StackingClassifier 具有在特征子集上训练的基础模型

您需要在自定义函数之上使用 and（检查此答案）来按相应的列筛选数据并操作目标值。make_pipelineFunctionTransformer

下面是一个演示代码

from sklearn.datasets import load_iris
from sklearn.tree import DecisionTreeClassifier
from sklearn.linear_model import LogisticRegression
from sklearn.preprocessing import StandardScaler, LabelBinarizer
from sklearn.pipeline import make_pipeline
from sklearn.ensemble import StackingClassifier
X, y = load_iris(return_X_y=True)

# Functions to filter the X on the important features
def select_dim1(X):
    red_X1 = X[:,:2]
    return red_X1

def select_dim2(X):
    red_X2 = X[:,2:]
    return red_X2

def select_dim3(X):
    red_X3 = X[:,3:]
    return red_X3

# functions to label binarise separate classes in y
def y_0(y):
    y[y == 0] = 1
    return y
def y_1(y):
    y[(y == 0)|(y == 2)] = 0
    return y
def y_2(y):
    y[y < 2] = 0
    return y

# converting them to function transformer
from sklearn.preprocessing import FunctionTransformer
select_dim1_tr = FunctionTransformer(select_dim1)
select_dim2_tr = FunctionTransformer(select_dim2)
select_dim3_tr = FunctionTransformer(select_dim3)

select_y_0_tr = FunctionTransformer(y_0)
select_y_1_tr = FunctionTransformer(y_1)
select_y_2_tr = FunctionTransformer(y_2)

estimators = [
    ('dt1', make_pipeline(select_y_0_tr,select_dim1_tr, DecisionTreeClassifier(random_state=42))),
    ('dt2', make_pipeline(select_y_1_tr,select_dim2_tr, DecisionTreeClassifier(random_state=42))),
    ('dt3', make_pipeline(select_y_2_tr,select_dim3_tr, DecisionTreeClassifier(random_state=42)))
]

clf = StackingClassifier(estimators=estimators, final_estimator=LogisticRegression())

from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y, stratify=y, random_state=42)

clf.fit(X_train, y_train).score(X_test, y_test)

你可以以编程方式做你所描述的事情，但我不确定使用一个简单的随机森林在内部完成所有这些工作（特征、子选择和拟合等）会有什么好处。

下面是您所描述的实现。我使用了与您提到的完全相同的基础和堆叠模型：

import numpy as np
import pandas as pd
from sklearn.datasets import make_classification
from sklearn.model_selection import train_test_split
from sklearn.tree import DecisionTreeClassifier
from sklearn.linear_model import LogisticRegression
from sklearn.ensemble import StackingClassifier
from sklearn.base import clone
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import FunctionTransformer

def select_columns(X, columns):
    return X[columns]


X, y = make_classification(n_samples=1000, n_features=10, n_classes=3, n_informative=3)
df = pd.DataFrame(X, columns=list('ABCDEFGHIJ'))
X_train, X_test, y_train, y_test = train_test_split(df, y, test_size=0.3, random_state=42)


feature_subsets = {
    0: ['A', 'B', 'C'],
    1: ['D', 'E', 'F', 'G'],
    2: ['G', 'H', 'I', 'J']
}

# Base model
base_dt_model = DecisionTreeClassifier(random_state=42)

#One-vs-Rest classifiers with feature subsets
classifiers = []
for class_label, features in feature_subsets.items():

    model = clone(base_dt_model)
    
    # select features, then apply the unique model
    pipeline = Pipeline([
        ('feature_selection', FunctionTransformer(select_columns, kw_args={'columns': features})),
        ('classifier', model)
    ])
    
    classifiers.append(('dt_class_' + str(class_label), pipeline))

# Logistic Regression as the metaclassifier
stack = StackingClassifier(estimators=classifiers, final_estimator=LogisticRegression())

stack.fit(X_train, y_train)

y_pred = stack.predict(X_test)