python - Sublists in pandas -
i've got pandas dataframe, 1 of columns of looks this:
0 {u'funny': 2, u'useful': 0, u'cool': 0} 1 {u'funny': 370, u'useful': 487, u'cool': 296} 2 {u'funny': 2, u'useful': 17, u'cool': 1} 3 {u'funny': 1233, u'useful': 2452, u'cool': 1875} 4 {u'funny': 4, u'useful': 7, u'cool': 2} 5 {u'funny': 408, u'useful': 819, u'cool': 557} 6 {u'funny': 321, u'useful': 673, u'cool': 430} 7 nan 8 {u'funny': 0, u'useful': 8, u'cool': 0}
i want break column out separate columns, without iterating on of rows explicitly (there lot, plus it's bad practice).
dataframe.thiscolumn[1]["funny"]
returns 2, dataframe.thiscolumn[:]["funny"]
doesn't work....
how can break separate columns?
thanks.
you apply series constructor column:
in [11]: df out[11]: dics 0 {u'funny': 2, u'useful': 0, u'cool': 0} 1 {u'funny': 370, u'useful': 487, u'cool': 296} 2 {u'funny': 2, u'useful': 17, u'cool': 1} 3 {u'funny': 1233, u'useful': 2452, u'cool': 1875} 4 {u'funny': 4, u'useful': 7, u'cool': 2} 5 {u'funny': 408, u'useful': 819, u'cool': 557} 6 {u'funny': 321, u'useful': 673, u'cool': 430} 7 {u'funny': 0, u'useful': 8, u'cool': 0} in [12]: df['dics'].apply(pd.series) out[12]: cool funny useful 0 0 2 0 1 296 370 487 2 1 2 17 3 1875 1233 2452 4 2 4 7 5 557 408 819 6 430 321 673 7 0 0 8
however, think it's best split these when constructing dataframe, rather once in dataframe.
Comments
Post a Comment