Super().__init__(x, y) vs super(Model, self).__init__()

I’m new to pytorch and i have few questions.

while defining a model, we have to provide “super().init()” inside constructor. then what is difference between

  1. super(Model, self).init()
  2. super().init(x, y)

If I’m not mistaken, you had to use the first approach in Python2.x, while both approaches should work in Python3.x (not sure about the exact minor versions).

For backward-compatibility reasons, you will probably see a lot of the first approach (and I’m also used to it), which wouldn’t be necessary anymore, as Python2 is dead by now. :slight_smile: