[1] <- as.numeric(kv[2])
-    for (j in 2:length(res)) {
-      tkv <- strsplit(res[[j]][i], ":")[[1]]
-      stats[j] <- as.numeric(tkv[2])
-    }
-    ret <- paste(ret, sprintf("%f", mean(stats)), sep="")
-    if (showsd) {
-      ret <- paste(ret, sprintf("+%f", stats::sd(stats)), sep="")
-    }
-  }
-  return (ret)
-}
-
-# Shamelessly copied from caret::createFolds
-# and simplified by always returning an unnamed list of test indices
+# Creates CV folds stratified by the values of y.
+# It was borrowed from caret::createFolds and simplified
+# by always returning an unnamed list of fold indices.
 xgb.createFolds <- function(y, k = 10)
 {
-  if(is.numeric(y)) {
+  if (is.numeric(y)) {
     ## Group the numeric data based on their magnitudes
     ## and sample within those groups.
 
@@ -318,7 +244,7 @@ xgb.createFolds <- function(y, k = 10)
              include.lowest = TRUE)
   }
 
-  if(k < length(y)) {
+  if (k < length(y)) {
     ## reset levels so that the possible levels and
     ## the levels in the vector are the same
     y <- factor(as.character(y))
@@ -328,19 +254,84 @@ xgb.createFolds <- function(y, k = 10)
     ## For each class, balance the fold allocation as far
     ## as possible, then resample the remainder.
     ## The final assignment of folds is also randomized.
-    for(i in 1:length(numInClass)) {
+    for (i in seq_along(numInClass)) {
       ## create a vector of integers from 1:k as many times as possible without
       ## going over the number of samples in the class. Note that if the number
       ## of samples in a class is less than k, nothing is producd here.
-      seqVector <- rep(1:k, numInClass[i] %/% k)
+      seqVector <- rep(seq_len(k), numInClass[i] %/% k)
       ## add enough random integers to get  length(seqVector) == numInClass[i]
-      if(numInClass[i] %% k > 0) seqVector <- c(seqVector, sample(1:k, numInClass[i] %% k))
+      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
-      foldVector[which(y == dimnames(numInClass)$y[i])] <- sample(seqVector)
+      foldVector[y == dimnames(numInClass)$y[i]] <- sample(seqVector)
     }
-  } else foldVector <- seq(along = y)
+  } else {
+    foldVector <- seq(along = y)
+  }
 
   out <- split(seq(along = y), foldVector)
   names(out) <- NULL
   out
 }
+
+
+#
+# Deprectaion notice utilities ------------------------------------------------
+#
+
+#' Deprecation notices.
+#' 
+#' At this time, some of the parameter names were changed in order to make the code style more uniform.
+#' The deprecated parameters would be removed in the next release.
+#' 
+#' To see all the current deprecated and new parameters, check the \code{xgboost:::depr_par_lut} table.
+#' 
+#' A deprecation warning is shown when any of the deprecated parameters is used in a call. 
+#' An additional warning is shown when there was a partial match to a deprecated parameter 
+#' (as R is able to partially match parameter names).
+#' 
+#' @name xgboost-deprecated
+NULL
+
+# Lookup table for the deprecated parameters bookkeeping
+depr_par_lut <- matrix(c(
+  'print.every.n', 'print_every_n', 
+  'early.stop.round', 'early_stopping_rounds',
+  'training.data', 'data',
+  'with.stats', 'with_stats',
+  'numberOfClusters', 'n_clusters',
+  'features.keep', 'features_keep',
+  'plot.height','plot_height',
+  'plot.width','plot_width',
+  'n_first_tree', 'trees',
+  'dummy', 'DUMMY'
+), ncol = 2, byrow = TRUE)
+colnames(depr_par_lut) <- c('old', 'new')
+
+# Checks the dot-parameters for deprecated names
+# (including partial matching), gives a deprecation warning,
+# and sets new parameters to the old parameters' values within its parent frame.
+# WARNING: has side-effects
+check.deprecation <- function(..., env = parent.frame()) {
+  pars <- list(...)
+  # exact and partial matches
+  all_match <- pmatch(names(pars), depr_par_lut[,1])
+  # indices of matched pars' names
+  idx_pars <- which(!is.na(all_match))
+  if (length(idx_pars) == 0) return()
+  # indices of matched LUT rows
+  idx_lut <- all_match[idx_pars]
+  # which of idx_lut were the exact matches?
+  ex_match <- depr_par_lut[idx_lut,1] %in% names(pars)
+  for (i in seq_along(idx_pars)) {
+    pars_par <- names(pars)[idx_pars[i]]
+    old_par <- depr_par_lut[idx_lut[i], 1]
+    new_par <- depr_par_lut[idx_lut[i], 2]
+    if (!ex_match[i]) {
+      warning("'", pars_par, "' was partially matched to '", old_par,"'")
+    }
+    .Deprecated(new_par, old = old_par, package = 'xgboost')
+    if (new_par != 'NULL') {
+      eval(parse(text = paste(new_par, '<-', pars[[pars_par]])), envir = env)
+    }
+  }
+}

R-package/R/xgb.Booster.R

@@ -0,0 +1,612 @@
+# Construct an internal xgboost Booster and return a handle to it.
+# internal utility function
+xgb.Booster.handle <- function(params = list(), cachelist = list(), modelfile = NULL) {
+  if (typeof(cachelist) != "list" ||
+      !all(vapply(cachelist, inherits, logical(1), what = 'xgb.DMatrix'))) {
+    stop("cachelist must be a list of xgb.DMatrix objects")
+  }
+
+  handle <- .Call(XGBoosterCreate_R, cachelist)
+  if (!is.null(modelfile)) {
+    if (typeof(modelfile) == "character") {
+      .Call(XGBoosterLoadModel_R, handle, modelfile[1])
+    } else if (typeof(modelfile) == "raw") {
+      .Call(XGBoosterLoadModelFromRaw_R, handle, modelfile)
+    } else if (inherits(modelfile, "xgb.Booster")) {
+      bst <- xgb.Booster.complete(modelfile, saveraw = TRUE)
+      .Call(XGBoosterLoadModelFromRaw_R, handle, bst$raw)
+    } else {
+      stop("modelfile must be either character filename, or raw booster dump, or xgb.Booster object")
+    }
+  }
+  class(handle) <- "xgb.Booster.handle"
+  if (length(params) > 0) {
+    xgb.parameters(handle) <- params
+  }
+  return(handle)
+}
+
+# Convert xgb.Booster.handle to xgb.Booster
+# internal utility function
+xgb.handleToBooster <- function(handle, raw = NULL) {
+  bst <- list(handle = handle, raw = raw)
+  class(bst) <- "xgb.Booster"
+  return(bst)
+}
+
+# Check whether xgb.Booster.handle is null
+# internal utility function
+is.null.handle <- function(handle) {
+  if (is.null(handle)) return(TRUE)
+
+  if (!identical(class(handle), "xgb.Booster.handle"))
+    stop("argument type must be xgb.Booster.handle")
+
+  if (.Call(XGCheckNullPtr_R, handle))
+    return(TRUE)
+
+  return(FALSE)
+}
+
+# Return a verified to be valid handle out of either xgb.Booster.handle or xgb.Booster
+# internal utility function
+xgb.get.handle <- function(object) {
+  handle <- switch(class(object)[1],
+    xgb.Booster = object$handle,
+    xgb.Booster.handle = object,
+    stop("argument must be of either xgb.Booster or xgb.Booster.handle class")
+  )
+  if (is.null.handle(handle)) {
+    stop("invalid xgb.Booster.handle")
+  }
+  handle
+}
+
+#' Restore missing parts of an incomplete xgb.Booster object.
+#'
+#' It attempts to complete an \code{xgb.Booster} object by restoring either its missing
+#' raw model memory dump (when it has no \code{raw} data but its \code{xgb.Booster.handle} is valid)
+#' or its missing internal handle (when its \code{xgb.Booster.handle} is not valid
+#' but it has a raw Booster memory dump).
+#'
+#' @param object object of class \code{xgb.Booster}
+#' @param saveraw a flag indicating whether to append \code{raw} Booster memory dump data
+#'                when it doesn't already exist.
+#'
+#' @details
+#'
+#' While this method is primarily for internal use, it might be useful in some practical situations.
+#'
+#' E.g., when an \code{xgb.Booster} model is saved as an R object and then is loaded as an R object,
+#' its handle (pointer) to an internal xgboost model would be invalid. The majority of xgboost methods
+#' should still work for such a model object since those methods would be using
+#' \code{xgb.Booster.complete} internally. However, one might find it to be more efficient to call the
+#' \code{xgb.Booster.complete} function explicitely once after loading a model as an R-object.
+#' That would prevent further repeated implicit reconstruction of an internal booster model.
+#'
+#' @return
+#' An object of \code{xgb.Booster} class.
+#'
+#' @examples
+#'
+#' data(agaricus.train, package='xgboost')
+#' bst <- xgboost(data = agaricus.train$data, label = agaricus.train$label, max_depth = 2,
+#'                eta = 1, nthread = 2, nrounds = 2, objective = "binary:logistic")
+#' saveRDS(bst, "xgb.model.rds")
+#'
+#' bst1 <- readRDS("xgb.model.rds")
+#' # the handle is invalid:
+#' print(bst1$handle)
+#'
+#' bst1 <- xgb.Booster.complete(bst1)
+#' # now the handle points to a valid internal booster model:
+#' print(bst1$handle)
+#'
+#' @export
+xgb.Booster.complete <- function(object, saveraw = TRUE) {
+  if (!inherits(object, "xgb.Booster"))
+    stop("argument type must be xgb.Booster")
+
+  if (is.null.handle(object$handle)) {
+    object$handle <- xgb.Booster.handle(modelfile = object$raw)
+  } else {
+    if (is.null(object$raw) && saveraw)
+      object$raw <- xgb.save.raw(object$handle)
+  }
+  return(object)
+}
+
+#' Predict method for eXtreme Gradient Boosting model
+#'
+#' Predicted values based on either xgboost model or model handle object.
+#'
+#' @param object Object of class \code{xgb.Booster} or \code{xgb.Booster.handle}
+#' @param newdata takes \code{matrix}, \code{dgCMatrix}, local data file or \code{xgb.DMatrix}.
+#' @param missing Missing is only used when input is dense matrix. Pick a float value that represents
+#'        missing values in data (e.g., sometimes 0 or some other extreme value is used).
+#' @param outputmargin whether the prediction should be returned in the for of original untransformed
+#'        sum of predictions from boosting iterations' results. E.g., setting \code{outputmargin=TRUE} for
+#'        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 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 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 \code{predleaf = TRUE}.
+#' @param ... Parameters passed to \code{predict.xgb.Booster}
+#'
+#' @details
+#' Note that \code{ntreelimit} is not necessarily equal to the number of boosting iterations
+#' and it is not necessarily equal to the number of trees in a model.
+#' E.g., in a random forest-like model, \code{ntreelimit} would limit the number of trees.
+#' But for multiclass classification, while there are multiple trees per iteration,
+#' \code{ntreelimit} limits the number of boosting iterations.
+#'
+#' Also note that \code{ntreelimit} would currently do nothing for predictions from gblinear,
+#' since gblinear doesn't keep its boosting history.
+#'
+#' One possible practical applications of the \code{predleaf} option is to use the model
+#' as a generator of new features which capture non-linearity and interactions,
+#' e.g., as implemented in \code{\link{xgb.create.features}}.
+#'
+#' Setting \code{predcontrib = TRUE} allows to calculate contributions of each feature to
+#' individual predictions. For "gblinear" booster, feature contributions are simply linear terms
+#' (feature_beta * feature_value). For "gbtree" booster, feature contributions are SHAP
+#' 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/}.
+#'
+#' @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
+#' a \code{(nrows(newdata), num_class)} dimension matrix is returned, depending on
+#' the \code{reshape} value.
+#'
+#' When \code{predleaf = TRUE}, the output is a matrix object with the
+#' number of columns corresponding to the number of trees.
+#'
+#' When \code{predcontrib = TRUE} and it is not a multiclass setting, the output is a matrix object with
+#' \code{num_features + 1} columns. The last "+ 1" column in a matrix corresponds to bias.
+#' 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).
+#'
+#' @seealso
+#' \code{\link{xgb.train}}.
+#' 
+#' @references
+#'
+#' Scott M. Lundberg, Su-In Lee, "A Unified Approach to Interpreting Model Predictions", NIPS Proceedings 2017, \url{https://arxiv.org/abs/1705.07874}
+#'
+#' Scott M. Lundberg, Su-In Lee, "Consistent feature attribution for tree ensembles", \url{https://arxiv.org/abs/1706.06060}
+#'
+#' @examples
+#' ## binary classification:
+#'
+#' data(agaricus.train, package='xgboost')
+#' data(agaricus.test, package='xgboost')
+#' train <- agaricus.train
+#' test <- agaricus.test
+#'
+#' bst <- xgboost(data = train$data, label = train$label, max_depth = 2,
+#'                eta = 0.5, nthread = 2, nrounds = 5, objective = "binary:logistic")
+#' # use all trees by default
+#' pred <- predict(bst, test$data)
+#' # use only the 1st tree
+#' pred1 <- predict(bst, test$data, ntreelimit = 1)
+#'
+#' # Predicting tree leafs:
+#' # the result is an nsamples X ntrees matrix
+#' pred_leaf <- predict(bst, test$data, predleaf = TRUE)
+#' str(pred_leaf)
+#'
+#' # Predicting feature contributions to predictions:
+#' # the result is an nsamples X (nfeatures + 1) matrix
+#' pred_contr <- predict(bst, test$data, predcontrib = TRUE)
+#' str(pred_contr)
+#' # verify that contributions' sums are equal to log-odds of predictions (up to float precision):
+#' summary(rowSums(pred_contr) - qlogis(pred))
+#' # for the 1st record, let's inspect its features that had non-zero contribution to prediction:
+#' contr1 <- pred_contr[1,]
+#' contr1 <- contr1[-length(contr1)]    # drop BIAS
+#' contr1 <- contr1[contr1 != 0]        # drop non-contributing features
+#' contr1 <- contr1[order(abs(contr1))] # order by contribution magnitude
+#' old_mar <- par("mar")
+#' par(mar = old_mar + c(0,7,0,0))
+#' barplot(contr1, horiz = TRUE, las = 2, xlab = "contribution to prediction in log-odds")
+#' par(mar = old_mar)
+#'
+#'
+#' ## multiclass classification in iris dataset:
+#'
+#' lb <- as.numeric(iris$Species) - 1
+#' num_class <- 3
+#' set.seed(11)
+#' bst <- xgboost(data = as.matrix(iris[, -5]), label = lb,
+#'                max_depth = 4, eta = 0.5, nthread = 2, nrounds = 10, subsample = 0.5,
+#'                objective = "multi:softprob", num_class = num_class)
+#' # predict for softmax returns num_class probability numbers per case:
+#' pred <- predict(bst, as.matrix(iris[, -5]))
+#' str(pred)
+#' # reshape it to a num_class-columns matrix
+#' pred <- matrix(pred, ncol=num_class, byrow=TRUE)
+#' # convert the probabilities to softmax labels
+#' pred_labels <- max.col(pred) - 1
+#' # the following should result in the same error as seen in the last iteration
+#' sum(pred_labels != lb)/length(lb)
+#'
+#' # compare that to the predictions from softmax:
+#' set.seed(11)
+#' bst <- xgboost(data = as.matrix(iris[, -5]), label = lb,
+#'                max_depth = 4, eta = 0.5, nthread = 2, nrounds = 10, subsample = 0.5,
+#'                objective = "multi:softmax", num_class = num_class)
+#' pred <- predict(bst, as.matrix(iris[, -5]))
+#' str(pred)
+#' all.equal(pred, pred_labels)
+#' # prediction from using only 5 iterations should result
+#' # in the same error as seen in iteration 5:
+#' pred5 <- predict(bst, as.matrix(iris[, -5]), ntreelimit=5)
+#' sum(pred5 != lb)/length(lb)
+#'
+#'
+#' ## random forest-like model of 25 trees for binary classification:
+#'
+#' set.seed(11)
+#' bst <- xgboost(data = train$data, label = train$label, max_depth = 5,
+#'                nthread = 2, nrounds = 1, objective = "binary:logistic",
+#'                num_parallel_tree = 25, subsample = 0.6, colsample_bytree = 0.1)
+#' # Inspect the prediction error vs number of trees:
+#' lb <- test$label
+#' dtest <- xgb.DMatrix(test$data, label=lb)
+#' err <- sapply(1:25, function(n) {
+#'   pred <- predict(bst, dtest, ntreelimit=n)
+#'   sum((pred > 0.5) != lb)/length(lb)
+#' })
+#' plot(err, type='l', ylim=c(0,0.1), xlab='#trees')
+#'
+#' @rdname predict.xgb.Booster
+#' @export
+predict.xgb.Booster <- function(object, newdata, missing = NA, outputmargin = FALSE, ntreelimit = NULL,
+                                predleaf = FALSE, predcontrib = FALSE, approxcontrib = FALSE, reshape = FALSE, ...) {
+
+  object <- xgb.Booster.complete(object, saveraw = FALSE)
+  if (!inherits(newdata, "xgb.DMatrix"))
+    newdata <- xgb.DMatrix(newdata, missing = missing)
+  if (!is.null(object[["feature_names"]]) &&
+      !is.null(colnames(newdata)) &&
+      !identical(object[["feature_names"]], colnames(newdata)))
+    stop("Feature names stored in `object` and `newdata` are different!")
+  if (is.null(ntreelimit))
+    ntreelimit <- NVL(object$best_ntreelimit, 0)
+  if (NVL(object$params[['booster']], '') == 'gblinear')
+    ntreelimit <- 0
+  if (ntreelimit < 0)
+    stop("ntreelimit cannot be negative")
+
+  option <- 0L + 1L * as.logical(outputmargin) + 2L * as.logical(predleaf) + 4L * as.logical(predcontrib) + 8L * as.logical(approxcontrib)
+
+  ret <- .Call(XGBoosterPredict_R, object$handle, newdata, option[1], as.integer(ntreelimit))
+
+  n_ret <- length(ret)
+  n_row <- nrow(newdata)
+  npred_per_case <- n_ret / n_row
+
+  if (n_ret %% n_row != 0)
+    stop("prediction length ", n_ret, " is not multiple of nrows(newdata) ", n_row)
+
+  if (predleaf) {
+    ret <- if (n_ret == n_row) {
+      matrix(ret, ncol = 1)
+    } else {
+      matrix(ret, nrow = n_row, byrow = TRUE)
+    }
+  } else if (predcontrib) {
+    n_col1 <- ncol(newdata) + 1
+    n_group <- npred_per_case / n_col1
+    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 = dnames)
+    } else if (n_group == 1) {
+      matrix(ret, nrow = n_row, byrow = TRUE, dimnames = dnames)
+    } else {
+      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)
+  }
+  return(ret)
+}
+
+#' @rdname predict.xgb.Booster
+#' @export
+predict.xgb.Booster.handle <- function(object, ...) {
+
+  bst <- xgb.handleToBooster(object)
+
+  ret <- predict(bst, ...)
+  return(ret)
+}
+
+
+#' Accessors for serializable attributes of a model.
+#'
+#' These methods allow to manipulate the key-value attribute strings of an xgboost model.
+#'
+#' @param object Object of class \code{xgb.Booster} or \code{xgb.Booster.handle}.
+#' @param name a non-empty character string specifying which attribute is to be accessed.
+#' @param value a value of an attribute for \code{xgb.attr<-}; for \code{xgb.attributes<-}
+#'        it's a list (or an object coercible to a list) with the names of attributes to set
+#'        and the elements corresponding to attribute values.
+#'        Non-character values are converted to character.
+#'        When attribute value is not a scalar, only the first index is used.
+#'        Use \code{NULL} to remove an attribute.
+#'
+#' @details
+#' The primary purpose of xgboost model attributes is to store some meta-data about the model.
+#' Note that they are a separate concept from the object attributes in R.
+#' Specifically, they refer to key-value strings that can be attached to an xgboost model,
+#' stored together with the model's binary representation, and accessed later
+#' (from R or any other interface).
+#' In contrast, any R-attribute assigned to an R-object of \code{xgb.Booster} class
+#' would not be saved by \code{xgb.save} because an xgboost model is an external memory object
+#' and its serialization is handled externally.
+#' Also, setting an attribute that has the same name as one of xgboost's parameters wouldn't
+#' change the value of that parameter for a model.
+#' Use \code{\link{xgb.parameters<-}} to set or change model parameters.
+#'
+#' The attribute setters would usually work more efficiently for \code{xgb.Booster.handle}
+#' than for \code{xgb.Booster}, since only just a handle (pointer) would need to be copied.
+#' That would only matter if attributes need to be set many times.
+#' Note, however, that when feeding a handle of an \code{xgb.Booster} object to the attribute setters,
+#' the raw model cache of an \code{xgb.Booster} object would not be automatically updated,
+#' and it would be user's responsibility to call \code{xgb.save.raw} to update it.
+#'
+#' The \code{xgb.attributes<-} setter either updates the existing or adds one or several attributes,
+#' but it doesn't delete the other existing attributes.
+#'
+#' @return
+#' \code{xgb.attr} returns either a string value of an attribute
+#' or \code{NULL} if an attribute wasn't stored in a model.
+#'
+#' \code{xgb.attributes} returns a list of all attribute stored in a model
+#' or \code{NULL} if a model has no stored attributes.
+#'
+#' @examples
+#' data(agaricus.train, package='xgboost')
+#' train <- agaricus.train
+#'
+#' bst <- xgboost(data = train$data, label = train$label, max_depth = 2,
+#'                eta = 1, nthread = 2, nrounds = 2, objective = "binary:logistic")
+#'
+#' xgb.attr(bst, "my_attribute") <- "my attribute value"
+#' print(xgb.attr(bst, "my_attribute"))
+#' xgb.attributes(bst) <- list(a = 123, b = "abc")
+#'
+#' xgb.save(bst, 'xgb.model')
+#' bst1 <- xgb.load('xgb.model')
+#' print(xgb.attr(bst1, "my_attribute"))
+#' print(xgb.attributes(bst1))
+#'
+#' # deletion:
+#' xgb.attr(bst1, "my_attribute") <- NULL
+#' print(xgb.attributes(bst1))
+#' xgb.attributes(bst1) <- list(a = NULL, b = NULL)
+#' print(xgb.attributes(bst1))
+#'
+#' @rdname xgb.attr
+#' @export
+xgb.attr <- function(object, name) {
+  if (is.null(name) || nchar(as.character(name[1])) == 0) stop("invalid attribute name")
+  handle <- xgb.get.handle(object)
+  .Call(XGBoosterGetAttr_R, handle, as.character(name[1]))
+}
+
+#' @rdname xgb.attr
+#' @export
+`xgb.attr<-` <- function(object, name, value) {
+  if (is.null(name) || nchar(as.character(name[1])) == 0) stop("invalid attribute name")
+  handle <- xgb.get.handle(object)
+  if (!is.null(value)) {
+    # Coerce the elements to be scalar strings.
+    # Q: should we warn user about non-scalar elements?
+    if (is.numeric(value[1])) {
+      value <- format(value[1], digits = 17)
+    } else {
+      value <- as.character(value[1])
+    }
+  }
+  .Call(XGBoosterSetAttr_R, handle, as.character(name[1]), value)
+  if (is(object, 'xgb.Booster') && !is.null(object$raw)) {
+    object$raw <- xgb.save.raw(object$handle)
+  }
+  object
+}
+
+#' @rdname xgb.attr
+#' @export
+xgb.attributes <- function(object) {
+  handle <- xgb.get.handle(object)
+  attr_names <- .Call(XGBoosterGetAttrNames_R, handle)
+  if (is.null(attr_names)) return(NULL)
+  res <- lapply(attr_names, function(x) {
+    .Call(XGBoosterGetAttr_R, handle, x)
+  })
+  names(res) <- attr_names
+  res
+}
+
+#' @rdname xgb.attr
+#' @export
+`xgb.attributes<-` <- function(object, value) {
+  a <- as.list(value)
+  if (is.null(names(a)) || any(nchar(names(a)) == 0)) {
+    stop("attribute names cannot be empty strings")
+  }
+  # Coerce the elements to be scalar strings.
+  # Q: should we warn a user about non-scalar elements?
+  a <- lapply(a, function(x) {
+    if (is.null(x)) return(NULL)
+    if (is.numeric(x[1])) {
+      format(x[1], digits = 17)
+    } else {
+      as.character(x[1])
+    }
+  })
+  handle <- xgb.get.handle(object)
+  for (i in seq_along(a)) {
+    .Call(XGBoosterSetAttr_R, handle, names(a[i]), a[[i]])
+  }
+  if (is(object, 'xgb.Booster') && !is.null(object$raw)) {
+    object$raw <- xgb.save.raw(object$handle)
+  }
+  object
+}
+
+#' Accessors for model parameters.
+#'
+#' Only the setter for xgboost parameters is currently implemented.
+#'
+#' @param object Object of class \code{xgb.Booster} or \code{xgb.Booster.handle}.
+#' @param value a list (or an object coercible to a list) with the names of parameters to set
+#'        and the elements corresponding to parameter values.
+#'
+#' @details
+#' Note that the setter would usually work more efficiently for \code{xgb.Booster.handle}
+#' than for \code{xgb.Booster}, since only just a handle would need to be copied.
+#'
+#' @examples
+#' data(agaricus.train, package='xgboost')
+#' train <- agaricus.train
+#'
+#' bst <- xgboost(data = train$data, label = train$label, max_depth = 2,
+#'                eta = 1, nthread = 2, nrounds = 2, objective = "binary:logistic")
+#'
+#' xgb.parameters(bst) <- list(eta = 0.1)
+#'
+#' @rdname xgb.parameters
+#' @export
+`xgb.parameters<-` <- function(object, value) {
+  if (length(value) == 0) return(object)
+  p <- as.list(value)
+  if (is.null(names(p)) || any(nchar(names(p)) == 0)) {
+    stop("parameter names cannot be empty strings")
+  }
+  names(p) <- gsub("\\.", "_", names(p))
+  p <- lapply(p, function(x) as.character(x)[1])
+  handle <- xgb.get.handle(object)
+  for (i in seq_along(p)) {
+    .Call(XGBoosterSetParam_R, handle, names(p[i]), p[[i]])
+  }
+  if (is(object, 'xgb.Booster') && !is.null(object$raw)) {
+    object$raw <- xgb.save.raw(object$handle)
+  }
+  object
+}
+
+# Extract the number of trees in a model.
+# TODO: either add a getter to C-interface, or simply set an 'ntree' attribute after each iteration.
+# internal utility function
+xgb.ntree <- function(bst) {
+  length(grep('^booster', xgb.dump(bst)))
+}
+
+
+#' Print xgb.Booster
+#'
+#' Print information about xgb.Booster.
+#'
+#' @param x an xgb.Booster object
+#' @param verbose whether to print detailed data (e.g., attribute values)
+#' @param ... not currently used
+#'
+#' @examples
+#' data(agaricus.train, package='xgboost')
+#' train <- agaricus.train
+#' bst <- xgboost(data = train$data, label = train$label, max_depth = 2,
+#'                eta = 1, nthread = 2, nrounds = 2, objective = "binary:logistic")
+#' attr(bst, 'myattr') <- 'memo'
+#'
+#' print(bst)
+#' print(bst, verbose=TRUE)
+#'
+#' @method print xgb.Booster
+#' @export
+print.xgb.Booster <- function(x, verbose = FALSE, ...) {
+  cat('##### xgb.Booster\n')
+
+  valid_handle <- !is.null.handle(x$handle)
+  if (!valid_handle)
+    cat("Handle is invalid! Suggest using xgb.Booster.complete\n")
+
+  cat('raw: ')
+  if (!is.null(x$raw)) {
+    cat(format(object.size(x$raw), units = "auto"), '\n')
+  } else {
+    cat('NULL\n')
+  }
+  if (!is.null(x$call)) {
+    cat('call:\n  ')
+    print(x$call)
+  }
+
+  if (!is.null(x$params)) {
+    cat('params (as set within xgb.train):\n')
+    cat( '  ',
+         paste(names(x$params),
+               paste0('"', unlist(x$params), '"'),
+               sep = ' = ', collapse = ', '), '\n', sep = '')
+  }
+  # TODO: need an interface to access all the xgboosts parameters
+
+  attrs <- character(0)
+  if (valid_handle)
+    attrs <- xgb.attributes(x)
+  if (length(attrs) > 0) {
+    cat('xgb.attributes:\n')
+    if (verbose) {
+      cat( paste(paste0('  ',names(attrs)),
+                 paste0('"', unlist(attrs), '"'),
+                 sep = ' = ', collapse = '\n'), '\n', sep = '')
+    } else {
+      cat('  ', paste(names(attrs), collapse = ', '), '\n', sep = '')
+    }
+  }
+
+  if (!is.null(x$callbacks) && length(x$callbacks) > 0) {
+    cat('callbacks:\n')
+    lapply(callback.calls(x$callbacks), function(x) {
+      cat('  ')
+      print(x)
+    })
+  }
+
+  if (!is.null(x$feature_names))
+    cat('# of features:', length(x$feature_names), '\n')
+
+  cat('niter: ', x$niter, '\n', sep = '')
+  # TODO: uncomment when faster xgb.ntree is implemented
+  #cat('ntree: ', xgb.ntree(x), '\n', sep='')
+
+  for (n in setdiff(names(x), c('handle', 'raw', 'call', 'params', 'callbacks',
+                                'evaluation_log','niter','feature_names'))) {
+    if (is.atomic(x[[n]])) {
+      cat(n, ':', x[[n]], '\n', sep = ' ')
+    } else {
+      cat(n, ':\n\t', sep = ' ')
+      print(x[[n]])
+    }
+  }
+
+  if (!is.null(x$evaluation_log)) {
+    cat('evaluation_log:\n')
+    print(x$evaluation_log, row.names = FALSE, topn = 2)
+  }
+
+  invisible(x)
+}

R-package/R/xgb.DMatrix.R

@@ -1,14 +1,17 @@
-#' Contruct xgb.DMatrix object
+#' Construct xgb.DMatrix object
 #' 
-#' Contruct xgb.DMatrix object from dense matrix, sparse matrix or local file.
+#' Construct xgb.DMatrix object from either a dense matrix, a sparse matrix, or a local file.
+#' Supported input file formats are either a libsvm text file or a binary file that was created previously by
+#' \code{\link{xgb.DMatrix.save}}).
 #' 
-#' @param data a \code{matrix} object, a \code{dgCMatrix} object or a character 
-#'   indicating the data file.
-#' @param info a list of information of the xgb.DMatrix object
-#' @param missing Missing is only used when input is dense matrix, pick a float
-#'     value that represents missing value. Sometime a data use 0 or other extreme value to represents missing values.
-#
-#' @param ... other information to pass to \code{info}.
+#' @param data a \code{matrix} object (either numeric or integer), a \code{dgCMatrix} object, or a character 
+#'        string representing a filename.
+#' @param info a named list of additional information to store in the \code{xgb.DMatrix} object.
+#'        See \code{\link{setinfo}} for the specific allowed kinds of 
+#' @param missing a float value to represents missing values in data (used only when input is a dense matrix).
+#'        It is useful when a 0 or some other extreme value represents missing values in data.
+#' @param silent whether to suppress printing an informational message after loading from a file.
+#' @param ... the \code{info} data could be passed directly as parameters, without creating an \code{info} list.
 #' 
 #' @examples
 #' data(agaricus.train, package='xgboost')
@@ -17,28 +20,342 @@
 #' xgb.DMatrix.save(dtrain, 'xgb.DMatrix.data')
 #' dtrain <- xgb.DMatrix('xgb.DMatrix.data')
 #' @export
-xgb.DMatrix <- function(data, info = list(), missing = NA, ...) {
+xgb.DMatrix <- function(data, info = list(), missing = NA, silent = FALSE, ...) {
+  cnames <- NULL
   if (typeof(data) == "character") {
-    handle <- .Call("XGDMatrixCreateFromFile_R", data, as.integer(FALSE),
-                    PACKAGE = "xgboost")
+    if (length(data) > 1)
+      stop("'data' has class 'character' and length ", length(data),
+           ".\n  'data' accepts either a numeric matrix or a single filename.")
+    handle <- .Call(XGDMatrixCreateFromFile_R, data, as.integer(silent))
   } else if (is.matrix(data)) {
-    handle <- .Call("XGDMatrixCreateFromMat_R", data, missing,
-                    PACKAGE = "xgboost")
-  } else if (class(data) == "dgCMatrix") {
-    handle <- .Call("XGDMatrixCreateFromCSC_R", data@p, data@i, data@x,
-                    PACKAGE = "xgboost")
+    handle <- .Call(XGDMatrixCreateFromMat_R, data, missing)
+    cnames <- colnames(data)
+  } else if (inherits(data, "dgCMatrix")) {
+    handle <- .Call(XGDMatrixCreateFromCSC_R, data@p, data@i, data@x, nrow(data))
+    cnames <- colnames(data)
   } else {
-    stop(paste("xgb.DMatrix: does not support to construct from ",
-               typeof(data)))
+    stop("xgb.DMatrix does not support construction from ", typeof(data))
   }
-  dmat <- structure(handle, class = "xgb.DMatrix")
+  dmat <- handle
+  attributes(dmat) <- list(.Dimnames = list(NULL, cnames), class = "xgb.DMatrix")
 
