* [CI] Use Vault repository to re-gain access to devtoolset-4
* Use manylinux2010 tag
* Update Dockerfile.jvm
* Fix rename_whl.py
* Upgrade Pip, to handle manylinux2010 tag
* Update insert_vcomp140.py
* Update test_python.sh
* Set output margin to True for custom objective in Python and R.
* Add a demo for writing multi-class custom objective function.
* Run tests on selected demos.
* Group aware GPU weighted sketching.
* Distribute group weights to each data point.
* Relax the test.
* Validate input meta info.
* Fix metainfo copy ctor.
* Add inplace prediction for dask-cudf.
* Remove Dockerfile.release, since it's not used anywhere
* Use Conda exclusively in CUDF and GPU containers
* Improve cupy memory copying.
* Add skip marks to tests.
* Add mgpu-cudf category on the CI to run all distributed tests.
Co-authored-by: Hyunsu Cho <chohyu01@cs.washington.edu>
* Ensure that configured header (build_config.h) from dmlc-core is picked up by Rabit and XGBoost
* Check which Rabit target is being used
* Use CMake 3.13 in all Jenkins tests
* Upgrade CMake in Travis CI
* Install CMake using Kitware installer
* Remove existing CMake (3.12.4)
* Use devtoolset-6.
* [CI] Use devtoolset-6 because devtoolset-4 is EOL and no longer available
* CUDA 9.0 doesn't work with devtoolset-6; use devtoolset-4 for GPU build only
Co-authored-by: Hyunsu Cho <chohyu01@cs.washington.edu>
* Add bindings for serialization.
* Change `xgb.save.raw' into full serialization instead of simple model.
* Add `xgb.load.raw' for unserialization.
* Run devtools.
* fix type error
* Validate number of features.
* resolve comments
* add feature size for LabelPoint and DataBatch
* pass the feature size to native
* move feature size validating tests into a separate suite
* resolve comments
Co-authored-by: fis <jm.yuan@outlook.com>
* Robust regularization of AFT gradient and hessian
* Fix AFT doc; expose it to tutorial TOC
* Apply robust regularization to uncensored case too
* Revise unit test slightly
* Fix lint
* Update test_survival.py
* Use GradientPairPrecise
* Remove unused variables
* Set default dtor for SimpleDMatrix to initialize default copy ctor, which is
deleted due to unique ptr.
* Remove commented code.
* Remove warning for calling host function (std::max).
* Remove warning for initialization order.
* Remove warning for unused variables.
Normal prediction with DMatrix is now thread safe with locks. Added inplace prediction is lock free thread safe.
When data is on device (cupy, cudf), the returned data is also on device.
* Implementation for numpy, csr, cudf and cupy.
* Implementation for dask.
* Remove sync in simple dmatrix.
* [WIP] Add lower and upper bounds on the label for survival analysis
* Update test MetaInfo.SaveLoadBinary to account for extra two fields
* Don't clear qids_ for version 2 of MetaInfo
* Add SetInfo() and GetInfo() method for lower and upper bounds
* changes to aft
* Add parameter class for AFT; use enum's to represent distribution and event type
* Add AFT metric
* changes to neg grad to grad
* changes to binomial loss
* changes to overflow
* changes to eps
* changes to code refactoring
* changes to code refactoring
* changes to code refactoring
* Re-factor survival analysis
* Remove aft namespace
* Move function bodies out of AFTNormal and AFTLogistic, to reduce clutter
* Move function bodies out of AFTLoss, to reduce clutter
* Use smart pointer to store AFTDistribution and AFTLoss
* Rename AFTNoiseDistribution enum to AFTDistributionType for clarity
The enum class was not a distribution itself but a distribution type
* Add AFTDistribution::Create() method for convenience
* changes to extreme distribution
* changes to extreme distribution
* changes to extreme
* changes to extreme distribution
* changes to left censored
* deleted cout
* changes to x,mu and sd and code refactoring
* changes to print
* changes to hessian formula in censored and uncensored
* changes to variable names and pow
* changes to Logistic Pdf
* changes to parameter
* Expose lower and upper bound labels to R package
* Use example weights; normalize log likelihood metric
* changes to CHECK
* changes to logistic hessian to standard formula
* changes to logistic formula
* Comply with coding style guideline
* Revert back Rabit submodule
* Revert dmlc-core submodule
* Comply with coding style guideline (clang-tidy)
* Fix an error in AFTLoss::Gradient()
* Add missing files to amalgamation
* Address @RAMitchell's comment: minimize future change in MetaInfo interface
* Fix lint
* Fix compilation error on 32-bit target, when size_t == bst_uint
* Allocate sufficient memory to hold extra label info
* Use OpenMP to speed up
* Fix compilation on Windows
* Address reviewer's feedback
* Add unit tests for probability distributions
* Make Metric subclass of Configurable
* Address reviewer's feedback: Configure() AFT metric
* Add a dummy test for AFT metric configuration
* Complete AFT configuration test; remove debugging print
* Rename AFT parameters
* Clarify test comment
* Add a dummy test for AFT loss for uncensored case
* Fix a bug in AFT loss for uncensored labels
* Complete unit test for AFT loss metric
* Simplify unit tests for AFT metric
* Add unit test to verify aggregate output from AFT metric
* Use EXPECT_* instead of ASSERT_*, so that we run all unit tests
* Use aft_loss_param when serializing AFTObj
This is to be consistent with AFT metric
* Add unit tests for AFT Objective
* Fix OpenMP bug; clarify semantics for shared variables used in OpenMP loops
* Add comments
* Remove AFT prefix from probability distribution; put probability distribution in separate source file
* Add comments
* Define kPI and kEulerMascheroni in probability_distribution.h
* Add probability_distribution.cc to amalgamation
* Remove unnecessary diff
* Address reviewer's feedback: define variables where they're used
* Eliminate all INFs and NANs from AFT loss and gradient
* Add demo
* Add tutorial
* Fix lint
* Use 'survival:aft' to be consistent with 'survival:cox'
* Move sample data to demo/data
* Add visual demo with 1D toy data
* Add Python tests
Co-authored-by: Philip Cho <chohyu01@cs.washington.edu>
* Move thread local entry into Learner.
This is an attempt to workaround CUDA context issue in static variable, where
the CUDA context can be released before device vector.
* Add PredictionEntry to thread local entry.
This eliminates one copy of prediction vector.
* Don't define CUDA C API in a namespace.
* - create a gpu metrics (internal) registry
- the objective is to separate the cpu and gpu implementations such that they evolve
indepedently. to that end, this approach will:
- preserve the same metrics configuration (from the end user perspective)
- internally delegate the responsibility to the gpu metrics builder when there is a
valid device present
- decouple the gpu metrics builder from the cpu ones to prevent misuse
- move away from including the cuda file from within the cc file and segregate the code
via ifdef's
* Use pre-rounding based method to obtain reproducible floating point
summation.
* GPU Hist for regression and classification are bit-by-bit reproducible.
* Add doc.
* Switch to thrust reduce for `node_sum_gradient`.
* Add release note for 1.0.0
* Fix a small bug in the Python script that compiles the list of contributors
* Clarify governance of CI infrastructure; now PMC is formally in charge
* Address reviewer comment
* Fix typo
- move segment sorter to common
- this is the first of a handful of pr's that splits the larger pr #5326
- it moves this facility to common (from ranking objective class), so that it can be
used for metric computation
- it also wraps all the bald device pointers into span.
* Remove f-string, since it's not supported by Python 3.5 (#5330)
* Remove f-string, since it's not supported by Python 3.5
* Add Python 3.5 to CI, to ensure compatibility
* Remove duplicated matplotlib
* Show deprecation notice for Python 3.5
* Fix lint
* Fix lint
* Fix a unit test that mistook MINOR ver for PATCH ver
* Enforce only major version in JSON model schema
* Bump version to 1.1.0-SNAPSHOT
* Added a check call macro in jvm package, prevents executing other functions
from jvm when error occurred in XGBoost. For example, when prediction fails jvm
should not try to allocate memory based on the output prediction size.
Move this function into gbtree, and uses only updater for doing so. As now the predictor knows exactly how many trees to predict, there's no need for it to update the prediction cache.
* Move prediction cache into Learner.
* Clean-ups
- Remove duplicated cache in Learner and GBM.
- Remove ad-hoc fix of invalid cache.
- Remove `PredictFromCache` in predictors.
- Remove prediction cache for linear altogether, as it's only moving the
prediction into training process but doesn't provide any actual overall speed
gain.
- The cache is now unique to Learner, which means the ownership is no longer
shared by any other components.
* Changes
- Add version to prediction cache.
- Use weak ptr to check expired DMatrix.
- Pass shared pointer instead of raw pointer.
The setup.py is rewritten. This new script uses only Python code and provide customized
implementation of setuptools commands. This way users can run most of setuptools commands
just like any other Python libraries.
* Remove setup_pip.py
* Remove soft links.
* Define customized commands.
* Remove shell script.
* Remove makefile script.
* Update the doc for building from source.
* Make pip install xgboost*.tar.gz work by fixing build-python.sh
* Simplify install doc
* Add test
* Install Miniconda for Linux target too
* Build XGBoost only once in sdist
* Try importing xgboost after installation
* Don't set PYTHONPATH env var for sdist test
* Turn xgboost::DataType into C++11 enum class
* New binary serialization format for DMatrix::MetaInfo
* Fix clang-tidy
* Fix c++ test
* Implement new format proposal
* Move helper functions to anonymous namespace; remove unneeded field
* Fix lint
* Add shape.
* Keep only roundtrip test.
* Fix test.
* various fixes
* Update data.cc
Co-authored-by: Jiaming Yuan <jm.yuan@outlook.com>
* Simplify Scikit-Learn parameter management.
* Copy base class for removing duplicated parameter signatures.
* Set all parameters to None.
* Handle None in set_param.
* Extract the doc.
Co-authored-by: Jiaming Yuan <jm.yuan@outlook.com>
* Simplify DropTrees calling logic
* Add `training` parameter for prediction method.
* [Breaking]: Add `training` to C API.
* Change for R and Python custom objective.
* Correct comment.
Co-authored-by: Philip Hyunsu Cho <chohyu01@cs.washington.edu>
Co-authored-by: Jiaming Yuan <jm.yuan@outlook.com>
* Fix syncing DMatrix columns.
* notes for tree method.
* Enable feature validation for all interfaces except for jvm.
* Better tests for boosting from predictions.
* Disable validation on JVM.
* Disable parameter validation for now.
Scikit-Learn passes all parameters down to XGBoost, whether they are used or
not.
* Add option `validate_parameters`.
* - implementation of map ranking algorithm
- also effected necessary suggestions mentioned in the earlier ranking pr's
- made some performance improvements to the ndcg algo as well
* Add OpenMP as CMake target
* Require CMake 3.12, to allow linking OpenMP target to objxgboost
* Specify OpenMP compiler flag for CUDA host compiler
* Require CMake 3.16+ if the OS is Mac OSX
* Use AppleClang in Mac tests.
* Update dmlc-core
* Remove `learning_rates`.
It's been deprecated since we have callback.
* Set `before_iteration` of `reset_learning_rate` to False to preserve
the initial learning rate, and comply to the term "reset".
Closes#4709.
* Tests for various `tree_method`.
* Pass pointer to model parameters.
This PR de-duplicates most of the model parameters except the one in
`tree_model.h`. One difficulty is `base_score` is a model property but can be
changed at runtime by objective function. Hence when performing model IO, we
need to save the one provided by users, instead of the one transformed by
objective. Here we created an immutable version of `LearnerModelParam` that
represents the value of model parameter after configuration.
This PR fixes tree weights in dart being ignored when computing contributions.
* Fix ellpack page source link.
* Add tree weights to compute contribution.
- Install wget explicitly to match openssl.
- Install CMake explicitly.
- Use newer miniconda link.
- Reenable unittests.
- gcc@9 + xcode@10 for osx due to missing <_stdio.h>. Other versions of gcc should also work. But as homebrew pour gcc@9 after update by default, so I just stick with latest version.
- Disabled one external memory test for OSX. Not sure about the thread implementation in there and fixing external memory is beyond the scope of this PR.
- Use Python3 with conda in jvm package.
* Extract interaction constraints from split evaluator.
The reason for doing so is mostly for model IO, where num_feature and interaction_constraints are copied in split evaluator. Also interaction constraint by itself is a feature selector, acting like column sampler and it's inefficient to bury it deep in the evaluator chain. Lastly removing one another copied parameter is a win.
* Enable inc for approx tree method.
As now the implementation is spited up from evaluator class, it's also enabled for approx method.
* Removing obsoleted code in colmaker.
They are never documented nor actually used in real world. Also there isn't a single test for those code blocks.
* Unifying the types used for row and column.
As the size of input dataset is marching to billion, incorrect use of int is subject to overflow, also singed integer overflow is undefined behaviour. This PR starts the procedure for unifying used index type to unsigned integers. There's optimization that can utilize this undefined behaviour, but after some testings I don't see the optimization is beneficial to XGBoost.
This makes GPU Hist robust in distributed environment as some workers might not
be associated with any data in either training or evaluation.
* Disable rabit mock test for now: See #5012 .
* Disable dask-cudf test at prediction for now: See #5003
* Launch dask job for all workers despite they might not have any data.
* Check 0 rows in elementwise evaluation metrics.
Using AUC and AUC-PR still throws an error. See #4663 for a robust fix.
* Add tests for edge cases.
* Add `LaunchKernel` wrapper handling zero sized grid.
* Move some parts of allreducer into a cu file.
* Don't validate feature names when the booster is empty.
* Sync number of columns in DMatrix.
As num_feature is required to be the same across all workers in data split
mode.
* Filtering in dask interface now by default syncs all booster that's not
empty, instead of using rank 0.
* Fix Jenkins' GPU tests.
* Install dask-cuda from source in Jenkins' test.
Now all tests are actually running.
* Restore GPU Hist tree synchronization test.
* Check UUID of running devices.
The check is only performed on CUDA version >= 10.x, as 9.x doesn't have UUID field.
* Fix CMake policy and project variables.
Use xgboost_SOURCE_DIR uniformly, add policy for CMake >= 3.13.
* Fix copying data to CPU
* Fix race condition in cpu predictor.
* Fix duplicated DMatrix construction.
* Don't download extra nccl in CI script.
* Do not store built artifacts in the Jenkins master
* Add wheel renaming script
* Upload wheels to S3 bucket
* Use env.GIT_COMMIT
* Capture git hash correctly
* Add missing import in Jenkinsfile
* Address reviewer's comments
* Put artifacts for pull requests in separate directory
* No wildcard expansion in Windows CMD
* Use `UpdateAllowUnknown' for non-model related parameter.
Model parameter can not pack an additional boolean value due to binary IO
format. This commit deals only with non-model related parameter configuration.
* Add tidy command line arg for use-dmlc-gtest.
* - pairwise ranking objective implementation on gpu
- there are couple of more algorithms (ndcg and map) for which support will be added
as follow-up pr's
- with no label groups defined, get gradient is 90x faster on gpu (120m instance
mortgage dataset)
- it can perform by an order of magnitude faster with ~ 10 groups (and adequate cores
for the cpu implementation)
* Add JSON config to rank obj.
* Use CMake config file for representing version.
* Generate c and Python version file with CMake.
The generated file is written into source tree. But unless XGBoost upgrades
its version, there will be no actual modification. This retains compatibility
with Makefiles for R.
* Add XGBoost version the DMatrix binaries.
* Simplify prefetch detection in CMakeLists.txt
* Apply Configurable to objective functions.
* Apply Model to Learner and Regtree, gbm.
* Add Load/SaveConfig to objs.
* Refactor obj tests to use smart pointer.
* Dummy methods for Save/Load Model.
* Don't set_params at the end of set_state.
* Also fix another issue found in dask prediction.
* Add note about prediction.
Don't support other prediction modes at the moment.
