Pandas: Create several rows from column that is a list -

let's have this:

df = pd.dataframe({'key':[1,2,3], 'type':[[1,3],[1,2,3],[1,2]], 'value':[5,1,8]})  key type        value 1   [1, 3]      5 2   [1, 2, 3]   1 3   [1]         8 

where 1 of columns contains list of items. create several rows each row contains multiple types.

ontaining this:

key type  value 1   1     5 1   3     5 2   1     1 2   2     1 2   3     1 3   1     8 

i've been playing apply axis=1 can't find way return more 1 row per row of dataframe. extracting different 'types' , looping-concatenating seems ugly.

any ideas? thanks!!!

import itertools import pandas pd import numpy np  def melt_series(s):     lengths = s.str.len().values     flat = [i in itertools.chain.from_iterable(s.values.tolist())]     idx = np.repeat(s.index.values, lengths)     return pd.series(flat, idx, name=s.name)   melt_series(df.type).to_frame().join(df.drop('type', 1)).reindex_axis(df.columns, 1) 

---

setup

df = pd.dataframe({'key':[1,2,3],                    'type':[[1,3],[1,2,3],[1,2]],                    'value':[5,1,8]}) df