   info <- append(info, list(...))
-  if (length(info) == 0)
-    return(dmat)
-  for (i in 1:length(info)) {
+  for (i in seq_along(info)) {
     p <- info[i]
-    xgb.setinfo(dmat, names(p), p[[1]])
+    setinfo(dmat, names(p), p[[1]])
   }
   return(dmat)
 }
+
+
+# get dmatrix from data, label
+# internal helper method
+xgb.get.DMatrix <- function(data, label = NULL, missing = NA, weight = NULL) {
+  if (inherits(data, "dgCMatrix") || is.matrix(data)) {
+    if (is.null(label)) {
+      stop("label must be provided when data is a matrix")
+    }
+    dtrain <- xgb.DMatrix(data, label = label, missing = missing)
+    if (!is.null(weight)){
+      setinfo(dtrain, "weight", weight)
+    }
+  } else {
+    if (!is.null(label)) {
+      warning("xgboost: label will be ignored.")
+    }
+    if (is.character(data)) {
+      dtrain <- xgb.DMatrix(data[1])
+    } else if (inherits(data, "xgb.DMatrix")) {
+      dtrain <- data
+    } else if (inherits(data, "data.frame")) {
+      stop("xgboost doesn't support data.frame as input. Convert it to matrix first.")
+    } else {
+      stop("xgboost: invalid input data")
+    }
+  }
+  return (dtrain)
+}
+
+
+#' Dimensions of xgb.DMatrix
+#' 
+#' Returns a vector of numbers of rows and of columns in an \code{xgb.DMatrix}.
+#' @param x Object of class \code{xgb.DMatrix}
+#' 
+#' @details
+#' Note: since \code{nrow} and \code{ncol} internally use \code{dim}, they can also 
+#' be directly used with an \code{xgb.DMatrix} object.
+#'
+#' @examples
+#' data(agaricus.train, package='xgboost')
+#' train <- agaricus.train
+#' dtrain <- xgb.DMatrix(train$data, label=train$label)
+#' 
+#' stopifnot(nrow(dtrain) == nrow(train$data))
+#' stopifnot(ncol(dtrain) == ncol(train$data))
+#' stopifnot(all(dim(dtrain) == dim(train$data)))
+#' 
+#' @export
+dim.xgb.DMatrix <- function(x) {
+  c(.Call(XGDMatrixNumRow_R, x), .Call(XGDMatrixNumCol_R, x))
+}
+
+
+#' Handling of column names of \code{xgb.DMatrix}
+#' 
+#' Only column names are supported for \code{xgb.DMatrix}, thus setting of 
+#' row names would have no effect and returnten row names would be NULL.
+#' 
+#' @param x object of class \code{xgb.DMatrix}
+#' @param value a list of two elements: the first one is ignored
+#'        and the second one is column names 
+#' 
+#' @details
+#' Generic \code{dimnames} methods are used by \code{colnames}.
+#' Since row names are irrelevant, it is recommended to use \code{colnames} directly.
+#'
+#' @examples
+#' data(agaricus.train, package='xgboost')
+#' train <- agaricus.train
+#' dtrain <- xgb.DMatrix(train$data, label=train$label)
+#' dimnames(dtrain)
+#' colnames(dtrain)
+#' colnames(dtrain) <- make.names(1:ncol(train$data))
+#' print(dtrain, verbose=TRUE)
+#' 
+#' @rdname dimnames.xgb.DMatrix
+#' @export
+dimnames.xgb.DMatrix <- function(x) {
+  attr(x, '.Dimnames')
+}
+
+#' @rdname dimnames.xgb.DMatrix
+#' @export
+`dimnames<-.xgb.DMatrix` <- function(x, value) {
+  if (!is.list(value) || length(value) != 2L)
+    stop("invalid 'dimnames' given: must be a list of two elements")
+  if (!is.null(value[[1L]]))
+    stop("xgb.DMatrix does not have rownames")
+  if (is.null(value[[2]])) {
+    attr(x, '.Dimnames') <- NULL
+    return(x)
+  }
+  if (ncol(x) != length(value[[2]])) 
+    stop("can't assign ", length(value[[2]]), " colnames to a ", 
+         ncol(x), " column xgb.DMatrix")
+  attr(x, '.Dimnames') <- value
+  x
+}
+
+
+#' Get information of an xgb.DMatrix object
+#' 
+#' Get information of an xgb.DMatrix object
+#' @param object Object of class \code{xgb.DMatrix}
+#' @param name the name of the information field to get (see details)
+#' @param ... other parameters
+#' 
+#' @details
+#' The \code{name} field can be one of the following:
+#' 
+#' \itemize{
+#'     \item \code{label}: label Xgboost learn from ;
+#'     \item \code{weight}: to do a weight rescale ;
+#'     \item \code{base_margin}: base margin is the base prediction Xgboost will boost from ;
+#'     \item \code{nrow}: number of rows of the \code{xgb.DMatrix}.
+#'     
+#' }
+#' 
+#' \code{group} can be setup by \code{setinfo} but can't be retrieved by \code{getinfo}.
+#' 
+#' @examples
+#' data(agaricus.train, package='xgboost')
+#' train <- agaricus.train
+#' dtrain <- xgb.DMatrix(train$data, label=train$label)
+#' 
+#' labels <- getinfo(dtrain, 'label')
+#' setinfo(dtrain, 'label', 1-labels)
+#' 
+#' labels2 <- getinfo(dtrain, 'label')
+#' stopifnot(all(labels2 == 1-labels))
+#' @rdname getinfo
+#' @export
+getinfo <- function(object, ...) UseMethod("getinfo")
+
+#' @rdname getinfo
+#' @export
+getinfo.xgb.DMatrix <- function(object, name, ...) {
+  if (typeof(name) != "character" ||
+      length(name) != 1 ||
+      !name %in% c('label', 'weight', 'base_margin', 'nrow')) {
+    stop("getinfo: name must be one of the following\n",
+         "    'label', 'weight', 'base_margin', 'nrow'")
+  }
+  if (name != "nrow"){
+    ret <- .Call(XGDMatrixGetInfo_R, object, name)
+  } else {
+    ret <- nrow(object)
+  }
+  if (length(ret) == 0) return(NULL)
+  return(ret)
+}
+
+
+#' Set information of an xgb.DMatrix object
+#' 
+#' Set information of an xgb.DMatrix object
+#' 
+#' @param object Object of class "xgb.DMatrix"
+#' @param name the name of the field to get
+#' @param info the specific field of information to set
+#' @param ... other parameters
+#'
+#' @details
+#' The \code{name} field can be one of the following:
+#' 
+#' \itemize{
+#'     \item \code{label}: label Xgboost learn from ;
+#'     \item \code{weight}: to do a weight rescale ;
+#'     \item \code{base_margin}: base margin is the base prediction Xgboost will boost from ;
+#'     \item \code{group}: number of rows in each group (to use with \code{rank:pairwise} objective).
+#' }
+#' 
+#' @examples
+#' data(agaricus.train, package='xgboost')
+#' train <- agaricus.train
+#' dtrain <- xgb.DMatrix(train$data, label=train$label)
+#' 
+#' labels <- getinfo(dtrain, 'label')
+#' setinfo(dtrain, 'label', 1-labels)
+#' labels2 <- getinfo(dtrain, 'label')
+#' stopifnot(all.equal(labels2, 1-labels))
+#' @rdname setinfo
+#' @export
+setinfo <- function(object, ...) UseMethod("setinfo")
+
+#' @rdname setinfo
+#' @export
+setinfo.xgb.DMatrix <- function(object, name, info, ...) {
+  if (name == "label") {
+    if (length(info) != nrow(object))
+      stop("The length of labels must equal to the number of rows in the input data")
+    .Call(XGDMatrixSetInfo_R, object, name, as.numeric(info))
+    return(TRUE)
+  }
+  if (name == "weight") {
+    if (length(info) != nrow(object))
+      stop("The length of weights must equal to the number of rows in the input data")
+    .Call(XGDMatrixSetInfo_R, object, name, as.numeric(info))
+    return(TRUE)
+  }
+  if (name == "base_margin") {
+    # if (length(info)!=nrow(object))
+    #   stop("The length of base margin must equal to the number of rows in the input data")
+    .Call(XGDMatrixSetInfo_R, object, name, as.numeric(info))
+    return(TRUE)
+  }
+  if (name == "group") {
+    if (sum(info) != nrow(object))
+      stop("The sum of groups must equal to the number of rows in the input data")
+    .Call(XGDMatrixSetInfo_R, object, name, as.integer(info))
+    return(TRUE)
+  }
+  stop("setinfo: unknown info name ", name)
+  return(FALSE)
+}
+
+
+#' Get a new DMatrix containing the specified rows of
+#' orginal xgb.DMatrix object
+#'
+#' Get a new DMatrix containing the specified rows of
+#' orginal xgb.DMatrix object
+#' 
+#' @param object Object of class "xgb.DMatrix"
+#' @param idxset a integer vector of indices of rows needed
+#' @param colset currently not used (columns subsetting is not available)
+#' @param ... other parameters (currently not used)
+#' 
+#' @examples
+#' data(agaricus.train, package='xgboost')
+#' train <- agaricus.train
+#' dtrain <- xgb.DMatrix(train$data, label=train$label)
+#' 
+#' dsub <- slice(dtrain, 1:42)
+#' labels1 <- getinfo(dsub, 'label')
+#' dsub <- dtrain[1:42, ]
+#' labels2 <- getinfo(dsub, 'label')
+#' all.equal(labels1, labels2)
+#' 
+#' @rdname slice.xgb.DMatrix
+#' @export
+slice <- function(object, ...) UseMethod("slice")
+
+#' @rdname slice.xgb.DMatrix
+#' @export
+slice.xgb.DMatrix <- function(object, idxset, ...) {
+  if (!inherits(object, "xgb.DMatrix")) {
+    stop("object must be xgb.DMatrix")
+  }
+  ret <- .Call(XGDMatrixSliceDMatrix_R, object, idxset)
+
+  attr_list <- attributes(object)
+  nr <- nrow(object)
+  len <- sapply(attr_list, length)
+  ind <- which(len == nr)
+  if (length(ind) > 0) {
+    nms <- names(attr_list)[ind]
+    for (i in seq_along(ind)) {
+      attr(ret, nms[i]) <- attr(object, nms[i])[idxset]
+    }
+  }
+  return(structure(ret, class = "xgb.DMatrix"))
+}
+
+#' @rdname slice.xgb.DMatrix
+#' @export
+`[.xgb.DMatrix` <- function(object, idxset, colset = NULL) {
+  slice(object, idxset)
+}
+
+
+#' Print xgb.DMatrix
+#' 
+#' Print information about xgb.DMatrix. 
+#' Currently it displays dimensions and presence of info-fields and colnames.
+#' 
+#' @param x an xgb.DMatrix object
+#' @param verbose whether to print colnames (when present)
+#' @param ... not currently used
+#' 
+#' @examples
+#' data(agaricus.train, package='xgboost')
+#' train <- agaricus.train
+#' dtrain <- xgb.DMatrix(train$data, label=train$label)
+#' 
+#' dtrain
+#' print(dtrain, verbose=TRUE)
+#' 
+#' @method print xgb.DMatrix
+#' @export
+print.xgb.DMatrix <- function(x, verbose = FALSE, ...) {
+  cat('xgb.DMatrix  dim:', nrow(x), 'x', ncol(x), ' info: ')
+  infos <- c()
+  if(length(getinfo(x, 'label')) > 0) infos <- 'label'
+  if(length(getinfo(x, 'weight')) > 0) infos <- c(infos, 'weight')
+  if(length(getinfo(x, 'base_margin')) > 0) infos <- c(infos, 'base_margin')
+  if (length(infos) == 0) infos <- 'NA'
+  cat(infos)
+  cnames <- colnames(x)
+  cat('  colnames:')
+  if (verbose & !is.null(cnames)) {
+    cat("\n'")
+    cat(cnames, sep = "','")
+    cat("'")
+  } else {
+    if (is.null(cnames)) cat(' no')
+    else cat(' yes')
+  }
+  cat("\n")
+  invisible(x)
+}

R-package/R/xgb.DMatrix.save.R

@@ -2,8 +2,8 @@
 #' 
 #' Save xgb.DMatrix object to binary file
 #' 
-#' @param DMatrix the DMatrix object
-#' @param fname the name of the binary file.
+#' @param dmatrix the \code{xgb.DMatrix} object
+#' @param fname the name of the file to write.
 #' 
 #' @examples
 #' data(agaricus.train, package='xgboost')
@@ -12,15 +12,12 @@
 #' xgb.DMatrix.save(dtrain, 'xgb.DMatrix.data')
 #' dtrain <- xgb.DMatrix('xgb.DMatrix.data')
 #' @export
-xgb.DMatrix.save <- function(DMatrix, fname) {
-  if (typeof(fname) != "character") {
-    stop("xgb.save: fname must be character")
-  }
-  if (class(DMatrix) == "xgb.DMatrix") {
-    .Call("XGDMatrixSaveBinary_R", DMatrix, fname, as.integer(FALSE),
-          PACKAGE = "xgboost")
-    return(TRUE)
-  }
-  stop("xgb.DMatrix.save: the input must be xgb.DMatrix")
-  return(FALSE)
+xgb.DMatrix.save <- function(dmatrix, fname) {
+  if (typeof(fname) != "character")
+    stop("fname must be character")
+  if (!inherits(dmatrix, "xgb.DMatrix"))
+    stop("dmatrix must be xgb.DMatrix")
+  
+  .Call(XGDMatrixSaveBinary_R, dmatrix, fname[1], 0L)
+  return(TRUE)
 }

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

@@ -2,12 +2,9 @@
 #' 
 #' May improve the learning by adding new features to the training data based on the decision trees from a previously learned model.
 #' 
-#' @importFrom magrittr %>%
-#' @importFrom Matrix cBind
-#' @importFrom Matrix sparse.model.matrix
-#' 
 #' @param model decision tree boosting model learned on the original data
-#' @param training.data original data (usually provided as a \code{dgCMatrix} matrix)
+#' @param data original data (usually provided as a \code{dgCMatrix} matrix)
+#' @param ... currently not used
 #' 
 #' @return \code{dgCMatrix} matrix including both the original data and the new features.
 #'
@@ -17,15 +14,15 @@
 #' \strong{Practical Lessons from Predicting Clicks on Ads at Facebook}
 #' 
 #' \emph{(Xinran He, Junfeng Pan, Ou Jin, Tianbing Xu, Bo Liu, Tao Xu, Yan, xin Shi, Antoine Atallah, Ralf Herbrich, Stuart Bowers, 
-#' Joaquin Quiñonero Candela)}
+#' Joaquin Quinonero Candela)}
 #'  
 #' International Workshop on Data Mining for Online Advertising (ADKDD) - August 24, 2014
 #' 
-#' \url{https://research.facebook.com/publications/758569837499391/practical-lessons-from-predicting-clicks-on-ads-at-facebook/}.
+#' \url{https://research.fb.com/publications/practical-lessons-from-predicting-clicks-on-ads-at-facebook/}.
 #' 
 #' Extract explaining the method:
 #' 
-#' "\emph{We found that boosted decision trees are a powerful and very
+#' "We found that boosted decision trees are a powerful and very
 #' convenient way to implement non-linear and tuple transformations
 #' of the kind we just described. We treat each individual
 #' tree as a categorical feature that takes as value the
@@ -46,7 +43,7 @@
 #' based transformation as a supervised feature encoding that
 #' converts a real-valued vector into a compact binary-valued
 #' vector. A traversal from root node to a leaf node represents
-#' a rule on certain features.}"
+#' a rule on certain features."
 #' 
 #' @examples
 #' data(agaricus.train, package='xgboost')
@@ -54,13 +51,14 @@
 #' 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')
+#' param <- list(max_depth=2, eta=1, silent=1, objective='binary:logistic')
 #' nround = 4
 #'
 #' 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$label)
+#' accuracy.before <- sum((predict(bst, agaricus.test$data) >= 0.5) == agaricus.test$label) /
+#'                    length(agaricus.test$label)
 #' 
 #' # Convert previous features to one hot encoding
 #' new.features.train <- xgb.create.features(model = bst, agaricus.train$data)
@@ -73,19 +71,17 @@
 #' bst <- xgb.train(params = param, data = new.dtrain, nrounds = nround, nthread = 2)
 #' 
 #' # Model accuracy with new features
-#' accuracy.after <- sum((predict(bst, new.dtest) >= 0.5) == agaricus.test$label) / length(agaricus.test$label)
+#' accuracy.after <- sum((predict(bst, new.dtest) >= 0.5) == agaricus.test$label) /
+#'                   length(agaricus.test$label)
 #' 
 #' # Here the accuracy was already good and is now perfect.
-#' cat(paste("The accuracy was", accuracy.before, "before adding leaf features and it is now", accuracy.after, "!\n"))
+#' cat(paste("The accuracy was", accuracy.before, "before adding leaf features and it is now",
+#'           accuracy.after, "!\n"))
 #' 
 #' @export
-xgb.create.features <- function(model, training.data){
-  pred_with_leaf = predict(model, training.data, predleaf = TRUE)
-  cols <- list()
-  for(i in 1:length(trees)){
-    # max is not the real max but it s not important for the purpose of adding features
-    leaf.id <- sort(unique(pred_with_leaf[,i]))
-    cols[[i]] <- factor(x = pred_with_leaf[,i], level = leaf.id)
-  }
-  cBind(training.data, sparse.model.matrix( ~ . -1, as.data.frame(cols)))
-}
+xgb.create.features <- function(model, data, ...){
+  check.deprecation(...)
+  pred_with_leaf <- predict(model, data, predleaf = TRUE)
+  cols <- lapply(as.data.frame(pred_with_leaf), factor)
+  cbind(data, sparse.model.matrix( ~ . -1, cols))
+}

R-package/R/xgb.cv.R

@@ -1,17 +1,6 @@
 #' Cross Validation
 #' 
-#' The cross valudation function of xgboost
-#' 
-#' @importFrom data.table data.table
-#' @importFrom data.table as.data.table
-#' @importFrom magrittr %>%
-#' @importFrom data.table :=
-#' @importFrom data.table rbindlist
-#' @importFrom stringr str_extract_all
-#' @importFrom stringr str_extract
-#' @importFrom stringr str_split
-#' @importFrom stringr str_replace
-#' @importFrom stringr str_match
+#' The cross validation function of xgboost
 #' 
 #' @param params the list of parameters. Commonly used ones are:
 #' \itemize{
@@ -21,21 +10,23 @@
 #'     \item \code{binary:logistic} logistic regression for classification
 #'   }
 #'   \item \code{eta} step size of each boosting step
-#'   \item \code{max.depth} maximum depth of the tree
+#'   \item \code{max_depth} maximum depth of the tree
 #'   \item \code{nthread} number of thread used in training, if not set, all threads are used
 #' }
 #'
-#'   See \link{xgb.train} for further details.
+#'   See \code{\link{xgb.train}} for further details.
 #'   See also demo/ for walkthrough example in R.
-#' @param data takes an \code{xgb.DMatrix} or \code{Matrix} as the input.
+#' @param data takes an \code{xgb.DMatrix}, \code{matrix}, or \code{dgCMatrix} as the input.
 #' @param nrounds the max number of iterations
 #' @param nfold the original dataset is randomly partitioned into \code{nfold} equal size subsamples. 
-#' @param label option field, when data is \code{Matrix}
-#' @param missing Missing is only used when input is dense matrix, pick a float
-#'     value that represents missing value. Sometime a data use 0 or other extreme value to represents missing values.
-#' @param prediction A logical value indicating whether to return the prediction vector.
-#' @param showsd \code{boolean}, whether show standard deviation of cross validation
-#' @param metrics, list of evaluation metrics to be used in corss validation,
+#' @param label vector of response values. Should be provided only when data is an R-matrix.
+#' @param missing is only used when input is a dense matrix. By default is set to NA, which means 
+#'        that NA values should be considered as 'missing' by the algorithm. 
+#'        Sometimes, 0 or other extreme value might be used to represent missing values.
+#' @param prediction A logical value indicating whether to return the test fold predictions 
+#'        from each CV model. This parameter engages the \code{\link{cb.cv.predict}} callback.
+#' @param showsd \code{boolean}, whether to show standard deviation of cross validation
+#' @param metrics, list of evaluation metrics to be used in cross validation,
 #'   when it is not specified, the evaluation metric is chosen according to objective function.
 #'   Possible options are:
 #' \itemize{
@@ -43,35 +34,37 @@
 #'   \item \code{rmse} Rooted mean square error
 #'   \item \code{logloss} negative log-likelihood function
 #'   \item \code{auc} Area under curve
+#'   \item \code{aucpr} Area under PR curve
 #'   \item \code{merror} Exact matching error, used to evaluate multi-class classification
 #' }
 #' @param obj customized objective function. Returns gradient and second order 
-#'   gradient with given prediction and dtrain.
+#'        gradient with given prediction and dtrain.
 #' @param feval custimized evaluation function. Returns 
-#'   \code{list(metric='metric-name', value='metric-value')} with given 
-#'   prediction and dtrain.
-#' @param stratified \code{boolean} whether sampling of folds should be stratified by the values of labels in \code{data}
-#' @param folds \code{list} provides a possibility of using a list of pre-defined CV folds (each element must be a vector of fold's indices).
-#'   If folds are supplied, the nfold and stratified parameters would be ignored.
+#'        \code{list(metric='metric-name', value='metric-value')} with given 
+#'        prediction and dtrain.
+#' @param stratified a \code{boolean} indicating whether sampling of folds should be stratified 
+#'        by the values of outcome labels.
+#' @param folds \code{list} provides a possibility to use a list of pre-defined CV folds
+#'        (each element must be a vector of test fold's indices). When folds are supplied, 
+#'        the \code{nfold} and \code{stratified} parameters are ignored.
 #' @param verbose \code{boolean}, print the statistics during the process
-#' @param print.every.n Print every N progress messages when \code{verbose>0}. Default is 1 which means all messages are printed.
-#' @param early.stop.round If \code{NULL}, the early stopping function is not triggered. 
-#'     If set to an integer \code{k}, training with a validation set will stop if the performance 
-#'     keeps getting worse consecutively for \code{k} rounds.
-#' @param maximize If \code{feval} and \code{early.stop.round} are set, then \code{maximize} must be set as well.
-#'     \code{maximize=TRUE} means the larger the evaluation score the better.
-#'     
+#' @param print_every_n Print each n-th iteration evaluation messages when \code{verbose>0}.
+#'        Default is 1 which means all messages are printed. This parameter is passed to the 
+#'        \code{\link{cb.print.evaluation}} callback.
+#' @param early_stopping_rounds If \code{NULL}, the early stopping function is not triggered. 
+#'        If set to an integer \code{k}, training with a validation set will stop if the performance 
+#'        doesn't improve for \code{k} rounds.
+#'        Setting this parameter engages the \code{\link{cb.early.stop}} callback.
+#' @param maximize If \code{feval} and \code{early_stopping_rounds} are set,
+#'        then this parameter must be set as well.
+#'        When it is \code{TRUE}, it means the larger the evaluation score the better.
+#'        This parameter is passed to the \code{\link{cb.early.stop}} callback.
+#' @param callbacks a list of callback functions to perform various task during boosting.
+#'        See \code{\link{callbacks}}. Some of the callbacks are automatically created depending on the 
+#'        parameters' values. User can provide either existing or their own callback methods in order 
+#'        to customize the training process.
 #' @param ... other parameters to pass to \code{params}.
 #' 
-#' @return
-#' If \code{prediction = TRUE}, a list with the following elements is returned:
-#' \itemize{
-#'   \item \code{dt} a \code{data.table} with each mean and standard deviation stat for training set and test set
-#'   \item \code{pred} an array or matrix (for multiclass classification) with predictions for each CV-fold for the model having been trained on the data in all other folds.
-#' }
-#'
-#' If \code{prediction = FALSE}, just a \code{data.table} with each mean and standard deviation stat for training set and test set is returned.
-#'
 #' @details 
 #' The original sample is randomly partitioned into \code{nfold} equal size subsamples. 
 #' 
@@ -83,168 +76,230 @@
 #' 
 #' Adapted from \url{http://en.wikipedia.org/wiki/Cross-validation_\%28statistics\%29#k-fold_cross-validation}
 #'
+#' @return 
+#' An object of class \code{xgb.cv.synchronous} with the following elements:
+#' \itemize{
+#'   \item \code{call} a function call.
+#'   \item \code{params} parameters that were passed to the xgboost library. Note that it does not 
+#'         capture parameters changed by the \code{\link{cb.reset.parameters}} callback.
+#'   \item \code{callbacks} callback functions that were either automatically assigned or 
+#'         explicitly passed.
+#'   \item \code{evaluation_log} evaluation history storead as a \code{data.table} with the
+#'         first column corresponding to iteration number and the rest corresponding to the 
+#'         CV-based evaluation means and standard deviations for the training and test CV-sets.
+#'         It is created by the \code{\link{cb.evaluation.log}} callback.
+#'   \item \code{niter} number of boosting iterations.
+#'   \item \code{nfeatures} number of features in training data.
+#'   \item \code{folds} the list of CV folds' indices - either those passed through the \code{folds} 
+#'         parameter or randomly generated.
+#'   \item \code{best_iteration} iteration number with the best evaluation metric value
+#'         (only available with early stopping).
+#'   \item \code{best_ntreelimit} the \code{ntreelimit} value corresponding to the best iteration, 
+#'         which could further be used in \code{predict} method
+#'         (only available with early stopping).
+#'   \item \code{pred} CV prediction values available when \code{prediction} is set. 
+#'         It is either vector or matrix (see \code{\link{cb.cv.predict}}).
+#'   \item \code{models} a liost of the CV folds' models. It is only available with the explicit 
+#'         setting of the \code{cb.cv.predict(save_models = TRUE)} callback.
+#' }
+#'
 #' @examples
 #' data(agaricus.train, package='xgboost')
 #' dtrain <- xgb.DMatrix(agaricus.train$data, label = agaricus.train$label)
-#' history <- xgb.cv(data = dtrain, nround=3, nthread = 2, nfold = 5, metrics=list("rmse","auc"),
-#'                   max.depth =3, eta = 1, objective = "binary:logistic")
-#' print(history)
+#' cv <- xgb.cv(data = dtrain, nrounds = 3, nthread = 2, nfold = 5, metrics = list("rmse","auc"),
+#'                   max_depth = 3, eta = 1, objective = "binary:logistic")
+#' print(cv)
+#' print(cv, verbose=TRUE)
+#' 
 #' @export
 xgb.cv <- function(params=list(), data, nrounds, nfold, label = NULL, missing = NA,
                    prediction = FALSE, showsd = TRUE, metrics=list(),
-                   obj = NULL, feval = NULL, stratified = TRUE, folds = NULL, verbose = T, print.every.n=1L,
-                   early.stop.round = NULL, maximize = NULL, ...) {
-    if (typeof(params) != "list") {
-        stop("xgb.cv: first argument params must be list")
-    }
-    if(!is.null(folds)) {
-        if(class(folds) != "list" | length(folds) < 2) {
-            stop("folds must be a list with 2 or more elements that are vectors of indices for each CV-fold")
-        }
-        nfold <- length(folds)
-    }
-    if (nfold <= 1) {
-        stop("nfold must be bigger than 1")
-    }
-    dtrain <- xgb.get.DMatrix(data, label, missing)
-    dot.params <- list(...)
-    nms.params <- names(params)
-    nms.dot.params <- names(dot.params)
-    if (length(intersect(nms.params,nms.dot.params)) > 0)
-        stop("Duplicated defined term in parameters. Please check your list of params.")
-    params <- append(params, dot.params)
-    params <- append(params, list(silent=1))
-    for (mc in metrics) {
-        params <- append(params, list("eval_metric"=mc))
-    }
+                   obj = NULL, feval = NULL, stratified = TRUE, folds = NULL, 
+                   verbose = TRUE, print_every_n=1L,
+                   early_stopping_rounds = NULL, maximize = NULL, callbacks = list(), ...) {
 
-    # customized objective and evaluation metric interface
-    if (!is.null(params$objective) && !is.null(obj))
-        stop("xgb.cv: cannot assign two different objectives")
-    if (!is.null(params$objective))
-        if (class(params$objective) == 'function') {
-            obj <- params$objective
-            params[['objective']] <- NULL
-        }
-    # if (!is.null(params$eval_metric) && !is.null(feval))
-    #  stop("xgb.cv: cannot assign two different evaluation metrics")
-    if (!is.null(params$eval_metric))
-        if (class(params$eval_metric) == 'function') {
-            feval <- params$eval_metric
-            params[['eval_metric']] <- NULL
-        }
+  check.deprecation(...)
+  
+  params <- check.booster.params(params, ...)
+  # TODO: should we deprecate the redundant 'metrics' parameter?
+  for (m in metrics)
+    params <- c(params, list("eval_metric" = m))
+  
+  check.custom.obj()
+  check.custom.eval()
 
-    # Early Stopping
-    if (!is.null(early.stop.round)){
-        if (!is.null(feval) && is.null(maximize))
-            stop('Please set maximize to note whether the model is maximizing the evaluation or not.')
-        if (is.null(maximize) && is.null(params$eval_metric))
-            stop('Please set maximize to note whether the model is maximizing the evaluation or not.')
-        if (is.null(maximize))
-        {
-            if (params$eval_metric %in% c('rmse','logloss','error','merror','mlogloss')) {
-                maximize <- FALSE
-            } else {
-                maximize <- TRUE
-            }
-        }
+  #if (is.null(params[['eval_metric']]) && is.null(feval))
+  #  stop("Either 'eval_metric' or 'feval' must be provided for CV")
+  
+  # Check the labels
+  if ( (inherits(data, 'xgb.DMatrix') && is.null(getinfo(data, 'label'))) ||
+       (!inherits(data, 'xgb.DMatrix') && is.null(label)))
+    stop("Labels must be provided for CV either through xgb.DMatrix, or through 'label=' when 'data' is matrix")
+  
+  # CV folds
+  if(!is.null(folds)) {
+    if(!is.list(folds) || length(folds) < 2)
+      stop("'folds' must be a list with 2 or more elements that are vectors of indices for each CV-fold")
+    nfold <- length(folds)
+  } else {
+    if (nfold <= 1)
+      stop("'nfold' must be > 1")
+    folds <- generate.cv.folds(nfold, nrow(data), stratified, label, params)
+  }
+  
+  # Potential TODO: sequential CV
+  #if (strategy == 'sequential')
+  #  stop('Sequential CV strategy is not yet implemented')
 
-        if (maximize) {
-            bestScore <- 0
-        } else {
-            bestScore <- Inf
-        }
-        bestInd <- 0
-        earlyStopflag <- FALSE
+  # verbosity & evaluation printing callback:
+  params <- c(params, list(silent = 1))
+  print_every_n <- max( as.integer(print_every_n), 1L)
+  if (!has.callbacks(callbacks, 'cb.print.evaluation') && verbose) {
+    callbacks <- add.cb(callbacks, cb.print.evaluation(print_every_n, showsd = showsd))
+  }
+  # evaluation log callback: always is on in CV
+  evaluation_log <- list()
+  if (!has.callbacks(callbacks, 'cb.evaluation.log')) {
+    callbacks <- add.cb(callbacks, cb.evaluation.log())
+  }
+  # Early stopping callback
+  stop_condition <- FALSE
+  if (!is.null(early_stopping_rounds) &&
+      !has.callbacks(callbacks, 'cb.early.stop')) {
+    callbacks <- add.cb(callbacks, cb.early.stop(early_stopping_rounds, 
+                                                 maximize = maximize, verbose = verbose))
+  }
+  # CV-predictions callback
+  if (prediction &&
+      !has.callbacks(callbacks, 'cb.cv.predict')) {
+    callbacks <- add.cb(callbacks, cb.cv.predict(save_models = FALSE))
+  }
+  # Sort the callbacks into categories
+  cb <- categorize.callbacks(callbacks)
 
-        if (length(metrics) > 1)
-            warning('Only the first metric is used for early stopping process.')
-    }
+  
+  # create the booster-folds
+  dall <- xgb.get.DMatrix(data, label, missing)
+  bst_folds <- lapply(seq_along(folds), function(k) {
+    dtest  <- slice(dall, folds[[k]])
+    dtrain <- slice(dall, unlist(folds[-k]))
+    handle <- xgb.Booster.handle(params, list(dtrain, dtest))
+    list(dtrain = dtrain, bst = handle, watchlist = list(train = dtrain, test=dtest), index = folds[[k]])
+  })
+  rm(dall)
+  # a "basket" to collect some results from callbacks
+  basket <- list()
 
-    xgb_folds <- xgb.cv.mknfold(dtrain, nfold, params, stratified, folds)
-    obj_type <- params[['objective']]
-    mat_pred <- FALSE
-    if (!is.null(obj_type) && obj_type == 'multi:softprob')
-    {
-        num_class <- params[['num_class']]
-        if (is.null(num_class))
-            stop('must set num_class to use softmax')
-        predictValues <- matrix(0,xgb.numrow(dtrain),num_class)
-        mat_pred <- TRUE
-    }
-    else
-        predictValues <- rep(0,xgb.numrow(dtrain))
-    history <- c()
-    print.every.n <- max(as.integer(print.every.n), 1L)
-    for (i in 1:nrounds) {
-        msg <- list()
-        for (k in 1:nfold) {
-            fd <- xgb_folds[[k]]
-            succ <- xgb.iter.update(fd$booster, fd$dtrain, i - 1, obj)
-            msg[[k]] <- xgb.iter.eval(fd$booster, fd$watchlist, i - 1, feval) %>% str_split("\t") %>% .[[1]]
-        }
-        ret <- xgb.cv.aggcv(msg, showsd)
-        history <- c(history, ret)
-        if(verbose)
-            if (0 == (i - 1L) %% print.every.n)
-                cat(ret, "\n", sep="")
+  # extract parameters that can affect the relationship b/w #trees and #iterations
+  num_class <- max(as.numeric(NVL(params[['num_class']], 1)), 1)
+  num_parallel_tree <- max(as.numeric(NVL(params[['num_parallel_tree']], 1)), 1)
 
-        # early_Stopping
-        if (!is.null(early.stop.round)){
-            score <- strsplit(ret,'\\s+')[[1]][2 + length(metrics)]
-            score <- strsplit(score,'\\+|:')[[1]][[2]]
-            score <- as.numeric(score)
-            if ( (maximize && score > bestScore) || (!maximize && score < bestScore)) {
-                bestScore <- score
-                bestInd <- i
-            } else {
-                if (i - bestInd >= early.stop.round) {
-                    earlyStopflag <- TRUE
-                    cat('Stopping. Best iteration:', bestInd, '\n')
-                    break
-                }
-            }
-        }
-    }
+  # those are fixed for CV (no training continuation)
+  begin_iteration <- 1
+  end_iteration <- nrounds
+  
+  # synchronous CV boosting: run CV folds' models within each iteration
+  for (iteration in begin_iteration:end_iteration) {
+    
+    for (f in cb$pre_iter) f()
+    
+    msg <- lapply(bst_folds, function(fd) {
+      xgb.iter.update(fd$bst, fd$dtrain, iteration - 1, obj)
+      xgb.iter.eval(fd$bst, fd$watchlist, iteration - 1, feval)
+    })
+    msg <- simplify2array(msg)
+    bst_evaluation <- rowMeans(msg)
+    bst_evaluation_err <- sqrt(rowMeans(msg^2) - bst_evaluation^2)
+    
+    for (f in cb$post_iter) f()
+    
+    if (stop_condition) break
+  }
+  for (f in cb$finalize) f(finalize = TRUE)
 
-    if (prediction) {
-        for (k in 1:nfold) {
-            fd <- xgb_folds[[k]]
-            if (!is.null(early.stop.round) && earlyStopflag) {
-              res <- xgb.iter.eval(fd$booster, fd$watchlist, bestInd - 1, feval, prediction)
-            } else {
-              res <- xgb.iter.eval(fd$booster, fd$watchlist, nrounds - 1, feval, prediction)
-            }
-            if (mat_pred) {
-                pred_mat <- matrix(res[[2]],num_class,length(fd$index))
-                predictValues[fd$index,] <- t(pred_mat)
-            } else {
-                predictValues[fd$index] <- res[[2]]
-            }
-        }
-    }
+  # the CV result
+  ret <- list(
+    call = match.call(),
+    params = params,
+    callbacks = callbacks,
+    evaluation_log = evaluation_log,
+    niter = end_iteration,
+    nfeatures = ncol(data),
+    folds = folds
+  )
+  ret <- c(ret, basket)
 
-    colnames <- str_split(string = history[1], pattern = "\t")[[1]] %>% .[2:length(.)] %>% str_extract(".*:") %>% str_replace(":","") %>% str_replace("-", ".")
-    colnamesMean <- paste(colnames, "mean")
-    if(showsd) colnamesStd <- paste(colnames, "std")
-
-    colnames <- c()
-    if(showsd) for(i in 1:length(colnamesMean)) colnames <- c(colnames, colnamesMean[i], colnamesStd[i])
-    else colnames <- colnamesMean
-
-    type <- rep(x = "numeric", times = length(colnames))
-    dt <- utils::read.table(text = "", colClasses = type, col.names = colnames) %>% as.data.table
-    split <- str_split(string = history, pattern = "\t")
-
-    for(line in split) dt <- line[2:length(line)] %>% str_extract_all(pattern = "\\d*\\.+\\d*") %>% unlist %>% as.numeric %>% as.list %>% {rbindlist( list( dt, .), use.names = F, fill = F)}
-
-    if (prediction) {
-        return( list( dt = dt,pred = predictValues))
-    }
-    return(dt)
+  class(ret) <- 'xgb.cv.synchronous'
+  invisible(ret)
 }
 
-# Avoid error messages during CRAN check.
-# The reason is that these variables are never declared
-# They are mainly column names inferred by Data.table...
-globalVariables(".")
+
+
+#' Print xgb.cv result
+#' 
+#' Prints formatted results of \code{xgb.cv}.
+#' 
+#' @param x an \code{xgb.cv.synchronous} object
+#' @param verbose whether to print detailed data
+#' @param ... passed to \code{data.table.print}
+#' 
+#' @details
+#' When not verbose, it would only print the evaluation results, 
+#' including the best iteration (when available).
+#' 
+#' @examples
+#' data(agaricus.train, package='xgboost')
+#' train <- agaricus.train
+#' cv <- xgb.cv(data = train$data, label = train$label, nfold = 5, max_depth = 2,
+#'                eta = 1, nthread = 2, nrounds = 2, objective = "binary:logistic")
+#' print(cv)
+#' print(cv, verbose=TRUE)
+#' 
+#' @rdname print.xgb.cv
+#' @method print xgb.cv.synchronous
+#' @export
+print.xgb.cv.synchronous <- function(x, verbose = FALSE, ...) {
+  cat('##### xgb.cv ', length(x$folds), '-folds\n', sep = '')
+  
+  if (verbose) {
+    if (!is.null(x$call)) {
+      cat('call:\n  ')
+      print(x$call)
+    }
+    if (!is.null(x$params)) {
+      cat('params (as set within xgb.cv):\n')
+      cat( '  ', 
+           paste(names(x$params), 
+                 paste0('"', unlist(x$params), '"'),
+                 sep = ' = ', collapse = ', '), '\n', sep = '')
+    }
+    if (!is.null(x$callbacks) && length(x$callbacks) > 0) {
+      cat('callbacks:\n')
+      lapply(callback.calls(x$callbacks), function(x) {
+        cat('  ')
+        print(x)
+      })
+    }
+    
+    for (n in c('niter', 'best_iteration', 'best_ntreelimit')) {
+      if (is.null(x[[n]])) 
+        next
+      cat(n, ': ', x[[n]], '\n', sep = '')
+    }
+
+    if (!is.null(x$pred)) {
+      cat('pred:\n')
+      str(x$pred)
+    }
+  }
+
+  if (verbose) 
+    cat('evaluation_log:\n')
+  print(x$evaluation_log, row.names = FALSE, ...)
+  
+  if (!is.null(x$best_iteration)) {
+    cat('Best iteration:\n')
+    print(x$evaluation_log[x$best_iteration], row.names = FALSE, ...)
+  }
+  invisible(x)
+}

R-package/R/xgb.dump.R

@@ -1,70 +1,72 @@
-#' Save xgboost model to text file
+#' Dump an xgboost model in text format.
 #' 
-#' Save a xgboost model to text file. Could be parsed later.
+#' Dump an xgboost model in text format.
 #' 
-#' @importFrom magrittr %>%
-#' @importFrom stringr str_replace
-#' @importFrom data.table fread
-#' @importFrom data.table :=
-#' @importFrom data.table setnames
 #' @param model the model object.
-#' @param fname the name of the text file where to save the model text dump. If not provided or set to \code{NULL} the function will return the model as a \code{character} vector.
-#' @param fmap feature map file representing the type of feature. 
+#' @param fname the name of the text file where to save the model text dump. 
+#'        If not provided or set to \code{NULL}, the model is returned as a \code{character} vector.
+#' @param fmap feature map file representing feature types.
 #'        Detailed description could be found at 
 #'        \url{https://github.com/dmlc/xgboost/wiki/Binary-Classification#dump-model}.
 #'        See demo/ for walkthrough example in R, and
 #'        \url{https://github.com/dmlc/xgboost/blob/master/demo/data/featmap.txt} 
 #'        for example Format.
-#' @param with.stats whether dump statistics of splits 
-#'        When this option is on, the model dump comes with two additional statistics:
+#' @param with_stats whether to dump some additional statistics about the splits.
+#'        When this option is on, the model dump contains two additional values:
 #'        gain is the approximate loss function gain we get in each split;
 #'        cover is the sum of second order gradient in each node.
+#' @param dump_format either 'text' or 'json' format could be specified.
+#' @param ... currently not used
 #'
 #' @return
-#' if fname is not provided or set to \code{NULL} the function will return the model as a \code{character} vector. Otherwise it will return \code{TRUE}.
+#' If fname is not provided or set to \code{NULL} the function will return the model
+#' as a \code{character} vector. Otherwise it will return \code{TRUE}.
 #'
 #' @examples
 #' data(agaricus.train, package='xgboost')
 #' data(agaricus.test, package='xgboost')
 #' train <- agaricus.train
 #' test <- agaricus.test
-#' bst <- xgboost(data = train$data, label = train$label, max.depth = 2, 
-#'                eta = 1, nthread = 2, nround = 2,objective = "binary:logistic")
+#' 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'
-#' xgb.dump(bst, 'xgb.model.dump', 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))
+#' print(xgb.dump(bst, with_stats = TRUE))
+#' 
+#' # print in JSON format:
+#' cat(xgb.dump(bst, with_stats = TRUE, dump_format='json'))
+#' 
 #' @export
-xgb.dump <- function(model = NULL, fname = NULL, fmap = "", with.stats=FALSE) {
-  if (class(model) != "xgb.Booster") {
-    stop("model: argument must be type xgb.Booster")
+xgb.dump <- function(model, fname = NULL, fmap = "", with_stats=FALSE,
+                     dump_format = c("text", "json"), ...) {
+  check.deprecation(...)
+  dump_format <- match.arg(dump_format)
+  if (!inherits(model, "xgb.Booster"))
+    stop("model: argument must be of type xgb.Booster")
+  if (!(is.null(fname) || is.character(fname)))
+    stop("fname: argument must be a character string (when provided)")
+  if (!(is.null(fmap) || is.character(fmap)))
+    stop("fmap: argument must be a character string (when provided)")
+  
+  model <- xgb.Booster.complete(model)
+  model_dump <- .Call(XGBoosterDumpModel_R, model$handle, NVL(fmap, "")[1], as.integer(with_stats),
+                      as.character(dump_format))
+
+  if (is.null(fname)) 
+    model_dump <- stri_replace_all_regex(model_dump, '\t', '')
+  
+  if (dump_format == "text")
+    model_dump <- unlist(stri_split_regex(model_dump, '\n'))
+  
+  model_dump <- grep('^\\s*$', model_dump, invert = TRUE, value = TRUE)
+  
+  if (is.null(fname)) {
+    return(model_dump)
   } else {
-    model <- xgb.Booster.check(model)
-  }
-  if (!(class(fname) %in% c("character", "NULL") && length(fname) <= 1)) {
-    stop("fname: argument must be type character (when provided)")
-  }
-  if (!(class(fmap) %in% c("character", "NULL") && length(fname) <= 1)) {
-    stop("fmap: argument must be type character (when provided)")
-  }
-
-  longString <- .Call("XGBoosterDumpModel_R", model$handle, fmap, as.integer(with.stats), PACKAGE = "xgboost")
-
-  dt <- fread(paste(longString, collapse = ""), sep = "\n", header = F)
-
-  setnames(dt, "Lines")
-
-  if(is.null(fname)) {
-    result <- dt[Lines != "0"][, Lines := str_replace(Lines, "^\t+", "")][Lines != ""][, paste(Lines)]
-    return(result)
-  } else {
-    result <- dt[Lines != "0"][Lines != ""][, paste(Lines)] %>% writeLines(fname)
+    writeLines(model_dump, fname[1])
     return(TRUE)
   }
 }
-
-# Avoid error messages during CRAN check.
-# The reason is that these variables are never declared
-# They are mainly column names inferred by Data.table...
-globalVariables(c("Lines", "."))

