Jiaming Yuan 0dd97c206b

Move thread local entry into Learner. (#5396) ...

* 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.

2020-03-07 15:37:39 +08:00

..

load_pickle.py

Fix wrapping GPU ID and prevent data copying. (#5160)

2019-12-27 16:51:08 +08:00

test_from_columnar.py

Move thread local entry into Learner. (#5396)

2020-03-07 15:37:39 +08:00

test_from_cupy.py

Make some GPU tests deterministic (#5229)

2020-01-26 11:53:07 +13:00

test_gpu_basic_models.py

Deterministic GPU histogram. (#5361)

2020-03-04 15:13:28 +08:00

test_gpu_interaction_constraints.py

[Breaking] Remove learning_rates in Python. (#5155)

2019-12-24 14:25:48 +08:00

test_gpu_linear.py

Make HostDeviceVector single gpu only (#4773)

2019-08-26 09:51:13 +12:00

test_gpu_pickling.py

Enforce correct data shape. (#5191)

2020-01-13 15:48:17 +08:00

test_gpu_prediction.py

Enforce correct data shape. (#5191)

2020-01-13 15:48:17 +08:00

test_gpu_ranking.py

implementation of map ranking algorithm on gpu (#5129)

2019-12-27 12:05:37 +13:00

test_gpu_training_continuation.py

JSON configuration IO. (#5111)

2019-12-15 17:31:53 +08:00

test_gpu_updaters.py

Testing hist_util (#5251)

2020-02-14 14:36:43 +13:00

test_gpu_with_dask.py

Tests for empty dmatrix. (#5159)

2019-12-26 11:51:54 +08:00

test_gpu_with_sklearn.py

Enforce correct data shape. (#5191)

2020-01-13 15:48:17 +08:00

test_monotonic_constraints.py

Refactor Python tests. (#3897)

2018-11-15 13:56:33 +13:00