Pandas TimedeltaIndex Problem

[Christoph S]

Hi

I'm trying to plot some data relative to the start-time of the data. Therefore I substract the first timestamp form the DataFrame index and set the result as the new index.
The DataFrame with the new index looks fine, but when I try to plot the data or just save it to a file I get an error. Do I have to create the new DataFrame in a different way,
or is this a bug?
Pandas version is 0.16.1, Python is 3.4.1

Thank you for your answers!

A simple example:

Python 3.4.1 (default, Nov  3 2014, 14:38:10)
Type "copyright", "credits" or "license" for more information.

IPython 3.0.0 -- An enhanced Interactive Python.
?         -> Introduction and overview of IPython's features.
%quickref -> Quick reference.
help      -> Python's own help system.
object?   -> Details about 'object', use 'object??' for extra details.

In [1]: import pandas as pd

In [2]: df = pd.DataFrame({'a': [1,2,3,4], 'b': [1,2,3,4]}, index = pd.date_range('20150606', periods=4, freq='H'))

In [3]: df
Out[3]:
                     a  b
2015-06-06 00:00:00  1  1
2015-06-06 01:00:00  2  2
2015-06-06 02:00:00  3  3
2015-06-06 03:00:00  4  4

In [4]: df.index = df.index - df.index[0]

In [5]: df
Out[5]:
          a  b
00:00:00  1  1
01:00:00  2  2
02:00:00  3  3
03:00:00  4  4

In [6]: df.plot()
---------------------------------------------------------------------------
MemoryError                               Traceback (most recent call last)
....
MemoryError:

In [7]: df.to_csv('out.txt')
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
.....
TypeError: Argument 'data_index' has incorrect type (expected numpy.ndarray, got list)

In [8]: pd.__version__
Out[8]: '0.16.1'

[Jeff]

This is not implemented. Simply set the index to the str values of the index. See here https://github.com/pydata/pandas/issues/8711