Make a Function That Accepts An Unspecified Number of Columns & Filters in Python-CodePudding

Let's say we have a dataframe df:

column1 column2 column3
0   A1  1   2
1   A2  2   3
2   A3  3   4

I'd like to make a function that does filtering(s). However, I don't know how many columns I'll be filtering. Depending on the dataset, I could use two columns or just one. For instance, I want to keep column2 that's greater or equal than 2 i.e., df[df.column2 >= 2]. However, at different times, I want to do two filterings: keep column2 that's greater or equal than 2 i.e., df[df.column2 >= 2] AND keep column3 that's greater or equal than 4 i.e., df[df.column3 >= 4].

How does one capture all this? I think the col parameter should have an asterisk and a num paramter that specifies the number to filter. However, I don't know how to set an inequality as a parameter.

def select_filters(*col, num):
    return df.col

CodePudding user response：

There are two methods to approach this kind of problem, the first is to use a list as your parameter as such:

def select_filters(col, num):
   for i in range(num):
      #Do whatever your filtering is here
      return col[num]

select_filters([df.column1,df.column2,..],2)

Another option for a variable number of parameters would look like this:

def select_filters(num,*col):
      #Do whatever your filtering is here

However, you must ensure that the variable col comes AFTER num, and not before. I am unsure about your exact filtering, but this should be (hopefully) enough to solve your issue if I understood correctly.