RuntimeError: weight should contain 32 elements not 1 in mutag_gin.py example

[Gori-LV]

Question

I am trying out the example of HIN using MUTAG dataset but run into this error:

Traceback (most recent call last):
  File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/IPython/core/interactiveshell.py", line 3418, in run_code
    exec(code_obj, self.user_global_ns, self.user_ns)
  File "<ipython-input-2-96b1080fe156>", line 1, in <module>
    runfile('/Users/gori/Google Drive/heter-master/pytorch_geometric_hetero/hin_mutag.py', wdir='/Users/gori/Google Drive/heter-master/pytorch_geometric_hetero')
  File "/Applications/PyCharm CE.app/Contents/plugins/python-ce/helpers/pydev/_pydev_bundle/pydev_umd.py", line 198, in runfile
    pydev_imports.execfile(filename, global_vars, local_vars)  # execute the script
  File "/Applications/PyCharm CE.app/Contents/plugins/python-ce/helpers/pydev/_pydev_imps/_pydev_execfile.py", line 18, in execfile
    exec(compile(contents+"\n", file, 'exec'), glob, loc)
  File "/Users/gori/Google Drive/heter-master/pytorch_geometric_hetero/hin_mutag.py", line 91, in <module>
    loss = train()
  File "/Users/gori/Google Drive/heter-master/pytorch_geometric_hetero/hin_mutag.py", line 70, in train
    out = model(data.x, data.edge_index, data.batch)
  File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/Users/gori/Google Drive/heter-master/pytorch_geometric_hetero/hin_mutag.py", line 53, in forward
    x = conv(x, edge_index).relu()
  File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/torch_geometric/nn/conv/gin_conv.py", line 70, in forward
    return self.nn(out)
  File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/torch_geometric/nn/models/mlp.py", line 56, in forward
    x = norm(x)
  File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/torch/nn/modules/batchnorm.py", line 135, in forward
    return F.batch_norm(
  File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/torch/nn/functional.py", line 2146, in batch_norm
    return torch.batch_norm(
RuntimeError: weight should contain 32 elements not 1

Environment

• PyG version: 2.0.2
• PyTorch version: 1.8.0
• OS: MacOS
• Python version: 3.8
• How you installed PyTorch and PyG (conda, pip, source): pip

I suspect it is caused by inconsistent version of PyG and other libs too perhaps, but I can't upgrade them at the moment for avoiding affecting project works that are done already. Can anyone help on solving this issue without upgrading libraries? A billion thanks!

[rusty1s]

I think it is best to always use the examples of the respective version that you use, e.g., https://github.com/pyg-team/pytorch_geometric/blob/2.0.2/examples/mutag_gin.py. Can you confirm that running this example works for you?

[Gori-LV]

it's working perfectly fine! thank you!