DMLC/XGBoost commits | Ecosyste.ms: OpenCollective

More support for column split in gpu predictor (#9562)

github.com/dmlc/xgboost - d8c3cc92ae677c320dc15c8090bf3b41e7725039 authored over 1 year ago

fix dupliate gpu check (#9578)

github.com/dmlc/xgboost - a343ae3b3452e42617a60fedd448d9a969148560 authored over 1 year ago

Fix default metric configuration. (#9575)

github.com/dmlc/xgboost - 300f9ace06b9d3c66c3376a5188894f82d127e46 authored over 1 year ago

Utilities and cleanups for socket. (#9576)

- Use c++-17 nodiscard and nested ns.
- Add bind method to socket.
- Remove rabit parameters.

github.com/dmlc/xgboost - b438d684d27286f09ec2864a2ce7da28f32acc0f authored over 1 year ago

[R] Fix method name. (#9577)

github.com/dmlc/xgboost - 5abe50ff8ca3acccd3b11487d82283f550e78e1e authored over 1 year ago

[doc] Fix typo in python_packaging.rst (#9573)

github.com/dmlc/xgboost - f90d034a86784f4f07417d1d28a77b4a189acc89 authored over 1 year ago

Allow JVM-Package to access inplace predict method (#9167)

---------

Co-authored-by: Stephan T. Lavavej <[email protected]>
Co-authored-by: Jiaming Yuan ...

github.com/dmlc/xgboost - d05ea589fb5dca2aa1ace6c9fbf7b5fbce896ed3 authored over 1 year ago

Support pandas 2.1.0. (#9557)

github.com/dmlc/xgboost - 9027686cac6b91f90e76c70b5aa6511f96c6e2c6 authored over 1 year ago

Add tests for gpu_approx (#9553)

github.com/dmlc/xgboost - 66a08327782e141003e2babd40487080dc3b66c4 authored over 1 year ago

[pyspark] support gpu transform (#9542)

---------

Co-authored-by: Jiaming Yuan <[email protected]>

github.com/dmlc/xgboost - 6c791b5b478e1b005aa9d954d20239f312b7f068 authored over 1 year ago

Add GPU support to NVFlare demo (#9552)

github.com/dmlc/xgboost - 0f35493b650070dbce7117261da60ff2f317df59 authored over 1 year ago

[CI] bump `setup-r` action version. (#9544)

github.com/dmlc/xgboost - 3b9e5909fb4d8f66727c9f5ce2566b89bd9dd19f authored over 1 year ago

Fix inplace predict with fallback when base margin is used. (#9536)

- Copy meta info from proxy DMatrix.
- Use `std::call_once` to emit less warnings.

github.com/dmlc/xgboost - adea842c838095b92bda2dde962a28a606513e57 authored over 1 year ago

[R] reformat build scripts (#9540)

github.com/dmlc/xgboost - d159ee85474b288694b4b68048489dedcd49c7a8 authored over 1 year ago

[pyspark] rework transform to reuse same code (#9292)

github.com/dmlc/xgboost - 419e052314f7bb5c14b0f4b82ea57e7ee72e7e9c authored over 1 year ago

add HTML files to gitignore (#9541)

github.com/dmlc/xgboost - 98e45f7b54f0b62564f8430e4a1467d4d51adf07 authored over 1 year ago

Support vertical federated learning with `gpu_hist` (#9539)

github.com/dmlc/xgboost - c928dd4ff500743fdf0ffd335464ccbf360f285d authored over 1 year ago

Support column split in gpu hist updater (#9384)

github.com/dmlc/xgboost - 9bab06cbca1ea78b1f17b20b2487d6d71359f1fa authored over 1 year ago

[rabit] Improved connection handling. (#9531)

- Enable timeout.
- Report connection error from the system.
- Handle retry for both tracker...

github.com/dmlc/xgboost - ccfc90e4c6aa3d3328b092b34065f5acdc6bce8a authored over 1 year ago

Bump com.nvidia:rapids-4-spark_2.12 in /jvm-packages (#9517)

Bumps com.nvidia:rapids-4-spark_2.12 from 23.08.0 to 23.08.1.

