v3.0-alpha

What's Changed

• Feature: orbital generation with spherical-wave reference states by @jinzx10 in #84
• Feature&Refactor: support jy as the basis to calculate the reference wavefunction by @kirk0830 in #89
• Feature: angular momemtum analysis by @jinzx10 in #90
• Feature: support zeta_notation = auto for fit_basis = jy by @kirk0830 in #91
• Create unittest.yaml by @kirk0830 in #92
• update unittest.yaml by @kirk0830 in #94
• Refactor: seperate the coef_opt task into jy and pw by @kirk0830 in #93
• Refactor: moving all import line to the top of files by @kirk0830 in #95
• Refactor: add unittest for spillage/api.py by @kirk0830 in #96
• Feature: support the nbands autoset for monomer jy case by @kirk0830 in #97
• Fix: make example more feasiable and correct the datparse read_runnin' function by @kirk0830 in #98
• Feature: substitute "wll" with svd method inferring nzeta by @kirk0830 in #99
• Fix: correct the way to orthogonalize the wfc by @kirk0830 in #100
• Fix: leverage the threshold for nzeta_infer by @kirk0830 in #101
• Devel: program a minimal workflow for investigating the effect of nband, nzeta and initgen on orbital optimization by @kirk0830 in #102
• Create issue_label_bot.yaml by @kirk0830 in #105
• Refactor: fine tune stage 2, support the jy-space truncation loss estimation on fixed nzeta by @kirk0830 in #103

