v0.6.10

DFTK v0.6.10

Diff since v0.6.9

Closed issues:

• Switch to ReTest (#508)
• Unexpected error when using Crystallographic Information Framework (CIF) File Input (#863)

Merged pull requests:

• API change for Ewald and pairwise computations (#853) (@epolack)
• Extend details on precompilation and developer setup (#862) (@mfherbst)
• Parse element by number (#864) (@mfherbst)
• Fix GPU CI (again) (#865) (@mfherbst)
• Fix GPU tests (#867) (@vchuravy)
• Fix for Ewald (#869) (@epolack)
• CompatHelper: bump compat for Spglib to 0.7, (keep existing compat) (#870) (@github-actions[bot])
• Collection of some documentation extensions and bugfixes (#873) (@mfherbst)

v0.6.9

DFTK v0.6.9

Diff since v0.6.8

Merged pull requests:

• Fix various smaller bugs (#860) (@mfherbst)

v0.6.8

DFTK v0.6.8

Diff since v0.6.7

Closed issues:

• Initial guess density currently broken on GPU (#841)
• unit and equation of calculated force (#847)
• Minus sign in Fourier transform of projectors (#855)

Merged pull requests:

• Fix initial density guess generation for GPU runs (#850) (@zhubonan)
• Fix sign in Fourier transform of projectors (#856) (@antoine-levitt)
• Switch to PrecompileTools (#857) (@mfherbst)
• Fix GPU CI by enforcing to run on CUDA 11 (#859) (@mfherbst)

v0.6.7

DFTK v0.6.7

Diff since v0.6.6

Closed issues:

• Remove remaining ASE-dependent code (#756)
• Mixed Reduced and Cartesian coords in apply_symop ? (#837)
• Precompilation causes warning (#843)
• Rounding issues cause bad wannier interface (#844)
• how to customize/define an xc functional (#848)

Merged pull requests:

• Less ambiguous SymOp doc (#840) (@epolack)
• Fix LAPACK non-orthogonality (#842) (@antoine-levitt)
• Fix rounding in symmetry (#845) (@antoine-levitt)
• Precompilation warning (#846) (@antoine-levitt)

v0.6.6

DFTK v0.6.6

Diff since v0.6.5

Merged pull requests:

• Need to depend on AtomsIOPython in tests (#838) (@mfherbst)
• More graceful numerical checks (#839) (@mfherbst)

v0.6.5

DFTK v0.6.5

Diff since v0.6.4

Closed issues:

• Support ARM / aarch64 / "Apple silicon" M-series processors (#820)
• Julia Nightly tests failing (#826)
• Usage Support: Adding DFTK.jl to Matsci Discourse Community (#833)

Merged pull requests:

• Non-linear core correction and pseudopotential density guess (#766) (@azadoks)
• Fixing some documentation layouts & duplicate (#815) (@epolack)
• Model update constructor (#816) (@mfherbst)
• MPI Allreduce fallbacks for aarch64 support (#821) (@azadoks)
• Fix fast test execution (#827) (@epolack)
• Fix Hamiltonian application for nonstandard models (#829) (@antoine-levitt)
• Bump actions/checkout from 2 to 3 (#830) (@dependabot[bot])
• Bump codecov/codecov-action from 2 to 3 (#831) (@dependabot[bot])
• Fixing documentation links (#832) (@epolack)
• Update introductory resources and publications (#834) (@mfherbst)

v0.6.4

DFTK v0.6.4

Diff since v0.6.3

Merged pull requests:

• Integrate with AtomsBase 0.3 (#814) (@mfherbst)

v0.6.3

DFTK v0.6.3

Diff since v0.6.1

Closed issues:

• Latency (#402)
• LDOS mixing for non-monotonic smearings (#787)

Merged pull requests:

• Implement more flexible Fermi level algorithm interface (#792) (@mfherbst)
• Add simplistic time per SCF step printing (#806) (@mfherbst)
• Add precompilation (#809) (@antoine-levitt)
• Vastly improve GPU support (#813) (@mfherbst)

v0.6.1

DFTK v0.6.1

Diff since v0.6.0

Closed issues:

• Crash with PyPlot (#808)

Merged pull requests:

• Remove load_xyz functions (#805) (@mfherbst)

v0.6.0

DFTK v0.6.0

Diff since v0.5.15

Note that this release increases the minor version and contains a number of breaking changes:

• The self_consistent_field and potential_mixing SCF functions now check convergence in the density by default. This means that self_consistent_field(basis; tol=1e-6) now converges until the change in density between subsequent steps is below 1e-6. Before it used to converge until the change in energy was less than tol. To recover the old behaviour use self_consistent_field(basis; is_converged=DFTK.ScfConvergenceEnergy(tol)). (#800)

• The load_lattice, load_atoms and load_positions functions are now deprecated and will be removed in DFTK 0.6.1. Users are advised to use AtomsIO for the purpose of loading structures from external files. See https://docs.dftk.org/v0.6/examples/atomsbase/ and https://github.com/mfherbst/AtomsIO.jl for examples.

• Instead of using NLSolve's Anderson acceleration we have switched to a built-in implementation, which we found more stable in practice. (#791)
  The NLSolve-been approach has been removed completely.

• The deprecated G_to_r and r_to_G functions for FFTs have now been removed in favour of ifft, fft and similar. (#804)

• The deprecated n_bands and n_ep_extra arguments from self_consistent_field have now been removed. (#804)

Closed issues:

• Struct for atoms (#118)
• GPU discussion (#350)
• SCF default tolerance should be on the density (#780)
• Run GPU CI only after activation (#803)

Merged pull requests:

• Bugfix for computations of δψ and supercells (#775) (@epolack)
• Switch default scf solver to builtin Anderson (#791) (@epolack)
• Add developper documentation for GPU computations (#793) (@GVigne)
• Correct PW parameterization comment (#795) (@epolack)
• Add GPU CI via gitlab (#796) (@mfherbst)
• Switch to density tolerance by default (#800) (@mfherbst)
• Make cigpu badge a link to ci pipelines in gitlab (#801) (@carstenbauer)
• Rely on AtomsIO for structure parsing (#802) (@mfherbst)
• Remove deprecated features (#804) (@mfherbst)