10
0
mirror of https://github.com/QuantumPackage/qp2.git synced 2024-11-15 18:43:51 +01:00

NO aba // ok

This commit is contained in:
Abdallah Ammar 2023-06-10 11:24:06 +02:00
parent 6e31ca280d
commit d9921922fc
2 changed files with 101 additions and 61 deletions

View File

@ -1104,12 +1104,20 @@ BEGIN_PROVIDER [ double precision, no_aba_contraction, (mo_num,mo_num,mo_num,mo_
!$OMP int2_grad1_u12_bimo_t, final_weight_at_r_vector, &
!$OMP no_aba_contraction)
allocate(tmp_3d(mo_num,mo_num,mo_num), tmp_2d(mo_num,mo_num))
allocate(tmp1(n_points_final_grid,3,mo_num), tmp2(n_points_final_grid,mo_num))
allocate(tmpval_1(n_points_final_grid), tmpval_2(n_points_final_grid))
allocate(tmpvec_1(n_points_final_grid,3), tmpvec_2(n_points_final_grid,3))
tmp_3d = 0.d0
tmp_2d = 0.d0
tmp1 = 0.d0
tmp2 = 0.d0
tmpval_1 = 0.d0
tmpval_2 = 0.d0
tmpvec_1 = 0.d0
tmpvec_2 = 0.d0
!$OMP DO
do ii = 1, Ne(2)
@ -1147,7 +1155,9 @@ BEGIN_PROVIDER [ double precision, no_aba_contraction, (mo_num,mo_num,mo_num,mo_
do p1 = 1, mo_num
do h2 = 1, mo_num
do p2 = 1, mo_num
!$OMP CRITICAL
no_aba_contraction(p2,h2,p1,h1) = no_aba_contraction(p2,h2,p1,h1) + tmp_3d(p2,h2,p1)
!$OMP END CRITICAL
enddo
enddo
enddo
@ -1177,7 +1187,9 @@ BEGIN_PROVIDER [ double precision, no_aba_contraction, (mo_num,mo_num,mo_num,mo_
do h2 = 1, mo_num
do p2 = 1, mo_num
!$OMP CRITICAL
no_aba_contraction(p2,h2,p1,h1) = no_aba_contraction(p2,h2,p1,h1) + tmp_2d(p2,h2)
!$OMP END CRITICAL
enddo
enddo
@ -1195,28 +1207,40 @@ BEGIN_PROVIDER [ double precision, no_aba_contraction, (mo_num,mo_num,mo_num,mo_
!$OMP END PARALLEL
allocate(tmp_3d(mo_num,mo_num,mo_num), tmp_2d(mo_num,mo_num))
allocate(tmp1(n_points_final_grid,3,mo_num), tmp2(n_points_final_grid,mo_num))
allocate(tmpval_1(n_points_final_grid), tmpval_2(n_points_final_grid))
allocate(tmpvec_1(n_points_final_grid,3), tmpvec_2(n_points_final_grid,3))
! purely open-shell part
if(Ne(2) < Ne(1)) then
!$OMP PARALLEL &
!$OMP DEFAULT (NONE) &
!$OMP PRIVATE (ipoint, h1, p1, h2, p2, i, ii, &
!$OMP tmp_3d, tmp_2d, tmp1, tmp2, &
!$OMP tmpval_1, tmpval_2, tmpvec_1, tmpvec_2) &
!$OMP SHARED (n_points_final_grid, Ne, occ, mo_num, &
!$OMP mos_l_in_r_array_transp, mos_r_in_r_array_transp, &
!$OMP int2_grad1_u12_bimo_t, final_weight_at_r_vector, &
!$OMP no_aba_contraction)
Allocate(tmp_3d(mo_num,mo_num,mo_num), tmp_2d(mo_num,mo_num))
Allocate(tmp1(n_points_final_grid,3,mo_num), tmp2(n_points_final_grid,mo_num))
Allocate(tmpval_1(n_points_final_grid), tmpval_2(n_points_final_grid))
Allocate(tmpvec_1(n_points_final_grid,3), tmpvec_2(n_points_final_grid,3))
Tmp_3d = 0.d0
Tmp_2d = 0.d0
Tmp1 = 0.d0
Tmp2 = 0.d0
Tmpval_1 = 0.d0
Tmpval_2 = 0.d0
Tmpvec_1 = 0.d0
Tmpvec_2 = 0.d0
!$OMP DO
do ii = Ne(2) + 1, Ne(1)
i = occ(ii,1)
do h1 = 1, mo_num
