How to drop Pandas dataframe columns based on another boolean array's column values?

Question

I am trying to drop the columns of a Pandas Dataframe based on the value of the columns of a second Boolean array (that has the same length).

For example suppose the pandas dataframe

[value1, value2, value3]

and a boolean array

[True, False, True].

The result of the dropping operation on the Pandas dataframe would be

[value1, value3] 

Notice that we dropped the second column that has value False on the boolean array.

I guess something like this would work fine:

for i in range(len(boolean_array)):
    if boolean_array[i] == False:
        df = df.drop(df.columns[i], axis=1)

but I am pretty sure there is an easier way.

Answer 1

Use boolean indexing with DataFrame.loc for filter columns by boolean mask:

df = pd.DataFrame({'value1':[1,2],
                   'value2':[10,20],
                   'value3':[100,200]})

arr = np.array([True, False, True])

df1 = df.loc[:, arr]
print (df1)
   value1  value3
0       1     100
1       2     200