CODEWITHSUNDEEP
pythonpandas
Nils Gudat
Nils Gudat

Reputation: 13800

Efficiently access property of objects in pandas column

Say I have the following DataFrame:

df = pd.DataFrame("x":[pd.Timestamp("2016-03-18")])

How can I access the properties of the datetime object stored in column x? That is, I want e.g.

df.x[0].weekofyear # returns 9

But for the whole column. Clearly, df.x.weekofyear wouldn't work, as df.x is a series and not a datetime object, and df.x.apply(pd.Timestamp.weekofyear) won't work either, as weekofyear is an attribute and not a function.

The solution I could think of was defining a function to access the attribute and then apply this, as:

def get_week(x):
    return x.weekofyear

df.x.apply(get_week) # returns series of weeks

It seems that this is a bit roundabout and verbose, and my experience is that whenever I do something like this, pandas already has a super efficient built-in way to do it - is there one for this case?

Upvotes: 3

Views: 2330

Answers (1)

EdChum
EdChum

Reputation: 394159

use vectorised .dt.weekofyear

So df['x'].dt.weekofyear will return for whole column:

In [119]:
df = pd.DataFrame({'dates': pd.date_range(dt.datetime(2016,1,1), dt.datetime(2016,1,20))})
df

Out[119]:
        dates
0  2016-01-01
1  2016-01-02
2  2016-01-03
3  2016-01-04
4  2016-01-05
5  2016-01-06
6  2016-01-07
7  2016-01-08
8  2016-01-09
9  2016-01-10
10 2016-01-11
11 2016-01-12
12 2016-01-13
13 2016-01-14
14 2016-01-15
15 2016-01-16
16 2016-01-17
17 2016-01-18
18 2016-01-19
19 2016-01-20

In [120]:    
df['dates'].dt.weekofyear

Out[120]:
0     53
1     53
2     53
3      1
4      1
5      1
6      1
7      1
8      1
9      1
10     2
11     2
12     2
13     2
14     2
15     2
16     2
17     3
18     3
19     3
Name: dates, dtype: int64

Upvotes: 3

Related Questions