R-package/R/xgb.ggplot.R

@@ -0,0 +1,135 @@
+# ggplot backend for the xgboost plotting facilities
+
+
+#' @rdname xgb.plot.importance
+#' @export
+xgb.ggplot.importance <- function(importance_matrix = NULL, top_n = NULL, measure = NULL, 
+                                  rel_to_first = FALSE, n_clusters = c(1:10), ...) {
+  
+  importance_matrix <- xgb.plot.importance(importance_matrix, top_n = top_n, measure = measure,
+                                           rel_to_first = rel_to_first, plot = FALSE, ...)
+  if (!requireNamespace("ggplot2", quietly = TRUE)) {
+    stop("ggplot2 package is required", call. = FALSE)
+  }
+  if (!requireNamespace("Ckmeans.1d.dp", quietly = TRUE)) {
+    stop("Ckmeans.1d.dp package is required", call. = FALSE)
+  }
+  
+  clusters <- suppressWarnings(
+    Ckmeans.1d.dp::Ckmeans.1d.dp(importance_matrix$Importance, n_clusters)
+  )
+  importance_matrix[, Cluster := as.character(clusters$cluster)]
+
+  plot <-
+    ggplot2::ggplot(importance_matrix, 
+                    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() + 
+    ggplot2::xlab("Features") + 
+    ggplot2::ggtitle("Feature importance") + 
+    ggplot2::theme(plot.title = ggplot2::element_text(lineheight = .9, face = "bold"), 
+                   panel.grid.major.y = ggplot2::element_blank())
+  return(plot)
+}
+
+
+#' @rdname xgb.plot.deepness
+#' @export
+xgb.ggplot.deepness <- function(model = NULL, which = c("2x1", "max.depth", "med.depth", "med.weight")) {
+
+  if (!requireNamespace("ggplot2", quietly = TRUE))
+    stop("ggplot2 package is required for plotting the graph deepness.", call. = FALSE)
+
+  which <- match.arg(which)
+  
+  dt_depths <- xgb.plot.deepness(model = model, plot = FALSE)
+  dt_summaries <- dt_depths[, .(.N, Cover = mean(Cover)), Depth]
+  setkey(dt_summaries, 'Depth')
+
+  if (which == "2x1") {
+    p1 <-
+      ggplot2::ggplot(dt_summaries) +
+      ggplot2::geom_bar(ggplot2::aes(x = Depth, y = N), stat = "Identity") +
+      ggplot2::xlab("") +
+      ggplot2::ylab("Number of leafs") +
+      ggplot2::ggtitle("Model complexity") +
+      ggplot2::theme(
+        plot.title = ggplot2::element_text(lineheight = 0.9, face = "bold"),
+        panel.grid.major.y = ggplot2::element_blank(),
+        axis.ticks = ggplot2::element_blank(),
+        axis.text.x = ggplot2::element_blank()
+      )
+  
+    p2 <- 
+      ggplot2::ggplot(dt_summaries) +
+      ggplot2::geom_bar(ggplot2::aes(x = Depth, y = Cover), stat = "Identity") + 
+      ggplot2::xlab("Leaf depth") +
+      ggplot2::ylab("Weighted cover")
+  
+    multiplot(p1, p2, cols = 1)
+    return(invisible(list(p1, p2)))
+    
+  } else if (which == "max.depth") {
+    p <-
+      ggplot2::ggplot(dt_depths[, max(Depth), Tree]) +
+      ggplot2::geom_jitter(ggplot2::aes(x = Tree, y = V1),
+                           height = 0.15, alpha=0.4, size=3, stroke=0) + 
+      ggplot2::xlab("tree #") +
+      ggplot2::ylab("Max tree leaf depth")
+    return(p)
+    
+  } else if (which == "med.depth") {
+    p <-
+      ggplot2::ggplot(dt_depths[, median(as.numeric(Depth)), Tree]) +
+      ggplot2::geom_jitter(ggplot2::aes(x = Tree, y = V1),
+                           height = 0.15, alpha=0.4, size=3, stroke=0) + 
+      ggplot2::xlab("tree #") +
+      ggplot2::ylab("Median tree leaf depth")
+    return(p)
+
+  } else if (which == "med.weight") {
+    p <-
+      ggplot2::ggplot(dt_depths[, median(abs(Weight)), Tree]) +
+      ggplot2::geom_point(ggplot2::aes(x = Tree, y = V1),
+                          alpha=0.4, size=3, stroke=0) + 
+      ggplot2::xlab("tree #") +
+      ggplot2::ylab("Median absolute leaf weight")
+    return(p)
+  }
+}
+
+# Plot multiple ggplot graph aligned by rows and columns.
+# ... the plots
+# cols number of columns
+# internal utility function
+multiplot <- function(..., cols = 1) {
+  plots <- list(...)
+  num_plots = length(plots)
+  
+  layout <- matrix(seq(1, cols * ceiling(num_plots / cols)),
+                   ncol = cols, nrow = ceiling(num_plots / cols))
+  
+  if (num_plots == 1) {
+    print(plots[[1]])
+  } else {
+    grid::grid.newpage()
+    grid::pushViewport(grid::viewport(layout = grid::grid.layout(nrow(layout), ncol(layout))))
+    for (i in 1:num_plots) {
+      # Get the i,j matrix positions of the regions that contain this subplot
+      matchidx <- as.data.table(which(layout == i, arr.ind = TRUE))
+      
+      print(
+        plots[[i]], vp = grid::viewport(
+          layout.pos.row = matchidx$row,
+          layout.pos.col = matchidx$col
+        )
+      )
+    }
+  }
+}
+
+globalVariables(c(
+  "Cluster", "ggplot", "aes", "geom_bar", "coord_flip", "xlab", "ylab", "ggtitle", "theme",
+  "element_blank", "element_text", "V1", "Weight"
+))

R-package/R/xgb.importance.R

@@ -1,110 +1,134 @@
-#' Show importance of features in a model
+#' Importance of features in a model.
 #' 
-#' Create a \code{data.table} of the most important features of a model. 
+#' Creates a \code{data.table} of feature importances in a model.
 #' 
-#' @importFrom data.table data.table
-#' @importFrom data.table setnames
-#' @importFrom data.table :=
-#' @importFrom magrittr %>%
-#' @importFrom Matrix colSums
-#' @importFrom Matrix cBind
-#' @importFrom Matrix sparseVector
-#' 
-#' @param feature_names names of each feature as a \code{character} vector. Can be extracted from a sparse matrix (see example). If model dump already contains feature names, this argument should be \code{NULL}.
-#' @param model generated by the \code{xgb.train} function.
-#' @param data the dataset used for the training step. Will be used with \code{label} parameter for co-occurence computation. More information in \code{Detail} part. This parameter is optional.
-#' @param label the label vetor used for the training step. Will be used with \code{data} parameter for co-occurence computation. More information in \code{Detail} part. This parameter is optional.
-#' @param target a function which returns \code{TRUE} or \code{1} when an observation should be count as a co-occurence and \code{FALSE} or \code{0} otherwise. Default function is provided for computing co-occurences in a binary classification. The \code{target} function should have only one parameter. This parameter will be used to provide each important feature vector after having applied the split condition, therefore these vector will be only made of 0 and 1 only, whatever was the information before. More information in \code{Detail} part. This parameter is optional.
-#'
-#' @return A \code{data.table} of the features used in the model with their average gain (and their weight for boosted tree model) in the model.
+#' @param feature_names character vector of feature names. If the model already
+#'       contains feature names, those would be used when \code{feature_names=NULL} (default value).
+#'       Non-null \code{feature_names} could be provided to override those in the model.
+#' @param model object of class \code{xgb.Booster}.
+#' @param trees (only for the gbtree booster) an integer vector of tree indices that should be included
+#'          into the importance calculation. If set to \code{NULL}, all trees of the model are parsed.
+#'          It could be useful, e.g., in multiclass classification to get feature importances 
+#'          for each class separately. IMPORTANT: the tree index in xgboost models
+#'          is zero-based (e.g., use \code{trees = 0:4} for first 5 trees).
+#' @param data deprecated.
+#' @param label deprecated.
+#' @param target deprecated.
 #'
 #' @details 
-#' This function is for both linear and tree models.
 #' 
-#' \code{data.table} is returned by the function. 
-#' The columns are :
+#' This function works for both linear and tree models.
+#' 
+#' For linear models, the importance is the absolute magnitude of linear coefficients. 
+#' For that reason, in order to obtain a meaningful ranking by importance for a linear model, 
+#' the features need to be on the same scale (which you also would want to do when using either 
+#' L1 or L2 regularization).
+#' 
+#' @return
+#' 
+#' For a tree model, a \code{data.table} with the following columns:
 #' \itemize{
-#'   \item \code{Features} name of the features as provided in \code{feature_names} or already present in the model dump;
-#'   \item \code{Gain} contribution of each feature to the model. For boosted tree model, each gain of each feature of each tree is taken into account, then average per feature to give a vision of the entire model. Highest percentage means important feature to predict the \code{label} used for the training (only available for tree models);
-#'   \item \code{Cover} metric of the number of observation related to this feature (only available for tree models);
-#'   \item \code{Weight} percentage representing the relative number of times a feature have been taken into trees.
+#'   \item \code{Features} names of the features used in the model;
+#'   \item \code{Gain} represents fractional contribution of each feature to the model based on
+#'        the total gain of this feature's splits. Higher percentage means a more important 
+#'        predictive feature.
+#'   \item \code{Cover} metric of the number of observation related to this feature;
+#'   \item \code{Frequency} percentage representing the relative number of times
+#'        a feature have been used in trees.
 #' }
 #' 
-#' If you don't provide \code{feature_names}, index of the features will be used instead.
+#' A linear model's importance \code{data.table} has the following columns:
+#' \itemize{
+#'   \item \code{Features} names of the features used in the model;
+#'   \item \code{Weight} the linear coefficient of this feature;
+#'   \item \code{Class} (only for multiclass models) class label.
+#' }
 #' 
-#' Because the index is extracted from the model dump (made on the C++ side), it starts at 0 (usual in C++) instead of 1 (usual in R).
-#' 
-#' Co-occurence count
-#' ------------------
-#' 
-#' The gain gives you indication about the information of how a feature is important in making a branch of a decision tree more pure. However, with this information only, you can't know if this feature has to be present or not to get a specific classification. In the example code, you may wonder if odor=none should be \code{TRUE} to not eat a mushroom.
-#' 
-#' Co-occurence computation is here to help in understanding this relation between a predictor and a specific class. It will count how many observations are returned as \code{TRUE} by the \code{target} function (see parameters). When you execute the example below, there are 92 times only over the 3140 observations of the train dataset where a mushroom have no odor and can be eaten safely.
-#' 
-#' If you need to remember one thing only: until you want to leave us early, don't eat a mushroom which has no odor :-)
+#' If \code{feature_names} is not provided and \code{model} doesn't have \code{feature_names}, 
+#' index of the features will be used instead. Because the index is extracted from the model dump
+#' (based on C++ code), it starts at 0 (as in C/C++ or Python) instead of 1 (usual in R).
 #' 
 #' @examples
+#' 
+#' # binomial classification using gbtree:
 #' data(agaricus.train, package='xgboost')
+#' bst <- xgboost(data = agaricus.train$data, label = agaricus.train$label, max_depth = 2, 
+#'                eta = 1, nthread = 2, nrounds = 2, objective = "binary:logistic")
+#' xgb.importance(model = bst)
 #' 
-#' bst <- xgboost(data = agaricus.train$data, label = agaricus.train$label, max.depth = 2, 
-#'                eta = 1, nthread = 2, nround = 2,objective = "binary:logistic")
+#' # binomial classification using gblinear:
+#' bst <- xgboost(data = agaricus.train$data, label = agaricus.train$label, booster = "gblinear", 
+#'                eta = 0.3, nthread = 1, nrounds = 20, objective = "binary:logistic")
+#' xgb.importance(model = bst)
 #' 
-#' # agaricus.train$data@@Dimnames[[2]] represents the column names of the sparse matrix.
-#' xgb.importance(agaricus.train$data@@Dimnames[[2]], model = bst)
-#' 
-#' # Same thing with co-occurence computation this time
-#' xgb.importance(agaricus.train$data@@Dimnames[[2]], model = bst, data = agaricus.train$data, label = agaricus.train$label)
+#' # multiclass classification using gbtree:
+#' nclass <- 3
+#' nrounds <- 10
+#' mbst <- xgboost(data = as.matrix(iris[, -5]), label = as.numeric(iris$Species) - 1,
+#'                max_depth = 3, eta = 0.2, nthread = 2, nrounds = nrounds,
+#'                objective = "multi:softprob", num_class = nclass)
+#' # all classes clumped together:
+#' xgb.importance(model = mbst)
+#' # inspect importances separately for each class:
+#' xgb.importance(model = mbst, trees = seq(from=0, by=nclass, length.out=nrounds))
+#' xgb.importance(model = mbst, trees = seq(from=1, by=nclass, length.out=nrounds))
+#' xgb.importance(model = mbst, trees = seq(from=2, by=nclass, length.out=nrounds))
 #' 
+#' # multiclass classification using gblinear:
+#' mbst <- xgboost(data = scale(as.matrix(iris[, -5])), label = as.numeric(iris$Species) - 1,
+#'                booster = "gblinear", eta = 0.2, nthread = 1, nrounds = 15,
+#'                objective = "multi:softprob", num_class = nclass)
+#' xgb.importance(model = mbst)
+#'
 #' @export
-xgb.importance <- function(feature_names = NULL, model = NULL, data = NULL, label = NULL, target = function(x) ( (x + label) == 2)){
-  if (!class(feature_names) %in% c("character", "NULL")) {
-    stop("feature_names: Has to be a vector of character or NULL if the model already contains feature name. Look at this function documentation to see where to get feature names.")
-  }
-
-  if (class(model) != "xgb.Booster") {
-    stop("model: Has to be an object of class xgb.Booster model generaged by the xgb.train function.")
-  }
-
-  if((is.null(data) & !is.null(label)) | (!is.null(data) & is.null(label))) {
-    stop("data/label: Provide the two arguments if you want co-occurence computation or none of them if you are not interested but not one of them only.")
-  }
-
-  if(class(label) == "numeric"){
-    if(sum(label == 0) / length(label) > 0.5) label <- as(label, "sparseVector")
-  }
+xgb.importance <- function(feature_names = NULL, model = NULL, trees = NULL,
+                           data = NULL, label = NULL, target = NULL){
   
-  treeDump <- function(feature_names, text, keepDetail){
-    if(keepDetail) groupBy <- c("Feature", "Split", "MissingNo") else groupBy <- "Feature"
-    xgb.model.dt.tree(feature_names = feature_names, text = text)[,"MissingNo" := Missing == No ][Feature != "Leaf",.(Gain = sum(Quality), Cover = sum(Cover), Frequency = .N), by = groupBy, with = T][,`:=`(Gain = Gain / sum(Gain), Cover = Cover / sum(Cover), Frequency = Frequency / sum(Frequency))][order(Gain, decreasing = T)]
-  }
+  if (!(is.null(data) && is.null(label) && is.null(target)))
+    warning("xgb.importance: parameters 'data', 'label' and 'target' are deprecated")
   
-  linearDump <- function(feature_names, text){
-    weights <- which(text == "weight:") %>% {a =. + 1; text[a:length(text)]} %>% as.numeric
-    if(is.null(feature_names)) feature_names <- seq(to = length(weights))
-    data.table(Feature = feature_names, Weight = weights)
-  }
-
-  model.text.dump <- xgb.dump(model = model, with.stats = T)
+  if (!inherits(model, "xgb.Booster"))
+    stop("model: must be an object of class xgb.Booster")
   
-  if(model.text.dump[2] == "bias:"){
-    result <- model.text.dump %>% linearDump(feature_names, .)
-    if(!is.null(data) | !is.null(label)) warning("data/label: these parameters should only be provided with decision tree based models.")
-  }  else {
-    result <- treeDump(feature_names, text = model.text.dump, keepDetail = !is.null(data))
+  if (is.null(feature_names) && !is.null(model$feature_names))
+    feature_names <- model$feature_names
+  
+  if (!(is.null(feature_names) || is.character(feature_names)))
+    stop("feature_names: Has to be a character vector")
 
-    # Co-occurence computation
-    if(!is.null(data) & !is.null(label) & nrow(result) > 0) {
-      # Take care of missing column
-      a <- data[, result[MissingNo == T,Feature], drop=FALSE] != 0
-      # Bind the two Matrix and reorder columns
-      c <- data[, result[MissingNo == F,Feature], drop=FALSE] %>% cBind(a,.) %>% .[,result[,Feature]]
-      rm(a)
-      # Apply split
-      d <- data[, result[,Feature], drop=FALSE] < as.numeric(result[,Split])
-      apply(c & d, 2, . %>% target %>% sum) -> vec
-
-      result <- result[, "RealCover" := as.numeric(vec), with = F][, "RealCover %" := RealCover / sum(label)][,MissingNo := NULL]
+  model_text_dump <- xgb.dump(model = model, with_stats = TRUE)
+  
+  # linear model
+  if(model_text_dump[2] == "bias:"){
+    weights <- which(model_text_dump == "weight:") %>%
+               {model_text_dump[(. + 1):length(model_text_dump)]} %>%
+               as.numeric
+    
+    num_class <- NVL(model$params$num_class, 1)
+    if(is.null(feature_names)) 
+      feature_names <- seq(to = length(weights) / num_class) - 1
+    if (length(feature_names) * num_class != length(weights))
+      stop("feature_names length does not match the number of features used in the model")
+    
+    result <- if (num_class == 1) {
+      data.table(Feature = feature_names, Weight = weights)[order(-abs(Weight))]
+    } else {
+      data.table(Feature = rep(feature_names, each = num_class),
+                 Weight = weights,
+                 Class = seq_len(num_class) - 1)[order(Class, -abs(Weight))]
     }
+  } else { 
+  # tree model
+    result <- xgb.model.dt.tree(feature_names = feature_names,
+                                text = model_text_dump,
+                                trees = trees)[
+      Feature != "Leaf", .(Gain = sum(Quality), 
+                           Cover = sum(Cover), 
+                           Frequency = .N), by = Feature][
+      ,`:=`(Gain = Gain / sum(Gain), 
+            Cover = Cover / sum(Cover),
+            Frequency = Frequency / sum(Frequency))][
+      order(Gain, decreasing = TRUE)]
   }
   result
 }
@@ -112,4 +136,4 @@ xgb.importance <- function(feature_names = NULL, model = NULL, data = NULL, labe
 # Avoid error messages during CRAN check.
 # The reason is that these variables are never declared
 # They are mainly column names inferred by Data.table...
-globalVariables(c(".", "Feature", "Split", "No", "Missing", "MissingNo", "RealCover"))
+globalVariables(c(".", ".N", "Gain", "Cover", "Frequency", "Feature", "Class"))

R-package/R/xgb.load.R

@@ -1,16 +1,31 @@
 #' Load xgboost model from binary file
 #' 
-#' Load xgboost model from the binary model file
+#' Load xgboost model from the binary model file. 
 #' 
-#' @param modelfile the name of the binary file.
+#' @param modelfile the name of the binary input file.
+#' 
+#' @details 
+#' The input file is expected to contain a model saved in an xgboost-internal binary format
+#' using either \code{\link{xgb.save}} or \code{\link{cb.save.model}} in R, or using some 
+#' appropriate methods from other xgboost interfaces. E.g., a model trained in Python and 
+#' saved from there in xgboost format, could be loaded from R.
+#' 
+#' Note: a model saved as an R-object, has to be loaded using corresponding R-methods,
+#' not \code{xgb.load}.
+#' 
+#' @return 
+#' An object of \code{xgb.Booster} class.
+#' 
+#' @seealso 
+#' \code{\link{xgb.save}}, \code{\link{xgb.Booster.complete}}. 
 #' 
 #' @examples
 #' data(agaricus.train, package='xgboost')
 #' data(agaricus.test, package='xgboost')
 #' train <- agaricus.train
 #' test <- agaricus.test
-#' bst <- xgboost(data = train$data, label = train$label, max.depth = 2, 
-#'                eta = 1, nthread = 2, nround = 2,objective = "binary:logistic")
+#' bst <- xgboost(data = train$data, label = train$label, max_depth = 2, 
+#'                eta = 1, nthread = 2, nrounds = 2,objective = "binary:logistic")
 #' xgb.save(bst, 'xgb.model')
 #' bst <- xgb.load('xgb.model')
 #' pred <- predict(bst, test$data)
@@ -19,13 +34,13 @@ xgb.load <- function(modelfile) {
   if (is.null(modelfile))
     stop("xgb.load: modelfile cannot be NULL")
 
-  handle <- xgb.Booster(modelfile = modelfile)
-  # re-use modelfile if it is raw so we donot need to serialize
+  handle <- xgb.Booster.handle(modelfile = modelfile)
+  # re-use modelfile if it is raw so we do not need to serialize
   if (typeof(modelfile) == "raw") {
     bst <- xgb.handleToBooster(handle, modelfile)
   } else {
     bst <- xgb.handleToBooster(handle, NULL)
   }
-  bst <- xgb.Booster.check(bst)
+  bst <- xgb.Booster.complete(bst, saveraw = TRUE)
   return(bst)
 }

R-package/R/xgb.model.dt.tree.R

@@ -1,153 +1,159 @@
-#' Parse boosted tree model text dump
+#' Parse a boosted tree model text dump
 #' 
-#' Parse a boosted tree model text dump and return a \code{data.table}.
+#' Parse a boosted tree model text dump into a \code{data.table} structure.
 #' 
-#' @importFrom data.table data.table
-#' @importFrom data.table set
-#' @importFrom data.table rbindlist
-#' @importFrom data.table copy
-#' @importFrom data.table :=
-#' @importFrom magrittr %>%
-#' @importFrom magrittr not
-#' @importFrom magrittr add
-#' @importFrom stringr str_extract
-#' @importFrom stringr str_split
-#' @importFrom stringr str_trim
-#' @param feature_names names of each feature as a character vector. Can be extracted from a sparse matrix (see example). If the model already contains feature names, this argument should be \code{NULL} (default value).
-#' @param model object created by the \code{xgb.train} function.
-#' @param text \code{character} vector generated by the \code{xgb.dump} function. Model dump must include the gain per feature and per tree (parameter \code{with.stats = TRUE} in function \code{xgb.dump}).
-#' @param n_first_tree limit the plot to the \code{n} first trees. If set to \code{NULL}, all trees of the model are plotted. Performance can be low depending of the size of the model.
+#' @param feature_names character vector of feature names. If the model already
+#'          contains feature names, those would be used when \code{feature_names=NULL} (default value).
+#'          Non-null \code{feature_names} could be provided to override those in the model.
+#' @param model object of class \code{xgb.Booster}
+#' @param text \code{character} vector previously generated by the \code{xgb.dump} 
+#'          function  (where parameter \code{with_stats = TRUE} should have been set).
+#'          \code{text} takes precedence over \code{model}.
+#' @param trees an integer vector of tree indices that should be parsed.
+#'          If set to \code{NULL}, all trees of the model are parsed.
+#'          It could be useful, e.g., in multiclass classification to get only
+#'          the trees of one certain class. IMPORTANT: the tree index in xgboost models
+#'          is zero-based (e.g., use \code{trees = 0:4} for first 5 trees).
+#' @param use_int_id a logical flag indicating whether nodes in columns "Yes", "No", "Missing" should be
+#'          represented as integers (when FALSE) or as "Tree-Node" character strings (when FALSE).
+#' @param ... currently not used.
 #'
-#' @return A \code{data.table} of the features used in the model with their gain, cover and few other information.
+#' @return 
+#' A \code{data.table} with detailed information about model trees' nodes.
 #'
-#' @details 
-#' General function to convert a text dump of tree model to a \code{data.table}. 
-#' 
-#' The purpose is to help user to explore the model and get a better understanding of it.
-#' 
 #' The columns of the \code{data.table} are:
 #' 
 #' \itemize{
-#' \item \code{ID}: unique identifier of a node ;
-#'  \item \code{Feature}: feature used in the tree to operate a split. When Leaf is indicated, it is the end of a branch ;
-#'  \item \code{Split}: value of the chosen feature where is operated the split ;
-#'  \item \code{Yes}: ID of the feature for the next node in the branch when the split condition is met ;
-#'  \item \code{No}: ID of the feature for the next node in the branch when the split condition is not met ;
-#'  \item \code{Missing}: ID of the feature for the next node in the branch for observation where the feature used for the split are not provided ;
-#'  \item \code{Quality}: it's the gain related to the split in this specific node ;
-#'  \item \code{Cover}: metric to measure the number of observation affected by the split ;
-#'  \item \code{Tree}: ID of the tree. It is included in the main ID ;
-#'  \item \code{Yes.Feature}, \code{No.Feature}, \code{Yes.Cover}, \code{No.Cover}, \code{Yes.Quality} and \code{No.Quality}: data related to the pointer in \code{Yes} or \code{No} column ;
+#'  \item \code{Tree}: integer ID of a tree in a model (zero-based index)
+#'  \item \code{Node}: integer ID of a node in a tree (zero-based index)
+#'  \item \code{ID}: character identifier of a node in a model (only when \code{use_int_id=FALSE})
+#'  \item \code{Feature}: for a branch node, it's a feature id or name (when available);
+#'              for a leaf note, it simply labels it as \code{'Leaf'}
+#'  \item \code{Split}: location of the split for a branch node (split condition is always "less than")
+#'  \item \code{Yes}: ID of the next node when the split condition is met
+#'  \item \code{No}: ID of the next node when the split condition is not met
+#'  \item \code{Missing}: ID of the next node when branch value is missing
+#'  \item \code{Quality}: either the split gain (change in loss) or the leaf value
+#'  \item \code{Cover}: metric related to the number of observation either seen by a split
+#'                      or collected by a leaf during training.
 #' } 
-#'   
+#' 
+#' When \code{use_int_id=FALSE}, columns "Yes", "No", and "Missing" point to model-wide node identifiers
+#' in the "ID" column. When \code{use_int_id=TRUE}, those columns point to node identifiers from 
+#' the corresponding trees in the "Node" column.
+#' 
 #' @examples
+#' # Basic use:
+#' 
 #' data(agaricus.train, package='xgboost')
 #' 
-#' bst <- xgboost(data = agaricus.train$data, label = agaricus.train$label, max.depth = 2, 
-#'                eta = 1, nthread = 2, nround = 2,objective = "binary:logistic")
+#' bst <- xgboost(data = agaricus.train$data, label = agaricus.train$label, max_depth = 2, 
+#'                eta = 1, nthread = 2, nrounds = 2,objective = "binary:logistic")
 #' 
-#' # agaricus.train$data@@Dimnames[[2]] represents the column names of the sparse matrix.
-#' xgb.model.dt.tree(feature_names = agaricus.train$data@@Dimnames[[2]], model = bst)
+#' (dt <- xgb.model.dt.tree(colnames(agaricus.train$data), bst))
 #' 
+#' # This bst model already has feature_names stored with it, so those would be used when 
+#' # feature_names is not set:
+#' (dt <- xgb.model.dt.tree(model = bst))
+#' 
+#' # How to match feature names of splits that are following a current 'Yes' branch:
+#' 
+#' merge(dt, dt[, .(ID, Y.Feature=Feature)], by.x='Yes', by.y='ID', all.x=TRUE)[order(Tree,Node)]
+#'  
 #' @export
-xgb.model.dt.tree <- function(feature_names = NULL, model = NULL, text = NULL, n_first_tree = NULL){
-
-  if (!class(feature_names) %in% c("character", "NULL")) {
-    stop("feature_names: Has to be a vector of character or NULL if the model dump already contains feature name. Look at this function documentation to see where to get feature names.")
-  }
-
-  if (class(model) != "xgb.Booster" & class(text) != "character") {
-    "model: Has to be an object of class xgb.Booster model generaged by the xgb.train function.\n" %>%
-      paste0("text: Has to be a vector of character or NULL if a path to the model dump has already been provided.") %>%
-      stop()
-  }
-
-  if (!class(n_first_tree) %in% c("numeric", "NULL") | length(n_first_tree) > 1) {
-    stop("n_first_tree: Has to be a numeric vector of size 1.")
-  }
-
-  if(is.null(text)){		
-    text <- xgb.dump(model = model, with.stats = T)
+xgb.model.dt.tree <- function(feature_names = NULL, model = NULL, text = NULL,
+                              trees = NULL, use_int_id = FALSE, ...){
+  check.deprecation(...)
+  
+  if (!inherits(model, "xgb.Booster") && !is.character(text)) {
+    stop("Either 'model' must be an object of class xgb.Booster\n",
+         "  or 'text' must be a character vector with the result of xgb.dump\n",
+         "  (or NULL if 'model' was provided).")
   }
   
-  position <- str_match(text, "booster") %>% is.na %>% not %>% which %>% c(length(text) + 1)
-
-  extract <- function(x, pattern)  str_extract(x, pattern) %>% str_split("=") %>% lapply(function(x) x[2] %>% as.numeric) %>% unlist
-
-  n_round <- min(length(position) - 1, n_first_tree)
-
-  addTreeId <- function(x, i) paste(i,x,sep = "-")
-
-  allTrees <- data.table()
-
-  anynumber_regex <- "[-+]?[0-9]*\\.?[0-9]+([eE][-+]?[0-9]+)?"
-  for (i in 1:n_round){
-
-    tree <- text[(position[i] + 1):(position[i + 1] - 1)]
-
-    # avoid tree made of a leaf only (no split)
-    if(length(tree) < 2) next
-
-    treeID <- i - 1
-
-    notLeaf <- str_match(tree, "leaf") %>% is.na
-    leaf <- notLeaf %>% not %>% tree[.]
-    branch <- notLeaf %>% tree[.]
-    idBranch <- str_extract(branch, "\\d*:") %>% str_replace(":", "") %>% addTreeId(treeID)
-    idLeaf <- str_extract(leaf, "\\d*:") %>% str_replace(":", "") %>% addTreeId(treeID)
-    featureBranch <- str_extract(branch, "f\\d*<") %>% str_replace("<", "") %>% str_replace("f", "") %>% as.numeric
-    if(!is.null(feature_names)){
-      featureBranch <- feature_names[featureBranch + 1]
-    }
-    featureLeaf <- rep("Leaf", length(leaf))
-    splitBranch <- str_extract(branch, paste0("<",anynumber_regex,"\\]")) %>% str_replace("<", "") %>% str_replace("\\]", "")
-    splitLeaf <- rep(NA, length(leaf))
-    yesBranch <- extract(branch, "yes=\\d*") %>% addTreeId(treeID)
-    yesLeaf <- rep(NA, length(leaf))
-    noBranch <- extract(branch, "no=\\d*") %>% addTreeId(treeID)
-    noLeaf <- rep(NA, length(leaf))
-    missingBranch <- extract(branch, "missing=\\d+") %>% addTreeId(treeID)
-    missingLeaf <- rep(NA, length(leaf))
-    qualityBranch <- extract(branch, paste0("gain=",anynumber_regex))
-    qualityLeaf <- extract(leaf, paste0("leaf=",anynumber_regex))
-    coverBranch <- extract(branch, "cover=\\d*\\.*\\d*")
-    coverLeaf <- extract(leaf, "cover=\\d*\\.*\\d*")
-    dt <- data.table(ID = c(idBranch, idLeaf), Feature = c(featureBranch, featureLeaf), Split = c(splitBranch, splitLeaf), Yes = c(yesBranch, yesLeaf), No = c(noBranch, noLeaf), Missing = c(missingBranch, missingLeaf), Quality = c(qualityBranch, qualityLeaf), Cover = c(coverBranch, coverLeaf))[order(ID)][,Tree := treeID]
-
-    allTrees <- rbindlist(list(allTrees, dt), use.names = T, fill = F)
+  if (is.null(feature_names) && !is.null(model) && !is.null(model$feature_names))
+    feature_names <- model$feature_names
+  
+  if (!(is.null(feature_names) || is.character(feature_names))) {
+    stop("feature_names: must be a character vector")
   }
+  
+  if (!(is.null(trees) || is.numeric(trees))) {
+    stop("trees: must be a vector of integers.")
+  }
+  
+  if (is.null(text)){
+    text <- xgb.dump(model = model, with_stats = TRUE)
+  }
+  
+  if (length(text) < 2 ||
+      sum(stri_detect_regex(text, 'yes=(\\d+),no=(\\d+)')) < 1) {
+    stop("Non-tree model detected! This function can only be used with tree models.")
+  }
+  
+  position <- which(!is.na(stri_match_first_regex(text, "booster")))
+  
+  add.tree.id <- function(node, tree) if (use_int_id) node else paste(tree, node, sep = "-")
+  
+  anynumber_regex <- "[-+]?[0-9]*\\.?[0-9]+([eE][-+]?[0-9]+)?"
+  
+  td <- data.table(t = text)
+  td[position, Tree := 1L]
+  td[, Tree := cumsum(ifelse(is.na(Tree), 0L, Tree)) - 1L]
+  
+  if (is.null(trees)) {
+    trees <- 0:max(td$Tree)
+  } else {
+    trees <- trees[trees >= 0 & trees <= max(td$Tree)]
+  }
+  td <- td[Tree %in% trees & !grepl('^booster', t)]
+  
+  td[, Node := stri_match_first_regex(t, "(\\d+):")[,2] %>% as.integer ]
+  if (!use_int_id) td[, ID := add.tree.id(Node, Tree)]
+  td[, isLeaf := !is.na(stri_match_first_regex(t, "leaf"))]
 
-  yes <- allTrees[!is.na(Yes), Yes]
-
-  set(allTrees, i = which(allTrees[, Feature] != "Leaf"),
-      j = "Yes.Feature",
-      value = allTrees[ID %in% yes, Feature])
-
-  set(allTrees, i = which(allTrees[, Feature] != "Leaf"),
-      j = "Yes.Cover",
-      value = allTrees[ID %in% yes, Cover])
-
-  set(allTrees, i = which(allTrees[, Feature] != "Leaf"),
-      j = "Yes.Quality",
-      value = allTrees[ID %in% yes, Quality])
-  no <- allTrees[!is.na(No), No]
-
-  set(allTrees, i = which(allTrees[, Feature] != "Leaf"),
-      j = "No.Feature",
-      value = allTrees[ID %in% no, Feature])
-
-  set(allTrees, i = which(allTrees[, Feature] != "Leaf"),
-      j = "No.Cover",
-      value = allTrees[ID %in% no, Cover])
-
-  set(allTrees, i = which(allTrees[, Feature] != "Leaf"),
-      j = "No.Quality",
-      value = allTrees[ID %in% no, Quality])
-
-  allTrees
+  # parse branch lines
+  branch_rx <- paste0("f(\\d+)<(", anynumber_regex, ")\\] yes=(\\d+),no=(\\d+),missing=(\\d+),",
+                      "gain=(", anynumber_regex, "),cover=(", anynumber_regex, ")")
+  branch_cols <- c("Feature", "Split", "Yes", "No", "Missing", "Quality", "Cover")
+  td[isLeaf == FALSE, 
+     (branch_cols) := {
+      # skip some indices with spurious capture groups from anynumber_regex
+      xtr <- stri_match_first_regex(t, branch_rx)[, c(2,3,5,6,7,8,10), drop = FALSE]
+      xtr[, 3:5] <- add.tree.id(xtr[, 3:5], Tree)
+      lapply(seq_len(ncol(xtr)), function(i) xtr[,i])
+    }]
+  # assign feature_names when available
+  if (!is.null(feature_names)) {
+    if (length(feature_names) <= max(as.numeric(td$Feature), na.rm = TRUE))
+      stop("feature_names has less elements than there are features used in the model")
+    td[isLeaf == FALSE, Feature := feature_names[as.numeric(Feature) + 1] ]
+  }
+  
+  # parse leaf lines
+  leaf_rx <- paste0("leaf=(", anynumber_regex, "),cover=(", anynumber_regex, ")")
+  leaf_cols <- c("Feature", "Quality", "Cover")
+  td[isLeaf == TRUE,
+     (leaf_cols) := {
+      xtr <- stri_match_first_regex(t, leaf_rx)[, c(2,4)]
+      c("Leaf", lapply(seq_len(ncol(xtr)), function(i) xtr[,i]))
+    }]
+  
+  # convert some columns to numeric
+  numeric_cols <- c("Split", "Quality", "Cover")
+  td[, (numeric_cols) := lapply(.SD, as.numeric), .SDcols = numeric_cols]
+  if (use_int_id) {
+    int_cols <- c("Yes", "No", "Missing")
+    td[, (int_cols) := lapply(.SD, as.integer), .SDcols = int_cols]
+  }
+  
+  td[, t := NULL]
+  td[, isLeaf := NULL]
+  
+  td[order(Tree, Node)]
 }
 
