tests failed

[beew]

🐛 Describe the bug

After installing pytorch_geometric I ran the tests and got a lot of failures for some tests.

There are these

================================== FAILURES ===================================
_________________ test_pyg_lib_and_torch_sparse_homo_equality __________________

    @withPackage('pyg_lib', 'torch_sparse')
    def test_pyg_lib_and_torch_sparse_homo_equality():
        edge_index = get_random_edge_index(20, 20, 100)
        adj = to_torch_csr_tensor(edge_index.flip([0]), size=(20, 20))
        colptr, row = adj.crow_indices(), adj.col_indices()

        seed = torch.arange(10)

        sample = torch.ops.pyg.neighbor_sample
        out1 = sample(colptr, row, seed, [-1, -1], None, None, None, True)
        sample = torch.ops.torch_sparse.neighbor_sample
        out2 = sample(colptr, row, seed, [-1, -1], False, True)

        row1, col1, node_id1, edge_id1 = out1[:4]
        node_id2, row2, col2, edge_id2 = out2
        assert torch.equal(node_id1, node_id2)
>       assert torch.equal(row1, row2)
E       assert False
E        +  where False = <built-in method equal of type object at 0x7f81b2330e20>(tensor([ 0,  0,  0,  0,  1,  1,  1,  1,  1,  2,  2,  2,  2,  3,  3,  3,  3,  4,\n         4,  4,  4,  4,  4,  4,  5,  5,  5,  5,  6,  6,  7,  7,  7,  7,  7,  8,\n         8,  8,  8,  8,  8,  9,  9,  9,  9, 10, 10, 10, 10, 10, 10, 11, 11, 11,\n        11, 12, 12, 13, 13, 13, 13, 13, 14, 14, 14, 14, 15, 15, 15, 15, 16, 16,\n        16, 16, 16, 17, 17, 17, 18, 18, 18, 18, 18, 18]), tensor([ 0,  8, 10, 11,  3,  6,  7, 12, 13,  2,  7, 14, 15,  4, 10, 16, 15,  1,\n         2,  5, 17, 12, 15, 11,  3,  4,  7, 11,  1,  4,  3,  8, 10, 14, 11,  3,\n         4,  8, 17, 13, 18,  2,  4,  7, 17,  0,  2, 17, 12, 16, 13, 10, 12, 15,\n        18,  7,  9,  0,  6, 10, 12, 11, 16, 13, 11, 18,  2,  4,  5, 18,  5,  7,\n        17, 14, 18,  9, 16, 11,  3,  8,  9, 16, 15, 18]))
E        +    where <built-in method equal of type object at 0x7f81b2330e20> = torch.equal

test/loader/test_neighbor_loader.py:540: AssertionError
________________ test_pyg_lib_and_torch_sparse_hetero_equality _________________

    @withPackage('pyg_lib', 'torch_sparse')
    def test_pyg_lib_and_torch_sparse_hetero_equality():
        edge_index = get_random_edge_index(20, 10, 50)
        adj = to_torch_csr_tensor(edge_index.flip([0]), size=(10, 20))
        colptr1, row1 = adj.crow_indices(), adj.col_indices()

        edge_index = get_random_edge_index(10, 20, 50)
        adj = to_torch_csr_tensor(edge_index.flip([0]), size=(20, 10))
        colptr2, row2 = adj.crow_indices(), adj.col_indices()

        node_types = ['paper', 'author']
        edge_types = [('paper', 'to', 'author'), ('author', 'to', 'paper')]
        colptr_dict = {
            'paper__to__author': colptr1,
            'author__to__paper': colptr2,
        }
        row_dict = {
            'paper__to__author': row1,
            'author__to__paper': row2,
        }
        seed_dict = {'paper': torch.arange(1)}
        num_neighbors_dict = {
            'paper__to__author': [-1, -1],
            'author__to__paper': [-1, -1],
        }

        sample = torch.ops.pyg.hetero_neighbor_sample
>       out1 = sample(node_types, edge_types, colptr_dict, row_dict, seed_dict,
                      num_neighbors_dict, None, None, None, True, False, True,
                      False, "uniform", True)

test/loader/test_neighbor_loader.py:572: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 

self = <OpOverloadPacket(op='pyg.hetero_neighbor_sample')>
args = (['paper', 'author'], [('paper', 'to', 'author'), ('author', 'to', 'paper')], {'author__to__paper': tensor([ 0,  3,  8... 8, 15, 17, 19,  8, 14])}, {'paper': tensor([0])}, {'author__to__paper': [-1, -1], 'paper__to__author': [-1, -1]}, ...)
kwargs = {}

    def __call__(self, *args, **kwargs):
        # overloading __call__ to ensure torch.ops.foo.bar()
        # is still callable from JIT
        # We save the function ptr as the `op` attribute on
        # OpOverloadPacket to access it here.
>       return self._op(*args, **kwargs or {})
E       RuntimeError: pyg::hetero_neighbor_sample() expected at most 14 argument(s) but received 15 argument(s). Declaration: pyg::hetero_neighbor_sample(str[] node_types, (str, str, str)[] edge_types, Dict(str, Tensor) rowptr_dict, Dict(str, Tensor) col_dict, Dict(str, Tensor) seed_dict, Dict(str, int[]) num_neighbors_dict, Dict(str, Tensor)? time_dict=None, Dict(str, Tensor)? seed_time_dict=None, bool csc=False, bool replace=False, bool directed=True, bool disjoint=False, str temporal_strategy="uniform", bool return_edge_id=True) -> (Dict(str, Tensor), Dict(str, Tensor), Dict(str, Tensor), Dict(str, Tensor)?, Dict(str, int[]), Dict(str, int[]))

../../lib/python3.10/site-packages/torch/_ops.py:502: RuntimeError
________________________ test_index_select_out_of_range ________________________

    def test_index_select_out_of_range():
        with pytest.raises(IndexError, match="out of range"):
>           index_select(torch.randn(3, 5), torch.tensor([0, 2, 3]))

test/loader/test_utils.py:16: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 

value = tensor([[ 0.9994,  0.1966,  0.1530,  1.1206, -0.4225],
        [ 1.7494,  0.4751,  0.0236, -0.0791, -1.5103],
        [ 1.2039, -2.7800,  0.6247, -0.5723,  0.4597]])
index = tensor([0, 2, 3]), dim = 0

    def index_select(
        value: FeatureTensorType,
        index: Tensor,
        dim: int = 0,
    ) -> Tensor:
        r"""Indexes the :obj:`value` tensor along dimension :obj:`dim` using the
        entries in :obj:`index`.

