After reading Parse dates when YYYYMMDD and HH are in separate columns using pandas in Python and Using python pandas to parse CSV with date in format Year, Day, Hour, Min, Sec
I still am not able to parse dates with separated columns for year, month, day and hour. My data looks like this (zeroth column is ID, first is year, second is month, third is day, fourth is hour and fifth is value)
50136 2011 1 1 21 9792
50136 2011 1 1 22 9794
50136 2011 1 1 23 9796
50136 2011 1 1 0 9798
50136 2011 1 1 1 9799
50136 2011 1 1 2 9802
I've tried following:
df = pd.read_csv(file, parse_dates = {'date': [1, 2, 3, 4]}, , index_col='date')
, but then I get index not as timestamp but as unicode(?)
In [17]: print df.head()
Out [17]:
0 5
date
2011 1 1 21 50136 9792
2011 1 1 22 50136 9794
2011 1 1 23 50136 9796
2011 1 1 0 50136 9798
2011 1 1 1 50136 9799
In [18]: print df.index
Out [18]:
Index([u'2011 1 1 21', u'2011 1 1 22', u'2011 1 1 23', u'2011 1 1 0', u'2011 1 1 1', u'2011 1 1 2'], dtype=object)
I'm obviously doing something wrong, but I can't figure it out. Any advise is really appreciated.