基于列的sklearn分层抽样

Question

我有一个相当大的 CSV 文件，其中包含我读入 pandas 数据框的亚马逊评论数据。我想将数据拆分为 80-20（训练测试），但在这样做时我想确保拆分数据按比例代表一列（类别）的值，即所有不同类别的评论都出现在火车中并按比例测试数据。

数据如下所示：

**ReviewerID**       **ReviewText**        **Categories**       **ProductId**

1212                   good product         Mobile               14444425
1233                   will buy again       drugs                324532
5432                   not recomended       dvd                  789654123

我正在使用以下代码来执行此操作：

import pandas as pd
Meta = pd.read_csv('C:\Users\xyz\Desktop\WM Project\Joined.csv')
import numpy as np
from sklearn.cross_validation import train_test_split

train, test = train_test_split(Meta.categories, test_size = 0.2, stratify=y)

它给出以下错误

NameError: name 'y' is not defined

由于我对 python 还比较陌生，所以我不知道我做错了什么，也不知道这段代码是否会根据列类别进行分层。当我从 train-test split 中删除分层选项和类别列时，它似乎工作正常。

任何帮助将不胜感激。

Answer 1

sklearn.model_selection.train_test_split

stratify : array-like or None (default is None)

If not None, data is split in a stratified fashion, using this as the class labels.

根据 API 文档，我认为您必须尝试 X_train, X_test, y_train, y_test = train_test_split(Meta_X, Meta_Y, test_size = 0.2, stratify=Meta_Y)。

Meta_X, Meta_Y 应该由你正确分配（我认为 Meta_Y 根据你的代码应该是 Meta.categories）。

Answer 2

    >>> import pandas as pd
    >>> Meta = pd.read_csv('C:\Users\*****\Downloads\so\Book1.csv')
    >>> import numpy as np
    >>> from sklearn.model_selection import train_test_split
    >>> y = Meta.pop('Categories')
    >>> Meta
        ReviewerID      ReviewText  ProductId
        0        1212    good product   14444425
        1        1233  will buy again     324532
        2        5432  not recomended  789654123
    >>> y
        0    Mobile
        1     drugs
        2       dvd
        Name: Categories, dtype: object
    >>> X = Meta
    >>> X_train, X_test, y_train, y_test = train_test_split( X, y, test_size=0.33, random_state=42, stratify=y)
    >>> X_test
        ReviewerID    ReviewText  ProductId
        0        1212  good product   14444425

Answer 3

我不确定为什么没有人提到 StratifiedShuffleSplit

from sklearn.model_selection import StratifiedShuffleSplit
split = StratifiedShuffleSplit(n_splits=10, test_size=0.2, random_state=42)
for train_index, test_index in split.split(df, df['Categories']):
    strat_train_set = df.loc[train_index]
    strat_test_set = df.loc[test_index]

有关文档，请参阅 StratifiedShuffleSplit

基于列的sklearn分层抽样

sklearn stratified sampling based on a column

python

pandas

scikit-learn

sklearn-pandas