list pandas map 索引及行号性能

    ind = pd.Index(range(993))


    print timeit.timeit("ind.get_loc(23)",setup="from __main__ import ind",)
    ind2 = set(range(9999123))
    print timeit.timeit("23431232 in ind2",setup="from __main__ import ind2",)


    ind3 = range(99)
    print timeit.timeit("23431232 in ind3",setup="from __main__ import ind3",)



 

    print timeit.timeit("ind3.index(12)",setup="from __main__ import ind3",)

 

 

输出：

4.46973529269

0.135495255209
5.70757211201

1.13445516574

 

 

 

    ind = pd.Index(range(9912213))
    time0=time.time()
    print ind.get_loc(2223)
 

    print  'correction time costed {0:.2f}'.format(time.time() - time0)




    a = {}


    for i,v in enumerate(ind):
        a[v]=i


    time0=time.time()
    print a[2233]

 

    print  'correction time costed {0:.5f}'.format(time.time() - time0)

 

 

输出：

    ind = pd.Index(range(9912213))
    time0=time.time()
    print ind.get_loc(2223)


    print  'correction time costed {0:.2f}'.format(time.time() - time0)




    a = {}


    for i,v in enumerate(ind):
        a[v]=i


    time0=time.time()
    print a[2233]


    print  'correction time costed {0:.5f}'.format(time.time() - time0)

 

输出：

2223

correction time costed 0.65
2233
correction time costed 0.00100

总结，a in b,b是set最快

寻找a在b中的行号，即index 。 list最慢，估计在遍历把 map最快，index 太慢