        Args:
            value (torch.Tensor or np.ndarray): The input tensor.
            index (torch.Tensor): The 1-D tensor containing the indices to index.
            dim (int, optional): The dimension in which to index.
                (default: :obj:`0`)

        .. warning::

            :obj:`index` is casted to a :obj:`torch.int64` tensor internally, as
            `PyTorch currently only supports indexing
            <https://github.com/pytorch/pytorch/issues/61819>`_ via
            :obj:`torch.int64`.
        """
        # PyTorch currently only supports indexing via `torch.int64`:
        # https://github.com/pytorch/pytorch/issues/61819
        index = index.to(torch.int64)

        if isinstance(value, Tensor):
            out: Optional[Tensor] = None
            if torch.utils.data.get_worker_info() is not None:
                # If we are in a background process, we write directly into a
                # shared memory tensor to avoid an extra copy:
                size = list(value.shape)
                size[dim] = index.numel()
                numel = math.prod(size)
                if torch_geometric.typing.WITH_PT20:
                    storage = value.untyped_storage()._new_shared(
                        numel * value.element_size())
                else:
                    storage = value.storage()._new_shared(numel)
                out = value.new(storage).view(size)

>           return torch.index_select(value, dim, index, out=out)
E           RuntimeError: INDICES element is out of DATA bounds, id=3 axis_dim=3

../../lib/python3.10/site-packages/torch_geometric/loader/utils.py:73: RuntimeError
=============================== warnings summary ===============================
../../lib/python3.10/site-packages/torch/utils/tensorboard/__init__.py:4
  /home/bernard/opt/python310/lib/python3.10/site-packages/torch/utils/tensorboard/__init__.py:4: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
    if not hasattr(tensorboard, "__version__") or LooseVersion(

../../lib/python3.10/site-packages/torch/utils/tensorboard/__init__.py:6
  /home/bernard/opt/python310/lib/python3.10/site-packages/torch/utils/tensorboard/__init__.py:6: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
    ) < LooseVersion("1.15"):

../../lib/python3.10/site-packages/torch_geometric/contrib/__init__.py:8
  /home/bernard/opt/python310/lib/python3.10/site-packages/torch_geometric/contrib/__init__.py:8: UserWarning: 'torch_geometric.contrib' contains experimental code and is subject to change. Please use with caution.
    warnings.warn(

test/data/test_batch.py::test_torch_sparse_batch[layout0]
  /home/bernard/opt/python310/Examples/pytorch_geometric/test/data/test_batch.py:517: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:54.)
    out = to_edge_index(batch.adj.to_sparse(layout=torch.sparse_csr))

test/data/test_batch.py::test_torch_nested_batch
  /home/bernard/opt/python310/Examples/pytorch_geometric/test/data/test_batch.py:530: UserWarning: The PyTorch API of nested tensors is in prototype stage and will change in the near future. (Triggered internally at ../aten/src/ATen/NestedTensorImpl.cpp:177.)
    x1 = nested_tensor([torch.randn(3), torch.randn(4)])

test/explain/algorithm/test_captum.py: 6 warnings
test/explain/algorithm/test_captum_explainer.py: 40 warnings
test/explain/algorithm/test_captum_hetero.py: 6 warnings
  /home/bernard/opt/python310/lib/python3.10/site-packages/captum/attr/_core/guided_backprop_deconvnet.py:64: UserWarning: Setting backward hooks on ReLU activations.The hooks will be removed after the attribution is finished
    warnings.warn(

test/explain/algorithm/test_captum_explainer.py: 56 warnings
test/explain/algorithm/test_captum_hetero.py: 8 warnings
  /home/bernard/opt/python310/lib/python3.10/site-packages/captum/_utils/gradient.py:57: UserWarning: Input Tensor 0 did not already require gradients, required_grads has been set automatically.
    warnings.warn(

test/explain/algorithm/test_captum_explainer.py: 8 warnings
test/explain/algorithm/test_captum_hetero.py: 8 warnings
  /home/bernard/opt/python310/lib/python3.10/site-packages/captum/_utils/gradient.py:57: UserWarning: Input Tensor 1 did not already require gradients, required_grads has been set automatically.
    warnings.warn(

test/nn/dense/test_linear.py::test_hetero_linear[device0]
  /home/bernard/opt/python310/lib/python3.10/site-packages/torch/jit/_check.py:172: UserWarning: The TorchScript type system doesn't support instance-level annotations on empty non-base types in `__init__`. Instead, either 1) use a type annotation in the class body, or 2) wrap the type in `torch.jit.Attribute`.
    warnings.warn("The TorchScript type system doesn't support "

test/transforms/test_node_property_split.py::test_node_property_split[popularity]
test/transforms/test_node_property_split.py::test_node_property_split[locality]
test/transforms/test_node_property_split.py::test_node_property_split[locality]
  /home/bernard/opt/python310/lib/python3.10/site-packages/networkx/algorithms/link_analysis/pagerank_alg.py:108: DeprecationWarning: networkx.pagerank_scipy is deprecated and will be removed in NetworkX 3.0, use networkx.pagerank instead.
    return pagerank_scipy(

-- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html
- generated Nunit xml file: /home/bernard/opt/python310/Examples/pytorch_geometric/test-output.xml -
=========================== short test summary info ============================
FAILED test/loader/test_neighbor_loader.py::test_pyg_lib_and_torch_sparse_homo_equality - assert False
FAILED test/loader/test_neighbor_loader.py::test_pyg_lib_and_torch_sparse_hetero_equality - RuntimeError: pyg::hetero_neighbor_sample() expected at most 14 argument(s)...
FAILED test/loader/test_utils.py::test_index_select_out_of_range - RuntimeError: INDICES element is out of DATA bounds, id=3 axis_dim=3

But a lot more are from test/explain/algorithm/test_graphmask_explainer.py. They are all like this

========================== short test summary info ============================
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-node-model-MaskType.object-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-node-model-MaskType.object-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-node-model-MaskType.common_attributes-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-node-model-MaskType.common_attributes-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-node-model-MaskType.attributes-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-node-model-MaskType.attributes-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-node-phenomenon-MaskType.object-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-node-phenomenon-MaskType.object-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-node-phenomenon-MaskType.common_attributes-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-node-phenomenon-MaskType.common_attributes-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-node-phenomenon-MaskType.attributes-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-node-phenomenon-MaskType.attributes-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-edge-model-MaskType.object-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-edge-model-MaskType.object-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-edge-model-MaskType.common_attributes-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-edge-model-MaskType.common_attributes-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-edge-model-MaskType.attributes-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-edge-model-MaskType.attributes-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-edge-phenomenon-MaskType.object-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-edge-phenomenon-MaskType.object-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-edge-phenomenon-MaskType.common_attributes-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-edge-phenomenon-MaskType.common_attributes-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-edge-phenomenon-MaskType.attributes-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-edge-phenomenon-MaskType.attributes-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-graph-model-MaskType.object-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-graph-model-MaskType.object-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-graph-model-MaskType.common_attributes-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-graph-model-MaskType.common_attributes-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-graph-model-MaskType.attributes-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-graph-model-MaskType.attributes-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-graph-phenomenon-MaskType.object-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-graph-phenomenon-MaskType.object-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-graph-phenomenon-MaskType.common_attributes-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-graph-phenomenon-MaskType.common_attributes-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-graph-phenomenon-MaskType.attributes-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-probs-graph-phenomenon-MaskType.attributes-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-raw-node-model-MaskType.object-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-raw-node-model-MaskType.object-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-raw-node-model-MaskType.common_attributes-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-raw-node-model-MaskType.common_attributes-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-raw-node-model-MaskType.attributes-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-raw-node-model-MaskType.attributes-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-raw-node-phenomenon-MaskType.object-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-raw-node-phenomenon-MaskType.object-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-raw-node-phenomenon-MaskType.common_attributes-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-raw-node-phenomenon-MaskType.common_attributes-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-raw-node-phenomenon-MaskType.attributes-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-raw-node-phenomenon-MaskType.attributes-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-raw-edge-model-MaskType.object-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-raw-edge-model-MaskType.object-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-raw-edge-model-MaskType.common_attributes-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-raw-edge-model-MaskType.common_attributes-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-raw-edge-model-MaskType.attributes-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-raw-edge-model-MaskType.attributes-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-raw-edge-phenomenon-MaskType.object-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-raw-edge-phenomenon-MaskType.object-None] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...
FAILED test/explain/algorithm/test_graphmask_explainer.py::test_graph_mask_explainer_binary_classification[None-raw-edge-phenomenon-MaskType.common_attributes-MaskType.object] - IndexError: Dimension out of range (expected to be in range of [-2, 1], but...

