diff options
author | Xiangrui Meng <meng@databricks.com> | 2015-03-12 01:39:04 -0700 |
---|---|---|
committer | Xiangrui Meng <meng@databricks.com> | 2015-03-12 01:39:04 -0700 |
commit | 0cba802adf15f5ab8da24dd1e8a5e7214cc4e148 (patch) | |
tree | a5dc95dc4f0fbf156cacffc6af130666711b3f98 /mllib | |
parent | 712679a7b447346a365b38574d7a86d56a93f767 (diff) | |
download | spark-0cba802adf15f5ab8da24dd1e8a5e7214cc4e148.tar.gz spark-0cba802adf15f5ab8da24dd1e8a5e7214cc4e148.tar.bz2 spark-0cba802adf15f5ab8da24dd1e8a5e7214cc4e148.zip |
[SPARK-5814][MLLIB][GRAPHX] Remove JBLAS from runtime
The issue is discussed in https://issues.apache.org/jira/browse/SPARK-5669. Replacing all JBLAS usage by netlib-java gives us a simpler dependency tree and less license issues to worry about. I didn't touch the test scope in this PR. The user guide is not modified to avoid merge conflicts with branch-1.3. srowen ankurdave pwendell
Author: Xiangrui Meng <meng@databricks.com>
Closes #4699 from mengxr/SPARK-5814 and squashes the following commits:
48635c6 [Xiangrui Meng] move netlib-java version to parent pom
ca21c74 [Xiangrui Meng] remove jblas from ml-guide
5f7767a [Xiangrui Meng] Merge remote-tracking branch 'apache/master' into SPARK-5814
c5c4183 [Xiangrui Meng] merge master
0f20cad [Xiangrui Meng] add mima excludes
e53e9f4 [Xiangrui Meng] remove jblas from mllib runtime
ceaa14d [Xiangrui Meng] replace jblas by netlib-java in graphx
fa7c2ca [Xiangrui Meng] move jblas to test scope
Diffstat (limited to 'mllib')
9 files changed, 84 insertions, 86 deletions
diff --git a/mllib/pom.xml b/mllib/pom.xml index b5c949e155..a76704a8c2 100644 --- a/mllib/pom.xml +++ b/mllib/pom.xml @@ -59,6 +59,7 @@ <groupId>org.jblas</groupId> <artifactId>jblas</artifactId> <version>${jblas.version}</version> + <scope>test</scope> </dependency> <dependency> <groupId>org.scalanlp</groupId> @@ -116,7 +117,7 @@ <dependency> <groupId>com.github.fommil.netlib</groupId> <artifactId>all</artifactId> - <version>1.1.2</version> + <version>${netlib.java.version}</version> <type>pom</type> </dependency> </dependencies> diff --git a/mllib/src/main/scala/org/apache/spark/ml/recommendation/ALS.scala b/mllib/src/main/scala/org/apache/spark/ml/recommendation/ALS.scala index 7bb69df653..e3515ee81a 100644 --- a/mllib/src/main/scala/org/apache/spark/ml/recommendation/ALS.scala +++ b/mllib/src/main/scala/org/apache/spark/ml/recommendation/ALS.scala @@ -26,7 +26,6 @@ import scala.util.hashing.byteswap64 import com.github.fommil.netlib.BLAS.{getInstance => blas} import com.github.fommil.netlib.LAPACK.{getInstance => lapack} -import org.jblas.DoubleMatrix import org.netlib.util.intW import org.apache.spark.