From ccfa362ddec1bc942785798dea41c3aac52df60f Mon Sep 17 00:00:00 2001 From: Xinghao <pxinghao@gmail.com> Date: Sun, 28 Jul 2013 10:33:57 -0700 Subject: [PATCH] Change *_LocalRandomSGD to *LocalRandomSGD --- .../classification/LogisticRegression.scala | 10 ++++---- .../spark/mllib/classification/SVM.scala | 10 ++++---- .../scala/spark/mllib/regression/Lasso.scala | 10 ++++---- .../LogisticRegressionSuite.scala | 23 +++---------------- .../spark/mllib/classification/SVMSuite.scala | 4 ++-- .../spark/mllib/regression/LassoSuite.scala | 8 +++---- 6 files changed, 24 insertions(+), 41 deletions(-) diff --git a/mllib/src/main/scala/spark/mllib/classification/LogisticRegression.scala b/mllib/src/main/scala/spark/mllib/classification/LogisticRegression.scala index 40b96fbe3a..1b093187f2 100644 --- a/mllib/src/main/scala/spark/mllib/classification/LogisticRegression.scala +++ b/mllib/src/main/scala/spark/mllib/classification/LogisticRegression.scala @@ -53,7 +53,7 @@ class LogisticRegressionModel( } } -class LogisticRegression_LocalRandomSGD private (var stepSize: Double, var miniBatchFraction: Double, +class LogisticRegressionLocalRandomSGD private (var stepSize: Double, var miniBatchFraction: Double, var numIters: Int) extends Logging { @@ -138,7 +138,7 @@ class LogisticRegression_LocalRandomSGD private (var stepSize: Double, var miniB * NOTE(shivaram): We use multiple train methods instead of default arguments to support * Java programs. */ -object LogisticRegression_LocalRandomSGD { +object LogisticRegressionLocalRandomSGD { /** * Train a logistic regression model given an RDD of (label, features) pairs. We run a fixed number @@ -163,7 +163,7 @@ object LogisticRegression_LocalRandomSGD { initialWeights: Array[Double]) : LogisticRegressionModel = { - new LogisticRegression_LocalRandomSGD(stepSize, miniBatchFraction, numIterations).train(input, initialWeights) + new LogisticRegressionLocalRandomSGD(stepSize, miniBatchFraction, numIterations).train(input, initialWeights) } /** @@ -185,7 +185,7 @@ object LogisticRegression_LocalRandomSGD { miniBatchFraction: Double) : LogisticRegressionModel = { - new LogisticRegression_LocalRandomSGD(stepSize, miniBatchFraction, numIterations).train(input) + new LogisticRegressionLocalRandomSGD(stepSize, miniBatchFraction, numIterations).train(input) } /** @@ -233,7 +233,7 @@ object LogisticRegression_LocalRandomSGD { } val sc = new SparkContext(args(0), "LogisticRegression") val data = MLUtils.loadLabeledData(sc, args(1)) - val model = LogisticRegression_LocalRandomSGD.train(data, args(4).toInt, args(2).toDouble, args(3).toDouble) + val model = LogisticRegressionLocalRandomSGD.train(data, args(4).toInt, args(2).toDouble, args(3).toDouble) sc.stop() } diff --git a/mllib/src/main/scala/spark/mllib/classification/SVM.scala b/mllib/src/main/scala/spark/mllib/classification/SVM.scala index 2cd1d668eb..76844f6b9c 100644 --- a/mllib/src/main/scala/spark/mllib/classification/SVM.scala +++ b/mllib/src/main/scala/spark/mllib/classification/SVM.scala @@ -53,7 +53,7 @@ class SVMModel( -class SVM_LocalRandomSGD private (var stepSize: Double, var regParam: Double, var miniBatchFraction: Double, +class SVMLocalRandomSGD private (var stepSize: Double, var regParam: Double, var miniBatchFraction: Double, var numIters: Int) extends Logging { @@ -138,7 +138,7 @@ class SVM_LocalRandomSGD private (var stepSize: Double, var regParam: Double, va */ -object SVM_LocalRandomSGD { +object SVMLocalRandomSGD { /** * Train a SVM model given an RDD of (label, features) pairs. We run a fixed number @@ -163,7 +163,7 @@ object SVM_LocalRandomSGD { initialWeights: Array[Double]) : SVMModel = { - new SVM_LocalRandomSGD(stepSize, regParam, miniBatchFraction, numIterations).train(input, initialWeights) + new