CODEWITHSUNDEEP
pythonpandasdataframe
David Jackson
David Jackson

Reputation: 101

python splitting the rows and columns

I am totally new to using python I am using a data frame in which one row needs to be split into two rows. Both the rows will have same data except for one column

Upvotes: 1

Views: 110

Answers (3)

ansev
ansev

Reputation: 30920

Another approach:

new_df = (df.reindex(df.index.repeat(len(df.columns)-1))
            .assign(Column4 = df[df.columns.difference(['Column1'])].stack().values))
print(new_df)
   Column1 Column2 Column3 Column4
0        1      A1      A2      A1
0        1      A1      A2      A2
1        2      B1      B2      B1
1        2      B1      B2      B2

Upvotes: 1

anky
anky

Reputation: 75080

here is another way without apply using df.stack , series.droplevel followed by df.join:

m = df.set_index('Column1')
m = m.join(m.stack().droplevel(1).to_frame('Column4')).reset_index()

print(m)

   Column1 Column2 Column3 Column4
0        1      A1      A2      A1
1        1      A1      A2      A2
2        2      B1      B2      B1
3        2      B1      B2      B2

Upvotes: 2

Sergey Bushmanov
Sergey Bushmanov

Reputation: 25189

Suppose we have a df:

df
   Column1 Column2 Column3
0        1      A1      A2
1        2      B1      B2

Then we can do:

df["Column4"] = df.apply(lambda x: [x[1],x[2]], axis=1)
df = df.explode("Column4")
print(df)

   Column1 Column2 Column3 Column4
0        1      A1      A2      A1
0        1      A1      A2      A2
1        2      B1      B2      B1
1        2      B1      B2      B2

If contents of Column1 is an issue we can go one step further:

df["Column1"] = range(1, df.shape[0]+1)
print(df)

   Column1 Column2 Column3 Column4
0        1      A1      A2      A1
0        2      A1      A2      A2
1        3      B1      B2      B1
1        4      B1      B2      B2

Upvotes: 3

Related Questions