* Move get transpose into cc.
* Clean up headers in host device vector, remove thrust dependency.
* Move span and host device vector into public.
* Install c++ headers.
* Short notes for c and c++.
Co-Authored-By: Philip Hyunsu Cho <chohyu01@cs.washington.edu>
* Add BigDenseMatrix
* ability to create DMatrix with bigger than Integer.MAX_VALUE size arrays
* uses sun.misc.Unsafe
* make DMatrix test work from a jar as well
* Add public group getter for java and scala
* Remove unnecessary param from javadoc
* Fix typo
* Fix another typo
* Add semicolon
* Fix javadoc return statement
* Fix missing return statement
* Add a unit test
* Restrict access to `cfg_` in gbm.
* Verify having correct updaters.
* Remove `grow_global_histmaker`
This updater is the same as `grow_histmaker`. The former is not in our
document so we just remove it.
* Initial support for cudf integration.
* Add two C APIs for consuming data and metainfo.
* Add CopyFrom for SimpleCSRSource as a generic function to consume the data.
* Add FromDeviceColumnar for consuming device data.
* Add new MetaInfo::SetInfo for consuming label, weight etc.
* Refactor configuration [Part II].
* General changes:
** Remove `Init` methods to avoid ambiguity.
** Remove `Configure(std::map<>)` to avoid redundant copying and prepare for
parameter validation. (`std::vector` is returned from `InitAllowUnknown`).
** Add name to tree updaters for easier debugging.
* Learner changes:
** Make `LearnerImpl` the only source of configuration.
All configurations are stored and carried out by `LearnerImpl::Configure()`.
** Remove booster in C API.
Originally kept for "compatibility reason", but did not state why. So here
we just remove it.
** Add a `metric_names_` field in `LearnerImpl`.
** Remove `LazyInit`. Configuration will always be lazy.
** Run `Configure` before every iteration.
* Predictor changes:
** Allocate both cpu and gpu predictor.
** Remove cpu_predictor from gpu_predictor.
`GBTree` is now used to dispatch the predictor.
** Remove some GPU Predictor tests.
* IO
No IO changes. The binary model format stability is tested by comparing
hashing value of save models between two commits
* bump scala to 2.12 which requires java 8 and also newer flink and akka
* put scala version in artifactId
* fix appveyor
* fix for scaladoc issue that looks like https://github.com/scala/bug/issues/10509
* fix ci_build
* update versions in generate_pom.py
* fix generate_pom.py
* apache does not have a download for spark 2.4.3 distro using scala 2.12 yet, so for now i use a tgz i put on s3
* Upload spark-2.4.3-bin-scala2.12-hadoop2.7.tgz to our own S3
* Update Dockerfile.jvm_cross
* Update Dockerfile.jvm_cross
* Reorganize contributor's doc
* Address comments from @trivialfis
* Address @sriramch's comment: include ABI compatibility guarantee
* Address @rongou's comment
* Postpone ABI compatibility guarantee for now
* provide the readme
* update for format
* reformat
* reformat -2
* update again
* update format
* update w.r.t yinlou's comments
* Add kubernetes tutorial to Table of Contents
* Style edit
* Fix#4630, #4421: Preserve correct ordering between metrics, and always use last metric for early stopping
* Clarify semantics of early stopping in presence of multiple valid sets and metrics
* Add a test
* Fix lint
* _maybe_pandas_xxx should return their arguments unchanged if no pandas installed
* Tests should not assume pandas is installed
* Mark tests which require pandas as such
* Fix external memory for get column batches.
This fixes two bugs:
* Use PushCSC for get column batches.
* Don't remove the created temporary directory before finishing test.
* Check all pages.
* Add to documentation how to build native unit tests
* Add instructions to run Python tests and to use Docker container [skip ci]
* Fix link to pytest chapter
* Add link to Google Test [skip ci]
* Set PYTHONPATH [skip ci]
* Revise test_python.sh for running tests locally
* Update test_python.sh
* Place Docker recommendation notice in a prominent place [skip ci]
* Initial performance optimizations for xgboost
* remove includes
* revert float->double
* fix for CI
* fix for CI
* fix for CI
* fix for CI
* fix for CI
* fix for CI
* fix for CI
* fix for CI
* fix for CI
* fix for CI
* Check existence of _mm_prefetch and __builtin_prefetch
* Fix lint
* optimizations for CPU
* appling comments in review
* add some comments, code refactoring
* fixing issues in CI
* adding runtime checks
* remove 1 extra check
* remove extra checks in BuildHist
* remove checks
* add debug info
* added debug info
* revert changes
* added comments
* Apply suggestions from code review
Co-Authored-By: Philip Hyunsu Cho <chohyu01@cs.washington.edu>
* apply review comments
* Remove unused function CreateNewNodes()
* Add descriptive comment on node_idx variable in QuantileHistMaker::Builder::BuildHistsBatch()
* Implement tree model dump with a code generator.
* Split up generators.
* Implement graphviz generator.
* Use pattern matching.
* [Breaking] Return a Source in `to_graphviz` instead of Digraph in Python package.
Co-Authored-By: Philip Hyunsu Cho <chohyu01@cs.washington.edu>
* - do not create device vectors for the entire sparse page while computing histograms...
- while creating the compressed histogram indices, the row vector is created for the entire
sparse page batch. this is needless as we only process chunks at a time based on a slice
of the total gpu memory
- this pr will allocate only as much as required to store the ppropriate row indices and the entries
* - do not dereference row_ptrs once the device_vector has been created to elide host copies of those counts
- instead, grab the entry counts directly from the sparsepage
* - set the appropriate device before freeing device memory...
- pr #4532 added a global memory tracker/logger to keep track of number of (de)allocations
and peak memory usage on a per device basis.
- this pr adds the appropriate check to make sure that the (de)allocation counts and memory usages
makes sense for the device. since verbosity is typically increased on debug/non-retail builds.
* - pre-create cub allocators and reuse them
- create them once and not resize them dynamically. we need to ensure that these allocators
are created and destroyed exactly once so that the appropriate device id's are set
This is part 1 of refactoring configuration.
* Move tree heuristic configurations.
* Split up declarations and definitions for GBTree.
* Implement UseGPU in gbm.
* - training with external memory - part 2 of 2
- when external memory support is enabled, building of histogram indices are
done incrementally for every sparse page
- the entire set of input data is divided across multiple gpu's and the relative
row positions within each device is tracked when building the compressed histogram buffer
- this was tested using a mortgage dataset containing ~ 670m rows before 4xt4's could be
saturated
* Fix C++11 config parser
* Use raw strings to improve readability of regex
* Fix compilation for GCC 5.x
Co-authored-by: Jiaming Yuan <jm.yuan@outlook.com>
* simplify the config.h file
* revise config.h
* revised config.h
* revise format
* revise format issues
* revise whitespace issues
* revise whitespace namespace format issues
* revise namespace format issues
* format issues
* format issues
* format issues
* format issues
* Revert submodule changes
* minor change
* Update src/common/config.h
Co-Authored-By: Philip Hyunsu Cho <chohyu01@cs.washington.edu>
* address format issue from trivialfis
* Use correct cub submodule
* - training with external memory part 1 of 2
- this pr focuses on computing the quantiles using multiple gpus on a
dataset that uses the external cache capabilities
- there will a follow-up pr soon after this that will support creation
of histogram indices on large dataset as well
- both of these changes are required to support training with external memory
- the sparse pages in dmatrix are taken in batches and the the cut matrices
are incrementally built
- also snuck in some (perf) changes related to sketches aggregation amongst multiple
features across multiple sparse page batches. instead of aggregating the summary
inside each device and merged later, it is aggregated in-place when the device
is working on different rows but the same feature
* Only define `gpu_id` and `n_gpus` in `LearnerTrainParam`
* Pass LearnerTrainParam through XGBoost vid factory method.
* Disable all GPU usage when GPU related parameters are not specified (fixes XGBoost choosing GPU over aggressively).
* Test learner train param io.
* Fix gpu pickling.
* - fix issues with training with external memory on cpu
- use the batch size to determine the correct number of rows in a batch
- use the right number of threads in omp parallalization if the batch size
is less than the default omp max threads (applicable for the last batch)
* - handle scenarios where last batch size is < available number of threads
- augment tests such that we can test all scenarios (batch size <, >, = number of threads)
* adding support for matrix slicing with query ID for cross-validation
* hail mary test of unrar installation for windows tests
* trying to modify tests to run in Github CI
* Remove dependency on wget and unrar
* Save error log from R test
* Relax assertion in test_training
* Use int instead of bool in C function interface
* Revise R interface
* Add XGDMatrixSliceDMatrixEx and keep old XGDMatrixSliceDMatrix for API compatibility
* Add CMake option to use bundled gtest from dmlc-core, so that it is easy to build XGBoost with gtest on Windows
* Consistently apply OpenMP flag to all targets. Force enable OpenMP when USE_CUDA is turned on.
* Insert vcomp140.dll into Windows wheels
* Add C++ and Python tests for CPU and GPU targets (CUDA 9.0, 10.0, 10.1)
* Prevent spurious msbuild failure
* Add GPU tests
* Upgrade dmlc-core
* Fix#4462: Use /MT flag consistently for MSVC target
* First attempt at Windows CI
* Distinguish stages in Linux and Windows pipelines
* Try running CMake in Windows pipeline
* Add build step
* Automatically set maximize_evaluation_metrics if not explicitly given.
* When custom_eval is set, require maximize_evaluation_metrics.
* Update documents on early stop in XGBoost4J-Spark.
* Fix code error.
* Make CMakeLists.txt compatible with CMake 3.3; require CMake 3.11 for MSVC
* Use CMake 3.12 when sanitizer is enabled
* Disable funroll-loops for MSVC
* Use cmake version in container name
* Add missing arg
* Fix egrep use in ci_build.sh
* Display CMake version
* Do not set OpenMP_CXX_LIBRARIES for MSVC
* Use cmake_minimum_required()
* Use feature interaction constraints to narrow search space for split candidates.
* fix clang-tidy broken at updater_quantile_hist.cc:535:3
* make const
* fix
* try to fix exception thrown in java_test
* fix suspected mistake which cause EvaluateSplit error
* try fix
* Fix bug: feature ID and node ID swapped in argument
* Rename CheckValidation() to CheckFeatureConstraint() for clarity
* Do not create temporary vector validFeatures, to enable parallelism
* Combine thread launches into single launch per tree for gpu_hist
algorithm.
* Address deprecation warning
* Add manual column sampler constructor
* Turn off omp dynamic to get a guaranteed number of threads
* Enable openmp in cuda code
* All Linux tests are now in Jenkins CI
* Tests are now de-coupled from builds. We can now build XGBoost with one version of CUDA/JDK and test it with another version of CUDA/JDK
* Builds (compilation) are significantly faster because 1) They use C5 instances with faster CPU cores; and 2) build environment setup is cached using Docker containers
* fix the nan and non-zero missing value handling
* fix nan handling part
* add missing value
* Update MissingValueHandlingSuite.scala
* Update MissingValueHandlingSuite.scala
* stylistic fix
* [jvm-packages][hot-fix] fix column mismatch caused by zip actions at XGBooostModel.transformInternal
* apply minibatch in prediction
* an iterator-compatible minibatch prediction
* regressor impl
* continuous working on mini-batch prediction of xgboost4j-spark
* Update Booster.java
* Refactor CMake scripts.
* Remove CMake CUDA wrapper.
* Bump CMake version for CUDA.
* Use CMake to handle Doxygen.
* Split up CMakeList.
* Export install target.
* Use modern CMake.
* Remove build.sh
* Workaround for gpu_hist test.
* Use cmake 3.12.
* Revert machine.conf.
* Move CLI test to gpu.
* Small cleanup.
* Support using XGBoost as submodule.
* Fix windows
* Fix cpp tests on Windows
* Remove duplicated find_package.
* [r-package] cut CI-time dependency on craigcitro/r-travis (fixes#4348)
* Install R
* Install R on OSX
* Remove gfortran symlink
* Specify CRAN repo
* added more R dependencies needed for testing
* removed heavy R dependencies in CI
* fixed bug in env var, removed unnecessary apt installs of R
* fix to R installs
The old NativeLibLoader had a short-circuit load path which modified
java.library.path and attempted to load the xgboost library from outside
the jar first, falling back to loading the library from inside the jar.
This path is a no-op every time when using XGBoost outside of it's
source tree. Additionally it triggers an illegal reflective access
warning in the module system in 9, 10, and 11.
On Java 12 the ClassLoader fields are not accessible via reflection
(separately from the illegal reflective acces warning), and so it fails
in a way that isn't caught by the code which falls back to loading the
library from inside the jar.
This commit removes that code path and always loads the xgboost library
from inside the jar file as it's a valid technique across multiple JVM
implementations and works with all versions of Java.
* Fix Histogram allocation.
nidx_map is cleared after `Reset`, but histogram data size isn't changed hence
histogram recycling is used in later iterations. After a reset(building new
tree), newly allocated node will start from 0, while recycling always choose
the node with smallest index, which happens to be our newly allocated node 0.
* When building pull requests, use Docker cache for master branch
Docker build caches are per-branch, so new pull requests will initially
have no build cache, causing the Docker containers to be built from
scratch. New pull requests should use the cache associated with the
master branch. This makes sense, since most pull requests do not modify
the Dockerfile.
* Add comments
* make the assignments of HostDeviceVector exception safe.
* storing a dummy GPUDistribution instance in HDV for CPU based code.
* change testxgboost binary location to build directory.
* Make train in xgboost4j respect print params
Previously no setting in params argument of Booster::train would prevent
the Rabit.trackerPrint call. This can fill up a lot of screen space in
the case that many folds are being trained.
* Setting "silent" in this map to "true", "True", a non-zero integer, or
a string that can be parsed to such an int will prevent printing.
* Setting "verbose_eval" to "False" or "false" will prevent printing.
* Setting "verbose_eval" to an int (or a String parseable to an int) n
will result in printing every n steps, or no printing is n is zero.
This is to match the python behaviour described here:
https://www.kaggle.com/c/rossmann-store-sales/discussion/17499
* Fixed 'slient' typo in xgboost4j test
* private access on two methods
* Optimisations for gpu_hist.
* Use streams to overlap operations.
* ColumnSampler now uses HostDeviceVector to prevent repeatedly copying feature vectors to the device.
* Brought the silent parameter for the SKLearn-like API back, marked it deprecated.
- added deprecation notice and warning
- removed silent from the tests for the SKLearn-like API
* Improved multi-node multi-GPU random forests.
- removed rabit::Broadcast() from each invocation of column sampling
- instead, syncing the PRNG seed when a ColumnSampler() object is constructed
- this makes non-trivial column sampling significantly faster in the distributed case
- refactored distributed GPU tests
- added distributed random forests tests
* Upgrade gtest for clang-tidy.
* Use CMake to install GTest instead of mv.
* Don't enforce clang-tidy to return 0 due to errors in thrust.
* Add a small test for tidy itself.
* Reformat.
* Added SKLearn-like random forest Python API.
- added XGBRFClassifier and XGBRFRegressor classes to SKL-like xgboost API
- also added n_gpus and gpu_id parameters to SKL classes
- added documentation describing how to use xgboost for random forests,
as well as existing caveats
* Initial commit to support multi-node multi-gpu xgboost using dask
* Fixed NCCL initialization by not ignoring the opg parameter.
- it now crashes on NCCL initialization, but at least we're attempting it properly
* At the root node, perform a rabit::Allreduce to get initial sum_gradient across workers
* Synchronizing in a couple of more places.
- now the workers don't go down, but just hang
- no more "wild" values of gradients
- probably needs syncing in more places
* Added another missing max-allreduce operation inside BuildHistLeftRight
* Removed unnecessary collective operations.
* Simplified rabit::Allreduce() sync of gradient sums.
* Removed unnecessary rabit syncs around ncclAllReduce.
