Maybe done

Author: jkbradley

examples/src/main/java/org/apache/spark/examples/ml/JavaDeveloperApiExample.java

@@ -28,7 +28,6 @@
 import org.apache.spark.ml.classification.ClassificationModel;
 import org.apache.spark.ml.param.IntParam;
 import org.apache.spark.ml.param.ParamMap;
-import org.apache.spark.ml.param.Params;
 import org.apache.spark.ml.param.Params$;
 import org.apache.spark.mllib.linalg.BLAS;
 import org.apache.spark.mllib.linalg.Vector;
@@ -100,10 +99,12 @@ public static void main(String[] args) throws Exception {
 /**
  * Example of defining a type of {@link Classifier}.
  *
- * NOTE: This is private since it is an example.  In practice, you may not want it to be private.
+ * Note: Some IDEs (e.g., IntelliJ) will complain that this will not compile due to
+ *       {@link org.apache.spark.ml.param.Params#set} using incompatible return types.
+ *       However, this should still compile and run successfully.
  */
 class MyJavaLogisticRegression
-    extends Classifier<Vector, MyJavaLogisticRegression, MyJavaLogisticRegressionModel> implements Params {
+    extends Classifier<Vector, MyJavaLogisticRegression, MyJavaLogisticRegressionModel> {
 
   /**
    * Param for max number of iterations
@@ -144,10 +145,12 @@ public MyJavaLogisticRegressionModel train(DataFrame dataset, ParamMap paramMap)
 /**
  * Example of defining a type of {@link ClassificationModel}.
  *
- * NOTE: This is private since it is an example.  In practice, you may not want it to be private.
+ * Note: Some IDEs (e.g., IntelliJ) will complain that this will not compile due to
+ *       {@link org.apache.spark.ml.param.Params#set} using incompatible return types.
+ *       However, this should still compile and run successfully.
  */
 class MyJavaLogisticRegressionModel
-    extends ClassificationModel<Vector, MyJavaLogisticRegressionModel> implements Params {
+    extends ClassificationModel<Vector, MyJavaLogisticRegressionModel> {
 
   private MyJavaLogisticRegression parent_;
   public MyJavaLogisticRegression parent() { return parent_; }

mllib/src/main/scala/org/apache/spark/ml/param/params.scala

@@ -43,7 +43,7 @@ class Param[T] (val parent: Params, val name: String, val doc: String, val isVal
   extends Serializable {
 
   def this(parent: Params, name: String, doc: String) =
-    this(parent, name, doc, ParamValidate.default[T])
+    this(parent, name, doc, ParamValidate.alwaysTrue[T])
 
   /**
    * Assert that the given value is valid for this parameter.
@@ -96,8 +96,8 @@ class Param[T] (val parent: Params, val name: String, val doc: String, val isVal
  */
 object ParamValidate {
 
-  /** Default validation always return true */
-  def default[T]: T => Boolean = (_: T) => true
+  /** (private[param]) Default validation always return true */
+  private[param] def alwaysTrue[T]: T => Boolean = (_: T) => true
 
   /**
    * Private method for checking numerical types and converting to Double.
@@ -176,7 +176,7 @@ class DoubleParam(parent: Params, name: String, doc: String, isValid: Double =>
   extends Param[Double](parent, name, doc, isValid) {
 
   def this(parent: Params, name: String, doc: String) =
-    this(parent, name, doc, ParamValidate.default[Double])
+    this(parent, name, doc, ParamValidate.alwaysTrue[Double])
 
   override def w(value: Double): ParamPair[Double] = super.w(value)
 }
@@ -186,7 +186,7 @@ class IntParam(parent: Params, name: String, doc: String, isValid: Int => Boolea
   extends Param[Int](parent, name, doc) {
 
   def this(parent: Params, name: String, doc: String) =
-    this(parent, name, doc, ParamValidate.default[Int])
+    this(parent, name, doc, ParamValidate.alwaysTrue[Int])
 
