public static function ex3()
{
header('Content-Type: application/json');
// Load files and parse them
$a = new SparseMatrix();
$b = new SparseMatrix();
$a->parseFile(self::A);
$b->parseFile(self::B);
$x = array();
$n = $a->Count();
for ($i = 0; $i <= $n; $i++) {
$x[$i - 1] = $i;
}
$res_aorix = self::multiplyMatrixWithVector($a->Matrix(), $x);
print_r($res_aorix);
//TODO: response json
echo json_encode(array("axb" => json_encode($res_aorix)));
}
/**
* Computes the beta vector from the given examples and labels. The
* examples are represented as a sparse matrix where each row is an example
* and each column a feature, and the labels as an array where each value
* is either 1 or -1, corresponding to a positive or negative example. Note
* that the first feature (column 0) corresponds to an intercept term, and
* is equal to 1 for every example.
*
* @param object $X SparseMatrix of training examples
* @param array $y example labels
*/
function train(SparseMatrix $X, $y)
{
$n = $X->columns();
$p = array_fill(0, $n, 0);
$a = array_fill(0, $n, 0);
$this->beta = array_fill(0, $n, 0.0);
$beta =& $this->beta;
foreach ($X as $i => $row) {
foreach ($row as $j => $Xij) {
if ($y[$i] == 1) {
$p[$j] += 1;
} else {
$a[$j] += 1;
}
}
}
$beta[0] = $this->logit($p[0], $a[0]);
for ($j = 1; $j < $n; $j++) {
$beta[$j] = $this->logit($p[$j], $a[$j]) - $beta[0];
}
}
/**
* Converts a SparseMatrix into an InvertedData instance. The data is
* duplicated.
*
* @param object $X SparseMatrix instance to convert
*/
function __construct(SparseMatrix $X)
{
$this->rows = $X->rows();
$this->columns = $X->columns();
$this->data = array();
$this->index = array();
foreach ($X as $i => $row) {
foreach ($row as $j => $Xij) {
$this->data[] = array($j, $i, $Xij);
}
}
sort($this->data);
$lastVar = -1;
foreach ($this->data as $dataOffset => $x) {
$currVar = $x[0];
if ($currVar != $lastVar) {
for ($var = $lastVar + 1; $var <= $currVar; $var++) {
$this->index[$var] = $dataOffset;
}
$lastVar = $currVar;
}
}
}
/**
* Given two sets of row indices, returns two new sparse matrices
* consisting of the corresponding rows.
*
* @param array $a_indices row indices for first new sparse matrix
* @param array $b_indices row indices for second new sparse matrix
* @return array array with two entries corresponding to the first and
* second new matrices
*/
function partition($a_indices, $b_indices)
{
$a = new SparseMatrix(count($a_indices), $this->n);
$b = new SparseMatrix(count($b_indices), $this->n);
$new_i = 0;
foreach ($a_indices as $i) {
$a->setRow($new_i++, $this->data[$i]);
}
$new_i = 0;
foreach ($b_indices as $i) {
$b->setRow($new_i++, $this->data[$i]);
}
return array($a, $b);
}
/**
* ex1 method
* @param void
* @return void
*/
public static function ex1()
{
header('Content-Type: application/json');
// Load files and parse them
define("TEST", 0);
$p = 7;
$kmax = 10000;
$epsilon = pow(10, -$p);
$epsilon2 = pow(10, 8);
