Hi,
I want to know if we can see the applied modules given the output tensor.
More precisely, given the following code
import torch
import torch.nn as nn
my_input = torch.rand(4,12)
my_ln = nn.Linear(12, 4)
my_output = my_ln(my_input)
Can we extract nn.Linear
module just from my_output
tensor? I think we can extract the operations applied with grad_fn
, but I am asking about modules.