List of objects as numpy array

Question

I have a class with some fields, and I want it to be represented as numpy array.

class Circle:
    def __init__(self, cX, cY, rad, color):
        self.cX = cX
        self.cY = cY
        self.rad = rad
        self.color = color

    # some methods...

I know that in order to print I should define an __str__() method. Is there a similar mechanism for numpy arrays? In particular, I want something like this:

list_of_objs = [Circle() for i in range(100)]
numpy_representatin = numpy.asarray(list_of_objs) # shape have to be (4, 100)

Thanks in advance!

UPD: forgot to mention, that all fields are assumed to be of the same type - int

Jens Lindahl · Accepted Answer · 2020-04-06 10:01:53Z

3

I think the following should help you: Writing custom array containers

As mentioned in the above link:

"We can convert to a numpy array using numpy.array or numpy.asarray, which will call its __array__-method to obtain a standard numpy.ndarray."

Thus, your class could look something like

class Circle:
    def __init__(self, cX, cY, rad, color):
        self.cX = cX
        self.cY = cY
        self.rad = rad
        self.color = color

    def __array__(self):
        return np.array([self.cX, self.cY, self.rad, self.color], np.int32)

    # some methods...

edited Apr 6, 2020 at 10:01

answered Apr 6, 2020 at 9:54

Jens Lindahl

4104 silver badges13 bronze badges

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

Collectives™ on Stack Overflow

List of objects as numpy array

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related