!$OMP PARALLEL &
!$OMP DEFAULT (NONE) &
!$OMP PRIVATE (ipoint) &
!$OMP SHARED (n_points_final_grid, i, h1, &
!$OMP mos_l_in_r_array_transp, mos_r_in_r_array_transp, &
!$OMP int2_grad1_u12_bimo_t, final_weight_at_r_vector, &
!$OMP tmpval_1, tmpval_2, tmpvec_1, tmpvec_2)
!$OMP DO
do ipoint = 1, n_points_final_grid
tmpval_1(ipoint) = final_weight_at_r_vector(ipoint) * mos_l_in_r_array_transp(ipoint,i) * mos_r_in_r_array_transp(ipoint, i)
tmpval_2(ipoint) = final_weight_at_r_vector(ipoint) * mos_l_in_r_array_transp(ipoint,i) * mos_r_in_r_array_transp(ipoint,h1)
@ -1227,16 +1251,7 @@ BEGIN_PROVIDER [ double precision, no_aba_contraction, (mo_num,mo_num,mo_num,mo_
tmpvec_2(ipoint,2) = final_weight_at_r_vector(ipoint) * int2_grad1_u12_bimo_t(ipoint,2,i,h1) * mos_r_in_r_array_transp(ipoint, i)
tmpvec_2(ipoint,3) = final_weight_at_r_vector(ipoint) * int2_grad1_u12_bimo_t(ipoint,3,i,h1) * mos_r_in_r_array_transp(ipoint, i)
enddo
!$OMP END DO
!$OMP END PARALLEL
!$OMP PARALLEL &
!$OMP DEFAULT (NONE) &
!$OMP PRIVATE (p1, ipoint) &
!$OMP SHARED (mo_num, n_points_final_grid, h1, i, &
!$OMP mos_l_in_r_array_transp, int2_grad1_u12_bimo_t, &
!$OMP tmpval_1, tmpval_2, tmpvec_1, tmpvec_2, tmp1)
!$OMP DO
do p1 = 1, mo_num
do ipoint = 1, n_points_final_grid
tmp1(ipoint,1,p1) = mos_l_in_r_array_transp(ipoint,p1) * (tmpvec_1(ipoint,1) - tmpvec_2(ipoint,1)) &
@ -1247,82 +1262,65 @@ BEGIN_PROVIDER [ double precision, no_aba_contraction, (mo_num,mo_num,mo_num,mo_
+ tmpval_1(ipoint) * int2_grad1_u12_bimo_t(ipoint,3,p1,h1) - tmpval_2(ipoint) * int2_grad1_u12_bimo_t(ipoint,3,p1,i)
enddo
enddo
!$OMP END DO
!$OMP END PARALLEL
call dgemm( 'T', 'N', mo_num*mo_num, mo_num, 3*n_points_final_grid, 0.5d0 &
, int2_grad1_u12_bimo_t(1,1,1,1), 3*n_points_final_grid &
, tmp1(1,1,1), 3*n_points_final_grid &
, 0.d0, tmp_3d(1,1,1), mo_num*mo_num)
!$OMP PARALLEL DO PRIVATE(p1,h2,p2)
do p1 = 1, mo_num
do h2 = 1, mo_num
do p2 = 1, mo_num
!$OMP CRITICAL
no_aba_contraction(p2,h2,p1,h1) = no_aba_contraction(p2,h2,p1,h1) + tmp_3d(p2,h2,p1)
!$OMP END CRITICAL
enddo
enddo
enddo
!$OMP END PARALLEL DO
do p1 = 1, mo_num
! to minimize the number of operations
!$OMP PARALLEL &
!$OMP DEFAULT (NONE) &
!$OMP PRIVATE (ipoint) &
!$OMP SHARED (n_points_final_grid, i, h1, p1, &
!$OMP int2_grad1_u12_bimo_t, final_weight_at_r_vector, &
!$OMP tmpval_1)
!$OMP DO
do ipoint = 1, n_points_final_grid
tmpval_1(ipoint) = final_weight_at_r_vector(ipoint) * ( int2_grad1_u12_bimo_t(ipoint,1, i,i) * int2_grad1_u12_bimo_t(ipoint,1,p1,h1) &
+ int2_grad1_u12_bimo_t(ipoint,2, i,i) * int2_grad1_u12_bimo_t(ipoint,2,p1,h1) &
+ int2_grad1_u12_bimo_t(ipoint,3, i,i) * int2_grad1_u12_bimo_t(ipoint,3,p1,h1) &
- int2_grad1_u12_bimo_t(ipoint,1,p1,i) * int2_grad1_u12_bimo_t(ipoint,1, i,h1) &
