Releases: pyscf/gpu4pyscf
Releases · pyscf/gpu4pyscf
v1.3.0
v1.2.1
What's Changed
- Change the license from GPL v3.0 to Apache 2.0
- Support direct SCF algorithms with multi-GPU
- Change the default conv_tol_cpscf = 1e-3 / batch of atoms to conv_tol_cpscf = 1e-6 / atom
- Add PBC HF and DFT with k-points, UHF/UKS, and density fitting
Improvements
- Fix numerical instability in complex-valued TDHF diagonalization
- Improve PCM and QMMM with
int1e_gridskernel - Support non-symmetric
int3c2eintegral - Optimize Hessian calculation with direct SCF
- Improve the numerical stability of
int3c2efor point charge - Add CI workflow for multi-GPU
Bugfixes
- Fix non-contiguous array error in p2p transfer between GPUs.
- Fix bugs in NMR calculations
Merry Christmas!
Full Changelog: v1.2.0...v1.2.1
v1.2.0
New Features
- Spin-conserved TDA and TDDFT methods
- Spin-flip TDA method.
- J-engine using McMuchie Davidson integral algorithm
- Support Multi-GPU density fitting energy, gradients and Hessian computation.
- Second order SCF solver
Improvements
- Support non-hermitian density matrix in J/K builder
- Secondary grids for CPHF solver
- 3-center integral computation efficiency for gradients and hessian
- One-electron Coulomb integrals against point charges and Gaussian charge distributions on grids.
- Automatically apply SCF initial guess from existing wavefunction
v1.1.0
New Features
- Add esp charge and resp charge by @wxj6000 in #208
- New Rys kernel by @sunqm in #221
- Optimize nuclear gradients using new Rys kernel by @sunqm in #224
- GPU kernel for analytical hessian by @sunqm in #227
- Add QM/MM by @MoleOrbitalHybridAnalyst in #218
Improvements
- Improved compatiability with pyscf 2.7.0 by @wxj6000 in #216
- Add skipping SCF cycles by @kvkarandashev in #229
- Skip building gint, gvhf, ... when building libxc by @wxj6000 in #210
Bugfix
- Typo in build_wheels.sh by @wxj6000 in #209
- Typo in dft_driver.py by @wxj6000 in #220
- Bugfix: cusolver error when specifying gpu by @wxj6000 in #213
- Bugfix: error in int2c2e by @wxj6000 in #212
- Bugfix: inconsistent gradient with CPU. Improved to_cpu, uks gradient, and grid_response by @wxj6000 in #230
- Bugfix: recompute int3c2e in DF UHF by @wxj6000 in #226
New Contributors
- @MoleOrbitalHybridAnalyst made their first contribution in #218
- @kvkarandashev made their first contribution in #229
Full Changelog: v1.0.2...v1.1.0
Contributors
sunqm, wxj6000, and 2 other contributors
Assets 2
v1.0.2
What's Changed
- Bugfix: append data in h5 file by @wxj6000 in #200
- Support customized CHELPG radii by @wxj6000 in #202
- Add cupy installation guide for developer installation instructions by @henryw7 in #204
- Bugfix: save density when spin unrestricted by @wxj6000 in #205
- Add chkfile support for pysisyphus by @henryw7 in #203
New Contributors
Full Changelog: v1.0.1...v1.0.2
Contributors
wxj6000 and henryw7
Assets 2
v1.0.1
What's Changed
- Bugfix in rks.reset by @wxj6000 in #191. The bug leads to the failure of geometry optimization with direct SCF (#190)
- Bugfix when CUDA unified memory is disabled. Removed CUDA unified memory in libxc, and reduced the overhead in calling libxc @wxj6000 in #180, #189
- Bugfix and Improvement in opt_driver by @wxj6000 in #187 #197
- Support SMD in opt_driver and dft driver @liuyu-chem1996 in #196
- Support thermo calculation in dft_driver @liuyu-chem1996 in #192
New Contributors
- @liuyu-chem1996 made their first contribution in #192
Full Changelog: v1.0...v1.0.1
Contributors
wxj6000 and liuyu-chem1996
Assets 2
v1.0
GPU4PySCF v0.7.9 release
New Features
- Gradient and Hessian for SMD model @wxj6000 in #86
- Unrestricted HF and DFT @puzhichen @wxj6000 in #87 #128 #102 #118 #117 #119 #92
- Incore CCSD @sunqm in #93
- Incore FCI fci.direct_spin1 by @sunqm in #109
- MP2 and DF MP2 @wxj6000 in #123
- IR, NMR, and polarizability @puzhichen in #147
- Cartesian atomic orbitals @wxj6000 in #100
Improvements
- Support Cupy v13 and cutensor 2.0
- Support SM 90 in PyPi binary package
- Improve Krylov subspace solver in CPHF by @wxj6000 in #156
- Rys roots up to 9 by @wxj6000 in #125
- Refactor GPU4PySCF classes @wxj6000 @sunqm in #139 #91 #132
- Nightly build by @wxj6000 in #138 #140
- Improve memory efficiency in PCM model @wxj6000 in #150
- Support specifying gpu in multi-GPU environment by @wxj6000 in #158
- Release LibXC in a separate package gpu4pyscf-libxc-cuda @wxj6000 in #110
- Improve cart2sph with cuda kernel by @wxj6000 in #137
- Align setups with pyscf by @wxj6000 in #146
- Reduce pair data reads in cuda kernels by @wxj6000 in #157
Bugfixes
- Fix a bug in shell level screening by @wxj6000 in #161
- Fix double counting dispersion correction between PySCF v2.5 and GPU4PySCF v0.6.17 #111
New Contributors
- @wenyan4work made their first contribution in #108
Full Changelog: v0.6.17...v0.7.9
Contributors
sunqm, wxj6000, and 2 other contributors
Assets 2
GPU4PySCF v0.6.17 release
New Features
- SMD solvent model
- Built-in serial DFTD3 and DFTD4, disable openmp for compatibility
- Unrestricted DFT (w/o density fitting)
- Zero-copy np array
Improvements
- Improve grids generation, vxc integration, AO evaluation of DFT.
- Reduced the initialization overhead in cupy arrays.
- Optimized 2nd derivative of hcore
- timer_debug2 for profiling cuda kernels
Bugfixes
- Fixed a bug in traspose_sum cuda kernel
- Fixed a bug in screen_index
New Contributors
- @t0saki made their first contribution in #79
- @puzhichen made their first contribution in #78
Full Changelog: v0.6.9...v0.6.17
Contributors
t0saki and puzhichen
Assets 2
GPU4PySCF v0.6.9 release
Features
- PCM SCF, analytical gradient, semi-analytical Hessian
- Analytical gradient for Direct SCF
- Workflow for PyPI binary release, support cuda 11.x and cuda 12.x
- 'to_cpu', send GPU4PySCF object to cpu PySCF
- Support various tensor contraction engine, cutensor, cuquantum, cupy, and opt_einsum.
- Chelpg charge
Improvements
- Compute Vxc with sparse AO
- Support analytical Hessian up to 168 atoms
- Improve the accuracy of Rys roots
Bugfixes
- Fixed a bug in dumping log file