@@ -183,7 +183,7 @@ discard block |
||
183 | 183 | |
184 | 184 | /** |
185 | 185 | * @param array $records |
186 | - * @return DecisionTreeLeaf[] |
|
186 | + * @return null|DecisionTreeLeaf |
|
187 | 187 | */ |
188 | 188 | protected function getBestSplit($records) |
189 | 189 | { |
@@ -359,7 +359,6 @@ discard block |
||
359 | 359 | /** |
360 | 360 | * Used to set predefined features to consider while deciding which column to use for a split, |
361 | 361 | * |
362 | - * @param array $features |
|
363 | 362 | */ |
364 | 363 | protected function setSelectedFeatures(array $selectedFeatures) |
365 | 364 | { |
@@ -397,7 +396,6 @@ discard block |
||
397 | 396 | * each column in the given dataset. The importance values are |
398 | 397 | * normalized and their total makes 1.<br/> |
399 | 398 | * |
400 | - * @param array $labels |
|
401 | 399 | * @return array |
402 | 400 | */ |
403 | 401 | public function getFeatureImportances() |
@@ -437,7 +435,6 @@ discard block |
||
437 | 435 | * |
438 | 436 | * @param int $column |
439 | 437 | * @param DecisionTreeLeaf |
440 | - * @param array $collected |
|
441 | 438 | * |
442 | 439 | * @return array |
443 | 440 | */ |
@@ -1,6 +1,6 @@ discard block |
||
1 | 1 | <?php |
2 | 2 | |
3 | -declare(strict_types=1); |
|
3 | +declare(strict_types = 1); |
|
4 | 4 | |
5 | 5 | namespace Phpml\Classification; |
6 | 6 | |
@@ -112,7 +112,7 @@ discard block |
||
112 | 112 | protected function getColumnTypes(array $samples) |
113 | 113 | { |
114 | 114 | $types = []; |
115 | - for ($i=0; $i<$this->featureCount; $i++) { |
|
115 | + for ($i = 0; $i < $this->featureCount; $i++) { |
|
116 | 116 | $values = array_column($samples, $i); |
117 | 117 | $isCategorical = $this->isCategoricalColumn($values); |
118 | 118 | $types[] = $isCategorical ? self::NOMINAL : self::CONTINUOS; |
@@ -136,7 +136,7 @@ discard block |
||
136 | 136 | // otherwise group the records so that we can classify the leaf |
137 | 137 | // in case maximum depth is reached |
138 | 138 | $leftRecords = []; |
139 | - $rightRecords= []; |
|
139 | + $rightRecords = []; |
|
140 | 140 | $remainingTargets = []; |
141 | 141 | $prevRecord = null; |
142 | 142 | $allSame = true; |
@@ -154,12 +154,12 @@ discard block |
||
154 | 154 | if ($split->evaluate($record)) { |
155 | 155 | $leftRecords[] = $recordNo; |
156 | 156 | } else { |
157 | - $rightRecords[]= $recordNo; |
|
157 | + $rightRecords[] = $recordNo; |
|
158 | 158 | } |
159 | 159 | |
160 | 160 | // Group remaining targets |
161 | 161 | $target = $this->targets[$recordNo]; |
162 | - if (! array_key_exists($target, $remainingTargets)) { |
|
162 | + if (!array_key_exists($target, $remainingTargets)) { |
|
163 | 163 | $remainingTargets[$target] = 1; |
164 | 164 | } else { |
165 | 165 | $remainingTargets[$target]++; |
@@ -175,7 +175,7 @@ discard block |
||
175 | 175 | $split->leftLeaf = $this->getSplitLeaf($leftRecords, $depth + 1); |
176 | 176 | } |
177 | 177 | if ($rightRecords) { |
178 | - $split->rightLeaf= $this->getSplitLeaf($rightRecords, $depth + 1); |
