To do this I need to make TensorViews wrapped by an RC, and make some trait to generalise over the two.
This will be pretty important for calculating jacobians -- since I will want to flatten everything before doing the chain rule, but need to make sure that the same underlying tensor has the grad set.
To do this I need to make TensorViews wrapped by an RC, and make some trait to generalise over the two.
This will be pretty important for calculating jacobians -- since I will want to flatten everything before doing the chain rule, but need to make sure that the same underlying tensor has the grad set.