Using pandas, how do I subsample a large DataFrame by group in an efficient manner?

I am trying to subsample rows of a DataFrame according to a grouping. Here is an example. Say I define the following data:

from pandas import *
df = DataFrame({'group1' : ["a","b","a","a","b","c","c","c","c",
                            "c","a","a","a","b","b","b","b"],
       开发者_运维百科         'group2' : [1,2,3,4,1,3,5,6,5,4,1,2,3,4,3,2,1],
                'value'  : ["apple","pear","orange","apple",
                            "banana","durian","lemon","lime",
                            "raspberry","durian","peach","nectarine",
                            "banana","lemon","guava","blackberry","grape"]})

If I group by group1 and group2, then the number of rows in each group is here:

In [190]: df.groupby(['group1','group2'])['value'].agg({'count':len})
Out[190]: 
      count
a  1  2    
   2  1    
   3  2    
   4  1    
b  1  2    
   2  2    
   3  1    
   4  1    
c  3  1    
   4  1    
   5  2    
   6  1    

(If there is an even more concise way to compute that, please tell.)

I now want to construct a DataFrame that has one randomly selected row from each group. My proposal is to do it like so:

In [215]: from random import choice
In [216]: grouped = df.groupby(['group1','group2'])
In [217]: subsampled = grouped.apply(lambda x: df.reindex(index=[choice(range(len(x)))]))
In [218]: subsampled.index = range(len(subsampled))
In [219]: subsampled
Out[219]: 
    group1  group2  value
0   b       2       pear 
1   a       1       apple
2   b       2       pear 
3   a       1       apple
4   a       1       apple
5   a       1       apple
6   a       1       apple
7   a       1       apple
8   a       1       apple
9   a       1       apple
10  a       1       apple
11  a       1       apple

which works. However, my real data has about 2.5 million rows and 12 columns. If I do this the dirty way by building my own data structures, I can complete this operation in a matter of seconds. However, my implementation above does not finish within 30 minutes (and does not appear to be memory-limited). As a side note, when I tried implementing this in R, I first tried plyr, which also did not finish in a reasonable amount of time; however, a solution using data.table finished very rapidly.

How do I get this to work rapidly with pandas? I want to love this package, so please help!

I tested with apply, it seems that when there are many sub groups, it's very slow. the groups attribute of grouped is a dict, you can choice index directly from it:

subsampled = df.ix[(choice(x) for x in grouped.groups.itervalues())]

EDIT: As of pandas version 0.18.1, itervalues no longer works on groupby objects - you can just use .values:

subsampled = df.ix[(choice(x) for x in grouped.groups.values())]