Using a lambda function for pandas.DataFrame boolean indexing reports TypeError

Question

I have a list of words called sowpods and I need to verify which combination of letters exist either as a word or within a word.

For example, if my letters are ['r', 't', 'e', 'f'] , one of the possible combinations is 're' which is within 'red' , therefore the word 'red' should be kept.

I already have some code that can figure out all of the possible combinations, but now I want to find how to add all of the words that fit the requirements to a list.

I have done the following:

import pandas as pd

sowpods = pd.read_csv('sowpods.csv', names=['Word'])

possible_combination = 'RE'
possible_words = pd.DataFrame([], columns=['Word'])

comb_in_word = lambda _: True if (possible_combination in _) else False # ------ line 8

sowpods_bool = sowpods['Word'].apply(comb_in_word) # --------------------------- line 10
possible_words.append(sowpods.loc[sowpods_bool, 'Word'])

But then I get:

 File "c:\tests.py", line 10, in <module>
    sowpods_bool = sowpods['Word'].apply(comb_in_word)
  File "C:\Python38-32\lib\site-packages\pandas\core\series.py", line 3848, in apply
    mapped = lib.map_infer(values, f, convert=convert_dtype)
  File "pandas\_libs\lib.pyx", line 2329, in pandas._libs.lib.map_infer
  File "c:\Users\lenovo\OneDrive\Prog\Projects\Scrabble\tests.py", line 8, in <lambda>
    comb_in_word = lambda _: True if possible_combination in _ else False
TypeError: argument of type 'float' is not iterable

I tested my lambda function in a more controlled environment and it worked fine, so I'm confident that the error's not coming from there.

I don't understand why I get this error when I'm not iterating through anything myself. I get that pandas is iterating through the DataFrame's column, but it shouldn't do an error where it's using floats instead of integers.

Edit:

[In]
print(sowpods.head())
[Out]
      Word
0      AA
1     AAH
2   AAHED
3  AAHING
4    AAHS

[In]
print(sowpods.dtypes)
[Out]
Word    object
dtype: object

Answer 1

In the list of words there were 'NA' and 'NULL' , which Pandas represented as NaN s. I had to specify keep_default_na=False :

sowpods = pd.read_csv('projects/scrabble/sowpods_en.csv', names=['Word'], keep_default_na=False)