Roman

Thanks for the heads up on `vdt`. I just started playing with it. However, it seems to me that the functionality of `vdt` is already available in `Vc`, unless I...

NumPy's explicit mode is not conventional Einstein summation. It is provided for convenience. If you want to disable summation over an index use two indices instead of one and use...

Okay so I do find the argument of letting the user specify the shape of the output tensor quite convincing and we should/can support this relatively easily. We can implement...

The explicit `einsum` mode is implemented in Fastor. This feature however requires C++17 support for the time being (will be backported to C++14 in the future), so you need to...

Apologies for the late response. I'll put support for this feature post next big release which should be in April

Most likely a bug. I will take a look ASAP. Thanks for reporting.

This is certainly a bug. I will have a look. Thanks for reporting.

This has been requested before. If you have time, please feel free to prepare a PR. I would be happy to merge it under `experimental` directory

The noalias should imply do no checks. If noalias is implemented in the same way as it behaves for views today then the user can just do `evaluate` which makes...

Absolutely, go ahead! The only requirement we have is that you need to add unittests for it and it has to be sufficiently fast. Look at the signature of `lu`...