VarianceThreshold::fit() - Code Metrics - Inspection of "Implement VarianceThreshold - simple baseline appr..." - php-ai/php-ml - Measure and Improve Code Quality continuously with Scrutinizer

Passed

Pull Request — master (#228)

by Arkadiusz

created 2018-02-10 16:37 UTC

VarianceThreshold::fit() A

↳ Parent: VarianceThreshold

Complexity

Conditions	3
Paths	3

Size

Total Lines	12
Code Lines	7

Duplication

Lines	0
Ratio	0 %

Importance

Changes

Metric	Value
dl	0
loc	12
rs	9.4285
c	0
b	0
f	0
cc	3
eloc	7
nc	3
nop	1

<?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;
        $this->variances = [];
        $this->keepColumns = [];
    }

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

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

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


1			<?php
2
3			declare(strict_types=1);
4
5			namespace Phpml\FeatureSelection;
6
7			use Phpml\Exception\InvalidArgumentException;
8			use Phpml\Math\Matrix;
9			use Phpml\Math\Statistic\Variance;
10			use Phpml\Transformer;
11
12			final class VarianceThreshold implements Transformer
13			{
14			/**
15			* @var float
16			*/
17			private $threshold;
18
19			/**
20			* @var array
21			*/
22			private $variances = [];
23
24			/**
25			* @var array
26			*/
27			private $keepColumns = [];
28
29			public function __construct(float $threshold = 0.0)
30			{
31			if ($threshold < 0) {
32			throw new InvalidArgumentException('Threshold can\'t be lower than zero');
33			}
34
35			$this->threshold = $threshold;
36			$this->variances = [];
37			$this->keepColumns = [];
38			}
39
40			public function fit(array $samples): void
41			{
42			$this->variances = array_map(function (array $column) {
43			return Variance::population($column);
44			}, Matrix::transposeArray($samples));
45
46			foreach ($this->variances as $column => $variance) {
47			if ($variance > $this->threshold) {
48			$this->keepColumns[$column] = true;
49			}
50			}
51			}
52
53			public function transform(array &$samples): void
54			{
55			foreach ($samples as &$sample) {
56			$sample = array_values(array_intersect_key($sample, $this->keepColumns));
57			}
58			}
59			}
60

php-ai / php-ml

Pull Request — master (#228)

VarianceThreshold::fit() A

Complexity

Size

Duplication

Importance

Duplication Side-by-Side

Filter issues like