Word Order Randomizer Transformer.

docs/transformers/word-randomizer.md

@@ -0,0 +1,22 @@
+<span style="float:right;"><a href="https://github.com/RubixML/ML/blob/master/src/Transformers/WordOrderRandomizer.php">[source]</a></span>
+
+# Word Randomizer
+Splits the given text based on a separator and then shuffles it.
+
+**Interfaces:** [Transformer](api.md#transformer)
+
+**Data Type Compatibility:** Categorical
+
+## Parameters
+| # | Name      | Default | Type   | Description                                                     |
+|---|-----------|---------|--------|-----------------------------------------------------------------|
+| 1 | separator | ' '     | string | Should the transformer split the string based on ' ' character? |
+
+## Example
+```php
+use Rubix\ML\Transformers\WordOrderRandomizer;
+$transformer = new WordOrderRandomizer();
+```
+
+## Additional Methods
+This transformer does not have any additional methods.

src/Transformers/WordOrderRandomizer.php

@@ -0,0 +1,85 @@
+<?php
+
+namespace Rubix\ML\Transformers;
+
+use Rubix\ML\DataType;
+
+use function is_string;
+use function array_walk;
+
+/**
+ * Word Order Randomizer
+ *
+ * This transformer shuffles the words in a given string.
+ *
+ * @category    Machine Learning
+ * @package     Rubix/ML
+ * @author      Stylianos Tzourelis
+ */
+class WordOrderRandomizer implements Transformer
+{
+    /**
+     * The separator to split the string with.
+     *
+     * @var string
+     */
+    protected string $separator;
+
+    /**
+     * @param string $separator
+     */
+    public function __construct(string $separator = ' ')
+    {
+        $this->separator = $separator;
+    }
+
+    /**
+     * Return the data types that this transformer is compatible with.
+     *
+     * @internal
+     *
+     * @return list<\Rubix\ML\DataType>
+     */
+    public function compatibility() : array
+    {
+        return DataType::all();
+    }
+
+    /**
+     * Transform the dataset in place.
+     *
+     * @param array<mixed[]> $samples
+     */
+    public function transform(array &$samples) : void
+    {
+        array_walk($samples, [$this, 'randomize']);
+    }
+
+    /**
+     * Randomize the text in a sample.
+     *
+     * @param list<mixed> $sample
+     */
+    private function randomize(array &$sample) : void
+    {
+        foreach ($sample as &$value) {
+            if (is_string($value) && !empty($this->separator)) {
+                $value = explode($this->separator, $value);
+                shuffle($value);
+                $value = implode($this->separator, $value);
+            }
+        }
+    }
+
+    /**
+     * Return the string representation of the object.
+     *
+     * @internal
+     *
+     * @return string
+     */
+    public function __toString() : string
+    {
+        return 'Word Order Randomizer';
+    }
+}

tests/Transformers/WordOrderRandomizerTest.php

@@ -0,0 +1,61 @@
+<?php
+
+namespace Rubix\ML\Tests\Transformers;
+
+use Rubix\ML\Datasets\Unlabeled;
+use Rubix\ML\Transformers\Transformer;
+use PHPUnit\Framework\TestCase;
+use Rubix\ML\Transformers\WordOrderRandomizer;
+
+/**
+ * @group Transformers
+ * @covers \Rubix\ML\Transformers\WordOrderRandomizer
+ */
+class WordOrderRandomizerTest extends TestCase
+{
+    /**
+     * @var \Rubix\ML\Datasets\Unlabeled
+     */
+    protected $dataset;
+
+    /**
+     * @var \Rubix\ML\Transformers\WordOrderRandomizer
+     */
+    protected $transformer;
+
+    /**
+     * @before
+     */
+    protected function setUp() : void
+    {
+        $this->dataset = Unlabeled::quick([
+            ['Red dining chair.'],
+            ['Blue,cotton,pillow'],
+        ]);
+
+        $this->transformer = new WordOrderRandomizer();
+    }
+
+    /**
+     * @test
+     */
+    public function build() : void
+    {
+        $this->assertInstanceOf(WordOrderRandomizer::class, $this->transformer);
+        $this->assertInstanceOf(Transformer::class, $this->transformer);
+    }
+
+    /**
+     * @test
+     */
+    public function transform() : void
+    {
+        $this->dataset->apply($this->transformer);
+
+        foreach (explode(' ', 'Red dining chair.') as $word) {
+            $this->assertTrue(str_contains($this->dataset->samples()[0][0], $word));
+        }
+
+        $this->assertEquals(['Blue,cotton,pillow'], $this->dataset->samples()[1]);
+    }
+}