First Commit

vendor/php-ai/php-ml/src/Dataset/ArrayDataset.php

@@ -0,0 +1,62 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Phpml\Dataset;
+
+use Phpml\Exception\InvalidArgumentException;
+
+class ArrayDataset implements Dataset
+{
+    /**
+     * @var array
+     */
+    protected $samples = [];
+
+    /**
+     * @var array
+     */
+    protected $targets = [];
+
+    /**
+     * @throws InvalidArgumentException
+     */
+    public function __construct(array $samples, array $targets)
+    {
+        if (count($samples) !== count($targets)) {
+            throw new InvalidArgumentException('Size of given arrays does not match');
+        }
+
+        $this->samples = $samples;
+        $this->targets = $targets;
+    }
+
+    public function getSamples(): array
+    {
+        return $this->samples;
+    }
+
+    public function getTargets(): array
+    {
+        return $this->targets;
+    }
+
+    /**
+     * @param int[] $columns
+     */
+    public function removeColumns(array $columns): void
+    {
+        foreach ($this->samples as &$sample) {
+            $this->removeColumnsFromSample($sample, $columns);
+        }
+    }
+
+    private function removeColumnsFromSample(array &$sample, array $columns): void
+    {
+        foreach ($columns as $index) {
+            unset($sample[$index]);
+        }
+
+        $sample = array_values($sample);
+    }
+}

vendor/php-ai/php-ml/src/Dataset/CsvDataset.php

@@ -0,0 +1,52 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Phpml\Dataset;
+
+use Phpml\Exception\FileException;
+
+class CsvDataset extends ArrayDataset
+{
+    /**
+     * @var array
+     */
+    protected $columnNames = [];
+
+    /**
+     * @throws FileException
+     */
+    public function __construct(string $filepath, int $features, bool $headingRow = true, string $delimiter = ',', int $maxLineLength = 0)
+    {
+        if (!file_exists($filepath)) {
+            throw new FileException(sprintf('File "%s" missing.', basename($filepath)));
+        }
+
+        $handle = fopen($filepath, 'rb');
+        if ($handle === false) {
+            throw new FileException(sprintf('File "%s" can\'t be open.', basename($filepath)));
+        }
+
+        if ($headingRow) {
+            $data = fgetcsv($handle, $maxLineLength, $delimiter);
+            $this->columnNames = array_slice((array) $data, 0, $features);
+        } else {
+            $this->columnNames = range(0, $features - 1);
+        }
+
+        $samples = $targets = [];
+        while ($data = fgetcsv($handle, $maxLineLength, $delimiter)) {
+            $samples[] = array_slice($data, 0, $features);
+            $targets[] = $data[$features];
+        }
+
+        fclose($handle);
+
+        parent::__construct($samples, $targets);
+    }
+
+    public function getColumnNames(): array
+    {
+        return $this->columnNames;
+    }
+}

vendor/php-ai/php-ml/src/Dataset/Dataset.php

@@ -0,0 +1,12 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Phpml\Dataset;
+
+interface Dataset
+{
+    public function getSamples(): array;
+
+    public function getTargets(): array;
+}

vendor/php-ai/php-ml/src/Dataset/Demo/GlassDataset.php

@@ -0,0 +1,28 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Phpml\Dataset\Demo;
+
+use Phpml\Dataset\CsvDataset;
+
+/**
+ * Classes: 6
+ * Samples per class:
+ *      70 float processed building windows
+ *      17 float processed vehicle windows
+ *      76 non-float processed building windows
+ *      13 containers
+ *      9 tableware
+ *      29 headlamps
+ * Samples total: 214
+ * Features per sample: 9.
+ */
+class GlassDataset extends CsvDataset
+{
+    public function __construct()
+    {
+        $filepath = __DIR__.'/../../../data/glass.csv';
+        parent::__construct($filepath, 9, true);
+    }
+}

vendor/php-ai/php-ml/src/Dataset/Demo/IrisDataset.php

@@ -0,0 +1,22 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Phpml\Dataset\Demo;
+
+use Phpml\Dataset\CsvDataset;
+
+/**
+ * Classes: 3
+ * Samples per class: 50
+ * Samples total: 150
+ * Features per sample: 4.
+ */
+class IrisDataset extends CsvDataset
+{
+    public function __construct()
+    {
+        $filepath = __DIR__.'/../../../data/iris.csv';
+        parent::__construct($filepath, 4, true);
+    }
+}

