Error on running examples/proteins_diff_pool.py

[ylf1109]

Running shows the error as:

AssertionError                            Traceback (most recent call last)
C:\Users\ADMINI~1\AppData\Local\Temp/ipykernel_15128/437079654.py in <module>
    113     device = torch.device('cpu')
    114 
--> 115 model = Net().to(device)
    116 optimizer = torch.optim.Adam(model.parameters(), lr=0.001)
    117 

C:\Users\ADMINI~1\AppData\Local\Temp/ipykernel_15128/437079654.py in __init__(self)
     85 
     86         self.lin1 = torch.nn.Linear(3 * 64, 64)
---> 87         self.lin2 = torch.nn.Linear(64, dataset.num_classes)
     88 
     89     def forward(self, x, adj, mask=None):

~\AppData\Local\conda\conda\envs\pytorch\lib\site-packages\torch_geometric\data\in_memory_dataset.py in num_classes(self)
     62         if self.transform is None:
     63             return self._infer_num_classes(self.data.y)
---> 64         return super().num_classes
     65 
     66     def len(self) -> int:

~\AppData\Local\conda\conda\envs\pytorch\lib\site-packages\torch_geometric\data\dataset.py in num_classes(self)
    150     def num_classes(self) -> int:
    151         r"""Returns the number of classes in the dataset."""
--> 152         y = torch.cat([data.y for data in self], dim=0)
    153         # Do not fill cache for `InMemoryDataset`:
    154         if hasattr(self, '_data_list') and self._data_list is not None:

~\AppData\Local\conda\conda\envs\pytorch\lib\site-packages\torch_geometric\data\dataset.py in <listcomp>(.0)
    150     def num_classes(self) -> int:
    151         r"""Returns the number of classes in the dataset."""
--> 152         y = torch.cat([data.y for data in self], dim=0)
    153         # Do not fill cache for `InMemoryDataset`:
    154         if hasattr(self, '_data_list') and self._data_list is not None:

~\AppData\Local\conda\conda\envs\pytorch\lib\site-packages\torch_geometric\data\dataset.py in __getitem__(self, idx)
    238 
    239             data = self.get(self.indices()[idx])
--> 240             data = data if self.transform is None else self.transform(data)
    241             return data
    242 

~\AppData\Local\conda\conda\envs\pytorch\lib\site-packages\torch_geometric\transforms\to_dense.py in __call__(self, data)
     28             num_nodes = orig_num_nodes
     29         else:
---> 30             assert orig_num_nodes <= self.num_nodes
     31             num_nodes = self.num_nodes
     32 

AssertionError: 

What should I change to make it work?

[wsad1]

The latest version of the example worked, but I have pyg installed from master.

[ylf1109]

Thank you, I reinstalled PyG from master and it works fine!