mirrors/z3
3
0
Fork 0
mirror of https://github.com/Z3Prover/z3 synced 2025-11-03 21:09:11 +00:00
Code Activity

Sparse matrix kernel (#6035)

Browse source 
* Subtle bug in kernel computation

Coefficient was being passed by reference and, therefore, was
being changed indirectly.

In the process, updated the code to be more generic to avoid rational
computation in the middle of matrix manipulation.

* sparse_matrix: fixed handling of 0 in add_var() and add()

particularly in add_var(), without the fix the user is responsible for checking
coefficients for 0.
This commit is contained in:
Arie Gurfinkel 2022-05-13 17:30:35 -07:00 • committed by GitHub
parent 6f7be77e2b
commit 7f62fa2b66
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
2 changed files with 57 additions and 51 deletions
Download patch file Download diff file Expand all files Collapse all files

106
src/math/simplex/sparse_matrix_ops.h
View file

@ -7,7 +7,7 @@ Module Name:
Abstract:
Author:
Nikolaj Bjorner (nbjorner) 2014-01-15
@ -23,59 +23,63 @@ Notes:
namespace simplex {
class sparse_matrix_ops {
public:
static void kernel(sparse_matrix<mpq_ext>& M, vector<vector<rational>>& K) {
rational D;
vector<unsigned> d, c;
unsigned n = M.num_vars(), m = M.num_rows();
auto& mgr = M.get_manager();
c.resize(m, 0u);
d.resize(n, 0u);
class sparse_matrix_ops {
public:
template <typename Ext>
static void kernel(sparse_matrix<Ext> &M, vector<vector<rational>> &K) {
using scoped_numeral = typename Ext::scoped_numeral;
for (unsigned k = 0; k < n; ++k) {
d[k] = 0;
for (auto [row, row_entry] : M.get_rows(k)) {
if (c[row.id()] != 0)
continue;
auto& m_jk = row_entry->m_coeff;
if (mpq_manager<false>::is_zero(m_jk))
continue;
D = rational(-1) / m_jk;
M.mul(row, D.to_mpq());
for (auto [row_i, row_i_entry] : M.get_rows(k)) {
if (row_i.id() == row.id())
continue;
auto& m_ik = row_i_entry->m_coeff;
// row_i += m_ik * row
M.add(row_i, m_ik, row);
}
c[row.id()] = k + 1;
d[k] = row.id() + 1;
break;
vector<unsigned> d, c;
unsigned n_vars = M.num_vars(), n_rows = M.num_rows();
c.resize(n_rows, 0u);
d.resize(n_vars, 0u);
auto &m = M.get_manager();
scoped_numeral m_ik(m);
scoped_numeral D(m);
for (unsigned k = 0; k < n_vars; ++k) {
d[k] = 0;
for (auto [row, row_entry] : M.get_rows(k)) {
if (c[row.id()] != 0) continue;
auto &m_jk = row_entry->m_coeff;
if (mpq_manager<false>::is_zero(m_jk)) continue;
// D = rational(-1) / m_jk;
m.set(D, m_jk);
m.inv(D);
m.neg(D);
M.mul(row, D);
for (auto [row_i, row_i_entry] : M.get_rows(k)) {
if (row_i.id() == row.id()) continue;
m.set(m_ik, row_i_entry->m_coeff);
// row_i += m_ik * row
M.add(row_i, m_ik, row);
}
c[row.id()] = k + 1;
d[k] = row.id() + 1;
break;
}
for (unsigned k = 0; k < n; ++k) {
if (d[k] != 0)
continue;
K.push_back(vector<rational>());
for (unsigned i = 0; i < n; ++i) {
if (d[i] > 0) {
auto r = sparse_matrix<mpq_ext>::row(d[i]-1);
K.back().push_back(rational(M.get_coeff(r, k)));
}
else if (i == k)
K.back().push_back(rational(1));
else
K.back().push_back(rational(0));
}
}
}
};
}
for (unsigned k = 0; k < n_vars; ++k) {
if (d[k] != 0) continue;
K.push_back(vector<rational>());
for (unsigned i = 0; i < n_vars; ++i) {
if (d[i] > 0) {
auto r = sparse_matrix<mpq_ext>::row(d[i] - 1);
K.back().push_back(rational(M.get_coeff(r, k)));
} else if (i == k)
K.back().push_back(rational(1));
else
K.back().push_back(rational(0));
}
}
}
static void kernel(sparse_matrix<mpq_ext> &M, vector<vector<rational>> &K) {
kernel<mpq_ext>(M, K);
}
};
} // namespace simplex