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

* 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

python-package/xgboost/sklearn.py

@@ -64,6 +64,8 @@ class XGBModel(XGBModelBase):
         Number of trees to fit.
     verbosity : int
         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
         Specify the learning task and the corresponding learning objective or
         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,
-                 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,
                  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,
@@ -146,6 +148,7 @@ class XGBModel(XGBModelBase):
         self.learning_rate = learning_rate
         self.n_estimators = n_estimators
         self.verbosity = verbosity
+        self.silent = silent
         self.objective = objective
         self.booster = booster
         self.gamma = gamma
@@ -240,6 +243,16 @@ class XGBModel(XGBModelBase):
         else:
             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:
             xgb_params.pop('nthread', None)
         return xgb_params
@@ -588,7 +601,8 @@ class XGBClassifier(XGBModel, XGBClassifierBase):
     __doc__ = "Implementation of the scikit-learn API for XGBoost classification.\n\n" \
         + '\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',
                  n_jobs=1, nthread=None, gamma=0, min_child_weight=1, max_delta_step=0,
                  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):
         super(XGBClassifier, self).__init__(
             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,
             min_child_weight=min_child_weight, max_delta_step=max_delta_step,
             subsample=subsample, colsample_bytree=colsample_bytree,
@@ -874,7 +888,8 @@ class XGBRFClassifier(XGBClassifier):
               + "for XGBoost random forest classification.\n\n"\
               + '\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,
                  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,
@@ -882,7 +897,7 @@ class XGBRFClassifier(XGBClassifier):
                  missing=None, **kwargs):
         super(XGBRFClassifier, self).__init__(
             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,
             min_child_weight=min_child_weight, max_delta_step=max_delta_step,
             subsample=subsample, colsample_bytree=colsample_bytree,
@@ -912,7 +927,8 @@ class XGBRFRegressor(XGBRegressor):
               + "for XGBoost random forest regression.\n\n"\
               + '\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,
                  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,
@@ -920,7 +936,7 @@ class XGBRFRegressor(XGBRegressor):
                  missing=None, **kwargs):
         super(XGBRFRegressor, self).__init__(
             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,
             min_child_weight=min_child_weight, max_delta_step=max_delta_step,
             subsample=subsample, colsample_bytree=colsample_bytree,
@@ -952,6 +968,8 @@ class XGBRanker(XGBModel):
             Number of boosted trees to fit.
         verbosity : int
             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
             Specify the learning task and the corresponding learning objective.
             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,
-                 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,
                  subsample=1, colsample_bytree=1, colsample_bylevel=1, colsample_bynode=1,
                  reg_alpha=0, reg_lambda=1, scale_pos_weight=1,
@@ -1045,7 +1063,7 @@ class XGBRanker(XGBModel):
 
         super(XGBRanker, self).__init__(
             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,
             min_child_weight=min_child_weight, max_delta_step=max_delta_step,
             subsample=subsample, colsample_bytree=colsample_bytree,

tests/python/test_with_sklearn.py

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