Skip to content

GitLab

Switch branch/tag
  1. 29 Jan, 2021 1 commit
  3. 28 Jan, 2021 1 commit
  5. 27 Jan, 2021 1 commit
  7. 26 Jan, 2021 5 commits
  9. 25 Jan, 2021 3 commits
  11. 24 Jan, 2021 2 commits
  13. 22 Jan, 2021 4 commits
  15. 20 Jan, 2021 1 commit
  17. 19 Jan, 2021 2 commits
  19. 18 Jan, 2021 1 commit
  21. 15 Jan, 2021 1 commit
  23. 11 Jan, 2021 1 commit
  25. 03 Jan, 2021 1 commit
  27. 29 Dec, 2020 1 commit
  29. 24 Dec, 2020 1 commit
    • Belinda Trotta's avatar
      Trees with linear models at leaves (#3299) · fcfd4132
      Belinda Trotta authored 
      * Add Eigen library.

* Working for simple test.

* Apply changes to config params.

* Handle nan data.

* Update docs.

* Add test.

* Only load raw data if boosting=gbdt_linear

* Remove unneeded code.

* Minor updates.

* Update to work with sk-learn interface.

* Update to work with chunked datasets.

* Throw error if we try to create a Booster with an already-constructed dataset having incompatible parameters.

* Save raw data in binary dataset file.

* Update docs and fix parameter checking.

* Fix dataset loading.

* Add test for regularization.

* Fix bugs when saving and loading tree.

* Add test for load/save linear model.

* Remove unneeded code.

* Fix case where not enough leaf data for linear model.

* Simplify code.

* Speed up code.

* Speed up code.

* Simplify code.

* Speed up code.

* Fix bugs.

* Working version.

* Store feature data column-wise (not fully working yet).

* Fix bugs.

* Speed up.

* Speed up.

* Remove unneeded code.

* Small speedup.

* Speed up.

* Minor updates.

* Remove unneeded code.

* Fix bug.

* Fix bug.

* Speed up.

* Speed up.

* Simplify code.

* Remove unneeded code.

* Fix bug, add more tests.

* Fix bug and add test.

* Only store numerical features

* Fix bug and speed up using templates.

* Speed up prediction.

* Fix bug with regularisation

* Visual studio files.

* Working version

* Only check nans if necessary

* Store coeff matrix as an array.

* Align cache lines

* Align cache lines

* Preallocation coefficient calculation matrices

* Small speedups

* Small speedup

* Reverse cache alignment changes

* Change to dynamic schedule

* Update docs.

* Refactor so that linear tree learner is not a separate class.

* Add refit capability.

* Speed up

* Small speedups.

* Speed up add prediction to score.

* Fix bug

* Fix bug and speed up.

* Speed up dataload.

* Speed up dataload

* Use vectors instead of pointers

* Fix bug

* Add OMP exception handling.

* Change return type of LGBM_BoosterGetLinear to bool

* Change return type of LGBM_BoosterGetLinear back to int, only parameter type needed to change

* Remove unused internal_parent_ property of tree

* Remove unused parameter to CreateTreeLearner

* Remove reference to LinearTreeLearner

* Minor style issues

* Remove unneeded check

* Reverse temporary testing change

* Fix Visual Studio project files

* Restore LightGBM.vcxproj.filters

* Speed up

* Speed up

* Simplify code

* Update docs

* Simplify code

* Initialise storage space for max num threads

* Move Eigen to include directory and delete unused files

* Remove old files.

* Fix so it compiles with mingw

* Fix gpu tree learner

* Change AddPredictionToScore back to const

* Fix python lint error

* Fix C++ lint errors

* Change eigen to a submodule

* Update comment

* Add the eigen folder

* Try to fix build issues with eigen

* Remove eigen files

* Add eigen as submodule

* Fix include paths

* Exclude eigen files from Python linter

* Ignore eigen folders for pydocstyle

* Fix C++ linting errors

* Fix docs

* Fix docs

* Exclude eigen directories from doxygen

* Update manifest to include eigen

* Update build_r to include eigen files

* Fix compiler warnings

* Store raw feature data as float

* Use float for calculating linear coefficients

* Remove eigen directory from GLOB

* Don't compile linear model code when building R package

* Fix doxygen issue

* Fix lint issue

* Fix lint issue

* Remove uneeded code

* Restore delected lines

* Restore delected lines

* Change return type of has_raw to bool

* Update docs

* Rename some variables and functions for readability

* Make tree_learner parameter const in AddScore

* Fix style issues

* Pass vectors as const reference when setting tree properties

* Make temporary storage of serial_tree_learner mutable so we can make the object's methods const

* Remove get_raw_size, use num_numeric_features instead

* Fix typo

* Make contains_nan_ and any_nan_ properties immutable again

* Remove data_has_nan_ property of tree

* Remove temporary test code

* Make linear_tree a dataset param

* Fix lint error

* Make LinearTreeLearner a separate class

* Fix lint errors

* Fix lint error

* Add linear_tree_learner.o

* Simulate omp_get_max_threads if openmp is not available

* Update PushOneData to also store raw data.

