groupby_sum.py

import numpy as np
import random

# dict to hold results                                                                    
counts = {}

# a hack to generate random key, value pairs.                                             
# 5k keys, 100k values                                                                    
x=[i%5000 for i in xrange(100000)]
random.shuffle(x)
kv = np.array(zip(x, np.random.uniform(0, 1, 100000)))

# groupby sum                                                                             
for k, v in kv:
    try:
	counts[k] += v
    except KeyError:
	counts[k] = v