Let's say I have a data frame called df
x count
d 2
e 3
f 2
Count would be the counter column and the # times I want it to repeat.
How would I expand it to make it
x count
d 2
d 2
e 3
e 3
e 3
f 2
f 2
I've already tried numpy.repeat(df,df.iloc['count']) and it errors out
You can use np.repeat()
import pandas as pd
import numpy as np
# your data
# ========================
df
x count
0 d 2
1 e 3
2 f 2
# processing
# ==================================
np.repeat(df.values, df['count'].values, axis=0)
array([['d', 2],
['d', 2],
['e', 3],
['e', 3],
['e', 3],
['f', 2],
['f', 2]], dtype=object)
pd.DataFrame(np.repeat(df.values, df['count'].values, axis=0), columns=['x', 'count'])
x count
0 d 2
1 d 2
2 e 3
3 e 3
4 e 3
5 f 2
6 f 2
You could use .loc with repeat like
In [295]: df.loc[df.index.repeat(df['count'])].reset_index(drop=True)
Out[295]:
x count
0 d 2
1 d 2
2 e 3
3 e 3
4 e 3
5 f 2
6 f 2
Or, using pd.Series.repeat you can
In [278]: df.set_index('x')['count'].repeat(df['count']).reset_index()
Out[278]:
x count
0 d 2
1 d 2
2 e 3
3 e 3
4 e 3
5 f 2
6 f 2