- this improves performance _significantly_ (7x faster for overall training,
20x faster for xgboost proper)
* pulling in latest xgboost
* removing changes to updater_quantile_hist.cc
* changing use_nccl_opg initialization, removing unnecessary if statements
* added definition for opaque ncclUniqueId struct to properly encapsulate GetUniqueId
* placing struct defintion in guard to avoid duplicate code errors
* addressing linting errors
* removing
* removing additional arguments to AllReduer initialization
* removing distributed flag
* making comm init symmetric
* removing distributed flag
* changing ncclCommInit to support multiple modalities
* fix indenting
* updating ncclCommInitRank block with necessary group calls
* fix indenting
* adding print statement, and updating accessor in vector
* improving print statement to end-line
* generalizing nccl_rank construction using rabit
* assume device_ordinals is the same for every node
* test, assume device_ordinals is identical for all nodes
* test, assume device_ordinals is unique for all nodes
* changing names of offset variable to be more descriptive, editing indenting
* wrapping ncclUniqueId GetUniqueId() and aesthetic changes
* adding synchronization, and tests for distributed
* adding to tests
* fixing broken #endif
* fixing initialization of gpu histograms, correcting errors in tests
* adding to contributors list
* adding distributed tests to jenkins
* fixing bad path in distributed test
* debugging
* adding kubernetes for distributed tests
* adding proper import for OrderedDict
* adding urllib3==1.22 to address ordered_dict import error
* added sleep to allow workers to save their models for comparison
* adding name to GPU contributors under docs
* Fix early stop with xgboost4j-spark
* Update XGBoost.java
* Update XGBoost.java
* Update XGBoost.java
To use -Float.MAX_VALUE as the lower bound, in case there is positive metric.
* Only update best score if the current score is better (no update when equal)
* Update xgboost-spark tutorial to fix early stopping docs.
* Fix test_gpu_coordinate.
* Use `gpu_coord_descent` in test.
* Reduce number of running rounds.
* Remove nthread.
* Use githubusercontent for r-appveyor.
* Use githubusercontent in travis r tests.
* Prevent empty quantiles
* Revise and improve unit tests for quantile hist
* Remove unnecessary comment
* Add #2943 as a test case
* Skip test if no sklearn
* Revise misleading comments
* Add checks for group size.
* Simple docs.
* Search group index during hist cut matrix initialization.
Co-authored-by: Jiaming Yuan <jm.yuan@outlook.com>
Co-authored-by: Philip Hyunsu Cho <chohyu01@cs.washington.edu>
* Fix broken R test: Install Homebrew GCC
Missing GCC Fortran causes installation failure of a dependency package
(igraph)
* Register gfortran system-wide
* Use correct keg
* Set env vars to change compiler choice
* Do not break other Mac builds
* Nuclear option: symlink gfortran
* Use /usr/local/bin instead of /usr/bin
* Symlink library path too
* Update run_test.sh
* Remove GHistRow, GHistEntry, GHistIndexRow.
* Remove kSimpleStats.
* Remove CheckInfo, SetLeafVec in GradStats and in SKStats.
* Clean up the GradStats.
* Cleanup calcgain.
* Move LossChangeMissing out of common.
* Remove [] operator from GHistIndexBlock.
* Basic script for using compilation database.
* Add `GENERATE_COMPILATION_DATABASE' to CMake.
* Rearrange CMakeLists.txt.
* Add basic python clang-tidy script.
* Remove modernize-use-auto.
* Add clang-tidy to Jenkins
* Refine logic for correct path detection
In Jenkins, the project root is of form /home/ubuntu/workspace/xgboost_PR-XXXX
* Run clang-tidy in CUDA 9.2 container
* Use clang_tidy container
* Enable xgb_model parameter in XGClassifier scikit-learn API
https://github.com/dmlc/xgboost/issues/3049
* add test_XGBClassifier_resume():
test for xgb_model parameter in XGBClassifier API.
* Update test_with_sklearn.py
* Fix lint
* Fix failing Travis CI on Mac
Use Homebrew Addon + latest Mac image
* Use long command for pytest
* Downgrade OSX image to xcode9.3, to use Java 8
* Install pytest in Python 2 environment
* Remove clang-tidy from Travis
- ./testxgboost (without filters) failed if run on a multi-GPU machine because
the memory was allocated on the current device, but device 0
was always passed into LaunchN
* Initial performance optimizations for xgboost
* remove includes
* revert float->double
* fix for CI
* fix for CI
* fix for CI
* fix for CI
* fix for CI
* fix for CI
* fix for CI
* fix for CI
* fix for CI
* fix for CI
* Check existence of _mm_prefetch and __builtin_prefetch
* Fix lint
* Updates to Booster to support other feature importances
* Add returns for Java methods
* Pass Scala style checks
* Pass Java style checks
* Fix indents
* Use class instead of enum
* Return map string double
* A no longer broken build, thanks to mvn package local build
* Add a unit test to increase code coverage back
* Address code review on main code
* Add more unit tests for different feature importance scores
* Address more CR
* Use Span in gpu coordinate.
* Use Span in device code.
* Fix shard size calculation.
- Use lower_bound instead of upper_bound.
* Check empty devices.
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* fix scalastyle error
* fix scalastyle error
* wrap iterators
* enable copartition training and validationset
* add parameters
* converge code path and have init unit test
* enable multi evals for ranking
* unit test and doc
* update example
* fix early stopping
* address the offline comments
* udpate doc
* test eval metrics
* fix compilation issue
* fix example
* Unify logging facilities.
* Enhance `ConsoleLogger` to handle different verbosity.
* Override macros from `dmlc`.
* Don't use specialized gamma when building with GPU.
* Remove verbosity cache in monitor.
* Test monitor.
* Deprecate `silent`.
* Fix doc and messages.
* Fix python test.
* Fix silent tests.
* Ensure lists cannot be passed into DMatrix
The documentation does not include lists as an allowed type for the data inputted into DMatrix. Despite this, a list can be passed in without an error. This change would prevent a list form being passed in directly.
* update description of early stopping rounds
the description of early stopping round was quite inconsistent in the scikit-learn api section since the fit paragraph tells that when early stopping rounds occurs, the last iteration is returned not the best one, but the predict paragraph tells that when the predict is called without ntree_limit specified, then ntree_limit is equals to best_ntree_limit.
Thus, when reading the fit part, one could think that it is needed to specify what is the best iter when calling the predict, but when reading the predict part, then the best iter is given by default, it is the last iter that you have to specify if needed.
* Update sklearn.py
* Update sklearn.py
fix doc according to the python_lightweight_test error
* Port elementwise metrics to GPU.
* All elementwise metrics are converted to static polymorphic.
* Create a reducer for metrics reduction.
* Remove const of Metric::Eval to accommodate CubMemory.
- Improved GPU performance logging
- Only use one execute shards function
- Revert performance regression on multi-GPU
- Use threads to launch NCCL AllReduce
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* fix scalastyle error
* fix scalastyle error
* update version
* 0.82
* fix early stopping condition
* remove unused
* update comments
* udpate comments
* update test
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* fix scalastyle error
* fix scalastyle error
* update version
* 0.82
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* fix scalastyle error
* fix scalastyle error
* wrap iterators
* remove unused code
* refactor
* fix typo
* use gain for sklearn feature_importances_
`gain` is a better feature importance criteria than the currently used `weight`
* added importance_type to class
* fixed test
* white space
* fix variable name
* fix deprecation warning
* fix exp array
* white spaces
* Enable running objectives with 0 GPU.
* Enable 0 GPU for objectives.
* Add doc for GPU objectives.
* Fix some objectives defaulted to running on all GPUs.
* Make C++ unit tests run and pass on Windows
* Fix logic for external memory. The letter ':' is part of drive letter,
so remove the drive letter before splitting on ':'.
* Cosmetic syntax changes to keep MSVC happy.
* Fix lint
* Add Windows guard
* Fix#3342 and h2oai/h2o4gpu#625: Save predictor parameters in model file
This allows pickled models to retain predictor attributes, such as
'predictor' (whether to use CPU or GPU) and 'n_gpu' (number of GPUs
to use). Related: h2oai/h2o4gpu#625Closes#3342.
TODO. Write a test.
* Fix lint
* Do not load GPU predictor into CPU-only XGBoost
* Add a test for pickling GPU predictors
* Make sample data big enough to pass multi GPU test
* Update test_gpu_predictor.cu
* Fix#3747: Add coef_ and intercept_ as properties of sklearn wrapper
Scikit-learn expects linear learners to expose `coef_` and `intercept_`
as properties.
Closes#3747.
* Fix lint
* Clean up logic for converting tree_method to updater sequence
* Use C++11 enum class for extra safety
Compiler will give warnings if switch statements don't handle all
possible values of C++11 enum class.
Also allow enum class to be used as DMLC parameter.
* Fix compiler error + lint
* Address reviewer comment
* Better docstring for DECLARE_FIELD_ENUM_CLASS
* Fix lint
* Add C++ test to see if tree_method is recognized
* Fix clang-tidy error
* Add test_learner.h to R package
* Update comments
* Fix lint error
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* fix scalastyle error
* fix scalastyle error
* documenting tracker
* Make it a separate note
The `save_model()` and `load_model()` method only saves the part of the model
that's common to all language interfaces and do not preserve Python-specific
attributes, such as `feature_names`. More crucially, label encoder is not
preserved either; this is needed for the scikit-learn wrapper, since you may
have string labels.
Fix: Explicitly recommend pickling as the way to save scikit-learn model
objects.
* Multi-GPU support in GPUPredictor.
- GPUPredictor is multi-GPU
- removed DeviceMatrix, as it has been made obsolete by using HostDeviceVector in DMatrix
* Replaced pointers with spans in GPUPredictor.
* Added a multi-GPU predictor test.
* Fix multi-gpu test.
* Fix n_rows < n_gpus.
* Reinitialize shards when GPUSet is changed.
* Tests range of data.
* Remove commented code.
* Remove commented code.
* Enable auto-locking of issues closed long ago
Issues that were closed more than 90 days ago will be locked automatically so
that no additional comments would be allowed. We will use a bot to do
this: https://probot.github.io/apps/lock/
Background: As a maintainer, I often see people leaving comments to old issue
posts that were closed long ago. Those comments are hard to discover and assist
with, since they get buried under list of other active issues.
With the change, users who want to follow up with an old issue would be asked
to file a new issue.
* Exempt `feature-request` from auto locking
* Disable comment to avoid triggering notification
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* fix scalastyle error
* fix scalastyle error
* temp
* add method for classifier and regressor
* update tutorial
* address the comments
* update
A privilege escalation vulnerability (CVE-2017-15288) has been
identified in the Scala compilation daemon. See
https://nvd.nist.gov/vuln/detail/CVE-2017-15288
Fix: Upgrade Scala to 2.11.12.
**Symptom** Apple Clang's implementation of `std::shuffle` expects doesn't work
correctly when it is run with the random bit generator for R package:
```cpp
CustomGlobalRandomEngine::result_type
CustomGlobalRandomEngine::operator()() {
return static_cast<result_type>(
std::floor(unif_rand() * CustomGlobalRandomEngine::max()));
}
```
Minimial reproduction of failure (compile using Apple Clang 10.0):
```cpp
std::vector<int> feature_set(100);
std::iota(feature_set.begin(), feature_set.end(), 0);
// initialize with 0, 1, 2, 3, ..., 99
std::shuffle(feature_set.begin(), feature_set.end(), common::GlobalRandom());
// This returns 0, 1, 2, ..., 99, so content didn't get shuffled at all!!!
```
Note that this bug is platform-dependent; it does not appear when GCC or
upstream LLVM Clang is used.
**Diagnosis** Apple Clang's `std::shuffle` expects 32-bit integer
inputs, whereas `CustomGlobalRandomEngine::operator()` produces 64-bit
integers.
**Fix** Have `CustomGlobalRandomEngine::operator()` produce 32-bit integers.
Closes#3523.
* Split building histogram into separated class.
* Extract `InitCompressedRow` definition.
* Basic tests for gpu-hist.
* Document the code more verbosely.
* Removed `HistCutUnit`.
* Removed some duplicated copies in `GPUHistMaker`.
* Implement LCG and use it in tests.
* Added some instructions on using MinGW-built XGBoost with python.
* Changes according to the discussion and some additions
* Fixed wording and removed redundancy.
* Even more fixes
* Fixed links. Removed redundancy.
* Some fixes according to the discussion
* fixes
* Some fixes
* fixes
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* fix scalastyle error
* fix scalastyle error
* sparjJobThread
* update
* fix issue when spark job execution thread cannot return before we execute first()
* Implement Transform class.
* Add tests for softmax.
* Use Transform in regression, softmax and hinge objectives, except for Cox.
* Mark old gpu objective functions deprecated.
* static_assert for softmax.
* Split up multi-gpu tests.
* DMatrix refactor 2
* Remove buffered rowset usage where possible
* Transition to c++11 style iterators for row access
* Transition column iterators to C++ 11
* Add multi-GPU unit test environment
* Better assertion message
* Temporarily disable failing test
* Distinguish between multi-GPU and single-GPU CPP tests
* Consolidate Python tests. Use attributes to distinguish multi-GPU Python tests from single-CPU counterparts
* 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
* add back train method but mark as deprecated
* fix scalastyle error
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* fix scalastyle error
* instrumentation
* use log console
* better measurement
* fix erros in example
* update histmaker
* add a demo of multi-class classification R version
* add a demo of multi-class classification result
* add intro to the demo readme
* Delete train.md
* Update README.md
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* remove copy paste error
* added test, commented out right now
* reinstated test
* added fix for checking encryption settings
* fix by using RDD conf
* fix compilation
* renamed conf
* use SparkSession if available
* fix message
* nop
* code review fixes
* Fix#3397: early_stop callback does not maximize metric of form NDCG@n-
Early stopping callback makes splits with '-' letter, which interferes
with metrics of form NDCG@n-. As a result, XGBoost tries to minimize
NDCG@n-, where it should be maximized instead.
Fix. Specify maxsplit=1.
* Python 2.x compatibility fix
* Add scikit-learn tests
Goal is to pass scikit-learn's check_estimator() for XGBClassifier,
XGBRegressor, and XGBRanker. It is actually not possible to do so
entirely, since check_estimator() assumes that NaN is disallowed,
but XGBoost allows for NaN as missing values. However, it is always
good ideas to add some checks inspired by check_estimator().
* Fix lint
* Fix lint
* add interaction constraints
* enable both interaction and monotonic constraints at the same time
* fix lint
* add R test, fix lint, update demo
* Use dmlc::JSONReader to express interaction constraints as nested lists; Use sparse arrays for bookkeeping
* Add Python test for interaction constraints
* make R interaction constraints parameter based on feature index instead of column names, fix R coding style
* Fix lint
* Add BlueTea88 to CONTRIBUTORS.md
* Short circuit when no constraint is specified; address review comments
* Add tutorial for feature interaction constraints
* allow interaction constraints to be passed as string, remove redundant column_names argument
* Fix typo
* Address review comments
* Add comments to Python test
- previously, vec_ in DeviceShard wasn't updated on copy; as a result,
the shards continued to refer to the old HostDeviceVectorImpl object,
which resulted in a dangling pointer once that object was deallocated
* Fix#3648: XGBClassifier.predict() should return margin scores when output_margin=True
* Fix tests to reflect correct implementation of XGBClassifier.predict(output_margin=True)
* Fix flaky test test_with_sklearn.test_sklearn_api_gblinear
* Replaced std::vector with HostDeviceVector in MetaInfo and SparsePage.
- added distributions to HostDeviceVector
- using HostDeviceVector for labels, weights and base margings in MetaInfo
- using HostDeviceVector for offset and data in SparsePage
- other necessary refactoring
* Added const version of HostDeviceVector API calls.
