Search notes:

torch.nn.Parameter

torch.nn.Parameter is a subclass of torch.Tensor.

If an instance of torch.nn.Parameter is assigned to an attribute of a torch.nn.Module, the parameters are added to the list of the module's parameters (and can be accessed through the module's .parameter() method:

import torch

class M(torch.nn.Module):

  def __init__(self):
      super().__init__()

m = M()

m.p1 = torch.nn.Parameter(torch.tensor([ [1.1,2.2] , [3.3,4.4] ]))
m.p2 = torch.nn.Parameter(torch.tensor([ [5.5,6.6] , [7.7,8.8] ]))

for param in (m.parameters()):
    print(param)

Index