I've implemented a scaled-down version of Shor's for factoring 15. The modular exponentiation step is definitely the trickiest to optimize on real hardware.
3 Comments
Join the Research Discussion
Sign in to share your insights, vote on theories, and engage with the QuantumSpark community.
I've seen similar convergence issues in recent Qiskit versions. Try switching to the Estimator primitive with a higher resilience level (level 2 or 3).