Examples of some more detailed logs on test_graphmask_explainer errors (there are too many)

_ test_graph_mask_explainer_binary_classification[None-probs-node-model-MaskType.object-MaskType.object] _

edge_mask_type = <MaskType.object: 'object'>
node_mask_type = <MaskType.object: 'object'>, explanation_type = 'model'
task_level = 'node', return_type = 'probs', index = None

    @pytest.mark.parametrize('edge_mask_type', edge_mask_types)
    @pytest.mark.parametrize('node_mask_type', node_mask_types)
    @pytest.mark.parametrize('explanation_type', ['model', 'phenomenon'])
    @pytest.mark.parametrize('task_level', ['node', 'edge', 'graph'])
    @pytest.mark.parametrize('return_type', ['probs', 'raw'])
    @pytest.mark.parametrize('index', [None, 2, torch.arange(3)])
    def test_graph_mask_explainer_binary_classification(
        edge_mask_type,
        node_mask_type,
        explanation_type,
        task_level,
        return_type,
        index,
    ):
        model_config = ModelConfig(
            mode='binary_classification',
            task_level=task_level,
            return_type=return_type,
        )

        model = GCN(model_config)

        target = None
        if explanation_type == 'phenomenon':
            with torch.no_grad():
                out = model(x, edge_index, batch, edge_label_index)
                if model_config.return_type == ModelReturnType.raw:
                    target = (out > 0).long().view(-1)
                if model_config.return_type == ModelReturnType.probs:
                    target = (out > 0.5).long().view(-1)

        explainer = Explainer(
            model=model,
            algorithm=GraphMaskExplainer(2, epochs=5, log=False),
            explanation_type=explanation_type,
            node_mask_type=node_mask_type,
            edge_mask_type=edge_mask_type,
            model_config=model_config,
        )

>       explanation = explainer(
            x,
            edge_index,
            target=target,
            index=index,
            batch=batch,
            edge_label_index=edge_label_index,
        )

test/explain/algorithm/test_graphmask_explainer.py:133: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
../../lib/python3.10/site-packages/torch_geometric/explain/explainer.py:204: in __call__
    explanation = self.algorithm(
../../lib/python3.10/site-packages/torch/nn/modules/module.py:1501: in _call_impl
    return forward_call(*args, **kwargs)
../../lib/python3.10/site-packages/torch_geometric/explain/algorithm/graphmask_explainer.py:133: in forward
    edge_mask = self._explain(model, index=index)
../../lib/python3.10/site-packages/torch_geometric/explain/algorithm/graphmask_explainer.py:515: in _explain
    result = self.gates[(i * 4) + 1][j](partial)
../../lib/python3.10/site-packages/torch/nn/modules/module.py:1501: in _call_impl
    return forward_call(*args, **kwargs)
../../lib/python3.10/site-packages/torch/nn/modules/normalization.py:190: in forward
    return F.layer_norm(
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 

input = tensor([[-0.1499],
        [-0.2280],
        [-0.2280],
        [-0.1674],
        [-0.1674],
        [-0.0529],
    ...80],
        [-0.1674],
        [-0.0529],
        [ 0.0813],
        [-0.0922],
        [-0.0853],
        [-0.1870]])
normalized_shape = (1,), weight = Parameter containing:
tensor([1.])
bias = Parameter containing:
tensor([0.]), eps = 1e-05

    def layer_norm(
        input: Tensor,
        normalized_shape: List[int],
        weight: Optional[Tensor] = None,
        bias: Optional[Tensor] = None,
        eps: float = 1e-5,
    ) -> Tensor:
        r"""Applies Layer Normalization for last certain number of dimensions.

        See :class:`~torch.nn.LayerNorm` for details.
        """
        if has_torch_function_variadic(input, weight, bias):
            return handle_torch_function(
                layer_norm, (input, weight, bias), input, normalized_shape, weight=weight, bias=bias, eps=eps
            )
>       return torch.layer_norm(input, normalized_shape, weight, bias, eps, torch.backends.cudnn.enabled)
E       IndexError: Dimension out of range (expected to be in range of [-2, 1], but got 2)

../../lib/python3.10/site-packages/torch/nn/functional.py:2515: IndexError
_ test_graph_mask_explainer_binary_classification[None-probs-node-model-MaskType.object-None] _

edge_mask_type = None, node_mask_type = <MaskType.object: 'object'>
explanation_type = 'model', task_level = 'node', return_type = 'probs'
index = None

    @pytest.mark.parametrize('edge_mask_type', edge_mask_types)
    @pytest.mark.parametrize('node_mask_type', node_mask_types)
    @pytest.mark.parametrize('explanation_type', ['model', 'phenomenon'])
    @pytest.mark.parametrize('task_level', ['node', 'edge', 'graph'])
    @pytest.mark.parametrize('return_type', ['probs', 'raw'])
    @pytest.mark.parametrize('index', [None, 2, torch.arange(3)])
    def test_graph_mask_explainer_binary_classification(
        edge_mask_type,
        node_mask_type,
        explanation_type,
        task_level,
        return_type,
        index,
    ):
        model_config = ModelConfig(
            mode='binary_classification',
            task_level=task_level,
            return_type=return_type,
        )

        model = GCN(model_config)

        target = None
        if explanation_type == 'phenomenon':
            with torch.no_grad():
                out = model(x, edge_index, batch, edge_label_index)
                if model_config.return_type == ModelReturnType.raw:
                    target = (out > 0).long().view(-1)
                if model_config.return_type == ModelReturnType.probs:
                    target = (out > 0.5).long().view(-1)

        explainer = Explainer(
            model=model,
            algorithm=GraphMaskExplainer(2, epochs=5, log=False),
            explanation_type=explanation_type,
            node_mask_type=node_mask_type,
            edge_mask_type=edge_mask_type,
            model_config=model_config,
        )

