fixed spelling errors in variable names

src/test/scala/au/csiro/variantspark/algo/ClassificattionSplitterTest.scala → src/test/scala/au/csiro/variantspark/algo/ClassificationSplitterTest.scala

@@ -6,7 +6,7 @@ import org.junit.Test;
 import org.apache.spark.mllib.linalg.Vectors
 import au.csiro.pbdava.ssparkle.common.utils.Logging
 
-abstract class ClassificattionSplitterTest extends Logging {
+abstract class ClassificationSplitterTest extends Logging {
 
   def splitter(labels: Array[Int], nLabels: Int = 2): ClassificationSplitter
 
@@ -24,13 +24,13 @@ abstract class ClassificattionSplitterTest extends Logging {
 
 
   @Test
-  def testConstansLabelSplit() {
+  def testConstantsLabelSplit() {
     val splitInfo = splitter(Array(1, 1, 1, 1)).findSplit(Vectors.dense(0.0, 1.0, 2.0, 3.0).toArray, Range(0, 4).toArray)
     assertEquals(SplitInfo(0, 0.0, 0.0, 0.0), splitInfo)
   }
 
   @Test
-  def testConstantsValuesSplist() {
+  def testConstantsValuesSplit() {
     val splitInfo = splitter(Array(0, 1, 0, 1)).findSplit(Vectors.dense(1.0, 1.0, 1.0, 1.0).toArray, Range(0, 4).toArray)
     assertNull(splitInfo)
   }
@@ -60,16 +60,16 @@ abstract class ClassificattionSplitterTest extends Logging {
 
 }
 
-class JClassificationSplitterTest extends ClassificattionSplitterTest {
+class JClassificationSplitterTest extends ClassificationSplitterTest {
   def splitter(labels: Array[Int], nLabels: Int = 2) = new JClassificationSplitter(labels, nLabels, 4)
 }
 
 
-class JClassificationSplitterUnboundedTest extends ClassificattionSplitterTest {
+class JClassificationSplitterUnboundedTest extends ClassificationSplitterTest {
   def splitter(labels: Array[Int], nLabels: Int = 2) = new JClassificationSplitter(labels, nLabels)
 }
 
-class JConfusionClassificationSplitterTest extends ClassificattionSplitterTest {
+class JConfusionClassificationSplitterTest extends ClassificationSplitterTest {
   def splitter(labels: Array[Int], nLabels: Int = 2) = new JConfusionClassificationSplitter(labels, nLabels, 4)
 }
 

src/test/scala/au/csiro/variantspark/algo/PairWiseDistanceTest.scala

@@ -35,7 +35,7 @@ class PairWiseDistanceTest extends SparkTest {
 
 
   @Test
-  def testCorrectlCalculatesPairWiseDistance2D() {
+  def testCorrectlyCalculatesPairWiseDistance2D() {
     val input = sc.parallelize(List(Array[Byte](0, 1), Array[Byte](0, 2), Array[Byte](1, 1)))
     val result = PairwiseDistance().compute(input)
     assertEquals(1, result.length)
@@ -44,7 +44,7 @@ class PairWiseDistanceTest extends SparkTest {
 
 
   @Test
-  def testCorrectlCalculatesPairWiseDistance3d() {
+  def testCorrectlyCalculatesPairWiseDistance3d() {
     val input = sc.parallelize(List(Array[Byte](0, 1, 1), Array[Byte](0, 2, 0), Array[Byte](0, 1, 0), Array[Byte](0, 2, 1)), 2)
     val result = PairwiseDistance().compute(input)
     assertEquals(3, result.length)

src/test/scala/au/csiro/variantspark/algo/TreeDataCollector.scala

@@ -27,8 +27,8 @@ class TreeDataCollector(treeStream: Stream[PredictiveModelWithImportance[Vector]
   }
 
