From 66159c35010af35098dd1ec75475bb5d4d0fd6ca Mon Sep 17 00:00:00 2001 From: Reynold Xin Date: Wed, 8 Apr 2015 11:32:44 -0700 Subject: [SPARK-6765] Fix test code style for mllib. So we can turn style checker on for test code. Author: Reynold Xin Closes #5411 from rxin/test-style-mllib and squashes the following commits: d8a2569 [Reynold Xin] [SPARK-6765] Fix test code style for mllib. --- .../scala/org/apache/spark/ml/param/TestParams.scala | 2 +- .../spark/mllib/classification/NaiveBayesSuite.scala | 2 +- .../StreamingLogisticRegressionSuite.scala | 2 +- .../apache/spark/mllib/clustering/KMeansSuite.scala | 10 +++++++--- .../org/apache/spark/mllib/clustering/LDASuite.scala | 4 ++-- .../spark/mllib/clustering/StreamingKMeansSuite.scala | 4 ++-- .../apache/spark/mllib/random/RandomRDDsSuite.scala | 5 +++-- .../apache/spark/mllib/recommendation/ALSSuite.scala | 3 +++ .../spark/mllib/regression/RidgeRegressionSuite.scala | 2 +- .../regression/StreamingLinearRegressionSuite.scala | 2 +- .../org/apache/spark/mllib/util/TestingUtils.scala | 14 +++++++------- .../apache/spark/mllib/util/TestingUtilsSuite.scala | 18 +++++++++++------- 12 files changed, 40 insertions(+), 28 deletions(-) (limited to 'mllib') diff --git a/mllib/src/test/scala/org/apache/spark/ml/param/TestParams.scala b/mllib/src/test/scala/org/apache/spark/ml/param/TestParams.scala index 1a65883d78..ce52f2f230 100644 --- a/mllib/src/test/scala/org/apache/spark/ml/param/TestParams.scala +++ b/mllib/src/test/scala/org/apache/spark/ml/param/TestParams.scala @@ -28,7 +28,7 @@ class TestParams extends Params { def setInputCol(value: String): this.type = { set(inputCol, value); this } def getInputCol: String = get(inputCol) - override def validate(paramMap: ParamMap) = { + override def validate(paramMap: ParamMap): Unit = { val m = this.paramMap ++ paramMap require(m(maxIter) >= 0) require(m.contains(inputCol)) diff --git a/mllib/src/test/scala/org/apache/spark/mllib/classification/NaiveBayesSuite.scala b/mllib/src/test/scala/org/apache/spark/mllib/classification/NaiveBayesSuite.scala index f9fe3e006c..ea89b17b7c 100644 --- a/mllib/src/test/scala/org/apache/spark/mllib/classification/NaiveBayesSuite.scala +++ b/mllib/src/test/scala/org/apache/spark/mllib/classification/NaiveBayesSuite.scala @@ -102,7 +102,7 @@ class NaiveBayesSuite extends FunSuite with MLlibTestSparkContext { def validateModelFit( piData: Array[Double], thetaData: Array[Array[Double]], - model: NaiveBayesModel) = { + model: NaiveBayesModel): Unit = { def closeFit(d1: Double, d2: Double, precision: Double): Boolean = { (d1 - d2).abs <= precision } diff --git a/mllib/src/test/scala/org/apache/spark/mllib/classification/StreamingLogisticRegressionSuite.scala b/mllib/src/test/scala/org/apache/spark/mllib/classification/StreamingLogisticRegressionSuite.scala index d50c43d439..5683b55e85 100644 --- a/mllib/src/test/scala/org/apache/spark/mllib/classification/StreamingLogisticRegressionSuite.scala +++ b/mllib/src/test/scala/org/apache/spark/mllib/classification/StreamingLogisticRegressionSuite.scala @@ -30,7 +30,7 @@ import org.apache.spark.streaming.TestSuiteBase class StreamingLogisticRegressionSuite extends FunSuite with TestSuiteBase { // use longer wait time to ensure job completion - override def maxWaitTimeMillis = 30000 + override def maxWaitTimeMillis: Int = 30000 // Test if we can accurately learn B for Y = logistic(BX) on streaming data test("parameter accuracy") { diff --git a/mllib/src/test/scala/org/apache/spark/mllib/clustering/KMeansSuite.scala