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

Brought the silent parameter for the SKLearn-like API back, marked it deprecated. (#4255)

Browse Source 
* Brought the silent parameter for the SKLearn-like API back, marked it deprecated.

- added deprecation notice and warning
- removed silent from the tests for the SKLearn-like API
This commit is contained in:
Andy Adinets 2019-03-13 21:45:08 +01:00 committed by Rory Mitchell
parent b833b642ec
commit 4352fcdb15
2 changed files with 32 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

36
python-package/xgboost/sklearn.py
View File

@ -64,6 +64,8 @@ class XGBModel(XGBModelBase):
Number of trees to fit. Number of trees to fit.
verbosity : int verbosity : int
The degree of verbosity. Valid values are 0 (silent) - 3 (debug). The degree of verbosity. Valid values are 0 (silent) - 3 (debug).
silent : boolean
Whether to print messages while running boosting. Deprecated. Use verbosity instead.
objective : string or callable objective : string or callable
Specify the learning task and the corresponding learning objective or Specify the learning task and the corresponding learning objective or
a custom objective function to be used (see note below). a custom objective function to be used (see note below).
@ -134,7 +136,7 @@ class XGBModel(XGBModelBase):
""" """
def __init__(self, max_depth=3, learning_rate=0.1, n_estimators=100, def __init__(self, max_depth=3, learning_rate=0.1, n_estimators=100,
verbosity=1, objective="reg:linear", booster='gbtree', verbosity=1, silent=None, objective="reg:linear", booster='gbtree',
n_jobs=1, nthread=None, gamma=0, min_child_weight=1, n_jobs=1, nthread=None, gamma=0, min_child_weight=1,
max_delta_step=0, subsample=1, colsample_bytree=1, colsample_bylevel=1, max_delta_step=0, subsample=1, colsample_bytree=1, colsample_bylevel=1,
colsample_bynode=1, reg_alpha=0, reg_lambda=1, scale_pos_weight=1, colsample_bynode=1, reg_alpha=0, reg_lambda=1, scale_pos_weight=1,
@ -146,6 +148,7 @@ class XGBModel(XGBModelBase):
self.learning_rate = learning_rate self.learning_rate = learning_rate
self.n_estimators = n_estimators self.n_estimators = n_estimators
self.verbosity = verbosity self.verbosity = verbosity
self.silent = silent
self.objective = objective self.objective = objective
self.booster = booster self.booster = booster
self.gamma = gamma self.gamma = gamma
@ -240,6 +243,16 @@ class XGBModel(XGBModelBase):
else: else:
xgb_params['nthread'] = n_jobs xgb_params['nthread'] = n_jobs
if 'silent' in xgb_params and xgb_params['silent'] is not None:
warnings.warn('The silent parameter is deprecated.'
'Please use verbosity instead.'
'silent is depreated', DeprecationWarning)
# TODO(canonizer): set verbosity explicitly if silent is removed from xgboost,
# but remains in this API
else:
# silent=None shouldn't be passed to xgboost
xgb_params.pop('silent', None)
if xgb_params['nthread'] <= 0: if xgb_params['nthread'] <= 0:
xgb_params.pop('nthread', None) xgb_params.pop('nthread', None)
return xgb_params return xgb_params
@ -588,7 +601,8 @@ class XGBClassifier(XGBModel, XGBClassifierBase):
__doc__ = "Implementation of the scikit-learn API for XGBoost classification.\n\n" \ __doc__ = "Implementation of the scikit-learn API for XGBoost classification.\n\n" \
+ '\n'.join(XGBModel.__doc__.split('\n')[2:]) + '\n'.join(XGBModel.__doc__.split('\n')[2:])
def __init__(self, max_depth=3, learning_rate=0.1, n_estimators=100, verbosity=1, def __init__(self, max_depth=3, learning_rate=0.1, n_estimators=100,
verbosity=1, silent=None,
objective="binary:logistic", booster='gbtree', objective="binary:logistic", booster='gbtree',
n_jobs=1, nthread=None, gamma=0, min_child_weight=1, max_delta_step=0, n_jobs=1, nthread=None, gamma=0, min_child_weight=1, max_delta_step=0,
subsample=1, colsample_bytree=1, colsample_bylevel=1, subsample=1, colsample_bytree=1, colsample_bylevel=1,
@ -596,7 +610,7 @@ class XGBClassifier(XGBModel, XGBClassifierBase):
base_score=0.5, random_state=0, seed=None, missing=None, **kwargs): base_score=0.5, random_state=0, seed=None, missing=None, **kwargs):
super(XGBClassifier, self).__init__( super(XGBClassifier, self).__init__(
max_depth=max_depth, learning_rate=learning_rate, n_estimators=n_estimators, max_depth=max_depth, learning_rate=learning_rate, n_estimators=n_estimators,
verbosity=verbosity, objective=objective, booster=booster, verbosity=verbosity, silent=silent, objective=objective, booster=booster,
n_jobs=n_jobs, nthread=nthread, gamma=gamma, n_jobs=n_jobs, nthread=nthread, gamma=gamma,
min_child_weight=min_child_weight, max_delta_step=max_delta_step, min_child_weight=min_child_weight, max_delta_step=max_delta_step,
subsample=subsample, colsample_bytree=colsample_bytree, subsample=subsample, colsample_bytree=colsample_bytree,
@ -874,7 +888,8 @@ class XGBRFClassifier(XGBClassifier):
+ "for XGBoost random forest classification.\n\n"\ + "for XGBoost random forest classification.\n\n"\
+ '\n'.join(XGBModel.__doc__.split('\n')[2:]) + '\n'.join(XGBModel.__doc__.split('\n')[2:])
