Convert multiple boolean columns which names start with string `abc_` at once into integer dtype

Become part of the top 3% of the developers by applying to Toptal https://topt.al/25cXVn

--

Track title: CC E Schuberts Piano Sonata D 784 in A

--

Chapters
00:00 Question
00:28 Accepted answer (Score 6)
00:43 Answer 2 (Score 7)
01:16 Answer 3 (Score 5)
01:31 Thank you

--

Full question
https://stackoverflow.com/questions/4827...

--

Content licensed under CC BY-SA
https://meta.stackexchange.com/help/lice...

--

Tags
#python #pandas #dataframe

#avk47

ANSWER 1

Score 7

Option 1: converting all boolean (dtype == 'bool') columns

df.loc[:, df.dtypes.eq('bool')] = df.loc[:, df.dtypes.eq('bool')].astype(np.int8)

Option 2: if only those boolean columns that start with abc_ should be converted:

col_mask = df.dtypes.eq('bool') & df.columns.str.contains('^abc_')
df.loc[:, col_mask] = df.loc[:, col_mask].astype(np.int8)

Option 3: converting only by column names

df.loc[:, df.columns.str.match(r'^abc_.*$')] = \
    df.filter(regex=r'^abc_').astype(np.int8)

ACCEPTED ANSWER

Score 6

You can do this with filter and an in-place update.

df.update(df.filter(regex='^abc_').astype(int))

ANSWER 3

Score 5

By using str.contains

df.loc[:,df.columns.str.contains('abc_')]=df.loc[:,df.columns.str.contains('abc_')].astype(int)