Creating dummy variables from a string column in pandas-CodePudding

So I have a pandas df as follows and my goal is to take the MATCHUP column and make it several more dummy columns.

INDICATOR MATCHUP 
1         [   "APPLE",   "GRAPE" ]
1         [   "APPLE",   "GRAPE" ]
0         [   "GRAPE",   "BANANA" ]
0         [   "PEAR",   "ORANGE" ]
1         [   "ORANGE",   "APPLE" ]

Here's a dict of how it looks:

{'INDICATOR': [1, 1, 0, 0, 1],
 'MATCHUP': ['[   "APPLE",   "GRAPE" ]',
  '[   "APPLE",   "GRAPE" ]',
  '[   "GRAPE",   "BANANA" ]',
  '[   "PEAR",   "ORANGE" ]',
  '[   "ORANGE",   "APPLE" ]']}

So given this df, I would like to create some dummy variables to identify if a value appears in the MATCHUP.

Final outcome:

INDICATOR MATCHUP                    APPLE GRAPE BANANA PEAR ORANGE
1         [   "APPLE",   "GRAPE" ]   1     1     0      0    0 
1         [   "APPLE",   "GRAPE" ]   1     1     0      0    0
0         [   "GRAPE",   "BANANA" ]  0     1     1      0    0
0         [   "PEAR",   "ORANGE" ]   0     0     0      1    1
1         [   "ORANGE",   "APPLE" ]  1     0     0      0    1

Is there a way to accomplish this using pandas? I attempted to accomplish this using this but I think the spacing in the MATCHUP column make this method unviable.

CodePudding user response：

Check explode with str.get_dummies

import ast
df = df.join(df['MATCHUP'].map(ast.literal_eval).explode().str.get_dummies().groupby(level=0).sum())