Open jakirkham opened 6 years ago
This appears to be better suited at xnd
. Migrating the issue over there. Sorry for the noise.
Apart from integration with standard C++ containers, I think a lot of work has been put into making xtensor fast. Especially, gumath kernels could benefit from SIMD support and lazy evaluation.
From @jakirkham on August 1, 2018 17:23
The xtensor library and surrounding ecosystem provides a quite powerful way of working with N-D arrays in C++ and plays nicely with standard C++ containers. Of course C remains the lingua franca in the programming world both for library interfaces and for working on a wide range of systems. It would be interesting to explore in what ways xnd can interoperate with xtensor.
xref: https://github.com/QuantStack/xtensor/issues/1030
Copied from original issue: plures/ndtypes#48