Olivier Gauthé

I find it surprising that `size(slice(view))` output type differs from any other combination of `view` and `getindex`, including e.g. `size(view(m1, BlockArrays.Block(1, 1))[1:1,..])`. However I agree that `LabelledInteger` can act as...

issue: `TensorAlgebra.contract` does not support views of `BlockSparseArrays` main at `141399083198d3022a5daaf6e7da3e37ad894c4d` ```julia using BlockArrays: BlockArrays using NDTensors.BlockSparseArrays: BlockSparseArrays using NDTensors.GradedAxes: GradedAxes using NDTensors.TensorAlgebra: TensorAlgebra ba = BlockArrays.BlockArray{Float64}(undef, [2],[2]) g =...

One gets a different error after reshaping the view: ```julia x = reshape(v, (16,1)) # Ok f(reshape(v, (16,1))) # ERROR: Not implemented ``` ```julia ERROR: Not implemented Stacktrace: [1] error(s::String)...

As I understand, you redefined `GradedUnitRange` as a `BlockedUnitRange` with an extra vector of `LabelledInteger` to label each sector. I understand this design is simpler, I was previously wondering why...

It seems to me that no solution is optimal and a tradeoff needs to be made. To specify my position, I currently lean towards option 2), explicitly splitting the two...

The current version implements fusion rules for simple categories, ordered category product and dictionary category product. The dictionary category product is still a a work in progress as it is...

All stability issues are now solved. I still need to check the behavior of `dual`.

I made the requested changes: all explicit type comparison are replaced by dispatch. I propose conventions for `tensor_product`, `fusion_product` and `dual`. I leave the unification of `SU2` and `SU{2}` for...

I had a look at the failures in CI. They all come from julia 1.6, all tests pass with julia 1.10. Failures boil down to two cases: first a return...

Thank you for the clues, I ended up using another way. I also removed a few `@inferred` from the tests and now they all pass with julia 1.6. I understand...