>       explanation = explainer(
            x,
            edge_index,
            target=target,
            index=index,
            batch=batch,
            edge_label_index=edge_label_index,
        )

test/explain/algorithm/test_graphmask_explainer.py:133: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
../../lib/python3.10/site-packages/torch_geometric/explain/explainer.py:204: in __call__
    explanation = self.algorithm(
../../lib/python3.10/site-packages/torch/nn/modules/module.py:1501: in _call_impl
    return forward_call(*args, **kwargs)
../../lib/python3.10/site-packages/torch_geometric/explain/algorithm/graphmask_explainer.py:133: in forward
    edge_mask = self._explain(model, index=index)
../../lib/python3.10/site-packages/torch_geometric/explain/algorithm/graphmask_explainer.py:515: in _explain
    result = self.gates[(i * 4) + 1][j](partial)
../../lib/python3.10/site-packages/torch/nn/modules/module.py:1501: in _call_impl
    return forward_call(*args, **kwargs)
../../lib/python3.10/site-packages/torch/nn/modules/normalization.py:190: in forward
    return F.layer_norm(
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 

input = tensor([[ 0.0005],
        [ 0.0181],
        [ 0.0181],
        [ 0.0122],
        [ 0.0122],
        [ 0.0016],
    ...81],
        [ 0.0122],
        [ 0.0016],
        [-0.0261],
        [-0.0473],
        [-0.0218],
        [-0.0230]])
normalized_shape = (1,), weight = Parameter containing:
tensor([1.])
bias = Parameter containing:
tensor([0.]), eps = 1e-05

    def layer_norm(
        input: Tensor,
        normalized_shape: List[int],
        weight: Optional[Tensor] = None,
        bias: Optional[Tensor] = None,
        eps: float = 1e-5,
    ) -> Tensor:
        r"""Applies Layer Normalization for last certain number of dimensions.

        See :class:`~torch.nn.LayerNorm` for details.
        """
        if has_torch_function_variadic(input, weight, bias):
            return handle_torch_function(
                layer_norm, (input, weight, bias), input, normalized_shape, weight=weight, bias=bias, eps=eps
            )
>       return torch.layer_norm(input, normalized_shape, weight, bias, eps, torch.backends.cudnn.enabled)
E       IndexError: Dimension out of range (expected to be in range of [-2, 1], but got 2)

../../lib/python3.10/site-packages/torch/nn/functional.py:2515: IndexError
_ test_graph_mask_explainer_binary_classification[None-probs-node-model-MaskType.common_attributes-MaskType.object] _

edge_mask_type = <MaskType.object: 'object'>
node_mask_type = <MaskType.common_attributes: 'common_attributes'>
explanation_type = 'model', task_level = 'node', return_type = 'probs'
index = None

    @pytest.mark.parametrize('edge_mask_type', edge_mask_types)
    @pytest.mark.parametrize('node_mask_type', node_mask_types)
    @pytest.mark.parametrize('explanation_type', ['model', 'phenomenon'])
    @pytest.mark.parametrize('task_level', ['node', 'edge', 'graph'])
    @pytest.mark.parametrize('return_type', ['probs', 'raw'])
    @pytest.mark.parametrize('index', [None, 2, torch.arange(3)])
    def test_graph_mask_explainer_binary_classification(
        edge_mask_type,
        node_mask_type,
        explanation_type,
        task_level,
        return_type,
        index,
    ):
        model_config = ModelConfig(
            mode='binary_classification',
            task_level=task_level,
            return_type=return_type,
        )

        model = GCN(model_config)

        target = None
        if explanation_type == 'phenomenon':
            with torch.no_grad():
                out = model(x, edge_index, batch, edge_label_index)
                if model_config.return_type == ModelReturnType.raw:
                    target = (out > 0).long().view(-1)
                if model_config.return_type == ModelReturnType.probs:
                    target = (out > 0.5).long().view(-1)

        explainer = Explainer(
            model=model,
            algorithm=GraphMaskExplainer(2, epochs=5, log=False),
            explanation_type=explanation_type,
            node_mask_type=node_mask_type,
            edge_mask_type=edge_mask_type,
            model_config=model_config,
        )

>       explanation = explainer(
            x,
            edge_index,
            target=target,
            index=index,
            batch=batch,
            edge_label_index=edge_label_index,
        )

test/explain/algorithm/test_graphmask_explainer.py:133: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
../../lib/python3.10/site-packages/torch_geometric/explain/explainer.py:204: in __call__
    explanation = self.algorithm(
../../lib/python3.10/site-packages/torch/nn/modules/module.py:1501: in _call_impl
    return forward_call(*args, **kwargs)
../../lib/python3.10/site-packages/torch_geometric/explain/algorithm/graphmask_explainer.py:133: in forward
    edge_mask = self._explain(model, index=index)
../../lib/python3.10/site-packages/torch_geometric/explain/algorithm/graphmask_explainer.py:515: in _explain
    result = self.gates[(i * 4) + 1][j](partial)
../../lib/python3.10/site-packages/torch/nn/modules/module.py:1501: in _call_impl
    return forward_call(*args, **kwargs)
../../lib/python3.10/site-packages/torch/nn/modules/normalization.py:190: in forward
    return F.layer_norm(
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 

input = tensor([[ 0.2942],
        [ 0.3121],
        [ 0.3121],
        [ 0.6925],
        [ 0.6925],
        [-1.1122],
    ...21],
        [ 0.6925],
        [-1.1122],
        [-0.4176],
        [-0.3225],
        [ 0.4694],
        [ 0.0182]])
normalized_shape = (1,), weight = Parameter containing:
tensor([1.])
bias = Parameter containing:
tensor([0.]), eps = 1e-05

    def layer_norm(
        input: Tensor,
        normalized_shape: List[int],
        weight: Optional[Tensor] = None,
        bias: Optional[Tensor] = None,
        eps: float = 1e-5,
    ) -> Tensor:
        r"""Applies Layer Normalization for last certain number of dimensions.

        See :class:`~torch.nn.LayerNorm` for details.
        """
        if has_torch_function_variadic(input, weight, bias):
            return handle_torch_function(
                layer_norm, (input, weight, bias), input, normalized_shape, weight=weight, bias=bias, eps=eps
            )
>       return torch.layer_norm(input, normalized_shape, weight, bias, eps, torch.backends.cudnn.enabled)
E       IndexError: Dimension out of range (expected to be in range of [-2, 1], but got 2)

../../lib/python3.10/site-packages/torch/nn/functional.py:2515: IndexError
_ test_graph_mask_explainer_binary_classification[None-probs-node-model-MaskType.common_attributes-None] _

edge_mask_type = None
node_mask_type = <MaskType.common_attributes: 'common_attributes'>
explanation_type = 'model', task_level = 'node', return_type = 'probs'
index = None

