Phpml\FeatureSelection\VarianceThreshold

Complexity

Total Complexity

7

Size/Duplication

Total Lines	43
Duplicated Lines	0 %

Importance

Changes	1
Bugs	0	Features	0

3 Methods

Rating	Name	Duplication	Size	Complexity
A	__construct()	0	7	2
A	fit()	0	9	3
A	transform()	0	4	2

<?php

declare(strict_types=1);

namespace Phpml\FeatureSelection;

use Phpml\Exception\InvalidArgumentException;
use Phpml\Math\Matrix;
use Phpml\Math\Statistic\Variance;
use Phpml\Transformer;

final class VarianceThreshold implements Transformer
{
    /**
     * @var float
     */
    private $threshold;

    /**
     * @var array
     */
    private $variances = [];

    /**
     * @var array
     */
    private $keepColumns = [];

    public function __construct(float $threshold = 0.0)
    {
        if ($threshold < 0) {
            throw new InvalidArgumentException('Threshold can\'t be lower than zero');
        }

        $this->threshold = $threshold;
    }

    public function fit(array $samples, ?array $targets = null): void
    {
        $this->variances = array_map(static function (array $column): float {
            return Variance::population($column);
        }, Matrix::transposeArray($samples));

        foreach ($this->variances as $column => $variance) {
            if ($variance > $this->threshold) {
                $this->keepColumns[$column] = true;
            }
        }
    }

    public function transform(array &$samples, ?array &$targets = null): void
    {
        foreach ($samples as &$sample) {
            $sample = array_values(array_intersect_key($sample, $this->keepColumns));
        }
    }
}