I was trying to convert a dictionary of {'a': [1, 2, 3], 'b': [4, 5]} into:
0 | 1 |
|:-----------|------------:|
| a | 1 |
| a | 2 |
| a | 3 |
| b | 4 |
| b | 5 |
But haven't found any way to achieve that without modifying the dictionary itself. Is there an easier way to do that? Any help will be really appreciated! Thank you!
you can expand the dict into list of tuples and then convert into df as follows;
ddd = {'a': [1, 2, 3], 'b': [4, 5]}
df = pd.DataFrame([(k,v) for k in ddd for v in ddd[k]])
Using pd.DataFrame.from_dict
pd.DataFrame.from_dict(d,'index').stack().reset_index(level=0)
Out[194]:
level_0 0
0 a 1.0
1 a 2.0
2 a 3.0
0 b 4.0
1 b 5.0
You can use numpy.concatenate and numpy.repeat:
import numpy as np
d = {'a': [1, 2, 3], 'b': [4, 5]}
values = np.concatenate(list(d.values()))
lens = list(map(len, d.values()))
res = pd.DataFrame({0: np.repeat(list(d), lens),
1: values})
Result
0 1
0 a 1
1 a 2
2 a 3
3 b 4
4 b 5
This seems tougher than it should be, but remember you need to create arrays for each series / column. In essence, we need to repeat dictionary keys for series 0 and concatenate dictionary values for series 1.