CODEWITHSUNDEEP
pythonpandasdatetime
Eoin Dowling
Eoin Dowling

Reputation: 433

Obtain the values of another Series by the timestamps associated with a filtered Series

I have two <class 'pandas.core.series.Series'> which I will refer to as a and b. len(a) returns 2400, as does b. I filtered a by some predicate, e.g. a[lambda x: x == 0] and this will yield a new series where the (timestamp) indices of a are preserved.

time
2014-06-29 17:00:00    0.0
2014-06-29 18:00:00    0.0
2014-06-29 19:00:00    0.0
2014-06-29 20:00:00    0.0
2014-06-29 21:00:00    0.0
                      ... 
2014-10-15 22:00:00    0.0
2014-10-15 23:00:00    0.0
2014-10-16 00:00:00    0.0
2014-10-16 01:00:00    0.0
2014-10-16 02:00:00    0.0
Name: response, Length: 1323, dtype: float64

What I am seeking to do, is to look-up the values associated with these indices (timestamps) on b.

Upvotes: 1

Views: 55

Answers (1)

jezrael
jezrael

Reputation: 863166

IIUC use Index.isin for filter by DatetimeIndexes:

new = a[lambda x: x == 0]

#same working
new = a[a == 0]
b[b.index.isin(new.index)]

Upvotes: 1

Related Questions