---
updated-dependencies:
- d...

github.com/dmlc/xgboost - 2462e22cd425d8918a4c507e0318d969e4e298bd authored over 1 year ago

Use the new `DeviceOrd` in the linalg module. (#9527)

github.com/dmlc/xgboost - ddf2e688219c0e5510a6931b7dac37e34ac14762 authored over 1 year ago

Fix GPU categorical split memory allocation. (#9529)

github.com/dmlc/xgboost - 942b957eef8e81c071ba543847eea8f4d2806df6 authored over 1 year ago

[R] Support multi-class custom objective. (#9526)

github.com/dmlc/xgboost - be6a55295695cb00c6d5b009dd466c3aa351b224 authored over 1 year ago

[rabit] Drop support for MPI backend. (#9525)

- Add checks in cmake.
- Remove mpi related code.

github.com/dmlc/xgboost - 90ef250ea10618865193bd0ca5c6491fc589d741 authored over 1 year ago

[R] Fix integer inputs with NA. (#9522)

github.com/dmlc/xgboost - c3574d932f5555c090ac9e6c4ee6b82baf3ace7a authored over 1 year ago

[rabit] Small cleanup to tracker initialization. (#9524)

- Remove recover related code.
- Clean startup, no need to consider previously connected nodes.

github.com/dmlc/xgboost - 1b87a1d8f8c1b372f1101badc7b3fdf03499ea28 authored over 1 year ago

Remove the deprecated Python rabit module. (#9523)

github.com/dmlc/xgboost - 209335b18cea53814aa78daee02079f1f39ee72a authored over 1 year ago

[dask] Filter models on worker. (#9518)

github.com/dmlc/xgboost - aa86bd5207964882a6a11eb401387747c8b1f18c authored over 1 year ago

Use matrix for gradient. (#9508)

- Use the `linalg::Matrix` for storing gradients.
- New API for the custom objective.
- Cust...

github.com/dmlc/xgboost - 972730cde05569e00e8e0576aea8dcd546f0f1dd authored over 1 year ago

Support column split in GPU evaluate splits (#9511)

github.com/dmlc/xgboost - 6103dca0bbc96b635eee8c7a69dbf2e25404faa0 authored over 1 year ago

Delay the check for vector leaf. (#9509)

github.com/dmlc/xgboost - 8c10af45a0886f2e400114dc4dbf4f9ffb4334f1 authored over 1 year ago

Fix device dispatch for linear updater. (#9507)

github.com/dmlc/xgboost - 3c09399f2992ebb73ac63b548cd6e469690600dc authored over 1 year ago

mitigate flaky test with distributed l1 error. (#9499)

github.com/dmlc/xgboost - 302bbdc958ca134238a3ee6ffbc52c817b9c8041 authored over 1 year ago

Drop support for loading remote files. (#9504)

github.com/dmlc/xgboost - 044fea1281e3910a70368068a8d72ccbe6b406f0 authored over 1 year ago

Bump scala-collection-compat_2.12 from 2.10.0 to 2.11.0 in /jvm-packages (#9311)

Bumps [scala-collection-compat_2.12](https://github.com/scala/scala-collection-compat) from 2.10...

github.com/dmlc/xgboost - d779a11af9979b222b15931709c41638a7cf5475 authored over 1 year ago

Deprecate the command line interface. (#9485)

---------

Co-authored-by: Philip Hyunsu Cho <[email protected]>

github.com/dmlc/xgboost - e6cf7a12785ab73f8aac79052a876ab40c65684f authored over 1 year ago

Fix release script for RC [skip ci] (#9505)

github.com/dmlc/xgboost - 38a3e1b858c12e19e1584474f61ba0dd50582e4d authored over 1 year ago

Bump spark.version.gpu in /jvm-packages/xgboost4j-spark-gpu (#9328)

Bumps `spark.version.gpu` from 3.3.2 to 3.4.1.

Updates `spark-core_2.12` from 3.3.2 to 3.4.1
...

github.com/dmlc/xgboost - 74d5056c614a0121dcda509318134e37c214705b authored over 1 year ago

[R] Differentiate dev version with release version. (#9503)

Use 2.1.0.0 as development version, we will change it to 2.1.0.1 during release.

github.com/dmlc/xgboost - db87d481bc1ffa3023bbe19213af37f0a0574cfd authored over 1 year ago

Bump org.apache.commons:commons-lang3 in /jvm-packages (#9489)

Bumps org.apache.commons:commons-lang3 from 3.12.0 to 3.13.0.