   override def batchPredict(indexedData: RDD[(Vector, Long)], models: Seq[PredictiveModelWithImportance[Vector]], indexes: Seq[Array[Int]]): Seq[Array[Int]] = {
-    //TODO I should be prjecting with indexes here
-    //but it doed not matter in this case
+    //TODO I should be projecting with indexes here
+    //but it does not matter in this case
     models.zip(indexes).map { case (model, indexes) => model.predictIndexed(indexedData) }
   }
 

src/test/scala/au/csiro/variantspark/algo/WideDecisionTreeIntegratedTest.scala

@@ -46,7 +46,7 @@ class WideDecisionTreeIntegratedTest extends SparkTest {
     assertArrayEquals(expected, prediction)
 
 
-    // check variable imporcances
+    // check variable importances
     val expectedImportances = CsvParser.parse(CsvFile("src/test/data/CNAE-9_R_importance.csv")).withRowIndex(0).withColIndex(0)
       .firstCol(s"maxdepth_${maxDepth}").mapValues(CsvParser.parseDouble).values.toSeq.toArray
 

src/test/scala/au/csiro/variantspark/algo/WideDecisionTreeModelTest.scala

@@ -13,7 +13,7 @@ class WideDecisionTreeModelTest extends SparkTest {
 
   @Test
   def testCorrectlyPredictsComplexTree() {
-    // lets build a tree with two variables and 5 nodes
+    // let's build a tree with 2 variables and 5 nodes
     val decisionTreeModel = new WideDecisionTreeModel(
       SplitNode(majorityLabel = 0, size = 10, nodeImpurity = 0.0, splitVariableIndex = 1L, splitPoint = 1.0, impurityReduction = 0.0,
         left = LeafNode(1, 0, 0.0),
@@ -33,7 +33,7 @@ class WideDecisionTreeModelTest extends SparkTest {
 
   @Test
   def testCorrectlyIdentifiedVariabelImportanceForComplexTree() {
-    // lets build a tree with two variables and 5 nodes
+    // let's build a tree with 2 variables and 5 nodes
     val decisionTreeModel = new WideDecisionTreeModel(
       SplitNode(majorityLabel = 0, size = 10, nodeImpurity = 1.0, splitVariableIndex = 1L, splitPoint = 1.0, impurityReduction = 0.0,
         left = SplitNode(majorityLabel = 0, size = 4, nodeImpurity = 0.4, splitVariableIndex = 2L, splitPoint = 0.0, impurityReduction = 0.0,

src/test/scala/au/csiro/variantspark/algo/WideRadomForrestModelTest.scala

@@ -34,16 +34,16 @@ class WideRadomForrestModelTest extends SparkTest {
 
   @Test
   def whenOnePredictorPassesThePrediction() {
-    val assumedPreditions = Array(1, 2)
-    val model = new WideRandomForestModel(List(TestPredictorWithImportance(assumedPreditions, null).toMember), nLabels)
+    val assumedPredictions = Array(1, 2)
+    val model = new WideRandomForestModel(List(TestPredictorWithImportance(assumedPredictions, null).toMember), nLabels)
     val prediction = model.predict(testData)
-    assertArrayEquals(assumedPreditions, prediction)
+    assertArrayEquals(assumedPredictions, prediction)
   }
 
   @Test
-  def whenManyPreditorsThenPredictsByVoting() {
-    val assumedPreditions = List(Array(1, 0), Array(1, 2), Array(1, 0))
-    val model = new WideRandomForestModel(assumedPreditions.map(TestPredictorWithImportance(_, null).toMember).toList, nLabels)
+  def whenManyPredictorsThenPredictsByVoting() {
+    val assumedPredictions = List(Array(1, 0), Array(1, 2), Array(1, 0))
+    val model = new WideRandomForestModel(assumedPredictions.map(TestPredictorWithImportance(_, null).toMember).toList, nLabels)
     val prediction = model.predict(testData)
     assertArrayEquals(Array(1, 0), prediction)
   }

src/test/scala/au/csiro/variantspark/input/generate/NoisyEfectLabelGeneratorTest.scala → src/test/scala/au/csiro/variantspark/input/generate/NoisyEffectLabelGeneratorTest.scala

