学习笔记--利用python进行数据分析

```

import json

from collectionsimport Counter

from collectionsimport defaultdict

from pandasimport  DataFrame,Series

import pandasas pd;import numpyas np

path='example.txt'

#print(open(path).readline())

records=[json.loads(line)for linein open(path)]#按行以json读取所有数据

#print(records[0]['tz'])

#for rec in records: #rec 是每一行

#  print(rec)

#  break

time_zones=[rec['tz']for recin recordsif 'tz' in rec]#遍历json，如果该行包含字段‘tz’，则读取该字段的内容

#print(time_zones[:10])

#统计不同tz内容的各多少个返回字典字段：内容

def get_counts(sequence):

counts={}

for xin sequence:

if xin counts:

counts[x]+=1

        else:

counts[x]=1

    return counts

#print(get_counts(time_zones))

#print(len(time_zones))

def get_counts2(sequence):

counts=defaultdict(int)#所有的值都会被初始化为0

    for xin sequence:

counts[x]+=1

    return counts

#前10个排序

def top_counts(count_dict,n=10):

value_key_pairs=[(count,tz)for tz,countin count_dict.items()]#count_dict.items()-字段：内容

    print(count_dict.items())

value_key_pairs.sort()#根据字段排序

    return value_key_pairs[-n:]

print(top_counts(get_counts(time_zones)))

#使用collections.counter类

counts=Counter(time_zones)

print(counts.most_common(10))

#使用padndas

frame=DataFrame(records)

#print(frame)

tz_counts=frame['tz'].value_counts()

print(tz_counts[:10])

clean_tz=frame['tz'].fillna('Missing')

clean_tz[clean_tz=='']='Unknown'

tz_counts=clean_tz.value_counts()

print(tz_counts[:10])

tz_counts[:10].plot(kind='barh',rot=0)

```