Simple Example with Planetoid class Fails at Data Fetch

[ajpowelsnl]

Example

A simple example fails on data fetch step.

Versions

$ conda -V
conda 24.11.2

$ conda list pytorch
# packages in environment at /opt/anaconda3:
#
# Name                    Version                   Build  Channel
pytorch                   2.4.1                  py3.12_0    pytorch
pytorch-scatter           2.1.2           py312_torch_2.4.0_cpu    pyg
pytorch-sparse            0.6.18          py312_torch_2.4.0_cpu    pyg
pytorch_geometric         2.5.3              pyhd8ed1ab_0    conda-forge

Failing Command

dataset = Planetoid(root='data/Planetoid', name='Cora', transform=NormalizeFeatures())

Error

In [32]: dataset = Planetoid(root="data/Planetoid", name="Cora", split="public", transform=NormalizeFeatures
    ...: ())
Downloading https://github.com/kimiyoung/planetoid/raw/master/data/ind.cora.x
Downloading https://github.com/kimiyoung/planetoid/raw/master/data/ind.cora.tx
Downloading https://github.com/kimiyoung/planetoid/raw/master/data/ind.cora.allx
Downloading https://github.com/kimiyoung/planetoid/raw/master/data/ind.cora.y
Downloading https://github.com/kimiyoung/planetoid/raw/master/data/ind.cora.ty
Downloading https://github.com/kimiyoung/planetoid/raw/master/data/ind.cora.ally
Downloading https://github.com/kimiyoung/planetoid/raw/master/data/ind.cora.graph
Downloading https://github.com/kimiyoung/planetoid/raw/master/data/ind.cora.test.index
Processing...
---------------------------------------------------------------------------
UnpicklingError                           Traceback (most recent call last)
Cell In[32], line 1
----> 1 dataset = Planetoid(root="data/Planetoid", name="Cora", split="public", transform=NormalizeFeatures())

File /opt/anaconda3/lib/python3.12/site-packages/torch_geometric/datasets/planetoid.py:102, in Planetoid.__init__(self, root, name, split, num_train_per_class, num_val, num_test, transform, pre_transform, force_reload)
     99 self.split = split.lower()
    100 assert self.split in ['public', 'full', 'geom-gcn', 'random']
--> 102 super().__init__(root, transform, pre_transform,
    103                  force_reload=force_reload)
    104 self.load(self.processed_paths[0])
    106 if split == 'full':

File /opt/anaconda3/lib/python3.12/site-packages/torch_geometric/data/in_memory_dataset.py:81, in InMemoryDataset.__init__(self, root, transform, pre_transform, pre_filter, log, force_reload)
     72 def __init__(
     73     self,
     74     root: Optional[str] = None,
   (...)
     79     force_reload: bool = False,
     80 ) -> None:
---> 81     super().__init__(root, transform, pre_transform, pre_filter, log,
     82                      force_reload)
     84     self._data: Optional[BaseData] = None
     85     self.slices: Optional[Dict[str, Tensor]] = None

File /opt/anaconda3/lib/python3.12/site-packages/torch_geometric/data/dataset.py:115, in Dataset.__init__(self, root, transform, pre_transform, pre_filter, log, force_reload)
    112     self._download()
    114 if self.has_process:
--> 115     self._process()

File /opt/anaconda3/lib/python3.12/site-packages/torch_geometric/data/dataset.py:260, in Dataset._process(self)
    257     print('Processing...', file=sys.stderr)
    259 fs.makedirs(self.processed_dir, exist_ok=True)
--> 260 self.process()
    262 path = osp.join(self.processed_dir, 'pre_transform.pt')
    263 fs.torch_save(_repr(self.pre_transform), path)

File /opt/anaconda3/lib/python3.12/site-packages/torch_geometric/datasets/planetoid.py:161, in Planetoid.process(self)
    160 def process(self) -> None:
--> 161     data = read_planetoid_data(self.raw_dir, self.name)
    163     if self.split == 'geom-gcn':
    164         train_masks, val_masks, test_masks = [], [], []

File /opt/anaconda3/lib/python3.12/site-packages/torch_geometric/io/planetoid.py:27, in read_planetoid_data(folder, prefix)
     25 def read_planetoid_data(folder: str, prefix: str) -> Data:
     26     names = ['x', 'tx', 'allx', 'y', 'ty', 'ally', 'graph', 'test.index']
---> 27     items = [read_file(folder, prefix, name) for name in names]
     28     x, tx, allx, y, ty, ally, graph, test_index = items
     29     train_index = torch.arange(y.size(0), dtype=torch.long)

File /opt/anaconda3/lib/python3.12/site-packages/torch_geometric/io/planetoid.py:107, in read_file(folder, prefix, name)
    105 with fsspec.open(path, 'rb') as f:
    106     warnings.filterwarnings('ignore', '.*`scipy.sparse.csr` name.*')
--> 107     out = pickle.load(f, encoding='latin1')
    109 if name == 'graph':
    110     return out

UnpicklingError: invalid load key, '<'.

Many thanks for any guidance you can offer.