@@ -20,14 +20,14 @@ class TestFeatureGenerator(val samples: Seq[Feature])(implicit sc: SparkContext)
 }
 
 
-class NoisyEfectLabelGeneratorTest extends SparkTest {
+class NoisyEffectLabelGeneratorTest extends SparkTest {
 
   @Test
-  def testResponseGenertion_var_0_2_prec_0_75() {
-    val freatureGenerator = OrdinalFeatureGenerator(3, 1000, 500)
-    val labelGenerator = new NoisyEfectLabelGenerator(freatureGenerator)(1, Map("v_0" -> 1.0, "v_1" -> 0.5, "v_2" -> 0.25), fractionVarianceExplained = 0.2, classThresholdPrecentile = 0.75)
-    val classes = labelGenerator.getLabels(freatureGenerator.sampleNames)
-    // we wouild expect 75% of samples in class 0
+  def testResponseGeneration_var_0_2_prec_0_75() {
+    val featureGenerator = OrdinalFeatureGenerator(3, 1000, 500)
+    val labelGenerator = new NoisyEfectLabelGenerator(featureGenerator)(1, Map("v_0" -> 1.0, "v_1" -> 0.5, "v_2" -> 0.25), fractionVarianceExplained = 0.2, classThresholdPrecentile = 0.75)
+    val classes = labelGenerator.getLabels(featureGenerator.sampleNames)
+    // we would expect 75% of samples in class 0
     assertEquals(0.75, classes.count(_ == 0).toDouble / classes.size, 0.01)
 
     val baseVariance = meanAndVariance(labelGenerator.baseContinuousResponse).variance
@@ -38,11 +38,11 @@ class NoisyEfectLabelGeneratorTest extends SparkTest {
 
   @Test
   def testMultiplicativeResponseGeneration_var_0_2_prec_0_75() {
-    val freatureGenerator = OrdinalFeatureGenerator(3, 1000, 500)
-    val labelGenerator = new NoisyEfectLabelGenerator(freatureGenerator)(1, Map("v_0" -> 1.0, "v_1" -> 0.5, "v_2" -> 0.25),
+    val featureGenerator = OrdinalFeatureGenerator(3, 1000, 500)
+    val labelGenerator = new NoisyEfectLabelGenerator(featureGenerator)(1, Map("v_0" -> 1.0, "v_1" -> 0.5, "v_2" -> 0.25),
       fractionVarianceExplained = 0.2, classThresholdPrecentile = 0.75, multiplicative = true)
-    val classes = labelGenerator.getLabels(freatureGenerator.sampleNames)
-    // we wouild expect 75% of samples in class 0
+    val classes = labelGenerator.getLabels(featureGenerator.sampleNames)
+    // we would expect 75% of samples in class 0
     assertEquals(0.75, classes.count(_ == 0).toDouble / classes.size, 0.01)
 
     val baseVariance = meanAndVariance(labelGenerator.baseContinuousResponse).variance
@@ -52,40 +52,40 @@ class NoisyEfectLabelGeneratorTest extends SparkTest {
 
   @Test
   def testAdditiveEffectCorrectness() {
-    val freatureGenerator = new TestFeatureGenerator(List(
+    val featureGenerator = new TestFeatureGenerator(List(
       Feature("v_0", Array[Byte](0, 1, 2, 0)),
       Feature("v_1", Array[Byte](0, 1, 2, 1)),
       Feature("v_2", Array[Byte](0, 1, 2, 2)),
       Feature("v_3", Array[Byte](2, 2, 2, 2))
     ))
-    val labelGenerator = new NoisyEfectLabelGenerator(freatureGenerator)(1, Map("v_0" -> 0.1, "v_1" -> 0.5, "v_2" -> 2.0),
+    val labelGenerator = new NoisyEfectLabelGenerator(featureGenerator)(1, Map("v_0" -> 0.1, "v_1" -> 0.5, "v_2" -> 2.0),
       fractionVarianceExplained = 0.2, classThresholdPrecentile = 0.75, multiplicative = false)
 