    @pytest.mark.parametrize('edge_mask_type', edge_mask_types)
    @pytest.mark.parametrize('node_mask_type', node_mask_types)
    @pytest.mark.parametrize('explanation_type', ['model', 'phenomenon'])
    @pytest.mark.parametrize('task_level', ['node', 'edge', 'graph'])
    @pytest.mark.parametrize('return_type', ['probs', 'raw'])
    @pytest.mark.parametrize('index', [None, 2, torch.arange(3)])
    def test_graph_mask_explainer_binary_classification(
        edge_mask_type,
        node_mask_type,
        explanation_type,
        task_level,
        return_type,
        index,
    ):
        model_config = ModelConfig(
            mode='binary_classification',
            task_level=task_level,
            return_type=return_type,
        )

        model = GCN(model_config)

        target = None
        if explanation_type == 'phenomenon':
            with torch.no_grad():
                out = model(x, edge_index, batch, edge_label_index)
                if model_config.return_type == ModelReturnType.raw:
                    target = (out > 0).long().view(-1)
                if model_config.return_type == ModelReturnType.probs:
                    target = (out > 0.5).long().view(-1)

        explainer = Explainer(
            model=model,
            algorithm=GraphMaskExplainer(2, epochs=5, log=False),
            explanation_type=explanation_type,
            node_mask_type=node_mask_type,
            edge_mask_type=edge_mask_type,
            model_config=model_config,
        )

>       explanation = explainer(
            x,
            edge_index,
            target=target,
            index=index,
            batch=batch,
            edge_label_index=edge_label_index,
        )

test/explain/algorithm/test_graphmask_explainer.py:133: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
../../lib/python3.10/site-packages/torch_geometric/explain/explainer.py:204: in __call__
    explanation = self.algorithm(
../../lib/python3.10/site-packages/torch/nn/modules/module.py:1501: in _call_impl
    return forward_call(*args, **kwargs)
../../lib/python3.10/site-packages/torch_geometric/explain/algorithm/graphmask_explainer.py:133: in forward
    edge_mask = self._explain(model, index=index)
../../lib/python3.10/site-packages/torch_geometric/explain/algorithm/graphmask_explainer.py:515: in _explain
    result = self.gates[(i * 4) + 1][j](partial)
../../lib/python3.10/site-packages/torch/nn/modules/module.py:1501: in _call_impl
    return forward_call(*args, **kwargs)
../../lib/python3.10/site-packages/torch/nn/modules/normalization.py:190: in forward
    return F.layer_norm(
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 

input = tensor([[-0.7821],
        [ 0.3164],
        [ 0.3164],
        [-0.6021],
        [-0.6021],
        [ 0.3111],
    ...64],
        [-0.6021],
        [ 0.3111],
        [-0.2455],
        [-0.5185],
        [-0.6260],
        [-0.8519]])
normalized_shape = (1,), weight = Parameter containing:
tensor([1.])
bias = Parameter containing:
tensor([0.]), eps = 1e-05

    def layer_norm(
        input: Tensor,
        normalized_shape: List[int],
        weight: Optional[Tensor] = None,
        bias: Optional[Tensor] = None,
        eps: float = 1e-5,
    ) -> Tensor:
        r"""Applies Layer Normalization for last certain number of dimensions.

        See :class:`~torch.nn.LayerNorm` for details.
        """
        if has_torch_function_variadic(input, weight, bias):
            return handle_torch_function(
                layer_norm, (input, weight, bias), input, normalized_shape, weight=weight, bias=bias, eps=eps
            )
>       return torch.layer_norm(input, normalized_shape, weight, bias, eps, torch.backends.cudnn.enabled)
E       IndexError: Dimension out of range (expected to be in range of [-2, 1], but got 2)

../../lib/python3.10/site-packages/torch/nn/functional.py:2515: IndexError
_ test_graph_mask_explainer_binary_classification[None-probs-node-model-MaskType.attributes-MaskType.object] _

edge_mask_type = <MaskType.object: 'object'>
node_mask_type = <MaskType.attributes: 'attributes'>, explanation_type = 'model'
task_level = 'node', return_type = 'probs', index = None

    @pytest.mark.parametrize('edge_mask_type', edge_mask_types)
    @pytest.mark.parametrize('node_mask_type', node_mask_types)
    @pytest.mark.parametrize('explanation_type', ['model', 'phenomenon'])
    @pytest.mark.parametrize('task_level', ['node', 'edge', 'graph'])
    @pytest.mark.parametrize('return_type', ['probs', 'raw'])
    @pytest.mark.parametrize('index', [None, 2, torch.arange(3)])
    def test_graph_mask_explainer_binary_classification(
        edge_mask_type,
        node_mask_type,
        explanation_type,
        task_level,
        return_type,
        index,
    ):
        model_config = ModelConfig(
            mode='binary_classification',
            task_level=task_level,
            return_type=return_type,
        )

        model = GCN(model_config)

        target = None
        if explanation_type == 'phenomenon':
            with torch.no_grad():
                out = model(x, edge_index, batch, edge_label_index)
                if model_config.return_type == ModelReturnType.raw:
                    target = (out > 0).long().view(-1)
                if model_config.return_type == ModelReturnType.probs:
                    target = (out > 0.5).long().view(-1)

        explainer = Explainer(
            model=model,
            algorithm=GraphMaskExplainer(2, epochs=5, log=False),
            explanation_type=explanation_type,
            node_mask_type=node_mask_type,
            edge_mask_type=edge_mask_type,
            model_config=model_config,
        )

>       explanation = explainer(
            x,
            edge_index,
            target=target,
            index=index,
            batch=batch,
            edge_label_index=edge_label_index,
        )

test/explain/algorithm/test_graphmask_explainer.py:133: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
../../lib/python3.10/site-packages/torch_geometric/explain/explainer.py:204: in __call__
    explanation = self.algorithm(
../../lib/python3.10/site-packages/torch/nn/modules/module.py:1501: in _call_impl
    return forward_call(*args, **kwargs)
../../lib/python3.10/site-packages/torch_geometric/explain/algorithm/graphmask_explainer.py:133: in forward
    edge_mask = self._explain(model, index=index)
../../lib/python3.10/site-packages/torch_geometric/explain/algorithm/graphmask_explainer.py:515: in _explain
    result = self.gates[(i * 4) + 1][j](partial)
../../lib/python3.10/site-packages/torch/nn/modules/module.py:1501: in _call_impl
    return forward_call(*args, **kwargs)
../../lib/python3.10/site-packages/torch/nn/modules/normalization.py:190: in forward
    return F.layer_norm(
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 

input = tensor([[-0.0378],
        [-0.0682],
        [-0.0682],
        [-0.0394],
        [-0.0394],
        [-0.0939],
    ...82],
        [-0.0394],
        [-0.0939],
        [-0.0893],
        [-0.0754],
        [-0.0550],
        [-0.0433]])
normalized_shape = (1,), weight = Parameter containing:
tensor([1.])
bias = Parameter containing:
tensor([0.]), eps = 1e-05

    def layer_norm(
        input: Tensor,
        normalized_shape: List[int],
        weight: Optional[Tensor] = None,
        bias: Optional[Tensor] = None,
        eps: float = 1e-5,
    ) -> Tensor:
        r"""Applies Layer Normalization for last certain number of dimensions.

