dynVCLVec(const int size_in, T scalar, const int ctx_id){
viennacl::context ctx;
// explicitly pull context for thread safe forking
ctx = viennacl::context(viennacl::ocl::get_context(static_cast<long>(ctx_id)));
// A.switch_memory_context(ctx);
// A = viennacl::scalar_vector<T>(size_in, scalar, ctx);
// A = static_cast<viennacl::vector_base<T> >(A);
viennacl::vector_base<T> A = viennacl::vector_base<T>(size_in, ctx);
viennacl::linalg::vector_assign(A, scalar);
size = size_in;
// ptr = &A;
begin = 1;
last = size_in;
viennacl::range temp_r(begin-1, last);
r = temp_r;
// shptr.reset(&A, [](decltype(&A) p) {
// std::cout << "Call delete from lambda...\n";
// // delete p;
// });
shared = false;
shared_type = 0;
shptr = std::make_shared<viennacl::vector_base<T> >(A);
};
dynVCLVec(viennacl::matrix_range<viennacl::matrix<T> > &mat, const int ctx_id) {
viennacl::context ctx;
// explicitly pull context for thread safe forking
ctx = viennacl::context(viennacl::ocl::get_context(static_cast<long>(ctx_id)));
viennacl::vector_base<T> A = viennacl::vector_base<T>(mat.size1() * mat.size2(), ctx);
viennacl::matrix_base<T> dummy(A.handle(),
mat.size1(), 0, 1, mat.size1(), //row layout
mat.size2(), 0, 1, mat.size2(), //column layout
true); // row-major
dummy = mat;
// shared = true;
size = A.size();
begin = 1;
last = size;
// ptr = &A;
viennacl::range temp_r(0, size);
r = temp_r;
shared = false;
shared_type = 0;
shptr = std::make_shared<viennacl::vector_base<T> >(A);
}
void
dynVCLVec<T>::setRange(int start, int end){
viennacl::range temp_r(start-1, end);
r = temp_r;
begin = start;
last = end;
}
dynVCLVec<T>::dynVCLVec(SEXP A_, int device_flag)
{
Eigen::Matrix<T, Eigen::Dynamic, 1> Am;
Am = Rcpp::as<Eigen::Matrix<T, Eigen::Dynamic, 1> >(A_);
// define device type to use
if(device_flag == 0){
//use only GPUs
long id = 0;
viennacl::ocl::set_context_device_type(id, viennacl::ocl::gpu_tag());
viennacl::ocl::switch_context(id);
}else{
// use only CPUs
long id = 1;
viennacl::ocl::set_context_device_type(id, viennacl::ocl::cpu_tag());
viennacl::ocl::switch_context(id);
}
int K = Am.size();
A = viennacl::vector<T>(K);
viennacl::copy(Am, A);
size = K;
begin = 1;
last = size;
ptr = &A;
viennacl::range temp_r(0, K);
r = temp_r;
}
// dynVCLVec(viennacl::vector_range<viennacl::vector_base<T> > vec, int ctx_id){
// viennacl::context ctx;
//
// // explicitly pull context for thread safe forking
// ctx = viennacl::context(viennacl::ocl::get_context(static_cast<long>(ctx_id)));
//
// A.switch_memory_context(ctx);
// A = vec;
//
// size = A.size();
// begin = 1;
// last = size;
// ptr = &A;
// viennacl::range temp_r(0, size);
// r = temp_r;
// }
dynVCLVec(SEXP A_, int ctx_id) {
#ifdef UNDEF
Eigen::Matrix<T, Eigen::Dynamic, 1> Am;
Am = Rcpp::as<Eigen::Matrix<T, Eigen::Dynamic, 1> >(A_);
int K = Am.size();
viennacl::context ctx;
// explicitly pull context for thread safe forking
ctx = viennacl::context(viennacl::ocl::get_context(static_cast<long>(ctx_id)));
// std::cout << "about to initialize" << std::endl;
viennacl::vector_base<T> A = viennacl::vector_base<T>(K, ctx);
// std::cout << "initialized vector" << std::endl;
viennacl::fast_copy(Am.data(), Am.data() + Am.size(), A.begin());
// viennacl::fast_copy(Am.begin(), Am.end(), A.begin());
// std::cout << "copied" << std::endl;
size = A.size();
begin = 1;
last = size;
// ptr = &A;
viennacl::range temp_r(0, size);
r = temp_r;
shared = false;
shared_type = 0;
shptr = std::make_shared<viennacl::vector_base<T> >(A);
#endif
}
// dynVCLVec(viennacl::vector_base<T> *vec) : ptr(vec){
// // A = vec;
//
// size = vec->size();
// begin = 1;
// last = size;
// viennacl::range temp_r(0, size);
// r = temp_r;
// shptr = std::make_shared<viennacl::vector_base<T> >(A);
// }
dynVCLVec(viennacl::matrix<T> *mat) : ptr_matrix(mat) {
shared = true;
shared_type = 0;
