我可以使用applymap来更改数据帧的变量名称

Question

If I want to change variable names in a data frame using pandas can I change the names without using pandas.df.rename() function but by using applymap() For example

Registrar   Enrolment Agency    State   District    Sub District    Pin Code    Gender
Allahabad Bank  Tera Software Ltd   Jharkhand   Ranchi  Namkum  834003  M
Allahabad Bank  Tera Software Ltd   Jharkhand   Ranchi  Ranchi  834004  F
Allahabad Bank  Vakrangee Softwares Limited Gujarat Surat   Nizar   394380  M

I need to fill spaces in the variable names in the above data frame with "_" and all the variable names should be in lower case

Answer 1

Function applymap is used for change all values of DataFrame element-wise, so for change column names are another methods.

I think you need list comprehension with python str function lower and replace (it does not work if NaN ):

df.columns = [col.lower().replace(' ', '_') for col in df.columns]
print (df)
        registrar     enrolment_agency            state district sub_district  \
0  Allahabad Bank    Tera Software Ltd        Jharkhand   Ranchi       Namkum   
1  Allahabad Bank    Tera Software Ltd        Jharkhand   Ranchi       Ranchi   
2  Allahabad Bank  Vakrangee Softwares  Limited Gujarat    Surat        Nizar   

   pin_code gender  
0    834003      M  
1    834004      F  
2    394380      M  

Or solution with pandas str function lower and replace :

df.columns = df.columns.str.replace(' ', '_').str.lower()
print (df)
        registrar     enrolment_agency            state district sub_district  \
0  Allahabad Bank    Tera Software Ltd        Jharkhand   Ranchi       Namkum   
1  Allahabad Bank    Tera Software Ltd        Jharkhand   Ranchi       Ranchi   
2  Allahabad Bank  Vakrangee Softwares  Limited Gujarat    Surat        Nizar   

   pin_code gender  
0    834003      M  
1    834004      F  
2    394380      M  

EDIT:

If need change column names by applymap it is impossible because this function is not implemented for Index ( column names ). But if really want something similar use map :

df.columns = df.columns.map(lambda col: col.lower().replace(' ', '_'))
print (df)
        registrar     enrolment_agency            state district sub_district  \
0  Allahabad Bank    Tera Software Ltd        Jharkhand   Ranchi       Namkum   
1  Allahabad Bank    Tera Software Ltd        Jharkhand   Ranchi       Ranchi   
2  Allahabad Bank  Vakrangee Softwares  Limited Gujarat    Surat        Nizar   

   pin_code gender  
0    834003      M  
1    834004      F  
2    394380      M  

Answer 2

You could do that in 3 ways.

Rewriting the columns

df.columns = [x.lower().replace(' ', '_') for x in df.columns] 

Using string methods

df.columns = df.columns.str.lower().str.replace(' ', '_')

Using rename

df.rename(columns=lambda x: x.lower().replace(' ', '_'))

Answer 3

Try this:

df.applymap(lambda x: str(x).lower().replace(' ', '_'))

applymap() will change the spaces to underscore in the data. If you want to change it in columns, then you can take jezrael's or John's answer.

Answer 4

Seriously... pick @JohnGalt's answer. rename is the way to go.

But because I like to try to add something other answers haven't covered yet:

Use split then join

df.columns = df.columns.str.lower().str.split().str.join('_')
df

Split and join has an advantage of replacing more than 1 space with underscore _