- const versions added to calls that can trigger data transfers, e.g. DevicePointer()
- updated the code that uses HostDeviceVector
- objective functions now accept const HostDeviceVector<bst_float>& for predictions
* Updated src/linear/updater_gpu_coordinate.cu.
* Added read-only state for HostDeviceVector sync.
- this means no copies are performed if both host and devices access
the HostDeviceVector read-only
* Fixed linter and test errors.
- updated the lz4 plugin
- added ConstDeviceSpan to HostDeviceVector
- using device % dh::NVisibleDevices() for the physical device number,
e.g. in calls to cudaSetDevice()
* Fixed explicit template instantiation errors for HostDeviceVector.
- replaced HostDeviceVector<unsigned int> with HostDeviceVector<int>
* Fixed HostDeviceVector tests that require multiple GPUs.
- added a mock set device handler; when set, it is called instead of cudaSetDevice()
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* interrupted exception is not rethrown
* Add XGBRanker to Python API doc
* Show inherited members of XGBRegressor in API doc, since XGBRegressor uses default methods from XGBModel
* Add table of contents to Python API doc
* Skip JVM doc download if not available
* Show inherited members for XGBRegressor and XGBRanker
* Expose XGBRanker to Python XGBoost module directory
* Add docstring to XGBRegressor.predict() and XGBRanker.predict()
* Fix rendering errors in Python docstrings
* Fix lint
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* fix update checkpoint func
* added xgbranker
* fixed predict method and ranking test
* reformatted code in accordance with pep8
* fixed lint error
* fixed docstring and added checks on objective
* added ranking demo for python
* fixed suffix in rank.py
* Add basic Span class based on ISO++20.
* Use Span<Entry const> instead of Inst in SparsePage.
* Add DeviceSpan in HostDeviceVector, use it in regression obj.
This pull request amends the broken #3062 allow Spark 2.2 to work.
Please note this won't work in Spark <=2.1 as sc.removeSparkListener was implemented in Spark 2.2. (So perhaps a more general method is better, although that is what was attempted in #3062)
This PR fixes: #3208, #3151 and the discussion in #1927.
I do find it strange that #3062 dose not work in Spark 2.2, it's probably due to some sort of public/private issue in the org.apache.spark.scheduler.LiveListenerBus class inheritance (In Spark itself). The error is: `java.lang.NoSuchMethodError: org.apache.spark.scheduler.LiveListenerBus.removeListener(Ljava/lang/Object;)V`
* Adding Java/Scala doc build to Jenkins CI
* Deploy built doc to S3 bucket
* Build doc only for branches
* Build doc first, to get doc faster for branch updates
* Have ReadTheDocs download doc tarball from S3
* Update JVM doc links
* Put doc build commands in a script
* Specify Spark 2.3+ requirement for XGBoost4J-Spark
* Build GPU wheel without NCCL, to reduce binary size
* Revert "Fix #3485, #3540: Don't use dropout for predicting test sets (#3556)"
This reverts commit 44811f2330.
* Document behavior of predict() for DART booster
* Add notice to parameter.rst
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* partial finish
* no test
* add test cases
* add test cases
* address comments
* add test for regressor
* fix typo
* Fix#3545: XGDMatrixCreateFromCSCEx silently discards empty trailing rows
Description: The bug is triggered when
1. The data matrix has empty rows at the bottom. More precisely, the rows
`n-k+1`, `n-k+2`, ..., `n` of the matrix have missing values in all
dimensions (`n` number of instances, `k` number of trailing rows)
2. The data matrix is given as Compressed Sparse Column (CSC) format.
Diagnosis: When the CSC matrix is converted to Compressed Sparse Row (CSR)
format (this is common format used for DMatrix), the trailing empty rows
are silently ignored. More specifically, the row pointer (`offset`) of the
newly created CSR matrix does not take account of these rows.
Fix: Modify the row pointer.
* Add regression test
The base margin will need to have length `[num_class] * [number of data points]`.
Otherwise, the array holding prediction results will be only partially
initialized, causing undefined behavior.
Fix: check the length of the base margin. If the length is not correct,
use the global bias (`base_score`) instead. Warn the user about the
substitution.
* Fix#3402: wrong fid crashes distributed algorithm
The bug was introduced by the recent DMatrix refactor (#3301). It was partially
fixed by #3408 but the example in #3402 was still failing. The example in #3402
will succeed after this fix is applied.
* Explicitly specify "this" to prevent compile error
* Add regression test
* Add distributed test to Travis matrix
* Install kubernetes Python package as dependency of dmlc tracker
* Add Python dependencies
* Add compile step
* Reduce size of regression test case
* Further reduce size of test
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* add new
* update doc
* finish Gang Scheduling
* more
* intro
* Add sections: Prediction, Model persistence and ML pipeline.
* Add XGBoost4j-Spark MLlib pipeline example
* partial finished version
* finish the doc
* adjust code
* fix the doc
* use rst
* Convert XGBoost4J-Spark tutorial to reST
* Bring XGBoost4J up to date
* add note about using hdfs
* remove duplicate file
* fix descriptions
* update doc
* Wrap HDFS/S3 export support as a note
* update
* wrap indexing_mode example in code block
This bring many goodies, including:
* Ability to specify delimiter and weight_column for CSV files:
```python
dtrain = xgboost.DMatrix('train.csv?format=csv&label_column=0&weight_column=1&delimiter= ')
```
* Ability to choose between 0-based and 1-based indexing for LIBSVM/LIBFM files:
```python
dtrain = xgboost.DMatrix('train.libsvm?indexing_mode=1') # use 1-based indexing
dtest = xgboost.DMatrix('test.libsvm') # use 0-based indexing (default)
dtest2 = xgboost.DMatrix('test2.libsvm?indexing_mode=-1') # use heuristic to detect 0-based / 1-based
```
* Fix a bug in float parsing (issue dmlc/dmlc-core#440)
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* consider spark.task.cpus when controlling parallelism
* fix bug
* fix conf setup
* calculate requestedCores within ParallelismController
* enforce spark.task.cpus = 1
* unify unit test case framework
* enable spark ui
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* consider missing value in prediction
* handle single prediction instance
* fix type conversion
* Fix bug of using list(x) function when x is string
list('abcdcba') = ['a', 'b', 'c', 'd', 'c', 'b', 'a']
* Allow feature_names/feature_types to be of any type
If feature_names/feature_types is iterable, e.g. tuple, list, then convert the value to list, except for string; otherwise construct a list with a single value
* Delete excess whitespace
* Fix whitespace to pass lint
* Expand histogram memory dynamically to prevent large allocations for large tree depths (e.g. > 15)
* Remove GPU memory allocation messages. These are misleading as a large number of allocations are now dynamic.
* Fix appveyor R test
* Save max_delta_step as an extra attribute of Booster
Fixes#3509 and #3026, where `max_delta_step` parameter gets lost during serialization.
* fix lint
* Use camel case for global constant
* disable local variable case in clang-tidy
* Added finding quantiles on GPU.
- this includes datasets where weights are assigned to data rows
- as the quantiles found by the new algorithm are not the same
as those found by the old one, test thresholds in
tests/python-gpu/test_gpu_updaters.py have been adjusted.
* Adjustments and improved testing for finding quantiles on the GPU.
- added C++ tests for the DeviceSketch() function
- reduced one of the thresholds in test_gpu_updaters.py
- adjusted the cuts found by the find_cuts_k kernel
Add `'total_gain'` and `'total_cover'` as possible `importance_type`
arguments to `Booster.get_score` in the Python package.
`get_score` already accepts a `'gain'` argument, which returns each
feature's average gain over all of its splits. `'total_gain'` does the
same, but returns a total rather than an average. This seems more
intuitively meaningful, and also matches the behavior of the R package's
`xgb.importance` function.
I also added an analogous `'total_cover'` command for consistency.
This should resolve#3484.
* Improved library loading a bit
* Fixed indentation.
* Fixes according to the discussion
* Moved the comment to a separate line.
* specified exception type
* Change doc build to reST exclusively
* Rewrite Intro doc in reST; create toctree
* Update parameter and contribute
* Convert tutorials to reST
* Convert Python tutorials to reST
* Convert CLI and Julia docs to reST
* Enable markdown for R vignettes
* Done migrating to reST
* Add guzzle_sphinx_theme to requirements
* Add breathe to requirements
* Fix search bar
* Add link to user forum
* Fail GPU CI after test failure
* Fix GPU linear tests
* Reduced number of GPU tests to speed up CI
* Remove static allocations of device memory
* Resolve illegal memory access for updater_fast_hist.cc
* Fix broken r tests dependency
* Update python install documentation for GPU
* Upgrading to NCCL2
* Part - II of NCCL2 upgradation
- Doc updates to build with nccl2
- Dockerfile.gpu update for a correct CI build with nccl2
- Updated FindNccl package to have env-var NCCL_ROOT to take precedence
* Upgrading to v9.2 for CI workflow, since it has the nccl2 binaries available
* Added NCCL2 license + copy the nccl binaries into /usr location for the FindNccl module to find
* Set LD_LIBRARY_PATH variable to pick nccl2 binary at runtime
* Need the nccl2 library download instructions inside Dockerfile.release as well
* Use NCCL2 as a static library
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* disable booster setup in spark
* check in parameter conversion
* fix compilation issue
* update exception type
* add qid for https://github.com/dmlc/xgboost/issues/2748
* change names
* change spaces
* change qid to bst_uint type
* change qid type to size_t
* change qid first to SIZE_MAX
* change qid type from size_t to uint64_t
* update dmlc-core
* fix qids name error
* fix group_ptr_ error
* Style fix
* Add qid handling logic to SparsePage
* New MetaInfo format + backward compatibility fix
Old MetaInfo format (1.0) doesn't contain qid field. We still want to be able
to read from MetaInfo files saved in old format. Also, define a new format
(2.0) that contains the qid field. This way, we can distinguish files that
contain qid and those that do not.
* Update MetaInfo test
* Simply group assignment logic
* Explicitly set qid=nullptr in NativeDataIter
NativeDataIter's callback does not support qid field. Users of NativeDataIter
will need to call setGroup() function separately to set group information.
* Save qids_ in SaveBinary()
* Upgrade dmlc-core submodule
* Add a test for reading qid
* Add contributor
* Check the size of qids_
* Document qid format
* allow arbitrary cross validation fold indices
- use training indices passed to `folds` parameter in `training.cv`
- update doc string
* add tests for arbitrary fold indices
* Refactor to allow for custom regularisation methods
* Implement compositional SplitEvaluator framework
* Fixed segfault when no monotone_constraints are supplied.
* Change pid to parentID
* test_monotone_constraints.py now passes
* Refactor ColMaker and DistColMaker to use SplitEvaluator
* Performance optimisation when no monotone_constraints specified
* Fix linter messages
* Fix a few more linter errors
* Update the amalgamation
* Add bounds check
* Add check for leaf node
* Fix linter error in param.h
* Fix clang-tidy errors on CI
* Fix incorrect function name
* Fix clang-tidy error in updater_fast_hist.cc
* Enable SSE2 for Win32 R MinGW
Addresses https://github.com/dmlc/xgboost/pull/3335#issuecomment-400535752
* Add contributor
CI tests were failing because wget prompts "the user" for a response
whenever the google test archive is already on the disk.
Fix: Use `-nc` option to skip download when the archive already
exists
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* maven central release
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* [jvm-packages] XGBoost Spark integration refactor. (#3313)
* XGBoost Spark integration refactor.
* Make corresponding update for xgboost4j-example
* Address comments.
* [jvm-packages] Refactor XGBoost-Spark params to make it compatible with both XGBoost and Spark MLLib (#3326)
* Refactor XGBoost-Spark params to make it compatible with both XGBoost and Spark MLLib
* Fix extra space.
* [jvm-packages] XGBoost Spark supports ranking with group data. (#3369)
* XGBoost Spark supports ranking with group data.
* Use Iterator.duplicate to prevent OOM.
* Update CheckpointManagerSuite.scala
* Resolve conflicts
* Use sparse page as singular CSR matrix representation
* Simplify dmatrix methods
* Reduce statefullness of batch iterators
* BREAKING CHANGE: Remove prob_buffer_row parameter. Users are instead recommended to sample their dataset as a preprocessing step before using XGBoost.
* GPU binning and compression.
- binning and index compression are done inside the DeviceShard constructor
- in case of a DMatrix with multiple row batches, it is first converted into a single row batch
Currently, `CLIPredict()` saves prediction results in default 6-digit precision which causes precision loss. This PR sets precision to a level so that the conversion back to `bst_float` is lossless.
Related: #3298.
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* update 0.80
* Fix print.xgb.Booster
valid_handle should be TRUE when x$handle is NOT null
* Update xgb.Booster.R
Modify is.null.handle to return TRUE for NULL handle
* Add option to use weights when evaluating metrics in validation sets
* Add test for validation-set weights functionality
* simplify case with no weights for test sets
* fix lint issues
* For CRAN submission, remove all #pragma's that suppress compiler warnings
A few headers in dmlc-core contain #pragma's that disable compiler warnings,
which is against the CRAN submission policy. Fix the problem by removing
the offending #pragma's as part of the command `make Rbuild`.
This addresses issue #3322.
* Fix script to improve Cygwin/MSYS compatibility
We need this to pass rmingw CI test
* Remove remove_warning_suppression_pragma.sh from packaged tarball
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* static glibc glibc++
* update to build with glib 2.12
* remove unsupported flags
* update version number
* remove properties
* remove unnecessary command
* update poms
* Update dmlc-core submodule
* Fix dense_parser to work with the latest dmlc-core
* Specify location of Google Test
* Add more source files in dmlc-minimum to get latest dmlc-core working
* Update dmlc-core submodule
* Adjust xgboost entries in .gitignore
They were overly broad. In particularly this was inconvenient when
working with tools such as fzf that use the .gitignore to decide what to
include. As written, we'd not look into /include/xgboost.
* Make cosmetic improvements to .gitignore
* Remove dmlc-core from .gitignore
This seems unnecessary and has the drawback that tools that use
.gitignore to know files to skip mean they won't look here, and being
able to inspect the submodule files with them is useful.
* Increase precision of bst_float values in tree dumps
* Increase precision of bst_float values in tree dumps
* Fix lint error and switch precision to right float variable
* Fix clang-tidy error
* Multi-GPU HostDeviceVector.
- HostDeviceVector instances can now span multiple devices, defined by GPUSet struct
- the interface of HostDeviceVector has been modified accordingly
- GPU objective functions are now multi-GPU
- GPU predicting from cache is now multi-GPU
- avoiding omp_set_num_threads() calls
- other minor changes
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* change version of jvm to keep consistent with other pkgs
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* update default spark version to 2.3
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* add back libsvm notes
* Now `make pippack` works without any manual action: it will produce
xgboost-[version].tar.gz, which one can use by typing
`pip3 install xgboost-[version].tar.gz`.
* Detect OpenMP-capable compilers (clang, gcc-5, gcc-7) on MacOS
* rank_metric: add AUC-PR
Implementation of the AUC-PR calculation for weighted data, proposed by Keilwagen, Grosse and Grau (https://doi.org/10.1371/journal.pone.0092209)
* rank_metric: fix lint warnings
* Implement tests for AUC-PR and fix implementation
* add aucpr to documentation for other languages
* fix rebase conflict
* [core] additional gblinear improvements
* [R] callback for gblinear coefficients history
* force eta=1 for gblinear python tests
* add top_k to GreedyFeatureSelector
* set eta=1 in shotgun test
* [core] fix SparsePage processing in gblinear; col-wise multithreading in greedy updater
* set sorted flag within TryInitColData
* gblinear tests: use scale, add external memory test
* fix multiclass for greedy updater
* fix whitespace
* fix typo
* Extended monotonic constraints support to 'hist' tree method.
* Added monotonic constraints tests.
