Github user mengxr commented on a diff in the pull request:
https://github.com/apache/spark/pull/268#discussion_r11169437
--- Diff:
mllib/src/main/scala/org/apache/spark/mllib/rdd/VectorRDDFunctions.scala ---
@@ -0,0 +1,170 @@
+/*
+ * 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.rdd
+
+import breeze.linalg.{axpy, Vector => BV}
+
+import org.apache.spark.mllib.linalg.{Vector, Vectors}
+import org.apache.spark.rdd.RDD
+
+/**
+ * Case class of the summary statistics, including mean, variance, count,
max, min, and non-zero
+ * elements count.
+ */
+case class VectorRDDStatisticalSummary(
+ mean: Vector,
+ variance: Vector,
+ count: Long,
+ max: Vector,
+ min: Vector,
+ nonZeroCnt: Vector) extends Serializable
+
+/**
+ * Case class of the aggregate value for collecting summary statistics
from RDD[Vector]. These
+ * values are relatively with
+ * [[org.apache.spark.mllib.rdd.VectorRDDStatisticalSummary
VectorRDDStatisticalSummary]], the
+ * latter is computed from the former.
+ */
+private case class VectorRDDStatisticalRing(
+ fakeMean: BV[Double],
+ fakeM2n: BV[Double],
+ totalCnt: Double,
+ nnz: BV[Double],
+ fakeMax: BV[Double],
+ fakeMin: BV[Double])
+
+/**
+ * Extra functions available on RDDs of
[[org.apache.spark.mllib.linalg.Vector Vector]] through an
+ * implicit conversion. Import `org.apache.spark.MLContext._` at the top
of your program to use
+ * these functions.
+ */
+class VectorRDDFunctions(self: RDD[Vector]) extends Serializable {
+
+ /**
+ * Aggregate function used for aggregating elements in a worker together.
+ */
+ private def seqOp(
+ aggregator: VectorRDDStatisticalRing,
+ currData: BV[Double]): VectorRDDStatisticalRing = {
+ aggregator match {
+ case VectorRDDStatisticalRing(prevMean, prevM2n, cnt, nnzVec,
maxVec, minVec) =>
+ currData.activeIterator.foreach {
+ case (id, value) =>
+ if (maxVec(id) < value) maxVec(id) = value
+ if (minVec(id) > value) minVec(id) = value
+
+ val tmpPrevMean = prevMean(id)
+ prevMean(id) = (prevMean(id) * cnt + value) / (cnt + 1.0)
+ prevM2n(id) += (value - prevMean(id)) * (value - tmpPrevMean)
+
+ nnzVec(id) += 1.0
+ }
+
+ VectorRDDStatisticalRing(
+ prevMean,
+ prevM2n,
+ cnt + 1.0,
+ nnzVec,
+ maxVec,
+ minVec)
+ }
+ }
+
+ /**
+ * Combine function used for combining intermediate results together
from every worker.
+ */
+ private def combOp(
+ statistics1: VectorRDDStatisticalRing,
+ statistics2: VectorRDDStatisticalRing): VectorRDDStatisticalRing = {
+ (statistics1, statistics2) match {
+ case (VectorRDDStatisticalRing(mean1, m2n1, cnt1, nnz1, max1, min1),
+ VectorRDDStatisticalRing(mean2, m2n2, cnt2, nnz2, max2, min2))
=>
+ val totalCnt = cnt1 + cnt2
+ val deltaMean = mean2 - mean1
+
+ mean2.activeIterator.foreach {
+ case (id, 0.0) =>
+ case (id, value) =>
+ mean1(id) = (mean1(id) * nnz1(id) + mean2(id) * nnz2(id)) /
(nnz1(id) + nnz2(id))
+ }
+
+ m2n2.activeIterator.foreach {
+ case (id, 0.0) =>
+ case (id, value) =>
+ m2n1(id) +=
+ value + deltaMean(id) * deltaMean(id) * nnz1(id) * nnz2(id)
/ (nnz1(id)+nnz2(id))
+ }
+
+ max2.activeIterator.foreach {
+ case (id, value) =>
+ if (max1(id) < value) max1(id) = value
+ }
+
+ min2.activeIterator.foreach {
+ case (id, value) =>
+ if (min1(id) > value) min1(id) = value
+ }
+
+ axpy(1.0, nnz2, nnz1)
+ VectorRDDStatisticalRing(mean1, m2n1, totalCnt, nnz1, max1, min1)
+ }
+ }
+
+ /**
+ * Compute full column-wise statistics for the RDD with the size of
Vector as input parameter.
+ */
+ def summarizeStatistics(size: Int): VectorRDDStatisticalSummary = {
+ val zeroValue = VectorRDDStatisticalRing(
+ BV.zeros[Double](size),
+ BV.zeros[Double](size),
+ 0.0,
+ BV.zeros[Double](size),
+ BV.fill(size)(Double.MinValue),
+ BV.fill(size)(Double.MaxValue))
+
+ val VectorRDDStatisticalRing(fakeMean, fakeM2n, totalCnt, nnz,
fakeMax, fakeMin) =
+ self.map(_.toBreeze).aggregate(zeroValue)(seqOp, combOp)
+
+ // solve real mean
+ val realMean = fakeMean :* nnz :/ totalCnt
+
+ // solve real m2n
+ val deltaMean = fakeMean
+ val realM2n = fakeM2n - ((deltaMean :* deltaMean) :* (nnz :* (nnz :-
totalCnt)) :/ totalCnt)
+
+ // remove the initial value in max and min, i.e. the Double.MaxValue
or Double.MinValue.
--- End diff --
You should check for every index where `nnz(i) != n`. Don't need to handle
Double.MinValue / Double.MaxValue explicitly.
---
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.
---