def __init__(self, max_depth=3, learning_rate=1, n_estimators=100, verbosity=1, def __init__(self, max_depth=3, learning_rate=1, n_estimators=100,
verbosity=1, silent=None,
objective="binary:logistic", n_jobs=1, nthread=None, gamma=0, objective="binary:logistic", n_jobs=1, nthread=None, gamma=0,
min_child_weight=1, max_delta_step=0, subsample=0.8, colsample_bytree=1, min_child_weight=1, max_delta_step=0, subsample=0.8, colsample_bytree=1,
colsample_bylevel=1, colsample_bynode=0.8, reg_alpha=0, reg_lambda=1, colsample_bylevel=1, colsample_bynode=0.8, reg_alpha=0, reg_lambda=1,
@ -882,7 +897,7 @@ class XGBRFClassifier(XGBClassifier):
missing=None, **kwargs): missing=None, **kwargs):
super(XGBRFClassifier, self).__init__( super(XGBRFClassifier, self).__init__(
max_depth=max_depth, learning_rate=learning_rate, n_estimators=n_estimators, max_depth=max_depth, learning_rate=learning_rate, n_estimators=n_estimators,
verbosity=verbosity, objective=objective, booster='gbtree', verbosity=verbosity, silent=silent, objective=objective, booster='gbtree',
n_jobs=n_jobs, nthread=nthread, gamma=gamma, n_jobs=n_jobs, nthread=nthread, gamma=gamma,
min_child_weight=min_child_weight, max_delta_step=max_delta_step, min_child_weight=min_child_weight, max_delta_step=max_delta_step,
subsample=subsample, colsample_bytree=colsample_bytree, subsample=subsample, colsample_bytree=colsample_bytree,
@ -912,7 +927,8 @@ class XGBRFRegressor(XGBRegressor):
+ "for XGBoost random forest regression.\n\n"\ + "for XGBoost random forest regression.\n\n"\
+ '\n'.join(XGBModel.__doc__.split('\n')[2:]) + '\n'.join(XGBModel.__doc__.split('\n')[2:])
def __init__(self, max_depth=3, learning_rate=1, n_estimators=100, verbosity=1, def __init__(self, max_depth=3, learning_rate=1, n_estimators=100,
verbosity=1, silent=None,
objective="reg:linear", n_jobs=1, nthread=None, gamma=0, objective="reg:linear", n_jobs=1, nthread=None, gamma=0,
min_child_weight=1, max_delta_step=0, subsample=0.8, colsample_bytree=1, min_child_weight=1, max_delta_step=0, subsample=0.8, colsample_bytree=1,
colsample_bylevel=1, colsample_bynode=0.8, reg_alpha=0, reg_lambda=1, colsample_bylevel=1, colsample_bynode=0.8, reg_alpha=0, reg_lambda=1,
@ -920,7 +936,7 @@ class XGBRFRegressor(XGBRegressor):
missing=None, **kwargs): missing=None, **kwargs):
super(XGBRFRegressor, self).__init__( super(XGBRFRegressor, self).__init__(
max_depth=max_depth, learning_rate=learning_rate, n_estimators=n_estimators, max_depth=max_depth, learning_rate=learning_rate, n_estimators=n_estimators,
verbosity=verbosity, objective=objective, booster='gbtree', verbosity=verbosity, silent=silent, objective=objective, booster='gbtree',
n_jobs=n_jobs, nthread=nthread, gamma=gamma, n_jobs=n_jobs, nthread=nthread, gamma=gamma,
min_child_weight=min_child_weight, max_delta_step=max_delta_step, min_child_weight=min_child_weight, max_delta_step=max_delta_step,
subsample=subsample, colsample_bytree=colsample_bytree, subsample=subsample, colsample_bytree=colsample_bytree,
@ -952,6 +968,8 @@ class XGBRanker(XGBModel):
Number of boosted trees to fit. Number of boosted trees to fit.
verbosity : int verbosity : int
The degree of verbosity. Valid values are 0 (silent) - 3 (debug). The degree of verbosity. Valid values are 0 (silent) - 3 (debug).
silent : boolean
Whether to print messages while running boosting. Deprecated. Use verbosity instead.
objective : string objective : string
Specify the learning task and the corresponding learning objective. Specify the learning task and the corresponding learning objective.
The objective name must start with "rank:". The objective name must start with "rank:".
@ -1037,7 +1055,7 @@ class XGBRanker(XGBModel):
""" """
def __init__(self, max_depth=3, learning_rate=0.1, n_estimators=100, def __init__(self, max_depth=3, learning_rate=0.1, n_estimators=100,
verbosity=1, objective="rank:pairwise", booster='gbtree', verbosity=1, silent=None, objective="rank:pairwise", booster='gbtree',
n_jobs=-1, nthread=None, gamma=0, min_child_weight=1, max_delta_step=0, n_jobs=-1, nthread=None, gamma=0, min_child_weight=1, max_delta_step=0,
subsample=1, colsample_bytree=1, colsample_bylevel=1, colsample_bynode=1, subsample=1, colsample_bytree=1, colsample_bylevel=1, colsample_bynode=1,
reg_alpha=0, reg_lambda=1, scale_pos_weight=1, reg_alpha=0, reg_lambda=1, scale_pos_weight=1,
@ -1045,7 +1063,7 @@ class XGBRanker(XGBModel):
super(XGBRanker, self).__init__( super(XGBRanker, self).__init__(
max_depth=max_depth, learning_rate=learning_rate, n_estimators=n_estimators, max_depth=max_depth, learning_rate=learning_rate, n_estimators=n_estimators,
verbosity=verbosity, objective=objective, booster=booster, verbosity=verbosity, silent=silent, objective=objective, booster=booster,
n_jobs=n_jobs, nthread=nthread, gamma=gamma, n_jobs=n_jobs, nthread=nthread, gamma=gamma,
min_child_weight=min_child_weight, max_delta_step=max_delta_step, min_child_weight=min_child_weight, max_delta_step=max_delta_step,
subsample=subsample, colsample_bytree=colsample_bytree, subsample=subsample, colsample_bytree=colsample_bytree,

