computing correlation function and expectation value

[amankumar]

I am computing two quantities which I expect should give me the same answer but they are not in the fermi Hubbard model.

Code: Select all

cc_ij = psi.correlation_function("Cdu", "Cu")+psi.correlation_function("Cdd", "Cd")


print(cc_ij[1][2])
print(psi.expectation_value_term([('Cdu', 1), ('Cu', 2)])+psi.expectation_value_term([('Cdd', 1), ('Cd', 2)]))

I expect that correlation function and expectation value should give the same answer. Is there any issue of projection being taken care in any one of them. Please let me know.

[Johannes]

This should really give the same. It's been a while since you wrote this - has this been resolved by now?
I quickly tried to reproduce this for a random MPS, but it seemed to work as expected. If you still find differences, it would be great if you can open an issue on github and provide a (minimal working example) or example MPS where this fails.