        See :class:`~torch.nn.LayerNorm` for details.
        """
        if has_torch_function_variadic(input, weight, bias):
            return handle_torch_function(
                layer_norm, (input, weight, bias), input, normalized_shape, weight=weight, bias=bias, eps=eps
            )
>       return torch.layer_norm(input, normalized_shape, weight, bias, eps, torch.backends.cudnn.enabled)
E       IndexError: Dimension out of range (expected to be in range of [-2, 1], but got 2)

../../lib/python3.10/site-packages/torch/nn/functional.py:2515: IndexError
_ test_graph_mask_explainer_binary_classification[None-probs-node-model-MaskType.attributes-None] _

edge_mask_type = None, node_mask_type = <MaskType.attributes: 'attributes'>
explanation_type = 'model', task_level = 'node', return_type = 'probs'
index = None

    @pytest.mark.parametrize('edge_mask_type', edge_mask_types)
    @pytest.mark.parametrize('node_mask_type', node_mask_types)
    @pytest.mark.parametrize('explanation_type', ['model', 'phenomenon'])
    @pytest.mark.parametrize('task_level', ['node', 'edge', 'graph'])
    @pytest.mark.parametrize('return_type', ['probs', 'raw'])
    @pytest.mark.parametrize('index', [None, 2, torch.arange(3)])
    def test_graph_mask_explainer_binary_classification(
        edge_mask_type,
        node_mask_type,
        explanation_type,
        task_level,
        return_type,
        index,
    ):
        model_config = ModelConfig(
            mode='binary_classification',
            task_level=task_level,
            return_type=return_type,
        )

        model = GCN(model_config)

        target = None
        if explanation_type == 'phenomenon':
            with torch.no_grad():
                out = model(x, edge_index, batch, edge_label_index)
                if model_config.return_type == ModelReturnType.raw:
                    target = (out > 0).long().view(-1)
                if model_config.return_type == ModelReturnType.probs:
                    target = (out > 0.5).long().view(-1)

        explainer = Explainer(
            model=model,
            algorithm=GraphMaskExplainer(2, epochs=5, log=False),
            explanation_type=explanation_type,
            node_mask_type=node_mask_type,
            edge_mask_type=edge_mask_type,
            model_config=model_config,
        )

>       explanation = explainer(
            x,
            edge_index,
            target=target,
            index=index,
            batch=batch,
            edge_label_index=edge_label_index,
        )

test/explain/algorithm/test_graphmask_explainer.py:133: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
../../lib/python3.10/site-packages/torch_geometric/explain/explainer.py:204: in __call__
    explanation = self.algorithm(
../../lib/python3.10/site-packages/torch/nn/modules/module.py:1501: in _call_impl
    return forward_call(*args, **kwargs)
../../lib/python3.10/site-packages/torch_geometric/explain/algorithm/graphmask_explainer.py:133: in forward
    edge_mask = self._explain(model, index=index)
../../lib/python3.10/site-packages/torch_geometric/explain/algorithm/graphmask_explainer.py:515: in _explain
    result = self.gates[(i * 4) + 1][j](partial)
../../lib/python3.10/site-packages/torch/nn/modules/module.py:1501: in _call_impl
    return forward_call(*args, **kwargs)
../../lib/python3.10/site-packages/torch/nn/modules/normalization.py:190: in forward
    return F.layer_norm(
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 

input = tensor([[ 0.0244],
        [ 0.0261],
        [ 0.0261],
        [-0.0057],
        [-0.0057],
        [-0.0456],
    ...61],
        [-0.0057],
        [-0.0456],
        [-0.0711],
        [-0.0126],
        [ 0.0970],
        [ 0.0654]])
normalized_shape = (1,), weight = Parameter containing:
tensor([1.])
bias = Parameter containing:
tensor([0.]), eps = 1e-05

    def layer_norm(
        input: Tensor,
        normalized_shape: List[int],
        weight: Optional[Tensor] = None,
        bias: Optional[Tensor] = None,
        eps: float = 1e-5,
    ) -> Tensor:
        r"""Applies Layer Normalization for last certain number of dimensions.

        See :class:`~torch.nn.LayerNorm` for details.
        """
        if has_torch_function_variadic(input, weight, bias):
            return handle_torch_function(
                layer_norm, (input, weight, bias), input, normalized_shape, weight=weight, bias=bias, eps=eps
            )
>       return torch.layer_norm(input, normalized_shape, weight, bias, eps, torch.backends.cudnn.enabled)
E       IndexError: Dimension out of range (expected to be in range of [-2, 1], but got 2)

../../lib/python3.10/site-packages/torch/nn/functional.py:2515: IndexError
_ test_graph_mask_explainer_binary_classification[None-probs-node-phenomenon-MaskType.object-MaskType.object] _

edge_mask_type = <MaskType.object: 'object'>
node_mask_type = <MaskType.object: 'object'>, explanation_type = 'phenomenon'
task_level = 'node', return_type = 'probs', index = None

    @pytest.mark.parametrize('edge_mask_type', edge_mask_types)
    @pytest.mark.parametrize('node_mask_type', node_mask_types)
    @pytest.mark.parametrize('explanation_type', ['model', 'phenomenon'])
    @pytest.mark.parametrize('task_level', ['node', 'edge', 'graph'])
    @pytest.mark.parametrize('return_type', ['probs', 'raw'])
    @pytest.mark.parametrize('index', [None, 2, torch.arange(3)])
    def test_graph_mask_explainer_binary_classification(
        edge_mask_type,
        node_mask_type,
        explanation_type,
        task_level,
        return_type,
        index,
    ):
        model_config = ModelConfig(
            mode='binary_classification',
            task_level=task_level,
            return_type=return_type,
        )

        model = GCN(model_config)

        target = None
        if explanation_type == 'phenomenon':
            with torch.no_grad():
                out = model(x, edge_index, batch, edge_label_index)
                if model_config.return_type == ModelReturnType.raw:
                    target = (out > 0).long().view(-1)
                if model_config.return_type == ModelReturnType.probs:
                    target = (out > 0.5).long().view(-1)

        explainer = Explainer(
            model=model,
            algorithm=GraphMaskExplainer(2, epochs=5, log=False),
            explanation_type=explanation_type,
            node_mask_type=node_mask_type,
            edge_mask_type=edge_mask_type,
            model_config=model_config,
        )