vendor/php-ai/php-ml/src/Dataset/Demo/WineDataset.php

@@ -0,0 +1,22 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Phpml\Dataset\Demo;
+
+use Phpml\Dataset\CsvDataset;
+
+/**
+ * Classes: 3
+ * Samples per class: class 1 59; class 2 71; class 3 48
+ * Samples total: 178
+ * Features per sample: 13.
+ */
+class WineDataset extends CsvDataset
+{
+    public function __construct()
+    {
+        $filepath = __DIR__.'/../../../data/wine.csv';
+        parent::__construct($filepath, 13, true);
+    }
+}

vendor/php-ai/php-ml/src/Dataset/FilesDataset.php

@@ -0,0 +1,47 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Phpml\Dataset;
+
+use Phpml\Exception\DatasetException;
+
+class FilesDataset extends ArrayDataset
+{
+    public function __construct(string $rootPath)
+    {
+        if (!is_dir($rootPath)) {
+            throw new DatasetException(sprintf('Dataset root folder "%s" missing.', $rootPath));
+        }
+
+        $this->scanRootPath($rootPath);
+    }
+
+    private function scanRootPath(string $rootPath): void
+    {
+        $dirs = glob($rootPath.DIRECTORY_SEPARATOR.'*', GLOB_ONLYDIR);
+
+        if ($dirs === false) {
+            throw new DatasetException(sprintf('An error occurred during directory "%s" scan', $rootPath));
+        }
+
+        foreach ($dirs as $dir) {
+            $this->scanDir($dir);
+        }
+    }
+
+    private function scanDir(string $dir): void
+    {
+        $target = basename($dir);
+
+        $files = glob($dir.DIRECTORY_SEPARATOR.'*');
+        if ($files === false) {
+            return;
+        }
+
+        foreach (array_filter($files, 'is_file') as $file) {
+            $this->samples[] = file_get_contents($file);
+            $this->targets[] = $target;
+        }
+    }
+}

vendor/php-ai/php-ml/src/Dataset/MnistDataset.php

@@ -0,0 +1,101 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Phpml\Dataset;
+
+use Phpml\Exception\InvalidArgumentException;
+
+/**
+ * MNIST dataset: http://yann.lecun.com/exdb/mnist/
+ * original mnist dataset reader: https://github.com/AndrewCarterUK/mnist-neural-network-plain-php
+ */
+final class MnistDataset extends ArrayDataset
+{
+    private const MAGIC_IMAGE = 0x00000803;
+
+    private const MAGIC_LABEL = 0x00000801;
+
+    private const IMAGE_ROWS = 28;
+
+    private const IMAGE_COLS = 28;
+
+    public function __construct(string $imagePath, string $labelPath)
+    {
+        $this->samples = $this->readImages($imagePath);
+        $this->targets = $this->readLabels($labelPath);
+
+        if (count($this->samples) !== count($this->targets)) {
+            throw new InvalidArgumentException('Must have the same number of images and labels');
+        }
+    }
+
+    private function readImages(string $imagePath): array
+    {
+        $stream = fopen($imagePath, 'rb');
+
+        if ($stream === false) {
+            throw new InvalidArgumentException('Could not open file: '.$imagePath);
+        }
+
+        $images = [];
+
+        try {
+            $header = fread($stream, 16);
+
+            $fields = unpack('Nmagic/Nsize/Nrows/Ncols', (string) $header);
+
+            if ($fields['magic'] !== self::MAGIC_IMAGE) {
+                throw new InvalidArgumentException('Invalid magic number: '.$imagePath);
+            }
+
+            if ($fields['rows'] != self::IMAGE_ROWS) {
+                throw new InvalidArgumentException('Invalid number of image rows: '.$imagePath);
+            }
+
+            if ($fields['cols'] != self::IMAGE_COLS) {
+                throw new InvalidArgumentException('Invalid number of image cols: '.$imagePath);
+            }
+
+            for ($i = 0; $i < $fields['size']; $i++) {
+                $imageBytes = fread($stream, $fields['rows'] * $fields['cols']);
+
+                // Convert to float between 0 and 1
+                $images[] = array_map(function ($b) {
+                    return $b / 255;
+                }, array_values(unpack('C*', (string) $imageBytes)));
+            }
+        } finally {
+            fclose($stream);
+        }
+
+        return $images;
+    }
+
+    private function readLabels(string $labelPath): array
+    {
+        $stream = fopen($labelPath, 'rb');
+
+        if ($stream === false) {
+            throw new InvalidArgumentException('Could not open file: '.$labelPath);
+        }
+
+        $labels = [];
+
+        try {
+            $header = fread($stream, 8);
+
+            $fields = unpack('Nmagic/Nsize', (string) $header);
+
+            if ($fields['magic'] !== self::MAGIC_LABEL) {
+                throw new InvalidArgumentException('Invalid magic number: '.$labelPath);
+            }
+
+            $labels = fread($stream, $fields['size']);
+        } finally {
+            fclose($stream);
+        }
+
+        return array_values(unpack('C*', (string) $labels));
+    }
+}