{Logging, Partitioner} @@ -361,14 +360,14 @@ object ALS extends Logging { private[recommendation] class NNLSSolver extends LeastSquaresNESolver { private var rank: Int = -1 private var workspace: NNLS.Workspace = _ - private var ata: DoubleMatrix = _ + private var ata: Array[Double] = _ private var initialized: Boolean = false private def initialize(rank: Int): Unit = { if (!initialized) { this.rank = rank workspace = NNLS.createWorkspace(rank) - ata = new DoubleMatrix(rank, rank) + ata = new Array[Double](rank * rank) initialized = true } else { require(this.rank == rank) @@ -385,7 +384,7 @@ object ALS extends Logging { val rank = ne.k initialize(rank) fillAtA(ne.ata, lambda * ne.n) - val x = NNLS.solve(ata, new DoubleMatrix(rank, 1, ne.atb: _*), workspace) + val x = NNLS.solve(ata, ne.atb, workspace) ne.reset() x.map(x => x.toFloat) } @@ -398,17 +397,16 @@ object ALS extends Logging { var i = 0 var pos = 0 var a = 0.0 - val data = ata.data while (i < rank) { var j = 0 while (j <= i) { a = triAtA(pos) - data(i * rank + j) = a - data(j * rank + i) = a + ata(i * rank + j) = a + ata(j * rank + i) = a pos += 1 j += 1 } - data(i * rank + i) += lambda + ata(i * rank + i) += lambda i += 1 } } diff --git a/mllib/src/main/scala/org/apache/spark/mllib/optimization/NNLS.scala b/mllib/src/main/scala/org/apache/spark/mllib/optimization/NNLS.scala index ccd93b318b..4766f77082 100644 --- a/mllib/src/main/scala/org/apache/spark/mllib/optimization/NNLS.scala +++ b/mllib/src/main/scala/org/apache/spark/mllib/optimization/NNLS.scala @@ -17,7 +17,9 @@ package org.apache.spark.mllib.optimization -import org.jblas.{DoubleMatrix, SimpleBlas} +import java.{util => ju} + +import com.github.fommil.netlib.BLAS.{getInstance => blas} /** * Object used to solve nonnegative least squares problems using a modified @@ -25,20 +27,20 @@ import org.jblas.{DoubleMatrix, SimpleBlas} */ private[spark] object NNLS { class Workspace(val n: Int) { - val scratch = new DoubleMatrix(n, 1) - val grad = new DoubleMatrix(n, 1) - val x = new DoubleMatrix(n, 1) - val dir = new DoubleMatrix(n, 1) - val lastDir = new DoubleMatrix(n, 1) - val res = new DoubleMatrix(n, 1) - - def wipe() { - scratch.fill(0.0) - grad.fill(0.0) - x.fill(0.0) - dir.fill(0.0) - lastDir.fill(0.0) - res.fill(0.0) + val scratch = new Array[Double](n) + val grad = new Array[Double](n) + val x = new Array[Double](n) + val dir = new Array[Double](n) + val lastDir = new Array[Double](n) + val res = new Array[Double](n) + + def wipe(): Unit = { + ju.Arrays.fill(scratch, 0.0) + ju.Arrays.fill(grad, 0.0) + ju.Arrays.fill(x, 0.0) + ju.Arrays.fill(dir, 0.0) + ju.Arrays.fill(lastDir, 0.0) + ju.Arrays.fill(res, 0.0) } } @@ -60,18 +62,18 @@ private[spark] object NNLS { * direction, however, while this method only uses a conjugate gradient direction if the last * iteration did not cause a previously-inactive constraint to become active. */ - def solve(ata: DoubleMatrix, atb: DoubleMatrix, ws: Workspace): Array[Double] = { + def solve(ata: Array[Double], atb: Array[Double], ws: Workspace): Array[Double] = { ws.wipe() - val n = atb.rows + val n = atb.length val scratch = ws.scratch // find the optimal unconstrained step - def steplen(dir: DoubleMatrix, res: DoubleMatrix): Double = { - val top = SimpleBlas.dot(dir, res) - SimpleBlas.gemv(1.0, ata, dir, 0.0, scratch) + def steplen(dir: Array[Double], res: Array[Double]): Double = { + val top = blas.ddot(n, dir, 1, res, 1) + blas.dgemv("N", n, n, 1.0, ata, n, dir, 1, 0.0, scratch, 1) // Push the denominator upward very slightly to avoid infinities and silliness - top / (SimpleBlas.dot(scratch, dir) + 1e-20) + top / (blas.ddot(n, scratch, 1, dir, 1) + 1e-20) } // stopping condition @@ -96,52 +98,52 @@ private[spark] object NNLS { var i = 0 while (iterno < iterMax) { // find the residual - SimpleBlas.gemv(1.0, ata, x, 0.0, res) - SimpleBlas.axpy(-1.0, atb, res) - SimpleBlas.copy(res, grad) + blas.dgemv("N", n, n, 1.0, ata, n, x, 1, 0.0, res, 1) + blas.daxpy(n, -1.0, atb, 1, res, 1) + blas.dcopy(n, res, 1, grad, 1) // project the gradient i = 0 while (i < n) { - if (grad.data(i) > 0.0 && x.data(i) == 0.0) { - grad.data(i) = 0.0 + if (grad(i) > 0.0 && x(i) == 0.0) { + grad(i) = 0.0 } i = i + 1 } - val ngrad = SimpleBlas.dot(grad, grad) + val ngrad = blas.ddot(n, grad, 1, grad, 1) - SimpleBlas.copy(grad, dir) + blas.dcopy(n, grad, 1, dir, 1) // use a CG direction under certain conditions var step = steplen(grad, res) var ndir = 0.0 - val nx = SimpleBlas.dot(x, x) + val nx = blas.ddot(n, x, 1, x, 1) if (iterno > lastWall + 1) { val alpha = ngrad / lastNorm - SimpleBlas.axpy(alpha, lastDir, dir) + blas.daxpy(n, alpha, lastDir, 1, dir, 1) val dstep = steplen(dir, res) - ndir = SimpleBlas.dot(dir, dir) + ndir = blas.ddot(n, dir, 1, dir, 1) if (stop(dstep, ndir, nx)) { // reject the CG step if it could lead to premature termination - SimpleBlas.copy(grad, dir) - ndir = SimpleBlas.dot(dir, dir) + blas.dcopy(n, grad, 1, dir, 1) + ndir = blas.ddot(n, dir, 1, dir, 1) } else { step = dstep } } else { - ndir = SimpleBlas.dot(dir, dir) + ndir = blas.ddot(n, dir, 1, dir, 1) } // terminate? if (stop(step, ndir, nx)) { - return x.data.clone + return x.clone } // don't run through the walls i = 0 while (i < n) { - if (step * dir.data(i) > x.data(i)) { - step = x.data(i) / dir.data(i) + if (step * dir(i) > x(i)) { + step = x(i) / dir(i) } i = i + 1 } @@ -149,19 +151,19 @@ private[spark] object NNLS { // take the step i = 0 while (i < n) { - if (step * dir.data(i) > x.data(i) * (1 - 1e-14)) { - x.data(i) = 0 + if (step * dir(i) > x(i) * (1 - 1e-14)) { + x(i) = 0 lastWall = iterno } else { - x.data(i) -= step * dir.data(i) + x(i) -= step * dir(i) } i = i + 1 } iterno = iterno + 1 - SimpleBlas.copy(dir, lastDir) + blas.dcopy(n, dir, 1, lastDir, 1) lastNorm = ngrad } - x.data.clone + x.clone } } diff --git a/mllib/src/main/scala/org/apache/spark/mllib/recommendation/MatrixFactorizationModel.scala b/mllib/src/main/scala/org/apache/spark/mllib/recommendation/MatrixFactorizationModel.scala index 5f5a996a87..36cbf060d9 100644 --- a/mllib/src/main/scala/org/apache/spark/mllib/recommendation/MatrixFactorizationModel.scala +++ b/mllib/src/main/scala/org/apache/spark/mllib/recommendation/MatrixFactorizationModel.scala @@ -21,10 +21,10 @@ import java.io.IOException import java.lang.{Integer => JavaInteger} import org.apache.hadoop.fs.Path -import org.jblas.DoubleMatrix import org.json4s._ import org.json4s.JsonDSL._ import org.json4s.jackson.JsonMethods._ +import com.github.fommil.netlib.BLAS.