I want to merge a Two columns in a database:
The two columns are
A
1
2
Na
Na
B
Na
Na
C
Na
Output
D
1
2
C
Na
Add a comment
|
2 Answers
You can use combine_first or fillna:
df['D'] = df['A'].combine_first(df['B'])
Or:
df['D'] = df['A'].fillna(df['B'])
Notice:
If one column is numeric and another not, get mixed types and some functions can be broken.
2 Comments
Abul
Thank you so much, but when I am trying to save to excel. It works fine and gets saved but in excel a warning appears in that column saying "Number formatted as text" How to get rid of it, any help?
jezrael
This is exactly what say notice - it is not recommnded, but if really need it, ignore warning.
Use combine_first
df.A.combine_first(df.B)
Consider the dataframe df
df = pd.DataFrame(dict(A=[1, 2, np.nan, np.nan], B=[np.nan, np.nan, 'C', np.nan]))
Then
df.A.combine_first(df.B)
0 1
1 2
2 C
3 NaN
Name: A, dtype: object
You can assign it to a new column 'C'
df.assign(C=df.A.combine_first(df.B))
A B C
0 1.0 NaN 1
1 2.0 NaN 2
2 NaN C C
3 NaN NaN NaN
2 Comments
Abul
Thanks for editing my question and making it presentable. Any idea how to convert the mixed types into one single type?
piRSquared
@Abul yes! but this question was about merging columns. I suggest you ask another question. We appreciate questions because they present more opportunities to earn upvotes. :-)