Define and use class methods from variable

Question

Let's say I have two classes with some methods.

First class:

@dataclass
class MyFirstClass:
    y1: List[float]
    y2: List[float]
    
    @property
    def my_func1(self) -> List[float]:
        return self.y1-self.y2

    @property
    def my_func2(self) -> List[float]:
        return self.y1+self.y2

    @property
    def my_func3(self) -> List[float]:
        return self.y1*self.y2

Second class:

@dataclass
class MySecondClass:
    list_vals: List[float]
    
    @property
    def norm1(self):
        return self.list_vals
 
    @property
    def norm2(self):
        return self.list_vals / numpy.mean(self.list_vals)

So my question is two fold. Say in my main I have something like:

def main():
    FUNCTION = <?>
    NORM = <?>

    y1 = [1, 2, 3, 4, 5]
    y2 = [3, 4, 5, 6, 7]

    test = MyFirstClass(y1, y2).my_func1

Then I just call MyFirstClass with the y1 and y2 lists and get an output.

But as can be seen in the beginning of main I have FUNCTION and NORM. Is there any way to call MyFirstClass from there, and then reuse it all through main, i.e. something like:

FUNCTION = MyFirstClass.my_func1

test = FUNCTION(y1, y2)

This doesn't work obviously. So how can one do that ?

Also, and this is probably a build upon the above, how can I, once again, choose the norm function to be used in the other class ? For instance, if I update MyFirstClass a bit to:

@dataclass
    class MyFirstClassUpdated:
        y1: List[float]
        y2: List[float]
        norm: NORM
        
        @property
        def my_func1(self) -> List[float]:
            return norm(self.y1)-norm(self.y2)
    
        @property
        def my_func2(self) -> List[float]:
            return norm(self.y1)+norm(self.y2)
    
        @property
        def my_func3(self) -> List[float]:
            return norm(self.y1)*norm(self.y2)

And then when the class is called it takes the NORM argument from the main function, i.e. something like:

FUNCTION = MyFirstClass.func1
NORM = MySecondClass.norm1

test = FUNCTION(y1, y2, NORM)

I have no idea if this is even possible without making the code "uglier".

y1 and y2 are arguments to MyFirstClass, not MyFirstClass.my_func1. The closest to what you are looking for is to make FUNCTION a bound method: FUNCTION = MyFirstClass(y1, y2).my_func1; FUNCTION(). — chepner
– chepner, Commented Jun 20, 2022 at 21:54

Barmar · Accepted Answer · 2022-06-20 21:53:42Z

1

You need to create an instance of MyFirstClass, and pass it as the self argument to the function.

FUNCTION = MyFirstClass.my_func1
mfc = MyFirstClass(y1, y2)
test = FUNCTION(mfc)

answered Jun 20, 2022 at 21:53

Barmar

789k57 gold badges555 silver badges669 bronze badges

Sign up to request clarification or add additional context in comments.

5 Comments

Denver Dang Over a year ago

That seemed to do the trick. But how would I go about providing the NORM function into that class as well ?

Barmar Over a year ago

You need to do it similarly. You have to make an instance of MySecondClass to call the NORM function on.

Barmar Over a year ago

Where is the list_vals argument to MySecondClass in the second example?

Denver Dang Over a year ago

I'm not sure I understand your last comment. In the second example the list_vals are just the self.y1 and self.y2. MySecondClass is unchanged compared to MyFirstClassUpdated. I just want to use the MySecondClass and its methods in the MyFirstClassUpdated class. That was not used in the first MyFirstClass.

Barmar Over a year ago

Why are you trying to do this with dataclasses? It seems like you just want ordinary functions that you can pass arguments to.

Collectives™ on Stack Overflow

Define and use class methods from variable

1 Answer 1

5 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

5 Comments

Your Answer

Sign up or log in

Post as a guest

Related