for scalac RBM.scala

Yusuke Sugomori · Yusuke Sugomori · commit eafd9bdfffc2 · 2013-03-21T15:35:53.000+09:00
diff --git a/java/DBN/src/DBN.java b/java/DBN/src/DBN.java
@@ -150,7 +150,7 @@ public void predict(int[] x, double[] y) {
 		log_layer.softmax(y);
 	}
 	
-	public static void main(String[] arg) {
+	private static void test_dbn() {
 		Random rng = new Random(123);
 		
 		double pretrain_lr = 0.1;
@@ -215,4 +215,8 @@ public static void main(String[] arg) {
 			System.out.println();
 		}
 	}
+	
+	public static void main(String[] args) {
+		test_dbn();
+	}
 }
diff --git a/java/LogisticRegression/src/LogisticRegression.java b/java/LogisticRegression/src/LogisticRegression.java
@@ -65,7 +65,7 @@ public void predict(int[] x, double[] y) {
 		softmax(y);
 	}	
 	
-	public static void main(String[] arg) {
+	private static void test_lr() {
 		double learning_rate = 0.1;
 		double n_epochs = 500;
 		
@@ -121,4 +121,8 @@ public static void main(String[] arg) {
 			System.out.println();
 		}
 	}
+	
+	public static void main(String[] args) {
+		test_lr();
+	}
 }
diff --git a/java/RBM/src/RBM.java b/java/RBM/src/RBM.java
@@ -163,7 +163,7 @@ public void reconstruct(int[] v, double[] reconstructed_v) {
 
 	
 	
-	public static void main(String[] arg) {
+	private static void test_rbm() {
 		Random rng = new Random(123);
 
 		double learning_rate = 0.1;
@@ -212,4 +212,9 @@ public static void main(String[] arg) {
 			System.out.println();
 		}
 	}
+	
+	public static void main(String[] args) {
+		test_rbm();
+	}
+	
 }
diff --git a/scala/RBM.scala b/scala/RBM.scala
@@ -1,3 +1,6 @@
+// $ scalac RBM.scala
+// $ scala RBM
+
 import scala.util.Random
 import scala.math
 
@@ -159,57 +162,61 @@ class RBM(val N: Int, val n_visible: Int, val n_hidden: Int,
 }
 
 
-def test_rbm() {
-  val rng: Random = new Random(123)
+object RBM {
+  def test_rbm() {
+    val rng: Random = new Random(123)
 
-  var learning_rate: Double = 0.1
-  val training_epochs: Int = 1000
-  val k: Int = 1
+    var learning_rate: Double = 0.1
+    val training_epochs: Int = 1000
+    val k: Int = 1
 
-  val train_N: Int = 6;
-  val test_N: Int = 2
-  val n_visible: Int = 6
-  val n_hidden: Int = 3
+    val train_N: Int = 6;
+    val test_N: Int = 2
+    val n_visible: Int = 6
+    val n_hidden: Int = 3
 
-  val train_X: Array[Array[Int]] = Array(
-    Array(1, 1, 1, 0, 0, 0),
-    Array(1, 0, 1, 0, 0, 0),
-    Array(1, 1, 1, 0, 0, 0),
-    Array(0, 0, 1, 1, 1, 0),
-    Array(0, 0, 1, 0, 1, 0),
-    Array(0, 0, 1, 1, 1, 0)
-  )
+    val train_X: Array[Array[Int]] = Array(
+      Array(1, 1, 1, 0, 0, 0),
+      Array(1, 0, 1, 0, 0, 0),
+      Array(1, 1, 1, 0, 0, 0),
+      Array(0, 0, 1, 1, 1, 0),
+      Array(0, 0, 1, 0, 1, 0),
+      Array(0, 0, 1, 1, 1, 0)
+    )
 
 
-  val rbm: RBM = new RBM(train_N, n_visible, n_hidden, rng=rng)
+    val rbm: RBM = new RBM(train_N, n_visible, n_hidden, rng=rng)
 