$a1 = new SparseMatrix();
$a2 = new SparseMatrix();
$a3 = new SparseMatrix();
$a4 = new SparseMatrix();
$m = array();
$b = array();
$n = array();
$diagonal = array();
$L = array();
$U = array();
$B = array();
$C = array();
$a1->parseFile(self::RareM1);
if (!TEST) {
$a2->parseFile(self::RareM2);
$a3->parseFile(self::RareM3);
$a4->parseFile(self::RareM4);
}
// Initialization
$m[0] = $a1->Matrix();
if (!TEST) {
$m[1] = $a2->Matrix();
$m[2] = $a3->Matrix();
$m[3] = $a4->Matrix();
}
$b[0] = $a1->Vector();
if (!TEST) {
$b[1] = $a2->Vector();
$b[2] = $a3->Vector();
$b[3] = $a4->Vector();
}
$n[0] = count($a1);
if (!TEST) {
$n[1] = count($a2);
$n[2] = count($a3);
$n[3] = count($a4);
}
$xc = array();
$nlen = count($n);
//Check elements on the diagonal and build the diagonal matrix at the same time
for ($q = 0; $q < $nlen; $q++) {
for ($i = 0; $i < $n[$q]; $i++) {
$el = $m[$q][$i]->FindCol($i);
if (!$el) {
echo "Error, diagonal element null on the matrix " . $i;
die;
} else {
$diagonal[$q][$i] = new SinglyList();
$diagonal[$q][$i]->Append($el->Value(), $i);
}
}
}
$output = "";
// build upper and lower matrix
for ($q = 0; $q < $nlen; $q++) {
for ($i = 0; $i < $n[$q]; $i++) {
$crwA = $m[$q][$i]->Tail();
$U[$q][$i] = new SinglyList();
$L[$q][$i] = new SinglyList();
while ($crwA !== null) {
$j = $crwA->Column();
$output .= $j . '-' . $i . '|';
if ($j < $i) {
$U[$q][$i]->Append($crwA->Value(), $j);
} elseif ($j > $i) {
$L[$q][$i]->Append($crwA->Value(), $j);
}
$crwA = $crwA->Next();
}
}
}
$resp = array();
// build b & c matrixes
for ($q = 0; $q < $nlen; $q++) {
$B[$q] = self::plusMatrix($L[$q], self::multiplyMatrixWithScalar($diagonal[$q], round(5 / 6, $p)));
$C[$q] = self::substractMatrixes(self::multiplyMatrixWithScalar($diagonal[$q], -round(1 / 6, $p)), $U[$q]);
}
unset($L);
unset($U);
unset($d);
for ($q = 0; $q < $nlen; $q++) {
for ($i = 0; $i < $n[$q]; $i++) {
if (TEST) {
$xc[$q][0][$i] = $i + 1;
} else {
$xc[$q][0][$i] = 0;
}
}
$k = 0;
do {
$xc[$q][$k + 1] = array();
for ($i = 0; $i < $n[$q]; $i++) {
$sub = self::computeBSubstraction($m[$q], $i, $xc[$q][$k + 1], $xc[$q][$k]);
// cu plus primul x1 da ca in exemplu
if (TEST) {
$xc[$q][$k + 1][$i] = -0.2 * $xc[$q][$k][$i] + 1.2 * round(($b[$q][$i] + $sub) / $m[$q][$i]->FindCol($i)->Value(), $p);
} else {
$xc[$q][$k + 1][$i] = -0.2 * $xc[$q][$k][$i] + 1.2 * round(($b[$q][$i] - $sub) / $m[$q][$i]->FindCol($i)->Value(), $p);
}
}
if (TEST) {
print_r($xc);
}
$dx = self::getStandardNorm(self::subtractVectors($xc[$q][$k + 1], $xc[$q][$k], $n[$q], $p), $n[$q]);
$k++;
} while ($dx >= $epsilon && $k <= $kmax && $dx <= $epsilon2);
if ($dx < $epsilon) {
$resp[$q] = $xc[$q][$k - 1];
} else {
$resp[$q] = "divergenta";
}
}
$norm = array();
for ($q = 0; $q < $nlen; $q++) {
@($norm[$q] = self::subtractVectors(self::multiplyMatrixWithVector($m[$q], $resp[$q]), $b[$q], $n[$q], $p));
}
echo json_encode(array("resp" => $resp, "norm" => $norm));
die;
}