计数：统计出现的次数

统计出现的次数的方法：

#coding=utf-8
__author__ = 'mac'

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import pylab as pl
import os

#
import json
path='/Users/mac/PycharmProjects/python2.7/BigData/pydata-book-2nd-edition/datasets/bitly_usagov/example.txt'
records=[json.loads(line) for line in open(path)]
print records[0]['tz']
time_zone=[rec['tz'] for rec in records if 'tz' in rec]
print time_zone[:10]

方法1：用纯python方法
def get_counts(sequence):
    counts={}
    for x in sequence:
        if x in counts:
            counts[x]+=1
        else:
            counts[x]=1
    return counts
# #测试数据
# a=['a','b',1,'a','c','a']
# print get_counts(a)
# print get_counts(time_zone[:10])

方法2：用collections的defaultdict将字典初始化为0
#或用collections库
from collections import defaultdict
def get_counts2(sequence):
    counts=defaultdict(int) #所有值都会被初始化为0
    for x in sequence:
        counts[x]+=1
    return counts
# #测试数据
# a=['a','b',1,'a','c','a']
# print get_counts2(a)

#计算time_zones的值
counts=get_counts(time_zone)
print counts['America/New_York']
print len(time_zone)

# a={'a':3,'b':5,'c':1,'d':3}
#处理排在前10的时区
def top_counts(count_dict,n=10):
    value_key_pairs=[(count,tz) for tz,count in count_dict.items()]
    print value_key_pairs
    value_key_pairs.sort() #sort()升序，故下面从最后面的开始倒着取
    return value_key_pairs[-n:]

print top_counts(counts)

方法3：用collections库中的Counter计数器，Counter.most_common从大到小排序取前10
#可以用collections的Counter类，它更简单
from collections import Counter
counts=Counter(time_zone)
#取counts中的前10的时区用most_common(10)
print counts.most_common(10)

方法4：引用数据类的库pandas的DataFrame,Series会带有index索引，和numpy中的array差不多
#使用pandas对时区进行计数
from pandas import DataFrame,Series
frame=DataFrame(records)
# print frame['tz'][:10]
print frame['tz']
tz_counts=frame['tz'].value_counts()
# print tz_counts

#用fillna函数可以替换缺失值(NAN),而未知值(空字符串)则可以通过布尔型数组索引加以替换：
clean_tz=frame['tz'].fillna('Missing')
# print clean_tz
clean_tz[clean_tz=='']='Unknown'
# print clean_tz
tz_counts=clean_tz.value_counts()
print tz_counts[:10]
print frame['a'][1]
#plot(tz_counts[:10],kind='barh',rot=0)
# pl.show()加上此才可以显示图形
results=Series([x.split()[0] for x in frame.a.dropna()]) #dropna()是删除缺失项
print results[:5]