>       explanation = explainer(
            x,
            edge_index,
            target=target,
            index=index,
            batch=batch,
            edge_label_index=edge_label_index,
        )

test/explain/algorithm/test_graphmask_explainer.py:133: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
../../lib/python3.10/site-packages/torch_geometric/explain/explainer.py:204: in __call__
    explanation = self.algorithm(
../../lib/python3.10/site-packages/torch/nn/modules/module.py:1501: in _call_impl
    return forward_call(*args, **kwargs)
../../lib/python3.10/site-packages/torch_geometric/explain/algorithm/graphmask_explainer.py:133: in forward
    edge_mask = self._explain(model, index=index)
../../lib/python3.10/site-packages/torch_geometric/explain/algorithm/graphmask_explainer.py:515: in _explain
    result = self.gates[(i * 4) + 1][j](partial)
../../lib/python3.10/site-packages/torch/nn/modules/module.py:1501: in _call_impl
    return forward_call(*args, **kwargs)
../../lib/python3.10/site-packages/torch/nn/modules/normalization.py:190: in forward
    return F.layer_norm(
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 

input = tensor([[ 0.0056],
        [ 0.0085],
        [ 0.0085],
        [ 0.0075],
        [ 0.0075],
        [ 0.0276],
    ...85],
        [ 0.0075],
        [ 0.0276],
        [ 0.0174],
        [ 0.0257],
        [-0.0016],
        [-0.0115]])
normalized_shape = (1,), weight = Parameter containing:
tensor([1.])
bias = Parameter containing:
tensor([0.]), eps = 1e-05

    def layer_norm(
        input: Tensor,
        normalized_shape: List[int],
        weight: Optional[Tensor] = None,
        bias: Optional[Tensor] = None,
        eps: float = 1e-5,
    ) -> Tensor:
        r"""Applies Layer Normalization for last certain number of dimensions.

        See :class:`~torch.nn.LayerNorm` for details.
        """
        if has_torch_function_variadic(input, weight, bias):
            return handle_torch_function(
                layer_norm, (input, weight, bias), input, normalized_shape, weight=weight, bias=bias, eps=eps
            )
>       return torch.layer_norm(input, normalized_shape, weight, bias, eps, torch.backends.cudnn.enabled)
E       IndexError: Dimension out of range (expected to be in range of [-2, 1], but got 2)

../../lib/python3.10/site-packages/torch/nn/functional.py:2515: IndexError
_ test_graph_mask_explainer_binary_classification[None-probs-node-phenomenon-MaskType.object-None] _

edge_mask_type = None, node_mask_type = <MaskType.object: 'object'>
explanation_type = 'phenomenon', task_level = 'node', return_type = 'probs'
index = None

    @pytest.mark.parametrize('edge_mask_type', edge_mask_types)
    @pytest.mark.parametrize('node_mask_type', node_mask_types)
    @pytest.mark.parametrize('explanation_type', ['model', 'phenomenon'])
    @pytest.mark.parametrize('task_level', ['node', 'edge', 'graph'])
    @pytest.mark.parametrize('return_type', ['probs', 'raw'])
    @pytest.mark.parametrize('index', [None, 2, torch.arange(3)])
    def test_graph_mask_explainer_binary_classification(
        edge_mask_type,
        node_mask_type,
        explanation_type,
        task_level,
        return_type,
        index,
    ):
        model_config = ModelConfig(
            mode='binary_classification',
            task_level=task_level,
            return_type=return_type,
        )

        model = GCN(model_config)

        target = None
        if explanation_type == 'phenomenon':
            with torch.no_grad():
                out = model(x, edge_index, batch, edge_label_index)
                if model_config.return_type == ModelReturnType.raw:
                    target = (out > 0).long().view(-1)
                if model_config.return_type == ModelReturnType.probs:
                    target = (out > 0.5).long().view(-1)

        explainer = Explainer(
            model=model,
            algorithm=GraphMaskExplainer(2, epochs=5, log=False),
            explanation_type=explanation_type,
            node_mask_type=node_mask_type,
            edge_mask_type=edge_mask_type,
            model_config=model_config,
        )

>       explanation = explainer(
            x,
            edge_index,
            target=target,
            index=index,
            batch=batch,
            edge_label_index=edge_label_index,
        )

test/explain/algorithm/test_graphmask_explainer.py:133: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
../../lib/python3.10/site-packages/torch_geometric/explain/explainer.py:204: in __call__
    explanation = self.algorithm(
../../lib/python3.10/site-packages/torch/nn/modules/module.py:1501: in _call_impl
    return forward_call(*args, **kwargs)
../../lib/python3.10/site-packages/torch_geometric/explain/algorithm/graphmask_explainer.py:133: in forward
    edge_mask = self._explain(model, index=index)
../../lib/python3.10/site-packages/torch_geometric/explain/algorithm/graphmask_explainer.py:515: in _explain
    result = self.gates[(i * 4) + 1][j](partial)
../../lib/python3.10/site-packages/torch/nn/modules/module.py:1501: in _call_impl
    return forward_call(*args, **kwargs)
../../lib/python3.10/site-packages/torch/nn/modules/normalization.py:190: in forward
    return F.layer_norm(
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 

input = tensor([[ 0.0648],
        [ 0.1313],
        [ 0.1313],
        [ 0.1122],
        [ 0.1122],
        [-0.0125],
    ...13],
        [ 0.1122],
        [-0.0125],
        [-0.0559],
        [ 0.0382],
        [ 0.0179],
        [ 0.0616]])
normalized_shape = (1,), weight = Parameter containing:
tensor([1.])
bias = Parameter containing:
tensor([0.]), eps = 1e-05

    def layer_norm(
        input: Tensor,
        normalized_shape: List[int],
        weight: Optional[Tensor] = None,
        bias: Optional[Tensor] = None,
        eps: float = 1e-5,
    ) -> Tensor:
        r"""Applies Layer Normalization for last certain number of dimensions.

        See :class:`~torch.nn.LayerNorm` for details.
        """
        if has_torch_function_variadic(input, weight, bias):
            return handle_torch_function(
                layer_norm, (input, weight, bias), input, normalized_shape, weight=weight, bias=bias, eps=eps
            )
>       return torch.layer_norm(input, normalized_shape, weight, bias, eps, torch.backends.cudnn.enabled)
E       IndexError: Dimension out of range (expected to be in range of [-2, 1], but got 2)

../../lib/python3.10/site-packages/torch/nn/functional.py:2515: IndexError

Environment

• PyG version: 2.4.0
• PyTorch version: 2.0.0+cpu
• OS: Ubuntu 22.04
• Python version: 3.10.9
• CUDA/cuDNN version:No cuda, use cpu
• How you installed PyTorch and PyG (conda, pip, source): pip
• Any other relevant information (e.g., version of torch-scatter): torch-scatter 2.1.1

[rusty1s]

I am not so sure about the GraphMaskExplainer test failures, but the neighbor loader tests should work if you upgrade to pyg-lib nightly.

[beew]

I am not so sure about the GraphMaskExplainer test failures, but the neighbor loader tests should work if you upgrade to pyg-lib nightly.

