0

我正在尝试使用 ssp ( https://github.com/russellizadi/ssp ) 重新减少数据集 Cora 的 SOTA。但是我得到了以下异常：

Traceback (most recent call last):
  File "D:/School/Papers/GNN/2021_05_06 Node Classification SOTA/ssp/experiments/gcn.py", line 96, in <module>
    dataset = get_planetoid_dataset(name=args.dataset, normalize_features=args.normalize_features, split=args.split)
  File "...\ssp\experiments\datasets.py", line 17, in get_planetoid_dataset
    dataset = Planetoid(path, name, split=split)
  File "...\site-packages\torch_geometric\datasets\planetoid.py", line 55, in __init__
    super(Planetoid, self).__init__(root, transform, pre_transform)
  File "...\site-packages\torch_geometric\data\in_memory_dataset.py", line 54, in __init__
    pre_filter)
  File "...\site-packages\torch_geometric\data\dataset.py", line 92, in __init__
    self._process()
  File "...\site-packages\torch_geometric\data\dataset.py", line 165, in _process
    self.process()
  File "...\site-packages\torch_geometric\datasets\planetoid.py", line 108, in process
    data = read_planetoid_data(self.raw_dir, self.name)
  File "...\site-packages\torch_geometric\io\planetoid.py", line 19, in read_planetoid_data
    items = [read_file(folder, prefix, name) for name in names]
  File "...\site-packages\torch_geometric\io\planetoid.py", line 19, in <listcomp>
    items = [read_file(folder, prefix, name) for name in names]
  File "...\site-packages\torch_geometric\io\planetoid.py", line 66, in read_file
    out = pickle.load(f, encoding='latin1')
EOFError: Ran out of input

我打印路径并确保我拥有所有数据集文件。所以我想知道为什么会发生这种情况，以及如何处理它？谢谢。

4

0 回答 0