size = mat->internal_size();
begin = 1;
last = size;
viennacl::range temp_r(0, size);
r = temp_r;
}
dynVCLVec<T>::dynVCLVec(Rcpp::XPtr<dynVCLVec<T> > dynVec)
{
size = dynVec->length();
begin = dynVec->start();
last = dynVec->end();
ptr = dynVec->getPtr();
viennacl::range temp_r(begin-1, last);
r = temp_r;
}
Rcpp::List RadiusSearch(Rcpp::NumericMatrix query_,
Rcpp::NumericMatrix ref_,
double radius,
int max_neighbour,
std::string build,
int cores,
int checks) {
const std::size_t n_dim = query_.ncol();
const std::size_t n_query = query_.nrow();
const std::size_t n_ref = ref_.nrow();
// Column major to row major
arma::mat query(n_dim, n_query);
{
arma::mat temp_q(query_.begin(), n_query, n_dim, false);
query = arma::trans(temp_q);
}
flann::Matrix<double> q_flann(query.memptr(), n_query, n_dim);
arma::mat ref(n_dim, n_ref);
{
arma::mat temp_r(ref_.begin(), n_ref, n_dim, false);
ref = arma::trans(temp_r);
}
flann::Matrix<double> ref_flann(ref.memptr(), n_ref, n_dim);
// Setting the flann index params
flann::IndexParams params;
if (build == "kdtree") {
params = flann::KDTreeSingleIndexParams(1);
} else if (build == "kmeans") {
params = flann::KMeansIndexParams(2, 10, flann::FLANN_CENTERS_RANDOM, 0.2);
} else if (build == "linear") {
params = flann::LinearIndexParams();
}
// Perform the radius search
flann::Index<flann::L2<double> > index(ref_flann, params);
index.buildIndex();
std::vector< std::vector<int> >
indices_flann(n_query, std::vector<int>(max_neighbour));
std::vector< std::vector<double> >
dists_flann(n_query, std::vector<double>(max_neighbour));
flann::SearchParams search_params;
search_params.cores = cores;
search_params.checks = checks;
search_params.max_neighbors = max_neighbour;
index.radiusSearch(q_flann, indices_flann, dists_flann, radius,
search_params);
return Rcpp::List::create(Rcpp::Named("indices") = indices_flann,
Rcpp::Named("distances") = dists_flann);
}
// margin - true = rows, false = cols
dynVCLVec(viennacl::matrix<T> *mat, const bool margin, int start) : ptr_matrix(mat) {
shared = true;
if(margin){
shared_type = 1;
size = mat->size2();
}else{
shared_type = 2;
size = mat->size1();
// viennacl::vector_base<T> A = viennacl::vector_base<T>(ptr_matrix->handle(), ptr_matrix->size1(), begin, ptr_matrix->internal_size2());
// ptr = &A;
}
begin = start;
last = size;
viennacl::range temp_r(0, size);
r = temp_r;
}
dynVCLVec(viennacl::vector_base<T> vec, int ctx_id) {
// viennacl::context ctx;
//
// // // explicitly pull context for thread safe forking
// ctx = viennacl::context(viennacl::ocl::get_context(static_cast<long>(ctx_id)));
//
// A.switch_memory_context(ctx);
viennacl::vector_base<T> A = vec;
size = A.size();
begin = 1;
last = size;
// ptr = &A;
viennacl::range temp_r(0, size);
r = temp_r;
shared = false;
shared_type = 0;
shptr = std::make_shared<viennacl::vector_base<T> >(A);
}
vector3 ViscositySmoothKernel::getLaplacian(vector3 *r_i, vector3 *r_j, float h_size)
{
//r_i表示产生作用的粒子位置
//r_j表示当前粒子位置
vector3 temp_r(r_i->x - r_j->x, r_i->y - r_j->y, r_i->z - r_j->z);
float r_len = temp_r.length();
float result;
if(r_len >= 0 && r_len <= h_size)
{
//result = (45*(h-r)/(PAI*h^6))
result = (float)(45*(h_size - r_len)/(PAI*pow(h_size, 6)));
}
else
{
result = 0;
}
temp_r = (temp_r.normalize())*result;
return temp_r;
}
vector3 Poly6SmoothKernel::getValue(vector3 *r_i, vector3 *r_j, float h_size)
{
//r_i表示产生作用的粒子位置
//r_j表示当前粒子位置
vector3 temp_r(r_i->x - r_j->x, r_i->y - r_j->y, r_i->z - r_j->z);
float r_len = temp_r.length();
float result;
if(r_len >= 0 && r_len <= h_size)
{
//result = (315/(64*PAI*h^9))*(h^2 - r^2)^3)
result = (float)((4.921875*pow((pow(h_size, 2) - pow(r_len, 2)),3))/(PAI*pow(h_size, 9)));
}
else
{
result = 0;
}
temp_r = (temp_r.normalize())*result;
return temp_r;
}
dynVCLVec<T>::dynVCLVec(int size_in, int device_flag)
