Pandas 기초7 (데이터프레임 생성 및 추가)

 

 

In [1]:

import pandas as pd

In [3]:

friend_list = [
    {'name':'John','age': 15, 'job': 'student'},
    {'name':'Jenny','age': 25, 'job': 'developer'},
    {'name':'Nate','age': 30, 'job': 'teacher'},
]
df = pd.DataFrame(friend_list, columns = ['name','age','job'])
df

Out[3]:

	name	age	job
0	John	15	student
1	Jenny	25	developer
2	Nate	30	teacher

 

열 생성¶

In [4]:

# salary열 생성
df['salary'] = 0
df

Out[4]:

	name	age	job	salary
0	John	15	student	0
1	Jenny	25	developer	0
2	Nate	30	teacher	0

In [6]:

# 연봉삽입(학생제외)
import numpy as np
df['salary'] = np.where(df['job'] != 'student', 'yes', 'no')
df

Out[6]:

	name	age	job	salary
0	John	15	student	no
1	Jenny	25	developer	yes
2	Nate	30	teacher	yes

In [7]:

#데이터 수정
friend_list = [
    {'name':'John','mid': 95, 'final': 85},
    {'name':'Jenny','mid': 85, 'final': 80},
    {'name':'Nate','mid': 30, 'final': 10},
]
df = pd.DataFrame(friend_list, columns = ['name','mid','final'])
df

Out[7]:

	name	mid	final
0	John	95	85
1	Jenny	85	80
2	Nate	30	10

In [10]:

#총점 열 생성
df['total'] = df['mid']+ df['final']
df

Out[10]:

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

In [13]:

#평균 열 생성
df['average'] = df['total']/2
df

Out[13]:

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

In [14]:

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

df['grades'] = grades
df

Out[14]:

	name	mid	final	total	average	grades
0	John	95	85	180	90.0	A
1	Jenny	85	80	165	82.5	B
2	Nate	30	10	40	20.0	F

In [16]:

# pass fail로 바꾸기
def pass_or_fail(row):
    if row != 'F':
        return "Pass"
    else:
        return "Fail"

In [19]:

df.grades = df.grades.apply(pass_or_fail)
df

Out[19]:

	name	mid	final	total	average	grades
0	John	95	85	180	90.0	Pass
1	Jenny	85	80	165	82.5	Pass
2	Nate	30	10	40	20.0	Pass

In [21]:

data_list = [
    {
        'yyyy-mm-dd': '2000-06-27'
    },
    {
        'yyyy-mm-dd': '2007-10-27'
    }
]

df = pd.DataFrame(data_list, columns = ['yyyy-mm-dd'])
df

Out[21]:

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

In [22]:

# -으로 나눠지는 0번째 인자(년도)만 추출하는 함수
def extract_year(row):
    return row.split('-')[0]

In [24]:

# 연도 열 생성
df['year'] = df['yyyy-mm-dd'].apply(extract_year)
df

Out[24]:

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

 

행 생성¶

In [25]:

#데이터 수정
friend_list = [
    {'name':'John','mid': 95, 'final': 85},
    {'name':'Jenny','mid': 85, 'final': 80},
    {'name':'Nate','mid': 30, 'final': 10},
]
df = pd.DataFrame(friend_list, columns = ['name','mid','final'])
df

Out[25]:

	name	mid	final
0	John	95	85
1	Jenny	85	80
2	Nate	30	10

In [27]:

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

Out[27]:

	name	mid	final
0	Ben	50	50

In [29]:

# df에 추가
df.append(df2)

Out[29]:

	name	mid	final
0	John	95	85
1	Jenny	85	80
2	Nate	30	10
0	Ben	50	50

In [30]:

# df 인덱스번호 이어서 추가
df.append(df2, ignore_index = True)

Out[30]:

	name	mid	final
0	John	95	85
1	Jenny	85	80
2	Nate	30	10
3	Ben	50	50

In [ ]: