I have a data frame df which looks like this:

I would like to use year and date as indexes so that I can retrive rows from a specific year or rows from a specific date given a year. For example, I would like to get data from year 2012. Also, I would like to get data from 2013-11-05. I probably need to use multiindex, but how can I set the multiindex with a dataframe like this?

CodePudding user response：

To make multiindex, use set_index() with list of column names

df = df.set_index(['year','date'])

               Time  observation1  observation2
year date                                      
2012 11-02  9:12:00     79.373668           224
     11-02  9:13:00    130.841316           477
     11-05  9:14:00     45.312814           835
2013 11-05  9:15:00    123.776946           623
     11-05  9:16:00     79.373668           224
     11-22  9:17:00    130.841316           477
     11-22  9:18:00     45.312814           835
2014 11-01  9:19:00    123.776946           623

But to make the selections you need, the following operations are enough without forming a multiindex

print(df[df.year.eq(2012)])
print(df[df.year.eq(2013) & df.date.eq('11-05')])

   year   date     Time  observation1  observation2
0  2012  11-02  9:12:00     79.373668           224
1  2012  11-02  9:13:00    130.841316           477
2  2012  11-05  9:14:00     45.312814           835

   year   date     Time  observation1  observation2
3  2013  11-05  9:15:00    123.776946           623
4  2013  11-05  9:16:00     79.373668           224