* Cast size to int

* Fix bug in ReshapeRaw

* Speed up code with multithreading

* Use OMP_NUM_THREADS

* Speed up with multithreading

* Update to use ArrayToString

* Fix tests

* Fix test

* Fix bug introduced in merge

* Minor updates

* Update docs
      fcfd4132
  31. 22 Dec, 2020 2 commits
    • Jan Stiborek's avatar
      [python] [dask] add initial dask integration (#3515) · d90a16d5
      Jan Stiborek authored 
      * migrated implementation from dask/dask-lightgbm

* relaxed tests

* tests skipped in case that MPI is used

* fixed python 2.7 import + tests disabled on windows

* python < 3.6 is not supported in tests

* tests enabled only for linux

* tests disabled for mpi interface

* dask version pinned to >= 2.0

* added @jameslamb as code owner

* added missing pandas dependency

* code refactoring, removed code duplication - lightgbm.dask.LGBMClassifier.fit is the same as lightgbm.dask.LGBMRegressor.fit

* fixed refactoring

* code deduplication - fit method moved into mixin class

* fixed CODEOWNERS

* removed unnecessary import

* skip the module execution on python < 3.6 and on platform different than linux.

* removed skip for python < 3.6

* review comments

* removed noqa, renamed API classes, renamed local variables
      d90a16d5
    • Nikita Titov's avatar
      [docs][ci] added docs about GPU support out of the box for Windows wheels and... · 1fb27443
      Nikita Titov authored 
      
[docs][ci] added docs about GPU support out of the box for Windows wheels and small refactoring for dual test (#3660)

* added docs about GPU support out of the box for Windows and small refactoring for dual test

* test

* Revert "test"

This reverts commit 45188103c2c9b3773bc62d97fbe865a2e5dd9120.

* fix docs

* fix docs

* hotfix config

* Apply suggestions from code review
Co-authored-by: default avatarTP Boudreau <tpboudreau@gmail.com>
Co-authored-by: default avatarTP Boudreau <tpboudreau@gmail.com>
      1fb27443
  33. 19 Dec, 2020 1 commit
    • Itamar Turner-Trauring's avatar
      [python] Windows wheels that support GPUs via OpenCL (#3403) · 222a7368
      Itamar Turner-Trauring authored 
      

* Add specialized OpenCL/Python package build path

* Refer to upstream OpenCL repository

* Reset build job count in setup.py

* Run test for dual CPU/GPU code in CI

* TEMPORARY: refer to OpenCL fork to ensure Linux CI builds succeed

* Try switching to OpenCL 2.0

* Add --opencl-python-pacakge option to wheel

* Fix some typos.

* Remove intermediate cmake target

* Restrict OpenCL headers to documented API version

* Refresh the environment after installation

* Try testing with two processes

* Use command line definition to activate integrated build

* Try newer OpenCL runtime

* If we hit OpenCL code path that's good enough

* Make tests run correctly

* Correct directory

* Flag reference to unofficial repo with FIXME

* TEMPORARY: update private repo tag for dependency

* Further post-merge fixes.

* Fix attribute name.

* Reformat with black.

* Rewrite config.h before library compilation for Windows wheel

* Run dual cpu/gpu test once only for Windows bdist CI

* Avoid C API tests when building gpu library for wheel

* TEMPORARY -- add clinfo tool to vm for debugging

* Increase compilation logging

* Rename OpenCL registry parameter after driver installation

* Install legacy OpenCL runtime

* ...

* Remove config.h change, clinfo installation, latest opencl runtime reference

* Add explicit script exit

* revert to 18.1 intel driver

* Run test_dual.py in two different modes

* Revert "Run test_dual.py in two different modes"

This reverts commit cbfec02659816f0cb321e1184994a9c2f37d2a77.

* Remove unused expected failure test

* Install OpenCL in separate powershell script

* fix setup.py imports

* activate conda env after session update

* separate update functions; exit opencl installation on error

* nothing; start pipeline

* download from original source

* specify tls version for installer download

* eliminate variable from command

* download opencl installer in parts

* remove duplicate setup.py command class entry

* download smaller parts

* remove redundant bdist wheel class

* Import chocolatey modules and use RefreshEnv

* insert pipeline cache for opencl installer

* rerun to ensure pipeline cache hit

* add comments, test another pipeline cache hit

* remove cache task; download full installer file from new location

* try another installer download timing

* reinsert pipeline cache

* retain cache task; download full file on cache miss

* test cache hit again

* remove todo
Co-authored-by: default avatarTP Boudreau <tpboudreau@gmail.com>
      222a7368
  35. 15 Dec, 2020 1 commit
  37. 11 Dec, 2020 1 commit
  39. 09 Dec, 2020 1 commit
    • Nikita Titov's avatar
      [python] Drop Python 2 support (#3581) · 44a6fb7f
      Nikita Titov authored 
      * Update setup.py

