Question

我有一个带月份列的df字符串：

Month      Value     Details
January      10        H12
April        12        J11
June         13        K03
May          08        Y21

我需要根据April to March模型来安排月份。哪种方法最简单？所需结果：

Month      Value     Details
April        12        J11
May          08        Y21
June         13        K03
January      10        H12

Answer 1

如果需要解决方案，如果缺少字典中的某些月份和所有月份，则可以正常工作，请使用Series.map和Series.argsort，然后按DataFrame.iloc更改顺序：

d = {'April':1,'May':2,'June':3,'July':4,'January':12}

df = df.iloc[df['Month'].map(d).argsort()]
print (df)
     Month  Value Details
1    April     12     J11
3      May      8     Y21
2     June     13     K03
0  January     10     H12

或使用ordered categoricals：

#add another months
c = ['April','May','June','July','January']
df['Month'] = pd.Categorical(df['Month'], categories=c, ordered=True)

df = df.sort_values('Month')
print (df)
     Month  Value Details
1    April     12     J11
3      May      8     Y21
2     June     13     K03
0  January     10     H12

Answer 2

您可以使用df.loc或df.reindex：

In [2048]: new_order = ['April','May','June','January']

In [2051]: df.set_index('Month', inplace=True)

In [2071]: df.loc[new_order].reset_index()    
Out[2071]: 
     Month  Value Details
0    April     12     J11
1      May      8     Y21
2     June     13     K03
3  January     10     H12

OR

In [2051]: df.reindex(new_order, axis=0).reset_index() 
Out[2071]: 
         Month  Value Details
    0    April     12     J11
    1      May      8     Y21
    2     June     13     K03
    3  January     10     H12

在熊猫中排列字符串月份行的最简单方法是什么

2 个答案: