permute() and movedim() in PyTorch

permute() or movedim() can permute the dimensions of a 1D or more D tensor without losing data as shown below:

*Memos:

• 
  permute() and movedim() can be used both from torch and a tensor.


• The 2nd argument of permute() is one or more dimensions with torch.


• The 1st argument of permute() is one or more dimensions with a tensor.


• The 2nd argument of movedim() is one or more dimensions. These must be unique with torch.


• The 1st argument of movedim() is …

data dimensions memos pytorch tensor torch