Collectives™ on Stack Overflow

Find centralized, trusted content and collaborate around the technologies you use most.

Teams

Q&A for work

Connect and share knowledge within a single location that is structured and easy to search.

class DeepGraphInfomax(torch.nn.Module):

def __init__(self, hidden_channels, pos_summary):#, encoder):#, summary, corruption):
    super().__init__()
    self.hidden_channels = hidden_channels
    #self.encoder = GCNEncoder() # needs to be defined either here or give it to here (my comment)
    self.pos_summary = pos_summary
    #self.corruption = corruption
    self.weight = Parameter(torch.Tensor(hidden_channels, hidden_channels))
    #self.reset_parameters()
    #pos_z = self.encoder(*args, **kwargs)
    #cor = self.corruption(*args, **kwargs)
    #cor = cor if isinstance(cor, tuple) else (cor, )
    #neg_z = self.encoder(*cor)
    summary = self.summary(pos_summary, *args, **kwargs)
    return summary# pos_z#, neg_z, summary
but running model() gives me the error:
TypeError: forward() missing 1 required positional argument:
'pos_summary'
                While instantiating DeepGraphInfomax, __init__ is expecting another parameter pos_summary; but you are not providing it.
– mshsayem
                Apr 2, 2022 at 13:02
model is an object since you instantiated DeepGraphInfomax.
model() calls the .__call__ function.
forward is called in the .__call__ function i.e. model().
Have a look at here.
The TypeError means that you should write input in forward function i.e.  model(data).
Here is an exmaple:
import torch
import torch.nn as nn
class example(nn.Module):
    def __init__(self):
        super().__init__()
        self.mlp = nn.Sequential(nn.Linear(5,2), nn.ReLU(), nn.Linear(2,1), nn.Sigmoid())
    def forward(self, x):
        return self.mlp(x)
# instantiate object
test = example()
input_data = torch.tensor([1.0,2.0,3.0,4.0,5.0])
# () and forward() are equal
print(test(input_data)) 
print(test.forward(input_data))
# outputs for me
#(tensor([0.5387], grad_fn=<SigmoidBackward>),
# tensor([0.5387], grad_fn=<SigmoidBackward>))
                yes, i understand that: but your last sentence is unclear to me: "The TypeError means that you should write input in forward function i.e. model(data)." so i should write model(data) as a positional argument in my forward function?
– Dr. S
                Apr 11, 2022 at 15:15
                You should write the "positional argument" in your forward function. The type error means that you miss the input of the forward function.
– Ethan Si
                Apr 12, 2022 at 10:56
        
Thanks for contributing an answer to Stack Overflow!
Please be sure to answer the question. Provide details and share your research!
But avoid …
Asking for help, clarification, or responding to other answers.
Making statements based on opinion; back them up with references or personal experience.
To learn more, see our tips on writing great answers.