vendor/php-ai/php-ml/src/Dataset/SvmDataset.php

@@ -0,0 +1,131 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Phpml\Dataset;
+
+use Phpml\Exception\DatasetException;
+use Phpml\Exception\FileException;
+
+class SvmDataset extends ArrayDataset
+{
+    public function __construct(string $filePath)
+    {
+        [$samples, $targets] = self::readProblem($filePath);
+
+        parent::__construct($samples, $targets);
+    }
+
+    private static function readProblem(string $filePath): array
+    {
+        $handle = self::openFile($filePath);
+
+        $samples = [];
+        $targets = [];
+        $maxIndex = 0;
+        while (false !== $line = fgets($handle)) {
+            [$sample, $target, $maxIndex] = self::processLine($line, $maxIndex);
+            $samples[] = $sample;
+            $targets[] = $target;
+        }
+
+        fclose($handle);
+
+        foreach ($samples as &$sample) {
+            $sample = array_pad($sample, $maxIndex + 1, 0);
+        }
+
+        return [$samples, $targets];
+    }
+
+    /**
+     * @return resource
+     */
+    private static function openFile(string $filePath)
+    {
+        if (!file_exists($filePath)) {
+            throw new FileException(sprintf('File "%s" missing.', basename($filePath)));
+        }
+
+        $handle = fopen($filePath, 'rb');
+        if ($handle === false) {
+            throw new FileException(sprintf('File "%s" can\'t be open.', basename($filePath)));
+        }
+
+        return $handle;
+    }
+
+    private static function processLine(string $line, int $maxIndex): array
+    {
+        $columns = self::parseLine($line);
+
+        $target = self::parseTargetColumn($columns[0]);
+        $sample = array_fill(0, $maxIndex + 1, 0);
+
+        $n = count($columns);
+        for ($i = 1; $i < $n; ++$i) {
+            [$index, $value] = self::parseFeatureColumn($columns[$i]);
+            if ($index > $maxIndex) {
+                $maxIndex = $index;
+                $sample = array_pad($sample, $maxIndex + 1, 0);
+            }
+
+            $sample[$index] = $value;
+        }
+
+        return [$sample, $target, $maxIndex];
+    }
+
+    private static function parseLine(string $line): array
+    {
+        $line = explode('#', $line, 2)[0];
+        $line = rtrim($line);
+        $line = str_replace("\t", ' ', $line);
+
+        return explode(' ', $line);
+    }
+
+    private static function parseTargetColumn(string $column): float
+    {
+        if (!is_numeric($column)) {
+            throw new DatasetException(sprintf('Invalid target "%s".', $column));
+        }
+
+        return (float) $column;
+    }
+
+    private static function parseFeatureColumn(string $column): array
+    {
+        $feature = explode(':', $column, 2);
+        if (count($feature) !== 2) {
+            throw new DatasetException(sprintf('Invalid value "%s".', $column));
+        }
+
+        $index = self::parseFeatureIndex($feature[0]);
+        $value = self::parseFeatureValue($feature[1]);
+
+        return [$index, $value];
+    }
+
+    private static function parseFeatureIndex(string $index): int
+    {
+        if (!is_numeric($index) || !ctype_digit($index)) {
+            throw new DatasetException(sprintf('Invalid index "%s".', $index));
+        }
+
+        if ((int) $index < 1) {
+            throw new DatasetException(sprintf('Invalid index "%s".', $index));
+        }
+
+        return (int) $index - 1;
+    }
+
+    private static function parseFeatureValue(string $value): float
+    {
+        if (!is_numeric($value)) {
+            throw new DatasetException(sprintf('Invalid value "%s".', $value));
+        }
+
+        return (float) $value;
+    }
+}