Fermi v0.5.0

Diff since v0.4.0

Breaking

TBLIS dropped as dependency, some fetures that relied on it may no longer work.

Updated [compat]

• Dependencies updated so Fermi will no longer force you to use old packages.

Tests

Tests were taking way too long to run, I decided to simplify them - they are faster now!

New methods

• FCI implemented! Two algorithms are available: 1) using sparse arrays and 2) determinant-based (alpha/beta strings).
• A little code for doing a quick Mulliken charges study was implemented.
• Direct RHF using density fitting is implemented!

Merged pull requests: - Tblis purge (#123) (@gustavojra) - Separating Vnuc from Fermi.Molecule for coming gradient work (#127) (@hfmull) - Gradient macro for general finite differences and analytic RHF gradients (#129) (@hfmull) - FCI + Fix to DirectRHF (#131) (@gustavojra) - New tag - 0.5.0 (#134) (@gustavojra) - new FCI method, eCC (#143) (@gustavojra) - Updading compact. Small clean up. Improved Molecules. (#145) (@gustavojra) - Update Project.toml (#146) (@gustavojra) - Compat added for KrylovKit and ArnoldiMethod (#147) (@gustavojra)

Closed issues: - Docs are way out of date (#29) - Improving Coupled Cluster Efficiency (#34) - Improve Error messages (#96) - Separate get_scf_alg into get_uhf_alg and get_rhf_alg (#102) - One- and two-body integrals from Hartree-Fock (#116)