Scala for Machine Learning

感谢分享！

好流弊的样子，下载看看。谢谢搂住

谢谢分享！

O(∩_∩)O谢谢

看看怎么样

谢谢分享

1. Measuring similarity
   
2. def manhattan[T <% Double, U <% Double](x: Array[T], y: Array[U]): Double = (x, y).zipped.foldLeft(0.0)((s, t) => s + Math.abs(t._1 - t._2))
   
3. 
   
4. def euclidean[T <% Double, U <% Double](x: Array[T], y: Array[U]): Double =  Math.sqrt((x, y).zipped.foldLeft(0.0)((s, t) => { val d = t._1 - t._2; s + d*d} ))
   
5. 
   
6. def cosine[T <% Double, U <% Double](x: Array[T], y: Array[U]): Double = {
   
7.   val zeros = (0.0, 0.0, 0.0)
   
8.   val norms = (x, y).zipped.foldLeft(zeros)((s, t) =>
   
9.      (s._1 + t._1*t._2, s._2 + t._1*t._1, s._3 + t._2*t._2))
   
10.   norms._1/Math.sqrt(norms._2*norms._3)

复制代码

1. Defining K-means
   
2. 
   
3. Let's declare the K-means algorithm class, KMeans, with its public methods.
   
4. 
   
5. The KMeans class takes the number of clusters, K, and the maximum number of iterations, maxIters, as parameters. The implicit conversion of type T to a Double is specified by the T <% Double view bound. The Ordering class has to be passed implicitly as a parameter because it is required by the sortWith method in the initialize and maxBy methods. The Manifest method is required to preserve the type erasure for Array[T] in the JVM:
   
6. 
   
7. class KMeans[T <% Double](K: Int, maxIters: Int, distance: (DblVector,Array[T]) => Double)(implicit order: Ordering[T], m: Manifest[T]) extends PipeOperator[XTSeries[Array[T]], List[Cluster[T]]] {
   
8.   def |> : PartialFunction[XTSeries[Array[T]], List[Cluster[T]]]
   
9.   def initialize(xt:XTSeries[Array[T]]): List[Cluster[T]]

复制代码