I have executed the training code, but I do not understand why the assertion statement is written as assert torch.equal(w, s) instead of assert s, s.
Additionally, after constructing the graph, I encountered a bug during training where the graph is missing the 'tffrac' property.
The traceback error is as follows:
Traceback (most recent call last):
File "/home/annq/source/HSG/HeterSumGraph/train.py", line 388, in
main()
File "/home/annq/source/HSG/HeterSumGraph/train.py", line 384, in main
setup_training(model, train_loader, valid_loader, valid_dataset, hps)
File "/home/annq/source/HSG/HeterSumGraph/train.py", line 72, in setup_training
run_training(model, train_loader, valid_loader, valset, hps, train_dir)
File "/home/annq/source/HSG/HeterSumGraph/train.py", line 115, in run_training
outputs = model.forward(G) # [n_snodes, 2]
File "/home/annq/source/HSG/HeterSumGraph/HiGraph.py", line 94, in forward
word_feature = self.set_wnfeature(graph) # [wnode, embed_size]
File "/home/annq/source/HSG/HeterSumGraph/HiGraph.py", line 150, in set_wnfeature
etf = graph.edges[wsedge_id].data["tffrac"]
File "/home/annq/anaconda3/envs/HSG_env/lib/python3.7/site-packages/dgl/view.py", line 128, in getitem
return self._graph.get_e_repr(self._edges)[key]
File "/home/annq/anaconda3/envs/HSG_env/lib/python3.7/site-packages/dgl/utils.py", line 284, in getitem
raise KeyError(key)
KeyError: 'tffrac'
Could you please clarify the purpose of assert torch.equal(w, s), and provide guidance on resolving the missing 'tffrac' property in the graph?
I have executed the training code, but I do not understand why the assertion statement is written as assert torch.equal(w, s) instead of assert s, s.
Additionally, after constructing the graph, I encountered a bug during training where the graph is missing the 'tffrac' property.
The traceback error is as follows:
Could you please clarify the purpose of assert torch.equal(w, s), and provide guidance on resolving the missing 'tffrac' property in the graph?