-
Notifications
You must be signed in to change notification settings - Fork 299
/
ConstantSparsityProjectionMethod.inl
124 lines (101 loc) · 4.77 KB
/
ConstantSparsityProjectionMethod.inl
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
/******************************************************************************
* SOFA, Simulation Open-Framework Architecture *
* (c) 2006 INRIA, USTL, UJF, CNRS, MGH *
* *
* This program is free software; you can redistribute it and/or modify it *
* under the terms of the GNU Lesser General Public License as published by *
* the Free Software Foundation; either version 2.1 of the License, or (at *
* your option) any later version. *
* *
* This program is distributed in the hope that it will be useful, but WITHOUT *
* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or *
* FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License *
* for more details. *
* *
* You should have received a copy of the GNU Lesser General Public License *
* along with this program. If not, see <http://www.gnu.org/licenses/>. *
*******************************************************************************
* Authors: The SOFA Team and external contributors (see Authors.txt) *
* *
* Contact information: contact@sofa-framework.org *
******************************************************************************/
#pragma once
#include <sofa/component/linearsystem/ConstantSparsityProjectionMethod.h>
#include <Eigen/Sparse>
#include <sofa/helper/ScopedAdvancedTimer.h>
#include <sofa/simulation/MainTaskSchedulerFactory.h>
#include <sofa/simulation/ParallelSparseMatrixProduct.h>
namespace sofa::component::linearsystem
{
template <class TMatrix>
ConstantSparsityProjectionMethod<TMatrix>::ConstantSparsityProjectionMethod()
: d_parallelProduct(initData(&d_parallelProduct, true, "parallelProduct", "Compute the matrix product in parallel"))
{}
template <class TMatrix>
ConstantSparsityProjectionMethod<TMatrix>::
~ConstantSparsityProjectionMethod() = default;
template <class TMatrix>
void ConstantSparsityProjectionMethod<TMatrix>::init()
{
Inherit1::init();
if (d_parallelProduct.getValue())
{
auto* taskScheduler = simulation::MainTaskSchedulerFactory::createInRegistry();
taskScheduler->init();
auto* matrixPrductKJ = new sofa::simulation::ParallelSparseMatrixProduct<
K_Type, J_Type, KJ_Type>();
m_matrixProductKJ = std::unique_ptr<sofa::simulation::ParallelSparseMatrixProduct<
K_Type, J_Type, KJ_Type>>(matrixPrductKJ);
matrixPrductKJ->taskScheduler = taskScheduler;
auto* matrixPrductJTKJ = new sofa::simulation::ParallelSparseMatrixProduct<
JT_Type, KJ_Type, JTKJ_Type>();
m_matrixProductJTKJ = std::unique_ptr<sofa::simulation::ParallelSparseMatrixProduct<
JT_Type, KJ_Type, JTKJ_Type>>(matrixPrductJTKJ);
matrixPrductJTKJ->taskScheduler = taskScheduler;
}
else
{
m_matrixProductKJ = std::make_unique<sofa::linearalgebra::SparseMatrixProduct<
K_Type, J_Type, KJ_Type>>();
m_matrixProductJTKJ = std::make_unique<sofa::linearalgebra::SparseMatrixProduct<
JT_Type, KJ_Type, JTKJ_Type>>();
}
}
template <class TMatrix>
void ConstantSparsityProjectionMethod<TMatrix>::computeProjection(
const Eigen::Map<Eigen::SparseMatrix<Block, Eigen::RowMajor>> KMap,
const sofa::type::fixed_array<std::shared_ptr<TMatrix>, 2> J,
Eigen::SparseMatrix<Block, Eigen::RowMajor>& JT_K_J)
{
if (J[0] && J[1])
{
const auto JMap0 = this->makeEigenMap(*J[0]);
const auto JMap1 = this->makeEigenMap(*J[1]);
m_matrixProductKJ->m_lhs = &KMap;
m_matrixProductKJ->m_rhs = &JMap1;
m_matrixProductKJ->computeProduct();
const JT_Type JMap0T = JMap0.transpose();
m_matrixProductJTKJ->m_lhs = &JMap0T;
m_matrixProductJTKJ->m_rhs = &m_matrixProductKJ->getProductResult();
m_matrixProductJTKJ->computeProduct();
JT_K_J = m_matrixProductJTKJ->getProductResult();
}
else if (J[0] && !J[1])
{
const auto JMap0 = this->makeEigenMap(*J[0]);
JT_K_J = JMap0.transpose() * KMap;
}
else if (!J[0] && J[1])
{
const auto JMap1 = this->makeEigenMap(*J[1]);
m_matrixProductKJ->m_lhs = &KMap;
m_matrixProductKJ->m_rhs = &JMap1;
m_matrixProductKJ->computeProduct();
JT_K_J = m_matrixProductKJ->getProductResult();
}
else
{
JT_K_J = KMap;
}
}
}