[WIP] [SPARK-1328] Add vector statistics #268

yinxusen · 2014-03-29T03:42:53Z

As with the new vector system in MLlib, we find that it is good to add some new APIs to precess the RDD[Vector]. Beside, the former implementation of computeStat is not stable which could loss precision, and has the possibility to cause Nan in scientific computing, just as said in the SPARK-1328.

APIs contain:

rowMeans(): RDD[Double]
rowNorm2(): RDD[Double]
rowSDs(): RDD[Double]
colMeans(): Vector
colMeans(size: Int): Vector
colNorm2(): Vector
colNorm2(size: Int): Vector
colSDs(): Vector
colSDs(size: Int): Vector
maxOption((Vector, Vector) => Boolean): Option[Vector]
minOption((Vector, Vector) => Boolean): Option[Vector]
rowShrink(): RDD[Vector]
colShrink(): RDD[Vector]

This is working in process now, and some more APIs will add to LabeledPoint. Moreover, the implicit declaration will move from MLUtils to MLContext later.

AmplabJenkins · 2014-03-29T03:47:22Z

Merged build triggered. Build is starting -or- tests failed to complete.

AmplabJenkins · 2014-03-29T03:47:32Z

Merged build started. Build is starting -or- tests failed to complete.

AmplabJenkins · 2014-03-29T04:39:35Z

Merged build finished. All automated tests passed.

AmplabJenkins · 2014-03-29T04:39:35Z

All automated tests passed.
Refer to this link for build results: https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/13565/

mengxr · 2014-03-29T06:43:31Z

@yinxusen Thanks for working on this! I don't think row statistics are important because they represent values for different features. For column statistics, instead of implementing each statistic separately, we can compute all common statistics like (n, nnz, mean, variance, max, min) in a single job. This adds little overhead to the computation.

Btw, Vector.toArray is an expensive operation for SparseVector. You should use breeze's axpy to aggregate vectors.

yinxusen · 2014-03-30T05:29:34Z

Yes, @mengxr I find it is cool to merge them together and rewrite the SD in a single pass, just like the page said. I'll fix it ASAP.

AmplabJenkins · 2014-03-30T08:47:22Z

Build triggered. Build is starting -or- tests failed to complete.

AmplabJenkins · 2014-03-30T08:47:28Z

Build started. Build is starting -or- tests failed to complete.

yinxusen · 2014-03-30T08:50:15Z

I remove all APIs and reserve a new one call statistics(), which can compute six statistical quantities in a single pass.

Row-wise statistical quantities are removed because they are meaningless.

Column-wise shrinkage and row-wise shrinkage are also removed, because they belong to feature/data selection. We should re-consider them and write in other semantic class such as LabeledPoint.

AmplabJenkins · 2014-03-30T09:39:47Z

Build finished. All automated tests passed.

AmplabJenkins · 2014-03-30T09:39:47Z

All automated tests passed.
Refer to this link for build results: https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/13585/

mengxr · 2014-03-30T17:59:35Z

mllib/src/main/scala/org/apache/spark/mllib/rdd/VectorRDDFunctions.scala

+   * }}},
+   * with the size of Vector as input parameter.
+   */
+  def statistics(size: Int): (Vector, Vector, Double, Vector, Vector, Vector) = {


Please wrap the return type to a class called VectorRDDStatisticalSummary that provides mean, variance, count: Long, min, max, and std. So later we can add more summary statistics to it. I would also recommend changing the method name to summarize or summarizeStatistics.

mengxr · 2014-03-30T18:23:39Z

@yinxusen I want to see whether we can improve the performance. First of all, the complexity should be O(nnz) instead of O(n d). This is a little bit tricky. Basically, we don't need to update zeros for a sparse vector. Suppose we have the following column:

1.0
0.0
2.0
0.0
3.0
0.0
0.0

To apply the formula for mean/variance, you need to do the update at the second row. This is actually not necessary because the summary statistics are invariant to the ordering. Imagine we have this column re-arranged:

1.0
2.0
3.0
0.0
0.0
0.0
0.0

which still have the same summary statistics. We can skip all zeros and do mean/variance updates based on the non-zero count instead of global count. After aggregate, we know we need to accumulate 4 zeros to the column, which is a constant-time operation.

I'm okay if you don't have enough time to make the change in this PR. Put a TODO and we can fix it later.

yinxusen · 2014-03-31T16:34:50Z

@mengxr I am not very sure of the concept of sparse vector. In your example, do you mean the column is Vector(1.0, 0.0, 2.0, 0.0, 3.0, 0.0, 0.0) or
RDD( Vector(1.0), Vector(0.0), Vector(2.0), Vector(0.0), Vector(3.0), Vector(0.0), Vector(0.0) )?

