Skip to content

Refactor&Feature: Package some parameters in DeePKS for simplification & update HR precalc output. #6706

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 13 commits into from
Nov 17, 2025
Merged

Refactor&Feature: Package some parameters in DeePKS for simplification & update HR precalc output. #6706

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
13 commits
Select commit Hold shift + click to select a range
78d732d
Package some parameters in DeePKS.
ErjieWu Nov 12, 2025
6b889c4
Simplify fpre.
ErjieWu Nov 12, 2025
27bd74d
Simplify orbpre.
ErjieWu Nov 12, 2025
327707b
Simplify pdm.
ErjieWu Nov 12, 2025
77c253d
Simplify spre.
ErjieWu Nov 13, 2025
f3a5dab
Simplify vdpre and vdrpre.
ErjieWu Nov 13, 2025
8b777a2
Simplify io.
ErjieWu Nov 13, 2025
83df3b7
Simplify force.
ErjieWu Nov 13, 2025
dbb8ebd
Merge branch 'develop' into refactor
ErjieWu Nov 13, 2025
1158136
clang-format change.
ErjieWu Nov 13, 2025
ced3357
clang-format change.
ErjieWu Nov 13, 2025
922b776
Change the output of HR precalc.
ErjieWu Nov 17, 2025
a35226f
Update and add new test for DeePKS HR.
ErjieWu Nov 17, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion source/source_io/output_mat_sparse.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,7 +27,7 @@ void output_mat_sparse(const bool& out_mat_hsR,
//! generate a file containing the Hamiltonian and S(overlap) matrices
if (out_mat_hsR)
{
output_HSR(ucell, istep, v_eff, pv, HS_Arrays, grid, kv, *p_dftu, p_ham);
output_HSR(ucell, istep, pv, HS_Arrays, grid, kv, *p_dftu, p_ham);
}

//! generate a file containing the kinetic energy matrix
Expand Down
3 changes: 0 additions & 3 deletions source/source_io/write_HS_R.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,7 +15,6 @@
template <typename TK>
void ModuleIO::output_HSR(const UnitCell& ucell,
const int& istep,
const ModuleBase::matrix& v_eff,
const Parallel_Orbitals& pv,
LCAO_HS_Arrays& HS_Arrays,
const Grid_Driver& grid, // mohan add 2024-04-06
Expand Down Expand Up @@ -304,7 +303,6 @@ void ModuleIO::output_TR(const int istep,
template void ModuleIO::output_HSR<double>(
const UnitCell& ucell,
const int& istep,
const ModuleBase::matrix& v_eff,
const Parallel_Orbitals& pv,
LCAO_HS_Arrays& HS_Arrays,
const Grid_Driver& grid,
Expand All @@ -324,7 +322,6 @@ template void ModuleIO::output_HSR<double>(
template void ModuleIO::output_HSR<std::complex<double>>(
const UnitCell& ucell,
const int& istep,
const ModuleBase::matrix& v_eff,
const Parallel_Orbitals& pv,
LCAO_HS_Arrays& HS_Arrays,
const Grid_Driver& grid,
Expand Down
1 change: 0 additions & 1 deletion source/source_io/write_HS_R.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,7 +15,6 @@ using TAC = std::pair<int, std::array<int, 3>>;
template <typename TK>
void output_HSR(const UnitCell& ucell,
const int& istep,
const ModuleBase::matrix& v_eff,
const Parallel_Orbitals& pv,
LCAO_HS_Arrays& HS_Arrays,
const Grid_Driver& grid, // mohan add 2024-04-06
Expand Down
2 changes: 1 addition & 1 deletion source/source_lcao/FORCE_gamma.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -231,10 +231,10 @@ void Force_LCAO<double>::ftable(const bool isforce,
gd,
*this->ParaV,
nks,
deepks.ld.deepks_param,
kv->kvec_d,
deepks.ld.phialpha,
deepks.ld.gedm,
deepks.ld.inl_index,
fvnl_dalpha,
isstress,
svnl_dalpha);
Expand Down
2 changes: 1 addition & 1 deletion source/source_lcao/FORCE_k.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -257,10 +257,10 @@ void Force_LCAO<std::complex<double>>::ftable(const bool isforce,
gd,
pv,
kv->get_nks(),
deepks.ld.deepks_param,
kv->kvec_d,
deepks.ld.phialpha,
deepks.ld.gedm,
deepks.ld.inl_index,
fvnl_dalpha,
isstress,
svnl_dalpha);
Expand Down
88 changes: 39 additions & 49 deletions source/source_lcao/LCAO_hamilt.hpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,8 +6,8 @@
#include "source_base/abfs-vector3_order.h"
#include "source_base/global_variable.h"
#include "source_base/timer.h"
#include "source_lcao/spar_exx.h"
#include "source_lcao/module_ri/RI_2D_Comm.h"
#include "source_lcao/spar_exx.h"

#include <RI/global/Global_Func-2.h>
#include <RI/ri/Cell_Nearest.h>
Expand All @@ -21,28 +21,27 @@
// Peize Lin add 2022.09.13

template <typename Tdata>
void sparse_format::cal_HR_exx(const UnitCell& ucell,
void sparse_format::cal_HR_exx(
const UnitCell& ucell,
const Parallel_Orbitals& pv,
LCAO_HS_Arrays& HS_Arrays,
const int& current_spin,
const double& sparse_threshold,
const int (&nmp)[3],
const std::vector<std::map<int,
std::map<std::pair<int, std::array<int, 3>>,
RI::Tensor<Tdata>>>>& Hexxs) {
const std::vector<std::map<int, std::map<std::pair<int, std::array<int, 3>>, RI::Tensor<Tdata>>>>& Hexxs)
{
ModuleBase::TITLE("sparse_format", "cal_HR_exx");
ModuleBase::timer::tick("sparse_format", "cal_HR_exx");

const Tdata frac = GlobalC::exx_info.info_global.hybrid_alpha;

std::map<int, std::array<double, 3>> atoms_pos;
for (int iat = 0; iat < ucell.nat; ++iat) {
atoms_pos[iat] = RI_Util::Vector3_to_array3(
ucell.atoms[ucell.iat2it[iat]]
.tau[ucell.iat2ia[iat]]);
for (int iat = 0; iat < ucell.nat; ++iat)
{
atoms_pos[iat] = RI_Util::Vector3_to_array3(ucell.atoms[ucell.iat2it[iat]].tau[ucell.iat2ia[iat]]);
}
const std::array<std::array<double, 3>, 3> latvec
= {RI_Util::Vector3_to_array3(ucell.a1), // too bad to use GlobalC here,
= {RI_Util::Vector3_to_array3(ucell.a1), // too bad to use GlobalC here,
RI_Util::Vector3_to_array3(ucell.a2),
RI_Util::Vector3_to_array3(ucell.a3)};

Expand All @@ -51,92 +50,83 @@ void sparse_format::cal_HR_exx(const UnitCell& ucell,
RI::Cell_Nearest<int, int, 3, double, 3> cell_nearest;
cell_nearest.init(atoms_pos, latvec, Rs_period);

const std::vector<int> is_list = (PARAM.inp.nspin != 4)
? std::vector<int>{current_spin}
: std::vector<int>{0, 1, 2, 3};
const std::vector<int> is_list
= (PARAM.inp.nspin != 4) ? std::vector<int>{current_spin} : std::vector<int>{0, 1, 2, 3};

for (const int is: is_list)
for (const int is: is_list)
{
int is0_b = 0;
int is1_b = 0;
std::tie(is0_b, is1_b) = RI_2D_Comm::split_is_block(is);

if (Hexxs.empty())
if (Hexxs.empty())
{
break;
}

for (const auto& HexxA: Hexxs[is])
for (const auto& HexxA: Hexxs[is])
{
const int iat0 = HexxA.first;
for (const auto& HexxB: HexxA.second)
for (const auto& HexxB: HexxA.second)
{
const int iat1 = HexxB.first.first;

const Abfs::Vector3_Order<int> R = RI_Util::array3_to_Vector3(
cell_nearest.get_cell_nearest_discrete(iat0,
iat1,
HexxB.first.second));
cell_nearest.get_cell_nearest_discrete(iat0, iat1, HexxB.first.second));