10
tests/python/test_with_sklearn.py
View File

@ -380,7 +380,7 @@ def test_sklearn_nfolds_cv():
params = { params = {
'max_depth': 2, 'max_depth': 2,
'eta': 1, 'eta': 1,
'silent': 1, 'verbosity': 0,
'objective': 'objective':
'multi:softprob', 'multi:softprob',
'num_class': 3 'num_class': 3
@ -408,7 +408,7 @@ def test_split_value_histograms():
y = digits_2class['target'] y = digits_2class['target']
dm = xgb.DMatrix(X, label=y) dm = xgb.DMatrix(X, label=y)
params = {'max_depth': 6, 'eta': 0.01, 'silent': 1, params = {'max_depth': 6, 'eta': 0.01, 'verbosity': 0,
'objective': 'binary:logistic'} 'objective': 'binary:logistic'}
gbdt = xgb.train(params, dm, num_boost_round=10) gbdt = xgb.train(params, dm, num_boost_round=10)
@ -604,7 +604,7 @@ def test_RFECV():
X, y = load_boston(return_X_y=True) X, y = load_boston(return_X_y=True)
bst = xgb.XGBClassifier(booster='gblinear', learning_rate=0.1, bst = xgb.XGBClassifier(booster='gblinear', learning_rate=0.1,
n_estimators=10, n_jobs=1, objective='reg:linear', n_estimators=10, n_jobs=1, objective='reg:linear',
random_state=0, silent=True) random_state=0, verbosity=0)
rfecv = RFECV( rfecv = RFECV(
estimator=bst, step=1, cv=3, scoring='neg_mean_squared_error') estimator=bst, step=1, cv=3, scoring='neg_mean_squared_error')
rfecv.fit(X, y) rfecv.fit(X, y)
@ -614,7 +614,7 @@ def test_RFECV():
bst = xgb.XGBClassifier(booster='gblinear', learning_rate=0.1, bst = xgb.XGBClassifier(booster='gblinear', learning_rate=0.1,
n_estimators=10, n_jobs=1, n_estimators=10, n_jobs=1,
objective='binary:logistic', objective='binary:logistic',
random_state=0, silent=True) random_state=0, verbosity=0)
rfecv = RFECV(estimator=bst, step=1, cv=3, scoring='roc_auc') rfecv = RFECV(estimator=bst, step=1, cv=3, scoring='roc_auc')
rfecv.fit(X, y) rfecv.fit(X, y)
@ -625,7 +625,7 @@ def test_RFECV():
n_estimators=10, n_jobs=1, n_estimators=10, n_jobs=1,
objective='multi:softprob', objective='multi:softprob',
random_state=0, reg_alpha=0.001, reg_lambda=0.01, random_state=0, reg_alpha=0.001, reg_lambda=0.01,
scale_pos_weight=0.5, silent=True) scale_pos_weight=0.5, verbosity=0)
rfecv = RFECV(estimator=bst, step=1, cv=3, scoring='neg_log_loss') rfecv = RFECV(estimator=bst, step=1, cv=3, scoring='neg_log_loss')
rfecv.fit(X, y) rfecv.fit(X, y)