Peter Borthwick

Hi @obackhouse , Thank you very much for you speedy reply. I should say that I don't have psi4/psi4numpy loaded on my computer so am providing 3rd-party integrals to the...

Hi, I'll have to check my spin-summation it's probably why I get incorrect results. I'm happy from your results that all is OK with psi4numpy now. I was an undergraduate(maths)/graduate(theoretical...

Hi @obackhouse, Can I just conform that if I do an IP calculation in 6-31g that for the eigenvalue 0.4541153021163 the eigenvectors are v_ip[:,2] and v_ip[:,3] and they split into...

Norms! I'm an idiot. Thank you so much. Peter.

Hi, I've tried your code and it's fine for 6-31g, but out of interest I tried sto-3g (given that on my computer it's the only basis small enough to run...

Hi @obackhouse, Yes that's the same guess as pyscf and much better. As for the solver try... de = np.linalg.norm(theta[:k] - theta_old) if de < tol: conv = True b...

No it's a nice compact solver that does the job really well. Thanks for this ADC code I was ploughing through Hodecker's thesis on ADC and trying my own implementation...

Hi Josh, I used the other approach using *real* for everything then switching to *complex* for the RT-TDSCF (still a work in progress!). The only place I had to coerse...

I've just checked and its 'eigvals' in the random phase approximation that despite having a real Hamiltonian returns (a few) complex eigenvalues. I'm going to re-write my code now to...

Yes the RPA Hamiltonian is not symmetric I know this because I had to write a QR algorithm in a BASIC Hartree-Fock program I wrote because the Jacobi doesn't do...