【发布时间】:2019-10-09 15:32:51
【问题描述】:
我需要提取在sklearn.ensemble.BaggingClassifier 中训练的每个模型的概率。这样做的原因是估计 XGBoostClassifier 模型的不确定性。
为此,我创建了一个继承自sklearn.ensemble.BaggingClassifier 的扩展类,并添加了一个新方法来获取这些概率。请注意这个问题与ModuleNotFoundError: No module named 'sklearn.utils._joblib'不同
我在下面展示了到目前为止我已经实现的一段代码:
必要的模块
from sklearn.ensemble import BaggingClassifier
from sklearn.ensemble.base import _partition_estimators
from sklearn.utils import check_array
from sklearn.utils.validation import check_is_fitted
import sklearn.utils as su
继承自BaggingClassifier的子类
class EBaggingClassifier(BaggingClassifier):
"""
Extends the class BaggingClassifier fromsklearn
"""
def __init__(self,
base_estimator=None,
n_estimators=10,
max_samples=1.0,
max_features=1.0,
bootstrap=True,
bootstrap_features=False,
oob_score=False,
warm_start=False,
n_jobs=1,
random_state=None,
verbose=0):
super().__init__(
base_estimator,
n_estimators,
max_samples,
max_features,
bootstrap,
bootstrap_features,
oob_score,
warm_start,
n_jobs,
random_state,
verbose)
允许计算每个估计量的概率的新方法定义如下。
def predict_proball(self, X):
"""
Computes the probability of each individual estimator
Parameters
----------
X : {array-like, sparse matrix} of shape = [n_samples, n_features]
The training input samples. Sparse matrices are accepted only if
they are supported by the base estimator.
Returns
-------
p : array of shape = [n_samples, n_classes]
The class probabilities of the input samples. The order of the
classes corresponds to that in the attribute `classes_`.
"""
check_is_fitted(self, "classes_")
# Check data
X = check_array(
X, accept_sparse=['csr', 'csc'], dtype=None,
force_all_finite=False
)
if self.n_features_ != X.shape[1]:
raise ValueError("Number of features of the model must "
"match the input. Model n_features is {0} and "
"input n_features is {1}."
"".format(self.n_features_, X.shape[1]))
# Parallel loop
n_jobs, n_estimators, starts = _partition_estimators(self.n_estimators,
self.n_jobs)
all_proba = su._joblib.Parallel(n_jobs=n_jobs, verbose=self.verbose,
**self._parallel_args())(
su._joblib.delayed(BaggingClassifier._parallel_predict_proba)(
self.estimators_[starts[i]:starts[i + 1]],
self.estimators_features_[starts[i]:starts[i + 1]],
X,
self.n_classes_)
for i in range(n_jobs))
return all_proba
我使用XGBoostClassifier 作为基础估计器来实例化这个类:
base_estimator = XGBoostClassifier(**params)
estimator = EBaggingClassifier(base_estimator=base_estimator, max_samples=0.8, n_estimators=10)
然后estimator 使用estimator.fit(X, y),其中X 和y 是pandas.DataFrame 对象。当我尝试运行 estimator.predict_proball(X) 时,我得到了
>>> estimator.predict_proball(X)
AttributeError: module 'sklearn.utils' has no attribute '_joblib'
有人知道为什么会这样吗?查看 BaggingClassifier script 函数 'sklearn.utils._joblib' 应该可用。
仅供参考:
>>> sklearn.__version__
'0.19.2'
【问题讨论】:
标签: python machine-learning scikit-learn ensemble-learning