* Fix the signature of NoConstraint::CalcSplitGain()
* Document monotonic constraint support in 'hist'
* Update signature of Update to account for latest refactor
* Support CSV file in DMatrix
We'd just need to expose the CSV parser in dmlc-core to the Python wrapper
* Revert extra code; document existing CSV support
CSV support is already there but undocumented
* Add notice about categorical features
* Replaced std::vector-based interfaces with HostDeviceVector-based interfaces.
- replacement was performed in the learner, boosters, predictors,
updaters, and objective functions
- only interfaces used in training were replaced;
interfaces like PredictInstance() still use std::vector
- refactoring necessary for replacement of interfaces was also performed,
such as using HostDeviceVector in prediction cache
* HostDeviceVector-based interfaces for custom objective function example plugin.
* Fix doc build
ReadTheDocs build has been broken for a while due to incompatibilities between
commonmark, recommonmark, and sphinx. See:
* "Recommonmark not working with Sphinx 1.6"
https://github.com/rtfd/recommonmark/issues/73
* "CommonMark 0.6.0 breaks compatibility"
https://github.com/rtfd/recommonmark/issues/24
For now, we fix the versions to get the build working again
* Fix search bar
* added mingw64 installation instruction, and library file copy.
* Change all `libxgboost.dll` to `xgboost.dll`
On Windows, the library file is called `xgboost.dll`, not `libxgboost.dll` as in the build doc previously
In line 461, the "size_t offset = 0;" should be declared before any calculation, otherwise will cause compilation error.
```
I:\Libraries\xgboost\src\c_api\c_api.cc(416): error C2146: Missing ";" before "offset" [I:\Libraries\xgboost\build\objxgboost.vcxproj]
```
* Add interaction effects and cox loss
* Minimize whitespace changes
* Cox loss now no longer needs a pre-sorted dataset.
* Address code review comments
* Remove mem check, rename to pred_interactions, include bias
* Make lint happy
* More lint fixes
* Fix cox loss indexing
* Fix main effects and tests
* Fix lint
* Use half interaction values on the off-diagonals
* Fix lint again
- thrust::copy() called from dvec::copy() for gpairs invoked a GPU kernel instead of
cudaMemcpy()
- this resulted in illegal memory access if the GPU running the kernel could not access
the data being copied
- new version of dvec::copy() for thrust::device_ptr iterators calls cudaMemcpy(),
avoiding the problem.
* Added GPU objective function and no-copy interface.
- xgboost::HostDeviceVector<T> syncs automatically between host and device
- no-copy interfaces have been added
- default implementations just sync the data to host
and call the implementations with std::vector
- GPU objective function, predictor, histogram updater process data
directly on GPU
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* tiny fix for empty partition in predict
* further fix
* [jvm-packages] Prevent dispose being called twice when finalize
* Convert SIGSEGV to XGBoostError
* Avoid creating a new SBooster with the same JBooster
* Address CR Comments
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* fix the pattern in dev script and version mismatch
After installing ``gcc@5``, ``CMAKE_C_COMPILER`` will not be set to gcc-5 in some macOS environment automatically and the installation of xgboost will still fail. Manually setting the compiler will solve the problem.
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* add dev script to update version and update versions
* add back train method but mark as deprecated
* add back train method but mark as deprecated
* fix scalastyle error
* fix scalastyle error
* update resource files
* Update SparkParallelismTracker.scala
* remove xgboost-tracker.properties
* [jvm-packages] Train Booster from an existing model
* Align Scala API with Java API
* Existing model should not load rabit checkpoint
* Address minor comments
* Implement saving temporary boosters and loading previous booster
* Add more unit tests for loadPrevBooster
* Add params to XGBoostEstimator
* (1) Move repartition out of the temp model saving loop (2) Address CR comments
* Catch a corner case of training next model with fewer rounds
* Address comments
* Refactor newly added methods into TmpBoosterManager
* Add two files which is missing in previous commit
* Rename TmpBooster to checkpoint
* [jvm-packages] Fixed test/train persistence
Prior to this patch both data sets were persisted in the same directory,
i.e. the test data replaced the training one which led to
* training on less data (since usually test < train) and
* test loss being exactly equal to the training loss.
Closes#2945.
* Cleanup file cache after the training
* Addressed review comments
* [R] fix finding R.exe with cmake on WIN when it is in PATH
* [R] appveyor config for R package
* [R] wrap the lines to make R check happier
* [R] install only binary dep-packages in appveyor
* [R] for MSVC appveyor, also build a binary for R package and keep as an artifact
* [R] fix predict contributions for data with no colnames
* [R] add a render parameter for xgb.plot.multi.trees; fixes#2628
* [R] update Rd's
* [R] remove unnecessary dep-package from R cmake install
* silence type warnings; readability
* [R] silence complaint about incomplete line at the end
* [R] initial version of xgb.plot.shap()
* [R] more work on xgb.plot.shap
* [R] enforce black font in xgb.plot.tree; fixes#2640
* [R] if feature names are available, check in predict that they are the same; fixes#2857
* [R] cran check and lint fixes
* remove tabs
* [R] add references; a test for plot.shap
* Fix#2905
* Fix gpu_exact test failures
* Fix bug in GPU prediction where multiple calls to batch prediction can produce incorrect results
* Fix GPU documentation formatting
* Some minor changes to the code style
Some minor changes to the code style in file basic_walkthrough.py
* coding style changes
* coding style changes arrcording PEP8
* Update basic_walkthrough.py
* Fix minor typo
* Minor edits to coding style
Minor edits to coding style following the proposals of PEP8.
* [jvm-packages] Exposed train-time evaluation metrics
They are accessible via 'XGBoostModel.summary'. The summary is not
serialized with the model and is only available after the training.
* Addressed review comments
* Extracted model-related tests into 'XGBoostModelSuite'
* Added tests for copying the 'XGBoostModel'
* [jvm-packages] Fixed a subtle bug in train/test split
Iterator.partition (naturally) assumes that the predicate is deterministic
but this is not the case for
r.nextDouble() <= trainTestRatio
therefore sometimes the DMatrix(...) call got a NoSuchElementException
and crashed the JVM due to lack of exception handling in
XGBoost4jCallbackDataIterNext.
* Make sure train/test objectives are different
I found the installation of the Python XGBoost package to be problematic as the documentation around compiler requirements was unclear, as discussed in #1501. I decided that I would improve the README.
- Implement colsampling, subsampling for gpu_hist_experimental
- Optimised multi-GPU implementation for gpu_hist_experimental
- Make nccl optional
- Add Volta architecture flag
- Optimise RegLossObj
- Add timing utilities for debug verbose mode
- Bump required cuda version to 8.0
In the refactor to add base margins, #2532, all of the labels were lost
when creating the dmatrix. This became obvious as metrics like ndcg
always returned 1.0 regardless of the results.
Change-Id: I88be047e1c108afba4784bd3d892bfc9edeabe55
Training a model with the experimental rank:ndcg objective incorrectly
returns a Classification model. Adjust the classification check to
not recognize rank:* objectives as classification.
While writing tests for isClassificationTask also turned up that
obj_type -> regression was incorrectly identified as a classification
task so the function was slightly adjusted to pass the new tests.
* Some minor changes to the code style
Some minor changes to the code style in file basic_walkthrough.py
* coding style changes
* coding style changes arrcording PEP8
* Update basic_walkthrough.py
* Fatal error if GPU algorithm selected without GPU support compiled
* Resolve type conversion warnings
* Fix gpu unit test failure
* Fix compressed iterator edge case
* Fix python unit test failures due to flake8 update on pip
Problem:
Fast histogram updater crashes whenever subsampling picks zero rows
Diagnosis:
Row set data structure uses "nullptr" internally to indicate a non-existent
row set. Since you cannot take the address of the first element of an empty
vector, a valid row set ends up getting "nullptr" as well.
Fix:
Use an arbitrary value (not equal to "nullptr") to bypass nullptr check.
* Only set OpenMP_CXX_FLAGS when OpenMP is found
I found this trying to get the Mac build working without OpenMP. Tips in
issue #2596 helped to point in the right direction.
* Revise check
* Trigger codecov
* Add SparkParallelismTracker to prevent job from hanging
* Code review comments
* Code Review Comments
* Fix unit tests
* Changes and unit test to catch the corner case.
* Update documentations
* Small improvements
* cancalAllJobs is problematic with scalatest. Remove it
* Code Review Comments
* Check number of executor cores beforehand, and throw exeception if any core is lost.
* Address CR Comments
* Add missing class
* Fix flaky unit test
* Address CR comments
* Remove redundant param for TaskFailedListener
* SHAP values for feature contributions
* Fix commenting error
* New polynomial time SHAP value estimation algorithm
* Update API to support SHAP values
* Fix merge conflicts with updates in master
* Correct submodule hashes
* Fix variable sized stack allocation
* Make lint happy
* Add docs
* Fix typo
* Adjust tolerances
* Remove unneeded def
* Fixed cpp test setup
* Updated R API and cleaned up
* Fixed test typo
* coding style update
Current coding style varies(for example: the mixed use of single quote and double quote), and it will be confusing, especially for new users.
This PR will try to follow proposal of PEP8, make the documents more readable.
* minor fix
* Allowed subsampling test from the training data frame/RDD
The implementation requires storing 1 - trainTestRatio points in memory
to make the sampling work.
An alternative approach would be to construct the full DMatrix and then
slice it deterministically into train/test. The peak memory consumption
of such scenario, however, is twice the dataset size.
* Removed duplication from 'XGBoost.train'
Scala callers can (and should) use names to supply a subset of
parameters. Method overloading is not required.
* Reuse XGBoost seed parameter to stabilize train/test splitting
* Added early stopping support to non-distributed XGBoost
Closes#1544
* Added early-stopping to distributed XGBoost
* Moved construction of 'watches' into a separate method
This commit also fixes the handling of 'baseMargin' which previously
was not added to the validation matrix.
* Addressed review comments
* [R] MSVC compatibility
* [GPU] allow seed in BernoulliRng up to size_t and scale to uint32_t
* R package build with cmake and CUDA
* R package CUDA build fixes and cleanups
* always export the R package native initialization routine on windows
* update the install instructions doc
* fix lint
* use static_cast directly to set BernoulliRng seed
* [R] demo for GPU accelerated algorithm
* tidy up the R package cmake stuff
* R pack cmake: installs main dependency packages if needed
* [R] version bump in DESCRIPTION
* update NEWS
* added short missing/sparse values explanations to FAQ
Current version of xgboost.readthedocs.io has a broken search box.
Enabling themes on ReadTheDocs is known to break the search function, as
reported in
[this document](https://github.com/rtfd/readthedocs.org/issues/1487). To get
around the bug, we replace the `searchtools.js` file with our custom version.
* Removal of redundant code/files.
* Removal of exact namespace in GPU plugin
* Revert double precision histograms to single precision for performance on Maxwell/Kepler
* Converted ml.dmlc.xgboost4j.LabeledPoint to Scala
This allows to easily integrate LabeledPoint with Spark DataFrame APIs,
which support encoding/decoding case classes out of the box. Alternative
solution would be to keep LabeledPoint in Java and make it a Bean by
generating boilerplate getters/setters. I have decided against that, even
thought the conversion in this PR implies a public API change.
I also had to remove the factory methods fromSparseVector and
fromDenseVector because a) they would need to be duplicated to support
overloaded calls with extra data (e.g. weight); and b) Scala would expose
them via mangled $.MODULE$ which looks ugly in Java.
Additionally, this commit makes it possible to switch to LabeledPoint in
all public APIs and effectively to pass initial margin/group as part of
the point. This seems to be the only reliable way of implementing distributed
learning with these data. Note that group size format used by single-node
XGBoost is not compatible with that scenario, since the partition split
could divide a group into two chunks.
* Switched to ml.dmlc.xgboost4j.LabeledPoint in RDD-based public APIs
Note that DataFrame-based and Flink APIs are not affected by this change.
* Removed baseMargin argument in favour of the LabeledPoint field
* Do a single pass over the partition in buildDistributedBoosters
Note that there is no formal guarantee that
val repartitioned = rdd.repartition(42)
repartitioned.zipPartitions(repartitioned.map(_ + 1)) { it1, it2, => ... }
would do a single shuffle, but in practice it seems to be always the case.
* Exposed baseMargin in DataFrame-based API
* Addressed review comments
* Pass baseMargin to XGBoost.trainWithDataFrame via params
* Reverted MLLabeledPoint in Spark APIs
As discussed, baseMargin would only be supported for DataFrame-based APIs.
* Cleaned up baseMargin tests
- Removed RDD-based test, since the option is no longer exposed via
public APIs
- Changed DataFrame-based one to check that adding a margin actually
affects the prediction
* Pleased Scalastyle
* Addressed more review comments
* Pleased scalastyle again
* Fixed XGBoost.fromBaseMarginsToArray
which always returned an array of NaNs even if base margin was not
specified. Surprisingly this only failed a few tests.
* repared serialization after update process; fixes#2545
* non-stratified folds in python could omit some data instances
* Makefile: fixes for older makes on windows; clean R-package too
* make cub to be a shallow submodule
* improve $(MAKE) recovery
* for MinGW, drop the 'lib' prefix from shared library name
* fix defines for 'g++ 4.8 or higher' to include g++ >= 5
* fix compile warnings
* [Appveyor] add MinGW with python; remove redundant jobs
* [Appveyor] also do python build for one of msvc jobs
* Deduplicated DataFrame creation in XGBoostDFSuite
* Extracted dermatology.data into MultiClassification
* Moved cache cleaning to SharedSparkContext
Cache files are prefixed with appName therefore this seems to be just the
place to delete them.
* Removed redundant JMatrix calls in xgboost4j-spark
* Slightly more readable buildDenseRDD in XGBoostGeneralSuite
* Generalized train/test DataFrame construction in XGBoostDFSuite
* Changed SharedSparkContext to setup a new context per-test
Hence the new name: PerTestSparkSession :)
* Fused Utils into PerTestSparkSession
* Whitespace fix in XGBoostDFSuite
* Ensure SparkSession is always eagerly created in PerTestSparkSession
* Renamed PerTestSparkSession->PerTest
because it was doing slightly more than creating/stopping the session.
Includes:
- Dockerfile changes
- Dockerfile clean up
- Fix execution privileges of files used from Dockerfile.
- New Dockerfile entrypoint to replace with_user script
- Defined a placeholders for CPU testing (script and Dockerfile)
- Jenkinsfile
- Jenkins file milestone defined
- Single source code checkout and propagation via stash/unstash
- Bash needs to be explicitly used in launching make build, since we need
access to environment
- Jenkinsfile build factory for cmake and make style of jobs
- Archivation of artifacts (*.so, *.whl, *.egg) produced by cmake build
Missing:
- CPU testing
- Python3 env build and testing
* [jvm-packages] Deduplicated train/test data access in tests
All datasets are now available via a unified API, e.g. Agaricus.test.
The only exception is the dermatology data which requires parsing a
CSV file.
* Inlined Utils.buildTrainingRDD
The default number of partitions for local mode is equal to the number
of available CPUs.
* Replaced dataset names with problem types
It has been reported that new parallel algorithm (#2493) results in excessive
message usage (see issue #2326). Until issues are resolved, XGBoost should use
the old parallel algorithm by default. The user would have to specify
`enable_feature_grouping=1` manually to enable the new algorithm.
* Patch to improve multithreaded performance scaling
Change parallel strategy for histogram construction.
Instead of partitioning data rows among multiple threads, partition feature
columns instead. Useful heuristics for assigning partitions have been adopted
from LightGBM project.
* Add missing header to satisfy MSVC
* Restore max_bin and related parameters to TrainParam
* Fix lint error
* inline functions do not require static keyword
* Feature grouping algorithm accepting FastHistParam
Feature grouping algorithm accepts many parameters (3+), and it gets annoying to
pass them one by one. Instead, simply pass the reference to FastHistParam. The
definition of FastHistParam has been moved to a separate header file to
accomodate this change.
