I read a csv with separate date and time columns. I generated an index from them in the following way:
data = p.read_csv(fileName,usecols=["date","time","price"])
data.set_index(["date","time"],inplace=True)
This however isn't very useful when I want to get the difference in days or hours between rows. How do I generate a single datetime index from the separate date and time columns?
I think you need parameter parse_dates with nested list with both columns and parameter index_col with new column created by concanecated columns names separated by _:
data = p.read_csv(fileName,
usecols=["date","time","price"],
parse_dates=[["date","time"]],
index_col=['date_time'])
Sample:
from pandas.compat import StringIO
temp=u"""date,time,price
2015-01-01,14:00:10,7
2014-01-01,10:20:10,1"""
#after testing replace 'StringIO(temp)' to 'filename.csv'
df = pd.read_csv(StringIO(temp),
usecols=["date","time","price"],
parse_dates=[["date","time"]],
index_col=['date_time'])
print (df)
price
date_time
2015-01-01 14:00:10 7
2014-01-01 10:20:10 1
print (df.index)
DatetimeIndex(['2015-01-01 14:00:10', '2014-01-01 10:20:10'],
dtype='datetime64[ns]',
name='date_time', freq=None)
pd.to_datetime.