Hendrik/xgboost
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Add notes to doc (#3765)

Browse Source
This commit is contained in:
Philip Hyunsu Cho 2018-10-07 14:09:09 -07:00 committed by GitHub
parent 91903ac5d4
commit c23783a0d1
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 26 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

20
python-package/xgboost/core.py
View File

@ -875,6 +875,10 @@ class Booster(object):
ctypes.byref(self.handle))) ctypes.byref(self.handle)))
self.set_param({'seed': 0}) self.set_param({'seed': 0})
self.set_param(params or {}) self.set_param(params or {})
if (params is not None) and ('booster' in params):
self.booster = params['booster']
else:
self.booster = 'gbtree'
if model_file is not None: if model_file is not None:
self.load_model(model_file) self.load_model(model_file)
@ -1379,6 +1383,12 @@ class Booster(object):
def get_fscore(self, fmap=''): def get_fscore(self, fmap=''):
"""Get feature importance of each feature. """Get feature importance of each feature.
.. note:: Feature importance is defined only for tree boosters
Feature importance is only defined when the decision tree model is chosen as base
learner (`booster=gbtree`). It is not defined for other base learner types, such
as linear learners (`booster=gblinear`).
Parameters Parameters
---------- ----------
fmap: str (optional) fmap: str (optional)
@ -1397,6 +1407,12 @@ class Booster(object):
* 'total_gain': the total gain across all splits the feature is used in. * 'total_gain': the total gain across all splits the feature is used in.
* 'total_cover': the total coverage across all splits the feature is used in. * 'total_cover': the total coverage across all splits the feature is used in.
.. note:: Feature importance is defined only for tree boosters
Feature importance is only defined when the decision tree model is chosen as base
learner (`booster=gbtree`). It is not defined for other base learner types, such
as linear learners (`booster=gblinear`).
Parameters Parameters
---------- ----------
fmap: str (optional) fmap: str (optional)
@ -1405,6 +1421,10 @@ class Booster(object):
One of the importance types defined above. One of the importance types defined above.
""" """
if self.booster != 'gbtree':
raise ValueError('Feature importance is not defined for Booster type {}'
.format(self.booster))
allowed_importance_types = ['weight', 'gain', 'cover', 'total_gain', 'total_cover'] allowed_importance_types = ['weight', 'gain', 'cover', 'total_gain', 'total_cover']
if importance_type not in allowed_importance_types: if importance_type not in allowed_importance_types:
msg = ("importance_type mismatch, got '{}', expected one of " + msg = ("importance_type mismatch, got '{}', expected one of " +

6
python-package/xgboost/sklearn.py
View File

@ -466,6 +466,12 @@ class XGBModel(XGBModelBase):
""" """
Feature importances property Feature importances property
.. note:: Feature importance is defined only for tree boosters
Feature importance is only defined when the decision tree model is chosen as base
learner (`booster=gbtree`). It is not defined for other base learner types, such
as linear learners (`booster=gblinear`).
Returns Returns
------- -------
feature_importances_ : array of shape ``[n_features]`` feature_importances_ : array of shape ``[n_features]``