Pandas 데이터분석 기초 실습 -5

행,열(로우, 칼럼) 생성 및 수정하기¶

In [3]:

import pandas as pd

In [4]:

friend_dict_list = [
    {'name':'Jone','age':15,'job':'student'},
    {'name':'Jenny','age':30,'job':'developer'},
    {'name':'Nate','age':30,'job':'teacher'}
]
column = [
    'name','age','job'
]
df = pd.DataFrame(friend_dict_list,columns = column)
df

Out[4]:

	name	age	job
0	Jone	15	student
1	Jenny	30	developer
2	Nate	30	teacher

In [5]:

df['salary'] = 0    #새로운 열 추가, 기본값= 0
df

Out[5]:

	name	age	job	salary
0	Jone	15	student	0
1	Jenny	30	developer	0
2	Nate	30	teacher	0

In [6]:

import numpy as np

In [7]:

df['salary'] = np.where(df['job'] != 'student','yes','no')

In [8]:

df.head()

Out[8]:

	name	age	job	salary
0	Jone	15	student	no
1	Jenny	30	developer	yes
2	Nate	30	teacher	yes

In [9]:

friend_dict_list = [
    {'name':'Jone','midterm':95,'final':85},
    {'name':'Jenny','midterm':85,'final':80},
    {'name':'Nate','midterm':30,'final':10}
]
column = ['name','midterm','final']

df = pd.DataFrame(friend_dict_list,columns =column)
df.head()

Out[9]:

	name	midterm	final
0	Jone	95	85
1	Jenny	85	80
2	Nate	30	10

In [10]:

df['total'] = df['midterm']+df['final']
df.head()

Out[10]:

	name	midterm	final	total
0	Jone	95	85	180
1	Jenny	85	80	165
2	Nate	30	10	40

In [11]:

df['average'] = df['total'] / 2
df.head()

Out[11]:

	name	midterm	final	total	average
0	Jone	95	85	180	90.0
1	Jenny	85	80	165	82.5
2	Nate	30	10	40	20.0

In [12]:

grades = []

for row in df['average']:
    if row >= 90:
        grades.append('A')
    elif row >= 80:
        grades.append('B')
    else:
        grades.append('F')

df['grade'] = grades    #리스트를 바로 추가
df.head()

Out[12]:

	name	midterm	final	total	average	grade
0	Jone	95	85	180	90.0	A
1	Jenny	85	80	165	82.5	B
2	Nate	30	10	40	20.0	F

In [13]:

#apply 함수 사용

def pass_or_fail(row):
    if row != 'F':
        return 'Pass'
    else:
        return 'Fail'

In [14]:

df.grade = df.grade.apply(pass_or_fail) #return value를 df의 grade에 넣음
df.head()

Out[14]:

	name	midterm	final	total	average	grade
0	Jone	95	85	180	90.0	Pass
1	Jenny	85	80	165	82.5	Pass
2	Nate	30	10	40	20.0	Fail

In [18]:

date_list = [
    {
        'yyyy-mm-dd':'2000-06-27'
    },
    {
        'yyyy-mm-dd':'2007-10-27'
    }
]
df = pd.DataFrame(date_list,columns= ['yyyy-mm-dd'])
df.head()

Out[18]:

	yyyy-mm-dd
0	2000-06-27
1	2007-10-27

In [19]:

def extract_year(row):
    return row.split('-')[0]

In [21]:

df['year'] = df['yyyy-mm-dd'].apply(extract_year)
df.head()

Out[21]:

	yyyy-mm-dd	year
0	2000-06-27	2000
1	2007-10-27	2007

행 수정¶

In [22]:

friend_dict_list = [
    {'name':'Jone','midterm':95,'final':85},
    {'name':'Jenny','midterm':85,'final':80},
    {'name':'Nate','midterm':30,'final':10}
]
column = ['name','midterm','final']

df = pd.DataFrame(friend_dict_list,columns =column)
df.head()

Out[22]:

	name	midterm	final
0	Jone	95	85
1	Jenny	85	80
2	Nate	30	10

In [24]:

df2 = pd.DataFrame([['Ben',50,50]],
columns=['name','midterm','final']
)
df2.head()

Out[24]:

	name	midterm	final
0	Ben	50	50

In [25]:

df.append(df2,ignore_index= True)   #index무시하고 넣음

Out[25]:

	name	midterm	final
0	Jone	95	85
1	Jenny	85	80
2	Nate	30	10
3	Ben	50	50

In [ ]: