php-ml/docs/machine-learning/classification/k-nearest-neighbors.md

40 lines
1023 B
Markdown
Raw Normal View History

2016-04-08 22:36:48 +00:00
# KNearestNeighbors Classifier
Classifier implementing the k-nearest neighbors algorithm.
2016-08-14 17:14:56 +00:00
## Constructor Parameters
2016-04-08 22:36:48 +00:00
* $k - number of nearest neighbors to scan (default: 3)
2017-01-05 20:06:10 +00:00
* $distanceMetric - Distance object, default Euclidean (see [distance documentation](../../math/distance.md))
2016-04-08 22:36:48 +00:00
```
$classifier = new KNearestNeighbors($k=4);
2016-04-16 19:41:37 +00:00
$classifier = new KNearestNeighbors($k=3, new Minkowski($lambda=4));
2016-04-08 22:36:48 +00:00
```
2016-08-14 17:14:56 +00:00
## Train
2016-04-08 22:36:48 +00:00
2016-05-02 11:49:19 +00:00
To train a classifier simply provide train samples and labels (as `array`). Example:
2016-04-08 22:36:48 +00:00
```
$samples = [[1, 3], [1, 4], [2, 4], [3, 1], [4, 1], [4, 2]];
$labels = ['a', 'a', 'a', 'b', 'b', 'b'];
$classifier = new KNearestNeighbors();
$classifier->train($samples, $labels);
```
You can train the classifier using multiple data sets, predictions will be based on all the training data.
2016-08-14 17:14:56 +00:00
## Predict
2016-04-08 22:36:48 +00:00
2016-05-02 11:49:19 +00:00
To predict sample label use `predict` method. You can provide one sample or array of samples:
2016-04-08 22:36:48 +00:00
```
$classifier->predict([3, 2]);
// return 'b'
$classifier->predict([[3, 2], [1, 5]]);
// return ['b', 'a']
```