如何按两列或更多列对python pandas中的数据帧进行排序？
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/17141558/
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
How to sort a dataFrame in python pandas by two or more columns?
提问by Rakesh Adhikesavan
Suppose I have a dataframe with columns a, band c, I want to sort the dataframe by column bin ascending order, and by column cin descending order, how do I do this?
假设我有一列的数据帧a，b并且c，我要通过列数据框排序b按升序，并经柱c降序排列，我该怎么办呢？
采纳答案by Andy Hayden
As of the 0.17.0 release, the sortmethod was deprecated in favor of sort_values. sortwas completely removed in the 0.20.0 release. The arguments (and results) remain the same:
从 0.17.0 版本开始，该sort方法已被弃用，而支持sort_values. sort在 0.20.0 版本中被完全删除。参数（和结果）保持不变：
df.sort_values(['a', 'b'], ascending=[True, False])
You can use the ascending argument of sort:
您可以使用 的升序参数sort：
df.sort(['a', 'b'], ascending=[True, False])
For example:
例如：
In [11]: df1 = pd.DataFrame(np.random.randint(1, 5, (10,2)), columns=['a','b'])
In [12]: df1.sort(['a', 'b'], ascending=[True, False])
Out[12]:
a b
2 1 4
7 1 3
1 1 2
3 1 2
4 3 2
6 4 4
0 4 3
9 4 3
5 4 1
8 4 1
As commented by @renadeen
正如@renadeen 所评论的
Sort isn't in place by default! So you should assign result of the sort method to a variable or add inplace=True to method call.
默认情况下没有排序！因此，您应该将 sort 方法的结果分配给变量或将 inplace=True 添加到方法调用中。
that is, if you want to reuse df1 as a sorted DataFrame:
也就是说，如果您想将 df1 重用为已排序的 DataFrame：
df1 = df1.sort(['a', 'b'], ascending=[True, False])
or
或者
df1.sort(['a', 'b'], ascending=[True, False], inplace=True)
回答by Kyle Heuton
As of pandas 0.17.0, DataFrame.sort()is deprecated, and set to be removed in a future version of pandas. The way to sort a dataframe by its values is now is DataFrame.sort_values
从 pandas 0.17.0 开始，DataFrame.sort()已弃用，并将在未来版本的 pandas 中删除。现在按值对数据框进行排序的方法是DataFrame.sort_values
As such, the answer to your question would now be
因此，您的问题的答案现在是
df.sort_values(['b', 'c'], ascending=[True, False], inplace=True)
回答by jpp
For large dataframes of numeric data, you may see a significant performance improvement via numpy.lexsort, which performs an indirect sort using a sequence of keys:
对于数字数据的大型数据帧，您可能会通过 看到显着的性能改进numpy.lexsort，它使用一系列键执行间接排序：
import pandas as pd
import numpy as np
np.random.seed(0)
df1 = pd.DataFrame(np.random.randint(1, 5, (10,2)), columns=['a','b'])
df1 = pd.concat([df1]*100000)
def pdsort(df1):
return df1.sort_values(['a', 'b'], ascending=[True, False])
def lex(df1):
arr = df1.values
return pd.DataFrame(arr[np.lexsort((-arr[:, 1], arr[:, 0]))])
assert (pdsort(df1).values == lex(df1).values).all()
%timeit pdsort(df1) # 193 ms per loop
%timeit lex(df1) # 143 ms per loop
One peculiarity is that the defined sorting order with numpy.lexsortis reversed: (-'b', 'a')sorts by series afirst. We negate series bto reflect we want this series in descending order.
一个特点是定义的排序顺序numpy.lexsort相反：首先(-'b', 'a')按系列排序a。我们否定系列b以反映我们希望按降序排列的系列。
Be aware that np.lexsortonly sorts with numeric values, while pd.DataFrame.sort_valuesworks with either string or numeric values. Using np.lexsortwith strings will give: TypeError: bad operand type for unary -: 'str'.
请注意，np.lexsort仅使用数字值排序，而pd.DataFrame.sort_values适用于字符串或数字值。np.lexsort与字符串一起使用将给出：TypeError: bad operand type for unary -: 'str'.
