-
Notifications
You must be signed in to change notification settings - Fork 5
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Solver continuation #87
Conversation
It's just the best way in the end.
There was an inconsistency in calculations which caused some things to explode in results value. 100-0 should only be the visual representation.
09bf43a
to
117a84e
Compare
Previously results were built incrimentally and wouldn't know how big the resultant bitstring may be, so you'd only get the 'unknown' values for ones further down the chain. 1, 1X, XX1, etc. Add in a quick mechanism for making sure measured qubits are known up-front to always build full strings.
Drastically improves runtime of evaluator.
When applying multi qubit operations sometimes each individual one would make the qubit drop out of entanglement, which would then trigger removal during iteration. This is a quick fix to that, but it needs some careful considerations about when qubits remove themselves from things. It may be easier to simply revert to the high level to do it for clusters.
117a84e
to
7778426
Compare
Initial tests show scalability looks good, 500qb with 20k gates took 10 minutes and there's a huge amount of improvements yet to make in regards to that. Going to verify/fix-up results synthesis as something's a little off there, then run some ambitious hybrid algorithms through it and make sure results align with much more precise simulations. |
The changes put in so far are enough breakage to warrant a version increase. Also increase QAOA available qubits for Qiskit simulation.
Going to merge this and start development of v0.2.0 directly on develop for a variety of reasons, mostly because develop is lacking a bunch of nice features that are locked to this branch, and even if I was to add them in isolation the changes would warrant a v0.2.0 release anyway. |
With the foundations in, working on more advanced entanglement complexities and fixing edge cases.