Jiaming Yuan d3ec116322

Revert ntree limit fix (#6616) (#6622) ...

The old (before fix) best_ntree_limit ignores the num_class parameters, which is incorrect. In before we workarounded it in c++ layer to avoid possible breaking changes on other language bindings. But the Python interpretation stayed incorrect. The PR fixed that in Python to consider num_class, but didn't remove the old workaround, so tree calculation in predictor is incorrect, see PredictBatch in CPUPredictor.

2021-01-20 04:20:07 +08:00

..

generate_models.py

Enable loading model from <1.0.0 trained with objective='binary:logitraw' (#6517) (#6524)

2020-12-18 04:10:09 +08:00

test_basic_models.py

Use pytest conventions consistently (#6337)

2020-11-19 17:00:15 -08:00

test_basic.py

[backport] Move metric configuration into booster. (#6504) (#6533)

2020-12-20 10:36:32 +08:00

test_callback.py

Fix handling of print period in EvaluationMonitor (#6499) (#6534)

2020-12-20 15:07:42 +08:00

test_cli.py

Fix CLI ranking demo. (#6439)

2020-12-03 21:28:28 -08:00

test_demos.py

Group CLI demo into subdirectory. (#6258)

2020-10-28 14:40:44 -07:00

test_dmatrix.py

Use pytest conventions consistently (#6337)

2020-11-19 17:00:15 -08:00

test_dt.py

Use pytest conventions consistently (#6337)

2020-11-19 17:00:15 -08:00

test_early_stopping.py

Use pytest conventions consistently (#6337)

2020-11-19 17:00:15 -08:00

test_eval_metrics.py

Use pytest conventions consistently (#6337)

2020-11-19 17:00:15 -08:00

test_interaction_constraints.py

Use pytest conventions consistently (#6337)

2020-11-19 17:00:15 -08:00

test_linear.py

Relax linear test. (#5849)

2020-07-03 07:49:53 +08:00

test_model_compatibility.py

Enable loading model from <1.0.0 trained with objective='binary:logitraw' (#6517) (#6524)

2020-12-18 04:10:09 +08:00

test_monotone_constraints.py

Use pytest conventions consistently (#6337)

2020-11-19 17:00:15 -08:00

test_openmp.py

Use pytest conventions consistently (#6337)

2020-11-19 17:00:15 -08:00

test_parse_tree.py

Use pytest conventions consistently (#6337)

2020-11-19 17:00:15 -08:00

test_pickling.py

Use pytest conventions consistently (#6337)

2020-11-19 17:00:15 -08:00

test_plotting.py

Use pytest conventions consistently (#6337)

2020-11-19 17:00:15 -08:00

test_predict.py

Revert ntree limit fix (#6616) (#6622)

2021-01-20 04:20:07 +08:00

test_ranking.py

Use pytest conventions consistently (#6337)

2020-11-19 17:00:15 -08:00

test_shap.py

Use pytest conventions consistently (#6337)

2020-11-19 17:00:15 -08:00

test_survival.py

[dask] Allow empty data matrix in AFT survival (#6379)

2020-11-12 17:49:58 -08:00

test_tracker.py

[dask] Supoort running on GKE. (#6343)

2020-11-11 18:04:34 +08:00

test_training_continuation.py

Revert ntree limit fix (#6616) (#6622)

2021-01-20 04:20:07 +08:00

test_tree_regularization.py

Use pytest conventions consistently (#6337)

2020-11-19 17:00:15 -08:00

test_updaters.py

Use pytest conventions consistently (#6337)

2020-11-19 17:00:15 -08:00

test_with_dask.py

[Backport] Rename data to X in predict_proba. (#6555) (#6586)

2021-01-10 16:05:17 +08:00

test_with_modin.py

Use pytest conventions consistently (#6337)

2020-11-19 17:00:15 -08:00

test_with_pandas.py

Use pytest conventions consistently (#6337)

2020-11-19 17:00:15 -08:00

test_with_shap.py

Update GPUTreeshap (#6163)

2020-09-28 09:43:47 +13:00

test_with_sklearn.py

Revert ntree limit fix (#6616) (#6622)

2021-01-20 04:20:07 +08:00

testing.py

Use pytest conventions consistently (#6337)

2020-11-19 17:00:15 -08:00