Dropping column from one dataframe based on column value of second dataframe in pandas

Question

I have 2 dataframes df1 and df2, both consisting of 8 columns each as seen below :

**df1**
╔══════════════════════════════════════════════════════════╗
║John ║ Mark ║ Jane ║ Natasha ║ Oliver ║ Tony ║ Judd ║ Ron ║
╚══════════════════════════════════════════════════════════╝


**df2**
╔══════════════════════════════════════════════════╗
║True ║True ║False ║True ║False ║False ║False ║True║
╚══════════════════════════════════════════════════╝

df1 has columns that are names of different people while df2 has column names that are boolean values. What I want to do is drop all columns in df1 that have a corresponding value of False in df2 . So the resulting output should look like this :

**output**
╔════════════════════════════╗
║John ║ Mark ║ Natasha ║ Ron ║
╚════════════════════════════╝

I am reading both the dataframes from csv files.

Any and all help would be appreciated.

Note : The actual dataframes have 500 columns each. Used 8 as an example for visualization purposes as well as to show that the dataframes have equal number of columns

Thanks in advance

Answer 1

You can, using basic indexing. However, when you parse your df2 , the column names have duplicates and are altered, so it requires a bit of cleaning.

Setup

names = ['John', 'Mark', 'Jane', 'Natasha', 'Oliver', 'Tony', 'Judd', 'Ron']
cols = ['TRUE', 'TRUE.1', 'FALSE', 'FALSE.1', 'TRUE.2', 'FALSE.2', 'FALSE.3', 'TRUE.3']

df1 = pd.DataFrame(columns=names)
df2 = pd.DataFrame(columns=cols)

---

df1.loc[:, df2.columns.str.contains('TRUE')]

Empty DataFrame
Columns: [John, Mark, Oliver, Ron]
Index: []