{getInstance => blas} import org.apache.spark.{Logging, SparkContext} import org.apache.spark.api.java.{JavaPairRDD, JavaRDD} @@ -70,9 +70,9 @@ class MatrixFactorizationModel( /** Predict the rating of one user for one product. */ def predict(user: Int, product: Int): Double = { - val userVector = new DoubleMatrix(userFeatures.lookup(user).head) - val productVector = new DoubleMatrix(productFeatures.lookup(product).head) - userVector.dot(productVector) + val userVector = userFeatures.lookup(user).head + val productVector = productFeatures.lookup(product).head + blas.ddot(userVector.length, userVector, 1, productVector, 1) } /** @@ -89,9 +89,7 @@ class MatrixFactorizationModel( } users.join(productFeatures).map { case (product, ((user, uFeatures), pFeatures)) => - val userVector = new DoubleMatrix(uFeatures) - val productVector = new DoubleMatrix(pFeatures) - Rating(user, product, userVector.dot(productVector)) + Rating(user, product, blas.ddot(uFeatures.length, uFeatures, 1, pFeatures, 1)) } } @@ -143,9 +141,8 @@ class MatrixFactorizationModel( recommendToFeatures: Array[Double], recommendableFeatures: RDD[(Int, Array[Double])], num: Int): Array[(Int, Double)] = { - val recommendToVector = new DoubleMatrix(recommendToFeatures) val scored = recommendableFeatures.map { case (id,features) => - (id, recommendToVector.dot(new DoubleMatrix(features))) + (id, blas.ddot(features.length, recommendToFeatures, 1, features, 1)) } scored.top(num)(Ordering.by(_._2)) } diff --git a/mllib/src/main/scala/org/apache/spark/mllib/util/LinearDataGenerator.scala b/mllib/src/main/scala/org/apache/spark/mllib/util/LinearDataGenerator.scala index 97f54aa62d..c9d33787b0 100644 --- a/mllib/src/main/scala/org/apache/spark/mllib/util/LinearDataGenerator.scala +++ b/mllib/src/main/scala/org/apache/spark/mllib/util/LinearDataGenerator.scala @@ -20,7 +20,7 @@ package org.apache.spark.mllib.util import scala.collection.JavaConversions._ import scala.util.Random -import org.jblas.DoubleMatrix +import com.github.fommil.netlib.BLAS.{getInstance => blas} import org.apache.spark.annotation.DeveloperApi import org.apache.spark.SparkContext @@ -72,11 +72,10 @@ object LinearDataGenerator { eps: Double = 0.1): Seq[LabeledPoint] = { val rnd = new Random(seed) - val weightsMat = new DoubleMatrix(1, weights.length, weights:_*) val x = Array.fill[Array[Double]](nPoints)( Array.fill[Double](weights.length)(2 * rnd.nextDouble - 1.0)) val y = x.map { xi => - new DoubleMatrix(1, xi.length, xi: _*).dot(weightsMat) + intercept + eps * rnd.nextGaussian() + blas.ddot(weights.length, xi, 1, weights, 1) + intercept + eps * rnd.nextGaussian() } y.zip(x).map(p => LabeledPoint(p._1, Vectors.dense(p._2))) } @@ -100,9 +99,9 @@ object LinearDataGenerator { eps: Double, nparts: Int = 2, intercept: Double = 0.0) : RDD[LabeledPoint] = { - org.jblas.util.Random.seed(42) + val random = new Random(42) // Random values distributed uniformly in [-0.5, 