private function readCategorys($dataDir)
{
$data = FileMappedInputStream::_getIntArray($dataDir . "/char.category");
$size = count($data) / 4;
$ary = array();
for ($i = 0; $i < $size; $i++) {
$ary[$i] = new Category($data[$i * 4], $data[$i * 4 + 1], $data[$i * 4 + 2] === 1, $data[$i * 4 + 3] === 1);
}
return $ary;
}
public function __construct($dataDir)
{
$this->trie = new Searcher($dataDir . "/word2id");
$this->data = FileMappedInputStream::_getString($dataDir . "/word.dat");
$this->indices = FileMappedInputStream::_getIntArray($dataDir . "/word.ary.idx");
$fmis = new FileMappedInputStream($dataDir . "/word.inf");
$wordCount = $fmis->size() / (4 + 2 + 2 + 2);
$this->dataOffsets = $fmis->getIntArrayInstance($wordCount);
//単語の素性データの開始位置
$this->leftIds = $fmis->getShortArrayInstance($wordCount);
//単語の左文脈ID
$this->rightIds = $fmis->getShortArrayInstance($wordCount);
//単語の右文脈ID
$this->costs = $fmis->getShortArrayInstance($wordCount);
//単語のコスト
$fmis->close();
}
