Question

我得到了以下DataFrame：

   r1       Angle1          r2        Angle      Res         System
0   1.250590    5.156790    5.724081  7.712978e+06   8.874775     #EN1
1   1.246085    5.678428    5.720825  3.353667e+04   8.813887     #EN2
2   1.250148    5.205040    5.700498  3.564115e+02   8.865881     #EN1
3   1.241692    6.256054    5.708635  8.518839e+06   8.828826     #EN3
4   1.247089    5.556371    5.718326  9.315014e+04   8.813084     #EN1
5   1.854719    0.936551    1.186143  1.853106e+01   8.669692     #EN2
6   1.260329    4.225127    5.687622  5.435705e+01   9.223529     #EN3
7   1.251378    5.072078    5.756885  1.449325e+01   8.893499     #EN2
8   1.037451   39.403842    1.340242  2.438089e+04  14.509603     #EN1

我想根据最后一个系统对其进行排序。我试图完成它：

    custom_dict = {'#EN1':0, '#EN2':1, '#EN3':2}
    s = df_ALL['System'].apply(lambda x: custom_dict[x])
    s.sort()
    df_ALL.set_index(s.index).sort()

但是我得到了相同的DataFrame作为回报。最重要的是弹出警告：

FutureWarning: sort is deprecated, use sort_values(inplace=True) for INPLACE sorting
  s.sort()
FutureWarning: sort(....) is deprecated, use sort_index(.....)
  df_ALL = df_ALL.set_index(s.index).sort()

愿任何人详细说明我做错了吗？

Answer 1

您似乎需要sort_values：

df = df.sort_values('System')
print (df)
         r1     Angle1        r2         Angle        Res System
0  1.250590   5.156790  5.724081  7.712978e+06   8.874775   #EN1
2  1.250148   5.205040  5.700498  3.564115e+02   8.865881   #EN1
4  1.247089   5.556371  5.718326  9.315014e+04   8.813084   #EN1
8  1.037451  39.403842  1.340242  2.438089e+04  14.509603   #EN1
1  1.246085   5.678428  5.720825  3.353667e+04   8.813887   #EN2
5  1.854719   0.936551  1.186143  1.853106e+01   8.669692   #EN2
7  1.251378   5.072078  5.756885  1.449325e+01   8.893499   #EN2
3  1.241692   6.256054  5.708635  8.518839e+06   8.828826   #EN3
6  1.260329   4.225127  5.687622  5.435705e+01   9.223529   #EN3

但如果需要按dict进行自定义排序，请使用ordered categorical：

custom_dict = {'#EN1':0, '#EN2':2,'#EN3':1}
#get order by sorted values of dict
cat = sorted(custom_dict, key=custom_dict.get)
print (cat)
['#EN1', '#EN3', '#EN2']

df['System'] = df['System'].astype('category', categories=cat, ordered=True)
df = df.sort_values('System')
print (df)
         r1     Angle1        r2         Angle        Res System
0  1.250590   5.156790  5.724081  7.712978e+06   8.874775   #EN1
2  1.250148   5.205040  5.700498  3.564115e+02   8.865881   #EN1
4  1.247089   5.556371  5.718326  9.315014e+04   8.813084   #EN1
8  1.037451  39.403842  1.340242  2.438089e+04  14.509603   #EN1
3  1.241692   6.256054  5.708635  8.518839e+06   8.828826   #EN3
6  1.260329   4.225127  5.687622  5.435705e+01   9.223529   #EN3
1  1.246085   5.678428  5.720825  3.353667e+04   8.813887   #EN2
5  1.854719   0.936551  1.186143  1.853106e+01   8.669692   #EN2
7  1.251378   5.072078  5.756885  1.449325e+01   8.893499   #EN2

在DataFrame中对列进行排序

1 个答案: