select DataFrame columns using boolean

Question

im quite new to python. i just wanna know what is the best way to select certain columns based on values on a row.

c = pd.DataFrame(np.arange(12).reshape(4,3))
c
Out[1]: 
   0   1   2
0  0   1   2
1  3   4   5
2  6   7   8
3  9  10  11

say i want to select the columns whose second row's value is larger than 3.

It seems c.loc[:,c.loc[1]>3] works for me. Is it the most efficient way? or any other way to achieve this?

cs95 · Accepted Answer · 2018-08-04 23:28:22Z

1

Is it the most efficient way?

Yes.

any other way to achieve this?

c.reindex(c.columns[c.loc[1] > 3], axis=1)

    1   2
0   1   2
1   4   5
2   7   8
3  10  11

Arguably not as good.

answered Aug 4, 2018 at 23:28

cs95

406k106 gold badges744 silver badges797 bronze badges

Sign up to request clarification or add additional context in comments.

1 Answer 1