Why is it necessary to load "shenzhen" state dict for GWN?

[KL4805]

Dear authors,

I followed your code and run it with graph-wavenet. In the file maml.py, I saw

        if self.model_name == 'GWN':
            maml_model = MetaGWN(self.model_args, self.task_args)
            maml_model.load_state_dict(torch.load('shenzhen_gwn_model.pkl'))
            print("model load successfully.")
            maml_model = maml_model.cuda()

I wonder why is it necessary to load the state dict, especially for GWN model?

[RobinLu1209]

Thank you for your interest in our work. We are very sorry that this is a legacy problem in the code debugging phase, for any model is not needed to load the model. We will comb the code again and update the codebase in the near future.