Python 为什么在 Pandas Series 上调用 .sort() 函数会就地对其值进行排序并且不返回任何内容？
声明：本页面是StackOverFlow热门问题的中英对照翻译，遵循CC BY-SA 4.0协议，如果您需要使用它，必须同样遵循CC BY-SA许可，注明原文地址和作者信息，同时你必须将它归于原作者（不是我）：StackOverFlow
原文地址: http://stackoverflow.com/questions/24049509/
Warning: these are provided under cc-by-sa 4.0 license. You are free to use/share it, But you must attribute it to the original authors (not me):
StackOverFlow
Why calling .sort() function on Pandas Series sorts its values in-place and returns nothing?
提问by meto
Sorry I think I am missing something very basic here:
对不起，我想我在这里遗漏了一些非常基本的东西：
>>> Series([3,4,0,3]).sort()
outputs None, while
输出无，而
>>> Series([3,4,0,3]).order()
2 0
0 3
3 3
1 4
dtype: int64
what am I missing with sort()?
我缺少 sort() 什么？
Thanks
谢谢
EDIT:
编辑：
Thanks for the answers, I do realize now that this is sorting in place. But I don't understand why
感谢您的回答，我现在意识到这是在排序。但我不明白为什么
>>> s = Series([3,4,0,3]).sort()
>>> s
does not return the sorted Series. If I understand the manualit should return the series sorted in place.
不返回排序的系列。如果我理解手册，它应该返回排序到位的系列。
采纳答案by gerrit
.sort()sorts in-place.
.sort()就地排序。
That means that after you call .sort(), your existing arrayhas been sorted. It doesn't return anything.
这意味着在您调用 之后.sort()，您现有的数组已被排序。它不返回任何东西。
To take an example from "core" Python:
以“核心”Python 为例：
In [175]: L = [2, 3, 1, 5]
In [176]: L.sort()
In [177]: print(L)
[1, 2, 3, 5]
It's the same for Pandas, as documented by Pandas.sort:
Pandas 也是如此，如Pandas.sort 所述：
Sort values and index labels by value, in place. For compatibility with ndarray API. No return value
就地按值对值和索引标签进行排序。为了与 ndarray API 兼容。无返回值
See also: What's the difference between Series.sort() and Series.order()?
回答by Christian Geier
As with most other python iterables, Series.sort() does actually return nothing but sorts the Series in place. See for example sorting a python list:
与大多数其他 python 迭代器一样， Series.sort() 实际上不返回任何内容，只是对系列进行了排序。例如，请参见对 Python 列表进行排序：
In [2]: foo = [5, 3, 9]
In [3]: foo.sort()
In [4]: foo
Out[4]: [3, 5, 9]
回答by YaOzI
In [1]: import pandas as pd
In [2]: s = pd.Series([3,4,0,3]).sort()
In [3]: s
Indeed In [3]will output nothing, as you can check:
IndeedIn [3]不会输出任何内容，您可以检查：
In [4]: type(s)
Out[4]: NoneType
The reason:
原因：
pd.Series([3,4,0,3])indeed return a pandas Seriestype object, BUT Series.sort()method return nothingbecause of inplacesorting. So the expression s = pd.Series([3,4,0,3]).sort(), sin LHS get nothing from RHS, thus In [3]: soutput nothing.
pd.Series([3,4,0,3])确实返回一个 PandasSeries类型的对象，但由于就地排序，Series.sort()方法不返回任何内容。所以表达式,在 LHS 从 RHS 什么也得不到，因此什么都不输出。s = pd.Series([3,4,0,3]).sort()sIn [3]: s
NOTE that:
注意：
After version 0.17.0, sorting by valuemethods pandas.Series.sort()and pandas.Series.order()are DEPRECATED, replaced by a unified pandas.Series.sort_values()API. See this answerfor more details.
版本之后0.17.0或更新版本，通过排序值的方法pandas.Series.sort()，并pandas.Series.order()已被弃用，取而代之的是一个统一的pandas.Series.sort_values()API。有关更多详细信息，请参阅此答案。
回答by Aniruddha Kalburgi
Both .sort()and order()functions are DEPRECATED
.sort_values()function is the replacement and here's the example on how to use it.
两者的.sort（）和顺序（）函数不推荐使用
.sort_values（）函数是更换，这里是如何使用它的例子。
Example:
例子：
import numpy as np
import pandas as pd
arr = np.array([1,3,4,2])
series = pd.Series(arr)
Ascending Order
Equivalent to .order() function from old versions.
升序
相当于旧版本的 .order() 函数。
ascending = series.sort_values()
Descending Order
Equivalent to .order(ascending=False)
降序
等效于 .order(ascending=False)
descending = series.sort_values(ascending=False)
In Place
Equivalent to .sort() from old versions.
就地
相当于旧版本的 .sort() 。
series.sort_values(inplace=True)
