在一定条件下填充pandas数据框列

Question

I have the following data frame with the index as date time stamps and I would like to create an extra column when p2>p1 set the p3('new column') to 1 otherwise set it to 0. I do not want to have to loop. I tried to use mask and where but i DID NOT GET IT RIGHT:

                             p1   p2
2018-01-15 07:15:00         -1.0  0.4
2018-01-15 07:30:00         -1.0  0.4
2018-01-15 07:45:00         -3.0  0.4
2018-01-15 08:00:00         -3.0  0.3
2018-01-15 08:15:00         -2.0  0.3
2018-01-15 08:30:00         -2.0  0.3
2018-01-15 08:45:00         -2.0  0.2
2018-01-15 09:00:00         -2.7  0.1
2018-01-15 09:15:00         -2.7  0.0
2018-01-15 09:30:00         -2.7  0.0
2018-01-15 09:45:00         -3.7  0.3
2018-01-15 10:00:00         -3.0  0.3
2018-01-15 10:15:00         -2.0  0.3
2018-01-15 10:30:00         -1.0  0.3

Answer 1

You can convert boolean mask to integers, True to 1 and False to 0 :

df['p3'] = (df.p2>df.p1).astype(int)
print (df)
                      p1   p2  p3
2018-01-15 07:15:00  1.0  0.4   0 <-changed first value in p1
2018-01-15 07:30:00 -1.0  0.4   1
2018-01-15 07:45:00 -3.0  0.4   1
2018-01-15 08:00:00 -3.0  0.3   1
2018-01-15 08:15:00 -2.0  0.3   1
2018-01-15 08:30:00 -2.0  0.3   1
2018-01-15 08:45:00 -2.0  0.2   1
2018-01-15 09:00:00 -2.7  0.1   1
2018-01-15 09:15:00 -2.7  0.0   1
2018-01-15 09:30:00 -2.7  0.0   1
2018-01-15 09:45:00 -3.7  0.3   1
2018-01-15 10:00:00 -3.0  0.3   1
2018-01-15 10:15:00 -2.0  0.3   1
2018-01-15 10:30:00 -1.0  0.3   1