search

chengsen / PyTorch_TextGCN

The PyTorch 1.6 and Python 3.7 implementation for the paper Graph Convolutional Networks for Text Classification
https://arxiv.org/abs/1809.05679
107 stars 21 forks source link

node features is a whole 1 array? #4

Open A11en0 opened 3 years ago

A11en0 commented 3 years ago

Hi, thanks for your wonderful implement by PyTorch. I have a question why does the node feature use an all 1 tensor(shape: (29426, 29426)) rather than one-hot?

image

A11en0 commented 3 years ago

And the size of the features should be N*D, which D is the length of vocabulary.