Prior to this commit XGBoostModel.predict produced an RDD with
an array of predictions for each partition, effectively changing
the shape wrt the input RDD. A more natural contract for prediction
API is that given an RDD it returns a new RDD with the same number
of elements. This allows the users to easily match inputs with
predictions.
This commit removes one layer of nesting in XGBoostModel.predict output.
Even though the change is clearly non-backward compatible, I still
think it is well justified. See discussion in 06bd5dca for motivation.
* Disabled excessive Spark logging in tests
* Fixed a singature of XGBoostModel.predict
Prior to this commit XGBoostModel.predict produced an RDD with
an array of predictions for each partition, effectively changing
the shape wrt the input RDD. A more natural contract for prediction
API is that given an RDD it returns a new RDD with the same number
of elements. This allows the users to easily match inputs with
predictions.
This commit removes one layer of nesting in XGBoostModel.predict output.
Even though the change is clearly non-backward compatible, I still
think it is well justified.
* Removed boxing in XGBoost.fromDenseToSparseLabeledPoints
* Inlined XGBoost.repartitionData
An if is more explicit than an opaque method name.
* Moved XGBoost.convertBoosterToXGBoostModel to XGBoostModel
* Check the input dimension in DMatrix.setBaseMargin
Prior to this commit providing an array of incorrect dimensions would
have resulted in memory corruption. Maybe backport this to C++?
* Reduced nesting in XGBoost.buildDistributedBoosters
* Ensured consistent naming of the params map
* Cleaned up DataBatch to make it easier to comprehend
* Made scalastyle happy
* Added baseMargin to XGBoost.train and trainWithRDD
* Deprecated XGBoost.train
It is ambiguous and work only for RDDs.
* Addressed review comments
* Revert "Fixed a singature of XGBoostModel.predict"
This reverts commit 06bd5dcae7780265dd57e93ed7d4135f4e78f9b4.
* Addressed more review comments
* Fixed NullPointerException in buildDistributedBoosters
* Fixed DLL name on Windows in ``xgboost.libpath``
* Added support for OS X to ``xgboost.libpath``
* Use .dylib for shared library on OS X
This does not affect the JNI library, because it is not trully
cross-platform in the Makefile-build anyway.
* Exposed prediction feature contribution on the Java side
* was not supplying the newly added argument
* Exposed from Scala-side as well
* formatting (keep declaration in one line unless exceeding 100 chars)
* [jvm-packages] Ensure the native library is loaded once
Previously any class using XGBoostJNI queried NativeLibLoader to make
sure the native library is loaded. This commit moves the initXGBoost
call to XGBoostJNI, effectively delegating the initialization to the class
loader.
Note also, that now XGBoostJNI would NOT suppress an IOException if it
occured in initXGBoost.
* [jvm-packages] Fused JNIErrorHandle with XGBoostJNI
There was no reason for having a separate class.
When using xgboost4j-spark I had executors getting killed much more
often than i would expect by yarn for overrunning their memory limits,
based on the memoryOverhead provided. It looks like a significant
amount of this is because dmatrix's were being created but not released,
because they were only released when the GC decided it was time to
cleanup the references.
Rather than waiting for the GC, relesae the DMatrix's when we know
they are no longer necessary.
* [jvm-packages] Fixed compilation on Windows
* [jvm-packages] Build the JNI bindings on Appveyor
* [jvm-packages] Build & test on OS X
* [jvm-packages] Re-applied the CMake build changes reverted by #2395
* Fixed Appveyor JVM build
* Muted Maven on Travis
* Don't link with libawt
* "linux2"->"linux"
Python2.x and 3.X use slightly different values for ``sys.platform``.
* Support for builing gpu-plugins to specific GPU architectures
1. Option GPU_COMPUTE_VER exposed from both Makefile and CMakeLists.txt
2. updater_gpu documentation updated accordingly
* Re-introduced GPU_COMPUTE_VER option in the cmake flow.
This seems to fix the compile-time, rdc=true and copy-constructor related
errors seen and discussed in PR #2390.
* [jvm-packages] Fixed JNI_OnLoad overload
It does not compile on Windows without proper export flags.
* [jvm-packages] Use JNI types directly where appropriate
* Removed lib hack from CMake build
Prior to this commit the CMake build use hardcoded lib prefix for
libxgboost and libxgboost4j. Unfortunatelly this did not play well with
Windows, which does not use the lib- prefix.
* [jvm-packages] Replaced create_jni.{bat,sh} with a Python version
This allows to have a single script for all platforms.
* [jvm-packages] Added all configuration options to create_jni.py
Use int32_t explicitly when serializing version field of dmatrix in binary
format. On ILP64 architectures, although very little, size of int is 64 bits.
* Integrating a faster version of grow_gpu plugin
1. Removed the older files to reduce duplication
2. Moved all of the grow_gpu files under 'exact' folder
3. All of them are inside 'exact' namespace to avoid any conflicts
4. Fixed a bug in benchmark.py while running only 'grow_gpu' plugin
5. Added cub and googletest submodules to ease integration and unit-testing
6. Updates to CMakeLists.txt to directly build cuda objects into libxgboost
* Added support for building gpu plugins through make flow
1. updated makefile and config.mk to add right targets
2. added unit-tests for gpu exact plugin code
* 1. Added support for building gpu plugin using 'make' flow as well
2. Updated instructions for building and testing gpu plugin
* Fix travis-ci errors for PR#2360
1. lint errors on unit-tests
2. removed googletest, instead depended upon dmlc-core provide gtest cache
* Some more fixes to travis-ci lint failures PR#2360
* Added Rory's copyrights to the files containing code from both.
* updated copyright statement as per Rory's request
* moved the static datasets into a script to generate them at runtime
* 1. memory usage print when silent=0
2. tests/ and test/ folder organization
3. removal of the dependency of googletest for just building xgboost
4. coding style updates for .cuh as well
* Fixes for compilation warnings
* add cuda object files as well when JVM_BINDINGS=ON
* [jvm-packages] Added libxgboost4j to CMake build
* [jvm-packages] Wired CMake build into create_jni.sh
* User newer CMake version on Travis
* Lowered CMake version constraints
* Fixed various quirks in the new CMake build
Don't use implicit conversions to c_int, which incidentally happen to work
on (some) 64-bit platforms, but:
* may lead to truncation of the input value to a 32-bit signed int,
* cause segfaults on some 32-bit architectures (tested on Ubuntu ARM,
but is also the likely cause of issue #1707).
Also, when passing references use explicit 64-bit integers, where needed,
instead of c_ulong, which is not guaranteed to be this large.
* Specified 'exec-maven-plugin' version
* Changed 'create_jni.sh' to fail on error
and also report each of the executed commands, which makes it easier
to debug.
for loop in create.new.tree.features was referencing length(trees) as the upper bound of the loop. trees is a base R dataset and not the model that the code is generating. Changed loop boundary to model$niter which should be the number of trees.
* Added kwargs support for Sklearn API
* Updated NEWS and CONTRIBUTORS
* Fixed CONTRIBUTORS.md
* Added clarification of **kwargs and test for proper usage
* Fixed lint error
* Fixed more lint errors and clf assigned but never used
* Fixed more lint errors
* Fixed more lint errors
* Fixed issue with changes from different branch bleeding over
* Fixed issue with changes from other branch bleeding over
* Added note that kwargs may not be compatible with Sklearn
* Fixed linting on kwargs note
* Added n_jobs and random_state to keep up to date with sklearn API.
Deprecated nthread and seed. Added tests for new params and
deprecations.
* Fixed docstring to reflect updates to n_jobs and random_state.
* Fixed whitespace issues and removed nose import.
* Added deprecation note for nthread and seed in docstring.
* Attempted fix of deprecation tests.
* Second attempted fix to tests.
* Set n_jobs to 1.
* [gblinear] add features contribution prediction; fix DumpModel bug
* [gbtree] minor changes to PredContrib
* [R] add feature contribution prediction to R
* [R] bump up version; update NEWS
* [gblinear] fix the base_margin issue; fixes#1969
* [R] list of matrices as output of multiclass feature contributions
* [gblinear] make order of DumpModel coefficients consistent: group index changes the fastest
* Fix compilation on OS X with GCC 7
Compilation failed with
In file included from src/tree/tree_updater.cc:6:0:
include/xgboost/tree_updater.h:75:46: error: 'function' is not a member of 'std'
std::function<TreeUpdater* ()> > {
caused by a missing <functional> include.
* Fixed another occurence of that issue spotted by @ClimberPG
* Add option to choose booster in scikit intreface (gbtree by default)
* Add option to choose booster in scikit intreface: complete docstring.
* Fix XGBClassifier to work with booster option
* Added test case for gblinear booster
* [R] add native routines registration
* c_api.h needs to include <cstdint> since it uses fixed width integer types
* [R] use registered native routines from R code
* [R] bump version; add info on native routine registration to the contributors guide
* make lint happy
* Add prediction of feature contributions
This implements the idea described at http://blog.datadive.net/interpreting-random-forests/
which tries to give insight in how a prediction is composed of its feature contributions
and a bias.
* Support multi-class models
* Calculate learning_rate per-tree instead of using the one from the first tree
* Do not rely on node.base_weight * learning_rate having the same value as the node mean value (aka leaf value, if it were a leaf); instead calculate them (lazily) on-the-fly
* Add simple test for contributions feature
* Check against param.num_nodes instead of checking for non-zero length
* Loop over all roots instead of only the first
* add back train method but mark as deprecated
* fix scalastyle error
* fix the persistence of XGBoostEstimator
* test persistence of a complete pipeline
* fix compilation issue
* do not allow persist custom_eval and custom_obj
* fix the failed tesl
* [R] make sure things work for a single split model; fixes#2191
* [R] add option use_int_id to xgb.model.dt.tree
* [R] add example of exporting tree plot to a file
* [R] set save_period = NULL as default in xgboost() to be the same as in xgb.train; fixes#2182
* [R] it's a good practice after CRAN releases to bump up package version in dev
* [R] allow xgb.DMatrix construction from integer dense matrices
* [R] xgb.DMatrix: silent parameter; improve documentation
* [R] xgb.model.dt.tree code style changes
* [R] update NEWS with parameter changes
* [R] code safety & style; handle non-strict matrix and inherited classes of input and model; fixes#2242
* [R] change to x.y.z.p R-package versioning scheme and set version to 0.6.4.3
* [R] add an R package versioning section to the contributors guide
* [R] R-package/README.md: clean up the redundant old installation instructions, link the contributors guide
Reported in issue #2165. Dynamic scheduling of OpenMP loops involve
implicit synchronization. To implement synchronization, libgomp uses futex
(fast userspace mutex), whereas MinGW uses kernel-space mutex, which is more
costly. With chunk size of 1, synchronization overhead may become prohibitive
on Windows machines.
Solution: use 'guided' schedule to minimize the number of syncs
Storing and then loading a model loses any eval_metric that was
provided. This causes implementations that always store/load, like
xgboost4j-spark, to be unable to eval with the desired metric.
This log appears to fire every time I ask the python package to make a prediction. It's the only log that fires from XGBoost. When we're getting predictions on millions of items a day in production, this log seems out of place.
* add back train method but mark as deprecated
* fix scalastyle error
* change class to object in examples
* fix compilation error
* small fix for cleanExternalCache
* add back train method but mark as deprecated
* fix scalastyle error
* change class to object in examples
* fix compilation error
* fix several issues in tests
* Bugfix 1: Fix segfault in multithreaded ApplySplitSparseData()
When there are more threads than rows in rowset, some threads end up
with empty ranges, causing them to crash. (iend - 1 needs to be
accessible as part of algorithm)
Fix: run only those threads with nonempty ranges.
* Add regression test for Bugfix 1
* Moving python_omp_test to existing python test group
Turns out you don't need to set "OMP_NUM_THREADS" to enable
multithreading. Just add nthread parameter.
* Bugfix 2: Fix corner case of ApplySplitSparseData() for categorical feature
When split value is less than all cut points, split_cond is set
incorrectly.
Fix: set split_cond = -1 to indicate this scenario
* Bugfix 3: Initialize data layout indicator before using it
data_layout_ is accessed before being set; this variable determines
whether feature 0 is included in feat_set.
Fix: re-order code in InitData() to initialize data_layout_ first
* Adding regression test for Bugfix 2
Unfortunately, no regression test for Bugfix 3, as there is no
way to deterministically assign value to an uninitialized variable.
* Add UpdatePredictionCache() option to updaters
Some updaters (e.g. fast_hist) has enough information to quickly compute
prediction cache for the training data. Each updater may override
UpdaterPredictionCache() method to update the prediction cache. Note: this
trick does not apply to validation data.
* Respond to code review
* Disable some debug messages by default
* Document UpdatePredictionCache() interface
* Remove base_margin logic from UpdatePredictionCache() implementation
* Do not take pointer to cfg, as reference may get stale
* Improve multi-threaded performance
* Use columnwise accessor to accelerate ApplySplit() step,
with support for a compressed representation
* Parallel sort for evaluation step
* Inline BuildHist() function
* Cache gradient pairs when building histograms in BuildHist()
* Add missing #if macro
* Respond to code review
* Use wrapper to enable parallel sort on Linux
* Fix C++ compatibility issues
* MSVC doesn't support unsigned in OpenMP loops
* gcc 4.6 doesn't support using keyword
* Fix lint issues
* Respond to code review
* Fix bug in ApplySplitSparseData()
* Attempting to read beyond the end of a sparse column
* Mishandling the case where an entire range of rows have missing values
* Fix training continuation bug
Disable UpdatePredictionCache() in the first iteration. This way, we can
accomodate the scenario where we build off of an existing (nonempty) ensemble.
* Add regression test for fast_hist
* Respond to code review
* Add back old version of ApplySplitSparseData
* Updated sklearn_parallel.py for soon-to-be-deprecated modules
* Updated predict_leaf_indices.py; Use python3 print() as other exmaples and removed unused module
This commit proposes a simpler single compiler specification for OSX and *nix. It also let's people override the setting on both systems, not just *nix.
I use the online prediction function(`inline void Predict(const SparseBatch::Inst &inst, ... ) const;`), the results obtained are different from the results of the batch prediction function(` virtual void Predict(DMatrix* data, ...) const = 0`). After the investigation found that the online prediction function using the `base_score_` parameters, and the batch prediction function is not used in this parameter. It is found that the `base_score_` values are different when the same model file is loaded many times.
```
1st times：base_score_: 6.69023e-21
2nd times：base_score_: -3.7668e+19
3rd times：base_score_: 5.40507e+07
```
Online prediction results are affected by `base_score_` parameters. After deleting the if condition(`if (out_preds->size() == 1)`) , the online prediction is consistent with the batch prediction results, and the xgboost prediction results are consistent with python version. Therefore, it is likely that the online prediction function is bug
* [jvm-packages] call setGroup for ranking task
* passing groupData through xgBoostConfMap
* fix original comment position
* make groupData param
* remove groupData variable, use xgBoostConfMap directly
* set default groupData value
* add use groupData tests
* reduce rank-demo size
* use TaskContext.getPartitionId() instead of mapPartitionsWithIndex
* add DF use groupData test
* remove unused varable
* add back train method but mark as deprecated
* fix scalastyle error
* first commit in scala binding for fast histo
* java test
* add missed scala tests
* spark training
* add back train method but mark as deprecated
* fix scalastyle error
* local change
* first commit in scala binding for fast histo
* local change
* fix df frame test
* add back train method but mark as deprecated
* fix scalastyle error
* change class to object in examples
* fix compilation error
* bump spark version to 2.1
* preserve num_class issues
* fix failed test cases
* rivising
* add multi class test
verbose_eval docs claim it will log the last iteration (http://xgboost.readthedocs.io/en/latest/python/python_api.html#xgboost.train). this is also consistent w/the behavior from 0.4. not a huge deal but I found it handy to see the last iter's result b/c my period is usually large.
this doesn't address logging the last stage found by early_stopping (as noted in docs) as I'm not sure how to do that.
