如何在 pandas dataframe 中过滤小写的行和单词？

Question

Hi I would like to know how to select rows which contains lower cases in the following dataframe:

ID     Name   Note
1      Fin    there IS A dog outside
2      Mik    NOTHING TO DECLARE
3      Lau    no house

What I would like to do is to filter rows where Note column contains at least one word in lower case:

ID     Name   Note
1      Fin    there IS A dog outside
3      Lau    no house

and collect in a list all the words in lower case: my_list=['there','dog','outside','no','house']

I have tried to filter rows is:

df1=df['Note'].str.lower()

For appending words in the list, I think I should first tokenise the string, then select all the terms in lower case. Am I right?

Answer 1

Use Series.str.contains for filter at least one lowercase character in boolean indexing :

df1 = df[df['Note'].str.contains(r'[a-z]')]
print (df1)
   ID Name                    Note
0   1  Fin  there IS A dog outside
2   3  Lau                no house

And then Series.str.extractall for extract lowercase words:

my_list = df1['Note'].str.extractall(r'(\b[a-z]+\b)')[0].tolist()
print (my_list)
['there', 'dog', 'outside', 'no', 'house']

Or use list comprehension with split sentences and filter by islower :

my_list = [y for x in df1['Note'] for y in x.split() if y.islower()]
print (my_list)
['there', 'dog', 'outside', 'no', 'house']