How to add a column to a pandas dataframe without column?

Question

I have a DataFrame wich has 2 'columns'. The first column does not seem to have a column name, the second one is named Speed.

Here is a MRE:

from io import StringIO  # to read strings as files for read_csv

import pandas as pd

parts = [
    '[Level1]\nLocation = "London"\nType= "GTHY66"\n'
    'Date = "16-11-2021"\nEnergy level = "Critical zero"\n',
    '0.000   26.788\n0.027   26.807\n0.053   26.860'
]

lvl2_lines = "Speed\n" + parts[1]

df_level2 = pd.read_csv(StringIO(lvl2_lines), sep='\t')
print(df_level2.columns)
print(df_level2)

This was my output when I did the print statements:

Index(['Speed'], dtype='object')
            Speed
0  0.000   26.788
1  0.027   26.807
2  0.053   26.860

This is my desired output:

Index(['Power', 'Speed'], dtype='object')
   Power   Speed
0  0.000  26.788
1  0.027  26.807
2  0.053  26.860

It look slike you have a multiindex perhaps? Try df.reset_index() — G. Anderson, Nov 17 '21 at 19:43
@not_speshal no because I don't want to rename. I want to add because I only have one column — Al-Andalus, Nov 17 '21 at 19:47
I already tried what @splash58 did but I received the following error: `ValueError: Length mismatch: Expected axis has 1 elements, new values have 2 elements` — Al-Andalus, Nov 17 '21 at 19:50
`df[['Power', 'Speed']] = df['Speed'].str.split(r'\s+', 1, expand=True)` — Henry Ecker, Nov 17 '21 at 19:54
[Change column type in pandas](https://stackoverflow.com/questions/15891038/change-column-type-in-pandas) with `df[['Power', 'Speed']] = df['Speed'].str.split(r'\s+', 1, expand=True).astype(float)` if needing numbers instead of strings. — Henry Ecker, Nov 17 '21 at 19:55

score 1 · Answer 1 · answered Nov 17 '21 at 19:54

1

You need to split the strings:

df[["Power", "Speed"]] = df["Speed"].str.split(expand=True).astype(float)

>>> df
    Speed  Power
0  26.788  0.000
1  26.807  0.027
2  26.860  0.053

answered Nov 17 '21 at 19:54

not_speshal

22,093
2
15
30

We could just specify both column names `lvl2_lines = "Speed\n" + parts[1]` -> `lvl2_lines = "Power Speed\n" + parts[1]` then there's no need to split after the fact or change dtype. But this would push the question into typo territory. And `df_level2 = pd.read_csv(StringIO(lvl2_lines), sep=r'\s+', engine='python')` to handle mixed spacing. – Henry Ecker Nov 17 '21 at 20:00
I see what you mean. That's a neat way to directly read correctly instead of manipulating later. I answered before OP posted that bit in his question tho :) – not_speshal Nov 17 '21 at 21:10

How to add a column to a pandas dataframe without column?

1 Answers1