   override def w(value: Int): ParamPair[Int] = super.w(value)
 }
@@ -196,7 +196,7 @@ class FloatParam(parent: Params, name: String, doc: String, isValid: Float => Bo
   extends Param[Float](parent, name, doc) {
 
   def this(parent: Params, name: String, doc: String) =
-    this(parent, name, doc, ParamValidate.default[Float])
+    this(parent, name, doc, ParamValidate.alwaysTrue[Float])
 
   override def w(value: Float): ParamPair[Float] = super.w(value)
 }
@@ -206,7 +206,7 @@ class LongParam(parent: Params, name: String, doc: String, isValid: Long => Bool
   extends Param[Long](parent, name, doc) {
 
   def this(parent: Params, name: String, doc: String) =
-    this(parent, name, doc, ParamValidate.default[Long])
+    this(parent, name, doc, ParamValidate.alwaysTrue[Long])
 
   override def w(value: Long): ParamPair[Long] = super.w(value)
 }
@@ -351,11 +351,13 @@ trait Params extends Identifiable with Serializable {
 
   /**
    * Sets default values for a list of params.
+   *
+   * Note: Java developers should use the single-parameter [[setDefault()]].
+   *
    * @param paramPairs  a list of param pairs that specify params and their default values to set
    *                    respectively. Make sure that the params are initialized before this method
    *                    gets called.
    */
-  @varargs
   protected final def setDefault(paramPairs: ParamPair[_]*): this.type = {
     paramPairs.foreach { p =>
       setDefault(p.param.asInstanceOf[Param[Any]], p.value)

mllib/src/test/java/org/apache/spark/ml/param/JavaParamsSuite.java

@@ -1,5 +1,6 @@
 package org.apache.spark.ml.param;
 
+import com.google.common.collect.Lists;
 import org.junit.After;
 import org.junit.Assert;
 import org.junit.Before;
@@ -29,7 +30,21 @@ public void tearDown() {
   public void testParams() {
     JavaTestParams testParams = new JavaTestParams();
     Assert.assertEquals(testParams.getMyIntParam(), 1);
-    testParams.setMyIntParam(2).setMyDoubleParam(0.4);
+    testParams.setMyIntParam(2).setMyDoubleParam(0.4).setMyStringParam("a");
     Assert.assertEquals(testParams.getMyDoubleParam(), 0.4, 0.0);
+    Assert.assertEquals(testParams.getMyStringParam(), "a");
+  }
+
+  @Test
+  public void testParamValidate() {
+    ParamValidate.alwaysTrue();
+    ParamValidate.gt(1.0);
+    ParamValidate.gtEq(1.0);
+    ParamValidate.lt(1.0);
+    ParamValidate.ltEq(1.0);
+    ParamValidate.inRange(0, 1, true, false);
+    ParamValidate.inRange(0, 1);
+    ParamValidate.inArray(Lists.newArrayList(0, 1, 3));
+    ParamValidate.inArray(Lists.newArrayList("a", "b"));
   }
 }

mllib/src/test/java/org/apache/spark/ml/param/JavaTestParams.java

@@ -40,6 +40,7 @@ public JavaTestParams() {
     List<String> validStrings = Lists.newArrayList("a", "b");
     myStringParam = new Param<String>(this, "myStringParam", "this is a string param",
       ParamValidate.inArray(validStrings));
-    setDefault(myIntParam.w(1), myDoubleParam.w(0.5));
+    setDefault(myIntParam, 1);
+    setDefault(myDoubleParam, 0.5);
   }
 }

mllib/src/test/scala/org/apache/spark/ml/param/ParamsSuite.scala

@@ -147,8 +147,8 @@ class ParamsSuite extends FunSuite {
   }
 
   test("ParamValidate") {
-    val default = ParamValidate.default[Int]
-    assert(default(1))
+    val alwaysTrue = ParamValidate.alwaysTrue[Int]
+    assert(alwaysTrue(1))
 
     val gt1Int = ParamValidate.gt[Int](1)
     assert(!gt1Int(1) && gt1Int(2))