[Python] Pandas Series 생성, 데이터 액세스, 산술 연산

Pandas 장점

- Allows the use of labels for rows and columns
- 기본적인 통계데이터 제공
- NaN values 를 알아서 처리함.
- 숫자 문자열을 알아서 로드함.
- 데이터셋들을 merge 할 수 있음.
- It integrates with NumPy and Matplotlib

 

Pandas Series 데이터 생성

>>> import pandas as pd 

>>> index = ['eggs', 'apples', 'milk', 'bread']
>>> data = [30, 6, 'Yes', 'No']

# data로 만들기
>>> x = pd.Series(data = data)
>>> x
0     30
1      6
2    Yes
3     No
dtype: object

# data와, index로 만들기
>>> groceries = pd.Series(data = data, index = index)
>>> groceries
eggs       30
apples      6
milk      Yes
bread      No
dtype: object

# data, index, columns로 만들기
>>> pd.DataFrame(groceries, columns=['h'])
	h
eggs	30
apples	6
milk	Yes
bread	No

pandas series 데이터 억세스

>>> groceries
eggs       30
apples      6
milk      Yes
bread      No
dtype: object
>>> groceries[0]
30
>>> groceries['eggs']
30
>>> groceries[['eggs', 'bread']]
eggs     30
bread    No
dtype: object
groceries[-1]
'No'
>>> groceries
eggs       30
apples      6
milk      Yes
bread      No
dtype: object
>>> groceries['apples' : ]
apples      6
milk      Yes
bread      No
dtype: object
>>> groceries['apples' : 'milk']
apples      6
milk      Yes
dtype: object
>>> groceries[1 : 2 + 1]
apples      6
milk      Yes
dtype: object

산술연산

>>> index = ['apples', 'oranges', 'bananas']
>>> data = [10, 6, 3,]
>>> fruits = pd.Series(data = data, index = index)
>>> fruits
apples     10
oranges     6
bananas     3
dtype: int64
# 전체 데이터를 5씩 더하기
>>> fruits = fruits + 5
>>> fruits
apples     15
oranges    11
bananas     8
dtype: int64
# oranges만 2 빼기
>>> fruits['oranges'] = fruits['oranges'] - 2
>>> fruits
apples     15
oranges     9
bananas     8
dtype: int64
# apples 와 bananas를 3씩 더하기
# 반영해 주세요
>>> fruits
apples     15
oranges     9
bananas     8
dtype: int64
>>> fruits[['apples', 'bananas']] = fruits[['apples', 'bananas']] + 3
>>> fruits
apples     18
oranges     9
bananas    11
dtype: int64