python - 按位 "&"ing 未指定数量的数据帧掩码？

Question

所以我习惯于像这样组合数据帧掩码: final_mask = mask1 & mask2

但是如果我想组合多个蒙版怎么办？例如，列表: [掩码1，掩码2，掩码3，掩码4，...，掩码20]

Answer 1

您可以使用pandas cookbook solution ，最后一段为 reduce :

df = pd.DataFrame({'AAA' : [4,5,6,7], 'BBB' : [10,20,30,40],'CCC' : [100,50,-30,-50]})
print (df)
   AAA  BBB  CCC
0    4   10  100
1    5   20   50
2    6   30  -30
3    7   40  -50

mask1 = df.AAA <= 5.5
mask2 = df.BBB == 10.0
mask3 = df.CCC > -40.0

masks = [mask1, mask2, mask3]
mask = functools.reduce(lambda x,y: x & y, masks)

print (df[mask])
   AAA  BBB  CCC
0    4   10  100

来自ayhan的另一个解决方案关于 1d 掩码的评论(掩码是 Series):

mask = np.logical_and.reduce(masks)

print (df[mask])
   AAA  BBB  CCC
0    4   10  100

如ayhan指出，第一个解决方案也适用于 2D 蒙版:

mask1 = df <= 5.5
mask2 = df < 1.0
mask3 = df > -40.0

masks = [mask1, mask2, mask3]
mask = functools.reduce(lambda x,y: x & y, masks)
print (mask)
     AAA    BBB    CCC
0  False  False  False
1  False  False  False
2  False  False   True
3  False  False  False

<小时/>

mask = np.logical_and.reduce(masks)
print (mask)

ValueError: cannot copy sequence with size 4 to array axis with dimension 3