使用递归特征消除时出现TypeError：__init__()参数数量不符

运行代码时出现错误：

TypeError: __init__() takes 2 positional arguments but 3 were given.

完整报错信息：

runfile('C:/Users/drash/OneDrive/Desktop/Howto Health/preprocessing 1.py', wdir='C:/Users/drash/OneDrive/Desktop/Howto Health')
Traceback (most recent call last):
File "C:\Users\drash\OneDrive\Desktop\Howto Health\preprocessing 1.py", line 60, in <module>
    rfe = RFE(model, 10)

TypeError: __init__() takes 2 positional arguments but 3 were given

相关代码片段：

features = ['EMB_CD3', 'EMB_CD45R0', 'EMB_LFA1', 'EMB_Perforin', 'EMB_Mac', 'EMB_HLA1', 'EMB_CD54', 
            'EMB_VCAM', 'Virus:1=Cox,4=B19V,6=HHV6,5=EBV,2=ADV(including double infections)', 'Viral_load_B19V-VP1_mRNA',
            'Viral_load_DNA_B19V','Virus_EBV', 'Virus_HHV-6', 'B19V-Typ1']
df_new = df[features].copy()
#KNN Imputer for missing values
imputer = KNNImputer(n_neighbors=3)
imputed = imputer.fit_transform(df_new)
df_imputed = pd.DataFrame(imputed, columns=df_new.columns)

X = df_imputed
Y = df['target'].astype(int)
#%% feature extraction using Recursive Feature Elimination
model = LogisticRegression(solver='lbfgs')
rfe = RFE(model, 10)
fit = rfe.fit(X, Y)
print("Num Features: %s" % (fit.n_features_))
print("Selected Features: %s" % (fit.support_))
print("Feature Ranking: %s" % (fit.ranking_))

这个错误是因为新版本scikit-learn中，RFE的构造函数不再支持通过位置参数传递特征数量。旧版写法RFE(model, 10)在新版中不再适用，n_features_to_select必须用关键字参数指定。

修正后的RFE初始化代码：

rfe = RFE(model, n_features_to_select=10)

完整修正后的特征提取部分代码：

#%% feature extraction using Recursive Feature Elimination
model = LogisticRegression(solver='lbfgs')
rfe = RFE(model, n_features_to_select=10)
fit = rfe.fit(X, Y)
print("Num Features: %s" % (fit.n_features_))
print("Selected Features: %s" % (fit.support_))
print("Feature Ranking: %s" % (fit.ranking_))

原理说明：scikit-learn 1.0+版本中，RFE的__init__方法签名变更，除第一个必选参数estimator外，其余参数都需要显式指定关键字，不能再用位置参数传递，否则会被识别为多余参数，触发TypeError。

内容的提问来源于stack exchange，提问作者drashti bhatt