If it is the case 1, then it is easy to rewrite it in O(nnz), otherwise, it will be difficult, because we cannot judge whether a column is sparse or not before we count the nnz. If the case 1 is your mean, then I think I should treat sparse vector different with the dense one with the following code:

RDD.take(1).head.type match { case DenseVector[Double] => xxx case SparseVector[Double] => xxx }.

mengxr · 2014-03-31T17:29:36Z

@yinxusen I mean a column. You don't need to check the type as you already use activeIterator to get the non-zero elements. The approach I suggested is to only update mean/variance/min/max for non-zeros. After we scan all the items, we know how many zeros we need to append to each column, and then we do a constant-time update for each column.

yinxusen · 2014-04-01T06:27:14Z

@mengxr Ah... I totally understand your mean. Code is on the way.

AmplabJenkins · 2014-04-01T13:42:22Z

Build triggered.

AmplabJenkins · 2014-04-01T13:42:29Z

Build started.

AmplabJenkins · 2014-04-01T13:44:31Z

Build finished.

AmplabJenkins · 2014-04-01T13:44:32Z

Refer to this link for build results: https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/13630/

yinxusen · 2014-04-01T14:23:41Z

mllib/src/test/scala/org/apache/spark/mllib/rdd/VectorRDDFunctionsSuite.scala

+    math.abs(lhs - rhs) / denominator < 0.3
+  }
+}
+


Build failed might be caught by this empty line in the end. I'll fix it in next commit with other problems.

… be evaluated only once

AmplabJenkins · 2014-04-10T22:48:11Z

Merged build triggered.

AmplabJenkins · 2014-04-10T22:54:57Z

Merged build started.

AmplabJenkins · 2014-04-10T23:33:23Z

Merged build finished. All automated tests passed.

AmplabJenkins · 2014-04-10T23:33:23Z

All automated tests passed.
Refer to this link for build results: https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/14025/

yinxusen · 2014-04-11T00:01:38Z

Conflict in MLUtils and RowMatrix. I think it is OK now.
2014-4-11 AM5:31于 "Matei Zaharia" notifications@github.com写道：

Hey, unfortunately this no longer merges cleanly. Do you mind rebasing it?
I think some conflicting changes happened in MLUtils.

—
Reply to this email directly or view it on GitHubhttps://github.com//pull/268#issuecomment-40145491
.

pwendell · 2014-04-12T02:44:13Z

Thanks, merged!

As with the new vector system in MLlib, we find that it is good to add some new APIs to precess the `RDD[Vector]`. Beside, the former implementation of `computeStat` is not stable which could loss precision, and has the possibility to cause `Nan` in scientific computing, just as said in the [SPARK-1328](https://spark-project.atlassian.net/browse/SPARK-1328). APIs contain: * rowMeans(): RDD[Double] * rowNorm2(): RDD[Double] * rowSDs(): RDD[Double] * colMeans(): Vector * colMeans(size: Int): Vector * colNorm2(): Vector * colNorm2(size: Int): Vector * colSDs(): Vector * colSDs(size: Int): Vector * maxOption((Vector, Vector) => Boolean): Option[Vector] * minOption((Vector, Vector) => Boolean): Option[Vector] * rowShrink(): RDD[Vector] * colShrink(): RDD[Vector] This is working in process now, and some more APIs will add to `LabeledPoint`. Moreover, the implicit declaration will move from `MLUtils` to `MLContext` later. Author: Xusen Yin <yinxusen@gmail.com> Author: Xiangrui Meng <meng@databricks.com> Closes #268 from yinxusen/vector-statistics and squashes the following commits: d61363f [Xusen Yin] rebase to latest master 16ae684 [Xusen Yin] fix minor error and remove useless method 10cf5d3 [Xusen Yin] refine some return type b064714 [Xusen Yin] remove computeStat in MLUtils cbbefdb [Xiangrui Meng] update multivariate statistical summary interface and clean tests 4eaf28a [Xusen Yin] merge VectorRDDStatistics into RowMatrix 48ee053 [Xusen Yin] fix minor error e624f93 [Xusen Yin] fix scala style error 1fba230 [Xusen Yin] merge while loop together 69e1f37 [Xusen Yin] remove lazy eval, and minor memory footprint 548e9de [Xusen Yin] minor revision 86522c4 [Xusen Yin] add comments on functions dc77e38 [Xusen Yin] test sparse vector RDD 18cf072 [Xusen Yin] change def to lazy val to make sure that the computations in function be evaluated only once f7a3ca2 [Xusen Yin] fix the corner case of maxmin 967d041 [Xusen Yin] full revision with Aggregator class 138300c [Xusen Yin] add new Aggregator class 1376ff4 [Xusen Yin] rename variables and adjust code 4a5c38d [Xusen Yin] add scala doc, refine code and comments 036b7a5 [Xusen Yin] fix the bug of Nan occur f6e8e9a [Xusen Yin] add sparse vectors test 4cfbadf [Xusen Yin] fix bug of min max 4e4fbd1 [Xusen Yin] separate seqop and combop out as independent functions a6d5a2e [Xusen Yin] rewrite for only computing non-zero elements 3980287 [Xusen Yin] rename variables 62a2c3e [Xusen Yin] use axpy and in-place if possible 9a75ebd [Xusen Yin] add case class to wrap return values d816ac7 [Xusen Yin] remove useless APIs c4651bb [Xusen Yin] remove row-wise APIs and refine code 1338ea1 [Xusen Yin] all-in-one version test passed cc65810 [Xusen Yin] add parallel mean and variance 9af2e95 [Xusen Yin] refine the code style ad6c82d [Xusen Yin] add shrink test e09d5d2 [Xusen Yin] add scala docs and refine shrink method 8ef3377 [Xusen Yin] pass all tests 28cf060 [Xusen Yin] fix error of column means 54b19ab [Xusen Yin] add new API to shrink RDD[Vector] 8c6c0e1 [Xusen Yin] add basic statistics

