CODEWITHSUNDEEP
pythonpandas
Foobard
Foobard

Reputation: 80

Creating a new columns in pandas based on the structure of the other

I have a table of data that looks like this:

F    M

f1   m1
f1   m2
f1   m3
f2   m1
f3   m1
f3   m2

and I need to add the third column made of objects of type str. So the table should look like so:

F    M   B

f1   m1  'b1'
f1   m2  'b2'
f1   m3  'b3'
f2   m1  'b1'
f3   m1  'b1'
f3   m2  'b2'

What would be the right way to do this in pandas? There are similar questions but I couldn't find a similar task and I can't figure out how to construct this B column.

EDIT:

Sorry for not being clear enough. All the values in the table (f{i} and m{i}) are some strings.

Upvotes: 0

Views: 26

Answers (1)

akuiper
akuiper

Reputation: 215107

Try groupby.cumcount which essentially gives the row number for each F, and then you can concatenate the row number with a letter such as b:

df['B'] = 'b'+df.groupby('F').cumcount().add(1).astype(str)

df
#    F   M   B
#0  f1  m1  b1
#1  f1  m2  b2
#2  f1  m3  b3
#3  f2  m1  b1
#4  f3  m1  b1
#5  f3  m2  b2

Upvotes: 1

Related Questions