Python之category-encoders：category-encoders库的简介、安装、使用方法之详细攻略

蔚落 2022-12-21 09:26 321阅读 0赞

**目录**

category-encoders库的简介

1、支持15种不同编码

category-encoders库的安装

category-encoders库的使用方法

1、一个无监督的例子

2、一个有监督的例子

--------------------

## category-encoders库的简介 ##

一组scikit-learn风格的转换器，用不同的技术将类别变量编码成数字。一组scikit-learn风格的转换器，用不同的技术将分类变量编码成数字。虽然序数编码、单热编码和散列编码在现有scikit-learn版本中具有类似的等价性，但这个库中的变形金刚都有一些有用的特性:

*  对pandas数据文件作为输入(可选作为输出)的一流支持
 *  可以显式配置数据中的哪些列按名称或索引编码，或推断非数字列，而不管输入类型
 *  可以根据训练集随意删除任何方差很低的列吗
 *  可移植性:培训数据转换器，pickle它，以后重用它，然后得到同样的东西。
 *  完全兼容sklearn管道，输入一个类似数组的数据集，像任何其他转换器

### 1、支持15种不同编码 ###

1.  [Backward Difference Coding][]
2.  [BaseN][]
3.  [Binary][]
4.  [CatBoost Encoder][]
5.  [Count Encoder][]
6.  [Generalized Linear Mixed Model Encoder][]
7.  [Hashing][]
8.  [Helmert Coding][]
9.  [James-Stein Encoder][]
10. [Leave One Out][]
11. [M-estimate][]
12. [One Hot][]
13. [Ordinal][]
14. [Polynomial Coding][]
15. [Sum Coding][]
16. [Target Encoder][]
17. [Weight of Evidence][]
18. [Wrappers][]

**文档**：[http://contrib.scikit-learn.org/category\_encoders/][http_contrib.scikit-learn.org_category_encoders]

## category-encoders库的安装 ##

pip install category-encoders

## category-encoders库的使用方法 ##

有两种类型的编码器:无监督和有监督的。

### 1、一个无监督的例子 ###

from category_encoders import *
    import pandas as pd
    from sklearn.datasets import load_boston
    
    # prepare some data
    bunch = load_boston()
    y = bunch.target
    X = pd.DataFrame(bunch.data, columns=bunch.feature_names)
    
    # use binary encoding to encode two categorical features
    enc = BinaryEncoder(cols=['CHAS', 'RAD']).fit(X)
    
    # transform the dataset
    numeric_dataset = enc.transform(X)

### 2、一个有监督的例子 ###

from category_encoders import *
    import pandas as pd
    from sklearn.datasets import load_boston
    
    # prepare some data
    bunch = load_boston()
    y_train = bunch.target[0:250]
    y_test = bunch.target[250:506]
    X_train = pd.DataFrame(bunch.data[0:250], columns=bunch.feature_names)
    X_test = pd.DataFrame(bunch.data[250:506], columns=bunch.feature_names)
    
    # use target encoding to encode two categorical features
    enc = TargetEncoder(cols=['CHAS', 'RAD'])
    
    # transform the datasets
    training_numeric_dataset = enc.fit_transform(X_train, y_train)
    testing_numeric_dataset = enc.transform(X_test)

[Backward Difference Coding]: http://contrib.scikit-learn.org/category_encoders/backward_difference.html
[BaseN]: http://contrib.scikit-learn.org/category_encoders/basen.html
[Binary]: http://contrib.scikit-learn.org/category_encoders/binary.html
[CatBoost Encoder]: http://contrib.scikit-learn.org/category_encoders/catboost.html
[Count Encoder]: http://contrib.scikit-learn.org/category_encoders/count.html
[Generalized Linear Mixed Model Encoder]: http://contrib.scikit-learn.org/category_encoders/glmm.html
[Hashing]: http://contrib.scikit-learn.org/category_encoders/hashing.html
[Helmert Coding]: http://contrib.scikit-learn.org/category_encoders/helmert.html
[James-Stein Encoder]: http://contrib.scikit-learn.org/category_encoders/jamesstein.html
[Leave One Out]: http://contrib.scikit-learn.org/category_encoders/leaveoneout.html
[M-estimate]: http://contrib.scikit-learn.org/category_encoders/mestimate.html
[One Hot]: http://contrib.scikit-learn.org/category_encoders/onehot.html
[Ordinal]: http://contrib.scikit-learn.org/category_encoders/ordinal.html
[Polynomial Coding]: http://contrib.scikit-learn.org/category_encoders/polynomial.html
[Sum Coding]: http://contrib.scikit-learn.org/category_encoders/sum.html
[Target Encoder]: http://contrib.scikit-learn.org/category_encoders/targetencoder.html
[Weight of Evidence]: http://contrib.scikit-learn.org/category_encoders/woe.html
[Wrappers]: http://contrib.scikit-learn.org/category_encoders/wrapper.html
[http_contrib.scikit-learn.org_category_encoders]: http://contrib.scikit-learn.org/category_encoders/