The confusion order of rows in SVD matrix ?

[buring]

Hi:
        I want to use SVD in my work. I tried some examples and have some
confusions. The input the 4*3 matrix as follows:
        2 0 0
        0 3 2
        0 3 1
        2 0 3
        My input file text as follows which is corresponding to the matrix
        0 0 2
        1 1 3
        1 2 2
        2 2 1
        2 1 3
        3 0 2
        3 2 3 
        After run the svd algorithm ,I tried to reCompute the input matrix 
through
U*S*V.T.But I found that ,The input matrix's row  is not as expected,I
printed it out:
        2 0 0
        0 3 1
        0 3 2
        2 0 3
        which rows 2 exchanged with rows 3,I confused on this.Can any one 
explain?

        My code is that :
        val inputData = sc.textFile(fname).map{
      line=>
        val parts = line.trim.split(' ')
        (parts(0).toLong,parts(1).toInt,parts(2).toDouble)
    }

    val dataRows = inputData.groupBy(_._1).map[(Long, Vector)]{ row =>
      val (indices, values) = row._2.map(e => (e._2, e._3)).unzip
      (row._1, new SparseVector(ncol, indices.toArray, values.toArray))
    }
    val data =
dataRows.take(dataRows.count().toInt).map(e=>e._1+e._2.toArray.mkString(";"))
    logInfo("----------------")
    logInfo(data(0))
    logInfo(data(1))
    logInfo(data(2))
    logInfo(data(3))

    Here is the print code ,and this matrix is the same as the reCompute
matrix U*S*V.T. But the order of rows has changed which I can't understand.I
want to have the same matrix as the input file. How to guarantee this?

    Thanks！



--
View this message in context: 
http://apache-spark-user-list.1001560.n3.nabble.com/The-confusion-order-of-rows-in-SVD-matrix-tp15337.html
Sent from the Apache Spark User List mailing list archive at Nabble.com.

---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscr...@spark.apache.org
For additional commands, e-mail: user-h...@spark.apache.org