---
updated-dependencies:
- d...

github.com/dmlc/xgboost - 5358e1ebf02a5149d59af2dc2016f345a2d63cba authored over 1 year ago

Bump spark.version from 3.4.0 to 3.4.1 in /jvm-packages/xgboost4j-spark (#9326)

Bumps `spark.version` from 3.4.0 to 3.4.1.

Updates `spark-core_2.12` from 3.4.0 to 3.4.1

U...

github.com/dmlc/xgboost - d016309a15aacf1d634995ac81568a14e7f1682b authored over 1 year ago

Return base score as intercept. (#9486)

github.com/dmlc/xgboost - 7f29a238e6041a052b015b4ad99c5ddfd89f2640 authored over 1 year ago

Bump hadoop.version from 3.3.5 to 3.3.6 in /jvm-packages (#9331)

Bumps `hadoop.version` from 3.3.5 to 3.3.6.

Updates `hadoop-hdfs` from 3.3.5 to 3.3.6

Upda...

github.com/dmlc/xgboost - 0bb87b5b35bf7632d9b292bbaa3bea022ef80bfc authored over 1 year ago

Fix safe_xgboost macro on c++ (#9501)

github.com/dmlc/xgboost - b74802dea9260c4c12d3b89d10dbbb6abb017856 authored over 1 year ago

Bump version to 2.1. (#9498)

github.com/dmlc/xgboost - 58530b1bc459b81e318062f17fa2d6175eda8532 authored over 1 year ago

[pyspark] hotfix for GPU setup validation (#9495)

* [pyspark] fix a bug of validating gpu configuration

---------

Co-authored-by: Jiaming Yu...

github.com/dmlc/xgboost - 68be454cfa5512a1c3a46eef80bcc48ca53fb4ba authored over 1 year ago

Fix version parsing with rc release. (#9493)

github.com/dmlc/xgboost - 5188e275135306652fb0fdb2b3a6a771c147ad8e authored over 1 year ago

Use hint for find nccl. (#9490)

github.com/dmlc/xgboost - f380c10a939b87f8d7d26502ce2cad888998934f authored over 1 year ago

Fix federated learning demos and tests (#9488)

github.com/dmlc/xgboost - 12fe2fc06c47064299bcb9693aab45878e610005 authored over 1 year ago

[doc] Quick note for the `device` parameter. [skip ci] (#9483)

github.com/dmlc/xgboost - b2e93d2742314cb0594171442d7a4c33b830dca3 authored over 1 year ago

[jvm-packages] Bump rapids version. (#9482)

github.com/dmlc/xgboost - c061e3ae50931242a6695dfd7ec7b3a33f3f2c5d authored over 1 year ago

[R] remove commented-out code (#9481)

github.com/dmlc/xgboost - b82e78c1693f63c49cb4d0be62a220698d43f880 authored over 1 year ago

Updated versions. Reorganised dependencies. (#9479)

github.com/dmlc/xgboost - 8463107013c3f9e6466fdde6d03aff8938546af1 authored over 1 year ago

Convert input to str for hypothesis note. (#9480)

github.com/dmlc/xgboost - 19b59938b7bf5bcc29b200268a5dd44e470b0705 authored over 1 year ago

[R] remove more uses of default values in internal functions (#9476)

github.com/dmlc/xgboost - e3f624d8e765d15eb0b560d5ad8b1c140c61c314 authored over 1 year ago

[R] [doc] remove documentation index entries for internal functions (#9477)

github.com/dmlc/xgboost - 2c84daeca7a7098b9ad3594426be3eac4707a675 authored over 1 year ago

[jvm-packages] throw exception when tree_method=approx and device=cuda (#9478)

---------

Co-authored-by: Jiaming Yuan <[email protected]>

github.com/dmlc/xgboost - 344f90b67ba0966c04ef05321eb99c127c3c2552 authored over 1 year ago

Handle special characters in JSON model dump. (#9474)

github.com/dmlc/xgboost - 05d70000968acaf7c16e72eaaa32cb3f86fa209d authored over 1 year ago

[CI] Update RAPIDS to latest stable (#9464)

* [CI] Update RAPIDS to latest stable

* [CI] Use CMake 3.26.4

---------

Co-authored-by:...

