Question

我有一个数据框：

df =>   A          B       C
      3.8314    60.6247   -1
      3.8167    60.6247   -2
      3.7524    60.6247   -1
      3.7407    60.6247   -1
      3.6939    60.7713   -1
      3.8899    60.7957   -2
      3.8723    60.7957   -3
      3.7144    60.7957   -1
      3.7904    62.4083   -7
      3.7758    62.4083   -1
      3.6676    62.4083   -6
      3.6588    62.4083   -6
      3.6471    62.4083   -5
      3.5828    62.6771   -6
      3.5681    62.6771   -1
      3.5272    62.6771   -7
      3.5418    62.7015   -1
      3.6383    62.9458   -7
      4.0010    63.3856   -2
      3.6997    63.3856   -2
      3.6822    63.3856   -2
      4.0185    63.4101   -2
      3.7027    63.9231   -2
      3.6851    63.9231   -3
      3.5535    63.9231   -3
      3.5389    63.9231   -3

如果B的值在两行或更多行的+/- 0.03范围内，并且这些行的A值落在彼此的+/- 0.026之内，我想取这些行的平均值来给出以下数据框：

df =>   A          B       C
      3.82405   60.6247   -1.5
      3.74655   60.6247   -1 
      3.84090   60.7835   -1.5   
      3.79335   60.7957   -2
      3.7831    62.4083   -4      
      3.65783   62.4083   -8.5
      3.57545   62.6771   -3.5
      3.5345    62.6771   -4  
      3.6383    62.9458   -7
      4.00975   63.39785  -2                
      3.69095   63.3856   -2                      
      3.6939    63.9231   -2.5            
      3.5462    63.9231   -3

关于如何做到这一点的任何想法？

Answer 1

试试这个。

df.groupby(((df.A.diff().abs().lt(0.026))&(df.B.diff().abs().lt(0.03))==False).cumsum()).mean()
Out[642]: 
           A        B         C
1   3.824050  60.6247 -1.500000
2   3.746550  60.6247 -1.000000
3   3.693900  60.7713 -1.000000
4   3.881100  60.7957 -2.500000
5   3.714400  60.7957 -1.000000
6   3.783100  62.4083 -4.000000
7   3.657833  62.4083 -5.666667
8   3.575450  62.6771 -3.500000
9   3.534500  62.6893 -4.000000
10  3.638300  62.9458 -7.000000
11  4.001000  63.3856 -2.000000
12  3.690950  63.3856 -2.000000
13  4.018500  63.4101 -2.000000
14  3.693900  63.9231 -2.500000
15  3.546200  63.9231 -3.000000

如何获取属于某个容差的数据框中的行的平均值

1 个答案: