10
0
mirror of https://github.com/LCPQ/quantum_package synced 2024-12-22 20:35:19 +01:00
This commit is contained in:
Yann Garniron 2017-01-04 10:06:41 +01:00
parent aebc386a3e
commit e1d014aa41
2 changed files with 410 additions and 4 deletions

View File

@ -18,7 +18,7 @@ IRPF90_FLAGS : --ninja --align=32
# 0 : Deactivate
#
[OPTION]
MODE : OPT ; [ OPT | PROFILE | DEBUG ] : Chooses the section below
MODE : DEBUG ; [ OPT | PROFILE | DEBUG ] : Chooses the section below
CACHE : 1 ; Enable cache_compile.py
OPENMP : 1 ; Append OpenMP flags

View File

@ -85,9 +85,10 @@ program fci_zmq
if(do_pt2_end)then
print*,'Last iteration only to compute the PT2'
threshold_selectors = 1.d0
threshold_generators = 0.9999d0
threshold_generators = 1d0 ! 0.9999d0
E_CI_before(1:N_states) = CI_energy(1:N_states)
call ZMQ_selection(0, pt2)
!call ZMQ_selection(0, pt2) pour non-stochastic
call ZMQ_pt2(pt2)
print *, 'Final step'
print *, 'N_det = ', N_det
print *, 'N_states = ', N_states
@ -103,8 +104,182 @@ program fci_zmq
call save_wavefunction
end
! subroutine ZMQ_pt2(pt2)
! use f77_zmq
! use selection_types
!
! implicit none
!
! character*(1000000) :: task
! integer(ZMQ_PTR) :: zmq_to_qp_run_socket
! type(selection_buffer) :: b
! integer :: i, N
! integer, external :: omp_get_thread_num
! double precision, intent(out) :: pt2(N_states)
!
! integer*8, allocatable :: bulk(:), tirage(:)
! integer, allocatable :: todo(:)
! double precision, allocatable :: pt2_detail(:,:), val(:,:), weight(:)
! double precision :: sume, sume2
! double precision :: tot_n
!
! allocate(bulk(N_det), tirage(N_det), todo(0:N_det), pt2_detail(N_states, N_det), val(N_states, N_det))
!
! sume = 0d0
! sume2 = 0d0
! tot_n = 0d0
! bulk = 0
! tirage = 0
! todo = 0
!
!
! N = 1
! provide nproc
! provide ci_electronic_energy
! call new_parallel_job(zmq_to_qp_run_socket,"pt2")
! call zmq_put_psi(zmq_to_qp_run_socket,1,ci_electronic_energy,size(ci_electronic_energy))
! call zmq_set_running(zmq_to_qp_run_socket)
! call create_selection_buffer(N, N*2, b)
!
! integer :: i_generator, i_generator_end, generator_per_task, step
!
! integer :: mergeN
! mergeN = 100
! call get_carlo_workbatch(tirage, weight, todo, bulk, 1d-2, mergeN)
! print *, "CARLO", todo(0), mergeN
!
! generator_per_task = todo(0)/1000 + 1
! do i=1,todo(0),generator_per_task
! i_generator_end = min(i+generator_per_task-1, todo(0))
! print *, "TASK", (i_generator_end-i+1), todo(i:i_generator_end)
! write(task,*) (i_generator_end-i+1), todo(i:i_generator_end)
! call add_task_to_taskserver(zmq_to_qp_run_socket,task)
! end do
! print *, "tasked"
! pt2_detail = 0d0
! !$OMP PARALLEL DEFAULT(shared) SHARED(b, pt2) PRIVATE(i) NUM_THREADS(nproc+1)
! i = omp_get_thread_num()
! if (i==0) then
! call pt2_collector(b, pt2_detail)
! else
! call pt2_slave_inproc(i)
! endif
! !$OMP END PARALLEL
! call end_parallel_job(zmq_to_qp_run_socket, 'pt2')
! print *, "daune"
! val += pt2_detail
! call perform_carlo(tirage, weight, bulk, val, sume, sume2, mergeN)
! tot_n = 0
! double precision :: sweight
! sweight = 0d0
! do i=1,N_det
! if(weight(i) /= 0) tot_n = tot_n + dfloat(bulk(i))
! sweight += weight(i)
! end do
! print *, "PT2_DETAIL", tot_n, sume/tot_n, sume, sume2
! pt2 = 0d0
! do i=1,N_det
! if(weight(i) /= 0d0) exit
! pt2(:) += pt2_detail(:,i)
! end do
! print *, "N_determinist = ", i-1
! end subroutine
subroutine ZMQ_pt2(pt2)
use f77_zmq
use selection_types
implicit none
character*(1000000) :: task
integer(ZMQ_PTR) :: zmq_to_qp_run_socket
type(selection_buffer) :: b
integer, external :: omp_get_thread_num
double precision, intent(out) :: pt2(N_states)
double precision :: pt2_detail(N_states, N_det_generators), comb(100000)
logical :: computed(N_det_generators)
integer :: tbc(0:N_det_generators)
integer :: i, Ncomb, generator_per_task, i_generator_end
integer, external :: pt2_find
provide nproc
call new_parallel_job(zmq_to_qp_run_socket,"pt2")
call zmq_put_psi(zmq_to_qp_run_socket,1,ci_electronic_energy,size(ci_electronic_energy))
call zmq_set_running(zmq_to_qp_run_socket)
call create_selection_buffer(1, 1*2, b)
call random_seed()
computed = .false.
tbc(0) = first_det_of_comb - 1
do i=1, tbc(0)
tbc(i) = i
computed(i) = .true.
end do
print *, "detererminist initial ", tbc(0)+1
!LOOP?
call get_carlo_workbatch(1d-3, computed, comb, Ncomb, tbc)
generator_per_task = tbc(0)/1000 + 1
print *, "TASK", tbc(0), tbc(1:tbc(0))
do i=1,tbc(0),generator_per_task
i_generator_end = min(i+generator_per_task-1, tbc(0))
!print *, "TASK", (i_generator_end-i+1), tbc(i:i_generator_end)
write(task,*) (i_generator_end-i+1), tbc(i:i_generator_end)
call add_task_to_taskserver(zmq_to_qp_run_socket,task)
end do
pt2_detail = 0d0
!$OMP PARALLEL DEFAULT(shared) SHARED(b, pt2) PRIVATE(i) NUM_THREADS(nproc+1)
i = omp_get_thread_num()
if (i==0) then
call pt2_collector(b, pt2_detail)
else
call pt2_slave_inproc(i)
endif
!$OMP END PARALLEL
call end_parallel_job(zmq_to_qp_run_socket, 'pt2')
double precision :: E0, avg, eqt
double precision :: sumabove(comb_teeth), sum2above(comb_teeth), Nabove(comb_teeth)
call do_carlo(tbc, Ncomb, comb, pt2_detail, sumabove, sum2above, Nabove)
tbc(0) = 0
!END LOOP?
integer :: tooth
!-8.091550677158776E-003
call get_first_tooth(computed, tooth)
E0 = sum(pt2_detail(1,:first_det_of_teeth(tooth)-1))
avg = E0 + (sumabove(tooth) / Nabove(tooth))
eqt = sqrt(1d0 / (Nabove(tooth)-1) * abs(sum2above(tooth) / Nabove(tooth) - (sumabove(tooth)/Nabove(tooth))**2))
print *, "PT2 ", avg, "+/-", eqt
pt2 = 0d0
end subroutine
subroutine do_carlo(tbc, Ncomb, comb, pt2_detail, sumabove, sum2above, Nabove)
integer, intent(in) :: tbc(0:N_det_generators), Ncomb
double precision, intent(in) :: comb(Ncomb), pt2_detail(N_states, N_det_generators)
double precision, intent(inout) :: sumabove(comb_teeth), sum2above(comb_teeth), Nabove(comb_teeth)
integer :: i, dets(comb_teeth)
double precision :: myVal, myVal2
do i=1,Ncomb
call get_comb(comb(i), dets)
myVal = 0d0
myVal2 = 0d0
do j=comb_teeth,1,-1
if(pt2_detail(1, dets(j)) == -1d0) print *, "uncalculatedidified", dets(j), pt2_detail(1, dets(j)-1:dets(j)+1)
myVal += pt2_detail(1, dets(j)) / weight(dets(j)) * comb_step
sumabove(j) += myVal
sum2above(j) += myVal**2
Nabove(j) += 1
end do
end do
end subroutine
subroutine ZMQ_selection(N_in, pt2)
use f77_zmq
@ -112,7 +287,7 @@ subroutine ZMQ_selection(N_in, pt2)
implicit none
character*(512) :: task
character*(1000000) :: task
integer(ZMQ_PTR) :: zmq_to_qp_run_socket
integer, intent(in) :: N_in
type(selection_buffer) :: b
@ -164,6 +339,72 @@ subroutine selection_slave_inproc(i)
call run_selection_slave(1,i,ci_electronic_energy)
end
subroutine pt2_slave_inproc(i)
implicit none
integer, intent(in) :: i
call run_pt2_slave(1,i,ci_electronic_energy)
end
subroutine pt2_collector(b, pt2_detail)
use f77_zmq
use selection_types
use bitmasks
implicit none
type(selection_buffer), intent(inout) :: b
double precision, intent(out) :: pt2_detail(N_states, N_det)
double precision :: pt2_mwen(N_states, N_det)
integer(ZMQ_PTR),external :: new_zmq_to_qp_run_socket
integer(ZMQ_PTR) :: zmq_to_qp_run_socket
integer(ZMQ_PTR), external :: new_zmq_pull_socket
integer(ZMQ_PTR) :: zmq_socket_pull
integer :: msg_size, rc, more
integer :: acc, i, j, robin, N, ntask
double precision, allocatable :: val(:)
integer(bit_kind), allocatable :: det(:,:,:)
integer, allocatable :: task_id(:)
integer :: done, Nindex
integer, allocatable :: index(:)
real :: time, time0
zmq_to_qp_run_socket = new_zmq_to_qp_run_socket()
zmq_socket_pull = new_zmq_pull_socket()
allocate(val(b%N), det(N_int, 2, b%N), task_id(N_det), index(N_det))
done = 0
more = 1
pt2_detail = -1d0
call CPU_TIME(time0)
do while (more == 1)
call pull_pt2_results(zmq_socket_pull, Nindex, index, pt2_mwen, task_id, ntask)
do i=1,Nindex
pt2_detail(:, index(i)) += pt2_mwen(:,i)
end do
!do i=1, N
! call add_to_selection_buffer(b, det(1,1,i), val(i))
!end do
do i=1, ntask
if(task_id(i) == 0) then
print *, "Error in collector"
endif
call zmq_delete_task(zmq_to_qp_run_socket,zmq_socket_pull,task_id(i),more)
end do
done += ntask
call CPU_TIME(time)
! print *, "DONE" , done, time - time0
end do
call end_zmq_to_qp_run_socket(zmq_to_qp_run_socket)
call end_zmq_pull_socket(zmq_socket_pull)
call sort_selection_buffer(b)
end subroutine
subroutine selection_collector(b, pt2)
use f77_zmq
use selection_types
@ -218,3 +459,168 @@ subroutine selection_collector(b, pt2)
call sort_selection_buffer(b)
end subroutine
integer function pt2_find(v, w)
implicit none
double precision :: v, w(N_det)
integer :: i,l,h
l = 0
h = N_det-1
do while(h >= l)
i = (h+l)/2
if(w(i+1) > v) then
h = i-1
else
l = i+1
end if
end do
pt2_find = l+1
end function
BEGIN_PROVIDER [ integer, comb_teeth ]
implicit none
comb_teeth = 10
END_PROVIDER
subroutine get_first_tooth(computed, first_teeth)
implicit none
logical, intent(in) :: computed(N_det_generators)
integer, intent(out) :: first_teeth
integer :: i
first_teeth = 1
do i=first_det_of_comb, N_det_generators
if(not(computed(i))) then
first_teeth = i
exit
end if
end do
do i=comb_teeth, 1, -1
if(first_det_of_teeth(i) < first_teeth) then
first_teeth = i
exit
end if
end do
end subroutine
subroutine get_carlo_workbatch(maxWorkload, computed, comb, Ncomb, tbc)
implicit none
double precision, intent(in) :: maxWorkload
double precision, intent(out) :: comb(N_det_generators)
integer, intent(inout) :: tbc(0:N_det_generators)
integer, intent(out) :: Ncomb
logical, intent(inout) :: computed(N_det_generators)
integer :: i, dets(comb_teeth)
double precision :: myWorkload
myWorkload = 0d0
do i=1,size(comb)
call RANDOM_NUMBER(comb(i))
comb(i) = comb(i) * comb_step
call add_comb(comb(i), computed, tbc, myWorkload)
Ncomb = i
if(myWorkload > maxWorkload) exit
end do
end subroutine
subroutine get_comb(stato, dets)
implicit none
double precision, intent(in) :: stato
integer, intent(out) :: dets(comb_teeth)
double precision :: curs
integer :: j
integer, external :: pt2_find
curs = 1d0 - stato
do j = comb_teeth, 1, -1
dets(j) = pt2_find(curs, cweight)
curs -= comb_step
end do
end subroutine
subroutine add_comb(comb, computed, tbc, workload)
implicit none
double precision, intent(in) :: comb
logical, intent(inout) :: computed(N_det_generators)
double precision, intent(inout) :: workload
integer, intent(inout) :: tbc(0:N_det_generators)
integer :: i, dets(comb_teeth)
call get_comb(comb, dets)
do i = 1, comb_teeth
if(not(computed(dets(i)))) then
tbc(0) += 1
tbc(tbc(0)) = dets(i)
workload += comb_workload(dets(i))
computed(dets(i)) = .true.
end if
end do
end subroutine
BEGIN_PROVIDER [ double precision, weight, (N_det_generators) ]
&BEGIN_PROVIDER [ double precision, cweight, (N_det_generators) ]
&BEGIN_PROVIDER [ double precision, comb_workload, (N_det_generators) ]
&BEGIN_PROVIDER [ double precision, comb_step ]
&BEGIN_PROVIDER [ integer, first_det_of_teeth, (comb_teeth) ]
&BEGIN_PROVIDER [ integer, first_det_of_comb ]
implicit none
integer :: i
double precision :: norm_left, stato
integer, external :: pt2_find
weight(1) = psi_coef_generators(1,1)**2
cweight(1) = psi_coef_generators(1,1)**2
do i=2,N_det_generators
weight(i) = psi_coef_generators(i,1)**2
cweight(i) = cweight(i-1) + psi_coef_generators(i,1)**2
end do
weight = weight / cweight(N_det_generators)
cweight = cweight / cweight(N_det_generators)
comb_workload = 1d0 / dfloat(N_det_generators)
norm_left = 1d0
comb_step = 1d0/dfloat(comb_teeth)
do i=1,N_det_generators
if(weight(i)/norm_left < comb_step/2d0) then
first_det_of_comb = i
exit
end if
norm_left -= weight(i)
end do
comb_step = 1d0 / dfloat(comb_teeth) * (1d0 - cweight(first_det_of_comb-1))
stato = 1d0 - comb_step + 1d-5
do i=comb_teeth, 1, -1
first_det_of_teeth(i) = pt2_find(stato, cweight)
stato -= comb_step
end do
print *, first_det_of_teeth(1), first_det_of_comb
if(first_det_of_teeth(1) /= first_det_of_comb) stop "comb provider"
END_PROVIDER