我有以下pandas DataFrame:
import pandas as pd
import numpy as np
df = pd.DataFrame({"first_column": [0,1,0]})
>>> df
first_column
0 0
1 0
2 0
3 1
4 1
5 1
6 0
7 0
8 1
9 1
10 0
11 0
12 0
13 0
14 1
15 1
16 1
17 1
18 1
19 0
20 0
first_column是0和1的二进制列.存在连续的“簇”,它们总是成对出现至少两个.
我的目标是创建一个列“计算”每组的行数:
>>> df
first_column counts
0 0 0
1 0 0
2 0 0
3 1 3
4 1 3
5 1 3
6 0 0
7 0 0
8 1 2
9 1 2
10 0 0
11 0 0
12 0 0
13 0 0
14 1 5
15 1 5
16 1 5
17 1 5
18 1 5
19 0 0
20 0 0
这听起来像df.loc()的工作,例如df.loc [df.first_column == 1] ……某事
我只是不确定如何考虑每个“群集”,以及如何用“行数”标记每个独特的群集.
怎么会这样做?
解决方法
这是NumPy的
cumsum和
bincount的一种方法 –
def cumsum_bincount(a):
# Append 0 & look for a [0,1] pattern. Form a binned array based off 1s groups
ids = a*(np.diff(np.r_[0,a])==1).cumsum()
# Get the bincount,index into the count with ids and finally mask out 0s
return a*np.bincount(ids)[ids]
样品运行 –
In [88]: df['counts'] = cumsum_bincount(df.first_column.values)
In [89]: df
Out[89]:
first_column counts
0 0 0
1 0 0
2 0 0
3 1 3
4 1 3
5 1 3
6 0 0
7 0 0
8 1 2
9 1 2
10 0 0
11 0 0
12 0 0
13 0 0
14 1 5
15 1 5
16 1 5
17 1 5
18 1 5
19 0 0
20 0 0
将前6个元素设置为1,然后测试 –
In [101]: df.first_column.values[:5] = 1
In [102]: df['counts'] = cumsum_bincount(df.first_column.values)
In [103]: df
Out[103]:
first_column counts
0 1 6
1 1 6
2 1 6
3 1 6
4 1 6
5 1 6
6 0 0
7 0 0
8 1 2
9 1 2
10 0 0
11 0 0
12 0 0
13 0 0
14 1 5
15 1 5
16 1 5
17 1 5
18 1 5
19 0 0
20 0 0
