forked from jorgecasas/php-ml
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathPipelineTest.php
144 lines (116 loc) · 3.68 KB
/
PipelineTest.php
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
<?php
declare(strict_types=1);
namespace Phpml\Tests;
use Phpml\Classification\SVC;
use Phpml\FeatureExtraction\TfIdfTransformer;
use Phpml\FeatureExtraction\TokenCountVectorizer;
use Phpml\ModelManager;
use Phpml\Pipeline;
use Phpml\Preprocessing\Imputer;
use Phpml\Preprocessing\Imputer\Strategy\MostFrequentStrategy;
use Phpml\Preprocessing\Normalizer;
use Phpml\Regression\SVR;
use Phpml\Tokenization\WordTokenizer;
use PHPUnit\Framework\TestCase;
class PipelineTest extends TestCase
{
public function testPipelineConstruction(): void
{
$transformers = [
new TfIdfTransformer(),
];
$estimator = new SVC();
$pipeline = new Pipeline($transformers, $estimator);
$this->assertEquals($transformers, $pipeline->getTransformers());
$this->assertEquals($estimator, $pipeline->getEstimator());
}
public function testPipelineEstimatorSetter(): void
{
$pipeline = new Pipeline([new TfIdfTransformer()], new SVC());
$estimator = new SVR();
$pipeline->setEstimator($estimator);
$this->assertEquals($estimator, $pipeline->getEstimator());
}
public function testPipelineWorkflow(): void
{
$transformers = [
new Imputer(null, new MostFrequentStrategy()),
new Normalizer(),
];
$estimator = new SVC();
$samples = [
[1, -1, 2],
[2, 0, null],
[null, 1, -1],
];
$targets = [
4,
1,
4,
];
$pipeline = new Pipeline($transformers, $estimator);
$pipeline->train($samples, $targets);
$predicted = $pipeline->predict([[0, 0, 0]]);
$this->assertEquals(4, $predicted[0]);
}
public function testPipelineTransformers(): void
{
$transformers = [
new TokenCountVectorizer(new WordTokenizer()),
new TfIdfTransformer(),
];
$estimator = new SVC();
$samples = [
'Hello Paul',
'Hello Martin',
'Goodbye Tom',
'Hello John',
'Goodbye Alex',
'Bye Tony',
];
$targets = [
'greetings',
'greetings',
'farewell',
'greetings',
'farewell',
'farewell',
];
$pipeline = new Pipeline($transformers, $estimator);
$pipeline->train($samples, $targets);
$expected = ['greetings', 'farewell'];
$predicted = $pipeline->predict(['Hello Max', 'Goodbye Mark']);
$this->assertEquals($expected, $predicted);
}
public function testSaveAndRestore(): void
{
$pipeline = new Pipeline([
new TokenCountVectorizer(new WordTokenizer()),
new TfIdfTransformer(),
], new SVC());
$pipeline->train([
'Hello Paul',
'Hello Martin',
'Goodbye Tom',
'Hello John',
'Goodbye Alex',
'Bye Tony',
], [
'greetings',
'greetings',
'farewell',
'greetings',
'farewell',
'farewell',
]);
$testSamples = ['Hello Max', 'Goodbye Mark'];
$predicted = $pipeline->predict($testSamples);
$filepath = tempnam(sys_get_temp_dir(), uniqid('pipeline-test', true));
$modelManager = new ModelManager();
$modelManager->saveToFile($pipeline, $filepath);
$restoredClassifier = $modelManager->restoreFromFile($filepath);
$this->assertEquals($pipeline, $restoredClassifier);
$this->assertEquals($predicted, $restoredClassifier->predict($testSamples));
unlink($filepath);
}
}