solves #issue82 by fixing the comparison between a Series and a list

[ThijsVroegh]

Issue

Fixes #issue82.

A previous bugfix around >>combined_df = combined_df[combined_df['category'] not in ['?', 'nan']] << was wrong in the sense that it was trying to check if the entire combined_df['category'] Series is "not in" the list ['?', 'nan']. This boils down to a comparison between a Series and a list, which led to a ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().

Description of changes

To properly filter the DataFrame by checking if each element in the category column is not in the list ['?', 'nan'], now the .isin() method is used in combination with the negation operator '~'

Includes

• Code changes
• Tests
• Documentation

[eriktks]

I have tested the issue:

• on the branches "master" and this branch,
• using the workflow displayed below
• with dutch_haliday_action_list.csv as input for the File widget
• letters.tab as input for the widget Corpus

Result: the issue occurred with the master branch but not with the current branch which shows that the issue was present but is now solved

Tip: the changed code is reasonable but I would remove the commented lines of code before merging

[ThijsVroegh]

@eriktks Thanks; I removed the old, commented-out lines of code.