Parameter setting of pre-trained model

[LuanDi1996]

No description provided.

[LuanDi1996]

Hi, thanks for sharing the great work.

I have a problem when I run the following command to predict with the pre-trained marginal prediction mode densetnt.raster.vehicle.1.zip here https://github.com/Tsinghua-MARS-Lab/M2I#marginal-trajectory-prediction. It seems like the parameter setting between the pre-trained model and script are different.

_RuntimeError: Error(s) in loading state_dict for VectorNet:
	size mismatch for decoder.goals_2D_decoder.mlp.linear.weight: copying a param with shape torch.Size([128, 384]) from checkpoint, the shape in current model is torch.Size([128, 512]).
	size mismatch for decoder.goals_2D_decoder.fc.weight: copying a param with shape torch.Size([1, 512]) from checkpoint, the shape in current model is torch.Size([1, 640])._