Is there an issue with the description of the shape of the returned tensor H?

benedekrozemberczki / pytorch_geometric_temporal

PyTorch Geometric Temporal: Spatiotemporal Signal Processing with Neural Machine Learning Models (CIKM 2021)

MIT License

2.67k stars 377 forks source link

Is there an issue with the description of the shape of the returned tensor H? #282

Open sonrisa07 opened 1 month ago

sonrisa07 commented 1 month ago

https://github.com/benedekrozemberczki/pytorch_geometric_temporal/blob/b2833bec8a0ba3e0072969cf4505e97c25315e78/torch_geometric_temporal/nn/attention/stgcn.py#L34-L37

Initially, X.permute(0, 3, 2, 1) is applied, and again at the end. Aren't the dimensions still consistent with the input shape: (batch_size, input_time_steps, num_nodes, in_channels)?