 # Avoid error messages during CRAN check.
 # The reason is that these variables are never declared
 # They are mainly column names inferred by Data.table...
-globalVariables(c("ID", "Tree", "Yes", ".", ".N", "Feature", "Cover", "Quality", "No", "Gain", "Frequency"))
+globalVariables(c("Tree", "Node", "ID", "Feature", "t", "isLeaf",".SD", ".SDcols"))

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

@@ -1,153 +1,143 @@
-#' Plot multiple graphs at the same time
-#' 
-#' Plot multiple graph aligned by rows and columns.
-#' 
-#' @importFrom data.table data.table
-#' @param cols number of columns
-#' @return NULL
-multiplot <- function(..., cols = 1) {
-  plots <- list(...)
-  numPlots = length(plots)
-  
-  layout <- matrix(seq(1, cols * ceiling(numPlots / cols)),
-                   ncol = cols, nrow = ceiling(numPlots / cols))
-  
-  if (numPlots == 1) {
-    print(plots[[1]])
-  } else {
-    grid::grid.newpage()
-    grid::pushViewport(grid::viewport(layout = grid::grid.layout(nrow(layout), ncol(layout))))
-    for (i in 1:numPlots) {
-      # Get the i,j matrix positions of the regions that contain this subplot
-      matchidx <- as.data.table(which(layout == i, arr.ind = TRUE))
-      
-      print(
-        plots[[i]], vp = grid::viewport(
-          layout.pos.row = matchidx$row,
-          layout.pos.col = matchidx$col
-        )
-      )
-    }
-  }
-}
-
-#' Parse the graph to extract vector of edges
-#' @param element igraph object containing the path from the root to the leaf.
-edge.parser <- function(element) {
-  edges.vector <- igraph::as_ids(element)
-  t <- tail(edges.vector, n = 1)
-  l <- length(edges.vector)
-  list(t,l)
-}
-
-#' Extract path from root to leaf from data.table
-#' @param dt.tree data.table containing the nodes and edges of the trees
-get.paths.to.leaf <- function(dt.tree) {
-  dt.not.leaf.edges <-
-    dt.tree[Feature != "Leaf",.(ID, Yes, Tree)] %>% list(dt.tree[Feature != "Leaf",.(ID, No, Tree)]) %>% rbindlist(use.names = F)
-  
-  trees <- dt.tree[,unique(Tree)]
-  
-  paths <- list()
-  for (tree in trees) {
-    graph <-
-      igraph::graph_from_data_frame(dt.not.leaf.edges[Tree == tree])
-    paths.tmp <-
-      igraph::shortest_paths(graph, from = paste0(tree, "-0"), to = dt.tree[Tree == tree &
-                                                                              Feature == "Leaf", c(ID)])
-    paths <- c(paths, paths.tmp$vpath)
-  }
-  paths
-}
-
 #' Plot model trees deepness
 #'
-#' Generate a graph to plot the distribution of deepness among trees.
-#'
-#' @importFrom data.table data.table
-#' @importFrom data.table rbindlist
-#' @importFrom data.table setnames
-#' @importFrom data.table :=
-#' @importFrom magrittr %>%
-#' @param model dump generated by the \code{xgb.train} function.
-#'
-#' @return Two graphs showing the distribution of the model deepness.
-#'
+#' Visualizes distributions related to depth of tree leafs.
+#' \code{xgb.plot.deepness} uses base R graphics, while \code{xgb.ggplot.deepness} uses the ggplot backend.
+#' 
+#' @param model either an \code{xgb.Booster} model generated by the \code{xgb.train} function
+#'        or a data.table result of the \code{xgb.model.dt.tree} function.
+#' @param plot (base R barplot) whether a barplot should be produced. 
+#'        If FALSE, only a data.table is returned.
+#' @param which which distribution to plot (see details).
+#' @param ... other parameters passed to \code{barplot} or \code{plot}.
+#' 
 #' @details
-#' Display both the number of \code{leaf} and the distribution of \code{weighted observations}
-#' by tree deepness level.
 #' 
-#' The purpose of this function is to help the user to find the best trade-off to set
-#' the \code{max.depth} and \code{min_child_weight} parameters according to the bias / variance trade-off.
-#' 
-#' See \link{xgb.train} for more information about these parameters.
-#'
-#' The graph is made of two parts:
-#'
+#' When \code{which="2x1"}, two distributions with respect to the leaf depth
+#' are plotted on top of each other:
 #' \itemize{
-#'  \item Count: number of leaf per level of deepness;
-#'  \item Weighted cover: noramlized weighted cover per leaf (weighted number of instances).
+#'  \item the distribution of the number of leafs in a tree model at a certain depth;
+#'  \item the distribution of average weighted number of observations ("cover") 
+#'        ending up in leafs at certain depth.
 #' }
+#' Those could be helpful in determining sensible ranges of the \code{max_depth} 
+#' and \code{min_child_weight} parameters.
+#' 
+#' When \code{which="max.depth"} or \code{which="med.depth"}, plots of either maximum or median depth
+#' per tree with respect to tree number are created. And \code{which="med.weight"} allows to see how
+#' a tree's median absolute leaf weight changes through the iterations.
 #'
-#' This function is inspired by the blog post \url{http://aysent.github.io/2015/11/08/random-forest-leaf-visualization.html}
+#' This function was inspired by the blog post
+#' \url{http://aysent.github.io/2015/11/08/random-forest-leaf-visualization.html}.
+#' 
+#' @return
+#' 
+#' Other than producing plots (when \code{plot=TRUE}), the \code{xgb.plot.deepness} function
+#' silently returns a processed data.table where each row corresponds to a terminal leaf in a tree model,
+#' and contains information about leaf's depth, cover, and weight (which is used in calculating predictions).
+#' 
+#' The \code{xgb.ggplot.deepness} silently returns either a list of two ggplot graphs when \code{which="2x1"}
+#' or a single ggplot graph for the other \code{which} options.
 #'
+#' @seealso 
+#' 
+#' \code{\link{xgb.train}}, \code{\link{xgb.model.dt.tree}}.
+#' 
 #' @examples
+#' 
 #' data(agaricus.train, package='xgboost')
 #'
-#' bst <- xgboost(data = agaricus.train$data, label = agaricus.train$label, max.depth = 15,
-#'                  eta = 1, nthread = 2, nround = 30, objective = "binary:logistic",
-#'                  min_child_weight = 50)
+#' # Change max_depth to a higher number to get a more significant result
+#' bst <- xgboost(data = agaricus.train$data, label = agaricus.train$label, max_depth = 6,
+#'                eta = 0.1, nthread = 2, nrounds = 50, objective = "binary:logistic",
+#'                subsample = 0.5, min_child_weight = 2)
 #'
-#' xgb.plot.deepness(model = bst)
+#' xgb.plot.deepness(bst)
+#' xgb.ggplot.deepness(bst)
+#' 
+#' xgb.plot.deepness(bst, which='max.depth', pch=16, col=rgb(0,0,1,0.3), cex=2)
+#' 
+#' xgb.plot.deepness(bst, which='med.weight', pch=16, col=rgb(0,0,1,0.3), cex=2)
 #'
+#' @rdname xgb.plot.deepness
 #' @export
-xgb.plot.deepness <- function(model = NULL) {
-  if (!requireNamespace("ggplot2", quietly = TRUE)) {
-    stop("ggplot2 package is required for plotting the graph deepness.",
-         call. = FALSE)
+xgb.plot.deepness <- function(model = NULL, which = c("2x1", "max.depth", "med.depth", "med.weight"),
+                              plot = TRUE, ...) {
+  
+  if (!(inherits(model, "xgb.Booster") || is.data.table(model)))
+    stop("model: Has to be either an xgb.Booster model generaged by the xgb.train function\n",
+         "or a data.table result of the xgb.importance function")
+
+  if (!requireNamespace("igraph", quietly = TRUE))
+    stop("igraph package is required for plotting the graph deepness.", call. = FALSE)
+
+  which <- match.arg(which)
+  
+  dt_tree <- model
+  if (inherits(model, "xgb.Booster"))
+    dt_tree <- xgb.model.dt.tree(model = model)
+  
+  if (!all(c("Feature", "Tree", "ID", "Yes", "No", "Cover") %in% colnames(dt_tree)))
+    stop("Model tree columns are not as expected!\n",
+         "  Note that this function works only for tree models.")
+  
+  dt_depths <- merge(get.leaf.depth(dt_tree), dt_tree[, .(ID, Cover, Weight = Quality)], by = "ID")
+  setkeyv(dt_depths, c("Tree", "ID"))
+  # count by depth levels, and also calculate average cover at a depth
+  dt_summaries <- dt_depths[, .(.N, Cover = mean(Cover)), Depth]
+  setkey(dt_summaries, "Depth")
+  
+  if (plot) {
+    if (which == "2x1") {
+      op <- par(no.readonly = TRUE)
+      par(mfrow = c(2,1),
+          oma = c(3,1,3,1) + 0.1,
+          mar = c(1,4,1,0) + 0.1)
+
+      dt_summaries[, barplot(N, border = NA, ylab = 'Number of leafs', ...)]
+
+      dt_summaries[, barplot(Cover, border = NA, ylab = "Weighted cover", names.arg = Depth, ...)]
+    
+      title("Model complexity", xlab = "Leaf depth", outer = TRUE, line = 1)
+      par(op)
+    } else if (which == "max.depth") {
+      dt_depths[, max(Depth), Tree][
+                , plot(jitter(V1, amount = 0.1) ~ Tree, ylab = 'Max tree leaf depth', xlab = "tree #", ...)]
+    } else if (which == "med.depth") {
+      dt_depths[, median(as.numeric(Depth)), Tree][
+                , plot(jitter(V1, amount = 0.1) ~ Tree, ylab = 'Median tree leaf depth', xlab = "tree #", ...)]
+    } else if (which == "med.weight") {
+      dt_depths[, median(abs(Weight)), Tree][
+                , plot(V1 ~ Tree, ylab = 'Median absolute leaf weight', xlab = "tree #", ...)]
+    }
   }
-  
-  if (!requireNamespace("igraph", quietly = TRUE)) {
-    stop("igraph package is required for plotting the graph deepness.",
-         call. = FALSE)
-  }
-  
-  if (!requireNamespace("grid", quietly = TRUE)) {
-    stop("grid package is required for plotting the graph deepness.",
-         call. = FALSE)
-  }
-  
-  if (class(model) != "xgb.Booster") {
-    stop("model: Has to be an object of class xgb.Booster model generaged by the xgb.train function.")
-  }
-  
-  dt.tree <- xgb.model.dt.tree(model = model)
-  
-  dt.edge.elements <- data.table()
-  paths <- get.paths.to.leaf(dt.tree)
-  
-  dt.edge.elements <-
-    lapply(paths, edge.parser) %>% rbindlist %>% setnames(c("last.edge", "size")) %>%
-    merge(dt.tree, by.x = "last.edge", by.y = "ID") %>% rbind(dt.edge.elements)
-  
-  dt.edge.summuize <-
-    dt.edge.elements[, .(.N, Cover = sum(Cover)), size][,Cover:= Cover / sum(Cover)]
-  
-  p1 <-
-    ggplot2::ggplot(dt.edge.summuize) + ggplot2::geom_line(ggplot2::aes(x = size, y = N, group = 1)) +
-    ggplot2::xlab("") + ggplot2::ylab("Count") + ggplot2::ggtitle("Model complexity") +
-    ggplot2::theme(
-      plot.title = ggplot2::element_text(lineheight = 0.9, face = "bold"),
-      panel.grid.major.y = ggplot2::element_blank(),
-      axis.ticks = ggplot2::element_blank(),
-      axis.text.x = ggplot2::element_blank()
-    )
-  
-  p2 <- 
-    ggplot2::ggplot(dt.edge.summuize) + ggplot2::geom_line(ggplot2::aes(x =size, y = Cover, group = 1)) + 
-    ggplot2::xlab("From root to leaf path length") + ggplot2::ylab("Weighted cover")
-  
-  multiplot(p1,p2,cols = 1)
+  invisible(dt_depths)
+}
+
+# Extract path depths from root to leaf
+# from data.table containing the nodes and edges of the trees.
+# internal utility function
+get.leaf.depth <- function(dt_tree) {
+  # extract tree graph's edges
+  dt_edges <- rbindlist(list(
+      dt_tree[Feature != "Leaf", .(ID, To = Yes, Tree)],
+      dt_tree[Feature != "Leaf", .(ID, To = No, Tree)]
+    ))
+  # whether "To" is a leaf:
+  dt_edges <- 
+    merge(dt_edges,
+          dt_tree[Feature == "Leaf", .(ID, Leaf = TRUE)],
+          all.x = TRUE, by.x = "To", by.y = "ID")
+  dt_edges[is.na(Leaf), Leaf := FALSE]
+
+  dt_edges[, {
+    graph <- igraph::graph_from_data_frame(.SD[,.(ID, To)])
+    # min(ID) in a tree is a root node
+    paths_tmp <- igraph::shortest_paths(graph, from = min(ID), to = To[Leaf == TRUE])
+    # list of paths to each leaf in a tree
+    paths <- lapply(paths_tmp$vpath, names)
+    # combine into a resulting path lengths table for a tree
+    data.table(Depth = sapply(paths, length), ID = To[Leaf == TRUE])
+  }, by = Tree]
 }
 
 # Avoid error messages during CRAN check.
@@ -155,6 +145,6 @@ xgb.plot.deepness <- function(model = NULL) {
 # They are mainly column names inferred by Data.table...
 globalVariables(
   c(
-    "Feature", "Count", "ggplot", "aes", "geom_bar", "xlab", "ylab", "ggtitle", "theme", "element_blank", "element_text", "ID", "Yes", "No", "Tree"
+    ".N", "N", "Depth", "Quality", "Cover", "Tree", "ID", "Yes", "No", "Feature", "Leaf", "Weight"
   )
 )

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

@@ -1,79 +1,125 @@
-#' Plot feature importance bar graph
+#' Plot feature importance as a bar graph
 #'
-#' Read a data.table containing feature importance details and plot it (for both GLM and Trees).
+#' Represents previously calculated feature importance as a bar graph.
+#' \code{xgb.plot.importance} uses base R graphics, while \code{xgb.ggplot.importance} uses the ggplot backend.
 #'
-#' @importFrom magrittr %>%
-#' @param importance_matrix a \code{data.table} returned by the \code{xgb.importance} function.
-#' @param numberOfClusters a \code{numeric} vector containing the min and the max range of the possible number of clusters of bars.
-#'
-#' @return A \code{ggplot2} bar graph representing each feature by a horizontal bar. Longer is the bar, more important is the feature. Features are classified by importance and clustered by importance. The group is represented through the color of the bar.
+#' @param importance_matrix a \code{data.table} returned by \code{\link{xgb.importance}}.
+#' @param top_n maximal number of top features to include into the plot.
+#' @param measure the name of importance measure to plot. 
+#'        When \code{NULL}, 'Gain' would be used for trees and 'Weight' would be used for gblinear.
+#' @param rel_to_first whether importance values should be represented as relative to the highest ranked feature.
+#'        See Details.
+#' @param left_margin (base R barplot) allows to adjust the left margin size to fit feature names.
+#'        When it is NULL, the existing \code{par('mar')} is used.
+#' @param cex (base R barplot) passed as \code{cex.names} parameter to \code{barplot}.
+#' @param plot (base R barplot) whether a barplot should be produced. 
+#'        If FALSE, only a data.table is returned.
+#' @param n_clusters (ggplot only) a \code{numeric} vector containing the min and the max range 
+#'        of the possible number of clusters of bars.
+#' @param ... other parameters passed to \code{barplot} (except horiz, border, cex.names, names.arg, and las).
 #'
 #' @details
-#' The purpose of this function is to easily represent the importance of each feature of a model.
-#' The function returns a ggplot graph, therefore each of its characteristic can be overriden (to customize it).
-#' In particular you may want to override the title of the graph. To do so, add \code{+ ggtitle("A GRAPH NAME")} next to the value returned by this function.
+#' The graph represents each feature as a horizontal bar of length proportional to the importance of a feature.
+#' Features are shown ranked in a decreasing importance order.
+#' It works for importances from both \code{gblinear} and \code{gbtree} models.
+#' 
+#' When \code{rel_to_first = FALSE}, the values would be plotted as they were in \code{importance_matrix}.
+#' For gbtree model, that would mean being normalized to the total of 1 
+#' ("what is feature's importance contribution relative to the whole model?").
+#' For linear models, \code{rel_to_first = FALSE} would show actual values of the coefficients.
+#' Setting \code{rel_to_first = TRUE} allows to see the picture from the perspective of 
+#' "what is feature's importance contribution relative to the most important feature?"
+#' 
+#' The ggplot-backend method also performs 1-D custering of the importance values, 
+#' with bar colors coresponding to different clusters that have somewhat similar importance values.
+#' 
+#' @return
+#' The \code{xgb.plot.importance} function creates a \code{barplot} (when \code{plot=TRUE})
+#' and silently returns a processed data.table with \code{n_top} features sorted by importance.
+#' 
+#' The \code{xgb.ggplot.importance} function returns a ggplot graph which could be customized afterwards.
+#' E.g., to change the title of the graph, add \code{+ ggtitle("A GRAPH NAME")} to the result.
 #'
+#' @seealso
+#' \code{\link[graphics]{barplot}}.
+#' 
 #' @examples
-#' data(agaricus.train, package='xgboost')
+#' data(agaricus.train)
 #'
-#' #Both dataset are list with two items, a sparse matrix and labels
-#' #(labels = outcome column which will be learned).
-#' #Each column of the sparse Matrix is a feature in one hot encoding format.
+#' bst <- xgboost(data = agaricus.train$data, label = agaricus.train$label, max_depth = 3,
+#'                eta = 1, nthread = 2, nrounds = 2, objective = "binary:logistic")
 #'
-#' bst <- xgboost(data = agaricus.train$data, label = agaricus.train$label, max.depth = 2,
-#'                eta = 1, nthread = 2, nround = 2,objective = "binary:logistic")
-#'
-#' #agaricus.train$data@@Dimnames[[2]] represents the column names of the sparse matrix.
-#' importance_matrix <- xgb.importance(agaricus.train$data@@Dimnames[[2]], model = bst)
-#' xgb.plot.importance(importance_matrix)
+#' importance_matrix <- xgb.importance(colnames(agaricus.train$data), model = bst)
+#' 
+#' xgb.plot.importance(importance_matrix, rel_to_first = TRUE, xlab = "Relative importance")
+#' 
+#' (gg <- xgb.ggplot.importance(importance_matrix, measure = "Frequency", rel_to_first = TRUE))
+#' gg + ggplot2::ylab("Frequency")
 #'
+#' @rdname xgb.plot.importance
 #' @export
-xgb.plot.importance <-
-  function(importance_matrix = NULL, numberOfClusters = c(1:10)) {
-    if (!"data.table" %in% class(importance_matrix))  {
-      stop("importance_matrix: Should be a data.table.")
-    }
-    if (!requireNamespace("ggplot2", quietly = TRUE)) {
-      stop("ggplot2 package is required for plotting the importance", call. = FALSE)
-    }
-    if (!requireNamespace("Ckmeans.1d.dp", quietly = TRUE)) {
-      stop("Ckmeans.1d.dp package is required for plotting the importance", call. = FALSE)
-    }
-    
-    if(isTRUE(all.equal(colnames(importance_matrix), c("Feature", "Gain", "Cover", "Frequency")))){
-      y.axe.name <- "Gain"
-    } else if(isTRUE(all.equal(colnames(importance_matrix), c("Feature", "Weight")))){
-      y.axe.name <- "Weight"
-    } else {
-      stop("Importance matrix is not correct (column names issue)")
-    }
-    
-    # To avoid issues in clustering when co-occurences are used
-    importance_matrix <-
-      importance_matrix[, .(Gain.or.Weight = sum(get(y.axe.name))), by = Feature]
-    
-    clusters <-
-      suppressWarnings(Ckmeans.1d.dp::Ckmeans.1d.dp(importance_matrix[,Gain.or.Weight], numberOfClusters))
-    importance_matrix[,"Cluster":= clusters$cluster %>% as.character]
-    
-    plot <-
-      ggplot2::ggplot(
-        importance_matrix, ggplot2::aes(
-          x = stats::reorder(Feature, Gain.or.Weight), y = Gain.or.Weight, width = 0.05
-        ), environment = environment()
-      ) + ggplot2::geom_bar(ggplot2::aes(fill = Cluster), stat = "identity", position =
-                              "identity") + ggplot2::coord_flip() + ggplot2::xlab("Features") + ggplot2::ylab(y.axe.name) + ggplot2::ggtitle("Feature importance") + ggplot2::theme(
-                                plot.title = ggplot2::element_text(lineheight = .9, face = "bold"), panel.grid.major.y = ggplot2::element_blank()
-                              )
-    
-    return(plot)
+xgb.plot.importance <- function(importance_matrix = NULL, top_n = NULL, measure = NULL, 
+                                rel_to_first = FALSE, left_margin = 10, cex = NULL, plot = TRUE, ...) {
+  check.deprecation(...)
+  if (!is.data.table(importance_matrix))  {
+    stop("importance_matrix: must be a data.table")
   }
 
+  imp_names <- colnames(importance_matrix)
+  if (is.null(measure)) {
+    if (all(c("Feature", "Gain") %in% imp_names)) {
+      measure <- "Gain"
+    } else if (all(c("Feature", "Weight") %in% imp_names)) {
+      measure <- "Weight"
+    } else {
+      stop("Importance matrix column names are not as expected!")
+    }
+  } else {
+    if (!measure %in% imp_names)
+      stop("Invalid `measure`")
+    if (!"Feature" %in% imp_names)
+      stop("Importance matrix column names are not as expected!")
+  }
+  
+  # also aggregate, just in case when the values were not yet summed up by feature
+  importance_matrix <- importance_matrix[, Importance := sum(get(measure)), by = Feature]
+  
+  # make sure it's ordered
+  importance_matrix <- importance_matrix[order(-abs(Importance))]
+  
+  if (!is.null(top_n)) {
+    top_n <- min(top_n, nrow(importance_matrix))
+    importance_matrix <- head(importance_matrix, top_n)
+  }
+  if (rel_to_first) {
+    importance_matrix[, Importance := Importance/max(abs(Importance))]
+  }
+  if (is.null(cex)) {
+    cex <- 2.5/log2(1 + nrow(importance_matrix))
+  }
+  
+  if (plot) {
+    op <- par(no.readonly = TRUE)
+    mar <- op$mar
+    if (!is.null(left_margin))
+      mar[2] <- left_margin
+    par(mar = mar)
+    
+    # reverse the order of rows to have the highest ranked at the top
+    importance_matrix[nrow(importance_matrix):1,
+                      barplot(Importance, horiz = TRUE, border = NA, cex.names = cex,
+                              names.arg = Feature, las = 1, ...)]
+    grid(NULL, NA)
+    # redraw over the grid
+    importance_matrix[nrow(importance_matrix):1,
+                      barplot(Importance, horiz = TRUE, border = NA, add = TRUE)]
+    par(op)
+  }
+  
+  invisible(importance_matrix)
+}
+
 # Avoid error messages during CRAN check.
 # The reason is that these variables are never declared
 # They are mainly column names inferred by Data.table...
-globalVariables(
-  c(
-    "Feature", "Gain.or.Weight", "Cluster", "ggplot", "aes", "geom_bar", "coord_flip", "xlab", "ylab", "ggtitle", "theme", "element_blank", "element_text", "Gain.or.Weight"
-  )
-)
+globalVariables(c("Feature", "Importance"))

R-package/R/xgb.plot.multi.trees.R

@@ -1,65 +1,77 @@
 #' Project all trees on one tree and plot it
-#' 
+#'
 #' Visualization of the ensemble of trees as a single collective unit.
 #'
-#' @importFrom data.table data.table
-#' @importFrom data.table rbindlist
-#' @importFrom data.table setnames
-#' @importFrom data.table :=
-#' @importFrom magrittr %>%
-#' @importFrom stringr str_detect
-#' @importFrom stringr str_extract
-#' 
-#' @param model dump generated by the \code{xgb.train} function.
-#' @param feature_names names of each feature as a \code{character} vector. Can be extracted from a sparse matrix (see example). If model dump already contains feature names, this argument should be \code{NULL}.
-#' @param features.keep number of features to keep in each position of the multi trees.
-#' @param plot.width width in pixels of the graph to produce
-#' @param plot.height height in pixels of the graph to produce
-#' 
-#' @return Two graphs showing the distribution of the model deepness.
+#' @param model produced by the \code{xgb.train} function.
+#' @param feature_names names of each feature as a \code{character} vector.
+#' @param features_keep number of features to keep in each position of the multi trees.
+#' @param plot_width width in pixels of the graph to produce
+#' @param plot_height height in pixels of the graph to produce
+#' @param render a logical flag for whether the graph should be rendered (see Value).
+#' @param ... currently not used
 #' 
 #' @details
-#' 
-#' This function tries to capture the complexity of gradient boosted tree ensemble 
-#' in a cohesive way. 
-#' 
-#' The goal is to improve the interpretability of the model generally seen as black box.
-#' The function is dedicated to boosting applied to decision trees only.
-#' 
-#' The purpose is to move from an ensemble of trees to a single tree only.
-#' 
-#' It takes advantage of the fact that the shape of a binary tree is only defined by 
-#' its deepness (therefore in a boosting model, all trees have the same shape). 
-#' 
+#'
+#' This function tries to capture the complexity of a gradient boosted tree model
+#' in a cohesive way by compressing an ensemble of trees into a single tree-graph representation.
+#' The goal is to improve the interpretability of a model generally seen as black box.
+#'
+#' Note: this function is applicable to tree booster-based models only.
+#'
+#' It takes advantage of the fact that the shape of a binary tree is only defined by
+#' its depth (therefore, in a boosting model, all trees have similar shape).
+#'
 #' Moreover, the trees tend to reuse the same features.
-#' 
-#' The function will project each tree on one, and keep for each position the 
-#' \code{features.keep} first features (based on Gain per feature measure).
-#' 
+#'
+#' The function projects each tree onto one, and keeps for each position the
+#' \code{features_keep} first features (based on the Gain per feature measure).
+#'
 #' This function is inspired by this blog post:
 #' \url{https://wellecks.wordpress.com/2015/02/21/peering-into-the-black-box-visualizing-lambdamart/}
 #'
+#' @return
+#'
+#' When \code{render = TRUE}:
+#' returns a rendered graph object which is an \code{htmlwidget} of class \code{grViz}.
+#' Similar to ggplot objects, it needs to be printed to see it when not running from command line.
+#'
+#' When \code{render = FALSE}:
+#' silently returns a graph object which is of DiagrammeR's class \code{dgr_graph}.
+#' This could be useful if one wants to modify some of the graph attributes
+#' before rendering the graph with \code{\link[DiagrammeR]{render_graph}}.
+#'
 #' @examples
+#'
 #' data(agaricus.train, package='xgboost')
 #'
-#' bst <- xgboost(data = agaricus.train$data, label = agaricus.train$label, max.depth = 15,
-#'                  eta = 1, nthread = 2, nround = 30, objective = "binary:logistic",
-#'                  min_child_weight = 50)
+#' bst <- xgboost(data = agaricus.train$data, label = agaricus.train$label, max_depth = 15,
+#'                eta = 1, nthread = 2, nrounds = 30, objective = "binary:logistic",
+#'                min_child_weight = 50, verbose = 0)
 #'
-#' p <- xgb.plot.multi.trees(model = bst, feature_names = agaricus.train$data@Dimnames[[2]], features.keep = 3)
+#' p <- xgb.plot.multi.trees(model = bst, features_keep = 3)
 #' print(p)
 #'
+#' \dontrun{
+#' # Below is an example of how to save this plot to a file.
+#' # Note that for `export_graph` to work, the DiagrammeRsvg and rsvg packages must also be installed.
+#' library(DiagrammeR)
+#' gr <- xgb.plot.multi.trees(model=bst, features_keep = 3, render=FALSE)
+#' export_graph(gr, 'tree.pdf', width=1500, height=600)
+#' }
+#'
 #' @export
-xgb.plot.multi.trees <- function(model, feature_names = NULL, features.keep = 5, plot.width = NULL, plot.height = NULL){
+xgb.plot.multi.trees <- function(model, feature_names = NULL, features_keep = 5, plot_width = NULL, plot_height = NULL,
+                                 render = TRUE, ...){
+  check.deprecation(...)
   tree.matrix <- xgb.model.dt.tree(feature_names = feature_names, model = model)
-  
+
   # first number of the path represents the tree, then the following numbers are related to the path to follow
   # root init
-  root.nodes <- tree.matrix[str_detect(ID, "\\d+-0"), ID]
-  tree.matrix[ID %in% root.nodes, abs.node.position:=root.nodes]
-  
+  root.nodes <- tree.matrix[stri_detect_regex(ID, "\\d+-0"), ID]
+  tree.matrix[ID %in% root.nodes, abs.node.position := root.nodes]
+
   precedent.nodes <- root.nodes
-  
+
   while(tree.matrix[,sum(is.na(abs.node.position))] > 0) {
     yes.row.nodes <- tree.matrix[abs.node.position %in% precedent.nodes & !is.na(Yes)]
     no.row.nodes <- tree.matrix[abs.node.position %in% precedent.nodes & !is.na(No)]
@@ -71,44 +83,66 @@ xgb.plot.multi.trees <- function(model, feature_names = NULL, features.keep = 5,
     precedent.nodes <- c(yes.nodes.abs.pos, no.nodes.abs.pos)
   }
   
-  tree.matrix[!is.na(Yes),Yes:= paste0(abs.node.position, "_0")]
-  tree.matrix[!is.na(No),No:= paste0(abs.node.position, "_1")]
+  tree.matrix[!is.na(Yes), Yes := paste0(abs.node.position, "_0")]
+  tree.matrix[!is.na(No), No := paste0(abs.node.position, "_1")]
   
+  remove.tree <- . %>% stri_replace_first_regex(pattern = "^\\d+-", replacement = "")
   
+  tree.matrix[,`:=`(abs.node.position = remove.tree(abs.node.position),
+                    Yes = remove.tree(Yes),
+                    No = remove.tree(No))]
   
-  remove.tree <- . %>% str_replace(pattern = "^\\d+-", replacement = "")
+  nodes.dt <- tree.matrix[
+        , .(Quality = sum(Quality))
+        , by = .(abs.node.position, Feature)
+      ][, .(Text = paste0(Feature[1:min(length(Feature), features_keep)],
+                          " (",
+                          format(Quality[1:min(length(Quality), features_keep)], digits=5),
+                          ")") %>%
+                   paste0(collapse = "\n"))
+        , by = abs.node.position]
   
-  tree.matrix[,`:=`(abs.node.position=remove.tree(abs.node.position), Yes=remove.tree(Yes), No=remove.tree(No))]
+  edges.dt <- tree.matrix[Feature != "Leaf", .(abs.node.position, Yes)] %>%
+    list(tree.matrix[Feature != "Leaf",.(abs.node.position, No)]) %>%
+    rbindlist() %>%
+    setnames(c("From", "To")) %>%
+    .[, .N, .(From, To)] %>%
+    .[, N:=NULL]
   
-  nodes.dt <- tree.matrix[,.(Quality = sum(Quality)),by = .(abs.node.position, Feature)][,.(Text =paste0(Feature[1:min(length(Feature), features.keep)], " (", Quality[1:min(length(Quality), features.keep)], ")") %>% paste0(collapse = "\n")), by=abs.node.position]
-  edges.dt <- tree.matrix[Feature != "Leaf",.(abs.node.position, Yes)] %>% list(tree.matrix[Feature != "Leaf",.(abs.node.position, No)]) %>% rbindlist() %>% setnames(c("From", "To")) %>% .[,.N,.(From, To)] %>% .[,N:=NULL]
-  
-  nodes <- DiagrammeR::create_nodes(nodes = nodes.dt[,abs.node.position],
-                                    label = nodes.dt[,Text],
-                                    style = "filled",
-                                    color = "DimGray",
-                                    fillcolor= "Beige",
-                                    shape = "oval",
-                                    fontname = "Helvetica"
+  nodes <- DiagrammeR::create_node_df(
+    n = nrow(nodes.dt),
+    label = nodes.dt[,Text]
   )
   
-  edges <- DiagrammeR::create_edges(from = edges.dt[,From],
-                                    to = edges.dt[,To],
-                                    color = "DimGray", 
-                                    arrowsize = "1.5", 
-                                    arrowhead = "vee",
-                                    fontname = "Helvetica",
-                                    rel = "leading_to")
+  edges <- DiagrammeR::create_edge_df(
+    from = match(edges.dt[,From], nodes.dt[,abs.node.position]),
+    to = match(edges.dt[,To], nodes.dt[,abs.node.position]),
+    rel = "leading_to")
   
-  graph <- DiagrammeR::create_graph(nodes_df = nodes,
-                                    edges_df = edges,
-                                    graph_attrs = "rankdir = LR")
-  
-  DiagrammeR::render_graph(graph, width = plot.width, height = plot.height)  
+  graph <- DiagrammeR::create_graph(
+      nodes_df = nodes,
+      edges_df = edges,
+      attr_theme = NULL
+      ) %>%
+    DiagrammeR::add_global_graph_attrs(
+      attr_type = "graph",
+      attr  = c("layout", "rankdir"),
+      value = c("dot", "LR")
+      ) %>%
+    DiagrammeR::add_global_graph_attrs(
+      attr_type = "node",
+      attr  = c("color", "fillcolor", "style", "shape", "fontname"),
+      value = c("DimGray", "beige", "filled", "rectangle", "Helvetica")
+      ) %>%
+    DiagrammeR::add_global_graph_attrs(
+      attr_type = "edge",
+      attr  = c("color", "arrowsize", "arrowhead", "fontname"),
+      value = c("DimGray", "1.5", "vee", "Helvetica"))
+
+  if (!render) return(invisible(graph))
+
+  DiagrammeR::render_graph(graph, width = plot_width, height = plot_height)
 }
 
-globalVariables(
-  c(
-    "Feature", "no.nodes.abs.pos", "ID", "Yes", "No", "Tree", "yes.nodes.abs.pos", "abs.node.position"
-  )
-)
+globalVariables(c(".N", "N", "From", "To", "Text", "Feature", "no.nodes.abs.pos",
+                  "ID", "Yes", "No", "Tree", "yes.nodes.abs.pos", "abs.node.position"))

