Applying function to pandas series receives series instead of row with swifter

 `def split_fk_keys(desc):
        desc_split = desc.split('_')

        if len(desc_split) > 1:
            return "{}{}".format(desc_split[0], "Key")
        else:
            return desc`

This one works in plain pandas:

   `relations['column_name'].apply(split_fk_keys)`

This one doesn’t work with swifter, relations[‘column_name’] returns a Series, and that series is passed as argument to the split_fk_keys method instead of each row in that series:

   `relations['column_name'].swifter.apply(split_fk_keys)`

Same problem when tried it this way (should make no difference as it’s actually same as above):

  `relations['column_name'].swifter.apply(lambda row: split_fk_keys(row))`

Or even:

  `relations.swifter.apply(lambda row: split_fk_keys(row['column_name']))`

[dabla]

I’ll test it in my spare time, currently the project is working using plain pandas without swifter. Nice to see there is progression on this issue, good work! 😃

[jmcarpenter2]

Didn’t mean to close it. Was hoping to get some feedback from the folks here first. Version 1.1.3 is now available! Please try and let me know 😃