Check if pandas column value is inside another column's list-CodePudding

I have a pandas column like this where amount is a string column:

id      amount    possible_amount
0        1.00       ['1.00', '2.00', '3.00']
1       45.00       ['100.00', '45.00']
2       37.00       ['29.00', '38.00']

I want to create a new column called 'match' whose value will be True if amount is in the possible_amount list and False otherwise. So expected results for example above is:

id      amount    possible_amount                     match
0        1.00       ['1.00', '2.00', '3.00']           True
1       45.00       ['100.00', '45.00']                True
2       37.00       ['29.00', '38.00']                 False

I've tried couple different ways, below being one of them. Also tried using str.contains() to no avail.

df['match'] = np.where(df['amount'].isin(df['possible_amount']), True, False)

But this only returns all False in match.

CodePudding user response：

Convert values to floats and compare in list comprehension:

df['match'] = [a in list(map(float, b)) for a, b in zip(df['amount'],df['possible_amount'])]
print (df)
   id  amount     possible_amount  match
0   0     1.0  [1.00, 2.00, 3.00]   True
1   1    45.0     [100.00, 45.00]   True
2   2    37.0      [29.00, 38.00]  False

Another idea, obviously slowier:

df['match'] = (df.explode('possible_amount')
                 .assign(possible_amount = lambda x: x['possible_amount'].astype(float),
                         new = lambda x: x['possible_amount'].eq(x['amount']))
                 .groupby(level=0)['new']
                 .any()
         )

print (df)
   id  amount     possible_amount  match
0   0     1.0  [1.00, 2.00, 3.00]   True
1   1    45.0     [100.00, 45.00]   True
2   2    37.0      [29.00, 38.00]  False