get AttributeError: grad_fn when use make_dot

[Pinnh]

when I use following code to visualize the torch pretrained resnet18 model, I get AttributeError: grad_fn, at line: add_nodes(var.grad_fn)

input=torch.randn(1,3,224,224)  
model=models.resnet18(pretrained=True)
for param in model.parameters():
     param.requires_grad = True
params = model.state_dict()
y=model(Variable(input))
print y
dot=make_dot(y, params)

[szagoruyko]

@Pinnh that's because model.state_dict() returns a dict of tensors, whereas make_dot expects a dict of Variables. I don't think there's a method that returns such dict for a model, so for now the best solution would probably be to make the params argument optional.