Question

我正在使用Php-ai/Php-ml框架。在example they give中，AI仅使用了一个无用的功能，但在git主页上，他们还给出了使用多个功能的示例：

$samples = [[1, 3], [1, 4], [2, 4], [3, 1], [4, 1], [4, 2]];
$labels = ['a', 'a', 'a', 'b', 'b', 'b'];

$classifier = new KNearestNeighbors();
$classifier->train($samples, $labels);

echo $classifier->predict([3, 2]);

基于该示例仅提供一个功能，而第二示例提供了两个功能。我试图重新创建这个：

我正在尝试使用两个功能来重新创建它。我当前的代码段如下所示：

public function train(Request $request) {

    # CSV File
    $file = $request->file('dataframe');

    # Features + 1 will be the labels column
    $dataset = new CsvDataset($file, (int) $request->features);

    $vectorizer = new TokenCountVectorizer(new WordTokenizer());
    $tfIdfTransformer = new TfIdfTransformer();

    $finalSamples = [];
    for($i = 0; $i <= $request->features -1; $i++):
        $samples = [];
        foreach ($dataset->getSamples() as $sample)
            $samples[] = $sample[$i];

        $vectorizer->fit($samples);
        $vectorizer->transform($samples);

        $tfIdfTransformer->fit($samples);
        $tfIdfTransformer->transform($samples);

        $finalSamples[] = $samples;
    endfor;

    # This gives us an output of Array[ 0 => [Feature 1, Feature 2], 1 => [Feature 1, Feature 2], ... ] like shown on example two.

    $result = [];
    foreach($finalSamples as $arr)
        foreach($arr as $k => $v)
            $result[$k][] = $v;

    $dataset = new ArrayDataset($result, $dataset->getTargets());
    $randomSplit = new StratifiedRandomSplit($dataset, 0.1);

    $classifier = new SVC(Kernel::RBF, 10000);

    # Train with half of the data frame
    $classifier->train($randomSplit->getTrainSamples(), $randomSplit->getTrainLabels());

    $predictedLabels = $classifier->predict($randomSplit->getTestSamples());
    $inputLabels     = $randomSplit->getTestLabels();
}

我的CSV文件如下：

"SibSp","Parch","Survived",
"1", "1", "1",
"3", "3", "1",
"4", "1", "0"
"4", "0", "1",
"5", "2", "0"
"3", "1", "0",
"2", "2", "1",
"0", "0", "1"

现在的问题是当我可视化数据时，我喜欢这样做：

$newDataFrame   = [];
$incorrect      = 0;

for($i = 0; $i <= count($inputLabels) -1; $i++):
    $newDataFrame[] = (object) ['input' => $inputLabels[$i], 'output' => $predictedLabels[$i]];
    if($inputLabels[$i] != $predictedLabels[$i]) $incorrect++;
ndfor;

$correct = count($inputLabels) - $incorrect;
$score   = round((float)Accuracy::score(isset($request->train) ? $randomSplit->getTestLabels() : $inputLabels, $predictedLabels) * 100 );

数据总是以1个正确，1个不正确的形式出现，得分为50（％）。

我如何使用此分类器来使用多个功能，而不仅仅是一个？我认为问题出在构建ArrayDataSet时，但是我不知道它有什么问题。

如何在PHP的Php-ai / Php-ml中使用多个功能进行预测

0 个答案: