Github user mengxr commented on a diff in the pull request:
https://github.com/apache/spark/pull/166#discussion_r10739905
--- Diff:
mllib/src/main/scala/org/apache/spark/mllib/optimization/GradientDescentWithLocalUpdate.scala
---
@@ -0,0 +1,147 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.spark.mllib.optimization
+
+import org.apache.spark.Logging
+import org.apache.spark.rdd.RDD
+
+import org.jblas.DoubleMatrix
+
+import scala.collection.mutable.ArrayBuffer
+import scala.util.Random
+
+/**
+ * Class used to solve an optimization problem using Gradient Descent.
+ * @param gradient Gradient function to be used.
+ * @param updater Updater to be used to update weights after every
iteration.
+ */
+class GradientDescentWithLocalUpdate(gradient: Gradient, updater: Updater)
+ extends GradientDescent(gradient, updater) with Logging
+{
+ private var numLocalIterations: Int = 1
+
+ /**
+ * Set the number of local iterations. Default 1.
+ */
+ def setNumLocalIterations(numLocalIter: Int): this.type = {
+ this.numLocalIterations = numLocalIter
+ this
+ }
+
+ override def optimize(data: RDD[(Double, Array[Double])],
initialWeights: Array[Double])
+ : Array[Double] = {
+
+ val (weights, stochasticLossHistory) =
GradientDescentWithLocalUpdate.runMiniBatchSGD(
+ data,
+ gradient,
+ updater,
+ stepSize,
+ numIterations,
+ numLocalIterations,
+ regParam,
+ miniBatchFraction,
+ initialWeights)
+ weights
+ }
+
+}
+
+// Top-level method to run gradient descent.
+object GradientDescentWithLocalUpdate extends Logging {
+ /**
+ * Run BSP+ gradient descent in parallel using mini batches.
+ *
+ * @param data - Input data for SGD. RDD of form (label, [feature
values]).
+ * @param gradient - Gradient object that will be used to compute the
gradient.
+ * @param updater - Updater object that will be used to update the model.
+ * @param stepSize - stepSize to be used during update.
+ * @param numOuterIterations - number of outer iterations that SGD
should be run.
+ * @param numInnerIterations - number of inner iterations that SGD
should be run.
+ * @param regParam - regularization parameter
+ * @param miniBatchFraction - fraction of the input data set that should
be used for
+ * one iteration of SGD. Default value 1.0.
+ *
+ * @return A tuple containing two elements. The first element is a
column matrix containing
+ * weights for every feature, and the second element is an array
containing the stochastic
+ * loss computed for every iteration.
+ */
+ def runMiniBatchSGD(
+ data: RDD[(Double, Array[Double])],
+ gradient: Gradient,
+ updater: Updater,
+ stepSize: Double,
+ numOuterIterations: Int,
+ numInnerIterations: Int,
+ regParam: Double,
+ miniBatchFraction: Double,
+ initialWeights: Array[Double]) : (Array[Double], Array[Double]) = {
+
+ val stochasticLossHistory = new ArrayBuffer[Double](numOuterIterations)
+
+ val numExamples: Long = data.count()
+ val numPartition = data.partitions.length
+ val miniBatchSize = numExamples * miniBatchFraction / numPartition
+
+ // Initialize weights as a column vector
+ var weights = new DoubleMatrix(initialWeights.length, 1,
initialWeights: _*)
+ var regVal = 0.0
+
+ for (i <- 1 to numOuterIterations) {
+ val weightsAndLosses = data.mapPartitions { iter =>
+ var iterReserved = iter
+ val localLossHistory = new ArrayBuffer[Double](numInnerIterations)
+
+ for (j <- 1 to numInnerIterations) {
+ val (iterCurrent, iterNext) = iterReserved.duplicate
--- End diff --
@yinxusen I think this approach will certainly run OOM if data is too big
to fit into memory. You can set a small executor memory and test some data
without caching.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
