New eigen solve interface #71
Conversation
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #71 +/- ##
==========================================
+ Coverage 90.17% 91.87% +1.70%
==========================================
Files 18 18
Lines 1699 1711 +12
==========================================
+ Hits 1532 1572 +40
+ Misses 167 139 -28 ☔ View full report in Codecov by Sentry. |
| return ([QuantumObject(res.vectors[:, k], OperatorKet, res.dims) for k in 1:length(res.values)], Val(:vectors)) | ||
| end | ||
| end | ||
| Base.iterate(res::EigsolveResult, ::Val{:vectors}) = (res.vectors, Val(:done)) |
There was a problem hiding this comment.
Should we return already a QuantumObject? In this way
E, V, U = eigenstates(H)U is already a QuantumObject.
There was a problem hiding this comment.
I was thinking of returning U as a Qobj in the beginning
but I noticed that for sparse solvers: U is usually a non-squared matrix.
if we do it in this way, it will make the return object type unstable (depends on sparse=true or false).
There was a problem hiding this comment.
Yes, we can leave it like this
|
|
| return ([QuantumObject(res.vectors[:, k], OperatorKet, res.dims) for k in 1:length(res.values)], Val(:vectors)) | ||
| end | ||
| end | ||
| Base.iterate(res::EigsolveResult, ::Val{:vectors}) = (res.vectors, Val(:done)) |
There was a problem hiding this comment.
Yes, we can leave it like this
close #39
Summary of this PR:
struct EigsolveResultiterateforEigsolveResultshowforEigsolveResulteigenenergies(A::QuantumObject; sparse::Bool=false, kwargs...)eigenstates(A::QuantumObject; sparse::Bool=false, kwargs...)Also, the fields in
EigsolveResultlikevalsandvecshave been renamed tovaluesandvectors, repectively. Make the variable name same asLinearAlgebra.Eigen.