* Update .appveyor.yml

* Update .travis.yml

* Update .vsts-ci.yml

* Update __init__.py

* Update test.sh

* Update test_windows.ps1

* Update advanced_example.py

* Update requirements_base.txt

* Update conf.py

* Update conf.py

* Update test_engine.py

* Update utils.py

* Update dockerfile-r

* Update README.md

* Update dockerfile.gpu

* Update test_consistency.py

* Update basic.py

* Update compat.py

* Update engine.py

* Update sklearn.py

* Update sklearn.py

* Update callback.py

* Update setup.py

* Update __init__.py

* Update plotting.py

* Update sklearn.py

* Update engine.py

* Update compat.py

* Update callback.py

* Update basic.py

* Update compat.py

* Update basic.py

* Update basic.py

* Update compat.py

* Update compat.py

* Update plotting.py

* Update engine.py

* Update basic.py

* Update sklearn.py

* Update compat.py

* Update engine.py

* Update engine.py

* Update callback.py

* Update basic.py

* Update basic.py

* Update basic.py

* Update basic.py

* Update basic.py

* Update sklearn.py

* Update sklearn.py

* Update plotting.py

* Update sklearn.py

* Update compat.py

* Update compat.py

* Update engine.py

* Update plotting.py

* Update sklearn.py

* Update basic.py

* Update basic.py

* Update basic.py

* Update basic.py

* Update compat.py

* Update compat.py

* Update compat.py

* Update engine.py

* Update basic.py

* Update compat.py

* Update basic.py

* Update basic.py

* Update basic.py

* Update compat.py

* Update compat.py

* Update basic.py

* Update basic.py

* Update .vsts-ci.yml

* Update .vsts-ci.yml

* Update conf.py

* Revert "Update dockerfile-r"

This reverts commit 4ff6ffc7e3eeda24cc6a59a3bb0c973f02d9d71c.
      44a6fb7f
  41. 08 Dec, 2020 1 commit
    • Alberto Ferreira's avatar
      Fix model locale issue and improve model R/W performance. (#3405) · 792c9303
      Alberto Ferreira authored 
      * Fix LightGBM models locale sensitivity and improve R/W performance.

When Java is used, the default C++ locale is broken. This is true for
Java providers that use the C API or even Python models that require JEP.

This patch solves that issue making the model reads/writes insensitive
to such settings.
To achieve it, within the model read/write codebase:
 - C++ streams are imbued with the classic locale
 - Calls to functions that are dependent on the locale are replaced
 - The default locale is not changed!

This approach means:
 - The user's locale is never tampered with, avoiding issues such as
    https://github.com/microsoft/LightGBM/issues/2979 with the previous
    approach https://github.com/microsoft/LightGBM/pull/2891
 - Datasets can still be read according the user's locale
 - The model file has a single format independent of locale

Changes:
 - Add CommonC namespace which provides faster locale-independent versions of Common's methods
 - Model code makes conversions through CommonC
 - Cleanup unused Common methods
 - Performance improvements. Use fast libraries for locale-agnostic conversion:
   - value->string: https://github.com/fmtlib/fmt
   - string->double: https://github.com/lemire/fast_double_parser (10x
      faster double parsing according to their benchmark)

Bugfixes:
 - https://github.com/microsoft/LightGBM/issues/2500
 - https://github.com/microsoft/LightGBM/issues/2890
 - https://github.com/ninia/jep/issues/205

 (as it is related to LGBM as well)

* Align CommonC namespace

* Add new external_libs/ to python setup

* Try fast_double_parser fix #1

Testing commit e09e5aad828bcb16bea7ed0ed8322e019112fdbe

If it works it should fix more LGBM builds

* CMake: Attempt to link fmt without explicit PUBLIC tag

* Exclude external_libs from linting

* Add exernal_libs to MANIFEST.in

* Set dynamic linking option for fmt.

* linting issues

* Try to fix lint includes

* Try to pass fPIC with static fmt lib

* Try CMake P_I_C option with fmt library

* [R-package] Add CMake support for R and CRAN

* Cleanup CMakeLists

* Try fmt hack to remove stdout

* Switch to header-only mode

* Add PRIVATE argument to target_link_libraries

* use fmt in header-only mode

* Remove CMakeLists comment

* Change OpenMP to PUBLIC linking in Mac

* Update fmt submodule to 7.1.2

* Use fmt in header-only-mode

* Remove fmt from CMakeLists.txt

* Upgrade fast_double_parser to v0.2.0

* Revert "Add PRIVATE argument to target_link_libraries"

This reverts commit 3dd45dde7b92531b2530ab54522bb843c56227a7.

* Address James Lamb's comments

* Update R-package/.Rbuildignore
Co-authored-by: default avatarJames Lamb <jaylamb20@gmail.com>

* Upgrade to fast_double_parser v0.3.0 - Solaris support

* Use legacy code only in Solaris

* Fix lint issues

* Fix comment

* Address StrikerRUS's comments (solaris ifdef).

* Change header guards
Co-authored-by: default avatarJames Lamb <jaylamb20@gmail.com>
      792c9303
  43. 07 Dec, 2020 1 commit
  45. 30 Nov, 2020 1 commit
  47. 26 Nov, 2020 1 commit
  49. 24 Nov, 2020 1 commit
  51. 19 Nov, 2020 1 commit
  53. 15 Nov, 2020 1 commit
  55. 10 Nov, 2020 1 commit