Add independent header to csv file with pandas

user_10164308 0
Viewed 1496

I have a csv dataframe which I want to save with an extra header, apart from the columns. pandas must read my file without the header:

pd.read_csv('file.csv', header=2)

After editing the csv file, I would like to save it with a new header, e.g. it is supposed to look like this:

no rows = 4
no cols = 3        
index, col1, col2, col3
0, A, B, C
1, D, E, F
2, G, H, I
3, J, L, M

But the the header argument in the to_csv function seems to be dependent on the columns as aliases according to the docs (https://pandas.pydata.org/pandas-docs/stable/generated/pandas.DataFrame.to_csv.html)

The header would count each row as one cell, therefore a 1col x 2row df. I thought about appending the two, but due to unequal columns it wouldn't work. Is there an easy way to add an independent header to a csv-file?

I couldn't find a post with this problem, any help will be appreciated!

pandas dataframe header csv python
1 Answers

So, when you want to read the file without header in to pandas. Do:

df = pd.read_csv('file.csv', header=None)

Now, after doing all sort of processing in df, you can assign column_names as per your choice like:

col_list = ['col1','col2','col3']
df.columns = col_list

Now, write this df back to csv using to_csv:

df.to_csv('/new_file.csv', index=None, header=None)

You can paste your 2-line header to the top of this file by writing it like this:

        with open('/new_file.csv', 'r+') as f:
        content = f.read()
        f.seek(0, 0)
        f.write('my_2-line_header' + '\n' + content)

Hope this helps.

user_58208140
Related