Python using drop() method?

Question

Let's say I have a DataFrame with the following columns: TAG, ALIAS, COMMENT, TYPE.

Is there a way to drop all columns that are not COMMENT without having to type the following line?

df.drop(["TAG","ALIAS","TYPE"], 1)

Is there a way to type an if statement somewhere in there and drop anything that is not the column call e.g., COMMENT?

I've misread the question so I reopened. However, if you only want to select one column, why don't you just use `df['COMMENT']`? — ayhan, May 04 '17 at 21:15

Vaishali · Accepted Answer · 2017-05-04T23:05:33.023

3

If you know that you only want the column COMMENT, just for for

df = df['COMMENT']

If you are looking for various columns starting with COMMENT, say COMMENT1, COMMENT2 etc, you can use filter

df = df.filter(like = 'COMMENT')

As @piRsquared suggested, another method of selecting columns would be

df = df[['COMMENT']]

This would be especially needed if you want to select multiple columns

df = df[['COMMENT', 'COMMENT1']]

edited May 04 '17 at 23:05

answered May 04 '17 at 21:20

Vaishali

1

So I don't have add an answer `df = df[['COMMENT']]` – piRSquared May 04 '17 at 23:02

score 3 · Answer 2 · answered May 04 '17 at 23:06

3

I changed my mind on adding an answer. This is obnoxious as what A-Za-z and ayhan have said makes way more sense...

... However, that doesn't stop me from posting this

df.drop(df.columns.difference(['COMMENT']), 1)

answered May 04 '17 at 23:06

piRSquared

2 Answers2