The question about modifying the hyper-parameters horizon

[wen1tao1]

In the source code, the horizon is set to 32, but my path planning task requires a larger horizon. When I increase the horizon, there will be an error when training the value function. How can I solve this problem? Do I need to modify the network structure?
The error is as follows:
Testing forward... Traceback (most recent call last):
File "/media/workspace/zhaowentao/diffuser-main/scripts/train_values_val.py", line 134, in
loss, _ = diffusion.loss(*batch)
File "/media/workspace/zhaowentao/diffuser-main/diffuser/models/diffusion.py", line 229, in loss
return self.p_losses(x, *args, t)
File "/media/workspace/zhaowentao/diffuser-main/diffuser/models/diffusion.py", line 243, in p_losses
pred = self.model(x_noisy, cond, t)
File "/media/workspace/zhaowentao/Anaconda3/envs/df_new/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1553, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/media/workspace/zhaowentao/Anaconda3/envs/df_new/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1562, in _call_impl
return forward_call(*args, **kwargs)
File "/media/workspace/zhaowentao/diffuser-main/diffuser/models/temporal.py", line 232, in forward
out = self.final_block(torch.cat([x, t], dim=-1))
File "/media/workspace/zhaowentao/Anaconda3/envs/df_new/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1553, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/media/workspace/zhaowentao/Anaconda3/envs/df_new/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1562, in _call_impl
return forward_call(*args, **kwargs)
File "/media/workspace/zhaowentao/Anaconda3/envs/df_new/lib/python3.8/site-packages/torch/nn/modules/container.py", line 219, in forward
input = module(input)
File "/media/workspace/zhaowentao/Anaconda3/envs/df_new/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1553, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/media/workspace/zhaowentao/Anaconda3/envs/df_new/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1562, in _call_impl
return forward_call(*args, **kwargs)
File "/media/workspace/zhaowentao/Anaconda3/envs/df_new/lib/python3.8/site-packages/torch/nn/modules/linear.py", line 117, in forward
return F.linear(input, self.weight, self.bias)
RuntimeError: mat1 and mat2 shapes cannot be multiplied (1x160 and 224x96)