Changing forward of nn.Sequential

1

For illustration, I have condensed my code to the following snippet

import torch.nn as nn
import torch

class Model(nn.Module):
    def __init__(self):
        super(Model, self).__init__()
        self.func = nn.Linear(10, 5)
        
    def forward(self, x, y):
        print(y)
        return self.func(x)

model = nn.Sequential(*[Model()])

x = y = torch.randn(size=(20, 10))

model(x, y)

I get following error - TypeError: forward() takes 2 positional arguments but 3 were given

This is because nn.Sequential has its own forward which is accepting only one argument. How can I change/extend forward’s definition so that it accepts two arguments. Obviously I can accept both x and y as a list under single argument but my use case would make it look messy.

2

Maybe I’m wrong but why beating around the bush? Why create a Model and added in Sequential? If you give us more details it would be helpful? What is you exactly look for?

3

In my case the class is a Layer.

4

hey, try to use this:

import torch.nn as nn
import torch

class Model(nn.Module):
    def __init__(self):
        super(Model, self).__init__()
        self.func = nn.Linear(10, 5)
        
    def forward(self, x):
        print(x[1])
        return self.func(x[0])

model = nn.Sequential(*[Model()])

x = y = torch.randn(size=(20, 10))

model((x, y))