b/mllib/src/test/scala/org/apache/spark/mllib/clustering/KMeansSuite.scala index 7bf250eb5a..0f2b26d462 100644 --- a/mllib/src/test/scala/org/apache/spark/mllib/clustering/KMeansSuite.scala +++ b/mllib/src/test/scala/org/apache/spark/mllib/clustering/KMeansSuite.scala @@ -199,9 +199,13 @@ class KMeansSuite extends FunSuite with MLlibTestSparkContext { test("k-means|| initialization") { case class VectorWithCompare(x: Vector) extends Ordered[VectorWithCompare] { - @Override def compare(that: VectorWithCompare): Int = { - if(this.x.toArray.foldLeft[Double](0.0)((acc, x) => acc + x * x) > - that.x.toArray.foldLeft[Double](0.0)((acc, x) => acc + x * x)) -1 else 1 + override def compare(that: VectorWithCompare): Int = { + if (this.x.toArray.foldLeft[Double](0.0)((acc, x) => acc + x * x) > + that.x.toArray.foldLeft[Double](0.0)((acc, x) => acc + x * x)) { + -1 + } else { + 1 + } } } diff --git a/mllib/src/test/scala/org/apache/spark/mllib/clustering/LDASuite.scala b/mllib/src/test/scala/org/apache/spark/mllib/clustering/LDASuite.scala index 302d751eb8..15de10fd13 100644 --- a/mllib/src/test/scala/org/apache/spark/mllib/clustering/LDASuite.scala +++ b/mllib/src/test/scala/org/apache/spark/mllib/clustering/LDASuite.scala @@ -19,7 +19,7 @@ package org.apache.spark.mllib.clustering import org.scalatest.FunSuite -import org.apache.spark.mllib.linalg.{DenseMatrix, Matrix, Vectors} +import org.apache.spark.mllib.linalg.{Vector, DenseMatrix, Matrix, Vectors} import org.apache.spark.mllib.util.MLlibTestSparkContext import org.apache.spark.mllib.util.TestingUtils._ @@ -141,7 +141,7 @@ private[clustering] object LDASuite { (terms.toArray, termWeights.toArray) } - def tinyCorpus = Array( + def tinyCorpus: Array[(Long, Vector)] = Array( Vectors.dense(1, 3, 0, 2, 8), Vectors.dense(0, 2, 1, 0, 4), Vectors.dense(2, 3, 12, 3, 1), diff --git a/mllib/src/test/scala/org/apache/spark/mllib/clustering/StreamingKMeansSuite.scala b/mllib/src/test/scala/org/apache/spark/mllib/clustering/StreamingKMeansSuite.scala index 850c9fce50..f90025d535 100644 --- a/mllib/src/test/scala/org/apache/spark/mllib/clustering/StreamingKMeansSuite.scala +++ b/mllib/src/test/scala/org/apache/spark/mllib/clustering/StreamingKMeansSuite.scala @@ -27,7 +27,7 @@ import org.apache.spark.util.random.XORShiftRandom class StreamingKMeansSuite extends FunSuite with TestSuiteBase { - override def maxWaitTimeMillis = 30000 + override def maxWaitTimeMillis: Int = 30000 test("accuracy for single center and equivalence to grand average") { // set parameters @@ -59,7 +59,7 @@ class StreamingKMeansSuite extends FunSuite with TestSuiteBase { // estimated center from streaming should exactly match the arithmetic mean of all data points // because the decay factor is set to 1.0 val grandMean = - input.flatten.map(x => x.toBreeze).reduce(_+_) / (numBatches * numPoints).toDouble + input.flatten.map(x => x.toBreeze).reduce(_ + _) / (numBatches * numPoints).toDouble assert(model.latestModel().clusterCenters(0) ~== Vectors.dense(grandMean.toArray) absTol 1E-5) } diff --git a/mllib/src/test/scala/org/apache/spark/mllib/random/RandomRDDsSuite.scala b/mllib/src/test/scala/org/apache/spark/mllib/random/RandomRDDsSuite.scala index 6395188a08..63f2ea916d 100644 --- a/mllib/src/test/scala/org/apache/spark/mllib/random/RandomRDDsSuite.scala +++ b/mllib/src/test/scala/org/apache/spark/mllib/random/RandomRDDsSuite.scala @@ -181,7 +181,8 @@ class RandomRDDsSuite extends FunSuite with MLlibTestSparkContext with Serializa val poisson = RandomRDDs.poissonVectorRDD(sc, poissonMean, rows, cols, parts, seed) testGeneratedVectorRDD(poisson, rows, cols, parts, poissonMean, math.sqrt(poissonMean), 