Sample test 생성



In [42]:

    
import random 

alldata = []
all_words = "apple banana orange mango blueberry pizza cheese naruto".split()
for _ in range(5):
    fruit = {}
    for word in all_words: 
        n = random.randint(0,5) * 10
        if n != 0:
            fruit[word] = n
    # print fruit
    alldata.append(fruit)

alldata









    Out[42]:





[{'apple': 50,
  'banana': 30,
  'orange': 40,
  'mango': 40,
  'blueberry': 30,
  'pizza': 40,
  'cheese': 10,
  'naruto': 20},
 {'banana': 20,
  'orange': 20,
  'blueberry': 40,
  'pizza': 30,
  'cheese': 30,
  'naruto': 10},
 {'apple': 50, 'orange': 30, 'blueberry': 40, 'cheese': 50, 'naruto': 40},
 {'banana': 20, 'orange': 20, 'mango': 50, 'cheese': 10, 'naruto': 50},
 {'banana': 30,
  'orange': 10,
  'mango': 40,
  'blueberry': 40,
  'cheese': 50,
  'naruto': 50}]

pandas 에 추가



In [43]:

    
import pandas as pd



In [44]:

    
df = pd.DataFrame()
df['all'] = alldata
df









    Out[44]:







  
    
      
      all
    
  
  
    
      0
      {'apple': 50, 'banana': 30, 'orange': 40, 'man...
    
    
      1
      {'banana': 20, 'orange': 20, 'blueberry': 40, ...
    
    
      2
      {'apple': 50, 'orange': 30, 'blueberry': 40, '...
    
    
      3
      {'banana': 20, 'orange': 20, 'mango': 50, 'che...
    
    
      4
      {'banana': 30, 'orange': 10, 'mango': 40, 'blu...

함수 정의



In [45]:

    
def getCount(item, **kwargs):
    column = kwargs['name']
    if column in item:
        return item[column]
    else:
        return 0



In [46]:

    
i = df.loc[0]['all']
i









    Out[46]:





{'apple': 50,
 'banana': 30,
 'orange': 40,
 'mango': 40,
 'blueberry': 30,
 'pizza': 40,
 'cheese': 10,
 'naruto': 20}



In [47]:

    
getCount(i, name='apple')









    Out[47]:





50

apply에 사용하기



In [48]:

    
df['all']









    Out[48]:





0    {'apple': 50, 'banana': 30, 'orange': 40, 'man...
1    {'banana': 20, 'orange': 20, 'blueberry': 40, ...
2    {'apple': 50, 'orange': 30, 'blueberry': 40, '...
3    {'banana': 20, 'orange': 20, 'mango': 50, 'che...
4    {'banana': 30, 'orange': 10, 'mango': 40, 'blu...
Name: all, dtype: object



In [49]:

    
cols = ['apple', 'banana']
for col in cols:
    df[col] = df['all'].apply(getCount, name = col)



In [50]:

    
df









    Out[50]:







  
    
      
      all
      apple
      banana
    
  
  
    
      0
      {'apple': 50, 'banana': 30, 'orange': 40, 'man...
      50
      30
    
    
      1
      {'banana': 20, 'orange': 20, 'blueberry': 40, ...
      0
      20
    
    
      2
      {'apple': 50, 'orange': 30, 'blueberry': 40, '...
      50
      0
    
    
      3
      {'banana': 20, 'orange': 20, 'mango': 50, 'che...
      0
      20
    
    
      4
      {'banana': 30, 'orange': 10, 'mango': 40, 'blu...
      0
      30



In [ ]:

	all
0	{'apple': 50, 'banana': 30, 'orange': 40, 'man...
1	{'banana': 20, 'orange': 20, 'blueberry': 40, ...
2	{'apple': 50, 'orange': 30, 'blueberry': 40, '...
3	{'banana': 20, 'orange': 20, 'mango': 50, 'che...
4	{'banana': 30, 'orange': 10, 'mango': 40, 'blu...