Full Changelog: v2.0+...v3.0-alpha

SIAB-v2.0+

What's Changed

• Basic utilities for orbital generation by @jinzx10 in #1
• First merge for front-end refactor by @kirk0830 in #2
• Refactor: frontend and backend interface update by @kirk0830 in #3
• INPUT and STRU file IO by @jinzx10 in #4
• Feature: connect with old pytorch.SWAT under new workflow by @kirk0830 in #5
• delete dependencies on torchvision and torchaudio by @kirk0830 in #6
• Add annotations and support nspin 2 cases in mixing aspect by @kirk0830 in #7
• add timing step-by-step in spillage optimization task by @kirk0830 in #9
• swap tab to whitspaces, add citation information, add deltaSpill output by @kirk0830 in #10
• smoothing-free spherical Bessel basis by @jinzx10 in #8
• Breaks down the build() function by @jinzx10 in #11
• Fix: enable the user setting of spillage optimization max_steps by @kirk0830 in #12
• Doc: update readme, add additional note on linking mkl with pytorch by @kirk0830 in #13
• Fix: for case pslnc big core pseudopotential, polarization is designed to … by @kirk0830 in #14
• correct read_input by @kirk0830 in #15
• fix bug in max_step by @kirk0830 in #16
• Fix: totally fix the bug for generating numerical atomic orbital whose corresponding pp has valence configuration with not consecutive l by @kirk0830 in #17
• Bijective map between composite & linearized index by @jinzx10 in #18
• Basis consistency by @jinzx10 in #19
• Add a function for reading orb_matrix data files by @jinzx10 in #20
• Enhancement: add preconditioning for blscan function by @kirk0830 in #21
• Optimization of generalized spillage by @jinzx10 in #22
• Enable selected configs in optimization by @jinzx10 in #23
• Fix argument name & restart issue by @jinzx10 in #24
• Feature: support multiprocess parallelization & update docs & update input example by @kirk0830 in #25
• Refactor: use uuid to identify more folders and temporary files by @kirk0830 in #26
• Refactor: make restart more robust by fixing cache_dir during orbital generation by @kirk0830 in #27
• Feature: support spillage_coeff keywords by @kirk0830 in #28
• Feature: support a new keyword spill_guess for making initial guess of sphbes coefs by @kirk0830 in #29
• Fix: set monomer to have the maximal lmax and support any possible INPUT param by @kirk0830 in #30
• Fix: recover the use of auto on nbands by @kirk0830 in #31
• Fix: update initial guess of Morse fitting problem by @kirk0830 in #32
• Feature: enable spill_thr param and update example by @kirk0830 in #33
• Fix: temporarily disable the param spill_thr by @kirk0830 in #34
• Fix: change pseudo_dir to absolute one by @kirk0830 in #35
• Refactor: print nthreads when fall back to run serially by @kirk0830 in #36
• Optimization of generalized spillage by @jinzx10 in #37
• Feature: enable the bfgs optimizer from front-end by @kirk0830 in #38
• Fix: add print header and fix auto-infer of nzeta by @kirk0830 in #39
• Feature: support tree topology of orbital dependence by @kirk0830 in #40
• Enhance: improve software stability and add more print by @kirk0830 in #41
• Feature: support more reference structures by @kirk0830 in #42
• Enhance: support more shapes of reference structures by @kirk0830 in #43
• Fix: fix for the case where there is no zeta functions in sublayer by @kirk0830 in #44
• Feature: support default value of dimer bond length by @kirk0830 in #45
• Fix: correct the possible-to-be-negative spillage by center molecule by @kirk0830 in #46
• Feature: add default bond lengths for trimer by @kirk0830 in #47
• Feature: support nbands_ref to be occ+/-%d and all for bfgs optimizer by @kirk0830 in #48
• Fix: improve stability of program on more complicated example by @kirk0830 in #49
• Fix: fix the case with old SIAB_INPUT by @kirk0830 in #50
• Feature: add output of ORBITAL_RESULTS.txt for bfgs optimizer by @kirk0830 in #51
• Fix: fix the bug when there is only one l by @kirk0830 in #52
• Correct plot.py by @kirk0830 in #53
• Update plot.py by @kirk0830 in #54
• Support spill_coefs in BFGS optimization by @jinzx10 in #55
• Create label.yml by @kirk0830 in #56
• Fix: fix the bug for element F case by @kirk0830 in #57
• Feature: enhance the nbands autoset for monomer by @kirk0830 in #58
• Delete .github/workflows/label.yml by @kirk0830 in #59
• Enhance: update the nbands default behavior of initial guess by @kirk0830 in #60
• Feature: support jY basis direct output by @kirk0830 in #61
• Fix: update the way to set nbands default by @kirk0830 in #64
• Fix: fix the bug with use endswith on list instance by @kirk0830 in #65
• Fix: fix the nbands autoset for Barium case by @kirk0830 in #66
• Fix: correct the upper bound of loop of lmax, no larger than 3 (f-orb) by @kirk0830 in #67
• Update README.md by @kirk0830 in #68
• Feature: support vwr pseudopotential by @kirk0830 in #69
• Refactor: move the nao2gto to io folder by @kirk0830 in #70
• Fix: make possible for GTH pseudopotential corresponding orbital generation by @kirk0830 in #72
• Fix: align the range of pseudopotential format supported with ABACUS by @kirk0830 in #73
• Feature: support mixed reference system by @kirk0830 in #74
• Refactor: remove nao2gto from orbital generation code by @kirk0830 in #75
• Fix: fix the severe bug after supporting mixed reference systems by @kirk0830 in #77
• Feature: support entry run by @kirk0830 in #78
• Update README.md by @kirk0830 in #79
• Update pyproject.toml by @kirk0830 in #80
• Add version requirements on scipy, python by @kirk0830 in #81
• Refactor: add explicit dependency on matplotlib by @kirk0830 in #83
• Update README.md by @kirk0830 in #82
• Refactor: delete out-dated files by @kirk0830 in #86
• Enhance: enhance the exception handling on pp and improve the way to parse pp by @kirk0830 in #85

New Contributors

• @jinzx10 made their first contribution in #1
• @kirk0830 made their first contribution in #2

Full Changelog: https://github.com/kirk0830/ABACUS-ORBGEN/commits/v2.0+