- int2_grad1_u12_bimo_t(ipoint,2,p1,i) * int2_grad1_u12_bimo_t(ipoint,2, i,h1) &
- int2_grad1_u12_bimo_t(ipoint,3,p1,i) * int2_grad1_u12_bimo_t(ipoint,3, i,h1) )
tmpval_1(ipoint) = final_weight_at_r_vector(ipoint) * &
( int2_grad1_u12_bimo_t(ipoint,1, i,i) * int2_grad1_u12_bimo_t(ipoint,1,p1,h1) &
+ int2_grad1_u12_bimo_t(ipoint,2, i,i) * int2_grad1_u12_bimo_t(ipoint,2,p1,h1) &
+ int2_grad1_u12_bimo_t(ipoint,3, i,i) * int2_grad1_u12_bimo_t(ipoint,3,p1,h1) &
- int2_grad1_u12_bimo_t(ipoint,1,p1,i) * int2_grad1_u12_bimo_t(ipoint,1, i,h1) &
- int2_grad1_u12_bimo_t(ipoint,2,p1,i) * int2_grad1_u12_bimo_t(ipoint,2, i,h1) &
- int2_grad1_u12_bimo_t(ipoint,3,p1,i) * int2_grad1_u12_bimo_t(ipoint,3, i,h1) )
enddo
!$OMP END DO
!$OMP END PARALLEL
!$OMP PARALLEL &
!$OMP DEFAULT (NONE) &
!$OMP PRIVATE (h2, ipoint) &
!$OMP SHARED (mo_num, n_points_final_grid, &
!$OMP mos_r_in_r_array_transp, &
!$OMP tmpval_1, tmp2)
!$OMP DO
do h2 = 1, mo_num
do ipoint = 1, n_points_final_grid
tmp2(ipoint,h2) = mos_r_in_r_array_transp(ipoint,h2) * tmpval_1(ipoint)
enddo
enddo
!$OMP END DO
!$OMP END PARALLEL
call dgemm( 'T', 'N', mo_num, mo_num, n_points_final_grid, 0.5d0 &
, mos_l_in_r_array_transp(1,1), n_points_final_grid &
, tmp2(1,1), n_points_final_grid &
, 0.d0, tmp_2d(1,1), mo_num)
!$OMP PARALLEL DO PRIVATE(h2,p2)
do h2 = 1, mo_num
do p2 = 1, mo_num
!$OMP CRITICAL
no_aba_contraction(p2,h2,p1,h1) = no_aba_contraction(p2,h2,p1,h1) + tmp_2d(p2,h2)
!$OMP END CRITICAL
enddo
enddo
!$OMP END PARALLEL DO
enddo ! p1
enddo ! h1
enddo !i
endif
!$OMP END DO
deallocate(tmp_2d, tmp_3d)
deallocate(tmp1, tmp2)
deallocate(tmpval_1, tmpval_2)
deallocate(tmpvec_1, tmpvec_2)
deallocate(tmp_3d, tmp_2d)
deallocate(tmp1, tmp2)
deallocate(tmpval_1, tmpval_2)
deallocate(tmpvec_1, tmpvec_2)
!$OMP END PARALLEL
endif
no_aba_contraction = -0.5d0 * no_aba_contraction
call sum_A_At(no_aba_contraction(1,1,1,1), mo_num*mo_num)

View File

@ -18,7 +18,8 @@ program tc_bi_ortho
! call timing_single
! call timing_double
call test_no()
!call test_no()
call test_no_aba()
end
subroutine test_h_u0
@ -297,4 +298,45 @@ end
! ---
subroutine test_no_aba()
implicit none
integer :: i, j, k, l
double precision :: accu, contrib, new, ref, thr
print*, ' testing no_aba_contraction ...'
thr = 1d-8
PROVIDE no_aba_contraction_v0
PROVIDE no_aba_contraction
accu = 0.d0
do i = 1, mo_num
do j = 1, mo_num
do k = 1, mo_num
do l = 1, mo_num
new = no_aba_contraction (l,k,j,i)
ref = no_aba_contraction_v0(l,k,j,i)
contrib = dabs(new - ref)
accu += contrib
if(contrib .gt. thr) then
print*, ' problem on no_aba_contraction'
print*, l, k, j, i
print*, ref, new, contrib
stop
endif
enddo
enddo
enddo
enddo
print*, ' accu on no_aba_contraction = ', accu / dble(mo_num)**4
return
end
! ---