Question about trainning of IND

[chuinixiongkou]

Hello author, thank you for the great work you are doing.I am training the ind dataset, when epoch==step1, the dimension of the tensor is wrong, I have no problem with the previous training generations when I debug, what is wrong?

File "train.py", line 144, in
loss, _, _, _, _, _, count = net(data, temp=temp, type=type, device=device)
File "/home//miniconda3/envs/py38/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(input, kwargs)
File "/home//exp/TDOR-main/model/full_model.py", line 71, in forward
waypts, waypt_lengths = self.net_p.sample_policy(pi, waypts_e,num_samples, temp=temp)
File "/home//exp/TDOR-main/model/policy_network.py", line 44, in sample_policy
state=waypts_e[:,:1].repeat(1,num_samples,1)
RuntimeError: Number of dimensions of repeat dims can not be smaller than number of dimensions of tensor

sorry for my question! I have fixed the problem.