Question

这是一个简单的数据框

import pandas as pd
import numpy as np
dates = pd. date_range(' 20130101' , periods=14)
data = pd.DataFrame({'a':[1,0,0,1,0,0,0,1,1,0,0,1,0,0],'b':[0,0,1,0,0,1,0,0,0,0,1,0,1,0]},index=dates)

现在我想在以下条件下添加列＆＃39; c＆＃39;

if a = 1, c = 1
if b = 1, c = 0
if a = 0 and b = 0, c = c.shift(1) 约束：不存在同时a = 1和b = 1的情况。

这是一个简单的问题，但很难解决......

有什么好主意吗？

Answer 1

你需要的IIUC：

data['c'] = np.where(data.a == 1, 1,
            np.where(data.b == 1, 0, np.nan))
print (data)
            a  b    c
2013-01-01  1  0  1.0
2013-01-02  0  0  NaN
2013-01-03  0  1  0.0
2013-01-04  1  0  1.0
2013-01-05  0  0  NaN
2013-01-06  0  1  0.0
2013-01-07  0  0  NaN
2013-01-08  1  0  1.0
2013-01-09  1  0  1.0
2013-01-10  0  0  NaN
2013-01-11  0  1  0.0
2013-01-12  1  0  1.0
2013-01-13  0  1  0.0
2013-01-14  0  0  NaN

然后我不确定是否需要bfill或ffill：

data['c'] = data['c'].bfill()
print (data)
            a  b    c
2013-01-01  1  0  1.0
2013-01-02  0  0  0.0
2013-01-03  0  1  0.0
2013-01-04  1  0  1.0
2013-01-05  0  0  0.0
2013-01-06  0  1  0.0
2013-01-07  0  0  1.0
2013-01-08  1  0  1.0
2013-01-09  1  0  1.0
2013-01-10  0  0  0.0
2013-01-11  0  1  0.0
2013-01-12  1  0  1.0
2013-01-13  0  1  0.0
2013-01-14  0  0  NaN

data['c'] = data['c'].ffill()
print (data)
            a  b    c
2013-01-01  1  0  1.0
2013-01-02  0  0  1.0
2013-01-03  0  1  0.0
2013-01-04  1  0  1.0
2013-01-05  0  0  1.0
2013-01-06  0  1  0.0
2013-01-07  0  0  0.0
2013-01-08  1  0  1.0
2013-01-09  1  0  1.0
2013-01-10  0  0  1.0
2013-01-11  0  1  0.0
2013-01-12  1  0  1.0
2013-01-13  0  1  0.0
2013-01-14  0  0  0.0

Answer 2

替代

data.assign(
    c=np.where(v.sum(1, keepdims=1), (np.diff(v[:, ::-1]) + 1) / 2, np.nan)
).ffill()

            a  b    c
2013-01-01  1  0  1.0
2013-01-02  0  0  1.0
2013-01-03  0  1  0.0
2013-01-04  1  0  1.0
2013-01-05  0  0  1.0
2013-01-06  0  1  0.0
2013-01-07  0  0  0.0
2013-01-08  1  0  1.0
2013-01-09  1  0  1.0
2013-01-10  0  0  1.0
2013-01-11  0  1  0.0
2013-01-12  1  0  1.0
2013-01-13  0  1  0.0
2013-01-14  0  0  0.0

如何在pandas数据帧中的列之间进行条件计算？

2 个答案: