dgl._ffi.base.DGLError: Cannot concatenate column test_mask with shape Scheme(shape=(), dtype=torch.bool) and shape Scheme(shape=(), dtype=torch.uint8)

[nahsimai]

Hello Julian,

I tried to follow the installation steps mentioned in your repository. However, while trying to generate the results, I get the following error message:

Processed 5980000 tuples, found 2831587 valid tuples.
Adding reverse edges ...
Creating one whole graph ...
Total #nodes: 1885136
Total #edges: 5668682
Convert to heterograph ...
#Node types: 7
#Canonical edge types: 108
#Unique edge type names: 96
Load training/validation/testing split ...
Done saving data into cached files.
Traceback (most recent call last):
  File "main.py", line 192, in <module>
    metric = train(config)
  File "main.py", line 85, in train
    graph, ndata=["labels", "test_mask", "train_mask"])
  File "/home/bourne/anaconda3/envs/scene/lib/python3.7/site-packages/dgl/convert.py", line 974, in to_homogeneous
    comb_nf = combine_frames(G._node_frames, range(len(G.ntypes)), col_names=ndata)
  File "/home/bourne/anaconda3/envs/scene/lib/python3.7/site-packages/dgl/heterograph.py", line 6188, in combine_frames
    (key, frame.schemes[key], scheme))
dgl._ffi.base.DGLError: Cannot concatenate column test_mask with shape Scheme(shape=(), dtype=torch.bool) and shape Scheme(shape=(), dtype=torch.uint8)

The error message occurs across multiple datasets. There seems to be a missing shape value somewhere. Do you have any pointers how to solve this?