|
178 | + $split->rightLeaf = $this->getSplitLeaf($rightRecords, $depth + 1); |
|
179 | 179 | } |
180 | 180 | } |
181 | 181 | return $split; |
@@ -234,7 +234,7 @@ discard block |
||
234 | 234 | protected function getSelectedFeatures() |
235 | 235 | { |
236 | 236 | $allFeatures = range(0, $this->featureCount - 1); |
237 | - if ($this->numUsableFeatures == 0 && ! $this->selectedFeatures) { |
|
237 | + if ($this->numUsableFeatures == 0 && !$this->selectedFeatures) { |
|
238 | 238 | return $allFeatures; |
239 | 239 | } |
240 | 240 | |
@@ -270,7 +270,7 @@ discard block |
||
270 | 270 | $countMatrix[$label][$rowIndex]++; |
271 | 271 | } |
272 | 272 | $giniParts = [0, 0]; |
273 | - for ($i=0; $i<=1; $i++) { |
|
273 | + for ($i = 0; $i <= 1; $i++) { |
|
274 | 274 | $part = 0; |
275 | 275 | $sum = array_sum(array_column($countMatrix, $i)); |
276 | 276 | if ($sum > 0) { |
@@ -292,7 +292,7 @@ discard block |
||
292 | 292 | // Detect and convert continuous data column values into |
293 | 293 | // discrete values by using the median as a threshold value |
294 | 294 | $columns = []; |
295 | - for ($i=0; $i<$this->featureCount; $i++) { |
|
295 | + for ($i = 0; $i < $this->featureCount; $i++) { |
|
296 | 296 | $values = array_column($samples, $i); |
297 | 297 | if ($this->columnTypes[$i] == self::CONTINUOS) { |
298 | 298 | $median = Mean::median($values); |
@@ -52,7 +52,7 @@ |
||
52 | 52 | * If normalizeInputs is set to true, then every input given to the algorithm will be standardized |
53 | 53 | * by use of standard deviation and mean calculation |
54 | 54 | * |
55 | - * @param int $learningRate |
|
55 | + * @param double $learningRate |
|
56 | 56 | * @param int $maxIterations |
57 | 57 | */ |
58 | 58 | public function __construct(float $learningRate = 0.001, int $maxIterations = 1000, |
@@ -4,9 +4,6 @@ |
||
4 | 4 | |
5 | 5 | namespace Phpml\Classification\Linear; |
6 | 6 | |
7 | -use Phpml\Helper\Predictable; |
|
8 | -use Phpml\Helper\Trainable; |
|
9 | -use Phpml\Classification\Classifier; |
|
10 | 7 | use Phpml\Classification\Linear\Perceptron; |
11 | 8 | use Phpml\Preprocessing\Normalizer; |
12 | 9 |
@@ -1,6 +1,6 @@ discard block |
||
1 | 1 | <?php |
2 | 2 | |
3 | -declare(strict_types=1); |
|
3 | +declare(strict_types = 1); |
|
4 | 4 | |
5 | 5 | namespace Phpml\Classification\Linear; |
6 | 6 | |
@@ -16,12 +16,12 @@ discard block |
||
16 | 16 | /** |
17 | 17 | * Batch training is the default Adaline training algorithm |
18 | 18 | */ |
19 | - const BATCH_TRAINING = 1; |
|
19 | + const BATCH_TRAINING = 1; |
|
20 | 20 | |
21 | 21 | /** |
22 | 22 | * Online training: Stochastic gradient descent learning |
23 | 23 | */ |
24 | - const ONLINE_TRAINING = 2; |
|
24 | + const ONLINE_TRAINING = 2; |
|
25 | 25 | |
26 | 26 | /** |
27 | 27 | * The function whose result will be used to calculate the network error |
@@ -62,7 +62,7 @@ discard block |
||
62 | 62 | $this->normalizer = new Normalizer(Normalizer::NORM_STD); |
63 | 63 | } |
64 | 64 | |
