1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
declare(strict_types=1); |
4
|
|
|
|
5
|
|
|
namespace Phpml\Classification\Ensemble; |
6
|
|
|
|
7
|
|
|
use Phpml\Classification\Ensemble\Bagging; |
8
|
|
|
use Phpml\Classification\DecisionTree; |
9
|
|
|
use Phpml\Classification\NaiveBayes; |
10
|
|
|
use Phpml\Classification\Classifier; |
11
|
|
|
|
12
|
|
|
class RandomForest extends Bagging |
13
|
|
|
{ |
14
|
|
|
/** |
15
|
|
|
* @var float|string |
16
|
|
|
*/ |
17
|
|
|
protected $featureSubsetRatio = 'log'; |
18
|
|
|
|
19
|
|
|
/** |
20
|
|
|
* @var array |
21
|
|
|
*/ |
22
|
|
|
protected $columnNames = null; |
23
|
|
|
|
24
|
|
|
/** |
25
|
|
|
* Initializes RandomForest with the given number of trees. More trees |
26
|
|
|
* may increase the prediction performance while it will also substantially |
27
|
|
|
* increase the processing time and the required memory |
28
|
|
|
* |
29
|
|
|
* @param type $numClassifier |
30
|
|
|
*/ |
31
|
|
|
public function __construct($numClassifier = 50) |
32
|
|
|
{ |
33
|
|
|
parent::__construct($numClassifier); |
34
|
|
|
|
35
|
|
|
$this->setSubsetRatio(1.0); |
36
|
|
|
} |
37
|
|
|
|
38
|
|
|
/** |
39
|
|
|
* This method is used to determine how many of the original columns (features) |
40
|
|
|
* will be used to construct subsets to train base classifiers.<br> |
41
|
|
|
* |
42
|
|
|
* Allowed values: 'sqrt', 'log' or any float number between 0.1 and 1.0 <br> |
43
|
|
|
* |
44
|
|
|
* Default value for the ratio is 'log' which results in log(numFeatures, 2) + 1 |
45
|
|
|
* features to be taken into consideration while selecting subspace of features |
46
|
|
|
* |
47
|
|
|
* @param mixed $ratio string or float should be given |
48
|
|
|
* @return $this |
49
|
|
|
* @throws Exception |
50
|
|
|
*/ |
51
|
|
|
public function setFeatureSubsetRatio($ratio) |
52
|
|
|
{ |
53
|
|
|
if (is_float($ratio) && ($ratio < 0.1 || $ratio > 1.0)) { |
54
|
|
|
throw new \Exception("When a float given, feature subset ratio should be between 0.1 and 1.0"); |
55
|
|
|
} |
56
|
|
|
if (is_string($ratio) && $ratio != 'sqrt' && $ratio != 'log') { |
57
|
|
|
throw new \Exception("When a string given, feature subset ratio can only be 'sqrt' or 'log' "); |
58
|
|
|
} |
59
|
|
|
$this->featureSubsetRatio = $ratio; |
60
|
|
|
return $this; |
61
|
|
|
} |
62
|
|
|
|
63
|
|
|
/** |
64
|
|
|
* RandomForest algorithm is usable *only* with DecisionTree |
65
|
|
|
* |
66
|
|
|
* @param string $classifier |
67
|
|
|
* @param array $classifierOptions |
68
|
|
|
* @return $this |
69
|
|
|
*/ |
70
|
|
|
public function setClassifer(string $classifier, array $classifierOptions = []) |
71
|
|
|
{ |
72
|
|
|
if ($classifier != DecisionTree::class) { |
73
|
|
|
throw new \Exception("RandomForest can only use DecisionTree as base classifier"); |
74
|
|
|
} |
75
|
|
|
|
76
|
|
|
return parent::setClassifer($classifier, $classifierOptions); |
77
|
|
|
} |
78
|
|
|
|
79
|
|
|
/** |
80
|
|
|
* This will return an array including an importance value for |
81
|
|
|
* each column in the given dataset. Importance values for a column |
82
|
|
|
* is the average importance of that column in all trees in the forest |
83
|
|
|
* |
84
|
|
|
* @return array |
85
|
|
|
*/ |
86
|
|
|
public function getFeatureImportances() |
87
|
|
|
{ |
88
|
|
|
// Traverse each tree and sum importance of the columns |
89
|
|
|
$sum = []; |
90
|
|
|
foreach ($this->classifiers as $tree) { |
91
|
|
|
/* @var $tree DecisionTree */ |
92
|
|
|
$importances = $tree->getFeatureImportances(); |
93
|
|
|
|
94
|
|
|
foreach ($importances as $column => $importance) { |
95
|
|
|
if (array_key_exists($column, $sum)) { |
96
|
|
|
$sum[$column] += $importance; |
97
|
|
|
} else { |
98
|
|
|
$sum[$column] = $importance; |
99
|
|
|
} |
100
|
|
|
} |
101
|
|
|
} |
102
|
|
|
|
103
|
|
|
// Normalize & sort the importance values |
104
|
|
|
$total = array_sum($sum); |
105
|
|
|
foreach ($sum as &$importance) { |
106
|
|
|
$importance /= $total; |
107
|
|
|
} |
108
|
|
|
|
109
|
|
|
arsort($sum); |
110
|
|
|
|
111
|
|
|
return $sum; |
112
|
|
|
} |
113
|
|
|
|
114
|
|
|
/** |
115
|
|
|
* A string array to represent the columns is given. They are useful |
116
|
|
|
* when trying to print some information about the trees such as feature importances |
117
|
|
|
* |
118
|
|
|
* @param array $names |
119
|
|
|
* @return $this |
120
|
|
|
*/ |
121
|
|
|
public function setColumnNames(array $names) |
122
|
|
|
{ |
123
|
|
|
$this->columnNames = $names; |
124
|
|
|
|
125
|
|
|
return $this; |
126
|
|
|
} |
127
|
|
|
|
128
|
|
|
/** |
129
|
|
|
* @param DecisionTree $classifier |
130
|
|
|
* @param int $index |
131
|
|
|
* @return DecisionTree |
132
|
|
|
*/ |
133
|
|
|
protected function initSingleClassifier($classifier, $index) |
134
|
|
|
{ |
135
|
|
|
if (is_float($this->featureSubsetRatio)) { |
136
|
|
|
$featureCount = (int)($this->featureSubsetRatio * $this->featureCount); |
137
|
|
|
} elseif ($this->featureCount == 'sqrt') { |
138
|
|
|
$featureCount = (int)sqrt($this->featureCount) + 1; |
139
|
|
|
} else { |
140
|
|
|
$featureCount = (int)log($this->featureCount, 2) + 1; |
141
|
|
|
} |
142
|
|
|
|
143
|
|
|
if ($featureCount >= $this->featureCount) { |
144
|
|
|
$featureCount = $this->featureCount; |
145
|
|
|
} |
146
|
|
|
|
147
|
|
|
if ($this->columnNames === null) { |
148
|
|
|
$this->columnNames = range(0, $this->featureCount - 1); |
149
|
|
|
} |
150
|
|
|
|
151
|
|
|
return $classifier |
152
|
|
|
->setColumnNames($this->columnNames) |
153
|
|
|
->setNumFeatures($featureCount); |
154
|
|
|
} |
155
|
|
|
} |
156
|
|
|
|