0.5] - val w = DoubleMatrix.rand(nfeatures, 1).subi(0.5) + val w = Array.fill(nfeatures)(random.nextDouble() - 0.5) val data: RDD[LabeledPoint] = sc.parallelize(0 until nparts, nparts).flatMap { p => val seed = 42 + p diff --git a/mllib/src/main/scala/org/apache/spark/mllib/util/MFDataGenerator.scala b/mllib/src/main/scala/org/apache/spark/mllib/util/MFDataGenerator.scala index b76fbe89c3..0c5b4f9d04 100644 --- a/mllib/src/main/scala/org/apache/spark/mllib/util/MFDataGenerator.scala +++ b/mllib/src/main/scala/org/apache/spark/mllib/util/MFDataGenerator.scala @@ -17,13 +17,14 @@ package org.apache.spark.mllib.util +import java.{util => ju} + import scala.language.postfixOps import scala.util.Random -import org.jblas.DoubleMatrix - -import org.apache.spark.annotation.DeveloperApi import org.apache.spark.SparkContext +import org.apache.spark.annotation.DeveloperApi +import org.apache.spark.mllib.linalg.{BLAS, DenseMatrix} import org.apache.spark.rdd.RDD /** @@ -72,24 +73,25 @@ object MFDataGenerator { val sc = new SparkContext(sparkMaster, "MFDataGenerator") - val A = DoubleMatrix.randn(m, rank) - val B = DoubleMatrix.randn(rank, n) - val z = 1 / scala.math.sqrt(scala.math.sqrt(rank)) - A.mmuli(z) - B.mmuli(z) - val fullData = A.mmul(B) + val random = new ju.Random(42L) + + val A = DenseMatrix.randn(m, rank, random) + val B = DenseMatrix.randn(rank, n, random) + val z = 1 / math.sqrt(rank) + val fullData = DenseMatrix.zeros(m, n) + BLAS.gemm(z, A, B, 1.0, fullData) val df = rank * (m + n - rank) val sampSize = scala.math.min(scala.math.round(trainSampFact * df), scala.math.round(.99 * m * n)).toInt val rand = new Random() val mn = m * n - val shuffled = rand.shuffle(1 to mn toList) + val shuffled = rand.shuffle((0 until mn).toList) val omega = shuffled.slice(0, sampSize) val ordered = omega.sortWith(_ < _).toArray val trainData: RDD[(Int, Int, Double)] = sc.parallelize(ordered) - .map(x => (fullData.indexRows(x - 1), fullData.indexColumns(x - 1), fullData.get(x - 1))) + .map(x => (x % m, x / m, fullData.values(x))) // optionally add gaussian noise if (noise) { @@ -105,7 +107,7 @@ object MFDataGenerator { val testOmega = shuffled.slice(sampSize, sampSize + testSampSize) val testOrdered = testOmega.sortWith(_ < _).toArray val testData: RDD[(Int, Int, Double)] = sc.parallelize(testOrdered) - .map(x => (fullData.indexRows(x - 1), fullData.indexColumns(x - 1), fullData.get(x - 1))) + .map(x => (x % m, x / m, fullData.values(x))) testData.map(x => x._1 + "," + x._2 + "," + x._3).saveAsTextFile(outputPath) } diff --git a/mllib/src/main/scala/org/apache/spark/mllib/util/SVMDataGenerator.scala b/mllib/src/main/scala/org/apache/spark/mllib/util/SVMDataGenerator.scala index 7db97e6bac..a8e30cc9d7 100644 --- a/mllib/src/main/scala/org/apache/spark/mllib/util/SVMDataGenerator.scala +++ b/mllib/src/main/scala/org/apache/spark/mllib/util/SVMDataGenerator.scala @@ -19,7 +19,7 @@ package org.apache.spark.mllib.util import scala.util.Random -import org.jblas.DoubleMatrix +import com.github.fommil.netlib.BLAS.