D2STGNN
D2STGNN copied to clipboard
zezhishao
Code for our VLDB'22 paper Decoupled Dynamic Spatial-Temporal Graph Neural Network for Traffic Forecasting.
RuntimeError: mat1 and mat2 shapes cannot be multiplied (307x12 and 24x64) 首先十分感谢作者开源，想请问我修改了seq_length为24，程序不能运行，也没有找到修改的地方，希望作者帮忙解惑，万分感谢
Hi, I am trying to reproduce the results. METR-LA and PEMS-BAY looks fine. But on PEMS08 the results is not good as expected. Is the adjacency matrix used for PEMS08...
Unexpected key(s) in state_dict: "layers.0.estimation_gate.FC1.weight", "layers.0.estimation_gate.FC1.bias", "layers.0.estimation_gate.FC2.weight", "layers.0.estimation_gate.FC2.bias", "layers.0.inh_layer.sub_and_norm.ln.weight", "layers.0.inh_layer.sub_and_norm.ln.bias", "layers.1.estimation_gate.FC1.weight", "layers.1.estimation_gate.FC1.bias", "layers.1.estimation_gate.FC2.weight", "layers.1.estimation_gate.FC2.bias", "layers.1.inh_layer.sub_and_norm.ln.weight", "layers.1.inh_layer.sub_and_norm.ln.bias", "layers.2.estimation_gate.FC1.weight", "layers.2.estimation_gate.FC1.bias", "layers.2.estimation_gate.FC2.weight", "layers.2.estimation_gate.FC2.bias", "layers.2.inh_layer.sub_and_norm.ln.weight", "layers.2.inh_layer.sub_and_norm.ln.bias", "layers.3.estimation_gate.FC1.weight", "layers.3.estimation_gate.FC1.bias", "layers.3.estimation_gate.FC2.weight", "layers.3.estimation_gate.FC2.bias", "layers.3.inh_layer.sub_and_norm.ln.weight", "layers.3.inh_layer.sub_and_norm.ln.bias", "layers.4.estimation_gate.FC1.weight", "layers.4.estimation_gate.FC1.bias",...
