在Scikit Learn中运行SelectKBest后获取功能名称的最简单方法

Question

我正在尝试使用scikit-learn的SelectKBest特征进行监督机器学习实验，但我不确定如何在找到最佳特征后创建新的数据框架:

让我们假设我想进行实验，选择5个最佳特征:

from sklearn.feature_selection import SelectKBest, f_classif
select_k_best_classifier = SelectKBest(score_func=f_classif, k=5).fit_transform(features_dataframe, targeted_class)

现在，如果我添加一行:

import pandas as pd
dataframe = pd.DataFrame(select_k_best_classifier)

我收到一个没有功能名称的新数据框(只有从0到4的索引)，但是我想用新选择的功能创建一个数据框，如下所示:

dataframe = pd.DataFrame(fit_transofrmed_features, columns=features_names)

我的问题是如何创建features_names列表?

我知道我应该使用:

 select_k_best_classifier.get_support()

返回一个布尔值数组，其中true值索引表示在原始数据框中应该被选择的列。

我应该如何使用这个布尔数组的所有功能名称的数组，我可以通过方法feature_names = list(features_dataframe.columns.values) ?

Answer 1

不需要循环。

# Create and fit selector
selector = SelectKBest(f_classif, k=5)
selector.fit(features_df, target)
# Get columns to keep and create new dataframe with those only
cols_idxs = selector.get_support(indices=True)
features_df_new = features_df.iloc[:,cols_idxs]

Answer 2

对我来说，这段代码工作得很好，更"python化":

mask = select_k_best_classifier.get_support()
new_features = features_dataframe.columns[mask]

Answer 3

您可以这样做:

mask = select_k_best_classifier.get_support() #list of booleans
new_features = [] # The list of your K best features
for bool_val, feature in zip(mask, feature_names):
    if bool_val:
        new_features.append(feature)

然后更改功能的名称:

dataframe = pd.DataFrame(fit_transofrmed_features, columns=new_features)

Answer 4

下面的代码将帮助您找到最重要的K个特征及其f值。设，X是pandas数据框架，其中的列是所有的特征，y是类标签的列表。

import pandas as pd
from sklearn.feature_selection import SelectKBest, f_classif
#Suppose, we select 5 features with top 5 Fisher scores
selector = SelectKBest(f_classif, k = 5)
#New dataframe with the selected features for later use in the classifier. fit() method works too, if you want only the feature names and their corresponding scores
X_new = selector.fit_transform(X, y)
names = X.columns.values[selector.get_support()]
scores = selector.scores_[selector.get_support()]
names_scores = list(zip(names, scores))
ns_df = pd.DataFrame(data = names_scores, columns=['Feat_names', 'F_Scores'])
#Sort the dataframe for better visualization
ns_df_sorted = ns_df.sort_values(['F_Scores', 'Feat_names'], ascending = [False, True])
print(ns_df_sorted)

Answer 5

根据chi2选择最佳10个功能;

from sklearn.feature_selection import SelectKBest, chi2
KBest = SelectKBest(chi2, k=10).fit(X, y) 

使用get_support()获取特性

f = KBest.get_support(1) #the most important features

创建新的df名为X_new;

X_new = X[X.columns[f]] # final features`

Answer 6

在Scikit-learn 1.0中，变压器具有get_feature_names_out方法，这意味着您可以编写

dataframe = pd.DataFrame(fit_transformed_features, columns=transformer.get_features_names_out())

Answer 7

还有另一种替代方法，但是，它没有上述解决方案快。

# Use the selector to retrieve the best features
X_new = select_k_best_classifier.fit_transform(train[feature_cols],train['is_attributed'])
# Get back the kept features as a DataFrame with dropped columns as all 0s
selected_features = pd.DataFrame(select_k_best_classifier.inverse_transform(X_new),
                            index=train.index,
                            columns= feature_cols)
selected_columns = selected_features.columns[selected_features.var() !=0]

Answer 8

# Fit the SelectKBest instance
select_k_best_classifier = SelectKBest(score_func=f_classif, k=5).fit(features_dataframe, targeted_class)
# Extract the required features
new_features  = select_k_best_classifier.get_feature_names_out(features_names)

Answer 9

假设您要选择10个最佳特性:

import pandas as pd
from sklearn.feature_selection import SelectKBest
selector = SelectKBest(score_func=chi2, k = 10)
selector.fit_transform(X, y)
features_names = selector.feature_names_in_
print(features_names)