#coding=utf-8
import pandas as pd
import numpy as np

data={
    'company':['B','A','B','B'],
    'gender':['female','female','male','male'],
    'num':[40,31,28,28]
}
#传入字典创建DataFrame对象
df=pd.DataFrame(data)
print(df)
'''
  company  gender  num
0       B  female   40
1       A  female   31
2       B    male   28
3       B    male   28
'''




#举例1、
#将列名 num -> age 修改成age，将索引 0-3 修改为 A-D
print(df.rename(columns={"num":"age"},index={0:"A",1:"B",2:"C",3:"D"}))
'''
  company  gender  age
A       B  female   40
B       A  female   31
C       B    male   28
D       B    male   28

注：此方式只是生成了一个副本，并未真正修改原df对象，如果要修改原df对象，需要传递参数inplace=True
'''
df.rename(columns={"num":"age"},index={0:"A",1:"B",2:"C",3:"D"},inplace=True)
print(df)
'''
  company  gender  age
A       B  female   40
B       A  female   31
C       B    male   28
D       B    male   28
'''




#举例2、set_index()将一列或多列设置为索引
print(df.set_index('company'))
'''
将company此列设置为索引
         gender  age
company             
B        female   40
A        female   31
B          male   28
B          male   28
'''

print(df.set_index(['company','gender']))
'''
将company和gender 同时设置为索引
company gender     
B       female   40
A       female   31
B       male     28
        male     28
'''

df.set_index(['company','gender'],inplace=True)
print(df)
'''
传入inplace=True ，使其真正修改原df对象
company gender     
B       female   40
A       female   31
B       male     28
        male     28
'''





#举例3、重置索引reset_index()
print(df.reset_index(drop=False))
'''
参数解释：
drop=False 删除原索引，但是原索引的列会被保留

  company  gender  age
0       B  female   40
1       A  female   31
2       B    male   28
3       B    male   28
'''

print(df.reset_index(drop=True))
'''
参数解释：
drop=True 删除原索引，原索引的列也会被删除
   age
0   40
1   31
2   28
3   28
'''

df.reset_index(drop=False,inplace=True)
print(df)
'''
inplace=True，使其真正修改原df对象
  company  gender  age
0       B  female   40
1       A  female   31
2       B    male   28
3       B    male   28
'''




#举例4、去重，drop_duplicates(subset,keep)
'''
参数解释：
subset=？指定对于某列进行去重判断，然后删除重复的行
keep=first/last 
     first表示对于重复的行删除时，保留排在首位的，默认为first
     last表示对于重复的行删除时，保留排在末尾的。
'''
print(df.drop_duplicates(subset=["company"]))
'''
解释：指定company作为重复的值判断，删除第3和4行
  company  gender  age
0       B  female   40
1       A  female   31
'''

print(df.drop_duplicates())
'''
解释：未指定某列，则将所有列作为重复的中判断，删除第4行
  company  gender  age
0       B  female   40
1       A  female   31
2       B    male   28
'''

print(df.drop_duplicates(subset=["company"],keep="last"))
'''
解释：指定company作为重复的值判断，同时指定keep="last"，保留排在末尾的行所以保留了第4行
  company  gender  age
1       A  female   31
3       B    male   28
'''





#举例5、isin() 常用于构建bool值，对DataFrame的数据进行筛选
print(df)
'''
  company  gender  age
0       B  female   40
1       A  female   31
2       B    male   28
3       B    male   28
'''
print(df.gender.isin(['male']))
print(df.loc[df.gender.isin(['male'])])
'''
0    False
1    False
2     True
3     True
Name: gender, dtype: bool

  company gender  age
2       B   male   28
3       B   male   28
'''

print(df.loc[df.company.isin(['A'])])
'''
  company  gender  age
1       A  female   31
'''