Pandas 常用功能整理

1.Create DataFrame

df=pd.DataFrame([Series1,Series2],index=['index1','index2'])
df=DataFrame(np.arange(9).reshape((3,3)),index=['a','b','c'],columns=['A','B','C'])
df=pd.DataFrame([List1,List2],index=['index1','index2'])
df=pd.DataFrame({'col1':'1','col2':'2','col3':3},index=['id1','id2'])
df=se.to_frame(series)
df=read_csv(path,encoding='utf-8',sep=';')

2.Query

df.index
df.index[0] #输出索引
df.index.get_loc['index']

df.iloc
df.iloc[0]
df.iloc[:,1:3]
df.iloc[[1,2]]

df.loc
df.loc['index']
df.loc[:,'col']
df.loc['index1':'index3','col1':col3']

df.dtypes
df.index.dtype
df.'col'.dtype

df.'col'.unique()
df['col'].isnull()

3.Data Edit

delete
df.drop(['col1','col2'],axis=1)
df.drop(['index1','index2'],axis=0)
del df['col']
df.dropna(how='any'/'all')

add
df.fillna(int) #直接填充输入的int值在空值处
df.fillna({col:int,col2:int,col3:int})
df.fillna(method='ffill')
df.fillna(method='bfill',limit=3)

df.append(df2)
df.insert(0,'col',value)
df. concat([df1,df2],key=['x','y'],join='inner')
df. concat([df1,df2],key=['x','y'],join_axes[df1.index])
pd.merge(df1,df2,how='left',on=['col1','col2'])
pd.join(df2,lsuffix='_l', rsuffix='_r') #合并的列表有相同列名情况
df.where(df['col']>condition) #返回原！数据同样格式，不符合条件的整行返回NaN
df.melt(id_vars=col1,value_vars=[('col2', 'col3')])
pd.cut(df['col'],bis=[1,3,5,7]) #分段用，左开右闭

4.Index/Colums Edit

df.set_index() #设置普通列为index,或复合index
df.reset_index(drop=False/True) #增加/还原原列数据
df.reindex（fill_value=0）
df.rename({'col1':'new_col1','col2':'new_col2'},axis=o/1)
df.droplevel #同复合index，复合column一起用
df.get_dummies(prefix=['col1','col2'])
df['col'].str.lower/uper()
df['col'].replace('old','new')
df['col'].str.split(' ', n, expend=True); #分隔成n列
df.extract #正则