Numpy溢出

今天在看书时，按照书上的例子敲了一段代码，发现跟书上的结果有出入，具体代码如下：

#coding = utf-8
__author__ = 'Administrator'

import sys
from datetime import datetime
import numpy as np

"""
vectorsum.py
"""
def numpysum(n):
    a = np.arange(n) ** 2
    b = np.arange(n) ** 3
    c = a + b
    return c

def pythonsum(n):
    a = range(n)
    b = range(n)
    c = []

    for i in range(len(a)):
        a[i] = a[i] ** 2
        b[i] = b[i] ** 3
        c.append(a[i] + b[i])

    return c


size = int(sys.argv[1])

start = datetime.now()
c = pythonsum(size)
delta = datetime.now() - start

print "The last 2 elements of the sum",c[-2:]
print "pythonsum elapsed time in microseconds",delta.microseconds

start1 = datetime.now()
c = numpysum(size)
delta1 = datetime.now() - start1

print "The last 2 elements of the sum",c[-2:]
print "numpysum elapsed time in microseconds ",delta1.microseconds

执行 python vectorsum 10000 时发生如下错误

vectorsum.py:13: RuntimeWarning: invalid value encountered in power
  b = np.arange(n) ** 3
The last 2 elements of the sum [999500079996L, 999800010000L]
pythonsum elapsed time in microseconds 17000
The last 2 elements of the sum [-2047523644 -2047503647]
numpysum elapsed time in microseconds  2000

上面的结果是对的  下面的numpy结果错了，然后国外的网站上提供了一个解决办法就是修改下numpysum函数修改完是这样的：

def numpysum(n):
    a = np.arange(n,dtype=object) ** 2
    b = np.arange(n,dtype=object) ** 3
    c = a + b
    return c

完了之后结果就一致了，vectorsum.py 10000
The last 2 elements of the sum [999500079996L, 999800010000L]
pythonsum elapsed time in microseconds 16000
The last 2 elements of the sum [999500079996L 999800010000L]
numpysum elapsed time in microseconds  12000

主要原因貌似是由于python版本用的32位的。