CODEWITHSUNDEEP
python
madsthaks
madsthaks

Reputation: 2181

Attempting to replicate each row in a pandas dataframe a certain number of times

Sample Data:

X  | a | c
41 | 5 | 2
54 | 3 | 1

Transformation:

X  | a | c
41 | 5 | 2
41 | 5 | 2
41 | 5 | 2
41 | 5 | 2
41 | 5 | 2
41 | 5 | 2
41 | 5 | 2
54 | 3 | 1
54 | 3 | 1
54 | 3 | 1
54 | 3 | 1
54 | 3 | 1
54 | 3 | 1
54 | 3 | 1

I've done this in R using the following command:

data_final[rep(1:nrow(data_final),each=nrow(expand.grid(0:6)),]

but I'm not quite sure how I'd do this in python.

Any help would be much appreciated

Upvotes: 1

Views: 80

Answers (3)

cmaher
cmaher

Reputation: 5215

This can be done by using .repeat on your index, passing the resulting array to .reindex, and then resetting your index with drop=True:

df.reindex(df.index.repeat(6)).reset_index(drop=True)

#      X  a  c
# 0   41  5  2
# 1   41  5  2
# 2   41  5  2
# 3   41  5  2
# 4   41  5  2
# 5   41  5  2
# 6   54  3  1
# 7   54  3  1
# 8   54  3  1
# 9   54  3  1
# 10  54  3  1
# 11  54  3  1

Upvotes: 0

akrun
akrun

Reputation: 887128

We can try

pd.concat([dat]*6, ignore_index = True).sort_values('X').reset_index(drop = True)
#     X  a  c
#0   41  5  2
#1   41  5  2
#2   41  5  2
#3   41  5  2
#4   41  5  2
#5   41  5  2
#6   54  3  1
#7   54  3  1
#8   54  3  1
#9   54  3  1
#10  54  3  1
#11  54  3  1

Upvotes: 1

jezrael
jezrael

Reputation: 862671

pandas + numpy solution with numpy.repeat:

If all columns have same dtypes:

df = pd.DataFrame(np.repeat(df.values, 7, axis=0), columns=df.columns)
print (df)
     X  a  c
0   41  5  2
1   41  5  2
2   41  5  2
3   41  5  2
4   41  5  2
5   41  5  2
6   41  5  2
7   54  3  1
8   54  3  1
9   54  3  1
10  54  3  1
11  54  3  1
12  54  3  1
13  54  3  1

If not:

df = df.loc[np.repeat(df.index, 7)].reset_index(drop=True)
print (df)
     X  a  c  d
0   41  5  2  a
1   41  5  2  a
2   41  5  2  a
3   41  5  2  a
4   41  5  2  a
5   41  5  2  a
6   41  5  2  a
7   54  3  1  r
8   54  3  1  r
9   54  3  1  r
10  54  3  1  r
11  54  3  1  r
12  54  3  1  r
13  54  3  1  r

Upvotes: 3

Related Questions