Currently we only support slicing for HostRingTensor or native HostTensor.
This issue is relating to add support for slicing HostBitTensor(BitArray). One can solve it by borrowing the slicing implementation from ndarray and adapt it to BitArray or convert the BitArray to an ArrayBase, call the ndarray slice mechanism and then convert data back to BitArray.
The downside of the latter method is that it increases the memory footprint.
Currently we only support slicing for
HostRingTensoror nativeHostTensor.This issue is relating to add support for slicing
HostBitTensor(BitArray). One can solve it by borrowing the slicing implementation from ndarray and adapt it toBitArrayor convert theBitArrayto anArrayBase, call the ndarray slice mechanism and then convert data back toBitArray.The downside of the latter method is that it increases the memory footprint.