github.com/dmlc/xgboost - f03463c45b9256d27009bba192a558e4d7b188c2 authored over 1 year ago

[doc] Document the current status of some features. (#9469)

github.com/dmlc/xgboost - fd4335d0bfb5795abeedf6f166bf3292d74fe5ed authored over 1 year ago

Test scikit-learn model IO with gblinear. (#9459)

github.com/dmlc/xgboost - 801116c30707d818e084cf95201db5f0ec361b17 authored over 1 year ago

Normalize file system path. (#9463)

github.com/dmlc/xgboost - bb5618339615338b3bb29aa5c208688f993a21ff authored over 1 year ago

Fix pyspark parameter. (#9460)

- Don't pass the `use_gpu` parameter to the learner.
- Fix GPU approx with PySpark.

github.com/dmlc/xgboost - bdc1a3c1780da9989e90caac527a29678e716ac8 authored over 1 year ago

[R] remove default values in internal booster manipulation functions (#9461)

github.com/dmlc/xgboost - 428f6cbbe20ccf65d6bdada13bcb0187cc67da04 authored over 1 year ago

Update README.md (#9462)

github.com/dmlc/xgboost - d6385355815005625115b648f6b7dc861eacd47e authored over 1 year ago

[R] remove default values in internal utility functions (#9457)

github.com/dmlc/xgboost - 44bd2981b2be1a40253945b7e02b297a297b140b authored over 1 year ago

[Doc] fix typos in documentation (#9458)

github.com/dmlc/xgboost - 9dbb71490c4737a55ed8b12296f53dfbf5b910c7 authored over 1 year ago

[R] [CI] use lintr 3.1.0 (#9456)

github.com/dmlc/xgboost - 4359356d46db7d30633ee13391233bb4839b7ca5 authored over 1 year ago

Use `realloc` for histogram cache and expose the cache limit. (#9455)

github.com/dmlc/xgboost - 1caa93221a11f0dd160731c10006ac6acb11da17 authored over 1 year ago

Fix links in R doc. (#9450)

github.com/dmlc/xgboost - a57371ef7c58acefe4bc794a23089a90734cc1a1 authored over 1 year ago

Use `weakref` instead of `id` for `DataIter` cache. (#9445)

- Fix case where Python reuses id from freed objects.
- Small optimization to column matrix wit...

github.com/dmlc/xgboost - f05a23b41cb9b3e55fe9f53c3ca26203c1d42748 authored over 1 year ago

[pyspark] add logs for training (#9449)

github.com/dmlc/xgboost - d495a180d8d486b11f5c4e8903a614d8604cdd13 authored over 1 year ago

Update R docs based on deprecated parameters/behaviour (#9437)

github.com/dmlc/xgboost - 7f854848d3df0709be360122ea0c78d5dc7c6cbf authored over 1 year ago

Fix clang warnings. (#9447)

- static function in header. (which is marked as unused due to translation unit
visibility).
-...

github.com/dmlc/xgboost - f05294a6f2e7d065ce7d479b9ad503e677f1b981 authored over 1 year ago

[R] Handle UTF-8 paths on Windows (#9448)

github.com/dmlc/xgboost - 819098a48f2e58a92894bd0e39f191ea4e199323 authored over 1 year ago

[CI] Check compiler warnings. (#9444)

github.com/dmlc/xgboost - c1b2cff874aca6aef3d8322255acc6721c95aa51 authored over 1 year ago

Handle UTF-8 paths correctly on Windows platform (#9443)

* Fix round-trip serialization with UTF-8 paths

* Add compiler version check

* Add comment...

github.com/dmlc/xgboost - 7ce090e775e017f7850a279befddb1d694c0c9fe authored over 1 year ago

Use lambda function in `ParallelFor2D`. (#9441)

github.com/dmlc/xgboost - 97fd5207dd1dbff4fce0df2d844ead19d2bf93cd authored over 1 year ago

Bound the size of the histogram cache. (#9440)

- A new histogram collection with a limit in size.
- Unify histogram building logic between h...

github.com/dmlc/xgboost - 54029a59af1632bd953e058855e0277292d8bee0 authored over 1 year ago

Explicitly specify libcudart_static in CMake config (#9436)

github.com/dmlc/xgboost - 5bd163aa259d2fa19054fe47c6564f5aeb88ed40 authored over 1 year ago