0.1) - val exponential = RandomRDDs.exponentialVectorRDD(sc, exponentialMean, rows, cols, parts, seed) + val exponential = + RandomRDDs.exponentialVectorRDD(sc, exponentialMean, rows, cols, parts, seed) testGeneratedVectorRDD(exponential, rows, cols, parts, exponentialMean, exponentialMean, 0.1) val gamma = RandomRDDs.gammaVectorRDD(sc, gammaShape, gammaScale, rows, cols, parts, seed) @@ -197,7 +198,7 @@ private[random] class MockDistro extends RandomDataGenerator[Double] { // This allows us to check that each partition has a different seed override def nextValue(): Double = seed.toDouble - override def setSeed(seed: Long) = this.seed = seed + override def setSeed(seed: Long): Unit = this.seed = seed override def copy(): MockDistro = new MockDistro } diff --git a/mllib/src/test/scala/org/apache/spark/mllib/recommendation/ALSSuite.scala b/mllib/src/test/scala/org/apache/spark/mllib/recommendation/ALSSuite.scala index 8775c0ca9d..b3798940dd 100644 --- a/mllib/src/test/scala/org/apache/spark/mllib/recommendation/ALSSuite.scala +++ b/mllib/src/test/scala/org/apache/spark/mllib/recommendation/ALSSuite.scala @@ -203,6 +203,7 @@ class ALSSuite extends FunSuite with MLlibTestSparkContext { * @param numProductBlocks number of product blocks to partition products into * @param negativeFactors whether the generated user/product factors can have negative entries */ + // scalastyle:off def testALS( users: Int, products: Int, @@ -216,6 +217,8 @@ class ALSSuite extends FunSuite with MLlibTestSparkContext { numUserBlocks: Int = -1, numProductBlocks: Int = -1, negativeFactors: Boolean = true) { + // scalastyle:on + val (sampledRatings, trueRatings, truePrefs) = ALSSuite.generateRatings(users, products, features, samplingRate, implicitPrefs, negativeWeights, negativeFactors) diff --git a/mllib/src/test/scala/org/apache/spark/mllib/regression/RidgeRegressionSuite.scala b/mllib/src/test/scala/org/apache/spark/mllib/regression/RidgeRegressionSuite.scala index 43d61151e2..d6c93cc0e4 100644 --- a/mllib/src/test/scala/org/apache/spark/mllib/regression/RidgeRegressionSuite.scala +++ b/mllib/src/test/scala/org/apache/spark/mllib/regression/RidgeRegressionSuite.scala @@ -35,7 +35,7 @@ private object RidgeRegressionSuite { class RidgeRegressionSuite extends FunSuite with MLlibTestSparkContext { - def predictionError(predictions: Seq[Double], input: Seq[LabeledPoint]) = { + def predictionError(predictions: Seq[Double], input: Seq[LabeledPoint]): Double = { predictions.zip(input).map { case (prediction, expected) => (prediction - expected.label) * (prediction - expected.label) }.reduceLeft(_ + _) / predictions.size diff --git a/mllib/src/test/scala/org/apache/spark/mllib/regression/StreamingLinearRegressionSuite.scala b/mllib/src/test/scala/org/apache/spark/mllib/regression/StreamingLinearRegressionSuite.scala index 24fd8df691..26604dbe6c 100644 --- a/mllib/src/test/scala/org/apache/spark/mllib/regression/StreamingLinearRegressionSuite.scala +++ b/mllib/src/test/scala/org/apache/spark/mllib/regression/StreamingLinearRegressionSuite.scala @@ -29,7 +29,7 @@ import org.apache.spark.streaming.TestSuiteBase class StreamingLinearRegressionSuite extends FunSuite with TestSuiteBase { // use longer wait time to ensure job completion - override def maxWaitTimeMillis = 20000 + override def maxWaitTimeMillis: Int = 20000 // Assert that two values are equal within tolerance epsilon def assertEqual(v1: Double, v2: Double, epsilon: Double) { diff --git