{
// define device type to use
if(device_flag == 0){
//use only GPUs
long id = 0;
viennacl::ocl::set_context_device_type(id, viennacl::ocl::gpu_tag());
viennacl::ocl::switch_context(id);
}else{
// use only CPUs
long id = 1;
viennacl::ocl::set_context_device_type(id, viennacl::ocl::cpu_tag());
viennacl::ocl::switch_context(id);
}
A = viennacl::zero_vector<T>(size_in);
begin = 1;
last = size_in;
ptr = &A;
viennacl::range temp_r(begin-1, last);
r = temp_r;
}
vector3 ViscositySmoothKernel::getGrads(vector3 *r_i, vector3 *r_j, float h_size)
{
//r_i表示产生作用的粒子位置
//r_j表示当前粒子位置
vector3 temp_r(r_i->x - r_j->x, r_i->y - r_j->y, r_i->z - r_j->z);
float r_len = temp_r.length();
float result;
if(r_len >= 0 && r_len <= h_size)
{
//result = (15*(-3*r^2/(2*h^3) + 2*r/h^2 - h/(2*r^2))/(2*PAI*h^3))
result = (float)(7.5*(-3*r_len*r_len/(2*pow(h_size, 3)) + 2*r_len/(h_size*h_size)
- h_size/(2*r_len*r_len))/(PAI*pow(h_size, 3)));
}
else
{
result = 0;
}
temp_r = (temp_r.normalize())*result;
return temp_r;
}
vector3 Poly6SmoothKernel::getLaplacian(vector3 *r_i, vector3 *r_j, float h_size)
{
//r_i表示产生作用的粒子位置
//r_j表示当前粒子位置
vector3 temp_r(r_i->x - r_j->x, r_i->y - r_j->y, r_i->z - r_j->z);
float r_len = temp_r.length();
float result;
if(r_len >= 0 && r_len <= h_size)
{
//result = (315/(64*PAI*h^9))*(-6)*(h^2 - r^2)*(h^2 + 3*r^2))
result = (float)((-29.53125*(h_size*h_size - r_len*r_len)*
(h_size*h_size + 3*r_len*r_len))/(PAI*pow(h_size, 9)));
}
else
{
result = 0;
}
temp_r = (temp_r.normalize())*result;
return temp_r;
}
// dynVCLVec(Eigen::Matrix<T, Eigen::Dynamic,1> &A_);
// dynVCLVec(Eigen::Map<Eigen::Matrix<T, Eigen::Dynamic, 1> > &A_, int size_);
dynVCLVec(const int size_in, const int ctx_id) {
viennacl::context ctx;
// explicitly pull context for thread safe forking
ctx = viennacl::context(viennacl::ocl::get_context(static_cast<long>(ctx_id)));
viennacl::vector_base<T> A = viennacl::vector_base<T>(size_in, ctx);
// A = viennacl::zero_vector<T>(size_in, ctx);
// A = static_cast<viennacl::vector_base<T> >(A);
viennacl::linalg::vector_assign(A, (T)(0));
begin = 1;
last = size_in;
// ptr = &A;
viennacl::range temp_r(begin-1, last);
r = temp_r;
shared = false;
shared_type = 0;
shptr = std::make_shared<viennacl::vector_base<T> >(A);
}
mat Wavefunction::Jastrow_Gradient_Ratio(const mat &r, const int &number_particles, const double &beta)
{
mat Jastrow_Gradient_Radio = zeros(number_particles, dimension);
rowvec r_12_vec(3);
rowvec temp_r(3);
double r_12, arg, Spin_variable;
double bb;
for (uint i = 0; i < number_particles; i++)
{
for (uint j = 0; j < i; j++)
{
temp_r(0)=0;
temp_r(1) =0;
temp_r(2)=0;
r_12_vec = r.row(i) - r.row(j);
r_12 = 0;
for (uint k = 0; k < dimension; k++)
{
r_12 += r_12_vec(k) * r_12_vec(k);
}
r_12 = sqrt(r_12);
//Finner spin variablen
/*
if (((i < number_particles/2) && (j < number_particles/2)) || ((i > number_particles/2) && (j > number_particles/2)))
{
Spin_variable = 1.0/4;
}
else
{
Spin_variable = 1.0/2;
}
*/
Spin_variable = seta(i,j,number_particles);
bb = 1+beta*r_12;
//bb = number_atoms * bb; //number_atoms brukes av TEKNISKE grunner
arg = Spin_variable / (bb*bb);
Jastrow_Gradient_Radio.row(i) += arg * r_12_vec / r_12;
}
for (uint j = i+1; j < number_particles; j++)
{
r_12_vec = r.row(i) - r.row(j);
r_12 = 0;
for (uint k = 0; k < dimension; k++)
{
r_12 += r_12_vec(k) * r_12_vec(k);
}
r_12 = sqrt(r_12);
//Finner spin variablen
/*
if (((i < number_particles/2) && (j < number_particles/2)) || ((i > number_particles/2) && (j > number_particles/2)))
{
Spin_variable = 1.0/4;
}
else
{
Spin_variable = 1.0/2;
}
*/
Spin_variable = seta(i,j,number_particles);
bb = 1+beta*r_12;
//bb = number_atoms * bb; //number_atoms brukes av TEKNISKE grunner
arg = Spin_variable / (bb*bb);
Jastrow_Gradient_Radio.row(i) += r_12_vec*arg/r_12;
}
}
return Jastrow_Gradient_Radio;
}