SVMLocalRandomSGD(stepSize, regParam, miniBatchFraction, numIterations).train(input, initialWeights) } /** @@ -185,7 +185,7 @@ object SVM_LocalRandomSGD { miniBatchFraction: Double) : SVMModel = { - new SVM_LocalRandomSGD(stepSize, regParam, miniBatchFraction, numIterations).train(input) + new SVMLocalRandomSGD(stepSize, regParam, miniBatchFraction, numIterations).train(input) } /** @@ -233,7 +233,7 @@ object SVM_LocalRandomSGD { } val sc = new SparkContext(args(0), "SVM") val data = MLUtils.loadLabeledData(sc, args(1)) - val model = SVM_LocalRandomSGD.train(data, args(4).toInt, args(2).toDouble, args(3).toDouble) + val model = SVMLocalRandomSGD.train(data, args(4).toInt, args(2).toDouble, args(3).toDouble) sc.stop() } diff --git a/mllib/src/main/scala/spark/mllib/regression/Lasso.scala b/mllib/src/main/scala/spark/mllib/regression/Lasso.scala index 64364323a2..1952658bb2 100644 --- a/mllib/src/main/scala/spark/mllib/regression/Lasso.scala +++ b/mllib/src/main/scala/spark/mllib/regression/Lasso.scala @@ -53,7 +53,7 @@ class LassoModel( } -class Lasso_LocalRandomSGD private (var stepSize: Double, var regParam: Double, var miniBatchFraction: Double, +class LassoLocalRandomSGD private (var stepSize: Double, var regParam: Double, var miniBatchFraction: Double, var numIters: Int) extends Logging { @@ -138,7 +138,7 @@ class Lasso_LocalRandomSGD private (var stepSize: Double, var regParam: Double, * * */ -object Lasso_LocalRandomSGD { +object LassoLocalRandomSGD { /** * Train a Lasso model given an RDD of (label, features) pairs. We run a fixed number @@ -163,7 +163,7 @@ object Lasso_LocalRandomSGD { initialWeights: Array[Double]) : LassoModel = { - new Lasso_LocalRandomSGD(stepSize, regParam, miniBatchFraction, numIterations).train(input, initialWeights) + new LassoLocalRandomSGD(stepSize, regParam, miniBatchFraction, numIterations).train(input, initialWeights) } /** @@ -185,7 +185,7 @@ object Lasso_LocalRandomSGD { miniBatchFraction: Double) : LassoModel = { - new Lasso_LocalRandomSGD(stepSize, regParam, miniBatchFraction, numIterations).train(input) + new LassoLocalRandomSGD(stepSize, regParam, miniBatchFraction, numIterations).train(input) } /** @@ -233,7 +233,7 @@ object Lasso_LocalRandomSGD { } val sc = new SparkContext(args(0), "Lasso") val data = MLUtils.loadLabeledData(sc, args(1)) - val model = Lasso_LocalRandomSGD.train(data, args(4).toInt, args(2).toDouble, args(3).toDouble) + val model = LassoLocalRandomSGD.train(data, args(4).toInt, args(2).toDouble, args(3).toDouble) sc.stop() } diff --git a/mllib/src/test/scala/spark/mllib/classification/LogisticRegressionSuite.scala b/mllib/src/test/scala/spark/mllib/classification/LogisticRegressionSuite.scala index 827ca66330..144b8b1bc7 100644 --- a/mllib/src/test/scala/spark/mllib/classification/LogisticRegressionSuite.scala +++ b/mllib/src/test/scala/spark/mllib/classification/LogisticRegressionSuite.scala @@ -1,6 +1,3 @@ -<<<<<<< HEAD:mllib/src/test/scala/spark/mllib/classification/LogisticRegressionSuite.scala -package spark.mllib.classification -======= /* * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with @@ -18,8 +15,7 @@ package spark.mllib.classification * limitations under the License. */ -package spark.mllib.regression ->>>>>>> FETCH_HEAD:mllib/src/test/scala/spark/mllib/regression/LogisticRegressionSuite.scala +package spark.mllib.classification import scala.util.Random @@ -37,13 +33,6 @@ class LogisticRegressionSuite extends FunSuite with BeforeAndAfterAll { System.clearProperty("spark.driver.port") } -<<<<<<< HEAD:mllib/src/test/scala/spark/mllib/classification/LogisticRegressionSuite.scala - // Test if we can correctly learn