Dr. Kashif Rasul 143475b27b use gain for sklearn feature_importances_ (#3876) ...

* 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

2018-11-13 03:30:40 -08:00

..

regression_test_utilities.py

Fixed issue 3605. (#3628)

2018-08-28 10:50:52 -07:00

test_basic_models.py

Revert "Fix #3485, #3540: Don't use dropout for predicting test sets" (#3563)

2018-08-08 09:48:55 -07:00

test_basic.py

Fix #3714: preserve feature names when slicing DMatrix (#3766)

2018-10-08 01:04:33 -07:00

test_dt.py

Allow import via python datatable. (#3272)

2018-06-20 13:16:18 -07:00

test_early_stopping.py

fix DeprecationWarning on sklearn.cross_validation (#2075)

2017-03-17 08:38:22 -05:00

test_eval_metrics.py

fix DeprecationWarning on sklearn.cross_validation (#2075)

2017-03-17 08:38:22 -05:00

test_interaction_constraints.py

[TREE] add interaction constraints (#3466)

2018-09-04 09:35:39 -07:00

test_linear.py

Added finding quantiles on GPU. (#3393)

2018-07-27 14:03:16 +12:00

test_monotone_constraints.py

'hist': Montonic Constraints (#3085)

2018-03-05 16:45:49 -08:00

test_openmp.py

Fix bugs in multithreaded ApplySplitSparseData() (#2161)

2017-04-02 11:37:39 -07:00

test_plotting.py

[PYTHON] Refactor trainnig API to use callback

2016-05-19 21:31:23 -07:00

test_shap.py

Add total_gain and total_cover importance measures (#3498)

2018-07-23 00:30:55 -07:00

test_sparse_dmatrix.py

More robust DMatrix creation from a sparse matrix (#1606)

2016-09-25 10:01:22 -07:00

test_training_continuation.py

[PYTHON] Refactor trainnig API to use callback

2016-05-19 21:31:23 -07:00

test_tree_regularization.py

Add LASSO (#3429)

2018-07-15 16:38:26 +12:00

test_updaters.py

Refactor python tests (#3410)

2018-06-27 11:20:27 +12:00

test_with_pandas.py

Fix issue 2670 (#2671)

2017-09-19 15:49:41 -04:00

test_with_sklearn.py

use gain for sklearn feature_importances_ (#3876)

2018-11-13 03:30:40 -08:00

testing.py

Allow import via python datatable. (#3272)

2018-06-20 13:16:18 -07:00