* [jvm-packages] Scala implementation of the Rabit tracker.
A Scala implementation of RabitTracker that is interface-interchangable with the
Java implementation, ported from `tracker.py` in the
[dmlc-core project](https://github.com/dmlc/dmlc-core).
* [jvm-packages] Updated Akka dependency in pom.xml.
* Refactored the RabitTracker directory structure.
* Fixed premature stopping of connection handler.
Added a new finite state "AwaitingPortNumber" to explicitly wait for the
worker to send the port, and close the connection. Stopping the actor
prematurely sends a TCP RST to the worker, causing the worker to crash
on AssertionError.
* Added interface IRabitTracker so that user can switch implementations.
* Default timeout duration changes.
* Dependency for Akka tests.
* Removed the main function of RabitTracker.
* A skeleton for testing Akka-based Rabit tracker.
* waitFor() in RabitTracker no longer throws exceptions.
* Completed unit test for the 'start' command of Rabit tracker.
* Preliminary support for Rabit Allreduce via JNI (no prepare function support yet.)
* Fixed the default timeout duration.
* Use Java container to avoid serialization issues due to intermediate wrappers.
* Added tests for Allreduce/model training using Scala Rabit tracker.
* Added spill-over unit test for the Scala Rabit tracker.
* Fixed a typo.
* Overhaul of RabitTracker interface per code review.
- Removed methods start() waitFor() (no arguments) from IRabitTracker.
- The timeout in start(timeout) is now worker connection timeout, as tcp
socket binding timeout is less intuitive.
- Dropped time unit from start(...) and waitFor(...) methods; the default
time unit is millisecond.
- Moved random port number generation into the RabitTrackerHandler.
- Moved all Rabit-related classes to package ml.dmlc.xgboost4j.scala.rabit.
* More code refactoring and comments.
* Unified timeout constants. Readable tracker status code.
* Add comments to indicate that allReduce is for tests only. Removed all other variants.
* Removed unused imports.
* Simplified signatures of training methods.
- Moved TrackerConf into parameter map.
- Changed GeneralParams so that TrackerConf becomes a standalone parameter.
- Updated test cases accordingly.
* Changed monitoring strategies.
* Reverted monitoring changes.
* Update test case for Rabit AllReduce.
* Mix in UncaughtExceptionHandler into IRabitTracker to prevent tracker from hanging due to exceptions thrown by workers.
* More comprehensive test cases for exception handling and worker connection timeout.
* Handle executor loss due to unknown cause: the newly spawned executor will attempt to connect to the tracker. Interrupt tracker in such case.
* Per code-review, removed training timeout from TrackerConf. Timeout logic must be implemented explicitly and externally in the driver code.
* Reverted scalastyle-config changes.
* Visibility scope change. Interface tweaks.
* Use match pattern to handle tracker_conf parameter.
* Minor clarification in JNI code.
* Clearer intent in match pattern to suppress warnings.
* Removed Future from constructor. Block in start() and waitFor() instead.
* Revert inadvertent comment changes.
* Removed debugging information.
* Updated test cases that are a bit finicky.
* Added comments on the reasoning behind the unit tests for testing Rabit tracker robustness.
* Fixed BufferUnderFlow bug in decoding tracker 'print' command.
* Merge conflicts resolution.
* A fix regarding the compatibility with python 2.6
the syntax of {n: self.attr(n) for n in attr_names} is illegal in python 2.6
* Update core.py
add a space after comma
As discussed in issue #1978, tree_method=hist ignores the parameter
param.num_roots; it simply assumes that the tree has only one root. In
particular, when InitData() method initializes row_set_collection_, it simply
assigns all rows to node 0, the value that's hard-coded.
For now, the updater will simply fail when num_roots exceeds 1. I will revise
the updater soon to support multiple roots.
* [R] xgb.save must work when handle in nil but raw exists
* [R] print.xgb.Booster should still print other info when handle is nil
* [R] rename internal function xgb.Booster to xgb.Booster.handle to make its intent clear
* [R] rename xgb.Booster.check to xgb.Booster.complete and make it visible; more docs
* [R] storing evaluation_log should depend only on watchlist, not on verbose
* [R] reduce the excessive chattiness of unit tests
* [R] only disable some tests in windows when it's not 64-bit
* [R] clean-up xgb.DMatrix
* [R] test xgb.DMatrix loading from libsvm text file
* [R] store feature_names in xgb.Booster, use them from utility functions
* [R] remove non-functional co-occurence computation from xgb.importance
* [R] verbose=0 is enough without a callback
* [R] added forgotten xgb.Booster.complete.Rd; cran check fixes
* [R] update installation instructions
* added the max_features parameter to the plot_importance function.
* renamed max_features parameter to max_num_features for better understanding
* removed unwanted character in docstring
* Support histogram-based algorithm + multiple tree growing strategy
* Add a brand new updater to support histogram-based algorithm, which buckets
continuous features into discrete bins to speed up training. To use it, set
`tree_method = fast_hist` to configuration.
* Support multiple tree growing strategies. For now, two policies are supported:
* `grow_policy=depthwise` (default): favor splitting at nodes closest to the
root, i.e. grow depth-wise.
* `grow_policy=lossguide`: favor splitting at nodes with highest loss change
* Improve single-threaded performance
* Unroll critical loops
* Introduce specialized code for dense data (i.e. no missing values)
* Additional training parameters: `max_leaves`, `max_bin`, `grow_policy`, `verbose`
* Adding a small test for hist method
* Fix memory error in row_set.h
When std::vector is resized, a reference to one of its element may become
stale. Any such reference must be updated as well.
* Resolve cross-platform compilation issues
* Versions of g++ older than 4.8 lacks support for a few C++11 features, e.g.
alignas(*) and new initializer syntax. To support g++ 4.6, use pre-C++11
initializer and remove alignas(*).
* Versions of MSVC older than 2015 does not support alignas(*). To support
MSVC 2012, remove alignas(*).
* For g++ 4.8 and newer, alignas(*) is enabled for performance benefits.
* Some old compilers (MSVC 2012, g++ 4.6) do not support template aliases
(which uses `using` to declate type aliases). So always use `typedef`.
* Fix a host of CI issues
* Remove dependency for libz on osx
* Fix heading for hist_util
* Fix minor style issues
* Add missing #include
* Remove extraneous logging
* Enable tree_method=hist in R
* Renaming HistMaker to GHistBuilder to avoid confusion
* Fix R integration
* Respond to style comments
* Consistent tie-breaking for priority queue using timestamps
* Last-minute style fixes
* Fix issuecomment-271977647
The way we quantize data is broken. The agaricus data consists of all
categorical values. When NAs are converted into 0's,
`HistCutMatrix::Init` assign both 0's and 1's to the same single bin.
Why? gmat only the smallest value (0) and an upper bound (2), which is twice
the maximum value (1). Add the maximum value itself to gmat to fix the issue.
* Fix issuecomment-272266358
* Remove padding from cut values for the continuous case
* For categorical/ordinal values, use midpoints as bin boundaries to be safe
* Fix CI issue -- do not use xrange(*)
* Fix corner case in quantile sketch
Signed-off-by: Philip Cho <chohyu01@cs.washington.edu>
* Adding a test for an edge case in quantile sketcher
max_bin=2 used to cause an exception.
* Fix fast_hist test
The test used to require a strictly increasing Test AUC for all examples.
One of them exhibits a small blip in Test AUC before achieving a Test AUC
of 1. (See bottom.)
Solution: do not require monotonic increase for this particular example.
[0] train-auc:0.99989 test-auc:0.999497
[1] train-auc:1 test-auc:0.999749
[2] train-auc:1 test-auc:0.999749
[3] train-auc:1 test-auc:0.999749
[4] train-auc:1 test-auc:0.999749
[5] train-auc:1 test-auc:0.999497
[6] train-auc:1 test-auc:1
[7] train-auc:1 test-auc:1
[8] train-auc:1 test-auc:1
[9] train-auc:1 test-auc:1
* option to shuffle data in mknfolds
* removed possibility to run as stand alone test
* split function def in 2 lines for lint
* option to shuffle data in mknfolds
* removed possibility to run as stand alone test
* split function def in 2 lines for lint
* fix cran check
* change required R version because of utils::globalVariables
* temporary commit, monotone not working
* fix test
* fix doc
* fix doc
* fix cran note and warning
* improve checks
* fix urls
* fix cran check
* add cleanup and bump up version number
* use clean in build
* Update Makefile
* [R-package] JSON tree dump interface
* [R-package] precision bugfix in xgb.attributes
* [R-package] bugfix for cb.early.stop called from xgb.cv
* [R-package] a bit more clarity on labels checking in xgb.cv
* [R-package] test JSON dump for gblinear as well
* whitespace lint
* [jvm-packages] Scala implementation of the Rabit tracker.
A Scala implementation of RabitTracker that is interface-interchangable with the
Java implementation, ported from `tracker.py` in the
[dmlc-core project](https://github.com/dmlc/dmlc-core).
* [jvm-packages] Updated Akka dependency in pom.xml.
* Refactored the RabitTracker directory structure.
* Fixed premature stopping of connection handler.
Added a new finite state "AwaitingPortNumber" to explicitly wait for the
worker to send the port, and close the connection. Stopping the actor
prematurely sends a TCP RST to the worker, causing the worker to crash
on AssertionError.
* Added interface IRabitTracker so that user can switch implementations.
* Default timeout duration changes.
* Dependency for Akka tests.
* Removed the main function of RabitTracker.
* A skeleton for testing Akka-based Rabit tracker.
* waitFor() in RabitTracker no longer throws exceptions.
* Completed unit test for the 'start' command of Rabit tracker.
* Preliminary support for Rabit Allreduce via JNI (no prepare function support yet.)
* Fixed the default timeout duration.
* Use Java container to avoid serialization issues due to intermediate wrappers.
* Added tests for Allreduce/model training using Scala Rabit tracker.
* Added spill-over unit test for the Scala Rabit tracker.
* Fixed a typo.
* Overhaul of RabitTracker interface per code review.
- Removed methods start() waitFor() (no arguments) from IRabitTracker.
- The timeout in start(timeout) is now worker connection timeout, as tcp
socket binding timeout is less intuitive.
- Dropped time unit from start(...) and waitFor(...) methods; the default
time unit is millisecond.
- Moved random port number generation into the RabitTrackerHandler.
- Moved all Rabit-related classes to package ml.dmlc.xgboost4j.scala.rabit.
* More code refactoring and comments.
* Unified timeout constants. Readable tracker status code.
* Add comments to indicate that allReduce is for tests only. Removed all other variants.
* Removed unused imports.
* Simplified signatures of training methods.
- Moved TrackerConf into parameter map.
- Changed GeneralParams so that TrackerConf becomes a standalone parameter.
- Updated test cases accordingly.
* Changed monitoring strategies.
* Reverted monitoring changes.
* Update test case for Rabit AllReduce.
* Mix in UncaughtExceptionHandler into IRabitTracker to prevent tracker from hanging due to exceptions thrown by workers.
* More comprehensive test cases for exception handling and worker connection timeout.
* Handle executor loss due to unknown cause: the newly spawned executor will attempt to connect to the tracker. Interrupt tracker in such case.
* Per code-review, removed training timeout from TrackerConf. Timeout logic must be implemented explicitly and externally in the driver code.
* Reverted scalastyle-config changes.
* Visibility scope change. Interface tweaks.
* Use match pattern to handle tracker_conf parameter.
* Minor clarification in JNI code.
* Clearer intent in match pattern to suppress warnings.
* Removed Future from constructor. Block in start() and waitFor() instead.
* Revert inadvertent comment changes.
* Removed debugging information.
* Updated test cases that are a bit finicky.
* Added comments on the reasoning behind the unit tests for testing Rabit tracker robustness.
The GetWeight is a wrapper which sets the correct weight
if the weights vector is not provided. Hence accessing the default
weights vector is not recommended.
Update the code coverage of the project on codecov for easy viewing.
Also the gcov on travis uses a different version which cannot
find the directory of the given files, and it needs to be specified
in the -o flag. Hence now we loop over the list of files and
run them independently.
* [CORE] allow updating trees in an existing model
* [CORE] in refresh updater, allow keeping old leaf values and update stats only
* [R-package] xgb.train mod to allow updating trees in an existing model
* [R-package] added check for nrounds when is_update
* [CORE] merge parameter declaration changes; unify their code style
* [CORE] move the update-process trees initialization to Configure; rename default process_type to 'default'; fix the trees and trees_to_update sizes comparison check
* [R-package] unit tests for the update process type
* [DOC] documentation for process_type parameter; improved docs for updater, Gamma and Tweedie; added some parameter aliases; metrics indentation and some were non-documented
* fix my sloppy merge conflict resolutions
* [CORE] add a TreeProcessType enum
* whitespace fix
* fix cran check
* change required R version because of utils::globalVariables
* temporary commit, monotone not working
* fix test
* fix doc
* fix doc
* fix cran note and warning
* improve checks
* fix urls
* Fix various typos
* Add override to functions that are overridden
gcc gives warnings about functions that are being overridden by not
being marked as oveirridden. This fixes it.
* Use bst_float consistently
Use bst_float for all the variables that involve weight,
leaf value, gradient, hessian, gain, loss_chg, predictions,
base_margin, feature values.
In some cases, when due to additions and so on the value can
take a larger value, double is used.
This ensures that type conversions are minimal and reduces loss of
precision.
* Allow using learning_rates parameter when doing CV
- Create a new `callback_cv` method working when called from `xgb.cv()`
- Rename existing `callback` into `callback_train` and make it the default callback
- Get the logic out of the callbacks and place it into a common helper
* Add a learning_rates parameter to cv()
* lint
* remove caller explicit reference
* callback is aware of its calling context
* remove caller argument
* remove learning_rates param
* restore learning_rates for training, but deprecated
* lint
* lint line too long
* quick example for predefined callbacks
In ecb3a271be the silent argument
in XGDMatrixCreateFromFile of c_api.cc was always overridden to
be false. This disabled the functionality to hide log messages.
This commit reverts that part to enable the hiding of log messages.
* add back train method but mark as deprecated
* fix scalastyle error
* change class to object in examples
* fix compilation error
* update methods in test cases to be consistent
* add blank lines
* fix
On Unix systems, it's common for programs to read their input from stdin, and
write their output to stdout. Messages should be written to stderr, where they
won't corrupt a program's output, and where they can be seen by the user even
if the output is being redirected.
This is mostly a problem when XGBoost is being used from Python or from another
program.
* add support for tweedie regression
* added back readme line that was accidentally deleted
* fixed linting errors
* add support for tweedie regression
* added back readme line that was accidentally deleted
* fixed linting errors
* rebased with upstream master and added R example
* changed parameter name to tweedie_variance_power
* linting error fix
* refactored tweedie-nloglik metric to be more like the other parameterized metrics
* added upper and lower bound check to tweedie metric
* add support for tweedie regression
* added back readme line that was accidentally deleted
* fixed linting errors
* added upper and lower bound check to tweedie metric
* added back readme line that was accidentally deleted
* rebased with upstream master and added R example
* rebased again on top of upstream master
* linting error fix
* added upper and lower bound check to tweedie metric
* rebased with master
* lint fix
* removed whitespace at end of line 186 - elementwise_metric.cc
* Fix typos and messages in docs
* parameter.md: Add docs for updater_seq
Mention the updater_seq parameter which sets the order of the tree
updaters to run and also specifies which ones to run. This can be
useful when pruning is not required or even a custom plugin is
being built along with xgboost.
* Add format to the params accepted by DumpModel
Currently, only the test format is supported when trying to dump
a model. The plan is to add more such formats like JSON which are
easy to read and/or parse by machines. And to make the interface
for this even more generic to allow other formats to be added.
Hence, we make some modifications to make these function generic
and accept a new parameter "format" which signifies the format of
the dump to be created.