A, B where Y = logistic(A + B*X) - test("LogisticRegression_LocalRandomSGD") { - val nPoints = 10000 - val rnd = new Random(42) - -======= // Generate input of the form Y = logistic(offset + scale*X) def generateLogisticInput( offset: Double, @@ -51,7 +40,6 @@ class LogisticRegressionSuite extends FunSuite with BeforeAndAfterAll { nPoints: Int, seed: Int): Seq[(Double, Array[Double])] = { val rnd = new Random(seed) ->>>>>>> FETCH_HEAD:mllib/src/test/scala/spark/mllib/regression/LogisticRegressionSuite.scala val x1 = Array.fill[Double](nPoints)(rnd.nextGaussian()) // NOTE: if U is uniform[0, 1] then ln(u) - ln(1-u) is Logistic(0,1) @@ -91,12 +79,7 @@ class LogisticRegressionSuite extends FunSuite with BeforeAndAfterAll { val testRDD = sc.parallelize(testData, 2) testRDD.cache() -<<<<<<< HEAD:mllib/src/test/scala/spark/mllib/classification/LogisticRegressionSuite.scala - val lr = new LogisticRegression_LocalRandomSGD().setStepSize(10.0) - .setNumIterations(20) -======= - val lr = new LogisticRegression().setStepSize(10.0).setNumIterations(20) ->>>>>>> FETCH_HEAD:mllib/src/test/scala/spark/mllib/regression/LogisticRegressionSuite.scala + val lr = new LogisticRegressionLocalRandomSGD().setStepSize(10.0).setNumIterations(20) val model = lr.train(testRDD) @@ -128,7 +111,7 @@ class LogisticRegressionSuite extends FunSuite with BeforeAndAfterAll { testRDD.cache() // Use half as many iterations as the previous test. - val lr = new LogisticRegression().setStepSize(10.0).setNumIterations(10) + val lr = new LogisticRegressionLocalRandomSGD().setStepSize(10.0).setNumIterations(10) val model = lr.train(testRDD, initialWeights) diff --git a/mllib/src/test/scala/spark/mllib/classification/SVMSuite.scala b/mllib/src/test/scala/spark/mllib/classification/SVMSuite.scala index 50cf260f49..0d781c310c 100644 --- a/mllib/src/test/scala/spark/mllib/classification/SVMSuite.scala +++ b/mllib/src/test/scala/spark/mllib/classification/SVMSuite.scala @@ -19,7 +19,7 @@ class SVMSuite extends FunSuite with BeforeAndAfterAll { System.clearProperty("spark.driver.port") } - test("SVM_LocalRandomSGD") { + test("SVMLocalRandomSGD") { val nPoints = 10000 val rnd = new Random(42) @@ -46,7 +46,7 @@ class SVMSuite extends FunSuite with BeforeAndAfterAll { writer_data.write("\n")}) writer_data.close() - val svm = new SVM_LocalRandomSGD().setStepSize(1.0) + val svm = new SVMLocalRandomSGD().setStepSize(1.0) .setRegParam(1.0) .setNumIterations(100) diff --git a/mllib/src/test/scala/spark/mllib/regression/LassoSuite.scala b/mllib/src/test/scala/spark/mllib/regression/LassoSuite.scala index 9836ac54c1..0c39e1e09b 100644 --- a/mllib/src/test/scala/spark/mllib/regression/LassoSuite.scala +++ b/mllib/src/test/scala/spark/mllib/regression/LassoSuite.scala @@ -17,7 +17,7 @@ class LassoSuite extends FunSuite with BeforeAndAfterAll { System.clearProperty("spark.driver.port") } - test("Lasso_LocalRandomSGD") { + test("LassoLocalRandomSGD") { val nPoints = 10000 val rnd = new Random(42) @@ -36,14 +36,14 @@ class LassoSuite extends FunSuite with BeforeAndAfterAll { val testRDD = sc.parallelize(testData, 2) testRDD.cache() - val ls = new Lasso_LocalRandomSGD().setStepSize(1.0) + val ls = new LassoLocalRandomSGD().setStepSize(1.0) .setRegParam(0.01) .setNumIterations(20) val model = ls.train(testRDD) - val weight0 = model.weights.get(0) - val weight1 = model.weights.get(1) + val weight0 = model.weights(0) + val weight1 = model.weights(1) assert(weight0 >= -1.60 && weight0 <= -1.40, weight0 + " not in [-1.6, -1.4]") assert(weight1 >= -1.0e-3 && weight1 <= 1.0e-3, weight1 + " not in [-0.001, 0.001]") assert(model.intercept >= 1.9 && model.intercept <= 2.1, model.intercept + " not in [1.9, 2.1]") -- GitLab