-    val classes = labelGenerator.getLabels(freatureGenerator.sampleNames)
+    val classes = labelGenerator.getLabels(featureGenerator.sampleNames)
     assertEquals(DenseVector[Double](-2.6, 0, 2.6, 1.9), labelGenerator.baseContinuousResponse)
   }
 
   @Test
-  def testMultiptlicativeEffectCorrectness() {
-    val freatureGenerator = new TestFeatureGenerator(List(
+  def testMultiplicativeEffectCorrectness() {
+    val featureGenerator = new TestFeatureGenerator(List(
       Feature("v_0", Array[Byte](0, 1, 2, 0)),
       Feature("v_1", Array[Byte](0, 1, 2, 1)),
       Feature("v_2", Array[Byte](0, 1, 2, 2)),
       Feature("v_3", Array[Byte](2, 2, 2, 2))
     ))
-    val labelGenerator = new NoisyEfectLabelGenerator(freatureGenerator)(1, Map("v_0" -> 0.1, "v_1" -> 0.5, "v_2" -> 2.0),
+    val labelGenerator = new NoisyEfectLabelGenerator(featureGenerator)(1, Map("v_0" -> 0.1, "v_1" -> 0.5, "v_2" -> 2.0),
       fractionVarianceExplained = 0.2, classThresholdPrecentile = 0.75, multiplicative = true)
 
-    val classes = labelGenerator.getLabels(freatureGenerator.sampleNames)
+    val classes = labelGenerator.getLabels(featureGenerator.sampleNames)
     assertEquals(DenseVector[Double](-0.1, 1.0, 0.1, -0.2), labelGenerator.baseContinuousResponse)
   }
 
 
   @Test
-  def testResponseGenertion_var_0_5_prec_0_50() {
-    val freatureGenerator = OrdinalFeatureGenerator(3, 1000, 500)
-    val labelGenerator = new NoisyEfectLabelGenerator(freatureGenerator)(1, Map("v_0" -> 1.0, "v_1" -> 0.5, "v_2" -> 0.25), fractionVarianceExplained = 0.5, classThresholdPrecentile = 0.5)
-    val classes = labelGenerator.getLabels(freatureGenerator.sampleNames)
+  def testResponseGeneration_var_0_5_prec_0_50() {
+    val featureGenerator = OrdinalFeatureGenerator(3, 1000, 500)
+    val labelGenerator = new NoisyEfectLabelGenerator(featureGenerator)(1, Map("v_0" -> 1.0, "v_1" -> 0.5, "v_2" -> 0.25), fractionVarianceExplained = 0.5, classThresholdPrecentile = 0.5)
+    val classes = labelGenerator.getLabels(featureGenerator.sampleNames)
     // we wouild expect 75% of samples in class 0
     assertEquals(0.5, classes.count(_ == 0).toDouble / classes.size, 0.01)
 

src/test/scala/au/csiro/variantspark/utils/SampleTest.scala

@@ -33,7 +33,7 @@ class SampleTest {
   }
 
   @Test
-  def testFreactionSampleWithoutReplacement() {
+  def testFractionSampleWithoutReplacement() {
     val nSize = 100
     val fraction = 0.5
     val sample = Sample.fraction(nSize, fraction, false)
@@ -47,7 +47,7 @@ class SampleTest {
 
 
   @Test
-  def testFreactionSampleWithReplacement() {
+  def testFractionSampleWithReplacement() {
     val nSize = 100
     val fraction = 0.5
     val sample = Sample.fraction(nSize, fraction, true)