As with the new vector system in MLlib, we find that it is good to add some new APIs to precess the `RDD[Vector]`. Beside, the former implementation of `computeStat` is not stable which could loss precision, and has the possibility to cause `Nan` in scientific computing, just as said in the [SPARK-1328](https://spark-project.atlassian.net/browse/SPARK-1328). APIs contain: * rowMeans(): RDD[Double] * rowNorm2(): RDD[Double] * rowSDs(): RDD[Double] * colMeans(): Vector * colMeans(size: Int): Vector * colNorm2(): Vector * colNorm2(size: Int): Vector * colSDs(): Vector * colSDs(size: Int): Vector * maxOption((Vector, Vector) => Boolean): Option[Vector] * minOption((Vector, Vector) => Boolean): Option[Vector] * rowShrink(): RDD[Vector] * colShrink(): RDD[Vector] This is working in process now, and some more APIs will add to `LabeledPoint`. Moreover, the implicit declaration will move from `MLUtils` to `MLContext` later. Author: Xusen Yin <yinxusen@gmail.com> Author: Xiangrui Meng <meng@databricks.com> Closes apache#268 from yinxusen/vector-statistics and squashes the following commits: d61363f [Xusen Yin] rebase to latest master 16ae684 [Xusen Yin] fix minor error and remove useless method 10cf5d3 [Xusen Yin] refine some return type b064714 [Xusen Yin] remove computeStat in MLUtils cbbefdb [Xiangrui Meng] update multivariate statistical summary interface and clean tests 4eaf28a [Xusen Yin] merge VectorRDDStatistics into RowMatrix 48ee053 [Xusen Yin] fix minor error e624f93 [Xusen Yin] fix scala style error 1fba230 [Xusen Yin] merge while loop together 69e1f37 [Xusen Yin] remove lazy eval, and minor memory footprint 548e9de [Xusen Yin] minor revision 86522c4 [Xusen Yin] add comments on functions dc77e38 [Xusen Yin] test sparse vector RDD 18cf072 [Xusen Yin] change def to lazy val to make sure that the computations in function be evaluated only once f7a3ca2 [Xusen Yin] fix the corner case of maxmin 967d041 [Xusen Yin] full revision with Aggregator class 138300c [Xusen Yin] add new Aggregator class 1376ff4 [Xusen Yin] rename variables and adjust code 4a5c38d [Xusen Yin] add scala doc, refine code and comments 036b7a5 [Xusen Yin] fix the bug of Nan occur f6e8e9a [Xusen Yin] add sparse vectors test 4cfbadf [Xusen Yin] fix bug of min max 4e4fbd1 [Xusen Yin] separate seqop and combop out as independent functions a6d5a2e [Xusen Yin] rewrite for only computing non-zero elements 3980287 [Xusen Yin] rename variables 62a2c3e [Xusen Yin] use axpy and in-place if possible 9a75ebd [Xusen Yin] add case class to wrap return values d816ac7 [Xusen Yin] remove useless APIs c4651bb [Xusen Yin] remove row-wise APIs and refine code 1338ea1 [Xusen Yin] all-in-one version test passed cc65810 [Xusen Yin] add parallel mean and variance 9af2e95 [Xusen Yin] refine the code style ad6c82d [Xusen Yin] add shrink test e09d5d2 [Xusen Yin] add scala docs and refine shrink method 8ef3377 [Xusen Yin] pass all tests 28cf060 [Xusen Yin] fix error of column means 54b19ab [Xusen Yin] add new API to shrink RDD[Vector] 8c6c0e1 [Xusen Yin] add basic statistics