Upgrade to pyg-lib nightly still a lot of failed tests

============================================================================================= short test summary info =============================================================================================
FAILED test/data/lightning/test_datamodule.py::test_lightning_hetero_link_data_custom_store - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_cache.py::test_cached_loader[device0] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_cache.py::test_cached_loader_transform[device0] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_imbalanced_sampler.py::test_neighbor_loader_with_imbalanced_sampler - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_imbalanced_sampler.py::test_hetero_neighbor_loader_with_imbalanced_sampler - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_homo_link_neighbor_loader_basic[None-None-directional-device0] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_homo_link_neighbor_loader_basic[None-None-bidirectional-device0] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_homo_link_neighbor_loader_basic[None-1.0-directional-device0] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_homo_link_neighbor_loader_basic[None-1.0-bidirectional-device0] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_homo_link_neighbor_loader_basic[True-None-directional-device0] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_homo_link_neighbor_loader_basic[True-None-bidirectional-device0] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_homo_link_neighbor_loader_basic[True-1.0-directional-device0] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_homo_link_neighbor_loader_basic[True-1.0-bidirectional-device0] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_homo_link_neighbor_loader_basic[False-None-directional-device0] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_homo_link_neighbor_loader_basic[False-None-bidirectional-device0] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_homo_link_neighbor_loader_basic[False-1.0-directional-device0] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_homo_link_neighbor_loader_basic[False-1.0-bidirectional-device0] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_hetero_link_neighbor_loader_basic[None-directional] - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_hetero_link_neighbor_loader_basic[None-bidirectional] - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_hetero_link_neighbor_loader_basic[1.0-directional] - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_hetero_link_neighbor_loader_basic[1.0-bidirectional] - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_hetero_link_neighbor_loader_loop[directional] - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_hetero_link_neighbor_loader_loop[bidirectional] - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_link_neighbor_loader_edge_label - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_temporal_homo_link_neighbor_loader[1] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_temporal_hetero_link_neighbor_loader - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_custom_hetero_link_neighbor_loader - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_homo_link_neighbor_loader_no_edges - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_hetero_link_neighbor_loader_no_edges - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_homo_link_neighbor_loader_triplet[1-False-False] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_homo_link_neighbor_loader_triplet[1-False-True] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_homo_link_neighbor_loader_triplet[1-True-True] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_homo_link_neighbor_loader_triplet[2-False-False] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_homo_link_neighbor_loader_triplet[2-False-True] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_homo_link_neighbor_loader_triplet[2-True-True] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_hetero_link_neighbor_loader_triplet[1-False-False] - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_hetero_link_neighbor_loader_triplet[1-False-True] - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_hetero_link_neighbor_loader_triplet[1-True-True] - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_hetero_link_neighbor_loader_triplet[2-False-False] - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_hetero_link_neighbor_loader_triplet[2-False-True] - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_hetero_link_neighbor_loader_triplet[2-True-True] - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_link_neighbor_loader.py::test_link_neighbor_loader_mapping - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_homo_neighbor_loader_basic[None-dtype0-SubgraphType.directional-device0] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_homo_neighbor_loader_basic[None-dtype0-SubgraphType.bidirectional-device0] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_homo_neighbor_loader_basic[None-dtype1-SubgraphType.directional-device0] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_homo_neighbor_loader_basic[None-dtype1-SubgraphType.bidirectional-device0] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_homo_neighbor_loader_basic[True-dtype0-SubgraphType.directional-device0] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_homo_neighbor_loader_basic[True-dtype0-SubgraphType.bidirectional-device0] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_homo_neighbor_loader_basic[True-dtype1-SubgraphType.directional-device0] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_homo_neighbor_loader_basic[True-dtype1-SubgraphType.bidirectional-device0] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_homo_neighbor_loader_basic[False-dtype0-SubgraphType.directional-device0] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_homo_neighbor_loader_basic[False-dtype0-SubgraphType.bidirectional-device0] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_homo_neighbor_loader_basic[False-dtype1-SubgraphType.directional-device0] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_homo_neighbor_loader_basic[False-dtype1-SubgraphType.bidirectional-device0] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_hetero_neighbor_loader_basic[dtype0-SubgraphType.directional] - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_hetero_neighbor_loader_basic[dtype0-SubgraphType.bidirectional] - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_hetero_neighbor_loader_basic[dtype1-SubgraphType.directional] - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_hetero_neighbor_loader_basic[dtype1-SubgraphType.bidirectional] - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_homo_neighbor_loader_on_cora[SubgraphType.directional] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_homo_neighbor_loader_on_cora[SubgraphType.bidirectional] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_hetero_neighbor_loader_on_cora[SubgraphType.directional] - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_hetero_neighbor_loader_on_cora[SubgraphType.bidirectional] - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_temporal_hetero_neighbor_loader_on_cora - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_custom_neighbor_loader - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_temporal_custom_neighbor_loader_on_cora - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_pyg_lib_and_torch_sparse_homo_equality - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_pyg_lib_and_torch_sparse_hetero_equality - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_memmap_neighbor_loader - RuntimeError: Caught RuntimeError in DataLoader worker process 0.
FAILED test/loader/test_neighbor_loader.py::test_homo_neighbor_loader_sampled_info - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_hetero_neighbor_loader_sampled_info - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_neighbor_loader_mapping - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_weighted_homo_neighbor_loader - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_neighbor_loader.py::test_weighted_hetero_neighbor_loader - RuntimeError: pyg::hetero_neighbor_sample() Expected a value of type 'Optional[Dict[str, Tensor]]' for argument 'edge_weight_dict' but instead found type 'bool'.
FAILED test/loader/test_utils.py::test_index_select_out_of_range - RuntimeError: INDICES element is out of DATA bounds, id=3 axis_dim=3
FAILED test/loader/test_zip_loader.py::test_zip_loader[True] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/loader/test_zip_loader.py::test_zip_loader[False] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/nn/conv/test_pna_conv.py::test_pna_conv_get_degree_histogram_neighbor_loader - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/nn/models/test_basic_gnn.py::test_basic_gnn_inference[None] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/nn/models/test_basic_gnn.py::test_basic_gnn_inference[last] - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/nn/models/test_basic_gnn.py::test_trim_to_layer - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/nn/models/test_basic_gnn.py::test_basic_gnn_cache - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.
FAILED test/utils/test_trim_to_layer.py::test_trim_to_layer_with_neighbor_loader - RuntimeError: pyg::neighbor_sample() Expected a value of type 'Optional[Tensor]' for argument 'edge_weight' but instead found type 'bool'.

All tests passed for 2.3.1 so I am downgrading.

[rusty1s]

Are you sure you are testing on latest PyG? It looks like you use an earlier version of PyG to test against.

[beew]

Are you sure you are testing on latest PyG? It looks like you use an earlier version of PyG to test against.

I am pretty sure. The GraphMaskExplainer test is not in pyg < 2.4.0

[rusty1s]

Sorry, maybe this was confusing. Are you running the latest tests against PyG == 2.4.0 or against the latest version of PyG from master?

[beew]

Sorry, maybe this was confusing. Are you running the latest tests against PyG == 2.4.0 or against the latest version of PyG from master?

2.4.0. I grabbed the tests from the latest release tarball (instead of cloning from git)

[rusty1s]

Got it, if you are running on fixed 2.4.0, then you will need pyg-lib==0.3.*.

[sandeepmukh]

I'm seeing this Neighborloader issue even with pyg-lib==0.3.1. Is there another potential fix?

[rusty1s]

Are you sure you are not on pyg-lib==0.4.0 (just released yesterday)? Support for it will be added in the next PyG version which we will release ASAP.

[sandeepmukh]

Yes—I was originally, but I downgraded after seeing your message above.