HS_Arrays.all_R_coor.insert(R);

const RI::Tensor<Tdata>& Hexx = HexxB.second;

for (size_t iw0 = 0; iw0 < Hexx.shape[0]; ++iw0)
for (size_t iw0 = 0; iw0 < Hexx.shape[0]; ++iw0)
{
const int iwt0 = RI_2D_Comm::get_iwt(ucell,iat0, iw0, is0_b);
const int iwt0 = RI_2D_Comm::get_iwt(ucell, iat0, iw0, is0_b);
const int iwt0_local = pv.global2local_row(iwt0);

if (iwt0_local < 0)
if (iwt0_local < 0)
{
continue;
}

for (size_t iw1 = 0; iw1 < Hexx.shape[1]; ++iw1)
for (size_t iw1 = 0; iw1 < Hexx.shape[1]; ++iw1)
{
const int iwt1 = RI_2D_Comm::get_iwt(ucell,iat1, iw1, is1_b);
const int iwt1 = RI_2D_Comm::get_iwt(ucell, iat1, iw1, is1_b);
const int iwt1_local = pv.global2local_col(iwt1);

if (iwt1_local < 0)
if (iwt1_local < 0)
{
continue;
}

if (std::abs(Hexx(iw0, iw1)) > sparse_threshold)
if (std::abs(Hexx(iw0, iw1)) > sparse_threshold)
{
if (PARAM.inp.nspin == 1 || PARAM.inp.nspin == 2)
if (PARAM.inp.nspin == 1 || PARAM.inp.nspin == 2)
{
auto& HR_sparse_ptr
= HS_Arrays
.HR_sparse[current_spin][R][iwt0];
auto& HR_sparse_ptr = HS_Arrays.HR_sparse[current_spin][R][iwt0];
double& HR_sparse = HR_sparse_ptr[iwt1];
HR_sparse += RI::Global_Func::convert<double>(
frac * Hexx(iw0, iw1));
if (std::abs(HR_sparse) <= sparse_threshold)
HR_sparse += RI::Global_Func::convert<double>(frac * Hexx(iw0, iw1));
if (std::abs(HR_sparse) <= sparse_threshold)
{
HR_sparse_ptr.erase(iwt1);
}
}
else if (PARAM.inp.nspin == 4)
}
else if (PARAM.inp.nspin == 4)
{
auto& HR_sparse_ptr
= HS_Arrays.HR_soc_sparse[R][iwt0];

std::complex<double>& HR_sparse
= HR_sparse_ptr[iwt1];

HR_sparse += RI::Global_Func::convert<
std::complex<double>>(frac * Hexx(iw0, iw1));

if (std::abs(HR_sparse) <= sparse_threshold)
auto& HR_sparse_ptr = HS_Arrays.HR_soc_sparse[R][iwt0];

std::complex<double>& HR_sparse = HR_sparse_ptr[iwt1];

HR_sparse += RI::Global_Func::convert<std::complex<double>>(frac * Hexx(iw0, iw1));

if (std::abs(HR_sparse) <= sparse_threshold)
{
HR_sparse_ptr.erase(iwt1);
}
}
else
}
else
{
throw std::invalid_argument(std::string(__FILE__) + " line "
+ std::to_string(__LINE__));
+ std::to_string(__LINE__));
}
}
}
Expand Down
Loading
Loading