65 | - if (! in_array($trainingType, [self::BATCH_TRAINING, self::ONLINE_TRAINING])) { |
|
65 | + if (!in_array($trainingType, [self::BATCH_TRAINING, self::ONLINE_TRAINING])) { |
|
66 | 66 | throw new \Exception("Adaline can only be trained with batch and online/stochastic gradient descent algorithm"); |
67 | 67 | } |
68 | 68 | $this->trainingType = $trainingType; |
@@ -103,7 +103,7 @@ discard block |
||
103 | 103 | $sum = array_sum($updates); |
104 | 104 | |
105 | 105 | // Updates all weights at once |
106 | - for ($i=0; $i <= $this->featureCount; $i++) { |
|
106 | + for ($i = 0; $i <= $this->featureCount; $i++) { |
|
107 | 107 | if ($i == 0) { |
108 | 108 | $this->weights[0] += $this->learningRate * $sum; |
109 | 109 | } else { |
@@ -1,6 +1,6 @@ discard block |
||
1 | 1 | <?php |
2 | 2 | |
3 | -declare(strict_types=1); |
|
3 | +declare(strict_types = 1); |
|
4 | 4 | |
5 | 5 | namespace Phpml\Preprocessing; |
6 | 6 | |
@@ -12,7 +12,7 @@ discard block |
||
12 | 12 | { |
13 | 13 | const NORM_L1 = 1; |
14 | 14 | const NORM_L2 = 2; |
15 | - const NORM_STD= 3; |
|
15 | + const NORM_STD = 3; |
|
16 | 16 | |
17 | 17 | /** |
18 | 18 | * @var int |
@@ -117,7 +117,7 @@ discard block |
||
117 | 117 | foreach ($sample as $feature) { |
118 | 118 | $norm2 += $feature * $feature; |
119 | 119 | } |
120 | - $norm2 = sqrt((float)$norm2); |
|
120 | + $norm2 = sqrt((float) $norm2); |
|
121 | 121 | |
122 | 122 | if (0 == $norm2) { |
123 | 123 | $sample = array_fill(0, count($sample), 1); |
@@ -20,7 +20,7 @@ discard block |
||
20 | 20 | */ |
21 | 21 | protected static $errorFunction = 'outputClass'; |
22 | 22 | |
23 | - /** |
|
23 | + /** |
|
24 | 24 | * @var array |
25 | 25 | */ |
26 | 26 | protected $samples = []; |
@@ -78,7 +78,7 @@ discard block |
||
78 | 78 | $this->maxIterations = $maxIterations; |
79 | 79 | } |
80 | 80 | |
81 | - /** |
|
81 | + /** |
|
82 | 82 | * @param array $samples |
83 | 83 | * @param array $targets |
84 | 84 | */ |
@@ -1,6 +1,6 @@ discard block |
||
1 | 1 | <?php |
2 | 2 | |
3 | -declare(strict_types=1); |
|
3 | +declare(strict_types = 1); |
|
4 | 4 | |
5 | 5 | namespace Phpml\Classification\Linear; |
6 | 6 | |
@@ -123,7 +123,7 @@ discard block |
||
123 | 123 | // Update bias |
124 | 124 | $this->weights[0] += $update * $this->learningRate; // Bias |
125 | 125 | // Update other weights |
126 | - for ($i=1; $i <= $this->featureCount; $i++) { |
|
126 | + for ($i = 1; $i <= $this->featureCount; $i++) { |
|
127 | 127 | $this->weights[$i] += $update * $sample[$i - 1] * $this->learningRate; |
128 | 128 | } |
129 | 129 | } |
@@ -169,6 +169,6 @@ discard block |
||
169 | 169 | { |
170 | 170 | $predictedClass = $this->outputClass($sample); |
171 | 171 | |
172 | - return $this->labels[ $predictedClass ]; |
|
172 | + return $this->labels[$predictedClass]; |
|
173 | 173 | } |
174 | 174 | } |
@@ -1,6 +1,6 @@ |
||
1 | 1 | <?php |
2 | 2 | |
3 | -declare(strict_types=1); |
|
3 | +declare(strict_types = 1); |
|
4 | 4 | |
5 | 5 | namespace Phpml\Classification\Linear; |
6 | 6 |