andrewmusselman
commented
2 months ago - research if/how these are implemented in each backend
- research the different methodologies to assess scope
- enumerate strategies e.g. Basis, Amplitude, etc
- document constraints e.g. whole-dataset vs individual
- compatibility with various kernel methodologies
- research what data structures are needed to keep computation backend-agnostic
- decide what needs to be built in qumat
- integrate backend implementations
- build utilities from scratch
rawkintrevo