Change the default kernel type in SVC to Kernel::RBF (#267)

* Change the default kernel type in SVC to Kernel::RBF * Update CHANGELOG.md
UDPress · Jun 20, 2018 · ab22cc5 · ab22cc5
1 parent 46fa2c2
commit ab22cc5
Show file tree

Hide file tree

Showing 3 changed files with 3 additions and 2 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -4,6 +4,7 @@ CHANGELOG
 This changelog references the relevant changes done in PHP-ML library.
 
 * Unreleased
+    * feature [Dataset] changed the default kernel type in SVC to Kernel::RBF (#267)
     * feature [Clustering] added KMeans associative clustering (#262)
     * feature [Dataset] added removeColumns function to ArrayDataset (#249)
     * feature [Dataset] added a SvmDataset class for SVM-Light (or LibSVM) format files (#237)

diff --git a/docs/machine-learning/classification/svc.md b/docs/machine-learning/classification/svc.md
@@ -4,7 +4,7 @@ Classifier implementing Support Vector Machine based on libsvm.
 
 ### Constructor Parameters
 
-* $kernel (int) - kernel type to be used in the algorithm (default Kernel::LINEAR)
+* $kernel (int) - kernel type to be used in the algorithm (default Kernel::RBF)
 * $cost (float) - parameter C of C-SVC (default 1.0)
 * $degree (int) - degree of the Kernel::POLYNOMIAL function (default 3)
 * $gamma (float) - kernel coefficient for ‘Kernel::RBF’, ‘Kernel::POLYNOMIAL’ and ‘Kernel::SIGMOID’. If gamma is ‘null’ then 1/features will be used instead.

diff --git a/src/Classification/SVC.php b/src/Classification/SVC.php
@@ -11,7 +11,7 @@
 class SVC extends SupportVectorMachine implements Classifier
 {
     public function __construct(
-        int $kernel = Kernel::LINEAR,
+        int $kernel = Kernel::RBF,
         float $cost = 1.0,
         int $degree = 3,
         ?float $gamma = null,