pandas映射问题

使用案例1

import pandas as pd
df = pd.DataFrame({'key1' : ['a', 'a', 'b', 'b', 'a'],
                   'key2' : ['one', 'two', 'one', 'two', 'one'],
                   'data1' : np.arange(5),
                   'data2' : np.arange(5,10)})
print(df)
df['data1'] = df['data1'].map(lambda x : "%.3f"%x) #lambda在这里其实是在定义一个简单的函数，一个没有函数名的函数。
df# 将数字保持三位有效数字

  key1 key2  data1  data2
0    a  one      0      5
1    a  two      1      6
2    b  one      2      7
3    b  two      3      8
4    a  one      4      9
key1     object
key2     object
data1     int64
data2     int64
dtype: object

	key1	key2	data1	data2
0	a	one	0.000	5
1	a	two	1.000	6
2	b	one	2.000	7
3	b	two	3.000	8
4	a	one	4.000	9

使用案例2

df = pd.DataFrame({'key1' : ['a', 'a', 'b', 'b', 'a'],
                   'key2' : ['one', 'two', 'one', 'two', 'one'],
                   'data1' : np.arange(5),
                   'data2' : np.arange(5,10)})
print(df)
df['key1'] = df['key1'].map({'a':'c',"b":"d"})
df

  key1 key2  data1  data2
0    a  one      0      5
1    a  two      1      6
2    b  one      2      7
3    b  two      3      8
4    a  one      4      9

	key1	key2	data1	data2
0	c	one	0	5
1	c	two	1	6
2	d	one	2	7
3	d	two	3	8
4	c	one	4	9

使用案例3

df = pd.DataFrame({'key1' : ['a', 'a', 'b', 'b', 'a'],
                   'key2' : ['one', 'two', 'one', 'two', 'one'],
                   'data1' : np.arange(5),
                   'data2' : np.arange(5,10)})
print(df)
df['key1'] = df['key1'].map({'a':'b',"b":"a"})# 互相映射
df

  key1 key2  data1  data2
0    a  one      0      5
1    a  two      1      6
2    b  one      2      7
3    b  two      3      8
4    a  one      4      9

	key1	key2	data1	data2
0	b	one	0	5
1	b	two	1	6
2	a	one	2	7
3	a	two	3	8
4	b	one	4	9

使用案例4

df = pd.DataFrame({'key1' : ['a', 'a', 'b', 'b', 'a'],
                   'key2' : ['one', 'two', 'one', 'two', 'one'],
                   'data1' : np.arange(5),
                   'data2' : np.arange(5,10)})
print(df)
df['key1'] = df['key1'].map({'a':'aaaa'})# 互相映射
print(df)##
## 可以看到这里的问题是如果映射的值不是全部的key,那么不完整的key会被映射成Nan

  key1 key2  data1  data2
0    a  one      0      5
1    a  two      1      6
2    b  one      2      7
3    b  two      3      8
4    a  one      4      9
   key1 key2  data1  data2
0  aaaa  one      0      5
1  aaaa  two      1      6
2   NaN  one      2      7
3   NaN  two      3      8
4  aaaa  one      4      9

使用案例5

df = pd.DataFrame({'key1' : ['a', 'a', 'b', 'b', 'a'],
                   'key2' : ['one', 'two', 'one', 'two', 'one'],
                   'data1' : np.arange(5),
                   'data2' : np.arange(5,10)})
print(df)
df['key1'] = df['key1'].map({'a':'aaaa'}).fillna(df['key1'])
print(df)##
## 加入fillna就可以做到粗存在映射的键值保持原来一样

  key1 key2  data1  data2
0    a  one      0      5
1    a  two      1      6
2    b  one      2      7
3    b  two      3      8
4    a  one      4      9
   key1 key2  data1  data2
0  aaaa  one      0      5
1  aaaa  two      1      6
2     b  one      2      7
3     b  two      3      8
4  aaaa  one      4      9

类别变量映射成整型变量

import pandas as pd 
df = pd.DataFrame({'x':['good', 'bad', 'good', 'great']}, dtype='category')

df["level"] = df["x"].cat.codes
print(df)

map1=dict( zip(df["x"].cat.codes,df['x'] ) )
print(map1)

df["xx"] =df["level"].map(map1) ## 重新映射回去，判断是否争取
print(df)

这里使用cat.codes进行映射，但是要想获得这个映射，那么就应该使用dict的zip