I have a data frame as the one generated by the script below - bringing in dataframe "data".
Ideally I would like to generate a new dataframe that combines the id and a sequence of 1 : value.
d = {'id': ['a', 'b','c'], 'value': [1, 2,1]}
data = pd.DataFrame(data=d)
data
This means that the ideal output would be:
|------|---------|
| ID | value |
|------|---------|
| a | 1 |
| b | 1 |
| b | 2 |
| c | 1 |
|------|---------|
Use Index.repeat by column value and reassign values by counter by GroupBy.cumcount:
#if not default RangeIndex
#data = data.reset_index(drop=True)
df = data.loc[data.index.repeat(data['value'])]
df['value'] = df.groupby(level=0).cumcount() + 1
df = df.reset_index(drop=True)
print (df)
id value
0 a 1
1 b 1
2 b 2
3 c 1
Alternative solution with DataFrame.assign:
df = (data.loc[data.index.repeat(data['value'])]
.assign(value=lambda x: x.groupby(level=0).cumcount() + 1)
.reset_index(drop=True))