{getInstance => blas} import org.apache.spark.annotation.DeveloperApi import org.apache.spark.SparkContext @@ -51,8 +51,7 @@ object SVMDataGenerator { val sc = new SparkContext(sparkMaster, "SVMGenerator") val globalRnd = new Random(94720) - val trueWeights = new DoubleMatrix(1, nfeatures + 1, - Array.fill[Double](nfeatures + 1)(globalRnd.nextGaussian()):_*) + val trueWeights = Array.fill[Double](nfeatures + 1)(globalRnd.nextGaussian()) val data: RDD[LabeledPoint] = sc.parallelize(0 until nexamples, parts).map { idx => val rnd = new Random(42 + idx) @@ -60,7 +59,7 @@ object SVMDataGenerator { val x = Array.fill[Double](nfeatures) { rnd.nextDouble() * 2.0 - 1.0 } - val yD = new DoubleMatrix(1, x.length, x: _*).dot(trueWeights) + rnd.nextGaussian() * 0.1 + val yD = blas.ddot(trueWeights.length, x, 1, trueWeights, 1) + rnd.nextGaussian() * 0.1 val y = if (yD < 0) 0.0 else 1.0 LabeledPoint(y, Vectors.dense(x)) } diff --git a/mllib/src/test/scala/org/apache/spark/mllib/optimization/NNLSSuite.scala b/mllib/src/test/scala/org/apache/spark/mllib/optimization/NNLSSuite.scala index 82c327bd49..22855e4e8f 100644 --- a/mllib/src/test/scala/org/apache/spark/mllib/optimization/NNLSSuite.scala +++ b/mllib/src/test/scala/org/apache/spark/mllib/optimization/NNLSSuite.scala @@ -55,7 +55,7 @@ class NNLSSuite extends FunSuite { for (k <- 0 until 100) { val (ata, atb) = genOnesData(n, rand) - val x = new DoubleMatrix(NNLS.solve(ata, atb, ws)) + val x = new DoubleMatrix(NNLS.solve(ata.data, atb.data, ws)) assert(x.length === n) val answer = DoubleMatrix.ones(n, 1) SimpleBlas.axpy(-1.0, answer, x) @@ -79,7 +79,7 @@ class NNLSSuite extends FunSuite { val goodx = Array(0.13025, 0.54506, 0.2874, 0.0, 0.028628) val ws = NNLS.createWorkspace(n) - val x = NNLS.solve(ata, atb, ws) + val x = NNLS.solve(ata.data, atb.data, ws) for (i <- 0 until n) { assert(x(i) ~== goodx(i) absTol 1E-3) assert(x(i) >= 0) @@ -104,7 +104,7 @@ class NNLSSuite extends FunSuite { val ws = NNLS.createWorkspace(n) - val x = new DoubleMatrix(NNLS.solve(ata, atb, ws)) + val x = new DoubleMatrix(NNLS.solve(ata.data, atb.data, ws)) val obj = computeObjectiveValue(ata, atb, x) assert(obj < refObj + 1E-5) diff --git a/mllib/src/test/scala/org/apache/spark/mllib/stat/KernelDensitySuite.scala b/mllib/src/test/scala/org/apache/spark/mllib/stat/KernelDensitySuite.scala index f6a1e19f50..16ecae23dd 100644 --- a/mllib/src/test/scala/org/apache/spark/mllib/stat/KernelDensitySuite.scala +++ b/mllib/src/test/scala/org/apache/spark/mllib/stat/KernelDensitySuite.scala @@ -21,9 +21,9 @@ import org.scalatest.FunSuite import org.apache.commons.math3.distribution.NormalDistribution -import org.apache.spark.mllib.util.LocalClusterSparkContext +import org.apache.spark.mllib.util.MLlibTestSparkContext -class KernelDensitySuite extends FunSuite with LocalClusterSparkContext { +class KernelDensitySuite extends FunSuite with MLlibTestSparkContext { test("kernel density single sample") { val rdd = sc.parallelize(Array(5.0)) val evaluationPoints = Array(5.0, 6.0) |