* Fix typos and errors in docs
* plugin: Mention all the register macros available
Document the register macros currently available to the plugin
writers so they know what exactly can be extended using hooks.
* sparce_page_source: Use same arg name in .h and .cc
* gbm: Add JSON dump
The dump_format argument can be used to specify what type
of dump file should be created. Add functionality to dump
gblinear and gbtree into a JSON file.
The JSON file has an array, each item is a JSON object for the tree.
For gblinear:
- The item is the bias and weights vectors
For gbtree:
- The item is the root node. The root node has a attribute "children"
which holds the children nodes. This happens recursively.
* core.py: Add arg dump_format for get_dump()
The `cd ..;` in the one liner takes you up a directory instead of into the xgboost directory. This will cause that step of the installation to fail. It seems like you are meant to enter the xgboost directory as you did in the instructions for installing xgboost without openmp.
* add back train method but mark as deprecated
* fix scalastyle error
* change class to object in examples
* fix compilation error
* fix mis configuration
* make DMatrix._init_from_npy2d only copy data when necessary
When creating DMatrix from a 2d ndarray, it can unnecessarily copy the input data. This can be problematic when the data is already very large--running out of memory. The copy is temporary (going out of scope at the end of this function) but it still adds to peak memory usage.
``numpy.array`` copies its input no matter what by default. By adding ``copy=False``, it will only do so when necessary. Since XGDMatrixCreateFromMat is readonly on the input buffer, this copy is not needed.
Also added comments explaining when a copy can happen (if data ordering/layout is wrong or if type is not 32-bit float).
* remove whitespace
* correct CalcDCG in rank_metric.cc
DCG use log base-2, however `std::log` returns log base-e.
* correct CalcDCG in rank_obj.cc
DCG use log base-2, however `std::log` returns log base-e.
* use std::log2 instead of std::log
make it more elegant
* use std::log2 instead of std::log
make it more elegant
*Fix 1439
*Fix python_wrapper when eval set name contain '-' will cause early_stop maximize variable con't set to True propely
Change-Id: Ib0595afd4ae7b445a84c00a3a8faeccc506c6d13
* Changes for Mingw64 compilation to ensure long is a consistent size.
Mainly impacts the Java API which would not compile, but there may be
silent errors on Windows with large datasets before this patch (as long
is 32-bits when compiled with mingw64 even in 64-bit mode).
* Adding ifdefs to ensure it still compiles on MacOS
* Makefile and create_jni.bat changes for Windows.
* Switching XGDMatrixCreateFromCSREx JNI call to use size_t cast
* Fixing lint error, adding profile switching to jvm-packages build to make create-jni.bat get called, adding myself to Contributors.Md
add_library(libxgboost SHARED ${SOURCES}) builds a library named
liblibxgboost.so; However, simply changing it to add_library(xgboost ...)
won't work, as add_executable(xgboost ...) and add_library(xgbboost ...)
will then have the same target name. This patch correctly handles the
same-name situation through SET_TARGET_PROPERTIES.
* add scikit-learn v0.18 compatibility
import KFold & StratifiedKFold from sklearn.model_selection instead of sklearn.cross_validation
* change DeprecationWarning to ImportError
DeprecationWarning isn't an exception, so it should work the other way around.
ml.dmlc.xgboost4j.scala.spark.XGBoost.scala:51
values is empty when we meet it at first time, so values(0) throw an IndexOutOfBoundsException.
It should be dVector.values(i) instead of values(i).
* bump up to scala 2.11
* framework of data frame integration
* test consistency between RDD and DataFrame
* order preservation
* test order preservation
* example code and fix makefile
* improve type checking
* improve APIs
* user docs
* work around travis CI's limitation on log length
* adjust test structure
* integrate with Spark -1 .x
* spark 2.x integration
* remove spark 1.x implementation but provide instructions on how to downgrade
* [TREE] Experimental version of monotone constraint
* Allow default detection of montone option
* loose the condition of strict check
* Update gbtree.cc
Fixed to work with future versions of visual studio i.e., 2015
MSVC has it's own section for setting compile parameters, it shouldn't need to fall into section below i.e., checking for c++11 as this is definitely already supported, though this isn't an issue for Visual Studio 2012, it breaks for later versions
of visual studio i.e., 2015 when the default c++ is version 14. Though still backward compatible with c++11
* test consistency of prediction functions between DMatrix and RDD
* remove APIs with DMatrix from xgboost-spark
* fix compilation error in xgboost4j-example
* fix test cases
Currently xgboost can only be installed by running:
python setup.py install
Now it can be packaged (in binary form) as a wheel and installed like:
pip install xgboost-0.6-py2-none-any.whl
distutils and wheel install `data_files` differently than setuptools.
setuptools will install the `data_files` in the package directory whereas the
others install it in `sys.prefix`. By adding `sys.prefix` to the list of
directories to check for the shared library, xgboost can now be distributed as
a wheel.
* Fixed OpenMP installation on MacOSX with gcc-6
- Modified makefile from gcc-5 to gcc-6
- Removed deprecated install instructions from doc (gcc-5 was automatically forced if available in makefile on OSX)
* Fixed OpenMP installation on MacOSX with gcc-6
- Modified makefile from gcc-5 to gcc-6
- Removed deprecated install instructions from doc (gcc-5 was automatically forced if available in makefile on OSX)
make math better, specifically, unify the notation for Theta or theta. changed basic linear model notation from weight w to theta to make more consistent. Changed Obj function notation also
* force gcc-5 or clang-omp for Mac OS, prepare for pip pack
* add sklearn dep, make -j4
* finalize PyPI submission
* revert to Xcode clang for passing build #1468
* force to clang, try to solve cmake travis error
* remove sklearn dependency
* [R] do not remove zero coefficients from gblinear dump
* [R] switch from stringr to stringi
* fix#1399
* [R] separate ggplot backend, add base r graphics, cleanup, more plots, tests
* add missing include in amalgamation - fixes building R package in linux
* add forgotten file
* [R] fix DESCRIPTION
* [R] fix travis check issue and some cleanup
* Add deviance metric for gamma regression
* Simplify the computation of nloglik for gamma regression
* Add a description for gamma-deviance
* Minor fix
* Add support for Gamma regression
* Use base_score to replace the lp_bias
* Remove the lp_bias config block
* Add a demo for running gamma regression in Python
* Typo fix
* Revise the description for objective
* Add a script to generate the autoclaims dataset
* create dmatrix with specified missing value
* update dmlc-core
* support for predict method in spark package
repartitioning
work around
* add more elements to work around training set empty partition issue
* added new function to calculate other feature importances
* added capability to plot other feature importance measures
* changed plotting default to fscore
* added info on importance_type to boilerplate comment
* updated text of error statement
* added self module name to fix call
* added unit test for feature importances
* style fixes
This error message can be hard to understand when there are several fields, as shown in the example below. This improves the error message, letting the user know which fields were unexpected or missing.
import xgboost as xgb
import pandas as pd
train = pd.DataFrame({'a':[1], 'b':[2], 'c':[3], 'd':[4], 'f':[2], 'g':2, 'etc etc etc':[11]})
dtrain = xgb.DMatrix(train.drop('d', axis=1), train.d)
test = pd.DataFrame({'a':[1], 'b':[2], 'c':[1], 'd':[4], 'e':[2], 'f':[2], 'g':2, 'etc etc etc':[11]})
dtest = xgb.DMatrix(test)
modl = xgb.train({}, dtrain)
modl.predict(dtest)
# ValueError: feature_names mismatch: [u'a', u'b', u'c', u'etc etc etc', u'f', u'g'] [u'a', u'b', u'c', u'd', u'e', u'etc etc etc', u'f', u'g']
Thanks for participating in the XGBoost community! We use https://discuss.xgboost.ai for any general usage questions and discussions. The issue tracker is used for actionable items such as feature proposals discussion, roadmaps, and bug tracking. You are always welcomed to post on the forum first :)
Issues that are inactive for a period of time may get closed. We adopt this policy so that we won't lose track of actionable issues that may fall at the bottom of the pile. Feel free to reopen a new one if you feel there is an additional problem that needs attention when an old one gets closed.
For bug reports, to help the developer act on the issues, please include a description of your environment, preferably a minimum script to reproduce the problem.
For feature proposals, list clear, small actionable items so we can track the progress of the change.
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 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.
* [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).
- 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.
Project Management Committee(PMC)
----------
The Project Management Committee(PMC) consists group of active committers that moderate the discussion, manage the project release, and proposes new committer/PMC members.
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.
* [Tianqi Chen](https://github.com/tqchen), University of Washington
- Tianqi is a Ph.D. student working on large-scale machine learning. He is the creator of the project.
- Michael is a lawyer and data scientist in France. He is the creator of XGBoost interactive analysis module in R.
* [Yuan Tang](https://github.com/terrytangyuan), Ant Financial
- Yuan is a software engineer in Ant Financial. He contributed mostly in R and Python packages.
* [Nan Zhu](https://github.com/CodingCat), Uber
- Nan is a software engineer in Uber. He contributed mostly in JVM packages.
* [Jiaming Yuan](https://github.com/trivialfis)
- Jiaming contributed to the GPU algorithms. He has also introduced new abstractions to improve the quality of the C++ codebase.
* [Hyunsu Cho](http://hyunsu-cho.io/), NVIDIA
- Hyunsu is the maintainer of the XGBoost Python package. He also manages the Jenkins continuous integration system (https://xgboost-ci.net/). He is the initial author of the CPU 'hist' updater.
* [Rory Mitchell](https://github.com/RAMitchell), University of Waikato
- Rory is a Ph.D. student at University of Waikato. He is the original creator of the GPU training algorithms. He improved the CMake build system and continuous integration.
* [Hongliang Liu](https://github.com/phunterlau)
Committers
----------
Committers are people who have made substantial contribution to the project and granted write access to the project.
* [Tong He](https://github.com/hetong007), Amazon AI
- Tong is an applied scientist in Amazon AI. He is the maintainer of XGBoost R package.
- Sergei is a software engineer in Criteo. He contributed mostly in JVM packages.
* [Scott Lundberg](http://scottlundberg.com/), University of Washington
- Scott is a Ph.D. student at University of Washington. He is the creator of SHAP, a unified approach to explain the output of machine learning models such as decision tree ensembles. He also helps maintain the XGBoost Julia package.
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
--------------------
* [Full List of Contributors](https://github.com/dmlc/xgboost/graphs/contributors)
- To contributors: please add your name to the list when you submit a patch to the project:)
* [Kailong Chen](https://github.com/kalenhaha)
- Kailong is an early contributor of xgboost, he is creator of ranking objectives in xgboost.
- Kailong is an early contributor of XGBoost, he is creator of ranking objectives in XGBoost.
* [Skipper Seabold](https://github.com/jseabold)
- Skipper is the major contributor to the scikit-learn module of xgboost.
- Skipper is the major contributor to the scikit-learn module of XGBoost.
* [Zygmunt Zając](https://github.com/zygmuntz)
- Zygmunt is the master behind the early stopping feature frequently used by kagglers.
#' \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 train_folds \code{list} list specifying which indicies to use for training. If \code{NULL}
#' (the default) all indices not specified in \code{folds} will be used for training.
#' @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.
#'
#' Of the \code{nfold} subsamples, a single subsample is retained as the validation data for testing the model, and the remaining \code{nfold - 1} subsamples are used as training data.
#'
#' @details
#' The original sample is randomly partitioned into \code{nfold} equal size subsamples.
#'
#' Of the \code{nfold} subsamples, a single subsample is retained as the validation data for testing the model, and the remaining \code{nfold - 1} subsamples are used as training data.
#'
#' The cross-validation process is then repeated \code{nrounds} times, with each of the \code{nfold} subsamples used exactly once as the validation data.
#'
#'
#' All observations are used for both training and validation.
#'
#'
#' 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 stored 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 list of the CV folds' models. It is only available with the explicit
#' setting of the \code{cb.cv.predict(save_models = TRUE)} callback.
#' 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 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}.
#' 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).
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.")
#' 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
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.")
#' 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 clustering of the importance values,
#' with bar colors corresponding 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.
#' 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).
#' 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}}.
#' An advanced interface for training xgboost model. Look at \code{\link{xgboost}} function for a simpler interface.
#'
#' @param params the list of parameters.
#'
#' \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
#'
#'
#' 2.1. Parameter for Tree Booster
#'
#'
#' \itemize{
#' \item \code{eta} control the learning rate: scale the contribution of each tree by a factor of \code{0 < eta < 1} when it is added to the current approximation. Used to prevent overfitting by making the boosting process more conservative. Lower value for \code{eta} implies larger value for \code{nrounds}: low \code{eta} value means model more robust to overfitting but slower to compute. Default: 0.3
#' \item \code{gamma} minimum loss reduction required to make a further partition on a leaf node of the tree. the larger, the more conservative the algorithm will be.
#' \item \code{gamma} minimum loss reduction required to make a further partition on a leaf node of the tree. the larger, the more conservative the algorithm will be.
#' \item \code{max_depth} maximum depth of a tree. Default: 6
#' \item \code{min_child_weight} minimum sum of instance weight (hessian) needed in a child. If the tree partition step results in a leaf node with the sum of instance weight less than min_child_weight, then the building process will give up further partitioning. In linear regression mode, this simply corresponds to minimum number of instances needed to be in each node. The larger, the more conservative the algorithm will be. Default: 1
#' \item \code{subsample} subsample ratio of the training instance. Setting it to 0.5 means that xgboost randomly collected half of the data instances to grow trees and this will prevent overfitting. It makes computation shorter (because less data to analyse). It is advised to use this parameter with \code{eta} and increase \code{nround}. Default: 1
#' \item \code{subsample} subsample ratio of the training instance. Setting it to 0.5 means that xgboost randomly collected half of the data instances to grow trees and this will prevent overfitting. It makes computation shorter (because less data to analyse). It is advised to use this parameter with \code{eta} and increase \code{nrounds}. Default: 1
#' \item \code{colsample_bytree} subsample ratio of columns when constructing each tree. Default: 1
#' \item \code{num_parallel_tree} Experimental parameter. number of trees to grow per round. Useful to test Random Forest through Xgboost (set \code{colsample_bytree < 1}, \code{subsample < 1} and \code{round = 1}) accordingly. Default: 1
#' \item \code{monotone_constraints} A numerical vector consists of \code{1}, \code{0} and \code{-1} with its length equals to the number of features in the training data. \code{1} is increasing, \code{-1} is decreasing and \code{0} is no constraint.
#' \item \code{interaction_constraints} A list of vectors specifying feature indices of permitted interactions. Each item of the list represents one permitted interaction where specified features are allowed to interact with each other. Feature index values should start from \code{0} (\code{0} references the first column). Leave argument unspecified for no interaction constraints.
#' }
#'
#'
#' 2.2. Parameter for Linear Booster
#'
#'
#' \itemize{
#' \item \code{lambda} L2 regularization term on weights. Default: 0
#' \item \code{lambda_bias} L2 regularization term on bias. Default: 0
#' \item \code{alpha} L1 regularization term on weights. (there is no L1 reg on bias because it is not important). Default: 0
#' }
#'
#' 3. Task Parameters
#'
#'
#' 3. Task Parameters
#'
#' \itemize{
#' \item \code{objective} specify the learning task and the corresponding learning objective, users can pass a self-defined function to it. The default objective options are below:
#' \itemize{
#' \item \code{reg:linear} linear regression (Default).
#' \item \code{reg:squarederror} Regression with squared loss (Default).
#' \item \code{reg:logistic} logistic regression.
#' \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 obj customized objective function. Returns gradient and second order
#' @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}.
#'
#' @details
#' This is the training function 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.
#' @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.
#'
#' Parallelization is automatically enabled if \code{OpenMP} is present.
#' @details
#' These are the training functions for \code{xgboost}.
#'
#' 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 following 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{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{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.
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) -
# 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).
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')
if (file.exists('xgb.model')) file.remove('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))
}
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