-  var i: Int = 0
-  var j: Int = 0
+    var i: Int = 0
+    var j: Int = 0
 
-  // train
-  var epoch: Int = 0
-  for(epoch <- 0 until training_epochs) {
-    for(i <- 0 until train_N) {
-      rbm.contrastive_divergence(train_X(i), learning_rate, k)
+    // train
+    var epoch: Int = 0
+    for(epoch <- 0 until training_epochs) {
+      for(i <- 0 until train_N) {
+        rbm.contrastive_divergence(train_X(i), learning_rate, k)
+      }
     }
-  }
 
-  // test data
-  val test_X: Array[Array[Int]] = Array(
-    Array(1, 1, 0, 0, 0, 0),
-    Array(0, 0, 0, 1, 1, 0)
-  )
+    // test data
+    val test_X: Array[Array[Int]] = Array(
+      Array(1, 1, 0, 0, 0, 0),
+      Array(0, 0, 0, 1, 1, 0)
+    )
 
-  val reconstructed_X: Array[Array[Double]] = Array.ofDim[Double](test_N, n_visible)
-  for(i <- 0 until test_N) {
-    rbm.reconstruct(test_X(i), reconstructed_X(i))
-    for(j <- 0 until n_visible) {
-      printf("%.5f", reconstructed_X(i)(j))
+    val reconstructed_X: Array[Array[Double]] = Array.ofDim[Double](test_N, n_visible)
+    for(i <- 0 until test_N) {
+      rbm.reconstruct(test_X(i), reconstructed_X(i))
+      for(j <- 0 until n_visible) {
+        printf("%.5f ", reconstructed_X(i)(j))
+      }
+      println()
     }
-    println()
-  }
 
-}
+  }
 
-test_rbm()
+  def main(args: Array[String]) {
+    test_rbm()
+  }
 
+}

Original file line number	Diff line number	Diff line change
`@@ -150,7 +150,7 @@ public void predict(int[] x, double[] y) {`
`150`	`150`	`log_layer.softmax(y);`
`151`	`151`	`}`
`152`	`152`
`153`		`- public static void main(String[] arg) {`
	`153`	`+ private static void test_dbn() {`
`154`	`154`	`Random rng = new Random(123);`
`155`	`155`
`156`	`156`	`double pretrain_lr = 0.1;`
`@@ -215,4 +215,8 @@ public static void main(String[] arg) {`
`215`	`215`	`System.out.println();`
`216`	`216`	`}`
`217`	`217`	`}`
	`218`	`+`
	`219`	`+ public static void main(String[] args) {`
	`220`	`+ test_dbn();`
	`221`	`+ }`
`218`	`222`	`}`
Original file line number	Diff line number	Diff line change
`@@ -65,7 +65,7 @@ public void predict(int[] x, double[] y) {`
`65`	`65`	`softmax(y);`
`66`	`66`	`}`
`67`	`67`
`68`		`- public static void main(String[] arg) {`
	`68`	`+ private static void test_lr() {`
`69`	`69`	`double learning_rate = 0.1;`
`70`	`70`	`double n_epochs = 500;`
`71`	`71`
`@@ -121,4 +121,8 @@ public static void main(String[] arg) {`
`121`	`121`	`System.out.println();`
`122`	`122`	`}`
`123`	`123`	`}`
	`124`	`+`
	`125`	`+ public static void main(String[] args) {`
	`126`	`+ test_lr();`
	`127`	`+ }`
`124`	`128`	`}`
Original file line number	Diff line number	Diff line change
`@@ -163,7 +163,7 @@ public void reconstruct(int[] v, double[] reconstructed_v) {`
`163`	`163`
`164`	`164`
`165`	`165`
`166`		`- public static void main(String[] arg) {`
	`166`	`+ private static void test_rbm() {`
`167`	`167`	`Random rng = new Random(123);`
`168`	`168`
`169`	`169`	`double learning_rate = 0.1;`
`@@ -212,4 +212,9 @@ public static void main(String[] arg) {`
`212`	`212`	`System.out.println();`
`213`	`213`	`}`
`214`	`214`	`}`
	`215`	`+`
	`216`	`+ public static void main(String[] args) {`
	`217`	`+ test_rbm();`
	`218`	`+ }`
	`219`	`+`
`215`	`220`	`}`