Home/ Questions/Q 8509117
In Process

The Archive Base Latest Questions

Editorial Team
  • 0
Asked: 2026-06-11T03:20:53+00:00

I’d like to create a pandas DataFrame from the following csv file: ……….. EUR

  • 0

I’d like to create a pandas DataFrame from the following csv file:

...........
EUR MS 3M;20111025;7d;1.1510;
EUR MS 3M;20111024;7d;1.1530;
EUR MS 3M;20111025;1m;1.1580;
EUR MS 3M;20111024;1m;1.1590;
...........

The DataFrame ideally would have a name given as the first column value (“EUR MS 3M”), an index composed by values in the second column (“20111025”), and where the column names for the DataFrame would be taken from the third column (“7d”, “1m” etc.), with correspondent values given in the last csv column (“1.150” etc.).
I have tried with different methods, but couldn’t sort this thing out the proper way. I think the first thing I should do should be to ‘unstack’ the values in the csv, in order to have an aligned index first, and then create a DataFrame, but really don’t how…
Anyone more expert than me have any clue? I’m started learning pandas only few weeks ago…
Thanks for your kind help!

(I’m editing the question to make thing a bit clearer:
I’d like to get a dataframe named EUR MS 3M containing columns such as: 

index    7d     1m 
20111024 1.1530 1.1590 
20111025 1.1510 1.1580

hope it is a bit clearer now. Thanks)

Share

Leave an answer

You must login to add an answer.

Need An Account,

1 Answer

  1. Editorial Team

    How about something like this:

    In [103]: print data
EUR MS 3M;20111025;7d;1.1510
EUR MS 3M;20111024;7d;1.1530
EUR MS 3M;20111025;1m;1.1580
EUR MS 3M;20111024;1m;1.1590

In [104]: frame = pd.read_csv(StringIO(data), sep=';', 
                              names=['frame_name', 'index', 'column', 'value'])
In [105]: name = frame.ix[0, 'frame_name']

In [106]: rs = frame.ix[:, 1:].pivot('index', 'column', 'value')

In [107]: rs.name = name

In [108]: rs
Out[108]:
column       1m     7d
index
20111024  1.159  1.153
20111025  1.158  1.151
    • 0