CODEWITHSUNDEEP
python
Talha Anwar
Talha Anwar

Reputation: 2959

Call name of function having multiple parameters from a python dictionary to another function

I am trying to call function from python dictionary one by one into another function

feature={
1:[perm_entropy, 3, True],
2:[svd_entropy, 3, 1, True],
3:[petrosian_fd],
4:[katz_fd],
5:[higuchi_fd],
6:[hjorth_mobility],
7:[hjorth_complexity]}

i am trying to call above function as follows

sub_25=[]
for i in feature.values():
    sub_25.append(sub25.groupby('trial')[col1].agg(i))

Except the first 2 function, all other functions are working correctly. I thinks it is because of extra parameters they have. These two function work well, if they are called separately as follow

sub_25.append(sub25.groupby('trial')[col1].agg(perm_entropy, 3, True))

How can i deal these two functions? EEG data

Upvotes: 0

Views: 67

Answers (2)

Yayati Sule
Yayati Sule

Reputation: 1631

What you want is partially applied function for your use-case. Python's functools package can take care of this. You can change your code as per the following example:

from functools import partial

def addthree(a,b,c): 
    return a+b+c

partial_sum = partial(addthree,1,2)

fdict = {1: partial_sum}

fdict[1](4) // this returns 7 on the console

Similarly you can write perm_entropy with its required arguments and pass your values later

new_perm_entropy = partial(perm_entropy,3,True)
feature = {1: new_perm_entropy,..}
feature[1](your_args_here)

You can read up on Partially Applied Functions for further clarifications.

Upvotes: 1

SyntaxVoid
SyntaxVoid

Reputation: 2633

From the documentation (https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.agg.html), the pandas agg method accepts a function as the first argument, optional axis second, and then positional arguments which will be passed to the function specified.

To do this in your for loop, you should use * to unpack i since i is a list of the arguments you need to pass to agg.

sub_25=[]
for i in feature.values():
    sub_25.append(sub25.groupby('trial')[col1].agg(*i))

For example, imagine i = [perm_entropy, 3, True], then sub_25.append(sub25.groupby('trial')[col1].agg(*i)) is equivalent to sub_25.append(sub25.groupby('trial')[col1].agg(perm_entropy, 3, True)) which is what you are after.

Upvotes: 0

Related Questions