如何获取不在索引中（但在这些值之间）的pandas中的数据元素？

Question

I have a Series of discrete states a system can be set to. The setting is done at discrete time steps, eg

import pandas as pd
states = pd.Series([1, 1, 0, 0, 1, 0, 1, 0], index=pd.date_range(
'2015-01-01', freq='2Min', periods=8))

This results in:

2015-01-01 00:00:00    1
2015-01-01 00:02:00    1
2015-01-01 00:04:00    0
2015-01-01 00:06:00    0
2015-01-01 00:08:00    1
2015-01-01 00:10:00    0
2015-01-01 00:12:00    1
2015-01-01 00:14:00    0
Freq: 2T, dtype: int64

This means the system is in state 1 in the first four minutes, starting from minute 5 in state 0 and so on. Now it is necessary to answer in which state the system is at 3 minutes and 34 seconds. Quite obvious states['2015-01-01 00:03:45'] won't work but raise a key error.

Is there an elegant solution to do so? I assume I could do

index = np.where(states.index < '2015-01-01 00:03:45')[-1][-1]
state = states.iloc[index]

to get these data but I consider this as rather ugly and I'd be happy for any suggestions for improvement.

Answer 1

Since your Series has a DatetimeIndex, you could use the asof method :

In [11]: states.asof('2015-01-01 00:03:45')
Out[14]: 1

In [15]: states.asof('2015-01-01 00:04:00')
Out[15]: 0