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

@@ -0,0 +1,217 @@
+#' SHAP contribution dependency plots
+#'
+#' Visualizing the SHAP feature contribution to prediction dependencies on feature value.
+#' 
+#' @param data data as a \code{matrix} or \code{dgCMatrix}.
+#' @param shap_contrib a matrix of SHAP contributions that was computed earlier for the above 
+#'          \code{data}. When it is NULL, it is computed internally using \code{model} and \code{data}.
+#' @param features a vector of either column indices or of feature names to plot. When it is NULL,
+#'          feature importance is calculated, and \code{top_n} high ranked features are taken.
+#' @param top_n when \code{features} is NULL, top_n [1, 100] most important features in a model are taken.
+#' @param model an \code{xgb.Booster} model. It has to be provided when either \code{shap_contrib}
+#'          or \code{features} is missing.
+#' @param trees passed to \code{\link{xgb.importance}} when \code{features = NULL}.
+#' @param target_class is only relevant for multiclass models. When it is set to a 0-based class index,
+#'          only SHAP contributions for that specific class are used.
+#'          If it is not set, SHAP importances are averaged over all classes.
+#' @param approxcontrib passed to \code{\link{predict.xgb.Booster}} when \code{shap_contrib = NULL}.
+#' @param subsample a random fraction of data points to use for plotting. When it is NULL,
+#'          it is set so that up to 100K data points are used.
+#' @param n_col a number of columns in a grid of plots.
+#' @param col color of the scatterplot markers.
+#' @param pch scatterplot marker.
+#' @param discrete_n_uniq a maximal number of unique values in a feature to consider it as discrete.
+#' @param discrete_jitter an \code{amount} parameter of jitter added to discrete features' positions.
+#' @param ylab a y-axis label in 1D plots.
+#' @param plot_NA whether the contributions of cases with missing values should also be plotted.
+#' @param col_NA a color of marker for missing value contributions.
+#' @param pch_NA a marker type for NA values.
+#' @param pos_NA a relative position of the x-location where NA values are shown:
+#'          \code{min(x) + (max(x) - min(x)) * pos_NA}.
+#' @param plot_loess whether to plot loess-smoothed curves. The smoothing is only done for features with
+#'          more than 5 distinct values.
+#' @param col_loess a color to use for the loess curves.
+#' @param span_loess the \code{span} paramerer in \code{\link[stats]{loess}}'s call.
+#' @param which whether to do univariate or bivariate plotting. NOTE: only 1D is implemented so far.
+#' @param plot whether a plot should be drawn. If FALSE, only a lits of matrices is returned.
+#' @param ... other parameters passed to \code{plot}.
+#' 
+#' @details
+#' 
+#' These scatterplots represent how SHAP feature contributions depend of feature values.
+#' The similarity to partial dependency plots is that they also give an idea for how feature values
+#' affect predictions. However, in partial dependency plots, we usually see marginal dependencies
+#' of model prediction on feature value, while SHAP contribution dependency plots display the estimated
+#' contributions of a feature to model prediction for each individual case.
+#' 
+#' When \code{plot_loess = TRUE} is set, feature values are rounded to 3 significant digits and
+#' weighted LOESS is computed and plotted, where weights are the numbers of data points
+#' at each rounded value.
+#' 
+#' Note: SHAP contributions are shown on the scale of model margin. E.g., for a logistic binomial objective,
+#' the margin is prediction before a sigmoidal transform into probability-like values.
+#' Also, since SHAP stands for "SHapley Additive exPlanation" (model prediction = sum of SHAP
+#' contributions for all features + bias), depending on the objective used, transforming SHAP
+#' contributions for a feature from the marginal to the prediction space is not necessarily
+#' a meaningful thing to do.
+#' 
+#' @return
+#' 
+#' In addition to producing plots (when \code{plot=TRUE}), it silently returns a list of two matrices:
+#' \itemize{
+#'  \item \code{data} the values of selected features;
+#'  \item \code{shap_contrib} the contributions of selected features.
+#' }
+#'
+#' @references
+#'
+#' Scott M. Lundberg, Su-In Lee, "A Unified Approach to Interpreting Model Predictions", NIPS Proceedings 2017, \url{https://arxiv.org/abs/1705.07874}
+#'
+#' Scott M. Lundberg, Su-In Lee, "Consistent feature attribution for tree ensembles", \url{https://arxiv.org/abs/1706.06060}
+#'
+#' @examples
+#' 
+#' data(agaricus.train, package='xgboost')
+#' data(agaricus.test, package='xgboost')
+#'
+#' bst <- xgboost(agaricus.train$data, agaricus.train$label, nrounds = 50, 
+#'                eta = 0.1, max_depth = 3, subsample = .5,
+#'                method = "hist", objective = "binary:logistic", nthread = 2, verbose = 0)
+#'
+#' xgb.plot.shap(agaricus.test$data, model = bst, features = "odor=none")
+#' contr <- predict(bst, agaricus.test$data, predcontrib = TRUE)
+#' xgb.plot.shap(agaricus.test$data, contr, model = bst, top_n = 12, n_col = 3)
+#'
+#' # multiclass example - plots for each class separately:
+#' nclass <- 3
+#' nrounds <- 20
+#' x <- as.matrix(iris[, -5])
+#' set.seed(123)
+#' is.na(x[sample(nrow(x) * 4, 30)]) <- TRUE # introduce some missing values
+#' mbst <- xgboost(data = x, label = as.numeric(iris$Species) - 1, nrounds = nrounds,
+#'                 max_depth = 2, eta = 0.3, subsample = .5, nthread = 2,
+#'                 objective = "multi:softprob", num_class = nclass, verbose = 0)
+#' trees0 <- seq(from=0, by=nclass, length.out=nrounds)
+#' col <- rgb(0, 0, 1, 0.5)
+#' xgb.plot.shap(x, model = mbst, trees = trees0, target_class = 0, top_n = 4,
+#'               n_col = 2, col = col, pch = 16, pch_NA = 17)
+#' xgb.plot.shap(x, model = mbst, trees = trees0 + 1, target_class = 1, top_n = 4,
+#'               n_col = 2, col = col, pch = 16, pch_NA = 17)
+#' xgb.plot.shap(x, model = mbst, trees = trees0 + 2, target_class = 2, top_n = 4,
+#'               n_col = 2, col = col, pch = 16, pch_NA = 17)
+#' 
+#' @rdname xgb.plot.shap
+#' @export
+xgb.plot.shap <- function(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, 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, ...) {
+  
+  if (!is.matrix(data) && !inherits(data, "dgCMatrix"))
+    stop("data: must be either matrix or dgCMatrix")
+
+  if (is.null(shap_contrib) && (is.null(model) || !inherits(model, "xgb.Booster")))
+    stop("when shap_contrib is not provided, one must provide an xgb.Booster model")
+
+  if (is.null(features) && (is.null(model) || !inherits(model, "xgb.Booster")))
+    stop("when features are not provided, one must provide an xgb.Booster model to rank the features")
+
+  if (!is.null(shap_contrib) &&
+      (!is.matrix(shap_contrib) || nrow(shap_contrib) != nrow(data) || ncol(shap_contrib) != ncol(data) + 1))
+    stop("shap_contrib is not compatible with the provided data")
+  
+  nsample <- if (is.null(subsample)) min(100000, nrow(data)) else as.integer(subsample * nrow(data))
+  idx <- sample(1:nrow(data), nsample)
+  data <- data[idx,]
+
+  if (is.null(shap_contrib)) {
+    shap_contrib <- predict(model, data, predcontrib = TRUE, approxcontrib = approxcontrib)
+  } else {
+    shap_contrib <- shap_contrib[idx,]
+  }
+
+  which <- match.arg(which)
+  if (which == "2d")
+    stop("2D plots are not implemented yet")
+
+  if (is.null(features)) {
+    imp <- xgb.importance(model = model, trees = trees)
+    top_n <- as.integer(top_n[1])
+    if (top_n < 1 && top_n > 100)
+      stop("top_n: must be an integer within [1, 100]")
+    features <- imp$Feature[1:min(top_n, NROW(imp))]
+  }
+  
+  if (is.character(features)) {
+    if (is.null(colnames(data)))
+      stop("Either provide `data` with column names or provide `features` as column indices")
+    features <- match(features, colnames(data))
+  }
+  
+  if (n_col > length(features)) n_col <- length(features)
+
+  if (is.list(shap_contrib)) { # multiclass: either choose a class or merge
+    shap_contrib <- if (!is.null(target_class)) shap_contrib[[target_class + 1]]
+                    else Reduce("+", lapply(shap_contrib, abs))
+  }
+
+  shap_contrib <- shap_contrib[, features, drop = FALSE]
+  data <- data[, features, drop = FALSE]
+  cols <- colnames(data)
+  if (is.null(cols)) cols <- colnames(shap_contrib)
+  if (is.null(cols)) cols <- paste0('X', 1:ncol(data))
+  colnames(data) <- cols
+  colnames(shap_contrib) <- cols
+  
+  if (plot && which == "1d") {
+    op <- par(mfrow = c(ceiling(length(features) / n_col), n_col),
+              oma = c(0,0,0,0) + 0.2,
+              mar = c(3.5,3.5,0,0) + 0.1,
+              mgp = c(1.7, 0.6, 0))
+    for (f in cols) {
+      ord <- order(data[, f])
+      x <- data[, f][ord]
+      y <- shap_contrib[, f][ord]
+      x_lim <- range(x, na.rm = TRUE)
+      y_lim <- range(y, na.rm = TRUE)
+      do_na <- plot_NA && any(is.na(x))
+      if (do_na) {
+        x_range <- diff(x_lim)
+        loc_na <- min(x, na.rm = TRUE) + x_range * pos_NA
+        x_lim <- range(c(x_lim, loc_na))
+      }
+      x_uniq <- unique(x)
+      x2plot <- x
+      # add small jitter for discrete features with <= 5 distinct values
+      if (length(x_uniq) <= discrete_n_uniq)
+        x2plot <- jitter(x, amount = discrete_jitter * min(diff(x_uniq), na.rm = TRUE))
+      plot(x2plot, y, pch = pch, xlab = f, col = col, xlim = x_lim, ylim = y_lim, ylab = ylab, ...)
+      grid()
+      if (plot_loess) {
+        # compress x to 3 digits, and mean-aggredate y
+        zz <- data.table(x = signif(x, 3), y)[, .(.N, y=mean(y)), x]
+        if (nrow(zz) <= 5) {
+          lines(zz$x, zz$y, col = col_loess)
+        } else {
+          lo <- stats::loess(y ~ x, data = zz, weights = zz$N, span = span_loess)
+          zz$y_lo <- predict(lo, zz, type = "link")
+          lines(zz$x, zz$y_lo, col = col_loess)
+        }
+      }
+      if (do_na) {
+        i_na <- which(is.na(x))
+        x_na <- rep(loc_na, length(i_na))
+        x_na <- jitter(x_na, amount = x_range * 0.01)
+        points(x_na, y[i_na], pch = pch_NA, col = col_NA)
+      }
+    }
+    par(op)
+  }
+  if (plot && which == "2d") {
+    # TODO
+  }
+  invisible(list(data = data, shap_contrib = shap_contrib))
+}

R-package/R/xgb.plot.tree.R

@@ -2,83 +2,137 @@
 #' 
 #' Read a tree model text dump and plot the model. 
 #' 
-#' @importFrom data.table data.table
-#' @importFrom data.table :=
-#' @importFrom magrittr %>%
-#' @param feature_names names of each feature as a \code{character} vector. Can be extracted from a sparse matrix (see example). If model dump already contains feature names, this argument should be \code{NULL}.
-#' @param model generated by the \code{xgb.train} function. Avoid the creation of a dump file.
-#' @param n_first_tree limit the plot to the n first trees. If \code{NULL}, all trees of the model are plotted. Performance can be low for huge models.
-#' @param plot.width  the width of the diagram in pixels.
-#' @param plot.height	the height of the diagram in pixels.
-#'
-#' @return A \code{DiagrammeR} of the model.
+#' @param feature_names names of each feature as a \code{character} vector.
+#' @param model produced by the \code{xgb.train} function.
+#' @param trees an integer vector of tree indices that should be visualized.
+#'          If set to \code{NULL}, all trees of the model are included.
+#'          IMPORTANT: the tree index in xgboost model is zero-based
+#'          (e.g., use \code{trees = 0:2} for the first 3 trees in a model).
+#' @param plot_width  the width of the diagram in pixels.
+#' @param plot_height	the height of the diagram in pixels.
+#' @param render a logical flag for whether the graph should be rendered (see Value).
+#' @param show_node_id a logical flag for whether to show node id's in the graph.
+#' @param ... currently not used.
 #'
 #' @details 
 #' 
 #' The content of each node is organised that way:
 #' 
 #' \itemize{
-#'  \item \code{feature} value;
-#'  \item \code{cover}: the sum of second order gradient of training data classified to the leaf, if it is square loss, this simply corresponds to the number of instances in that branch. Deeper in the tree a node is, lower this metric will be;
-#'  \item \code{gain}: metric the importance of the node in the model.
+#'  \item Feature name.
+#'  \item \code{Cover}: The sum of second order gradient of training data classified to the leaf.
+#'        If it is square loss, this simply corresponds to the number of instances seen by a split
+#'        or collected by a leaf during training.
+#'        The deeper in the tree a node is, the lower this metric will be.
+#'  \item \code{Gain} (for split nodes): the information gain metric of a split
+#'        (corresponds to the importance of the node in the model).
+#'  \item \code{Value} (for leafs): the margin value that the leaf may contribute to prediction.
 #' } 
+#' The tree root nodes also indicate the Tree index (0-based).
 #' 
-#' The function uses \href{http://www.graphviz.org/}{GraphViz} library for that purpose.
+#' The "Yes" branches are marked by the "< split_value" label.
+#' The branches that also used for missing values are marked as bold
+#' (as in "carrying extra capacity").
+#' 
+#' This function uses \href{http://www.graphviz.org/}{GraphViz} as a backend of DiagrammeR.
 #'  
+#' @return
+#' 
+#' When \code{render = TRUE}:
+#' returns a rendered graph object which is an \code{htmlwidget} of class \code{grViz}.
+#' Similar to ggplot objects, it needs to be printed to see it when not running from command line.
+#' 
+#' When \code{render = FALSE}:
+#' silently returns a graph object which is of DiagrammeR's class \code{dgr_graph}.
+#' This could be useful if one wants to modify some of the graph attributes
+#' before rendering the graph with \code{\link[DiagrammeR]{render_graph}}.
+#'
 #' @examples
 #' data(agaricus.train, package='xgboost')
 #' 
-#' bst <- xgboost(data = agaricus.train$data, label = agaricus.train$label, max.depth = 2, 
-#'                eta = 1, nthread = 2, nround = 2,objective = "binary:logistic")
+#' bst <- xgboost(data = agaricus.train$data, label = agaricus.train$label, max_depth = 3,
+#'                eta = 1, nthread = 2, nrounds = 2,objective = "binary:logistic")
+#' # plot all the trees
+#' xgb.plot.tree(model = bst)
+#' # plot only the first tree and display the node ID:
+#' xgb.plot.tree(model = bst, trees = 0, show_node_id = TRUE)
 #' 
-#' # agaricus.train$data@@Dimnames[[2]] represents the column names of the sparse matrix.
-#' xgb.plot.tree(feature_names = agaricus.train$data@@Dimnames[[2]], model = bst)
+#' \dontrun{
+#' # Below is an example of how to save this plot to a file. 
+#' # Note that for `export_graph` to work, the DiagrammeRsvg and rsvg packages must also be installed.
+#' library(DiagrammeR)
+#' gr <- xgb.plot.tree(model=bst, trees=0:1, render=FALSE)
+#' export_graph(gr, 'tree.pdf', width=1500, height=1900)
+#' export_graph(gr, 'tree.png', width=1500, height=1900)
+#' }
 #' 
 #' @export
-xgb.plot.tree <- function(feature_names = NULL, model = NULL, n_first_tree = NULL, plot.width = NULL, plot.height = NULL){
-
-  if (class(model) != "xgb.Booster") {
-    stop("model: Has to be an object of class xgb.Booster model generaged by the xgb.train function.")
+xgb.plot.tree <- function(feature_names = NULL, model = NULL, trees = NULL, plot_width = NULL, plot_height = NULL,
+                          render = TRUE, show_node_id = FALSE, ...){
+  check.deprecation(...)
+  if (!inherits(model, "xgb.Booster")) {
+    stop("model: Has to be an object of class xgb.Booster")
   }
 
   if (!requireNamespace("DiagrammeR", quietly = TRUE)) {
     stop("DiagrammeR package is required for xgb.plot.tree", call. = FALSE)
   }
   
-  allTrees <- xgb.model.dt.tree(feature_names = feature_names, model = model, n_first_tree = n_first_tree)
-  
-  allTrees[, label:= paste0(Feature, "\nCover: ", Cover, "\nGain: ", Quality)]
-  allTrees[, shape:= "rectangle"][Feature == "Leaf", shape:= "oval"]
-  allTrees[, filledcolor:= "Beige"][Feature == "Leaf", filledcolor:= "Khaki"]
-  
-  # rev is used to put the first tree on top.
-  nodes <- DiagrammeR::create_nodes(nodes = allTrees[,ID] %>% rev,
-                 label = allTrees[,label] %>% rev,
-                 style = "filled",
-                 color = "DimGray",
-                 fillcolor= allTrees[,filledcolor] %>% rev,
-                 shape = allTrees[,shape] %>% rev,
-                 data = allTrees[,Feature] %>% rev,
-                 fontname = "Helvetica"
-                 )
-  
-  edges <- DiagrammeR::create_edges(from = allTrees[Feature != "Leaf", c(ID)] %>% rep(2),
-                        to = allTrees[Feature != "Leaf", c(Yes, No)],
-                        label = allTrees[Feature != "Leaf", paste("<",Split)] %>% c(rep("",nrow(allTrees[Feature != "Leaf"]))),
-                        color = "DimGray", 
-                        arrowsize = "1.5", 
-                        arrowhead = "vee",
-                        fontname = "Helvetica",
-                        rel = "leading_to")
+  dt <- xgb.model.dt.tree(feature_names = feature_names, model = model, trees = trees)
 
-  graph <- DiagrammeR::create_graph(nodes_df = nodes,
-                        edges_df = edges,
-                        graph_attrs = "rankdir = LR")
+  dt[, label:= paste0(Feature, "\nCover: ", Cover, ifelse(Feature == "Leaf", "\nValue: ", "\nGain: "), Quality)]
+  if (show_node_id)
+    dt[, label := paste0(ID, ": ", label)]
+  dt[Node == 0, label := paste0("Tree ", Tree, "\n", label)]
+  dt[, shape:= "rectangle"][Feature == "Leaf", shape:= "oval"]
+  dt[, filledcolor:= "Beige"][Feature == "Leaf", filledcolor:= "Khaki"]
+  # in order to draw the first tree on top:
+  dt <- dt[order(-Tree)]
   
-  DiagrammeR::render_graph(graph, width = plot.width, height = plot.height)
+  nodes <- DiagrammeR::create_node_df(
+    n         = nrow(dt),
+    ID        = dt$ID,
+    label     = dt$label,
+    fillcolor = dt$filledcolor,
+    shape     = dt$shape,
+    data      = dt$Feature,
+    fontcolor = "black")
+  
+  edges <- DiagrammeR::create_edge_df(
+    from  = match(dt[Feature != "Leaf", c(ID)] %>% rep(2), dt$ID),
+    to    = match(dt[Feature != "Leaf", c(Yes, No)], dt$ID),
+    label = dt[Feature != "Leaf", paste("<", Split)] %>%
+            c(rep("", nrow(dt[Feature != "Leaf"]))),
+    style = dt[Feature != "Leaf", ifelse(Missing == Yes, "bold", "solid")] %>%
+            c(dt[Feature != "Leaf", ifelse(Missing == No, "bold", "solid")]),
+    rel   = "leading_to")
+
+  graph <- DiagrammeR::create_graph(
+      nodes_df = nodes,
+      edges_df = edges,
+      attr_theme = NULL
+      ) %>%
+    DiagrammeR::add_global_graph_attrs(
+      attr_type = "graph",
+      attr  = c("layout", "rankdir"),
+      value = c("dot", "LR")
+      ) %>%
+    DiagrammeR::add_global_graph_attrs(
+      attr_type = "node",
+      attr  = c("color", "style", "fontname"),
+      value = c("DimGray", "filled", "Helvetica")
+      ) %>%
+    DiagrammeR::add_global_graph_attrs(
+      attr_type = "edge",
+      attr  = c("color", "arrowsize", "arrowhead", "fontname"),
+      value = c("DimGray", "1.5", "vee", "Helvetica"))
+  
+  if (!render) return(invisible(graph))
+  
+  DiagrammeR::render_graph(graph, width = plot_width, height = plot_height)
 }
 
 # Avoid error messages during CRAN check.
 # The reason is that these variables are never declared
 # They are mainly column names inferred by Data.table...
-globalVariables(c("Feature", "ID", "Cover", "Quality", "Split", "Yes", "No", ".", "shape", "filledcolor", "label"))
+globalVariables(c("Feature", "ID", "Cover", "Quality", "Split", "Yes", "No", "Missing", ".", "shape", "filledcolor", "label"))

R-package/R/xgb.save.R

@@ -1,31 +1,42 @@
 #' Save xgboost model to binary file
 #' 
-#' Save xgboost model from xgboost or xgb.train
+#' Save xgboost model to a file in binary format.
 #' 
-#' @param model the model object.
-#' @param fname the name of the binary file.
+#' @param model model object of \code{xgb.Booster} class.
+#' @param fname name of the file to write.
+#' 
+#' @details 
+#' This methods allows to save a model in an xgboost-internal binary format which is universal 
+#' among the various xgboost interfaces. In R, the saved model file could be read-in later
+#' using either the \code{\link{xgb.load}} function or the \code{xgb_model} parameter 
+#' of \code{\link{xgb.train}}.
+#' 
+#' Note: a model can also be saved as an R-object (e.g., by using \code{\link[base]{readRDS}} 
+#' or \code{\link[base]{save}}). However, it would then only be compatible with R, and 
+#' corresponding R-methods would need to be used to load it.
+#' 
+#' @seealso 
+#' \code{\link{xgb.load}}, \code{\link{xgb.Booster.complete}}. 
 #' 
 #' @examples
 #' data(agaricus.train, package='xgboost')
 #' data(agaricus.test, package='xgboost')
 #' train <- agaricus.train
 #' test <- agaricus.test
-#' bst <- xgboost(data = train$data, label = train$label, max.depth = 2, 
-#'                eta = 1, nthread = 2, nround = 2,objective = "binary:logistic")
+#' bst <- xgboost(data = train$data, label = train$label, max_depth = 2, 
+#'                eta = 1, nthread = 2, nrounds = 2,objective = "binary:logistic")
 #' xgb.save(bst, 'xgb.model')
 #' bst <- xgb.load('xgb.model')
 #' pred <- predict(bst, test$data)
 #' @export
 xgb.save <- function(model, fname) {
-  if (typeof(fname) != "character") {
-    stop("xgb.save: fname must be character")
+  if (typeof(fname) != "character")
+    stop("fname must be character")
+  if (!inherits(model, "xgb.Booster")) {
+    stop("model must be xgb.Booster.",
+         if (inherits(model, "xgb.DMatrix")) " Use xgb.DMatrix.save to save an xgb.DMatrix object." else "")
   }
-  if (class(model) == "xgb.Booster") {
-    model <- xgb.Booster.check(model)
-    .Call("XGBoosterSaveModel_R", model$handle, fname, PACKAGE = "xgboost")
-    return(TRUE)
-  }
-  stop("xgb.save: the input must be xgb.Booster. Use xgb.DMatrix.save to save
-       xgb.DMatrix object.")
-  return(FALSE)
+  model <- xgb.Booster.complete(model, saveraw = FALSE)
+  .Call(XGBoosterSaveModel_R, model$handle, fname[1])
+  return(TRUE)
 }

R-package/R/xgb.save.raw.R

@@ -10,20 +10,14 @@
 #' data(agaricus.test, package='xgboost')
 #' train <- agaricus.train
 #' test <- agaricus.test
-#' bst <- xgboost(data = train$data, label = train$label, max.depth = 2, 
-#'                eta = 1, nthread = 2, nround = 2,objective = "binary:logistic")
+#' bst <- xgboost(data = train$data, label = train$label, max_depth = 2, 
+#'                eta = 1, nthread = 2, nrounds = 2,objective = "binary:logistic")
 #' raw <- xgb.save.raw(bst)
 #' bst <- xgb.load(raw)
 #' pred <- predict(bst, test$data)
+#'
 #' @export
 xgb.save.raw <- function(model) {
-  if (class(model) == "xgb.Booster"){
-    model <- model$handle
-  }
-  if (class(model) == "xgb.Booster.handle") {
-    raw <- .Call("XGBoosterModelToRaw_R", model, PACKAGE = "xgboost")
-    return(raw)
-  }
-  stop("xgb.raw: the input must be xgb.Booster.handle. Use xgb.DMatrix.save to save
-       xgb.DMatrix object.")
+  model <- xgb.get.handle(model)
+  .Call(XGBoosterModelToRaw_R, model)
 }

R-package/R/xgb.train.R

@@ -1,14 +1,16 @@
 #' eXtreme Gradient Boosting Training
 #' 
-#' An advanced interface for training xgboost model. Look at \code{\link{xgboost}} function for a simpler interface.
+#' \code{xgb.train} is an advanced interface for training an xgboost model.
+#' The \code{xgboost} function is a simpler wrapper for \code{xgb.train}.
 #'
 #' @param params the list of parameters. 
+#'        The complete list of parameters is available at \url{http://xgboost.readthedocs.io/en/latest/parameter.html}.
+#'        Below is a shorter summary:
 #' 
 #' 1. General Parameters
 #' 
 #' \itemize{
-#'   \item \code{booster} which booster to use, can be \code{gbtree} or \code{gblinear}. Default: \code{gbtree}
-#'   \item \code{silent} 0 means printing running messages, 1 means silent mode. Default: 0
+#'   \item \code{booster} which booster to use, can be \code{gbtree} or \code{gblinear}. Default: \code{gbtree}.
 #' }
 #'  
 #' 2. Booster Parameters
@@ -23,6 +25,7 @@
 #'   \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.
 #' }
 #' 
 #' 2.2. Parameter for Linear Booster
@@ -43,68 +46,151 @@
 #'     \item \code{binary:logistic} logistic regression for binary classification. Output probability.
 #'     \item \code{binary:logitraw} logistic regression for binary classification, output score before logistic transformation.
 #'     \item \code{num_class} set the number of classes. To use only with multiclass objectives.
-#'     \item \code{multi:softmax} set xgboost to do multiclass classification using the softmax objective. Class is represented by a number and should be from 0 to \code{num_class}.
-#'     \item \code{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 probabilities of each data point belonging to each class.
+#'     \item \code{multi:softmax} set xgboost to do multiclass classification using the softmax objective. Class is represented by a number and should be from 0 to \code{num_class - 1}.
+#'     \item \code{multi:softprob} same as softmax, but prediction outputs a vector of ndata * nclass elements, which can be further reshaped to ndata, nclass matrix. The result contains predicted probabilities of each data point belonging to each class.
 #'     \item \code{rank:pairwise} set xgboost to do ranking task by minimizing the pairwise loss.
 #'   }
 #'   \item \code{base_score} the initial prediction score of all instances, global bias. Default: 0.5
 #'   \item \code{eval_metric} evaluation metrics for validation data. Users can pass a self-defined function to it. Default: metric will be assigned according to objective(rmse for regression, and error for classification, mean average precision for ranking). List is provided in detail section.
 #' }
 #' 
-#' @param data takes an \code{xgb.DMatrix} as the input.
-#' @param nrounds the max number of iterations
-#' @param watchlist what information should be printed when \code{verbose=1} or
-#'   \code{verbose=2}. Watchlist is used to specify validation set monitoring
-#'   during training. For example user can specify
-#'    watchlist=list(validation1=mat1, validation2=mat2) to watch
-#'    the performance of each round's model on mat1 and mat2
-#'
+#' @param data training dataset. \code{xgb.train} accepts only an \code{xgb.DMatrix} as the input.
+#'        \code{xgboost}, in addition, also accepts \code{matrix}, \code{dgCMatrix}, or name of a local data file.
+#' @param nrounds max number of boosting iterations.
+#' @param watchlist named list of xgb.DMatrix datasets to use for evaluating model performance.
+#'        Metrics specified in either \code{eval_metric} or \code{feval} will be computed for each
+#'        of these datasets during each boosting iteration, and stored in the end as a field named 
+#'        \code{evaluation_log} in the resulting object. When either \code{verbose>=1} or 
+#'        \code{\link{cb.print.evaluation}} callback is engaged, the performance results are continuously
+#'        printed out during the training. 
+#'        E.g., specifying \code{watchlist=list(validation1=mat1, validation2=mat2)} allows to track
+#'        the performance of each round's model on mat1 and mat2.
 #' @param obj customized objective function. Returns gradient and second order 
-#'   gradient with given prediction and dtrain, 
+#'        gradient with given prediction and dtrain.
 #' @param feval custimized evaluation function. Returns 
-#'   \code{list(metric='metric-name', value='metric-value')} with given 
-#'   prediction and dtrain,
-#' @param verbose If 0, xgboost will stay silent. If 1, xgboost will print 
-#'   information of performance. If 2, xgboost will print information of both
-#' @param print.every.n Print every N progress messages when \code{verbose>0}. Default is 1 which means all messages are printed.
-#' @param early.stop.round If \code{NULL}, the early stopping function is not triggered. 
-#'     If set to an integer \code{k}, training with a validation set will stop if the performance 
-#'     keeps getting worse consecutively for \code{k} rounds.
-#' @param maximize If \code{feval} and \code{early.stop.round} are set, then \code{maximize} must be set as well.
-#'     \code{maximize=TRUE} means the larger the evaluation score the better.
-#' @param save_period save the model to the disk in every \code{save_period} rounds, 0 means no such action.
+#'        \code{list(metric='metric-name', value='metric-value')} with given 
+#'        prediction and dtrain.
+#' @param verbose If 0, xgboost will stay silent. If 1, it will print information about performance.
+#'        If 2, some additional information will be printed out.
+#'        Note that setting \code{verbose > 0} automatically engages the 
+#'        \code{cb.print.evaluation(period=1)} callback function.
+#' @param print_every_n Print each n-th iteration evaluation messages when \code{verbose>0}.
+#'        Default is 1 which means all messages are printed. This parameter is passed to the 
+#'        \code{\link{cb.print.evaluation}} callback.
+#' @param early_stopping_rounds If \code{NULL}, the early stopping function is not triggered. 
+#'        If set to an integer \code{k}, training with a validation set will stop if the performance 
+#'        doesn't improve for \code{k} rounds.
+#'        Setting this parameter engages the \code{\link{cb.early.stop}} callback.
+#' @param maximize If \code{feval} and \code{early_stopping_rounds} are set,
+#'        then this parameter must be set as well.
+#'        When it is \code{TRUE}, it means the larger the evaluation score the better.
+#'        This parameter is passed to the \code{\link{cb.early.stop}} callback.
+#' @param save_period when it is non-NULL, model is saved to disk after every \code{save_period} rounds,
+#'        0 means save at the end. The saving is handled by the \code{\link{cb.save.model}} callback.
 #' @param save_name the name or path for periodically saved model file.
+#' @param xgb_model a previously built model to continue the training from.
+#'        Could be either an object of class \code{xgb.Booster}, or its raw data, or the name of a 
+#'        file with a previously saved model.
+#' @param callbacks a list of callback functions to perform various task during boosting.
+#'        See \code{\link{callbacks}}. Some of the callbacks are automatically created depending on the 
+#'        parameters' values. User can provide either existing or their own callback methods in order 
+#'        to customize the training process.
 #' @param ... other parameters to pass to \code{params}.
+#' @param label vector of response values. Should not be provided when data is 
+#'        a local data file name or an \code{xgb.DMatrix}.
+#' @param missing by default is set to NA, which means that NA values should be considered as 'missing'
+#'        by the algorithm. Sometimes, 0 or other extreme value might be used to represent missing values.
+#'        This parameter is only used when input is a dense matrix.
+#' @param weight a vector indicating the weight for each row of the input.
 #' 
 #' @details 
-#' This is the training function for \code{xgboost}. 
+#' These are the training functions for \code{xgboost}. 
 #' 
-#' It supports advanced features such as \code{watchlist}, customized objective function (\code{feval}),
-#' therefore it is more flexible than \code{\link{xgboost}} function.
+#' The \code{xgb.train} interface supports advanced features such as \code{watchlist}, 
+#' customized objective and evaluation metric functions, therefore it is more flexible 
+#' than the \code{xgboost} interface.
 #'
 #' Parallelization is automatically enabled if \code{OpenMP} is present. 
 #' Number of threads can also be manually specified via \code{nthread} parameter.
 #' 
-#' \code{eval_metric} parameter (not listed above) is set automatically by Xgboost but can be overriden by parameter. Below is provided the list of different metric optimized by Xgboost to help you to understand how it works inside or to use them with the \code{watchlist} parameter.
+#' The evaluation metric is chosen automatically by Xgboost (according to the objective)
+#' when the \code{eval_metric} parameter is not provided.
+#' User may set one or several \code{eval_metric} parameters. 
+#' Note that when using a customized metric, only this single metric can be used.
+#' The folloiwing is the list of built-in metrics for which Xgboost provides optimized implementation:
 #'   \itemize{
 #'      \item \code{rmse} root mean square error. \url{http://en.wikipedia.org/wiki/Root_mean_square_error}
 #'      \item \code{logloss} negative log-likelihood. \url{http://en.wikipedia.org/wiki/Log-likelihood}
-#'      \item \code{mlogloss} multiclass logloss. \url{https://www.kaggle.com/wiki/MultiClassLogLoss}
-#'      \item \code{error} Binary classification error rate. It is calculated as \code{(wrong cases) / (all cases)}. For the predictions, the evaluation will regard the instances with prediction value larger than 0.5 as positive instances, and the others as negative instances.
-#'      \item \code{merror} Multiclass classification error rate. It is calculated as \code{(wrong cases) / (all cases)}.
+#'      \item \code{mlogloss} multiclass logloss. \url{http://wiki.fast.ai/index.php/Log_Loss}
+#'      \item \code{error} Binary classification error rate. It is calculated as \code{(# wrong cases) / (# all cases)}.
+#'            By default, it uses the 0.5 threshold for predicted values to define negative and positive instances.
+#'            Different threshold (e.g., 0.) could be specified as "error@0."
+#'      \item \code{merror} Multiclass classification error rate. It is calculated as \code{(# wrong cases) / (# all cases)}.
 #'      \item \code{auc} Area under the curve. \url{http://en.wikipedia.org/wiki/Receiver_operating_characteristic#'Area_under_curve} for ranking evaluation.
+#'      \item \code{aucpr} Area under the PR curve. \url{https://en.wikipedia.org/wiki/Precision_and_recall} for ranking evaluation.
 #'      \item \code{ndcg} Normalized Discounted Cumulative Gain (for ranking task). \url{http://en.wikipedia.org/wiki/NDCG}
 #'   }
-#'   
-#' Full list of parameters is available in the Wiki \url{https://github.com/dmlc/xgboost/wiki/Parameters}.
 #' 
-#' This function only accepts an \code{\link{xgb.DMatrix}} object as the input.
+#' The following callbacks are automatically created when certain parameters are set:
+#' \itemize{
+#'   \item \code{cb.print.evaluation} is turned on when \code{verbose > 0};
+#'         and the \code{print_every_n} parameter is passed to it.
+#'   \item \code{cb.evaluation.log} is on when \code{watchlist} is present.
+#'   \item \code{cb.early.stop}: when \code{early_stopping_rounds} is set.
+#'   \item \code{cb.save.model}: when \code{save_period > 0} is set.
+#' }
 #' 
+#' @return 
+#' An object of class \code{xgb.Booster} with the following elements:
+#' \itemize{
+#'   \item \code{handle} a handle (pointer) to the xgboost model in memory.
+#'   \item \code{raw} a cached memory dump of the xgboost model saved as R's \code{raw} type.
+#'   \item \code{niter} number of boosting iterations.
+#'   \item \code{evaluation_log} evaluation history storead as a \code{data.table} with the
+#'         first column corresponding to iteration number and the rest corresponding to evaluation
+#'         metrics' values. It is created by the \code{\link{cb.evaluation.log}} callback.
+#'   \item \code{call} a function call.
+#'   \item \code{params} parameters that were passed to the xgboost library. Note that it does not 
+#'         capture parameters changed by the \code{\link{cb.reset.parameters}} callback.
+#'   \item \code{callbacks} callback functions that were either automatically assigned or 
+#'         explicitely passed.
+#'   \item \code{best_iteration} iteration number with the best evaluation metric value
+#'         (only available with early stopping).
+#'   \item \code{best_ntreelimit} the \code{ntreelimit} value corresponding to the best iteration, 
+#'         which could further be used in \code{predict} method
+#'         (only available with early stopping).
+#'   \item \code{best_score} the best evaluation metric value during early stopping.
+#'         (only available with early stopping).
+#'   \item \code{feature_names} names of the training dataset features
+#'         (only when comun names were defined in training data).
+#'   \item \code{nfeatures} number of features in training data.
+#' }
+#' 
+#' @seealso
+#' \code{\link{callbacks}},
+#' \code{\link{predict.xgb.Booster}},
+#' \code{\link{xgb.cv}}
+#' 
+#' @references
+#'
+#' Tianqi Chen and Carlos Guestrin, "XGBoost: A Scalable Tree Boosting System",
+#' 22nd SIGKDD Conference on Knowledge Discovery and Data Mining, 2016, \url{https://arxiv.org/abs/1603.02754}
+#'
 #' @examples
 #' data(agaricus.train, package='xgboost')
+#' data(agaricus.test, package='xgboost')
+#' 
 #' dtrain <- xgb.DMatrix(agaricus.train$data, label = agaricus.train$label)
-#' dtest <- dtrain
-#' watchlist <- list(eval = dtest, train = dtrain)
+#' dtest <- xgb.DMatrix(agaricus.test$data, label = agaricus.test$label)
+#' watchlist <- list(train = dtrain, eval = dtest)
+#' 
+#' ## A simple xgb.train example:
+#' param <- list(max_depth = 2, eta = 1, silent = 1, nthread = 2, 
+#'               objective = "binary:logistic", eval_metric = "auc")
+#' bst <- xgb.train(param, dtrain, nrounds = 2, watchlist)
+#' 
+#' 
+#' ## An xgb.train example where custom objective and evaluation metric are used:
 #' logregobj <- function(preds, dtrain) {
 #'    labels <- getinfo(dtrain, "label")
 #'    preds <- 1/(1 + exp(-preds))
@@ -117,121 +203,169 @@
 #'   err <- as.numeric(sum(labels != (preds > 0)))/length(labels)
 #'   return(list(metric = "error", value = err))
 #' }
-#' param <- list(max.depth = 2, eta = 1, silent = 1, objective=logregobj,eval_metric=evalerror)
-#' bst <- xgb.train(param, dtrain, nthread = 2, nround = 2, watchlist)
+#' 
+#' # These functions could be used by passing them either:
+#' #  as 'objective' and 'eval_metric' parameters in the params list:
+#' param <- list(max_depth = 2, eta = 1, silent = 1, nthread = 2, 
+#'               objective = logregobj, eval_metric = evalerror)
+#' bst <- xgb.train(param, dtrain, nrounds = 2, watchlist)
+#' 
+#' #  or through the ... arguments:
+#' param <- list(max_depth = 2, eta = 1, silent = 1, nthread = 2)
+#' bst <- xgb.train(param, dtrain, nrounds = 2, watchlist,
+#'                  objective = logregobj, eval_metric = evalerror)
+#' 
+#' #  or as dedicated 'obj' and 'feval' parameters of xgb.train:
+#' bst <- xgb.train(param, dtrain, nrounds = 2, watchlist,
+#'                  obj = logregobj, feval = evalerror)
+#' 
+#' 
+#' ## An xgb.train example of using variable learning rates at each iteration:
+#' param <- list(max_depth = 2, eta = 1, silent = 1, nthread = 2,
+#'               objective = "binary:logistic", eval_metric = "auc")
+#' my_etas <- list(eta = c(0.5, 0.1))
+#' bst <- xgb.train(param, dtrain, nrounds = 2, watchlist,
+#'                  callbacks = list(cb.reset.parameters(my_etas)))
+#' 
+#' ## Early stopping:
+#' bst <- xgb.train(param, dtrain, nrounds = 25, watchlist,
+#'                  early_stopping_rounds = 3)
+#' 
+#' ## An 'xgboost' interface example:
+#' bst <- xgboost(data = agaricus.train$data, label = agaricus.train$label, 
+#'                max_depth = 2, eta = 1, nthread = 2, nrounds = 2, 
+#'                objective = "binary:logistic")
+#' pred <- predict(bst, agaricus.test$data)
+#' 
+#' @rdname xgb.train
 #' @export
-xgb.train <- function(params=list(), data, nrounds, watchlist = list(),
-                      obj = NULL, feval = NULL, verbose = 1, print.every.n=1L,
-                      early.stop.round = NULL, maximize = NULL,
-                      save_period = 0, save_name = "xgboost.model", ...) {
+xgb.train <- function(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(), ...) {
+  
+  check.deprecation(...)
+  
+  params <- check.booster.params(params, ...)
+
+  check.custom.obj()
+  check.custom.eval()
+  
+  # data & watchlist checks
   dtrain <- data
-  if (typeof(params) != "list") {
-    stop("xgb.train: first argument params must be list")
-  }
-  if (class(dtrain) != "xgb.DMatrix") {
-    stop("xgb.train: second argument dtrain must be xgb.DMatrix")
-  }
-  if (verbose > 1) {
-    params <- append(params, list(silent = 0))
-  } else {
-    params <- append(params, list(silent = 1))
-  }
-  if (length(watchlist) != 0 && verbose == 0) {
-    warning('watchlist is provided but verbose=0, no evaluation information will be printed')
+  if (!inherits(dtrain, "xgb.DMatrix")) 
+    stop("second argument dtrain must be xgb.DMatrix")
+  if (length(watchlist) > 0) {
+    if (typeof(watchlist) != "list" ||
+        !all(vapply(watchlist, inherits, logical(1), what = 'xgb.DMatrix')))
+      stop("watchlist must be a list of xgb.DMatrix elements")
+    evnames <- names(watchlist)
+    if (is.null(evnames) || any(evnames == ""))
+      stop("each element of the watchlist must have a name tag")
   }
 
