How to display the table header with df.take()?

Question

The default behaviour for df.take() on Watson Studio Jupyter notebooks does not display the header for the table whereas df.show() does:

I would like to use df.take() because the output is much easier on the eye :) - Is it possible to get df.take() to display the header?

NOTE:

I don't want to use show().
I want to use take() and I want take() to display the header.

take will return you the Array[Row] but show is the api to give you the tabular view of the dataframe. What exactly your expected output. if you want to get limited records then you can use limit api with show — Chandan Ray, Aug 30 '18 at 18:19
In scala, `take` expects a number and outputs a number of rows. So, no, you can't use `take`. Why not just use Pandas `take` and call it a day? You will get the header. — Dr G., Aug 31 '18 at 05:17

score 1 · Answer 1 · answered Aug 30 '18 at 17:08

1

Limit can be used:

df.limit(5).show()

answered Aug 30 '18 at 17:08

pasha701

6,831
1
15
22

I want to display the header with `take()` - I don't want to output the table with `show()` – Chris Snow Aug 30 '18 at 19:23

score 1 · Accepted Answer · answered Aug 31 '18 at 05:45

No, it is not possible to let take show the headers. Header information is in the dataframe, but take returns a plain list or array, not a dataframe.

Python: DataFrame.take
Scala: Dataset.take

As suggested in another answer, limit would return a dataframe with header information, on which you could call show.

Python: Dataframe.limit
Scala: Dataset.limit

But since you insist on using take and on not using show, the answer is No. take is not meant to do what you want to accomplish.

How to display the table header with df.take()?

2 Answers2