a/mllib/src/test/scala/org/apache/spark/mllib/util/TestingUtils.scala b/mllib/src/test/scala/org/apache/spark/mllib/util/TestingUtils.scala index e957fa5d25..352193a678 100644 --- a/mllib/src/test/scala/org/apache/spark/mllib/util/TestingUtils.scala +++ b/mllib/src/test/scala/org/apache/spark/mllib/util/TestingUtils.scala @@ -95,16 +95,16 @@ object TestingUtils { /** * Comparison using absolute tolerance. */ - def absTol(eps: Double): CompareDoubleRightSide = CompareDoubleRightSide(AbsoluteErrorComparison, - x, eps, ABS_TOL_MSG) + def absTol(eps: Double): CompareDoubleRightSide = + CompareDoubleRightSide(AbsoluteErrorComparison, x, eps, ABS_TOL_MSG) /** * Comparison using relative tolerance. */ - def relTol(eps: Double): CompareDoubleRightSide = CompareDoubleRightSide(RelativeErrorComparison, - x, eps, REL_TOL_MSG) + def relTol(eps: Double): CompareDoubleRightSide = + CompareDoubleRightSide(RelativeErrorComparison, x, eps, REL_TOL_MSG) - override def toString = x.toString + override def toString: String = x.toString } case class CompareVectorRightSide( @@ -166,7 +166,7 @@ object TestingUtils { x.toArray.zip(y.toArray).forall(x => x._1 ~= x._2 relTol eps) }, x, eps, REL_TOL_MSG) - override def toString = x.toString + override def toString: String = x.toString } case class CompareMatrixRightSide( @@ -229,7 +229,7 @@ object TestingUtils { x.toArray.zip(y.toArray).forall(x => x._1 ~= x._2 relTol eps) }, x, eps, REL_TOL_MSG) - override def toString = x.toString + override def toString: String = x.toString } } diff --git a/mllib/src/test/scala/org/apache/spark/mllib/util/TestingUtilsSuite.scala b/mllib/src/test/scala/org/apache/spark/mllib/util/TestingUtilsSuite.scala index b0ecb33c28..59e6c77880 100644 --- a/mllib/src/test/scala/org/apache/spark/mllib/util/TestingUtilsSuite.scala +++ b/mllib/src/test/scala/org/apache/spark/mllib/util/TestingUtilsSuite.scala @@ -88,16 +88,20 @@ class TestingUtilsSuite extends FunSuite { assert(!(17.8 ~= 17.59 absTol 0.2)) // Comparisons of numbers very close to zero, and both side of zeros - assert(Double.MinPositiveValue ~== 4 * Double.MinPositiveValue absTol 5 * Double.MinPositiveValue) - assert(Double.MinPositiveValue !~== 6 * Double.MinPositiveValue absTol 5 * Double.MinPositiveValue) - - assert(-Double.MinPositiveValue ~== 3 * Double.MinPositiveValue absTol 5 * Double.MinPositiveValue) - assert(Double.MinPositiveValue !~== -4 * Double.MinPositiveValue absTol 5 * Double.MinPositiveValue) + assert( + Double.MinPositiveValue ~== 4 * Double.MinPositiveValue absTol 5 * Double.MinPositiveValue) + assert( + Double.MinPositiveValue !~== 6 * Double.MinPositiveValue absTol 5 * Double.MinPositiveValue) + + assert( + -Double.MinPositiveValue ~== 3 * Double.MinPositiveValue absTol 5 * Double.MinPositiveValue) + assert( + Double.MinPositiveValue !~== -4 * Double.MinPositiveValue absTol 5 * Double.MinPositiveValue) } test("Comparing vectors using relative error.") { - //Comparisons of two dense vectors + // Comparisons of two dense vectors assert(Vectors.dense(Array(3.1, 3.5)) ~== Vectors.dense(Array(3.130, 3.534)) relTol 0.01) assert(Vectors.dense(Array(3.1, 3.5)) !~== Vectors.dense(Array(3.135, 3.534)) relTol 0.01) assert(Vectors.dense(Array(3.1, 3.5)) ~= Vectors.dense(Array(3.130, 3.534)) relTol 0.01) @@ -130,7 +134,7 @@ class TestingUtilsSuite extends FunSuite { test("Comparing vectors using absolute error.") { - //Comparisons of two dense vectors + // Comparisons of two dense vectors assert(Vectors.dense(Array(3.1, 3.5, 0.0)) ~== Vectors.dense(Array(3.1 + 1E-8, 3.5 + 2E-7, 1E-8)) absTol 1E-6) -- cgit v1.2.3