CODEWITHSUNDEEP
pythonpandas
psowa001
psowa001

Reputation: 823

How to extract initials from name in DataFrame

I want to extract initials from user names and next replace names with initials

s = pd.DataFrame(['Robert Flitz', 'Hans Malek', 'Marek J. Beer'])

when i try use this function 

def initials(name):
    init = ""
    for n in name.str.split():
        for i in n:
            init+=i[0]
    return init

it returns single string

initials(s[0])
'RFHMMJB'

I want to look it like

'RF'
'HM'
'MJB'

Any ideas?

Upvotes: 4

Views: 1548

Answers (4)

Allen Qin
Allen Qin

Reputation: 19957

You can also use a regular expression:

s[0].str.replace('[^A-Z]', '')
0     RF
1     HM
2    MJB

Upvotes: 4

Jim Eisenberg
Jim Eisenberg

Reputation: 1500

A more pandas approach would probably be:

def initials(name):
    init = ""
    for n in name.split():
        init+=n[0]
    return init
s[1] = s[0].apply(initials)

this gives:

               0    1
0   Robert Flitz   RF
1     Hans Malek   HM
2  Marek J. Beer  MJB

You can then get a list of initials with list(s[1]) if you need it in that format.

Upvotes: 0

akshika47
akshika47

Reputation: 43

You can edit your initials function. Append a new line to the init variable after the end of the inner loop.

def initials(name):
    init = ""
    for n in name.str.split():
        for i in n:
            init+=i[0]
        init+= "\n"
    return init

Upvotes: 0

Kshitij Saxena
Kshitij Saxena

Reputation: 940

s[0].apply(lambda x: ''.join(i[0] for i in x.split()))

Output:

0    RF
1    HM
2    MJB
Name: 0, dtype: object

Change it numpy array

np.array(s[0].apply(lambda x: ''.join(i[0] for i in x.split())))

Output:

array(['RF', 'HM', 'MJB'], dtype=object)

Upvotes: 1

Related Questions