-  fit.call <- match.call()
-  dot.params <- list(...)
-  nms.params <- names(params)
-  nms.dot.params <- names(dot.params)
-  if (length(intersect(nms.params,nms.dot.params)) > 0)
-    stop("Duplicated term in parameters. Please check your list of params.")
-  params <- append(params, dot.params)
-
-  # customized objective and evaluation metric interface
-  if (!is.null(params$objective) && !is.null(obj))
-    stop("xgb.train: cannot assign two different objectives")
-  if (!is.null(params$objective))
-    if (class(params$objective) == 'function') {
-      obj <- params$objective
-      params$objective <- NULL
-    }
-  if (!is.null(params$eval_metric) && !is.null(feval))
-    stop("xgb.train: cannot assign two different evaluation metrics")
-  if (!is.null(params$eval_metric))
-    if (class(params$eval_metric) == 'function') {
-      feval <- params$eval_metric
-      params$eval_metric <- NULL
-    }
-
-  # Early stopping
-  if (!is.null(early.stop.round)){
-    if (!is.null(feval) && is.null(maximize))
-      stop('Please set maximize to note whether the model is maximizing the evaluation or not.')
-    if (length(watchlist) == 0)
-      stop('For early stopping you need at least one set in watchlist.')
-    if (is.null(maximize) && is.null(params$eval_metric))
-      stop('Please set maximize to note whether the model is maximizing the evaluation or not.')
-    if (is.null(maximize))
-    {
-      if (params$eval_metric %in% c('rmse','logloss','error','merror','mlogloss')) {
-        maximize <- FALSE
-      } else {
-        maximize <- TRUE
-      }
-    }
-
-    if (maximize) {
-      bestScore <- 0
-    } else {
-      bestScore <- Inf
-    }
-    bestInd <- 0
-    earlyStopflag = FALSE
-
-    if (length(watchlist) > 1)
-      warning('Only the first data set in watchlist is used for early stopping process.')
+  # evaluation printing callback
+  params <- c(params, list(silent = ifelse(verbose > 1, 0, 1)))
+  print_every_n <- max( as.integer(print_every_n), 1L)
+  if (!has.callbacks(callbacks, 'cb.print.evaluation') &&
+      verbose) {
+    callbacks <- add.cb(callbacks, cb.print.evaluation(print_every_n))
   }
+  # evaluation log callback:  it is automatically enabled when watchlist is provided
+  evaluation_log <- list()
+  if (!has.callbacks(callbacks, 'cb.evaluation.log') &&
+      length(watchlist) > 0) {
+    callbacks <- add.cb(callbacks, cb.evaluation.log())
+  }
+  # Model saving callback
+  if (!is.null(save_period) &&
+      !has.callbacks(callbacks, 'cb.save.model')) {
+    callbacks <- add.cb(callbacks, cb.save.model(save_period, save_name))
+  }
+  # Early stopping callback
+  stop_condition <- FALSE
+  if (!is.null(early_stopping_rounds) &&
+      !has.callbacks(callbacks, 'cb.early.stop')) {
+    callbacks <- add.cb(callbacks, cb.early.stop(early_stopping_rounds, 
+                                                 maximize = maximize, verbose = verbose))
+  }
+  # Sort the callbacks into categories
+  cb <- categorize.callbacks(callbacks)
 
-  handle <- xgb.Booster(params, append(watchlist, dtrain))
+  # The tree updating process would need slightly different handling
+  is_update <- NVL(params[['process_type']], '.') == 'update'
+
+  # Construct a booster (either a new one or load from xgb_model)
+  handle <- xgb.Booster.handle(params, append(watchlist, dtrain), xgb_model)
   bst <- xgb.handleToBooster(handle)
-  print.every.n <- max( as.integer(print.every.n), 1L)
-  for (i in 1:nrounds) {
-    succ <- xgb.iter.update(bst$handle, dtrain, i - 1, obj)
-    if (length(watchlist) != 0) {
-      msg <- xgb.iter.eval(bst$handle, watchlist, i - 1, feval)
-      if (0 == ( (i - 1) %% print.every.n))
-	    cat(paste(msg, "\n", sep = ""))
-      if (!is.null(early.stop.round))
-      {
-        score <- strsplit(msg,':|\\s+')[[1]][3]
-        score <- as.numeric(score)
-        if ( (maximize && score > bestScore) || (!maximize && score < bestScore)) {
-          bestScore <- score
-          bestInd <- i
-        } else {
-          earlyStopflag = TRUE
-          if (i - bestInd >= early.stop.round) {
-            cat('Stopping. Best iteration:', bestInd, '\n')
-            break
-          }
-        }
-      }
-    }
-    if (save_period > 0) {
-      if (i %% save_period == 0) {
-        xgb.save(bst, save_name)
-      }
+
+  # extract parameters that can affect the relationship b/w #trees and #iterations
+  num_class <- max(as.numeric(NVL(params[['num_class']], 1)), 1)
+  num_parallel_tree <- max(as.numeric(NVL(params[['num_parallel_tree']], 1)), 1)
+
+  # When the 'xgb_model' was set, find out how many boosting iterations it has
+  niter_init <- 0
+  if (!is.null(xgb_model)) {
+    niter_init <- as.numeric(xgb.attr(bst, 'niter')) + 1
+    if (length(niter_init) == 0) {
+      niter_init <- xgb.ntree(bst) %/% (num_parallel_tree * num_class)
     }
   }
-  bst <- xgb.Booster.check(bst)
+  if(is_update && nrounds > niter_init)
+    stop("nrounds cannot be larger than ", niter_init, " (nrounds of xgb_model)")
 
-  if (!is.null(early.stop.round)) {
-    bst$bestScore <- bestScore
-    bst$bestInd <- bestInd
+  # TODO: distributed code
+  rank <- 0
+  
+  niter_skip <- ifelse(is_update, 0, niter_init)
+  begin_iteration <- niter_skip + 1
+  end_iteration <- niter_skip + nrounds
+  
+  # the main loop for boosting iterations
+  for (iteration in begin_iteration:end_iteration) {
+    
+    for (f in cb$pre_iter) f()
+    
+    xgb.iter.update(bst$handle, dtrain, iteration - 1, obj)
+    
+    bst_evaluation <- numeric(0)
+    if (length(watchlist) > 0)
+      bst_evaluation <- xgb.iter.eval(bst$handle, watchlist, iteration - 1, feval)
+    
+    xgb.attr(bst$handle, 'niter') <- iteration - 1
+
+    for (f in cb$post_iter) f()
+
+    if (stop_condition) break
+  }
+  for (f in cb$finalize) f(finalize = TRUE)
+  
+  bst <- xgb.Booster.complete(bst, saveraw = TRUE)
+  
+  # store the total number of boosting iterations
+  bst$niter = end_iteration
+
+  # store the evaluation results
+  if (length(evaluation_log) > 0 &&
+      nrow(evaluation_log) > 0) {
+    # include the previous compatible history when available
+    if (inherits(xgb_model, 'xgb.Booster') &&
+        !is_update &&
+        !is.null(xgb_model$evaluation_log) &&
+        isTRUE(all.equal(colnames(evaluation_log),
+                         colnames(xgb_model$evaluation_log)))) {
+      evaluation_log <- rbindlist(list(xgb_model$evaluation_log, evaluation_log))
+    }
+    bst$evaluation_log <- evaluation_log
   }
 
-  attr(bst, "call") <- fit.call
-  attr(bst, "params") <- params
+  bst$call <- match.call()
+  bst$params <- params
+  bst$callbacks <- callbacks
+  if (!is.null(colnames(dtrain)))
+    bst$feature_names <- colnames(dtrain)
+  bst$nfeatures <- ncol(dtrain)
+
   return(bst)
 }

R-package/R/xgboost.R

@@ -1,83 +1,26 @@
-#' eXtreme Gradient Boosting (Tree) library
-#' 
-#' A simple interface for training xgboost model. Look at \code{\link{xgb.train}} function for a more advanced interface.
-#' 
-#' @param data takes \code{matrix}, \code{dgCMatrix}, local data file or 
-#'   \code{xgb.DMatrix}. 
-#' @param label the response variable. User should not set this field,
-#'    if data is local data file or  \code{xgb.DMatrix}. 
-#' @param params the list of parameters.
-#' 
-#' Commonly used ones are:
-#' \itemize{
-#'   \item \code{objective} objective function, common ones are
-#'   \itemize{
-#'     \item \code{reg:linear} linear regression
-#'     \item \code{binary:logistic} logistic regression for classification
-#'   }
-#'   \item \code{eta} step size of each boosting step
-#'   \item \code{max.depth} maximum depth of the tree
-#'   \item \code{nthread} number of thread used in training, if not set, all threads are used
-#' }
-#'   
-#'   Look at \code{\link{xgb.train}} for a more complete list of parameters or \url{https://github.com/dmlc/xgboost/wiki/Parameters} for the full list.
-#'   
-#'   See also \code{demo/} for walkthrough example in R.
-#' 
-#' @param nrounds the max number of iterations
-#' @param verbose If 0, xgboost will stay silent. If 1, xgboost will print 
-#'   information of performance. If 2, xgboost will print information of both
-#'   performance and construction progress information
-#' @param print.every.n Print every N progress messages when \code{verbose>0}. Default is 1 which means all messages are printed.
-#' @param missing Missing is only used when input is dense matrix, pick a float 
-#'     value that represents missing value. Sometimes a data use 0 or other extreme value to represents missing values.
-#' @param weight a vector indicating the weight for each row of the input.
-#' @param early.stop.round If \code{NULL}, the early stopping function is not triggered. 
-#'     If set to an integer \code{k}, training with a validation set will stop if the performance 
-#'     keeps getting worse consecutively for \code{k} rounds.
-#' @param maximize If \code{feval} and \code{early.stop.round} are set, then \code{maximize} must be set as well.
-#'     \code{maximize=TRUE} means the larger the evaluation score the better.
-#' @param save_period save the model to the disk in every \code{save_period} rounds, 0 means no such action.
-#' @param save_name the name or path for periodically saved model file.
-#' @param ... other parameters to pass to \code{params}.
-#' 
-#' @details 
-#' This is the modeling function for Xgboost.
-#' 
-#' Parallelization is automatically enabled if \code{OpenMP} is present.
-#' 
-#' Number of threads can also be manually specified via \code{nthread} parameter.
-#' 
-#' @examples
-#' data(agaricus.train, package='xgboost')
-#' data(agaricus.test, package='xgboost')
-#' train <- agaricus.train
-#' test <- agaricus.test
-#' bst <- xgboost(data = train$data, label = train$label, max.depth = 2, 
-#'                eta = 1, nthread = 2, nround = 2, objective = "binary:logistic")
-#' pred <- predict(bst, test$data)
-#' 
+# Simple interface for training an xgboost model that wraps \code{xgb.train}.
+# Its documentation is combined with xgb.train.
+#
+#' @rdname xgb.train
 #' @export
 xgboost <- function(data = NULL, label = NULL, missing = NA, weight = NULL,
                     params = list(), nrounds,
-                    verbose = 1, print.every.n = 1L, early.stop.round = NULL,
-                    maximize = NULL, save_period = 0, save_name = "xgboost.model", ...) {
+                    verbose = 1, print_every_n = 1L, 
+                    early_stopping_rounds = NULL, maximize = NULL, 
+                    save_period = NULL, save_name = "xgboost.model",
+                    xgb_model = NULL, callbacks = list(), ...) {
+
   dtrain <- xgb.get.DMatrix(data, label, missing, weight)
 
-  params <- append(params, list(...))
-
-  if (verbose > 0) {
-    watchlist <- list(train = dtrain)
-  } else {
-    watchlist <- list()
-  }
-
-  bst <- xgb.train(params, dtrain, nrounds, watchlist, verbose = verbose, print.every.n=print.every.n,
-                   early.stop.round = early.stop.round, maximize = maximize,
-                   save_period = save_period, save_name = save_name)
+  watchlist <- list(train = dtrain)
 
+  bst <- xgb.train(params, dtrain, nrounds, watchlist, verbose = verbose, print_every_n = print_every_n,
+                   early_stopping_rounds = early_stopping_rounds, maximize = maximize,
+                   save_period = save_period, save_name = save_name,
+                   xgb_model = xgb_model, callbacks = callbacks, ...)
   return(bst)
 }
+
 #' Training part from Mushroom Data Set
 #' 
 #' This data set is originally from the Mushroom data set,
@@ -131,3 +74,40 @@ NULL
 #' @format A list containing a label vector, and a dgCMatrix object with 1611 
 #' rows and 126 variables
 NULL
+
+# Various imports
+#' @importClassesFrom Matrix dgCMatrix dgeMatrix
+#' @importFrom Matrix colSums
+#' @importFrom Matrix sparse.model.matrix
+#' @importFrom Matrix sparseVector
+#' @importFrom Matrix sparseMatrix
+#' @importFrom Matrix t
+#' @importFrom data.table data.table
+#' @importFrom data.table is.data.table
+#' @importFrom data.table as.data.table
+#' @importFrom data.table :=
+#' @importFrom data.table rbindlist
+#' @importFrom data.table setkey
+#' @importFrom data.table setkeyv
+#' @importFrom data.table setnames
+#' @importFrom magrittr %>%
+#' @importFrom stringi stri_detect_regex
+#' @importFrom stringi stri_match_first_regex
+#' @importFrom stringi stri_replace_first_regex
+#' @importFrom stringi stri_replace_all_regex
+#' @importFrom stringi stri_split_regex
+#' @importFrom utils object.size str tail
+#' @importFrom stats predict
+#' @importFrom stats median
+#' @importFrom utils head
+#' @importFrom graphics barplot
+#' @importFrom graphics lines
+#' @importFrom graphics points
+#' @importFrom graphics grid
+#' @importFrom graphics par
+#' @importFrom graphics title
+#' @importFrom grDevices rgb
+#' 
+#' @import methods
+#' @useDynLib xgboost, .registration = TRUE
+NULL

R-package/README.md

@@ -1,8 +1,14 @@
-R package for xgboost
-=====================
+XGBoost R Package for Scalable GBM
+==================================
 
-[![CRAN Status Badge](http://www.r-pkg.org/badges/version/xgboost)](http://cran.r-project.org/web/packages/xgboost)
-[![CRAN Downloads](http://cranlogs.r-pkg.org/badges/xgboost)](http://cran.rstudio.com/web/packages/xgboost/index.html)
+[![CRAN Status Badge](http://www.r-pkg.org/badges/version/xgboost)](https://cran.r-project.org/web/packages/xgboost)
+[![CRAN Downloads](http://cranlogs.r-pkg.org/badges/xgboost)](https://cran.rstudio.com/web/packages/xgboost/index.html)
+[![Documentation Status](https://readthedocs.org/projects/xgboost/badge/?version=latest)](http://xgboost.readthedocs.org/en/latest/R-package/index.html)
+
+Resources
+---------
+* [XGBoost R Package Online Documentation](http://xgboost.readthedocs.org/en/latest/R-package/index.html)
+  - Check this out for detailed documents, examples and tutorials.
 