Remove Koffie Labs from Sponsors list (#9434)

github.com/dmlc/xgboost - 7fc57f39743352990b5be66dae11c9f15e0f74c4 authored over 1 year ago

Switch back to the GPUIDX macro (#9438)

github.com/dmlc/xgboost - bde1ebc209e749775ccced5155c8b1b4e7af240e authored over 1 year ago

[Doc] Clarify the output behavior of reg:logistic (#9435)

github.com/dmlc/xgboost - 1aabc690ec3969c3e62666cd702808d870e8671c authored over 1 year ago

Change training stage from ResultStage to ShuffleMapStage (#9423)

github.com/dmlc/xgboost - 04c99683c3b4f27afb8a877c67721ccd6715881a authored over 1 year ago

Unify the code path between local and distributed training. (#9433)

This removes the need for a local histogram space during distributed training, which cuts the ca...

github.com/dmlc/xgboost - 1332ff787ff620c0d12230b4f3af4d695022e483 authored over 1 year ago

Raise if expected workers are not alive in `xgboost.dask.train` (#9421)

github.com/dmlc/xgboost - f958e326832c9acc20f4c9548132f036cf1785af authored over 1 year ago

Accept only keyword arguments in data iterator. (#9431)

github.com/dmlc/xgboost - 71299888473b59310b3a14e1080364472438573e authored over 1 year ago

Small cleanup for histogram routines. (#9427)

* Small cleanup for histogram routines.

- Extract hist train param from GPU hist.
- Make his...

github.com/dmlc/xgboost - e93a2748230656de9770aa880157dc71c35af2da authored over 1 year ago

Clean up MGPU C++ tests (#9430)

github.com/dmlc/xgboost - c2b85ab68a29e1f691ebe61f9d1180cb7484b81f authored over 1 year ago

[CI] Update github actions. (#9428)

github.com/dmlc/xgboost - a9da2e244a0102732693a454bab4a0a9154a0414 authored over 1 year ago

Initial GPU support for the approx tree method. (#9414)

github.com/dmlc/xgboost - 912e341d575f107be1cc2631271fd0737b75dfba authored over 1 year ago

[jvm-packages] automatically set the max/min direction for best score (#9404)

github.com/dmlc/xgboost - 8f0efb4ab3ab21e527ba1f32c03cac7cd1ac8bd2 authored over 1 year ago

Retry switching to per-thread default stream (#9416)

github.com/dmlc/xgboost - 7579905e187fbbc2cec48299f7ea36a0365ba4d8 authored over 1 year ago

[doc] Fix typo in prediction.rst (#9415)

Typo for `pred_contribs` and `pred_interactions`

github.com/dmlc/xgboost - 54579da4d7f23f9620ad562b640b82ac429b5a2e authored over 1 year ago

Revert "Switch to per-thread default stream (#9396)" (#9413)

This reverts commit f7f673b00c15458fb4dd74a2a0d2ba80369c5faf.

github.com/dmlc/xgboost - 3a9996173e5209dff0e71ea8b2e85c954339ab59 authored over 1 year ago

[jvm-packages] set device to cuda when tree method is "gpu_hist" (#9412)

github.com/dmlc/xgboost - 1b657a55138f60dbf46ec9e5c80c05aae3cf2c4d authored over 1 year ago

Implement sketching with Hessian on GPU. (#9399)

- Prepare for implementing approx on GPU.
- Unify the code path between weighted and uniform sk...

github.com/dmlc/xgboost - a196443a0734df3cf646c951858a294a0325c697 authored over 1 year ago

Define `best_iteration` only if early stopping is used. (#9403)

* Define `best_iteration` only if early stopping is used.

This is the behavior specified by t...

github.com/dmlc/xgboost - 851cba931ebafd5ee38e57fb980dcaf465dd87fc authored over 1 year ago

[breaking] Remove support for single string feature info. (#9401)

- Input must be a sequence of strings.
- Improve validation error message.

github.com/dmlc/xgboost - 01e00efc53003290051ad0b3db0c85a6ba74666b authored over 1 year ago

Document for device ordinal. (#9398)

- Rewrite GPU demos. notebook is converted to script to avoid committing additional png plots.
...

github.com/dmlc/xgboost - 275da176ba097fa035e3deaca68400812d2a9dfa authored over 1 year ago