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
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
|
/*
* Copyright (c) 2009 Samit Basu
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 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 General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*
*/
#include "SparseCCS.hpp"
#include "Math.hpp"
#include "Algorithms.hpp"
// Normally, the i,j,v representation for a sparse matrix has
// a great degree of redundancy in one of the components. If
// j is the column coordinate then a vector such as:
// 1 3
// 3 3
// 5 3
// In: 1 1 2 2 3 3 3 3 4 5 5
// out:
QVector<uint32> CompressCCSCols(const QVector<uint32> &cols, index_t colcount) {
QVector<uint32> y(int(colcount+1));
// Scan through the columns
int mp=0;
for (unsigned col=1;col<=unsigned(colcount+1);col++) {
while ((mp < cols.size()) && (cols[mp] < col)) mp++;
y[col-1] = mp;
}
return y;
}
QVector<uint32> DecompressCCSCols(const QVector<uint32> &colstart, index_t colcount) {
QVector<uint32> x;
for (int p=1;p<colstart.size();p++) {
for (unsigned n=0;n<unsigned(colstart[p] - colstart[p-1]);n++)
x << p;
}
return x;
}
template <typename T>
static void TSparseMatToIJV(const SparseMatrix<T>&A, QVector<index_t> &rows,
QVector<index_t> &cols, QVector<T> &Adata) {
ConstSparseIterator<T> iter(&A);
while (iter.isValid()) {
cols << iter.col();
rows << iter.row();
Adata << iter.value();
iter.next();
}
}
template <typename T>
static void TSparseMatToIJV(const SparseMatrix<T> &Areal, const SparseMatrix<T> &Aimag,
QVector<index_t> &rows, QVector<index_t> &cols,
QVector<T> &Areal_part, QVector<T> &Aimag_part) {
ConstComplexSparseIterator<T> iter(&Areal,&Aimag);
while (iter.isValid()) {
cols << iter.col();
rows << iter.row();
Areal_part << iter.realValue();
Aimag_part << iter.imagValue();
iter.next();
}
}
template <typename T>
static Array TSparseToIJV(const Array &A, Array &rows, Array &cols) {
QVector<index_t> adata_rows;
QVector<index_t> adata_cols;
QVector<T> adata_real;
QVector<T> adata_imag;
if (A.allReal()) {
TSparseMatToIJV(A.constRealSparse<T>(),adata_rows,adata_cols,adata_real);
rows = Array(ToBasicArray(adata_rows));
cols = Array(ToBasicArray(adata_cols));
return Array(ToBasicArray(adata_real));
} else {
TSparseMatToIJV(A.constRealSparse<T>(),A.constImagSparse<T>(),
adata_rows,adata_cols,adata_real,adata_imag);
rows = Array(ToBasicArray(adata_rows));
cols = Array(ToBasicArray(adata_cols));
return Array(ToBasicArray(adata_real),ToBasicArray(adata_imag));
}
}
#define MacroSparseToIJV(ctype,cls) \
case cls: return TSparseToIJV<ctype>(a,rows,cols);
Array SparseToIJV(const Array &a, Array &rows, Array &cols) {
switch (a.dataClass()) {
default: throw Exception("unhandled case for sparse-to-ijv");
MacroExpandCases(MacroSparseToIJV);
}
}
#undef MacroSparseToIJV
// Array IJVForm::indices() {
// return Add(Multiply(Subtract(m_cols,Array(double(1))),
// Array(double(m_size_rows))),m_rows);
// }
template <typename T>
static SparseMatrix<T> TIJVToSparse(const BasicArray<index_t> &ip,
const BasicArray<index_t> &jp,
const BasicArray<T> &dp,
index_t rows, index_t cols) {
index_t ii = (ip.isScalar() == 1) ? 0 : 1;
index_t jj = (jp.isScalar() == 1) ? 0 : 1;
index_t dd = (dp.isScalar() == 1) ? 0 : 1;
SparseMatrix<T> retval(NTuple(rows,cols));
index_t len = qMax(ip.length(),qMax(jp.length(),dp.length()));
for (index_t i=1;i<=len;i++)
retval[NTuple(ip[(i-1)*ii+1],jp[(i-1)*jj+1])] = dp[(i-1)*dd+1];
return retval;
}
template <typename T>
static Array TIJVToSparse(const BasicArray<index_t> &ip,
const BasicArray<index_t> &jp,
const Array &dat, index_t rows,
index_t cols) {
if (dat.allReal())
return Array(TIJVToSparse(ip,jp,dat.constReal<T>(),rows,cols));
else
return Array(TIJVToSparse(ip,jp,dat.constReal<T>(),rows,cols),
TIJVToSparse(ip,jp,dat.constImag<T>(),rows,cols));
}
#define MacroIJVToSparse(ctype,cls) \
case cls: return TIJVToSparse<ctype>(ip,jp,dat,rows,cols);
Array IJVToSparse(const BasicArray<index_t> &ip,
const BasicArray<index_t> &jp,
const Array &dat,
index_t rows, index_t cols) {
switch (dat.dataClass()) {
default: throw Exception("unhandled type for ijv-to-sparse");
MacroExpandCases(MacroIJVToSparse);
}
}
Array IJVToSparse(const BasicArray<index_t> &ip,
const BasicArray<index_t> &jp,
const Array &dat) {
return IJVToSparse(ip,jp,dat.asDenseArray(),MaxValue(ip),MaxValue(jp));
}
|
