-
Notifications
You must be signed in to change notification settings - Fork 0
All issues
Issue creation is restricted in this repository
Issues
is:issue state:open
is:issue state:open
Search results
MPS gate application does not keep canonical form, so bond splits truncate on the wrong singular values
bugSomething isn't workingSomething isn't workingcorrectnessGolden Rule #1Golden Rule #1highKinda messed upKinda messed upmpsRelated to MPS SimRelated to MPS SimStatus: Open.#128 In verycareful/lindblad;MPS backend is far slower than Aer once the bond cap binds
lowNot the priority right now. I hate this colour btwNot the priority right now. I hate this colour btwmpsRelated to MPS SimRelated to MPS SimperformanceGolden Rule #2Golden Rule #2Status: Open.#124 In verycareful/lindblad;Should a failed run() throw uniformly, or keep the per-backend error channel?
questionFurther information is requestedFurther information is requestedStatus: Open.#121 In verycareful/lindblad;GF(2) elimination: no measurement says which method wins at either call site, so one default rests on one data point
performanceGolden Rule #2Golden Rule #2Status: Open.#115 In verycareful/lindblad;Observers cannot be cloned per thread, so watched runs cannot follow the shot loop into parallelism
performanceGolden Rule #2Golden Rule #2Status: Open.#114 In verycareful/lindblad;Simulators parallelise inside gate kernels but never across shots, so many-shot runs of small circuits leave cores idle
performanceGolden Rule #2Golden Rule #2Status: Open.#111 In verycareful/lindblad;Randomised property testing covers only the Clifford backend
dmSlide into the Density Matrix SimSlide into the Density Matrix SimmpsRelated to MPS SimRelated to MPS SimstatevectorRelated to Statevector SimRelated to Statevector SimStatus: Open.#109 In verycareful/lindblad;Statevector::sample_counts allocates 8 * 2^n bytes on every call
performanceGolden Rule #2Golden Rule #2statevectorRelated to Statevector SimRelated to Statevector SimStatus: Open.#106 In verycareful/lindblad;Survey whether the tableau performance techniques apply to the other backends
performanceGolden Rule #2Golden Rule #2questionFurther information is requestedFurther information is requestedStatus: Open.#105 In verycareful/lindblad;Clifford sampling path: coset expansion, small-support shortcut, and counts assembly
cliffordRelated to Clifford SimRelated to Clifford SimperformanceGolden Rule #2Golden Rule #2Status: Open.#104 In verycareful/lindblad;Terminal-measurement circuits could drop the destabilizer half, but Result::final_state is public
cliffordRelated to Clifford SimRelated to Clifford SimquestionFurther information is requestedFurther information is requestedStatus: Open.#103 In verycareful/lindblad;Two Clifford hot paths cost several times what their operation counts predict
cliffordRelated to Clifford SimRelated to Clifford SimperformanceGolden Rule #2Golden Rule #2questionFurther information is requestedFurther information is requestedStatus: Open.#102 In verycareful/lindblad;