/**
* 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;
}
}
}