mengxr reviewed Mar 30, 2014
View reviewed changes

mengxr mentioned this pull request Apr 1, 2014

[SPARK-1212, Part II] Support sparse data in MLlib #245

Closed

yinxusen reviewed Apr 1, 2014
View reviewed changes

yinxusen and others added 14 commits April 11, 2014 06:26

change def to lazy val to make sure that the computations in function…

18cf072

… be evaluated only once

test sparse vector RDD

dc77e38

add comments on functions

86522c4

minor revision

548e9de

remove lazy eval, and minor memory footprint

69e1f37

merge while loop together

1fba230

fix scala style error

e624f93

fix minor error

48ee053

merge VectorRDDStatistics into RowMatrix

4eaf28a

update multivariate statistical summary interface and clean tests

cbbefdb

remove computeStat in MLUtils

b064714

refine some return type

10cf5d3

fix minor error and remove useless method

16ae684

rebase to latest master

d61363f

asfgit closed this in fdfb45e Apr 12, 2014

mccheah pushed a commit to mccheah/spark that referenced this pull request Oct 12, 2017

Correct publish location (apache#268)

1a24446

Igosuki pushed a commit to Adikteev/spark that referenced this pull request Jul 31, 2018

Removed spark command.json because it is no longer needed (apache#268)

4c71ffb

bzhaoopenstack pushed a commit to bzhaoopenstack/spark that referenced this pull request Sep 11, 2019

Unify the network name used by openlab jobs (apache#268)

df49a9b

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[WIP] [SPARK-1328] Add vector statistics #268

[WIP] [SPARK-1328] Add vector statistics #268

yinxusen commented Mar 29, 2014

AmplabJenkins commented Mar 29, 2014

AmplabJenkins commented Mar 29, 2014

AmplabJenkins commented Mar 29, 2014

AmplabJenkins commented Mar 29, 2014

mengxr commented Mar 29, 2014

yinxusen commented Mar 30, 2014

AmplabJenkins commented Mar 30, 2014

AmplabJenkins commented Mar 30, 2014

yinxusen commented Mar 30, 2014

AmplabJenkins commented Mar 30, 2014

AmplabJenkins commented Mar 30, 2014

mengxr Mar 30, 2014

mengxr commented Mar 30, 2014

yinxusen commented Mar 31, 2014

mengxr commented Mar 31, 2014

yinxusen commented Apr 1, 2014

AmplabJenkins commented Apr 1, 2014

AmplabJenkins commented Apr 1, 2014

AmplabJenkins commented Apr 1, 2014

AmplabJenkins commented Apr 1, 2014

yinxusen Apr 1, 2014

AmplabJenkins commented Apr 10, 2014

AmplabJenkins commented Apr 10, 2014

AmplabJenkins commented Apr 10, 2014

AmplabJenkins commented Apr 10, 2014

yinxusen commented Apr 11, 2014

pwendell commented Apr 12, 2014

[WIP] [SPARK-1328] Add vector statistics #268

[WIP] [SPARK-1328] Add vector statistics #268

Conversation

yinxusen commented Mar 29, 2014

AmplabJenkins commented Mar 29, 2014

AmplabJenkins commented Mar 29, 2014

AmplabJenkins commented Mar 29, 2014

AmplabJenkins commented Mar 29, 2014

mengxr commented Mar 29, 2014

yinxusen commented Mar 30, 2014

AmplabJenkins commented Mar 30, 2014

AmplabJenkins commented Mar 30, 2014

yinxusen commented Mar 30, 2014

AmplabJenkins commented Mar 30, 2014

AmplabJenkins commented Mar 30, 2014

mengxr Mar 30, 2014

Choose a reason for hiding this comment

mengxr commented Mar 30, 2014

yinxusen commented Mar 31, 2014

mengxr commented Mar 31, 2014

yinxusen commented Apr 1, 2014

AmplabJenkins commented Apr 1, 2014

AmplabJenkins commented Apr 1, 2014

AmplabJenkins commented Apr 1, 2014

AmplabJenkins commented Apr 1, 2014

yinxusen Apr 1, 2014

Choose a reason for hiding this comment

AmplabJenkins commented Apr 10, 2014

AmplabJenkins commented Apr 10, 2014

AmplabJenkins commented Apr 10, 2014

AmplabJenkins commented Apr 10, 2014

yinxusen commented Apr 11, 2014

pwendell commented Apr 12, 2014