Python pandas 를 이용한 데이타 분석
import pandas as pd
temp = pd.Series([-20, -10, 10,20])
print(temp)
temp = pd.Series([-20, -10, 10,20],index=['jan','feb','mar','apr'])
temp
data = {
'이름' : ['채치수', '정대만','송태섭','서태웅','강백호','변덕규','황태산','윤대협'],
'학교' : ['북산고','북산고','북산고','북산고','북산고', '능남고','능남고','능남고'],
'키' : [197, 184, 168, 187, 188, 202, 188, 190],
'국어' : [90, 40, 80, 40, 15, 80, 55, 100],
'sw특기' : ['Python', 'Java', 'Javascript', '','', 'c', 'python', 'c#']
}
data['이름']
import pandas as pd
df = pd.DataFrame(data)
df
df.index.name = '지원번호'
df.reset_index(drop=True, inplace = True)
df = pd.DataFrame(data, index=['1번','2번','3번','4번','5번','6번','7번','8번'])
df.set_index('이름')
df.set_index('이름', inplace=True)
df.sort_index()
df.sort_index(ascending=False)
df.to_csv('score.csv', encoding = 'utf-8-sig')
df.to_csv('score.txt', sep='\t')
df.to_excel('score.xlsx')
df = pd.read_csv('score.csv', skiprows=[0,1,2])
df = pd.read_csv('score.csv', nrows=3, skiprows=1)
df = pd.read_csv('score.txt',sep='\t', index_col = '지원번호')
df = pd.read_excel('score.xlsx')
df
df.describe()
df.info()
df.head()
df.head(7)
df.tail()
df.values
df.index
df.columns
df.shape
df['키'].describe()
df['키'].min()
df['키'].max()
df['키'].nlargest(3)
df['키'].mean()
df['키'].sum()
df['학교'].unique()
df['학교'].nunique()
df[['키','이름']]
df.columns[0]
df[df.columns[0]]
df.columns[-1]
df['국어'][1:5]
df[3:5]
df.loc[1]
df.loc[[0,1],'국어']
df.loc[[0,1],['국어','키']]
df.loc[1:5, '이름':'국어']
df.iloc[0]
df.iloc[4]
df.iloc[0:5]
df.iloc[0,2]
df.iloc[[0,1],2]
df.iloc[[0,1],[2,3,4]]
df['키']>=185
filt = (df['키'] >= 185)
df[filt]
df[~filt]
df[df['키']>=185]
df.loc[df['키']>=185,'국어']
df.loc[df['키']>=185,['이름','국어']]
df.loc[(df['키']>=185) & (df['학교']=='북산고')]
df.loc[(df['키']<170) | (df['키']>200)]
filt = df['이름'].str.startswith('송')
df[filt]
filt2 = df['이름'].str.contains('태')
df[filt2]
df[~filt2]
langs = ['python', 'Java']
filt3 = df['sw특기'].isin(langs)
df[filt3]
langs = ['python','java']
filt = df['sw특기'].str.lower().isin(langs)
df[filt]
filt = df['sw특기'].str.contains('Java', na = False)
df[filt]
df.fillna('')
df.fillna('없음')
df['학교'] = np.nan
df.fillna('모름')
df.fillna('오름', inplace = True)
df['sw특기'].fillna('확인중', inplace=True)
df = pd.read_excel('score.xlsx')
df.dropna()
df.dropna(axis='index', how="any")
df.dropna(axis='columns')
df.sort_values('키')
df.sort_values('키', ascending=False)
df.sort_values(['키','국어'], ascending=False)
df.sort_values(['키', '국어'], ascending=[True, False])
df['키'].sort_values()
df.sort_index()
df.sort_index(ascending=False)
df['학교'].replace({'북산고':'상북고', '능남고':'무슨고'})
df['sw특기'] = df['sw특기'].str.lower()
df['학교']=df['학교']+'등학교'
df['총합'] = [455,205, 280, 325, 90, 440, 240, 465]
df['결과']='fail'
df.loc[df['총합'] > 400, '결과'] ='pass'
df.drop(columns=['총합'])
df.drop(index=4)
filt = df['국어'] < 80
df[filt]
df[filt].index
df.drop(index=df[filt].index)
df.loc[8]=['9번','이정환','해남고등학교',184,90,'Kotlin',450,'pass']
df['학교'].replace({'북산고등학교등학교등학교':'북산고등학교','능남고등학교등학교등학교':'능남고등학교'}, inplace=True)
df.loc[3,'sw특기']='python'
df.loc[4,['학교','sw특기']]=['능남고등학교','c']
cols = list(df.columns)
df = df[[cols[-1]]+cols[0:-1]]
df.columns = ['result','number','name','graduate', 'height','grade','skill', 'total']
df['학교']=df['학교']+'등학교'
df['영어']=[85,35,75,60,20,100,65,85]
def add_cm(height):
return str(height) + 'cm'
df['키'] = df['키'].apply(add_cm)
def capital(lang):
if pd.notnull(lang):
return lang.capitalize()
return lang
df['sw특기']=df['sw특기'].apply(capital)
df['sw특기'].str.capitalize()
df.groupby('학교')
df.groupby('학교').get_group('북산고')
df.groupby('학교').get_group('능남고')
df.groupby('학교').mean()
df.groupby('학교').size()
df.groupby('학교').size()['능남고']
df.groupby('학교')['키'].mean()
df.groupby('학교')[['키', '국어']].mean()
df['학년']=[3,3,2,1,1,3,2,2]
df.groupby(['학교','학년']).mean()
df.groupby('학년').mean()
df.groupby('학년').mean().sort_values('키',ascending=False)
df.groupby('학교')['sw특기'].count()
school = df.groupby('학교')
school['학년'].value_counts()
school['학년'].value_counts().loc['북산고']
school['학년'].value_counts(normalize=True).loc['북산고']