Removing header index from dataframe

Question

I have a data-frame (df):

       0       1         2            3                    4     5
0   6894   JONES     STEVE            D  2017-01-03 00:00:00  None
1  13555   SMITH     JENNY            E  2017-04-01 00:00:00  None

I have a header called new_header that I am applying to the data-frame which looks like:

0                     Identifier
1                        Surname
2                  First name(s)
3                 Transferred to
4                  Transfer Date
5                         Status

I apply new_header with the following:

df.columns = new_header

and I get the following output:

0               Identifier Surname  ...          Transfer Date Status
0                     6894   JONES  ...    2017-01-03 00:00:00   None
1                    13555   SMITH  ...    2017-04-01 00:00:00   None

In the top left corner of df you can see a 0. How can I remove the 0 so my data-frame looks like:

                Identifier Surname  ...          Transfer Date Status
0                     6494   JONES  ...    2017-01-03 00:00:00   None
1                    13595   SMITH  ...    2017-04-01 00:00:00   None

what if you do df.columns = new_header.values ?

anky
– anky

2019-05-09 15:29:08 +00:00
Commented May 9, 2019 at 15:29 — anky
– anky, Commented May 9, 2019 at 15:29
df.columns.name = None should do it

Vaishali
– Vaishali

2019-05-09 15:30:09 +00:00
Commented May 9, 2019 at 15:30 — Vaishali
– Vaishali, Commented May 9, 2019 at 15:30
Yes! Thank you so much!

Stacey
– Stacey

2019-05-09 15:30:12 +00:00
Commented May 9, 2019 at 15:30 — Stacey
– Stacey, Commented May 9, 2019 at 15:30

piRSquared · Accepted Answer · 2019-05-09 15:46:41Z

You are attempting to rename columns with a pandas.Series object. When doing so, the name of the Series translates to the name of the resulting pandas.Index object. You can solve this in various ways.

Assign to `df.columns` something that doesn't have a name as suggested by @anky_91

df.columns = new_header.values

df

   Identifier Surname First name(s) Transferred to        Transfer Date Status
0        6894   JONES         STEVE              D  2017-01-03 00:00:00   None
1       13555   SMITH         JENNY              E  2017-04-01 00:00:00   None

Assign to `df.columns` a series whose name is `None`

df.columns = new_header.rename(None)

df

   Identifier Surname First name(s) Transferred to        Transfer Date Status
0        6894   JONES         STEVE              D  2017-01-03 00:00:00   None
1       13555   SMITH         JENNY              E  2017-04-01 00:00:00   None

Rename the axis after the fact

df.columns = new_header
df.rename_axis(None, 1)

   Identifier Surname First name(s) Transferred to        Transfer Date Status
0        6894   JONES         STEVE              D  2017-01-03 00:00:00   None
1       13555   SMITH         JENNY              E  2017-04-01 00:00:00   None

Use `rename` with the `new_header` as a renaming mapping

This assumes the index of new_header lines up with what is being renamed in df.
Also, I had to get tricky and make sure that they were the same type.

df.rename(columns=int).rename(columns=new_header)

   Identifier Surname First name(s) Transferred to        Transfer Date Status
0        6894   JONES         STEVE              D  2017-01-03 00:00:00   None
1       13555   SMITH         JENNY              E  2017-04-01 00:00:00   None

nassim · Accepted Answer · 2019-05-09 15:33:02Z

0

You can get or set the index by name

df.index.name = ''

see the docs

answered May 9, 2019 at 15:33

nassim

1,5811 gold badge16 silver badges28 bronze badges

Collectives™ on Stack Overflow

Removing header index from dataframe

2 Answers 2

Assign to `df.columns` something that doesn't have a name as suggested by @anky_91

Assign to `df.columns` a series whose name is `None`

Rename the axis after the fact

Use `rename` with the `new_header` as a renaming mapping

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Assign to df.columns something that doesn't have a name as suggested by @anky_91

Assign to df.columns a series whose name is None

Rename the axis after the fact

Use rename with the new_header as a renaming mapping

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related

Assign to `df.columns` something that doesn't have a name as suggested by @anky_91

Assign to `df.columns` a series whose name is `None`

Use `rename` with the `new_header` as a renaming mapping