LCPQ/qp2
9
1
Fork 0
mirror of https://github.com/QuantumPackage/qp2.git synced 2024-06-01 18:25:17 +02:00
Code Releases Activity

Improved weights in selection

Browse Source
This commit is contained in:
Anthony Scemama 2021-02-13 16:19:36 +01:00
parent 112473e242
commit 13fa8d19b6
1 changed files with 1 additions and 35 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

36
src/cipsi/selection_weight.irp.f
View File

@ -33,53 +33,21 @@ subroutine update_pt2_and_variance_weights(pt2_data, N_st)
double precision :: avg, element, dt, x
integer :: k
! integer, save :: i_iter=0
! integer, parameter :: i_itermax = 1
! double precision, allocatable, save :: memo_variance(:,:), memo_pt2(:,:)
pt2(:) = pt2_data % pt2(:)
variance(:) = pt2_data % variance(:)
! if (i_iter == 0) then
! allocate(memo_variance(N_st,i_itermax), memo_pt2(N_st,i_itermax))
! memo_pt2(:,:) = 1.d0
! memo_variance(:,:) = 1.d0
! endif
!
! i_iter = i_iter+1
! if (i_iter > i_itermax) then
! i_iter = 1
! endif
!
! dt = 2.0d0
avg = sum(pt2(1:N_st)) / dble(N_st) + 1.d-32 ! Avoid future division by zero
! do k=1,N_st
! element = exp(dt*(pt2(k)/avg -1.d0))
! element = min(2.0d0 , element)
! element = max(0.5d0 , element)
! memo_pt2(k,i_iter) = element
! pt2_match_weight(k) *= product(memo_pt2(k,:))
!enddo
dt = 1.0d0 * selection_factor
dt = 2.d0 !* selection_factor
do k=1,N_st
element = exp(dt*(pt2(k)/avg - 1.d0))
element = min(2.0d0 , element)
element = max(0.5d0 , element)
print *, k, element
pt2_match_weight(k) *= element
enddo
avg = sum(variance(1:N_st)) / dble(N_st) + 1.d-32 ! Avoid future division by zero
! do k=1,N_st
! element = exp(dt*(variance(k)/avg -1.d0))
! element = min(2.0d0 , element)
! element = max(0.5d0 , element)
! memo_variance(k,i_iter) = element
! variance_match_weight(k) *= product(memo_variance(k,:))
! enddo
do k=1,N_st
element = exp(dt*(variance(k)/avg -1.d0))
@ -94,8 +62,6 @@ subroutine update_pt2_and_variance_weights(pt2_data, N_st)
variance_match_weight(:) = 1.d0
endif
print *, 'XXX', n_det, pt2_match_weight(1), pt2_match_weight(2)
threshold_davidson_pt2 = min(1.d-6, &
max(threshold_davidson, 1.e-1 * PT2_relative_error * minval(abs(pt2(1:N_states)))) )