InstanceNorm1d returns 0-filled tensor to 2D tensor.This is because InstanceNorm1d reshapes inputs to(1, N * C, ...) from (N, C,...) and this makesvariances 0

[moTcream]

ValueError Traceback (most recent call last)
Cell In[7], line 1
----> 1 train(save="test")

Cell In[6], line 52
49 target = target[:, ind, :]
51 # Compute displacement prediction
---> 52 disp_pred = predictor(source, target, net)
54 # Compute loss
55 loss = netloss(disp, disp_pred)

Cell In[3], line 152
150 def sLBP_GF(kpts_fixed, kpts_moving, net, f=1):
151 # geometric features
--> 152 kpts_fixed_feat, kpts_moving_feat = net(kpts_fixed, kpts_moving, k)
153 kpts_fixed_disp_pred = inference(kpts_fixed,kpts_moving,kpts_fixed_feat,kpts_moving_feat, f)
154 return kpts_fixed_disp_pred

File ~/anaconda3/envs/deftrans/lib/python3.8/site-packages/torch/nn/modules/module.py:1102, in Module._call_impl(self, *input, **kwargs)
1098 # If we don't have any hooks, we want to skip the rest of the logic in
1099 # this function, and just call forward.
1100 if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks
1101 or _global_forward_hooks or _global_forward_pre_hooks):
-> 1102 return forward_call(*input, **kwargs)
...
138 if input.dim() != 3:
139 raise ValueError('expected 3D input (got {}D input)'
140 .format(input.dim()))

ValueError: InstanceNorm1d returns 0-filled tensor to 2D tensor.This is because InstanceNorm1d reshapes inputs to(1, N * C, ...) from (N, C,...) and this makesvariances 0
InstanceNorm1d returns 0-filled tensor to 2D tensor.This is because InstanceNorm1d reshapes inputs to(1, N * C, ...) from (N, C,...) and this makesvariances 0