 Installation
 ------------
@@ -13,11 +19,7 @@ We are [on CRAN](https://cran.r-project.org/web/packages/xgboost/index.html) now
 install.packages('xgboost')
 ```
 
-For up-to-date version, please install from github. Windows user will need to install [RTools](http://cran.r-project.org/bin/windows/Rtools/) first.
-
-```r
-devtools::install_github('dmlc/xgboost',subdir='R-package')
-```
+For more detailed installation instructions, please see [here](http://xgboost.readthedocs.org/en/latest/build.html#r-package-installation).
 
 Examples
 --------
@@ -25,20 +27,7 @@ Examples
 * Please visit [walk through example](demo).
 * See also the [example scripts](../demo/kaggle-higgs) for Kaggle Higgs Challenge, including [speedtest script](../demo/kaggle-higgs/speedtest.R) on this dataset and the one related to [Otto challenge](../demo/kaggle-otto), including a [RMarkdown documentation](../demo/kaggle-otto/understandingXGBoostModel.Rmd).
 
-Notes
------
+Development
+-----------
 
-If you face an issue installing the package using  ```devtools::install_github```, something like this (even after updating libxml and RCurl as lot of forums say) -
-
-```
-devtools::install_github('dmlc/xgboost',subdir='R-package')
-Downloading github repo dmlc/xgboost@master
-Error in function (type, msg, asError = TRUE)  :
-  Peer certificate cannot be authenticated with given CA certificates
-```
-To get around this you can build the package locally as mentioned [here](https://github.com/dmlc/xgboost/issues/347) -
-```
-1. Clone the current repository and set your workspace to xgboost/R-package/
-2. Run R CMD INSTALL --build . in terminal to get the tarball.
-3. Run install.packages('path_to_the_tarball',repo=NULL) in R to install.
-```
+* See the [R Package section](https://xgboost.readthedocs.io/en/latest/how_to/contribute.html#r-package) of the contributors guide.

R-package/cleanup

@@ -0,0 +1,3 @@
+#!/bin/sh
+
+rm -f src/Makevars

R-package/configure.ac

@@ -0,0 +1,31 @@
+### configure.ac					-*- Autoconf -*-
+
+AC_PREREQ(2.62)
+
+AC_INIT([xgboost],[0.6-3],[],[xgboost],[])
+
+OPENMP_CXXFLAGS=""
+
+if test `uname -s` = "Linux"
+then
+  OPENMP_CXXFLAGS="\$(SHLIB_OPENMP_CFLAGS)"
+fi
+
+if test `uname -s` = "Darwin"
+then
+  OPENMP_CXXFLAGS="\$(SHLIB_OPENMP_CFLAGS)"
+  ac_pkg_openmp=no
+  AC_MSG_CHECKING([whether OpenMP will work in a package])
+  AC_LANG_CONFTEST(
+  [AC_LANG_PROGRAM([[#include <omp.h>]], [[ return omp_get_num_threads (); ]])])
+  PKG_CFLAGS="${OPENMP_CFLAGS}" PKG_LIBS="${OPENMP_CFLAGS}" "$RBIN" CMD SHLIB conftest.c 1>&AS_MESSAGE_LOG_FD 2>&AS_MESSAGE_LOG_FD && "$RBIN" --vanilla -q -e "dyn.load(paste('conftest',.Platform\$dynlib.ext,sep=''))" 1>&AS_MESSAGE_LOG_FD 2>&AS_MESSAGE_LOG_FD && ac_pkg_openmp=yes
+  AC_MSG_RESULT([${ac_pkg_openmp}])
+  if test "${ac_pkg_openmp}" = no; then
+    OPENMP_CXXFLAGS=''
+  fi
+fi
+
+AC_SUBST(OPENMP_CXXFLAGS)
+AC_CONFIG_FILES([src/Makevars])
+AC_OUTPUT
+

R-package/demo/00Index

@@ -9,3 +9,6 @@ create_sparse_matrix            Create Sparse Matrix
 predict_leaf_indices            Predicting the corresponding leaves
 early_stopping                  Early Stop in training
 poisson_regression              Poisson Regression on count data
+tweedie_regression              Tweddie Regression
+gpu_accelerated                 GPU-accelerated tree building algorithms
+

R-package/demo/README.md

@@ -8,6 +8,7 @@ XGBoost R Feature Walkthrough
 * [Generalized Linear Model](generalized_linear_model.R)
 * [Cross validation](cross_validation.R)
 * [Create a sparse matrix from a dense one](create_sparse_matrix.R)
+* [Use GPU-accelerated tree building algorithms](gpu_accelerated.R)
 
 Benchmarks
 ====

R-package/demo/basic_walkthrough.R

@@ -1,7 +1,8 @@
 require(xgboost)
 require(methods)
+
 # we load in the agaricus dataset
-# In this example, we are aiming to predict whether a mushroom can be eaten
+# In this example, we are aiming to predict whether a mushroom is edible
 data(agaricus.train, package='xgboost')
 data(agaricus.test, package='xgboost')
 train <- agaricus.train
@@ -15,33 +16,33 @@ class(train$data)
 # note: we are putting in sparse matrix here, xgboost naturally handles sparse input
 # use sparse matrix when your feature is sparse(e.g. when you are using one-hot encoding vector)
 print("Training xgboost with sparseMatrix")
-bst <- xgboost(data = train$data, label = train$label, max.depth = 2, eta = 1, nround = 2,
+bst <- xgboost(data = train$data, label = train$label, max_depth = 2, eta = 1, nrounds = 2,
                nthread = 2, objective = "binary:logistic")
 # alternatively, you can put in dense matrix, i.e. basic R-matrix
 print("Training xgboost with Matrix")
-bst <- xgboost(data = as.matrix(train$data), label = train$label, max.depth = 2, eta = 1, nround = 2,
+bst <- xgboost(data = as.matrix(train$data), label = train$label, max_depth = 2, eta = 1, nrounds = 2,
                nthread = 2, objective = "binary:logistic")
 
 # you can also put in xgb.DMatrix object, which stores label, data and other meta datas needed for advanced features
 print("Training xgboost with xgb.DMatrix")
 dtrain <- xgb.DMatrix(data = train$data, label = train$label)
-bst <- xgboost(data = dtrain, max.depth = 2, eta = 1, nround = 2, nthread = 2, 
+bst <- xgboost(data = dtrain, max_depth = 2, eta = 1, nrounds = 2, nthread = 2, 
                objective = "binary:logistic")
 
 # Verbose = 0,1,2
 print("Train xgboost with verbose 0, no message")
-bst <- xgboost(data = dtrain, max.depth = 2, eta = 1, nround = 2,
+bst <- xgboost(data = dtrain, max_depth = 2, eta = 1, nrounds = 2,
                nthread = 2, objective = "binary:logistic", verbose = 0)
 print("Train xgboost with verbose 1, print evaluation metric")
-bst <- xgboost(data = dtrain, max.depth = 2, eta = 1, nround = 2,
+bst <- xgboost(data = dtrain, max_depth = 2, eta = 1, nrounds = 2,
                nthread = 2, objective = "binary:logistic", verbose = 1)
 print("Train xgboost with verbose 2, also print information about tree")
-bst <- xgboost(data = dtrain, max.depth = 2, eta = 1, nround = 2,
+bst <- xgboost(data = dtrain, max_depth = 2, eta = 1, nrounds = 2,
                nthread = 2, objective = "binary:logistic", verbose = 2)
 
 # you can also specify data as file path to a LibSVM format input
 # since we do not have this file with us, the following line is just for illustration
-# bst <- xgboost(data = 'agaricus.train.svm', max.depth = 2, eta = 1, nround = 2,objective = "binary:logistic")
+# bst <- xgboost(data = 'agaricus.train.svm', max_depth = 2, eta = 1, nrounds = 2,objective = "binary:logistic")
 
 #--------------------basic prediction using xgboost--------------
 # you can do prediction using the following line
@@ -64,8 +65,8 @@ raw = xgb.save.raw(bst)
 # load binary model to R
 bst3 <- xgb.load(raw)
 pred3 <- predict(bst3, test$data)
-# pred2 should be identical to pred
-print(paste("sum(abs(pred3-pred))=", sum(abs(pred2-pred))))
+# pred3 should be identical to pred
+print(paste("sum(abs(pred3-pred))=", sum(abs(pred3-pred))))
 
 #----------------Advanced features --------------
 # to use advanced features, we need to put data in xgb.DMatrix
@@ -77,19 +78,19 @@ watchlist <- list(train=dtrain, test=dtest)
 # to train with watchlist, use xgb.train, which contains more advanced features
 # watchlist allows us to monitor the evaluation result on all data in the list 
 print("Train xgboost using xgb.train with watchlist")
-bst <- xgb.train(data=dtrain, max.depth=2, eta=1, nround=2, watchlist=watchlist,
+bst <- xgb.train(data=dtrain, max_depth=2, eta=1, nrounds=2, watchlist=watchlist,
                  nthread = 2, objective = "binary:logistic")
 # we can change evaluation metrics, or use multiple evaluation metrics
 print("train xgboost using xgb.train with watchlist, watch logloss and error")
-bst <- xgb.train(data=dtrain, max.depth=2, eta=1, nround=2, watchlist=watchlist,
-                 eval.metric = "error", eval.metric = "logloss",
+bst <- xgb.train(data=dtrain, max_depth=2, eta=1, nrounds=2, watchlist=watchlist,
+                 eval_metric = "error", eval_metric = "logloss",
                  nthread = 2, objective = "binary:logistic")
 
 # xgb.DMatrix can also be saved using xgb.DMatrix.save
 xgb.DMatrix.save(dtrain, "dtrain.buffer")
 # to load it in, simply call xgb.DMatrix
 dtrain2 <- xgb.DMatrix("dtrain.buffer")
-bst <- xgb.train(data=dtrain2, max.depth=2, eta=1, nround=2, watchlist=watchlist,
+bst <- xgb.train(data=dtrain2, max_depth=2, eta=1, nrounds=2, watchlist=watchlist,
                  nthread = 2, objective = "binary:logistic")
 # information can be extracted from xgb.DMatrix using getinfo
 label = getinfo(dtest, "label")
@@ -98,11 +99,12 @@ err <- as.numeric(sum(as.integer(pred > 0.5) != label))/length(label)
 print(paste("test-error=", err))
 
 # You can dump the tree you learned using xgb.dump into a text file
-xgb.dump(bst, "dump.raw.txt", with.stats = T)
+dump_path = file.path(tempdir(), 'dump.raw.txt')
+xgb.dump(bst, dump_path, with_stats = T)
 
 # Finally, you can check which features are the most important.
 print("Most important features (look at column Gain):")
-imp_matrix <- xgb.importance(feature_names = train$data@Dimnames[[2]], model = bst)
+imp_matrix <- xgb.importance(feature_names = colnames(train$data), model = bst)
 print(imp_matrix)
 
 # Feature importance bar plot by gain

R-package/demo/boost_from_prediction.R

@@ -11,8 +11,8 @@ watchlist <- list(eval = dtest, train = dtrain)
 #
 print('start running example to start from a initial prediction')
 # train xgboost for 1 round
-param <- list(max.depth=2,eta=1,nthread = 2, silent=1,objective='binary:logistic')
-bst <- xgb.train( param, dtrain, 1, watchlist )
+param <- list(max_depth=2, eta=1, nthread = 2, silent=1, objective='binary:logistic')
+bst <- xgb.train(param, dtrain, 1, watchlist)
 # Note: we need the margin value instead of transformed prediction in set_base_margin
 # do predict with output_margin=TRUE, will always give you margin values before logistic transformation
 ptrain <- predict(bst, dtrain, outputmargin=TRUE)

R-package/demo/caret_wrapper.R

@@ -24,7 +24,7 @@ df[,ID:=NULL]
 #-------------Basic Training using XGBoost in caret Library-----------------
 # Set up control parameters for caret::train
 # Here we use 10-fold cross-validation, repeating twice, and using random search for tuning hyper-parameters.
-fitControl <- trainControl(method = "cv", number = 10, repeats = 2, search = "random")
+fitControl <- trainControl(method = "repeatedcv", number = 10, repeats = 2, search = "random")
 # train a xgbTree model using caret::train
 model <- train(factor(Improved)~., data = df, method = "xgbTree", trControl = fitControl)
 

R-package/demo/create_sparse_matrix.R

@@ -65,11 +65,10 @@ output_vector = df[,Y:=0][Improved == "Marked",Y:=1][,Y]
 
 # Following is the same process as other demo
 cat("Learning...\n")
-bst <- xgboost(data = sparse_matrix, label = output_vector, max.depth = 9,
-               eta = 1, nthread = 2, nround = 10,objective = "binary:logistic")
+bst <- xgboost(data = sparse_matrix, label = output_vector, max_depth = 9,
+               eta = 1, nthread = 2, nrounds = 10, objective = "binary:logistic")
 
-# sparse_matrix@Dimnames[[2]] represents the column names of the sparse matrix.
-importance <- xgb.importance(feature_names = sparse_matrix@Dimnames[[2]], model = bst)
+importance <- xgb.importance(feature_names = colnames(sparse_matrix), model = bst)
 print(importance)
 # According to the matrix below, the most important feature in this dataset to predict if the treatment will work is the Age. The second most important feature is having received a placebo or not. The sex is third. Then we see our generated features (AgeDiscret). We can see that their contribution is very low (Gain column).
 

R-package/demo/cross_validation.R

@@ -6,7 +6,7 @@ dtrain <- xgb.DMatrix(agaricus.train$data, label = agaricus.train$label)
 dtest <- xgb.DMatrix(agaricus.test$data, label = agaricus.test$label)
 
 nround <- 2
-param <- list(max.depth=2,eta=1,silent=1,nthread = 2, objective='binary:logistic')
+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
@@ -19,7 +19,7 @@ cat('running cross validation, disable standard deviation display\n')
 # [iteration]  metric_name:mean_value+std_value
 # std_value is standard deviation of the metric
 xgb.cv(param, dtrain, nround, nfold=5,
-       metrics={'error'}, showsd = FALSE)
+       metrics='error', showsd = FALSE)
 
 ###
 # you can also do cross validation with cutomized loss function
@@ -40,12 +40,12 @@ evalerror <- function(preds, dtrain) {
   return(list(metric = "error", value = err))
 }
 
-param <- list(max.depth=2,eta=1,silent=1,
+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 = nround, nfold = 5)
 
 # do cross validation with prediction values for each fold
 res <- xgb.cv(params = param, data = dtrain, nrounds = nround, nfold = 5, prediction = TRUE)
-res$dt
+res$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, silent=1, 
+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, silent=1, 
+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, silent=1)
+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
@@ -34,7 +34,7 @@ print ('start training with early Stopping setting')
 
 bst <- xgb.train(param, dtrain, num_round, watchlist, 
                  objective = logregobj, eval_metric = evalerror, maximize = FALSE,
-                 early.stop.round = 3)
+                 early_stopping_round = 3)
 bst <- xgb.cv(param, dtrain, num_round, nfold = 5, 
               objective = logregobj, eval_metric = evalerror,
-              maximize = FALSE, early.stop.round = 3)
+              maximize = FALSE, early_stopping_rounds = 3)

R-package/demo/gpu_accelerated.R

@@ -0,0 +1,45 @@
+# An example of using GPU-accelerated tree building algorithms
+# 
+# NOTE: it can only run if you have a CUDA-enable GPU and the package was 
+#       specially compiled with GPU support.
+#
+# For the current functionality, see 
+# https://xgboost.readthedocs.io/en/latest/gpu/index.html
+#
+
+library('xgboost')
+
+# Simulate N x p random matrix with some binomial response dependent on pp columns
+set.seed(111)
+N <- 1000000
+p <- 50
+pp <- 25
+X <- matrix(runif(N * p), ncol = p)
+betas <- 2 * runif(pp) - 1
+sel <- sort(sample(p, pp))
+m <- X[, sel] %*% betas - 1 + rnorm(N)
+y <- rbinom(N, 1, plogis(m))
+
+tr <- sample.int(N, N * 0.75)
+dtrain <- xgb.DMatrix(X[tr,], label = y[tr])
+dtest <- xgb.DMatrix(X[-tr,], label = y[-tr])
+wl <- list(train = dtrain, test = dtest)
+
+# An example of running 'gpu_hist' algorithm
+# which is
+# - similar to the 'hist'
+# - the fastest option for moderately large datasets
+# - current limitations: max_depth < 16, does not implement guided loss
+# You can use tree_method = 'gpu_exact' for another GPU accelerated algorithm,
+# which is slower, more memory-hungry, but does not use binning.
+param <- list(objective = 'reg:logistic', eval_metric = 'auc', subsample = 0.5, nthread = 4,
+              max_bin = 64, tree_method = 'gpu_hist')
+pt <- proc.time()
+bst_gpu <- xgb.train(param, dtrain, watchlist = wl, nrounds = 50)
+proc.time() - pt
+
+# Compare to the 'hist' algorithm:
+param$tree_method <- 'hist'
+pt <- proc.time()
+bst_hist <- xgb.train(param, dtrain, watchlist = wl, nrounds = 50)
+proc.time() - pt

R-package/demo/predict_first_ntree.R

@@ -5,7 +5,7 @@ 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)
 
-param <- list(max.depth=2,eta=1,silent=1,objective='binary:logistic')
+param <- list(max_depth=2, eta=1, silent=1, objective='binary:logistic')
 watchlist <- list(eval = dtest, train = dtrain)
 nround = 2
 

R-package/demo/predict_leaf_indices.R

@@ -10,7 +10,7 @@ data(agaricus.test, package='xgboost')
 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')
+param <- list(max_depth=2, eta=1, silent=1, objective='binary:logistic')
 nround = 4
 
 # training the model for two rounds
@@ -27,12 +27,12 @@ head(pred_with_leaf)
 create.new.tree.features <- function(model, original.features){
   pred_with_leaf <- predict(model, original.features, predleaf = TRUE)
   cols <- list()
-  for(i in 1:length(trees)){
+  for(i in 1:model$niter){
     # max is not the real max but it s not important for the purpose of adding features
     leaf.id <- sort(unique(pred_with_leaf[,i]))
     cols[[i]] <- factor(x = pred_with_leaf[,i], level = leaf.id)
   }
-  cBind(original.features, sparse.model.matrix( ~ . -1, as.data.frame(cols)))
+  cbind(original.features, sparse.model.matrix( ~ . -1, as.data.frame(cols)))
 }
 
 # Convert previous features to one hot encoding

R-package/demo/runall.R

@@ -10,3 +10,5 @@ demo(predict_leaf_indices)
 demo(early_stopping)
 demo(poisson_regression)
 demo(caret_wrapper)
+demo(tweedie_regression)
+#demo(gpu_accelerated) # can only run when built with GPU support

R-package/demo/tweedie_regression.R

@@ -0,0 +1,49 @@
+library(xgboost)
+library(data.table)
+library(cplm)
+
+data(AutoClaim)
+
+# auto insurance dataset analyzed by Yip and Yau (2005)
+dt <- data.table(AutoClaim)
+
+# exclude these columns from the model matrix
+exclude <-  c('POLICYNO', 'PLCYDATE', 'CLM_FREQ5', 'CLM_AMT5', 'CLM_FLAG', 'IN_YY')
+
+# retains the missing values
+# NOTE: this dataset is comes ready out of the box
+options(na.action = 'na.pass')
+x <- sparse.model.matrix(~ . - 1, data = dt[, -exclude, with = F])
+options(na.action = 'na.omit')
+
+# response
+y <- dt[, CLM_AMT5]
+
+d_train <- xgb.DMatrix(data = x, label = y, missing = NA)
+
+# the tweedie_variance_power parameter determines the shape of 
+# distribution
+# - closer to 1 is more poisson like and the mass
+#   is more concentrated near zero 
+# - closer to 2 is more gamma like and the mass spreads to the 
+#   the right with less concentration near zero
+
+params <- list(
+  objective = 'reg:tweedie',
+  eval_metric = 'rmse', 
+  tweedie_variance_power = 1.4,
+  max_depth = 6,
+  eta = 1)
+
+bst <- xgb.train(
+  data = d_train, 
+  params = params, 
+  maximize = FALSE,
+  watchlist = list(train = d_train), 
+  nrounds = 20)
+
+var_imp <- xgb.importance(attr(x, 'Dimnames')[[2]], model = bst)
+
+preds <- predict(bst, d_train)
+
+rmse <- sqrt(sum(mean((y - preds)^2)))

R-package/man/agaricus.test.Rd

@@ -29,4 +29,3 @@ Bache, K. & Lichman, M. (2013). UCI Machine Learning Repository
 School of Information and Computer Science.
 }
 \keyword{datasets}
-

R-package/man/agaricus.train.Rd

@@ -29,4 +29,3 @@ Bache, K. & Lichman, M. (2013). UCI Machine Learning Repository
 School of Information and Computer Science.
 }
 \keyword{datasets}
-

R-package/man/callbacks.Rd

@@ -0,0 +1,37 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/callbacks.R
+\name{callbacks}
+\alias{callbacks}
+\title{Callback closures for booster training.}
+\description{
+These are used to perform various service tasks either during boosting iterations or at the end.
+This approach helps to modularize many of such tasks without bloating the main training methods, 
+and it offers .
+}
+\details{
+By default, a callback function is run after each boosting iteration.
+An R-attribute \code{is_pre_iteration} could be set for a callback to define a pre-iteration function.
+
+When a callback function has \code{finalize} parameter, its finalizer part will also be run after 
+the boosting is completed.
+
+WARNING: side-effects!!! Be aware that these callback functions access and modify things in 
+the environment from which they are called from, which is a fairly uncommon thing to do in R.
+
+To write a custom callback closure, make sure you first understand the main concepts about R envoronments.
+Check either R documentation on \code{\link[base]{environment}} or the 
+\href{http://adv-r.had.co.nz/Environments.html}{Environments chapter} from the "Advanced R" 
+book by Hadley Wickham. Further, the best option is to read the code of some of the existing callbacks -
+choose ones that do something similar to what you want to achieve. Also, you would need to get familiar 
+with the objects available inside of the \code{xgb.train} and \code{xgb.cv} internal environments.
+}
+\seealso{
+\code{\link{cb.print.evaluation}},
+\code{\link{cb.evaluation.log}},
+\code{\link{cb.reset.parameters}},
+\code{\link{cb.early.stop}},
+\code{\link{cb.save.model}},
+\code{\link{cb.cv.predict}},
+\code{\link{xgb.train}},
+\code{\link{xgb.cv}}
+}

R-package/man/cb.cv.predict.Rd

@@ -0,0 +1,43 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/callbacks.R
+\name{cb.cv.predict}
+\alias{cb.cv.predict}
+\title{Callback closure for returning cross-validation based predictions.}
+\usage{
+cb.cv.predict(save_models = FALSE)
+}
+\arguments{
+\item{save_models}{a flag for whether to save the folds' models.}
+}
+\value{
+Predictions are returned inside of the \code{pred} element, which is either a vector or a matrix,
+depending on the number of prediction outputs per data row. The order of predictions corresponds 
+to the order of rows in the original dataset. Note that when a custom \code{folds} list is 
+provided in \code{xgb.cv}, the predictions would only be returned properly when this list is a 
+non-overlapping list of k sets of indices, as in a standard k-fold CV. The predictions would not be 
+meaningful when user-profided folds have overlapping indices as in, e.g., random sampling splits.
+When some of the indices in the training dataset are not included into user-provided \code{folds},
+their prediction value would be \code{NA}.
+}
+\description{
+Callback closure for returning cross-validation based predictions.
+}
+\details{
+This callback function saves predictions for all of the test folds,
+and also allows to save the folds' models.
+
+It is a "finalizer" callback and it uses early stopping information whenever it is available,
+thus it must be run after the early stopping callback if the early stopping is used.
+
+Callback function expects the following values to be set in its calling frame:
+\code{bst_folds},
+\code{basket},
+\code{data},
+\code{end_iteration},
+\code{params},
+\code{num_parallel_tree},
+\code{num_class}.
+}
+\seealso{
+\code{\link{callbacks}}
+}

R-package/man/cb.early.stop.Rd

@@ -0,0 +1,62 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/callbacks.R
+\name{cb.early.stop}
+\alias{cb.early.stop}
+\title{Callback closure to activate the early stopping.}
+\usage{
+cb.early.stop(stopping_rounds, maximize = FALSE, metric_name = NULL,
+  verbose = TRUE)
+}
+\arguments{
+\item{stopping_rounds}{The number of rounds with no improvement in 
+the evaluation metric in order to stop the training.}
+
+\item{maximize}{whether to maximize the evaluation metric}
+
+\item{metric_name}{the name of an evaluation column to use as a criteria for early
+stopping. If not set, the last column would be used.
+Let's say the test data in \code{watchlist} was labelled as \code{dtest}, 
+and one wants to use the AUC in test data for early stopping regardless of where 
+it is in the \code{watchlist}, then one of the following would need to be set:
+\code{metric_name='dtest-auc'} or \code{metric_name='dtest_auc'}.
+All dash '-' characters in metric names are considered equivalent to '_'.}
+
+\item{verbose}{whether to print the early stopping information.}
+}
+\description{
+Callback closure to activate the early stopping.
+}
+\details{
+This callback function determines the condition for early stopping 
+by setting the \code{stop_condition = TRUE} flag in its calling frame.
+
+The following additional fields are assigned to the model's R object:
+\itemize{
+\item \code{best_score} the evaluation score at the best iteration
+\item \code{best_iteration} at which boosting iteration the best score has occurred (1-based index)
+\item \code{best_ntreelimit} to use with the \code{ntreelimit} parameter in \code{predict}.
+     It differs from \code{best_iteration} in multiclass or random forest settings.
+}
+
+The Same values are also stored as xgb-attributes:
+\itemize{
+\item \code{best_iteration} is stored as a 0-based iteration index (for interoperability of binary models)
+\item \code{best_msg} message string is also stored.
+}
+
+At least one data element is required in the evaluation watchlist for early stopping to work.
+
+Callback function expects the following values to be set in its calling frame:
+\code{stop_condition},
+\code{bst_evaluation},
+\code{rank},
+\code{bst} (or \code{bst_folds} and \code{basket}),
+\code{iteration},
+\code{begin_iteration},
+\code{end_iteration},
+\code{num_parallel_tree}.
+}
+\seealso{
+\code{\link{callbacks}},
+\code{\link{xgb.attr}}
+}

R-package/man/cb.evaluation.log.Rd

@@ -0,0 +1,31 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/callbacks.R
+\name{cb.evaluation.log}
+\alias{cb.evaluation.log}
+\title{Callback closure for logging the evaluation history}
+\usage{
+cb.evaluation.log()
+}
+\description{
+Callback closure for logging the evaluation history
+}
+\details{
+This callback function appends the current iteration evaluation results \code{bst_evaluation}
+available in the calling parent frame to the \code{evaluation_log} list in a calling frame.
+
+The finalizer callback (called with \code{finalize = TURE} in the end) converts 
+the \code{evaluation_log} list into a final data.table.
+
+The iteration evaluation result \code{bst_evaluation} must be a named numeric vector. 
+
+Note: in the column names of the final data.table, the dash '-' character is replaced with 
+the underscore '_' in order to make the column names more like regular R identifiers.
+
+Callback function expects the following values to be set in its calling frame:
+\code{evaluation_log},
+\code{bst_evaluation},
+\code{iteration}.
+}
+\seealso{
+\code{\link{callbacks}}
+}

R-package/man/cb.gblinear.history.Rd

@@ -0,0 +1,95 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/callbacks.R
+\name{cb.gblinear.history}
+\alias{cb.gblinear.history}
+\title{Callback closure for collecting the model coefficients history of a gblinear booster
+during its training.}
+\usage{
+cb.gblinear.history(sparse = FALSE)
+}
+\arguments{
+\item{sparse}{when set to FALSE/TURE, a dense/sparse matrix is used to store the result.
+Sparse format is useful when one expects only a subset of coefficients to be non-zero,
+when using the "thrifty" feature selector with fairly small number of top features
+selected per iteration.}
+}
+\value{
+Results are stored in the \code{coefs} element of the closure.
+The \code{\link{xgb.gblinear.history}} convenience function provides an easy way to access it.
+With \code{xgb.train}, it is either a dense of a sparse matrix.
+While with \code{xgb.cv}, it is a list (an element per each fold) of such matrices.
+}
+\description{
+Callback closure for collecting the model coefficients history of a gblinear booster
+during its training.
+}
+\details{
+To keep things fast and simple, gblinear booster does not internally store the history of linear
+model coefficients at each boosting iteration. This callback provides a workaround for storing
+the coefficients' path, by extracting them after each training iteration.
+
+Callback function expects the following values to be set in its calling frame:
+\code{bst} (or \code{bst_folds}).
+}
+\examples{
+#### Binary classification:
+#
+# In the iris dataset, it is hard to linearly separate Versicolor class from the rest
+# without considering the 2nd order interactions:
+require(magrittr)
+x <- model.matrix(Species ~ .^2, iris)[,-1]
+colnames(x)
+dtrain <- xgb.DMatrix(scale(x), label = 1*(iris$Species == "versicolor"))
+param <- list(booster = "gblinear", objective = "reg:logistic", eval_metric = "auc",
+              lambda = 0.0003, alpha = 0.0003, nthread = 2)
+# For 'shotgun', which is a default linear updater, using high eta values may result in
+# unstable behaviour in some datasets. With this simple dataset, however, the high learning
+# rate does not break the convergence, but allows us to illustrate the typical pattern of
+# "stochastic explosion" behaviour of this lock-free algorithm at early boosting iterations.
+bst <- xgb.train(param, dtrain, list(tr=dtrain), nrounds = 200, eta = 1.,
+                 callbacks = list(cb.gblinear.history()))
+# Extract the coefficients' path and plot them vs boosting iteration number:
+coef_path <- xgb.gblinear.history(bst)
+matplot(coef_path, type = 'l')
+
+# With the deterministic coordinate descent updater, it is safer to use higher learning rates.
+# Will try the classical componentwise boosting which selects a single best feature per round:
+bst <- xgb.train(param, dtrain, list(tr=dtrain), nrounds = 200, eta = 0.8,
+                 updater = 'coord_descent', feature_selector = 'thrifty', top_k = 1,
+                 callbacks = list(cb.gblinear.history()))
+xgb.gblinear.history(bst) \%>\% matplot(type = 'l')
+#  Componentwise boosting is known to have similar effect to Lasso regularization.
+# Try experimenting with various values of top_k, eta, nrounds,
+# as well as different feature_selectors.
+
+# For xgb.cv:
+bst <- xgb.cv(param, dtrain, nfold = 5, nrounds = 100, eta = 0.8,
+             callbacks = list(cb.gblinear.history()))
+# coefficients in the CV fold #3
+xgb.gblinear.history(bst)[[3]] \%>\% matplot(type = 'l')
+
+
+#### Multiclass classification:
+#
+dtrain <- xgb.DMatrix(scale(x), label = as.numeric(iris$Species) - 1)
+param <- list(booster = "gblinear", objective = "multi:softprob", num_class = 3,
+              lambda = 0.0003, alpha = 0.0003, nthread = 2)
+# For the default linear updater 'shotgun' it sometimes is helpful
+# to use smaller eta to reduce instability
+bst <- xgb.train(param, dtrain, list(tr=dtrain), nrounds = 70, eta = 0.5,
+                 callbacks = list(cb.gblinear.history()))
+# Will plot the coefficient paths separately for each class:
+xgb.gblinear.history(bst, class_index = 0) \%>\% matplot(type = 'l')
+xgb.gblinear.history(bst, class_index = 1) \%>\% matplot(type = 'l')
+xgb.gblinear.history(bst, class_index = 2) \%>\% matplot(type = 'l')
+
+# CV:
+bst <- xgb.cv(param, dtrain, nfold = 5, nrounds = 70, eta = 0.5,
+              callbacks = list(cb.gblinear.history(FALSE)))
+# 1st forld of 1st class
+xgb.gblinear.history(bst, class_index = 0)[[1]] \%>\% matplot(type = 'l')
+
+}
+\seealso{
+\code{\link{callbacks}}, \code{\link{xgb.gblinear.history}}.
+}

R-package/man/cb.print.evaluation.Rd

@@ -0,0 +1,29 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/callbacks.R
+\name{cb.print.evaluation}
+\alias{cb.print.evaluation}
+\title{Callback closure for printing the result of evaluation}
+\usage{
+cb.print.evaluation(period = 1, showsd = TRUE)
+}
+\arguments{
+\item{period}{results would be printed every number of periods}
+
+\item{showsd}{whether standard deviations should be printed (when available)}
+}
+\description{
+Callback closure for printing the result of evaluation
+}
+\details{
+The callback function prints the result of evaluation at every \code{period} iterations.
+The initial and the last iteration's evaluations are always printed.
+
+Callback function expects the following values to be set in its calling frame:
+\code{bst_evaluation} (also \code{bst_evaluation_err} when available),
+\code{iteration},
+\code{begin_iteration},
+\code{end_iteration}.
+}
+\seealso{
+\code{\link{callbacks}}
+}

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

@@ -0,0 +1,36 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/callbacks.R
+\name{cb.reset.parameters}
+\alias{cb.reset.parameters}
+\title{Callback closure for restetting the booster's parameters at each iteration.}
+\usage{
+cb.reset.parameters(new_params)
+}
+\arguments{
+\item{new_params}{a list where each element corresponds to a parameter that needs to be reset.
+Each element's value must be either a vector of values of length \code{nrounds} 
+to be set at each iteration, 
+or a function of two parameters \code{learning_rates(iteration, nrounds)} 
+which returns a new parameter value by using the current iteration number 
+and the total number of boosting rounds.}
+}
+\description{
+Callback closure for restetting the booster's parameters at each iteration.
+}
+\details{
+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{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:
+\code{bst} or \code{bst_folds},
+\code{iteration},
+\code{begin_iteration},
+\code{end_iteration}.
+}
+\seealso{
+\code{\link{callbacks}}
+}

R-package/man/cb.save.model.Rd

@@ -0,0 +1,33 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/callbacks.R
+\name{cb.save.model}
+\alias{cb.save.model}
+\title{Callback closure for saving a model file.}
+\usage{
+cb.save.model(save_period = 0, save_name = "xgboost.model")
+}
+\arguments{
+\item{save_period}{save the model to disk after every 
+\code{save_period} iterations; 0 means save the model at the end.}
+
+\item{save_name}{the name or path for the saved model file.
+It can contain a \code{\link[base]{sprintf}} formatting specifier 
+to include the integer iteration number in the file name.
+E.g., with \code{save_name} = 'xgboost_%04d.model', 
+the file saved at iteration 50 would be named "xgboost_0050.model".}
+}
+\description{
+Callback closure for saving a model file.
+}
+\details{
+This callback function allows to save an xgb-model file, either periodically after each \code{save_period}'s or at the end.
+
+Callback function expects the following values to be set in its calling frame:
+\code{bst},
+\code{iteration},
+\code{begin_iteration},
+\code{end_iteration}.
+}
+\seealso{
+\code{\link{callbacks}}
+}

R-package/man/dim.xgb.DMatrix.Rd

@@ -0,0 +1,28 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/xgb.DMatrix.R
+\name{dim.xgb.DMatrix}
+\alias{dim.xgb.DMatrix}
+\title{Dimensions of xgb.DMatrix}
+\usage{
+\method{dim}{xgb.DMatrix}(x)
+}
+\arguments{
+\item{x}{Object of class \code{xgb.DMatrix}}
+}
+\description{
+Returns a vector of numbers of rows and of columns in an \code{xgb.DMatrix}.
+}
+\details{
+Note: since \code{nrow} and \code{ncol} internally use \code{dim}, they can also 
+be directly used with an \code{xgb.DMatrix} object.
+}
+\examples{
+data(agaricus.train, package='xgboost')
+train <- agaricus.train
+dtrain <- xgb.DMatrix(train$data, label=train$label)
+
+stopifnot(nrow(dtrain) == nrow(train$data))
+stopifnot(ncol(dtrain) == ncol(train$data))
+stopifnot(all(dim(dtrain) == dim(train$data)))
+
+}

R-package/man/dimnames.xgb.DMatrix.Rd

@@ -0,0 +1,35 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/xgb.DMatrix.R
+\name{dimnames.xgb.DMatrix}
+\alias{dimnames.xgb.DMatrix}
+\alias{dimnames<-.xgb.DMatrix}
+\title{Handling of column names of \code{xgb.DMatrix}}
+\usage{
+\method{dimnames}{xgb.DMatrix}(x)
+
+\method{dimnames}{xgb.DMatrix}(x) <- value
+}
+\arguments{
+\item{x}{object of class \code{xgb.DMatrix}}
+
+\item{value}{a list of two elements: the first one is ignored
+and the second one is column names}
+}
+\description{
+Only column names are supported for \code{xgb.DMatrix}, thus setting of 
+row names would have no effect and returnten row names would be NULL.
+}
+\details{
+Generic \code{dimnames} methods are used by \code{colnames}.
+Since row names are irrelevant, it is recommended to use \code{colnames} directly.
+}
+\examples{
+data(agaricus.train, package='xgboost')
+train <- agaricus.train
+dtrain <- xgb.DMatrix(train$data, label=train$label)
+dimnames(dtrain)
+colnames(dtrain)
+colnames(dtrain) <- make.names(1:ncol(train$data))
+print(dtrain, verbose=TRUE)
+
+}

R-package/man/edge.parser.Rd

@@ -1,15 +0,0 @@
-% Generated by roxygen2: do not edit by hand
-% Please edit documentation in R/xgb.plot.deepness.R
-\name{edge.parser}
-\alias{edge.parser}
-\title{Parse the graph to extract vector of edges}
-\usage{
-edge.parser(element)
-}
-\arguments{
-\item{element}{igraph object containing the path from the root to the leaf.}
-}
-\description{
-Parse the graph to extract vector of edges
-}
-

R-package/man/get.paths.to.leaf.Rd

@@ -1,15 +0,0 @@
-% Generated by roxygen2: do not edit by hand
-% Please edit documentation in R/xgb.plot.deepness.R
-\name{get.paths.to.leaf}
-\alias{get.paths.to.leaf}
-\title{Extract path from root to leaf from data.table}
-\usage{
-get.paths.to.leaf(dt.tree)
-}
-\arguments{
-\item{dt.tree}{data.table containing the nodes and edges of the trees}
-}
-\description{
-Extract path from root to leaf from data.table
-}
-

R-package/man/getinfo.Rd

@@ -1,42 +1,45 @@
 % Generated by roxygen2: do not edit by hand
-% Please edit documentation in R/getinfo.xgb.DMatrix.R
-\docType{methods}
+% Please edit documentation in R/xgb.DMatrix.R
 \name{getinfo}
 \alias{getinfo}
-\alias{getinfo,xgb.DMatrix-method}
+\alias{getinfo.xgb.DMatrix}
 \title{Get information of an xgb.DMatrix object}
 \usage{
 getinfo(object, ...)
 
-\S4method{getinfo}{xgb.DMatrix}(object, name)
+\method{getinfo}{xgb.DMatrix}(object, name, ...)
 }
 \arguments{
 \item{object}{Object of class \code{xgb.DMatrix}}
 
 \item{...}{other parameters}
 
-\item{name}{the name of the field to get}
+\item{name}{the name of the information field to get (see details)}
 }
 \description{
 Get information of an xgb.DMatrix object
 }
 \details{
-The information can be one of the following:
+The \code{name} field can be one of the following:
 
 \itemize{
     \item \code{label}: label Xgboost learn from ;
     \item \code{weight}: to do a weight rescale ;
     \item \code{base_margin}: base margin is the base prediction Xgboost will boost from ;
     \item \code{nrow}: number of rows of the \code{xgb.DMatrix}.
+    
 }
+
+\code{group} can be setup by \code{setinfo} but can't be retrieved by \code{getinfo}.
 }
 \examples{
 data(agaricus.train, package='xgboost')
 train <- agaricus.train
 dtrain <- xgb.DMatrix(train$data, label=train$label)
+
 labels <- getinfo(dtrain, 'label')
 setinfo(dtrain, 'label', 1-labels)
+
 labels2 <- getinfo(dtrain, 'label')
 stopifnot(all(labels2 == 1-labels))
 }
-

R-package/man/multiplot.Rd

@@ -1,15 +0,0 @@
-% Generated by roxygen2: do not edit by hand
-% Please edit documentation in R/xgb.plot.deepness.R
-\name{multiplot}
-\alias{multiplot}
-\title{Plot multiple graphs at the same time}
-\usage{
-multiplot(..., cols = 1)
-}
-\arguments{
-\item{cols}{number of columns}
-}
-\description{
-Plot multiple graph aligned by rows and columns.
-}
-

R-package/man/nrow-xgb.DMatrix-method.Rd

@@ -1,23 +0,0 @@
-% Generated by roxygen2: do not edit by hand
-% Please edit documentation in R/nrow.xgb.DMatrix.R
-\docType{methods}
-\name{nrow,xgb.DMatrix-method}
-\alias{nrow,xgb.DMatrix-method}
-\title{Number of xgb.DMatrix rows}
-\usage{
-\S4method{nrow}{xgb.DMatrix}(x)
-}
-\arguments{
-\item{x}{Object of class \code{xgb.DMatrix}}
-}
-\description{
-\code{nrow} return the number of rows present in the \code{xgb.DMatrix}.
-}
-\examples{
-data(agaricus.train, package='xgboost')
-train <- agaricus.train
-dtrain <- xgb.DMatrix(train$data, label=train$label)
-stopifnot(nrow(dtrain) == nrow(train$data))
-
-}
-

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

@@ -1,53 +0,0 @@
-% Generated by roxygen2: do not edit by hand
-% Please edit documentation in R/predict.xgb.Booster.R
-\docType{methods}
-\name{predict,xgb.Booster-method}
-\alias{predict,xgb.Booster-method}
-\title{Predict method for eXtreme Gradient Boosting model}
-\usage{
-\S4method{predict}{xgb.Booster}(object, newdata, missing = NA,
-  outputmargin = FALSE, ntreelimit = NULL, predleaf = FALSE)
-}
-\arguments{
-\item{object}{Object of class "xgb.Boost"}
-
-\item{newdata}{takes \code{matrix}, \code{dgCMatrix}, local data file or 
-\code{xgb.DMatrix}.}
-
-\item{missing}{Missing is only used when input is dense matrix, pick a float 
-value that represents missing value. Sometime a data use 0 or other extreme value to represents missing values.}
-
-\item{outputmargin}{whether the prediction should be shown in the original
-value of sum of functions, when outputmargin=TRUE, the prediction is 
-untransformed margin value. In logistic regression, outputmargin=T will
-output value before logistic transformation.}
-
-\item{ntreelimit}{limit number of trees used in prediction, this parameter is
-only valid for gbtree, but not for gblinear. set it to be value bigger 
-than 0. It will use all trees by default.}
-
-\item{predleaf}{whether predict leaf index instead. If set to TRUE, the output will be a matrix object.}
-}
-\description{
-Predicted values based on xgboost model object.
-}
-\details{
-The option \code{ntreelimit} purpose is to let the user train a model with lots 
-of trees but use only the first trees for prediction to avoid overfitting 
-(without having to train a new model with less trees).
-
-The option \code{predleaf} purpose is inspired from §3.1 of the paper 
-\code{Practical Lessons from Predicting Clicks on Ads at Facebook}.
-The idea is to use the model as a generator of new features which capture non linear link 
-from original features.
-}
-\examples{
-data(agaricus.train, package='xgboost')
-data(agaricus.test, package='xgboost')
-train <- agaricus.train
-test <- agaricus.test
-bst <- xgboost(data = train$data, label = train$label, max.depth = 2, 
-               eta = 1, nthread = 2, nround = 2,objective = "binary:logistic")
-pred <- predict(bst, test$data)
-}
-

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

@@ -1,18 +0,0 @@
-% Generated by roxygen2: do not edit by hand
-% Please edit documentation in R/predict.xgb.Booster.handle.R
-\docType{methods}
-\name{predict,xgb.Booster.handle-method}
-\alias{predict,xgb.Booster.handle-method}
-\title{Predict method for eXtreme Gradient Boosting model handle}
-\usage{
-\S4method{predict}{xgb.Booster.handle}(object, ...)
-}
-\arguments{
-\item{object}{Object of class "xgb.Boost.handle"}
-
-\item{...}{Parameters pass to \code{predict.xgb.Booster}}
-}
-\description{
-Predicted values based on xgb.Booster.handle object.
-}
-

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

@@ -0,0 +1,172 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/xgb.Booster.R
+\name{predict.xgb.Booster}
+\alias{predict.xgb.Booster}
+\alias{predict.xgb.Booster.handle}
+\title{Predict method for eXtreme Gradient Boosting model}
+\usage{
+\method{predict}{xgb.Booster}(object, newdata, missing = NA,
+  outputmargin = FALSE, ntreelimit = NULL, predleaf = FALSE,
+  predcontrib = FALSE, approxcontrib = FALSE, reshape = FALSE, ...)
+
+\method{predict}{xgb.Booster.handle}(object, ...)
+}
+\arguments{
+\item{object}{Object of class \code{xgb.Booster} or \code{xgb.Booster.handle}}
+
+\item{newdata}{takes \code{matrix}, \code{dgCMatrix}, local data file or \code{xgb.DMatrix}.}
+
+\item{missing}{Missing is only used when input is dense matrix. Pick a float value that represents
+missing values in data (e.g., sometimes 0 or some other extreme value is used).}
+
+\item{outputmargin}{whether the prediction should be returned in the for of original untransformed
+sum of predictions from boosting iterations' results. E.g., setting \code{outputmargin=TRUE} for
+logistic regression would result in predictions for log-odds instead of probabilities.}
+
+\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 instead.}
+
+\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{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 \code{predleaf = TRUE}.}
+
+\item{...}{Parameters passed to \code{predict.xgb.Booster}}
+}
+\value{
+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
+a \code{(nrows(newdata), num_class)} dimension matrix is returned, depending on
+the \code{reshape} value.
+
+When \code{predleaf = TRUE}, the output is a matrix object with the
+number of columns corresponding to the number of trees.
+
+When \code{predcontrib = TRUE} and it is not a multiclass setting, the output is a matrix object with
+\code{num_features + 1} columns. The last "+ 1" column in a matrix corresponds to bias.
+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).
+}
+\description{
+Predicted values based on either xgboost model or model handle object.
+}
+\details{
+Note that \code{ntreelimit} is not necessarily equal to the number of boosting iterations
+and it is not necessarily equal to the number of trees in a model.
+E.g., in a random forest-like model, \code{ntreelimit} would limit the number of trees.
+But for multiclass classification, while there are multiple trees per iteration,
+\code{ntreelimit} limits the number of boosting iterations.
+
+Also note that \code{ntreelimit} would currently do nothing for predictions from gblinear,
+since gblinear doesn't keep its boosting history.
+
+One possible practical applications of the \code{predleaf} option is to use the model
+as a generator of new features which capture non-linearity and interactions,
+e.g., as implemented in \code{\link{xgb.create.features}}.
+
+Setting \code{predcontrib = TRUE} allows to calculate contributions of each feature to
+individual predictions. For "gblinear" booster, feature contributions are simply linear terms
+(feature_beta * feature_value). For "gbtree" booster, feature contributions are SHAP
+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/}.
+}
+\examples{
+## binary classification:
+
+data(agaricus.train, package='xgboost')
+data(agaricus.test, package='xgboost')
+train <- agaricus.train
+test <- agaricus.test
+
+bst <- xgboost(data = train$data, label = train$label, max_depth = 2,
+               eta = 0.5, nthread = 2, nrounds = 5, objective = "binary:logistic")
+# use all trees by default
+pred <- predict(bst, test$data)
+# use only the 1st tree
+pred1 <- predict(bst, test$data, ntreelimit = 1)
+
+# Predicting tree leafs:
+# the result is an nsamples X ntrees matrix
+pred_leaf <- predict(bst, test$data, predleaf = TRUE)
+str(pred_leaf)
+
+# Predicting feature contributions to predictions:
+# the result is an nsamples X (nfeatures + 1) matrix
+pred_contr <- predict(bst, test$data, predcontrib = TRUE)
+str(pred_contr)
+# verify that contributions' sums are equal to log-odds of predictions (up to float precision):
+summary(rowSums(pred_contr) - qlogis(pred))
+# for the 1st record, let's inspect its features that had non-zero contribution to prediction:
+contr1 <- pred_contr[1,]
+contr1 <- contr1[-length(contr1)]    # drop BIAS
+contr1 <- contr1[contr1 != 0]        # drop non-contributing features
+contr1 <- contr1[order(abs(contr1))] # order by contribution magnitude
+old_mar <- par("mar")
+par(mar = old_mar + c(0,7,0,0))
+barplot(contr1, horiz = TRUE, las = 2, xlab = "contribution to prediction in log-odds")
+par(mar = old_mar)
+
+
+## multiclass classification in iris dataset:
+
+lb <- as.numeric(iris$Species) - 1
+num_class <- 3
+set.seed(11)
+bst <- xgboost(data = as.matrix(iris[, -5]), label = lb,
+               max_depth = 4, eta = 0.5, nthread = 2, nrounds = 10, subsample = 0.5,
+               objective = "multi:softprob", num_class = num_class)
+# predict for softmax returns num_class probability numbers per case:
+pred <- predict(bst, as.matrix(iris[, -5]))
+str(pred)
+# reshape it to a num_class-columns matrix
+pred <- matrix(pred, ncol=num_class, byrow=TRUE)
+# convert the probabilities to softmax labels
+pred_labels <- max.col(pred) - 1
+# the following should result in the same error as seen in the last iteration
+sum(pred_labels != lb)/length(lb)
+
+# compare that to the predictions from softmax:
+set.seed(11)
+bst <- xgboost(data = as.matrix(iris[, -5]), label = lb,
+               max_depth = 4, eta = 0.5, nthread = 2, nrounds = 10, subsample = 0.5,
+               objective = "multi:softmax", num_class = num_class)
+pred <- predict(bst, as.matrix(iris[, -5]))
+str(pred)
+all.equal(pred, pred_labels)
+# prediction from using only 5 iterations should result
+# in the same error as seen in iteration 5:
+pred5 <- predict(bst, as.matrix(iris[, -5]), ntreelimit=5)
+sum(pred5 != lb)/length(lb)
+
+
+## random forest-like model of 25 trees for binary classification:
+
+set.seed(11)
+bst <- xgboost(data = train$data, label = train$label, max_depth = 5,
+               nthread = 2, nrounds = 1, objective = "binary:logistic",
+               num_parallel_tree = 25, subsample = 0.6, colsample_bytree = 0.1)
+# Inspect the prediction error vs number of trees:
+lb <- test$label
+dtest <- xgb.DMatrix(test$data, label=lb)
+err <- sapply(1:25, function(n) {
+  pred <- predict(bst, dtest, ntreelimit=n)
+  sum((pred > 0.5) != lb)/length(lb)
+})
+plot(err, type='l', ylim=c(0,0.1), xlab='#trees')
+
+}
+\references{
+Scott M. Lundberg, Su-In Lee, "A Unified Approach to Interpreting Model Predictions", NIPS Proceedings 2017, \url{https://arxiv.org/abs/1705.07874}
+
+Scott M. Lundberg, Su-In Lee, "Consistent feature attribution for tree ensembles", \url{https://arxiv.org/abs/1706.06060}
+}
+\seealso{
+\code{\link{xgb.train}}.
+}

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

@@ -0,0 +1,29 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/xgb.Booster.R
+\name{print.xgb.Booster}
+\alias{print.xgb.Booster}
+\title{Print xgb.Booster}
+\usage{
+\method{print}{xgb.Booster}(x, verbose = FALSE, ...)
+}
+\arguments{
+\item{x}{an xgb.Booster object}
+
+\item{verbose}{whether to print detailed data (e.g., attribute values)}
+
+\item{...}{not currently used}
+}
+\description{
+Print information about xgb.Booster.
+}
+\examples{
+data(agaricus.train, package='xgboost')
+train <- agaricus.train
+bst <- xgboost(data = train$data, label = train$label, max_depth = 2,
+               eta = 1, nthread = 2, nrounds = 2, objective = "binary:logistic")
+attr(bst, 'myattr') <- 'memo'
+
+print(bst)
+print(bst, verbose=TRUE)
+
+}

R-package/man/print.xgb.DMatrix.Rd

@@ -0,0 +1,28 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/xgb.DMatrix.R
+\name{print.xgb.DMatrix}
+\alias{print.xgb.DMatrix}
+\title{Print xgb.DMatrix}
+\usage{
+\method{print}{xgb.DMatrix}(x, verbose = FALSE, ...)
+}
+\arguments{
+\item{x}{an xgb.DMatrix object}
+
+\item{verbose}{whether to print colnames (when present)}
+
+\item{...}{not currently used}
+}
+\description{
+Print information about xgb.DMatrix. 
+Currently it displays dimensions and presence of info-fields and colnames.
+}
+\examples{
+data(agaricus.train, package='xgboost')
+train <- agaricus.train
+dtrain <- xgb.DMatrix(train$data, label=train$label)
+
+dtrain
+print(dtrain, verbose=TRUE)
+
+}

R-package/man/print.xgb.cv.Rd

@@ -0,0 +1,31 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/xgb.cv.R
+\name{print.xgb.cv.synchronous}
+\alias{print.xgb.cv.synchronous}
+\title{Print xgb.cv result}
+\usage{
+\method{print}{xgb.cv.synchronous}(x, verbose = FALSE, ...)
+}
+\arguments{
+\item{x}{an \code{xgb.cv.synchronous} object}
+
+\item{verbose}{whether to print detailed data}
+
+\item{...}{passed to \code{data.table.print}}
+}
+\description{
+Prints formatted results of \code{xgb.cv}.
+}
+\details{
+When not verbose, it would only print the evaluation results, 
+including the best iteration (when available).
+}
+\examples{
+data(agaricus.train, package='xgboost')
+train <- agaricus.train
+cv <- xgb.cv(data = train$data, label = train$label, nfold = 5, max_depth = 2,
+               eta = 1, nthread = 2, nrounds = 2, objective = "binary:logistic")
+print(cv)
+print(cv, verbose=TRUE)
+
+}

R-package/man/setinfo.Rd

@@ -1,14 +1,13 @@
 % Generated by roxygen2: do not edit by hand
-% Please edit documentation in R/setinfo.xgb.DMatrix.R
-\docType{methods}
+% Please edit documentation in R/xgb.DMatrix.R
 \name{setinfo}
 \alias{setinfo}
-\alias{setinfo,xgb.DMatrix-method}
+\alias{setinfo.xgb.DMatrix}
 \title{Set information of an xgb.DMatrix object}
 \usage{
 setinfo(object, ...)
 
-\S4method{setinfo}{xgb.DMatrix}(object, name, info)
+\method{setinfo}{xgb.DMatrix}(object, name, info, ...)
 }
 \arguments{
 \item{object}{Object of class "xgb.DMatrix"}
@@ -23,22 +22,22 @@ setinfo(object, ...)
 Set information of an xgb.DMatrix object
 }
 \details{
-It can be one of the following:
+The \code{name} field can be one of the following:
 
 \itemize{
     \item \code{label}: label Xgboost learn from ;
     \item \code{weight}: to do a weight rescale ;
     \item \code{base_margin}: base margin is the base prediction Xgboost will boost from ;
-    \item \code{group}.
+    \item \code{group}: number of rows in each group (to use with \code{rank:pairwise} objective).
 }
 }
 \examples{
 data(agaricus.train, package='xgboost')
 train <- agaricus.train
 dtrain <- xgb.DMatrix(train$data, label=train$label)
+
 labels <- getinfo(dtrain, 'label')
 setinfo(dtrain, 'label', 1-labels)
 labels2 <- getinfo(dtrain, 'label')
-stopifnot(all(labels2 == 1-labels))
+stopifnot(all.equal(labels2, 1-labels))
 }
-

R-package/man/slice.xgb.DMatrix.Rd

@@ -1,22 +1,26 @@
 % Generated by roxygen2: do not edit by hand
-% Please edit documentation in R/slice.xgb.DMatrix.R
-\docType{methods}
+% Please edit documentation in R/xgb.DMatrix.R
 \name{slice}
 \alias{slice}
-\alias{slice,xgb.DMatrix-method}
+\alias{slice.xgb.DMatrix}
+\alias{[.xgb.DMatrix}
 \title{Get a new DMatrix containing the specified rows of
 orginal xgb.DMatrix object}
 \usage{
 slice(object, ...)
 
-\S4method{slice}{xgb.DMatrix}(object, idxset, ...)
+\method{slice}{xgb.DMatrix}(object, idxset, ...)
+
+\method{[}{xgb.DMatrix}(object, idxset, colset = NULL)
 }
 \arguments{
 \item{object}{Object of class "xgb.DMatrix"}
 
-\item{...}{other parameters}
+\item{...}{other parameters (currently not used)}
 
 \item{idxset}{a integer vector of indices of rows needed}
+
+\item{colset}{currently not used (columns subsetting is not available)}
 }
 \description{
 Get a new DMatrix containing the specified rows of
@@ -26,6 +30,11 @@ orginal xgb.DMatrix object
 data(agaricus.train, package='xgboost')
 train <- agaricus.train
 dtrain <- xgb.DMatrix(train$data, label=train$label)
-dsub <- slice(dtrain, 1:3)
-}
 
+dsub <- slice(dtrain, 1:42)
+labels1 <- getinfo(dsub, 'label')
+dsub <- dtrain[1:42, ]
+labels2 <- getinfo(dsub, 'label')
+all.equal(labels1, labels2)
+
+}

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

@@ -0,0 +1,49 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/xgb.Booster.R
+\name{xgb.Booster.complete}
+\alias{xgb.Booster.complete}
+\title{Restore missing parts of an incomplete xgb.Booster object.}
+\usage{
+xgb.Booster.complete(object, saveraw = TRUE)
+}
+\arguments{
+\item{object}{object of class \code{xgb.Booster}}
+
+\item{saveraw}{a flag indicating whether to append \code{raw} Booster memory dump data
+when it doesn't already exist.}
+}
+\value{
+An object of \code{xgb.Booster} class.
+}
+\description{
+It attempts to complete an \code{xgb.Booster} object by restoring either its missing
+raw model memory dump (when it has no \code{raw} data but its \code{xgb.Booster.handle} is valid)
+or its missing internal handle (when its \code{xgb.Booster.handle} is not valid
+but it has a raw Booster memory dump).
+}
+\details{
+While this method is primarily for internal use, it might be useful in some practical situations.
+
+E.g., when an \code{xgb.Booster} model is saved as an R object and then is loaded as an R object,
+its handle (pointer) to an internal xgboost model would be invalid. The majority of xgboost methods
+should still work for such a model object since those methods would be using
+\code{xgb.Booster.complete} internally. However, one might find it to be more efficient to call the
+\code{xgb.Booster.complete} function explicitely once after loading a model as an R-object.
+That would prevent further repeated implicit reconstruction of an internal booster model.
+}
+\examples{
+
+data(agaricus.train, package='xgboost')
+bst <- xgboost(data = agaricus.train$data, label = agaricus.train$label, max_depth = 2,
+               eta = 1, nthread = 2, nrounds = 2, objective = "binary:logistic")
+saveRDS(bst, "xgb.model.rds")
+
+bst1 <- readRDS("xgb.model.rds")
+# the handle is invalid:
+print(bst1$handle)
+
+bst1 <- xgb.Booster.complete(bst1)
+# now the handle points to a valid internal booster model:
+print(bst1$handle)
+
+}

R-package/man/xgb.DMatrix.Rd

@@ -2,23 +2,28 @@
 % Please edit documentation in R/xgb.DMatrix.R
 \name{xgb.DMatrix}
 \alias{xgb.DMatrix}
-\title{Contruct xgb.DMatrix object}
+\title{Construct xgb.DMatrix object}
 \usage{
-xgb.DMatrix(data, info = list(), missing = NA, ...)
+xgb.DMatrix(data, info = list(), missing = NA, silent = FALSE, ...)
 }
 \arguments{
-\item{data}{a \code{matrix} object, a \code{dgCMatrix} object or a character 
-indicating the data file.}
+\item{data}{a \code{matrix} object (either numeric or integer), a \code{dgCMatrix} object, or a character 
+string representing a filename.}
 
-\item{info}{a list of information of the xgb.DMatrix object}
+\item{info}{a named list of additional information to store in the \code{xgb.DMatrix} object.
+See \code{\link{setinfo}} for the specific allowed kinds of}
 
-\item{missing}{Missing is only used when input is dense matrix, pick a float
-value that represents missing value. Sometime a data use 0 or other extreme value to represents missing values.}
+\item{missing}{a float value to represents missing values in data (used only when input is a dense matrix).
+It is useful when a 0 or some other extreme value represents missing values in data.}
 
-\item{...}{other information to pass to \code{info}.}
+\item{silent}{whether to suppress printing an informational message after loading from a file.}
+
+\item{...}{the \code{info} data could be passed directly as parameters, without creating an \code{info} list.}
 }
 \description{
-Contruct xgb.DMatrix object from dense matrix, sparse matrix or local file.
+Construct xgb.DMatrix object from either a dense matrix, a sparse matrix, or a local file.
+Supported input file formats are either a libsvm text file or a binary file that was created previously by
+\code{\link{xgb.DMatrix.save}}).
 }
 \examples{
 data(agaricus.train, package='xgboost')
@@ -27,4 +32,3 @@ dtrain <- xgb.DMatrix(train$data, label=train$label)
 xgb.DMatrix.save(dtrain, 'xgb.DMatrix.data')
 dtrain <- xgb.DMatrix('xgb.DMatrix.data')
 }
-

R-package/man/xgb.DMatrix.save.Rd

@@ -4,12 +4,12 @@
 \alias{xgb.DMatrix.save}
 \title{Save xgb.DMatrix object to binary file}
 \usage{
-xgb.DMatrix.save(DMatrix, fname)
+xgb.DMatrix.save(dmatrix, fname)
 }
 \arguments{
-\item{DMatrix}{the DMatrix object}
+\item{dmatrix}{the \code{xgb.DMatrix} object}
 
-\item{fname}{the name of the binary file.}
+\item{fname}{the name of the file to write.}
 }
 \description{
 Save xgb.DMatrix object to binary file
@@ -21,4 +21,3 @@ dtrain <- xgb.DMatrix(train$data, label=train$label)
 xgb.DMatrix.save(dtrain, 'xgb.DMatrix.data')
 dtrain <- xgb.DMatrix('xgb.DMatrix.data')
 }
-

R-package/man/xgb.attr.Rd

@@ -0,0 +1,85 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/xgb.Booster.R
+\name{xgb.attr}
+\alias{xgb.attr}
+\alias{xgb.attr<-}
+\alias{xgb.attributes}
+\alias{xgb.attributes<-}
+\title{Accessors for serializable attributes of a model.}
+\usage{
+xgb.attr(object, name)
+
+xgb.attr(object, name) <- value
+
+xgb.attributes(object)
+
+xgb.attributes(object) <- value
+}
+\arguments{
+\item{object}{Object of class \code{xgb.Booster} or \code{xgb.Booster.handle}.}
+
+\item{name}{a non-empty character string specifying which attribute is to be accessed.}
+
+\item{value}{a value of an attribute for \code{xgb.attr<-}; for \code{xgb.attributes<-}
+it's a list (or an object coercible to a list) with the names of attributes to set
+and the elements corresponding to attribute values.
+Non-character values are converted to character.
+When attribute value is not a scalar, only the first index is used.
+Use \code{NULL} to remove an attribute.}
+}
+\value{
+\code{xgb.attr} returns either a string value of an attribute
+or \code{NULL} if an attribute wasn't stored in a model.
+
+\code{xgb.attributes} returns a list of all attribute stored in a model
+or \code{NULL} if a model has no stored attributes.
+}
+\description{
+These methods allow to manipulate the key-value attribute strings of an xgboost model.
+}
+\details{
+The primary purpose of xgboost model attributes is to store some meta-data about the model.
+Note that they are a separate concept from the object attributes in R.
+Specifically, they refer to key-value strings that can be attached to an xgboost model,
+stored together with the model's binary representation, and accessed later
+(from R or any other interface).
+In contrast, any R-attribute assigned to an R-object of \code{xgb.Booster} class
+would not be saved by \code{xgb.save} because an xgboost model is an external memory object
+and its serialization is handled externally.
+Also, setting an attribute that has the same name as one of xgboost's parameters wouldn't
+change the value of that parameter for a model.
+Use \code{\link{xgb.parameters<-}} to set or change model parameters.
+
+The attribute setters would usually work more efficiently for \code{xgb.Booster.handle}
+than for \code{xgb.Booster}, since only just a handle (pointer) would need to be copied.
+That would only matter if attributes need to be set many times.
+Note, however, that when feeding a handle of an \code{xgb.Booster} object to the attribute setters,
+the raw model cache of an \code{xgb.Booster} object would not be automatically updated,
+and it would be user's responsibility to call \code{xgb.save.raw} to update it.
+
+The \code{xgb.attributes<-} setter either updates the existing or adds one or several attributes,
+but it doesn't delete the other existing attributes.
+}
+\examples{
+data(agaricus.train, package='xgboost')
+train <- agaricus.train
+
+bst <- xgboost(data = train$data, label = train$label, max_depth = 2,
+               eta = 1, nthread = 2, nrounds = 2, objective = "binary:logistic")
+
+xgb.attr(bst, "my_attribute") <- "my attribute value"
+print(xgb.attr(bst, "my_attribute"))
+xgb.attributes(bst) <- list(a = 123, b = "abc")
+
+xgb.save(bst, 'xgb.model')
+bst1 <- xgb.load('xgb.model')
+print(xgb.attr(bst1, "my_attribute"))
+print(xgb.attributes(bst1))
+
+# deletion:
+xgb.attr(bst1, "my_attribute") <- NULL
+print(xgb.attributes(bst1))
+xgb.attributes(bst1) <- list(a = NULL, b = NULL)
+print(xgb.attributes(bst1))
+
+}

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

@@ -4,12 +4,14 @@
 \alias{xgb.create.features}
 \title{Create new features from a previously learned model}
 \usage{
-xgb.create.features(model, training.data)
+xgb.create.features(model, data, ...)
 }
 \arguments{
 \item{model}{decision tree boosting model learned on the original data}
 
-\item{training.data}{original data (usually provided as a \code{dgCMatrix} matrix)}
+\item{data}{original data (usually provided as a \code{dgCMatrix} matrix)}
+
+\item{...}{currently not used}
 }
 \value{
 \code{dgCMatrix} matrix including both the original data and the new features.
@@ -23,15 +25,15 @@ This is the function inspired from the paragraph 3.1 of the paper:
 \strong{Practical Lessons from Predicting Clicks on Ads at Facebook}
 
 \emph{(Xinran He, Junfeng Pan, Ou Jin, Tianbing Xu, Bo Liu, Tao Xu, Yan, xin Shi, Antoine Atallah, Ralf Herbrich, Stuart Bowers, 
-Joaquin Quiñonero Candela)}
+Joaquin Quinonero Candela)}
  
 International Workshop on Data Mining for Online Advertising (ADKDD) - August 24, 2014
 
-\url{https://research.facebook.com/publications/758569837499391/practical-lessons-from-predicting-clicks-on-ads-at-facebook/}.
+\url{https://research.fb.com/publications/practical-lessons-from-predicting-clicks-on-ads-at-facebook/}.
 
 Extract explaining the method:
 
-"\emph{We found that boosted decision trees are a powerful and very
+"We found that boosted decision trees are a powerful and very
 convenient way to implement non-linear and tuple transformations
 of the kind we just described. We treat each individual
 tree as a categorical feature that takes as value the
@@ -52,7 +54,7 @@ We can understand boosted decision tree
 based transformation as a supervised feature encoding that
 converts a real-valued vector into a compact binary-valued
 vector. A traversal from root node to a leaf node represents
-a rule on certain features.}"
+a rule on certain features."
 }
 \examples{
 data(agaricus.train, package='xgboost')
@@ -60,13 +62,14 @@ data(agaricus.test, package='xgboost')
 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')
+param <- list(max_depth=2, eta=1, silent=1, objective='binary:logistic')
 nround = 4
 
 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$label)
+accuracy.before <- sum((predict(bst, agaricus.test$data) >= 0.5) == agaricus.test$label) /
+                   length(agaricus.test$label)
 
 # Convert previous features to one hot encoding
 new.features.train <- xgb.create.features(model = bst, agaricus.train$data)
@@ -79,10 +82,11 @@ watchlist <- list(train = new.dtrain)
 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)
+accuracy.after <- sum((predict(bst, new.dtest) >= 0.5) == agaricus.test$label) /
+                  length(agaricus.test$label)
 
 # Here the accuracy was already good and is now perfect.
-cat(paste("The accuracy was", accuracy.before, "before adding leaf features and it is now", accuracy.after, "!\\n"))
+cat(paste("The accuracy was", accuracy.before, "before adding leaf features and it is now",
+          accuracy.after, "!\\n"))
 
 }
-

R-package/man/xgb.cv.Rd

@@ -6,8 +6,9 @@
 \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 = T,
-  print.every.n = 1L, early.stop.round = NULL, maximize = 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:
@@ -18,29 +19,31 @@ xgb.cv(params = list(), data, nrounds, nfold, label = NULL, missing = NA,
     \item \code{binary:logistic} logistic regression for classification
   }
   \item \code{eta} step size of each boosting step
-  \item \code{max.depth} maximum depth of the tree
+  \item \code{max_depth} maximum depth of the tree
   \item \code{nthread} number of thread used in training, if not set, all threads are used
 }
 
-  See \link{xgb.train} for further details.
+  See \code{\link{xgb.train}} for further details.
   See also demo/ for walkthrough example in R.}
 
-\item{data}{takes an \code{xgb.DMatrix} or \code{Matrix} as the input.}
+\item{data}{takes an \code{xgb.DMatrix}, \code{matrix}, or \code{dgCMatrix} as the input.}
 
 \item{nrounds}{the max number of iterations}
 
 \item{nfold}{the original dataset is randomly partitioned into \code{nfold} equal size subsamples.}
 
-\item{label}{option field, when data is \code{Matrix}}
+\item{label}{vector of response values. Should be provided only when data is an R-matrix.}
 
-\item{missing}{Missing is only used when input is dense matrix, pick a float
-value that represents missing value. Sometime a data use 0 or other extreme value to represents missing values.}
+\item{missing}{is only used when input is a dense matrix. By default is set to NA, which means 
+that NA values should be considered as 'missing' by the algorithm. 
+Sometimes, 0 or other extreme value might be used to represent missing values.}
 
-\item{prediction}{A logical value indicating whether to return the prediction vector.}
+\item{prediction}{A logical value indicating whether to return the test fold predictions 
+from each CV model. This parameter engages the \code{\link{cb.cv.predict}} callback.}
 
-\item{showsd}{\code{boolean}, whether show standard deviation of cross validation}
+\item{showsd}{\code{boolean}, whether to show standard deviation of cross validation}
 
-\item{metrics, }{list of evaluation metrics to be used in corss validation,
+\item{metrics, }{list of evaluation metrics to be used in cross validation,
   when it is not specified, the evaluation metric is chosen according to objective function.
   Possible options are:
 \itemize{
@@ -48,6 +51,7 @@ value that represents missing value. Sometime a data use 0 or other extreme valu
   \item \code{rmse} Rooted mean square error
   \item \code{logloss} negative log-likelihood function
   \item \code{auc} Area under curve
+  \item \code{aucpr} Area under PR curve
   \item \code{merror} Exact matching error, used to evaluate multi-class classification
 }}
 
@@ -58,35 +62,65 @@ gradient with given prediction and dtrain.}
 \code{list(metric='metric-name', value='metric-value')} with given 
 prediction and dtrain.}
 
-\item{stratified}{\code{boolean} whether sampling of folds should be stratified by the values of labels in \code{data}}
+\item{stratified}{a \code{boolean} indicating whether sampling of folds should be stratified 
+by the values of outcome labels.}
 
-\item{folds}{\code{list} provides a possibility of using a list of pre-defined CV folds (each element must be a vector of fold's indices).
-If folds are supplied, the nfold and stratified parameters would be ignored.}
+\item{folds}{\code{list} provides a possibility to use a list of pre-defined CV folds
+(each element must be a vector of test fold's indices). When folds are supplied, 
+the \code{nfold} and \code{stratified} parameters are ignored.}
 
 \item{verbose}{\code{boolean}, print the statistics during the process}
 
-\item{print.every.n}{Print every N progress messages when \code{verbose>0}. Default is 1 which means all messages are printed.}
+\item{print_every_n}{Print each n-th iteration evaluation messages when \code{verbose>0}.
+Default is 1 which means all messages are printed. This parameter is passed to the 
+\code{\link{cb.print.evaluation}} callback.}
 
-\item{early.stop.round}{If \code{NULL}, the early stopping function is not triggered. 
+\item{early_stopping_rounds}{If \code{NULL}, the early stopping function is not triggered. 
 If set to an integer \code{k}, training with a validation set will stop if the performance 
-keeps getting worse consecutively for \code{k} rounds.}
+doesn't improve for \code{k} rounds.
+Setting this parameter engages the \code{\link{cb.early.stop}} callback.}
 
-\item{maximize}{If \code{feval} and \code{early.stop.round} are set, then \code{maximize} must be set as well.
-\code{maximize=TRUE} means the larger the evaluation score the better.}
+\item{maximize}{If \code{feval} and \code{early_stopping_rounds} are set,
+then this parameter must be set as well.
+When it is \code{TRUE}, it means the larger the evaluation score the better.
+This parameter is passed to the \code{\link{cb.early.stop}} callback.}
+
+\item{callbacks}{a list of callback functions to perform various task during boosting.
+See \code{\link{callbacks}}. Some of the callbacks are automatically created depending on the 
+parameters' values. User can provide either existing or their own callback methods in order 
+to customize the training process.}
 
 \item{...}{other parameters to pass to \code{params}.}
 }
 \value{
-If \code{prediction = TRUE}, a list with the following elements is returned:
+An object of class \code{xgb.cv.synchronous} with the following elements:
 \itemize{
-  \item \code{dt} a \code{data.table} with each mean and standard deviation stat for training set and test set
-  \item \code{pred} an array or matrix (for multiclass classification) with predictions for each CV-fold for the model having been trained on the data in all other folds.
+  \item \code{call} a function call.
+  \item \code{params} parameters that were passed to the xgboost library. Note that it does not 
+        capture parameters changed by the \code{\link{cb.reset.parameters}} callback.
+  \item \code{callbacks} callback functions that were either automatically assigned or 
+        explicitly passed.
+  \item \code{evaluation_log} evaluation history storead as a \code{data.table} with the
+        first column corresponding to iteration number and the rest corresponding to the 
+        CV-based evaluation means and standard deviations for the training and test CV-sets.
+        It is created by the \code{\link{cb.evaluation.log}} callback.
+  \item \code{niter} number of boosting iterations.
+  \item \code{nfeatures} number of features in training data.
+  \item \code{folds} the list of CV folds' indices - either those passed through the \code{folds} 
+        parameter or randomly generated.
+  \item \code{best_iteration} iteration number with the best evaluation metric value
+        (only available with early stopping).
+  \item \code{best_ntreelimit} the \code{ntreelimit} value corresponding to the best iteration, 
+        which could further be used in \code{predict} method
+        (only available with early stopping).
+  \item \code{pred} CV prediction values available when \code{prediction} is set. 
+        It is either vector or matrix (see \code{\link{cb.cv.predict}}).
+  \item \code{models} a liost of the CV folds' models. It is only available with the explicit 
+        setting of the \code{cb.cv.predict(save_models = TRUE)} callback.
 }
-
-If \code{prediction = FALSE}, just a \code{data.table} with each mean and standard deviation stat for training set and test set is returned.
 }
 \description{
-The cross valudation function of xgboost
+The cross validation function of xgboost
 }
 \details{
 The original sample is randomly partitioned into \code{nfold} equal size subsamples. 
@@ -102,8 +136,9 @@ Adapted from \url{http://en.wikipedia.org/wiki/Cross-validation_\%28statistics\%
 \examples{
 data(agaricus.train, package='xgboost')
 dtrain <- xgb.DMatrix(agaricus.train$data, label = agaricus.train$label)
-history <- xgb.cv(data = dtrain, nround=3, nthread = 2, nfold = 5, metrics=list("rmse","auc"),
-                  max.depth =3, eta = 1, objective = "binary:logistic")
-print(history)
-}
+cv <- xgb.cv(data = dtrain, nrounds = 3, nthread = 2, nfold = 5, metrics = list("rmse","auc"),
+                  max_depth = 3, eta = 1, objective = "binary:logistic")
+print(cv)
+print(cv, verbose=TRUE)
 
+}

R-package/man/xgb.dump.Rd

@@ -2,44 +2,55 @@
 % Please edit documentation in R/xgb.dump.R
 \name{xgb.dump}
 \alias{xgb.dump}
-\title{Save xgboost model to text file}
+\title{Dump an xgboost model in text format.}
 \usage{
-xgb.dump(model = NULL, fname = NULL, fmap = "", with.stats = FALSE)
+xgb.dump(model, fname = NULL, fmap = "", with_stats = FALSE,
+  dump_format = c("text", "json"), ...)
 }
 \arguments{
 \item{model}{the model object.}
 
-\item{fname}{the name of the text file where to save the model text dump. If not provided or set to \code{NULL} the function will return the model as a \code{character} vector.}
+\item{fname}{the name of the text file where to save the model text dump. 
+If not provided or set to \code{NULL}, the model is returned as a \code{character} vector.}
 
-\item{fmap}{feature map file representing the type of feature. 
+\item{fmap}{feature map file representing feature types.
 Detailed description could be found at 
 \url{https://github.com/dmlc/xgboost/wiki/Binary-Classification#dump-model}.
 See demo/ for walkthrough example in R, and
 \url{https://github.com/dmlc/xgboost/blob/master/demo/data/featmap.txt} 
 for example Format.}
 
-\item{with.stats}{whether dump statistics of splits 
-When this option is on, the model dump comes with two additional statistics:
+\item{with_stats}{whether to dump some additional statistics about the splits.
+When this option is on, the model dump contains two additional values:
 gain is the approximate loss function gain we get in each split;
 cover is the sum of second order gradient in each node.}
+
+\item{dump_format}{either 'text' or 'json' format could be specified.}
+
+\item{...}{currently not used}
 }
 \value{
-if fname is not provided or set to \code{NULL} the function will return the model as a \code{character} vector. Otherwise it will return \code{TRUE}.
+If fname is not provided or set to \code{NULL} the function will return the model
+as a \code{character} vector. Otherwise it will return \code{TRUE}.
 }
 \description{
-Save a xgboost model to text file. Could be parsed later.
+Dump an xgboost model in text format.
 }
 \examples{
 data(agaricus.train, package='xgboost')
 data(agaricus.test, package='xgboost')
 train <- agaricus.train
 test <- agaricus.test
-bst <- xgboost(data = train$data, label = train$label, max.depth = 2, 
-               eta = 1, nthread = 2, nround = 2,objective = "binary:logistic")
+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'
-xgb.dump(bst, 'xgb.model.dump', 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))
-}
+print(xgb.dump(bst, with_stats = TRUE))
 
+# print in JSON format:
+cat(xgb.dump(bst, with_stats = TRUE, dump_format='json'))
+
+}

R-package/man/xgb.gblinear.history.Rd

@@ -0,0 +1,29 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/callbacks.R
+\name{xgb.gblinear.history}
+\alias{xgb.gblinear.history}
+\title{Extract gblinear coefficients history.}
+\usage{
+xgb.gblinear.history(model, class_index = NULL)
+}
+\arguments{
+\item{model}{either an \code{xgb.Booster} or a result of \code{xgb.cv()}, trained
+using the \code{cb.gblinear.history()} callback.}
+
+\item{class_index}{zero-based class index to extract the coefficients for only that
+specific class in a multinomial multiclass model. When it is NULL, all the
+coeffients are returned. Has no effect in non-multiclass models.}
+}
+\value{
+For an \code{xgb.train} result, a matrix (either dense or sparse) with the columns
+corresponding to iteration's coefficients (in the order as \code{xgb.dump()} would
+return) and the rows corresponding to boosting iterations.
+
+For an \code{xgb.cv} result, a list of such matrices is returned with the elements
+corresponding to CV folds.
+}
+\description{
+A helper function to extract the matrix of linear coefficients' history
+from a gblinear model created while using the \code{cb.gblinear.history()}
+callback.
+}

R-package/man/xgb.importance.Rd

@@ -2,64 +2,94 @@
 % Please edit documentation in R/xgb.importance.R
 \name{xgb.importance}
 \alias{xgb.importance}
-\title{Show importance of features in a model}
+\title{Importance of features in a model.}
 \usage{
-xgb.importance(feature_names = NULL, model = NULL, data = NULL,
-  label = NULL, target = function(x) ((x + label) == 2))
+xgb.importance(feature_names = NULL, model = NULL, trees = NULL,
+  data = NULL, label = NULL, target = NULL)
 }
 \arguments{
-\item{feature_names}{names of each feature as a \code{character} vector. Can be extracted from a sparse matrix (see example). If model dump already contains feature names, this argument should be \code{NULL}.}
+\item{feature_names}{character vector of feature names. If the model already
+contains feature names, those would be used when \code{feature_names=NULL} (default value).
+Non-null \code{feature_names} could be provided to override those in the model.}
 
-\item{model}{generated by the \code{xgb.train} function.}
+\item{model}{object of class \code{xgb.Booster}.}
 
-\item{data}{the dataset used for the training step. Will be used with \code{label} parameter for co-occurence computation. More information in \code{Detail} part. This parameter is optional.}
+\item{trees}{(only for the gbtree booster) an integer vector of tree indices that should be included
+into the importance calculation. If set to \code{NULL}, all trees of the model are parsed.
+It could be useful, e.g., in multiclass classification to get feature importances 
+for each class separately. IMPORTANT: the tree index in xgboost models
+is zero-based (e.g., use \code{trees = 0:4} for first 5 trees).}
 
-\item{label}{the label vetor used for the training step. Will be used with \code{data} parameter for co-occurence computation. More information in \code{Detail} part. This parameter is optional.}
+\item{data}{deprecated.}
 
-\item{target}{a function which returns \code{TRUE} or \code{1} when an observation should be count as a co-occurence and \code{FALSE} or \code{0} otherwise. Default function is provided for computing co-occurences in a binary classification. The \code{target} function should have only one parameter. This parameter will be used to provide each important feature vector after having applied the split condition, therefore these vector will be only made of 0 and 1 only, whatever was the information before. More information in \code{Detail} part. This parameter is optional.}
+\item{label}{deprecated.}
+
+\item{target}{deprecated.}
 }
 \value{
-A \code{data.table} of the features used in the model with their average gain (and their weight for boosted tree model) in the model.
+For a tree model, a \code{data.table} with the following columns:
+\itemize{
+  \item \code{Features} names of the features used in the model;
+  \item \code{Gain} represents fractional contribution of each feature to the model based on
+       the total gain of this feature's splits. Higher percentage means a more important 
+       predictive feature.
+  \item \code{Cover} metric of the number of observation related to this feature;
+  \item \code{Frequency} percentage representing the relative number of times
+       a feature have been used in trees.
+}
+
+A linear model's importance \code{data.table} has the following columns:
+\itemize{
+  \item \code{Features} names of the features used in the model;
+  \item \code{Weight} the linear coefficient of this feature;
+  \item \code{Class} (only for multiclass models) class label.
+}
+
+If \code{feature_names} is not provided and \code{model} doesn't have \code{feature_names}, 
+index of the features will be used instead. Because the index is extracted from the model dump
+(based on C++ code), it starts at 0 (as in C/C++ or Python) instead of 1 (usual in R).
 }
 \description{
-Create a \code{data.table} of the most important features of a model.
+Creates a \code{data.table} of feature importances in a model.
 }
 \details{
-This function is for both linear and tree models.
+This function works for both linear and tree models.
 
-\code{data.table} is returned by the function. 
-The columns are :
-\itemize{
-  \item \code{Features} name of the features as provided in \code{feature_names} or already present in the model dump;
-  \item \code{Gain} contribution of each feature to the model. For boosted tree model, each gain of each feature of each tree is taken into account, then average per feature to give a vision of the entire model. Highest percentage means important feature to predict the \code{label} used for the training (only available for tree models);
-  \item \code{Cover} metric of the number of observation related to this feature (only available for tree models);
-  \item \code{Weight} percentage representing the relative number of times a feature have been taken into trees.
-}
-
-If you don't provide \code{feature_names}, index of the features will be used instead.
-
-Because the index is extracted from the model dump (made on the C++ side), it starts at 0 (usual in C++) instead of 1 (usual in R).
-
-Co-occurence count
-------------------
-
-The gain gives you indication about the information of how a feature is important in making a branch of a decision tree more pure. However, with this information only, you can't know if this feature has to be present or not to get a specific classification. In the example code, you may wonder if odor=none should be \code{TRUE} to not eat a mushroom.
-
-Co-occurence computation is here to help in understanding this relation between a predictor and a specific class. It will count how many observations are returned as \code{TRUE} by the \code{target} function (see parameters). When you execute the example below, there are 92 times only over the 3140 observations of the train dataset where a mushroom have no odor and can be eaten safely.
-
-If you need to remember one thing only: until you want to leave us early, don't eat a mushroom which has no odor :-)
+For linear models, the importance is the absolute magnitude of linear coefficients. 
+For that reason, in order to obtain a meaningful ranking by importance for a linear model, 
+the features need to be on the same scale (which you also would want to do when using either 
+L1 or L2 regularization).
 }
 \examples{
+
+# binomial classification using gbtree:
 data(agaricus.train, package='xgboost')
+bst <- xgboost(data = agaricus.train$data, label = agaricus.train$label, max_depth = 2, 
+               eta = 1, nthread = 2, nrounds = 2, objective = "binary:logistic")
+xgb.importance(model = bst)
 
-bst <- xgboost(data = agaricus.train$data, label = agaricus.train$label, max.depth = 2, 
-               eta = 1, nthread = 2, nround = 2,objective = "binary:logistic")
+# binomial classification using gblinear:
+bst <- xgboost(data = agaricus.train$data, label = agaricus.train$label, booster = "gblinear", 
+               eta = 0.3, nthread = 1, nrounds = 20, objective = "binary:logistic")
+xgb.importance(model = bst)
 
-# agaricus.train$data@Dimnames[[2]] represents the column names of the sparse matrix.
-xgb.importance(agaricus.train$data@Dimnames[[2]], model = bst)
+# multiclass classification using gbtree:
+nclass <- 3
+nrounds <- 10
+mbst <- xgboost(data = as.matrix(iris[, -5]), label = as.numeric(iris$Species) - 1,
+               max_depth = 3, eta = 0.2, nthread = 2, nrounds = nrounds,
+               objective = "multi:softprob", num_class = nclass)
+# all classes clumped together:
+xgb.importance(model = mbst)
+# inspect importances separately for each class:
+xgb.importance(model = mbst, trees = seq(from=0, by=nclass, length.out=nrounds))
+xgb.importance(model = mbst, trees = seq(from=1, by=nclass, length.out=nrounds))
+xgb.importance(model = mbst, trees = seq(from=2, by=nclass, length.out=nrounds))
 
-# Same thing with co-occurence computation this time
-xgb.importance(agaricus.train$data@Dimnames[[2]], model = bst, data = agaricus.train$data, label = agaricus.train$label)
+# multiclass classification using gblinear:
+mbst <- xgboost(data = scale(as.matrix(iris[, -5])), label = as.numeric(iris$Species) - 1,
+               booster = "gblinear", eta = 0.2, nthread = 1, nrounds = 15,
+               objective = "multi:softprob", num_class = nclass)
+xgb.importance(model = mbst)
 
 }
-

R-package/man/xgb.load.Rd

@@ -7,20 +7,34 @@
 xgb.load(modelfile)
 }
 \arguments{
-\item{modelfile}{the name of the binary file.}
+\item{modelfile}{the name of the binary input file.}
+}
+\value{
+An object of \code{xgb.Booster} class.
 }
 \description{
-Load xgboost model from the binary model file
+Load xgboost model from the binary model file.
+}
+\details{
+The input file is expected to contain a model saved in an xgboost-internal binary format
+using either \code{\link{xgb.save}} or \code{\link{cb.save.model}} in R, or using some 
+appropriate methods from other xgboost interfaces. E.g., a model trained in Python and 
+saved from there in xgboost format, could be loaded from R.
+
+Note: a model saved as an R-object, has to be loaded using corresponding R-methods,
+not \code{xgb.load}.
 }
 \examples{
 data(agaricus.train, package='xgboost')
 data(agaricus.test, package='xgboost')
 train <- agaricus.train
 test <- agaricus.test
-bst <- xgboost(data = train$data, label = train$label, max.depth = 2, 
-               eta = 1, nthread = 2, nround = 2,objective = "binary:logistic")
+bst <- xgboost(data = train$data, label = train$label, max_depth = 2, 
+               eta = 1, nthread = 2, nrounds = 2,objective = "binary:logistic")
 xgb.save(bst, 'xgb.model')
 bst <- xgb.load('xgb.model')
 pred <- predict(bst, test$data)
 }
-
+\seealso{
+\code{\link{xgb.save}}, \code{\link{xgb.Booster.complete}}.
+}

R-package/man/xgb.model.dt.tree.Rd

@@ -2,54 +2,76 @@
 % Please edit documentation in R/xgb.model.dt.tree.R
 \name{xgb.model.dt.tree}
 \alias{xgb.model.dt.tree}
-\title{Parse boosted tree model text dump}
+\title{Parse a boosted tree model text dump}
 \usage{
 xgb.model.dt.tree(feature_names = NULL, model = NULL, text = NULL,
-  n_first_tree = NULL)
+  trees = NULL, use_int_id = FALSE, ...)
 }
 \arguments{
-\item{feature_names}{names of each feature as a character vector. Can be extracted from a sparse matrix (see example). If the model already contains feature names, this argument should be \code{NULL} (default value).}
+\item{feature_names}{character vector of feature names. If the model already
+contains feature names, those would be used when \code{feature_names=NULL} (default value).
+Non-null \code{feature_names} could be provided to override those in the model.}
 
-\item{model}{object created by the \code{xgb.train} function.}
+\item{model}{object of class \code{xgb.Booster}}
 
-\item{text}{\code{character} vector generated by the \code{xgb.dump} function. Model dump must include the gain per feature and per tree (parameter \code{with.stats = TRUE} in function \code{xgb.dump}).}
+\item{text}{\code{character} vector previously generated by the \code{xgb.dump} 
+function  (where parameter \code{with_stats = TRUE} should have been set).
+\code{text} takes precedence over \code{model}.}
 
-\item{n_first_tree}{limit the plot to the \code{n} first trees. If set to \code{NULL}, all trees of the model are plotted. Performance can be low depending of the size of the model.}
+\item{trees}{an integer vector of tree indices that should be parsed.
+If set to \code{NULL}, all trees of the model are parsed.
+It could be useful, e.g., in multiclass classification to get only
+the trees of one certain class. IMPORTANT: the tree index in xgboost models
+is zero-based (e.g., use \code{trees = 0:4} for first 5 trees).}
+
+\item{use_int_id}{a logical flag indicating whether nodes in columns "Yes", "No", "Missing" should be
+represented as integers (when FALSE) or as "Tree-Node" character strings (when FALSE).}
+
+\item{...}{currently not used.}
 }
 \value{
-A \code{data.table} of the features used in the model with their gain, cover and few other information.
-}
-\description{
-Parse a boosted tree model text dump and return a \code{data.table}.
-}
-\details{
-General function to convert a text dump of tree model to a \code{data.table}. 
-
-The purpose is to help user to explore the model and get a better understanding of it.
+A \code{data.table} with detailed information about model trees' nodes.
 
 The columns of the \code{data.table} are:
 
 \itemize{
-\item \code{ID}: unique identifier of a node ;
- \item \code{Feature}: feature used in the tree to operate a split. When Leaf is indicated, it is the end of a branch ;
- \item \code{Split}: value of the chosen feature where is operated the split ;
- \item \code{Yes}: ID of the feature for the next node in the branch when the split condition is met ;
- \item \code{No}: ID of the feature for the next node in the branch when the split condition is not met ;
- \item \code{Missing}: ID of the feature for the next node in the branch for observation where the feature used for the split are not provided ;
- \item \code{Quality}: it's the gain related to the split in this specific node ;
- \item \code{Cover}: metric to measure the number of observation affected by the split ;
- \item \code{Tree}: ID of the tree. It is included in the main ID ;
- \item \code{Yes.Feature}, \code{No.Feature}, \code{Yes.Cover}, \code{No.Cover}, \code{Yes.Quality} and \code{No.Quality}: data related to the pointer in \code{Yes} or \code{No} column ;
+ \item \code{Tree}: integer ID of a tree in a model (zero-based index)
+ \item \code{Node}: integer ID of a node in a tree (zero-based index)
+ \item \code{ID}: character identifier of a node in a model (only when \code{use_int_id=FALSE})
+ \item \code{Feature}: for a branch node, it's a feature id or name (when available);
+             for a leaf note, it simply labels it as \code{'Leaf'}
+ \item \code{Split}: location of the split for a branch node (split condition is always "less than")
+ \item \code{Yes}: ID of the next node when the split condition is met
+ \item \code{No}: ID of the next node when the split condition is not met
+ \item \code{Missing}: ID of the next node when branch value is missing
+ \item \code{Quality}: either the split gain (change in loss) or the leaf value
+ \item \code{Cover}: metric related to the number of observation either seen by a split
+                     or collected by a leaf during training.
+} 
+
+When \code{use_int_id=FALSE}, columns "Yes", "No", and "Missing" point to model-wide node identifiers
+in the "ID" column. When \code{use_int_id=TRUE}, those columns point to node identifiers from 
+the corresponding trees in the "Node" column.
 }
+\description{
+Parse a boosted tree model text dump into a \code{data.table} structure.
 }
 \examples{
+# Basic use:
+
 data(agaricus.train, package='xgboost')
 
-bst <- xgboost(data = agaricus.train$data, label = agaricus.train$label, max.depth = 2, 
-               eta = 1, nthread = 2, nround = 2,objective = "binary:logistic")
+bst <- xgboost(data = agaricus.train$data, label = agaricus.train$label, max_depth = 2, 
+               eta = 1, nthread = 2, nrounds = 2,objective = "binary:logistic")
 
-# agaricus.train$data@Dimnames[[2]] represents the column names of the sparse matrix.
-xgb.model.dt.tree(feature_names = agaricus.train$data@Dimnames[[2]], model = bst)
+(dt <- xgb.model.dt.tree(colnames(agaricus.train$data), bst))
 
+# This bst model already has feature_names stored with it, so those would be used when 
+# feature_names is not set:
+(dt <- xgb.model.dt.tree(model = bst))
+
+# How to match feature names of splits that are following a current 'Yes' branch:
+
+merge(dt, dt[, .(ID, Y.Feature=Feature)], by.x='Yes', by.y='ID', all.x=TRUE)[order(Tree,Node)]
+ 
 }
-

R-package/man/xgb.parameters.Rd

@@ -0,0 +1,31 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/xgb.Booster.R
+\name{xgb.parameters<-}
+\alias{xgb.parameters<-}
+\title{Accessors for model parameters.}
+\usage{
+xgb.parameters(object) <- value
+}
+\arguments{
+\item{object}{Object of class \code{xgb.Booster} or \code{xgb.Booster.handle}.}
+
+\item{value}{a list (or an object coercible to a list) with the names of parameters to set
+and the elements corresponding to parameter values.}
+}
+\description{
+Only the setter for xgboost parameters is currently implemented.
+}
+\details{
+Note that the setter would usually work more efficiently for \code{xgb.Booster.handle}
+than for \code{xgb.Booster}, since only just a handle would need to be copied.
+}
+\examples{
+data(agaricus.train, package='xgboost')
+train <- agaricus.train
+
+bst <- xgboost(data = train$data, label = train$label, max_depth = 2,
+               eta = 1, nthread = 2, nrounds = 2, objective = "binary:logistic")
+
+xgb.parameters(bst) <- list(eta = 0.1)
+
+}

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

@@ -1,46 +1,74 @@
 % Generated by roxygen2: do not edit by hand
-% Please edit documentation in R/xgb.plot.deepness.R
-\name{xgb.plot.deepness}
+% Please edit documentation in R/xgb.ggplot.R, R/xgb.plot.deepness.R
+\name{xgb.ggplot.deepness}
+\alias{xgb.ggplot.deepness}
 \alias{xgb.plot.deepness}
 \title{Plot model trees deepness}
 \usage{
-xgb.plot.deepness(model = NULL)
+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, ...)
 }
 \arguments{
-\item{model}{dump generated by the \code{xgb.train} function.}
+\item{model}{either an \code{xgb.Booster} model generated by the \code{xgb.train} function
+or a data.table result of the \code{xgb.model.dt.tree} function.}
+
+\item{which}{which distribution to plot (see details).}
+
+\item{plot}{(base R barplot) whether a barplot should be produced. 
+If FALSE, only a data.table is returned.}
+
+\item{...}{other parameters passed to \code{barplot} or \code{plot}.}
 }
 \value{
-Two graphs showing the distribution of the model deepness.
+Other than producing plots (when \code{plot=TRUE}), the \code{xgb.plot.deepness} function
+silently returns a processed data.table where each row corresponds to a terminal leaf in a tree model,
+and contains information about leaf's depth, cover, and weight (which is used in calculating predictions).
+
+The \code{xgb.ggplot.deepness} silently returns either a list of two ggplot graphs when \code{which="2x1"}
+or a single ggplot graph for the other \code{which} options.
 }
 \description{
-Generate a graph to plot the distribution of deepness among trees.
+Visualizes distributions related to depth of tree leafs.
+\code{xgb.plot.deepness} uses base R graphics, while \code{xgb.ggplot.deepness} uses the ggplot backend.
 }
 \details{
-Display both the number of \code{leaf} and the distribution of \code{weighted observations}
-by tree deepness level.
-
-The purpose of this function is to help the user to find the best trade-off to set
-the \code{max.depth} and \code{min_child_weight} parameters according to the bias / variance trade-off.
-
-See \link{xgb.train} for more information about these parameters.
-
-The graph is made of two parts:
-
+When \code{which="2x1"}, two distributions with respect to the leaf depth
+are plotted on top of each other:
 \itemize{
- \item Count: number of leaf per level of deepness;
- \item Weighted cover: noramlized weighted cover per leaf (weighted number of instances).
+ \item the distribution of the number of leafs in a tree model at a certain depth;
+ \item the distribution of average weighted number of observations ("cover") 
+       ending up in leafs at certain depth.
 }
+Those could be helpful in determining sensible ranges of the \code{max_depth} 
+and \code{min_child_weight} parameters.
 
-This function is inspired by the blog post \url{http://aysent.github.io/2015/11/08/random-forest-leaf-visualization.html}
+When \code{which="max.depth"} or \code{which="med.depth"}, plots of either maximum or median depth
+per tree with respect to tree number are created. And \code{which="med.weight"} allows to see how
+a tree's median absolute leaf weight changes through the iterations.
+
+This function was inspired by the blog post
+\url{http://aysent.github.io/2015/11/08/random-forest-leaf-visualization.html}.
 }
 \examples{
+
 data(agaricus.train, package='xgboost')
 
-bst <- xgboost(data = agaricus.train$data, label = agaricus.train$label, max.depth = 15,
-                 eta = 1, nthread = 2, nround = 30, objective = "binary:logistic",
-                 min_child_weight = 50)
+# Change max_depth to a higher number to get a more significant result
+bst <- xgboost(data = agaricus.train$data, label = agaricus.train$label, max_depth = 6,
+               eta = 0.1, nthread = 2, nrounds = 50, objective = "binary:logistic",
+               subsample = 0.5, min_child_weight = 2)
 
-xgb.plot.deepness(model = bst)
+xgb.plot.deepness(bst)
+xgb.ggplot.deepness(bst)
+
+xgb.plot.deepness(bst, which='max.depth', pch=16, col=rgb(0,0,1,0.3), cex=2)
+
+xgb.plot.deepness(bst, which='med.weight', pch=16, col=rgb(0,0,1,0.3), cex=2)
 
 }
-
+\seealso{
+\code{\link{xgb.train}}, \code{\link{xgb.model.dt.tree}}.
+}